Intel-XE Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Nirmoy Das <nirmoy.das@linux.intel.com>
To: Lucas De Marchi <lucas.demarchi@intel.com>,
	intel-xe@lists.freedesktop.org
Cc: Jonathan Cavitt <jonathan.cavitt@intel.com>,
	Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Subject: Re: [PATCH 3/3] drm/xe: Stop accumulating LRC timestamp on job_free
Date: Mon, 28 Oct 2024 18:23:25 +0100	[thread overview]
Message-ID: <d2ac9d9a-76d2-483e-bdb0-2e81954549d0@linux.intel.com> (raw)
In-Reply-To: <20241026170952.94670-5-lucas.demarchi@intel.com>


On 10/26/2024 7:08 PM, Lucas De Marchi wrote:
> The exec queue timestamp is only really useful when it's being queried
> through the fdinfo. There's no need to update it so often, on every
> job_free. Tracing a simple app like vkcube running shows an update
> rate of ~ 120Hz.
>
> The update on job_free() is used to cover a gap: if exec
> queue is created and destroyed rapidily, before a new query, the
> timestamp still needs to be accumulated and accounted on the xef.
> Initial implementation in commit 6109f24f87d7 ("drm/xe: Add helper to
> accumulate exec queue runtime") couldn't do it on the exec_queue_fini
> since the xef could be gone at that point. However since commit
> ce8c161cbad4 ("drm/xe: Add ref counting for xe_file") the xef is
> refcounted and the exec queue has a reference.
>
> Improve the fix in commit 2149ded63079 ("drm/xe: Fix use after free when
> client stats are captured") by reducing the frequency in which the
> update is needed.
>
> Fixes: 2149ded63079 ("drm/xe: Fix use after free when client stats are captured")
> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>

Reviewed-by: Nirmoy Das <nirmoy.das@intel.com>
> ---
>  drivers/gpu/drm/xe/xe_exec_queue.c | 6 ++++++
>  drivers/gpu/drm/xe/xe_guc_submit.c | 2 --
>  2 files changed, 6 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_exec_queue.c b/drivers/gpu/drm/xe/xe_exec_queue.c
> index b15ca84b2422..bc2fc917e0de 100644
> --- a/drivers/gpu/drm/xe/xe_exec_queue.c
> +++ b/drivers/gpu/drm/xe/xe_exec_queue.c
> @@ -260,8 +260,14 @@ void xe_exec_queue_fini(struct xe_exec_queue *q)
>  {
>  	int i;
>  
> +	/*
> +	 * Before releasing our ref to lrc and xef, accumulate our run ticks
> +	 */
> +	xe_exec_queue_update_run_ticks(q);
> +
>  	for (i = 0; i < q->width; ++i)
>  		xe_lrc_put(q->lrc[i]);
> +
>  	__xe_exec_queue_free(q);
>  }
>  
> diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
> index e5d7c767a744..ebe4665d9159 100644
> --- a/drivers/gpu/drm/xe/xe_guc_submit.c
> +++ b/drivers/gpu/drm/xe/xe_guc_submit.c
> @@ -747,8 +747,6 @@ static void guc_exec_queue_free_job(struct drm_sched_job *drm_job)
>  {
>  	struct xe_sched_job *job = to_xe_sched_job(drm_job);
>  
> -	xe_exec_queue_update_run_ticks(job->q);
> -
>  	trace_xe_sched_job_free(job);
>  	xe_sched_job_put(job);
>  }

  reply	other threads:[~2024-10-28 17:23 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-10-26 17:08 [PATCH 0/3] drm/xe: Fix races on fdinfo Lucas De Marchi
2024-10-26 17:08 ` [PATCH 1/3] drm/xe: Add trace to lrc timestamp update Lucas De Marchi
2024-10-28 14:37   ` Cavitt, Jonathan
2024-10-26 17:08 ` [PATCH 2/3] drm/xe: Accumulate exec queue timestamp on destroy Lucas De Marchi
2024-10-28 14:38   ` Cavitt, Jonathan
2024-10-28 20:33   ` Umesh Nerlige Ramappa
2024-10-28 21:59     ` Matthew Brost
2024-10-28 22:17       ` Cavitt, Jonathan
2024-10-28 23:05       ` Lucas De Marchi
2024-10-28 22:32     ` Lucas De Marchi
2024-10-29 17:27       ` Umesh Nerlige Ramappa
2024-10-29 17:58         ` Lucas De Marchi
2024-10-29 19:03           ` Umesh Nerlige Ramappa
2024-10-29 19:12           ` Lucas De Marchi
2024-10-29 19:31             ` Umesh Nerlige Ramappa
2024-10-29 19:53               ` Lucas De Marchi
2024-10-26 17:08 ` [PATCH 3/3] drm/xe: Stop accumulating LRC timestamp on job_free Lucas De Marchi
2024-10-28 17:23   ` Nirmoy Das [this message]
2024-10-28 20:29   ` Cavitt, Jonathan
2024-10-26 17:15 ` ✓ CI.Patch_applied: success for drm/xe: Fix races on fdinfo Patchwork
2024-10-26 17:16 ` ✗ CI.checkpatch: warning " Patchwork
2024-10-26 17:17 ` ✓ CI.KUnit: success " Patchwork
2024-10-26 17:29 ` ✓ CI.Build: " Patchwork
2024-10-26 17:31 ` ✓ CI.Hooks: " Patchwork
2024-10-26 17:32 ` ✓ CI.checksparse: " Patchwork
2024-10-26 17:57 ` ✓ CI.BAT: " Patchwork
2024-10-27 16:45 ` ✗ CI.FULL: failure " Patchwork
2024-10-29 20:02 ` ✗ CI.Patch_applied: failure for drm/xe: Fix races on fdinfo (rev2) Patchwork

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=d2ac9d9a-76d2-483e-bdb0-2e81954549d0@linux.intel.com \
    --to=nirmoy.das@linux.intel.com \
    --cc=intel-xe@lists.freedesktop.org \
    --cc=jonathan.cavitt@intel.com \
    --cc=lucas.demarchi@intel.com \
    --cc=umesh.nerlige.ramappa@intel.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