Mail Archives: cygwin/2002/07/27/22:55:53
> Thus spoke: "Gary R. Van Sickle" <g dot r dot vansickle at worldnet
dot att
> dot net>
>
>>Slower than... not running at all?
>
>
> Almost, and yes, to the point where cygwin is so slow that it is not a
> practical foundation for program development. Since the .10 revision,
> literally seconds go by between simple operations such as when
executing a
> .sh list of file copies.
>
> Under DOS, images of the Harbour compiler (http://www.harbour.org) are
> downloaded from Sourceforge, then cygwin (gcc) is used to build images of
> the compiler, all the while keeping cygwin current. Cygwin has proved an
> excellent platform, until the last two major revisions. What used to
take
> only several minutes to run through the makefile now can extend beyond
> twenty!
>
> The computer here is an old Pentium 166 with 64 mbytes of memory, but the
> response has been quite adequate. I would expect even a faster
system would
> not fully mask whatever is inhibiting the current performance of cygwin.
>
> Sorry I can't be more helpful or more specific.
>
> Bill
> billsmith AT ispwest DOT com
>
>
Quite a while back I posted a bash shell script that developed a major
performance problem with the .11 release. I skipped the .10 release, so
maybe it was a problem there, too. The few people who ran the shell
script said it worked just fine on their machines. I tried it on some
other cygwin installations, and indeed, it did work fine. So, there
seems to be something different about one of my installations. The one
I've noticed performance problems with is running WinME, and the ones
that worked fine were Win NT 4.0. Given the number of people
undoubtedly running WinME, I'd be very surprised if it were as simple as
that...
Since then I've been monitoring this list for performance issues. There
have been several, but it looks to me like it must be a very small
minority of us who are experiencing these weird problems. Certainly
none of the primary cygwin "kernel" developers must be seeing them, or
else I'd guess they would have dug into the problem already.
Is there someone who knows their way around the cygwin kernel enough to
help us try to fix this problem? Maybe someone could suggest some
specific tests that might shed some light on what is different about
this small percentage of machines that seem prone to the performance
problems. I like cygwin a lot, but I find it very frustrating that I
can't get a handle on what's causing the poor performance. I'm sure
there are several of us who would be very appreciative of any help.
Thanks.
-Russ
jorg928 AT softhome DOT net
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Bug reporting: http://cygwin.com/bugs.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
- Raw text -