Package Base Details: linux-lqx

Git Clone URL: (read-only, click to copy)
Keywords: bbr2 bfq futex muqss proton zen
Submitter: akurei
Maintainer: sir_lucjan (damentz)
Last Packager: damentz
Votes: 147
Popularity: 2.44
First Submitted: 2011-08-08 16:08 (UTC)
Last Updated: 2022-08-18 18:39 (UTC)

Pinned Comments

damentz commented on 2020-08-31 15:22 (UTC) (edited on 2021-12-21 18:25 (UTC) by damentz)

Official binaries of linux-lqx, linux-lqx-headers, and linux-lqx-docs are now available:

Signing key import instructions: sudo pacman-key --keyserver hkps:// --recv-keys 9AE4078033F8024D && sudo pacman-key --lsign-key 9AE4078033F8024D

Latest Comments

pipes80 commented on 2022-08-03 18:03 (UTC)

Spectre V2 : WARNING: Unprivileged eBPF is enabled with eIBRS on

after upgrade on linux-lqx 5.18.16 i cannot boot for problrm with nvidia-drivers, i search problem with journactl and i have find that message too. it's ok or it's a problem

MajorBarnulf commented on 2022-07-20 00:09 (UTC)


Oh; I did not knew about that file, my bad... Thank you for your time anyway.

sir_lucjan commented on 2022-07-19 16:55 (UTC)


I hope you know that this can be set globally in makepkg.conf

MajorBarnulf commented on 2022-07-19 16:08 (UTC)

hi, would it be possible to have the kernel be compiled on multiple threads by default? I think it will benefit everyone

lnoferin commented on 2022-05-08 13:16 (UTC)

@damentz yes, I found the problem was in my configuration for the keyservers. Thanks the same!

damentz commented on 2022-05-07 23:31 (UTC)

@lnoferin, sounds like the keyserver was down or your network connection is blocking access. The ability pull down keys is also required for adding third party repos unless you disable signing.

We can't solve your network issues, so that's something you'll need to figure out on your end.

lnoferin commented on 2022-04-28 10:34 (UTC) (edited on 2022-04-28 10:34 (UTC) by lnoferin)

@sir_lucjan Tried but it failed:

gpg --recv-keys 9AE4078033F8024D && gpg --lsign-key 9AE4078033F8024D
gpg: keyserver receive failed: Connection refused

sir_lucjan commented on 2022-04-28 09:41 (UTC) (edited on 2022-04-28 09:41 (UTC) by sir_lucjan)


Try gpg --recv-keys 9AE4078033F8024D && gpg --lsign-key 9AE4078033F8024D as normal user.

lnoferin commented on 2022-04-28 09:37 (UTC)

I cannot build the package on a manjaro because this error:

pamac build linux-lqx                          1 ✘  1m 3s  
Cloning linux-lqx build files...
Generating linux-lqx information...
Checking linux-lqx dependencies...
Resolving dependencies...
Checking inter-conflicts...

To build (1):
  linux-lqx  5.17.5.lqx1-1    AUR

Edit build files : [e] 
Apply transaction ? [e/y/N] y

Building linux-lqx...
==> Making package: linux-lqx 5.17.5.lqx1-1 (gio 28 apr 2022, 11:00:20)
==> Checking runtime dependencies...
==> Checking buildtime dependencies...
==> Retrieving sources...
  -> Found linux-5.17.tar.xz
  -> Found linux-5.17.tar.sign
  -> Found 5.17-9.tar.gz
==> Validating source files with sha512sums...
    linux-5.17.tar.xz ... Passed
    linux-5.17.tar.sign ... Skipped
    5.17-9.tar.gz ... Passed
==> Verifying source file signatures with gpg...
    linux-5.17.tar ... FAILED (unknown public key 38DBBDC86092693E)
==> ERROR: One or more PGP signatures could not be verified!
Error: Failed to build linux-lqx

I followed the pinned comment in this page but nothing changed:

pacman-key --keyserver hkps:// --recv-keys 9AE4078033F8024D && pacman-key --lsign-key 9AE4078033F8024D
gpg: key 9AE4078033F8024D: "Steven Barrett <>" not changed
gpg: Total number processed: 1
gpg:              unchanged: 1
  -> Locally signed 1 keys.
==> Updating trust database...
gpg: next trustdb check due at 2022-05-02

damentz commented on 2022-04-08 22:37 (UTC)

I failed over to using my build scripts on my local system. Liquorix repo is up-to-date again. Builds are performed clean room in docker and produce the same artifacts as my cloud build project.

damentz commented on 2022-04-08 20:33 (UTC)

If anyone's wondering why the official packages are not updated, I'm working with AWS support on my account that does the binary builds, my ability to run CodeBuild projects got randomly revoked. I'll post a notification once the builds are back online and repository updated.

damentz commented on 2022-03-09 13:00 (UTC)

Sorry about that, I had a short deadline to push out an update and this one was more complicated due to a breakage with PDS. In my rush, I missed adding the patch in git itself, but didn't notice due to the way my docker builds work for Debian and Ubuntu.

New update is out, appears to be building. Builds for Debian and Ubuntu ran yesterday successfully so I'm assuming this one is good too.

AlPiotrowicz commented on 2022-03-09 11:54 (UTC)

Confirming the error reported by @Borov

Borov commented on 2022-03-09 10:06 (UTC)

Please check

Patching sources with v5.16.13-lqx2.patch patch: **** Can't open patch file /home/user/.cache/yay/linux-lqx/src/liquorix-package-5.16-15/linux-liquorix/debian/patches/zen/v5.16.13-lqx2.patch : No such file or directory

AlPiotrowicz commented on 2022-01-31 21:24 (UTC)

Confirming it works

sir_lucjan commented on 2022-01-31 21:00 (UTC)

Try running makepkg -Rs unless you have deleted the src and pkg directories

AlPiotrowicz commented on 2022-01-31 20:51 (UTC)

@sir_lucjan Is it possible to resume the failed (in the place it failed that matter) compilation some easy way?

sir_lucjan commented on 2022-01-31 20:21 (UTC)

Let us know if it works.

Nie ma za co :D

AlPiotrowicz commented on 2022-01-31 20:20 (UTC)

@sir_lucjan Dzieki za ekspresowy fix :)

sir_lucjan commented on 2022-01-31 20:13 (UTC) (edited on 2022-01-31 20:14 (UTC) by sir_lucjan)


Sorry for the mistake, I reversed the last change.

P.S Komunikaty błędów zgłaszaj jednak po angielsku, nie każdy developer mówi w naszym rodzimym języku :D

AlPiotrowicz commented on 2022-01-31 20:00 (UTC) (edited on 2022-01-31 20:17 (UTC) by AlPiotrowicz)

Hello, I experience:

==> Rozpoczynanie package_linux-lqx-headers()...
Installing build files...
install: cannot stat 'tools/bpf/resolve_btfids/resolve_btfids': No such file or directory
==> BŁĄD: Wystąpił błąd w package_linux-lqx-headers().
Probably PKGBUILD needs one more if statement

Althorion commented on 2022-01-16 17:59 (UTC)

The current version patches and builds correctly. No idea what, if anything, was wrong with the previous one…

damentz commented on 2022-01-13 23:58 (UTC) (edited on 2022-01-14 00:16 (UTC) by damentz)

@Althorion mine also patches and begins building:

Maybe github is sending you a different version of the tagged archive? That or your ISP is intercepting your download and sending you something else.

EDIT: In your asciinema, I see that your makepkg passes the integrity checks, so most likely your system is corrupting the extracted contents when extracting either archive.

sir_lucjan commented on 2022-01-12 21:34 (UTC)

I can not confirm.

Althorion commented on 2022-01-12 21:14 (UTC) (edited on 2022-01-12 21:14 (UTC) by Althorion)

@damentz I am. I’ve both tried makepkg --clean, and removing the folder and cloning the package from scratch. Previous versions still build fine, if I check them out instead.

That’s how it looks on my end:

damentz commented on 2022-01-12 17:34 (UTC)

@theriddick, I missed your last comment; Liquorix uses PDS as the process scheduler with a 1000hz tick rate and 2ms scheduler deadline, so depending on how prone latte dock is to race conditions (things happening out of order, artificial delays on errors, etc), Latte Dock will initialize slower. It might be worth your time to submit a bug report since it's most certainly an assumption depending on an order of operations CFS guarantees most of the time, but is invalid with PDS.

MuQSS (and as an extension PDS), tends to expose programs with bugs in multithreading far more often than CFS. This goes for buggy kernel drivers too. Fortunately, things have gotten a lot better than when MuQSS was first introduced and PDS appears to overall have less show stopping bugs than its predecessor.

damentz commented on 2022-01-12 17:29 (UTC)

@Althorion, are you building from a clean environment? Last automated build ran successfully 21 hours ago:

Althorion commented on 2022-01-12 17:09 (UTC)

The current version (5.15.14.lqx1-1) fails to patch properly:

theriddick commented on 2021-11-02 03:53 (UTC) (edited on 2021-11-02 03:54 (UTC) by theriddick)

I notice that latte-dock takes a few extra seconds to load with this kernel. Not a show stopper but just a odd curiosity issue. Also my wireless keyboard is disabled for a few seconds on login also (happens with default kernel), wonder if that is related to something.

(These things don't happen with xanmod btw)

damentz commented on 2021-11-02 03:18 (UTC)

@theriddick, once Liquorix is rebased to 5.15, it will have it.

In the meantime, looks like there's a DKMS package that gets you the same functionality but a different module name, ntfs3:

theriddick commented on 2021-11-02 02:50 (UTC) (edited on 2021-11-02 03:53 (UTC) by theriddick)

Guessing this does not include NTFS3? (yes using ntfs3-dkms atm)

JoseskVolpe commented on 2021-10-10 19:21 (UTC)

¡It works, thanks! ÛwÛ

damentz commented on 2021-09-27 01:10 (UTC)

@JoseskVolpe, what happens when you read the pinned comment and follow its instructions?

JoseskVolpe commented on 2021-09-27 01:02 (UTC)

¿Can't we have a pre-compiled package? This takes forever to compile every update lol

brunomiguel commented on 2021-09-25 17:38 (UTC)

I was able to build it in a clean chroot. The error, though, persisted with a normal build. Yet, after rebooting, it built normally.

damentz commented on 2021-09-25 15:47 (UTC) (edited on 2021-09-25 15:48 (UTC) by damentz)

@brunomiguel, most likely a unique configuration on your system is breaking the build. I recommend trying to build in a new vanilla system, chroot, or container.

The binary builds (as listed in the pinned comment), always build from clean room using the official Arch Linux docker image and have had no issues. Same goes for when I test using local builds on my desktop.

If you want to run Liquorix but can't build for some reason, I highly recommend you using the official Liquorix repository for Arch.

brunomiguel commented on 2021-09-25 15:43 (UTC) (edited on 2021-09-25 16:37 (UTC) by brunomiguel)

@vp1981 I have 12GB of RAM and 200GiB of disk space (with only 55GB used, more than enough to compile this kernel). I even monitored the RAM and disk space used when compiling and it never ran out of RAM or disk space.

update: i can build it in a clean chroot

vp1981 commented on 2021-09-25 09:08 (UTC)

@brunomiguel, wild guess: do you have enough memory? Free space?

brunomiguel commented on 2021-09-23 02:45 (UTC)

I made no changes, I'm using the original PKGBUILD

vp1981 commented on 2021-09-23 02:04 (UTC)

Hello brunomiguel.

I have just built linux-lqx with GCC. Are you using exactly the same PKGBUILD and config as the AUR package or did you add/change something yourself?

brunomiguel commented on 2021-09-23 01:30 (UTC)

I'm building the latest version but I get the following error

ZSTD22  arch/x86/boot/compressed/vmlinux.bin.zst
  CC      arch/x86/boot/compressed/misc.o
FATAL: modpost: drivers/net/wwan/rpmsg_wwan_ctrl: sizeof(struct rpmsg_device_id)=32 is not a modulo of the size of section __mod_rpmsg__<identifier>_device_table=120.
Fix definition of struct rpmsg_device_id in mod_devicetable.h
make[1]: *** [scripts/Makefile.modpost:150: modules-only.symvers] Erro 1
make: *** [Makefile:1770: modules] Erro 2

damentz commented on 2021-09-20 00:35 (UTC)

@vp1981, that is correct, Liquorix now uses PDS. It's spiritually very similar to MuQSS.

From my early testing, single threaded performance is much higher than MuQSS, but multithreaded performance suffers a bit. However, the delta is not big enough to be an issue for Liquorix's target behavior and the single threaded performance is arguably more important for common activities most people do on their systems.

vp1981 commented on 2021-09-19 23:48 (UTC)

So, there is no more MUQSS, right?

vp1981 commented on 2021-08-22 12:40 (UTC)

Hello. Something strange is going on with key servers. I couldn't find key 9AE4078033F8024D on but found it on, didn't try to import it using that server though.

damentz commented on 2021-06-22 01:15 (UTC)

Wow, that's unfortunate. I just updated the instructions to use instead.

arnamak commented on 2021-06-22 00:03 (UTC)

It would appear that, as of today, the keyserver listed in the pinned comment no longer works.

According to the status page for, this is due to further GDPR takedown requests.

Is there an alternate keyserver we can use?

damentz commented on 2021-06-07 14:55 (UTC)

Yes, that is strange. This issue would come back occasionally when updates were pulled from upstream Arch, but that was due to pulling a function that would try to dynamically determine the kernel version.

This time around nothing like that is happening, but whatever is determining the version isn't pulling from $pkgver like it should.

@sirlucjan, do you have any ideas? Maybe upstream Arch has already deviated and we need to pull its updates.

ryshglene commented on 2021-06-07 06:07 (UTC)

Upgraded to 5.12.9.lqx2-1, and yet the kernel name remains 5.12.9-lqx1-1-lqx?

damentz commented on 2021-06-05 22:02 (UTC) (edited on 2021-06-05 22:02 (UTC) by damentz)

Kido commented on 2021-06-05 16:25
RTL8111/8168/8411 is broken in linux-lqx 5.12.9.lqx2-1

What is broken about these devices and since which kernel?

Kido commented on 2021-06-05 21:25 (UTC)

RTL8111/8168/8411 is broken in linux-lqx 5.12.9.lqx2-1

freenestor commented on 2021-03-11 02:55 (UTC)

makepkg hangs at

... Removing scripts/mod/modpost.o Removing scripts/mod/file2alias.o Removing scripts/mod/sumversion.o Removing scripts/kconfig/confdata.o Removing scripts/kconfig/preprocess.o Removing scripts/kconfig/expr.o Removing scripts/kconfig/util.o Removing scripts/kconfig/ Removing scripts/kconfig/symbol.o Removing scripts/kconfig/lexer.lex.o Removing scripts/kconfig/conf.o Stripping build tools... #hangs here

Any idea?

Willy1992 commented on 2021-03-08 14:49 (UTC)

@damentz yeah i saw it, recompiled with default config and it works, seems like using my current kernel config broke it, perhaps i need to take a look with nconfig

damentz commented on 2021-03-07 19:59 (UTC)

@Willy1992, the binaries built successfully 2 hours ago (check pinned comment). Are you not using the configuration in the AUR package?

Willy1992 commented on 2021-03-07 19:42 (UTC)

A few hours ago compiled succesfully 5.11-3, but 5.11-4 fails with:

kernel/sched/MuQSS.c:141:36: error: 'CONFIG_SHARERQ' undeclared here (not in a function); did you mean 'CONFIG_SPARSE_IRQ'? 141 | static int rqshare __read_mostly = CONFIG_SHARERQ; / Default RQSHARE_MC / | ^~~~~~~~~~~~~~ | CONFIG_SHMEM

brunomiguel commented on 2021-03-05 09:56 (UTC)

I've installed this kernel a few minutes ago, but I can't get it to boot (I'm using GRUB, if that's relevant, and compressing the images with LZ4 -1). In GRUB, it doesn't pass the "Loading initial ramdisk ..." message, although the file is the correct place.

sir_lucjan commented on 2021-02-10 19:26 (UTC)

Binutils 2.36.1 should be released soon and the bug may be fixed there. I haven't checked, but if someone wants to, she/he should look through the sources.

lewiji commented on 2021-02-10 18:59 (UTC)

@sir_lucjan thanks for the link, I will follow that bug.

I do use qemu/virtmanager as well but use Virtualbox sometimes for some specific old Win95/Win3.1 games and software as the integration is much better.

For now I can just use another kernel version for that until they've figured it out. Cheers.

sir_lucjan commented on 2021-02-10 12:10 (UTC)

I'm definitely not sure about that, which is why I wrote "probably" :) I haven't used VB for years so I personally haven't encountered this error.

Perhaps it will go better with virt-manager? It's also a GUI program, so there's always an alternative.

Anarconda commented on 2021-02-10 12:06 (UTC)

@sir_lucjan Are you sure about that? In my /var/lib/dkms/vboxhost you can see this:

drwxr-xr-x 1 root root 144 10.02.2021 08:17 6.1.18_OSE/ lrwxrwxrwx 1 root root 33 08.02.2021 21:48 kernel-5.10.14-arch1-1-x86_64 -> 6.1.18_OSE/5.10.14-arch1-1/x86_64/ lrwxrwxrwx 1 root root 46 08.02.2021 13:47 kernel-5.10.14-hardened1-1-hardened-x86_64 -> 6.1.18_OSE/5.10.14-hardened1-1-hardened/x86_64/ lrwxrwxrwx 1 root root 36 08.02.2021 21:46 kernel-5.10.14-zen1-1-zen-x86_64 -> 6.1.18_OSE/5.10.14-zen1-1-zen/x86_64/

In other words, for my other installed kernels the module it's been created.

sir_lucjan commented on 2021-02-10 11:48 (UTC)

Probably the bug has to do with the latest binutils. Similar problems occur on other kernels as well.

lewiji commented on 2021-02-10 11:45 (UTC)

Same issue as Anarconda, the make.log says:

DKMS make.log for vboxhost-6.1.18_OSE for kernel 5.10.14-lqx2-1-lqx (x86_64)
Wed 10 Feb 11:41:21 GMT 2021
make: Entering directory '/usr/lib/modules/5.10.14-lqx2-1-lqx/build'
  CC [M]  /var/lib/dkms/vboxhost/6.1.18_OSE/build/vboxnetadp/linux/VBoxNetAdp-linux.o
  CC [M]  /var/lib/dkms/vboxhost/6.1.18_OSE/build/vboxnetadp/VBoxNetAdp.o
  CC [M]  /var/lib/dkms/vboxhost/6.1.18_OSE/build/vboxnetflt/linux/VBoxNetFlt-linux.o
  CC [M]  /var/lib/dkms/vboxhost/6.1.18_OSE/build/vboxnetflt/VBoxNetFlt.o
  CC [M]  /var/lib/dkms/vboxhost/6.1.18_OSE/build/vboxnetflt/SUPR0IdcClient.o
  CC [M]  /var/lib/dkms/vboxhost/6.1.18_OSE/build/vboxdrv/linux/SUPDrv-linux.o
  CC [M]  /var/lib/dkms/vboxhost/6.1.18_OSE/build/vboxnetflt/SUPR0IdcClientComponent.o
  CC [M]  /var/lib/dkms/vboxhost/6.1.18_OSE/build/vboxdrv/SUPDrv.o
  CC [M]  /var/lib/dkms/vboxhost/6.1.18_OSE/build/vboxnetflt/linux/SUPR0IdcClient-linux.o
  CC [M]  /var/lib/dkms/vboxhost/6.1.18_OSE/build/vboxdrv/SUPDrvGip.o
  CC [M]  /var/lib/dkms/vboxhost/6.1.18_OSE/build/vboxdrv/SUPDrvSem.o
  CC [M]  /var/lib/dkms/vboxhost/6.1.18_OSE/build/vboxdrv/SUPDrvTracer.o
  CC [M]  /var/lib/dkms/vboxhost/6.1.18_OSE/build/vboxdrv/SUPLibAll.o
  CC [M]  /var/lib/dkms/vboxhost/6.1.18_OSE/build/vboxdrv/r0drv/alloc-r0drv.o
  CC [M]  /var/lib/dkms/vboxhost/6.1.18_OSE/build/vboxdrv/r0drv/initterm-r0drv.o
  CC [M]  /var/lib/dkms/vboxhost/6.1.18_OSE/build/vboxdrv/r0drv/memobj-r0drv.o
  CC [M]  /var/lib/dkms/vboxhost/6.1.18_OSE/build/vboxdrv/r0drv/mpnotification-r0drv.o
  CC [M]  /var/lib/dkms/vboxhost/6.1.18_OSE/build/vboxdrv/r0drv/powernotification-r0drv.o
  CC [M]  /var/lib/dkms/vboxhost/6.1.18_OSE/build/vboxdrv/r0drv/linux/assert-r0drv-linux.o
  CC [M]  /var/lib/dkms/vboxhost/6.1.18_OSE/build/vboxdrv/r0drv/linux/alloc-r0drv-linux.o
  CC [M]  /var/lib/dkms/vboxhost/6.1.18_OSE/build/vboxdrv/r0drv/linux/initterm-r0drv-linux.o
/var/lib/dkms/vboxhost/6.1.18_OSE/build/vboxdrv/SUPDrvTracer.o: warning: objtool: .text+0x7: indirect jump found in RETPOLINE build
/var/lib/dkms/vboxhost/6.1.18_OSE/build/vboxdrv/SUPDrvTracer.o: warning: objtool: supdrvTracerProbeFireStub() is missing an ELF size annotation
/var/lib/dkms/vboxhost/6.1.18_OSE/build/vboxdrv/SUPDrvTracer.o: warning: objtool: missing symbol for insn at offset 0x0

make[2]: *** [scripts/ /var/lib/dkms/vboxhost/6.1.18_OSE/build/vboxdrv/SUPDrvTracer.o] Error 255
make[2]: *** Deleting file '/var/lib/dkms/vboxhost/6.1.18_OSE/build/vboxdrv/SUPDrvTracer.o'
make[2]: *** Waiting for unfinished jobs....
  LD [M]  /var/lib/dkms/vboxhost/6.1.18_OSE/build/vboxnetadp/vboxnetadp.o
  LD [M]  /var/lib/dkms/vboxhost/6.1.18_OSE/build/vboxnetflt/vboxnetflt.o
make[1]: *** [scripts/ /var/lib/dkms/vboxhost/6.1.18_OSE/build/vboxdrv] Error 2
make: *** [Makefile:1801: /var/lib/dkms/vboxhost/6.1.18_OSE/build] Error 2
make: Leaving directory '/usr/lib/modules/5.10.14-lqx2-1-lqx/build'

Anarconda commented on 2021-02-10 07:22 (UTC)

Usually vboxhost module is build correctly but with 5.10.14 series I'm getting this:

==> dkms install --no-depmod -m vboxhost -v 6.1.18_OSE -k 5.10.14-lqx2-1-lqx Error! Bad return status for module build on kernel: 5.10.14-lqx2-1-lqx (x86_64) Consult /var/lib/dkms/vboxhost/6.1.18_OSE/build/make.log for more information. ==> Warning, `dkms install --no-depmod -m vboxhost -v 6.1.18_OSE -k 5.10.14-lqx2-1-lqx' returned 10

What can I do? Thanks in advance.

damentz commented on 2021-01-30 18:13 (UTC)

@Anarconda, the .12 update was pretty small with no block patches. You might have a bad drive that only sometimes has issues at boot. I recommend checking the smart data on your drives to see if anything is primed to fail soon.

Anarconda commented on 2021-01-30 17:32 (UTC)

Just to let you know the following messages are shown by dmesg with .12

"[ 161.707995] ata8.00: failed command: READ FPDMA QUEUED [ 161.707999] ata8.00: cmd 60/00:18:20:b7:84/0a:00:2c:00:00/40 tag 3 ncq dma 1310720 in res 40/00:20:20:c1:84/00:00:2c:00:00/40 Emask 0x10 (ATA bus error) [ 161.708000] ata8.00: status: { DRDY } [ 161.708001] ata8.00: failed command: READ FPDMA QUEUED [ 161.708004] ata8.00: cmd 60/00:20:20:c1:84/02:00:2c:00:00/40 tag 4 ncq dma 262144 in res 40/00:20:20:c1:84/00:00:2c:00:00/40 Emask 0x10 (ATA bus error)"

reverting to .11 messages are gone. Thanks for your efforts.

gardotd426 commented on 2020-12-31 05:06 (UTC) (edited on 2020-12-31 05:17 (UTC) by gardotd426)

This is likely a Virtualbox issue, not a linux-lqx issue. They have to have their modules be able to be built against this version of the kernel. Usually with custom kernels it takes a bit. I've noticed it for well over a year when using kernels like linux-zen, linux-tkg, etc., and sometimes even regular linux, for the first few versions of a new kernel (like up to X.Y.7 or so)

Angel_Caido commented on 2020-12-31 04:40 (UTC) (edited on 2020-12-31 04:41 (UTC) by Angel_Caido)

Virtualbox Machines not working properly after upgrading to the most recent version of Linux-lqx [5.10 and up].
Loading modules: modprobe: FATAL: Module vboxnetadp not found in directory /lib/modules/5.10.4-lqx1-1-lqx
modprobe: FATAL: Module vboxnetflt not found in directory /lib/modules/5.10.4-lqx1-1-lqx
modprobe: FATAL: Module vboxdrv not found in directory /lib/modules/5.10.4-lqx1-1-lqx

Anarconda commented on 2020-12-29 08:12 (UTC)

I can confirm messages are gone. Thanks again.

damentz commented on 2020-12-29 01:58 (UTC)

@Anarconda, next release (5.10.3.lqx2-1), should fix it. The issue appears to be a BPF feature that depends on userspace tool built with the kernel. The error itself is completely benign otherwise.

sir_lucjan commented on 2020-12-29 01:41 (UTC) (edited on 2020-12-29 01:42 (UTC) by sir_lucjan)


Currently, the package is managed by the creator of patchset liquorix, damentz, co-maintainer. I will let him know if it still does not work. Unfortunately - I don't use patchset liquorix but my solutions. I haven't observed this error at my system and we will have to look into it if the problem persists - currently AUR notifications come with a very long delay.

Anarconda commented on 2020-12-28 20:41 (UTC)

@sir_lucjan. After update to version 5.10.3 and reboot I can read the following message:

libbpf: failed to find valid kernel BTF libbpf: Error loading vmlinux BTF: -3 libbpf: failed to load object 'iterators_bpf' libbpf: failed to load BPF skeleton 'iterators_bpf': -3 Failed to load could be due to wrong endianness

What can I do solve this? Thanks in advance. Regards.

sir_lucjan commented on 2020-12-14 14:48 (UTC)


It is NOT obsolete. The lqx patchset will only appear when a new MUQSS is released.

damentz commented on 2020-11-25 02:47 (UTC)

error: required key missing from keyring

both downloading from custom repo and compiling from source tells me i dont have pgp keys

@bonkmaykrq, have you tried getting the keys? The pinned comment gives you the template to start with. If you can't figure it out, I recommend using the binaries and following the key import instructions pinned as the first comment.

bonkmaykrq commented on 2020-11-24 16:25 (UTC)

error: required key missing from keyring

both downloading from custom repo and compiling from source tells me i dont have pgp keys

LibbleGlobGlob commented on 2020-11-15 04:42 (UTC)

@damentz Eyy, that worked really well! I also fixed some changes in the conf where I used systemd hook instead of the udev and the cat compression for "faster boot time" that didn't really worked for me. Thanks !

damentz commented on 2020-11-14 19:13 (UTC)

@LibbleGlobGlob, that sounds like you're missing autofs4 in your initramfs. Try adding the module to your mkinitcpio.conf and rebuilding with mkinitcpio -P.

LibbleGlobGlob commented on 2020-11-14 08:40 (UTC)

I'm getting a "Failed to insert module 'autofs4': No such file or directory" during boot. But it is functional and works as expected. How do I get rid of this message though?

gardotd426 commented on 2020-10-31 13:16 (UTC)

It's not "incompatible." It works fine, the only thing missing is CUDA.

Even then, you can get full functionality back if you use TK-Glitch's nvidia-all PKGBUILD, CUDA and all.

@samuk190 the real issue is likely that you haven't rebuilt your Nvidia drivers after installing a new kernel version. You always have to do this, when going from 5.8 to 5.9, etc.

Re-install the Nvidia drivers. If that doesn't work, use

sir_lucjan commented on 2020-10-31 12:44 (UTC)

samuk190 commented on 2020-10-31 12:18 (UTC)

Error: nvidia fb hangs on boot. Manjaro "mikah"

I guess its something related with nvidia but I tested this kernel on xubuntu before and I didnt have any issues.

damentz commented on 2020-09-28 15:05 (UTC)

@xuanruiqi, should be fixed in next release, you can look at this github issue for reference:

xuanruiqi commented on 2020-09-28 14:27 (UTC)

OK, I found the GitHub issue: Apparently you can't enable AMD microcode support w/o enabling Intel microcode support. Weird.

xuanruiqi commented on 2020-09-28 14:25 (UTC)

I'm getting this while trying to compile the kernel:

ld: arch/x86/kernel/cpu/microcode/core.o: in function `reload_store':
core.c:(.text+0x166): undefined reference to `force_ucode_load'
ld: core.c:(.text+0x196): undefined reference to `force_ucode_load'
ld: arch/x86/kernel/cpu/microcode/core.o: in function `reload_store.cold':
core.c:(.text.unlikely+0x22): undefined reference to `force_ucode_load'
ld: core.c:(.text.unlikely+0xa4): undefined reference to `force_ucode_load'
ld: core.c:(.text.unlikely+0xda): undefined reference to `force_ucode_load'

Anyone having the same problem?

damentz commented on 2020-09-23 15:56 (UTC)

@thaewrapt, added cpio back to makedepends.

thaewrapt commented on 2020-09-23 11:48 (UTC)

For reference:

thaewrapt commented on 2020-09-23 11:46 (UTC) (edited on 2020-09-23 11:46 (UTC) by thaewrapt)

Failed to build without cpio:

make[1]: *** [kernel/Makefile:145: kernel/kheaders_data.tar.xz] Error 127
make: *** [Makefile:1750: kernel] Error 2
make: *** Waiting for unfinished jobs....

asasione commented on 2020-09-18 11:52 (UTC) (edited on 2020-09-18 11:52 (UTC) by asasione)

I am getting a systemd error on this kernel, I'm using the official binaries: Failed to find module 'autofs4'

damentz commented on 2020-09-13 01:21 (UTC)

The modules required for Anbox are now included, but there's still more instructions. Read the wiki:

enihcam commented on 2020-09-12 17:59 (UTC)

according to changelog, Anbox support is added. How to enable it?

sir_lucjan commented on 2020-09-06 13:25 (UTC)

Paste log with errors.

duff commented on 2020-09-06 13:23 (UTC) (edited on 2020-09-06 13:24 (UTC) by duff)

Latest Version (2020-09-05 14:29) does not build :(

CC [M] fs/xfs/scrub/agheader_repair.o CC [M] fs/xfs/scrub/bitmap.o AR fs/built-in.a CC [M] fs/xfs/scrub/repair.o LD [M] fs/xfs/xfs.o ==> FEHLER: Ein Fehler geschah in build(). Breche ab... error making: linux-lqx (linux-lqx linux-lqx-docs linux-lqx-headers)

enihcam commented on 2020-09-03 23:44 (UTC)

sir, that was a long while ago. i don't see why cpio break the build.

btw, one of the reasons i like this package is that PKGBUILD content is clean and neat. :)

sir_lucjan commented on 2020-09-03 13:59 (UTC)

I remember that cpio was added in response to this request:

damentz commented on 2020-09-03 13:56 (UTC)

You're probably right, I don't see them in the official PKGBUILD for linux. I'll remove them on the next update and specify package(s) that are actually required, liked base-devel.

enihcam commented on 2020-09-03 13:39 (UTC)

Two unnecessary makedepends: 'inetutils' 'cpio' ?

damentz commented on 2020-08-31 15:22 (UTC) (edited on 2021-12-21 18:25 (UTC) by damentz)

Official binaries of linux-lqx, linux-lqx-headers, and linux-lqx-docs are now available:

Signing key import instructions: sudo pacman-key --keyserver hkps:// --recv-keys 9AE4078033F8024D && sudo pacman-key --lsign-key 9AE4078033F8024D

commented on 2020-05-28 06:09 (UTC)

I installed the kernel and the headers from Manjaro Pamac but I don't get an option to choose it when booting

coremanfrank commented on 2020-05-11 03:02 (UTC)

this is worked for me $ gpg --keyserver --recv-key 38DBBDC86092693E

damentz commented on 2020-05-06 20:11 (UTC)

Ok great, glad you got it working.

Fenrihr commented on 2020-05-06 19:48 (UTC)

@damentz the problem was the pacman GUI in Manjaro ... i resolved with in the terminal (but i not remember how at this momment)

damentz commented on 2020-05-06 19:43 (UTC) (edited on 2020-05-06 19:50 (UTC) by damentz)

@Fenrihr, did you do any research on what that output means?

Fenrihr commented on 2020-05-02 11:31 (UTC)

Hi! i'm getting the msg: "the key ABAF11C65A2970B130ABE3C479BE3E4300411886 could not be imported"

henrik_er commented on 2020-03-13 09:21 (UTC)


Wasn't aware that they were interconnected. Thanks!

damentz commented on 2020-03-13 07:23 (UTC)


henrik_er commented on 2020-03-13 06:00 (UTC)

Hi @sir_lucjan and @damentz I am relatively new to kernel compiling, but any reason the "make ..." in the PKGBUILD cannot be changed to "make -j$(nproc) ..." I have modified it manually at each update, and it greatly improves the compilation speed as it utilizes all cores. I have not seen any issues.

sir_lucjan commented on 2020-02-08 11:01 (UTC)


No. Liquorix patch must be applied to the kernel 5.X, not 5.X.Y

j1simon commented on 2020-02-08 09:48 (UTC)

If it's 5.5.2, shouldn't it download that version of the kernel repository and not 5.5 (as it does now)?

==> Making package: linux-lqx 5.5.2_1-1 (Sat Feb  8 10:46:24 2020)
==> Checking runtime dependencies...
==> Checking buildtime dependencies...
==> Retrieving sources...
  -> Downloading linux-5.5.tar.xz...
Initializing download:
File size: 110713660 bytes
Opening output file linux-5.5.tar.xz.part
Starting download
Downloaded 105.585 Megabyte(s) in 1 second(s). (79320.54 KB/s)
  -> Downloading linux-5.5.tar.sign...
Initializing download:
File size: 983 bytes
Opening output file linux-5.5.tar.sign.part
Starting download
Downloaded 983 byte(s) in 0 second(s). (9.58 KB/s)
  -> Downloading 5.5-3.tar.gz...
Initializing download:

frogec commented on 2019-12-10 08:26 (UTC)

I had problems with gpg keys not importing (I stoped the command after 20 minutes). When installing with yay first decline importing of the keys.

==> PGP keys need importing ==> Import? [Y/n] n

Than it will give you the following error:

linux-5.3.tar ... FAILED (unknown public key <KEY>)

==> ERROR: One or more PGP signatures could not be verified! Error downloading sources: linux-lqx

Use command to add key:

$ gpg --recv-key <KEY>


Retry installing. The verification passed for me.

Freso commented on 2019-12-08 01:04 (UTC) (edited on 2019-12-08 01:06 (UTC) by Freso)

@gardotd426 That error is not related to this package, but something on your system.

gardotd426 commented on 2019-12-07 21:59 (UTC)

Can't install this for the life of me:

==> Import? [Y/n] y
:: Importing keys with gpg......
gpg: key 79BE3E4300411886: 150215 signatures not checked due to missing keys
gpg: error writing keyring '/home/matt/.gnupg/pubring.kbx': Provided object is too large
gpg: key 79BE3E4300411886: keyblock too large, retrying with self-sigs-only
gpg: key 79BE3E4300411886: public key "Linus Torvalds <>" imported
gpg: Total number processed: 1
gpg:               imported: 1
==> Error: Problem importing keys

bparsons commented on 2019-12-04 14:40 (UTC)

FWIW, I've worked through the suspend/resume issues with 5.4 on my Thinkpad P50. Let me know if I can help.

Freso commented on 2019-12-02 23:56 (UTC)

@digitalone See the commit message:

digitalone commented on 2019-12-02 23:16 (UTC)

Why you downgraded to 5.3?

gardotd426 commented on 2019-11-22 05:39 (UTC)

This might be a bit nitpicky (but also kind of funny), but the PKGBUILD near the end when it lists the devs, it says "Linux Torvalds"

Agafron commented on 2019-11-05 23:08 (UTC)

@sir_lucjan , thanks! now i must refuse lqx, waitin for sync, on arch I have no cross, because laziness see into, sorry

sir_lucjan commented on 2019-11-05 17:21 (UTC)


I checked Manjaro repositories a moment ago and the required versions of kmod and mkinitcpio are not even in the unstable repository. What's more, Manjaro developers have rebuilt these packages in their own way and at this point most AUR kernels are completely incompatible with Manjaro.

sir_lucjan commented on 2019-10-31 19:16 (UTC)

1) This is the right thing to do. Extramodules have been abandoned in Arch IN KERNEL and transferred to individual drivers and modules. At least it looks like it.

2) No. I used to use Virtualbox once, but I don't need to build any module for a long time now.

Agafron commented on 2019-10-31 19:14 (UTC) (edited on 2019-10-31 19:15 (UTC) by Agafron)

I don't have nvidia so I'm not able to check how and if it will work.
ok, have you r8186 or any driver which need build separately?

Agafron commented on 2019-10-31 19:12 (UTC)

See on PKGBUILD from Arch repos
I do, but _extradir="/usr/lib/modules/$(</usr/src/linux/version)/extramodules" stay

sir_lucjan commented on 2019-10-31 18:12 (UTC)


Yes, extramodules was dropped.

See on PKGBUILD from Arch repos - I don't have nvidia so I'm not able to check how and if it will work.

Agafron commented on 2019-10-31 18:00 (UTC) (edited on 2019-10-31 18:02 (UTC) by Agafron)

@sir_lucjan, if I understand right, i can modify PKGBUILD for nvidia-dkms,
I'll need change path /usr/src/linux on /usr/src/linux-lqx
but I don`t know where i need put build driver modules? Dir "extramodules" is in out?

sir_lucjan commented on 2019-10-30 23:36 (UTC)

The packages are more likely to go to Manjaro but when - I have no idea. Maybe it will take a few days, maybe a few weeks - not more, I think. I mean the new mkinitcpio rather surely and the kmod patched by Archa also but one hundred percent surety can not be. On the example of extramodules you can see that the changes are not always accepted 100%.

Agafron commented on 2019-10-30 23:31 (UTC)


Agafron commented on 2019-10-30 23:29 (UTC)

I think, That philosophy in manjaro, not so far from philosophy in Arch and will be synchronized))

sir_lucjan commented on 2019-10-30 23:26 (UTC) (edited on 2019-10-30 23:26 (UTC) by sir_lucjan)

New kmod and mkinitcpio will introduce new hooks in place of the existing ones. Therefore, the old ones are removed as redundant. As Manjaro will not have this for some time, neither initramfs nor depmods will be built. This will make it impossible to install and run the kernel. So it is better to keep this in mind. I don't do it out of malice - but the priority is to support Arch Linux, which is quite understandable, right?

Agafron commented on 2019-10-30 23:19 (UTC) (edited on 2019-10-30 23:21 (UTC) by Agafron)

ok, thanks, I'll keep it on mind

sir_lucjan commented on 2019-10-30 23:14 (UTC)

By the way, new kmod and mkinitcpio will soon appear in the stable repos. This will force the introduction of changes that will prevent the use of kernels from AUR in Manjaro until the moment when these packages do not appear in the Manjaro repository. So keep this in mind.

Although the kernel should compile without any problems and the problem will occur only during the installation.

sir_lucjan commented on 2019-10-30 23:06 (UTC) (edited on 2019-10-30 23:09 (UTC) by sir_lucjan)

Do you use Manjaro? Well, I'm not going to maintain compatibility with Manjaro just with Arch Linux. Arch abandoned extramodules so linux-lqx also. So if you want to continue using kernel from AUR then you have two choices:

1) You're using Arch and that's it.

2) You're modifying PKGBUILDs from Arch Linux repo and that's it.

Agafron commented on 2019-10-30 23:02 (UTC) (edited on 2019-10-30 23:04 (UTC) by Agafron)

nvidia driver, PKGBUILD
r8168 driver, -
PS OS manjaro

sir_lucjan commented on 2019-10-30 22:49 (UTC)

Arch abandoned the extramodules so all the kernels were synchronized with the stock. I don't know which driver you're installing, but check out the new philosophy in Arch. If you tell me what you want to install, there may be a solution.

sir_lucjan commented on 2019-10-30 22:45 (UTC)

Please look at:

Agafron commented on 2019-10-30 22:43 (UTC) (edited on 2019-10-30 22:43 (UTC) by Agafron)

I'm trying to build a driver, but it gives an error:
cat: /usr/lib/modules/extramodules-lqx/version: No such file or directory

Filip98 commented on 2019-10-23 17:13 (UTC)

@damentz, "I choose a lazy person to do a hard job. Because a lazy person will find an easy way to do it." something like... this?: (forgot the standard platform) It lacks sed commands to modify linux-lqx's PKGBUILD into a new one for the bin version but this beta should build linux-lqx minus not importing the pgp keys atm

damentz commented on 2019-10-22 14:39 (UTC)

@Filip98, purely out of laziness. If you can point me to a guide that utilizes some type of chroot to build AUR packages and how to host them properly, I'll take a look. I'd prefer the binaries to be on for trust reasons.

Filip98 commented on 2019-10-22 08:31 (UTC)

This package has been available for a while but there's no -bin version, is there any specific reason? if not would it be okay for me to set up a linux-lqx-bin package? presumably using the precompiled ones for e.g Ubuntu isn't the solution.

damentz commented on 2019-10-22 01:34 (UTC)

@XopmoH, looks like you ran out of disk space or you got filesystem corruption in the middle of the build. Make sure your build directory has enough space.

@zebulon, Liquorix prefers MuQSS over CFS for process scheduling and has major other configuration differences to the "linux-zen" kernel. You can run a diff on the two configurations to see. Some of the more obvious differences are on

sir_lucjan commented on 2019-10-21 19:48 (UTC) (edited on 2019-10-21 19:49 (UTC) by sir_lucjan)

Did you use some kind of aur-helper? Besides, why do you associate a compilation error with a browser error? It never happened to me - and I've compiled thousands of kernels already - that a compilation error would cause some kind of regression on a running system... I don't think it's an exaggeration, so I ask you honestly - what does it have to do with the compilation error?

XopmoH commented on 2019-10-21 19:36 (UTC) (edited on 2019-10-21 19:40 (UTC) by XopmoH)

It did not work to build this core! An error occurred during installation from AUR:

Assembler message:
Error: couldn't open /tmp/ccWnsYb5.s to read: Not found file or directory
make[2]: *** [scripts/ fs/btrfs/extent-tree.o] Error 1
make[1]: *** [scripts/ fs/btrfs] Error 2
make: *** [Makefile:1081: fs] Error 2

After that, when I opened the browser in full screen, a message popped up: "An error occurred and the system could not recover. You need to end the session and log in again."

Screenshots with a build error and a full-screen error can be viewed here:!4WgCxYpC!zquOhI2U8fqLTp5yv3KpCQ

zebulon commented on 2019-10-21 07:06 (UTC)

Hi, first thank you very much for your excellent contribution. One question though: what is the difference between linux-zen and linux-lqx? Looking at the Github sources, I am struggling understanding which patches or settings differ from linux-zen. Thanks a lot in advance.

toki1990 commented on 2019-09-05 23:44 (UTC)

@damentz "_makegconfig=1" worked. new window(included menu config) opened

damentz commented on 2019-09-05 23:43 (UTC)

@toki1990, this comment section is not AUR helper technical support. Try using the official arch wiki article[1] to install linux-lqx, then you can ask for help in the Arch support forums if you run into trouble.


toki1990 commented on 2019-09-05 23:15 (UTC)

@damentz I'm using pamac-manager for aur installation. i tried "enable _makemenuconfig=" , "_makemenuconfig=enable" enable_makemenuconfig=Y and more in PKGBUILD. I cant see config menu. Can u teach me step by step.

damentz commented on 2019-09-05 20:48 (UTC)

@toki1990, enable _makemenuconfig= in PKGBUILD and set the cpu architecture to your own, or use native. The kernel Make scripts ignore your environment's configuration for good reason.

toki1990 commented on 2019-09-05 18:07 (UTC)

Hi @damentz, i'm edited makepkg.conf. I'm set "-march=x86-64 -mtune=native". Is it enough for build for special architure. I'm reading "-march -mtune" method is cancelled. How i can optimize my build for amd zen (ryzen r7 1700)

Democrab commented on 2019-08-30 16:09 (UTC)

@damentz, Thank you very much for the updates and extra information, I really appreciate it.

damentz commented on 2019-08-30 16:03 (UTC)

@Democrab, Valve's fsync patch set was merged on v5.2.10-lqx2. Seems like the discussion died down so I went ahead and merged it instead of waiting for the next version. You can see more details in the zen-kernel github issue:

damentz commented on 2019-08-16 22:33 (UTC)


I'm waiting for a second version of the patches in this issue:

There's a lot of discussion about them so I'd rather merge the v2 rather than v1.

Democrab commented on 2019-08-15 20:43 (UTC)

Given the amount of desktop users who game and the fact that this is orientated for desktop users, is there any chance of incorporating Valve's fsync kernel patches into this? I've been adding them in myself with great success in the games I've been playing via Proton.

damentz commented on 2019-07-27 19:28 (UTC)

@wailord284, just added the cpio make dependency and bumped the version.

wailord284 commented on 2019-07-27 18:39 (UTC)

Hi, Can you please add cpio as a dependency? I get this error without it:

./kernel/ line 77: cpio: command not found make[1]: [kernel/Makefile:132: kernel/kheaders_data.tar.xz] Error 127 make[1]: Waiting for unfinished jobs....

Thank you

Arzte commented on 2019-06-23 23:11 (UTC)

Thanks for looking into it! I appropriate it greatly!!

damentz commented on 2019-06-23 23:06 (UTC)

Done, just pushed out a new commit. No need to bump the pkgrel since the change is technically a no-op if you use makepkg. Any new build of this package will include the change, so yay users should get the benefits immediately.

sir_lucjan commented on 2019-06-23 22:46 (UTC)


Sure. We could drop pkgver() if you want.

damentz commented on 2019-06-23 22:44 (UTC)

@sir_lucjan, I separated the commit to change pkgver() from the update to the version updates, so we're still tracking the latest version of Liquorix despite your revert.

According to the wiki, if you define a pkgver() function, it'll be used to automatically update pkgver on execution of makepkg with the code you define. How about we just delete the function entirely then? I don't use that code myself, so all it does is interfere with the value I put into pkgver.

sir_lucjan commented on 2019-06-23 22:25 (UTC)


Currently pkgver() generates the correct version number because regardless of the position in PKGBUILD it starts anyway after the prepare() function where these values are declared. If prepare() was following pkgver() then you would be absolutely right. pkgver() is before prepare() in PKGBUILD only for cosmetic reasons and it doesn't matter for the operation; anyway it is after prepare() and before build().

Freso commented on 2019-06-23 21:57 (UTC)

I don’t think $_minor and $_patchrel get carried over from prepare() to other functions (such as pkgver()). You may need to define them again inside pkgver().

sir_lucjan commented on 2019-06-23 20:46 (UTC)


I've reverted

pkgver() dosen't work properly; I've downgrade lqxpatchset (-13 --> -11) and pkgver() doesn't set a proper pkgver.


Please check your mailbox.

Freso commented on 2019-06-23 20:32 (UTC)

If all pkgver() does is return $pkgver, why not just remove the pkgver() function entirely? The pkgver() is a helper function to set $pkgver, so it makes no sense to have it set $pkgver to $pkgver

Arzte commented on 2019-06-13 21:41 (UTC)

Sure! Using makepkg, if you do your build in a separate command, it'll build using the wrong version. As an example, makepkg --cleanbuild --nobuild followed by makepkg --clean --force --noextract --noprepare will cause the second command to use the wrong version (because it "updates" it to 5.1._-1 from 5.1.9_4-1)

sir_lucjan commented on 2019-06-13 21:04 (UTC)


We don't support yay and aur-helpers.

Please use a makepkg.

Arzte commented on 2019-06-13 21:01 (UTC) (edited on 2019-06-13 21:05 (UTC) by Arzte)

Heya this package is currently broken when using the AUR helper yay. Yay runs functions separately (Eg prepare separate from build) and so any variable set in say prepare aren't kept for later on. This means that when this is building because _minor and _patchrel are set in prepare, this package can't currently be installed using yay, as it builds under a weird version.

An example of this would be with version 5.1.9_4-1 of this package, yay will error out when trying to install it, because makepkg would've built it under 5.1._-1, instead of 5.1.9_4-1.

(The yay issue discussing this is here)

To clarify this is because the variables are in separate functions, so they don't carry on when the PKGBUILD functions are executed separately with makepkg.

Democrab commented on 2019-06-04 13:24 (UTC)

I've been building/using this kernel for a few months on Manjaro, using an AUR Helper with zero problems in that entire timeframe. Any problems that @megapro17 is having are related to his specific system.

Thank you for maintaining this.

damentz commented on 2019-05-14 02:01 (UTC)

@Agafron, thanks for the feedback.

Agafron commented on 2019-05-07 17:38 (UTC)

@megapro17, @damentz, I build this kernel on manjaro, distributive haven`t problems with compiling

Terence commented on 2019-03-08 16:07 (UTC)

@damentz ok great thanks. BTW don't forget to check your matrix client from time to time ;)

damentz commented on 2019-03-08 15:14 (UTC)

@Terence, I merged the patches into 4.20/master on Zen Kernel yesterday. Next release of Liquorix will include them.

Terence commented on 2019-03-07 15:04 (UTC) (edited on 2019-03-07 19:01 (UTC) by Terence)

@sir_lucjan @damentz would it be possible to merge bfq-mq v10 into the kernel? The new version seems to greatly improve throughput performances but only made it for inclusion into 5.2

sir_lucjan commented on 2019-02-06 10:29 (UTC)


I've compiled with makepkg.

Please fix your hands. And if it possible, your system too.

sir_lucjan commented on 2019-02-06 10:21 (UTC)


We don't support Manjaro. That PKGBUILD doesn't use wget so..... problem exists between keyboard and chair.

damentz commented on 2019-02-05 21:45 (UTC)

@simona, Have you tried submitting your build errors to the yay maintainers? Makepkg is the core Arch utility for building AUR packages. If your helper is creating new problems that makepkg does not cause, then the AUR helper has a bug.

Also, looking at yay's github issue page, I'm seeing open issues about yay trying to install packages for the wrong architecture after building:

When we say we don't support AUR helpers, it's because our AUR package comment section is not for AUR helper tech support. That belongs in forums that the developers of the AUR helpers provide support, like their github or AUR package page.

@megapro17, Have you opened up a thread on the Manjaro forum? If only Manjaro users are unable to build a package in the AUR, then there's something specific to Manjaro that's causing it. You should provide them feedback to they can look into it. I'm sure other Manjaro users would benefit too.

megapro17 commented on 2019-02-05 17:26 (UTC)

I tried to install this kernel on fresh manjaro, and it just frozen after wget phase. Maintainer please fix your hands. And if it possible, package too.

sir_lucjan commented on 2019-01-16 11:47 (UTC) (edited on 2019-01-16 11:49 (UTC) by sir_lucjan)

The problem IS your aur helper. I've compiled with makepkg:



$ ls   
4.19-10.tar.gz  PKGBUILD             linux-lqx-4.19.15_1-2-x86_64.pkg.tar.xz           linux-lqx-headers-4.19.15_1-2-x86_64.pkg.tar.xz      linux.preset
60-linux.hook   config.last          linux-lqx-4.19.15_1-2-x86_64.pkg.tar.xz.sig       linux-lqx-headers-4.19.15_1-2-x86_64.pkg.tar.xz.sig  pkg
90-linux.hook   linux-4.19.tar.sign  linux-lqx-docs-4.19.15_1-2-x86_64.pkg.tar.xz      linux.install                                        src
99-linux.hook   linux-4.19.tar.xz    linux-lqx-docs-4.19.15_1-2-x86_64.pkg.tar.xz.sig  linux.install.pkg

So.... please try with makepkg or "will simply stop using this kernel"


simona commented on 2019-01-16 11:11 (UTC)

the problem is any aur helper I use and this is probably not a bug in the helper. all the other packages (and I use a lot of them) work perfectly with all my aur helper (sporadically someone does not work with some specific aur helper). although it is legitimate to ignore this specific problem, more generally I do not think it is right to declare war on aur helpers as if they were absolute evil. therefore I will simply stop using this kernel or wait for the problem to be "alone" over time. good day.

sir_lucjan commented on 2019-01-14 15:25 (UTC)

I don't support aur-helpers.

Please use makepkg instead yay.

simona commented on 2019-01-14 15:20 (UTC)

==> Uscita dall'ambiente di fakeroot. ==> Compilazione terminata: linux-lqx 4.19._-1 (lun 14 gen 2019 14:12:33 CET) ==> Pulizia in corso... caricamento dei pacchetti in corso... errore: '/home/simona/.cache/yay/linux-lqx/linux-lqx-4.19.15_1-1-x86_64.pkg.tar.xz': impossibile trovare o leggere il pacchetto errore: '/home/simona/.cache/yay/linux-lqx/linux-lqx-docs-4.19.15_1-1-x86_64.pkg.tar.xz': impossibile trovare o leggere il pacchetto errore: '/home/simona/.cache/yay/linux-lqx/linux-lqx-headers-4.19.15_1-1-x86_64.pkg.tar.xz': impossibile trovare o leggere il pacchetto

sir_lucjan commented on 2019-01-13 17:11 (UTC)

Please paste a log with error.

simona commented on 2019-01-12 10:54 (UTC)

compile but at the end do not create package...

Agafron commented on 2018-12-21 20:37 (UTC)

@damentz, I can`t initialize rules on this kernel, any comand had answer 'No chain/target/match by that name', Its after update on 4.19-lqx? on the custom kernel iptables works ok. OS manjaro. which info can I give you to fix this problem?

damentz commented on 2018-12-21 18:10 (UTC)

@Agafron, there shouldn't be anything different with iptables on liquorix. On my work laptop, I'm able to use UFW to block incoming connections besides those to port 22, and sshuttle to create poor mans VPNs utilizing iptables to redirect traffic.

Maybe you forgot to initialize your rules?

Agafron commented on 2018-12-19 20:25 (UTC)

@damentz, hi! I have some problems with iptables on this version

sudo iptables -L

iptables: No chain/target/match by that name.

what parameters affect it?

damentz commented on 2018-12-01 20:42 (UTC)

@Filip98, I'll re-enable hierarchical scheduling once Paolo corrects an oops when it's enabled. You can see the progress on the bfq-iosched google group here:!topic/bfq-iosched/5wkrbgXf4VM

Filip98 commented on 2018-11-27 23:00 (UTC)

This kernel seems to have an issue with libvirt(Qemu/KVM) I've digged a little bit around the net: and looks like it requires the following to be enabled in the kernel: CONFIG_BLK_CGROUP CONFIG_CFQ_GROUP_IOSCHED CONFIG_BLK_DEV_THROTTLING I switched to my alternative kernel and it runs fine, I didn't recompile to confirm that this fixes the issue tho

damentz commented on 2018-11-20 14:43 (UTC) (edited on 2018-11-20 15:21 (UTC) by damentz)

@Terence, I briefly experimented again with 1000hz but it was a huge drain on battery life. MuQSS 0.173 (and possibly 0.180 with 4.19), doesn't support tickless idle correctly, and ends up feeding cpufreq with bad idle data. The periodic forces cpufreq to ingest idle time accounting data, letting the cpu properly rest while idle, but the periodic tick prevents the CPU from staying in the CPU's lowest state (C6 / C7). It's measurable, and I mentioned some data points in my commit reverting back to 250hz here:

Until I figure out a better compromise with 1000hz, you'll need to manually go in on every build and raise the frequency to 1000hz for your system. It's perfectly fine for desktops but a huge problem on laptops. Both my personal and work laptops lose about 20%-30% battery life for idle / casual loads. In my testing, 250hz preserves cpu temperatures as well as tickless idle, so that'll work for now.

Also, I re-enabled IRQ time accounting as it seems that it was originally added so that the process scheduler (CFS in this case), had more accurate data about how much cpu time each task consumed. It should have some benefits for MuQSS as well.

Agafron commented on 2018-11-19 19:30 (UTC)

@sir_lucjan, ok, thanks!

sir_lucjan commented on 2018-11-19 19:26 (UTC)


I don't think so.

Agafron commented on 2018-11-19 19:24 (UTC)

@sir_lucjan, ok, need I reinstall package?

$ uname -r


sir_lucjan commented on 2018-11-19 19:12 (UTC)



Agafron commented on 2018-11-19 19:06 (UTC)

@damentz, I have some diff of versions:

1 aur/linux-lqx 4.18.19_2-2 [installed: 4.18.20_1-1] (114) (1,86)

A desktop oriented kernel and modules with Liquorix patches

2 aur/linux-lqx-docs 4.18.19_2-2 [installed: 4.18.20_1-1] (114) (1,86)

Kernel hackers manual - HTML documentation that comes with the linux-lqx.

3 aur/linux-lqx-headers 4.18.19_2-2 [installed: 4.18.20_1-1] (114) (1,86)

Its PKGBUILD mistake?

damentz commented on 2018-11-18 18:39 (UTC) (edited on 2018-11-18 18:40 (UTC) by damentz)

@Terence, that's good information. While you were comparing results, I was measuring the overhead of 1000hz vs 250hz.

With 250hz, the timer consumes about 2-3ms, every second. Increasing this to 1000hz almost quadruples it. At idle, the timer tick consumes 8ms/s, but 5ms/s when the system is under full load (highest frequency). 8ms is less than 1% of the cpu usage of one core on a processor.

On the flip-side, one change I made with switching to 250hz is to re-instate a sampling down factor on ondemand. While the system is idle, it consumes about 3-5ms/s worth of processing to determine the next frequency. While the system is under full load, or near full load, ondemand consumes less than 1ms/s.

One thing I was wondering if you could try is adding rqshare=mc to your kernel parameters. Liquorix is currently configured with smt runqueue sharing, which improves throughput by looking for tasks in the order of cache locality. This might be influencing your underruns in a bad way.

Terence commented on 2018-11-18 15:22 (UTC)

@damentz I tried your suggestion but it didn't seem to improve things. I have between 3 and 4 times less xruns when using 1000Hz tick rate compared to the now default 250Hz. My test settings were a sample rate of 48000 Hz with 64 as the buffer size giving a latency of 1.3ms. I used the performance governor. I normally use 128 as the buffer size but I increased it in order to get more frequent xruns. An other observation is the schedutil is closer to the performance governor in terms of xruns at 1000Hz but there is a bigger gap between both at 250Hz.

I'm still wondering if it wouldn't make sense to keep 250hz as the default as my usage is quite an edge case I would say.

damentz commented on 2018-11-17 19:15 (UTC)

@Terence, try setting rr_interval back to 3 in /proc/sys/kernel. If that doesn't help, then it's the kernel tick frequency.

And thanks for reporting. Con has mentioned that the kernel tick frequency shouldn't matter and that MuQSS is "tickless". This may prove the opposite and that you still need a high frequency tick rate to support soft-realtime applications properly.

Let me know which of the two settings helps you the most - I'll update the configuration accordingly.

Terence commented on 2018-11-16 12:27 (UTC)

@damentz your changes introduced by while being saner defaults degrades realtime audio/midi performances (more frequent xruns). Would only changing back the tick to 1000hz locally enough?

sir_lucjan commented on 2018-11-13 19:47 (UTC)


Because 4.18.18-lqx patchset includes 4.18.18 from upstream.

Agafron commented on 2018-11-13 19:43 (UTC) (edited on 2018-11-13 19:49 (UTC) by Agafron)

Downloading linux-4.18.tar.xz





Why is downloading linux-4.18.tar.xz instead of linux-4.18.18.tar.xz?

CybeX commented on 2018-11-13 10:00 (UTC) (edited on 2018-11-13 10:00 (UTC) by CybeX)

If anyone has an issue in building the kernel relating to:

 Verifying source file signatures with gpg...
    linux-4.18.tar ... FAILED (unknown public key 79BE3E4300411886)
==> ERROR:</font> One or more PGP signatures could not be verified!
Use the following:
gpg --keyserver hkps:// --recv-keys 79BE3E4300411886

gaougalos commented on 2018-11-10 17:16 (UTC)

When installing this kernel i get ERROR: module not found: `usbhid'

mapcode.mind commented on 2018-10-11 20:40 (UTC)

i'm happy to report that the package compiled and installed successfully. @damentz : your comment is informative; I didn't know that having not used makepkg before. I was unsuccessfully trying to edit the PKGBUILD variables to make it create the correct file. Thanks a lot for your input.

@sir_lucjan : about manjaro not being arch, i know that. I was thinking of using manjaro as a stepping stone towards arch, coming from a debian system. But, i have no complaints with the system i have now. Nor can i figure out a compelling enough reason to switch. In fact, this is the first issue i have had with aur since i installed manjaro, and i am pretty sure that this was an issue with aur helper yay and not manjaro itself. Thanks a lot for your comments and suggestions; they were really helpful.

damentz commented on 2018-10-11 20:29 (UTC) (edited on 2018-10-11 20:29 (UTC) by damentz)

There's probably multiple reasons why makepkg -sric solved your problem:

-s = --syncdeps - installs missing dependencies

-c = --clean - removes artifacts and work directories before building

-r = --rmdeps - removes installed dependecies required for build

It could just be you were missing dependencies and your work directory was dirty, had previous folders and files that were interfering with your build.

sir_lucjan commented on 2018-10-11 20:04 (UTC)

Manjaro != Arch

AUR is designed for Arch, not Manjaro

mapcode.mind commented on 2018-10-11 20:00 (UTC)

i am running manjaro and have base-devel already installed. You know it's really surprising, but this time i tried makepkg -sric, the first line was what it is supposed to be :

==> Making package: linux-lqx 4.18.13_1-1 (Fri 12 Oct 2018 01:21:27 IST)
==> Checking runtime dependencies...
==> Checking buildtime dependencies...
==> Retrieving sources...
  -> Downloading linux-4.18.tar.xz...

I guess it would work this time. I'll let you know after it finishes compiling. Thank you for the makepkg suggestion.

sir_lucjan commented on 2018-10-11 19:50 (UTC) (edited on 2018-10-11 19:51 (UTC) by sir_lucjan)

Do you use Arch Linux or fork? Do you install base-devel? I've installed linux-lqx without problem:

mapcode.mind commented on 2018-10-11 19:49 (UTC)

i just tried makepkg and the first line itself is instructive :

➜  linux-lqx git:(master) ✗ makepkg
==> Making package: linux-lqx 4.18._-1 (Fri 12 Oct 2018 01:17:44 IST)
==> Checking runtime dependencies...
==> Checking buildtime dependencies...
==> Retrieving sources...
  -> Found linux-4.18.tar.xz
  -> Found linux-4.18.tar.sign
  -> Found 4.18-16.tar.gz
  -> Found 60-linux.hook
  -> Found 90-linux.hook
  -> Found 99-linux.hook
  -> Found linux.preset
==> Validating source files with sha512sums...
    linux-4.18.tar.xz ... Passed
    linux-4.18.tar.sign ... Skipped
    4.18-16.tar.gz ... Passed
    60-linux.hook ... Passed
    90-linux.hook ... Passed
    99-linux.hook ... Passed
    linux.preset ... Passed
==> Verifying source file signatures with gpg...
    linux-4.18.tar ... Passed
==> Extracting sources...
  -> Extracting linux-4.18.tar.xz with bsdtar
==> ERROR: Aborted by user! Exiting...

sir_lucjan commented on 2018-10-11 19:46 (UTC) (edited on 2018-10-11 19:48 (UTC) by sir_lucjan)

Try with makepkg.

git clone
cd linux-lqx.git
makepkg -sric

mapcode.mind commented on 2018-10-11 19:44 (UTC)

i am using aur helper yay. I edit the PKGBUILD to use nconfig at times and to set use_Current=y.

sir_lucjan commented on 2018-10-11 19:43 (UTC) (edited on 2018-10-11 19:45 (UTC) by sir_lucjan)

Do you use aur helpers or makepkg?

I don't support aur-helpers. Try with makepkg.

mapcode.mind commented on 2018-10-11 19:41 (UTC) (edited on 2018-10-11 19:43 (UTC) by mapcode.mind)

@sir_lucjan : i was able to install linux-lqx with yay until now. I'd try using makepkg if the issue persists.

here are the logs for the latest build i tried:

-> Adding symlink...
  -> Fixing permissions...
==> Tidying install...
  -> Removing libtool files...
  -> Purging unwanted files...
  -> Removing static library files...
  -> Compressing man and info pages...
==> Checking for packaging issues...
==> Creating package "linux-lqx-headers"...
  -> Generating .PKGINFO file...
  -> Generating .BUILDINFO file...
  -> Generating .MTREE file...
  -> Compressing package...
==> Starting package_linux-lqx-docs()...
  -> Installing documentation...
  -> Adding symlink...
  -> Fixing permissions...
==> Tidying install...
  -> Removing libtool files...
  -> Purging unwanted files...
  -> Removing static library files...
  -> Compressing man and info pages...
==> Checking for packaging issues...
==> Creating package "linux-lqx-docs"...
  -> Generating .PKGINFO file...
  -> Generating .BUILDINFO file...
  -> Generating .MTREE file...
  -> Compressing package...
==> Leaving fakeroot environment.
==> Finished making: linux-lqx 4.18._-1 (Fri 12 Oct 2018 01:06:46 IST)
==> Cleaning up...
loading packages...
error: '/home/mapcode/.cache/yay/linux-lqx/linux-lqx-4.18.13_1-1-x86_64.pkg.tar.xz': could not find or read package
error: '/home/mapcode/.cache/yay/linux-lqx/linux-lqx-headers-4.18.13_1-1-x86_64.pkg.tar.xz': could not find or read package

sir_lucjan commented on 2018-10-11 15:07 (UTC)

@ mapcode.mind

You should use makepkg.

damentz commented on 2018-10-11 14:41 (UTC)

@mapcode.mind, try the latest version.

mapcode.mind commented on 2018-10-11 14:17 (UTC) (edited on 2018-10-11 14:18 (UTC) by mapcode.mind)

➜  ~ yay -Syua
:: Searching AUR for updates...

:: 2 Packages to upgrade.
2  aur/linux-lqx          4.18._-1 -> 4.18.12_2-2
1  aur/linux-lqx-headers  4.18._-1 -> 4.18.12_2-2
==> Packages to not upgrade: (eg: 1 2 3, 1-3, ^4 or repo name)
==> ^C⏎   

This is the output after i manually installed the .pkg.tar.xz. I'll recompile the kernel and paste logs.

sir_lucjan commented on 2018-10-11 10:35 (UTC)

I can not confirm:

sir_lucjan commented on 2018-10-11 10:35 (UTC)

I can not confirm:

mapcode.mind commented on 2018-10-11 08:19 (UTC)

i wish i could, but then i'll have to recompile it. But, i remember that the build process finished successfully, and a linux-lqx-4.18.-1.pkg.tar.xz file was created. But, at the final step, when this .pkg.tar.xz file was to be installed, the script looked for linux-lqx-4.18.12_2-2.pkg.tar.xz, couldn't find it, and exited with error. I manually was able to install the linux-lqx-4.18.-1.pkg.tar.xz file then with pacman - U.

sir_lucjan commented on 2018-10-10 21:51 (UTC)

Could you paste logs?

mapcode.mind commented on 2018-10-10 21:47 (UTC) (edited on 2018-10-10 21:48 (UTC) by mapcode.mind)

i am facing some issues when installing it. Somehow the versioning is not being assigned properly in the pkgbuild. the pkgbuild builds it as 4.18._-1 instead of 4.18.12_2-2, and then exits complaining at the last installation stage that it couldn't find 4.18.12_2-2.

I was able to manually install the .pkg.tar.xz files

Terence commented on 2018-10-10 16:33 (UTC)


sir_lucjan commented on 2018-10-10 15:22 (UTC)


I'll test

damentz commented on 2018-10-10 15:11 (UTC)


I don't see a problem keeping linux-lqx-docs, it's just htmldocs that's the problem.

Check your email, I sent you a patch to disable htmldocs by default, and allow one to enable it as they need it.

sir_lucjan commented on 2018-10-10 14:23 (UTC) (edited on 2018-10-10 14:29 (UTC) by sir_lucjan)


Probably I'll drop docs package, so tell me what you think.

damentz commented on 2018-10-10 14:17 (UTC)

@sir_lucjan, @Terence's comment reminded me - why don't we just turn off htmldocs? They were added to the primary Arch linux PKGBUILD, but I genuinely don't think enough people use it to make it a build requirement. At least, they only existed in the docs package just a month or two ago, I don't think anyone will care if we turn them back off.

sir_lucjan commented on 2018-10-10 12:15 (UTC)

@ Terence

Please look at:

makedepends=(xmlto kmod inetutils bc libelf git python-sphinx graphviz)

Could you paste logs?

Terence commented on 2018-10-10 12:03 (UTC)

@sir_lucjan Please add texlive-latexextra to the makedepends otherwise generation of documentation fails.

sir_lucjan commented on 2018-09-21 10:11 (UTC)

@ Dill

OK. Damentz has updated sources. Please read pinned comment.

DIlI commented on 2018-09-21 05:48 (UTC) (edited on 2018-09-21 05:49 (UTC) by DIlI)


Verified config.amd64 NOT SUCCESSFUL Could you please fix it? Thank you!

ronoverdrive commented on 2018-09-20 23:02 (UTC)

Have the CPUFreq drivers been removed? I've been unable to change governors or adjust the ondemand ramp up settings and the drivers don't seem to be present even though I have them all check marked in XConfig.

Terence commented on 2018-09-08 18:54 (UTC)

@sir_lucjan: simona is probably using an aur helper and has also modified the config through one of the config tools available in the PKGBUILD, he just needs to delete the file (and make a backup if necessary).

sir_lucjan commented on 2018-09-08 13:15 (UTC)


I can not confirm:

[lucjan@archlinux ~]$ export LANG=C
[lucjan@archlinux ~]$ git cl aur:linux-lqx
Cloning into 'linux-lqx'...
remote: Enumerating objects: 1097, done.
remote: Counting objects: 100% (1097/1097), done.
remote: Compressing objects: 100% (527/527), done.
remote: Total 1097 (delta 639), reused 1017 (delta 570)
Receiving objects: 100% (1097/1097), 351.51 KiB | 3.38 MiB/s, done.
Resolving deltas: 100% (639/639), done.
[lucjan@archlinux ~]$ cd linux-lqx
[lucjan@archlinux linux-lqx]$ makepkg -sr --sign
==> Making package: linux-lqx 4.18.6-1 (Sat Sep  8 15:14:30 2018)
==> Checking runtime dependencies...
==> Checking buildtime dependencies...
==> Retrieving sources...
  -> Downloading linux-4.18.tar.xz...
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
100   178  100   178    0     0    917      0 --:--:-- --:--:-- --:--:--   917
100 97.0M  100 97.0M    0     0  3810k      0  0:00:26  0:00:26 --:--:-- 3771k
  -> Downloading linux-4.18.tar.sign...
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
100   178  100   178    0     0   1022      0 --:--:-- --:--:-- --:--:--  1022
100   455  100   455    0     0   1350      0 --:--:-- --:--:-- --:--:--  1350
  -> Downloading 4.18.6-1.patch.gz...
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
100  625k  100  625k    0     0   598k      0  0:00:01  0:00:01 --:--:--  598k
  -> Downloading config.amd64...
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
100  201k  100  201k    0     0   246k      0 --:--:-- --:--:-- --:--:--  246k
  -> Found 60-linux.hook
  -> Found 90-linux.hook
  -> Found 99-linux.hook
  -> Found linux.preset
==> Validating source files with sha512sums...
    linux-4.18.tar.xz ... Passed
    linux-4.18.tar.sign ... Skipped
    4.18.6-1.patch.gz ... Passed
    config.amd64 ... Passed
    60-linux.hook ... Passed
    90-linux.hook ... Passed
    99-linux.hook ... Passed
    linux.preset ... Passed
==> Verifying source file signatures with gpg...
    linux-4.18.tar ... Passed
==> Extracting sources...
  -> Extracting linux-4.18.tar.xz with bsdtar

sir_lucjan commented on 2018-09-08 13:12 (UTC)

  1. Please read first comment
  2. Only english!!!

simona commented on 2018-09-08 12:28 (UTC)

==> Validazione di source file con sha512sums... linux-4.18.tar.xz ... Verificato linux-4.18.tar.sign ... Ignorato 4.18.6-1.patch.gz ... Verificato config.amd64 ... NON RIUSCITO

sir_lucjan commented on 2018-09-08 11:17 (UTC)

Bump to 4.18.6-1.

PS. Latest commit got its name wrong, my bad.

Terence commented on 2018-09-03 15:34 (UTC) (edited on 2018-09-05 12:13 (UTC) by Terence)

Would you mind updating it based on ?

Edit: nevermind:

sir_lucjan commented on 2018-08-24 18:09 (UTC)


damentz commented on 2018-08-24 17:56 (UTC)


@Terence is correct here, if you update the package, you need to bump pkgrel. It may install the same kernel, but the package is now distinctly different and does at least one thing different, and thus is a different version.

sir_lucjan commented on 2018-08-24 17:42 (UTC)


Please look at:

Did you see?

sir_lucjan commented on 2018-08-24 17:38 (UTC)


I'll bump pkgrel later.

Terence commented on 2018-08-24 17:36 (UTC)

@damentz @sir_lucjan I'm not saying pkgrel needs to be 3 to match the lqxrel but because the PKGBUILD has been modified by 5c3a7fec1607 and thus needs to have its pkgrel updated.

damentz commented on 2018-08-24 17:13 (UTC)


The pkgrel variable determines the version of the package, which gets incremented on every update. The lqxrel variable is used to find the liquorix release / patch to pull everything the package needs. It's only coincidental that they're identical at times.

sir_lucjan commented on 2018-08-24 16:36 (UTC)

@Terence Why? pkgrel != lqxrel

Terence commented on 2018-08-24 16:35 (UTC)

@sir_lucjan doesn't 5c3a7fec1607 needs pkgrel=3 ?

sir_lucjan commented on 2018-08-22 12:47 (UTC)



# pkgname=('linux-lqx' 'linux-lqx-headers' 'linux-lqx-docs')

Terence commented on 2018-08-22 12:42 (UTC) (edited on 2018-08-22 12:42 (UTC) by Terence)

@sir_lucjan you forgot to bump to 4.17.17-3 :)

Filip98 commented on 2018-08-08 11:04 (UTC)

Thank you for the responses :).

damentz commented on 2018-07-20 17:00 (UTC) (edited on 2018-07-20 17:03 (UTC) by damentz)

@Filip98, you're probably compiling the kernel with one thread, check the recommendations here :

markusk commented on 2018-07-20 16:57 (UTC)

@Filip98: The compile time can be quite a while on slow/medium hardware and therefore it "is normal for that kind of hardware". You can speed it up a bit by using ccache (

Lubuntu is faster as your were most likely where using a precompiled package from some kernel-repo (launchpad?). The AUR system is source based in most of the cases.

Filip98 commented on 2018-07-20 13:22 (UTC)

It seems to be compiling itself for hours with a list of the kernel files e.g "CC [M] sound/.../*.o" going at it one by one, Is this normal behavior? I'm reluctant to think about repeating this for each update. This wasn't the case while I was using Lubuntu, I'm using aurman if it matters.

damentz commented on 2018-06-27 15:26 (UTC)

@sir_lucjan, thanks for the heads up

@DungSherlock, 'pkg' will use the Liquorix configuration used in the binary kernels from, while old will take your running configuration and execute 'make oldconfig', asking you to select options for new or updated menu items. If you're comfortable with the latter and are only interested in the kernel sources, then 'old' is the way to go. If you want the behavior you get from Liquorix binary kernels (Debian Sid, the major Ubuntu's), then choose 'pkg.

A commonly used option, since you're building from source, is to actually choose 'm'. This will let you adjust the 'pkg' configuration before the build starts. One advantage of this is you can optimize the kernel for your processor, giving you some marginal gains depending on what you're using the kernel for.

DungSherlock commented on 2018-06-27 15:13 (UTC)

Different build option _config="pkg" and _config="old"? What is advice?

sir_lucjan commented on 2018-06-27 10:08 (UTC) (edited on 2018-06-27 10:11 (UTC) by sir_lucjan)


Probably MuQSS doesn't work properly with 4.16.17.

damentz commented on 2018-06-27 03:47 (UTC)

@sir_lucjan, please keep the patch at 4.16.16-2 for now, there's a stable patch in 4.16.17+ that's preventing some of my own systems from booting, but I currently don't have the time to figure out why only Arch is affected.

Seeing as how Arch itself has already moved to 4.17, it might be best to update linux-lqx once there's a 4.17 release - that'll probably include the fix.

markusk commented on 2018-05-07 13:51 (UTC)

@sir_lucjan : thx for looking into it. i cleaned out all my download caches and build locations and now it works ... odd ...

sir_lucjan commented on 2018-05-07 12:41 (UTC)


I can't confirm:

[lucjan@archlinux roboczy-3]$ export LANG=C
[lucjan@archlinux roboczy-3]$ git clone
Cloning into 'linux-lqx'...
remote: Counting objects: 985, done.
remote: Compressing objects: 100% (491/491), done.
remote: Total 985 (delta 561), reused 909 (delta 494)
Receiving objects: 100% (985/985), 300.70 KiB | 1.65 MiB/s, done.
Resolving deltas: 100% (561/561), done.
[lucjan@archlinux roboczy-3]$ cd linux-lqx    
[lucjan@archlinux linux-lqx]$ mkaurball
==> Making package: linux-lqx 4.16.7-1 (Mon May  7 14:35:37 CEST 2018)
==> Retrieving sources...
  -> Downloading linux-4.16.tar.xz...
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
100   178  100   178    0     0   1379      0 --:--:-- --:--:-- --:--:--  1379
100 98.2M  100 98.2M    0     0  2174k      0  0:00:46  0:00:46 --:--:-- 3411k
  -> Downloading linux-4.16.tar.sign...
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
100   178  100   178    0     0    912      0 --:--:-- --:--:-- --:--:--   912
100   455  100   455    0     0   1146      0 --:--:-- --:--:-- --:--:--  1146
  -> Downloading 4.16.7-1.patch.gz...
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
100  644k  100  644k    0     0   624k      0  0:00:01  0:00:01 --:--:--  625k
  -> Downloading config.amd64...
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
100  202k  100  202k    0     0   227k      0 --:--:-- --:--:-- --:--:--  227k
  -> Found 60-linux.hook
  -> Found 90-linux.hook
  -> Found 99-linux.hook
  -> Found linux.preset
==> Validating source files with sha512sums...
    linux-4.16.tar.xz ... Passed
    linux-4.16.tar.sign ... Skipped
    4.16.7-1.patch.gz ... Passed
    config.amd64 ... Passed
    60-linux.hook ... Passed
    90-linux.hook ... Passed
    99-linux.hook ... Passed
    linux.preset ... Passed
==> Verifying source file signatures with gpg...
    linux-4.16.tar ... Passed
==> Entering fakeroot environment...
==> Creating source package...
  -> Adding PKGBUILD...
  -> Generating .SRCINFO file...
  -> Adding 60-linux.hook...
  -> Adding 90-linux.hook...
  -> Adding 99-linux.hook...
  -> Adding linux.preset...
  -> Adding install file (linux.install)...
  -> Compressing source package...
==> Leaving fakeroot environment.
==> Source package created: linux-lqx (Mon May  7 14:36:39 CEST 2018)

sir_lucjan commented on 2018-05-06 16:52 (UTC)

linux-lqx from github:

damentz commented on 2018-05-06 16:15 (UTC)

Hi navjack27,

Intel p-states are disabled deliberately because they return lower frequencies than ondemand during multicore loads, and keep the processor frequency / voltage ramped up too high during idle loads with MuQSS. Due to the nature of how MuQSS works, you'll almost never benefit from TurboBoost 3.0's single core boost speeds, since MuQSS is more likely to reschedule task on a neighboring logical or physical core than CFS is.

You have the option to change the configuration of the AUR package before building using menuconfig - you can always force enable it yourself if your particular system.

sir_lucjan commented on 2018-05-06 16:01 (UTC)


Please report @damentz.

navjack27 commented on 2018-05-06 15:59 (UTC)

PLEASE enable intel p-states in the .config. There shouldn't be any reason why my laptop is limited to 1800MHz. This is a serious oversight. I'm going through the rest of the .config and will point out any other inconsistencies with modern computing.

sir_lucjan commented on 2018-04-27 19:42 (UTC) (edited on 2018-04-27 19:43 (UTC) by sir_lucjan)

@ Plexcon

  1. In English, mate, in English
  2. Please read previous comments.

Plexcon commented on 2018-04-27 19:39 (UTC) (edited on 2018-04-27 19:43 (UTC) by Plexcon)

==> Verificando las firmas de las fuentes con gpg... linux-4.15.tar ... HA FALLADO (clave pública desconocida 79BE3E4300411886) ==> ERROR: ¡No se ha podido verificar alguna de las firmas PGP!


gpg --recv-keys 79BE3E4300411886

damentz commented on 2018-04-18 17:36 (UTC) (edited on 2018-04-18 17:37 (UTC) by damentz)


I updated the configuration last night with the recommendations in the wiki. The latest update @sir_lucjan pushed includes the change.


sir_lucjan commented on 2018-04-17 11:00 (UTC)


Please mail to damentz.

sxyzy1016 commented on 2018-04-17 04:04 (UTC)

Maybe should set "CONFIG_EFIVAR_FS=y" and "CONFIG_EFI_VARS=n" in .config file. Always got into a mount error " unknow filesystem type 'efivarfs' " at the very beginning of booting up.

sir_lucjan commented on 2018-03-08 15:11 (UTC) (edited on 2018-03-08 15:15 (UTC) by sir_lucjan)


My bad. Fixed

Ashraaf commented on 2018-03-08 15:01 (UTC) (edited on 2018-03-08 15:02 (UTC) by Ashraaf)

warning: cannot resolve "linux-lqx", a dependency of "linux-lqx-test-headers"

warning: cannot resolve "linux-lqx", a dependency of "linux-lqx-test-docs"

Does the test version really need to depend on main version?

bruceutut commented on 2018-03-04 06:28 (UTC)

I use linux-lqx with ath10k driver, and wlan driver seems unable to load.

demsg shows logs for linux-lqx: [ 66.527476] ath10k_pci 0000:01:00.0: Direct firmware load for ath10k/cal-pci-0000:01:00.0.bin failed with error -2 [ 66.527490] ath10k_pci 0000:01:00.0: Falling back to user helper [ 127.967562] ath10k_pci 0000:01:00.0: Direct firmware load for ath10k/QCA9377/hw1.0/firmware-6.bin failed with error -2 [ 127.967565] ath10k_pci 0000:01:00.0: Falling back to user helper

Official kernel reports error too, but driver successfully loads. Logs here,

dmesg for official kernel: [ 2.895267] ath10k_pci 0000:01:00.0: Direct firmware load for ath10k/pre-cal-pci-0000:01:00.0.bin failed with error -2 [ 2.895313] ath10k_pci 0000:01:00.0: Direct firmware load for ath10k/cal-pci-0000:01:00.0.bin failed with error -2 [ 2.895979] ath10k_pci 0000:01:00.0: Direct firmware load for ath10k/QCA9377/hw1.0/firmware-6.bin failed with error -2 [ 2.898589] ath10k_pci 0000:01:00.0: qca9377 hw1.0 target 0x05020000 chip_id 0x003820ff sub 17aa:4035 [ 2.898597] ath10k_pci 0000:01:00.0: kconfig debug 1 debugfs 1 tracing 1 dfs 0 testmode 0 [ 2.900001] ath10k_pci 0000:01:00.0: firmware ver WLAN.TF.1.0-00267-1 api 5 features ignore-otp crc32 79cea2c7 [ 2.965420] ath10k_pci 0000:01:00.0: board_file api 2 bmi_id N/A crc32 8aedfa4a

sir_lucjan commented on 2018-03-01 00:20 (UTC)


Please try again.

kotajacob commented on 2018-03-01 00:11 (UTC)

Is the website down for anyone else? That might need updated if it is...

simona commented on 2018-02-11 14:20 (UTC)

For some strange thing "pacaur -S linux-lqx" fail check while "yaourt -S linux-lqx" was ok....

sir_lucjan commented on 2018-01-29 21:56 (UTC)


not out of date until lqx for 4.15.x is released.

sir_lucjan commented on 2018-01-18 11:51 (UTC)



gaougalos commented on 2018-01-18 08:02 (UTC)

Is there a repo for this package?

sir_lucjan commented on 2018-01-12 20:25 (UTC)


Try makepkg. I support only makepkg.

cokomoko commented on 2018-01-12 20:21 (UTC)


I am using yaourt.

I do not think I'm experiencing full disk or file system corruption.

tmpfs 3.9G 1.1G 2.9G 27% /tmp

damentz commented on 2018-01-12 19:40 (UTC)


"sha512_mb.o: file not recognized: Dosya biçemi tanınmıyor"

Maybe you have a full disk or filesystem corruption?

sir_lucjan commented on 2018-01-12 19:38 (UTC) (edited on 2018-01-12 19:39 (UTC) by sir_lucjan)


Do you use yaourt, pacaur or makepkg?

Please use English.

cokomoko commented on 2018-01-12 19:36 (UTC)

CC [M] arch/x86/crypto/sha256-mb/sha256_mb.o AS [M] arch/x86/crypto/sha256-mb/sha256_mb_mgr_flush_avx2.o CC [M] arch/x86/crypto/sha256-mb/sha256_mb_mgr_init_avx2.o AS [M] arch/x86/crypto/sha256-mb/sha256_mb_mgr_submit_avx2.o AS [M] arch/x86/crypto/sha256-mb/sha256_x8_avx2.o LD [M] arch/x86/crypto/sha256-mb/sha256-mb.o AR arch/x86/crypto/sha512-mb/built-in.o CC [M] arch/x86/crypto/sha512-mb/sha512_mb.o AS [M] arch/x86/crypto/sha512-mb/sha512_mb_mgr_flush_avx2.o CC [M] arch/x86/crypto/sha512-mb/sha512_mb_mgr_init_avx2.o AS [M] arch/x86/crypto/sha512-mb/sha512_mb_mgr_submit_avx2.o AS [M] arch/x86/crypto/sha512-mb/sha512_x4_avx2.o LD [M] arch/x86/crypto/sha512-mb/sha512-mb.o arch/x86/crypto/sha512-mb/sha512_mb.o: file not recognized: Dosya biçemi tanınmıyor make[3]: [scripts/ arch/x86/crypto/sha512-mb/sha512-mb.o] Error 1 make[2]: [scripts/ arch/x86/crypto/sha512-mb] Error 2 make[1]: [scripts/ arch/x86/crypto] Error 2 make: [Makefile:1032: arch/x86] Error 2

Where's the problem?

sir_lucjan commented on 2018-01-06 22:05 (UTC)


Did you try with 4.14.12?

hydranix commented on 2018-01-04 02:16 (UTC) (edited on 2018-01-04 13:30 (UTC) by hydranix)

I keep getting error during kernel build related to arrays overflowing in loops.

Anybody familiar with why this might be? I've been trying to track it down for a while now.

Various files fail with "array subscript is above array bounds"

simona commented on 2018-01-03 22:44 (UTC)

Using yaourt instead of pacaur problem was solved.

sir_lucjan commented on 2018-01-02 11:36 (UTC) (edited on 2018-01-02 11:38 (UTC) by sir_lucjan)

@simona Please paste full log. Please use English. Did you understand me?

sir_lucjan commented on 2017-12-29 15:56 (UTC)


Please paste full log. Please use English.

simona commented on 2017-12-29 15:53 (UTC)

==> ERRORE: Uno o più file non hanno superato il controllo di validità!

sir_lucjan commented on 2017-12-25 16:20 (UTC)

@hobbledehoy899 4.14.8-3 is the newest lqx patch:

sir_lucjan commented on 2017-12-09 16:45 (UTC) (edited on 2017-12-09 17:25 (UTC) by sir_lucjan)

Damentz sometimes updates sources and checksums do not match. Please let me know.

sir_lucjan commented on 2017-12-09 16:42 (UTC) (edited on 2017-12-14 20:53 (UTC) by sir_lucjan)

I know :P

damentz has updated sources -->

4.14.4-2.patch.gz 2017-12-08 20:54 
config.amd64    2017-12-08 20:43

I'll update ASAP. Be patient

markusk commented on 2017-12-09 16:32 (UTC)


i can confirm he config.amd64 error ! config.amd64 ... FAILED

sir_lucjan commented on 2017-12-04 11:00 (UTC)

If you have your custom config - no. If you don't have - yes.

VoltureX commented on 2017-12-03 15:11 (UTC)

Do I need config.amd64 or is everything there just by downloading the kernel and patch? Thanks. Volt.

Al3s commented on 2017-11-30 18:02 (UTC)

I edited PKGBUILD to skip the validation for config.amd64 :p

sir_lucjan commented on 2017-11-30 17:59 (UTC)

@Al3s Full log:

sir_lucjan commented on 2017-11-30 17:56 (UTC)

updpkgsums oraz makepkg -g I've tried to reproduce ... but you saw what happened :P

Al3s commented on 2017-11-30 17:53 (UTC)

I don't understand. If the problem is on my end how would I fix it?

sir_lucjan commented on 2017-11-30 17:51 (UTC)

@Al3s ==> Validating source files with sha512sums... linux-4.14.tar.xz ... Passed linux-4.14.tar.sign ... Skipped 4.14.2-1.patch.gz ... Passed config.amd64 ... Passed 60-linux.hook ... Passed 90-linux.hook ... Passed 99-linux.hook ... Passed linux.preset ... Passed It seems to be OK.

Al3s commented on 2017-11-30 17:42 (UTC)

Hi. There seems to be a problem validating config.amd64

sir_lucjan commented on 2017-11-29 11:22 (UTC)

@catweazel I've dropped nvidia-lqx. Maybe someone will adopt nvidia-lqx. Secondary question: You should edit /etc/makepkg.conf and: #MAKEFLAGS="-j2" The MAKEFLAGS option can be used to specify additional options for make. Users with multi-core/multi-processor systems can specify the number of jobs to run simultaneously. This can be accomplished with the use of nproc to determine the number of available processors, e.g. -j4 (where 4 is the output of nproc).

catweazel commented on 2017-11-29 05:47 (UTC) (edited on 2017-11-29 05:49 (UTC) by catweazel)

aur/nvidia-lqx 384.69-1 will not install the nVidia driver. The script expects a kernel version >= 4.12 and < 4.13. I've tried several times to install linux-lqx 4.13.16-1, along with linux-lqx-headers 4.13.16-1 and nvidia-lqx 384.69-1. The kernel and headers install fine, but nvidia-lqx always fails. Can this be fixed? Secondary question: Where in the install scripts can I tell the compiler to use 16 threads? My machine should compile linux-lqx in no more than 15 minutes flat but it takes close to 5 hours, which is very painful. Thanks for your assistance, and thank you for maintaining this package.

kode54 commented on 2017-11-16 02:48 (UTC)

Manually updated the script for 4.13.13-1 on my end, you can probably do the same easily, as the maintainer of the scripts. Not sure how to fix the docs and headers packages including the depmod hook when they shouldn't.

sir_lucjan commented on 2017-11-07 22:41 (UTC)

Read previous comments.

ilkyest commented on 2017-11-07 22:37 (UTC)

linux-4.13.tar ... FAILED (unknown public key 79BE3E4300411886)

sir_lucjan commented on 2017-09-26 07:53 (UTC)

I'll update linux-lqx tomorrow.

sir_lucjan commented on 2017-09-17 09:06 (UTC)


DungSherlock commented on 2017-09-16 23:59 (UTC)

>>> Updating module dependencies. Please wait ... depmod: ERROR: Bad version passed %KERNVER% error: command failed to execute correctly How to fix this!?

sir_lucjan commented on 2017-09-16 09:23 (UTC)

damentz: I afraid I can't -I can add an info.

damentz commented on 2017-09-16 03:49 (UTC)

@DungSherlock gpg --recv-keys 79BE3E4300411886 @sirlucjan Maybe we can add this part to the pkgbuild?

DungSherlock commented on 2017-09-16 03:27 (UTC)

==> Verifying source file signatures with gpg... linux-4.12.tar ... FAILED (unknown public key 79BE3E4300411886) ==> ERROR: One or more PGP signatures could not be verified! ==> ERROR: Makepkg was unable to build linux-lqx.

commented on 2017-09-13 23:29 (UTC)

@damentz Thanks ! I remove elevator=bfq for linux-lqx as well as linux-lts-ck 4.9 I have the same dmesg as yours, which bfq-sq is the default io scheduler.

commented on 2017-09-13 23:29 (UTC)

@damentz Thanks ! I remove elevator=bfq for linux-lqx as well as linux-lts-ck 4.9 I have the same dmesg as yours, which bfq-sq is the default io scheduler.

damentz commented on 2017-09-13 23:22 (UTC) (edited on 2017-09-13 23:27 (UTC) by damentz)

@liubenyuan The parameter "elevator=bfq" probably doesn't do what you think. The 'bfq' IO scheduler is now used for mainline BFQ in the blk-mq subsystem, and is selected by through some simple logic in the code, overriding what you pick. Also, the 'elevator' parameter only affects legacy block. For 4.12, I switched from blk-mq back to blk-sq (legacy block) due to issues with suspend on 4.12. In order to get bfq working, I backported bfq-sq (the new names for bfq on blk-sq). So in order to get the same behavior, which is already default, you must add "elevator=bfq-sq" for now. This will probably change though, so I recommend not adding any explicit elevator on Liquorix since this is subject to change kernel version to kernel version as bugs are fixed. Also note, you can see which elevator is default by checking dmesg. However, the output in dmesg does not differentiate between blk-mq and blk-sq, so you'll need to do your research if you have any problems: $ dmesg | grep 'io scheduler' [ 9.418222] io scheduler noop registered [ 9.418222] io scheduler deadline registered [ 9.418238] io scheduler cfq registered [ 9.418239] io scheduler mq-deadline registered [ 9.418239] io scheduler kyber registered [ 9.418252] io scheduler bfq registered [ 9.418263] io scheduler bfq-sq registered (default)

commented on 2017-09-13 23:14 (UTC)

@sir_lucjan The error message says that it can not mount 'efivars'. And it stays at saying Start version 243. And when I remove elevator=bfq in my boot config, linux-lqx can boot! The boot command of my linux-lqx was #options root=/dev/sda2 ro quiet ipv6.disable=1 elevator=bfq init=/usr/lib/systemd/systemd acpi_osi=Linux acpi_backlight=vendor BTW, linux-bfq boots fine using 4.11.12

sir_lucjan commented on 2017-09-13 15:59 (UTC)

You should paste this error message.

commented on 2017-09-13 15:51 (UTC)

My linux-lqx 4.12.12-1 failed to start after upgrading from 4.11.11. How could I debug it?

sir_lucjan commented on 2017-09-04 11:24 (UTC)

@@cpatrick08 - 4.12.10 is the newest version:

sir_lucjan commented on 2017-09-04 09:43 (UTC)

@cpatrick08 - Please don't flag out-of-date until the new version is moved to [core]

sir_lucjan commented on 2017-08-20 11:03 (UTC)

@wudu - Now checksums are OK but when Damentz will push a new version... :D

sir_lucjan commented on 2017-08-20 10:56 (UTC) (edited on 2017-08-20 10:59 (UTC) by sir_lucjan)

@wudu - I've checked sums and everything was ok (look at my previous comment). Damentz has updated configs files when he's pushed new version of lqx patchset (4.12.8-3). I always check sums but sometimes Damentz bump his patch or change config. I'll updated sources ASAP.

wudu commented on 2017-08-20 10:49 (UTC) (edited on 2017-08-20 10:50 (UTC) by wudu)

4.12.8-1 checksum failed config.i386 ... FAILED config.i386-pae ... FAILED config.amd64 ... FAILED Why don't you check your PKGBUILD as maintainer? Don't get it

sir_lucjan commented on 2017-08-17 12:34 (UTC)

I've checked and seems to be OK...

sir_lucjan commented on 2017-08-17 10:34 (UTC)


EgoistAnarchist commented on 2017-08-17 03:44 (UTC)

4.12.8-1 patch release. Should/Can this be updated?

bruceutut commented on 2017-08-10 02:53 (UTC)

4.11.12 patch release, update plz. thanks

whoami commented on 2017-06-05 15:32 (UTC)

updpkgsums works for me. Thanks.

sir_lucjan commented on 2017-06-05 15:27 (UTC)

@atommixz: Probably damentz has changed sources. Please make: updpkgsums I'll upload a new version (4.11.3-3 or 4.11.4-1)

whoami commented on 2017-06-05 15:21 (UTC)

do rm -rf linux-lqx git clone cd linux-lqx makepkg -si and have 4.11.3-2.patch.gz ... FAILED config.i386 ... FAILED config.i386-pae ... FAILED config.amd64 ... FAILED

raku-cat commented on 2017-06-05 06:12 (UTC)

Anyone had issues with a complete lock up on resume from suspend to disk after the jump to 4.11.x?

erkexzcx commented on 2017-06-04 06:23 (UTC)

Wondering if there is anyone who would like to create an unofficial repo with precompiled packages for/with lqx patchset :D

damentz commented on 2017-06-02 17:03 (UTC) (edited on 2017-06-02 17:32 (UTC) by damentz)

@sir_lucjan, I have a relatively crude script I've been using to push out patches and config files to It always attempts to gzip the patch and push it out, causing the checksum to change every time I compress. I checked the diff between 4.11-9 and 4.11-8, and the only things changed were the debian/changelog file and the configurations, so even though the checksum changed, the patch is essentially identical. If you want to update the package to use the github releases, the only files you need to worry about are: Liquorix release: Name of tag in Kernel version: linux-liquorix/debian/config/defines (value after "abiname: ") Patch name: linux-liquorix/debian/patches/series (patch file prefixed with "zen/") Config files: linux-liquorix/debian/config/kernelarch-x86/config-arch-32 # 32-bit linux-liquorix/debian/config/kernelarch-x86/config-arch-32-pae # 32-bit + PAE linux-liquorix/debian/config/kernelarch-x86/config-arch-64 # 64-bit EDIT: You can find the latest tag tarball URLs (releases doesn't seem to work for me), by running: curl "" | grep -Eo 'https://.*tarball[^"]+'. Using this method, the only thing you would need to change on every release is the tag name, everything else can be derived from the package contents.

id942701 commented on 2017-06-02 08:30 (UTC)

@sir_lucjan Thanks, it works. But hash sum mismatch looks suspicious.

sir_lucjan commented on 2017-06-01 14:20 (UTC)

@damentz: Why :)? @id942701, @Freso updpkgsums

damentz commented on 2017-06-01 13:35 (UTC)

Hmm, it looks like I may have re-diffed 4.11.3-2 by accident. Either way, this package should be pulling the sources from the GitHub releases page here:

id942701 commented on 2017-06-01 13:22 (UTC)

==> Making package: linux-lqx 4.11.3-2 (Thu Jun 1 14:14:09 +01 2017) Validating source files with sha512sums... linux-4.11.tar.xz ... Passed linux-4.11.tar.sign ... Skipped 4.11.3-2.patch.gz ... FAILED config.i386 ... FAILED config.i386-pae ... FAILED config.amd64 ... FAILED linux.preset ... Passed 90-linux.hook ... Passed ==> ERROR: One or more files did not pass the validity check! :: failed to verify linux-lqx-headers,linux-lqx integrity

Freso commented on 2017-06-01 07:11 (UTC)

The sha512 sum for 4.11.3-2.patch.gz seems to be incorrect. If I sha512sum it, I get "06e01068577baa5bd5d747e108c50e3af7f23b7d393edce7cef1c74aa1083ed9f457c3e9bb2a84884539f2d22dc53b04b3b0606494502d64d643b6dd8c06324c".

sir_lucjan commented on 2017-05-21 13:47 (UTC)

Bump to 4.11.2

sir_lucjan commented on 2017-05-16 17:49 (UTC)

Bump to 4.10.16-1

sir_lucjan commented on 2017-05-10 15:42 (UTC)

kazuo - you should ask damentz.

whoami commented on 2017-04-28 08:34 (UTC)

Now I understand what my problem with linux-lqx. He can't mount /boot on ext2 with this parametrs and going to emergency mode - "rw,relatime,block_validity,barrier,user_xattr,acl,discard", but he can with this "rw,relatime,user_xattr,acl" I thing, long string was in fresh archlinux install. Hooooray!!! Linux atmxz 4.10.11-1-lqx #1 ZEN SMP PREEMPT Thu Apr 27 21:18:06 MSK 2017 x86_64 GNU/Linux

ronoverdrive commented on 2017-03-10 00:30 (UTC)

@Pival81: common problem, just delete your old config.* files and let makepkg or whatever AUR package manager you use redownload them.

Pival81 commented on 2017-03-09 14:41 (UTC)

==> Validating source files with sha512sums... linux-4.9.tar.xz ... Passed linux-4.9.tar.sign ... Skipped 4.9.13-1.patch.gz ... Passed config.i386 ... FAILED config.i386-pae ... FAILED config.amd64 ... FAILED linux.preset ... Passed 99-linux.hook ... Passed ==> ERROR: One or more files did not pass the validity check! ==> ERROR: Makepkg was unable to build linux-lqx. ==> Restart building linux-lqx ? [y/N] ==> ---------------------------------- ==>

whoami commented on 2017-02-25 21:55 (UTC) (edited on 2017-02-25 22:00 (UTC) by whoami)

It's just clear/new SSD disk with old am3 motherboard sata 3.1 on-board controller and clear/new ArchLinux installation with /boot on ext2, swap and / on ext4. Everithing should work, but only dis kernel not works for me. You are guys forget about it. It's not metter anymore for me.

damentz commented on 2017-02-23 16:19 (UTC)

@sir_lucjan One thing I've noticed regarding the %KERNVER% is that it occurs when installing linux-lqx-headers and linux-lqx-docs only. I think those packages don't need to be installed the same way linux-lqx is. @atommixz I think your issue might be caused by a configuration on your system. Most likely, your initramfs does not contain the modules you need to access your drives. Seeing as how no on else is experiencing the problem, have you tried installing linux-lqx on a brand new arch installation to remove the chance a local modification is breaking your boot?

sir_lucjan commented on 2017-02-20 22:13 (UTC)

[lucek@lucek ~]$ uname -a Linux lucek 4.9.11-1-lqx #1 ZEN SMP PREEMPT Mon Feb 20 22:59:38 CET 2017 x86_64 GNU/Linux [lucek@lucek ~]$ pacman -Qs linux-lqx local/linux-lqx 4.9.11-1 A desktop oriented kernel and modules with Liquorix patches local/linux-lqx-docs 4.9.11-1 Kernel hackers manual - HTML documentation that comes with the linux-lqx kernel local/linux-lqx-headers 4.9.11-1 Header files and scripts to build modules for linux-lqx. [lucek@lucek ~]$ systemctl status display-manager ● lightdm-plymouth.service - Light Display Manager Loaded: loaded (/usr/lib/systemd/system/lightdm-plymouth.service; enabled; ve Active: active (running) since pon 2017-02-20 23:09:55 CET; 1min 59s ago Main PID: 397 (lightdm) Tasks: 4 (limit: 4915) CGroup: /system.slice/lightdm-plymouth.service ├─397 /usr/bin/lightdm └─409 /usr/lib/xorg-server/Xorg :0 -seat seat0 -auth /run/lightdm/roo lut 20 23:09:54 lucek systemd[1]: Starting Light Display Manager... lut 20 23:09:55 lucek systemd[1]: Started Light Display Manager. lut 20 23:09:58 lucek lightdm[442]: pam_unix(lightdm-greeter:session): sessio lut 20 23:10:07 lucek lightdm[568]: pam_unix(lightdm:session): session opened As you can see....

whoami commented on 2017-02-20 20:40 (UTC)

I'm don't found linux-bfq in repos/aur/wiki, so I'm compile linux-ck. It's boot properly immediately. With default /etc/mkinitcpio.conf. $ systemctl status boot.mount ● boot.mount - /boot Loaded: loaded (/etc/fstab; generated; vendor preset: disabled) Active: active (mounted) since Mon 2017-02-20 23:34:23 MSK; 4min 58s ago Where: /boot What: /dev/sdc2 Docs: man:fstab(5) man:systemd-fstab-generator(8) Process: 273 ExecMount=/usr/bin/mount /dev/disk/by-uuid/8fd6580e-35ab-4edf-a0be-b0dcfe691 Tasks: 0 (limit: 4915) CGroup: /system.slice/boot.mount $ uname -a Linux atmxz 4.9.11-1-ck #1 SMP PREEMPT Mon Feb 20 23:23:26 MSK 2017 x86_64 GNU/Linux It's soooo faaaast if compared with default arch kernel. It's just WOW effect.

sir_lucjan commented on 2017-02-20 10:24 (UTC)

Could you try with another kernel, for example linux-bfq?

whoami commented on 2017-02-19 20:14 (UTC)

u can scroll my msgs

sir_lucjan commented on 2017-02-19 19:57 (UTC)

Show me: ls -l /boot

whoami commented on 2017-02-19 19:48 (UTC)

Works for me on default kernel. Error msg only can't mount /boot. And running Single Mode. ls -l /boot empty on boot in linux-lqx. I compile it as makepkg -sifC # systemctl status boot.mount ● boot.mount - /boot Loaded: loaded (/etc/fstab; generated; vendor preset: disabled) Active: failed (Result: exit-code) since Sat 2017-02-18 22:22:59 MSK; 2min 25s ago Where: /boot What: /dev/disk/by-uuid/8fd6580e-35ab-4edf-a0be-b0dcfe691886 Docs: man:fstab(5) man:systemd-fstab-generator(8) Process: 305 ExecMount=/usr/bin/mount /dev/disk/by-uuid/8fd6580e-35ab-4edf-a0be-b0dcfe691886 /boot -t ext2 -o rw,relatime,block_validity,barrier,user_xattr,acl (code=exited, status=32) Feb 18 22:22:59 atmxz systemd[1]: Mounting /boot... Feb 18 22:22:59 atmxz systemd[1]: boot.mount: Mount process exited, code=exited status=32 Feb 18 22:22:59 atmxz systemd[1]: Failed to mount /boot. Feb 18 22:22:59 atmxz systemd[1]: boot.mount: Unit entered failed state. $ ls -lA /boot/ total 46736 drwxr-xr-x 6 root root 1024 Feb 18 22:37 grub -rw-r--r-- 1 root root 14881152 Feb 18 22:54 initramfs-linux-fallback.img -rw-r--r-- 1 root root 15447532 Feb 19 19:34 initramfs-linux-lqx-fallback.img -rw-r--r-- 1 root root 3954396 Feb 19 19:34 initramfs-linux-lqx.img -rw-r--r-- 1 root root 4072024 Feb 18 22:53 initramfs-linux.img drwx------ 2 root root 12288 Feb 18 08:02 lost+found -rw-r--r-- 1 root root 4863776 Feb 6 15:01 vmlinuz-linux -rw-r--r-- 1 root root 4428576 Feb 19 19:18 vmlinuz-linux-lqx /boot/grub/grub.cfg (generate as # grub-mkconfig -o /boot/grub/grub.cfg ... menuentry 'Arch Linux' --class arch --class gnu-linux --class gnu --class os $menuentry_id_option 'gnulinux-simple-bd56223e-d6fd-4170-bb6c-f7de646ac28d' { load_video set gfxpayload=keep insmod gzio insmod part_gpt insmod ext2 set root='hd2,gpt2' if [ x$feature_platform_search_hint = xy ]; then search --no-floppy --fs-uuid --set=root --hint-bios=hd2,gpt2 --hint-efi=hd2,gpt2 --hint-baremetal=ahci2,gpt2 8fd6580e-35ab-4edf-a0be-b0dcfe691886 else search --no-floppy --fs-uuid --set=root 8fd6580e-35ab-4edf-a0be-b0dcfe691886 fi echo 'Loading Linux linux-lqx ...' linux /vmlinuz-linux-lqx root=UUID=bd56223e-d6fd-4170-bb6c-f7de646ac28d rw quiet echo 'Loading initial ramdisk ...' initrd /initramfs-linux-lqx.img } ... /etc/fstab # /dev/sdc4 UUID=bd56223e-d6fd-4170-bb6c-f7de646ac28d / ext4 rw,relatime,data=ordered 0 1 # /dev/sdc2 UUID=8fd6580e-35ab-4edf-a0be-b0dcfe691886 /boot ext2 rw,relatime,block_validity,barrier,user_xattr,acl 0 2 ...

sir_lucjan commented on 2017-02-19 18:14 (UTC)

I know this "depmod error" but I don't have problem with booting. Did you try with another kernels? Could you show me error's message?

whoami commented on 2017-02-19 18:11 (UTC)

No I'm not try 4.9.7 cause I'm new guy with this kernel. Should I do this?

sir_lucjan commented on 2017-02-19 17:31 (UTC)

Can anyone confirm this 'boot error'?

sir_lucjan commented on 2017-02-19 17:26 (UTC)

sir_lucjan commented on 2017-02-19 17:25 (UTC)

Did you try with 4.9.7?

whoami commented on 2017-02-19 17:24 (UTC)

I have issue "depmod: ERROR: Bad version passed %KERNVER%" while install linux-lqx-headers-4.9.11-1 and linux-lqx-docs-4.9.11-1 but it installed. Also I can't boot 4.9.8, 4.9.10, 4.9.11. I have /boot on ext2 with GPT Partition table # /dev/sdc4 UUID=bd56223e-d6fd-4170-bb6c-f7de646ac28d / ext4 rw,relatime,data=ordered 0 1 # /dev/sdc2 UUID=8fd6580e-35ab-4edf-a0be-b0dcfe691886 /boot ext2 rw,relatime,block_validity,barrier,user_xattr,acl 0 2 # systemctl status boot.mount ● boot.mount - /boot Loaded: loaded (/etc/fstab; generated; vendor preset: disabled) Active: failed (Result: exit-code) since Sat 2017-02-18 22:22:59 MSK; 2min 25s ago Where: /boot What: /dev/disk/by-uuid/8fd6580e-35ab-4edf-a0be-b0dcfe691886 Docs: man:fstab(5) man:systemd-fstab-generator(8) Process: 305 ExecMount=/usr/bin/mount /dev/disk/by-uuid/8fd6580e-35ab-4edf-a0be-b0dcfe691886 /boot -t ext2 -o rw,relatime,block_validity,barrier,user_xattr,acl (code=exited, status=32) Feb 18 22:22:59 atmxz systemd[1]: Mounting /boot... Feb 18 22:22:59 atmxz systemd[1]: boot.mount: Mount process exited, code=exited status=32 Feb 18 22:22:59 atmxz systemd[1]: Failed to mount /boot. Feb 18 22:22:59 atmxz systemd[1]: boot.mount: Unit entered failed state. also I add MODULES="ext2" into mkinitcpio.conf and do mkinitcpio -p linux-lqx, but still can't boot. UUID is correct.

sir_lucjan commented on 2017-02-18 19:57 (UTC)

@atommixz: Please don't flag as out to date when it's up to date. The newest liquorix patch is 4.9.10 (18th of Feb, 20:57):;O=D

jeroni commented on 2017-02-10 09:28 (UTC) (edited on 2017-02-10 09:38 (UTC) by jeroni)

Thanks @spinvis I already tried it as @sir_lucjan suggested, but I got the same error by using makepkg. UPDATE: cleaning the code worked fine!

spinvis commented on 2017-02-10 07:43 (UTC)

Thanks sir_lucjan, works perfect. @jeroni, use makepkg instead.

jeroni commented on 2017-02-09 22:45 (UTC)

I also have the same error as @spinvis patching file virt/lib/irqbypass.c ==> ERROR: Se produjo un fallo en prepare().

sir_lucjan commented on 2017-02-08 18:30 (UTC)

Works fine. You should use a makepkg.

spinvis commented on 2017-02-08 18:23 (UTC)

linux-lqx 4.9.8-1 keeps giving me the following error patching file virt/kvm/arm/vgic/vgic-init.c patching file virt/kvm/arm/vgic/vgic-v2.c patching file virt/kvm/arm/vgic/vgic-v3.c patching file virt/lib/irqbypass.c ==> ERROR: A failure occurred in prepare(). Aborting... :: bc is now an orphan package :: failed to build linux-lqx package(s) I also see a lot of the following messages (not sure if it is relevant) in the output further up: The next patch would create the file kernel/skip_list.c, which already exists! Skipping patch. 1 out of 1 hunk ignored

sir_lucjan commented on 2017-01-04 15:40 (UTC)

@BenCranston Did previous version work correctly?

sir_lucjan commented on 2017-01-03 19:57 (UTC)

@metak: currently nvidia 304xx doens't build without an update to 304.134 and currently nvidia-304xx-utils in [extra] is stuck on 304.132. This means I can't push the 4.9 update until our devs put 304.134 in the repos.

BenCranston commented on 2017-01-02 19:21 (UTC)

4.8.15-1 green screens in early boot on both yaourt and makepkg builds. Boot looks like this: Loading Linux linux-lqx ... Loading initial ramdisk ... starting version 232 ROOT: clean, 1057370/37994496 files, 51054981/151977984 block *green screen lockup at this point* I am using mdadm raid on this volume but no LVM Let me know what I can provide to help out. Stock arch linux kernel boots just fine from grub menu

sir_lucjan commented on 2016-12-10 12:20 (UTC)

Bump to 4.8.13

sir_lucjan commented on 2016-12-05 15:52 (UTC)

Bump to 4.8.12

sir_lucjan commented on 2016-12-02 15:37 (UTC)

I'll update with 4.8.12 :)

Freso commented on 2016-12-02 08:14 (UTC)

Both and work fine with https:// AFAICT. Perhaps update $sources array to use HTTPS?

sir_lucjan commented on 2016-12-01 15:45 (UTC)

Não há problema

the_wolf commented on 2016-12-01 15:42 (UTC)

rsrsrs ... Sorry. It is very clear the same. I have to pay more attention. Hug ...

sir_lucjan commented on 2016-12-01 15:27 (UTC)

"_config="pkg" # "local": compile only probed modules(" It's pretty clear. BUG:

the_wolf commented on 2016-12-01 04:00 (UTC)

When I went to install the kernel-headers of this error. [root@XXXY:linux-lqx]# pacman -U linux-lqx-headers-4.8.11-1-x86_64.pkg.tar.xz carregando pacotes... resolvendo dependências... procurando por pacotes conflitantes... Pacotes (1) linux-lqx-headers-4.8.11-1 Tamanho total instalado: 50,11 MiB :: Continuar a instalação? [S/n] (1/1) verificando chaves no chaveiro [#################################################################] 100% (1/1) verificando integridade do pacote [#################################################################] 100% (1/1) carregando arquivos do pacote [#################################################################] 100% (1/1) verificando conflitos de arquivo [#################################################################] 100% (1/1) verificando espaço em disco disponível [#################################################################] 100% :: Processando alterações do pacote... (1/1) instalando linux-lqx-headers [#################################################################] 100% >>> Updating module dependencies. Please wait ... depmod: ERROR: Bad version passed %KERNVER% erro: comando não executado corretamente :: Executando hooks pós-transação... (1/1) Install DKMS modules

the_wolf commented on 2016-12-01 02:34 (UTC)

Sorry, it does not have the local parameter in the description. hug

sir_lucjan commented on 2016-11-30 18:24 (UTC)

the_wolf Please read pkgbuild carefully :)

the_wolf commented on 2016-11-30 18:11 (UTC)

To implement, is it possible ?? hug

sir_lucjan commented on 2016-11-22 21:36 (UTC)

I'll try fix it. Be patient.

kazuo commented on 2016-11-22 16:52 (UTC)

I'm getting a error on install of linux-lqx-headers depmod: ERROR: Bad version passed %KERNVER% Looks like to sed magic is not working? Looking at the .INSTALL it still have the placeholder tags (the main kernel is ok, but both the headers and the docs have the placeholders)

sir_lucjan commented on 2016-11-18 16:42 (UTC)

Add pacman-hook

sir_lucjan commented on 2016-11-13 19:01 (UTC)


hydranix commented on 2016-11-13 18:54 (UTC) appears to be down. Is this temporary?

sir_lucjan commented on 2016-10-31 12:32 (UTC)

Hi @xxxtonixxx, thx for the info. Fixed.

xxxtonixxx commented on 2016-10-31 11:59 (UTC)

Hi @sir_lucjan ! modprobed-db doesn't work because it's modprobed-db no modprobed_db

sir_lucjan commented on 2016-10-30 10:35 (UTC)

I'll bump to 4.8.5-2 tomorrow.

sir_lucjan commented on 2016-10-26 15:27 (UTC)

@damentz Done. Could you check?

Anthony25 commented on 2016-10-26 15:11 (UTC) (edited on 2016-10-26 20:05 (UTC) by Anthony25)

@damentz: already done since 4.8.3-2, thanks Edit: my bad, misread your question

damentz commented on 2016-10-26 15:10 (UTC)

Just a heads up on the latest package, MuQSS is now the default in Liquorix, so the PKGBUILD should probably be updated to handle both CFS or MuQSS being enabled, and swap the config lines as needed.

sir_lucjan commented on 2016-10-22 18:32 (UTC) (edited on 2016-10-22 18:32 (UTC) by sir_lucjan)

You're welcome. Please let me know if you'll got an error.

Anthony25 commented on 2016-10-22 18:29 (UTC)

@sir_lucjan: All good, many thanks! Plus, this version seems to fix the freezes I got with BFS.

sir_lucjan commented on 2016-10-22 18:05 (UTC)

@Anthony25 Could you try and report?

sir_lucjan commented on 2016-10-22 17:59 (UTC) (edited on 2016-10-22 17:59 (UTC) by sir_lucjan)

@Anthony25 Please don't flag package without reason

Anthony25 commented on 2016-10-22 17:41 (UTC) (edited on 2016-10-22 18:02 (UTC) by Anthony25)

Hello, You might want to update your PKGBUILD for options related to BFS, as Muqss has been merged in linux-zen, replacing BFS: Thanks! Edit: flagged the pkg out-of-date for this reason, because enabling BFS in the PKGBUILD will do nothing on 4.8.3. I'm sorry about that.

TheGreatMcPain commented on 2016-10-20 19:18 (UTC) (edited on 2016-10-21 00:20 (UTC) by TheGreatMcPain)

I'm also getting freezing with BFS. On my i5 3570k desktop it freezes when running x264, or simply "sudo poweroff". On my Pentium P6200 laptop it will freeze when opening chrome, or wiggling the terminal around. I could not find any useful info in journalctl though. Edit: After rebuilding with CFS the freezing still happens on my Desktop, not sure on the laptop yet, this time it "unfreezes" after a hour of being frozen.

Anthony25 commented on 2016-10-17 21:52 (UTC)

Hello, I get some freezes at boot when enabling BFS, when I did not have any issue on 4.7.7. Going back to CFS fixes the issue. I don't know how to get some logs, but on 10 boots, 8 or 9 are failing/freezing. I saw that CK released some recent patches for BFS (like 10 days ago) that fixed some freezes. Are they merged in liquorix yet?

sir_lucjan commented on 2016-10-17 15:52 (UTC)

Bump to 4.8.2

sir_lucjan commented on 2016-10-17 09:27 (UTC)

No problem. I'll push a new tarball in a couple hours.

metak commented on 2016-10-16 02:16 (UTC) (edited on 2016-10-16 04:13 (UTC) by metak)

Sorry about that sir_lucjan. Flagged it before I saw your comment. Edit: updated pkgbuild file to 4.8.1 and it builds fine on my end.

damentz commented on 2016-10-16 00:30 (UTC)

@neognomic, net/ipv4/tcp_input.c is not changed by by Zen or by Liquorix. All I can suspect is that your environment may be contributing to the build error and connectivity issues. Try the 4.8 series when sir_lucjan bumps it next week, if this kernel specifically is still causing issues for you, I'll take a look.

neognomic commented on 2016-10-04 00:28 (UTC)

This is apparently more than just a warning: [code] CC net/ipv4/tcp_input.o net/ipv4/tcp_input.c: In function ‘tcp_data_queue’: net/ipv4/tcp_input.c:4283:49: warning: array subscript is above array bounds [-Warray-bounds] tp->selective_acks[i-1] = tp->selective_acks[i]; [/code] as the network is broken. ~messages~ "WIRED CONNECTION DEACTIVATED" "IP Configuration UNAVAILABLE" ...unable to obtain ipaddress. I suppose linux44, linux47, linux-mainline and linux-ck working as expected would be ample to indicate that it is this kernel. { Only mod to PKGBUILD was for localmodconfig(same as used for linux-ck and linux-mainline(linux48)] }. Thanks anyway; LQX looks almost ideal for desktop usage.

dmitry3l commented on 2016-09-11 08:24 (UTC)

@MrEvilBreakfast apply this patch

MrEvilBreakfast commented on 2016-09-10 17:50 (UTC)

Hi, I just compiled the kernel on my netbook and it runs beautifully, but I have a question. How can I stop the absolute flood of console messages with this kernel? I use the console a lot and I find it a little irritating that this kernel is so verbose. Thanks, and keep up the good work!

sir_lucjan commented on 2016-09-07 08:23 (UTC)

Did you change default config? Did you try with older version?

UnixMan commented on 2016-09-07 05:35 (UTC)

I'm getting this error: nano /etc/default/grub added: GRUB_CMDLINE_LINUX_DEFAULT="zswap.enabled=1 quiet splash" grub-mkconfig -o /boot/grub/grub.cfg OS: Arch Linux x86_64 CPU: Intel(R) Celeron(R) CPU 847 @ 1.10GHz

sir_lucjan commented on 2016-08-15 15:13 (UTC)

Bump to 4.7.0

sir_lucjan commented on 2016-08-12 10:36 (UTC)

Bump to 4.6.6

sir_lucjan commented on 2016-07-26 14:13 (UTC)

updpkgsums && makepkg -si

ltorvalds024 commented on 2016-07-26 11:47 (UTC)

I am getting this error: $ git clone $ cd linux-lqx $ makepkg -si ... ==> Validating source files with sha512sums... linux-4.6.tar.xz ... Passed linux-4.6.tar.sign ... Skipped 4.6.4-2.patch.gz ... Passed config.i386 ... FAILED config.i386-pae ... FAILED config.amd64 ... FAILED linux-lqx.preset ... Passed ==> ERROR: One or more files did not pass the validity check! ...

eeleater commented on 2016-07-13 21:30 (UTC) (edited on 2016-07-13 21:30 (UTC) by eeleater)

I built it on my server now, no idea what was wrong with my laptop, I was using makepkg on both btw.

sir_lucjan commented on 2016-07-13 16:32 (UTC)

eelater: I've compiled without errors. Did you use makepkg or yaourt?

eeleater commented on 2016-07-12 21:50 (UTC)

/bin/sh: scripts/basic/fixdep: No such file or directory make[3]: *** [scripts/ drivers/acpi/acpica/nsxfeval.o] Error 1 make[3]: *** Waiting for unfinished jobs.... make[3]: *** No rule to make target 'sound/firewire/bebob/bebob_maudio.c', needed by 'sound/firewire/bebob/bebob_maudio.o'. Stop. make[3]: *** Waiting for unfinished jobs.... CC [M] sound/firewire/bebob/bebob_focusrite.o gcc: error: sound/firewire/bebob/bebob_focusrite.c: No such file or directory gcc: fatal error: no input files compilation terminated. make[3]: *** [scripts/ sound/firewire/bebob/bebob_focusrite.o] Error 1 make[3]: *** No rule to make target 'sound/firewire/digi00x/digi00x-pcm.c', needed by 'sound/firewire/digi00x/digi00x-pcm.o'. Stop. make[3]: *** Waiting for unfinished jobs.... CC [M] sound/firewire/digi00x/digi00x-proc.o CC [M] drivers/atm/solos-pci.o gcc: error: sound/firewire/digi00x/digi00x-proc.c: No such file or directory gcc: fatal error: no input files compilation terminated. In file included from sound/firewire/bebob/bebob_yamaha.c:9:0: sound/firewire/bebob/./bebob.h:30:20: fatal error: ../lib.h: No such file or directory make[3]: *** [scripts/ sound/firewire/digi00x/digi00x-proc.o] Error 1 compilation terminated. make[2]: *** [scripts/ sound/firewire/digi00x] Error 2 make[2]: *** Waiting for unfinished jobs.... make[3]: *** [scripts/ sound/firewire/bebob/bebob_yamaha.o] Error 1 make[2]: *** [scripts/ sound/firewire/bebob] Error 2 make[1]: *** [scripts/ sound/firewire] Error 2 make: *** [Makefile:965: sound] Error 2 make: *** Waiting for unfinished jobs.... CC [M] drivers/atm/idt77105.o In file included from <command-line>:0:0: ././include/linux/kconfig.h:4:32: fatal error: generated/autoconf.h: No such file or directory #include <generated/autoconf.h> ^ compilation terminated. /bin/sh: scripts/basic/fixdep: No such file or directory make[2]: *** [scripts/ drivers/ata/sata_via.o] Error 1 make[1]: *** [scripts/ drivers/ata] Error 2 make[1]: *** Waiting for unfinished jobs.... make[2]: *** [scripts/ drivers/atm/solos-pci.o] Error 1 make[2]: *** Waiting for unfinished jobs.... In file included from <command-line>:0:0: ././include/linux/kconfig.h:4:32: fatal error: generated/autoconf.h: No such file or directory #include <generated/autoconf.h> ^ compilation terminated. make[2]: *** [scripts/ drivers/atm/idt77105.o] Error 1 CC [M] drivers/atm/adummy.o CC [M] fs/btrfs/volumes.o CC drivers/acpi/hed.o CC [M] fs/btrfs/async-thread.o In file included from <command-line>:0:0: ././include/linux/kconfig.h:4:32: fatal error: generated/autoconf.h: No such file or directory #include <generated/autoconf.h> ^ compilation terminated. In file included from <command-line>:0:0: ././include/linux/kconfig.h:4:32: fatal error: generated/autoconf.h: No such file or directory #include <generated/autoconf.h> ^ compilation terminated. In file included from <command-line>:0:0: ././include/linux/kconfig.h:4:32: fatal error: generated/autoconf.h: No such file or directory #include <generated/autoconf.h> ^ compilation terminated. make[2]: *** [scripts/ drivers/atm/adummy.o] Error 1 make[2]: *** [scripts/ drivers/acpi/hed.o] Error 1 make[2]: *** Waiting for unfinished jobs.... /bin/sh: scripts/basic/fixdep: No such file or directory make[3]: *** [scripts/ drivers/acpi/acpica/nsxfname.o] Error 1 make[2]: scripts/ No such file or directory make[2]: *** No rule to make target 'scripts/'. Stop. make[1]: *** [scripts/ fs/cachefiles] Error 2 make[1]: *** Waiting for unfinished jobs.... make[2]: *** [scripts/ drivers/acpi/acpica] Error 2 make[1]: *** [scripts/ drivers/acpi] Error 2 CC [M] fs/btrfs/ioctl.o cc1: fatal error: ./include/linux/kconfig.h: No such file or directory compilation terminated. make[2]: *** [scripts/ fs/btrfs/async-thread.o] Error 1 make[2]: *** Waiting for unfinished jobs.... CC [M] fs/btrfs/locking.o make[2]: *** [scripts/ fs/btrfs/volumes.o] Error 1 cc1: fatal error: ./include/linux/kconfig.h: No such file or directory compilation terminated. CC [M] fs/btrfs/orphan.o make[2]: *** [scripts/ fs/btrfs/ioctl.o] Error 1 cc1: fatal error: fs/btrfs/orphan.c: No such file or directory compilation terminated. cc1: fatal error: ./include/linux/kconfig.h: No such file or directory compilation terminated. Assembler messages: Fatal error: can't create fs/btrfs/.tmp_orphan.o: No such file or directory make[2]: *** [scripts/ fs/btrfs/orphan.o] Error 1 make[2]: *** [scripts/ fs/btrfs/locking.o] Error 1 fs/btrfs/extent_io.c:5789:1: fatal error: opening dependency file fs/btrfs/.extent_io.o.d: No such file or directory } ^ compilation terminated. make[2]: *** [scripts/ fs/btrfs/extent_io.o] Error 1 make[1]: *** [scripts/ fs/btrfs] Error 2 make: *** [Makefile:965: fs] Error 2 drivers/atm/eni.c:2340:1: fatal error: opening dependency file drivers/atm/.eni.o.d: No such file or directory MODULE_LICENSE("GPL"); ^~~~~~~~~~~~~~ compilation terminated. make[2]: *** [scripts/ drivers/atm/eni.o] Error 1 drivers/atm/idt77252.c:3804:1: fatal error: opening dependency file drivers/atm/.idt77252.o.d: No such file or directory MODULE_DESCRIPTION("IDT77252 ABR SAR Driver"); ^~~~~~~~~~~~~~~~~~ compilation terminated. make[2]: *** [scripts/ drivers/atm/idt77252.o] Error 1 make[1]: *** [scripts/ drivers/atm] Error 2 make: *** [Makefile:965: drivers] Error 2 ==> ERROR: A failure occurred in build(). Aborting... ==> ERROR: Makepkg was unable to build linux-lqx. What to do?

sir_lucjan commented on 2016-07-12 15:35 (UTC)

Not a bug.

Pival81 commented on 2016-07-12 15:32 (UTC)

==> Verifying source file signatures with gpg... linux-4.6.tar ... FAILED (unknown public key 79BE3E4300411886) ==> ERROR: One or more PGP signatures could not be verified! ==> ERROR: Makepkg was unable to build linux-lqx. ==> Restart building linux-lqx ? [y/N] ==> ---------------------------------- ==>

sir_lucjan commented on 2016-07-12 14:33 (UTC)

Bump to 4.6.4

damentz commented on 2016-07-01 00:40 (UTC) (edited on 2016-07-01 00:42 (UTC) by damentz)

@sir_lucjan I don't specifically remember changing the configuration since the last release, but I do recommend updating the package to use github tags here: Each tag on github has an associated zip and tar.gz file, like so: In each of these is the config folder, linux-liquorix/debian/config/kernelarch-x86. I do understand though that this will take time to implement. If I get some free time I'll try my stab at updating the kernel package to use this method. EDIT: Forgot to mention, the patch is also available in the folder, linux-liquorix/debian/patches/zen.

damentz commented on 2016-07-01 00:35 (UTC)

@Junio_Calu Nice catch, I wasn't aware that the graysky2 patches added their own native architecture configuration. I committed a change to the zen source to remove the ZEN specific option since it duplicates the option in graysky2's (which is preferred since it adds more specific architectures). Change commit ID: Next version of the liquorix package will have this change merged.

sir_lucjan commented on 2016-06-30 08:46 (UTC) (edited on 2016-06-30 08:47 (UTC) by sir_lucjan)

@saburouta Probably damentz has changed config files.

saburouta commented on 2016-06-30 04:32 (UTC)

Is there a reason the config files do not match the checksum array? Does that happen often?

JunioCalu commented on 2016-06-29 02:26 (UTC) (edited on 2016-06-29 02:26 (UTC) by JunioCalu)

damentz, I chose CONFIG_MNATIVE option, so it is also necessary to mark CONFIG_X86_MARCH_NATIVE ?

damentz commented on 2016-06-29 01:22 (UTC)

This patch is already merged per this commit: Customize with menuconfig or xconfig and you'll see new options in the processor architecture section.

JunioCalu commented on 2016-06-29 01:20 (UTC) (edited on 2016-06-29 01:23 (UTC) by JunioCalu)

sir_lucjan, could add support to "kernel_gcc_patch" created by graysky2 as an option?

sir_lucjan commented on 2016-06-19 20:32 (UTC)

Bump to 4.6.2-2

damentz commented on 2016-06-19 20:19 (UTC)

Released a new patch that removes CKO from the kernel source, this should fix the build error that eeleater and metak confirmed. Also, I've been pushing my changes to github here: Each release is tagged and downloadable as a zip or tar.gz file. For instance, the latest release as of this post can be downloaded from this link: Inside are the patches and configuration. Just an FYI in case handling github releases through the AUR is easier than tracking the files from

sir_lucjan commented on 2016-06-19 18:54 (UTC)

Vanilla linux, linux-bfq, linux-ck - build withouts errors. I've reported bug.

metak commented on 2016-06-19 18:40 (UTC)

Same error as @eeleater.

eeleater commented on 2016-06-19 17:44 (UTC) (edited on 2016-06-19 18:49 (UTC) by eeleater)

drivers/hwtracing/stm/console.c: In function ‘stm_console_link’: drivers/hwtracing/stm/console.c:51:20: error: assignment from incompatible pointer type [-Werror=incompatible-pointer-types] sc->console.write = stm_console_write; ^ cc1: some warnings being treated as errors make[2]: *** [scripts/ drivers/hwtracing/stm/console.o] Error 1 make[2]: *** Waiting for unfinished jobs.... make: *** [Makefile:965: drivers] Error 2 ==> ERROR: A failure occurred in build(). Aborting... ==> ERROR: Makepkg was unable to build linux-lqx. ==> Restart building linux-lqx ? [y/N] ==> ----------------------------------

sir_lucjan commented on 2016-06-19 16:16 (UTC)

Bump to 4.6.2 mariojuniorjp Show me your modprobed.db files.

mariojuniorjp commented on 2016-06-11 16:49 (UTC)


sir_lucjan commented on 2016-06-11 10:28 (UTC)

Did you use modprobed-db?

mariojuniorjp commented on 2016-06-10 23:01 (UTC) (edited on 2016-06-10 23:02 (UTC) by mariojuniorjp)


sir_lucjan commented on 2016-06-10 22:04 (UTC)

lsmod | grep ath9k

mariojuniorjp commented on 2016-06-10 21:54 (UTC)

The wireless stopped working after compilation with localmodconfig. Both the driver and the chipset of the module are loaded. The wpa_supplicant.service remains inactive but loaded. I enabled and I disabled the networmanager, but nothing decided. 06:00.0 Network controller: Qualcomm Atheros AR93xx Wireless Network Adapter (rev 01) Subsystem: Qualcomm Atheros Device 3112 Kernel driver in use: ath9k Kernel modules: ath9k

sir_lucjan commented on 2016-06-03 10:44 (UTC)

Bump to 4.5.6

kasei commented on 2016-05-24 04:13 (UTC)

I put this package to archlinuxcn. So I just want make sure there is no download error. Thanks for help.

frol commented on 2016-05-22 07:15 (UTC)

@kasei 4.5.5-1 works fine for me. Make sure you cleaned makepkg cache (/etc/makepkg.conf:SRCDEST)

sir_lucjan commented on 2016-05-22 07:12 (UTC)

Use "updpkgsums". I'll fix this tomorrow.

kasei commented on 2016-05-21 14:00 (UTC)

Validity check failed again. Validating source files with sha512sums... linux-4.5.tar.xz ... Passed linux-4.5.tar.sign ... Skipped 4.5.5-1.patch.gz ... Passed config.i386 ... FAILED config.i386-pae ... FAILED config.amd64 ... FAILED linux-lqx.preset ... Passed

sir_lucjan commented on 2016-05-20 12:05 (UTC)

Bump to 4.5.5

sir_lucjan commented on 2016-05-11 14:50 (UTC)

Bump to 4.5.3-2

sir_lucjan commented on 2016-05-11 08:31 (UTC)


kasei commented on 2016-05-11 04:57 (UTC)

Validity check failed. Validating source files with sha512sums... linux-4.5.tar.xz ... Passed linux-4.5.tar.sign ... Skipped 4.5.3-1.patch.gz ... Passed config.i386 ... FAILED config.i386-pae ... FAILED config.amd64 ... FAILED linux-lqx.preset ... Passed

spangry commented on 2016-05-10 20:20 (UTC)

@damentz Some googling around reveals very little about this driver (only get 4 pages back). It's for a Atmel Wlan card of some kind. Probably safe to remove until it compiles properly. By the looks of it, the non-compilation issue has been encountered before:

damentz commented on 2016-05-10 15:01 (UTC)

@spangry, I noticed this too recently, I think it's GCC 6.x that's causing the build failure. On Debian, this module builds successfully with the official Liquorix packages using pbuilder. If the hardware this module supports is rare and no one actually has it, I'm ok disabling it in the official configuration. It _is_ a staging device driver, after all.

spangry commented on 2016-05-10 14:41 (UTC)

Hey is anyone else getting errors when compiling? Build was failing for me on one of the staging drivers (wilc_spi). After removing this via menuconfig it compiled fine.

sir_lucjan commented on 2016-05-08 07:13 (UTC)

Bump to 4.5.3

sir_lucjan commented on 2016-05-03 09:21 (UTC)

@damentz: Done. :)

damentz commented on 2016-05-03 03:04 (UTC)

@sir_lucjan, can you revert the change? Per the wiki, we shouldn't be override user configured MAKEFLAGS.

disastro commented on 2016-05-03 02:25 (UTC)

makepkg.conf :/ And like the wiki says, the only reason to override that should be -j1 for race conditions. num_cores is just wrong and overriding user preference and ${MAKEFLAGS} is unnecessary.

damentz commented on 2016-05-03 01:53 (UTC)

I'm not too familiar with makepkg, I just started using Arch seriously on a laptop of mine and wanted the build to use all the cores. If you know a better way of doing it, feel free to submit a patch to @sir_lucjan. The ${MAKEFLAGS} would be ideal since we don't need to modify the make command.

disastro commented on 2016-05-03 01:31 (UTC) (edited on 2016-05-03 01:34 (UTC) by disastro)

Why is -j being set? That's what MAKEFLAGS is for... EDIT: And I hadn't even noticed before that ${MAKEFLAGS} was being passed too, even though makepkg already does that. :/

damentz commented on 2016-05-03 00:14 (UTC) (edited on 2016-05-03 00:23 (UTC) by damentz)

@sir_lucjan, I believe the build bug is fixed. It was due to a stable patch that changes quite a bit about how the SCHEDSTATS config option works and requires some serious porting. For now, when selecting BFS, CONFIG_SCHEDSTATS turns off. I'll remove this workaround in v4.6 when Con ports it officially. @eeleater, awesome! We merged a change to the PKGBUILD that tries to get the total number of cores on your system and use that as the job parameter during the various make commands. The more cores you have the faster you will build.

eeleater commented on 2016-05-02 21:03 (UTC)

That compiled real fast now :D

sir_lucjan commented on 2016-05-02 09:19 (UTC)

@damentz Could you check ;)?

damentz commented on 2016-04-29 03:53 (UTC) (edited on 2016-04-29 03:54 (UTC) by damentz)

Uploaded a new patch that reverts the 'wastedcores' patch set, marked this package as out-of-date. Basically, this patch set improved efficiency on large systems, but didn't provide any measurable benefit to regular and small systems, and had the side effect of breaking BFS builds.

BrianAllred commented on 2016-04-29 02:49 (UTC)

So, any word on a fix out there for those of us that build with BFS?

ronoverdrive commented on 2016-04-27 23:37 (UTC)

sir_lucjan: yeah it compiles with BFS disabled.

sir_lucjan commented on 2016-04-27 10:45 (UTC)

@ronoverdrive: Could you try without BFS?

ronoverdrive commented on 2016-04-27 03:36 (UTC)

sir_lucjan: Yes I enable BFS.

kbrownlees commented on 2016-04-26 20:25 (UTC)

The section mismatch issue seems to be related to BFS on my laptop, hit the issue when building with BFS but worked fine without.

frol commented on 2016-04-26 17:01 (UTC)

+1 everything is fine on my custom old-config. (no BFS, no KSM)

shivik commented on 2016-04-26 16:58 (UTC)

No problems here also. BFS enabled.

sir_lucjan commented on 2016-04-26 14:39 (UTC) (edited on 2016-04-26 14:40 (UTC) by sir_lucjan)

Sorry, I can't reproduce this error - I build linux-lqx succesfull. Did you change config file / PKGBUILD, for example with BFS?

ronoverdrive commented on 2016-04-26 14:36 (UTC)

Still getting the same error as last time.

sir_lucjan commented on 2016-04-26 10:53 (UTC)

Bump to 4.5.2-3

sir_lucjan commented on 2016-04-26 09:48 (UTC)

DEPMOD 4.5.2-3-lqx ==> Tidying install... -> Removing libtool files... -> Purging unwanted files... -> Removing static library files... -> Compressing man and info pages... ==> Checking for packaging issue... ==> Creating package "linux-lqx"... -> Generating .PKGINFO file... -> Generating .BUILDINFO file... -> Adding install file... -> Generating .MTREE file... -> Compressing package... ==> Starting package_linux-lqx-headers()... ==> Tidying install... -> Removing libtool files... -> Purging unwanted files... -> Removing static library files... -> Compressing man and info pages... ==> Checking for packaging issue... ==> Creating package "linux-lqx-headers"... -> Generating .PKGINFO file... -> Generating .BUILDINFO file... -> Adding install file... -> Generating .MTREE file... -> Compressing package... ==> Starting package_linux-lqx-docs()... ==> Tidying install... -> Removing libtool files... -> Purging unwanted files... -> Removing static library files... -> Compressing man and info pages... ==> Checking for packaging issue... ==> Creating package "linux-lqx-docs"... -> Generating .PKGINFO file... -> Generating .BUILDINFO file... -> Adding install file... -> Generating .MTREE file... -> Compressing package... ==> Leaving fakeroot environment. ==> Finished making: linux-lqx 4.5.2-3 (Tue Apr 26 11:47:38 CEST 2016) ==> Cleaning up... I'll bump later.

sir_lucjan commented on 2016-04-26 09:11 (UTC)

4.5.2-3 is out and I'll try. Be patient.

BrianAllred commented on 2016-04-25 23:48 (UTC)

Build breaks at modpost with the following error: LINK vmlinux LD vmlinux.o MODPOST vmlinux.o WARNING: modpost: Found 1 section mismatch(es). To see full details build your kernel with: 'make CONFIG_DEBUG_SECTION_MISMATCH=y' GEN .version CHK include/generated/compile.h UPD include/generated/compile.h CC init/version.o LD init/built-in.o arch/x86/built-in.o: In function `set_cpu_sibling_map': (.text+0x46942): undefined reference to `sched_max_numa_distance' Makefile:938: recipe for target 'vmlinux' failed make: *** [vmlinux] Error 1 If I make with CONFIG_DEBUG_SECTION_MISMATCH=y, then I get the following output: LINK vmlinux LD vmlinux.o MODPOST vmlinux.o WARNING: vmlinux.o(.data+0x6f0a0): Section mismatch in reference from the variable logo_clut224 to the variable .init.rodata:logo_arch_clut224 The variable logo_clut224 references the variable __initconst logo_arch_clut224 If the reference is valid then annotate the variable with __init* or __refdata (see linux/init.h) or name the variable: *_template, *_timer, *_sht, *_ops, *_probe, *_probe_one, *_console GEN .version CHK include/generated/compile.h UPD include/generated/compile.h CC init/version.o LD init/built-in.o arch/x86/built-in.o: In function `set_cpu_sibling_map': (.text+0x46942): undefined reference to `sched_max_numa_distance' Makefile:938: recipe for target 'vmlinux' failed make: *** [vmlinux] Error 1 I have my kernel .config file if needed. I tried building with pkgconfig set to both 'old' and 'x' and then reselecting my boot logo (that's all I change).

ronoverdrive commented on 2016-04-25 23:48 (UTC)

Fails to build: arch/x86/build-in.o: In function 'set_cpu_sibling_map': (.text+0x4a841): undefined reference to 'sched_max_numa_distance' Makefile:938: recipe for target 'vmlinux' failed make: *** [vmlinux] Error 1

sir_lucjan commented on 2016-04-25 11:33 (UTC)

Bump to 4.5.2-2

sir_lucjan commented on 2016-04-23 09:54 (UTC)

Fixed. Thx for an info.

FadeMind commented on 2016-04-23 07:18 (UTC)

Typo is : find "${pkgdir}" -type f -exec chmod 4.5 {} \; VALID is find "${pkgdir}" -type f -exec chmod 444 {} \; not 644 or other.

Anthony25 commented on 2016-04-23 00:15 (UTC)

Hello, In your PKGBUILD, line 393, I think there is a typo with the chmod parameter, and you might mean 644 instead of 4.5.

sir_lucjan commented on 2016-04-20 22:04 (UTC)

Bump to 4.5.2

sir_lucjan commented on 2016-04-15 10:36 (UTC)

Bump to 4.5.1

sir_lucjan commented on 2016-04-13 11:11 (UTC)

Bump to 4.4.7

sir_lucjan commented on 2016-04-12 17:42 (UTC)

Thx. Fixed

disastro commented on 2016-04-12 17:36 (UTC)

You forgot to push a .SRCINFO for 4.4.6-3

sir_lucjan commented on 2016-04-12 16:58 (UTC)

Bump to 4.4.6-3

sir_lucjan commented on 2016-04-11 11:24 (UTC)

Bump to 4.4.6-2

sir_lucjan commented on 2016-03-21 09:54 (UTC)

Bump to 4.4.6

sir_lucjan commented on 2016-03-13 22:17 (UTC)

Bump to 4.4.5-3

sir_lucjan commented on 2016-03-12 20:38 (UTC)

Bump to 4.4.5-2

sir_lucjan commented on 2016-03-08 14:30 (UTC)

I don't know.

frol commented on 2016-03-08 14:18 (UTC)

Does anyone know when site will be up and running again? Liquorix Home page is completely unreachable and `/sources/4.4.4-3.patch.gz` path returns 'Service unavailable'.

sir_lucjan commented on 2016-03-08 12:19 (UTC) (edited on 2016-03-08 12:20 (UTC) by sir_lucjan)

Bump to 4.4.4-3

sir_lucjan commented on 2016-03-07 14:53 (UTC)

Bump to 4.4.4-2

sir_lucjan commented on 2016-03-02 11:04 (UTC)

Bump to 4.4.3-3

FadeMind commented on 2016-02-27 13:56 (UTC) (edited on 2016-02-27 14:02 (UTC) by FadeMind)


sir_lucjan commented on 2016-02-27 10:22 (UTC) (edited on 2016-02-27 10:31 (UTC) by sir_lucjan)

@FadeMind Do you mean nvidia, virtualbox and other?

FadeMind commented on 2016-02-27 10:16 (UTC)

@sir_lucjan FYI: tl;dr Standalone packages for external modules are not needed any more. Users should using dkms for build it - ALPM hooks manage this stuff.

sir_lucjan commented on 2016-02-27 10:03 (UTC)

Bump to 4.4.3-2

sir_lucjan commented on 2016-02-21 06:22 (UTC)

Super :D

eeleater commented on 2016-02-21 00:57 (UTC)

yeah, was a temporary glitch, went through compilation without problems just now

eeleater commented on 2016-02-21 00:28 (UTC)

I use yaourt, should I try it with makepkg directly?

sir_lucjan commented on 2016-02-21 00:08 (UTC)

Do you use yaourt or makepkg?

sir_lucjan commented on 2016-02-21 00:02 (UTC)

I'll check it soon (I mean monday). Be patient.

eeleater commented on 2016-02-20 23:59 (UTC) (edited on 2016-02-20 23:59 (UTC) by eeleater)

Nope nothing of the sort, just straight up tried upgrading linux-lqx. I am now trying to have it compile again, might have been only a temporary glitch.

eeleater commented on 2016-02-20 23:35 (UTC) (edited on 2016-02-20 23:35 (UTC) by eeleater)

Do you know what to do here? I get this when compiling the newest linux-lqx: objdump: 'sound/pci/oxygen/.tmp_oxygen_io.o': No such file mv: cannot stat 'sound/pci/oxygen/.tmp_oxygen_io.o': No such file or directory scripts/ recipe for target 'sound/pci/oxygen/oxygen_io.o' failed make[3]: *** [sound/pci/oxygen/oxygen_io.o] Error 1 scripts/ recipe for target 'sound/pci/oxygen' failed make[2]: *** [sound/pci/oxygen] Error 2 scripts/ recipe for target 'sound/pci' failed make[1]: *** [sound/pci] Error 2 Makefile:945: recipe for target 'sound' failed make: *** [sound] Error 2

sir_lucjan commented on 2016-02-20 22:02 (UTC)

Bump to 4.4.2

sir_lucjan commented on 2016-02-17 10:50 (UTC)

Bump to 4.4.1-3 (lqx 4.4.1-2)

sir_lucjan commented on 2016-02-05 08:29 (UTC)


ronoverdrive commented on 2016-02-05 00:06 (UTC)

I'm getting the following error: error: failed to commit transaction (conflicting files) linux-lqx-headers: /usr/lib/modules/4.4.1-1-lqx/build/Documentation/kbuild/Kconfig.recursion-issue-01 exists in filesystem linux-lqx-headers: /usr/lib/modules/4.4.1-1-lqx/build/Documentation/kbuild/Kconfig.recursion-issue-02 exists in filesystem linux-lqx-headers: /usr/lib/modules/4.4.1-1-lqx/build/Documentation/kbuild/ exists in filesystem Errors occurred, no packages were upgraded.

sir_lucjan commented on 2016-02-04 17:50 (UTC)

No problem, mate ;)

funkmuscle commented on 2016-02-04 17:45 (UTC)

and there we go, back to normal.. building as we speak.. thank you for the superb job sir_lucjan!

sir_lucjan commented on 2016-02-04 17:28 (UTC)

Bump to 4.4.1

funkmuscle commented on 2016-02-04 16:29 (UTC)

this is by far the best. bfq doesn't do it. this one is close to vanilla as possible with some tweaks.. I'll just wait till 4.4.1 as I never had to compile the earlier versions. If this is how he's doing them, then I'll compile.. thanx

sir_lucjan commented on 2016-02-04 13:24 (UTC)

@funmuscle Try linux-bfq :D

funkmuscle commented on 2016-02-04 12:52 (UTC)

haha, thanx sir_lucjan :-) I haven't compiled a kernel in about 10+ years. I love the kernel, I hope it won't be like this from here on..

sir_lucjan commented on 2016-02-04 09:54 (UTC)

Ask damentz ;)

funkmuscle commented on 2016-02-04 01:46 (UTC)

why is this build asking so much? * * Restart config... * * * General setup * Tune kernel for interactivity (ZEN_INTERACTIVE) [Y/n/?] y Cross-compiler tool prefix (CROSS_COMPILE) [] Compile also drivers which will not load (COMPILE_TEST) [N/y/?] n Local version - append to kernel release (LOCALVERSION) [-lqx] -lqx Automatically append version information to the version string (LOCALVERSION_AUTO) [N/y/?] n Kernel compression mode 1. Gzip (KERNEL_GZIP) 2. Bzip2 (KERNEL_BZIP2) 3. LZMA (KERNEL_LZMA) > 4. XZ (KERNEL_XZ) 5. LZO (KERNEL_LZO) 6. LZ4 (KERNEL_LZ4) choice[1-6?]: 4 Default hostname (DEFAULT_HOSTNAME) [(none)] (none) Support for paging of anonymous memory (swap) (SWAP) [Y/n/?] y System V IPC (SYSVIPC) [Y/n/?] y POSIX Message Queues (POSIX_MQUEUE) [Y/n/?] y kdbus interprocess communication (KDBUS) [N/m/y/?] n Enable process_vm_readv/writev syscalls (CROSS_MEMORY_ATTACH) [Y/n/?] y open by fhandle syscalls (FHANDLE) [Y/n/?] y uselib syscall (USELIB) [N/y/?] n Auditing support (AUDIT) [Y/?] y

sir_lucjan commented on 2016-02-02 19:20 (UTC)

Bump to 4.4.0

frol commented on 2016-01-27 03:45 (UTC)

BTW, the prepackaged config has `CONFIG_FB_NVIDIA=m`, which seem to cause a freeze on resume from suspend. I haven't tested it thoroughly, but resume feature got fixed when I did `rmmod nvidiafb`. Stock Arch Linux has this module disabled.

frol commented on 2016-01-26 12:28 (UTC)

FYI, I have succeeded in getting a stable Liquorix kernel (4.4.0-2) using `_config="local"` mode (thus, using prepackaged config.amd64 as a base) for my Lenovo Y50-70 laptop (Intel HD 4600 + Nvidia GTX 860M). There were only two issues on my way this time: 1. WiFi couldn't initialize due to the missing CONFIG_PACKET (I was getting "l2_packet_init: socket(PF_PACKET): Address family not supported by protocol" error in journalctl). This was caused by `local` mode since prepackaged config.amd64 has CONFIG_PACKET enabled as a module, but I since I was moving from Arch stock where this option was enabled straight to a kernel instead of a module, modprobed-db couldn't know that this option is needed. 2. The combination of the default encoding for FAT FS (CONFIG_FAT_DEFAULT_IOCHARSET) being utf-8 in Liquorix prepackaged config.amd64, and modprobed-db stripping almost all `CONFIG_NLS_*` (including UTF8) resulted in that I couldn't mount an MMC card (and wouldn't probably able to mount USB flash drive either). I have changed CONFIG_FAT_DEFAULT_IOCHARSET to `iso8859-1` (just to be safe, as this is makes more sense to me, and stock Arch Linux kernel uses it), and also re-enabled `CONFIG_NLS_*` options.

sir_lucjan commented on 2016-01-23 17:19 (UTC)

Bump to 4.3.3-8

sir_lucjan commented on 2016-01-23 17:01 (UTC)

I will bump to 4.4 when linux 4.4 hits to [core]

frol commented on 2016-01-23 16:42 (UTC)

I just want to leave quick notes here: 1. xconfig in Kernel 4.4 has been ported to Qt5, so the build requirement might be changed from `qt4` to `qt5-base`. 2. Current Liquorix patch for 4.4 Kernel (4.4.0-2) does not include BFS scheduler.

sir_lucjan commented on 2016-01-20 09:01 (UTC)

Fix CVE-2016-0728

sir_lucjan commented on 2016-01-13 14:08 (UTC)

Bump to 4.3.3-5

frol commented on 2016-01-01 06:23 (UTC)

@Dea993 You have to import Linus Torvalds' public key:

Dea993 commented on 2015-12-28 13:00 (UTC)

==> Validazione dei sorgenti con sha512sums in corso... linux-4.3.tar.xz ... Verificato linux-4.3.tar.sign ... Ignorato 4.3.3-2.patch.gz ... Verificato config.i386 ... Verificato config.i386-pae ... Verificato config.amd64 ... Verificato linux-lqx.preset ... Verificato ==> Validazione delle firme dei sorgenti con gpg in corso... linux-4.3.tar ... NON RIUSCITO (chiave pubblica sconosciuta 79BE3E4300411886) ==> ERRORE: Una o più firme PGP non possono essere verificate! ==> ERRORE: Makepkg non è riuscito a compilare linux-lqx.

lembang commented on 2015-12-28 06:20 (UTC)

@frol sorry havent check the page for awhile. I tried your idea, but no luck, with liquorix's default config; even worse the new config causing me have a kernel panic, no idea if its because I use LVM on LUKS on that laptop, so for now I give up using liquorix on my t420;am still using it on my g73sw. Once I have enough time to play around with it, I will post it here.

afader commented on 2015-12-27 06:39 (UTC)

For some reason, with this kernel I can't get an IPv4 address from my router. I use the linux-ck kernel with no problem with my router and it gets a 192.168.1.x address from dhcpcd. When I switch to the -lqx kernel I only get an ipv6 address. I can't use only IPv6 because my ISP only provides IPv4 and the tunnel is slow. I am compiling with localmodconfig but I went in and checked a bunch of IP-related modules in the config, but it didn't help.

frol commented on 2015-12-22 12:14 (UTC)

@damentz, I haven't succeeded in moving from liquorix default config to a working one (I just couldn't make my wifi (Intel Corporation Dual Band Wireless-AC 7260) work well, it somehow gets initialized correctly only 20+ minutes after a boot). Thus, I cannot even tell how stable the kernel is as it is not acceptable to me waiting 20 minutes before I can start using my laptop... My config (posted in the previous comment) is originated from Arch upstream kernel config, which I have manually stripped.

damentz commented on 2015-12-21 05:46 (UTC)

@frol, just to clear up any possible FUD, CONFIG_INTEL_IOMMU_DEFAULT_ON was disabled the same day you posted your message, ironically, for a different bug I encountered on a different system of mine. Here was the update from the official debian changelog: linux-liquorix (4.3-7) unstable; urgency=medium * disable CONFIG_INTEL_IOMMU_DEFAULT_ON - DMA remapping is unstable on most intel systems where IOMMU is enabled in the BIOS. Causes crashes and freezes on modern Haswell+ systems. - Known bug that was solved on Ubuntu kernels: * update version to 3.dmz.2 -- Steven Barrett <> Wed, 16 Dec 2015 22:48:02 -0600 What else did you change that is making your system more stable?

frol commented on 2015-12-16 09:53 (UTC)

@lembang You know what, I have tried to compile the kernel with the packaged configs and they seem to be broken for me also. At least dmesg has scary messages about IOMMU-related stuff (`DMAR:[fault reason 06] PTE Read access is not set`), and at the very least WiFi won't work for no reason (modules are loaded and everything seems to be fine, but no WiFi device can be found). Also, 4.3.2-2, in contrast to 4.3.2-1, doesn't work with my old config either. Here is my working kernel config - BTW, It seems that CONFIG_INTEL_IOMMU_DEFAULT_ON causes some issues by itself, so you may start by trying switching it off by passing via grub options `intel_iommu=off` (other available options are `=pt`, `=igfx_off`, and, obviously, `=on`). Stock Arch kernel has CONFIG_INTEL_IOMMU_DEFAULT_ON disabled, but liquorix's prepackaged config has it enabled.

frol commented on 2015-12-15 15:18 (UTC)

@lembang I'm sorry for my delayed responses. I have disabled a number of unnecessary for my Lenovo y50-70 laptop options, and I use BFS process scheduler, but starting with the built-in ('pkg') kernel config didn't work for me (WiFi was misbehaving for no reason at all), so I started with 'old' config running stock Arch kernel and later I was slimming it gradually.

lembang commented on 2015-12-10 07:56 (UTC)

@frol that is interesting. I tested this kernel on my thinkpad t420(Intel HD 3000) ,asus g751JM (HD 4600 + GTX 860m), asus x200ma (baytrail) all will have a random freeze. It only working fine on my Asus G73SW (GTX 560M). Did you do any kernel parameter?

frol commented on 2015-12-10 07:51 (UTC)

@lembang I have Intel HD 4600 + Nvidia GTX 860M (though, I rarely use Nvidia and it is almost always turned off by bumblebee+bbswitch). They have been working absolutely fine on this kernel since I installed the kernel this September.

lembang commented on 2015-12-08 23:11 (UTC)

Anyone in here actually having a problem with lqx kernel + intel HD graphics card. I found that all the device that is running intel HD GPU will have a random freeze. I have followed all the kernel flag instruction it doenst help me with anything. I found it works perfect on my other devices which run NVIDIA GPU

JunioCalu commented on 2015-12-08 18:53 (UTC)

Thanks for the clarification @sir_lucjan, I will use the patch without fear now :D

sir_lucjan commented on 2015-11-22 17:52 (UTC)

@Junio_Calu: If you want - no ;) You can add this patch to PKGBUILD.

JunioCalu commented on 2015-11-22 03:39 (UTC)

@sir_lucjan, is there any restriction on the use of graysky2 patch (enable_additional_cpu_optimizations_for_gcc_v4.9+_kernel_v3.15+.patch )?

sir_lucjan commented on 2015-11-16 18:14 (UTC) (edited on 2015-11-16 18:15 (UTC) by sir_lucjan)

Try nvidia-lqx or nvidia-dkms from [testing].

sir_lucjan commented on 2015-11-16 18:13 (UTC)

Paste: pacman -Qs lqx

afader commented on 2015-11-16 17:25 (UTC)

/var/lib/dkms/nvidia/355.11/build/nvidia/nv-procfs.c: In function ‘nv_procfs_read_registry’: /var/lib/dkms/nvidia/355.11/build/nvidia/nv-procfs.c:363:12: error: void value not ignored as it ought to be return seq_printf(s, "Binary: \"%s\"\n", registry_keys); ^ /var/lib/dkms/nvidia/355.11/build/nvidia/nv-procfs.c: In function ‘nv_procfs_read_text_file’: /var/lib/dkms/nvidia/355.11/build/nvidia/nv-procfs.c:563:12: error: void value not ignored as it ought to be return seq_puts(s, s->private); ^ /var/lib/dkms/nvidia/355.11/build/nvidia/nv-procfs.c:564:1: warning: control reaches end of non-void function [-Wreturn-type] } ^ /var/lib/dkms/nvidia/355.11/build/nvidia/nv-procfs.c: In function ‘nv_procfs_read_registry’: /var/lib/dkms/nvidia/355.11/build/nvidia/nv-procfs.c:364:1: warning: control reaches end of non-void function [-Wreturn-type] } ^ scripts/ recipe for target '/var/lib/dkms/nvidia/355.11/build/nvidia/nv-procfs.o' failed make[2]: *** [/var/lib/dkms/nvidia/355.11/build/nvidia/nv-procfs.o] Error 1 make[2]: *** Waiting for unfinished jobs.... Makefile:1379: recipe for target '_module_/var/lib/dkms/nvidia/355.11/build' failed make[1]: *** [_module_/var/lib/dkms/nvidia/355.11/build] Error 2 make[1]: Leaving directory '/usr/lib/modules/4.3.0-1-lqx/build' Makefile:81: recipe for target 'modules' failed make: *** [modules] Error 2

afader commented on 2015-11-16 17:24 (UTC)

I'm having trouble building nvidia-dkms on 4.3 -> Running build hook: [nvidia] Building nvidia modules for 4.3.0-1-lqx kernel... Kernel preparation unnecessary for this kernel. Skipping... Building module: cleaning build area... 'make' -j8 NV_EXCLUDE_BUILD_MODULES='__EXCLUDE_MODULES' KERNEL_UNAME=4.3.0-1-lqx modules....(bad exit status: 2) Error! Bad return status for module build on kernel: 4.3.0-1-lqx (x86_64) Consult /var/lib/dkms/nvidia/355.11/build/make.log for more information.

sir_lucjan commented on 2015-11-13 15:52 (UTC)

Bump to 4.3

sir_lucjan commented on 2015-11-10 16:51 (UTC)

Bump to 4.2.6

sir_lucjan commented on 2015-10-29 15:48 (UTC)


sir_lucjan commented on 2015-10-29 10:15 (UTC)

I will fix that later.

frol commented on 2015-10-29 03:06 (UTC)

Weird, 4.2.5-1.patch.gz doesn't match sha512sum. For me, it is 4503310f5ad1725b7093d9fec35c87af4ccd7c3397f3b8edc232066fc4b525f6bf384841e308a8ae969fb650dbfd1b094f63bc049f471b6474316d14395a9a

damentz commented on 2015-10-29 00:15 (UTC)

@vania, the old versions were removed as part of my sync scripts a long time ago when I found out that one of the previous maintainers to the AUR was not properly maintaining this package. I got tired of constantly logging in and marking this package as obsolete, so as a compromise, I just delete non-current patches from the sources folder on Since this isn't an issue anymore I can turn that off and just sync all patches instead. If you want to stay up-to-date, you can subscribe to the atom feed here:

sir_lucjan commented on 2015-10-28 17:03 (UTC)

Link to "old" version (4.1.12):

sir_lucjan commented on 2015-10-28 15:58 (UTC)

Bump to 4.2.5

vania commented on 2015-10-28 14:32 (UTC)

I not speek good english, tx..

sir_lucjan commented on 2015-10-28 13:32 (UTC)

You can ask him if you want :)

vania commented on 2015-10-28 12:42 (UTC)

It will not be the case to ask the author to leave the "old patches" a little longer? The patch 4.2.4 has already been deleted ..

sir_lucjan commented on 2015-10-24 15:35 (UTC)

Bump to 4.2.4

sir_lucjan commented on 2015-10-05 15:47 (UTC)

Link to "old" version (4.1.10):

sir_lucjan commented on 2015-10-04 17:15 (UTC)

Bump to 4.2.3

sir_lucjan commented on 2015-09-30 15:43 (UTC)

Link to "old" version (4.1.9):

sir_lucjan commented on 2015-09-30 15:38 (UTC)

Bump to 4.2.2

sir_lucjan commented on 2015-09-24 05:44 (UTC)

Link to "old" version (4.1.8):

sir_lucjan commented on 2015-09-24 05:42 (UTC)

Hi. It's probably an upstream bug: It look like linux-firmware from [core] doesn't include iwlwifi-7260-14.ucode.

pejas commented on 2015-09-23 19:11 (UTC)

Hi, After an upgrade to 4.2.1-1 my wifi stopped working. dmesg shows: [ 61.640993] iwlwifi 0000:03:00.0: Direct firmware load for iwlwifi-7260-14.ucode failed with error -2 it looks like missing firmware (no 14): $ sudo updatedb $ locate iwlwifi-7260 /usr/lib/firmware/iwlwifi-7260-10.ucode /usr/lib/firmware/iwlwifi-7260-12.ucode /usr/lib/firmware/iwlwifi-7260-13.ucode /usr/lib/firmware/iwlwifi-7260-7.ucode /usr/lib/firmware/iwlwifi-7260-8.ucode /usr/lib/firmware/iwlwifi-7260-9.ucode no packages to upgrade after: yaourt -Syua I don't know if it's arch/aur problem or linux/lqx in general. I think Intel Wireless 7260 is quiet popular... please let me know if I should fill it upstream or can help etc. Best Regards pp

sir_lucjan commented on 2015-09-22 15:40 (UTC)

Version 4.1.8:

sir_lucjan commented on 2015-09-22 15:23 (UTC)

Bump to 4.2.1

sir_lucjan commented on 2015-09-22 05:30 (UTC)

4.2.1 is out. I'll bump later.

petejones commented on 2015-09-22 03:39 (UTC)

I'm getting a 404 error when I try to build package.

ilikenwf commented on 2015-09-21 14:02 (UTC)

I know what LQX is, I've been following zen for probably 8? years, when it was known as the kamikaze patchset over on gentoo forums. Thanks for the clarification :) It makes me happy to see so much mainstream use of this kernel.

sir_lucjan commented on 2015-09-21 13:50 (UTC)

Steven Barrett (damentz) wrote: "To put it plainly, Liquorix is a distribution of Zen. Zen is just the primary source code that contains many changes under the zen tune branch, plus many out of tree patches that have not made it to mainline. Liquorix takes the latest zen patches, adds a configuration on top, and packages that for Debian Unstable / Testing."

sir_lucjan commented on 2015-09-21 11:05 (UTC)

1. I know --> 2010 :) 2. Diiff between -zen and -lqx config (for example 64 bit). 3. linux-lqx includes 'config 32-pae'.

ilikenwf commented on 2015-09-21 07:25 (UTC)

The real question, is what are the differences between this and the zen-kernel in the [extra] repo? Zen and lqx use the same git repo, and I'm assuming that zen-kernel in [extra] is the same, since the maintainer is one of the Zen/lqx devs (@heftig).

sir_lucjan commented on 2015-09-20 19:43 (UTC)

Bump to 4.2.0

sir_lucjan commented on 2015-09-14 18:51 (UTC)

Bump to 4.1.7

sir_lucjan commented on 2015-09-13 19:29 (UTC)

Bump to 4.1.6-2

sir_lucjan commented on 2015-09-11 06:02 (UTC)

[ 0.317183] io scheduler bfq registered (default) Check: # cat /sys/block/sda/queue/scheduler

DungSherlock commented on 2015-09-11 01:53 (UTC)

[ 0.316623] io scheduler noop registered [ 0.316768] io scheduler deadline registered [ 0.316984] io scheduler cfq registered [ 0.317183] io scheduler bfq registered (default) [ 0.327078] BFQ I/O-scheduler: v7r8 [ 0.652154] CFS CPU scheduler. BFQ not working? How to fix this?!

sir_lucjan commented on 2015-09-07 05:29 (UTC)

@MrEvilBreakfast Why out-of-date?

sir_lucjan commented on 2015-08-11 17:54 (UTC)

Bump to 4.1.5

MrEvilBreakfast commented on 2015-08-11 08:40 (UTC)

I spotted 4.1.5. I grabbed the patch myself, modified the 4.1.4 pkgbuild and got 4.1.5 running nicely.

sir_lucjan commented on 2015-08-04 17:03 (UTC)

Bump to 4.1.4

sir_lucjan commented on 2015-07-22 18:09 (UTC)

Bump to 4.1.3

lembang commented on 2015-07-22 03:26 (UTC)

4.1.3 is on the upstream

sir_lucjan commented on 2015-07-17 20:45 (UTC)

Version 4.0.8:

damentz commented on 2015-07-12 21:22 (UTC)

Woops, accidentally flagged as out of date, you can ignore that.

sir_lucjan commented on 2015-07-12 19:18 (UTC)

Bump to 4.1.2

keren_sky commented on 2015-07-12 12:16 (UTC)

Anyone getting a 404 on the 4.1.1-2.patch.gz link since y'day?

sir_lucjan commented on 2015-07-06 14:52 (UTC)

@afader: No. @ Freso: Fixed

Freso commented on 2015-07-06 03:43 (UTC)

4.1.1-1.patch.gz, config.i386-pae, and config.amd64 doesn't pass checksum tests.

afader commented on 2015-07-06 00:49 (UTC)

Is there still aufs in this package?

sir_lucjan commented on 2015-07-05 12:02 (UTC)

Bump to 4.1.1

sir_lucjan commented on 2015-07-03 18:27 (UTC)

@Det linux-lqx is up-to-date

sir_lucjan commented on 2015-06-23 17:38 (UTC)

Bump to 4.0.6-1

MrEvilBreakfast commented on 2015-05-24 19:00 (UTC)

It's fine now.

MrEvilBreakfast commented on 2015-05-24 17:21 (UTC)

Oh how I wish it were that simple! I'm sure your PKGBUILD is fine. If you're at all curious as to what's going on my end,

sir_lucjan commented on 2015-05-24 10:47 (UTC)

You should use a makepkg without root privileges.

MrEvilBreakfast commented on 2015-05-24 08:08 (UTC)

Well it may not be the PKGBUILD's fault. No AUR kernel will compile for me. Disregard me.

MrEvilBreakfast commented on 2015-05-24 05:08 (UTC)

I'm getting patching errors... ... patching file tools/perf/util/cloexec.h patching file tools/perf/util/symbol-elf.c patching file tools/power/x86/turbostat/Makefile patching file virt/kvm/arm/vgic.c patching file virt/kvm/kvm_main.c ==> ERROR: A failure occurred in prepare(). Aborting... Naturally, running makepkg --noprepare results in a healthy stream of errors.

sir_lucjan commented on 2015-05-19 08:33 (UTC)

Did you try for example xconfig?

thoffmeyer commented on 2015-05-19 02:17 (UTC)

I don't know if this is a menuconfig/nconfig glitch or to do with nVidia drivers, but I cannot for the life of me get menuconfig/nconfig working in fb or terminal without just massivly glitching out.. Not sure if anyone else has this problem just massive flickers..

sir_lucjan commented on 2015-05-18 08:18 (UTC)

Bump to 4.0.4

sir_lucjan commented on 2015-05-12 15:43 (UTC)

Bump to 4.0.2-2

sir_lucjan commented on 2015-05-08 13:08 (UTC)

Bump to 4.0.2-1

sir_lucjan commented on 2015-05-06 14:44 (UTC)

BFS for 4.0 sometimes doesn't work properly. Please don't use "BFS=yes".

Freso commented on 2015-05-06 09:00 (UTC)

I get if I set $_use_BFS to "yes" but leave all other options as default in the PKGBUILD ("no", "pkg", "", "yes" (BFS), "no", "no"). It compiled fine with $_use_BFS set to "no".

sir_lucjan commented on 2015-05-05 15:35 (UTC)

Bump to 4.0.1-3 New version of liquorix patchset

sir_lucjan commented on 2015-04-30 09:47 (UTC)

Bump to 4.0.1-1

sir_lucjan commented on 2015-04-29 15:43 (UTC)

Probably yes.

hpstg commented on 2015-04-29 15:35 (UTC)

Is CFS the default scheduler again? └──> ~ $ >> sudo dmesg | grep scheduler [ 0.312382] io scheduler noop registered [ 0.312383] io scheduler deadline registered [ 0.312403] io scheduler cfq registered [ 0.312420] io scheduler bfq registered (default) [ 0.312421] BFQ I/O-scheduler version: v7r7 [ 0.326239] CFS CPU scheduler.

sir_lucjan commented on 2015-04-27 12:34 (UTC)

Bump to 4.0.0-1

sir_lucjan commented on 2015-04-21 08:47 (UTC)

Bump to 3.19.5-1

sir_lucjan commented on 2015-04-14 16:12 (UTC)

Bump to 3.19.4-1

sir_lucjan commented on 2015-04-11 16:24 (UTC)


topgun commented on 2015-04-11 16:19 (UTC)

I imported the key, but I keep getting these errors ==> Validating source files with sha512sums... linux-3.19.tar.xz ... Passed linux-3.19.tar.sign ... Skipped 3.19.3-1.patch.gz ... FAILED config.i386 ... Passed config.i386-pae ... Passed config.amd64 ... Passed linux-lqx.preset ... Passed 0001-fix-btrfs-mount-deadlock.patch ... Passed 0001-fixup-drm.patch ... Passed ==> ERROR: One or more files did not pass the validity check! ==> ERROR: Makepkg was unable to build linux-lqx.

sir_lucjan commented on 2015-04-02 08:32 (UTC)

Bump to 3.19.3-1

sir_lucjan commented on 2015-03-28 17:44 (UTC)

gpg --keyserver --recv-keys 79BE3E4300411886

AACO commented on 2015-03-28 17:40 (UTC)

I just tried to install this package and got an error saying that one or more pgp signatures couldnt be verified. ==> Building and installing package ==> Making package: linux-lqx 3.19.2-1 (Sat Mar 28 17:36:49 WET 2015) ==> Checking runtime dependencies... ==> Checking buildtime dependencies... ==> Retrieving sources... -> Downloading linux-3.19.tar.xz... ...... -> Downloading linux-3.19.tar.sign... ...... -> Downloading 3.19.2-1.patch.gz... ...... -> Downloading config.i386... ...... -> Downloading config.i386-pae... ...... -> Downloading config.amd64... ...... -> Found linux-lqx.preset ==> Validating source files with sha512sums... linux-3.19.tar.xz ... Passed linux-3.19.tar.sign ... Skipped 3.19.2-1.patch.gz ... Passed config.i386 ... Passed config.i386-pae ... Passed config.amd64 ... Passed linux-lqx.preset ... Passed ==> Verifying source file signatures with gpg... linux-3.19.tar ... FAILED (unknown public key 79BE3E4300411886) ==> ERROR: One or more PGP signatures could not be verified! ==> ERROR: Makepkg was unable to build linux-lqx. ==> Restart building linux-lqx ? [y/N] ==> ---------------------------------- ==>

sir_lucjan commented on 2015-03-25 07:53 (UTC)

Try now.

Freso commented on 2015-03-25 07:24 (UTC)

I keep getting FAILED sources for the config.* and 3.19.2-1.patch.gz files when makepkg --verifysources'ing. Am I continuously pulling down bad files, or did they get ninja-modified upstream?

sir_lucjan commented on 2015-03-23 07:54 (UTC)

Bump to 3.19.2-1

cokomoko commented on 2015-03-17 16:04 (UTC)

@damentz and @sir_lucjan Turkish my system.So there is an error in the compilation. modversions.h file compilation looks as follows: src/linux-3.18/include/config/modversIon.h so src/linux-3.18/include/config/modversion.h file could not be found: drivers/staging/unisys/channels/channel.c:18:32: ölümcül hata: config/modversions.h: Böyle bir dosya ya da dizin yok #include <config/modversions.h> ^ derleme sonlandırıldı. I fix compilation was successfully terminated as modversion.h hands: DEPMOD 3.18.9-2-lqx ==> Kurulum düzenleniyor... -> İstenmeyen dosyalar kaldırılıyor... -> libtool dosyaları kaldırılıyor... -> static library dosyaları kaldırılıyor... -> Kılavuz ve bilgi sayfaları sıkıştırılıyor... ==> "linux-lqx" paketi oluşturuluyor... -> .PKGINFO dosyası üretiliyor... -> install dosyası ekleniyor... -> .MTREE dosyası oluşturuluyor... -> Paket sıkıştırılıyor... ==> package_linux-lqx-headers() işlemine başlanıyor... ==> Kurulum düzenleniyor... -> İstenmeyen dosyalar kaldırılıyor... -> libtool dosyaları kaldırılıyor... -> static library dosyaları kaldırılıyor... -> Kılavuz ve bilgi sayfaları sıkıştırılıyor... ==> "linux-lqx-headers" paketi oluşturuluyor... -> .PKGINFO dosyası üretiliyor... -> install dosyası ekleniyor... -> .MTREE dosyası oluşturuluyor... -> Paket sıkıştırılıyor... ==> package_linux-lqx-docs() işlemine başlanıyor... ==> Kurulum düzenleniyor... -> İstenmeyen dosyalar kaldırılıyor... -> libtool dosyaları kaldırılıyor... -> static library dosyaları kaldırılıyor... -> Kılavuz ve bilgi sayfaları sıkıştırılıyor... ==> "linux-lqx-docs" paketi oluşturuluyor... -> .PKGINFO dosyası üretiliyor... -> install dosyası ekleniyor... -> .MTREE dosyası oluşturuluyor... -> Paket sıkıştırılıyor... ==> fakeroot ortamından çıkılıyor. ==> linux-lqx 3.18.9-2 (Sal Mar 17 19:44:20 EET 2015) paketinin derlenmesi tamamlandı I think this is a concern for the compilation of Turkish users. Thank you all.

pavbaranov commented on 2015-03-17 08:09 (UTC)

It builds without any problems for me.

damentz commented on 2015-03-16 18:36 (UTC)

sir_lucjan, ok thanks. cokomoko, what happens when you try building with LANG=C? Your output is in Turkish and there's a chance that your language settings could be interfering with the build.

sir_lucjan commented on 2015-03-16 17:59 (UTC)

@damentz Yes Package has built successfully (~tenfold).

damentz commented on 2015-03-16 17:51 (UTC)

sir_lucjan, can you reproduce from a vanilla Arch Linux installation? I wonder if there's a dependency missing from this build that's only obvious on brand new installations with minimal package sets installed.

cokomoko commented on 2015-03-16 17:41 (UTC)

Do not have users who experience any problems? Am I the only who have problems?

napcok commented on 2015-03-16 17:23 (UTC)

Builds fine here

sir_lucjan commented on 2015-03-16 16:04 (UTC)

Once again (and probably last time): DEPMOD 3.18.9-2-lqx ==> Tidying install... -> Purging unwanted files... -> Removing libtool files... -> Removing static library files... -> Compressing man and info pages... ==> Creating package "linux-lqx"... -> Generating .PKGINFO file... -> Adding install file... -> Generating .MTREE file... -> Compressing package... ==> Starting package_linux-lqx-headers()... ==> Tidying install... -> Purging unwanted files... -> Removing libtool files... -> Removing static library files... -> Compressing man and info pages... ==> Creating package "linux-lqx-headers"... -> Generating .PKGINFO file... -> Adding install file... -> Generating .MTREE file... -> Compressing package... ==> Starting package_linux-lqx-docs()... ==> Tidying install... -> Purging unwanted files... -> Removing libtool files... -> Removing static library files... -> Compressing man and info pages... ==> Creating package "linux-lqx-docs"... -> Generating .PKGINFO file... -> Adding install file... -> Generating .MTREE file... -> Compressing package... ==> Leaving fakeroot environment. ==> Finished making: linux-lqx 3.18.9-2 (Mon Mar 16 17:04:01 CET 2015) ==> Cleaning up... Regards.

commented on 2015-03-16 15:56 (UTC)

Package has built successfully ============ DEPMOD 3.18.9-2-lqx ==> Sprzątam instalację... -> Usuwam niechciane pliki... -> Usuwam libtool plików... -> Usuwanie statycznych plików bibliotek -> Kompresuję strony man oraz info... ==> Tworzenie pakietu "linux-lqx"... -> Generuję plik .PKGINFO... -> Dodawanie pliku install... -> Generowanie pliku .MTREE... -> Kompresuję pakiet... ==> Zaczynam package_linux-lqx-headers()... ==> Sprzątam instalację... -> Usuwam niechciane pliki... -> Usuwam libtool plików... -> Usuwanie statycznych plików bibliotek -> Kompresuję strony man oraz info... ==> Tworzenie pakietu "linux-lqx-headers"... -> Generuję plik .PKGINFO... -> Dodawanie pliku install... -> Generowanie pliku .MTREE... -> Kompresuję pakiet... ==> Zaczynam package_linux-lqx-docs()... ==> Sprzątam instalację... -> Usuwam niechciane pliki... -> Usuwam libtool plików... -> Usuwanie statycznych plików bibliotek -> Kompresuję strony man oraz info... ==> Tworzenie pakietu "linux-lqx-docs"... -> Generuję plik .PKGINFO... -> Dodawanie pliku install... -> Generowanie pliku .MTREE... -> Kompresuję pakiet... ==> Opuszczam środowisko fakeroot. ==> Skończyłem tworzenie: linux-lqx 3.18.9-2 (pon, 16 mar 2015, 16:52:53 CET) ===============

commented on 2015-03-16 15:55 (UTC)

Package has build without any crash ... [quote] DEPMOD 3.18.9-2-lqx ==> Sprzątam instalację... -> Usuwam niechciane pliki... -> Usuwam libtool plików... -> Usuwanie statycznych plików bibliotek -> Kompresuję strony man oraz info... ==> Tworzenie pakietu "linux-lqx"... -> Generuję plik .PKGINFO... -> Dodawanie pliku install... -> Generowanie pliku .MTREE... -> Kompresuję pakiet... ==> Zaczynam package_linux-lqx-headers()... ==> Sprzątam instalację... -> Usuwam niechciane pliki... -> Usuwam libtool plików... -> Usuwanie statycznych plików bibliotek -> Kompresuję strony man oraz info... ==> Tworzenie pakietu "linux-lqx-headers"... -> Generuję plik .PKGINFO... -> Dodawanie pliku install... -> Generowanie pliku .MTREE... -> Kompresuję pakiet... ==> Zaczynam package_linux-lqx-docs()... ==> Sprzątam instalację... -> Usuwam niechciane pliki... -> Usuwam libtool plików... -> Usuwanie statycznych plików bibliotek -> Kompresuję strony man oraz info... ==> Tworzenie pakietu "linux-lqx-docs"... -> Generuję plik .PKGINFO... -> Dodawanie pliku install... -> Generowanie pliku .MTREE... -> Kompresuję pakiet... ==> Opuszczam środowisko fakeroot. ==> Skończyłem tworzenie: linux-lqx 3.18.9-2 (pon, 16 mar 2015, 16:52:53 CET) [/quote]

sir_lucjan commented on 2015-03-16 14:41 (UTC)

You're right - that is very interesting. I tried many times - you saw logs...

cokomoko commented on 2015-03-16 14:35 (UTC)

I tried to build both desktop PCs and notebooks linux-lqx 3.18.9-2, but gives the same error in the same place. whereas linux-ck and linux-bfs was built smoothly: aur/linux-ck 3.19.1-1 [installed] (266) aur/linux-ck-headers 3.19.1-1 [installed] (266) aur/nvidia-340xx-ck 340.76-4 [installed] (4) aur/vhba-module-ck 20140928-2 [installed] (17) aur/linux-bfs 3.19.1-1 [installed] (17) aur/linux-bfs-docs 3.19.1-1 [installed] (17) aur/linux-bfs-headers 3.19.1-1 [installed] (17) aur/nvidia-bfs-340xx 340.76-4 [installed] (1) very interesting... linux-lqx not be built, so sorry

cokomoko commented on 2015-03-16 04:47 (UTC)

I'll try 3.18.9-2

sir_lucjan commented on 2015-03-16 04:12 (UTC)

@cokomoko shivik wrote "Builds fine for me" Bulids fine for me too - I tried many times.

cokomoko commented on 2015-03-16 02:31 (UTC)

@sir_lucjan I have successfully installed the linux-bfs package: DEPMOD 3.19.1-1-bfs ==> Kurulum düzenleniyor... -> İstenmeyen dosyalar kaldırılıyor... -> libtool dosyaları kaldırılıyor... -> Statik kütüphane dosyaları kaldırılıyor... -> Kılavuz ve bilgi sayfaları sıkıştırılıyor... ==> "linux-bfs" paketi oluşturuluyor... -> .PKGINFO dosyası üretiliyor... -> install dosyası ekleniyor... -> .MTREE dosyası oluşturuluyor... -> Paket sıkıştırılıyor... ==> package_linux-bfs-headers() işlemine başlanıyor... ==> Kurulum düzenleniyor... -> İstenmeyen dosyalar kaldırılıyor... -> libtool dosyaları kaldırılıyor... -> Statik kütüphane dosyaları kaldırılıyor... -> Kılavuz ve bilgi sayfaları sıkıştırılıyor... ==> "linux-bfs-headers" paketi oluşturuluyor... -> .PKGINFO dosyası üretiliyor... -> .MTREE dosyası oluşturuluyor... -> Paket sıkıştırılıyor... ==> package_linux-bfs-docs() işlemine başlanıyor... ==> Kurulum düzenleniyor... -> İstenmeyen dosyalar kaldırılıyor... -> libtool dosyaları kaldırılıyor... -> Statik kütüphane dosyaları kaldırılıyor... -> Kılavuz ve bilgi sayfaları sıkıştırılıyor... ==> "linux-bfs-docs" paketi oluşturuluyor... -> .PKGINFO dosyası üretiliyor... -> .MTREE dosyası oluşturuluyor... -> Paket sıkıştırılıyor... ==> fakeroot ortamından çıkılıyor. ==> linux-bfs 3.19.1-1 (Pzt Mar 16 04:27:18 UTC 2015) paketinin derlenmesi tamamlandı

cokomoko commented on 2015-03-15 21:02 (UTC)

I tried giving the same error in the same place on both arch and manjaro. 3 times over the arch, one time I tried to build on linux-lqx manjaro but it does not build. I'll try to build linux-bfs results you writer

sir_lucjan commented on 2015-03-15 20:23 (UTC)

Try with another kernel (for example linux-bfs) and let me know. Can someone else confirm this bug? P.S: AUR doesn't support Manjaro :P

cokomoko commented on 2015-03-15 20:17 (UTC)

I installed manjaro, I try to build linux-lqx 3.18.9-1 but still gives the same error: LD drivers/staging/unisys/channels/built-in.o CC [M] drivers/staging/unisys/channels/channel.o drivers/staging/unisys/channels/channel.c:18:32: ölümcül hata: config/modversions.h: Böyle bir dosya ya da dizin yok #include <config/modversions.h> ^ derleme sonlandırıldı. scripts/ recipe for target 'drivers/staging/unisys/channels/channel.o' failed make[4]: *** [drivers/staging/unisys/channels/channel.o] Error 1 scripts/ recipe for target 'drivers/staging/unisys/channels' failed make[3]: *** [drivers/staging/unisys/channels] Error 2 scripts/ recipe for target 'drivers/staging/unisys' failed make[2]: *** [drivers/staging/unisys] Error 2 scripts/ recipe for target 'drivers/staging' failed make[1]: *** [drivers/staging] Error 2 Makefile:937: recipe for target 'drivers' failed make: *** [drivers] Error 2 ==> HATA: build() içinde bir hata oluştu. Çıkılıyor... [cokomoko@linux aur-linux-lqx]$ uname -a Linux linux #1 SMP PREEMPT Mon Dec 1 19:15:43 UTC 2014 x86_64 GNU/Linux where is the problem?

sir_lucjan commented on 2015-03-15 19:32 (UTC)

Bump to 3.18.9-2 Add: From ### DO NOT MERGE OUT THIS IF STATEMENT ### IT AFFECTS USERS WHO STRIP OUT THE DVB STUFF SO THE OFFICIAL ARCH CODE HAS A CP ### LINE THAT CAUSES MAKEPKG TO END IN AN ERROR ### if [ -d include/config/dvb/ ]; then mkdir -p "${pkgdir}/usr/lib/modules/${_kernver}/build/include/config/dvb/" cp include/config/dvb/*.h "${pkgdir}/usr/lib/modules/${_kernver}/build/include/config/dvb/" fi

sir_lucjan commented on 2015-03-15 18:49 (UTC)

@shivik I sent a mail to Freso with (this) solution. I will bump shortly.

shivik commented on 2015-03-15 18:43 (UTC)

Builds fine for me, but check this out: From ### DO NOT MERGE OUT THIS IF STATEMENT ### IT AFFECTS USERS WHO STRIP OUT THE DVB STUFF SO THE OFFICIAL ARCH CODE HAS A CP ### LINE THAT CAUSES MAKEPKG TO END IN AN ERROR ### if [ -d include/config/dvb/ ]; then mkdir -p "${pkgdir}/usr/lib/modules/${_kernver}/build/include/config/dvb/" cp include/config/dvb/*.h "${pkgdir}/usr/lib/modules/${_kernver}/build/include/config/dvb/" fi

sir_lucjan commented on 2015-03-15 18:15 (UTC)

@ Freso Check your mailbox ;)

Freso commented on 2015-03-15 18:04 (UTC)

Disabling DVB related things in .config is a perfectly valid thing to do, so the PKGBUILD shouldn't fail if DVB things aren't available. The 3.18.9-1 PKGBUILD still has this issue.

sir_lucjan commented on 2015-03-15 05:15 (UTC)

Probably your system is misconfigured.

cokomoko commented on 2015-03-14 21:21 (UTC)

package was not built LD drivers/staging/unisys/built-in.o LD drivers/staging/unisys/channels/built-in.o CC [M] drivers/staging/unisys/channels/channel.o drivers/staging/unisys/channels/channel.c:18:32: ölümcül hata: config/modversions.h: Böyle bir dosya ya da dizin yok #include <config/modversions.h> ^ derleme sonlandırıldı. scripts/ recipe for target 'drivers/staging/unisys/channels/channel.o' failed make[4]: *** [drivers/staging/unisys/channels/channel.o] Error 1 scripts/ recipe for target 'drivers/staging/unisys/channels' failed make[3]: *** [drivers/staging/unisys/channels] Error 2 scripts/ recipe for target 'drivers/staging/unisys' failed make[2]: *** [drivers/staging/unisys] Error 2 scripts/ recipe for target 'drivers/staging' failed make[1]: *** [drivers/staging] Error 2 Makefile:937: recipe for target 'drivers' failed make: *** [drivers] Error 2 ==> HATA: build() içinde bir hata oluştu. Çıkılıyor... ==> HATA:makepkg linux-lqx'i inşa edemedi. ==> linux-lqx yeniden inşa edilsin mi ? [e/H] ==> ----------------------------------------- I'll wait for the next linux-lqx package...

cokomoko commented on 2015-03-14 18:42 (UTC)

@ sir_lucjan I did not modify the file @ damentz my machine a little old:Toshiba Satellite A500-19W linux-lqx package installation will find 5-6 hours The results of the installation finishes, I'll write

damentz commented on 2015-03-14 18:01 (UTC)

cokomoko, at this point we suspect you have a tainted build environment or your system is misconfigured. The errors you're showing us is that files are missing after they were built. Perhaps you have some serious filesystem corruption? If you keep having trouble, I would try building this package in a virtual machine or on another linux host, just to be sure it's not your current system that's breaking the build.

sir_lucjan commented on 2015-03-14 17:26 (UTC)

Did you modify a config file?

cokomoko commented on 2015-03-14 17:17 (UTC)

@ sir_lucjan I'll try to re-establish the pack I will write the results here

sir_lucjan commented on 2015-03-14 16:50 (UTC)

@cokomoko DEPMOD 3.18.9-1-lqx ==> Tidying install... -> Purging unwanted files... -> Removing libtool files... -> Removing static library files... -> Compressing man and info pages... ==> Creating package "linux-lqx"... -> Generating .PKGINFO file... -> Adding install file... -> Generating .MTREE file... -> Compressing package... ==> Starting package_linux-lqx-headers()... ==> Tidying install... -> Purging unwanted files... -> Removing libtool files... -> Removing static library files... -> Compressing man and info pages... ==> Creating package "linux-lqx-headers"... -> Generating .PKGINFO file... -> Adding install file... -> Generating .MTREE file... -> Compressing package... ==> Starting package_linux-lqx-docs()... ==> Tidying install... -> Purging unwanted files... -> Removing libtool files... -> Removing static library files... -> Compressing man and info pages... ==> Creating package "linux-lqx-docs"... -> Generating .PKGINFO file... -> Adding install file... -> Generating .MTREE file... -> Compressing package... ==> Leaving fakeroot environment. ==> Finished making: linux-lqx 3.18.9-1 (Sat Mar 14 17:49:48 CET 2015) ==> Cleaning up...

sir_lucjan commented on 2015-03-14 15:54 (UTC)

1.export LANG=C and then give me logs. I don't understand. 2. Did you change a config?

cokomoko commented on 2015-03-14 15:46 (UTC)

LD drivers/staging/unisys/channels/built-in.o CC [M] drivers/staging/unisys/channels/channel.o drivers/staging/unisys/channels/channel.c:18:32: ölümcül hata: config/modversions.h: Böyle bir dosya ya da dizin yok #include <config/modversions.h> ^ derleme sonlandırıldı. scripts/ recipe for target 'drivers/staging/unisys/channels/channel.o' failed make[4]: *** [drivers/staging/unisys/channels/channel.o] Error 1 scripts/ recipe for target 'drivers/staging/unisys/channels' failed make[3]: *** [drivers/staging/unisys/channels] Error 2 scripts/ recipe for target 'drivers/staging/unisys' failed make[2]: *** [drivers/staging/unisys] Error 2 scripts/ recipe for target 'drivers/staging' failed make[1]: *** [drivers/staging] Error 2 Makefile:937: recipe for target 'drivers' failed make: *** [drivers] Error 2 ==> HATA: build() içinde bir hata oluştu. Çıkılıyor... ┌─[cokomoko]──[cokomoko]:~/aur-linux-lqx$ └──>>

sir_lucjan commented on 2015-03-14 05:29 (UTC)

Please DON'T build in tmp directory. You should use "makepkg -sric"

cokomoko commented on 2015-03-13 22:32 (UTC)

Package install failed:

sir_lucjan commented on 2015-03-13 18:17 (UTC)

Bump to v3.18.9-1

sir_lucjan commented on 2015-03-11 17:10 (UTC)

Yes, you're right ;) ==> Tidying install... -> Purging unwanted files... -> Removing libtool files... -> Removing static library files... -> Compressing man and info pages... ==> Creating package "linux-lqx"... -> Generating .PKGINFO file... -> Adding install file... -> Generating .MTREE file... -> Compressing package... ==> Starting package_linux-lqx-headers()... ==> Tidying install... -> Purging unwanted files... -> Removing libtool files... -> Removing static library files... -> Compressing man and info pages... ==> Creating package "linux-lqx-headers"... -> Generating .PKGINFO file... -> Adding install file... -> Generating .MTREE file... -> Compressing package... ==> Starting package_linux-lqx-docs()... ==> Tidying install... -> Purging unwanted files... -> Removing libtool files... -> Removing static library files... -> Compressing man and info pages... ==> Creating package "linux-lqx-docs"... -> Generating .PKGINFO file... -> Adding install file... -> Generating .MTREE file... -> Compressing package... ==> Leaving fakeroot environment. ==> Finished making: linux-lqx 3.18.8-1 (Wed Mar 11 18:10:01 CET 2015) ==> Cleaning up...

Freso commented on 2015-03-11 15:10 (UTC)

While running package_linux-lqx-headers() it errored out due to not being able to stat() 'include/config/dvb/*.h'. I disabled all DVB things in the .config, so I'm guessing this might be related?

sir_lucjan commented on 2015-03-03 18:47 (UTC)

Bump to v3.18.8-1

sir_lucjan commented on 2015-02-28 09:22 (UTC)

Bump to v3.18.7-2

sir_lucjan commented on 2015-02-13 09:01 (UTC)

Bump to v3.18.7-1

sir_lucjan commented on 2015-02-11 10:19 (UTC)

@Det linux-lqx is up to date.

sir_lucjan commented on 2015-02-08 17:51 (UTC)

JohnRobson commented on 2015-02-08 17:48 (UTC)

Thank you. I manually added the PGP Key: $ gpg --keyserver --recv-keys 79BE3E4300411886 Please, remember: source=(${pkgname}-${pkgver}.tar.xz{,.sig}) validpgpkeys=('F37CDAB708E65EA183FD1AF625EF0A436C2A4AFF') # Carlos O'Donell

sir_lucjan commented on 2015-02-07 19:23 (UTC)

Bump to v3.18.6-1

sir_lucjan commented on 2015-01-31 20:53 (UTC)

Bump to v3.18.5-1

fhdrin commented on 2015-01-27 04:07 (UTC)

Despite reading everything over and over and still not getting it; I figured it out: enter this --> gpg --keyserver --recv-keys 79BE3E4300411886

sir_lucjan commented on 2015-01-19 10:43 (UTC)

Bump to v3.18.3-1 Note: This PKGBUILD now conforms to the pacman-4.2 standards of being able to verify pgp signatures of upstream files. See the following for info on how to add the needed keys or how to disable this feature entirely,[1] 1.

sir_lucjan commented on 2015-01-11 15:28 (UTC)

Bump to v3.18.2-1 Note: This PKGBUILD now conforms to the pacman-4.2 standards of being able to verify pgp signatures of upstream files. See the following for info on how to add the needed keys or how to disable this feature entirely,[1] 1.

bigc commented on 2015-01-11 10:31 (UTC)

Patch at updated to 3.18.2-1

sir_lucjan commented on 2015-01-09 13:54 (UTC)

sir_lucjan commented on 2015-01-08 22:39 (UTC)

Note: This PKGBUILD now conforms to the pacman-4.2 standards of being able to verify pgp signatures of upstream files. See the following for info on how to add the needed keys or how to disable this feature entirely,[1] 1.

sir_lucjan commented on 2014-12-26 11:48 (UTC)

Done :) Update to 3.18.1

clfarron4 commented on 2014-12-25 23:01 (UTC)

Oop, looks like 3.18.1-1 has been released, just in time for Christmas :)

sir_lucjan commented on 2014-12-24 23:25 (UTC)

Update to 3.17.7-2

sir_lucjan commented on 2014-12-19 10:52 (UTC)

Update to 3.17.7-1

sir_lucjan commented on 2014-12-08 10:45 (UTC)

Update to 3.17.6-1

sir_lucjan commented on 2014-11-24 10:59 (UTC)

Update to 3.17.4-1

sir_lucjan commented on 2014-11-18 10:02 (UTC)


cargoan commented on 2014-11-18 09:55 (UTC)

wrong KERNEL_VERSION variable in linux-lqx.install (3.17.2-2-lqx) must be 3.17.3-2-lqx

sir_lucjan commented on 2014-11-17 16:12 (UTC)

Bump to v3.17.3-2 Changelog: Added fix FS#42689

sir_lucjan commented on 2014-11-15 19:13 (UTC)

Update to 3.17.3-1

sir_lucjan commented on 2014-11-14 21:05 (UTC)

@FadeMind I really wish the 'flag out-of-date' button required a reason to be entered like the one on the official packages to help maintainers understand why. For example, this package could be dated for ONLY ONE reason: liquorix patch is out of the date, Thank you for attention. Best regards.

sir_lucjan commented on 2014-11-14 20:49 (UTC)

@FadeMind Damn... 3.17.2 is the NEWEST version.... 3.17.3 comming soon. Now, don't pretend you don't understand me.

sir_lucjan commented on 2014-11-14 20:38 (UTC)

@FadeMind No comment. 3.17.2 is the NEWEST version of lqx patch at the moment. Prywatne słówko: zapoznaj się z tematem, nim wyskoczysz przed szereg.

sir_lucjan commented on 2014-11-07 15:35 (UTC)

Bump to 3.17.2-3 Add package linux-lqx-docs pkgbase=linux-lqx pkgname=('linux-lqx' 'linux-lqx-headers' 'linux-lqx-docs')

sir_lucjan commented on 2014-11-07 11:41 (UTC)

Attention! There is no BFS released for 3.17!

sir_lucjan commented on 2014-11-07 11:14 (UTC)

Bump to 3.17.2-2 Little changes in PKGBUILD

sir_lucjan commented on 2014-11-06 18:45 (UTC)

Only one change: Separate packages for kernel and headers. pkgbase=linux-lqx pkgname=('linux-lqx' 'linux-lqx-headers')

sir_lucjan commented on 2014-11-06 18:10 (UTC)

shivik - thank you very much :) It is a great honour for me.

shivik commented on 2014-11-06 18:06 (UTC)

For sometime now it's been buggin' me that I can't do much for *lqx packages. So naturally I decided to pass the torch. Congratulations to the new maintainer - sir_lucjan! Wish you all the best! Thank you guys for sticking around and special thanks to damentz for keeping this alive!

lembang commented on 2014-11-06 06:06 (UTC)

Anyone in here have a problem with libdrm with intel graphics card? I'm staying with kernel 3.16.6 because libdram seems always failed to load for me.

shivik commented on 2014-10-23 11:24 (UTC)

Well, 3.17 is here. Old stuff can be found here:

shivik commented on 2014-10-10 09:39 (UTC)

Thanks, damentz! Included in the PKGBUILD. Its use is controlled with _use_32bit_pae variable.

damentz commented on 2014-10-10 01:11 (UTC)

Configurations have updated for 32-bit. There is also a new 32-bit config file named config.i386-pae for convenience - this comes directly from the debian kernel package.

shivik commented on 2014-09-22 16:26 (UTC)

Open PKGBUILD file with a text editor and you'll see custom="no" option.On the right of that you'll see all available options. Just replace "no" with "m" or "x", or whatever you like.

RayManzarek1984 commented on 2014-09-22 10:00 (UTC)

Is it possible to enter menuconfig before compilation process to edit a custom configuration for my own kernel?

Anonymo commented on 2014-08-25 03:50 (UTC)

silvik commented on 2014-08-16 19:27 (UTC) is out

clfarron4 commented on 2014-08-10 18:29 (UTC)

shivik commented on 2014-08-04 05:11 (UTC)

I naver gave it much thought really. It takes about 10 minutes for me to build the package,so I never bothered. But, you can take a look at this

joko commented on 2014-08-03 11:14 (UTC)

Hello, could you make an example on how to use optimally this PKGBUILD? I would like to minimize downloading and compilation time. So far, I am using the same directory for PKGBUILDs of different version and I have enabled modprobed-db. Although I don't have to re-download the source files of the current minor version (e.g. 3.15), I cannot use the same source directory, since files there are already patched, so I download pkg and src directories. I guess I could save some time by using previously compiled files that haven't changed. Any ideas?

clfarron4 commented on 2014-07-22 14:09 (UTC)

I make it now. That was quick.

shivik commented on 2014-07-22 04:12 (UTC)

(3.15.6-1.patch.gz 2014-07-21 18:23 163K) 3.15.6-1 is out, but it's 163K big so I assume something went wrong. Also no BFS,BFQ(which were present in 3.15.5-1)...

Cape commented on 2014-06-22 02:17 (UTC)

Can't download 3.14.7-1.patch.gz (server return 404). I see 3.14.8-1.patch.gz at . Out of date already?

yachi commented on 2014-06-17 08:42 (UTC)

==> Validating source files with sha512sums... linux-3.14.tar.xz ... Passed 3.14.7-1.patch.gz ... FAILED config.i386 ... Passed config.amd64 ... Passed linux-lqx.preset ... Passed ==> ERROR: One or more files did not pass the validity check!

commented on 2014-05-13 02:32 (UTC)

I can boot linux-lqx with 3.14.3 !!

commented on 2014-05-13 00:36 (UTC)

@clfarron4 what do you mean by [core] ? I will give 3.14.3 a try

alaviss commented on 2014-05-11 01:49 (UTC)

Liquorix 3.14.3 released

commented on 2014-05-10 01:18 (UTC)

Well, something must have changed from 3.13 -> 3.14, my x230 box failed to boot, only with a blank screen. I made bfs and ksm enabled by default, and below are my boot loader script: title Arch Linux (LQX) linux /vmlinuz-linux-lqx initrd /initramfs-linux-lqx.img options root=/dev/sda2 ro quiet ipv6.disable=1 elevator=noop init=/usr/lib/systemd/systemd acpi_osi=Linux acpi_backlight=vendor logo.nologo

shivik commented on 2014-05-09 11:02 (UTC)

Although it's still 3.14.2-1 checksum changed with the upload of the new config files. Fixed.Should be good now.

commented on 2014-05-09 08:51 (UTC)

==> Validating source files with sha512sums... linux-3.14.tar.xz ... Passed 3.14.2-1.patch.gz ... FAILED config.i386 ... Passed config.amd64 ... Passed linux-lqx.preset ... Passed strange, the 3.14.2-1.patch.gz failed, any suggestion ?

damentz commented on 2014-05-08 12:46 (UTC)

Sorry, there was a typo on my sync script. Configurations should be correct now.

clfarron4 commented on 2014-05-08 10:16 (UTC)

@damentz: If I can reproduce the kernel panic with linux-lqx (because the panics seem to be random, and I've only had the one with LQX), I'll certainly post over there. This post should give you an idea of what I got (but I only managed to reproduce it with linux-ck and linux-lts-ck so far):

shivik commented on 2014-05-07 05:39 (UTC)

damentz, yes of course I checked the 3.14 folder the timestamps are different,but the checksums are the same. from 3.13.11-1 - config.amd64 2014-04-23 23:41 154K from 3.13.11-1 - config.i386 2014-04-23 23:41 157K from 3.14.2-1 - config.amd64 2014-05-01 22:33 154K from 3.14.2-1 - config.i386 2014-05-01 22:33 157K Plus the PKGBUILD looks in the right place ... __basekernel=3.14 ...${__basekernel}/config.i386${__basekernel}/config.amd64 ... I get a lot of these (the first 3) I assumed that everyone else will get those. Maybe there is something wrong on my part,but I just deleted everything and ran a makepkg -S.PKGBUILD redownloaded everything and no checksum failed. The config files have been the same from 3.13.7-1 to 3.13.11-1. And if you check the PKGBUILDS from here Checksums are the same as the current from the sources/3.14 folder.

damentz commented on 2014-05-06 22:14 (UTC)

shivik, the config files are in the sources/3.14 folder. clfarron4, can you kick start a post on the liquorix forum? Preferably, I think running 'inxi -bDPxx' will help the most. Make sure you're running the latest version too, h2 fixed some bugs w/ partition / drive detection recently.

clfarron4 commented on 2014-05-06 12:58 (UTC)

Question: Is anyone having random kernel panics caused by BFQ with "elevator=bfq" set?

shivik commented on 2014-05-02 06:27 (UTC)

The first 3.14 lqx patch is out, but the config files are the ones from 3.13, so you'll get lots config related questions during build.Just hit "Enter" to each one. That is until damentz uploads the new configs for 3.14. And there is no BFS released for 3.14(if that's a show stopper for you) 3.13 and older stuff can be found here:

shivik commented on 2014-04-20 20:29 (UTC)

Sorry guys.I was in a hurry for Easter day and forgot to clean up after some experiments. My fault entirely. Redownload and test.Should be OK now.

Roshless commented on 2014-04-20 19:47 (UTC)

Not only you. Tried 2 times since morning, same error.

fhdrin commented on 2014-04-20 19:30 (UTC)

compile errors with latest (3.13.10-1) lqx kernel. just me or everyone?

alaviss commented on 2014-04-20 02:00 (UTC)

Liquorix 3.13.10-1 is out! Please update the package

clfarron4 commented on 2014-04-08 21:20 (UTC)

Thanks! Builds now!

shivik commented on 2014-04-08 16:16 (UTC)

Yes,there is a problem. The guy reports that this works: --- linux-3.13.9-uksm_orig/mm/uksm.c 2014-04-04 14:56:29.738462696 -0300 +++ linux-3.13.9-devel/mm/uksm.c 2014-04-04 14:55:36.165528224 -0300 @@ -1604,7 +1604,11 @@ static struct page *page_trans_compound_anon(struct page *page) { if (PageTransCompound(page)) { - struct page *head = compound_trans_head(page); + /*benatto: old compound_trans_head + * was merged into standard + * compound_head function on linux/mm.h + */ + struct page *head = compound_head(page); /* * head may actually be splitted and freed from under * us but it's ok here.

clfarron4 commented on 2014-04-08 13:47 (UTC)

When building 3.13.9 with KSM, I got this error: Any ideas? It built fine with 3.13.6 (didn't build with KSM for 3.13.7 or 3.13.8).

cl00t commented on 2014-04-06 17:05 (UTC)

Flagged out of date as 3.13.9-1.patch.gz is out

thoffmeyer commented on 2014-03-12 13:10 (UTC)

@clfarron4: Thank you brother. I see it is now updated. I do know that lqx doesn't follow the [core] kernel updates or packages.

clfarron4 commented on 2014-03-11 12:00 (UTC)

@thoffmeyer: The LQX kernel follows its own development cycle which releases kernels at different times to when the individual parts are released. Put another way, when [core] releases a version, LQX doesn't immediately jump to that version.

shivik commented on 2014-03-09 11:47 (UTC)

From 3.13.5-1.patch.gz 2014-02-25 07:49 663K

shivik commented on 2014-03-09 11:22 (UTC)

Show me where is this 3.13.6.

thoffmeyer commented on 2014-03-09 08:28 (UTC)

3.13.6 is out along with a new patch for BFS. I would love to use this a my kernel but until BFS gets forked in and updated.. :/, flagged out of date.

clfarron4 commented on 2014-03-04 09:22 (UTC)

@Det: BFS and what not.

shivik commented on 2014-03-04 09:01 (UTC)

linux-lqx follows So...not out-of-date yet.

Det commented on 2014-03-03 21:56 (UTC)

CK patches..?

zoopp commented on 2014-02-27 12:54 (UTC)

Might someone have an idea on why systemd complains about failing to insert 'autofs4' and 'ipv6' module when the kernel is built with '_config="local"'? I have been running modprobed-db for a few months by now and setup a cron script that stores loaded modules every 5 minutes. Everything works fine if _config is set to pkg.

clfarron4 commented on 2014-02-25 17:36 (UTC)

The 3.13.5 patchset does not contain the BFS CPU scheduler because Con hasn't released it for the 3.13 series yet.

shivik commented on 2014-02-24 06:45 (UTC)

3.13.5-1.patch.gz is 35 bytes. I'll wait for damentz to reupload.

shivik commented on 2014-02-06 22:11 (UTC)

Not out-of-date!

kleku commented on 2014-02-04 16:35 (UTC)

That's really important.

shivik commented on 2014-02-04 16:34 (UTC)

in the respective config file(config.amd64 or config.i386) change CONFIG_LOGO=y to CONFIG_LOGO=n.But then checksum will fail and you'll have to change it accordingly in PKGBUILD. Or build package with menuconfig,nconfig or xconfig enabled in PKGBUILD and then search and disable "Boot logo".

commented on 2014-02-04 12:01 (UTC)

Hi, how to prevent the bootup logo from building into the kernel ? I have add logo.nologo to the kernel command line, however it still shows a splash screen.

shivik commented on 2014-01-30 07:08 (UTC)

Yeap... 3.12.9-2.patch.gz 2014-01-29 00:13 870K 3.12.9-2.patch.gz 2014-01-29 13:55 870K Fixed.

commented on 2014-01-29 20:31 (UTC)

5160298ee23bead800b918dec8306af6a14ccb112d3c34f6438019bdfe0dedad6fb4894d3a7058610c153414a3d30f383ce34494ba5d5f6f95cb6ce3b6372489 3.12.9-2.patch.gz

commented on 2014-01-29 20:26 (UTC)

==> Validating source files with sha512sums 3.12.9-2.patch.gz ... FAILED

shivik commented on 2014-01-28 19:26 (UTC)

linux-lqx.install gets overwritten every time the package is built,so not really a problem. If you build the package you'll get a linux-lqx.install with KERNEL_VERSION 3.12.9-1 (or whatever the current version)

commented on 2014-01-28 10:36 (UTC)

need to FIX - > KERNEL_VERSION=3.12.8-1-lqx in linux-lqx.install

shivik commented on 2014-01-18 07:46 (UTC)

There is an option now in PKGBUILD for those who wish to use BFS. Just set _use_BFS to "yes" and you're good to go.

damentz commented on 2014-01-18 01:44 (UTC)

joko, CFS has less functional bugs than BFS. After a few years of going back and forth due to new random bugs, I decided to stick to CFS and work with it instead. One nice feature of CFS that BFS doesn't have is autogrouping, taking advantage of cgroups to share CPU bandwidth more fairly. This is most obvious when building software with multiple threads - the build tool will be grouped and will be treated as a single process, preventing a DoS on your own system. This is important for desktops where increased latency through interactivity is considered a big problem. Lastly, since BFS has been released, many of the bugs that once infested CFS have been fixed. If you read benchmarks, there's no drastic difference between the two schedulers. That leaves behavior, which I don't believe is really an issue anymore. I can be building a kernel, watching youtube, and playing an intensive AAA game at the same time and I can't tell the difference. The only difference with BFS is that you have access to sched_batch, which you can use to lower the priority of a build program that uses multiple threads. In CFS, you have auto grouping, which solves that problem nicely without any action of the user. So there you go, that's the reality of CFS vs BFS. Behavior is mostly identical and throughput is close enough to not matter.

joko commented on 2014-01-18 00:04 (UTC)

Hello, does this kernel use the BFS scheduler? I have compiled this package quickly via yaourt and it seems it doesn't: $ dmesg | grep scheduler [ 0.636725] io scheduler noop registered [ 0.636725] io scheduler deadline registered [ 0.636743] io scheduler cfq registered [ 0.636757] io scheduler bfq registered (default) [ 0.702179] CFS CPU scheduler. Any reason why it is not used?

KairiTech commented on 2014-01-12 21:27 (UTC)

Appreciate your efforts. I'll keep working at it to find an easier way to keep track of the mobo-specific kernels I create. Thanks again for your efforts.

shivik commented on 2014-01-12 20:09 (UTC)

I did a pacman -Rns linux-lqx prior to installing the rebuilt linux-lqx(with the custom name). Maybe this has something to do.I mean if you haven't done that yourself... This is all I can think of for now. Making that work for your case turns out to be beyond my knowledge. Next step is a completely modified PKGBUILD just for your needs.

KairiTech commented on 2014-01-12 18:00 (UTC)

No errors here either but I end up in emergency mode when I re-boot. I've had it work before without any changes to the kernel version but it doesn't seem to work now.

shivik commented on 2014-01-11 17:48 (UTC)

Yeah, two of my experiments must have mixed up 'cause I made a whole new PKGBUILD just for that and probably installed one on top of the other... I just recompiled and I have what you have,but I have a working system and I got no errors during kernel install - initramfs went fine,etc. I use nvidia driver and it is working. I don't use LVM, so no way to test that. The way this is supposed to work is just let you append a string to the kernel version and everything else should be normal - i.e. no need to mess with your bootloader config. Is your system working with this setup?

KairiTech commented on 2014-01-11 14:01 (UTC)

I don't seem to get extramodules-3.12-lqx-M5A99X_EVO or initramfs-linux-lqx-M5A99X_EVO-fallback.img vmlinuz-linux-lqx-M5A99X_EVO initramfs-linux-lqx-M5A99X_EVO.img ls -l /lib/modules 3.12.6-1-ARCH 3.12.6-1-lqx-M5A99X_EVO extramodules-3.12-ARCH extramodules-3.12-lqx ls /boot EFI initramfs-linux-lqx-fallback.img initramfs-linux-lqx.img vmlinuz-linux-lqx

shivik commented on 2014-01-09 10:09 (UTC)

OK, I added an option _append_kernel_custom_string to the PKGBUILD. Change that to your desire.It works on my system,no errors. Changed, rebuilt the kernel and here's my uname -r now :) [shivik@archenemy ~]$ uname -r 3.12.6-1-lqx-M5A99X_EVO My /lib/modules looks like this now: [shivik@archenemy modules]$ ls 3.12.6-1-ARCH 3.12.6-1-lqx-M5A99X_EVO extramodules-3.12-ARCH extramodules-3.12-lqx-M5A99X_EVO Keep in mind that this changes your /boot kernel and initramfs filenames,so you have to edit your bootloader config file accordingly. [shivik@archenemy boot]$ ls initramfs-linux-fallback.img initramfs-linux.img vmlinuz-linux initramfs-linux-lqx-M5A99X_EVO-fallback.img lost+found vmlinuz-linux-lqx-M5A99X_EVO initramfs-linux-lqx-M5A99X_EVO.img syslinux

KairiTech commented on 2014-01-09 00:42 (UTC)

As suggested I appended -M5A99X_EVO in the PKGBUILD instead of running xconfig but got the error below instead: ==> Finished making: linux-lqx-M5A99X_EVO 3.12.6-1 (Wed Jan 8 19:37:51 EST 2014) ==> Installing package linux-lqx-M5A99X_EVO with pacman -U... loading packages... resolving dependencies... looking for inter-conflicts... Packages (1): linux-lqx-M5A99X_EVO-3.12.6-1 Total Installed Size: 0.00 MiB :: Proceed with installation? [Y/n] Y (1/1) checking keys in keyring [##############################] 100% (1/1) checking package integrity [##############################] 100% (1/1) loading package files [##############################] 100% (1/1) checking for file conflicts [##############################] 100% (1/1) checking available disk space [##############################] 100% (1/1) installing linux-lqx-M5A99X_EVO [##############################] 100% >>> Updating module dependencies. Please wait ... >>> Generating initial ramdisk, using mkinitcpio. Please wait... ==> Building image from preset: /etc/mkinitcpio.d/linux-lqx.preset: 'default' -> -k /boot/vmlinuz-linux-lqx -c /etc/mkinitcpio.conf -g /boot/initramfs-linux-lqx.img ==> ERROR: specified kernel image does not exist: `/boot/vmlinuz-linux-lqx' ==> Building image from preset: /etc/mkinitcpio.d/linux-lqx.preset: 'fallback' -> -k /boot/vmlinuz-linux-lqx -c /etc/mkinitcpio.conf -g /boot/initramfs-linux-lqx-fallback.img -S autodetect ==> ERROR: specified kernel image does not exist: `/boot/vmlinuz-linux-lqx' :: linux-lqx cleaned

shivik commented on 2014-01-08 06:24 (UTC)

What about it? I just tried it out of curiosity. Removed nvidia, nvidia-lqx, installed nvidia-dkms, reboot, works.

ilkyest commented on 2014-01-08 01:15 (UTC)

What about the correct way to make nvidia driver working with this kernel? nvidia-dkms?

shivik commented on 2014-01-07 18:01 (UTC)

Basically you want to change -lqx to -lqx-M5A99X_EVO, right? So instead of changing -lqx in xconfig try and change _kernelname in PKGBUILD to -lqx-M5A99X_EVO and see what happens. I haven't tested or given it much thought, but give it a go. It will either work or it won't :)

KairiTech commented on 2014-01-07 15:20 (UTC)

I use the "local" modprobed_db option to compile only the modules required for my mobo but this generates the error below during the re-creation of the initramfs. Funny thing is I am still able to boot without any issues even though I am using LVM2. depmod: ERROR: could not open directory /lib/modules/3.12.6-1-lqx: No such file or directory depmod: FATAL: could not search modules: No such file or directory Running build hook: [lvm2] module not found: `dm-snapshot' module not found: `dm-mirror' I'm sure it has something to do with the fact that I append '-M5A99X_EVO' to the ‘-lqx’ kernel version when I run xconfigure. I do end up with the modules directory below but mkinitcpio does seem able to find it and I thought that the package file would be renamed too but it isn't. /lib/modules/3.12.6-1-lqx-M5A99X_EVO/ /lib/modules/extramodules-3.12-lqx/ So my question is should I change the kernel version to include my mobo name in some other way?

KairiTech commented on 2014-01-07 15:05 (UTC)

shivik commented on 2013-12-27 12:56 (UTC)

This happens when damentz updates lqx with the same version. 3.12.6-1 updated to 3.12.6-1,but different checksums. 3.10.22-1.patch.gz 2013-12-24 03:21 1.0M 3.12.6-1.patch.gz 2013-12-24 03:21 794K config.i386 2013-12-24 03:21 156K config.amd64 2013-12-24 03:21 152K 3.10.22-1.patch.gz 2013-12-27 01:03 1.0M 3.12.6-1.patch.gz 2013-12-27 01:03 794K config.i386 2013-12-27 01:03 156K config.amd64 2013-12-27 01:03 152K I check a few times a day.

PerfectGentleman commented on 2013-12-27 10:43 (UTC)

==> Validating source files with sha512sums... linux-3.12.tar.xz ... Passed 3.12.6-1.patch.gz ... FAILED config.i386 ... Passed config.amd64 ... Passed linux-lqx.preset ... Passed ==> ERROR: One or more files did not pass the validity check!

damentz commented on 2013-12-20 21:35 (UTC)

kleku, when this package updates, can you try again? I made a change I was waiting to bundle with 3.12.6 (bundling updates together prevents massive number of updates and work for package maintainers). You'll have to wait though for shivik to update this package. Basically, I removed the Intel P-State cpufreq driver (final confirmation was Valve citing it as a problem for smooth performance on Steam), set ondemand as default governor, and set the ondemand sampling down factor to 10 (increases throughput during high load). My suspicion was that the Intel P-State cpufreq driver wasn't properly saving power when it could. You should be better off with ondemand.

kleku commented on 2013-12-20 11:08 (UTC)

Powertop on my notebook with standard kernel: The battery reports a discharge rate of 10.2 W and with -lqx kernel: The battery reports a discharge rate of 20.1 W :/ I'm using TLP.

damentz commented on 2013-12-02 07:04 (UTC)

Actually, I'm just on vacation from work and everything else for Thanksgiving. I'll update to 3.12 later this week - I can't responsibly update to 3.12 without having time to both use the kernel for an extended time and respond to feedback.

shivik commented on 2013-12-02 06:52 (UTC)

I think the answer lies below in the comments. damentz: "The only difference is that I do a bit of quality control and respond to feedback. It's hard to do that where there isn't a "product", or something to install and test immediately without wasting an hour of time compiling or figuring out what configuration to use." damentz test stuff to some extent prior to release. I haven't had any problems(even minor) with linux-lqx for more than a year. I don't even have rcu_preempt wrong time accounting and I enable BFS every time. So that would explain the delays to me. Not that it matters so much.It's just an enthusiast thing. I don't mind waitng :)

PerfectGentleman commented on 2013-12-02 04:38 (UTC)

I just wonder if LQX and Zen share the same codebase, why Zen is 3.12.2 and LQX is still 3.11.9 ?

damentz commented on 2013-12-02 02:24 (UTC)

Well, this package only has a 1 liter engine, so driving up hills is a hit or miss.

PerfectGentleman commented on 2013-12-02 00:05 (UTC)

why so slow?

ilikenwf commented on 2013-11-22 05:24 (UTC)

Just reminding folks that LQX and Zen share the same codebase, LQX just uses a more optimized config than a default kernel config....

shivik commented on 2013-11-21 06:18 (UTC)

No problem.Glad it worked.

saburouta commented on 2013-11-21 02:39 (UTC)

That was it, of course. Sorry I made you quote yourself. I thought it would be enough to nuke the packer* dirs in /tmp. Tarball worked.

shivik commented on 2013-11-20 06:28 (UTC)

Current version is 3.11.8-1. Your warnigs say 3.11.7-1. You have some sort of version mix up. As always delete everything and "Get the tarball,extract into a new directory and try again. You can copy your linux-3.11.tar.xz to avoid downloading that again."

saburouta commented on 2013-11-19 16:11 (UTC)

[user@host ~]$ sudo depmod -a depmod: WARNING: could not open /lib/modules/3.11.7-1-lqx/modules.order: No such file or directory depmod: WARNING: could not open /lib/modules/3.11.7-1-lqx/modules.builtin: No such file or directory

shivik commented on 2013-11-17 12:53 (UTC)

Fixed. There were two 3.11.8-1 updates: 3.11.8-1.patch.gz 2013-11-17 00:03 791K 3.11.8-1.patch.gz 2013-11-17 02:13 791K

shivik commented on 2013-11-15 06:29 (UTC)

@AnAkkk - why out of date? still says 3.11.7-1

Wintershade commented on 2013-09-22 23:13 (UTC)

Alright, thanks man :) I deleted everything and dl'ed the fresh tarball. It works now.

shivik commented on 2013-09-22 21:04 (UTC)

I updated the package to 3.11. 3.10 stuff can be found here:

shivik commented on 2013-09-22 19:45 (UTC)

I checked and all checksums check out. (Deleted configs,3.10.12-1 patch,etc and did a makepkg -S) Actually config.* files have been the same for a while. So very strange to get checksum error there. Must be something on your part. Get the tarball,extract into a new directory and try again. You can copy your linux-3.10.tar.xz to avoid downloading that again.

Wintershade commented on 2013-09-22 17:33 (UTC)

Hey guys, a quick question. I just pulled all the source files from here (twice, manually and through yaourt), and I'm getting this error... ==> Validating source files with sha512sums... linux-3.10.tar.xz ... Passed 3.10.12-1.patch.gz ... FAILED config.i386 ... FAILED config.amd64 ... FAILED linux-lqx.preset ... Passed ==> ERROR: One or more files did not pass the validity check! so supposedly the 3.10.12-1 patch, as well as the config.* files aren't what they're supposed to be. What gives? Am I missing something, or doing something wrong? TIA

jjdekroon commented on 2013-09-22 08:45 (UTC)

Okay, thanks for clearing that up, and I understand your decision. Also I have a remark on the latest kernel source. It seems aufs doesn't function on my system. I saw it is included in the kernel configuration. Greetings PsychoGame

damentz commented on 2013-09-22 00:22 (UTC)

Will BFS make a comeback? Probably not. The bugs in BFS are minor but noticeable. This biggest bug that's yet to be fixed is time keeping. This was documented heavily on the liquorix forums and lead me to choosing CFS over BFS with heavy configuration modifications. To answer your last question, just choose the zen-tune custom configuration, it will set rr_interval to 3 and change a few more defaults to improve performance.

jjdekroon commented on 2013-09-21 20:57 (UTC)

I have a small question about bfs. Will BFS make a comeback as standard scheduler in the builds in the future? Are the bugs which made the switch from BFS to CFS still existing in the latest kernel releases? Also if I want to compile the Liquorix kernel myself with BFS, what configuration do you recommend for the lowest latency? Greetings PsychoGame

shivik commented on 2013-08-16 16:05 (UTC)

linux-lqx.install gets overwritten every time you build, so not really a problem. If you build end then check linux-lqx.install you'll see that it says 3.6.10(or whatever the current version). Reuploaded anyway.

brenix commented on 2013-08-16 13:28 (UTC)

After update to 3.10.6, the linux-lqx.install file still contains the 3.10.5 version..just a FYI

damentz commented on 2013-08-11 02:04 (UTC)

Sorry, configs have been updated. Thanks for catching that, must have resynced my configs on an incomplete master branch earlier last month.

shivik commented on 2013-08-05 07:14 (UTC)

config* files need to be updated. I get a lot of config related options during build time, which are not present in the respective config.* file.

shivik commented on 2013-07-28 06:55 (UTC)

Thanks, damentz! Nvidia works on 3.10 with Here's the PKGBUILD for 3.9.11 kernel:

damentz commented on 2013-07-28 02:34 (UTC)

3.10 patch is out, but there's no proper support from nvidia and fglrx yet. Unless you guys have an easy work around, I'll recommend sticking to 3.9 until you can install an nvidia driver with sgfxi.

shivik commented on 2013-05-29 05:53 (UTC)

Nice catch. Added inetutils and bc to makedepends.

silvik commented on 2013-05-28 22:08 (UTC)

also, ths might be of interest: something about kmod and inetutils and bc.

silvik commented on 2013-05-28 22:02 (UTC)

finally found it! at first i thought that MAKEFLAGS="-j4" option is the problem, so i commented it out. oddly enough, i receided another error message, not about xfs, but about missing bc command. after installing bc everyhting is fine, j4 or not. so, it looks like bc is needed to build 3.9 kernel...

shivik commented on 2013-05-28 15:07 (UTC)

Hmm...I got nothing. And I'm using xfs on two of my hard drives.Builds and runs fine. Try a clean build and all options to default.

silvik commented on 2013-05-28 12:43 (UTC)

build fails, no idea why: CC [M] fs/xfs/xfs_quotaops.o CC [M] fs/xfs/xfs_rtalloc.o CC [M] fs/xfs/xfs_acl.o CC [M] fs/xfs/xfs_stats.o CC [M] fs/xfs/xfs_sysctl.o CC [M] fs/xfs/xfs_ioctl32.o LD [M] fs/xfs/xfs.o LD fs/built-in.o ==> ERROR: A failure occurred in build(). Aborting... ==> ERROR: Makepkg was unable to build linux-lqx.

shivik commented on 2013-04-23 19:28 (UTC)

Slip of the mouse. Mistake of course :) Fixed.

PerfectGentleman commented on 2013-04-23 14:41 (UTC)

Category: modules - why?

shivik commented on 2013-04-07 15:07 (UTC)

What is "_config" set to in your PKGBUILD?

antihero commented on 2013-04-07 14:50 (UTC)

Something has gone wrong: ==> Running make prepare HOSTCC scripts/basic/fixdep SHIPPED scripts/kconfig/ HOSTCC scripts/kconfig/conf.o SHIPPED scripts/kconfig/zconf.lex.c SHIPPED scripts/kconfig/zconf.hash.c HOSTCC scripts/kconfig/ HOSTLD scripts/kconfig/conf scripts/kconfig/conf --silentoldconfig Kconfig * * Restart config... * * * Firmware Drivers * BIOS Enhanced Disk Drive calls determine boot disk (EDD) [N/m/y/?] n EFI Variable Support via sysfs (EFI_VARS) [M/y/?] m Register efivars backend for pstore (EFI_VARS_PSTORE) [Y/n/?] (NEW) aborted! Console input/output is redirected. Run 'make oldconfig' to update configuration. make[2]: *** [silentoldconfig] Error 1 make[1]: *** [silentoldconfig] Error 2 SYSHDR arch/x86/syscalls/../include/generated/uapi/asm/unistd_32.h SYSHDR arch/x86/syscalls/../include/generated/uapi/asm/unistd_64.h make: *** No rule to make target `include/config/auto.conf', needed by `include/config/kernel.release'. Stop. make: *** Waiting for unfinished jobs.... SYSHDR arch/x86/syscalls/../include/generated/uapi/asm/unistd_x32.h SYSTBL arch/x86/syscalls/../include/generated/asm/syscalls_32.h ==> ERROR: A failure occurred in build(). Aborting... :: linux-lqx cleaned

poincare commented on 2013-03-26 08:45 (UTC)

Many thanks for putting this together and keeping it alive!!

ilikenwf commented on 2013-03-11 21:55 (UTC)

Damentz, thought so. I like zen because I'm a bit of a ricer :P Either way, you and heftig are saints for keeping the project alive after all it's been through. Thanks, and keep up the amazing work!

shivik commented on 2013-03-11 06:58 (UTC)

@ilkyest - How up-to-date is your system? Have you read this Most likely this is the problem as kernel builds and installs fine for everyone else including me.

damentz commented on 2013-03-11 03:21 (UTC)

Hi ilikenwf, both heftig and I work on Zen Sources. You're right, Liquorix is zen + configuration. The only difference is that I do a bit of quality control and respond to feedback. It's hard to do that where there isn't a "product", or something to install and test immediately without wasting an hour of time compiling or figuring out what configuration to use.

ilkyest commented on 2013-03-10 21:37 (UTC)

>>> Updating module dependencies. Please wait ... depmod: ERROR: could not open directory /lib/modules/3.7.10-1-lqx: No such file or directory depmod: FATAL: could not search modules: No such file or directory >>> Generating initial ramdisk, using mkinitcpio. Please wait... /usr/bin/mkinitcpio: line 248: /etc/mkinitcpio.d/linux-lqx.preset: No such file or directory ==> ERROR: Preset not found: `/etc/mkinitcpio.d/linux-lqx.preset' Also... these don't create "linux-headers

ilikenwf commented on 2013-03-10 20:35 (UTC)

Damentz does most of the work on lqx and zen both, I think... I was in IRC and someone told me it was the same, except for the default configs provided...

shivik commented on 2013-03-10 15:59 (UTC)

Yes, I've overlooked that qt change. Works fine with with qt4. Wrong checksum has been fixed when silvernode reported it.

ilkyest commented on 2013-03-10 13:27 (UTC)

sha512 sum mistaken.... here I have 7795d4f3d2aacd1d442017997d211afc9d1ee22a4ae5a069439634ed70e9f7daa33fb879d97e1e5d858934a902c7592ceba228c8162d6211e20e601f897ff72b and.... arch linux has 3 qt. qt3; qt4, and qt5. None of them solves "dependencies issue" I worked around it changing pkg build. Put the sha512sum on pkgbuild (no more validity check error) and changing. Since "x" asks for qt, I put "n" and change the comand line, that call xconfig instead nconfig works fine, even without qt (that I don't know what version the kernel calls)

shivik commented on 2013-03-10 10:49 (UTC)


silvernode commented on 2013-03-10 10:44 (UTC)

Failed validating sha512sums 3.7.10-1.patch.gz ... FAILED It should be replaced with: 7795d4f3d2aacd1d442017997d211afc9d1ee22a4ae5a069439634ed70e9f7daa33fb879d97e1e5d858934a902c7592ceba228c8162d6211e20e601f897ff72b

shivik commented on 2013-03-09 20:38 (UTC)

You can say it is based on zen. It uses modified zen sources,other patches,different kernel configuration as far as I know.

ilikenwf commented on 2013-03-09 19:31 (UTC)

Isn't this bascially the same as linux-zen?

shivik commented on 2013-03-04 16:45 (UTC)

It doesn't matter that much, Anonymo. If you guys like damentz's liquorix and what I'm doing then I'll keep doing it :)

Anonymo commented on 2013-03-04 07:48 (UTC)

@shivik. That Det guy has been a nuisance ever since I was maintaining this kernel too. I think he has his own packages, so I don't understand why he is constantly marking stuff out of date. Wish there was a block user feature on AUR

shivik commented on 2013-03-01 18:48 (UTC)

Thanks, for the heads up.Fixed.

silvik commented on 2013-03-01 17:48 (UTC)

looks like the kernel source link changed:${__basekernel}.tar.xz

shivik commented on 2013-02-26 14:27 (UTC)

I got that and other config questions as well. We'll probably have new config files for 3.7.9. Just hit enter and the build process will continue.

silvik commented on 2013-02-26 11:26 (UTC)

I get a dialog at make prepare: Choose SLAB allocator 1. SLAB (SLAB) (NEW) > 2. SLUB (Unqueued Allocator) (SLUB) choice[1-2?]:

shivik commented on 2013-02-26 07:14 (UTC)

config files are the same as 3.7.8 config files at the time I update to 3.7.9. I'll fix those when I come back from work if damentz updates them in the meanwhile.

damentz commented on 2013-02-21 18:22 (UTC)

Det, I don't know if you're trolling or if you're really that stupid. Package maintenance is a time consuming and inconvenient task. We don't decide when upstream updates their software, but we can decide when to begin working on updating the package. The more random and frequent the updates, the more difficult the maintenance is. Remember that Shivik is updating this package out of his own good will and self interest. Are you asserting that your personal interests are more important than his?

shivik commented on 2013-02-21 17:30 (UTC)

What's that suppose to mean,Det? Sometimes you mark it as out of date.So? Which is of no use to me anyway,because I'll update regardless of the package marked as out-of-date or not. For me it is not a question of seeing if there's a new patchset,but if I can use my computer to do it. Meaning: I update whenever I can. And since when does that become a problem? Anyway,Det,if you want to become the maintainer just let me know. I'm letting go of the bone right away.

Det commented on 2013-02-21 15:50 (UTC)

> I update whenever there's a.. You update whenever I tell you to.

shivik commented on 2013-02-16 06:16 (UTC)

I update whenever there's a new patch here: linux-lqx follows patches released there. Keep in mind the time zone difference, meaning I may be sleeping when damentz releases a new patchset revision, so it will be a few hours sometimes. Also I go to work, you know :)

commented on 2013-02-16 00:44 (UTC)

Can you update this to 3.7.8?

shivik commented on 2013-02-14 06:31 (UTC)

Why is this flagged out-of-date? I don't see 3.7.7 in

Joermungand commented on 2013-02-08 11:20 (UTC)

Thank you, shivik!

shivik commented on 2013-02-08 11:11 (UTC)

Yes,because 3.7.6-2 is out.Updated.

Joermungand commented on 2013-02-08 11:04 (UTC)

Patch 3.7.6-1 is no longer available at the remote location.

shivik commented on 2013-02-03 15:48 (UTC)

BFS is back with 3.7.5-2 patch.

damentz commented on 2013-01-02 06:38 (UTC)

Thank you shivik, I was trying to workaround some bugs that presented themselves in the forum after I pushed out the 3.7 kernel. The only thing left that I cannot find a solution for is some very weird time accounting bug in BFS. For the time being, CFS is the configured scheduler until someone fixes the new bugs in 3.7 for BFS. If anyone doesn't already know, the forum is located here:

shivik commented on 2013-01-02 06:31 (UTC)

damentz, if you are reading this - Thanks for your work!

shivik commented on 2013-01-02 06:31 (UTC)

They keep coming. -5 is out. damentz is working hard on 3.7

msx commented on 2013-01-02 02:27 (UTC)

It was impossible to compile -3 for me but I didn't report back because I was trying to make it run. OTOH -4 compiles and runs smooth!

shivik commented on 2012-12-30 07:23 (UTC)

3.7.1-3 compiles and runs with no problems.

shivik commented on 2012-12-29 07:25 (UTC)

So far I can tell that there is some problem iwth 3.7.1-2 patch. 3.7.1-1 is OK.Lets hope damentz will run into the same problem fix it soon. You can grap the pkgsrc for 3.7.1-1 from here And the 3.7.1-1 patch from here You can use the current config.i386 and config.amd64 SHA512 sums should be changed to(3rd and 4th are the relevant one for the config files): sha512sums=('91d65d658e9b6d0c0f03533e631ef1cc46c79ec5c1425a0914e1fe97ac4069230ee2f7d947df5cf76dd8865eb78f149d46140a3ec6509de21094ef70f27f6f61' '6ccc3d28b212ace866830e0e600e765c97f5179b178b57c164d92fd451247e495a5f90ce9c047f8d355c4cf8df67f820bdadf928a857b720a43946c4048d5587' '5151ee59093283f02c77c89959432269b16f0bd1b65d0e8da2b41403b182f1a31f239de00ace911aa668455b9b20647804809bcd2b41134be52420a469db5ccb' 'e9c34c725b0c92650e9ef0c8de7bdacf044f2b09aa8a59bd686120093bdab47cfaf25c85f1d93f5d7543c8bdf1f437908b79e056bb6ad72649cca2d22ba45b0f' 'fe4dcd7b5ec06ec3ec4aa631531469f58f6a7111e2d33affa98a1b8a8d230c5fa7e25ffdf770fe5ce61f249b0ec0ecd69df2858c4029acee0efaadff957858fe')

sknd commented on 2012-12-28 14:43 (UTC)

I'm unable to build, it fails: CC [M] net/wireless/lib80211_crypt_ccmp.o CC [M] net/wireless/lib80211_crypt_tkip.o LD net/wireless/built-in.o LD [M] net/wireless/cfg80211.o LD net/built-in.o ==> BŁĄD: Wystąpił błąd w build(). Przerywam... The build failed. Any ideas? What should i do?

shivik commented on 2012-12-27 17:21 (UTC)


msx commented on 2012-12-27 12:39 (UTC)

New sha512sums: 4bdb4006d268c90661fdc41844b1072add213a6ce69dc3fae7253a159c3efff087214fecc4caee6a337fe68b1e14da048ff80c7cd6db18b8194fd0d9df8b3813 config.i386 2dcebac5cd67489119240da6b732d9c9c50f2b535800e46183594002bbc50b4b6bb9e4633a9bf6cb729ea422a515737413d2cc96f284a3aeb18651405e614e5f config.amd64

commented on 2012-12-23 02:17 (UTC)

does someone know how to extract xz archive from the git tree..... changing shasum everytime isn't fun

shivik commented on 2012-12-15 07:26 (UTC)

Try one from here 3.5.4-1 should work.Not sure about 3.5.5-1,but try it.

commented on 2012-12-15 02:37 (UTC)

Author, please, can you give old 3.5.1 i686 kernel?

MatejLach commented on 2012-11-07 16:39 (UTC)

@shivik - Thanks for keeping the Liquorix kernel so up-to date!

shivik commented on 2012-10-21 05:45 (UTC)

Yes.It is built-in. [shivik@archenemy linux-lqx]$ grep -i acpi_cpufreq config.* config.amd64:CONFIG_X86_ACPI_CPUFREQ=y config.i386:CONFIG_X86_ACPI_CPUFREQ=y

martadinata666 commented on 2012-10-21 01:11 (UTC)

i can't insert acpi_cpufreq module on this kernel? but the cpupower still working, is it built-in on the kernel?

shivik commented on 2012-10-17 05:56 (UTC)

@engelus, in theory - yes.(CONFIG_X86_MARCH_NATIVE is still experimental) But you can also try CONFIG_MCORE2, which is supposed to optimize for i3,i5,i7 Some reading:

engelus commented on 2012-10-17 02:19 (UTC)

@shivik: If I want to enable specific optimizations for my cpu (core i3), in the archive config.i386 (I am building x86 kernel) should I have to coment line CONFIG_X86_GENERIC=y and uncoment line CONFIG_X86_MARCH_NATIVE ? Is that correct? Thank you.

silvik commented on 2012-09-18 14:39 (UTC)

looks like there's a config problem, 2 questions are asked regarding layer7 iptables match. after responding "m", looks like it's working, so not a big deal.

msx commented on 2012-09-17 14:12 (UTC)

Thanks for the feedback. Here sensors is working too but I still having the error message when running sensors-detect to config the utility: /usr/share/webapps # sensors radeon-pci-0100 Adapter: PCI adapter temp1: -128.0°C <-- wtf, lol! coretemp-isa-0000 Adapter: ISA adapter Core 0: +52.0°C (high = +95.0°C, crit = +105.0°C) Core 2: +52.0°C (high = +95.0°C, crit = +105.0°C) acpitz-virtual-0 Adapter: Virtual device temp1: +54.0°C Btw, so far community-testing/VirtualBox 4.2.0 with aur/virtualbox-hook with linux-lqx 3.5.3 is working great.

engelus commented on 2012-09-13 02:33 (UTC)

@msx I am using pure systemd system with lqx without any problems since 3.4.9

shivik commented on 2012-09-12 20:18 (UTC)

Config files are supplied by damentz - the author of the lqx patchset. Actually I just checked both config files for 3.5.3-2 lqx and they both have CONFIG_CGROUPS=y. [shivik@archenemy linux-lqx]$ grep -i cgroups config.* config.amd64:CONFIG_CGROUPS=y config.i386:CONFIG_CGROUPS=y and this is from my running 3.5.3-2 lqx kernel [shivik@archenemy linux-lqx]$ zcat /proc/config.gz | grep -i cgroups CONFIG_CGROUPS=y Click one of the links below,hit Ctrl+F and type cgroups You'll see that you have CONFIG_CGROUPS=y. msx,you say you haven't made the switch to systemd. Me neither, yet with fully updated system my sensors work. [shivik@archenemy ~]$ sensors it8720-isa-0228 Adapter: ISA adapter in0: +1.28 V (min = +0.00 V, max = +4.08 V) in1: +1.52 V (min = +0.00 V, max = +4.08 V) in2: +3.38 V (min = +0.00 V, max = +4.08 V) +5V: +2.96 V (min = +0.00 V, max = +4.08 V) in4: +3.09 V (min = +0.00 V, max = +4.08 V) in5: +2.19 V (min = +0.00 V, max = +4.08 V) in6: +4.08 V (min = +0.00 V, max = +4.08 V) ALARM 5VSB: +2.96 V (min = +0.00 V, max = +4.08 V) Vbat: +3.22 V fan1: 325 RPM (min = 0 RPM) fan2: 520 RPM (min = 0 RPM) fan3: 0 RPM (min = 0 RPM) fan4: 0 RPM (min = 0 RPM) temp1: +38.0°C (low = +127.0°C, high = +127.0°C) sensor = thermistor temp2: +36.0°C (low = +127.0°C, high = +60.0°C) sensor = thermal diode temp3: +39.0°C (low = +127.0°C, high = +127.0°C) sensor = thermistor cpu0_vid: +0.375 V intrusion0: ALARM k10temp-pci-00c3 Adapter: PCI adapter temp1: +25.5°C (high = +70.0°C) (crit = +67.0°C, hyst = +62.0°C)

msx commented on 2012-09-12 17:15 (UTC)

Hi @shivik, according to this post: user @nanix says "If you are running a custom kernel make sure you compile with CONFIG_CGROUPS = y, systemd requires it. The aur lqx no longer has this flagged.". Now, after a bit research I found these two interesting articles regarding this flag, and, where says that: a. CONFIG_CGROUPS isn't supported in the 3.5 kernel branch (1st link) b. it is needed with systemd (2nd link) Will you please consider enabling this flag in the future? While I personally didn't the jump yet to systemd it's certainly a matter of time until it become the de-facto init handler. Btw, all of this showed up when I tried to run the sensors-detect tool of lm_sensors application and got the error message: "Failed to get D-Bus connection: No connection to service manager", in my case not because systemd but because CONFIG_CGROUPS isn't available in the current kernel branch (found in Linux kernels: 2.6.24–2.6.39, 3.0–3.4, 3.6-rc+HEAD).

engelus commented on 2012-09-04 23:10 (UTC)

@shivik sorry for the delay I was testing stock and linux-ck and having the same problem, also slim brokes on logout. I Will search in the formums to see is anyone is having the same issue. Also i will revert the both kernel stock and lqx to 3.4.9 and see if the problem is present too. Thanks.

shivik commented on 2012-09-03 05:53 (UTC)

Everything rock solid here. Engelus,sounds like a power-saving issue of some sort. Are you sure it's the kernel and not other packages that you upgraded along with the kernel? Try the stock arch kernel or some other custom kernel(like linux-pf or linux-bfs) and see if you'll get the same problem.

engelus commented on 2012-09-03 01:57 (UTC)

I am having a strange issue with 3.5.3-2. If I leave my laptop on for a while without any kind of activity then the wifi is disabled and the compositing (xcompmgr) disabled and tint2 closed. So I have to relog to things work again. With 3.4.9 I havent any of this problems. Any ideas?

msx commented on 2012-09-01 21:09 (UTC)

@shivik: still having the same issue here: ERROR: could not insert 'vboxdrv': Exec format error ERROR: could not insert 'vboxnetadp': Exec format error ERROR: could not insert 'vboxnetflt': Exec format error ERROR: could not insert 'vboxpci': Exec format error I'm wondering if any of you guys have this issue with VirtualBox and Liquorix!? As a side note, the whole 3.4 branch worked okay and also VirtualBox is working great on linux-pf. @engelus thanks for the feedback, I will try uninstalling and then reinstalling all VirtualBox packages...

engelus commented on 2012-09-01 21:06 (UTC)

I am using the latest lqx build: [engelus@hades ~]$ uname -a Linux hades 3.5.3-2-lqx #1 ZEN SMP PREEMPT Fri Aug 31 20:06:34 ART 2012 i686 GNU/Linux

msx commented on 2012-09-01 20:48 (UTC)

Hey guys, hi! @engelus: are you running the latest Liquorix or anything from the 3.4 branch? @shivik: thanks, I'm gonna install the new x86_64 from your repo and post back my results! Kind regards, Martin

shivik commented on 2012-09-01 20:11 (UTC)

Doesn't matter the PKGBUILD is the same for 32- and 64-bit(only slightly different config* files). It only matters when it comes down to precompiled package binaries. I rebuilt the x86_64 version in the repo.That's why I was asking. But since you didn't have problems it's not the PKGBUILD.

engelus commented on 2012-09-01 20:00 (UTC)

To clarify I am using i686 kernel

shivik commented on 2012-09-01 19:52 (UTC)

Now I see engelus's post, so its probably not the PKGBUILD then.

shivik commented on 2012-09-01 19:48 (UTC)

@msx,is it the same error? I modified the PKGBUILD a bit just now.Fiddled with the "provides" section. You can try it if you want(or if you have a fast computer the compile time shouldn't be a problem). If you're running 64bit,you can try the precompiled package from the "repo".I just re-packaged that.

engelus commented on 2012-09-01 19:44 (UTC)

@msx I upated yesterday the kernel and I havent got any problem rebuilding the modules for virtualbox, I used: sudo dkms install vboxhost/4.1.20

msx commented on 2012-09-01 16:50 (UTC)

VirtualBox problem again with latest 3.5.3-2, anyone else?

shivik commented on 2012-09-01 05:39 (UTC)

I kinda set a repo :) It's not the final thing.More of a proof of concept. Hosting sucks,cuts out randomly while you are downloading,so you have to run pacman -Syu until it downloads the whole thing. Initial set up was with package signing,but it pacman kept throwing errors,so I have to find info about how to do it the right way. Anyway here goes.Add the following to your /etc/pacman.conf: [liquorix] SigLevel = Never Server =$arch Then run pacman -Syy to sync with new repo. Precompiled linux-lqx and nvidia-lqx.

shivik commented on 2012-08-31 16:21 (UTC)

Got it! Deleted my previous comment. Waiting for the build proccess to complete now.

shivik commented on 2012-08-31 06:19 (UTC)

I'll update in about 6 hours. Sorry guys - duty calls.

shivik commented on 2012-08-29 20:20 (UTC)

supercool :)

engelus commented on 2012-08-29 19:57 (UTC)

shivik thanks for your help if do what you said scaling works, also discovered that changing the value of /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor to ondemand also works :D and thanks again for your help

shivik commented on 2012-08-29 19:39 (UTC)

Very strange.I use powernow-k8(which is also built-in). Working out of the box. Can you try this(as root or sudo): 1.pacman -S cpupower 2.then edit /etc/conf.d/cpupower and uncomment governor='ondemand' 3.rc.d start cpupower (or /etc/rc.d/cpupower start) and check if scaling is working

engelus commented on 2012-08-29 11:37 (UTC)

Well the error for the module of vbox was that I havent installed virtualbox-source package, so problem resolved, sorry for that. --- I know cpu scaling is working because in stock when I log in the cpu are at 0.97 Ghz (I use conky for monitoring) my machine have an i3-350M @ 2,26 GHz, but when I log in in lqx the cpu are always at top GHz.

msx commented on 2012-08-29 11:32 (UTC)

@engelus: did you visit this forum post? May be I'm forgetting something or so, but following that instructions you shouldn't have any problem.

shivik commented on 2012-08-29 07:06 (UTC)

Hmm...everything is fine with your modules dir so vbox is _supposed_ to work... Does virtualbox work with the stock kernel? What errors do you get with lqx? As for cpu frequency scaling it should work out of the box with lqx for acpi-cpufreq. And you're right the governor is also in the kernel. So no need to load anything as module. LQX / srock kernel CONFIG_X86_ACPI_CPUFREQ=y / CONFIG_X86_ACPI_CPUFREQ=m Is scaling working on your machine or you are just concern,because you don't see acpi-cpufreq in lsmod?

engelus commented on 2012-08-28 23:53 (UTC)

[engelus@hades ~]$ ls /usr/lib/modules 3.4.9-1-ARCH 3.4.9-3-lqx extramodules-3.4-ARCH extramodules-3.4-lqx yep my sistem is up to date i suffer from versionist :D ---- I havent installed cpufrequtils because in stock kernel on demand governor is started at boot. lsmod from stock: lsmod from lqx:

shivik commented on 2012-08-28 21:11 (UTC)

Paste the output of "ls /usr/lib/modules" Also is your system updated pacman -Syu? acpi-cpufreq is built in the kernel.It's not a module. This is why you get the error.It's already in the kernel. Do you have cpufrequtils installed(this is replaced by cpupower by the way).

engelus commented on 2012-08-28 14:27 (UTC)

Hi i am using the kernel 3.4.9-3 it compiled succesfully but having problems with Virtual Box solution provided by msx doesnt work for me. Another problem the module acpi-cpufreq used by intel chipset for cpu frecuency scaling isnt loading at boot. if i try to load with modprobe doesnt report errors but if i try to unload it it gives me "libkmod: kmod_module_get_holders: could not open '/sys/module/acpi_cpufreq/holders': No such file or directory Error: Module acpi_cpufreq is in use" Thanks and sorry for my bad english

shivik commented on 2012-08-26 16:10 (UTC)

Try it with the default PKGBUILD options to see if it's a confguration problem or some change in the source code.

trizen commented on 2012-08-26 15:58 (UTC)

32-bit and a custom configuration ( ). I tried to compile it with _config="old" (with the same configuration that worked just fine with 3.4.7-1).

shivik commented on 2012-08-26 15:54 (UTC)

No ideas for the moment.Compiled fine on my machine. Are you runing 32 or 64 bit? Any custom configuration?

trizen commented on 2012-08-26 09:55 (UTC)

I'm trying to upgrade from 3.4.7-1 to 3.4.9-3, but I got this error: Any ideas?

commented on 2012-08-20 10:47 (UTC)

@shivik hi apologize for the delay in responding. i try CONFIG_PACKET=y and work! thanks a lot for your time again

shivik commented on 2012-08-15 13:53 (UTC)

@msx, nice to see your problem solved :)

msx commented on 2012-08-14 21:33 (UTC)

@shivik: tnx pal, everything's right now =) In case any user of this kernel keeps having problems with VirtualBox I will simplify here the steps needed to resolve this: 1. got to /usr/lib/modules 2. localize and rename any -lqx directory except the main one, in my case 3.4.8-1-lqx 3. sudo dkms uninstall vboxhost/4.1.18 && sudo dkms install vboxhost/4.1.18 Modules should load as usual now. And shivik, thanks for your great job packaging Liquorix!

shivik commented on 2012-08-14 05:24 (UTC)

@nmapper, modprobe_db is the reason why modprobe af_packet doesn't work. So do this then: Get the stock linux-lqx and don't change _config(leave it as _config="pkg") Then change CONFIG_PACKET=m to CONFIG_PACKET=y in config.i386 if you're running 32-bit or in config.amd64 if you're running 64-bit. Recompile the kernel and tell me if it works. @msx, see this: Might be the same problem.

commented on 2012-08-14 04:38 (UTC)

@shivit, yes i do it just like you say it. but, modprobe not found a module called af_packet. i imagine that module not found because i only compile modules listed in /var/log/modprobed.dd, thats why i had to add that module in modprobed_db. - Ok, i'll try to do this again, and notify you - thanks for you time

msx commented on 2012-08-13 20:23 (UTC)

@shivit, guys: VirtualBox modules compilation doesn't work with this linux-lqx release. The modules themselves seems to compile ok with: # dkms install vboxhost/4.1.18 but then I can't modprobe them: # modprobe {module} ERROR: could not insert 'vboxdrv': Exec format error ERROR: could not insert 'vboxnetadp': Exec format error ERROR: could not insert 'vboxnetflt': Exec format error ERROR: could not insert 'vboxpci': Exec format error This does not happen with stock kernel. Anyone else?

shivik commented on 2012-08-13 18:35 (UTC)

No, just modprobe af_packet as in my previous comment on the fly(don't edit anything,just issue the command modprobe af_packet) and then try to connect to your wifi. No recompiling,no rebooting. If it works then you need to modify your kernel config and change CONFIG_PACKET=m to CONFIG_PACKET=y. You can also try and edit configs that come with linux-lqx and change CONFIG_PACKET=m to CONFIG_PACKET=y. config.i386 if you're running 32-bit config.amd64 if you're running 64-bit Change _config to _config="pkg" and then recompile the kernel and tell me if it works,but try the first one first.

commented on 2012-08-13 18:06 (UTC)

- ok, i add "af_packet" as module in my /var/log/modprobed.db, recompile, and doesn't function. - i recompile linux-lqx "as is" without touch anything in PKGBUILD and function. I assume that is not a problem with the package. - sure is an problem with my modules. I'll try to found this and i post the solution. Thanks a lot again!

shivik commented on 2012-08-13 05:29 (UTC)

Troubleshooting starts assuming you are not an idiot :) I'll need more info,but try this first: (as root) "modprobe af_packet" (or sudo) "sudo modprobe af_packet"

commented on 2012-08-12 20:04 (UTC)

Hi, im compile this package with "_config="local"" in PKGBUILD. I have this modules in my /var/log/modprobed.db I have this wifi card "05:00.0 Network controller: Broadcom Corporation BCM4313 802.11b/g/n Wireless LAN Controller (rev 01)". When i boot into this kernel, my wifi (Network Manager, and Wicd) tells me that password is invalid. and i tried to connect to my wifi with wpa_passphrase and wpa_supplicant and i have this error: "l2_packet_init: socket(PF_PACKET): Address family not supported by protocol". Any idea? its a bug? i'm an idiot? thanks a lot!

commented on 2012-08-02 19:32 (UTC)

Thanks! It turns out that I had made the switch from /lib to /usr/lib. Everything worked once I upgraded. --rennis250

shivik commented on 2012-07-21 16:35 (UTC)

KARCH has nothing to do with that.I'm running 64bit also. I don't actually know how yaourt works,but since kmod 9-2 modules are now in /usr/lib not in /lib. There's a line in the PKGBUILD that moves the entire tree to the new location("mv "$pkgdir/lib" "$pkgdir/usr""). Have you read this: ? Is your system updated so it reflects the new changes? Whats your output of " pacman -Q kmod "? You can try and build linux-lqx with makepkg. 1.Download the tarball: 2.Extract the linux-lqx folder to a subfolder of your liking. 3.Open a terminal window,navigate into that folder and type" makepkg "

commented on 2012-07-21 02:13 (UTC)

Hello, I'm receiving a curious error: MKDIR /tmp/yaourt-tmp-rje/aur-linux-lqx/pkg/linux-lqx/lib/firmware/cpia2/ INSTALL /tmp/yaourt-tmp-rje/aur-linux-lqx/pkg/linux-lqx/lib/firmware/cpia2/stv0672_vp4.bin DEPMOD 3.4.4-2-lqx ERROR: could not open directory /tmp/yaourt-tmp-rje/aur-linux-lqx/pkg/linux-lqx/lib/modules/3.4.4-2-lqx: No such file or directory FATAL: could not search modules: No such file or directory ==> ERROR: A failure occurred in package_linux-lqx(). Aborting... ==> ERROR: Makepkg was unable to build linux-lqx. I took a look at the package_linux-lqx function and noticed that KARCH is set to x86 by default. I'm running an x64 system, if that has any bearing. Any ideas? Thanks, rennis250

shivik commented on 2012-07-14 15:10 (UTC)

Updated. Now includes changes, which come with the new kmod 9-2.Full story here:

doug commented on 2012-07-08 09:23 (UTC)

I modified the PKGBUILD so it works with new kmod from testing (at least for x64 systems)

shivik commented on 2012-07-06 17:27 (UTC)

Thanks, Harry! I didn't know actually. Doh...great timing...I have so much to do right now...

commented on 2012-07-05 22:31 (UTC)

@shivik - Just wondering if you were following the kmod stuff in testing? A discussion has started (with links to relevant arch discussions) over on the linux-ck AUR page. I guess it will effect those using custom kernels. Just thought I would give you a heads up just in case. Thanks

commented on 2012-06-25 07:43 (UTC)

I don't know why, but my kernel didn't become bigger, (still 3,2Mb), and boot-up time haven't improved... Probably, I've done something wrong("nomod" in PKGBUILD and then remove the check mark near "Use modules" in "xconfig").. I am also thinking about SSD but I am too lazy to buy it, copy system from HDD to SSD...)

shivik commented on 2012-06-22 04:59 (UTC)

How's your kernel size? What about the boot-up time? Did it improve? I've struggled in the past to achieve faster boot time and faster loading programs and found the ultimate tweak - it's called SSD :) One of my all time best upgrades I've done.

commented on 2012-06-21 08:36 (UTC)

Thank you. I have tried it, and i don't know why, kernel config still contains modules...

shivik commented on 2012-06-19 16:24 (UTC)

OK, I'm not sure how this is supposed to work,but it got me 18.5 meg kernel package after compile. It did ask me some questions prior to compile(don't know if that's normal). It compiled fine and built the package with no errors,but I didn't try and boot it. Just change _config="pkg" to _config="nomod" and give it a shot.

commented on 2012-06-18 13:00 (UTC)

Thanks a lot.

shivik commented on 2012-06-18 10:48 (UTC)

@Vekt, I'll dig into that as soon as I have more time.

commented on 2012-06-18 09:26 (UTC)

Please, add a posibility to compile it with "make localyesconfig". I wouldn't like to have any modules in my system in order to make it boot faster, but when i tried to change "localmodconfig" to "localyesconfig" in the PKGBUILD i got an errror at the end of compilation(The present kernel configuration has modules disabled. Type 'make config' and enable loadable module support. Then build a kernel with module support enabled.) Thank you.

shivik commented on 2012-06-17 06:25 (UTC)

Guess you're right:

Freso commented on 2012-06-16 18:33 (UTC)

@shivik: What you're describing is very much what I was expecting. What happens though, is that "make oldconfig" seems to be run as a fall-back. See After actually having read the log, it looks like a previously valid statement in the .config, "ZCACHE=m", is now invalid, and thus makes the process bark. I admit though, that this sounds more like an upstream bug, than an issue with the PKGBUILD - but now I've mentioned it here, just in case.

shivik commented on 2012-06-11 13:54 (UTC)

_config="old" uses configuration from your running kernel(which is in /proc/config.gz (try "zcat /proc/config.gz" in a shell)). _config="x" gives a gui to config.i386(or config.amd64) that comes with lqx. I tried it and it first copies the configuration from the running kernel and then runs xconfig on top of that. So in case you set these two options together it just follows the order in the PKGBUILD script, where zcat /proc/config.gz > ./.config is 1st and xconfig comes second. Generally, you have to use one thing or the other. Edit: {line removed} In fact there is no command `make oldconfig` in the lqx PKGBUILD. Can you tell me what are you trying to achieve so I can have a better understanding of your idea and maybe help you?

Freso commented on 2012-06-10 21:13 (UTC)

Having set '_custom="x"' and '_config="old"', it seems to run `make oldconfig` - which I'm not really interested in wading through. Any suggestions?

shivik commented on 2012-06-06 06:54 (UTC)

linux-lqx 3.3.8-1 PKGBUILD:

MatejLach commented on 2012-05-25 15:18 (UTC)

Ok, thanks.

shivik commented on 2012-05-25 12:35 (UTC)

There were two "provides" statements in the PKGBUILD by mistake. I left the the one with "linux=". Try it now and see if it works.

MatejLach commented on 2012-05-24 20:31 (UTC)

I am not sure why virtualbox requires the standard 'linux' package from the repos, when. Shouldn't 'provides' take care of this?

shivik commented on 2012-05-10 14:46 (UTC)

No problem ;)

commented on 2012-05-10 13:32 (UTC)

well that's embarassing ahah, i totally forgot about that, it works just fine now , thanks for the help and sorry if i wasted you're time with something so stupid :P

shivik commented on 2012-05-10 13:16 (UTC)

see this line -> -=- kernel /boot/vmlinuz-linux root=UUID=3f92645f-7064-41bb-a033-53c4339f7195 resume=UUID=38df3f39-8626-4976-9365-0f0b7ff22908 pcie_aspm=force i915.i915_enable_rc6=1 i915.i915_enable_fbc=1 i915.lvds_downclock=1 ro initrd /boot/initramfs-linux-lqx.img -=- it says to boot vmlinuz-linux(which is your stock kernel),as opposed to vmlinuz-linux-lqx(which is linux-lqx) Can you edit your /boot/grub/menu.lst(_only the LQX section_) so it becomes like this(you can copy&paste actually,but remove the old entries if you do): # (0) Arch Linux title Arch Linux LQX kernel root (hd0,5) kernel /vmlinuz-linux-lqx root=/dev/sda6 fastboot resume=/dev/sda5 pcie_aspm=force i915.i915_enable_rc6=1 i915.i915_enable_fbc=1 i915.lvds_downclock=1 ro initrd /initramfs-linux-lqx.img

commented on 2012-05-10 12:43 (UTC)

here it is, i left the stock kernel as the fallback

shivik commented on 2012-05-10 12:38 (UTC)

Can post your /boot/grub/menu.lst or at least the LQX section?

commented on 2012-05-10 11:07 (UTC)

nope, i tried but it still can't find the devices

shivik commented on 2012-05-09 20:42 (UTC)

Maybe the same problem:

shivik commented on 2012-05-09 20:24 (UTC)

I'm out of ideas. I had that same problem back when 3.2 was out and solved it with the fastboot option. It was not related specifically to linux-lqx,though...(linux-ck for example too was droppin' me to a shell)

commented on 2012-05-09 14:36 (UTC)

@shivik I tried but it still can't find the device :/

shivik commented on 2012-05-09 13:53 (UTC)

@clod89, try: mkinitcpio -p linux-lqx (as root or sudo) 2.add "fastboot" ot your /boot/grub/menu.lst lqx kernel line like so: kernel /vmlinuz-linux-lqx root=/dev/sda3 fastboot elevator=deadline ro Above is just an example do not copy the line just add fastboot to yours and reboot...

commented on 2012-05-09 09:39 (UTC)

I just tried 3.3.5, it compiled fine but it can't find neither the resume nor the root partition at boot and it just drops a recovery shell, booting the stock kernel works fine.

shivik commented on 2012-04-22 05:09 (UTC)

Glad it worked for you :)

msx commented on 2012-04-22 03:26 (UTC)

I compiled the kernel without the CONFIG_I2O_SCSI module and it did compile ok so next thing I did was clean the /tmp directory and then yaourt it again: everything worked smooth this time :)

msx commented on 2012-04-18 17:50 (UTC)

Excellent, thanks for the help, I will look upon that :)

shivik commented on 2012-04-18 06:43 (UTC)

@msx,I can't tell what might be the problem in your case from this error. It just says that it fails while building the i2o_scsi module. I'm running 3.3.2-0 now - compiled and installed with no problems. Are you running 32 or 64 bit? Did you "makepkg" in an empty folder or there were leftovers from previous lqx compiles? Is you system up-to-date (pacman -Syu)? For the sake of experiment try and disable CONFIG_I2O_SCSI=m(just put a # in front of the line in your config.i386(if 32bit) or config.amd64(if 64bit)) and run makepkg again. And maybe try the 3.2.14-lqx here and see if you got the same error: No one had problems with that,so give it a shot.

msx commented on 2012-04-18 03:48 (UTC)

@shivik: Thanks for your aur package, I've been using it for a while until recently when I switched to pf-kernel (btw, I agree Liquorix feels slicker...) Now, I have this error message when upgrading the package: compilation terminated. make[1]: *** [drivers/message/i2o/i2o_scsi.mod.o] Error 1 make: *** [modules] Error 2 ==> ERROR: A failure occurred in build(). Aborting... ==> ERROR: Makepkg was unable to build linux-lqx. ==> Restart building linux-lqx ? [y/N] ==> ---------------------------------- ==> :( @shadowed87: Yeah, Liqourix is a fatless marvel that uses ~100mb less than stock/-ck kernels and extends your laptop's battery life. You may want to check the wiki for other means to get most out of your battery specially cpufreq (with acpid), laptop-mode, aur/powertop2 and so on.

shivik commented on 2012-04-17 15:19 (UTC)

Updated to 3.3.2 using latest 3.2.1 configs. Package version is -0,so I can bump to -1 once damentz updates the config files on the liquorix website.

shivik commented on 2012-04-17 06:21 (UTC)

3.3.2-1 patch is out,but config folder is empty so I can't update the lqx package yet. You can't compile 3.3.1-1 for the same reason. Let's hope damentz will fix it soon. Index of /sources/3.3 Name Last modified Size Description Parent

shivik commented on 2012-04-11 20:17 (UTC)

3.2.14 PKGBUILD: nvidia module:

shadowed87 commented on 2012-04-05 10:59 (UTC)

yes it feels really slick, i only modified the config according to this

shivik commented on 2012-03-31 04:51 (UTC)

I think it won't be any different than the stock arch kernel in that aspect. Try it and see if you like it.

shadowed87 commented on 2012-03-30 15:14 (UTC)

is this kernel suitable for a laptop aswell?? i mean about power management...are there any drawbacks?

shivik commented on 2012-03-15 05:49 (UTC)

No.You don't need to change anything to build for x86_64. If you're running x86_64 it will build for x86_64 and produce a 64-bit binary Don't worry about it. KARCH=x86 means "pc paltform" in this case(does not define 32 or 64 bit). It just tells the script to use the correct kernel sources.

commented on 2012-03-15 00:39 (UTC)

So to build this for x86_64 all I need to do is change KARCH to "x86_64" in the PKGBUILD?

Freso commented on 2012-02-29 14:07 (UTC)

Like graysky's. But it looks like AUR is already in a Git repos, so I guess I can just clone this from there. :)

shivik commented on 2012-02-28 21:04 (UTC)

You mean like graysky's PKG build commits here: Or the liquorix patches that damentz does?

Freso commented on 2012-02-28 19:49 (UTC)

Is the development of this PKGBUILD going on in a Git repository anywhere by chance?

shivik commented on 2012-02-22 06:11 (UTC)

Updated to 3.2.7-1. config files are the same as 3.2.6-1 at the time I update the PKGBUILD. I don't know if damentz is going to update those or leave them the same for 3.2.7-1 In case he updates you'll get config.i386 ... FAILED config.amd64 ... FAILED until I update the PKGBUILD again.

shivik commented on 2012-02-15 05:38 (UTC)

config files were updated again

shivik commented on 2012-02-06 22:26 (UTC)

Updated checksums. I should probably bump package version too,but that will brake the sync with the lqx patch...

doug commented on 2012-02-06 13:18 (UTC)

config files were updated, so they fail validity check

Wintershade commented on 2012-02-01 10:05 (UTC)

@shivik - it does indeed - at the moment. I haven't tested all the modules yet (didn't get the chance to do so), what currently interests me most is the open-source virtualbox module, and the module from the nvidia-all package. I prefer nvidia-all to nvidia-lqx due to the number of kernels I test in my free time - although the -lqx is the "main" kernel I use. So when I find the time to test more, I'll drop you more feedback here. Thanks again :) you're doing a great job.

shivik commented on 2012-01-31 20:57 (UTC)

No problem:) Glad to hear it does the job for you.

Wintershade commented on 2012-01-31 10:11 (UTC)

@shivik, sorry for the late feedback, I was busy as hell for the past few days, didn't even see your updates. Anyway, it's working fine so far :) thanks for all your effort!

shivik commented on 2012-01-28 21:24 (UTC)

main PKGBUILD now uses extramodules. --- 3.2.2-1:oldstyle PKGBUILD: oldstyle nvidia drivers:

commented on 2012-01-28 14:14 (UTC)

Thanks, shivik.

shivik commented on 2012-01-25 12:22 (UTC)

I've decided to migrate linux-lqx PKGBUILD to accommodate extramodules. this is the final PKGBUILD before I replace it in AUR: nvidia modules for it: It works for me but, can you please test it on your PC's. It also addresses the 3.2.0-lqx issue.So it now shows the correct version in uname -r. I can share details if someone's interested. So... This one features: 1.the latest patch from damentz(BFS,BFQ,ZEN,etc.root exploit closed too) 2.updated to work with modprobed_db 3.custom configuration via menuconfig, nconfig, xconfig 4.ability to use current kernel config for the new one using /proc/config.gz 5.fixed naming in uname -r

doug commented on 2012-01-25 12:15 (UTC)

I noticed that version number bug with 3.1.0, maybe it started with 3.0, but I'm sure that it was OK on 2.6.x.x versions

shivik commented on 2012-01-25 10:07 (UTC)

@computerquip, it sure is no problem.As for the reason I'm digging into it right now, but it is always been that way for lqx. Run a pacman -Q linux-lqx and you'll see that the correct version is installed. @akurei,thanks :) I'm trying almost my best. By the way I can give the ownership to someone else if one is so inclined. I don't consider myself skillful enough. I was actually updating and disowning the package right away, but one day I just forgot to disown it and when I came back from work saw I'm still listed as the maintainer and said to myself "what the hell".

computerquip commented on 2012-01-25 06:34 (UTC)

Quick question, "uname -r" says the kernel is of version 3.2.0-lqx. Is that normal?

akurei commented on 2012-01-25 01:52 (UTC)

Been somewhat regretful of orphaning this package, but now I am glad you took over this shivik. My condolesences. This is madness ;) : You do a great job there!

shivik commented on 2012-01-24 17:58 (UTC)

experimental lqx 3.2.1-5 with extramodules - nvidia module for it -

shivik commented on 2012-01-24 06:43 (UTC)

Done. Thanks for your work, damentz! P.S. I was actually waiting for -5, but you always release your patches after I go to bed :)

damentz commented on 2012-01-24 02:42 (UTC)

@shivik: please update to 3.2.1-5 soon. This contains a patch to a recently discovered local privilege escalation vulnerability. More information here:

silvik commented on 2012-01-23 16:49 (UTC)

@damentz: thanks for including brcmsmac in the latest release.

shivik commented on 2012-01-23 06:58 (UTC)

experimental lqx 3.2.1-4 with extramodules - nvidia module for it -

commented on 2012-01-22 23:50 (UTC)

Out of date. Current patch is for 3.2.1-4, won't build.

shivik commented on 2012-01-21 17:30 (UTC)

@harryNID, request 1:Done request 2:Done Regarding the kernel configuration - it's what damentz sets up for us.Take a look here

commented on 2012-01-21 16:56 (UTC)

@shivik, Could you please take a look at the modprobed_db code. The new version of modprobed_db does not use the reload_database script any longer. For reference check out "modprobed_db" and "linux-ck" packages. It does require that sudo be set up however. Example: (Updated for this package from "linux-ck" PKGBUILD) ### Optionally load needed modules for the make localmodconfig # See if [ $_config = "local" ]; then msg "If you have modprobe_db installed, running it in recall mode now" if [ -e /usr/bin/modprobed_db ]; then [[ ! -x /usr/bin/sudo ]] && echo "Cannot call modprobe with sudo. Install via pacman -S sudo and configure to work with this user." && exit 1 sudo /usr/bin/modprobed_db recall fi msg "Running Steven Rostedt's make localmodconfig now" make localmodconfig else yes "" | make config fi Also, would it be possible to add nconfig along with menuconfig and xconfig. It has a nicer interface than plain menuconfig (IMO). Example: (Since I use nconfig I have "y" set to that but it's up to you) _menu="y" # menuconfig option [m = make menuconfig; x = make xconfig; nconf or "y" = nconfig n = none] if [ $_menu = "m" ]; then msg "Running make menuconfig" make menuconfig fi if [ $_menu = "x" ]; then msg "Running make xconfig" make xconfig fi if [ $_menu = "nconf" -o $_menu = "y" ]; then msg "Running make nconfig" make nconfig fi One last question, Is it just me or does the this new kernel config version have more things being built directly into the kernel? Is it the lqx config? It just seems like there is more things I have to manually go through and deselect from being built directly in than in previous versions so I can get a stripped down kernel. Thanks for your time!!

shivik commented on 2012-01-21 07:02 (UTC)

Oops, I think I got it.You have to use this nvidia PKGBUILD with the extramodules lqx I used that on both machines,but never included it in the original tar.gz, because I didn't know you use nvidia at that time. It will be good if others test if extramodules lqx works for them. If it does I'll replace the main lqx package with one that uses extramodules.

shivik commented on 2012-01-21 06:29 (UTC)

@Wintershade,looks like you have some linux-headers problem.At least this is what "Unable to determine the target kernel version." I get when I mess up the location of the linux headers when I'm tinkering around experimenting. ""nvidia: disagrees about version of symbol module_layout" - this I get when I try to load a nvidia module that is compiled for a different version of the kernel. for example 3.1.5 module on 3.1.8 kernel(pure example there with versions).Or maybe (3.1 module on 3.2 kernel). What's inside your "ls /lib/modules/extramodules-3.2-lqx/".I get a "nvidia.ko.gz" and "version". I checked the extramodules kernel on another machine at home and it does work there as well(I only have nvidia module),but it boots to gui no problem. I'm actually still running the extramodule kernel that I did for you on 19th of jan. That doesn't prove anything 100% of course,but is a starting point where to look.

silvik commented on 2012-01-21 00:14 (UTC)

@damentz: no, everything else works fine on my hardware, just like 3.1 did. thanks. this broadcom thing is new to 3.2, some new driver design decision made by kernel devs. looks like in 3.3 things will change again:

Wintershade commented on 2012-01-20 23:22 (UTC)

@shivik - oops! I'm sorry, but my previous comment turned out to be a false positive. at least regarding nvidia-all, which is supposed to install into the /lib/modules/extramodules-3.2-lqx. The error I'm receiving is "nvidia: disagrees about version of symbol module_layout. Even with intel_iommu=off the error persists. OTOH, now even the nvidia-lqx fails to compile, stating "Unable to determine the target kernel version." (I'm in command line, using links now.) So what gives?

damentz commented on 2012-01-20 23:18 (UTC)

@silvik: well, I can go ahead and change that directly. Is there anything else you noticed missing from the 3.2 kernel? You can also visit the Liquorix forum if you have any other feedback.

silvik commented on 2012-01-20 23:04 (UTC)

looks like lqx config does not enable the new broadcom drivers (brcmsmac) as mainline arch does. if you have a broadcom wifi (like me) you need to edit the PKGBUILD to enable menuconfig (_menu="m"). in the menuconfig disable BCMA support (in Broadcom specific AMBA section) and enable Broadcom IEEE802.11n PCIe SoftMAC WLAN driver (in Network devices / Wireless LAN) is it possible to make this small change in the lqx package config? thanks.

Wintershade commented on 2012-01-20 18:45 (UTC)

@shivik, it compiles and works well. virtualbox module doesn't work due to different issues on my system - however a package nvidia-all compiles and works like a charm (I'm using it rather than your nvidia-lqx due to a number of other kernels I'm testing whenever I can). So anyway, absolutely positive feedback from me. Kudos!

shivik commented on 2012-01-19 15:05 (UTC)

@Wintershade, there you go: Compiled and installed fine on my system. So it works for me. Give it a shot and let me know if it works for you.

shivik commented on 2012-01-19 11:01 (UTC)

For anyone having issues booting with with nvidia driver use amd_iommu=off or intel_iommu=off(depending on your motherboard chipset). Chances are high that you'll be able to boot to graphical environment. @Wintershade - I'll work on that.

Wintershade commented on 2012-01-18 19:59 (UTC)

Could you please add support for external modules (i.e. /lib/modules/extramodules-XY)? Without those, external modules (such as virtualbox driver) won't work...

shivik commented on 2012-01-18 06:35 (UTC)

That's because the package was out of date(a 3.2.1-2 patch was released). It's updated now so you shouldn't have problems.

commented on 2012-01-17 20:58 (UTC)

I am trying it install the liquorix kernel and I get the following error. I think it is trying to download 3.2.1-1.patch.gz which doesn't exist. ==> Making package: linux-lqx 3.2.1-1 (Tue Jan 17 14:53:14 CST 2012) ==> Checking runtime dependencies... warning: config file /etc/pacman.conf, line 33: directive 'ShowSize' in section 'options' not recognized. ==> Checking buildtime dependencies... ==> Retrieving Sources... -> Downloading linux-3.2.tar.xz... % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 100 62.0M 100 62.0M 0 0 305k 0 0:03:28 0:03:28 --:--:-- 300k -> Downloading 3.2.1-1.patch.gz... % 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 3.2.1-1.patch.gz Aborting...

shivik commented on 2012-01-17 18:04 (UTC)

No problem.Just exit and the build process will continue normally. It's back to how it was before(no menuconfig). This nvidia issue is driving me crazy. Downgraded to 3.1.8 and the problem is gone. The moment I switch to 3.2.1 and I get a black screen.

silvik commented on 2012-01-17 16:42 (UTC)

i get the make menuconfig configuration dialog. is this ok?

shivik commented on 2012-01-17 07:06 (UTC)


commented on 2012-01-16 19:46 (UTC)

please can this use the linux-3.2.tar.xz archive like . Thank you.

shivik commented on 2012-01-16 17:54 (UTC)

Fixed. Wrong checksums were left from the experimental configs when I was trying to fix the nvidia problem.

silvik commented on 2012-01-16 15:07 (UTC)

==> Validating source files with sha512sums... linux-3.2.tar.bz2 ... Passed 3.2.1-1.patch.gz ... Passed config.i386 ... FAILED config.amd64 ... FAILED linux-lqx.preset ... Passed ==> ERROR: One or more files did not pass the validity check! ==> ERROR: Makepkg was unable to build linux-lqx.

shivik commented on 2012-01-15 21:03 (UTC)

I just tried to compile linux-ck and nvidia-ck and got the same result. So it's either something with my hardware or something in the 3.2 tree.

shivik commented on 2012-01-15 20:27 (UTC)

3.2.1 is out,but I can't get the nvidia driver to work with it. I get a black screen when it starts loading X. I adapted the linux-ck PKGBUILD both for kernel and nvidia for the -lqx - no good.

shivik commented on 2012-01-11 15:38 (UTC)

Updated to 3.1.8.

Wintershade commented on 2012-01-10 16:54 (UTC)

nope, it's still (officially) maintained my Mavirick. @Mavirick, please update the package or disown it so that somebody else can do it if you cannot or don't want to anymore.

commented on 2012-01-09 19:23 (UTC)

Has this package been orphaned or something?

damentz commented on 2012-01-09 01:18 (UTC)

3.1.8 is now out.

Wintershade commented on 2012-01-07 10:28 (UTC)

3.1.7 is released. this package is out of date indeed. check out

commented on 2012-01-04 01:05 (UTC)

Please stop marking out of date as there is no released from

commented on 2011-12-30 12:21 (UTC)

Updated the Kernel with the patches, tested and verified it working. Sorry for the delay, and an early Happy New Years to you all.

commented on 2011-12-30 01:56 (UTC)

Aye, I noticed it was out the other day. I am currently traveling and will try and get something up in the next day or two.

commented on 2011-12-29 18:00 (UTC)

3.1.6 is out."3.1.6-1.patch.gz 24-Dec-2011 12:54 685K "

aericson commented on 2011-12-20 17:33 (UTC)

Anyone managed to get truecrypt working with it?

akurei commented on 2011-12-04 23:31 (UTC)

You're welcome =) Have fun!

commented on 2011-12-04 22:15 (UTC)

Thanks for pointing out that msg akurei I completely forgot about it when I was modding the PKGBUILD for upload.

silvik commented on 2011-12-04 21:40 (UTC)

thanks for adopting lqx, Mavirick. that's the power of open source, in a few days problems get fixed. updating...

akurei commented on 2011-12-04 21:31 (UTC)

You might want to remove: msg "" msg "This package has been orphaned. Feel free to adopt." msg ""

commented on 2011-12-04 18:52 (UTC)

We're cookin' now :) Thumbs up!

commented on 2011-12-04 06:06 (UTC)

I believe I applied to adopt it, so I suppose I am just waiting.

commented on 2011-12-04 06:03 (UTC)

But how can you take over when this RcrdBrt guy is nowhere to be found? Maybe we should signal the archlinux staff?

commented on 2011-12-04 01:09 (UTC)

I've got everything updated and tested and ready to upload.

commented on 2011-12-04 00:38 (UTC)

I'm in the process of creating the updated PKG Build and I will be taking over the maintenance of the package for the time being.

akurei commented on 2011-12-03 20:34 (UTC)

3.1.4-1 has been released by LQX now =)

commented on 2011-12-02 22:48 (UTC)

A 3.1.4-1 patch of lqx has not been released yet, once it is I will work on getting an updated version.

commented on 2011-12-02 18:25 (UTC)

Yeah, linux-lqx is a good kernel and doesn't deserve this. Maybe this guy clicked on "Adopt packages" by accident?

silvik commented on 2011-12-02 13:13 (UTC)

hey Mavirick / td123 / someone with sufficient skills and time, why don't you take ownership of this package or create linux-liquorix or something like this and maintain it properly? it's already 2 versions behind official arch kernel. the maintainer doesn't care anyway.

commented on 2011-11-30 07:07 (UTC)

Thanks,Mavirick! Too bad the _new_ maintainer doesn't seem to care. It's been 3 days since lqx 3.1.3 is out and still no update.

commented on 2011-11-30 06:47 (UTC)

I've created and tested a new pkgbuild for linux-lqx-3.1.3-1 it requires the new mkinitcpio 0.8 which moves mkinitcpio from /sbin/ to /usr/bin Here is the PKGBUILD: Here is the Modified linux-lqx.install:

shivik commented on 2011-11-23 07:13 (UTC)

td123, what I meant was that you can use a different name for your kernel for ex. linux-td123, linux-td, linux-lqxtd (and still base it on the -lqx)- you get my point. I believe in diversity so I welcome your unique approach, but still I don't think linux-lqx should be changed. That's why I suggested a new AUR package.

td123 commented on 2011-11-23 06:36 (UTC)

Ok, fine, I guess I wont take ownership of this package then.. I did manage to rebase against the [core] linux PKGBUILD. You can find the package at: fyi, this package is *a lot* cleaner and perhaps more bug free than the one provided on this page. Possibly easier to maintain. I haven't experienced any problems so far with my package.

akurei commented on 2011-11-22 22:31 (UTC)

I know, I am not maintining this anymore, and thus should not interfere, but please td123, leave the options for make xconfig and make menuconfig etc in. They're useful for everyone and don't harm anyone.

td123 commented on 2011-11-22 21:49 (UTC)

There is a good reason those patches do get applied to the arch kernel (fix freezes on certain hardware). The arch kernel does the compression also, so I think it's safe to say that this kernel should be using _compress_modules="yes" for all cases. I can't think of a reason why you wouldn't want this. The problem with creating another linux-lqx pkg in the aur is that it would be a duplicate. I guess for now, I will just try reworking the PKGBUILD and see what you guys think before posting it here...

shivik commented on 2011-11-22 19:15 (UTC)

I'm kind of against this. The main idea is to get the base kernel with the liquorix patchset applied only. As for the options - almost all of the other AUR kernels have those and they're kind of useful sometimes. Maybe _compress_modules is useless... A PKGBUILD cleanup would be nice though Your ideas are by no means bad,but I think this package should be left the way it is. Maybe you should create a new kernel based on the -lqx one and realize your ideas.

td123 commented on 2011-11-22 18:14 (UTC)

I'll try to rebase this kernel against the latest linux in [core]. My current thoughts are what to do about patches that get applied? Also, I'm planning on removing some if not all of the options provided at the top: _menu, _config, _compress_modules And general cleanup of the PKGBUILD so that future maintenance would be a lot easier. Does anyone have any opinions for or against this? If I can't update the PKGBUILD in about a week, I will just give up since that will probably mean I don't have enough time to figure all of this out.

shivik commented on 2011-10-31 06:29 (UTC)

I tried it - 18 hours running already with no problems. I can update the PKGBUILD here if there's no one else to do it.

commented on 2011-10-30 23:13 (UTC)

anyone tried substituting the 8 patch for the 6 this PKGBUILD calls for? I'd try but exams taking up soo much time... Anyhow thanks.

damentz commented on 2011-10-29 21:05 (UTC)

You might want to look at this post: LVM requires you to have at least three numbers in your kernel version. For instance, 3.0-6.dmz.1-liquorix-amd64 broke support for LVM and other disk layer user space, but adding in a second 0 for compatibility (3.0.0-6.dmz.1-liquorix-amd64) with the old 2.6.n format works around this bug. You can check with `uname -r`, make sure you have three numbers in the kernel version (3.0.0 for instance). I suspect that you only need to modify your package scripts to accommodate this change.

akurei commented on 2011-10-29 20:51 (UTC)

When using linux-lqx, I am not able to snapshot a lvm volume. Also while booting, there are error messages like "hey i cannot find your LVM volumes, but i will let you boot anyway". This problem only appears on linux-lqx. All other kernels work fine. I double, triple, quadruple-checked the .config files for dm_*, and still this error persists. I simply cannot run my LVM Backupscript when I am running the lqx kernel. My lvm backupscript: URL: RAW:

damentz commented on 2011-10-29 20:45 (UTC)

You're the first to report this bug, can you tell me more about it on the Liquorix forums?

akurei commented on 2011-10-29 17:47 (UTC)

Orphaned this package, because I got tired of the dm_snapshot bug with this. I am using linux-pf now. Feel free to adopt.

akurei commented on 2011-10-28 13:03 (UTC)

Changed that.

Alir3z4 commented on 2011-10-28 00:21 (UTC) is online now, i changed the source line in the PKGBUILD and it's worked as well. but the mirror link isn't working. tnx

akurei commented on 2011-09-12 00:24 (UTC)

Pushed to 3.0.4-2. I change the download-server for as long as is down.

akurei commented on 2011-09-11 20:10 (UTC) is down, so I am currently looking for a solution. Stay tuned.

commented on 2011-09-11 17:42 (UTC)

The build command should be: make ${MAKEFLAGS} bzImage modules

akurei commented on 2011-09-06 12:33 (UTC)

I probably won't touch the "default" config files from because, well, they're the default ones.

commented on 2011-09-05 09:47 (UTC)

@capoeira you probably need to set the "Packet Socket" option in Networking. I had the same problem and i think setting this fixed it. BTW, in the configfiles there is 'CONFIG_FAT_DEFAULT_IOCHARSET="utf8"'. I don't think this is good since you can't mount most FAT-Filesystems with that (and the Kernel warns about that in dmesg). It should probably be set to 'iso8859-1' (thats what the stock kernel has)

capoeira commented on 2011-09-04 17:04 (UTC)

Can anybody here help me out? I'm new to this localmod config thing. I have this script running wich records all used modules. When I compile this kernel with localmod my wifi doesn't work (it works installing the whole kernel). When I try to conect manualy (WICD-GUI) to my router I get this st like this "couldn't conect, wrong key-phrase". The key isn't wrong it's the same wich is working now I'm posting this. What could that be? Also on my first attemp to compile with localmod I got a "couldn't load X25, not found" message. So I deleted it from the module-list. Can anybody give me a clue?

commented on 2011-09-02 15:32 (UTC)

It must have been a bug in grub2 itself. After 1.99-4 update everything's back to normal.

akurei commented on 2011-09-01 18:09 (UTC)

@bfo: Sorry, I am not using GRUB2 atm.

akurei commented on 2011-09-01 15:41 (UTC)

Pushed to 3.0.4-1

commented on 2011-08-30 14:57 (UTC)

Does anyone have problems with GRUB2 not detecting vmlinuz-linux-lqx-fallback and both *-lqx initramfs images? Only vmlinuz-linux-lqx gets on the list while generating GRUB2 config, but without the corresponding ramfs.

akurei commented on 2011-08-26 22:58 (UTC)

Nice =) Almost three days without a notification. Feels like vacation ;)

akurei commented on 2011-08-24 19:26 (UTC)

Sorry, I did not keep the old "config.*"-files.

commented on 2011-08-24 19:18 (UTC)

Hmmm, that's interesting. I wonder what Damentz changed?

akurei commented on 2011-08-24 10:02 (UTC)

Pushed to 3.0.3-3.

akurei commented on 2011-08-24 09:58 (UTC)

Well, this time it's not my fault ;) The files were changed on the server: ______Name____________________Last_modified______Size__ ______config.i386_____________22-Aug-2011_18:18__135K__ ______config.amd64____________22-Aug-2011_18:18__128K__ I'm going to fix that asap.

capoeira commented on 2011-08-24 01:34 (UTC)

the two files still fail the check

akurei commented on 2011-08-19 22:29 (UTC)

Updated to 3.0.3-2, thanks disastrophe! =)

commented on 2011-08-19 18:34 (UTC)


akurei commented on 2011-08-19 17:46 (UTC)

Seems like makepkg still had the old config files and did not download the new ones, despite me running makepkg -o in advance. I though by running makepkg -o and makepkg -g afterwards it would download the files and then generate the checksums. But instead it used the files in my local cache ... wtf? Welle anyway I am going to fix that as soon as I come home.

commented on 2011-08-19 14:32 (UTC)

-> Found linux-lqx.preset ==> Validating source files with sha512sums... linux-3.0.tar.bz2 ... Passed 3.0.3-1.patch.gz ... Passed config.i386 ... FAILED config.amd64 ... FAILED linux-lqx.preset ... Passed ==> ERROR: One or more files did not pass the validity check! ==> ERROR: Makepkg was unable to build linux-lqx.

akurei commented on 2011-08-19 12:25 (UTC)

Updated to 3.0.3-1

akurei commented on 2011-08-16 23:43 (UTC)

Updated to 3.0.1-1

akurei commented on 2011-08-16 23:16 (UTC)

Already testing. I am out geocaching at the moment. When I come home the test build should be finished and I will upload the package.

commented on 2011-08-16 23:11 (UTC)

lqx pathset is out for 3.0~ ;)

akurei commented on 2011-08-09 15:48 (UTC)

Seems like I forgot the sed-lines in the PKGBUILD on lines 142 and 143. Pushed to -6

commented on 2011-08-09 15:19 (UTC)

.4-5 also works fine but it still comes out named linux-lqx.img,

akurei commented on 2011-08-09 10:34 (UTC)

Fixed a minor glitch (initramfs in /boot/ was named linux-lqx.img and linux-lqx-fallback.img instead of the canonical initramfs-linux-lqx.img and initramfs-linux-lqx-fallback.img) Please update your bootloader (GRUB menu.lst) to reflect these changes.

commented on 2011-08-09 01:26 (UTC)

Works good and yeah, thanks for all the hard work. You da man!

flan_suse commented on 2011-08-08 21:32 (UTC)

Thanks for the new PKGBUILD. Glad to see AUR packages keeping up with the Arch updates. I'll wait until LQX 3.0 before compiling this, but just wanted to say thanks!