[lkml]   [1999]   [Apr]   [20]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: hda: lost interrupt
Aaron Lehmann wrote:
> I get this message on my console about every 12 hours. I think its an SiS
> chipset. It is doing UDMA. I'm running 2.2.5ac6, i dont think this ever
> happened before i ugreaded to it.
> Is something wrong with my chipset or drive? Is there a bug in 2.2.5ac6
> ide?
> KeyID 1024D/73348CA0
> Fingerprint 8EFC 7F10 F26C 55A8 458A 38B0 890F 384F 7334 8CA0
> Public key available at
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to
> Please read the FAQ at

Let's see if my $.02 worth makes any sense. It has generally been
the consensus that this is a hdrive on its way south. Since this
problem seems to be cropping up for more people and because of another
filesystem oddity I've noticed: I, at least, am beginning to wonder.
Unfortunately I am NOT a kernel coder.
I (as did some others) got this message using stock (kernel-2.2.5,
binutils-, egcs-1.1.2, glibc-2.1.1pre1) during periods of
high disk io. Seems to always be hda which gets this error. In my
case, nothing Linux is on hda, but it is the master for hdb which
contains /root.
Now, (kernel-2.2.6, binutils-, egcs-1.1.2,
glibc-2.1.1pre1) the message has changed, but occurs under the same
conditions: hda: unexpected interrupt. This is completely repeatable
for me, AMD5x86-WB w/48M RAM 128M swap(hdc). From a clean boot, do
anything that eats free memory, ie. compile a program. After this
completes, I 'startx' and I get this message. It doesn't have to be
from a clean boot, just anytime free pages are low. I can only start X
from a completely clean reboot if I start it immediately on login. I'm
wondering if something is occurring outside a process/thread that is
supposed to be under a locked condition.
The other oddity I've noticed is that dirty? inodes don't seem to be
getting cleared. I only noticed this because my root partition is
getting a little thin: ~11M. It happened while reinstalling
glibc-2.1.1p1 over itself. Installation went fine but the script failed because it wouldn't pick up the
libc*.so links in /usr/lib. I check everything it suggested and reran
the installation which resulted in a "no space on device" error. I
checked, and sure enough, 0 free on /. It seems to me that this
shouldn't happen because it was copying identical files over top of the
others and it removes the *.new as it goes. Anyway, I rebooted,
noticing as shutdown was running, that /lib and / were still in use.
After reboot I had my 11M back. I ran the glibc install again and noted
the free space on / was 4M. Installation test still failed for whatever
reason.(dangit, if the links show up in and /usr/lib is in
egcs' search path why won't ld find them) :( I digress. I continued my
normal routines from the console (can't start X except on a clean boot,
remember) for the next 2 days and that 7M of disk was never freed. I
gave up, rebooted and voila', I had 11M free again. Still showed / and
/usr in use on shutdown.l sigh
Any ideas how to give you any usable information? Currently planning
on recompiling EVERYthing even tho' most of it is less than 2 weeks old.

Thanks for your ear,


For who among men knows the thoughts of a man except the man's spirit
within him? In the same way, no one knows the thoughts of God except
the Spirit of God. We have not received the spirit of the world but the
Spirit who is from God, that we may understand what God has freely given
-- I Corinthians 2:11,12

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to
Please read the FAQ at

 \ /
  Last update: 2005-03-22 13:51    [W:0.030 / U:4.540 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site