# Intel SYCL and LLVM
**URL:** https://discourse.julialang.org/t/intel-sycl-and-llvm/30046
**Category:** Community
**Created:** [October 18, 2019, 8:45am UTC](https://discourse.julialang.org/t/intel-sycl-and-llvm/30046 "2019-10-18T08:45:34Z")
**Posts on this page:** 2
**Page:** 1
### Author:  [@johnh](https://discourse.julialang.org/u/johnh)
#### Post date: [October 18, 2019, 8:45am UTC](https://discourse.julialang.org/t/intel-sycl-and-llvm/30046/1 "2019-10-18T08:45:34Z")
I am unfamilair with Intel SYCL . However does this discussion on Intel LLVM compilers have any relevance to Julia?
[https://github.com/intel/intel-graphics-compiler/issues/62#issuecomment-509912798](https://github.com/intel/intel-graphics-compiler/issues/62#issuecomment-509912798)
---
### Author:  [@jpsamaroo](https://discourse.julialang.org/u/jpsamaroo)
#### Post date: [October 18, 2019, 12:18pm UTC](https://discourse.julialang.org/t/intel-sycl-and-llvm/30046/2 "2019-10-18T12:18:12Z")
All that really matters for Julia support of Intel GPUs is having a functioning LLVM target (like AMDGPU or NVPTX) that we can emit native code for, and a compute runtime that can accept native code and launch it as a kernel on the target GPU (as well as all the other things a good GPU runtime does). Both of these things seem to exist, although I don’t think it’s worth starting on supporting this in Julia until Intel upstreams their LLVM changes to an LLVM release that Julia can run with.