wallnuss

7 ms·

1.
Enzyme: Performing Automatic Differentiation on LLVM IR for C/C++(enzyme.mit.edu)
2 points by wallnuss 6y ago|0 comments
2.
by wallnuss 6y ago
Julia has an interesting split here, it does the lowering into SSA from in pure Julia and then has a codegen steps that translates the SSA from into LLVM IR, but for that second step we do use the C++ API. We have very robust bindings to th
3.
by wallnuss 6y ago
Arcanist is a but of a bother, but please do submit the patch.
4.
by wallnuss 6y ago
Having recently had the pleasure of having to debug JIT-compiled code with an ABI mismatch, I can't overstate how useful `rr` ( https://github.com/mozilla/rr ) can be to debug assembly. The ability to `rsi` e.g. rev
5.
by wallnuss 6y ago
Julia is particularly sensitive to LLVM compile times. LLVM 11 did improve those over LLVM 10 (which had noticeable regressions versus LLVM 9). As an example see https://github.com/JuliaLang/julia/issues/37915
6.
by wallnuss 6y ago
I don't disagree with the sentiment, but using the moon as an example is a bit far-fetched, as well as the metric being time. I thinkl it would be fair to say: Using the same amount of resources, how hard is it to reach a given point?
7.
Reliably Building Binariews for Distribution(binarybuilder.org)
1 points by wallnuss 6y ago|0 comments
8.
by wallnuss 6y ago
It all depends on which optimizations you enable and LLVM is very flexible, albeit sometimes you still spend 20% of your time in ISel (Instruction Selection)...
9.
by wallnuss 6y ago
It's being worked on. I am rather excited for the upcoming work that makes the parser replaceable and allows us to actually give good syntax errors! There is some discussion about making error printing more configurable so that one can
10.
by wallnuss 6y ago
I recently spend two weeks on and off hunting down a bug on a platform that didn't support `rr`. I am fairly confident to say that if I had `rr` available it would have taken me a couple of hours at most. Being able to run backwards fr
11.
KPTI/Meltdown mitigations lead to slowdown of up to 800% on Linux(brendangregg.com)
2 points by wallnuss 7y ago|0 comments
12.
by wallnuss 7y ago
We are indeed interested in targeting AMD GPUs. There is a prototype backend available at https://github.com/JuliaGPU/AMDGPUnative.jl and we are closely following the status of SPIR-V and Intel GPUs in LLVM. The focus
13.
Speculative JIT Compilation in LLVM(preejackie.github.io)
1 points by wallnuss 7y ago|0 comments
14.
by wallnuss 7y ago
Über just means above. So "über dir" is "above you". Urban dictionary has uber in english to mean superior, but that is not the original German meaning.
15.
by wallnuss 7y ago
It's not about Upwork intentionally skewing the results as you pointed out, but that there is a potentially sampling bias. Upwork might only attract certain kinds of programming jobs and those are less likely to use Go and Rust.
16.
by wallnuss 7y ago
As far as I know he bought that license for one of his student at the time, Alan Edelman, who then ended up co-creating the Julia language.
17.
by wallnuss 7y ago
I especially use rr as an exploration tool. It allows me to ask not the question "Where does the program go from here", but rather "Where did it come from". I couldn't do my daily work without rr, since often I work
18.
by wallnuss 7y ago
It does by default so you can use gdb/lldb and perf on Julia code.
19.
by wallnuss 8y ago
If Intel had one, they would have further consolidated their hold on the HPC/Supercomputer market. Summit is an interesting supercomputer because it's NVidia/Mellanox/IBM instead of the Intel hegemony that Cori II was, i
20.
by wallnuss 8y ago
I particularly enjoy that Flux.jl is an AD framework and a couple lines of Julia code defining convenience functions for ML, and it is all written in Julia so it invites exploration. Tim Besard gave a talk at a Tensorflow meetup recently th
21.
by wallnuss 8y ago
You can revoke OAuth tokens at https://accounts.google.com/b/0/IssuedAuthSubTokens?hl=en
22.
by wallnuss 8y ago
Take a look at CUDAnative.jl ( https://github.com/JuliaGPU/CUDAnative.jl ) which uses the NVPTX LLVM backend to compile Julia code to the GPU. What you described sounds very similar to that and could definitely made to w
23.
by wallnuss 8y ago
One of the author here. Working on this paper was fun, one of the most interesting lessons I learned was that divergence on GPUs has dramatically fallen in cost on modern architectures.
24.
by wallnuss 8y ago
There is also llvm-mca, which is kinda neat since it uses the I formation that LLVM has about the op cost. Contrasting that to IACA or OACA is valuable as well.
25.
ZStd is faster than Leviathan(cbloomrants.blogspot.com)
2 points by wallnuss 8y ago|0 comments
26.
by wallnuss 8y ago
Julia has `widemul` which takes two `Int64` and produces an `Int128`. It is also not to hard to actually add your own primitive type for `Int256` (with the caveat that it needs support from LLVM, which I haven't checked)
27.
by wallnuss 8y ago
Also the report (in German) [1] that the original release deadline was May 7th, but Intel has asked for another 14 days of leniency. [1] https://www.heise.de/security/meldung/Spectre-NG-Intel-versc...
28.
by wallnuss 8y ago
One of the things that Nvidia quite successfully did (and Intel to some extent) is a ridiculously good academic outreach program. Making it easy (and cheap) for students and researchers to use you hardware leads to long-term buy-in.
29.
by wallnuss 8y ago
IACA is awesome to understand the behaviour of your own hot loops. Sadly it only works on Intel. Luckily LLVM has recently merged llvm-mca [1] (machine-code analyser) which hopefully will in time bring all the features of IACA and more to o
30.
by wallnuss 9y ago
Working on it :) I will probably announce it on twitter (@vchuravy) once I managed to find time to finish it.

More ›