intel/compute-runtime
1
0
Fork 0
mirror of https://github.com/intel/compute-runtime.git synced 2026-01-03 06:49:52 +08:00
Code Issues Packages Projects Releases Wiki Activity
16,924 Commits 43 Branches 655 Tags
716c557cdb7bf00e2dcbbf30b90032ff3e3dc993
Commit Graph

2756 Commits

Author SHA1 Message Date
Jitendra Sharma
0d53b21d3c refactor: correct variable naming 
Signed-off-by: Jitendra Sharma <jitendra.sharma@intel.com>
 2024-04-11 11:12:56 +02:00
Kamil Kopryk
cfc3b6a8c2 refactor: add heaplessModeEnabled argument to isHeaplessStateInitEnabled 
Signed-off-by: Kamil Kopryk <kamil.kopryk@intel.com>
 2024-04-11 10:08:10 +02:00
Bartosz Dunajski
f62b49a237 feature: device->host fence in in-order L0 path 
Related-To: NEO-10417

Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2024-04-10 19:30:06 +02:00
John Falkowski
10c7a9552d fix: correct memory copy region 
Resolves:  NEO-11103

Signed-off-by: John Falkowski <john.falkowski@intel.com>
 2024-04-10 09:41:53 +02:00
Mateusz Jablonski
cb2b572e94 feature: add support for null aub mode 
In this mode AUB csr will be created, however, no aub file will be created

Related-To: NEO-11097
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2024-04-09 16:59:42 +02:00
Bartosz Dunajski
945deb8f67 feature: memory copy device to host fence for host visible events in l0 path 
Related-To: NEO-10417

Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2024-04-09 16:11:23 +02:00
Compute-Runtime-Validation
9568ee47e7 Revert "fix: remove compiler cache legacy implementation" 
This reverts commit 864f42116c.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
 2024-04-09 02:06:51 +02:00
Zbigniew Zdanowicz
0ad5316f53 fix: reset patched scratch per thread size values after command list reset 
Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2024-04-08 22:11:18 +02:00
Morek, Szymon
035bcee57a refactor: Remove unused bool 
Signed-off-by: Morek, Szymon <szymon.morek@intel.com>
 2024-04-08 17:50:12 +02:00
Weronika Kapusta
864f42116c fix: remove compiler cache legacy implementation 
Related-To: NEO-10679
Signed-off-by: Kapusta, Weronika <weronika.kapusta@intel.com>
 2024-04-08 16:40:57 +02:00
Zbigniew Zdanowicz
bbd3efc13b test: change tests for command list global stateless mode 
- add module object to appended kernel to avoid null dereference
- disable global stateless in tests and fixtures expecting private surface heap
- initialize correctly command queue for possible global stateless heap
- add initialized internal queue to immediate command lists
- do not parse command buffer when no commands are dispatched and
- no commands are expected to be found

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2024-04-08 15:02:40 +02:00
Zbigniew Zdanowicz
1f35eaf7d8 feature: enable scratch address patching on regular command lists 
Related-To: NEO-10381

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2024-04-05 14:12:12 +02:00
Zbigniew Zdanowicz
a468827c12 feature: improve patching of scratch inline pointer 
Related-To: NEO-10381

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2024-04-05 11:24:00 +02:00
Kacper Nowak
b8d19ccfb5 refactor: Move logic for querying maxMemoryFillPatternSize to product helper 
Signed-off-by: Kacper Nowak <kacper.nowak@intel.com>
 2024-04-05 09:35:01 +02:00
Zbigniew Zdanowicz
73d558058c feature: add heapless and global stateless scratch address patching 
Related-To: NEO-10381

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2024-04-03 17:04:35 +02:00
Naklicki, Mateusz
5a9853c7d0 fix: properly initialize data for event packets 
Signed-off-by: Naklicki, Mateusz <mateusz.naklicki@intel.com>
 2024-04-02 17:20:47 +02:00
Bartosz Dunajski
f0a24a650c feature: sync dispatch cleanup section 
Related-To: NEO-8171

Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2024-04-02 15:16:21 +02:00
Kamil Kopryk
4eae28bd64 feature: introduce heapless state init in L0 
Related-To: NEO-7824
Signed-off-by: Kamil Kopryk <kamil.kopryk@intel.com>
 2024-04-02 12:34:53 +02:00
Bartosz Dunajski
50908a0809 feature: full sync dispatch mode initialization path 
Related-To: NEO-8171

Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2024-04-02 11:07:09 +02:00
Mateusz Hoppe
f86d4220a5 feature: add bindless samplers support to level zero 
- samplers using bindless adressing require patching bindless offsets to
sampler states on kernel's cross thread data

Related-To: NEO-10505

Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
 2024-03-29 08:07:28 +01:00
Young Jin Yoon
d6a14d4ed5 feature: support explicit memory locking 
Added lockMemory in context to explicitly locking memory,
Added a boolean flag in graphics_allocation to indicate the allocation
is locked, and modified memory_operations_handler to add lock().

Related-To: NEO-8277
Signed-off-by: Young Jin Yoon <young.jin.yoon@intel.com>
 2024-03-29 07:31:22 +01:00
Wenju He
15420aa094 fix: fix image 1D/2DArray in appendImageCopyFromMemoryExt/ToMemoryExt 
This fixes SYCL bindless image e2e test read_write_unsampled_array.cpp
when copy engine is used.

Related-To: GSD-4137
Signed-off-by: Wenju He <wenju.he@intel.com>
 2024-03-29 07:20:11 +01:00
Kacper Nowak
35ca2bbea9 refactor: refactor logic for querying atomic FP capabilities 
- Move logic from product helper to compiler product helper
- Add method for adjusting fp16 and extra capabilities using release
helper (if present).


Signed-off-by: Kacper Nowak <kacper.nowak@intel.com>
 2024-03-28 15:27:45 +01:00
Zbigniew Zdanowicz
94cf31033c refactor: change queue interfaces to provide different scratch controller 
Related-To: NEO-10381

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2024-03-28 13:50:04 +01:00
Wenju He
b0b82171ab fix: fix potential integer overflow in CommandListCoreFamily getInputBufferSize 
Related-To: GSD-4137
Signed-off-by: Wenju He <wenju.he@intel.com>
 2024-03-28 10:29:40 +01:00
Bartosz Dunajski
0fbd2b51f5 feature: dont assign queue id for limited sync dispatch 
Related-To: NEO-8171

Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2024-03-27 15:46:26 +01:00
Bartosz Dunajski
6c7a568f64 feature: handle limited sync dispatch initialization 
Related-To: NEO-8171

Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2024-03-27 13:44:11 +01:00
Zbigniew Zdanowicz
3560e8a735 feature: use stateless only kernels in command lists of stateless mode 
Related-To: NEO-10381

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2024-03-27 13:30:47 +01:00
Bartosz Dunajski
fce016ee7a refactor: add new semaphore param 
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2024-03-26 15:46:47 +01:00
Zbigniew Zdanowicz
00445d9285 feature: add debugger functionality to global stateless feature 
Related-To: NEO-10381

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2024-03-26 14:11:41 +01:00
Bartosz Dunajski
74dcb5a9e3 feature: initial support for sync dispatch token allocation 
Related-To: NEO-8171

Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2024-03-26 12:07:58 +01:00
Wenju He
8acb4678fb feature: implement zeCommandListAppendImageCopyToMemoryExt/FromMemoryExt 
This PR implements these two APIs defined in Level Zero spec.
They are required for SYCL bindless tests read_write_*D_subregion.cpp.

Related-To: GSD-4137
Signed-off-by: Wenju He <wenju.he@intel.com>
 2024-03-26 08:55:25 +01:00
Aravind Gopalakrishnan
af65721cde refactor: Add tlb flush before copy for regular command lists 
Add tlb flush under debug key also for regular command lists

Signed-off-by: Aravind Gopalakrishnan <Aravind.Gopalakrishnan@intel.com>
 2024-03-26 07:03:28 +01:00
Bartosz Dunajski
1377437b3f fix: in-order compact events handling 
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2024-03-26 00:15:53 +01:00
Bartosz Dunajski
97ce15614b feature: mechanism to assign sync queue id in ImplicitScaling mode 
Related-To: NEO-8171

Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2024-03-25 20:30:32 +01:00
Young Jin Yoon
068f6a25c6 Revert "feature: support explicit memory locking" 
This reverts commit 27a3307bb0.

Signed-off-by: Young Jin Yoon <young.jin.yoon@intel.com>
 2024-03-25 20:10:20 +01:00
Mateusz Jablonski
78a4a92b44 refactor: reorder members to reduce internal padding in structs 
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2024-03-25 15:50:00 +01:00
Filip Hazubski
270e8af729 performance: pass extension function name by reference 
Signed-off-by: Filip Hazubski <filip.hazubski@intel.com>
 2024-03-25 10:17:08 +01:00
Young Jin Yoon
27a3307bb0 feature: support explicit memory locking 
Added lockMemory in context to explicitly locking memory,
Added a boolean flag in graphics_allocation to indicate the allocation
is locked, and modified memory_operations_handler to add lock().
Change the logic to work correctly with makeResident() when lock() is
called previously for the same memory region

Related-To: NEO-8277
Signed-off-by: Young Jin Yoon <young.jin.yoon@intel.com>
 2024-03-25 09:49:18 +01:00
Bartosz Dunajski
9aa81bae75 feature: initial support to enable synchronized dispatch 
Related-To: NEO-8171

Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2024-03-22 17:23:58 +01:00
Katarzyna Cencelewska
da7b03dd15 fix: to always use grfs count in calculateNumThreadsPerThreadGroup 
grf size != grf count

Related-To: GSD-8437
Signed-off-by: Katarzyna Cencelewska <katarzyna.cencelewska@intel.com>
 2024-03-22 11:03:18 +01:00
Wenju He
c0f30bece4 refactor: remove redundant allocateBindlessSlot from Image getDeviceOffset 
This is a NFC change.
Bindless slot is already allocated in image initialize function.

Related-To: NEO-10352
Signed-off-by: Wenju He <wenju.he@intel.com>
 2024-03-22 10:37:16 +01:00
Katarzyna Cencelewska
dd1d52259e refactor: add param rootDeviceEnvironment to calculateNumThreadsPerThreadGroup 
Signed-off-by: Katarzyna Cencelewska <katarzyna.cencelewska@intel.com>
 2024-03-21 22:25:14 +01:00
Zbigniew Zdanowicz
bf7436c2a2 refactor: change interface in level zero hw helper methods 
Related-To: NEO-10062

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2024-03-21 21:55:10 +01:00
Zbigniew Zdanowicz
b3be51fc35 refactor: add checker for stateless only kernels 
Related-To: NEO-10381

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2024-03-21 21:54:55 +01:00
Zbigniew Zdanowicz
12affba420 feature: add override key to change command list update capability 
Related-To: NEO-10062

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2024-03-21 17:11:27 +01:00
Mateusz Hoppe
0cec5ccc32 fix: check if any of devices has BindlessHeapHelper enabled 
- report Bindless_images extension in driver only when at least one
device has global bindless heaps enabled

Related-To: NEO-10352

Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
 2024-03-20 20:50:58 +01:00
Mateusz Hoppe
27b930cabc refactor: allow default setting for UseExternalAllocatorForSshAndDsh 
- value of -1 selects driver default setting for external allocator

Related-To: NEO-7063

Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
 2024-03-20 12:29:56 +01:00
Mateusz Jablonski
a2742492ab feature: enable xe drm detection by default 
driver is built with xe drm support by default

added cmake flag to control xe eu debug API support
NEO_ENABLE_XE_EU_DEBUG_SUPPORT

This flag is disabled by default and uapi-eu-debug headers are not
needed for driver compilation as these headers are not a part of
upstream kernel yet.

Related-To: NEO-10780

Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2024-03-19 08:47:13 +01:00
Mateusz Jablonski
19dcc80e44 Revert "build: enable xe drm detection by default" 
This reverts commit 973757a58d.

Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2024-03-18 09:04:59 +01:00