Netfilter-Devel Archive mirror
 help / color / mirror / Atom feed
From: Jeremy Sowden <jeremy@azazel.net>
To: Pablo Neira Ayuso <pablo@netfilter.org>
Cc: Netfilter Devel <netfilter-devel@vger.kernel.org>
Subject: Re: [PATCH nft v2 0/2] Support for variables in map expressions
Date: Fri, 5 Apr 2024 21:38:56 +0100	[thread overview]
Message-ID: <20240405203856.GB1083504@celephais.dreamlands> (raw)
In-Reply-To: <Zg6NUHYLHYbIgKtq@calendula>

[-- Attachment #1: Type: text/plain, Size: 2694 bytes --]

On 2024-04-04, at 13:21:52 +0200, Pablo Neira Ayuso wrote:
> On Wed, Apr 03, 2024 at 01:09:35PM +0100, Jeremy Sowden wrote:
> > The first patch replaces the current assertion failure for invalid
> > mapping expression in stateful-object statements with an error message.
> > This brings it in line with map statements.
> > 
> > It is possible to use a variable to initialize a map, which is then used
> > in a map statement, but if one tries to use the variable directly, nft
> > rejects it.  The second patch adds support for doing this.
> 
> Thanks. I can trigger crashes, e.g.
> 
> define quota_map = "1.2.3.4"
> 
> table ip x {
>         chain y {
>                 quota name ip saddr map $quota_map
>         }
> }
> 
> src/mnl.c:1759:2: runtime error: member access within misaligned address 0x000100000001 for type 'struct expr', which requires 8 byte alignment
> 0x000100000001: note: pointer points here
> <memory cannot be printed>
> src/netlink.c:121:10: runtime error: member access within misaligned address 0x000100000001 for type 'const struct expr', which requires 8 byte alignment
> 0x000100000001: note: pointer points here
> <memory cannot be printed>
> AddressSanitizer:DEADLYSIGNAL
> =================================================================
> ==150056==ERROR: AddressSanitizer: SEGV on unknown address 0x00009fff8009 (pc 0x7f58e67d8624 bp 0x7ffd57d17eb0 sp 0x7ffd57d17c40 T0)
> ==150056==The signal is caused by a READ memory access.
>     #0 0x7f58e67d8624 in alloc_nftnl_setelem src/netlink.c:121
>     #1 0x7f58e67c3d12 in mnl_nft_setelem_batch src/mnl.c:1760
>     #2 0x7f58e67c45d9 in mnl_nft_setelem_add src/mnl.c:1805
>     #3 0x7f58e687df1e in __do_add_elements src/rule.c:1425
>     #4 0x7f58e687e528 in do_add_set src/rule.c:1471
>     #5 0x7f58e687e7aa in do_command_add src/rule.c:1491
>     #6 0x7f58e688fdb3 in do_command src/rule.c:2599
>     #7 0x7f58e679d417 in nft_netlink src/libnftables.c:42
>     #8 0x7f58e67a514a in __nft_run_cmd_from_filename src/libnftables.c:729
>     #9 0x7f58e67a639c in nft_run_cmd_from_filename src/libnftables.c:807
>     #10 0x557c9d25b3b0 in main src/main.c:536
>     #11 0x7f58e5846249 in __libc_start_call_main ../sysdeps/nptl/libc_start_call_main.h:58
>     #12 0x7f58e5846304 in __libc_start_main_impl ../csu/libc-start.c:360
>     #13 0x557c9d258460 in _start (/usr/sbin/nft+0x9460)
> 
> AddressSanitizer can not provide additional info.
> SUMMARY: AddressSanitizer: SEGV src/netlink.c:121 in alloc_nftnl_setelem
> ==150056==ABORTING
> 
> I think this is lacking more validation.

Agreed.  Should have done more testing.  Apologies!  Will follow up.

J.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

      reply	other threads:[~2024-04-05 20:39 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-04-03 12:09 [PATCH nft v2 0/2] Support for variables in map expressions Jeremy Sowden
2024-04-03 12:09 ` [PATCH nft v2 1/2] evaluate: handle invalid mapping expressions in stateful object statements gracefully Jeremy Sowden
2024-04-03 12:09 ` [PATCH nft v2 2/2] evaluate: add support for variables in map expressions Jeremy Sowden
2024-04-04 11:21 ` [PATCH nft v2 0/2] Support " Pablo Neira Ayuso
2024-04-05 20:38   ` Jeremy Sowden [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20240405203856.GB1083504@celephais.dreamlands \
    --to=jeremy@azazel.net \
    --cc=netfilter-devel@vger.kernel.org \
    --cc=pablo@netfilter.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).