Andreas Noack
Andreas Noack
Yes. I can do that once Intel's registration system works again. Just tried and it currently isn't working.
Ref https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Possible-issue-with-mkl-sparse-x-mv-when-using-CSC/m-p/1220155#M30192
Not really. On 30 October, I was told that > We will check the issue ASAP and get back to you. and I haven't heard more. I've just asked for...
Intel has just gotten back to me to tell me that they were sorry for not replying earlier. No further details. To be continued.
Unfortunately not. Actually, I asked for an update four days ago but haven't heard back from Intel.
Some progress. They just replied that they have run the reproducer I shared, that they are able to reproduce the bug, and that > The issue has been escalated to...
Just got the following update from Intel > The issue is in progress.
Today's news from Intel > The Issue has been fixed and will be available with MKL 2021.U3
Please avoid calling the BLAS wrappers directly if possible. It would be better to call a generic version.
It's a general problem. See also https://github.com/JuliaLang/julia/issues/25565. In Julia, we have generally associated *vector* with 1d arrays and *linear maps* with 2d arrays but, as you mention, it doesn't have...