Re: [PATCH] perf tests switch-tracking: Fix timestamp comparison

[Leo Yan <leo.yan@arm.com>]

Hi Ian, Namhyung,

[ - Mailing list ]

On Wed, Apr 02, 2025 at 10:05:16AM +0100, Leo Yan wrote:
> On Tue, Apr 01, 2025 at 12:54:12PM -0700, Ian Rogers wrote:
> > On Tue, Apr 1, 2025 at 2:14 AM Leo Yan <leo.yan@arm.com> wrote:
> > >
> > > On Mon, Mar 31, 2025 at 01:18:31PM -0700, Ian Rogers wrote:
> > >
> > > [...]
> > >
> > > > I'm reminded of a Java check I wrote for this:
> > >
> > > Nice short article.
> > >
> > > > In clang -Wshorten-64-to-32 looks to cover this. I'll see if we can
> > > > clean those warnings up a bit.
> > >
> > > I checked a bit and seems GCC has no this flag, but it makes sense for
> > > me to enable the flag for Clang.
> > >
> > > > Reviewed-by: Ian Rogers <irogers@google.com>
> > >
> > > Thanks a lot, Ian.
> > 
> > I made a small variation to the change in:
> > https://lore.kernel.org/lkml/20250401182347.3422199-10-irogers@google.com/
> > to avoid a subtract and just directly compare the values.

Do you mind to pick up my this patch? :) Our internal CI reports the
test case 109_Track_with_sched_switch failure daily, I am just wandering
if we could apply the fix quickly.

Ian is working on a patch series for resolving the Clang warning which
also includes a fix [1], if Ian could extract the fix for the compar()
function in switch-tracking.c, this either would be fine for me.

Thanks a lot for your helping!

Leo

[1] https://lore.kernel.org/linux-perf-users/20250401182347.3422199-10-irogers@google.com/