@hatookov profile picture
hatoo💛 @hatookov

Rustacean

Joined October 2017
  • Tweets 3,571
  • Following 729
  • Followers 528
284 Photos and videos
わざわざ#[inline(never)]している == 後のカスタムrustc_codegenで関数名でマッチしてなにかしたいと思って良さそう
1
124
denoはベンチマーク勝負してるイメージあんまりなかったから意外。もっと争え…！
1
166
oha使われとるやん！😉
Deno@deno_land
Jun 11
Deno 2.8.3 is out! Deno.serve() is faster: we shipped a new HTTP/1.1 fast path, plus fixes for streaming bodies, header handling, and compression negotiation. Also a big LSP cleanup, 170 fixes in total. `deno desktop` is coming soon!
4
15
3,830
hatoo💛 retweeted
Lisan al Gaib
@scaling01
Jun 11
Fable 5 refused 200 out of 200 ProgramBench tasks lmao
125
184
5,164
409,749
ワイ、github.com/NVlabs/cuda-oxide をみてNvidia株買い増しを決意

GitHub - NVlabs/cuda-oxide: cuda-oxide is an experimental Rust-to-CUDA compiler that lets you write...

cuda-oxide is an experimental Rust-to-CUDA compiler that lets you write (SIMT) GPU kernels in safe(ish), idiomatic Rust. It compiles standard Rust code directly to PTX — no DSLs, no foreign languag...

github.com
1
5
129
あまり公言できないけど俺はHTTP/2とかHTTP/3は難しすぎてもうやめて欲しいってうっすら思ってるよ
1
130
hatoo💛 retweeted
Calif
@calif_io
Jun 2
Introducing HTTP/2 Bomb: a remote DoS in nginx, Apache httpd, Microsoft IIS, Envoy, and Cloudflare Pingora. A single client pins 32GB of server memory in 10s. Found by Codex. Blog post: blog.calif.io/p/codex-discov… PoCs: github.com/califio/publicati…

Codex Discovered a Hidden HTTP/2 Bomb

14 years ago, I helped break HTTP header compression, then was asked to review the fix, which became part of HTTP/2. Life has come full circle: today we're releasing an attack I missed.

blog.calif.io
25
437
2,013
184,850
小さいコンピュータ好きなのでCardputerZeroバックします
106
たった今、@Kickstarter_JP で CardputerZero, Pocket Raspberry Pi Computer for Makers のバッカーになりました kickstarter.com/projects/m5s…

CardputerZero, Pocket Raspberry Pi Computer for Makers

Powered by CM0 | All-in-One Design | Standard Linux I/O | Grove & Advanced I/O Ports | 100 Extra M5 Modules | Credit Card Size

kickstarter.com
2
186
hatoo💛 retweeted
hatoo💛@hatookov
Apr 16
Fluorite今どんな感じですか……？ fluorite.game/
1
1
303
ProgramBenchのテストケースに俺が書いたプログラムがあったのでテストケース見てみた｜hatoo zenn.dev/hatoo/articles/e783… #zenn

ProgramBenchのテストケースに俺が書いたプログラムがあったのでテストケース見てみた

zenn.dev
3
1
746
Hugging Faceのdatasetってクローラーから隠すために意図的にtarballで配布してんの？見づらいんだが
191
ProgramBenchとかいうLLMベンチマークにohaが入ってるってマジ！？ > We evaluate 9 LMs and find that none fully resolve any task, with the best model passing 95% of tests on only 3% of tasks. LLM君もうちょっと頑張って❤️ arxiv.org/abs/2605.03546
1
3
267
なんかTUIのテストもある。草
114
GPT 5.5 (xhigh)は$5.02でohaのtest 78.6%passか…やるやんけ programbench.com/run/gpt-5-5…
193
うおおおお
TaKO8Ki@TaKOBKi
May 12
最近取り組んでいたRustのclosureの中で使えるmove expression `move(expr)`の初期実装がマージされた。
3
340
hatoo💛 retweeted
Jarred Sumner
@jarredsumner
May 5
Zig has been great for Bun. This branch is a fun experiment, and not a decision - here’s what I wrote on HN:
I work on Bun and this is my branch This whole thread is an overreaction. 302 comments about code that does not work. We haven’t committed to rewriting. There’s a very high chance all this code gets thrown out completely. I’m curious to see what a working version of this looks, what it feels like, how it performs and if/how hard it’d be to get it to pass Bun’s test suite and be maintainable. I’d like to be able to compare a viable Rust version and a Zig version side by side.

ALT I work on Bun and this is my branch This whole thread is an overreaction. 302 comments about code that does not work. We haven’t committed to rewriting. There’s a very high chance all this code gets thrown out completely. I’m curious to see what a working version of this looks, what it feels like, how it performs and if/how hard it’d be to get it to pass Bun’s test suite and be maintainable. I’d like to be able to compare a viable Rust version and a Zig version side by side.

Luke Parker
@LukeParkerDev
May 4
what!? is Bun moving from Zig to Rust??? github.com/oven-sh/bun/commi…
55
64
1,620
370,765
GPU最適化でAMDのイベントに招待されたことあります！俺を雇ってください！！ 世界最大55万台のGPUを持つxAI、実は6万台分しか使えていないことが判明 xenospectrum.com/xai-colossu…

世界最大55万台のGPUを持つxAI、実は6万台分しか使えていないことが判明 | XenoSpectrum

xAIは世界最大規模のAIクラスター「Colossus」を保有するが、その計算能力のわずか11%しか活用できておらず、新社長が2ヶ月以内に50%への改善を宣言した。これは、急速なハードウェア拡張に対しソフトウェア整備が追いつかず、MetaやGoogleに比べて実効的なGPU稼働率が著しく低いという構造的な課題を露呈している。

xenospectrum.com
3
217
hatoo💛 retweeted
Luke Parker
@LukeParkerDev
May 4
what!? is Bun moving from Zig to Rust??? github.com/oven-sh/bun/commi…

phase-a: draft batch big-1 (12 files) · oven-sh/bun@9f3917e

github.com
52
100
1,107
1,586,342
directory traversalって怖くね？？ 多分おれはLeanかなんかで証明されようが怖いと思う github.com/tower-rs/tower-ht…

tower-http/tower-http/src/services/fs/serve_dir/mod.rs at 91bdd3b5828260593f8bc3ff588a592bd966cebd...

HTTP specific Tower utilities. Contribute to tower-rs/tower-http development by creating an account on GitHub.

github.com
1
2
315
Load more