diff --git a/windows/hcsshim.go b/windows/hcsshim.go
index 031a6359f..98ddee816 100644
--- a/windows/hcsshim.go
+++ b/windows/hcsshim.go
@@ -10,6 +10,7 @@ import (
 	"strings"
 
 	"github.com/Microsoft/hcsshim"
+	"github.com/Microsoft/opengcs/client"
 	"github.com/containerd/containerd/errdefs"
 	"github.com/containerd/containerd/log"
 	specs "github.com/opencontainers/runtime-spec/specs-go"
@@ -19,29 +20,46 @@ import (
 // newContainerConfig generates a hcsshim container configuration from the
 // provided OCI Spec
 func newContainerConfig(ctx context.Context, owner, id string, spec *specs.Spec) (*hcsshim.ContainerConfig, error) {
-	if len(spec.Windows.LayerFolders) == 0 {
-		return nil, errors.Wrap(errdefs.ErrInvalidArgument,
-			"spec.Windows.LayerFolders cannot be empty")
-	}
-
 	var (
-		layerFolders = spec.Windows.LayerFolders
-		conf         = &hcsshim.ContainerConfig{
-			SystemType:                 "Container",
-			Name:                       id,
-			Owner:                      owner,
-			HostName:                   spec.Hostname,
-			IgnoreFlushesDuringBoot:    spec.Windows.IgnoreFlushesDuringBoot,
-			AllowUnqualifiedDNSQuery:   spec.Windows.Network.AllowUnqualifiedDNSQuery,
-			EndpointList:               spec.Windows.Network.EndpointList,
-			NetworkSharedContainerName: spec.Windows.Network.NetworkSharedContainerName,
+		conf = &hcsshim.ContainerConfig{
+			SystemType: "Container",
+			Name:       id,
+			Owner:      owner,
+			HostName:   spec.Hostname,
 		}
 	)
 
-	if spec.Windows.CredentialSpec != nil {
-		conf.Credentials = spec.Windows.CredentialSpec.(string)
+	if spec.Windows.Network != nil {
+		conf.AllowUnqualifiedDNSQuery = spec.Windows.Network.AllowUnqualifiedDNSQuery
+		conf.EndpointList = spec.Windows.Network.EndpointList
+		conf.NetworkSharedContainerName = spec.Windows.Network.NetworkSharedContainerName
+		if spec.Windows.Network.DNSSearchList != nil {
+			conf.DNSSearchList = strings.Join(spec.Windows.Network.DNSSearchList, ",")
+		}
 	}
 
+	return conf, nil
+}
+
+// newWindowsContainerConfig generates a hcsshim Windows container
+// configuration from the provided OCI Spec
+func newWindowsContainerConfig(ctx context.Context, owner, id string, spec *specs.Spec) (*hcsshim.ContainerConfig, error) {
+	conf, err := newContainerConfig(ctx, owner, id, spec)
+	if err != nil {
+		return nil, err
+	}
+	conf.IgnoreFlushesDuringBoot = spec.Windows.IgnoreFlushesDuringBoot
+
+	if len(spec.Windows.LayerFolders) < 1 {
+		return nil, errors.Wrap(errdefs.ErrInvalidArgument,
+			"spec.Windows.LayerFolders must have at least 1 layers")
+	}
+	var (
+		layerFolders    = spec.Windows.LayerFolders
+		homeDir         = filepath.Dir(layerFolders[0])
+		layerFolderPath = filepath.Join(homeDir, id)
+	)
+
 	// TODO: use the create request Mount for those
 	for _, layerPath := range layerFolders {
 		_, filename := filepath.Split(layerPath)
@@ -55,6 +73,60 @@ func newContainerConfig(ctx context.Context, owner, id string, spec *specs.Spec)
 		})
 	}
 
+	var (
+		di = hcsshim.DriverInfo{
+			Flavour: 1, // filter driver
+			HomeDir: homeDir,
+		}
+	)
+	conf.LayerFolderPath = layerFolderPath
+
+	// TODO: Once there is a snapshotter for windows, this can be deleted.
+	// The R/W Layer should come from the Rootfs Mounts provided
+	//
+	// Windows doesn't support creating a container with a readonly
+	// filesystem, so always create a RW one
+	if err = hcsshim.CreateSandboxLayer(di, id, layerFolders[0], layerFolders); err != nil {
+		return nil, errors.Wrapf(err, "failed to create sandbox layer for %s: layers: %#v, driverInfo: %#v",
+			id, layerFolders, di)
+	}
+	defer func() {
+		if err != nil {
+			removeLayer(ctx, conf.LayerFolderPath)
+		}
+	}()
+
+	if err = hcsshim.ActivateLayer(di, id); err != nil {
+		return nil, errors.Wrapf(err, "failed to activate layer %s", conf.LayerFolderPath)
+	}
+
+	if err = hcsshim.PrepareLayer(di, id, layerFolders); err != nil {
+		return nil, errors.Wrapf(err, "failed to prepare layer %s", conf.LayerFolderPath)
+	}
+
+	conf.VolumePath, err = hcsshim.GetLayerMountPath(di, id)
+	if err != nil {
+		return nil, errors.Wrapf(err, "failed to getmount path for layer %s: driverInfo: %#v", id, di)
+	}
+
+	if spec.Windows.HyperV != nil {
+		conf.HvPartition = true
+		for _, layerPath := range layerFolders {
+			utilityVMPath := spec.Windows.HyperV.UtilityVMPath
+			_, err := os.Stat(utilityVMPath)
+			if err == nil {
+				conf.HvRuntime = &hcsshim.HvRuntime{ImagePath: utilityVMPath}
+				break
+			} else if !os.IsNotExist(err) {
+				return nil, errors.Wrapf(err, "failed to access layer %s", layerPath)
+			}
+		}
+	}
+
+	if spec.Windows.CredentialSpec != nil {
+		conf.Credentials = spec.Windows.CredentialSpec.(string)
+	}
+
 	if len(spec.Mounts) > 0 {
 		mds := make([]hcsshim.MappedDir, len(spec.Mounts))
 		for i, mount := range spec.Mounts {
@@ -72,59 +144,51 @@ func newContainerConfig(ctx context.Context, owner, id string, spec *specs.Spec)
 		conf.MappedDirectories = mds
 	}
 
-	if spec.Windows.Network.DNSSearchList != nil {
-		conf.DNSSearchList = strings.Join(spec.Windows.Network.DNSSearchList, ",")
+	return conf, nil
+}
+
+// newLinuxConfig generates a hcsshim Linux container configuration from the
+// provided OCI Spec
+func newLinuxConfig(ctx context.Context, owner, id string, spec *specs.Spec) (*hcsshim.ContainerConfig, error) {
+	conf, err := newContainerConfig(ctx, owner, id, spec)
+	if err != nil {
+		return nil, err
 	}
 
-	if spec.Windows.HyperV != nil {
-		conf.HvPartition = true
-		for _, layerPath := range layerFolders {
-			utilityVMPath := spec.Windows.HyperV.UtilityVMPath
-			_, err := os.Stat(utilityVMPath)
-			if err == nil {
-				conf.HvRuntime = &hcsshim.HvRuntime{ImagePath: utilityVMPath}
-				break
-			} else if !os.IsNotExist(err) {
-				return nil, errors.Wrapf(err, "failed to access layer %s", layerPath)
-			}
-		}
-	}
+	conf.ContainerType = "Linux"
+	conf.HvPartition = true
 
+	if len(spec.Windows.LayerFolders) < 1 {
+		return nil, errors.Wrap(errdefs.ErrInvalidArgument,
+			"spec.Windows.LayerFolders must have at least 1 layer")
+	}
 	var (
-		err error
-		di  = hcsshim.DriverInfo{
-			Flavour: 1, // filter driver
-			HomeDir: filepath.Dir(layerFolders[0]),
-		}
+		layerFolders = spec.Windows.LayerFolders
 	)
 
-	// TODO: Once there is a snapshotter for windows, this can be deleted.
-	// The R/W Layer should come from the Rootfs Mounts provided
-	//
-	// Windows doesn't support creating a container with a readonly
-	// filesystem, so always create a RW one
-	if err = hcsshim.CreateSandboxLayer(di, id, layerFolders[0], layerFolders); err != nil {
-		return nil, errors.Wrapf(err, "failed to create sandbox layer for %s: layers: %#v, driverInfo: %#v",
-			id, layerFolders, di)
+	config := &client.Config{}
+	if err := config.GenerateDefault(nil); err != nil {
+		return nil, err
 	}
-	conf.LayerFolderPath = filepath.Join(di.HomeDir, id)
-	defer func() {
+
+	conf.HvRuntime = &hcsshim.HvRuntime{
+		ImagePath:           config.KirdPath,
+		LinuxKernelFile:     config.KernelFile,
+		LinuxInitrdFile:     config.InitrdFile,
+		LinuxBootParameters: config.BootParameters,
+	}
+
+	// TODO: use the create request Mount for those
+	for _, layerPath := range layerFolders {
+		_, filename := filepath.Split(layerPath)
+		guid, err := hcsshim.NameToGuid(filename)
 		if err != nil {
-			removeLayer(ctx, conf.LayerFolderPath)
+			return nil, errors.Wrapf(err, "unable to get GUID for %s", filename)
 		}
-	}()
-
-	if err = hcsshim.ActivateLayer(di, id); err != nil {
-		return nil, errors.Wrapf(err, "failed to activate layer %s", conf.LayerFolderPath)
-	}
-
-	if err = hcsshim.PrepareLayer(di, id, layerFolders); err != nil {
-		return nil, errors.Wrapf(err, "failed to prepare layer %s", conf.LayerFolderPath)
-	}
-
-	conf.VolumePath, err = hcsshim.GetLayerMountPath(di, id)
-	if err != nil {
-		return nil, errors.Wrapf(err, "failed to getmount path for layer %s: driverInfo: %#v", id, di)
+		conf.Layers = append(conf.Layers, hcsshim.Layer{
+			ID:   guid.ToString(),
+			Path: filepath.Join(layerPath, "layer.vhd"),
+		})
 	}
 
 	return conf, nil
@@ -165,21 +229,23 @@ func removeLayer(ctx context.Context, path string) error {
 	return nil
 }
 
-func newProcessConfig(spec *specs.Process, pset *pipeSet) *hcsshim.ProcessConfig {
+func newProcessConfig(processSpec *specs.Process, pset *pipeSet) *hcsshim.ProcessConfig {
 	conf := &hcsshim.ProcessConfig{
 		EmulateConsole:   pset.src.Terminal,
 		CreateStdInPipe:  pset.stdin != nil,
 		CreateStdOutPipe: pset.stdout != nil,
 		CreateStdErrPipe: pset.stderr != nil,
-		User:             spec.User.Username,
-		CommandLine:      strings.Join(spec.Args, " "),
+		User:             processSpec.User.Username,
 		Environment:      make(map[string]string),
-		WorkingDirectory: spec.Cwd,
-		ConsoleSize:      [2]uint{spec.ConsoleSize.Height, spec.ConsoleSize.Width},
+		WorkingDirectory: processSpec.Cwd,
+	}
+
+	if processSpec.ConsoleSize != nil {
+		conf.ConsoleSize = [2]uint{processSpec.ConsoleSize.Height, processSpec.ConsoleSize.Width}
 	}
 
 	// Convert OCI Env format to HCS's
-	for _, s := range spec.Env {
+	for _, s := range processSpec.Env {
 		arr := strings.SplitN(s, "=", 2)
 		if len(arr) == 2 {
 			conf.Environment[arr[0]] = arr[1]
@@ -188,3 +254,15 @@ func newProcessConfig(spec *specs.Process, pset *pipeSet) *hcsshim.ProcessConfig
 
 	return conf
 }
+
+func newWindowsProcessConfig(processSpec *specs.Process, pset *pipeSet) *hcsshim.ProcessConfig {
+	conf := newProcessConfig(processSpec, pset)
+	conf.CommandLine = strings.Join(processSpec.Args, " ")
+	return conf
+}
+
+func newLinuxProcessConfig(processSpec *specs.Process, pset *pipeSet) (*hcsshim.ProcessConfig, error) {
+	conf := newProcessConfig(processSpec, pset)
+	conf.CommandArgs = processSpec.Args
+	return conf, nil
+}
diff --git a/windows/runtime.go b/windows/runtime.go
index ce69dabff..74f499e28 100644
--- a/windows/runtime.go
+++ b/windows/runtime.go
@@ -243,7 +243,7 @@ func (r *windowsRuntime) newTask(ctx context.Context, namespace, id string, spec
 		conf *hcsshim.ContainerConfig
 		nsid = namespace + "-" + id
 	)
-	if conf, err = newContainerConfig(ctx, hcsshimOwner, nsid, spec); err != nil {
+	if conf, err = newWindowsContainerConfig(ctx, hcsshimOwner, nsid, spec); err != nil {
 		return nil, err
 	}
 	defer func() {
diff --git a/windows/task.go b/windows/task.go
index 86078f71d..33e5b7041 100644
--- a/windows/task.go
+++ b/windows/task.go
@@ -111,7 +111,7 @@ func (t *task) Info() runtime.TaskInfo {
 }
 
 func (t *task) Start(ctx context.Context) error {
-	conf := newProcessConfig(t.spec.Process, t.io)
+	conf := newWindowsProcessConfig(t.spec.Process, t.io)
 	p, err := t.newProcess(ctx, t.id, conf, t.io)
 	if err != nil {
 		return err
@@ -194,7 +194,7 @@ func (t *task) Exec(ctx context.Context, id string, opts runtime.ExecOpts) (runt
 		return nil, err
 	}
 
-	conf := newProcessConfig(spec, pset)
+	conf := newWindowsProcessConfig(spec, pset)
 	p, err := t.newProcess(ctx, id, conf, pset)
 	if err != nil {
 		return nil, err