gnufx

50 ms·

601.
by gnufx 6y ago
Hoare said "I don’t know what the language of the year 2000 will look like, but I know it will be called Fortran". He also said "ALGOL 60 is alive and well and living in Fortran 90", which is a decent compliment from him
602.
by gnufx 6y ago
While Goto typically gets the plaudits (and obviously did good work there), you should credit van de Geijn in that area long term.
603.
by gnufx 6y ago
The storage association rules have nothing to do with pointers like that. Code will typically break because arguments are passed with copy-in/copy-out semantics, not by reference.
604.
by gnufx 6y ago
It does, indeed, but that's not the only way to associate storage. Using INTENT to specify whether an array is updated is useful.
605.
by gnufx 6y ago
Mostly the compiler won't catch violation of the storage association rules. Toolpack had a useful tool for strict F77 code which was a boon fixing stuff not to break with the Alliant compiler, which was the first one the project I inh
606.
by gnufx 6y ago
Profiling and -fopt-info should tell you why. (If you care about speed, use them anyway!) I'd be surprised if it's vectorization improvements, other than a specific bug fix.
607.
by gnufx 6y ago
It's perhaps worth saaying that 30 years ago code in the wild might well not have compiled and run correctly on a different system, and is more likely to today. There tended to be more non-standard features and ignoring of the standar
608.
by gnufx 6y ago
Fortran arrays are indexed as you choose. Vectors would be arrays. Array operations will lose if the compiler doesn't scalarization/deforestation properly (if deforestation is an appropriate word when dealing with arrays rather th
609.
by gnufx 6y ago
Thanks from those of us feeling relatively young for once, even if we have read McCracken. The best failure I know was due to beer fermenting in the warm space above the fancy disk of an old Sigma 3; with hilarious consequences. No opportun
610.
by gnufx 6y ago
Well, there was the shocking news here not long ago that some OS\360(?) backwards compatibility had been dropped. What's the world coming to. Relevant to Fortran, there's a version of ESSL that I recall on OS\370 now for POWER9 (
611.
by gnufx 6y ago
The main advantage usually quoted over C is the optimization opportunities from the storage association rules, in particular. GCC got the alias analysis for that as far back as the egcs version. (The flame war preceding egcs actually aros
612.
by gnufx 6y ago
And not only on such codes. I've posted variations on this before: In my experience GNU Fortran was always competitive with proprietary compilers at around the 20% level on average once the scheduling was sorted for a new architecture
613.
by gnufx 6y ago
Yes, I switched to it from K-9 for the work O365 we now have to use, sigh. (There is a large XOAUTH patch for K-9, but it hasn't been incorporated as far as I know.) Not having that n the F-Droid version is a pain, but it does phone h
614.
by gnufx 6y ago
Ho, ho. Graphene has left the lab, but it's only got as far as the hype mill, not the factory. https://www.manchester.ac.uk/discover/news/the-university-un... was a notable success though; and allegedly grap
615.
by gnufx 6y ago
The V in SVE is for vector, the S isn't for SIMD, and it's length-agnostic; I don't know how similar it is to the RISC-V vector extension. Think CDC, Cray, NEC, not AMD/Intel. I guess the recent innovation in that spac
616.
by gnufx 6y ago
Surely the barrier to using a new architecture is being able to boot a kernel and run (say) the GNU toolchain, as demonstrated with RISC-V. Then you just compile your code, assuming it doesn't contain assembler, or something. Whether
617.
by gnufx 6y ago
For something some of us would consider more like dancing(!), there's https://www.scienceceilidh.com/ . UK folk-type dancing has a large STEMy contingent: beware dances written by Cambridge mathematicians if you'r
618.
by gnufx 6y ago
It's widely used on the right side of the Atlantic too, at least in technical circles. What's unfortunate is that -- especially in the wider world -- now everyone just talks about "coding", the relatively trivial part o
619.
by gnufx 6y ago
I'm not sure what to make of the article but, since it includes BLAS, it might be worth pointing out that scientific subroutine libraries are one of the most successful examples of reuse, particularly compared with more trendy things a
620.
by gnufx 6y ago
So what's the problem, and what's the alternative anyway (in the absence of FIDO support in the case of ssh)?
621.
by gnufx 6y ago
> It is sad that with PGP being so horribly broken, it still casts its shadow over security products. How is ssh with a PGP key horribly broken? There's no shadow over Solo/Somu from where I'm standing.
622.
by gnufx 6y ago
Ah, I didn't know about the PGP code size. Point taken about the cost, and I hope you can sell to organizations. I'm surprised I can buy a competitor for £8, but then I can't run the proprietary programming software or, I a
623.
by gnufx 6y ago
I'd forgotten the name PrimeHPC-FX. You would think they'd be worth looking at if you knew about K. I don't remember exactly what Bristol have got now. (Its not so easy to get detailed info on UK Tier 2 systems, even if yo
624.
by gnufx 6y ago
I've spent a lot of my life around Toxteth and Moss Side, and that doesn't affect my threat model relevant to phishing and FIDO. Mugging is probably less of a threat than break-in anyway, as far as compromised hardware goes. You
625.
by gnufx 6y ago
I don't think the kickstarter is relevant now. You just buy them from solokeys.com. There's been Somu since that, and Solo2 is in the offing, with plenty of storage, apparently. I bought Somu partly for convenience, and partly
626.
by gnufx 6y ago
K was basically a one-off, but you could buy systems with similar technology, though they weren't common outside Japan. A64FX machines are appearing in the West now. There was a recent substantial award for one to a UK academic/
627.
by gnufx 6y ago
Fugaku, for which A64FX was designed, has no DDR memory, just HBM, but yes because of bandwidth. Its Tofu fabric has decent bandwidth but particularly low latency.
628.
by gnufx 6y ago
> The arrival of Intel (and AMD with the Opteron) on the TOP-500 supercomputer list demonstrated the importance of channel bandwidth (memory and I/O). Top500 is based on the HPL benchmark. Its problems are specifically not being se
629.
by gnufx 6y ago
There may be only a couple of CORAL-size systems (Summit et al) a year but there are many >~$1M HPC systems sold. Intel have so far failed to deliver for CORAL.
630.
by gnufx 6y ago
If they're better engineered than Sun's, it's impressive.

More ›