Package Details: palemoon

Git Clone URL: (read-only)
Package Base: palemoon
Description: Open source web browser based on Firefox focusing on efficiency.
Upstream URL:
Keywords: browser goanna web
Licenses: GPL, MPL, LGPL
Submitter: artiom
Maintainer: WorMzy
Last Packager: WorMzy
Votes: 121
Popularity: 1.574377
First Submitted: 2014-06-05 10:54
Last Updated: 2019-07-04 22:45

Latest Comments

« First ‹ Previous ... 2 3 4 5 6 7 8 9 10 11 12 ... Next › Last »

fabertawe commented on 2017-11-12 11:29

I don't build in a clean chroot! CPU is an Intel i7 4790k, 'makepkg.conf' reads...
CFLAGS="-march=native -mtune=generic -O2 -pipe -fstack-protector-strong -fno-plt"

'c++config.h' (untouched by me) shows...
#define _GLIBCXX_USE_C99_MATH 1

sekret commented on 2017-11-12 11:07

Me too, I recently switched my HD into another laptop, maybe it makes a difference if I build gcc with this CPU? I don't use custom build flags, so I can't imagine that's the reason, but well, I'll try and report back.

WorMzy commented on 2017-11-12 11:04

Okay, that's interesting. My clean chroot build also fails with gcc5.5, with the same errors as Wolf. It'd be interesting to see if we could identify what causes this build failure. Runical amd fabertawe, I'm guessing you don't modify c++config.h, so the bug report I suspected is probably a red herring (although that edit results in a successful build for me). Just as a sanity check, could you all check your gcc5 package's /usr/lib/gcc/x86_64-pc-linux-gnu/5.5.0/include/c++/x86_64-pc-linux-gnu/bits/c++config.h and see whether _GLIBCXX_USE_C99_MATH is defined (check around line #1346).

Also, if you don't mind sharing what sort of CPU you use, and whether you use any custom C{,XX,PP}FLAGS, that might help.

Either leave a comment here, ping me on irc (WorMzy@freenode), or drop me an email. Any help would be appreciated.

@sekret: on closer inspection, your build failure is different to the ones me and wolf are (were?) experiencing. It's interesting to see three different outcomes from clean-chroot builds!

I'm going to try rebuilding gcc5.5 again and see if palemoon builds with the new package.

sekret commented on 2017-11-12 09:17

I build in a clean chroot too and palemoon failed to build with gcc5 from the AUR.

runical commented on 2017-11-12 09:06

@fabertawe: I think this is his specific issue. GCC5 built just fine for me and the same goes for palemoon. I do build in a clean chroot though, so that removes a lot of the issues anyway.

fabertawe commented on 2017-11-11 13:19

I'm confused... Pale Moon builds for me with gcc5 5.5.0-2 from the AUR. Is it failing for everyone else or is wolf's a specific issue?

wolf commented on 2017-11-11 11:37

Can confirm 5.4 from works just fine, maybe you could add info about this (5.5 not working) into that pinned comment :)

WorMzy commented on 2017-11-09 22:22

Yes, unfortunately I haven't had time to do anything about this yet.

If you want to help out, please check other packages which use gcc5 to build and see if they fail with the same errors when you use v5.5.0. Then see if they build successfully when you edit /usr/lib/gcc/x86_64-pc-linux-gnu/5.5.0/include/c++/x86_64-pc-linux-gnu/bits/c++config.h, changing line #1346 to read:

#define _GLIBCXX_USE_C99_MATH 1

Alternatively, if you can find a way to override c++config.h on a case-by-case basis, that doesn't involve editing source code (upstream is getting arsey about from-source builds as it is [1]), that would be a big help.


sekret commented on 2017-11-09 20:37

Trying to build palemoon 27.6.0-1 with gcc5 5.5.0-2 from the AUR still fails.

39:00.72 make[5]: *** [/build/palemoon/src/Pale-Moon/config/ Unified_cpp_dom_canvas0.o] Error 1
39:00.72 make[4]: *** [/build/palemoon/src/Pale-Moon/config/ dom/canvas/target] Error 2
39:00.72 make[4]: *** Waiting for unfinished jobs....
39:00.72 Unified_cpp_dom_media3.o
39:11.72 In file included from /build/palemoon/src/pmbuild/dom/media/Unified_cpp_dom_media1.cpp:20:0:
39:11.72 /build/palemoon/src/Pale-Moon/dom/media/MP3Demuxer.cpp: At global scope:
39:11.72 Warning: -Wunused-variable in /build/palemoon/src/Pale-Moon/dom/media/MP3Demuxer.cpp: ‘mozilla::mp3::vbr_header::TYPE_STR’ defined but not used
39:11.72 /build/palemoon/src/Pale-Moon/dom/media/MP3Demuxer.cpp:893:20: warning: ‘mozilla::mp3::vbr_header::TYPE_STR’ defined but not used [-Wunused-variable]
39:11.72 static const char* TYPE_STR[3] = {"NONE", "XING", "VBRI"};
39:11.72 ^
39:32.51 UnifiedBindings7.o
39:36.68 UnifiedBindings8.o
39:37.43 libdom_media.a.desc
39:37.55 UnifiedBindings9.o
40:29.59 StructuredClone.o
40:51.58 Unified_cpp_dom_bindings0.o
41:15.21 libdom_bindings.a.desc
41:15.26 make[3]: *** [/build/palemoon/src/Pale-Moon/config/ compile] Error 2
41:15.27 make[2]: *** [/build/palemoon/src/Pale-Moon/config/ default] Error 2
41:15.27 make[1]: *** [/build/palemoon/src/Pale-Moon/ realbuild] Error 2
41:15.27 make: *** [ build] Error 2
41:15.31 114 compiler warnings present.
41:15.82 Notification center failed: Install the python dbus module to get a notification when the build finishes.
==> ERROR: A failure occurred in build().

I'll try with the old gcc5...

WorMzy commented on 2017-10-31 13:02

I think this is related to this bug: which is actually a bug in glibc 2.26 (gcc 5.4.0 has _GLIBCXX_USE_C99_MATH defined in c++config.h, so doesn't trigger it)

Should be fixed in glibc 2.27, but since that's many months away (Feb 2018), I'll see if I can find time to find a workaround for the interim. This workaround may be prodding the gcc5 maintainer and seeing if they'd readd the _GLIBCXX_USE_C99_MATH definition to c++config.h until glibc 2.27 lands. \o/

Temporary workaround: use the old gcc 5.4.0 packages from the ALA