I'm having some troubles with SATA and Linux, I have seagate SATA disk.
DMA seems to be disabled on the disk.
The following is from the default fedora core 3 installation (kernel -2.6.9-1.667), the test gave me different results every time, it kept increasing all the time by approx. 14MB/sec.

#/sbin/lsmod
sata_sil 7813 3
libata 39813 1 sata_sil
sd_mod 13521 4
scsi_mod 76993 2 libata,sd_mod

#hdparm -tT /dev/sda
/dev/sda:
Timing cached reads: 2116 MB in 2.00 seconds = 1058.16 MB/sec
BLKFLSBUF failed: Operation not supported
HDIO_DRIVE_CMD(null) (wait for flush complete) failed: Operation not supported
Timing buffered disk reads: 46 MB in 3.11 seconds = 14.79 MB/sec
BLKFLSBUF failed: Operation not supported
HDIO_DRIVE_CMD(null) (wait for flush complete) failed: Operation not supported

#hdparm -tT /dev/sda
/dev/sda:
Timing cached reads: 2028 MB in 2.00 seconds = 1014.15 MB/sec
BLKFLSBUF failed: Operation not supported
HDIO_DRIVE_CMD(null) (wait for flush complete) failed: Operation not supported
Timing buffered disk reads: 88 MB in 3.09 seconds = 28.46 MB/sec
BLKFLSBUF failed: Operation not supported
HDIO_DRIVE_CMD(null) (wait for flush complete) failed: Operation not supported

and after some more times

# hdparm -tT /dev/sda
/dev/sda:
Timing cached reads: 1988 MB in 2.00 seconds = 993.65 MB/sec
BLKFLSBUF failed: Operation not supported
HDIO_DRIVE_CMD(null) (wait for flush complete) failed: Operation not supported
Timing buffered disk reads: 400 MB in 3.06 seconds = 130.65 MB/sec
BLKFLSBUF failed: Operation not supported
HDIO_DRIVE_CMD(null) (wait for flush complete) failed: Operation not supported

[root@localhost source]# hdparm -i /dev/sda
/dev/sda:
HDIO_GET_IDENTITY failed: Inappropriate ioctl for device
---------------------------------------------------------------------------
After recompiling the new 2.6.10 kernel:
#/sbin/lsmod
sata_sil 7813 3
libata 39813 1 sata_sil
sd_mod 13521 4
scsi_mod 76993 2 libata,sd_mod

[root@localhost source]# hdparm -tT /dev/sda
/dev/sda:
Timing cached reads: 2120 MB in 2.00 seconds = 1059.63 MB/sec
HDIO_DRIVE_CMD(null) (wait for flush complete) failed: Inappropriate ioctl for device
Timing buffered disk reads: 44 MB in 3.09 seconds = 14.23 MB/sec
HDIO_DRIVE_CMD(null) (wait for flush complete) failed: Inappropriate ioctl for device

Here th result is stuck to 14 MB/sec
[root@localhost source]# hdparm -i /dev/sda
/dev/sda:
HDIO_GET_IDENTITY failed: Inappropriate ioctl for device

This is a part of # dmesg

SCSI subsystem initialized
libata version 1.10 loaded.
sata_sil version 0.8
ACPI: PCI Interrupt Link [LNK3] enabled at IRQ 11
PCI: setting IRQ 11 as level-triggered
ACPI: PCI interrupt 0000:01:0b.0[A] -> GSI 11 (level, low) -> IRQ 11
ata1: SATA max UDMA/100 cmd 0xE0802080 ctl 0xE080208A bmdma 0xE0802000 irq 11
ata2: SATA max UDMA/100 cmd 0xE08020C0 ctl 0xE08020CA bmdma 0xE0802008 irq 11
ata1: dev 0 cfg 49:2f00 82:346b 83:7d01 84:4003 85:3469 86:3c01 87:4003 88:207f
ata1: dev 0 ATA, max UDMA/133, 312581808 sectors: lba48
ata1(0): applying Seagate errata fix
ata1: dev 0 configured for UDMA/100
scsi0 : sata_sil
ata2: no device found (phy stat 00000000)
scsi1 : sata_sil
Vendor: ATA Model: ST3160023AS Rev: 3.18
Type: Direct-Access ANSI SCSI revision: 05
SCSI device sda: 312581808 512-byte hdwr sectors (160042 MB)
SCSI device sda: drive cache: write back
SCSI device sda: 312581808 512-byte hdwr sectors (160042 MB)
SCSI device sda: drive cache: write back
sda: sda1 sda2 < sda5 sda6 sda7 sda8 sda9 >
Attached scsi disk sda at scsi0, channel 0, id 0, lun 0
------------------------------------------------------------------------
Can you please advise, is it Seagate mistake? (I have heared that there are some troubles with Seagate SATA disks) or is it SATA SiI 3112 error?
Can anybody with similar setup post results of hdparm -tT /dev/sda?

Thanks in advance

I have a same problem, the temporary solution was use 2.4 kernel, where I get Timing buffered disk reads around 60 MB/s.

I probably found better solution. The main problem is that a seagate disks are blacklisted in libata and siimage driver and driver runs with seagate errata fix.
Now I run on 2.6.10 kernel with siimage driver, where i do walkaround seagate errata fix, and disk works very well. I think that seagate fix is only for Sil 3112 and asus a7n8x have Sil3112a.
I am going to try do walkaround in libata driver , if I will be success I post reply.

Has anyone actually found any solution with coexistance of sata_sil and kernel 2.6??
The problem is stil valid because neither Slackware 10.2 with kernel test26.s nor suse 9.3 recognize this module. The problem is NOT is SATA disk but Serial CONTROLLER of Silicon Image that powers the serial HD.
I read all the threads of LQ on the "sata_sil" subject and still no answer has been given to that one. Of course, it's possible to still use 2.4, but come on...

01:0b.0 RAID bus controller: Silicon Image, Inc. SiI 3112 [SATALink/SATARaid] Serial ATA Controller (rev 02)

This is my SATA chipset and it works fine with a 10.2. As i said in another thread, you probably have to enable SATA raid in your bios, even if you don't have one. My MoBo is an ABIT-NF7. My kernel a 2.6.13.

You probably didn't get the problem. SuSe as well as Slackware see the device and recognize it as Silicon Image SATA Controller, but while loading driver "sata_sil" box just hangs. This is when loading with kernel "sata.i" and "test26.s". When loading kernel 2.4 though, system doesn't see the SATA controller at all.
Slackware 10.0, however, recognizes this module and loads the driver with no problem whatsoever. As well as SuSE 9.1 .
This is the only reason I switched to Debian-Etch, because it has no problems with loading driver "sata_sil" in kernel 2.6

This is strange. I made the install with sata.i, then i installed a custom kernel with no problem. Did you tried to make your own bootdisks?

If you mean FDD then I have no such device. Obsolete

SUCCESS!! On SuSE 10 though, not 9.3.
I used "irqpoll" option on boot-install and it recognized "Silicon Image S-ATA Controller" and loaded "sata_sil" driver. Same trick with 9.3 doesn't work though, system still hangs while
"Searching for info file".......