Mail Archives: cygwin/2003/10/03/09:01:48
Hi.
Yes, I've read http://cygwin.com/faq/faq_4.html#SEC33
No, I have no network share in path 0=)
I'm writing this message because it *seems* to be that this issue is not
very old, but I couln't say if this is exact, expected, or always was.
$ cat do-something.sh
#!/bin/bash
for ((i=0; i<10; i++)); do echo -n $i; done; echo
On CygWin:
$ time for ((i=0; i<10; i++)); do echo -n $i; done; echo
0123456789
real 0m0.001s
user 0m0.000s
sys 0m0.000s
$ time ./do-something.sh
0123456789
real 0m0.686s
user 0m0.062s
sys 0m0.015s
While, e.g. on FreeBSD:
$ time for ((i=0; i<10; i++)); do echo -n $i; done; echo
0123456789
real 0m0.000s
user 0m0.000s
sys 0m0.000s
$ time ./do-something.sh
0123456789
real 0m0.004s
user 0m0.001s
sys 0m0.002s
OK, I know that Windows doesn't have pipes and cygwin, in emulating
them, *cannot be* as fast, BUT I don't remember CygWin taking almost 1
second to spawn, in the past.
But, I repeat, maybe I only forgot the past... but it really didn't seem
to take "5 minutes" to complete a typical cofngiure script (which
involves LOTS of spawning.)
Is is not, btw, the "priority" sisue that it was some time ago, as the
700ms time is the same either the CPU being 0% or 100% (with idle tasks).
Seems really strange to me... ?_?
Even a single "echo" takes up to 500ms:
$ time /bin/echo 1
1
real 0m0.442s
user 0m0.030s
sys 0m0.030s
--
Lapo 'Raist' Luchini
lapo AT lapo DOT it (PGP & X.509 keys available)
http://www.lapo.it (ICQ UIN: 529796)
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
- Raw text -