562be5c864
SWDEV-401847 - Update atomicMin/Max for float and double.
...
Change-Id: Ib5be459b8a24f0739e299ed12c9f877f8baa02b1
[ROCm/hipother commit: 01dc9ce02e
2023-05-29 14:27:16 -04:00
790f51e1f3
SWDEV-401850 - Fix hipCreateChannelDescHalf APIs
...
- Address ChannelDescHalf1 and ChannelDescHalf2 APIs not returning
correct Channel Descriptors in amd headers
- Add missing hipCreateChannelDescHalf APIs in nvidia headers
Change-Id: I558847425c2459d3cde329ca2e926d882bb0a9dd
[ROCm/hipother commit: e5e4a16996
2023-05-25 04:36:24 -04:00
efaf0b2aed
SWDEV-367537 - Add __hip_bfloat16 and vector definitions to match __nv_bfloat16
...
Change-Id: I1c0f9f5f278c2c3b4e175d9f08831ba458ed856e
[ROCm/hipother commit: ea190a2571
2023-05-11 04:45:13 -04:00
5b0467b9a3
SWDEV-394488 - cudaStreamGetCaptureInfo_v2 is undefined in CUDA 12.0
...
hipStreamPerThrdCompilerOptn.cc test fails to build with cudaStreamGetCaptureInfo_v2
in CUDA 12.0.
fix was to change runtime API cudaStreamGetCaptureInfo_v2
to Driver cuStreamGetCaptureInfo_v2
Change-Id: I44a0110770d3246f5345092acae301c9a2f6d520
[ROCm/hipother commit: e5e95946a1
2023-05-05 10:10:07 -04:00
04dd3d66b0
SWDEV-393199 - Added new include file for opengl interop mappings for nvidia
...
Change-Id: I2e955a9dd06539939188a5f2e1dde4f173af1202
[ROCm/hipother commit: 0f95d1d31d
2023-05-03 19:43:24 -04:00
d15029d3c0
SWDEV-390170 - guard deprecated runtime apis on CUDA 12.0
...
Change-Id: I4ab967c276dfa98f75d508b9484e8d1324b4e4a7
[ROCm/hipother commit: cedc75f7e8
2023-04-05 16:36:53 +00:00
f32fc707da
SWDEV-387173 - Added cuda mappings for hipGraphMemAttr* enums
...
Change-Id: Ie5485ec518c6d107901a429ac3614e22caa9196b
[ROCm/hipother commit: cafad8f51e
2023-03-09 08:08:06 -05:00
60d87305c5
SWDEV-330658 - Added flag hipHostRegisterReadOnly
...
Change-Id: Idb59dc6187e99512546dfeafde44c08ae85f6057
[ROCm/hipother commit: 81717d6329
2023-03-09 00:20:24 -05:00
09c16355f4
SWDEV-372153 - Add hipStreamGetDevice Implementation
...
Change-Id: Ifd1f13e311e8221ca6d94cf27f9131eb97678067
[ROCm/hipother commit: 8ce9ec60f8
2023-03-01 04:35:49 -05:00
5fe52ac30f
SWDEV-384100 - HIP support for CUDA 12.0
...
Apps are failing to build due to undefined deprecated texture APIs
Change-Id: I1fb64adc4bc0ba6ee6ecaa65d54b34da0327e6a3
[ROCm/hipother commit: 3fc34b8bf5
2023-02-21 12:23:48 -05:00
7323f7a4cf
SWDEV-338733 - Implement hipArrayGet* APIs
...
- Add implementation for hipArrayGetInfo, hipArrayGetDescriptor &
hipArray3DGetDescriptor APIs
Change-Id: I181a472066006bc3bd0d987408ea67e218310983
[ROCm/hipother commit: bb635fd5ff
2023-02-16 23:04:29 -05:00
ea22aea19b
SWDEV-369961 - Add mappings for enum hipGraphInstantiateFlags and types.
...
Change-Id: I03fa8221684b08a0c44e12aaf2eb7ce5281be629
[ROCm/hipother commit: cf18c50814
2023-01-15 12:17:21 +00:00
a4f481f796
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/hipother commit: ec92380b66
2023-01-13 13:06:59 -05:00
dc62bdd371
SWDEV-374405 - Added OpenGL interoperability APIs for nvidia
...
Change-Id: I1d6ad5c610ab6ee805bd07dacea316b3242ed129
[ROCm/hipother commit: 9c531b6a67
2023-01-12 05:22:14 -05:00
d0cb83e7a8
SWDEV-368881 - Implement hipModuleLaunchCooperative* APIs
...
- Add implementation for hipModuleLaunchCooperativeKernel, hipModuleLaunchCooperativeKernelMultiDevice APIs
Change-Id: I2a80264e458549211c34b413397a08b57d021147
[ROCm/hipother commit: 9808909320
2023-01-10 13:50:20 -05:00
bb291a85fd
SWDEV-369042 - updating to GetError to match cuda
...
Change-Id: I0ec2330443b26cb1c8cedba942f31fb1267cd09d
[ROCm/hipother commit: c1fe38aff6
2023-01-06 05:33:41 -05:00
e30e8d2a62
SWDEV-369955 - Support hipDrvGetErrorString and hipDrvGetErrorName on Nvidia headers
...
Change-Id: I54cfa8faed2f45f310682625f94952aa3c2ee27f
[ROCm/hipother commit: b222f4f13d
2022-12-02 11:52:07 -05:00
16acc1ce57
SWDEV-368477 - hipOccupancyMaxPotentialBlockSizeVariableSMemWithFlags nvidia mapping
...
SWDEV-369618 - hipOccupancyMaxPotentialBlockSizeVariableSMemWithFlags invalid flag scenario
Change-Id: I2edbab5d3eeacf80bc215f0a77a46d733f4b4ec9
[ROCm/hipother commit: d1fc24e583
2022-11-29 12:38:39 -05:00
2c1a3a1dbc
SWDEV-348820 - Fix hipPointerAttribute_t incompatibility
...
- Use hipPointerAttribute_t.type instead of hipPointerAttribute_t.memoryType
Change-Id: I3bf1c0758bdae987213ba7de62247dd420ef4cc9
[ROCm/hipother commit: 6767dbd769
2022-11-29 00:53:24 -05:00
42dc455ced
SWDEV-369557 - Added enum mapping for external[Semaphore/Memory]HandleType
...
Change-Id: If1b95b084f7fa312e6cdad4ccf433127671a35ce
[ROCm/hipother commit: 9f1655969b
2022-11-28 13:32:31 -05:00
5222cc89fd
SWDEV-325711 - Added few graph API support for nvidia path
...
Change-Id: I4eb6e77220e201c85fdaa3a004590c4c5d5ea6c6
[ROCm/hipother commit: b95b2ba704
2022-11-25 09:08:32 -05:00
540e6be0a4
SWDEV-366088 - Default case returns -1 as enum value.
...
Change-Id: I8dd5cac254c99e4de9f6d492e71060e39b069507
[ROCm/hipother commit: 6ce46810d9
2022-11-15 22:12:12 -05:00
d78774a328
SWDEV-362199 - add vulkan interop apis in nvidia
...
Change-Id: I6c5de6174a80be862b6d869927ed704dd1421b41
[ROCm/hipother commit: b8532f6dd4
2022-11-04 16:49:21 -04:00
965272f038
SWDEV-349226 - Add the hipDataTypes mapping to cudaDataTypes
...
Change-Id: Iadeb2675adf314c02dcc5d722581678ec36f8632
[ROCm/hipother commit: 4d63e5cabe
2022-10-14 06:44:30 -04:00
65104e694f
SWDEV-356551 - Add support for device attributes.
...
Change-Id: Ic0f2b9cb510b027e0b8d1b3fb6d984e0d2600b15
[ROCm/hipother commit: 5f582f5615
2022-10-11 12:06:44 -04:00
a6bd98310e
SWDEV-360001 - Add hipLaunchHostFunc in Nvidia path
...
Change-Id: I5c733b416642174a089a86c420b03e8775e206d6
[ROCm/hipother commit: 9ef5e96734
2022-10-11 06:18:52 -04:00
589c26c999
SWDEV-356651 - Merge hipJit options between hiprtc and hip runtime.
...
Change-Id: I73b66eaf081ce63b6b7ce45f71880877953cf81c
[ROCm/hipother commit: 17c0b17346
2022-09-22 12:53:58 -04:00
9e82256533
SWDEV-353672 - add missing defines for cuda
...
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com >
Change-Id: I8e2e0a3b8845d1920e6a40e3ed676c657ab90276
[ROCm/hipother commit: f7c0a424fb
2022-08-29 08:59:23 -04:00
0e161d5520
SWDEV-325711: Add userobject functions for graph
...
- add user obj APIs for creating release and retain of user onbjects
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com >
Change-Id: I0bf2999c77e44269565b27c31c7c1461f8a160a2
[ROCm/hipother commit: d6cee2332a
2022-08-17 14:31:43 -04:00
94bb7e57d4
SWDEV-341992 - Add hipRefTex data types mapping on nvidia path
...
Change-Id: Icec2ec110f2647bfb4c571ff16dd3852cfc9fe4a
[ROCm/hipother commit: 5a485e3393
2022-08-10 02:45:25 -04:00
17090bc850
SWDEV-338376 - CUDA VMM API mappings
...
Change-Id: I47595dbf57fcce352d23842dbbc2b98b4ec97fb5
[ROCm/hipother commit: 013fa23988
2022-07-27 02:35:13 -04:00
942141e39a
SWDEV-347345 - add missing func for nvdia
...
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com >
Change-Id: Ie2a70778d01948083d724a9d20e1f17541178ec5
[ROCm/hipother commit: bd1ffabfc1
2022-07-24 21:38:14 -04:00
290e599ce3
SWDEV-344135 - Enable stream ordered memory and mempool tests
...
- Implement CUDA mappings for stream ordered memory allocator and memory pool APIs
Change-Id: I2434118ff043527ec7c3389cd5175e1e21d032bf
[ROCm/hipother commit: a44d54f279
2022-07-21 00:05:39 -04:00
5e4c00c4fb
SWDEV-346448 - fix the missing defines
...
for nvdi compile
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com >
Change-Id: Ifd0e50a6f8636609fa61680cc719a2f82fb1320c
[ROCm/hipother commit: b875936c5d
2022-07-20 11:31:50 -04:00
a369b4730b
SWDEV-333033 - add safe and unsafe atomic min and max including gfx940 and add missing nvidia support
...
Change-Id: I829a0a5fd49c510e77eabbcb92d1a415ef6b5a4c
[ROCm/hipother commit: 13d1a1b0e4
2022-07-14 10:22:42 -04:00
d163746f0c
SWDEV-321698 - add new enum entries to match cuda
...
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com >
Change-Id: If0a17b3234fd90b1f553a075e02280258000d36a
[ROCm/hipother commit: e7f7073e96
2022-07-13 10:22:06 -04:00
0dbbf52d8e
SWDEV-341174 - Fixed typo
...
Change-Id: Ib605390bd17d014539d57488f0633db44d7c1fad
[ROCm/hipother commit: 038114c098
2022-07-11 12:24:18 -04:00
1dfcc22d96
SWDEV-345723 - Fixed test build issue
...
Change-Id: Id9545c9523bfad6b5afefcdc1f705d6043094ae4
[ROCm/hipother commit: 36e047814b
2022-07-11 12:23:34 -04:00
e0ec532c31
SWDEV-341174 - fixed warning on cuda
...
Change-Id: I5d017bcc154a2985e23893c0d933438bf0fc1958
[ROCm/hipother commit: 3289950436
2022-06-27 04:49:53 -04:00
97b4b0506e
SWDEV-340918 - make hip_runtime.h work with g++
...
Change-Id: Ief3e6dc075837dfccae3defad7da696130ed05bd
[ROCm/hipother commit: 95e6e11bf0
2022-06-20 23:32:58 -04:00
77d84b36e8
SWDEV-322688 - Added hip math constants header.
...
Change-Id: I9363f82159a53c6822e943c4950fc7f6c350b326
[ROCm/hipother commit: 53190b0a65
2022-06-16 23:56:45 -04:00
e6e86b0789
SWDEV-336532 - Convert Managed Memory in hipPointerGetAttributes on Nvidia platform
...
Change-Id: I3bc6180040b6ad48f06ffdc7d01ca6fd9fe32c0e
[ROCm/hipother commit: 70c90e5975
2022-06-16 00:43:15 -04:00
d1e003b718
SWDEV-338376 - Added hipMemGetAllocationGranularity, hipMemCreate and hipMemRelease for CUDA
...
Change-Id: If8c6a00958bb39752eb202c30a7960d13de225ae
[ROCm/hipother commit: cb53be08fc
2022-06-15 23:59:56 -04:00
c6d10a3bf0
SWDEV-325711 - add Graph memory APIs skeleton
...
- hipDeviceGet/SetGraphMemAttr
- hipDeviceGraphMemTrim
- there is no memory pool for graphs currently
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com >
Change-Id: I11db76ea7ea1c7732175fc93264448052357e8dc
[ROCm/hipother commit: 9e3c65ccbd
2022-06-15 14:47:16 -04:00
92423493b1
SWDEV-321698 - remove Memcpy1D type
...
- Memcpy1D node type is not complying with cuda
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com >
Change-Id: If8113f5e699de0c62d98effc4580a2e0fee9a950
[ROCm/hipother commit: 9583567388
2022-06-13 08:31:02 -04:00
b0d323fe42
SWDEV-318833 - Get and Set attribute for kernel nodes
...
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com >
Change-Id: Ie51aa4b56661cbb8c5b4eb1dbaad327377084ffb
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com >
[ROCm/hipother commit: f874762659
2022-06-09 09:47:05 -04:00
a87da88ec3
SWDEV-319818 - Release restriction on hipResourceViewDesc
...
Let more types support hipResourceViewDesc.
Add missing symbols to fix cuda build failing.
Change-Id: Ife694cc6491427093863252e257e820b0bb4fa0f
[ROCm/hipother commit: 3fc1dca59f
2022-05-10 15:57:31 -04:00
de06c6e4b2
SWDEV-332811 - Clean up and extend HIP unsafe atomic add
...
Update HIP's unsafeAtomicAdd to:
- Compile properly even when not compiling for gfx90a
- Fall back to safe atomic add on non-gfx90a architectures
- use flat atomic add for FP64 on gfx90a, instead of dynamically
checking memory spaces.
In addition, when the compiler is passed -munsafe-fp-atomics, it
will define __AMDGCN_UNSAFE_FP_ATOMICS__. When this happens, the
compiler is requesting that the HIP headers force all HIP
atomicAdd() calls on floats or doubles to use their unsafe versions.
This patch thus causes unsafeAtomicAdd() calls when that define
is seen. This call to unsafeAtomicAdd() is also done for atomicSub(),
since that calls atomicAdd underneath. This is not done for
system-scope atomicAdd because, on gfx90a, system-scope atomic FP
add instructions would need to target fine-grained memory, which is
always unsafe.
This patch also creates safeAtomicAdd() functions for float and double.
These functions will create a standalone safe atomic, even when the
application is compiled with -munsafe-fp-atomics.
Finally, this patch adds wrappers in the Nvidia path of HIP so that
these HIP functions call through to atomicAdd there as well.
Change-Id: I8af0621d3d28ea30c9278bfeea7393d03bbdac6d
[ROCm/hipother commit: b712a51e74
2022-04-29 07:19:58 -04:00
c010936428
SWDEV-326789 - Mark texture reference APIs as deprecated
...
Change-Id: I7a39cc1f036cd185ac3aab6c5f44730f16225f69
[ROCm/hipother commit: 0f378ffc09
2022-04-05 00:04:34 -04:00
681471c584
SWDEV-311271 - Initial mempool implementation
...
HIP_MEM_POOL_SUPPORT controls memory pool support in runtime.
Currently it's disabled by default. The initial change doesn't
include: IPC, MGPU, virtual memory alloc, suballoc, defragmentation,
internal dependencies.
Change-Id: Ibed8528ebec698b045ebb247e49c0ecd6e587ed7
[ROCm/hipother commit: 3b89440eb1
2022-03-31 09:21:37 -04:00
Jaydeep Patel
Satyanvesh Dittakavi
Jatin Chaudhary
Ajay
Ioannis Assiouras
Anusha GodavarthySurya
Rakesh Roy
German Andryeyev
pghafari
kjayapra-amd
sdashmiz
Brian Sumner
Sarbojit Sarkar
Maneesh Gupta
Tao Sang
Joseph Greathouse