[Bug middle-end/63311] [4.9/5 Regression] -O1 optimization introduces valgrind warning

[Joost.VandeVondele at mat dot ethz.ch]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63311

Joost VandeVondele <Joost.VandeVondele at mat dot ethz.ch> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |Joost.VandeVondele at mat dot ethz
                   |                            |.ch

--- Comment #5 from Joost VandeVondele <Joost.VandeVondele at mat dot ethz.ch> ---
(In reply to Jakub Jelinek from comment #3)
> Started with r208687 , so possibly a latent bug is no longer latent after
> the FE changes.

It also reproduces with a C instead of with a Fortran testcase (as derived from
the .original dump). To be attached in a second:

> gcc -O1 -std=c11 -g PR63311.c -lm && valgrind ./a.out
==36918== Conditional jump or move depends on uninitialised value(s)
==36918==    at 0x40056C: test (PR63311.c:41)
==36918==    by 0x40061F: main (PR63311.c:130)
==36918== ERROR SUMMARY: 4 errors from 1 contexts (suppressed: 6 from 6)

> gcc -O0 -std=c11 -g PR63311.c -lm && valgrind ./a.out
==36936== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 6 from 6)