badf2b0880
SWDEV-301667 - Refactor graph code
...
- Remove Last graph node optimization and instead submit a barrier NOP
packet always. This simplifies the code.
Change-Id: Ied443173ba47a08b6df148ac7e3ead712acda11c
2024-05-28 06:28:17 +00:00
3d540ec113
SWDEV-457755 - Add TS only for kernel packets in the Accumulate command
...
Change-Id: I1b2f01c5763761808f49802fa117abc6306a22aa
2024-05-28 06:28:17 +00:00
51e4368723
SWDEV-459778 - Remove CPU wait for profiler
...
- No cpu wait is needed when profiler is attached, Doing this changes
the application profile when roctracer is attached.
Change-Id: I2b9cfc48d697cf5ed54bb6a240d8c12bdb079171
2024-05-28 06:28:17 +00:00
c7461436d6
SWDEV-459254 - Change return type to support values > 128.
...
Change-Id: Ib4c9a3718c4cb26b24503fe4d9af3e63e151e843
2024-05-28 06:28:17 +00:00
dd1dd86fd7
SWDEV-459254 - Overwrite cacheline size to 256 for gfx12, as it is used for kernarg alignment.
...
Change-Id: Ia6acf312ee84f6dde1c830fc21f10d3a8a9de5ee
2024-05-28 06:28:17 +00:00
2b746de6de
SWDEV-451594 - Fallback to host kernel args on older devices
...
On gfx8, gfx9 devices before MI100 and gfx10.0 or gfx10.1
none of the memory ordering workarounds for device kernel arguments
can be applied. Use host kernel arguments on these devices.
Change-Id: I9be6fbfe4b3986eb7d9f83998334df5f03fd4124
2024-05-28 06:28:17 +00:00
5bb30d7718
SWDEV-460581 - Fixed memory leak in hipIpcCloseMemHandle
...
hsa_amd_ipc_memory_detach is called with an invalid mapped pointer.
Changed to pass the svm pointer of the owner memory instead.
Change-Id: I8203c6e2d718efb8ca3b028309bc78caff8d4c7d
2024-05-28 06:28:17 +00:00
87619af175
SWDEV-459556 - Revert "SWDEV-442724 - Remove uses of -amdgpu-early-inline-all to enable function calls"
...
This reverts commit a09204388a
2024-05-28 06:28:17 +00:00
5b0bfdcbad
SWDEV-460242 - Add system memory suballocator
...
Switch commands creation to the new suballocator to avoid
frequent expensive OS calls
Change-Id: I3597c811820e577c15708bad8b8a41aa53acc400
2024-05-28 06:28:17 +00:00
f5d0ee650b
SWDEV-460151 - Enabling gfx12 in HIP runtime.
...
Change-Id: I374ea7c3119b0c61f9846a862c4a448ddb179748
2024-05-28 06:28:07 +00:00
2eb30376ba
SWDEV-451945 - Remove ShouldLoadPlatform function
...
Change-Id: Iabb4071bb77201576bc2c0488a04f4fa188815df
2024-05-06 10:42:59 -04:00
4a9d24a211
SWDEV-301667 - Pass reference to kernel name
...
Change-Id: I21abe109ddfabfe7640bf78a96c81a1317d31952
2024-05-05 16:38:20 -04:00
1bd8b4eaf5
SWDEV-433371 - Build needs comgr >= 2.8
...
Unbundle compressed code objects needs comgr 2.8 or higher
Change-Id: I23942d2038e19b02c3ea5d3c9c1fe5367db87136
2024-05-03 09:34:49 +00:00
6cb7b6ec6b
SWDEV-451594 - Change device kernel args to use HDP flush by default
...
The Readback and Avoid HDP Flush memory ordering workaround is
used as a fallback solution only when HDP flush register is invalid
Change-Id: Ic284eba1f95ed22b0270d3abeb904fb902015b1a
2024-05-02 19:35:13 +00:00
e53df57ffe
SWDEV-433371 - use comgr to unbundle code objects
...
1.Make runtime use comgr to unbundle code objects
2.Support compressed/uncompressed modes
3.Remove HIP_USE_RUNTIME_UNBUNDLER and
HIPRTC_USE_RUNTIME_UNBUNDLER to simplify logics
4.Add comgr wrapper for
amd_comgr_action_info_set_bundle_entry_ids()
Change-Id: Ic41b1ad1b64cca1e31986437983a5146d52a7329
2024-05-01 16:09:12 -04:00
948ca5a931
SWDEV-301667 - Add LOG_TS mask
...
- Add LOG_TS mask for printing signal times
- Read raw ticks from signals
Change-Id: Ibdd0bf06c790729f6c65083a4784c97a3c3219e0
2024-04-30 12:24:48 -04:00
1d48f2a1ab
SWDEV-456279 - Adding new hip flag to access contiguous memory and pass the flag to HSA API.
...
Change-Id: I1bafeaa3096395c729723af958d609bc41e7845c
2024-04-30 05:25:38 -04:00
7a371503b2
SWDEV-311271 - Enable mempools under Linux
...
Change-Id: I7fda94e61121f9d3a30f4ad185b8a97712922f3c
2024-04-29 18:06:34 -04:00
0e1a0572e6
SWDEV-413997 - Changes to use GlobalContext in views.
...
Change-Id: I1f8411eae9ed49632667e244a25f223fed92c720
2024-04-29 16:41:39 -04:00
49b4aef8c9
SWDEV-413997 - Handling cases where multiple phys_mem is mapped into a single virtual address.
...
Change-Id: Ie678e607a64f2e5c35a10b9083185f041c5527ac
2024-04-28 08:28:12 -04:00
5c1804aa14
SWDEV-353281 - Corret VA unmap
...
Make sure graph mempool unmaps VA on release
Change-Id: Id3f1bd8d0115b533ae60aa5ba3676b8bf7e5b961
2024-04-26 09:37:01 -04:00
cffff4e1cb
SWDEV-457859 - Initialize isWGPMode_ in WorkGroupInfo
...
Change-Id: Ie3f3c0bcea84368c1b0607fd52b4bc7cae41c512
2024-04-25 16:36:55 -04:00
9fdddb7c5d
SWDEV-447691 - Correct handle type for DX12 semaphore
...
Change-Id: Id23882286cb2a0d0472964ffc501ab27b7dc7f00
2024-04-25 11:24:56 -04:00
5a715ed160
Switch luxmark to lightning compiler for all ASICs
...
Change-Id: Idcd37628a2167f0bd2db2a83132a1862cbd051b0
2024-04-23 10:00:39 -04:00
74ffc5f0d5
SWDEV-413997 - Cleanup fixes for Virtual Memory Management.
...
Change-Id: I9a4a4d9087b5daf15e3ba31e786d34db431212a1
2024-04-22 10:58:06 -04:00
0ccdb3e160
SWDEV-440746 - Release last command on terminate
...
Change-Id: Ib6a9b8fc9a8692eb17b39b854cefd92c6b59733f
2024-04-22 09:57:38 -04:00
7448113cfc
SWDEV-440746 - Remove obsolete code
...
The "optimized" version of memcpy is outdated and
was used in win32 only.
Change-Id: I7f2e0e9051e37cec95438266824b5b0025c324c6
2024-04-22 09:56:42 -04:00
fb217fa9e0
SWDEV-453180 - Add UUID support for HIP_VISIBLE_DEVICES on Windows
...
- UUID needs to be specified in the format GPU-<body>, <body> encodes UUID as a 16 chars
- Convert set UUID in HIP_VISIBLE_DEVICES to device index internally
- Then use existing device index logic for HIP_VISIBLE_DEVICES
Change-Id: I654f492a49cd4d7a9b7339360ab558165240caa5
2024-04-20 02:39:19 -04:00
329ba271fa
SWDEV-440746 - Wait for signal before release
...
Change-Id: I9e2aefdbcbba153c7f1080d80aab7a345eaf1eb4
2024-04-19 18:33:28 -04:00
fd81490bb8
SWDEV-440746 - Don't set CL_SUBMITTED twice
...
Change-Id: I9ba34454f7487d6bc0d398b322a147cbac6c6443
2024-04-19 17:36:51 -04:00
bf74ef4025
SWDEV-451594 - Implement Readback and Avoid HDP Flush workaround for device kernel args
...
Change-Id: I6d41a089a17f55306e7ff402588a1e831b20a7a7
2024-04-19 09:29:20 -04:00
56ebf5157a
SWDEV-413997 - VMM IPC implementation for Linux.
...
Change-Id: Icfeb83ca51e96be35abb67a94d6e3e1a1ca5a934
2024-04-18 11:28:13 -04:00
8942939fac
SWDEV-455346 - End wait if HostcallListener terminates.
...
Change-Id: I21ec8eadb189147c579ec65acf68de40d604686b
2024-04-18 00:04:00 -04:00
62559a6e5a
SWDEV-440746 - Fix the hostcall buffer creation
...
Avoid a deadlock on the host call buffer creation. Since the buffer will be
allocated in the queue thread, then use direct device memory allocation
skipping the global context lock.
Change-Id: I09b55ee03bb42ab5d320c152b52a8c842c5fdcc1
2024-04-17 12:37:23 -04:00
5ddca5854c
SWDEV-455699 - removing HW_REG_TRAPSTS for gfx12
...
Change-Id: I5f8b030eefdb37d3d51da3e135e5aa0f18ad9018
2024-04-16 19:46:21 -04:00
c95a75a2bf
SWDEV-444670 - Enable teardown class
...
Force implicit runtime teardown with a global destructor.
Change-Id: Iabe63dedf5b94fefc98668585c45a61607120669
2024-04-16 12:00:06 -04:00
52db98edd9
SWDEV-453180 - Add UUID support for HIP_VISIBLE_DEVICES on Linux
...
- UUID is Ascii string with a maximum of 21 chars which uniquely identifies a GPU
- Convert set UUID in HIP_VISIBLE_DEVICES to device index internally
- Then use existing device index logic for HIP_VISIBLE_DEVICES
Change-Id: I8cab4fe42459f8209b97f909300789e6e687b9ac
2024-04-13 22:07:19 -04:00
d52d16c8e6
SWDEV-413997 - Fixing multiple device cases.
...
Change-Id: I10ad3fbfca887e92cd81f68392fa1acf753cbd2b
2024-04-13 06:14:03 -04:00
7de7da4016
SWDEV-455254 - Reduce blit kernels signature
...
Remove offset from blit kernels, since it can be applied in setup.
Change-Id: I06b585068d68a0ee8e125ddf46a36fccb372f30d
2024-04-12 14:45:55 -04:00
35c80dd482
SWDEV-424956 - Fix half vector printf issue
...
Refactor PrintfDbg::outputArgument() to remove potential risk.
Fix half vector printf issue on all devices.
Fix FEAT-56794 as well.
Change-Id: Iae39359d2128588def2e43d77fe58e868b8e71ff
2024-04-12 14:25:44 -04:00
d52168b46d
SWDEV-436754 - Use glbctx instead so that ref count increments for multi devices and chunk decommit gets delayed.
...
Change-Id: Ia4b0d5fbfa8f198776e52d14de8b22c6942f740d
2024-04-12 00:04:34 -04:00
f0c7ecf617
SWDEV-455254 - Add kernel arg optimization
...
Add kernel arguments optimization into blit path.
Enabled by default on MI300.
Change-Id: I2694a81b90d48ad07d86dfe4c0c64fe187bada8e
2024-04-10 18:08:37 -04:00
2b8634bada
SWDEV-446298 - Adding error code to the logs on p2p hsa api failure.
...
Change-Id: Ic41b1ad1b64cca0e31986337a83a5146d52a7328
2024-04-10 06:00:00 -04:00
3f0bcf7834
SWDEV-301667 - Fix SDMA mask reuse
...
If we are using the mask returned by getLastUsedSdmaEngine() then we
need to apply the SDMA Read/Write mask to it before using with HSA
copy_on_engine API.
Change-Id: I6e5dc6c187eeb3c61ee159e9d2a0fa7b4737c06e
2024-04-08 15:42:52 -04:00
dbac2976e4
SWDEV-451964 - Limit gpu single allocation percentage for gfx940 only
...
Change-Id: Iadcdadd734e7aeeb23742e426353defa972d3ad5
2024-04-05 09:43:42 -04:00
d7f352dbed
SWDEV-453301 - Remove the option to write multiple packets in dispatchGenericAqlPacket
...
Dispatching multiple packets with ring the doorbell once is not supported by the lower layers
Change-Id: I7665a2dcdd4ef9e47dadfe410180fed64c5a4ee0
2024-04-05 05:28:10 -04:00
bc80802c1a
SWDEV-446726 - Disable large bar for 32 bit windows
...
When large bar is enabled, persistent memory leads to overallocation for 32 bit architecture.
Change-Id: Iae39359d8128588de02e42d77fe58e868b8e71fd
2024-04-03 15:36:41 -04:00
411960a131
SWDEV-451687 - Fix alloc message values in AMD_LOG_LEVEL for 32 bit
...
Change-Id: Icbe67024297c92bf59139b6a2ccd2ba3674f60b1
2024-04-01 13:32:20 -04:00
2f3ad43c4a
SWDEV-449922 - Remove a global lock for queues on wait
...
The new logic has a lock for PAL call and doesn't require the lock for queues.
Change-Id: I61b67c3c4abd2ede44809de1d6beed756766032e
2024-03-28 18:17:46 -04:00
95e3958748
SWDEV-449922 - Remove per queue residency update
...
alwaysResident setting doesn't require per queue residency tracking.
Thus, the logic can be skipped to avoid the lock of queues.
Change-Id: Ib5cff5b79d3ecb8c2f2eb2565cf069f9a69438b0
2024-03-28 16:16:37 -04:00
Saleel Kudchadker
Satyanvesh Dittakavi
kjayapra-amd
Ioannis Assiouras
Rakesh Roy
German Andryeyev
Alex Xie
Maneesh Gupta
taosang2
Jaydeep Patel
Konstantin Zhuravlyov
pghafari
Sourabh Betigeri
cadolphe