On Thursday, February 29, 2024 at 01:21:01 PM EST, Kate Deplaix via Cygwin <cygwin@cygwin.com> wrote: 
> At some point of the make process, one job will infinit loop (taking 100% of one core), while the rest of the mingw jobs will idle and never finish. After some time Cygwin Terminal isn't 
> responsive anymore, and I'm not able to open another one or to spawn any other processes that are relying on the cygwin dll. I was able to observe this behaviour every time i tried 
> consistently.

I have repro'd this problem with the exception that there was no 100% cpu usage.  Looks like an exception handling loop:

(2c20.4df8): Access violation - code c0000005 (first chance)
First chance exceptions are reported before any exception handling.
This exception may be expected and handled.
cygwin1!dirname+0x198e:
00007ffe`7dbf95ad 458b5c2408      mov     r11d,dword ptr [r12+8] ds:00000001`a0030008=????????
0:000> g
(2c20.4df8): Access violation - code c0000005 (first chance)
First chance exceptions are reported before any exception handling.
This exception may be expected and handled.
cygwin1!dirname+0x198e:
00007ffe`7dbf95ad 458b5c2408      mov     r11d,dword ptr [r12+8] ds:00000001`a0030008=????????
0:000> g
(2c20.4df8): Access violation - code c0000005 (first chance)
First chance exceptions are reported before any exception handling.
This exception may be expected and handled.
cygwin1!dirname+0x198e:
00007ffe`7dbf95ad 458b5c2408      mov     r11d,dword ptr [r12+8] ds:00000001`a0030008=????????
0:000> g
(2c20.4df8): Access violation - code c0000005 (first chance)
First chance exceptions are reported before any exception handling.
This exception may be expected and handled.
cygwin1!dirname+0x198e:
00007ffe`7dbf95ad 458b5c2408      mov     r11d,dword ptr [r12+8] ds:00000001`a0030008=????????


STACK_TEXT:  
00000007`ffffc8c0 00007ffe`7dc57f74     : 00000008`00039b40 00000008`00039b40 00000000`00010002 0000000a`00000298 : cygwin1!dirname+0x198e
00000007`ffffc980 00007ffe`7dc56a78     : 00000000`00f2f2f2 00000000`00000000 00000007`ffffcd30 00007ffe`e44773ae : cygwin1!cuserid+0x16e16
00000007`ffffc9c0 00007ffe`7dc581d3     : 00ff783b`00767676 00000000`005b0720 00000000`00000002 00000008`00004870 : cygwin1!cuserid+0x1591a
00000007`ffffca40 00007ffe`7dbb954d     : 00020000`00000050 00000000`000a3dd0 00000000`000ab7b0 00000000`00000000 : cygwin1!cuserid+0x17075
00000007`ffffca70 00007ffe`7dbb99c2     : 00000007`ffffcc8a 00007ffe`7dba7d37 00000007`ffffcd30 00000007`00000c90 : cygwin1!dlfork+0x2695
00000007`ffffcb70 00007ffe`7dba7187     : 315c615c`305c625c 2f00735c`2e5c625c 5f5c6c5c`665c615c 745c735c`6e5c695c : cygwin1!dlfork+0x2b0a
00000007`ffffcbf0 00007ffe`7dba5e08     : 00000000`00000000 00000000`00000000 00000000`00000000 00000000`00000000 : cygwin1!cygwin_dll_init+0x381
00000007`ffffcd80 00007ffe`7dba5e86     : 00000000`00000000 00000000`00000000 00000000`00000000 00000000`00000000 : cygwin1!_assert+0x23f0
00000007`ffffcdd0 00000000`00000000     : 00007ffe`7dd0604f 00000008`0003a210 00000000`00000000 00000008`00039400 : cygwin1!_assert+0x246e


gdb does not connect...

Other interesting thing, the "bad" gcc has cons0 tty attached:

18907       1   18907      11296  cons0   1610619958 16:19:23 /usr/bin/x86_64-w64-mingw32-gcc

Kevin