Testing compact flash cards as root drive

Author: Peter Apian-Bennewitz, pab-opto.de,

Find below what I found while testing multiple compact flash Toshiba cards:

dmesg

hda: TOSHIBA THNCF256MMA, ATA DISK drive
hdc: TOSHIBA CD-ROM XM-6702B, ATAPI CD/DVD-ROM drive
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
ide1 at 0x170-0x177,0x376 on irq 15
hda: 500736 sectors (256 MB) w/2KiB Cache, CHS=978/16/32, DMA
hdc: ATAPI 48X CD-ROM drive, 128kB Cache, UDMA(33)
Uniform CD-ROM driver Revision: 3.12
Partition check:
hda:hda: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hda: dma_intr: error=0x04 { DriveStatusError }
hda: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hda: dma_intr: error=0x04 { DriveStatusError }
hda: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hda: dma_intr: error=0x04 { DriveStatusError }
hda: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hda: dma_intr: error=0x04 { DriveStatusError }
hda: DMA disabled
This systems boots from SCSI Adaptec card.

Problem

With DMA disabled, I/O errors show up after a while (1/2h). running fsck on the device results in:
Nov 12 12:06:09 t kernel: hda: hda1
Nov 12 12:06:16 t kernel: hda: hda1
Nov 12 12:06:16 t kernel: hda: read_intr: status=0x59 { DriveReady SeekComplete DataRequest Error }
Nov 12 12:06:16 t kernel: hda: read_intr: error=0x40 { UncorrectableError }, LBAsect=131115, sector=131082
Nov 12 12:06:16 t kernel: end_request: I/O error, dev 03:01 (hda), sector 131082
Nov 12 12:06:16 t kernel: hda: read_intr: status=0x59 { DriveReady SeekComplete DataRequest Error }
Nov 12 12:06:16 t kernel: hda: read_intr: error=0x40 { UncorrectableError }, LBAsect=131115, sector=131082
Nov 12 12:06:16 t kernel: end_request: I/O error, dev 03:01 (hda), sector 131082

According to Fabio Parodi, the problem is fixed by increasing WAIT_DRQ in ./include/linux/ide.h, which is already set beyond ATA limits, but not enough for some CF cards. This doesn't seem to fix the above problem.
Increasing WAIT_READY by a factor of 10 doesn't seem to fix it either.

Meanwhile (20.11.02), I'm waiting for new Toshiba CF cards with Toshiba's MBA controller instead of their MMA controller.

Toshiba MBA cards

work nicely- no problems any more.

I'm happy to share experience, your mileage may vary.


Please bark at info@pab-opto.de in case you feel upset or interested.
updated: