- 15 Jan, 2021 4 commits
-
-
Colin Ian King authored
Currently the kmalloc allocation for config is not being null checked and could potentially lead to a null pointer dereference. Fix this by adding the missing null check. Addresses-Coverity: ("Dereference null return value") Fixes: 2df7af93 ("drm/vkms: Add vkms_config type") Signed-off-by:Colin Ian King <colin.king@canonical.com> Reviewed-by:
Sumera Priyadarsini <sylphrenadin@gmail.com> Signed-off-by:
Melissa Wen <melissa.srw@gmail.com> Link: https://patchwork.freedesktop.org/patch/msgid/20210115130911.71073-1-colin.king@canonical.com
-
Nirmoy Das authored
Remove unused space_needed variable. Fixes: 453f617a ("drm/amdgpu: Resize BAR0 to the maximum available size, even if it doesn't cover VRAM") Signed-off-by:
Nirmoy Das <nirmoy.das@amd.com> Reviewed-by:
Alex Deucher <alexander.deucher@amd.com> Reviewed-by:
Christian König <christian.koenig@amd.com> Signed-off-by:
-
Thomas Zimmermann authored
The function vc4_prime_import_sg_table() is an otherwise empty wrapper around CMA's drm_gem_cma_prime_import_sg_table(). Removing it in favor of the latter allows to initialize vc4_drm_driver with CMA's initializer macro. Signed-off-by:
Thomas Zimmermann <tzimmermann@suse.de> Reviewed-by:
Daniel Vetter <daniel.vetter@ffwll.ch> Acked-by:
Maxime Ripard <mripard@kernel.org> Link: https://patchwork.freedesktop.org/patch/msgid/20210114084949.29014-1-tzimmermann@suse.de
-
Thomas Zimmermann authored
The file is not in use. It got re-added by a rebased patch. Removing it. Signed-off-by:
Tian Tao <tiantao6@hisilicon.com> Reported-by:
-
- 14 Jan, 2021 14 commits
-
-
Ville Syrjälä authored
The new >8k CEA modes have dotclocks reaching 5.94 GHz, which means our clock*1000 will now overflow the 32bit unsigned integer. Switch to 64bit maths to avoid it. Cc: stable@vger.kernel.org Reported-by:
Randy Dunlap <rdunlap@infradead.org> Signed-off-by:
Ville Syrjälä <ville.syrjala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20201022194256.30978-1-ville.syrjala@linux.intel.comTested-by:
Chris Wilson <chris@chris-wilson.co.uk>
-
Zack Rusin authored
We can't be setting the display_id register to an invalid value because that makes our device reset the fb which causes nasty flicker (due to destruction and creation of a new fb). Also we can't be using the BITS_PER_PIXEL register if the 8BIT_EMULATION is not supported. Signed-off-by:
Zack Rusin <zackr@vmware.com> Reviewed-by:
Martin Krastev <krastevm@vmware.com> Reviewed-by:
Roland Scheidegger <sroland@vmware.com> Link: https://patchwork.freedesktop.org/patch/414041/?series=85516&rev=2
-
Zack Rusin authored
Lets try to cleanup the usage of the term FIFO which we used for both our MMIO based cmd queue processing and for general command processing which could have been using command buffers interface. We're going to rename the functions which are processing commands (and work either via MMIO or command buffers) as _cmd_ and functions which operate on the MMIO based commands as FIFO to match the SVGA device naming. Signed-off-by:
-
Zack Rusin authored
Throttling was used before fencing to implement early vsync support in the xorg state tracker a long time ago. The xorg state tracker has been removed years ago and no one else has ever used throttling. It's time to remove this code, it hasn't been used or tested in years. Signed-off-by:
-
Zack Rusin authored
Instead of doing it in multiple spots lets centralize the code to handle pci resources. This also cleans up the error handling a bit and will make it a lot easier to add additional svga versions to the driver. Signed-off-by:
-
Zack Rusin authored
Going forward the svga device might reuse mmio for general register accesses, in order to prepare for that we need to cleanup our naming and handling of fifo specific mmio reads and writes. As part of this work lets switch to managed mapping of the fifo mmio to make the error handling cleaner. Signed-off-by:
-
Zack Rusin authored
To cleanup some of the error handling and prepare for some other work lets switch to a managed drm device. It will let us get a better handle on some of the error paths. Signed-off-by:
-
Zack Rusin authored
Before drm got helpers for removing conflicting pci framebuffer devices we implemented something known as "stealth" mode which allowed vmwgfx to run even if it couldn't reserve pci resources. We can just switch to regular drm helpers instead of keeping the stealth mode alive as it makes our code a lot cleaner. Signed-off-by:
-
Roland Scheidegger authored
Reviewed-by:
-
Sumera Priyadarsini authored
Fix typo in intro chapter in drm_vblank.c. Change 'sacn' to 'scan'. Signed-off-by:
-
Mauro Carvalho Chehab authored
A function has a different name between their prototype and its kernel-doc markup: ../include/drm/drm_crtc.h:1257: warning: expecting prototype for drm_crtc_alloc_with_planes(). Prototype was for drmm_crtc_alloc_with_planes() instead Signed-off-by:
Mauro Carvalho Chehab <mchehab+huawei@kernel.org> Acked-by:
Simon Ser <contact@emersion.fr> Signed-off-by:
-
Giulio Benetti authored
During commit 88bc4178 ("drm: Use new DRM_BUS_FLAG_*_(DRIVE|SAMPLE)_(POS|NEG)EDGE flags") DRM_BUS_FLAG_* macros have been changed to avoid ambiguity but just because of this ambiguity previous DRM_BUS_FLAG_PIXDATA_(POS/NEG)EDGE were used meaning _SAMPLE_ not _DRIVE_. This leads to DLCK inversion and need to fix but instead of swapping phase values, let's adopt an easier approach Maxime suggested: It turned out that bit 26 of SUN4I_TCON0_IO_POL_REG is dedicated to invert DCLK polarity and this makes things really easier than before. So let's handle DCLK polarity by adding SUN4I_TCON0_IO_POL_DCLK_DRIVE_NEGEDGE as bit 26 and activating according to bus_flags the same way it is done for all the other signals polarity. Fixes: 88bc4178 ("drm: Use new DRM_BUS_FLAG_*_(DRIVE|SAMPLE)_(POS|NEG)EDGE flags") Suggested-by:
Maxime Ripard <maxime@cerno.tech> Signed-off-by:
Giulio Benetti <giulio.benetti@micronovasrl.com> Signed-off-by:
-
Tian Tao authored
Fix the following errors: divers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c: In function ‘hibmc_hw_map’: drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c:213:25: error: ‘dev’ undeclared (first use in this function); Fixes: 4d4dad21 ("drm/hibmc: Remove references to struct drm_device.pdev") Signed-off-by:
-
Thomas Zimmermann authored
The function drm_need_swiotbl() needs mem_encrypt_active() from <linux/mem_encrypt.h>. The include got lost when refactoring the code recently. Signed-off-by:
kernel test robot <lkp@intel.com> Cc: Thomas Zimmermann <tzimmermann@suse.de> Cc: Christian König <christian.koenig@amd.com> Cc: Daniel Vetter <daniel.vetter@ffwll.ch> Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Maxime Ripard <mripard@kernel.org> Cc: David Airlie <airlied@linux.ie> Cc: Daniel Vetter <daniel@ffwll.ch> Cc: dri-devel@lists.freedesktop.org Link: https://patchwork.freedesktop.org/patch/msgid/20210114080535.17132-1-tzimmermann@suse.de
-
- 13 Jan, 2021 18 commits
-
-
Lukasz Luba authored
Devfreq framework supports 2 modes for monitoring devices. Use delayed timer as default instead of deferrable timer in order to monitor the GPU status regardless of CPU idle. Signed-off-by:
Lukasz Luba <lukasz.luba@arm.com> Reviewed-by:
Steven Price <steven.price@arm.com> Signed-off-by:
-
Thomas Zimmermann authored
The field is only relevant for legacy DRM drivers. Its only non-legacy user in the DRM core is in drm_file.c. This code is now protected by CONFIG_DRM_LEGACY. Radeon, the only driver that used the field, has been changed to maintain it's own copy. Signed-off-by:
-
Thomas Zimmermann authored
Moves struct drm_device.hose into struct radeon_device. The field in struct DRM device is only for legacy drivers. Signed-off-by:
-
Thomas Zimmermann authored
CONFIG_DRM_VM gets selected by CONFIG_DRM_LEGACY, but nothing else. So remove it and build drm_vm.o as part of CONFIG_DRM_LEGACY. Signed-off-by:
-
Thomas Zimmermann authored
The file contains I/O-memory functions that are only used by legacy drivers. Signed-off-by:
-
Thomas Zimmermann authored
The function is declared in drm_cache.h. I also removed the curly braces from the for loop to adhere to kernel coding style. Signed-off-by:
-
Thomas Zimmermann authored
The AGP wrapper functions serve no purpose. They used to handle builds that have CONFIG_AGP unset. But their callers are all in drm_agpsupport.c, which only gets build with CONFIG_AGP. v2: * clarify CONFIG_AGP in commit description (Daniel) Signed-off-by:
-
Thomas Zimmermann authored
Moving vc4's mmap code from vc4_mmap() into a GEM object function allows for the use drm_gem_mmap() and drm_gem_prime_mmap(). The content of vc4_drm_fpos can then be generated by DEFINE_DRM_GEM_FOPS(). The actual mmap implementation is just a check if the BO is a validated shader plus the default CMA mmap code. Signed-off-by:
-
Thomas Zimmermann authored
Rearrange the code to make BO functions static. This will also help with streamlining the BO's mmap implementation. Signed-off-by:
-
Thomas Zimmermann authored
Validated shaders cannot be exported. There's no need for testing this in the BO's vmap implementation. Call drm_gem_cma_vmap() directly instead. Signed-off-by:
-
Jernej Skrabec authored
DE3 supports 10-bit formats, so it's only naturally to also support BT2020 encoding. Add support for it. Signed-off-by:
Jernej Skrabec <jernej.skrabec@siol.net> Signed-off-by:
-
Jernej Skrabec authored
YUV to RGB matrices are almost identical to YVU to RGB matrices. They only have second and third column reversed. Do that reversion in code in order to lower amount of static data and redundancy. Signed-off-by:
-
Jernej Skrabec authored
Rework DE3 CSC macros to take just one coordinate instead of two. This will make its usage easier in subsequent commit. Signed-off-by:
-
Sebastian Reichel authored
Standard DRM panel driver for DSI command mode panel used by omapfb2 is also available now. Just like the other panels its module name clashes with the module from drivers/video/fbdev/omap2/omapfb/displays, part of the deprecated omapfb2 fbdev driver. As omapfb2 can only be compiled when the omapdrm driver is disabled, and the DRM panel drivers are useless in that case, make the omapfb2 panel depend on the standard DRM panels being disabled to fix the name clash. Fixes: cf64148a ("drm/panel: Move OMAP's DSI command mode panel driver") Reported-by:
Stephen Rothwell <sfr@canb.auug.org.au> Signed-off-by:
Sebastian Reichel <sebastian.reichel@collabora.com> Acked-by:
Sam Ravnborg <sam@ravnborg.org> Signed-off-by:
Tomi Valkeinen <tomi.valkeinen@ideasonboard.com> Link: https://patchwork.freedesktop.org/patch/msgid/20210108112441.14609-1-sebastian.reichel@collabora.com
-
Thomas Zimmermann authored
Using struct drm_device.pdev is deprecated. Convert nouveau to struct drm_device.dev. No functional changes. v3: * fix nv04_dfp_update_backlight() as well (Jeremy) Signed-off-by:
Jeremy Cline <jcline@redhat.com> Cc: Ben Skeggs <bskeggs@redhat.com> Link: https://patchwork.freedesktop.org/patch/msgid/20210107080748.4768-8-tzimmermann@suse.de
-
Thomas Zimmermann authored
Using struct drm_device.pdev is deprecated. Convert hibmc to struct drm_device.dev. No functional changes. v3: * rebased Signed-off-by:
-
Thomas Zimmermann authored
Using struct drm_device.pdev is deprecated. Convert amdgpu to struct drm_device.dev. No functional changes. v3: * rebased Signed-off-by:
-
Thomas Zimmermann authored
Adhere to kernel coding style. Signed-off-by:
-
- 12 Jan, 2021 4 commits
-
-
Sumera Priyadarsini authored
Update vkms documentation to contain usage of `modinfo` command and steps to load vkms with module options enabled. Signed-off-by:
-
Sumera Priyadarsini authored
Add enable_writeback feature to vkms_config as a module. Signed-off-by:
-
Sumera Priyadarsini authored
Currently, data for the device instance is held by vkms_device. Add a separate type, vkms_config to contain configuration details for the device and various modes to be later used by configfs. This config data stays constant once the device is created. Accordingly, add vkms_create and vkms_destroy to initialize/destroy device through configfs. Currently, they are being called from vkms_init and vkms_exit, but will be evoked from configfs later on. When configfs is added, device configuration will be tracked by configfs and only vkms device lifetime will be handled by vkms_init and vkms_exit functions. Modify usage of enable_cursor feature to reflect the changes in relevant files. Co-developed-by:
Daniel Vetter <danvet.vetter@ffwl.ch> Signed-off-by:
-
Simon Ser authored
The comment says the layout and options use 8 bits, and the shift uses 8 bits. However the mask is 0xf, ie. 0b00001111 (4 bits). This could be surprising when introducing new layouts or options that take more than 4 bits, as this would silently drop the high bits. Make the masks consistent with the comment and the shift. Found when writing a drm_info patch [1]. [1]: https://github.com/ascent12/drm_info/pull/67Signed-off-by:
Neil Armstrong <narmstrong@baylibre.com> Link: https://patchwork.freedesktop.org/patch/msgid/20210110125103.15447-1-contact@emersion.fr
-
