kernel/linux.git/drivers/usb/gadget/function/uvc_queue.c, branch v5.15.208

usb: gadget: uvc: fix sg handling in error case

2022-11-03T14:59:11+00:00

commit 0a0a2760b04814428800d48281a447a7522470ad upstream. If there is a transmission error the buffer will be returned too early, causing a memory fault as subsequent requests for that buffer are still queued up to be sent. Refactor the error handling to wait for the final request to come in before reporting back the buffer to userspace for all transfer types (bulk/isoc/isoc_sg). This ensures userspace knows if the frame was successfully sent. Fixes: e81e7f9a0eb9 ("usb: gadget: uvc: add scatter gather support") Cc: Signed-off-by: Dan Vacura Link: https://lore.kernel.org/r/20221018215044.765044-4-w36195@motorola.com Signed-off-by: Greg Kroah-Hartman

usb: gadget: uvc: giveback vb2 buffer on req complete

2022-10-29T08:12:53+00:00

commit 9b969f93bcef9b3d9e92f1810e22bbd6c344a0e5 upstream. On uvc_video_encode_isoc_sg the mapped vb2 buffer is returned to early. Only after the last usb_request worked with the buffer it is allowed to give it back to vb2. This patch fixes that. Signed-off-by: Michael Grzeschik Link: https://lore.kernel.org/r/20220402232744.3622565-3-m.grzeschik@pengutronix.de Cc: Dan Vacura Signed-off-by: Greg Kroah-Hartman

usb: gadget: uvc: rework uvcg_queue_next_buffer to uvcg_complete_buffer

2022-10-29T08:12:53+00:00

commit 61aa709ca58a0dbeeb817bfa9230c1a92979f2c6 upstream. The function uvcg_queue_next_buffer is used different than its name suggests. The return value nextbuf is never used by any caller. This patch reworks the function to its actual purpose, by removing the unused code and renaming it. The function name uvcg_complete_buffer makes it more clear that it is actually marking the current video buffer as complete. Signed-off-by: Michael Grzeschik Link: https://lore.kernel.org/r/20220402232744.3622565-2-m.grzeschik@pengutronix.de Cc: Dan Vacura Signed-off-by: Greg Kroah-Hartman

usb: gadget: uvc: calculate the number of request depending on framesize

2022-08-25T09:40:31+00:00

[ Upstream commit 87d76b5f1d8eeb49efa16e2018e188864cbb9401 ] The current limitation of possible number of requests being handled is dependent on the gadget speed. It makes more sense to depend on the typical frame size when calculating the number of requests. This patch is changing this and is using the previous limits as boundaries for reasonable minimum and maximum number of requests. For a 1080p jpeg encoded video stream with a maximum imagesize of e.g. 800kB with a maxburst of 8 and an multiplier of 1 the resulting number of requests is calculated to 49. 800768 1 nreqs = ------ * -------------- ~= 49 2 (1024 * 8 * 1) Tested-by: Dan Vacura Signed-off-by: Michael Grzeschik Link: https://lore.kernel.org/r/20220529223848.105914-2-m.grzeschik@pengutronix.de Signed-off-by: Greg Kroah-Hartman Signed-off-by: Sasha Levin

usb: gadget: uvc: Fix crash when encoding data for usb request

2022-05-09T07:14:28+00:00

commit 71d471e3faf90c9674cadc7605ac719e82cb7fac upstream. During the uvcg_video_pump() process, if an error occurs and uvcg_queue_cancel() is called, the buffer queue will be cleared out, but the current marker (queue->buf_used) of the active buffer (no longer active) is not reset. On the next iteration of uvcg_video_pump() the stale buf_used count will be used and the logic of min((unsigned int)len, buf->bytesused - queue->buf_used) may incorrectly calculate a nbytes size, causing an invalid memory access. [80802.185460][ T315] configfs-gadget gadget: uvc: VS request completed with status -18. [80802.185519][ T315] configfs-gadget gadget: uvc: VS request completed with status -18. ... uvcg_queue_cancel() is called and the queue is cleared out, but the marker queue->buf_used is not reset. ... [80802.262328][ T8682] Unable to handle kernel paging request at virtual address ffffffc03af9f000 ... ... [80802.263138][ T8682] Call trace: [80802.263146][ T8682] __memcpy+0x12c/0x180 [80802.263155][ T8682] uvcg_video_pump+0xcc/0x1e0 [80802.263165][ T8682] process_one_work+0x2cc/0x568 [80802.263173][ T8682] worker_thread+0x28c/0x518 [80802.263181][ T8682] kthread+0x160/0x170 [80802.263188][ T8682] ret_from_fork+0x10/0x18 [80802.263198][ T8682] Code: a8c12829 a88130cb a8c130 Fixes: d692522577c0 ("usb: gadget/uvc: Port UVC webcam gadget to use videobuf2 framework") Cc: Signed-off-by: Dan Vacura Link: https://lore.kernel.org/r/20220331184024.23918-1-w36195@motorola.com Signed-off-by: Greg Kroah-Hartman

usb: gadget: uvc: add scatter gather support

2021-07-27T13:59:19+00:00

This patch adds support for scatter gather transfers. If the underlying gadgets sg_supported == true, then the videeobuf2-dma-sg is used and the encode routine maps all scatter entries to separate scatterlists for the usb gadget. When streaming 1080p with request size of 1024 times 3 bytes top shows a difference of about 6.4% CPU load applying this patch: PID USER PR NI VIRT RES %CPU %MEM TIME+ S COMMAND 64 root 0 -20 0.0m 0.0m 7.7 0.0 0:01.25 I [kworker/u5:0-uvcvideo] 83 root 0 -20 0.0m 0.0m 4.5 0.0 0:03.71 I [kworker/u5:3-uvcvideo] 307 root -51 0 0.0m 0.0m 3.8 0.0 0:01.05 S [irq/51-dwc3] vs. 64 root 0 -20 0.0m 0.0m 5.8 0.0 0:01.79 I [kworker/u5:0-uvcvideo] 306 root -51 0 0.0m 0.0m 3.2 0.0 0:01.97 S [irq/51-dwc3] 82 root 0 -20 0.0m 0.0m 0.6 0.0 0:01.86 I [kworker/u5:1-uvcvideo] Signed-off-by: Michael Grzeschik Link: https://lore.kernel.org/r/20210628155311.16762-5-m.grzeschik@pengutronix.de Signed-off-by: Greg Kroah-Hartman

usb: gadget: uvc: make uvc_num_requests depend on gadget speed

2021-07-27T13:59:18+00:00

While sending bigger images is possible with USB_SPEED_SUPER it is better to use more isochronous requests in flight. This patch makes the number uvc_num_requests dynamic by changing it depending on the gadget speed. Reviewed-by: Paul Elder Reviewed-by: Laurent Pinchart Signed-off-by: Michael Grzeschik Link: https://lore.kernel.org/r/20210628155311.16762-3-m.grzeschik@pengutronix.de Signed-off-by: Greg Kroah-Hartman

usb: gadget: uvc: constify vb2_ops structure

2018-11-26T07:06:32+00:00

The vb2_ops structure can be const as it is only stored in the ops field of a vb2_queue structure and this field is const. Done with the help of Coccinelle. Reviewed-by: Laurent Pinchart Signed-off-by: Julia Lawall Signed-off-by: Felipe Balbi

media: videobuf2-v4l2: integrate with media requests

2018-08-31T15:19:30+00:00

This implements the V4L2 part of the request support. The main change is that vb2_qbuf and vb2_prepare_buf now have a new media_device pointer. This required changes to several drivers that did not use the vb2_ioctl_qbuf/prepare_buf helper functions. Signed-off-by: Hans Verkuil Reviewed-by: Mauro Carvalho Chehab Signed-off-by: Mauro Carvalho Chehab

media: annotate ->poll() instances

2017-11-27T21:20:06+00:00

Signed-off-by: Al Viro