[PATCH v2] scsi: fix async probe regression

[PATCH v2] scsi: fix async probe regression

[Dan Williams]

Commit a7a20d1 "[SCSI] sd: limit the scope of the async probe domain"
moved sd probe work out of reach of wait_for_device_probe().  Allow it
to be synced via scsi_complete_async_scans().

Cc: David Miller <davem@davemloft.net>
Cc: Alan Stern <stern@rowland.harvard.edu>
Reported-by: Meelis Roos <mroos@linux.ee>
Tested-by: Meelis Roos <mroos@linux.ee>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---

 v2: James: don't wrapper scsi_complete_async_scans()

 Kept the Tested-by tag since this is equivalent to v1.

 --
 Dan

 drivers/scsi/scsi_scan.c |    7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/scsi/scsi_scan.c b/drivers/scsi/scsi_scan.c
index 01b0374..2e5fe58 100644
--- a/drivers/scsi/scsi_scan.c
+++ b/drivers/scsi/scsi_scan.c
@@ -147,7 +147,7 @@ int scsi_complete_async_scans(void)
 
 	do {
 		if (list_empty(&scanning_hosts))
-			return 0;
+			goto out;
 		/* If we can't get memory immediately, that's OK.  Just
 		 * sleep a little.  Even if we never get memory, the async
 		 * scans will finish eventually.
@@ -179,8 +179,11 @@ int scsi_complete_async_scans(void)
 	}
  done:
 	spin_unlock(&async_scan_lock);
-
 	kfree(data);
+
+ out:
+	async_synchronize_full_domain(&scsi_sd_probe_domain);
+
 	return 0;
 }
 

Re: [PATCH v2] scsi: fix async probe regression

[walt]

On 05/25/2012 10:55 AM, Dan Williams wrote:
> Commit a7a20d1 "[SCSI] sd: limit the scope of the async probe domain"
> moved sd probe work out of reach of wait_for_device_probe().  Allow it
> to be synced via scsi_complete_async_scans().

Hi Dan.  After applying version 2 of a7a20d1 I can boot again, but there's
a catch :)

My grub.cfg boot menu entry uses the root=PARTUUID=<guid> syntax, but that
has stopped working.  After a7a20d1(v2) I have to use root=/dev/sda5 or I
get the familiar "can't find block device (foo,bar)" kernel panic.

Any idea what would cause that particular kind of breakage?

Thanks!

Re: [PATCH v2] scsi: fix async probe regression

[Dan Williams]

On Tue, May 29, 2012 at 5:53 PM, walt <w41ter@gmail.com> wrote:
> On 05/25/2012 10:55 AM, Dan Williams wrote:
>> Commit a7a20d1 "[SCSI] sd: limit the scope of the async probe domain"
>> moved sd probe work out of reach of wait_for_device_probe().  Allow it
>> to be synced via scsi_complete_async_scans().
>
> Hi Dan.  After applying version 2 of a7a20d1 I can boot again, but there's
> a catch :)

Ah you took me off the cc, so I missed this originally.

>
> My grub.cfg boot menu entry uses the root=PARTUUID=<guid> syntax, but that
> has stopped working.  After a7a20d1(v2) I have to use root=/dev/sda5 or I
> get the familiar "can't find block device (foo,bar)" kernel panic.

What is a7a20d1(v2)?  That's the original commit that broke things, so
I'm confused.

> Any idea what would cause that particular kind of breakage?

The fix should be equivalent for both root= options because if the
device can be found by /dev/sda5 I don't see anything related to the
async subsystem that would prevent it from being looked up by uuid??

--
Dan
--
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html