[Bug middle-end/93505] [8/9 Regression] wrong code or ICE with __builtin_bswap64() and rotation at -Og

2020-02-13 Thread cvs-commit at gcc dot gnu.org 
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93505

--- Comment #19 from CVS Commits  ---
The releases/gcc-9 branch has been updated by Jakub Jelinek
:

https://gcc.gnu.org/g:329475795c6eeaa2b122672091c9119b9d6c5564

commit r9-8217-g329475795c6eeaa2b122672091c9119b9d6c5564
Author: Jakub Jelinek 
Date:   Thu Jan 30 21:28:17 2020 +0100

combine: Punt on out of range rotate counts [PR93505]

What happens on this testcase is with the out of bounds rotate we get:
Trying 13 -> 16:
   13: r129:SI=r132:DI#0<-<0x20
  REG_DEAD r132:DI
   16: r123:DI=r129:SI<0
  REG_DEAD r129:SI
Successfully matched this instruction:
(set (reg/v:DI 123 [  ])
(const_int 0 [0]))
during combine.  So, perhaps we could also change simplify-rtx.c to punt
if it is out of bounds rather than trying to optimize anything.
Or, but probably GCC11 material, if we decide that ROTATE/ROTATERT doesn't
have out of bounds counts or introduce targetm.rotate_truncation_mask,
we should truncate the argument instead of punting.
Punting is better for backports though.

2020-01-30  Jakub Jelinek  

PR middle-end/93505
* combine.c (simplify_comparison) : Punt on out of range
rotate counts.

* gcc.c-torture/compile/pr93505.c: New test.

[Bug middle-end/93505] [8/9 Regression] wrong code or ICE with __builtin_bswap64() and rotation at -Og

2020-02-03 Thread segher at gcc dot gnu.org 
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93505

--- Comment #18 from Segher Boessenkool  ---
PR93512 is marked as enhancement, but if we don't fix this it is a regression.

[Bug middle-end/93505] [8/9 Regression] wrong code or ICE with __builtin_bswap64() and rotation at -Og

2020-01-31 Thread jakub at gcc dot gnu.org 
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93505

Jakub Jelinek  changed:

   What|Removed |Added

   Priority|P1  |P2

[Bug middle-end/93505] [8/9 Regression] wrong code or ICE with __builtin_bswap64() and rotation at -Og

2020-01-31 Thread jakub at gcc dot gnu.org 
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93505

Jakub Jelinek  changed:

   What|Removed |Added

Summary|[8/9/10 Regression] wrong   |[8/9 Regression] wrong code
   |code or ICE with|or ICE with
   |__builtin_bswap64() and |__builtin_bswap64() and
   |rotation at -Og |rotation at -Og

--- Comment #17 from Jakub Jelinek  ---
The wrong-code is now fixed on the trunk.  What we want to do further depends
on whether we want to make ROTATE/ROTATERT imply truncation (always or only on
selected targets), let's defer that part to PR93512.