Raul Metsma wrote:
http://rtedev.com/~raul/threads/threads.c
Compile this program under mingw
gcc -o threads.exe threads.c
./threads 600
now compile
http://rtedev.com/~raul/threads/main.c
with cygwin
and when I execute this, then threads.exe will eat 100% CPU
Is there any chance get this
No, I tested with just two cmd windows. The joke is that I don't see
the Cygwin process doing anything suspicious. In my test I just started
tcsh under strace to see if it begins to behave weird when the threads
application is running, but it just waits in ready_for_read for user input.
Sorry no Makefile based project, but in Visual C
http://rtedev.com/~raul/threads/Threads.zip
Now when I execute this program and run some amount threads (like 600)
and then run the while(1) sleep(1);, which is compiled
under cygwin, then the first program will use 100% CPU.
Actually my CPU usage
Hello,
On Thu, 25 Aug 2005, Raul Metsma wrote:
Sorry no Makefile based project, but in Visual C
http://rtedev.com/~raul/threads/Threads.zip
By doing this you are seriously reducing the chance that someone will look
at your testcase.
--
Unsubscribe info:
Ohh never get this right :(
Actually I am more system administrator than programmer
Lets try again:
http://rtedev.com/~raul/threads/threads.c
Compile this program under mingw
gcc -o threads.exe threads.c
./threads 600
now compile
http://rtedev.com/~raul/threads/main.c
with cygwin
and when I
On Aug 25 12:30, Raul Metsma wrote:
Ohh never get this right :(
Actually I am more system administrator than programmer
Lets try again:
http://rtedev.com/~raul/threads/threads.c
Compile this program under mingw
gcc -o threads.exe threads.c
./threads 600
now compile
Original Message
From: Corinna Vinschen
Sent: 25 August 2005 11:03
On Aug 25 12:30, Raul Metsma wrote:
Ohh never get this right :(
Actually I am more system administrator than programmer
Lets try again:
http://rtedev.com/~raul/threads/threads.c
Compile this program under mingw
Seems this is something related CPU-s or total Windows thread count
I tested it on Windwos 2003, where are 2x2,4GHz Intel Xeon
There occurs this issue on 800 threads
Maybe is this something related with OS?
Raul Metsma
Corinna Vinschen wrote:
On Aug 25 12:30, Raul Metsma wrote:
Ohh never get
Hello,
On Thu, 25 Aug 2005, Raul Metsma wrote:
Ohh never get this right :(
Actually I am more system administrator than programmer
Lets try again:
http://rtedev.com/~raul/threads/threads.c
Compile this program under mingw
gcc -o threads.exe threads.c
./threads 600
now compile
On Aug 25 13:40, Raul Metsma wrote:
Seems this is something related CPU-s or total Windows thread count
I tested it on Windwos 2003, where are 2x2,4GHz Intel Xeon
There occurs this issue on 800 threads
You're right, when raising the number of threads, I see this test
gradually taking more CPU
Hello,
On Thu, 25 Aug 2005, Corinna Vinschen wrote:
On Aug 25 13:40, Raul Metsma wrote:
Seems this is something related CPU-s or total Windows thread count
I tested it on Windwos 2003, where are 2x2,4GHz Intel Xeon
There occurs this issue on 800 threads
You're right, when raising the
On Aug 25 14:59, Pavel Tsekov wrote:
Hello,
On Thu, 25 Aug 2005, Corinna Vinschen wrote:
On Aug 25 13:40, Raul Metsma wrote:
Seems this is something related CPU-s or total Windows thread count
I tested it on Windwos 2003, where are 2x2,4GHz Intel Xeon
There occurs this issue on
You're on W2K, I'm on XP, maybe there's a difference? I can easily
reproduce nearly 0% CPU with 1000 threads if no Cygwin process is
present and nearly 99% if a Cygwin process is present.
I have XP SP2 laptop and it occures on 200 threads
Raul Metsma
--
Unsubscribe info:
Original Message
From: Corinna Vinschen
Sent: 25 August 2005 13:09
Even if we can't do much about it, it would be interesting to find out
how Cygwin affects CPU usage of non-Cygwin processes.
When I moved from 2k to XP I noticed that running big cross compiles in
the background
Hello,
On Thu, 25 Aug 2005, Corinna Vinschen wrote:
Here when the app is compiled with -mno-cygwin it takes 100% cpu when it
allocates 400+ threads - both from Windows Command Prompt and Cygwin bash.
After all the testcase consist almost entirely of Win32 calls.
You're on W2K, I'm on XP,
Corinna Vinschen wrote:
You're on W2K, I'm on XP, maybe there's a difference? I can easily
reproduce nearly 0% CPU with 1000 threads if no Cygwin process is
present and nearly 99% if a Cygwin process is present.
Even if we can't do much about it, it would be interesting to find out
how
On Aug 25 15:40, Pavel Tsekov wrote:
Hello,
On Thu, 25 Aug 2005, Corinna Vinschen wrote:
Here when the app is compiled with -mno-cygwin it takes 100% cpu when it
allocates 400+ threads - both from Windows Command Prompt and Cygwin bash.
After all the testcase consist almost entirely
There seems to be some kind issues with cygwin sleep()
and Windows threads.
I created one test program which creates 600 threads
with native windows Sleep(1).
Now when I compile simple c program under cygwin
while(1) { sleep(1); }
and execute it, the first program will eat 100% CPU.
It does
Original Message
From: Raul Metsma
Sent: 24 August 2005 17:11
Now when I compile simple c program under cygwin
while(1) { sleep(1); }
and execute it, the first program will eat 100% CPU.
I can't reproduce this. When I tried it, sleep worked as expected, and
the program
Original Message
From: raul
Sent: 24 August 2005 18:59
Raul, please keep these posts on the mailing list.
http://cygwin.com/acronyms#PPIOSPE
I uploaded sample programs to http://rtedev.com/~raul/threads/
threads.exe is written in delphi and Threads2.exe in visual c++
That was
Just tested it at home, seems like it doesnt occure on Windows Mediacenter
I send tomorrow the test cases
Algne kiri -
Kellelt: Jason Pyeron [EMAIL PROTECTED]
Kellele: [EMAIL PROTECTED]
Kuupäev: Kolmapäev, 24 August, 2005, 14:59
Pealkiri: Re: RE: Problems with Windows threads and cygwin
One more note, I tested cygwin dll versions 1.5.12 - 1.5.18,
all have the same problem
I'm running 1.5.18-1 here, and it doesn't have the problem.
Also, nobody else has reported such a problem. There's no
way that a
function as fundamental as sleep (...) could have been broken
raul wrote:
[snip]
how about 6 windows 2003 servers, my laptop with XP and couple more
desktops wit windows 2000 and xp
[snip]
Are you running the same compiled program? or have you compiled on more than one
machine?
I can corroborate that there is no problem with sleep() and latest Cygwin
23 matches
Mail list logo
