diff options
| author | Elvis Wang <elvis.wang@sifive.com> | 2025-09-02 11:43:27 +0800 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2025-09-02 11:43:27 +0800 |
| commit | 8fdae0c7daf3831c48be4c69e799da627337bddd (patch) | |
| tree | 2465b8b71811493b5253b67b4176d552bda5b387 /llvm/test/CodeGen/DirectX | |
| parent | 0849dc11d9844a4df5f18c46831da3859e3ae4c1 (diff) | |
| download | llvm-8fdae0c7daf3831c48be4c69e799da627337bddd.zip llvm-8fdae0c7daf3831c48be4c69e799da627337bddd.tar.gz llvm-8fdae0c7daf3831c48be4c69e799da627337bddd.tar.bz2 | |
[Reland] "[RISCV][TTI] Implement getAddressComputationCost() in RISCV TTI. #149955" (#156386)
This patch implements the `getAddressComputationCost()` in RISCV TTI
which
make the gather/scatter with address calculation more expansive that
stride cost.
Note that the only user of `getAddressComputationCost()` with vector
type is in `VPWidenMemoryRecipe::computeCost()`. So this patch make some
LV tests changes.
I've checked the tests changes in LV and seems those changes can be
divided into two groups.
* gather/scatter with uniform vector ptr, seems can be optimized to
masked.load.
* can optimize to stride load/store.
----
After #155739 landed, the assertion (cost mis-aligned) is fixed.
I've tested llvm-test-suite w/ rva23u64 and rva23u64_zvl1024b locally
and no assertion occurred.
Diffstat (limited to 'llvm/test/CodeGen/DirectX')
0 files changed, 0 insertions, 0 deletions