Re: [PATCH RFC/RFT] perf bpf skels: Stop using vmlinux.h generated from BTF, use subset of used structs + CO-RE. was Re: BPF skels in perf .Re: [GIT PULL] perf tools changes for v6.4

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Fri, May 5, 2023 at 2:52 PM Arnaldo Carvalho de Melo <acme@xxxxxxxxxx> wrote:
>
> Em Fri, May 05, 2023 at 02:21:56PM -0700, Andrii Nakryiko escreveu:
> > On Fri, May 5, 2023 at 2:15 PM Jiri Olsa <olsajiri@xxxxxxxxx> wrote:
> > >
> > > On Fri, May 05, 2023 at 01:46:30PM -0700, Ian Rogers wrote:
> > > > On Fri, May 5, 2023 at 1:43 PM Jiri Olsa <olsajiri@xxxxxxxxx> wrote:
> > > > >
> > > > > On Fri, May 05, 2023 at 10:04:47AM -0700, Ian Rogers wrote:
> > > > > > On Fri, May 5, 2023 at 9:56 AM Arnaldo Carvalho de Melo <acme@xxxxxxxxxx> wrote:
> > > > > > >
> > > > > > > Em Fri, May 05, 2023 at 10:33:15AM -0300, Arnaldo Carvalho de Melo escreveu:
> > > > > > > > Em Fri, May 05, 2023 at 01:03:14AM +0200, Jiri Olsa escreveu:
> > > > > > > > That with the preserve_access_index isn't needed, we need just the
> > > > > > > > fields that we access in the tools, right?
> > > > > > >
> > > > > > > I'm now doing build test this in many distro containers, without the two
> > > > > > > reverts, i.e. BPF skels continue as opt-out as in my pull request, to
> > > > > > > test build and also for the functionality tests on the tools using such
> > > > > > > bpf skels, see below, no touching of vmlinux nor BTF data during the
> > > > > > > build.
> > > > > > >
> > > > > > > - Arnaldo
> > > > > > >
> > > > > > > From 882adaee50bc27f85374aeb2fbaa5b76bef60d05 Mon Sep 17 00:00:00 2001
> > > > > > > From: Arnaldo Carvalho de Melo <acme@xxxxxxxxxx>
> > > > > > > Date: Thu, 4 May 2023 19:03:51 -0300
> > > > > > > Subject: [PATCH 1/1] perf bpf skels: Stop using vmlinux.h generated from BTF,
> > > > > > >  use subset of used structs + CO-RE
> > > > > > >
> > > > > > > Linus reported a build break due to using a vmlinux without a BTF elf
> > > > > > > section to generate the vmlinux.h header with bpftool for use in the BPF
> > > > > > > tools in tools/perf/util/bpf_skel/*.bpf.c.
> > > > > > >
> > > > > > > Instead add a vmlinux.h file with the structs needed with the fields the
> > > > > > > tools need, marking the structs with __attribute__((preserve_access_index)),
> > > > > > > so that libbpf's CO-RE code can fixup the struct field offsets.
> > > > > > >
> > > > > > > In some cases the vmlinux.h file that was being generated by bpftool
> > > > > > > from the kernel BTF information was not needed at all, just including
> > > > > > > linux/bpf.h, sometimes linux/perf_event.h was enough as non-UAPI
> > > > > > > types were not being used.
> > > > > > >
> > > > > > > To keep te patch small, include those UAPI headers from the trimmed down
> > > > > > > vmlinux.h file, that then provides the tools with just the structs and
> > > > > > > the subset of its fields needed for them.
> > > > > > >
> > > > > > > Testing it:
> > > > > > >
> > > > > > >   # perf lock contention -b find / > /dev/null
> > > > >
> > > > > I tested perf lock con -abv -L rcu_state sleep 1
> > > > > and needed fix below
> > > > >
> > > > > jirka
> > > >
> > > > I thought this was fixed by:
> > > > https://lore.kernel.org/lkml/20230427234833.1576130-1-namhyung@xxxxxxxxxx/
> > > > but I think that is just in perf-tools-next.
> > >
> > > ah ok, missed that one
> >
> > Please try validating with veristat to check if all of perf's .bpf.o
> > files are successful. Veristat is part of selftests and can be built
> > with just `make -C tools/testing/selftests/bpf veristat`. After that;
> >
> >  sudo ~/bin/veristat tools/perf/util/bpf_skel/.tmp/*.bpf.o
> >
> > This is a surer way to check that BPF object files are ok at least on
> > your currently running kernel, than trying to exercise each BPF
> > program through perf commands.
>
> [acme@quaco perf-tools]$ sudo tools/testing/selftests/bpf/veristat /tmp/build/perf-tools/util/bpf_skel/.tmp/*.bpf.o
> Processing 'bperf_cgroup.bpf.o'...
> Processing 'bperf_follower.bpf.o'...
> Processing 'bperf_leader.bpf.o'...
> Processing 'bpf_prog_profiler.bpf.o'...
> Processing 'func_latency.bpf.o'...
> Processing 'kwork_trace.bpf.o'...
> Processing 'lock_contention.bpf.o'...
> Processing 'off_cpu.bpf.o'...
> Processing 'sample_filter.bpf.o'...
> File                     Program                          Verdict  Duration (us)   Insns  States  Peak states
> -----------------------  -------------------------------  -------  -------------  ------  ------  -----------
> bperf_cgroup.bpf.o       on_cgrp_switch                   success           6479   17025     417          174
> bperf_cgroup.bpf.o       trigger_read                     success           6370   17025     417          174
> bperf_follower.bpf.o     fexit_XXX                        failure              0       0       0            0
> bperf_leader.bpf.o       on_switch                        success            360      49       3            3
> bpf_prog_profiler.bpf.o  fentry_XXX                       failure              0       0       0            0
> bpf_prog_profiler.bpf.o  fexit_XXX                        failure              0       0       0            0
> func_latency.bpf.o       func_begin                       success            351      69       6            6
> func_latency.bpf.o       func_end                         success            318     158      15           15
> kwork_trace.bpf.o        latency_softirq_entry            success            334     108      10           10
> kwork_trace.bpf.o        latency_softirq_raise            success            896    1993      34           34
> kwork_trace.bpf.o        latency_workqueue_activate_work  success            333      46       4            4
> kwork_trace.bpf.o        latency_workqueue_execute_start  success           1112    2219      41           41
> kwork_trace.bpf.o        report_irq_handler_entry         success           1067    2118      34           34
> kwork_trace.bpf.o        report_irq_handler_exit          success            334     110      10           10
> kwork_trace.bpf.o        report_softirq_entry             success            897    1993      34           34
> kwork_trace.bpf.o        report_softirq_exit              success            329     108      10           10
> kwork_trace.bpf.o        report_workqueue_execute_end     success           1124    2219      41           41
> kwork_trace.bpf.o        report_workqueue_execute_start   success            295      46       4            4
> lock_contention.bpf.o    collect_lock_syms                failure              0       0       0            0
> lock_contention.bpf.o    contention_begin                 failure              0       0       0            0
> lock_contention.bpf.o    contention_end                   failure              0       0       0            0
> off_cpu.bpf.o            on_newtask                       success            387      37       3            3
> off_cpu.bpf.o            on_switch                        success            536     220      20           20
> sample_filter.bpf.o      perf_sample_filter               success         190443  190237   11173          923
> -----------------------  -------------------------------  -------  -------------  ------  ------  -----------
> Done. Processed 9 files, 0 programs. Skipped 24 files, 0 programs.
> [acme@quaco perf-tools]$
>
> What extra info can we get from these "failure" lines?

you can ask for verifier log with -v (or very detailed verifier log
with -vl2). And you can narrow down to single program at a time with
-f <prog_name>:

sudo ./veristat -v <path-to-bpf.o> -f <prog_name>

Not every possible program can be correctly loaded by veristat (e.g.,
if fentry/fexit doesn't specify target function). But in the above
case everything from lock_contention.bpf.o is definitely worth
checking.

>
> - Arnaldo





[Index of Archives]     [Linux Samsung SoC]     [Linux Rockchip SoC]     [Linux Actions SoC]     [Linux for Synopsys ARC Processors]     [Linux NFS]     [Linux NILFS]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]


  Powered by Linux