Menu
▾
▴
Re: [Pearpc-devel] [patch] MMU translations property
|
From: James M. <jam...@tr...> - 2008-06-10 08:46:33
|
Cassy Foesch wrote:
> Well, is this something "new" described my OpenBIOS, or is it actually
> something that was originally defined in OpenFirmware?
>
> 2008/6/10 James Molloy <jam...@tr...
> <mailto:jam...@tr...>>:
>
> Cassy Foesch wrote:
> > This is an OpenBios feature? Or OpenFirmware? Or where are the
> > specifications written?
> >
> > Basically, the OpenFirmware implementation on PearPC is pretty
> buggy,
> > and incomplete. If it does just run twice, then it doesn't look
> like
> > there would be any problems with even an inefficient version
> running.
> > It sounds like it works pretty good, and I'll have to examine the
> > code a little more, but so far, it looks pretty good.
> >
> > 2008/6/9 James Molloy <jam...@tr...
> <mailto:jam...@tr...>
> > <mailto:jam...@tr... <mailto:jam...@tr...>>>:
> >
> > /Apologies for the possible repost - I checked the mailing list
> > archives
> > and my message didn't seem to appear, so I've now subscribed
> to the
> > mailing list to check if it does./
> >
> > Hi,
> >
> > I'm using PearPC to assist me in developing a hobby OS for
> the PowerPC
> > platform (in addition to real hardware) and noticed that PearPC
> > doesn't
> > expose the "translations" property of the MMU firmware
> device node.
> >
> > The translations property is documented in the OpenBios PowerPC
> > bindings
> > document and is aimed at helping the client program
> replicate the HTAB
> > before taking full control of memory management. I took the
> liberty of
> > creating this property myself.
> >
> > As far as I can tell, PearPC doesn't maintain an explicit
> list of
> > mmu->map() calls, so my algorithm searches the HTAB for
> every page in
> > the effective address space and manually constructs a list of
> > them. It's
> > O(n), but since it's only intended to be called twice (once
> from the
> > getproplen call and once from the getprop call) I don't see that
> > as much
> > of a handicap. It executes seemingly instantaneously on my
> Core2Duo
> > running the x86_64 JITC cpu.
> >
> > Anyway, let me know if it's not up to coding standard or
> there are too
> > few comments and I'll change them.
> >
> > Cheers,
> >
> > James Molloy
> >
> > Index: src/io/prom/promdt.cc
> > ==============================
> > =====================================
> > RCS file: /cvsroot/pearpc/pearpc/src/io/prom/promdt.cc,v
> > retrieving revision 1.29
> > diff -u -r1.29 promdt.cc
> > --- src/io/prom/promdt.cc 16 Aug 2005 12:26:43 -0000 1.29
> > +++ src/io/prom/promdt.cc 8 Jun 2008 09:48:24 -0000
> > @@ -1032,6 +1032,95 @@
> > }
> > }
> >
> > +/*
> > + * PromPropMMU: Implements the MMU's "translations" property,
> > which allows
> > + * client programs to replicate the HTAB contents before
> taking over
> > + * memory management.
> > + */
> > +PromPropMMU::PromPropMMU(const char *aName)
> > + :PromProp(aName), nValidTranslations(0)
> > +{
> > +}
> > +
> > +PromPropMMU::~PromPropMMU()
> > +{
> > +}
> > +
> > +uint32 PromPropMMU::getValueLen()
> > +{
> > + calculateTranslations();
> > + return nValidTranslations*sizeof(Translation);
> > +}
> > +
> > +uint32 PromPropMMU::getValue(uint32 buf, uint32 buflen)
> > +{
> > + calculateTranslations();
> > + if (nValidTranslations*sizeof(Translation) < buflen)
> > + {
> > + buflen = nValidTranslations*sizeof(Translation);
> > + }
> > +
> > + // Convert all values to BE.
> > + for (int i = 0; i < nValidTranslations; i++)
> > + {
> > + translations[i].virt =
> ppc_word_to_BE(translations[i].virt);
> > + translations[i].phys =
> ppc_word_to_BE(translations[i].phys);
> > + translations[i].size =
> ppc_word_to_BE(translations[i].size);
> > + translations[i].mode =
> ppc_word_to_BE(translations[i].mode);
> > + }
> > +
> > + uint32 phys;
> > + ppc_prom_effective_to_physical(phys, buf);
> > + ppc_dma_write(phys, (void*)translations, buflen);
> > + return 0;
> > +}
> > +
> > +uint32 PromPropMMU::setValue(uint32 buf, uint32 buflen)
> > +{
> > + return -1;
> > +}
> > +
> > +void PromPropMMU::calculateTranslations()
> > +{
> > + nValidTranslations = 0;
> > + uint32 phys;
> > + // For every page in the address space, check for a mapping.
> > + for (uint32 i = 0; i <= 0xFFFFF; i++)
> > + {
> > + uint32 ea = i << 12;
> > + if (ppc_prom_effective_to_physical(phys, ea))
> > + {
> > + // A valid translation exists. Add to the
> translations list.
> > + addTranslation(ea, phys);
> > + }
> > + }
> > +}
> > +
> > +void PromPropMMU::addTranslation(uint32 ea, uint32 phys)
> > +{
> > + // Can the given translation be appended to translation
> > 'nValidTranslations-1'?
> > + if ( (nValidTranslations > 0) &&
> > + (translations[nValidTranslations-1].virt +
> > translations[nValidTranslations-1].size == ea) &&
> > + (translations[nValidTranslations-1].phys +
> > translations[nValidTranslations-1].size == phys) )
> > + {
> > + // Modify 'nValidTranslations'.
> > + translations[nValidTranslations-1].size += 0x1000;
> > + }
> > + else
> > + {
> > + // Else add a new translation.
> > + translations[nValidTranslations].virt = ea;
> > + translations[nValidTranslations].phys = phys;
> > + translations[nValidTranslations].size = 0x1000;
> > + translations[nValidTranslations].mode = 0x6a;
> > + nValidTranslations++;
> > + if (nValidTranslations == NUM_TRANSLATIONS)
> > + {
> > + IO_PROM_ERR("PromPropMMU::calculateTranslations - too
> many
> > translations.");
> > + }
> > + }
> > +}
> > +
> >
> /*******************************************************************************
> > *
> > */
> > @@ -1111,6 +1200,8 @@
> > gPromRoot->addNode(cpus);
> > gPromRoot->addNode(kbd);
> >
> > + mmu->addProp(new PromPropMMU("translations"));
> > +
> > PromNode *cpu = new PromNode("PowerPC,G4");
> > cpus->addNode(cpu);
> > cpus->addProp(new PromPropInt("#size-cells", 0));
> > Index: src/io/prom/promdt.h
> >
> ===================================================================
> > RCS file: /cvsroot/pearpc/pearpc/src/io/prom/promdt.h,v
> > retrieving revision 1.3
> > diff -u -r1.3 promdt.h
> > --- src/io/prom/promdt.h 21 May 2004 22:29:43 -0000 1.3
> > +++ src/io/prom/promdt.h 8 Jun 2008 09:48:24 -0000
> > @@ -272,6 +272,30 @@
> > virtual uint32 setValue(uint32 buf, uint32
> buflen);
> > };
> >
> > +#define NUM_TRANSLATIONS 256
> > +class PromPropMMU: public PromProp {
> > +public:
> > + PromPropMMU(const char *name);
> > + virtual ~PromPropMMU();
> > +
> > + virtual uint32 getValueLen();
> > + virtual uint32 getValue(uint32 buf,
> uint32 buflen);
> > + virtual uint32 setValue(uint32 buf, uint32
> buflen);
> > +
> > +private:
> > + void calculateTranslations();
> > + void addTranslation(uint32 ea, uint32 phys);
> > + struct Translation
> > + {
> > + uint32 virt;
> > + uint32 size;
> > + uint32 phys;
> > + uint32 mode;
> > + };
> > + Translation translations[NUM_TRANSLATIONS];
> > + uint32 nValidTranslations;
> > +};
> > +
> > #define FIND_DEVICE_FIND 0
> > #define FIND_DEVICE_OPEN 1
> > PromNode *findDevice(const char *aPathName, int type,
> > PromInstanceHandle *ret);
> >
> >
> -------------------------------------------------------------------------
> > Check out the new SourceForge.net Marketplace.
> > It's the best place to buy or sell services for
> > just about anything Open Source.
> > http://sourceforge.net/services/buy/index.php
> > _______________________________________________
> > Pearpc-devel mailing list
> > Pea...@li...
> <mailto:Pea...@li...>
> > <mailto:Pea...@li...
> <mailto:Pea...@li...>>
> > https://lists.sourceforge.net/lists/listinfo/pearpc-devel
> >
> >
> >
> >
> > --
> > Cassy
> >
> ------------------------------------------------------------------------
> >
> >
> -------------------------------------------------------------------------
> > Check out the new SourceForge.net Marketplace.
> > It's the best place to buy or sell services for
> > just about anything Open Source.
> > http://sourceforge.net/services/buy/index.php
> >
> ------------------------------------------------------------------------
> >
> > _______________________________________________
> > Pearpc-devel mailing list
> > Pea...@li...
> <mailto:Pea...@li...>
> > https://lists.sourceforge.net/lists/listinfo/pearpc-devel
> >
> Hi Cassy,
>
> Thanks for the reply. The property is described in the PPC binding for
> OpenBIOS, available at http://openbios.org/Bindings. To quote from
> section 5.1.7 - Memory management unit properties:
>
> To aid a client in "taking over" the translation mechanism and
> still
> interact with Open Firmware (via the client interface), the client
> needs to know what translations have been established by Open
> Firmware. The following standard translation shall exist within the
> package to which the "mmu" property of the /chosen package refers.
>
> "Translations"
> This property, consisting of sets of translations, defines the
> currently active translations that have been established by Open
> Firmware (e.g. using map). Each set has the following format:
> (virt size phys mode)
> Each value is encoded as with encode-int
>
>
> As I said, it seems to run pretty quickly even though it's O(n),
> however
> I've only run it on a core 2 duo and a pentium 4, neither of which are
> particularly slow (the P4 is pushing it though :) )
>
> Cheers,
>
> James
>
> -------------------------------------------------------------------------
> Check out the new SourceForge.net Marketplace.
> It's the best place to buy or sell services for
> just about anything Open Source.
> http://sourceforge.net/services/buy/index.php
> _______________________________________________
> Pearpc-devel mailing list
> Pea...@li...
> <mailto:Pea...@li...>
> https://lists.sourceforge.net/lists/listinfo/pearpc-devel
>
>
>
>
> --
> Cassy
> ------------------------------------------------------------------------
>
> -------------------------------------------------------------------------
> Check out the new SourceForge.net Marketplace.
> It's the best place to buy or sell services for
> just about anything Open Source.
> http://sourceforge.net/services/buy/index.php
> ------------------------------------------------------------------------
>
> _______________________________________________
> Pearpc-devel mailing list
> Pea...@li...
> https://lists.sourceforge.net/lists/listinfo/pearpc-devel
>
Hi,
That's a little difficult to answer, as I don't know the full history of
OpenFirmware/OpenBIOS.
What I /can/ tell you is:
(a) It's defined as a "shall" in the OpenBIOS PPC binding (as I've
already quoted).
(b) It's implemented in OpenFirmware version 3 as used by the Apple
iBook and iMacs.
Further to that I have no real idea what you're asking - sorry.
James
|