ide-tape flaky in 2.2.15

From: Ed Tomlinson (tomlins@cam.org)
Date: Thu May 25 2000 - 06:32:41 EST


I have been having 'fun' with ide-tape. If I run with DMA enabled I get:

May 24 23:20:34 oscar kernel: hdd: lost interrupt
May 24 23:20:34 oscar kernel: ide-tape: The tape wants to issue more interrupts in DMA mode
May 24 23:20:34 oscar kernel: ide-tape: DMA disabled, reverting to PIO
May 24 23:20:34 oscar kernel: hdd: DMA disabled
May 24 23:20:35 oscar kernel: hdd: ATAPI reset complete
May 24 23:20:35 oscar kernel: ide-tape: ht0: I/O error, pc = a, key = 6, asc = 29, ascq = 0
May 24 23:20:35 oscar kernel: ide-tape: ht0: I/O error, pc = a, key = 2, asc = 4, ascq = 1
May 24 23:20:37 oscar last message repeated 239 times
May 24 23:20:37 oscar kernel: ide-tape: ht0: I/O error, pc = 10, key = 2, asc = 4, ascq = 1
May 24 23:20:37 oscar kernel: ide-tape: Couldn't write a filemark

Which looks like something is expecting an interrupt but times out, and then
when the interrupt arrives it gives up. So I try PIO :-( and get:

May 24 21:41:41 oscar kernel: ide-tape: ht0: DSC timeout
May 24 21:41:42 oscar kernel: hdd: ATAPI reset complete
May 24 21:41:42 oscar kernel: ide-tape: ht0: I/O error, pc = a, key = 2, asc = 4, ascq = 1
May 24 21:41:44 oscar last message repeated 240 times
May 24 21:41:44 oscar kernel: ide-tape: ht0: I/O error, pc = 10, key = 2, asc = 4, ascq = 1
May 24 21:41:44 oscar kernel: ide-tape: Couldn't write a filemark
May 24 22:40:21 oscar kernel: ide-tape: Block location is unknown to the tape
May 24 22:40:21 oscar kernel: ide-tape: Block location is unknown to the tape

Which looks to be the same problem. I have four tapes. I get this error with
all of them. This implies is not the media...

The tape is reconized as:

May 24 21:18:11 oscar kernel: VP_IDE: IDE controller on PCI bus 00 dev 39
May 24 21:18:11 oscar kernel: VP_IDE: not 100% native mode: will probe irqs later
May 24 21:18:11 oscar kernel: ide0: BM-DMA at 0xe000-0xe007, BIOS settings: hda:DMA, hdb:DMA
May 24 21:18:11 oscar kernel: ide0: VIA Bus-Master (U)DMA Timing Config Success
May 24 21:18:11 oscar kernel: ide1: BM-DMA at 0xe008-0xe00f, BIOS settings: hdc:DMA, hdd:DMA
May 24 21:18:11 oscar kernel: ide1: VIA Bus-Master (U)DMA Timing Config Success
May 24 21:18:11 oscar kernel: hda: QUANTUM FIREBALLP KA13.6, ATA DISK drive
May 24 21:18:11 oscar kernel: hdb: CD-ROM 50X, ATAPI CDROM drive
May 24 21:18:11 oscar kernel: hdc: QUANTUM FIREBALL SE4.3A, ATA DISK drive
May 24 21:18:11 oscar kernel: hdd: HP COLORADO 20GB, ATAPI TAPE drive
May 24 21:18:11 oscar kernel: ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
May 24 21:18:11 oscar kernel: ide1 at 0x170-0x177,0x376 on irq 15
May 24 21:18:11 oscar kernel: hda: QUANTUM FIREBALLP KA13.6, 13216MB w/371kB Cache, CHS=1684/255/63, UDMA
May 24 21:18:11 oscar kernel: hdc: QUANTUM FIREBALL SE4.3A, 4110MB w/80kB Cache, CHS=14848/9/63, UDMA
May 24 21:18:11 oscar kernel: hdb: ATAPI 32X CD-ROM drive, 128kB Cache
May 24 21:18:11 oscar kernel: Uniform CDROM driver Revision: 2.56
May 24 21:18:11 oscar kernel: ide-tape: hdd: overriding capabilities->speed (assuming 650KB/sec)
May 24 21:18:11 oscar kernel: ide-tape: hdd: overriding capabilities->max_speed (assuming 650KB/sec)
May 24 21:18:11 oscar kernel: ide-tape: hdd <-> ht0, 650KBps, 13*32kB buffer, 3200kB pipeline, 220ms tDSC

ide-tape is compiled into the kernel (2.2.15). I have changed the constant to increase the number of
buffers from 200 to 800. Also, when the drive works, it easily does 950KBps.

I get these errors about in about 2/3s of my backups. Restores also have problems.
I have tried the ide patches from Andre's site. These are no better - I cannot restore
at all using them. Would really like this to work dependably.

TIA,

Ed Tomlinson <tomlins@cam.org>

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Wed May 31 2000 - 21:00:14 EST