Mailinglist Archive: opensuse-bugs (15100 mails)

[bugzilla_noreply@xxxxxxxxxx]

https://bugzilla.novell.com/show_bug.cgi?id=325917#c20


Thomas Renninger <trenn@xxxxxxxxxx> changed:

           What    |Removed                                         |Added
----------------------------------------------------------------------------
         AssignedTo|trenn@xxxxxxxxxx                                |teheo@xxxxxxxxxx
             Status|NEEDINFO                                        |NEW
      Info Provider|sragavan@xxxxxxxxxx                             |




--- Comment #20 from Thomas Renninger <trenn@xxxxxxxxxx>  2007-09-25 05:46:42 MST ---
Disabling C3 and C4 helps:
processor.max_cstate=2

First I thought this comes from no_hz and clocksource patches, but this could
be timer unrelated because:

The disk LED is on all the time (when deeper sleep states are entered)
hdparm -I |grep -i dma      shows:
 DMA: mdma0 mdma1 mdma2 udma0 udma1 *udma2 udma3 udma4 udma5

While on 10.2 udma5 is on by default:
 DMA: mdma0 mdma1 mdma2 udma0 udma1 udma2 udma3 udma4 *udma5

Trying to set it to udma5 using
hdparm -X udma5 /dev/sda
results in:

-------------------------
/dev/sda:
 setting xfermode to 69 (UltraDMA mode5)
SG_IO: bad/missing ATA_16 sense data::  70 00 05 00 00 00 00 0a 00 00 00 00 24
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
 HDIO_DRIVE_CMD(setxfermode) failed: Input/output error
-------------------------

syslog output is:
-------------------------
kernel: ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
kernel: ata1.00: cmd ca/00:08:78:6a:2f/00:00:00:00:00/e1 tag 0 cdb 0x1e data
4096 out
kernel:          res 40/00:03:00:00:00/00:00:00:00:00/b0 Emask 0x4 (timeout)
kernel: ata1: soft resetting link
kernel: ata1.00: configured for UDMA/33
kernel: ata1.01: configured for UDMA/33
kernel: ata1: EH complete
kernel: sd 0:0:0:0: [sda] 117210240 512-byte hardware sectors (60012 MB)
kernel: sd 0:0:0:0: [sda] Write Protect is off
kernel: sd 0:0:0:0: [sda] Mode Sense: 00 3a 00 00
kernel: sd 0:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't
support DPO or FUA
kernel: sd 0:0:0:0: [sda] 117210240 512-byte hardware sectors (60012 MB)
kernel: sd 0:0:0:0: [sda] Write Protect is off
kernel: sd 0:0:0:0: [sda] Mode Sense: 00 3a 00 00
kernel: sd 0:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't
support DPO or FUA
-------------------------


The machine works half way also in C3/C4 as long as no disk access is done.
I expect the real culprit is that the default dma mode (udma5?) cannot be set
for some reason. Then there happens dma interferences with deeper sleep states,
causing the whole machine to hang for some time.

Tejun can you help? I reassign this to you for now, don't know how to debug
this further. Srinivasa (thanks a lot for your help!) can give you ssh access
to the machine. Best you ping each other via GW messenger.


-- 
Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.