intel/compute-runtime
1
0
Fork 0
mirror of https://github.com/intel/compute-runtime.git synced 2025-12-26 15:03:02 +08:00
Code Issues Packages Projects Releases Wiki Activity
1,197 Commits 42 Branches 655 Tags
c3be3d9ef3134d4587b2c402667ab43217b2193d
Commit Graph

144 Commits

Author SHA1 Message Date
Pawel Wilma
8c1db4fb2f Code cleanup for reduced GPU address space 
Change-Id: Ibce79ddbe1f03dac1813b5dc2356a9db86b60200
Signed-off-by: Pawel Wilma <pawel.wilma@intel.com>
 2018-09-10 16:16:06 +02:00
Dunajski, Bartosz
7ec0989eea Obtain new TimestampPacket before releasing old node 
Change-Id: I243d50e32967a0accea2d73fdd5ff4935bc03449
Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2018-09-10 15:58:00 +02:00
Dunajski, Bartosz
24136414e6 Use Semaphore to wait for dependencies from different device 
Change-Id: I154f377c77847e93d5b188a5b2252e74d9d70b75
Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2018-09-10 10:00:49 +02:00
Dunajski, Bartosz
d04614dce3 Use Semaphore to wait for dependencies on the same device 
Change-Id: Ib04c960c50183c080d02753815ece80b58d1980e
Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2018-09-07 22:34:44 +02:00
Mrozek, Michal
393ce116e7 Remove flushWaitList method. 
- No longer needed.

Change-Id: I9e255067fb4b0d52a42f6a49145b3a8d591b5e74
 2018-09-07 15:27:37 +02:00
Dunajski, Bartosz
eb6b574561 Improve TimestmapPacket write commands size estimation 
Change-Id: Idf80401f4360342a162d54aba7ffbe6dfcf714e8
Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2018-09-06 16:13:07 +02:00
Dunajski, Bartosz
b74280beb6 Check EnableTimestampPacket debug variable once and set as CSR mode 
Change-Id: Ia6e7caa96f3b46b30590fb46a1fb37fa153adeb4
 2018-09-06 11:19:02 +02:00
Dunajski, Bartosz
1b04210c6a Extend TimestampPacket data with Submit field 
Change-Id: Ia2a23003476db06afeee25ca7766cfd1bc6da4a3
Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2018-09-06 10:37:03 +02:00
Pawel Wilma
4a12deea2b Add support for reduced GPU address space 
Change-Id: I9ebbc8c51039bb533b44c6b80e717e1489a20a43
Signed-off-by: Pawel Wilma <pawel.wilma@intel.com>
 2018-09-03 13:51:36 +02:00
Filip Hazubski
85d7e0e4c8 Refactored GPGPU walker for easier maintenance 
Change-Id: Ie10272f363d3e583893c46123bb7a56a5fd9ac45
Signed-off-by: Filip Hazubski <filip.hazubski@intel.com>
 2018-09-03 11:04:24 +02:00
Mateusz Jablonski
92bfd2e3d2 Move OsContext to Device 
Change-Id: I030b65372fbdc075423d22720e9da34ac65b8e68
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2018-09-03 10:42:26 +02:00
Dunajski, Bartosz
c014c49102 Pass TimestampPacket to dispatchWalker during enqueue call 
Change-Id: Ia2b9206d4c2bb69232bfe9e7d68d195f3a558926
 2018-08-31 19:57:47 +02:00
Dunajski, Bartosz
570c0843ef Handle TimestampPacketNode residency 
Change-Id: I1769d67426ca704b600931b58d3f505bef0e893d
Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2018-08-30 11:53:56 +02:00
Dunajski, Bartosz
41914d3058 Make enqueue blocking if parent kernel requires aux translation 
Change-Id: I678e1045d84f15e30223a99438bbb7057e172cff
Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2018-08-29 19:33:00 +02:00
Dunajski, Bartosz
da0f9381dc Timestamp Packet ownership 
- Tag allocator: reference count tracking
- Obtain tag by command queue and pass to Event if exist during enqueue
- Handle Timestamp Packet lifetime on Event and CmdQueue destruction

Change-Id: I9a5969830ea7a9d729e6f70519d8c28ff70fcf06
Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2018-08-29 13:47:46 +02:00
Dunajski, Bartosz
a807b9a90b Initial implementation of Timestamp Packet write 
Change-Id: Ic498bcf9795f54fbb5fb5a8d07ed17fa70dc4f1a
Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2018-08-28 08:27:13 +02:00
Milczarek, Slawomir
f6f9c0f456 Add notify functions to enqueue read buffer and image calls 
This commit adds notifications to enqueue read buffer and image calls
and setters/getters to mark/check if an allocation is dumpable.

Change-Id: I123f24752d2a86abcf934e0d404f4e0ecf1729cc
 2018-08-22 21:43:17 +02:00
Maciej Dziuban
b91c14f70e Delete Device::getBuiltIns() 
Change-Id: I9d1968dfb2ba4a56020fd17152119add726106e1
Signed-off-by: Maciej Dziuban <maciej.dziuban@intel.com>
 2018-08-22 16:54:53 +02:00
Mateusz Jablonski
6286f245a1 Fix generation local ids for image layout with local workgroup size 12x12x1 
Change-Id: Ib723b132b570d8cfb3f72f32ddadde869607c354
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2018-08-22 14:32:07 +02:00
Maciej Dziuban
e0e48203d2 Move BuiltIns to ExecutionEnvironment 
Change-Id: Ib2a1b82cc7858c898bb32820aad106a01d1325ad
Signed-off-by: Maciej Dziuban <maciej.dziuban@intel.com>
 2018-08-21 23:15:47 +02:00
Dunajski, Bartosz
931b462ee1 Disable NonAux to Aux translation for Parent Kernel 
Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
Change-Id: I863608fe3652e7e777a1e841d79b5b56e7362a3f
 2018-08-21 15:12:25 +02:00
Mateusz Jablonski
7afba8d50b Cleanup after adding new local ids layout for images 
- Apply layout for images only when Z size is equal to 1
- Fix generating local ids for local workgroup size
  when any size is not power of 2
- Revert commit c53c09da45

Change-Id: Ie745782fafce2facbd877e3e33e4ba347cb2b09e
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2018-08-21 11:27:30 +02:00
Dunajski, Bartosz
044255e9bd Pick Main Kernel for LWS and numWG in dispatchWalker() 
Change-Id: I4fd0746ec77890ceacbf333966bb00a4ea99b186
Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2018-08-20 13:51:35 +02:00
Dunajski, Bartosz
c7a49666d5 Refactor querying Main and Parent Kernel from MultiDispatchInfo 
Change-Id: I723d91f2f445bc7af1bcb0de46f8ac07837f3449
Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2018-08-17 13:51:32 +02:00
Mrozek, Michal
c53c09da45 Limit local work sizes where local ids limit is applied. 
Change-Id: Id9a84d6a7d4530344771f48fd278cff9ab2dd927
 2018-08-16 12:34:09 +02:00
Mateusz Jablonski
47f3dad619 Apply (2/4)x4x1 layout when generating local ids for kernel with images only 
- For SIMD8 apply 2x4x1 layout
- For SIMD16/SIMD32 apply 4x4x1 layout

Change-Id: I31bceb49387011c66da5f96ad2a71125b96d4cda
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
 2018-08-14 12:22:20 +02:00
Dunajski, Bartosz
a5950500a3 Aux translation [4/n]: Lock BuiltIn Kernel + refactor BuiltIns locking 
Change-Id: Ic7dc9b86a4aa5f93f1c4bcdf80b9598ecdff9713
Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2018-08-14 10:56:16 +02:00
Lukasz Towarek
95e28faca0 Fix division by zero in enqueueKernel 
Change-Id: I8e7d3db39805133a5af545e65a94fb19433a2a41
 2018-08-14 09:02:17 +02:00
Dunajski, Bartosz
6ca84c278a Aux translation [3/n]: Dispatch AuxTranslation builtin when required 
Change-Id: I9bd0294de7980ac01ebb3c2d696eba6fd6a456ec
 2018-08-13 12:15:30 +02:00
Dunajski, Bartosz
b4f53fdfa7 Pick applicable buffers for aux translation 
Change-Id: I60a28cd9e0dec61120b1ae5c42dfe0cb852eb387
 2018-08-08 09:23:51 +02:00
Chodor, Jaroslaw
c10d0d79f5 Workgroup walk order 
Change-Id: Id02db6a383e21dc17be64655e7f51a84103b2e0b
 2018-08-07 13:54:10 +02:00
Mrozek, Michal
d80dbb1ae0 Do not take ownership on device. 
- Command Stream Receiver should be used instead for locking.
- Remove not needed synchronization in clSetUserEventStatus

Change-Id: I17050dc70cb0be03b2003043a9666ba8df1a83c9
 2018-08-07 09:29:50 +02:00
Dunajski, Bartosz
ec6f0f9f86 Aux translation [1/n] 
- Mark Kernel for aux translation
- Initial implementation of dispatchAuxTranslation for future use

Change-Id: Ifca1c9a893876eecc5678cdc4f564b2bfcae959a
 2018-08-07 09:07:25 +02:00
Mateusz Jablonski
9ae4f390d1 Remove command queue, completion stamp and device from mem obj 
Remove setCompletionStamp function from Surface

Change-Id: I25f3040a91892495e55cb4924f1538276de6264e
 2018-08-01 16:17:13 +02:00
Mrozek, Michal
f60847b64e Pass device to flushTask. 
- do not obtain it from memory manager

Change-Id: Icc7c03dc925c69ec5932c5812151ac28dc34d20d
 2018-08-01 14:11:06 +02:00
Dunajski, Bartosz
239ebf9eab Improve AllocationType operations: dont do bit operations on enums 
Change-Id: Ie70ca9e2a93ec80b1cd655bad622db9e12abb7f7
Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
 2018-07-20 09:12:58 +02:00
Mrozek, Michal
2a1feaffe9 Initialize stack variables. 
Change-Id: I3fda7163cb649fe4754c46d83c641921593e1823
 2018-07-19 16:19:17 +02:00
Mrozek, Michal
1dc50172d3 Remove redundant casts. 
Change-Id: I9bfe4615ad0eea739ad0a780426abc007163c961
 2018-07-17 19:02:36 +02:00
Hoppe, Mateusz
55a045ebe1 Refactor graphics memory allocation scheme 
- replace createGraphicsAllocationWithRequiredBitness with more general
methodallocateGraphicsMemoryInPreferredPool based on passed
 AllocationData
- proper flags for allocation selected based on AllocationType

- remove allocateGraphicsMemory(size_t size, size_t alignment)
and use allocateGraphicsMemory(size_t size) instead where default
alignment is sufficient, otherwise use full options version:
allocateGraphicsMemory(size_t size, size_t alignment,
 bool forcePin, bool uncacheable)

Change-Id: I2da891f372ee181253cb840568a61b33c0d71fc9
 2018-07-11 15:48:05 +02:00
Hoppe, Mateusz
684b1d75ba Refactor GraphicsAllocation::AllocationType and allocationType enums 
- change GraphicsAllocatoin::AllocationType to scoped enumeration
so that ALLOCATION_TYPE_ prefix in every enum value can be removed
- all accesses are typed (example AllocationType::IMAGE)
- Rename allocationType to AllocationUsage to eliminate confusion
with multiple AllocationType enums / types

Change-Id: I16003297ecfcb0aaa5779ad00706c5d983914bbe
 2018-07-06 13:00:08 +02:00
Milczarek, Slawomir
eb1b5ded9c Add support for AUB subcapture (filter and toggle modes) 
This commit adds a capability to selectively enable/disable AUB capture,
i.e. by toggling the registry key from the outside or specifying the filter
with a kernel name and/or kernel start index and kernel end index.

Change-Id: Ib5d39c21863fbc4a95aa73c949b9779ff993de0f
 2018-06-15 13:02:27 +02:00
Artur Harasimiuk
75ab0c6fe1 Switch clang-format to 6.0 
Change-Id: Id96d1f47fb3d479d10d1022f1259dc030a148192
Signed-off-by: Artur Harasimiuk <artur.harasimiuk@intel.com>
 2018-06-14 09:45:00 +02:00
Zdunowski, Piotr
0cc10e47cc Use device instead of context when programing surface state. 
Change-Id: I67615036d373cf905762a43a92562bf3d84854a5
 2018-06-11 17:20:11 +02:00
Woloszyn, Wojciech
8a488ad52f Fix reported row/slicePitch for mip-maps 
- use information from gmm correctly
- modify computation on gen8

Change-Id: Iaefcc20ce9436ef70cd2f4bc36654932c4b5af49
 2018-05-22 10:36:54 +02:00
Zdanowicz, Zbigniew
33fee15711 Change interface to pass Interface Descriptor Data as pointer 
Change-Id: I0f33109b800a7607206954bb1e5cb0826290e6f3
 2018-05-11 14:35:53 +02:00
Mrozek, Michal
34ff5852eb Add capability to csr to allow N:1 aggregation when ooq is created. 
- This allows applications to force the N:1 aggregation by creating out
of order queue.
- That switches csr to N:1 submission model where commands from multiple
command streams may be aggregated.
- That forces scenarios returning an event to be aggregated as well.

Change-Id: I8fd8d7f88bb2665234ee90870133120b206710a8
 2018-04-26 15:41:20 +02:00
Mrozek, Michal
8d2df3c332 Move indirect heaps from command queues to csr. 
-This is required to enable N:1 submission model.
-If heaps are coming from different command queues that always
mean that STATE_BASE_ADDRESS needs to be reloaded
-In order to not emit any non pipelined state in CSR, this change
moves the ownership of IndirectHeap to one centralized place which is
CommandStreamReceiver
-This way when there are submissions from multiple command queues then
they reuse the same heaps, therefore preventing SBA reload

Change-Id: I5caf5dc5cb05d7a2d8766883d9bc51c29062e980
 2018-04-26 14:05:40 +02:00
Pawel Wilma
a0c044e6d2 Extend batch buffer flattening in AubCSR to BatchedDispatch mode 
- batch buffer flatening in batched mode
    - added MI_USER_INTERRUPT command
    - added GUC Work Queue Item

Change-Id: I35142da34b30d3006bb4ffc1521db7f6ebe68ebc
 2018-04-26 12:45:02 +02:00
Zdanowicz, Zbigniew
d94b853c32 Separate HW commands from class declaration and interface 
Change-Id: Ia49098171f4b7814c42a35686354713a322c9df7
 2018-04-25 14:03:00 +02:00
Mrozek, Michal
ce8c44cae3 Add check for local work group size in clEnqueueNDRangeKernel call. 
- Incoming local work group size cannot exceed device capabilities.

Change-Id: I89a7503155c71443e3ebc630debb5d5b466c6cb5
 2018-04-20 08:16:16 +02:00