Void (glibc) is faster. The musl version is even faster. Compared to Void, Alpine feels like booting up my Pentium III in 2002. If you think Alpine is fast (lol) more power to ya.

>>103222886
>What makes it so fast,
It doesn't have any features

>>103222886
do you have some actual benchmark instead of teenage incel neet boot times?

>>103223102
nigga wut? the only differences between alpine and void are coreutils, init and package manager
if you're only talking about boot speed then maybe it's just openrc being shit, because my non-openrc alpine rig goes from LUKS prompt to GUI in literally under a second

smaller != faster
not when it comes to resident memory.
not when it comes to binary size (hint to all the retards who build with `-Os` for desktop targets).

comparing boot times is the most retarded metric ever. doubly so with different init systems involved.

musl with its built-in allocator (both the old one and the new one) is in fact SLOW and SHIT performance wise, although that doesn't manifest in all workloads. and the only (global) solution to this is integrating a good allocator directly into libc, but i'm not aware of anyone doing that other that chimera.

there are specific implementation details that may make musl slightly slower (or faster) than glibc in specific workloads. but the built-in allocator is the biggest and most impactful difference between the two.

>>103222886
Half of my boot time is from systemd loading stupid services I probably don't need but some I do and I don't feel like researching every one to figure out if I truly need it.

Actually the longest part is me entering the FDE password but I don't count that as actual boot time.

>>103223258
I dont understand how you make a memory allocator slower then glibc's, Even my own memory allocator is faster then it. And it's just a simple explicit free list allocator.

The desktop is horribly slow when I use Alpine on a Raspberry Pi. I could smoothly move windows around on Raspbian at 60Hz.