* [DONTCOMMIT/WIP] gspca: Check device presence before accessing hardware
@ 2019-02-28 15:30 Ezequiel Garcia
0 siblings, 0 replies; only message in thread
From: Ezequiel Garcia @ 2019-02-28 15:30 UTC (permalink / raw)
To: linux-media; +Cc: Hans Verkuil, kernel, ccr, Ezequiel Garcia
sd_desc->stop0 and sd_desc->stopN hooks are meant to access
the device hardware, stopping the transfer. Thus, hardware
presence should be checked before calling them.
This means that sd_desc.start is not allowed to allocate resources,
since sd_desc.stop0 and sd_desc.stopN won't be releasing when
the USB device is disconnected.
TODO: Check what's the deal with the konica device, which is allocating
URB memory in sd_desc.start. This seems wrong.
Signed-off-by: Ezequiel Garcia <ezequiel@collabora.com>
---
drivers/media/usb/gspca/gspca.c | 11 ++++++-----
drivers/media/usb/gspca/gspca.h | 6 +++---
2 files changed, 9 insertions(+), 8 deletions(-)
diff --git a/drivers/media/usb/gspca/gspca.c b/drivers/media/usb/gspca/gspca.c
index 4d7517411cc2..98b68a7a52c0 100644
--- a/drivers/media/usb/gspca/gspca.c
+++ b/drivers/media/usb/gspca/gspca.c
@@ -745,15 +745,16 @@ static void gspca_stream_off(struct gspca_dev *gspca_dev)
{
gspca_dev->streaming = false;
gspca_dev->usb_err = 0;
- if (gspca_dev->sd_desc->stopN)
+ if (gspca_dev->present && gspca_dev->sd_desc->stopN)
gspca_dev->sd_desc->stopN(gspca_dev);
destroy_urbs(gspca_dev);
gspca_input_destroy_urb(gspca_dev);
- gspca_set_alt0(gspca_dev);
- if (gspca_dev->present)
+ if (gspca_dev->present) {
+ gspca_set_alt0(gspca_dev);
gspca_input_create_urb(gspca_dev);
- if (gspca_dev->sd_desc->stop0)
- gspca_dev->sd_desc->stop0(gspca_dev);
+ if (gspca_dev->sd_desc->stop0)
+ gspca_dev->sd_desc->stop0(gspca_dev);
+ }
gspca_dbg(gspca_dev, D_STREAM, "stream off OK\n");
}
diff --git a/drivers/media/usb/gspca/gspca.h b/drivers/media/usb/gspca/gspca.h
index b0ced2e14006..e0ca120d2e31 100644
--- a/drivers/media/usb/gspca/gspca.h
+++ b/drivers/media/usb/gspca/gspca.h
@@ -105,13 +105,13 @@ struct sd_desc {
cam_cf_op config; /* called on probe */
cam_op init; /* called on probe and resume */
cam_op init_controls; /* called on probe */
- cam_op start; /* called on stream on after URBs creation */
+ cam_op start; /* called on stream on after URBs creation. Can't allocate any resources here! */
cam_pkt_op pkt_scan;
/* optional operations */
cam_op isoc_init; /* called on stream on before getting the EP */
cam_op isoc_nego; /* called when URB submit failed with NOSPC */
- cam_v_op stopN; /* called on stream off - main alt */
- cam_v_op stop0; /* called on stream off & disconnect - alt 0 */
+ cam_v_op stopN; /* called on stream off, only if usb device is present - main alt */
+ cam_v_op stop0; /* called on stream off, only if usb device is present - alt 0 */
cam_v_op dq_callback; /* called when a frame has been dequeued */
cam_get_jpg_op get_jcomp;
cam_set_jpg_op set_jcomp;
--
2.20.1
^ permalink raw reply related [flat|nested] only message in thread
only message in thread, other threads:[~2019-02-28 15:30 UTC | newest]
Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2019-02-28 15:30 [DONTCOMMIT/WIP] gspca: Check device presence before accessing hardware Ezequiel Garcia
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox