Re: sched_getaffinity() always returns the full cpu mask

2025-03-08 Thread Christian Franke via Cygwin
Brian Inglis via Cygwin wrote: On 2025-03-08 10:11, Christian Franke via Cygwin wrote: $ grep processor /proc/cpuinfo | tail -1 # i7-14700K processor   : 27 coreutils nproc should show the equivalent result 28 Works as expected: $ nproc 28 $ busybox nproc 28 -- Problem reports:

python 3.12.9 hangs on Flask

2025-03-08 Thread Masamichi Hosoda via Cygwin
Hi. I've found python 3.12.9 hangs on Flask. After starting the Flask HTTP server, the first HTTP access works fine but hangs immediately afterward. Once hung, CTRL+C and sending any signals have no effect. Here's a reproduction procedure. ``` $ uname -srvmo CYGWIN_NT-10.0-19045 3.5.7-1.x86_64 2

Re: sched_getaffinity() always returns the full cpu mask

2025-03-08 Thread Mark Geisert via Cygwin
Hi Christian, On 3/8/2025 9:11 AM, Christian Franke via Cygwin wrote: Testcase: $ grep processor /proc/cpuinfo | tail -1 # i7-14700K processor   : 27 $ sleep 60 & taskset 0x1 sleep 60 & [1] 62094 [2] 62095 $ taskset -p 62094 pid 62094's current affinity mask: fff $ taskset -p 62095 p

sched_getaffinity() always returns the full cpu mask

2025-03-08 Thread Christian Franke via Cygwin
Testcase: $ grep processor /proc/cpuinfo | tail -1 # i7-14700K processor   : 27 $ sleep 60 & taskset 0x1 sleep 60 & [1] 62094 [2] 62095 $ taskset -p 62094 pid 62094's current affinity mask: fff $ taskset -p 62095 pid 62095's current affinity mask: fff Another test shows that the

Re: mv error building LilyPond tests, maybe MAX_PATH related

2025-03-08 Thread Marco Atzeri via Cygwin
On 08/03/2025 11:39, Marco Atzeri wrote: On 08/03/2025 06:10, Saul Tobin via Cygwin wrote: Hello, Based on this, it appears to me that when an in-place rename is performed by `mv`, it calls into a Win32 API that enforces the 260 character MAX_PATH. I can see in the trace that the error is o

Testing with UNC paths && GNU mkinstalldirs not working with (Cygwin) UNC dirs ...

2025-03-08 Thread Roland Mainz via Cygwin
HI! While doing testing with UNC paths I found that GNU mkinstalldirs does not handle UNC paths - for example creating subdirs in //foo.bar.com/chickmonster/dir1/dir2/ will *partially* appear in /foo.bar.com/chickmonster/dir1/dir2/ Workaround is in https://github.com/kofemann/ms-nfs41-client

Re: Testing with UNC paths && GNU mkinstalldirs not working with (Cygwin) UNC dirs ...

2025-03-08 Thread Marco Atzeri via Cygwin
On 08/03/2025 19:22, Roland Mainz via Cygwin wrote: HI! - Does anyone know whether there is a single upstream for GNU mkinstalldirs, or do the various GNU projects just copy the script around ? Bye, Roland I would bet on Automake /usr/share/automake-1.16/mkinstalldirs -- Problem rep

Re: Testing with UNC paths && GNU mkinstalldirs not working with (Cygwin) UNC dirs ...

2025-03-08 Thread Brian Inglis via Cygwin
On 2025-03-08 11:22, Roland Mainz via Cygwin wrote: While doing testing with UNC paths I found that GNU mkinstalldirs does not handle UNC paths - for example creating subdirs in //foo.bar.com/chickmonster/dir1/dir2/ will *partially* appear in /foo.bar.com/chickmonster/dir1/dir2/ Workaround is in

Re: sched_getaffinity() always returns the full cpu mask

2025-03-08 Thread Brian Inglis via Cygwin
On 2025-03-08 10:11, Christian Franke via Cygwin wrote: $ grep processor /proc/cpuinfo | tail -1 # i7-14700K processor   : 27 coreutils nproc should show the equivalent result 28 -- Take care. Thanks, Brian Inglis Calgary, Alberta, Canada La perfection est atteinte

Re: mv error building LilyPond tests, maybe MAX_PATH related

2025-03-08 Thread Marco Atzeri via Cygwin
On 08/03/2025 06:10, Saul Tobin via Cygwin wrote: Hello, I've encountered a kind of bizarre error when trying to build GNU LilyPond regression tests on Cygwin 3.5.7-1 on Windows 11. For one and only one folder, `mv` fails with `mv: cannot move './out-test' to './out-test-baseline': Permission de

Re: cygwin 3.6.0: No signals received after swapcontext() is used

2025-03-08 Thread Christian Franke via Cygwin
Corinna Vinschen via Cygwin wrote: On Mar 6 13:24, Christian Franke via Cygwin wrote: Found because 'stress-ng --context 1 ...' always hangs. The attached testcase uses the example from Linux swapcontext(3) to call the context functions. Just tested with 3.5.3 and it doesn't work there, eithe