Why we use float? Long should be enough, not?

Yes, it includes the intr data. if that we need add another one hypercall :); currently if not support, the HV will return failure; it should be OK. actually, I have a re-factor idea on current

Originally, we output "Time Consumed" with RTC cycles. It's not easy to read. So, this patch convert it to milliseconds. BTW, this patch also output "Total run time" with ms instead of original

Originally, we don't format the output of analyser well. It is hard to read the result. This patch make every entry of the result align with the corresponding title to make it easier for users to

Kaige Fu (2): tools: acrnalyze: Output event time cost with milliseconds tools: acrnalyze: Make the result easier to read tools/acrntrace/scripts/irq_analyze.py | 6 ++---

Patch title are wrong. Will fix it and resend them. Sorry for the noise.

Under multiple VQ use case, it is possible to have concurrent requests. Added support to return multiple vq index from all vcpu. Signed-off-by: Ong Hock Yu <ong.hock.yu@...> ---

Originally, we don't format the output of analyser well. It is hard to read the result. This patch make every entry of the result align with the corresponding title to make it easier for users to

Originally, we output "Time Consumed" with RTC cycles. It's not easy to read. So, this patch convert it to milliseconds. BTW, this patch also output "Total run time" with ms instead of original

This patch mainly focus on: - Output "Time Consumed" and "Total run time" with milliseconds. - Format the output to make the result entry align with the title. Kaige Fu (2): HV: acrntrace:

Agenda (8/15): ACRN ACPIManagement WWTopicPresentatorStatusWW21ACRN roadmap introductionRen, JackDoneWW22Patch submission process ACRN feature list introductionWang, Hongbo Ren, jackDoneWW23Memory

DM or VM monitor can use this ioctl/hypercall to get the UOS interrupt count data, to check if it in normal status, and give a response. Signed-off-by: Minggui Cao <minggui.cao@...> ---

intr monitor thread will check UOS interrupt status, first to get intr data, and check if "interrupt storm" happen as the condition setting, if happened, it will handle it as the policy

this hypercall can be used by DM or VMM manager to monitor UOS interrupt work status; for example, to check if there is an "interrupt storm", and do some reactions according to its

add the UOS interrupt data check first. *** BLURB HERE *** Minggui Cao (2): HV: add hypercall to monitor UOS interrupt data DM: add monitor thread to check intrrupt status

Hi OH, Remove this line. This might casue a lots of logs for normal case. The corner case happens rarely. How about while (1) { vcpu = find_first_bit(ioreqs_map, dev->_ctx.max_vcpu); if (vcpu

Arindam Roy <arindam.roy@...> writes: Reviewed-by: Junjie Mao <junjie.mao@...> Please also include the scope of this patch in your commit log. I think this patch covers the const

Will fix it when PR.

xgwu: ok, will change. thx. xguw: ok, will change. thx. xgwu: change it to: native_assign_ports -> port_map_tbl native_assign_info -> native_dev_info In futurn, should refine