Kernel driver for Sil 3112 (sata_sil.c) problems

Kernel driver for Sil 3112 (sata_sil.c) problems

[kacer]

Dear Jeff,
sorry for disturbing you, but I tried nearly everything to get this SATA
controller (Silicon Image Sil 3112) working, but it always crashes while
writing data (few hundred MBs or few GBs). In dmesg I can see something like
that:

end_request: I/O error, dev sdb, sector 459620760 
Buffer I/O error on device dm-3, logical block 155125763 
lost page write due to I/O error on dm-3 
ATA: abnormal status 0xD8 on port 0xE08180C7 
ATA: abnormal status 0xD8 on port 0xE08180C7 
ATA: abnormal status 0xD8 on port 0xE08180C7 

ata2: command 0x35 timeout, stat 0x58 host_stat 0x0 
ata2: status=0x58 { DriveReady SeekComplete DataRequest } 
SCSI error : <1 0 0 0> return code = 0x8000002 
sdb: Current: sense key=0xb 
    ASC=0x47 ASCQ=0x0 
end_request: I/O error, dev sdb, sector 459620744 
Buffer I/O error on device dm-3, logical block 155125761 
lost page write due to I/O error on dm-3 
ATA: abnormal status 0x58 on port 0xE08180C7 
ATA: abnormal status 0x58 on port 0xE08180C7 
ATA: abnormal status 0x58 on port 0xE08180C7

And same problems with /dev/sda. In this moment system is locked and nearly
not working, the only thing what I can do is HW reset...

I have Sil 3112 in Asus TUSL2-C (i815) motherboard, 2x256 MB RAM, there are
2 WD Raid Edition 400 GB drives on controller (2 of 4 ports used), BIOS of
controller is 4.2.0, I tried several kernels from 2.6.12 up to 2.6.15.4.
Nothing changed and I'm still getting same error messages. If you need some
other additional info or cooperation, just write me, I'm feeling desperate,
I don't know that to do to get that working. I tried to search the internet
for some suggestions, but I was not able to fix the the problems.

Sincerely
Jiri Bartosik

Re: Kernel driver for Sil 3112 (sata_sil.c) problems

[Tejun Heo]

kacer wrote:
> Dear Jeff,
> sorry for disturbing you, but I tried nearly everything to get this SATA
> controller (Silicon Image Sil 3112) working, but it always crashes while
> writing data (few hundred MBs or few GBs). In dmesg I can see something like
> that:
> 
> end_request: I/O error, dev sdb, sector 459620760 
> Buffer I/O error on device dm-3, logical block 155125763 
> lost page write due to I/O error on dm-3 
> ATA: abnormal status 0xD8 on port 0xE08180C7 
> ATA: abnormal status 0xD8 on port 0xE08180C7 
> ATA: abnormal status 0xD8 on port 0xE08180C7 
> 
> ata2: command 0x35 timeout, stat 0x58 host_stat 0x0 
> ata2: status=0x58 { DriveReady SeekComplete DataRequest } 
> SCSI error : <1 0 0 0> return code = 0x8000002 
> sdb: Current: sense key=0xb 
>     ASC=0x47 ASCQ=0x0 
> end_request: I/O error, dev sdb, sector 459620744 
> Buffer I/O error on device dm-3, logical block 155125761 
> lost page write due to I/O error on dm-3 
> ATA: abnormal status 0x58 on port 0xE08180C7 
> ATA: abnormal status 0x58 on port 0xE08180C7 
> ATA: abnormal status 0x58 on port 0xE08180C7
> 
> And same problems with /dev/sda. In this moment system is locked and nearly
> not working, the only thing what I can do is HW reset...
> 
> I have Sil 3112 in Asus TUSL2-C (i815) motherboard, 2x256 MB RAM, there are
> 2 WD Raid Edition 400 GB drives on controller (2 of 4 ports used), BIOS of
> controller is 4.2.0, I tried several kernels from 2.6.12 up to 2.6.15.4.
> Nothing changed and I'm still getting same error messages. If you need some
> other additional info or cooperation, just write me, I'm feeling desperate,
> I don't know that to do to get that working. I tried to search the internet
> for some suggestions, but I was not able to fix the the problems.
> 

Hello, Jiri.

'lspci -n' please.

-- 
tejun

RE: Kernel driver for Sil 3112 (sata_sil.c) problems

[kacer]

Hi Tejun,

here is lspci:
00:00.0 Host bridge: Intel Corporation 82815 815 Chipset Host Bridge and
Memory Controller Hub (rev 04)
00:01.0 PCI bridge: Intel Corporation 82815 815 Chipset AGP Bridge (rev 04)
00:1e.0 PCI bridge: Intel Corporation 82801 PCI Bridge (rev 05)
00:1f.0 ISA bridge: Intel Corporation 82801BA ISA Bridge (LPC) (rev 05)
00:1f.1 IDE interface: Intel Corporation 82801BA IDE U100 (rev 05)
00:1f.2 USB Controller: Intel Corporation 82801BA/BAM USB (Hub #1) (rev 05)
00:1f.3 SMBus: Intel Corporation 82801BA/BAM SMBus (rev 05)
00:1f.4 USB Controller: Intel Corporation 82801BA/BAM USB (Hub #2) (rev 05)
01:00.0 VGA compatible controller: ATI Technologies Inc 3D Rage IIC AGP (rev
3a)
02:0a.0 Mass storage controller: Triones Technologies, Inc.
HPT366/368/370/370A/372/372N (rev 04)
02:0c.0 Mass storage controller: Triones Technologies, Inc.
HPT366/368/370/370A/372/372N (rev 04)
02:0d.0 PCI bridge: Tiger Jet Network Inc. Unknown device 0600 (rev 01)
02:0e.0 Ethernet controller: Intel Corporation 82541GI/PI Gigabit Ethernet
Controller
03:04.0 Mass storage controller: Silicon Image, Inc. SiI 3112
[SATALink/SATARaid] Serial ATA Controller (rev 02)
03:05.0 Mass storage controller: Silicon Image, Inc. SiI 3112
[SATALink/SATARaid] Serial ATA Controller (rev 02)

lspci -n:
00:00.0 0600: 8086:1130 (rev 04)
00:01.0 0604: 8086:1131 (rev 04)
00:1e.0 0604: 8086:244e (rev 05)
00:1f.0 0601: 8086:2440 (rev 05)
00:1f.1 0101: 8086:244b (rev 05)
00:1f.2 0c03: 8086:2442 (rev 05)
00:1f.3 0c05: 8086:2443 (rev 05)
00:1f.4 0c03: 8086:2444 (rev 05)
01:00.0 0300: 1002:475a (rev 3a)
02:0a.0 0180: 1103:0004 (rev 04)
02:0c.0 0180: 1103:0004 (rev 04)
02:0d.0 0604: e159:0600 (rev 01)
02:0e.0 0200: 8086:1076
03:04.0 0180: 1095:3112 (rev 02)
03:05.0 0180: 1095:3112 (rev 02)

According tests I have found that problems are while writing data, not while
reading.

Jiri Bartosik

-----Original Message-----
From: Tejun Heo [mailto:htejun@gmail.com] 
Sent: Saturday, February 25, 2006 1:26 PM
To: kacer
Cc: jgarzik@pobox.com; linux-ide@vger.kernel.org
Subject: Re: Kernel driver for Sil 3112 (sata_sil.c) problems

Hello, Jiri.

'lspci -n' please.

-- 
tejun

Re: Kernel driver for Sil 3112 (sata_sil.c) problems

[Jeff Garzik]

kacer wrote:
> Dear Jeff,
> sorry for disturbing you, but I tried nearly everything to get this SATA
> controller (Silicon Image Sil 3112) working, but it always crashes while
> writing data (few hundred MBs or few GBs). In dmesg I can see something like
> that:

hrm, sounds like heat or power...

	Jeff

Re: Kernel driver for Sil 3112 (sata_sil.c) problems

[Tejun Heo]

Jeff Garzik wrote:
> kacer wrote:
> 
>> Dear Jeff,
>> sorry for disturbing you, but I tried nearly everything to get this SATA
>> controller (Silicon Image Sil 3112) working, but it always crashes while
>> writing data (few hundred MBs or few GBs). In dmesg I can see 
>> something like
>> that:
> 
> 
> hrm, sounds like heat or power...
> 

He said he was using 2 out of 4 ports. Gotta be 3114. It *might* be 
R_ERR on DMA activate FIS.

-- 
tejun

RE: Kernel driver for Sil 3112 (sata_sil.c) problems

[kacer]

Hi,
the controller is Kouwell 571S-4 , it is based on 2 chips Sil 3112, each
chip provides 2 ports. I don't think it's matter of power or heat, server is
huge, but has good cooling and 550 W Enermax PSU. My friend has same
problems and he tested it, problem persists no matter how many HDDs (1-4) or
PCI cards in computer.

For Tejun: I will provide that list from lspci as soon as machine will be
up, unfortunately machine's OS is stucked in error and I can't reset it now.

For more info: there is Gentoo linux on box, most packages on latest builds,
kernel is from vanilla sources with custom configuration, but I tried also
several kernels from gentoo sources.

Jiri Bartosik


-----Original Message-----
From: Tejun Heo [mailto:htejun@gmail.com] 
Sent: Saturday, February 25, 2006 2:08 PM
To: Jeff Garzik
Cc: kacer; linux-ide@vger.kernel.org
Subject: Re: Kernel driver for Sil 3112 (sata_sil.c) problems

Jeff Garzik wrote:
> kacer wrote:
> 
>> Dear Jeff,
>> sorry for disturbing you, but I tried nearly everything to get this SATA
>> controller (Silicon Image Sil 3112) working, but it always crashes while
>> writing data (few hundred MBs or few GBs). In dmesg I can see 
>> something like
>> that:
> 
> 
> hrm, sounds like heat or power...
> 

He said he was using 2 out of 4 ports. Gotta be 3114. It *might* be 
R_ERR on DMA activate FIS.

-- 
tejun