From: Peter Zijlstra <peterz@infradead.org>
To: Stephane Eranian <eranian@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>,
Ingo Molnar <mingo@kernel.org>, Jiri Olsa <jolsa@redhat.com>,
LKML <linux-kernel@vger.kernel.org>,
tonyj@suse.com, nelson.dsouza@intel.com
Subject: Re: [PATCH 1/8] perf/x86/intel: Fix memory corruption
Date: Wed, 3 Apr 2019 09:32:43 +0200 [thread overview]
Message-ID: <20190403073243.GA4038@hirez.programming.kicks-ass.net> (raw)
In-Reply-To: <CABPqkBQ=hGNY8nd5WTONaMepH1ma=dBN=AdWOyOF_xe1K5JGHA@mail.gmail.com>
On Fri, Mar 22, 2019 at 12:04:55PM -0700, Stephane Eranian wrote:
> On Thu, Mar 21, 2019 at 10:51 AM Thomas Gleixner <tglx@linutronix.de> wrote:
> >
> > On Thu, 21 Mar 2019, Stephane Eranian wrote:
> > > On Thu, Mar 21, 2019 at 9:45 AM Thomas Gleixner <tglx@linutronix.de> wrote:
> > > >
> > > > On Thu, 21 Mar 2019, Peter Zijlstra wrote:
> > > > > Subject: perf/x86/intel: Initialize TFA MSR
> > > > >
> > > > > Stephane reported that we don't initialize the TFA MSR, which could lead
> > > > > to trouble if the RESET value is not 0 or on kexec.
> > > >
> > > > That sentence doesn't parse.
> > > >
> > > > Stephane reported that the TFA MSR is not initialized by the kernel, but
> > > > the TFA bit could set by firmware or as a leftover from a kexec, which
> > > > makes the state inconsistent.
> > > >
> > > Correct. This is what I meant.
> > > The issue is what does the kernel guarantee when it boots?
> > >
> > > I see:
> > > static bool allow_tsx_force_abort = true;
> > >
> > > Therefore you must ensure the MSR is set to reflect that state on boot.
> > > So you have to force it to that value to be in sync which is what your
> > > new patch is doing.
> >
> > The initial state should be that the MSR TFA bit is 0. The software state
> > is a different beast.
> >
> > allow_tsx_force_abort
> >
> > false Do not set MSR TFA bit (Make TSX work with PMC3) and
> > exclude PMC3 from being used.
> >
> > true Set the MSR TFA bit when PMC3 is used by perf, clear it
> > when PMC3 is not longer in use.
> >
> I would expect this description to be included in the source code where the
> allow_tsx_force_abort variable is defined
That part is easy I suppose.
> and somewhere in the kernel Documentation because it is not trivial to
> understand what the control actually does and the guarantees you have
> when you toggle it.
But here we seem to be sorely lacking, that is, there is no sysfs/perf
documentation at all.
In any case, feel free to send a patch :-)
next prev parent reply other threads:[~2019-04-03 7:32 UTC|newest]
Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-03-14 13:01 [RFC][PATCH 0/8] perf/x86: event scheduling cleanups Peter Zijlstra
2019-03-14 13:01 ` [PATCH 1/8] perf/x86/intel: Fix memory corruption Peter Zijlstra
2019-03-15 11:29 ` [tip:perf/urgent] " tip-bot for Peter Zijlstra
2019-03-19 6:29 ` [PATCH 1/8] " Stephane Eranian
2019-03-19 11:05 ` Peter Zijlstra
2019-03-19 17:52 ` Stephane Eranian
2019-03-19 18:20 ` Peter Zijlstra
2019-03-20 20:47 ` Stephane Eranian
2019-03-20 20:52 ` Stephane Eranian
2019-03-20 22:22 ` Peter Zijlstra
2019-03-21 12:38 ` Peter Zijlstra
2019-03-21 16:45 ` Thomas Gleixner
2019-03-21 17:10 ` Peter Zijlstra
2019-03-21 17:17 ` Thomas Gleixner
2019-03-21 18:20 ` Peter Zijlstra
2019-03-21 19:42 ` Tony Jones
2019-03-21 19:47 ` DSouza, Nelson
2019-03-21 20:07 ` Peter Zijlstra
2019-03-21 23:16 ` DSouza, Nelson
2019-03-22 22:14 ` DSouza, Nelson
2019-03-21 17:23 ` Stephane Eranian
2019-03-21 17:51 ` Thomas Gleixner
2019-03-22 19:04 ` Stephane Eranian
2019-04-03 7:32 ` Peter Zijlstra [this message]
2019-04-03 10:40 ` [tip:perf/urgent] perf/x86/intel: Initialize TFA MSR tip-bot for Peter Zijlstra
2019-04-03 11:30 ` Thomas Gleixner
2019-04-03 12:23 ` Vince Weaver
2019-03-14 13:01 ` [RFC][PATCH 2/8] perf/x86/intel: Simplify intel_tfa_commit_scheduling() Peter Zijlstra
2019-03-14 13:01 ` [RFC][PATCH 3/8] perf/x86: Simplify x86_pmu.get_constraints() interface Peter Zijlstra
2019-03-19 21:21 ` Stephane Eranian
2019-03-14 13:01 ` [RFC][PATCH 4/8] perf/x86: Remove PERF_X86_EVENT_COMMITTED Peter Zijlstra
2019-03-19 20:48 ` Stephane Eranian
2019-03-19 21:00 ` Peter Zijlstra
2019-03-20 13:14 ` Peter Zijlstra
2019-03-20 12:23 ` Peter Zijlstra
2019-03-14 13:01 ` [RFC][PATCH 5/8] perf/x86/intel: Optimize intel_get_excl_constraints() Peter Zijlstra
2019-03-19 23:43 ` Stephane Eranian
2019-03-14 13:01 ` [RFC][PATCH 6/8] perf/x86: Clear ->event_constraint[] on put Peter Zijlstra
2019-03-19 21:50 ` Stephane Eranian
2019-03-20 12:25 ` Peter Zijlstra
2019-03-14 13:01 ` [RFC][PATCH 7/8] perf/x86: Optimize x86_schedule_events() Peter Zijlstra
2019-03-19 23:55 ` Stephane Eranian
2019-03-20 13:11 ` Peter Zijlstra
2019-03-20 19:30 ` Stephane Eranian
2019-03-14 13:01 ` [RFC][PATCH 8/8] perf/x86: Add sanity checks to x86_schedule_events() Peter Zijlstra
2019-03-15 7:15 ` [RFC][PATCH 0/8] perf/x86: event scheduling cleanups Stephane Eranian
2019-03-15 7:15 ` Stephane Eranian
2019-03-15 8:01 ` 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=20190403073243.GA4038@hirez.programming.kicks-ass.net \
--to=peterz@infradead.org \
--cc=eranian@google.com \
--cc=jolsa@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=nelson.dsouza@intel.com \
--cc=tglx@linutronix.de \
--cc=tonyj@suse.com \
/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).