On 18.03.2008 22:17, Ward Vandewege wrote:
On Tue, Mar 18, 2008 at 10:10:41PM +0100, Ronald Hoogenboom wrote:
On Mon, 2008-03-17 at 16:45 -0400, Ward Vandewege wrote:
... except for those of us who've soldered a socket onto a rev1 board of course! I often use larger chips on a v1.
And, I presume the 512K limit doesn't apply for them...
That's correct. It works at least with 1MB chips; I have not tried larger.
So, this should really be spi-only.
Yes, but how? Having a separate V2-board target afterall? Or Config.lb files specific for V1/V2 boards, where the V1 uses rom_stream (CONFIG_ROM_PAYLOAD=1) and V2 uses piorom_stream (CONFIG_PIOROM_PAYLOAD=1)? Which is the least ugly....
Hrm. I'd like to avoid splitting up the targets, but perhaps we have to. Carl-Daniel, thoughts?
To be honest, most of this probably could be worked around a bit easier with less code changes if decompression wasn't changed at all. Is there any reason we can't copy (in the sense of "read byte from SPI, copy it somewhere") the ROM contents to a given location in RAM, then decompress it with normal lzma functions? That would reduce code differences to copy/nocopy and decompression source address. As a bonus, it would speed up booting a lot in the SPI case.
Regards, Carl-Daniel