All work
- Finalize upstream of lib thermal in kernel tools directoryPERF-61Resolved issue: PERF-61Daniel Lezcano
- Prepare and sent latency nice v2 based on community feedbackPERF-60Resolved issue: PERF-60Vincent Guittot
- submit v1 libthermal in kernel toolsPERF-59Resolved issue: PERF-59Daniel Lezcano
- study where to merge libthermal: kernel tools directory or other placePERF-58Resolved issue: PERF-58Daniel Lezcano
- Latency nice 1st submission on lkmlPERF-57Resolved issue: PERF-57
- Take into account IPIs : 1st submissionPERF-56Resolved issue: PERF-56Ulf Hansson
- Start discussing with android SoC/OEM/ODM about the generic thermal hal in AOSPPERF-55Resolved issue: PERF-55Daniel Lezcano
- Capture Temp behavior on PocophonePERF-54Resolved issue: PERF-54Daniel Lezcano
- Minimize latency impact from ktime_get() in genpd - first submissionPERF-53Resolved issue: PERF-53Ulf Hansson
- Restart latency nice discussion on mailing listPERF-52Resolved issue: PERF-52Vincent Guittot
- Setup a test that highlights the benefit of latency nice and share results with interested membersPERF-51Resolved issue: PERF-51Vincent Guittot
- Evaluate Display pipeline LatencyPERF-50Resolved issue: PERF-50
- Add support for hierarchical multiple idle states to genpd - first submissionPERF-49Resolved issue: PERF-49Ulf Hansson
- Rebase and wait to be merged devfreq DTPMPERF-48Resolved issue: PERF-48Daniel Lezcano
- Ready to be merged DTPM hierarchy descriptionPERF-47Resolved issue: PERF-47Daniel Lezcano
- Finalize Audio latency evaluationPERF-46Resolved issue: PERF-46Paolo Valente
- Reproduce the hierarchical locking bug in genpdPERF-45Resolved issue: PERF-45Ulf Hansson
- Prepare 1st PoC of latency nice featurePERF-44Resolved issue: PERF-44Vincent Guittot
- Investigate support for CPUIdle and suspend-to-idle in OP-TEE for stm32mp1PERF-43Resolved issue: PERF-43Gabriel Fernandez
- Scheduler Latency and interactive taskPERF-42Resolved issue: PERF-42Vincent Guittot
- Android Power HAL updatePERF-41Resolved issue: PERF-41Daniel Lezcano
- Android Thermal HALPERF-40Resolved issue: PERF-40Daniel Lezcano
- Improve the idlestate selection in the genpd governorPERF-39Resolved issue: PERF-39Ulf Hansson
- Minimize latency impact with runtime PM for CPU devicesPERF-38Resolved issue: PERF-38Ulf Hansson
- Minimize latency impact from ktime_get() in genpdPERF-37Resolved issue: PERF-37Ulf Hansson
- Support PSCI OS-initiated mode in TF-APERF-36Resolved issue: PERF-36Ulf Hansson
- Energy and performance optimizations for cluster idlingPERF-35Resolved issue: PERF-35Ulf Hansson
- Add support for hierarchical multiple idle states to genpdPERF-34Resolved issue: PERF-34Ulf Hansson
- Temperature and power monitoring daemonPERF-33Resolved issue: PERF-33Daniel Lezcano
- Android thermal HAL connection with new thermal notification mechanismPERF-32Resolved issue: PERF-32Daniel Lezcano
- Add missing features in Deadline CPU schedulerPERF-31Resolved issue: PERF-31Paolo Valente
- Evaluate the current feature status of Deadline CPU schedulerPERF-30Resolved issue: PERF-30Paolo Valente
- DTPM devfreq power vs loadPERF-29Resolved issue: PERF-29Daniel Lezcano
- DTPM CPU power against loadPERF-28Resolved issue: PERF-28Daniel Lezcano
- DTPM energy model based devfreqPERF-27Resolved issue: PERF-27Daniel Lezcano
- DTPM kernel power rebalancingPERF-26Resolved issue: PERF-26Daniel Lezcano
- DTPM energy model based CPUPERF-25Resolved issue: PERF-25Daniel Lezcano
- DTPM generic frameworkPERF-24Resolved issue: PERF-24Daniel Lezcano
- Enable PSCI OSI in TFAPERF-23Resolved issue: PERF-23Gabriel Fernandez
- Address requirements from GKI 2.0 for cluster idlingPERF-22Resolved issue: PERF-22Ulf Hansson
- Optimize tasks placement on NUMA systemPERF-14Resolved issue: PERF-14Vincent Guittot
- Evaluate and optimize interaction with other systemsPERF-13Resolved issue: PERF-13
- Enabling new scheduler features on Arm serverPERF-12Resolved issue: PERF-12Viresh Kumar
- Load balance improvementPERF-6Resolved issue: PERF-6Vincent Guittot
- prototype sched_idle usage for androidPERF-7Resolved issue: PERF-7Viresh Kumar
- Fixup the issue with the hierarchical locking mechanism in genpdPERF-11Resolved issue: PERF-11Ulf Hansson
- profile and improve performance of scheduler for all type of Arm systemPERF-5Resolved issue: PERF-5Vincent Guittot
- Get next wake up event of cluster/power domainPERF-8Resolved issue: PERF-8Daniel Lezcano
- Bus scaling enhancementsPERF-4Resolved issue: PERF-4Georgi Djakov
- Using utilization information in cooling devicesPERF-3Resolved issue: PERF-3Viresh Kumar
MBed TLS running slow on Open CI
Description
Environment
Engineering Progress Update
Attachments
is a dependency for
Details
Details
Assignee
Reporter
Upstream
Share Visibility
Time tracking
Priority
Checklist
Checklist
Sentry
Sentry
Activity
Glen Valante May 19, 2023 at 12:54 PM
To date we are seeing reasonable performance. We are closing this issue .
Glen Valante April 5, 2023 at 1:12 AM
Update:
We have adjusted the size of the executors to: T3.medium/T3.large/T3.medium. The Master is still C6a.4xlarge. We are getting consistent numbers and will monitor.
After TF-M/TF-A releases in April, we will look for an opportunity to migrate the tests back to Production from Stage.
Moving the issue to blocked while we monitor and wait till next month when we migrate it back to production.
Arthur She March 29, 2023 at 6:20 PM
The ARM CC license server issue has been resolved by moving the flexnet back to scaleway from AWS.
Below is 's explanation
Currently: I have moved flexnet back to the old server at scaleway in order to get the tests running again.
The problem: When we setup DNS for 'flexnet.tf.o' to point to the public IP address, traffic from the mbed-tls instances and ci.staging are being routed outside of the AWS network and thus VPC security group and back in through the public IP's interface. This means that traffic from any AWS assets is no longer granted the privileges of the security group and is subject to the same firewall policies as traffic from the rest of the internet... so it gets blocked.
If you attempt to reach the flexnet server by its internal AWS IP address from another AWS instance, it works just fine with our existing rules.
The solution: We need to find a way to keep AWS instances from using the public IP interface for the server and ensure they get routed through the private IP. Something similar to putting an /etc/hosts in to preempt the DNS settings.
Since I'm not sure what the right way to do this is, I decided to fall back to the old flexnet server since we know it's operational. Arthur ran a test and it passed, so this should hold us for now.
Arthur She March 24, 2023 at 3:33 AM
Bence merged a patch that resolved the performance issue on the OpenCI.
And we’re having ARM CC license server issue, all_armcc-build_armcc test failed on all builds. The log showed “Unable to connect to the license server"
The MbedTLS team is noticing that jobs are running extremely slow, examples:
https://ci.trustedfirmware.org/view/TF-M/job/tf-m-nightly/922/
No indication of why this is happening.