I’ve started noticing articles and YouTube videos touting the benefits of branchless programming, making it sound like this is a hot new technique (or maybe a hot old technique) that everyone should be using. But it seems like it’s only really applicable to data processing applications (as opposed to general programming) and there are very few times in my career where I’ve needed to use, much less optimize, data processing code. And when I do, I use someone else’s library.
How often does branchless programming actually matter in the day to day life of an average developer?
Because of the performance improvements from vectorization, and the fact that GPUs are particularly well suited to that? Or are GPUs particularly bad at branches.
How often do a few nanoseconds in the inner loop matter?
Looking at all the software out there, the vast majority of it is games, apps, and websites. Applications where performance is critical, such as control systems, operating systems, databases, numerical analysis, etc, are relatively rare compared to apps/etc. So statistically speaking the majority of developers must be working on the latter (which is what I mean by an “average developer”). In my experience working on apps there are exceedingly few times where micro-optimizations matter (as in things like assembly and/or branchless programming as opposed to macro-optimizations such as avoiding unnecessary looping/nesting/etc).
Edit: I can imagine it might matter a lot more for games, such as in shaders or physics calculations. I’ve never worked on a game so my knowledge of that kind of work is rather lacking.
Yes. GPUs don’t have per-core branching, they have dozens of cores running the same instructions. So if some cores should run the if branch and some run the else branch, all cores in the group will execute both branches, and mask out the one they shouldn’t have run. I also think they they don’t have the advanced branch prediction CPUs have.
https://en.wikipedia.org/wiki/Single_instruction,_multiple_threads
Fintech. Stock exchanges will go to extreme lengths to appease their wolves of Wallstreet.
Also if you branch on a GPU, the compiler has to reserve enough registers to walk through both branches (handwavey), which means lower occupancy.
Often you have no choice, or removing the branch leaves you with just as much code so it’s irrelevant. But sometimes it matters. If you know that a particular draw call will always use one side of the branch but not the other, a typical optimization is to compile a separate version of the shader that removes the unused branch and saves on registers
It doesn’t matter until you need it. And when you need it, it’s the difference between life and death