Messages in this thread | | | Date | Thu, 30 Jul 1998 19:44:07 +0200 (CEST) | From | Rik van Riel <> | Subject | Re: Corruption Stats: Suggested Blacklist from the data |
| |
On Thu, 30 Jul 1998, Joel D. Elkins wrote: > Linus wrote: > >>And now that I started thinking about it, I also remembered that tar > >>sometimes fails to unpack otherwise good gzipped tarballs (if I repeat the > >>command, everything is (usually) swell). > > > >But the above does sound very strange indeed. It's not necessarily the disk, > >it could be memory or something else (the fact that it succeeds the second > >time in fact implies that the image on the disk is fine, and that the > >potential corruption happened somewhere else). > > If I remember the original post, he has a SCSI Quantum Fireball on which he was > seeing these strange corruption problems. I can report very similar problems
I've seen errors like these too. Using scsi-setup to carefully twiddle the mode pages of my SCSI devices cured this.
For IDE, using hdparm to set a more safe/conservative mode works too. I don't disable DMA, but I use no multcount or do some other 'marginal' thing. Most things like this are, IMHO, solved by seemingly irrelevant things and not the 'main suspects'...
Ingo, Mark, Linus: would this indicate a 'marginal/irrelevant' kernel bug too? (ie. are we looking in the wrong place with DMA or SMP and is timing or bookkeeping the problem; it would be very frustrating to completely rewrite the DMA code when some IDE timings table contains a typo)
Rik. +-------------------------------------------------------------------+ | Linux memory management tour guide. H.H.vanRiel@phys.uu.nl | | Scouting Vries cubscout leader. http://www.phys.uu.nl/~riel/ | +-------------------------------------------------------------------+
- 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.altern.org/andrebalsa/doc/lkml-faq.html
| |