From: Hicks, Jamey (Jamey.Hicks@hp.com)
Date: Tue Jul 30 2002 - 15:19:31 EDT
This proposal sounds good to me, for all the registers whose behavior is the same across the various implementations.
I'm CC's the bootldr list so we'll have an archive and so other bootldr hackers will know what is going on.
Thanks,
jamey
> -----Original Message-----
> From: Carl Worth [mailto:cworth@east.isi.edu]
> Sent: Tuesday, July 30, 2002 2:25 PM
> To: Vaibhav Bhandari
> Cc: Hicks, Jamey; cworth@east.isi.edu
> Subject: boot-arm.s etc
>
>
> The essential idea was to basically rename boot-sa1100.s to
> boot-arm.s. Similarly, sa1100.h would be split into two files: arm.h
> and sa1100.h, where arm.h contains any definitions that are still
> valid for the PXA250. Then the only new file would be pxa250.h which
> would define the new register mappings.
>
> Minimal machinery is necessary to support this, corrected slightly
> from what Vaibhav had in his last email:
>
> <Makefile>
> ifeq ($(CONFIG_ARCH), h3900)
> ARCH_DEFS += -DCONFIG_PXA250
>
> <boot-arm.s>
> #include "arm.h"
>
> #ifdef CONFIG_PXA250
> # include "pxa250.h"
> #else
> # include "sa1100.h"
> #endif
>
> > the registers in the code will go as REGNAME_REG instead of
> > <ARCH>_RERGNAME_REG
>
> That would be a fairly automatic modification to both the .h and .s
> files.
>
> We considered doing something like the following somewhere along the
> way:
>
> <sa1100.h>
> #define SA1100_REGNAME_REG <value>
>
> <pxa250.h>
> #define PXA250_REGNAME_REG <other-value>
>
> <arm.h>
> #ifdef CONFIG_PXA250
> # define REGNAME_REG PXA250_REGNAME_REG
> #else
> # define REGNAME_REG SA100_REGNAME_REG
> #else
>
> But that makes it more work to add a new architecture. Simpler would
> be:
>
> <sa1100.h>
> #define REGNAME_REG <value>
>
> <pxa250.h>
> #define REGNAME_REG <other-value>
>
> Comments?
>
> -Carl
>
> --
> Carl Worth
> USC Information Sciences Institute cworth@east.isi.edu
> 3811 N. Fairfax Dr. #200, Arlington VA 22203 703-812-3725
>
This archive was generated by hypermail 2.1.5 : Fri Jan 17 2003 - 17:47:04 EST