Cygwin 3.4.3 and 3.5.0... hangs in make, top, procps, ls /proc/PID/...

[Takashi Yano]

On Mon, 2 Jan 2023 11:32:01 +0900
Takashi Yano wrote:
> On Thu, 29 Dec 2022 21:59:45 -0700
> Brian Inglis wrote:
> > I got some hangs (deadlock?) between (parallel?) make jobs, top, procps, and 
> > even ls /proc/*/ when trying to cygport all check curl or look at the process 
> > statuses when builds hung under Cygwin 3.4.3 and 3.5.0-0.69...
> > 
> > Had to revert to a Cygwin 3.4.0-344 test build from Dec 16 sources to build and 
> > check curl etc.
> > 
> > Only external differences seems to be recent dll sizes are *MUCH* smaller:
> > 
> > $ ls -glort /bin/cygwin1*.dll
> > -rwxr-xr-x 1 3588124 Sep  5 05:17 /bin/cygwin1-prev.dll
> > -rwxr-xr-x 1 3561202 Dec 16 00:33 /bin/cygwin1-test.dll
> > -rwxr-xr-x 1 3561202 Dec 16 00:33 /bin/cygwin1.dll
> > -rwxr-xr-x 1 2951221 Dec 16 05:50 /bin/cygwin1-save.dll
> > -rwxr-xr-x 1 2951221 Dec 16 05:50 /bin/cygwin1-curr.dll
> > -rwxr-xr-x 1 2940981 Dec 23 03:03 /bin/cygwin1-69.dll
> > 
> > and tars:
> > -rwxr-xr-x 1 3563651 2022-12-04 06:09 usr/bin/cygwin1.dll cygwin-3.4.0-1
> > -rwxr-xr-x 1 3563526 2022-12-10 13:12 usr/bin/cygwin1.dll cygwin-3.4.1-1
> > -rwxr-xr-x 1 3564074 2022-12-11 05:06 usr/bin/cygwin1.dll cygwin-3.4.2-1
> > -rwxr-xr-x 1 2951221 2022-12-16 05:50 usr/bin/cygwin1.dll cygwin-3.4.3-1
> > -rwxr-xr-x 1 2940981 2022-12-23 03:03 usr/bin/cygwin1.dll cygwin-3.5.0-0.69...
> > 
> > Anyone seen this issue or anything come to mind to cause or help diagnose this?
> > 
> > Running MS Win 10.0.19044.2364 21H2 patched to date.
> 
> I have looked into this issue a bit, and found that
> q->sigtls becomes sometimes NULL and access violation
> occurs at the following code.
> 
> winsup/cygwin/sigproc.cc: 1378
> 		if (q->sigtls->sigmask & (bit = SIGTOMASK (q->si.si_signo)))
> 		  {
> 		    tl_entry = cygheap->find_tls (q->si.si_signo, issig_wait);
> 
> I'm not sure why this happens, however it seems that
> the following patch fixes the issue.
> 
> diff --git a/winsup/cygwin/sigproc.cc b/winsup/cygwin/sigproc.cc
> index ce36c8be3..90eaa2a47 100644
> --- a/winsup/cygwin/sigproc.cc
> +++ b/winsup/cygwin/sigproc.cc
> @@ -1375,6 +1375,8 @@ wait_sig (VOID *)
>  	    *pack.mask = 0;
>  	    while ((q = q->next))
>  	      {
> +		if (q->sigtls == NULL)
> +		  continue;
>  		if (q->sigtls->sigmask & (bit = SIGTOMASK (q->si.si_signo)))
>  		  {
>  		    tl_entry = cygheap->find_tls (q->si.si_signo, issig_wait);
> 
> Corinna, could you please have a look?

Do we need to guard here as well?

	case __SIGPENDING:
	  {
	    unsigned bit;

	    *pack.mask = 0;
	    tl_entry = cygheap->find_tls (pack.sigtls);
	    while ((q = q->next))
	      {
		/* Skip thread-specific signals for other threads. */
		if (q->sigtls && pack.sigtls != q->sigtls)
		  continue;
------->	if (pack.sigtls->sigmask & (bit = SIGTOMASK (q->si.si_signo)))
		  *pack.mask |= bit;
	      }
	    cygheap->unlock_tls (tl_entry);
	  }
	  break;

-- 
Takashi Yano <takashi.yano@nifty.ne.jp>