Hi Guys,
We use a Samsung ATA flash as /dev/hda. Filesystem is ext3 but the same
occurs with minix. It worked fine up to kernel 2.4.10. But from kernel
2.4.11 (2.4.12) onward a troublesome problem crept in.
We fixed it by modifying the timeout value of WAIT_DRQ.
IDE specs allow up to 20ms, standard Linux kernel allows up to
50ms, but Samsung ATA CF needs 100ms.
What should we do? We have four options:
A - keep this patch for us. Stop bothering.
B - #define WAIT_DRQ (10*HZ/100) in standard kernel tree
C - put a parameter to be adjusted at boot time
D - write some code that autoadjusts WAIT_DRQ depending on device type
We think that option B should not harm other users - after all,
this timeout only occurs in case of errors.
If you are interested in details, this is an excerpt from dmesg:
Linux version 2.4.16 (root@devel-linux) (gcc version 2.95.2 19991024
(release))
...
ide: Assuming 33MHz system bus speed for PIO modes; override with
idebus=xx
CS5530: IDE controller on PCI bus 00 dev 92
CS5530: chipset revision 0
CS5530: not 100% native mode: will probe irqs later
ide0: BM-DMA at 0xfc00-0xfc07, BIOS settings: hda:pio, hdb:pio
ide1: BM-DMA at 0xfc08-0xfc0f, BIOS settings: hdc:pio, hdd:pio
hda: SAMSUNG ATA CF, ATA DISK drive
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
hda: 123392 sectors (63 MB) w/1KiB Cache, CHS=964/4/32
Partition check:
hda: hda1
...
After one minute from the boot, the ide driver
prints these messages:
hda: status timeout: status=0xd0 { Busy }
hda: no DRQ after issuing WRITE
ide0: reset: master: error (0x00?)
hda: status error: status=0x59 { DriveReady SeekComplete DataRequest Error }
hda: status error: error=0x00 { }
hda: drive not ready for command
hda: status error: status=0x58 { DriveReady SeekComplete DataRequest }
hda: drive not ready for command
hda: status error: status=0x58 { DriveReady SeekComplete DataRequest }
hda: drive not ready for command
hda: status error: status=0x58 { DriveReady SeekComplete DataRequest }
hda: drive not ready for command
ide0: reset: master: error (0x00?)
end_request: I/O error, dev 03:01 (hda), sector 2
end_request: I/O error, dev 03:01 (hda), sector 16
end_request: I/O error, dev 03:01 (hda), sector 18
end_request: I/O error, dev 03:01 (hda), sector 20
end_request: I/O error, dev 03:01 (hda), sector 22
end_request: I/O error, dev 03:01 (hda), sector 24
end_request: I/O error, dev 03:01 (hda), sector 26
end_request: I/O error, dev 03:01 (hda), sector 28
end_request: I/O error, dev 03:01 (hda), sector 30
The system becomes unstable.
Fabio Parodi
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Fri Dec 07 2001 - 21:00:33 EST