Manuel Drehwald
6de0591c0b
Split ol mapper into more specific to/kernel/from mapper and move init_all_rtls into global ctor
2026-02-07 17:34:39 -08:00
Marcelo Domínguez
212c8c3811
Remove dummy loads
2026-02-04 15:26:56 +01:00
Manuel Drehwald
35ce8ab120
adjust testcase for new logic
2026-01-27 10:43:21 -08:00
Stuart Cook
1262ff906b
Rollup merge of #150288 - offload-bench-fix, r=ZuseZ4
...
Add scalar support for offload
This PR adds scalar support to the offload feature. The scalar management has two main parts:
On the host side, each scalar arg is casted to `ix` type, zero extended to `i64` and passed to the kernel like that.
On the device, the each scalar arg (`i64` at that point), is truncated to `ix` and then casted to the original type.
r? @ZuseZ4
2026-01-20 18:00:08 +11:00
Marcelo Domínguez
307a4fcdf8
Add scalar support for both host and device
2026-01-19 22:28:42 +01:00
Manuel Drehwald
5c85d522d0
Generate global openmp metadata to trigger llvm openmp-opt pass
2026-01-16 14:57:32 -05:00
Manuel Drehwald
fa584faca5
Update test and verify that tgt_(un)register_lib have the right type
2026-01-04 06:58:31 -08:00
Marcelo Domínguez
58e2610f71
Expose workgroup/thread dims as intrinsic args
2026-01-02 11:50:32 +01:00
Marcelo Domínguez
41a24c4b58
Add offload test for control flow handling
2025-12-31 13:11:28 +01:00
Marcelo Domínguez
04c2d2be13
Remove region_id unnamed attr
2025-12-19 13:27:14 +01:00
Marcelo Domínguez
3e4944d573
Split runtime global logic and cache kernel specific one
2025-12-19 13:27:13 +01:00
Marcelo Domínguez
5128ce10a0
Implement offload intrinsic
2025-11-25 20:04:27 +01:00
Manuel Drehwald
b56d555a36
fix host code
2025-10-19 09:28:39 -07:00
Guillaume Gomez
a27f3e3fd1
Rename tests/codegen into tests/codegen-llvm
2025-07-22 14:28:48 +02:00
