linux-perf-users.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* Re: [acme:tmp.perf-tools-next] [perf build] a980755beb: perf-sanity-tests.perf.make.fail
       [not found] <202303222158.8975b36d-oliver.sang@intel.com>
@ 2023-03-22 16:03 ` Ian Rogers
  2023-03-29  6:17   ` Oliver Sang
  0 siblings, 1 reply; 4+ messages in thread
From: Ian Rogers @ 2023-03-22 16:03 UTC (permalink / raw)
  To: kernel test robot
  Cc: oe-lkp, lkp, linux-perf-users, Arnaldo Carvalho de Melo,
	Adrian Hunter, Alexander Shishkin, Andres Freund, Ingo Molnar,
	Jiri Olsa, Leo Yan, Mark Rutland, Martin Liška,
	Namhyung Kim, Nathan Chancellor, Nick Desaulniers,
	Pavithra Gurushankar, Peter Zijlstra, Quentin Monnet,
	Roberto Sassu, Stephane Eranian, Tiezhu Yang, Tom Rix,
	Yang Jihong, linux-kernel

On Wed, Mar 22, 2023 at 7:20 AM kernel test robot <oliver.sang@intel.com> wrote:
>
>
> Greeting,
>
> FYI, we noticed perf-sanity-tests.perf.make.fail due to commit (built with gcc-11):
>
> commit: a980755beb5aca9002e1c95ba519b83a44242b5b ("perf build: Make BUILD_BPF_SKEL default, rename to NO_BPF_SKEL")
> https://git.kernel.org/cgit/linux/kernel/git/acme/linux.git tmp.perf-tools-next
>
> in testcase: perf-sanity-tests
> version: perf-x86_64-e8d018dd0257-1_20230320
> with following parameters:
>
>         perf_compiler: clang
>
>
>
> on test machine: 4 threads Intel(R) Core(TM) i5-6500 CPU @ 3.20GHz (Skylake) with 32G memory
>
> caused below changes (please refer to attached dmesg/kmsg for entire log/backtrace):
>
>
>
>
> If you fix the issue, kindly add following tag
> | Reported-by: kernel test robot <oliver.sang@intel.com>
> | Link: https://lore.kernel.org/oe-lkp/202303222158.8975b36d-oliver.sang@intel.com
>
>
> [   89.510180][  T246]   LINK    /usr/src/perf_selftests-x86_64-rhel-8.3-func-a980755beb5aca9002e1c95ba519b83a44242b5b/tools/perf/util/bpf_skel/.tmp/boo
> tstrap/bpftool
> [   89.510191][  T246]
> [   89.801665][  T246]   GEN     /usr/src/perf_selftests-x86_64-rhel-8.3-func-a980755beb5aca9002e1c95ba519b83a44242b5b/tools/perf/util/bpf_skel/vmlinux.
> h
> [   89.801677][  T246]
> [   91.387572][  T248] libbpf: failed to find '.BTF' ELF section in /usr/src/perf_selftests-x86_64-rhel-8.3-func-a980755beb5aca9002e1c95ba519b83a44242b5
> b/tools/perf/util/bpf_skel/.tmp/bootstrap/bpftool
> [   91.387584][  T248]
> [   91.601268][  T248] Error: failed to load BTF from /usr/src/perf_selftests-x86_64-rhel-8.3-func-a980755beb5aca9002e1c95ba519b83a44242b5b/tools/perf/u
> til/bpf_skel/.tmp/bootstrap/bpftool: No data available
> [   91.601281][  T248]
> [   91.626292][  T248] make[2]: *** [Makefile.perf:1075: /usr/src/perf_selftests-x86_64-rhel-8.3-func-a980755beb5aca9002e1c95ba519b83a44242b5b/tools/per
> f/util/bpf_skel/vmlinux.h] Error 195
> [   91.626303][  T248]
> [   91.648913][  T248] make[2]: *** Deleting file '/usr/src/perf_selftests-x86_64-rhel-8.3-func-a980755beb5aca9002e1c95ba519b83a44242b5b/tools/perf/util
> /bpf_skel/vmlinux.h'
> [   91.648922][  T248]
> [   91.667624][  T248] make[2]: *** Waiting for unfinished jobs....
> [   91.667631][  T248]
> [   98.262749][  T248] make[1]: *** [Makefile.perf:236: sub-make] Error 2
> [   98.262761][  T248]
> [   98.266203][  T246] make: Leaving directory '/usr/src/perf_selftests-x86_64-rhel-8.3-func-a980755beb5aca9002e1c95ba519b83a44242b5b/tools/perf'
> [   98.270224][  T248] make: *** [Makefile:70: all] Error 2
> [   98.272060][  T246]
> [   98.284492][  T248]
> [   98.290078][  T246] make perf failed

Hi Oliver,

Could we add NO_BPF_SKEL=1 to the build flags? The idea with changing
the default is that is what most users want and so we want the BPF
skeleton to be opt-out rather than opt-in.

Thanks,
Ian

>
> To reproduce:
>
>         git clone https://github.com/intel/lkp-tests.git
>         cd lkp-tests
>         sudo bin/lkp install job.yaml           # job file is attached in this email
>         bin/lkp split-job --compatible job.yaml # generate the yaml file for lkp run
>         sudo bin/lkp run generated-yaml-file
>
>         # if come across any failure that blocks the test,
>         # please remove ~/.lkp and /lkp dir to run from a clean state.
>
>
>
> --
> 0-DAY CI Kernel Test Service
> https://github.com/intel/lkp-tests
>
>

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [acme:tmp.perf-tools-next] [perf build] a980755beb: perf-sanity-tests.perf.make.fail
  2023-03-22 16:03 ` [acme:tmp.perf-tools-next] [perf build] a980755beb: perf-sanity-tests.perf.make.fail Ian Rogers
@ 2023-03-29  6:17   ` Oliver Sang
  2023-03-29 16:54     ` Ian Rogers
  0 siblings, 1 reply; 4+ messages in thread
From: Oliver Sang @ 2023-03-29  6:17 UTC (permalink / raw)
  To: Ian Rogers
  Cc: oe-lkp, lkp, linux-perf-users, Arnaldo Carvalho de Melo,
	Adrian Hunter, Alexander Shishkin, Andres Freund, Ingo Molnar,
	Jiri Olsa, Leo Yan, Mark Rutland, Martin Liška,
	Namhyung Kim, Nathan Chancellor, Nick Desaulniers,
	Pavithra Gurushankar, Peter Zijlstra, Quentin Monnet,
	Roberto Sassu, Stephane Eranian, Tiezhu Yang, Tom Rix,
	Yang Jihong, linux-kernel

hi Ian,

On Wed, Mar 22, 2023 at 09:03:28AM -0700, Ian Rogers wrote:
> On Wed, Mar 22, 2023 at 7:20 AM kernel test robot <oliver.sang@intel.com> wrote:
> >
> >
> > Greeting,
> >
> > FYI, we noticed perf-sanity-tests.perf.make.fail due to commit (built with gcc-11):
> >
> > commit: a980755beb5aca9002e1c95ba519b83a44242b5b ("perf build: Make BUILD_BPF_SKEL default, rename to NO_BPF_SKEL")
> > https://git.kernel.org/cgit/linux/kernel/git/acme/linux.git tmp.perf-tools-next
> >
> > in testcase: perf-sanity-tests
> > version: perf-x86_64-e8d018dd0257-1_20230320
> > with following parameters:
> >
> >         perf_compiler: clang
> >
> > [   98.290078][  T246] make perf failed
> 
> Hi Oliver,
> 
> Could we add NO_BPF_SKEL=1 to the build flags? The idea with changing
> the default is that is what most users want and so we want the BPF
> skeleton to be opt-out rather than opt-in.

Thanks a lot for guidance!

I added NO_BPF_SKEL=1 then make perf can pass upon this commit.
I also tried to add BUILD_BPF_SKEL=1 while make perf for parent, it will fail.

since you said in commit message
"BPF skeleton support is now key to a number of perf features."
we will start to fix make issue.

Thanks!

> 
> Thanks,
> Ian
> 
> >
> > To reproduce:
> >
> >         git clone https://github.com/intel/lkp-tests.git
> >         cd lkp-tests
> >         sudo bin/lkp install job.yaml           # job file is attached in this email
> >         bin/lkp split-job --compatible job.yaml # generate the yaml file for lkp run
> >         sudo bin/lkp run generated-yaml-file
> >
> >         # if come across any failure that blocks the test,
> >         # please remove ~/.lkp and /lkp dir to run from a clean state.
> >
> >
> >
> > --
> > 0-DAY CI Kernel Test Service
> > https://github.com/intel/lkp-tests
> >
> >
> 

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [acme:tmp.perf-tools-next] [perf build] a980755beb: perf-sanity-tests.perf.make.fail
  2023-03-29  6:17   ` Oliver Sang
@ 2023-03-29 16:54     ` Ian Rogers
  2023-03-29 17:44       ` Arnaldo Carvalho de Melo
  0 siblings, 1 reply; 4+ messages in thread
From: Ian Rogers @ 2023-03-29 16:54 UTC (permalink / raw)
  To: Oliver Sang
  Cc: oe-lkp, lkp, linux-perf-users, Arnaldo Carvalho de Melo,
	Adrian Hunter, Alexander Shishkin, Andres Freund, Ingo Molnar,
	Jiri Olsa, Leo Yan, Mark Rutland, Martin Liška,
	Namhyung Kim, Nathan Chancellor, Nick Desaulniers,
	Pavithra Gurushankar, Peter Zijlstra, Quentin Monnet,
	Roberto Sassu, Stephane Eranian, Tiezhu Yang, Tom Rix,
	Yang Jihong, linux-kernel

On Tue, Mar 28, 2023 at 11:17 PM Oliver Sang <oliver.sang@intel.com> wrote:
>
> hi Ian,
>
> On Wed, Mar 22, 2023 at 09:03:28AM -0700, Ian Rogers wrote:
> > On Wed, Mar 22, 2023 at 7:20 AM kernel test robot <oliver.sang@intel.com> wrote:
> > >
> > >
> > > Greeting,
> > >
> > > FYI, we noticed perf-sanity-tests.perf.make.fail due to commit (built with gcc-11):
> > >
> > > commit: a980755beb5aca9002e1c95ba519b83a44242b5b ("perf build: Make BUILD_BPF_SKEL default, rename to NO_BPF_SKEL")
> > > https://git.kernel.org/cgit/linux/kernel/git/acme/linux.git tmp.perf-tools-next
> > >
> > > in testcase: perf-sanity-tests
> > > version: perf-x86_64-e8d018dd0257-1_20230320
> > > with following parameters:
> > >
> > >         perf_compiler: clang
> > >
> > > [   98.290078][  T246] make perf failed
> >
> > Hi Oliver,
> >
> > Could we add NO_BPF_SKEL=1 to the build flags? The idea with changing
> > the default is that is what most users want and so we want the BPF
> > skeleton to be opt-out rather than opt-in.
>
> Thanks a lot for guidance!
>
> I added NO_BPF_SKEL=1 then make perf can pass upon this commit.
> I also tried to add BUILD_BPF_SKEL=1 while make perf for parent, it will fail.
>
> since you said in commit message
> "BPF skeleton support is now key to a number of perf features."
> we will start to fix make issue.
>
> Thanks!

This is great Oliver, many thanks!
Ian

> >
> > Thanks,
> > Ian
> >
> > >
> > > To reproduce:
> > >
> > >         git clone https://github.com/intel/lkp-tests.git
> > >         cd lkp-tests
> > >         sudo bin/lkp install job.yaml           # job file is attached in this email
> > >         bin/lkp split-job --compatible job.yaml # generate the yaml file for lkp run
> > >         sudo bin/lkp run generated-yaml-file
> > >
> > >         # if come across any failure that blocks the test,
> > >         # please remove ~/.lkp and /lkp dir to run from a clean state.
> > >
> > >
> > >
> > > --
> > > 0-DAY CI Kernel Test Service
> > > https://github.com/intel/lkp-tests
> > >
> > >
> >

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [acme:tmp.perf-tools-next] [perf build] a980755beb: perf-sanity-tests.perf.make.fail
  2023-03-29 16:54     ` Ian Rogers
@ 2023-03-29 17:44       ` Arnaldo Carvalho de Melo
  0 siblings, 0 replies; 4+ messages in thread
From: Arnaldo Carvalho de Melo @ 2023-03-29 17:44 UTC (permalink / raw)
  To: Ian Rogers, Oliver Sang
  Cc: oe-lkp, lkp, linux-perf-users, Arnaldo Carvalho de Melo,
	Adrian Hunter, Alexander Shishkin, Andres Freund, Ingo Molnar,
	Jiri Olsa, Leo Yan, Mark Rutland, Martin Liška,
	Namhyung Kim, Nathan Chancellor, Nick Desaulniers,
	Pavithra Gurushankar, Peter Zijlstra, Quentin Monnet,
	Roberto Sassu, Stephane Eranian, Tiezhu Yang, Tom Rix,
	Yang Jihong, linux-kernel



On March 29, 2023 1:54:39 PM GMT-03:00, Ian Rogers <irogers@google.com> wrote:
>On Tue, Mar 28, 2023 at 11:17 PM Oliver Sang <oliver.sang@intel.com> wrote:
>>
>> hi Ian,
>>
>> On Wed, Mar 22, 2023 at 09:03:28AM -0700, Ian Rogers wrote:
>> > On Wed, Mar 22, 2023 at 7:20 AM kernel test robot <oliver.sang@intel.com> wrote:
>> > >
>> > >
>> > > Greeting,
>> > >
>> > > FYI, we noticed perf-sanity-tests.perf.make.fail due to commit (built with gcc-11):
>> > >
>> > > commit: a980755beb5aca9002e1c95ba519b83a44242b5b ("perf build: Make BUILD_BPF_SKEL default, rename to NO_BPF_SKEL")
>> > > https://git.kernel.org/cgit/linux/kernel/git/acme/linux.git tmp.perf-tools-next
>> > >
>> > > in testcase: perf-sanity-tests
>> > > version: perf-x86_64-e8d018dd0257-1_20230320
>> > > with following parameters:
>> > >
>> > >         perf_compiler: clang
>> > >
>> > > [   98.290078][  T246] make perf failed
>> >
>> > Hi Oliver,
>> >
>> > Could we add NO_BPF_SKEL=1 to the build flags? The idea with changing
>> > the default is that is what most users want and so we want the BPF
>> > skeleton to be opt-out rather than opt-in.
>>
>> Thanks a lot for guidance!
>>
>> I added NO_BPF_SKEL=1 then make perf can pass upon this commit.
>> I also tried to add BUILD_BPF_SKEL=1 while make perf for parent, it will fail.
>>
>> since you said in commit message
>> "BPF skeleton support is now key to a number of perf features."
>> we will start to fix make issue.


So, does that mean that you guys will not use NO_BPF_SKEL=1 and instead will try to have what is needed to build with the default: with BPF skel and the features enabled by them?

If that is the case: great!

- Arnaldo 

>>
>> Thanks!
>
>This is great Oliver, many thanks!
>Ian
>
>> >
>> > Thanks,
>> > Ian
>> >
>> > >
>> > > To reproduce:
>> > >
>> > >         git clone https://github.com/intel/lkp-tests.git
>> > >         cd lkp-tests
>> > >         sudo bin/lkp install job.yaml           # job file is attached in this email
>> > >         bin/lkp split-job --compatible job.yaml # generate the yaml file for lkp run
>> > >         sudo bin/lkp run generated-yaml-file
>> > >
>> > >         # if come across any failure that blocks the test,
>> > >         # please remove ~/.lkp and /lkp dir to run from a clean state.
>> > >
>> > >
>> > >
>> > > --
>> > > 0-DAY CI Kernel Test Service
>> > > https://github.com/intel/lkp-tests
>> > >
>> > >
>> >

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2023-03-29 17:44 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
     [not found] <202303222158.8975b36d-oliver.sang@intel.com>
2023-03-22 16:03 ` [acme:tmp.perf-tools-next] [perf build] a980755beb: perf-sanity-tests.perf.make.fail Ian Rogers
2023-03-29  6:17   ` Oliver Sang
2023-03-29 16:54     ` Ian Rogers
2023-03-29 17:44       ` Arnaldo Carvalho de Melo

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).