[Bug tree-optimization/107053] ones bits is not tracked and popcount is not tracked

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

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107053

Andrew Pinski  changed:

   What|Removed |Added

   Target Milestone|--- |14.0

[Bug tree-optimization/107053] ones bits is not tracked and popcount is not tracked

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

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107053

Aldy Hernandez  changed:

   What|Removed |Added

 Status|NEW |RESOLVED
 Resolution|--- |FIXED

--- Comment #5 from Aldy Hernandez  ---
fixed in trunk

[Bug tree-optimization/107053] ones bits is not tracked and popcount is not tracked

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

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107053

--- Comment #4 from CVS Commits  ---
The master branch has been updated by Aldy Hernandez :

https://gcc.gnu.org/g:137fb7077f7711e2e09ee8f82650fe7d93de6a4d

commit r14-2479-g137fb7077f7711e2e09ee8f82650fe7d93de6a4d
Author: Aldy Hernandez 
Date:   Fri Jun 30 20:40:02 2023 +0200

[range-op] Take known set bits into account in popcount [PR107053]

This patch teaches popcount about known set bits which are now
available in the irange.

PR tree-optimization/107053

gcc/ChangeLog:

* gimple-range-op.cc (cfn_popcount): Use known set bits.

gcc/testsuite/ChangeLog:

* gcc.dg/tree-ssa/pr107053.c: New test.

[Bug tree-optimization/107053] ones bits is not tracked and popcount is not tracked

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

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107053

Aldy Hernandez  changed:

   What|Removed |Added

   See Also||https://gcc.gnu.org/bugzill
   ||a/show_bug.cgi?id=107043
 CC||amacleod at redhat dot com

--- Comment #3 from Aldy Hernandez  ---
This is a variant of PR107043 which I discuss in comment #3.  We can't get this
because we don't track the or-mask, the known 1-bits.

[Bug tree-optimization/107053] ones bits is not tracked and popcount is not tracked

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

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107053

--- Comment #2 from Andrew Pinski  ---
Note the correct testcase is:
void link_failure();
void f(int a)
{
a |= 0x300;
int b =  __builtin_popcount(a);
if (b < 2)
link_failure();
}

--- CUT ---
the range of b should be 2..31 (assuming int is 32bits).
With this change clang is able to optimize it.

For the original testcase I think it can be folded to:
void link_failure();
void f(int a)
{
int a1 &= ~0x300;
if (a1 != 0)
link_failure();
}
but that might be worse unless popcount can be removed. (If I did this
correctly).

[Bug tree-optimization/107053] ones bits is not tracked and popcount is not tracked

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

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107053

Aldy Hernandez  changed:

   What|Removed |Added

 Status|UNCONFIRMED |NEW
 Ever confirmed|0   |1
   Last reconfirmed||2022-09-27

[Bug tree-optimization/107053] ones bits is not tracked and popcount is not tracked

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

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107053

--- Comment #1 from Andrew Pinski  ---
This testcase is not optimized by clang/llvm.