forked from mirrors/gotosocial
acc333c40b
When GTS is running in a container runtime which has configured CPU or memory limits or under an init system that uses cgroups to impose CPU and memory limits the values the Go runtime sees for GOMAXPROCS and GOMEMLIMIT are still based on the host resources, not the cgroup. At least for the throttling middlewares which use GOMAXPROCS to configure their queue size, this can result in GTS running with values too big compared to the resources that will actuall be available to it. This introduces 2 dependencies which can pick up resource contraints from the current cgroup and tune the Go runtime accordingly. This should result in the different queues being appropriately sized and in general more predictable performance. These dependencies are a no-op on non-Linux systems or if running in a cgroup that doesn't set a limit on CPU or memory. The automatic tuning of GOMEMLIMIT can be disabled by either explicitly setting GOMEMLIMIT yourself or by setting AUTOMEMLIMIT=off. The automatic tuning of GOMAXPROCS can similarly be counteracted by setting GOMAXPROCS yourself.
155 lines
3.6 KiB
Go
155 lines
3.6 KiB
Go
/*
|
|
Copyright The containerd Authors.
|
|
|
|
Licensed under the Apache License, Version 2.0 (the "License");
|
|
you may not use this file except in compliance with the License.
|
|
You may obtain a copy of the License at
|
|
|
|
http://www.apache.org/licenses/LICENSE-2.0
|
|
|
|
Unless required by applicable law or agreed to in writing, software
|
|
distributed under the License is distributed on an "AS IS" BASIS,
|
|
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
See the License for the specific language governing permissions and
|
|
limitations under the License.
|
|
*/
|
|
|
|
package cgroups
|
|
|
|
import (
|
|
"io/ioutil"
|
|
"math"
|
|
"os"
|
|
"path/filepath"
|
|
"strconv"
|
|
"strings"
|
|
|
|
v1 "github.com/containerd/cgroups/stats/v1"
|
|
specs "github.com/opencontainers/runtime-spec/specs-go"
|
|
)
|
|
|
|
type rdmaController struct {
|
|
root string
|
|
}
|
|
|
|
func (p *rdmaController) Name() Name {
|
|
return Rdma
|
|
}
|
|
|
|
func (p *rdmaController) Path(path string) string {
|
|
return filepath.Join(p.root, path)
|
|
}
|
|
|
|
func NewRdma(root string) *rdmaController {
|
|
return &rdmaController{
|
|
root: filepath.Join(root, string(Rdma)),
|
|
}
|
|
}
|
|
|
|
func createCmdString(device string, limits *specs.LinuxRdma) string {
|
|
var cmdString string
|
|
|
|
cmdString = device
|
|
if limits.HcaHandles != nil {
|
|
cmdString = cmdString + " " + "hca_handle=" + strconv.FormatUint(uint64(*limits.HcaHandles), 10)
|
|
}
|
|
|
|
if limits.HcaObjects != nil {
|
|
cmdString = cmdString + " " + "hca_object=" + strconv.FormatUint(uint64(*limits.HcaObjects), 10)
|
|
}
|
|
return cmdString
|
|
}
|
|
|
|
func (p *rdmaController) Create(path string, resources *specs.LinuxResources) error {
|
|
if err := os.MkdirAll(p.Path(path), defaultDirPerm); err != nil {
|
|
return err
|
|
}
|
|
|
|
for device, limit := range resources.Rdma {
|
|
if device != "" && (limit.HcaHandles != nil || limit.HcaObjects != nil) {
|
|
limit := limit
|
|
return retryingWriteFile(
|
|
filepath.Join(p.Path(path), "rdma.max"),
|
|
[]byte(createCmdString(device, &limit)),
|
|
defaultFilePerm,
|
|
)
|
|
}
|
|
}
|
|
return nil
|
|
}
|
|
|
|
func (p *rdmaController) Update(path string, resources *specs.LinuxResources) error {
|
|
return p.Create(path, resources)
|
|
}
|
|
|
|
func parseRdmaKV(raw string, entry *v1.RdmaEntry) {
|
|
var value uint64
|
|
var err error
|
|
|
|
parts := strings.Split(raw, "=")
|
|
switch len(parts) {
|
|
case 2:
|
|
if parts[1] == "max" {
|
|
value = math.MaxUint32
|
|
} else {
|
|
value, err = parseUint(parts[1], 10, 32)
|
|
if err != nil {
|
|
return
|
|
}
|
|
}
|
|
if parts[0] == "hca_handle" {
|
|
entry.HcaHandles = uint32(value)
|
|
} else if parts[0] == "hca_object" {
|
|
entry.HcaObjects = uint32(value)
|
|
}
|
|
}
|
|
}
|
|
|
|
func toRdmaEntry(strEntries []string) []*v1.RdmaEntry {
|
|
var rdmaEntries []*v1.RdmaEntry
|
|
for i := range strEntries {
|
|
parts := strings.Fields(strEntries[i])
|
|
switch len(parts) {
|
|
case 3:
|
|
entry := new(v1.RdmaEntry)
|
|
entry.Device = parts[0]
|
|
parseRdmaKV(parts[1], entry)
|
|
parseRdmaKV(parts[2], entry)
|
|
|
|
rdmaEntries = append(rdmaEntries, entry)
|
|
default:
|
|
continue
|
|
}
|
|
}
|
|
return rdmaEntries
|
|
}
|
|
|
|
func (p *rdmaController) Stat(path string, stats *v1.Metrics) error {
|
|
|
|
currentData, err := ioutil.ReadFile(filepath.Join(p.Path(path), "rdma.current"))
|
|
if err != nil {
|
|
return err
|
|
}
|
|
currentPerDevices := strings.Split(string(currentData), "\n")
|
|
|
|
maxData, err := ioutil.ReadFile(filepath.Join(p.Path(path), "rdma.max"))
|
|
if err != nil {
|
|
return err
|
|
}
|
|
maxPerDevices := strings.Split(string(maxData), "\n")
|
|
|
|
// If device got removed between reading two files, ignore returning
|
|
// stats.
|
|
if len(currentPerDevices) != len(maxPerDevices) {
|
|
return nil
|
|
}
|
|
|
|
currentEntries := toRdmaEntry(currentPerDevices)
|
|
maxEntries := toRdmaEntry(maxPerDevices)
|
|
|
|
stats.Rdma = &v1.RdmaStat{
|
|
Current: currentEntries,
|
|
Limit: maxEntries,
|
|
}
|
|
return nil
|
|
}
|