5 Commits

Author SHA1 Message Date
Jennifer Yu
eace13928b Back out test that failed.
But I can not reproduce the problem on my local machine. My local machine run:

222 0x5a6780
222 0x7fffbef9400e
222 0x5a677e 0x5a6780 0x7fffbef936c8
222 0x376f8e 0x376f90 0x7fffbef94008
222 0x281f20
222 0x7fffbef9400e
PASSED
2022-11-04 17:23:05 -07:00
Jennifer Yu
de14befa77 Remove redundant loads.
It is caused by regenerate captured var value when processing the
has_device_addr, the captured var value has been generated in
GenerateOpenMPCapturedVars and passed as Arg in generateInfoForCapture.
The fix just use Arg instead regenerated just same as is_device_ptr
2022-11-04 15:22:25 -07:00
Jennifer Yu
48ffd40ba2 [Clang][OpenMP] Codegen generation for has_device_addr claues.
This patch add codegen support for the has_device_addr clause. It use
the same logic of is_device_ptr. But passing &var instead pointer to var
to kernal.

Differential Revision: https://reviews.llvm.org/D134268
2022-09-20 21:12:30 -07:00
Ron Lieberman
d5b5289561 revert 684f76643 [Clang][OpenMP] Codegen generation for has_device_addr claues.
breaks amdgpu buildbot
2022-09-20 01:37:27 +00:00
Jennifer Yu
684f766431 [Clang][OpenMP] Codegen generation for has_device_addr claues.
Summary: This patch add codegen support for the has_device_addr clause.  It
use the same logic of is_device_ptr.

Differential Revision: https://reviews.llvm.org/D134186
2022-09-19 16:14:57 -07:00