AMDGPU: Skip unexpected CFG in SIOptimizeVGPRLiveRange

There are some cases that we use si_if/si_else in unatural way.
Just skip them.

Fixes: https://github.com/llvm/llvm-project/issues/55922

Revi

AMDGPU: Skip unexpected CFG in SIOptimizeVGPRLiveRange

There are some cases that we use si_if/si_else in unatural way.
Just skip them.

Fixes: https://github.com/llvm/llvm-project/issues/55922

Reviewed by: critson

Differential Revision: https://reviews.llvm.org/D128193

show more ...

[AMDGPU] Graceful abort for waterfalls in SIOptimizeVGPRLiveRange

If the CFG structure of a waterfall loop is not the expected shape
then gracefully abort traversing the IR for the given loop.
This

[AMDGPU] Graceful abort for waterfalls in SIOptimizeVGPRLiveRange

If the CFG structure of a waterfall loop is not the expected shape
then gracefully abort traversing the IR for the given loop.
This applies to nest waterfall loops which are not supported by
the VGPR live range optimizer.

Reviewed By: ruiling

Differential Revision: https://reviews.llvm.org/D123480

show more ...

[AMDGPU] Split waterfall loop exec manipulation

Split waterfall loops into multiple blocks so that exec mask
manipulation (s_and_saveexec) does not occur in the middle of
a block.

VGPR live range o

[AMDGPU] Split waterfall loop exec manipulation

Split waterfall loops into multiple blocks so that exec mask
manipulation (s_and_saveexec) does not occur in the middle of
a block.

VGPR live range optimizer is updated to handle waterfall loops
spanning multiple blocks.

Reviewed By: ruiling

Differential Revision: https://reviews.llvm.org/D122200

show more ...

AMDGPU: Fix LiveVariables error after optimizing VGPR ranges

This was not removing the block from the live set depending on the
specific depth first visit order. Fixes a verifier error in the OpenCL

AMDGPU: Fix LiveVariables error after optimizing VGPR ranges

This was not removing the block from the live set depending on the
specific depth first visit order. Fixes a verifier error in the OpenCL
conformance tests.

show more ...

AMDGPU: Clear NoPHIs property in SIOptimizeVGPRLiveRanges

Fixes verifier error when writing MIR tests that didn't have phis to
begin with.

[AMDGPU][NFC] Fix typos

Differential Revision: https://reviews.llvm.org/D113672

[AMDGPU] Improve killed check for vgpr optimization

The killed flag is not always set. E.g. when a variable is used in a
loop, it is never marked as killed, although it is unused in following
basic

[AMDGPU] Improve killed check for vgpr optimization

The killed flag is not always set. E.g. when a variable is used in a
loop, it is never marked as killed, although it is unused in following
basic blocks. Also, we try to deprecate kill flags and not use them.

Check if the register is live in the endif block. If not, consider it
killed in the then and else blocks.

The vgpr-liverange tests have two new tests with loops
(pre-committed, so the diff is visible).
I also needed to change the subtarget to gfx10.1, otherwise calls
are not working.

Differential Revision: https://reviews.llvm.org/D106291

show more ...

[AMDGPU] Optimize VGPR LiveRange in waterfall loops

The loops are run exactly once per lane, so VGPRs do not need to be
saved. Use the SIOptimizeVGPRLiveRange pass to add phi nodes that take
undef w

[AMDGPU] Optimize VGPR LiveRange in waterfall loops

The loops are run exactly once per lane, so VGPRs do not need to be
saved. Use the SIOptimizeVGPRLiveRange pass to add phi nodes that take
undef when coming from the loop.

There is still a shortcoming:
Return values from a function call in the loop are copied because their
live range conflicts with the live range of arguments, even if arguments
are only IMPLICIT_DEF after the phi insertion.

Differential Revision: https://reviews.llvm.org/D105192

show more ...

[AMDGPU] Fix typo

[NFC] Wrap entire assert-only block in LLVM_DEBUG

[AMDGPU] Add Optimize VGPR LiveRange Pass.

This pass aims to optimize VGPR live-range in a typical divergent if-else
control flow. For example:

def(a)
if(cond)
use(a)
... // A
else
use(a)

As

[AMDGPU] Add Optimize VGPR LiveRange Pass.

This pass aims to optimize VGPR live-range in a typical divergent if-else
control flow. For example:

def(a)
if(cond)
use(a)
... // A
else
use(a)

As AMDGPU access vgpr with respect to active-mask, we can mark `a` as
dead in region A. For details, please refer to the comments in
implementation file.

The pass is enabled by default, the frontend can disable it through
"-amdgpu-opt-vgpr-liverange=false".

Differential Revision: https://reviews.llvm.org/D102212

show more ...