On 09/24/2012 02:06:55 PM, Tom Rini wrote:
On Mon, Sep 24, 2012 at 01:50:07PM -0500, Scott Wood wrote:
On 09/21/2012 07:13:24 PM, Tom Rini wrote:
On Fri, Sep 21, 2012 at 07:01:10PM -0500, Scott Wood wrote:
This is based on a merge of top-of-tree plus the 85xx tree and
the spl
framework branch (it doesn't use the SPL framework code because
of size
constraints, but I wanted to make sure there were no conflicts
with it).
Tested on P2020RDB-PC_NAND and P1021RDB-PC_36BIT_NAND.
How close (or far) is this from using the framework due to size? And I will give the whole series a proper read and comment Monday.
Configuring for P2020RDB-PC_NAND - Board: p1_p2_rdb_pc, Options: P2020RDB,NAND text data bss dec hex filename 393179 31912 267088 692179 a8fd3 /tmp/u-boot//u-boot 3612 388 0 4000 fa0
/tmp/u-boot//spl/u-boot-spl
--------------------- SUMMARY ---------------------------- Boards compiled: 1
Size limit is 4096 bytes.
Right. But that's with your changes yes?
Yes, pretty much the same as with the old nand_spl.
Do you have the how-much for using the common framework was? Or just going (and I agree, it won't fit today) by the 96 bytes to spare that a new framework won't fit?
The latter. When I get to a board with IFC (which has an 8K limit), such as p1010rdb, I'll give it a try.
-Scott