[CRYPTO] xcbc: Fix crash when ipsec uses xcbc-mac with big data chunk

The kernel crashes when ipsec passes a udp packet of about 14XX bytes
of data to aes-xcbc-mac.

It seems the first xxxx bytes of the data are in first sg entry,
and remaining xx bytes are in next sg entry. But we don't
check next sg entry to see if we need to go look the page up.

I noticed in hmac.c, we do a scatterwalk_sg_next(), to do this check
and possible lookup, thus xcbc.c needs to use this routine too.

A 15-hour run of an ipsec stress test sending streams of tcp and
udp packets of various sizes,  using this patch and
aes-xcbc-mac completed successfully, so hopefully this fixes the
problem.

Signed-off-by: Joy Latten <latten@austin.ibm.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

diff --git a/crypto/xcbc.c b/crypto/xcbc.c
index 2feb0f239c38179c579cbead01e7fe168d4ea0dd..b63b633e549ce7d42370c7747cdc1fbad1f8fcb8 100644 (file)
--- a/crypto/xcbc.c
+++ b/crypto/xcbc.c
@@ -116,13 +116,11 @@ static int crypto_xcbc_digest_update2(struct hash_desc *pdesc,
        struct crypto_xcbc_ctx *ctx = crypto_hash_ctx_aligned(parent);
        struct crypto_cipher *tfm = ctx->child;
        int bs = crypto_hash_blocksize(parent);
-       unsigned int i = 0;
 
-       do {
-
-               struct page *pg = sg_page(&sg[i]);
-               unsigned int offset = sg[i].offset;
-               unsigned int slen = sg[i].length;
+       for (;;) {
+               struct page *pg = sg_page(sg);
+               unsigned int offset = sg->offset;
+               unsigned int slen = sg->length;
 
                if (unlikely(slen > nbytes))
                        slen = nbytes;
@@ -182,8 +180,11 @@ static int crypto_xcbc_digest_update2(struct hash_desc *pdesc,
                        offset = 0;
                        pg++;
                }
-               i++;
-       } while (nbytes>0);
+
+               if (!nbytes)
+                       break;
+               sg = scatterwalk_sg_next(sg);
+       }
 
        return 0;
 }