Bug 43196
Summary: | COMRESET failed (errno=-16) on Intel SSD | ||
---|---|---|---|
Product: | IO/Storage | Reporter: | keino |
Component: | Serial ATA | Assignee: | Jeff Garzik (jgarzik) |
Status: | RESOLVED OBSOLETE | ||
Severity: | normal | CC: | alan, cj.wijtmans, j.keil, jonware, linux, stefan |
Priority: | P1 | ||
Hardware: | All | ||
OS: | Linux | ||
URL: | https://bugs.launchpad.net/ubuntu/+source/linux/+bug/991040 | ||
Kernel Version: | 3.4.0-030400rc4-generic | Subsystem: | |
Regression: | No | Bisected commit-id: |
Description
keino
2012-05-02 18:10:37 UTC
Exactly the same problem occurs for me with a Crucial M4 SSD. As well as in Ubuntu 12.04, I have seen issue in Mageia 2 & Arch (both using kernel 3.4.2) The issue can not be seen in kernel 2.6.37.6 as shipped with the Slackware 13.37. Thanks Have you tried setting libahci.skip_host_reset=1 as kernel boot parameter? This fixed the comreset issue for me. However, I don't know if it has other implications. There's also an email thread on the linux kernel list on this issue: http://comments.gmane.org/gmane.linux.kernel/1247563 (In reply to comment #2) > There's also an email thread on the linux kernel list on this issue: > > http://comments.gmane.org/gmane.linux.kernel/1247563 Kernel 3.5 has a fix for that issue in rev 66583c9, please see if that fixes your problem. That issue was only introduced in 3.3, so it doesn't account for the original report, as Ubuntu 12.04 ships with 3.2. Additionally, I'm seeing this behaviour on 3.5.0 when resuming from suspend. Hardware: SATA controller [0106]: Intel Corporation Panther Point 6 port SATA Controller [AHCI mode] [8086:1e03] (rev 04) (prog-if 01 [AHCI 1.0]) ATA device, with non-removable media Model Number: C400-MTFDDAK256MAM Firmware Revision: 05TD Transport: Serial, ATA8-AST, SATA 1.0a, SATA II Extensions, SATA Rev 2.5, SATA Rev 2.6, SATA Rev 3.0 So, Ivy Bridge AHCI controller, and Micron C400 (AKA Crucial M4) SSD. I've tried reverting 66583c9, d408e2b and 7faa33d, but the problem still exists here. So (for me at least), that rules out this being related to those the changesets discussed in the above thread. I have a little difficulty testing on < 3.3-RC1 because of Ivy Bridge hangs (and I need this laptop functional for work, but it's the only system I can reproduce on). Happy to take some advice on what bisects to try. I'm going to try an SSD firmware update in case that's the culprit in my case, but it's interesting to me that all the reporters so far are citing SSDs being the issue. Unfortunately, I've not been able to obtain a firmware update for my hardware at this stage, happy to perform any testing suggested. I suffered this problem too on a Arch Linux Kernel using a Intel SSD X25-M 80GB: $ uname -a Linux luegisland 3.9.5-1-ARCH #1 SMP PREEMPT Sat Jun 8 09:22:45 CEST 2013 x86_64 GNU/Linux I could resolve this issue by updating the firmware (used to be 2CV102HD, now the version is 2CV102M3). I get this every time i resume from sleep(to RAM). On linux 4.0.2. However obviously ata1 is my ssd and ata3 is my WD 640GB green drive. [47517.277041] ata1: SATA link up 6.0 Gbps (SStatus 133 SControl 300) [47517.279050] ata1.00: configured for UDMA/133 [47517.284047] ata7: SATA link down (SStatus 0 SControl 300) [47517.287044] ata8: SATA link down (SStatus 0 SControl 300) [47522.325048] ata3: link is slow to respond, please be patient (ready=0) [47527.014272] ata3: COMRESET failed (errno=-16) [47528.491405] ata3: SATA link up 3.0 Gbps (SStatus 123 SControl 300) [47528.522316] ata3.00: configured for UDMA/133 |