[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: [Stretch] Status for architecture qualification



On Tue, Jun 07, 2016 at 11:38:06AM -0700, Martin Michlmayr wrote:
>* Steve McIntyre <steve@einval.com> [2016-06-06 15:14]:
>> However, I will admit (again) that armel is starting to lose upstream
>> support in some cases. I'm tempted to suggest that Stretch should be
>> the last release for armel for that reason.
>
>Which upstream problems do you see?

There's a few projects that have abandoned claiming to support
anything below ARMv7. The one that comes to mind most readily is
libv8, but there have been others.

>And do you know how long ARM/Linaro are planning to support it
>upstream?

*Personally* I reckon ARM are never going to stop supporting older
CPUs in toolchains etc., however any new development will obviously be
targeting newer v7 and v8 stuff. Linaro have never spent any time
supporting pre-v7, as none of the Linaro members care about anything
older than v7. Most are pushing for v8 primarily now.

However, I'd predict most of the issues are going to be in other
projects that neither ARM nor Linaro are working on. Now that even the
Pi has moved on from v6, there's not going to be much push for
supporting older CPU versions. I've had quite a few conversations with
folks who are confused why we still target v4t...

>I'm torn at the moment.  On the one hand, we have a lot of armel users
>and mainstream armel hardware was sold until fairly recently.  On the
>other hand, assuming LTS jessie will support armel, that might be long
>enough for the hardware to get mostly (or finally :) obsolete.

Yeah. As I've said, a partial architecture would be a good route here
but I've not seen anybody working on such an option.

>I definitely think we should make a decision one or the other way
>and document it appropriately if we intend to drop armel after
>stretch.

Nod.

-- 
Steve McIntyre, Cambridge, UK.                                steve@einval.com
"When C++ is your hammer, everything looks like a thumb." -- Steven M. Haflich


Reply to: