Skip to content

Add --disable-hook flag to cdi generate command #1077

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 3 commits into
base: main
Choose a base branch
from
+236 −11
Open

Add --disable-hook flag to cdi generate command #1077

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
96ecfd0
Add --disable-hook flag to cdi generate command
ArangoGutierrez May 12, 2025
d657e0a
rename flag to disable-hooks
ArangoGutierrez May 14, 2025
b6ac316
Add all option to disable all hooks
ArangoGutierrez May 14, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
21 changes: 19 additions & 2 deletions cmd/nvidia-ctk/cdi/generate/generate.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -57,6 +57,7 @@ type options struct {

configSearchPaths cli.StringSlice
librarySearchPaths cli.StringSlice
disableHooks cli.StringSlice

csv struct {
files cli.StringSlice
Expand Down Expand Up @@ -176,6 +177,12 @@ func (m command) build() *cli.Command {
Usage: "Specify a pattern the CSV mount specifications.",
Destination: &opts.csv.ignorePatterns,
},
&cli.StringSliceFlag{
Name: "disable-hook",
Usage: "Comma-separated list of hooks to skip when generating the CDI specification.",
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Technically the flag can also be specified multiple times.

Value: cli.NewStringSlice(),
Destination: &opts.disableHooks,
},
}

return &c
Expand Down Expand Up @@ -262,7 +269,7 @@ func (m command) generateSpec(opts *options) (spec.Interface, error) {
deviceNamers = append(deviceNamers, deviceNamer)
}

cdilib, err := nvcdi.New(
initOpts := []nvcdi.Option{
nvcdi.WithLogger(m.logger),
nvcdi.WithDriverRoot(opts.driverRoot),
nvcdi.WithDevRoot(opts.devRoot),
Expand All @@ -276,7 +283,17 @@ func (m command) generateSpec(opts *options) (spec.Interface, error) {
nvcdi.WithCSVIgnorePatterns(opts.csv.ignorePatterns.Value()),
// We set the following to allow for dependency injection:
nvcdi.WithNvmlLib(opts.nvmllib),
)
}

if len(opts.disableHooks.Value()) > 0 {
for _, hook := range opts.disableHooks.Value() {
for _, hookName := range nvcdi.NewHookName(hook) {
initOpts = append(initOpts, nvcdi.WithDisabledHook(hookName))
}
}
}

cdilib, err := nvcdi.New(initOpts...)
if err != nil {
return nil, fmt.Errorf("failed to create CDI library: %v", err)
}
Expand Down
166 changes: 166 additions & 0 deletions cmd/nvidia-ctk/cdi/generate/generate_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -26,6 +26,7 @@ import (
"github.com/NVIDIA/go-nvml/pkg/nvml/mock/dgxa100"
testlog "github.com/sirupsen/logrus/hooks/test"
"github.com/stretchr/testify/require"
"github.com/urfave/cli/v2"

"github.com/NVIDIA/nvidia-container-toolkit/internal/test"
)
Expand All @@ -36,6 +37,9 @@ func TestGenerateSpec(t *testing.T) {
require.NoError(t, err)

driverRoot := filepath.Join(moduleRoot, "testdata", "lookup", "rootfs-1")
disableHook1 := cli.NewStringSlice("enable-cuda-compat")
disableHook2 := cli.NewStringSlice("enable-cuda-compat", "update-ldcache")
disableHook3 := cli.NewStringSlice("all")

logger, _ := testlog.NewNullLogger()
testCases := []struct {
Expand Down Expand Up @@ -112,6 +116,168 @@ containerEdits:
- nosuid
- nodev
- bind
`,
},
{
description: "disableHooks1",
options: options{
format: "yaml",
mode: "nvml",
vendor: "example.com",
class: "device",
driverRoot: driverRoot,
disableHooks: *disableHook1,
},
expectedOptions: options{
format: "yaml",
mode: "nvml",
vendor: "example.com",
class: "device",
nvidiaCDIHookPath: "/usr/bin/nvidia-cdi-hook",
driverRoot: driverRoot,
disableHooks: *disableHook1,
},
expectedSpec: `---
cdiVersion: 0.5.0
kind: example.com/device
devices:
- name: "0"
containerEdits:
deviceNodes:
- path: /dev/nvidia0
hostPath: {{ .driverRoot }}/dev/nvidia0
- name: all
containerEdits:
deviceNodes:
- path: /dev/nvidia0
hostPath: {{ .driverRoot }}/dev/nvidia0
containerEdits:
env:
- NVIDIA_VISIBLE_DEVICES=void
deviceNodes:
- path: /dev/nvidiactl
hostPath: {{ .driverRoot }}/dev/nvidiactl
hooks:
- hookName: createContainer
path: /usr/bin/nvidia-cdi-hook
args:
- nvidia-cdi-hook
- create-symlinks
- --link
- libcuda.so.1::/lib/x86_64-linux-gnu/libcuda.so
- hookName: createContainer
path: /usr/bin/nvidia-cdi-hook
args:
- nvidia-cdi-hook
- update-ldcache
- --folder
- /lib/x86_64-linux-gnu
mounts:
- hostPath: {{ .driverRoot }}/lib/x86_64-linux-gnu/libcuda.so.999.88.77
containerPath: /lib/x86_64-linux-gnu/libcuda.so.999.88.77
options:
- ro
- nosuid
- nodev
- bind
`,
},
{
description: "disableHooks2",
options: options{
format: "yaml",
mode: "nvml",
vendor: "example.com",
class: "device",
driverRoot: driverRoot,
disableHooks: *disableHook2,
},
expectedOptions: options{
format: "yaml",
mode: "nvml",
vendor: "example.com",
class: "device",
nvidiaCDIHookPath: "/usr/bin/nvidia-cdi-hook",
driverRoot: driverRoot,
disableHooks: *disableHook2,
},
expectedSpec: `---
cdiVersion: 0.5.0
kind: example.com/device
devices:
- name: "0"
containerEdits:
deviceNodes:
- path: /dev/nvidia0
hostPath: {{ .driverRoot }}/dev/nvidia0
- name: all
containerEdits:
deviceNodes:
- path: /dev/nvidia0
hostPath: {{ .driverRoot }}/dev/nvidia0
containerEdits:
env:
- NVIDIA_VISIBLE_DEVICES=void
deviceNodes:
- path: /dev/nvidiactl
hostPath: {{ .driverRoot }}/dev/nvidiactl
hooks:
- hookName: createContainer
path: /usr/bin/nvidia-cdi-hook
args:
- nvidia-cdi-hook
- create-symlinks
- --link
- libcuda.so.1::/lib/x86_64-linux-gnu/libcuda.so
mounts:
- hostPath: {{ .driverRoot }}/lib/x86_64-linux-gnu/libcuda.so.999.88.77
containerPath: /lib/x86_64-linux-gnu/libcuda.so.999.88.77
options:
- ro
- nosuid
- nodev
- bind
`,
},
{
description: "disableHooksAll",
options: options{
format: "yaml",
mode: "nvml",
vendor: "example.com",
class: "device",
driverRoot: driverRoot,
disableHooks: *disableHook3,
},
expectedOptions: options{
format: "yaml",
mode: "nvml",
vendor: "example.com",
class: "device",
nvidiaCDIHookPath: "/usr/bin/nvidia-cdi-hook",
driverRoot: driverRoot,
disableHooks: *disableHook3,
},
expectedSpec: `---
cdiVersion: 0.5.0
kind: example.com/device
devices:
- name: "0"
containerEdits:
deviceNodes:
- path: /dev/nvidia0
hostPath: {{ .driverRoot }}/dev/nvidia0
- name: all
containerEdits:
deviceNodes:
- path: /dev/nvidia0
hostPath: {{ .driverRoot }}/dev/nvidia0
containerEdits:
env:
- NVIDIA_VISIBLE_DEVICES=void
deviceNodes:
- path: /dev/nvidiactl
hostPath: {{ .driverRoot }}/dev/nvidiactl
`,
},
}
Expand Down
29 changes: 29 additions & 0 deletions pkg/nvcdi/api.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -44,4 +44,33 @@ const (
// HookEnableCudaCompat refers to the hook used to enable CUDA Forward Compatibility.
// This was added with v1.17.5 of the NVIDIA Container Toolkit.
HookEnableCudaCompat = HookName("enable-cuda-compat")
// HookCreateSymlinks refers to the hook used create symlinks inside the
// directory path to be mounted into a container.
HookCreateSymlinks = HookName("create-symlinks")
// HookUpdateLDCache refers to the hook used to Update the dynamic linker
// cache inside the directory path to be mounted into a container.
HookUpdateLDCache = HookName("update-ldcache")
)

// NewHookName takes a string and returns a []HookName, empty if the HookName
// is invalid and all Hooks if the string is "all"
func NewHookName(hookName string) []HookName {
if hookName == "" {
return []HookName{}
}

if hookName == "all" {
return []HookName{HookEnableCudaCompat, HookCreateSymlinks, HookUpdateLDCache}
}

switch hookName {
case string(HookEnableCudaCompat):
return []HookName{HookEnableCudaCompat}
case string(HookCreateSymlinks):
return []HookName{HookCreateSymlinks}
case string(HookUpdateLDCache):
return []HookName{HookUpdateLDCache}
default:
return nil
}
}
31 changes: 22 additions & 9 deletions pkg/nvcdi/driver-nvml.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -99,21 +99,34 @@ func (l *nvcdilib) NewDriverLibraryDiscoverer(version string) (discover.Discover

var discoverers []discover.Discover

driverDotSoSymlinksDiscoverer := discover.WithDriverDotSoSymlinks(
libraries,
version,
l.nvidiaCDIHookPath,
)
discoverers = append(discoverers, driverDotSoSymlinksDiscoverer)
if l.HookIsSupported(HookCreateSymlinks) {
driverDotSoSymlinksDiscoverer := discover.WithDriverDotSoSymlinks(
libraries,
version,
l.nvidiaCDIHookPath,
)
discoverers = append(discoverers, driverDotSoSymlinksDiscoverer)
}

if l.HookIsSupported(HookEnableCudaCompat) {
// TODO: The following should use the version directly.
cudaCompatLibHookDiscoverer := discover.NewCUDACompatHookDiscoverer(l.logger, l.nvidiaCDIHookPath, l.driver)
cudaCompatLibHookDiscoverer := discover.NewCUDACompatHookDiscoverer(
l.logger,
l.nvidiaCDIHookPath,
l.driver,
)
discoverers = append(discoverers, cudaCompatLibHookDiscoverer)
}

updateLDCache, _ := discover.NewLDCacheUpdateHook(l.logger, libraries, l.nvidiaCDIHookPath, l.ldconfigPath)
discoverers = append(discoverers, updateLDCache)
if l.HookIsSupported(HookUpdateLDCache) {
updateLDCache, _ := discover.NewLDCacheUpdateHook(
l.logger,
libraries,
l.nvidiaCDIHookPath,
l.ldconfigPath,
)
discoverers = append(discoverers, updateLDCache)
}

d := discover.Merge(discoverers...)

Expand Down