Why should kernel switch off DMA in hdd if problem on hdc?
From: Jan Panteltje (pNaonStpealmtje_at_yahoo.com)
Date: 05/31/05
- Next message: Netocrat: "Re: problem in iptables when i install 2.6.10 on RH9"
- Previous message: Sergei Organov: "Re: unexpected abstraction penalty in C++ (second version)"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Tue, 31 May 2005 17:31:20 GMT
I added a bigger harddisk (120 GB), so now both IDE
channels are full on this Tyan mobo (old one).
Everytime I try to mount a filesystem on hdc (the new disk),
I get:
May 31 19:16:59 panteltje kernel: for (ide1(22,7))
May 31 19:17:00 panteltje kernel: hdc: dma_intr: status=0x51 { DriveReady SeekComplete Error }
May 31 19:17:00 panteltje kernel: hdc: dma_intr: error=0x84 { DriveStatusError BadCRC }
May 31 19:17:01 panteltje kernel: hdc: dma_intr: status=0x51 { DriveReady SeekComplete Error }
May 31 19:17:01 panteltje kernel: hdc: dma_intr: error=0x84 { DriveStatusError BadCRC }
May 31 19:17:01 panteltje kernel: hdc: dma_intr: status=0x51 { DriveReady SeekComplete Error }
May 31 19:17:01 panteltje kernel: hdc: dma_intr: error=0x84 { DriveStatusError BadCRC }
May 31 19:17:02 panteltje kernel: hdc: dma_intr: status=0x51 { DriveReady SeekComplete Error }
May 31 19:17:02 panteltje kernel: hdc: dma_intr: error=0x84 { DriveStatusError BadCRC }
May 31 19:17:02 panteltje kernel: hdd: DMA disabled
----------------------------------???
May 31 19:17:02 panteltje kernel: ide1: reset: success
May 31 19:17:03 panteltje kernel: hdc: dma_intr: status=0x51 { DriveReady SeekComplete Error }
May 31 19:17:03 panteltje kernel: hdc: dma_intr: error=0x84 { DriveStatusError BadCRC }
Why switch of DMA for hdd (DVD drive), if a problem with hdc?
OK they are on the same cable.
Switching DMA back on with hdparm -d /dev/hdd is possible, and after the mount
there are zero errors again on all drives (with DMA on or OFF on hdd).
I had a 40GB in the same system and no problems.
Look:
panteltje:~# mount
/dev/hda7 on / type ext2 (rw)
proc on /proc type proc (rw)
devpts on /dev/pts type devpts (rw,mode=0620,gid=5)
/dev/hda5 on /boot type ext2 (rw)
shmfs on /dev/shm type shm (rw)
/dev/hdb1 on /video type ext2 (rw)
/dev/hdb2 on /reiser type reiserfs (rw)
/dev/hdc1 on /huge type ext2 (rw)
/dev/hdc3 on /r2 type reiserfs (rw)
/dev/hda1 on /big type vfat (rw,uid=500,gid=100)
/proc/bus/usb on /proc/bus/usb type usbdevfs (rw)
/dev/hdc7 on /mnt/lfs type reiserfs (rw)
/dev/hdc6 on /mnt/lfs/boot type reiserfs (rw)
panteltje:~# df
Filesystem 1k-blocks Used Available Use% Mounted on
/dev/hda7 9076364 8165052 450252 95% /
/dev/hda5 23302 5613 16486 26% /boot
shmfs 193388 0 193388 0% /dev/shm
/dev/hdb1 28842748 24969120 2408504 92% /video
/dev/hdb2 9775248 3732664 6042584 39% /reiser
/dev/hdc1 39371864 6738756 30633112 19% /huge
/dev/hdc3 37217132 32840 37184292 1% /r2
/dev/hda1 29288192 28603504 684688 98% /big
/dev/hdc7 39448760 32840 39415920 1% /mnt/lfs
/dev/hdc6 49960 32840 17120 66% /mnt/lfs/boot
So, I actually want to build a new Linux From Scratch on this hdc...
For a very strange reason, Flash 7 plugin only comes for a news libc6
version then I use.
I have even firefox compiled with old libc6 ....
But the flash does not come as source.
So, if people do not release as source you have to buy or make a new system
everytime .. that will never work, as different vendors will have different
libc versions.
That really sucks, it is sabotage.
Or they should link in libc statically too?
All beats me :-)
- Next message: Netocrat: "Re: problem in iptables when i install 2.6.10 on RH9"
- Previous message: Sergei Organov: "Re: unexpected abstraction penalty in C++ (second version)"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|
