Re: include tag in uname -s for ARM64 hosts

2024-11-30 Thread Jon Turney via Cygwin
On 20/11/2024 19:34, Jeremy Drake via Cygwin wrote: I mentioned this briefly on cygwin-patches during the review of the fix for a deadlock when running under emulation on ARM64 hosts, but I wanted to discuss it more thoroughly here now that that is merged. It seems like it would be useful, perha

Re: include tag in uname -s for ARM64 hosts

2024-11-21 Thread Jeremy Drake via Cygwin
On Thu, 21 Nov 2024, Corinna Vinschen via Cygwin wrote: > IMHO: > > if (IsWow64Process2 (GetCurrentProcess (), &emulated, &hosted)) > && emulated != IMAGE_FILE_MACHINE_UNKNOWN) > if (hosted == IMAGE_FILE_MACHINE_AMD64) > strcat (sysname, "-WOW64"); > else if (hosted == IMAGE_

Re: include tag in uname -s for ARM64 hosts

2024-11-21 Thread Corinna Vinschen via Cygwin
On Nov 21 10:15, Jeremy Drake via Cygwin wrote: > On Thu, 21 Nov 2024, Corinna Vinschen via Cygwin wrote: > > > > i686 x86_64 -WOW64-x64 (or omit arch to match existing?) > > > > Just -WOW64 > > > > For backward compat and it's a sole representative of an intel-intel > > emulation anyway. I

Re: include tag in uname -s for ARM64 hosts

2024-11-21 Thread Jeremy Drake via Cygwin
On Thu, 21 Nov 2024, Corinna Vinschen via Cygwin wrote: > > i686x86_64 -WOW64-x64 (or omit arch to match existing?) > > Just -WOW64 > > For backward compat and it's a sole representative of an intel-intel > emulation anyway. I think we may safely ignore "WOW" > > > i686ARM64 -W

Re: include tag in uname -s for ARM64 hosts

2024-11-21 Thread Corinna Vinschen via Cygwin
On Nov 20 11:34, Jeremy Drake via Cygwin wrote: > I mentioned this briefly on cygwin-patches during the review of the fix > for a deadlock when running under emulation on ARM64 hosts, but I wanted > to discuss it more thoroughly here now that that is merged. > > It seems like it would be useful, p

Re: include tag in uname -s for ARM64 hosts

2024-11-20 Thread Jeremy Drake via Cygwin
On Wed, 20 Nov 2024, Jeremy Drake via Cygwin wrote: > CygwinHostSuffix > Not worth worrying about: > ARM ARM64 -WOW64-ARM (?) Oops, this should have been -WOW64-ARM64 since the host is ARM64. But like I said, this case would never happen anyway. -- Problem reports: https:/