| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
|
|
|
| |
Summary:
We should not change volatile loads/stores in promoting alloca to vector.
Reviewers:
arsenm
Differential Revision:
http://reviews.llvm.org/D33107
llvm-svn: 302943
|
|
|
|
|
|
|
|
|
|
|
|
| |
Currently the default C calling convention functions are treated
the same as compute kernels. Make this explicit so the default
calling convention can be changed to a non-kernel.
Converted with perl -pi -e 's/define void/define amdgpu_kernel void/'
on the relevant test directories (and undoing in one place that actually
wanted a non-kernel).
llvm-svn: 298444
|
|
|
|
|
|
| |
The colon is important.
llvm-svn: 292761
|
|
|
|
|
|
|
| |
If the load has a pointer type, we don't want to change
its type.
llvm-svn: 270000
|
|
llvm-svn: 264214
|