- 01 Aug, 2023 20 commits
-
-
Miquel Raynal authored
A very basic debugging rule when a device is connected for the first time is to access a read-only register which contains known data in order to ensure the communication protocol is properly working. This driver lacked any read helper which is often a critical piece for speeding-up bring-ups. Add a read helper and use it to verify the communication with the panel is working as soon as possible in order to inform the user early if this is not the case. As this panel may work with no MISO line, the check is discarded in this case. Upon error, we do not fail probing but just warn the user, in case the DT description would be lacking the Rx bus width (which is likely on old descriptions) in order to avoid breaking existing devices. Signed-off-by:
Miquel Raynal <miquel.raynal@bootlin.com> Acked-by:
Sam Ravnborg <sam@ravnborg.org> Acked-by:
Maxime Ripard <mripard@kernel.org> Reviewed-by:
Sebastian Reichel <sre@kernel.org> Tested-by: Sebastian Reichel <sre@kernel.org> # no MISO line Signed-off-by:
Neil Armstrong <neil.armstrong@linaro.org> Link: https://patchwork.freedesktop.org/patch/msgid/20230714013756.1546769-20-sre@kernel.org
-
Miquel Raynal authored
This panel from Emerging Display Technologies Corporation features an ST7789V2 LCD controller panel inside which is almost identical to what the Sitronix panel driver supports. In practice, the module physical size is specific, and experiments show that the display will malfunction if any of the following situation occurs: * Pixel clock is above 3MHz * Pixel clock is not inverted I could not properly identify the reasons behind these failures, scope captures show valid input signals. Signed-off-by:
-
Miquel Raynal authored
The Sitronix datasheet explains BIT(1) of the RGBCTRL register as the DOTCLK/PCLK edge used to sample the data lines: “0” The data is input on the positive edge of DOTCLK “1” The data is input on the negative edge of DOTCLK IOW, this bit implies a falling edge and not a high state. Correct the definition to ease the comparison with the datasheet. Signed-off-by: -
Miquel Raynal authored
The Sitronix controller expects 9-bit words, provide this as default at probe time rather than specifying this in each and every access. Signed-off-by:
-
Miquel Raynal authored
The ST7789V LCD controller supports regular SPI wiring, as well as no Rx data line at all. The operating system needs to know whether it can read registers from the device or not. Let's detail this specific design possibility by bounding the spi-rx-bus-width property. Signed-off-by:
Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org> Reviewed-by:
-
Miquel Raynal authored
The ST7789V LCD controller is also embedded in the ET028013DMA panel. Add a compatible string to describe this other panel. Signed-off-by:
-
Sebastian Reichel authored
UNI-T UTi260b has a Inanbo T28CP45TN89 v17 panel. I could not find proper documentation for the panel apart from a technical drawing, but according to the vendor U-Boot it is based on a Sitronix st7789v chip. I generated the init sequence by modifying the default one until proper graphics output has been seen on the device. Reviewed-by:
Michael Riesch <michael.riesch@wolfvision.net> Signed-off-by:
-
Sebastian Reichel authored
Add polarity information via mode and bus flags, so that they are no longer hardcoded and forward the information to the DRM stack. This is required for adding panels with different settings. Reviewed-by:
-
Sebastian Reichel authored
While the default panel uses invert mode, some panels require non-invert mode instead. Reviewed-by:
-
Sebastian Reichel authored
Add support for describing the media bus format in the panel configuration and expose that to userspace. Since both supported formats (RGB565 and RGB666) are using 6 bits per color also hardcode that information. Reviewed-by:
-
Sebastian Reichel authored
Move the panel size information to the mode struct, so that different panel sizes can be specified depending on the panel type. Reviewed-by:
-
Sebastian Reichel authored
Avoid hard-coding the default_mode and supply it from match data. One additional layer of abstraction has been introduced, which will be needed for specifying other panel information (e.g. bus flags) in the next steps. Reviewed-by:
-
Sebastian Reichel authored
Improve error handling in the probe routine, so that probe defer errors are captured in /sys/kernel/debug/devices_deferred Reviewed-by:
-
Sebastian Reichel authored
st7789v_spi_write initializes a message with just a single transfer, spi_sync_transfer can be used for that. Reviewed-by:
-
Sebastian Reichel authored
The reset pin might not be software controllable from the SoC, so make it optional. Reviewed-by:
-
Sebastian Reichel authored
ST7789V_COLMOD_RGB_FMT_18BITS and ST7789V_COLMOD_CTRL_FMT_18BITS are unused in favour of MIPI_DCS_PIXEL_FMT_18BIT, remove them. Reviewed-by:
-
Sebastian Reichel authored
SPI device drivers should also have a SPI ID table. Reviewed-by:
-
Sebastian Reichel authored
Add compatible value for Inanbo t28cp45tn89 and make reset GPIO non mandatory, since it might not be connected to the CPU. Reviewed-by:
Rob Herring <robh@kernel.org> Signed-off-by:
-
Sebastian Reichel authored
Shenzhen INANBO Electronic Technology Co., Ltd. manufacturers TFT/OLED LCD panels. Reviewed-by:
-
Arthur Grillo authored
The drm_exec tests where crashing[0] because of a null dereference. This is caused by a new access of the `driver` attribute of `struct drm_driver` on drm_gem_private_object_init(). Alloc the drm_device to fix that. [0] [15:05:24] ================== drm_exec (6 subtests) =================== [15:05:24] [PASSED] sanitycheck ^CERROR:root:Build interruption occurred. Cleaning console. [15:05:50] [ERROR] Test: drm_exec: missing expected subtest! [15:05:50] BUG: kernel NULL pointer dereference, address: 00000000000000b0 [15:05:50] #PF: supervisor read access in kernel mode [15:05:50] #PF: error_code(0x0000) - not-present page [15:05:50] PGD 0 P4D 0 [15:05:50] Oops: 0000 [#1] PREEMPT NOPTI [15:05:50] CPU: 0 PID: 23 Comm: kunit_try_catch Tainted: G N 6.4.0-rc7-02032-ge6303f32 #69 [15:05:50] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.2-1.fc37 04/01/2014 [15:05:50] RIP: 0010:drm_gem_private_object_init+0x60/0xc0 Fixes: e6303f32 ("drm: manager to keep track of GPUs VA mappings") Signed-off-by:
Arthur Grillo <arthurgrillo@riseup.net> Tested-by:
Danilo Krummrich <dakr@redhat.com> Acked-by:
Maíra Canal <mcanal@igalia.com> Signed-off-by:
-
- 31 Jul, 2023 20 commits
-
-
Dmitry Osipenko authored
Add sync object DRM UAPI support to VirtIO-GPU driver. Sync objects support is needed by native context VirtIO-GPU Mesa drivers, it also will be used by Venus and Virgl contexts. Reviewed-by; Emil Velikov <emil.velikov@collabora.com> Signed-off-by:
Dmitry Osipenko <dmitry.osipenko@collabora.com> Tested-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com> # amdgpu nctx Tested-by: Rob Clark <robdclark@gmail.com> # freedreno nctx Reviewed-by:
Rob Clark <robdclark@gmail.com> Acked-by:
Gurchetan Singh <gurchetansingh@chromium.org> Acked-by:
Gerd Hoffmann <kraxel@redhat.com> Link: https://patchwork.freedesktop.org/patch/msgid/20230416115237.798604-4-dmitry.osipenko@collabora.com
-
Thomas Zimmermann authored
Deferred-I/O generator macros generate callbacks for struct fb_ops that operate on memory ranges in I/O address space or system address space. Rename the macros to use the _IOMEM_ and _SYSMEM_ infixes of their underlying helpers. Adapt all users. No functional changes. Signed-off-by:
Thomas Zimmermann <tzimmermann@suse.de> Reviewed-by:
Helge Deller <deller@gmx.de> Link: https://patchwork.freedesktop.org/patch/msgid/20230729193157.15446-5-tzimmermann@suse.de
-
Thomas Zimmermann authored
Change the infix for fbdev's DMA-memory helpers from _DMA_ to _DMAMEM_. The helpers perform operations within DMA-able memory, but they don't perform DMA operations. Naming should make this clear. Adapt all users. No functional changes. Signed-off-by:
-
Thomas Zimmermann authored
Change the infix for fbdev's system-memory helpers from _SYS_ to _SYSMEM_. The helpers perform operations within system memory, but not on the state of the operating system itself. Naming should make this clear. Adapt all users. No functional changes. Suggested-by:
-
Thomas Zimmermann authored
Change the infix for fbdev's I/O-memory helpers from _IO_ to _IOMEM_ to distiguish them from other types of I/O, such as file operations. The helpers operate on memory ranges in the I/O address space and the naming should make this clear. Adapt all users. No functional changes. Suggested-by:
-
Marek Vasut authored
The connector type and pixel format are missing for this panel, add them to prevent various drivers from failing to determine either of those parameters. Fixes: 7ee933a1 ("drm/panel: simple: Add support for AUO T215HVN01") Signed-off-by:
Marek Vasut <marex@denx.de> Reviewed-by:
-
Guillaume La Roque authored
This driver support the Startek KD070FHFID015, which is a 7-inch TFT LCD display using MIPI DSI interface. Signed-off-by:
Guillaume La Roque <glaroque@baylibre.com> Reviewed-by:
Alexandre Mergnat <amergnat@baylibre.com> Signed-off-by:
-
Alexandre Mergnat authored
The Startek KD070FHFID015 is a 7-inch TFT LCD display with a resolution of 1024 x 600 pixels. Reviewed-by:
Conor Dooley <conor.dooley@microchip.com> Signed-off-by:
-
Matus Gajdos authored
Add support for TDO TL050HDV35-H1311A LCD panel. Signed-off-by:
Matus Gajdos <matuszpd@gmail.com> Reviewed-by:
-
Matus Gajdos authored
Add support for TDO TL050HDV35-H1311A LCD panel. Signed-off-by:
-
Dmitry Baryshkov authored
Specify bpc value for the powertip_ph800480t013_idf02 panel to stop drm code from complaining about unexpected bpc value (0). Signed-off-by:
Dmitry Baryshkov <dmitry.baryshkov@linaro.org> Reviewed-by:
-
Arnd Bergmann authored
The newly added driver only builds when DRM_DISPLAY_DP_HELPER is enabled: x86_64-linux-ld: drivers/gpu/drm/panel/panel-visionox-r66451.o: in function `visionox_r66451_enable': panel-visionox-r66451.c:(.text+0x105): undefined reference to `drm_dsc_pps_payload_pack' Select both CONFIG_DRM_DISPLAY_DP_HELPER and CONFIG_DRM_DISPLAY_HELPER to ensure the helper function is always available. Fixes: a6dfab27 ("drm/panel: Add driver for Visionox r66451 panel") Signed-off-by:
Arnd Bergmann <arnd@arndb.de> Reviewed-by:
Jessica Zhang <quic_jesszhan@quicinc.com> Signed-off-by:
-
Maxime Ripard authored
We've had a couple of tests that weren't really obvious, nor did they document what they were supposed to test. Document that to make it hopefully more obvious. Reviewed-by:
Javier Martinez Canillas <javierm@redhat.com> Reviewed-by:
Maíra Canal <mairacanal@riseup.net> Link: https://lore.kernel.org/r/20230728-kms-kunit-actions-rework-v3-11-952565ccccfe@kernel.orgSigned-off-by:
-
Maxime Ripard authored
Now that we have a helper that takes care of an atomic state allocation and cleanup, we can migrate to it to simplify our tests. Reviewed-by:
-
Maxime Ripard authored
The new helper to init the locking context allows to remove some boilerplate. Reviewed-by:
-
Maxime Ripard authored
The *_mock_device functions allocate a DRM device that needs to be released using drm_dev_unregister. Now that we have a kunit release action API, we can switch to it and don't require any kind of garbage collection from the caller. Reviewed-by:
-
Maxime Ripard authored
Calling drm_kunit_helper_free_device() to clean up the resources allocated by drm_kunit_helper_alloc_device() is now optional and not needed in most cases. Remove it. Reviewed-by:
-
Maxime Ripard authored
As we gain more tests, boilerplate to allocate an atomic state and free it starts to be there more and more as well. In order to reduce the allocation boilerplate, we can create a helper to create that atomic state, and call an action when the test is done. This will also clean up the exit path. Reviewed-by:
-
Maxime Ripard authored
As we get more and more tests, the locking context initialisation creates more and more boilerplate, both at creation and destruction. Let's create a helper that will allocate, initialise a context, and register kunit actions to clean up once the test is done. Reviewed-by:
-
Maxime Ripard authored
Calling drm_kunit_helper_free_device() to clean up the resources allocated by drm_kunit_helper_alloc_device() is now optional and not needed in most cases. Remove it. Reviewed-by:
-
