The change enables VM support in graphs on Windows. That allows
to avoid caching of all allocations at the cost of map/unmap
overhead during memory create/destroy.
Change-Id: I792be00fba099e5e5d3cd44a963e1dfd6976a86d
GraphMemcpyNodeSetParamsFrom/ToSymbol APIs neew to check device id for
original src/dst is same as what is passed in while set.
Change-Id: If0b610808223dce9115562bb5e9b31c8eaa2df22
- before removing node from graph all edges should be removed and rest
of graph updated
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com>
Change-Id: Ide0afcc964f87f13cf407c971e22497433e3b1ed
- if graph is already instantiated there cant be a second instance in
case nodes are free or alloc
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com>
Change-Id: I51e2ca8ade24799da96ed126a26c5ea3bad6f452
- invalid to synchronize or query the execution status of a capturing stream but non-capturing streams can.
- If any local thread has an ongoing or concurrent capture sequence initiated with hipStreamCaptureModeGlobal, it is prohibited from unsafe calls
Change-Id: Ifa641e807216d3b7c3e8c2fb1be2f7a50bd641df
- test fails intermittently because object is not cleared correctly
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com>
Change-Id: I88daf3dc08bb83d6d3f047ff48a63c8f856fb0bf
- remove node dependency before checking parents
- reduce edge level acording to new value of node
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com>
Change-Id: Id4bff1684f7e0b42beeebc4d2e009bfdb507fb5f
- nodes should belong to graph
- num of passed dependency cant be larger than graph nodes
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com>
Change-Id: Ia6f2283546bc44edee705b0483bfe506b7b1177a
- stream capture should be done before any sync APIs.
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com>
Change-Id: I3d65f67ee68777be71f97f48d460ccaefdd4e1af
Add memory allocation support in graph. Current implementation uses
cache from mempool to hold the allocations which belong to the graph.
Also the resource tracking is disabled at this moment because mempool
operates with hip::Stream objects, but graph has execution with
amd::HostQueue objects.
Change-Id: I54fe3250126d24f5a26ada975f37d429bb4ef17b
- GetNode should retrun nodes based on the dependency between them
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com>
Change-Id: Ie8ecdd976f8dc9d9b5dc2adc6ae8f879912c8169
- return implicit error if null sream is passed while blocing stream is
in operation
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com>
Change-Id: I4d867a0bddd855e2c0d3eb9b769d5d36412c20c0
- Aggregate all TLS(Thread Local Storage) variables into a single class
- This is to improve cache accesses per thread
Change-Id: Ic8361eaeae290fff00254684e309471958365eb9
For hipGraphKernelNode, remove func_;
and reorganize functions to naturely support mGPU;
For hipGraphMemcpyNode, make EnqueueCommands() support different
queues' sync
Change-Id: I22708923f454adf4456ff99d25559daffed8c20d
German
Jaydeep Patel
sdashmiz
Sourabh Betigeri
Ioannis Assiouras
Jatin Chaudhary
pghafari
Anusha GodavarthySurya
Rakesh Roy
Jiabao Xie
Tao Sang