Changes in drm_internal.h

1f828b4d - drm/client: Make client support optional

Mon, 14 Oct 2024 08:55:25 +0000

drm/client: Make client support optionalOnly build client code if DRM_CLIENT has been selected. Automatiallydo so if one of the default clients has been enabled. If client supporthas been disabled, the helpers for client-related events are empty andthe regular client functions are not present.Amdgpu has an internal DRM client, so it has to select DRM_CLIENT byitself unconditionally.v3:- provide empty drm_client_debugfs_init() if DRM_CLIENT=n (kernel test robot)Signed-off-by: Thomas Zimmermann Cc: Alex Deucher Cc: "Christian König" Cc: Xinhui Pan Reviewed-by: Jonathan Cavitt Link: https://patchwork.freedesktop.org/patch/msgid/20241014085740.582287-12-tzimmermann@suse.de List of files: /linux-6.15/drivers/gpu/drm/drm_internal.h

45c4d994 - accel: Use XArray instead of IDR for minors

Fri, 23 Aug 2024 16:30:47 +0000

accel: Use XArray instead of IDR for minorsAccel minor management is based on DRM (and is also using structdrm_minor internally), since DRM is using XArray for minors, it makessense to also convert accel.As the two implementations are identical (only difference being theunderlying xarray), move the accel_minor_* functionality to DRM.Signed-off-by: Michał Winiarski Acked-by: James Zhu Acked-by: Christian König Link: https://patchwork.freedesktop.org/patch/msgid/20240823163048.2676257-3-michal.winiarski@intel.comSigned-off-by: Christian König List of files: /linux-6.15/drivers/gpu/drm/drm_internal.h

b5757a5b - drm: Remove struct drm_driver.lastclose

Mon, 12 Aug 2024 08:28:29 +0000

drm: Remove struct drm_driver.lastcloseThe lastclose callback in struct drm_driver is unused. Remove it. Alsoupdate documentation.v2:- update to use drm_lastclose()- fix typo in documentationSigned-off-by: Thomas Zimmermann Reviewed-by: Daniel Vetter Link: https://patchwork.freedesktop.org/patch/msgid/20240812083000.337744-9-tzimmermann@suse.de List of files: /linux-6.15/drivers/gpu/drm/drm_internal.h

4e765920 - drm/edid: move all internal declarations to drm_crtc_internal.h

Tue, 16 Apr 2024 09:19:55 +0000

drm/edid: move all internal declarations to drm_crtc_internal.hThe declarations for internal EDID functions are a bit scattered. Putthem all in drm_crtc_internal.h.Reviewed-by: Thomas Zimmermann Link: https://patchwork.freedesktop.org/patch/msgid/9ae137ea51f9cc2ccb3899b0acda553e6a8ce2db.1713259151.git.jani.nikula@intel.comSigned-off-by: Jani Nikula List of files: /linux-6.15/drivers/gpu/drm/drm_internal.h

a7802784 - drm/gem: Acquire reservation lock in drm_gem_{pin/unpin}()

Tue, 27 Feb 2024 10:14:56 +0000

drm/gem: Acquire reservation lock in drm_gem_{pin/unpin}()Acquire the buffer object's reservation lock in drm_gem_pin() andremove locking the drivers' GEM callbacks where necessary. Same forunpin().DRM drivers and memory managers modified by this patch will now havecorrect dma-buf locking semantics: the caller is responsible forholding the reservation lock when calling the pin or unpin callback.DRM drivers and memory managers that are not modified will now beprotected against concurent invocation of their pin and unpin callbacks.PRIME does not implement struct dma_buf_ops.pin, which requiresthe caller to hold the reservation lock. It does implement structdma_buf_ops.attach, which requires to callee to acquire thereservation lock. The PRIME code uses drm_gem_pin(), so locksare now taken as specified. Same for unpin and detach.The patch harmonizes GEM pin and unpin to have non-interruptiblereservation locking across all drivers, as is already the case forvmap and vunmap. This affects gem-shmem, gem-vram, loongson, qxl andradeon.Signed-off-by: Thomas Zimmermann Reviewed-by: Christian König Reviewed-by: Zack Rusin Reviewed-by: Dmitry Osipenko Tested-by: Dmitry Osipenko # virtio-gpuLink: https://patchwork.freedesktop.org/patch/msgid/20240227113853.8464-10-tzimmermann@suse.de List of files: /linux-6.15/drivers/gpu/drm/drm_internal.h

b5d7cb76 - drm: add missing header guards to drm_internal.h

Wed, 06 Mar 2024 18:31:07 +0000

drm: add missing header guards to drm_internal.hIncluding the file twice leads to errors.Reviewed-by: Thomas Zimmermann Signed-off-by: Jani Nikula Link: https://patchwork.freedesktop.org/patch/msgid/6e744360513e581765147ea7b1e693f4bffe03a9.1709749576.git.jani.nikula@intel.com List of files: /linux-6.15/drivers/gpu/drm/drm_internal.h

2504c7ec - drm: Remove source code for non-KMS drivers

Wed, 22 Nov 2023 12:09:41 +0000

drm: Remove source code for non-KMS driversRemove all remaining source code for non-KMS drivers. These drivershave been removed in v6.3 and won't comeback.Signed-off-by: Thomas Zimmermann Reviewed-by: David Airlie Reviewed-by: Daniel Vetter Acked-by: Alex Deucher Link: https://patchwork.freedesktop.org/patch/msgid/20231122122449.11588-13-tzimmermann@suse.de List of files: /linux-6.15/drivers/gpu/drm/drm_internal.h

6bb0814b - drm: Remove the legacy DRM_IOCTL_MODESET_CTL ioctl

Wed, 22 Nov 2023 12:09:38 +0000

drm: Remove the legacy DRM_IOCTL_MODESET_CTL ioctlDRM drivers with user-space mode setting have been removed in Linuxv6.3. [1] Now remove the ioctl entry points for these drivers. Invokingany of the ioctl ops will unconditionally return -EINVAL to user space.Invoking DRM_IOCTL_MODESET_CTL is different from the other legacyioctl ops as it returns 0 even without CONFIG_DRM_LEGACY set. From theoriginal commit 29935554b384 ("drm: Disallow DRM_IOCTL_MODESET_CTL forKMS drivers") it is not apparent how or why the operation differs fromthe others. It is likely just an oversight in commit 61ae227032e7("drm: allow removal of legacy codepaths (v4.1)"), which alloweddisabling leagacy ioctls in the first place. Still keep this removalseparate from the other ioctls to allow an easy revert, if necessary.Signed-off-by: Thomas Zimmermann Link: https://patchwork.freedesktop.org/series/111602/ # [1]Reviewed-by: David Airlie Reviewed-by: Daniel Vetter Acked-by: Alex Deucher Link: https://patchwork.freedesktop.org/patch/msgid/20231122122449.11588-10-tzimmermann@suse.de List of files: /linux-6.15/drivers/gpu/drm/drm_internal.h

caf525ed - drm/encoder: register per-encoder debugfs dir

Sun, 03 Dec 2023 11:53:13 +0000

drm/encoder: register per-encoder debugfs dirEach of connectors and CRTCs used by the DRM device provides debugfsdirectory, which is used by several standard debugfs files and canfurther be extended by the driver. Add such generic debugfs directoriesfor encoder.Reviewed-by: Neil Armstrong Acked-by: Maxime Ripard Signed-off-by: Dmitry Baryshkov Link: https://patchwork.freedesktop.org/patch/msgid/20231203115315.1306124-2-dmitry.baryshkov@linaro.org List of files: /linux-6.15/drivers/gpu/drm/drm_internal.h

8af46811 - drm/edid: add helpers to get/set struct cea_sad from/to 3-byte sad

Tue, 31 Oct 2023 10:16:42 +0000

drm/edid: add helpers to get/set struct cea_sad from/to 3-byte sadAdd helpers to pack/unpack SADs. Both ways and non-static, as follow-upwork needs them.v2: Add include to get the declarationsCc: Mitul Golani Reviewed-by: Mitul Golani Signed-off-by: Jani Nikula Link: https://patchwork.freedesktop.org/patch/msgid/21d657ca854ce26423b461c0bb71e7a0727ba437.1698747331.git.jani.nikula@intel.com List of files: /linux-6.15/drivers/gpu/drm/drm_internal.h

21b6c281 - drm/sysfs: Register "ddc" symlink later

Tue, 29 Aug 2023 11:39:10 +0000

drm/sysfs: Register "ddc" symlink laterCurrently drm_sysfs_connector_add() attempts to registerthe "ddc" symlink (based one connector->ddc) before thedriver's .early_register() hook has been called. That istoo early for i915 which only fully registers the aux chand associated i2c bus from said hook (to prevent halfinitialized stuff getting exposed to userspace). Thiscauses my attempt at using drm_connector_init_with_ddc()to fail, and the entire connector disappears from sysfson account of sysfs_create_link() failing.To fix that split the sysfs symlink stuff into separatefunctions (drm_sysfs_connector_add_late() anddrm_sysfs_connector_remove_early()) which are calledon the opposite side of the .later_register() and.early_unregister() hooks.Cc: Andrzej Pietrasiewicz Cc: Daniel Vetter Cc: Andrzej Hajda Cc: Emil Velikov Cc: Sam Ravnborg Cc: Neil Armstrong Signed-off-by: Ville Syrjälä Link: https://patchwork.freedesktop.org/patch/msgid/20230829113920.13713-3-ville.syrjala@linux.intel.comReviewed-by: Jani Nikula Acked-by: Thomas Zimmermann #irc List of files: /linux-6.15/drivers/gpu/drm/drm_internal.h

8e455145 - drm/debugfs: rework drm_debugfs_create_files implementation v2

Tue, 29 Aug 2023 11:01:15 +0000

drm/debugfs: rework drm_debugfs_create_files implementation v2Use managed memory allocation for this. That allows us to not keeptrack of all the files any more.v2: keep drm_debugfs_cleanup(), but rename to drm_debugfs_unregister(), we still need to cleanup the symlinkSigned-off-by: Christian König Link: https://patchwork.freedesktop.org/patch/msgid/20230829110115.3442-6-christian.koenig@amd.comReviewed-by: Andi Shyti List of files: /linux-6.15/drivers/gpu/drm/drm_internal.h

ec9c7073 - drm/debugfs: remove dev->debugfs_list and debugfs_mutex v2

Tue, 29 Aug 2023 11:01:14 +0000

drm/debugfs: remove dev->debugfs_list and debugfs_mutex v2The mutex was completely pointless in the first place since anyparallel adding of files to this list would result in randombehavior since the list is filled and consumed multiple times.Completely drop that approach and just create the files directly butreturn -ENODEV while opening the file when the minors are notregistered yet.v2: rebase on debugfs directory rework, limit access before minors are registered.Signed-off-by: Christian König Link: https://patchwork.freedesktop.org/patch/msgid/20230829110115.3442-5-christian.koenig@amd.comReviewed-by: Andi Shyti List of files: /linux-6.15/drivers/gpu/drm/drm_internal.h

0b30d57a - drm/debugfs: rework debugfs directory creation v5

Tue, 29 Aug 2023 11:01:13 +0000

drm/debugfs: rework debugfs directory creation v5Instead of the per minor directories only create a single debugfsdirectory for the whole device directly when the device is initialized.For DRM devices each minor gets a symlink to the per device directoryfor now until we can be sure that this isn't useful any more in any way.Accel devices create only the per device directory and also drops the midlayer callback to create driver specific files.v2: cleanup accel component as wellv3: fix typo when debugfs is disabledv4: call drm_debugfs_dev_fini() during release as well, some kerneldoc typos fixedv5: rebased and one more kerneldoc fixSigned-off-by: Christian König Link: https://patchwork.freedesktop.org/patch/msgid/20230829110115.3442-4-christian.koenig@amd.comReviewed-by: Andi Shyti List of files: /linux-6.15/drivers/gpu/drm/drm_internal.h

c7a47229 - drm/syncobj: add IOCTL to register an eventfd

Fri, 14 Jul 2023 11:13:03 +0000

drm/syncobj: add IOCTL to register an eventfdIntroduce a new DRM_IOCTL_SYNCOBJ_EVENTFD IOCTL which signals aneventfd from a syncobj.This is useful for Wayland compositors to handle wait-before-submit.Wayland clients can send a timeline point to the compositorbefore the point has materialized yet, then compositors can waitfor the point to materialize via this new IOCTL.The existing DRM_IOCTL_SYNCOBJ_TIMELINE_WAIT IOCTL is not suitablebecause it blocks. Compositors want to integrate the wait withtheir poll(2)-based event loop.Requirements for new uAPI:- User-space patch: https://gitlab.freedesktop.org/wlroots/wlroots/-/merge_requests/4262- IGT: https://lists.freedesktop.org/archives/igt-dev/2023-July/057893.htmlv2:- Wait for fence when flags is zero- Improve documentation (Pekka)- Rename IOCTL (Christian)- Fix typo in drm_syncobj_add_eventfd() (Christian)v3:- Link user-space + IGT patches- Add reference from overview docsv4: fix IOCTL number conflict with GETFB2 (Nicholas Choi, Vitaly Prosyak)Signed-off-by: Simon Ser Reviewed-by: Christian König Acked-by: Pekka Paalanen Cc: Jason Ekstrand Cc: Daniel Vetter Cc: Bas Nieuwenhuizen Cc: Daniel Stone Cc: James Jones Cc: Austin Shafer Cc: Vitaly Prosyak Link: https://patchwork.freedesktop.org/patch/msgid/20230714111257.11940-1-contact@emersion.fr List of files: /linux-6.15/drivers/gpu/drm/drm_internal.h

96a7b60f - drm: remove dumb_destroy callback

Thu, 26 Jan 2023 10:26:38 +0000

drm: remove dumb_destroy callbackNot used by any driver any more.Signed-off-by: Christian König Reviewed-by: Thomas Zimmermann Tested-by: Thomas Zimmermann Link: https://patchwork.freedesktop.org/patch/msgid/20230126102814.8722-2-christian.koenig@amd.com List of files: /linux-6.15/drivers/gpu/drm/drm_internal.h

dbb23cf5 - drm/debugfs: create debugfs late register functions

Mon, 19 Dec 2022 12:06:17 +0000

drm/debugfs: create debugfs late register functionsAlthough the device-centered debugfs functions can track requests forthe addition of DRM debugfs files at any time and have them added allat once during drm_dev_register(), they are not able to create debugfsfiles for modeset components, as they are registered after the primaryand the render drm_minor are registered.So, create a drm_debugfs_late_register() function, which is responsiblefor dealing with the creation of all the debugfs files for modesetcomponents at once. Therefore, the functions drm_debugfs_add_file()and drm_debugfs_add_files() can be used in late_register hooks.Signed-off-by: Maíra Canal Reviewed-by: Melissa Wen Acked-by: Daniel Vetter Signed-off-by: Maíra Canal Link: https://patchwork.freedesktop.org/patch/msgid/20221219120621.15086-4-mcanal@igalia.com List of files: /linux-6.15/drivers/gpu/drm/drm_internal.h

4979524f - drm: Fix potential null-ptr-deref in drm_vblank_destroy_worker()

Tue, 01 Nov 2022 07:07:16 +0000

drm: Fix potential null-ptr-deref in drm_vblank_destroy_worker()drm_vblank_init() call drmm_add_action_or_reset() withdrm_vblank_init_release() as action. If __drmm_add_action() failed, willdirectly call drm_vblank_init_release() with the vblank whose worker isNULL. As the resule, a null-ptr-deref will happen inkthread_destroy_worker(). Add the NULL check before callingdrm_vblank_destroy_worker().BUG: null-ptr-derefKASAN: null-ptr-deref in range [0x0000000000000068-0x000000000000006f]CPU: 5 PID: 961 Comm: modprobe Not tainted 6.0.0-11331-gd465bff130bf-dirtyRIP: 0010:kthread_destroy_worker+0x25/0xb0 Call Trace: drm_vblank_init_release+0x124/0x220 [drm] ? drm_crtc_vblank_restore+0x8b0/0x8b0 [drm] __drmm_add_action_or_reset+0x41/0x50 [drm] drm_vblank_init+0x282/0x310 [drm] vkms_init+0x35f/0x1000 [vkms] ? 0xffffffffc4508000 ? lock_is_held_type+0xd7/0x130 ? __kmem_cache_alloc_node+0x1c2/0x2b0 ? lock_is_held_type+0xd7/0x130 ? 0xffffffffc4508000 do_one_initcall+0xd0/0x4f0 ... do_syscall_64+0x35/0x80 entry_SYSCALL_64_after_hwframe+0x46/0xb0Fixes: 5e6c2b4f9161 ("drm/vblank: Add vblank works")Signed-off-by: Shang XiaoJing Reviewed-by: Lyude Paul Signed-off-by: Lyude Paul Link: https://patchwork.freedesktop.org/patch/msgid/20221101070716.9189-3-shangxiaojing@huawei.com List of files: /linux-6.15/drivers/gpu/drm/drm_internal.h

ea2aa97c - drm/gem: Fix GEM handle release errors

Fri, 19 Aug 2022 07:28:34 +0000

drm/gem: Fix GEM handle release errorsCurrently we are assuming a one to one mapping between dmabuf andGEM handle when releasing GEM handles.But that is not always true, since we would create extra handles for theGEM obj in cases like gem_open() and getfb{,2}().A similar issue was reported at:https://lore.kernel.org/all/20211105083308.392156-1-jay.xu@rock-chips.com/Another problem is that the imported dmabuf might not always havegem_obj->dma_buf set, which would cause leaks indrm_gem_remove_prime_handles().Let's fix these for now by using handle to find the exact map to remove.Signed-off-by: Jeffy Chen Reviewed-by: Christian König Signed-off-by: Christian König Link: https://patchwork.freedesktop.org/patch/msgid/20220819072834.17888-1-jeffy.chen@rock-chips.com List of files: /linux-6.15/drivers/gpu/drm/drm_internal.h

7938f421 - dma-buf-map: Rename to iosys-map

Fri, 04 Feb 2022 17:05:41 +0000

dma-buf-map: Rename to iosys-mapRename struct dma_buf_map to struct iosys_map and corresponding APIs.Over time dma-buf-map grew up to more functionality than the one used bydma-buf: in fact it's just a shim layer to abstract system memory, thatcan be accessed via regular load and store, from IO memory that needs tobe acessed via arch helpers.The idea is to extend this API so it can fulfill other needs, internalto a single driver. Example: in the i915 driver it's desired to sharethe implementation for integrated graphics, which uses mostly systemmemory, with discrete graphics, which may need to access IO memory.The conversion was mostly done with the following semantic patch: @r1@ @@ - struct dma_buf_map + struct iosys_map @r2@ @@ ( - DMA_BUF_MAP_INIT_VADDR + IOSYS_MAP_INIT_VADDR | - dma_buf_map_set_vaddr + iosys_map_set_vaddr | - dma_buf_map_set_vaddr_iomem + iosys_map_set_vaddr_iomem | - dma_buf_map_is_equal + iosys_map_is_equal | - dma_buf_map_is_null + iosys_map_is_null | - dma_buf_map_is_set + iosys_map_is_set | - dma_buf_map_clear + iosys_map_clear | - dma_buf_map_memcpy_to + iosys_map_memcpy_to | - dma_buf_map_incr + iosys_map_incr ) @@ @@ - #include + #include Then some files had their includes adjusted and some comments wereupdate to remove mentions to dma-buf-map.Since this is not specific to dma-buf anymore, move the documentation tothe "Bus-Independent Device Accesses" section.v2: - Squash patchesv3: - Fix wrong removal of dma-buf.h from MAINTAINERS - Move documentation from dma-buf.rst to device-io.rstv4: - Change documentation title and levelSigned-off-by: Lucas De Marchi Acked-by: Christian König Acked-by: Sumit Semwal Acked-by: Thomas Zimmermann Link: https://patchwork.freedesktop.org/patch/msgid/20220204170541.829227-1-lucas.demarchi@intel.com List of files: /linux-6.15/drivers/gpu/drm/drm_internal.h