Re: sata_sil24 stability and performance

[Tejun Heo <htejun@gmail.com>]

Denys Dmytriyenko wrote:
> Tejun et al,
> 
> I wanted to thank you all for the support you provided. I ended up replacing 
> MoBo+CPU to Intel based ones (and previously mentioned single-rail PSU) and 
> now everything works rock solid for over a week now! Thanks again.
> 
>>> BTW, issuing "smartctl -a" on a drive in standby, throws this exception:
>>>
>>> Mar 20 18:16:53 [kernel] ata10.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
>>> Mar 20 18:16:53 [kernel] ata10.00: cmd b0/da:00:00:4f:c2/00:00:00:00:00/00 tag 0 cdb 0x0 data 0 
>>> Mar 20 18:16:53 [kernel]          res 40/00:01:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout)
>>> Mar 20 18:16:53 [kernel] ata10: soft resetting port
>>> Mar 20 18:16:54 [kernel] ata10: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
>>> Mar 20 18:16:54 [kernel] ata10.00: configured for UDMA/100
>>> Mar 20 18:16:54 [kernel] ata10: EH complete
>>> Mar 20 18:16:54 [kernel] sd 9:0:0:0: [sdj] 976773168 512-byte hardware sectors (500108 MB)
>>> Mar 20 18:16:54 [kernel] sd 9:0:0:0: [sdj] Write Protect is off
>>> Mar 20 18:16:54 [kernel] sd 9:0:0:0: [sdj] Mode Sense: 00 3a 00 00
>>> Mar 20 18:16:54 [kernel] sd 9:0:0:0: [sdj] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
>> Which kernel version and how did you put the drive into sleep?
> 
> This timeout exception is still there - 2.6.24.4, drive goes to sleep on 
> timeout of 30 minutes (hdpram -S 241)

Yeah, this is still under discussion.  It seems smartd will need to 
issue CHECK POWER before issuing other commands.

-- 
tejun