On 12/01/2024 14:09, Jon Turney wrote:
> Pre-format a command to be executed on a fatal error to run 'dumper'
> (using an absolute path).
> 
> Factor out executing a pre-formatted command, so we can use that for
> invoking the JIT debugger in try_to_debug() (if error_start is present
> in the CYGWIN env var) and to invoke dumper when a fatal error occurs.
> 

So, there is a small problem with this change: because dumper itself 
terminates the dumped process, it doesn't go on to exit with the 
signal+128 exit status.

(In fact, it seems to exit with status 0 when terminated by an attached 
debugger terminating, which isn't great)

That's relatively easy to fix: just use the '-n' option to dumper so it 
detaches before exiting, to prevent that terminating the dumped process, 
but then we run into the difficulties of reliably detecting that dumper 
has attached and done it's work, so it's safe for us to exit.

Attached patch does that, and documents the expectations on the 
error_start command a bit more clearly.

Even then this is clearly not totally bullet-proof. Maybe the right 
thing to do is add a suitable timeout here, so even if we fail to notice 
the DebugActiveProcess() (or there's a custom JIT debugger which just 
writes the fact a process crashed to a logfile or something), we'll exit 
eventually?