intel/compute-runtime
1
0
Fork 0
mirror of https://github.com/intel/compute-runtime.git synced 2025-12-23 11:03:02 +08:00
Code Issues Packages Projects Releases Wiki Activity
18,647 Commits 42 Branches 655 Tags
a017025e67ffaf5e9214accfe80cbba5a51caf7d
Commit Graph

245 Commits

Author SHA1 Message Date
Szymon Morek
d477efeb40 fix: disable staging when cpu copy is allowed 
Related-To: NEO-14027

If buffer cpu copy is allowed, it means that whole
transfer can be done on cpu memcpy.
In that case, disable staging as it's less performant.

Signed-off-by: Szymon Morek <szymon.morek@intel.com>
 2025-02-14 14:05:06 +01:00
Venevtsev, Igor
3a8f748d55 fix: return correct error code for deprecated clSetCommadQueueProperties 
Resolves: NEO-14046
Signed-off-by: Venevtsev, Igor <igor.venevtsev@intel.com>
 2025-02-14 13:41:40 +01:00
Mateusz Jablonski
6ec92ac082 refactor: move groupDevices logic to shared code 
update platforms used in unit test

Related-To: NEO-14062
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2025-02-13 21:42:43 +01:00
Szymon Morek
be2ac667fc performance: enable staging for reads from image 
Related-To: NEO-14026

Also, provide performance hint if ptr from
user does not meet restrictions.

Signed-off-by: Szymon Morek <szymon.morek@intel.com>
 2025-02-13 18:15:59 +01:00
Mateusz Jablonski
756c540275 refactor: remove pre-gen9 code 
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2025-02-04 12:29:05 +01:00
Szymon Morek
ac78c07899 refactor: unify duplicated code for staging images 
Pass command type instead of maintaining
two methods for reads and writes

Signed-off-by: Szymon Morek <szymon.morek@intel.com>
 2025-01-17 09:52:35 +01:00
Jaroslaw Warchulski
4dd1c80362 refactor: use enum class instead of enum for DeviceHierarchyMode 
Related-To: GSD-9560
Signed-off-by: Jaroslaw Warchulski <jaroslaw.warchulski@intel.com>
 2025-01-16 15:49:44 +01:00
Szymon Morek
0b4fe9a0df performance: add staging transfers for cl buffers 
Related-To: NEO-13529

Signed-off-by: Szymon Morek <szymon.morek@intel.com>
 2025-01-15 12:24:43 +01:00
Jaroslaw Warchulski
6a7bee67e9 refactor: unify device hierarchy mode 
Related-To: GSD-9560
Signed-off-by: Jaroslaw Warchulski <jaroslaw.warchulski@intel.com>
 2025-01-14 21:08:44 +01:00
Jaroslaw Warchulski
a180afa2e6 fix: respect affinity mask with COMBINED hierarchy in OCL 
Related-To: GSD-9560
Signed-off-by: Jaroslaw Warchulski <jaroslaw.warchulski@intel.com>
 2025-01-13 15:38:39 +01:00
Szymon Morek
2815d4167d fix: don't use staging for map allocations 
Related-To: NEO-13572

Signed-off-by: Szymon Morek <szymon.morek@intel.com>
 2025-01-07 13:30:11 +01:00
Szymon Morek
548f93c55a Revert "performance: enable staging read for images" 
This reverts commit 6aeb8dffa0.

Signed-off-by: Szymon Morek <szymon.morek@intel.com>
 2024-12-23 17:42:12 +01:00
Szymon Morek
6aeb8dffa0 performance: enable staging read for images 
Related-To: NEO-12968

Signed-off-by: Szymon Morek <szymon.morek@intel.com>
 2024-12-20 15:02:28 +01:00
Szymon Morek
6c4eb322b1 performance: introduce staging reads from image 
Related-To: NEO-12968

Signed-off-by: Szymon Morek <szymon.morek@intel.com>
 2024-12-11 14:43:45 +01:00
Lukasz Jobczyk
7832b115a4 fix: Select csr once for staging buffer memcpy 
Resolves: NEO-13083

Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
 2024-11-22 15:42:09 +01:00
Jaroslaw Warchulski
723e1e7d29 fix: expose tiles as devices in OCL with combined hierarchy 
Related-To: GSD-8708
Signed-off-by: Jaroslaw Warchulski <jaroslaw.warchulski@intel.com>
 2024-11-21 15:23:06 +01:00
Szymon Morek
a25e973205 performance: write image through staging chunks 
Related-To: NEO-12968

Signed-off-by: Szymon Morek <szymon.morek@intel.com>
 2024-11-06 16:45:22 +01:00
Maciej Plewka
df7e5191ed fix: check if tracing_exit was called before return from api call 
Related-To: GSD-9415, NEO-12719
Signed-off-by: Maciej Plewka <maciej.plewka@intel.com>
 2024-11-06 11:31:31 +01:00
Maciej Plewka
75595e687e fix: add missing tracing_exit before return in api.cpp 
Signed-off-by: Maciej Plewka <maciej.plewka@intel.com>

Related-To: GSD-9415, NEO-12719
 2024-11-05 11:59:05 +01:00
Maciej Plewka
1dafb66463 fix: return success from release calls after platform teardown 
Related-To: NEO-11282
Signed-off-by: Maciej Plewka <maciej.plewka@intel.com>
 2024-09-26 19:30:26 +02:00
Mateusz Jablonski
d5812f49d7 refactor: remove not needed vme builtin related code 
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2024-09-16 15:27:49 +02:00
Bartosz Dunajski
4f1262645b refactor: pass extra walker params 
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2024-09-10 15:30:03 +02:00
Michal Mrozek
20d6910b66 performance: move usm pool init to first alloc call 
Signed-off-by: Michal Mrozek <michal.mrozek@intel.com>
 2024-07-18 16:07:22 +02:00
Szymon Morek
dbd96372be performance: adjust staging buffer usage 
Related-To: NEO-11928

Don't copy through staging buffer if dst usm allocation
was not used before and transfer would be splitted.
Also, don't use staging buffers for mapped ocl buffers.

Signed-off-by: Szymon Morek <szymon.morek@intel.com>
 2024-07-10 10:19:18 +02:00
Szymon Morek
29e3eb512c performance: non-usm copy through staging buffers 
Related-To: NEO-11501

Signed-off-by: Szymon Morek <szymon.morek@intel.com>
 2024-06-25 07:18:53 +02:00
Alicja Lukaszewicz
b4c31492d3 fix: add missing tracing to all functions in api.cpp 
Related-To: NEO-11576

Signed-off-by: Alicja Lukaszewicz <alicja.lukaszewicz@intel.com>
 2024-06-06 12:14:21 +02:00
Aleksander Czerwionka
4667f55ea2 fix: add missing checks to return proper OpenCL error from API 
changes affect clEnqueueNDCountKernelINTEL and
clGetKernelMaxConcurrentWorkGroupCountINTEL

Related-To: NEO-9688

Signed-off-by: Aleksander Czerwionka <aleksander.czerwionka@intel.com>
 2024-05-07 20:32:36 +02:00
Jaroslaw Warchulski
ea057abd36 fix: allow empty set to be passed to clSetKernelExecInfo 
Related-To: GSD-8538
Signed-off-by: Jaroslaw Warchulski <jaroslaw.warchulski@intel.com>
 2024-04-16 14:11:04 +02:00
Dominik Dabek
0d69dfe165 refactor: remove comment 
Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
 2024-04-08 11:43:19 +02:00
Dominik Dabek
af1620a308 fix(ocl): allocation info from pool svm ptr 
Fix querying allocation info from pooled svm ptr.
Handle requested allocation alignment.
Refactor sorted vector usage.
Do not associate device with host pool allocation.

Related-To: NEO-9700

Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
 2024-01-05 15:20:01 +01:00
Compute-Runtime-Validation
5535ef3049 Revert "performance(ocl): enable usm pool allocator" 
This reverts commit 7bc8424a69.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
 2023-12-29 05:54:07 +01:00
Dominik Dabek
7bc8424a69 performance(ocl): enable usm pool allocator 
Enable opencl usm pool allocator by default

Related-To: NEO-9700

Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
 2023-12-28 13:14:41 +01:00
Dominik Dabek
2fe3804cc2 performance(ocl): add usm allocation pooling flag 
EnableDeviceUsmAllocationPool and EnableHostUsmAllocationPool for device
and host allocations respectively.

Pool size will be set to flag value * MB.

Allocation size threshold to be pooled is 1MB.

Pools are created per context.

Related-To: NEO-9700

Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
 2023-12-27 11:42:01 +01:00
Dominik Dabek
239831f097 fix: return error for 0 size usm allocations 
according to both level zero and opencl specs, usm allocations with
size=0 should return invalid/unsupported buffer size errors

Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
 2023-12-22 17:26:39 +01:00
Mateusz Jablonski
dd1b9d6abc refactor: correct naming of enum class constants 8/n 
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2023-12-19 08:18:18 +01:00
Mateusz Jablonski
27fbdde4c5 refactor: correct naming of unified memory enums 
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2023-12-13 15:58:21 +01:00
Maciej Plewka
f8eefbd020 fix: return error code for usm param passed to clSetKernelExecInfo 
Related-To: NEO-8202

Signed-off-by: Maciej Plewka <maciej.plewka@intel.com>
 2023-12-05 09:14:00 +01:00
Baj, Tomasz
fe0c7f1004 fix: when size is 0 then don't copy 
Related-To: NEO-8381

Signed-off-by: Baj, Tomasz <tomasz.baj@intel.com>
 2023-12-04 11:18:00 +01:00
Mateusz Jablonski
c3ac7b78bd refactor: correct variable naming 
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2023-12-01 02:18:46 +01:00
Mateusz Jablonski
c9664e6bad refactor: rename global debug manager to debugManager 
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2023-11-30 13:00:59 +01:00
Baj, Tomasz
10a97548c2 refactor: remove ReturnSubDevicesAsApiDevices from OCL and L0 
Related-To: NEO-9437

Signed-off-by: Baj, Tomasz <tomasz.baj@intel.com>
 2023-11-09 15:43:06 +01:00
Jablonski, Mateusz
ac5f64f5c6 fix: fix compilation error in clang on Windows (2/n) 
Signed-off-by: Jablonski, Mateusz <mateusz.jablonski@intel.com>
 2023-10-24 15:59:06 +02:00
Kacper Nowak
0f161c1f19 fix(ocl): Add missing tracing mechanism for clCreateSubDevices API call 
Related-To: NEO-8898
Signed-off-by: Kacper Nowak <kacper.nowak@intel.com>
 2023-09-25 10:09:47 +02:00
Kacper Nowak
66c030b5a0 fix: Remove unneeded cast from clCloneKernel API call 
Related-To: NEO-8330
Signed-off-by: Kacper Nowak <kacper.nowak@intel.com>
 2023-09-08 14:21:12 +02:00
Kacper Nowak
43321dd463 fix: Remove unneeded cast from clLinkProgram API call 
Related-To: NEO-8330
Signed-off-by: Kacper Nowak <kacper.nowak@intel.com>
 2023-09-01 15:36:51 +02:00
Kacper Nowak
1afaf37f78 refactor: Unify GTPin initialization logic between APIs 
Add support for new GTPin loading logic in OCL path, similar to existing
in L0 - invoking exposed, dedicated API call (OpenGTPinOCL).
- Move logic to shared, including unit tests
- Check whether instrumentation is required on
  context creation and if yes,
  make a call to OpenGTPinOCL function.
  Handle potential errors gracefully without exiting.
Signed-off-by: Kacper Nowak <kacper.nowak@intel.com>
 2023-08-09 17:15:24 +02:00
Compute-Runtime-Validation
65df34bbc7 Revert "refactor: Unify GTPin initialization logic between APIs" 
This reverts commit 68a5108e05.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
 2023-08-07 04:31:52 +02:00
Kacper Nowak
68a5108e05 refactor: Unify GTPin initialization logic between APIs 
Add support for new GTPin loading logic in OCL path, similar to existing
in L0 - invoking exposed, dedicated API call (OpenGTPinOCL).
- Move logic to shared, including unit tests
- Check whether instrumentation is required on
  clGetPlatformIDs API call, and if yes,
  make a call to OpenGTPinOCL function.
Signed-off-by: Kacper Nowak <kacper.nowak@intel.com>
 2023-08-05 18:22:08 +02:00
Neil R Spruit
8efd8853fb feature: L0 Device Hierarchy support for Composite & Flat modes 
Related-To: LOCI-4615

- Added Support for users to set ZE_FLAT_DEVICE_HIERARCHY to either FLAT
or COMPOSITE to change how devices are returned in zeDeviceGet and
clGetDeviceIDs.
- COMPOSITE is default behavior that exists today.
- FLAT returns all sub devices which have no sub devices and all root
devices that have no sub devices in zeDeviceGet ie with all devices
flattened out in order.
- Added zeDeviceGetRootDevice for one to retrieve the Root Device for
any SubDevice.

Signed-off-by: Neil R Spruit <neil.r.spruit@intel.com>
 2023-07-20 01:00:54 +02:00
Kacper Nowak
1fc435e529 fix(ocl): Remove unneeded cast from ClCreateContextFromType API call 
Remove not needed c-style cl_context handle casting on
clCreateContextFromType API call. This bug is currently also visible
when using OCL tracing API.
Related-To: NEO-8011
Signed-off-by: Kacper Nowak <kacper.nowak@intel.com>
 2023-06-19 14:07:40 +02:00