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
12,838 Commits 42 Branches 655 Tags
0204761add07a8b208683f012c4dad9259ca4ced
Commit Graph

3859 Commits

Author SHA1 Message Date
Mateusz Hoppe
0204761add feature: gpu assert implementation 
- allocate assert buffer when kernel has assert
- track assert kernels in cmdlists and cmdqueues
- check and print assert at sync calls: cmdqueue synchronize(), fence
synchronize(), event hostSynchronize(), synchronous imm cmdlists
append()

Related-To: NEO-5753

Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
 2023-03-15 19:22:09 +01:00
Zbigniew Zdanowicz
e645f58b65 [fix] Do not reset state heap position for command list reset 
- state base address tracking allows to reuse base address state
- surface state slots can be reused after sba reload or cache flush
- to avoid cache flush after each reset, then allow to gradualy consume heaps
- only until natural heap depletion and then dispatch reload of sba state

Related-To : NEO-5055

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2023-03-15 19:04:20 +01:00
Filip Hazubski
0bee81c0c0 refactor: Move isLinearStoragePreferred function from gfx to product helper 
Signed-off-by: Filip Hazubski <filip.hazubski@intel.com>
 2023-03-15 18:51:59 +01:00
Fabian Zwolinski
93a30f002b L0 Debugger - check debug_eu entry. 
Related-To: NEO-7790
Signed-off-by: Fabian Zwolinski <fabian.zwolinski@intel.com>
 2023-03-15 16:14:49 +01:00
Lukasz Jobczyk
79075a3e92 Make command buffers resident at cmd container init 
Signed-off-by: Lukasz Jobczyk <lukasz.jobczyk@intel.com>
 2023-03-15 16:11:29 +01:00
Milczarek, Slawomir
eff7439eca [perf] Set local memory pool for kmd-migrated shared allocation 
Select ccs engine for usm device and shared allocations
(i.e. for transfers from local to local).

Related-To: NEO-7252

Signed-off-by: Milczarek, Slawomir <slawomir.milczarek@intel.com>
 2023-03-15 16:02:37 +01:00
Mateusz Jablonski
e3c5cbc22f fix ocl/ocloc: expose cl_ext_float_atomics for OCL 2.0+ 
Related-To: NEO-7800
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2023-03-15 14:48:17 +01:00
Dunajski, Bartosz
686790274d Disable OCL RelaxedOrdering if tag update is required 
Related-To: NEO-7458

Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2023-03-15 14:44:05 +01:00
Mateusz Hoppe
e62c5e25d5 refactor: change debugging enabled to debugging mode 
Related-To: NEO-7630

Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
 2023-03-15 13:41:41 +01:00
Mateusz Jablonski
db719b2acb refactor: rename function getExtensions->getDeviceExtensions 
Related-To: NEO-7800
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2023-03-15 12:46:46 +01:00
Compute-Runtime-Validation
3e1d931296 Revert "L0 Debugger - check debug_eu entry" 
This reverts commit 9f935276a0.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
 2023-03-15 12:28:08 +01:00
Mateusz Jablonski
028e1bb1b6 fix ocloc: define extensions in one place 
unify exposed device extensions between OCL and ocloc

Related-To: NEO-7800
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2023-03-15 11:14:46 +01:00
Dunajski, Bartosz
60741b05aa Add missing struct member initialization 
Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2023-03-15 10:09:05 +01:00
Michal Mrozek
593b3cf4fd Revert "[performance] do not perform migrations if not needed." 
Signed-off-by: Michal Mrozek <michal.mrozek@intel.com>
 2023-03-14 19:08:19 +01:00
Zbigniew Zdanowicz
b6d0a64638 Revert: "compute events select device memory for xe hpg family" 
This reverts commit 072233d170.

Resolves: HSD-18028222329

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2023-03-14 17:18:59 +01:00
Mateusz Jablonski
faba16f657 refactor: move extensions string creation to CompilerProductHelper 
Related-To: NEO-7800
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2023-03-14 16:27:21 +01:00
Dunajski, Bartosz
477448c097 RelaxedOrdering dispatch for OCL 2/n 
Related-To: NEO-7458

Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2023-03-14 15:48:06 +01:00
Mateusz Jablonski
39eab81c83 refactor: add separate helper for extension support 
cl_intel_subgroup_split_matrix_multiply_accumulate

Related-To: NEO-7800
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2023-03-14 15:08:29 +01:00
Fabian Zwolinski
1e4c91fb08 Do not disable scratch pages when dbgr is enabled 
Related-To: NEO-7990
Signed-off-by: Fabian Zwolinski <fabian.zwolinski@intel.com>
 2023-03-14 15:03:18 +01:00
Mateusz Jablonski
79281f4995 disallow using special mode of operation in XY_FAST_COLOR_BLT 
Resolved: NEO-7803
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2023-03-14 14:10:39 +01:00
Mateusz Jablonski
340f932ca2 refactor: move GfxCoreHelper::getExtensions to CompilerProductHelper 
Related-To: NEO-7800
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2023-03-14 13:56:19 +01:00
Zbigniew Zdanowicz
86c91847cc [perf] change stream properties interfaces allowing fine grain selective update 
Related-To: NEO-5055

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2023-03-14 13:46:49 +01:00
Mateusz Jablonski
483f9fe340 refactor: move extensions support logic to compiler product helper 
- cl_intel_subgroup_local_block_io
- cl_intel_dot_accumulate
- cl_intel_create_buffer_with_properties
- cl_khr_subgroup_named_barrier
- cl_intel_subgroup_extended_block_read

Related-To: NEO-7800
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2023-03-14 12:09:26 +01:00
Dunajski, Bartosz
97a35debde RelaxedOrdering dispatch for OCL 1/n 
Related-To: NEO-7458

Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2023-03-13 17:37:12 +01:00
Mateusz Jablonski
7e46668fc7 fix dg2: restore logic of setting number of L3 bank count 
this commit reverts this logic from 5b2e1f619c

Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2023-03-13 16:22:55 +01:00
Filip Hazubski
3cfba5b73d refactor(cmake): check images support per platform instead of per core 
Signed-off-by: Filip Hazubski <filip.hazubski@intel.com>
 2023-03-13 16:14:39 +01:00
Zhenjie Pan
00b675643e fix: missed error handler of SubmissionStatus::FAILED 
Related-To: NEO-7802

Signed-off-by: Pan Zhenjie <zhenjie.pan@intel.com>
 2023-03-13 16:08:27 +01:00
Fabian Zwolinski
9f935276a0 L0 Debugger - check debug_eu entry 
Related-To: NEO-7790
Signed-off-by: Fabian Zwolinski <fabian.zwolinski@intel.com>
 2023-03-13 14:46:28 +01:00
Maciej Plewka
98cbf75adc Remove AIL for filmora on DG2 
Signed-off-by: Maciej Plewka <maciej.plewka@intel.com>
 2023-03-13 11:17:51 +01:00
Raiyan Latif
d5c909c9f9 Fix calculation of number of Ray-Tracing stacks 
MaxDualSubSlicesSupported is filled inside GT_SYSTEM_INFO
structure when querying the KMD appropriately with the
number of enabled DualSubSlices. However we need to find
the highest index of the last enabled DualSubSlice.

For proper allocation of thread scratch space, allocation
has to be done based on native die config (including unfused
or non-enabled DualSubSlices). Since HW doesn't provide us a
way to know the exact native die config, in SW we need to
allocate RT stacks with enough size based on the last used
DualSubSlice.

The IsDynamicallyPopulated field in GT_SYSTEM_INFO is used to
indicate if system details are populated either via Fuse reg.
or hard-coded. Based on this field's value, we calcuate the
numRtStacks appropriately.

Related-To: LOCI-3954

Signed-off-by: Raiyan Latif <raiyan.latif@intel.com>
 2023-03-13 10:48:10 +01:00
Compute-Runtime-Validation
e3a80f0bc1 Revert "Enable state base address tracking" 
This reverts commit 8b9078127f.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
 2023-03-12 07:38:57 +01:00
Zbigniew Zdanowicz
f3324964f6 [perf] initialize stream properties only once without further check 
Related-To: NEO-5055

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2023-03-10 17:50:49 +01:00
Zbigniew Zdanowicz
8b9078127f Enable state base address tracking 
Related-To: NEO-5055

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2023-03-10 17:32:40 +01:00
Zbigniew Zdanowicz
24c8f089ed [perf] add state compute mode dirty flag to allow selective properties update 
- full properties update is time intesive task and must be done only once
- selective update can be done after initial update
- dirty flag will allow to distinguish initial update is done

Related-To: NEO-5055

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2023-03-10 17:27:08 +01:00
Zbigniew Zdanowicz
d93f00e075 [perf] simplify getting indirect heap memory location 
Related-To: NEO-5055

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2023-03-10 17:17:01 +01:00
Kamil Kopryk
5a99635232 Don't check addressing mode for gtpin surface 
If gtpin is used, then don't check addressing mode
of the last explicit arg, which is
gtpin's surface.

Related-To: NEO-6075
Signed-off-by: Kamil Kopryk <kamil.kopryk@intel.com>
 2023-03-10 16:01:24 +01:00
Dunajski, Bartosz
ab3ae9215c Remove SIMD size from scratch/private allocation size calculation 
Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2023-03-10 15:54:14 +01:00
Cencelewska, Katarzyna
398c7b2d29 refactor, remove typo in struct name 
change name of EncodeSempahore to EncodeSemaphore
Signed-off-by: Cencelewska, Katarzyna <katarzyna.cencelewska@intel.com>
 2023-03-10 15:44:25 +01:00
Mateusz Jablonski
89cd4bd4a5 refactor: move getters for device extension support to compiler product helper 
isMatrixMultiplyAccumulateSupported
isBFloat16ConversionSupported

Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2023-03-10 15:41:26 +01:00
Kamil Kopryk
fa8579602f refactor: rename product helper files n/n 
Related-To: NEO-7703
Signed-off-by: Kamil Kopryk <kamil.kopryk@intel.com>
 2023-03-10 13:24:38 +01:00
Kamil Kopryk
f8e17842ed refactor: rename enable product helper files 
Related-To: NEO-7703
Signed-off-by: Kamil Kopryk <kamil.kopryk@intel.com>
 2023-03-10 11:58:58 +01:00
Pan, Zhenjie
b82c2c59cc fix: wrong batch buffer end address in direct submission 
Related-To: NEO-7795

Signed-off-by: Pan Zhenjie <zhenjie.pan@intel.com>
 2023-03-10 10:31:06 +01:00
Kamil Kopryk
63ad1a2e86 refactor: rename product helper cpp files 
Related-To: NEO-7703
Signed-off-by: Kamil Kopryk <kamil.kopryk@intel.com>
 2023-03-10 10:30:32 +01:00
Kacper Kasper
48337d815e Add float atomics hw_info to all platforms 
Signed-off-by: Kacper Kasper <kacper.k.kasper@intel.com>
 2023-03-10 10:30:14 +01:00
Compute-Runtime-Validation
2e0c68af43 Revert "refactor: rename enable product helper files" 
This reverts commit a6dfb3a6c2.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
 2023-03-09 20:01:12 +01:00
Pan, Zhenjie
ad49a6e3b0 fix: truncated invalid kernel reason message 
IGC will add "invalid_kernel: uses-fp64-math" into kernel binary for fp64
unsupported platform. "uses-fp64-math" will be truncated to "uses" due to "-"
is not a valid identifier. It makes application can't get "uses-fp64-math"
through zeKernelGetSourceAttributes.

Related-To: NEO-7794

Signed-off-by: Pan Zhenjie <zhenjie.pan@intel.com>
 2023-03-09 17:58:09 +01:00
Kamil Kopryk
57e68624f3 refactor: rename product helper wddm and drm files 
Related-To: NEO-7703
Signed-off-by: Kamil Kopryk <kamil.kopryk@intel.com>
 2023-03-09 17:51:01 +01:00
Zbigniew Zdanowicz
0950f5a23e Set global heap size to constant value 
Related-To: NEO-5055

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2023-03-09 17:17:32 +01:00
Dunajski, Bartosz
d7aba6747c Move isRelaxedOrderingDispatchAllowed to common helper 
Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2023-03-09 17:11:53 +01:00
Kamil Kopryk
a6dfb3a6c2 refactor: rename enable product helper files 
Related-To: NEO-7703
Signed-off-by: Kamil Kopryk <kamil.kopryk@intel.com>
 2023-03-09 16:55:31 +01:00