Re: ASAN:DEADLYSIGNAL

[Ben Abbott]

> On Jun 17, 2016, at 10:46 PM, Ben Abbott <address@hidden> wrote:
> 
> John,
> 
> I don’t know if you’ll have any insight on this, but after bisecting, I 
> isolated the changeset below as responsible for the error that follows. I’ve 
> seen ASAN errors in the past. They were eventually fixed by subsequent 
> changesets, but I don’t recall the details.
> 
> The first bad revision is:
> changeset:   21926:24215a16f3b2
> user:        John W. Eaton <address@hidden>
> date:        Thu Jun 16 10:02:49 2016 -0400
> summary:     hide dirent.h header
> 
>  GEN      doc/interpreter/plot-figureproperties.texi
> ASAN:DEADLYSIGNAL
> =================================================================
> ==21503==ERROR: AddressSanitizer: SEGV on unknown address 0x000000000040 (pc 
> 0x7fff8d3994cd bp 0x7fff5159a380 sp 0x7fff5159a300 T0)
> 
> Ben

I’m still seeing ASAN:DEADLYSIGNAL

$ hg tip
changeset:   21952:a5924b0cbf42
bookmark:    @
tag:         tip
user:        Rik <address@hidden>
date:        Sun Jun 19 06:23:54 2016 -0700
summary:     Return "Screensize" for root graphics property "MonitorPositions" 
(partial fix bug #48239).

  GEN      doc/interpreter/voronoi.txt
ASAN:DEADLYSIGNAL
=================================================================
==68284==ERROR: AddressSanitizer: SEGV on unknown address 0x000000000040 (pc 
0x7fff8d3994cd bp 0x7fff51568360 sp 0x7fff515682e0 T0)
^Coctave exited with signal 2
Makefile:25955: recipe for target 'all-recursive' failed
make[1]: *** [all-recursive] Interrupt: 2
Makefile:9810: recipe for target 'all' failed
make: *** [all] Interrupt: 2

I’m lost on how to dig deeper. Any advice on how to isolate the problem?

Ben