[Bug c++/92617] Invalid loop optimization: no exit condition

2019-11-21 Thread pinskia at gcc dot gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92617 --- Comment #6 from Andrew Pinski --- In c++ (unlike C) return from a non void function is undefined even if the return value is not used. In c, only of the return value is used it would be undefined.

[Bug c++/92617] Invalid loop optimization: no exit condition

2019-11-21 Thread pinskia at gcc dot gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92617 --- Comment #7 from Andrew Pinski --- Not you can get a runtime error if you use -fsanitizer=undefined.

[Bug c++/92617] Invalid loop optimization: no exit condition

2019-11-21 Thread piotr.seweryn at gmail dot com
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92617 --- Comment #5 from Piotr Seweryn --- Obviously there is no return statement and warning was issued, however endless loop isn't the expected result, don't you agree? And falling-through to the next piece of code is even worse, coze we are falling

[Bug c++/92617] Invalid loop optimization: no exit condition

2019-11-21 Thread rguenth at gcc dot gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92617 Richard Biener changed: What|Removed |Added Status|UNCONFIRMED |RESOLVED Resolution|---

[Bug c++/92617] Invalid loop optimization: no exit condition

2019-11-21 Thread piotr.seweryn at gmail dot com
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92617 --- Comment #3 from Piotr Seweryn --- Similar problem occurs when -O2 or -O3 is used, there is also an endless loop, however different assembler code is generated. I can also provided appropriate .ii and .s files for O2 and O3 case.

[Bug c++/92617] Invalid loop optimization: no exit condition

2019-11-21 Thread piotr.seweryn at gmail dot com
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92617 --- Comment #2 from Piotr Seweryn --- Created attachment 47324 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=47324&action=edit Assembler file

[Bug c++/92617] Invalid loop optimization: no exit condition

2019-11-21 Thread piotr.seweryn at gmail dot com
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92617 --- Comment #1 from Piotr Seweryn --- Created attachment 47323 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=47323&action=edit Pre-processed file