uma_zcreate() call from kern_mbuf.c - bug?

[Dima Dorfman]

Brian Somers <brian at Awfulhak.org> wrote:
> Of course m_get() et. al. seem to manage to get MSIZE aligned pointers back
> from uma_zalloc_arg(zone_mbuf...) anyway, but surely that's an implementation
> side effect and the align argument should be corrected.

This change looks right to me, but I'm hardly an expert here.

> There should probably also be a KASSERT here to ensure that MSIZE is sane

This might be better as a CTASSERT:

Index: kern_mbuf.c
===================================================================
RCS file: /ref/cvsf/src/sys/kern/kern_mbuf.c,v
retrieving revision 1.3
diff -u -r1.3 kern_mbuf.c
--- kern_mbuf.c	2 Aug 2004 00:18:35 -0000	1.3
+++ kern_mbuf.c	10 Sep 2004 06:53:10 -0000
@@ -123,6 +123,9 @@
 static void	mb_reclaim(void *);
 static void	mbuf_init(void *);
 
+/* Ensure that MSIZE doesn't break dtom(). */
+CTASSERT((((MSIZE - 1) ^ MSIZE) + 1) >> 1 == MSIZE);
+
 /*
  * Initialize FreeBSD Network buffer allocation.
  */
@@ -135,7 +138,7 @@
 	 * Configure UMA zones for Mbufs, Clusters, and Packets.
 	 */
 	zone_mbuf = uma_zcreate("Mbuf", MSIZE, mb_ctor_mbuf, mb_dtor_mbuf,
-	    NULL, NULL, UMA_ALIGN_PTR, UMA_ZONE_MAXBUCKET);
+	    NULL, NULL, MSIZE - 1, UMA_ZONE_MAXBUCKET);
 	zone_clust = uma_zcreate("MbufClust", MCLBYTES, mb_ctor_clust,
 	    mb_dtor_clust, NULL, NULL, UMA_ALIGN_PTR, UMA_ZONE_REFCNT);
 	if (nmbclusters > 0)