diff options
| author | Jonathan Cohen <joncoh@apple.com> | 2025-07-17 19:21:57 +0300 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2025-07-17 19:21:57 +0300 |
| commit | e8a891b0f9d2a742ac3904116aaec2c7c9231b24 (patch) | |
| tree | eeacf9b6edae62c869a25bf2d55597bdbc63e612 /llvm/lib/IR/Module.cpp | |
| parent | a6fb3b3c18fd48a2eaaa8c969edbc013b9276a09 (diff) | |
| download | llvm-e8a891b0f9d2a742ac3904116aaec2c7c9231b24.zip llvm-e8a891b0f9d2a742ac3904116aaec2c7c9231b24.tar.gz llvm-e8a891b0f9d2a742ac3904116aaec2c7c9231b24.tar.bz2 | |
[AArch64][Machine-Combiner] Split gather patterns into neon regs to multiple vectors (#142941)
This changes optimizes gather-like sequences, where we load values
separately into lanes of a neon vector. Since each load has serial
dependency, when performing multiple i32 loads into a 128 bit vector for example, it
is more profitable to load into separate vector registers and zip them.
rdar://151851094
Diffstat (limited to 'llvm/lib/IR/Module.cpp')
0 files changed, 0 insertions, 0 deletions