[systemd-devel] [PATCH] Add sabridge for socket activation of traditional daemons

[David Strauss]

On Tue, Oct 15, 2013 at 4:32 PM, Zbigniew Jędrzejewski-Szmek
<zbyszek at in.waw.pl> wrote:
> I'm not sure if I understand the question correctly. First of all,
> the socket is in non-blocking mode, no? And anyway, even if it was in
> blocking mode, in the time between an incomplete send was done, the
> buffer could have been emptied by the kernel, so you cannot assume
> that the next send will block.

That's correct, and that's the response I got on Stack Exchange. I
figured it was the safe approach, anyway. The code now does that.

Specifically, it loops until it hits EAGAIN or EWOULDBLOCK, even if it
got a partial send(). This also works well under load tests.

-- 
David Strauss
   | david at davidstrauss.net
   | +1 512 577 5827 [mobile]