Re: Non-packed structures in IP headers

[Florian Weimer <fw@deneb.enyo.de>]

* Carles Cufi:

>> * Carles Cufi:
>> 
>> > I was looking through the structures for IPv{4,6} packet headers and
>> > noticed that several of those that seem to be used to parse a packet
>> > directly from the wire are not declared as packed. This surprised me
>> > because, although I did find that provisions are made so that the
>> > alignment of the structure, it is still technically possible for the
>> > compiler to inject padding bytes inside those structures, since AFAIK
>> > the C standard makes no guarantees about padding unless it's
>> > instructed to pack the structure.
>> 
>> The C standards do not make such guarantees, but the platform ABI
>> standards describe struct layout and ensure that there is no padding.
>> Linux relies on that not just for networking, but also for the userspace
>> ABI, support for separately compiled kernel modules, and in other places.
>
> That makes sense, but aren't ABI standards different for every
> architecture? For example, I checked the Arm AAPCS[1] and it states:
>
> "The size of an aggregate shall be the smallest multiple of its
> alignment that is sufficient to hold all of its members."
>
> Which, unless I am reading this wrong, means that the compiler would
> indeed insert padding if the size of the IP headers structs was not
> a multiple of 4. In this particular case, the struct sizes for the
> IP headers are 20 and 40 bytes respectively, so there will be no
> padding inserted. But I only checked a single architecture's ABI (or
> Procedure Call Standard) documentation, is this true for all archs?

For structure layout in memory, there is a large overlap between ABIs.
There is divergence around long long (which is easily avoided by
adding padding manually), and potentially bit fileds (but I haven't
looked at that).

Things only get weird for pass-by-value structs and unions and return
types.