1
0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-10-23 13:02:52 +02:00
llvm-mirror/test/CodeGen/AMDGPU/elf.ll
Marek Olsak e93050ba9d AMDGPU/SI: Increase SGPR limit to 96 on Tonga/Iceland
Summary:
This is the setting of the Vulkan closed source driver.

It decreases the max wave count from 10 to 8.

26010 shaders in 14650 tests
Totals:
VGPRS: 829593 -> 808440 (-2.55 %)
Spilled SGPRs: 81878 -> 42226 (-48.43 %)
Spilled VGPRs: 367 -> 358 (-2.45 %)
Scratch VGPRs: 1764 -> 1748 (-0.91 %) dwords per thread
Code Size: 36677864 -> 35923932 (-2.06 %) bytes

There is a massive decrease in SGPR spilling in general and -7.4% spilled
VGPRs for DiRT Showdown (= SGPRs spilled to scratch?)

Reviewers: arsenm, tstellarAMD, nhaehnle

Subscribers: arsenm, llvm-commits, kzhuravl

Differential Revision: https://reviews.llvm.org/D23034

llvm-svn: 277867
2016-08-05 21:23:29 +00:00

35 lines
1.7 KiB
LLVM

; RUN: llc < %s -march=amdgcn -mcpu=SI -verify-machineinstrs -filetype=obj | llvm-readobj -s -symbols -file-headers - | FileCheck --check-prefix=ELF %s
; RUN: llc < %s -march=amdgcn -mcpu=SI -verify-machineinstrs -o - | FileCheck --check-prefix=CONFIG --check-prefix=TYPICAL %s
; RUN: llc < %s -march=amdgcn -mcpu=tonga -verify-machineinstrs -filetype=obj | llvm-readobj -s -symbols -file-headers - | FileCheck --check-prefix=ELF %s
; RUN: llc < %s -march=amdgcn -mcpu=tonga -verify-machineinstrs -o - | FileCheck --check-prefix=CONFIG --check-prefix=TONGA %s
; RUN: llc < %s -march=amdgcn -mcpu=carrizo -verify-machineinstrs -filetype=obj | llvm-readobj -s -symbols -file-headers - | FileCheck --check-prefix=ELF %s
; RUN: llc < %s -march=amdgcn -mcpu=carrizo -verify-machineinstrs -o - | FileCheck --check-prefix=CONFIG --check-prefix=TYPICAL %s
; Test that we don't try to produce a COFF file on windows
; RUN: llc < %s -mtriple=amdgcn-pc-mingw -mcpu=SI -verify-machineinstrs -filetype=obj | llvm-readobj -s -symbols -file-headers - | FileCheck --check-prefix=ELF %s
; ELF: Format: ELF64
; ELF: OS/ABI: AMDGPU_HSA (0x40)
; ELF: Machine: EM_AMDGPU (0xE0)
; ELF: Name: .AMDGPU.config
; ELF: Type: SHT_PROGBITS
; ELF: Symbol {
; ELF: Name: test
; ELF: Binding: Global
; CONFIG: .section .AMDGPU.config
; CONFIG-NEXT: .long 45096
; TYPICAL-NEXT: .long 0
; TONGA-NEXT: .long 704
; CONFIG: .p2align 8
; CONFIG: test:
define amdgpu_ps void @test(i32 %p) {
%i = add i32 %p, 2
%r = bitcast i32 %i to float
call void @llvm.SI.export(i32 15, i32 0, i32 1, i32 12, i32 0, float %r, float %r, float %r, float %r)
ret void
}
declare void @llvm.SI.export(i32, i32, i32, i32, i32, float, float, float, float)