Package Details: chromium-dev 104.0.5110.0-1

Git Clone URL: https://aur.archlinux.org/chromium-dev.git (read-only, click to copy)
Package Base: chromium-dev
Description: The open-source project behind Google Chrome (Dev Channel)
Upstream URL: http://www.chromium.org
Keywords: browser web
Licenses: BSD
Submitter: None
Maintainer: sl1pkn07
Last Packager: sl1pkn07
Votes: 159
Popularity: 0.005713
First Submitted: 2010-05-17 09:04 (UTC)
Last Updated: 2022-06-11 16:49 (UTC)

Dependencies (30)

Required by (0)

Sources (10)

Latest Comments

DAC324 commented on 2022-06-22 08:11 (UTC) (edited on 2022-06-22 10:57 (UTC) by DAC324)

Nevermind. Looks like enable_nacl=true adds some comprehensive testing if 'disable_fieldtrial_testing_config=false' - in addition to the tests performed for the browser.

I had `'disable_fieldtrial_testing_config=false' due to a misunderstanding on my side as I thought it might be related to enabling VAAPIVideoDecoder but that's not the case.

So you can safely keep 'disable_fieldtrial_testing_config=true'

With that, you can build chromium with Native Client enabled and working (just tested it with one of the SSH extensions).

sl1pkn07 commented on 2022-06-21 18:31 (UTC) (edited on 2022-06-21 18:32 (UTC) by sl1pkn07)

another hidden dependency. great

and yes. working. at least, in my pc

greetings

DAC324 commented on 2022-06-21 18:22 (UTC)

Set icu to be using the bundled version. Chromium still does not build when enable_nacl is set to true:

ninja: Entering directory `out/Release'
ninja: error: '../../content/test/data/web_ui_test.test-mojom', needed by 'gen/mojom-webui/content/test/data/web_ui_test.test-mojom-webui.js', missing and no known rule to make it

DAC324 commented on 2022-06-21 15:58 (UTC)

OK, Thank you very much. Looks like there are more and more components that cannot be unbundled anymore.

I predict chromium will soon be like emacs [1] but here we can say

Chrome/Chromium? Yes, they are great operating systems! All they are lacking is a decent browser :D

[1] https://news.ycombinator.com/item?id=7978048

sl1pkn07 commented on 2022-06-21 14:26 (UTC)

As i put in the pkgbuild. nacl needs bundles ICU instead of system ones

DAC324 commented on 2022-06-21 14:24 (UTC) (edited on 2022-06-21 14:25 (UTC) by DAC324)

I have just seen that the following statements have been added to PKGBUILD to enable building the Native Client:

  # https://crbug.com/1306914.
  sed '/<limits.h>/a#undef PTHREAD_STACK_MIN\n#define PTHREAD_STACK_MIN 16384' -i native_client/src/shared/platform/posix/nacl_threads.c
  sed '/man.h>/a#undef SIGSTKSZ\n#define SIGSTKSZ 8192' -i native_client/src/trusted/service_runtime/posix/nacl_signal_stack.c

Was anybody successful yet in enabling Native Client? Reason I am asking is that the NaCl build process ignores $CFLAGS, using its own set of compiler flags. As such, -I/usr/include is ignored as well, resulting in the NaCl build process being unable to find system headers:

In file included from irt_x64/gen/shim_headers/icui18n_shim/third_party/icu/source/i18n/unicode/calendar.h:5:
/usr/include/unicode/calendar.h:30:10: fatal error: 'unicode/utypes.h' file not found
#include "unicode/utypes.h"
         ^~~~~~~~~~~~~~~~~~
1 error generated.

DAC324 commented on 2022-03-15 09:07 (UTC) (edited on 2022-03-16 17:51 (UTC) by DAC324)

I just compiled the 101.0.4945.0 sources using this package's PKGBUILD. If you leave nacl enabled with that version, it does not compile:

../../native_client/src/shared/platform/posix/nacl_threads.c:32:59: error: function-like macro '__sysconf' is not defined
#if !defined(__native_client__) && NACL_KERN_STACK_SIZE < PTHREAD_STACK_MIN
                                                          ^
/usr/include/bits/pthread_stack_min-dynamic.h:26:30: note: expanded from macro 'PTHREAD_STACK_MIN'
#   define PTHREAD_STACK_MIN __sysconf (__SC_THREAD_STACK_MIN_VALUE)
                             ^
../../native_client/src/shared/platform/posix/nacl_threads.c:47:18: warning: comparison of integers of different signs: 'size_t' (aka 'unsigned long') and 'long' [-Wsign-compare]
  if (stack_size < PTHREAD_STACK_MIN) {
      ~~~~~~~~~~ ^ ~~~~~~~~~~~~~~~~~
1 warning and 1 error generated.

Also, third_party/tcmalloc has to be commented out as it does not seem to be still included in 101.0.4945.0.

Due to the 101.0.4945.0 not being able to compile with nacl enabled, I for now had to disable that by setting 'enable_nacl=false'.

As a consequence, all entries in _nacl_libs have to be commented out. Finally, headless_lib.pak is not there in 101.0.4945.0 so it has to be commented out, either.

sl1pkn07 commented on 2022-03-14 16:01 (UTC) (edited on 2022-03-14 16:02 (UTC) by sl1pkn07)

is a typo

about nacl. im still use it because https://chrome.google.com/webstore/detail/secure-shell/iodihamcpbpeioajjeobimgagajmlibd needs it

your patch provided below fix the issue using system icu AND enable nacl?

DAC324 commented on 2022-03-14 15:55 (UTC)

The _keeplibs variable currently reads as follows:

...
           'third_party/swiftshader/third_party/SPIRV-Headers/include/spirv/unified1'
           'third_party/tcmalloc'
           'third_party/tensorflow-text'
           'third_party/tflite'
           'third_party/tflite/src/third_party/eigen3'
           'third_party/tflite/src/third_party/fft2d'
           'third_party/tcmalloc'
...

Is there a special reason why 'third_party/tcmalloc' is included twice?

DAC324 commented on 2022-02-04 22:35 (UTC)

Just confirmed: Chromium 99.0.4844.9 runs with

   'enable_nacl=false'
   'enable_nacl_nonsfi=false'
   'enable_hangout_services_extension=false'
   'enable_widevine=false'

and with the system ICU. Thanks for the good work providing this package!

DAC324 commented on 2022-02-04 18:31 (UTC)

It is also possible to use system icu after this patch to the sources:

--- a/base/BUILD.gn 2021-08-04 18:26:08.618031942 +0200
+++ b/base/BUILD.gn 2021-08-04 18:27:08.884244626 +0200
@@ -762,7 +762,6 @@
     "third_party/cityhash/city.h",
     "third_party/cityhash_v103/src/city_v103.cc",
     "third_party/cityhash_v103/src/city_v103.h",
-    "third_party/icu/icu_utf.h",
     "third_party/nspr/prtime.cc",
     "third_party/nspr/prtime.h",
     "third_party/superfasthash/superfasthash.c",
--- a/base/files/file_path.cc   2021-08-04 18:27:50.660596514 +0200
+++ b/base/files/file_path.cc   2021-08-04 18:30:29.379400090 +0200
@@ -26,7 +26,7 @@

 #if defined(OS_APPLE)
 #include "base/mac/scoped_cftyperef.h"
-#include "base/third_party/icu/icu_utf.h"
+#include "unicode/utf.h"
 #endif

 #if defined(OS_WIN)
@@ -1186,9 +1186,9 @@
                                             int* index) {
   int codepoint = 0;
   while (*index < length && codepoint == 0) {
-    // CBU8_NEXT returns a value < 0 in error cases. For purposes of string
+    // U8_NEXT returns a value < 0 in error cases. For purposes of string
     // comparison, we just use that value and flag it with DCHECK.
-    CBU8_NEXT(string, *index, length, codepoint);
+    U8_NEXT(string, *index, length, codepoint);
     DCHECK_GT(codepoint, 0);
     if (codepoint > 0) {
       // Check if there is a subtable for this upper byte.
--- a/base/json/json_parser.cc  2021-08-04 18:35:06.730641327 +0200
+++ b/base/json/json_parser.cc  2021-08-04 18:36:54.193163718 +0200
@@ -19,7 +19,7 @@
 #include "base/strings/stringprintf.h"
 #include "base/strings/utf_string_conversion_utils.h"
 #include "base/strings/utf_string_conversions.h"
-#include "base/third_party/icu/icu_utf.h"
+#include <unicode/utf.h>
 #include "third_party/abseil-cpp/absl/types/optional.h"

 namespace base {
@@ -639,9 +639,9 @@

   // If this is a high surrogate, consume the next code unit to get the
   // low surrogate.
-  if (CBU16_IS_SURROGATE(code_unit16_high)) {
+  if (U16_IS_SURROGATE(code_unit16_high)) {
     // Make sure this is the high surrogate.
-    if (!CBU16_IS_SURROGATE_LEAD(code_unit16_high)) {
+    if (!U16_IS_SURROGATE_LEAD(code_unit16_high)) {
       if ((options_ & JSON_REPLACE_INVALID_CHARACTERS) == 0)
         return false;
       *out_code_point = kUnicodeReplacementPoint;
@@ -665,7 +665,7 @@
     if (!UnprefixedHexStringToInt(*escape_sequence, &code_unit16_low))
       return false;

-    if (!CBU16_IS_TRAIL(code_unit16_low)) {
+    if (!U16_IS_TRAIL(code_unit16_low)) {
       if ((options_ & JSON_REPLACE_INVALID_CHARACTERS) == 0)
         return false;
       *out_code_point = kUnicodeReplacementPoint;
@@ -673,12 +673,12 @@
     }

     uint32_t code_point =
-        CBU16_GET_SUPPLEMENTARY(code_unit16_high, code_unit16_low);
+        U16_GET_SUPPLEMENTARY(code_unit16_high, code_unit16_low);

     *out_code_point = code_point;
   } else {
     // Not a surrogate.
-    DCHECK(CBU16_IS_SINGLE(code_unit16_high));
+    DCHECK(U16_IS_SINGLE(code_unit16_high));

     *out_code_point = code_unit16_high;
   }
--- a/base/json/string_escape.cc    2021-08-04 17:30:48.973063734 +0200
+++ b/base/json/string_escape.cc    2021-08-04 18:46:25.742182189 +0200
@@ -14,7 +14,7 @@
 #include "base/strings/stringprintf.h"
 #include "base/strings/utf_string_conversion_utils.h"
 #include "base/strings/utf_string_conversions.h"
-#include "base/third_party/icu/icu_utf.h"
+#include <unicode/utf.h>

 namespace base {

--- a/base/strings/pattern.cc   2021-08-04 18:47:56.511496789 +0200
+++ b/base/strings/pattern.cc   2021-08-04 18:51:26.059914447 +0200
@@ -4,13 +4,13 @@

 #include "base/strings/pattern.h"

-#include "base/third_party/icu/icu_utf.h"
+#include <unicode/utf.h>

 namespace base {

 namespace {

-constexpr bool IsWildcard(base_icu::UChar32 character) {
+constexpr bool IsWildcard(UChar32 character) {
   return character == '*' || character == '?';
 }

@@ -55,9 +55,9 @@
       // Check if the chars match, if so, increment the ptrs.
       const CHAR* pattern_next = *pattern;
       const CHAR* string_next = *string;
-      base_icu::UChar32 pattern_char = next(&pattern_next, pattern_end);
+      UChar32 pattern_char = next(&pattern_next, pattern_end);
       if (pattern_char == next(&string_next, string_end) &&
-          pattern_char != CBU_SENTINEL) {
+          pattern_char != U_SENTINEL) {
         *pattern = pattern_next;
         *string = string_next;
         continue;
@@ -121,20 +121,20 @@
 }

 struct NextCharUTF8 {
-  base_icu::UChar32 operator()(const char** p, const char* end) {
-    base_icu::UChar32 c;
+  UChar32 operator()(const char** p, const char* end) {
+    UChar32 c;
     int offset = 0;
-    CBU8_NEXT(*p, offset, end - *p, c);
+    U8_NEXT(*p, offset, end - *p, c);
     *p += offset;
     return c;
   }
 };

 struct NextCharUTF16 {
-  base_icu::UChar32 operator()(const char16_t** p, const char16_t* end) {
-    base_icu::UChar32 c;
+  UChar32 operator()(const char16_t** p, const char16_t* end) {
+    UChar32 c;
     int offset = 0;
-    CBU16_NEXT(*p, offset, end - *p, c);
+    U16_NEXT(*p, offset, end - *p, c);
     *p += offset;
     return c;
   }
--- a/base/strings/string_util.cc   2021-08-04 18:53:29.892372670 +0200
+++ b/base/strings/string_util.cc   2021-08-04 18:55:34.808145957 +0200
@@ -27,7 +27,7 @@
 #include "base/strings/string_util_internal.h"
 #include "base/strings/utf_string_conversion_utils.h"
 #include "base/strings/utf_string_conversions.h"
-#include "base/third_party/icu/icu_utf.h"
+#include <unicode/utf.h>
 #include "build/build_config.h"

 namespace base {
@@ -173,19 +173,19 @@
   }
   DCHECK_LE(byte_size,
             static_cast<uint32_t>(std::numeric_limits<int32_t>::max()));
-  // Note: This cast is necessary because CBU8_NEXT uses int32_ts.
+  // Note: This cast is necessary because U8_NEXT uses int32_ts.
   int32_t truncation_length = static_cast<int32_t>(byte_size);
   int32_t char_index = truncation_length - 1;
   const char* data = input.data();

-  // Using CBU8, we will move backwards from the truncation point
+  // Using U8, we will move backwards from the truncation point
   // to the beginning of the string looking for a valid UTF8
   // character.  Once a full UTF8 character is found, we will
   // truncate the string to the end of that character.
   while (char_index >= 0) {
     int32_t prev = char_index;
-    base_icu::UChar32 code_point = 0;
-    CBU8_NEXT(data, char_index, truncation_length, code_point);
+    UChar32 code_point = 0;
+    U8_NEXT(data, char_index, truncation_length, code_point);
     if (!IsValidCharacter(code_point) ||
         !IsValidCodepoint(code_point)) {
       char_index = prev - 1;
--- a/ui/gfx/utf16_indexing.cc  2021-08-04 19:02:42.921692352 +0200
+++ b/ui/gfx/utf16_indexing.cc  2021-08-04 19:03:40.487934831 +0200
@@ -5,13 +5,13 @@
 #include "ui/gfx/utf16_indexing.h"

 #include "base/check_op.h"
-#include "base/third_party/icu/icu_utf.h"
+#include <unicode/utf.h>

 namespace gfx {

 bool IsValidCodePointIndex(const std::u16string& s, size_t index) {
   return index == 0 || index == s.length() ||
-    !(CBU16_IS_TRAIL(s[index]) && CBU16_IS_LEAD(s[index - 1]));
+    !(U16_IS_TRAIL(s[index]) && U16_IS_LEAD(s[index - 1]));
 }

 ptrdiff_t UTF16IndexToOffset(const std::u16string& s, size_t base, size_t pos) {
--- a/base/strings/utf_string_conversion_utils.cc.orig  2021-08-04 18:56:28.594425021 +0200
+++ b/base/strings/utf_string_conversion_utils.cc   2021-08-04 19:54:54.358528026 +0200
@@ -4,7 +4,7 @@

 #include "base/strings/utf_string_conversion_utils.h"

-#include "base/third_party/icu/icu_utf.h"
+#include <unicode/utf.h>
 #include "build/build_config.h"

 namespace base {
@@ -19,7 +19,7 @@
   // use a signed type for code_point.  But this function returns false
   // on error anyway, so code_point_out is unsigned.
   int32_t code_point;
-  CBU8_NEXT(src, *char_index, src_len, code_point);
+  U8_NEXT(src, *char_index, src_len, code_point);
   *code_point_out = static_cast<uint32_t>(code_point);

   // The ICU macro above moves to the next char, we want to point to the last
@@ -34,16 +34,16 @@
                           int32_t src_len,
                           int32_t* char_index,
                           uint32_t* code_point) {
-  if (CBU16_IS_SURROGATE(src[*char_index])) {
-    if (!CBU16_IS_SURROGATE_LEAD(src[*char_index]) ||
+  if (U16_IS_SURROGATE(src[*char_index])) {
+    if (!U16_IS_SURROGATE_LEAD(src[*char_index]) ||
         *char_index + 1 >= src_len ||
-        !CBU16_IS_TRAIL(src[*char_index + 1])) {
+        !U16_IS_TRAIL(src[*char_index + 1])) {
       // Invalid surrogate pair.
       return false;
     }

     // Valid surrogate pair.
-    *code_point = CBU16_GET_SUPPLEMENTARY(src[*char_index],
+    *code_point = U16_GET_SUPPLEMENTARY(src[*char_index],
                                           src[*char_index + 1]);
     (*char_index)++;
   } else {
@@ -77,30 +77,30 @@
   }


-  // CBU8_APPEND_UNSAFE can append up to 4 bytes.
+  // U8_APPEND_UNSAFE can append up to 4 bytes.
   size_t char_offset = output->length();
   size_t original_char_offset = char_offset;
-  output->resize(char_offset + CBU8_MAX_LENGTH);
+  output->resize(char_offset + U8_MAX_LENGTH);

-  CBU8_APPEND_UNSAFE(&(*output)[0], char_offset, code_point);
+  U8_APPEND_UNSAFE(&(*output)[0], char_offset, code_point);

-  // CBU8_APPEND_UNSAFE will advance our pointer past the inserted character, so
+  // U8_APPEND_UNSAFE will advance our pointer past the inserted character, so
   // it will represent the new length of the string.
   output->resize(char_offset);
   return char_offset - original_char_offset;
 }

 size_t WriteUnicodeCharacter(uint32_t code_point, std::u16string* output) {
-  if (CBU16_LENGTH(code_point) == 1) {
+  if (U16_LENGTH(code_point) == 1) {
     // Thie code point is in the Basic Multilingual Plane (BMP).
     output->push_back(static_cast<char16_t>(code_point));
     return 1;
   }
   // Non-BMP characters use a double-character encoding.
   size_t char_offset = output->length();
-  output->resize(char_offset + CBU16_MAX_LENGTH);
-  CBU16_APPEND_UNSAFE(&(*output)[0], char_offset, code_point);
-  return CBU16_MAX_LENGTH;
+  output->resize(char_offset + U16_MAX_LENGTH);
+  U16_APPEND_UNSAFE(&(*output)[0], char_offset, code_point);
+  return U16_MAX_LENGTH;
 }

 // Generalized Unicode converter -----------------------------------------------
--- a/base/json/string_escape.cc    2021-08-04 23:24:06.001124181 +0200
+++ b/base/json/string_escape.cc    2021-08-04 23:24:49.897457457 +0200
@@ -92,7 +92,7 @@
   for (int32_t i = 0; i < length; ++i) {
     uint32_t code_point;
     if (!ReadUnicodeCharacter(str.data(), length, &i, &code_point) ||
-        code_point == static_cast<decltype(code_point)>(CBU_SENTINEL) ||
+        code_point == static_cast<decltype(code_point)>(U_SENTINEL) ||
         !IsValidCodepoint(code_point)) {
       code_point = kReplacementCodePoint;
       did_replacement = true;
--- a/base/strings/string_util_internal.h   2021-08-04 23:33:49.036681262 +0200
+++ b/base/strings/string_util_internal.h   2021-08-04 23:35:06.626091464 +0200
@@ -11,7 +11,7 @@
 #include "base/notreached.h"
 #include "base/ranges/algorithm.h"
 #include "base/strings/string_piece.h"
-#include "base/third_party/icu/icu_utf.h"
+#include <unicode/utf.h>

 namespace base {

@@ -233,7 +233,7 @@

   while (char_index < src_len) {
     int32_t code_point;
-    CBU8_NEXT(src, char_index, src_len, code_point);
+    U8_NEXT(src, char_index, src_len, code_point);
     if (!Validator(code_point))
       return false;
   }
--- a/base/strings/escape.cc    2021-08-05 09:17:12.432399855 +0200
+++ b/base/strings/escape.cc    2021-08-05 09:17:54.222088806 +0200
@@ -7,7 +7,7 @@
 #include "base/strings/string_piece.h"
 #include "base/strings/string_util.h"
 #include "base/strings/utf_string_conversion_utils.h"
-#include "base/third_party/icu/icu_utf.h"
+#include <unicode/utf.h>

 namespace base {

@@ -84,14 +84,14 @@
                                   std::string* unescaped_out) {
   DCHECK(unescaped_out->empty());

-  unsigned char bytes[CBU8_MAX_LENGTH];
+  unsigned char bytes[U8_MAX_LENGTH];
   if (!UnescapeUnsignedByteAtIndex(escaped_text, index, &bytes[0]))
     return false;

   size_t num_bytes = 1;

   // If this is a lead byte, need to collect trail bytes as well.
-  if (CBU8_IS_LEAD(bytes[0])) {
+  if (U8_IS_LEAD(bytes[0])) {
     // Look for the last trail byte of the UTF-8 character.  Give up once
     // reach max character length number of bytes, or hit an unescaped
     // character. No need to check length of escaped_text, as
@@ -99,7 +99,7 @@
     while (num_bytes < size(bytes) &&
            UnescapeUnsignedByteAtIndex(escaped_text, index + num_bytes * 3,
                                        &bytes[num_bytes]) &&
-           CBU8_IS_TRAIL(bytes[num_bytes])) {
+           U8_IS_TRAIL(bytes[num_bytes])) {
       ++num_bytes;
     }
   }

--- a/ui/base/ime/character_composer.cc 2021-08-05 10:22:00.819616049 +0200
+++ b/ui/base/ime/character_composer.cc 2021-08-05 10:23:03.695813471 +0200
@@ -13,7 +13,7 @@
 #include "base/strings/string_util.h"
 #include "base/strings/utf_string_conversion_utils.h"
 #include "base/strings/utf_string_conversions.h"
-#include "base/third_party/icu/icu_utf.h"
+#include <unicode/utf.h>
 #include "ui/events/event.h"
 #include "ui/events/keycodes/dom/dom_key.h"
 #include "ui/events/keycodes/dom/keycode_converter.h"
@@ -37,12 +37,12 @@
 bool UTF32CharacterToUTF16(uint32_t character, std::u16string* output) {
   output->clear();
   // Reject invalid character. (e.g. codepoint greater than 0x10ffff)
-  if (!CBU_IS_UNICODE_CHAR(character))
+  if (!U_IS_UNICODE_CHAR(character))
     return false;
   if (character) {
-    output->resize(CBU16_LENGTH(character));
+    output->resize(U16_LENGTH(character));
     size_t i = 0;
-    CBU16_APPEND_UNSAFE(&(*output)[0], i, character);
+    U16_APPEND_UNSAFE(&(*output)[0], i, character);
   }
   return true;
 }
--- a/components/filename_generation/filename_generation.cc 2021-08-05 10:37:39.502739983 +0200
+++ b/components/filename_generation/filename_generation.cc 2021-08-05 10:38:18.405784645 +0200
@@ -11,7 +11,7 @@
 #include "base/strings/string_util.h"
 #include "base/strings/sys_string_conversions.h"
 #include "base/strings/utf_string_conversions.h"
-#include "base/third_party/icu/icu_utf.h"
+#include <unicode/utf.h>
 #include "build/build_config.h"
 #include "build/chromeos_buildflags.h"
 #include "components/url_formatter/url_formatter.h"
@@ -168,7 +168,7 @@
 #elif defined(OS_WIN)
   // UTF-16.
   DCHECK(name.size() > limit);
-  truncated = name.substr(0, CBU16_IS_TRAIL(name[limit]) ? limit - 1 : limit);
+  truncated = name.substr(0, U16_IS_TRAIL(name[limit]) ? limit - 1 : limit);
 #else
 // We cannot generally assume that the file name encoding is in UTF-8 (see
 // the comment for FilePath::AsUTF8Unsafe), hence no safe way to truncate.
--- a/net/cert/internal/parse_name.cc   2021-08-05 10:40:08.241635390 +0200
+++ b/net/cert/internal/parse_name.cc   2021-08-05 10:40:37.068087718 +0200
@@ -11,7 +11,7 @@
 #include "base/strings/utf_string_conversion_utils.h"
 #include "base/strings/utf_string_conversions.h"
 #include "base/sys_byteorder.h"
-#include "base/third_party/icu/icu_utf.h"
+#include <unicode/utf.h>

 namespace net {

@@ -36,7 +36,7 @@

     // BMPString only supports codepoints in the Basic Multilingual Plane;
     // surrogates are not allowed.
-    if (CBU_IS_SURROGATE(c))
+    if (U_IS_SURROGATE(c))
       return false;
   }
   return base::UTF16ToUTF8(in_16bit.data(), in_16bit.size(), out);
@@ -56,7 +56,7 @@
   for (const uint32_t c : in_32bit) {
     // UniversalString is UCS-4 in big-endian order.
     uint32_t codepoint = base::NetToHost32(c);
-    if (!CBU_IS_UNICODE_CHAR(codepoint))
+    if (!U_IS_UNICODE_CHAR(codepoint))
       return false;

     base::WriteUnicodeCharacter(codepoint, out);
--- a/net/base/escape.cc    2021-08-05 10:41:42.997597240 +0200
+++ b/net/base/escape.cc    2021-08-05 10:42:10.937389282 +0200
@@ -11,7 +11,7 @@
 #include "base/strings/string_util.h"
 #include "base/strings/utf_string_conversion_utils.h"
 #include "base/strings/utf_string_conversions.h"
-#include "base/third_party/icu/icu_utf.h"
+#include <unicode/utf.h>
 #include "build/build_config.h"

 namespace net {
--- a/tools/gn/src/base/third_party/icu/icu_utf.cc  2021-08-05 10:43:07.293636317 +0200
+++ b/tools/gn/src/base/third_party/icu/icu_utf.cc  2021-08-05 10:43:58.736586415 +0200
@@ -19,7 +19,7 @@
 *   that would otherwise be too long as macros.
 */

-#include "base/third_party/icu/icu_utf.h"
+#include <unicode/utf.h>

 namespace base_icu {

@@ -36,7 +36,7 @@
   } else if (strict == -3) {
     return 0xfffd;
   } else {
-    return CBU_SENTINEL;
+    return U_SENTINEL;
   }
 }

@@ -78,13 +78,13 @@
     // U8_NEXT() handles shorter valid sequences inline.
     uint8_t t1 = s[i], t2, t3;
     c &= 7;
-    if (CBU8_IS_VALID_LEAD4_AND_T1(c, t1) && ++i != length &&
+    if (U8_IS_VALID_LEAD4_AND_T1(c, t1) && ++i != length &&
         (t2 = s[i] - 0x80) <= 0x3f && ++i != length &&
         (t3 = s[i] - 0x80) <= 0x3f) {
       ++i;
       c = (c << 18) | ((t1 & 0x3f) << 12) | (t2 << 6) | t3;
       // strict: forbid non-characters like U+fffe
-      if (strict <= 0 || !CBU_IS_UNICODE_NONCHAR(c)) {
+      if (strict <= 0 || !U_IS_UNICODE_NONCHAR(c)) {
         *pi = i;
         return c;
       }
@@ -93,12 +93,12 @@
     c &= 0xf;
     if (strict != -2) {
       uint8_t t1 = s[i], t2;
-      if (CBU8_IS_VALID_LEAD3_AND_T1(c, t1) && ++i != length &&
+      if (U8_IS_VALID_LEAD3_AND_T1(c, t1) && ++i != length &&
           (t2 = s[i] - 0x80) <= 0x3f) {
         ++i;
         c = (c << 12) | ((t1 & 0x3f) << 6) | t2;
         // strict: forbid non-characters like U+fffe
-        if (strict <= 0 || !CBU_IS_UNICODE_NONCHAR(c)) {
+        if (strict <= 0 || !U_IS_UNICODE_NONCHAR(c)) {
           *pi = i;
           return c;
         }
--- a/tools/gn/src/base/files/file_path.cc  2021-08-05 10:45:22.262630525 +0200
+++ b/tools/gn/src/base/files/file_path.cc  2021-08-05 10:46:05.652306974 +0200
@@ -18,7 +18,7 @@

 #if defined(OS_MACOSX)
 #include "base/mac/scoped_cftyperef.h"
-#include "base/third_party/icu/icu_utf.h"
+#include <unicode/utf.h>
 #endif

 #if defined(OS_WIN)
--- a/tools/gn/src/base/strings/utf_string_conversion_utils.cc  2021-08-05 10:47:09.295165556 +0200
+++ b/tools/gn/src/base/strings/utf_string_conversion_utils.cc  2021-08-05 10:47:52.798174297 +0200
@@ -4,7 +4,7 @@

 #include "base/strings/utf_string_conversion_utils.h"

-#include "base/third_party/icu/icu_utf.h"
+#include <unicode/utf.h>
 #include "util/build_config.h"

 namespace base {
@@ -19,7 +19,7 @@
   // use a signed type for code_point.  But this function returns false
   // on error anyway, so code_point_out is unsigned.
   int32_t code_point;
-  CBU8_NEXT(src, *char_index, src_len, code_point);
+  U8_NEXT(src, *char_index, src_len, code_point);
   *code_point_out = static_cast<uint32_t>(code_point);

   // The ICU macro above moves to the next char, we want to point to the last
@@ -34,16 +34,16 @@
                           int32_t src_len,
                           int32_t* char_index,
                           uint32_t* code_point) {
-  if (CBU16_IS_SURROGATE(src[*char_index])) {
-    if (!CBU16_IS_SURROGATE_LEAD(src[*char_index]) ||
-        *char_index + 1 >= src_len || !CBU16_IS_TRAIL(src[*char_index + 1])) {
+  if (U16_IS_SURROGATE(src[*char_index])) {
+    if (!U16_IS_SURROGATE_LEAD(src[*char_index]) ||
+        *char_index + 1 >= src_len || !U16_IS_TRAIL(src[*char_index + 1])) {
       // Invalid surrogate pair.
       return false;
     }

     // Valid surrogate pair.
     *code_point =
-        CBU16_GET_SUPPLEMENTARY(src[*char_index], src[*char_index + 1]);
+        U16_GET_SUPPLEMENTARY(src[*char_index], src[*char_index + 1]);
     (*char_index)++;
   } else {
     // Not a surrogate, just one 16-bit word.
@@ -62,30 +62,30 @@
     return 1;
   }

-  // CBU8_APPEND_UNSAFE can append up to 4 bytes.
+  // U8_APPEND_UNSAFE can append up to 4 bytes.
   size_t char_offset = output->length();
   size_t original_char_offset = char_offset;
-  output->resize(char_offset + CBU8_MAX_LENGTH);
+  output->resize(char_offset + U8_MAX_LENGTH);

-  CBU8_APPEND_UNSAFE(&(*output)[0], char_offset, code_point);
+  U8_APPEND_UNSAFE(&(*output)[0], char_offset, code_point);

-  // CBU8_APPEND_UNSAFE will advance our pointer past the inserted character, so
+  // U8_APPEND_UNSAFE will advance our pointer past the inserted character, so
   // it will represent the new length of the string.
   output->resize(char_offset);
   return char_offset - original_char_offset;
 }

 size_t WriteUnicodeCharacter(uint32_t code_point, std::u16string* output) {
-  if (CBU16_LENGTH(code_point) == 1) {
+  if (U16_LENGTH(code_point) == 1) {
     // Thie code point is in the Basic Multilingual Plane (BMP).
     output->push_back(static_cast<char16_t>(code_point));
     return 1;
   }
   // Non-BMP characters use a double-character encoding.
   size_t char_offset = output->length();
-  output->resize(char_offset + CBU16_MAX_LENGTH);
-  CBU16_APPEND_UNSAFE(&(*output)[0], char_offset, code_point);
-  return CBU16_MAX_LENGTH;
+  output->resize(char_offset + U16_MAX_LENGTH);
+  U16_APPEND_UNSAFE(&(*output)[0], char_offset, code_point);
+  return U16_MAX_LENGTH;
 }

 // Generalized Unicode converter -----------------------------------------------
--- a/tools/gn/src/base/strings/string_util.cc  2021-08-05 10:48:59.417677129 +0200
+++ b/tools/gn/src/base/strings/string_util.cc  2021-08-05 10:49:36.237402265 +0200
@@ -25,7 +25,7 @@
 #include "base/macros.h"
 #include "base/strings/utf_string_conversion_utils.h"
 #include "base/strings/utf_string_conversions.h"
-#include "base/third_party/icu/icu_utf.h"
+#include <unicode/utf.h>
 #include "util/build_config.h"

 namespace base {
@@ -297,19 +297,19 @@
   }
   DCHECK_LE(byte_size,
             static_cast<uint32_t>(std::numeric_limits<int32_t>::max()));
-  // Note: This cast is necessary because CBU8_NEXT uses int32_ts.
+  // Note: This cast is necessary because U8_NEXT uses int32_ts.
   int32_t truncation_length = static_cast<int32_t>(byte_size);
   int32_t char_index = truncation_length - 1;
   const char* data = input.data();

-  // Using CBU8, we will move backwards from the truncation point
+  // Using U8, we will move backwards from the truncation point
   // to the beginning of the string looking for a valid UTF8
   // character.  Once a full UTF8 character is found, we will
   // truncate the string to the end of that character.
   while (char_index >= 0) {
     int32_t prev = char_index;
-    base_icu::UChar32 code_point = 0;
-    CBU8_NEXT(data, char_index, truncation_length, code_point);
+    UChar32 code_point = 0;
+    U8_NEXT(data, char_index, truncation_length, code_point);
     if (!IsValidCharacter(code_point) || !IsValidCodepoint(code_point)) {
       char_index = prev - 1;
     } else {
@@ -453,7 +453,7 @@

   while (char_index < src_len) {
     int32_t code_point;
-    CBU8_NEXT(src, char_index, src_len, code_point);
+    U8_NEXT(src, char_index, src_len, code_point);
     if (!IsValidCharacter(code_point))
       return false;
   }
--- a/tools/gn/src/base/strings/string_split.cc 2021-08-05 10:51:54.316371187 +0200
+++ b/tools/gn/src/base/strings/string_split.cc 2021-08-05 10:52:21.186170479 +0200
@@ -8,7 +8,7 @@

 #include "base/logging.h"
 #include "base/strings/string_util.h"
-#include "base/third_party/icu/icu_utf.h"
+#include <unicode/utf.h>

 namespace base {

--- a/tools/gn/src/base/json/string_escape.cc   2021-08-05 10:53:00.515876682 +0200
+++ b/tools/gn/src/base/json/string_escape.cc   2021-08-05 10:53:32.152306995 +0200
@@ -14,7 +14,7 @@
 #include "base/strings/stringprintf.h"
 #include "base/strings/utf_string_conversion_utils.h"
 #include "base/strings/utf_string_conversions.h"
-#include "base/third_party/icu/icu_utf.h"
+#include <unicode/utf.h>

 namespace base {

@@ -92,7 +92,7 @@
   for (int32_t i = 0; i < length; ++i) {
     uint32_t code_point;
     if (!ReadUnicodeCharacter(str.data(), length, &i, &code_point) ||
-        code_point == static_cast<decltype(code_point)>(CBU_SENTINEL) ||
+        code_point == static_cast<decltype(code_point)>(U_SENTINEL) ||
         !IsValidCharacter(code_point)) {
       code_point = kReplacementCodePoint;
       did_replacement = true;
--- a/tools/gn/src/base/json/json_parser.cc 2021-08-05 10:54:43.705105681 +0200
+++ b/tools/gn/src/base/json/json_parser.cc 2021-08-05 10:55:16.498193950 +0200
@@ -17,7 +17,7 @@
 #include "base/strings/stringprintf.h"
 #include "base/strings/utf_string_conversion_utils.h"
 #include "base/strings/utf_string_conversions.h"
-#include "base/third_party/icu/icu_utf.h"
+#include <unicode/utf.h>
 #include "base/values.h"

 namespace base {
@@ -569,10 +569,10 @@

   // If this is a high surrogate, consume the next code unit to get the
   // low surrogate.
-  if (CBU16_IS_SURROGATE(code_unit16_high)) {
+  if (U16_IS_SURROGATE(code_unit16_high)) {
     // Make sure this is the high surrogate. If not, it's an encoding
     // error.
-    if (!CBU16_IS_SURROGATE_LEAD(code_unit16_high))
+    if (!U16_IS_SURROGATE_LEAD(code_unit16_high))
       return false;

     // Make sure that the token has more characters to consume the
@@ -588,18 +588,18 @@
     if (!HexStringToInt(*escape_sequence, &code_unit16_low))
       return false;

-    if (!CBU16_IS_TRAIL(code_unit16_low))
+    if (!U16_IS_TRAIL(code_unit16_low))
       return false;

     uint32_t code_point =
-        CBU16_GET_SUPPLEMENTARY(code_unit16_high, code_unit16_low);
+        U16_GET_SUPPLEMENTARY(code_unit16_high, code_unit16_low);
     if (!IsValidCharacter(code_point))
       return false;

     *out_code_point = code_point;
   } else {
     // Not a surrogate.
-    DCHECK(CBU16_IS_SINGLE(code_unit16_high));
+    DCHECK(U16_IS_SINGLE(code_unit16_high));
     if (!IsValidCharacter(code_unit16_high)) {
       if ((options_ & JSON_REPLACE_INVALID_CHARACTERS) == 0) {
         return false;

--- a/tools/cfi/ignores.txt 2021-08-26 16:07:35.744982216 +0200
+++ b/tools/cfi/ignores.txt 2021-08-26 16:08:04.958072127 +0200
@@ -124,6 +124,7 @@
 type:sandbox::BrokerServices
 type:sandbox::TargetPolicy
 type:sandbox::TargetServices
+type:icu_60::*

 #############################################################################
 # Disabled indirect calls

Of course, you will then also have to include icu in the depends() and _use_system() arrays in PKGBUILD.

Additionally, the compile flags have to be amended (best place will be right below the _use_system() declaration):

  if [[ "${_use_system[*]}" =~ "icu" ]]; then
    _flags+=('icu_use_data_file=false')
  fi

DAC324 commented on 2022-02-04 16:16 (UTC) (edited on 2022-02-04 18:14 (UTC) by DAC324)

Works well with 99.0.4844.9 (adjusted pkgver() in PKGBUILD).

What I found in the Chromium developer forums: To speed up the compile process and to streamline the package, you can omit certain build options by setting them to false:

# Most developers don't normally need to test Native Client capabilities and can speed up the build by disabling it.
   'enable_nacl=false'
   'enable_nacl_nonsfi=false'
# not everybody needs these
   'enable_hangout_services_extension=false'
   'enable_widevine=false'

Of course, if the Native Client is not built, the following section in PKGBUILD has to be commented out:

  # Install NaCL.
  _nacl_libs=(
             # 'nacl_helper'
             # 'nacl_helper_bootstrap'
             # 'nacl_irt_x86_64.nexe'
              )
  for i in "${_nacl_libs[@]}"; do
    install -Dm755 "${i}" "${pkgdir}/usr/lib/chromium-dev/${i}"
    case "$i" in
      nacl_irt_x86_64.nexe)
        ;;
      *)
        strip $STRIP_BINARIES "${pkgdir}/usr/lib/chromium-dev/${i}"
        ;;
    esac
  done

I am currently running 99.0.4844.0 without the Native Client and could not observe any problems so far.

Finally, it seems to make sense to set an alias like this

alias touch='echo -n >>'

before the Build all command (LC_ALL=C ninja -C out/Release -v chrome chrome_sandbox chromedriver). This also speeds up things a little bit as calling touch as an external command takes much longer than using the echo internal command.

sl1pkn07 commented on 2022-01-27 17:09 (UTC)

new version (99.0.4840.0) is in build process right now

chazy commented on 2022-01-27 17:03 (UTC)

Looks like gentoo removed a patch file:

https://raw.githubusercontent.com/gentoo/gentoo/master/www-client/chromium/files/chromium-97-fix-tag-dragging-i3.patch

Compiling works if you replace the url with one pointing to a previous version of the repository (where the file still exists).

Whether or not the patch is still needed, that's up to the maintainer to determine (there's a reason why gentoo decided to delete the patch).

alex19EP commented on 2022-01-07 01:07 (UTC)

fails to build if lto is enabled. to fix I advise to synchronize with chromium from extra

DAC324 commented on 2021-11-02 17:17 (UTC) (edited on 2021-11-02 17:18 (UTC) by DAC324)

Hardware accelerated video decoding using vaapi currently does not seem to work (see log from chromium-dev, on the chrome://gpu page):

[4210:4961:1102/135750.651953:FATAL:vaapi_wrapper.cc(2358)] : Check failed: sequence_checker_.CalledOnValidSequence(). 
#0 0x7f8c160035fe base::debug::CollectStackTrace() 
#1 0x7f8c15f30efb base::debug::StackTrace::StackTrace() 
#2 0x7f8c15f4c0ea logging::LogMessage::~LogMessage() 
#3 0x7f8c15f4cc85 logging::LogMessage::~LogMessage() 
#4 0x7f8c064c84fa media::VaapiWrapper::SubmitBuffers() 
#5 0x7f8c064a038f (/usr/lib/chromium-dev/libservice.so+0x23238e) 
#6 0x7f8c064d4fa4 media::H264Decoder::StartNewFrame() 
#7 0x7f8c064d7904 media::H264Decoder::Decode() 
#8 0x7f8c064aa50f media::VaapiVideoDecodeAccelerator::DecodeTask() 
#9 0x7f8c15fb30bc base::TaskAnnotator::RunTask() 
#10 0x7f8c15fc95be base::sequence_manager::internal::ThreadControllerWithMessagePumpImpl::DoWorkImpl() 
#11 0x7f8c15fc92fc base::sequence_manager::internal::ThreadControllerWithMessagePumpImpl::DoWork() 
#12 0x7f8c15fc9989 base::sequence_manager::internal::ThreadControllerWithMessagePumpImpl::DoWork() 
#13 0x7f8c15f569db base::MessagePumpDefault::Run() 
#14 0x7f8c15fc9c26 base::sequence_manager::internal::ThreadControllerWithMessagePumpImpl::Run() 
#15 0x7f8c15f89061 base::RunLoop::Run() 
#16 0x7f8c15fea29b base::Thread::Run() 
#17 0x7f8c15fea451 base::Thread::ThreadMain() 
#18 0x7f8c1601894c (/usr/lib/chromium-dev/libbase.so+0x22294b) 
#19 0x7f8c0700b259 start_thread 
#20 0x7f8c069f25e3 __GI___clone Task trace: 
#0 0x7f8c064ac67e media::VaapiVideoDecodeAccelerator::AssignPictureBuffers() 
#1 0x7f8c0d82075a (/usr/lib/chromium-dev/libmedia_mojo_services.so+0x9f759) 
#2 0x7f8c064ab814 media::VaapiVideoDecodeAccelerator::TryFinishSurfaceSetChange() 
#3 0x7f8c064aa80b media::VaapiVideoDecodeAccelerator::DecodeTask() 
#4 0x7f8c064aa01b media::VaapiVideoDecodeAccelerator::QueueInputBuffer() Task trace buffer limit hit, update PendingTask::kTaskBacktraceLength to increase. Crash keys: "last-video-decoder" = "name=VDAVideoDecoder:codec=h264:profile=3:size=1280x720:cs=[1,1,2,1]:hdrm=0" "num-video-decoders" = "1"
GpuProcessHost: The GPU process crashed!

DAC324 commented on 2021-09-28 10:13 (UTC) (edited on 2021-09-28 11:24 (UTC) by DAC324)

This package does not compile if -march is set to something different than -march=x86-64 in CCFLAGS in /etc/makepkg.conf. If you encounter errors like below, you will have to check if that flag is set to the generic -march=x86-64.

[3293/56075] /var/tmp/pamac-build-gerd/chromium-dev/src/chromium-96.0.4651.0/third_party/llvm-build/Release+Asserts/bin/clang++ -MMD -MF obj/skia/skia_core_and_effects/SkOpts.o.d -DUSE_UDEV -DUSE_AURA=1 -DUSE_GLIB=1 -DUSE_NSS_CERTS=1 -DUSE_OZONE=1 -DUSE_X11=1 -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_GNU_SOURCE -DCR_CLANG_REVISION=\"llvmorg-14-init-3940-gafc45ff0-2\" -DCOMPONENT_BUILD -D_LIBCPP_ABI_UNSTABLE -D_LIBCPP_ABI_VERSION=Cr -D_LIBCPP_ENABLE_NODISCARD -DCR_LIBCXX_REVISION=79a2e924d96e2fc1e4b937c42efd08898fa472d7 -DNDEBUG -DNVALGRIND -DDYNAMIC_ANNOTATIONS_ENABLED=0 -DSK_UNTIL_CRBUG_1187654_IS_FIXED -DSK_CODEC_DECODES_PNG -DSK_CODEC_DECODES_WEBP -DSK_ENCODE_PNG -DSK_ENCODE_WEBP -DSK_ENABLE_SKSL -DSK_USER_CONFIG_HEADER=\"../../skia/config/SkUserConfig.h\" -DSK_GL -DSK_CODEC_DECODES_JPEG -DSK_ENCODE_JPEG -DSK_HAS_WUFFS_LIBRARY -DSK_VULKAN_HEADER=\"../../skia/config/SkVulkanConfig.h\" -DSK_VULKAN=1 -DSKIA_DLL -DSKCMS_API=__attribute__\(\(visibility\(\"default\"\)\)\) -DSK_SUPPORT_GPU=1 -DSK_GPU_WORKAROUNDS_HEADER=\"gpu/config/gpu_driver_bug_workaround_autogen.h\" -DVK_USE_PLATFORM_XCB_KHR -DSKIA_IMPLEMENTATION=1 -DSK_GAMMA_EXPONENT=1.2 -DSK_GAMMA_CONTRAST=0.2 -DSK_DEFAULT_FONT_CACHE_LIMIT=20971520 -I../.. -Igen -I../../buildtools/third_party/libc++ -I../../third_party/skia -I../../third_party/wuffs/src/release/c -I../../third_party/vulkan/include -I../../third_party/vulkan-deps/vulkan-headers/src/include -fno-delete-null-pointer-checks -fno-ident -fno-strict-aliasing --param=ssp-buffer-size=4 -fstack-protector -funwind-tables -fPIC -pthread -fcolor-diagnostics -fmerge-all-constants -fcrash-diagnostics-dir=../../tools/clang/crashreports -mllvm -instcombine-lower-dbg-declare=0 -ffp-contract=off -fcomplete-member-pointers -m64 -march=x86-64 -msse3 -Wno-builtin-macro-redefined -D__DATE__= -D__TIME__= -D__TIMESTAMP__= -ffile-compilation-dir=. -no-canonical-prefixes -fno-omit-frame-pointer -g0 -ftrivial-auto-var-init=pattern -fvisibility=hidden -Xclang -add-plugin -Xclang find-bad-constructs -Xclang -plugin-arg-find-bad-constructs -Xclang raw-ptr-template-as-trivial-member -Xclang -plugin-arg-find-bad-constructs -Xclang check-ipc -Wheader-hygiene -Wstring-conversion -Wtautological-overlap-compare -O2 -fdata-sections -ffunction-sections -fno-unique-section-names -Wall -Wno-unused-variable -Wno-c++11-narrowing -Wno-misleading-indentation -Wno-missing-field-initializers -Wno-unused-parameter -Wloop-analysis -Wno-unneeded-internal-declaration -Wenum-compare-conditional -Wno-psabi -Wno-ignored-pragma-optimize -Wno-builtin-assume-aligned-alignment -Wno-unused-but-set-parameter -Wno-unused-but-set-variable -std=c++17 -std=c++14 -fno-trigraphs -Wno-trigraphs -fno-exceptions -fno-rtti -nostdinc++ -isystem../../buildtools/third_party/libc++/trunk/include -isystem../../buildtools/third_party/libc++abi/trunk/include -fvisibility-inlines-hidden -std=c++17 -march=native -mtune=native -O2 -pipe -fstack-protector-strong -fno-plt -fexceptions -Wp,-D_FORTIFY_SOURCE=2 -Wformat -Werror=format-security -fstack-clash-protection -fcf-protection -I/usr/include -D__DATE__=  -D__TIME__=  -D__TIMESTAMP__= -march=native -mtune=native -O2 -pipe -fstack-protector-strong -fno-plt  -Wp,-D_FORTIFY_SOURCE=2 -Wformat -Werror=format-security -fstack-clash-protection -fcf-protection -I/usr/include -Wp,-D_GLIBCXX_ASSERTIONS -Wno-builtin-macro-redefined -c ../../third_party/skia/src/core/SkOpts.cpp -o obj/skia/skia_core_and_effects/SkOpts.o
FAILED: obj/skia/skia_core_and_effects/SkOpts.o 
/var/tmp/pamac-build-gerd/chromium-dev/src/chromium-96.0.4651.0/third_party/llvm-build/Release+Asserts/bin/clang++ -MMD -MF obj/skia/skia_core_and_effects/SkOpts.o.d -DUSE_UDEV -DUSE_AURA=1 -DUSE_GLIB=1 -DUSE_NSS_CERTS=1 -DUSE_OZONE=1 -DUSE_X11=1 -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_GNU_SOURCE -DCR_CLANG_REVISION=\"llvmorg-14-init-3940-gafc45ff0-2\" -DCOMPONENT_BUILD -D_LIBCPP_ABI_UNSTABLE -D_LIBCPP_ABI_VERSION=Cr -D_LIBCPP_ENABLE_NODISCARD -DCR_LIBCXX_REVISION=79a2e924d96e2fc1e4b937c42efd08898fa472d7 -DNDEBUG -DNVALGRIND -DDYNAMIC_ANNOTATIONS_ENABLED=0 -DSK_UNTIL_CRBUG_1187654_IS_FIXED -DSK_CODEC_DECODES_PNG -DSK_CODEC_DECODES_WEBP -DSK_ENCODE_PNG -DSK_ENCODE_WEBP -DSK_ENABLE_SKSL -DSK_USER_CONFIG_HEADER=\"../../skia/config/SkUserConfig.h\" -DSK_GL -DSK_CODEC_DECODES_JPEG -DSK_ENCODE_JPEG -DSK_HAS_WUFFS_LIBRARY -DSK_VULKAN_HEADER=\"../../skia/config/SkVulkanConfig.h\" -DSK_VULKAN=1 -DSKIA_DLL -DSKCMS_API=__attribute__\(\(visibility\(\"default\"\)\)\) -DSK_SUPPORT_GPU=1 -DSK_GPU_WORKAROUNDS_HEADER=\"gpu/config/gpu_driver_bug_workaround_autogen.h\" -DVK_USE_PLATFORM_XCB_KHR -DSKIA_IMPLEMENTATION=1 -DSK_GAMMA_EXPONENT=1.2 -DSK_GAMMA_CONTRAST=0.2 -DSK_DEFAULT_FONT_CACHE_LIMIT=20971520 -I../.. -Igen -I../../buildtools/third_party/libc++ -I../../third_party/skia -I../../third_party/wuffs/src/release/c -I../../third_party/vulkan/include -I../../third_party/vulkan-deps/vulkan-headers/src/include -fno-delete-null-pointer-checks -fno-ident -fno-strict-aliasing --param=ssp-buffer-size=4 -fstack-protector -funwind-tables -fPIC -pthread -fcolor-diagnostics -fmerge-all-constants -fcrash-diagnostics-dir=../../tools/clang/crashreports -mllvm -instcombine-lower-dbg-declare=0 -ffp-contract=off -fcomplete-member-pointers -m64 -march=x86-64 -msse3 -Wno-builtin-macro-redefined -D__DATE__= -D__TIME__= -D__TIMESTAMP__= -ffile-compilation-dir=. -no-canonical-prefixes -fno-omit-frame-pointer -g0 -ftrivial-auto-var-init=pattern -fvisibility=hidden -Xclang -add-plugin -Xclang find-bad-constructs -Xclang -plugin-arg-find-bad-constructs -Xclang raw-ptr-template-as-trivial-member -Xclang -plugin-arg-find-bad-constructs -Xclang check-ipc -Wheader-hygiene -Wstring-conversion -Wtautological-overlap-compare -O2 -fdata-sections -ffunction-sections -fno-unique-section-names -Wall -Wno-unused-variable -Wno-c++11-narrowing -Wno-misleading-indentation -Wno-missing-field-initializers -Wno-unused-parameter -Wloop-analysis -Wno-unneeded-internal-declaration -Wenum-compare-conditional -Wno-psabi -Wno-ignored-pragma-optimize -Wno-builtin-assume-aligned-alignment -Wno-unused-but-set-parameter -Wno-unused-but-set-variable -std=c++17 -std=c++14 -fno-trigraphs -Wno-trigraphs -fno-exceptions -fno-rtti -nostdinc++ -isystem../../buildtools/third_party/libc++/trunk/include -isystem../../buildtools/third_party/libc++abi/trunk/include -fvisibility-inlines-hidden -std=c++17 -march=native -mtune=native -O2 -pipe -fstack-protector-strong -fno-plt -fexceptions -Wp,-D_FORTIFY_SOURCE=2 -Wformat -Werror=format-security -fstack-clash-protection -fcf-protection -I/usr/include -D__DATE__=  -D__TIME__=  -D__TIMESTAMP__= -march=native -mtune=native -O2 -pipe -fstack-protector-strong -fno-plt  -Wp,-D_FORTIFY_SOURCE=2 -Wformat -Werror=format-security -fstack-clash-protection -fcf-protection -I/usr/include -Wp,-D_GLIBCXX_ASSERTIONS -Wno-builtin-macro-redefined -c ../../third_party/skia/src/core/SkOpts.cpp -o obj/skia/skia_core_and_effects/SkOpts.o
In file included from ../../third_party/skia/src/core/SkOpts.cpp:46:
../../third_party/skia/src/opts/SkRasterPipeline_opts.h:3137:44: error: cannot initialize a parameter of type 'avx::F' (aka 'V<float>') with an lvalue of type '__m128' (vector of 4 'float' values)
    return join<F>(SK_OPTS_NS::rcp_precise(lo), SK_OPTS_NS::rcp_precise(hi));
                                           ^~
../../third_party/skia/src/opts/SkRasterPipeline_opts.h:369:25: note: passing argument to parameter 'v' here
    SI F rcp_precise (F v) {
                        ^
../../third_party/skia/src/opts/SkRasterPipeline_opts.h:3137:73: error: cannot initialize a parameter of type 'avx::F' (aka 'V<float>') with an lvalue of type '__m128' (vector of 4 'float' values)
    return join<F>(SK_OPTS_NS::rcp_precise(lo), SK_OPTS_NS::rcp_precise(hi));
                                                                        ^~
../../third_party/skia/src/opts/SkRasterPipeline_opts.h:369:25: note: passing argument to parameter 'v' here
    SI F rcp_precise (F v) {
                        ^
2 errors generated.

gnaggnoyil commented on 2021-07-28 14:09 (UTC) (edited on 2021-07-28 17:40 (UTC) by gnaggnoyil)

Official Gentoo ebuild repository has updated filename of their patches in www-client/chromium/files. Please update corresponding src items in PKGBUILD.

DAC324 commented on 2021-04-26 12:32 (UTC)

Starting with v91, chromium-dev does not display the default font correctly in the UI and in web pages. In order to circumvent that, chromium-dev has to be started with the --no-sandbox -test-type command line options.

Already filed an issue (https://bugs.chromium.org/p/chromium/issues/detail?id=1193445) but it looks like there is no solution yet.

DAC324 commented on 2021-04-26 09:07 (UTC) (edited on 2021-04-28 13:25 (UTC) by DAC324)

If anybody wants to use Chromium Dev in the version 92.0.4487.7 - here's a patch to PKGBUILD to enable this version:

--- PKGBUILD_91.0.4472.10   2021-04-26 10:55:51.019695028 +0200
+++ PKGBUILD    2021-04-28 15:08:05.278179868 +0200
@@ -15,7 +15,7 @@
 ## -- Package and components information -- ##
 ##############################################
 pkgname=chromium-dev
-pkgver=91.0.4472.10
+pkgver=92.0.4484.7
 pkgrel=1
 pkgdesc="The open-source project behind Google Chrome (Dev Channel)"
 arch=('x86_64')
@@ -28,7 +28,7 @@
 #          'minizip'
          'nss'
          'pciutils'
-#          're2'
+         're2'
          'snappy'
          'xdg-utils'
 #          'protobuf'
@@ -76,17 +76,15 @@
 source=(
         #"https://gsdview.appspot.com/chromium-browser-official/chromium-${pkgver}.tar.xz"
         "https://commondatastorage.googleapis.com/chromium-browser-official/chromium-${pkgver}.tar.xz"
+        # "chromium-browser-official/chromium-${pkgver}.tar.xz"
         'git+https://github.com/foutrelis/chromium-launcher.git'
         'chromium-dev.svg'
-        # Patch form Gentoo.
+        # Patch from Gentoo.
         'https://raw.githubusercontent.com/gentoo/gentoo/master/www-client/chromium/files/chromium-89-EnumTable-crash.patch'
         'https://raw.githubusercontent.com/gentoo/gentoo/master/www-client/chromium/files/chromium-shim_headers.patch'
-        'https://raw.githubusercontent.com/gentoo/gentoo/master/www-client/chromium/files/chromium-91-system-icu.patch'
-        'https://raw.githubusercontent.com/gentoo/gentoo/master/www-client/chromium/files/chromium-91-ThemeService-crash.patch'
         # Misc Patches.
         'fix_hevc_in_non_cromeos_r1.patch'
         # Patch from crbug.com (chromium bugtracker), chromium-review.googlesource.com / Gerrit or Arch chromium package.
-        'https://raw.githubusercontent.com/archlinux/svntogit-packages/packages/chromium/trunk/chromium-glibc-2.33.patch'
         'https://raw.githubusercontent.com/archlinux/svntogit-packages/packages/chromium/trunk/use-oauth2-client-switches-as-default.patch'
         )
 sha256sums=(
@@ -94,15 +92,13 @@
             "$(curl -sL https://commondatastorage.googleapis.com/chromium-browser-official/chromium-${pkgver}.tar.xz.hashes | grep sha256 | cut -d ' ' -f3)"
             'SKIP'
             'dd2b5c4191e468972b5ea8ddb4fa2e2fa3c2c94c79fc06645d0efc0e63ce7ee1'
-            # Patch form Gentoo
+            # Patch from Gentoo
             '61066e638ee795e888d12cc2c4af42c8cf3fe9f473b1dc0b78c9e87faf5eacda'
             'fabf66cfb15449011a20e377d600573b6338cc4c52e3f28f80e0541772659e8b'
-            '1276018471004688356ef70be708ce96a816ffca6625ed7cc99cb985b6e15559'
-            '26cc77232d06e55b50a253617fb87218b0058ba2b0271fef5845bcc474f6a117'
             # Misc Patches
-            '70f7a1736aabd63bbdf5060c015b220ede7d5d19837408eb4a5172099aecd862'
+             '70f7a1736aabd63bbdf5060c015b220ede7d5d19837408eb4a5172099aecd862'
+            # 9d685d75c9384ebd4a0976796db607b36d8500c97fb5f1fb572b0356a3b39987
             # Patch from crbug (chromium bugtracker) or Arch chromium package
-            '2fccecdcd4509d4c36af873988ca9dbcba7fdb95122894a9fdf502c33a1d7a4b'
             'e393174d7695d0bafed69e868c5fbfecf07aa6969f3b64596d0bae8b067e1711'
             )
 install=chromium-dev.install
@@ -119,7 +115,7 @@
 #_google_default_client_id="413772536636.apps.googleusercontent.com"
 #_google_default_client_secret="0ZChLK6AxeA3Isu96MkwqDR4"

-# List of third-party components needed for build chromium. The rest is remove by remove_bundled_libraries srcipt in prepare().
+# List of third-party components needed for build chromium. The rest is removed by remove_bundled_libraries srcipt in prepare().
 _keeplibs=(
            'base/third_party/cityhash'
            'base/third_party/double_conversion'
@@ -183,6 +179,7 @@
            'third_party/dav1d'
            'third_party/dawn'
            'third_party/dawn/third_party/khronos'
+           'third_party/dawn/third_party/tint'  # new in Chromium 92
            'third_party/depot_tools'
            'third_party/devscripts'
            'third_party/devtools-frontend'
@@ -190,6 +187,7 @@
            'third_party/devtools-frontend/src/front_end/third_party/axe-core'
            'third_party/devtools-frontend/src/front_end/third_party/chromium'
            'third_party/devtools-frontend/src/front_end/third_party/codemirror'
+           'third_party/devtools-frontend/src/front_end/third_party/diff' # new in Chromium 92
            'third_party/devtools-frontend/src/front_end/third_party/fabricjs'
            'third_party/devtools-frontend/src/front_end/third_party/i18n'
            'third_party/devtools-frontend/src/front_end/third_party/intl-messageformat'
@@ -198,8 +196,10 @@
            'third_party/devtools-frontend/src/front_end/third_party/lodash-isequal'
            'third_party/devtools-frontend/src/front_end/third_party/marked'
            'third_party/devtools-frontend/src/front_end/third_party/puppeteer'
+           'third_party/devtools-frontend/src/third_party/typescript' # new in Chromium 92
            'third_party/devtools-frontend/src/front_end/third_party/wasmparser'
            'third_party/devtools-frontend/src/third_party'
+           'third_party/devtools-frontend/src/inspector_overlay' # new in Chromium 92
            'third_party/dom_distiller_js'
            'third_party/eigen3'
            'third_party/emoji-segmenter'
@@ -317,7 +317,7 @@
            'third_party/tflite/src/third_party/fft2d'
            'third_party/tflite-support'
            'third_party/tcmalloc'
-           'third_party/tint'
+#           'third_party/tint'
            'third_party/ruy'
            'third_party/ukey2'
            'third_party/unrar'
@@ -356,19 +356,26 @@
            'third_party/speech-dispatcher'
            'third_party/usb_ids'
            'third_party/xdg-utils'
+#           'third_party/xstream'
            )

 _keeplibs+=(
             'third_party/icu' # https://crbug.com/678661.
             )

+if [ "${_use_wayland}" = "1" ]; then
+  _keeplibs+=(
+              'third_party/minigbm'
+             )
+fi
+
 # Set build flags.
 _flags=(
         "custom_toolchain=\"//build/toolchain/linux/unbundle:default\""
         "host_toolchain=\"//build/toolchain/linux/unbundle:default\""
+        "target_os=\"linux\""
         'is_debug=false'
         'is_official_build=false'
-        'chrome_pgo_phase=0' # unsupported instrumentation profile format version
         'is_component_build=true'
         'enable_widevine=true'
         'enable_hangout_services_extension=true'
@@ -387,7 +394,10 @@
         'treat_warnings_as_errors=false'
         'enable_nacl=true'
         'enable_nacl_nonsfi=true'
-        'use_custom_libcxx=true' # use true if you want use bundled RE2
+# Most developers don't normally need to test Native Client capabilities and can speed up the build by disabling it.
+#        'enable_nacl=false'
+#        'enable_nacl_nonsfi=false'
+        'use_custom_libcxx=false' # use true if you want use bundled RE2
         'use_vaapi=true'
         'enable_platform_hevc=true'
         'enable_platform_ac3_eac3_audio=true'
@@ -406,13 +416,13 @@
 if [ "${_use_wayland}" = "1" ]; then
   _flags+=(
            'ozone_platform_wayland=true'
+           'use_xkbcommon=true'
            'use_system_libdrm=true'
-           'use_system_minigbm=true'
            'use_system_libwayland=true'
            'use_v4l2_codec=true'
-           'use_xkbcommon=true'
-           "ozone_platform=\"wayland"\"
+           "ozone_platform=\"wayland\""
            'rtc_use_pipewire=true'
+           'rtc_link_pipewire=true'
            'rtc_pipewire_version="0.3"' # will be the default in Chromium 90
            )
 fi
@@ -420,23 +430,23 @@
 # Set the bundled/external components.
 # TODO: need ported to GN as GYP doing before. see status page: https://crbug.com/551343.
 _use_system=(
-#              'ffmpeg'       # I'm not sure why, but all videos stop playback if use system ffmpeg.
+             'ffmpeg'       # I'm not sure why, but all videos stop playback if use system ffmpeg.
              'flac'
              'fontconfig'
              'freetype'
              'harfbuzz-ng'
-#              'icu'          # https://crbug.com/678661.
+             'icu'          # use bundled version if compile fails: https://crbug.com/678661.
              'libdrm'
-#              'libevent'     # Get segfaults and other problems https://bugs.gentoo.org/593458.
+             'libevent'     # Use the bundled version if there are segfaults and other problems https://bugs.gentoo.org/593458.
              'libjpeg'
              'libpng'
-#              'libvpx'
+             'libvpx'       # Use bundled version in case of compile errors
              'libwebp'
              'libxml'
              'libxslt'
              'openh264'
              'opus'
-#              're2'
+             're2'
              'snappy'
 #              'yasm'
 #              'zlib'         # NaCL needs it
@@ -466,6 +476,8 @@

 _clang_path="${BUILDDIR}${_builddir}/src/chromium-${pkgver}/third_party/llvm-build/Release+Asserts/bin/"

+export FC_CONFIG_FILE="/etc/fonts/fonts.conf"
+export FC_CONFIG_DIR="/etc/fonts"
 export CC="${_clang_path}clang"
 export CXX="${_clang_path}clang++"
 export AR="${_clang_path}llvm-ar"
@@ -473,6 +485,7 @@
 export RANLIB=/usr/bin/true
 _lld="ld.lld"

+
 ################################################

 prepare() {
@@ -489,7 +502,7 @@
   2to3 -w --no-diff third_party/ffmpeg/chromium/scripts/build_ffmpeg.py

   # Remove most bundled libraries. Some are still needed.
-  msg2 "Removing unnecessary components to save disk space."
+  msg2 "Removing unnecessary components to save disk space: "
   build/linux/unbundle/remove_bundled_libraries.py ${_keeplibs[@]} --do-remove

   msg2 "Changing bundle libraries to system ones."
@@ -523,9 +536,9 @@
       -e "s|'app_name': 'Chromium|&-dev|g" \
       -i components/policy/tools/template_writers/writer_configuration.py

-  # Fix(?) the name of the sandbox.
-  sed -e 's|chrome-sandbox|chrome_sandbox|g'\
-      -i sandbox/linux/suid/client/setuid_sandbox_host.cc
+# Fix(?) the name of the sandbox.
+#  sed -e 's|chrome-sandbox|chrome_sandbox|g'\
+#      -i sandbox/linux/suid/client/setuid_sandbox_host.cc

   # If use ccache, set it.
   if check_buildoption ccache y; then
@@ -549,31 +562,28 @@
       -i third_party/blink/renderer/core/xml/*.cc \
       -i third_party/blink/renderer/core/xml/parser/xml_document_parser.cc \
       -i third_party/libxml/chromium/libxml_utils.cc
-
+  # Fix Java assertion (https://groups.google.com/a/chromium.org/g/chromium-checkins/c/gEBqCY8Bqrc?pli=1)
   ln -s /usr/bin/java third_party/jdk/current/bin/java
+  # patch -p1 -i "${srcdir}/java-assertion.patch"

   # Fix build if enable HEVC in non-chromeOS system
+  msg2 "Fix HEVC in non-ChromeOS"
   patch -p1 -i "${srcdir}/fix_hevc_in_non_cromeos_r1.patch"

   # # Patch from Gentoo
+  msg2 "Patches from Gentoo:"
+  msg2 "1. EnumTable Crash"
   patch -p1 -i "${srcdir}/chromium-89-EnumTable-crash.patch"
+  msg2 "2. Shim Headers"
   patch -p1 -i "${srcdir}/chromium-shim_headers.patch"
-  patch -p1 -i "${srcdir}/chromium-91-ThemeService-crash.patch"
-  patch -p1 -i "${srcdir}/chromium-91-system-icu.patch"

   # # Patch from crbug.com (chromium bugtracker), chromium-review.googlesource.com / Gerrit or Arch chromium package.
-
   # Use the --oauth2-client-id= and --oauth2-client-secret= switches for
   # setting GOOGLE_DEFAULT_CLIENT_ID and GOOGLE_DEFAULT_CLIENT_SECRET at
   # runtime -- this allows signing into Chromium without baked-in values
+  msg2 "Use OAuth2 client switches as default"
   patch -p1 -i "${srcdir}/use-oauth2-client-switches-as-default.patch"

-  # https://crbug.com/1164975
-  patch -p1 -i "${srcdir}/chromium-glibc-2.33.patch"
-  sed '34i#define AT_EMPTY_PATH 0x1000' -i sandbox/linux/seccomp-bpf-helpers/sigsys_handlers.cc
-
-  # Upstream fixes
-
   # Setup nodejs dependency.
   mkdir -p third_party/node/linux/node-linux-x64/bin/
   ln -sf /usr/bin/node third_party/node/linux/node-linux-x64/bin/node

sl1pkn07 commented on 2021-04-22 13:49 (UTC)

thanks @hexhu

switch to bundled SPIRV headers

greetings

hexhu commented on 2021-04-22 04:30 (UTC) (edited on 2021-04-22 04:30 (UTC) by hexhu)

@sl1pkn07 worked like a charm:

using spirv-headers-git: compiles

only appending to _keeplibs: compiles

with both: compiles

sl1pkn07 commented on 2021-04-21 13:50 (UTC)

try to use spirv-headers-git in [AUR], or add this line in the _keeplibs array

'third_party/swiftshader/third_party/SPIRV-Headers/include/spirv/unified1'

greetings

hexhu commented on 2021-04-21 01:00 (UTC)

@sl1pkn07 somehow the error persists: https://pastebin.aosc.io/paste/Ygt6e8sGQpiYV6qgK6LOWg

thanks for the help!

sl1pkn07 commented on 2021-04-19 17:55 (UTC)

@hexhu try now

greetings

hexhu commented on 2021-04-19 17:23 (UTC) (edited on 2021-04-19 17:23 (UTC) by hexhu)

Using the latest PKGBUILD, but having error about namespace 'spv' missing members (OpAsmINTEL, OpAsmTargetINTEL, OpAsmCallINTEL, etc.) when compiling obj/third_party/swiftshader/src/Pipeline/Pipeline/SpirvShaderInstructions.o.d

Full error: https://pastebin.aosc.io/paste/E3f7Ek1fFVk958pJZdh2FQ

DAC324 commented on 2021-03-29 12:22 (UTC) (edited on 2021-03-29 12:23 (UTC) by DAC324)

If somebody wants to use this package with the latest Chromium-Dev:

Here is a preliminary patch for PKGBUILD. Just patch your PKGBUILD and build the package with makepkg the usual way.

--- PKGBUILD    2021-03-29 14:13:43.053395575 +0200
+++ PKGBUILD    2021-03-29 14:19:00.907521838 +0200
@@ -15,7 +15,7 @@
 ## -- Package and components information -- ##
 ##############################################
 pkgname=chromium-dev
-pkgver=89.0.4381.6
+pkgver=91.0.4455.2
 pkgrel=1
 pkgdesc="The open-source project behind Google Chrome (Dev Channel)"
 arch=('x86_64')
@@ -75,12 +75,12 @@
 source=(
         #"https://gsdview.appspot.com/chromium-browser-official/chromium-${pkgver}.tar.xz"
         "https://commondatastorage.googleapis.com/chromium-browser-official/chromium-${pkgver}.tar.xz"
+        # "chromium-browser-official/chromium-${pkgver}.tar.xz"
         'git+https://github.com/foutrelis/chromium-launcher.git'
         'chromium-dev.svg'
-        # Patch form Gentoo.

         # Misc Patches.
-        'fix_hevc_in_non_cromeos.patch'
+        # 'fix_hevc_in_non_cromeos.patch'

         # Patch from crbug.com (chromium bugtracker), chromium-review.googlesource.com / Gerrit or Arch chromium package.
         )
@@ -89,10 +89,9 @@
             "$(curl -sL https://commondatastorage.googleapis.com/chromium-browser-official/chromium-${pkgver}.tar.xz.hashes | grep sha256 | cut -d ' ' -f3)"
             'SKIP'
             'dd2b5c4191e468972b5ea8ddb4fa2e2fa3c2c94c79fc06645d0efc0e63ce7ee1'
-            # Patch form Gentoo

             # Misc Patches
-            '814246e08b63884dae096fe317f837259294118b5336e8da081b19eeb7991977'
+            # '814246e08b63884dae096fe317f837259294118b5336e8da081b19eeb7991977'
             # Patch from crbug (chromium bugtracker) or Arch chromium package
             )
 install=chromium-dev.install
@@ -193,12 +192,16 @@
            'third_party/devtools-frontend/src/third_party'
            'third_party/dom_distiller_js'
            'third_party/emoji-segmenter'
+           'third_party/farmhash'
+           'third_party/fdlibm'
            'third_party/ffmpeg'
+           'third_party/fft2d'
            'third_party/flatbuffers'
            'third_party/fusejs'
            'third_party/google_input_tools'
            'third_party/google_input_tools/third_party/closure_library'
            'third_party/google_input_tools/third_party/closure_library/third_party/closure'
+           'third_party/google-truth'
            'third_party/googletest'
            'third_party/harfbuzz-ng/utils'
            'third_party/hunspell'
@@ -212,12 +215,13 @@
            'third_party/libXNVCtrl'
            'third_party/libaddressinput'
            'third_party/libaom'
-           'third_party/libaom/source/libaom/third_party/vector'
-           'third_party/libaom/source/libaom/third_party/x86inc'
+           'third_party/libaom/source/libaom/'
            'third_party/libavif'
+           'third_party/libgav1'
            'third_party/libgifcodec'
            'third_party/libjingle'
            'third_party/libphonenumber'
+           'third_party/libpng'
            'third_party/libsecret'
            'third_party/libsrtp'
            'third_party/libsync'
@@ -227,10 +231,12 @@
            'third_party/libvpx'
            'third_party/libvpx/source/libvpx/third_party/x86inc'
            'third_party/libwebm'
+           'third_party/libwebp'
            'third_party/libx11'
            'third_party/libxcb-keysyms'
            'third_party/libxml/chromium'
            'third_party/libyuv'
+           'third_party/libzip'
            'third_party/llvm'
            'third_party/lottie'
            'third_party/lss'
@@ -249,6 +255,7 @@
            'third_party/openscreen'
            'third_party/openscreen/src/third_party/mozilla'
            'third_party/openscreen/src/third_party/tinycbor/src/src'
+           'third_party/opus'
            'third_party/ots'
            'third_party/pdfium'
            'third_party/pdfium/third_party/agg23'
@@ -261,6 +268,8 @@
            'third_party/pdfium/third_party/libtiff'
            'third_party/pdfium/third_party/skia_shared'
            'third_party/perfetto'
+           'third_party/perfetto/protos/third_party/chromium'
+           'third_party/perfetto/src/protozero'
            'third_party/pffft'
            'third_party/ply'
            'third_party/polymer'
@@ -272,10 +281,11 @@
            'third_party/qcms'
            'third_party/re2'
            'third_party/rnnoise'
+           'third_party/ruy'
            'third_party/s2cellid'
            'third_party/schema_org'
            'third_party/securemessage'
-           'third_party/shaka-player'
+#           'third_party/shaka-player'
            'third_party/shell-encryption'
            'third_party/simplejson'
            'third_party/skia'
@@ -284,21 +294,46 @@
            'third_party/skia/third_party/skcms'
            'third_party/smhasher'
            'third_party/sqlite'
+           'third_party/sqlite4java'
            'third_party/swiftshader'
+           'third_party/swiftshader/third_party/angle'
            'third_party/swiftshader/third_party/astc-encoder'
+           'third_party/swiftshader/third_party/boost'
+#           'third_party/swiftshader/third_party/benchmark'
+#           'third_party/swiftshader/third_party/cppdap'
+#           'third_party/swiftshader/third_party/googletest'
+#           'third_party/swiftshader/third_party/json'
+           'third_party/swiftshader/third_party/libbacktrace'
+#           'third_party/swiftshader/third_party/glslang'
+           'third_party/swiftshader/third_party/llvm-10.0'
            'third_party/swiftshader/third_party/llvm-subzero'
            'third_party/swiftshader/third_party/marl'
+#           'third_party/swiftshader/third_party/PowerVR_Examples
+           'third_party/swiftshader/third_party/SPIRV-Headers'
+           'third_party/swiftshader/third_party/SPIRV-Tools'
            'third_party/swiftshader/third_party/subzero'
            'third_party/tcmalloc'
+           'third_party/tensorflow-text'
+           'third_party/tflite'
+           'third_party/tflite/src'
+           'third_party/tflite/src/tensorflow'
+           'third_party/tflite/src/tensorflow/lite'
+           'third_party/tflite/src/tensorflow/lite/kernels'
+           'third_party/tflite/src/tensorflow/lite/kernels/internal'
+           'third_party/tflite-support'
            'third_party/tint'
+#           'third_party/turbine/src'
+#           'third_party/ub-uiautomator/lib'
            'third_party/ukey2'
            'third_party/unrar'
            'third_party/usrsctp'
+           'third_party/utf'
            'third_party/vulkan-deps'
            'third_party/vulkan_memory_allocator'
            'third_party/wayland'
            'third_party/web-animations-js'
            'third_party/webdriver'
+           'third_party/webgpu-cts'
            'third_party/webrtc'
            'third_party/webrtc/common_audio/third_party/ooura'
            'third_party/webrtc/common_audio/third_party/spl_sqrt_floor'
@@ -307,6 +342,7 @@
            'third_party/webrtc/modules/third_party/g722'
            'third_party/webrtc/rtc_base/third_party/base64'
            'third_party/webrtc/rtc_base/third_party/sigslot'
+           'third_party/weston/src'
            'third_party/widevine'
            'third_party/woff2'
            'third_party/wuffs'
@@ -324,10 +360,11 @@

            # gyp -> gn leftovers.
            'base/third_party/libevent'
-           'third_party/adobe'
+#           'third_party/adobe'
            'third_party/speech-dispatcher'
            'third_party/usb_ids'
            'third_party/xdg-utils'
+           'third_party/xstream'
            )

 _keeplibs+=(
@@ -401,8 +438,8 @@
              'libdrm'
 #              'libevent'     # Get segfaults and other problems https://bugs.gentoo.org/593458.
              'libjpeg'
-             'libpng'
-#              'libvpx'
+#             'libpng'        # compile errors with 91.0.4455.2
+#             'libvpx'
              'libwebp'
              'libxml'
              'libxslt'
@@ -462,8 +499,9 @@
   2to3 -w --no-diff third_party/ffmpeg/chromium/scripts/build_ffmpeg.py

   # Remove most bundled libraries. Some are still needed.
-  msg2 "Removing unnecessary components to save disk space."
-  build/linux/unbundle/remove_bundled_libraries.py ${_keeplibs[@]} --do-remove
+  msg2 "Removing unnecessary components to save disk space: "
+  msg2 "Not possible anymore as all bundled third-party libraries are configured as being mandatory for compiling!"
+#  build/linux/unbundle/remove_bundled_libraries.py ${_keeplibs[@]} --do-remove

   msg2 "Changing bundle libraries to system ones."
   build/linux/unbundle/replace_gn_files.py --system-libraries ${_use_system[@]}
@@ -496,9 +534,9 @@
       -e "s|'app_name': 'Chromium|&-dev|g" \
       -i components/policy/tools/template_writers/writer_configuration.py

-  # Fix(?) the name of the sandbox.
-  sed -e 's|chrome-sandbox|chrome_sandbox|g'\
-      -i sandbox/linux/suid/client/setuid_sandbox_host.cc
+# Fix(?) the name of the sandbox.
+#  sed -e 's|chrome-sandbox|chrome_sandbox|g'\
+#      -i sandbox/linux/suid/client/setuid_sandbox_host.cc

   # If use ccache, set it.
   if check_buildoption ccache y; then
@@ -517,14 +555,14 @@
     -i tools/generate_shim_headers/generate_shim_headers.py

   # https://crbug.com/893950.
-  sed -e 's/\<xmlMalloc\>/malloc/' \
-      -e 's/\<xmlFree\>/free/' \
-      -i third_party/blink/renderer/core/xml/*.cc \
-      -i third_party/blink/renderer/core/xml/parser/xml_document_parser.cc \
-      -i third_party/libxml/chromium/libxml_utils.cc
+  # sed -e 's/\<xmlMalloc\>/malloc/' \
+  #    -e 's/\<xmlFree\>/free/' \
+  #    -i third_party/blink/renderer/core/xml/*.cc \
+  #    -i third_party/blink/renderer/core/xml/parser/xml_document_parser.cc \
+  #    -i third_party/libxml/chromium/libxml_utils.cc

   # Fix build if enable HEVC in non-chromeOS system
-  patch -p1 -i "${srcdir}/fix_hevc_in_non_cromeos.patch"
+  # patch -p1 -i "${srcdir}/fix_hevc_in_non_cromeos.patch"

   # # Patch from Gentoo

@@ -555,7 +593,7 @@
 }

 build() {
-  # set python 2 for buiuld pnacl.
+  # set python 2 for build pnacl.
   export PNACLPYTHON=/usr/bin/python2

   msg2 "Build the Launcher"

DAC324 commented on 2021-03-24 10:27 (UTC)

Flagged Out of Date. Does that mean that this package is no longer maintained or developed? If that is the case, are there any alternatives?

sl1pkn07 commented on 2021-01-15 16:12 (UTC)

can you post the output?

greetings

hexhu commented on 2021-01-15 15:31 (UTC) (edited on 2021-01-15 15:32 (UTC) by hexhu)

For anyone encountering compile errors about a missing spirv header: spirv-headers should have been a makedepends for this package.

java-runtime-headless should also be a makedepends, as in extra/chromium's PKGBUILD. See my bug report a few comments below for details.

DAC324 commented on 2021-01-13 17:59 (UTC)

@auruser123: PKGBUILD has been fixed. Either try reinstalling this package or download the new PKGBUILD and build the package locally using makepkg and install it afterwards.

auruser123 commented on 2021-01-08 08:47 (UTC) (edited on 2021-01-08 08:49 (UTC) by auruser123)

Fails to install for me:

-> Cloning chromium-launcher git repo...
Cloning into bare repository '$HOME/.cache/rua/build/chromium-dev/chromium-launcher'...
remote: Enumerating objects: 233, done.
remote: Total 233 (delta 0), reused 0 (delta 0), pack-reused 233
Receiving objects: 100% (233/233), 38.26 KiB | 1.06 MiB/s, done.
Resolving deltas: 100% (118/118), done.
  -> Found chromium-dev.svg
  -> Downloading chromium-skia-harmony-r2.patch...
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
  0     0    0     0    0     0      0      0 --:--:-- --:--:-- --:--:--     0
curl: (22) The requested URL returned error: 404
==> ERROR: Failure while downloading https://git.archlinux.org/svntogit /packages.git/plain/trunk/chromium-skia-harmony.patch?h=packages/chromium
    Aborting...

Been failing like this for a couple of weeks already, so the fact that nobody else has raised this is suspicious. Is there a different package that I should be using instead? I just need the last development version of chrome because I'm on wayland.

hexhu commented on 2020-10-05 19:52 (UTC) (edited on 2020-10-05 19:54 (UTC) by hexhu)

Java should be added as a dependency. Can anyone confirm this?

compiling mojo_bindings_lite.js seems to require java

After yay -Rns java-runtime-common jre-openjdk-headless, yay -S chromium-dev:

[1943/48789] /usr/bin/python2 ../../third_party/closure_compiler/js_binary.py --compiler ../../third_party/closure_compiler/compiler/compiler.jar --output gen/mojo/public/js/mojo_bindings_lite.js --deps gen/mojo/public/js/bindings_lite_sources.js_library --sources ../../mojo/public/js/bindings_lite.js --flags jscomp_error=accessControls jscomp_error=checkTypes jscomp_error=checkVars jscomp_error=constantProperty jscomp_error=deprecated jscomp_error=externsValidation jscomp_error=globalThis jscomp_error=invalidCasts jscomp_error=misplacedTypeAnnotation jscomp_error=missingProperties jscomp_error=missingReturn jscomp_error=nonStandardJsDocs jscomp_error=suspiciousCode jscomp_error=undefinedNames jscomp_error=undefinedVars jscomp_error=unknownDefines jscomp_error=uselessCode jscomp_error=visibility compilation_level=ADVANCED_OPTIMIZATIONS language_in=ECMASCRIPT_2017 language_out=ECMASCRIPT_2015 generate_exports export_local_property_definitions isolation_mode=IIFE --externs ../../third_party/closure_compiler/externs/chrome.js ../../third_party/closure_compiler/externs/polymer-1.0.js ../../third_party/closure_compiler/externs/mojo_core.js ../../third_party/closure_compiler/externs/pending.js
FAILED: gen/mojo/public/js/mojo_bindings_lite.js 
/usr/bin/python2 ../../third_party/closure_compiler/js_binary.py --compiler ../../third_party/closure_compiler/compiler/compiler.jar --output gen/mojo/public/js/mojo_bindings_lite.js --deps gen/mojo/public/js/bindings_lite_sources.js_library --sources ../../mojo/public/js/bindings_lite.js --flags jscomp_error=accessControls jscomp_error=checkTypes jscomp_error=checkVars jscomp_error=constantProperty jscomp_error=deprecated jscomp_error=externsValidation jscomp_error=globalThis jscomp_error=invalidCasts jscomp_error=misplacedTypeAnnotation jscomp_error=missingProperties jscomp_error=missingReturn jscomp_error=nonStandardJsDocs jscomp_error=suspiciousCode jscomp_error=undefinedNames jscomp_error=undefinedVars jscomp_error=unknownDefines jscomp_error=uselessCode jscomp_error=visibility compilation_level=ADVANCED_OPTIMIZATIONS language_in=ECMASCRIPT_2017 language_out=ECMASCRIPT_2015 generate_exports export_local_property_definitions isolation_mode=IIFE --externs ../../third_party/closure_compiler/externs/chrome.js ../../third_party/closure_compiler/externs/polymer-1.0.js ../../third_party/closure_compiler/externs/mojo_core.js ../../third_party/closure_compiler/externs/pending.js
../../third_party/closure_compiler/compiler/compiler.jar --jscomp_error=accessControls --jscomp_error=checkTypes --jscomp_error=checkVars --jscomp_error=constantProperty --jscomp_error=deprecated --jscomp_error=externsValidation --jscomp_error=globalThis --jscomp_error=invalidCasts --jscomp_error=misplacedTypeAnnotation --jscomp_error=missingProperties --jscomp_error=missingReturn --jscomp_error=nonStandardJsDocs --jscomp_error=suspiciousCode --jscomp_error=undefinedNames --jscomp_error=undefinedVars --jscomp_error=unknownDefines --jscomp_error=uselessCode --jscomp_error=visibility --compilation_level=ADVANCED_OPTIMIZATIONS --language_in=ECMASCRIPT_2017 --language_out=ECMASCRIPT_2015 --generate_exports --export_local_property_definitions --isolation_mode=IIFE --externs=../../third_party/closure_compiler/externs/mojo_core.js --externs=../../third_party/closure_compiler/externs/pending.js --externs=../../third_party/closure_compiler/externs/chrome.js --externs=../../third_party/closure_compiler/externs/polymer-1.0.js --js_output_file gen/mojo/public/js/mojo_bindings_lite.js --js gen/mojo/public/interfaces/bindings/interface_control_messages.mojom-lite-for-compile.js gen/mojo/public/interfaces/bindings/native_struct.mojom-lite-for-compile.js gen/mojo/public/interfaces/bindings/pipe_control_messages.mojom-lite-for-compile.js ../../mojo/public/js/compile_preamble.js ../../mojo/public/js/bindings_lite.js ../../mojo/public/js/interface_support.js
/bin/sh: java: command not found

DAC324 commented on 2020-09-28 10:57 (UTC) (edited on 2020-09-28 11:18 (UTC) by DAC324)

How do I include vaapi and vdpau in this package?

warigan commented on 2020-08-17 23:14 (UTC) (edited on 2020-08-18 02:29 (UTC) by warigan)

Edit: I was able to build chromium-vaapi after replacing gn-git with gn, though it doesn't help me with this package since it depends on gn-git.

I tried building chromium-vaapi and I get the same issue there as well. I guess there is something wrong with my system/environment, you can disregard my previous comment.

sl1pkn07 commented on 2020-08-17 14:37 (UTC) (edited on 2020-08-17 14:37 (UTC) by sl1pkn07)

that sound like MacOS. but, why?

warigan commented on 2020-08-17 13:58 (UTC) (edited on 2020-08-17 14:04 (UTC) by warigan)

I get an error when building. I'm assuming there's a missing dependency, but I'm not sure what to look for.

Defining config //third_party/webrtc:common_objc(//build/toolchain/linux/unbundle:default)
ERROR at //third_party/webrtc/BUILD.gn:417:12: Unsupported value in libs.
  libs = [ "Foundation.framework" ]
       ^---------------------
Use frameworks to list framework dependencies.
See //third_party/webrtc_overrides/BUILD.gn:8:16: which caused the file to be included.
  configs += [ "//third_party/webrtc:common_config" ]
           ^-----------------------------------
Defining target //third_party/webrtc/rtc_base:safe_conversions(//build/toolchain/linux/unbundle:default)
Loading //third_party/webrtc/rtc_base/system/BUILD.gn (referenced from //third_party/webrtc_overrides/BUILD.gn:142)
==> ERROR: A failure occurred in build().
    Aborting...

StackException commented on 2020-07-06 17:15 (UTC) (edited on 2020-07-06 17:16 (UTC) by StackException)

libva.so.2 is missing, chromium-dev doesn't launch without missing so (shared object) files at least by default, this has been going on since about a month now, pacman -S libva worksaround this

Scimmia commented on 2020-05-20 17:12 (UTC)

The chromium-widevine optdep isn't doing anything anymore, is it? Without the patch, it won't look for the bundled plugin.

T4cC0re commented on 2020-05-04 09:56 (UTC)

In the configuration, as it is in the PKGBUILD (or maybe because of patches) this also needs libva as a dependency to build in a clean chroot

sl1pkn07 commented on 2020-04-03 20:27 (UTC) (edited on 2020-04-03 20:31 (UTC) by sl1pkn07)

update you gn-git

EDIT: fixed typo in the PKGBUILD

hugegameartgd commented on 2020-04-03 18:54 (UTC) (edited on 2020-04-05 14:41 (UTC) by hugegameartgd)

Now I get:

Defining target //: chromium_builder_asan (// build / toolchain / linux / unbundle: default)
ERROR at //BUILD.gn:1376:5: Unknown function.
    filter_exclude (["$ root_build_dir / foo"],
    ^ -------------
==> ERROR: An error happened in build ().
    Cancel ...
Error making: chromium-dev
[user @ archlinux ~] $

Edit: Thanks again, that fixed it :+1:

hugegameartgd commented on 2020-03-09 17:05 (UTC)

Thanks! 82.0.4077.0 compiled for me fine now.

hugegameartgd commented on 2020-03-07 10:55 (UTC)

Some sha256 sums failed and clang-format-new- commit number must be updated. Here is the modified PKGBUILD:

diff --git a/PKGBUILD b/PKGBUILD
index 8fbb2af..8f4f551 100644
--- a/PKGBUILD
+++ b/PKGBUILD
@@ -81,17 +81,16 @@ source=(
         'clang-format.zip::https://chromium-review.googlesource.com/changes/chromium%2Ftools%2Fbuild~2071716/revisions/4/files/scripts%2Fslave%2Frecipe_modules%2Fchromium%2Fresources%2Fclang-format/download'
         )
 sha256sums=(
-            #"$(curl -sL https://gsdview.appspot.com/chromium-browser-official/chromium-${pkgver}.tar.xz.hashes | grep sha256 | cut -d ' ' -f3)"
-            "$(curl -sL https://commondatastorage.googleapis.com/chromium-browser-official/chromium-${pkgver}.tar.xz.hashes | grep sha256 | cut -d ' ' -f3)"
             'SKIP'
-            'dd2b5c4191e468972b5ea8ddb4fa2e2fa3c2c94c79fc06645d0efc0e63ce7ee1'
+            'SKIP'
+            'SKIP'
             # Patch form Gentoo

             # Misc Patches
-            '1b93388254c9d780365e4639d494bfa337a7924426c12f7362a1f7e8e7fad014'
+            'SKIP'
             # Patch from crbug (chromium bugtracker) or Arch chromium package
-            '771292942c0901092a402cc60ee883877a99fb804cb54d568c8c6c94565a48e1'
-            '743eb4f5b1396c0e56323b67917e16aa458fa9ea7bccc7eb58fa5675dcc4b0c3'
+            'SKIP'
+            'SKIP'
             )
 install=chromium-dev.install

@@ -519,7 +518,7 @@ prepare() {
   patch -p0 -i "${srcdir}/chromium-skia-harmony-r2.patch"

   # https://crbug.com/1052503.
-  install -Dm755 "${srcdir}/clang-format_new-fd3b94d6928fa15be2c01854f52bda3c" buildtools/linux64/clang-format
+  install -Dm755 "${srcdir}/clang-format_new-4bc475ee92f6290bad252f19b6f88d3f" buildtools/linux64/clang-format

   # Setup nodejs dependency.
   mkdir -p third_party/node/linux/node-linux-x64/bin/

Compilation/linking still fails:

. DPPAPI_THUNK_IMPLEMENTATION -DUSE_AURA = 1 -DOFFICIAL_BUILD -D_FILE_OFFSET_BITS = 64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_GNU_SOURCE -DCR_CLANG_REVISION = \ "n341867-c2900381-1 \" -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -D_FORTIFY_SOURCE = 2 -DNDEBUG -DNVALGRIND -DU_USING_ICU_NAMESPACE = 0 -DU_ENABLE_DYLOAD = 0 - DUSE_CHROMIUM_ICU = 1 -DU_STATIC_IMPLEMENTATION -DICU_UTIL_DATA_IMPL = ICU_UTIL_DATA_FILE -DUCHAR_TYPE = uint16_t -DGOOGLE_PROTOBUF_NO_RTTI -DGOOGLE_PROTOBUF_NO_STATIC_INITIALIZER -DHAVE_PTHREAD -I ../ .. -Iirt_x64 / gen -I ../../ third_party / khronos -I ../../ gpu -I ../../ third_party / ced / src -I ../../ third_party / icu / source / common -I ../../ third_party / icu / source / i18n -I ../ .. / third_party / protobuf / src -fno-strict-aliasing -fcolor-diagnostics - fmerge-all-constants -U__STRICT_ANSI__ -integrated-as -m64 -march = x86-64 -fasynchronous-unwind-tables -fomit-frame-pointer -Wall -Wextra -Wimplicit-fallthrough -Wextra-semi -Wno-missing-field- initializers -Wno-unused-parameter -Wno-c ++ 11-narrowing -Wno-unneeded-internal-declaration -Os -ffunction-sections -fdata-sections -fomit-frame-pointer -g0 -fvisibility = hidden -Wno-shorten -64-to-32 -std = c ++ 14 -fno-exceptions -fno-rtti -fvisibility-inlines-hidden -c ../../ppapi/shared_impl/ppb_audio_config_shared.cc -o irt_x64 / obj / ppapi / shared_impl / shared_impl / ppb_audio_config_shared.o
. DPPAPI_THUNK_IMPLEMENTATION -DUSE_AURA = 1 -DOFFICIAL_BUILD -D_FILE_OFFSET_BITS = 64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_GNU_SOURCE -DCR_CLANG_REVISION = \ "n341867-c2900381-1 \" -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -D_FORTIFY_SOURCE = 2 -DNDEBUG -DNVALGRIND -DU_USING_ICU_NAMESPACE = 0 -DU_ENABLE_DYLOAD = 0 - DUSE_CHROMIUM_ICU = 1 -DU_STATIC_IMPLEMENTATION -DICU_UTIL_DATA_IMPL = ICU_UTIL_DATA_FILE -DUCHAR_TYPE = uint16_t -DGOOGLE_PROTOBUF_NO_RTTI -DGOOGLE_PROTOBUF_NO_STATIC_INITIALIZER -DHAVE_PTHREAD -I ../ .. -Iirt_x64 / gen -I ../../ third_party / khronos -I ../../ gpu -I ../../ third_party / ced / src -I ../../ third_party / icu / source / common -I ../../ third_party / icu / source / i18n -I ../ .. / third_party / protobuf / src -fno-strict-aliasing -fcolor-diagnostics -fm erge-all-constants -U__STRICT_ANSI__ -integrated-as -m64 -march = x86-64 -fasynchronous-unwind-tables -fomit-frame-pointer -Wall -Wextra -Wimplicit-fallthrough -Wextra-semi -Wno-missing-field- initializers -Wno-unused-parameter -Wno-c ++ 11-narrowing -Wno-unneeded-internal-declaration -Os -ffunction-sections -fdata-sections -fomit-frame-pointer -g0 -fvisibility = hidden -Wno-shorten -64-to-32 -std = c ++ 14 -fno-exceptions -fno-rtti -fvisibility-inlines-hidden -c ../../ppapi/shared_impl/ppb_device_ref_shared.cc -o irt_x64 / obj / ppapi / shared_impl / shared_impl / ppb_device_ref_shared.o
. DPPAPI_THUNK_IMPLEMENTATION -DUSE_AURA = 1 -DOFFICIAL_BUILD -D_FILE_OFFSET_BITS = 64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_GNU_SOURCE -DCR_CLANG_REVISION = \ "n341867-c2900381-1 \" -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -D_FORTIFY_SOURCE = 2 -DNDEBUG -DNVALGRIND -DU_USING_ICU_NAMESPACE = 0 -DU_ENABLE_DYLOAD = 0 - DUSE_CHROMIUM_ICU = 1 -DU_STATIC_IMPLEMENTATION -DICU_UTIL_DATA_IMPL = ICU_UTIL_DATA_FILE -DUCHAR_TYPE = uint16_t -DGOOGLE_PROTOBUF_NO_RTTI -DGOOGLE_PROTOBUF_DITOI -Iirt_x64 / gen -I ../../ third_party / khronos -I ../../ gpu -I ../../ third_party / ced / src -I ../../ third_party / icu / source / common -I ../../ third_party / icu / source / i18n -I ../../ third_party / protobuf / src -fno-strict-aliasing -fcolor-diagnostics -fmerge-all-constants -U__STRICT_ANSI__ -integrated -as -m64 -march = x86-64 -fasynchronous-unwind-tables -fomit-frame-pointer -Wall -Wextra -Wimplicit-fallthrough -Wextra-semi -Wno-missing-field-initializers -Wno-unused-parameter -Wno -c ++ 11-narrowing -Wno-unneeded-internal-declaration -Os -ffunction-sections -fdata-sections -fomit-frame-pointer -g0 -fvisibility = hidden -Wno-shorten-64-to-32 -std = c ++ 14 -fno-exceptions -fno-rtti -fvisibility-inlines-hidden -c ../../ppapi/shared_impl/ppb_audio_shared.cc -o irt_x64 / obj / ppapi / shared_impl / shared_impl / ppb_audio_shared.o
. DPPAPI_THUNK_IMPLEMENTATION -DUSE_AURA = 1 -DOFFICIAL_BUILD -D_FILE_OFFSET_BITS = 64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_GNU_SOURCE -DCR_CLANG_REVISION = \ "n341867-c2900381-1 \" -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -D_FORTIFY_SOURCE = 2 -DNDEBUG -DNVALGRIND -DU_USING_ICU_NAMESPACE = 0 -DU_ENABLE_DYLOAD = 0 - DUSE_CHROMIUM_ICU = 1 -DU_STATIC_IMPLEMENTATION -DICU_UTIL_DATA_IMPL = ICU_UTIL_DATA_FILE -DUCHAR_TYPE = uint16_t -DGOOGLE_PROTOBUF_NO_RTTI -DGOOGLE_PROTOBUF_NO_STATIC_INITIALIZER -DHAVE_PTHREAD -I ../ .. -Iirt_x64 / gen -I ../../ third_party / khronos -I ../../ gpu -I ../../ third_party / ced / src -I ../../ third_party / icu / source / common -I ../../ third_party / icu / source / i18n -I ../ .. / third_party / protobuf / src -fno-strict-aliasing -fcolor-diagnostics -f merge-all-constants -U__STRICT_ANSI__ -integrated-as -m64 -march = x86-64 -fasynchronous-unwind-tables -fomit-frame-pointer -Wall -Wextra -Wimplicit-fallthrough -Wextra-semi -Wno-missing-field- initializers -Wno-unused-parameter -Wno-c ++ 11-narrowing -Wno-unneeded-internal-declaration -Os -ffunction-sections -fdata-sections -fomit-frame-pointer -g0 -fvisibility = hidden -Wno-shorten -64-to-32 -std = c ++ 14 -fno-exceptions -fno-rtti -fvisibility-inlines-hidden -c ../../ppapi/shared_impl/ppb_input_event_shared.cc -o irt_x64 / obj / ppapi / shared_impl / shared_impl / ppb_input_event_shared.o
. DPPAPI_THUNK_IMPLEMENTATION -DUSE_AURA = 1 -DOFFICIAL_BUILD -D_FILE_OFFSET_BITS = 64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_GNU_SOURCE -DCR_CLANG_REVISION = \ "n341867-c2900381-1 \" -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -D_FORTIFY_SOURCE = 2 -DNDEBUG -DNVALGRIND -DU_USING_ICU_NAMESPACE = 0 -DU_ENABLE_DYLOAD = 0 - DUSE_CHROMIUM_ICU = 1 -DU_STATIC_IMPLEMENTATION -DICU_UTIL_DATA_IMPL = ICU_UTIL_DATA_FILE -DUCHAR_TYPE = uint16_t -DGOOGLE_PROTOBUF_NO_RTTI -DGOOGLE_PROTOBUF_NO_STATIC_INITIALIZER -DHAVE_PTHREAD -I ../ .. -Iirt_x64 / gen -I ../../ third_party / khronos -I ../../ gpu -I ../../ third_party / ced / src -I ../../ third_party / icu / source / common -I ../../ third_party / icu / source / i18n -I ../ .. / third_party / protobuf / src -fno-strict-aliasing -fcolor-diagnostics - fmerge-all-constants -U__STRICT_ANSI__ -integrated-as -m64 -march = x86-64 -fasynchronous-unwind-tables -fomit-frame-pointer -Wall -Wextra -Wimplicit-fallthrough -Wextra-semi -Wno-missing-field- initializers -Wno-unused-parameter -Wno-c ++ 11-narrowing -Wno-unneeded-internal-declaration -Os -ffunction-sections -fdata-sections -fomit-frame-pointer -g0 -fvisibility = hidden -Wno-shorten -64-to-32 -std = c ++ 14 -fno-exceptions -fno-rtti -fvisibility-inlines-hidden -c ../../ppapi/shared_impl/ppb_message_loop_shared.cc -o irt_x64 / obj / ppapi / shared_impl / shared_impl / ppb_message_loop_shared.o
. DPPAPI_THUNK_IMPLEMENTATION -DUSE_AURA = 1 -DOFFICIAL_BUILD -D_FILE_OFFSET_BITS = 64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_GNU_SOURCE -DCR_CLANG_REVISION = \ "n341867-c2900381-1 \" -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -D_FORTIFY_SOURCE = 2 -DNDEBUG -DNVALGRIND -DU_USING_ICU_NAMESPACE = 0 -DU_ENABLE_DYLOAD = 0 - DUSE_CHROMIUM_ICU = 1 -DU_STATIC_IMPLEMENTATION -DICU_UTIL_DATA_IMPL = ICU_UTIL_DATA_FILE -DUCHAR_TYPE = uint16_t -DGOOGLE_PROTOBUF_NO_RTTI -DGOOGLE_PROTOBUF_DITOI -Iirt_x64 / gen -I ../../ third_party / khronos -I ../../ gpu -I ../../ third_party / ced / src -I ../../ third_party / icu / source / common -I ../../ third_party / icu / source / i18n -I ../../ third_party / protobuf / src -fno-strict-aliasing -fcolor-diagnostics -fmerge-all-constants -U__STRICT_ANSI__ -integrated -as -m64 -march = x86-64 -fasynchronous-unwind-tables -fomit-frame-pointer -Wall -Wextra -Wimplicit-fallthrough -Wextra-semi -Wno-missing-field-initializers -Wno-unused-parameter -Wno -c ++ 11-narrowing -Wno-unneeded-internal-declaration -Os -ffunction-sections -fdata-sections -fomit-frame-pointer -g0 -fvisibility = hidden -Wno-shorten-64-to-32 -std = c ++ 14 -fno-exceptions -fno-rtti -fvisibility-inlines-hidden -c ../../ppapi/shared_impl/ppb_graphics_3d_shared.cc -o irt_x64 / obj / ppapi / shared_impl / shared_impl / ppb_graphics_3d_shared.o
. DPPAPI_THUNK_IMPLEMENTATION -DUSE_AURA = 1 -DOFFICIAL_BUILD -D_FILE_OFFSET_BITS = 64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_GNU_SOURCE -DCR_CLANG_REVISION = \ "n341867-c2900381-1 \" -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -D_FORTIFY_SOURCE = 2 -DNDEBUG -DNVALGRIND -DU_USING_ICU_NAMESPACE = 0 -DU_ENABLE_DYLOAD = 0 - DUSE_CHROMIUM_ICU = 1 -DU_STATIC_IMPLEMENTATION -DICU_UTIL_DATA_IMPL = ICU_UTIL_DATA_FILE -DUCHAR_TYPE = uint16_t -DGOOGLE_PROTOBUF_NO_RTTI -DGOOGLE_PROTOBUF_NO_STATIC_INITIALIZER -DHAVE_PTHREAD -I ../ .. -Iirt_x64 / gen -I ../../ third_party / khronos -I ../../ gpu -I ../../ third_party / ced / src -I ../../ third_party / icu / source / common -I ../../ third_party / icu / source / i18n -I ../ .. / third_party / protobuf / src -fno-strict-aliasing -fcolor-diagnostics -fmer ge-all-constants -U__STRICT_ANSI__ -integrated-as -m64 -march = x86-64 -fasynchronous-unwind-tables -fomit-frame-pointer -Wall -Wextra -Wimplicit-fallthrough -Wextra-semi -Wno-missing-field- initializers -Wno-unused-parameter -Wno-c ++ 11-narrowing -Wno-unneeded-internal-declaration -Os -ffunction-sections -fdata-sections -fomit-frame-pointer -g0 -fvisibility = hidden -Wno-shorten -64-to-32 -std = c ++ 14 -fno-exceptions -fno-rtti -fvisibility-inlines-hidden -c ../../ppapi/shared_impl/ppb_instance_shared.cc -o irt_x64 / obj / ppapi / shared_impl / shared_impl / ppb_instance_shared.o
. DPPAPI_THUNK_IMPLEMENTATION -DUSE_AURA = 1 -DOFFICIAL_BUILD -D_FILE_OFFSET_BITS = 64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_GNU_SOURCE -DCR_CLANG_REVISION = \ "n341867-c2900381-1 \" -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -D_FORTIFY_SOURCE = 2 -DNDEBUG -DNVALGRIND -DU_USING_ICU_NAMESPACE = 0 -DU_ENABLE_DYLOAD = 0 - DUSE_CHROMIUM_ICU = 1 -DU_STATIC_IMPLEMENTATION -DICU_UTIL_DATA_IMPL = ICU_UTIL_DATA_FILE -DUCHAR_TYPE = uint16_t -DGOOGLE_PROTOBUF_NO_RTTI -DGOOGLE_PROTOBUF_DITOI -Iirt_x64 / gen -I ../../ third_party / khronos -I ../../ gpu -I ../../ third_party / ced / src -I ../../ third_party / icu / source / common -I ../../ third_party / icu / source / i18n -I ../../ third_party / protobuf / src -fno-strict-aliasing -fcolor-diagnostics -fmerge-all-constants -U__STRICT_ANSI__ -integrated -as -m64 -march = x86-64 -fasynchronous-unwind-tables -fomit-frame-pointer -Wall -Wextra -Wimplicit-fallthrough -Wextra-semi -Wno-missing-field-initializers -Wno-unused-parameter -Wno -c ++ 11-narrowing -Wno-unneeded-internal-declaration -Os -ffunction-sections -fdata-sections -fomit-frame-pointer -g0 -fvisibility = hidden -Wno-shorten-64-to-32 -std = c ++ 14 -fno-exceptions -fno-rtti -fvisibility-inlines-hidden -c ../../ppapi/shared_impl/ppb_tcp_socket_shared.cc -o irt_x64 / obj / ppapi / shared_impl / shared_impl / ppb_tcp_socket_shared.o
[24833/27589] /run/media/user/compile/yay/chromium-dev/src/chromium-82.0.4068.4/third_party/llvm-build/Release+Asserts/bin/clang++ -Wl, - fatal-warnings - Wl, - build-id = sha1 -fPIC -Wl, -z, noexecstack -Wl, -z, relro -Wl, -z, now -Wl, -z, defs -Wl, - as-needed -fuse- ld = / run / media / user / compile / yay / chromium-dev / src / chromium-82.0.4068.4 / third_party / llvm-build / Release + Asserts / bin / ld.lld -Wl, - icf = all -Wl , - color-diagnostics -flto = thin -Wl, - thinlto-jobs = 8 -Wl, - thinlto-cache-dir = thinlto-cache -Wl, - thinlto-cache-policy, cache_size = 10 \% : cache_size_bytes = 10g: cache_size_files = 100000 -Wl, - lto-O0 -fwhole-program-vtables -m64 -Wl, -O2 -Wl, - gc-sections -rdynamic -nostdlib ++ -fsanitize = cfi-vcall -fsanitize = cfi-icall -pie -Wl, - disable-new-dtags -Wl, - icf = none -Wl, -O1, - sort-common, - as-needed, -z, relro, -z, now -o "./v8_context_snapshot_generator" -Wl, - start-group @ "./ v8_context_snapshot_generator.rsp" -Wl, - end-group -ldl -lpthread -lrt -lgmodule-2.0 -lglib-2.0 -lgobject-2.0 - lgthread-2.0 -lsmime3 -lnss3 -ln ssutil3 -lplds4 -lplc4 -lnspr4 -latomic -ljpeg -lpng16 -lz -lxml2 -lxslt -licui18n -licuuc -licudata -llzma -lm -lresolv -lwebpdemux -lwebpmux -lwebp -lfrebype -lexpatlll-fontcon -lexpatlset-font lopus -lopenh264 -lvpx -lX11 -lXcomposite -lXdamage -lXext -lXfixes -lXrender -lXrandr -lX11-xcb -lxcb -lxcb-dri3 -lXcursor -lXi -lXtst -ldrm -lpci -lglas-1 lpulse -lsnappy
FAILED: v8_context_snapshot_generator
/run/media/user/compile/yay/chromium-dev/src/chromium-82.0.4068.4/third_party/llvm-build/Release+Asserts/bin/clang++ -Wl, - fatal-warnings -Wl, - build -id = sha1 -fPIC -Wl, -z, noexecstack -Wl, -z, relro -Wl, -z, now -Wl, -z, defs -Wl, - as-needed -fuse-ld = / run / media / user / compile / yay / chromium-dev / src / chromium-82.0.4068.4 / third_party / llvm-build / Release + Asserts / bin / ld.lld -Wl, - icf = all -Wl, - color- diagnostics -flto = thin -Wl, - thinlto-jobs = 8 -Wl, - thinlto-cache-dir = thinlto-cache -Wl, - thinlto-cache-policy, cache_size = 10 \%: cache_size_bytes = 10g: cache_size_files = 100000 -Wl, - lto-O0 -fwhole-program-vtables -m64 -Wl, -O2 -Wl, - gc-sections -rdynamic -nostdlib ++ -fsanitize = cfi-vcall -fsanitize = cfi-icall -pie -Wl, - disable-new-dtags -Wl, - icf = none -Wl, -O1, - sort-common, - as-needed, -z, relro, -z, now -o "./ v8_context_snapshot_generator "-Wl, - start-group @" ./ v8_context_snapshot_generator.rsp "-Wl, - end-group -ldl -lpthread -lrt -lgmodule-2.0 -lglib-2.0 -lgobject-2.0 -lgthread-2.0 -lsmime3 -lnss3 -lnssutil3 -lplds 4 -lplc4 -lnspr4 -latomic -ljpeg -lpng16 -lz -lxml2 -lxslt -licui18n -licuuc -licudata -llzma -lm -lresolv -lwebpdemux -lwebpmux -lwebp -lfreetype -lexpat -lfontconb -lharfluzz lopenh264 -lvpx -lX11 -lXcomposite -lXdamage -lXext -lXfixes -lXrender -lXrandr -lX11-xcb -lxcb -lxcb-dri3 -lXcursor -lXi -lXtst -ldrm -lpci -lgbm -lpsl lsnappy
clang: warning: Unknown CUDA version 10.2. Assuming the latest supported version 10.1 [-Wunknown-cuda-version]
ld.lld: error: undefined symbol: GrTextContext :: CanDrawAsDistanceFields (SkPaint const &, SkFont const &, SkMatrix const &, SkSurfaceProps const &, bool, GrTextContext :: Options const &)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (SkGlyphRunListPainter :: processGlyphRunList (SkGlyphRunList const &, SkMatrix const &, SkSurfacePropstext const &, bain, Grin Options Gr &

ld.lld: error: undefined symbol: GrTextContext :: ComputeScalerContextFlags (GrColorInfo const &)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)

ld.lld: error: undefined symbol: GrTextBlob :: Key :: Key ()
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)

ld.lld: error: undefined symbol: GrTextBlob :: GetKey (GrTextBlob const &)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextBlobCache :: add (sk_sp <GrTextBlob>))

ld.lld: error: undefined symbol: GrTextBlob :: Key :: operator == (GrTextBlob :: Key const &) const
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)

ld.lld: error: undefined symbol: GrTextContext :: ComputeCanonicalColor (SkPaint const &, bool)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)

ld.lld: error: undefined symbol: GrTextBlob :: mustRegenerate (SkPaint const &, bool, SkMaskFilterBase :: BlurRec const &, SkMatrix const &, SkPoint)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)

ld.lld: error: undefined symbol: GrTextBlob :: Make (SkGlyphRunList const &, GrStrikeCache *, SkMatrix const &, unsigned int, bool)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)

ld.lld: error: undefined symbol: GrTextBlob :: setupKey (GrTextBlob :: Key const &, SkMaskFilterBase :: BlurRec const &, SkPaint const &)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)

ld.lld: error: undefined symbol: GrTextBlob :: ~ GrTextBlob ()
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)
>>> referenced 9 more times

ld.lld: error: undefined symbol: GrTextBlob :: operator delete (void *)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceProps const &, SkGlyphRunList const &))
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)
>>> referenced 9 more times

ld.lld: error: undefined symbol: GrTextBlob :: flush (GrTextTarget *, SkSurfaceProps const &, GrDistanceFieldAdjustTable const *, SkPaint const &, SkRGBA4f <(SkAlphaType) 2> const &, GrClip const &, SkMatrix const &, SkPoint)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)

ld.lld: error: undefined symbol: GrTextBlob :: size () const
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextContext :: drawGlyphRunList (GrRecordingContext *, GrTextTarget *, GrClip const &, SkMatrix const &, SkSurfaceGropph & const &)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextBlobCache :: add (sk_sp <GrTextBlob>))

ld.lld: error: undefined symbol: GrTextBlobCache :: checkPurge (GrTextBlob *)
>>> referenced by SkGlyphRunPainter.cpp
>>> thinlto-cache / llvmcache-9F4D2CF1E8D4561F4723A900DDC656AC85F62C67: (GrTextBlobCache :: add (sk_sp <GrTextBlob>))

ld.lld: error: undefined symbol: GrTextContext :: SanitizeOptions (GrTextContext :: Options *)
>>> referenced by SkRemoteGlyphCache.cpp
>>> thinlto-cache / llvmcache-C7BD7A705AE321A0FB3F69AA460055BF79484783: (SkTextBlobCacheDiffCanvas :: TrackLayerDevice :: drawGlyphRunList (SkGlyphRunList const &))

ld.lld: error: undefined symbol: GrTextContext :: InitDistanceFieldPaint (SkPaint const &)
>>> referenced by SkStrikeSpec.cpp
>>> thinlto-cache / llvmcache-BE60AB13125AC3E7B71B0149933C46A338EA1A1A: (SkStrikeSpec :: MakeSDFT (SkFont const &, SkPaint const &, SkSurfaceProps const &, SkMatrix const &, GrTextContext :: Options const &))

ld.lld: error: undefined symbol: GrTextContext :: InitDistanceFieldFont (SkFont const &, SkMatrix const &, GrTextContext :: Options const &, float *)
>>> referenced by SkStrikeSpec.cpp
>>> thinlto-cache / llvmcache-BE60AB13125AC3E7B71B0149933C46A338EA1A1A: (SkStrikeSpec :: MakeSDFT (SkFont const &, SkPaint const &, SkSurfaceProps const &, SkMatrix const &, GrTextContext :: Options const &))

ld.lld: error: undefined symbol: GrTextContext :: InitDistanceFieldMinMaxScale (float, SkMatrix const &, GrTextContext :: Options const &)
>>> referenced by SkStrikeSpec.cpp
>>> thinlto-cache / llvmcache-BE60AB13125AC3E7B71B0149933C46A338EA1A1A: (SkStrikeSpec :: MakeSDFT (SkFont const &, SkPaint const &, SkSurfaceProps const &, SkMatrix const &, GrTextContext :: Options const &))

ld.lld: error: undefined symbol: GrTextBlobCache :: PostPurgeBlobMessage (unsigned int, unsigned int)
>>> referenced by SkTextBlob.cpp
>>> thinlto-cache / llvmcache-E4D630BCD8492FCD1F909B34F48752293E20D157: (SkTextBlob :: ~ SkTextBlob ())

ld.lld: error: undefined symbol: GrTextContext :: Make (GrTextContext :: Options const &)
>>> referenced by GrDrawingManager.cpp
>>> thinlto-cache / llvmcache-D6F158900D61CFE4439D9DEA6D3CEAB9A062558E: (GrDrawingManager :: getTextContext ())

ld.lld: error: too many errors emitted, stopping now (use -error-limit = 0 to see all errors)
clang: error: linker command failed with exit code 1 (use -v to see invocation)
. DPPAPI_THUNK_IMPLEMENTATION -DUSE_AURA = 1 -DOFFICIAL_BUILD -D_FILE_OFFSET_BITS = 64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_GNU_SOURCE -DCR_CLANG_REVISION = \ "n341867-c2900381-1 \" -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -D_FORTIFY_SOURCE = 2 -DNDEBUG -DNVALGRIND -I ../ .. -Iirt_x64 / gen -fno-strict-aliasing -fcolor-diagnostics -fmerge-all-constants -U__STRICT_ANSI__ -integrated-as -m64 -march = x86-64 -fasynchronous-unwind-tables -fomit- frame-pointer -Wall -Wextra -Wimplicit-fallthrough -Wextra-semi -Wno-missing-field-initializers -Wno-unused-parameter -Wno-c ++ 11-narrowing -Wno-unneeded-internal-declaration -Os -ffunction -sections -fdata-sections -fomit-frame-pointer -g0 -fvisibility = hidden -std = c ++ 14 -fno-exceptions -fno-rtti -fvisibility-inlines-hidden -c ../../ppapi/thunk /ppb_file_io_thunk.cc -o irt_x64 / obj / ppapi / thunk / thunk / ppb_file_io_thunk.o
. DPPAPI_THUNK_IMPLEMENTATION -DUSE_AURA = 1 -DOFFICIAL_BUILD -D_FILE_OFFSET_BITS = 64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_GNU_SOURCE -DCR_CLANG_REVISION = \ "n341867-c2900381-1 \" -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -D_FORTIFY_SOURCE = 2 -DNDEBUG -DNVALGRIND -I ../ .. -Iirt_x64 / gen -fno-strict-aliasing -fcolor-diagnostics -fmerge-all-constants -U__STRICT_ANSI__ -integrated-as -m64 -march = x86-64 -fasynchronous-unwind-tables -fomit-frame-pointer -Wall -Wextra -Wimplicit -fallthrough -Wextra-semi -Wno-missing-field-initializers -Wno-unused-parameter -Wno-c ++ 11-narrowing -Wno-unneeded-internal-declaration -Os -ffunction-sections -fdata-sections -fomit- frame-pointer -g0 -fvisibility = hidden -std = c ++ 14 -fno-exceptions -fno-rtti -fvisibility-inl ines-hidden -c ../../ppapi/thunk/ppb_file_io_private_thunk.cc -o irt_x64 / obj / ppapi / thunk / thunk / ppb_file_io_private_thunk.o
. DPPAPI_THUNK_IMPLEMENTATION -DUSE_AURA = 1 -DOFFICIAL_BUILD -D_FILE_OFFSET_BITS = 64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_GNU_SOURCE -DCR_CLANG_REVISION = \ "n341867-c2900381-1 \" -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -D_FORTIFY_SOURCE = 2 -DNDEBUG -DNVALGRIND -I ../ .. -Iirt_x64 / gen -fno-strict-aliasing -fcolor-diagnostics -fmerge-all-constants -U__STRICT_ANSI__ -integrated-as -m64 -march = x86-64 -fasynchronous-unwind-tables -fomit-frame-pointer -Wall -Wextra -Wimplicit -fallthrough -Wextra-semi -Wno-missing-field-initializers -Wno-unused-parameter -Wno-c ++ 11-narrowing -Wno-unneeded-internal-declaration -Os -ffunction-sections -fdata-sections -fomit- frame-pointer -g0 -fvisibility = hidden -std = c ++ 14 -fno-exceptions -fno-rtti -fvisibilit y-inlines-hidden -c ../../ppapi/thunk/ppb_file_chooser_trusted_thunk.cc -o irt_x64 / obj / ppapi / thunk / thunk / ppb_file_chooser_trusted_thunk.o
. DPPAPI_THUNK_IMPLEMENTATION -DUSE_AURA = 1 -DOFFICIAL_BUILD -D_FILE_OFFSET_BITS = 64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_GNU_SOURCE -DCR_CLANG_REVISION = \ "n341867-c2900381-1 \" -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -D_FORTIFY_SOURCE = 2 -DNDEBUG -DNVALGRIND -I ../ .. -Iirt_x64 / gen -fno-strict-aliasing -fcolor-diagnostics -fmerge-all-constants -U__STRICT_ANSI__ -integrated-as -m64 -march = x86-64 -fasynchronous-unwind-tables -fomit-frame-pointer -Wall -Wextra -Wimplicit -fallthrough -Wextra-semi -Wno-missing-field-initializers -Wno-unused-parameter -Wno-c ++ 11-narrowing -Wno-unneeded-internal-declaration -Os -ffunction-sections -fdata-sections -fomit- frame-pointer -g0 -fvisibility = hidden -std = c ++ 14 -fno-exceptions -fno-rtti -fvisibility-inlines-hi dden -c ../../ppapi/thunk/ppb_file_ref_thunk.cc -o irt_x64 / obj / ppapi / thunk / thunk / ppb_file_ref_thunk.o
. DPPAPI_THUNK_IMPLEMENTATION -DUSE_AURA = 1 -DOFFICIAL_BUILD -D_FILE_OFFSET_BITS = 64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_GNU_SOURCE -DCR_CLANG_REVISION = \ "n341867-c2900381-1 \" -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -D_FORTIFY_SOURCE = 2 -DNDEBUG -DNVALGRIND -DU_USING_ICU_NAMESPACE = 0 -DU_ENABLE_DYLOAD = 0 - DUSE_CHROMIUM_ICU = 1 -DU_STATIC_IMPLEMENTATION -DICU_UTIL_DATA_IMPL = ICU_UTIL_DATA_FILE -DUCHAR_TYPE = uint16_t -DGOOGLE_PROTOBUF_NO_RTTI -DGOOGLE_PROTOBUF_DITOI -Iirt_x64 / gen -I ../../ third_party / khronos -I ../../ gpu -I ../../ third_party / ced / src -I ../../ third_party / icu / source / common -I ../../ third_party / icu / source / i18n -I ../../ third_party / protobuf / src -fno-strict-aliasing -fcolor-diagnostics -fmerge-all-constants -U__STRICT_ANSI__ -integrated -as -m64 -march = x86-64 -fasynchronous-unwind-tables -fomit-frame-pointer -Wall -Wextra -Wimplicit-fallthrough -Wextra-semi -Wno-missing-field-initializers -Wno-unused-parameter -Wno -c ++ 11-narrowing -Wno-unneeded-internal-declaration -Os -ffunction-sections -fdata-sections -fomit-frame-pointer -g0 -fvisibility = hidden -Wno-shorten-64-to-32 -std = c ++ 14 -fno-exceptions -fno-rtti -fvisibility-inlines-hidden -c ../../ppapi/shared_impl/ppb_opengles2_shared.cc -o irt_x64 / obj / ppapi / shared_impl / shared_impl / ppb_opengles2_shared.o
ninja: build stopped: subcommand failed.
==> ERROR: An error happened in build ().
    Cancel ...
[user @ archlinux chromium-dev] $

sl1pkn07 commented on 2020-02-13 19:38 (UTC)

no problem

danshick commented on 2020-02-13 19:32 (UTC)

Would you be willing to add the appropriate build flags and optional dependencies for pipewire screen sharing support under wayland?

For reference: https://src.fedoraproject.org/rpms/chromium/pull-request/19#request_diff

misc commented on 2020-01-02 05:38 (UTC) (edited on 2020-01-08 13:14 (UTC) by misc)

There's a bunch of patches in the gentoo rep that could be included, esp. the one that fixes use_system libxml.

Other bits…

  • use_system ffmpeg has been working for a long time again

  • libvpx & libwebp should also work

  • if you reduce the build commands in build() to ninja, ie. move the rest into prepare(), you can restart (makepkg -e) the compilation just fine if it crashed (out of memory etc. – recommend running with "earlyoom -m 2") – assuming it did start. If one compiles in /tmp it may also be worthwhile to prepend it with "rm -f tools/clang/crashreports/*", as those reports can be large in size (or at least were during the 'jumbo' period).

  • this patch… https://pastebin.com/6nb1LwmU …addresses three issues I had:

  • ccache wouldn't actually work with the PKGBUILD no matter what I tried (check "ccache -s" during chromium compilation and see if the counters actually change). I had to "patch" it in – "" as the comment in that gni has it seem like that's actually the intended way nowadays. CCACHE_CPP2 isn't needed, btw.

  • I can build/run with use_system icu just fine after these changes (bogus dependency & hardcoded include that the unbundle script forgets).

  • These two media/gpu/test lines (again bogus dependency) needed to be removed or else compilation wouldn't start.

sl1pkn07 commented on 2019-08-29 09:31 (UTC)

I never test

ecocode commented on 2019-08-29 09:21 (UTC)

Does this package work with swaywm (wayland) ?

j1simon commented on 2019-06-23 08:38 (UTC)

Are the patches updated?

j1simon commented on 2019-06-22 10:07 (UTC)

I think there is another error with this variable name:

...
_use_wayland=0           # Build Wayland NOTE: extremely experimental and don't work at this moment
...
if [ "${_wayland}" = "1" ]; then
...

j1simon commented on 2019-06-22 09:20 (UTC)

I think I know what the problem is with the last error: https://bugs.chromium.org/p/chromium/issues/detail?id=964411#c15

If I compile with gn works but it fails with gn-git due this change.

j1simon commented on 2019-06-21 21:11 (UTC)

It doesn't compile. Problem with vulkan:

...
ERROR at //third_party/swiftshader/src/Vulkan/BUILD.gn:102:5: Only source, header, and object files belong in the sources of a shared_library. //third_party/swiftshader/src/Vulkan/libvk_swiftshader.lds is not one of the valid types.
    "libvk_swiftshader.lds",
    ^----------------------
See //third_party/swiftshader/src/swiftshader.gni:49:3: whence it was called.
  shared_library(target_name) {
  ^----------------------------
See //third_party/swiftshader/src/Vulkan/BUILD.gn:35:1: whence it was called.
swiftshader_shared_library("swiftshader_libvulkan") {
^----------------------------------------------------
See //gpu/vulkan/x/BUILD.gn:44:5: which caused the file to be included.
    "//third_party/swiftshader/src/Vulkan:swiftshader_libvulkan",
    ^-----------------------------------------------------------
...

SolarAquarion commented on 2019-06-21 20:58 (UTC)

you need spirv-headers as a make dependency

garwol commented on 2019-06-14 17:30 (UTC)

Defining target //media/mojo/interfaces:constants_shared__generator(//build/toolchain/linux/unbundle:default)
Defining target //media/mojo/interfaces:constants_shared_cpp_sources(//build/toolchain/linux/unbundle:default)
Defining target //media/mojo/interfaces:constants_shared(//build/toolchain/linux/unbundle:default)
Running //third_party/glslang/src/BUILD.gn with toolchain //build/toolchain/linux/unbundle:default
Loading //build_overrides/glslang.gni (referenced from //third_party/glslang/src/BUILD.gn:34)
Defining config //third_party/glslang/src:glslang_public(//build/toolchain/linux/unbundle:default)
Defining target //third_party/glslang/src:glslang_sources(//build/toolchain/linux/unbundle:default)
ERROR at //third_party/glslang/src/BUILD.gn:109:5: Only source, header, and object files belong in the sources of a source_set. //third_party/glslang/src/glslang/MachineIndependent/glslang.y is not one of the valid types.
    "glslang/MachineIndependent/glslang.y",
    ^-------------------------------------
See //third_party/angle/src/tests/BUILD.gn:271:17: which caused the file to be included.
      deps += [ "${angle_glslang_dir}:glslang_sources" ]
                ^-------------------------------------
Loading //third_party/blink/renderer/modules/exported/BUILD.gn (referenced from //third_party/blink/renderer/controller/BUILD.gn:170)
==> BŁĄD: Wystąpił błąd w build().

walkindude commented on 2019-05-19 17:24 (UTC) (edited on 2019-05-20 01:33 (UTC) by walkindude)

Defining config //google_apis/gcm:proto_config(//build/toolchain/linux/unbundle:default) Defining target //ui/gfx:native_widget_types(//build/toolchain/linux/unbundle:default) Defining target //google_apis/gcm:proto(//build/toolchain/linux/unbundle:default) ERROR at //third_party/protobuf/proto_library.gni:369:15: Only source, header, and object files belong in the sources of a static_library. //out/Release/pyproto/google_apis/gcm/protocol/mcs_pb2.py is not one of the valid types. sources = get_target_outputs(":$action_name") ^--------------------------------- See //google_apis/gcm/BUILD.gn:78:1: whence it was called. proto_library("proto") { ^----------------------- See //BUILD.gn:89:7: which caused the file to be included. "//google_apis/gcm:gcm_unit_tests", ^--------------------------------- Defining target //ui/gfx:selection_bound(//build/toolchain/linux/unbundle:default) Loading //chrome/test/BUILD.gn (referenced from //BUILD.gn:104) Defining target //ui/gfx:selection_bound_sources(//build/toolchain/linux/unbundle:default) ==> ERROR: A failure occurred in build(). Aborting... ==> ERROR: Build failed, check /home/walkindude/chroot/walkindude/build

Edit: downgraded gn-git , build succeeded

SolarAquarion commented on 2019-05-19 02:21 (UTC)

patch -p0 -i "${srcdir}/chromium-skia-harmony.patch" No such patch

sl1pkn07 commented on 2019-05-17 17:18 (UTC)

Yep. Know it. This nigth upload the changes with new version.

Silvered commented on 2019-05-17 17:06 (UTC)

There is an error on line 497 in PKGBUILD change: patch -p0 -i "${srcdir}/chromium-skia-harmony.patch" to patch -p0 -i "${srcdir}/chromium-skia-harmony_r1.patch"

sl1pkn07 commented on 2019-04-22 13:49 (UTC)

try now @helvetica @garwol

helvetica commented on 2019-04-21 21:48 (UTC)

I had the same issue, and installing extra/nasm manually with pacman fixed it for me. Should probably update the PKGBUILD to include it as a makedep as it seems the one included by the source bundle is broken.

garwol commented on 2019-04-10 19:36 (UTC)

Intel(R) Core(TM)2 Quad CPU Q9550 so its not amd i have -march=native flag in makepkg.conf, maybe it causes errors? i will try build it tomorrow again

sl1pkn07 commented on 2019-04-10 14:39 (UTC)

only for test, try adding this in the line n538

-- --disable-asm

after the --branding ChromeOS

like

chromium/scripts/build_ffmpeg.py linux x64 --branding ChromeOS -- --disable-asm

question: what cpu you use?

greetings

garwol commented on 2019-04-10 04:09 (UTC) (edited on 2019-04-10 04:09 (UTC) by garwol)

FAILED: obj/third_party/ffmpeg/ffmpeg_nasm/aacpsdsp.o /usr/bin/python2 ../../build/gn_run_binary.py nasm -DPIC -felf64 -DARCH_X86_64 -W+error=all -P../../third_party/ffmpeg/chromium/config/ChromeOS/linux/x64/config.asm -I../../third_party/ffmpeg/chromium/config/ChromeOS/linux/x64/ -I../../third_party/ffmpeg/libavcodec/x86/ -I../../third_party/ffmpeg/libavutil/x86/ -I../../third_party/ffmpeg/ -I./ -I../../ -Igen/ -DPIC -MD obj/third_party/ffmpeg/ffmpeg_nasm/aacpsdsp.o.d -o obj/third_party/ffmpeg/ffmpeg_nasm/aacpsdsp.o ../../third_party/ffmpeg/libavcodec/x86/aacpsdsp.asm ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:1011: error: unknown 'cpu' type 'amdnop' ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:978: ... from macro `INIT_XMM' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:895: ... from macro `INIT_CPUFLAGS' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:116: ... from macro `CPUNOP' defined here ../../third_party/ffmpeg/libavcodec/x86/aacpsdsp.asm:55: error: unknown 'cpu' type 'amdnop' ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:978: ... from macro `INIT_XMM' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:895: ... from macro `INIT_CPUFLAGS' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:116: ... from macro `CPUNOP' defined here ../../third_party/ffmpeg/libavcodec/x86/aacpsdsp.asm:56: error: unknown 'cpu' type 'amdnop' ../../third_party/ffmpeg/libavcodec/x86/aacpsdsp.asm:35: ... from macro `PS_ADD_SQUARES' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:707: ... from macro `cglobal' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:757: ... from macro `cglobal_internal' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:978: ... from macro `INIT_XMM' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:895: ... from macro `INIT_CPUFLAGS' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:116: ... from macro `CPUNOP' defined here ../../third_party/ffmpeg/libavcodec/x86/aacpsdsp.asm:57: error: unknown 'cpu' type 'amdnop' ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:978: ... from macro `INIT_XMM' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:895: ... from macro `INIT_CPUFLAGS' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:116: ... from macro `CPUNOP' defined here ../../third_party/ffmpeg/libavcodec/x86/aacpsdsp.asm:58: error: unknown 'cpu' type 'amdnop' ../../third_party/ffmpeg/libavcodec/x86/aacpsdsp.asm:35: ... from macro `PS_ADD_SQUARES' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:707: ... from macro `cglobal' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:757: ... from macro `cglobal_internal' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:978: ... from macro `INIT_XMM' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:895: ... from macro `INIT_CPUFLAGS' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:116: ... from macro `CPUNOP' defined here ../../third_party/ffmpeg/libavcodec/x86/aacpsdsp.asm:64: error: unknown 'cpu' type 'amdnop' ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:978: ... from macro `INIT_XMM' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:895: ... from macro `INIT_CPUFLAGS' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:116: ... from macro `CPUNOP' defined here ../../third_party/ffmpeg/libavcodec/x86/aacpsdsp.asm:65: error: unknown 'cpu' type 'amdnop' ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:707: ... from macro `cglobal' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:757: ... from macro `cglobal_internal' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:978: ... from macro `INIT_XMM' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:895: ... from macro `INIT_CPUFLAGS' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:116: ... from macro `CPUNOP' defined here ../../third_party/ffmpeg/libavcodec/x86/aacpsdsp.asm:93: error: unknown 'cpu' type 'amdnop' ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:978: ... from macro `INIT_XMM' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:895: ... from macro `INIT_CPUFLAGS' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:116: ... from macro `CPUNOP' defined here ../../third_party/ffmpeg/libavcodec/x86/aacpsdsp.asm:94: error: unknown 'cpu' type 'amdnop' ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:707: ... from macro `cglobal' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:757: ... from macro `cglobal_internal' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:978: ... from macro `INIT_XMM' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:895: ... from macro `INIT_CPUFLAGS' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:116: ... from macro `CPUNOP' defined here ../../third_party/ffmpeg/libavcodec/x86/aacpsdsp.asm:126: error: unknown 'cpu' type 'amdnop' ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:978: ... from macro `INIT_XMM' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:895: ... from macro `INIT_CPUFLAGS' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:116: ... from macro `CPUNOP' defined here ../../third_party/ffmpeg/libavcodec/x86/aacpsdsp.asm:127: error: unknown 'cpu' type 'amdnop' ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:707: ... from macro `cglobal' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:757: ... from macro `cglobal_internal' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:978: ... from macro `INIT_XMM' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:895: ... from macro `INIT_CPUFLAGS' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:116: ... from macro `CPUNOP' defined here ../../third_party/ffmpeg/libavcodec/x86/aacpsdsp.asm:174: error: unknown 'cpu' type 'amdnop' ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:978: ... from macro `INIT_XMM' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:895: ... from macro `INIT_CPUFLAGS' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:116: ... from macro `CPUNOP' defined here ../../third_party/ffmpeg/libavcodec/x86/aacpsdsp.asm:175: error: unknown 'cpu' type 'amdnop' ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:707: ... from macro `cglobal' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:757: ... from macro `cglobal_internal' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:978: ... from macro `INIT_XMM' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:895: ... from macro `INIT_CPUFLAGS' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:116: ... from macro `CPUNOP' defined here ../../third_party/ffmpeg/libavcodec/x86/aacpsdsp.asm:393: error: unknown 'cpu' type 'amdnop' ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:978: ... from macro `INIT_XMM' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:895: ... from macro `INIT_CPUFLAGS' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:116: ... from macro `CPUNOP' defined here ../../third_party/ffmpeg/libavcodec/x86/aacpsdsp.asm:394: error: unknown 'cpu' type 'amdnop' ../../third_party/ffmpeg/libavcodec/x86/aacpsdsp.asm:281: ... from macro `HYBRID_SYNTHESIS_DEINT' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:707: ... from macro `cglobal' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:757: ... from macro `cglobal_internal' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:978: ... from macro `INIT_XMM' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:895: ... from macro `INIT_CPUFLAGS' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:116: ... from macro `CPUNOP' defined here ../../third_party/ffmpeg/libavcodec/x86/aacpsdsp.asm:395: error: unknown 'cpu' type 'amdnop' ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:978: ... from macro `INIT_XMM' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:895: ... from macro `INIT_CPUFLAGS' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:116: ... from macro `CPUNOP' defined here ../../third_party/ffmpeg/libavcodec/x86/aacpsdsp.asm:396: error: unknown 'cpu' type 'amdnop' ../../third_party/ffmpeg/libavcodec/x86/aacpsdsp.asm:281: ... from macro `HYBRID_SYNTHESIS_DEINT' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:707: ... from macro `cglobal' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:757: ... from macro `cglobal_internal' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:978: ... from macro `INIT_XMM' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:895: ... from macro `INIT_CPUFLAGS' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:116: ... from macro `CPUNOP' defined here ../../third_party/ffmpeg/libavcodec/x86/aacpsdsp.asm:484: error: unknown 'cpu' type 'amdnop' ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:978: ... from macro `INIT_XMM' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:895: ... from macro `INIT_CPUFLAGS' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:116: ... from macro `CPUNOP' defined here ../../third_party/ffmpeg/libavcodec/x86/aacpsdsp.asm:485: error: unknown 'cpu' type 'amdnop' ../../third_party/ffmpeg/libavcodec/x86/aacpsdsp.asm:435: ... from macro `PS_HYBRID_ANALYSIS' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:707: ... from macro `cglobal' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:757: ... from macro `cglobal_internal' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:978: ... from macro `INIT_XMM' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:895: ... from macro `INIT_CPUFLAGS' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:116: ... from macro `CPUNOP' defined here ../../third_party/ffmpeg/libavcodec/x86/aacpsdsp.asm:486: error: unknown 'cpu' type 'amdnop' ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:978: ... from macro `INIT_XMM' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:895: ... from macro `INIT_CPUFLAGS' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:116: ... from macro `CPUNOP' defined here ../../third_party/ffmpeg/libavcodec/x86/aacpsdsp.asm:487: error: unknown 'cpu' type 'amdnop' ../../third_party/ffmpeg/libavcodec/x86/aacpsdsp.asm:435: ... from macro `PS_HYBRID_ANALYSIS' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:707: ... from macro `cglobal' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:757: ... from macro `cglobal_internal' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:978: ... from macro `INIT_XMM' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:895: ... from macro `INIT_CPUFLAGS' defined here ../../third_party/ffmpeg/libavutil/x86/x86inc.asm:116: ... from macro `CPUNOP' defined here nasm failed with exit code 1 [8492/39926] /home/garwol/Desktop/temp/yaourt-tmp-garwol/aur-chromium-dev/src/chromium-75.0.3753.4/third_party/llvm-build/Release+Asserts/bin/clang++ -MMD -MF obj/extensions/browser/value_store/value_store/lazy_leveldb.o.d -DUSE_UDEV -DUSE_AURA=1 -DUSE_GLIB=1 -DUSE_NSS_CERTS=1 -DUSE_X11=1 -DFULL_SAFE_BROWSING -DSAFE_BROWSING_CSD -DSAFE_BROWSING_DB_LOCAL -DOFFICIAL_BUILD -DCHROMIUM_BUILD -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_GNU_SOURCE -DCR_CLANG_REVISION=\"356356-3\" -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -D_FORTIFY_SOURCE=2 -DNDEBUG -DNVALGRIND -DDYNAMIC_ANNOTATIONS_ENABLED=0 -DGLIB_VERSION_MAX_ALLOWED=GLIB_VERSION_2_32 -DGLIB_VERSION_MIN_REQUIRED=GLIB_VERSION_2_26 -DLEVELDB_PLATFORM_CHROMIUM=1 -DLEVELDB_PLATFORM_CHROMIUM=1 -I../.. -Igen -Igen/shim_headers/re2_shim -Igen/shim_headers/snappy_shim -I../../third_party/leveldatabase -I../../third_party/leveldatabase/src -I../../third_party/leveldatabase/src/include -fprofile-sample-use=../../chrome/android/profiles/afdo.prof -fno-strict-aliasing --param=ssp-buffer-size=4 -fstack-protector -funwind-tables -fPIC -pthread -fcolor-diagnostics -fmerge-all-constants -fcrash-diagnostics-dir=../../tools/clang/crashreports -Xclang -mllvm -Xclang -instcombine-lower-dbg-declare=0 -flto=thin -fsplit-lto-unit -fwhole-program-vtables -m64 -march=x86-64 -no-canonical-prefixes -Wall -Wextra -Wimplicit-fallthrough -Wthread-safety -Wextra-semi -Wno-missing-field-initializers -Wno-unused-parameter -Wno-c++11-narrowing -Wno-unneeded-internal-declaration -Wno-undefined-var-template -Wno-ignored-pragma-optimize -O2 -fno-ident -fdata-sections -ffunction-sections -fno-omit-frame-pointer -g0 -fsanitize=cfi-vcall -fsanitize-blacklist=../../tools/cfi/blacklist.txt -fsanitize=cfi-icall -fvisibility=hidden -Xclang -add-plugin -Xclang find-bad-constructs -Xclang -plugin-arg-find-bad-constructs -Xclang check-ipc -Wheader-hygiene -Wstring-conversion -Wtautological-overlap-compare -I/usr/include/glib-2.0 -I/usr/lib/glib-2.0/include -I/usr/lib/libffi-3.2.1/include -std=c++14 -fno-exceptions -fno-rtti -fvisibility-inlines-hidden -D_FORTIFY_SOURCE=2 -D__DATE__= -D__TIME__= -D__TIMESTAMP__= -DNO_UNWIND_TABLES -march=native -O3 -pipe -fstack-protector-strong -fno-plt -Wno-builtin-macro-redefined -fno-unwind-tables -fno-asynchronous-unwind-tables -c ../../extensions/browser/value_store/lazy_leveldb.cc -o obj/extensions/browser/value_store/value_store/lazy_leveldb.o [8493/39926] /home/garwol/Desktop/temp/yaourt-tmp-garwol/aur-chromium-dev/src/chromium-75.0.3753.4/third_party/llvm-build/Release+Asserts/bin/clang++ -MMD -MF obj/components/services/leveldb/public/interfaces/interfaces/leveldb.mojom-test-utils.o.d -DUSE_UDEV -DUSE_AURA=1 -DUSE_GLIB=1 -DUSE_NSS_CERTS=1 -DUSE_X11=1 -DFULL_SAFE_BROWSING -DSAFE_BROWSING_CSD -DSAFE_BROWSING_DB_LOCAL -DOFFICIAL_BUILD -DCHROMIUM_BUILD -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_GNU_SOURCE -DCR_CLANG_REVISION=\"356356-3\" -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -D_FORTIFY_SOURCE=2 -DNDEBUG -DNVALGRIND -DDYNAMIC_ANNOTATIONS_ENABLED=0 -DGLIB_VERSION_MAX_ALLOWED=GLIB_VERSION_2_32 -DGLIB_VERSION_MIN_REQUIRED=GLIB_VERSION_2_26 -DU_USING_ICU_NAMESPACE=0 -DU_ENABLE_DYLOAD=0 -DUSE_CHROMIUM_ICU=1 -DU_STATIC_IMPLEMENTATION -DICU_UTIL_DATA_IMPL=ICU_UTIL_DATA_FILE -DUCHAR_TYPE=uint16_t -DLEVELDB_PLATFORM_CHROMIUM=1 -DLEVELDB_PLATFORM_CHROMIUM=1 -I../.. -Igen -Igen/shim_headers/re2_shim -Igen/shim_headers/snappy_shim -I../../third_party/ced/src -I../../third_party/icu/source/common -I../../third_party/icu/source/i18n -I../../third_party/leveldatabase -I../../third_party/leveldatabase/src -I../../third_party/leveldatabase/src/include -fprofile-sample-use=../../chrome/android/profiles/afdo.prof -fno-strict-aliasing --param=ssp-buffer-size=4 -fstack-protector -funwind-tables -fPIC -pthread -fcolor-diagnostics -fmerge-all-constants -fcrash-diagnostics-dir=../../tools/clang/crashreports -Xclang -mllvm -Xclang -instcombine-lower-dbg-declare=0 -flto=thin -fsplit-lto-unit -fwhole-program-vtables -m64 -march=x86-64 -no-canonical-prefixes -Wall -Wextra -Wimplicit-fallthrough -Wthread-safety -Wextra-semi -Wno-missing-field-initializers -Wno-unused-parameter -Wno-c++11-narrowing -Wno-unneeded-internal-declaration -Wno-undefined-var-template -Wno-ignored-pragma-optimize -O2 -fno-ident -fdata-sections -ffunction-sections -fno-omit-frame-pointer -g0 -fsanitize=cfi-vcall -fsanitize-blacklist=../../tools/cfi/blacklist.txt -fsanitize=cfi-icall -fvisibility=hidden -Xclang -add-plugin -Xclang find-bad-constructs -Xclang -plugin-arg-find-bad-constructs -Xclang check-ipc -Wheader-hygiene -Wstring-conversion -Wtautological-overlap-compare -I/usr/include/glib-2.0 -I/usr/lib/glib-2.0/include -I/usr/lib/libffi-3.2.1/include -std=c++14 -fno-exceptions -fno-rtti -fvisibility-inlines-hidden -D_FORTIFY_SOURCE=2 -D__DATE__= -D__TIME__= -D__TIMESTAMP__= -DNO_UNWIND_TABLES -march=native -O3 -pipe -fstack-protector-strong -fno-plt -Wno-builtin-macro-redefined -fno-unwind-tables -fno-asynchronous-unwind-tables -c gen/components/services/leveldb/public/interfaces/leveldb.mojom-test-utils.cc -o obj/components/services/leveldb/public/interfaces/interfaces/leveldb.mojom-test-utils.o [8494/39926] /usr/bin/python2 ../../build/gn_run_binary.py nasm -DPIC -felf64 -P ../../third_party/dav1d/config/linux/x64/config.asm -I../../third_party/dav1d/libdav1d/src// -I../../third_party/dav1d/config/linux/x64/ -I./ -I../../ -Igen/ -MD obj/third_party/dav1d/dav1d_asm/mc_ssse3.o.d -o obj/third_party/dav1d/dav1d_asm/mc_ssse3.o ../../third_party/dav1d/libdav1d/src/x86/mc_ssse3.asm [8495/39926] /usr/bin/python2 ../../build/gn_run_binary.py nasm -DPIC -felf64 -P ../../third_party/dav1d/config/linux/x64/config.asm -I../../third_party/dav1d/libdav1d/src// -I../../third_party/dav1d/config/linux/x64/ -I./ -I../../ -Igen/ -MD obj/third_party/dav1d/dav1d_asm/mc.o.d -o obj/third_party/dav1d/dav1d_asm/mc.o ../../third_party/dav1d/libdav1d/src/x86/mc.asm [8496/39926] /home/garwol/Desktop/temp/yaourt-tmp-garwol/aur-chromium-dev/src/chromium-75.0.3753.4/third_party/llvm-build/Release+Asserts/bin/clang++ -MMD -MF obj/components/services/leveldb/public/interfaces/interfaces/leveldb.mojom.o.d -DUSE_UDEV -DUSE_AURA=1 -DUSE_GLIB=1 -DUSE_NSS_CERTS=1 -DUSE_X11=1 -DFULL_SAFE_BROWSING -DSAFE_BROWSING_CSD -DSAFE_BROWSING_DB_LOCAL -DOFFICIAL_BUILD -DCHROMIUM_BUILD -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_GNU_SOURCE -DCR_CLANG_REVISION=\"356356-3\" -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -D_FORTIFY_SOURCE=2 -DNDEBUG -DNVALGRIND -DDYNAMIC_ANNOTATIONS_ENABLED=0 -DGLIB_VERSION_MAX_ALLOWED=GLIB_VERSION_2_32 -DGLIB_VERSION_MIN_REQUIRED=GLIB_VERSION_2_26 -DU_USING_ICU_NAMESPACE=0 -DU_ENABLE_DYLOAD=0 -DUSE_CHROMIUM_ICU=1 -DU_STATIC_IMPLEMENTATION -DICU_UTIL_DATA_IMPL=ICU_UTIL_DATA_FILE -DUCHAR_TYPE=uint16_t -DLEVELDB_PLATFORM_CHROMIUM=1 -DLEVELDB_PLATFORM_CHROMIUM=1 -I../.. -Igen -Igen/shim_headers/re2_shim -Igen/shim_headers/snappy_shim -I../../third_party/ced/src -I../../third_party/icu/source/common -I../../third_party/icu/source/i18n -I../../third_party/leveldatabase -I../../third_party/leveldatabase/src -I../../third_party/leveldatabase/src/include -fprofile-sample-use=../../chrome/android/profiles/afdo.prof -fno-strict-aliasing --param=ssp-buffer-size=4 -fstack-protector -funwind-tables -fPIC -pthread -fcolor-diagnostics -fmerge-all-constants -fcrash-diagnostics-dir=../../tools/clang/crashreports -Xclang -mllvm -Xclang -instcombine-lower-dbg-declare=0 -flto=thin -fsplit-lto-unit -fwhole-program-vtables -m64 -march=x86-64 -no-canonical-prefixes -Wall -Wextra -Wimplicit-fallthrough -Wthread-safety -Wextra-semi -Wno-missing-field-initializers -Wno-unused-parameter -Wno-c++11-narrowing -Wno-unneeded-internal-declaration -Wno-undefined-var-template -Wno-ignored-pragma-optimize -O2 -fno-ident -fdata-sections -ffunction-sections -fno-omit-frame-pointer -g0 -fsanitize=cfi-vcall -fsanitize-blacklist=../../tools/cfi/blacklist.txt -fsanitize=cfi-icall -fvisibility=hidden -Xclang -add-plugin -Xclang find-bad-constructs -Xclang -plugin-arg-find-bad-constructs -Xclang check-ipc -Wheader-hygiene -Wstring-conversion -Wtautological-overlap-compare -I/usr/include/glib-2.0 -I/usr/lib/glib-2.0/include -I/usr/lib/libffi-3.2.1/include -std=c++14 -fno-exceptions -fno-rtti -fvisibility-inlines-hidden -D_FORTIFY_SOURCE=2 -D__DATE__= -D__TIME__= -D__TIMESTAMP__= -DNO_UNWIND_TABLES -march=native -O3 -pipe -fstack-protector-strong -fno-plt -Wno-builtin-macro-redefined -fno-unwind-tables -fno-asynchronous-unwind-tables -c gen/components/services/leveldb/public/interfaces/leveldb.mojom.cc -o obj/components/services/leveldb/public/interfaces/interfaces/leveldb.mojom.o ninja: build stopped: subcommand failed. ==> BŁĄD: Wystąpił błąd w build(). Przerywam... ==> BŁĄD:Makepkg nie był w stanie zbudować chromium-dev. ==> Powtórzyć budowanie chromium-dev?[t/N]

sl1pkn07 commented on 2019-04-09 20:07 (UTC)

can you paste more log?

greetings

garwol commented on 2019-04-09 19:44 (UTC)

failing with a lot of errors with ffmpeg like ../../third_party/ffmpeg/libavcodec/x86/aacpsdsp.asm:485: error: unknown 'cpu' type 'amdnop'

mrturcot commented on 2019-03-28 01:52 (UTC) (edited on 2019-03-28 01:53 (UTC) by mrturcot)

success with https://aur.archlinux.org/cgit/aur.git/tree/PKGBUILD?h=chromium-dev#n453 thanks!!

==> Finished making: chromium-dev 74.0.3729.28-1 (Wed 27 Mar 2019 06:51:13 PM PDT)

sl1pkn07 commented on 2019-03-27 20:24 (UTC)

https://aur.archlinux.org/cgit/aur.git/tree/?h=chromium-dev

mrturcot commented on 2019-03-27 20:21 (UTC)

$ makepkg -si
==> ERROR: install file (chromium-dev.install) does not exist or is not a regular file.

sl1pkn07 commented on 2019-03-27 20:16 (UTC)

https://aur.archlinux.org/cgit/aur.git/tree/PKGBUILD?h=chromium-dev#n453

mrturcot commented on 2019-03-27 20:14 (UTC)

Edited previous comment to provide more context, thanks

sl1pkn07 commented on 2019-03-27 20:11 (UTC)

in where?

mrturcot commented on 2019-03-27 20:10 (UTC) (edited on 2019-03-27 20:13 (UTC) by mrturcot)

Getting this this now

SyntaxError: Missing parentheses in call to 'print'. Did you mean print('''Python version 2.6 or 2.7 required

tail end of build https://pastebin.com/aFLRc5DJ that error keeps repeating.

sl1pkn07 commented on 2019-03-27 14:35 (UTC)

@garwol try now

garwol commented on 2019-03-27 13:09 (UTC)

failing on chromium-widevine-r4.patch checksum validating

==> Sprawdzanie plików source za pomocą sha256sums... chromium-74.0.3729.28.tar.xz ... Zgadza się chromium-launcher ... Pominięto chromium-dev.svg ... Zgadza się enable-vaapi.patch ... Pominięto chromium-ffmpeg-clang.patch ... Zgadza się chromium-widevine-r4.patch ... NIE ZGADZA SIĘ chromium-skia-harmony.patch ... Zgadza się ==> BŁĄD: Przynajmniej jeden plik nie jest poprawny! ==> BŁĄD:Makepkg nie był w stanie zbudować chromium-dev. ==> Powtórzyć budowanie chromium-dev?[t/N]

SolarAquarion commented on 2019-02-24 13:35 (UTC)

-mtune=generic -O2 -pipe -fstack-protector-strong -fno-plt -Wno-builtin-macro-redefined -fno-unwind-tables -fno-asynchronous-unwind-tables -c ../../media/audio/pulse/pulse_util.cc -o obj/media/audio/audio/pulse_util.o ../../media/audio/pulse/pulse_util.cc:55:29: error: expected ';' after top level declarator pa_channel_position Chromium-devToPAChannelPosition(Channels channel) { ^ ; ../../media/audio/pulse/pulse_util.cc:285:49: error: use of undeclared identifier 'devToPAChannelPosition' channel_map.map[channel_index] = Chromium-devToPAChannelPosition(ch); ^ ../../media/audio/pulse/pulse_util.cc:356:3: error: unknown type name 'ScopedPropertyList' ScopedPropertyList property_list; ^ ../../media/audio/pulse/pulse_util.cc:465:3: error: unknown type name 'ScopedPropertyList' ScopedPropertyList property_list; ^ 4 errors generated.

sl1pkn07 commented on 2019-02-24 13:35 (UTC)

do'o!!

fixed

SolarAquarion commented on 2019-02-24 13:20 (UTC) (edited on 2019-02-24 13:20 (UTC) by SolarAquarion)

FAILED: obj/media/audio/audio/pulse_util.o /home/solaraquarion/build/chromium-dev/src/chromium-74.0.3710.0/third_party/llvm-build/Release+Asserts/bin/clang++ -MMD -MF obj/media/audio/audio/pulse_util.o.d -DUSE_UDEV -DUSE_AURA=1 -DUSE_GLIB=1 -DUSE_NSS_CERTS=1 -DUSE_X11=1 -DFULL_SAFE_BROWSING -DSAFE_BROWSING_CSD -DSAFE_BROWSING_DB_LOCAL -DOFFICIAL_BUILD -DCHROMIUM_BUILD -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_GNU_SOURCE -DCR_CLANG_REVISION=\"353250-1\" -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -D_FORTIFY_SOURCE=2 -DNDEBUG -DNVALGRIND -DDYNAMIC_ANNOTATIONS_ENABLED=0 -DUSE_ALSA -DIS_MEDIA_IMPL -DUSE_PULSEAUDIO -DGLIB_VERSION_MAX_ALLOWED=GLIB_VERSION_2_32 -DGLIB_VERSION_MIN_REQUIRED=GLIB_VERSION_2_26 -DWEBP_EXTERN=extern -DSK_HAS_PNG_LIBRARY -DSK_HAS_WEBP_LIBRARY -DSK_HAS_JPEG_LIBRARY -DSK_VULKAN_HEADER=\"../../skia/config/SkVulkanConfig.h\" -DSK_VULKAN=1 -DSK_SUPPORT_GPU=1 -DSK_GPU_WORKAROUNDS_HEADER=\"gpu/config/gpu_driver_bug_workaround_autogen.h\" -DVK_NO_PROTOTYPES -I../.. -Igen -Igen/shim_headers/libpng_shim -I../../third_party/libwebp/src -Igen/shim_headers/zlib_shim -Igen/shim_headers/libdrm_shim -I../../third_party/libyuv/include -Igen/shim_headers/opus_shim -I../../skia/config -I../../skia/ext -I../../third_party/skia/include/c -I../../third_party/skia/include/codec -I../../third_party/skia/include/config -I../../third_party/skia/include/core -I../../third_party/skia/include/docs -I../../third_party/skia/include/effects -I../../third_party/skia/include/encode -I../../third_party/skia/include/gpu -I../../third_party/skia/include/pathops -I../../third_party/skia/include/ports -I../../third_party/skia/include/utils -I../../third_party/vulkan/include -I../../third_party/skia/third_party/vulkanmemoryallocator -I../../third_party/skia/src/gpu -I../../third_party/skia/src/sksl -I../../third_party/skia/modules/skottie/include -I../../third_party/vulkan/include -fprofile-sample-use=../../chrome/android/profiles/afdo.prof -fno-strict-aliasing --param=ssp-buffer-size=4 -fstack-protector -funwind-tables -fPIC -pthread -fcolor-diagnostics -fmerge-all-constants -Xclang -mllvm -Xclang -instcombine-lower-dbg-declare=0 -flto=thin -fsplit-lto-unit -fwhole-program-vtables -m64 -march=x86-64 -no-canonical-prefixes -Wall -Wextra -Wimplicit-fallthrough -Wthread-safety -Wno-missing-field-initializers -Wno-unused-parameter -Wno-c++11-narrowing -Wno-unneeded-internal-declaration -Wno-undefined-var-template -Wno-ignored-pragma-optimize -O2 -fno-ident -fdata-sections -ffunction-sections -fno-omit-frame-pointer -g0 -fsanitize=cfi-vcall -fsanitize-blacklist=../../tools/cfi/blacklist.txt -fsanitize=cfi-icall -fvisibility=hidden -Xclang -add-plugin -Xclang find-bad-constructs -Xclang -plugin-arg-find-bad-constructs -Xclang check-ipc -Wheader-hygiene -Wstring-conversion -Wtautological-overlap-compare -Wexit-time-destructors -D_REENTRANT -Wno-shorten-64-to-32 -I/usr/include/glib-2.0 -I/usr/lib/glib-2.0/include -I/usr/lib/libffi-3.2.1/include -std=c++14 -fno-exceptions -fno-rtti -fvisibility-inlines-hidden -D_FORTIFY_SOURCE=2 -D__DATE__= -D__TIME__= -D__TIMESTAMP__= -DNO_UNWIND_TABLES -march=x86-64 -mtune=generic -O2 -pipe -fstack-protector-strong -fno-plt -Wno-builtin-macro-redefined -fno-unwind-tables -fno-asynchronous-unwind-tables -c ../../media/audio/pulse/pulse_util.cc -o obj/media/audio/audio/pulse_util.o ../../media/audio/pulse/pulse_util.cc:55:29: error: expected ';' after top level declarator pa_channel_position Chromium-devToPAChannelPosition(Channels channel) { ^ ; ../../media/audio/pulse/pulse_util.cc:285:49: error: use of undeclared identifier 'devToPAChannelPosition' channel_map.map[channel_index] = Chromium-devToPAChannelPosition(ch); ^ ../../media/audio/pulse/pulse_util.cc:356:3: error: unknown type name 'ScopedPropertyList' ScopedPropertyList property_list; ^ ../../media/audio/pulse/pulse_util.cc:465:3: error: unknown type name 'ScopedPropertyList' ScopedPropertyList property_list; ^ 4 errors generated.

sl1pkn07 commented on 2019-02-17 02:30 (UTC)

Try remove and download again

SolarAquarion commented on 2019-02-16 23:59 (UTC)

the patch is failing for me

sl1pkn07 commented on 2019-02-16 15:15 (UTC)

└───╼  makepkg
==> Creando el paquete: chromium-dev 74.0.3702.0-1 (sáb 16 feb 2019 16:14:40 CET)
==> Comprobando dependencias mientras se ejecuta...
==> Comprobando dependencias mientras se compila...
==> Recibiendo las fuentes...
  -> chromium-74.0.3702.0.tar.xz ha sido encontrado 
  -> Actualizando chromium-launcher del repositorio git...
Extrayendo origin
  -> chromium-dev.svg ha sido encontrado 
  -> chromium-compiler-r7.patch ha sido encontrado 
  -> chromium-widevine-r4.patch ha sido encontrado 
  -> enable-vaapi.patch ha sido encontrado 
  -> chromium-ffmpeg-clang.patch ha sido encontrado 
  -> Descargando chromium-skia-harmony.patch...
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
100   523  100   523    0     0   1629      0 --:--:-- --:--:-- --:--:--  1629
==> Validando los archivos source con sha256sums...
    chromium-74.0.3702.0.tar.xz ... Aprobado
    chromium-launcher ... Omitido
    chromium-dev.svg ... Aprobado
    chromium-compiler-r7.patch ... Aprobado
    chromium-widevine-r4.patch ... Aprobado
    enable-vaapi.patch ... Omitido
    chromium-ffmpeg-clang.patch ... Aprobado
    chromium-skia-harmony.patch ... Aprobado

SolarAquarion commented on 2019-02-16 15:02 (UTC)

==> Validating source files with sha256sums... chromium-74.0.3702.0.tar.xz ... Passed chromium-launcher ... Skipped chromium-dev.svg ... Passed chromium-compiler-r7.patch ... Passed chromium-widevine-r4.patch ... Passed enable-vaapi.patch ... Skipped chromium-ffmpeg-clang.patch ... Passed chromium-skia-harmony.patch ... FAILED ==> ERROR: One or more files did not pass the validity check!

when i do updpkgsums

can't find file to patch at input line 3 Perhaps you used the wrong -p or --strip option? The text leading up to this was:


|--- qtwebengine-opensource-src-5.9.1/src/3rdparty/chromium/third_party/skia/src/ports/SkFontHost_FreeType.cpp.orig 2017-10-10 17:42:06.956950985 +0200 |+++ qtwebengine-opensource-src-5.9.1/src/3rdparty/chromium/third_party/skia/src/ports/SkFontHost_FreeType.cpp 2017-10-10 17:46:05.824187787 +0200


File to patch:

sl1pkn07 commented on 2019-02-16 00:47 (UTC)

done

walkindude commented on 2019-02-16 00:21 (UTC) (edited on 2019-02-16 00:33 (UTC) by walkindude)

==> Making package: chromium-dev 74.0.3702.0-1 (Sat 16 Feb 2019 03:15:11 MSK) ==> Retrieving sources... -> Found chromium-74.0.3702.0.tar.xz -> Updating chromium-launcher git repo... Fetching origin -> Found chromium-dev.svg -> Found chromium-compiler-r7.patch -> Downloading chromium-webrtc-r0.patch... % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0 curl: (22) The requested URL returned error: 404 Not Found ==> ERROR: Failure while downloading https://raw.githubusercontent.com/gentoo/gentoo/master/www-client/chromium/files/chromium-webrtc-r0.patch Aborting... ==> ERROR: Could not download sources.

sl1pkn07 commented on 2019-02-14 20:22 (UTC)

log please

SolarAquarion commented on 2019-02-14 20:21 (UTC)

Chromium-Skia-Harmony fails to do stuff

sl1pkn07 commented on 2018-12-05 07:04 (UTC)

For what??

petronny commented on 2018-12-05 05:42 (UTC)

Hi java-runtime is needed as a make dependency. Please add it.

sl1pkn07 commented on 2018-11-09 15:47 (UTC)

@petronny done

petronny commented on 2018-11-09 06:38 (UTC)

  -> Found chromium-webrtc-r0.patch
  -> Downloading chromium-ffmpeg-ebp-r1.patch...
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
  0     0    0     0    0     0      0      0 --:--:--  0:00:01 --:--:--     0
curl: (22) The requested URL returned error: 404 Not Found
==> ERROR: Failure while downloading https://raw.githubusercontent.com/gentoo/gentoo/master/www-client/chromium/files/chromium-ffmpeg-ebp-r1.patch
    Aborting...
==> ERROR: Could not download sources.

This patch exists no more.

petronny commented on 2018-10-26 06:56 (UTC)

I get

../../third_party/blink/renderer/platform/fonts/shaping/harfbuzz_face.cc:316:5: warning: 'hb_font_funcs_set_glyph_func' is deprecated [-Wdeprecated-declarations]
    hb_font_funcs_set_glyph_func(funcs, HarfBuzzGetGlyph, nullptr, nullptr);
    ^
/usr/include/harfbuzz/hb-deprecated.h:53:11: note: 'hb_font_funcs_set_glyph_func' has been explicitly marked deprecated here
HB_EXTERN HB_DEPRECATED_FOR(hb_font_funcs_set_nominal_glyph_func or hb_font_funcs_set_variation_glyph_func) void
          ^
/usr/include/harfbuzz/hb-common.h:79:30: note: expanded from macro 'HB_DEPRECATED_FOR'
#define HB_DEPRECATED_FOR(f) HB_DEPRECATED
                             ^
/usr/include/harfbuzz/hb-common.h:67:38: note: expanded from macro 'HB_DEPRECATED'
#define HB_DEPRECATED __attribute__((__deprecated__))
                                     ^
../../third_party/blink/renderer/platform/fonts/shaping/harfbuzz_face.cc:319:5: error: no matching function for call to 'hb_font_funcs_set_glyph_h_advances_func'
    hb_font_funcs_set_glyph_h_advances_func(
    ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
/usr/include/harfbuzz/hb-font.h:307:1: note: candidate function not viable: no known conversion from 'void (hb_font_t *, void *, unsigned int, hb_codepoint_t *, unsigned int, hb_position_t *, unsigned int, void *)' (aka 'void (hb_font_t *, void *, unsigned int, unsigned int *, unsigned int, int *, unsigned int, void *)') to 'hb_font_get_glyph_h_advances_func_t' (aka 'void (*)(hb_font_t *, void *, unsigned int, const unsigned int *, unsigned int, int *, unsigned int, void *)') for 2nd argument
hb_font_funcs_set_glyph_h_advances_func (hb_font_funcs_t *ffuncs,
^
1 warning and 1 error generated.

No idea how to fix it.

sl1pkn07 commented on 2018-07-31 19:11 (UTC)

fixed

woodchip commented on 2018-07-31 18:27 (UTC)

I'm getting the following: ==> Validating source files with sha256sums... chromium-69.0.3497.12.tar.xz ... Passed chromium-launcher ... Skipped chromium-dev.svg ... Passed chromium-compiler-r4.patch ... Passed chromium-webrtc-r0.patch ... Passed chromium-ffmpeg-ebp-r1.patch ... FAILED chromium-ffmpeg-clang.patch ... Passed chromium-intel-vaapi_r18.diff.base64 ... Passed chromium-widevine-r2.patch ... Passed chromium-skia-harmony.patch ... Passed fix_mixup_between_DIP_pixel_coordinates.diff.base64 ... Passed ==> ERROR: One or more files did not pass the validity check! What am I doing wrong?

bm456 commented on 2018-07-31 00:07 (UTC)

hevc = enables h265 playback (h265 is 30% faster than x26 and is a bit lighter)

ac3-eac3 = dolby audio (both are at chromium's code)

https://github.com/henrypp/chromium/blob/master/args_gn.md

sl1pkn07 commented on 2018-07-30 19:42 (UTC)

that flags come from?

bm456 commented on 2018-07-30 19:34 (UTC)

Hi

can we test these new GN flags? at dev channel?

may be interesting to get it

enable_hevc_demuxing = true

enable_ac3_eac3_audio_demuxing = true

sl1pkn07 commented on 2018-07-22 18:42 (UTC) (edited on 2018-07-22 18:50 (UTC) by sl1pkn07)

gn is only and exlusive for build

before, gn is builded with clang because is a little mess configure compiler for each component

bm456 commented on 2018-07-22 17:45 (UTC) (edited on 2018-07-22 18:07 (UTC) by bm456)

Hi

can you use my GN (its better cause it uses the SAME build env as chromium)

use the same env will cause "forgiven & strange" refs + "random & unexplanaible" crashes

https://aur.archlinux.org/packages/gn-chromium/

bm456 commented on 2018-07-22 03:01 (UTC)

Hi

gentoo had fixeed their ebuilds

and we need to know how add GN :)

gn-git + gentoo patches https://github.com/gentoo/gentoo/blob/master/dev-util/gn/gn-0.1445.ebuild

bm456 commented on 2018-07-18 22:48 (UTC) (edited on 2018-07-18 22:54 (UTC) by bm456)

What's can we do?

69.0.3486.0 crashes with and without gentoo's fixes

and 69.0.3493.3 had removed GN

https://omahaproxy.appspot.com/changelog?old_version=69.0.3486.0&new_version=69.0.3493.3

the last working tarball is 69.0.3473.0 (i'm using it)

laenco commented on 2018-06-26 07:34 (UTC)

got ethernal gpu crash on 69.0.3464.0-1 with nvidia gpu (1060). Resetting flags changes nothing. Does anyone have the same problem?

sl1pkn07 commented on 2018-06-24 17:28 (UTC)

yust a typo. fixed

Scimmia commented on 2018-06-24 15:58 (UTC)

You added a dep without even checking if it existed, or if the package you just uploaded could be installed?

SolarAquarion commented on 2018-06-24 01:48 (UTC)

In file included from ../../media/gpu/vaapi/vaapi_video_encode_accelerator.h:20: ../../media/gpu/vaapi/va_surface.h:11:10: fatal error: 'va/va.h' file not found

include <va va.h=""></va>
     ^~~~~~~~~

It depends on libva

j1simon commented on 2018-06-22 12:17 (UTC)

https://raw.githubusercontent.com/gentoo/gentoo/master/www-client/chromium/files/chromium-compiler-r1.patch doesn't exist.

SolarAquarion commented on 2018-06-20 21:15 (UTC)

https://github.com/gentoo/gentoo/commit/e6f520f2cdff24e23a3f34c99db6ef73f6bffe43#diff-e511ef19ad7a09d06ca73d83b73de8eb

petronny commented on 2018-05-12 17:37 (UTC)

Please apply this patch to fix:

In file included from ../../components/crash/content/app/breakpad_linux.cc:48:
In file included from ../../third_party/breakpad/breakpad/src/client/linux/handler/exception_handler.h:42:
In file included from ../../third_party/breakpad/breakpad/src/client/linux/minidump_writer/minidump_writer.h:41:
In file included from ../../third_party/breakpad/breakpad/src/client/linux/minidump_writer/linux_dumper.h:51:
In file included from ../../third_party/breakpad/breakpad/src/client/linux/dump_writer_common/thread_info.h:37:
In file included from ../../third_party/breakpad/breakpad/src/common/memory_allocator.h:50:
In file included from ../../third_party/lss/linux_syscall_support.h:111:
../../third_party/crashpad/crashpad/compat/linux/sys/ptrace.h:25:35: error: redefinition of 'PTRACE_GET_THREAD_AREA' as different kind of symbol
static constexpr __ptrace_request PTRACE_GET_THREAD_AREA =
                                  ^
/usr/include/sys/ptrace.h:110:3: note: previous definition is here
  PTRACE_GET_THREAD_AREA = 25,
  ^
1 error generated.

More: https://groups.google.com/a/chromium.org/forum/#!topic/crashpad-dev/VsBPzKxVQfk

SolarAquarion commented on 2018-05-11 23:07 (UTC)

@sl1pkn07 gentoo haas nonsfi disabled

SolarAquarion commented on 2018-05-09 02:27 (UTC) (edited on 2018-05-09 14:55 (UTC) by SolarAquarion)

That won't work. libevent hasn't been touched in 6 years. An interesting proposal is to just do a git clone of git master and send the updated code to third_party/libevent?

Like rsync * to third_party/libevent. It may work, forgot to do autogen.sh.

A large part of Chromium needs to be rebased on newest libevent

SolarAquarion commented on 2018-05-09 02:17 (UTC)

https://pastebin.com/mcgDSgHc I get it.

Try libevent-git and use the system version of it to build?

sl1pkn07 commented on 2018-05-09 01:49 (UTC) (edited on 2018-05-12 18:33 (UTC) by sl1pkn07)

each console have his own method, for example konsole/yakuake is right-mouse -> search

BTW, build with 'enable_nacl_nonsfi=false' do the trick in my case, but break the secure shell app (nacl app) :/

SolarAquarion commented on 2018-05-08 21:58 (UTC)

How do i do that?

sl1pkn07 commented on 2018-05-08 20:14 (UTC)

search FAILED word in your log

SolarAquarion commented on 2018-05-08 20:12 (UTC)

So my compilation error is unique?

sl1pkn07 commented on 2018-05-08 20:12 (UTC)

nope. my issue is with the libevent part

seems native_client_nosfi is never set, and pnacl-clang not have this define

SolarAquarion commented on 2018-05-08 20:09 (UTC)

Is this your issue

[33109/35282] /usr/bin/python2 ../../v8/tools/run.py ./mksnapshot --turbo_instruction_scheduling --embedded_src gen/v8/embedded.cc --embedded_variant Default --random-seed 314159265 --startup_blob snapshot_blob.bin [33110/35282] ../../native_client/toolchain/linux_x86/pnacl_newlib/bin/x86_64-nacl-clang++ -MMD -MF irt_x64/obj/components/tracing/tracing/tracing_messages.o.d -DNACL_TC_REV=b48266b051f596802addff7c8d1fc6fb9cf6bea7 -DTRACING_IMPLEMENTATION -DV8_DEPRECATION_WARNINGS -DNO_TCMALLOC -DFULL_SAFE_BROWSING -DSAFE_BROWSING_CSD -DSAFE_BROWSING_DB_LOCAL -DOFFICIAL_BUILD -DCHROMIUM_BUILD -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -DSTDC_CONSTANT_MACROS -DSTDC_FORMAT_MACROS -DNDEBUG -DNVALGRIND -DGOOGLE_PROTOBUF_NO_RTTI -DGOOGLE_PROTOBUF_NO_STATIC_INITIALIZER -DHAVE_PTHREAD -I../.. -Iirt_x64/gen -I../../third_party/protobuf/src -fno-strict-aliasing -fcolor-diagnostics -fmerge-all-constants -integrated-as -fasynchronous-unwind-tables -fomit-frame-pointer -Wall -Wextra -Wimplicit-fallthrough -Wthread-safety -Wno-missing-field-initializers -Wno-unused-parameter -Wno-c++11-narrowing -Wno-covered-switch-default -Wno-unneeded-internal-declaration -Wno-inconsistent-missing-override -Os -ffunction-sections -fdata-sections -fomit-frame-pointer -fvisibility=hidden -std=gnu++14 -fno-exceptions -fno-rtti -fvisibility-inlines-hidden -c ../../components/tracing/common/tracing_messages.cc -o irt_x64/obj/components/tracing/tracing/tracing_messages.o [33111/35282] ../../native_client/toolchain/linux_x86/pnacl_newlib/bin/x86_64-nacl-clang++ -MMD -MF irt_x64/obj/gpu/ipc/common/command_buffer_traits_sources/gpu_command_buffer_traits.o.d -DNACL_TC_REV=b48266b051f596802addff7c8d1fc6fb9cf6bea7 -DV8_DEPRECATION_WARNINGS -DNO_TCMALLOC -DFULL_SAFE_BROWSING -DSAFE_BROWSING_CSD -DSAFE_BROWSING_DB_LOCAL -DOFFICIAL_BUILD -DCHROMIUM_BUILD -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -DSTDC_CONSTANT_MACROS -DSTDC_FORMAT_MACROS -DNDEBUG -DNVALGRIND -DGPU_IMPLEMENTATION -DGOOGLE_PROTOBUF_NO_RTTI -DGOOGLE_PROTOBUF_NO_STATIC_INITIALIZER -DHAVE_PTHREAD -I../.. -Iirt_x64/gen -I../../third_party/khronos -I../../gpu -I../../third_party/protobuf/src -fno-strict-aliasing -fcolor-diagnostics -fmerge-all-constants -integrated-as -fasynchronous-unwind-tables -fomit-frame-pointer -Wall -Wextra -Wimplicit-fallthrough -Wthread-safety -Wno-missing-field-initializers -Wno-unused-parameter -Wno-c++11-narrowing -Wno-covered-switch-default -Wno-unneeded-internal-declaration -Wno-inconsistent-missing-override -Os -ffunction-sections -fdata-sections -fomit-frame-pointer -fvisibility=hidden -std=gnu++14 -fno-exceptions -fno-rtti -fvisibility-inlines-hidden -c ../../gpu/ipc/common/gpu_command_buffer_traits.cc -o irt_x64/obj/gpu/ipc/common/command_buffer_traits_sources/gpu_command_buffer_traits.o [33112/35282] ../../native_client/toolchain/linux_x86/pnacl_newlib/bin/x86_64-nacl-clang++ -MMD -MF irt_x64/obj/ipc/ipc/ipc_channel_common.o.d -DNACL_TC_REV=b48266b051f596802addff7c8d1fc6fb9cf6bea7 -DIS_IPC_IMPL -DV8_DEPRECATION_WARNINGS -DNO_TCMALLOC -DFULL_SAFE_BROWSING -DSAFE_BROWSING_CSD -DSAFE_BROWSING_DB_LOCAL -DOFFICIAL_BUILD -DCHROMIUM_BUILD -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -DSTDC_CONSTANT_MACROS -DSTDC_FORMAT_MACROS -DNDEBUG -DNVALGRIND -DGOOGLE_PROTOBUF_NO_RTTI -DGOOGLE_PROTOBUF_NO_STATIC_INITIALIZER -DHAVE_PTHREAD -I../.. -Iirt_x64/gen -I../../third_party/protobuf/src -fno-strict-aliasing -fcolor-diagnostics -fmerge-all-constants -integrated-as -fasynchronous-unwind-tables -fomit-frame-pointer -Wall -Wextra -Wimplicit-fallthrough -Wthread-safety -Wno-missing-field-initializers -Wno-unused-parameter -Wno-c++11-narrowing -Wno-covered-switch-default -Wno-unneeded-internal-declaration -Wno-inconsistent-missing-override -Os -ffunction-sections -fdata-sections -fomit-frame-pointer -fvisibility=hidden -std=gnu++14 -fno-exceptions -fno-rtti -fvisibility-inlines-hidden -c ../../ipc/ipc_channel_common.cc -o irt_x64/obj/ipc/ipc/ipc_channel_common.o [33113/35282] ../../native_client/toolchain/linux_x86/pnacl_newlib/bin/x86_64-nacl-clang++ -MMD -MF irt_x64/obj/components/tracing/tracing/child_trace_message_filter.o.d -DNACL_TC_REV=b48266b051f596802addff7c8d1fc6fb9cf6bea7 -DTRACING_IMPLEMENTATION -DV8_DEPRECATION_WARNINGS -DNO_TCMALLOC -DFULL_SAFE_BROWSING -DSAFE_BROWSING_CSD -DSAFE_BROWSING_DB_LOCAL -DOFFICIAL_BUILD -DCHROMIUM_BUILD -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -DSTDC_CONSTANT_MACROS -DSTDC_FORMAT_MACROS -DNDEBUG -DNVALGRIND -DGOOGLE_PROTOBUF_NO_RTTI -DGOOGLE_PROTOBUF_NO_STATIC_INITIALIZER -DHAVE_PTHREAD -I../.. -Iirt_x64/gen -I../../third_party/protobuf/src -fno-strict-aliasing -fcolor-diagnostics -fmerge-all-constants -integrated-as -fasynchronous-unwind-tables -fomit-frame-pointer -Wall -Wextra -Wimplicit-fallthrough -Wthread-safety -Wno-missing-field-initializers -Wno-unused-parameter -Wno-c++11-narrowing -Wno-covered-switch-default -Wno-unneeded-internal-declaration -Wno-inconsistent-missing-override -Os -ffunction-sections -fdata-sections -fomit-frame-pointer -fvisibility=hidden -std=gnu++14 -fno-exceptions -fno-rtti -fvisibility-inlines-hidden -c ../../components/tracing/child/child_trace_message_filter.cc -o irt_x64/obj/components/tracing/tracing/child_trace_message_filter.o [33114/35282] ../../native_client/toolchain/linux_x86/pnacl_newlib/bin/x86_64-nacl-clang++ -MMD -MF irt_x64/obj/ipc/mojom/ipc.mojom.o.d -DNACL_TC_REV=b48266b051f596802addff7c8d1fc6fb9cf6bea7 -DIS_IPC_MOJOM_IMPL -DV8_DEPRECATION_WARNINGS -DNO_TCMALLOC -DFULL_SAFE_BROWSING -DSAFE_BROWSING_CSD -DSAFE_BROWSING_DB_LOCAL -DOFFICIAL_BUILD -DCHROMIUM_BUILD -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -DSTDC_CONSTANT_MACROS -DSTDC_FORMAT_MACROS -DNDEBUG -DNVALGRIND -I../.. -Iirt_x64/gen -fno-strict-aliasing -fcolor-diagnostics -fmerge-all-constants -integrated-as -fasynchronous-unwind-tables -fomit-frame-pointer -Wall -Wextra -Wimplicit-fallthrough -Wthread-safety -Wno-missing-field-initializers -Wno-unused-parameter -Wno-c++11-narrowing -Wno-covered-switch-default -Wno-unneeded-internal-declaration -Wno-inconsistent-missing-override -Os -ffunction-sections -fdata-sections -fomit-frame-pointer -fvisibility=hidden -std=gnu++14 -fno-exceptions -fno-rtti -fvisibility-inlines-hidden -c irt_x64/gen/ipc/ipc.mojom.cc -o irt_x64/obj/ipc/mojom/ipc.mojom.o

sl1pkn07 commented on 2018-05-08 19:40 (UTC)

oh, is a newer version. now in test

sl1pkn07 commented on 2018-05-08 15:10 (UTC) (edited on 2018-05-08 15:14 (UTC) by sl1pkn07)

https://pastebin.com/gccHmTE6

SolarAquarion commented on 2018-05-08 14:44 (UTC)

/home/solaraquarion/build/chromium-dev/src/chromium-68.0.3418.2/third_party/ffmpeg/libavcodec/libopus.c:22:10: fatal error: 'opus_defines.h' file not found

include <opus_defines.h></opus_defines.h>
     ^~~~~~~~~~~~~~~~

/home/solaraquarion/build/chromium-dev/src/chromium-68.0.3418.2/third_party/ffmpeg/libavcodec/libopusdec.c:22:10: error: 'opus.h' file not found with <angled> include; use "quotes" instead </angled>

include <opus.h></opus.h>
     ^~~~~~~~
     "opus.h"

1 error generated. <br> CC libavcodec/lsp.o make: [/home/solaraquarion/build/chromium-dev/src/chromium-68.0.3418.2/third_party/ffmpeg/ffbuild/common.mak:60: libavcodec/libopus.o] Error 1 make: Waiting for unfinished jobs.... /home/solaraquarion/build/chromium-dev/src/chromium-68.0.3418.2/third_party/ffmpeg/libavcodec/libopusdec.c:23:10: fatal error: 'opus_multistream.h' file not found

include <opus_multistream.h></opus_multistream.h>
     ^~~~~~~~~~~~~~~~~~~~

I'm getting this with the build

sl1pkn07 commented on 2018-05-08 04:22 (UTC)

See #c3 on the bugreport

SolarAquarion commented on 2018-05-08 01:06 (UTC)

But Chromium is at this? 68.0.3409.2

sl1pkn07 commented on 2018-05-06 10:34 (UTC)

stuck in here: https://bugs.chromium.org/p/chromium/issues/detail?id=837441

SolarAquarion commented on 2018-05-04 03:08 (UTC)

https://github.com/gentoo/gentoo/commit/ba6e7372f34d1760309b03615ad069fda5d916a7#diff-e511ef19ad7a09d06ca73d83b73de8eb

yangci20973 commented on 2018-04-29 14:14 (UTC) (edited on 2018-04-30 13:04 (UTC) by yangci20973)

i managed to compile every file by:

  1. change _clang_path as @jrudess suggested
  2. apply a patch found in https://groups.google.com/a/chromium.org/forum/#!topic/crashpad-dev/VsBPzKxVQfk (solved @SolarAquarion's last build error)

still, the build process can't be accomplished.

got a error similar to https://github.com/pocl/pocl/issues/360, no valuable hint found on the internet.

==============================================================================

update: finished building by apply this patch: https://pastebin.com/unuPmuCE

sl1pkn07 commented on 2018-04-25 17:46 (UTC)

more log please. search the word 'FAILED'

SolarAquarion commented on 2018-04-23 13:33 (UTC)

Now i get this build failure

[9062/34740] /home/solaraquarion/build/chromium-dev/src/chromium-67.0.3396.10/third_party/llvm-build/Release+Asserts/bin/clang++ -MMD -MF obj/components/services/font/lib/font_service_app.o.d -DV8_DEPRECATION_WARNINGS -DUSE_UDEV -DUSE_AURA=1 -DUSE_GLIB=1 -DUSE_NSS_CERTS=1 -DUSE_X11=1 -DFULL_SAFE_BROWSING -DSAFE_BROWSING_CSD -DSAFE_BROWSING_DB_LOCAL -DOFFICIAL_BUILD -DCHROMIUM_BUILD -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -DCR_CLANG_REVISION=\"328716-2\" -DSTDC_CONSTANT_MACROS -DSTDC_FORMAT_MACROS -DNDEBUG -DNVALGRIND -DDYNAMIC_ANNOTATIONS_ENABLED=0 -DGLIB_VERSION_MAX_ALLOWED=GLIB_VERSION_2_32 -DGLIB_VERSION_MIN_REQUIRED=GLIB_VERSION_2_26 -DSK_IGNORE_LINEONLY_AA_CONVEX_PATH_OPTS -DSK_HAS_PNG_LIBRARY -DSK_HAS_WEBP_LIBRARY -DSK_HAS_JPEG_LIBRARY -DSK_VULKAN_HEADER=\"../../skia/config/SkVulkanConfig.h\" -DSK_VULKAN=1 -DSK_SUPPORT_GPU=1 -DU_USING_ICU_NAMESPACE=0 -DU_ENABLE_DYLOAD=0 -DU_STATIC_IMPLEMENTATION -DICU_UTIL_DATA_IMPL=ICU_UTIL_DATA_FILE -DUCHAR_TYPE=uint16_t -DGOOGLE_PROTOBUF_NO_RTTI -DGOOGLE_PROTOBUF_NO_STATIC_INITIALIZER -DHAVE_PTHREAD -I../.. -Igen -Igen/shim_headers/libpng_shim -Igen/shim_headers/zlib_shim -I../../skia/config -I../../skia/ext -I../../third_party/skia/include/c -I../../third_party/skia/include/config -I../../third_party/skia/include/core -I../../third_party/skia/include/effects -I../../third_party/skia/include/encode -I../../third_party/skia/include/gpu -I../../third_party/skia/include/images -I../../third_party/skia/include/lazy -I../../third_party/skia/include/pathops -I../../third_party/skia/include/pdf -I../../third_party/skia/include/pipe -I../../third_party/skia/include/ports -I../../third_party/skia/include/utils -I/usr/include -I../../third_party/skia/src/gpu -I../../third_party/skia/src/sksl -I../../third_party/skia/include/gpu/vk -I../../third_party/vulkan/include -I../../third_party/ced/src -I../../third_party/icu/source/common -I../../third_party/icu/source/i18n -I../../third_party/protobuf/src -fno-strict-aliasing -fmerge-all-constants --param=ssp-buffer-size=4 -fstack-protector -funwind-tables -fPIC -pipe -pthread -fcolor-diagnostics -no-canonical-prefixes -flto=thin -fwhole-program-vtables -m64 -march=x86-64 -Wall -Wextra -Wimplicit-fallthrough -Wthread-safety -Wno-missing-field-initializers -Wno-unused-parameter -Wno-c++11-narrowing -Wno-covered-switch-default -Wno-unneeded-internal-declaration -Wno-inconsistent-missing-override -Wno-undefined-var-template -Wno-nonportable-include-path -Wno-address-of-packed-member -Wno-unused-lambda-capture -Wno-user-defined-warnings -Wno-enum-compare-switch -Wno-null-pointer-arithmetic -Wno-ignored-pragma-optimize -O2 -fno-ident -fdata-sections -ffunction-sections -fno-omit-frame-pointer -g0 -fsanitize=cfi-vcall -fsanitize-blacklist=../../tools/cfi/blacklist.txt -fvisibility=hidden -Xclang -load -Xclang ../../third_party/llvm-build/Release+Asserts/lib/libFindBadConstructs.so -Xclang -add-plugin -Xclang find-bad-constructs -Xclang -plugin-arg-find-bad-constructs -Xclang no-realpath -Xclang -plugin-arg-find-bad-constructs -Xclang check-enum-max-value -Xclang -plugin-arg-find-bad-constructs -Xclang check-ipc -Wheader-hygiene -Wstring-conversion -Wtautological-overlap-compare -isystem../../../../../../../../usr/include/glib-2.0 -isystem../../../../../../../../usr/lib/glib-2.0/include -std=gnu++14 -fno-exceptions -fno-rtti -fvisibility-inlines-hidden -D_FORTIFY_SOURCE=2 -DDATE= -DTIME= -DTIMESTAMP= -DNO_UNWIND_TABLES -march=x86-64 -mtune=generic -O2 -pipe -fstack-protector-strong -Wno-builtin-macro-redefined -fno-unwind-tables -fno-asynchronous-unwind-tables -c ../../components/services/font/font_service_app.cc -o obj/components/services/font/lib/font_service_app.o ninja: build stopped: subcommand failed. ==> ERROR: A failure occurred in build().

SolarAquarion commented on 2018-04-23 13:16 (UTC)

that fixes it for me

jrudess commented on 2018-04-23 04:00 (UTC) (edited on 2018-04-23 04:00 (UTC) by jrudess)

I have the same build error as SolarAquarion. This edit to pkgbuild fixes it for me:

Old: _clang_path="${BUILDDIR}/${pkgname}/src/chromium-${pkgver}/third_party/llvm-build/Release+Asserts/bin/"

New: _clang_path="${BUILDDIR}/src/chromium-${pkgver}/third_party/llvm-build/Release+Asserts/bin/"

sl1pkn07 commented on 2018-04-22 20:31 (UTC)

where install the script tools/clang/scripts/update.py, the bundled clang in your system?

SolarAquarion commented on 2018-04-22 20:07 (UTC)

-> Download external build components from google Downloading https://commondatastorage.googleapis.com/chromium-browser-clang/Linux_x64/clang-328716-2.tgz .......... Done. ==> Removing existing $pkgdir/ directory... ==> Starting build()... -> Build the Launcher make: Entering directory '/home/solaraquarion/build/chromium-dev/src/chromium-launcher' /home/solaraquarion/build/chromium-dev/chromium-dev/src/chromium-67.0.3396.10/third_party/llvm-build/Release+Asserts/bin/clang -o chromium-dev -D_FORTIFY_SOURCE=2 -DDATE= -DTIME= -DTIMESTAMP= -DNO_UNWIND_TABLES -DLAUNCHER_VERSION=\"v6-2-g0d47114\" -DCHROMIUM_NAME=\"chromium-dev\" -DCHROMIUM_BINARY=\"/usr/lib/chromium-dev/chromium-dev\" -DCHROMIUM_VENDOR="\"Arch Linux\"" -DPEPPER_FLASH_DIR=\"/usr/lib/PepperFlash\" -march=x86-64 -mtune=generic -O2 -pipe -fstack-protector-strong -Wno-builtin-macro-redefined -fno-unwind-tables -fno-asynchronous-unwind-tables -pthread -I/usr/include/json-glib-1.0 -I/usr/include/glib-2.0 -I/usr/lib/glib-2.0/include -Wl,-O1,--sort-common,--as-needed,-z,relro,-z,now launcher.c -ljson-glib-1.0 -lgio-2.0 -lgobject-2.0 -lglib-2.0 /bin/sh: /home/solaraquarion/build/chromium-dev/chromium-dev/src/chromium-67.0.3396.10/third_party/llvm-build/Release+Asserts/bin/clang: No such file or directory make: *** [Makefile:23: chromium-dev] Error 127 make: Leaving directory '/home/solaraquarion/build/chromium-dev/src/chromium-launcher'

prazola commented on 2018-04-09 23:30 (UTC)

You don't need an account to see the video. Try this one: https://www.facebook.com/ESLProLeagueCSGO/videos/2023893594542313/

sl1pkn07 commented on 2018-04-05 13:22 (UTC)

can you paste the link of youtube video? i don't have netflix/facebukakke account

prazola commented on 2018-04-05 12:15 (UTC)

Netflix previews and some youtube and facebook videos.

sl1pkn07 commented on 2018-04-04 04:43 (UTC)

Please post the video source. I never see this before in chromium

prazola commented on 2018-04-04 03:33 (UTC)

Some videos are wrongly decoded and colors are broken. This doesn't happen in precompiled chromium builds, what could it be? http://www.imagebam.com/image/bb5c01807045363

walkindude commented on 2018-04-01 00:53 (UTC)

yes, used system clang (_use_bundled_clang=0), but guess I messed up with llvm/clang packages from svn and not - chroot (arch-nspawn/makechrootpkg) helped and now it works, but not before I clean up whole chroot thing/delete snapshot folder/reboot (not quite sure what really helped), otherwise even with lld installed it just kept telling me about ffmpeg problem already discussed here.

sl1pkn07 commented on 2018-03-31 17:02 (UTC) (edited on 2018-03-31 17:02 (UTC) by sl1pkn07)

you setup the pkgbuild for use system clang, rigth? what version you use?

walkindude commented on 2018-03-31 16:01 (UTC)

https://paste.ee/p/hwjiq

sl1pkn07 commented on 2018-03-31 15:22 (UTC)

more log please

walkindude commented on 2018-03-31 12:47 (UTC) (edited on 2018-03-31 12:53 (UTC) by walkindude)

tried with lld and lld from llvm-svn

[409/411] rm -f libevent.a && ar rcsT libevent.a base/third_party/libevent/buffer.o base/third_party/libevent/evbuffer.o base/third_party/libevent/evdns.o base/third_party/libevent/event.o base/third_party/libevent/event_tagging.o base/third_party/libevent/evrpc.o base/third_party/libevent/evutil.o base/third_party/libevent/http.o base/third_party/libevent/log.o base/third_party/libevent/poll.o base/third_party/libevent/select.o base/third_party/libevent/signal.o base/third_party/libevent/strlcpy.o base/third_party/libevent/epoll.o [410/411] rm -f gn_lib.a && ar rcsT gn_lib.a tools/gn/visibility.o tools/gn/source_dir.o tools/gn/command_help.o tools/gn/scope_per_file_provider.o tools/gn/ninja_copy_target_writer.o tools/gn/parser_fuzzer.o tools/gn/function_get_target_outputs.o tools/gn/builder_record.o tools/gn/ninja_build_writer.o tools/gn/visual_studio_writer.o tools/gn/label_pattern.o tools/gn/function_write_file.o tools/gn/switches.o tools/gn/functions_target.o tools/gn/ninja_create_bundle_target_writer.o tools/gn/settings.o tools/gn/input_conversion.o tools/gn/xml_element_writer.o tools/gn/input_file.o tools/gn/command_desc.o tools/gn/desc_builder.o tools/gn/function_template.o tools/gn/parser.o tools/gn/string_utils.o tools/gn/bundle_data_target_generator.o tools/gn/pool.o tools/gn/function_set_defaults.o tools/gn/function_set_default_toolchain.o tools/gn/function_forward_variables_from.o tools/gn/binary_target_generator.o tools/gn/ninja_target_writer.o tools/gn/item.o tools/gn/runtime_deps.o tools/gn/label.o tools/gn/args.o tools/gn/function_read_file.o tools/gn/c_include_iterator.o tools/gn/tool.o tools/gn/ninja_writer.o tools/gn/ninja_bundle_data_target_writer.o tools/gn/inherited_libraries.o tools/gn/toolchain.o tools/gn/build_settings.o tools/gn/analyzer.o tools/gn/scope.o tools/gn/ninja_toolchain_writer.o tools/gn/qt_creator_writer.o tools/gn/config_values_extractors.o tools/gn/source_file.o tools/gn/value.o tools/gn/bundle_data.o tools/gn/config_values.o tools/gn/bundle_file_rule.o tools/gn/command_ls.o tools/gn/command_format.o tools/gn/ninja_action_target_writer.o tools/gn/input_file_manager.o tools/gn/function_rebase_path.o tools/gn/function_get_path_info.o tools/gn/command_path.o tools/gn/action_target_generator.o tools/gn/ninja_group_target_writer.o tools/gn/builder.o tools/gn/variables.o tools/gn/standard_out.o tools/gn/xcode_object.o tools/gn/substitution_pattern.o tools/gn/ninja_binary_target_writer.o tools/gn/command_check.o tools/gn/tokenizer.o tools/gn/function_foreach.o tools/gn/setup.o tools/gn/template.o tools/gn/function_get_label_info.o tools/gn/function_process_file_template.o tools/gn/deps_iterator.o tools/gn/commands.o tools/gn/command_gen.o tools/gn/operators.o tools/gn/lib_file.o tools/gn/command_analyze.o tools/gn/path_output.o tools/gn/output_file.o tools/gn/substitution_list.o tools/gn/filesystem_utils.o tools/gn/value_extractors.o tools/gn/location.o tools/gn/err.o tools/gn/xcode_writer.o tools/gn/command_clean.o tools/gn/loader.o tools/gn/header_checker.o tools/gn/functions.o tools/gn/import_manager.o tools/gn/function_exec_script.o tools/gn/command_args.o tools/gn/source_file_type.o tools/gn/token.o tools/gn/target.o tools/gn/ninja_utils.o tools/gn/config.o tools/gn/target_generator.o tools/gn/parse_tree.o tools/gn/visual_studio_utils.o tools/gn/test_with_scope.o tools/gn/escape.o tools/gn/eclipse_writer.o tools/gn/exec_process.o tools/gn/pattern.o tools/gn/substitution_writer.o tools/gn/command_refs.o tools/gn/group_target_generator.o tools/gn/json_project_writer.o tools/gn/copy_target_generator.o tools/gn/create_bundle_target_generator.o tools/gn/scheduler.o tools/gn/config_values_generator.o tools/gn/action_values.o tools/gn/parse_node_value_adapter.o tools/gn/function_toolchain.o tools/gn/substitution_type.o tools/gn/trace.o [411/411] clang++ -Wl,-O1,--sort-common,--as-needed,-z,relro,-lm,-lpthread -nodefaultlibs -o gn -Wl,--start-group tools/gn/gn_main.o libc++.a xdg_user_dirs.a libc++abi.a base.a gn_lib.a libevent.a dynamic_annotations.a -Wl,--end-group -lc -lgcc_s -lm -lpthread -lrt -latomic FAILED: gn clang++ -Wl,-O1,--sort-common,--as-needed,-z,relro,-lm,-lpthread -nodefaultlibs -o gn -Wl,--start-group tools/gn/gn_main.o libc++.a xdg_user_dirs.a libc++abi.a base.a gn_lib.a libevent.a dynamic_annotations.a -Wl,--end-group -lc -lgcc_s -lm -lpthread -lrt -latomic tools/gn/gn_main.o: file not recognized: File format not recognized clang-7: error: linker command failed with exit code 1 (use -v to see invocation) ninja: build stopped: subcommand failed. Command '['ninja', '-C', '/home/walkindude/yaourt/yaourt-tmp-walkindude/aur-chromium-dev/src/chromium-67.0.3381.0/out_bootstrap', '-w', 'dupbuild=err', '-v', 'gn']' returned non-zero exit status 1 ==> ERROR: A failure occurred in build(). Aborting... ==> ERROR: Makepkg was unable to build chromium-dev. ==> Restart building chromium-dev ? [y/N] ==> ------------------------------------- ==>

sl1pkn07 commented on 2018-03-30 16:05 (UTC)

i'm not sure if is better use lld or disable it in the build environment with 'use_lld=false'

any thoughts?

laenco commented on 2018-03-30 16:02 (UTC)

yep, it's completely ok now. Thank you twice!)

sl1pkn07 commented on 2018-03-30 16:00 (UTC)

try now

laenco commented on 2018-03-30 15:58 (UTC) (edited on 2018-03-30 15:59 (UTC) by laenco)

UPD: Sorry, found and fixed the issue. The lld package suddenly was not installed, nor present in chromium requirements.

Thank you for help!)

laenco commented on 2018-03-30 15:51 (UTC) (edited on 2018-03-30 15:51 (UTC) by laenco)

Got exactly the same error

$ /tmp/makepkg/chromium-dev/src/chromium-67.0.3381.0/third_party/ffmpeg/configure ......
clang is unable to create an executable file.
If clang is a cross-compiler, use the --enable-cross-compile option.
Only do this if you know what cross compiling means.
C compiler test failed.

If you think configure made a mistake, make sure you are using the latest
version from Git.  If the latest version fails, report the problem to the
ffmpeg-user@ffmpeg.org mailing list or IRC #ffmpeg on irc.freenode.net.
Include the log file "ffbuild/config.log" produced by configure as this will help
solve the problem.

But I've found that error could be avoided by removing --extra-ldflags="-fuse-ld=lld"

I'll try to compile this way.

sl1pkn07 commented on 2018-03-30 15:18 (UTC) (edited on 2018-03-30 15:21 (UTC) by sl1pkn07)

can you run this by hand?

/tmp/makepkg/chromium-dev/src/chromium-67.0.3381.0/third_party/ffmpeg/configure --disable-everything --disable-all --disable-doc --disable-htmlpages --disable-manpages --disable-podpages --disable-txtpages --disable-static --enable-avcodec --enable-avformat --enable-avutil --enable-fft --enable-rdft --enable-static --enable-libopus --disable-bzlib --disable-iconv --disable-lzo --disable-network --disable-schannel --disable-sdl2 --disable-symver --disable-xlib --disable-zlib --disable-securetransport --disable-faan --disable-alsa --disable-autodetect --enable-decoder=vorbis,libopus,flac --enable-decoder=pcm_u8,pcm_s16le,pcm_s24le,pcm_s32le,pcm_f32le,mp3 --enable-decoder=pcm_s16be,pcm_s24be,pcm_mulaw,pcm_alaw --enable-demuxer=ogg,matroska,wav,flac,mp3,mov --enable-parser=opus,vorbis,flac,mpegaudio --extra-cflags="-I/usr/include/opus" --disable-linux-perf --x86asmexe=yasm --optflags="-O2" --enable-decoder=theora,vp8 --enable-parser=vp3,vp8 --enable-lto --enable-pic --cc=clang --cxx=clang++ --ld=clang --extra-ldflags="-fuse-ld=lld" --enable-decoder=aac,h264 --enable-demuxer=aac --enable-parser=aac,h264 --enable-decoder=mpeg4 --enable-parser=h263,mpeg4video --enable-demuxer=avi --enable-demuxer=amr --enable-decoder=amrnb,amrwb --enable-decoder=gsm_ms --enable-parser=gsm

laenco commented on 2018-03-30 07:11 (UTC) (edited on 2018-03-30 07:12 (UTC) by laenco)

Unable to compile latest change:

 -> Configure bundled ffmpeg
System information:
Host OS       : linux
Target OS     : linux
Host arch     : x64
Target arch   : x64
Parallel jobs : 8

Chromium skipped
Chrome skipped
ChromeOS configure/build:
Running /tmp/makepkg/chromium-dev/src/chromium-67.0.3381.0/third_party/ffmpeg/configure
 --disable-everything
 --disable-all
 --disable-doc
 --disable-htmlpages
 --disable-manpages
 --disable-podpages
 --disable-txtpages
 --disable-static
 --enable-avcodec
 --enable-avformat
 --enable-avutil
 --enable-fft
 --enable-rdft
 --enable-static
 --enable-libopus
 --disable-bzlib
 --disable-iconv
 --disable-lzo
 --disable-network
 --disable-schannel
 --disable-sdl2
 --disable-symver
 --disable-xlib
 --disable-zlib
 --disable-securetransport
 --disable-faan
 --disable-alsa
 --disable-autodetect
 --enable-decoder=vorbis,libopus,flac
 --enable-decoder=pcm_u8,pcm_s16le,pcm_s24le,pcm_s32le,pcm_f32le,mp3
 --enable-decoder=pcm_s16be,pcm_s24be,pcm_mulaw,pcm_alaw
 --enable-demuxer=ogg,matroska,wav,flac,mp3,mov
 --enable-parser=opus,vorbis,flac,mpegaudio
 --extra-cflags=-I/usr/include/opus
 --disable-linux-perf
 --x86asmexe=yasm
 --optflags="-O2"
 --enable-decoder=theora,vp8
 --enable-parser=vp3,vp8
 --enable-lto
 --enable-pic
 --cc=clang
 --cxx=clang++
 --ld=clang
 --extra-ldflags=-fuse-ld=lld
 --enable-decoder=aac,h264
 --enable-demuxer=aac
 --enable-parser=aac,h264
 --enable-decoder=mpeg4
 --enable-parser=h263,mpeg4video
 --enable-demuxer=avi
 --enable-demuxer=amr
 --enable-decoder=amrnb,amrwb
 --enable-decoder=gsm_ms
 --enable-parser=gsm
clang is unable to create an executable file.
If clang is a cross-compiler, use the --enable-cross-compile option.
Only do this if you know what cross compiling means.
C compiler test failed.

If you think configure made a mistake, make sure you are using the latest
version from Git.  If the latest version fails, report the problem to the
ffmpeg-user@ffmpeg.org mailing list or IRC #ffmpeg on irc.freenode.net.
Include the log file "ffbuild/config.log" produced by configure as this will help
solve the problem.
Traceback (most recent call last):
  File "chromium/scripts/build_ffmpeg.py", line 769, in <module>
    sys.exit(main(sys.argv[1:]))
  File "chromium/scripts/build_ffmpeg.py", line 347, in main
    configure_args, options=options)
  File "chromium/scripts/build_ffmpeg.py", line 761, in ConfigureAndBuild
    do_build_ffmpeg('ChromeOS', chrome_os_flags)
  File "chromium/scripts/build_ffmpeg.py", line 741, in do_build_ffmpeg
    options.config_only, branding, configure_flags)
  File "chromium/scripts/build_ffmpeg.py", line 224, in BuildFFmpeg
    [os.path.join(FFMPEG_DIR, 'configure')] + configure_flags, cwd=config_dir)
  File "chromium/scripts/build_ffmpeg.py", line 93, in PrintAndCheckCall
    subprocess.check_call(argv, *args, **kwargs)
  File "/usr/lib/python2.7/subprocess.py", line 186, in check_call
    raise CalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command '['/tmp/makepkg/chromium-dev/src/chromium-67.0.3381.0/third_party/ffmpeg/configure', '--disable-everything', '--disable-all', '--disable-doc', '--disable-htmlpages', '--disable-manpages', '--disable-podpages', '--disable-txtpages', '--disable-static', '--enable-avcodec', '--enable-avformat', '--enable-avutil', '--enable-fft', '--enable-rdft', '--enable-static', '--enable-libopus', '--disable-bzlib', '--disable-iconv', '--disable-lzo', '--disable-network', '--disable-schannel', '--disable-sdl2', '--disable-symver', '--disable-xlib', '--disable-zlib', '--disable-securetransport', '--disable-faan', '--disable-alsa', '--disable-autodetect', '--enable-decoder=vorbis,libopus,flac', '--enable-decoder=pcm_u8,pcm_s16le,pcm_s24le,pcm_s32le,pcm_f32le,mp3', '--enable-decoder=pcm_s16be,pcm_s24be,pcm_mulaw,pcm_alaw', '--enable-demuxer=ogg,matroska,wav,flac,mp3,mov', '--enable-parser=opus,vorbis,flac,mpegaudio', '--extra-cflags=-I/usr/include/opus', '--disable-linux-perf', '--x86asmexe=yasm', '--optflags="-O2"', '--enable-decoder=theora,vp8', '--enable-parser=vp3,vp8', '--enable-lto', '--enable-pic', '--cc=clang', '--cxx=clang++', '--ld=clang', '--extra-ldflags=-fuse-ld=lld', '--enable-decoder=aac,h264', '--enable-demuxer=aac', '--enable-parser=aac,h264', '--enable-decoder=mpeg4', '--enable-parser=h263,mpeg4video', '--enable-demuxer=avi', '--enable-demuxer=amr', '--enable-decoder=amrnb,amrwb', '--enable-decoder=gsm_ms', '--enable-parser=gsm']' returned non-zero exit status 1

collapse block don't work here (>_<)' : sorry for long log.

0ri2N commented on 2018-03-22 20:12 (UTC)

Repository: https://github.com/archlinuxcn/repo/blob/master/README.md

francoism90 commented on 2018-03-18 17:52 (UTC)

Any repo available? I have tried to build the package but it takes hours for me.

francoism90 commented on 2018-03-18 10:18 (UTC)

Thanks! :)

sl1pkn07 commented on 2018-03-17 20:36 (UTC)

done

francoism90 commented on 2018-03-17 20:15 (UTC)

==> Validating source files with sha256sums... chromium-66.0.3350.0.tar.xz ... Passed chromium-launcher ... Skipped chromium-dev.svg ... Passed chromium-FORTIFY_SOURCE-r2.patch ... Passed chromium-clang-r2.patch ... Passed chromium-intel-vaapi_r16.diff.base64 ... FAILED chromium-widevine-r1.patch ... Passed chromium-exclude_unwind_tables_r3.patch ... Passed

Hmm, I'm I the only one getting the validate error?

bm456 commented on 2018-02-02 19:12 (UTC) (edited on 2018-02-02 19:13 (UTC) by bm456)

@prazola

may be a segfault?

please read this (use lesser jobs) http://www.linuxfromscratch.org/blfs/view/svn/xsoft/chromium.html

recomended: 1.6gb per thread

sl1pkn07 commented on 2018-01-27 17:53 (UTC)

@misc that patch is knowed by upstream?

misc commented on 2018-01-27 00:20 (UTC) (edited on 2018-01-27 12:30 (UTC) by misc)

Haven't followed the AUR changes for this package in so long, here's my patches if they're of any use. All apply to chromium git.

  1. https://pastebin.com/p1eBiP9n General build fixes, the one to media/base/decoder_buffer.h is required with next ffmpeg

  2. https://pastebin.com/r02Y4UaG Undoes freetype & harfbuzz being combined to one target, with this it's possible to use system libs for both again.

  3. https://pastebin.com/VB7zXAfM Working VAAPI patch. Somehow GPU usage no longer shows for me in task manager, doubt that's related though.

prazola commented on 2018-01-22 13:17 (UTC)

Compiled, but says this at runtime: ../../third_party/tcmalloc/chromium/src/tcmalloc.cc:289] Attempt to free invalid pointer 0x5596ee2da030

bm456 commented on 2018-01-13 19:35 (UTC)

pd

can you add an alert (at nijna line) about possible oom-related troubles?

just like this warning (at my benchmarks, with an amd fx cpu i estimate it at >1.46GB per job), they say 1.25 at their warning because they're using gcc instead of google's clang trunk (clang 6)

http://www.linuxfromscratch.org/blfs/view/svn/xsoft/chromium.html

feel free to use my alert https://aur.archlinux.org/cgit/aur.git/commit/?h=inox-hard&id=1c7db197a086

bm456 commented on 2018-01-09 23:22 (UTC)

somebody failed again lol

https://bugs.chromium.org/p/chromium/issues/detail?id=800553

sl1pkn07 commented on 2017-12-30 21:54 (UTC)

that is because you add zero logs of the problem, only the solution

bm456 commented on 2017-12-30 21:43 (UTC) (edited on 2017-12-30 21:51 (UTC) by bm456)

yeah, but they prefer make "jokes", or just ignore it (since agust many peple reported some bugs if you disable webrtc and google ignored it)

just like "please give me your chrome://crash log"

ehen its a build issue, not a run issue https://bugs.chromium.org/p/chromium/issues/detail?id=797448

sl1pkn07 commented on 2017-12-30 21:31 (UTC) (edited on 2017-12-30 21:33 (UTC) by sl1pkn07)

What do they do?

i only support official patches (and fix for the patches iself)

bm456 commented on 2017-12-30 21:27 (UTC)

i report the issues, and google (sometimes) fix it

but if google doesnt fix it, i mantain the patches

example

its your package, with my patches (only the 4 mandatory patches, the rest of patches are my features)

https://p.teknik.io/Raw/WfNkb

sl1pkn07 commented on 2017-12-30 21:22 (UTC)

these patch is knowed by upstream?

bm456 commented on 2017-12-30 21:20 (UTC) (edited on 2017-12-30 21:21 (UTC) by bm456)

no

but i'm the mantainer of a chromium-deblobbed version (if offers a deblobbed chromium-based browser, and has stable, beta and dev channels)

and i need patch it, its just share patches (patches that fixes google errors)

just like https://raw.githubusercontent.com/bn0785ac/in-dev/master/narnia4.patch

and i could help you to mantain it

sl1pkn07 commented on 2017-12-25 21:10 (UTC) (edited on 2017-12-25 21:11 (UTC) by sl1pkn07)

why? are you crhromium developer?

bm456 commented on 2017-12-25 20:46 (UTC)

@sl1pkn07 please add me as comantainer

sl1pkn07 commented on 2017-12-21 23:27 (UTC)

@bm456

please send you patches to upstream

Greetings

bm456 commented on 2017-12-13 19:01 (UTC) (edited on 2017-12-15 23:06 (UTC) by bm456)

i've tested my chromium-dev "deblobbed" with imagemagick 7

and works fine

PD my patches (narnia1 and gnb are still useful for 64.0.3282.24)

EDIT: Builds fine with ImageMagick7

EDIT: If somebody wants chromium 65 i'll upload it at some hours at inox-dev

bm456 commented on 2017-11-30 02:14 (UTC)

if you need i've the patches for 64.0.3278.0 https://raw.githubusercontent.com/bn0785ac/in-dev/master/gnb.patch https://raw.githubusercontent.com/bn0785ac/in-dev/master/narnia1.patch gnb.patch fixes an mojom error narnia1.patch fixes a human upstream error

xuiqzy commented on 2017-11-15 13:49 (UTC)

Computing //:All(//build/toolchain/linux:clang_x64) Computing //chrome/test/media_router:media_router_tests(//build/toolchain/linux:clang_x64) Done. Made 7437 targets from 1419 files in 3824ms ninja: Entering directory `out/Release' ninja: error: '../../native_client/toolchain/linux_x86/pnacl_newlib/bin/x86_64-nacl-objcopy', needed by 'irt_x64/obj/ppapi/native_client/nacl_irt_debuglink.inputdeps.stamp', missing and no known rule to make it ==> ERROR: A failure occurred in build(). Aborting... Also it took several tries until the download from commondatastorage.googleapis.com didn't abort!

sl1pkn07 commented on 2017-10-19 14:58 (UTC)

https://git.archlinux.org/svntogit/packages.git/plain/trunk/chromium-gcc-r1.patch?h=packages/chromium&id=58e831f91d201ccf0634862842661efa72500e6b

JoFa commented on 2017-10-19 14:44 (UTC)

==> ERROR: Failure while downloading https://git.archlinux.org/svntogit/packages.git/plain/trunk/chromium-gcc-r1.patch?h=packages/chromium Aborting...

sl1pkn07 commented on 2017-10-09 13:26 (UTC)

IOError: [Errno 28] No space left on device try to change the BUILDDIR variable in /etc/makepkg.conf

zfkerr commented on 2017-10-09 13:19 (UTC) (edited on 2017-10-09 13:21 (UTC) by zfkerr)

@sl1pkn07, I can't to build this package: .................................................. INFO: Extracting unsandboxed_runtime_arm_linux.tgz (50/52) |------------------------------------------------| .................................................. INFO: Extracting unsandboxed_runtime_x86_32_linux.tgz (51/52) |------------------------------------------------| .................................................. INFO: Extracting unsandboxed_runtime_x86_64_linux.tgz (52/52) |------------------------------------------------| .................................................. INFO: Extracting package (pnacl_translator) to directory: /home/user/chromium-dev/src/chromium-63.0.3230.0/native_client/toolchain/linux_x86/pnacl_translator INFO: Extracting sandboxed_translators.tgz (1/1) |------------------------------------------------| .................................................. Updating Clang to 313786-1... Creating directory /home/user/chromium-dev/src/chromium-63.0.3230.0/third_party/llvm-build Downloading prebuilt clang Downloading https://commondatastorage.googleapis.com/chromium-browser-clang/Linux_x64/clang-313786-1.tgz Traceback (most recent call last): File "tools/clang/scripts/update.py", line 868, in <module> sys.exit(main()) File "tools/clang/scripts/update.py", line 864, in main return UpdateClang(args) File "tools/clang/scripts/update.py", line 414, in UpdateClang DownloadAndUnpack(cds_full_url, LLVM_BUILD_DIR) File "tools/clang/scripts/update.py", line 138, in DownloadAndUnpack DownloadUrl(url, f) File "tools/clang/scripts/update.py", line 108, in DownloadUrl output_file.write(chunk) IOError: [Errno 28] No space left on device ==> ERROR: A failure occurred in prepare(). Aborting...

laenco commented on 2017-09-28 14:49 (UTC)

all ok, thank you!

sl1pkn07 commented on 2017-09-28 14:45 (UTC)

try now

laenco commented on 2017-09-28 14:42 (UTC)

Oops 'chromium-gn-bootstrap-r20.patch.base64::https://chromium-review.googlesource.com/changes/686415/revisions/8ddf0976e671bc0aad990378e7676d3397b28e3d/patch?downloadbase' "--downloadbase" is not a valid option Seems it should be: 'chromium-gn-bootstrap-r20.patch.base64::https://chromium-review.googlesource.com/changes/686415/revisions/8ddf0976e671bc0aad990378e7676d3397b28e3d/patch?download'

sl1pkn07 commented on 2017-09-13 20:41 (UTC)

yes, you can edit the pkgbuild manually to update the max-res patch sums and no at the moment. i'm waiting to solve this error https://bugs.chromium.org/p/chromium/issues/detail?id=763650 and use the new release (63.0.3213.3) greetings

CaTa commented on 2017-09-13 20:37 (UTC)

Thanks for responding, So the sha256sums array in this repository needs to update the current entry ['69958012fa3af965ce15a5d108e912e18acdb7c577ed6b53ec179624dfc4d1e3'] to reflect the correct sha2656 has of the new patch found on https://github.com/saiarcot895/chromium-ubuntu-build/blob/master/debian/patches/specify-max-resolution.patch? Is that something I can help with?

sl1pkn07 commented on 2017-09-13 17:33 (UTC)

Seems the max resolution patch is updated in upstream

CaTa commented on 2017-09-13 15:27 (UTC)

Hi All, I'm getting a checksum failure on the "specify-max-resolution.patch" as shown below. I tried installing on a fresh pc and it also had the same issue. Is this something I'm doing wrong or a configuration issue on my side? ==> Validating source files with sha256sums... chromium-62.0.3198.0.tar.xz ... Passed chromium-launcher ... Skipped chromium-dev.svg ... Passed chromium-FORTIFY_SOURCE-r2.patch ... Passed chromium-gn-bootstrap-r17.patch ... Passed chromium-gcc5-r2.patch ... Passed minizip.patch ... Passed chromium-intel-vaapi_r2.patch ... Passed chromium-intel-vaapi-fix.patch ... Passed specify-max-resolution.patch ... FAILED breakpad-use-ucontext_t.patch ... Passed chromium-gcc-r1.patch ... Passed chromium-blink-gcc7-r2.patch ... Passed chromium-widevine-r1.patch ... Passed ==> ERROR: One or more files did not pass the validity check!

nullie commented on 2017-09-06 18:17 (UTC)

Okay, probably you're right. I was using pacaur.

nullie commented on 2017-09-06 18:16 (UTC)

Okay, probably you're right. I was using pacaur.

sl1pkn07 commented on 2017-09-06 14:20 (UTC) (edited on 2017-09-06 14:21 (UTC) by sl1pkn07)

right python path is exported in prepare() function https://aur.archlinux.org/cgit/aur.git/tree/PKGBUILD?h=chromium-dev#n384 you have something wrong in your pacman helper/installation try with plain makepkg

nullie commented on 2017-09-06 14:13 (UTC)

I've made a fix From 3c9a4a661b29041fabe51482a244486542b31438 Mon Sep 17 00:00:00 2001 From: Ilia Novoselov <ilya.novoselov@gmail.com> Date: Wed, 6 Sep 2017 19:11:31 +0500 Subject: [PATCH] Fix PNaCl python version --- PKGBUILD | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/PKGBUILD b/PKGBUILD index 04113e3..8cc69bd 100644 --- a/PKGBUILD +++ b/PKGBUILD @@ -530,7 +530,7 @@ build() { out/Release/gn gen out/Release -v --args="${_flags[*]} ${_debug_flag}" --script-executable=/usr/bin/python2 # Build all with ninja. - LC_ALL=C ninja -C out/Release -v pdf chrome chrome_sandbox chromedriver widevinecdmadapter clearkeycdm + LC_ALL=C PNACLPYTHON=python2 ninja -C out/Release -v pdf chrome chrome_sandbox chromedriver widevinecdmadapter clearkeycdm } package() { -- 2.14.1

nullie commented on 2017-09-06 07:33 (UTC)

I've got an error trying to build 62.0.3192.0-1. Does it look like my local issue? The problem seems that wrong python is used, it's common for many builds that they expect python to be python2, not python3. [30959/32706] ../../native_client/toolchain/linux_x86/pnacl_newlib/bin/pnacl-clang -MMD -MF newlib_pnacl_nonsfi/obj/base/third_party/libevent/libevent/evutil.o.d -DNACL_TC_REV=efa3f5d8ef135ed2463a75ac4630d1c448021400 -DHAVE_CONFIG_H -DV8_DEPRECATION_WARNINGS -DUSE_OPENSSL_CERTS=1 -DNO_TCMALLOC -DFULL_SAFE_BROWSING -DSAFE_BROWSING_CSD -DSAFE_BROWSING_DB_LOCAL -DCHROMIUM_BUILD -DFIELDTRIAL_TESTING_ENABLED -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -DNDEBUG -DNVALGRIND -I../../base/third_party/libevent/nacl_nonsfi -I../../native_client/src/public/linux_syscalls -I../.. -Inewlib_pnacl_nonsfi/gen -fno-strict-aliasing -Wno-builtin-macro-redefined -D__DATE__= -D__TIME__= -D__TIMESTAMP__= -fcolor-diagnostics --pnacl-allow-translate -arch x86-32-nonsfi --pnacl-bias=x86-32-nonsfi --target=i686-unknown-nacl -O2 -fno-ident -fdata-sections -ffunction-sections -fomit-frame-pointer -g0 -fvisibility=hidden -Wno-unused-variable -Wno-missing-field-initializers -Wno-unused-parameter -Wno-c++11-narrowing -Wno-covered-switch-default -Wno-unneeded-internal-declaration -Wno-inconsistent-missing-override -std=gnu11 -c ../../base/third_party/libevent/evutil.c -o newlib_pnacl_nonsfi/obj/base/third_party/libevent/libevent/evutil.o FAILED: newlib_pnacl_nonsfi/obj/base/third_party/libevent/libevent/evutil.o ../../native_client/toolchain/linux_x86/pnacl_newlib/bin/pnacl-clang -MMD -MF newlib_pnacl_nonsfi/obj/base/third_party/libevent/libevent/evutil.o.d -DNACL_TC_REV=efa3f5d8ef135ed2463a75ac4630d1c448021400 -DHAVE_CONFIG_H -DV8_DEPRECATION_WARNINGS -DUSE_OPENSSL_CERTS=1 -DNO_TCMALLOC -DFULL_SAFE_BROWSING -DSAFE_BROWSING_CSD -DSAFE_BROWSING_DB_LOCAL -DCHROMIUM_BUILD -DFIELDTRIAL_TESTING_ENABLED -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -DNDEBUG -DNVALGRIND -I../../base/third_party/libevent/nacl_nonsfi -I../../native_client/src/public/linux_syscalls -I../.. -Inewlib_pnacl_nonsfi/gen -fno-strict-aliasing -Wno-builtin-macro-redefined -D__DATE__= -D__TIME__= -D__TIMESTAMP__= -fcolor-diagnostics --pnacl-allow-translate -arch x86-32-nonsfi --pnacl-bias=x86-32-nonsfi --target=i686-unknown-nacl -O2 -fno-ident -fdata-sections -ffunction-sections -fomit-frame-pointer -g0 -fvisibility=hidden -Wno-unused-variable -Wno-missing-field-initializers -Wno-unused-parameter -Wno-c++11-narrowing -Wno-covered-switch-default -Wno-unneeded-internal-declaration -Wno-inconsistent-missing-override -std=gnu11 -c ../../base/third_party/libevent/evutil.c -o newlib_pnacl_nonsfi/obj/base/third_party/libevent/libevent/evutil.o File "../../native_client/toolchain/linux_x86/pnacl_newlib/bin/pydir/loader.py", line 12 The environment variable PNACLPYTHON can override the python found in PATH''' ^ SyntaxError: Missing parentheses in call to 'print'

sl1pkn07 commented on 2017-08-15 12:27 (UTC)

is for making a selector for active the vaapi patch if detect the intel and AMD (and if nouveau is supported) drivers

xsmile commented on 2017-08-15 12:21 (UTC)

Don't know about nouveau.

sl1pkn07 commented on 2017-08-15 12:03 (UTC)

also with nouveau?

xsmile commented on 2017-08-15 11:56 (UTC) (edited on 2017-08-15 12:00 (UTC) by xsmile)

It would have been too good to be true if it worked without issues. I'm not up-to-date about the current status of vaapi_patch_r2.patch but the official commit definitely works with Intel and apparently helped an AMD user use VA-API while it didn't work with vaapi_patch_r2.patch. Now we know it causes issues with NVIDIA. EDIT: It would be great to collect reports and forward them to Daniel Charles (Google) while the official patchset is still being reviewed.

sl1pkn07 commented on 2017-08-15 10:13 (UTC)

with the vaapi patch https://sl1pkn07.wtf/paste/view/511741b3 :/

xsmile commented on 2017-08-14 18:06 (UTC) (edited on 2017-08-14 18:40 (UTC) by xsmile)

@sl1pkn07: Got it, I was missing the custom_toolchain flag. This way the compiler options are appended to the default ones set by GN and should override them, if necessary.

sl1pkn07 commented on 2017-08-14 16:45 (UTC) (edited on 2017-08-14 16:47 (UTC) by sl1pkn07)

└───╼ cat /etc/makepkg.conf | grep -e CFLAGS -e CXXFLAGS CFLAGS="-march=native -O2 -pipe -fstack-protector-strong -fno-plt" CXXFLAGS="-march=native -O2 -pipe -fstack-protector-strong -fno-plt" DEBUG_CFLAGS="-g -fvar-tracking-assignments" DEBUG_CXXFLAGS="-g -fvar-tracking-assignments" [106/32501] /usr/bin/clang++ -Qunused-arguments -MMD -MF obj/base/base/profiler.o.d -DUSE_SYMBOLIZE -DV8_DEPRECATION_WARNINGS -DUSE_UDEV -DUSE_AURA=1 -DUSE_PANGO=1 -DUSE_CAIRO=1 -DUSE_GLIB=1 -DUSE_NSS_CERTS=1 -DUSE_X11=1 -DFULL_SAFE_BROWSING -DSAFE_BROWSING_CSD -DSAFE_BROWSING_DB_LOCAL -DCHROMIUM_BUILD -DFIELDTRIAL_TESTING_ENABLED -DCR_CLANG_REVISION=\"308728-3\" -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -DNDEBUG -DNVALGRIND -DDYNAMIC_ANNOTATIONS_ENABLED=0 -DBASE_IMPLEMENTATION -DGLIB_VERSION_MAX_ALLOWED=GLIB_VERSION_2_32 -DGLIB_VERSION_MIN_REQUIRED=GLIB_VERSION_2_26 -I../.. -Igen -I/usr/include/glib-2.0 -I/usr/lib/glib-2.0/include -fno-strict-aliasing --param=ssp-buffer-size=4 -fstack-protector -Wno-builtin-macro-redefined -D__DATE__= -D__TIME__= -D__TIMESTAMP__= -funwind-tables -fPIC -pipe -pthread -fcolor-diagnostics -m64 -march=x86-64 -Wall -Wextra -Wno-missing-field-initializers -Wno-unused-parameter -Wno-c++11-narrowing -Wno-covered-switch-default -Wno-unneeded-internal-declaration -Wno-inconsistent-missing-override -Wno-undefined-var-template -Wno-nonportable-include-path -Wno-address-of-packed-member -Wno-unused-lambda-capture -Wno-user-defined-warnings -fomit-frame-pointer -g0 -fvisibility=hidden -Wheader-hygiene -Wstring-conversion -Wtautological-overlap-compare -Wno-char-subscripts -Wexit-time-destructors -Wexit-time-destructors -O2 -fno-ident -fdata-sections -ffunction-sections -std=gnu++14 -fno-rtti -fno-exceptions -fvisibility-inlines-hidden -D_FORTIFY_SOURCE=2 -march=native -O2 -pipe -fstack-protector-strong -Wno-unknown-warning-option -c ../../base/debug/profiler.cc -o obj/base/base/profiler.o try with this (ugly) patch (for the PKGBUILD) https://sl1pkn07.wtf/paste/view/51cf7227 good luck!

xsmile commented on 2017-08-14 16:23 (UTC)

@sl1pkn07: The flags are used during GN compilation but not in the next compilation stage. For instance -march=native defaults to -march=x86-64, regardless of the settings in /etc/makepkg.conf. Thanks for testing btw.

sl1pkn07 commented on 2017-08-14 16:15 (UTC)

tnx now in testing also test a ugly hack for this https://crbug.com/382487

xsmile commented on 2017-08-14 16:13 (UTC)

@sl1pkn07: https://sl1pkn07.wtf/paste/view/918a2c94

sl1pkn07 commented on 2017-08-14 16:08 (UTC) (edited on 2017-08-14 16:08 (UTC) by sl1pkn07)

please use another downloadeable pastebin, for example https://sl1pkn07.wtf/paste

xsmile commented on 2017-08-14 15:48 (UTC)

@sl1pkn07: Use this patch for VA-API: https://0bin.net/paste/hIETCNLbL3hbqxln#MLskYbo-lMWhren2mQa5fo6Q0hI4Aakz9SlJUSsJ5Pf System Clang is used, this is not the issue. 64-bit and I already tried it. Environment variables are ignored, including CFLAGS, CXXFLAGS, CC and CXX.

sl1pkn07 commented on 2017-08-14 14:30 (UTC) (edited on 2017-08-14 14:31 (UTC) by sl1pkn07)

- for missing -i in L#463 and L#464, yes, do'h - for vaapi patch,this patchset can't apply in this version, maybe in the next release - about clang, 32 or 64 bits? is is 64bits, try edit the L#483 and set the system clang path. but i can't make sure it works

xsmile commented on 2017-08-14 13:31 (UTC)

@sl1pkn07: The two GCC patches are missing a "-i" before the file path and pause makepkg when disabling Clang. To enable VA-API with the current dev build, use the latest patchset from https://chromium-review.googlesource.com/c/532294 instead of vaapi_patch_r2.patch, it should apply without conflicts. Additionally 'use_vaapi=true' needs to be added to the flags (already present). See the comment section of https://aur.archlinux.org/packages/chromium-vaapi/ for success reports. Another issue - I tried building with 'clang_base_path="/usr"' to use system clang, but environment variables like CFLAGS are ignored. Is there any way to modify the variables while building with system clang without patching the source?

sl1pkn07 commented on 2017-08-13 07:32 (UTC)

never, i think

nicman23 commented on 2017-08-12 16:12 (UTC)

isn't vaapi in master?

sl1pkn07 commented on 2017-08-05 15:46 (UTC)

vaapi patch needs update https://chromium-review.googlesource.com/c/569529

trialuser commented on 2017-08-05 03:30 (UTC)

vaapi patch fails patching file build/config/compiler/BUILD.gn Hunk #1 succeeded at 1217 (offset 4 lines). patching file tools/gn/bootstrap/bootstrap.py patching file content/browser/accessibility/browser_accessibility_auralinux.cc patching file chrome/browser/about_flags.cc Hunk #1 succeeded at 1479 (offset 435 lines). patching file content/common/BUILD.gn Hunk #1 succeeded at 26 with fuzz 1 (offset 13 lines). Hunk #2 succeeded at 585 with fuzz 1 (offset 17 lines). patching file content/common/sandbox_linux/bpf_gpu_policy_linux.cc Hunk #1 succeeded at 25 with fuzz 1. Hunk #2 succeeded at 42 with fuzz 2. Hunk #5 succeeded at 322 (offset -7 lines). Hunk #6 succeeded at 365 (offset -7 lines). patching file content/common/va_wayland.sigs patching file content/gpu/BUILD.gn Hunk #1 succeeded at 103 (offset 11 lines). patching file content/gpu/gpu_main.cc Hunk #1 succeeded at 84 (offset -1 lines). Hunk #2 succeeded at 135 (offset -6 lines). patching file content/public/common/content_switches.cc Hunk #1 succeeded at 997 (offset -3 lines). patching file content/public/common/content_switches.h Hunk #1 succeeded at 283 (offset -8 lines). patching file gpu/config/software_rendering_list.json Hunk #1 succeeded at 371 (offset -24 lines). patching file media/BUILD.gn Hunk #1 FAILED at 396. Hunk #2 succeeded at 318 with fuzz 1 (offset -349 lines). 1 out of 2 hunks FAILED -- saving rejects to file media/BUILD.gn.rej patching file media/gpu/BUILD.gn Hunk #2 FAILED at 212. Hunk #3 FAILED at 221. Hunk #4 succeeded at 285 (offset 26 lines). 2 out of 4 hunks FAILED -- saving rejects to file media/gpu/BUILD.gn.rej patching file media/gpu/gpu_video_decode_accelerator_factory.cc Hunk #2 succeeded at 87 (offset 1 line). Hunk #3 succeeded at 140 (offset 1 line). Hunk #4 succeeded at 221 (offset -1 lines). patching file media/gpu/gpu_video_decode_accelerator_factory.h Hunk #1 succeeded at 105 (offset 4 lines). patching file media/gpu/ipc/service/BUILD.gn Hunk #1 succeeded at 50 (offset 3 lines). patching file media/gpu/ipc/service/gpu_video_decode_accelerator.cc Hunk #2 succeeded at 175 (offset 3 lines). patching file media/gpu/va_stub_header.fragment patching file media/gpu/vaapi_video_decode_accelerator.cc Hunk #1 succeeded at 364 (offset 6 lines). Hunk #2 succeeded at 384 (offset 6 lines). Hunk #3 succeeded at 400 (offset 6 lines). patching file media/gpu/vaapi_wrapper.cc Hunk #1 succeeded at 190 (offset 2 lines). Hunk #2 succeeded at 342 (offset 2 lines). Hunk #3 succeeded at 411 (offset 2 lines). Hunk #4 succeeded at 435 (offset 2 lines).

misc commented on 2017-06-16 22:46 (UTC)

You can switch to system freetype next build. Reason for the corruption I saw is this freetype commit… https://git.savannah.gnu.org/cgit/freetype/freetype2.git/commit/?id=ab2599ea3f09ba8da4f50b877021d23241d22609 … which will only be included in the next freetype version, by which time Skia should have been updated. Saves some 4 MiB.

commented on 2017-06-15 19:56 (UTC)

hours later, build is successful and package updated. yeah i had to install them during the build process, openh264,ncurses and sandbox.

sl1pkn07 commented on 2017-06-15 16:35 (UTC)

because you need install the package also into the chroot durning build

commented on 2017-06-15 16:31 (UTC)

i have the package installed yet its still throwing that error

sl1pkn07 commented on 2017-06-15 16:19 (UTC)

https://aur.archlinux.org/packages/openh264/ (?)

commented on 2017-06-15 16:12 (UTC)

tried building it in chroot but it cant seem to find a package i have installed ==> Making package: chromium-dev 61.0.3128.3-1 (Thu Jun 15 12:06:41 EDT 2017) ==> Checking runtime dependencies... ==> Installing missing dependencies... error: target not found: openh264

walkindude commented on 2017-06-13 22:43 (UTC)

never mind, built in chroot, seems there are some incompatibilities with mesa-git

sl1pkn07 commented on 2017-06-12 21:28 (UTC)

build fine here

walkindude commented on 2017-06-12 21:12 (UTC)

https://pastebin.com/DrM0SpJT

commented on 2017-06-08 04:21 (UTC)

still throwing that error even after removing ffmpeg2.8 https://pastebin.com/PxaT7ReS

sl1pkn07 commented on 2017-06-08 01:30 (UTC)

sounds like chromium trying build against ffmpeg2.8 instead of ffmpeg in you setup try remove temporarily ffmpeg 2.8 and try again

commented on 2017-06-08 00:46 (UTC)

ffmpeg 1:3.3.1-9 ffmpeg2.8 2.8.11-9

commented on 2017-06-08 00:46 (UTC)

ffmpeg 1:3.3.1-9 ffmpeg2.8 2.8.11-9

sl1pkn07 commented on 2017-06-07 15:04 (UTC)

what version of ffmpeg you use?

commented on 2017-06-07 14:57 (UTC)

ninja: build stopped: subcommand failed. logs https://pastebin.com/TujES9aL

sl1pkn07 commented on 2017-06-04 14:37 (UTC)

use makepkg

ardadem commented on 2017-06-04 12:18 (UTC)

how can i fix this problem? : http://paste.ubuntu.com/24771505/

misc commented on 2017-06-02 11:47 (UTC) (edited on 2017-06-02 11:52 (UTC) by misc)

New versions support system freetype, but enabling it causes fonts to render in garbage quality for me. Also GCC 7.0.1 fails with (at least): https://bugs.chromium.org/p/chromium/issues/detail?id=691681

sl1pkn07 commented on 2017-06-02 06:19 (UTC)

More log please

commented on 2017-06-02 05:10 (UTC)

i keep on getting this error trying to build the new update. ninja: build stopped: subcommand failed.

sl1pkn07 commented on 2017-05-13 18:53 (UTC)

no. you need rebuild chromium-dev symlink missing libraries is mala praxis

commented on 2017-05-13 18:48 (UTC)

If you find an error saying /usr/lib/chromium-dev/chromium-dev: error while loading shared libraries: libwebp.so.6: cannot open shared object file: No such file or directory you can fix it like this sudo ln -s /usr/lib64/libwebp.so.7 /usr/lib64/libwebp.so.6

commented on 2017-05-12 03:59 (UTC)

Nice job on the very quick update :)

axfelix commented on 2017-05-08 17:37 (UTC)

Does this build not contain the new support for native notifications on Linux? Is that somehow only in Chrome rather than Chromium?

sl1pkn07 commented on 2017-05-03 13:57 (UTC)

@M4P9be3w0u done

M4P9be3w0u commented on 2017-05-02 22:52 (UTC)

I had to modify to the file "vaapi_patch_r0.patch" to compile version 60.0.3080.5-1 with vaapi enabled. Find the patch file that worked for me here: https://pastebin.com/qMH50eKS . (Disclaimer: not sure if this is the right place to poste this. May not work for eveyone.)

LordAro commented on 2017-04-18 09:56 (UTC)

oh indeed, that's how i built the package in the end. the question is whether the PKGBUILD *should* be able to handle being called in separate steps. Should probably take this to a mailing list...

sl1pkn07 commented on 2017-04-17 16:53 (UTC) (edited on 2017-04-17 16:57 (UTC) by sl1pkn07)

try with plain makepkg ;) (without options)

LordAro commented on 2017-04-17 16:43 (UTC)

I'm not convinced it is pacaur's bug though - it's this PKGBUILD that failed to build with makepkg -o && makepkg -es . I'll be quite happy to raise a bug report with pacaur if you can show that your PKGBUILD is doing it "correctly" (with documentation or whatever)

sl1pkn07 commented on 2017-04-16 12:33 (UTC)

open bug in pacaur. nothing to do in the pkgbuild

LordAro commented on 2017-04-16 10:28 (UTC)

@sl1pkn07 ah, found the issue. pacaur calls makepkg -o followed by makepkg -es Since your PKGBUILD only sets the $PATH in the prepare() function, the changes don't get added to the build() and other functions. Seems like the PATH change should be added either more globally (can you do that?) or to the other functions as needed

sl1pkn07 commented on 2017-04-11 13:57 (UTC)

done

deathangel908 commented on 2017-04-11 13:54 (UTC) (edited on 2017-04-11 13:56 (UTC) by deathangel908)

Just in case someone faced with the issue below - reinstall 'python2-six': ----------- FAILED: gen/content/browser/tracing/about_tracing.js gen/content/browser/tracing/about_tracing.html /usr/bin/python2 ../../third_party/catapult/tracing/bin/generate_about_tracing_contents --outdir gen/content/browser/tracing Traceback (most recent call last): File "../../third_party/catapult/tracing/bin/generate_about_tracing_contents", line 14, in <module> sys.exit(generate_about_tracing_contents.Main(sys.argv[1:])) File "/run/media/andrew/ext4/chromium-dev/src/chromium-59.0.3063.4/third_party/catapult/tracing/tracing_build/generate_about_tracing_contents.py", line 32, in Main load_sequence = vulcanizer.CalcLoadSequenceForModuleNames(names) File "/run/media/andrew/ext4/chromium-dev/src/chromium-59.0.3063.4/third_party/catapult/third_party/py_vulcanize/py_vulcanize/project.py", line 138, in CalcLoadSequenceForModuleNames name in module_names] File "/run/media/andrew/ext4/chromium-dev/src/chromium-59.0.3063.4/third_party/catapult/third_party/py_vulcanize/py_vulcanize/resource_loader.py", line 140, in LoadModule m.Parse() File "/run/media/andrew/ext4/chromium-dev/src/chromium-59.0.3063.4/third_party/catapult/third_party/py_vulcanize/py_vulcanize/html_module.py", line 29, in Parse raise Exception('While parsing %s: %s' % (self.name, str(ex))) Exception: While parsing tracing.ui.extras.about_tracing.about_tracing: Couldn't find a tree builder with the features you requested: html5lib. Do you need to install a parser library ---------- >>> import bs4 >>> bs4.BeautifulSoup(html, 'html5lib') ---------- >>> import html5lib Traceback (most recent call last): File "<stdin>", line 1, in <module> File "/usr/lib/python2.7/site-packages/html5lib/__init__.py", line 16, in <module> from .html5parser import HTMLParser, parse, parseFragment File "/usr/lib/python2.7/site-packages/html5lib/html5parser.py", line 2, in <module> from six import with_metaclass, viewkeys, PY3 ImportError: cannot import name viewkeys -------------------

digitalone commented on 2017-04-03 09:52 (UTC)

vaapi patch still doesn't work for AMD/Radeon cards?

sl1pkn07 commented on 2017-03-25 17:08 (UTC)

@kasei Done @lordaro seems is problem of you helper

kasei commented on 2017-03-25 16:46 (UTC)

missing depend harfbuzz-icu?

LordAro commented on 2017-03-25 13:35 (UTC)

Found my issue - # Set Python2 path. mkdir -p python-path ln -sf /usr/bin/python2 "${srcdir}/python-path/python" export PATH="${srcdir}/python-path:$PATH" conflicts with msg2 "Make sure use Python2" find . -name '*.py' -exec sed -r 's|/usr/bin/python$|&2|g' -i {} + find . -name '*.py' -exec sed -r 's|/usr/bin/env python$|&2|g' -i {} + changing the symlink to python2 sorts the issue

prazola commented on 2017-03-19 12:05 (UTC)

Compiled OK with clang and vaapi patch, but only software backend for me. r9 390x radeonsi/amdgpu. Here the starting log: [CODE] chromium-dev [17819:17845:0319/130308.484427:ERROR:connection_factory_impl.cc(386)] Failed to connect to MCS endpoint with error -111 [17905:17905:0319/130308.557405:ERROR:sandbox_linux.cc(343)] InitializeSandbox() called with multiple threads in process gpu-process. Received signal 11 SEGV_MAPERR 000000000000 #0 0x55fc13c399b7 <unknown> #1 0x55fc13c3952f <unknown> #2 0x7f800af00fe0 <unknown> r8: 000055fc1b1b4c68 r9: 0000000000000000 r10: 0000000000000036 r11: 00007f7ff84252cc r12: 00007ffcf0a59470 r13: 000055fc1b1e25f0 r14: 000055fc1b1b4c68 r15: 000055fc1b1b4ba0 di: 000055fc1b037c50 si: 0000000000000000 bp: 00007ffcf0a59828 bx: 000055fc1b1b4c60 dx: 0000000000000000 ax: 000055fc1b037c50 cx: 0000000000000001 sp: 00007ffcf0a590e8 ip: 0000000000000000 efl: 0000000000010202 cgf: 002b000000000033 erf: 0000000000000014 trp: 000000000000000e msk: 0000000000000000 cr2: 0000000000000000 [end of stack trace] Calling _exit(1). Core file will not be generated. [17819:17845:0319/130308.560396:ERROR:browser_gpu_channel_host_factory.cc(113)] Failed to launch GPU process. [17819:17845:0319/130308.560424:ERROR:browser_gpu_channel_host_factory.cc(113)] Failed to launch GPU process. [17819:17845:0319/130308.560453:ERROR:browser_gpu_channel_host_factory.cc(113)] Failed to launch GPU process. [17819:17845:0319/130308.561244:ERROR:browser_gpu_channel_host_factory.cc(113)] Failed to launch GPU process. [17819:17845:0319/130308.561370:ERROR:browser_gpu_channel_host_factory.cc(113)] Failed to launch GPU process. [/CODE]

LordAro commented on 2017-03-12 12:07 (UTC)

Getting the following build failure (from a fresh git checkout) http://pastebin.com/riBUdhvC

sl1pkn07 commented on 2017-03-10 00:09 (UTC) (edited on 2017-03-25 17:09 (UTC) by sl1pkn07)

And others,like breeze-icons by KDE/plasma, neither of the two if rename the files/entry to chromium, this make conflicts with chromium from [extra] (if is already installed) and show nothing in plasma with breeze-icons set you can make a workground creating a symlink in your favorite icon set from chromium to chromium-dev greetings

axfelix commented on 2017-03-09 22:46 (UTC)

Thanks for your hard work maintaining this package, it doesn't seem fun :) Thoughts on changing Icon= in the desktop file to just chromium, rather than chromium-dev? Most icon sets don't have a separate icon for chromium-dev.

cmsigler commented on 2017-03-09 14:13 (UTC) (edited on 2017-03-09 14:13 (UTC) by cmsigler)

chromium-dev 58.0.3029.6-1 builds for me on x86_64 with _use_clang=1 and _use_ccache=1 :) Happy-happy joy-joy @sl1pkn07: How in the world do you keep this AUR PKGBUILD working? How do you do constant maintenance on that beast?! Thank you for your hard work. Clemmitt

misc commented on 2017-03-02 23:08 (UTC) (edited on 2017-03-13 09:07 (UTC) by misc)

Welp, no idea why the previous hashes are garbage. Proper ones: f3aa7226 color: Update ICC histograms based on results 41d5c604 Add UMAs for numerical approximation of table-based transfer functions. 8958419e color: Add analytic transfer functions in shaders 1d14a779 color: Fix accidental partial color conversion ecdfdf5b color: Add support for shader generation As for enable_vaapi_on_linux.diff: http://pastebin.com/raw/je2ZQb4h There's sweet FA reason to append _${pkgver} when all you're doing is downloading an unchanging file, btw. *edit:* Video is fixed in the next version

cmsigler commented on 2017-03-02 17:09 (UTC) (edited on 2017-03-03 03:51 (UTC) by cmsigler)

UPDATE: For 58.0.3026.3, header file third_party/libvpx/source/libvpx/vpx/vp8dx.h is needed to declare VPXD_GET_LAST_QUANTIZER. Without this declaration third_party/webrtc/modules/video_coding/codecs/vp8/vp8_impl.cc (and vp9_impl.cc) can't compile. Also, when building with g++ there's a problem with function vpx_codec_control not being defined. During the build process the source tree's vp8dx.h file is removed and a stub header out/Release/gen/shim_headers/libvpx_shim/third_party/libvpx/source/libvpx/vpx/vp8dx.h is constructed. I'm not very good at understanding the mechanics of this sort of complex build process so I'm not sure what needs to be done to patch it. NOTE: chromium-gn-bootstrap-r1.patch and fix_668446_r1.diff are no longer needed. They've been upstreamed. HTH. Clemmitt

sl1pkn07 commented on 2017-02-28 22:42 (UTC)

@misc. can you post the link of these commits? i can't found it in the chromium git page

sl1pkn07 commented on 2017-02-28 13:04 (UTC)

yep. the vaapi part is totally broken you can talk with upstream https://github.com/saiarcot895/chromium-ubuntu-build greetings

digitalone commented on 2017-02-28 11:54 (UTC) (edited on 2017-02-28 12:17 (UTC) by digitalone)

It does not work with vaapi patch: patching file media/ffmpeg/ffmpeg_common.h Hunk #2 succeeded at 35 (offset 4 lines). patching file media/filters/ffmpeg_demuxer.cc Hunk #1 succeeded at 1195 (offset 10 lines). patching file tools/gn/bootstrap/bootstrap.py patching file chrome/browser/about_flags.cc Hunk #1 FAILED at 976. 1 out of 1 hunk FAILED -- saving rejects to file chrome/browser/about_flags.cc.rej patching file content/common/BUILD.gn Hunk #2 succeeded at 558 (offset -12 lines). patching file content/common/sandbox_linux/bpf_gpu_policy_linux.cc patching file content/common/va_wayland.sigs patching file content/gpu/BUILD.gn Hunk #1 succeeded at 92 (offset 2 lines). patching file content/gpu/gpu_main.cc Hunk #1 succeeded at 85 with fuzz 1 (offset 6 lines). Hunk #2 succeeded at 141 (offset 7 lines). patching file content/public/common/content_switches.cc Hunk #1 succeeded at 1003 (offset 5 lines). patching file content/public/common/content_switches.h patching file gpu/config/software_rendering_list_json.cc patching file media/BUILD.gn Hunk #1 succeeded at 389 (offset 13 lines). Hunk #2 succeeded at 660 (offset -12 lines). patching file media/gpu/BUILD.gn patching file media/gpu/gpu_video_decode_accelerator_factory.cc patching file media/gpu/gpu_video_decode_accelerator_factory.h patching file media/gpu/ipc/service/BUILD.gn patching file media/gpu/ipc/service/gpu_video_decode_accelerator.cc patching file media/gpu/va_stub_header.fragment patching file media/gpu/vaapi_video_decode_accelerator.cc Hunk #1 FAILED at 359. Hunk #2 FAILED at 379. Hunk #3 FAILED at 394. 3 out of 3 hunks FAILED -- saving rejects to file media/gpu/vaapi_video_decode_accelerator.cc.rej patching file media/gpu/vaapi_wrapper.cc Hunk #1 FAILED at 194. Hunk #2 succeeded at 340 (offset -7 lines). Hunk #3 succeeded at 409 (offset -14 lines). Hunk #4 succeeded at 433 (offset -14 lines). 1 out of 4 hunks FAILED -- saving rejects to file media/gpu/vaapi_wrapper.cc.rej

LordAro commented on 2017-02-15 00:04 (UTC)

Looks great, ta (I think I only had trouble with python & node, but all 3 works as well)

sl1pkn07 commented on 2017-02-14 23:36 (UTC) (edited on 2017-02-14 23:37 (UTC) by sl1pkn07)

oh. i see done(?)

LordAro commented on 2017-02-14 23:12 (UTC)

Can you add a couple of -f to the ln commands? Getting "failed to create symbolic link" every time I try to rebuild this is a bit boring...

digitalone commented on 2017-02-13 21:53 (UTC)

Can someone upload a binary package with vaapi patch? Thanks.

cmsigler commented on 2017-02-11 13:56 (UTC) (edited on 2017-02-25 16:29 (UTC) by cmsigler)

One final update on building with (chromium's version of) clang -- See chromium Issue 685244, https://bugs.chromium.org/p/chromium/issues/detail?id=685244 Next UPDATE: The in-tree version of clang has been updated to llvm/clang r295762. As the clang crash was fixed in r291064 it's expected clang can be used to build the tree under Linux again. This assumes this in-tree update is not reverted for some reason. So far, so good. This clang update was committed to the tree 24 Feb 2017. AFAICT, a dev channel update to chromium 58.0.3021.3 or later should contain this update. chromium-58.0.3018.3.tar.xz would not. Clemmitt

sl1pkn07 commented on 2017-02-09 19:47 (UTC)

for tht i back to gcc (see the commit log)

cmsigler commented on 2017-02-09 17:46 (UTC) (edited on 2017-02-10 17:29 (UTC) by cmsigler)

@sl1pkn07 UPDATE 2: Having researched, this clang build bug is perplexing to me. The clang version d'l'ded for the build is somehow tied into the chromium tree. See src/chromium-58.0.3004.3/tools/clang/scripts/update.py where it says "Do NOT CHANGE this if you don't know what you're doing" and the URL given. But... it's impossible to build chromium with this buggy revision of clang, right? Shouldn't they update clang throughout their own tree? Thank you for answering my comment :) P.S.: I finally spent the time required to verify this hadn't been reported as a chromium bug, then filed a bug report there myself: chromium Issue 690972 Original comment: Hi, For chromium vers. 58.0.3000.4 and 58.0.3004.3 I tried to build with _use_ccache=1 (may not help speed things up very much?) and _use_clang=1. My builds crashed (OneObsession also saw clang failure), in both cases on this file: src/chromium-58.0.3004.3/chrome/common/chrome_content_client.cc. I looked back and I used clang to build ver. 56.0.2897.0 successfully. HTH. UPDATE: This is llvm.org/bugs/ Bug 31043 which is fixed in clang r291064. The clang binary automatically downloaded from Google is clang-289944-2.tgz which hasn't had this bug fixed. Clemmitt

sl1pkn07 commented on 2017-02-08 21:52 (UTC)

Poll: Use GTK3 by default? Remove the PAX related in the PKGBUILD? #discuss

BrettBlackham commented on 2017-02-08 18:14 (UTC)

You should add wget to the list of required packages. Cheers. -> Update libaddressinput strings. -> Update nodejs deps ./third_party/node/update_node_binaries: line 86: wget: command not found ==> ERROR: A failure occurred in prepare(). Aborting...

OneObsession commented on 2017-02-02 19:12 (UTC)

Hi, i was not able to compile chromium with clang. Here is the output: Command '['/tmp/tmpFGCpwT/gn', 'gen', '/home/user/Downloads/chromium-dev/src/chromium-56.0.2924.87/out/Release', '--args=is_debug=false enable_widevine=true enable_hangout_services_extension=false ffmpeg_branding="ChromeOS" proprietary_codecs=true google_api_key="AIzaSyDwr302FpOSkGRpLlUpPThNTDPbXcIn_FM" google_default_client_id="413772536636.apps.googleusercontent.com" google_default_client_secret="0ZChLK6AxeA3Isu96MkwqDR4" fieldtrial_testing_like_official_build=false remove_webcore_debug_symbols=true use_gconf=false use_gio=false use_gnome_keyring=true use_gtk3=false use_pulseaudio=true link_pulseaudio=true use_kerberos=true use_cups=true use_sysroot=false use_gold=false use_allocator="none" linux_use_bundled_binutils=false fatal_linker_warnings=false treat_warnings_as_errors=false enable_nacl=true enable_nacl_nonsfi=true custom_toolchain="/home/user/Downloads/chromium-dev/src:default" is_clang=true clang_use_chrome_plugins=true is_debug=false']' returned non-zero exit status -4

yacuken commented on 2017-01-20 07:52 (UTC)

full build log on pastebin http://pastebin.com/ZrXDhf2X

sl1pkn07 commented on 2017-01-20 06:57 (UTC)

please paste more log, seems is incomplete greetings

yacuken commented on 2017-01-20 06:47 (UTC)

ERROR at //build/config/linux/pkg_config.gni:85:17: Script returned non-zero exit code. pkgresult = exec_script(pkg_config_script, args, "value") ^---------- Current dir: /home/yacuken/chromium-dev/src/chromium-57.0.2984.0/out/Release/ Command: python -- /home/yacuken/chromium-dev/src/chromium-57.0.2984.0/build/config/linux/pkg-config.py gmodule-2.0 gtk+-2.0 gthread-2.0 Returned 1. stderr: Package gtk+-2.0 was not found in the pkg-config search path. Perhaps you should add the directory containing `gtk+-2.0.pc' to the PKG_CONFIG_PATH environment variable No package 'gtk+-2.0' found Could not run pkg-config. See //build/config/linux/gtk2/BUILD.gn:15:1: whence it was called. pkg_config("gtk2_internal_config") { ^----------------------------------- See //chrome/browser/ui/libgtkui/BUILD.gn:147:12: which caused the file to be included. "//build/config/linux/gtk2", ^--------------------------

sl1pkn07 commented on 2017-01-20 06:46 (UTC)

Please paste the log (for add to the upstream bugreport) Thanks

yacuken commented on 2017-01-20 06:42 (UTC)

gtk2 still needs as makedepends even if use gtk3.

sl1pkn07 commented on 2017-01-16 13:55 (UTC) (edited on 2017-01-16 13:58 (UTC) by sl1pkn07)

try to change the line #168 from gnome-keyring to libgnome-keyring edit: pushed

yacuken commented on 2017-01-16 13:37 (UTC) (edited on 2017-01-16 13:39 (UTC) by yacuken)

ERROR at //build/config/linux/pkg_config.gni:85:17: Script returned non-zero exit code. pkgresult = exec_script(pkg_config_script, args, "value") ^---------- Current dir: /tmp/yaourt-tmp-yacuken/aur-chromium-dev/src/chromium-57.0.2979.0/out/Release/ Command: python -- /tmp/yaourt-tmp-yacuken/aur-chromium-dev/src/chromium-57.0.2979.0/build/config/linux/pkg-config.py gnome-keyring-1 Returned 1. stderr: Package gnome-keyring-1 was not found in the pkg-config search path. Perhaps you should add the directory containing `gnome-keyring-1.pc' to the PKG_CONFIG_PATH environment variable No package 'gnome-keyring-1' found Could not run pkg-config. Add libgnome-keyring to makedepends (if use gnome-keyring). Thanks

sl1pkn07 commented on 2017-01-13 23:29 (UTC)

try build with internal jinja2,markupsafe and ply in the next release i will use instead of system ones, because is only for make, not for run

LordAro commented on 2017-01-13 22:52 (UTC)

I'm getting a (several) compile failure around 10000 files in: gen/v8/src/inspector/protocol/HeapProfiler.h:294:13: error: static assertion failed: property total should not be set yet static_assert(!(STATE & TotalSet), "property total should not be set yet"); any ideas?

neeks commented on 2017-01-12 16:12 (UTC) (edited on 2017-01-12 16:34 (UTC) by neeks)

The build tag 57.0.2970.0 happened right at the beginning of real GTK3 integration support. If you use the default adwaita theme - specifically the light version - none of the GTK3 menu text gets styled and the text will appear white on a white background. The next tagged version should have much better support, but until then you can update `~/.config/gtk-3.0/gtk.css` with the following to make the menu item's visible. This solution is just meant as temporary as the styling won't match up to the theme completely because this version's gtk3 css selector targeting isn't working great with 3.20+: ``` /* Seperator */ .menu .menuitem { color: rgba(0,0,0.15); } /* Menu Item Text Color */ .menu .menuitem * { color: rgba(0,0,0,.75); } /* Menu Item on hover, can't get the text to pick up a different color */ .menu .menuitem:hover { background-color: #4a90d9; } /* Make the menu border more subtle */ .menu { border: 1px solid rgba(0,0,0,.05); } ``` You can follow Chromium's gtk3 progress here: https://bugs.chromium.org/p/chromium/issues/detail?id=132847

joekiller commented on 2017-01-10 20:56 (UTC)

I got the error chromuim /usr/lib/libharfbuzz.so.0: undefined reference to `FT_Get_Var_Blend_Coordinates' and had to rebuild libharfbuzz per instructions I found at https://github.com/jaagr/polybar/issues/310#issuecomment-271353372 which states to rebuild: $ abs extra/harfbuzz $ cp -r /var/abs/extra/harfbuzz /tmp $ # increase patch version in PKGBUILD $ cd /tmp/harfbuzz $ makepkg -si

lwin commented on 2017-01-07 14:17 (UTC)

Can confirm vaapi patch is broken for the current 57.0.2970.0-1 release: patching file media/gpu/gpu_video_decode_accelerator_factory.cc patching file media/gpu/gpu_video_decode_accelerator_factory.h patching file media/gpu/ipc/service/BUILD.gn patching file media/gpu/ipc/service/gpu_video_decode_accelerator.cc patching file media/gpu/va_stub_header.fragment patching file media/gpu/vaapi_video_decode_accelerator.cc patching file media/gpu/vaapi_wrapper.cc ==> ERROR: A failure occurred in prepare(). Aborting...

sl1pkn07 commented on 2017-01-07 10:56 (UTC)

sounds vaapi patch is broken again ¬¬

OneObsession commented on 2017-01-07 10:54 (UTC)

I can not compile chromium because the build fails in prepare(). Patching the files seem to cause errors. Here is the output: patching file chrome/browser/about_flags.cc Hunk #1 FAILED at 976. 1 out of 1 hunk FAILED -- saving rejects to file chrome/browser/about_flags.cc.rej patching file content/common/BUILD.gn Hunk #2 succeeded at 568 (offset -2 lines). patching file content/common/sandbox_linux/bpf_gpu_policy_linux.cc patching file content/common/va_wayland.sigs patching file content/gpu/BUILD.gn patching file content/gpu/gpu_main.cc Hunk #1 succeeded at 80 (offset 1 line). Hunk #2 succeeded at 135 (offset 1 line). patching file content/public/common/content_switches.cc Hunk #1 succeeded at 1003 (offset 5 lines). patching file content/public/common/content_switches.h patching file gpu/config/software_rendering_list_json.cc patching file media/BUILD.gn Hunk #1 succeeded at 377 (offset 1 line). Hunk #2 succeeded at 673 (offset 1 line). patching file media/gpu/BUILD.gn patching file media/gpu/gpu_video_decode_accelerator_factory.cc patching file media/gpu/gpu_video_decode_accelerator_factory.h patching file media/gpu/ipc/service/BUILD.gn patching file media/gpu/ipc/service/gpu_video_decode_accelerator.cc patching file media/gpu/va_stub_header.fragment patching file media/gpu/vaapi_video_decode_accelerator.cc patching file media/gpu/vaapi_wrapper.cc ==> FEHLER: Ein Fehler geschah in prepare(). Breche ab...

usuariopolivalen commented on 2016-12-28 22:29 (UTC)

with what graphics is useful the vaapi option? the other chromium with vaapi(samcv repo) gives me problems with amd.. it is only for intel graphics(i got this functioning in intel pc's only)? the only thing where I saw differences between the two graphics (apart of the performance of amd over intel) is the fail of sandbox on amd with gallium(= no hardware accel).. can it be the problem of the accel with that cards?

marcelhuber commented on 2016-12-13 13:43 (UTC)

harfbuzz-icu is missing in dependencies; build fails on minimal archlinux installation

sl1pkn07 commented on 2016-11-24 18:06 (UTC)

need wait for upstream vaapi patch update

prazola commented on 2016-11-24 17:26 (UTC)

patching file media/gpu/ipc/service/gpu_video_decode_accelerator.cc Hunk #1 FAILED at 58. Hunk #2 FAILED at 171. 2 out of 2 hunks FAILED -- saving rejects to file media/gpu/ipc/service/gpu_video_decode_accelerator.cc.rej

sl1pkn07 commented on 2016-11-23 17:03 (UTC)

Unfortunately, GTK2 is still required when building

normanu commented on 2016-11-23 16:25 (UTC)

Mm ok fixed ncurses5-compat-libs by importing the PGP key. In the PKGBUILD is have the variable _use_gtk3=1 Is it normal that I still need GTK2 installed? Package gtk+-2.0 was not found in the pkg-config search path. Perhaps you should add the directory containing `gtk+-2.0.pc' to the PKG_CONFIG_PATH environment variable No package 'gtk+-2.0' found Could not run pkg-config.

sl1pkn07 commented on 2016-11-23 11:30 (UTC)

https://aur.archlinux.org/packages/ncurses5-compat-libs/

normanu commented on 2016-11-23 09:36 (UTC)

Chromium-dev doesn't seem to want to build for me because of a dependency for ncurses5-compat-libs With yaourt it will try to install it, but finds ncurses6-compat-libs and can't match the pgp code. How do install this? Need the va-api enabled for Chrome, to use Moonlight Gamestream :) (strange if it is default in ChromeOS that they don't enable it for other distro's)

commented on 2016-11-16 16:28 (UTC)

Thanks!

sl1pkn07 commented on 2016-11-16 15:48 (UTC)

try to set _debug_mode= to 1 and leave untouch the options=()

commented on 2016-11-16 15:46 (UTC)

Hi, would it be possible to use GCC instead of Clang as the compiler? When trying to set options=('debug' '!strip') the compilation fails because Clang doesn't support -fvar-tracking-assignments.

sl1pkn07 commented on 2016-11-08 06:27 (UTC) (edited on 2016-11-08 06:29 (UTC) by sl1pkn07)

Try remove '-v' in the line 490 and 500

44trent3 commented on 2016-11-08 03:19 (UTC)

Why is everything so...verbose in my terminal when I install this? I sort of want to be able to see how far it is, but I can't because my terminal window is spammed with the full compile commands of every source file

sl1pkn07 commented on 2016-10-28 12:58 (UTC)

because don't work in all systems

BigEars431 commented on 2016-10-28 06:05 (UTC)

why not just have vaapi enabled by default?

BigEars431 commented on 2016-10-28 06:05 (UTC)

why not just have vaapi enabled by default?

deemde commented on 2016-10-27 18:21 (UTC)

I am getting a weird error with 55 and 56. As soon as I press any key on my keyboard, chromium immediately shuts down / crashes. Has anyone experienced similar problems?

BigEars431 commented on 2016-10-26 10:39 (UTC)

how's the 56.0.2897.0-2 PKGBUILD update going? how can I help? in my own little way =)

sl1pkn07 commented on 2016-10-16 13:02 (UTC) (edited on 2016-10-16 13:03 (UTC) by sl1pkn07)

depend of you aur helper anyway, ncurses5-compat-libs is in makedepends because is only used by bundled clang (compiler)

archerie commented on 2016-10-16 11:57 (UTC)

@sl1pkn07 Installing that removes the error. Shouldn't it be specified as part of this package's dependencies so it doesn't have to be installed separately?

sl1pkn07 commented on 2016-10-16 11:05 (UTC)

@BigEars431 https://github.com/saiarcot895/chromium-ubuntu-build/issues/2 paste here you problem

sl1pkn07 commented on 2016-10-16 11:03 (UTC)

ncurses5-compat-libs is in [AUR]

archerie commented on 2016-10-16 10:35 (UTC) (edited on 2016-10-16 10:35 (UTC) by archerie)

Installation fails for me. How to fix it? ==> Making package: chromium-dev 55.0.2883.11-1 (Sun Oct 16 12:33:00 CEST 2016) ==> Checking runtime dependencies... ==> Checking buildtime dependencies... ==> Installing missing dependencies... error: target not found: ncurses5-compat-libs ==> ERROR: 'pacman' failed to install missing dependencies. :: failed to build chromium-dev package(s)

BigEars431 commented on 2016-10-16 07:24 (UTC)

I still don't get the vaapi integration... did I do anything wrong? I'm on an Intel-Only Lenovo Thinkpad X1 Carbon...Compiled successfully with the default PKGBUILD

prazola commented on 2016-10-09 11:45 (UTC)

Thi is the starting log: [12243:12243:1009/134131:ERROR:sandbox_linux.cc(343)] InitializeSandbox() called with multiple threads in process gpu-process. Received signal 11 SEGV_MAPERR 000000000000 #0 0x5630fbfe1d47 <unknown> #1 0x7f51228a2080 <unknown> r8: 0000563101c26a00 r9: 0000000000000000 r10: 0000000000000036 r11: 00007f51102102cc r12: 00007ffddb3d7ff0 r13: 00005631036ae7e0 r14: 0000563101c26a00 r15: 00005631036a5010 di: 0000563103527600 si: 0000000000000000 bp: 00007ffddb3d8300 bx: 0000563101c269f8 dx: 0000000000000000 ax: 0000563103527600 cx: 0000000000000001 sp: 00007ffddb3d7ac8 ip: 0000000000000000 efl: 0000000000010206 cgf: 002b000000000033 erf: 0000000000000014 trp: 000000000000000e msk: 0000000000000000 cr2: 0000000000000000 [end of stack trace] [12096:12163:1009/134132:ERROR:connection_factory_impl.cc(367)] Failed to connect to MCS endpoint with error -111

sl1pkn07 commented on 2016-10-08 20:51 (UTC) (edited on 2016-10-08 20:51 (UTC) by sl1pkn07)

please paste the console log (the part related of libva) into https://github.com/saiarcot895/chromium-ubuntu-build/issues/2 greetings

prazola commented on 2016-10-08 16:26 (UTC) (edited on 2016-10-08 16:39 (UTC) by prazola)

Nope, r9 390x (radeonsi). I'll reinstall everything va related. EDIT: [URL]http://www.imagebam.com/image/d0ee1f508528021[/URL]

sl1pkn07 commented on 2016-10-08 16:01 (UTC) (edited on 2016-10-08 16:02 (UTC) by sl1pkn07)

nvidia user? install 'libva-vdpau-driver-chromium'

prazola commented on 2016-10-08 15:52 (UTC)

Still software only in chrome:gpu for me. :(

sl1pkn07 commented on 2016-10-03 17:58 (UTC)

the new push only sync with the new version, for me no need apply the patch (works OOB), you can try if set 1 '_patch_vaapi'

prazola commented on 2016-10-03 17:45 (UTC)

vaapi and hw acceleration working now?

BigEars431 commented on 2016-10-01 16:33 (UTC) (edited on 2016-10-01 16:33 (UTC) by BigEars431)

YAY! Thank you so much for maintaining this package, wishing the vaapi patch is enabled by default though... so the people maintaining the archlinuxcn repo doesn't have to change the PKGBUILD everytime a new release is out. That's where I get my chromium-dev binary to make it a faster process for me as a web developer (no need to compile hehe), but overall Thanks @sl1pkn07 :)

felixsanz commented on 2016-10-01 14:11 (UTC)

Now it compiles, thanks

sl1pkn07 commented on 2016-10-01 11:05 (UTC)

i found a typo, try now

felixsanz commented on 2016-10-01 10:04 (UTC)

Well my installation is not broken as there is no installation at all. I can remove the entire chromium-dev folder, repeat, and same thing happens, so it's clearly an issue with the AUR package

NSDragon commented on 2016-09-30 22:07 (UTC)

For some reason makepkg isn't installing the makedepends for me at all (I have no idea why). After manually installing them all, chromium compiles just fine.

sl1pkn07 commented on 2016-09-30 19:24 (UTC)

Seems your instalación is Broken,because ninja is mandatory and is set in makedepends

felixsanz commented on 2016-09-30 19:02 (UTC)

$ ninja fish: Unknown command 'ninja'

sl1pkn07 commented on 2016-09-30 17:50 (UTC) (edited on 2016-09-30 17:51 (UTC) by sl1pkn07)

is strange, because this File "tools/gn/bootstrap/bootstrap.py", line 44, in check_call subprocess.check_call(cmd, cwd=GN_ROOT, **kwargs) launch this: ['ninja', '-C', '/tmp/tmpKAfkXT', '-v', 'gn'] are you sure have right installed ninja? is set in makedepends

felixsanz commented on 2016-09-30 16:02 (UTC)

yes, my disk have space and user have permissions to /tmp

NSDragon commented on 2016-09-30 15:17 (UTC)

I'm also unable to compile. /tmp has 7.9 GiB free, and I am able to write to it as a normal user. -> Starting building Chromium... Building gn manually in a temporary directory for bootstrapping... DEBUG:root:Running: /home/nsdragon/aur/chromium-dev/src/chromium-55.0.2873.4/build/write_buildflag_header.py --output base/allocator/features.h --gen-dir /tmp/tmpBGiRNA/gen --definitions /tmp/tmpBGiRNA/gen/base/allocator/features.h.tmp DEBUG:root:Running: /home/nsdragon/aur/chromium-dev/src/chromium-55.0.2873.4/build/write_buildflag_header.py --output base/debug/debugging_flags.h --gen-dir /tmp/tmpBGiRNA/gen --definitions /tmp/tmpBGiRNA/gen/base/debug/debugging_flags.h.tmp DEBUG:root:Running: ninja -C /tmp/tmpBGiRNA -v gn Traceback (most recent call last): File "tools/gn/bootstrap/bootstrap.py", line 750, in <module> sys.exit(main(sys.argv[1:])) File "tools/gn/bootstrap/bootstrap.py", line 120, in main return run_build(tempdir, options) File "tools/gn/bootstrap/bootstrap.py", line 71, in run_build build_gn_with_ninja_manually(tempdir, options) File "tools/gn/bootstrap/bootstrap.py", line 195, in build_gn_with_ninja_manually check_call(cmd) File "tools/gn/bootstrap/bootstrap.py", line 44, in check_call subprocess.check_call(cmd, cwd=GN_ROOT, **kwargs) File "/usr/lib/python2.7/subprocess.py", line 536, in check_call retcode = call(*popenargs, **kwargs) File "/usr/lib/python2.7/subprocess.py", line 523, in call return Popen(*popenargs, **kwargs).wait() File "/usr/lib/python2.7/subprocess.py", line 711, in __init__ errread, errwrite) File "/usr/lib/python2.7/subprocess.py", line 1343, in _execute_child raise child_exception OSError: [Errno 2] No such file or directory ==> ERROR: A failure occurred in build(). Aborting...

sl1pkn07 commented on 2016-09-30 11:22 (UTC)

@LordAro try now sorry and greetings

LordAro commented on 2016-09-30 09:37 (UTC)

I agree, but even so I think the SRCINFO file should be a minimum subset of the dependencies, i.e. the ones that are guaranteed to be needed. Currently `makepkg --printsrcinfo` will be parsing the PKGBUILD based on your current install, rather than mine or anyone else's I wonder if pulseaudio could just be made an optional dependency, with the caveat of it having to be installed at compile time (which covers most cases, I think)

sl1pkn07 commented on 2016-09-30 09:14 (UTC)

Then the problem is pacaur, because the pkgbuild iself detect if you have installed PulseAudio, then add it to depends automatically Try to build with makepkg

LordAro commented on 2016-09-30 09:07 (UTC)

pacaur is trying to get me to install pulseaudio as a dependency of chromium-dev, i believe because it's been added to .SRCINFO, can it (and other optional dependencies) be removed?

sl1pkn07 commented on 2016-09-30 07:49 (UTC) (edited on 2016-09-30 07:50 (UTC) by sl1pkn07)

have you space in /tmp? can you write anything in /tmp as user? because: -> Starting building Chromium... Building gn manually in a temporary directory for bootstrapping... DEBUG:root:Running: /tmp/makepkg/chromium-dev/src/chromium-55.0.2873.4/build/write_buildflag_header.py --output base/allocator/features.h --gen-dir /tmp/tmphiecKP/gen --definitions /tmp/tmphiecKP/gen/base/allocator/features.h.tmp DEBUG:root:Running: /tmp/makepkg/chromium-dev/src/chromium-55.0.2873.4/build/write_buildflag_header.py --output base/debug/debugging_flags.h --gen-dir /tmp/tmphiecKP/gen --definitions /tmp/tmphiecKP/gen/base/debug/debugging_flags.h.tmp DEBUG:root:Running: ninja -C /tmp/tmphiecKP -v gn ninja: Entering directory `/tmp/tmphiecKP' [1/330] /tmp/makepkg/chromium-dev/src/chromium-55.0.2873.4/third_party/llvm-build/Release+Asserts/bin/clang -Qunused-arguments -MMD -MF base/third_party/libevent/evbuffer.o.d -I/tmp/tmphiecKP/gen -I/tmp/makepkg/chromium-dev/src/chromium-55.0.2873.4 -I/tmp/makepkg/chromium-dev/src/chromium-55.0.2873.4/base/third_party/libevent/linux -march=native -O2 -pipe -fstack-protector-strong -DNO_TCMALLOC -O2 -g0 -D_FILE_OFFSET_BITS=64 -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -pthread -pipe -fno-exceptions -march=native -O2 -pipe -fstack-protector-strong -DNO_TCMALLOC -O2 -g0 -D_FILE_OFFSET_BITS=64 -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -pthread -pipe -fno-exceptions -DHAVE_CONFIG_H -c /tmp/makepkg/chromium-dev/src/chromium-55.0.2873.4/base/third_party/libevent/evbuffer.c -o base/third_party/libevent/evbuffer.o [2/330] /tmp/makepkg/chromium-dev/src/chromium-55.0.2873.4/third_party/llvm-build/Release+Asserts/bin/clang -Qunused-arguments -MMD -MF base/third_party/libevent/evutil.o.d -I/tmp/tmphiecKP/gen -I/tmp/makepkg/chromium-dev/src/chromium-55.0.2873.4 -I/tmp/makepkg/chromium-dev/src/chromium-55.0.2873.4/base/third_party/libevent/linux -march=native -O2 -pipe -fstack-protector-strong -DNO_TCMALLOC -O2 -g0 -D_FILE_OFFSET_BITS=64 -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -pthread -pipe -fno-exceptions -march=native -O2 -pipe -fstack-protector-strong -DNO_TCMALLOC -O2 -g0 -D_FILE_OFFSET_BITS=64 -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -pthread -pipe -fno-exceptions -DHAVE_CONFIG_H -c /tmp/makepkg/chromium-dev/src/chromium-55.0.2873.4/base/third_party/libevent/evutil.c -o base/third_party/libevent/evutil.o

felixsanz commented on 2016-09-30 05:31 (UTC)

@sl1pkn07 i've seen that you uploaded another version, but it still doesn't compile. -> Starting building Chromium... Building gn manually in a temporary directory for bootstrapping... DEBUG:root:Running: /home/me/builds/chromium-dev/src/chromium-55.0.2873.4/build/write_buildflag_header.py --output base/allocator/features.h --gen-dir /tmp/tmpad8LG8/gen --definitions /tmp/tmpad8LG8/gen/base/allocator/features.h.tmp DEBUG:root:Running: /home/me/builds/chromium-dev/src/chromium-55.0.2873.4/build/write_buildflag_header.py --output base/debug/debugging_flags.h --gen-dir /tmp/tmpad8LG8/gen --definitions /tmp/tmpad8LG8/gen/base/debug/debugging_flags.h.tmp DEBUG:root:Running: ninja -C /tmp/tmpad8LG8 -v gn Traceback (most recent call last): File "tools/gn/bootstrap/bootstrap.py", line 750, in <module> sys.exit(main(sys.argv[1:])) File "tools/gn/bootstrap/bootstrap.py", line 120, in main return run_build(tempdir, options) File "tools/gn/bootstrap/bootstrap.py", line 71, in run_build build_gn_with_ninja_manually(tempdir, options) File "tools/gn/bootstrap/bootstrap.py", line 195, in build_gn_with_ninja_manually check_call(cmd) File "tools/gn/bootstrap/bootstrap.py", line 44, in check_call subprocess.check_call(cmd, cwd=GN_ROOT, **kwargs) File "/usr/lib/python2.7/subprocess.py", line 536, in check_call retcode = call(*popenargs, **kwargs) File "/usr/lib/python2.7/subprocess.py", line 523, in call return Popen(*popenargs, **kwargs).wait() File "/usr/lib/python2.7/subprocess.py", line 711, in __init__ errread, errwrite) File "/usr/lib/python2.7/subprocess.py", line 1343, in _execute_child raise child_exception OSError: [Errno 2] No such file or directory ==> ERROR: A failure occurred in build(). Aborting...

sl1pkn07 commented on 2016-09-28 18:35 (UTC) (edited on 2016-09-29 07:38 (UTC) by sl1pkn07)

nope. because ncurses-compat-libs is used only in exclusively by bundled clang, and clang is only for make, not for runtime and about icudat, you can wait a little time? the new version is out and now is in test/build greetings edited

felixsanz commented on 2016-09-28 18:22 (UTC)

Fixed that issue installing ncurses5-compat-libs, thanks! But... shouldn't ncurses5-compat-libs be a dep of chromium-dev? Anyway, it still doesn't compile, now this: ninja: Entering directory `out/Release' ninja: error: '../../third_party/icu/common/icudtl.dat', needed by 'icudtl.dat', missing and no known rule to make it ==> ERROR: A failure occurred in build(). Aborting...

sl1pkn07 commented on 2016-09-28 17:11 (UTC)

@felixsanz try again

joekiller commented on 2016-09-28 14:11 (UTC) (edited on 2016-09-28 14:21 (UTC) by joekiller)

@felixsanz libtinfo has moved on to .6. To get .5 you can install ncurses5-compat-libs. Note that to install that package you must import the author's key first. IE: gpg --keyserver pgp.mit.edu --recv-keys 702353E0F7E48EDB It should compile then. I had the same problem.

felixsanz commented on 2016-09-28 06:41 (UTC)

@sl1pkn07 same error :(

sl1pkn07 commented on 2016-09-28 04:14 (UTC)

@felixsanz try now

felixsanz commented on 2016-09-28 02:06 (UTC)

Any idea of why this doesn't compile? /home/me/builds/chromium-dev/src/chromium-55.0.2868.3/third_party/llvm-build/Release+Asserts/bin/clang: error while loading shared libraries: libtinfo.so.5: cannot open shared object file: No such file or directory ninja: build stopped: subcommand failed. Command '['ninja', '-C', '/tmp/tmpiPDkdY', '-v', 'gn']' returned non-zero exit status 1 ==> ERROR: A failure occurred in build(). Aborting...

sl1pkn07 commented on 2016-09-27 19:00 (UTC)

Now pushed the option for enable the vaapi patch (disabled by default, for enable need set 1 the option _patch_vaapi in the PKGBUILD) for me, if apply the patch i get segfault about the GPU process and turn off the HW acceleration. sorry the inconvenience greetings

sl1pkn07 commented on 2016-09-27 15:41 (UTC) (edited on 2016-09-27 16:49 (UTC) by sl1pkn07)

@axfelix @banzr try now @axfelix I've add the vaapi patch, but now from the PPA repository instead of github @banzr I've disabled the built with bundled clang if the system is 32bit. but should able build with the system clang @all, the pkgrel is not bumped because the package works as-is, if anyone want build for include the changes, only need reinstall the package greetings

sl1pkn07 commented on 2016-09-26 22:47 (UTC)

need rework by upstream (the patch) try to ask in here https://github.com/saiarcot895/chromium-ubuntu-build but for me, work out of the box (without patch) try also you can tweak the chrome://flags greetings

axfelix commented on 2016-09-26 22:15 (UTC) (edited on 2016-09-26 22:17 (UTC) by axfelix)

is the VAAPI patch missing now? I just installed and I'm only getting software video decoding, and it doesn't seem to be in the list of sources anymore... though congrats on getting it to build again :)

sl1pkn07 commented on 2016-09-26 14:15 (UTC) (edited on 2016-09-26 15:30 (UTC) by sl1pkn07)

mm sound is only compatible with 64bits, because https://commondatastorage.googleapis.com/chromium-browser-clang/tools/cmake343_Linux.tgz only ship the 64bits of cmake (needed for build the bundled clang) I,ve work to this, at the moment, the solution is edit the PKGBUILD and set to 0 _use_bundled_clang in the top of the script greetings EDITED

commented on 2016-09-26 07:14 (UTC)

I get the following error during build(). I'm running on an i686 platform. Could there be a 32 bit flag set to 64bit by accident that I need to manually set? ...llvm-build-tools/cmake343/bin/cmake: cannot execute binary file: Exec format error Failed.

sl1pkn07 commented on 2016-09-23 21:03 (UTC)

55.0.2868.3 in build process. if all ok, tomorrow i'll push the new version!

misc commented on 2016-09-23 20:40 (UTC)

Ah, you meant chrome:gpu Yeah now that I see, all green except "Native GpuMemoryBuffers". Yay!

sl1pkn07 commented on 2016-09-23 20:24 (UTC)

fixed my site

misc commented on 2016-09-23 19:58 (UTC) (edited on 2016-09-23 19:58 (UTC) by misc)

That compilation aborts (with the same pkgbuild as before) unless zlib is removed from the gn system array and its _keeplibs adjusted accordingly. Also your site 522's.

sl1pkn07 commented on 2016-09-23 19:23 (UTC) (edited on 2016-09-23 19:38 (UTC) by sl1pkn07)

builded with bunbled libevent and it works!!! include HW acceleration!! (?) https://sl1pkn07.wtf/paste/view/raw/96c84e62 habemus new working version!!! AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAa!!!!!!!!!!! what you mean with the zlib part? greetings

misc commented on 2016-09-23 19:12 (UTC)

My 55.0.2868.0 frankenbuild (bunch of git versions atop) seems to almost work again, getting reproducible crashes but note of the previous issues. However… 1. system zlib doesn't work as gn file is broken 2. disabled system libevent, according to a gentoo bug it breaks tcmalloc (which indeed works again) 3. vaapi patch still needs to be redone

sl1pkn07 commented on 2016-09-23 13:15 (UTC) (edited on 2016-09-23 19:13 (UTC) by sl1pkn07)

Is not dead, but is hard to run (without errors) again with the New upstream build system (GN)

BigEars431 commented on 2016-09-23 12:18 (UTC)

Is this dead? I sure hope not.. really enjoying the VAAPI patch =)

sl1pkn07 commented on 2016-09-14 21:26 (UTC)

https://sl1pkn07.wtf/paste/view/raw/7ad7f6c6

LordAro commented on 2016-09-14 21:02 (UTC) (edited on 2016-09-14 21:03 (UTC) by LordAro)

"==> ERROR: unset-madv_free.patch was not found in the build directory and is not a URL." I did have a "MADV_FREE.patch" in my working dir (from somewhere, it's not in the git repo), but that failed the verification, so I presume I need something else... EDIT: sl1pkn07: that's with your dropbox link

sl1pkn07 commented on 2016-09-09 20:46 (UTC) (edited on 2016-09-09 20:47 (UTC) by sl1pkn07)

nope. testing 55.2853.0. some strange crash in a random pages builded with gcc and bundled clang. same fails is not ready for release :( sorry guys if any can help. this is the updated pkgbuild https://dl.dropboxusercontent.com/u/6596386/chromium-dev.tar.gz with updated patches for this version greetings

sl1pkn07 commented on 2016-09-09 16:23 (UTC)

Finally can build with GN!. Testing now the versión 54.2840.14, but some extensión Crash the browser. Now have in build the 55.2853.0. if all is good can update the package at last!

misc commented on 2016-09-09 15:55 (UTC) (edited on 2016-09-09 15:57 (UTC) by misc)

If anyone manages to compile a GN version which * doesn't consistently hang (eg. wikipedia, imgur, anything using flash), * downright crash (so far one tab group at a time), and * has all extensions work (eg. ublock, tempermonkey and RES do not) do tell and share the PKGBUILD please. Also if you manage to compile with tcmalloc, cause for me it has chromium still (55.0.2853.0) instantly crash on start.

uzer commented on 2016-09-06 18:56 (UTC) (edited on 2016-09-06 18:57 (UTC) by uzer)

@sl1pkn07 The package builds now, but I'm running into a similar issue that LordAro described: $ chromium-dev Received signal 4 ILL_ILLOPN 555f2a4644e1 #0 0x555f26453407 <unknown> #1 0x7fb2e43b2080 <unknown> #2 0x555f2a4644e1 <unknown> #3 0x555f270812b0 <unknown> #4 0x555f2708181f <unknown> #5 0x555f2707f6b2 <unknown> #6 0x555f2708023d <unknown> #7 0x555f2779dc59 <unknown> #8 0x555f2779a550 <unknown> #9 0x555f277853ef <unknown> #10 0x555f27784f8c <unknown> #11 0x555f294d3e3b <unknown> #12 0x555f294d68e3 <unknown> #13 0x555f294cf8c9 <unknown> #14 0x555f294cf087 <unknown> #15 0x555f2a89e9ef <unknown> #16 0x555f26f19395 <unknown> #17 0x555f264c0014 <unknown> #18 0x555f2a90548b <unknown> #19 0x555f2a90444c <unknown> #20 0x555f264c0014 <unknown> #21 0x555f2646e6f4 <unknown> #22 0x555f2646ea38 <unknown> #23 0x555f2646eddb <unknown> #24 0x555f2647063a <unknown> #25 0x555f264874eb <unknown> #26 0x555f2646dbf7 <unknown> #27 0x555f294f6bd5 <unknown> #28 0x555f264218bb <unknown> #29 0x555f26422883 <unknown> #30 0x555f26421460 <unknown> #31 0x555f25fc9b37 ChromeMain #32 0x7fb2d9d33291 __libc_start_main #33 0x555f25fc99da _start r8: 0000000000000000 r9: 0000000000000000 r10: 0000000000000022 r11: 0000000000000202 r12: 00003f6523a1c1e0 r13: 00003f6523a1c1f8 r14: 00003f6523a68000 r15: 00003f6523a1c1f8 di: 000013194d081000 si: 000000000001e000 bp: 0000000000000004 bx: 00003f6523a1c200 dx: 0000000000000008 ax: ffffffffffffffff cx: ffffffffffffff58 sp: 00007ffdfebc4e40 ip: 0000555f2a4644e1 efl: 0000000000010286 cgf: 002b000000000033 erf: 0000000000000000 trp: 0000000000000006 msk: 0000000000000000 cr2: 0000000000000000 [end of stack trace] libva info: VA-API version 0.39.2 libva info: va_getDriverName() returns 0 libva info: Trying to open /usr/lib/dri/i965_drv_video.so libva info: Found init function __vaDriverInit_0_39 libva info: va_openDriver() returns 0 libva info: VA-API version 0.39.2 libva info: va_getDriverName() returns 0 libva info: Trying to open /usr/lib/dri/i965_drv_video.so libva info: Found init function __vaDriverInit_0_39 libva info: va_openDriver() returns 0

sl1pkn07 commented on 2016-09-05 04:16 (UTC)

try now

uzer commented on 2016-09-05 04:07 (UTC)

enable_vaapi_on_linux.diff ... FAILED

sl1pkn07 commented on 2016-09-01 15:36 (UTC) (edited on 2016-09-01 15:40 (UTC) by sl1pkn07)

in next release, maybe, if finish build OK, is released with GN instead of GYP for the moment: https://bugs.chromium.org/p/chromium/issues/detail?id=643193

misc commented on 2016-08-25 22:23 (UTC)

Here's mine: https://dl.dropboxusercontent.com/u/4607702/chromium-dev.tar.xz Actually the one I used for compiling is still a bit different, hope I didn't break anything in this version. Anyway, I threw out all the stuff I don't need, so it's more of a template only. package() follows extra's chromium, build() is reduced to almost just ninja since that's where the breakages commonly occur, and it's thus useful to have a minimal build() for "makepkg -e".

sl1pkn07 commented on 2016-08-22 04:12 (UTC)

is my idea, but GN don't have all options like GYP (disable gnome stuff, for example)

misc commented on 2016-08-21 18:36 (UTC)

idk, but might be worth switching to gn over gyp

sl1pkn07 commented on 2016-08-19 04:44 (UTC)

any help with this? Updating projects from gyp files... gyp: Dependency '/tmp/makepkg/chromium-dev/src/chromium-54.0.2824.2/components/components.gyp:infobars#target' not found while trying to load target /tmp/makepkg/chromium-dev/src/chromium-54.0.2824.2/components/components.gyp:autofill_core_test_support#target

KittyKatt commented on 2016-08-19 04:37 (UTC)

This was failing to build for me until I used the enable_vaapi_on_linux.diff in saiarcot895's 2785 branch. There was a change made to gpu_video_decode_accelerator_factory that needed a new patch file that is in that branch.

wonko commented on 2016-08-13 08:28 (UTC)

@LordAro Same problem here. Running from the terminal get: [19900:19900:0813/102741:ERROR:browser_main_loop.cc(255)] gdkwindow-x11.c:5555 drawable is not a native X11 window (http://crbug.com/329991) [19900:19900:0813/102741:ERROR:browser_main_loop.cc(255)] gdkwindow-x11.c:5555 drawable is not a native X11 window (http://crbug.com/329991) [19900:19900:0813/102741:ERROR:browser_main_loop.cc(255)] gdkwindow-x11.c:5555 drawable is not a native X11 window (http://crbug.com/329991) Received signal 4 ILL_ILLOPN 562ca9489541 #0 0x562ca53c4cc7 <unknown> #1 0x7f6990380080 <unknown> #2 0x562ca9489541 <unknown> #3 0x562ca6022fc0 <unknown> #4 0x562ca6023533 <unknown> #5 0x562ca6021332 <unknown> #6 0x562ca6021ed1 <unknown> #7 0x562ca67568ba <unknown> #8 0x562ca6753160 <unknown> #9 0x562ca673de4f <unknown> #10 0x562ca673d9dc <unknown> #11 0x562ca84c1e9b <unknown> #12 0x562ca84c4913 <unknown> #13 0x562ca84bd919 <unknown> #14 0x562ca84bd0c7 <unknown> #15 0x562ca98d413f <unknown> #16 0x562ca5eb4c45 <unknown> #17 0x562ca54334b4 <unknown> #18 0x562ca993be3b <unknown> #19 0x562ca993adec <unknown> #20 0x562ca54334b4 <unknown> #21 0x562ca53e03b4 <unknown> #22 0x562ca53e06f8 <unknown> #23 0x562ca53e0a9b <unknown> #24 0x562ca53e22ea <unknown> #25 0x562ca53f952c <unknown> #26 0x562ca53df8b7 <unknown> #27 0x562ca84e4b99 <unknown> #28 0x562ca53925eb <unknown> #29 0x562ca53935c3 <unknown> #30 0x562ca53921b0 <unknown> #31 0x562ca4f312f7 ChromeMain #32 0x7f6985acc291 __libc_start_main #33 0x562ca4f3119a _start r8: 0000000000000000 r9: 0000000000000000 r10: 0000000000000022 r11: 0000000000000202 r12: 00002d3302e1c1e0 r13: 0000000000000001 r14: 00002d3302e68000 r15: 00002d3302e1c1f8 di: 0000397017601000 si: 000000000001e000 bp: 0000000000000004 bx: 00002d3302e1c200 dx: 0000000000000008 ax: ffffffffffffffff cx: ffffffffffffff58 sp: 00007ffe0263fd20 ip: 0000562ca9489541 efl: 0000000000010286 cgf: 002b000000000033 erf: 0000000000000000 trp: 0000000000000006 msk: 0000000000000000 cr2: 0000000000000000 [end of stack trace] libva info: VA-API version 0.39.2 libva info: va_getDriverName() returns 0 libva info: Trying to open /usr/lib/dri/i965_drv_video.so libva info: Found init function __vaDriverInit_0_39 libva info: va_openDriver() returns 0 libva info: VA-API version 0.39.2 libva info: va_getDriverName() returns 0 libva info: Trying to open /usr/lib/dri/i965_drv_video.so libva info: Found init function __vaDriverInit_0_39 libva info: va_openDriver() returns 0

LordAro commented on 2016-08-11 18:07 (UTC) (edited on 2016-08-11 20:31 (UTC) by LordAro)

For the last week or so chromium-dev has been straight up broken for me. It starts up, but "Aw snap"s everywhere and all the extensions crash as well. The following appears if I run it from the terminal: https://gist.github.com/anonymous/ad7235321c92324872613689ff510766 (I compiled with debugging to see if it revealed anything) I've tried downgrading, but with the same results, so presumably it's some dependency that's been messed up. Anyone else been seeing this? EDIT: google-chrome-dev 54.x.x.x is fine, couldn't get chromium 54 from below to compile

cdesai commented on 2016-08-09 07:00 (UTC)

I have done an update to 54.0.2816.0 (latest linux dev version as of now), available at https://git.io/v6n4Y It's based on the gentoo 54.0.2810.2 ebuild. I have updated the vaapi patch, there was a file rename. Getting an error with the clang compilation, http://sprunge.us/IccW Leaving this comment here in case anybody wants to pick up on this.

misc commented on 2016-08-05 18:14 (UTC)

@prazola Actually it's worth checking if a manual symlink suffices, just don't forget to remove it eventually.

constroy commented on 2016-08-05 12:48 (UTC)

How to disable the speech-dispatcher?

sl1pkn07 commented on 2016-08-05 04:06 (UTC)

yep :(

prazola commented on 2016-08-05 01:11 (UTC)

Should I recompile with extra/libvpx? /usr/lib/chromium-dev/chromium-dev: error while loading shared libraries: libvpx.so.3: cannot open shared object file: No such file or directory

sl1pkn07 commented on 2016-08-04 13:46 (UTC)

-> Descargando enable_vaapi_on_linux.diff… % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 100 26784 100 26784 0 0 64317 0 --:--:-- --:--:-- --:--:-- 64230 -> minizip.patch ha sido encontrado -> chromium-widevine-r1.patch ha sido encontrado ==> Validando los archivos source con sha1sums… chromium-53.0.2785.8.tar.xz ... Aprobado chromium-launcher-3.tar.gz ... Aprobado chromium-dev.svg ... Aprobado chromium-system-ffmpeg-r2.patch ... Aprobado chromium-system-jinja-r11.patch ... Aprobado chromium-system-zlib-r0.patch ... Aprobado enable_vaapi_on_linux.diff ... Aprobado minizip.patch ... Aprobado chromium-widevine-r1.patch ... Aprobado ==> Eliminando el directorio $srcdir/…

ogarcia commented on 2016-08-04 09:38 (UTC)

==> Validating source files with sha1sums... chromium-53.0.2785.8.tar.xz ... Passed chromium-launcher-3.tar.gz ... Passed chromium-dev.svg ... Passed chromium-system-ffmpeg-r2.patch ... Passed chromium-system-jinja-r11.patch ... Passed chromium-system-zlib-r0.patch ... Passed enable_vaapi_on_linux.diff ... FAILED minizip.patch ... Passed chromium-widevine-r1.patch ... Passed There is an error in sha1sum of enable_vaapi_on_linux.diff

usuariopolivalen commented on 2016-08-03 23:31 (UTC)

has this chromium hardware acceleration?

usuariopolivalen commented on 2016-08-03 23:29 (UTC)

error of compiling => Compilando e instalando el paquete ==> Creando el paquete: chromium-dev 53.0.2783.2-1 (jue ago 4 01:27:37 CEST 2016) ==> Comprobando dependencias mientras se ejecuta… ==> Comprobando dependencias mientras se compila… ==> Recibiendo las fuentes… -> chromium-53.0.2783.2.tar.xz ha sido encontrado -> chromium-launcher-3.tar.gz ha sido encontrado -> chromium-dev.svg ha sido encontrado -> chromium-system-ffmpeg-r2.patch ha sido encontrado -> chromium-system-jinja-r11.patch ha sido encontrado -> chromium-system-zlib-r0.patch ha sido encontrado -> chromium-linker-warnings-r0.patch ha sido encontrado -> enable_vaapi_on_linux.diff ha sido encontrado -> minizip.patch ha sido encontrado -> chromium-widevine-r1.patch ha sido encontrado ==> Validando los archivos source con sha1sums… chromium-53.0.2783.2.tar.xz ... Aprobado chromium-launcher-3.tar.gz ... Aprobado chromium-dev.svg ... Aprobado chromium-system-ffmpeg-r2.patch ... Aprobado chromium-system-jinja-r11.patch ... Aprobado chromium-system-zlib-r0.patch ... Aprobado chromium-linker-warnings-r0.patch ... Aprobado enable_vaapi_on_linux.diff ... HA FALLADO minizip.patch ... Aprobado chromium-widevine-r1.patch ... Aprobado ==> ERROR: ¡Uno o más archivos no superaron el control de validación! ==> ERROR: Makepkg no ha podido compilar chromium-dev. ==> ¿Reiniciar la compilación de chromium-dev? [s/N] ==> ------------------------------------------------ ==>

misc commented on 2016-07-31 18:41 (UTC) (edited on 2016-07-31 19:16 (UTC) by misc)

For the dumb dbus/dbus.h failure, this http://pastebin.com/raw/RVNp60mk reverts the commits "Geolocation cleanup: run clang-format" and "Reland 3: Geolocation: move from content/browser to device/". Probably some minor gyp oversight that's the actual issue. This applies to git head though, so maybe you only need to revert the latter commit, which can be found at https://chromium.googlesource.com/chromium/src.git/+/4e216e71b1e151982bc0b08919b3264febf6e46a Same for this dumb failure at "skia_commit_hash.h": http://pastebin.com/raw/H6WzZ5e9 Also a bunch of changes to the catapult lines in _neces*s*ary, mine now being: 'third_party/catapult' 'third_party/catapult/third_party/py_vulcanize' 'third_party/catapult/third_party/py_vulcanize/third_party/rcssmin' 'third_party/catapult/third_party/py_vulcanize/third_party/rjsmin' 'third_party/catapult/third_party/polymer' 'third_party/catapult/tracing/third_party/d3' 'third_party/catapult/tracing/third_party/gl-matrix' 'third_party/catapult/tracing/third_party/jszip' 'third_party/catapult/tracing/third_party/mannwhitneyu'

LordAro commented on 2016-07-28 07:34 (UTC)

-> Starting building Chromium... Updating projects from gyp files... ninja: Entering directory `out/Release' ninja: error: '../../native_client/toolchain/linux_x86/pnacl_newlib/pnacl_newlib.json', needed by 'gen/tc_irt/lib64/libppapi_proxy_nacl.a', missing and no known rule to make it ==> ERROR: A failure occurred in build(). Aborting... ==> ERROR: Makepkg was unable to build chromium-dev.

sl1pkn07 commented on 2016-06-26 15:29 (UTC)

@misc thanks for the patch. but without patch build for me ok greetings

misc commented on 2016-06-23 15:45 (UTC)

Dunno if 53.0.2774.3 is already affected by… ../../content/public/common/common_param_traits_macros.h:25:92: fatal error: third_party/WebKit/public/platform/modules/permissions/permission_status.mojom.h: file not found … but anyway, solved for me with this patch: http://pastebin.com/raw/xz5ghExY

ogarcia commented on 2016-06-17 09:44 (UTC)

@sl1pkn07 The error is here http://pastebin.com/CtSr29jS The full log can be donwloaded from here https://db.tt/FX5GtMOR

sl1pkn07 commented on 2016-06-16 15:02 (UTC)

@ogracia: please paste in a pastebin the error greetings

ogarcia commented on 2016-06-16 13:55 (UTC) (edited on 2016-06-16 13:56 (UTC) by ogarcia)

@sl1pkn07 I don't know why, but if I try to make package without gtk2 installed gives me missing gtk2 error :( You can make test with a clean build environment using chroot: https://wiki.archlinux.org/index.php/DeveloperWiki:Building_in_a_Clean_Chroot

TingPing commented on 2016-06-15 23:00 (UTC)

Just wanted to note that: If _use_bundled_clang is disabled system clang will fail to build. Probably requires newer development version. If _use_clang is disabled the resulting binary just constantly crashes with memory access errors.

sl1pkn07 commented on 2016-06-14 18:48 (UTC) (edited on 2016-06-14 18:51 (UTC) by sl1pkn07)

gtk2 for what? _use_gtk3=1 # If set 1, then build with GTK3 support, if set 0, then build with GTK2 .... ... .. # Build with GTK3? if [ "${_use_gtk3}" = "1" ]; then depends+=('gtk3') _launcher_gtk='GTK=3' optdepends+=('libappindicator-gtk3: Needed for show systray icon in the panel in plasma-next (KF5)') elif [ "${_use_gtk3}" = "0" ]; then depends+=('gtk2') optdepends+=('libappindicator-gtk2: Needed for show systray icon in the panel in plasma-next (KF5)') fi

ogarcia commented on 2016-06-14 18:40 (UTC)

You forget gtk2 as build dependency

bryanparadis commented on 2016-06-01 00:30 (UTC) (edited on 2016-06-01 00:37 (UTC) by bryanparadis)

When building this package make sure you have at least 9GB of space on /tmp. Build finished but makepkg didn't have enough room to generate the packages with a 7.8GB /tmp. Managed to recover from it without rebuilding. /tmp will be by default 50% of your ram. You can override this by adding an fstab entry with the size option.

neeks commented on 2016-05-19 01:56 (UTC) (edited on 2016-05-19 01:56 (UTC) by neeks)

I've been getting errors the past two weeks in `third_party/WebKit/Source/build/scripts/make_css_value_keywords.py` and `third_party/WebKit/Source/build/scripts/make_css_property_names.py`. Turns out that the gperf package - 3.0.4.7 - that was pushed to `extra` 16 days ago is to blame. Downgrading gperf to 3.0.4.6 solved those specific build errors.

j1simon commented on 2016-05-18 18:01 (UTC)

Yes, I have disabled the [multilib] repository. $ pacaur -y chromium-dev :: resolving dependencies... :: no results found for lib32-gcc-libs :: no results found for lib32-zlib chrome://nacl -> http://pix.toile-libre.org/upload/original/1463594452.png

sl1pkn07 commented on 2016-05-18 17:55 (UTC)

in a clean environment?

j1simon commented on 2016-05-18 17:54 (UTC)

@sl1pkn07 I've compiled it with the line ("makedepends_x86_64=('lib32-gcc-libs' 'lib32-zlib')") commented without problems and it compiles NaCl plugin.

sl1pkn07 commented on 2016-05-18 13:22 (UTC)

because is need for build the NaCL stack

j1simon commented on 2016-05-18 10:21 (UTC)

Why "makedepends_x86_64=('lib32-gcc-libs' 'lib32-zlib')"?

LinguinePenguiny commented on 2016-04-23 08:09 (UTC)

Compile proxy/plugin_message_filter.cc [FLAGS I REMOVED] Traceback (most recent call last): File "../native_client/build/build_nexe.py", line 1136, in <module> sys.exit(Main(sys.argv)) File "../native_client/build/build_nexe.py", line 1108, in Main out = build.Compile(filename) File "../native_client/build/build_nexe.py", line 574, in Compile self.CleanOutput(outd) File "/tmp/yaourt-tmp-aaron/aur-chromium-dev/src/chromium-51.0.2704.19/native_client/build/build_nexe_tools.py", line 81, in CleanOutput if IsFile(out): File "/tmp/yaourt-tmp-aaron/aur-chromium-dev/src/chromium-51.0.2704.19/native_client/build/build_nexe_tools.py", line 41, in IsFile return os.path.isfile(FixPath(path)) File "/tmp/yaourt-tmp-aaron/aur-chromium-dev/src/chromium-51.0.2704.19/native_client/build/build_nexe_tools.py", line 32, in FixPath path = os.path.normpath(os.path.join(os.getcwd(), path)) OSError: [Errno 2] No such file or directory ninja: build stopped: subcommand failed. ==> ERROR: A failure occurred in build(). It's not compiling, still.

sl1pkn07 commented on 2016-04-22 13:23 (UTC)

@juston_li done tnx

juston_li commented on 2016-04-22 02:17 (UTC) (edited on 2016-04-22 02:17 (UTC) by juston_li)

install: cannot stat 'chromium-51.0.2704.19/chrome/installer/linux/common/default-app.tempate': No such file or directory Looks like just a typo on line 627 tempate -> template

wicast commented on 2016-04-21 13:09 (UTC)

MimeType is missing in .desktop file. xdg-open is not working well.

axfelix commented on 2016-04-19 16:33 (UTC)

h264 VAAPI hardware decoding is working again with the most recent build (51.0.2704)! Thanks, folks.

axfelix commented on 2016-04-15 16:51 (UTC)

Just confirmed it's back to working for flash video containers, at least, just not HTML5. I think this was the case up until Chromium 44-45 or so, when HTML5 h264 would actually work with the vaapi patches, but it seems that lasted a fairly short time.

axfelix commented on 2016-04-15 16:38 (UTC)

Build works with the minizip patch and it's back to showing Video Decode as hardware accelerated in chrome://gpu with the new vaapi patch (it went back to showing software only in 50) but GPU decoding still isn't kicking in after that change, in 51.

misc commented on 2016-04-10 17:08 (UTC)

Patch for the minizip update: http://pastebin.com/raw/QCqSDam5 If I understand correctly, the old version interpreted a 0 as sign to pick strcmp, the new one directly takes a function pointer and falls to strcmp when it's NULL.

guiodic commented on 2016-04-06 19:34 (UTC)

"Seems like GPU video decoding patches may have stopped working between 49 and 50..." Yes, it's true. But when I run chromium from the cli I get the va-api message. Nevertheless chromium takes a lot of cpu. Maybe an upstream bug related to the switch to GTK3.

axfelix commented on 2016-03-24 22:16 (UTC)

Seems like GPU video decoding patches may have stopped working between 49 and 50...

sl1pkn07 commented on 2016-03-06 22:39 (UTC)

yeah!

misc commented on 2016-03-06 21:51 (UTC)

I don't have the line '-Dicu_use_data_file_flag=0', just try after killing it entirely.

sl1pkn07 commented on 2016-03-06 21:32 (UTC)

obj/mojo/shell/runner/mojo_runner_host_lib.init.o: In function `mojo::shell::CallLibraryEarlyInitialization(void*)': init.cc:(.text._ZN4mojo5shell30CallLibraryEarlyInitializationEPv+0x16): undefined reference to `base::i18n::GetRawIcuMemory()' mmmm

moscar commented on 2016-03-06 21:11 (UTC)

This currently fails to build if gtk2 is not installed: Updating projects from gyp files... Package gdk-2.0 was not found in the pkg-config search path. Perhaps you should add the directory containing `gdk-2.0.pc' to the PKG_CONFIG_PATH environment variable No package 'gdk-2.0' found gyp: Call to 'pkg-config --cflags gdk-2.0' returned exit status 1 while in /sources/chromium-dev/src/chromium-50.0.2633.3/build/linux/system.gyp. ==> ERROR: A failure occurred in build(). I'm guessing the `-Duse_gtk3` flag is not enough, or something else is missing.

misc commented on 2016-03-06 15:39 (UTC)

One revert and a workaround I need to compile ATM. Dunno if the reverted commit already made it into this tarball. http://pastebin.com/raw/2FKGi944 ← revert http://pastebin.com/raw/nz3KrTbc ← system_icu=1 issue

moscar commented on 2016-01-30 06:05 (UTC)

imagemagick should be in makedepends since the convert command is used here: https://aur.archlinux.org/cgit/aur.git/tree/PKGBUILD?h=chromium-dev#n453

juston_li commented on 2016-01-27 17:40 (UTC) (edited on 2016-01-27 17:46 (UTC) by juston_li)

The constant was removed in a cleanup patch that sl1pkn07 linked in the issue tracker https://chromium.googlesource.com/chromium/src/+/8921a7faaf42c6e216ad3de32093a0c3a121e696. Looks to be a rather silly and trivial error, not sure why they haven't fixed or how it got though review. The commit simply removed the declaration of "kDefaultTintFrameInactive" but left a reference to it on line 1226. Not sure if the constant was accidentally removed or they forgot to remove the reference, I'll try and look into it but safer way to fix is just to add the constant again. Here's the simple patch tested working: https://drive.google.com/file/d/0BxTqqkmPl8mTODFLM29nM3lFc3c/view?usp=sharing

prazola commented on 2016-01-27 16:30 (UTC)

"kDefaultTintFrameInactive" should be a constant indicating a color. Have you checked if the constant is missing in the some theme source inside the project?

sl1pkn07 commented on 2016-01-25 01:04 (UTC)

@corepoint before, the patch is applied with this: https://aur.archlinux.org/cgit/aur.git/commit/?h=chromium-dev&id=cf179b8e640c8ed6cc0600fae730d6e60f7d1515 (the sed line) now, using the chromium [extra] patch. (more elaborated than my "patch") for widevine(netflix)support in this package you need install this package: https://aur.archlinux.org/packages/chromium-widevine-dev/ I not have netflix account, but working with this example page http://www.dash-player.com/demo/drm-test-area/ greetings

CorePoint commented on 2016-01-24 17:55 (UTC)

@sl1pkn07 Hey, can you elaborate? They don't seem to have widevine in their repository, so i'm out of luck?

sl1pkn07 commented on 2016-01-22 05:54 (UTC)

congrats! greetings

juston_li commented on 2016-01-22 05:44 (UTC)

Built fine this time Thanks

juston_li commented on 2016-01-22 04:12 (UTC)

32GB ram Though it might have been full as I had just built 2623.0 then it got updated to 2623.13. If yaourt didn't clear it then having both packages might have been a bit too much =p I'll try again with /tmp clear Thanks

sl1pkn07 commented on 2016-01-22 03:56 (UTC) (edited on 2016-01-22 04:00 (UTC) by sl1pkn07)

maybe a insufficient space? :S http://wstaw.org/m/2016/01/22/Screenshot_20160122_045937.png

juston_li commented on 2016-01-22 02:45 (UTC)

2623.13 build fails at the end for me [19398/19407] =/ http://pastebin.com/pW14MEcK

sl1pkn07 commented on 2016-01-15 17:51 (UTC) (edited on 2016-01-15 17:54 (UTC) by sl1pkn07)

@Prazola: in this moment, no https://code.google.com/p/chromium/issues/detail?id=575038 @CorePoint: that patch is applied, but in another way

prazola commented on 2016-01-15 17:50 (UTC)

No more GTK3 default in last build? :D

CorePoint commented on 2016-01-13 10:37 (UTC) (edited on 2016-01-13 10:37 (UTC) by CorePoint)

Hey, i would really like to have a binary for this in the aur too. The package in the archlinuxcn fails too use widevine. When i install chromium-widevine-dev from the aur, it shows up in chrome://plugins but not in chrome://components. So netflix shows its usual error "missing widevine, make sure to enable it blah". It seems they didn't inculde this patch. https://code.google.com/p/chromium/issues/detail?id=429452#c16 Just telling, because maybe nobody saw a reason to make a chromiun-dev binary aur, and now has a reason. Have a nice day

sl1pkn07 commented on 2015-12-19 10:52 (UTC)

@misc with pnacl?

misc commented on 2015-12-18 18:40 (UTC)

I've been able to use the -lite.tar.xz tarballs without issue, which are still around 450 MiB.

sl1pkn07 commented on 2015-12-16 16:51 (UTC)

yes and no https://groups.google.com/a/chromium.org/forum/#!topic/chromium-packagers/c6lVrV8IuwA/discussion

prazola commented on 2015-12-16 16:11 (UTC)

The source tar.xz is about 1.4GB instead of the usal 380~ MB. Is it ok? :/

axfelix commented on 2015-12-06 20:47 (UTC)

@skzzorz thanks for finding that link -- looks like it is indeed explicitly disabled on mesa.

sflor commented on 2015-12-01 17:15 (UTC)

Does anybody know if there's a binary build with GTK3 support?

sl1pkn07 commented on 2015-11-12 17:50 (UTC)

nope, chromium-dev still need bundled libvpx (it needs headers not included in libvpx 1.5) 'vpx/vpx_codec_impl_bottom.h', 'vpx/vpx_codec_impl_top.h', greetings

prazola commented on 2015-11-12 14:57 (UTC)

Does 48.0.2560.0-1 build support libvpx 1.5?

prazola commented on 2015-11-10 16:21 (UTC)

Wow, latest nightly builds are a bit to heavy :D https://download-chromium.appspot.com/dl/Linux_x64 The executable reach 2gb when uncompressed. Do they have some build problem?

skzzorz commented on 2015-11-08 01:36 (UTC)

@axfelix a qucik search at cs.chromium.org leads to: https://code.google.com/p/chromium/codesearch#chromium/src/content/browser/gpu/browser_gpu_memory_buffer_manager.cc&sq=package:chromium&l=102&rcl=1446928296 it doesn't look like it's disabled by OS-related macro.

axfelix commented on 2015-11-05 22:04 (UTC)

I'm noticing that the --enable-native-gpu-memory-buffers flag doesn't seem to be working on Linux in newest chromium-dev despite the fact that this functionality *should* now be there; is it possible that this is blacklisted at compile time on non-ChromeOS platforms the way that vaapi was? Might be fun to have a look into that...

axfelix commented on 2015-11-03 23:52 (UTC)

Thanks!

skzzorz commented on 2015-10-29 09:59 (UTC)

@quartz55 @axfelix Pre-compiled builds for x86-64 with VA-API patch can be found at archlinuxcn's unofficial community repo at repo.archlinuxcn.org

ethail commented on 2015-10-27 11:46 (UTC) (edited on 2015-10-27 21:59 (UTC) by ethail)

Looking at the post on the chrome releases blog [0] for this release, shouldn't we be using 48.0.2541.0 instead of 48.0.2540.0? Not that it matters that much, as a release should happen shortly, I guess, but maybe for a future we could pay a little more attention when two versions targeted at diferent OSes are released. [0] http://googlechromereleases.blogspot.com.es/2015/10/dev-channel-update_20.html EDIT (a few hours later): I realized my tone could get misinterpreted, by "we" I mean everyone who uses this package and cares even a little about it, not just the maintainer.

prazola commented on 2015-10-21 12:10 (UTC)

I have a script for precompiled snapshot. I prefer to compile this package myself.

danbruegge commented on 2015-10-20 07:41 (UTC)

@axfelix +1 for Precompiled builds

quartz55 commented on 2015-10-17 16:52 (UTC)

@axfelix Precompiled builds would be awesome, it does take quite a while for the builds to finish :p

axfelix commented on 2015-09-08 00:36 (UTC)

Any interest in doing precompiled builds? I used this because I wanted the vaapi patch (super thrilled to someone maintaining it in the AUR, think you're the only one), and it works great, but man, a Chromium build is not for the faint of heart :)

sl1pkn07 commented on 2015-08-31 20:17 (UTC)

You can't wait?

mortan commented on 2015-08-31 19:06 (UTC)

@sl1pkn07: Any plans to update this package?

sl1pkn07 commented on 2015-07-18 23:30 (UTC)

yes. i know the situation. for the moment, just remove python2-ply, build chromium, and then reinstall. the patch for fix the issue still no merge https://github.com/dabeaz/ply/issues/66 https://bugs.archlinux.org/task/45625

emerino commented on 2015-07-18 23:22 (UTC)

Just to give you a heads up, current python-ply 3.6 WON'T work, it'll cause an error at compile time. You have to either downgrade to 3.4 or use python-ply's git version. https://github.com/dabeaz/ply/issues/63

andre.vmatos commented on 2015-07-06 23:37 (UTC)

Anyone has a repository or could post somewhere x86_64 version of this package, please? Thank you.

prazola commented on 2015-06-06 16:16 (UTC)

I was missing some gcc 32bit libs, I'm recompiling right now.

prazola commented on 2015-06-06 14:15 (UTC)

Wow, no, not installed. Idk why, I'm upgrading from a previous version of you package, really strange. Thx Core2 stop on "Compile posix/nacl_timestamp.c" without informations about the error.

sl1pkn07 commented on 2015-06-06 13:59 (UTC)

you have installed bison? (strange, because is part of base-devel (mandatory to build any AUR package)

prazola commented on 2015-06-06 13:23 (UTC)

i7 error http://pastebin.com/T5F4Z1iV

sl1pkn07 commented on 2015-06-06 12:39 (UTC)

log?

prazola commented on 2015-06-06 12:10 (UTC)

Build problems on two different platforms (native, o3) ninja subcommand failed. Build dir is in my home. Config 1: core2 p8600 - 4GB ddr2 - ssd 240gb Config 2: i7 4790k - 16GB ddr3 - ssd 120gb Config 3: i5 450m - 8GB ddr3 - hdd 640gb -> build OK! Python2-ply not installed, so what should I do?

vdemin commented on 2015-06-03 06:51 (UTC)

@foutrelis, you are right. I changed location of the build folder from tmpfs to hard drive and decreased the tmpfs size, and compilation finished successfully. Thank you for the help!

foutrelis commented on 2015-06-02 20:44 (UTC)

> c++: internal compiler error: Killed (program cc1plus) That indicates that you're running out of memory during compilation.

sl1pkn07 commented on 2015-06-02 20:34 (UTC)

build ok here (x86_64) snip the clang build log http://sl1pkn07.wtf/paste/view/561d7015

vdemin commented on 2015-06-02 20:03 (UTC)

I can't build the package on x86_64: http://pastebin.com/CZJsQh60

hotaru commented on 2015-05-06 16:13 (UTC)

yeah, if you remove '--insecure' it should be ok.

sl1pkn07 commented on 2015-05-06 15:59 (UTC)

oh. you mean the '--insecure' flag?. is removed in the next version (now in build/test)

hotaru commented on 2015-05-06 15:01 (UTC)

> sha1sums=("$(curl -sL "https://commondatastorage.googleapis.com/chromium-browser-official/chromium-${pkgver}.tar.xz.hashes" --insecure | grep sha1 | cut -d " " -f3)" doesn't that completely negate the point of using sha1sums?

sl1pkn07 commented on 2015-05-03 12:29 (UTC)

yes. that is but for the moment, the plugin don't work in a linux platform

juston_li commented on 2015-05-03 03:34 (UTC)

Hmm seems like widevine build included again, got this error: error: failed to commit transaction (conflicting files) chromium-dev: /usr/lib/chromium-dev/libwidevinecdm.so exists in filesystem Removed'chromium-widevine-dev' package then I could upgrade. chromium-dev shows widevine in chrome://plugins

eggze commented on 2015-04-29 11:44 (UTC)

It helped, thank you.

sl1pkn07 commented on 2015-04-27 13:23 (UTC)

uninstall python2-ply first

eggze commented on 2015-04-27 12:31 (UTC)

Build failed on my machine with similar error as katta's one: FAILED: cd ../../ui/accessibility; python ../../tools/json_schema_compiler/compiler.py "ax_enums.idl" "--root=../.." "--destdir=../../out/Release/gen" "--namespace=ui" "--generator=cpp" "--include-rules=" Traceback (most recent call last): File "../../tools/json_schema_compiler/compiler.py", line 191, in <module> opts.namespace, opts.impl_dir, include_rules) File "../../tools/json_schema_compiler/compiler.py", line 53, in GenerateSchema api_def = schema_loader.LoadSchema(schema) File "/home/jprokop/Downloads/builds/chromium-dev/src/chromium-44.0.2376.0/tools/json_schema_compiler/schema_loader.py", line 90, in LoadSchema api_defs = idl_schema.Load(schema_path) File "/home/jprokop/Downloads/builds/chromium-dev/src/chromium-44.0.2376.0/tools/json_schema_compiler/idl_schema.py", line 525, in Load return Process(contents, filename) File "/home/jprokop/Downloads/builds/chromium-dev/src/chromium-44.0.2376.0/tools/json_schema_compiler/idl_schema.py", line 535, in Process idl = idl_parser.IDLParser().ParseData(contents, filename) File "/home/jprokop/Downloads/builds/chromium-dev/src/chromium-44.0.2376.0/tools/json_schema_compiler/../../ppapi/generators/idl_parser.py", line 906, in __init__ IDLLexer.__init__(self) File "/home/jprokop/Downloads/builds/chromium-dev/src/chromium-44.0.2376.0/tools/json_schema_compiler/../../ppapi/generators/idl_lexer.py", line 215, in __init__ self.lexobj = lex.lex(object=self, lextab=None, optimize=0) File "/usr/lib/python2.7/site-packages/ply/lex.py", line 893, in lex if '.' not in lextab: TypeError: argument of type 'NoneType' is not iterable ninja: build stopped: subcommand failed. My version of python2-ply is 3.6-1

sl1pkn07 commented on 2015-04-26 23:20 (UTC)

ok, is because chromium build system don't like the new python2-ply 1.5.0 try now

katta commented on 2015-04-26 15:48 (UTC)

build failed: [1226/17650] cd ../../third_party/zlib; python ../../tools/generate_shim_headers/generate_shim_headers.py --headers-root . --output-directory ../../out/Release/gen/shim_headers/zlib/target zlib.h --generate FAILED: cd ../../third_party/WebKit/Source/bindings/scripts; python blink_idl_parser.py ../../../../../out/Release/gen/blink/bindings/scripts Traceback (most recent call last): File "blink_idl_parser.py", line 456, in <module> sys.exit(main(sys.argv)) File "blink_idl_parser.py", line 452, in main parser = BlinkIDLParser(outputdir=outputdir, rewrite_tables=True) File "blink_idl_parser.py", line 428, in __init__ picklefile=picklefile) File "/usr/lib/python2.7/site-packages/ply/yacc.py", line 3232, in yacc read_signature = lr.read_pickle(picklefile) File "/usr/lib/python2.7/site-packages/ply/yacc.py", line 1990, in read_pickle in_f = open(filename, 'rb') IOError: [Errno 2] No such file or directory: '../../../../../out/Release/gen/blink/bindings/scripts/parsetab.pickle' ninja: build stopped: subcommand failed.

c00w commented on 2015-04-24 03:05 (UTC)

I don't think HIDPI support is working anymore.

sl1pkn07 commented on 2015-04-19 10:06 (UTC)

@juston_li fixed also now use a chromium-launcher tarball instead of git and fixed minnor issues greetings

juston_li commented on 2015-04-19 00:27 (UTC)

Got this error trying to run chromium-dev, compiled with GCC: Can't locate File/BaseDir.pm in @INC (you may need to install the File::BaseDir module) (@INC contains: /usr/lib/perl5/site_perl /usr/share/perl5/site_perl /usr/lib/perl5/vendor_perl /usr/share/perl5/vendor_perl /usr/lib/perl5/core_perl /usr/share/perl5/core_perl .) at /usr/bin/chromium-dev line 16. BEGIN failed--compilation aborted at /usr/bin/chromium-dev line 16. Fixed by installing perl-file-basedir

sl1pkn07 commented on 2015-04-18 21:12 (UTC)

update to 44 series - now uses a launcher like a chromium in [extra]. please check the instruction after installation - fix the NaCL installation - for the moment now use a system clang instead of build (fail build) - build widevine lib not need anymore. please install a chromium-widevine/chromium-widevine-dev instead greetings

walkindude commented on 2015-04-10 19:11 (UTC)

@Valeth no need, looks like the latest build solved this issue

Valeth commented on 2015-04-09 19:41 (UTC)

@walkindude try to launch chromium-dev with the MESA_GL_VERSION_OVERRIDE environment variable set to 4.0

sl1pkn07 commented on 2015-04-08 19:58 (UTC)

ok. in the next release greetings @walkindude. sorry. try add the issue in crbug.com (chromium issue tracker). i can't help you :(

juston_li commented on 2015-04-08 19:41 (UTC)

Should probably make ccache usage optional by making it a build option. It takes up about 1GB to storage compiling with GCC so I feel it should require explicit action to enable. That aside, I've been using ccache for awhile now and if you don't mind giving up some disk space, ccache really speeds things up. Compile times from 40-50m down to 5m with ccache on an SSD

walkindude commented on 2015-04-08 14:45 (UTC)

... [8293:8293:0408/174044:ERROR:logger.cc(46)] Too many GL errors, not reporting any more for this context. use --disable-gl-error-limit to see all errors. [8293:8293:0408/174044:ERROR:gles2_cmd_decoder.cc(11546)] [GroupMarkerNotSet(crbug.com/242999)!:D0D63D32FF3B0000]GL ERROR :GL_INVALID_OPERATION : glTexStorage2DEXT: <- error from previous GL command [8293:8293:0408/174044:ERROR:shader_manager.cc(128)] Shader translator allowed/produced an invalid shader unless the driver is buggy: --Log from shader translator-- --original-shader-- #define TexCoordPrecision highp attribute TexCoordPrecision vec4 a_position; attribute TexCoordPrecision vec2 a_texCoord; attribute float a_index; uniform mat4 matrix; uniform TexCoordPrecision vec2 quad[4]; uniform TexCoordPrecision vec4 vertexTexTransform; varying TexCoordPrecision vec2 v_texCoord; void main() { vec2 pos = quad[int(a_index)]; gl_Position = matrix * vec4(pos, a_position.z, a_position.w); v_texCoord = a_texCoord * vertexTexTransform.zw + vertexTexTransform.xy; } --translated-shader-- #version 150 in vec4 a_position; in vec2 a_texCoord; in float a_index; uniform mat4 matrix; uniform vec2 quad[4]; uniform vec4 vertexTexTransform; out vec2 v_texCoord; void main(){ vec2 pos = quad[int(clamp(float(int(a_index)), 0.0, float(3)))]; (gl_Position = (matrix * vec4(pos, a_position.z, a_position.w))); (v_texCoord = ((a_texCoord * vertexTexTransform.zw) + vertexTexTransform.xy)); } --info-log-- 0:1(10): error: GLSL 1.50 is not supported. Supported versions are: 1.10, 1.20, 1.30, 1.00 ES, and 3.00 ES [8293:8293:0408/174044:ERROR:shader_manager.cc(128)] Shader translator allowed/produced an invalid shader unless the driver is buggy: --Log from shader translator-- --original-shader-- #define TexCoordPrecision mediump #define SamplerType sampler2D #define TextureLookup texture2D #define ApplyBlendMode(X, Y) (X) precision mediump float; varying TexCoordPrecision vec2 v_texCoord; uniform SamplerType s_texture; uniform float alpha; void main() { vec4 texColor = TextureLookup(s_texture, v_texCoord); gl_FragColor = ApplyBlendMode(texColor * alpha, 0.0); } --translated-shader-- #version 150 out vec4 webgl_FragColor; in vec2 v_texCoord; uniform sampler2D s_texture; uniform float alpha; void main(){ vec4 texColor = texture(s_texture, v_texCoord); (webgl_FragColor = (texColor * alpha)); } --info-log-- 0:1(10): error: GLSL 1.50 is not supported. Supported versions are: 1.10, 1.20, 1.30, 1.00 ES, and 3.00 ES [8251:8251:0408/174054:ERROR:extension_downloader.cc(695)] Invalid URL: '' for extension dgpdioedihjhncjafcpgbbjdpbbkikmi [8251:8488:0408/174055:ERROR:get_updates_processor.cc(243)] PostClientToServerMessage() failed during GetUpdates from the latest build I have: walkindude@Desktop> glxinfo | grep version ~ server glx version string: 1.4 client glx version string: 1.4 GLX version: 1.4 OpenGL core profile version string: 3.3 (Core Profile) Mesa 10.6.0-devel (git-eb51c6d) OpenGL core profile shading language version string: 3.30 OpenGL version string: 3.0 Mesa 10.6.0-devel (git-eb51c6d) OpenGL shading language version string: 1.30 OpenGL ES profile version string: OpenGL ES 3.0 Mesa 10.6.0-devel (git-eb51c6d) OpenGL ES profile shading language version string: OpenGL ES GLSL ES 3.00 no more chromium for me?

harryharryharry commented on 2015-04-06 10:25 (UTC)

@sl1pkn07 Yes I already changed the makeflags. But it's okay, at least I now know I should take a beer in the meantime :)

sl1pkn07 commented on 2015-04-05 22:27 (UTC)

@harryharryharry yes. is normal. but can try edit the /etc/makepkg.conf and edit/comment out the MAKEFLAGS= variable with -jyour_number_of_cores (-j4 for quad-core or -j8 if have quad-core with HT) (if you had it before) the pkgbuild has been fixed greething and sorry

harryharryharry commented on 2015-04-05 21:35 (UTC)

@sl1pkn07 I had the same error as rekoil, gokcen & daurnimator. Adding ccache as a dependency solved it for me. Thanks! The build took over 6 hours (with quad-core, 4GB ram) is that normal ?

sl1pkn07 commented on 2015-04-05 12:02 (UTC)

oh. try to add 'ccache' to a makedepends()

daurnimator commented on 2015-04-05 03:50 (UTC)

I get same error: -> Starting building Chromium... Updating projects from gyp files... Traceback (most recent call last): File "../../tools/compile_test/compile_test.py", line 65, in <module> sys.exit(DoMain(sys.argv[1:])) File "../../tools/compile_test/compile_test.py", line 52, in DoMain stderr=subprocess.PIPE) File "/usr/lib/python2.7/subprocess.py", line 710, in __init__ errread, errwrite) File "/usr/lib/python2.7/subprocess.py", line 1335, in _execute_child raise child_exception OSError: [Errno 2] No such file or directory gyp: Call to 'python ../../tools/compile_test/compile_test.py --code "int main() { return 0; }" --run-linker --on-success "-lwebp -lwebpdemux" --on-failure "-lwebp" -- -lwebpdemux' returned exit status 1. ==> ERROR: A failure occurred in build(). Aborting... ==> ERROR: Makepkg was unable to build .

illuser commented on 2015-04-05 01:31 (UTC)

Getting the same error, even with python2-lxml added to dependencies.

sl1pkn07 commented on 2015-04-04 11:11 (UTC)

@gokcen @rekoil add 'python2-lxml' to dependencies greetings

gokcen commented on 2015-04-04 10:21 (UTC)

Getting the same error here. FWIW, _use_system_clang=1 does not fix the error.

rekoil commented on 2015-04-03 18:07 (UTC)

I'm having the following issue when attempting to compile the latest chromium-dev. Not sure exactly what is going on, perhaps there's some dependency missing? Can't really test because it keeps starting from scratch every time I run makepkg. -> Starting building Chromium... Updating projects from gyp files... Traceback (most recent call last): File "../../tools/compile_test/compile_test.py", line 65, in <module> sys.exit(DoMain(sys.argv[1:])) File "../../tools/compile_test/compile_test.py", line 52, in DoMain stderr=subprocess.PIPE) File "/usr/lib/python2.7/subprocess.py", line 710, in __init__ errread, errwrite) File "/usr/lib/python2.7/subprocess.py", line 1335, in _execute_child raise child_exception OSError: [Errno 2] No such file or directory gyp: Call to 'python ../../tools/compile_test/compile_test.py --code "int main() { return 0; }" --run-linker --on-success "-lwebp -lwebpdemux" --on-failure "-lwebp" -- -lwebpdemux' returned exit status 1. ==> ERROR: A failure occurred in build(). Aborting...

gokcen commented on 2015-03-21 07:38 (UTC)

I hope hidpi support will get better with the latest fix released: https://code.google.com/p/chromium/issues/detail?id=143619#c71

sl1pkn07 commented on 2015-03-15 22:55 (UTC)

@jebarb please fill bugreport in http://crbug.com greetings

jebarb commented on 2015-03-15 22:29 (UTC)

@sl1pkn07 Package finally built with hidpi flag enabled. Everything works, but the menus are nearly unusable because the text size is about 3x everything else. Right click has the same issue, and the boxes for both are popping up in the wrong place. The menu pops up in the middle of the screen, and the right click menu is all over the place.

jebarb commented on 2015-03-15 19:57 (UTC)

@sl1pkn07 Looks like I was somehow running out of memory, causing segfaults. Nothing useful in any logs. Computer was becoming 99% unresponsive. Strange, because it was failing on a system with 12G memory with 6G tmpfs. Bumped it up to 8G and added a swapfile, now the build is moving along fine.

sl1pkn07 commented on 2015-03-14 08:13 (UTC)

@jbarb For first error. Please paste more log (search 'FAILED' string). For second error, i'll investigare. One cuestión. You use 32 or 64 bits? Greetings

jebarb commented on 2015-03-13 23:55 (UTC)

Package is failing to build with packaged clang, system clang, and gcc on two devices. Error with packaged clang: [1837/2229] Building CXX object tools/clang/lib/ASTMatchers/Dynamic/CMakeFiles/clangDynamicASTMatchers.dir/Registry.cpp.o FAILED: /usr/bin/c++ -DCLANG_ENABLE_ARCMT -DCLANG_ENABLE_OBJC_REWRITER -DCLANG_ENABLE_STATIC_ANALYZER -DGTEST_HAS_RTTI=0 -D_DEBUG -D_GNU_SOURCE -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -D__STDC_LIMIT_MACROS -fPIC -fvisibility-inlines-hidden -Wall -W -Wno-unused-parameter -Wwrite-strings -Wcast-qual -Wno-missing-field-initializers -pedantic -Wno-long-long -Wno-maybe-uninitialized -Wno-comment -std=c++11 -ffunction-sections -fdata-sections -fno-common -Woverloaded-virtual -fno-strict-aliasing -O3 -Itools/clang/lib/ASTMatchers/Dynamic -I/tmp/pacaurtmp-jebarb/chromium-dev/src/chromium-42.0.2311.22/third_party/llvm/tools/clang/lib/ASTMatchers/Dynamic -I/tmp/pacaurtmp-jebarb/chromium-dev/src/chromium-42.0.2311.22/third_party/llvm/tools/clang/include -Itools/clang/include -Iinclude -I/tmp/pacaurtmp-jebarb/chromium-dev/src/chromium-42.0.2311.22/third_party/llvm/include -UNDEBUG -fno-exceptions -fno-rtti -MMD -MT tools/clang/lib/ASTMatchers/Dynamic/CMakeFiles/clangDynamicASTMatchers.dir/Registry.cpp.o -MF tools/clang/lib/ASTMatchers/Dynamic/CMakeFiles/clangDynamicASTMatchers.dir/Registry.cpp.o.d -o tools/clang/lib/ASTMatchers/Dynamic/CMakeFiles/clangDynamicASTMatchers.dir/Registry.cpp.o -c /tmp/pacaurtmp-jebarb/chromium-dev/src/chromium-42.0.2311.22/third_party/llvm/tools/clang/lib/ASTMatchers/Dynamic/Registry.cpp c++: internal compiler error: Killed (program cc1plus) Please submit a full bug report, with preprocessed source if appropriate. See <https://bugs.archlinux.org/> for instructions. [1837/2229] Building CXX object tools/clang/lib/Sema/CMakeFiles/clangSema.dir/SemaDeclCXX.cpp.o ninja: build stopped: subcommand failed. ==> ERROR: A failure occurred in build(). Error with system clang (gcc error is similar): -> Starting building Chromium... File "build/gyp_chromium", line 118 except SyntaxError, e: ^ SyntaxError: invalid syntax ==> ERROR: A failure occurred in build().

ethail commented on 2015-03-09 22:54 (UTC)

Would it be too much trouble to add some kind of config option that results on using ccache when the system clang is enabled? It would have to change lines 364 onwards for something like this: msg2 "Setup system clang" export CC="ccache clang -Qunused-arguments" export CXX="ccache clang++ -Qunused-arguments" export CCACHE_CPP2=yes export CCACHE_SLOPPINESS=time_macros export CXXFLAGS="${CXXFLAGS} -Wno-unknown-warning-option" For those that have set-up ccache, the change could result in a speed-up of subsequent rebuilds or updates. I know anyone could make that changes manually every time or something like that, but having it as a config option in the PKGBUILD seemed a better way to handle it. Thanks in advance

sl1pkn07 commented on 2015-02-27 10:38 (UTC)

Ops! I whill fix when go to home Greerings

ledti commented on 2015-02-27 04:51 (UTC)

I noticed that the chromium-dev wrapper looks for /etc/chromium/default rather than /etc/chromium-dev/default, which makes chromium-dev incompatible with all the pepper-flash packages unless you also have chromium installed (or you edit the wrapper). Is that intentional?

sl1pkn07 commented on 2015-02-25 05:42 (UTC)

ouch!

misc commented on 2015-02-25 01:15 (UTC)

Just change the COMPLETE to SUCCESS in the hunk. Not sure if there weren't any other changes between 2311 and 2314, though.

sl1pkn07 commented on 2015-02-24 21:42 (UTC)

Thanks @misc but your patch, but don't work in 42.0.2311.4 (with some changes like remove this http://sl1pkn07.wtf/paste/view/670f462a) Hunk #11 FAILED at 465. Hunk #12 succeeded at 477 (offset -6 lines). 1 out of 12 hunks FAILED -- saving rejects to file ui/file_manager/file_manager/background/js/media_import_handler.js.rej http://sl1pkn07.wtf/paste/view/29ac699d and please don't use pastebin, that breaks the EOL greetings

misc commented on 2015-02-24 16:43 (UTC)

For those who don't want the Google Analytics crap built into their Chromium: http://pastebin.com/raw.php?i=ZUJS8ecT

sl1pkn07 commented on 2015-02-20 15:36 (UTC)

how you update chromium dev? (which helper i mean) because if '_use_widevine' is set to '1', then set makedepend directly on 'chromium-widevine-dev' see line 90: # Use Widevide? [ "${_use_widevine}" = "1" ] && makedepends+=('chromium-widevine-dev') greetings

drizz commented on 2015-02-20 15:03 (UTC)

When building: ==> Starting prepare()... -> Fix to save configuration in ~/.config/chromium-dev -> Add Widevine Support /tmp/pacaurtmp-/chromium-dev/PKGBUILD: line 322: /usr/lib/chromium-dev/widevine.version: No such file or directory ==> ERROR: A failure occurred in prepare(). Aborting...

sl1pkn07 commented on 2015-02-19 20:15 (UTC)

do'h, jesuchrist now really fixed ><

sl1pkn07 commented on 2015-02-19 05:27 (UTC)

OW men!! http://upload.wikimedia.org/wikipedia/commons/f/f4/The_Scream.jpg yes. the pkgbuild have a typo in prepare() (widevide instead of widevine) fixed very sorry :S

juston_li commented on 2015-02-19 00:54 (UTC)

After it finishes compiling I get: install: cannot stat ‘libwidevinecdmadapter.so’: No such file or directory

sl1pkn07 commented on 2015-02-18 21:11 (UTC)

- now support Google Widevine plugin (active by default. to disable set '_use_widevine' to 0) - now can handle more configuration files in /etc/chromium-dev. useful for plugins. for use their own conf files instead edit the general conf file. see https://bugs.archlinux.org/task/43804 - libpdf now integrated into a binary greetings

sl1pkn07 commented on 2015-02-10 14:55 (UTC)

32 or 64 bits?

shackra commented on 2015-02-10 07:10 (UTC)

FAILED with 1: ../../../../out/Release/gen/sdk/linux_x86/pnacl_newlib/bin/pnacl-clang -c posix/nacl_timestamp.c -o ../../../../out/Release/obj/native _client/src/shared/platform/platform_lib.gen/pnacl_newlib-pnacl/platform_lib/nacl_timestamp_362f7b12.o -MD -MF ../../../../out/Release/obj/native_cli ent/src/shared/platform/platform_lib.gen/pnacl_newlib-pnacl/platform_lib/nacl_timestamp_362f7b12.o.d -O2 -g -Wall -fdiagnostics-show-option -gline-ta bles-only -Wno-unused-function -Wno-char-subscripts -Wno-c++11-extensions -Wno-unnamed-type-template-args -Wno-extra-semi -Wno-unused-private-field - Wno-char-subscripts -Wno-unused-function -D__STDC_LIMIT_MACROS=1 -D__STDC_FORMAT_MACROS=1 -D_GNU_SOURCE=1 -D_DEFAULT_SOURCE=1 -D_BSD_SOURCE=1 -D_POSI X_C_SOURCE=199506 -D_XOPEN_SOURCE=600 -DDYNAMIC_ANNOTATIONS_ENABLED=1 -DDYNAMIC_ANNOTATIONS_PREFIX=NACL_ -DV8_DEPRECATION_WARNINGS -D_FILE_OFFSET_BIT S=64 -D_DEFAULT_SOURCE=1 -D_BSD_SOURCE=1 -D_POSIX_C_SOURCE=199506 -D_XOPEN_SOURCE=600 -D_GNU_SOURCE=1 -D__STDC_LIMIT_MACROS=1 -DCHROMIUM_BUILD -DCR_C LANG_REVISION=223108 -DTOOLKIT_VIEWS=1 -DUI_COMPOSITOR_IMAGE_TRANSPORT -DUSE_AURA=1 -DUSE_ASH=1 -DUSE_PANGO=1 -DUSE_CAIRO=1 -DUSE_DEFAULT_RENDER_THEM E=1 -DUSE_LIBJPEG_TURBO=1 -DUSE_X11=1 -DUSE_CLIPBOARD_AURAX11=1 -DENABLE_ONE_CLICK_SIGNIN -DENABLE_PRE_SYNC_BACKUP -DENABLE_REMOTING=1 -DENABLE_WEBRT C=1 -DUSE_PROPRIETARY_CODECS -DENABLE_PEPPER_CDMS -DENABLE_CONFIGURATION_POLICY -DENABLE_NOTIFICATIONS -DUSE_UDEV -DDONT_EMBED_BUILD_METADATA -DENABL E_TASK_MANAGER=1 -DENABLE_EXTENSIONS=1 -DENABLE_PLUGINS=1 -DENABLE_SESSION_SERVICE=1 -DENABLE_THEMES=1 -DENABLE_AUTOFILL_DIALOG=1 -DENABLE_BACKGROUND =1 -DENABLE_GOOGLE_NOW=1 -DCLD_VERSION=2 -DENABLE_PRINTING=1 -DENABLE_BASIC_PRINTING=1 -DENABLE_PRINT_PREVIEW=1 -DENABLE_SPELLCHECK=1 -DENABLE_CAPTIV E_PORTAL_DETECTION=1 -DENABLE_APP_LIST=1 -DENABLE_SETTINGS_APP=1 -DENABLE_SUPERVISED_USERS=1 -DENABLE_MDNS=1 -DENABLE_SERVICE_DISCOVERY=1 -DV8_USE_EX TERNAL_STARTUP_DATA -DUSE_LIBPCI=1 -DUSE_GLIB=1 -DUSE_NSS=1 -DNACL_WINDOWS=0 -DNACL_OSX=0 -DNACL_LINUX=0 -DNACL_ANDROID=0 -DNACL_BUILD_ARCH=pnacl -I. ./../../../out/Release/gen/tc_pnacl_newlib/include -I../../../.. -I../../../../out/Release/gen -I../../../src/third_party -I../../../.. -DNDEBUG -std =gnu99 Unknown clang plugin argument: check-weak-ptr-factory-order ninja: build stopped: subcommand failed.

sl1pkn07 commented on 2015-02-06 21:39 (UTC)

no problem ;) thanks for you I found various errors in the pkgbuild ;) greetings

Lynthium commented on 2015-02-06 20:38 (UTC)

@sl1pkn07 Ye, fully up to date 64bit. The build was with the bundled clang. I couldn't find the "virdual" in the file the error mentioned though. Guess it's just a gremlin poking its head up in my system (or in this insanely complicated build system). Building it with system clang right now. It's past 14.5k/17.2k mark so it should be fine. My apologies for wasting your time with this :(

sl1pkn07 commented on 2015-02-06 20:31 (UTC)

build ok with GCC

sl1pkn07 commented on 2015-02-06 19:57 (UTC)

@Lynden build ok with system clang. trying with GCC....

sl1pkn07 commented on 2015-02-06 19:37 (UTC)

@chrisdunder with your option spawn a lot of warning: unknown warning option '-Wno-inconsistent-missing-override' [-Wunknown-warning-option] XD

sl1pkn07 commented on 2015-02-06 19:19 (UTC)

update pkgbuild to fix build bundled ffmpeg with system clang

sl1pkn07 commented on 2015-02-06 18:58 (UTC)

you system is up to date? 32 or 64 bits? build fine here (with bundled clang). build with system clang in progress.. greetings

Lynthium commented on 2015-02-06 16:27 (UTC)

It fails at 10553 / 17202 with In file included from ../../third_party/WebKit/Source/core/fetch/ResourcePtr.h:29: ../../third_party/WebKit/Source/core/fetch/Resource.h:191:5: error: unknown type name 'virdual' virdual void willFollowRedirect(ResourceRequest&, const ResourceResponse&); ^ ../../third_party/WebKit/Source/core/fetch/Resource.h:191:13: error: expected member name or ';' after declaration specifiers virdual void willFollowRedirect(ResourceRequest&, const ResourceResponse&); ~~~~~~~ ^ 2 errors generated.

Lynthium commented on 2015-02-06 15:49 (UTC)

@sl1pkn07 Thanks, it builds fine now :)

sl1pkn07 commented on 2015-02-06 15:15 (UTC)

@Lynden try now @misc thanks for the info! @chrisdunder added cmake to makedepends if use bundled clang (fixed typo (c&p code xd) and added -Wno-inconsistent-missing-override if use system clang python path ('export PATH') still in prepare() greetings

Lynthium commented on 2015-02-06 14:58 (UTC)

Addition: the previous comment fixes the ffmpeg build, but the chromium build shoots itself in the foot still later on. -> Starting building Chromium... Updating projects from gyp files... ninja: Entering directory `out/Release' ninja: warning: multiple rules generate obj/third_party/zlib/google/zip.zip.o. builds involving this target will not be correct; continuing anyway ninja: warning: multiple rules generate obj/third_party/zlib/google/zip.zip_internal.o. builds involving this target will not be correct; continuing anyway ninja: warning: multiple rules generate obj/third_party/zlib/google/zip.zip_reader.o. builds involving this target will not be correct; continuing anyway ninja: error: '/data/chromium-dev/src/flapper_version.h', needed by 'gen/flapper_version.h', missing and no known rule to make it "/data/chromium-dev/src" doesn't exist, it should be "/tmp/makepkg/chromium-dev/src".

Lynthium commented on 2015-02-06 14:50 (UTC)

"$(pwd)/src" in _build_ffmpeg_args (line 84) breaks the ffmpeg build if you have a custom BUILDDIR set in makepkg.conf, since the src is no longer created in the current pwd. Changing it to "${srcdir}" and moving said statement into prepare() works, but it's prob not the cleanest solution.

misc commented on 2015-02-06 09:41 (UTC)

Just a heads-up, libpdf will be integrated directly into the chromium binary in the next release, leading to space savings of some 7 MiB. That and they just enabled multithreading for the gold linker, which consistently crashes on my machine. Doubt that's gonna make it into the release, though: https://codereview.chromium.org/894463003

sl1pkn07 commented on 2015-02-06 04:20 (UTC)

@chrisdunder thanks for the suggestions. one question: why need move 'export PATH' to to build()? greetings

chrisdunder commented on 2015-02-05 23:01 (UTC)

Can you move the `export PATH` line from prepare() to build()? Also cmake is a builddep if use_system_clang==0. Also can you add -Wno-unknown-warning-option to CXXFLAGS if use_system_clang==1? Otherwise it spams "warning: unknown warning option '-Wno-inconsistent-missing-override'" with the latest (system) clang. From one to another, Thanks for maintaining this!

TNC commented on 2015-01-28 12:18 (UTC)

Worked like a charm, thank you! greetings

TNC commented on 2015-01-27 15:25 (UTC)

I'll try that and keep you updated. greetings

sl1pkn07 commented on 2015-01-27 15:16 (UTC)

Update (no need rebuild) pkgbuild to add support to use system instead of bundled clang @TNC try to set "'use_system_clang=' to '1'. or if you want use GCC, set '_use_clang' to '0' greetings

TNC commented on 2015-01-27 09:55 (UTC)

the newest version failed to build twice: /backward -internal-isystem /usr/local/include -internal-isystem /usr/bin/../lib/clang/3.5.1/include -internal-externc-isystem /include -internal-externc-isystem /usr/include -O3 -Wall -Wno-unused-parameter -Wno-missing-field-initializers -Wno-reserved-user-defined-literal -Wheader-hygiene -Wno-char-subscripts -Wno-unneeded-internal-declaration -Wno-covered-switch-default -Wstring-conversion -Wno-c++11-narrowing -Wno-deprecated-register -Wno-inconsistent-missing-override -Wglobal-constructors -Wsign-compare -std=gnu++11 -fdeprecated-macro -fdebug-compilation-dir /tmp/yaourt-tmp-mpd/aur-chromium-dev/src/chromium-41.0.2272.12/out/Release -ferror-limit 19 -fmessage-length 0 -fvisibility hidden -fvisibility-inlines-hidden -pthread -stack-protector 2 -stack-protector-buffer-size 4 -stack-protector-buffer-size 4 -mstackrealign -fno-rtti -fno-threadsafe-statics -fobjc-runtime=gcc -fdiagnostics-show-option -fcolor-diagnostics -vectorize-loops -vectorize-slp -load /tmp/yaourt-tmp-mpd/aur-chromium-dev/src/chromium-41.0.2272.12/third_party/llvm-build/Release+Asserts/lib/libFindBadConstructs.so -add-plugin find-bad-constructs -plugin-arg-find-bad-constructs strict-virtual-specifiers -load /tmp/yaourt-tmp-mpd/aur-chromium-dev/src/chromium-41.0.2272.12/third_party/llvm-build/Release+Asserts/lib/libBlinkGCPlugin_14.so -add-plugin blink-gc-plugin -o obj/third_party/WebKit/Source/wtf/wtf.DataLog.o -x c++ ../../third_party/WebKit/Source/wtf/DataLog.cpp clang: error: unable to execute command: Segmentation fault (core dumped) clang: error: clang frontend command failed due to signal (use -v to see invocation) clang version 3.5.1 (tags/RELEASE_351/final) Target: x86_64-unknown-linux-gnu Thread model: posix clang: note: diagnostic msg: PLEASE submit a bug report to http://llvm.org/bugs/ and include the crash backtrace, preprocessed source, and associated run script. clang: note: diagnostic msg: ******************** PLEASE ATTACH THE FOLLOWING FILES TO THE BUG REPORT: Preprocessed source(s) and associated run script(s) are located at: clang: note: diagnostic msg: /tmp/DataLog-8e4e9f.cpp clang: note: diagnostic msg: /tmp/DataLog-8e4e9f.sh clang: note: diagnostic msg: ******************** warning: unknown warning option '-Wno-inconsistent-missing-override' [-Wunknown-warning-option] 1 warning generated. warning: unknown warning option '-Wno-inconsistent-missing-override' [-Wunknown-warning-option] 1 warning generated. ninja: build stopped: subcommand failed. ==> FEHLER: Ein Fehler geschah in build(). Breche ab... I can provide the 2 mentioned files if needed. I tried to built the package with clang.

sl1pkn07 commented on 2015-01-25 21:11 (UTC)

update for add patch like a [extra] chromium. now fix open pages like https://en.wikipedia.org/wiki/List_of_Unicode_characters greetings and sorry the inconvenience

sl1pkn07 commented on 2015-01-22 14:48 (UTC)

update to 41.0.2272.12 now uses bundled clang greetings

commented on 2015-01-17 04:51 (UTC)

Builds fails for me with following error message: [3581/17051] cd ../../third_party/WebKit/Source/core; python ../build/scripts/make_token_matcher.py ../core/css/parser/BisonCSSParser-in.cpp ../../../../out/Release/gen/blink/core/BisonCSSParser.cpp FAILED: cd ../../third_party/WebKit/Source/core; python ../build/scripts/rule_bison.py "css/parser/CSSGrammar.y" ../../../../out/Release/gen/blink/core bison Traceback (most recent call last): File "../build/scripts/rule_bison.py", line 82, in <module> returnCode = subprocess.call([bisonExe, '-d', '-p', prefix, inputFile, '-o', outputCpp]) File "/usr/lib/python2.7/subprocess.py", line 522, in call return Popen(*popenargs, **kwargs).wait() File "/usr/lib/python2.7/subprocess.py", line 710, in __init__ errread, errwrite) File "/usr/lib/python2.7/subprocess.py", line 1335, in _execute_child raise child_exception OSError: [Errno 2] No such file or directory ninja: build stopped: subcommand failed. ==> ERROR: A failure occurred in build(). Aborting... Tried twice, same output.

commented on 2015-01-17 04:48 (UTC)

Build fails for me with following error message: [3581/17051] cd ../../third_party/WebKit/Source/core; python ../build/scripts/make_token_matcher.py ../core/css/parser/BisonCSSParser-in.cpp ../../../../out/Release/gen/blink/core/BisonCSSParser.cpp FAILED: cd ../../third_party/WebKit/Source/core; python ../build/scripts/rule_bison.py "css/parser/CSSGrammar.y" ../../../../out/Release/gen/blink/core bison Traceback (most recent call last): File "../build/scripts/rule_bison.py", line 82, in <module> returnCode = subprocess.call([bisonExe, '-d', '-p', prefix, inputFile, '-o', outputCpp]) File "/usr/lib/python2.7/subprocess.py", line 522, in call return Popen(*popenargs, **kwargs).wait() File "/usr/lib/python2.7/subprocess.py", line 710, in __init__ errread, errwrite) File "/usr/lib/python2.7/subprocess.py", line 1335, in _execute_child raise child_exception OSError: [Errno 2] No such file or directory ninja: build stopped: subcommand failed. ==> ERROR: A failure occurred in build(). Aborting...

commented on 2015-01-17 04:47 (UTC)

Build fails for me with following error message: atcher.py ../core/css/parser/BisonCSSParser-in.cpp ../../../../out/Release/gen/blink/core/BisonCSSParser.cpp FAILED: cd ../../third_party/WebKit/Source/core; python ../build/scripts/rule_bison.py "css/parser/CSSGrammar.y" ../../../../out/Release/gen/blink/core bison Traceback (most recent call last): File "../build/scripts/rule_bison.py", line 82, in <module> returnCode = subprocess.call([bisonExe, '-d', '-p', prefix, inputFile, '-o', outputCpp]) File "/usr/lib/python2.7/subprocess.py", line 522, in call return Popen(*popenargs, **kwargs).wait() File "/usr/lib/python2.7/subprocess.py", line 710, in __init__ errread, errwrite) File "/usr/lib/python2.7/subprocess.py", line 1335, in _execute_child raise child_exception OSError: [Errno 2] No such file or directory ninja: build stopped: subcommand failed. ==> ERROR: A failure occurred in build(). Aborting...

sl1pkn07 commented on 2014-12-31 17:46 (UTC)

not free space. to build chromium-dev need 2Gb or more

tajidinabd commented on 2014-12-31 17:36 (UTC)

When trying to build this package in the prepare field all kinds of errors complaining about write failed as you can see. Is there a permissions problem? chromium-41.0.2251.0/rlz/lib/rlz_lib.h: Write failed chromium-41.0.2251.0/rlz/lib/lib_values.h: Write failed chromium-41.0.2251.0/rlz/lib/financial_ping.h: Write failed chromium-41.0.2251.0/rlz/lib/machine_id_unittest.cc: Write failed chromium-41.0.2251.0/rlz/lib/rlz_lib.cc: Write failed chromium-41.0.2251.0/rlz/test/rlz_test_helpers.cc: Write failed chromium-41.0.2251.0/rlz/test/rlz_unittest_main.cc: Write failed chromium-41.0.2251.0/rlz/test/rlz_test_helpers.h: Write failed bsdtar: Error exit delayed from previous errors. ==> FEHLER: Ein Fehler geschah in prepare(). Excuse me for the german basically it is saying it failed in the prepare(). any idea the situation

TNC commented on 2014-12-18 10:21 (UTC)

For some reason the build was fine, but I didn't install or update any packages... greetings

TNC commented on 2014-12-16 21:53 (UTC)

As I mentioned, I forgot to write the missing files down. I will recompile the package with gcc, so you have to be patient. :-) Greetings

sl1pkn07 commented on 2014-12-16 10:32 (UTC)

Can you tell me what files is missing? Greetings

TNC commented on 2014-12-16 09:13 (UTC)

Build did fail with gcc on my amd c60, but with clang it went through. With gcc the build failed because of some missing python files. I can't tell you which, because I forgot to write it down. A failed build is really infuriating, because on this slow machine, the build takes over 12h. :-) greetings

sl1pkn07 commented on 2014-11-21 01:40 (UTC)

update to 41 branch - switch back to GCC (with clang fail build) greeetings

sparse commented on 2014-11-20 14:06 (UTC)

Is there any repo for this package?

sl1pkn07 commented on 2014-10-29 12:09 (UTC)

git download over 3Gb in first time. and need a lot of work to "mount" all sources to build same like a tar.xz sources

Nefelim4ag commented on 2014-10-29 11:15 (UTC)

Sorry but, may be you can create a git version of this package? (Downloading of 250 Mb not good, and i use DEVELSRCDIR="/var/cache/src-csv-cache" in yaourt for cache VCS, and i think, not only i, its useful)

sl1pkn07 commented on 2014-09-29 06:29 (UTC)

@skelet nope, both is from [core] (base & base-devel group)

skelet commented on 2014-09-29 06:03 (UTC)

It seems it needs some additional dependencies: * patch * diffutils (for cmp command)

sl1pkn07 commented on 2014-09-26 01:37 (UTC)

please insert this after line 105 pwd $srcdir ls $srcdir exit try now and paste here the output

sl1pkn07 commented on 2014-09-26 01:30 (UTC)

@roukoswarf what version of pacman you use? git?

roukoswarf commented on 2014-09-26 01:28 (UTC)

Package fails prepare on latest versions, path needs ../ removed on decompress line, as it will fail to find the archive.

abandonedaccount commented on 2014-09-21 17:17 (UTC)

You're right. I understand it now: remove_bundled_libraries.py <preserved-directories> Thank you.

sl1pkn07 commented on 2014-09-20 19:10 (UTC)

that array in "Remove unnecessary components to save space" is for exceptions (only apply in all thrid_party folders)

sl1pkn07 commented on 2014-09-20 19:07 (UTC)

thats a test I did. nothing happens fixed greetings

abandonedaccount commented on 2014-09-20 19:05 (UTC)

chromium-system-jinja-r7.patch seems to be patching only files in third_party/WebKit folder which is later removed via msg2 "Remove unnecessary components to save space"

abandonedaccount commented on 2014-09-20 18:59 (UTC)

in this line: find . -not \( -path ./git -prune \) -type f -exec sed -i -r \ ./git should be ./.git or else sed will dive into it - nobody was affected by this unless they manually git checkout the repo instead of the archive

sl1pkn07 commented on 2014-09-18 05:22 (UTC)

fixed. sorry

elric_ed commented on 2014-09-17 18:43 (UTC)

Got the following error while running makepkg ==> Validating source files with sha1sums... chromium-39.0.2159.4.tar.xz ... Passed chromium-dev.desktop ... Passed chromium-dev.sh ... Passed chromium-dev.default ... Passed chromium-dev.xml ... Passed chromium-system-jinja-r7.patch ... FAILED ==> ERROR: One or more files did not pass the validity check! updated sha1sum for chromium-system-jinja-r7.patch on line 54 of PKGBUILD with c24d14029714d2295f3220a7173a5a7362f578a2 It is building now. Will report back if it fails!

juston_li commented on 2014-09-17 18:29 (UTC)

chromium-system-jinja-r7.patch ... FAILED ==> ERROR: One or more files did not pass the validity check!

abandonedaccount commented on 2014-09-07 19:47 (UTC)

that pacman-git bug is tracked here: https://bugs.archlinux.org/task/41862

abandonedaccount commented on 2014-09-07 17:28 (UTC)

I'll try to find out why it isn't softlinked for me(likely pacman-git), I'll let you know what I find. I've tested so far with untouched makepkg.conf and pacman.conf but the error persists, pwd is: /tmp/yaourt-tmp-emacs/aur-chromium-dev/src and all other files are softlinked. Btw, thank you for maintaining this package! Much appreciated!

sl1pkn07 commented on 2014-09-07 17:20 (UTC)

maybe is a pacman-git problem all my test is with pacman 4.1.2-6 and yaourt 1.5-1

sl1pkn07 commented on 2014-09-07 17:15 (UTC)

is softlinked. add that option versions ago. always work the softlink is added if the files exist in sources array

abandonedaccount commented on 2014-09-07 17:01 (UTC)

I think the reason the tar.gz isn't softlinked is because of this line: noextract=("chromium-${pkgver}.tar.xz") haven't tested yet

sl1pkn07 commented on 2014-09-07 16:51 (UTC)

from yaourt (same test) http://sl1pkn07.wtf/paste/view/9a74f85f

sl1pkn07 commented on 2014-09-07 16:44 (UTC)

makepkg and yaourt softlink the sources from SRCDEST to srcdir http://sl1pkn07.wtf/paste/view/9dfbc763 tested with edit the pkgbuild and add pwd ls -la exit in lines 102, 103 and 104 (http://wstaw.org/m/2014/09/07/plasma-desktopil2384.png) all with untouch /etc/makepkg.conf

abandonedaccount commented on 2014-09-07 16:32 (UTC)

I was afraid that was the case which means it may be my (newer)pacman version: Pacman v4.1.2-406-g3e19-dirty - libalpm v8.0.2 putting an echo `pwd` before the bsdtar command shows current folder to be: /tmp/makepkg/chromium-dev/src while the .tar.gz resides in /tmp/sources/ I've the following /etc/makepkg.conf settings: BUILDDIR=/tmp/makepkg PKGDEST=/tmp/packages SRCDEST=/tmp/sources SRCPKGDEST=/tmp/srcpackages testing with SRCDEST unset in /etc/makepkg.conf leaves the downloaded tar.xz file in /tmp/yaourt-tmp-emacs/aur-chromium-dev and still fails. Rebuilding pacman yields the same version. makepkg (pacman) 4.1.2-406-g3e19-dirty (August 7 commit was last) was using yaourt 1.5 but now(next) using yaourt-git: yaourt 1.4-1-gb6b8 nothing changed. bsdtar: Error opening archive: Failed to open 'chromium-39.0.2145.4.tar.xz' pwd is: /tmp/makepkg/chromium-dev/src archive is actually in /tmp/sources/ or when SRCDEST above isn't set(and running yaourt in a new terminal) then archive is in: /tmp/yaourt-tmp-emacs/aur-chromium-dev/ where PKGBUILD &co. are also downloaded

sl1pkn07 commented on 2014-09-07 15:14 (UTC)

working for me (makepkg(local) and yaourt(online))

abandonedaccount commented on 2014-09-07 14:49 (UTC)

had to replace this: bsdtar -xf "chromium-${pkgver}.tar.xz" with this: bsdtar -xf "${SRCDEST}/chromium-${pkgver}.tar.xz" or else it would fail to find that file and fail.

sl1pkn07 commented on 2014-08-24 22:24 (UTC)

im not sure. but libwidevine is include directly in chrome binary └───╼ strings chrome |grep widevine libwidevinecdmadapter.so application/x-ppapi-widevine-cdm

jeagoss commented on 2014-08-24 21:25 (UTC)

No. I would like to copy the libwidevinecdm*.so components from the binary build to the opensource build and have them register properly. Kind of like how the flash plugin is copied currently. I've been doing minimal searching and not turning up much on how to enable recognition for the components.

sl1pkn07 commented on 2014-08-24 21:23 (UTC)

@jeagos you mean this? http://www.chromium.org/administrators/pre-installed-extensions (docs.crx, drive.crx youtube.crx, etc..)

sl1pkn07 commented on 2014-08-23 21:02 (UTC)

Nope. Sorry. But i can have a look when back from my little flash holiday Regards

jeagoss commented on 2014-08-23 20:29 (UTC)

Would you happen to know if there is a way to register the widevine components from the binary build with the open source build?

sl1pkn07 commented on 2014-08-22 03:24 (UTC)

38.0.2125.8 release today.... bah!!

sl1pkn07 commented on 2014-08-22 03:04 (UTC)

If this package build/compact in one minute i don't make this effusive apology. but this take an a hour and more.. I can not allow such a silly error like this. I do not want waste the time to other users. greetings :)

jeagoss commented on 2014-08-22 02:08 (UTC)

Heh. Everyone forgets stuff sometimes. No need to apologize. Thanks for fixing it :)

sl1pkn07 commented on 2014-08-21 19:51 (UTC)

fixed in pkgbuild i'm so sorry :S NO NEED REBUILD ALL. only just copy the file with the right name and before next update remove that file sorry the inconvenience

sl1pkn07 commented on 2014-08-21 19:41 (UTC)

BIG DO'H! my faut! please rename /usr/lib/chromium-dev/chrome_100_percent to /usr/lib/chromium-dev/chrome_100_percent.pak BIG SORRY :S

jeagoss commented on 2014-08-21 19:08 (UTC)

This last build has resulted in an incredibly unusable package. When loading chromium, Almost all UI elements are Red. Plugins are very crashy. I've done a bit of searching and nothing seems to be being reported. Is this a known issue? Has anyone else run into it?

sl1pkn07 commented on 2014-06-19 13:30 (UTC)

@matchulinux try now (disable ffmpeg patch)

matchulinux commented on 2014-06-19 12:45 (UTC)

It seems like the sha1sum for the file chromium-ffmpeg-r1.patch is wrong. ==> Validating source files with sha1sums... [...] chromium-ffmpeg-r1.patch ... FAILED [...] ==> ERROR: One or more files did not pass the validity check! PS: Thanks for maintaining this package :)

sl1pkn07 commented on 2014-06-06 09:47 (UTC)

@lordmocha that error is problem from google side. and for re2 related... i think is because makepkg (makepkg -s) cannot handle packages from AUR directly greetings

lordmocha commented on 2014-06-06 04:03 (UTC)

IOError: Could not download URL (https://storage.googleapis.com/nativeclient-once/object/llvm_i686_linux_3c0f38277820277b0ede45b2b22fb54fcd9ded5c.tgz): The read operation timed out ==> ERROR: A failure occurred in prepare(). Aborting... and i also had to install re2 from aur first, it didn't autoget it as a dependency

sl1pkn07 commented on 2014-06-05 22:27 (UTC)

try add -Drelease_extra_cflags='march=native' into _flags= array greetings

sudsbud commented on 2014-06-05 21:57 (UTC)

I was wondering how I might specify some more compile options - for instance I want to specify march=native. I tried exporting C and CXX FLAGS and editing the PKGBUILD along with editing my makepkg.conf but it didn't seem to work. How should I specify flags for this build?

sl1pkn07 commented on 2014-06-05 10:09 (UTC)

"warning" is a internal command from makepkg, like "msg" and "msg2" nothing to do

pfrenssen commented on 2014-06-05 08:26 (UTC)

I'm getting the following error: /tmp/pacaurtmp-pieter/chromium-dev/PKGBUILD: line 28: warning: command not found It tries to execute the command called "warning" which is not present on my system. You can either list the package which supplies "warning" as a dependency, or just use echo instead of warning.

sl1pkn07 commented on 2014-05-30 21:27 (UTC)

my bad. I upload new pkgbuild with disable unittest please, you can safe rebuild with "makepkg -fR" (only compress package, no rebuild everything) greetings and sorry :S

DarkMind commented on 2014-05-30 19:20 (UTC)

[1444/1444] ThreadCollisionTest.MTSynchedScopedRecursiveBookCriticalSectionTest (4000 ms) Retrying 4 tests (retry #1) [1445/1448] TraceEventTestFixture.TraceContinuousSampling (0 ms) [ RUN ] FileUtilTest.ChangeFilePermissionsAndWrite ../../base/file_util_unittest.cc:772: Failure Value of: WriteFile(file_name, kData.data(), kData.length()) Actual: 5 Expected: -1 ../../base/file_util_unittest.cc:773: Failure Value of: PathIsWritable(file_name) Actual: true Expected: false [ FAILED ] FileUtilTest.ChangeFilePermissionsAndWrite (0 ms) [1446/1448] FileUtilTest.ChangeFilePermissionsAndWrite (0 ms) [ RUN ] FileUtilTest.ChangeFilePermissionsAndRead ../../base/file_util_unittest.cc:732: Failure Value of: ReadFile(file_name, buffer, buffer_size) Actual: 5 Expected: -1 [ FAILED ] FileUtilTest.ChangeFilePermissionsAndRead (0 ms) [1447/1448] FileUtilTest.ChangeFilePermissionsAndRead (0 ms) [ RUN ] FileUtilTest.ChangeDirectoryPermissionsAndEnumerate ../../base/file_util_unittest.cc:819: Failure Value of: c1.size() Actual: 1 Expected: 0 ../../base/file_util_unittest.cc:820: Failure Value of: GetPosixFilePermissions(file_name, &mode) Actual: true Expected: false [ FAILED ] FileUtilTest.ChangeDirectoryPermissionsAndEnumerate (0 ms) [1448/1448] FileUtilTest.ChangeDirectoryPermissionsAndEnumerate (0 ms) Retrying 3 tests (retry #2) [ RUN ] FileUtilTest.ChangeFilePermissionsAndWrite ../../base/file_util_unittest.cc:772: Failure Value of: WriteFile(file_name, kData.data(), kData.length()) Actual: 5 Expected: -1 ../../base/file_util_unittest.cc:773: Failure Value of: PathIsWritable(file_name) Actual: true Expected: false [ FAILED ] FileUtilTest.ChangeFilePermissionsAndWrite (0 ms) [1449/1451] FileUtilTest.ChangeFilePermissionsAndWrite (0 ms) [ RUN ] FileUtilTest.ChangeFilePermissionsAndRead ../../base/file_util_unittest.cc:732: Failure Value of: ReadFile(file_name, buffer, buffer_size) Actual: 5 Expected: -1 [ FAILED ] FileUtilTest.ChangeFilePermissionsAndRead (1 ms) [1450/1451] FileUtilTest.ChangeFilePermissionsAndRead (0 ms) [ RUN ] FileUtilTest.ChangeDirectoryPermissionsAndEnumerate ../../base/file_util_unittest.cc:819: Failure Value of: c1.size() Actual: 1 Expected: 0 ../../base/file_util_unittest.cc:820: Failure Value of: GetPosixFilePermissions(file_name, &mode) Actual: true Expected: false [ FAILED ] FileUtilTest.ChangeDirectoryPermissionsAndEnumerate (0 ms) [1451/1451] FileUtilTest.ChangeDirectoryPermissionsAndEnumerate (0 ms) Retrying 3 tests (retry #3) [ RUN ] FileUtilTest.ChangeFilePermissionsAndWrite ../../base/file_util_unittest.cc:772: Failure Value of: WriteFile(file_name, kData.data(), kData.length()) Actual: 5 Expected: -1 ../../base/file_util_unittest.cc:773: Failure Value of: PathIsWritable(file_name) Actual: true Expected: false [ FAILED ] FileUtilTest.ChangeFilePermissionsAndWrite (1 ms) [1452/1454] FileUtilTest.ChangeFilePermissionsAndWrite (0 ms) [ RUN ] FileUtilTest.ChangeFilePermissionsAndRead ../../base/file_util_unittest.cc:732: Failure Value of: ReadFile(file_name, buffer, buffer_size) Actual: 5 Expected: -1 [ FAILED ] FileUtilTest.ChangeFilePermissionsAndRead (0 ms) [1453/1454] FileUtilTest.ChangeFilePermissionsAndRead (0 ms) [ RUN ] FileUtilTest.ChangeDirectoryPermissionsAndEnumerate ../../base/file_util_unittest.cc:819: Failure Value of: c1.size() Actual: 1 Expected: 0 ../../base/file_util_unittest.cc:820: Failure Value of: GetPosixFilePermissions(file_name, &mode) Actual: true Expected: false [ FAILED ] FileUtilTest.ChangeDirectoryPermissionsAndEnumerate (0 ms) [1454/1454] FileUtilTest.ChangeDirectoryPermissionsAndEnumerate (0 ms) 3 tests failed: FileUtilTest.ChangeDirectoryPermissionsAndEnumerate FileUtilTest.ChangeFilePermissionsAndRead FileUtilTest.ChangeFilePermissionsAndWrite Tests took 23 seconds. ==> ERROR: Se produjo un error en check(). Cancelando...

sl1pkn07 commented on 2014-05-23 14:10 (UTC)

update to 37.0.2008.2 - new 37 series - include own libpdf builded from sources (see "pdfium" in http://build.chromium.org/f/chromium/perf/dashboard/ui/changelog.html?url=/trunk/src&range=269467:272007&mode=html). no need chromium-libpdf(-dev) anymore enjoy!

sl1pkn07 commented on 2014-05-22 12:03 (UTC)

@dz0ny: what happens??

sl1pkn07 commented on 2014-05-21 21:21 (UTC)

OW! fixed >< thanks

trizen commented on 2014-05-21 21:05 (UTC)

In chromium-dev.sh, at line 6, should be '/etc/chromium-dev/default', instead of /etc/chromium/default', otherwise chromium-dev will not start.

sl1pkn07 commented on 2014-05-21 12:18 (UTC)

@Vi0L0 i'll test 8K (with you images in crbug) images with this new version. the scroll and resize back to smooth

sl1pkn07 commented on 2014-05-21 12:13 (UTC)

update to 36.0.1985.18 - Back to use clang (more general stablity, specially with nvidia drivers 337.19)) - Remove PepperFlash and libpdf: if want use this plugins install chromium-pepper-flash(-dev) and chromium-libpdf(-dev) - adjust dependency list - Back to use system ICU greetings

sl1pkn07 commented on 2014-05-14 21:23 (UTC)

update to 36.0.1985.2 - use bundled jsoncpp - back to use GCC(4.9) instead of Clang (if set "_use_clang" to "1", then use arch clang instead of rebuild bundled clang (this use "-Dclang_use_chrome_plugins=0" to work) - set different exclude unittest for clang/GCC (ones work with clang, others not work with GCC) greetings

Vi0L0 commented on 2014-05-12 20:39 (UTC)

Sorry for delay. Bug report is here [1], feel free to add your two cents ;) [1] https://code.google.com/p/chromium/issues/detail?id=372538

sl1pkn07 commented on 2014-05-09 07:51 (UTC)

@vi0l0, same behavior with Nvidia (beta) here :S

Vi0L0 commented on 2014-05-08 17:27 (UTC)

Guys, maybe you noticed performance problem when resizing big image to its actual size? Take this 8k image for example: http://www.glif.is/publications/maps/GLIF_5-11_World_8k.jpg Resize it, and try to move around it. I'm catching almost a freeze, but only on this one images tab. I'm thinking about posting bug report, but decided to ask here at first. BTW: compiled while fglrx(catalyst) was installed, it's lagging on both fglrx and oss ati(-git).

sl1pkn07 commented on 2014-05-05 19:44 (UTC)

reger. then update now

trizen commented on 2014-05-05 17:58 (UTC)

After I've added the "--force-local-build --without-android" arguments to "tools/clang/scripts/update.sh", I couldn't compile clang with clang-3.4 (doh), so I uninstalled clang and tried again so it builds using gcc-3.9.0. At this moment, clang is still compiling... Update: the process completed successfully -- chromium is working now. So the solution on an x86 arch, is to compile clang locally with gcc, then compile chromium with the locally built clang.

sl1pkn07 commented on 2014-05-05 12:05 (UTC)

@trizen: nope, need disable clang plugins (-Dclang_use_chrome_plugins=0") but fail build http://sl1pkn07.no-ip.com/paste/view/ef1c1eca related: https://bbs.archlinux.org/viewtopic.php?id=181094 :S about --force-local-build. i think is not a solution. but i'll make test greetings

trizen commented on 2014-05-05 10:25 (UTC)

No, I don't use any of that. By looking at "src/build/tools/clang/scripts/update.sh", you can see at line 207 that it actually downloads a x64 version of clang: if [ "${OS}" = "Linux" ]; then CDS_FULL_URL="${CDS_URL}/Linux_x64/${CDS_FILE}" Where ${CDS_URL} is: https://commondatastorage.googleapis.com/chromium-browser-clang I also tried to link /usr/bin/clang, /usr/bin/clang++ and /usr/bin/llvm-symbolizer into src/build/third_party/llvm-build/Release+Asserts/bin, but got another error: error: unable to load plugin '/home/swampyx/Others/Build/chromium-dev/src/build/tools/clang/scripts/../../../third_party/llvm-build/Release+Asserts/lib/libFindBadConstructs.so': '/home/swampyx/Others/Build/chromium-dev/src/build/tools/clang/scripts/../../../third_party/llvm-build/Release+Asserts/lib/libFindBadConstructs.so: cannot open shared object file: No such file or directory' ninja: build stopped: subcommand failed. ==> ERROR: A failure occurred in build(). Aborting... So, it needs clang-3.5, I guess. I'm thinking about passing the "--force-local-build" argument to "tools/clang/scripts/update.sh" to see if this works...

sl1pkn07 commented on 2014-05-05 10:12 (UTC)

no, is for both. you use selinux/apparmor/something?

trizen commented on 2014-05-05 09:47 (UTC)

Thanks. It works now, but I made a mistake compiling it with gcc-4.9.0. Trying again now with clang. Update: It doesn't work for me (maybe clang is for x64?): /bin/sh: ../../third_party/llvm-build/Release+Asserts/bin/clang: cannot execute binary file: Exec format error ninja: build stopped: subcommand failed. ==> ERROR: A failure occurred in build(). Aborting...

sl1pkn07 commented on 2014-05-04 22:08 (UTC)

@tizen try now greetings and sorry

trizen commented on 2014-05-04 20:38 (UTC)

On x86, "https://dl.google.com/linux/direct/google-chrome-unstable_current_i386.rpm" returns 404 - Not found. Using line 83 seems to fix the problem.

sl1pkn07 commented on 2014-05-04 16:36 (UTC)

updated to 36.0.1964.2 - now build with clang (provide by chromium) (Turn back to GCC when fix run with GCC 4.9) - now NaCl/PNaCl toolchain download by chromium internal tool - pepperflash/libpdf provide from official common download site instead Fedora repository (still use RPM sources) greetings and sorry the delay

sl1pkn07 commented on 2014-05-03 22:25 (UTC)

i've tried build with clang, the error are gone (36.0.1964.2) pnacl/nacl build ok for me in the new pkgbuild i've made have several changes respect this (use internal tool provide by chromium sources instead download manually) and add one flag: -Dlinux_use_bundled_binutils=0 about .rpm (flash/pdf plugin), 36.0.1964.2 x86_64 not found (i386 ok) in google official fedora repository. this time use the google-chrome-unstable_current_${_rpm_arch}.rpm from official download site when finish test if all work ok, i'll upload the update if any want test the update: https://dl.dropboxusercontent.com/u/6596386/chromium-dev-36.0.1964.2-1.src.tar.gz greetings

misc commented on 2014-05-03 20:38 (UTC)

There's 36.0.1964.5 now, maybe try that?

misc commented on 2014-05-03 20:32 (UTC)

It's gone with 1974.0 (r267764, actually), so it's either somebody doing a bisect to find the fix, or waiting for the next dev release. Also, I can no longer get [p]nacl to build and had to disable it (-Ddisable_nacl=1 and remove the pnacl line + adjust package(), obviously). Luckily Flash nowadays works without it, I wasn't aware of that.

sl1pkn07 commented on 2014-05-03 20:14 (UTC)

have problems with GCC 4.9. build OK. but... http://wstaw.org/m/2014/05/03/plasma-desktopF11541.png all plugins/extensions stop working and all pages (include internal) gets like screenshot console log: http://sl1pkn07.no-ip.com/paste/view/64746be0 gdb: http://sl1pkn07.no-ip.com/paste/view/914bee2a dmesg: tons of http://sl1pkn07.no-ip.com/paste/view/83a83e4b any ideas?

misc commented on 2014-05-03 09:12 (UTC)

If that's still the case, use the deb files instead. For x64: https://dl.google.com/linux/chrome/deb/pool/main/g/google-chrome-unstable/google-chrome-unstable_36.0.1964.2-1_amd64.deb Needs to be extracted, and the two bsdtar lines adjusted to bsdtar -xf data.tar.lzma opt/google/chrome-unstable/[…]

fluxboxer commented on 2014-04-25 13:52 (UTC)

curl: (22) The requested URL returned error: 404 Not Found

fluxboxer commented on 2014-04-25 13:37 (UTC)

All right, will try tmpfs /tmp tmpfs nodev,nosuid,size=3G 0 0

fluxboxer commented on 2014-04-25 13:25 (UTC)

I guess so. Sorry for botherin you but should I chage my fstab entry: tmpfs /tmp tmpfs nodev,nosuid to anything else? (Have SSD)

sl1pkn07 commented on 2014-04-25 13:08 (UTC)

@qanjaro are you sure have a sufficient free space?, uncompress tar need +/- 2Gb

fluxboxer commented on 2014-04-25 10:23 (UTC)

Extracting chromium-36.0.1941.0.tar.xz with bsdtar chromium-36.0.1941.0/tools/deps2git/PRESUBMIT.py: Write failed chromium-36.0.1941.0/tools/deps2git/deps2git.py: Write failed chromium-36.0.1941.0/tools/deps2git/git_tools.py: Write failed chromium-36.0.1941.0/tools/deps2git/svn_to_git_public.py: Write failed chromium-36.0.1941.0/tools/deps2git/WATCHLISTS: Write failed chromium-36.0.1941.0/tools/deps2git/buildspec_to_git.py: Write failed chromium-36.0.1941.0/tools/symsrc/README.chromium: Write failed chromium-36.0.1941.0/tools/symsrc/COPYING-pefile: Write failed chromium-36.0.1941.0/tools/symsrc/img_fingerprint.py: Write failed chromium-36.0.1941.0/tools/symsrc/pefile.py: Write failed chromium-36.0.1941.0/tools/symsrc/pdb_fingerprint_from_img.py: Write failed chromium-36.0.1941.0/tools/symsrc/source_index.py: Write failed chromium-36.0.1941.0/tools/set_default_handler/DEPS: Write failed chromium-36.0.1941.0/tools/set_default_handler/set_default_handler.gyp: Write failed chromium-36.0.1941.0/tools/set_default_handler/set_default_handler_main.cc: Write failed chromium-36.0.1941.0/tools/linux/dump-static-initializers.py: Write failed chromium-36.0.1941.0/tools/linux/procfs.py: Write failed chromium-36.0.1941.0/tools/linux/PRESUBMIT.py: Write failed chromium-36.0.1941.0/tools/linux/tests/procfs_tests.py: Write failed chromium-36.0.1941.0/tools/memory_watcher/DEPS: Write failed chromium-36.0.1941.0/tools/memory_watcher/mini_disassembler.h: Write failed chromium-36.0.1941.0/tools/memory_watcher/call_stack.cc: Write failed chromium-36.0.1941.0/tools/memory_watcher/memory_watcher.gyp: Write failed chromium-36.0.1941.0/tools/memory_watcher/hotkey.h: Write failed chromium-36.0.1941.0/tools/memory_watcher/scripts/memtrace.pl: Write failed chromium-36.0.1941.0/tools/memory_watcher/scripts/memprof.pl: Write failed chromium-36.0.1941.0/tools/memory_watcher/scripts/summary.pl: Write failed chromium-36.0.1941.0/tools/memory_watcher/scripts/finditem.pl: Write failed chromium-36.0.1941.0/tools/memory_watcher/memory_hook.h: Write failed chromium-36.0.1941.0/tools/memory_watcher/preamble_patcher.cc: Write failed chromium-36.0.1941.0/tools/memory_watcher/README: Write failed chromium-36.0.1941.0/tools/memory_watcher/mini_disassembler_types.h: Write failed chromium-36.0.1941.0/tools/memory_watcher/mini_disassembler.cc: Write failed chromium-36.0.1941.0/tools/memory_watcher/ia32_modrm_map.cc: Write failed chromium-36.0.1941.0/tools/memory_watcher/memory_watcher.h: Write failed chromium-36.0.1941.0/tools/memory_watcher/preamble_patcher_with_stub.cc: Write failed chromium-36.0.1941.0/tools/memory_watcher/memory_watcher.cc: Write failed chromium-36.0.1941.0/tools/memory_watcher/ia32_opcode_map.cc: Write failed chromium-36.0.1941.0/tools/memory_watcher/dllmain.cc: Write failed chromium-36.0.1941.0/tools/memory_watcher/preamble_patcher.h: Write failed chromium-36.0.1941.0/tools/memory_watcher/call_stack.h: Write failed chromium-36.0.1941.0/tools/memory_watcher/memory_hook.cc: Write failed chromium-36.0.1941.0/rlz/DEPS: Write failed chromium-36.0.1941.0/rlz/rlz.gyp: Write failed chromium-36.0.1941.0/rlz/OWNERS: Write failed chromium-36.0.1941.0/rlz/mac/lib/machine_id_mac.cc: Write failed chromium-36.0.1941.0/rlz/mac/lib/rlz_value_store_mac.h: Write failed chromium-36.0.1941.0/rlz/mac/lib/rlz_value_store_mac.mm: Write failed chromium-36.0.1941.0/rlz/chromeos/lib/machine_id_chromeos.cc: Write failed chromium-36.0.1941.0/rlz/chromeos/lib/rlz_value_store_chromeos.cc: Write failed chromium-36.0.1941.0/rlz/chromeos/lib/rlz_value_store_chromeos.h: Write failed chromium-36.0.1941.0/rlz/examples/rlz_id.cc: Write failed chromium-36.0.1941.0/rlz/ios/lib/machine_id_ios.cc: Write failed chromium-36.0.1941.0/rlz/win/dll/dll_main.cc: Write failed chromium-36.0.1941.0/rlz/win/dll/exports.cc: Write failed chromium-36.0.1941.0/rlz/win/lib/lib_mutex.cc: Write failed chromium-36.0.1941.0/rlz/win/lib/rlz_value_store_registry.h: Write failed chromium-36.0.1941.0/rlz/win/lib/machine_deal.cc: Write failed chromium-36.0.1941.0/rlz/win/lib/rlz_value_store_registry.cc: Write failed chromium-36.0.1941.0/rlz/win/lib/process_info.h: Write failed chromium-36.0.1941.0/rlz/win/lib/machine_id_win.cc: Write failed chromium-36.0.1941.0/rlz/win/lib/lib_mutex.h: Write failed chromium-36.0.1941.0/rlz/win/lib/registry_util.cc: Write failed chromium-36.0.1941.0/rlz/win/lib/machine_deal_test.cc: Write failed chromium-36.0.1941.0/rlz/win/lib/registry_util.h: Write failed chromium-36.0.1941.0/rlz/win/lib/process_info.cc: Write failed chromium-36.0.1941.0/rlz/win/lib/machine_deal.h: Write failed chromium-36.0.1941.0/rlz/win/lib/rlz_lib_win.cc: Write failed chromium-36.0.1941.0/rlz/lib/string_utils.h: Write failed chromium-36.0.1941.0/rlz/lib/assert.cc: Write failed chromium-36.0.1941.0/rlz/lib/crc32.h: Write failed chromium-36.0.1941.0/rlz/lib/string_utils_unittest.cc: Write failed chromium-36.0.1941.0/rlz/lib/machine_id.cc: Write failed chromium-36.0.1941.0/rlz/lib/financial_ping.cc: Write failed chromium-36.0.1941.0/rlz/lib/lib_values_unittest.cc: Write failed chromium-36.0.1941.0/rlz/lib/machine_id.h: Write failed chromium-36.0.1941.0/rlz/lib/assert.h: Write failed chromium-36.0.1941.0/rlz/lib/lib_values.cc: Write failed chromium-36.0.1941.0/rlz/lib/rlz_enums.h: Write failed chromium-36.0.1941.0/rlz/lib/crc32_wrapper.cc: Write failed chromium-36.0.1941.0/rlz/lib/crc8_unittest.cc: Write failed chromium-36.0.1941.0/rlz/lib/rlz_lib_clear.cc: Write failed chromium-36.0.1941.0/rlz/lib/financial_ping_test.cc: Write failed chromium-36.0.1941.0/rlz/lib/rlz_lib_test.cc: Write failed chromium-36.0.1941.0/rlz/lib/crc32_unittest.cc: Write failed chromium-36.0.1941.0/rlz/lib/string_utils.cc: Write failed chromium-36.0.1941.0/rlz/lib/recursive_cross_process_lock_posix.h: Write failed chromium-36.0.1941.0/rlz/lib/rlz_value_store.h: Write failed chromium-36.0.1941.0/rlz/lib/crc8.h: Write failed chromium-36.0.1941.0/rlz/lib/crc8.cc: Write failed chromium-36.0.1941.0/rlz/lib/recursive_cross_process_lock_posix.cc: Write failed chromium-36.0.1941.0/rlz/lib/rlz_lib.h: Write failed chromium-36.0.1941.0/rlz/lib/lib_values.h: Write failed chromium-36.0.1941.0/rlz/lib/financial_ping.h: Write failed chromium-36.0.1941.0/rlz/lib/machine_id_unittest.cc: Write failed chromium-36.0.1941.0/rlz/lib/rlz_lib.cc: Write failed chromium-36.0.1941.0/rlz/test/rlz_test_helpers.cc: Write failed chromium-36.0.1941.0/rlz/test/rlz_unittest_main.cc: Write failed chromium-36.0.1941.0/rlz/test/rlz_test_helpers.h: Write failed bsdtar: Error exit delayed from previous errors.

misc commented on 2014-04-22 19:11 (UTC)

In package()'s "Extract and install PepperFlash" section, makepkg cd's into $srcdir (where pepper-flash resides), yet wants to copy that file from the parent directory (ie, ../), and thus fails.

crondog commented on 2014-04-18 11:29 (UTC)

@sl1pkn07 Just tried. That only changes the font on webpages etc

sl1pkn07 commented on 2014-04-18 10:52 (UTC)

@trizen. Thanks. added @crondog. try with https://chrome.google.com/webstore/detail/advanced-font-settings/caclkomlalccbpcdllchkeecicepbmbm

crondog commented on 2014-04-18 03:24 (UTC)

So i finally got around to updating from 34->36. Anyone know how to reduce the size of the omnibox font? It's twice the size of the bookmark fonts and setting the font size with lxappearance does not change the omnibox font size

trizen commented on 2014-04-17 23:40 (UTC)

Minor fix: 'pnacl_public_x86_64_libpnacl_irt_shim_a' should be 'pnacl_public_x86_32_libpnacl_irt_shim_a' on a 32-bit machine. PKGBUILD: http://pastebin.com/XDUfHtku

sl1pkn07 commented on 2014-04-17 11:12 (UTC)

try now

walkindude commented on 2014-04-17 10:45 (UTC)

-> 1337's g0 7h3 p4r7y! Updating projects from gyp files... Running build/landmines.py... ninja: Entering directory `out/Release' ninja: warning: multiple rules generate obj/third_party/zlib/google/zip.zip.o. builds involving this target will not be correct; continuing anyway ninja: warning: multiple rules generate obj/third_party/zlib/google/zip.zip_internal.o. builds involving this target will not be correct; continuing anyway ninja: warning: multiple rules generate obj/third_party/zlib/google/zip.zip_reader.o. builds involving this target will not be correct; continuing anyway ninja: error: 'gen/sdk/toolchain/linux_x86_newlib/stamp.untar', needed by 'gen/sdk/toolchain/linux_x86_newlib/stamp.prep', missing and no known rule to make it ==> ERROR: A failure occurred in build(). Aborting... ==> ERROR: Makepkg was unable to build chromium-dev.

sl1pkn07 commented on 2014-04-16 07:20 (UTC)

NOTICE in the 36.0.1941.0, GTK port has been complety remove in sources from now, AURA is now set by default see version log: http://build.chromium.org/f/chromium/perf/dashboard/ui/changelog.html?url=/trunk/src&range=263744:262849&mode=html

ledti commented on 2014-04-06 09:33 (UTC)

I managed to get around the error by building with clang.

ledti commented on 2014-04-06 00:57 (UTC)

For the last two updates, trying to build an aura enabled build results in the following error near the very end: libremoting client plugin.an: error adding symbols: Archive has no index; run ranlib to add one Any tips?

sl1pkn07 commented on 2014-04-04 00:57 (UTC)

https://code.google.com/p/chromium/issues/detail?id=359177 :(

WhoNeedszZz commented on 2014-03-28 14:54 (UTC)

Error occurs at this point: " -> L3ts g0 th3 p4rTy!"

WhoNeedszZz commented on 2014-03-28 14:52 (UTC)

Since the build is out of date I tried updating it myself, but I'm running into this error: Exception: Failed to open ../third_party/dom_distiller_js/js/domdistiller.js while trying to flatten ../third_party/dom_distiller_js/js/domdistiller.js. (No such file or directory) Searching google for this returns only one useless result.

sl1pkn07 commented on 2014-02-14 19:22 (UTC)

I've tried, but have a lots of problems with for example D&D from the chromium to the dolphin (the desktop get unresponsive/slow when switch between aplications (alt+tab))

jeagoss commented on 2014-02-14 19:06 (UTC)

I've been using Aura with a clang build for a while now. (The default on my repo.) I haven't run into any issues recently. Even with themes.

Nordlicht commented on 2014-02-14 15:12 (UTC)

Here is an interesting discussion about the "Aura on Linux" Topic: https://code.google.com/p/chromium/issues/detail?id=317859#c71

sl1pkn07 commented on 2014-02-14 08:09 (UTC)

DAMN!!!!!! https://wiki.archlinux.org/index.php/Creating_Packages#The_check.28.29_function option=('!check') wrong!!! (fixed now in wiki) please, if use yaourt to update package: DON'T CANCEL THE INSTALLATION IF FAILS!!!!! just go to yaourt temporal directory (/tmp/yaourt-tmp->user<) (our your tmp path defined in yaourt.rc file) and type: makepkg -fR to avoid the rebuild everiting and only start repackage if you build without AUR helper only need run the same command i've upload new PKGBUILD por solve this sorry and grettings :S

sl1pkn07 commented on 2014-02-14 06:15 (UTC)

update to 34.0.1838.2 - The pepperflash flags now store in own file (/etc/chromium-dev/pepper-flash) instead in /etc/chromium-dev/default. I've changed the wrapper launcher for handle this (/usr/bin/chromium-dev). - Unittest is now "stable", but disable by default. only for debugging @ All users with PaX environment: _use_pax=1 works? greetings

jeagoss commented on 2014-02-08 03:08 (UTC)

I don't think aura decoration should be turned on by default. However, I think the PKGEXT variable should be removed. People need to control that in their /etc/makepkg.conf. Not in a PKGBUILD file.

Nordlicht commented on 2014-02-07 17:42 (UTC)

"POLL: active Aura decorations by default in next release? - Yes - No" Devinitly no. There are a lot of showstoppers atm (Dropdown fields, Theme bugs, etc.). I would wait a while, until the bigger bugs are solved.

sl1pkn07 commented on 2014-02-07 03:24 (UTC)

@Lucy new found: https://code.google.com/p/chromium/wiki/NinjaBuild#Parallelism in next release remove all ${makeflags} greetings

sl1pkn07 commented on 2014-02-07 03:18 (UTC)

updated to 34.0.1825.4 @Lucy according of this: http://martine.github.io/ninja/manual.html#_running_ninja ninja by default no need parse -j. anyway, i've added method to use "-j N" instead "-jN" from $MAKEFLAGS (remember: to use this variable is need comment out MAKEFLAGS in /etc/makepkg.conf) for @ALL POLL: active Aura decorations by default in next release? - Yes - No greetings

commented on 2014-02-04 15:52 (UTC)

MAKEFLAGS shouldn't be passed to ninja (lines 359 and 364) as it doesn't support the same flags as make, so it fails with what's normally a valid MAKEFLAGS.

sl1pkn07 commented on 2014-02-01 21:07 (UTC)

works for me http://privatepaste.com/a55def467d

lockheed commented on 2014-02-01 20:03 (UTC)

There is a problem with downloading the package: -> Downloading chromium-34.0.1809.0.tar.xz... --2014-02-01 21:01:38-- https://commondatastorage.googleapis.com/chromium-browser-official/chromium-34.0.1809.0.tar.xz Resolving commondatastorage.googleapis.com (commondatastorage.googleapis.com)... 127.0.0.1, 2a00:1450:4001:802::100b Connecting to commondatastorage.googleapis.com (commondatastorage.googleapis.com)|127.0.0.1|:443... failed: Connection refused. Connecting to commondatastorage.googleapis.com (commondatastorage.googleapis.com)|2a00:1450:4001:802::100b|:443... failed: Address family not supported by protocol. ==> ERROR: Failure while downloading chromium-34.0.1809.0.tar.xz Aborting... I found a misplaced " inside the file, but correcting it did not help.

sl1pkn07 commented on 2014-01-30 23:45 (UTC)

thanks @jeagoss! udate to 34.0.1809.0

jeagoss commented on 2014-01-30 21:19 (UTC)

And here is the full modification set. http://src.chromium.org/viewvc/chrome?view=revision&revision=247501

jeagoss commented on 2014-01-30 21:19 (UTC)

Here is the fix. http://src.chromium.org/viewvc/chrome/trunk/src/chrome/browser/ui/gtk/autofill/autofill_popup_view_gtk.cc?r1=242491&r2=247501

misc commented on 2014-01-30 15:50 (UTC)

The GTK issue was fixed after 1809 already, but dunno by which commit.

sl1pkn07 commented on 2014-01-30 03:05 (UTC)

https://dl.dropboxusercontent.com/u/6596386/chromium-dev-34.0.1809.0-1.src.tar.gz

jeagoss commented on 2014-01-30 02:14 (UTC)

Do you host copies of the PKGBUILD anywhere? I'd like to give it a go and see if I can figure anything out.

sl1pkn07 commented on 2014-01-30 02:07 (UTC)

fail build with GTK decorations, but build OK with AURA decorations https://code.google.com/p/chromium/issues/detail?id=339158 update package anyway?

sl1pkn07 commented on 2014-01-17 14:59 (UTC)

if the hares waiting... update to 34.x series

sl1pkn07 commented on 2014-01-09 22:53 (UTC)

waiting for libwebp update....

dimaqq commented on 2013-12-11 11:17 (UTC)

If anyone wants a prebuilt 64-bit arch package with aura, I can offer this: chromium-dev-33.0.1729.3-1-x86_64.pkg.tar.xz (45MB, md5 183b4b03ee45354ac132f483ab02b1cc) built-in: aura, ncl, pepper-flash, pdf can be installed and ran side-by-side with regular chromium, profiles are not shared. however, requires that you uninstall chromium's separately packaged pepper-flash and pdf. feel free to reinstall these later and disable in chromium-dev. https://docs.google.com/file/d/0B3d3mgl1sAwLOUdwOEhDNU12alk/edit?pli=1

dimaqq commented on 2013-12-11 08:53 (UTC)

D'oh! I'm so stupid... ran makepkg and I see these two files: -rw-r--r-- 1 root root 185371392 Dec 10 14:53 chromium-33.0.1729.3.tar.xz -rw-r--r-- 1 root root 45142436 Dec 10 19:39 chromium-dev-33.0.1729.3-1-x86_64.pkg.tar.xz and I was trying to install the first instead of the second...

sl1pkn07 commented on 2013-12-02 01:29 (UTC)

because ninja and python2-ply is stored in AUR

govno.ebanoe commented on 2013-12-02 00:45 (UTC)

Please add to dependencies "ninja" and "python2-ply" packages. On clean system, makepkg says "Not found in resitories".

sl1pkn07 commented on 2013-11-23 17:32 (UTC)

finaly update to 33 series add _use_aura= if you want use new Aura Decoration greetings

misc commented on 2013-11-18 16:59 (UTC)

The gnome-keyring issue appears fixed in the next release.

misc commented on 2013-11-15 18:05 (UTC)

No, I meant the behavior that is controlled by the "Use System Title Bar and Borders" option, which is gone/disabled in 1707.0. It's back/fixed in the version I'm currently running (1711.0); dunno if the keyring issue was fixed too, though.

sl1pkn07 commented on 2013-11-14 17:58 (UTC)

@ Misc "native window decorations" means Aura? -Duse_aura=1 -Dtoolkit_uses_gtk=0 ?

misc commented on 2013-11-13 17:11 (UTC)

BTW the native window decorations are entirely broken, thus theming/Linux integration messed up, and for whatever reason compilation took a bunch of GB more, so you might want to skip this 1707.

misc commented on 2013-11-13 16:42 (UTC)

Apparently somebody killed an option check, because gnome-keyring was mandatory for my compilation of 1707 — meaning I had to enable Duse_gconf, Dlinux_link_gnome_keyring and Duse_gnome_keyring, which also implies installing libgnome-keyring. Moreover, I had to hardcode the include to <gnome-keyring.h> in chrome/browser/password_manager/native_backend_gnome_x.{h,cc} to "/usr/include/gnome-keyring-1/gnome-keyring.h" or it wouldn't find it.

sl1pkn07 commented on 2013-11-01 18:02 (UTC)

try now. update to 32.0.1687.2 fixed missing "$" in _use_pax variables

test0 commented on 2013-11-01 17:55 (UTC)

I circumvented the problem by setting -Dv8_use_snapshot=0.

sl1pkn07 commented on 2013-11-01 15:51 (UTC)

ow. missing "$" fixed in next release (in 2 or 3 hours)

test0 commented on 2013-11-01 15:42 (UTC)

Does not compile when grsecurity/pax mprotect is active even with _use_pax set. Either mksnapshot is never marked with paxctl, or (which is more probable) it is needlessly recompiled and overwritten.

sl1pkn07 commented on 2013-10-29 15:12 (UTC)

why? not works with ninja-git?

grimboy commented on 2013-10-29 00:15 (UTC)

Relies on too specific a version of ninja.

sl1pkn07 commented on 2013-10-17 19:04 (UTC)

Need rebuild again

juston_li commented on 2013-10-17 18:57 (UTC)

icu update today seems to breaks chromium /usr/lib/chromium-dev/chromium: error while loading shared libraries: libicui18n.so.51: cannot open shared object file: No such file or directory

sl1pkn07 commented on 2013-10-16 18:21 (UTC)

ninja and python2-ply set in makedepends

Nordlicht commented on 2013-10-16 17:31 (UTC)

New dependencies: ninja, python2-ply. Sadly, it doesn't build. I only get the error: "ninja: build stopped: subcommand failed."

juston_li commented on 2013-10-02 06:19 (UTC)

chromium-system-icu-r0.patch has a sha1sum mismatch, doesn't matter since PKGBUILD doesn't use it, commented out. On a separate note, Do you know how I would go about includeing https://code.google.com/p/chromium/wiki/ChromotingBuildInstructions in the build. Thanks!

sl1pkn07 commented on 2013-09-28 14:01 (UTC)

update 31.0.1650.0 - sync with gentoo ebuild - back chromedriver - now use bundle V8

sl1pkn07 commented on 2013-09-10 20:14 (UTC)

update 31.0.1622.0 - fix build with libwebp (disable for the moment) - disable build with system ffmpeg (solve gmail, google+ and other fail render) - change method to remove unused components (now use python tool include in sources) greetings

crondog commented on 2013-09-08 01:27 (UTC)

Just to let others know (if you didn't know already) I managed to get it working with use_system_ffmpeg=0. You need to remove the chromium-system-ffmpeg-r7.patch and uncomment install -Dm755 libffmpegsumo.so "${pkgdir}/usr/lib/chromium-dev/libffmpegsumo.so" I found this out from http://packages.gentoo.org/package/www-client/chromium

commented on 2013-09-04 23:21 (UTC)

related issue on chromium bug tracker: https://code.google.com/p/chromium/issues/detail?can=2&start=0&num=100&q=&colspec=ID%20Pri%20M%20Iteration%20ReleaseBlock%20Cr%20Status%20Owner%20Summary%20OS%20Modified&groupby=&sort=&id=278876

t4c commented on 2013-09-04 12:34 (UTC)

media-video/ffmpeg-1.2.2

sl1pkn07 commented on 2013-09-04 10:26 (UTC)

What version of ffmpeg use gentoo?.

t4c commented on 2013-09-04 10:08 (UTC)

Same here, Chromium crashed on google.*/ig, gmail and many more pages. Only on Archlinux, no problems with devchannel on Gentoo Linux, so it seems to be AUR problem.

commented on 2013-09-02 18:07 (UTC)

Crashes on G+, Gmail, Youtube (with HTML5 active) and many other web-site. Test on 30.x.x.x and 31.x.x.x I have investigating the issue. This problem with ffmpeg. I build with use_system_ffmpeg=1 and also with use_system_ffmpeg=0. Every time crash. I have create test page here: http://legendar.info/tmp/html5video/index.html (click on any button crash the page). Nothing interesting in traceback, but only one in dmesg: [18062.849091] Media[2262]: segfault at 0 ip 00007f6ed98181c0 sp 00007f6eb7ffe7f8 error 4 in libavutil.so.52.38.100[7f6ed980a000+35000] I also try build ffmpeg-full from aur. Nothing helps =(

juston_li commented on 2013-09-01 23:12 (UTC)

Crashes on Google+ for me too, otherwise fine.

crondog commented on 2013-09-01 08:50 (UTC)

Just wondering if anyone else is having crashes on gmail and google plus while using version >=30.x.x.x. I am currently running 29.0.1547.0 with no problems.

misc commented on 2013-08-12 01:39 (UTC)

Seems like they're gonna drop the custom path and filename for the sandbox in the next release. Its install line will need to be changed from chromium-sandbox to chrome-sandbox.

shaunren commented on 2013-07-30 14:41 (UTC)

@t4c same here.

t4c commented on 2013-07-29 09:04 (UTC)

@xavier114fch update you v8-dev package. I get a segfault with aur/chromium-dev 30.0.1573.2-0: (gdb) run Starting program: /usr/lib/chromium-dev/chromium warning: no loadable sections found in added symbol-file system-supplied DSO at 0x7ffff7ffa000 warning: Could not load shared library symbols for linux-vdso.so.1. Do you need "set solib-search-path" or "set sysroot"? [Thread debugging using libthread_db enabled] Using host libthread_db library "/usr/lib/libthread_db.so.1". [New Thread 0x7fffdf247700 (LWP 12248)] [New Thread 0x7fffdea46700 (LWP 12249)] [New Thread 0x7fffde245700 (LWP 12250)] [New Thread 0x7ffff7ff9700 (LWP 12251)] [New Thread 0x7ffff7f7d700 (LWP 12252)] [New Thread 0x7fffdda44700 (LWP 12253)] [New Thread 0x7ffff7fd8700 (LWP 12254)] [New Thread 0x7fffdc8db700 (LWP 12255)] [New Thread 0x7fffbffff700 (LWP 12256)] [New Thread 0x7fffbf7fe700 (LWP 12257)] [New Thread 0x7fffbeffd700 (LWP 12258)] [New Thread 0x7fffbe7fc700 (LWP 12259)] [New Thread 0x7fffbdffb700 (LWP 12260)] [New Thread 0x7fffbd7fa700 (LWP 12261)] [New Thread 0x7fffbc88b700 (LWP 12262)] Program received signal SIGSEGV, Segmentation fault. 0x00007fffed0edbe5 in __memcpy_ssse3_back () from /usr/lib/libc.so.6

xavier114fch commented on 2013-07-27 04:28 (UTC)

Build error: http://pastebin.com/UR12nZ2h

xavier114fch commented on 2013-07-26 03:13 (UTC)

30.0.1573.2 is available now. https://gsdview.appspot.com/chromium-browser-official/chromium-30.0.1573.2.tar.xz V8 3.20.7 is also there. https://gsdview.appspot.com/chromium-browser-official/v8-3.20.7.tar.bz2

sl1pkn07 commented on 2013-07-25 00:01 (UTC)

chromium-30.0.1573.2.tar.xz no aviable on https://gsdview.appspot.com/chromium-browser-official/ or https://commondatastorage.googleapis.com/chromium-browser-official/ no sources = no party :( PS: send comment from Euskal Encounter 21

xavier114fch commented on 2013-07-24 06:22 (UTC)

Latest version (30.0.1568.0) is available now: https://gsdview.appspot.com/chromium-browser-official/chromium-30.0.1568.0.tar.xz v8-dev also needs to update to 3.20.6.0.

sl1pkn07 commented on 2013-07-23 13:28 (UTC)

https://code.google.com/p/chromium/issues/detail?id=262123

misc commented on 2013-07-23 11:06 (UTC)

Looks like they're generated at about one tarball per hour. Since minor version bumps are included, this could take another day or two.

xavier114fch commented on 2013-07-22 02:59 (UTC)

Seems they are coming back, but the version lags behind the latest one shown on their release calendar. http://www.chromium.org/developers/calendar

sl1pkn07 commented on 2013-07-16 21:28 (UTC)

again. no sources (what fu*cking happen?)

sl1pkn07 commented on 2013-07-08 22:39 (UTC)

no sources again ¬¬

sl1pkn07 commented on 2013-06-27 10:14 (UTC)

fixed

xavier114fch commented on 2013-06-27 02:27 (UTC)

Checksum failed for the main xz package. The correct one is: f868be3b5939dc1aa83fe64fa1ca24b625dfd8e8

Det commented on 2013-06-26 09:31 (UTC)

Does ninja change so harshly upon every bugfix release that 'ninja>=1.3.4' is overdoing it?

n1f7 commented on 2013-06-20 12:26 (UTC)

Thanks for that

sl1pkn07 commented on 2013-06-20 12:20 (UTC)

ok, update pkgbuild

n1f7 commented on 2013-06-20 12:13 (UTC)

I had old version of ninja. I was able to build chromium only after I updated to recent ninja-1.3.4-1, but build should warn about outdated ninja instead of throwing error. I might misunderstood something, sorry for inconvenience

sl1pkn07 commented on 2013-06-20 12:04 (UTC)

mmm why you need mark the version in ninja array?

n1f7 commented on 2013-06-20 11:23 (UTC)

It should be 'ninja=1.3.4' then

sl1pkn07 commented on 2013-06-20 11:16 (UTC)

it's makedepend makedepends=('elfutils' 'gperf' 'mesa' 'ninja' 'perl-json' 'python2-jinja' 'python2-ply' 'python2-simplejson' 'subversion' 'xdg-utils' 'yasm') # 'libsrtp'

n1f7 commented on 2013-06-20 11:13 (UTC)

Nope, it's working with ninja-1.3.4-1 but it's not listed as a dependancy

sl1pkn07 commented on 2013-06-20 10:56 (UTC)

what version of ninja? not working with ninja-1.3.4-1?

n1f7 commented on 2013-06-20 10:45 (UTC)

Please also add versioned dependency to ninja, this should fix the problem

n1f7 commented on 2013-06-20 10:34 (UTC)

I got this ninja error again Updating projects from gyp files... ninja: Entering directory `out/Release' ninja: error: build.ninja:17: unexpected variable 'deps' deps = gcc ^ near here ==> ERROR: A failure occurred in build(). Aborting...

sl1pkn07 commented on 2013-06-20 10:08 (UTC)

roger

test0 commented on 2013-06-20 08:37 (UTC)

(Which actually means: Please add a versioned dependency.)

test0 commented on 2013-06-20 08:33 (UTC)

Funny, I got a build error somewhere else. (v8-related)

n1f7 commented on 2013-06-20 05:03 (UTC)

I have this error: -> Fix to really use python2. ==> Prepare NaCl SDK ==> Starting build()... -> Building Chromium... Updating projects from gyp files... ninja: Entering directory `out/Release' ninja: error: build.ninja:17: unexpected variable 'deps' deps = gcc ^ near here ==> ERROR: A failure occurred in build(). Aborting...

sl1pkn07 commented on 2013-06-16 14:03 (UTC)

roger!

misc commented on 2013-06-16 13:43 (UTC)

With 29.0.1539.0 they've *finally* resumed to provide new tarballs — without filling in the missing ones, however. Anyway, in order to build three changes to the PKGBUILD are required: 1. Exclude third_party/usrsctp/* 2. Exclude third_party/lzma_sdk/* 3. Add perl-json to makedepends.

test0 commented on 2013-06-08 17:04 (UTC)

The dropbox-hosted tarball works for me.

sl1pkn07 commented on 2013-06-08 08:00 (UTC)

https://dl.dropboxusercontent.com/u/6596386/chromium-dev-29.0.1521.3-0.src.tar.gz try with this pkgbuild

sl1pkn07 commented on 2013-06-08 07:57 (UTC)

install harfbuzz-icu

juston_li commented on 2013-06-08 04:58 (UTC)

Failed to open Chormium after upgrading harfbuzz too. /usr/lib/chromium-dev/chromium: symbol lookup error: /usr/lib/chromium-dev/chromium: undefined symbol: hb_icu_get_unicode_funcs Downgraded harfbuzz

xavier114fch commented on 2013-06-08 03:47 (UTC)

Fail to build after harfbuzz upgrade. extra/harfbuzz 0.9.18-1 [installed] OpenType text shaping engine http://pastie.org/8021767

sl1pkn07 commented on 2013-06-07 11:46 (UTC)

no sources = no party :(

misc commented on 2013-06-06 20:38 (UTC)

Crash reporting is not available for Chromium. From their FAQ page: > Chrome handles its own crashes by linking in Breakpad, whereas Chromium does not link in Breakpad. Therefore, when a Chromium crash occurs, it is simply ignored. You can still get a classic trace, see the ArchWiki "Debug - Getting Traces" for that. (Warning — I did that one, the compilation took ages and the package grew to little less than 900 MiB.)

juston_li commented on 2013-05-30 14:17 (UTC)

Perhaps a dumb question but is crash reporting disabled? I can't seem to find a way to retrieve crash ID. Say's "Crash reporting is disabled. Crash reporting is not available in Chromium." And the report crash statistics is not in settings. Thanks for the packages by the way

sl1pkn07 commented on 2013-05-10 01:02 (UTC)

update to 28.0.1500.5 - toolchain_linux -> no need anymore, thanks @misc for the tip note for: * Instead of patching libvpx, adding "-Dlibvpx_source=/usr/include" to _flags does the same. I prefer patch, and none of the two ways is necessary en next version https://code.google.com/p/chromium/issues/detail?id=226860#c18 greetings

misc commented on 2013-05-09 18:01 (UTC)

Two notes: * toolchain_linux_x86.tar.bz2 isn't necessary, at least I've never downloaded it. * Instead of patching libvpx, adding "-Dlibvpx_source=/usr/include" to _flags does the same.

sl1pkn07 commented on 2013-05-08 23:12 (UTC)

update to 28.0.1500.3 - add initial support to unit test (disable for default, need huge test (for me get out of memory (?¿)) - add some patch from gentoo build - add part of code from @misc pkgbuild (pnacl related) (tnx bro) - back need lib32-gcc-libsa and lib32-zlib. need active multilib repository in pacman.conf (finaly make more test (without trolling me) with chrooted installation, yes, only need these two packages, no need more.) and add warning message if detect x64 systems - now use own v8 packgae called v8-dev, this package is updated when released dev channel thanks all for the patient greetings and enjoy! http://www.youtube.com/watch?v=EsMNqUVv3mQ

sl1pkn07 commented on 2013-05-08 16:05 (UTC)

jummm. I think I tested the pkgbuild only with lib32-zlib and lib32-gcc and fail with missing files http://paste.rufian.eu/6YXA/raw/, only build with rebuild chroot with gcc-multilib (include denpendencies) well, I try pkgbuild again later when resolve build new version (ICU related?) thanks for your pkgbuild @misc. i take some parts to include in my pkgbuild (pnacl related) greetings.

misc commented on 2013-05-08 15:06 (UTC)

Here's my PKGBUILD; I can compile with nacl/pnacl enabled using just the "pure" x64 gcc (though lib32-zlib and lib32-gcc-libs are installed): http://pastie.org/7818390 It's a bit of a frankenbuild with some changes, not all are annotated.

sl1pkn07 commented on 2013-05-08 13:24 (UTC)

https://code.google.com/p/chromium/issues/detail?id=238984 :/

sl1pkn07 commented on 2013-05-08 08:49 (UTC)

lib32-zlib/lib32-gcc-libs is part of [multilib], I test build in a chrooted environment [1], these libraries not need, only need gcc-multilib (or multilib-devel)[2] (in new instalation) [1] https://wiki.archlinux.org/index.php/DeveloperWiki:Building_in_a_Clean_Chroot [2] https://www.archlinux.org/groups/x86_64/multilib-devel/

TheWaffleGuy commented on 2013-05-08 03:56 (UTC)

I've always compiled chromium with just lib32-zlib and lib32-gcc-libs, like the old PKGBUILD and it worked fine on my x86_64 system.

sl1pkn07 commented on 2013-05-07 22:30 (UTC)

roger

misc commented on 2013-05-07 22:25 (UTC)

The gold linker is now stable enough to be used for Chromium; it's much faster and requires much less memory. (It's also possible to rm -rf the chromium-build/third_party folder after make if one needs the space for package(), like I used to do with my 3G tmpfs.) To use it, simply add gold/* to the third_party exclusions and toggle the two use_gold flags.

sl1pkn07 commented on 2013-05-05 15:14 (UTC)

@macau if is possible fill bug in chromium bug tracker http://crbug.com greetings

sl1pkn07 commented on 2013-05-05 13:02 (UTC)

build with clang is broken for the moment. use gcc instead

commented on 2013-05-05 11:19 (UTC)

Trying compilation with clang. with clean pkgbuild i catch a bug at start: /usr/lib64/gcc/x86_64-unknown-linux-gnu/4.8.0/../../../../include/c++/4.8.0/type_traits:268:39: error: use of undeclared identifier '__float128' struct __is_floating_point_helper<__float128> winth -std=c++11 flag compilation go ok, but then i catch other bug: In file included from sync/engine/apply_control_data_updates.cc:11: In file included from ./sync/syncable/directory.h:12: ./base/file_util.h:416:11: error: use of undeclared identifier 'typeof' if (HANDLE_EINTR(close(*x)) < 0) ^ ./base/posix/eintr_wrapper.h:24:3: note: expanded from macro 'HANDLE_EINTR' typeof(x) eintr_wrapper_result; \ ^ In file included from sync/engine/apply_control_data_updates.cc:11: In file included from ./sync/syncable/directory.h:12: ./base/file_util.h:416:11: error: use of undeclared identifier 'eintr_wrapper_result' ./base/posix/eintr_wrapper.h:26:5: note: expanded from macro 'HANDLE_EINTR' eintr_wrapper_result = (x); \ ^ In file included from sync/engine/apply_control_data_updates.cc:11: In file included from ./sync/syncable/directory.h:12: ./base/file_util.h:416:11: error: use of undeclared identifier 'eintr_wrapper_result' ./base/posix/eintr_wrapper.h:27:12: note: expanded from macro 'HANDLE_EINTR' } while (eintr_wrapper_result == -1 && errno == EINTR); \ ^ In file included from sync/engine/apply_control_data_updates.cc:11: In file included from ./sync/syncable/directory.h:12: ./base/file_util.h:416:11: error: use of undeclared identifier 'eintr_wrapper_result' ./base/posix/eintr_wrapper.h:28:3: note: expanded from macro 'HANDLE_EINTR' eintr_wrapper_result; \ ^ In file included from sync/engine/apply_control_data_updates.cc:11: In file included from ./sync/syncable/directory.h:12: ./base/file_util.h:416:35: error: invalid operands to binary expression ('void' and 'int') if (HANDLE_EINTR(close(*x)) < 0) ~~~~~~~~~~~~~~~~~~~~~~~ ^ ~ 5 errors generated.

sl1pkn07 commented on 2013-05-05 05:23 (UTC)

@jeagos, tested the package in chroot build yes, need multilib-devel (tested only with gcc-multilib) in makedepends but this have a problem, this is not a normal makedepend, because afected all system. Little explain (my english is limited): other normal makedepends, after build package only need remove that with pacman and go. Install multilib-devel implies active [multilib] and alternate play with [core] gcc (and depends) before make the pkgbuild (for resolve cycle dependency) if want have the system clean (more people want pure 64bits systems) I only see two ways: 1) add warning in pkgbuild to need active [multilib] before build package (with all that implies) 2) change method to build Nacl components. this method is back to use https://www.archlinux.org/packages/extra/x86_64/nacl-toolchain-newlib/. I don't know if this package is out of date, and build in separate package with recent version of source stored in http://gsdview.appspot.com/nativeclient-archive2/x86_toolchain/ take a LOOOOOONNNGGGGG time (i remember need bit more time than chomium) then: build Nacl with actual method: pros: - all code up to date - no need huge size dependency, nacl-toolchain-newlib have 102Mb (uncompress) only for build (no need for run). need all time when need build chromium. Toolchain include in the actual method is cleaned after build by helpers, like yaourt (if use remote install). cons: - need enable multilib repos and install multilib-devel group before build chromium - if install multilib-devel is complicated (but not impossible) go back older method: pros: - no need multilib (for pure 64bits system) - the nacl-toolchain-newlib package is maintained by distribution cons: - code outdated - need install this package when need build chromium ladies and gentlemen, open poll :) Notice in next version: the actual v8 package in extra is "mess". don't build chromium with that (now build with a static lib provide by chromium bundle). I make a new package, calling "v8-dev" with some changes than the [extra] v8 package (use the real method to build this lib, like a Gentoo package) (and yes, if any ask, this package is a little bit clone like gentoo chromium-dev package). this package update when dev channel is released if all people have any question, commentary or insult xd. please say it :) gretings and sorry for the huge post (and of course, for my "burn brain" english)

jeagoss commented on 2013-05-04 07:05 (UTC)

multilib-devel only does the following binutils-multilib - replaces binutils gcc-multilib - replaces gcc gcc-libs-multilib - replaces gcc-libs libtool-multilib - replaces libtool lib32-gcc-libs - addition lib32-glibc - addition lib32-zlib is not needed As always, it is a pleasure working with you :)

sl1pkn07 commented on 2013-05-04 06:53 (UTC)

no, all my systems have multilib actived. yes, is possible for build nacl components need add lib32 makedepends or multilib-devel like you say, in older pkgbuilds have in makedepends array this packages: ('lib32-zlib' 'lib32-gcc-libs') can you test if only need that packages instead all multilib-devel? (for save donwnloads tons of dependencies) i need mount a chroot, but don't have space and time. i try to do something in the next few days thanks for the (regression?) report greetings and sorry my bad english :S

jeagoss commented on 2013-05-04 06:29 (UTC)

I have no idea if this has been mentioned before, but I ran into a problem compiling this on a vanilla x64 install. The build kept erroring out with a 'No such file or directory' error with x86_64-nacl-gcc. To fix the problem, I installed multilib-devel. Once the multilib tools were installed, it compiled fine. I've reproduced this on 3 different systems now and fixed it with multilib-devel. @sl1pn07 - Have you built this on a system without multilib-devel? If not, should it be considered a builddep?

sl1pkn07 commented on 2013-05-02 02:57 (UTC)

update to 28.1490.2 adapt pkgbuild to new system to set use_system_foo libraries greetings

n1f7 commented on 2013-04-18 14:11 (UTC)

Ok. Got it working with gcc and without -flto flag

n1f7 commented on 2013-04-18 10:24 (UTC)

Thanks for update, but still segfaulting on both gcc and clang builds

sl1pkn07 commented on 2013-04-17 12:01 (UTC)

sorry the delay, yesterday don't have good day :( finaly update to 28.0.1478.0 - fix build with GCC 4.8 - clean dependencies array - rework several PKGBUILD structure for pacman 4.1 - "fix" in pkgbuild build with clang (make segfault) - disable build with internal sqlite - disable pnacl support - other forgotten changes... greetings

sl1pkn07 commented on 2013-04-14 04:13 (UTC)

libpdf is is include directly in the pkgbuild by option (enable), no need external package

emmerkar commented on 2013-04-14 02:42 (UTC)

Compiled without problems, but segfaulting at launch. Removing the package gives me this message: chromium-libpdf optionally requires chromium-dev: development version of default browser, requires modifying the PKGBUILD

n1f7 commented on 2013-04-09 08:19 (UTC)

Thanks. Also segfaulting

sl1pkn07 commented on 2013-04-09 07:23 (UTC)

http://sl1pkn07.no-ip.com/paste/view/8b0255a0

n1f7 commented on 2013-04-09 07:15 (UTC)

Please provide updated pkgbuild, I get this with clang http://sl1pkn07.no-ip.com/paste/view/ee1d38ab

sl1pkn07 commented on 2013-04-09 06:57 (UTC)

fail build 28.0.1469.0 with gcc4.8 build with Clang OK. but get Segment Fault :S

emmerkar commented on 2013-04-09 06:32 (UTC)

thanks for the fix :)

emmerkar commented on 2013-04-09 06:30 (UTC)

thanks for the fix :)

emmerkar commented on 2013-04-09 06:30 (UTC)

Well it is a intel core duo P8600 2.40GHz nmot veri new but with enough power to compile. Anyway i'll retry it with longer time to compile. Thanks

emmerkar commented on 2013-04-09 06:29 (UTC)

Well it is a intel core duo P8600 2.40GHz nmot veri new but with enough power to compile. Anyway i'll retry it with longer time to compile. Thanks

sl1pkn07 commented on 2013-04-08 12:23 (UTC)

thanks for the little pkgbuild fix

n1f7 commented on 2013-04-08 11:45 (UTC)

quickfix for PKGBUILD http://sl1pkn07.no-ip.com/paste/view/f7aace3c

sl1pkn07 commented on 2013-04-08 10:30 (UTC)

https://code.google.com/p/chromium/issues/detail?id=227506 any help?

n1f7 commented on 2013-04-08 06:59 (UTC)

gcc 4.8 is fine. Do you have enough processor power? On my atom N570(4 threads) compilation takes around 14-15 hours (without linkage).

sl1pkn07 commented on 2013-04-08 06:37 (UTC)

in next build clean dependencies i don't test with gcc4.8 please wait

emmerkar commented on 2013-04-08 06:23 (UTC)

I forgot to say that for ttf-font the package should be ttf-ms-fonts Emmerkar

emmerkar commented on 2013-04-08 06:12 (UTC)

I cannot com,pile cjhromium-dev because of a strange behaviour. it begins building the package and then it acts like it entered a compiliign loop. I launched the building process yesterday and this moring it was still there and i had the im,presison it was repeting some routine. After 10 hrs is too much i guess :) here is my configuration: 3.8.5-1-ARCH x86_64 gcc (GCC) 4.8.0 alsa-lib 1.0.26-1 desktop-file-utils 0.21-1 expat 2.1.0-1 ffmpeg 1:1.2-2 flac 1.2.1-3 gpsd 3.7-1 gtk2 2.24.17-1 harfbuzz 0.9.15-1 hicolor-icon-theme 0.12-2 icu 51.1-1 jsoncpp 0.5.0-3 libevent 2.0.21-2 errore: impossibile trovare il pacchetto 'libjpeg' libpng 1.5.14-1 errore: impossibile trovare il pacchetto 'libusb' libwebp 0.3.0-1 libxslt 1.1.26-3 libxss 1.2.2-1 minizip 1.2.7-1 nspr 4.9.5-1 nss 3.14.3-2 opus 1.0.2-1 protobuf 2.5.0-3 speech-dispatcher 0.8-1 speex 1.2rc1-3 sqlite 3.7.16.1-1 errore: impossibile trovare il pacchetto 'ttf-font' errore: impossibile trovare il pacchetto 'udev' Libjpeg missing becuase is installed libusb missing because libusb compact is isntalled instead ttf-fon missing at all in official and AUR repositories udev missing at all in official and AUR repositories Maybe the prob is in the gcc version? Emmerkar

sl1pkn07 commented on 2013-03-28 18:55 (UTC)

updated: "fixed" by gentoo patch and disable pnacl support greetings

sl1pkn07 commented on 2013-03-27 03:53 (UTC)

waiting to fix http://crbug.com/224169 greetings

misc commented on 2013-03-09 12:54 (UTC)

Patchie for speech-dispatcher 0.8: http://pastie.org/6429359

sl1pkn07 commented on 2013-03-07 06:15 (UTC)

oks :) thanks for testing, now wait any 32bit build status to launch release greetings

wuffleton commented on 2013-03-07 04:42 (UTC)

Looks like something under chrome://flags was affecting it for me. Tested with a clean profile and it worked, and resetting all of those to default helped as well.

sl1pkn07 commented on 2013-03-07 04:01 (UTC)

works for me

wuffleton commented on 2013-03-07 03:36 (UTC)

Yeah, that type of dropdown menu. It appears as a couple of white pixels for me.

sl1pkn07 commented on 2013-03-07 03:35 (UTC)

scriptsafe: http://wstaw.org/m/2013/03/07/plasma-desktopC18744.png

sl1pkn07 commented on 2013-03-07 03:34 (UTC)

dropdown menu lite this? http://wstaw.org/m/2013/03/07/plasma-desktopa18744.png

wuffleton commented on 2013-03-07 03:16 (UTC)

Testing the prerelease PKGBUILD, I've encountered a pretty nasty bug: Dropdown menus no longer appear on extension buttons. Renders things like ScriptSafe pretty useless since they rely on those dropdown menus. Other than that, everything else seems to build fine and work properly on x86_64.

sl1pkn07 commented on 2013-03-06 14:40 (UTC)

https://dl.dropbox.com/u/6596386/chromium-dev/chromium-dev-27.0.1430.0-1.src.tar.gz please test, new enable build with system mesa (in special, 32 bits users) @det, You could have waited a little XD

Det commented on 2013-03-06 14:25 (UTC)

:D. I'm not saying you're incompetent if I'm flagging your package. I'm just letting people know there's a new one available. Like now.

sl1pkn07 commented on 2013-03-05 20:01 (UTC)

it's simple, if don't update package is for i found problems in build or "all" other fail (for example, sourcecode don't ship in google "ftp") and i repeat, i build the package by myself before upload any update (of course, i'm human, make fails, like latest revision with jsoncpp), install and test some time. you like lost 40min in build time to fail in stupid typo or bad apply change? and yes, i following dev-channel RSS besides, the date of launch new version almost always grabs me sleeping (time difference) You can rest assured. this package don't dead greetings and sorry for my enlish

Det commented on 2013-03-05 11:09 (UTC)

And that has anything to do with pkg-config or is somehow a bad thing because? Forgive me a little bit here but if you're following the Dev Channel why is it a bad thing that you're actually doing so?

sl1pkn07 commented on 2013-03-04 19:32 (UTC)

all my email out-of-date reported is only by you

Det commented on 2013-03-04 19:19 (UTC)

@sl1pkn07, why should I care about a pkg-config bug not reported by me, not commented by me, not linked by me and not ever, once, mentioned by me?

sl1pkn07 commented on 2013-03-02 11:29 (UTC)

yes, i see http://wstaw.org/m/2013/03/02/plasma-desktopj22940.png for my it's not bad, but could be better

wuffleton commented on 2013-03-02 03:01 (UTC)

Is it just me, or does the new version ignore system font settings and/or uses a horrid font for tab titles? I've noticed that the tab title text is much blurrier and narrower in this build.

sl1pkn07 commented on 2013-03-02 00:24 (UTC)

are?? not include?? :S sorry guys. i don't know in my local copy have include it :S updating

jeagoss commented on 2013-03-02 00:14 (UTC)

I experienced the same issue as SirWuffleton. After adding jsoncpp to the depends array and adding jsoncpp to my system from AUR, everything worked just fine. sl1pkn07, you might update the depends array to include jsoncpp

sl1pkn07 commented on 2013-03-01 22:41 (UTC)

jsoncpp is as dependency

wuffleton commented on 2013-03-01 22:34 (UTC)

In that case, could it be that jsoncpp is a dependency? There exists a jsoncpp package in the AUR, and the presence of such a package is usually implied by use_system_X=1 I'm currently testing this theory to see if that's the issue.

sl1pkn07 commented on 2013-03-01 22:29 (UTC)

change use_system_jsoncpp=1 to use_system_jsoncpp=0 but build without any probelm for me :S (always build, install and test chromium by myseft before update, is some cases several times, if have error don't upload)

wuffleton commented on 2013-03-01 22:28 (UTC)

@sl1pkno7: It's only happening for me on the new version (27.0.1425.2-1) The previous version didn't have this issue and compiled without a problem.

n1f7 commented on 2013-03-01 22:22 (UTC)

@sl1pkn07 I have the same error after last update.

sl1pkn07 commented on 2013-03-01 22:20 (UTC)

before or after new version?

wuffleton commented on 2013-03-01 22:03 (UTC)

Current version will not build on my x86_64 box. It fails with lines in reference to libjingle's json component: http://pastie.org/6362611

sl1pkn07 commented on 2013-02-27 11:23 (UTC)

@Det: https://bugs.archlinux.org/task/34046

sl1pkn07 commented on 2013-02-25 14:38 (UTC)

yea, i've upload pkgbuild with the change days ago

shosca commented on 2013-02-25 14:24 (UTC)

It builds with use_system_mesa=0

sl1pkn07 commented on 2013-02-23 19:24 (UTC)

http://paste.ubuntu.com/5559153 quick fix build with mesa 9.1 in 64bits (?) use_system_mesa=${_use_mesa} to use_system_mesa=0 (need test, don't set out-of-date package)

sl1pkn07 commented on 2013-02-23 14:09 (UTC)

aw, sorry fail upload src.tar.gx sorry the inconvenience thanks Det, for once your "out-of-date" for something worth /me runs

commented on 2013-02-21 21:20 (UTC)

@sl1pkn07 works for me

sl1pkn07 commented on 2013-02-20 09:56 (UTC)

https://dl.dropbox.com/u/6596386/chromium-dev/chromium-dev-26.0.1410.10-1.src.tar.gz please test

commented on 2013-02-20 09:49 (UTC)

The gl_bindings_skia_in_process.o fail occurs on 32 bit

sl1pkn07 commented on 2013-02-20 04:09 (UTC)

oh, i have this isuue! (close cromium keep process running) back linux_use_tcmalloc = 0 thanks for the report!

ipha commented on 2013-02-20 03:16 (UTC)

Seems like the tcmalloc bug is back(still there?) I have to compile it with "linux_use_tcmalloc=0" or it will leave a few processes running when it exits.

sl1pkn07 commented on 2013-02-20 02:48 (UTC)

https://bugs.gentoo.org/show_bug.cgi?id=457130 @all "gl_bindings_skia_in_process.o" fail users 32 or 64 bits?

commented on 2013-02-19 17:33 (UTC)

I think I'm gettin the same problem Architeckt had. Here it is: http://pastie.org/6227128

sl1pkn07 commented on 2013-02-17 21:05 (UTC)

if you're interested https://mailman.archlinux.org/pipermail/arch-dev-public/2013-February/024459.html

shosca commented on 2013-02-17 20:15 (UTC)

No its my own package https://github.com/shosca/pkgbuild-xorg-git/blob/master/mesa-git/PKGBUILD

sl1pkn07 commented on 2013-02-17 19:36 (UTC)

with libgles from extra/testing?

shosca commented on 2013-02-17 19:13 (UTC)

I use mesa-git package: http://pastie.org/6205353 Just did a clean build with your changes and still get the same error.

sl1pkn07 commented on 2013-02-17 18:50 (UTC)

@shosca how you built the package? GLES2/gl2ext.h is part of libgles, and this is in makedepends

shosca commented on 2013-02-17 18:43 (UTC)

With those changes: ==> Prepare NaCl SDK ==> Building Chromium... -> Building build project... Updating projects from gyp files... Traceback (most recent call last): File "generate_bindings.py", line 1827, in <module> sys.exit(main(sys.argv[1:])) File "generate_bindings.py", line 1778, in main print ResolveHeader(header, options.header_paths) File "generate_bindings.py", line 1763, in ResolveHeader raise Exception('Header %s not found.' % header) Exception: Header GLES2/gl2ext.h not found. gyp: Call to 'python generate_bindings.py --header-paths=../../third_party/mesa/MesaLib/include:../../third_party/khronos --inputs' returned exit status 1. while loading dependencies of gpu/gpu.gyp while loading dependencies of third_party/WebKit/Source/WebCore/WebCore.gyp/WebCore.gyp while loading dependencies of third_party/WebKit/Source/WebKit/chromium/WebKit.gyp while loading dependencies of content/browser/devtools/devtools_resources.gyp while loading dependencies of chrome/chrome_resources.gyp while loading dependencies of chrome/chrome.gyp while loading dependencies of build/all.gyp while trying to load build/all.gyp ==> ERROR: A failure occurred in build(). Aborting...

sl1pkn07 commented on 2013-02-17 07:55 (UTC)

uhmmmmmmm... thi is weird... i build chromium without any problem try this: in: msg "Remove unnecesary components" add: \! -path 'third_party/mesa/*' \ and in GYP_DEFINES="\ change "use_system_mesa=1" to "use_system_mesa=0" greetings

shosca commented on 2013-02-17 00:20 (UTC)

Build fails with http://pastie.org/6079040 I've fixed it with http://pastie.org/6199586 but now fails with http://pastie.org/6199590

sl1pkn07 commented on 2013-02-16 18:20 (UTC)

added thanks for the report @architeckt, please, test if this fix your problem greetings

thorbert commented on 2013-02-16 18:13 (UTC)

makedepends seems to be missing libegl as well.

commented on 2013-02-10 11:13 (UTC)

build fails: http://pastie.org/6108981

sl1pkn07 commented on 2013-02-06 15:31 (UTC)

strange, before post this update i test the build some times, build without error in all times http://paste.ubuntu.com/1616902/ (build log, finished a few minutes ago)

shosca commented on 2013-02-06 12:59 (UTC)

build fails with this > http://pastie.org/6079040

sl1pkn07 commented on 2013-02-06 02:36 (UTC)

try now

wuffleton commented on 2013-02-06 01:31 (UTC)

The new version doesn't compile on my box, complaining about a missing header: http://pastie.org/pastes/6076223/text

sl1pkn07 commented on 2013-02-06 00:23 (UTC)

finaly updated \*O*/ - back to use internal ffmpeg (wait to update system ffmpeg 1.1.1) - back to use internal libvpx (need update to support new VP9 codec) - back to use internal V8 engine (need downgrade ?¿?¿) - remove and add some patches - remove old sorry for this long wait, but 26.0.1397.2 (dev channel for build) not allowed in google servers greetings

sl1pkn07 commented on 2013-02-05 19:03 (UTC)

yes, is typo

nasedo commented on 2013-02-05 18:33 (UTC)

A typo? use_sistem_harfbuzz=1 \

commented on 2013-01-28 08:00 (UTC)

Any idea why chromium always fails to build with the same exact error? Here's the log: http://dpaste.com/895001/

eduardosm commented on 2013-01-20 12:21 (UTC)

I started with a new profile and now works, but it still stucks on some pages when I use the back button.

moonman commented on 2013-01-19 19:23 (UTC)

For those who are trying to build this for ARM, there are extra patches required. The topic is here: http://archlinuxarm.org/forum/viewtopic.php?f=18&t=4878

sl1pkn07 commented on 2013-01-19 14:44 (UTC)

@eduardosm http://wstaw.org/m/2013/01/19/works_1.png

sl1pkn07 commented on 2013-01-19 14:39 (UTC)

ned more info to build in ARM architecture please use http://paste.ubuntu.com/1549160 and paste the output here

eduardosm commented on 2013-01-19 14:30 (UTC)

You cannot access the settings either, it stucks like this: https://docs.google.com/file/d/0B-CglhwBY0JqLUxWejFGOV85Smc/edit

eduardosm commented on 2013-01-19 14:19 (UTC)

This last version (26.0.1386.0-1) is quite defective. Some pages freeze before loading. I do not recommend updating.

rpodgorny commented on 2013-01-19 13:20 (UTC)

...for me, it fails to build on armv6h (raspberry pi): ==> Building Chromium... -> Building build project... Updating projects from gyp files... gyp: Undefined variable asm_sources in third_party/ffmpeg/ffmpeg.gyp while loading dependencies of third_party/WebKit/Source/WebCore/WebCore.gyp/WebCore.gyp while loading dependencies of third_party/WebKit/Source/WebKit/chromium/WebKit.gyp while loading dependencies of content/browser/devtools/devtools_resources.gyp while loading dependencies of chrome/chrome_resources.gyp while loading dependencies of chrome/chrome.gyp while loading dependencies of build/all.gyp while trying to load build/all.gyp ==> ERROR: A failure occurred in build(). Aborting...

moonman commented on 2013-01-18 04:48 (UTC)

lol no worries. Thanks for keeping it up to date. Trying to build your version for ARM as the official one doesn't have the patch to use system ffmpeg and included one just won't build. They really messed it up...

sl1pkn07 commented on 2013-01-17 15:07 (UTC)

fuck fixed & sorry

moonman commented on 2013-01-17 14:55 (UTC)

use_system_protobuf=1 --> no slash so something doesn't get used, probably the line after...

misc commented on 2013-01-16 13:12 (UTC)

tcmalloc has been working fine for me for months now.

jleclanche commented on 2013-01-16 04:30 (UTC)

""" # Fix compilation with glib-2.31.6 (http://crbug.com/109527) sed 's|glib/gutils.h|glib.h|' -i "${srcdir}/chromium-build/ui/base/l10n/l10n_util.cc" """ This was fixed a while back, glib/gutils.h no longer appears in the file.

sl1pkn07 commented on 2013-01-11 03:11 (UTC)

update to 25.0.1364.29 - finaly, build with system ffmpeg support - fix build with system libpng 1.5x - build with opus support (?) (html5test say no) - add API keys from Chromium package in [extra] - add/remove some depend & makedepend - "clean" Pkgbuild Poll: enable by default NaCL support in next release: yes? no? greetings

sl1pkn07 commented on 2012-12-25 13:36 (UTC)

fixed

jeagoss commented on 2012-12-25 09:32 (UTC)

In your nacl install section, there is an error on the 32bit build. The following line: [ "${CARCH}" = "i686" ] && install -Dm755 out/Release/nacl_irt_{,_srpc}x86_32.nexe "${pkgdir}/usr/lib/${pkgname}" nacl_irt_{,_srpc}x86_32.nexe needs to be changed to nacl_irt{,_srpc}_x86_32.nexe

eherranzr commented on 2012-12-13 21:27 (UTC)

Yesterday it didn't build, today it did. Maybe some upgrade. Thanks

sl1pkn07 commented on 2012-12-12 19:16 (UTC)

need more info. post more lines before error message

eherranzr commented on 2012-12-12 19:10 (UTC)

Get the following error while building (in /home/emilio/Chorradas): FAILED with 1: /home/emilio/Chorradas/yaourt-tmp-emilio/aur-chromium-dev/src/chromium-build/out/Release/obj/gen/sdk/toolchain/linux_x86_newlib/bin/x86_64-nacl-g++ -c proxy/ppapi_messages.cc -o /home/emilio/Chorradas/yaourt-tmp-emilio/aur-chromium-dev/src/chromium-build/out/Release/obj.target/ppapi_ipc_untrusted/geni/newlib-x86-64/ppapi_ipc_untrusted/ppapi_messages.o -MD -MF /home/emilio/Chorradas/yaourt-tmp-emilio/aur-chromium-dev/src/chromium-build/out/Release/obj.target/ppapi_ipc_untrusted/geni/newlib-x86-64/ppapi_ipc_untrusted/ppapi_messages.o.d -m64 -mtls-use-call -fomit-frame-pointer -fasynchronous-unwind-tables -O2 -g -Wall -Werror -fdiagnostics-show-option -D__linux__ -D__STDC_LIMIT_MACROS=1 -D__STDC_FORMAT_MACROS=1 -D_GNU_SOURCE=1 -D_BSD_SOURCE=1 -D_POSIX_C_SOURCE=199506 -D_XOPEN_SOURCE=600 -DDYNAMIC_ANNOTATIONS_ENABLED=1 -DDYNAMIC_ANNOTATIONS_PREFIX=NACL_ -DNACL_BUILD_ARCH=x86 -DNACL_PPAPI_IPC_PROXY -DENABLE_PEPPER_THREADING -D_FORTIFY_SOURCE=2 -D_FILE_OFFSET_BITS=64 -DNO_TCMALLOC -DCHROMIUM_BUILD -DUSE_DEFAULT_RENDER_THEME=1 -DUSE_LIBJPEG_TURBO=1 -DUSE_NSS=1 -DENABLE_ONE_CLICK_SIGNIN -DGTK_DISABLE_SINGLE_INCLUDES=1 -DENABLE_REMOTING=1 -DENABLE_WEBRTC=1 -DUSE_PROPRIETARY_CODECS -DENABLE_PEPPER_THREADING -DENABLE_CONFIGURATION_POLICY -DENABLE_INPUT_SPEECH -DENABLE_NOTIFICATIONS -DENABLE_GPU=1 -DENABLE_EGLIMAGE=1 -DUSE_SKIA=1 -DENABLE_TASK_MANAGER=1 -DENABLE_WEB_INTENTS=1 -DENABLE_EXTENSIONS=1 -DENABLE_PLUGIN_INSTALLATION=1 -DENABLE_PLUGINS=1 -DENABLE_PROTECTOR_SERVICE=1 -DENABLE_SESSION_SERVICE=1 -DENABLE_THEMES=1 -DENABLE_BACKGROUND=1 -DENABLE_AUTOMATION=1 -DENABLE_GOOGLE_NOW=1 -DENABLE_LANGUAGE_DETECTION=1 -DENABLE_PRINTING=1 -DENABLE_CAPTIVE_PORTAL_DETECTION=1 -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -DNACL_WINDOWS=0 -DNACL_OSX=0 -DNACL_LINUX=0 -I/home/emilio/Chorradas/yaourt-tmp-emilio/aur-chromium-dev/src/chromium-build/out/Release/obj/gen/tc_newlib/include -I.. -I../ppapi -I.. -I../third_party/khronos make: *** [out/Release/obj/gen/tc_newlib/lib64/libppapi_ipc_untrusted.a] Error 1

sl1pkn07 commented on 2012-12-08 16:15 (UTC)

ow :S

matchulinux commented on 2012-12-08 16:09 (UTC)

I think you should remove wget from makedepends since it's not needed anymore

Det commented on 2012-11-27 18:45 (UTC)

@lessdevrandom, that is the stupidest thing I've heard in a long time. E: Nor frankly do I understand what's up with the whole "next version" syndrome. You don't have to raise the pkgver/rel every time you do something.

sl1pkn07 commented on 2012-11-18 02:07 (UTC)

oks, in next release add curl instead wget

wuffleton commented on 2012-11-18 00:09 (UTC)

@lessdevrandom: Actually, it's still pretty easy to just use curl (took all of a couple minutes to replace the wget commands with curl) @sl1pkn07: Consider using this PKGBUILD instead? (Replaced wget with curl and eliminated a couple unnecessary uses of cat that I saw) http://pastie.org/5394424

commented on 2012-11-17 22:55 (UTC)

That is because wget is scripted into the PKGBUILD on lines 66, 67, 68 and 90. It was easier to add wget as a build dependency than to edit the PKGBUILD to use curl.

Det commented on 2012-11-17 14:43 (UTC)

Why wouldn't you just use curl?

sl1pkn07 commented on 2012-11-16 02:08 (UTC)

thanks for the report

commented on 2012-11-16 01:52 (UTC)

I needed to add wget to makedepends in the PKGBUILD, but it otherwise builds and runs perfectly.

sl1pkn07 commented on 2012-11-06 12:37 (UTC)

python-simplejson move to community whit name python2-simplejson change this in new release

wuffleton commented on 2012-11-06 09:37 (UTC)

Is it just me or did the makedepend 'python-simplejson' vanish from the AUR with the updated interface? Packer keeps spitting errors about it and I can't even find a trace of it in the aur-mirror-git.

sl1pkn07 commented on 2012-11-04 10:00 (UTC)

in the pkgbuild have this (with little changes): rm -f *.sha1hash* wget -q "${_nacl_sdk_path}"/naclsdk_linux_x86.tgz.sha1hash wget -q "${_nacl_sdk_path}"/naclsdk_pnacl_linux_x86.tgz.sha1hash wget -q "${_toolchain_path}"/toolchain_linux_x86.tar.bz2.sha1hash _nacl_sdk_sha1sum="$(cat naclsdk_linux_x86.tgz.sha1hash)" _pnacl_sdk_sha1sum="$(cat naclsdk_pnacl_linux_x86.tgz.sha1hash)" _toolchain_sha1sum="$(cat toolchain_linux_x86.tar.bz2.sha1hash)" rm -f *.sha1hash* [ -e "${srcdir}/../naclsdk_linux_x86.tgz" ] && [ "$(sha1sum "${srcdir}/../naclsdk_linux_x86.tgz" | cut -d " " -f1)" != "${_nacl_sdk_sha1sum}" ] && rm -fr "${srcdir}/../naclsdk_linux_x86.tgz" [ -e "${srcdir}/../naclsdk_pnacl_linux_x86.tgz" ] && [ "$(sha1sum "${srcdir}/../naclsdk_pnacl_linux_x86.tgz" | cut -d " " -f1)" != "${_pnacl_sdk_sha1sum}" ] && rm -fr "${srcdir}/../naclsdk_pnacl_linux_x86.tgz" [ -e "${srcdir}/../toolchain_linux_x86.tar.bz2" ] && [ "$(sha1sum "${srcdir}/../toolchain_linux_x86.tar.bz2" | cut -d " " -f1)" != "${_toolchain_sha1sum}" ] && rm -fr "${srcdir}/../toolchain_linux_x86.tar.bz2" source+=("${_nacl_sdk_path}/naclsdk_linux_x86.tgz" "${_nacl_sdk_path}/naclsdk_pnacl_linux_x86.tgz" "${_toolchain_path}/toolchain_linux_x86.tar.bz2") sha1sums+=("${_nacl_sdk_sha1sum}" "${_pnacl_sdk_sha1sum}" "${_toolchain_sha1sum}") steps for this: - download the sha1file (from google) for these files and make variable with file contains. and add in sha1sums pkgbuild array - search if exists file --- if yes -> get the sha1 hash from existing file and compare with sha1has variable (get from sha1file download from google) if not same hash ------- if yes (not same hash) -> remove existing file and jump to next file ------- if no (same hash) -> jump to next file --- if no -> jump to next file make this for prevent you say i test this: https://wiki.archlinux.org/index.php/PKGBUILD#source Tip: You can specify a different name for the downloaded file - if the downloaded file has a different name for some reason like the URL had a GET parameter - using the following syntax: filename::fileuri, for example $pkgname-$pkgver.zip::http://199.91.152.193/7pd0l2tpkidg/jg2e1cynwii/Warez_collection_16.4.exe if working (the build need symlink nacl/pnacl/toolchain zips with the specific names (like download files)), add in next release greetings

pumbur commented on 2012-11-03 21:43 (UTC)

.... ==> Validating source files with sha1sums... .... pulse_audio_fix.patch ... Passed naclsdk_linux_x86.tgz ... FAILED naclsdk_pnacl_linux_x86.tgz ... FAILED toolchain_linux_x86.tar.bz2 ... Passed https://bugs.archlinux.org/index.php?do=details&task_id=29106

sl1pkn07 commented on 2012-10-31 14:42 (UTC)

@Det: what report?

jeagoss commented on 2012-10-31 01:28 (UTC)

@Det I apologize for not clarifying....

Det commented on 2012-10-31 00:00 (UTC)

@jeagoss, not as much out-of-date as meant for the _stable_ channel chromium. @sl1pkn07, jeagoss's patch (http://archrepo.jeago.com/sources/chromium-dev/pulse_audio_fix.patch) fixes http://crbug.com/157876? Well, shouldn't this be mentioned in the report?

crondog commented on 2012-10-30 06:32 (UTC)

Hey sl1pkn07 I have been building chromium-dev for a while now and have always had to fiddle with the PKGBUILD manually to get it working with BUILDDIR=/tmp/makepkg. I have finally made a PKGBUILD which keeps your functionality but is able to cope with the builddir being in a different location. --- PKGBUILD 2012-10-28 14:28:27.000000000 +1100 +++ PKGBUILD.new 2012-10-30 17:26:09.405889398 +1100 @@ -92,6 +92,7 @@ source+=("http://dl.google.com/linux/chrome/rpm/stable/"${_rpm_arch}"/google-chrome-unstable-"${pkgver}"-"${_rpm_build}"."${_rpm_arch}".rpm") sha1sums+=("${_rpm_sha1}") noextract+=("google-chrome-unstable-"${pkgver}"-"${_rpm_build}"."${_rpm_arch}".rpm") + _before_srcdir="$(pwd)" fi # Are we in Gnome? @@ -372,7 +373,7 @@ install -Dm755 "${srcdir}"/"${pkgname}".sh "${pkgdir}"/usr/bin/"${pkgname}" install -Dm644 LICENSE "${pkgdir}"/usr/share/licenses/"${pkgname}"/LICENSE - install -Dm644 "${srcdir}"/../default "${pkgdir}"/etc/chromium-dev/default + install -Dm644 "${srcdir}"/default "${pkgdir}"/etc/chromium-dev/default if [ "${_use_nacl}" = 1 ]; then msg2 "Adding NaCl components" @@ -406,7 +407,7 @@ for i in "${srcdir}"/opt/google/chrome/PepperFlash/*; do install -m644 "$i" "${_chromium_home}"/PepperFlash; done chmod 775 "${_chromium_home}"/PepperFlash/libpepflashplayer.so _flash_version="$(cat "${_chromium_home}"/PepperFlash/manifest.json | grep version | sed 's|[a-z,": ]*||g')" - sed -e "s|use_pepperflash=0|use_pepperflash=1|" -e "s|version=0|version=${_flash_version}|" -i "${srcdir}"/../"${pkgname}".install + sed -e "s|use_pepperflash=0|use_pepperflash=1|" -e "s|version=0|version=${_flash_version}|" -i $_before_srcdir/"${pkgname}".install rm -fr "${srcdir}"/opt fi

sl1pkn07 commented on 2012-10-28 11:29 (UTC)

update to 24.0.1305.3 - fix http://crbug.com/157876 (thanks to @jeagoss) - back to use internal copy of V8 sorry for delay

jeagoss commented on 2012-10-27 18:55 (UTC)

I have updated to 24.0.1305.3. In order to get it to compile on Arch, you'll need the following patch. http://archrepo.jeago.com/sources/chromium-dev/pulse_audio_fix.patch Also, for all of you experiencing crashes on Facebook, Google Reader, etc... the V8 library included in Arch is out of date. Try installing the v8-svn package from aur. Doing so has solved my problems with crashing. (You might need to recompile chromium after updating v8.)

sl1pkn07 commented on 2012-10-12 09:08 (UTC)

@hard try install v8 from community

sl1pkn07 commented on 2012-10-12 09:05 (UTC)

@cpatrick try now

HarD commented on 2012-10-12 06:46 (UTC)

In file included from ./webkit/plugins/webview_plugin.h:15:0, from webkit/plugins/webview_plugin.cc:5: ./third_party/WebKit/Source/WebKit/chromium/public/WebFrameClient.h:46:16: fatal error: v8.h: No such file or directory compilation aborted. make: *** [out/Release/obj.target/glue/webkit/plugins/webview_plugin.o] Error 1

cpatrick08 commented on 2012-10-12 01:16 (UTC)

error installing curl: (22) The requested URL returned error: 404 Not Found ==> ERROR: Failure while downloading google-chrome-unstable-24.0.1290.1-159713.x86_64.rpm

sl1pkn07 commented on 2012-09-21 13:26 (UTC)

the _toochain_rev is determine by hand because i search the revision by date release. the 23.0.1271.1 released 20 september 2012. then search toolchain revision release same day make this for possible incompatibility in buld

misc commented on 2012-09-21 12:49 (UTC)

The latest _toochain_rev can be dynamically determined eg. with: curl -s 'https://gsdview.appspot.com/nativeclient-archive2/x86_toolchain/?marker=x86_toolchain%2Fr9801%40' | grep toolchain | tail -n2 | head -n1 | cut -d '>' -f 3 | cut -c 2-5 (Note: For some weird reason that Google server acts up every now and then, requiring one to add/remove the "%40".)

commented on 2012-09-08 05:32 (UTC)

crashes some sites: facebook, orkut and some albums picasa

sl1pkn07 commented on 2012-09-05 08:38 (UTC)

adblock and autopagerize fails for me

wuffleton commented on 2012-09-05 08:33 (UTC)

Is anyone else experiencing constant crashes of some of their extensions with the latest version of chromium? Adblock and Flag for Chrome won't load at all for me and simply crash every time I try to reload them.

sl1pkn07 commented on 2012-08-31 10:09 (UTC)

good fix -> if "sudo pacman -Qo /usr/lib/libudev.so.0" result is don't provide any package. remove libudev.so.0 (rm /usr/lib/libudev.so) and rebuild chromium-dev. if other program fail to use libudev.so.0, rebuild these program (all packages in core/extra/community/testing use libudev.so.1 instead libudev.so.0) (probably AUR package) ugly fix -> sudo rm /usr/lib/libudev.so.0 && ln -s /usr/lib/libudev.so.1.1.0 /usr/lib/libudev.so.0

lockheed commented on 2012-08-31 08:56 (UTC)

$ chromium-browser /opt/chromium-browser/chrome: error while loading shared libraries: libudev.so.0: cannot open shared object file: No such file or directory And I already have all this in /usr/lib/ libudev.a libudev.so.0 libudev.so.1.1.0 libudev.so libudev.so.1

sl1pkn07 commented on 2012-08-28 19:14 (UTC)

finaly update to 23.0.1246.0 - use V8 from extra (finaly update to latest version) - use nacl-toolchain-newlib from extra instead external build (keep old code in PKGBUILD) - add new makedependencies - Remove unnecesary files (resources folder) greetings and sorry the inconveniencies (i'm back to the holidays)

misc commented on 2012-08-25 20:28 (UTC)

Wrote a few lines that move the nacl_sdk download to pacman, thus requiring only one when it does not change over various Chromium releases. The relevant parts: 1. At the top of the PKGBUILD: _nacl_sdk=$(curl -s 'https://commondatastorage.googleapis.com/nativeclient-mirror/nacl/nacl_sdk/naclsdk_manifest2.json' | grep linux -B 4 | tail -n 5 | head -n 1 | cut -d '/' -f 7) (Note: The json file does change its structure occasionally. You get the most recent Chromium tarball with eg.: "curl -s 'https://gsdview.appspot.com/chromium-browser-official/?marker=chromium-23.0.1245.0.tar.bz' | grep chromium-2 | tail -1 | cut -d '>' -f 2 | cut -d '/' -f 3 | cut -c 10-20" To get then the version for the Chrome rpm with the Pepper Flash plugin, use "_rpm_ver="$(cat filelists.xml | grep -e "unstable" | cut -d \" -f10)"" as replacement variable.) 2. In source, add: naclsdk_linux-$_nacl_sdk.tar.bz2::http://commondatastorage.googleapis.com/nativeclient-mirror/nacl/nacl_sdk/$_nacl_sdk/naclsdk_linux.tar.bz2 The first part adds the revision to the downloaded file's name, so pacman can know if any present one is outdated. 3. Within the "if [ "${_use_nacl}" = 1 ]; then" paragraph change: i) only rm -fr sdk_cache ii) add below that "mv $srcdir/pepper_2? $srcdir/nacl_sdk/"${_pepper_version}"" That should be it. However, I did not care about the pacman checksum, since I build with --skipchecksums.

sl1pkn07 commented on 2012-07-24 19:57 (UTC)

finally! update new version - Fix bison 2.6 build http://crbug.com/138243 - Fix Nacl build fail https://crbug.com/137628 - Fix build libyuv when use external libjpg-turbo build

sl1pkn07 commented on 2012-07-24 19:38 (UTC)

https://projects.archlinux.org/svntogit/packages.git/tree/trunk/chromium-20.0.1132.57-bison-2.6-remove-yyparse-decl.patch?h=packages/chromium

misc commented on 2012-07-24 19:21 (UTC)

Hacky bison 2.6 build fix in Chromium 20.0.1132.57-2

wuffleton commented on 2012-07-24 14:35 (UTC)

Just wondering, does anyone have Chromium working with the --use-pulseaudio flag? I downgraded bison and recompiled with PulseAudio present on my system, but it seems to segfault as soon as it tries to play something (I tested with a HTML5 Youtube Video). It outputs this when run from the command line: Assertion 's' failed at pulse/stream.c:2050, function pa_stream_set_write_callback(). Aborting. zsh: abort chromium-dev --use-pulseaudio

sl1pkn07 commented on 2012-07-21 22:49 (UTC)

yes. i see in https://code.google.com/p/chromium/issues/detail?id=138243 but the latest version in core is 2.6. to fix build need downgrade bison out of repository or wait external patch (the attachment diff is generate code output from bison) i prefeer wait external patch

misc commented on 2012-07-21 22:16 (UTC)

Had it happen now, too. Following the Gentoo people, a revert to bison 2.5.1 fixed it.

sl1pkn07 commented on 2012-07-20 13:48 (UTC)

shit. other fail build with disable NACL https://code.google.com/p/chromium/issues/detail?id=138243 :S

misc commented on 2012-07-17 20:47 (UTC)

Not sure about that NaCL breakage as I don't use it. However, as for that tcmalloc-fix the one in extra's Chromium appears less tinker-ish than mine. Also, the recent addition libc_urandom_override.cc (note: it came with 1208.0 IIRC) lacks an include for GCC 4.7. Link points to both fixes in one patch: http://pastie.org/4273855 Moreover, 1210.0+ needs "\! -path 'third_party/hyphen/*' \".

sl1pkn07 commented on 2012-07-17 04:40 (UTC)

fail build https://code.google.com/p/chromium/issues/detail?id=137628

sl1pkn07 commented on 2012-07-08 18:47 (UTC)

zankius!

misc commented on 2012-07-08 18:44 (UTC)

Here's the diff that I've been using to make tcmalloc work again: http://pastie.org/4221566

misc commented on 2012-07-06 17:59 (UTC)

Appears that tcmalloc is again broken with glibc 2.16; compilation fails rather instantly (worked fine with 2.15). No problem for this package though, as it's already disabled in the PKGBUILD.

wuffleton commented on 2012-06-22 08:34 (UTC)

If you're referring to a segfault when attempting to print, then I've been experiencing that for the past few versions. I'm also on x86_64. Here's my journal output from after the crash: Jun 22 01:30:52 Night kernel: [181718.631052] chromium[2032] general protection ip:7f1d1baac40e sp:7fffeb168d70 error:0 in libgtk-x11-2.0.so.0.2400.10[7f1d1b85b000+422000]

sl1pkn07 commented on 2012-06-21 00:27 (UTC)

"oops" when print screen?. same issue here :/

jk121960 commented on 2012-06-20 18:57 (UTC)

Hi, I am using the latest build 21.0.1171.0-1 when I click to print it just closes all instances that are open. I am using Arch-64 Awesome desktop, GTK2 themeing using lxappearance. On Chromium I am using sync profiling. I hope I am supposed to post this here of not please direct me as to where. If there is anything else you need please let me know. thanks --jerry

misc commented on 2012-06-18 17:45 (UTC)

With r142750 (in 1180) the missing include for memory_details_linux has been added. IOW, amazingly no patches at all are needed — for now. "* \! -path 'third_party/qcms/*' \" has become necessary, though.

sl1pkn07 commented on 2012-06-14 00:03 (UTC)

* \! -path 'third_party/libwebp/*' \ is include in this pkgbuild (+ use_system_libwebp=0) thansk for the info greetings

misc commented on 2012-06-13 22:36 (UTC)

I've been able to compile without setting use_system-libvpx at all. However as of 1173, the next version will need: * use_system_libwebp=0 * \! -path 'third_party/trace-viewer/*' \ * \! -path 'third_party/libwebp/*' \ * No more sqlite patch as it has been integrated. Also, there is no need for the remaining GCC 4.7 patch/additional cstddef includes anymore. Right now the memory_details_linux patch is the only one truly required.

sl1pkn07 commented on 2012-06-12 17:49 (UTC)

update to 21.0.1171.0 - remove disable_glibc (fail build) - set use_system-libvpx to 0 (if set 1 fail build) - add patch to fix memory_details_linux.cc build - remove patch chromium-expat-r0.patch - change patch gcc47v3.patch to gcc47v4.patch - remove patch ipc-fix.diff greetings

wuffleton commented on 2012-06-08 08:17 (UTC)

Looks like my ALSA issue was due to chromium not liking the IPC key I had set for dmixer. Changing it from 4096 to 3072 resolved the issue. I'm currently compiling 21.0.1167.0 to see if the fix has been merged yet. Besides a few of the patches becoming unnecessary or requiring edits, it would seem that disable_glibc needs to be set to 0 in order to compile. The bug that required it to be set to 1 appears to have been resolved, and if set to 1 now, the build will fail. (Just a heads up for any other people wanting try the bleeding-edge tarballs for bug fixes)

jeagoss commented on 2012-06-07 22:39 (UTC)

The extensions problem will be fixed in the next revision. http://src.chromium.org/viewvc/chrome?view=rev&revision=140954

wuffleton commented on 2012-06-05 18:24 (UTC)

Any other pure-ALSA users having issues with the new version and HTML5 Audio/Video? It floods messages similar to this on the command line when attempting to do anything that makes noise: [13373:13386:5014272109:ERROR:alsa_output.cc(670)] ALSA returned 4073 of 3763 frames available. [13373:13386:5014457904:ERROR:alsa_output.cc(670)] ALSA returned 4058 of 3763 frames available.

sl1pkn07 commented on 2012-06-03 17:07 (UTC)

no crash for me, but if left click in plugins don't show menu of these plugins

misc commented on 2012-06-03 11:37 (UTC)

Me too. Also, Chromium crashes if I try to deactivate such extensions (eg. ScriptNo) via right click menu.

jeagoss commented on 2012-06-03 01:12 (UTC)

I've noticed the same behavior as DarkSniper. Trying to step through it in gdb.

wuffleton commented on 2012-06-03 01:04 (UTC)

Anyone else having weird behavior with extension menus in this version? I've found that the menus for extensions like adblock and ghostery (which have toolbar buttons next to the wrench menu) appear on the wrong side of the window if at all.

sl1pkn07 commented on 2012-05-24 13:07 (UTC)

http://code.google.com/p/chromium/issues/detail?id=129221 ?

sl1pkn07 commented on 2012-05-23 21:48 (UTC)

mmm no for me (for the moment)

misc commented on 2012-05-23 20:32 (UTC)

Anybody else getting permanent crashes when closing a certain tab?

sl1pkn07 commented on 2012-05-23 07:10 (UTC)

at last!! update to 21.0.1145.0 - remove pulseaudio patch - mod gcc47.diff to gcc47v3.diff - add symlink needed for build with Nacl - add patch for http://crbug.com/129173 (include modifications in gcc47.diff) - add changes sugsted by @misc greetings and enjoy!

misc commented on 2012-05-22 20:59 (UTC)

Odd. All I can tell is that with issue10378057_5035.diff reverted, compilation for me ends successful, otherwise I get exactly the same error. Maybe applying this patch instead (ie, without the revert) will solve it. Haven't got any time left to test it today: https://pastebin.com/raw.php?i=VbFK7MjL (Like last time, pastebin ate the last two lines again; one with two spaces, one blank.)

sl1pkn07 commented on 2012-05-22 19:52 (UTC)

yea. the fail appear with changes you say: - "\! -path 'third_party/zlib/contrib/minizip/*' \" → "\! -path 'third_party/zlib/*' \" (change in new pkgbuild) - "\! -path 'third_party/libxml/*' \" → find&delete or compilation fails. (done in 1132.11) - Patch fragment for base/message_pump_libevent.cc needs to be removed as include is no longer missing. (change new pkgbuild, rejected gcc47.diff patch, new gcc47v2.diff: http://paste.kde.org/484514) - Like Chromium in Extra, "CFLAGS+=' -Wno-c++0x-compat'" has to be added or else GCC 4.7 prints a couple of thousand useless warnings. (done in 1132.8 or 11. this add new tons of messages: cc1: warning: command line option '-Wno-c++0x-compat' is valid for C++/ObjC++ but not for C [enabled by default]) - This commit must be reverted (or perhaps a new include is missing) for compilation to not fail: http://codereview.chromium.org/10378057. (change in new pkgbuild) changes by me: add symbolic link from native_client/toolchain/linux_x86_newlib → native_client/toolchain/linux_x86 (fail build without this) but fails (see post below) need other change? the new pkgbuild http://paste.kde.org/484568 (issue10378057_5035.diff: http://codereview.chromium.org/download/issue10378057_5035.diff)

misc commented on 2012-05-22 16:28 (UTC)

There's a reason that I wrote those comments, you know.

sl1pkn07 commented on 2012-05-22 15:05 (UTC)

g++ '-D_FILE_OFFSET_BITS=64' '-DCHROMIUM_BUILD' '-DUSE_NSS=1' '-DGTK_DISABLE_SINGLE_INCLUDES=1' '-DENABLE_REMOTING=1' '-DENABLE_WEBRTC=1' '-DENABLE_P2P_APIS=1' '-DUSE_PROPRIETARY_CODECS' '-DENABLE_CONFIGURATION_POLICY' '-DENABLE_INPUT_SPEECH' '-DENABLE_NOTIFICATIONS' '-DENABLE_GPU=1' '-DENABLE_EGLIMAGE=1' '-DUSE_SKIA=1' '-DENABLE_TASK_MANAGER=1' '-DENABLE_WEB_INTENTS=1' '-DENABLE_EXTENSIONS=1' '-DENABLE_PLUGIN_INSTALLATION=1' '-DENABLE_PROTECTOR_SERVICE=1' '-DENABLE_SESSION_SERVICE=1' '-DENABLE_THEMES=1' '-DENABLE_BACKGROUND=1' '-DENABLE_PROMO_RESOURCE_SERVICE=1' '-DENABLE_AUTOMATION=1' '-D__STDC_FORMAT_MACROS' '-DIPC_IMPLEMENTATION' '-DNDEBUG' '-DNVALGRIND' '-DDYNAMIC_ANNOTATIONS_ENABLED=0' -I. -pthread -fno-exceptions -fno-strict-aliasing -Wall -Wno-unused-parameter -Wno-missing-field-initializers -fvisibility=hidden -pipe -fPIC -pthread -I/usr/include/glib-2.0 -I/usr/lib/glib-2.0/include -pthread -I/usr/include/glib-2.0 -I/usr/lib/glib-2.0/include -I/usr/include/gtk-2.0 -I/usr/lib/gtk-2.0/include -I/usr/include/atk-1.0 -I/usr/include/cairo -I/usr/include/gdk-pixbuf-2.0 -I/usr/include/pango-1.0 -I/usr/include/pixman-1 -I/usr/include/freetype2 -I/usr/include/libpng15 -O2 -fno-ident -fdata-sections -ffunction-sections -march=x86-64 -mtune=native -O2 -pipe -fstack-protector --param=ssp-buffer-size=4 -D_FORTIFY_SOURCE=2 -Wno-c++0x-compat -fno-rtti -fno-threadsafe-statics -fvisibility-inlines-hidden -Wsign-compare -MMD -MF out/Release/.deps/out/Release/obj.target/ipc/ipc/ipc_sync_message.o.d.raw -c -o out/Release/obj.target/ipc/ipc/ipc_sync_message.o ipc/ipc_sync_message.cc ipc/ipc_platform_file.cc: In function 'IPC::PlatformFileForTransit IPC::GetFileHandleForProcess(base::PlatformFile, base::ProcessHandle, bool)': ipc/ipc_platform_file.cc:39:43: error: '::dup' has not been declared make: *** [out/Release/obj.target/ipc/ipc/ipc_platform_file.o] Error 1 make: *** Waiting for unfinished jobs....

misc commented on 2012-05-21 16:33 (UTC)

Still need to revert that commit with 1045.0. @softwayer: Can happen, but should not (tarballs are indicatory: 1084.46 184.0M vs. 1132.11 185.4M).

Softwayer commented on 2012-05-18 13:40 (UTC)

So the version of latest stable build is bigger than of the unstable one?

sl1pkn07 commented on 2012-05-18 13:30 (UTC)

in dev-channel? no http://googlechromereleases.blogspot.com.es/2012/05/dev-channel-update_17.html

sl1pkn07 commented on 2012-05-18 13:25 (UTC)

@Softwayer out of date?? why???

Softwayer commented on 2012-05-18 13:23 (UTC)

The latest version is 21.0.1141.0, isn't it?

sl1pkn07 commented on 2012-05-18 10:06 (UTC)

oh my bad! oks. is for active clang option instead GCC sorry. change now

FredBezies commented on 2012-05-18 09:26 (UTC)

A little better, but got a building problem later : chrome/browser/policy/policy_path_parser_linux.cc:35:36: error: use of undeclared identifier 'geteuid' struct passwd* user = getpwuid(geteuid()); ^ chrome/browser/policy/policy_path_parser_linux.cc:45:9: error: use of undeclared identifier 'gethostname' if (gethostname(machinename, 255) == 0) { ^ 2 errors generated. make: *** [out/Release/obj.target/browser/chrome/browser/policy/policy_path_parser_linux.o] Erreur 1

FredBezies commented on 2012-05-18 07:59 (UTC)

Didn't think about this problem. Bad letters :]

sl1pkn07 commented on 2012-05-18 02:46 (UTC)

@fredbezies Téléchargements <- try in other place (without accents)

misc commented on 2012-05-17 20:02 (UTC)

≥1140: "\! -path 'third_party/zlib/contrib/minizip/*' \" → "\! -path 'third_party/zlib/*' \"

FredBezies commented on 2012-05-16 17:11 (UTC)

Crash on building : ==> Patch, update and copy NaCl SDK ~/Téléchargements/chromium-dev/src/nacl_sdk ~/Téléchargements/chromium-dev/src/chromium-build Traceback (most recent call last): File "/home/fred/Téléchargements/chromium-dev/src/nacl_sdk/sdk_tools/sdk_update.py", line 1060, in <module> sys.exit(main(sys.argv[1:])) File "/home/fred/Téléchargements/chromium-dev/src/nacl_sdk/sdk_tools/sdk_update.py", line 1053, in main InvokeCommand(args) File "/home/fred/Téléchargements/chromium-dev/src/nacl_sdk/sdk_tools/sdk_update.py", line 1042, in InvokeCommand command(options, args[1:]) File "/home/fred/Téléchargements/chromium-dev/src/nacl_sdk/sdk_tools/sdk_update.py", line 899, in Update bundle_path = os.path.join(options.sdk_root_dir, bundle_name) File "/usr/lib/python2.7/posixpath.py", line 71, in join path += '/' + b UnicodeDecodeError: 'ascii' codec can't decode byte 0xc3 in position 12: ordinal not in range(128) ==> ERREUR : Une erreur s'est produite dans build(). Abandon...

misc commented on 2012-05-16 14:18 (UTC)

≥1138: 4. "\! -path 'third_party/libxml/*' \" to find&delete or compilation fails.

misc commented on 2012-05-16 11:59 (UTC)

Applies to 1133 (or so) til at least 1137.5: 1. Patch fragment for base/message_pump_libevent.cc needs to be removed as include is no longer missing. 2. This commit must be reverted (or perhaps a new include is missing) for compilation to not fail: http://codereview.chromium.org/10378057 3. Like Chromium in Extra, "CFLAGS+=' -Wno-c++0x-compat'" has to be added or else GCC 4.7 prints a couple of thousand useless warnings.

sl1pkn07 commented on 2012-05-12 18:43 (UTC)

sl1pkn07@sL1pkn07 ~ $ locate localefwd.h /home/sl1pkn07/aplicaciones/chromium-dev/src/nacl_sdk/pepper_20/toolchain/linux_x86_glibc/x86_64-nacl/include/c++/4.4.3/bits/localefwd.h /home/sl1pkn07/aplicaciones/chromium-dev/src/nacl_sdk/pepper_20/toolchain/linux_x86_newlib/x86_64-nacl/include/c++/4.4.3/bits/localefwd.h /usr/include/c++/4.7.0/bits/localefwd.h sl1pkn07@sL1pkn07 ~ $ pkgfile localefwd.h core/gcc community/arm-wince-cegcc-gcc community/mingw32-gcc multilib/gcc-multilib sl1pkn07@sL1pkn07 ~ $ yaourt -Qo /usr/include/c++/4.7.0/bits/localefwd.h /usr/include/c++/4.7.0/bits/localefwd.h es propiedad de gcc-multilib 4.7.0-6 sl1pkn07@sL1pkn07 ~ $ in this case. localefwd.h is provided by pepper_20 download by chromium-dev (if nacl is active) regarding .pacnew file.. in the old packages, the variable CHROMIUM_FLAGS=" " is empty . in new default only remove the space between " ". only have hard modifications the chromium-dev.install file i'm tested this in last night with all possibility: uninstall all, fresh install with/out pepperfrash, update pachage with/out pepperflash when initial fresh install have pepperflash and not, edit by hand the version of pepperflah to view update version take effect, etc. all test sucessfull (include if have other flags before/afther pepperflash flags) of course. nothig is perfect :) gretings

wuffleton commented on 2012-05-12 17:37 (UTC)

20.0.1132.3-1's compile succeeded for me, but I noticed that the .pacnew file for /etc/default didn't work. When I removed the flags relating to pepper-flash (the new default is blank), it was not detected by chromium. Regarding the compile issue: I do not have this file either, but it compiled just fine on my box.

imrehg commented on 2012-05-12 16:17 (UTC)

compilaton of 20.0.1132.3-1 fails with: In file included from ./remoting/protocol/authentication_method.h:16:0, from remoting/protocol/authentication_method.cc:5: /usr/lib/gcc/x86_64-unknown-linux-gnu/4.7.0/../../../../include/c++/4.7.0/string:45:69: fatal error: /usr/include/cairo/bits/localefwd.h: Permission denied compilation terminated. make: *** [out/Release/obj.target/remoting_protocol/remoting/protocol/authentication_method.o] Error 1 Actually, the file /usr/include/cairo/bits/localefwd.h does not exists in my system, is that a missing dependency? What would it be?

sl1pkn07 commented on 2012-05-09 16:28 (UTC)

with new compilation (rev2) fails for me :S

sl1pkn07 commented on 2012-05-09 15:11 (UTC)

oh! i see. yes i'm use the raw patch (http://chromium.googlecode.com/issues/attachment?aid=1225250019000&name=sqlite-3.7.6.3-fix-out-of-scope-memory-reference.patch&token=jqQlLFZFmSnbOxt60rCB6H5mAIo%3A1336575820598) oks. update pkgbuild to v2 sorry the inconvenient :S

wuffleton commented on 2012-05-09 14:41 (UTC)

Looks like you're using the raw patch (which came directly from sqlite) you need to patch third_party/sqlite/amalgamation/sqlite3.c in order for it to be included in the build. It was mentioned on the bug report comments and the new chromium PKGBUILD does this. The edited patch I posted earlier is pointed at that file (or you could just do what the PKGBUILD in extra is doing). Things have been stable for me with this patch included.

sl1pkn07 commented on 2012-05-09 13:03 (UTC)

i think is my faul to apply sqlite patch. i'm test new pkgbuild with chages like chromium package from extra please patient

sl1pkn07 commented on 2012-05-09 12:46 (UTC)

ag! still fails :S

wuffleton commented on 2012-05-09 02:01 (UTC)

Just a heads up, a patch that fixes the segfaults was posted on the bug report page ( http://crbug.com/122525 ). I edited the patch since a later comment indicated that the patch needed to be used on a different file in order to be included in the build. You can get that patch here: http://dl.dropbox.com/u/6464618/sqlite-3.7.6.3-fix-out-of-scope-memory-reference.patch Then just add this to the PKGBUILD and you're good to go: patch --silent -Np0 -i "${srcdir}"/sqlite-3.7.6.3-fix-out-of-scope-memory-reference.patch Also, 20.0.1130.1 is out according to the RPM list.

fishburn commented on 2012-05-07 15:23 (UTC)

That's strange that I'm the only one having the problem. After searching /usr/lib for *expat* before and after installing pyxml, it looks like: /usr/lib/python2.7/lib-dynload/pyexpat.so is superseded by: /usr/lib/python2.7/site-packages/_xmlplus/parsers/pyexpat.so If I rename the new pyexpat.so, the same error returns. I have no idea how this arose, but reinstalling all of the relevant packages didn't change anything. Its not a big problem since its working now, but it is strange.

jeagoss commented on 2012-05-07 08:46 (UTC)

I definitely don't have pyxml installed, and chromium-dev compiles for me every time. It seems there might have been some kind of install problem that the installation of pyxml might have repaired.

wuffleton commented on 2012-05-07 08:22 (UTC)

I can confirm this. I have been able to to build chromium-dev successfully and have never needed pyxml. I'm running x86_64 if it makes a difference. My output of the stuff sl1pkn07 ran: [dark@Night ~]% python2 -c "import xml.parsers.expat" [dark@Night ~]% packer -Qs xml core/expat 2.1.0-1 An XML parser library extra/libplist 1.8-2 A library to handle Apple Property List format whereas it's binary or XML extra/libxml++ 2.34.2-1 C++ bindings to libxml2 extra/libxml2 2.7.8-2 XML parsing library, version 2 extra/libxslt 1.1.26-3 XML stylesheet transformation library extra/perl-xml-parser 2.41-2 Expat-based XML parser module for perl extra/perl-xml-simple 2.18-4 Simple XML parser for perl extra/raptor 2.0.7-2 A C library that parses RDF/XML/N-Triples into RDF triples extra/xcb-proto 1.7.1-1 XML-XCB protocol descriptions

sl1pkn07 commented on 2012-05-07 07:02 (UTC)

i don't have install pyxml. and build chromium-dev sucessfull in all version sl1pkn07@sL1pkn07 ~ $ python2 -c "import xml.parsers.expat" sl1pkn07@sL1pkn07 ~ $ my yaourt xml output http://paste.kde.org/470684 anyone can confirm this?

fishburn commented on 2012-05-07 00:02 (UTC)

@sl1pkn07 build.gyp would fail with a python error regarding libexpat.so (from expat.py) Without pyxml installed, attempting to load expat would produce an error (even though expat was installed): [name@host ~]$ python2 -c "import xml.parsers.expat" Traceback (most recent call last): File "<string>", line 1, in <module> File "/usr/lib/python2.7/xml/parsers/expat.py", line 4, in <module> from pyexpat import * ImportError: /usr/lib/python2.7/lib-dynload/pyexpat.so: undefined symbol: XML_SetHashSalt After installing pyxml, this command no longer produced an error and chromium-dev would build successfully. My machine configuration isn't very exotic so it would be strange if I am the only one with this dependency.

sl1pkn07 commented on 2012-05-06 12:40 (UTC)

@cornholio are you sure? sl1pkn07@sL1pkn07 ~ $ yaourt pyxml 1 extra/pyxml 0.8.4-9 Python XML parsing library

misc commented on 2012-05-06 09:28 (UTC)

Haven't had any crashes anymore again with 1027. Might be worth updating to that (or 1028 that's out in one/two hours) for those who have issues with 1123.4.

fishburn commented on 2012-05-06 01:01 (UTC)

pyxml is a make dependency

misc commented on 2012-05-05 10:19 (UTC)

I'm getting some seemingly random crashes since 1024 or so, too. Doubt they're related to ICU though, for a few builds after the patch Chromium was entirely stable again.

sl1pkn07 commented on 2012-05-04 16:30 (UTC)

@imhumn fixed. thanks for report!

wuffleton commented on 2012-05-04 14:44 (UTC)

Looks like that icu patch didn't solve the crashes after all. While my last build seemed to crash maybe once a day tops, the new version seems to be crashing just as much as before: May 03 18:25:31 Night kernel: Chrome_HistoryT[11640] general protection ip:7f6a891cbaa7 sp:7f6a6eff2090 error:0 in chromium[7f6a87bae000+3b05000] May 03 18:29:03 Night kernel: Chrome_HistoryT[15473] general protection ip:7fe03147baa7 sp:7fe0172a2090 error:0 in chromium[7fe02fe5e000+3b05000] May 03 18:29:53 Night kernel: Chrome_HistoryT[16084] general protection ip:7fcd88f12aa7 sp:7fcd6ed39290 error:0 in chromium[7fcd878f5000+3b05000]

commented on 2012-05-04 14:28 (UTC)

The build fails because one of the build steps calls the svnversion command, which provided by the subversion package. Please update buildtime dependencies accordingly!

sl1pkn07 commented on 2012-05-03 22:09 (UTC)

er..... ops

commented on 2012-05-03 21:20 (UTC)

The PKGBUILD uploaded now has use_system_icu=1 but the "\! -path 'third_party/icu/*' \" is not removed. Did you manually do that before you built with the patched ICU? Or is it still broken despite?

sl1pkn07 commented on 2012-05-03 21:13 (UTC)

OPS! :S for me still broken :S

sl1pkn07 commented on 2012-05-03 20:55 (UTC)

@darksnipper SUCESS! don't crash! thanks man! update to 20.0.1123.2 and change pepper_19 to pepper_20 please. use ICU build by @darksniper (can you upload to AUR?) greetings

wuffleton commented on 2012-05-03 19:27 (UTC)

@sl1pkn07: Yeah, I just downloaded icu from the ABS and put the patch above ./configure like misc said (it needed -Np4 not -Np1 though). Here's the makepkg source archive I used for icu if you need it: http://dl.dropbox.com/u/6464618/icu-49.1.1-1.src.tar.gz For the chromium-dev pkgbuild, I set use_system_icu=1, added icu to the depends array and removed the line "\! -path 'third_party/icu/*' \" and it built just fine.

sl1pkn07 commented on 2012-05-03 19:15 (UTC)

hi. the patch provide by @misc not work in 20.0.1123.2 (don't apply patch) without system ICU (use_system_icu=0 and withouth \! -path 'third_party/icu/*' \) i test download icu 4.9.1.1 from sources (http://download.icu-project.org/files/icu4c/49.1.1/icu4c-49_1_1-src.tgz) and put on third_party folder (with patch by @misc)... don't build: http://paste.kde.org/468140 howto get fix @darksnipper? download extra ICU from ABS (or pkgbuild) and build with patch? (and build chromium-dev with use_system_icu=1) greetings: sorry the delay, yesterday my pc exploded

wuffleton commented on 2012-05-01 20:12 (UTC)

I can confirm that the icu patch fixes the crashes with icu 49.1.1 and chromium 20.0.1120.0. Even after heavily abusing the omnibox for a good 5 minutes, I could not get it to crash :D

wuffleton commented on 2012-05-01 19:17 (UTC)

Yep, that's definitely the problem. My ldd output shows no links to libicu since I built chromium using a modified version of the current PKGBUILD. I've made the appropriate changes to the PKGBUILD and am recompiling chromium now. I'll test it and post back with my results when it's done :)

misc commented on 2012-05-01 19:09 (UTC)

Forgot that patch files can be used to revert them, too ("-R" instead of "-N"). :) So I just tested 50-svn with said patch reverted, and indeed it crashed again. Which also means that if 50-svn or a patched 49.1.1 still crashes for you, it's a different bug — in which case you should of course submit your own ticket.

misc commented on 2012-05-01 18:58 (UTC)

> Do you think that a recompile of chromium against the new ICU would help? It should not; if Chromium links dynamically to ICU (check with eg "ldd /usr/lib/chromium-dev/chromium | grep libicu" — I get three links to /usr/lib/libicu[…].so.50, whereas you'd get *.49, of course) all that patch ought to have changed would be some of those files in /usr/lib. > would this prevent the patched ICU from being used by chromium? Indeed it does. I'm afraid that if you have not changed back "use_system_icu=0" to 1 and removed that above "\! -path 'third_party/icu/*' \" line, your Chromium compiles most likely linked to the ICU version shipped with the Chromium tarball (some past version, 4.5 or so I think). Just use ldd again to check — those three links would probably not show at all in that case, or perhaps be a self-reference to chromium itself.

wuffleton commented on 2012-05-01 17:33 (UTC)

@misc: That patch applied cleanly to 49.1.1's source and the compile was successful, however, I can still replicate the crash in the 20.0.1120.0 package I built earlier. Do you think that a recompile of chromium against the new ICU would help? (Also, I noticed that the PKGBUILD configures chromium to not use the system's ICU, would this prevent the patched ICU from being used by chromium?)

misc commented on 2012-05-01 16:48 (UTC)

pedberg suggests that the following fix is what solved the crashes for me with ICU 50-svn, and could readily be applied to 49.1.1 as well: http://pastebin.com/raw.php?i=FyTd1ea4 Just add it as eg "patch -Np1 -i <path>" before ./configure in build() of ICU's PKGBUILD. If Chromium has been compiled with 49.1.1 then there should be no need to recompile it — or the other packages — to test (just close the programs or better restart, of course).

wuffleton commented on 2012-04-30 08:02 (UTC)

Looks like it was just stubborn with its crashes before. I'm experiencing the exact same issues as before with the normal build.

wuffleton commented on 2012-04-30 03:32 (UTC)

Derp. I just realized that leaving Ccache enabled while bug testing is a terrible idea. I built icu-49.1.1-1 and chromium-dev 20.0.1120.0 with debug symbols and ccache disabled with the intent of getting valgrind/gdb traces, however I can't reproduce the crash anymore. I was unable to get GDB to start chromium, but the crash I was able to cause through valgrind doesn't look related. (I was using a script to spam the omnibox rather than just typing random stuff into it as I did before. Here's my trace if anyone else can make sense of it: http://pastebin.com/0REr7xRX ) I'm going to try clearing CCache and compiling the normal version of chromium-dev to confirm that it doesn't exhibit the bug under normal use.

misc commented on 2012-04-29 17:04 (UTC)

The crashes of ICU 49.1.1 are indeed gone for me with ICU 50-svn (C20.0.1122.0). Note that if you install the latter you'll also have to recompile all packages that depend on it for those (and those that in turn depend on them, etc.) to work again. Again, additional valgrind & gdb traces will help to locate and properly fix the bug(s?).

misc commented on 2012-04-29 13:20 (UTC)

@DarkSniper: Thanks for testing. Maybe it's an ICU 49 bug then, after all — I'll test it again next time. @all: There's a ticket for it on ICU's bugtracker; the dev has asked for valgrind traces with ICU compiled for debug (see eg the Arch Wiki page "Debug - Getting Traces" for instructions): http://bugs.icu-project.org/trac/ticket/9276

wuffleton commented on 2012-04-29 04:39 (UTC)

@misc: I've compiled version 20.0.1120.0 against those package versions and it still experiences the same segfault as before on my end.

misc commented on 2012-04-28 14:43 (UTC)

Note for my previous comment: Preferably with extra's icu-49.1.1-1, gcc 4.7.0-5 and chromium-dev 20.0.1120.0. Rationale: Given that it works now for me without any apparently relevant code change on ICU's side, I'm trying to figure out whether this was actually a bug actually in eg a prior GCC 4.7 snapshot or Chromium itself.

misc commented on 2012-04-28 13:46 (UTC)

Not getting anymore crashes with icu (svn). Can someone confirm, please?

sl1pkn07 commented on 2012-04-26 17:43 (UTC)

ow. sorry :S very thanks

wuffleton commented on 2012-04-26 17:12 (UTC)

Just a note on the force flags sl1pkno7 mentioned: I'd recommend that you use only -Sf/-Uf if recompiling the package is that inconvenient for you. It will override the file ownership check that would prevent you from upgrading otherwise, though using -dd (skipping the dependency checks) is dangerous and unnecessary in this situation.

sl1pkn07 commented on 2012-04-26 10:30 (UTC)

oks. now fixed in pkgbuild upload missing resources (not need rebuild again the package) http://dl.dropbox.com/u/6596386/theme_resources_standard.pak http://dl.dropbox.com/u/6596386/ui_resources_standard.pak put in /usr/lib/chromium-dev in the next update. force instalation by -Sfdd (or rebuild now with the changes) greetings

misc commented on 2012-04-26 10:06 (UTC)

Hurr. I should have looked at the PKGBUILD first: Line 306 just doesn't copy those two new files. Once amended to install -Dm644 out/Release/{chrome,resources,content_resources,ui_resources_standard,theme_resources_standard}.pak "${_chromium_home}"/ all's well again. (Except for that unrelated memory corruption in "third_party/tcmalloc/chromium/src/free_list.cc:117" on exit.)

misc commented on 2012-04-25 20:27 (UTC)

Faulty themes started with 1115.0; there are a couple of commits concerned with them between it and 1114.0. Upon start (and somewhere before finishing initialization) Chromium now also complains about not finding some resource files in /usr/lib/chromium-dev.

sl1pkn07 commented on 2012-04-25 17:23 (UTC)

is bug. i think is related by our friend gcc 4.7 http://sl1pkn07.no-ip.com/looooooooooooooooooooooooonnnnnnnnnnnnnnngggggggggggglog.txt

yetAnotherZero commented on 2012-04-25 16:48 (UTC)

@imhumn I got the same problem.

commented on 2012-04-25 16:15 (UTC)

Giant angry red bar at the top of the screen, lack of icons, and this error message: (chromium:1528): GdkPixbuf-WARNING **: GdkPixbufLoader finalized without calling gdk_pixbuf_loader_close() - this is not allowed. You must explicitly end the data stream to the loader before dropping the last reference. Feature? or bug? hahaha... Rebuilding now to see if this was just a bad build on my system, or if it's Chromium's fault. May wanna hold off on upgrading in the meantime.

yetAnotherZero commented on 2012-04-21 20:06 (UTC)

@DarkSniper Thanks for that. I recompiled it and it's awesome. I just did a pacman -Rdd gcc gcc-libs libtool lib32-gcc-libs then pacman -U on all those pkg versions you mentioned below. After the compile finished I did a pacman -Syu to get my system back up to par Thanks again

wuffleton commented on 2012-04-20 18:12 (UTC)

@speed: I compiled it in a chroot using GCC4.6 with the Clang option and the crashes are gone for me. People reported that it doesn't crash if compiled under GCC 4.6, but downgrading GCC on your working install could potentially cause catastrophic breakage. A chroot seems to be the best (and only) existing workaround at this point. I followed this guide for the chroot creation: https://wiki.archlinux.org/index.php/DeveloperWiki:Building_in_a_Clean_Chroot (Beware: If you're using BTRFS, it makes a subvolume for the chroot, so you have to use btrfs sub delete if you wish to remove the chroot, as rm -rf won't work) Here's my package list for the chroot, which has everything you need to compile chromium (replace the base base-devel sudo part of the mkarchroot command with this): http://pastebin.com/vj6NL75r Get version 4.6.3-1 of core/gcc, core/gcc-libs, and multilib/lib32-gcc-libs (x86_64 only) and version 2.4.2-4 of core/libtool from http://schlunix.org/archlinux/ and install them into your chroot with "pacman -r /path/to/your/chroot/root -U <paths to downloaded packages here>" Before running mkchrootpkg, you'll probably need this edited PKGBUILD (It's up to you whether to use clang or not): http://pastebin.com/D7W50u6R The '"${srcdir}"/../' bits in package() don't work with mkchrootpkg because it doesn't store the source files one level above $srcdir. I've changed instances of this to '"${srcdir}"/', since install seems to treat the symlinks as normal files. I have not tested this yet (I copied the source files to $srcdir/../ manually), but it should work as intended. Hope this helps at least a few other users until the chromium team fixes GCC 4.7 support!

yetAnotherZero commented on 2012-04-20 17:21 (UTC)

Anyone have feedback regarding the omnibar crashes? It's a little sad everytime I forget and try to use it...

wuffleton commented on 2012-04-12 08:53 (UTC)

Using those options didn't seem to fix the segfault. After putting the omnibox through some abuse, it still crashes. I'm getting something slightly different on the command line now, though the dmesg output seems to be very similar if not identical to the previous segfault. From Command Line: [2695:2695:120527473823:ERROR:content_settings_pref_provider.cc(493)] Invalid pattern strings: ,* [2695:2695:120527475053:ERROR:content_settings_pref_provider.cc(363)] Invalid pattern strings: ,* [2695:2730:120535207943:ERROR:diagnostic_error_delegate.h(31)] sqlite error 11, errno 0: database disk image is malformed (This one doesn't always occur) Segmentation fault (core dumped) From dmesg: [120553.729454] Chrome_HistoryT[2730] general protection ip:7f1df11d6247 sp:7f1dd88722a0 error:0 in chromium[7f1defabb000+4524000]

sl1pkn07 commented on 2012-04-12 06:29 (UTC)

in my case. with ICU 4.9 from testing (with use_system_icu=1) Program received signal SIGSEGV, Segmentation fault. 0x00007ffff3f16354 in icu_49::RegexMatcher::reset(icu_49::UnicodeString const&) () from /usr/lib/libicui18n.so.49

wuffleton commented on 2012-04-12 05:54 (UTC)

@ sl1pkn07: Awesome! Compiling now to see if it fixes the issue.

sl1pkn07 commented on 2012-04-12 04:52 (UTC)

i think found problem described by @ DarkSniper (segfaults in omnibox) please test with: use_system_icu=0 and add "\! -path 'third_party/icu/*' \ in" in remove unnecesary components greetings

jeagoss commented on 2012-04-12 02:28 (UTC)

The patches still need testing to make sure I got everything. So please test away.

sl1pkn07 commented on 2012-04-12 02:13 (UTC)

@jeagoss http://i.imgur.com/uIMfV.png

jeagoss commented on 2012-04-12 01:23 (UTC)

I have a working PKGBUILD and patches to make chromium compile with pulseaudio. I also commented on your bug thread with the patch. You can download the patches and PKGBUILD from here. http://archrepo.jeago.com/sources/chromium-dev/

commented on 2012-04-11 14:52 (UTC)

Thanks, as always for the updates! One suggested change to the PKGBUILD: if building with clang and you use plugins, both from and not from the Web Store, clang_use_chrome_plugins=1 will also have to be added to the GYP_DEFINES variable. Was using the clang build earlier, and this was necessary to see my plugins in-browser. AFAIK, this shouldn't cause any problems if you don't use plugins. See [https://code.google.com/p/chromium/wiki/Clang] for more informations.

sl1pkn07 commented on 2012-04-11 00:33 (UTC)

update to 20.0.1096.1 now back 64bit pepperflash!!! now use pkgbuild sugested by @nagisa now pepperflash 32 and 64 bits use same paths /usr/lib/chromium-dev/Pepperflash/libpepflashplayer.so notes for @nagisa - In /etc/makepkg.conf #-- Make Flags: change this for DistCC/SMP systems MAKEFLAGS="-j8" no "need" fast build option. only uncomment and edit makepkg.conf (accoirding you cores) to enable multicore compilation and ${MAKEFLAGS} dont work. use $MAKEFLAGS instead - The disable pulseaudio is for http://crbug.com/119903. i dont know the fail build is related by http://crbug.com/111392 patch. (still fail build with/out patches) - All ffmpeg system libs/ffmpegsumo build is according by http://sources.gentoo.org/cgi-bin/viewvc.cgi/gentoo-x86/www-client/chromium/ (make by phajdan.jr , Chromium developer) i comment all build with ffmpeg system libs for use in the future. active now don't have changes in perfomance. - all no common variables warned by namcap. change to: "_<variable>" any other suggestion are welcome greetings and PD: really sorry my english :S

nagisa commented on 2012-04-06 19:37 (UTC)

I made (didn't test it too much tough) new PKGBUILD based on your PKGBUILD. Maybe you want to take some parts from it or even take whole PKGBUILD. https://gist.github.com/7df6bd01b6e5f5d8bc15

commented on 2012-04-06 14:43 (UTC)

That switch didn't work for me when Pepper Flash first hit this PKGBUILD. I had to point Chromium directly to the file with another switch. It's possible that's a Windows switch, not sure.

misc commented on 2012-04-06 07:38 (UTC)

Well, the only change that commit actually requires (assuming, again, that it's really the only one responsible) is to start Chromium with --enable-bundled-ppapi-flash .

misc commented on 2012-04-06 07:23 (UTC)

Couldn't you just include a patch that reverts http://codereview.chromium.org/9969131/diff/1/chrome_content_client.cc ? Seems like that's the only change responsible for disabling it. Haven't tested it though since I don't use pepper yet. No idea why they did that (and only on 1084), either.

sl1pkn07 commented on 2012-04-06 01:23 (UTC)

http://dl.dropbox.com/u/6596386/google-chrome-unstable-19.0.1084.9-130290.x86_64.rpm for older flash 64

sl1pkn07 commented on 2012-04-06 01:18 (UTC)

update to 19.0.1084.15 DONT HAVE PEPPERFLASH ON 64BITS BRRRRRRRRRRRRRRRRRR!!!!!!!!!! òó see http://build.chromium.org/f/chromium/perf/dashboard/ui/changelog.html?url=/branches/1084/src&range=130290:130829&mode=html

jeagoss commented on 2012-04-05 22:25 (UTC)

It appears in the latest release that google has renamed the flash plugin for the 32bit version. So there is no longer a need to have a 32bit/64bit check. The flash plugin is named exactly the same and located in the same place as the 64bit version.

sl1pkn07 commented on 2012-04-05 20:53 (UTC)

yes, i have this issue. but i dont know is related :S i think related by GCC 4.7 (with gcc 4.6.x don't appear) , but dont have idea fix it. in compialation don't show output error

wuffleton commented on 2012-04-05 20:26 (UTC)

This version has been segfaulting occasionally for me when typing something in the omnibox. Has anyone else been experiencing similar behavior? The output seems to indicate something related to history, so I tried clearing my history, but it didn't seem to have an effect, as it's still segfaulting. Here's the relevant log output from my Systemd journal: Apr 05 13:19:48 Night kernel: Chrome_HistoryT[7870] general protection ip:7fc64046e397 sp:7fc6269e42a0 error:0 in chromium[7fc63eea6000+39a5000] Apr 05 13:19:48 Night[20497]: Process 7835 (Chrome_HistoryT) dumped core.

commented on 2012-04-04 08:20 (UTC)

Tested again. Works fine. My fault, i guess. Sorry for inconvenience.

sl1pkn07 commented on 2012-03-29 14:38 (UTC)

testing. no fail for me (gcc 4.6.3 and disable NaCL) http://dl.dropbox.com/u/6596386/log greetings

commented on 2012-03-29 10:34 (UTC)

Latest build without NaCL fails with: Updating projects from gyp files... Traceback (most recent call last): File "build/gyp_chromium", line 171, in <module> sys.exit(gyp.main(args)) File "/tmp/yaourt-tmp-dm/aur-chromium-dev/src/chromium-build/tools/gyp/pylib/gyp/__init__.py", line 471, in main options.circular_check) File "/tmp/yaourt-tmp-dm/aur-chromium-dev/src/chromium-build/tools/gyp/pylib/gyp/__init__.py", line 111, in Load depth, generator_input_info, check, circular_check) File "/tmp/yaourt-tmp-dm/aur-chromium-dev/src/chromium-build/tools/gyp/pylib/gyp/input.py", line 2292, in Load depth, check) File "/tmp/yaourt-tmp-dm/aur-chromium-dev/src/chromium-build/tools/gyp/pylib/gyp/input.py", line 433, in LoadTargetBuildFile includes, depth, check) File "/tmp/yaourt-tmp-dm/aur-chromium-dev/src/chromium-build/tools/gyp/pylib/gyp/input.py", line 433, in LoadTargetBuildFile includes, depth, check) File "/tmp/yaourt-tmp-dm/aur-chromium-dev/src/chromium-build/tools/gyp/pylib/gyp/input.py", line 433, in LoadTargetBuildFile includes, depth, check) File "/tmp/yaourt-tmp-dm/aur-chromium-dev/src/chromium-build/tools/gyp/pylib/gyp/input.py", line 414, in LoadTargetBuildFile "Unable to find targets in build file %s" % build_file_path Exception: Unable to find targets in build file ppapi/native_client/native_client.gyp while loading dependencies of chrome/chrome.gyp while loading dependencies of chrome/browser/sync/tools/sync_tools.gyp while loading dependencies of build/all.gyp while trying to load build/all.gyp

sl1pkn07 commented on 2012-03-27 11:55 (UTC)

oks. fix and update

misc commented on 2012-03-27 11:23 (UTC)

Looks like pastebin cut off the last two lines as they are "empty", which has patch complain about syntax (not an error though). So what's missing after "class SkBitmap;": One line with one space, then another line that's entirely empty.

sl1pkn07 commented on 2012-03-27 11:08 (UTC)

@misc (Stripping trailing CRs from patch.) patching file ui/gfx/codec/jpeg_codec.h patch unexpectedly ends in middle of line Hunk #1 succeeded at 8 with fuzz 2 (offset 1 line).

sl1pkn07 commented on 2012-03-27 06:53 (UTC)

try now

wuffleton commented on 2012-03-27 01:42 (UTC)

nacl_sdk.zip seems to have a bad checksum. I'm repeatedly getting 6819b50db888bb4c3c215c3c96f32eff431b6024 but the PKGBUILD expects e4991439b4832a5a80134a452fb824992fc9818d. Changing the checksum doesn't seem to cause any problems.

sl1pkn07 commented on 2012-03-26 23:48 (UTC)

ains.... update - disable (temporally, even fix build) pulseaudio support http://crbug.com/119903 - now use sha1sums instead md5sums. to work for script for download automatically official chrome RPM for libpdf/pepperflash plugins (search official yum repository (http://dl.google.com/linux/chrome/rpm/stable/"${_arch}"/repodata/filelists.xml.gz) and download RPM according this) - add support (not tested) to GCC 4.7.0 (testing/multilib repository) (patch provided by @misc). please test it. (only apply if detect GCC 4.7.0 has been installed) please report more bugs, fix, patches, ideas, encouragement, spit, blablabla greetings

Det commented on 2012-03-26 20:59 (UTC)

But so pulseaudio still needs to be disabled?

misc commented on 2012-03-25 15:07 (UTC)

Well that wasn't hard to fix. Here's my patch to have Chromium (tested with 1079.0 & 1080.0) compile with GCC 4.7 — just a bunch of missing includes (those adding cstddef are those that still apply from extra's chromium package to solve warnings for GCC 4.6; aren't necessary but don't hurt either): http://pastebin.com/CQu4GwC9

misc commented on 2012-03-24 16:15 (UTC)

1078.0 built fine for me with pulseaudio disabled, with GCC 4.7 however I'm getting prototype and missing declaration errors (for 1079.0).

sl1pkn07 commented on 2012-03-24 02:56 (UTC)

fail build 19.0.1077.3 http://crbug.com/119903

commented on 2012-03-20 15:55 (UTC)

Just wanted to put it out there that Arch's default CFLAGS appear to work just fine, even when compiling Nacl, Pepper Flash, and the libpdf extras. There should be no need to remove the stack-smashing flags at the moment, AFAIK. There may be a few outstanding quirks regarding using the --enable-pulseaudio switch for 64-bit users using PulseAudio and Pepper Flash together, but I'm looking into it.

nagisa commented on 2012-03-16 22:14 (UTC)

Building without _system_ffmpeg_libs is OK, however with them (without patching, because it rejects) I get out/Release/obj.target/ffmpeg/geni/ffmpeg_stubs.cc:19:10: fatal error: 'libavformat/url.h' file not found #include <libavformat/url.h> For now I will bear with builing ffmpeg_sumo, but it would be great to see system_ffmpeg_libs working again soon.

sl1pkn07 commented on 2012-03-16 18:41 (UTC)

@Nagisa oks. now disable (with comment) all references to build/install system FFmpeg library.

nagisa commented on 2012-03-16 18:27 (UTC)

You should also add a note near "compile with system ffmpeg" option that compiling with system ffmpeg currently fails and you can remove patch_for_ffmpeg-git.patch as it doesn't work anyway.

sl1pkn07 commented on 2012-03-16 18:23 (UTC)

:D thanks for this report! (need more of this). updated and added lib32-gcc-libs makedepend (if nacl is acctived) greetings and little cosmetics

commented on 2012-03-16 16:01 (UTC)

Great work on getting all of this PKGBUILD fun worked out! I've been until now unable to compile Nacl, even with safe CFLAGS (took out the stack-smashing protection flags manually). Was receiving fatal errors that some gcc library files were missing. lib32-gcc-libs needs to be a dependency for 64-bit users if building Nacl, it seems.

sl1pkn07 commented on 2012-03-16 15:23 (UTC)

do'h fixed

jeagoss commented on 2012-03-16 12:56 (UTC)

On line 344 of the PKGBUILD, you need to change the first ${pkgdir} to ${srcdir}. This change allows for full compilation on i686. OLD: install -m775 "${pkgdir}"/opt/google/chrome/libgcflashplayer.so "${pkgdir}"/usr/lib/chromium-dev/libgcflashplayer.so NEW: install -m775 "${srcdir}"/opt/google/chrome/libgcflashplayer.so "${pkgdir}"/usr/lib/chromium-dev/libgcflashplayer.so

sl1pkn07 commented on 2012-03-16 12:19 (UTC)

@jeagoss try now (update pkgbuild) in 32bits, the PapperFlash library is named "libgcflashplayer.so". now install in /usr/lib/chromium-dev/libgcflashplayer.so (if detect 32bits) (like a RPM package) @imhumn edit /etc/chromium-dev/default and add "--ppapi-flash-path=/usr/lib/chromium-dev/PepperFlash/libpepflashplayer.so --ppapi-flash-version=11.2.31.114" in CHROMIUM_FLAGS (--enable-greasemonkey --enable-sync not need) i think fixed now in chromium-dev.install script. @all new pkgbuild enable compile "chromiumdrive" (chromedrive). for new major updates include it for default please test it greetings

jeagoss commented on 2012-03-16 10:15 (UTC)

It appears that the i386 versions of chrome do not actually include the pepperflash directory opt/google/chrome/PepperFlash in the rpm archive. I also checked the i386 versions of the deb's that are available, and the same seems to be true. Due to i386 not including the plugin, it causes the build to fail. I'm not sure if google is maintaining an i386 version of the plugin or not. I've been trying to figure this out. Until you can find a reliable source for the plugin, you might consider disabling the install on i686 architectures.

commented on 2012-03-16 02:35 (UTC)

Nope. The CHROMIUM_FLAGS entry in /etc/chromium-dev/default is empty.

sl1pkn07 commented on 2012-03-16 01:28 (UTC)

#imhumn please. in /etc/chromium-dev/default exist this line?: CHROMIUM_FLAGS=" --ppapi-flash-path=/usr/lib/chromium-dev/PepperFlash/libpepflashplayer.so --ppapi-flash-version=11.2.31.114 --enable-greasemonkey --enable-sync" greetings

commented on 2012-03-16 01:07 (UTC)

@sl1pkn07: Thanks! Pepper Flash is working just fine using the --ppapi-flash-path="/path/to/plugin" command-line switch.

sl1pkn07 commented on 2012-03-16 00:43 (UTC)

#imhumm fixed (but not affected. typo present in all scripts)

acfrazier commented on 2012-03-15 18:52 (UTC)

curl: (22) The requested URL returned error: 404 ==> ERROR: Failure while downloading google-chrome-unstable-19.0.1068.0-126342.x86_64.rpm I'm assuming this is due to the fact that we're now on 19.0.1068.1, so I'll wait for an update.

commented on 2012-03-15 15:33 (UTC)

@sl1pkn07: Minor typo in your sed command around the end of the PKGBUILD, where the PKGBUILD-internal variable is written _use_papperflash instead of 'pepperflash'. Not sure if that's gonna cause issues in the installation of the Pepper plugin. I can't find it in the browser, despite the file existing, and the --enable-bundled-ppapi-flash switch being on. Does the Pepper Flash plugin require Nacl perhaps? I'd rebuild to test this, but I'm getting a 404 Not Found on the .rpm download after having successfully built your new PKGBUILD once. URL change?

sl1pkn07 commented on 2012-03-15 06:09 (UTC)

upadate to 19.0.1068.0 add new options: - Add "_use_libpdf" option to install (enable by default) "libpdf.so" from Official Chrome-Browser unstable (RPM) - Add "_use_pepperflash"option to install (enable by default) "PepperFlash" (libpepflashplayer.so) from Official Chrome-Browser unstable (RPM) need new dependence: openssl098 - Add support custom flags in /etc/chromium-dev/default (like chromium PPA packages from Ubuntu .deb) (is need to install PepperFlash plugin) please. test it greetings

sl1pkn07 commented on 2012-03-11 18:19 (UTC)

fixed http://i.imgur.com/rn8Zh.png

Det commented on 2012-03-11 17:41 (UTC)

By the way, there's a couple of typos in the PKGBUILD. It says "Enviorement" instead of "Environment" in the "Automatic Build Options" and the Gnome/Pulseaudio (sub)sections.

sl1pkn07 commented on 2012-03-09 01:06 (UTC)

@nagisa, the ffmpeg patch is provide by @misc. please ask him if you have working patch. please send me greetings

nagisa commented on 2012-03-08 15:37 (UTC)

In addition to yet unfixed problem reported in 24th February there's another problem now which is related with patches. There's no `third_party/ffmpeg/avcodec-53.sigs` anymore and the file is in `third_party/ffmpeg/chromium/avcodec-54.sigs` now. Also trying to patch this file simply fails with yet another reject. As with previous case, failing patch is called `patch_for_ffmpeg-git.patch`

nagisa commented on 2012-02-24 15:53 (UTC)

By the way reject looks like this: https://gist.github.com/9db1661d4b1474dbb9dc

nagisa commented on 2012-02-24 15:37 (UTC)

I'm getting patch unexpectedly ends in middle of line patch unexpectedly ends in middle of line 1 out of 1 hunk FAILED -- saving rejects to file ui/base/l10n/l10n_util.cc.rej when patching for ffmpeg-git. Also, is there a way to get rid of that nasty lib32-zlib dependency and only use lib-zlib, when compiling with NaCl? @sl1pkn07, I have been using this nickname way before I watched Clannad, so... not. Thanks for update.

sl1pkn07 commented on 2012-02-24 02:55 (UTC)

update to 19.0.1049.3 (sorry for the delay :S) - Add changes suggested by @Nagisa (Dango Daikazoku?) - Add third_party/adobe/ in remove unnecesary components exception - Delete chromium-revert-jpeg-swizzle-r2.patch. no need anymore (fixed by libjpeg-turbo 1.2.0) greetings

nagisa commented on 2012-02-16 14:25 (UTC)

To better integrate with Gnome Shell and provided .desktop file you should change line exec /usr/lib/chromium-dev/chromium "$@" to exec /usr/lib/chromium-dev/chromium --class=chromium-dev "$@" in chromium-dev.sh Here's new file https://gist.github.com/a6def8dfd86fc9ac4a6f and it's md5 is 48e11fa61360ec851c375e6deede11e2

sl1pkn07 commented on 2012-02-12 00:30 (UTC)

update to 19.0.1036.7 - remove some patches - add GYP_DEFINES linux_use_gold_binary=0 linux_use_gold_flags=0 greetings

misc commented on 2012-02-11 17:49 (UTC)

Finally — after "python2 build/gyp_chromium -f make --depth=. build/all.gyp": sed -i 's/LINK.target ?= $(LINK)/LINK.target ?= $(LINK) -fuse-linker-plugin/g' Makefile sed -i 's/LINK.host ?= g++/LINK.host ?= g++ -fuse-linker-plugin/g' Makefile I'm currently compiling .1039; if it fails I'll add another comment.

misc commented on 2012-02-10 12:28 (UTC)

Gold accessible only as ld.gold, not vanilla ld, so one needs two plugin parameters for gcc. However, reading build/common.gypi makes it clear that the script searches for its own gold in third_party, so I excluded it with third_party/gold/* from removal — to no avail, linking still fails with unrecognized option. Adding "-B/usr/bin/ld.gold" to CFLAGS & CXXFLAGS before calling make didn't solve it, either.

misc commented on 2012-02-10 10:48 (UTC)

@Miche: No idea what's included in that 10GB figure — perhaps all revisions of the source files that ever existed — but as for the tarballs, they only contain one. Thus the ball (~190MB) and its extracted contents (~900MB) take up ~ 1GB on my /tmp, which has up to 2GB allocated of my 4GB RAM. So far compilation has worked hassle-free — but only if no other memory intensive programs run. @sl1pkn07: The glib2 patch appears superfluous as of 1036; as for the ffmpeg one, I doubt it's enough for the current git. However, there's another issue now, at least on my machine: Compilation aborts with "/usr/bin/ld: unrecognized option '--icf=none'". Identical code folding is a feature added by (and afaik only available to) Google's gold linker, which ought to be accessible since Arch's binutils are compiled with the --enable-gold parameter — so no idea why it fails.

commented on 2012-02-07 11:06 (UTC)

Chromium wiki says "It takes about 10GB or so of disk space to check out and build the source tree. This number grows over time." I tried twice and the process end with a "no space left on the device" error. The second time i set the build dir in my home (yaourt -S chromium-dev --tmp ~/tmp) but didn't worked anyway. Did you got this issue? Where do you build it?

acfrazier commented on 2012-02-04 02:57 (UTC)

Thank you, I appreciate your work on this package.

sl1pkn07 commented on 2012-02-03 20:30 (UTC)

this morning release 1025.2 (yes, i sure) now look new release http://feedproxy.google.com/~r/GoogleChromeReleases/~3/kvVrvuQCnDQ/dev-channel-update.html updating in progress

acfrazier commented on 2012-02-03 19:49 (UTC)

http://omahaproxy.appspot.com/viewer Latest Linux dev release is 18.0.1025.3 I'm not sure if it builds with the current pkgbuild, testing now..

misc commented on 2012-01-30 22:04 (UTC)

1020< needs \! -path 'third_party/libsrtp/*' \ in "Remove unnecesary components". Btw, my patch is actually two in one, the last two changes are for glib2 >= 2.31 but should be downwards compatible.

sl1pkn07 commented on 2012-01-28 20:18 (UTC)

apply patch prvide in http://crbug.com/111392 now compile upload and updane new version add patch provided by @misc (only active if use system ffmpeg libs) greetings

acfrazier commented on 2012-01-25 23:03 (UTC)

It built just fine on my machine, I don't know why you are seeing that error.

sl1pkn07 commented on 2012-01-25 22:30 (UTC)

nacl-g++ /home/sl1pkn07/aplicaciones/chromium-dev/src/chromium-build/out/Release/obj.target/ppruntime_lib/geni/newlib-x86-64/ppruntime_lib/plugin_ppb_audio.o In file included from ./base/memory/weak_ptr.h:63:0, from ./media/audio/pulse/pulse_output.h:25, from media/audio/pulse/pulse_output.cc:5: ./base/logging.h: In function 'std::string* logging::CheckEQImpl(const t1&, const t2&, const char*) [with t1 = scoped_refptr<base::MessageLoopProxy>, t2 = MessageLoop*, std::string = std::basic_string<char>]': media/audio/pulse/pulse_output.cc:129:3: instantiated from here ./base/logging.h:556:1: error: no match for 'operator==' in 'v1 == v2' ./base/logging.h:556:1: note: candidates are: ./base/logging.h:556:1: note: operator==(MessageLoop*, MessageLoop*) <built-in> ./base/logging.h:556:1: note: no known conversion for argument 1 from 'const scoped_refptr<base::MessageLoopProxy>' to 'MessageLoop*' ./base/logging.h:556:1: note: operator==(base::MessageLoopProxy*, base::MessageLoopProxy*) <built-in> ./base/logging.h:556:1: note: no known conversion for argument 2 from 'MessageLoop* const' to 'base::MessageLoopProxy*' ./base/memory/scoped_ptr.h:203:6: note: template<class C> bool operator==(C*, const scoped_ptr<C>&) ./base/memory/scoped_ptr.h:315:6: note: template<class C> bool operator==(C*, const scoped_array<C>&) ./base/memory/scoped_ptr.h:441:6: note: template<class C, class FP> bool operator==(C*, const scoped_ptr_malloc<C, FP>&) CXX(target) out/Release/obj.target/media/media/base/byte_queue.o ./base/logging.h:556:1: warning: control reaches end of non-void function [-Wreturn-type] make: *** [out/Release/obj.target/media/media/audio/pulse/pulse_output.o] Error 1

acfrazier commented on 2012-01-25 19:08 (UTC)

MD5 for chromium-18.0.1017.2.tar.bz2: 186593a3f943e3d9bdf89c7899f382a3 Builds fine only changing the package version/MD5, and leaving the rest of the things as-is.

misc commented on 2012-01-24 15:58 (UTC)

My bad, I misread an announcement.

jeagoss commented on 2012-01-24 08:25 (UTC)

Why is this flagged out of date? The latest official dev release is the version stated.

misc commented on 2012-01-22 10:09 (UTC)

Here's a little patch to make it compile with ffmpeg-git & glib2 >= 2.31: http://pastebin.com/raw.php?i=ixheuye6

sl1pkn07 commented on 2012-01-19 12:43 (UTC)

oks. thanks :)

jeagoss commented on 2012-01-19 11:37 (UTC)

That looks to be the only dependency that wasn't there.

sl1pkn07 commented on 2012-01-19 11:05 (UTC)

@jeagoss please. test package with namcap and post the results (for needed add other dependencies) greetings

sl1pkn07 commented on 2012-01-19 11:01 (UTC)

roger. update pkgbuild don't up pkgrel because is the first issue by this. i supposed all user have install dbus-glib package thanks @jeagoss greetings

jeagoss commented on 2012-01-19 10:42 (UTC)

dbus-glib should be added as a dependency. I was building from a clean system, and the build fails out in the gyp stage if it is missing.

sl1pkn07 commented on 2012-01-18 19:28 (UTC)

update to 18.0.1010.1 fixed: http://crbug.com/103360 http://crbug.com/41887 add udev and icu to dependency greetings

pfreire commented on 2012-01-15 20:41 (UTC)

Thanks Det, that was it. now i have another error, but I believe it's because of the CFLAGS i'm using just like in this problem http://code.google.com/p/chromium/issues/detail?id=104569. They advise to use safe CFLAGS. FAILED with 1: ../native_client/toolchain/linux_x86_newlib/x86_64-nacl/bin/gcc -c src/untrusted/stubs/crti_x86_64.S -o /tmp/yaourt-tmp-root/aur-chromium-dev/src/chromium-build/out/Release/obj/gen/tc_newlib/lib64/crti.o -std=gnu99 -m64 -O3 -fomit-frame-pointer -mtls-use-call -DNACL_BUILD_ARCH=x86 -D__linux__ -D__STDC_LIMIT_MACROS=1 -D__STDC_FORMAT_MACROS=1 -D_GNU_SOURCE=1 -D_BSD_SOURCE=1 -D_POSIX_C_SOURCE=199506 -D_XOPEN_SOURCE=600 -DDYNAMIC_ANNOTATIONS_ENABLED=1 -DDYNAMIC_ANNOTATIONS_PREFIX=NACL_ -DNACL_BUILD_SUBARCH=64 -I/tmp/yaourt-tmp-root/aur-chromium-dev/src/chromium-build/out/Release/obj/gen/tc_newlib/include -I.. -I../ppapi -I.. -I../ppapi make: *** [out/Release/obj/gen/tc_newlib/lib64/crti.o] Error 255

Det commented on 2012-01-15 15:30 (UTC)

It's in [multilib]. Dunno why the "Native Client" (NACL) needs it.

pfreire commented on 2012-01-15 15:27 (UTC)

Hello, I get this error (using yaourt): ==> Building and installing package ==> Install or build missing dependencies for chromium-dev: error: target not found: lib32-zlib Where is lib32-zlib, and why do I need it on a x64 system?

sl1pkn07 commented on 2012-01-12 16:18 (UTC)

http://crbug.com/106954 now fixed! thanks phajdan.jr!

sl1pkn07 commented on 2012-01-11 20:18 (UTC)

compile chromium take several minutes, force compile all for a "little" revision (incrementing pkgrel number) is a "dirty trick" (sorry, my english not good) only incrementing pkgrel number if a make several changes (not include update package)

imrehg commented on 2012-01-11 15:09 (UTC)

Just two more thoughts: - chromium-revert-jpeg-swizzle-r0.patch seems to be completely commented out anyway, I guess shouldn't be included - would be great if package fixes would be noted with incremented pkgrel number, it gets confusing that I downloaded the PKGBUILD 3 or 4 times today, all worked (or not-worked) differently but nominally they are all the "same". Other then these things, thanks a lot for maintaining it! For a while I tried to keep track of all the changes that needed to be taken care of to get Chromium to build, but I had much much less patience with it. :)

sl1pkn07 commented on 2012-01-11 14:26 (UTC)

yesyes. fixed in pkgbuild :S

imrehg commented on 2012-01-11 14:23 (UTC)

Had to except "third_party/libjpeg_turbo/" in this latest build, otherwise got compilation error.

sl1pkn07 commented on 2012-01-11 12:14 (UTC)

fuck! http://crbug.com/106954 not fixed >< back GYP_Define to "use_libjpeg=0"

sl1pkn07 commented on 2012-01-11 11:58 (UTC)

update to 18.0.1003.1 fixed: http://crbug.com/104569 (no need patch anymore) http://crbug.com/108228 (fixed in 18.) @darksniper http://crbug.com/106954 (no need patch anymore) back mv instead cp -R add exception for "third_party/libyuv/" (thanks @imrehg) remove Debug code greetings

wuffleton commented on 2012-01-08 21:09 (UTC)

That was exactly the issue, built fine [without eating up loads of space] after changing that =D However, it seems that this version still contains this bug, whereas the latest stable one fixed it: http://crbug.com/108228

sl1pkn07 commented on 2012-01-07 21:33 (UTC)

try this: in build() mv instead cp -R only source code need 852MiB

wuffleton commented on 2012-01-07 16:15 (UTC)

Just wondering... Is it just me, or does building chromium now take an outrageous amount of disk space? On earlier builds I was able to build the whole thing in my 2GB tmpfs, now I can't even extract the source without running out of space. For some odd reason, the source alone it's taking up about 700 megs more on /tmp than it does on my hard drive, and I'm pretty sure that BTRFS's compression isn't responsible for such a big difference in size. I bumped /tmp's size up to 2.5 gigs and it barely got to the end of the NACL SDK thing before running out of space again. Any ideas on what might be causing this?

imrehg commented on 2012-01-02 11:22 (UTC)

For newer versions I had to add an exception for "third_party/libyuv/" in the "Remove unnecesary components" section, otherwise the build breaks. With this small change (and disabling nacl) I could successfully build 18.0.993.0 By the way, is that section really necessary? Can't the build system decide what it needs the build and what not? It took a while to track down while a build breaks with "missing header file", when it is originally there.

sl1pkn07 commented on 2011-12-28 14:50 (UTC)

install with yaourt. its more easy yaourt -Sa nameofpackage or with edit /etc/yaourtrc AURSEARCH=1 , install with simply yaourt -S nameofpackage whist this install all necesary dependencies in all packages example this: yaourt -S chromium-dev https://wiki.archlinux.org/index.php/Yaourt

darnir commented on 2011-12-28 11:27 (UTC)

Okay, my mistake. I had only made the package, not installed it. Still new to Arch, figuring my way out.. Once I ran makepkg -si, it installed libwebp, and no problem thence. Though, shouldnt that be automated when I run makepkg -si for the chromium-dev package?

sl1pkn07 commented on 2011-12-28 10:43 (UTC)

are you sure install libwebp?. try install with yaourt (yaourt -Sa libwebp)

darnir commented on 2011-12-28 08:04 (UTC)

While installing, I ran makepkg -s, and got an error: error: target not found: libwebp => Error: 'pacman' failed to install missing dependencies. So, I followed the link to the libwebp package and manually installed it. Yet I keep getting this error and cannot install the chromium-dev package.

sl1pkn07 commented on 2011-12-16 19:56 (UTC)

@oraerk in compile time. buid 64 and 32bits plugin in same time

commented on 2011-12-16 14:51 (UTC)

According to Chromium docs NACL should be supported in 64bit mode. Why use lib32-zlib?

sl1pkn07 commented on 2011-12-14 12:50 (UTC)

New PKGBUILD Update - Remove Build Options: _use_cups and _use_krb now build by default without touch any config - Now NACL suppot buid by default. disable with build option _use_nacl this is for change system build NACL (see http://projects.archlinux.org/svntogit/packages.git/tree/trunk/PKGBUILD?h=packages/chromium) thanks to @Evangelos for guide me to the light XD sorry for the any conveniences :S

sl1pkn07 commented on 2011-12-14 08:22 (UTC)

update to fix post below and remember (now fix) _use_nacl _use_cups _use_kerb ALL set to 0, to enable set to 1 (to this edit PKGBUILD)

sl1pkn07 commented on 2011-12-14 08:14 (UTC)

edit /etc/makepkg.con and uncomment MAKEFLAGS entry add -jX (X is a number of cored you use. the first core is 1 (example quad core = -j4, dual core = -j2, etc)

naelstrof commented on 2011-12-14 06:09 (UTC)

I get the following error while building chromium-dev: http://j.mp/vrlCvv

sl1pkn07 commented on 2011-12-13 22:54 (UTC)

new PKGBUILD update add in GYP_DEFINES use_system_vpx : disable (0). prepare to use libvpx system libs (don't use system lib, waiting TODO) use_system_hunspell : prepare to use hunspell system libs (upstream changes needed) use_system_ssl : prepare to use openssl system libs (http://crbug.com/58087) BUILD OPTIONS _use_kerb (kerberos) : enable/disable support to build Kerberos environment fix install NACL libs and components when _use_nacl=1 some cosmetic changes greetings

sl1pkn07 commented on 2011-12-13 18:53 (UTC)

okok. update to 17.0.963.6 changes in pkgbuild: add in GYP_DEFINES: use_system_v8=0 : prepare to use V8 system libs (when update V8 extra package to +3.7.6) use_system_sqlite=0 : prepare to use sqlite3 system libs (not working) use_system_flac=1 : use system FLAC libs (and add FLAC dependency) use_system_libwebp=1 : use system libwebp libs (and add libwebp dependency) use_system_libjpeg=0 : back config (see http://code.google.com/p/chromium/issues/detail?id=106954) ffmpeg_branding=Chrome BUILD OPTIONS _use_cups : enable support to build CUPS environment _use_pulseaudio : Now is automatic when detect pulseaudio server installed _use_nacl : Now WORKS when set to 1. thanks @jeagoss. for more info see http://code.google.com/p/chromium/issues/detail?id=104569 Back Don't use type 'settings' in gyp files: http://code.google.com/p/chromium/issues/detail?id=96629 Add Remove unnecesary components Add More internal patches Mod Force usage Python2 greetings

Det commented on 2011-12-13 07:47 (UTC)

Same thing: not yet mentioned in the blog but .6 is the latest Dev one.

sl1pkn07 commented on 2011-12-13 00:36 (UTC)

@big_bum the latest DEV-CHANNEL is 17.0.963.2 http://googlechromereleases.blogspot.com/search/label/Dev%20updates @jeagoss O_O i test with this 3 days ago and don't compile :S http://code.google.com/p/chromium/issues/detail?id=104569#c19 that if, without CFLAGS edit and without "filter-sp.patch". only CXXFLAGS whitout -fastak-protector don't test with CFLAGS, CXXFLAGS and patch (in test now) waiting result... EDIT: WORKS O_o THANKS!!!!

jeagoss commented on 2011-12-12 20:33 (UTC)

When building with NaCl, I found something that might work as a fix for the missing symbols error. I had to add the following lines export CXXFLAGS="`echo ${CXXFLAGS} | sed 's| -fstack-protector||'`" export CFLAGS="`echo ${CFLAGS} | sed 's| -fstack-protector||'`" I put them right under line 99 where you link the newlib folder into the srcdir I tested the changes on x86_64 and i686. This was the only way I could get the build to complete with _use_nacl=1

big_bum commented on 2011-12-12 18:15 (UTC)

Isn't the latest version 18.0.969.0 ?

sl1pkn07 commented on 2011-12-12 16:36 (UTC)

try now (update)

TamCore commented on 2011-12-12 15:12 (UTC)

resolving dependencies... warning: cannot resolve "gconf libgnome-keyring", a dependency of "chromium-dev" :: The following package cannot be upgraded due to unresolvable dependencies: chromium-dev

sl1pkn07 commented on 2011-12-11 22:12 (UTC)

update libjpeg-turbo not longer need (http://code.google.com/p/chromium/issues/detail?id=106954) perl-switch not longer need add use_system_libexpat and use_system_speex GYP_defines (and speex dependency) sorry for delay

sl1pkn07 commented on 2011-11-10 17:38 (UTC)

fixed

kfgz commented on 2011-11-10 12:45 (UTC)

Md5sum of chromium-17.0.932.0.tar.bz2 is not correct. It should be afe0e00c9745a145bdc910c12497780e

sl1pkn07 commented on 2011-11-10 10:38 (UTC)

ok. fixed

imrehg commented on 2011-11-10 02:05 (UTC)

New location to tarballs (according to the README in the previous directory), like http://commondatastorage.googleapis.com/chromium-browser-official/chromium-17.0.934.0.tar.bz2

sl1pkn07 commented on 2011-11-07 06:25 (UTC)

sorry for the dalay this weekend broken one of my HDD. fortunately restore all of data update to 17.0.928.0

sl1pkn07 commented on 2011-10-19 22:18 (UTC)

update to 16.0.912.0 fix build with pulseaudio change use gyp flags to build correctly with pulseaudio support. disable by default, to enable edit "_use_pulseaudio" variable to 1 greetings

sl1pkn07 commented on 2011-10-06 22:05 (UTC)

why i use it. :)

misc commented on 2011-10-06 17:20 (UTC)

[Not a complaint:] Why is "Build With Pulseaudio" enabled by default?

terminalmage commented on 2011-09-26 12:16 (UTC)

http://googlechromereleases.blogspot.com/2011/09/dev-channel-update_22.html "Due to known bug, Linux isn't updated." http://code.google.com/p/chromium/issues/detail?id=97655 I would upgrade at your own risk. This isn't an "official" release for Linux.

sl1pkn07 commented on 2011-09-26 12:10 (UTC)

update to 16.0.889.0

sl1pkn07 commented on 2011-09-16 18:46 (UTC)

update to 15.0.874.15

sl1pkn07 commented on 2011-09-14 14:13 (UTC)

update to 15.0.874.12

sl1pkn07 commented on 2011-09-09 21:00 (UTC)

update to 15.0.874.5

sl1pkn07 commented on 2011-09-09 19:13 (UTC)

doh! the RSS dont mark the update update later, previous compile by myselft to test build correctly

Det commented on 2011-09-09 19:04 (UTC)

So does 15.0.874.5 build?

sl1pkn07 commented on 2011-09-09 19:00 (UTC)

oks, thanks greetings

terminalmage commented on 2011-09-09 18:58 (UTC)

Yes, I was able to build on i686 with the updated patch.

sl1pkn07 commented on 2011-09-09 18:01 (UTC)

archtaku. whit the patch mod fix your problem?

terminalmage commented on 2011-09-09 16:33 (UTC)

Just a note to people who tried to build 15.0.874.0 before the patch was modified to include i686: If you're getting patch errors, remove/rename the "src" dir.

sl1pkn07 commented on 2011-09-09 15:50 (UTC)

if have fails post in comment. dot mark outdated please. im not soothsayer greetings

sl1pkn07 commented on 2011-09-08 19:55 (UTC)

try now i'm modiffied patch to make same fix on 32bits (before only patch 64 bits) greetings and sorry

terminalmage commented on 2011-09-08 19:45 (UTC)

Updated PKGBUILD does not work. CARCH is correctly identified as i686 and patch is not applied, but build error is still the same.

terminalmage commented on 2011-09-08 19:36 (UTC)

Didn't work.

sl1pkn07 commented on 2011-09-08 19:25 (UTC)

mmm oks comment patch -p1 -i ../nacl.gypi.patch line fix later thanks

terminalmage commented on 2011-09-08 19:07 (UTC)

Build works on x86_64 but fails on i686: http://pastebin.com/fYg5LUGY

sl1pkn07 commented on 2011-09-08 18:12 (UTC)

update dev channel 15.0.874.0 add patch for error compile (see http://code.google.com/p/chromium/issues/detail?id=94518) more configs for gnome users (add "use_gnome_keyring" and "linux_link_gnome_keyring" gyps flags) sorry for delay

sl1pkn07 commented on 2011-08-30 21:29 (UTC)

probe this patches. but not works (and anhother patches and script fixes) for this need help :S

terminalmage commented on 2011-08-30 15:51 (UTC)

Looks like this has been submitted to their bug tracker. May want to follow this bug report: http://code.google.com/p/chromium/issues/detail?id=94518

sl1pkn07 commented on 2011-08-30 14:28 (UTC)

please. need help! http://paste.kde.org/116143/

sl1pkn07 commented on 2011-08-20 18:27 (UTC)

update to update to 15.0.854.0 fixed don't close tabs

sl1pkn07 commented on 2011-08-18 01:56 (UTC)

back to 849 http://code.google.com/p/chromium/issues/detail?id=93086

sl1pkn07 commented on 2011-08-18 01:45 (UTC)

update to 15.0.854.0

sl1pkn07 commented on 2011-08-12 16:04 (UTC)

update to 15.0.849.0 - gyp_define "use_system_libevent=0" (fail build with "1") - don't install "product_logo_64.png" icon

sl1pkn07 commented on 2011-08-08 12:47 (UTC)

fixed. add gyp_define "include_pulse_audio" sorry for this :S

imrehg commented on 2011-08-08 09:23 (UTC)

Looking at the changes, it only adds the option to the PKGBUILD, but "_use_pulseaudio" is not actually used in the GYP_DEFINES, so nothing changes in the build, except the "depends" of the package. Did I miss something?

sl1pkn07 commented on 2011-08-07 20:21 (UTC)

update to 14.0.835.18 addep "_use_pulseaudio" option by default is 1 (enable), if you NOT use pulseaudio, set to 0 (disable, don't add pulseaudio in depends)

Meow commented on 2011-08-06 11:32 (UTC)

pulseaudio also needs to be included in dependencies, please add it.

sl1pkn07 commented on 2011-07-29 04:40 (UTC)

update dev channel 14.0.835.8

sl1pkn07 commented on 2011-07-27 14:12 (UTC)

update dev channel 14.0.835.4 - Change method force use Python2

imrehg commented on 2011-07-22 07:43 (UTC)

sl1pkn07: Still need to add to the PKGBUILD because if the package does not use 'make ${MAKEFLAGS}', then the line in makepkg.conf does not matter at all. Once it is included, the parameter value can be added manually (like I did in the example), or automatically like using the conf you referenced.

chmielu commented on 2011-07-19 18:28 (UTC)

@piotrek91666 pacman -S base-devel

piotrek91666 commented on 2011-07-19 18:19 (UTC)

I have error: | *** Starting to parse CSSPropertyNames.in... | | *** Starting to parse CSSValueKeywords.in... | ACTION webcore_bindings_sources_MathMLNames out/Release/obj/gen/webkit/MathMLNames.cpp ACTION webcore_bindings_sources_UserAgentStyleSheets out/Release/obj/gen/webkit/UserAgentStyleSheets.h ACTION webcore_bindings_sources_XLinkNames out/Release/obj/gen/webkit/XLinkNames.cpp ACTION webcore_bindings_sources_XMLNSNames out/Release/obj/gen/webkit/XMLNSNames.cpp ACTION webcore_bindings_sources_XMLNames out/Release/obj/gen/webkit/XMLNames.cpp ACTION webcore_bindings_sources_tokenizer out/Release/obj/gen/webkit/tokenizer.cpp RULE webcore_bindings_sources_bison_0 out/Release/obj/gen/webkit/CSSGrammar.cpp Traceback (most recent call last): File "scripts/action_maketokenizer.py", line 101, in <module> sys.exit(main(sys.argv)) File "scripts/action_maketokenizer.py", line 89, in main p1 = subprocess.Popen(['flex', '-t', flexInput], stdout=subprocess.PIPE) File "/usr/lib/python2.7/subprocess.py", line 679, in __init__ errread, errwrite) File "/usr/lib/python2.7/subprocess.py", line 1228, in _execute_child raise child_exception OSError: [Errno 2] No such file or directory RULE webcore_bindings_sources_bison_1 out/Release/obj/gen/webkit/XPathGrammar.cpp make: *** [out/Release/obj/gen/webkit/tokenizer.cpp] Błąd 1 make: *** Oczekiwanie na niezakończone zadania.... Traceback (most recent call last): File "scripts/rule_bison.py", line 72, in <module> returnCode = subprocess.call(['bison', '-d', '-p', prefix, inputFile, '-o', outputCpp]) File "/usr/lib/python2.7/subprocess.py", line 493, in call return Popen(*popenargs, **kwargs).wait() File "/usr/lib/python2.7/subprocess.py", line 679, in __init__ errread, errwrite) File "/usr/lib/python2.7/subprocess.py", line 1228, in _execute_child raise child_exception OSError: [Errno 2] No such file or directory make: *** [out/Release/obj/gen/webkit/CSSGrammar.cpp] Błąd 1 Traceback (most recent call last): File "scripts/rule_bison.py", line 72, in <module> returnCode = subprocess.call(['bison', '-d', '-p', prefix, inputFile, '-o', outputCpp]) File "/usr/lib/python2.7/subprocess.py", line 493, in call return Popen(*popenargs, **kwargs).wait() File "/usr/lib/python2.7/subprocess.py", line 679, in __init__ errread, errwrite) File "/usr/lib/python2.7/subprocess.py", line 1228, in _execute_child raise child_exception OSError: [Errno 2] No such file or directory make: *** [out/Release/obj/gen/webkit/XPathGrammar.cpp] Błąd 1

sl1pkn07 commented on 2011-07-19 14:20 (UTC)

update Dev channel. 14.0.825 - GCC 4.6 patch not need (commented in pkgbuild) - build ffmpegsumo.so by default (to use system ffmpeg libs change "_use_ffmpeg_system_libs" to "1"

sl1pkn07 commented on 2011-07-17 18:57 (UTC)

yes. I'd added this option because i'd had the same problem. i dont know why occur this, i think this occur because ffmpeg what is installed in the system uses new API (chromium uses old API) the difference between ffmpeg or internal ffmpeg is minimal: (ffmpeg uses for decoder, not for encode) http://i.imgur.com/rLj8j.png (compile with ffmpegsumo.so) greetings

blkqi commented on 2011-07-17 18:23 (UTC)

compile was OK with _use_ffmpeg_system_libs=0 with _use_ffmpeg_system_libs=1 I get the av_malloc() conflict posted below.

sl1pkn07 commented on 2011-07-16 19:14 (UTC)

this not depend of your /etc/makepkg.conf? #-- Make Flags: change this for DistCC/SMP systems MAKEFLAGS="-j9"

imrehg commented on 2011-07-16 10:44 (UTC)

Hi, could you add ${MAKEFLAGS} to the build line, like "make BUILDTYPE=${_buildtype} ${MAKEFLAGS} chrome chrome_sandbox"? It's the same as in core/kernel26, and lets me build things much faster on a multicore system (e.g. running "MAKEFLAGS=-j8 makepkg"). Cheers!

solsticedhiver commented on 2011-07-14 13:31 (UTC)

why "update-desktop-database -q" is not run in post_install in .install file as for chromium ?

sl1pkn07 commented on 2011-07-13 15:10 (UTC)

update. thanks huulivoide for the "fix" xd

Huulivoide commented on 2011-07-13 12:11 (UTC)

Turn the system_icu off till it gets fixed, upstream, or someone writes a patch, unless your working on a patch alredy ^^

sl1pkn07 commented on 2011-07-12 12:25 (UTC)

fail build new release Dev channel CXX(target) out/Release/obj.target/i18n_api/v8/src/extensions/experimental/number-format.o v8/src/extensions/experimental/number-format.cc: En la función ‘icu_46::DecimalFormat* v8::internal::CreateFormatterFromSkeleton(const icu_46::Locale&, const icu_46::UnicodeString&, UErrorCode*)’: v8/src/extensions/experimental/number-format.cc:254:5: error: ‘EStyles’ no es un miembro de ‘icu_46::NumberFormat’ v8/src/extensions/experimental/number-format.cc:254:32: error: expected ‘;’ before ‘style’ v8/src/extensions/experimental/number-format.cc:257:9: error: ‘style’ no se declaró en este ámbito v8/src/extensions/experimental/number-format.cc:257:17: error: ‘kCurrencyStyle’ no es un miembro de ‘icu_46::NumberFormat’ v8/src/extensions/experimental/number-format.cc:260:17: error: ‘kIsoCurrencyStyle’ no es un miembro de ‘icu_46::NumberFormat’ v8/src/extensions/experimental/number-format.cc:263:17: error: ‘kPluralCurrencyStyle’ no es un miembro de ‘icu_46::NumberFormat’ v8/src/extensions/experimental/number-format.cc:267:55: error: ‘style’ no se declaró en este ámbito CXX(target) out/Release/obj.target/i18n_api/gen/i18n-js.o AR(target) out/Release/obj.target/remoting/libremoting_base.a make: *** [out/Release/obj.target/i18n_api/v8/src/extensions/experimental/number-format.o] Error 1

sl1pkn07 commented on 2011-07-12 04:24 (UTC)

new DEV channel release. update later

sl1pkn07 commented on 2011-07-11 12:47 (UTC)

media/ffmpeg/ffmpeg_common.cc: En la función ‘base::TimeDelta media::ConvertFromTimeBase(const AVRational&, int64)’: media/ffmpeg/ffmpeg_common.cc:15:70: error: ‘av_rescale_q’ no se declaró en este ámbito media/ffmpeg/ffmpeg_common.cc: En la función ‘int64 media::ConvertToTimeBase(const AVRational&, const base::TimeDelta&)’: media/ffmpeg/ffmpeg_common.cc:21:73: error: ‘av_rescale_q’ no se declaró en este ámbito media/ffmpeg/ffmpeg_common.cc:22:1: aviso: el control alcanza el final de una función que no es void [-Wreturn-type] make: *** [out/Release/obj.target/media/media/ffmpeg/ffmpeg_common.o] Error 1

Huulivoide commented on 2011-07-11 09:49 (UTC)

Here this should include all http://paste.pocoo.org/show/435997/

Huulivoide commented on 2011-07-11 09:45 (UTC)

Oh yeh I forgot the stubs dir I remember there were two places I did modifications in but as it was few weeks ago I didn't rememeber that let me update the patch

sl1pkn07 commented on 2011-07-10 16:39 (UTC)

fail for me out/Release/obj.target/geni/ffmpeg_stubs.cc:41:64: error: la declaración de la función C ‘void* av_malloc(unsigned int)’ genera un conflicto con /usr/include/libavutil/mem.h:73:7: error: declaración previa de ‘void* av_malloc(size_t)’ aquí CXX(target) out/Release/obj.target/wtf/third_party/WebKit/Source/JavaScriptCore/wtf/DecimalNumber.o test with libav-git and using system ffmpeg (_use_ffmpeg_system_libs=1)

Huulivoide commented on 2011-07-10 09:55 (UTC)

http://paste.pocoo.org/show/435053/ This patch should fix some of the symbols but im not sure if this is all, please install some svn or git liav pacxkage and test it ^^

sl1pkn07 commented on 2011-06-28 13:35 (UTC)

update dev: 14.0.803 and patches

sl1pkn07 commented on 2011-06-27 19:59 (UTC)

what patch? link please

Huulivoide commented on 2011-06-22 09:18 (UTC)

could you add a patch for newer ffmpeg-revisions? It only recuires few changes to change depricated funktion names with new ones. Like CODEC_TYPE_VIDEO becomes AVMEDIA_TYPE_VIDEO, these are alredy so old that they hawe been completely removed from ffmpeg for some time.

sl1pkn07 commented on 2011-06-21 13:52 (UTC)

update dev channel: 14.0.797.0 - update patch for gcc46 and glib 2.14 - add option: "_use_ffmpeg_system_libs=" 0 = build "ffmpegsumo.so" library 1 = use system ffmpeg greetings

joshdmiller commented on 2011-06-19 22:42 (UTC)

I am using glib 1.2.10 still, so I commented out the tmalloc patch line and everything's groovy now.

sl1pkn07 commented on 2011-06-19 18:59 (UTC)

update: when using Glib 2.14 [testing] apply the patch for compile with glib 2.14, if use glib 2.10 [core], skip this patch

sl1pkn07 commented on 2011-06-19 15:59 (UTC)

what glib version using you?

joshdmiller commented on 2011-06-19 13:58 (UTC)

I still get tcmalloc compilation errors: CXX(target) out/Release/obj.target/allocator/third_party/tcmalloc/chromium/src/thread_cache.o third_party/tcmalloc/chromium/src/tcmalloc.cc:379:36: error: conflicting declaration ‘void* (* volatile __malloc_hook)(size_t, const void*)’ /usr/include/malloc.h:165:16: error: ‘__malloc_hook’ has a previous declaration as ‘void* (* __malloc_hook)(size_t, const void*)’ third_party/tcmalloc/chromium/src/tcmalloc.cc:387:47: error: conflicting declaration ‘void* (* volatile __realloc_hook)(void*, size_t, const void*)’ /usr/include/malloc.h:167:16: error: ‘__realloc_hook’ has a previous declaration as ‘void* (* __realloc_hook)(void*, size_t, const void*)’ third_party/tcmalloc/chromium/src/tcmalloc.cc:393:60: error: conflicting declaration ‘void (* volatile __free_hook)(void*, const void*)’ /usr/include/malloc.h:163:15: error: ‘__free_hook’ has a previous declaration as ‘void (* __free_hook)(void*, const void*)’ third_party/tcmalloc/chromium/src/tcmalloc.cc:1587:64: error: conflicting declaration ‘void* (* volatile __memalign_hook)(size_t, size_t, const void*)’ /usr/include/malloc.h:169:16: error: ‘__memalign_hook’ has a previous declaration as ‘void* (* __memalign_hook)(size_t, size_t, const void*)’ make: *** [out/Release/obj.target/allocator/third_party/tcmalloc/chromium/src/tcmalloc.o] Error 1 make: *** Waiting for unfinished jobs.... WARNING - duplicate keys exist in shortcut group IDD_CHROME_FRAME_FIND_DIALOG

sl1pkn07 commented on 2011-06-18 22:48 (UTC)

update and adopt this package fixed with patch for Glib 2.14 greetings

commented on 2011-06-17 12:03 (UTC)

These days Stable is good enough for me. Have fun.

sl1pkn07 commented on 2011-06-14 16:11 (UTC)

+1 to jim1960 error :S

KittyKatt commented on 2011-06-14 04:16 (UTC)

Getting the same error as jim1960. Waiting for a fix, perhaps.

commented on 2011-06-09 10:58 (UTC)

@yvonney: Both gperf and libgnome-keyring are in makedepends, meaning that they are needed at build time but could be uninstalled afterwards.

jim1960 commented on 2011-06-08 00:49 (UTC)

I just upgrade glibc to 2.14-1 then compile error: third_party/tcmalloc/chromium/src/tcmalloc.cc:379:36: 錯誤:相互衝突的宣告 「void* (* __malloc_hook)(size_t, const void*)」 /usr/include/malloc.h:176:39: 錯誤:「__malloc_hook」早先被宣告為 「void* (* volatile __malloc_hook)(size_t, const void*)」 third_party/tcmalloc/chromium/src/tcmalloc.cc:387:47: 錯誤:相互衝突的宣告 「void* (* __realloc_hook)(void*, size_t, const void*)」 /usr/include/malloc.h:179:39: 錯誤:「__realloc_hook」 早先被宣告為 「void* (* volatile __realloc_hook)(void*, size_t, const void*)」 third_party/tcmalloc/chromium/src/tcmalloc.cc:393:50: 錯誤:相互衝突的宣告 「void (* __free_hook)(void*, const void*)」 /usr/include/malloc.h:173:38: 錯誤:「__free_hook」 早先被宣告為 「void (* volatile __free_hook)(void*, const void*)」 third_party/tcmalloc/chromium/src/tcmalloc.cc:1587:54: 錯誤:相互衝突的宣告 「void* (* __memalign_hook)(size_t, size_t, const void*)」 /usr/include/malloc.h:183:39: 錯誤:「__memalign_hook」 早先被宣告為 「void* (* volatile __memalign_hook)(size_t, size_t, const void*)」 make: *** [out/Release/obj.target/allocator/third_party/tcmalloc/chromium/src/tcmalloc.o] Error 1

yvonney commented on 2011-06-07 04:00 (UTC)

Upon install the dev version wants to install: gperf libgnome-keyring are they now essential to the dev build? forum posting suggestion great. big thanks for all the help! no luck yet though that's ok for now.

Vi0L0 commented on 2011-06-07 00:42 (UTC)

@yvonney: i just dunno why you are getting this: install file (chromium-dev.install) does not exist. error. chromium-dev.install surelly is there, so maybe there is some problem with your filesystem? I dunno, i think that you should write new post about this issue on the forum. Ps. ah sorry you were writing about aurget, not yourt... i'm kinda sleepy (-.-)Zzz then i have no idea is there possibility of editing PKGBUILD before building when using aurget...

Vi0L0 commented on 2011-06-07 00:28 (UTC)

@yvonney: i'm not yaourt user, although i'm pretty sure that with it you can edit PKGBUILD before building package. So edit PKGBUILD of perl-switch and simply remove "make test &&" You will find it in: make &&make test && line inside build() function. Then exit editor and it shall build package.

commented on 2011-06-06 12:02 (UTC)

Fixed build with icu 4.8

yvonney commented on 2011-06-06 00:05 (UTC)

THANKS everyone! can you advise? WHICH 'test' line to comment? Still getting: '==> ERROR: install file (chromium-dev.install) does not exist.' # for n in avcodec avdevice avfilter avformat avutil postproc swscale; do # if test -e /usr/lib/lib${n}.so.[0-9]; then # f=`echo /usr/lib/lib${n}.so.[0-9]` # else # f=`echo /usr/lib/lib${n}.so.[0-9][0-9]` # fi # f=`basename "$f"` # ln -s ../$f "${chromium_home}/${f}" # done

Vi0L0 commented on 2011-06-05 23:19 (UTC)

yes, with use_system_icu=0 it's compiling fine

Vi0L0 commented on 2011-06-05 22:45 (UTC)

edit: or to change from: use_system_icu=1 \ to: use_system_icu=0 \

Vi0L0 commented on 2011-06-05 22:42 (UTC)

@alub: it's caused by icu which were updated lately: icu (4.6.1-2 -> 4.8-1) when i tried re-compile with new icu i'm getting error :/ i don't have time to investigate so i just copied needed libs from older icu package. If you don't have compiled on older icu chromium-dev package then i guess it's good idea to remove: use_system_icu=1 \ line from chromium-dev's PKGBUILD

alub commented on 2011-06-05 22:33 (UTC)

With 13.0.782.1-2, i have the following error: /usr/lib/chromium-dev/chromium: error while loading shared libraries: libicui18n.so.46: cannot open shared object file: No such file or directory

sl1pkn07 commented on 2011-06-05 19:19 (UTC)

comment the line "make test" this is the "tune pkgbuild" need

yvonney commented on 2011-06-05 04:49 (UTC)

ERROR: install file (chromium-dev.install) does not exist.

yvonney commented on 2011-06-05 04:47 (UTC)

runnin testing repo also. is it that perl version is too new in testing repo? could use a solution: Test Summary Report ------------------- t/given.t (Wstat: 0 Tests: 293 Failed: 2) Failed tests: 2-3 Files=3, Tests=590, 0 wallclock secs ( 0.06 usr 0.01 sys + 0.49 cusr 0.01 csys = 0.57 CPU) Result: FAIL Failed 1/3 test programs. 2/590 subtests failed. make: *** [test_dynamic] Error 255 ==> ERROR: A failure occurred in build(). Aborting... notice: perl-switch failed while building, remove source files

Vi0L0 commented on 2011-06-04 01:07 (UTC)

@sl1pkn07: same here (testing), thanks for solution

sl1pkn07 commented on 2011-06-03 19:03 (UTC)

hello. (again xd) fail build Can't locate Switch.pm in @INC (@INC contains: /usr/lib/perl5/site_perl /usr/share/perl5/site_perl /usr/lib/perl5/vendor_perl /usr/share/perl5/vendor_perl /usr/lib/perl5/core_perl /usr/share/perl5/core_perl .) at ../make-hash-tools.pl line 23. BEGIN failed--compilation aborted at ../make-hash-tools.pl line 23. Can't locate Switch.pm in @INC (@INC contains: /usr/lib/perl5/site_perl /usr/share/perl5/site_perl /usr/lib/perl5/vendor_perl /usr/share/perl5/vendor_perl /usr/lib/perl5/core_perl /usr/share/perl5/core_perl .) at ../make-hash-tools.pl line 23. BEGIN failed--compilation aborted at ../make-hash-tools.pl line 23. make: *** [out/Release/obj/gen/webkit/DocTypeStrings.cpp] Error 2 make: *** Se espera a que terminen otras tareas.... make: *** [out/Release/obj/gen/webkit/ColorData.cpp] Error 2 /tmp/yaourt-tmp-sl1pkn07/aur-chromium-dev/src/chromium-13.0.782.1/third_party/WebKit/Source/WebCore/css/CSSGrammar.y:946.5: aviso: a `;' might be needed at the end of action code /tmp/yaourt-tmp-sl1pkn07/aur-chromium-dev/src/chromium-13.0.782.1/third_party/WebKit/Source/WebCore/css/CSSGrammar.y:946.5: aviso: future versions of Bison will not add the `;' WARNING - duplicate keys exist in shortcut group IDD_CHROME_FRAME_FIND_DIALOG need install perl-switch (with tune pkgbuild) im using [testing]

sl1pkn07 commented on 2011-05-15 08:58 (UTC)

i used FFmpeg-MT fork from Mplayer2 repository http://git.mplayer2.org/mplayer2-build/ -> http://repo.or.cz/w/FFMpeg-mirror/mplayer-patches.git

commented on 2011-05-14 14:34 (UTC)

@sl4pkn07: Cannot reproduce. $ pacman -Q ffmpeg ffmpeg 20110330-2

sl1pkn07 commented on 2011-05-14 13:31 (UTC)

Fail Build (ffmpeg) http://paste.kde.org/67309

imrehg commented on 2011-05-14 07:18 (UTC)

Hi, current version needs to disable building the native client for the moment, as it has to download a precompiled executable, but it's not a simple thing just yet. As e.g. Gentoo did, add the following to the gyp setting list: disable_nacl=1 Also, gcc46.patch needs a little update because one of the files (third_party/libjingle/source/talk/base/sigslot.h) changed within the patched region, here's a patched version https://github.com/imrehg/aur/blob/72233efe203ff3dae2fe4dbac57037a29e2f1069/usedpackages/chromium/gcc46.patch

commented on 2011-05-01 17:44 (UTC)

Apparently the path has changed from build.chromium.org/buildbot/official/ to build.chromium.org/official/. Not a problem as the former is set to redirect to the new address, just a warning.

commented on 2011-04-19 07:50 (UTC)

12.0.733.0-2: Pulled gcc 4.6 build fix from http://trac.webkit.org/changeset/84123

Vi0L0 commented on 2011-04-18 18:53 (UTC)

@mvorozhtsov: yes, you are right, i just compiled it on virtual machine with gcc 4.5 and it's fine. You may download x86_64 package from: http://catalyst.apocalypsus.net/other_pkgs/chromium-dev-12.0.733.0-1-x86_64.pkg.tar.xz

commented on 2011-04-18 12:19 (UTC)

It's http://code.google.com/p/chromium/issues/detail?id=46411 I can't even build 12.0.725.0 anymore. Does it happen with gcc 4.5? I can't downgrade my machine because I use GNOME 3.

Vi0L0 commented on 2011-04-18 11:11 (UTC)

I am also getting failure, although mine looks kinda different: http://hpaste.org/45804 + i'm using [testing]

sl1pkn07 commented on 2011-04-18 10:30 (UTC)

fail build: http://paste.kde.org/13156

commented on 2011-04-08 07:56 (UTC)

Use Arch about 2 years and this is the first package that requires base-devel:)

commented on 2011-04-08 05:19 (UTC)

@4ybaka https://wiki.archlinux.org/index.php/AUR#Getting_started

commented on 2011-04-07 16:04 (UTC)

@mvorozhtsov Thanks, I haven't some of these packages. Why them not included in dependencies?

commented on 2011-04-07 13:31 (UTC)

@4ybaka $ pacman -S base-devel

commented on 2011-04-07 13:12 (UTC)

Build failed on last update... Traceback (most recent call last): File "scripts/action_maketokenizer.py", line 101, in <module> sys.exit(main(sys.argv)) File "scripts/action_maketokenizer.py", line 89, in main p1 = subprocess.Popen(['flex', '-t', flexInput], stdout=subprocess.PIPE) File "/usr/lib/python2.7/subprocess.py", line 672, in __init__ errread, errwrite) File "/usr/lib/python2.7/subprocess.py", line 1202, in _execute_child raise child_exception OSError: [Errno 2] No such file or directory make: *** [out/Release/obj/gen/webkit/tokenizer.cpp] Error 1

commented on 2011-04-06 07:11 (UTC)

Yes, I'm on testing. I assume then that the issue lies with my toolchain.

commented on 2011-04-05 07:45 (UTC)

@robert_paxon: Not me. Are you on [testing]?

commented on 2011-04-04 21:35 (UTC)

This compilation error has been present since 12.0.712.0-1: CXX(target) out/Release/obj.target/browser/chrome/browser/search_engines/template_url_prepopulate_data.o In file included from chrome/browser/search_engines/template_url_prepopulate_data.cc:5:0: ./chrome/browser/search_engines/template_url_prepopulate_data.h:30:29: error: ‘size_t’ has not been declared make: *** [out/Release/obj.target/browser/chrome/browser/search_engines/template_url_prepopulate_data.o] Error 1 ------------------------- I assume it's completely an upstream issue but I'm posting to see if others are getting the same error.

ydasilva commented on 2011-03-18 22:14 (UTC)

@mvorozhtsov my bad then :)

commented on 2011-03-15 11:23 (UTC)

@dl.zerocool This package tracks the Dev channel (http://www.chromium.org/getting-involved/dev-channel, announcements are published at http://googlechromereleases.blogspot.com/) and NOT supposed to provide the "last-published-tarball" version.

ydasilva commented on 2011-03-15 10:31 (UTC)

To get the new version : pkgver=11.0.696.3 to pkgver=12.0.703.0 You can always check the last version on : http://build.chromium.org/official/ Shall you do a small script to update the pkgbuild daily :) (And change md5 signature of course)

commented on 2011-03-13 21:21 (UTC)

It appears not to work with the NaCl demos. ex, http://nacl-gallery.appspot.com/life/life.html Note, I have --no-sandbox and I have enabled NaCl in about:flags. [6949:6949:110715291860:ERROR:gpu_info_collector_linux.cc(178)] Failed to locate libpci [6972:6972:110715410744:ERROR:renderer_main.cc(299)] Running without renderer sandbox [6974:6974:110715477114:ERROR:renderer_main.cc(299)] Running without renderer sandbox Moonlight: 3.99.0.1.99 Moonlight: Attempting to load libmoonloaderxpi [6977:6977:110715706753:ERROR:renderer_main.cc(299)] Running without renderer sandbox [6981:6981:110715884523:ERROR:renderer_main.cc(299)] Running without renderer sandbox [6986:6986:110716197973:ERROR:renderer_main.cc(299)] Running without renderer sandbox [6997:6997:110721749765:ERROR:renderer_main.cc(299)] Running without renderer sandbox [7000:7000:110753067670:ERROR:renderer_main.cc(299)] Running without renderer sandbox [7003:7003:110753212924:ERROR:native_library_linux.cc(32)] dlopen failed when trying to open default_plugin: default_plugin: cannot open shared object file: No such file or directory [7005:7005:110767985590:ERROR:renderer_main.cc(299)] Running without renderer sandbox [7019:7019:110913395047:ERROR:renderer_main.cc(299)] Running without renderer sandbox [7022:7022:110918882152:ERROR:renderer_main.cc(299)] Running without renderer sandbox [7025:7025:110932539981:ERROR:renderer_main.cc(299)] Running without renderer sandbox [7029:7029:111125708219:ERROR:renderer_main.cc(299)] Running without renderer sandbox

Huulivoide commented on 2011-03-12 12:49 (UTC)

ffmpeg version must be lower than 20110215 as there was a bigger change in the code base and chromium hasnt been updated to reflect it yet.

commented on 2011-03-09 13:54 (UTC)

Beware of https://bugs.archlinux.org/task/23204

commented on 2011-02-18 12:50 (UTC)

11.0.672.2: Disabled system ICU for now as it breaks linking for me.

commented on 2011-02-11 23:33 (UTC)

Now, after nearly two hours of compiling, it works very smooth :)

commented on 2011-02-01 14:42 (UTC)

There was no .11 tarball so I packaged .12. @Det: "next update" = "when I have time".

Det commented on 2011-02-01 13:55 (UTC)

Out of curiosity, is there a specific reason you want to wait until the next release each time there's something to be changed in the PKGBUILD? If you checked/verified that (10.0.648).11 was already released, then it's of course fine to just say that you'll be doing both changes on the same go, though.

commented on 2011-02-01 13:21 (UTC)

lastchange.py uses git, I'll add it to makedepends in the next update.

terminalmage commented on 2011-01-31 19:36 (UTC)

I have the same error when building on i686, only with out/Release/obj.target/geni/webkit_version.h instead of out/Release/obj/gen/chrome/grit/component_extension_resources.h. Didn't have a problem building on x86_64 though.

jpaechnatz commented on 2011-01-31 06:59 (UTC)

...fails to compile... RULE chrome_extra_resources_grit_0 out/Release/obj/gen/chrome/grit/component_extension_resources.h Traceback (most recent call last): File "./lastchange.py", line 149, in <module> sys.exit(main()) File "./lastchange.py", line 134, in main version_info = FetchVersionInfo(opts.default_lastchange) File "./lastchange.py", line 84, in FetchVersionInfo if IsGitSVN(directory): File "./lastchange.py", line 31, in IsGitSVN cwd=directory) File "/usr/lib/python2.7/subprocess.py", line 486, in call return Popen(*popenargs, **kwargs).wait() File "/usr/lib/python2.7/subprocess.py", line 672, in __init__ errread, errwrite) File "/usr/lib/python2.7/subprocess.py", line 1202, in _execute_child raise child_exception OSError: [Errno 2] No such file or directory make: *** [out/Release/obj/gen/build/LASTCHANGE.always] Fehler 1 make: *** Warte auf noch nicht beendete Prozesse... WARNING - duplicate keys exist in shortcut group IDD_CHROME_FRAME_FIND_DIALOG sk duplicates: &N fr duplicates: &R ro duplicates: &G nl duplicates: &O no duplicates: &N pt-PT duplicates: &P sw duplicates: &F ca duplicates: &A de duplicates: &N tr duplicates: &A sv duplicates: &S da duplicates: &O lv duplicates: &R, &U pl duplicates: &W cs duplicates: &N fi duplicates: &K es-419 duplicates: &A id duplicates: &T es duplicates: &B

commented on 2011-01-25 08:23 (UTC)

I've also tried to compile without 'disable_sse2=1' on x86_64 Intel Core 2 Duo, Intel Core i5, Intel Xeon. Everything seems to work fine.

commented on 2011-01-15 07:06 (UTC)

OK, I'll add this in the next update.

Vi0L0 commented on 2011-01-14 17:54 (UTC)

i've compiled it on x86_64 with sse2 and everything is ok - so i could recommend it to other x86_64 users. ofcourse you can leave it how it looks now - if somebody wants he can always remove 'disable_sse2=1' but maybe a good consensus here is to copy solution from [1] extra's chromium: $([ "${CARCH}" == 'i686' ] && echo '-Ddisable_sse2=1') [1] http://projects.archlinux.org/svntogit/packages.git/tree/chromium/repos/extra-x86_64/PKGBUILD

commented on 2011-01-14 16:08 (UTC)

It seems that enabling SSE2 results in wrong CFLAGS on some systems: http://bugs.gentoo.org/show_bug.cgi?id=336871 At least it is forcibly disabled in Gentoo for that reason: http://sources.gentoo.org/cgi-bin/viewvc.cgi/gentoo-x86/www-client/chromium/chromium-10.0.634.0.ebuild?revision=1.1&view=markup

Vi0L0 commented on 2011-01-14 15:41 (UTC)

hmm... i can only see disable_sse2=1 and afaik sse2 is supported by all x86_64 cpus - so we don't need any check for this arch ofcourse if im right, but i should be - sse2 is known from 2001's pentium 4 cpus, and amd added such support in their athlon64 cpus...

Det commented on 2011-01-14 13:31 (UTC)

@mvorozhtsov, can't you add a check of /proc/cpuinfo whether sse is supported? :p

Malix commented on 2011-01-14 10:07 (UTC)

@mvorozhtsov: ok, sounds reasonable. so if I have SSE capable cpu, I assume I should change that to zero. Does makepkg.conf override that if I use -march=native, for example.

commented on 2011-01-14 09:59 (UTC)

@Malix: Not all i686 CPUs support SSE.

Malix commented on 2011-01-13 15:25 (UTC)

why do we disable_sse=1 in this PKGBUILD, is there some negative effect of using it? Also, is there some place where I could do some reading on these options, tried to search them with google but didn't find anything worthwhile.

Det commented on 2011-01-06 07:36 (UTC)

@theislainman, you are free to push the *Flag Out-of-date* button too :). As for mvorozhtsov, please check out the minor tweaks I made to your PKGBUILD on the previous comment (in case you missed it).

commented on 2011-01-06 05:56 (UTC)

http://googlechromereleases.blogspot.com/2011/01/dev-channel-update.html?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+GoogleChromeReleases+%28Google+Chrome+Releases%29

Det commented on 2010-12-23 14:26 (UTC)

You are free to take whatever liberties you like with your _own_ packages. For good reasons is always even better.

commented on 2010-12-23 12:55 (UTC)

@Det: I looked up the RegEx bug they mention on the blog and it seems to affect Linux machines too, so I took the liberty of upgrading to the newer version.

Det commented on 2010-12-23 11:15 (UTC)

That'd be nice. The only possible drawbacks with system libraries would be that they were either too new or buggy. I don't think sqlite is.

Malix commented on 2010-12-23 11:00 (UTC)

couldn't we use more system libraries etc, for example: Sqlite ("use_system_sqlite=1")? Seems to work, but is there any drawbacks to this? Afaik we're currently compiling stuff most of us have already installed.

Det commented on 2010-12-22 16:17 (UTC)

Another btw: here's some minor improvements to your PKGBUILD: http://aur.pastebin.com/CzMuGTTz

Det commented on 2010-12-22 13:32 (UTC)

Btw. 10.0.612.3 is an 'incorrect' version (http://googlechromereleases.blogspot.com/2010/12/dev-channel-update_16.html): "Update: The Dev channel has been updated to 10.0.612.3 for _Windows_ and _Mac_ to correct a RegEx regression."

Det commented on 2010-12-20 18:39 (UTC)

Please remove the svn makedependency on the same go (in case you missed c00kiemon5ter's comment).

terminalmage commented on 2010-12-20 18:30 (UTC)

@mvorozhtsov: Looks like they finally posted it

commented on 2010-12-20 08:33 (UTC)

Waiting for the 10.0.612.1 tarball at http://build.chromium.org/buildbot/official/

Det commented on 2010-12-14 15:37 (UTC)

No.

c00kiemon5ter commented on 2010-12-14 14:19 (UTC)

also I can build this without svn installed, is it really a dep ? makedepends=('svn' .. )

Det commented on 2010-12-13 19:13 (UTC)

@dolby, you thought correct: http://code.google.com/p/chromium/issues/detail?id=13322

commented on 2010-12-13 19:05 (UTC)

I thought gconf wasnt required now: http://phajdan-jr.blogspot.com/2010/11/www-clientchromium-gconf-no-longer.html

alphazo commented on 2010-12-12 21:23 (UTC)

Never mind, Chromed Bird and Proxy Switchy are working back on 9.0.597.16.

alphazo commented on 2010-12-12 20:33 (UTC)

Since 9.0.596, "Chromed Bird" and "Proxy Switchy!" extensions have stopped working (icon is there but nothing happens when clicking on it). Does anyone have them running with the latest version (Dec. 12)?

Det commented on 2010-12-10 16:40 (UTC)

@c00kiemon5ter, I can't because there is nothing to correct.

c00kiemon5ter commented on 2010-12-10 01:59 (UTC)

@fettouchi as ioni pointed in #archlinux -dev is different from 'latest snapshot' ie -dev is this: http://googlechromereleases.blogspot.com/ "The Dev channel has been updated to 9.0.597.10 @ Monday, December 6, 2010" while at the same time http://build.chromium.org/buildbot/official/ already included -10.0.60x this is supposed to be following -dev so I dont think it'll be updated to -10.x soon correct me if I'm wrong

fettouhi commented on 2010-12-08 11:03 (UTC)

I can see that 10.0.x is out. Are you going to switch to that branch soon? does the google webstore and ebook store work with chromium or do you need the official google chrome browser?

commented on 2010-12-05 21:30 (UTC)

@Det: Thank you for the answer. I will wait ;)

Det commented on 2010-12-05 19:12 (UTC)

You're just gonna have to wait until they fix hardware acceleration again. It didn't work with Egl/Gles2 enabled mesa 7.9 for me and it doesn't work with mesa-git either. With both I just get some Egl symbol lookup error about some undefined symbol or something. If I then enable accelerated 2d canvas the extension LastPass stops working - while the accelerated 2d canvas doesn't work either anyway. If you really want to you can rebuild your mesa with "--enable-gles2" but I really don't think this issue is just on my system.

commented on 2010-12-05 15:47 (UTC)

So, what is the solution of the problem about libGLESv2? Thanks ;) Now, i'm using the package of chromium from official repository :(

Det commented on 2010-12-05 08:31 (UTC)

Ahhh, of course it was _Gles2_ not _Egl_. Ok then, my fault the whole thing then. I dunno how I got the idea that "libGLESv2.so" would suddenly be a Mesa Egl component but no harm done... except for archtaku I guess. Sorry about that ;).

commented on 2010-12-05 08:19 (UTC)

alphazo problem is with missing libGLESv2.so, not libffmpegsumo.so. I don't know how he managed to pull this dependency though, looking at media/media.gyp, the default backend is x11, not GLES. I have neither of these shared libraries on my system and chromium is running fine (it complains about libffmpegfumo.so, but that's fine). Maybe it has something to do with enabling experimental 3D/acceleration features (I don't use them).

Det commented on 2010-12-05 08:07 (UTC)

So when alphazo said he cannot start chromium because of the two errors - Mesa Egl and libffmpegsumo.so - he actually could?

commented on 2010-12-05 08:00 (UTC)

Just to clear things up. libffmpegsumo.so is entirely optional and it's absence can't cause any failure.

Koshon commented on 2010-12-04 22:41 (UTC)

to Det: it's not for update, only for who want.

Det commented on 2010-12-04 19:28 (UTC)

Stay in your pants, it's not about being hard to understand. I'm just wondering how can you bypass the Mesa Egl requirement by copying libffmpegsumo.so (Ffmpeg component) to the install folder.

terminalmage commented on 2010-12-04 18:56 (UTC)

Det: Because it _was_ less work. I already had chromium installed, and even if I didn't, installing it and copying the .so _is_ less work than installing one package and recompiling another. I'm fully aware it's not the cleanest method of resolving the issue, which is why I added that it was a dirty hack. Why is this so hard for you to understand?

Det commented on 2010-12-04 18:35 (UTC)

archtaku: Hemm, in that case I _really_ wonder why did you say "but less work than installing an extra package and rebuilding mesa." Koshon: if you mean this package should be updated, you are wrong. Check here the latest stable/beta/_dev_ channel releases: http://googlechromereleases.blogspot.com/

terminalmage commented on 2010-12-04 18:25 (UTC)

Det: You say this as if you think it's all news to me.

Koshon commented on 2010-12-04 16:43 (UTC)

10.0.601.0 md5sum c84b48d9468eeeb71ddc85e4b4f74d4f use of health..

Det commented on 2010-12-04 10:06 (UTC)

@archtaku, if you get an error about a missing mesa feature, your 'dirty hacks' with moving chromium files around aren't going to help. Additionally not everybody even has two chromiums installed at the time and the problem with manually doing stuff is that to keep up doing it you need to do it yourself - whilst packages are auto-updating. @alphazo, that happens with me too. I just don't use accelerated compositing/2d canvas for now. Dunno about your extensions, though. At least with me LastPass started working again when I disabled accelerated 2d canvas (about:flags or remove the manual launcher switch).

alphazo commented on 2010-12-03 22:51 (UTC)

I went the recompilation way (mesa) and installed chromium-dev-codecs-ffmpeg-nonfree-bin. Chromium-dev starts again with only one error message though: /usr/lib/chromium-dev/chromium --type=gpu-process --channel=5542.0x58382b8.1442616187: symbol lookup error: /usr/lib/egl/egl_dri2.so: undefined symbol: _glapi_get_proc_address Now I have two extensions that stopped working on this release (9.0.597): Chromed Bird and Proxy Switch!

alphazo commented on 2010-12-03 20:27 (UTC)

Still no luck even with having the .so file in /usr/lib/chromium-dev. I still have this last issue: [5652:5652:3751961263:ERROR:base/native_library_linux.cc(32)] dlopen failed when trying to open libGLESv2.so: libGLESv2.so: cannot open shared object file: No such file or directory

commented on 2010-12-03 20:11 (UTC)

bortol ~ $ chromium-dev [14546:14546:29055732666:ERROR:base/native_library_linux.cc(32)] dlopen failed when trying to open libGLESv2.so: libGLESv2.so: cannot open shared object file: No such file or directory chromium-dev-codecs-ffmpeg-nonfree-bin [14471:14471:29057246618:ERROR:chrome/browser/renderer_host/render_sandbox_host_linux.cc(625)] sendmsg: Connessione rifiutata Segmentation fault I have already installed chromium-dev-codecs-ffmpeg-nonfree-bin from aur.

terminalmage commented on 2010-12-03 17:39 (UTC)

Instead of doing that, I copied the .so from /usr/lib/chromium/ to /usr/lib/chromium-dev/. A dirty hack, yes, but less work than installing an extra package and rebuilding mesa.

Det commented on 2010-12-03 10:54 (UTC)

You need the ffmpeg codecs from the package chromium-dev-codecs-ffmpeg-nonfree-bin (should be marked as a dependency) and rebuild mesa with "--enable-gles2" and "--enable-egl". Maybe [extra]'s mesa will be built with those flags in the end anyway... just have to wait.

alphazo commented on 2010-12-03 07:30 (UTC)

I have been using chromium-dev for quite some time now. However chromium-dev 9.0.597.0-1 compiles fine but I cannot start it: [1:1:287546422:ERROR:base/native_library_linux.cc(32)] dlopen failed when trying to open /usr/lib/chromium-dev/libffmpegsumo.so: /usr/lib/chromium-dev/libffmpegsumo.so: cannot open shared object file: No such file or directory [3:3:288278871:ERROR:base/native_library_linux.cc(32)] dlopen failed when trying to open /usr/lib/chromium-dev/libffmpegsumo.so: /usr/lib/chromium-dev/libffmpegsumo.so: cannot open shared object file: Permission denied [6171:6171:288841360:ERROR:base/native_library_linux.cc(32)] dlopen failed when trying to open libGLESv2.so: libGLESv2.so: cannot open shared object file: No such file or directory I looked inside /usr/lib/chromium-dev/ and there is no such libffmpegsumo.so.

Det commented on 2010-11-25 14:12 (UTC)

@bsergik, for the 999999th time, flex, bison, automake, autoconf, etc. is included in the group "base-devel" that consists of the most common stuff needed to build stuff from source. Spread the world.

bsergik commented on 2010-11-25 12:13 (UTC)

Please add flex and bison as a makedepends. Otherwise build fails. ACTION Executing flex on compiler/glslang.l out/Release/obj.target/geni/glslang.cpp /bin/sh: flex: command not found make: *** [out/Release/obj.target/geni/glslang.cpp] Error 127 Aborting... ACTION Executing bison on compiler/glslang.y out/Release/obj.target/geni/glslang_tab.cpp /bin/sh: bison: command not found make: *** [out/Release/obj.target/geni/glslang_tab.cpp] Error 127 Aborting...

tomk commented on 2010-11-14 12:34 (UTC)

9.0.580.0 requires libxtst in depends.

Det commented on 2010-11-05 16:48 (UTC)

9.0.570.1 brought only new flash for Chrome users. Not really a reason for rebuild...: http://googlechromereleases.blogspot.com/2010/11/dev-channel-update_04.html

Det commented on 2010-10-30 07:40 (UTC)

@Dogmeat, he probably means this package. Unless Jesses uses accelerated 2d canvas there's no help with 8.0.552.23 either: http://googlechromereleases.blogspot.com/2010/10/dev-channel-update_29.html It was removed from "about:flags" with .23 but it can, of course, be re-enabled with the command line switch: "--enable-accelerated-2d-canvas"

commented on 2010-10-29 23:55 (UTC)

Jesses: sometimes with 9.0.563.0.

Huulivoide commented on 2010-10-29 17:43 (UTC)

Anyone elsa having lock ups whit youtube?

Vi0L0 commented on 2010-10-28 13:58 (UTC)

i686 package: http://www.mediafire.com/file/pw3mtcrgq84j71u/chromium-dev-8.0.552.18-1-i686.pkg.tar.xz + gpu acceleration can be enabled in about:flags tab

Det commented on 2010-10-22 13:02 (UTC)

Didn't the Ffmpeg stuff come with Google Chrome and for "chromium-browser-bin" users through the "chromium-codecs-ffmpeg-nonfree-bin" package?

commented on 2010-10-22 05:55 (UTC)

@hokapoka I may be wrong, but I think ffmpegsumo stuff is installed only when you build the ffmpeg bundled with Chromium, and this PKGBUILD uses the Arch version.

Det commented on 2010-10-21 17:00 (UTC)

@mvorozhtsov, of course, sorry about that.

commented on 2010-10-21 16:30 (UTC)

Not sure if it's just my machines but when starting chromium-dev on all 4 of my boxes I was getting the following error : [1:1:6238048324:ERROR:base/native_library_linux.cc(28)] dlopen failed when trying to open /usr/lib/chromium-dev/libffmpegsumo.so: /usr/lib/chromium-dev/libffmpegsum o.so: cannot open shared object file: No such file or directory The browser appeared to function perfectly without, I assume it's to do with some video playback but haven't tried to see what it might be effecting. I located the file from the binary install of chromium from the extra repos and copied it into place : # sudo cp /usr/lib/chromium/libffmpegsumo.so /usr/lib/chromium-dev/ Restarted chromium-dev and the error hasn't re-appeared. It's a crappy hack I know, but I thought that I would pass it on. Thanks again mvorozhtsov for maintaing this package.

commented on 2010-10-21 15:24 (UTC)

@Det As you can see at http://googlechromereleases.blogspot.com/ >Update: The Dev channel has been updated to 8.0.552.5 for Windows, Mac, and ChromeFrame But not for Linux.

Det commented on 2010-10-21 15:14 (UTC)

To whoever unflagged this as Out-of-date have a look here before doing it again (8.0.552.5 released): http://googlechromereleases.blogspot.com/

jackoneill commented on 2010-10-20 18:37 (UTC)

Commenté par: Det sur Fri, 15 Oct 2010 10:52:46 +0000 You get a binary package with google-chrome/-dev/-beta.

Kwpolska commented on 2010-10-20 18:36 (UTC)

Is there any binary package for chromium-dev?

commented on 2010-10-20 12:33 (UTC)

Ouch, I didn't know that, thanks for replying so fast.

commented on 2010-10-20 12:19 (UTC)

Also, packages already included in base-devel should not be included in makedepends. base-devel had been created to avoid cluttering of makedepends field.

commented on 2010-10-20 12:18 (UTC)

@ TrashMaster You should have done just pacman -S base-devel base-devel group is mandatory for those who want to build something on arch (it's written in wiki). I suggest to install the whole group.

commented on 2010-10-20 12:14 (UTC)

It also depends on patch, bison, and another packages. I've got to restart the build of chromium tree times, installing missing dependencies.

commented on 2010-10-19 07:41 (UTC)

8.0.552.0-2: Compatibility with the new python environment.

Det commented on 2010-10-15 10:52 (UTC)

You get a binary package with google-chrome/-dev/-beta.

dcelasun commented on 2010-10-14 20:51 (UTC)

Is there a place where I can find a binary package for this? It just takes too much time for every update.

alphazo commented on 2010-10-14 13:11 (UTC)

Oops.. My mistake. I had "-g" in my /etc/makepkg.conf. I enabled it to track down a bug on Thunderbird. Thanks for the tip!

alphazo commented on 2010-10-14 13:07 (UTC)

Why is chromium-dev-8.0.552.0-1 package so fat (155MB) and why does it require 787MB to install?

Malix commented on 2010-10-14 13:06 (UTC)

@alphazo did you edit the PKGBUILD and disabled stripping the debug symbols? atleast on my box the package is roughly 17MB

tomk commented on 2010-10-13 18:57 (UTC)

I guess I cheated then. :) I did (briefly) consider wading through it myself for python stuff, but then I decided to rebuild testing/python2 with a /usr/bin/python symlink. After that, the build was successful. It runs fine too.. mostly. That is, it seems to be working as expected so far, but I got a "no such file or directory" error about /usr/lib/chromium-dev/libffmpegsumo.so - and there is indeed no such file or directory in the package I created. extra/chromium has that lib btw.

demonicmaniac commented on 2010-10-05 12:53 (UTC)

Hi. There's some issues with this PKGBUILD when you have testing enabled as testing to moved to python3. The steps suggested by Allan in his blog don't suffice to get it built as some .py files explicitly call python instead of python2 and I can't be arsed to wade through this mess to fix it >.> Perhaps somebody with more patience might be able to.

commented on 2010-10-05 07:17 (UTC)

7.0.536.2-2: Added fix for http://code.google.com/p/chromium/issues/detail?id=57167

mutantmonkey commented on 2010-10-02 20:35 (UTC)

My apologies; I must have neglected to install base-devel after my reinstall.

commented on 2010-10-02 18:16 (UTC)

mutantmonkey, bison is in base-devel group, thus it should not be included in makedepends. base-devel is a requirement to compile anything on arch.

mutantmonkey commented on 2010-10-02 16:41 (UTC)

Please add bison to makedepends, I wasn't able to build without installing it.

commented on 2010-09-29 14:43 (UTC)

@jpaechnatz: Looks like http://code.google.com/p/chromium/issues/detail?id=57275, except I use -Duse_system_ssl=0. It seems you build somehow gets hooked up with the system nss in spite of that.

jpaechnatz commented on 2010-09-29 07:42 (UTC)

Since 7.0.517.8 I wasn't able to compile a newer version...caused by similar failures like: CXX(target) out/Release/obj.target/net/net/spdy/spdy_http_stream.o net/socket/ssl_client_socket_nss.cc: In Funktion »int net::<unnamed>::MapNSPRError(PRErrorCode)«: net/socket/ssl_client_socket_nss.cc:212:10: Fehler: »SSL_ERROR_WEAK_SERVER_KEY« wurde in diesem Gültigkeitsbereich nicht definiert make: *** [out/Release/obj.target/net/net/socket/ssl_client_socket_nss.o] Fehler 1 make: *** Warte auf noch nicht beendete Prozesse... Breche ab ... Do I miss a package? I already checked for nss/nspr, but they are current (maybe to new?) Thanks.

terminalmage commented on 2010-09-25 01:57 (UTC)

New official release: 7.0.517.13

Det commented on 2010-09-23 18:05 (UTC)

Well, at least the "google devs is lazy" and "only one excuse for not updating" parts made my day :).

terminalmage commented on 2010-09-22 22:54 (UTC)

Again, sl1pkn07, if you want a newer version than the latest official dev release, download the tarball (http://aur.archlinux.org/packages/chromium-dev/chromium-dev.tar.gz), unpack, modify the release number in the PKGBUILD, and run makepkg. It's not that hard. But, don't expect every tarball to build because, as baghera said, the tarballs are auto-generated and aren't guaranteed to build.

commented on 2010-09-22 18:37 (UTC)

@ sl1pkn07 You should look a bit more closer at the link you provided. There are _at least_ two tarballs a day. Those are _not_ all official release, they're official tarballs generated by a buildbot. Sometimes those tarballs are broken and won't compile, while official releases tracked on the blog brings code in a consistent state. Google devs are not lazy updating the blog.

sl1pkn07 commented on 2010-09-22 18:29 (UTC)

you no look the link address? http://build.chromium.org/buildbot/>official</chromium-7.0.530.0.tar.bz2 the google devs is lazy for updating the blog (of course, working in the program) last post in dev channel: "Thursday, September 16, 2010 | 17:52" the #malix post is only one excuse for not updating this package if i have want a new version, download from the SVN... o wait: http://build.chromium.org/buildbot/official/chromium-7.0.517.8.log greetings

terminalmage commented on 2010-09-22 16:53 (UTC)

@sl1pkn07: Please read the comments before flagging out-of-date. The PKGBUILD is only updated for official dev releases. (see http://googlechromereleases.blogspot.com/) If you want a newer version, download the PKGBUILD, update the release number, and build it.

sl1pkn07 commented on 2010-09-22 16:12 (UTC)

http://build.chromium.org/buildbot/official/chromium-7.0.530.0.tar.bz2 yaes, is out of date

Malix commented on 2010-09-16 06:23 (UTC)

@FredBezies this package follows the dev-channel releases, last released is what the pkg offers (see: http://googlechromereleases.blogspot.com/2010/09/dev-channel-update.html). so not out of date.

FredBezies commented on 2010-09-16 06:20 (UTC)

7.0.524.0 is the last one available. Please update.

commented on 2010-09-14 19:55 (UTC)

yay, i have managed to build it but youtube html5 didn't work (BSOD: http://i.imgur.com/9iPhF.png) @hokapoka: Thanks man, like you said it took +/- 1.2 GB hard disk and 3 hours to build (P4, 768 MB RAM)

commented on 2010-09-14 11:16 (UTC)

Mikhail Many thanks for this AUR, works a treat on x86 and x64. I only had one issue, the md5sum wasn't correct. I tried 2 different downloads of the tarball from the buildbot repos and neither was the same as in the PKBUILD. Had to use : c2b142a835eed3ed7fa382612f3d9b4a It appears that the DateTime stamp on the Tarball is later than the last update of the package, maybe they rebuilt it after you created the package. Again, many thanks.

commented on 2010-09-14 11:12 (UTC)

bintang, eh? 10Gb... where did you get that size from? I've just checked my "chromium-dev" build folder and it was 1.2 Gb If you want I can send you a copy of my "chromium-dev-7.0.517.0-1-i686.pkg.tar.xz", it just finished building on my old P4 took ~3 hours. I've never tried to install AURs that have been built on other machines before, so I can't be sure it'll work, you're very welcome to try it. Let me know if you want a link to it.

commented on 2010-09-14 09:34 (UTC)

@hokapoka, alright i'll try ^_^ But do i really need 10 GB hard disk space? I think this is my biggest concern, only 2 gigs left ^_^

commented on 2010-09-14 09:19 (UTC)

bintang, just fire it off overnight regardless how slow your box is it should be complete by the morning, at least it does on one of my old (>5 years old) P4's. That being said there is a another package that appears to have a binary version available, but I had issues some dependencies. It was this one, http://aur.archlinux.org/packages.php?ID=27031, currently flagged out of date. Personally I'd prefer to use this (chromium-dev) from source over the that binary AUR. HTH

commented on 2010-09-13 22:30 (UTC)

Would anyone share their i686 build (or maybe set up a repo.. hehe)? My computer is too slow.

Malix commented on 2010-09-12 20:24 (UTC)

@dcelasun I'd guess it took roughly 20 minutes or so on a phenom2 945, using MAKEFLAGS="-j4" in /etc/makepkg.conf

dcelasun commented on 2010-09-12 20:20 (UTC)

Quick question, how long did it take for you to compile this? On what hardware? It's been a while since I started it and I wonder if, somewhere, I did something wrong.

thermionix commented on 2010-09-11 00:42 (UTC)

Just modified the pkgbuild, don't know if anyones interested. This re-enables phantom tabs, a great feature which was removed recently http://pastebin.com/AaWMx1qX -pkgbuild http://pastebin.com/uiAAc5PV -patch

commented on 2010-09-05 14:34 (UTC)

@joint007 Try to upgrade gcc to 4.5.1-6 and rebuild the package: $ rm -rf src $ makepkg -s -f I also recommend to disable custom CFLAGS/CXXFLAGS if you are using any.

commented on 2010-09-05 14:23 (UTC)

@mvorozhtsov This is second time build for this version chromium. gcc version is 4.5.0-6.

commented on 2010-09-05 13:54 (UTC)

@joint007 Looks like http://code.google.com/p/chromium/issues/detail?id=52748 but that bug is already fixed in the current version. What version of gcc are you using? Your log is rather short, did you build from scratch (i.e. no src dir)?

commented on 2010-09-05 12:18 (UTC)

Hello, i have a problem: http://pastebin.com/jZwZYnsg Ideas?(sorry for my very bad english)

jackoneill commented on 2010-09-01 08:26 (UTC)

@ViOLO: I see, thanks.

Vi0L0 commented on 2010-09-01 01:10 (UTC)

@cantabile: i tested only with .509, but Gillian00 said it's working fine with .503 : https://bbs.archlinux.org/viewtopic.php?pid=818805#p818805

jackoneill commented on 2010-08-31 20:13 (UTC)

@ViOLO: In that bbs post you offer a source tarball for version 7.0.508; is it necessary to use _that_ version (or higher) or will this one (7.0.503) work as well?

Malix commented on 2010-08-30 21:37 (UTC)

@Vi0L0 thanks for the link, I tried experimenting with the commandline switches but managed to trigger some ridiculous memoryleaks. I probably was doing strong overdose with them. anyway, now works like a charm. fluid youtube-video among other things <3

Vi0L0 commented on 2010-08-30 21:04 (UTC)

GPU acceleration in newest chromium works great. At least with catalyst driver, i guess it will also work great with nvidia one. More here: https://bbs.archlinux.org/viewtopic.php?pid=818410#p818410

commented on 2010-08-19 17:46 (UTC)

@JASN If you look a few comments below, you can see that mvorozhtsov updates this package following official release of dev channel (or beta if newer). You can see official release here http://googlechromereleases.blogspot.com/ Version 7.0.497.0 doesn't build because that tarball is automatically generated by buildbot, even if trunk is broken.

commented on 2010-08-19 08:29 (UTC)

New version 7.0.497.0 is out, although it didn't build for me with that PKGBUILD

commented on 2010-08-15 06:39 (UTC)

Updated, the issue with make should be fixed now. See http://code.google.com/p/chromium/issues/detail?id=51288

stricte commented on 2010-08-14 22:08 (UTC)

the same with "chrome/common_constants.target.mk:6: *** brakujący separator. Stop." on x86_64

commented on 2010-08-14 14:48 (UTC)

@gaunt I'm on x86_64 myself, cannot reproduce it. Try to remove the src dir. Is anyone else having this issue?

commented on 2010-08-14 14:36 (UTC)

As of this timestamp, (490.1) doesn't build on x86_64 with the error: chrome/common_constants.target.mk:6: *** missing separator. Stop. Aborting...

commented on 2010-08-12 13:18 (UTC)

@Malix: I package only officially promoted (via one of the Stable/Beta/Dev channels) versions.

Malix commented on 2010-08-12 13:07 (UTC)

Just out of curiosity, there seems to be chromium-6.0.491.0 sources at http://build.chromium.org/buildbot/official/ why did you update this to .472.33 instead of .491.0 ?

commented on 2010-08-12 10:33 (UTC)

Now beta channel have higher version than dev channel (6.0.472.33 vs 6.0.472.25), will you update this package or stick with this version?

jackoneill commented on 2010-08-12 09:23 (UTC)

Thank you mvorozhtsov.

commented on 2010-08-12 09:20 (UTC)

@cantabile 1) yes 2) no, the folders in $HOME are different.

jackoneill commented on 2010-08-12 08:00 (UTC)

Hi! Two questions: 1 - if my cpu has sse2, can I use disable_sse2=0 ? 2 - an earlier comment says "This package can be installed alongside with the chromium package from [extra]." I assume that means there are no file conflicts. But what about the user's profile? Do they both use the same profile? The same folder(s) in my home?

commented on 2010-08-09 13:32 (UTC)

flex is part of base-devel group, so it should _not_ be listed as makedepend, that's the purpose of base-devel. If not, also gcc, make, fakeroot and others should be listed as makedepends.

markuman commented on 2010-08-09 13:29 (UTC)

flex is a make dependecie

commented on 2010-07-29 11:58 (UTC)

Hi, I flagged out of date since 6.0.472.11 is out. Can you also move libgnome-keyring to makedepends/optdepends please?

commented on 2010-07-29 11:57 (UTC)

Hi, I flagged out of date since 6.0.472.11 is out. Can you also move libgnome-keyring to optdepends please?

xenoterracide commented on 2010-07-26 10:15 (UTC)

yeah that worked down to 57 MB thanks

xenoterracide commented on 2010-07-26 10:02 (UTC)

oh didn't think about strip. I'd turned that off to try debugging some other less stable package. I'll try rebuilding with it on.

commented on 2010-07-26 09:58 (UTC)

The problem is that the package doesn't get stripped. Possible causes include: - DEBUG=yes _at the time of building_ - nostrip in options=(...) (possibly via /etc/makepkg.conf) - Failure of actual invocation of strip(1) (not enough disk space?) - Some other magic

xenoterracide commented on 2010-07-26 09:42 (UTC)

-rwxr-xr-x 1 root root 743860615 Jul 24 09:54 /usr/lib/chromium-dev/chromium

xenoterracide commented on 2010-07-26 09:42 (UTC)

I don't have DEBUG enabled

commented on 2010-07-26 09:35 (UTC)

What does 'ls -l /usr/lib/chromium-dev/chromium' print?

xenoterracide commented on 2010-07-26 09:28 (UTC)

I don't have DEBUG enabled

commented on 2010-07-26 09:08 (UTC)

Check your environment. If the environment contains DEBUG=yes, then a debug version of the package will be built.

xenoterracide commented on 2010-07-26 08:58 (UTC)

well in my case... Installed Size : 738648.00 K wtf

commented on 2010-07-26 06:18 (UTC)

@xenoterracide $ pacman -Qi chromium-dev | grep Size Installed Size : 59132.00 K

xenoterracide commented on 2010-07-24 14:13 (UTC)

it says 700MB install is that right? that doesn't seem right? also why install as -dev instead of just providing the regular chromium

OttoA commented on 2010-07-11 15:15 (UTC)

@baghera: it seems as if it is only bundled with google chrome at the moment. It is bundled with the google-chrome-dev package

commented on 2010-07-04 21:57 (UTC)

In 6.0.453.1 log I see * The PDF plugin is now available (turn it on in about:plugins). but I can't find it in about:plugins. Missing file in package maybe?

commented on 2010-07-02 19:08 (UTC)

@ Limoto As you can see in the link you posted, there are at least 2 releases a day, but those tarballs are not intended to be all packaged (they are generated by buildbot). If you check http://googlechromereleases.blogspot.com/ you can see which of those tarballs are actually official releases, intended as tarball with code in a consistent state, and no breakage (apart from known issues).

Limoto commented on 2010-07-02 16:00 (UTC)

http://build.chromium.org/buildbot/official/ and scrool down ;-)

commented on 2010-06-22 23:56 (UTC)

To the one who flagged this out of date: where is the new release? http://googlechromereleases.blogspot.com/ still reports 6.0.437.3 as the last release.

sl1pkn07 commented on 2010-06-20 15:50 (UTC)

wow. i dont touch nothing. now compile ok! :S (without change pkgbuild/makepkg.conf) (with gcc from testing)

sl1pkn07 commented on 2010-06-20 13:41 (UTC)

@mvorozhtsov CXXFLAGS="-march=x86-64 -mtune=generic -O2 -pipe" not have -Werror in this config

commented on 2010-06-20 12:56 (UTC)

@gaelic: Didn't check that - but also didn't change the PKGBUILD at all. So i'd say the PKGBUILD is missing the dependency for its configuration.

commented on 2010-06-20 11:43 (UTC)

Good PKGBUILD!

polslinux commented on 2010-06-20 08:40 (UTC)

there's a new release

commented on 2010-06-20 06:19 (UTC)

@sl1pkn07: You got -Werror in your CXXFLAGS. Check your /etc/makepkg.conf.

sl1pkn07 commented on 2010-06-19 20:36 (UTC)

CXX(target) out/Release/obj.target/glue/webkit/glue/plugins/pepper_device_context_2d.o cc1plus: warnings being treated as errors webkit/glue/plugins/pepper_device_context_2d.cc: En la función ‘PP_Resource pepper::<sinnombre>::Create(PP_Module, int32_t, int32_t, bool)’: webkit/glue/plugins/pepper_device_context_2d.cc:69:12: error: se convierte al tipo ‘long int’ que no es puntero desde NULL webkit/glue/plugins/pepper_device_context_2d.cc:73:12: error: se convierte al tipo ‘long int’ que no es puntero desde NULL make: *** [out/Release/obj.target/glue/webkit/glue/plugins/pepper_device_context_2d.o] Error 1 on arch64bits

gaelic commented on 2010-06-19 18:43 (UTC)

hmm. i think libgnome-keyring is only optional, just like kwallet.

commented on 2010-06-19 13:38 (UTC)

The PKGBUILD is missing the dependency libgnome-keyring and fails in the buildprocess due to gnome-keyring-1.pc not being found. Thanks for the packagebuild anyway!

commented on 2010-06-19 08:46 (UTC)

6.0.437.1-1: Switched to the system ffmpeg.

gaelic commented on 2010-06-17 14:49 (UTC)

Anyone enabled kwallet support in chromium? http://code.google.com/p/chromium/issues/detail?id=12351

gaelic commented on 2010-06-17 14:48 (UTC)

@mvorozhtsov: thx. my /usr/lib/pkgconfig was missing a lot of pc files (and i don't know why) I did a pacman -Ql gtk2 and saw the .pc files of the package, but as mentioned no .pc files in the config_path. A reinstall did it, but not only gkt, also cairo, pango, glib, etc.

jinks commented on 2010-06-17 06:12 (UTC)

Another question: Since the new ffmpeg came out today with WebM support, can we use_system_ffmpeg=1 now?

jinks commented on 2010-06-17 06:11 (UTC)

If you want to play back h264, you have to add "proprietary_codecs=1" to the GYP_DEFINES variable. Also, 6.0.437.1 is out.

commented on 2010-06-17 04:37 (UTC)

@gaelic: There is no "devel" packages on Arch, gtk+-2.0.pc is a part of the gtk2 package (just do $ pacman -Ql gtk2 to see all the contents). Perhaps your environment variables are screwed (PKG_CONFIG_PATH and PKG_CONFIG_* in general).

gaelic commented on 2010-06-16 09:01 (UTC)

I have problems installing because pkg-config is unable to find gtk+-2.0.pc and gconf***.pc. But I don't see any devel packages for gtk+2? Any suggestions?

commented on 2010-06-11 20:48 (UTC)

All packages in base-devel should NOT be listed as (make)depends, base-devel has been created for this reason, it's a requirement for those who want to compile something on arch.

daneel commented on 2010-06-11 20:30 (UTC)

Please add "flex" and "bison" (core/base-devel) to makedepends. Thanks.

commented on 2010-06-11 13:30 (UTC)

Just add export CC=/usr/bin/gcc-4.4 export CXX=/usr/bin/g++-4.4 before GYP_DEFINES and it should work properly.

sl1pkn07 commented on 2010-06-11 13:28 (UTC)

oks. thanks. but what option edit the on the pkgbuild for use gcc44?

commented on 2010-06-11 13:15 (UTC)

You can: 1. Downgrade gcc , gcc-libs , mpfr and libmpc to core version or 2. install gcc44 from aur and build chromium-dev using that gcc version or 3. Wait a few hours for gcc 4.5.0-6 to hit testing :) Allan is gonna fix this soon, commits for 4.5.0-6 are in trunk yet (see http://mailman.archlinux.org/pipermail/arch-commits/2010-June/091885.html )

commented on 2010-06-11 13:14 (UTC)

You can: 1. Downgrade gcc , gcc-libs , mpfr and libmpc to core version or 2. install gcc44 from aur and build chromium-dev using that gcc version or 3. Wait a few hours for gcc 4.5.0-6 to hit testing :) Allan is gonna fix this soon, commits for 4.5.0-6 are in trunk yet.

sl1pkn07 commented on 2010-06-11 13:09 (UTC)

yes., i'm use testing howto fix without uninstall testing? thanks

commented on 2010-06-11 12:52 (UTC)

@ sl1pkn07 Are you using testing? gcc 4.5.0-5 in testing is broken atm , see http://mailman.archlinux.org/pipermail/arch-dev-public/2010-June/017066.html

sl1pkn07 commented on 2010-06-11 12:50 (UTC)

6.0.427 make me all "¡Oh, no!" pages

commented on 2010-05-28 19:31 (UTC)

I also tried to build this package (and -beta) with gcc-4.5.0-3 from [testing], and I got the same results, on x86_64 as well. I didn't try on i686.

commented on 2010-05-28 15:03 (UTC)

The last gcc snapshot that got into [core] (as 4.5.0-2) makes chromium go OOM on any page. At least on x86_64 with -O2 (Debug version works fine). Is it the case for i686? For now I added gcc44 to makedeps on x86_64.

commented on 2010-05-28 11:15 (UTC)

Hi mvorozhtsov, I just tried to build this package in a clean chroot, and during sources checkout I got an error about missing git executable. Adding git to makedeps and rebuilding fixed the issue.

commented on 2010-05-23 04:58 (UTC)

why is it SO LARGE???, with my connection this is painful PAINFUL

commented on 2010-05-17 09:04 (UTC)

This package can be installed alongside with the chromium package from [extra].