Performance addon operator code base move to NTO

Makefile

@@ -18,11 +18,10 @@ TUNED_COMMIT:=682c47c0a9eb5596c2d396b6d0dae4e297414c50
 TUNED_DIR:=daemon
 
 # API-related variables
-API_TYPES_DIR:=./pkg/apis/tuned/v1
-API_TYPES:=$(wildcard $(API_TYPES_DIR)/*_types.go)
+API_TYPES_DIR:=pkg/apis
+API_TYPES:=$(shell find $(API_TYPES_DIR) -name \*_types.go)
 API_ZZ_GENERATED:=zz_generated.deepcopy
-API_TYPES_GENERATED:=$(API_TYPES_DIR)/$(API_ZZ_GENERATED).go
-API_GO_HEADER_FILE:=pkg/apis/header.go.txt
+API_GO_HEADER_FILE:=$(API_TYPES_DIR)/header.go.txt
 
 # Container image-related variables
 IMAGE_BUILD_CMD=podman build --no-cache
@@ -33,6 +32,10 @@ ORG=openshift
 TAG=$(shell git rev-parse --abbrev-ref HEAD)
 IMAGE=$(REGISTRY)/$(ORG)/origin-cluster-node-tuning-operator:$(TAG)
 
+# PAO variables
+CLUSTER ?= "ci"
+PAO_CRD_APIS :=$(addprefix ./$(API_TYPES_DIR)/pao/,v2 v1 v1alpha1)
+
 all: build
 
 # Do not put any includes above the "all" target.  We want the default target to build
@@ -59,25 +62,25 @@ $(BINDATA): $(GOBINDATA_BIN) $(ASSETS)
 
 pkg/generated: $(API_TYPES)
 	$(GO) run k8s.io/code-generator/cmd/deepcopy-gen \
-	  --input-dirs $(PACKAGE)/pkg/apis/tuned/v1 \
+	  --input-dirs $(PACKAGE)/$(API_TYPES_DIR)/tuned/v1,$(PACKAGE)/$(API_TYPES_DIR)/pao/v1alpha1,$(PACKAGE)/$(API_TYPES_DIR)/pao/v1,$(PACKAGE)/$(API_TYPES_DIR)/pao/v2 \
 	  -O $(API_ZZ_GENERATED) \
 	  --go-header-file $(API_GO_HEADER_FILE) \
-	  --bounding-dirs $(PACKAGE)/pkg/apis \
+	  --bounding-dirs $(PACKAGE)/$(API_TYPES_DIR) \
 	  --output-base tmp
 	$(GO) run k8s.io/code-generator/cmd/client-gen \
 	  --clientset-name versioned \
 	  --input-base '' \
-	  --input $(PACKAGE)/pkg/apis/tuned/v1 \
+	  --input $(PACKAGE)/$(API_TYPES_DIR)/tuned/v1 \
 	  --go-header-file $(API_GO_HEADER_FILE) \
 	  --output-package $(PACKAGE)/pkg/generated/clientset \
 	  --output-base tmp
 	$(GO) run k8s.io/code-generator/cmd/lister-gen \
-	  --input-dirs $(PACKAGE)/pkg/apis/tuned/v1 \
+	  --input-dirs $(PACKAGE)/$(API_TYPES_DIR)/tuned/v1 \
 	  --go-header-file $(API_GO_HEADER_FILE) \
 	  --output-package $(PACKAGE)/pkg/generated/listers \
 	  --output-base tmp
 	$(GO) run k8s.io/code-generator/cmd/informer-gen \
-	  --input-dirs $(PACKAGE)/pkg/apis/tuned/v1 \
+	  --input-dirs $(PACKAGE)/$(API_TYPES_DIR)/tuned/v1 \
 	  --versioned-clientset-package $(PACKAGE)/pkg/generated/clientset/versioned \
 	  --listers-package $(PACKAGE)/pkg/generated/listers \
 	  --go-header-file $(API_GO_HEADER_FILE) \
@@ -86,7 +89,6 @@ pkg/generated: $(API_TYPES)
 	tar c tmp | tar x --strip-components=4
 	touch $@
 
-
 $(GOBINDATA_BIN):
 	$(GO) build -o $(GOBINDATA_BIN) ./vendor/github.com/kevinburke/go-bindata/go-bindata
 
@@ -130,7 +132,8 @@ local-image-push:
 # $2 - apis
 # $3 - manifests
 # $4 - output
-$(call add-crd-gen,tuned,$(API_TYPES_DIR),./manifests,./manifests)
+$(call add-crd-gen,tuned,./$(API_TYPES_DIR)/tuned/v1,./manifests,./manifests)
+$(call add-crd-gen,pao,$(PAO_CRD_APIS),./manifests,./manifests)
 
 # This will include additional actions on the update and verify targets to ensure that profile patches are applied
 # to manifest files
@@ -141,3 +144,29 @@ $(call add-crd-gen,tuned,$(API_TYPES_DIR),./manifests,./manifests)
 $(call add-profile-manifests,manifests,./profile-patches,./manifests)
 
 .PHONY: all build deepcopy crd-schema-gen test-e2e verify verify-gofmt clean local-image local-image-push
+
+# PAO
+
+.PHONY: cluster-deploy-pao
+cluster-deploy-pao:
+	@echo "Deploying PAO artifacts"
+	CLUSTER=$(CLUSTER) hack/deploy.sh
+
+.PHONY: cluster-label-worker-cnf
+cluster-label-worker-cnf:
+	@echo "Adding worker-cnf label to worker nodes"
+	hack/label-worker-cnf.sh
+
+.PHONY: pao-functests
+pao-functests: cluster-label-worker-cnf pao-functests-only
+
+.PHONY: pao-functests-only
+pao-functests-only:
+	@echo "Cluster Version"
+	hack/show-cluster-version.sh
+	hack/run-functests.sh
+
+.PHONY: cluster-clean-pao
+cluster-clean-pao:
+	@echo "Cleaning up performance addons artifacts"
+	hack/clean-deploy.sh

assets/pao/assets.go

@@ -0,0 +1,19 @@
+package assets
+
+import (
+	"embed"
+)
+
+var (
+	// Configs contains all files that placed under the configs directory
+	//go:embed configs
+	Configs embed.FS
+
+	// Scripts contains all files that placed under the scripts directory
+	//go:embed scripts
+	Scripts embed.FS
+
+	// Tuned contains all files that placed under the tuned directory
+	//go:embed tuned
+	Tuned embed.FS
+)

assets/pao/configs/99-low-latency-hooks.json

@@ -0,0 +1,11 @@
+{
+  "version": "1.0.0",
+  "hook": {
+    "path": "/usr/local/bin/low-latency-hooks.sh",
+    "args": ["low-latency-hooks.sh", "{{.RPSMask}}"]
+  },
+  "when": {
+    "always": true
+  },
+  "stages": ["prestart"]
+}

assets/pao/configs/99-netdev-rps.rules

@@ -0,0 +1 @@
+SUBSYSTEM=="net", ACTION=="add", ENV{DEVPATH}!="/devices/virtual/net/veth*", TAG+="systemd", ENV{SYSTEMD_WANTS}="update-rps@%k.service"

assets/pao/configs/99-runtimes.conf

@@ -0,0 +1,20 @@
+{{if .ReservedCpus}}
+[crio.runtime]
+infra_ctr_cpuset = "{{.ReservedCpus}}"
+{{end}}
+
+# We should copy paste the default runtime because this snippet will override the whole runtimes section
+[crio.runtime.runtimes.runc]
+runtime_path = ""
+runtime_type = "oci"
+runtime_root = "/run/runc"
+
+# The CRI-O will check the allowed_annotations under the runtime handler and apply high-performance hooks when one of
+# high-performance annotations presents under it.
+# We should provide the runtime_path because we need to inform that we want to re-use runc binary and we
+# do not have high-performance binary under the $PATH that will point to it.
+[crio.runtime.runtimes.high-performance]
+runtime_path = "/bin/runc"
+runtime_type = "oci"
+runtime_root = "/run/runc"
+allowed_annotations = ["cpu-load-balancing.crio.io", "cpu-quota.crio.io", "irq-load-balancing.crio.io"]

assets/pao/scripts/hugepages-allocation.sh

@@ -0,0 +1,26 @@
+#!/usr/bin/env bash
+
+set -euo pipefail
+
+nodes_path="/sys/devices/system/node"
+hugepages_file="${nodes_path}/node${NUMA_NODE}/hugepages/hugepages-${HUGEPAGES_SIZE}kB/nr_hugepages"
+
+if [ ! -f "${hugepages_file}" ]; then
+  echo "ERROR: ${hugepages_file} does not exist"
+  exit 1
+fi
+
+timeout=60
+sample=1
+current_time=0
+while [ "$(cat "${hugepages_file}")" -ne "${HUGEPAGES_COUNT}" ]; do
+  echo "${HUGEPAGES_COUNT}" >"${hugepages_file}"
+
+  current_time=$((current_time + sample))
+  if [ $current_time -gt $timeout ]; then
+    echo "ERROR: ${hugepages_file} does not have the expected number of hugepages ${HUGEPAGES_COUNT}"
+    exit 1
+  fi
+
+  sleep $sample
+done

assets/pao/scripts/low-latency-hooks.sh

@@ -0,0 +1,35 @@
+#!/usr/bin/env bash
+
+mask="${1}"
+[ -n "${mask}" ] || { logger "${0}: The rps-mask parameter is missing" ; exit 0; }
+
+pid=$(jq '.pid' /dev/stdin 2>&1)
+[[ $? -eq 0 && -n "${pid}" ]] || { logger "${0}: Failed to extract the pid: ${pid}"; exit 0; }
+
+ns=$(ip netns identify "${pid}" 2>&1)
+[[ $? -eq 0 && -n "${ns}" ]] || { logger "${0} Failed to identify the namespace: ${ns}"; exit 0; }
+
+# Updates the container veth RPS mask on the node
+netns_link_indexes=$(ip netns exec "${ns}" ip -j link | jq ".[] | select(.link_index != null) | .link_index")
+for link_index in ${netns_link_indexes}; do
+  container_veth=$(ip -j link | jq ".[] | select(.ifindex == ${link_index}) | .ifname" | tr -d '"')
+  echo ${mask} > /sys/devices/virtual/net/${container_veth}/queues/rx-0/rps_cpus
+done
+
+# Updates the RPS mask for the interface inside of the container network namespace
+mode=$(ip netns exec "${ns}" [ -w /sys ] && echo "rw" || echo "ro" 2>&1)
+[ $? -eq 0 ] || { logger "${0} Failed to determine if the /sys is writable: ${mode}"; exit 0; }
+
+if [ "${mode}" = "ro" ]; then
+    res=$(ip netns exec "${ns}" mount -o remount,rw /sys 2>&1)
+    [ $? -eq 0 ] || { logger "${0}: Failed to remount /sys as rw: ${res}"; exit 0; }
+fi
+
+# /sys/class/net can't be used recursively to find the rps_cpus file, use /sys/devices instead
+res=$(ip netns exec "${ns}" find /sys/devices -type f -name rps_cpus -exec sh -c "echo ${mask} | cat > {}" \; 2>&1)
+[[ $? -eq 0 && -z "${res}" ]] || logger "${0}: Failed to apply the RPS mask: ${res}"
+
+if [ "${mode}" = "ro" ]; then
+    ip netns exec "${ns}" mount -o remount,ro /sys
+    [ $? -eq 0 ] || exit 1 # Error out so the pod will not start with a writable /sys
+fi

assets/pao/scripts/set-rps-mask.sh

@@ -0,0 +1,36 @@
+#!/usr/bin/env bash
+
+dev=$1
+[ -n "${dev}" ] || { echo "The device argument is missing" >&2 ; exit 1; }
+
+mask=$2
+[ -n "${mask}" ] || { echo "The mask argument is missing" >&2 ; exit 1; }
+
+dev_dir="/sys/class/net/${dev}"
+
+function find_dev_dir {
+  systemd_devs=$(systemctl list-units -t device | grep sys-subsystem-net-devices | cut -d' ' -f1)
+
+  for systemd_dev in ${systemd_devs}; do
+    dev_sysfs=$(systemctl show "${systemd_dev}" -p SysFSPath --value)
+
+    dev_orig_name="${dev_sysfs##*/}"
+    if [ "${dev_orig_name}" = "${dev}" ]; then
+      dev_name="${systemd_dev##*-}"
+      dev_name="${dev_name%%.device}"
+      if [ "${dev_name}" = "${dev}" ]; then # disregard the original device unit
+              continue
+      fi
+
+      echo "${dev} device was renamed to $dev_name"
+      dev_dir="/sys/class/net/${dev_name}"
+      break
+    fi
+  done
+}
+
+[ -d "${dev_dir}" ] || find_dev_dir                # the net device was renamed, find the new name
+[ -d "${dev_dir}" ] || { sleep 5; find_dev_dir; }  # search failed, wait a little and try again
+[ -d "${dev_dir}" ] || { echo "${dev_dir}" directory not found >&2 ; exit 0; } # the interface disappeared, not an error
+
+find "${dev_dir}"/queues -type f -name rps_cpus -exec sh -c "echo ${mask} | cat > {}" \;

assets/pao/tuned/openshift-node-performance

@@ -0,0 +1,132 @@
+[main]
+summary=Openshift node optimized for deterministic performance at the cost of increased power consumption, focused on low latency network performance. Based on Tuned 2.11 and Cluster node tuning (oc 4.5)
+include=openshift-node,cpu-partitioning
+
+# Inheritance of base profiles legend:
+# cpu-partitioning -> network-latency -> latency-performance
+# https://github.com/redhat-performance/tuned/blob/master/profiles/latency-performance/tuned.conf
+# https://github.com/redhat-performance/tuned/blob/master/profiles/network-latency/tuned.conf
+# https://github.com/redhat-performance/tuned/blob/master/profiles/cpu-partitioning/tuned.conf
+
+# All values are mapped with a comment where a parent profile contains them.
+# Different values will override the original values in parent profiles.
+
+[variables]
+#> isolated_cores take a list of ranges; e.g. isolated_cores=2,4-7
+{{if .IsolatedCpus}}
+isolated_cores={{.IsolatedCpus}}
+{{end}}
+
+not_isolated_cores_expanded=${f:cpulist_invert:${isolated_cores_expanded}}
+
+[cpu]
+#> latency-performance
+#> (override)
+force_latency=cstate.id:1|3
+governor=performance
+energy_perf_bias=performance
+min_perf_pct=100
+
+[service]
+service.stalld=start,enable
+
+[vm]
+#> network-latency
+transparent_hugepages=never
+
+{{if not .GloballyDisableIrqLoadBalancing}}
+[irqbalance]
+#> Override the value set by cpu-partitioning with an empty one
+banned_cpus=""
+{{end}}
+
+[scheduler]
+runtime=0
+group.ksoftirqd=0:f:11:*:ksoftirqd.*
+group.rcuc=0:f:11:*:rcuc.*
+sched_rt_runtime_us=-1
+sched_min_granularity_ns=10000000
+sched_migration_cost_ns=5000000
+numa_balancing=0
+{{if not .GloballyDisableIrqLoadBalancing}}
+default_irq_smp_affinity = ignore
+{{end}}
+
+[sysctl]
+#> cpu-partitioning #realtime
+kernel.hung_task_timeout_secs = 600
+#> cpu-partitioning #realtime
+kernel.nmi_watchdog = 0
+#> realtime
+kernel.sched_rt_runtime_us = -1
+# cpu-partitioning and realtime for RHEL disable it (= 0)
+# OCP is too dynamic when partitioning and needs to evacuate
+#> scheduled timers when starting a guaranteed workload (= 1)
+kernel.timer_migration = 1
+#> network-latency
+kernel.numa_balancing=0
+net.core.busy_read=50
+net.core.busy_poll=50
+net.ipv4.tcp_fastopen=3
+#> cpu-partitioning  #realtime
+vm.stat_interval = 10
+
+# ktune sysctl settings for rhel6 servers, maximizing i/o throughput
+#
+# Minimal preemption granularity for CPU-bound tasks:
+# (default: 1 msec#  (1 + ilog(ncpus)), units: nanoseconds)
+#> latency-performance
+kernel.sched_min_granularity_ns=10000000
+
+# If a workload mostly uses anonymous memory and it hits this limit, the entire
+# working set is buffered for I/O, and any more write buffering would require
+# swapping, so it's time to throttle writes until I/O can catch up.  Workloads
+# that mostly use file mappings may be able to use even higher values.
+#
+# The generator of dirty data starts writeback at this percentage (system default
+# is 20%)
+#> latency-performance
+vm.dirty_ratio=10
+
+# Start background writeback (via writeback threads) at this percentage (system
+# default is 10%)
+#> latency-performance
+vm.dirty_background_ratio=3
+
+# The swappiness parameter controls the tendency of the kernel to move
+# processes out of physical memory and onto the swap disk.
+# 0 tells the kernel to avoid swapping processes out of physical memory
+# for as long as possible
+# 100 tells the kernel to aggressively swap processes out of physical memory
+# and move them to swap cache
+#> latency-performance
+vm.swappiness=10
+
+# The total time the scheduler will consider a migrated process
+# "cache hot" and thus less likely to be re-migrated
+# (system default is 500000, i.e. 0.5 ms)
+#> latency-performance
+kernel.sched_migration_cost_ns=5000000
+
+[selinux]
+#> Custom (atomic host)
+avc_cache_threshold=8192
+
+{{if .NetDevices}}
+{{.NetDevices}}
+{{end}}
+
+[bootloader]
+# set empty values to disable RHEL initrd setting in cpu-partitioning
+initrd_remove_dir=
+initrd_dst_img=
+initrd_add_dir=
+# overrides cpu-partitioning cmdline
+cmdline_cpu_part=+nohz=on rcu_nocbs=${isolated_cores} tuned.non_isolcpus=${not_isolated_cpumask} intel_pstate=disable nosoftlockup
+{{if .StaticIsolation}}
+cmdline_realtime=+tsc=nowatchdog intel_iommu=on iommu=pt isolcpus=domain,managed_irq,${isolated_cores} systemd.cpu_affinity=${not_isolated_cores_expanded}
+{{else}}
+cmdline_realtime=+tsc=nowatchdog intel_iommu=on iommu=pt isolcpus=managed_irq,${isolated_cores} systemd.cpu_affinity=${not_isolated_cores_expanded}
+{{end}}
+cmdline_hugepages=+{{if .DefaultHugepagesSize}} default_hugepagesz={{.DefaultHugepagesSize}} {{end}} {{if .Hugepages}} {{.Hugepages}} {{end}}
+cmdline_additionalArg=+{{if .AdditionalArgs}} {{.AdditionalArgs}} {{end}}