Re: [lwip-users] Problem with loading a website

[Sergio R. Caprile]

You need to understand that a browser "loading a website" is not 
actually a sequential process.
A browser opens as many simultaneous connections as it can (in some 
cases you can limit this), so you will get many connection requests, 
many GETs, and you need to be able to satisfy that.
The problem (I think) is that you are probably using statics and 
thinking of one connection, one file at a time, which is probably OK, or 
not, depending on the seek time of your media; and your "entire file 
read" operation is "not reentrant".
Anyway, if "your filesystem" can't have more than one "file" open at the 
same time, or can't read from more than one "file" at a time, you need 
to be able to tell so to the using application and it must be able to 
cope with that.
See the word '_custom' there ? That means you will develop strategies so 
your filesystem works with the web server, like limiting the number of 
simultaneous connections, delaying opening files, delaying read 
requests, etc. You can write a state machine there if you need to, 
providing state variables depend on the caller (e.g.: not static but in 
a structure passed as a parameter, like the file handler itself).

That opens at least two possibilities:
fs_file_extension can handle whatever you need to keep your internal 
state, like for example "waiting for other read to finish"
enabling LWIP_HTTPD_FS_ASYNC_READ will give you the possibility to wait 
on a read. I didn't study this interface and can't give you further advice.