Intel-XE Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Rodrigo Vivi <rodrigo.vivi@intel.com>
To: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Cc: <intel-xe@lists.freedesktop.org>, <kernel-dev@igalia.com>
Subject: Re: [PATCH v12 02/13] drm/xe/xelp: Quiesce memory traffic before invalidating AuxCCS
Date: Wed, 1 Oct 2025 11:47:27 -0400	[thread overview]
Message-ID: <aN1NDy4Mfcs3A-eL@intel.com> (raw)
In-Reply-To: <20250923100812.88257-3-tvrtko.ursulin@igalia.com>

On Tue, Sep 23, 2025 at 11:07:55AM +0100, Tvrtko Ursulin wrote:
> According to i915 commit
> ad8ebf12217e ("drm/i915/gt: Ensure memory quiesced before invalidation")
> quiescing of the memory traffic is required before invalidating the AuxCCS
> tables.
> 
> Add an extra pipe control flush to achieve that.
> 
> Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>

Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>

> ---
>  drivers/gpu/drm/xe/xe_ring_ops.c       | 10 +++++++++-
>  drivers/gpu/drm/xe/xe_ring_ops_types.h |  2 +-
>  2 files changed, 10 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/gpu/drm/xe/xe_ring_ops.c b/drivers/gpu/drm/xe/xe_ring_ops.c
> index 25259b2784bc..af0141c6f809 100644
> --- a/drivers/gpu/drm/xe/xe_ring_ops.c
> +++ b/drivers/gpu/drm/xe/xe_ring_ops.c
> @@ -358,10 +358,18 @@ static void __emit_job_gen12_render_compute(struct xe_sched_job *job,
>  	struct xe_gt *gt = job->q->gt;
>  	struct xe_device *xe = gt_to_xe(gt);
>  	bool lacks_render = !(gt->info.engine_mask & XE_HW_ENGINE_RCS_MASK);
> +	const bool aux_ccs = has_aux_ccs(xe);
>  	u32 mask_flags = 0;
>  
>  	i = emit_copy_timestamp(lrc, dw, i);
>  
> +	/*
> +	 * On AuxCCS platforms the invalidation of the Aux table requires
> +	 * quiescing the memory traffic beforehand.
> +	 */
> +	if (aux_ccs)
> +		i = emit_render_cache_flush(job, dw, i);
> +
>  	dw[i++] = preparser_disable(true);
>  	if (lacks_render)
>  		mask_flags = PIPE_CONTROL_3D_ARCH_FLAGS;
> @@ -372,7 +380,7 @@ static void __emit_job_gen12_render_compute(struct xe_sched_job *job,
>  	i = emit_pipe_invalidate(mask_flags, job->ring_ops_flush_tlb, dw, i);
>  
>  	/* hsdes: 1809175790 */
> -	if (has_aux_ccs(xe))
> +	if (aux_ccs)
>  		i = emit_aux_table_inv(gt, CCS_AUX_INV, dw, i);
>  
>  	dw[i++] = preparser_disable(false);
> diff --git a/drivers/gpu/drm/xe/xe_ring_ops_types.h b/drivers/gpu/drm/xe/xe_ring_ops_types.h
> index d7e3e150a9a5..477dc7defd72 100644
> --- a/drivers/gpu/drm/xe/xe_ring_ops_types.h
> +++ b/drivers/gpu/drm/xe/xe_ring_ops_types.h
> @@ -8,7 +8,7 @@
>  
>  struct xe_sched_job;
>  
> -#define MAX_JOB_SIZE_DW 58
> +#define MAX_JOB_SIZE_DW 70
>  #define MAX_JOB_SIZE_BYTES (MAX_JOB_SIZE_DW * 4)
>  
>  /**
> -- 
> 2.48.0
> 

  reply	other threads:[~2025-10-01 15:47 UTC|newest]

Thread overview: 45+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-09-23 10:07 [PATCH v12 00/13] AuxCCS handling and render compression modifiers Tvrtko Ursulin
2025-09-23 10:07 ` [PATCH v12 01/13] drm/xe/xelpg: Flush CCS when flushing caches Tvrtko Ursulin
2025-09-23 10:07 ` [PATCH v12 02/13] drm/xe/xelp: Quiesce memory traffic before invalidating AuxCCS Tvrtko Ursulin
2025-10-01 15:47   ` Rodrigo Vivi [this message]
2025-09-23 10:07 ` [PATCH v12 03/13] drm/xe/xelp: Support auxccs invalidation on blitter Tvrtko Ursulin
2025-09-23 10:07 ` [PATCH v12 04/13] drm/xe/xelp: Use MI_FLUSH_DW_CCS on auxccs platforms Tvrtko Ursulin
2025-09-23 10:07 ` [PATCH v12 05/13] drm/xe/xelp: Wait for AuxCCS invalidation to complete Tvrtko Ursulin
2025-09-23 10:07 ` [PATCH v12 06/13] drm/xe: Export xe_emit_aux_table_inv Tvrtko Ursulin
2025-09-23 10:08 ` [PATCH v12 07/13] drm/xe/xelp: Add AuxCCS invalidation to the indirect context workarounds Tvrtko Ursulin
2025-09-23 10:08 ` [PATCH v12 08/13] drm/xe: Flush GGTT writes after populating DPT Tvrtko Ursulin
2025-09-23 10:08 ` [PATCH v12 09/13] drm/xe: Handle DPT in system memory Tvrtko Ursulin
2025-09-23 10:08 ` [PATCH v12 10/13] drm/xe/display: Add support for AuxCCS Tvrtko Ursulin
2025-09-23 10:08 ` [PATCH v12 11/13] drm/xe: Force flush system memory AuxCCS framebuffers before scan out Tvrtko Ursulin
2025-09-23 10:19   ` Ville Syrjälä
2025-09-23 10:48     ` Tvrtko Ursulin
2025-09-23 12:01       ` Ville Syrjälä
2025-09-23 12:25         ` Tvrtko Ursulin
2025-09-23 13:20           ` Ville Syrjälä
2025-09-23 14:40             ` Tvrtko Ursulin
2025-09-23 14:52               ` Ville Syrjälä
2025-09-24 13:09                 ` Tvrtko Ursulin
2025-09-24 22:35                   ` Ville Syrjälä
2025-09-25  7:24                     ` Tvrtko Ursulin
2025-09-25 10:08                       ` Tvrtko Ursulin
2025-09-26  7:41                         ` Ville Syrjälä
2025-09-26 19:35                           ` Ville Syrjälä
2025-10-02 14:01                             ` Tvrtko Ursulin
2025-10-02 14:36                               ` Tvrtko Ursulin
2025-10-02 16:23                               ` Ville Syrjälä
2025-10-02 17:04                                 ` Tvrtko Ursulin
2025-10-02 17:14                                   ` Ville Syrjälä
2025-10-02 22:02                                     ` Ville Syrjälä
2025-09-23 10:44   ` [PATCH v13 " Tvrtko Ursulin
2025-09-23 10:08 ` [PATCH v12 12/13] drm/xe: Do not use stolen memory for DPT on IGFX and AuxCCS Tvrtko Ursulin
2025-09-23 10:08 ` [PATCH v12 13/13] drm/i915/display: Expose AuxCCS frame buffer modifiers for Xe Tvrtko Ursulin
2025-09-23 10:15 ` ✗ CI.checkpatch: warning for AuxCCS handling and render compression modifiers (rev15) Patchwork
2025-09-23 10:16 ` ✓ CI.KUnit: success " Patchwork
2025-09-23 11:15 ` ✓ Xe.CI.BAT: " Patchwork
2025-09-23 11:21 ` ✗ CI.checkpatch: warning for AuxCCS handling and render compression modifiers (rev16) Patchwork
2025-09-23 11:22 ` ✓ CI.KUnit: success " Patchwork
2025-09-23 12:03 ` ✓ Xe.CI.BAT: " Patchwork
2025-09-23 13:26 ` ✗ Xe.CI.Full: failure for AuxCCS handling and render compression modifiers (rev15) Patchwork
2025-09-23 14:12 ` ✗ Xe.CI.Full: failure for AuxCCS handling and render compression modifiers (rev16) Patchwork
2025-09-23 20:12 ` [PATCH v12 00/13] AuxCCS handling and render compression modifiers Ville Syrjälä
2025-09-24  7:59   ` Tvrtko Ursulin

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=aN1NDy4Mfcs3A-eL@intel.com \
    --to=rodrigo.vivi@intel.com \
    --cc=intel-xe@lists.freedesktop.org \
    --cc=kernel-dev@igalia.com \
    --cc=tvrtko.ursulin@igalia.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