Re: [PATCH gwproxy v8 2/2] gwproxy: refactor code base to add experimental raw DNS backend

[Ahmad Gani <reyuki@gnuweeb.org>]

On Sun, Sep 7, 2025 at 2:06 PM Ammar Faizi wrote:
> On Sun, Sep 07, 2025 at 01:39:05PM +0700, Ahmad Gani wrote:
> > I think 65536 is a pretty low memory footprint in modern systems; as Sir
> > Alviro noted, it's around 0.25 or 0.5 MB.
> >
> > Maybe it's more beneficial to allocate at once in the stack rather than
> > add instructions for growing/shrinking the allocated memory?
>
> Yes, it's small. But I prefer to save the space for it. The memory
> saving is very beneficial for a small Linux device like Raspberry Pi,
> 4G modems, or even your lovely home routers.

Unfortunately I can't install gwproxy on my lovely fiberhome HG6243C routers :(

> Also, when you have 65536 clients in a single thread, that doesn't
> necessarily mean your DNS mapping are fully used from index 0 to 65535.
> A DNS query is a short-lived object. You only need to keep it until the
> DNS query receives a response.
>
> Having 65536 outstanding DNS queries is very unlikely even with 100K
> concurrent connections. Especially given the cache feature, you will
> skip the DNS resolution if identical names are accessed very often. I
> believe allocating it lazily is a good way for this design.
>
> If the initial size, 16, feels too low for you. You may consider adding
> an option to customize the initial size via CLI args.

Fair enough, I will try to adopt this proposal as well. Thanks for the idea!

--
Ahmad Gani