[PATCH] D104997: [WIP][AMDGPU] Deduce attributes with the Attributor

[Johannes Doerfert via Phabricator via llvm-commits]

jdoerfert added inline comments.


================
Comment at: llvm/lib/Target/AMDGPU/AMDGPUAttributor.cpp:27-31
+    "amdgpu-work-item-id-x",  "amdgpu-work-item-id-y",
+    "amdgpu-work-item-id-z",  "amdgpu-work-group-id-x",
+    "amdgpu-work-group-id-y", "amdgpu-work-group-id-z",
+    "amdgpu-dispatch-ptr",    "amdgpu-dispatch-id",
+    "amdgpu-queue-ptr",       "amdgpu-implicitarg-ptr"};
----------------
arsenm wrote:
> jdoerfert wrote:
> > arsenm wrote:
> > > These are the attributes as they exist now, but I do think these need to be inverted to be more sound. Assuming they are present is the conservative direction, so ideally we would operate on a no-* basis
> > I'd recommend to do that in a follow up as this is tested on it's own first, wdyt?
> Sure, but I would assume inference works naturally in the negative direction?
yes, it should. Not sure I would say the direction of the inference is different but the lattice we use for the state is reversed, still pulling information into kernels (transitively).


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D104997/new/

https://reviews.llvm.org/D104997