kernel.git/drivers/gpu/drm/msm/msm_fb.c, branch linux-rolling-stable

drm/msm: Pass along the format info from .fb_create() to drm_helper_mode_fill_fb_struct()

2025-07-16T17:11:05Z

Plumb the format info from .fb_create() all the way to drm_helper_mode_fill_fb_struct() to avoid the redundant lookups. For the fbdev case a manual drm_get_format_info() lookup is needed. Cc: Rob Clark Cc: Abhinav Kumar Cc: Dmitry Baryshkov Cc: Sean Paul Cc: Marijn Suijten Cc: linux-arm-msm@vger.kernel.org Cc: freedreno@lists.freedesktop.org Reviewed-by: Thomas Zimmermann Acked-by: Dmitry Baryshkov Signed-off-by: Ville Syrjälä Link: https://patchwork.freedesktop.org/patch/msgid/20250701090722.13645-16-ville.syrjala@linux.intel.com

drm: Allow the caller to pass in the format info to drm_helper_mode_fill_fb_struct()

2025-07-16T17:04:45Z

Soon all drivers should have the format info already available in the places where they call drm_helper_mode_fill_fb_struct(). Allow it to be passed along into drm_helper_mode_fill_fb_struct() instead of doing yet another redundant lookup. Start by always passing in NULL and still doing the extra lookup. The actual changes to avoid the lookup will follow. Done with cocci (with some manual fixups): @@ identifier dev, fb, mode_cmd; expression get_format_info; @@ void drm_helper_mode_fill_fb_struct(struct drm_device *dev, struct drm_framebuffer *fb, + const struct drm_format_info *info, const struct drm_mode_fb_cmd2 *mode_cmd) { ... - fb->format = get_format_info; + fb->format = info ?: get_format_info; ... } @@ identifier dev, fb, mode_cmd; @@ void drm_helper_mode_fill_fb_struct(struct drm_device *dev, struct drm_framebuffer *fb, + const struct drm_format_info *info, const struct drm_mode_fb_cmd2 *mode_cmd); @@ expression dev, fb, mode_cmd; @@ drm_helper_mode_fill_fb_struct(dev, fb + ,NULL ,mode_cmd); Cc: Alex Deucher Cc: Liviu Dudau Cc: Russell King Cc: Inki Dae Cc: Seung-Woo Kim Cc: Kyungmin Park Cc: Patrik Jakobsson Cc: Rob Clark Cc: Abhinav Kumar Cc: Dmitry Baryshkov Cc: Sean Paul Cc: Marijn Suijten Cc: Lyude Paul Cc: Danilo Krummrich Cc: Tomi Valkeinen Cc: Thierry Reding Cc: Mikko Perttunen Cc: Gerd Hoffmann Cc: Dmitry Osipenko Cc: Gurchetan Singh Cc: Chia-I Wu Cc: Zack Rusin Cc: Broadcom internal kernel review list Cc: amd-gfx@lists.freedesktop.org Cc: linux-arm-msm@vger.kernel.org Cc: freedreno@lists.freedesktop.org Cc: nouveau@lists.freedesktop.org Cc: linux-tegra@vger.kernel.org Cc: virtualization@lists.linux.dev Reviewed-by: Thomas Zimmermann Reviewed-by: Laurent Pinchart Reviewed-by: Dmitry Baryshkov Reviewed-by: Liviu Dudau Acked-by: Alex Deucher Acked-by: Rodrigo Vivi Signed-off-by: Ville Syrjälä Link: https://patchwork.freedesktop.org/patch/msgid/20250701090722.13645-6-ville.syrjala@linux.intel.com

drm: Pass the format info to .fb_create()

2025-07-16T17:03:14Z

Pass along the format information from the top to .fb_create() so that we can avoid redundant (and somewhat expensive) lookups in the drivers. Done with cocci (with some manual fixups): @@ identifier func =~ ".*create.*"; identifier dev, file, mode_cmd; @@ struct drm_framebuffer *func( struct drm_device *dev, struct drm_file *file, + const struct drm_format_info *info, const struct drm_mode_fb_cmd2 *mode_cmd) { ... ( - const struct drm_format_info *info = drm_get_format_info(...); | - const struct drm_format_info *info; ... - info = drm_get_format_info(...); ) <... - if (!info) - return ...; ...> } @@ identifier func =~ ".*create.*"; identifier dev, file, mode_cmd; @@ struct drm_framebuffer *func( struct drm_device *dev, struct drm_file *file, + const struct drm_format_info *info, const struct drm_mode_fb_cmd2 *mode_cmd) { ... } @find@ identifier fb_create_func =~ ".*create.*"; identifier dev, file, mode_cmd; @@ struct drm_framebuffer *fb_create_func( struct drm_device *dev, struct drm_file *file, + const struct drm_format_info *info, const struct drm_mode_fb_cmd2 *mode_cmd); @@ identifier find.fb_create_func; expression dev, file, mode_cmd; @@ fb_create_func(dev, file + ,info ,mode_cmd) @@ expression dev, file, mode_cmd; @@ drm_gem_fb_create(dev, file + ,info ,mode_cmd) @@ expression dev, file, mode_cmd; @@ drm_gem_fb_create_with_dirty(dev, file + ,info ,mode_cmd) @@ expression dev, file_priv, mode_cmd; identifier info, fb; @@ info = drm_get_format_info(...); ... fb = dev->mode_config.funcs->fb_create(dev, file_priv + ,info ,mode_cmd); @@ identifier dev, file_priv, mode_cmd; @@ struct drm_mode_config_funcs { ... struct drm_framebuffer *(*fb_create)(struct drm_device *dev, struct drm_file *file_priv, + const struct drm_format_info *info, const struct drm_mode_fb_cmd2 *mode_cmd); ... }; v2: Fix kernel docs (Laurent) Fix commit msg (Geert) Cc: Alex Deucher Cc: Liviu Dudau Cc: Maxime Ripard Cc: Russell King Cc: Inki Dae Cc: Seung-Woo Kim Cc: Kyungmin Park Cc: Patrik Jakobsson Cc: Chun-Kuang Hu Cc: Philipp Zabel Cc: Rob Clark Cc: Abhinav Kumar Cc: Dmitry Baryshkov Cc: Sean Paul Cc: Marijn Suijten Cc: Marek Vasut Cc: Stefan Agner Cc: Lyude Paul Cc: Danilo Krummrich Cc: Tomi Valkeinen Cc: Dave Airlie Cc: Gerd Hoffmann Cc: Kieran Bingham Cc: Biju Das Cc: Sandy Huang Cc: "Heiko Stübner" Cc: Andy Yan Cc: Thierry Reding Cc: Mikko Perttunen Cc: Dave Stevenson Cc: "Maíra Canal" Cc: Raspberry Pi Kernel Maintenance Cc: Dmitry Osipenko Cc: Gurchetan Singh Cc: Chia-I Wu Cc: Zack Rusin Cc: Broadcom internal kernel review list Cc: Oleksandr Andrushchenko Cc: amd-gfx@lists.freedesktop.org Cc: linux-arm-msm@vger.kernel.org Cc: freedreno@lists.freedesktop.org Cc: nouveau@lists.freedesktop.org Cc: virtualization@lists.linux.dev Cc: spice-devel@lists.freedesktop.org Cc: linux-renesas-soc@vger.kernel.org Cc: linux-tegra@vger.kernel.org Cc: Laurent Pinchart Reviewed-by: Geert Uytterhoeven Reviewed-by: Thomas Zimmermann Reviewed-by: Dmitry Baryshkov Acked-by: Liviu Dudau Reviewed-by: Laurent Pinchart Acked-by: Alex Deucher Acked-by: Rodrigo Vivi Signed-off-by: Ville Syrjälä Link: https://patchwork.freedesktop.org/patch/msgid/20250701090722.13645-5-ville.syrjala@linux.intel.com

drm: Pass pixel_format+modifier directly to drm_get_format_info()

2025-07-16T17:01:13Z

Decouple drm_get_format_info() from struct drm_mode_fb_cmd2 and just pass the pixel format+modifier combo in by hand. We may want to use drm_get_format_info() outside of the normal addfb paths where we won't have a struct drm_mode_fb_cmd2, and creating a temporary one just for this seems silly. Done with cocci: @@ identifier dev, mode_cmd; @@ struct drm_format_info * drm_get_format_info(struct drm_device *dev, - const struct drm_mode_fb_cmd2 *mode_cmd + u32 pixel_format, u64 modifier ) { <... ( - mode_cmd->pixel_format + pixel_format | - mode_cmd->modifier[0] + modifier ) ...> } @@ identifier dev, mode_cmd; @@ struct drm_format_info * drm_get_format_info(struct drm_device *dev, - const struct drm_mode_fb_cmd2 *mode_cmd + u32 pixel_format, u64 modifier ); @@ expression dev, mode_cmd; @@ - drm_get_format_info(dev, mode_cmd) + drm_get_format_info(dev, mode_cmd->pixel_format, mode_cmd->modifier[0]) v2: Fix kernel docs (Laurent) Drop drm_mode_fb_cmd2 forward declaration (Thomas) Cc: Liviu Dudau Cc: Russell King Cc: Inki Dae Cc: Seung-Woo Kim Cc: Kyungmin Park Cc: Patrik Jakobsson Cc: Chun-Kuang Hu Cc: Philipp Zabel Cc: Rob Clark Cc: Abhinav Kumar Cc: Dmitry Baryshkov Cc: Sean Paul Cc: Marijn Suijten Cc: Marek Vasut Cc: Stefan Agner Cc: Lyude Paul Cc: Danilo Krummrich Cc: Tomi Valkeinen Cc: Alex Deucher Cc: Sandy Huang Cc: "Heiko Stübner" Cc: Andy Yan Cc: Thierry Reding Cc: Mikko Perttunen Cc: linux-arm-msm@vger.kernel.org Cc: freedreno@lists.freedesktop.org Cc: nouveau@lists.freedesktop.org Cc: amd-gfx@lists.freedesktop.org Cc: linux-tegra@vger.kernel.org Reviewed-by: Thomas Zimmermann Reviewed-by: Laurent Pinchart Reviewed-by: Liviu Dudau Acked-by: Alex Deucher Signed-off-by: Ville Syrjälä Link: https://patchwork.freedesktop.org/patch/msgid/20250701090722.13645-3-ville.syrjala@linux.intel.com

drm/msm: Defer VMA unmap for fb unpins

2025-07-05T00:48:38Z

With the conversion to drm_gpuvm, we lost the lazy VMA cleanup, which means that fb cleanup/unpin when pageflipping to new scanout buffers immediately unmaps the scanout buffer. This is costly (with tlbinv, it can be 4-6ms for a 1080p scanout buffer, and more for higher resolutions)! To avoid this, introduce a vma_ref, which is incremented whenever userspace has a GEM handle or dma-buf fd. When unpinning if the vm is the kms->vm we defer tearing down the VMA until the vma_ref drops to zero. If the buffer is still part of a flip-chain then userspace will be holding some sort of reference to the BO, either via a GEM handle and/or dma-buf fd. So this avoids unmapping the VMA when there is a strong possibility that it will be needed again. Signed-off-by: Rob Clark Tested-by: Antonino Maniscalco Reviewed-by: Antonino Maniscalco Patchwork: https://patchwork.freedesktop.org/patch/661538/

drm/msm: Use drm_gpuvm types more

2025-07-05T00:48:35Z

Most of the driver code doesn't need to reach in to msm specific fields, so just use the drm_gpuvm/drm_gpuva types directly. This should hopefully improve commonality with other drivers and make the code easier to understand. Signed-off-by: Rob Clark Signed-off-by: Rob Clark Tested-by: Antonino Maniscalco Reviewed-by: Antonino Maniscalco Patchwork: https://patchwork.freedesktop.org/patch/661483/

drm/msm: Refcount framebuffer pins

2025-07-04T18:09:45Z

We were already keeping a refcount of # of prepares (pins), to clear the iova array. Use that to avoid unpinning the iova until the last cleanup (unpin). This way, when msm_gem_unpin_iova() actually tears down the mapping, we won't have problems if the fb is being scanned out on another display (for example). Signed-off-by: Rob Clark Tested-by: Antonino Maniscalco Reviewed-by: Antonino Maniscalco Patchwork: https://patchwork.freedesktop.org/patch/661477/

drm/msm: Stop passing vm to msm_framebuffer

2025-07-04T18:09:45Z

The fb only deals with kms->vm, so make that explicit. This will start letting us refcount the # of times the fb is pinned, so we can only unpin the vma after last user of the fb is done. Having a single reference count really only works if there is only a single vm. Signed-off-by: Rob Clark Tested-by: Antonino Maniscalco Reviewed-by: Antonino Maniscalco Patchwork: https://patchwork.freedesktop.org/patch/661476/

drm/msm: Rename msm_gem_address_space -> msm_gem_vm

2025-07-04T18:09:44Z

Re-aligning naming to better match drm_gpuvm terminology will make things less confusing at the end of the drm_gpuvm conversion. This is just rename churn, no functional change. Signed-off-by: Rob Clark Reviewed-by: Dmitry Baryshkov Signed-off-by: Rob Clark Tested-by: Antonino Maniscalco Reviewed-by: Antonino Maniscalco Patchwork: https://patchwork.freedesktop.org/patch/661466/

drm/msm: drop msm_kms_funcs::get_format() callback

2024-04-23T09:56:30Z

Now as all subdrivers were converted to use common database of formats, drop the get_format() callback and use mdp_get_format() directly. Reviewed-by: Abhinav Kumar Signed-off-by: Dmitry Baryshkov Patchwork: https://patchwork.freedesktop.org/patch/590431/ Link: https://lore.kernel.org/r/20240420-dpu-format-v2-9-9e93226cbffd@linaro.org