Hi Simon,
On 31.08.22 19:44, Simon Glass wrote:
<snip>
It is needed because bootstage is called before driver model is ready. In fact it can be used to time driver model things.
I see, makes sense. This brings up my next questions though, why isn't CONFIG_TIMER_EARLY enough in this case? AFAICT it's targeted exactly for this early (pre DM) bootstage. From drivers/timer/Kconfig:
config TIMER_EARLY bool "Allow timer to be used early in U-Boot" depends on TIMER # initr_bootstage() requires a timer and is called before initr_dm() # so only the early timer is available default y if X86 && BOOTSTAGE help In some cases the timer must be accessible before driver model is active. Examples include when using CONFIG_TRACE to trace U-Boot's execution before driver model is set up. Enable this option to use an early timer. These functions must be supported by your timer driver: timer_early_get_count() and timer_early_get_rate().
So again, do we really need timer_get_boot_us() or isn't it enough to select TIMER_EARLY when BOOTSTAGE is enabled?
The timer is for milliseconds but for bootstage we need microseconds.
Perhaps the ultimate solution here is to support a microsecond timer through the TIMER api and use the TIMER_EARLY thing to provide timer_get_boot_us(), perhaps renaming to timer_early_get_us() ?
Yes, sounds like a plan. We should consolidate these implementations. Let me think about it and perhaps do some basic implementations and tests for a while.
Thanks, Stefan