netdev - RE: [B.A.T.M.A.N.] [PATCH] Fix ARM BUILD_BUG

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <AE90C24D6B3A694183C094C60CF0A2F6026B7453@saturn3.aculab.com>
Date:	Mon, 2 Dec 2013 13:10:12 -0000
From:	"David Laight" <David.Laight@...LAB.COM>
To:	"Antonio Quartulli" <antonio@...hcoding.com>,
	"Russell King - ARM Linux" <linux@....linux.org.uk>
Cc:	"The list for a Better Approach To Mobile Ad-hoc Networking" 
	<b.a.t.m.a.n@...ts.open-mesh.org>, <netdev@...r.kernel.org>,
	"David Miller" <davem@...emloft.net>
Subject: RE: [B.A.T.M.A.N.] [PATCH] Fix ARM BUILD_BUG_ON() errors withbatman-adv

> From: Antonio Quartulli
> On 02/12/13 11:24, David Laight wrote:
> > ...
> >> The only solutions are: use the GCC packed attribute, redesign
> >> the structures...
> >
> > It is probably enough to mark the inner structure containing the
> > three byte fields 'packed'. Marking it aligned(1) might also have
> > the desired effect. The outer structure should then be ok. But
> > would need to use a specially named attribute so it doesn't get
> > removed.
> 
> This may work with the structures I reported in a previous email, but
> it is not a good solution for us because we have other more complex
> substructs that cannot be packed that way.

You really don't want to have structures that need to be (say) 7 bytes
long and contain a 32bit value that will always be correctly aligned.

IIRC adding __attribute__((aligned(1))) to a structure isn't the same
as marking it 'packed' - it doesn't normally reduce the alignment
for structures (but can be used to force a reduced alignment on a
structure member - eg a 4byte aligned 64bit integer).

I don't have an arm compiler handy.

> I think we will simply duplicate the members and avoid substructs in
> our packets.

Can be easiest...


	David



--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html