mirror of
https://github.com/llvm/llvm-project.git
synced 2025-04-27 01:16:05 +00:00
f93aa5157a
These generic targets include multiple GPUs and will, in the future, provide a way to build once and run on multiple GPU, at the cost of less optimization opportunities. Note that this is just doing the compiler side of things, device libs an runtimes/loader/etc. don't know about these targets yet, so none of them actually work in practice right now. This is just the initial commit to make LLVM aware of them. This contains the documentation changes for both this change and #76954 as well.
19 lines
1.2 KiB
LLVM
19 lines
1.2 KiB
LLVM
; RUN: not llc -march=amdgcn -mcpu=gfx9-generic --amdhsa-code-object-version=5 -o - %s 2>&1 | FileCheck --check-prefix=GFX9-V5 %s
|
|
; RUN: not llc -march=amdgcn -mcpu=gfx10.1-generic --amdhsa-code-object-version=5 -o - %s 2>&1 | FileCheck --check-prefix=GFX101-V5 %s
|
|
; RUN: not llc -march=amdgcn -mcpu=gfx10.3-generic --amdhsa-code-object-version=5 -o - %s 2>&1 | FileCheck --check-prefix=GFX103-V5 %s
|
|
; RUN: not llc -march=amdgcn -mcpu=gfx11-generic --amdhsa-code-object-version=5 -o - %s 2>&1 | FileCheck --check-prefix=GFX11-V5 %s
|
|
|
|
; RUN: llc -march=amdgcn -mcpu=gfx9-generic --amdhsa-code-object-version=6 -o - %s
|
|
; RUN: llc -march=amdgcn -mcpu=gfx10.1-generic --amdhsa-code-object-version=6 -o - %s
|
|
; RUN: llc -march=amdgcn -mcpu=gfx10.3-generic --amdhsa-code-object-version=6 -o - %s
|
|
; RUN: llc -march=amdgcn -mcpu=gfx11-generic --amdhsa-code-object-version=6 -o - %s
|
|
|
|
; GFX9-V5: gfx9-generic is only available on code object version 6 or better
|
|
; GFX101-V5: gfx10.1-generic is only available on code object version 6 or better
|
|
; GFX103-V5: gfx10.3-generic is only available on code object version 6 or better
|
|
; GFX11-V5: gfx11-generic is only available on code object version 6 or better
|
|
|
|
define void @foo() {
|
|
ret void
|
|
}
|