Re: Dropping support for the smallest armel machines

[Ansgar Burchardt <ansgar@debian.org>]

Hi,

[ Please CC me in replies. I'm not subscribed to the list. ]

On 06/27/2013 19:22, Ben Hutchings wrote:
> On Wed, Jun 26, 2013 at 04:39:15PM +0100, Ben Hutchings wrote:
>> Bet they're slower than QEMU versatile emulation on an x86.  And DSA
>> loves virtual machines. :-)
> 
> This speed comparison might not be true, and I know people worry about
> accuracy of emulation and being able to reproduce architecture-
> specific bugs.
> 
> The other option that has been suggested repeatedly is to put armel
> chroots on ARMv7 hardware.  Unaligned accesses behave differently on
> v7, but they weren't consistent between different v5 implementations
> (http://www.heyrick.co.uk/armwiki/Unaligned_data_access) so I don't
> think this is critical.

We could try setting up armel chroots on ARMv7 hardware (the armhf
buildds) for experimental and maybe wheezy-backports as well.

This way packages entering the next release would still be built on
ARMv5 hardware, but we could see how well it works in experimental.

For -backports the packages should already have been built in unstable,
though there's still a small risk of bugs if built on ARMv7 hardware
instead of v5. However there's currently no buildds for
wheezy-backports/armel at all (as far as I know)...

Ansgar