- 10 Sep, 2020 40 commits
-
-
Robert Foss authored
OV8856 supports 1632x1224 @ 60 FPS, which is added in this commit. Signed-off-by:
Robert Foss <robert.foss@linaro.org> Signed-off-by:
Sakari Ailus <sakari.ailus@linux.intel.com> Signed-off-by:
Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
-
Sakari Ailus authored
Document that it is possible to provide defaults for multiple bus types to v4l2_fwnode_endpoint_parse and v4l2_fwnode_endpoint_alloc_parse. Also underline the fact that detecting the bus type without bus-type property is only for the old drivers. Also correct capitalisation of BT.656. Signed-off-by:
Laurent Pinchart <laurent.pinchart@ideasonboard.com> Signed-off-by:
-
Sakari Ailus authored
The bus specific parameters were a union. This made providing bus specific defaults impossible as the memory used to store the defaults for multiple different busses was the same. Make it struct instead. It's not large so the size isn't really an issue. Signed-off-by:
-
Sakari Ailus authored
The maximum is currently four (4). No short is needed. Signed-off-by:
-
Laurent Pinchart authored
v4l2_async_notifier_add_subdev() requires the asd to be allocated dynamically, but the max9286 driver embeds it in the max9286_source structure. This causes memory corruption when the notifier is destroyed at remove time with v4l2_async_notifier_cleanup(). Fix this issue by registering the asd with v4l2_async_notifier_add_fwnode_subdev(), which allocates it dynamically internally. A new max9286_asd structure is introduced, to store a pointer to the corresonding max9286_source that needs to be accessed from bound and unbind callbacks. There's no need to take an extra explicit reference to the fwnode anymore as v4l2_async_notifier_add_fwnode_subdev() does so internally. While at it, use %u instead of %d to print the unsigned index in the error message from the v4l2_async_notifier_add_fwnode_subdev() error path. Fixes: 66d8c9d2 ("media: i2c: Add MAX9286 driver") Signed-off-by:
Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com> Signed-off-by:
-
Laurent Pinchart authored
v4l2_async_notifier_add_subdev() requires the asd to be allocated dynamically, but the rcar-csi2 driver embeds it in the rcar_csi2 structure. This causes memory corruption when the notifier is destroyed at remove time with v4l2_async_notifier_cleanup(). Fix this issue by registering the asd with v4l2_async_notifier_add_fwnode_subdev(), which allocates it dynamically internally. Fixes: 769afd21 ("media: rcar-csi2: add Renesas R-Car MIPI CSI-2 receiver driver") Signed-off-by:
-
Laurent Pinchart authored
v4l2_async_notifier_add_subdev() requires the asd to be allocated dynamically, but the rcar-drif driver embeds it in the rcar_drif_graph_ep structure. This causes memory corruption when the notifier is destroyed at remove time with v4l2_async_notifier_cleanup(). Fix this issue by registering the asd with v4l2_async_notifier_add_fwnode_subdev(), which allocates it dynamically internally. Fixes: d079f94c ("media: platform: Switch to v4l2_async_notifier_add_subdev") Signed-off-by:
-
Laurent Pinchart authored
The fwnode reference corresponding to the endpoint is leaked in an error path of the rcar_drif_parse_subdevs() function. Fix it, and reorganize fwnode reference handling in the function to release references early, simplifying error paths. Signed-off-by:
-
Laurent Pinchart authored
The v4l2_async_notifier_add_subdev() function requires the asd pointer it receives to be allocated dynamically, but doesn't explicitly say so. Only one driver out of 13 get its right (atmel-sama5d2-isc.c, but with memory leaks in the error paths), clearly showing we have an issue. Update the v4l2_async_notifier_add_subdev() documentation to clearly state the allocation requirement. Whether this will be enough to avoid new offending code isn't certain, but it's a good first step nonetheless. Fixes: 9ca46531 ("media: v4l: fwnode: Support generic parsing of graph endpoints in a device") Signed-off-by:
-
Lad Prabhakar authored
Fail probe if unsupported bus_type is detected. Signed-off-by:
Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com> Reviewed-by:
Jacopo Mondi <jacopo@jmondi.org> Signed-off-by:
-
Lad Prabhakar authored
Enable support for BT656 mode. Signed-off-by:
Biju Das <biju.das.jz@bp.renesas.com> Reviewed-by:
-
Lad Prabhakar authored
Configure HVP lines in s_power callback instead of configuring everytime in ov5640_set_stream_dvp(). Alongside also disable MIPI in DVP mode. Signed-off-by:
-
Lad Prabhakar authored
During testing this sensor on iW-RainboW-G21D-Qseven platform in 8-bit DVP mode with rcar-vin bridge noticed the capture worked fine for the first run (with yavta), but for subsequent runs the bridge driver waited for the frame to be captured. Debugging further noticed the data lines were enabled/disabled in stream on/off callback and dumping the register contents 0x3017/0x3018 in ov5640_set_stream_dvp() reported the correct values, but yet frame capturing failed. To get around this issue data lines are enabled in s_power callback. (Also the sensor remains in power down mode if not streaming so power consumption shouldn't be affected) Fixes: f22996db ("media: ov5640: add support of DVP parallel interface") Signed-off-by:
-
Lad Prabhakar authored
In preparation for adding DVP configuration in s_power callback move mipi configuration into separate function Signed-off-by:
-
Lad Prabhakar authored
Keep the sensor in software power down mode and wake up only in ov5640_set_stream_dvp() callback. Signed-off-by:
-
Sakari Ailus authored
pr_info() was accidentally used to print the link frequencies whereas the rest of the information is printed on debug level. Fix that by using pr_debug() also for link frequencies. Signed-off-by:
-
Sakari Ailus authored
Alan Chiang's e-mail address no longer works. Remove it. Signed-off-by:
-
Andy Shevchenko authored
Update Copyright year to cover the previous changes and at the same time address indentation issues. Signed-off-by:
Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by:
-
Andy Shevchenko authored
There are assignments inside the functions which are useless. Drop them for good. Signed-off-by:
-
Andy Shevchenko authored
pcim_iomap_table() won't fail if previous pcim_iomap_regions() hasn't. Since we check pcim_iomap_regions() for failure the check close to pcim_iomap_table() is bogus and not needed. Signed-off-by:
-
Andy Shevchenko authored
It's unclear why driver repeats the code from PCI core. Drop it for good. Signed-off-by:
-
Andy Shevchenko authored
We may use special helper macro to poll IO till condition or timeout occurs. Signed-off-by:
-
Andy Shevchenko authored
Refactor cio2_buffer_done() to get rid of infinite loop by replacing it by one with clear exit condition. This change also allows to check for an error ahead. Signed-off-by:
-
Andy Shevchenko authored
There are few nice macros in pfn.h, some of which we may use here. Signed-off-by:
-
Andy Shevchenko authored
It's quite unlikely that another page size will be supported, but in any case there is still an inconsistency between custom page size definition and generic macros used in the driver. Switch over to the generic PAGE_SIZE for sake of the consistency. Signed-off-by:
-
Andy Shevchenko authored
This constant is used in several places in the code, define it for better maintenance. Signed-off-by:
-
Andy Shevchenko authored
The code looks more nicer if we use: while (i--) instead: for (i = i - 1; i >= 0; i--) This would also allow making 'i' unsigned again. Signed-off-by:
-
Dinghao Liu authored
When devm_ioremap_resource() fails, isp should be freed just like other error paths in isp_probe. Fixes: 8644cdf9 ("[media] omap3isp: Replace many MMIO regions by two") Signed-off-by:
Dinghao Liu <dinghao.liu@zju.edu.cn> Signed-off-by:
-
Christophe JAILLET authored
The intent here is to reset the whole 'scaler_coeffs_luma' array, not just the first element. Fixes: e11110a5 ("media: staging/intel-ipu3: css: Compute and program ccs") Signed-off-by:
Christophe JAILLET <christophe.jaillet@wanadoo.fr> Reviewed-by:
Bingbu Cao <bingbu.cao@intel.com> Signed-off-by:
-
Adam Goode authored
The Color Matching Descriptor has been present in USB cameras since the original version of UVC, but it has never been fully exposed in Linux. This change informs V4L2 of all of the UVC colorspace parameters: color primaries, transfer characteristics, and YCbCr encoding. videodev2.h doesn't have values for all the possible UVC color settings, so it is mapped as closely as possible. Signed-off-by:
Adam Goode <agoode@google.com> Signed-off-by:
-
Adam Goode authored
bFrameIndex and bFormatIndex can be negotiated by the camera during probing, resulting in the camera choosing a different format than expected. v4l2 can already accommodate such changes, but the code was not updating the proper fields. Without such a change, v4l2 would potentially interpret the payload incorrectly, causing corrupted output. This was happening on the Elgato HD60 S+, which currently always renegotiates to format 1. As an aside, the Elgato firmware is buggy and should not be renegotating, but it is still a valid thing for the camera to do. Both macOS and Windows will properly probe and read uncorrupted images from this camera. With this change, both qv4l2 and chromium can now read uncorrupted video from the Elgato HD60 S+. [Add blank lines, remove periods at the of messages] Signed-off-by:
-
Laurent Pinchart authored
UBSAN reports a shift-out-of-bounds warning in uvc_get_le_value(). The report is correct, but the issue should be harmless as the computed value isn't used when the shift is negative. This may however cause incorrect behaviour if a negative shift could generate adverse side effects (such as a trap on some architectures for instance). Regardless of whether that may happen or not, silence the warning as a full WARN backtrace isn't nice. Reported-by:
Bart Van Assche <bvanassche@acm.org> Fixes: c0efd232 ("V4L/DVB (8145a): USB Video Class driver") Signed-off-by:
-
Daniel W. S. Almeida authored
Fixes the following coccinelle report: drivers/media/usb/uvc/uvc_ctrl.c:1860:5-11: ERROR: invalid reference to the index variable of the iterator on line 1854 by adding a boolean variable to check if the loop has found the Found using - Coccinelle (http://coccinelle.lip6.fr) [Replace cursor variable with bool found] Signed-off-by:
Daniel W. S. Almeida <dwlsalmeida@gmail.com> Signed-off-by:
-
Alexander A. Klimov authored
Rationale: Reduces attack surface on kernel devs opening the links for MITM as HTTPS traffic is much harder to manipulate. Deterministic algorithm: For each file: If not .svg: For each line: If doesn't contain `\bxmlns\b`: If neither `\bgnu\.org/license`, nor `\bmozilla\.org/MPL\b`: If both the HTTP and HTTPS versions return 200 OK and serve the same content: Replace HTTP with HTTPS. Signed-off-by:Alexander A. Klimov <grandmaster@al2klimov.de> Signed-off-by:
-
Hans de Goede authored
There is only 1 error exit in uvc_ctrl_add_info(), so using goto style error handling is not necessary. Also the kfree(ctrl->uvc_data) on error is not necessary, because the only error exit is for the kzalloc() of ctrl->uvc_data failing. Remove all the error handling cruft and simply do "return -ENOMEM" on kzalloc() failure. Signed-off-by:
Hans de Goede <hdegoede@redhat.com> Signed-off-by:
-
Hans de Goede authored
uvc_ctrl_add_info() calls uvc_ctrl_get_flags() which will override the fixed-up flags set by uvc_ctrl_fixup_xu_info(). uvc_ctrl_init_xu_ctrl() already calls uvc_ctrl_get_flags() before calling uvc_ctrl_add_info(), so the uvc_ctrl_get_flags() call in uvc_ctrl_add_info() is not necessary for xu ctrls. This commit moves the uvc_ctrl_get_flags() call for normal controls from uvc_ctrl_add_info() to uvc_ctrl_init_ctrl(), so that we no longer call uvc_ctrl_get_flags() twice for xu controls and so that we no longer override the fixed-up flags set by uvc_ctrl_fixup_xu_info(). This fixes the xu motor controls not working properly on a Logitech 046d:08cc, and presumably also on the other Logitech models which have a quirk for this in the uvc_ctrl_fixup_xu_info() function. Cc: stable@vger.kernel.org Signed-off-by:
-
Laurent Pinchart authored
The media controller core prints a warning when an entity is registered without a function being set. This affects the uvcvideo driver, as the warning was added without first addressing the issue in existing drivers. The problem is harmless, but unnecessarily worries users. Fix it by mapping UVC entity types to MC entity functions as accurately as possible using the existing functions. Fixes: b50bde4e ("[media] v4l2-subdev: use MEDIA_ENT_T_UNKNOWN for new subdevs") Signed-off-by:
Kieran Bingham <kieran.bingham@ideasonboard.com> Signed-off-by:
-
Dmitry Buzdyk authored
Add HEVC GUID and assotiate with HEVC pixel format so that frame based format descriptors recognized by the UVC video driver. Signed-off-by:
Dmitry Buzdyk <dima.buzdyk@gmail.com> Signed-off-by:
-
Jun Chen authored
The formula (2) is to convert from SOF to host clock, it should be fix as "TS = ((TS2 - TS1) * SOF + TS1 * SOF2 - TS2 * SOF1) / (SOF2 - SOF1)" Signed-off-by:
Jun Chen <jun.chen@vatics.com> Signed-off-by:
-
Robin Murphy authored
Since commit 9495b7e9 ("driver core: platform: Initialize dma_parms for platform devices"), struct platform_device already provides a dma_parms structure, so we can save allocating another one. Also the DMA segment size is simply a size, not a bitmask. Signed-off-by:
Robin Murphy <robin.murphy@arm.com> Reviewed-by:
Kieran Bingham <kieran.bingham+renesas@ideasonboard.com> Signed-off-by:
-
