Thread (17 messages) 17 messages, 5 authors, 2014-09-13

Re: [PATCH net-next v4 2/9] net: filter: keep original BPF program around

From: Alexei Starovoitov <hidden>
Date: 2014-09-12 03:51:19

On Thu, Sep 11, 2014 at 8:27 PM, Eric Dumazet [off-list ref] wrote:
From: Eric Dumazet <edumazet@google.com>

On Fri, 2014-03-28 at 18:58 +0100, Daniel Borkmann wrote:
quoted
In order to open up the possibility to internally transform a BPF program
into an alternative and possibly non-trivial reversible representation, we
need to keep the original BPF program around, so that it can be passed back
to user space w/o the need of a complex decoder.

The reason for that use case resides in commit a8fc92778080 ("sk-filter:
Add ability to get socket filter program (v2)"), that is, the ability
to retrieve the currently attached BPF filter from a given socket used
mainly by the checkpoint-restore project, for example.

Therefore, we add two helpers sk_{store,release}_orig_filter for taking
care of that. In the sk_unattached_filter_create() case, there's no such
possibility/requirement to retrieve a loaded BPF program. Therefore, we
can spare us the work in that case.

This approach will simplify and slightly speed up both, sk_get_filter()
and sock_diag_put_filterinfo() handlers as we won't need to successively
decode filters anymore through sk_decode_filter(). As we still need
sk_decode_filter() later on, we're keeping it around.

Joint work with Alexei Starovoitov.

Signed-off-by: Alexei Starovoitov <redacted>
Signed-off-by: Daniel Borkmann <redacted>
Cc: Pavel Emelyanov <redacted>
---
Note that this patch added a possible use after free.

Following patch should be sent to stable trees only, as 3.17+
incidentally fixed this with
commit 278571baca2aecf5fb5cb5c8b002dbfa0a6c524c
    net: filter: simplify socket charging

[PATCH] net: filter: fix possible use after free

If kmemdup() fails, we free fp->orig_prog and return -ENOMEM

sk_attach_filter()
 -> sk_filter_uncharge(sk, fp)
  -> sk_filter_release(fp)
   -> call_rcu(&fp->rcu, sk_filter_release_rcu)
    -> sk_filter_release_rcu()
     -> sk_release_orig_filter()
        fprog = fp->orig_prog; // not NULL, but points to freed memory
          kfree(fprog->filter); // use after free, potential corruption
          kfree(fprog); // double free or corruption

Note: This was fixed in 3.17+ with commit 278571baca2a
("net: filter: simplify socket charging")

Found by AddressSanitizer

Signed-off-by: Eric Dumazet <edumazet@google.com>
Fixes: a3ea269b8bcdb ("net: filter: keep original BPF program around")
Wow. that's a tricky one. Thanks!
Acked-by: Alexei Starovoitov <redacted>

Indeed 3.17 is ok, but the fix is small enough, so wouldn't
hurt to apply it everywhere. Just for consistency?
quoted hunk ↗ jump to hunk
---
diff --git a/net/core/filter.c b/net/core/filter.c
index 1dbf6462f766..3139f966a178 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -1318,6 +1318,7 @@ static int sk_store_orig_filter(struct sk_filter *fp,
        fkprog->filter = kmemdup(fp->insns, fsize, GFP_KERNEL);
        if (!fkprog->filter) {
                kfree(fp->orig_prog);
+               fp->orig_prog = NULL;
                return -ENOMEM;
 }

Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help