Bug 55441 - sata_sil & sata_via make silent coruptions on FS
Summary: sata_sil & sata_via make silent coruptions on FS
Alias: None
Product: Drivers
Classification: Unclassified
Component: Other (show other bugs)
Hardware: All Linux
: P1 blocking
Assignee: drivers_other
Depends on:
Reported: 2013-03-19 10:10 UTC by Sue
Modified: 2013-11-13 21:35 UTC (History)
2 users (show)

See Also:
Kernel Version: 3.4.33-2.24-desktop (and never) x86_64
Tree: Mainline
Regression: No


Description Sue 2013-03-19 10:10:34 UTC
::sata_sil & sata_via make silent coruptions on FS


sata_sil & sata_via make corupptions on opensuse 12.2, 12.3 and Knoppix 7.0.5
in the X86_64 variant. Testing with knoppix in the i386 variant show no problems.

For both drivers and cards are the same problem:

Kernel use automaticly the right modul, activate them and mount the HD.
But by FS-operations (R/W, copy files..) all seems right else the HD are very slow. No errors, no suspects appears. Only when checked the files shasum we see that most files are corrupted. (EXT4)

The drive speed are ~6MB/s when wrong and ~35MB/s when use on i386 (knoppix)

Cards are PCI2STATA, eg:

> lspci:

RAID bus controller: Silicon Image, Inc. SiI 3112 [SATALink/SATARaid]
Serial ATA Controller (rev 02)
Flags: bus master, 66MHz, medium devsel, latency 32, IRQ 19
Memory at febffc00 (32-bit, non-prefetchable) [size=512]
Expansion ROM at feb00000 [disabled] [size=512K]
Capabilities: [60] Power Management version 2
Kernel driver in use: sata_sil

The card has an own interrupt:

> cat /proc/interrupts
 19:         16      35480   IO-APIC-fasteoi   sata_sil

Please help me to use the drivers, and sorry for bad english.

What can i do to help you to fix this bug?

Comment 1 Sue 2013-03-28 15:35:48 UTC
Also tested with btrfs: same problem.

Please help urgend, we cant upgrade our storage!
Comment 2 Sue 2013-04-04 13:25:13 UTC

Can everyone read this BUG-Report?
Comment 3 Alan 2013-11-13 21:35:28 UTC
This isn't a support forum and your bug report really doesn't contain any information that would enable someone to debug it, especially as its quite likely to be a hardware problem.

Note You need to log in before you can comment on or make changes to this bug.