On Mon, 14 Mar 2016, Molina wrote:

> You are totally right. Using the default one (32KB) gives me the same
> performance, so it was obviously a mistake to increase its value. I updated
> the changes in my repository [1].

Great. It is also good to get that theory tested and confirmed so that we know
for sure it still holds.

> On a related matter, the server might indicate via SETTINGS that he’s not
> able to handle the default value CURL has at the moment, which is the same
> default nghttp2 has, i.e. 100 parallel streams. If that happens, should CURL
> decrease (or increase!) its own NGHTTP2_SETTINGS_MAX_CONCURRENT_STREAMS to
> the minimum of both client and server?

Oh yes it should! That's indeed an oversight! Can you take a look at that?

A question on your updated commit at:
https://github.com/curl/curl/compare/master...Moliholy:http2-window

You add a call nghttp2_submit_window_update() for the upgrade case. Is that
really necessary? The nghttp2_session_upgrade() call done just a couple of
lines before that is already sending the settings that includes the large
initial window size. Surely we then don't need to bump the window size
immediately?

-- 
  / daniel.haxx.se

-------------------------------------------------------------------
List admin: https://cool.haxx.se/list/listinfo/curl-library
Etiquette: https://curl.haxx.se/mail/etiquette.html
Received on 2016-03-17