[ECOS] ARM HAL CYGHWR_HAL_ROM_VADDR question

[ECOS] ARM HAL CYGHWR_HAL_ROM_VADDR question

[Mike Sweeney]

Can someone please explain to me what CYGHWR_HAL_ROM_VADDR is used for
in the ARM HAL? I see the comment about it in vectors.S but it makes
no sense to me. It further confuses me because it affects the
UNMAPPED_PTR macro which is only used when setting up the reset vector
and not the other exception vectors. Can someone shed some light on
what problem this solves?

-- 
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss

Re: [ECOS] ARM HAL CYGHWR_HAL_ROM_VADDR question

[Andrew Lunn]

On Sun, Jun 03, 2007 at 11:10:07AM -0400, Mike Sweeney wrote:
> Can someone please explain to me what CYGHWR_HAL_ROM_VADDR is used for
> in the ARM HAL? I see the comment about it in vectors.S but it makes
> no sense to me. It further confuses me because it affects the
> UNMAPPED_PTR macro which is only used when setting up the reset vector
> and not the other exception vectors. Can someone shed some light on
> what problem this solves?

Think about how the system works when at boot time the ROM is at
0x0. Then early in the boot sequence the ROM is remapped to a high
address and RAM is mapped to 0x0. Applications which run in ROM have
to be linked with the high address, but the first part of the boot
code, in particular the reset vector, needs to be linked for 0x0. If i
remember correctl, CYGHWR_HAL_ROM_VADDR is involved in this.

        Andrew

-- 
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss

Re: [ECOS] ARM HAL CYGHWR_HAL_ROM_VADDR question

[Mike Sweeney]

On 6/3/07, Andrew Lunn <andrew@lunn.ch> wrote:
> On Sun, Jun 03, 2007 at 11:10:07AM -0400, Mike Sweeney wrote:
> > Can someone please explain to me what CYGHWR_HAL_ROM_VADDR is used for
> > in the ARM HAL? I see the comment about it in vectors.S but it makes
> > no sense to me. It further confuses me because it affects the
> > UNMAPPED_PTR macro which is only used when setting up the reset vector
> > and not the other exception vectors. Can someone shed some light on
> > what problem this solves?
>
> Think about how the system works when at boot time the ROM is at
> 0x0. Then early in the boot sequence the ROM is remapped to a high
> address and RAM is mapped to 0x0. Applications which run in ROM have
> to be linked with the high address, but the first part of the boot
> code, in particular the reset vector, needs to be linked for 0x0. If i
> remember correctl, CYGHWR_HAL_ROM_VADDR is involved in this.
>
>         Andrew

Thanks for the reply. I figured that's what it is there for, but I
don't understand based on how it is being used in the existing HAL
code how it actually works to fix that issue. I don't understand how
the UNMAPPED macro trickery works. An explantation of an existing
platform that uses this would be great.

-- 
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss