Module Name:    src
Committed By:   martin
Date:           Mon Aug 21 12:18:17 UTC 2023

Modified Files:
        src/sys/net/npf [netbsd-10]: npf_tableset.c

Log Message:
Pull up following revision(s) (requested by riastradh in ticket #332):

        sys/net/npf/npf_tableset.c: revision 1.41

npf(9): Drop table lock around copyout.

It is forbidden to hold a spin lock around copyout, and t_lock is a
spin lock.

We need t_lock in order to iterate over the list of entries.
However, during copyout itself, we only need to ensure that the
object we're copying out isn't freed by npf_table_remove or
npf_table_gc.

Fortunately, the only caller of npf_table_list, npf_table_remove, and
npf_table_gc is npfctl_table, and it serializes all of them by the
npf config lock.  So we can safely drop t_lock across copyout.

PR kern/57136
PR kern/57181


To generate a diff of this commit:
cvs rdiff -u -r1.38 -r1.38.4.1 src/sys/net/npf/npf_tableset.c

Please note that diffs are not public domain; they are subject to the
copyright notices on the relevant files.

Modified files:

Index: src/sys/net/npf/npf_tableset.c
diff -u src/sys/net/npf/npf_tableset.c:1.38 src/sys/net/npf/npf_tableset.c:1.38.4.1
--- src/sys/net/npf/npf_tableset.c:1.38	Sat Apr  9 23:38:33 2022
+++ src/sys/net/npf/npf_tableset.c	Mon Aug 21 12:18:17 2023
@@ -46,7 +46,7 @@
 
 #ifdef _KERNEL
 #include <sys/cdefs.h>
-__KERNEL_RCSID(0, "$NetBSD: npf_tableset.c,v 1.38 2022/04/09 23:38:33 riastradh Exp $");
+__KERNEL_RCSID(0, "$NetBSD: npf_tableset.c,v 1.38.4.1 2023/08/21 12:18:17 martin Exp $");
 
 #include <sys/param.h>
 #include <sys/types.h>
@@ -762,15 +762,17 @@ table_ent_copyout(const npf_addr_t *addr
 }
 
 static int
-table_generic_list(const npf_table_t *t, void *ubuf, size_t len)
+table_generic_list(npf_table_t *t, void *ubuf, size_t len)
 {
 	npf_tblent_t *ent;
 	size_t off = 0;
 	int error = 0;
 
 	LIST_FOREACH(ent, &t->t_list, te_listent) {
+		mutex_exit(&t->t_lock);
 		error = table_ent_copyout(&ent->te_addr,
 		    ent->te_alen, ent->te_preflen, ubuf, len, &off);
+		mutex_enter(&t->t_lock);
 		if (error)
 			break;
 	}

Reply via email to