This is the mail archive of the
ecos-discuss@sources.redhat.com
mailing list for the eCos project.
Re: sdram size
- From: Mark Salter <msalter at redhat dot com>
- To: carl dot huang at mic dot com dot tw
- Cc: souto at contronics dot com, ecos-discuss at sources dot redhat dot com
- Date: Wed, 22 May 2002 08:08:53 -0400
- Subject: Re: [ECOS] sdram size
- References: <A56A8E5E02CBD511881800306E002CCA04F026@mrspdc.mrs.com.cn>
>>>>> CARL HUANG writes:
> Hello, Jose:.
> Thanks for yor reply. That's just what I want.
> I read "PC SDRAM Serial Presence Detect (SPD) Specification "and almost know
> the code.
> But I still have a question: The slave address for SDRAM is oxA2 (from
> hal_platform_setup.h). Why define this address to oxA2?
Actually, the comment in the code is wrong. The actual SDRAM_DEVID
value is 0xAE. The address is derived from fixed bits in hw and
the setting of switch S1D1. The default setting of S1D1 results in
an address of 0xAE for the SDRAM SPD. The code won't work for any
other settings of S1D1 as there is no way for the code to read the
switch settings. I have no idea why the switch is provided.
> The slave address
> for I2c is 0x02, why define this address to 0x02?
This is arbitrary and not necessary. The hal_platform_setup code does
not use the I2C controller in slave mode, so this address never comes
into play.
--Mark
--
Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos
and search the list archive: http://sources.redhat.com/ml/ecos-discuss