While looking at PR 44608 and finding it was already fixed, I noticed instead we should produce ABSU at the gimple level but don't currently. Take: unsigned abssat2 (int x) { unsigned int y = x; if (x < 0) y = -y; return y; }
Confirmed.
Mine, let me try to figure out the best way to fix this.
(In reply to Andrew Pinski from comment #2) > Mine, let me try to figure out the best way to fix this. Basically the patterns starting with: /* abs/negative simplifications moved from fold_cond_expr_with_comparison, Need to handle (A - B) case as fold_cond_expr_with_comparison does. Need to handle UN* comparisons. need to add nop_convertN? handling. There is 6 patterns.
Another one missed: ``` unsigned abssat2 (unsigned x) { int y = x; if (y < 0) x = -x; return x; } ``` I have a patch for the this case, I think. Working on the other case now.
Actually using bitwise_equal_p makes this easier and no need for the nop_convert ... I have a patch where I just need to add testcases too.
Created attachment 56116 [details] Patch which I am testing
Created attachment 56118 [details] Patch which I am testing which allows for more (independent of the other) This patch improves the case where we have a conversion at the end (after the negate expression). And with both of these patches together we are able to catch almost all. There is one more phi-opt patch needed to catch the case where the conversion before the negate expression is conditional but that is for another day.
Patch: https://gcc.gnu.org/pipermail/gcc-patches/2023-October/633043.html
The trunk branch has been updated by Andrew Pinski <pinskia@gcc.gnu.org>: https://gcc.gnu.org/g:c7609acb8a8210188d21b2cd72ecc6d3b2de2ab8 commit r14-4662-gc7609acb8a8210188d21b2cd72ecc6d3b2de2ab8 Author: Andrew Pinski <pinskia@gmail.com> Date: Sun Oct 15 10:36:56 2023 -0700 MATCH: Improve `A CMP 0 ? A : -A` set of patterns to use bitwise_equal_p. This improves the `A CMP 0 ? A : -A` set of match patterns to use bitwise_equal_p which allows an nop cast between signed and unsigned. This allows catching a few extra cases which were not being caught before. OK? Bootstrapped and tested on x86_64-linux-gnu with no regressions. gcc/ChangeLog: PR tree-optimization/101541 * match.pd (A CMP 0 ? A : -A): Improve using bitwise_equal_p. gcc/testsuite/ChangeLog: PR tree-optimization/101541 * gcc.dg/tree-ssa/phi-opt-36.c: New test. * gcc.dg/tree-ssa/phi-opt-37.c: New test.
Fixed.
The trunk branch has been updated by Andrew Pinski <pinskia@gcc.gnu.org>: https://gcc.gnu.org/g:0fc13e8c0e39c51e82deb93f324d9d86ad8d7460 commit r14-4889-g0fc13e8c0e39c51e82deb93f324d9d86ad8d7460 Author: Andrew Pinski <pinskia@gmail.com> Date: Sun Oct 15 19:15:38 2023 +0000 Improve factor_out_conditional_operation for conversions and constants In the case of a NOP conversion (precisions of the 2 types are equal), factoring out the conversion can be done even if int_fits_type_p returns false and even when the conversion is defined by a statement inside the conditional. Since it is a NOP conversion there is no zero/sign extending happening which is why it is ok to be done here; we were trying to prevent an extra sign/zero extend from being moved away from definition which no-op conversions are not. Bootstrapped and tested on x86_64-linux-gnu with no regressions. gcc/ChangeLog: PR tree-optimization/104376 PR tree-optimization/101541 * tree-ssa-phiopt.cc (factor_out_conditional_operation): Allow nop conversions even if it is defined by a statement inside the conditional. gcc/testsuite/ChangeLog: PR tree-optimization/101541 * gcc.dg/tree-ssa/phi-opt-39.c: New test.