Re: MTIME update frequency
On Fri, Nov 19, 2021 at 02:32:45PM -0600, Vedvyas Shanbhogue wrote:
On Fri, Nov 19, 2021 at 03:27:40PM -0500, Jonathan Behrens wrote:The ISA specification also says:On Fri, Nov 19, 2021 at 3:02 PM Ved Shanbhogue <ved@...> wrote:This states synchronized to one tick of the real-time clock. It does notOn Fri, Nov 19, 2021 at 12:42:13PM -0500, Jonathan Behrens wrote:The requirement seems to come from the Zicntr extension described in theOn Fri, Nov 19, 2021 at 12:27 PM Greg Favor <gfavor@...>wrote:toThe current specification says that the MTIME values seen by two HARTsincrement inbe not apart by more than 1. It does say the MTIME must alwaysoneunits of 1. I do not think the specification mandates incrementing by- canon each clock tick. Presently it says the tick - the update frequencyis 10be 100 MHz or 10 MHz or somewhere in between. If the update frequencyfrequencyMHz then the MTIME increment per clock must be 10. If the updatethatis 100 Mhz then MTIME increment per clock is 1. So is your concern isdiffersI think the argument is that you technically violate the ISA spec if youan adder that adds 10 to MTIME per clock tick is hard?
The execution environment should provide a means of determining the
period of the real-time counter (seconds/tick).
If the period of the counter is in units of seconds/tick, then 1 count (1
LSB) is 1 tick.