On Thu, Oct 31, 2024 at 12:07:42AM -0500, Lucas De Marchi wrote:
> On Wed, Oct 23, 2024 at 12:07:58AM -0500, Lucas De Marchi wrote:
> > On Tue, Oct 22, 2024 at 11:52:10PM +0200, Peter Zijlstra wrote:
> > > On Fri, Oct 18, 2024 at 02:46:31PM -0500, Lucas De Marchi wrote:
> > >
> > > > I will give t
On Wed, Oct 23, 2024 at 12:07:58AM -0500, Lucas De Marchi wrote:
On Tue, Oct 22, 2024 at 11:52:10PM +0200, Peter Zijlstra wrote:
On Fri, Oct 18, 2024 at 02:46:31PM -0500, Lucas De Marchi wrote:
I will give this a try with i915 and/or xe.
Less horrible version here:
git://git.kernel.org/pub/
On Tue, Oct 22, 2024 at 11:52:10PM +0200, Peter Zijlstra wrote:
On Fri, Oct 18, 2024 at 02:46:31PM -0500, Lucas De Marchi wrote:
I will give this a try with i915 and/or xe.
Less horrible version here:
git://git.kernel.org/pub/scm/linux/kernel/git/peterz/queue.git
perf/pmu-unregister
I've
On Fri, Oct 18, 2024 at 02:46:31PM -0500, Lucas De Marchi wrote:
> I will give this a try with i915 and/or xe.
Less horrible version here:
git://git.kernel.org/pub/scm/linux/kernel/git/peterz/queue.git
perf/pmu-unregister
I've just pushed it out to the robots, but it builds, passes perf test
On Wed, Oct 16, 2024 at 02:03:02PM +0200, Peter Zijlstra wrote:
On Mon, Oct 14, 2024 at 09:25:19PM +0200, Peter Zijlstra wrote:
Let me ponder that a little bit.
So I did the thing on top of the get/put thing that would allow you to
get rid of the ->closed thing, and before I was finished I al
On Mon, Oct 14, 2024 at 09:25:19PM +0200, Peter Zijlstra wrote:
> Let me ponder that a little bit.
So I did the thing on top of the get/put thing that would allow you to
get rid of the ->closed thing, and before I was finished I already hated
all of it :-(
The thing is, if you're going to the ef
On Mon, Oct 14, 2024 at 09:25:19PM +0200, Peter Zijlstra wrote:
On Mon, Oct 14, 2024 at 01:20:34PM -0500, Lucas De Marchi wrote:
On Mon, Oct 14, 2024 at 07:32:46PM +0200, Peter Zijlstra wrote:
> I'm confused.. probably because I still don't have any clue about
> drivers and the above isn't re
On Mon, Oct 14, 2024 at 01:20:34PM -0500, Lucas De Marchi wrote:
> On Mon, Oct 14, 2024 at 07:32:46PM +0200, Peter Zijlstra wrote:
> > I'm confused.. probably because I still don't have any clue about
> > drivers and the above isn't really telling me much either.
> >
> > I don't see how you get r
On Mon, Oct 14, 2024 at 07:32:46PM +0200, Peter Zijlstra wrote:
On Tue, Oct 08, 2024 at 01:34:59PM -0500, Lucas De Marchi wrote:
If a pmu is unregistered while there's an active event, perf will still
access the pmu via event->pmu, even after the event is destroyed. This
makes it difficult for d
On Tue, Oct 08, 2024 at 01:34:59PM -0500, Lucas De Marchi wrote:
> If a pmu is unregistered while there's an active event, perf will still
> access the pmu via event->pmu, even after the event is destroyed. This
> makes it difficult for drivers like i915 that can be unbound from the
> HW.
>
>
el.org/pub/scm/linux/kernel/git/perf/perf-tools-next.git
perf-tools-next
patch link:
https://lore.kernel.org/r/20241008183501.1354695-4-lucas.demarchi%40intel.com
patch subject: [PATCH 3/5] perf: Add pmu get/put
config: x86_64-kexec
(https://download.01.org/0day-ci/archive/20241009/20241009184
If a pmu is unregistered while there's an active event, perf will still
access the pmu via event->pmu, even after the event is destroyed. This
makes it difficult for drivers like i915 that can be unbound from the
HW.
BUG: KASAN: use-after-free in exclusive_event_destroy+0xd8/0xf0
R
12 matches
Mail list logo
