undefined | Better HN

0 pointsfphhotchips1y ago0 comments

> There's still plenty of hardware-specific APIs, you still debug assembly when something crashes, you still optimize databases for specific storage technologies and multimedia transcoders for specific CPU architectures...

You might, maybe, but an increasing proportion of developers:

- Don't have access to the assembly to debug it

- Don't even know what storage tech their database is sitting on

- Don't know or even control what CPU architecture their code is running on.

My job is debugging and performance profiling other people's code, but the vast majority of that is looking at query plans. If I'm really stumped, I'll look at the C++, but I've not yet once looked at assembly for it.

0 comments

daveguy1y ago

This makes sense to me. When I optimize, the most significant gains I find are algorithmic. Whether it's an extra call, a data structure that needs to be tweaked, or just utilizing a library that operates closer to silicon. I rarely need to go to assembly or even a lower level language to get acceptable performance. The only exception is occasionally getting into architecture specifics of a GPU. At this point, optimizing compilers are excellent and probably have more architecture details baked into them than I will ever know. Thank you, compiler programmers!

almostgotcaught1y ago

> At this point, optimizing compilers are excellent

the only people that say this are people who don't work on compilers. ask anyone that actually does and they'll tell you most compiler are pretty mediocre (tend to miss a lot of optimization opportunities), some compilers are horrendous, and a few are good in a small domain (matmul).

mandevil1y ago

It's more that the God of Moore's Law have given us so many transistors that we are essentially always I/O blocked, so it effectively doesn't matter how good our assembly is for all but the most specialized of applications. Good assembly, bad assembly, whatever, the point is that your thread is almost always going to be blocked waiting for I/O (disk, network, human input) rather than something that a fancy optimization of the loop that enables better branch prediction can fix.

almostgotcaught1y ago

> It's more that the God of Moore's Law have given us so many transistors that we are essentially always I/O blocked

this is again just more brash confidence without experience. you're wrong. this is a post about GPUs and so i'll tell you that as a GPU compiler engineer i spend my entire day (work day) staring/thinking about asm in order to affect register pressure and ilp and load/store efficiency etc.

> rather than something that a fancy optimization of the loop

a fancy loop optimization (pipelinig) can fix some problems (load/store efficiency) but create other problems (register pressure). the fundamental fact is NFL theorem applies here fully: you cannot optimize for all programs uniformly.

https://en.wikipedia.org/wiki/No_free_lunch_theorem

1 more reply

davemp1y ago

> At this point, optimizing compilers are excellent and probably have more architecture details baked into them than I will ever know.

While modern compilers are great, you’d be surprised about the seemingly obvious optimizations compilers can’t do because of language semantics or the code transformations would be infeasible to detect.

I type versions of functions into godbolt all the time and it’s very interesting to see what code is/isn’t equivalent after O3 passes

fpoling1y ago

The need to expose SSE instruction to system languages tells that compilers are not good at translating straightforward code into optimal machine code. And using SSE properly allows often to speed up the code by several times.

j / k navigate · click thread line to collapse

0 comments

daveguy1y ago

almostgotcaught1y ago

> At this point, optimizing compilers are excellent

mandevil1y ago

almostgotcaught1y ago

> It's more that the God of Moore's Law have given us so many transistors that we are essentially always I/O blocked

> rather than something that a fancy optimization of the loop

https://en.wikipedia.org/wiki/No_free_lunch_theorem

1 more reply

davemp1y ago

> At this point, optimizing compilers are excellent and probably have more architecture details baked into them than I will ever know.

I type versions of functions into godbolt all the time and it’s very interesting to see what code is/isn’t equivalent after O3 passes

fpoling1y ago

j / k navigate · click thread line to collapse