bug#46881: 28.0.50; pdumper dumping causes way too many syscalls

[Pip Cet]

On Fri, Mar 5, 2021 at 7:19 AM Eli Zaretskii <eliz@gnu.org> wrote:
> > From: Pip Cet <pipcet@gmail.com>
> > Date: Fri, 5 Mar 2021 02:30:13 +0000
> > Cc: Eli Zaretskii <eliz@gnu.org>, Paul Eggert <eggert@cs.ucla.edu>, 
> > 46881@debbugs.gnu.org
> >
> > > It looks fine, but wouldn't dumping to a FILE* (with internal buffering)
> > > do the same basic thing in a simpler way?
> >
> > I initially set out to do that, but decided against it. We don't just
> > write sequentially (when FILE I/O helps, a little), we also have the
> > seek-and-fixup phase, and it didn't seem any simpler at that point..
>
> I'm not sure I understand: what's wrong with fseek?

Nothing, assuming you're fine with the current performance. Many libcs
aren't going to be smart enough to avoid I/O when you fseek through a
"large" file and write a word here and there, and my suspicion is that
would include glibc.

Also, we're not currently using fseek-and-write anywhere in Emacs.

We're talking about a file which Emacs is going to have to keep in
memory anyway, when reading the dump. The only case in which there
might be a problem is if the build machine has significantly less
available memory than the machine we intend to run on, and I just
don't think that's going to happen.

Pip