Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 4 Dec 1998 19:49:50 +1100
From:      Bruce Evans <bde@zeta.org.au>
To:        ache@nagual.pp.ru, dillon@apollo.backplane.com
Cc:        cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG
Subject:   Re: cvs commit: src/sys/netinet icmp_var.h
Message-ID:  <199812040849.TAA14855@godzilla.zeta.org.au>

next in thread | raw e-mail | index | archive | help
>On Thu, Dec 03, 1998 at 08:27:55PM -0800, Matthew Dillon wrote:
>>     Due to the extern badport_bandlim() I have to keep the #include in
>>     icmp_var.h ... but it isn't that bad, only three kernel files depend
>>     on icmp_var.h.  I agree that doing the #include'ing in the two C files
>>     is cleaner, but what to do about the extern then ?
>
>Even in this variant you can safely remove opt_....h from icmp_var.h and
>include it only into really affected *.c files instead (before
>icmp_var.h). For kernel files which not needs this stuff but include
>icmp_var.h even badport_...() prototype will be not picked which is good. 

Another rule:
- options headers shall not affect the contents of other headers.

The extern prototype probably needs to be unconditional.  There's no
better place for it than icmp_var.h.  It's too bad that icmp_var.h
isn't actually a file for kernel variables (it mainly gives the API
for icmp statistics).

Bruce

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe cvs-all" in the body of the message



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199812040849.TAA14855>