Filter
Exclude
Time range
-
Minimum likes
Spotlight: Chris Lattner (@clattner_llvm) 🙌 Original author of LLVM, who also started Clang, Swift and MLIR, and co-founded Modular, maker of the Mojo language and MAX. So many builders use his work every day!
…dstatesofamericasuperintelligence.com/people/chris-l…
#LLVM #Mojo #AI
0
0
0
28
Replying to @Wesley__Lewis
0
0
0
2
#ptx asm in #llvm ir: redplait.blogspot.com/2026/09/ptx-as…
llvm doc lies about operand constraint codes
you can swap operand and this makes parsing/comparison of instructions non-trivial
nvidia's llvm backend cicc uses ~5-7% of instructions with undocumented attributes not avail in clang
0
0
2
501
Unstable stream update: 13th September 2026

Decent update to our unstable stream this week.

We landed LLVM 23.1.1, Cosmic 1.8.0, Plasma 6.7.5, Linux 7.2.5 and Nvidia Driver 615.71.09 amongst a load of update updates and additions.

aerynos.discourse.group/t/unstable-str…

#aerynOS #linux #LLVM
0
0
8
333
This week is the real finale! 🎉

The #SwiftMentorshipProgram 2026 wraps up tomorrow with the Demo Showcase and I get to present what I've been building with my mentor Augusto these last 10 weeks. 🚀

From my first bug-fix task to working on #LLVM itself and improving n the most intense learning sprint I've had in years. Tomorrow I finally get to show it off. 🧠

Part 1 - September 16, 17:00-18:00 CEST
Come cheer on this year's mentees - myself included! Full details and the Webex link are on the Swift Forums: 👇

forums.swift.org/t/swift-mentor…

#SwiftMentorshipProgram #Swift #DevDiary
0
0
2
185
We're bringing back unikernels! No HAL, no FSP, no #clang. Straight #llvm ELF for ARM M-33 from #cleflang #iot @RenesasGlobal
1
0
5
123
We are working in providing with cool material where people can learn about advance topics and with new techniques!
Learn how to write your own tooling using #LLVM, the theory is the one used by bigger and common reverse engineering tools.
#RevEng #LLVM #IR #BinaryAnalysis
Program Analysis has always been an arcane art. We shed light on it, you can build your own IR and lift to another language like a charm!
Tame the Dragon - MBA, Obfuscation and RE will get 100x speed!!!

PRE-SALE will end soon!
1
4
17
1,644
4 days to go for our early bird registrations to close!

Please register on compilertech.org if you haven’t already. If you are a student, we have travel reimbursements for you.

Register on : compilertech.org

#IICT #compilertech #llvm #mlir #iict2026 #compilertechnology
1
1
6
230
My mentor in the #SwiftMentorshipProgram is Augusto, who works on the Swift compiler and actively contributes to #LLVM. 🛠️

The program runs 10 weeks and wraps up on September 11

I'm really proud of what I've done. Here's a sneak peek of my goal. ✨

Curious?
1
0
5
772
2026 EuroLLVM Developers’ Meeting Talks
Videos: youtube.com/playlist?list=…
Slides: llvm.org/devmtg/2026-04/
@llvmorg #LLVM #MLIR
0
4
14
1,353
From our most recent #Stockholm #Cpp Meetup, a walkthrough of the #LLVM compiler pipeline, done by a LLVM release manager

youtu.be/g96voy6M6xo
0
0
2
131
Just got back from presenting at EuroLLVM 2026 in Dublin 🇮🇪

My work was on Runtime vendor-agnostic dispatch of ML kernel binaries via MLIR OffloadBinary.

#LLVM #MLIR #Compilers #GPU
0
1
4
151
Most compiler content is too theoretical.

I’m building CompilerSutra to fix that — practical deep dives on LLVM, MLIR, and performance ⚙️

If you're interested 👇
docs.google.com/forms/d/e/1FAI…

What topics do you want next?
#llvm #compiler #tutorial
0
0
1
34
Most people think performance = instruction count.
Not true.
Same instructions.
Same code.
GCC is ~25% faster than Clang here.
Reason?
AGU pressure + scheduling.

Compilers don’t just generate code
they decide how the CPU gets fed.

godbolt.org/z/E4havxWcW

#clang #gcc #llvm
1
0
3
98
LLVM 22.1 cross-platform compiler system is out with major backend upgrades, LLDB improvements, and new ThinLTO features.
linuxiac.com/llvm-22-1-rele…

#LLVM #OpenSource
0
2
20
801
Current status and coming soon:

Compiling flash-attention kernels with @bazelbuild and #llvm for all Nvidia Compute Capabilities, 1 action per sm_xx per kernel, fully parallel, hermetic and remote, in 5 minutes.

🫶 @buildbuddy for providing the cores <3
2
7
66
13,965
A massive new compiler toolchain release just dropped! #LLVM/#Clang 22.1 introduces critical infrastructure for Intel's next-gen Nova Lake CPUs, finally optimizes for AMD Zen 4, and fully deprecates Google Native Client. For Rerad more: 👉 tinyurl.com/5499jmvp
0
0
1
45