- 27 Jun, 2024 6 commits
-
-
Jacopo Mondi authored
Add bindings for the Raspberry Pi PiSP Back End memory-to-memory image signal processor. Datasheet: https://datasheets.raspberrypi.com/camera/raspberry-pi-image-signal-processor-specification.pdfSigned-off-by:
Jacopo Mondi <jacopo.mondi@ideasonboard.com> Reviewed-by:
Rob Herring <robh@kernel.org> Reviewed-by:
Naushir Patuck <naush@raspberrypi.com> Signed-off-by:
Sakari Ailus <sakari.ailus@linux.intel.com> Signed-off-by:
Hans Verkuil <hverkuil-cisco@xs4all.nl>
-
Jacopo Mondi authored
Add Raspberry Pi compressed RAW Bayer formats. The compression algorithm description is provided by Nick Hollinghurst <nick.hollinghurst@raspberrypi.com> from Raspberry Pi. Signed-off-by:
-
Jacopo Mondi authored
Add format description for the PiSP Back End configuration parameter buffer. Signed-off-by:
-
Jacopo Mondi authored
Add the Raspberry Pi PiSP Back End uAPI header. The header defines the data type used to configure the PiSP Back End ISP. The detailed description of the types and of the ISP configuration procedure is available at https://datasheets.raspberrypi.com/camera/raspberry-pi-image-signal-processor-specification.pdfSigned-off-by:
-
Jacopo Mondi authored
Add BGR48 and RGB48 16-bit per component image formats. Signed-off-by:
Kieran Bingham <kieran.bingham@ideasonboard.com> Reviewed-by:
-
Jacopo Mondi authored
The Y10P, Y12P and Y14P format variants are packed according to the RAW10, RAW12 and RAW14 formats as defined by the MIPI CSI-2 specification. Document it. Signed-off-by:
Laurent Pinchart <laurent.pinchart@ideasonboard.com> Reviewed-by:
-
- 26 Jun, 2024 16 commits
-
-
Hans Verkuil authored
While queue_setup was correct for CREATE_BUFS support for video devices, for VBI, SDR and touch devices it was wrong. This was found after adding new v4l2-compliance tests. Signed-off-by: -
Hans Verkuil authored
In several places a division by fmt->vdownsampling[p] was missing in the sizeimage[p] calculation, causing incorrect behavior for multiplanar formats were some planes are smaller than the first plane. Found by new v4l2-compliance tests. Signed-off-by: -
Laurent Pinchart authored
Store the active formats and crop rectangle in the subdevice active state. This simplifies implementation of the format and selection accessors, and allows using the v4l2_subdev_get_fmt() helper to implement the .get_fmt() operation. The active configuration that is used in the .process_frame() handler is still stored in the vimc_scaler_device structure. The driver could instead access the active state in the .process_frame() handler, but the required locking could interfere with the real time constraints of the frame processing. This data would be stored in registers in the .s_stream() handler for real hardware, storing it in dedicated storage thus mimics a real driver. To differentiate them from the rest of the device private data, move the corresponding fields to a sub-structure of vimc_scaler_device named hw. Signed-off-by:
Shuah Khan <skhan@linuxfoundation.org> Signed-off-by:
-
Laurent Pinchart authored
Store the active formats and crop rectangle in the subdevice active state. This simplifies implementation of the format and selection accessors, and allows using the v4l2_subdev_get_fmt() helper to implement the .get_fmt() operation. The active configuration that is used in the .process_frame() handler is still stored in the vimc_debayer_device structure. The driver could instead access the active state in the .process_frame() handler, but the required locking could interfere with the real time constraints of the frame processing. This data would be stored in registers in the .s_stream() handler for real hardware, storing it in dedicated storage thus mimics a real driver. To differentiate them from the rest of the device private data, move the corresponding fields to a sub-structure of vimc_debayer_device named hw. Signed-off-by:
-
Laurent Pinchart authored
Store the active formats and crop rectangle in the subdevice active state. This simplifies implementation of the format and selection accessors, and allows using the v4l2_subdev_get_fmt() helper to implement the .get_fmt() operation. The active configuration that is used in the .process_frame() handler is still stored in the vimc_sensor_device structure. The driver could instead access the active state in the .process_frame() handler, but the required locking could interfere with the real time constraints of the frame processing. This data would be stored in registers in the .s_stream() handler for real hardware, storing it in dedicated storage thus mimics a real driver. To differentiate them from the rest of the device private data, move the corresponding fields to a sub-structure of vimc_sensor_device named hw. Signed-off-by:
-
Laurent Pinchart authored
Finalize subdev initialization for all subdevs that provide a .init_state() operation. This creates an active state for all those subdevs, which subsequent patches will use to simplify the implementation of individual vimc entities. Signed-off-by:
-
Laurent Pinchart authored
Initialize the subdev internal_ops field in the vimc_ent_sd_register() function. This handles the internal ops the same way as the subdev ops, and prepares for moving to the V4L2 subdev active state API. Signed-off-by:
-
Laurent Pinchart authored
The vic_sca_pad enum's name has been shortened to the extreme for no good reason. Rename it to vimc_scaler_pad. Signed-off-by:
-
Laurent Pinchart authored
The ent_config array contains data that is never modified. Make it const. Signed-off-by:
-
Laurent Pinchart authored
The vimc_ent_type structure contains static pointers to functions, and no other information that need to be modified after initialization. Make them const to avoid the risk of arbitrary code execution following an overflow that would overwrite the structure's contents. Signed-off-by:
-
Laurent Pinchart authored
The .init_state() operations of the debayer and sensor entities iterate over the entity's pads. In practice, the iteration covers a single pad only. Access the pad directly and remove the loops. Signed-off-by:
-
Hans Verkuil authored
Merge tag 'tags/next-media-20240626' of git://git.kernel.org/pub/scm/linux/kernel/git/pinchartl/linux.git - fix typo in v4l2-subdev.h - imx-pxp bug fix - media i2c Kconfig: add missing FW_UPLOAD select Signed-off-by:
-
Hans Verkuil authored
Merge tag 'tags/next-media-renesas-20240619' of git://git.kernel.org/pub/scm/linux/kernel/git/pinchartl/linux.git - Conversion of max9286 and adv748x to V4L2 subdev active state - Cleanups and fixes for the Renesas R-Car VSP and VIN drivers - Miscellaneous cleanups to V4L2 core Signed-off-by:
-
Laurent Pinchart authored
Replace the incorrect reference to the v4l2_subdev_enable_stream() function with the correct v4l2_subdev_enable_streams() spelling. Fixes: d0749adb ("media: v4l2-subdev: Add subdev .(enable|disable)_streams() operations") Reviewed-by:
Tomi Valkeinen <tomi.valkeinen@ideasonboard.com> Link: https://lore.kernel.org/r/20240619225343.15873-1-laurent.pinchart@ideasonboard.comSigned-off-by:
-
Kory Maincent authored
FW_LOADER config only selects the firmware loader API, but we also need the sysfs_upload symbols for firmware_upload_unregister() and firmware_upload_register() to function properly. Fixes: 7a52ab41 ("media: i2c: Add driver for THine THP7312") Cc: stable@vger.kernel.org Signed-off-by:
Kory Maincent <kory.maincent@bootlin.com> Reviewed-by:
Paul Elder <paul.elder@ideasonboard.com> Link: https://lore.kernel.org/r/20240620102544.1918105-1-kory.maincent@bootlin.comSigned-off-by:
-
Harshit Mogalapalli authored
devm_regmap_init_mmio() can fail, add a check and bail out in case of error. Fixes: 4e5bd3fd ("media: imx-pxp: convert to regmap") Cc: stable@vger.kernel.org Signed-off-by:
Harshit Mogalapalli <harshit.m.mogalapalli@oracle.com> Reviewed-by:
-
- 24 Jun, 2024 13 commits
-
-
Wolfram Sang authored
The email address bounced. I couldn't find a newer one in recent git history, so update MAINTAINERS accordingly. Signed-off-by:
Wolfram Sang <wsa+renesas@sang-engineering.com> Signed-off-by:
-
Samuel Wein authored
Fix flags in X1 Yoga example. MEDIA_LNK_FL_DYNAMIC (0x4 in the link flag) was removed in V4 Intel IPU6 and IPU6 input system drivers. Added -V flag to media-ctl commands for X1 Yoga, lower-case v only makes it verbose upper-case V sets the format. Signed-off-by:
Samuel Wein <sam@samwein.com> [Sakari Ailus: Align subject line, rewrap commit message.] Signed-off-by:
-
Zhi Mao authored
Add a V4L2 sub-device driver for Galaxycore GC05A2 image sensor. Reviewed-by:
AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com> Signed-off-by:
Zhi Mao <zhi.mao@mediatek.com> [Sakari Ailus: Fold in MAINTAINERS change.] Signed-off-by:
-
Zhi Mao authored
Add YAML device tree binding for GC05A2 CMOS image sensor, and the relevant MAINTAINERS entries. Reviewed-by:
Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org> Reviewed-by:
-
Zhi Mao authored
Add a V4L2 sub-device driver for Galaxycore GC08A3 image sensor. Reviewed-by:
-
Zhi Mao authored
Add YAML device tree binding for GC08A3 CMOS image sensor, and the relevant MAINTAINERS entries. Reviewed-by:
-
Hans Verkuil authored
smatch generated this warning: drivers/media/usb/b2c2/flexcop-usb.c:199 flexcop_usb_memory_req() warn: iterator 'i' not incremented and indeed the function is not using i or updating buf. The reason this always worked is that this function is called to write just 6 bytes (a MAC address) to the USB device, and so in practice there is only a single chunk written. If we ever would need to write more than one chunk, this function would fail since each chunk would read from or write to the same buf address. Rewrite the function to properly handle this. Signed-off-by: -
Devarsh Thakkar authored
Use v4l2-jpeg core API to import reference quantization and huffman tables used for JPEG Encoding. Signed-off-by:
Devarsh Thakkar <devarsht@ti.com> Reviewed-by:
Chen-Yu Tsai <wenst@chromium.org> Acked-by:
Nicolas Dufresne <nicolas.dufresne@collabora.com> Signed-off-by:
Sebastian Fricke <sebastian.fricke@collabora.com> Signed-off-by:
-
Devarsh Thakkar authored
Use exported huffman and quantization tables from v4l2-jpeg core library. Signed-off-by:
-
Devarsh Thakkar authored
Enable documentation for v4l2-jpeg header related helper functions which are useful for parsing jpeg files while decoding or creating jpeg headers while encoding. Signed-off-by:
-
Devarsh Thakkar authored
Export reference quantization and huffman tables as provided in ITU-T.81 so that they can be re-used by other JPEG drivers. These are example tables provided in ITU-T.81 as reference tables and the JPEG encoders are free to use either these or their own proprietary tables. Also add necessary prefixes to be used for huffman tables in global header file. Signed-off-by:
-
Devarsh Thakkar authored
This adds support for stateful V4L2 M2M based driver for Imagination E5010 JPEG Encoder [1] which supports baseline encoding with two different quantization tables and compression ratio as demanded. Support for both contiguous and non-contiguous YUV420 and YUV422 semiplanar formats is added along with alignment restrictions as required by the hardware. System and runtime PM hooks are added in the driver along with v4l2 crop and selection API support. Minimum resolution supported is 64x64 and Maximum resolution supported is 8192x8192. All v4l2-compliance tests are passing [2] : v4l2-compliance -s -f -a -d /dev/video0 -e /dev/video1 Total for e5010 device /dev/video0: 79, Succeeded: 79, Failed: 0, Warnings: 0 NOTE: video1 here is VIVID test pattern generator Also tests [3] were run manually to verify below driver features: - Runtime Power Management - Multi-instance JPEG Encoding - DMABUF import, export support - NV12, NV21, NV16, NV61 video format support - Compression quality S_CTRL - Cropping support using S_SELECTION Existing V4L2 M2M based JPEG drivers namely s5p-jpeg, imx-jpeg and rcar_jpu were referred while making this. TODO: Add MMU and memory tiling support [1]: AM62A TRM (Section 7.6 is for JPEG Encoder) : Link: https://www.ti.com/lit/pdf/spruj16 [2]: v4l2-compliance test : Link: https://gist.github.com/devarsht/1f039c631ca953a57f405cfce1b69e49 [3]: E5010 JPEG Encoder Manual tests : Performance: Link: https://gist.github.com/devarsht/c40672944fd71c9a53ab55adbfd9e28b Functionality: Link: https://gist.github.com/devarsht/8e88fcaabff016bb2bac83d89c9d23ce Compression Quality: Link: https://gist.github.com/devarsht/cbcc7cd97e8c48ba1486caa2b7884655 Multi Instance: Link: https://gist.github.com/devarsht/22c2fca08cd3441fb40f2c7a4cebc95a Crop support: Link: https://gist.github.com/devarsht/de6f5142f678bb1a5338abfd9f814abd Runtime PM: Link: https://gist.github.com/devarsht/70cd95d4440ddc678489d93885ddd4ddCo-developed-by:
David Huang <d-huang@ti.com> Signed-off-by:
Benjamin Gaignard <benjamin.gaignard@collabora.com> Signed-off-by:
-
Devarsh Thakkar authored
Add dt-bindings for Imagination E5010 JPEG Encoder [1] which is implemented as stateful V4L2 M2M driver. The device supports baseline encoding with two different quantization tables and compression ratio as demanded. Minimum resolution supported is 64x64 and Maximum resolution supported is 8192x8192. Link: https://www.ti.com/lit/pdf/spruj16 [1] (Section 7.6 JPEG Encoder) Co-developed-by:
-
- 21 Jun, 2024 5 commits
-
-
Ivan Bornyakov authored
Add initial support for the Wave515 multi-decoder IP. For now it is only able to decode HEVC Main/Main10 profile videos into YUV420. This was tested on FPGA prototype, so wave5_dt_ids[] was not expanded. Users of the real hardware with Wave515 IP will have to * provide firmware specific to their SoC * add struct wave5_match_data like this: static const struct wave5_match_data platform_name_wave515_data = { .flags = WAVE5_IS_DEC, .fw_name = "cnm/wave515_platform_name_fw.bin", .sram_size = (71 * 1024), }; * add item to wave5_dt_ids[] like this: { .compatible = "vendor,soc-wave515", .data = &platform_name_wave515_data, }, * describe new compatible in Documentation/devicetree/bindings/media/cnm,wave521c.yaml Signed-off-by:Ivan Bornyakov <brnkv.i1@gmail.com> Signed-off-by:
-
Ivan Bornyakov authored
Move the excessive "sram-size" device-tree property to the device match data. Also change the SRAM memory allocation strategy, instead of allocating exactly sram_size bytes, allocate all available SRAM memory up to sram_size. Add the placeholders wave5_vpu_dec_validate_sec_axi() and wave5_vpu_enc_validate_sec_axi() to validate that the allocated SRAM memory is sufficient to decode/encode bitstream with a given resolution. Signed-off-by:
-
Ivan Bornyakov authored
Implement a separate setup routine for interrupts to reduce code duplication. Also enable interrupts based on vpu_attr->support_encoders and vpu_attr->support_decoders fields to facilitate support for other Wave5xx IPs, because not all of them are both encoders and decoders. Signed-off-by:
-
Ivan Bornyakov authored
Add initial support for optional reset lines. For now, simply deassert resets while probing the driver and assert them back when removing the driver. Signed-off-by:
-
Ivan Bornyakov authored
Add support for decoding HEVC Main10 profile by scaling the FBC buffer stride and size by a factor of (bitdepth / 8). Signed-off-by:
-
