On Tue, 27 Jan 2015, Nathan Rosenblum wrote:

> If the underlying socket write returns EAGAIN while writing the the HTTP GET
> request, will libcurl invoke the socket callback to indicate its interest in
> writeable-socket notifications?

No. It only calls the callback once for what it wants, and it should not call
the callback again until it wants to wait for something different.

> Perhaps somebody with more familiarity can clarify or point out where I've
> gone wrong? For many concurrent transfers I am burning considerable CPU time
> servicing useless writeable-socket events and would love to avoid them.

It sounds like a "normal" bug to me, that it has asked to get told about a
situation that it shouldn't at that point in the transfer. Wouldn't you agree?

-- 
  / daniel.haxx.se
-------------------------------------------------------------------
List admin: http://cool.haxx.se/list/listinfo/curl-library
Etiquette:  http://curl.haxx.se/mail/etiquette.html