In D143814#4120672, @mehdi_amini wrote:

Thanks for the update!

Do you need someone to land this or do you have commit access?

Address comments.

Rebase.

• Add comment for MUL.
• Add MIN(U)/MAX(U).

In D140530#4016511, @HsiangKai wrote:

In D140530#4015004, @asi-sc wrote:

Have you had a chance to make some performance measurements?

Quoted from D138107,

I run C/C++ benchmarks in SPECrate 2017 on Fujitsu A64FX processor, which has two pipelines for integer operations and SIMD/FP operations each. 511.povray_r had 4% improvement. Other benchmarks (int: 500, 502, 505, 520, 523, 525, 531, 541, 557; fp: 508, 510, 519, 538, 544) were within 1% up/down. For a synthetic benchmark, it doubled the performance.

I have no performance number for RISC-V multiple issue machines. I am not sure what the impact is of the patch. Is there anyone can help to measure it?

Remove unnecessary fast-math flags in the test cases.

In D140530#4015004, @asi-sc wrote:

Have you had a chance to make some performance measurements?

No need to check RmReassoc and FmNsz for fmin/fmax.

clang format

• Move FRM operand checking to hasReassociableSibling.
• Add ADD, SUB, ADDW, SUBW to getInverseOpcode and add test cases for the inverse cases.

Remove MULH, MULHU from the list.

clang format

Add default case for the switch statement.

Out of date.

In D132536#3771057, @aemerson wrote:

Are there existing IR tests that we can add run lines to?

Make the test lines consistent.

Address comments.

Address comments.

Revert the change and modify the comments only.

Add predicate for (uaddo X, C).

Take the return value of UpdateNodeOperands().

Update failed PowerPC test cases.

Move new test cases to the end of the test file to remove unnecessary noises.

Add test cases.

Update test cases.

It is not safe to do this peephole for ADD*W.

Update ProgrammersManual.rst according to @dexonsmith's suggestions.

Address dexonsmith's comments.