net: fix up truesize of cloned skb in skb_prepare_for_shift()
authorMarco Elver <elver@google.com>
Mon, 1 Feb 2021 16:04:20 +0000 (17:04 +0100)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Sun, 7 Mar 2021 11:34:05 +0000 (12:34 +0100)
commit97ff09a7ed484fef2b1bbc103857444b7332fca8
tree863aec24e2a9dc3a2585bc1a1a3269197bcc954e
parente00420943aef3cd07311a968cf0a8cfad24fe8d9
net: fix up truesize of cloned skb in skb_prepare_for_shift()

commit 097b9146c0e26aabaa6ff3e5ea536a53f5254a79 upstream.

Avoid the assumption that ksize(kmalloc(S)) == ksize(kmalloc(S)): when
cloning an skb, save and restore truesize after pskb_expand_head(). This
can occur if the allocator decides to service an allocation of the same
size differently (e.g. use a different size class, or pass the
allocation on to KFENCE).

Because truesize is used for bookkeeping (such as sk_wmem_queued), a
modified truesize of a cloned skb may result in corrupt bookkeeping and
relevant warnings (such as in sk_stream_kill_queues()).

Link: https://lkml.kernel.org/r/X9JR/J6dMMOy1obu@elver.google.com
Reported-by: syzbot+7b99aafdcc2eedea6178@syzkaller.appspotmail.com
Suggested-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20210201160420.2826895-1-elver@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
net/core/skbuff.c