Re: [lmi] 'wine test_coding_rules.exe' segfault

[Vadim Zeitlin]

On Fri, 15 May 2020 21:34:34 +0000 Greg Chicares <gchicares@sbcglobal.net> 
wrote:

GC> On 2020-05-15 17:57, Vadim Zeitlin wrote:
GC> > On Fri, 15 May 2020 17:18:07 +0000 Greg Chicares 
<gchicares@sbcglobal.net> wrote:
GC> > 
GC> > GC> To reproduce: build lmi with gcc_i686-w64-mingw32 version 9.3, then:
GC> [...]
GC> > GC> Segmentation fault
GC> > GC> make: [GNUmakefile:467: check_concinnity] Error 139 (ignored)
GC> > GC> 
GC> > GC> OTOH, it works (no segfault) with gcc-8.3 .
GC> 
GC> I awoke from my nap thinking that the cause might be something other
GC> than gcc--maybe instead it's wine, for example.

 Inspired by the recent "gcc performance regression" thread, I've decided
to finally test whether this could be indeed the case -- and it is. This is
definitely not a gcc problem, because the same binary which crashes with
this error under Wine (5.0-4) works fine under a native MSW system, so it
seems reasonable to suspect Wine here too.

 If this problem is still relevant (I know that you've found a workaround,
but it could still happen in some other place), I could build the debug
version of Wine 5 and try to find the problem in it -- please let me know
if I should do this.

 Thanks,
VZ

pgpe1YJLzXhLx.pgp
Description: PGP signature