* SCSI strangeness on Ultra2
@ 1999-10-06 18:08 Ryan Cresawn
1999-10-06 18:51 ` David S. Miller
` (3 more replies)
0 siblings, 4 replies; 5+ messages in thread
From: Ryan Cresawn @ 1999-10-06 18:08 UTC (permalink / raw)
To: ultralinux
I'm having some trouble on my Ultra2 with the SCSI busses. If I am
accessing the ZIP 250 drive on the second SCSI bus then it will really
confuse the whole system. Keyboard doesn't take input, programs
freeze their operation, X freezes for minutes and then begins to
work. If you would like more info let me know.
*** output of `uname -a' ***
Linux grimlock.larc.nasa.gov 2.2.12 #1 Wed Oct 6 10:06:57 EDT 1999 sparc64 unknown
*** trimmed output of `dmesg' ***
esp0: IRQ 3,7e0 SCSI ID 7 Clk 40MHz CCF=8 TOut 167 NCR53C9XF(espfast)
esp1: IRQ 7,7c3 SCSI ID 7 Clk 25MHz CCF=5 TOut 167 NCR53C90A(esp100a)
ESP: Total of 2 ESP hosts found, 2 actually in use.
scsi0 : Sparc ESP366-HME
scsi1 : Sparc ESP100A (NCR53C90A)
scsi : 2 hosts.
Vendor: FUJITSU Model: MAB3045S SUN4.2G Rev: 1806
Type: Direct-Access ANSI SCSI revision: 02
Detected scsi disk sda at scsi0, channel 0, id 1, lun 0
Vendor: FUJITSU Model: MAE3182LC Rev: 0108
Type: Direct-Access ANSI SCSI revision: 02
Detected scsi disk sdb at scsi0, channel 0, id 2, lun 0
Vendor: TOSHIBA Model: XM5701TASUN12XCD Rev: 2395
Type: CD-ROM ANSI SCSI revision: 02
Detected scsi CD-ROM sr0 at scsi0, channel 0, id 6, lun 0
Vendor: IOMEGA Model: ZIP 250 Rev: H.41
Type: Direct-Access ANSI SCSI revision: 02
Detected scsi removable disk sdc at scsi1, channel 0, id 5, lun 0
scsi : detected 1 SCSI cdrom 3 SCSI disks total.
esp0: target 6 asynchronous
Uniform CDROM driver Revision: 2.55
esp0: target 1 [period 100ns offset 15 20.00MHz FAST-WIDE SCSI-II]
SCSI device sda: hdwr sector= 512 bytes. Sectors= 8385121 [4094 MB] [4.1 GB]
esp0: target 2 [period 100ns offset 15 20.00MHz FAST-WIDE SCSI-II]
SCSI device sdb: hdwr sector= 512 bytes. Sectors= 35700480 [17431 MB] [17.4 GB]
scsi : aborting command due to timeout : pid 60, scsi1, channel 0, id 5, lun 0 Read Capacity 00 00 00 00 00 00 00
00 00
esp1: Aborting command
esp1: dumping state
esp1: dma -- cond_reg<94000010> addr<0e010001>
esp1: SW [sreg<06> sstep<01> ireg<10>]
esp1: HW reread [sreg<06> sstep<c1> ireg<10>]
esp1: current command [tgt<05> lun<00> pphase<MSGOUT> cphase<MSGOUTDONE>]
esp1: disconnected
SCSI host 1 abort (pid 60) timed out - resetting
SCSI bus is being reset for host 1 channel 0.
esp1: Resetting scsi bus
esp1: Gross error sreg@
esp1: SCSI bus reset interrupt
esp1: illegal command
esp1: dumping state
esp1: dma -- cond_reg<96000201> addr<0e01000b>
esp1: SW [sreg<06> sstep<01> ireg<40>]
esp1: HW reread [sreg<93> sstep<84> ireg<18>]
esp1: current command [tgt<05> lun<00> pphase<UNISSUED> cphase<SLCTNORM>]
esp1: disconnected
esp1: Forcing async and disabling disconnect for target 5
esp1: Resetting scsi bus
esp1: SCSI bus reset interrupt
esp1: SCSI bus reset interrupt
--
Ryan
^ permalink raw reply [flat|nested] 5+ messages in thread* Re: SCSI strangeness on Ultra2
1999-10-06 18:08 SCSI strangeness on Ultra2 Ryan Cresawn
@ 1999-10-06 18:51 ` David S. Miller
1999-10-06 19:04 ` Ryan Cresawn
` (2 subsequent siblings)
3 siblings, 0 replies; 5+ messages in thread
From: David S. Miller @ 1999-10-06 18:51 UTC (permalink / raw)
To: ultralinux
From: Ryan Cresawn <cresawn@vaquero.LaRC.NASA.GOV>
Date: Wed, 6 Oct 1999 14:08:24 -0400 (EDT)
I'm having some trouble on my Ultra2 with the SCSI busses. If I am
accessing the ZIP 250 drive on the second SCSI bus then it will
really confuse the whole system. Keyboard doesn't take input,
programs freeze their operation, X freezes for minutes and then
begins to work. If you would like more info let me know.
Aparently the ZIP drive does not like to negotiate sync/wide SCSI
parameters, that is why the SCSI bus locked up and the chain needed
to be reset.
Happily all of my error recovery seems to work and you get things
working after the timeout is hit and the scsi bus is reset.
Later,
David S. Miller
davem@redhat.com
^ permalink raw reply [flat|nested] 5+ messages in thread* Re: SCSI strangeness on Ultra2
1999-10-06 18:08 SCSI strangeness on Ultra2 Ryan Cresawn
1999-10-06 18:51 ` David S. Miller
@ 1999-10-06 19:04 ` Ryan Cresawn
1999-10-06 19:20 ` David S. Miller
1999-10-06 19:31 ` Ryan Cresawn
3 siblings, 0 replies; 5+ messages in thread
From: Ryan Cresawn @ 1999-10-06 19:04 UTC (permalink / raw)
To: ultralinux
David S. Miller writes:
> From: Ryan Cresawn <cresawn@vaquero.LaRC.NASA.GOV>
> Date: Wed, 6 Oct 1999 14:08:24 -0400 (EDT)
>
> I'm having some trouble on my Ultra2 with the SCSI busses. If I am
> accessing the ZIP 250 drive on the second SCSI bus then it will
> really confuse the whole system. Keyboard doesn't take input,
> programs freeze their operation, X freezes for minutes and then
> begins to work. If you would like more info let me know.
>
> Aparently the ZIP drive does not like to negotiate sync/wide SCSI
> parameters, that is why the SCSI bus locked up and the chain needed
> to be reset.
>
> Happily all of my error recovery seems to work and you get things
> working after the timeout is hit and the scsi bus is reset.
The renegotiation information that appeared in dmesg was present at
boot time and was never repeated. The problem I had was that since I
was trying to copy about 40 MBs to the ZIP drive at once it froze
almost everything for minutes at a time, then would release it, then
would freeze again, then would release it...you get the idea.
Anyways, this pattern continued for maybe 30-45 minutes until the copy
operation was complete. It's true that I'm happy the computer didn't
permanently lock up, but I was unhappy to find my parallel port ZIP
drive on my PC to be much faster.
--
Ryan
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: SCSI strangeness on Ultra2
1999-10-06 18:08 SCSI strangeness on Ultra2 Ryan Cresawn
1999-10-06 18:51 ` David S. Miller
1999-10-06 19:04 ` Ryan Cresawn
@ 1999-10-06 19:20 ` David S. Miller
1999-10-06 19:31 ` Ryan Cresawn
3 siblings, 0 replies; 5+ messages in thread
From: David S. Miller @ 1999-10-06 19:20 UTC (permalink / raw)
To: ultralinux
From: Ryan Cresawn <cresawn@vaquero.LaRC.NASA.GOV>
Date: Wed, 6 Oct 1999 15:04:41 -0400 (EDT)
The renegotiation information that appeared in dmesg was present at
boot time and was never repeated. The problem I had was that since
I was trying to copy about 40 MBs to the ZIP drive at once it froze
almost everything for minutes at a time, then would release it,
then would freeze again, then would release it...you get the idea.
The reset policy when a device locks up the SCSI bus is to
turn off all "features" to that device, which includes synchronous
data transfers and disconnect. The lack of disconnect is almost
surely what is causing the problem, when a read/write is issued
to the drive, that SCSI bus is effectively unusable by any other
SCSI device on it until the ZIP drive completes the read/write
request.
Ho hum... please give this kernel patch a try, it will allow
disconnects to removable disk media and avoid the negotations, which
should solve the bus reset and the freeze up problems for you:
--- drivers/scsi/esp.c.~1~ Wed Sep 22 10:31:15 1999
+++ drivers/scsi/esp.c Wed Oct 6 12:17:18 1999
@@ -1520,7 +1520,7 @@
if((idprom->id_machtype = (SM_SUN4C | SM_4C_SS1)) ||
(idprom->id_machtype = (SM_SUN4C | SM_4C_SS1PLUS))) {
/* But we are nice and allow tapes to disconnect. */
- if(SDptr->type = TYPE_TAPE)
+ if(SDptr->type = TYPE_TAPE || SDptr->removable)
SDptr->disconnect = 1;
else
SDptr->disconnect = 0;
@@ -1538,7 +1538,8 @@
*/
if(esp->erev = fashme && !SDptr->wide) {
if(!SDptr->borken &&
- SDptr->type != TYPE_ROM) {
+ SDptr->type != TYPE_ROM &&
+ SDptr->removable = 0) {
build_wide_nego_msg(esp, 16);
SDptr->wide = 1;
esp->wnip = 1;
@@ -1556,6 +1557,11 @@
"CDROM.\n", esp->esp_id));
cdrom_hwbug_wkaround = 1;
build_sync_nego_msg(esp, 0, 0);
+ } else if (SDptr->removable != 0) {
+ ESPMISC(("esp%d: Not negotiating sync/wide but "
+ "allowing disconnect for removable media.\n",
+ esp->esp_id));
+ build_sync_nego_msg(esp, 0, 0);
} else {
build_sync_nego_msg(esp, esp->sync_defp, 15);
}
@@ -1589,7 +1595,9 @@
* Therefore _no_ disconnects for SCSI1 targets
* thank you very much. ;-)
*/
- if(((SDptr->scsi_level < 3) && (SDptr->type != TYPE_TAPE)) ||
+ if(((SDptr->scsi_level < 3) &&
+ (SDptr->type != TYPE_TAPE) &&
+ SDptr->removable = 0) ||
cdrom_hwbug_wkaround || SDptr->borken) {
ESPMISC((KERN_INFO "esp%d: Disabling DISCONNECT for target %d "
"lun %d\n", esp->esp_id, SCptr->target, SCptr->lun));
^ permalink raw reply [flat|nested] 5+ messages in thread* Re: SCSI strangeness on Ultra2
1999-10-06 18:08 SCSI strangeness on Ultra2 Ryan Cresawn
` (2 preceding siblings ...)
1999-10-06 19:20 ` David S. Miller
@ 1999-10-06 19:31 ` Ryan Cresawn
3 siblings, 0 replies; 5+ messages in thread
From: Ryan Cresawn @ 1999-10-06 19:31 UTC (permalink / raw)
To: ultralinux
David S. Miller writes:
> From: Ryan Cresawn <cresawn@vaquero.LaRC.NASA.GOV>
> Date: Wed, 6 Oct 1999 15:04:41 -0400 (EDT)
>
> The renegotiation information that appeared in dmesg was present at
> boot time and was never repeated. The problem I had was that since
> I was trying to copy about 40 MBs to the ZIP drive at once it froze
> almost everything for minutes at a time, then would release it,
> then would freeze again, then would release it...you get the idea.
>
> The reset policy when a device locks up the SCSI bus is to
> turn off all "features" to that device, which includes synchronous
> data transfers and disconnect. The lack of disconnect is almost
> surely what is causing the problem, when a read/write is issued
> to the drive, that SCSI bus is effectively unusable by any other
> SCSI device on it until the ZIP drive completes the read/write
> request.
Just so you know, I didn't have the cable I needed to connect the ZIP
250 to esp0 but I did have another SCSI card handy. I installed esp1
and connected the ZIP 250 without trouble. There are no other SCSI
devices attached to it. The three other SCSI devices are on esp0.
Does this jive with your reasoning as to why the computer was
freezing?
Thanks for the patch. I'll apply it and let you know how it goes.
--
Ryan
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~1999-10-06 19:31 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
1999-10-06 18:08 SCSI strangeness on Ultra2 Ryan Cresawn
1999-10-06 18:51 ` David S. Miller
1999-10-06 19:04 ` Ryan Cresawn
1999-10-06 19:20 ` David S. Miller
1999-10-06 19:31 ` Ryan Cresawn
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.