From: olof@lixom•net (Olof Johansson)
To: Steve Munroe <sjmunroe@us•ibm.com>
Cc: linuxppc-dev list <linuxppc-dev@ozlabs•org>,
Paul Mackerras <paulus@samba•org>,
Anton Blanchard <anton@samba•org>,
Ulrich Weigand <Ulrich.Weigand@de•ibm.com>
Subject: Re: Saving to 32 bits of GPRs in signal context
Date: Wed, 30 May 2007 15:52:39 -0500 [thread overview]
Message-ID: <20070530205239.GA9462@lixom.net> (raw)
In-Reply-To: <OF3CF62163.CD3DCA16-ON862572EB.006C0F2C-862572EB.006CBB1C@us.ibm.com>
On Wed, May 30, 2007 at 02:47:37PM -0500, Steve Munroe wrote:
>
> Kumar Gala <galak@kernel•crashing.org> wrote on 05/30/2007 12:31:32 AM:
>
> >
> > On May 29, 2007, at 9:54 PM, Steve Munroe wrote:
> >
> > >
> > > Kumar Gala <galak@kernel•crashing.org> wrote on 05/29/2007 07:43:05
> > > PM:
> > >
> > >>
> > >> On May 29, 2007, at 6:46 PM, Olof Johansson wrote:
> > >>
> > >>> On Wed, May 30, 2007 at 07:32:33AM +1000, Benjamin Herrenschmidt
> > >>> wrote:
> > >>>> On Tue, 2007-05-29 at 08:10 -0500, Kumar Gala wrote:
> > >>>>> This is all problematic since some 64-bit implementations may not
> > >>>>> guarantee the upper bits are valid when in 32-bit mode. Look
> > >>>>> at the
> > >>>>> 'Computation Modes' section in the architecture specs 2.03 or
> > >>>>> greater
> > >>>>> for embedded processors.
> > >>>>
> > >>>> Yuck. Well, we might need to export a spearate CPU feature bit to
> > >>>> indicate that it's the case then.
> > >>>
> > >>> No need for a new bit, you should be able to key off of
> > >>> PPC_FEATURE_64
> > >>> && !PPC_FEATURE_BOOKE.
> > >>
> > >> Nope, the architecture allows embedded to behave like server parts
> > >> and support the full 64-bit registers. We really should have a new
> > >> feature bit so that if someone has an implementation of an embedded
> > >> part that supports the functionality, they get the benefit.
> > >>
> > > When such exists we can add a bit, until then we can wait. The current
> > > 32-bit AT_HWCAP is almost full. so we should not allocate bits on
> > > speculation.
> >
> > Understandable.. dare I ask about a few of the current AT_HWCAPs we
> > do have:
> >
> > #define PPC_FEATURE_POWER4 0x00080000
> > #define PPC_FEATURE_POWER5 0x00040000
> > #define PPC_FEATURE_POWER5_PLUS 0x00020000
> > #define PPC_FEATURE_ARCH_2_05 0x00001000
> > #define PPC_FEATURE_PA6T 0x00000800
> > #define PPC_FEATURE_POWER6_EXT 0x00000200
> >
> > What exactly are we using these for? Can we not use platform for
> > some of these?
> >
> These are poorly named ISA versions
>
> PPC_FEATURE_POWER4 == PPC_FEATURE_ARCH_2_0
> PPC_FEATURE_POWER5 == PPC_FEATURE_ARCH_2_02
> PPC_FEATURE_POWER5+ == PPC_FEATURE_ARCH_2_03
>
> Ask Olof about this but I think
> PPC_FEATURE_PA6T == PPC_FEATURE_ARCH_2_04
> but I think it is more then 2_04 and less than 2_05.
The problem is that IBM has never (before) had to care about what was
implementation and what was architecture. The implementation WAS the
architecture up until POWER5+, and the PPC ISA went lock-step with the
new server processor releases.
PA6T is 2.04 + a few 2.05 bits, give or take. But it's not equivalent of
POWER6 (nor is it equivalent of POWER5+, since they implement different
optional features of the architecture).
I'm not sure just how to make this scale down the road -- if we are to
use a PPC_FEATURE_* for every optional feature in the ISA, we'll run
out of bits in no time. If we end up using a flag per implementation,
it probably won't be quite as bad, but I'm guessing the actual code that
uses it will get hairier.
-Olof
next prev parent reply other threads:[~2007-05-30 20:47 UTC|newest]
Thread overview: 63+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-05-29 7:24 Saving to 32 bits of GPRs in signal context Benjamin Herrenschmidt
2007-05-29 7:52 ` Dan Malek
2007-05-29 8:05 ` Benjamin Herrenschmidt
2007-05-29 9:26 ` Gabriel Paubert
2007-05-29 9:44 ` Benjamin Herrenschmidt
2007-05-29 13:12 ` Segher Boessenkool
2007-05-29 14:00 ` Steve Munroe
2007-05-29 14:08 ` Ulrich Weigand
2007-05-29 14:17 ` Kumar Gala
2007-05-29 14:38 ` Segher Boessenkool
2007-05-29 19:04 ` Becky Bruce
2007-05-30 10:04 ` Christoph Hellwig
2007-05-30 12:13 ` Kumar Gala
2007-05-30 12:30 ` Segher Boessenkool
2007-05-29 14:31 ` Segher Boessenkool
2007-05-29 14:51 ` Steve Munroe
2007-05-29 21:44 ` Benjamin Herrenschmidt
2007-05-29 23:16 ` Steve Munroe
2007-05-29 23:19 ` Benjamin Herrenschmidt
2007-05-30 7:34 ` Hiroyuki Machida
2007-05-30 11:40 ` Segher Boessenkool
2007-05-30 11:48 ` Benjamin Herrenschmidt
2007-05-30 3:37 ` Paul Mackerras
2007-05-30 5:32 ` Kumar Gala
2007-05-30 11:44 ` Benjamin Herrenschmidt
2007-05-30 12:15 ` Kumar Gala
2007-05-30 12:48 ` Hiroyuki Machida
2007-05-30 12:58 ` Benjamin Herrenschmidt
2007-05-30 18:09 ` Steve Munroe
2007-05-30 21:02 ` Gabriel Paubert
2007-05-30 21:41 ` Steve Munroe
2007-05-30 12:01 ` Segher Boessenkool
2007-05-30 11:59 ` Segher Boessenkool
2007-05-30 12:01 ` Benjamin Herrenschmidt
2007-05-30 12:07 ` Segher Boessenkool
2007-05-30 12:09 ` Benjamin Herrenschmidt
2007-05-30 12:36 ` Segher Boessenkool
2007-05-29 14:28 ` Segher Boessenkool
2007-05-29 21:37 ` Benjamin Herrenschmidt
2007-05-29 21:38 ` Benjamin Herrenschmidt
2007-05-29 13:04 ` Segher Boessenkool
2007-05-29 14:28 ` Arnd Bergmann
2007-05-29 14:43 ` Segher Boessenkool
2007-05-29 15:54 ` Geert Uytterhoeven
2007-05-29 18:48 ` Arnd Bergmann
2007-05-29 21:27 ` Benjamin Herrenschmidt
2007-05-29 21:45 ` Felix Domke
2007-05-30 11:23 ` Benjamin Herrenschmidt
2007-05-30 11:52 ` Felix Domke
2007-05-30 13:14 ` Segher Boessenkool
2007-05-30 11:54 ` Segher Boessenkool
2007-05-30 12:07 ` Felix Domke
2007-05-31 5:39 ` Benjamin Herrenschmidt
2007-05-29 13:10 ` Kumar Gala
2007-05-29 21:32 ` Benjamin Herrenschmidt
2007-05-29 23:46 ` Olof Johansson
2007-05-30 0:43 ` Kumar Gala
2007-05-30 2:54 ` Steve Munroe
2007-05-30 5:31 ` Kumar Gala
2007-05-30 19:47 ` Steve Munroe
2007-05-30 20:52 ` Olof Johansson [this message]
2007-05-30 21:33 ` Steve Munroe
2007-05-29 13:53 ` Ulrich Weigand
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20070530205239.GA9462@lixom.net \
--to=olof@lixom$(echo .)net \
--cc=Ulrich.Weigand@de$(echo .)ibm.com \
--cc=anton@samba$(echo .)org \
--cc=linuxppc-dev@ozlabs$(echo .)org \
--cc=paulus@samba$(echo .)org \
--cc=sjmunroe@us$(echo .)ibm.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox