[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: State of the overlay tree branch?

From: Stefan Monnier
Subject: Re: State of the overlay tree branch?
Date: Fri, 23 Mar 2018 10:39:04 -0400
User-agent: Gnus/5.13 (Gnus v5.13) Emacs/27.0.50 (gnu/linux)

>> > I'd be interested to see a comparison with a code that ignores the
>> > markers entirely, and uses just these 4:
>> >   CONSIDER (BUF_PT (b), BUF_PT_BYTE (b));
>> >   CONSIDER (BUF_GPT (b), BUF_GPT_BYTE (b));
>> >   CONSIDER (BUF_ZV (b), BUF_ZV_BYTE (b));
>> I tried that, and in the synthetic benchmark which tries to reproduce
>> Sebastian's lsp-mode situation the result was indeed much better, but
>> then in other benchmarks it caused very significant slowdowns.
> In what benchmarks did it cause significant slowdowns?

Can't remember exactly, I think it was bad enough for a test case
which seemed pretty realistic so I discarded that option.
Basically what I remember is that I got the impression that it would
probably harm more users than the problem at hand.


PS: BTW, the number of markers is not the only issue: the order in which
they are created also matters.  Maybe we should try Sebastian's test
case but creating the markers/overlays in random order (and if that
helps, we could get a similar effect by making the GC randomly shuffle
the buffers's list of markers ;-).

reply via email to

[Prev in Thread] Current Thread [Next in Thread]