23 Commits

Author SHA1 Message Date
Jun Wang
86842e1f72
[AMDGPU] New clang option for emitting a waitcnt instruction after each memory instruction (#79236)
This patch introduces a new command-line option for clang, namely,
amdgpu-precise-mem-op (or precise-memory in the backend). When this option is specified, a waitcnt
instruction is generated after each memory load/store instruction. The
counter values are always 0, but which counters are involved depends on
the memory instruction.

---------

Co-authored-by: Jun Wang <jun.wang7@amd.com>
2024-04-10 10:47:04 -07:00
Fangrui Song
2f79bb1046 [Driver][test] Add -nogpulib to some AMDGPU tests
An upcoming change D156363 will change -### to exit with code 1 if
hasErrorOccurred. Some AMDGPU tests would exit with code 0 due to not
found GPU installation.
2023-07-28 19:57:25 -07:00
pvanhout
30e7cd4877 [clang] Remove legacy -m(no)-code-object-v3 options
Code object V2 and V3 have been deprecated for a long time.
They're now scheduled to be removed completely from LLVM in the coming weeks/months.

There is no reason to support those legacy options anymore as they've
also been deprecated for a long time.

Reviewed By: #amdgpu, yaxunl, artem.tamazov

Differential Revision: https://reviews.llvm.org/D145671
2023-03-27 09:33:59 +02:00
Stanislav Mekhanoshin
a8d9d50762 [AMDGPU] gfx90a support
Differential Revision: https://reviews.llvm.org/D96906
2021-02-17 16:01:32 -08:00
Yaxun (Sam) Liu
0b81d9a992 [AMDGPU] add -mcode-object-version=n
Add option -mcode-object-version=n to control code object version for
AMDGPU.

Differential Revision: https://reviews.llvm.org/D91310
2020-12-07 18:08:37 -05:00
Yaxun (Sam) Liu
40ad476a32 [clang][AMDGPU] rename sram-ecc as sramecc
As backend renamed sram-ecc to sramecc, this patch makes
corresponding change in clang.

Differential Revision: https://reviews.llvm.org/D86217
2020-12-07 18:05:47 -05:00
Konstantin Zhuravlyov
67f189e93c Make sure both cc1 and cc1as process -m[no-]code-object-v3
Differential Revision: https://reviews.llvm.org/D89478
2020-10-15 14:03:26 -04:00
Konstantin Zhuravlyov
3fdf3b1539 AMDGPU: Update AMDHSA code object version handling
Differential Revision: https://reviews.llvm.org/D89076
2020-10-14 13:04:27 -04:00
Konstantin Zhuravlyov
e2eaa91451 AMDGPU: Remove -mamdgpu-debugger-abi option
It has been unsupported for few years now.

Differential Revision: https://reviews.llvm.org/D89125
2020-10-13 12:20:28 -04:00
Yaxun (Sam) Liu
36501b180a Emit predefined macro for wavefront size for amdgcn
Also fix the issue of multiple -m[no-]wavefrontsize64
options to make the last one wins.

Differential Revision: https://reviews.llvm.org/D88370
2020-10-02 10:17:21 -04:00
Yaxun (Sam) Liu
7546b29e76 [HIP] Support target id by --offload-arch
This patch introduces support of target id by
-offload-arch.

Differential Revision: https://reviews.llvm.org/D60620
2020-08-18 23:43:53 -04:00
Stanislav Mekhanoshin
8a8131a3f6 [AMDGPU] gfx1010 wave32 clang support
Differential Revision: https://reviews.llvm.org/D63209

llvm-svn: 363341
2019-06-13 23:47:59 +00:00
Stanislav Mekhanoshin
91792f1b93 [AMDGPU] gfx1010 clang target
Differential Revision: https://reviews.llvm.org/D61875

llvm-svn: 360634
2019-05-13 23:15:59 +00:00
Matt Arsenault
d91bb4831b AMDGPU: Don't emit debugger subtarget features
Keep the flag around for compatability.

llvm-svn: 354624
2019-02-21 21:31:43 +00:00
Konstantin Zhuravlyov
08c41659ef AMDGPU: Add sram-ecc feature options
Differential Revision: https://reviews.llvm.org/D53223

llvm-svn: 346178
2018-11-05 22:44:59 +00:00
Konstantin Zhuravlyov
7122595f81 AMDGPU: Add options to enable/disable code object v3
Differential Revision: https://reviews.llvm.org/D53386

llvm-svn: 344711
2018-10-17 21:39:12 +00:00
Konstantin Zhuravlyov
37e9739a58 AMDGPU: Remove amdgpu-debugger-reserve-regs feature
llvm-svn: 335287
2018-06-21 20:27:47 +00:00
Konstantin Zhuravlyov
bfd6c1c016 AMDGPU: Add -mxnack/-mno-xnack options that set +/-xnack feature
Differential Revision: https://reviews.llvm.org/D39878

llvm-svn: 317917
2017-11-10 19:28:25 +00:00
Serge Pavlov
b43573b9a4 Driver must return non-zero code on errors in command line
This is recommit of r302775, reverted in r302777 due to a fail in
clang-tidy. Original mesage is below.

Now if clang driver is given wrong arguments, in some cases it
continues execution and returns zero code. This change fixes this
behavior.

The fix revealed some errors in clang test set.

File test/Driver/gfortran.f90 added in r118203 checks forwarding
gfortran flags to GCC. Now driver reports error on this file, because
the option -working-directory implemented in clang differs from the
option with the same name implemented in gfortran, in clang the option
requires argument, in gfortran does not.

In the file test/Driver/arm-darwin-builtin.c clang is called with
options -fbuiltin-strcat and -fbuiltin-strcpy. These option were removed
in r191435 and now clang reports error on this test.

File arm-default-build-attributes.s uses option -verify, which is not
supported by driver, it is cc1 option.

Similarly, the file split-debug.h uses options -fmodules-embed-all-files
and -fmodule-format=obj, which are not supported by driver.

Other revealed errors are mainly mistypes.

Differential Revision: https://reviews.llvm.org/D33013

llvm-svn: 303756
2017-05-24 14:57:17 +00:00
Serge Pavlov
738d3b97af Reverted r302775
llvm-svn: 302777
2017-05-11 08:25:22 +00:00
Serge Pavlov
c5cc230587 Driver must return non-zero code on errors in command line
Now if clang driver is given wrong arguments, in some cases it
continues execution and returns zero code. This change fixes this
behavior.

The fix revealed some errors in clang test set.

File test/Driver/gfortran.f90 added in r118203 checks forwarding
gfortran flags to GCC. Now driver reports error on this file, because
the option -working-directory implemented in clang differs from the
option with the same name implemented in gfortran, in clang the option
requires argument, in gfortran does not.

In the file test/Driver/arm-darwin-builtin.c clang is called with
options -fbuiltin-strcat and -fbuiltin-strcpy. These option were removed
in r191435 and now clang reports error on this test.

File arm-default-build-attributes.s uses option -verify, which is not
supported by driver, it is cc1 option.

Similarly, the file split-debug.h uses options -fmodules-embed-all-files
and -fmodule-format=obj, which are not supported by driver.

Other revealed errors are mainly mistypes.

Differential Revision: https://reviews.llvm.org/D33013

llvm-svn: 302775
2017-05-11 08:00:33 +00:00
Konstantin Zhuravlyov
e6dbd902c0 [AMDGPU] Remove individual debugger options + update features
Differential Revision: http://reviews.llvm.org/D20336

llvm-svn: 270895
2016-05-26 19:36:39 +00:00
Konstantin Zhuravlyov
5124bf8edc [AMDGPU] Add debugger related target options
Differential Revision: http://reviews.llvm.org/D18748

llvm-svn: 266133
2016-04-12 21:42:15 +00:00