This patch makes the write completion handler use the is_channel_healthy flag to prevent the hdm from scheduling a second clear_halt workqueue in case an endpoint reported a STALL condition. Signed-off-by: Christian Gromm <christian.gromm@xxxxxxxxxxxxx> --- drivers/staging/most/hdm-usb/hdm_usb.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/staging/most/hdm-usb/hdm_usb.c b/drivers/staging/most/hdm-usb/hdm_usb.c index c2ba5a1..57d6c6d 100644 --- a/drivers/staging/most/hdm-usb/hdm_usb.c +++ b/drivers/staging/most/hdm-usb/hdm_usb.c @@ -418,6 +418,7 @@ static void hdm_write_completion(struct urb *urb) case -EPIPE: dev_warn(dev, "Broken OUT pipe detected\n"); most_stop_enqueue(&mdev->iface, channel); + mdev->is_channel_healthy[channel] = false; mbo->status = MBO_E_INVAL; mdev->clear_work[channel].pipe = urb->pipe; schedule_work(&mdev->clear_work[channel].ws); @@ -582,6 +583,7 @@ static void hdm_read_completion(struct urb *urb) case -EPIPE: dev_warn(dev, "Broken IN pipe detected\n"); most_stop_enqueue(&mdev->iface, channel); + mdev->is_channel_healthy[channel] = false; mbo->status = MBO_E_INVAL; mdev->clear_work[channel].pipe = urb->pipe; schedule_work(&mdev->clear_work[channel].ws); @@ -938,6 +940,7 @@ static void wq_clear_halt(struct work_struct *wq_obj) if (usb_clear_halt(mdev->usb_device, pipe)) dev_warn(&mdev->usb_device->dev, "Failed to reset endpoint.\n"); + mdev->is_channel_healthy[channel] = true; most_resume_enqueue(&mdev->iface, channel); mutex_unlock(&mdev->io_mutex); } -- 1.9.1 _______________________________________________ devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxx http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
