Push kernel lock down to umb_rtrequest().

We are going to unlock PF_ROUTE sockets. This means `if_rtrequest'
handler will be performed without kernel lock.

umb_rtrequest() calls umb_send_inet_proposal() which touches kernel lock
protected `ipv{4,6}dns' array. Also umb_rtrequest() is the only handler
which requires kernel lock to be held. So push the lock down to
umb_rtrequest() instead of grab it around `if_rtrequest' call.

This hunk was commited separately for decreases PF_ROUTE sockets
unlocking diff.

ok gerhard@ deraadt@

diff --git a/sys/dev/usb/if_umb.c b/sys/dev/usb/if_umb.c
index 655a7f5..eba04af 100644 (file)
--- a/sys/dev/usb/if_umb.c
+++ b/sys/dev/usb/if_umb.c
@@ -1,4 +1,4 @@
-/*     $OpenBSD: if_umb.c,v 1.42 2021/03/30 20:58:19 sthen Exp $ */
+/*     $OpenBSD: if_umb.c,v 1.43 2021/04/01 08:39:52 mvs Exp $ */
 
 /*
  * Copyright (c) 2016 genua mbH
@@ -1053,10 +1053,12 @@ umb_rtrequest(struct ifnet *ifp, int req, struct rtentry *rt)
        struct umb_softc *sc = ifp->if_softc;
 
        if (req == RTM_PROPOSAL) {
+               KERNEL_LOCK();
                umb_send_inet_proposal(sc, AF_INET);
 #ifdef INET6
                umb_send_inet_proposal(sc, AF_INET6);
 #endif
+               KERNEL_UNLOCK();
                return;
        }