Cliff Click @ VMSS16: Bits of Advice For the VM Writer on HackerTube: recommended by 2 Hacker News commenters across 4 separate months, 2017–2021.
talk · advanced · evergreen
Cliff Click, author of HotSpot's JIT compiler, shares hard-won advice for people writing virtual machines at the VMSS16 workshop. Topics include interpreters, JITs and performance lessons.
HN recommends it, and Cliff Click's talks in general, as practical wisdom from someone who built HotSpot and worked on Azul's VMs.
“I strongly recommend following and listening to videos by Cliff Click who wrote HotSpot and worked on Azul and other VMs.”
6 Hacker News comments link this video. Each one links back to its thread.
I strongly recommend following and listening to videos by Cliff Click who wrote HotSpot and worked on Azul and other VMs. Cliff also has Coffee Compiler Club sessions on Fridays. - [link] - ▶ this video
in “Ask HN: Best Books on the JVM?” · View on HN
I highly recommend folks to view this talk by the venerable Cliff Click called "Bits of Advice For the VM Writer" [1] Of particular interest are Continuation Style Passing Interpreters [2] and how much of a speedup you can get by going to raw assembly. The talk has lessons Cliff learned from his years on the JVM but those can be adapted easily to other runtimes. [1] ▶ this video [2] [link]
in “Building Fast Interpreters in Rust” · View on HN
Very poor, there's no mention of continuation style passing interpreters. If anyone's curious about tales from the battlefield, highly recommend Cliff Click's talks (any of them) [1] [1] ▶ this video
in “The Structure and Performance of Efficient Interpreters [pdf]” · View on HN
I'm also a fan of Cliff and other Rice alumni; here are a few of his videos that I like. [0] The Sea of Nodes and the HotSpot JIT [1] A JVM Does That? [2] A Crash Course in Modern Hardware [3] Bits of Advice for the VM Writer [0] [link] [1] [link] [2] [link] [3] ▶ this video
in “Intel to disable TSX by default on more CPUs with new microcode” · View on HN
He doesn't give a lot of details, but somewhere in the first 20 minutes of this talk (probably 15 minutes in), Cliff Click says that Hotspot had to create fair locks in user-space because kernel provided locks weren't adequate. ▶ this video
in “On mutex performance and WTF::Lock” · View on HN
Thank you for the video, very interesting. The time for what you mentioned is here: ▶ this video (from 18:24 to 24:00)
in “On mutex performance and WTF::Lock” · View on HN