LISP
This started out as a mere stub while I put together some more resources, and over the years became a quite large list of LISP implementations I have kept an eye on.(...)
#compilers #functional #implementations #interpreters #language #lisp #programming #resources #scheme
https://taoofmac.com/space/dev/lisp?utm_content=atom&utm_source=mastodon&utm_medium=social
LISP
This started out as a mere stub while I put together some more resources, and over the years became a quite large list of LISP implementations I have kept an eye on.(...)
#compilers #functional #implementations #interpreters #language #lisp #programming #resources #scheme
https://taoofmac.com/space/dev/lisp?utm_content=atom&utm_source=mastodon&utm_medium=social
Comparison of C/POSIX standard library implementations for Linux
https://www.etalabs.net/compare_libcs.html
#HackerNews #C #POSIX #Linux #Comparison #Libc #Implementations #Programming
Karatsuba Matrix Multiplication and Its Efficient Hardware Implementations
https://arxiv.org/abs/2501.08889
#HackerNews #Karatsuba #Matrix #Multiplication #Efficient #Hardware #Implementations #Algorithm #Research
While the Karatsuba algorithm reduces the complexity of large integer multiplication, the extra additions required minimize its benefits for smaller integers of more commonly-used bitwidths. In this work, we propose the extension of the scalar Karatsuba multiplication algorithm to matrix multiplication, showing how this maintains the reduction in multiplication complexity of the original Karatsuba algorithm while reducing the complexity of the extra additions. Furthermore, we propose new matrix multiplication hardware architectures for efficiently exploiting this extension of the Karatsuba algorithm in custom hardware. We show that the proposed algorithm and hardware architectures can provide real area or execution time improvements for integer matrix multiplication compared to scalar Karatsuba or conventional matrix multiplication algorithms, while also supporting implementation through proven systolic array and conventional multiplier architectures at the core. We provide a complexity analysis of the algorithm and architectures and evaluate the proposed designs both in isolation and in an end-to-end deep learning accelerator system compared to baseline designs and prior state-of-the-art works implemented on the same type of compute platform, demonstrating their ability to increase the performance-per-area of matrix multiplication hardware.
You're talking about how the way memory/object tracking and reference counting is done in the Python core, which in one implementation of Python is itself written in C, correct?
Responses:
1) Not all Python #implementations are in C. Some are inherently memory-safe.
2) In #CPython, the object lifetimes and #reference counting rules are well-understood and #documented, are maintained by a large core of developers who understand them, and have been tested with trial-by-fire by millions of users and millions of Python programs over decades. If you write an application in C or another memory-unsafe language and do your #memory handling yourself, it's code by one person, tested by at most a few people, and newly written - maybe not documented, and probably not with all the #bugs already shaken out of it.
Analogy: you need a boiler to operate your steam engine. You can use one designed by an engineer who has been doing boilers for twenty years, and manufactured by someone who has done nothing but that for the same length of time -- or you can cobble one together by yourself for the first time, not knowing the engineering behind it, and just hope it doesn't explode the first time you bring it up to pressure.
'QDax: A Library for Quality-Diversity and Population-based Algorithms with Hardware Acceleration', by Felix Chalumeau et al.
http://jmlr.org/papers/v25/23-1027.html
#qdax #qd #implementations
How do we call a #test that can test two or more #implementations of an #interface with the same test code?
They call it "Testing interface contracts" here: https://www.baeldung.com/java-junit-verify-interface-contract
IIRC, I heared a different term in the past, but I cannot remember it. #followerPower
I haven't. I sent a message to the #qmail list, replying to something or other, some time back, in which I described the lasting legacy of qmail as its #design, not its #codebase, and suggested perhaps a group project to maybe modernize a couple of design things - small! - and then start with fresh #implementations of the components to add #features or whatever - and noted they don't all have to be in the same implementation language etc.
[...]
Tao of Mac
CosicBe
Hacker News
C.
JMLR
π§π Greg Cocks
dtanzer