2012/4/7 Sarah Sharp <sarah.a.sharp@xxxxxxxxxxxxxxx>:
> On Fri, Apr 06, 2012 at 11:14:46AM -0700, Greg KH wrote:
>> On Fri, Apr 06, 2012 at 10:16:16AM -0700, Sarah Sharp wrote:
>> > On Fri, Apr 06, 2012 at 08:10:44AM -0700, Greg KH wrote:
>> > > That was a bug in 3.4-rc1, it is now fixed in Linus's tree.
>> >
>> > Any chance you can pull linus' tree into your usb-linus branch? That
>> > will make testing these bug fix patches easier.
>>
>> My branch right now is at 3.4-rc1 with no patches on it. That doesn't
>> stop you from putting your patches on whatever you want right now for
>> testing. You can always rebase when sending to me if needed. I'm not
>> going to be able to get to any patches until Monday, so there's no rush.
>
> Ok, I've pushed an updated branch that's based on the HEAD commit of
> Linus' master branch. Please re-test, and hopefully you won't run into
> any non-xHCI related hangs.
>
> Elric, I'll take a closer look at your USB 3.0 device oops on Monday.
> Any chance you can use git bisect to figure out which patch caused the
> warning? I'm wondering if it's related to some SCSI core changes, or
> perhaps it's caused by the ring expansion patches?
Hi Sarah,
I found the reason caused the oops. It is not a xHCI driver issue but a SCSI
core issue.
778 static int scsi_send_eh_cmnd(struct scsi_cmnd *scmd, unsigned char *cmnd,
779 int cmnd_size, int timeout, unsigned
sense_bytes)
780 {
781 struct scsi_device *sdev = scmd->device;
782 struct scsi_driver *sdrv = scsi_cmd_to_driver(scmd);
783 struct Scsi_Host *shost = sdev->host;
784 DECLARE_COMPLETION_ONSTACK(done);
785 unsigned long timeleft;
786 struct scsi_eh_save ses;
787 int rtn;
The oops will occur while the driver run to line 782. It was added in 3.4-rc1,
the Commit ID is 18a4d0a22ed6c54b67af7718c305cd010f09ddf8. And
the subject is "[SCSI] Handle disk devices which can not process medium".
I am not familiar with the SCSI core. It seems like scsi_cmd_to_driver() has
an issue. But the issue is very random. It only happened while the device
was reset by mass storage driver due to some scsi commands failed and
re-enumerated in device initialization process. And even if the device is
reset, the issue doesn't always occur. So should we forward the mail to
linux-scsi?
Best Regards,
Elric
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
