Jaydeep Patel
4b3f70a6b4
SWDEV-380405 - Make hipMemcpyPeer async w.r.t. host and let dst device wait till copy finishes by src device.
...
Change-Id: Idbe25404d0c31b93436e190b548f12f327f47a04
[ROCm/clr commit: 5f231c55e3 ]
2023-02-07 00:34:45 -05:00
German Andryeyev
395d85ac2e
SWDEV-353281 - Switch Graph to operate with stream
...
MemPool was designed to use hip::Stream, but graph implementation uses
amd::HostQueue. Hence switch graph to hip::Stream management.
Change-Id: Ia319389de45e4c3c6043d17473279a6f27a13140
[ROCm/clr commit: 0fe03b6494 ]
2023-02-06 10:11:55 -05:00
Matthew Arsenault
716ddd1d98
SWDEV-1 - Reapply "SWDEV-1 - Directly call __builtin_amdgcn_fence"
...
This reverts commit 7b3cf2e9d7 .
Resubmit after https://github.com/ROCmSoftwarePlatform/rocPRIM/commit/ba8a86166ec9c4cf3a40ecde7017feeb1484e7f6
Change-Id: I194559484a5a0d7f4443381fe77218ea73d2def9
[ROCm/clr commit: 8c7391398a ]
2023-02-03 10:54:27 -05:00
Jaydeep Patel
5981bd5c22
SWDEV-379395 - Use getQueue as it handles null stream and flags can be combination.
...
Change-Id: If37854a9fc0fc57acb30e652953a7f283831cad8
[ROCm/clr commit: 01c3299425 ]
2023-02-01 12:03:42 -05:00
Jaydeep Patel
c6da0aa014
SWDEV-377804 - Initial commit to support hipGraphInstantiateFlagAutoFreeOnLaunch
...
Change-Id: I7a35becb6c98a6ff70264e141317d98be7457a37
[ROCm/clr commit: f9e27bcdd4 ]
2023-02-01 11:51:39 -05:00
Matt Arsenault
c77979a365
SWDEV-1 - Use proper builtins for math intrinsics
...
Change-Id: If0e420332c718abe0e9f6316c467b6b29b80b183
[ROCm/clr commit: 14477c24d0 ]
2023-02-01 07:19:45 -05:00
sdashmiz
a081bdd157
SWDEV-361623 - correct remove edge behaviour
...
- remove node dependency before checking parents
- reduce edge level acording to new value of node
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com >
Change-Id: Id4bff1684f7e0b42beeebc4d2e009bfdb507fb5f
[ROCm/clr commit: 34d087da93 ]
2023-01-31 12:48:59 -05:00
Rahul Garg
b02b0b13f0
SWDEV-379678 - Remove catch build option
...
Change-Id: Idedf5e61da1e6c7fbdd9a65e57220875a3b24112
[ROCm/clr commit: 118e7144f8 ]
2023-01-30 11:04:03 -05:00
sdashmiz
aaf82fb91b
SWDEV-374368 - dependency list should not have duplicate
...
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com >
Change-Id: I67e1c8203f10916b98408e2004e6c64e96b1933b
[ROCm/clr commit: b251c8fdb2 ]
2023-01-30 10:28:35 -05:00
sdashmiz
483f22536f
SWDEV-374389 - correct update dependency behaviour
...
- nodes should belong to graph
- num of passed dependency cant be larger than graph nodes
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com >
Change-Id: Ia6f2283546bc44edee705b0483bfe506b7b1177a
[ROCm/clr commit: 23edcd3e23 ]
2023-01-30 10:05:52 -05:00
Ajay GunaShekar
01fe2ea09e
SWDEV-372757 - Don't destroy null queue in MT
...
This reverts commit 3afbe3a155 .
Reason for revert: patch does not fix all stream hangs. So another patch was merged which fixes all issues.
Change-Id: I332d1ea29c23747b46b7667fe3e34e0ceefd2b23
[ROCm/clr commit: 1f4176062b ]
2023-01-25 23:41:16 -05:00
sdashmiz
e72db4a957
SWDEV-379151 - correct error status
...
- if the stream is null ptr and there is capture ongoing return an error
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com >
Change-Id: Iff33e4aebc253fb7bb56daf5c455722d726c6705
[ROCm/clr commit: a4dd362a1d ]
2023-01-25 15:06:38 -05:00
Rakesh Roy
e8b7717abb
SWDEV-375004 - Fix rocprim test failure
...
- For !__HIPCC_RTC__ case, operator<<(std::ostream& os, const hip_bfloat16& bf16) calls itself in endless recursion
- Convert hip_bfloat16 to float to fix this
Change-Id: I252a656817550caf43c587cebf461474f12b8c0c
[ROCm/clr commit: 8b00f56577 ]
2023-01-25 00:01:02 -05:00
Jaydeep Patel
43565f0496
SWDEV-379125 - Check only if there is attr set for graph kernel node.
...
Change-Id: I8768b33ad27e75eb753d99d682edf60b31b240df
[ROCm/clr commit: 70bfc6145b ]
2023-01-24 09:51:12 +00:00
Satyanvesh Dittakavi
d96114fd22
SWDEV-372396 - Address the errors in hiprtc headers with -Werror
...
Change-Id: Ied553216dfc2ff8b5c72de617cb47b6ca8f52063
[ROCm/clr commit: 0ccd41050b ]
2023-01-23 17:25:42 +00:00
Maneesh Gupta
7b3cf2e9d7
SWDEV-378321 - Revert "SWDEV-1 - Directly call __builtin_amdgcn_fence"
...
This reverts commit 1de57c06b3 .
Reason for revert: Root cause for regression SWDEV-378321
Change-Id: If3116c5ad1a23db1840e5fc019d2fe997b37d229
[ROCm/clr commit: 5b39c346ad ]
2023-01-19 13:01:27 -05:00
German
e95245aaf2
SWDEV-377991 - Rmove Liquidflash logic
...
Change-Id: I1c2169c1a8c7449c59cc15074af039c48fc1ef31
[ROCm/clr commit: b685930bee ]
2023-01-19 09:58:07 -05:00
pghafari
5fa035dbd4
SWDEV-369551 - invalid stream check in hipMemcpy
...
Change-Id: I6762af93cd566a38f7d04994b87a76e744b19043
[ROCm/clr commit: 76b0b4d635 ]
2023-01-18 20:02:28 -05:00
pghafari
de904d8716
SWDEV-374381 - updated error check
...
for hipGraphAddMemcpyNodeFrom/ToSymbol
Change-Id: I4de063a5a9307148bf4da0e8a2edc1d3b01a8b9a
[ROCm/clr commit: d491ee7385 ]
2023-01-18 19:59:32 -05:00
Anusha GodavarthySurya
f681854310
SWDEV-376407 - Track manual nodes added during capture
...
Change-Id: Ic44f842c303eed0eff08b6ee5922976d96a52c71
[ROCm/clr commit: d422ad0ec1 ]
2023-01-18 03:27:41 -05:00
Anusha GodavarthySurya
e333f99dec
SWDEV-373173 - Handle kernel compilation WGP/cumode in occupancy calculation
...
Change-Id: Id0225683a578c3275a35f8f807a57b6379746392
[ROCm/clr commit: ce8f441c14 ]
2023-01-18 03:27:30 -05:00
Anusha GodavarthySurya
a7a887ab05
SWDEV-373173 - Correct max VGPRs and VGPR Granularity in Occupancy calculation
...
Change-Id: I36010f3991f809c61f0c4388b8daa8d3e9b8cc9d
[ROCm/clr commit: 27c5189bca ]
2023-01-18 03:27:21 -05:00
Maneesh Gupta
a2a134968c
SWDEV-1 - Use VERSION file instead of hipvars.pm for versioning
...
Change-Id: I9974aec388c9a89f4ed79a974dea99ec1e8257b8
[ROCm/clr commit: e0516a6984 ]
2023-01-18 02:39:12 -05:00
pghafari
5959057299
SWDEV-374361, SWDEV-374362, SWDEV-374364, SWDEV-374365 - error checking
...
Change-Id: I0b729976ed794239fa54e1ad2965133b319d2f5d
[ROCm/clr commit: a9cbff0a49 ]
2023-01-17 23:10:22 -05:00
sdashmiz
11f97823cd
SWDEV-374374 - memset memory mapping should not change
...
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com >
Change-Id: I4cc3b1ac5a59fe5c9ae060f4d5d189dc63ca450c
[ROCm/clr commit: 52bf09aac6 ]
2023-01-17 14:41:24 -05:00
German Andryeyev
36d240ff89
SWDEV-355281 - Use offset from FindMemObj
...
Change-Id: I53ce2c1ef2e3a580b8b12312b693c2476f0885d2
[ROCm/clr commit: 39db0ec297 ]
2023-01-17 10:01:49 -05:00
pghafari
829bbe4933
SWDEV-369570 - MaxSharedMemoryPerBloc check
...
Change-Id: I74d216a5d9316758bb611eb241ba85222194ce8b
[ROCm/clr commit: 2e60edbe0e ]
2023-01-17 07:52:15 -05:00
pghafari
7261e8fa39
SWDEV-374380 - hipGraphAddMemcpyNode1D errorCheck
...
Change-Id: I50501de139e3898783c1cc9c91c5ed3ed2564f3b
[ROCm/clr commit: 236a725d67 ]
2023-01-17 07:49:23 -05:00
Satyanvesh Dittakavi
f94a34e128
SWDEV-373231 - Support hip atomics header with HIPRTC
...
Change-Id: Ic500dfb66ed5ef0c4b672dfd0595409d0e8caaf5
[ROCm/clr commit: 842b9d0b56 ]
2023-01-17 00:31:07 -05:00
sdashmiz
42a08627ef
SWDEV-360031 - check for stream capture finish.
...
- stream capture should be done before any sync APIs.
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com >
Change-Id: I3d65f67ee68777be71f97f48d460ccaefdd4e1af
[ROCm/clr commit: be966acb0c ]
2023-01-16 13:09:24 -05:00
sdashmiz
4f74c83b94
SWDEV-374390 - add correct error
...
- null stream should return error on host function
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com >
Change-Id: I17659ad7af0814091dbf6cb4a2cb2644514551f0
[ROCm/clr commit: 338307ef2c ]
2023-01-16 13:08:23 -05:00
Jaydeep Patel
d74f646173
SWDEV-374360 - Handle free for external memory.
...
Change-Id: Ibfaf30eef371499a3c1f931a73cce76c83d8f7f3
[ROCm/clr commit: 7eecbdd9db ]
2023-01-16 13:07:01 -05:00
Jaydeep Patel
ffcff763d2
SWDEV-375366 - SWDEV-375351 - Handle HtoH case for graph mem cpy impl.
...
Change-Id: I5a8c3c3c22db045f714b0443b8d70a8c6b4a8cea
[ROCm/clr commit: fdecc1b351 ]
2023-01-16 11:37:57 -05:00
Jaydeep Patel
c9fa0aad8b
SWDEV-369961 - Add mappings for enum hipGraphInstantiateFlags and types.
...
Change-Id: I03fa8221684b08a0c44e12aaf2eb7ce5281be629
[ROCm/clr commit: 8bd45ae5da ]
2023-01-15 12:17:21 +00:00
Ioannis Assiouras
1974ef22cf
SWDEV-374403 - Added input sanity checks for hipLaunchCooperativeKernelMultiDevice and hipExtLaunchMultiKernelMultiDevice
...
Change-Id: Iebef5ae68439d87549fdf9559de51c6c79c64392
[ROCm/clr commit: 0d2e0d1621 ]
2023-01-13 19:48:42 -05:00
Matt Arsenault
967ad97993
SWDEV-1 - Use proper builtin for groupstaticsize
...
Remove the now empty header for asm intrinsic wrappers.
Change-Id: I63495dc6ed899b1ea823062f2be2fb064af8d4f9
[ROCm/clr commit: 8615a89f24 ]
2023-01-13 13:14:10 -05:00
German Andryeyev
9c811dc241
SWDEV-353281 - Initial support of memalloc in graph
...
Add memory allocation support in graph. Current implementation uses
cache from mempool to hold the allocations which belong to the graph.
Also the resource tracking is disabled at this moment because mempool
operates with hip::Stream objects, but graph has execution with
amd::HostQueue objects.
Change-Id: I54fe3250126d24f5a26ada975f37d429bb4ef17b
[ROCm/clr commit: eef47ca24a ]
2023-01-13 13:06:59 -05:00
Satyanvesh Dittakavi
1087f13ee1
SWDEV-369552 - SWDEV-369553 - SWDEV-369555 - Fix hipMemAdvise
...
- API should return error if the size passed is more than the
bytes allocated
- Passing invalid device should return error except for the
advise hipMemAdviseSetReadMostly
Change-Id: I1c4c15136b825215bf1e1f5ee7a9a08e5895d870
[ROCm/clr commit: 873bc03034 ]
2023-01-13 12:53:31 -05:00
Rakesh Roy
9045dbab4b
SWDEV-374369 - Fix hipGraphAdd*Node* APIs invalid behavior
...
- Return hipErrorInvalidValue if dependencies array contains a node from
different graph
Change-Id: I0dcfaaaeb50a89848ac546a0b86b185977e2128c
[ROCm/clr commit: a427d3cd10 ]
2023-01-13 12:09:42 -05:00
AravindanC
61f59d2285
SWDEV-352878 - Remove CLANG include dependencies
...
Change-Id: I46bba0fdf21a3541e08ff0c465ff076d478094f9
[ROCm/clr commit: 2228383fcb ]
2023-01-13 07:41:10 -08:00
Satyanvesh Dittakavi
6035b0e214
SWDEV-376386 - Support __hip_move_dpp with hipRTC
...
Change-Id: If9a6fe23d6df143a7f522823fca7183416048c6a
[ROCm/clr commit: c193503722 ]
2023-01-13 00:05:31 -05:00
Anusha GodavarthySurya
57d351bed6
SWDEV-364576 - use hipLimitMallocHeapSize to set initial heap buffer size
...
Change-Id: I2f1d3b229bc28571ce6c55f93f8ef73dda516b50
[ROCm/clr commit: 33eef6d95e ]
2023-01-12 23:34:29 -05:00
Matt Arsenault
1de57c06b3
SWDEV-1 - Directly call __builtin_amdgcn_fence
...
The device libraries provide hip.bc, which only contains
__atomic_work_item_fence. This is identical to OpenCL's
atomic_work_item_fence, except prefixed with __. This only existed to
switch over the unknown values the OpenCL function provides. There's
no need to go through the wrapper here since it's using constant
values, and without it we can drop one of the bitcode libraries. I'm
guessing this was a holdover from before the fence builtin existed.
Change-Id: I8b04614f96194eafddeba295215c12c3856e6c42
[ROCm/clr commit: 3e54bad5b3 ]
2023-01-12 12:04:09 -05:00
Matt Arsenault
ac1201f598
SWDEV-1 - Remove unused intrinsic declaration
...
The uses have moved to using the proper __builtin_amdgcn_ds_swizzle.
Change-Id: I82159ebf41fb91ac6a9d07b33311e569dce5e754
[ROCm/clr commit: 91b851fad7 ]
2023-01-12 12:04:09 -05:00
Ioannis Assiouras
bfa0b9f844
SWDEV-374405 - Added OpenGL interoperability APIs for nvidia
...
Change-Id: I1d6ad5c610ab6ee805bd07dacea316b3242ed129
[ROCm/clr commit: e97dcfbe48 ]
2023-01-12 05:22:14 -05:00
Jaydeep Patel
d12bfbbb4d
SWDEV-374383 - Check from and to if no of dependencies are more than 0, Check from and to node are belongs to graph pass in.
...
Change-Id: I0cdafb497147f9ac7f84d40203a21e9eefd595a3
[ROCm/clr commit: bb486b04c6 ]
2023-01-11 11:43:22 -05:00
Rakesh Roy
87db95fad7
SWDEV-368881 - Implement hipModuleLaunchCooperative* APIs
...
- Add implementation for hipModuleLaunchCooperativeKernel, hipModuleLaunchCooperativeKernelMultiDevice APIs
Change-Id: I2a80264e458549211c34b413397a08b57d021147
[ROCm/clr commit: b321e2afc6 ]
2023-01-10 13:50:20 -05:00
Ajay
433f99686b
SWDEV-369556 - Unit_hipMemGetAddressRange_Negative cases
...
- return value for invalid device ptr
- offset greater than size
Change-Id: I8bcf878c3c1c0a29836ea664bfb9048e1114b74e
[ROCm/clr commit: c6a7ce66bb ]
2023-01-10 06:57:54 -05:00
Maneesh Gupta
4f5f6a3696
SWDEV-345571 - Revert changes for warp shuffle functions
...
This reverts commit 609598cd15 and 149afa9444 .
Change-Id: I548f15055417453a7fd323a481ed410598830521
[ROCm/clr commit: 6c53a670c1 ]
2023-01-09 22:56:52 -05:00
Jiabao Xie
3ab54974dc
SWDEV-374375 - moving nullptr check into hipGraphNode::isValidNode
...
Change-Id: Ic9264521d6f0b36ffc948ca9feeaa70eb6e98eb6
[ROCm/clr commit: 03200953e5 ]
2023-01-09 16:27:38 -05:00