[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[lwip-devel] Another clarification about sys_arch_timeouts...

From: RAc
Subject: [lwip-devel] Another clarification about sys_arch_timeouts...
Date: Fri, 18 Jan 2013 05:24:17 -0800 (PST)

Is it correct to say that the sys_arch_timeouts mechanism is only needed if
my OS port must implement timeouts manually? For example, my FreeRTOS port
translates calls to sys_arch_sem_wait() to the native xSemaphoreTake() which
already takes a timeout as a parameter (the same applies to all wrapper
functions that implement timeouts), so I don't need to keep track of that
timeout myself. All implementations of ports I could find that do support
sys_arch_timeouts() appear to need it only when they implement timeouts
themselves, and since I couldn't find any reference to sys_arch_timeouts()
in the lwip kernel itself (timeouts relevant to the TXP stack appear to be
processed in the tcpip_thread), it seems as if the mechanism is optional?

I understand the theory about sys_arch_timeouts() implementing per thread
timeouts, but as far as I can tell, it's purely up to the port whether those
must be supported or not.


View this message in context: 
Sent from the lwip-devel mailing list archive at Nabble.com.

reply via email to

[Prev in Thread] Current Thread [Next in Thread]