From: Peter Zijlstra <peterz@infradead.org>
To: Stephane Eranian <eranian@google.com>
Cc: Ingo Molnar <mingo@kernel.org>,
Vince Weaver <vincent.weaver@maine.edu>,
Jiri Olsa <jolsa@redhat.com>, "Liang, Kan" <kan.liang@intel.com>,
LKML <linux-kernel@vger.kernel.org>,
Andrew Hunter <ahh@google.com>,
Maria Dimakopoulou <maria.n.dimakopoulou@gmail.com>
Subject: Re: [PATCH v2 01/11] perf,x86: Fix event/group validation
Date: Tue, 26 May 2015 12:12:37 +0200 [thread overview]
Message-ID: <20150526101237.GK3644@twins.programming.kicks-ass.net> (raw)
In-Reply-To: <CABPqkBTzQ0z6erv4dnotFKu=KystkSRJG2ovxBhsp5vFFFufqg@mail.gmail.com>
On Tue, May 26, 2015 at 02:24:38AM -0700, Stephane Eranian wrote:
> On Fri, May 22, 2015 at 6:40 AM, Peter Zijlstra <peterz@infradead.org> wrote:
> > On Fri, May 22, 2015 at 03:29:06PM +0200, Peter Zijlstra wrote:
> >> @@ -788,9 +788,9 @@ int x86_schedule_events(struct cpu_hw_ev
> >> x86_pmu.start_scheduling(cpuc);
> >>
> >> for (i = 0, wmin = X86_PMC_IDX_MAX, wmax = 0; i < n; i++) {
> >> + cpuc->event_constraint[i] = NULL;
> >
> > ^^^ that is new, which is esp. important in light of the
> > intel_get_event_constraints() hunk below, which would happily continue
> > life with a garbage constraint.
> >
> You've moved the constraint list from event to cpuc. Yet, it is still
> an array of pointers
> to constraints. So here you are saying, that in the case validate_group() is
> preempted and there is a context switch, there is still a risk of
> overwriting the
> constraint? I don't see how because validate_group() is using a fake_cpuc.
> So yes, the cpuc->event_constraint[] array is modified but it is not the same
> as the actual cpuc used by non-validate code. Or am I still missing something?
>
> When using dynamic constraints, we already have constraint storage in cpuc
> (to avoid calling kmalloc() in ctxsw context). Thus, I am wondering if it would
> not be easier to always use cpuc for constraint storage (no more pointers).
No; the problem here is repeated use of the cpuc (the real one). Say one
scheduling run installs a constraint pointer for event i. Then event i
gets removed and another installed in the same spot.
Then the next scheduling run will pick up the old pointer in
intel_get_event_constraints() as a base for the new one.
next prev parent reply other threads:[~2015-05-26 10:14 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-05-22 13:29 [PATCH v2 00/11] Various x86 pmu scheduling patches Peter Zijlstra
2015-05-22 13:29 ` [PATCH v2 01/11] perf,x86: Fix event/group validation Peter Zijlstra
2015-05-22 13:40 ` Peter Zijlstra
2015-05-26 9:24 ` Stephane Eranian
2015-05-26 10:12 ` Peter Zijlstra [this message]
2015-05-26 11:46 ` Stephane Eranian
2015-05-26 12:16 ` Peter Zijlstra
2015-05-26 12:25 ` Stephane Eranian
2015-05-26 13:22 ` Peter Zijlstra
2015-05-26 13:44 ` Stephane Eranian
2015-05-22 13:29 ` [PATCH v2 02/11] perf/x86: Improve HT workaround GP counter constraint Peter Zijlstra
2015-05-22 13:42 ` Peter Zijlstra
2015-05-26 9:37 ` Stephane Eranian
2015-05-26 10:15 ` Peter Zijlstra
2015-05-26 11:47 ` Stephane Eranian
2015-05-26 13:19 ` Peter Zijlstra
2015-05-26 16:07 ` Peter Zijlstra
2015-05-27 9:01 ` Stephane Eranian
2015-05-27 10:11 ` Peter Zijlstra
2015-05-27 11:39 ` Stephane Eranian
2015-05-27 10:13 ` Peter Zijlstra
2015-05-27 11:44 ` Stephane Eranian
2015-05-26 23:33 ` Andi Kleen
2015-05-27 7:48 ` Peter Zijlstra
2015-05-27 14:00 ` Andi Kleen
2015-05-22 13:29 ` [PATCH v2 03/11] perf/x86: Correct local vs remote sibling state Peter Zijlstra
2015-05-26 11:48 ` Stephane Eranian
2015-05-22 13:29 ` [PATCH v2 04/11] perf/x86: Use lockdep Peter Zijlstra
2015-05-22 13:29 ` [PATCH v2 05/11] perf/x86: Simplify dynamic constraint code somewhat Peter Zijlstra
2015-05-22 13:29 ` [PATCH v2 06/11] perf/x86: Make WARNs consistent Peter Zijlstra
2015-05-22 13:29 ` [PATCH v2 07/11] perf/x86: Move intel_commit_scheduling() Peter Zijlstra
2015-05-22 13:29 ` [PATCH v2 08/11] perf/x86: Remove pointless tests Peter Zijlstra
2015-05-22 13:29 ` [PATCH v2 09/11] perf/x86: Remove intel_excl_states::init_state Peter Zijlstra
2015-05-22 13:29 ` [PATCH v2 10/11] perf,x86: Simplify logic Peter Zijlstra
2015-05-22 13:29 ` [PATCH v2 11/11] perf/x86: Simplify put_exclusive_constraints Peter Zijlstra
2015-05-22 13:38 ` Peter Zijlstra
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20150526101237.GK3644@twins.programming.kicks-ass.net \
--to=peterz@infradead.org \
--cc=ahh@google.com \
--cc=eranian@google.com \
--cc=jolsa@redhat.com \
--cc=kan.liang@intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=maria.n.dimakopoulou@gmail.com \
--cc=mingo@kernel.org \
--cc=vincent.weaver@maine.edu \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).