Correctly reference the new amdocl target.
Do not use hardcoded paths to libamdocl64.so, instead query the path to
it from cmake.
Change-Id: I4738dd8d6d08f9fb72fdd725b6488b81342755a4
[ROCm/clr commit: 6046e6fea4]
Correctly reference the new amdocl target.
Do not use hardcoded paths to libamdocl64.so, instead query the path to
it from cmake.
Change-Id: I4738dd8d6d08f9fb72fdd725b6488b81342755a4
-Bsymbolic is required to make sure all AMD OpenCL runtime symbols are
resolved internally, otherwise ld might resolve them using symbols from
the OpenCL ICD. This will cause incorrect functions to be called.
Change-Id: I4ca0326a574ee2e537525bfda96ec20fabfe0158
[ROCm/clr commit: bc5c8df20b]
-Bsymbolic is required to make sure all AMD OpenCL runtime symbols are
resolved internally, otherwise ld might resolve them using symbols from
the OpenCL ICD. This will cause incorrect functions to be called.
Change-Id: I4ca0326a574ee2e537525bfda96ec20fabfe0158
move blacklisted test case from gfx902 iommuv2 to dgpu path.
Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
Change-Id: I8b101226ca8dcd0c12c484f5f6ce12fe73a75bdc
Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
(cherry picked from commit 9cf4377572321396225950b9a58beb549120c2a3)
This patch also includes additional cleanup logic to the installation
rules. The OpenCL ICD already defines its own installation rules, no
need to duplicate them.
Change-Id: Id1c386307a83b82e21f6fa132e41957d9b348fd3
[ROCm/clr commit: c6eadce180]
This patch also includes additional cleanup logic to the installation
rules. The OpenCL ICD already defines its own installation rules, no
need to duplicate them.
Change-Id: Id1c386307a83b82e21f6fa132e41957d9b348fd3
Remove extra -isystem causing --cuda-device-only flag
to be dropped. Fix missing __hipRTC_header_size symbol.
Add macros which were consumed during pre-processor
generation. Add INT_MAX and CHAR_BIT macros used in
HIP headers.
Change-Id: Id5143e3c8a2b1e7c78658ba84d5ab3b55ac1fa72
[ROCm/clr commit: 7c8e210c26]
Remove extra -isystem causing --cuda-device-only flag
to be dropped. Fix missing __hipRTC_header_size symbol.
Add macros which were consumed during pre-processor
generation. Add INT_MAX and CHAR_BIT macros used in
HIP headers.
Change-Id: Id5143e3c8a2b1e7c78658ba84d5ab3b55ac1fa72
Remove extra -isystem causing --cuda-device-only flag
to be dropped. Fix missing __hipRTC_header_size symbol.
Add macros which were consumed during pre-processor
generation. Add INT_MAX and CHAR_BIT macros used in
HIP headers.
Change-Id: Id5143e3c8a2b1e7c78658ba84d5ab3b55ac1fa72
[ROCm/hip commit: b982637965]
Remove extra -isystem causing --cuda-device-only flag
to be dropped. Fix missing __hipRTC_header_size symbol.
Add macros which were consumed during pre-processor
generation. Add INT_MAX and CHAR_BIT macros used in
HIP headers.
Change-Id: Id5143e3c8a2b1e7c78658ba84d5ab3b55ac1fa72
1st issue was that one of the ostream ops failed to print the
content of the struct.
2nd issue: get_ptr_type was called with args being src/dest
pointers while it should be the agents pointers for src/dest.
3rd issue: memcopies map used (recordid, procid, is_async)
as a key but this is not enough as some copies share same key,
so I added begin/end timestamps as a way to distinguish between them.
Change-Id: I7c6e80e74e30ea572f21612aaf0cf7efec6e91e6
1st issue was that one of the ostream ops failed to print the
content of the struct.
2nd issue: get_ptr_type was called with args being src/dest
pointers while it should be the agents pointers for src/dest.
3rd issue: memcopies map used (recordid, procid, is_async)
as a key but this is not enough as some copies share same key,
so I added begin/end timestamps as a way to distinguish between them.
Change-Id: I7c6e80e74e30ea572f21612aaf0cf7efec6e91e6
[ROCm/rocprofiler commit: 761bd6a86b]
Add macro debug_warning_n to stop printing a message after
N instances.
Change-Id: Id5f84b11eb63b3a20bd2bcb2ea8f10a066b457ef
[ROCm/ROCR-Runtime commit: ca8387768e]