This is an archive of the discontinued LLVM Phabricator instance.

[mlir] Make GpuAsyncRegion pass depend on async dialect.
ClosedPublic

Authored by csigg on Jan 11 2021, 3:20 AM.

Details

Summary

Do not cache gpu.async.token type so that the pass can be created before the GPU dialect is registered.

Diff Detail

Event Timeline

csigg created this revision.Jan 11 2021, 3:20 AM
csigg requested review of this revision.Jan 11 2021, 3:20 AM
csigg retitled this revision from Make GpuAsyncRegion pass depend on async dialect. to [mlir] Make GpuAsyncRegion pass depend on async dialect..Jan 11 2021, 4:54 AM
ftynse accepted this revision.Jan 11 2021, 4:56 AM
This revision is now accepted and ready to land.Jan 11 2021, 4:56 AM
This revision was automatically updated to reflect the committed changes.