Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1320733 > unrolled thread

Re: [PATCH v9 3/6] arm64/arm, numa, dt: adding numa dt binding implementation for arm64 platforms.

Started byWill Deacon <will.deacon@arm.com>
First post2016-01-28 15:40 +0100
Last post2016-01-28 19:50 +0100
Articles 4 — 2 participants

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.


Contents

  Re: [PATCH v9 3/6] arm64/arm, numa, dt: adding numa dt binding  implementation for arm64 platforms. Will Deacon <will.deacon@arm.com> - 2016-01-28 15:40 +0100
    Re: [PATCH v9 3/6] arm64/arm, numa, dt: adding numa dt binding  implementation for arm64 platforms. Ganapatrao Kulkarni <gpkulkarni@gmail.com> - 2016-01-28 18:20 +0100
      Re: [PATCH v9 3/6] arm64/arm, numa, dt: adding numa dt binding  implementation for arm64 platforms. Will Deacon <will.deacon@arm.com> - 2016-01-28 19:10 +0100
        Re: [PATCH v9 3/6] arm64/arm, numa, dt: adding numa dt binding  implementation for arm64 platforms. Ganapatrao Kulkarni <gpkulkarni@gmail.com> - 2016-01-28 19:50 +0100

#1320733 — Re: [PATCH v9 3/6] arm64/arm, numa, dt: adding numa dt binding implementation for arm64 platforms.

FromWill Deacon <will.deacon@arm.com>
Date2016-01-28 15:40 +0100
SubjectRe: [PATCH v9 3/6] arm64/arm, numa, dt: adding numa dt binding implementation for arm64 platforms.
Message-ID<qVVTQ-4Lm-3@gated-at.bofh.it>
On Tue, Jan 26, 2016 at 02:36:04PM -0600, Bjorn Helgaas wrote:
> Subject is "arm64/arm, numa, dt: adding ..."  What is the significance
> of the "arm" part?  The other patches only mention "arm64".
> 
> General comment: the code below has little, if anything, that is
> actually arm64-specific.  Maybe this is the first DT-based NUMA
> platform?  I don't see other similar code for other arches, so maybe
> it's too early to try to generalize it, but we should try to avoid
> adding duplicates of this code if/when other arches do show up.

Having it in the core code would allow us to share it with arch/arm/
fairly straightforwardly.

Will

[toc] | [next] | [standalone]


#1320892

FromGanapatrao Kulkarni <gpkulkarni@gmail.com>
Date2016-01-28 18:20 +0100
Message-ID<qVYoG-6D9-19@gated-at.bofh.it>
In reply to#1320733
Hi Will,


On Thu, Jan 28, 2016 at 8:09 PM, Will Deacon <will.deacon@arm.com> wrote:
> On Tue, Jan 26, 2016 at 02:36:04PM -0600, Bjorn Helgaas wrote:
>> Subject is "arm64/arm, numa, dt: adding ..."  What is the significance
>> of the "arm" part?  The other patches only mention "arm64".
>>
>> General comment: the code below has little, if anything, that is
>> actually arm64-specific.  Maybe this is the first DT-based NUMA
>> platform?  I don't see other similar code for other arches, so maybe
>> it's too early to try to generalize it, but we should try to avoid
>> adding duplicates of this code if/when other arches do show up.
>
> Having it in the core code would allow us to share it with arch/arm/
> fairly straightforwardly.
This binding can be used for arm too.
however at this moment it is the need of arm64 platforms.
can we please keep this to arm64 as it's too early to try to
generalize it(as Bjorn suggested)
I prefer to keep it as it is, otherwise ok.
Please suggest.
>
> Will
thanks
Ganapat

[toc] | [prev] | [next] | [standalone]


#1320931

FromWill Deacon <will.deacon@arm.com>
Date2016-01-28 19:10 +0100
Message-ID<qVZb3-7go-9@gated-at.bofh.it>
In reply to#1320892
On Thu, Jan 28, 2016 at 10:42:17PM +0530, Ganapatrao Kulkarni wrote:
> On Thu, Jan 28, 2016 at 8:09 PM, Will Deacon <will.deacon@arm.com> wrote:
> > On Tue, Jan 26, 2016 at 02:36:04PM -0600, Bjorn Helgaas wrote:
> >> Subject is "arm64/arm, numa, dt: adding ..."  What is the significance
> >> of the "arm" part?  The other patches only mention "arm64".
> >>
> >> General comment: the code below has little, if anything, that is
> >> actually arm64-specific.  Maybe this is the first DT-based NUMA
> >> platform?  I don't see other similar code for other arches, so maybe
> >> it's too early to try to generalize it, but we should try to avoid
> >> adding duplicates of this code if/when other arches do show up.
> >
> > Having it in the core code would allow us to share it with arch/arm/
> > fairly straightforwardly.
> This binding can be used for arm too.
> however at this moment it is the need of arm64 platforms.
> can we please keep this to arm64 as it's too early to try to
> generalize it(as Bjorn suggested)
> I prefer to keep it as it is, otherwise ok.
> Please suggest.

My suggestions time and time again on the NUMA patches from you have
consistently been around consolidation of existing code, or moving things
that aren't architecture-specific out of the architecture code.

Will

[toc] | [prev] | [next] | [standalone]


#1320944

FromGanapatrao Kulkarni <gpkulkarni@gmail.com>
Date2016-01-28 19:50 +0100
Message-ID<qVZNM-7xo-17@gated-at.bofh.it>
In reply to#1320931
On Thu, Jan 28, 2016 at 11:38 PM, Will Deacon <will.deacon@arm.com> wrote:
> On Thu, Jan 28, 2016 at 10:42:17PM +0530, Ganapatrao Kulkarni wrote:
>> On Thu, Jan 28, 2016 at 8:09 PM, Will Deacon <will.deacon@arm.com> wrote:
>> > On Tue, Jan 26, 2016 at 02:36:04PM -0600, Bjorn Helgaas wrote:
>> >> Subject is "arm64/arm, numa, dt: adding ..."  What is the significance
>> >> of the "arm" part?  The other patches only mention "arm64".
>> >>
>> >> General comment: the code below has little, if anything, that is
>> >> actually arm64-specific.  Maybe this is the first DT-based NUMA
>> >> platform?  I don't see other similar code for other arches, so maybe
>> >> it's too early to try to generalize it, but we should try to avoid
>> >> adding duplicates of this code if/when other arches do show up.
>> >
>> > Having it in the core code would allow us to share it with arch/arm/
>> > fairly straightforwardly.
>> This binding can be used for arm too.
>> however at this moment it is the need of arm64 platforms.
>> can we please keep this to arm64 as it's too early to try to
>> generalize it(as Bjorn suggested)
>> I prefer to keep it as it is, otherwise ok.
>> Please suggest.
>
> My suggestions time and time again on the NUMA patches from you have
> consistently been around consolidation of existing code, or moving things
> that aren't architecture-specific out of the architecture code.

thanks, i shall move this out to drivers/of
>
> Will

thanks
Ganapat

[toc] | [prev] | [standalone]


Back to top | Article view | linux.kernel


csiph-web