[PATCH] D31210: [AMDGPU] Add new address space mapping

Stanislav Mekhanoshin via Phabricator via cfe-commits cfe-commits at lists.llvm.org
Tue Mar 21 14:30:17 PDT 2017


rampitec added a comment.

I'm concerned about the default address space to be 64 bit. It would move alloca into generic address space effectively making private address to be 64 bit.
This may have very undesirable performance implications, like address arithmetic can become expensive 64 bit and only be truncated at load or store.
I realize you will use addrspacecast on an alloca's value, though I'm not sure that is sufficient to mitigate performance hit.
I believe such change shall not be made without a good performance comparison with the feature enabled, provided the very likely performance issues.


https://reviews.llvm.org/D31210





More information about the cfe-commits mailing list