This is the mail archive of the ecos-patches@sources.redhat.com mailing list for the eCos project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

RE: [ECOS] Samsung SNDS100 Eval. Board...


Jonathan,

Your change to hal_platform_setup.h seems fine.  I have made some additional
changes to that file (for the SNDS100 Rev 1.2 board) to get the SDRAM memory
configuration right.  I have attached a cvs diff (with your change too) for
reference.  I am still testing these additional changes.

These additional changes are (probably) specific to my SNDS board, and
perhaps others.  It really depends on what memory chips are installed on the
board (SRAM, DRAM, SDRAM), and on how they are jumpered.  My board has two
Samsung K4S641632D TC1L SDRAMs (16MB) installed jumpered to DRAMCON0 (jumper
block SDRAM position 0).  Other boards may be configured differently.  Of
note is that the previous (incorrect) configuration "appeared" to work.
This is probably because it did map the first 4MB of SDRAM, so if the code
only used the first 4MB of RAM, then no problem.

I have added a flash driver for the SNDS100 board (still testing), that
exposed the RAM configuration problem.  This flash driver is for the two
Winbond W29EE011 128Kx8 flash chips, configured as a 256KB 16-bit wide flash
array.  The flash driver also requires some additional changes to the
mlt_arm_snds_ram.* files (still testing) to define the ROM section.

Jay

-----Original Message-----
From: Jonathan Larmour [mailto:jifl@eCosCentric.com]
Sent: Friday, October 18, 2002 10:05 PM
To: eCos Patches List
Cc: Jay Foster
Subject: Re: [ECOS] Samsung SNDS100 Eval. Board...


Hi guys,

I'd like confirmation the attached patch is okay. It's to fix the below 
problem, which also affects the (similar) e7t.

Just not 100% sure if there's some other more appropriate conditional. 
Ordinarily I'd just go ahead, but I've got brain fade right now.

Jifl

Jay Foster wrote:
> 
> Another thing to check is if the RAM and ROM configuration registers are
> getting configured correctly.  Since I'm using GDB to load my RAM version
of
> the code, I do this with a .gdbinit file before loading the code.  The
> startup code for the ROM version must do this itself.  I noticed that the
> code that does this for the SNDS board only gets included if the GDB stubs
> are compiled in (see the PLATFORM_SETUP1 definition in the
> hal_platform_setup.h file for the snds platform). In fact, I had to
disable
> this code to get the RAM version of the RedBoot package to build (this
code
> can only be executed from ROM, and should only be included in the ROM
> versions).


-- 
eCosCentric       http://www.eCosCentric.com/       <info@eCosCentric.com>
--[ "You can complain because roses have thorns, or you ]--
--[  can rejoice because thorns have roses." -Lincoln   ]-- Opinions==mine

Attachment: snds100.platformsetup.pat
Description: Binary data


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]