The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

speedyx · 2013-04-08 21:56:45

Hallo, I'm new here!

Making my own linux-pf package I have today two problems with the 3.8.2 verion.

1) the name "linux-pf-nehalem " in PKGBUILD exceed with a blank character at the line 444. Just fixed with removing it.

2) at the end of the package's installing I obtain the following error:

==> ERROR: module not found: `usbhid'

I tried to include it in the IGNORE list of the modprobed_db.conf but the error persist.
I tried to disable all the checkboxes in the menuconf that refer to USB Hid but without success.
Have anyone any suggestion?

Last edited by speedyx (2013-04-09 04:47:51)

Thaodan · 2013-04-09 04:23:41

Thanks for report, I fixed my typo.
But I don't know anything about the error you got as I builded linux-pf it with ARCH module config.

nous · 2013-04-11 06:14:29

speedyx wrote:

Hallo, I'm new here!
Making my own linux-pf package I have today two problems with the 3.8.2 verion.
1) the name "linux-pf-nehalem " in PKGBUILD exceed with a blank character at the line 444. Just fixed with removing it.
2) at the end of the package's installing I obtain the following error:
==> ERROR: module not found: `usbhid'
I tried to include it in the IGNORE list of the modprobed_db.conf but the error persist.
I tried to disable all the checkboxes in the menuconf that refer to USB Hid but without success.
Have anyone any suggestion?

That's a mkinitcpio error, see this bug report for more info. Anyway, I don't think you need usbhid in MODULES; you can just add 'keyboard' to HOOKS.

jbernardo · 2013-04-12 05:07:01

Hi,
I am trying to update to 3.8.2-3 from the repository, but now it fails with "linux-pf: missing required signature". Should I just disable signature checking in pacman?

nous · 2013-04-12 06:30:21

jbernardo wrote:

Hi,
I am trying to update to 3.8.2-3 from the repository, but now it fails with "linux-pf: missing required signature". Should I just disable signature checking in pacman?

I don't sign the packages in my repo, you should set the SigLevel directive for [pfkernel] to Optional or Never. Note that you can set this per repo.

jbernardo · 2013-04-12 09:06:44

nous wrote:

I don't sign the packages in my repo, you should set the SigLevel directive for [pfkernel] to Optional or Never. Note that you can set this per repo.

Thanks, I've added the SigLevel directive to the pfkernel repo, and all is fine. I wasn't sure it was still supported, as it was removed from the default repositories when pacman last updated.

nous · 2013-04-12 17:00:19

jbernardo wrote:

nous wrote:
I don't sign the packages in my repo, you should set the SigLevel directive for [pfkernel] to Optional or Never. Note that you can set this per repo.
Thanks, I've added the SigLevel directive to the pfkernel repo, and all is fine. I wasn't sure it was still supported, as it was removed from the default repositories when pacman last updated.

I don't know why it was removed from the wiki list, I've added it back. I'll keep uploading latest binaries for as long as I have a few minutes of time to run my build scripts, unless someone else sets up a "better" repo (i.e. not a dropbox one) and builds CPU-specific packages besides the generic ones. As a side note, I recently saw at least two new repositories hosting generic linux-pf packages.

Thaodan · 2013-04-13 16:08:56

The problem is some that can host a server, personally I build very linux-pf build for sandy and ivry-bridge every version.

Hercules · 2013-04-23 18:24:00

Why does uname -r say 3.8.0 when install says 3.8.2-3?

jakob · 2013-04-24 13:33:02

Hercules wrote:

Why does uname -r say 3.8.0 when install says 3.8.2-3?

Have you rebooted? I've got the repo version here and it seems alright:

jakob ❯ uname -r && pacman -Qi linux-pf-core2 | grep Version                                              
3.8.2-pf
Version        : 3.8.2-3

Thaodan · 2013-04-25 15:53:57

Hercules wrote:

Why does uname -r say 3.8.0 when install says 3.8.2-3?

Do you striped the last version number while building?

kostett · 2013-04-28 10:33:01

> linux-pf-atom: missing required signature
what I can to do to fix it?

lucke · 2013-04-28 10:51:11

With linux-pf from the repo:

# pacman -Q linux-pf 
linux-pf 3.8.2-3
# pacman -Ql linux-pf
...
linux-pf /usr/lib/modules/3.8.0/
...

Thaodan · 2013-04-29 04:42:13

kostett wrote:

> linux-pf-atom: missing required signature
what I can to do to fix it?

See a few posts above.

Thaodan · 2013-05-05 10:37:20

updated to 3.9

jakob · 2013-05-06 13:31:51

Thaodan wrote:

updated to 3.9

Are there any plans to maintain a repository again or will we need to compile the packages ourselves? Not that I'm hereby asking for somebody to do that, just for information. I'm totally fine with compiling the kernel myself!

Thaodan · 2013-05-06 13:37:06

The problem is that I need a server to host the repo, Currently I upload the sandy and ivvybridge version to my repo, as I build them for my self.

abraxas88 · 2013-05-07 12:10:52

I am getting a build() error

  LD [M]  drivers/net/wireless/rtlwifi/rtl8723ae/rtl8723ae.o
  LD      drivers/net/wireless/built-in.o
  LD      drivers/net/built-in.o
  LD      drivers/built-in.o
==> ERRORE: Si è verificato un errore in build().
    L'operazione sta per essere interrotta...

It always stops after built-in.o.
What's going on? :\

Thorsten Reinbold · 2013-05-07 13:24:05

abraxas88 wrote:

I am getting a build() error

  LD [M]  drivers/net/wireless/rtlwifi/rtl8723ae/rtl8723ae.o
  LD      drivers/net/wireless/built-in.o
  LD      drivers/net/built-in.o
  LD      drivers/built-in.o
==> ERRORE: Si è verificato un errore in build().
    L'operazione sta per essere interrotta...

It always stops after built-in.o.
What's going on? :\

Maybe it would help if you would start building with "export LC_ALL="C"", so that people from outside italy could read your error messages.

abraxas88 · 2013-05-07 13:28:52

Thorsten Reinbold wrote:

Maybe it would help if you would start building with "export LC_ALL="C"", so that people from outside italy could read your error messages.

If it had said anything relevant I would have posted the translation. It simply says "There's been an error in build() [like I said], the operation has been interrupted".

Thaodan · 2013-05-07 13:38:28

Please provide more parts of your log,

omerglr · 2013-05-07 18:40:01

Another build error here:

ERROR: "follow_page_mask" [kernel/power/tuxonice_core.ko] undefined!
ERROR: "page_mapping" [kernel/power/tuxonice_core.ko] undefined!
  LZMA    arch/x86/boot/compressed/vmlinux.bin.lzma
make[1]: *** [__modpost] Hata 1
make: *** [modules] Error 2
make: *** Waiting for unfinished jobs....

and another:

CC      arch/x86/kernel/setup_percpu.o
kernel/power/tuxonice_bio_signature.c: In function 'remove_old_signature':
kernel/power/tuxonice_bio_signature.c:185:10: UYARI: argument to 'sizeof' in 'memcpy' call is the same pointer type 'char *' as the destination; expected 'char' or an explicit length [-Wsizeof-pointer-memaccess]
    sizeof(no_image_signature_contents));
          ^
  CC [M]  kernel/power/tuxonice_swap.o
  CC [M]  kernel/power/tuxonice_file.o
  CC      arch/x86/kernel/mpparse.o
  CC [M]  kernel/power/tuxonice_userui.o
  CC      arch/x86/kernel/trace_clock.o
  CC      arch/x86/kernel/machine_kexec_64.o
  LD      kernel/power/tuxonice_core.o
  LD [M]  kernel/power/tuxonice_bio.o
  LD      kernel/power/built-in.o
  AS      arch/x86/kernel/relocate_kernel_64.o
  CC      kernel/sched/bfs.o
  CC      arch/x86/kernel/crash.o
  CC      arch/x86/kernel/module.o
  CC      arch/x86/kernel/early_printk.o
  CC      arch/x86/kernel/hpet.o
kernel/sched/bfs.c: In function 'irqtime_account_hi_si':
kernel/sched/bfs.c:2333:2: hata: 'nsecs_to_cputime64' işlevinin örtük bildirimi [-Werror=implicit-function-declaration]
  latest_ns = nsecs_to_cputime64(this_cpu_read(cpu_hardirq_time));
  ^
kernel/sched/bfs.c: Üst düzeyde:
kernel/sched/bfs.c:2847:13: UYARI: 'account_idle_times' tanımlanmış ama kullanılmamış [-Wunused-function]
 static void account_idle_times(cputime_t cputime)
             ^
cc1: some warnings being treated as errors
make[2]: *** [kernel/sched/bfs.o] Hata 1
make[1]: *** [kernel/sched] Hata 2
make: *** [kernel] Error 2
make: *** Waiting for unfinished jobs....

ends with:

 AS      arch/x86/vdso/vdso32/sysenter.o
  LD      arch/x86/realmode/built-in.o
  CC      arch/x86/vdso/vdso32-setup.o
  VDSO    arch/x86/vdso/vdso.so.dbg
  VDSO    arch/x86/vdso/vdso32-int80.so.dbg
  VDSO    arch/x86/vdso/vdso32-syscall.so.dbg
  VDSO    arch/x86/vdso/vdso32-sysenter.so.dbg
  VDSOSYM arch/x86/vdso/vdso-syms.lds
  VDSOSYM arch/x86/vdso/vdso32-int80-syms.lds
  VDSOSYM arch/x86/vdso/vdso32-syscall-syms.lds
  OBJCOPY arch/x86/vdso/vdso.so
  OBJCOPY arch/x86/vdso/vdso32-int80.so
  OBJCOPY arch/x86/vdso/vdso32-syscall.so
  OBJCOPY arch/x86/vdso/vdso32-sysenter.so
  VDSOSYM arch/x86/vdso/vdso32-sysenter-syms.lds
  AS      arch/x86/vdso/vdso.o
  AS      arch/x86/vdso/vdso32.o
  VDSOSYM arch/x86/vdso/vdso32-syms.lds
  LD      arch/x86/vdso/built-in.o
  LD      arch/x86/built-in.o
==> ERROR: A failure occurred in build().
    Aborting...

Last edited by omerglr (2013-05-07 18:45:50)

Thaodan · 2013-05-08 13:45:30

Still without LANG=C ....

Last edited by Thaodan (2013-05-08 13:45:43)

jakob · 2013-05-08 16:25:38

Hm, I rebuilt it yesterday evening without any errors. Using localmodconfig and optimising for corei7-avx.

I've got a different problem: Although zgrep ONDEMAND /proc/config.gz outputs

CONFIG_CPU_FREQ_DEFAULT_GOV_ONDEMAND=y
CONFIG_CPU_FREQ_GOV_ONDEMAND=y

I can't activate the governor:

jakob ❯ cat scaling_available_governors                                                     /s/d/s/c/c/cpufreq
performance powersave

But that's most likely unrelated to pf and rather due to my personal build…

omerglr · 2013-05-08 18:04:44

Actually mine was with "export LC_ALL="C"" as you said Thaodan. I solved my compile problem. Disabled Tuxonice completely and disabled Full dynticks support from RCU subsystem.

Arch Linux

#676 2013-04-08 21:56:45

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#677 2013-04-09 04:23:41

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#678 2013-04-11 06:14:29

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#679 2013-04-12 05:07:01

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#680 2013-04-12 06:30:21

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#681 2013-04-12 09:06:44

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#682 2013-04-12 17:00:19

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#683 2013-04-13 16:08:56

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#684 2013-04-23 18:24:00

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#685 2013-04-24 13:33:02

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#686 2013-04-25 15:53:57

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#687 2013-04-28 10:33:01

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#688 2013-04-28 10:51:11

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#689 2013-04-29 04:42:13

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#690 2013-05-05 10:37:20

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#691 2013-05-06 13:31:51

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#692 2013-05-06 13:37:06

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#693 2013-05-07 12:10:52

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#694 2013-05-07 13:24:05

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#695 2013-05-07 13:28:52

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#696 2013-05-07 13:38:28

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#697 2013-05-07 18:40:01

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#698 2013-05-08 13:45:30

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#699 2013-05-08 16:25:38

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

#700 2013-05-08 18:04:44

Re: The linux-pf thread; BFS/CK, TuxOnIce, BFQ, AUFS3

Board footer