Re: [PATCH] perf test: Test 73 Sig_trap fails on s390

From: Leo Yan
Date: Tue Jan 18 2022 - 06:19:55 EST


Hi John,

On Tue, Jan 18, 2022 at 10:20:37AM +0000, John Garry wrote:

[...]

> > static int test__sigtrap(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
> > {
> > ...
> >
> > if (!BP_SIGNAL_IS_SUPPORTED) {
> > pr_debug("Test not supported on this architecture");
> > return TEST_SKIP;
> > }
> >
> > ...
> > }
> >
> > Since we have defined BP_SIGNAL_IS_SUPPORTED, I think we can reuse it at
> > here.
>
>
> Do you know any other architectures which would have this issue? Or a
> generic way to check for support?
>
> It's better to not have to add to this list arch-by-arch..

Yeah, it's ugly to add archs one by one. But I don't find an ABI can
be used to make decision if an arch supports signal handler for
breakpoint. Usually, it's architecture specific operations for signal
handling, see the code [1]; simply to say, architecture needs to disable
single step when call signal handler and restore single step after
return from signal handler.

[1] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/arch/x86/kernel/signal.c#n830

> > > And fails on my x86 broadwell machine:
> > >
> > > john@localhost:~/kernel-dev2/tools/perf> sudo ./perf test -v 73
> > > 73: Sigtrap :
> > > --- start ---
> > > test child forked, pid 22255
> > > FAILED sys_perf_event_open(): Argument list too long
> > > test child finished with -1
> > > ---- end ----
> > > Sigtrap: FAILED!
> > > john@localhost:~/kernel-dev2/tools/perf>
> > It is a bit suprise for the failure on x86, as I remembered x86 platform
> > can support signal handler with hw breakpoint. And from the error
> > "Argument list too long", it should be a different issue from other
> > archs.
>
> Yeah, I don't know what's going on here.

Seems to there have incompatible issue.
Maybe you could cleanup with "make clean" and then rebuild perf.

Thanks,
Leo