Removed redundant parenses.

Addressed comment.

Ping...

Addressed comment, removed few more getValue() calls.

Ping...

Ping...

Rebased...

Added comment/description for getTypeConversion()

Addressed comment.

Rebased.
I thought we should wait for additional opinions on the patch...

Ping...

Addresses comments.

Moved warning skipping into checkLivenessAtDef() to get back some checks.

In D126661#3691962, @MatzeB wrote:

I guess Jay already posted the liveness info. It seems this is related to CalcLiveRangeUtilBase::createDeadDef moving definitions to the early clobber slot when there's an early-clobber and a normal def. Judging by the longer explanation in that function this is probably deliberate and we have to blame the MachineVerifier here. But that means you should fix the machine verifier and not just disable a whole subset of tests!

Ping - the patch with D127136 are fixing https://github.com/RadeonOpenCompute/ROCm/issues/1486...

Using pipeliner instead of RC as dummy pass.

With D129208 + D130477 I get:

Failed Tests (3):
  LLVM :: CodeGen/AMDGPU/optimize-negated-cond-exec-masking-wave32.mir
  LLVM :: CodeGen/AMDGPU/optimize-negated-cond-exec-masking.mir
  LLVM :: CodeGen/Hexagon/expand-condsets-def-undef.mir

Yes, I was too optimistic: the patch fixes "Multiple connected components in live interval" issues in AMDGPU/* tests but reveals next issue related to incorrect updates of ranges.
Regarding Hexagon/expand-condsets-def-undef.mir test issue - it is steel about multiple connected components.

Gentle ping...

@MatzeB, @qcolombet, would you please have a look?

LiveIntervals dumps for the test, before the patch:
Before first move:

%1 [16r,32r:0) 0@16r  L0000000000000003 [16r,32r:0) 0@16r  L000000000000000C [16r,32r:0) 0@16r  weight:0.000000e+00
%2 [32r,32d:0) 0@32r  weight:0.000000e+00
%3 [48r,64r:0)[64r,80r:1) 0@48r 1@64r  L0000000000000003 [48r,80r:0) 0@48r  L000000000000000C [64r,80r:0) 0@64r  weight:0.000000e+00
%4 [80r,80d:0) 0@80r  weight:0.000000e+00
RegMasks:
********** MACHINEINSTRS **********
# Machine code for function func: NoPHIs

Jay, you are completely correct. Sorry I didn't mention the same info as in dependent review. Updated the description.

Beautified test.

Beautified test.

Addressed comments.

ping...

It seems to me Cost::RateFormula() form LSR is the only user that can be affected by the change. Would you please double-look the use case?

Fixed SLOW/FAST misplacement in test

Updated test as requested.

Needs a test.

Ping...

Thanks!

In D110100#3014761, @RKSimon wrote:

In D110100#3014729, @samparker wrote:

Do you think we should stop using default cost kinds entirely?

I think this would be for the best, and I thought I already changed most of the calls to be explicit.

@dfukalov Is this something you'd be willing to do in this patch? At least for getInstructionCost + getInterleavedMemoryOpCost (TTI + Impl)?

Fixed all clang-tidy warnings except "invalid case style".

Addressed comment.