[PATCH] D31210: [AMDGPU] Add new address space mapping
Yaxun Liu via Phabricator via cfe-commits
cfe-commits at lists.llvm.org
Wed Mar 22 12:16:12 PDT 2017
yaxunl added inline comments.
================
Comment at: lib/Basic/Targets.cpp:2029-2040
+ if (UseNew) {
+ Generic = 0;
+ Global = 1;
+ Local = 3;
+ Constant = 4;
+ Private = 5;
+ } else {
----------------
tstellar wrote:
> What are these values used for?
The new address space mapping will be used for all kernel languages, especially for C++-based kernel languages like HCC/OpenMP/CUDA.
We found the old address space mapping is unable to support C++ based kernel languages, that's the motivation of this change.
Eventually we will switch to the new address space mapping.
We temporarily keep two address space mapping during the transition,
================
Comment at: lib/Basic/Targets.cpp:2057
+ 4, // opencl_constant
+ 0, // opencl_generic
+ 1, // cuda_device
----------------
tstellar wrote:
> How will the backend deal with the fact that allocas return generic address space pointers?
Matt will upstream a patch which let alloca return a pointer to private address space.
https://reviews.llvm.org/D31210
More information about the cfe-commits
mailing list