From: "Gustavo A. R. Silva" <gustavo@embeddedor.com>
To: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Kees Cook <keescook@chromium.org>,
linux-hardening@vger.kernel.org,
"Gustavo A. R. Silva" <gustavoars@kernel.org>,
Keith Packard <keithpac@amazon.com>,
Andrew Morton <akpm@linux-foundation.org>,
linux-kernel@vger.kernel.org, linux-wireless@vger.kernel.org,
netdev@vger.kernel.org, dri-devel@lists.freedesktop.org,
linux-staging@lists.linux.dev, linux-block@vger.kernel.org,
linux-kbuild@vger.kernel.org, clang-built-linux@googlegroups.com
Subject: Re: [PATCH 19/64] ip: Use struct_group() for memcpy() regions
Date: Wed, 28 Jul 2021 01:31:16 -0500 [thread overview]
Message-ID: <e3193698-86d5-d529-e095-e09b4d52927b@embeddedor.com> (raw)
In-Reply-To: <YQD2/CA7zJU7MW6M@kroah.com>
On 7/28/21 01:19, Greg Kroah-Hartman wrote:
> On Wed, Jul 28, 2021 at 01:14:33AM -0500, Gustavo A. R. Silva wrote:
>>
>>
>> On 7/28/21 00:55, Greg Kroah-Hartman wrote:
>>> On Tue, Jul 27, 2021 at 01:58:10PM -0700, Kees Cook wrote:
>>>> In preparation for FORTIFY_SOURCE performing compile-time and run-time
>>>> field bounds checking for memcpy(), memmove(), and memset(), avoid
>>>> intentionally writing across neighboring fields.
>>>>
>>>> Use struct_group() in struct flowi4, struct ipv4hdr, and struct ipv6hdr
>>>> around members saddr and daddr, so they can be referenced together. This
>>>> will allow memcpy() and sizeof() to more easily reason about sizes,
>>>> improve readability, and avoid future warnings about writing beyond the
>>>> end of saddr.
>>>>
>>>> "pahole" shows no size nor member offset changes to struct flowi4.
>>>> "objdump -d" shows no meaningful object code changes (i.e. only source
>>>> line number induced differences.)
>>>>
>>>> Note that since this is a UAPI header, struct_group() has been open
>>>> coded.
>>>>
>>>> Signed-off-by: Kees Cook <keescook@chromium.org>
>>>> ---
>>>> include/net/flow.h | 6 ++++--
>>>> include/uapi/linux/if_ether.h | 12 ++++++++++--
>>>> include/uapi/linux/ip.h | 12 ++++++++++--
>>>> include/uapi/linux/ipv6.h | 12 ++++++++++--
>>>> net/core/flow_dissector.c | 10 ++++++----
>>>> net/ipv4/ip_output.c | 6 ++----
>>>> 6 files changed, 42 insertions(+), 16 deletions(-)
>>>>
>>>> diff --git a/include/net/flow.h b/include/net/flow.h
>>>> index 6f5e70240071..f1a3b6c8eae2 100644
>>>> --- a/include/net/flow.h
>>>> +++ b/include/net/flow.h
>>>> @@ -81,8 +81,10 @@ struct flowi4 {
>>>> #define flowi4_multipath_hash __fl_common.flowic_multipath_hash
>>>>
>>>> /* (saddr,daddr) must be grouped, same order as in IP header */
>>>> - __be32 saddr;
>>>> - __be32 daddr;
>>>> + struct_group(addrs,
>>>> + __be32 saddr;
>>>> + __be32 daddr;
>>>> + );
>>>>
>>>> union flowi_uli uli;
>>>> #define fl4_sport uli.ports.sport
>>>> diff --git a/include/uapi/linux/if_ether.h b/include/uapi/linux/if_ether.h
>>>> index a0b637911d3c..8f5667b2ea92 100644
>>>> --- a/include/uapi/linux/if_ether.h
>>>> +++ b/include/uapi/linux/if_ether.h
>>>> @@ -163,8 +163,16 @@
>>>>
>>>> #if __UAPI_DEF_ETHHDR
>>>> struct ethhdr {
>>>> - unsigned char h_dest[ETH_ALEN]; /* destination eth addr */
>>>> - unsigned char h_source[ETH_ALEN]; /* source ether addr */
>>>> + union {
>>>> + struct {
>>>> + unsigned char h_dest[ETH_ALEN]; /* destination eth addr */
>>>> + unsigned char h_source[ETH_ALEN]; /* source ether addr */
>>>> + };
>>>> + struct {
>>>> + unsigned char h_dest[ETH_ALEN]; /* destination eth addr */
>>>> + unsigned char h_source[ETH_ALEN]; /* source ether addr */
>>>> + } addrs;
>>>
>>> A union of the same fields in the same structure in the same way?
>>>
>>> Ah, because struct_group() can not be used here? Still feels odd to see
>>> in a userspace-visible header.
>>>
>>>> + };
>>>> __be16 h_proto; /* packet type ID field */
>>>> } __attribute__((packed));
>>>> #endif
>>>> diff --git a/include/uapi/linux/ip.h b/include/uapi/linux/ip.h
>>>> index e42d13b55cf3..33647a37e56b 100644
>>>> --- a/include/uapi/linux/ip.h
>>>> +++ b/include/uapi/linux/ip.h
>>>> @@ -100,8 +100,16 @@ struct iphdr {
>>>> __u8 ttl;
>>>> __u8 protocol;
>>>> __sum16 check;
>>>> - __be32 saddr;
>>>> - __be32 daddr;
>>>> + union {
>>>> + struct {
>>>> + __be32 saddr;
>>>> + __be32 daddr;
>>>> + } addrs;
>>>> + struct {
>>>> + __be32 saddr;
>>>> + __be32 daddr;
>>>> + };
>>>
>>> Same here (except you named the first struct addrs, not the second,
>>> unlike above).
>>>
>>>
>>>> + };
>>>> /*The options start here. */
>>>> };
>>>>
>>>> diff --git a/include/uapi/linux/ipv6.h b/include/uapi/linux/ipv6.h
>>>> index b243a53fa985..1c26d32e733b 100644
>>>> --- a/include/uapi/linux/ipv6.h
>>>> +++ b/include/uapi/linux/ipv6.h
>>>> @@ -130,8 +130,16 @@ struct ipv6hdr {
>>>> __u8 nexthdr;
>>>> __u8 hop_limit;
>>>>
>>>> - struct in6_addr saddr;
>>>> - struct in6_addr daddr;
>>>> + union {
>>>> + struct {
>>>> + struct in6_addr saddr;
>>>> + struct in6_addr daddr;
>>>> + } addrs;
>>>> + struct {
>>>> + struct in6_addr saddr;
>>>> + struct in6_addr daddr;
>>>> + };
>>>
>>> addrs first? Consistancy is key :)
>>
>> I think addrs should be second. In general, I think all newly added
>> non-anonymous structures should be second.
>
> Why not use a local version of the macro like was done in the DRM header
> file, to make it always work the same and more obvious what is
> happening? If I were a userspace developer and saw the above, I would
> think that the kernel developers have lost it :)
Then don't take a look at this[1]. :p
--
Gustavo
[1] https://git.kernel.org/linus/c0a744dcaa29e9537e8607ae9c965ad936124a4d
next prev parent reply other threads:[~2021-07-28 6:53 UTC|newest]
Thread overview: 158+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-07-27 20:57 [PATCH 00/64] Introduce strict memcpy() bounds checking Kees Cook
2021-07-27 20:57 ` [PATCH 01/64] media: omap3isp: Extract struct group for memcpy() region Kees Cook
2021-07-28 0:55 ` Gustavo A. R. Silva
2021-07-28 1:50 ` Kees Cook
2021-07-28 8:59 ` David Sterba
2021-07-28 9:14 ` Dan Carpenter
2021-07-28 21:37 ` Bart Van Assche
2021-07-28 21:37 ` David Sterba
2021-07-29 5:56 ` Greg Kroah-Hartman
2021-07-29 8:20 ` Dan Carpenter
2021-07-30 6:00 ` Kees Cook
2021-07-30 8:38 ` David Sterba
2021-07-30 9:00 ` Dan Carpenter
2021-07-30 16:44 ` Kees Cook
2021-07-30 17:08 ` Nick Desaulniers
2021-07-30 19:18 ` Kees Cook
2021-07-27 20:57 ` [PATCH 02/64] mac80211: Use flex-array for radiotap header bitmap Kees Cook
2021-07-28 7:35 ` Dan Carpenter
2021-07-28 9:23 ` David Sterba
2021-07-28 21:54 ` Kees Cook
2021-07-29 10:45 ` David Sterba
2021-07-30 6:06 ` Kees Cook
2021-07-28 21:20 ` Kees Cook
2021-07-28 23:14 ` Kees Cook
2021-07-28 23:33 ` Kees Cook
2021-07-29 8:25 ` Dan Carpenter
2021-07-27 20:57 ` [PATCH 03/64] rpmsg: glink: Replace strncpy() with strscpy_pad() Kees Cook
2021-07-28 2:07 ` Gustavo A. R. Silva
2021-07-27 20:57 ` [PATCH 04/64] stddef: Introduce struct_group() helper macro Kees Cook
2021-07-28 2:32 ` Gustavo A. R. Silva
2021-07-28 10:54 ` Rasmus Villemoes
2021-07-28 21:59 ` Kees Cook
2021-07-30 22:19 ` Williams, Dan J
2021-07-31 2:59 ` Kees Cook
[not found] ` <CAKwiHFheDv2pwsm6Fa+-KnOFyvk7bfZQjb2BQ-CSwH61gxgVYg@mail.gmail.com>
2021-07-31 15:10 ` Kees Cook
2021-07-27 20:57 ` [PATCH 05/64] skbuff: Switch structure bounds to struct_group() Kees Cook
2021-07-28 3:50 ` Gustavo A. R. Silva
2021-07-27 20:57 ` [PATCH 06/64] bnxt_en: Use struct_group_attr() for memcpy() region Kees Cook
2021-07-28 1:03 ` Michael Chan
2021-07-28 4:45 ` Gustavo A. R. Silva
2021-07-27 20:57 ` [PATCH 07/64] staging: rtl8192e: Use struct_group() " Kees Cook
2021-07-27 22:30 ` Larry Finger
2021-07-28 5:45 ` Greg Kroah-Hartman
2021-07-27 20:57 ` [PATCH 08/64] staging: rtl8192u: " Kees Cook
2021-07-28 5:45 ` Greg Kroah-Hartman
2021-07-27 20:58 ` [PATCH 09/64] staging: rtl8723bs: Avoid field-overflowing memcpy() Kees Cook
2021-07-28 5:46 ` Greg Kroah-Hartman
2021-07-27 20:58 ` [PATCH 10/64] lib80211: Use struct_group() for memcpy() region Kees Cook
2021-07-28 5:52 ` Greg Kroah-Hartman
2021-08-13 8:04 ` Johannes Berg
2021-08-13 15:49 ` Kees Cook
2021-08-13 19:44 ` Johannes Berg
2021-07-27 20:58 ` [PATCH 11/64] net/mlx5e: Avoid field-overflowing memcpy() Kees Cook
2021-07-27 20:58 ` [PATCH 12/64] mwl8k: Use struct_group() for memcpy() region Kees Cook
2021-07-27 20:58 ` [PATCH 13/64] libertas: " Kees Cook
2021-07-27 20:58 ` [PATCH 14/64] libertas_tf: " Kees Cook
2021-07-27 20:58 ` [PATCH 15/64] ipw2x00: " Kees Cook
2021-07-28 18:55 ` Stanislav Yakovlev
2021-07-27 20:58 ` [PATCH 16/64] thermal: intel: int340x_thermal: " Kees Cook
2021-07-27 20:58 ` [PATCH 17/64] iommu/amd: " Kees Cook
2021-07-27 20:58 ` [PATCH 18/64] cxgb3: " Kees Cook
2021-07-27 20:58 ` [PATCH 19/64] ip: Use struct_group() for memcpy() regions Kees Cook
2021-07-28 5:55 ` Greg Kroah-Hartman
2021-07-28 6:14 ` Gustavo A. R. Silva
2021-07-28 6:19 ` Greg Kroah-Hartman
2021-07-28 6:31 ` Gustavo A. R. Silva [this message]
2021-07-28 6:37 ` Gustavo A. R. Silva
2021-07-28 6:41 ` Greg Kroah-Hartman
2021-07-28 21:01 ` Kees Cook
2021-07-29 1:59 ` Bart Van Assche
2021-07-27 20:58 ` [PATCH 20/64] intersil: Use struct_group() for memcpy() region Kees Cook
2021-07-27 20:58 ` [PATCH 21/64] cxgb4: " Kees Cook
2021-07-27 20:58 ` [PATCH 22/64] bnx2x: " Kees Cook
2021-07-27 20:58 ` [PATCH 23/64] drm/amd/pm: " Kees Cook
2021-07-30 2:07 ` Alex Deucher
2021-07-27 20:58 ` [PATCH 24/64] staging: wlan-ng: " Kees Cook
2021-07-28 5:45 ` Greg Kroah-Hartman
2021-07-27 20:58 ` [PATCH 25/64] drm/mga/mga_ioc32: " Kees Cook
2021-07-28 5:56 ` Greg Kroah-Hartman
2021-07-29 12:11 ` Daniel Vetter
2021-07-31 4:20 ` Kees Cook
2021-07-27 20:58 ` [PATCH 26/64] net/mlx5e: " Kees Cook
2021-07-27 20:58 ` [PATCH 27/64] HID: cp2112: " Kees Cook
2021-07-27 20:58 ` [PATCH 28/64] compiler_types.h: Remove __compiletime_object_size() Kees Cook
2021-07-27 20:58 ` [PATCH 29/64] lib/string: Move helper functions out of string.c Kees Cook
2021-07-27 20:58 ` [PATCH 30/64] fortify: Move remaining fortify helpers into fortify-string.h Kees Cook
2021-07-27 20:58 ` [PATCH 31/64] fortify: Explicitly disable Clang support Kees Cook
2021-07-27 21:18 ` Nathan Chancellor
2021-07-27 21:47 ` Kees Cook
2021-07-27 20:58 ` [PATCH 32/64] fortify: Add compile-time FORTIFY_SOURCE tests Kees Cook
2021-07-27 20:58 ` [PATCH 33/64] lib: Introduce CONFIG_TEST_MEMCPY Kees Cook
2021-07-27 23:31 ` Bart Van Assche
2021-07-27 23:33 ` Randy Dunlap
2021-07-28 1:30 ` Kees Cook
2021-07-27 20:58 ` [PATCH 34/64] fortify: Detect struct member overflows in memcpy() at compile-time Kees Cook
2021-07-27 22:43 ` Nick Desaulniers
2021-07-28 1:47 ` Kees Cook
2021-07-28 11:19 ` Rasmus Villemoes
2021-07-30 2:39 ` Kees Cook
2021-07-27 20:58 ` [PATCH 35/64] fortify: Detect struct member overflows in memmove() " Kees Cook
2021-07-27 20:58 ` [PATCH 36/64] scsi: ibmvscsi: Avoid multi-field memset() overflow by aiming at srp Kees Cook
2021-07-28 1:39 ` Martin K. Petersen
2021-07-28 18:57 ` Kees Cook
2021-07-29 3:35 ` Martin K. Petersen
2021-07-30 19:11 ` Tyrel Datwyler
2021-07-30 18:16 ` Tyrel Datwyler
2021-07-27 20:58 ` [PATCH 37/64] string.h: Introduce memset_after() for wiping trailing members/padding Kees Cook
2021-07-27 20:58 ` [PATCH 38/64] xfrm: Use memset_after() to clear padding Kees Cook
2021-07-27 20:58 ` [PATCH 39/64] mac80211: Use memset_after() to clear tx status Kees Cook
2021-07-31 15:55 ` Kees Cook
2021-08-13 7:40 ` Johannes Berg
2021-08-13 16:08 ` Kees Cook
2021-08-13 18:19 ` Johannes Berg
2021-08-13 7:41 ` Johannes Berg
2021-07-27 20:58 ` [PATCH 40/64] net: 802: Use memset_after() to clear struct fields Kees Cook
2021-07-27 20:58 ` [PATCH 41/64] net: dccp: Use memset_after() for TP zeroing Kees Cook
2021-07-27 20:58 ` [PATCH 42/64] net: qede: Use memset_after() for counters Kees Cook
2021-07-31 16:07 ` Kees Cook
2021-07-27 20:58 ` [PATCH 43/64] ath11k: Use memset_after() for clearing queue descriptors Kees Cook
2021-07-27 20:58 ` [PATCH 44/64] iw_cxgb4: Use memset_after() for cpl_t5_pass_accept_rpl Kees Cook
2021-07-27 20:58 ` [PATCH 45/64] intel_th: msu: Use memset_after() for clearing hw header Kees Cook
2021-07-27 20:58 ` [PATCH 46/64] IB/mthca: Use memset_after() for clearing mpt_entry Kees Cook
2021-07-27 20:58 ` [PATCH 47/64] btrfs: Use memset_after() to clear end of struct Kees Cook
2021-07-28 9:42 ` David Sterba
2021-07-28 21:56 ` Kees Cook
2021-07-29 10:33 ` David Sterba
2021-07-31 15:25 ` Kees Cook
2021-08-09 11:20 ` David Sterba
2021-07-27 20:58 ` [PATCH 48/64] drbd: Use struct_group() to zero algs Kees Cook
2021-07-28 21:45 ` Bart Van Assche
2021-07-30 2:31 ` Kees Cook
2021-07-30 2:57 ` Bart Van Assche
2021-07-30 9:25 ` Lars Ellenberg
2021-07-30 15:32 ` Nick Desaulniers
2021-07-27 20:58 ` [PATCH 49/64] cm4000_cs: Use struct_group() to zero struct cm4000_dev region Kees Cook
2021-07-28 5:48 ` Greg Kroah-Hartman
2021-07-27 20:58 ` [PATCH 50/64] KVM: x86: Use struct_group() to zero decode cache Kees Cook
2021-07-27 20:58 ` [PATCH 51/64] tracing: Use struct_group() to zero struct trace_iterator Kees Cook
2021-07-27 20:58 ` [PATCH 52/64] dm integrity: Use struct_group() to zero struct journal_sector Kees Cook
2021-07-27 20:58 ` [PATCH 53/64] HID: roccat: Use struct_group() to zero kone_mouse_event Kees Cook
2021-07-27 20:58 ` [PATCH 54/64] ipv6: Use struct_group() to zero rt6_info Kees Cook
2021-07-29 18:58 ` Jakub Kicinski
2021-07-31 15:01 ` Kees Cook
2021-07-27 20:58 ` [PATCH 55/64] RDMA/mlx5: Use struct_group() to zero struct mlx5_ib_mr Kees Cook
2021-07-27 20:58 ` [PATCH 56/64] ethtool: stats: Use struct_group() to clear all stats at once Kees Cook
2021-07-27 20:58 ` [PATCH 57/64] netfilter: conntrack: Use struct_group() to zero struct nf_conn Kees Cook
2021-07-27 20:58 ` [PATCH 58/64] powerpc: Split memset() to avoid multi-field overflow Kees Cook
2021-08-05 11:36 ` Michael Ellerman
2021-07-27 20:58 ` [PATCH 59/64] fortify: Detect struct member overflows in memset() at compile-time Kees Cook
2021-07-27 20:58 ` [PATCH 60/64] fortify: Work around Clang inlining bugs Kees Cook
2021-07-27 20:58 ` [PATCH 61/64] Makefile: Enable -Warray-bounds Kees Cook
2021-07-27 20:58 ` [PATCH 62/64] netlink: Avoid false-positive memcpy() warning Kees Cook
2021-07-28 5:49 ` Greg Kroah-Hartman
2021-07-28 11:24 ` Rasmus Villemoes
2021-07-30 1:39 ` Kees Cook
2021-07-30 1:41 ` Kees Cook
2021-07-27 20:58 ` [PATCH 63/64] iwlwifi: dbg_ini: Split memcpy() to avoid multi-field write Kees Cook
2021-07-27 20:58 ` [PATCH 64/64] fortify: Add run-time WARN for cross-field memcpy() Kees Cook
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=e3193698-86d5-d529-e095-e09b4d52927b@embeddedor.com \
--to=gustavo@embeddedor.com \
--cc=akpm@linux-foundation.org \
--cc=clang-built-linux@googlegroups.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=gregkh@linuxfoundation.org \
--cc=gustavoars@kernel.org \
--cc=keescook@chromium.org \
--cc=keithpac@amazon.com \
--cc=linux-block@vger.kernel.org \
--cc=linux-hardening@vger.kernel.org \
--cc=linux-kbuild@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-staging@lists.linux.dev \
--cc=linux-wireless@vger.kernel.org \
--cc=netdev@vger.kernel.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 NNTP newsgroup(s).