Re: Q67 and QM67 COMRESET solved by skip_host_reset=1
From: Andrew Lutomirski <hidden>
Date: 2011-06-14 15:13:49
On Thu, May 26, 2011 at 9:12 AM, Andrew Lutomirski [off-list ref] wrote:
On Wed, May 25, 2011 at 5:36 AM, Tejun Heo [off-list ref] wrote:quoted
Cc'ing Intel people and Jeff. Hi! Any ideas what's going on with the following? In short, on sandy bridge ahci + intel g2 ssd, doing HOST_RESET during initialization makes the first reset timeout inducing 10 sec delay.On my desktop box, I have three SATA devices: a Sandforce-based SSD, a normal hard drive, and a CD. Only the SSD has the problem, and libahci.skip_host_reset=1 fixes it on this box as well. Here's the relevant bit of the log:
[ 7.431907] ata1: link is slow to respond, please be patient (ready=0) [ 12.121178] ata1: COMRESET failed (errno=-16) [ 12.428010] ata1: SATA link up 6.0 Gbps (SStatus 133 SControl 300) [ 12.540828] ata1.00: ATA-9: OCZ-VERTEX3, 2.02, max UDMA/133
*sigh* I assumed the fact that I had this problem on two similar motherboard chipsets with two different generations of disks from two vendors meant it was a chipset or driver bug. I assumed wrong. It was separate firmware bugs in the Sandforce and the Intel firmware. Both are fixed in newer versions. Sorry for the noise, Andy