Re: [PATCH] v4l: uvcvideo: Fix buffer completion size check
Hi Laurent,
On Thu, 2 Oct 2014, Laurent Pinchart wrote:
Hi Guennadi,
Ping ?
Sorry again for a delay, and unfortunately my eventual reply won't be very
helpful: we've modified our user-space in a way, that that path isn't
triggered anymore, so, I cannot easily verify your patch. In any case by
looking at it and judging by the fact, that sizeimage is anyway
initialised from dwMaxVideoFrameSize in VIDIOC_G_FMT, I think your patch
is correct.
Thanks
Guennadi
On Wednesday 01 October 2014 00:42:51 Laurent Pinchart wrote:
Commit e93e7fd9f5a3fffec7792dbcc4c3574653effda7 (v4l2: uvcvideo: Allow
using larger buffers) reworked the buffer size sanity check at buffer
completion time to use the frame size instead of the allocated buffer
size. However, it introduced two bugs in doing so:
- it assigned the allocated buffer size to the frame_size field, instead
of assigning the correct frame size
- it performed the assignment in the S_FMT handler, resulting in the
frame_size field being uninitialized if the userspace application
doesn't call S_FMT.
Fix both issues by removing the frame_size field and validating the
buffer size against the UVC video control dwMaxFrameSize.
Fixes: e93e7fd9f5a3 (v4l2: uvcvideo: Allow using larger buffers)
Signed-off-by: Laurent Pinchart laurent.pinch...@ideasonboard.com
---
drivers/media/usb/uvc/uvc_v4l2.c | 1 -
drivers/media/usb/uvc/uvc_video.c | 2 +-
drivers/media/usb/uvc/uvcvideo.h | 1 -
3 files changed, 1 insertion(+), 3 deletions(-)
Guennadi, could you please test and ack this ASAP, as the bug needs to be
fixed for v3.18-rc1 if possible ?
diff --git a/drivers/media/usb/uvc/uvc_v4l2.c
b/drivers/media/usb/uvc/uvc_v4l2.c index f205934..f33a067 100644
--- a/drivers/media/usb/uvc/uvc_v4l2.c
+++ b/drivers/media/usb/uvc/uvc_v4l2.c
@@ -318,7 +318,6 @@ static int uvc_v4l2_set_format(struct uvc_streaming
*stream, stream-ctrl = probe;
stream-cur_format = format;
stream-cur_frame = frame;
- stream-frame_size = fmt-fmt.pix.sizeimage;
done:
mutex_unlock(stream-mutex);
diff --git a/drivers/media/usb/uvc/uvc_video.c
b/drivers/media/usb/uvc/uvc_video.c index 9ace520..df81b9c 100644
--- a/drivers/media/usb/uvc/uvc_video.c
+++ b/drivers/media/usb/uvc/uvc_video.c
@@ -1143,7 +1143,7 @@ static int uvc_video_encode_data(struct uvc_streaming
*stream, static void uvc_video_validate_buffer(const struct uvc_streaming
*stream, struct uvc_buffer *buf)
{
- if (stream-frame_size != buf-bytesused
+ if (stream-ctrl.dwMaxVideoFrameSize != buf-bytesused
!(stream-cur_format-flags UVC_FMT_FLAG_COMPRESSED))
buf-error = 1;
}
diff --git a/drivers/media/usb/uvc/uvcvideo.h
b/drivers/media/usb/uvc/uvcvideo.h index f585c08..897cfd8 100644
--- a/drivers/media/usb/uvc/uvcvideo.h
+++ b/drivers/media/usb/uvc/uvcvideo.h
@@ -458,7 +458,6 @@ struct uvc_streaming {
struct uvc_format *def_format;
struct uvc_format *cur_format;
struct uvc_frame *cur_frame;
- size_t frame_size;
/* Protect access to ctrl, cur_format, cur_frame and hardware video
* probe control.
--
Regards,
Laurent Pinchart
--
To unsubscribe from this list: send the line unsubscribe linux-media in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: [PATCH] v4l: uvcvideo: Fix buffer completion size check
On Mon, 10 Nov 2014, Guennadi Liakhovetski wrote:
Hi Laurent,
On Thu, 2 Oct 2014, Laurent Pinchart wrote:
Hi Guennadi,
Ping ?
Sorry again for a delay, and unfortunately my eventual reply won't be very
helpful: we've modified our user-space in a way, that that path isn't
triggered anymore, so, I cannot easily verify your patch. In any case by
looking at it and judging by the fact, that sizeimage is anyway
initialised from dwMaxVideoFrameSize in VIDIOC_G_FMT, I think your patch
is correct.
Actgually, I did test it, but only in the normal case, not in the error
case, which was the problem. I.e. when the buffer size is equal to the
frame size. But since I tested this case with your patch applied and it
worked, this means, that dwMaxVideoFrameSize is indeed equal to the
expected image size, since it is unlikely, that it is derived from the
buffer size, right? So, I guess, we can count this as a
Tested-by: Guennadi Liakhovetski g.liakhovet...@gmx.de
(if you haven't pushed it upstream yet ;))
Thanks
Guennadi
On Wednesday 01 October 2014 00:42:51 Laurent Pinchart wrote:
Commit e93e7fd9f5a3fffec7792dbcc4c3574653effda7 (v4l2: uvcvideo: Allow
using larger buffers) reworked the buffer size sanity check at buffer
completion time to use the frame size instead of the allocated buffer
size. However, it introduced two bugs in doing so:
- it assigned the allocated buffer size to the frame_size field, instead
of assigning the correct frame size
- it performed the assignment in the S_FMT handler, resulting in the
frame_size field being uninitialized if the userspace application
doesn't call S_FMT.
Fix both issues by removing the frame_size field and validating the
buffer size against the UVC video control dwMaxFrameSize.
Fixes: e93e7fd9f5a3 (v4l2: uvcvideo: Allow using larger buffers)
Signed-off-by: Laurent Pinchart laurent.pinch...@ideasonboard.com
---
drivers/media/usb/uvc/uvc_v4l2.c | 1 -
drivers/media/usb/uvc/uvc_video.c | 2 +-
drivers/media/usb/uvc/uvcvideo.h | 1 -
3 files changed, 1 insertion(+), 3 deletions(-)
Guennadi, could you please test and ack this ASAP, as the bug needs to be
fixed for v3.18-rc1 if possible ?
diff --git a/drivers/media/usb/uvc/uvc_v4l2.c
b/drivers/media/usb/uvc/uvc_v4l2.c index f205934..f33a067 100644
--- a/drivers/media/usb/uvc/uvc_v4l2.c
+++ b/drivers/media/usb/uvc/uvc_v4l2.c
@@ -318,7 +318,6 @@ static int uvc_v4l2_set_format(struct uvc_streaming
*stream, stream-ctrl = probe;
stream-cur_format = format;
stream-cur_frame = frame;
- stream-frame_size = fmt-fmt.pix.sizeimage;
done:
mutex_unlock(stream-mutex);
diff --git a/drivers/media/usb/uvc/uvc_video.c
b/drivers/media/usb/uvc/uvc_video.c index 9ace520..df81b9c 100644
--- a/drivers/media/usb/uvc/uvc_video.c
+++ b/drivers/media/usb/uvc/uvc_video.c
@@ -1143,7 +1143,7 @@ static int uvc_video_encode_data(struct
uvc_streaming
*stream, static void uvc_video_validate_buffer(const struct uvc_streaming
*stream, struct uvc_buffer *buf)
{
- if (stream-frame_size != buf-bytesused
+ if (stream-ctrl.dwMaxVideoFrameSize != buf-bytesused
!(stream-cur_format-flags UVC_FMT_FLAG_COMPRESSED))
buf-error = 1;
}
diff --git a/drivers/media/usb/uvc/uvcvideo.h
b/drivers/media/usb/uvc/uvcvideo.h index f585c08..897cfd8 100644
--- a/drivers/media/usb/uvc/uvcvideo.h
+++ b/drivers/media/usb/uvc/uvcvideo.h
@@ -458,7 +458,6 @@ struct uvc_streaming {
struct uvc_format *def_format;
struct uvc_format *cur_format;
struct uvc_frame *cur_frame;
- size_t frame_size;
/* Protect access to ctrl, cur_format, cur_frame and hardware video
* probe control.
--
Regards,
Laurent Pinchart
--
To unsubscribe from this list: send the line unsubscribe linux-media in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
--
To unsubscribe from this list: send the line unsubscribe linux-media in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: [PATCH] v4l: uvcvideo: Fix buffer completion size check
Hi Laurent,
On Thu, 2 Oct 2014, Laurent Pinchart wrote:
Hi Guennadi,
Ping ?
Uhm, yes, that's bad. 1 month and a week. Sorry. I'll try to test this on
Monday.
Thanks
Guennadi
On Wednesday 01 October 2014 00:42:51 Laurent Pinchart wrote:
Commit e93e7fd9f5a3fffec7792dbcc4c3574653effda7 (v4l2: uvcvideo: Allow
using larger buffers) reworked the buffer size sanity check at buffer
completion time to use the frame size instead of the allocated buffer
size. However, it introduced two bugs in doing so:
- it assigned the allocated buffer size to the frame_size field, instead
of assigning the correct frame size
- it performed the assignment in the S_FMT handler, resulting in the
frame_size field being uninitialized if the userspace application
doesn't call S_FMT.
Fix both issues by removing the frame_size field and validating the
buffer size against the UVC video control dwMaxFrameSize.
Fixes: e93e7fd9f5a3 (v4l2: uvcvideo: Allow using larger buffers)
Signed-off-by: Laurent Pinchart laurent.pinch...@ideasonboard.com
---
drivers/media/usb/uvc/uvc_v4l2.c | 1 -
drivers/media/usb/uvc/uvc_video.c | 2 +-
drivers/media/usb/uvc/uvcvideo.h | 1 -
3 files changed, 1 insertion(+), 3 deletions(-)
Guennadi, could you please test and ack this ASAP, as the bug needs to be
fixed for v3.18-rc1 if possible ?
diff --git a/drivers/media/usb/uvc/uvc_v4l2.c
b/drivers/media/usb/uvc/uvc_v4l2.c index f205934..f33a067 100644
--- a/drivers/media/usb/uvc/uvc_v4l2.c
+++ b/drivers/media/usb/uvc/uvc_v4l2.c
@@ -318,7 +318,6 @@ static int uvc_v4l2_set_format(struct uvc_streaming
*stream, stream-ctrl = probe;
stream-cur_format = format;
stream-cur_frame = frame;
- stream-frame_size = fmt-fmt.pix.sizeimage;
done:
mutex_unlock(stream-mutex);
diff --git a/drivers/media/usb/uvc/uvc_video.c
b/drivers/media/usb/uvc/uvc_video.c index 9ace520..df81b9c 100644
--- a/drivers/media/usb/uvc/uvc_video.c
+++ b/drivers/media/usb/uvc/uvc_video.c
@@ -1143,7 +1143,7 @@ static int uvc_video_encode_data(struct uvc_streaming
*stream, static void uvc_video_validate_buffer(const struct uvc_streaming
*stream, struct uvc_buffer *buf)
{
- if (stream-frame_size != buf-bytesused
+ if (stream-ctrl.dwMaxVideoFrameSize != buf-bytesused
!(stream-cur_format-flags UVC_FMT_FLAG_COMPRESSED))
buf-error = 1;
}
diff --git a/drivers/media/usb/uvc/uvcvideo.h
b/drivers/media/usb/uvc/uvcvideo.h index f585c08..897cfd8 100644
--- a/drivers/media/usb/uvc/uvcvideo.h
+++ b/drivers/media/usb/uvc/uvcvideo.h
@@ -458,7 +458,6 @@ struct uvc_streaming {
struct uvc_format *def_format;
struct uvc_format *cur_format;
struct uvc_frame *cur_frame;
- size_t frame_size;
/* Protect access to ctrl, cur_format, cur_frame and hardware video
* probe control.
--
Regards,
Laurent Pinchart
--
To unsubscribe from this list: send the line unsubscribe linux-media in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: [PATCH] v4l: uvcvideo: Fix buffer completion size check
Hi Guennadi,
Ping ?
On Wednesday 01 October 2014 00:42:51 Laurent Pinchart wrote:
Commit e93e7fd9f5a3fffec7792dbcc4c3574653effda7 (v4l2: uvcvideo: Allow
using larger buffers) reworked the buffer size sanity check at buffer
completion time to use the frame size instead of the allocated buffer
size. However, it introduced two bugs in doing so:
- it assigned the allocated buffer size to the frame_size field, instead
of assigning the correct frame size
- it performed the assignment in the S_FMT handler, resulting in the
frame_size field being uninitialized if the userspace application
doesn't call S_FMT.
Fix both issues by removing the frame_size field and validating the
buffer size against the UVC video control dwMaxFrameSize.
Fixes: e93e7fd9f5a3 (v4l2: uvcvideo: Allow using larger buffers)
Signed-off-by: Laurent Pinchart laurent.pinch...@ideasonboard.com
---
drivers/media/usb/uvc/uvc_v4l2.c | 1 -
drivers/media/usb/uvc/uvc_video.c | 2 +-
drivers/media/usb/uvc/uvcvideo.h | 1 -
3 files changed, 1 insertion(+), 3 deletions(-)
Guennadi, could you please test and ack this ASAP, as the bug needs to be
fixed for v3.18-rc1 if possible ?
diff --git a/drivers/media/usb/uvc/uvc_v4l2.c
b/drivers/media/usb/uvc/uvc_v4l2.c index f205934..f33a067 100644
--- a/drivers/media/usb/uvc/uvc_v4l2.c
+++ b/drivers/media/usb/uvc/uvc_v4l2.c
@@ -318,7 +318,6 @@ static int uvc_v4l2_set_format(struct uvc_streaming
*stream, stream-ctrl = probe;
stream-cur_format = format;
stream-cur_frame = frame;
- stream-frame_size = fmt-fmt.pix.sizeimage;
done:
mutex_unlock(stream-mutex);
diff --git a/drivers/media/usb/uvc/uvc_video.c
b/drivers/media/usb/uvc/uvc_video.c index 9ace520..df81b9c 100644
--- a/drivers/media/usb/uvc/uvc_video.c
+++ b/drivers/media/usb/uvc/uvc_video.c
@@ -1143,7 +1143,7 @@ static int uvc_video_encode_data(struct uvc_streaming
*stream, static void uvc_video_validate_buffer(const struct uvc_streaming
*stream, struct uvc_buffer *buf)
{
- if (stream-frame_size != buf-bytesused
+ if (stream-ctrl.dwMaxVideoFrameSize != buf-bytesused
!(stream-cur_format-flags UVC_FMT_FLAG_COMPRESSED))
buf-error = 1;
}
diff --git a/drivers/media/usb/uvc/uvcvideo.h
b/drivers/media/usb/uvc/uvcvideo.h index f585c08..897cfd8 100644
--- a/drivers/media/usb/uvc/uvcvideo.h
+++ b/drivers/media/usb/uvc/uvcvideo.h
@@ -458,7 +458,6 @@ struct uvc_streaming {
struct uvc_format *def_format;
struct uvc_format *cur_format;
struct uvc_frame *cur_frame;
- size_t frame_size;
/* Protect access to ctrl, cur_format, cur_frame and hardware video
* probe control.
--
Regards,
Laurent Pinchart
--
To unsubscribe from this list: send the line unsubscribe linux-media in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
[PATCH] v4l: uvcvideo: Fix buffer completion size check
Commit e93e7fd9f5a3fffec7792dbcc4c3574653effda7 (v4l2: uvcvideo: Allow
using larger buffers) reworked the buffer size sanity check at buffer
completion time to use the frame size instead of the allocated buffer
size. However, it introduced two bugs in doing so:
- it assigned the allocated buffer size to the frame_size field, instead
of assigning the correct frame size
- it performed the assignment in the S_FMT handler, resulting in the
frame_size field being uninitialized if the userspace application
doesn't call S_FMT.
Fix both issues by removing the frame_size field and validating the
buffer size against the UVC video control dwMaxFrameSize.
Fixes: e93e7fd9f5a3 (v4l2: uvcvideo: Allow using larger buffers)
Signed-off-by: Laurent Pinchart laurent.pinch...@ideasonboard.com
---
drivers/media/usb/uvc/uvc_v4l2.c | 1 -
drivers/media/usb/uvc/uvc_video.c | 2 +-
drivers/media/usb/uvc/uvcvideo.h | 1 -
3 files changed, 1 insertion(+), 3 deletions(-)
Guennadi, could you please test and ack this ASAP, as the bug needs to be
fixed for v3.18-rc1 if possible ?
diff --git a/drivers/media/usb/uvc/uvc_v4l2.c b/drivers/media/usb/uvc/uvc_v4l2.c
index f205934..f33a067 100644
--- a/drivers/media/usb/uvc/uvc_v4l2.c
+++ b/drivers/media/usb/uvc/uvc_v4l2.c
@@ -318,7 +318,6 @@ static int uvc_v4l2_set_format(struct uvc_streaming *stream,
stream-ctrl = probe;
stream-cur_format = format;
stream-cur_frame = frame;
- stream-frame_size = fmt-fmt.pix.sizeimage;
done:
mutex_unlock(stream-mutex);
diff --git a/drivers/media/usb/uvc/uvc_video.c
b/drivers/media/usb/uvc/uvc_video.c
index 9ace520..df81b9c 100644
--- a/drivers/media/usb/uvc/uvc_video.c
+++ b/drivers/media/usb/uvc/uvc_video.c
@@ -1143,7 +1143,7 @@ static int uvc_video_encode_data(struct uvc_streaming
*stream,
static void uvc_video_validate_buffer(const struct uvc_streaming *stream,
struct uvc_buffer *buf)
{
- if (stream-frame_size != buf-bytesused
+ if (stream-ctrl.dwMaxVideoFrameSize != buf-bytesused
!(stream-cur_format-flags UVC_FMT_FLAG_COMPRESSED))
buf-error = 1;
}
diff --git a/drivers/media/usb/uvc/uvcvideo.h b/drivers/media/usb/uvc/uvcvideo.h
index f585c08..897cfd8 100644
--- a/drivers/media/usb/uvc/uvcvideo.h
+++ b/drivers/media/usb/uvc/uvcvideo.h
@@ -458,7 +458,6 @@ struct uvc_streaming {
struct uvc_format *def_format;
struct uvc_format *cur_format;
struct uvc_frame *cur_frame;
- size_t frame_size;
/* Protect access to ctrl, cur_format, cur_frame and hardware video
* probe control.
--
Regards,
Laurent Pinchart
--
To unsubscribe from this list: send the line unsubscribe linux-media in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: [PATCH] v4l: uvcvideo: Fix buffer completion size check
On Wednesday 01 October 2014 00:42:51 Laurent Pinchart wrote:
Commit e93e7fd9f5a3fffec7792dbcc4c3574653effda7 (v4l2: uvcvideo: Allow
using larger buffers) reworked the buffer size sanity check at buffer
completion time to use the frame size instead of the allocated buffer
size. However, it introduced two bugs in doing so:
- it assigned the allocated buffer size to the frame_size field, instead
of assigning the correct frame size
- it performed the assignment in the S_FMT handler, resulting in the
frame_size field being uninitialized if the userspace application
doesn't call S_FMT.
Fix both issues by removing the frame_size field and validating the
buffer size against the UVC video control dwMaxFrameSize.
Fixes: e93e7fd9f5a3 (v4l2: uvcvideo: Allow using larger buffers)
Signed-off-by: Laurent Pinchart laurent.pinch...@ideasonboard.com
---
drivers/media/usb/uvc/uvc_v4l2.c | 1 -
drivers/media/usb/uvc/uvc_video.c | 2 +-
drivers/media/usb/uvc/uvcvideo.h | 1 -
3 files changed, 1 insertion(+), 3 deletions(-)
Guennadi, could you please test and ack this ASAP, as the bug needs to be
fixed for v3.18-rc1 if possible ?
And, while we're at it, are you aware that the uvcvideo driver ignores the
sizeimage field in its S_FMT handler ? :-)
diff --git a/drivers/media/usb/uvc/uvc_v4l2.c
b/drivers/media/usb/uvc/uvc_v4l2.c index f205934..f33a067 100644
--- a/drivers/media/usb/uvc/uvc_v4l2.c
+++ b/drivers/media/usb/uvc/uvc_v4l2.c
@@ -318,7 +318,6 @@ static int uvc_v4l2_set_format(struct uvc_streaming
*stream, stream-ctrl = probe;
stream-cur_format = format;
stream-cur_frame = frame;
- stream-frame_size = fmt-fmt.pix.sizeimage;
done:
mutex_unlock(stream-mutex);
diff --git a/drivers/media/usb/uvc/uvc_video.c
b/drivers/media/usb/uvc/uvc_video.c index 9ace520..df81b9c 100644
--- a/drivers/media/usb/uvc/uvc_video.c
+++ b/drivers/media/usb/uvc/uvc_video.c
@@ -1143,7 +1143,7 @@ static int uvc_video_encode_data(struct uvc_streaming
*stream, static void uvc_video_validate_buffer(const struct uvc_streaming
*stream, struct uvc_buffer *buf)
{
- if (stream-frame_size != buf-bytesused
+ if (stream-ctrl.dwMaxVideoFrameSize != buf-bytesused
!(stream-cur_format-flags UVC_FMT_FLAG_COMPRESSED))
buf-error = 1;
}
diff --git a/drivers/media/usb/uvc/uvcvideo.h
b/drivers/media/usb/uvc/uvcvideo.h index f585c08..897cfd8 100644
--- a/drivers/media/usb/uvc/uvcvideo.h
+++ b/drivers/media/usb/uvc/uvcvideo.h
@@ -458,7 +458,6 @@ struct uvc_streaming {
struct uvc_format *def_format;
struct uvc_format *cur_format;
struct uvc_frame *cur_frame;
- size_t frame_size;
/* Protect access to ctrl, cur_format, cur_frame and hardware video
* probe control.
--
Regards,
Laurent Pinchart
--
To unsubscribe from this list: send the line unsubscribe linux-media in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
