dri-devel Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Qinyun Tan <qinyuntan@linux.alibaba.com>
To: dri-devel@lists.freedesktop.org
Cc: tzimmermann@suse.de, airlied@redhat.com, jfalempe@redhat.com,
	kraxel@redhat.com, dmitry.osipenko@collabora.com,
	hansg@kernel.org, maarten.lankhorst@linux.intel.com,
	mripard@kernel.org, simona@ffwll.ch,
	leandro.ribeiro@collabora.com, daniels@collabora.com,
	pekka.paalanen@collabora.com, virtualization@lists.linux.dev,
	spice-devel@lists.freedesktop.org, linux-kernel@vger.kernel.org,
	Qinyun Tan <qinyuntan@linux.alibaba.com>
Subject: [PATCH v2 0/4] drm: create blend mode property on alpha-capable planes of simple drivers
Date: Tue,  1 Sep 2026 16:32:30 +0800	[thread overview]
Message-ID: <20260901083234.1828755-1-qinyuntan@linux.alibaba.com> (raw)

Since commit 860e748bddcc ("drm: ensure blend mode supported if pixel
format with alpha exposed"), drm_mode_config_validate() warns when a
plane exposes an alpha pixel format but does not create the "pixel
blend mode" property. Several simple/virtual drivers trip this on
driver load. I hit the ast one on an ASPEED AST2600 BMC:

  [PLANE:37:plane-1] pixel format with alpha exposed but blend mode not setup
  WARNING: drivers/gpu/drm/drm_mode_config.c:872 at drm_mode_config_validate+0x48f/0x510 [drm]
  ...
  Call Trace:
   drm_dev_register+0x1ce/0x290 [drm]
   ast_pci_probe+0x19d/0x3f0 [ast]
   local_pci_probe+0x41/0x90

These four drivers share the same gap; this series covers ast, qxl,
virtio-gpu and vboxvideo. Each patch creates the "pixel blend mode"
property advertising the single blend mode the device actually
implements, based on the following:

  - ast: the hardware cursor blends with straight alpha per the
    ASPEED documentation (thanks Thomas), i.e. COVERAGE.
  - qxl: the SPICE protocol explicitly defines the ALPHA cursor type
    as a "pre-multiplied ARGB8888 pixmap" [1], i.e. PREMULTI.
  - virtio-gpu: the virtio spec does not define cursor alpha
    semantics; the host forwards the pixels verbatim and the remote
    cursor protocols among its frontends (SPICE alpha cursors, the
    VNC "Cursor With Alpha" encoding) define pre-multiplied alpha,
    i.e. PREMULTI.
  - vboxvideo: the host frontend loads the pointer shape verbatim
    into an unpremultiplied ARGB image (QImage::Format_ARGB32) before
    handing it to the host cursor APIs, i.e. COVERAGE.

This follows the same approach as the already-merged i915 and nouveau
fixes:

  67e955e073c6 ("drm/i915/display: expose blend mode on alpha-capable planes")
  df0311845915 ("drm/nouveau/kms/nv50-: Unconditionally create blend_mode prop for wndws")

The ast patch is verified on AST2600 hardware: the warning is gone
and the cursor plane exposes the "pixel blend mode" property. The
qxl, virtio and vboxvideo patches follow the identical pattern and
are compile-tested.

[1] https://www.spice-space.org/spice-protocol.html
    ("Cursor channel definition" section)

v2:
  - ast: advertise DRM_MODE_BLEND_COVERAGE instead of PREMULTI, per
    Thomas Zimmermann's review and the ASPEED documentation.
  - vboxvideo: advertise DRM_MODE_BLEND_COVERAGE instead of PREMULTI
    after checking how the VirtualBox host composites the pointer
    shape.
  - qxl, virtio: code unchanged; commit messages now cite the
    protocol-level justification for PREMULTI.

Qinyun Tan (4):
  drm/ast: create blend mode property on cursor plane
  drm/qxl: create blend mode property on primary and cursor planes
  drm/virtio: create blend mode property on cursor plane
  drm/vboxvideo: create blend mode property on planes

 drivers/gpu/drm/ast/ast_cursor.c       | 3 +++
 drivers/gpu/drm/qxl/qxl_display.c      | 4 ++++
 drivers/gpu/drm/vboxvideo/vbox_mode.c  | 4 ++++
 drivers/gpu/drm/virtio/virtgpu_plane.c | 4 ++++
 4 files changed, 15 insertions(+)

-- 
2.43.7


             reply	other threads:[~2026-09-03  7:33 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-09-01  8:32 Qinyun Tan [this message]
2026-09-01  8:32 ` [PATCH v2 1/4] drm/ast: create blend mode property on cursor plane Qinyun Tan
2026-09-01  8:46   ` sashiko-bot
2026-09-01 11:54   ` Thomas Zimmermann
2026-09-01 12:30     ` Thomas Zimmermann
2026-09-01 12:51       ` Qinyun Tan
2026-09-03  7:05         ` Thomas Zimmermann
2026-09-04  5:31           ` Qinyun Tan
2026-09-03  7:01   ` Thomas Zimmermann
2026-09-01  8:32 ` [PATCH v2 2/4] drm/qxl: create blend mode property on primary and cursor planes Qinyun Tan
2026-09-01 11:45   ` Thomas Zimmermann
2026-09-01  8:32 ` [PATCH v2 3/4] drm/virtio: create blend mode property on cursor plane Qinyun Tan
2026-09-01 11:46   ` Thomas Zimmermann
2026-09-01  8:32 ` [PATCH v2 4/4] drm/vboxvideo: create blend mode property on planes Qinyun Tan
2026-09-01  8:46   ` sashiko-bot
2026-09-03  7:00   ` Thomas Zimmermann

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=20260901083234.1828755-1-qinyuntan@linux.alibaba.com \
    --to=qinyuntan@linux.alibaba.com \
    --cc=airlied@redhat.com \
    --cc=daniels@collabora.com \
    --cc=dmitry.osipenko@collabora.com \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=hansg@kernel.org \
    --cc=jfalempe@redhat.com \
    --cc=kraxel@redhat.com \
    --cc=leandro.ribeiro@collabora.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=maarten.lankhorst@linux.intel.com \
    --cc=mripard@kernel.org \
    --cc=pekka.paalanen@collabora.com \
    --cc=simona@ffwll.ch \
    --cc=spice-devel@lists.freedesktop.org \
    --cc=tzimmermann@suse.de \
    --cc=virtualization@lists.linux.dev \
    /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