linux 2.4.0test1ac8+riels patch performs okay

linux 2.4.0test1ac8+riels patch performs okay

[Frank Dekervel]

[-- Attachment #1: Type: text/plain, Size: 909 bytes --]

hello,

just a little message about my positive experiences with ac8 + riels
patch (i didnot test without):

-> for the io performance it is the fastest development kernel i yet
tested. (got 6meg/sec where raw ac7 gave me about
     the half). this closely matches 2.2

-> it is much more responsive than any other development kernels i have
tested. offcourse this may be a bit subjective.
     even when under high load (eg starting netscape or soffice under
kde) my system stays responsive enough to
     continue working. This was not the case with ac7.

-> my collection of legal mp3's finally plays fluently !! woo hoo !


no need to say it survives the memtest suite.

keep on and thanks for the good work ...

kervel

PS

other kernels i tested :
ac7
ac7 + rogerL's patch
ac7 + riel #1 + rogerL's (crashed)
ac7 + riel #1
(ac8 + riels)

here is a vmstat of a mmap001 .. don't know it its of much use




[-- Attachment #2: mmap001.vmstat --]
[-- Type: text/plain, Size: 561 bytes --]

   procs                      memory    swap          io     system         cpu
 r  b  w   swpd   free   buff  cache  si  so    bi    bo   in    cs  us  sy  id
 0  0  0  21304   2988    308  18920  29  31   360    51  523   182   8   7  85
 0  0  0  21304   2960    308  18932   2   0     2     0  127   122   2   0  98
 1  6  2  21240    380    224  23440 619 422  1004  2240 5089   712   6  12  82
 1  4  0  21316    628    320  22036 530   0   881  4502 11796  1092   1   7  91
 0  0  0  21300  14140    320   8492  69   0   151     0  381   101   1   6  94

Re: linux 2.4.0test1ac8+riels patch performs okay

[John Fremlin]

Frank Dekervel <frank.dekervel@student.kuleuven.ac.be> writes:
> -> it is much more responsive than any other development kernels i have
> tested. offcourse this may be a bit subjective.

My patch (vs ac7+Rik3) improves responsiveness with several running
processes considerably (at least on my system). Could you test it?
(I posted it to linux-mm and linux-kernel on Sunday).

-- 

	http://altern.org/vii
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux.eu.org/Linux-MM/