Good day to all,
From an architecture point of view is it fair to assume there can be
only one timestamp generator per system or there can be more, something like one per CPU cluster? Either way, is the current practice expected to continue or there are plans to move to different design standards?
Thanks, Mathieu
Hi Mathieu,
Yes - only one timestamp generator per Coresight system is a fair assumption.
From A profile v8.4 onwards, any system that includes the self hosted
trace extensions can source the timestamp value from a counter in the core - which will enable it to be accurately correlated with system time. (see ARM Architecture Reference Manual for Armv8-A architecture profile.ARM DDI 0487F.b, section D3, AArch64 self hosted trace.)
Regards
Mike
On Tue, 7 Jul 2020 at 18:56, Mathieu Poirier mathieu.poirier@linaro.org wrote:
Good day to all,
From an architecture point of view is it fair to assume there can be only one timestamp generator per system or there can be more, something like one per CPU cluster? Either way, is the current practice expected to continue or there are plans to move to different design standards?
Thanks, Mathieu _______________________________________________ CoreSight mailing list CoreSight@lists.linaro.org https://lists.linaro.org/mailman/listinfo/coresight
On Wed, 8 Jul 2020 at 08:45, Mike Leach mike.leach@linaro.org wrote:
Hi Mathieu,
Yes - only one timestamp generator per Coresight system is a fair assumption.
Very well.
From A profile v8.4 onwards, any system that includes the self hosted trace extensions can source the timestamp value from a counter in the core - which will enable it to be accurately correlated with system time.
That is really good news.
Thanks for looking into this, Mathieu
(see ARM Architecture Reference Manual for Armv8-A architecture profile.ARM DDI 0487F.b, section D3, AArch64 self hosted trace.)
Regards
Mike
On Tue, 7 Jul 2020 at 18:56, Mathieu Poirier mathieu.poirier@linaro.org wrote:
Good day to all,
From an architecture point of view is it fair to assume there can be only one timestamp generator per system or there can be more, something like one per CPU cluster? Either way, is the current practice expected to continue or there are plans to move to different design standards?
Thanks, Mathieu _______________________________________________ CoreSight mailing list CoreSight@lists.linaro.org https://lists.linaro.org/mailman/listinfo/coresight
-- Mike Leach Principal Engineer, ARM Ltd. Manchester Design Centre. UK
-
Mathieu Poirier -
Mike Leach