On 7/12/2011 5:33 PM, Graeme Russ wrote: > Hi Reinhard, > > On 13/07/11 02:08, Reinhard Meyer wrote: >> Dear J. William Campbell, All > [snip] > >> Lets just keep the current functions udelay(us) and u32 get_timer(), the >> latter maybe without parameter. Remove all *masked() and *reset() functions > This is happening and has Wolfgang's 100% support. Patches are done, just > need to be rebased and resent > >> and lets change the timeout values used in code that also NIOS uses to >> be 20ms more than the hardware requires. It does not really matter if a >> *timeout* of 20ms is blown up to 40ms. > Can't do that - That would mean winding out common code (CFI, Network, > Serial, FPGA etc, etc) timeouts to global worst-case scenario unless you > want to add a #define and start littering the code with macro's - Urgh! Hi All, Yes, this is true. However, the time_elapsed_since routine can do this dynamically (i.e. add twice the timer resolution) . I think you had another function name (at_least) involved, but you can define time_elapsed_since as always compensating for the resolution. That will fix any resolution questions in a processor-specific way. It is either that or the ifdefs. One way or another, the resolution must be addressed. Up to now, the implicit resolution has been 1 ms, but we now know that is not general enough.
Best Regards, Bill Campbell > > Regards, > > Graeme > _______________________________________________ > U-Boot mailing list > U-Boot@lists.denx.de > http://lists.denx.de/mailman/listinfo/u-boot > > _______________________________________________ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot