https://bugzilla.novell.com/show_bug.cgi?id=647829 https://bugzilla.novell.com/show_bug.cgi?id=647829#c0 Summary: SCSI errors with 2.6.32 kernel Classification: openSUSE Product: openSUSE 11.4 Version: Factory Platform: x86-64 OS/Version: SLES 11 Status: NEW Severity: Critical Priority: P5 - None Component: Kernel AssignedTo: kernel-maintainers@forge.provo.novell.com ReportedBy: dmitri.zoguine@sun.com QAContact: qa@suse.de Found By: --- Blocker: --- User-Agent: Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; en-US; rv:1.9.2.8) Gecko/20100722 Firefox/3.6.8 The following errors are observed if running 2.6.32.19 kernel: Oct 15 17:55:29 sata22 kernel: [12762.839374] ata3.00: exception Emask 0x60 SAct 0x3 SErr 0x800 action 0x6 frozen Oct 15 17:55:29 sata22 kernel: [12762.846770] ata3.00: irq_stat 0x20000000, host bus error Oct 15 17:55:29 sata22 kernel: [12762.852158] ata3: SError: { HostInt } Oct 15 17:55:29 sata22 kernel: [12762.852161] ata3.00: failed command: WRITE FPDMA QUEUED Oct 15 17:55:29 sata22 kernel: [12762.852166] ata3.00: cmd 61/00:00:3b:a3:f8/08:00:01:00:00/40 tag 0 ncq 1048576 out Oct 15 17:55:29 sata22 kernel: [12762.852167] res 40/00:08:83:4d:d2/00:00:01:00:00/40 Emask 0x60 (host bus error) Oct 15 17:55:29 sata22 kernel: [12762.852169] ata3.00: status: { DRDY } Oct 15 17:55:29 sata22 kernel: [12762.852171] ata3.00: failed command: WRITE FPDMA QUEUED Oct 15 17:55:29 sata22 kernel: [12762.852175] ata3.00: cmd 61/40:08:83:4d:d2/00:00:01:00:00/40 tag 1 ncq 32768 out Oct 15 17:55:29 sata22 kernel: [12762.852176] res 40/00:08:83:4d:d2/00:00:01:00:00/40 Emask 0x60 (host bus error) Oct 15 17:55:29 sata22 kernel: [12762.852177] ata3.00: status: { DRDY } Oct 15 17:55:29 sata22 kernel: [12762.852181] ata3: hard resetting link .. Oct 15 17:56:05 sata22 kernel: [12798.654944] sd 2:0:0:0: [sdc] Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE Oct 15 17:56:05 sata22 kernel: [12798.662197] sd 2:0:0:0: [sdc] Sense Key : Aborted Command [current] [descriptor] Oct 15 17:56:05 sata22 kernel: [12798.669745] Descriptor sense data with sense descriptors (in hex): Oct 15 17:56:05 sata22 kernel: [12798.676004] 72 0b 00 00 00 00 00 0c 00 0a 80 00 00 00 00 00 Oct 15 17:56:05 sata22 kernel: [12798.676008] 01 e9 83 3b Oct 15 17:56:05 sata22 kernel: [12798.676010] sd 2:0:0:0: [sdc] Add. Sense: No additional sense information Oct 15 17:56:05 sata22 kernel: [12798.676012] sd 2:0:0:0: [sdc] CDB: Write(10): 2a 00 01 e9 6b 3b 00 08 00 00 Oct 15 17:56:05 sata22 kernel: [12798.676016] end_request: I/O error, dev sdc, sector 32074555 The same errors are not observed if running 2.6.27 kernel. Controller is Intel 631xESB with the latest firmware rev 09 00:1f.2 SATA controller: Intel Corporation 631xESB/632xESB SATA AHCI Controller (rev 09) (prog-if01 [AHCI 1.0]) I noticed that similar errors were reported in kernel bugzilla https://bugzilla.kernel.org/show_bug.cgi?id=14922 and they resulted in the kernel change for NVIDIA based cards: diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c index b343903..a6a736a 100644 (file) --- a/drivers/ata/ahci.c +++ b/drivers/ata/ahci.c @@ -3082,8 +3082,16 @@ static int ahci_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) ahci_save_initial_config(pdev, hpriv); /* prepare host */ - if (hpriv->cap & HOST_CAP_NCQ) - pi.flags |= ATA_FLAG_NCQ | ATA_FLAG_FPDMA_AA; + if (hpriv->cap & HOST_CAP_NCQ) { + pi.flags |= ATA_FLAG_NCQ; + /* Auto-activate optimization is supposed to be supported on + all AHCI controllers indicating NCQ support, but it seems + to be broken at least on some NVIDIA MCP79 chipsets. + Until we get info on which NVIDIA chipsets don't have this + issue, if any, disable AA on all NVIDIA AHCIs. */ + if (pdev->vendor != PCI_VENDOR_ID_NVIDIA) + pi.flags |= ATA_FLAG_FPDMA_AA; + } if (hpriv->cap & HOST_CAP_PMP) pi.flags |= ATA_FLAG_PMP; It might be that there are another issues with this feature introduced in 2.6.32 and I am now testing the quick workaround of dropping the flag ATA_FLAG_FPDMA_AA for all cards Reproducible: Sometimes Steps to Reproduce: Reproduced under heavy IO load -- 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.