Re: Re: [musl] The heap memory performance (malloc/free/realloc) is significantly degraded in musl 1.2 (compared to 1.1)

[Rich Felker <dalias@libc.org>]

On Tue, Sep 20, 2022 at 01:32:31AM +0800, baiyang wrote:
> Hi Rich,
> 
> Thanks for your reply.
> 
> > Unless you have an application that's explicitly using
> > malloc_usable_size all over the place, it's highly unlikely that this
> > is the cause of your real-world performance problems. 
> 
> 1. Yes, we have a real scenario where `malloc_usable_size` is called
> frequently: we need to optimize the realloc experience. We add an
> extra parameter to realloc - minimalCopyBytes: it represents the
> actual size of data that needs to be copied after fallback to
> malloc-copy-free mode. We will judge whether to call realloc or
> complete malloc-memcpy-free by ourself based on factors such as the
> size of the data that realloc needs to copy (obtained through
> `malloc_usable_size`), the size that we actually need to copy when
> we doing malloc-memcpy-free ourself (minimalCopyBytes) and the
> chance of merging chunks (small blocks) or mremap (large blocks) in
> the underlayer realloc. So, this is a real scenario, we need to call
> `malloc_usable_size` frequently.

Is there a reason you're relying on an unreliable and nonstandard
function (malloc_usable_size) to do this rather than your program
keeping track of its own knowledge of the allocated size? This is what
the C language expects you to do. For example if you have a structure
that contains a pointer to a dynamically sized buffer, normally you
store the size in a size_t member right next to that pointer, allowing
you to make these kind of decisions without having to probe anything.

> 2. As I mentioned before, this isn't just a problem with
> `malloc_usable_size`, since we actually include a full
> `malloc_usable_size` procedure in both `realloc` and `free`, it
> actually slows down The speed of other calls such as `free` and
> `realloc`. So this problem actually slows down not only the
> `malloc_usable_size` call itself, but also the realloc and free
> calls.

If this is affecting you too, that's a separate issue. But I can't
tell from what you've reported so far whether you're just claiming
this on a theoretical basis or whether you're actually experiencing
unacceptable performance.