kernel/drivers/gpu/drm/drm_modeset_lock.c, branch linux-5.1.y

drm/atomic: integrate modeset lock with private objects

2018-12-11T14:24:30Z

Follow the same pattern of locking as with other state objects. This avoids boilerplate in the driver. Signed-off-by: Rob Clark Signed-off-by: Boris Brezillon Reviewed-by: Daniel Vetter Link: https://patchwork.freedesktop.org/patch/msgid/20181022123122.30468-1-boris.brezillon@bootlin.com

drm: Add DRM_MODESET_LOCK_BEGIN/END helpers

2018-11-29T15:48:31Z

This patch adds a couple of helpers to remove the boilerplate involved in grabbing all of the modeset locks. I've also converted the obvious cases in drm core to use the helpers. The only remaining instance of drm_modeset_lock_all_ctx() is in drm_framebuffer. It's complicated by the state clear that occurs on deadlock. ATM, there's no way to inject code in the deadlock path with the helpers, so it's unfit for conversion. Changes in v2: - Relocate ret argument to the end of the list (Daniel) - Incorporate Daniel's doc suggestions (Daniel) Suggested-by: Daniel Vetter Reviewed-by: Daniel Vetter Signed-off-by: Sean Paul Link: https://patchwork.freedesktop.org/patch/msgid/20181129150423.239081-4-sean@poorly.run

drm/atomic: Call ww_acquire_done after drm_modeset_lock_all

2018-03-05T09:35:32Z

After we acquired all generic modeset locks in drm_modeset_lock_all, it's unsafe acquire any other so just mark acquisition as done. Atomic drivers shouldn't use drm_modeset_lock_all. Signed-off-by: Maarten Lankhorst Link: https://patchwork.freedesktop.org/patch/msgid/20180221152331.9212-1-maarten.lankhorst@linux.intel.com Reviewed-by: Daniel Vetter Acked-by: Harry Wentland

drm: Require __GFP_NOFAIL for the legacy drm_modeset_lock_all

2017-10-31T16:36:46Z

To acquire all modeset locks requires a ww_ctx to be allocated. As this is the legacy path and the allocation small, to reduce the changes required (and complex untested error handling) to the legacy drivers, we simply assume that the allocation succeeds. At present, it relies on the too-small-to-fail rule, but syzbot found that by injecting a failure here we would hit the WARN. Document that this allocation must succeed with __GFP_NOFAIL. Signed-off-by: Chris Wilson Cc: Daniel Vetter Reported-by: syzbot Signed-off-by: Daniel Vetter Reviewed-by: Ville Syrjälä Link: https://patchwork.freedesktop.org/patch/msgid/20171031115535.15166-1-chris@chris-wilson.co.uk

drm/atomic: Prepare drm_modeset_lock infrastructure for interruptible waiting, v2.

2017-09-13T07:50:52Z

When we want to make drm_atomic_commit interruptible, there are a lot of places that call the lock function, which we don't have control over. Rather than trying to convert every single one, it's easier to toggle interruptible waiting per acquire_ctx. If drm_modeset_acquire_init is called with DRM_MODESET_ACQUIRE_INTERRUPTIBLE, then we will perform interruptible waits in drm_modeset_lock and drm_modeset_backoff. Changes since v1: - Fix locking example in drm_modeset_lock.c to be compatible with interruptible waiting (xexaxo) and make it default. Uninterruptible waiting shouldn't happen except in corner cases, but the example will still apply if the flag is removed. - Add drm_modeset_lock_single_interruptible() to documentation. - Fix dead link to removed drm_modeset_lock_interruptible() in drm_modeset_lock(). Signed-off-by: Maarten Lankhorst Reviewed-by: Daniel Vetter #v1 Cc: Emil Velikov Link: https://patchwork.freedesktop.org/patch/msgid/20170912133749.6532-2-maarten.lankhorst@linux.intel.com Reviewed-by: Emil Velikov

drm: Improve kerneldoc for drm_modeset_lock

2017-07-26T11:45:08Z

Explain better when the drm_modeset_acquire_ctx parameter can be skipped for drm_modeset_lock() call. Signed-off-by: Liviu Dudau Cc: Daniel Vetter Signed-off-by: Daniel Vetter Link: https://patchwork.freedesktop.org/patch/msgid/20170720160748.12856-1-Liviu.Dudau@arm.com Signed-off-by: Daniel Vetter

drm: Remove drm_modeset_legacy_acquire_ctx and crtc->acquire_ctx

2017-04-05T07:26:45Z

With all the callers of drm_modeset_lock_crtc gone, and all the places it was formerly used properly wiring the acquire ctx through, we can remove this. The only hidden context magic we still have is now the global one. Reviewed-by: Harry Wentland Reviewed-by: Maarten Lankhorst Signed-off-by: Daniel Vetter Link: http://patchwork.freedesktop.org/patch/msgid/20170403083304.9083-4-daniel.vetter@ffwll.ch

drm: Remove drm_modeset_(un)lock_crtc

2017-04-05T07:26:07Z

The last user, the cursor ioctl, can just open-code this too. We simply have to move the acquire ctx dance from the universal function up into the top-level ioctl handler. Reviewed-by: Harry Wentland Reviewed-by: Maarten Lankhorst Signed-off-by: Daniel Vetter Link: http://patchwork.freedesktop.org/patch/msgid/20170403083304.9083-3-daniel.vetter@ffwll.ch

drm: Make drm_modeset_lock_crtc internal

2017-04-05T07:25:37Z

This is only for legacy paths that need to grab the crtc/plane lock combo. If you want to lock a crtc, just use drm_modeset_lock(). Reviewed-by: Harry Wentland Reviewed-by: Maarten Lankhorst Signed-off-by: Daniel Vetter Link: http://patchwork.freedesktop.org/patch/msgid/20170403083304.9083-2-daniel.vetter@ffwll.ch

drm/kms-core: Use recommened kerneldoc for struct member refs

2017-01-25T15:30:34Z

I just learned that &struct_name.member_name works and looks pretty even. It doesn't (yet) link to the member directly though, which would be really good for big structures or vfunc tables (where the per-member kerneldoc tends to be long). Also some minor drive-by polish where it makes sense, I read a lot of docs ... v2: Review from Eric. Cc: Jani Nikula Cc: Chris Wilson Reviewed-by: Eric Engestrom Signed-off-by: Daniel Vetter Link: http://patchwork.freedesktop.org/patch/msgid/20170125062657.19270-4-daniel.vetter@ffwll.ch