Date: Tue, 25 Aug 2015 18:14:34 -0700 From: John-Mark Gurney <jmg@funkthat.com> To: Justin Hibbits <jrh29@alumni.cwru.edu> Cc: "freebsd-arch@freebsd.org" <freebsd-arch@freebsd.org> Subject: Re: Devices with 36-bit paddr on 32-bit system Message-ID: <20150826011434.GO33167@funkthat.com> In-Reply-To: <CAHSQbTDsvB32%2BLyzHJO78VwUwAfUTMOUQp13BMCUpapSMT0fbg@mail.gmail.com> References: <CAHSQbTDsvB32%2BLyzHJO78VwUwAfUTMOUQp13BMCUpapSMT0fbg@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Justin Hibbits wrote this message on Mon, Aug 24, 2015 at 23:44 -0700:
> With my work porting FreeBSD to PowerPC e500mc and e5500, I have
> devices in my device tree mapped well above the 4GB mark
> (0xffexxxxxx), and have no idea how to properly address them for
> resources in rman. Do we already have a solution to support this?
> Part of the problem is the powerpc nexus does a straight convert to
> vm_offset_t of rman_get_start() (itself returning a u_long), and
> vm_offset_t is not necessarily equal to vm_paddr_t (on Book-E powerpc
> vm_offset_t is 32-bits, vm_paddr_t is 64-bits).
>
> Could rman be thought that resources aren't necessarily u_long? The
> only thought I have is to assume in the nexus code that the bottom 4
> bits of the 32-bit address are actually the top bits of the 36-bit
> address, and generate those in ofw_bus_reg_to_rl(), since the bottom
> 12 bits can be ignored anyways (pmap_mapdev() only maps full pages at
> a minimum). This seems kinda kludgy to me, though.
>
> Any thoughts?
I'd look at how i386's PAE does it, as this is exactly the same type
of issue that PAE has...
--
John-Mark Gurney Voice: +1 415 225 5579
"All that I will do, has been done, All that I have, has not."
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20150826011434.GO33167>