Page MenuHomePhabricator

[OPENMP][NVPTX]Fixed initialization of the data-sharing interface.
ClosedPublic

Authored by ABataev on Dec 27 2018, 6:54 AM.

Details

Summary

Avoid using of the atomic loop to wait for the completion of the
data-sharing interface initialization, use __shfl_sync instead for the
communication within the warp to signal other threads in the warp about
completion of the initialization.

Diff Detail

Repository
rL LLVM

Event Timeline

ABataev created this revision.Dec 27 2018, 6:54 AM
grokos added inline comments.Dec 27 2018, 8:11 AM
libomptarget/deviceRTLs/nvptx/src/data_sharing.cu
449 ↗(On Diff #179549)

Isn't FrameP always 8-bytes long?

ABataev marked an inline comment as done.Dec 27 2018, 8:12 AM
ABataev added inline comments.
libomptarget/deviceRTLs/nvptx/src/data_sharing.cu
449 ↗(On Diff #179549)

What if the architecture is 32 bits?

grokos accepted this revision.Dec 27 2018, 4:19 PM
grokos added inline comments.
libomptarget/deviceRTLs/nvptx/src/data_sharing.cu
449 ↗(On Diff #179549)

That would be true if we compiled the RTL on a 32-bit host system, but - at least at the moment - only 64-bit hosts are supported by libomptarget (x86_64, ppc64, aarch64). So the 32-bit case is unreachable, although admittedly it doesn't hurt to have this discrimination here as a safeguard for the future. On the other hand, how probable is it that we will ever support 32-bit systems? Anyway, the patch looks good, I'm leaving it up to you to decide whether the 32-bit scenario should be considered here.

This revision is now accepted and ready to land.Dec 27 2018, 4:19 PM
This revision was automatically updated to reflect the committed changes.