Page cache problem? (was Re: Problem with kernels 2.3.43/4/5)

From: Manfred Spraul (manfreds@colorfullife.com)
Date: Mon Feb 14 2000 - 18:22:28 EST

Ok I could reproduce it as well:
1) create a large file, larger that ram size
2) please don't use "dd if=/dev/zero", the file should be random,
otherwise md5sum won't detect twisted sectors.

3)
$ (md5sum file&);(md5sum file&);(md5sum file&);(md5sum file&);
< wait until it created the sum>
< repeat>

I replaced the file with "/dev/hda3", and the hash numbers were always
identical.
It seems that the problem is SMP specific, at least I couldn't reproduce
it on my AMD K6.

Any known page cache corruptions in these kernel? (filesystem: ext2) 

--
	Manfred

Rui Sousa wrote:
> 
> Test 1:
> Running four md5sum's on the same file (168 Mb) at the same time, they all give
> the
> same result. (same time means the time it takes to put the mouse over another
> window and press enter ~< 0.5s)
> 
> Wait a couple of seconds, and rerun the four md5sum's on the same file. They all
> give
> the same result but it's different from the one in the first run.
> 
> Each time I do another run they all give the same result but it usually is
> different
> from the previous one.
> 
> Test 2:
> Running one md5sum on the same file up to nine times, one after the other, I
> always
> got the same result.
> 
> Rui Sousa

-
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/

This archive was generated by hypermail 2b29 : Tue Feb 15 2000 - 21:00:28 EST