Re: [GIT PULL] x86/atomic changes for v2.6.35

["H. Peter Anvin" <hpa@zytor.com>]

On 05/19/2010 07:36 AM, Stephen Rothwell wrote:
> Hi,
> 
> On Wed, 19 May 2010 07:24:00 -0700 "H. Peter Anvin" <hpa@zytor.com> wrote:
>>
>> On 05/19/2010 04:46 AM, Geert Uytterhoeven wrote:
>>> <boilerplate>
>>> It's a pity this wasn't raised/resolved between its detection in linux-next and
>>> before it entered mainline...
>>> </boilerplate>
>>
>> As far as your boilerplate is concerned, I think Linus made it clear at
>> the Kernel Summit that is it not the obligation of x86/ARM/PowerPC to
>> slow down to not break the smaller architectures; it's the
>> responsibility of those architecture maintainers to keep up.  Sorry.
> 
> I don't think this reply has anything to do with the sentiments expr
> by Geert above.  My interpretation of his comments is just that it is a
> pity noone noticed the problem while it was only in linux-next and
> reported it widely (like on linux-arch) so something could have been done
> before it all Linus' tree.  There was no suggestion of slowing the pace
> of development.

It was discussed on linux-kernel -- note that there is no breakage for
smaller architectures unless you enable the test directly or via randconfig.

The other part is that generic atomic64_t has been available since
middle of 2009, and was *also* discussed extensively on linux-kernel --
in fact, several of the smaller architectures added support at that
time.  That the breakage occurred because of an inconsequential test
rather than real code is thus really nothing but fortunate.

	-hpa

-- 
H. Peter Anvin, Intel Open Source Technology Center
I work for Intel.  I don't speak on their behalf.