intel/compute-runtime
1
0
Fork 0
mirror of https://github.com/intel/compute-runtime.git synced 2025-12-28 00:03:14 +08:00
Code Issues Packages Projects Releases Wiki Activity
7,833 Commits 42 Branches 655 Tags
dfe578754b238b70782c384946ead2bf363cb545
Commit Graph

100 Commits

Author SHA1 Message Date
Zbigniew Zdanowicz
a869921c2a Remove RMW from gfx allocations 
Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2021-07-15 11:28:03 +02:00
Dominik Dabek
dc9b2351d5 Change patchGlobalOffset in l0 kernel to void 
Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
 2021-07-06 11:36:53 +02:00
Dominik Dabek
62f89b174a Add work_dim patching to l0 kernel 
Related-To: NEO-5931

Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
 2021-07-05 20:09:20 +02:00
Compute-Runtime-Validation
233aab88e7 Revert "Add support for ze_kernel_preferred_group_size_properties_t" 
This reverts commit 6b0b5ef2ff.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
 2021-07-03 11:19:38 +02:00
Jaime Arteaga
6b0b5ef2ff Add support for ze_kernel_preferred_group_size_properties_t 
Signed-off-by: Jaime Arteaga <jaime.a.arteaga.molina@intel.com>
 2021-07-02 12:10:33 +02:00
Jim Snow
2acc0fb3f6 Add memory backed buffer allocation for L0 ray tracing. 
This allocates the buffer on a per-device basis and enables ray
tracing on devices that support it when given a kernel with ray
tracing calls.

Signed-off-by: Jim Snow <jim.m.snow@intel.com>
 2021-07-02 11:56:18 +02:00
Maciej Plewka
4dda709b41 Fix add unaligned part to size in surface state 
Signed-off-by: Maciej Plewka <maciej.plewka@intel.com>
 2021-06-29 13:13:26 +02:00
Compute-Runtime-Validation
20ba56f254 Revert "Add support for ze_kernel_preferred_group_size_properties_t" 
This reverts commit ef07bbd041.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
 2021-06-28 17:21:33 +02:00
Jaime Arteaga
ef07bbd041 Add support for ze_kernel_preferred_group_size_properties_t 
Signed-off-by: Jaime Arteaga <jaime.a.arteaga.molina@intel.com>
 2021-06-28 13:15:50 +02:00
Maciej Plewka
0d973aaaa3 Programm smaller buffer size in SS when offset not patched 
Signed-off-by: Maciej Plewka <maciej.plewka@intel.com>
 2021-06-23 14:20:53 +02:00
Compute-Runtime-Validation
3137e82fca Revert "Remove redundant DualSubSliceCount checks" 
This reverts commit 5e85fe5042.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
 2021-06-17 13:47:49 +02:00
Filip Hazubski
5e85fe5042 Remove redundant DualSubSliceCount checks 
Signed-off-by: Filip Hazubski <filip.hazubski@intel.com>
 2021-06-17 11:11:56 +02:00
Maciej Plewka
689ceacfe6 Fix set allocation adress in SS when offset is patched 
Signed-off-by: Maciej Plewka <maciej.plewka@intel.com>
 2021-06-08 13:05:38 +02:00
Bartlomiej Wolny
b98148fb3f Adjust minWorkGroupSize when EUFusion is enabled 
Related-To: NEO-5260

Co-authored-by: Bartlomiej Wolny <bartlomiej.wolny@intel.com>
Signed-off-by: Maciej Dziuban <maciej.dziuban@intel.com>
 2021-06-08 11:43:26 +02:00
Mateusz Hoppe
32729f6feb Handle VM BIND events - extract ISA allocation info 
Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
 2021-05-20 13:19:23 +02:00
lgotszal
3bd4bca911 Copyright header update 
Dates corrected in copyright headers to reflect original publication date
(2018 for OpenCL, 2020 for Level Zero).

Signed-off-by: lgotszal <lukasz.gotszald@intel.com>
 2021-05-17 20:38:19 +02:00
Kamil Kopryk
14bc9fb0bf Ignore media block 2d patch token when media block support is disabled 
Signed-off-by: Kamil Kopryk <kamil.kopryk@intel.com>
Related-To: NEO-5820
 2021-05-17 11:38:46 +02:00
Mateusz Hoppe
1414247ed4 Register ELF before transferring ISA to Allocation 
Related-To: NEO-5857

Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
 2021-05-14 13:04:54 +02:00
Young Jin Yoon
bb53de1e4e Fix errors for incorrect work group size 
Related-To: LOCI-2084
Signed-off-by: Young Jin Yoon <young.jin.yoon@intel.com>
 2021-05-14 12:05:48 +02:00
Jaime Arteaga
128cd8a31c Add support for non-IPC P2P access to L0 
Signed-off-by: Jaime Arteaga <jaime.a.arteaga.molina@intel.com>
 2021-04-20 01:05:40 +02:00
Maciej Plewka
a51ee31500 Add method for unknown kernel arg 
Signed-off-by: Maciej Plewka <maciej.plewka@intel.com>
 2021-04-15 16:39:39 +02:00
Jaime Arteaga
a56b413392 Improve support for L0 uncached device allocations (2) 
Make sure UNCACHED flags are used in stateful paths.

Related-To: NEO-5500

Signed-off-by: Jaime Arteaga <jaime.a.arteaga.molina@intel.com>
 2021-04-12 09:47:30 +02:00
Jaime Arteaga
b510923e25 Fix error code for non-found kernel on zeKernelCreate() 
Signed-off-by: Jaime Arteaga <jaime.a.arteaga.molina@intel.com>
 2021-04-09 10:30:56 +02:00
Jaime Arteaga
ddca333045 Improve support for L0 uncached device allocations 
Make sure UNCACHED flags are translated into setting the MOCS index
for uncaching L3.

Related-To: NEO-5500

Signed-off-by: Jaime Arteaga <jaime.a.arteaga.molina@intel.com>
 2021-04-08 13:00:03 +02:00
Igor Venevtsev
bd32518d31 Add extra parameters to EncodeComputeMode::adjustComputeMode() method 
Signed-off-by: Igor Venevtsev <igor.venevtsev@intel.com>
 2021-03-31 16:51:55 +02:00
Mateusz Jablonski
8215395401 Simplify Context method 
return if context has multiple sub devices related to a given root device

Related-To: NEO-3691
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2021-03-30 10:22:15 +02:00
Mateusz Jablonski
35ff284944 Cleanup Kernel class 
move deviceVector to MultiDeviceKernel class
remove Device arg from Kernel's methods

Related-To: NEO-5001
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2021-03-24 09:17:41 +01:00
Filip Hazubski
8d55bfe21d Implement zeCommandListAppendLaunchCooperativeKernel 
Resolves: NEO-4725


Signed-off-by: Filip Hazubski <filip.hazubski@intel.com>
 2021-03-22 19:26:41 +01:00
Igor Venevtsev
52e118fe49 Add extra parameters to Buffer::setSurfaceState() method 
Signed-off-by: Igor Venevtsev <igor.venevtsev@intel.com>
 2021-03-16 10:17:59 +01:00
davidoli
8fdd1931a9 improve stub for zetKernelGetProfileInfo with ULT 
Signed-off-by: davidoli <david.olien@intel.com>
 2021-03-01 00:17:58 +01:00
Jaime Arteaga
722a80079b Add experimental_properties for indirect access (2) 
Add support to OpenCL to also make use of indirect access properties
provided by IGC and to make use of DisableIndirectAccess debug key.

Signed-off-by: Jaime Arteaga <jaime.a.arteaga.molina@intel.com>
 2021-02-25 11:41:24 +01:00
Mateusz Hoppe
6dd0f0c728 Relocate debug data 
Related-To: NEO-4769

Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
 2021-02-16 14:59:30 +01:00
Jaime Arteaga
1aca16a7ac Add experimental_properties for indirect access 
Read zebin experimental properties indicating whether a kernel has
indirect access and if so, ignore calls to zeKernelSetIndirectAccess.

Signed-off-by: Jaime Arteaga <jaime.a.arteaga.molina@intel.com>
 2021-02-10 14:23:43 +01:00
Igor Venevtsev
3df6110a17 Add extra parameters to setArgStateful() 
Signed-off-by: Igor Venevtsev <igor.venevtsev@intel.com>
 2021-02-05 12:24:27 +01:00
Jaime Arteaga
2fa27ebe73 Move ISA at kernel creation time (2) 
For internal kernels, move ISA only at creation time. For user
kernels, do it a module creation time.

Related-To: LOCI-2009

Signed-off-by: Jaime Arteaga <jaime.a.arteaga.molina@intel.com>
 2021-02-04 21:03:41 +01:00
Mateusz Hoppe
f7613a5827 Program surface state when debugging is enabled 
- surface state for debugSurface should be set when
kernel is compiled for debug

Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
 2021-02-02 17:53:12 +01:00
Jaime Arteaga
afffedebb2 Move ISA at kernel creation time 
Instead of moving the ISAs for all kernel in a module when the module
is created, move the ISA when the kernel is created, to avoid
unnecessary memory transfers.

Related-To: LOCI-2009

Signed-off-by: Jaime Arteaga <jaime.a.arteaga.molina@intel.com>
 2021-02-01 13:28:38 +01:00
Young Jin Yoon
4168ee8120 Print output from Kernel dtor to avoid segfault 
Related-to: LOCI-1912
Signed-off-by: Young Jin Yoon <young.jin.yoon@intel.com>
 2021-01-21 23:00:15 +01:00
Jaime Arteaga
05b5ad37ea Initialize kernel private surface when kernel is created 
This instead of when the associated module is created, to avoid
allocating memory for kernels that are never created nor used.

Signed-off-by: Jaime Arteaga <jaime.a.arteaga.molina@intel.com>
 2021-01-13 17:22:40 +01:00
Jaime Arteaga
08655a315c Revert "Initialize kernel private surface when kernel is created" 
This reverts commit be2a87fe98.

Signed-off-by: Jaime Arteaga <jaime.a.arteaga.molina@intel.com>
 2021-01-10 22:56:32 +01:00
Jaime Arteaga
be2a87fe98 Initialize kernel private surface when kernel is created 
This instead of when the associated module is created, to avoid
allocating memory for kernels that are never created nor used.

Signed-off-by: Jaime Arteaga <jaime.a.arteaga.molina@intel.com>
 2021-01-08 19:22:17 +01:00
Slawomir Milczarek
55f3c8f134 Add resolve capability for compressed USM device allocations 
Related-To: NEO-5107

Signed-off-by: Slawomir Milczarek <slawomir.milczarek@intel.com>
 2021-01-08 12:16:26 +01:00
Jaime Arteaga
d7ea713c5f Revert "Initialize kernel immutable data when kernel is created" 
This reverts commit a6ac10088c.

Signed-off-by: Jaime Arteaga <jaime.a.arteaga.molina@intel.com>
 2021-01-04 11:11:39 +01:00
Kamil Diedrich
e0da0e1868 Fix for block load and store 
Signed-off-by: Kamil Diedrich <kamil.diedrich@intel.com>
 2020-12-31 12:28:26 +01:00
Jaime Arteaga
a6ac10088c Initialize kernel immutable data when kernel is created 
This instead of when the associated module is created, to avoid
allocating memory for kernels that are never created nor used.

Signed-off-by: Jaime Arteaga <jaime.a.arteaga.molina@intel.com>
 2020-12-29 19:29:10 +01:00
Krystian Chmielewski
4948c39d39 Remove executionEnvironment from KernelInfo 
Related-To: NEO-3739
 2020-12-16 14:56:57 +01:00
Jim Snow
37cd49330c Implement ZE_CACHE_CONFIG_FLAG_LARGE_DATA for zeKernelSetCacheConfig 
Signed-off-by: Jim Snow <jim.m.snow@intel.com>
 2020-12-16 07:00:13 +01:00
Zbigniew Zdanowicz
b32b5784c2 Fix calculations for offseted addresses 
Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2020-12-11 20:25:54 +01:00
Jaime Arteaga
beb3c5ed05 Add support for global work offset extension in L0 
Add experimental extension to set global work offest in L0.
Current L0 specification does not have interface to export
experimental function symbols, so for now, applications need
to find the symbol like with dlsym on Linux.

A blackbox test showing functionality is also added.

Signed-off-by: Jaime Arteaga <jaime.a.arteaga.molina@intel.com>
 2020-12-09 07:33:40 +01:00
Zbigniew Zdanowicz
19bea962d0 Use imported host pointers in the driver when available 
Related-To: NEO-5126

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2020-12-03 13:52:14 +01:00