Re: Broken logic in rtrequest1(RTM_DELETE, ..., ret_nrt)

[Ruslan Ermilov]

On Sun, Dec 22, 2002 at 09:33:29PM +0200, Ruslan Ermilov wrote:
> Hi!
> 
> The attached patch fixes the logic of rtrequest1()
> when handling RTM_DELETE requests and the caller
> asked for a copy of the removed entry.
> 
> Reviewers are highly welcome.
> 
> This makes the code more natural, similar to the
> RTM_ADD and RTM_RESOLVE cases (wrt to handling
> the returned rtentry's refcnt), and reduces some
> code bloat in applications.
> 
> It also makes it safe to turn the RTF_UP bit before
> rt_fixdelete() walk -- that was the start of my
> looking at this part of code.
> 
Oops, missed one.


Cheers,
-- 
Ruslan Ermilov          Sysadmin and DBA,
[EMAIL PROTECTED]           Sunbay Software AG,
[EMAIL PROTECTED]          FreeBSD committer,
+380.652.512.251        Simferopol, Ukraine

http://www.FreeBSD.org  The Power To Serve
http://www.oracle.com   Enabling The Information Age

Index: route.c
===================================================================
RCS file: /home/ncvs/src/sys/net/route.c,v
retrieving revision 1.72
diff -u -p -r1.72 route.c
--- route.c     18 Dec 2002 11:45:27 -0000      1.72
+++ route.c     22 Dec 2002 19:39:54 -0000
@@ -1117,10 +1112,7 @@ rtinit(ifa, cmd, flags)
                         * If we are deleting, and we found an entry, then
                         * it's been removed from the tree.. now throw it away.
                         */
-                       if (rt->rt_refcnt <= 0) {
-                               rt->rt_refcnt++; /* make a 1->0 transition */
-                               rtfree(rt);
-                       }
+                       RTFREE(rt);
                } else if (cmd == RTM_ADD) {
                        /*
                         * We just wanted to add it.. we don't actually

msg07914/pgp00000.pgp
Description: PGP signature