|
[PATCH 0/7] Use NMI to nofity vCPUs with lapic-pt
ACRN hypervisor needs to kick vCPU off VMX non-root mode to do some operations in hypervisor, such as interrupt/exception injection, EPT flush etc. For non lapic-pt vCPUs, we can use IPI to do so. But
ACRN hypervisor needs to kick vCPU off VMX non-root mode to do some operations in hypervisor, such as interrupt/exception injection, EPT flush etc. For non lapic-pt vCPUs, we can use IPI to do so. But
|
By
Kaige Fu
· #25913
·
|
|
[PATCH 4/5] acrntrace: make the -f/--frequency mandatory
Will drop this patch when PR.
Will drop this patch when PR.
|
By
Kaige Fu
· #25804
·
|
|
[PATCH 5/5] acrntrace: Add opt to specify the cpus where we should capture the data
This patch adds one new option '-a' to specify the cpus where we should capture the trace data. If the this option is not set or set with wrong optarg, we will capture the trace data of all possible c
This patch adds one new option '-a' to specify the cpus where we should capture the trace data. If the this option is not set or set with wrong optarg, we will capture the trace data of all possible c
|
By
Kaige Fu
· #25800
·
|
|
[PATCH 4/5] acrntrace: make the -f/--frequency mandatory
Different board gets different frequncy. The correct frequency is essential for correct result. So, this patch make it mandatory instead of optional with a default value. Signed-off-by: Kaige Fu <kaig
Different board gets different frequncy. The correct frequency is essential for correct result. So, this patch make it mandatory instead of optional with a default value. Signed-off-by: Kaige Fu <kaig
|
By
Kaige Fu
· #25799
·
|
|
[PATCH 3/5] acrntrace: Use correct format for total run time
It is better to output it using float instead of int. Otherwise, we will get '0 sec' for total run time if it is less than one second. Signed-off-by: Kaige Fu <kaige.fu@...> --- misc/tools/acrnt
It is better to output it using float instead of int. Otherwise, we will get '0 sec' for total run time if it is less than one second. Signed-off-by: Kaige Fu <kaige.fu@...> --- misc/tools/acrnt
|
By
Kaige Fu
· #25798
·
|
|
[PATCH 2/5] acrntrace: break when finding the matching key
The key in the list is unique. So, it is better to break the loop when find the matching key instead of traversing the whole list. Signed-off-by: Kaige Fu <kaige.fu@...> --- misc/tools/acrntrace
The key in the list is unique. So, it is better to break the loop when find the matching key instead of traversing the whole list. Signed-off-by: Kaige Fu <kaige.fu@...> --- misc/tools/acrntrace
|
By
Kaige Fu
· #25797
·
|
|
[PATCH 1/5] acrntrace: Fix the incorrect total vmexit cnt issue
Originally, we assume that the vmenter will sit at the first line of the trace file. But if the vmexit comes earlier than vmenter. The total vmexit cnt will be set as 0 by the first vmenter. This patc
Originally, we assume that the vmenter will sit at the first line of the trace file. But if the vmexit comes earlier than vmenter. The total vmexit cnt will be set as 0 by the first vmenter. This patc
|
By
Kaige Fu
· #25796
·
|
|
[PATCH 0/5] Fixes and improvement of acrntrace
This patchset mainly for: - Fix the total vmexit cnt and total run time issue. - Make the -f/--frequency mandatory to get correct result. - Add opt to specify the cpus where we should capture the data
This patchset mainly for: - Fix the total vmexit cnt and total run time issue. - Make the -f/--frequency mandatory to get correct result. - Add opt to specify the cpus where we should capture the data
|
By
Kaige Fu
· #25795
·
|
|
[PATCH] HV: correct the formatting flag of hypcall_id
hypcall_id has a type of uint64_t and should use 'llx' as formatting flag instead of '%d'. Otherwise, we will get a confusing error log when not-allowed hypercall occurs. Without this patch: [96707209
hypcall_id has a type of uint64_t and should use 'llx' as formatting flag instead of '%d'. Otherwise, we will get a confusing error log when not-allowed hypercall occurs. Without this patch: [96707209
|
By
Kaige Fu
· #25399
·
|
|
[PATCH] HV: Fix poweroff issue of hard RTVM
Hi Fengwei,
By
Kaige Fu
· #25329
·
|
|
[PATCH] HV: Fix poweroff issue of hard RTVM
Hi Fengwei,
By
Kaige Fu
· #25322
·
|
|
[PATCH] HV: Fix poweroff issue of hard RTVM
We should use INIT signal to notify the vcpu threads when powering off the hard RTVM. To achive this, we should set the vcpu->thread_obj.notify_mode as SCHED_NOTIFY_INIT. Patch (27163df9 hv: sched: ad
We should use INIT signal to notify the vcpu threads when powering off the hard RTVM. To achive this, we should set the vcpu->thread_obj.notify_mode as SCHED_NOTIFY_INIT. Patch (27163df9 hv: sched: ad
|
By
Kaige Fu
· #25297
·
|
|
[PATCH] DM: samples: Add RT related kernel boot params
Hi Fuzhong,
By
Kaige Fu
· #24473
·
|
|
[PATCH] DM: samples: Add RT related kernel boot params
This patch adds the following kernel params to gain a more deterministic results of preempt-rt linux. clocksource=tsc processor.max_cstate=0 intel_idle.max_cstate=0 intel_pstate=disabled mce=ignore_ce
This patch adds the following kernel params to gain a more deterministic results of preempt-rt linux. clocksource=tsc processor.max_cstate=0 intel_idle.max_cstate=0 intel_pstate=disabled mce=ignore_ce
|
By
Kaige Fu
· #24456
·
|
|
[PATCH] hv: vcr: check guest cr3 before loading pdptrs
By
Kaige Fu
· #24108
·
|
|
[PATCH v3] HV: Enable vART support by intercepting TSC_ADJUST MSR
Thanks.
By
Kaige Fu
· #23933
·
|
|
[PATCH v3] HV: Enable vART support by intercepting TSC_ADJUST MSR
Hi Eddie,
By
Kaige Fu
· #23931
·
|
|
[PATCH v3] HV: Enable vART support by intercepting TSC_ADJUST MSR
Hi Eddie,
By
Kaige Fu
· #23928
·
|
|
[PATCH v3] HV: Enable vART support by intercepting TSC_ADJUST MSR
The policy of vART is that software in native can run in VM too. And in native side, the relationship between the ART hardware and TSC is: pTSC = (pART * M) / N + pAdjust The vART solution is: - Prese
The policy of vART is that software in native can run in VM too. And in native side, the relationship between the ART hardware and TSC is: pTSC = (pART * M) / N + pAdjust The vART solution is: - Prese
|
By
Kaige Fu
· #23923
·
|
|
[PATCH v2] HV: Enable vART support by intercepting TSC_ADJUST MSR
By
Kaige Fu
· #23905
·
|