linux-kernel - Re: [PATCH 0/3] drm/lima: expose usage statistics via fdinfo

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAKGbVbvaHgJbuh4O8K1CXhYn_PFwBzrkdXzM4YhFAMnuQ0ncUw@mail.gmail.com>
Date:   Sun, 2 Apr 2023 22:27:11 +0800
From:   Qiang Yu <yuq825@...il.com>
To:     dev@...xeye.de
Cc:     Erico Nunes <nunes.erico@...il.com>,
        David Airlie <airlied@...ux.ie>, linux-kernel@...r.kernel.org,
        dri-devel@...ts.freedesktop.org, lima@...ts.freedesktop.org
Subject: Re: [PATCH 0/3] drm/lima: expose usage statistics via fdinfo

> "df622729ddbf drm/scheduler: track GPU active time per entity" had to
> be reverted due to it introducing a use after free. I guess this
> patchset now conflicts with the revert.
>
I do get some build fail message on other branch. Do I need to revert this
patchset on drm-misc-next or left to branch maintainer to decide whether
to pick this patchset upstream?

Regards,
Qiang

>
> > On Mon, Mar 13, 2023 at 11:09 AM Qiang Yu <yuq825@...il.com> wrote:
> > >
> > > Patch set is:
> > > Reviewed-by: Qiang Yu <yuq825@...il.com>
> > >
> > > Looks like drm-misc-next does not contain "df622729ddbf drm/scheduler:
> > > track GPU active time per entity" yet.
> > > Will apply later.
> > >
> > > Regards,
> > > Qiang
> > >
> > > On Mon, Mar 13, 2023 at 7:31 AM Erico Nunes <nunes.erico@...il.com> wrote:
> > > >
> > > > Expose lima gp and pp usage stats through fdinfo, following
> > > > Documentation/gpu/drm-usage-stats.rst.
> > > > Borrowed from these previous implementations:
> > > >
> > > > "df622729ddbf drm/scheduler: track GPU active time per entity" added
> > > > usage time accounting to drm scheduler, which is where the data used
> > > > here comes from.
> > > >
> > > > Then the main implementation is based on these etnaviv commits:
> > > > "d306788b6e1b drm/etnaviv: allocate unique ID per drm_file" and
> > > > "97804a133c68 drm/etnaviv: export client GPU usage statistics via
> > > > fdinfo"
> > > >
> > > > Also "874442541133 drm/amdgpu: Add show_fdinfo() interface" since lima
> > > > has a context manager very similar to amdgpu and all contexts created
> > > > (and released) at the ctx_mgr level need to be accounted for.
> > > >
> > > > Tested with the generic "gputop" tool currently available as patches to
> > > > igt, a sample run with this patchset looks like this:
> > > >
> > > > DRM minor 128
> > > >     PID               NAME             gp                        pp
> > > >     4322   glmark2-es2-way |█████▊                  ||██████████████████      |
> > > >     3561            weston |▎                       ||███▌                    |
> > > >     4159          Xwayland |▏                       ||▉                       |
> > > >     4154          glxgears |▏                       ||▎                       |
> > > >     3661           firefox |▏                       ||▏                       |
> > > >
> > > >
> > > > Erico Nunes (3):
> > > >   drm/lima: add usage counting method to ctx_mgr
> > > >   drm/lima: allocate unique id per drm_file
> > > >   drm/lima: add show_fdinfo for drm usage stats
> > > >
> > > >  drivers/gpu/drm/lima/lima_ctx.c    | 30 ++++++++++++++++++++-
> > > >  drivers/gpu/drm/lima/lima_ctx.h    |  3 +++
> > > >  drivers/gpu/drm/lima/lima_device.h |  3 +++
> > > >  drivers/gpu/drm/lima/lima_drv.c    | 43 +++++++++++++++++++++++++++++-
> > > >  drivers/gpu/drm/lima/lima_drv.h    |  1 +
> > > >  5 files changed, 78 insertions(+), 2 deletions(-)
> > > >
> > > > --
> > > > 2.39.2
> > > >
>