RE: S2885 -- spare 4s in rebooting

[YhLu]

Spare 4s for rebooting

I add some comments in s2885/auto.c.

In auto.c call soft2_reset directly instead of jmp to __cpu_reset

#if 0
                        asm volatile ("jmp __cpu_reset");
#else 
                /* cpu reset also reset the memtroller ????
                        need soft_reset to reset all except keep HT link
freq and width */
                /* So we don't need to 
                        1. jmp to __cpu_reset
                        2. jmp to __main to copy ROM to ram (It costs some
time) 
                        3. call hardwaremain(), it will according to
boot_complete to issue hard_reset in southbridge. 
                                (Actually it is soft2_reset(); --- without
call hard_reset, the memory is corrupted.
                  We will call soft2_reset directly to spare time in 1 and 2
and 3.2
                */
                        distinguish_cpu_resets();
                        soft2_reset();
#endif

S2885 -- spare 4s in rebooting

YhLu  writes:
> Spare 4s for rebooting
>
> I add some comments in s2885/auto.c.
>
> In auto.c call soft2_reset directly instead of jmp to __cpu_reset

Hmm. That does seem to have some merit.  Where are you seeing the code
run slowly?  In particular I am wondering if this is an mtrr setup
problem or something else you are avoiding.

Eric