Re: PID-reuse races fix, introduced GCC validation brakage

[Pedro Alves]

On 03/30/2016 04:03 PM, Yvan Roux wrote:
> Hi,
> 
> https://lists.gnu.org/archive/html/dejagnu/2015-07/msg00000.html
> 
> this patch introduced  a new failure related to GDB testing, but this
> time in GCC guality part of the testsuite.  When
> gcc/testsuite/gcc.dg/guality/example.c is executed we have this set of
> processes started
> 
> PID  PPID     command
> 100  99          ./example.exe
> 101  99          cat
> 102  100        sh -c gdb -nx -nw --quiet > /dev/null 2>&1 ./example.exe
> 103  102        gdb -nx -nw --quiet ./example.exe
> 
> The issue is that when close_wait_program is called pid="100 101" and
> when the "wait" returns none of these 4 processes is actually killed,
> and the killing of the "kill pipeline" made these pids untouched and
> the validation hangs forever.

The questions I'm not seeing answered are:

 - How did this work before?

 - What exactly in the patch causes the regression?

AFAICS, before my patch we had:

- if { $pid > 0 } {
-    verbose "doing kill, pid is $pid"
-    # This is very, very nasty. SH, instead of EXPECT, is used
-    # to run this in the background since, on older CYGWINs, a
-    # strange file I/O error occures.
-    set pgid "-[join $pid { -}]"

while my patch changed it to:

+    if { $pid > 0 } {
+       # Tcl has no kill primitive, so we have to execute an external
+       # command in order to kill the process.
+       verbose "doing kill, pid is $pid"
+       # Prepend "-" to generate the "process group ID" needed by
+       # kill.
+       set pgid "-$pid"

I probably did that change just assuming pid is a number, given
the pid > 0 check.

So if you replace:

 + set pgid "-$pid"

with:

 - set pgid "-[join $pid { -}]"

I imagine it would get things back to how they were for you?

Thanks,
Pedro Alves