PERFORCE change 77201 for review

[Christian S.J. Peron]

http://perforce.freebsd.org/chv.cgi?CH=77201

Change 77201 by csjp at csjp_xor on 2005/05/20 00:17:30

	
	-replace a size_t for int in the checksum calculation routines
	-declare temp variable to stuff the size of the extended attribute in.
	 We have to do this because for some reason the data type representing
	 extended attribute size between vn_extattr_get and VOP_GETEXTATTR conflict.
	 This makes the mac_chkexec build happy on AMD64

Affected files ...

.. //depot/projects/trustedbsd/mac/sys/security/mac_chkexec/mac_chkexec.c#8 edit

Differences ...

==== //depot/projects/trustedbsd/mac/sys/security/mac_chkexec/mac_chkexec.c#8 (text+ko) ====

@@ -419,7 +419,7 @@
 	MD5_CTX	md5ctx;
 	u_quad_t b;
 	int error, count;
-	size_t resid;
+	int resid;
 	struct vattr va;
 	caddr_t bufobj;
 
@@ -465,7 +465,7 @@
 	SHA1_CTX sha1ctx;
 	u_quad_t b;
 	int error, count;
-	size_t resid;
+	int resid;
 	struct vattr va;
 	caddr_t bufobj;
 
@@ -510,15 +510,20 @@
 	char *depends, **ap, *paths[10];
 	int error, i, npaths;
 	struct nameidata nd;
-	size_t alen;
-	
+	int alen;
+	size_t ealen;
+
 	ASSERT_VOP_LOCKED(vp, "no vlock held");
 	error = VOP_GETEXTATTR(vp, MAC_CHKEXEC_ATTRN, MAC_CHKEXEC_DEP,
-	    NULL, &alen, NOCRED, curthread);
+	    NULL, &ealen, NOCRED, curthread);
 	if (error != 0 && error == ENOATTR)
 		return (0);
 	else if (error)
 		return (error);
+	/* XXX why are'nt extended attribute size specification types between
+	 * VOP_GETEXTATTR and vn_extattr_get the same?
+	 */
+	alen = ealen;
 	depends = malloc(alen + 1, M_CHKEXEC, M_WAITOK | M_ZERO);
 	error = vn_extattr_get(vp, IO_NODELOCKED, MAC_CHKEXEC_ATTRN,
 	    MAC_CHKEXEC_DEP, &alen, depends, curthread);
To Unsubscribe: send mail to majordomo at trustedbsd.org
with "unsubscribe trustedbsd-cvs" in the body of the message