6 Commits

Author SHA1 Message Date
Matt Arsenault
c7a0c2d0f7 AMDGPU: Report large stack usage for recursive calls
We were previously setting an ignored bit in the kernel headers. The
current behavior is to add the large amount on top of the statically
known size of a single stack frame. I'm not sure if we should just use
the large size as the entire reported size instead.
2021-11-10 20:02:01 -05:00
Anshil Gandhi
0567f03331 [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols
By default clang emits complete contructors as alias of base constructors if they are the same.
The backend is supposed to emit symbols for the alias, otherwise it causes undefined symbols.
@yaxunl observed that this issue is related to the llvm options `-amdgpu-early-inline-all=true`
and `-amdgpu-function-calls=false`. This issue is resolved by only inlining global values
with internal linkage. The `getCalleeFunction()` in AMDGPUResourceUsageAnalysis also had
to be extended to support aliases to functions. inline-calls.ll was corrected appropriately.

Reviewed By: yaxunl, #amdgpu

Differential Revision: https://reviews.llvm.org/D109707
2021-10-18 16:53:15 -06:00
Anshil Gandhi
1830ec94ac Revert "[HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols"
This reverts commit 03375a3fb33b11e1249d9c88070b7f33cb97802a.
2021-10-15 16:16:18 -06:00
Anshil Gandhi
03375a3fb3 [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols
By default clang emits complete contructors as alias of base constructors if they are the same.
The backend is supposed to emit symbols for the alias, otherwise it causes undefined symbols.
@yaxunl observed that this issue is related to the llvm options `-amdgpu-early-inline-all=true`
and `-amdgpu-function-calls=false`. This issue is resolved by only inlining global values
with internal linkage. The `getCalleeFunction()` in AMDGPUResourceUsageAnalysis also had
to be extended to support aliases to functions. inline-calls.ll was corrected appropriately.

Reviewed By: yaxunl, #amdgpu

Differential Revision: https://reviews.llvm.org/D109707
2021-10-15 11:39:15 -06:00
David Stuttard
69f7d81d0a [AMDGPU] Set number vgprs used in PS shaders based on input registers actually used
For PS shaders we can use the input SPI_PS_INPUT_ENA and SPI_PS_INPUT_ADDR
registers

Calculate the number of VGPR registers used as input VGPRs based on these
registers rather than the arguments passed in (this conservatively always
allocates the maximum).

Differential Revision: https://reviews.llvm.org/D101633

Change-Id: Idf7c060cbbd5f7e3300102c55ecee3c07f209de6
2021-10-08 14:24:35 +01:00
Sebastian Neubauer
2b08f6af62 [AMDGPU] Improve register computation for indirect calls
First, collect the register usage in each function, then apply the
maximum register usage of all functions to functions with indirect
calls.

This is more accurate than guessing the maximum register usage without
looking at the actual usage.

As before, assume that indirect calls will hit a function in the
current module.

Differential Revision: https://reviews.llvm.org/D105839
2021-07-20 13:48:50 +02:00