- Added the optimized multi stream path in graph execution. It uses a fixed number of async streams in the execution
- Optimize the launch latency, where commands
creation and execution is done at the same time
- Optimize the scheduling to use less barriers and waiting signals if
the same queue can be detected
- The new path is controlled by DEBUG_HIP_FORCE_GRAPH_QUEUES
environment variable, where 0 will use the original path and any other
value will force the number of asynchronous queues for execution
- DEBUG_HIP_FORCE_ASYNC_QUEUE can force single queue async
execution in graphs(applicable for Navi families only)
Change-Id: I7eb40bc15c45f508d6911868a6f6d4c3598d380e
[ROCm/clr commit: 9db52f9a46]
=> Added support to capture multiple AQL Packets.
=> Added Interface to callback to hip runtime from rocclr to allocate
kernel args from the graph kernel arg pool.
=> Enabled Support to capture memset node.
Change-Id: I7e1c2ba06927459e024653058af142bd82192c43
[ROCm/clr commit: bd3a35bde1]
hipDeviceSynchronize called from __hipUnregisterFatBinary
accesses static maps and monitors. This change ensures these ojects
are not destroyed before __hipUnregisterFatBinary is called.
Additionally it disables the teardown process for static build.
Change-Id: I46b58641d60efcf6637a8e99cdd786ffe9e2c77d
[ROCm/clr commit: 9b33db9b24]
This issue was happening because of incorrect usage of getStream call,
if we get the null stream first and then typecast it, and call on
getStream again, we lose the advantage of simply passing "nullptr" to
indicate NULL stream. Thus we enter the waitActiveStream call and add
barriers to sync across streams.
Change-Id: I94dc4e3ec927295b9e1ab6dee4b37d7d3e00b0cc
[ROCm/clr commit: cda4b7db1c]
Also in the scope of SWDEV-467540.
Fix sporadic crash in Unit_hipStreamAddCallback_MultipleThreads by
deferring release() of block_command.
The test will invoke 1000 threads on the same stream thus there
is a chance to free block_command too early in original code.
By deferring release() of block_command we can make sure block_command
is always valid during calling block_command->notifyCmdQueue().
Change-Id: I31555ee18e6958e34b89f04181867fa4e932a38c
[ROCm/clr commit: e3ef19e22a]
Creation of ReferenceCountedObject will increase reference count by 1.
Clear the commands from Node after capture so that they wont be reference later.
Change-Id: I1cc4085939cf65218ec2aa2e25ab6d737f7cacd3
[ROCm/clr commit: 6ae5d6896c]
Capture AQL packets during GraphInstantiation and enqueue AQL packets during graph launch.
Added support to capture single graph memset node.
Capture support for memset node is currently disabled.
Memset capture will be enabled when capture for multiple packets are supported..
Change-Id: I14dfbc41731025cc3a548a730558915def3fa384
[ROCm/clr commit: 346da4bb40]
On some platforms user can ask for extended shared memory for a
particular kernel in some cases. This feature does not exist on HIP at
the moment. So we are setting it to sharedMemPerBlock which is the
maximum user can expect for their kernels.
Change-Id: I81005cf0d1c9fb941e77d34fb8385241ffe5bdd0
[ROCm/clr commit: 4b95e7bc87]
For refactoring of childGraph to have its own graphExec,
kernelArgs needs to be separated from the graphExec object.
All the childNodes part of graph should share same kernelArg pool.
Otherwise we endup creating multiple device kernel arg memory chucks
for single graphExec.
Change-Id: I4029a46ebc1fa112d87df64ab1fecbf288fabe5e
[ROCm/clr commit: 35079e834e]
This change modifies the readback mechanism to use a pointer to volatile
instead of a volatile pointer. This ensures that the compiler does not
optimize away the read operation.
Change-Id: I79ff925d615aa8cc4f950e8ff4b7e608fcb179a4
[ROCm/clr commit: ea50d2c0c2]
- Introduce a lock when checking isUserObjectValid. We need a lock
here as one can remove the userObject T2, leading to buffer overflow
when checking ranges in T1.
Change-Id: I058144b8cc463c90ab6bf5cf96bf937897742917
[ROCm/clr commit: 6ac67afdd5]
If graph has multiple branches, End command is enqueued on launch stream which
makes sure all the internal parallel streams are finsihed.
When node is removed from the graph, indegree and outdegree are not getting update correctly for parent, child nodes and
resulting in endNode not having deps on parallel commands. Resulting in graph sync issues.
Change-Id: I33cc2f21220e1c017d88099b29b542e05b683f73
[ROCm/clr commit: 9ad7e79e50]
Resolved an issue where a freed virtual buffer was incorrectly
added to the global mapping causing an assertion error during
teardown process.
Change-Id: I4801157a28603ce9be1ca0131982b700ff884f7a
[ROCm/clr commit: 1c6b92627d]
Changed find_package call to prioritize the package that is
found under the rocm installation over other system locations
Change-Id: Ice93c94bbb9cdebd467d3e88bb2e4bfb7a1e76d9
[ROCm/clr commit: 6b9e89fe0c]
If the graph has kernels that does device side allocation, during packet capture, heap is
allocated because heap pointer has to be added to the AQL packet, and initialized during
graph launch.
Handle race with wait when 2 kernels with device heap are enqueued on multiple streams.
Change-Id: I45933b77fcaf7bc8fdf1bc906462e32b5d8d3688
[ROCm/clr commit: 57156c524d]
Adding a safety check prevents an invalid memory access
if timestamps and kernelNames vectors are of different size.
The patch also moves the addKernelNames for the accumulate command
into dispatchAqlPacket function.
Change-Id: Iea0927e1253800403a1ae3f3d72de1e7d96476c3
[ROCm/clr commit: d44f44a5b1]
- Update the intra socket weight for partitions within single socket as
it is changed to 13 by the driver.
- Use the PCIe function to distinguish the partitions of the same device
such as TPX mode in gfx942.
Change-Id: I8e64023d44e37c2dbb105cbb343441a48021ba7b
[ROCm/clr commit: 1815fc808d]
Jaydeep Patel
German Andryeyev
Anusha GodavarthySurya
Julia Jiang
Marko Arandjelovic
Vladana Stojiljkovic
Ioannis Assiouras
Saleel Kudchadker
taosang2
kjayapra-amd
Jatin Chaudhary
sdashmiz
Rahul Manocha
Branislav Brzak
Satyanvesh Dittakavi