clang/OpenCL: Fix not setting convergent on block invoke kernels
authorMatt Arsenault <Matthew.Arsenault@amd.com>
Wed, 11 Jan 2023 01:08:00 +0000 (20:08 -0500)
committerMatt Arsenault <arsenm2@gmail.com>
Mon, 30 Jan 2023 19:03:14 +0000 (15:03 -0400)
commit00f6a7f02f9c8d542ce8ff1c9c037d9fdb421b88
tree7704ac5aba0af7bae7b0ecf085603dd2a90bcf5b
parent52c28d7cf9018f3558268217f6cf91271eb4ac39
clang/OpenCL: Fix not setting convergent on block invoke kernels

Yet another example how convergent not being the default is dangerous
and backwards.
clang/lib/CodeGen/TargetInfo.cpp
clang/test/CodeGenOpenCL/amdgpu-enqueue-kernel.cl
clang/test/CodeGenOpenCL/cl20-device-side-enqueue.cl