d77a150bf9
SWDEV-361591 - clean up header files
...
Change-Id: I25ca37a94524da0521f5b856a9fa9bab73f103c1
[ROCm/clr commit: 06867f0c04
2022-10-21 10:23:55 -04:00
7347172bc3
SWDEV-2 - Change OpenCL version number from 3516 to 3517
...
[ROCm/clr commit: 904f3e3d75
2022-10-21 03:00:52 -04:00
25ff252d73
SWDEV-362371 - Uninitialized stream error checks
...
Change-Id: I7e04ef46f0cf2b0c45caf42d5b2ee31cb68e93eb
[ROCm/clr commit: bded3d8d1f
2022-10-20 21:57:11 -04:00
326049efb8
SWDEV-360706 - Fix WriteAqlArgAt() crash in X86
...
In X86 on Windows, sizeof(size_t)=4, but size=8, for
amd::KernelParameterDescriptor::HiddenGlobalOffsetX/Y/Z items.
Loose the condition to prevent crash.
Change-Id: I2216f71f4d4fd6dd3766023b1c821cb3d35d7848
[ROCm/clr commit: 3d281114fb
2022-10-20 18:04:49 -04:00
17948751c9
SWDEV-362451 - Disable OCLPinnedMemory for now.
...
Change-Id: I9a771cfc5751e06c5a494ad2e4036f9aa3efbe9a
[ROCm/clr commit: b7d64f9f5f
2022-10-20 16:03:36 -04:00
644fc9421c
SWDEV-361730 - Fix to match the kernel arg allocation sizes between gpu visible mem and host/runtime mem
...
Change-Id: I6495aa57e97b46373c1d90cc230435780fdcf3ef
[ROCm/clr commit: b2665ccc4b
2022-10-20 14:20:37 -04:00
e81e8965c9
SWDEV-360944 - palBlit setArgument desc.size_ --> argSize
...
Fixes AMF tests that were impacted on Windows
Change-Id: I0326e794ec5a34fdec7a5ff4ed79d5bd055997d2
[ROCm/clr commit: a5b550e828
2022-10-20 12:44:15 -04:00
860e33dbed
SWDEV-362046 - Report HIP_OPS activities using the ROCr driver_node_id instead of the device's index
...
The ROCclr assigns zero-based IDs to GPUs in the order they are
discovered. That zero-based ID is what is used to identify the GPU
on which the HIP_OPS activity took place.
When multiple ranks are used, each rank's first logical device always
has GPU ID 0, regardless of which physical device is selected with
CUDA_VISIBLE_DEVICES. Because of this, when merging trace files from
multiple ranks, GPU IDs from different processes may overlap.
The long term solution is to use the KFD's gpu_id which is stable
across APIs and processes. Unfortunately the gpu_id is not yet exposed
by the ROCr, so for now use the driver's node id.
Change-Id: Ib78854527d600d175bb76e2df0747c33f898c615
[ROCm/clr commit: 9a82118c85
2022-10-20 12:31:30 -04:00
a8c33d177a
SWDEV-357122 - fix failure in vdi so as to handle unreasonable input size in MallocManagedNegativeTest
...
Change-Id: I1ed1916b652afc67327b0935c3c60fc2a404df30
[ROCm/clr commit: dacd55f3d7
2022-10-20 12:19:22 -04:00
cc132b1aec
SWDEV-2 - Change OpenCL version number from 3515 to 3516
...
[ROCm/clr commit: 05b2bd7995
2022-10-20 03:00:06 -04:00
cecfda83f5
SWDEV-362891 - Add the ocltst install path to RUNPATH.
...
So ocltst can automatically find liboclruntime.so from the same directory as ocltst.
Change-Id: Id2f4f58b1c7f0304be76a3aeecd5eea2eec328de
[ROCm/clr commit: cbbf0d1edd
2022-10-19 18:17:17 -04:00
c8eadd18af
SWDEV-362882 - Unable to run clinfo/opencl tests
...
The explicit RUNPATH addtion has been
revoked after the file reorg. Hence
the rpath for discovery of the libs
has to be taken care by individual binaries
Current patch adds the run path for
descovery of the libOpenCL.so
Signed-off-by: Ashutosh Mishra <ashutosh.mishra@amd.com >
Change-Id: Ia3bf9202bea78ecfdb881e40b0eb7a0d4a303ed3
[ROCm/clr commit: 1754b043a3
2022-10-19 15:03:07 -04:00
89321974f3
SWDEV-345870 - Correct HIP path for new directory layout
...
With file reorganization changes HIP is installed in /opt/rocm-ver
Using HIP path as /opt/rocm-ver/hip will generate warnings
Use real path with symlinks resolved, so that HIP path always points to the installed path
Removed redundant code for finding hsa header files. find_dependency to hsa-runtime should handle hsa dependency
Change-Id: Iccea3c1c7297c705244bf752f38fbff71929d64c
[ROCm/clr commit: f3394cf7a5
2022-10-19 12:58:02 -04:00
b2d02fe704
SWDEV-322620 - Virtual Memory Management
...
Hooked up the start VA hint address to PAL.
Change-Id: I4301d3eaaabcc85faf63fb25291291dcc3e8ebd1
[ROCm/clr commit: 470c4d7335
2022-10-19 10:50:07 -04:00
18808d87a4
SWDEV-356557, SWDEV-356558 - matching cuda return
...
Change-Id: I3b7d50e07bd437bd23a495439d66940ff30f7f07
[ROCm/clr commit: d7b27989b4
2022-10-19 09:17:28 -04:00
3db95d8b68
SWDEV-2 - Change OpenCL version number from 3514 to 3515
...
[ROCm/clr commit: 5cad630425
2022-10-19 03:00:13 -04:00
db848fbcd2
SWDEV-362891 - Incorperate TestLog into ocltst.
...
So it's no longer needed to setup path to find TestLog
Change-Id: I0c7ae21e426f2d6d832f707b9d37661bbed67100
[ROCm/clr commit: cf9f9e84b8
2022-10-18 18:01:53 -04:00
c0c8aa5eb6
SWDEV-349226 - Add the hipDataTypes mapping to cudaDataTypes
...
Change-Id: Iadeb2675adf314c02dcc5d722581678ec36f8632
[ROCm/clr commit: 1c68c74bb7
2022-10-14 06:44:30 -04:00
9bfae25a7c
SWDEV-307411 - Require comment/message string in guarantee().
...
Change-Id: I55a699c5366f6a7d167ede4c9be2ec5b15cea9d0
[ROCm/clr commit: 951a5b9e7d
2022-10-13 16:21:21 -04:00
e380d85f27
SWDEV-355727 - Change in AMDGCN format in bitcode.
...
Change-Id: Ic32a3e7b866b35a321d8ececc2f6c63f9b23121d
[ROCm/clr commit: 1b902e1809
2022-10-13 14:19:03 -04:00
409c07e3d8
SWDEV-359548 - Enqueue marker if fence is dirty
...
The marker would flush any caches.
Change-Id: I5f990285fb3c3a3e8d5d0b2a2d5a9f1fdcea435c
[ROCm/clr commit: da88091ba4
2022-10-13 07:53:01 -04:00
bfe5591f69
SWDEV-2 - Change OpenCL version number from 3513 to 3514
...
[ROCm/clr commit: 4e0e0666d6
2022-10-13 03:00:09 -04:00
a0981e4ccd
SWDEV-357759, SWDEV-360041, SWDEV-361145 Fix Stream end capture on forked streams
...
Change-Id: If0dc6242d2d3ca680e37e14a5dea5cf68dc295df
[ROCm/clr commit: 039e26ee0f
2022-10-12 13:00:05 -04:00
69a1326d3f
SWDEV-318349 - Remove sync for null stream
...
Remove sync for null stream in hipGraphExec::Run()
Change-Id: Ieaaed1c15b4d258193d8341d4b17d9f03a9e4783
[ROCm/clr commit: 56e7c8b3a0
2022-10-12 09:46:18 -04:00
537b3dd67f
SWDEV-356570, SWDEV-356571, SWDEV-356572, SWDEV-356575 -
...
updating with current status of hipNotSupported and null checks
Change-Id: I3b1ce3d3e5ce9b69d1f5dacda35f5aa586714f5c
[ROCm/clr commit: 88c399a43d
2022-10-11 23:45:27 -04:00
ca2235f671
SWDEV-353548 - assert with different behavior for Release vs Debug
...
assert statement behavior
-Debug: crashes tests with SIGABORT
-Release: continues silently without
Change-Id: I7578eb16a7391ff7f9d68f1cae3bcea7f8225579
[ROCm/clr commit: be80bf5406
2022-10-11 14:00:28 -04:00
fc3f04fa7d
SWDEV-356551 - Add support for device attributes.
...
Change-Id: Ic0f2b9cb510b027e0b8d1b3fb6d984e0d2600b15
[ROCm/clr commit: 7ef4eada54
2022-10-11 12:06:44 -04:00
f664f42443
SWDEV-360001 - Add hipLaunchHostFunc in Nvidia path
...
Change-Id: I5c733b416642174a089a86c420b03e8775e206d6
[ROCm/clr commit: 0d67bd3f09
2022-10-11 06:18:52 -04:00
3d97da8522
SWDEV-356314 - Init managed vars on device kernel is launched
...
Change-Id: I9c30b7d0e2a490dbb2bd9c02dccaae17d76874ef
[ROCm/clr commit: 0a3e89c0cb
2022-10-11 00:18:30 -04:00
434df167df
SWDEV-359548 - Detect fence optimization
...
- Use a dirty flag to determine fence optimization
- If fence is dirty submit a marker at top level to sync.
Change-Id: I53fb19b5bb05b7c7b37c41637a6c7aaf870b639a
[ROCm/clr commit: 6405b6cdba
2022-10-10 14:15:16 -04:00
6ec1955d2c
Fix a build error when compiling with clang
...
Fix the following error:
hip_intercept.cpp:52:7: error: reinterpret_cast from 'const void *' to 'decltype(activity_prof::report_activity.load())' (aka 'int (*)(activity_domain_t, unsigned int, void *)') casts away qualifiers
reinterpret_cast<decltype(activity_prof::report_activity.load())>(function),
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
by replacing the 'const void *function' argument with the correct type.
Change-Id: I859d45ee01b7aaa1e46563cdc37de57b4159d330
[ROCm/clr commit: c3a340c21b
2022-10-10 12:30:36 -04:00
f149b21399
SWDEV-351969 - TLS Optimization
...
- Aggregate all TLS(Thread Local Storage) variables into a single class
- This is to improve cache accesses per thread
Change-Id: Ic8361eaeae290fff00254684e309471958365eb9
[ROCm/clr commit: 8b391ef18c
2022-10-09 22:08:14 -04:00
9fdcdf6cd3
SWDEV-2 - Change OpenCL version number from 3512 to 3513
...
[ROCm/clr commit: 621faa03ec
2022-10-08 03:00:09 -04:00
2be45a82ec
SWDEV-318349 - Fix hipGraphKernelNode and hipGraphMemcpyNode
...
For hipGraphKernelNode, remove func_;
and reorganize functions to naturely support mGPU;
For hipGraphMemcpyNode, make EnqueueCommands() support different
queues' sync
Change-Id: I22708923f454adf4456ff99d25559daffed8c20d
[ROCm/clr commit: f83ba8cd23
2022-10-07 09:07:56 -04:00
41d911d637
SWDEV-356482 - Improve logging
...
Change-Id: I0c6286fcdb9d94398ed66262e95e8837898ff9fe
[ROCm/clr commit: eef2f99513
2022-10-06 11:13:24 -04:00
3c01989266
SWDEV-2 - Change OpenCL version number from 3511 to 3512
...
[ROCm/clr commit: 3e3fa68657
2022-10-06 03:00:06 -04:00
f25d8dae77
SWDEV-357246 - Adds a missing return statement
...
Change-Id: I2216f71f4d4fb6dd3766023b0c821cb3d35d7849
[ROCm/clr commit: 84fbb30b7c
2022-10-05 16:29:32 -04:00
8ceb2ac5d6
SWDEV-357246 - Fixes to return with a error code instead of segfault in alloc failure case
...
Change-Id: Ie18d50f2f7ef3f4b2c0153152fca057232c1f8a2
[ROCm/clr commit: aeeab1c975
2022-10-04 09:38:42 -04:00
7bcdb29caa
SWDEV-351980 - Add FillBuffer byte count to the record
...
Change-Id: I90c791f5810b8a3f6b1d6a9e81c165b1a7515c92
[ROCm/clr commit: e00965df50
2022-09-30 21:20:14 -07:00
08d7b38b0f
SWDEV-359838 - Add a phase data pointer to the hip_api_data_t
...
To avoid using the thread local std::stack to remember the phase enter
timestamp, the tracer tool uses the phase data to store the timestamp.
Change-Id: I9e95637b41d6f0b2bd61016062ca07d6ba897652
[ROCm/clr commit: 47a29c748a
2022-09-30 14:55:01 -07:00
9c0157e582
SWDEV-2 - Change OpenCL version number from 3510 to 3511
...
[ROCm/clr commit: 800fe4b239
2022-09-30 03:00:10 -04:00
d4274a37c7
SWDEV-1 - Fix typo in compiler utils
...
Found by the Debian maintainers:
https://github.com/ROCm-Developer-Tools/ROCclr/issues/28
Signed-off-by: Jeremy Newton <Jeremy.Newton@amd.com >
Change-Id: I8c09913f7b7f4ad02f12a0630f15e66bdd5360a7
[ROCm/clr commit: 65fd4f5f07
2022-09-29 18:14:29 -04:00
a7f33f9237
SWDEV-357250 - hiprtc support with AMD_LOG_LEVEL when linked as separate library
...
Change-Id: Ieda83af85a9344b4e09d4fe85acc65386db79c39
[ROCm/clr commit: b425e27230
2022-09-23 05:50:07 -04:00
f1c029e147
SWDEV-2 - Change OpenCL version number from 3509 to 3510
...
[ROCm/clr commit: f93d921c4f
2022-09-23 03:00:11 -04:00
32838b0613
SWDEV-356651 - Merge hipJit options between hiprtc and hip runtime.
...
Change-Id: I73b66eaf081ce63b6b7ce45f71880877953cf81c
[ROCm/clr commit: ab89a70f5d
2022-09-22 12:53:58 -04:00
0dd9add8e1
SWDEV-352001 - Store last scopes for dispatch
...
- Store last fence scopes and use the last value to determine if we need a cache flush again. This helps cases where hipExtLaunchKernel API is
used.
- Purge code for ROC_EVENT_NO_FLUSH
Change-Id: I531cf9c9c60d5e2b3a9e265d0f52f79ed2fa8a8c
[ROCm/clr commit: 9b5cbd37a2
2022-09-22 11:34:10 -04:00
a045c28998
SWDEV-2 - Change OpenCL version number from 3508 to 3509
...
[ROCm/clr commit: ae46d57519
2022-09-22 03:00:08 -04:00
12c48303c8
Revert "SWDEV-351980 - Update hip_prof_str.h"
...
This reverts commit c878ce02f6835b7578dc
2022-09-21 13:04:38 -04:00
b0825090e9
SWDEV-355727 - Added more rules to target ID comparison in hiprtc.
...
Change-Id: I56137375c3eb47dd604c7a02bb88f7e3dbe6c04d
[ROCm/clr commit: d870bdb9b1
2022-09-21 12:18:13 -04:00
c878ce02f6
SWDEV-351980 - Update hip_prof_str.h
...
Change-Id: Ied27c137aff54e69dc14129e6daad9196479edf9
[ROCm/clr commit: f296e42803
2022-09-21 11:38:59 +00:00
Alex (Bin) Xie
Chauncey Hui
Satyanvesh Dittakavi
Tao Sang
jatang
Sourabh Betigeri
Ajay
Laurent Morichetti
Julia Jiang
Ashutosh Mishra
Ranjith Ramakrishnan
Christophe Paquot
pghafari
kjayapra-amd
Saleel Kudchadker
Anusha GodavarthySurya
Jaydeep Patel
Rakesh Roy
Jeremy Newton
Maneesh Gupta