Intel-GFX Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Andi Shyti <andi.shyti@linux.intel.com>
To: Nirmoy Das <nirmoy.das@intel.com>
Cc: intel-gfx@lists.freedesktop.org, dri-devel@lists.freedesktop.org,
	John.C.Harrison@intel.com
Subject: Re: [PATCH v2 2/2] drm/i915: Fix gt reset with GuC submission is disabled
Date: Tue, 23 Apr 2024 11:32:00 +0200	[thread overview]
Message-ID: <ZieAEGnCE88TY0GM@ashyti-mobl2.lan> (raw)
In-Reply-To: <20240422201951.633-2-nirmoy.das@intel.com>

Hi Nirmoy,

On Mon, Apr 22, 2024 at 10:19:51PM +0200, Nirmoy Das wrote:
> Currently intel_gt_reset() kills the GuC and then resets requested
> engines. This is problematic because there is a dedicated CSB FIFO
> which only GuC can access and if that FIFO fills up, the hardware
> will block on the next context switch until there is space that means
> the system is effectively hung. If an engine is reset whilst actively
> executing a context, a CSB entry will be sent to say that the context
> has gone idle. Thus if reset happens on a very busy system then
> killing GuC before killing the engines will lead to deadlock because
> of filled up CSB FIFO.

is this a fix?

> To address this issue, the GuC should be killed only after resetting
> the requested engines and before calling intel_gt_init_hw().
> 
> v2: Improve commit message(John)
> 
> Cc: John Harrison <john.c.harrison@intel.com>
> Signed-off-by: Nirmoy Das <nirmoy.das@intel.com>
> ---
>  drivers/gpu/drm/i915/gt/intel_reset.c | 16 ++++++++++++++--
>  1 file changed, 14 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/gpu/drm/i915/gt/intel_reset.c b/drivers/gpu/drm/i915/gt/intel_reset.c
> index b1393863ca9b..6161f7a3ff70 100644
> --- a/drivers/gpu/drm/i915/gt/intel_reset.c
> +++ b/drivers/gpu/drm/i915/gt/intel_reset.c
> @@ -879,8 +879,17 @@ static intel_engine_mask_t reset_prepare(struct intel_gt *gt)
>  	intel_engine_mask_t awake = 0;
>  	enum intel_engine_id id;
>  
> -	/* For GuC mode, ensure submission is disabled before stopping ring */
> -	intel_uc_reset_prepare(&gt->uc);
> +	/**
> +	 * For GuC mode with submission enabled, ensure submission
> +	 * is disabled before stopping ring.

nit: "stopping *the* ring"

> +	 *
> +	 * For GuC mode with submission disabled, ensure that GuC is not
> +	 * sanitized, do that after engine reset. reset_prepare()
> +	 * is followed by engine reset which in this mode requires GuC to
> +	 * process any CSB FIFO entries generated by the resets.
> +	 */
> +	if (intel_uc_uses_guc_submission(&gt->uc))
> +		intel_uc_reset_prepare(&gt->uc);
>  
>  	for_each_engine(engine, gt, id) {
>  		if (intel_engine_pm_get_if_awake(engine))
> @@ -1227,6 +1236,9 @@ void intel_gt_reset(struct intel_gt *gt,
>  
>  	intel_overlay_reset(gt->i915);
>  
> +	/* sanitize uC after engine reset */
> +	if (!intel_uc_uses_guc_submission(&gt->uc))
> +		intel_uc_reset_prepare(&gt->uc);

Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>

Thanks,
Andi

  parent reply	other threads:[~2024-04-23  9:32 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-04-22 20:19 [PATCH v2 1/2] drm/i915: Refactor confusing __intel_gt_reset() Nirmoy Das
2024-04-22 20:19 ` [PATCH v2 2/2] drm/i915: Fix gt reset with GuC submission is disabled Nirmoy Das
2024-04-23  0:09   ` John Harrison
2024-04-23  9:32   ` Andi Shyti [this message]
2024-04-23 10:45     ` Nirmoy Das
2024-04-23 14:42       ` Andi Shyti
2024-04-22 21:14 ` ✗ Fi.CI.SPARSE: warning for series starting with [v2,1/2] drm/i915: Refactor confusing __intel_gt_reset() Patchwork
2024-04-22 21:20 ` ✓ Fi.CI.BAT: success " Patchwork
2024-04-23  0:07 ` [PATCH v2 1/2] " John Harrison
2024-04-23  9:27 ` Andi Shyti
2024-04-23  9:49 ` ✗ Fi.CI.SPARSE: warning for series starting with [v2,1/2] drm/i915: Refactor confusing __intel_gt_reset() (rev2) Patchwork
2024-04-23  9:57 ` ✓ Fi.CI.BAT: success " Patchwork
2024-04-23 10:28 ` ✗ Fi.CI.IGT: failure for series starting with [v2,1/2] drm/i915: Refactor confusing __intel_gt_reset() Patchwork
2024-04-24  8:16 ` ✗ Fi.CI.IGT: failure for series starting with [v2,1/2] drm/i915: Refactor confusing __intel_gt_reset() (rev2) Patchwork
2024-04-24  8:56   ` Nirmoy Das
2024-04-24 17:06     ` Andi Shyti
2024-04-24 20:51       ` Das, Nirmoy

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=ZieAEGnCE88TY0GM@ashyti-mobl2.lan \
    --to=andi.shyti@linux.intel.com \
    --cc=John.C.Harrison@intel.com \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=intel-gfx@lists.freedesktop.org \
    --cc=nirmoy.das@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