Re: `message' not outputting the newline "atomically"

[Paul Eggert]

Eli Zaretskii wrote:
> What if Emacs crashes half way through that loop

What, due to a hardware error or a buffer overrun or something like that? In 
that case all bets are off: Emacs could output nonsense or truncated output 
regardless of whether it uses one system call or a hundred.

Besides, if we were allowed to hypothesize nonexistent bugs to argue against a 
change to Emacs, then we could easily argue against any change whatsoever. When 
thinking about changes to Emacs, it's more productive to consider real problems 
than imaginary ones.

> the number of system calls is not something we should be
> bothered with

If by "number of system calls" you're referring to performance, then I agree 
that stderr output typically is not a significant performance problem (this 
particular output certainly isn't). However, the motivation for line-buffering 
stderr is correctness, not performance.

Line-buffered stderr does a better job of interleaving diagnostics from parallel 
instances of Emacs, and this is what prompted the fix in question. The 
poorly-interleaved diagnostics have been a practical problem that has bugged me 
for a while, and bugs Lars and I assume others. In contrast, unbuffered stderr's 
only advantages mentioned so far have been theoretical.