tls: fix use-after-free in tls_push_record
authorDaniel Borkmann <daniel@iogearbox.net>
Fri, 15 Jun 2018 01:07:45 +0000 (03:07 +0200)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Tue, 26 Jun 2018 00:06:29 +0000 (08:06 +0800)
[ Upstream commit a447da7d00410278c90d3576782a43f8b675d7be ]

syzkaller managed to trigger a use-after-free in tls like the
following:

  BUG: KASAN: use-after-free in tls_push_record.constprop.15+0x6a2/0x810 [tls]
  Write of size 1 at addr ffff88037aa08000 by task a.out/2317

  CPU: 3 PID: 2317 Comm: a.out Not tainted 4.17.0+ #144
  Hardware name: LENOVO 20FBCTO1WW/20FBCTO1WW, BIOS N1FET47W (1.21 ) 11/28/2016
  Call Trace:
   dump_stack+0x71/0xab
   print_address_description+0x6a/0x280
   kasan_report+0x258/0x380
   ? tls_push_record.constprop.15+0x6a2/0x810 [tls]
   tls_push_record.constprop.15+0x6a2/0x810 [tls]
   tls_sw_push_pending_record+0x2e/0x40 [tls]
   tls_sk_proto_close+0x3fe/0x710 [tls]
   ? tcp_check_oom+0x4c0/0x4c0
   ? tls_write_space+0x260/0x260 [tls]
   ? kmem_cache_free+0x88/0x1f0
   inet_release+0xd6/0x1b0
   __sock_release+0xc0/0x240
   sock_close+0x11/0x20
   __fput+0x22d/0x660
   task_work_run+0x114/0x1a0
   do_exit+0x71a/0x2780
   ? mm_update_next_owner+0x650/0x650
   ? handle_mm_fault+0x2f5/0x5f0
   ? __do_page_fault+0x44f/0xa50
   ? mm_fault_error+0x2d0/0x2d0
   do_group_exit+0xde/0x300
   __x64_sys_exit_group+0x3a/0x50
   do_syscall_64+0x9a/0x300
   ? page_fault+0x8/0x30
   entry_SYSCALL_64_after_hwframe+0x44/0xa9

This happened through fault injection where aead_req allocation in
tls_do_encryption() eventually failed and we returned -ENOMEM from
the function. Turns out that the use-after-free is triggered from
tls_sw_sendmsg() in the second tls_push_record(). The error then
triggers a jump to waiting for memory in sk_stream_wait_memory()
resp. returning immediately in case of MSG_DONTWAIT. What follows is
the trim_both_sgl(sk, orig_size), which drops elements from the sg
list added via tls_sw_sendmsg(). Now the use-after-free gets triggered
when the socket is being closed, where tls_sk_proto_close() callback
is invoked. The tls_complete_pending_work() will figure that there's
a pending closed tls record to be flushed and thus calls into the
tls_push_pending_closed_record() from there. ctx->push_pending_record()
is called from the latter, which is the tls_sw_push_pending_record()
from sw path. This again calls into tls_push_record(). And here the
tls_fill_prepend() will panic since the buffer address has been freed
earlier via trim_both_sgl(). One way to fix it is to move the aead
request allocation out of tls_do_encryption() early into tls_push_record().
This means we don't prep the tls header and advance state to the
TLS_PENDING_CLOSED_RECORD before allocation which could potentially
fail happened. That fixes the issue on my side.

Fixes: 3c4d7559159b ("tls: kernel TLS support")
Reported-by: syzbot+5c74af81c547738e1684@syzkaller.appspotmail.com
Reported-by: syzbot+709f2810a6a05f11d4d3@syzkaller.appspotmail.com
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Dave Watson <davejwatson@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
net/tls/tls_sw.c

index 83f886d7c1f8b8b67b3d7097822875b609e8a42b..3c86614462f6d31c3c54caa7809fc82f92834e5f 100644 (file)
@@ -211,18 +211,12 @@ static void tls_free_both_sg(struct sock *sk)
 }
 
 static int tls_do_encryption(struct tls_context *tls_ctx,
-                            struct tls_sw_context *ctx, size_t data_len,
-                            gfp_t flags)
+                            struct tls_sw_context *ctx,
+                            struct aead_request *aead_req,
+                            size_t data_len)
 {
-       unsigned int req_size = sizeof(struct aead_request) +
-               crypto_aead_reqsize(ctx->aead_send);
-       struct aead_request *aead_req;
        int rc;
 
-       aead_req = kzalloc(req_size, flags);
-       if (!aead_req)
-               return -ENOMEM;
-
        ctx->sg_encrypted_data[0].offset += tls_ctx->prepend_size;
        ctx->sg_encrypted_data[0].length -= tls_ctx->prepend_size;
 
@@ -235,7 +229,6 @@ static int tls_do_encryption(struct tls_context *tls_ctx,
        ctx->sg_encrypted_data[0].offset -= tls_ctx->prepend_size;
        ctx->sg_encrypted_data[0].length += tls_ctx->prepend_size;
 
-       kfree(aead_req);
        return rc;
 }
 
@@ -244,8 +237,14 @@ static int tls_push_record(struct sock *sk, int flags,
 {
        struct tls_context *tls_ctx = tls_get_ctx(sk);
        struct tls_sw_context *ctx = tls_sw_ctx(tls_ctx);
+       struct aead_request *req;
        int rc;
 
+       req = kzalloc(sizeof(struct aead_request) +
+                     crypto_aead_reqsize(ctx->aead_send), sk->sk_allocation);
+       if (!req)
+               return -ENOMEM;
+
        sg_mark_end(ctx->sg_plaintext_data + ctx->sg_plaintext_num_elem - 1);
        sg_mark_end(ctx->sg_encrypted_data + ctx->sg_encrypted_num_elem - 1);
 
@@ -261,15 +260,14 @@ static int tls_push_record(struct sock *sk, int flags,
        tls_ctx->pending_open_record_frags = 0;
        set_bit(TLS_PENDING_CLOSED_RECORD, &tls_ctx->flags);
 
-       rc = tls_do_encryption(tls_ctx, ctx, ctx->sg_plaintext_size,
-                              sk->sk_allocation);
+       rc = tls_do_encryption(tls_ctx, ctx, req, ctx->sg_plaintext_size);
        if (rc < 0) {
                /* If we are called from write_space and
                 * we fail, we need to set this SOCK_NOSPACE
                 * to trigger another write_space in the future.
                 */
                set_bit(SOCK_NOSPACE, &sk->sk_socket->flags);
-               return rc;
+               goto out_req;
        }
 
        free_sg(sk, ctx->sg_plaintext_data, &ctx->sg_plaintext_num_elem,
@@ -284,6 +282,8 @@ static int tls_push_record(struct sock *sk, int flags,
                tls_err_abort(sk);
 
        tls_advance_record_sn(sk, tls_ctx);
+out_req:
+       kfree(req);
        return rc;
 }