lk/linux-kernel-test
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

USB: Dealias -110 code (more complete)

Browse Source 
The purpose of this patch is to split off the case when a device does
not reply on the lower level (which is reported by HC hardware), and
a case when the device accepted the request, but does not reply at
upper level. This redefinition allows to diagnose issues easier,
without asking the user if the -110 happened "immediately".

The usbmon splits such cases already thanks to its timestamp, but
it's not always available.

I adjusted all drivers which I found affected (by searching for "urb").
Out of tree drivers may suffer a little bit, but I do not expect much
breakage. At worst they may print a few messages.

Signed-off-by: Pete Zaitcev <zaitcev@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This commit is contained in:
Pete Zaitcev
2006-09-18 22:49:02 -07:00
committed by Greg Kroah-Hartman
parent ec17cf1cfe
commit 38e2bfc94e
24 changed files with 48 additions and 60 deletions
Download Patch File Download Diff File Expand all files Collapse all files

18
drivers/net/irda/irda-usb.c
View File

@ -671,10 +671,8 @@ static void irda_usb_net_timeout(struct net_device *netdev)
* Jean II */
done = 1;
break;
case -ECONNABORTED: /* -103 */
case -ECONNRESET: /* -104 */
case -ETIMEDOUT: /* -110 */
case -ENOENT: /* -2 (urb unlinked by us) */
case -ECONNRESET:
case -ENOENT: /* urb unlinked by us */
default: /* ??? - Play safe */
urb->status = 0;
netif_wake_queue(self->netdev);
@ -712,10 +710,8 @@ static void irda_usb_net_timeout(struct net_device *netdev)
* Jean II */
done = 1;
break;
case -ECONNABORTED: /* -103 */
case -ECONNRESET: /* -104 */
case -ETIMEDOUT: /* -110 */
case -ENOENT: /* -2 (urb unlinked by us) */
case -ECONNRESET:
case -ENOENT: /* urb unlinked by us */
default: /* ??? - Play safe */
if(skb != NULL) {
dev_kfree_skb_any(skb);
@ -845,14 +841,14 @@ static void irda_usb_receive(struct urb *urb, struct pt_regs *regs)
self->stats.rx_crc_errors++;
/* Also precursor to a hot-unplug on UHCI. */
/* Fallthrough... */
case -ECONNRESET: /* -104 */
case -ECONNRESET:
/* Random error, if I remember correctly */
/* uhci_cleanup_unlink() is going to kill the Rx
* URB just after we return. No problem, at this
* point the URB will be idle ;-) - Jean II */
case -ESHUTDOWN: /* -108 */
case -ESHUTDOWN:
/* That's usually a hot-unplug. Submit will fail... */
case -ETIMEDOUT: /* -110 */
case -ETIME:
/* Usually precursor to a hot-unplug on OHCI. */
default:
self->stats.rx_errors++;