[Bug c/100505] ICE: in record_in_finally_tree, at tree-eh.c:213

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100505

Andrew Pinski  changed:

   What|Removed |Added

 Resolution|--- |DUPLICATE
 Status|UNCONFIRMED |RESOLVED

--- Comment #4 from Andrew Pinski  ---
Dup of bug 100625. Even though PR 100625 is newer, it records which revision
fixed this bug too.

*** This bug has been marked as a duplicate of bug 100625 ***

[Bug c/100505] ICE: in record_in_finally_tree, at tree-eh.c:213

[rguenth at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100505

--- Comment #3 from Richard Biener  ---
(In reply to Chengnian Sun from comment #2)
> Should I avoid reporting test programs with __GIMPLE or __RTL?

I think so, at least if you're just mass-generating them somehow.  I think
it's OK to report ICEs without prior reported errors - those we'd
eventually want to fix.  So for __GIMPLE at least make sure to pass -fgimple
to get rid of the first reported error.

[Bug c/100505] ICE: in record_in_finally_tree, at tree-eh.c:213

[cnsun at uwaterloo dot ca via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100505

--- Comment #2 from Chengnian Sun  ---
Should I avoid reporting test programs with __GIMPLE or __RTL?

[Bug c/100505] ICE: in record_in_finally_tree, at tree-eh.c:213

[jakub at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100505

Jakub Jelinek  changed:

   What|Removed |Added

 CC||jakub at gcc dot gnu.org

--- Comment #1 from Jakub Jelinek  ---
__GIMPLE and __RTL expects correct input and handles just limited diagnostics.
I think especially error-recovery for these are WONTFIX.