Messages in this thread |  | | | Subject | Re: "attempt to access beyond end of device" yet again in 2.2.13pre14 | | Date | Tue, 5 Oct 1999 09:32:34 +0200 (MEST) | | From | (Rogier Wolff) |
| |
Andries mailing from his eternal Guest account wrote: > From: Ilpo Ruotsalainen <lonewolf@iki.fi>
> attempt to access beyond end of device > 08:03: rw=0, want=1852793648, limit=6554520 > 08:03: rw=0, want=795306343, limit=6554520 > 08:03: rw=0, want=796093551, limit=6554520 > 08:03: rw=0, want=946828138, limit=6554520
> Hmm - some days ago I muttered that it was a computed value > that is printed here, so unlikely to be anything recognizable, > but here it is > /con > fig/ > nls/ > iso8
It's the block number, not the sector number. The block numbers are of course stored somewhere in the filesystem. For example in an inode block (the quick access first few blocks) or an indirect block.
Still: Possible causes are abundant.
DMA into the wrong place (either overwriting the info we had cached or not overwriting what was in the intended buffer before)
Bit errors in memory. Pointer to data falls over. This should normally be accompanied by Oopsen, and sig11's during kernel compilation.
Bit errors while transferring data from disk. This time a bit pointing to an indirect block must've been changed, leading to the wrong block getting accessed as the indirect block.
Driver bug, scribbling random memory. Again I'd expect more Oopsen. Hmm. If the bug is to use memory that has just been freed, the bug will always have a valid pointer , previnting the oops... Hmm.
Roger.
-- ** R.E.Wolff@BitWizard.nl ** http://www.BitWizard.nl/ ** +31-15-2137555 ** *-- BitWizard writes Linux device drivers for any device you may have! --* ------ Microsoft SELLS you Windows, Linux GIVES you the whole house ------
- 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/
|  |