Hi Nicholas,
I recently ran a test on a LIO target system. I used a Windows 7 Pro
box with an ATTO card running iometer (www.iometer.org) hooked to a
Linux target running kernel 4.0.4 with a QLE-2562. Just doing a read
with a 5MB transfer size fails! I thought this was all fixed with
commit "11e764bd5ed4bb930e0ec5dd161df58307507347, 09-May-2012, Nicholas
Bellinger target: Remove max_sectors device attribute for modern
se_task less code" and maybe commit
"046ba64285a4389ae5e9a7dfa253c6bff3d7c341, 07-Jan-2015, Nicholas
Bellinger target: Drop arbitrary maximum I/O size limit" (and a few others)?
I checked the block sizes for the disk I was using:
# cat /sys/block/sdb/queue/max_hw_sectors_kb
32767
# cat /sys/block/sdb/queue/max_sectors_kb
32767
# cat /sys/block/sdb/queue/hw_sector_size
512
From what I understand, this block size should be the limiting factor
now? I was hoping to get closer to 30MB transfer sizes.
Basically, the read fails. No data is transferred. Watching the LsCroy
FC protocol analyzer, the read command is sent by the initiator and the
LIO target immediately drops the link. This is followed by a bunch of
low-level arbitration as the initiator tries to reconnect. During this,
the target console is printing "Error entry - invalid handle/queue."
messages (100s of these). Every now and then, the connection is
re-established and the command is resent with identical results. Here's
what is printed when the link gets re-established (LOOP UP):
...
Jun 29 17:04:34 fc82 kernel: qla2xxx [0000:01:00.1]-5030:9: Error entry
- invalid handle/queue.
Jun 29 17:04:34 fc82 kernel: qla2xxx [0000:01:00.1]-5030:9: Error entry
- invalid handle/queue.
Jun 29 17:04:34 fc82 kernel: qla2xxx [0000:01:00.1]-5030:9: Error entry
- invalid handle/queue.
Jun 29 17:04:34 fc82 kernel: qla2xxx [0000:01:00.1]-5030:9: Error entry
- invalid handle/queue.
Jun 29 17:04:34 fc82 kernel: qla2xxx [0000:01:00.1]-5030:9: Error entry
- invalid handle/queue.
Jun 29 17:04:34 fc82 kernel: qla2xxx [0000:01:00.1]-5030:9: Error entry
- invalid handle/queue.
Jun 29 17:04:34 fc82 kernel: qla2xxx [0000:01:00.1]-5030:9: Error entry
- invalid handle/queue.
Jun 29 17:04:34 fc82 kernel: qla2xxx [0000:01:00.1]-5030:9: Error entry
- invalid handle/queue.
Jun 29 17:04:35 fc82 kernel: qla2xxx [0000:01:00.1]-500a:9: LOOP UP
detected (8 Gbps).
Jun 29 17:04:35 fc82 kernel: qla2xxx [0000:01:00.1]-0121:9: Failed to
enable receiving of RSCN requests: 0x2.
Jun 29 17:04:36 fc82 kernel: qla2xxx [0000:01:00.1]-00af:9: Performing
ISP error recovery - ha=ffff880170914000.
Jun 29 17:04:39 fc82 kernel: qla2xxx [0000:01:00.1]-500a:9: LOOP UP
detected (8 Gbps).
Jun 29 17:04:39 fc82 kernel: qla2xxx [0000:01:00.1]-0121:9: Failed to
enable receiving of RSCN requests: 0x2.
Jun 29 17:04:41 fc82 kernel: qla2xxx [0000:01:00.1]-5030:9: Error entry
- invalid handle/queue.
Jun 29 17:04:41 fc82 kernel: qla2xxx [0000:01:00.1]-5030:9: Error entry
- invalid handle/queue.
Jun 29 17:04:41 fc82 kernel: qla2xxx [0000:01:00.1]-5030:9: Error entry
- invalid handle/queue.
Jun 29 17:04:41 fc82 kernel: qla2xxx [0000:01:00.1]-5030:9: Error entry
- invalid handle/queue.
...
Reads sized at 4MB work fine BTW.
This is a 4.0.4 Linux Kernel with targetcli_fb-2.1.fb38,
rtslib_fb-2.1.51, and configshell_fb-1.1.16.
Thanks for any suggestions/help.
Craig Watson
--
To unsubscribe from this list: send the line "unsubscribe target-devel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
