intel/compute-runtime
1
0
Fork 0
mirror of https://github.com/intel/compute-runtime.git synced 2025-12-30 01:35:20 +08:00
Code Issues Packages Projects Releases Wiki Activity
9,766 Commits 42 Branches 655 Tags
e6460e5534173a8783ffcf8405a9a9fd9f40e472
Commit Graph

156 Commits

Author SHA1 Message Date
Dominik Dabek
e6460e5534 Add setArgBuffer caching to l0 
When trying to set the same values we don't need to set it again.
The pointer and allocId need to be the same.

Related-To: NEO-6737

Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
 2022-03-11 14:20:15 +01:00
Filip Hazubski
dd01cff879 Unify logic determining thread arbitration policy value 
Related-To: NEO-6728

Signed-off-by: Filip Hazubski <filip.hazubski@intel.com>
 2022-03-08 13:14:56 +01:00
Filip Hazubski
80b520bc9b Change ThreadArbitrationPolicy enum type to int32_t 
Change ThreadArbitrationPolicy::NotPresent value to -1
Update initial values to ThreadArbitrationPolicy::NotPresent

Related-To: NEO-6728

Signed-off-by: Filip Hazubski <filip.hazubski@intel.com>
 2022-03-07 20:04:24 +01:00
Konstanty Misiak
cf1bc3a2ba Disable EU fusion based on kernel properties from compiler 
Related-To: NEO-6633

Signed-off-by: Konstanty Misiak <konstanty.misiak@intel.com>
 2022-02-28 18:50:38 +01:00
Jaime Arteaga
06a4d2cc02 Refactor support for L0 scheduling hints (XE_HP and later) 
Make sure STATE_COMPUTE_MODE is updated when passing the
scheduling hint for a kernel.

Signed-off-by: Jaime Arteaga <jaime.a.arteaga.molina@intel.com>
 2022-02-24 20:42:41 +01:00
Mateusz Jablonski
ea6f089e17 Unify implicit args programming across APIs 
Related-To: NEO-5081
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2022-02-23 11:52:47 +01:00
Mateusz Jablonski
4f71aaf595 Handle SIMD-1 scenario when programming local ids for implicit args 
according to implicit args design for SIMD-1 local ids are one-by-one

Resolves: NEO-6692
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2022-02-17 17:48:54 +01:00
Mateusz Hoppe
e1758a2206 Register ELF for every kernel allocation for zebin format 
- every isa allocation will have ELF linked
- fix for debug elf from patchtoken binary:
pass relocated ELF when exists
- simplify code

Related-To: NEO-5571

Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
 2022-02-08 17:33:53 +01:00
Bartosz Dunajski
4b0d986876 Move AllocationType enum out of GraphicsAllocation class 
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2022-02-04 17:49:09 +01:00
Mateusz Jablonski
b697d75695 Correct dimension order in local ids generated for implicit args 
when local ids are generated by HW, use same dim order for runtime generation
move common logic to separated file

Related-To: NEO-5081
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2022-02-04 12:46:59 +01:00
Compute-Runtime-Validation
6f62a784e1 Revert "Check IndirectStatelessCount from igc" 
This reverts commit 5e62df4f8e.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
 2022-02-04 12:15:37 +01:00
Dominik Dabek
5e62df4f8e Check IndirectStatelessCount from igc 
If kernel has no stateless indirect accesses don't set the
kernelHasIndirectAccess flag.
Don't make resident or migrate if kernel has no indirect accesses.
Changed initial values in KernelAttributes.

Related-To: NEO-6597

Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
 2022-02-02 12:28:30 +01:00
Dominik Dabek
07c75c2de3 Revert "Check IndirectStatelessCount from igc" 
This reverts commit 01f368ac147d99fd0dcb4d71a9a2543cd215adfd.

Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
 2022-01-27 16:06:13 +01:00
Dominik Dabek
63f406a58c Check IndirectStatelessCount from igc 
If kernel has no stateless indirect accesses don't set the
kernelHasIndirectAccess flag.
Don't make resident or migrate if kernel has no indirect accesses.
Changed initial values in KernelAttributes.

Related-To: NEO-6597

Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
 2022-01-26 11:10:04 +01:00
Maciej Plewka
6968bfdb33 Use correct enum values for sampler in clamp mode 
Signed-off-by: Maciej Plewka <maciej.plewka@intel.com>
 2022-01-20 18:15:53 +01:00
Zbigniew Zdanowicz
c36c083812 Refactor implicit scaling parameters for surface state 
Related-To: NEO-6589

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2022-01-17 09:30:58 +01:00
Compute-Runtime-Validation
182042b04d Revert "Update default thread arbitration policy" 
This reverts commit 8c3e9ace69.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
 2022-01-13 21:32:29 +01:00
Bartosz Dunajski
8c3e9ace69 Update default thread arbitration policy 
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2022-01-13 16:24:11 +01:00
Igor Venevtsev
71746a2fff Register zebin binary in L0 debugger 
Related-To: NEO-5571

Signed-off-by: Igor Venevtsev <igor.venevtsev@intel.com>
 2022-01-12 23:17:59 +01:00
Maciej Plewka
df7723411f Patch image descriptor fields with dword width 
Signed-off-by: Maciej Plewka <maciej.plewka@intel.com>
 2021-12-27 17:55:42 +01:00
Jaime Arteaga
7364288ed9 Revert "Use correct enum values for sampler in clamp mode" 
This reverts commit c8737c6b4b3c67a81f7b4d1ea0918a5f9daac87d.

Signed-off-by: Jaime Arteaga <jaime.a.arteaga.molina@intel.com>
 2021-12-22 22:17:57 +01:00
Maciej Plewka
c92a4acc01 Use correct enum values for sampler in clamp mode 
Signed-off-by: Maciej Plewka <maciej.plewka@intel.com>
 2021-12-22 19:40:30 +01:00
Jaime Arteaga
e5415dad5d Use base pointer when looking for peer allocations 
Signed-off-by: Jaime Arteaga <jaime.a.arteaga.molina@intel.com>
 2021-12-22 06:03:31 +01:00
Neil R Spruit
64ca4b3456 Allow Explicit RTDispatch args and remove multi free of rtMemoryBackedBuffer 
Signed-off-by: Neil R Spruit <neil.r.spruit@intel.com>
 2021-12-21 20:09:36 +01:00
Mateusz Hoppe
17f82bbe12 Fix double ISA transfer for user kernels in L0 
Related-To: NEO-6555

- ISA should only be copied once, after linking phase is complete

Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
 2021-12-21 07:54:51 +01:00
Jim Snow
2dfb7df63b Add RTDispatchGlobals allocation for ray tracing 
If a kernel has ray tracing calls, we allocate and initialize
per-device RTDispatchGlobals if needed, and hand off pointer to
the same into a running kernel via an implicit parameter.

Related-To: NEO-5384
Signed-off-by: Jim Snow <jim.m.snow@intel.com>
 2021-12-08 09:44:13 +01:00
Mateusz Jablonski
3de256b17b Correct L0 kernel tests 
respect memory copies for patching ISA
remove tests for nullptr kernel ISA

Related-To: NEO-5081, IGC-4707
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2021-12-03 07:47:50 +01:00
Spruit, Neil R
6d8502847e Set Uncached MOCS for List only given stateless uncached args 
Signed-off-by: Spruit, Neil R <neil.r.spruit@intel.com>
 2021-11-18 03:35:48 +01:00
Spruit, Neil R
9c6433e55e Fix to L3 Caching given unaligned memory & hostPtrCopy only for src host ptrs 
Signed-off-by: Spruit, Neil R <neil.r.spruit@intel.com>
 2021-11-16 00:44:46 +01:00
Zbigniew Zdanowicz
f90932cca7 Use references instead copy ctors 
Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2021-11-15 14:42:33 +01:00
Zbigniew Zdanowicz
f305e38b32 Refactor creation of buffer surface state 4/n 
Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2021-10-21 17:46:47 +02:00
Zbigniew Zdanowicz
e80ec1df9c Refactor creation of buffer surface state 3/n 
Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2021-10-21 16:57:21 +02:00
Zbigniew Zdanowicz
a7747fece5 Refactor creation of buffer surface state 2/n 
Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2021-10-21 16:09:54 +02:00
Zbigniew Zdanowicz
9d56939980 Refactor creation of buffer surface state 1/n 
Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2021-10-21 13:11:31 +02:00
Raiyan Latif
f5210ed89f Account for offset pointer in P2P memory copies 
Signed-off-by: Raiyan Latif <raiyan.latif@intel.com>
 2021-10-21 03:09:31 +02:00
Mateusz Hoppe
10e4b54986 Copy ISA before making it resident 
Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
 2021-10-12 21:51:06 +02:00
Raiyan Latif
0859f99d64 Enable zeDeviceCanAccessPeer 
- Propagate return value up from Batch Buffer exec
- Add zeDeviceCanAccessPeer functionality

Signed-off-by: Raiyan Latif <raiyan.latif@intel.com>
 2021-10-12 09:38:22 +02:00
Zbigniew Zdanowicz
96fef37834 Move shared local work size code to shared directory 
Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2021-10-04 14:04:28 +02:00
Jaime Arteaga
11980db8eb Create L0 list and module with correct NEO device on explicit mode 
When implicit scaling is not used, a root device in L0 defaults to
sub-device 0. In that case, all allocations need to be made against
that sub-device.

For list and module creation, we were passing the root NEO device
instead, and we were allocating their internal allocations using
the least-occupied bank, which meant we were allocating alternately
between the two sub-devices.

This patch makes sure the NEO device used on explicit mode for
the root device is sub-device 0.

Signed-off-by: Jaime Arteaga <jaime.a.arteaga.molina@intel.com>
 2021-10-01 00:43:53 +02:00
Zbigniew Zdanowicz
25c71a6c13 Move kernel info to shared directory 
Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2021-09-29 23:49:49 +02:00
Mateusz Jablonski
e82c2e4653 Patch printf buffer in implicit args in L0 
Related-To: NEO-5081
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2021-09-16 10:31:40 +02:00
Mateusz Jablonski
5a59a6ad2e Add support for implicit args in L0 
Related-To: NEO-5081
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2021-09-15 11:12:23 +02:00
Mateusz Jablonski
f958b053ab Merge patchWorkDim method's logic into setGroupCount method 
Related-To: NEO-5081
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2021-09-14 08:57:24 +02:00
Mateusz Jablonski
caddc63eec Remove not needed function 
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2021-09-07 12:24:35 +02:00
Bartosz Dunajski
0345d9f707 Improve queries with number of SubDevices 
Signed-off-by: Bartosz Dunajski <bartosz.dunajski@intel.com>
 2021-09-01 14:01:57 +02:00
Filip Hazubski
de1e4e0074 Add adjustMaxWorkGroupCount helper 
Signed-off-by: Filip Hazubski <filip.hazubski@intel.com>
 2021-08-27 15:39:48 +02:00
Jaime Arteaga
0bd41cc3ec Add support for ze_kernel_preferred_group_size_properties_t 
Related-To: LOCI-2313

Signed-off-by: Jaime Arteaga <jaime.a.arteaga.molina@intel.com>
 2021-08-25 00:10:13 +02:00
Compute-Runtime-Validation
c92ff15d59 Revert "Add support for ze_kernel_preferred_group_size_properties_t" 
This reverts commit b193475abb.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
 2021-08-24 04:44:59 +02:00
Jaime Arteaga
b193475abb Add support for ze_kernel_preferred_group_size_properties_t 
Related-To: LOCI-2313

Signed-off-by: Jaime Arteaga <jaime.a.arteaga.molina@intel.com>
 2021-08-23 19:23:40 +02:00
Compute-Runtime-Validation
0ae53851da Revert "Add support for ze_kernel_preferred_group_size_properties_t" 
This reverts commit 4586238c0e.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
 2021-08-17 04:28:15 +02:00