Commit Graph

3626 Commits

Author SHA1 Message Date
645600d141 Return error when there is no memory to evict
We want to return error code to the application instead of aborting when
we are not able to make more memory resident.

Related-To: NEO-7289
Signed-off-by: Fabian Zwolinski <fabian.zwolinski@intel.com>
2022-09-22 14:26:55 +02:00
81f2d04f5a correct and unify programming of front end disable overdispatch property support
Related-To: NEO-5019

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
2022-09-22 13:13:38 +02:00
ec3668fc18 Add initialization method to ioctl helpers
Signed-off-by: Naklicki, Mateusz <mateusz.naklicki@intel.com>
2022-09-22 11:55:59 +02:00
ddaf6c10cd Temporarily disable bufferL3CacheTest ult
Signed-off-by: Kamil Kopryk <kamil.kopryk@intel.com>
Related-To: IGC-5606

Test is validating 1st programmed surface state (it's expecting
buffer at 1st surface state), however
during patch tokens cleanup order of surface states and BTIs
will be changed.
2022-09-21 13:53:43 +02:00
43676ed02a L0Debug Win: initialize device with empty submission
In case of debuggable context device should be additionally
initialized by early empty submission issue.

Signed-off-by: Igor Venevtsev <igor.venevtsev@intel.com>
2022-09-21 12:02:34 +02:00
faf8d51f6d Remove fallback path for PAT index programming
Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
2022-09-21 10:46:43 +02:00
efac290ba3 Do not use selector copy engine
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
2022-09-20 21:49:00 +02:00
e00a5e0a1e Release main copy engine from device and sub device
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
2022-09-20 17:12:34 +02:00
99d63facb5 Clarify meaning of ForceDeviceId debug flag
this flag can be used only to override device id in AUB/TBX mode

Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
2022-09-20 13:15:15 +02:00
cfe51ff2ba Remove not used isSimulation functions
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
2022-09-20 11:01:55 +02:00
3d5e34f727 Reduce the size of masks to 4.
32 is not required.

Signed-off-by: Michal Mrozek <michal.mrozek@intel.com>
2022-09-19 21:53:40 +02:00
6b3ac9f1e2 refactor ocl patchtokens tests
Related-To: NEO-7131

Signed-off-by: Krystian Chmielewski <krystian.chmielewski@intel.com>
2022-09-19 17:46:02 +02:00
0192e8038f Check for GPU hang in path with wait for timestamps
Related-To: NEO-6868

Signed-off-by: Milczarek, Slawomir <slawomir.milczarek@intel.com>
2022-09-19 15:01:46 +02:00
b7a780868a Prepare OCL tests for switch to zebin
Signed-off-by: Krystian Chmielewski <krystian.chmielewski@intel.com>
2022-09-16 15:33:26 +02:00
3f61888912 Prepare ocloc tests for switch to zebin
Signed-off-by: Krystian Chmielewski <krystian.chmielewski@intel.com>
2022-09-16 12:44:59 +02:00
24b1cfbff5 Change internal copy engine to BCS3
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
2022-09-15 16:08:10 +02:00
218a98f7f7 Refactor of pipeline select programming
Adding new interface to cooperate with hw context state
Simplify programming removing unnecessary functions
Code optimization that stop using expensive call and instead
stores configuration parameter

Related-To: NEO-5019

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
2022-09-15 15:38:10 +02:00
cee520b311 simplify systolic mode code and reduce double implementation
Related-To: NEO-5019

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
2022-09-15 11:57:54 +02:00
c824bedc53 Add unit tests for gen file generation
gen file is not generated by default. We can get this file
by ocloc option -gen_file. This change add unit tests
for this option.

Related-To: NEO-5672

Signed-off-by: Mateusz Borzyszkowski mateusz.borzyszkowski@intel.com
2022-09-15 10:46:51 +02:00
f6bf669ee9 Use forward declarations in compression_selector.h
Signed-off-by: Patryk Wrobel <patryk.wrobel@intel.com>
2022-09-14 11:54:24 +02:00
09bf464327 Add option to disable zebin in ocloc
Signed-off-by: Krystian Chmielewski <krystian.chmielewski@intel.com>
2022-09-13 14:36:54 +02:00
9b19014cf1 Fix read of 64bit enviroment variables
change atoi to atoll

Signed-off-by: Cencelewska, Katarzyna <katarzyna.cencelewska@intel.com>
2022-09-09 14:00:10 +02:00
1f6c09ba1d zebin: sanitize scratch space size
Sanitize scratch space size to value programmable on GPU.

Signed-off-by: Krystian Chmielewski <krystian.chmielewski@intel.com>
2022-09-09 11:50:09 +02:00
417746182c Revert "Ocloc: Add -s to options string for non-spirv input with -g option pa...
This reverts commit 301be3c21b.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
2022-09-09 05:42:13 +02:00
ec1de69fee Do not enable basic WA.
LOAD_BALANCED is disabled so basic WA is not needed.

Signed-off-by: Michal Mrozek <michal.mrozek@intel.com>
2022-09-08 19:54:15 +02:00
3a7f266d66 Enable BCS split in OCL
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
2022-09-08 15:56:30 +02:00
4395e0c3a1 Use EXPECT_THROW instead of handmade try-catch in tests
Signed-off-by: Patryk Wrobel <patryk.wrobel@intel.com>
2022-09-08 15:32:52 +02:00
301be3c21b Ocloc: Add -s to options string for non-spirv input with -g option passed
Automatically add "-s" (source path) option if -g flag is present.
This applies only to non-spirv input.
Related-To: NEO-7285
Signed-off-by: Kacper Nowak <kacper.nowak@intel.com>
2022-09-08 15:19:05 +02:00
710c8cf5ef Linker: add support for symbols with local binding
This commits add support for relocating
symbols with local binding and of functional type
(STB_LOCAL, STT_FUNC).

Related-To: NEO-7299
Signed-off-by: Kacper Nowak <kacper.nowak@intel.com>
2022-09-08 14:50:29 +02:00
824c781ab5 Do not program extended WA.
It was only needed for LOAD_BALANCED scenarios, so with recent disabling
of this feature in KMD, it is no longer required.

Signed-off-by: Michal Mrozek <michal.mrozek@intel.com>
2022-09-08 13:51:17 +02:00
f65d2aeb87 Split copy along single dimension
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
2022-09-08 12:39:06 +02:00
a72213943e Change DG2 l1 cache policy to WB
With compiler LSC WAs this gives better performance.

If debugger is active, policy will not be changed ie.
will be WBP.

Related-To: NEO-7003

Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
2022-09-08 12:14:52 +02:00
2e9574c656 Fix printf for type BYTE and SHORT
Generated instructions writing to printf buffer require destination
address to be DWORD aligned. Because of that values of type BYTE (1B)
and SHORT (2B) need to be written as 4B value.
This change adds support for this. When trying to read value of type
BYTE or SHORT four bytes are actually read to be aligned with compiler
implementation.

Signed-off-by: Krystian Chmielewski <krystian.chmielewski@intel.com>
2022-09-08 11:37:18 +02:00
6ad16c9e72 Missing support for OCL debugging with L0
- make resident debugSurface
- allocate debugSurface with correct allocation type and size
- notify cmdQ create/destroy

Related-To: NEO-7075

Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
2022-09-07 13:30:05 +02:00
c0ce477f85 Revert "Enable BCS split WA in OCL"
This reverts commit abfcaf1265.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
2022-09-06 20:42:57 +02:00
16d9000429 Add option to change GRF mode
Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
2022-09-06 13:00:17 +02:00
b41eed8438 CMake: correct test kernels naming + minor improvements
- For test kernels compiled with options passed, change their naming to
following convention:
{basename}_{options_passed}_{suffix}.
- Correct CMake variables naming.
- Refactor logic of retrieving test kernels' data (also in compilers
mock)
- In relation to previous changes: do not generate unnecessary
.gen binary for L0 test kernel

Related-To: NEO-7285
Signed-off-by: Kacper Nowak <kacper.nowak@intel.com>
2022-09-06 11:19:46 +02:00
16798467ac Add api specific config for allocation cache
Currently disabled for both opencl and level zero

Related-To: NEO-6893

Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
2022-09-05 21:15:18 +02:00
57da7ad307 Don't generate gen file by default
Related-To: NEO-6613
Signed-off-by: Warchulski, Jaroslaw <jaroslaw.warchulski@intel.com>
2022-09-05 15:00:48 +02:00
abfcaf1265 Enable BCS split WA in OCL
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
2022-09-05 13:34:11 +02:00
0d6bef0753 Add BCS split to api specific config
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
2022-09-05 11:13:10 +02:00
20f49481f2 Revert "Enable BCS split WA in OCL"
This reverts commit d672920121.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
2022-09-03 20:25:10 +02:00
c8975e14fe Revert "Fix printf for type BYTE and SHORT"
This reverts commit b04c226767.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
2022-09-03 14:11:26 +02:00
e1c7e194ee ULT: remove gmock WA from opencl/shared tests
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
2022-09-02 17:58:08 +02:00
fa6bccdb30 Add new mt test for scenario with blitter
also change in other EnqueueKernelTest to use a proper
retValue per thread
Signed-off-by: Cencelewska, Katarzyna <katarzyna.cencelewska@intel.com>
2022-09-02 17:31:48 +02:00
2430ed5b9a Remove dead code from ULT
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
2022-09-02 17:15:17 +02:00
485ba234f3 Revert change DG2 l1 cache policy
This reverts cache policy back to WBP,
due to functional regressions

Related-To: NEO-7003

Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
2022-09-02 12:02:14 +02:00
01dc470d26 Fix BCS split ULTs on 32 bit
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
2022-09-02 09:56:21 +02:00
572fc24cf5 Revert "Wait on timestamps in Event::wait"
This reverts commit 5d4aca361f.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
2022-09-02 02:43:20 +02:00
b04c226767 Fix printf for type BYTE and SHORT
Generated instructions writing to printf buffer require destination
address to be DWORD aligned. Because of that values of type BYTE (1B)
and SHORT (2B) need to be written as 4B value.
This change adds support for this. When trying to read value of type
BYTE or SHORT four bytes are actually read to be aligned with compiler
implementation.

Signed-off-by: Krystian Chmielewski <krystian.chmielewski@intel.com>
2022-09-01 17:58:31 +02:00