Fangrui Song
b279f6b098
[NVPTX,test] Change llc -march= to -mtriple=
...
Similar to 806761a7629df268c8aed49657aeccffa6bca449
-mtriple= specifies the full target triple while -march= merely sets the
architecture part of the default target triple (e.g. Windows, macOS),
leaving a target triple which may not make sense.
Therefore, -march= is error-prone and not recommended for tests without
a target triple. The issue has been benign as we recognize
nvptx{,64}-apple-darwin as ELF instead of rejecting it outrightly.
2024-12-15 10:45:11 -08:00
Youngsuk Kim
0f0a96b862
[llvm][NVPTX] Strip unneeded '+0' in PTX load/store ( #113017 )
...
Remove the extraneous '+0' immediate offset part in PTX load/stores, to
improve readability of output PTX code.
2024-10-19 10:05:36 -04:00
Youngsuk Kim
5a0942cd74
[llvm][NVPTX] Don't emit unused var 'temp_param_reg' (NFC) ( #89004 )
...
Don't emit unused variable 'temp_param_reg' which has been around since
ae556d3ef72dfe5f40a337b7071f42b7bf5b66a4 .
2024-04-17 14:45:33 -04:00
Adrian Kuegel
f0a5e50550
[llvm][NVPTX] Add missing feature guard.
2024-03-19 06:53:14 +00:00
Alex MacLean
89b7b3b995
[NVPTX] support dynamic allocas with PTX alloca instruction ( #84585 )
...
Add support for dynamically sized alloca instructions with the PTX
alloca instruction introduced in PTX 7.3
([9.7.15.3. Stack Manipulation Instructions: alloca]
(https://docs.nvidia.com/cuda/parallel-thread-execution/index.html#stack-manipulation-instructions-alloca ))
2024-03-15 11:51:46 -07:00
Youngsuk Kim
f9304974cc
[llvm][NVPTX] Inform that 'DYNAMIC_STACKALLOC' is unsupported ( #74684 )
...
Catch unsupported path early up, and emit error with information.
Motivated by the following threads:
* https://discourse.llvm.org/t/nvptx-problems-with-dynamic-alloca/70745
* #64017
2023-12-14 22:06:22 -05:00
