pm24.git/drivers/gpu, branch v6.10-rc3

Merge tag 'drm-misc-next-fixes-2024-06-07' of https://gitlab.freedesktop.org/drm/misc/kernel into drm-fixes

2024-06-06T22:40:58Z

drm-misc-next-fixes for v6.10-rc3: - Single unused struct removal that should have been in -fixes. Signed-off-by: Dave Airlie From: Maarten Lankhorst Link: https://patchwork.freedesktop.org/patch/msgid/0251b6ae-bffa-44b2-b698-955712c25a27@linux.intel.com

Merge tag 'drm-misc-fixes-2024-06-07' of https://gitlab.freedesktop.org/drm/misc/kernel into drm-fixes

2024-06-06T22:37:25Z

drm-misc-fixes for v6.10-rc3: - Robustness fixes for vmwgfx. - Error check for of_drm_get_panel_orientation failing in sitronix-st7789v. Signed-off-by: Dave Airlie From: Maarten Lankhorst Link: https://patchwork.freedesktop.org/patch/msgid/d5645d00-a8cf-47d9-a2a0-4ff55842fc7d@linux.intel.com

Merge tag 'amd-drm-fixes-6.10-2024-06-06' of https://gitlab.freedesktop.org/agd5f/linux into drm-fixes

2024-06-06T22:22:59Z

amd-drm-fixes-6.10-2024-06-06: amdgpu: - Fix shutdown issues on some SMU 13.x platforms - Silence some UBSAN flexible array warnings Signed-off-by: Dave Airlie From: Alex Deucher Link: https://patchwork.freedesktop.org/patch/msgid/20240606192348.3620805-1-alexander.deucher@amd.com

drm/vmwgfx: Don't memcmp equivalent pointers

2024-06-06T02:38:40Z

These pointers are frequently the same and memcmp does not compare the pointers before comparing their contents so this was wasting cycles comparing 16 KiB of memory which will always be equal. Fixes: bb6780aa5a1d ("drm/vmwgfx: Diff cursors when using cmds") Signed-off-by: Ian Forbes Signed-off-by: Zack Rusin Link: https://patchwork.freedesktop.org/patch/msgid/20240328190716.27367-1-ian.forbes@broadcom.com

drm/vmwgfx: remove unused struct 'vmw_stdu_dma'

2024-06-06T02:20:25Z

'vmw_stdu_dma' is unused since commit 39985eea5a6d ("drm/vmwgfx: Abstract placement selection") Remove it. Signed-off-by: Dr. David Alan Gilbert Signed-off-by: Zack Rusin Link: https://patchwork.freedesktop.org/patch/msgid/20240517232858.230860-1-linux@treblig.org

drm/vmwgfx: Don't destroy Screen Target when CRTC is enabled but inactive

2024-06-06T02:11:42Z

drm_crtc_helper_funcs::atomic_disable can be called even when the CRTC is still enabled. This can occur when the mode changes or the CRTC is set as inactive. In the case where the CRTC is being set as inactive we only want to blank the screen. The Screen Target should remain intact as long as the mode has not changed and CRTC is enabled. This fixes a bug with GDM where locking the screen results in a permanent black screen because the Screen Target is no longer defined. Fixes: 7b0062036c3b ("drm/vmwgfx: Implement virtual crc generation") Signed-off-by: Ian Forbes Reviewed-by: Martin Krastev Signed-off-by: Zack Rusin Link: https://patchwork.freedesktop.org/patch/msgid/20240531203358.26677-1-ian.forbes@broadcom.com

drm/vmwgfx: Standardize use of kibibytes when logging

2024-06-06T02:10:49Z

Use the same standard abbreviation KiB instead of incorrect variants. Signed-off-by: Ian Forbes Signed-off-by: Zack Rusin Link: https://patchwork.freedesktop.org/patch/msgid/20240521184720.767-5-ian.forbes@broadcom.com

drm/vmwgfx: Remove STDU logic from generic mode_valid function

2024-06-06T02:10:48Z

STDU has its own mode_valid function now so this logic can be removed from the generic version. Fixes: 935f795045a6 ("drm/vmwgfx: Refactor drm connector probing for display modes") Signed-off-by: Ian Forbes Signed-off-by: Zack Rusin Link: https://patchwork.freedesktop.org/patch/msgid/20240521184720.767-4-ian.forbes@broadcom.com

drm/vmwgfx: 3D disabled should not effect STDU memory limits

2024-06-06T02:10:47Z

This limit became a hard cap starting with the change referenced below. Surface creation on the device will fail if the requested size is larger than this limit so altering the value arbitrarily will expose modes that are too large for the device's hard limits. Fixes: 7ebb47c9f9ab ("drm/vmwgfx: Read new register for GB memory when available") Signed-off-by: Ian Forbes Signed-off-by: Zack Rusin Link: https://patchwork.freedesktop.org/patch/msgid/20240521184720.767-3-ian.forbes@broadcom.com

drm/vmwgfx: Filter modes which exceed graphics memory

2024-06-06T02:10:46Z

SVGA requires individual surfaces to fit within graphics memory (max_mob_pages) which means that modes with a final buffer size that would exceed graphics memory must be pruned otherwise creation will fail. Additionally llvmpipe requires its buffer height and width to be a multiple of its tile size which is 64. As a result we have to anticipate that llvmpipe will round up the mode size passed to it by the compositor when it creates buffers and filter modes where this rounding exceeds graphics memory. This fixes an issue where VMs with low graphics memory (< 64MiB) configured with high resolution mode boot to a black screen because surface creation fails. Fixes: d947d1b71deb ("drm/vmwgfx: Add and connect connector helper function") Signed-off-by: Ian Forbes Signed-off-by: Zack Rusin Link: https://patchwork.freedesktop.org/patch/msgid/20240521184720.767-2-ian.forbes@broadcom.com