From: "Ville Syrjälä" <ville.syrjala@linux.intel.com>
To: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Cc: intel-xe@lists.freedesktop.org, kernel-dev@igalia.com
Subject: Re: [PATCH v5 05/16] drm/xe/xelp: Quiesce memory traffic before invalidating auxccs
Date: Fri, 4 Apr 2025 18:44:56 +0300 [thread overview]
Message-ID: <Z-_-ePLbnbZOALEN@intel.com> (raw)
In-Reply-To: <20250403190317.6064-6-tvrtko.ursulin@igalia.com>
On Thu, Apr 03, 2025 at 08:03:05PM +0100, Tvrtko Ursulin wrote:
> According to i915 before invalidating auxccs we must quiesce the memory
> traffic by an extra flush.
>
> Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
> ---
> drivers/gpu/drm/xe/xe_ring_ops.c | 14 ++++++++++----
> drivers/gpu/drm/xe/xe_ring_ops_types.h | 2 +-
> 2 files changed, 11 insertions(+), 5 deletions(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_ring_ops.c b/drivers/gpu/drm/xe/xe_ring_ops.c
> index 61295f93ffb7..3f075dd36c86 100644
> --- a/drivers/gpu/drm/xe/xe_ring_ops.c
> +++ b/drivers/gpu/drm/xe/xe_ring_ops.c
> @@ -172,7 +172,8 @@ static int emit_store_imm_ppgtt_posted(u64 addr, u64 value,
> return i;
> }
>
> -static int emit_render_cache_flush(struct xe_sched_job *job, u32 *dw, int i)
> +static int emit_render_cache_flush(struct xe_sched_job *job, bool flush_l3,
> + u32 *dw, int i)
> {
> struct xe_gt *gt = job->q->gt;
> struct xe_device *xe = gt_to_xe(gt);
> @@ -204,7 +205,7 @@ static int emit_render_cache_flush(struct xe_sched_job *job, u32 *dw, int i)
> * deals with Protected Memory which is not needed for
> * AUX CCS invalidation and lead to unwanted side effects.
> */
> - if (GRAPHICS_VERx100(xe) < 1270)
> + if (flush_l3 && GRAPHICS_VERx100(xe) < 1270)
> flags |= PIPE_CONTROL_FLUSH_L3;
>
> if (lacks_render)
> @@ -370,10 +371,15 @@ 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);
>
> + /* hsdes: 1809175790 */
> + if (aux_ccs)
> + i = emit_render_cache_flush(job, 0, dw, i);
s/0/false/
> +
> dw[i++] = preparser_disable(true);
> if (lacks_render)
> mask_flags = PIPE_CONTROL_3D_ARCH_FLAGS;
> @@ -384,7 +390,7 @@ 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);
> @@ -394,7 +400,7 @@ static void __emit_job_gen12_render_compute(struct xe_sched_job *job,
>
> i = emit_bb_start(batch_addr, ppgtt_flag, dw, i);
>
> - i = emit_render_cache_flush(job, dw, i);
> + i = emit_render_cache_flush(job, true, dw, i);
>
> if (job->user_fence.used)
> i = emit_store_imm_ppgtt_posted(job->user_fence.addr,
> 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
--
Ville Syrjälä
Intel
next prev parent reply other threads:[~2025-04-04 15:45 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-04-03 19:03 [PATCH v5 00/16] AuxCCS handling and render compression modifiers Tvrtko Ursulin
2025-04-03 19:03 ` [PATCH v5 01/16] drm/xe: Adjust ringbuf emission for maximum possible size Tvrtko Ursulin
2025-04-15 11:59 ` Francois Dugast
2025-04-15 12:40 ` Tvrtko Ursulin
2025-04-15 14:21 ` Lucas De Marchi
2025-04-03 19:03 ` [PATCH v5 02/16] drm/xe: Use emit_flush_imm_ggtt helper instead of open coding Tvrtko Ursulin
2025-04-03 19:03 ` [PATCH v5 03/16] drm/xe/xelpg: Flush CCS when flushing caches Tvrtko Ursulin
2025-04-03 19:03 ` [PATCH v5 04/16] drm/xe: Flush L3 when flushing render cache Tvrtko Ursulin
2025-04-03 19:03 ` [PATCH v5 05/16] drm/xe/xelp: Quiesce memory traffic before invalidating auxccs Tvrtko Ursulin
2025-04-04 15:44 ` Ville Syrjälä [this message]
2025-04-03 19:03 ` [PATCH v5 06/16] drm/xe/xelp: Support auxccs invalidation on blitter Tvrtko Ursulin
2025-04-03 19:03 ` [PATCH v5 07/16] drm/xe/xelp: Use MI_FLUSH_DW_CCS on auxccs platforms Tvrtko Ursulin
2025-04-03 19:03 ` [PATCH v5 08/16] drm/xe/xelp: Wait for AuxCCS invalidation to complete Tvrtko Ursulin
2025-04-03 19:03 ` [PATCH v5 09/16] drm/xe/xelp: Add AuxCCS invalidation to the buffer migration path Tvrtko Ursulin
2025-04-03 19:03 ` [PATCH v5 10/16] drm/xe: Use fb cached min alignment Tvrtko Ursulin
2025-04-03 19:03 ` [PATCH v5 11/16] drm/xe: Reduce DPT table alignment as in i915 Tvrtko Ursulin
2025-04-04 15:52 ` Ville Syrjälä
2025-04-03 19:03 ` [PATCH v5 12/16] drm/xe: Flush GGTT writes after populating DPT Tvrtko Ursulin
2025-04-03 19:03 ` [PATCH v5 13/16] drm/xe: Handle DPT in system memory Tvrtko Ursulin
2025-04-03 19:03 ` [PATCH v5 14/16] drm/xe: Force flush system memory AuxCCS framebuffers before scan out Tvrtko Ursulin
2025-04-04 16:20 ` Ville Syrjälä
2025-04-14 9:44 ` Tvrtko Ursulin
2025-04-03 19:03 ` [PATCH v5 15/16] drm/xe/display: Add support for AuxCCS Tvrtko Ursulin
2025-04-03 19:03 ` [PATCH v5 16/16] drm/i915/display: Expose AuxCCS frame buffer modifiers for Xe Tvrtko Ursulin
2025-04-04 2:19 ` ✓ CI.Patch_applied: success for AuxCCS handling and render compression modifiers (rev5) Patchwork
2025-04-04 2:19 ` ✗ CI.checkpatch: warning " Patchwork
2025-04-04 2:21 ` ✓ CI.KUnit: success " Patchwork
2025-04-04 2:37 ` ✓ CI.Build: " Patchwork
2025-04-04 2:40 ` ✓ CI.Hooks: " Patchwork
2025-04-04 2:41 ` ✓ CI.checksparse: " Patchwork
2025-04-04 3:27 ` ✓ Xe.CI.BAT: " Patchwork
2025-04-04 12:03 ` ✗ Xe.CI.Full: failure " 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=Z-_-ePLbnbZOALEN@intel.com \
--to=ville.syrjala@linux.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