intel/compute-runtime
1
0
Fork 0
mirror of https://github.com/intel/compute-runtime.git synced 2025-12-27 15:53:13 +08:00
Code Issues Packages Projects Releases Wiki Activity
14,250 Commits 42 Branches 655 Tags
1963ea69e5dcb485adef10a0ebc9ff19f548641f
Commit Graph

2214 Commits

Author SHA1 Message Date
Mateusz Jablonski
1963ea69e5 fix: add missing member initializations 
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2023-10-03 08:41:46 +02:00
Dunajski, Bartosz
b4733dedb7 feature: skip not needed event waits in in-order mode 
Related-To: NEO-7966

Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2023-10-03 08:30:30 +02:00
Naklicki, Mateusz
0461af492d fix: unify path for getting number of grfs per thread 
Related-To: NEO-8043

Signed-off-by: Naklicki, Mateusz <mateusz.naklicki@intel.com>
 2023-10-03 08:17:46 +02:00
Mateusz Jablonski
16d94a0b8c refactor: remove dead code 
Related-To: NEO-9038
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2023-10-03 07:34:31 +02:00
Dunajski, Bartosz
891cc2d09b feature: enable waiting for in-order events from regular CmdList 
This is prework. Functional only on immediate CmdLists

Related-To: NEO-8145

Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2023-10-02 16:42:10 +02:00
Spruit, Neil R
5a22477b83 fix: Update Relaxed Memory Size to check physical before global mem size 
Related-To: NEO-9012

- Allows for the memory size requested by the user to be within the
physical memory size if that is set, otherwise the limit is the global
memory size.

Signed-off-by: Spruit, Neil R <neil.r.spruit@intel.com>
 2023-10-02 16:02:35 +02:00
John Falkowski
56f05303c9 feature: add support for zeMemGet/SetAtomicAccessAttributeExp 
Resolves: NEO-8219

Signed-off-by: John Falkowski <john.falkowski@intel.com>
 2023-10-02 15:59:17 +02:00
Compute-Runtime-Validation
1ac37d4a49 Revert "refactor: Enable CSR heap sharing on Older Generation platforms" 
This reverts commit 58ff9c6d94.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
 2023-10-02 15:13:23 +02:00
Dunajski, Bartosz
5402c9a13f feature: pass in-order exec info from Regular CmdList to Event 
Related-To: NEO-8145

Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2023-09-29 15:29:19 +02:00
Hoppe, Mateusz
5c565efe28 feature: bindless global heap with debugger 
- program debugSurface's SurfaceState at the beginning of Bindless Surface
State Heap - SPECIAL_SSH
- ensure SPECIAL_SSH is resident

Related-To: NEO-7063

Signed-off-by: Hoppe, Mateusz <mateusz.hoppe@intel.com>
 2023-09-29 13:13:46 +02:00
Mateusz Jablonski
81b535d3f1 fix: correct handling ze group count in appendLaunchMultipleKernelsIndirect 
Related-To: HSD-18033883770
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2023-09-29 12:16:32 +02:00
Jitendra Sharma
58ff9c6d94 refactor: Enable CSR heap sharing on Older Generation platforms 
Related-To: LOCI-4312
Signed-off-by: Jitendra Sharma <jitendra.sharma@intel.com>
 2023-09-29 11:54:51 +02:00
Dunajski, Bartosz
c5f5f60f8d feature: cross CmdList and Event in-order data sharing [1/n] 
Related-To: NEO-7966

Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2023-09-29 11:14:18 +02:00
Mateusz Jablonski
b8c3dea8dd refactor: simplify KernelImmutableData dtor 
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2023-09-28 08:36:01 +02:00
Mateusz Jablonski
3a21b3b228 refactor: remove not needed code 
Related-To: NEO-7527

Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2023-09-28 07:52:39 +02:00
Mateusz Jablonski
cb730d11f4 refactor: pass ze group count as a reference, not a pointer 
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2023-09-28 07:41:04 +02:00
Dunajski, Bartosz
2aaf5a1f03 feature: enable RelaxedOrdering for in-order Events chaining 
Related-To: NEO-7966

Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2023-09-27 18:36:31 +02:00
Dunajski, Bartosz
4e8600d8d0 feature: initial support for RelaxedOrdering of in-order Events chaining 
Disabled by default.

Related-To: NEO-7966

Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2023-09-27 16:45:20 +02:00
Mateusz Jablonski
09044dfbaa refactor: remove not needed code 
Related-To: NEO-7527

Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2023-09-27 14:35:49 +02:00
Katarzyna Cencelewska
e71db368db fix: use proper pointer inside appendWriteGlobalTimestamp 
Resolves: NEO-8105
Signed-off-by: Katarzyna Cencelewska <katarzyna.cencelewska@intel.com>
 2023-09-27 14:00:37 +02:00
Dunajski, Bartosz
96a7f63371 feature: allow waiting for immediate in-order event from Regular CmdList 
Related-To: NEO-7966

Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2023-09-27 13:03:44 +02:00
Mateusz Jablonski
64fd270d56 refactor: pass reference to pointer as arg to findAllocationDataForRange 
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2023-09-27 12:20:23 +02:00
Dunajski, Bartosz
97b8cac1d1 feature: enable in-order regular cmd lists patching 
Related-To: NEO-7966

Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2023-09-26 09:19:05 +02:00
Dominik Dabek
eebf2bbd26 performance(ocl): timestamp packet count per gfx 
Add support for different timestamp packet counts per gfx family.
Change all packet counts to 1 except for xe-hpc.

Related-To: NEO-8154

Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
 2023-09-25 20:34:58 +02:00
Dunajski, Bartosz
529aa60563 performance: optimize RelaxedOrdering in-order Barrier programming 
Related-To: NEO-7966

Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2023-09-25 15:45:36 +02:00
Dunajski, Bartosz
c115eeb108 feature: dont program pipe_control for in-order barrier profiling 
Related-To: NEO-7966

Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2023-09-25 15:32:18 +02:00
Dunajski, Bartosz
202a33ffac feature: track InOrder allocation ownership 
CmdList can be released before Event. In this case, GfxAllocation
destruction must be deferred.

Related-To: NEO-7966

Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2023-09-25 15:03:25 +02:00
Dunajski, Bartosz
80d0c74605 fix: track registered CSR clients 2 
Related-To: NEO-8884

Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2023-09-25 09:35:55 +02:00
Compute-Runtime-Validation
ade538ce54 Revert "fix: track registered CSR clients" 
This reverts commit 53f635e392.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
 2023-09-24 10:07:26 +02:00
Dunajski, Bartosz
d515ba965c feature: return not ready state for not signaled in-order event 
Related-To: NEO-7966

Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2023-09-22 15:56:50 +02:00
Dunajski, Bartosz
480c058cb2 feature: in-order patching for ComputeWalker 
Related-To: NEO-7966

Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2023-09-22 15:00:44 +02:00
Compute-Runtime-Validation
f19898d178 Revert "fix: enable cache env variables for level-zero" 
This reverts commit 74689fa58a.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
 2023-09-22 13:10:02 +02:00
Dunajski, Bartosz
53f635e392 fix: track registered CSR clients 
In L0 its not possible to track objects relations. For example CmdList
may be removed before Event.
In such case, Event needs to safely skip unregister call, without
accessing CmdList/CmdQueue object.

Related-To: NEO-8884

Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2023-09-22 09:42:55 +02:00
Kacper Kasper
74689fa58a fix: enable cache env variables for level-zero 
Related-To: NEO-4262

Signed-off-by: Kacper Kasper <kacper.k.kasper@intel.com>
 2023-09-21 17:03:33 +02:00
Maciej Plewka
8658fdb04e fix: Use stack vec for api specific prefix 
Related-To: NEO-8388, GSD-6296

Signed-off-by: Maciej Plewka <maciej.plewka@intel.com>
 2023-09-21 16:10:54 +02:00
Dunajski, Bartosz
42496ac96d feature: initial support for patching regular in-order CmdList 
Related-To: NEO-7966

Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2023-09-21 14:20:50 +02:00
Maciej Bielski
97e7cda912 feature: Optimize intra-module kernel ISA allocations 
So far, there is a separate page allocated for each kernel's ISA within
`KernelImmutableData::initialize()`. Apparently the ISA blocks are often
much smaller than a 64k page, which leads to poor memory utilization and
was even observed to cause the device OOM error if a single module has
several keys.

Improve the situation by reusing the parent allocation (owned by the
module instance) for modules, which kernel ISAs can fit together within
a single 64k page. This improves the memory utilization on a single
module level.

Related-To: NEO-7788
Signed-off-by: Maciej Bielski <maciej.bielski@intel.com>
 2023-09-21 13:55:45 +02:00
Dunajski, Bartosz
b94f58abaa feature: debug flag to enable in-order events 
Related-To: NEO-7966

Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2023-09-21 11:22:48 +02:00
Latif, Raiyan
9bb559d468 feature: Ray Tracing Acceleration Structure (RTAS) Support 
Related-To: LOCI-3419

Signed-off-by: Latif, Raiyan <raiyan.latif@intel.com>
 2023-09-20 19:57:56 +02:00
Dunajski, Bartosz
020822fe29 feature: re-enable in-order semaphores for regular CmdLists 
Related-To: NEO-7966

Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2023-09-20 15:34:05 +02:00
Mateusz Hoppe
69f5ca6345 feature: bindless addressing - flush state cache after reusing SS slot 
- when Surface State is reused for new resource, State Cache needs to be
invalidated

Related-To: NEO-7063

Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
 2023-09-20 12:53:32 +02:00
Zbigniew Zdanowicz
09ab0ca1d2 performance: remove not needed validation loop 
- validation of arguments must be moved to validation layer

Related-To: NEO-7828

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2023-09-20 12:35:18 +02:00
Zbigniew Zdanowicz
40e169f8e2 refactor: add batch buffer attribute for explicit monitor fence dispatch 
Related-To: NEO-8395

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2023-09-20 12:18:32 +02:00
Zbigniew Zdanowicz
713511eea9 refactor: inline simple level zero queue method 
Related-To: NEO-7828

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2023-09-20 12:12:56 +02:00
Zbigniew Zdanowicz
453d01efb2 performance: remove extra loop processing command lists in execute call 
Related-To: NEO-7828

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
 2023-09-19 18:01:16 +02:00
Dunajski, Bartosz
9da60d8e07 feature: regular CmdList submission counter 
This is prework for in-order cmd lists

Related-To: NEO-7966

Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2023-09-19 16:26:38 +02:00
Dunajski, Bartosz
d3d5da1f72 feature: initial 64b in-order CmdList support 
Related-To: NEO-7966

Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2023-09-19 15:23:37 +02:00
Mateusz Jablonski
8da3879944 refactor: remove not needed code 
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2023-09-19 14:40:24 +02:00
Compute-Runtime-Validation
913a926fd4 Revert "feature: Optimize intra-module kernel ISA allocations" 
This reverts commit c348831470.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
 2023-09-19 14:16:05 +02:00
Maciej Bielski
c348831470 feature: Optimize intra-module kernel ISA allocations 
So far, there is a separate page allocated for each kernel's ISA within
`KernelImmutableData::initialize()`. Apparently the ISA blocks are often
much smaller than a 64k page, which leads to poor memory utilization and
was even observed to cause the device OOM error if a single module has
several keys.

Improve the situation by reusing the parent allocation (owned by the
module instance) for modules, which kernel ISAs can fit together within
a single 64k page. This improves the memory utilization on a single
module level.

Related-To: NEO-7788
Signed-off-by: Maciej Bielski <maciej.bielski@intel.com>
 2023-09-19 12:05:09 +02:00