I tried Linus's official pre4, and I get basically the same results. I
suspect that Linus's version is better because SOMETHING should be
swapped; I have lots of unused daemons like 'xfstt'.
On the other hand, Linus's version also swaps out at least 2 MB on
doing wc /usr/bin/* to make room for worthless page cache pages. This
may be a good thing... Perhaps it only goes up the "right" amount of
swap and then stops. However, some pieces of netscape needed to be
swapped back in afterwards, which I think should not happen. Perhaps
that is the best that we can do with only 1 bit of age information?
Anyway, here are my test results... I'm not quite sure how to interpret
the differences between the 'shared' and 'cached' values for Linus and
ArcA.
-----------Results: (64Mb, PPro, EIDE with 4.4Mb/s media r/w
speed)------
---**2.2.0-pre4-test1-latestarca:**
real 9m24.528s
user 8m8.710s
sys 0m29.720s
total used free shared buffers
cached
Mem: 63548 52600 10948 31772 6176
24868
-/+ buffers/cache: 21556 41992
Swap: 34236 0 34236
---**2.2.0-pre4:**
real 9m26.428s
user 8m5.010s
sys 0m29.540s
total used free shared buffers
cached
Mem: 63484 60804 2680 25868 6524
37992
-/+ buffers/cache: 16288 47196
Swap: 34236 8104 26132
--------------------------------------------------------------------
The 'free' output was made and stored immediately after 'time make
bzImage modules' finished.
telomere:/proc/sys/vm> cat pagecache
1 30 75
telomere:/proc/sys/vm> cat buffermem
1 20 60
(these are my own settings)
-- Benjamin Redelings I <>< http://sdcc13.ucsd.edu/~bredelin- 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/