On Wed, Dec 31, 2003 at 01:01:37AM +0900, Akira Kitada wrote: > Partition check: > hda:<4>hda: dma_timer_expiry: dma status == 0x61 > hda: error waiting for DMA > hda: dma timeout retry: status=0x58 { DriveReady SeekComplete DataRequest } > > blk: queue c04102a0, I/O limit 4095Mb (mask 0xffffffff) > hda1 hda2 hda3 <<4>hda: dma_timer_expiry: dma status == 0x61 > hda: error waiting for DMA > hda: dma timeout retry: status=0x5a { DriveReady SeekComplete DataRequest Index } > > hda5<4>hda: dma_timer_expiry: dma status == 0x61 > hda: error waiting for DMA > hda: dma timeout retry: status=0x58 { DriveReady SeekComplete DataRequest } > > hda6<4>hda: dma_timer_expiry: dma status == 0x61 > hda: error waiting for DMA > hda: dma timeout retry: status=0x58 { DriveReady SeekComplete DataRequest } > hda7 hda8 hda9 hda10 hda11 > > hdb: hdb1 < hdb5 >
Using ``hdparm -d0'' to disable DMA will be a meantime workaround. But not using DMA is slower than using it (when it works). There is something like dma=off kernel parameter, can't remember exactly. Have you tried to fiddle with the drives physicaly (swapping cables, putting on a different channel/controller)? -- Jan Minar Don't CC me, I'm subscribed.
pgp00000.pgp
Description: PGP signature