9c462f9a6d
Disable worker thread creation for direct dispatch
...
Change-Id: I28f08ab9352310c9bf843fcb803a48f95ddf4676
[ROCm/clr commit: e4f51e063b
2020-11-30 17:50:12 -05:00
50138a93a3
Add CommandKindString to the log
...
Change-Id: Ie23123a85cff82b1732da85f5bffbff6958c02e5
[ROCm/clr commit: d943cae31f
2020-10-26 09:16:03 -04:00
d0b6c2b538
Improve queueLock and lastCmdLock
...
Reduce the size of the queueLock and lastCmdLock critical sections
to improve lock contention performance. The smaller the critical
sections are the better.
lasCmdLock is still needed to guarantee that getLastEnqueueCommand_
can retain the command before it is swapped out and released.
Change-Id: Id35d4a77c035b2da0de4c15568b153d49e958bb7
[ROCm/clr commit: 080dcfe857
2020-09-01 18:09:31 -04:00
5f5f1a3a84
Fix indentation with clang-format
...
Change-Id: I7aeadef3c613d5efc31a98e666bfb819ae34bdf5
[ROCm/clr commit: c95c613edc
2020-09-01 18:09:19 -04:00
20845c2560
Revert "SWDEV-241977 [ROCm QA] Random Soft hang observed while running TF and Caffe2 benchmarks"
...
This reverts commit c6649a8c7f1c24072d13
2020-09-01 18:09:10 -04:00
c6649a8c7f
SWDEV-241977 [ROCm QA] Random Soft hang observed while running TF and Caffe2 benchmarks
...
Change-Id: I42016c11db15411b86e7b8130d6ba557bc22dbb7
[ROCm/clr commit: ce038f3163
2020-07-22 02:03:48 -04:00
f14d79c587
Make append and setLastQueuedCommand atomic
...
Two threads can enqueue to the same HostQueue (HostQueue::enqueue)
and result in last queued command being the first one reachine queue_.enqueue
NOTE: Temporarly make setLastQueuedCommand empty function to pass the build
Change-Id: Id09c3a28d184986f52b2ec86a2f6a18c40df1f0b
[ROCm/clr commit: 3d15a1e291
2020-07-14 18:22:45 -04:00
ebe5c3d381
Remove extra barriers
...
Don't flush current batch if the dependent wait is a nop
Change-Id: I8a8722b9011fe042c1a4ce195938290fc75e7c86
[ROCm/clr commit: c18892a590
2020-06-22 12:41:02 -04:00
3d2182f8ba
Revert "Avoid lock for last queued command"
...
This reverts commit 88c3f77bed44bc0cb35d
2020-06-04 18:08:17 -04:00
88c3f77bed
Avoid lock for last queued command
...
Use atomics for last queued command update
Change-Id: I759e9d78ea72f23c0d45dbede6250b231e122276
[ROCm/clr commit: dc4e09a63a
2020-05-29 11:06:55 -04:00
992fbe8215
Use a dedicated lock for last queued command set/get
...
Change-Id: If3d2144841c7863cf7afe2ca85aea62e0a3a33c7
[ROCm/clr commit: 0782acabb5
2020-05-28 12:49:39 -07:00
dee687d2d7
Add support for setting CU mask on ROCclr for ROCm backend
...
Change-Id: I0dbe2eeb33467fc0f24b26929119c10e9b455da7
[ROCm/clr commit: fed94b8604
2020-05-15 14:23:43 -04:00
17f6a41982
removing AMD emails per palamida scan
...
Change-Id: If7307f5b1f81a43f2725ec5abd3b8989cbddbcc5
[ROCm/clr commit: 1b6f21ad9a
2020-03-11 21:26:55 -04:00
ebc40e115c
SWDEV-193956 - [hipclang-vdi-rocm][perf]
...
Retain command before a wait.
Change-Id: I0ab8f1352af32c96567dad9e31b3687bdf407bf6
[ROCm/clr commit: 77ff161324
2020-02-21 16:43:21 -05:00
039621e7b9
SWDEV-193956 - [hipclang-vdi-rocm][perf]
...
~45% to 50% of Performance drop on rocBLAS_int8 test
Use the last command in the queue for a wait.
Add extra print information about processed commands.
Add an option to disable file location printing.
Change-Id: I4187883e1a90e571fde3128af98368108fda8785
[ROCm/clr commit: a66d09f5a3
2020-02-21 15:21:15 -05:00
e284923583
Update copyright info
...
Change-Id: Ia4f9ff0f5f873b4223a8cca154188bb0d2f1abba
[ROCm/clr commit: b4c6143a2f
2020-02-04 09:26:14 -08:00
011f3e945b
Merge branch 'origin/pghafari/vdi-prototype' into lmoriche/amd-master
...
Change-Id: Id3b833d405596735becb3346f3b08c6da57033fe
[ROCm/clr commit: 20c7173849
2020-01-30 20:12:13 -08:00
German Andryeyev
Jason Tang
Laurent Morichetti
Saleel Kudchadker
Alex Xie
Christophe Paquot
Aryan Salmanpour
Payam