[all-commits] [llvm/llvm-project] 5f6fec: AMDGPU: Fix handling of infinite loops in fragment...

[Connor Abbott via All-commits]

  Branch: refs/heads/release/10.x
  Home:   https://github.com/llvm/llvm-project
  Commit: 5f6fec2404c5135247ae9e4e515e8d9d3242f790
      https://github.com/llvm/llvm-project/commit/5f6fec2404c5135247ae9e4e515e8d9d3242f790
  Author: Connor Abbott <cwabbott0 at gmail.com>
  Date:   2020-02-04 (Tue, 04 Feb 2020)

  Changed paths:
    M llvm/lib/Target/AMDGPU/AMDGPUUnifyDivergentExitNodes.cpp
    A llvm/test/CodeGen/AMDGPU/kill-infinite-loop.ll

  Log Message:
  -----------
  AMDGPU: Fix handling of infinite loops in fragment shaders

Summary:
Due to the fact that kill is just a normal intrinsic, even though it's
supposed to terminate the thread, we can end up with provably infinite
loops that are actually supposed to end successfully. The
AMDGPUUnifyDivergentExitNodes pass breaks up these loops, but because
there's no obvious place to make the loop branch to, it just makes it
return immediately, which skips the exports that are supposed to happen
at the end and hangs the GPU if all the threads end up being killed.

While it would be nice if the fact that kill terminates the thread were
modeled in the IR, I think that the structurizer as-is would make a mess if we
did that when the kill is inside control flow. For now, we just add a null
export at the end to make sure that it always exports something, which fixes
the immediate problem without penalizing the more common case. This means that
we sometimes do two "done" exports when only some of the threads enter the
discard loop, but from tests the hardware seems ok with that.

This fixes dEQP-VK.graphicsfuzz.while-inside-switch with radv.

Reviewers: arsenm, nhaehnle

Subscribers: kzhuravl, jvesely, wdng, yaxunl, dstuttard, tpr, t-tye, hiraditya, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D70781

(cherry picked from commit 87d98c149504f9b0751189744472d7cc94883960)