Waheed Barghouthi

Results 18 comments of Waheed Barghouthi

I believe you just have to define the tunnel name and it will work just fine. for example:- `hans -c 127.0.0.1 -d tun0 -m 1321 -q`

Hello folks, I hope you are doing well. And I would like to thank @mefistotelis for this nice tool, I actually happened to have a laptop with this chip used...

> * `gfx1010` is probably best to be treated as `gfx900`, because even though gfx1030 is theoretically the closest, there are optimizations and LLVM builtins that gfx1010 cannot use while...

I just hit this error while trying to add the gfx1030 features to gfx1010, `Cached asm caps differ from derived asm caps for (10, 1, 0)` which is expected, as...

Thanks @trixirt , yes, I will also add this issue to it: below is actually the compileArgs command, and the first element is None, which i guess is the cxx_compiler(it...

Thanks for having a look @GZGavinZhao appreciate all your efforts. > RDNA1 and RDNA2 are not the same instruction set. For example, [consult](https://gpuopen.com/amd-gpu-architecture-programming-documentation/) section 6.3 of the RDNA1 and RDNA2...

> Note that only for CK, `gfx1010` has the same code path as `gfx900`. For Tensile/rocBLAS, `gfx1010` is being compiled normally, so it receives all the optimizations that `gfx1010` should...

Right now I'm just reading some weird stuff like this: `defm V_DOT2_F32_F16 : VOP3P_Real_gfx10_gfx11_gfx12_Triple;` this is basically in some llvm file I think this one, llvm-project/llvm/lib/Target/AMDGPU/VOP3PInstructions.td and it shows that...

I think I understand what you mean, but you did not understand what I mean :) It worked! right now here is my results for building for ck gfx1030;gfx1010;gfx803 with...

> Really nice! I'm curious, what did you change? Did you only update CK, or did you also update the TableGen files in LLVM as well? Thanks! I actually changed...