- 04 Jan, 2017 4 commits
-
-
Daniel Vetter authored
When writing the generic nonblocking commit code I assumed that through clever lifetime management I can assure that the completion (stored in drm_crtc_commit) only gets freed after it is completed. And that worked. I also wanted to make nonblocking helpers resilient against driver bugs, by having timeouts everywhere. And that worked too. Unfortunately taking boths things together results in oopses :( Well, at least sometimes: What seems to happen is that the drm event hangs around forever stuck in limbo land. The nonblocking helpers eventually time out, move on and release it. Now the bug I tested all this against is drivers that just entirely fail to deliver the vblank events like they should, and in those cases the event is simply leaked. But what seems to happen, at least sometimes, on i915 is that the event is set up correctly, but somohow the vblank fails to fire in time. Which means the event isn't leaked, it's still there waiting for eventually a vblank to fire. That tends to happen when re-enabling the pipe, and then the trap springs and the kernel oopses. The correct fix here is simply to refcount the crtc commit to make sure that the event sticks around even for drivers which only sometimes fail to deliver vblanks for some arbitrary reasons. Since crtc commits are already refcounted that's easy to do. References: https://bugs.freedesktop.org/show_bug.cgi?id=96781 Cc: Jim Rees <rees@umich.edu> Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Jani Nikula <jani.nikula@linux.intel.com> Reviewed-by:
Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Signed-off-by:
Daniel Vetter <daniel.vetter@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/20161221102331.31033-1-daniel.vetter@ffwll.ch
-
Arvind Yadav authored
Here, pci_iomap can fail, handle this case and return -ENOMEM. Signed-off-by:
Arvind Yadav <arvind.yadav.cs@gmail.com> Signed-off-by:
Daniel Vetter <daniel.vetter@ffwll.ch> Link: http://patchwork.freedesktop.org/patch/msgid/1483443027-13444-1-git-send-email-arvind.yadav.cs@gmail.com
-
Gabriel Krisman Bertazi authored
v2: - Replace discouraged with deprecated - Link to new initialization/teardown functions Signed-off-by:
Gabriel Krisman Bertazi <krisman@collabora.co.uk> Signed-off-by:
-
Randy Dunlap authored
Fix build errors in nouveau driver when CONFIG_LEDS_CLASS=m and CONFIG_DRM_NOUVEAU=y. If LEDS_CLASS is enabled, DRM_NOUVEAU is restricted to the same kconfig value as LEDS_CLASS. drivers/built-in.o: In function `nouveau_do_suspend': nouveau_drm.c:(.text+0x2030b1): undefined reference to `nouveau_led_suspend' drivers/built-in.o: In function `nouveau_do_resume': nouveau_drm.c:(.text+0x2034ca): undefined reference to `nouveau_led_resume' drivers/built-in.o: In function `nouveau_drm_unload': nouveau_drm.c:(.text+0x203a15): undefined reference to `nouveau_led_fini' drivers/built-in.o: In function `nouveau_drm_load': nouveau_drm.c:(.text+0x204423): undefined reference to `nouveau_led_init' BTW, this line in Kbuild: nouveau-$(CONFIG_LEDS_CLASS) += nouveau_led.o does nothing when CONFIG_LEDS_CLASS=m and CONFIG_DRM_NOUVEAU=y. Signed-off-by:
Randy Dunlap <rdunlap@infradead.org> Reported-by:
kbuild test robot <fengguang.wu@intel.com> Cc: Martin Peres <martin.peres@free.fr> Cc: Ben Skeggs <bskeggs@redhat.com> Signed-off-by:
-
- 02 Jan, 2017 2 commits
-
-
Jani Nikula authored
No reason not to be const. Reviewed-by:
Jani Nikula <jani.nikula@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/1482923186-22430-1-git-send-email-jani.nikula@intel.com
-
Daniel Vetter authored
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 ... Cc: Archit Taneja <architt@codeaurora.org> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by:
Archit Taneja <architt@codeaurora.org> [danvet: Remove spurious hunk that Archit spotted.] Signed-off-by:
-
- 30 Dec, 2016 24 commits
-
-
Daniel Vetter authored
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 Laurent: - Move misplaced doc change to the right patch. - Remove "DRM driver's", it's redundant. - Spotted 3 more places where where we could add prose reference with a real one. Cc: Laurent Pinchart <Laurent.pinchart@ideasonboard.com> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by:
Laurent Pinchart <laurent.pinchart@ideasonboard.com> Signed-off-by:
-
Daniel Vetter authored
If we store the fb funcs pointer, we can remove a bit of boilerplate. Also remove the _fbdev_ in the example code, since the fb_funcs->dirty callback has nothing to do with fbdev. It's a KMS feature, only used by the fbdev deferred_io support to implement flushing/upload. Cc: Noralf Trønnes <noralf@tronnes.org> Cc: Laurent Pinchart <laurent.pinchart@ideasonboard.com> Reviewed-by:
-
Daniel Vetter authored
I reported the include issue for tracepoints a while ago, but nothing seems to have happened. Now it bit us, since the drm_mm_print conversion was broken for armada. Fix it, so I can re-enable armada in the drm-misc build configs. v2: Rebase just the compile fix on top of Chris' build fix. Cc: Russell King <rmk+kernel@armlinux.org.uk> Cc: Chris Wilson <chris@chris-wilson.co.uk> Acked: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by:
-
Chris Wilson authored
Missed rename during v2 of b5c3714f ("drm/mm: Convert to drm_printer") Reported-by:
Chris Wilson <chris@chris-wilson.co.uk> Cc: Daniel Vetter <daniel.vetter@intel.com> Signed-off-by:
-
Chris Wilson authored
For a virtual device, drm_device.dev is NULL, so becareful not to dereference it unconditionally in core code such as drm_dev_register(). Fixes: 75f6dfe3 ("drm: Deduplicate driver initialization message") Signed-off-by:
-
Daniel Vetter authored
We forgot to clean this up when adding connector refcounting. Reviewed-by:
David Herrmann <dh.herrmann@gmail.com> Signed-off-by:
-
Daniel Vetter authored
Drive-by polish. Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Signed-off-by:
-
Daniel Vetter authored
The new cool is &struct foo (kernel-doc now copes with linebreaks), and structure members should be referenced using &foo.bar. Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by:
-
Daniel Vetter authored
I've forgotten to remove this when revamping the connector_list locking. Cc: seanpaul@chromium.org Reviewed-by:
Sean Paul <seanpaul@chromium.org> Signed-off-by:
-
Daniel Vetter authored
sed -e 's/\( \* .*\)struct &\([_a-z]*\)/\1\&struct \2/' -i Originally I wasnt a friend of this style because I thought a line-break between the "&struct" and "foo" part would break it. But a quick test shows that " * &struct \n * foo\n" works pefectly well with current kernel-doc. So time to mass-apply these changes! Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by:
-
Daniel Vetter authored
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). Cc: Sumit Semwal <sumit.semwal@linaro.org> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by:
-
Daniel Vetter authored
sed -e 's/\( \* .*\)struct &\([_a-z]*\)/\1\&struct \2/' -i Originally I wasnt a friend of this style because I thought a line-break between the "&struct" and "foo" part would break it. But a quick test shows that " * &struct \n * foo\n" works pefectly well with current kernel-doc. So time to mass-apply these changes! Cc: Sumit Semwal <sumit.semwal@linaro.org> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by:
-
Daniel Vetter authored
Added some boilerplate for the structs, documented members where they are relevant and plenty of markup for hyperlinks all over. And a few small wording polish. Note that the intro needs some more love after the DRM_MM_INSERT_* patch from Chris has landed. v2: Spelling fixes (Chris). v3: Use &struct foo instead of &foo structure (Chris). Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Reviewed-by:
-
Daniel Vetter authored
- Remove the outdated hunk about driver documentation which somehow got misplaced here in the split-up. - Collect all the testing&validation stuff together and give the CRC section a heading for prettier output. Cc: Tomeu Vizoso <tomeu.vizoso@collabora.com> Cc: Jani Nikula <jani.nikula@intel.com> Reviewed-by:
-
Gabriel Krisman Bertazi authored
ttm_global_reference was renamed to drm_global_reference. This updates the documentation to reflect that. While we are there, document the drm_global_reference API and update the initialization interface documentation. Signed-off-by:
-
Gabriel Krisman Bertazi authored
drm_ioctl_permit is exported but missed a kernel-doc style documentation. Signed-off-by:
-
Gabriel Krisman Bertazi authored
Signed-off-by:
-
Gabriel Krisman Bertazi authored
Several DRM drivers print the same initialization message right after drm_dev_register, so move that to common code. The exception is i915, which uses its own register handle, so let it keep its own message. Notice that this was tested only with Exynos, but looks simple enough for the other drivers. Signed-off-by:
-
Chris Wilson authored
Currently at the end of drm_core_init() we print [ 0.735185] [drm] Initialized which does not provide any user information and is only a breadcrumb for developers, so reduce it from info to debug. Signed-off-by:
-
Shawn Guo authored
Function rockchip_crtc_from_pipe() does the exactly same thing as what crtc helper drm_crtc_from_index() provides. Use the helper to save some code. Signed-off-by:
Shawn Guo <shawn.guo@linaro.org> Signed-off-by:
-
Shawn Guo authored
Function zx_find_crtc() does the exactly same thing as what crtc helper drm_crtc_from_index() provides. Use the helper to save some code. Signed-off-by:
-
Shawn Guo authored
It adds a crtc helper drm_crtc_from_index() to find the registered CRTC with a given index, just like drm_plane_from_index(). Signed-off-by:
-
Daniel Vetter authored
Including all drivers. I thought about keeping small compat functions to avoid having to change all drivers. But I really like the drm_printer idea, so figured spreading it more widely is a good thing. v2: Review from Chris: - Natural argument order and better name for drm_mm_print. - show_mm() macro in the selftest. Cc: Rob Clark <robdclark@gmail.com> Cc: Russell King <rmk+kernel@armlinux.org.uk> Cc: Alex Deucher <alexander.deucher@amd.com> Cc: Christian König <christian.koenig@amd.com> Cc: Lucas Stach <l.stach@pengutronix.de> Cc: Tomi Valkeinen <tomi.valkeinen@ti.com> Cc: Thierry Reding <thierry.reding@gmail.com> Cc: Jyri Sarha <jsarha@ti.com> Reviewed-by:
-
Daniel Vetter authored
Useful for dumping lots of data into dmesg, e.g. drm_mm. v2: Fixup export_symbol line, I misplaced a hunk (Chris). Cc: Rob Clark <robdclark@gmail.com> Reviewed-by:
-
- 29 Dec, 2016 1 commit
-
-
Daniel Vetter authored
Our style guide should only be the delta compared to the overall one. Cc: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by:
-
- 28 Dec, 2016 9 commits
-
-
Chris Wilson authored
A couple of parameters slipped through the kerneldoc net. Reported-by:
-
Daniel Vetter authored
Drivers need to take care. Motivated by a discussion between Mark and Rob on dri-devel. Cc: Mark yao <mark.yao@rock-chips.com> Cc: Rob Clark <robdclark@gmail.com> Reviewed-by:
-
Chris Wilson authored
Remove a superfluous helper as drm_mm_insert_node is equivalent to insert_node_in_range with a range of [0, U64_MAX]. Signed-off-by:
Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Signed-off-by:
-
Chris Wilson authored
mm->color_adjust() compares the hole with its neighbouring nodes. They only abutt before we restrict the hole, so we have to apply color_adjust before we apply the range restriction. Signed-off-by:
-
Chris Wilson authored
Insulate users from changes to the internal hole tracking within struct drm_mm_node by using an accessor for hole_follows. Signed-off-by:
-
Chris Wilson authored
Using mm->color_adjust makes the eviction scanner much tricker since we don't know the actual neighbours of the target hole until after it is created (after scanning is complete). To work out whether we need to evict the neighbours because they impact upon the hole, we have to then check the hole afterwards - requiring an extra step in the user of the eviction scanner when they apply color_adjust. v2: Massage kerneldoc. Signed-off-by:
-
Chris Wilson authored
Since we mandate a strict reverse-order of drm_mm_scan_remove_block() after drm_mm_scan_add_block() we can further simplify the list manipulations when generating the temporary scan-hole. v2: Highlight the games being played with the lists to track the scan holes without allocation. Signed-off-by:
-
Chris Wilson authored
For power-of-two alignments, we can avoid the 64bit divide and do a simple bitwise add instead. v2: s/alignment_mask/remainder_mask/ Signed-off-by:
-
Chris Wilson authored
Compute the minimal required hole during scan and only evict those nodes that overlap. This enables us to reduce the number of nodes we need to evict to the bare minimum. Signed-off-by:
-
