ESP8266 Webserver Project

Moderator: Sprite_tm

User avatar
By Sprite_tm
#5277 Okay, a small warning:

DO NOT USE SDK 0.9.4 to compile this... yet.

Seems the behaviour of espconn_sent has changed in this version, breaking esphttpd. I'm not sure if this change is intentional or not, but I've reported it: http://bbs.espressif.com/viewtopic.php?f=13&t=92 Until that's resolved, esphttpd will break when compiled against 0.9.4.
User avatar
By pvvx
#5279
Sprite_tm wrote:Okay, a small warning:

DO NOT USE SDK 0.9.4 to compile this... yet.

Seems the behaviour of espconn_sent has changed in this version, breaking esphttpd. I'm not sure if this change is intentional or not, but I've reported it: http://bbs.espressif.com/viewtopic.php?f=13&t=92 Until that's resolved, esphttpd will break when compiled against 0.9.4.

espconn_sent no changed in version 0.9.4.
Errors in esp-httpd to espconn_sent with the first versions. Transmit buffer can not be free until the end of data transfer to the client (up to confirm tcp stack - RX [ACK])
Not https://en.wikipedia.org/wiki/Nagle%27s_algorithm :)
User avatar
By clinkme
#5287
Sprite_tm wrote:Okay, a small warning:

DO NOT USE SDK 0.9.4 to compile this... yet.

Seems the behaviour of espconn_sent has changed in this version, breaking esphttpd. I'm not sure if this change is intentional or not, but I've reported it: http://bbs.espressif.com/viewtopic.php?f=13&t=92 Until that's resolved, esphttpd will break when compiled against 0.9.4.


Yes, I confirm this bug.
If you sequentially calls espconn_sent(), for example, two times,
only the first call actually sends the data.
This is not depends are the data local (in the stack), global or so on.