Re: Serious performance problems (new snapshot has some performance improvement)

2005-05-31 Thread Ralf Habacker
Am Dienstag, 31. Mai 2005 12:42 schrieb Gerrit P. Haase: > Ralf Habacker wrote: > > > Interesting, why is it faster when running a binary that doesn't > depend on cygwin1.dll after swapping the DLL? Some Win caching mechanism? > >>> > >>> > >>>I recognized this caching behavior with KDE

Re: Serious performance problems (new snapshot has some performance improvement)

2005-05-31 Thread Gerrit P. Haase
Ralf Habacker wrote: Interesting, why is it faster when running a binary that doesn't depend on cygwin1.dll after swapping the DLL? Some Win caching mechanism? I recognized this caching behavior with KDE/cygwin too. Under http://cvs.sourceforge.net/viewcvs.py/kde-cygwin/tools/fillmem/ the

Re: Serious performance problems (new snapshot has some performance improvement)

2005-05-31 Thread Ralf Habacker
Am Dienstag, 31. Mai 2005 12:05 schrieb Gerrit P. Haase: > [EMAIL PROTECTED] wrote: > > Am Montag, 30. Mai 2005 22:33 schrieb Gerrit P. Haase: > > > >>Anonymous wrote: > >> > >> > >>>My System: > >>>#Set-up: > >>> > >>>$ g++ cygspd.cc -o cygspd-basic > >>>$ g++ -O7 cygspd.cc -o cygspd-o7 > >>>$ g+

Re: Serious performance problems (new snapshot has some performance improvement)

2005-05-31 Thread Gerrit P. Haase
[EMAIL PROTECTED] wrote: Am Montag, 30. Mai 2005 22:33 schrieb Gerrit P. Haase: Anonymous wrote: My System: #Set-up: $ g++ cygspd.cc -o cygspd-basic $ g++ -O7 cygspd.cc -o cygspd-o7 $ g++ -fno-exceptions cygspd.cc -o cygspd-ne $ g++ -O7 -fno-exceptions cygspd.cc -o cygspd-ne-o7 $ g++ -mno-c

Re: Serious performance problems (new snapshot has some performance improvement)

2005-05-30 Thread Anonymous
There is no reason to expect any improvement in mingw programs from my change. Yeah i figured that one out. ;) I was just trying to rationalize the change that I did show in the original report. My timings don't show great improvement, and a 1 second margin of error due to windows caching doe

Re: Serious performance problems (new snapshot has some performance improvement)

2005-05-30 Thread Christopher Faylor
On Mon, May 30, 2005 at 08:00:35PM -0400, Tacvek wrote: >>Interesting, why is it faster when running a binary that doesn't depend >>on cygwin1.dll after swapping the DLL? Some Win caching mechanism? > >Repeating many times should minimize any caching: > >#New DLL >$ /bin/time -f "%E %S %U" cygspd-

Re: Serious performance problems (new snapshot has some performance improvement)

2005-05-30 Thread Tacvek
Interesting, why is it faster when running a binary that doesn't depend on cygwin1.dll after swapping the DLL? Some Win caching mechanism? Repeating many times should minimize any caching: #New DLL $ /bin/time -f "%E %S %U" cygspd-mingw-ne-o7 cygspd.dat 0:04.48 0.01 0.02 0:01.40 0.00 0.02 0:

Re: Serious performance problems (new snapshot has some performance improvement)

2005-05-30 Thread ralf . habacker
Am Montag, 30. Mai 2005 22:33 schrieb Gerrit P. Haase: > Anonymous wrote: > > > My System: > > #Set-up: > > > > $ g++ cygspd.cc -o cygspd-basic > > $ g++ -O7 cygspd.cc -o cygspd-o7 > > $ g++ -fno-exceptions cygspd.cc -o cygspd-ne > > $ g++ -O7 -fno-exceptions cygspd.cc -o cygspd-ne-o7 > > $ g++ -

Re: Serious performance problems (new snapshot has some performance improvement)

2005-05-30 Thread Gerrit P. Haase
Anonymous wrote: My System: #Set-up: $ g++ cygspd.cc -o cygspd-basic $ g++ -O7 cygspd.cc -o cygspd-o7 $ g++ -fno-exceptions cygspd.cc -o cygspd-ne $ g++ -O7 -fno-exceptions cygspd.cc -o cygspd-ne-o7 $ g++ -mno-cygwin cygspd.cc -o cygspd-mingw $ g++ -O7 -mno-cygwin cygspd.cc -o cygspd-mingw-o7 $

Re: Serious performance problems (new snapshot has some performance improvement)

2005-05-30 Thread Anonymous
My System: #Set-up: $ g++ cygspd.cc -o cygspd-basic $ g++ -O7 cygspd.cc -o cygspd-o7 $ g++ -fno-exceptions cygspd.cc -o cygspd-ne $ g++ -O7 -fno-exceptions cygspd.cc -o cygspd-ne-o7 $ g++ -mno-cygwin cygspd.cc -o cygspd-mingw $ g++ -O7 -mno-cygwin cygspd.cc -o cygspd-mingw-o7 $ g++ -fno-exception

Re: Serious performance problems (new snapshot has some performance improvement)

2005-05-29 Thread Christopher Faylor
On Sun, May 29, 2005 at 09:37:16PM +0200, Gerrit P. Haase wrote: >Christopher Faylor wrote: > >>On Sat, May 28, 2005 at 11:05:23PM -0400, Christopher Faylor wrote: >> >>>On Sat, May 28, 2005 at 08:18:46PM -0400, Christopher Faylor wrote: >>> I have an idea about how to work around this problem

Re: Serious performance problems (new snapshot has some performance improvement)

2005-05-29 Thread Gerrit P. Haase
Christopher Faylor wrote: On Sat, May 28, 2005 at 11:05:23PM -0400, Christopher Faylor wrote: On Sat, May 28, 2005 at 08:18:46PM -0400, Christopher Faylor wrote: I have an idea about how to work around this problem but I have to think about how dangerous it might be. Basically removing the

Re: Serious performance problems (new snapshot has some performance improvement)

2005-05-29 Thread Christopher Faylor
On Sat, May 28, 2005 at 11:05:23PM -0400, Christopher Faylor wrote: >On Sat, May 28, 2005 at 08:18:46PM -0400, Christopher Faylor wrote: >>I have an idea about how to work around this problem but I have to think >>about how dangerous it might be. Basically removing the signal handling >>wrapper ar

Re: Serious performance problems (new snapshot has some performance improvement)

2005-05-28 Thread Christopher Faylor
On Sat, May 28, 2005 at 08:18:46PM -0400, Christopher Faylor wrote: >I have an idea about how to work around this problem but I have to think >about how dangerous it might be. Basically removing the signal handling >wrapper around pthread_getspecific and pthread_setspecific. That may >work ok but