[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20220511200657.GB76023@worktop.programming.kicks-ass.net>
Date: Wed, 11 May 2022 22:06:57 +0200
From: Peter Zijlstra <peterz@...radead.org>
To: "Liang, Kan" <kan.liang@...ux.intel.com>
Cc: x86@...nel.org, eranian@...gle.com, linux-kernel@...r.kernel.org,
acme@...nel.org, mark.rutland@....com,
alexander.shishkin@...ux.intel.com, jolsa@...nel.org,
namhyung@...nel.org
Subject: Re: [PATCH 4/5] perf/x86: Change x86_pmu::limit_period signature
On Wed, May 11, 2022 at 01:47:06PM -0400, Liang, Kan wrote:
> > @@ -1386,19 +1387,14 @@ int x86_perf_event_set_period(struct per
> > hwc->last_period = period;
> > ret = 1;
> > }
> > - /*
> > - * Quirk: certain CPUs dont like it if just 1 hw_event is left:
> > - */
> > - if (unlikely(left < 2))
> > - left = 2;
> >
>
> Is the quirk accidentally deleted?
> We should still need the quirk for certain CPUs.
No, but I did forget to write about it in the Changelog :/
IIRC it was Nehalem that triggered that and that should now be covered
by nhm_limit_period().
Or are you aware of more machines that need this?
Anyway, perhaps this should be its own separate commit.
Powered by blists - more mailing lists