JackAKirk
JackAKirk
> Why is this not relying on the existing device-cache? I would expect that the new "device_impl" would not be created because of this code: > > https://github.com/intel/llvm/blob/f6420c769d473c00be736b8df361e10dd98a6009/sycl/source/detail/platform_impl.cpp#L219 > >...
> > that `createSyclObjFromImpl` does not call the `Platform->getOrMakeDeviceImpl` method that checks the device-cache. This is why we needed to call `device::get_devices` that does check the device-cache and as far...
> > `MDeviceCache` is a private member of `platform_impl` so I thought it could be controversial to let `make_device` access it directly? It is true that if `make_device` were to...
> @smaslov-intel What do you think about my above comment/questions?
btw the CUDA test failure is the now XFAIL'ed known flaky test Assert/assert_in_simultaneously_multiple_tus.cpp
ping @v-klochkov
/verify with https://github.com/intel/llvm-test-suite/pull/1280
/verify with https://github.com/intel/llvm-test-suite/pull/1280
> Looks good to me. Thank you. Before proceeding to merge. Please add a test verifying the test case being fixed by this PR. > > If there is an...
Can this be merge now?