[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <54fe1ebc-6b3a-178c-d7d2-fb5486f54e8d@embeddedor.com>
Date: Fri, 23 Jun 2023 08:14:34 -0600
From: "Gustavo A. R. Silva" <gustavo@...eddedor.com>
To: "Maciej W. Rozycki" <macro@...am.me.uk>,
Thomas Bogendoerfer <tsbogend@...ha.franken.de>
Cc: "Gustavo A. R. Silva" <gustavoars@...nel.org>,
linux-mips@...r.kernel.org, linux-kernel@...r.kernel.org,
linux-hardening@...r.kernel.org
Subject: Re: [PATCH][next] MIPS: dec: prom: Address -Warray-bounds warning
On 6/23/23 07:41, Maciej W. Rozycki wrote:
> On Fri, 23 Jun 2023, Thomas Bogendoerfer wrote:
>
>>> Zero-length arrays are deprecated, and we are replacing them with flexible
>>> array members instead. So, replace zero-length array with flexible-array
>>> member in struct memmap.
>
> Technically it is a semantics bug fix actually, as the TURBOchannel
> firmware specification (from Jan 1993) says it's:
>
> typedef struct{ int pagesize; unsigned char bitmap[];}memmap;
> int getbitmap(memmap *map);
>
> (formatting preserved as in the document) so it should have always been a
> flexible array member. Maybe old (2.x) GCC versions had an issue with it
> or something, as otherwise I can't imagine why whoever added our typedef
> did it differently from the spec.
Apparently, flexible-array members were supported in some 2.x versions as an
extension to the language; and it was not until GCC 3.0 that they were fully
supported by the compiler.
>
>> applied to mips-next.
>
> Not sure if you can retrofit it, but:
>
> Acked-by: Maciej W. Rozycki <macro@...am.me.uk>
Thanks!
--
Gustavo
Powered by blists - more mailing lists