Alessandro Suardi wrote:
>
> Seems we're not moving from the 8.6 MB/s marker :(
>
> 2.3.99pre3 idebus=33 36.99 8.60
> multcount=0
> 2.3.99pre3 idebus=33 36.89 8.56
> multcount=8
>
strace seems to actually point to read() being slower... also the
RH6.2 upgrade sees my 2.3.99pre3 even slower than it was on RH6.1 :(
note I run my tests under X / Netscape but disconnected from my ppp
connection and with _no_ activity at all in all cases.
[asuardi@dogbert rc.d]$ head -12 /tmp/hdparm-strace-2.2.14.out-m8
execve("/sbin/hdparm", ["hdparm", "-Tt", "/dev/hda"], [/* 20 vars */]) = 0
/dev/hda:
Timing buffer-cache reads: 128 MB in 3.46 seconds = 36.99 MB/sec
Timing buffered disk reads: 64 MB in 6.27 seconds = 10.21 MB/sec
% time seconds usecs/call calls errors syscall
------ ----------- ----------- --------- --------- ----------------
98.79 5.206393 26837 194 read
1.07 0.056186 14047 4 ioctl
0.07 0.003492 1164 3 sync
0.04 0.001892 7 257 lseek
0.02 0.000913 457 2 shmdt
[asuardi@dogbert rc.d]$ head -12 /tmp/hdparm-strace-2.3.99p3-m8.out
execve("/sbin/hdparm", ["hdparm", "-Tt", "/dev/hda"], [/* 20 vars */]) = 0
/dev/hda:
Timing buffer-cache reads: 128 MB in 3.50 seconds = 36.57 MB/sec
Timing buffered disk reads: 64 MB in 8.11 seconds = 7.89 MB/sec
% time seconds usecs/call calls errors syscall
------ ----------- ----------- --------- --------- ----------------
99.06 5.473274 28213 194 read
0.38 0.021126 4225 5 ioctl
0.32 0.017505 2918 6 write
0.16 0.008821 2940 3 sync
0.04 0.002022 8 257 lseek
from my lilo.conf:
boot=/dev/hda
map=/boot/map
install=/boot/boot.b
append="reboot=warm idebus=33" <--- is this okay ?
prompt
Ciao,
--alessandro <asuardi@uninetcom.it> <asuardi@it.oracle.com>
Linux 2.2.14/2.3.99-pre3 glibc-2.1.3.15 gcc-2.95.2 binutils-2.9.5.0.32
-
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 : Fri Apr 07 2000 - 21:00:08 EST