[Bug rtl-optimization/80025] [5/6/7 Regression] ICE w/ -O2 (-O3, -Ofast) -g -ftracer (infinite recursion in rtx_equal_for_cselib_1)

["asolokha at gmx dot com" <gcc-bugzilla at gcc dot gnu dot org>]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80025

--- Comment #6 from Arseny Solokha <asolokha at gmx dot com> ---
(In reply to Martin Liška from comment #5)
> That should gently report stack
> overflow?

Exactly.

<trimmed>

==3238== Stack overflow in thread #1: can't grow stack to 0xffe001000
==3238== Can't extend stack to 0xffe000fd8 during signal delivery for thread 1:
==3238==   no stack segment
==3238== 
==3238== Process terminating with default action of signal 11 (SIGSEGV):
dumping core
==3238==  Access not within mapped region at address 0xFFE000FD8
==3238== Stack overflow in thread #1: can't grow stack to 0xffe001000
==3238==    at 0x768549: rtx_equal_for_cselib_1(rtx_def*, rtx_def*,
machine_mode) (in /usr/libexec/gcc/x86_64-pc-linux-gnu/7.0.0-alpha20170312/cc1)
==3238==  If you believe this happened as a result of a stack
==3238==  overflow in your program's main thread (unlikely but
==3238==  possible), you can try to increase the size of the
==3238==  main thread stack using the --main-stacksize= flag.
==3238==  The main thread stack size used in this run was 16777216.
==3238== Stack overflow in thread #1: can't grow stack to 0xffe001000

<trimmed>