Sourabh Betigeri
c839dbc470
SWDEV-246104: Added new performance test to verify stream concurrency
...
Change-Id: Ieceaafd4d404b137936eb8aceecff27bbba42026
[ROCm/clr commit: 90e4f2062a ]
2020-08-20 13:42:55 -04:00
Aryan Salmanpour
dee8fed222
[HipPerf] add a test for measuring create/copy/destroy for hip streams
...
SWDEV-245290/SWDEV-247767
Change-Id: Id52e3d2f7afee52763da107eb5ed04627942d9f0
[ROCm/clr commit: 29bca5b949 ]
2020-08-20 13:28:36 -04:00
Todd tiantuo Li
005dcecd5c
move hipFuncSetCacheConfig from hip_device to hip_module
...
Change-Id: If5a930e1210e76531f5ba9a0e5f5ec98ad473a19
[ROCm/clr commit: 3ac8483124 ]
2020-08-17 17:39:01 -04:00
Aryan Salmanpour
20fe2fd7fd
SWDEV-248057 - fix the calculation of allGridSize used in multi_grid_group() API
...
Change-Id: Ib470094e28dcacaa4769dc5c7ab08924f5b7fa41
[ROCm/clr commit: 359c0fbb76 ]
2020-08-17 12:52:40 -04:00
Jatin Chaudhary
6b597909ce
Adding Anyorder flag to HIP
...
Change-Id: Ie20931541b3febe23fa9ac36ebc0c90de75a5f0a
[ROCm/clr commit: 628a308475 ]
2020-08-17 08:46:52 -04:00
Rahul Garg
608fded501
Fix hipDispatchEnqueueRateMT sample
...
Change-Id: I7ade0dcf8b32ac90a311022d12339740b2b3df44
[ROCm/clr commit: ee3dbfac49 ]
2020-08-14 18:56:39 -04:00
Todd tiantuo Li
4885fa0e48
SWDEV-240803 - add hipFuncSetAttribute and hipFuncAttribute
...
Change-Id: I3f4d67b19d89fd348fa5b884af4a2542ee4aba60
[ROCm/clr commit: fb43f21044 ]
2020-08-14 17:39:29 -04:00
kjayapra-amd
cddec46bb7
SWDEV-237658 - Do not attach events to user invisible Marker.
...
Change-Id: I3a74fd2614d6e2e1d51543471bbd5ae51a713a47
[ROCm/clr commit: ea15c75c55 ]
2020-08-14 10:55:34 -04:00
Freddy Paul
f959aadea0
Enable developers to build hip without CMAKE_PREFIX_PATH
...
With recent changes in ROCclr to take HSA cmake target it require
HIP to have CMAKE_PREFIX_PATH added for install location for HSA,
amd_comgr,llvm.
User can provide CMAKE_PREFIX_PATH if above
packages are built/installed at custom location as:
cmake -DCMAKE_PREFIX_PATH="<path_hsa> <path_amd_comgr> <path_llvm>"
For default ROCm installation adding a default path will help.
Change-Id: Iefa443dc198da76b57422a641c9821c63cde4922
[ROCm/clr commit: 95729c31e8 ]
2020-08-13 23:13:40 -04:00
Aryan Salmanpour
15eaef319d
[HipPerf] relocating/renaming some hip perf tests
...
Change-Id: Ie85d242cd68cd14a858e07ed27875a5196014688
[ROCm/clr commit: de1e14b252 ]
2020-08-13 17:28:11 -04:00
Aryan Salmanpour
1c52703636
[HipPerf] report performance based on wall time only for hipPerfDevMemReadSpeed/hipPerfDevMemWriteSpeed
...
Change-Id: I1fda2ec76da6fad6852d328e0a3fc39e28af57bb
[ROCm/clr commit: 1a2622dd60 ]
2020-08-13 14:23:58 -04:00
Jason Tang
91a2f86826
SWDEV-232197 - Remove refering to gfxipVersion_
...
Change-Id: I91b6b972e26b18dabf6deda5a142bff491c8bb45
[ROCm/clr commit: d5c97e0c31 ]
2020-08-12 16:34:18 -04:00
German Andryeyev
031cf9d32a
Allow prefetch on the null stream
...
Change-Id: I87e950460361fab65ab140988fc2288bcb194f08
[ROCm/clr commit: b73673bc74 ]
2020-08-12 13:07:43 -04:00
Rahul Garg
420ec4eeae
Return error from hipMallocManaged for size =0
...
Change-Id: Ida0cecf05e1df731176d41e59422d1e606cc9bd9
[ROCm/clr commit: 236ef94166 ]
2020-08-10 18:25:31 +00:00
Vlad Sytchenko
4a3b0190c5
Don't align 2D image buffer size
...
Cuda allows creating 2D textures from memory with unaligned size.
Aligning it will cause us to overshoot the size of the underlying
allocation.
Change-Id: Ia20dec2105e36ca4a8448c9c6299d0001b388d25
[ROCm/clr commit: 963c6fe4e9 ]
2020-08-09 15:53:52 -04:00
Julia jiang
ac119447fd
SWDEV-246878-Update document HIP_porting_guide
...
Change-Id: I7f4c92637f31da76293048133fff5af816c752ca
[ROCm/clr commit: e93f7b6a56 ]
2020-08-07 14:50:01 -04:00
Aryan Salmanpour
7ec4a9af8e
[HipPerf] add two subtests for measuring maximum device memory read/write speed
...
SWDEV-245290 / SWDEV-246220 [HIPPerf] Port OCLPerfDevMemWriteSpeed/OCLPerfDevMemReadSpeed into hip performance subtests
Change-Id: I5dc323c75cebbc17596dcb4ed9492e18c5246868
[ROCm/clr commit: b98727505d ]
2020-08-07 14:21:27 -04:00
David Salinas
dbeb78fa07
correct -x option in hip::device
...
Change-Id: I8c288e678b4357019144a024395cda1fcead61fc
[ROCm/clr commit: 996df70de9 ]
2020-08-07 12:33:43 -04:00
Aryan Salmanpour
0372607c1b
[HipPerf] add a test for measuring shared memory read speed
...
SWDEV-245290/SWDEV-247330
Change-Id: If78fa0879ff58aab84775e412a86665c7e5959e7
[ROCm/clr commit: 61dafa0397 ]
2020-08-07 10:07:05 -04:00
Yaxun (Sam) Liu
1ca3e4696e
Let hipcc not escape space between obj files extracted from library
...
hipcc extracts bundles from static libraries and pass them to clang. It should
not escape spaces between objects extracted from library.
Fixes: SWDEV-246544
Change-Id: Iad3902f28a43a3986917ee37a49f10382c329940
[ROCm/clr commit: c1a498c219 ]
2020-08-07 09:34:27 -04:00
Jatin Chaudhary
492c85b8fb
Bring .clang-format file in sync with ROCclr's
...
Change-Id: I1990722c0779d61f66d79a5d0649d1b64e42eaf7
[ROCm/clr commit: fb2760a8aa ]
2020-08-07 05:34:44 -04:00
Jatin Chaudhary
093f0490d6
Fixing the memchannel calculation
...
Change-Id: Ib5f5931cc0e1ee5b7cb356c2ede0cb951ee8c9de
[ROCm/clr commit: f0d9df0b71 ]
2020-08-07 05:34:23 -04:00
Jatin Chaudhary
bee94be69d
Cuda 11 changes: https://docs.nvidia.com/cuda/cuda-runtime-api/structcudaPointerAttributes.html\#structcudaPointerAttributes
...
Change-Id: I8a5389b88df286043c365a734983a4c5de352102
[ROCm/clr commit: 94bfd485ef ]
2020-08-07 05:33:33 -04:00
Saleel Kudchadker
fd7ea936fe
Enable queue profile only if we attach a profiler
...
Submit explicit profile marker for hipEventRecord to record
timestamps. Enable explicit signal profiling if the API specifies
start and stop events.
Toggle this with env var HIP_FORCE_QUEUE_PROFILING=0
Change-Id: Iae449a63ec3ebf6c2880e65d7b1dd1031a29018f
[ROCm/clr commit: dec95e58e3 ]
2020-08-06 11:28:16 -04:00
Payam
1cdf2afd5e
adding duration to print log for hip api calls
...
Change-Id: Icffa8fcacfb8ba24861d980b9553faeee4efec8c
[ROCm/clr commit: c24f884c76 ]
2020-08-06 06:42:21 -04:00
Sarbojit Sarkar
ee526f9212
Added free device memory info
...
Recently there were few OOM(out of Memory) issues poped. Adding trace will help know genuine OOM issue.
Change-Id: Ie2a56bba95cead24caf1af5c807e73c383c2f5d9
[ROCm/clr commit: bc76760452 ]
2020-08-06 01:16:36 -04:00
Anusha Godavarthy Surya
66be9e86d5
SWDEV-243961 Added overflow check for globalWorkSize
...
Change-Id: I153d99398eb4619baad8c0a52466e0d5d5f898ca
[ROCm/clr commit: 8481524416 ]
2020-08-05 11:18:07 -04:00
Aryan Salmanpour
9fa78dd816
[dtest] disable hipStreamCreateWithPriority test temporarily as it fails randomly on Jenkins
...
Change-Id: I4edcd555653ef8e7562245da3a6f310d60884fe0
[ROCm/clr commit: cc0f0bae74 ]
2020-07-30 16:18:04 -04:00
jujiang
d308b7006e
SWDEV-242207-Update deprecated APIs
...
Change-Id: Icb64227be0eb44497de99240c0044a162ff3c085
[ROCm/clr commit: 5ac8672fb2 ]
2020-07-30 15:15:41 -04:00
Rahul Garg
ce2da5238a
Support empty sources in add_executable and add_library
...
Change-Id: I1441f2c045aeda2bba99bffe1c99fcc4a59cea9b
[ROCm/clr commit: e3617167f5 ]
2020-07-30 13:34:46 -04:00
Rahul Garg
d66c70a8e1
Bump version to 3.8
...
Change-Id: I710ee1456aa108eb979eb5bef8f32081dda593d9
[ROCm/clr commit: 2140833421 ]
2020-07-30 13:30:55 -04:00
Jatin Chaudhary
6c217e7f69
Fixing/Disabling tests on nvcc path
...
Change-Id: I7ed4a1246943867482ff595d5f98bb25bd79074b
[ROCm/clr commit: d251c2db57 ]
2020-07-30 00:39:07 -04:00
Freddy Paul
0210995414
hip-rocclr:Remove dependency to rocm-utils
...
rocm-utils is a meta-pacakge that pull in lot more pacakges than
what is required for hip.
Correct dependency to set as of today will be rocminfo.
Change-Id: I065ca74b4520e0c9ce183d82d6026554ff43e1f0
[ROCm/clr commit: b0438e6e99 ]
2020-07-28 11:42:09 -04:00
Jason Tang
aa179b01a3
SWDEV-243576 - Fix hipMemcpy regression.
...
If the queue device doesn't match the device on any memory object, use the queue device from the memory object.
Change-Id: I5fdcf00494f8391574f4443332c01788b8da44ef
[ROCm/clr commit: e430506f70 ]
2020-07-28 03:04:00 -04:00
Jatin Chaudhary
380e2fe752
Removing redundant .hipInfo Install
...
Change-Id: Id8f7a4668d8b6965a47a464c94c19bc1036f3537
[ROCm/clr commit: 78fcbab8c2 ]
2020-07-28 02:01:53 -04:00
Rahul Garg
47c57a5a82
Remove libhip_hcc symlinks to libamdhip64
...
Change-Id: I7b5b47fc7ca9760316c2424c16b454d7919c665c
[ROCm/clr commit: c919a88795 ]
2020-07-27 23:58:29 +00:00
jujiang
e5b39e4056
SWDEV-242344-hipMemcpyAll failed on MGPU
...
Change-Id: I01671e322c84213964d3d48bd6b5dc704dba731a
[ROCm/clr commit: c1d65e57a1 ]
2020-07-27 13:30:30 -04:00
Tao Sang
d7c4f9f492
Remove __gnu_h2f_ieee and __gnu_f2h_ieee
...
Change-Id: Ide24d245d851e20961020323e52f33322a33fff9
[ROCm/clr commit: e9cc570be8 ]
2020-07-23 14:50:24 -04:00
Vlad Sytchenko
bddfc6a790
2D/3D copy optimizations
...
SWDEV-244798
If {src/dst} ptr is marked as hipMemoryTypeHost, check if the memory was
prepinned. If it was, upgrade the copy type to hipMemoryTypeDevice to
avoid extra pinning.
Change-Id: Id287ef5b14ae67dfbcf80c4caa1b08a311191948
[ROCm/clr commit: fe6b645f3b ]
2020-07-23 11:24:16 -04:00
Tao Sang
073b1e28d5
Apply constexpr on global constant varaibles
...
When HIP_ENABLE_DEFERRED_LOADING=0, many global variables will be
referenced but they are not initialized in that early time. The patch
will use constexpr to initialze global constant varables in compile
time.
Change-Id: I9837c42e6bab38a894ece4e0f34b64f81b38fef0
[ROCm/clr commit: f7cc49c0b2 ]
2020-07-22 22:15:03 -04:00
kjayapra-amd
db1fbf23d2
SWDEV-236178 - Remove use of old routines functions()/modules().
...
Change-Id: I3a1ed967227c91b0d8cdf39e1360ade685e0bc73
[ROCm/clr commit: fd783c1e88 ]
2020-07-22 14:55:02 -04:00
kjayapra-amd
7aa68bc3cb
SWDEV-240800 - Adding initial support for hipDeviceGetP2PAttribute. Some attr support still pending.
...
Change-Id: I0611aed136270db497dfa374144f6f5e35352a8f
[ROCm/clr commit: 311fc8c513 ]
2020-07-22 14:54:34 -04:00
Payam
d1e3736fc3
modify cmake to strip debug symbols from rel blds
...
Change-Id: Ia67261fc3c4a20f3f1bbe439a70384a3f6aefa91
[ROCm/clr commit: 274fded47d ]
2020-07-22 11:53:54 -04:00
Tao Sang
25c98ea4de
Support numa if libnume-dev is installed
...
Let rocclr decide numa dependence
Change-Id: I65bdfba7ec0d06b550f86632318bcfd1f765cfa9
[ROCm/clr commit: eba78710dc ]
2020-07-22 08:55:28 -04:00
Satyanvesh Dittakavi
89d36cba5f
SWDEV-244772 - Fix dtest hipMemcpyWithStreamMultiThread.cpp.
...
hipSetDevice is not used correctly to allocate on multiple devices in mGPU setup.
Due to which hipMalloc was called on the same device on multiple threads leading to out of memory issues on some devices with lesser memory.
Change-Id: I0e5b1bc028b9ecb11bd40c3a5edf715f8bd721ff
[ROCm/clr commit: 6f5e63ebac ]
2020-07-22 00:52:14 -04:00
Saleel Kudchadker
2bf550d3c1
Print devPtr and size for hip*Symbol* api
...
Change-Id: I07be6949cca005700ee1d1708d86220c793c9728
[ROCm/clr commit: 603546cd45 ]
2020-07-21 13:49:10 -04:00
Rahul Garg
7654a45737
Handle size 0 symbols
...
Change-Id: Ie282b37cf6bbccae4611dfc17c5c7f31977feea7
[ROCm/clr commit: 19ede4fd23 ]
2020-07-21 10:09:26 -04:00
Aryan Salmanpour
dfc8958316
Add missing stream null check for some hipStreamCreate APIs
...
Change-Id: I716d71e4ec59b0bd7922869bfa0ed908c22c289e
[ROCm/clr commit: e27d53043c ]
2020-07-20 18:07:03 -04:00
jujiang
362faab3f5
SWDEV-242337-Fix hipHostRegister test failure for MGPU
...
Change-Id: I748dac0c5a47fa00184f065d9ef716573aa70b12
[ROCm/clr commit: 797d2761d4 ]
2020-07-20 11:40:49 -04:00
Sudheer Kumar
6549bf654e
[dtest] Compare hipDeviceGetPCIBusId output with lscpi output
...
SWDEV-238517 for enhancing hip unit tests
Change-Id: I5507dd19fc22e7f8806655eefaa69fbd61c98f56
[ROCm/clr commit: 94c4462848 ]
2020-07-20 04:31:48 -04:00