With direct dispatch enqueue occurs before callback update and
it can't be tracked in the device backend
Change-Id: Ie8793e3ddb68cc5bb36348f7a8dcdbdc87a2487c
[ROCm/clr commit: b0402a95bf]
Some upstream will load hip multiple times thus Rocclr static lib will
be linked to hip::amdhip64 many times. This will bring some unexpected
issues such as in hipBLAS. The patch prevents his happening.
Change-Id: I6bb27659f74371dae6e59c59fd6bb2022cc062ff
[ROCm/clr commit: f65348cfbe]
Flags of NVCC_OPTIONS need be sent to linker. Because compiler and
linker flager are mixed in NVCC_OPTIONS.
Change-Id: I3db37b962808566ea145e3cbdefa66d373e2d360
[ROCm/clr commit: f727a048d3]
- Don't notify if the batch is empty, because that means
the current command was processed already.
- Disable pinning optimization to avoid a race condition on stall.
- TS marker submition requires extra AQL barrier
to track the status.
Change-Id: I17eff4ad12ac66cfe1bb44048bebb1891805279d
[ROCm/clr commit: 24299e25bd]
1. Fix the size of the memory when releasing.
2. Make sure we only count the device memory
Change-Id: Ib4dcda79f313c4ee9cc1c7bab53f8076bce5f583
[ROCm/clr commit: 639d67866c]
which can cause segmentation faults. Hence checking for its sanity
Signed-off-by: Ashutosh Mishra <ashutosh.mishra@amd.com>
Change-Id: I78b4d029f0926a1369a8ebbeb4aef951a8f1f1d7
[ROCm/clr commit: b9c0065e4f]
Skip notification for markers with direct dispatch only,
since they are blocking always
Change-Id: I6bb17650f73371dae6e29c59fd6bb2012cc062fd
[ROCm/clr commit: a9b0e20d26]
This reverts commit be82aa43fb.
Hold wall time related updates till direct dispatch is ready.
Change-Id: I53b232f6f51bc2fc71b6b639fe0081e2907e9707
[ROCm/clr commit: 62c278c39d]
Since the allocation can be a suballocation, we should print both the
VA range for the allocation and the underlying memory object.
Change-Id: Ic9c707bbb78113b366d1b2c688e6fd33bdc8fd94
[ROCm/clr commit: 9e8a2f3266]
[dtest] Tests for hipEvent related APIs
Added Negative scenarios for hipEvent related APIs
1. Verifying all hipEvent related APIs by passing nullptr.
2. Pass illegal/unknown flag to hipEventCreateWithFlags API
Change-Id: Ia0a24065d16fe0f5ee28a88e280c25c1be0c3590
[ROCm/clr commit: 2a0593258c]
Direct disaptch doesn't insert extra barriers for Markers if
AQL barrier was the last issued command already.
Change-Id: I00fbc658547d83dd3ee64ec391ed50e5f8a08e30
[ROCm/clr commit: 0587fb7450]
If MT is enabled, then a new callback can be received before the previous
command is processed, causing a conflict of 2 callbacks.
Change-Id: I5ff8f231208e8d62824d590d3c8e791e8e36affb
[ROCm/clr commit: fe7501a64b]
runtimeApi/event/hipEventElapsedTime will report invalid resource error
on cuda due to wrong calling sequence. The fix will arrange the calling
in right sequence.
Change-Id: I3db28a962888566ea135e3cbdefa68d373e2d369
[ROCm/clr commit: b9cf12e920]
Vladislav Sytchenko
German Andryeyev
Sarbojit Sarkar
Chauncey Hui
Tao Sang
Ravi C Akkenapally
kjayapra-amd
Julia Jiang
Jatin Chaudhary
Alex Xie
Ashutosh Mishra
Payam
Satyanvesh Dittakavi
Rahul Garg
Arya.Rafii
DURGESH KROTTAPALLI
Jason Tang