This is the mail archive of the
ecos-discuss@sourceware.cygnus.com
mailing list for the eCos project.
Re: [ECOS] MBX hal clock settings
- To: ecos-discuss@sourceware.cygnus.com
- Subject: Re: [ECOS] MBX hal clock settings
- From: Hugo 'NOx' Tyson <hmt@cygnus.co.ukx>
- Date: 27 Jul 1999 14:21:53 +0100
- DJ-Gateway: from newsgroup cygnus.sourceware.ecos.d
- Newsgroups: cygnus.sourceware.ecos.d
- Organization: Cygnus Solutions, Cambridge, UK
- References: <Pine.LNX.3.96.990726133034.8001A-100000@newt.home.net>
Jamie Guinan <guinan@bluebutton.com> writes:
> I was studying the MPC823 clocks, and I noticed what
> appears to be a small error.
>
> In packages/hal/powerpc/mbx/v1_2_3/src/mbx.S, the setting
> for the PLPRCR register at 50MHz uses 0x5fa in the MF
> (multiplication factor) field. Doing the math, I get,
>
> 50*1000*1000 / 0x5fa = 32,786
Interesting, I get 32680 as the nearest value here. But never mind...
> The crystal is suppoed to be 32,768 (not 86), so I belive
> the appropriate value should be 0x5f5.
>
> 50*1000*1000 / 0x5f5 = 32,768
>
> There is actually some remainder there, but with 0x5f5
> the frequency is low by 28.8Khz versus over by 135KHz
> with 0x5fa (assuming the crystal is accurate).
Yup; I agree, I've patched our sources, this will take effect as and
when... (next week for anoncvs users)
Of course, this gives us a performance hit... we were successfully
overclocking! ;-)
> (I didn't consider this a "support issue" so I just posted
> here - let me know if there is a better place to report
> bugs like this).
Small issues/typos like this, posting here works fine.
Thanks!
- Huge