1- I dont think we should be worried about names that are too long. I'd much rather have a long name than a non-descriptive one (or in this case a totally misleading one)
2- Changing the attribute shouldn't be that big of a deal.
3 - dimm_mask isn't exposed to the mainboard layer which is where we need it.


As for 'why we need it', we have a post card which displays the DIMM population map on boot for our ops guys...



On 10/20/05, yhlu <yinghailu@gmail.com> wrote:
1. that would be too long, do we need DIMM_SOCKETS_PER_CHANNEL_PER_NODE?
2. that is controller array is "static const", you need to change the attribute so you can change the value....that would break the code for ROMCC....
dimm_mask already did popppp for you.
 
for rev f.. I have made one CAR only raminit.c, it keep the dimmask and needed dimm spd info in struct mem_info...
So don't need to read pci_conf or smbus to get info....
it include dimm_mask, x4_mask, x16_mask, ecc_mask, registered_mask, is_Width128, is_opteron.....
 
YH
 
On 10/20/05, San Mehat < san@google.com> wrote:
Greets,

I've made some modifications locally here to the amdk8 memory code and wanted to just chuck them out there to see what you guys think:

1 - Renamed DIMM_SOCKETS define to DIMM_SOCKETS_PER_CHANNEL since thats what it actually is.
2 - Added 'channel0_pop' and 'channel1_pop' to 'struct mem_controller' and modified 'raminit.c' to fill this in with the physical DIMM population map.

For us knowing the detected DIMM population map *at the mainboard level* is very usefull.


Fire away ;)

-San


--
LinuxBIOS mailing list
LinuxBIOS@openbios.org
http://www.openbios.org/mailman/listinfo/linuxbios