llvm

llvm copied to clipboard

luszczewskakasia1

Initially reported by @slawekptak ### Describe the bug Setting external events onto an in-order queue from a queue that is a recording to a queue that is not recording to...

mmichel11

[WIP] Reproduce issue19033 urEnqueueKernelLaunchIncrementMultiDeviceTest fail

2

rafNNN

[UR][L0v2] Add initial record & replay implementation

3

kswiecicki

[UR][L0 v2] Implement zero-copy buffers for integrated gpus

1

Implementation of zero-copy buffer operation in integrated GPUs. The GPUs share memory with host - therefore some buffer operations can be zero-copy, unlike in discrete GPUs.

mateuszpn

Use perfect forwarding in queue inline implementations. In the same place use event move if possible. For sake of performance.

lslusarczyk

[L0v2] add submitted kernel vector compaction

1

L0v2 avoids internally tracking each kernel submission through an event for lifetime management. Instead, when a kernel is submitted to the queue, its handle is added to a vector, to...

pbalcer

[SYCL] Support sycl_ext_oneapi_platform_device_index

1

https://github.com/intel/llvm/blob/sycl/sycl/doc/extensions/proposed/sycl_ext_oneapi_platform_device_index.asciidoc

KornevNikita

Also report warning to user if device memory is out of resource.

zhaomaosu

zhaomaosu