re: kern/58168: IFF_OACTIVE semantics and replacement plan is unclear

[matthew green <mrg%eterna23.net@localhost>]

The following reply was made to PR kern/58168; it has been noted by GNATS.

From: matthew green <mrg%eterna23.net@localhost>
To: kern-bug-people%netbsd.org@localhost, gnats-admin%netbsd.org@localhost,
    netbsd-bugs%netbsd.org@localhost, gnats-bugs%netbsd.org@localhost
Cc: 
Subject: re: kern/58168: IFF_OACTIVE semantics and replacement plan is unclear
Date: Thu, 18 Apr 2024 16:01:06 +1000

 > in the old method, IFF_OACTIVE is an driver-internal thing to say
 > "i'm sending a packet".
 >
 > how it's used is upto each driver.
 >
 > replacement is to put it into driver-internal data, if you need it,
 > but it's not always wanted or useful.  as the TODO.smpnet says, it
 > needs to stop being part of if_flags, but what you do instead is a
 > driver specific answer.  (perhaps updating TODO.smpnet to say that
 > the semantics are driver specific would suffice here?)
 
 taylor pointed out that net/if.c does pay attention to this
 flag.  it won't call if_start if it is set, assuming that
 the device is already running the queue.
 
 that check, if needed, should also be moved into the device
 if_start() if not already there.
 
 
 .mrg.