projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: d41b3bf) | patch
[SCEV] Use umin_seq for symbolic max BE count
authorNikita Popov <npopov@redhat.com>
Wed, 7 Dec 2022 14:31:17 +0000 (15:31 +0100)
committerNikita Popov <npopov@redhat.com>
Wed, 7 Dec 2022 14:32:49 +0000 (15:32 +0100)
commit4d97a914d7aef6c95cd95669f99e5d32899e165a
tree659f5643d8ec3b847b85781e3f4395e1d20069edtree | snapshot
parentd41b3bf7c37d36f02e55662c31c14aecdcdf9b9bcommit | diff
[SCEV] Use umin_seq for symbolic max BE count

We were using umin_seq when computing the exact BE count, but not
when computing the symbolic max BE count.
llvm/lib/Analysis/ScalarEvolution.cpp diff | blob | history
llvm/test/Analysis/ScalarEvolution/symbolic_max_exit_count.ll diff | blob | history
Domain: System / Toolchain;