Re: [lwip-commits] [SCM] lwIP Contrib - Contributed code for Lightweight TCP/IP stack branch, master, updated. STABLE-2_0_0_RELEASE-191-g4c854d4

[Dirk Ziegelmeier]

goddammit :-))

​We'll get this right somewhere in space and time, I'm sure.​

On Tue, Jan 2, 2018 at 9:00 PM, address@hidden <address@hidden> wrote:

Joel Cunningham wrote:

Dirk,

I think there is an issue with the test implementation for win32:

#if LWIP_TCPIP_CORE_LOCKING
      LWIP_ASSERT("Function called without core lock", (current_thread_id
== lwip_tcpip_thread_id) || lwip_core_locked);
#else /* LWIP_TCPIP_CORE_LOCKING */
      LWIP_ASSERT("Function called from wrong thread", current_thread_id
== lwip_tcpip_thread_id);
#endif /* LWIP_TCPIP_CORE_LOCKING */

For the LWIP_TCPIP_CORE_LOCKING case I don't think the assertion should
check if the current thread is the TCPIP thread, but rather always check
lwip_core_locked.

There's another issue here. The code in place checks that *any* thread has locked the core. Any other thread (or even an ISR) happening to call a function at that time passes the check... ;-)

Cheers,
Simon