Commit 0010a3c
authored
[NFC][LoopVectorize] Add more partial reduction tests (#126525)
* Adds variants of dotp (dotp_i8_to_i64_has_neon_dotprod,
dotp_i16_to_i64_has_neon_dotprod) that show how the loop
vectoriser has generated fixed-width partial reductions
without any matching NEON udot instruction.
* Adds loops that could also benefit from partial
reductions once the work is done to recognise patterns
such as
%zext = zext i8 %load to i32
%acc.next = add i32 %acc, %zext
See zext_add_reduc_i8_i32, etc. I intend to follow up with
a patch to add support for vectorising such patterns.1 parent 1c583c1 commit 0010a3c
File tree
1 file changed
+1364
-97
lines changed- llvm/test/Transforms/LoopVectorize/AArch64
1 file changed
+1364
-97
lines changed
0 commit comments