- 21 Apr, 2022 3 commits
-
-
Zack Rusin authored
Switch to using the TTM resource manager debugfs helpers. It's exactly the same functionality but the debugfs code is shared with other drivers. The TTM resource managers need to stay valid for as long as the drm debugfs_root is valid. Signed-off-by:
Zack Rusin <zackr@vmware.com> Cc: Alex Deucher <alexander.deucher@amd.com> Cc: "Pan, Xinhui" <Xinhui.Pan@amd.com> Cc: David Airlie <airlied@linux.ie> Cc: Daniel Vetter <daniel@ffwll.ch> Cc: Felix Kuehling <Felix.Kuehling@amd.com> Cc: Nirmoy Das <nirmoy.das@amd.com> Cc: Thomas Zimmermann <tzimmermann@suse.de> Cc: amd-gfx@lists.freedesktop.org Link: https://patchwork.freedesktop.org/patch/msgid/20220412033526.369115-4-zack@kde.orgReviewed-by:
Christian König <christian.koenig@amd.com>
-
Zack Rusin authored
Use the newly added TTM's ability to automatically create debugfs entries for specified placements. This creates debugfs files that can be read to get information about various TTM resource managers which are used by vmwgfx. Signed-off-by:
-
Zack Rusin authored
Drivers duplicate the code required to add debugfs entries for various ttm resource managers. To fix it add common TTM resource manager debugfs code that each driver can reuse. Specific resource managers can overwrite ttm_resource_manager_func::debug to get more information from those debugfs entries. Signed-off-by:
-
- 20 Apr, 2022 8 commits
-
-
Javier Martinez Canillas authored
The ssd130x driver only provides the core support for these devices but it does not have any bus transport logic. Add a driver to interface over SPI. There is a difference in the communication protocol when using 4-wire SPI instead of I2C. For the latter, a control byte that contains a D/C# field has to be sent. This field tells the controller whether the data has to be written to the command register or to the graphics display data memory. But for 4-wire SPI that control byte is not used, instead a real D/C# line must be pulled HIGH for commands data and LOW for graphics display data. For this reason the standard SPI regmap can't be used and a custom .write bus handler is needed. Signed-off-by:
Javier Martinez Canillas <javierm@redhat.com> Acked-by:
Mark Brown <broonie@kernel.org> Reviewed-by:
Geert Uytterhoeven <geert+renesas@glider.be> Link: https://patchwork.freedesktop.org/patch/msgid/20220419214824.335075-6-javierm@redhat.com
-
Javier Martinez Canillas authored
These are declared in the ssd130x-i2c transport driver but the information is not I2C specific, and could be used by other SSD130x transport drivers. Move them to the ssd130x core driver and just set the OF device entries to an ID that could be used to lookup the correct device info from an array. While being there, also move the SSD130X_DATA and SSD130X_COMMAND control bytes. Since even though they are used by the I2C interface, they could also be useful for other transport protocols such as SPI. Suggested-by:
Chen-Yu Tsai <wens@kernel.org> Signed-off-by:
-
Javier Martinez Canillas authored
The current compatible strings for SSD130x I2C controllers contain an "fb" and "-i2c" suffixes. These have been deprecated and more correct ones were added, that don't encode a subsystem or bus used to interface the devices. Signed-off-by:
-
Javier Martinez Canillas authored
The Solomon SSD130x OLED displays can either have an I2C or SPI interface, add to the schema the properties and examples for OLED devices under SPI. Signed-off-by:
Rob Herring <robh@kernel.org> Link: https://patchwork.freedesktop.org/patch/msgid/20220419214824.335075-3-javierm@redhat.com
-
Javier Martinez Canillas authored
The current compatible strings for SSD130x I2C controllers contain both an "fb" and "-i2c" suffixes. It seems to indicate that are for a fbdev driver and also that are for devices that can be accessed over an I2C bus. But a DT is supposed to describe the hardware and not Linux implementation details. So let's deprecate those compatible strings and add new ones that only contain the vendor and device name, without any of these suffixes. These will just describe the device and can be matched by both I2C and SPI DRM drivers. The required properties should still be enforced for old ones. While being there, just drop the "sinowealth,sh1106-i2c" compatible string since that was never present in a released Linux version. Signed-off-by:
-
Thomas Zimmermann authored
A workaround makes fbdev hot-unplugging work for framebuffers without device. The only user for this feature was offb. As each OF framebuffer now has an associated platform device, the workaround hould no longer be triggered. Update it with a warning and rewrite the comment. Fbdev drivers that trigger the hot-unplug workaround really need to be fixed. Signed-off-by:
Thomas Zimmermann <tzimmermann@suse.de> Reviewed-by:
-
Thomas Zimmermann authored
Create a platform device for each OF-declared framebuffer and have offb bind to these devices. Allows for real hot-unplugging and other drivers besides offb. Originally, offb created framebuffer devices while initializing its module by parsing the OF device tree. No actual Linux device was set up. This tied OF framebuffers to offb and makes writing other drivers for the OF framebuffers complicated. The absence of a Linux device further prevented real hot-unplugging. Adding a distinct platform device for each OF framebuffer solves both problems. Specifically, a DRM driver can now provide graphics output for modern userspace. Some of the offb init code is now located in the OF initialization. There's now also an implementation of of_platform_default_populate_init(), which was missing before. The OF side creates different devices for either OF display nodes or BootX displays as they require different handling by the driver. The offb drivers picks up each type of device and runs the appropriate fbdev initialization. Tested with OF display nodes on qemu's ppc64le target. v3: * declare variable 'node' with function scope (Rob) v2: * run PPC code as part of existing initialization (Rob) * add a few more error warnings (Javier) Signed-off-by:
-
Xiaomeng Tong authored
Instead of exiting the loop as expected when an entry is found, the list_for_each_entry() continues until the traversal is complete. To avoid potential executing 'ret = gma_backlight_init(dev);' repeatly, goto outside the loop when found entry by replacing switch/case with if statement. Signed-off-by:
Xiaomeng Tong <xiam0nd.tong@gmail.com> Signed-off-by:
Patrik Jakobsson <patrik.r.jakobsson@gmail.com> [Fixed indentation] Link: https://patchwork.freedesktop.org/patch/msgid/20220413051105.5612-1-xiam0nd.tong@gmail.com
-
- 19 Apr, 2022 13 commits
-
-
Marek Vasut authored
The driver currently hard-codes DSI lane count to two, however the chip is capable of operating in 1..4 DSI lanes mode. Parse 'data-lanes' DT property and program the result into DSI_CTRL register. Signed-off-by:
Marek Vasut <marex@denx.de> Cc: Jagan Teki <jagan@amarulasolutions.com> Cc: Laurent Pinchart <laurent.pinchart@ideasonboard.com> Cc: Maxime Ripard <maxime@cerno.tech> Cc: Robert Foss <robert.foss@linaro.org> Cc: Sam Ravnborg <sam@ravnborg.org> Cc: Thomas Zimmermann <tzimmermann@suse.de> To: dri-devel@lists.freedesktop.org Signed-off-by:
Robert Foss <robert.foss@linaro.org> Link: https://patchwork.freedesktop.org/patch/msgid/20220407185617.179573-2-marex@denx.de
-
Marek Vasut authored
It is necessary to specify the number of connected/used DSI data lanes when using the DSI input port of this bridge. Document the 'data-lanes' property of the DSI input port. Signed-off-by:
-
Nícolas F. R. A. Prado authored
As defined in the anx7625 dt-binding, the analogix,lane0-swing and analogix,lane1-swing properties are uint8 arrays. Yet, the driver was reading the array as if it were of uint32 and masking to 8-bit before writing to the registers. This means that a devicetree written in accordance to the dt-binding would have its values incorrectly parsed. Fix the issue by reading the array as uint8 and storing them as uint8 internally, so that we can also drop the masking when writing the registers. Fixes: fd0310b6 ("drm/bridge: anx7625: add MIPI DPI input feature") Signed-off-by:
Nícolas F. R. A. Prado <nfraprado@collabora.com> Reviewed-by:
AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com> Signed-off-by:
-
Sandor Yu authored
General Parallel Audio (GPA) interface is one of the supported audio interface for synopsys HDMI module, which has verified for i.MX8MPlus platform. This is initial version for GPA. Signed-off-by:
Shengjiu Wang <shengjiu.wang@nxp.com> Signed-off-by:
Sandor Yu <Sandor.yu@nxp.com> Reviewed-by:
Neil Armstrong <narmstrong@baylibre.com> Signed-off-by:
-
Sandor Yu authored
PHY reset register(MC_PHYRSTZ) active high reset control for PHY GEN2, and active low reset control for PHY GEN1. Rename function dw_hdmi_phy_reset to dw_hdmi_phy_gen2_reset. Add dw_hdmi_phy_gen1_reset function for PHY GEN1. Signed-off-by:
Jernej Skrabec <jernej.skrabec@gmail.com> Signed-off-by:
-
Sandor Yu authored
HDMI1.4b specification section 6.5.3: Source shall only send GCPs with non-zero CD to sinks that indicate support for Deep Color. DW HDMI GCP default enabled, but only transmit CD and do not handle AVMUTE, PP norDefault_Phase (yet). Disable Auto GCP when 24-bit color for sinks that not support Deep Color. Signed-off-by:
-
Sandor Yu authored
i.MX8MPlus (v2.13a) has verified need the workaround to clear the overflow with one iteration. Only i.MX6Q(v1.30a) need the workaround with 4 iterations, the others versions later than v1.3a have been identified as needing the workaround with a single iteration. Default enable the workaround with one iteration for all versions later than v1.30a. Signed-off-by:
-
Marek Szyprowski authored
If panel_bridge_attach() happens after DRM device registration, the created connector will not be registered by the DRM core anymore. Fix this by registering it explicitly in such case. This fixes the following issue observed on Samsung Exynos4210-based Trats board with a DSI panel (the panel driver is registered after the Exynos DRM component device is bound): $ ./modetest -c -Mexynos could not get connector 56: No such file or directory Segmentation fault While touching this, move the connector reset() call also under the DRM device registered check, because otherwise it is not really needed. Fixes: 934aef88 ("drm: bridge: panel: Reset the connector state pointer") Signed-off-by:
Marek Szyprowski <m.szyprowski@samsung.com> Reviewed-by:
Jagan Teki <jagan@amarulasolutions.com> Signed-off-by:
-
Biju Das authored
Connector detection using poll method won't work in case of bridge attached to the encoder with the flag DRM_BRIDGE_ATTACH_NO_CONNECTOR, as the code defaults to HPD. Enable DRM_BRIDGE_OP_HPD based on HPD interrupt availability, so that it will fall back to polling, if HPD is not available. Signed-off-by:
Biju Das <biju.das.jz@bp.renesas.com> Reviewed-by:
-
Marek Vasut authored
Add driver for Lontium LT9211 Single/Dual-Link DSI/LVDS or Single DPI to Single-link/Dual-Link DSI/LVDS or Single DPI bridge. This chip is highly capable at converting formats, but sadly it is also highly undocumented. This driver is written without any documentation from Lontium and based only on shreds of information available in various obscure example codes, hence long runs of unknown register patches and lengthy delays in various places. Whichever register meaning could be divined from its behavior has at least a comment around it. Currently the only mode tested is Single-link DSI to Single-link LVDS. Dual-link LVDS might work as well, the register programming is in place, but is untested. Reviewed-by:
-
Marek Vasut authored
Add bindings for Lontium LT9211 Single/Dual-Link DSI/LVDS or Single DPI to Single-link/Dual-Link DSI/LVDS or Single DPI bridge. This chip is highly capable at converting formats, but sadly it is also highly undocumented. Reviewed-by:
-
Hsin-Yi Wang authored
Speaker may share I2S with DP and .get_eld callback will be called when speaker is playing. When HDMI wans't connected, the connector will be null. Instead of return an error, fill in empty ELD. Signed-off-by:
Hsin-Yi Wang <hsinyi@chromium.org> Reviewed-by:
Xin Ji <xji@analogixsemi.com> Signed-off-by:
-
Devarsh Thakkar authored
Soft reset the display subsystem controller on startup and wait for the reset to complete. This helps the scenario where display was already in use by some other core before the linux was booted. Signed-off-by:
Devarsh Thakkar <devarsht@ti.com> Reviewed-by:
Tomi Valkeinen <tomi.valkeinen@ideasonboard.com> Signed-off-by:
-
- 18 Apr, 2022 3 commits
-
-
Paul Cercueil authored
drm/drm-next has a build fix for the NewVision NV3052C panel (drivers/gpu/drm/panel/panel-newvision-nv3052c.c), which needs to be merged back to drm-misc-next, as it was failing to build there. Signed-off-by:Paul Cercueil <paul@crapouillou.net>
-
Tom Rix authored
Smatch reports this issue gf108.c:147:1: warning: symbol 'gf108_gr_fwif' was not declared. Should it be static? gf108_gr_fwif is only used in gf108.c. Single file variables should not be global so change gf108_gr_fwif's storage-class specifier to static. Signed-off-by:
Tom Rix <trix@redhat.com> Reviewed-by:
Lyude Paul <lyude@redhat.com> Signed-off-by:
-
Tom Rix authored
Smatch reports this issue base917c.c:26:1: warning: symbol 'base917c_format' was not declared. Should it be static? base917c_format is only used in base917.c. Single file variables should not be global so change base917c_format's storage-class specifier to static. Signed-off-by:
-
- 14 Apr, 2022 13 commits
-
-
Helge Deller authored
This reverts commit d6cd978f. It has been solved differently already. Signed-off-by:
Helge Deller <deller@gmx.de> Acked-by:
Paul Menzel <pmenzel@molgen.mpg.de> Link: https://patchwork.freedesktop.org/patch/msgid/20220210065824.368355-1-zhouzhouyi@gmail.com
-
Zhouyi Zhou authored
In function do_remove_conflicting_framebuffers, if device is NULL, there will be null pointer reference. The patch add a check to the if expression. Signed-off-by:
Zhouyi Zhou <zhouzhouyi@gmail.com> Acked-by:
-
Yang Guang authored
Use sysfs_emit instead of scnprintf or sprintf. Reported-by:
Zeal Robot <zealci@zte.com.cn> Signed-off-by:
Yang Guang <yang.guang5@zte.com.cn> Signed-off-by:
-
Imre Deak authored
Factor out from drm_dp_dpcd_read() a function to probe a DPCD address with a 1-byte read access. This will be needed by the next patch doing a read from an LTTPR address, which must happen without the preceding wake-up read in drm_dp_dpcd_read(). While at it add tracing for the 1 byte read even if the read was successful. v2: Add a probe function instead of exporting drm_dp_dpcd_access(). (Jani) v3: Add tracing for the 1-byte read even if the read was successful. (Khaled) Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Khaled Almahallawy <khaled.almahallawy@intel.com> Cc: dri-devel@lists.freedesktop.org Signed-off-by:
Imre Deak <imre.deak@intel.com> Reviewed-by:
Jani Nikula <jani.nikula@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20220411132539.984647-1-imre.deak@intel.com
-
Jani Nikula authored
Add some helpers to figure out the EDID extension block count, block count, size, pointers to blocks. Unfortunately, we'll need to cast away the const in a few places where we actually need to access the data. v3: fix (!edid_extension_block_count(edid) == 0) (kernel test robot) v2: fix s/j/i/ introduced in a rebase Signed-off-by:
Ville Syrjälä <ville.syrjala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/dc7b0850293d837439fb3914c8a9d81e39018b4b.1649685475.git.jani.nikula@intel.com
-
Jani Nikula authored
This will be useful in the future. Use fail label for fail exit. Signed-off-by:
-
Jani Nikula authored
We don't need override as a variable for anything. Signed-off-by:
-
Jani Nikula authored
Use the EDID block read helper also for extension block reads, making edid_block_read() the only place with the read retry logic. Note: We observe that drm_do_get_edid() does not use invalid extension blocks to flag the EDID as corrupt. Signed-off-by:
-
Jani Nikula authored
Convert drm_do_get_edit() from the base block read helper to the generic block read helper. There's quite a bit going on here, as the corrupt and null EDID information is moved back to the caller. As we see, they were not all that clear to begin with, and this change underlines that. Signed-off-by:
-
Jani Nikula authored
We have an abstraction for the EDID base block read, yet duplicating the retries and error handling for extension block reads. Introduce a more generic EDID block read helper. Switch to the helper piecemeal, starting with drm_edid_get_panel_id(), which doesn't need or have access to the connector anyway. The subtle change is switching from drm_edid_block_valid() to edid_block_check(). We also status print once, not for every attempt. Signed-off-by:
-
Jani Nikula authored
Make the callback a bit easier on the eye. Signed-off-by:
-
Jani Nikula authored
Avoid casting here and there, and make it const. Signed-off-by:
-
Jani Nikula authored
Unify debug log dumping. There's duplication in the error paths for EDID block validity checks, but this should be neglible. Signed-off-by:
-
