o Introduce new m_getcl() interface routine that allocates an mbuf

  and a cluster in one shot.
o Introduce MBP_PERSIST and MBP_PERSISTENT control bits to mb_alloc();
  MBP_PERSIST means "if you can allocate, then keep the cache lock
  held on exit," and MBP_PERSISTENT means "a cache lock is alredy held
  on entry, so allocate from the specified (already locked) cache."
  They may be used in combination.
o m_getcl() uses the MBP_PERSIST/MBP_PERSISTENT interface so that it
  doesn't drop the cache lock in between the mbuf and cluster allocations.
o m_getm(), which takes a size and allocates an mbuf + cluster "best fit"
  chain, has been moved from uipc_mbuf.c to subr_mbuf.c and shown how to
  use MBP_PERSIST/MBP_PERSISTENT to attempt to do a grouped allocation
  without dropping the cache lock in between.

Why this is good: much less bus-locked lock acquires/drops when they're
not needed.  Also, prototype for m_getcl():
struct mbuf * m_getcl(int how, short type, int flags);
"how" and "type" are self-explanatory.  "flags" may be M_PKTHDR, in
which case m_getcl() will make the mbuf a pkthdr-mbuf.

While I'm in subr_mbuf.c:
o Every exported routine now has a nice comment with a description of
  the expected arguments.  Eventually, mbuf(9) needs to be re-vamped
  but there's still more code to write/finalize before I get to that.
o internal macros have been changed a bit.
o consistently use 'short' for "type."  This somehow slipped through
  before (that 'type' was sometimes declared as int).

Alfred has been pushing for the MBP_PERSIST{,ENT} thing for almost a
year now.  Luigi asked for m_getcl(), and will probably MFC that
part of this commit.

TODO [Related]: teach mb_free() about MBP_PERSIST{, ENT}.

1 files changed, 0 insertions, 66 deletions

diff --git a/sys/kern/uipc_mbuf.c b/sys/kern/uipc_mbuf.c
index 27ca156..ea8f16d 100644
--- a/sys/kern/uipc_mbuf.c
+++ b/sys/kern/uipc_mbuf.c
@@ -62,72 +62,6 @@ SYSCTL_INT(_kern_ipc, KIPC_MAX_HDR, max_hdr, CTLFLAG_RW, &max_hdr, 0, "");
 SYSCTL_INT(_kern_ipc, KIPC_MAX_DATALEN, max_datalen, CTLFLAG_RW,
 	   &max_datalen, 0, "");
 
-/*
- * struct mbuf *
- * m_getm(m, len, how, type)
- *
- * This will allocate len-worth of mbufs and/or mbuf clusters (whatever fits
- * best) and return a pointer to the top of the allocated chain. If m is
- * non-null, then we assume that it is a single mbuf or an mbuf chain to
- * which we want len bytes worth of mbufs and/or clusters attached, and so
- * if we succeed in allocating it, we will just return a pointer to m.
- *
- * If we happen to fail at any point during the allocation, we will free
- * up everything we have already allocated and return NULL.
- *
- */
-struct mbuf *
-m_getm(struct mbuf *m, int len, int how, int type)
-{
-	struct mbuf *top, *tail, *mp, *mtail = NULL;
-
-	KASSERT(len >= 0, ("len is < 0 in m_getm"));
-
-	MGET(mp, how, type);
-	if (mp == NULL)
-		return (NULL);
-	else if (len > MINCLSIZE) {
-		MCLGET(mp, how);
-		if ((mp->m_flags & M_EXT) == 0) {
-			m_free(mp);
-			return (NULL);
-		}
-	}
-	mp->m_len = 0;
-	len -= M_TRAILINGSPACE(mp);
-
-	if (m != NULL)
-		for (mtail = m; mtail->m_next != NULL; mtail = mtail->m_next);
-	else
-		m = mp;
-
-	top = tail = mp;
-	while (len > 0) {
-		MGET(mp, how, type);
-		if (mp == NULL)
-			goto failed;
-
-		tail->m_next = mp;
-		tail = mp;
-		if (len > MINCLSIZE) {
-			MCLGET(mp, how);
-			if ((mp->m_flags & M_EXT) == 0)
-				goto failed;
-		}
-
-		mp->m_len = 0;
-		len -= M_TRAILINGSPACE(mp);
-	}
-
-	if (mtail != NULL)
-		mtail->m_next = top;
-	return (m);
-
-failed:
-	m_freem(top);
-	return (NULL);
-}
-
 void
 m_freem(struct mbuf *m)
 {