> Phil Harman wrote:
>
> > Bob Palowoda wrote:
> >
> >> There is probably nothing wrong with connection it
> just toke
> >> 123sec on a 3.2ghz machine. I just didn't wait
> long enough.
> >
> >
> >
> > Phew! 123sec is a long time. Maybe your machine was
> busy doing other stuff in the background?
> ^^^^^^^^^^^^^^^
> [EMAIL PROTECTED]:/usr/local/src/bench/libMicro/libMicro-0
> .3.0> exit
>
> That is where I get these huge 1 to 3 second delays.
Just to add a little to this. Through almost every iteration through the
test I hit these pollsys calls that take 3 seconds. What is happening
is I don't think they are recorded in the statistics because the delay
is too large. I ran Brenden's procsystime but it dosn't record pollsys
which I'm suprised as topsyscall shows pollsys. See below. Note
the wiatsys is the true time of the test. Also the conn_accept test
has the same problem EAGAIN after accept random 3 sec pollsys delays.
procsystime bin/connection -E -C 200 -L -S -W -N connection -B 256
Elapsed Times for command bin/connection -E -C 200 -L -S -W -N connection -B
256,
SYSCALL TIME (ns)
sigaltstack 2002
sigpending 2363
getgid 3281
setpgrp 3497
getuid 4814
getpid 5330
fcntl 8450
setcontext 9502
open64 13818
getrlimit 13936
sysconfig 14160
fstat64 14313
lwp_sigmask 16851
sysi86 23910
access 27718
brk 39219
getcwd 45198
pipe 45554
open 61739
schedctl 70021
munmap 73703
ioctl 81812
sigaction 90164
close 94536
resolvepath 95467
xstat 98955
stat64 112050
mmap 254948
memcntl 257899
fork1 666229
exece 738648
write 769508
read 7837105
waitsys 124837118120
[EMAIL PROTECTED]:/export/home0/local/src/bench/libMicro/libMicro-0.3.0> exit
This message posted from opensolaris.org
_______________________________________________
perf-discuss mailing list
[email protected]
