Recent submissions

Comment · 2 months ago
is this mainly about making host binaries self-contained for heterogenous workloads?

also, seems like this is mostly targeted towards HPC audience?

Comment · 3 months ago
Comment · 5 months ago
Comment · 5 months ago
> has been looping indefinitely

this can mean WebContent process is crashing

Comment · 6 months ago
the timer and urgency of this reminds me of the movie Armageddon where they had limited time to form a crew for a space mission.
Comment · 8 months ago
they have a watchdog loop, it runs periodically
Comment · 10 months ago
shouldn't the title be "CUDA Tile IR Open Sourced"?
Comment · 1 year ago
is there a document or reference implementation that describes the full algorithm? tiling, sorting, merging, and strip conversion.
Comment · 1 year ago
> In C++, it's an rvalue reference , which can be effectively thought of as an lvalue

hmm...this doesn't sound quite right? the comma operator's result in C++ is not an rvalue reference - it takes on exactly the value category of its right operand (which in this case is an lvalue)

Comment · 2 years ago
so, these are hand optimized primitives for specific model of nvidia gpus? do you still have to make launch/scheduling decisions to maximize occupancy? how does this approach scale to other target devices with specialized instruction sets and different architecture?
Comment · 2 years ago
can someone explain how is profiling tools like this written for GPU applications? wouldn't you need access to internal runtime api?

for ex. Apple wraps Metal buffers as "Debug" buffers to record allocations/deallocations.

Comment · 2 years ago
what kind of model architecture was used for this? is it safe to assume they used a transformer model or a variant of it?
Comment · 2 years ago
what's the purpose of this? is it one of those 'fun' problems to solve?
Comment · 2 years ago
Comment · 2 years ago
In a similar fashion, you'll see that JAX has frontend code being open-sourced, while device-related code is distributed as binaries. For example, if you're on Google's TPU, you'll see libtpu.so, and on macOS, you'll see pjrt_plugin_metal_1.x.dylib.

The main optimizations (scheduler, vectorizer, etc.) are hidden behind these shared libraries. If open-sourced, they might reveal hints about proprietary algorithms and provide clues to various hardware components, which could potentially be exploited.

Live from the official Hacker News API

A modern reader for Hacker News · SvelteKit + Skeleton