Poor 2.3.8 times

Jim Nance (jlnance@sailboat.mis.uncwil.edu)
Tue, 22 Jun 1999 23:31:14 -0400


Hello All,
I thought the new code in 2.3.7/8 might help with my mozilla build times.
I decided to benchmark this and I found that it actually seems to hurt.

My benchmark consists of going to a directory where mozilla has been built,
removing all the libraries, and timing how long it takes to rebuild them.
On this machine this involves a fair amount of swapping in addition to the
I/O necessary to create the libs. For both 2.2.10 and 2.3.8 I ran the
following set of commands 3 times:

find . -type f -name \*.a -o -name \*.so -print | xargs rm -f; sync
time make MAKE='make -j 2' libs

Here are my results:

2.2.10
1 185.360u 63.950s 7:00.66 59.2% 0+0k 0+0io 140517799pf+1248w
2 186.330u 63.290s 6:53.47 60.3% 0+0k 0+0io 140516809pf+0w
3 186.760u 62.470s 6:45.39 61.4% 0+0k 0+0io 140524569pf+0w

2.3.8
1 179.630u 64.180s 7:24.60 54.8% 0+0k 0+0io 140433611pf+0w
2 178.800u 65.080s 7:11.18 56.5% 0+0k 0+0io 140557286pf+0w
3 178.840u 64.470s 7:07.19 56.9% 0+0k 0+0io 140535261pf+32w

The machine I ran these tests on has a 133 MHz Cyrix CPU, 64M of ram,
and IDE disks. I dont know if it makes any difference, but the mozilla
source is on my /dev/hdc6 partition but the object and library files are
on a /dev/hda3 which is (obviously) on a different disk.

Hope someone finds this interesting/useful.

Jim

-
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/