From: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
To: intel-xe@lists.freedesktop.org
Cc: kernel-dev@igalia.com, Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Subject: [PATCH 10/12] drm/xe: Force flush system memory AuxCCS framebuffers before scan out
Date: Fri, 21 Feb 2025 10:17:29 +0000 [thread overview]
Message-ID: <20250221101736.78986-11-tvrtko.ursulin@igalia.com> (raw)
In-Reply-To: <20250221101736.78986-1-tvrtko.ursulin@igalia.com>
Even though frame buffer objects are created as write-combined, in
practice, on top of all the ring buffer flushing, an additional clflush
seems to be needed before display engine can coherently scan out the
AuxCCS compressed data without transient artifacts.
If for comparison we look at how i915 handles things (where AuxCCS works
fine), as it happens it has this same clflush before a frame buffer is
pinned for display for the first time, courtesy the dynamic tracking of
the buffer cache mode and setting the latter to uncached before handing
to display.
Since xe considers the buffer object caching mode as static we can
implement the same approach by adding a flag telling us if the buffer
was ever pinned for display and flush on the first pin. Subsequent re-pins
will not repeat the clflush but so far I have not observed any glitching
after the first pin.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
---
drivers/gpu/drm/xe/display/xe_fb_pin.c | 13 +++++++++++++
drivers/gpu/drm/xe/xe_bo_types.h | 14 +++++++++-----
2 files changed, 22 insertions(+), 5 deletions(-)
diff --git a/drivers/gpu/drm/xe/display/xe_fb_pin.c b/drivers/gpu/drm/xe/display/xe_fb_pin.c
index a32f3603751a..5e7813154733 100644
--- a/drivers/gpu/drm/xe/display/xe_fb_pin.c
+++ b/drivers/gpu/drm/xe/display/xe_fb_pin.c
@@ -331,6 +331,7 @@ static struct i915_vma *__xe_pin_fb_vma(const struct intel_framebuffer *fb,
struct i915_vma *vma = kzalloc(sizeof(*vma), GFP_KERNEL);
struct drm_gem_object *obj = intel_fb_bo(&fb->base);
struct xe_bo *bo = gem_to_xe_bo(obj);
+ bool first_pin;
int ret;
if (!vma)
@@ -362,6 +363,9 @@ static struct i915_vma *__xe_pin_fb_vma(const struct intel_framebuffer *fb,
if (ret)
goto err;
+ first_pin = !bo->display_pin;
+ bo->display_pin = true;
+
if (IS_DGFX(xe))
ret = xe_bo_migrate(bo, XE_PL_VRAM0);
else
@@ -382,6 +386,15 @@ static struct i915_vma *__xe_pin_fb_vma(const struct intel_framebuffer *fb,
/* Ensure DPT writes are flushed */
xe_device_l2_flush(xe);
+
+ /*
+ * Force flush frame buffer data for non-coherent display access when
+ * AuxCCS formats are used.
+ */
+ if (first_pin && !xe_bo_is_vram(bo) && !xe_bo_is_stolen(bo) &&
+ intel_fb_is_ccs_modifier(fb->base.modifier))
+ drm_clflush_sg(xe_bo_sg(bo));
+
return vma;
err_unpin:
diff --git a/drivers/gpu/drm/xe/xe_bo_types.h b/drivers/gpu/drm/xe/xe_bo_types.h
index 60c522866500..f518becb78c9 100644
--- a/drivers/gpu/drm/xe/xe_bo_types.h
+++ b/drivers/gpu/drm/xe/xe_bo_types.h
@@ -67,11 +67,6 @@ struct xe_bo {
struct llist_node freed;
/** @update_index: Update index if PT BO */
int update_index;
- /** @created: Whether the bo has passed initial creation */
- bool created;
-
- /** @ccs_cleared */
- bool ccs_cleared;
/**
* @cpu_caching: CPU caching mode. Currently only used for userspace
@@ -80,6 +75,15 @@ struct xe_bo {
*/
u16 cpu_caching;
+ /** @created: Whether the bo has passed initial creation */
+ bool created : 1;
+
+ /** @ccs_cleared */
+ bool ccs_cleared : 1;
+
+ /** @display_pin: Was it ever pinned to display */
+ bool display_pin : 1;
+
/** @vram_userfault_link: Link into @mem_access.vram_userfault.list */
struct list_head vram_userfault_link;
--
2.48.0
next prev parent reply other threads:[~2025-02-21 10:17 UTC|newest]
Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-02-21 10:17 [PATCH 00/12] AuxCCS handling and render compression modifiers Tvrtko Ursulin
2025-02-21 10:17 ` [PATCH 01/12] drm/xe: Fix MOCS debugfs LNCF readout Tvrtko Ursulin
2025-02-27 20:21 ` Rodrigo Vivi
2025-02-27 23:24 ` Matt Roper
2025-02-28 8:10 ` Tvrtko Ursulin
2025-02-21 10:17 ` [PATCH 02/12] drm/xe: Fix ring flush invalidation Tvrtko Ursulin
2025-02-27 23:46 ` Matt Roper
2025-02-28 9:08 ` Tvrtko Ursulin
2025-02-21 10:17 ` [PATCH 03/12] drm/xe: Pass flags directly to emit_flush_imm_ggtt Tvrtko Ursulin
2025-02-27 23:47 ` Matt Roper
2025-02-21 10:17 ` [PATCH 04/12] drm/xe: Add ring buffer handling for AuxCCS Tvrtko Ursulin
2025-02-21 10:17 ` [PATCH 05/12] drm/xe: Use correct type width for alignment in fb pinning code Tvrtko Ursulin
2025-02-28 14:57 ` Rodrigo Vivi
2025-02-21 10:17 ` [PATCH 06/12] drm/xe: Use fb cached min alignment Tvrtko Ursulin
2025-02-28 19:21 ` Rodrigo Vivi
2025-02-21 10:17 ` [PATCH 07/12] drm/xe: Reduce DPT table alignment as in i915 Tvrtko Ursulin
2025-02-28 19:25 ` Rodrigo Vivi
2025-02-21 10:17 ` [PATCH 08/12] drm/xe: Flush GGTT writes after populating DPT Tvrtko Ursulin
2025-02-28 19:27 ` Rodrigo Vivi
2025-03-04 14:23 ` Tvrtko Ursulin
2025-02-21 10:17 ` [PATCH 09/12] drm/xe: Handle DPT in system memory Tvrtko Ursulin
2025-02-21 10:17 ` Tvrtko Ursulin [this message]
2025-02-28 19:34 ` [PATCH 10/12] drm/xe: Force flush system memory AuxCCS framebuffers before scan out Rodrigo Vivi
2025-03-04 14:21 ` Tvrtko Ursulin
2025-02-21 10:17 ` [PATCH 11/12] drm/xe/display: Add support for AuxCCS Tvrtko Ursulin
2025-02-21 10:17 ` [PATCH 12/12] drm/xe/display: Expose AuxCCS frame buffer modifiers Tvrtko Ursulin
2025-02-28 19:32 ` Rodrigo Vivi
2025-02-21 10:25 ` ✓ CI.Patch_applied: success for AuxCCS handling and render compression modifiers (rev2) Patchwork
2025-02-21 10:25 ` ✗ CI.checkpatch: warning " Patchwork
2025-02-21 10:27 ` ✓ CI.KUnit: success " Patchwork
2025-02-21 10:53 ` ✓ CI.Build: " Patchwork
2025-02-21 10:56 ` ✓ CI.Hooks: " Patchwork
2025-02-21 10:57 ` ✓ CI.checksparse: " Patchwork
2025-02-21 11:16 ` ✓ Xe.CI.BAT: " Patchwork
2025-02-22 0:13 ` ✗ 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=20250221101736.78986-11-tvrtko.ursulin@igalia.com \
--to=tvrtko.ursulin@igalia.com \
--cc=intel-xe@lists.freedesktop.org \
--cc=kernel-dev@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