The 'MSG_MORE' state of the previous sendmsg() is fetched without the
socket lock held, so two sendmsg calls can race. This can be seen with a
large sendfile() as that now does a series of sendmsg() calls, and if a
write() comes in on the same socket at an inopportune time, it can flip the
state.
Fix this by moving the fetch of ctx->more inside the socket lock.
Fixes: c662b043cdca ("crypto: af_alg/hash: Support MSG_SPLICE_PAGES")
Reported-by: syzbot+689ec3afb1ef07b766b2@xxxxxxxxxxxxxxxxxxxxxxxxx
Link: https://lore.kernel.org/r/000000000000554b8205ffdea64e@xxxxxxxxxx/
Signed-off-by: David Howells <dhowells@xxxxxxxxxx>
Tested-by: syzbot+689ec3afb1ef07b766b2@xxxxxxxxxxxxxxxxxxxxxxxxx
cc: Herbert Xu <herbert@xxxxxxxxxxxxxxxxxxx>
cc: Paolo Abeni <pabeni@xxxxxxxxxx>
cc: "David S. Miller" <davem@xxxxxxxxxxxxx>
cc: Eric Dumazet <edumazet@xxxxxxxxxx>
cc: Jakub Kicinski <kuba@xxxxxxxxxx>
cc: linux-crypto@xxxxxxxxxxxxxxx
cc: netdev@xxxxxxxxxxxxxxx
---
crypto/algif_hash.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/crypto/algif_hash.c b/crypto/algif_hash.c
index 0ab43e149f0e..82c44d4899b9 100644
--- a/crypto/algif_hash.c
+++ b/crypto/algif_hash.c
@@ -68,13 +68,15 @@ static int hash_sendmsg(struct socket *sock, struct msghdr *msg,
struct hash_ctx *ctx = ask->private;
ssize_t copied = 0;
size_t len, max_pages, npages;
- bool continuing = ctx->more, need_init = false;
+ bool continuing, need_init = false;
int err;
max_pages = min_t(size_t, ALG_MAX_PAGES,
DIV_ROUND_UP(sk->sk_sndbuf, PAGE_SIZE));
lock_sock(sk);
+ continuing = ctx->more;
+
if (!continuing) {
/* Discard a previous request that wasn't marked MSG_MORE. */
hash_free_result(sk, ctx);
