[PATCH unfinished update] firewire: sbp2: parallelize login/inquiry,reconnect, and shutdown

From: Stefan Richter
Date: Tue Oct 12 2010 - 17:39:30 EST


So I changed the patch to the simple one below, but suddenly my
INIC-2430 device has trouble with inquiry in parallel to its two LUNs:

Oct 12 23:23:50 stein kernel: scsi22 : SBP-2 IEEE-1394
Oct 12 23:23:50 stein kernel: firewire_sbp2: Workarounds for fw7.0: 0x2 (firmware_revision 0x000241, model_id 0x000000)
Oct 12 23:23:50 stein kernel: scsi23 : SBP-2 IEEE-1394
Oct 12 23:23:50 stein kernel: firewire_sbp2: Workarounds for fw7.1: 0x2 (firmware_revision 0x000241, model_id 0x000000)
Oct 12 23:23:50 stein kernel: firewire_core: created device fw7: GUID 00027a0e440020c2, S800
Oct 12 23:23:50 stein kernel: firewire_sbp2: fw7.0: logged in to LUN 0000 (0 retries)
Oct 12 23:23:50 stein kernel: firewire_sbp2: fw7.1: logged in to LUN 0001 (0 retries)
Oct 12 23:23:52 stein kernel: scsi 23:0:0:1: CD-ROM TEAC DV-516GC XT11 PQ: 0 ANSI: 0
Oct 12 23:23:52 stein kernel: sr1: scsi3-mmc drive: 48x/48x cd/rw xa/form2 cdda tray
Oct 12 23:23:52 stein kernel: sr 23:0:0:1: Attached scsi CD-ROM sr1
Oct 12 23:23:52 stein kernel: sr 23:0:0:1: Attached scsi generic sg6 type 5
Oct 12 23:23:52 stein kernel: firewire_sbp2: fw7.0: orb reply timed out, rcode=0x11

When I then force a bus reset by hardware or software, the device comes
to its senses:

Oct 12 23:24:38 stein kernel: firewire_sbp2: fw7.0: logged in to LUN 0000 (0 retries)
Oct 12 23:24:38 stein kernel: firewire_sbp2: fw7.1: reconnected to LUN 0001 (0 retries)
Oct 12 23:24:38 stein kernel: scsi 22:0:0:0: CD-ROM TEAC DV-516GC XT11 PQ: 0 ANSI: 0
Oct 12 23:24:38 stein kernel: sr2: scsi3-mmc drive: 48x/48x cd/rw xa/form2 cdda tray
Oct 12 23:24:38 stein kernel: sr 22:0:0:0: Attached scsi CD-ROM sr2
Oct 12 23:24:38 stein kernel: sr 22:0:0:0: Attached scsi generic sg7 type 5

The initial Inquiry(?) timeout happens not always but quite often. It
did not happen when I tested the earlier version of the patch in which I
used system_nrt_wq.

Maybe I need ordered per-target workqueues. Not Signed-off-by me yet.

Tejun, can a workqueue job call destroy_workqueue on its own queue?
---
drivers/firewire/sbp2.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)

Index: b/drivers/firewire/sbp2.c
===================================================================
--- a/drivers/firewire/sbp2.c
+++ b/drivers/firewire/sbp2.c
@@ -1656,7 +1656,8 @@ MODULE_ALIAS("sbp2");

static int __init sbp2_init(void)
{
- sbp2_wq = create_singlethread_workqueue(KBUILD_MODNAME);
+ sbp2_wq = alloc_workqueue(KBUILD_MODNAME,
+ WQ_NON_REENTRANT | WQ_RESCUER, 0);
if (!sbp2_wq)
return -ENOMEM;


--
Stefan Richter
-=====-==-=- =-=- -==--
http://arcgraph.de/sr/

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/