CubeMX4.25 + STM32H743 + LwIP2.03
The first thing is to find what's limiting the speed ...
Profiling the code, and instrumenting the IRQ handler might provide insight.
I'd presume there is a lot of serialized mentality in the examples. Look for examples using threaded execution and that can handle a lot of inflight packets efficiently.
You might also try other stacks.
Note that 192.168.xxx.xxx addresses are LOCAL, millions of people use these same address ranges in there local networks and on routers, they aren't secret or accessible from the greater internet.
Clive One wrote: You might also try other stacks.
Clive One wrote:
Remember that the "Lw" in LwIP stands for Lightweight. So the key focus of this stack is on code economy - not necessarily on performance...
Also remember that LwIP is an independent 3rd-party tool - nothing specifically to do with ST.
For LwIP info & support, see: lwIP - A Lightweight TCP/IP stack - Summary [Savannah]
It is widely used - so some general googling may reveal potential optimisations not related to ST ...
The first thing you should tune in lwipopts.h is the conservative MTU size and the appropiate parameters.
I modified the CubeMX xml file to have the right values for every new project.
Retrieving data ...