at91sam9g20: Upcoming Patches

[Greg Ansley]

One big patch it is, it is easier for me as well.

I will send it after creating the patch against today's head and testing 
it on both boards (late today, probably).

I found that I can use a generic U-boot and configure the pins in the 
"bootcmd" and then the kernel can ignore the board specific issues. 
Obviously the board_xxx.c files are the way to go for a production build 
but I would really like to make it easy for people to try out FreeBSD on 
lots of boards without a lot of effort.

I intended to write a up a quick wiki with some examples and links to 
binaries to help get people started...

Greg

On 7/19/10 10:16 AM, Warner Losh wrote:
> I'd prefer one big patch, assuming it is almost ready to roll into the 
> tree. Chances ate I'll be committing it. Bit there might also be a 
> second effort that has patches in flight too.  There was some talk too 
> of investgating fdt to cope with the different families.
>
> Btw, when I was doing the original port I found the multiplexed pins 
> were rarely configured by the wide array of boot loaders...  You were 
> lucky if the memory and nic pins were configured.
>
> Warner
>
>
> On Jul 19, 2010, at 7:09 AM, Greg Ansley <gja at ansley.com> wrote:
>
>> During the next day or so I'll be submitting the following patches 
>> for the arm/at91 (head) tree with the goal of eventually being able 
>> to create a single "GENERIC" kernel that will boot on the at91rm9200, 
>> at91sam926? and at91sam9g20 (as long as boot sequence configures the 
>> multiplexed pins).
>>
>> I also have patches against 8-current if any one needs them.
>>
>> These patches have been tested on a KB9202C and an Atmel 
>> AT91SAM9G20EK Rev B.  I would appreciate testing (or loan of boards) 
>> for the at91sam926? family.
>>
>> I am assuming it would be best to submit separate PRs for each of the 
>> changes. If something else is preferred let me know.
>>
>> at91_mci.c:
>>
>>   * Compile and run time selection of "Slot B" of mci controller.
>>   * Compile and run time selection of 4 wire connection (4x bandwidth
>>     to slot).
>>   * No byte swapping on non-RM9200 targets.
>>   * Fix obscure segfault when bus_dma_load() fails in
>>     at91_mci_start_cmd().
>>
>> at91_pit.c:
>>
>>   * Support for full resolution timer counter.
>>   * Remove compile time address dependencies.
>>
>> at91_rst.c:  (NEW)
>>
>>   * Support SAM9 reset controller.
>>   * Implement shutdown() on NRST (reset button) and full reset if held
>>     for more the 5 seconds.
>>
>> at91_wdt.c: (NEW)
>>
>>   * Support SAM9 watchdog timer using the watchdog(9) style interface.
>>
>> if_ate.c:
>>
>>   * Support for sam9 "EMAC" controller.
>>   * Support for rmii interface to phy.
>>
>> at91.c & at91sam9.c:
>>
>>   * Eliminate separate at91sam9.c file.
>>   * Add new devices to at91sam9_devs table.
>>
>> at91_machdep.c & at at91sam9_machdep.c:
>>
>>   * Automatic chip type determination.
>>   * Remove compile time chip dependencies.
>>   * Eliminate separate at91sam9_machdep.c file.
>>
>> at91_pmc.c:
>>
>>   * Corrected support for all of the sam926? and sam9g20 chips.
>>   * Remove compile time chip dependencies.
>>
>> Greg Ansley
>>
>> _______________________________________________
>> freebsd-arm at freebsd.org mailing list
>> http://lists.freebsd.org/mailman/listinfo/freebsd-arm
>> To unsubscribe, send any mail to "freebsd-arm-unsubscribe at freebsd.org"
>>
>>