[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: "Significant Garbage Collection Improvement For Emacs" - sweep_conse
From: |
Po Lu |
Subject: |
Re: "Significant Garbage Collection Improvement For Emacs" - sweep_conses performance improved by 50%? |
Date: |
Sun, 30 Oct 2022 08:53:12 +0800 |
User-agent: |
Gnus/5.13 (Gnus v5.13) |
Stefan Monnier <monnier@iro.umontreal.ca> writes:
> Basically you're replacing 1kB blocks with 32kB blocks, which indeed
> seems very reasonable. Back when I introduced those blocks (and their
> associated bitmaps) I chose 1kB because I wanted something small enough
> that it couldn't be worse than what we had before, not because it was
> the best choice.
It may be worth nothing that I chose a cons and float block size of 4
times the page size while working with the incremental GC, which ends up
16k on most systems. However, there are also two pairs of markbits in
each block.
I haven't seen anything bad happen from that choice yet. But there,
BLOCK_ALIGN is gone in the first place, and it relies on valloc
returning suitably aligned memory.