3 comments:
File src/soc/amd/picasso/Makefile.inc:
Patch Set #4, Line 213: PSP_BIOSBIN_DEST=$(shell printf "%x" $(call int-subtract, $(call int-add, $(CONFIG_X86_RESET_VECTOR) 0x10) $(PSP_BIOSBIN_SIZE)))
I believe we would want to avoid new references to C_ENV_BOOTBLOCK_SIZE and take the actual uncompre […]
Re. amdfwtool, the comment at CB:33759 picasso/Makefile.inc#211 applies here too. Re. "hardware is wired at IP=fff0", yeah I'd suspect that's true but it certainly matches the documented behavior. However, that's not my goal here. The benefit to using C_ENV_BOOTBLOCK_SIZE and X86_RESET_VECTOR+0x10-n is to have those base/size values later when I need to reserve that RAM. An alternative _might_ be to parse the PSP table instead, but I prefer not to do that work until vboot is completely implemented and functioning.
Patch Set #4, Line 219: ## #$(shell printf "%d" $(CONFIG_STACK_SIZE))
whoops
Patch Set #4, Line 301: $(PSP_BIOSBIN_FILE): $(PSP_ELF_FILE) $(AMDCOMPRESS)
I am looking at CB:33401 where util/cbfstool/amdcompress. […]
Ack
It had started out in a brand new amd utility. I don't recall the reasoning, now, behind the request to put it in cbfstool. I'm not opposed to moving it to amdfwtool but we still have the challenge of how to put the biosbin outside of amdfw.rom (as mentioned elsewhere).
To view, visit change 37490. To unsubscribe, or for help writing mail filters, visit settings.