diff options
| author | David Green <david.green@arm.com> | 2020-09-12 13:51:42 +0100 |
|---|---|---|
| committer | David Green <david.green@arm.com> | 2020-09-12 13:51:42 +0100 |
| commit | c437446d90be17c3fe8a216a90ee442222f2fe9d (patch) | |
| tree | 5cde0bb35c20fd90dda5c85a5446fe0f37270d55 /clang/lib/CodeGen/CodeGenFunction.cpp | |
| parent | 36e2e2e12efb6b02ad07f502d61b9a95937edb08 (diff) | |
| download | llvm-c437446d90be17c3fe8a216a90ee442222f2fe9d.zip llvm-c437446d90be17c3fe8a216a90ee442222f2fe9d.tar.gz llvm-c437446d90be17c3fe8a216a90ee442222f2fe9d.tar.bz2 | |
[ARM] Recognize "double extend" reduction patterns
We can sometimes get code that does:
xe = zext i16 x to i32
ye = zext i16 y to i32
m = mul i32 xe, ye
me = zext i32 m to i64
r = vecreduce.add(me)
This "double extend" can trip up the reduction identification, but
should give identical results.
This extends the pattern matching to handle them.
Differential Revision: https://reviews.llvm.org/D87276
Diffstat (limited to 'clang/lib/CodeGen/CodeGenFunction.cpp')
0 files changed, 0 insertions, 0 deletions