From: Koen Martens <[email protected]>
Changes are:
- Added s3c2412-specific files to arch/arm/mach-s3c2410
- Added s3c2412 detection to arch/arm/mach-s3c2410/cpu.c
- Added CONFIG_CPU_S3C2412
- Added s3c2412 specific registers and register addresses to
various regs-*.h files in include/asm-arm/arch-s3c2410
All changes are preliminary, final documentation is not yet available
from samsung. We did test on actual hardware, but outside the linux
kernel (due to limited number of actual chips we have available and
lack of proper PCB with serial lines exported).
Signed-off-by: Koen Martens <[email protected]>
Have fun,
Koen Martens
--
K.F.J. Martens, Sonologic, http://www.sonologic.nl/
Networking, hosting, embedded systems, unix, artificial intelligence.
Public PGP key: http://www.metro.cx/pubkey-gmc.asc
Wondering about the funny attachment your mail program
can't read? Visit http://www.openpgp.org/
On Tue, Feb 28, 2006 at 03:18:00PM +0100, Koen Martens wrote:
> From: Koen Martens <[email protected]>
>
> Changes are:
> - Added s3c2412-specific files to arch/arm/mach-s3c2410
> - Added s3c2412 detection to arch/arm/mach-s3c2410/cpu.c
> - Added CONFIG_CPU_S3C2412
> - Added s3c2412 specific registers and register addresses to
> various regs-*.h files in include/asm-arm/arch-s3c2410
It would help to split these emails into smaller chunks than all the files
so that comments could be with the chunks that it goes with.
Kconfig.diff - fine, no problems here
Makefile.diff - going to have to remove those #s at somepoint
regs-clock.diff - didn't need to add s3c2412_get_pll, same as s3c2410_get_pll
regs-gpio.diff - group the s3c2412 changes together into larger #ifdefs
regs-irq.diff - looks fine
regs-rtc.diff - two sets of changes here? 2410 and 2412 additions
s3c2412_c.diff - looks fine
s3c2412_h.diff - looks fine
cpu.diff:
the 2412 is an ARM926EJS, so I'd expect the core-ID accessed
via the co-processor registers to be diffeernt, so this could give
the indication of what sort of CPU to try.
this also seems to have the S3C2442 ID patch from the patch-q in it,
please be careful about merging patches like that.
> All changes are preliminary, final documentation is not yet available
> from samsung. We did test on actual hardware, but outside the linux
> kernel (due to limited number of actual chips we have available and
> lack of proper PCB with serial lines exported).
[snip]
--
Ben
Q: What's a light-year?
A: One-third less calories than a regular year.
Ben Dooks wrote:
>It would help to split these emails into smaller chunks than all the files
>so that comments could be with the chunks that it goes with.
>
>
All right, i'll repost in a minute..
>Makefile.diff - going to have to remove those #s at somepoint
>
>
True, but i can't submit those as long as i use tomtom specific cpu-type
registers (see reply to other post later today).
>regs-clock.diff - didn't need to add s3c2412_get_pll, same as s3c2410_get_pll
>
>
Hmm, i disagree.. The term (mdiv+8) is multiplied by 2 for s3c2412, this
is not the case for s3c2410..
I changed it to use the 64 bit arithmetic like s3c2410_get_pll and
you'll notice the difference is this line:
+ fvco = (uint64_t)baseclk * ((mdiv + 8)<<1);
Maybe it is better to use one get_pll and add an extra parameter to
indicate the multiplication of (mdiv+8) ?
>regs-gpio.diff - group the s3c2412 changes together into larger #ifdefs
>
>
Done.
>regs-rtc.diff - two sets of changes here? 2410 and 2412 additions
>
>
Ok, cleaned that up..
>cpu.diff:
>
>the 2412 is an ARM926EJS, so I'd expect the core-ID accessed
>via the co-processor registers to be diffeernt, so this could give
>the indication of what sort of CPU to try.
>
>this also seems to have the S3C2442 ID patch from the patch-q in it,
>please be careful about merging patches like that.
>
>
Ok, cleaned that up a bit too.. As for the core-ID remark, i will try
and have a look at it today, but as it is my last day at this firm today
i can't make any guarantees. Maybe Dimitry Andric will pick this up when
i'm gone..
Best,
Koen
--
K.F.J. Martens, Sonologic, http://www.sonologic.nl/
Networking, hosting, embedded systems, unix, artificial intelligence.
Public PGP key: http://www.metro.cx/pubkey-gmc.asc
Wondering about the funny attachment your mail program
can't read? Visit http://www.openpgp.org/