cd9db5a2fa
SWDEV-505277 - Adds hipStreamBatchMemOp in the enum of hcc_map
...
Change-Id: I6e58dfbe4ba13db8717edc36020fefabc9ddbe23
2024-12-18 05:38:58 -05:00
fa63919a63
SWDEV-504340 - Move cast of cl_mem inside the condition
...
Change-Id: I9c91f5d945a8d8bd2b2f55e3d11ede66afe4eef7
2024-12-17 12:58:12 -05:00
e3efce20be
SWDEV-504650 - Switch to shared_mutex for events
...
Use shared mutex for events validation
Change-Id: Iff291c758d9edd65717c506150f3b9d39e5306ba
2024-12-17 11:04:58 -05:00
fccf0fa2f0
SWDEV-377518 - Fix bf16/fp8 header to be compileable with hiprtc
...
Change-Id: I2093a39d79a46da7e102266c04c2a71e03dcb88e
2024-12-17 08:57:15 -05:00
e8b2fdab96
SWDEV-483134 - Remove hipExtHostAlloc API
...
Change-Id: I60777ef5c56b60dd8100d0d794ca10fb3b96a555
2024-12-16 17:13:49 -05:00
078fe7e5de
SWDEV-503764 - Add wptr and rptr to ClPrint for dispatch barrier methods
...
- added wptr and rptr to ClPrint in dispatchBarrierPacket and dispatchBarrierValuePacket
Change-Id: I8a62289deb23c9f657a9b0ac6138bb55eafecba2
2024-12-16 16:45:30 -05:00
a808c4b23a
SWDEV-489255 - Update stack size limit in rocvirtual
...
Change-Id: I2aac9d211f64b3d6c121d8b010d215dcbdeac3aa
2024-12-16 09:30:39 -05:00
13e2e797c0
SWDEV-469422 - Derive GraphExec from Graph and ChildGraphNode from GraphExec
...
Change-Id: I54d67a1665355579bc249d8ff4f9806e9ee14588
2024-12-16 00:43:57 -05:00
3c863dad91
SWDEV-502543 - Update doxygen to surface funtions and Coop Group API
...
Change-Id: Id4df63b8ae64a1113f85d89aa250ac9f7cc8b9bb
2024-12-14 14:11:37 -05:00
8ffb1430dd
SWDEV-496037 - add Strix and Strix Halo to ocltst runtime test
...
Change-Id: Ia21afddf5223ecd132a06f37bb430961fb7a9341
2024-12-13 20:19:48 -05:00
f1c05e9026
SWDEV-421020 - Adds hipGraphAddBatchMemOp, SetGetParams and execSetParams APIs
...
Change-Id: Ieccecfe6173cc68fd3c01f86c99f7cc09fe194a3
2024-12-13 06:23:39 +00:00
93f1e8ff60
SWDEV-301667 - Clear dispatch indicator signal flag
...
Change-Id: I9028df0bb73289791d169e7f064a1d0f615236a5
2024-12-12 21:20:05 -05:00
03dbcd8ca7
SWDEV-440866 - [hip-roclr] Adds support to batch memory operations APIs
...
Change-Id: I5ac63a6626af8c2b4ac382c52dfe1aaf0b3716b8
2024-12-12 19:29:24 -05:00
3ad8f1b811
SWDEV-496667 - Support gfx9-4-generic target
...
Support gfx9-4-generic target to cover mi3XX.
Support features sramecc and xnack in generic target.
Improve some code formats.
Add more log on compiler.
Change-Id: I6b3c6af55c60cffd43ce6f17b75998f751b75713
2024-12-12 14:43:39 -05:00
537f2fffc9
SWDEV-301667 - Improve logging for hip_memory
...
Change-Id: Id624b2c91e6b701bc0ee561a0c193f2c66654890
2024-12-12 14:42:00 -05:00
cfcc743824
SWDEV-499997 - Unify ManagedBuffer and KernelArg buffer implementation
...
Change-Id: I95421c87904dd62d7ee214539a57c7bda1097ff4
2024-12-12 12:56:23 -05:00
28cbf2bc4f
SWDEV-469422 - Refacor childgraph node
...
Remove static functions in graph
Change-Id: I4df94915f81f250acaea60398aea32ef0ed658e2
2024-12-12 12:38:24 -05:00
c8afd7109d
SWDEV-496544 - Sync with stream if it is different than srcMemory stream.
...
Change-Id: I9c0f94a9531555278a51202ec7203961e1344c2e
2024-12-12 06:17:39 -05:00
23c21d5181
SWDEV-503761 - Reintroduce save-temps path for OpenCL
...
Change-Id: I5e111047242aed7d982b7a25c11ab52293af639d
2024-12-11 13:01:29 -05:00
072f94c204
Revert "SWDEV-491314 - enable _sync() functions with 64-bit mask argument"
...
This reverts commit 4e2fd192eb
2024-12-11 11:16:48 -05:00
14f58fc74d
SWDEV-501757 - Clean-up signal creation
...
Use hsa_amd_signal_create() if settings.system_scope_signal_ is true.
Change-Id: I6d440155dfbcd5bf03658583a93827cb1c56537c
2024-12-11 09:57:50 -05:00
98b33886cd
SWDEV-503299 - Do not use operator to check for nan
...
Some libs use __HIP_NO_HALF_OPERATORS__ and __HIP_NO_HALF_CONVERSIONS__
which results in operators being hidden and can cause errors.
Change-Id: I83c194d7d727cba30b46d7c296f7d396549f5fca
2024-12-11 00:33:44 -05:00
def369a010
SWDEV-489099 - Deprecate AMDGPU_TARGETS in lieu of supporting GPU_TARGETS
...
Change-Id: I184cd18e47b1618bcad0fadc8984de54d2a00a9b
2024-12-10 12:38:37 -05:00
03e4057fce
SWDEV-496392 - Remove references to ASIC Polaris22 in pal::Settings and pal::Device (it is being retired)
...
Change-Id: I6318abf3e46ed250d087a3d2266d2ae3d4c8c000
2024-12-10 06:01:32 -05:00
7c9c7a6332
SWDEV-498728 - Add backward compatibility for deprecated package rocm-opencl-icd-loader
...
Change-Id: Ic659639e3bb55bd90bd50acf28d8079ff7b084bc
2024-12-09 16:27:43 -05:00
f4b9d3b7bd
SWDEV-501757 - Use signals without interrupts
...
In active wait mode use signals without interrupts by default and switch
to the interrupts only if a callback is required.
Change-Id: Ibcde8f7d44c70f8fb8fa5e0a7fdd8b08a2982a8e
2024-12-09 15:16:15 -05:00
7d763fb803
SWDEV-498474 - APU: when total allocated is greater than 75% of invisible
...
Only when memory type is Local and the invisible memory is +ve
Should also fix SWDEV-490991
Change-Id: I78a4925a234ba90c63909bde5b7dc217568b4de3
2024-12-09 12:06:57 -05:00
89dfdc4dbe
SWDEV-490860 - Do signal_is_required detection post graph schedule
...
Change-Id: Iaf1067a811aeac3d16c08de954036e219b545e07
2024-12-09 03:57:44 -05:00
0d4823ff88
SWDEV-502532 - Exit graph launch in case of empty graph.
...
Change-Id: Ifb6ab14ca6810cbc1c9e38c59d1d9e7d367358d9
2024-12-07 12:27:53 -05:00
e80442fdbf
SWDEV-497759 - Fix memObj offset computation for hipHostRegister on Windows mgpu
...
On Windows, hipHostRegister may add a single object in the MemObjMap
that maps to memory that is allocated on different devices.
This change ensures that the offset that is returned from
getMemoryObject() is computed relative to the memory that is allocated
on the current device.
Change-Id: I5fd3af200bf6f4926fdeaea12dcb9d0154d3a843
2024-12-05 16:18:10 -05:00
78f62d3230
SWDEV-495609 - Change include path for rocclr/utils
...
- Header files inside rocclr/utils when included from hipamd or opencl should be included as #include "rocclr/utils/xxx.h" instead of "utils/xxx.h"
Change-Id: Ic0760c33b9d091f5620dec67e5482c9698d22093
2024-12-05 11:44:20 -05:00
5122b8c999
SWDEV-501779 - add correct function qualifier for fp16 functions
...
Some functions were __device__ only, but should be __host__ and
__device__, changed them to __HOST_DEVICE__.
Some functions were __HOST_DEVICE__ but were using ockl functions,
changed them to __device__ only.
Change-Id: Ife9e7abe60415bda68f5f9a101e6e7c39ad51064
2024-12-05 10:17:18 -05:00
f2dba978f5
SWDEV-490860 - Include signal_is_required in dot file dumps
...
Change-Id: Iec4b433b11fbecb71a4ce68beb7d6f681d25b8e6
2024-12-05 04:51:59 -05:00
231b2410a0
SWDEV-492049 - Remove the handle of Phy Mem from Memobj
...
The hipGraph will use VMM by default when allocating memory.
However, the handle of Phy mem has been added to Memobj by default.
Since the Memobj will track the whole address range from handle to
handle + size, this needs the system to reserve the whole address
range. If the system range have not reserved by the system, then it
will have the potential issue that clr finds the Memobj incorrectly.
This patch removes the handle from the Memobj to fix this potential
issue.
Change-Id: I2da38e6b2d11d0d48e1afe66c46899500c290624
2024-12-04 19:39:52 -05:00
7863eb92dc
SWDEV-497145 - Use rocr copyOnEngine API for staged copies
...
- Refactor blit code and clean ASAN instrumentation
- Use unified function for rocr copy
- Enable shader copy path for unpinned writeBuffer/readBuffer paths
- Set GPU_FORCE_BLIT_COPY_SIZE=16 which means we will use BLIT copy for
pinned copies or unpinned H2D/D2H copies < 16KB
Change-Id: I42045cca79234b340dbf53dafb93044199736ae4
2024-12-04 13:38:13 -05:00
efd3ea4b30
SWDEV-501403 - Switch to std::shared_mutex for streamSetLock
...
Shared mutex allows to have access to the list of streams from
multiple threads at the same time.
Change-Id: Ibee64b846cde03321d5b17dbee2829c0bab7e7d6
2024-12-04 12:06:51 -05:00
063f7ef32a
SWDEV-485945 - use union to convert values
...
this shows up in some compilers as warnings.
Change-Id: I862cd6baf2edb8161757adc54abb787530489481
2024-12-04 11:15:03 -05:00
6c755a4116
SWDEV-485672 - LOG_INFO corrected to LOG_ERROR for errors
...
Change-Id: I8ab5f2117dfd7725bd4ed8b178e370096aa31018
2024-12-04 01:18:01 -05:00
b89977d518
SWDEV-469422 - Always schedule graph nodes
...
Change-Id: Icc636527fa19e7bf3eb111bc4b1bb9a5f9acff73
2024-12-03 23:44:23 -05:00
6611cc015d
SWDEV-494149 - Improve hipGet/Set Device
...
Change-Id: If8975687a3ba9caadafc48a0066f19a4ebaab9e2
2024-12-03 13:36:38 -05:00
2ca644cf22
Revert "SWDEV-440866 - [hip-roclr] Adds support to batch memory operations APIs"
...
This reverts commit bd5d8e9baf
2024-12-02 16:46:12 -05:00
08aee16573
SWDEV-499794 - Update AQL packet after updating GraphNode
...
Change-Id: I332d70bdf42a276894a548a02d636e370c2ca08c
2024-12-02 12:29:35 -05:00
f0063ba8da
SWDEV-485827 release hostcall listener memory regardless of thread status
...
The early return if the thread is not alive causes memory leaks.
Neither doorbell_ or urilocator are released if the thread is not alive.
This change alters the logic so regardless of the thread status the
HostcallListener releases its memory.
Change-Id: Ie912360ec0e2ee257de9937b1a8d7375e6aebd83
2024-12-02 04:42:56 -05:00
4e2fd192eb
SWDEV-491314 - enable _sync() functions with 64-bit mask argument
...
Change-Id: Ieb13a9e1b2fc49ff225a05a51056d1212d95ae57
2024-12-01 10:16:59 -05:00
bd5d8e9baf
SWDEV-440866 - [hip-roclr] Adds support to batch memory operations APIs
...
Change-Id: I449ffca44bbb04d13348d112e896d603c70fd485
2024-11-30 17:54:32 -05:00
c47f9dda58
SWDEV-469422 - Cleanup graph code remove parallellists and nodewaitlists
...
Change-Id: I00c7b2894333bd13d47b913d3fcdd6e1ffcb741f
2024-11-30 04:40:51 -05:00
f1f4f40c5b
SWDEV-447973 - Support generic targets
...
Change-Id: I32db83843e45e0f013591493aafd7a532c881e16
2024-11-29 10:12:10 -05:00
b75b0d9a53
SWDEV-494612- Add capture support for hipLaunchCooperativeKernel
...
Change-Id: I6b3c6af55c60cffd43ce6f47b75998f750b75703
2024-11-29 08:17:41 -05:00
fb7ad8361c
SWDEV-489084 - Update max streams for graph
...
Change-Id: I6d0992b2e80ebf3184911593a4f3574327b2e9c3
2024-11-29 08:16:16 -05:00
06e6561eb5
SWDEV-489084 - Avoid using queue colliding with the graph launch stream
...
Change-Id: I3ecaf8836c8e0883441275139041c702aba0937e
2024-11-29 08:15:58 -05:00
Sourabh Betigeri
Saleel Kudchadker
German Andryeyev
Jatin Chaudhary
Ioannis Assiouras
Pengda Xie
Anusha GodavarthySurya
Istvan Kiss
Todd tiantuo Li
Tao Sang
Jaydeep Patel
Maneesh Gupta
Gerardo Hernandez
Ranjith Ramakrishnan
Ajay
Branislav Brzak
Marko Arandjelovic
Shane Xiao
Jimbo Xie
Aidan Belton-Schure
Sameer Sahasrabuddhe
Vladana Stojiljkovic