Merge release-20210705.0-11-ge3fdd1593 (automated)

91 files changed, 382 insertions, 423 deletions

diff --git a/pkg/sentry/arch/arch_aarch64.go b/pkg/sentry/arch/arch_aarch64.go
index 7def71fef..18942e848 100644
--- a/pkg/sentry/arch/arch_aarch64.go
+++ b/pkg/sentry/arch/arch_aarch64.go
@@ -26,7 +26,6 @@ import (
 	"gvisor.dev/gvisor/pkg/log"
 	"gvisor.dev/gvisor/pkg/sentry/arch/fpu"
 	rpb "gvisor.dev/gvisor/pkg/sentry/arch/registers_go_proto"
-	"gvisor.dev/gvisor/pkg/syserror"
 )
 
 // Registers represents the CPU registers for this architecture.
@@ -234,7 +233,7 @@ func (s *State) PtraceGetRegSet(regset uintptr, dst io.Writer, maxlen int) (int,
 	switch regset {
 	case _NT_PRSTATUS:
 		if maxlen < ptraceRegistersSize {
-			return 0, syserror.EFAULT
+			return 0, linuxerr.EFAULT
 		}
 		return s.PtraceGetRegs(dst)
 	default:
@@ -247,7 +246,7 @@ func (s *State) PtraceSetRegSet(regset uintptr, src io.Reader, maxlen int) (int,
 	switch regset {
 	case _NT_PRSTATUS:
 		if maxlen < ptraceRegistersSize {
-			return 0, syserror.EFAULT
+			return 0, linuxerr.EFAULT
 		}
 		return s.PtraceSetRegs(src)
 	default:
diff --git a/pkg/sentry/arch/arch_x86.go b/pkg/sentry/arch/arch_x86.go
index d13e12f8c..8d8bb8a8b 100644
--- a/pkg/sentry/arch/arch_x86.go
+++ b/pkg/sentry/arch/arch_x86.go
@@ -27,7 +27,6 @@ import (
 	"gvisor.dev/gvisor/pkg/log"
 	"gvisor.dev/gvisor/pkg/sentry/arch/fpu"
 	rpb "gvisor.dev/gvisor/pkg/sentry/arch/registers_go_proto"
-	"gvisor.dev/gvisor/pkg/syserror"
 )
 
 // Registers represents the CPU registers for this architecture.
@@ -354,7 +353,7 @@ func (s *State) PtraceGetRegSet(regset uintptr, dst io.Writer, maxlen int) (int,
 	switch regset {
 	case _NT_PRSTATUS:
 		if maxlen < ptraceRegistersSize {
-			return 0, syserror.EFAULT
+			return 0, linuxerr.EFAULT
 		}
 		return s.PtraceGetRegs(dst)
 	case _NT_PRFPREG:
@@ -371,7 +370,7 @@ func (s *State) PtraceSetRegSet(regset uintptr, src io.Reader, maxlen int) (int,
 	switch regset {
 	case _NT_PRSTATUS:
 		if maxlen < ptraceRegistersSize {
-			return 0, syserror.EFAULT
+			return 0, linuxerr.EFAULT
 		}
 		return s.PtraceSetRegs(src)
 	case _NT_PRFPREG:
diff --git a/pkg/sentry/arch/fpu/fpu_amd64.go b/pkg/sentry/arch/fpu/fpu_amd64.go
index f0ba26736..2e61b4955 100644
--- a/pkg/sentry/arch/fpu/fpu_amd64.go
+++ b/pkg/sentry/arch/fpu/fpu_amd64.go
@@ -21,9 +21,9 @@ import (
 
 	"golang.org/x/sys/unix"
 	"gvisor.dev/gvisor/pkg/cpuid"
+	"gvisor.dev/gvisor/pkg/errors/linuxerr"
 	"gvisor.dev/gvisor/pkg/hostarch"
 	"gvisor.dev/gvisor/pkg/sync"
-	"gvisor.dev/gvisor/pkg/syserror"
 )
 
 // initX86FPState (defined in asm files) sets up initial state.
@@ -70,7 +70,7 @@ const ptraceFPRegsSize = 512
 // PtraceGetFPRegs implements Context.PtraceGetFPRegs.
 func (s *State) PtraceGetFPRegs(dst io.Writer, maxlen int) (int, error) {
 	if maxlen < ptraceFPRegsSize {
-		return 0, syserror.EFAULT
+		return 0, linuxerr.EFAULT
 	}
 
 	return dst.Write((*s)[:ptraceFPRegsSize])
@@ -79,7 +79,7 @@ func (s *State) PtraceGetFPRegs(dst io.Writer, maxlen int) (int, error) {
 // PtraceSetFPRegs implements Context.PtraceSetFPRegs.
 func (s *State) PtraceSetFPRegs(src io.Reader, maxlen int) (int, error) {
 	if maxlen < ptraceFPRegsSize {
-		return 0, syserror.EFAULT
+		return 0, linuxerr.EFAULT
 	}
 
 	var f [ptraceFPRegsSize]byte
diff --git a/pkg/sentry/devices/tundev/tundev.go b/pkg/sentry/devices/tundev/tundev.go
index c4889017b..b4e2a6d91 100644
--- a/pkg/sentry/devices/tundev/tundev.go
+++ b/pkg/sentry/devices/tundev/tundev.go
@@ -26,7 +26,6 @@ import (
 	"gvisor.dev/gvisor/pkg/sentry/kernel"
 	"gvisor.dev/gvisor/pkg/sentry/socket/netstack"
 	"gvisor.dev/gvisor/pkg/sentry/vfs"
-	"gvisor.dev/gvisor/pkg/syserror"
 	"gvisor.dev/gvisor/pkg/tcpip/link/tun"
 	"gvisor.dev/gvisor/pkg/usermem"
 	"gvisor.dev/gvisor/pkg/waiter"
@@ -105,7 +104,7 @@ func (fd *tunFD) Ioctl(ctx context.Context, uio usermem.IO, args arch.SyscallArg
 		return 0, err
 
 	default:
-		return 0, syserror.ENOTTY
+		return 0, linuxerr.ENOTTY
 	}
 }
 
diff --git a/pkg/sentry/fs/dev/net_tun.go b/pkg/sentry/fs/dev/net_tun.go
index ae9aed7b5..1abf11142 100644
--- a/pkg/sentry/fs/dev/net_tun.go
+++ b/pkg/sentry/fs/dev/net_tun.go
@@ -25,7 +25,6 @@ import (
 	"gvisor.dev/gvisor/pkg/sentry/inet"
 	"gvisor.dev/gvisor/pkg/sentry/kernel"
 	"gvisor.dev/gvisor/pkg/sentry/socket/netstack"
-	"gvisor.dev/gvisor/pkg/syserror"
 	"gvisor.dev/gvisor/pkg/tcpip/link/tun"
 	"gvisor.dev/gvisor/pkg/usermem"
 	"gvisor.dev/gvisor/pkg/waiter"
@@ -126,7 +125,7 @@ func (n *netTunFileOperations) Ioctl(ctx context.Context, file *fs.File, io user
 		return 0, err
 
 	default:
-		return 0, syserror.ENOTTY
+		return 0, linuxerr.ENOTTY
 	}
 }
 
diff --git a/pkg/sentry/fs/dirent.go b/pkg/sentry/fs/dirent.go
index 8d7660e79..ad8ff227e 100644
--- a/pkg/sentry/fs/dirent.go
+++ b/pkg/sentry/fs/dirent.go
@@ -864,7 +864,7 @@ func direntReaddir(ctx context.Context, d *Dirent, it DirIterator, root *Dirent,
 	// Once we have written entries for "." and "..", future errors from
 	// IterateDir will be hidden.
 	if !IsDir(d.Inode.StableAttr) {
-		return 0, syserror.ENOTDIR
+		return 0, linuxerr.ENOTDIR
 	}
 
 	// This is a special case for lseek(fd, 0, SEEK_END).
diff --git a/pkg/sentry/fs/file_overlay.go b/pkg/sentry/fs/file_overlay.go
index 02208ac3d..06c07c807 100644
--- a/pkg/sentry/fs/file_overlay.go
+++ b/pkg/sentry/fs/file_overlay.go
@@ -408,7 +408,7 @@ func (f *overlayFileOperations) Ioctl(ctx context.Context, overlayFile *File, io
 		// copy up on any ioctl would be too drastic. In the future, it can have a
 		// list of ioctls that are safe to send to lower and a list that triggers a
 		// copy up.
-		return 0, syserror.ENOTTY
+		return 0, linuxerr.ENOTTY
 	}
 	return f.upper.FileOperations.Ioctl(ctx, f.upper, io, args)
 }
diff --git a/pkg/sentry/fs/fsutil/file.go b/pkg/sentry/fs/fsutil/file.go
index fe327235a..00b3bb29b 100644
--- a/pkg/sentry/fs/fsutil/file.go
+++ b/pkg/sentry/fs/fsutil/file.go
@@ -170,7 +170,7 @@ type FileNotDirReaddir struct{}
 
 // Readdir implements fs.FileOperations.FileNotDirReaddir.
 func (FileNotDirReaddir) Readdir(context.Context, *fs.File, fs.DentrySerializer) (int64, error) {
-	return 0, syserror.ENOTDIR
+	return 0, linuxerr.ENOTDIR
 }
 
 // FileNoFsync implements fs.FileOperations.Fsync for files that don't support
@@ -223,7 +223,7 @@ type FileNoIoctl struct{}
 
 // Ioctl implements fs.FileOperations.Ioctl.
 func (FileNoIoctl) Ioctl(context.Context, *fs.File, usermem.IO, arch.SyscallArguments) (uintptr, error) {
-	return 0, syserror.ENOTTY
+	return 0, linuxerr.ENOTTY
 }
 
 // FileNoSplice implements fs.FileOperations.ReadFrom and
diff --git a/pkg/sentry/fs/fsutil/inode.go b/pkg/sentry/fs/fsutil/inode.go
index d2b769007..7c2de04c1 100644
--- a/pkg/sentry/fs/fsutil/inode.go
+++ b/pkg/sentry/fs/fsutil/inode.go
@@ -237,7 +237,7 @@ func (i *InodeSimpleExtendedAttributes) SetXattr(_ context.Context, _ *fs.Inode,
 
 	_, ok := i.xattrs[name]
 	if ok && flags&linux.XATTR_CREATE != 0 {
-		return syserror.EEXIST
+		return linuxerr.EEXIST
 	}
 	if !ok && flags&linux.XATTR_REPLACE != 0 {
 		return linuxerr.ENODATA
@@ -332,47 +332,47 @@ type InodeNotDirectory struct{}
 
 // Lookup implements fs.InodeOperations.Lookup.
 func (InodeNotDirectory) Lookup(context.Context, *fs.Inode, string) (*fs.Dirent, error) {
-	return nil, syserror.ENOTDIR
+	return nil, linuxerr.ENOTDIR
 }
 
 // Create implements fs.InodeOperations.Create.
 func (InodeNotDirectory) Create(context.Context, *fs.Inode, string, fs.FileFlags, fs.FilePermissions) (*fs.File, error) {
-	return nil, syserror.ENOTDIR
+	return nil, linuxerr.ENOTDIR
 }
 
 // CreateLink implements fs.InodeOperations.CreateLink.
 func (InodeNotDirectory) CreateLink(context.Context, *fs.Inode, string, string) error {
-	return syserror.ENOTDIR
+	return linuxerr.ENOTDIR
 }
 
 // CreateHardLink implements fs.InodeOperations.CreateHardLink.
 func (InodeNotDirectory) CreateHardLink(context.Context, *fs.Inode, *fs.Inode, string) error {
-	return syserror.ENOTDIR
+	return linuxerr.ENOTDIR
 }
 
 // CreateDirectory implements fs.InodeOperations.CreateDirectory.
 func (InodeNotDirectory) CreateDirectory(context.Context, *fs.Inode, string, fs.FilePermissions) error {
-	return syserror.ENOTDIR
+	return linuxerr.ENOTDIR
 }
 
 // Bind implements fs.InodeOperations.Bind.
 func (InodeNotDirectory) Bind(context.Context, *fs.Inode, string, transport.BoundEndpoint, fs.FilePermissions) (*fs.Dirent, error) {
-	return nil, syserror.ENOTDIR
+	return nil, linuxerr.ENOTDIR
 }
 
 // CreateFifo implements fs.InodeOperations.CreateFifo.
 func (InodeNotDirectory) CreateFifo(context.Context, *fs.Inode, string, fs.FilePermissions) error {
-	return syserror.ENOTDIR
+	return linuxerr.ENOTDIR
 }
 
 // Remove implements fs.InodeOperations.Remove.
 func (InodeNotDirectory) Remove(context.Context, *fs.Inode, string) error {
-	return syserror.ENOTDIR
+	return linuxerr.ENOTDIR
 }
 
 // RemoveDirectory implements fs.InodeOperations.RemoveDirectory.
 func (InodeNotDirectory) RemoveDirectory(context.Context, *fs.Inode, string) error {
-	return syserror.ENOTDIR
+	return linuxerr.ENOTDIR
 }
 
 // Rename implements fs.FileOperations.Rename.
@@ -463,22 +463,22 @@ type InodeNoExtendedAttributes struct{}
 
 // GetXattr implements fs.InodeOperations.GetXattr.
 func (InodeNoExtendedAttributes) GetXattr(context.Context, *fs.Inode, string, uint64) (string, error) {
-	return "", syserror.EOPNOTSUPP
+	return "", linuxerr.EOPNOTSUPP
 }
 
 // SetXattr implements fs.InodeOperations.SetXattr.
 func (InodeNoExtendedAttributes) SetXattr(context.Context, *fs.Inode, string, string, uint32) error {
-	return syserror.EOPNOTSUPP
+	return linuxerr.EOPNOTSUPP
 }
 
 // ListXattr implements fs.InodeOperations.ListXattr.
 func (InodeNoExtendedAttributes) ListXattr(context.Context, *fs.Inode, uint64) (map[string]struct{}, error) {
-	return nil, syserror.EOPNOTSUPP
+	return nil, linuxerr.EOPNOTSUPP
 }
 
 // RemoveXattr implements fs.InodeOperations.RemoveXattr.
 func (InodeNoExtendedAttributes) RemoveXattr(context.Context, *fs.Inode, string) error {
-	return syserror.EOPNOTSUPP
+	return linuxerr.EOPNOTSUPP
 }
 
 // InodeNoopRelease implements fs.InodeOperations.Release as a noop.
@@ -513,7 +513,7 @@ type InodeNotAllocatable struct{}
 
 // Allocate implements fs.InodeOperations.Allocate.
 func (InodeNotAllocatable) Allocate(_ context.Context, _ *fs.Inode, _, _ int64) error {
-	return syserror.EOPNOTSUPP
+	return linuxerr.EOPNOTSUPP
 }
 
 // InodeNoopAllocate implements fs.InodeOperations.Allocate as a noop.
diff --git a/pkg/sentry/fs/gofer/path.go b/pkg/sentry/fs/gofer/path.go
index 958f46bd6..88d83060c 100644
--- a/pkg/sentry/fs/gofer/path.go
+++ b/pkg/sentry/fs/gofer/path.go
@@ -277,7 +277,7 @@ func (i *inodeOperations) Bind(ctx context.Context, dir *fs.Inode, name string,
 
 	s := i.session()
 	if s.overrides == nil {
-		return nil, syserror.EOPNOTSUPP
+		return nil, linuxerr.EOPNOTSUPP
 	}
 
 	// Stabilize the override map while creation is in progress.
diff --git a/pkg/sentry/fs/host/inode.go b/pkg/sentry/fs/host/inode.go
index 77bdf4925..5f6af2067 100644
--- a/pkg/sentry/fs/host/inode.go
+++ b/pkg/sentry/fs/host/inode.go
@@ -266,7 +266,7 @@ func (i *inodeOperations) Rename(ctx context.Context, inode *fs.Inode, oldParent
 
 // Bind implements fs.InodeOperations.Bind.
 func (i *inodeOperations) Bind(ctx context.Context, dir *fs.Inode, name string, data transport.BoundEndpoint, perm fs.FilePermissions) (*fs.Dirent, error) {
-	return nil, syserror.EOPNOTSUPP
+	return nil, linuxerr.EOPNOTSUPP
 }
 
 // BoundEndpoint implements fs.InodeOperations.BoundEndpoint.
diff --git a/pkg/sentry/fs/host/tty.go b/pkg/sentry/fs/host/tty.go
index c7010f35c..6f38b25c3 100644
--- a/pkg/sentry/fs/host/tty.go
+++ b/pkg/sentry/fs/host/tty.go
@@ -127,7 +127,7 @@ func (t *TTYFileOperations) Release(ctx context.Context) {
 func (t *TTYFileOperations) Ioctl(ctx context.Context, _ *fs.File, io usermem.IO, args arch.SyscallArguments) (uintptr, error) {
 	task := kernel.TaskFromContext(ctx)
 	if task == nil {
-		return 0, syserror.ENOTTY
+		return 0, linuxerr.ENOTTY
 	}
 
 	// Ignore arg[0].  This is the real FD:
@@ -168,7 +168,7 @@ func (t *TTYFileOperations) Ioctl(ctx context.Context, _ *fs.File, io usermem.IO
 
 		pidns := kernel.PIDNamespaceFromContext(ctx)
 		if pidns == nil {
-			return 0, syserror.ENOTTY
+			return 0, linuxerr.ENOTTY
 		}
 
 		t.mu.Lock()
@@ -193,7 +193,7 @@ func (t *TTYFileOperations) Ioctl(ctx context.Context, _ *fs.File, io usermem.IO
 			// drivers/tty/tty_io.c:tiocspgrp() converts -EIO from
 			// tty_check_change() to -ENOTTY.
 			if linuxerr.Equals(linuxerr.EIO, err) {
-				return 0, syserror.ENOTTY
+				return 0, linuxerr.ENOTTY
 			}
 			return 0, err
 		}
@@ -201,7 +201,7 @@ func (t *TTYFileOperations) Ioctl(ctx context.Context, _ *fs.File, io usermem.IO
 		// Check that calling task's process group is in the TTY
 		// session.
 		if task.ThreadGroup().Session() != t.session {
-			return 0, syserror.ENOTTY
+			return 0, linuxerr.ENOTTY
 		}
 
 		var pgIDP primitive.Int32
@@ -219,7 +219,7 @@ func (t *TTYFileOperations) Ioctl(ctx context.Context, _ *fs.File, io usermem.IO
 		pidns := task.PIDNamespace()
 		pg := pidns.ProcessGroupWithID(pgID)
 		if pg == nil {
-			return 0, syserror.ESRCH
+			return 0, linuxerr.ESRCH
 		}
 
 		// Check that new process group is in the TTY session.
@@ -284,7 +284,7 @@ func (t *TTYFileOperations) Ioctl(ctx context.Context, _ *fs.File, io usermem.IO
 		unimpl.EmitUnimplementedEvent(ctx)
 		fallthrough
 	default:
-		return 0, syserror.ENOTTY
+		return 0, linuxerr.ENOTTY
 	}
 }
 
diff --git a/pkg/sentry/fs/inotify.go b/pkg/sentry/fs/inotify.go
index cf4d6d118..ee28b0f99 100644
--- a/pkg/sentry/fs/inotify.go
+++ b/pkg/sentry/fs/inotify.go
@@ -122,7 +122,7 @@ func (*Inotify) Seek(context.Context, *File, SeekWhence, int64) (int64, error) {
 
 // Readdir implements FileOperatons.Readdir.
 func (*Inotify) Readdir(context.Context, *File, DentrySerializer) (int64, error) {
-	return 0, syserror.ENOTDIR
+	return 0, linuxerr.ENOTDIR
 }
 
 // Write implements FileOperations.Write.
@@ -223,7 +223,7 @@ func (i *Inotify) Ioctl(ctx context.Context, _ *File, io usermem.IO, args arch.S
 		return 0, err
 
 	default:
-		return 0, syserror.ENOTTY
+		return 0, linuxerr.ENOTTY
 	}
 }
 
diff --git a/pkg/sentry/fs/mounts.go b/pkg/sentry/fs/mounts.go
index 3de1c5561..10146af4e 100644
--- a/pkg/sentry/fs/mounts.go
+++ b/pkg/sentry/fs/mounts.go
@@ -24,7 +24,6 @@ import (
 	"gvisor.dev/gvisor/pkg/refs"
 	"gvisor.dev/gvisor/pkg/sentry/kernel/auth"
 	"gvisor.dev/gvisor/pkg/sync"
-	"gvisor.dev/gvisor/pkg/syserror"
 )
 
 // DefaultTraversalLimit provides a sensible default traversal limit that may
@@ -498,7 +497,7 @@ func (mns *MountNamespace) FindLink(ctx context.Context, root, wd *Dirent, path
 		if current != root {
 			if !IsDir(current.Inode.StableAttr) {
 				current.DecRef(ctx) // Drop reference from above.
-				return nil, syserror.ENOTDIR
+				return nil, linuxerr.ENOTDIR
 			}
 			if err := current.Inode.CheckPermission(ctx, PermMask{Execute: true}); err != nil {
 				current.DecRef(ctx) // Drop reference from above.
diff --git a/pkg/sentry/fs/proc/task.go b/pkg/sentry/fs/proc/task.go
index 7a7af31e4..edd62b857 100644
--- a/pkg/sentry/fs/proc/task.go
+++ b/pkg/sentry/fs/proc/task.go
@@ -47,7 +47,7 @@ import (
 // no longer in use.
 func getTaskMM(t *kernel.Task) (*mm.MemoryManager, error) {
 	if t.ExitState() == kernel.TaskExitDead {
-		return nil, syserror.ESRCH
+		return nil, linuxerr.ESRCH
 	}
 	var m *mm.MemoryManager
 	t.WithMuLocked(func(t *kernel.Task) {
@@ -64,7 +64,7 @@ func checkTaskState(t *kernel.Task) error {
 	case kernel.TaskExitZombie:
 		return linuxerr.EACCES
 	case kernel.TaskExitDead:
-		return syserror.ESRCH
+		return linuxerr.ESRCH
 	}
 	return nil
 }
@@ -282,7 +282,7 @@ func (e *exe) executable() (file fsbridge.File, err error) {
 		// (with locks held).
 		file = mm.Executable()
 		if file == nil {
-			err = syserror.ESRCH
+			err = linuxerr.ESRCH
 		}
 	})
 	return
@@ -332,14 +332,14 @@ func (e *cwd) Readlink(ctx context.Context, inode *fs.Inode) (string, error) {
 	cwd := e.t.FSContext().WorkingDirectory()
 	if cwd == nil {
 		// It could have raced with process deletion.
-		return "", syserror.ESRCH
+		return "", linuxerr.ESRCH
 	}
 	defer cwd.DecRef(ctx)
 
 	root := fs.RootFromContext(ctx)
 	if root == nil {
 		// It could have raced with process deletion.
-		return "", syserror.ESRCH
+		return "", linuxerr.ESRCH
 	}
 	defer root.DecRef(ctx)
 
@@ -474,7 +474,7 @@ func (m *memDataFile) Read(ctx context.Context, _ *fs.File, dst usermem.IOSequen
 	n, readErr := mm.CopyIn(ctx, hostarch.Addr(offset), buf, usermem.IOOpts{IgnorePermissions: true})
 	if n > 0 {
 		if _, err := dst.CopyOut(ctx, buf[:n]); err != nil {
-			return 0, syserror.EFAULT
+			return 0, linuxerr.EFAULT
 		}
 		return int64(n), nil
 	}
@@ -1004,7 +1004,7 @@ func (o *oomScoreAdj) GetFile(ctx context.Context, dirent *fs.Dirent, flags fs.F
 // Read implements fs.FileOperations.Read.
 func (f *oomScoreAdjFile) Read(ctx context.Context, _ *fs.File, dst usermem.IOSequence, offset int64) (int64, error) {
 	if f.t.ExitState() == kernel.TaskExitDead {
-		return 0, syserror.ESRCH
+		return 0, linuxerr.ESRCH
 	}
 	var buf bytes.Buffer
 	fmt.Fprintf(&buf, "%d\n", f.t.OOMScoreAdj())
@@ -1031,7 +1031,7 @@ func (f *oomScoreAdjFile) Write(ctx context.Context, _ *fs.File, src usermem.IOS
 	}
 
 	if f.t.ExitState() == kernel.TaskExitDead {
-		return 0, syserror.ESRCH
+		return 0, linuxerr.ESRCH
 	}
 	if err := f.t.SetOOMScoreAdj(v); err != nil {
 		return 0, err
diff --git a/pkg/sentry/fs/tty/master.go b/pkg/sentry/fs/tty/master.go
index 1cf869b62..88d6703a8 100644
--- a/pkg/sentry/fs/tty/master.go
+++ b/pkg/sentry/fs/tty/master.go
@@ -17,13 +17,13 @@ package tty
 import (
 	"gvisor.dev/gvisor/pkg/abi/linux"
 	"gvisor.dev/gvisor/pkg/context"
+	"gvisor.dev/gvisor/pkg/errors/linuxerr"
 	"gvisor.dev/gvisor/pkg/marshal/primitive"
 	"gvisor.dev/gvisor/pkg/sentry/arch"
 	"gvisor.dev/gvisor/pkg/sentry/fs"
 	"gvisor.dev/gvisor/pkg/sentry/fs/fsutil"
 	"gvisor.dev/gvisor/pkg/sentry/kernel"
 	"gvisor.dev/gvisor/pkg/sentry/unimpl"
-	"gvisor.dev/gvisor/pkg/syserror"
 	"gvisor.dev/gvisor/pkg/usermem"
 	"gvisor.dev/gvisor/pkg/waiter"
 )
@@ -157,7 +157,7 @@ func (mf *masterFileOperations) Ioctl(ctx context.Context, file *fs.File, io use
 	t := kernel.TaskFromContext(ctx)
 	if t == nil {
 		// ioctl(2) may only be called from a task goroutine.
-		return 0, syserror.ENOTTY
+		return 0, linuxerr.ENOTTY
 	}
 
 	switch cmd := args[1].Uint(); cmd {
@@ -201,7 +201,7 @@ func (mf *masterFileOperations) Ioctl(ctx context.Context, file *fs.File, io use
 		return mf.t.setForegroundProcessGroup(ctx, args, true /* isMaster */)
 	default:
 		maybeEmitUnimplementedEvent(ctx, cmd)
-		return 0, syserror.ENOTTY
+		return 0, linuxerr.ENOTTY
 	}
 }
 
diff --git a/pkg/sentry/fs/tty/replica.go b/pkg/sentry/fs/tty/replica.go
index 0e3eea3bd..ca5bc7535 100644
--- a/pkg/sentry/fs/tty/replica.go
+++ b/pkg/sentry/fs/tty/replica.go
@@ -17,12 +17,12 @@ package tty
 import (
 	"gvisor.dev/gvisor/pkg/abi/linux"
 	"gvisor.dev/gvisor/pkg/context"
+	"gvisor.dev/gvisor/pkg/errors/linuxerr"
 	"gvisor.dev/gvisor/pkg/marshal/primitive"
 	"gvisor.dev/gvisor/pkg/sentry/arch"
 	"gvisor.dev/gvisor/pkg/sentry/fs"
 	"gvisor.dev/gvisor/pkg/sentry/fs/fsutil"
 	"gvisor.dev/gvisor/pkg/sentry/kernel"
-	"gvisor.dev/gvisor/pkg/syserror"
 	"gvisor.dev/gvisor/pkg/usermem"
 	"gvisor.dev/gvisor/pkg/waiter"
 )
@@ -142,7 +142,7 @@ func (sf *replicaFileOperations) Ioctl(ctx context.Context, file *fs.File, io us
 	t := kernel.TaskFromContext(ctx)
 	if t == nil {
 		// ioctl(2) may only be called from a task goroutine.
-		return 0, syserror.ENOTTY
+		return 0, linuxerr.ENOTTY
 	}
 
 	switch cmd := args[1].Uint(); cmd {
@@ -179,7 +179,7 @@ func (sf *replicaFileOperations) Ioctl(ctx context.Context, file *fs.File, io us
 		return sf.si.t.setForegroundProcessGroup(ctx, args, false /* isMaster */)
 	default:
 		maybeEmitUnimplementedEvent(ctx, cmd)
-		return 0, syserror.ENOTTY
+		return 0, linuxerr.ENOTTY
 	}
 }
 
diff --git a/pkg/sentry/fsbridge/fs.go b/pkg/sentry/fsbridge/fs.go
index 36d7c110a..527bde181 100644
--- a/pkg/sentry/fsbridge/fs.go
+++ b/pkg/sentry/fsbridge/fs.go
@@ -24,7 +24,6 @@ import (
 	"gvisor.dev/gvisor/pkg/sentry/fs"
 	"gvisor.dev/gvisor/pkg/sentry/memmap"
 	"gvisor.dev/gvisor/pkg/sentry/vfs"
-	"gvisor.dev/gvisor/pkg/syserror"
 	"gvisor.dev/gvisor/pkg/usermem"
 )
 
@@ -139,7 +138,7 @@ func (l *fsLookup) OpenPath(ctx context.Context, path string, opts vfs.OpenOptio
 	// If they claim it's a directory, then make sure.
 	if strings.HasSuffix(path, "/") {
 		if d.Inode.StableAttr.Type != fs.Directory {
-			return nil, syserror.ENOTDIR
+			return nil, linuxerr.ENOTDIR
 		}
 	}
 
diff --git a/pkg/sentry/fsimpl/devpts/master.go b/pkg/sentry/fsimpl/devpts/master.go
index 1374fd3be..9a1a245dc 100644
--- a/pkg/sentry/fsimpl/devpts/master.go
+++ b/pkg/sentry/fsimpl/devpts/master.go
@@ -25,7 +25,6 @@ import (
 	"gvisor.dev/gvisor/pkg/sentry/kernel/auth"
 	"gvisor.dev/gvisor/pkg/sentry/unimpl"
 	"gvisor.dev/gvisor/pkg/sentry/vfs"
-	"gvisor.dev/gvisor/pkg/syserror"
 	"gvisor.dev/gvisor/pkg/usermem"
 	"gvisor.dev/gvisor/pkg/waiter"
 )
@@ -133,7 +132,7 @@ func (mfd *masterFileDescription) Ioctl(ctx context.Context, io usermem.IO, args
 	t := kernel.TaskFromContext(ctx)
 	if t == nil {
 		// ioctl(2) may only be called from a task goroutine.
-		return 0, syserror.ENOTTY
+		return 0, linuxerr.ENOTTY
 	}
 
 	switch cmd := args[1].Uint(); cmd {
@@ -178,7 +177,7 @@ func (mfd *masterFileDescription) Ioctl(ctx context.Context, io usermem.IO, args
 		return mfd.t.setForegroundProcessGroup(ctx, args, true /* isMaster */)
 	default:
 		maybeEmitUnimplementedEvent(ctx, cmd)
-		return 0, syserror.ENOTTY
+		return 0, linuxerr.ENOTTY
 	}
 }
 
diff --git a/pkg/sentry/fsimpl/devpts/replica.go b/pkg/sentry/fsimpl/devpts/replica.go
index 81572b991..e251897b4 100644
--- a/pkg/sentry/fsimpl/devpts/replica.go
+++ b/pkg/sentry/fsimpl/devpts/replica.go
@@ -24,7 +24,6 @@ import (
 	"gvisor.dev/gvisor/pkg/sentry/kernel"
 	"gvisor.dev/gvisor/pkg/sentry/kernel/auth"
 	"gvisor.dev/gvisor/pkg/sentry/vfs"
-	"gvisor.dev/gvisor/pkg/syserror"
 	"gvisor.dev/gvisor/pkg/usermem"
 	"gvisor.dev/gvisor/pkg/waiter"
 )
@@ -142,7 +141,7 @@ func (rfd *replicaFileDescription) Ioctl(ctx context.Context, io usermem.IO, arg
 	t := kernel.TaskFromContext(ctx)
 	if t == nil {
 		// ioctl(2) may only be called from a task goroutine.
-		return 0, syserror.ENOTTY
+		return 0, linuxerr.ENOTTY
 	}
 
 	switch cmd := args[1].Uint(); cmd {
@@ -180,7 +179,7 @@ func (rfd *replicaFileDescription) Ioctl(ctx context.Context, io usermem.IO, arg
 		return rfd.inode.t.setForegroundProcessGroup(ctx, args, false /* isMaster */)
 	default:
 		maybeEmitUnimplementedEvent(ctx, cmd)
-		return 0, syserror.ENOTTY
+		return 0, linuxerr.ENOTTY
 	}
 }
 
diff --git a/pkg/sentry/fsimpl/fuse/fusefs.go b/pkg/sentry/fsimpl/fuse/fusefs.go
index e9379e350..172cbd88f 100644
--- a/pkg/sentry/fsimpl/fuse/fusefs.go
+++ b/pkg/sentry/fsimpl/fuse/fusefs.go
@@ -394,7 +394,7 @@ func (i *inode) Open(ctx context.Context, rp *vfs.ResolvingPath, d *kernfs.Dentr
 	isDir := i.InodeAttrs.Mode().IsDir()
 	// return error if specified to open directory but inode is not a directory.
 	if !isDir && opts.Mode.IsDir() {
-		return nil, syserror.ENOTDIR
+		return nil, linuxerr.ENOTDIR
 	}
 	if opts.Flags&linux.O_LARGEFILE == 0 && atomic.LoadUint64(&i.size) > linux.MAX_NON_LFS {
 		return nil, linuxerr.EOVERFLOW
diff --git a/pkg/sentry/fsimpl/fuse/regular_file.go b/pkg/sentry/fsimpl/fuse/regular_file.go
index a0802cd32..6c4de3507 100644
--- a/pkg/sentry/fsimpl/fuse/regular_file.go
+++ b/pkg/sentry/fsimpl/fuse/regular_file.go
@@ -47,7 +47,7 @@ func (fd *regularFileFD) PRead(ctx context.Context, dst usermem.IOSequence, offs
 	//
 	// TODO(gvisor.dev/issue/2601): Support select preadv2 flags.
 	if opts.Flags&^linux.RWF_HIPRI != 0 {
-		return 0, syserror.EOPNOTSUPP
+		return 0, linuxerr.EOPNOTSUPP
 	}
 
 	size := dst.NumBytes()
@@ -151,7 +151,7 @@ func (fd *regularFileFD) pwrite(ctx context.Context, src usermem.IOSequence, off
 	//
 	// TODO(gvisor.dev/issue/2601): Support select preadv2 flags.
 	if opts.Flags&^linux.RWF_HIPRI != 0 {
-		return 0, offset, syserror.EOPNOTSUPP
+		return 0, offset, linuxerr.EOPNOTSUPP
 	}
 
 	inode := fd.inode()
diff --git a/pkg/sentry/fsimpl/gofer/filesystem.go b/pkg/sentry/fsimpl/gofer/filesystem.go
index 652e5fe77..14a97b468 100644
--- a/pkg/sentry/fsimpl/gofer/filesystem.go
+++ b/pkg/sentry/fsimpl/gofer/filesystem.go
@@ -189,7 +189,7 @@ func (fs *filesystem) renameMuUnlockAndCheckCaching(ctx context.Context, ds **[]
 // Postconditions: The returned dentry's cached metadata is up to date.
 func (fs *filesystem) stepLocked(ctx context.Context, rp *vfs.ResolvingPath, d *dentry, mayFollowSymlinks bool, ds **[]*dentry) (*dentry, bool, error) {
 	if !d.isDir() {
-		return nil, false, syserror.ENOTDIR
+		return nil, false, linuxerr.ENOTDIR
 	}
 	if err := d.checkPermissions(rp.Credentials(), vfs.MayExec); err != nil {
 		return nil, false, err
@@ -305,7 +305,7 @@ func (fs *filesystem) walkParentDirLocked(ctx context.Context, rp *vfs.Resolving
 		}
 	}
 	if !d.isDir() {
-		return nil, syserror.ENOTDIR
+		return nil, linuxerr.ENOTDIR
 	}
 	return d, nil
 }
@@ -333,7 +333,7 @@ func (fs *filesystem) resolveLocked(ctx context.Context, rp *vfs.ResolvingPath,
 		}
 	}
 	if rp.MustBeDir() && !d.isDir() {
-		return nil, syserror.ENOTDIR
+		return nil, linuxerr.ENOTDIR
 	}
 	return d, nil
 }
@@ -362,7 +362,7 @@ func (fs *filesystem) doCreateAt(ctx context.Context, rp *vfs.ResolvingPath, dir
 	}
 	name := rp.Component()
 	if name == "." || name == ".." {
-		return syserror.EEXIST
+		return linuxerr.EEXIST
 	}
 	if parent.isDeleted() {
 		return syserror.ENOENT
@@ -382,13 +382,13 @@ func (fs *filesystem) doCreateAt(ctx context.Context, rp *vfs.ResolvingPath, dir
 	// checks for writability fail below. Existence check is done by the creation
 	// RPCs themselves.
 	if child, ok := parent.children[name]; ok && child != nil {
-		return syserror.EEXIST
+		return linuxerr.EEXIST
 	}
 	checkExistence := func() error {
 		if child, err := fs.getChildLocked(ctx, parent, name, &ds); err != nil && !linuxerr.Equals(linuxerr.ENOENT, err) {
 			return err
 		} else if child != nil {
-			return syserror.EEXIST
+			return linuxerr.EEXIST
 		}
 		return nil
 	}
@@ -553,7 +553,7 @@ func (fs *filesystem) unlinkAt(ctx context.Context, rp *vfs.ResolvingPath, dir b
 			if child.cachedMetadataAuthoritative() {
 				if !child.isDir() {
 					vfsObj.AbortDeleteDentry(&child.vfsd) // +checklocksforce: see above.
-					return syserror.ENOTDIR
+					return linuxerr.ENOTDIR
 				}
 				for _, grandchild := range child.children {
 					if grandchild != nil {
@@ -574,7 +574,7 @@ func (fs *filesystem) unlinkAt(ctx context.Context, rp *vfs.ResolvingPath, dir b
 			if child != nil {
 				vfsObj.AbortDeleteDentry(&child.vfsd) // +checklocksforce: see above.
 			}
-			return syserror.ENOTDIR
+			return linuxerr.ENOTDIR
 		}
 	}
 	if parent.isSynthetic() {
@@ -645,7 +645,7 @@ func (fs *filesystem) GetDentryAt(ctx context.Context, rp *vfs.ResolvingPath, op
 	}
 	if opts.CheckSearchable {
 		if !d.isDir() {
-			return nil, syserror.ENOTDIR
+			return nil, linuxerr.ENOTDIR
 		}
 		if err := d.checkPermissions(rp.Credentials(), vfs.MayExec); err != nil {
 			return nil, err
@@ -767,7 +767,7 @@ func (fs *filesystem) MknodAt(ctx context.Context, rp *vfs.ResolvingPath, opts v
 		switch {
 		case err == nil:
 			// Step succeeded, another file exists.
-			return syserror.EEXIST
+			return linuxerr.EEXIST
 		case !linuxerr.Equals(linuxerr.ENOENT, err):
 			// Unexpected error.
 			return err
@@ -807,7 +807,7 @@ func (fs *filesystem) OpenAt(ctx context.Context, rp *vfs.ResolvingPath, opts vf
 	// support, and it isn't clear that there's any way to implement this in
 	// 9P.
 	if opts.Flags&linux.O_TMPFILE != 0 {
-		return nil, syserror.EOPNOTSUPP
+		return nil, linuxerr.EOPNOTSUPP
 	}
 	mayCreate := opts.Flags&linux.O_CREAT != 0
 	mustCreate := opts.Flags&(linux.O_CREAT|linux.O_EXCL) == (linux.O_CREAT | linux.O_EXCL)
@@ -830,7 +830,7 @@ func (fs *filesystem) OpenAt(ctx context.Context, rp *vfs.ResolvingPath, opts vf
 			return nil, syserror.EISDIR
 		}
 		if mustCreate {
-			return nil, syserror.EEXIST
+			return nil, linuxerr.EEXIST
 		}
 		if !start.cachedMetadataAuthoritative() {
 			// Refresh dentry's attributes before opening.
@@ -879,7 +879,7 @@ afterTrailingSymlink:
 		return nil, err
 	}
 	if mustCreate {
-		return nil, syserror.EEXIST
+		return nil, linuxerr.EEXIST
 	}
 	// Open existing child or follow symlink.
 	if child.isSymlink() && rp.ShouldFollowSymlink() {
@@ -894,7 +894,7 @@ afterTrailingSymlink:
 		goto afterTrailingSymlink
 	}
 	if rp.MustBeDir() && !child.isDir() {
-		return nil, syserror.ENOTDIR
+		return nil, linuxerr.ENOTDIR
 	}
 	child.IncRef()
 	defer child.DecRef(ctx)
@@ -1218,7 +1218,7 @@ func (fs *filesystem) RenameAt(ctx context.Context, rp *vfs.ResolvingPath, oldPa
 	newName := rp.Component()
 	if newName == "." || newName == ".." {
 		if opts.Flags&linux.RENAME_NOREPLACE != 0 {
-			return syserror.EEXIST
+			return linuxerr.EEXIST
 		}
 		return linuxerr.EBUSY
 	}
@@ -1272,7 +1272,7 @@ func (fs *filesystem) RenameAt(ctx context.Context, rp *vfs.ResolvingPath, oldPa
 		}
 	} else {
 		if opts.MustBeDir || rp.MustBeDir() {
-			return syserror.ENOTDIR
+			return linuxerr.ENOTDIR
 		}
 	}
 
@@ -1293,7 +1293,7 @@ func (fs *filesystem) RenameAt(ctx context.Context, rp *vfs.ResolvingPath, oldPa
 	var replacedVFSD *vfs.Dentry
 	if replaced != nil {
 		if opts.Flags&linux.RENAME_NOREPLACE != 0 {
-			return syserror.EEXIST
+			return linuxerr.EEXIST
 		}
 		replacedVFSD = &replaced.vfsd
 		if replaced.isDir() {
@@ -1305,7 +1305,7 @@ func (fs *filesystem) RenameAt(ctx context.Context, rp *vfs.ResolvingPath, oldPa
 			}
 		} else {
 			if rp.MustBeDir() || renamed.isDir() {
-				return syserror.ENOTDIR
+				return linuxerr.ENOTDIR
 			}
 		}
 	}
diff --git a/pkg/sentry/fsimpl/gofer/gofer.go b/pkg/sentry/fsimpl/gofer/gofer.go
index 2f85215d9..bcf989765 100644
--- a/pkg/sentry/fsimpl/gofer/gofer.go
+++ b/pkg/sentry/fsimpl/gofer/gofer.go
@@ -1292,7 +1292,7 @@ func (d *dentry) checkXattrPermissions(creds *auth.Credentials, name string, ats
 	// to the remote filesystem. This is inconsistent with Linux's 9p client,
 	// but consistent with other filesystems (e.g. FUSE).
 	if strings.HasPrefix(name, linux.XATTR_SECURITY_PREFIX) || strings.HasPrefix(name, linux.XATTR_SYSTEM_PREFIX) {
-		return syserror.EOPNOTSUPP
+		return linuxerr.EOPNOTSUPP
 	}
 	mode := linux.FileMode(atomic.LoadUint32(&d.mode))
 	kuid := auth.KUID(atomic.LoadUint32(&d.uid))
diff --git a/pkg/sentry/fsimpl/gofer/regular_file.go b/pkg/sentry/fsimpl/gofer/regular_file.go
index 7f458dd05..91405fe66 100644
--- a/pkg/sentry/fsimpl/gofer/regular_file.go
+++ b/pkg/sentry/fsimpl/gofer/regular_file.go
@@ -35,7 +35,6 @@ import (
 	"gvisor.dev/gvisor/pkg/sentry/usage"
 	"gvisor.dev/gvisor/pkg/sentry/vfs"
 	"gvisor.dev/gvisor/pkg/sync"
-	"gvisor.dev/gvisor/pkg/syserror"
 	"gvisor.dev/gvisor/pkg/usermem"
 )
 
@@ -132,7 +131,7 @@ func (fd *regularFileFD) PRead(ctx context.Context, dst usermem.IOSequence, offs
 	//
 	// TODO(gvisor.dev/issue/2601): Support select preadv2 flags.
 	if opts.Flags&^linux.RWF_HIPRI != 0 {
-		return 0, syserror.EOPNOTSUPP
+		return 0, linuxerr.EOPNOTSUPP
 	}
 
 	// Check for reading at EOF before calling into MM (but not under
@@ -202,7 +201,7 @@ func (fd *regularFileFD) pwrite(ctx context.Context, src usermem.IOSequence, off
 	//
 	// TODO(gvisor.dev/issue/2601): Support select pwritev2 flags.
 	if opts.Flags&^linux.RWF_HIPRI != 0 {
-		return 0, offset, syserror.EOPNOTSUPP
+		return 0, offset, linuxerr.EOPNOTSUPP
 	}
 
 	d := fd.dentry()
diff --git a/pkg/sentry/fsimpl/gofer/special_file.go b/pkg/sentry/fsimpl/gofer/special_file.go
index 690cde707..29afb67d9 100644
--- a/pkg/sentry/fsimpl/gofer/special_file.go
+++ b/pkg/sentry/fsimpl/gofer/special_file.go
@@ -191,7 +191,7 @@ func (fd *specialFileFD) PRead(ctx context.Context, dst usermem.IOSequence, offs
 	//
 	// TODO(gvisor.dev/issue/2601): Support select preadv2 flags.
 	if opts.Flags&^linux.RWF_HIPRI != 0 {
-		return 0, syserror.EOPNOTSUPP
+		return 0, linuxerr.EOPNOTSUPP
 	}
 
 	if d := fd.dentry(); d.cachedMetadataAuthoritative() {
@@ -271,7 +271,7 @@ func (fd *specialFileFD) pwrite(ctx context.Context, src usermem.IOSequence, off
 	//
 	// TODO(gvisor.dev/issue/2601): Support select pwritev2 flags.
 	if opts.Flags&^linux.RWF_HIPRI != 0 {
-		return 0, offset, syserror.EOPNOTSUPP
+		return 0, offset, linuxerr.EOPNOTSUPP
 	}
 
 	d := fd.dentry()
diff --git a/pkg/sentry/fsimpl/host/host.go b/pkg/sentry/fsimpl/host/host.go
index cdd33e0fe..9998bd0c3 100644
--- a/pkg/sentry/fsimpl/host/host.go
+++ b/pkg/sentry/fsimpl/host/host.go
@@ -492,7 +492,7 @@ func (i *inode) open(ctx context.Context, d *kernfs.Dentry, mnt *vfs.Mount, flag
 	case unix.S_IFSOCK:
 		if i.isTTY {
 			log.Warningf("cannot use host socket fd %d as TTY", i.hostFD)
-			return nil, syserror.ENOTTY
+			return nil, linuxerr.ENOTTY
 		}
 
 		ep, err := newEndpoint(ctx, i.hostFD, &i.queue)
@@ -585,7 +585,7 @@ func (f *fileDescription) PRead(ctx context.Context, dst usermem.IOSequence, off
 	//
 	// TODO(gvisor.dev/issue/2601): Support select preadv2 flags.
 	if opts.Flags&^linux.RWF_HIPRI != 0 {
-		return 0, syserror.EOPNOTSUPP
+		return 0, linuxerr.EOPNOTSUPP
 	}
 
 	i := f.inode
@@ -602,7 +602,7 @@ func (f *fileDescription) Read(ctx context.Context, dst usermem.IOSequence, opts
 	//
 	// TODO(gvisor.dev/issue/2601): Support select preadv2 flags.
 	if opts.Flags&^linux.RWF_HIPRI != 0 {
-		return 0, syserror.EOPNOTSUPP
+		return 0, linuxerr.EOPNOTSUPP
 	}
 
 	i := f.inode
@@ -701,7 +701,7 @@ func (f *fileDescription) writeToHostFD(ctx context.Context, src usermem.IOSeque
 	hostFD := f.inode.hostFD
 	// TODO(gvisor.dev/issue/2601): Support select pwritev2 flags.
 	if flags != 0 {
-		return 0, syserror.EOPNOTSUPP
+		return 0, linuxerr.EOPNOTSUPP
 	}
 	writer := hostfd.GetReadWriterAt(int32(hostFD), offset, flags)
 	n, err := src.CopyInTo(ctx, writer)
diff --git a/pkg/sentry/fsimpl/host/tty.go b/pkg/sentry/fsimpl/host/tty.go
index 5974dce64..7f6ce4ee5 100644
--- a/pkg/sentry/fsimpl/host/tty.go
+++ b/pkg/sentry/fsimpl/host/tty.go
@@ -148,7 +148,7 @@ func (t *TTYFileDescription) Write(ctx context.Context, src usermem.IOSequence,
 func (t *TTYFileDescription) Ioctl(ctx context.Context, io usermem.IO, args arch.SyscallArguments) (uintptr, error) {
 	task := kernel.TaskFromContext(ctx)
 	if task == nil {
-		return 0, syserror.ENOTTY
+		return 0, linuxerr.ENOTTY
 	}
 
 	// Ignore arg[0]. This is the real FD:
@@ -189,7 +189,7 @@ func (t *TTYFileDescription) Ioctl(ctx context.Context, io usermem.IO, args arch
 
 		pidns := kernel.PIDNamespaceFromContext(ctx)
 		if pidns == nil {
-			return 0, syserror.ENOTTY
+			return 0, linuxerr.ENOTTY
 		}
 
 		t.mu.Lock()
@@ -213,14 +213,14 @@ func (t *TTYFileDescription) Ioctl(ctx context.Context, io usermem.IO, args arch
 			// drivers/tty/tty_io.c:tiocspgrp() converts -EIO from tty_check_change()
 			// to -ENOTTY.
 			if linuxerr.Equals(linuxerr.EIO, err) {
-				return 0, syserror.ENOTTY
+				return 0, linuxerr.ENOTTY
 			}
 			return 0, err
 		}
 
 		// Check that calling task's process group is in the TTY session.
 		if task.ThreadGroup().Session() != t.session {
-			return 0, syserror.ENOTTY
+			return 0, linuxerr.ENOTTY
 		}
 
 		var pgIDP primitive.Int32
@@ -238,7 +238,7 @@ func (t *TTYFileDescription) Ioctl(ctx context.Context, io usermem.IO, args arch
 		pidns := task.PIDNamespace()
 		pg := pidns.ProcessGroupWithID(pgID)
 		if pg == nil {
-			return 0, syserror.ESRCH
+			return 0, linuxerr.ESRCH
 		}
 
 		// Check that new process group is in the TTY session.
@@ -303,7 +303,7 @@ func (t *TTYFileDescription) Ioctl(ctx context.Context, io usermem.IO, args arch
 		unimpl.EmitUnimplementedEvent(ctx)
 		fallthrough
 	default:
-		return 0, syserror.ENOTTY
+		return 0, linuxerr.ENOTTY
 	}
 }
 
diff --git a/pkg/sentry/fsimpl/kernfs/filesystem.go b/pkg/sentry/fsimpl/kernfs/filesystem.go
index 20d2526ad..a97473f7d 100644
--- a/pkg/sentry/fsimpl/kernfs/filesystem.go
+++ b/pkg/sentry/fsimpl/kernfs/filesystem.go
@@ -40,7 +40,7 @@ import (
 // Postcondition: Caller must call fs.processDeferredDecRefs*.
 func (fs *Filesystem) stepExistingLocked(ctx context.Context, rp *vfs.ResolvingPath, d *Dentry, mayFollowSymlinks bool) (*Dentry, error) {
 	if !d.isDir() {
-		return nil, syserror.ENOTDIR
+		return nil, linuxerr.ENOTDIR
 	}
 	// Directory searchable?
 	if err := d.inode.CheckPermissions(ctx, rp.Credentials(), vfs.MayExec); err != nil {
@@ -170,7 +170,7 @@ func (fs *Filesystem) walkExistingLocked(ctx context.Context, rp *vfs.ResolvingP
 		}
 	}
 	if rp.MustBeDir() && !d.isDir() {
-		return nil, syserror.ENOTDIR
+		return nil, linuxerr.ENOTDIR
 	}
 	return d, nil
 }
@@ -197,7 +197,7 @@ func (fs *Filesystem) walkParentDirLocked(ctx context.Context, rp *vfs.Resolving
 		}
 	}
 	if !d.isDir() {
-		return nil, syserror.ENOTDIR
+		return nil, linuxerr.ENOTDIR
 	}
 	return d, nil
 }
@@ -215,13 +215,13 @@ func checkCreateLocked(ctx context.Context, creds *auth.Credentials, name string
 		return err
 	}
 	if name == "." || name == ".." {
-		return syserror.EEXIST
+		return linuxerr.EEXIST
 	}
 	if len(name) > linux.NAME_MAX {
 		return linuxerr.ENAMETOOLONG
 	}
 	if _, ok := parent.children[name]; ok {
-		return syserror.EEXIST
+		return linuxerr.EEXIST
 	}
 	if parent.VFSDentry().IsDead() {
 		return syserror.ENOENT
@@ -318,7 +318,7 @@ func (fs *Filesystem) GetDentryAt(ctx context.Context, rp *vfs.ResolvingPath, op
 
 	if opts.CheckSearchable {
 		if !d.isDir() {
-			return nil, syserror.ENOTDIR
+			return nil, linuxerr.ENOTDIR
 		}
 		if err := d.inode.CheckPermissions(ctx, rp.Credentials(), vfs.MayExec); err != nil {
 			return nil, err
@@ -345,7 +345,7 @@ func (fs *Filesystem) GetParentDentryAt(ctx context.Context, rp *vfs.ResolvingPa
 // LinkAt implements vfs.FilesystemImpl.LinkAt.
 func (fs *Filesystem) LinkAt(ctx context.Context, rp *vfs.ResolvingPath, vd vfs.VirtualDentry) error {
 	if rp.Done() {
-		return syserror.EEXIST
+		return linuxerr.EEXIST
 	}
 	fs.mu.Lock()
 	defer fs.processDeferredDecRefs(ctx)
@@ -390,7 +390,7 @@ func (fs *Filesystem) LinkAt(ctx context.Context, rp *vfs.ResolvingPath, vd vfs.
 // MkdirAt implements vfs.FilesystemImpl.MkdirAt.
 func (fs *Filesystem) MkdirAt(ctx context.Context, rp *vfs.ResolvingPath, opts vfs.MkdirOptions) error {
 	if rp.Done() {
-		return syserror.EEXIST
+		return linuxerr.EEXIST
 	}
 	fs.mu.Lock()
 	defer fs.processDeferredDecRefs(ctx)
@@ -426,7 +426,7 @@ func (fs *Filesystem) MkdirAt(ctx context.Context, rp *vfs.ResolvingPath, opts v
 // MknodAt implements vfs.FilesystemImpl.MknodAt.
 func (fs *Filesystem) MknodAt(ctx context.Context, rp *vfs.ResolvingPath, opts vfs.MknodOptions) error {
 	if rp.Done() {
-		return syserror.EEXIST
+		return linuxerr.EEXIST
 	}
 	fs.mu.Lock()
 	defer fs.processDeferredDecRefs(ctx)
@@ -512,7 +512,7 @@ func (fs *Filesystem) OpenAt(ctx context.Context, rp *vfs.ResolvingPath, opts vf
 			return nil, syserror.EISDIR
 		}
 		if mustCreate {
-			return nil, syserror.EEXIST
+			return nil, linuxerr.EEXIST
 		}
 		if err := d.inode.CheckPermissions(ctx, rp.Credentials(), ats); err != nil {
 			return nil, err
@@ -577,7 +577,7 @@ afterTrailingSymlink:
 	}
 	// Open existing file or follow symlink.
 	if mustCreate {
-		return nil, syserror.EEXIST
+		return nil, linuxerr.EEXIST
 	}
 	if rp.ShouldFollowSymlink() && child.isSymlink() {
 		targetVD, targetPathname, err := child.inode.Getlink(ctx, rp.Mount())
@@ -681,7 +681,7 @@ func (fs *Filesystem) RenameAt(ctx context.Context, rp *vfs.ResolvingPath, oldPa
 	newName := rp.Component()
 	if newName == "." || newName == ".." {
 		if noReplace {
-			return syserror.EEXIST
+			return linuxerr.EEXIST
 		}
 		return linuxerr.EBUSY
 	}
@@ -693,7 +693,7 @@ func (fs *Filesystem) RenameAt(ctx context.Context, rp *vfs.ResolvingPath, oldPa
 	case linuxerr.Equals(linuxerr.EEXIST, err):
 		if noReplace {
 			// Won't overwrite existing node since RENAME_NOREPLACE was requested.
-			return syserror.EEXIST
+			return linuxerr.EEXIST
 		}
 		dst = dstDir.children[newName]
 		if dst == nil {
@@ -774,7 +774,7 @@ func (fs *Filesystem) RmdirAt(ctx context.Context, rp *vfs.ResolvingPath) error
 		return err
 	}
 	if !d.isDir() {
-		return syserror.ENOTDIR
+		return linuxerr.ENOTDIR
 	}
 	if d.inode.HasChildren() {
 		return linuxerr.ENOTEMPTY
@@ -844,7 +844,7 @@ func (fs *Filesystem) StatFSAt(ctx context.Context, rp *vfs.ResolvingPath) (linu
 // SymlinkAt implements vfs.FilesystemImpl.SymlinkAt.
 func (fs *Filesystem) SymlinkAt(ctx context.Context, rp *vfs.ResolvingPath, target string) error {
 	if rp.Done() {
-		return syserror.EEXIST
+		return linuxerr.EEXIST
 	}
 	fs.mu.Lock()
 	defer fs.processDeferredDecRefs(ctx)
diff --git a/pkg/sentry/fsimpl/kernfs/inode_impl_util.go b/pkg/sentry/fsimpl/kernfs/inode_impl_util.go
index df29c8838..a42fc79b4 100644
--- a/pkg/sentry/fsimpl/kernfs/inode_impl_util.go
+++ b/pkg/sentry/fsimpl/kernfs/inode_impl_util.go
@@ -511,7 +511,7 @@ func (o *OrderedChildren) insert(name string, child Inode, static bool) error {
 	o.mu.Lock()
 	defer o.mu.Unlock()
 	if _, ok := o.set[name]; ok {
-		return syserror.EEXIST
+		return linuxerr.EEXIST
 	}
 	s := &slot{
 		name:   name,
diff --git a/pkg/sentry/fsimpl/overlay/filesystem.go b/pkg/sentry/fsimpl/overlay/filesystem.go
index 77f9affc1..5e89928c5 100644
--- a/pkg/sentry/fsimpl/overlay/filesystem.go
+++ b/pkg/sentry/fsimpl/overlay/filesystem.go
@@ -138,7 +138,7 @@ func (fs *filesystem) renameMuUnlockAndCheckDrop(ctx context.Context, ds **[]*de
 // * !rp.Done().
 func (fs *filesystem) stepLocked(ctx context.Context, rp *vfs.ResolvingPath, d *dentry, mayFollowSymlinks bool, ds **[]*dentry) (*dentry, lookupLayer, error) {
 	if !d.isDir() {
-		return nil, lookupLayerNone, syserror.ENOTDIR
+		return nil, lookupLayerNone, linuxerr.ENOTDIR
 	}
 	if err := d.checkPermissions(rp.Credentials(), vfs.MayExec); err != nil {
 		return nil, lookupLayerNone, err
@@ -438,7 +438,7 @@ func (fs *filesystem) walkParentDirLocked(ctx context.Context, rp *vfs.Resolving
 		d = next
 	}
 	if !d.isDir() {
-		return nil, syserror.ENOTDIR
+		return nil, linuxerr.ENOTDIR
 	}
 	return d, nil
 }
@@ -458,7 +458,7 @@ func (fs *filesystem) resolveLocked(ctx context.Context, rp *vfs.ResolvingPath,
 		d = next
 	}
 	if rp.MustBeDir() && !d.isDir() {
-		return nil, syserror.ENOTDIR
+		return nil, linuxerr.ENOTDIR
 	}
 	return d, nil
 }
@@ -480,7 +480,7 @@ func (fs *filesystem) doCreateAt(ctx context.Context, rp *vfs.ResolvingPath, dir
 	}
 	name := rp.Component()
 	if name == "." || name == ".." {
-		return syserror.EEXIST
+		return linuxerr.EEXIST
 	}
 	if parent.vfsd.IsDead() {
 		return syserror.ENOENT
@@ -495,14 +495,14 @@ func (fs *filesystem) doCreateAt(ctx context.Context, rp *vfs.ResolvingPath, dir
 
 	// Determine if a file already exists at name.
 	if _, ok := parent.children[name]; ok {
-		return syserror.EEXIST
+		return linuxerr.EEXIST
 	}
 	childLayer, err := fs.lookupLayerLocked(ctx, parent, name)
 	if err != nil {
 		return err
 	}
 	if childLayer.existsInOverlay() {
-		return syserror.EEXIST
+		return linuxerr.EEXIST
 	}
 
 	if !dir && rp.MustBeDir() {
@@ -593,7 +593,7 @@ func (fs *filesystem) GetDentryAt(ctx context.Context, rp *vfs.ResolvingPath, op
 	}
 	if opts.CheckSearchable {
 		if !d.isDir() {
-			return nil, syserror.ENOTDIR
+			return nil, linuxerr.ENOTDIR
 		}
 		if err := d.checkPermissions(rp.Credentials(), vfs.MayExec); err != nil {
 			return nil, err
@@ -783,7 +783,7 @@ func (fs *filesystem) OpenAt(ctx context.Context, rp *vfs.ResolvingPath, opts vf
 			return nil, syserror.EISDIR
 		}
 		if mustCreate {
-			return nil, syserror.EEXIST
+			return nil, linuxerr.EEXIST
 		}
 		if start.isRegularFile() && mayWrite {
 			if err := start.copyUpLocked(ctx); err != nil {
@@ -823,7 +823,7 @@ afterTrailingSymlink:
 	}
 	// Open existing child or follow symlink.
 	if mustCreate {
-		return nil, syserror.EEXIST
+		return nil, linuxerr.EEXIST
 	}
 	if child.isSymlink() && rp.ShouldFollowSymlink() {
 		target, err := child.readlink(ctx)
@@ -837,7 +837,7 @@ afterTrailingSymlink:
 		goto afterTrailingSymlink
 	}
 	if rp.MustBeDir() && !child.isDir() {
-		return nil, syserror.ENOTDIR
+		return nil, linuxerr.ENOTDIR
 	}
 	if child.isRegularFile() && mayWrite {
 		if err := child.copyUpLocked(ctx); err != nil {
@@ -1034,7 +1034,7 @@ func (fs *filesystem) RenameAt(ctx context.Context, rp *vfs.ResolvingPath, oldPa
 	newName := rp.Component()
 	if newName == "." || newName == ".." {
 		if opts.Flags&linux.RENAME_NOREPLACE != 0 {
-			return syserror.EEXIST
+			return linuxerr.EEXIST
 		}
 		return linuxerr.EBUSY
 	}
@@ -1074,7 +1074,7 @@ func (fs *filesystem) RenameAt(ctx context.Context, rp *vfs.ResolvingPath, oldPa
 		}
 	} else {
 		if opts.MustBeDir || rp.MustBeDir() {
-			return syserror.ENOTDIR
+			return linuxerr.ENOTDIR
 		}
 	}
 
@@ -1100,7 +1100,7 @@ func (fs *filesystem) RenameAt(ctx context.Context, rp *vfs.ResolvingPath, oldPa
 	}
 	if replaced != nil {
 		if opts.Flags&linux.RENAME_NOREPLACE != 0 {
-			return syserror.EEXIST
+			return linuxerr.EEXIST
 		}
 		replacedVFSD = &replaced.vfsd
 		if replaced.isDir() {
@@ -1118,7 +1118,7 @@ func (fs *filesystem) RenameAt(ctx context.Context, rp *vfs.ResolvingPath, oldPa
 			}
 		} else {
 			if rp.MustBeDir() || renamed.isDir() {
-				return syserror.ENOTDIR
+				return linuxerr.ENOTDIR
 			}
 		}
 	}
@@ -1310,7 +1310,7 @@ func (fs *filesystem) RmdirAt(ctx context.Context, rp *vfs.ResolvingPath) error
 		return err
 	}
 	if !child.isDir() {
-		return syserror.ENOTDIR
+		return linuxerr.ENOTDIR
 	}
 	if err := parent.mayDelete(rp.Credentials(), child); err != nil {
 		return err
@@ -1536,7 +1536,7 @@ func (fs *filesystem) UnlinkAt(ctx context.Context, rp *vfs.ResolvingPath) error
 		return syserror.EISDIR
 	}
 	if rp.MustBeDir() {
-		return syserror.ENOTDIR
+		return linuxerr.ENOTDIR
 	}
 	vfsObj := rp.VirtualFilesystem()
 	mntns := vfs.MountNamespaceFromContext(ctx)
@@ -1659,7 +1659,7 @@ func (fs *filesystem) getXattr(ctx context.Context, d *dentry, creds *auth.Crede
 	// Return EOPNOTSUPP when fetching an overlay attribute.
 	// See fs/overlayfs/super.c:ovl_own_xattr_get().
 	if isOverlayXattr(opts.Name) {
-		return "", syserror.EOPNOTSUPP
+		return "", linuxerr.EOPNOTSUPP
 	}
 
 	// Analogous to fs/overlayfs/super.c:ovl_other_xattr_get().
@@ -1697,7 +1697,7 @@ func (fs *filesystem) setXattrLocked(ctx context.Context, d *dentry, mnt *vfs.Mo
 	// Return EOPNOTSUPP when setting an overlay attribute.
 	// See fs/overlayfs/super.c:ovl_own_xattr_set().
 	if isOverlayXattr(opts.Name) {
-		return syserror.EOPNOTSUPP
+		return linuxerr.EOPNOTSUPP
 	}
 
 	// Analogous to fs/overlayfs/super.c:ovl_other_xattr_set().
@@ -1742,7 +1742,7 @@ func (fs *filesystem) removeXattrLocked(ctx context.Context, d *dentry, mnt *vfs
 	// Linux passes the remove request to xattr_handler->set.
 	// See fs/xattr.c:vfs_removexattr().
 	if isOverlayXattr(name) {
-		return syserror.EOPNOTSUPP
+		return linuxerr.EOPNOTSUPP
 	}
 
 	if err := mnt.CheckBeginWrite(); err != nil {
diff --git a/pkg/sentry/fsimpl/proc/task.go b/pkg/sentry/fsimpl/proc/task.go
index 4577bfe2e..2717e1359 100644
--- a/pkg/sentry/fsimpl/proc/task.go
+++ b/pkg/sentry/fsimpl/proc/task.go
@@ -26,7 +26,6 @@ import (
 	"gvisor.dev/gvisor/pkg/sentry/kernel/auth"
 	"gvisor.dev/gvisor/pkg/sentry/mm"
 	"gvisor.dev/gvisor/pkg/sentry/vfs"
-	"gvisor.dev/gvisor/pkg/syserror"
 )
 
 // taskInode represents the inode for /proc/PID/ directory.
@@ -50,7 +49,7 @@ var _ kernfs.Inode = (*taskInode)(nil)
 
 func (fs *filesystem) newTaskInode(ctx context.Context, task *kernel.Task, pidns *kernel.PIDNamespace, isThreadGroup bool, fakeCgroupControllers map[string]string) (kernfs.Inode, error) {
 	if task.ExitState() == kernel.TaskExitDead {
-		return nil, syserror.ESRCH
+		return nil, linuxerr.ESRCH
 	}
 
 	contents := map[string]kernfs.Inode{
diff --git a/pkg/sentry/fsimpl/proc/task_files.go b/pkg/sentry/fsimpl/proc/task_files.go
index 3cd9628c7..5526cac1e 100644
--- a/pkg/sentry/fsimpl/proc/task_files.go
+++ b/pkg/sentry/fsimpl/proc/task_files.go
@@ -73,7 +73,7 @@ func checkTaskState(t *kernel.Task) error {
 	case kernel.TaskExitZombie:
 		return linuxerr.EACCES
 	case kernel.TaskExitDead:
-		return syserror.ESRCH
+		return linuxerr.ESRCH
 	}
 	return nil
 }
@@ -110,7 +110,7 @@ var _ dynamicInode = (*auxvData)(nil)
 // Generate implements vfs.DynamicBytesSource.Generate.
 func (d *auxvData) Generate(ctx context.Context, buf *bytes.Buffer) error {
 	if d.task.ExitState() == kernel.TaskExitDead {
-		return syserror.ESRCH
+		return linuxerr.ESRCH
 	}
 	m, err := getMMIncRef(d.task)
 	if err != nil {
@@ -160,7 +160,7 @@ var _ dynamicInode = (*cmdlineData)(nil)
 // Generate implements vfs.DynamicBytesSource.Generate.
 func (d *cmdlineData) Generate(ctx context.Context, buf *bytes.Buffer) error {
 	if d.task.ExitState() == kernel.TaskExitDead {
-		return syserror.ESRCH
+		return linuxerr.ESRCH
 	}
 	m, err := getMMIncRef(d.task)
 	if err != nil {
@@ -228,7 +228,7 @@ func (d *cmdlineData) Generate(ctx context.Context, buf *bytes.Buffer) error {
 		if int(arEnvv.Length()) > remaining {
 			end, ok := arEnvv.Start.AddLength(uint64(remaining))
 			if !ok {
-				return syserror.EFAULT
+				return linuxerr.EFAULT
 			}
 			arEnvv.End = end
 		}
@@ -486,7 +486,7 @@ func (fd *memFD) PRead(ctx context.Context, dst usermem.IOSequence, offset int64
 	n, readErr := m.CopyIn(ctx, hostarch.Addr(offset), buf, usermem.IOOpts{IgnorePermissions: true})
 	if n > 0 {
 		if _, err := dst.CopyOut(ctx, buf[:n]); err != nil {
-			return 0, syserror.EFAULT
+			return 0, linuxerr.EFAULT
 		}
 		return int64(n), nil
 	}
@@ -763,7 +763,7 @@ var _ vfs.WritableDynamicBytesSource = (*oomScoreAdj)(nil)
 // Generate implements vfs.DynamicBytesSource.Generate.
 func (o *oomScoreAdj) Generate(ctx context.Context, buf *bytes.Buffer) error {
 	if o.task.ExitState() == kernel.TaskExitDead {
-		return syserror.ESRCH
+		return linuxerr.ESRCH
 	}
 	fmt.Fprintf(buf, "%d\n", o.task.OOMScoreAdj())
 	return nil
@@ -785,7 +785,7 @@ func (o *oomScoreAdj) Write(ctx context.Context, src usermem.IOSequence, offset
 	}
 
 	if o.task.ExitState() == kernel.TaskExitDead {
-		return 0, syserror.ESRCH
+		return 0, linuxerr.ESRCH
 	}
 	if err := o.task.SetOOMScoreAdj(v); err != nil {
 		return 0, err
@@ -825,7 +825,7 @@ func (s *exeSymlink) Readlink(ctx context.Context, _ *vfs.Mount) (string, error)
 	root := vfs.RootFromContext(ctx)
 	if !root.Ok() {
 		// It could have raced with process deletion.
-		return "", syserror.ESRCH
+		return "", linuxerr.ESRCH
 	}
 	defer root.DecRef(ctx)
 
@@ -857,7 +857,7 @@ func (s *exeSymlink) Getlink(ctx context.Context, _ *vfs.Mount) (vfs.VirtualDent
 		// (with locks held).
 		exec = mm.Executable()
 		if exec == nil {
-			err = syserror.ESRCH
+			err = linuxerr.ESRCH
 		}
 	})
 	if err != nil {
@@ -901,7 +901,7 @@ func (s *cwdSymlink) Readlink(ctx context.Context, _ *vfs.Mount) (string, error)
 	root := vfs.RootFromContext(ctx)
 	if !root.Ok() {
 		// It could have raced with process deletion.
-		return "", syserror.ESRCH
+		return "", linuxerr.ESRCH
 	}
 	defer root.DecRef(ctx)
 
@@ -921,7 +921,7 @@ func (s *cwdSymlink) Getlink(ctx context.Context, _ *vfs.Mount) (vfs.VirtualDent
 	cwd := s.task.FSContext().WorkingDirectoryVFS2()
 	if !cwd.Ok() {
 		// It could have raced with process deletion.
-		return vfs.VirtualDentry{}, "", syserror.ESRCH
+		return vfs.VirtualDentry{}, "", linuxerr.ESRCH
 	}
 	return cwd, "", nil
 }
@@ -1124,7 +1124,7 @@ func (d *taskCgroupData) Generate(ctx context.Context, buf *bytes.Buffer) error
 	// exit this file show a task is in no cgroups, which is incorrect. Instead,
 	// once a task has left its cgroups, we return an error.
 	if d.task.ExitState() >= kernel.TaskExitInitiated {
-		return syserror.ESRCH
+		return linuxerr.ESRCH
 	}
 
 	d.task.GenerateProcTaskCgroup(buf)
diff --git a/pkg/sentry/fsimpl/sys/kcov.go b/pkg/sentry/fsimpl/sys/kcov.go
index d06aea162..51f0bf3d8 100644
--- a/pkg/sentry/fsimpl/sys/kcov.go
+++ b/pkg/sentry/fsimpl/sys/kcov.go
@@ -24,7 +24,6 @@ import (
 	"gvisor.dev/gvisor/pkg/sentry/kernel/auth"
 	"gvisor.dev/gvisor/pkg/sentry/memmap"
 	"gvisor.dev/gvisor/pkg/sentry/vfs"
-	"gvisor.dev/gvisor/pkg/syserror"
 	"gvisor.dev/gvisor/pkg/usermem"
 )
 
@@ -90,7 +89,7 @@ func (fd *kcovFD) Ioctl(ctx context.Context, uio usermem.IO, args arch.SyscallAr
 		}
 		return 0, fd.kcov.DisableTrace(ctx)
 	default:
-		return 0, syserror.ENOTTY
+		return 0, linuxerr.ENOTTY
 	}
 }
 
diff --git a/pkg/sentry/fsimpl/tmpfs/filesystem.go b/pkg/sentry/fsimpl/tmpfs/filesystem.go
index 2370fc95e..8b04df038 100644
--- a/pkg/sentry/fsimpl/tmpfs/filesystem.go
+++ b/pkg/sentry/fsimpl/tmpfs/filesystem.go
@@ -46,7 +46,7 @@ func (fs *filesystem) Sync(ctx context.Context) error {
 func stepLocked(ctx context.Context, rp *vfs.ResolvingPath, d *dentry) (*dentry, error) {
 	dir, ok := d.inode.impl.(*directory)
 	if !ok {
-		return nil, syserror.ENOTDIR
+		return nil, linuxerr.ENOTDIR
 	}
 	if err := d.inode.checkPermissions(rp.Credentials(), vfs.MayExec); err != nil {
 		return nil, err
@@ -113,7 +113,7 @@ func walkParentDirLocked(ctx context.Context, rp *vfs.ResolvingPath, d *dentry)
 	}
 	dir, ok := d.inode.impl.(*directory)
 	if !ok {
-		return nil, syserror.ENOTDIR
+		return nil, linuxerr.ENOTDIR
 	}
 	return dir, nil
 }
@@ -133,7 +133,7 @@ func resolveLocked(ctx context.Context, rp *vfs.ResolvingPath) (*dentry, error)
 		d = next
 	}
 	if rp.MustBeDir() && !d.inode.isDir() {
-		return nil, syserror.ENOTDIR
+		return nil, linuxerr.ENOTDIR
 	}
 	return d, nil
 }
@@ -162,13 +162,13 @@ func (fs *filesystem) doCreateAt(ctx context.Context, rp *vfs.ResolvingPath, dir
 	}
 	name := rp.Component()
 	if name == "." || name == ".." {
-		return syserror.EEXIST
+		return linuxerr.EEXIST
 	}
 	if len(name) > linux.NAME_MAX {
 		return linuxerr.ENAMETOOLONG
 	}
 	if _, ok := parentDir.childMap[name]; ok {
-		return syserror.EEXIST
+		return linuxerr.EEXIST
 	}
 	if !dir && rp.MustBeDir() {
 		return syserror.ENOENT
@@ -221,7 +221,7 @@ func (fs *filesystem) GetDentryAt(ctx context.Context, rp *vfs.ResolvingPath, op
 	}
 	if opts.CheckSearchable {
 		if !d.inode.isDir() {
-			return nil, syserror.ENOTDIR
+			return nil, linuxerr.ENOTDIR
 		}
 		if err := d.inode.checkPermissions(rp.Credentials(), vfs.MayExec); err != nil {
 			return nil, err
@@ -313,7 +313,7 @@ func (fs *filesystem) MknodAt(ctx context.Context, rp *vfs.ResolvingPath, opts v
 func (fs *filesystem) OpenAt(ctx context.Context, rp *vfs.ResolvingPath, opts vfs.OpenOptions) (*vfs.FileDescription, error) {
 	if opts.Flags&linux.O_TMPFILE != 0 {
 		// Not yet supported.
-		return nil, syserror.EOPNOTSUPP
+		return nil, linuxerr.EOPNOTSUPP
 	}
 
 	// Handle O_CREAT and !O_CREAT separately, since in the latter case we
@@ -348,7 +348,7 @@ func (fs *filesystem) OpenAt(ctx context.Context, rp *vfs.ResolvingPath, opts vf
 			return nil, syserror.EISDIR
 		}
 		if mustCreate {
-			return nil, syserror.EEXIST
+			return nil, linuxerr.EEXIST
 		}
 		start.IncRef()
 		defer start.DecRef(ctx)
@@ -402,7 +402,7 @@ afterTrailingSymlink:
 		return fd, nil
 	}
 	if mustCreate {
-		return nil, syserror.EEXIST
+		return nil, linuxerr.EEXIST
 	}
 	// Is the file mounted over?
 	if err := rp.CheckMount(ctx, &child.vfsd); err != nil {
@@ -419,7 +419,7 @@ afterTrailingSymlink:
 		goto afterTrailingSymlink
 	}
 	if rp.MustBeDir() && !child.inode.isDir() {
-		return nil, syserror.ENOTDIR
+		return nil, linuxerr.ENOTDIR
 	}
 	child.IncRef()
 	defer child.DecRef(ctx)
@@ -513,7 +513,7 @@ func (fs *filesystem) RenameAt(ctx context.Context, rp *vfs.ResolvingPath, oldPa
 	newName := rp.Component()
 	if newName == "." || newName == ".." {
 		if opts.Flags&linux.RENAME_NOREPLACE != 0 {
-			return syserror.EEXIST
+			return linuxerr.EEXIST
 		}
 		return linuxerr.EBUSY
 	}
@@ -552,7 +552,7 @@ func (fs *filesystem) RenameAt(ctx context.Context, rp *vfs.ResolvingPath, oldPa
 		}
 	} else {
 		if opts.MustBeDir || rp.MustBeDir() {
-			return syserror.ENOTDIR
+			return linuxerr.ENOTDIR
 		}
 	}
 
@@ -562,7 +562,7 @@ func (fs *filesystem) RenameAt(ctx context.Context, rp *vfs.ResolvingPath, oldPa
 	replaced, ok := newParentDir.childMap[newName]
 	if ok {
 		if opts.Flags&linux.RENAME_NOREPLACE != 0 {
-			return syserror.EEXIST
+			return linuxerr.EEXIST
 		}
 		replacedDir, ok := replaced.inode.impl.(*directory)
 		if ok {
@@ -574,10 +574,10 @@ func (fs *filesystem) RenameAt(ctx context.Context, rp *vfs.ResolvingPath, oldPa
 			}
 		} else {
 			if rp.MustBeDir() {
-				return syserror.ENOTDIR
+				return linuxerr.ENOTDIR
 			}
 			if renamed.inode.isDir() {
-				return syserror.ENOTDIR
+				return linuxerr.ENOTDIR
 			}
 		}
 	} else {
@@ -661,7 +661,7 @@ func (fs *filesystem) RmdirAt(ctx context.Context, rp *vfs.ResolvingPath) error
 	}
 	childDir, ok := child.inode.impl.(*directory)
 	if !ok {
-		return syserror.ENOTDIR
+		return linuxerr.ENOTDIR
 	}
 	if len(childDir.childMap) != 0 {
 		return linuxerr.ENOTEMPTY
@@ -767,7 +767,7 @@ func (fs *filesystem) UnlinkAt(ctx context.Context, rp *vfs.ResolvingPath) error
 		return syserror.EISDIR
 	}
 	if rp.MustBeDir() {
-		return syserror.ENOTDIR
+		return linuxerr.ENOTDIR
 	}
 	mnt := rp.Mount()
 	if err := mnt.CheckBeginWrite(); err != nil {
diff --git a/pkg/sentry/fsimpl/tmpfs/regular_file.go b/pkg/sentry/fsimpl/tmpfs/regular_file.go
index ce595ce85..0f2ac6144 100644
--- a/pkg/sentry/fsimpl/tmpfs/regular_file.go
+++ b/pkg/sentry/fsimpl/tmpfs/regular_file.go
@@ -34,7 +34,6 @@ import (
 	"gvisor.dev/gvisor/pkg/sentry/usage"
 	"gvisor.dev/gvisor/pkg/sentry/vfs"
 	"gvisor.dev/gvisor/pkg/sync"
-	"gvisor.dev/gvisor/pkg/syserror"
 	"gvisor.dev/gvisor/pkg/usermem"
 )
 
@@ -375,7 +374,7 @@ func (fd *regularFileFD) PRead(ctx context.Context, dst usermem.IOSequence, offs
 	//
 	// TODO(gvisor.dev/issue/2601): Support select preadv2 flags.
 	if opts.Flags&^(linux.RWF_HIPRI|linux.RWF_DSYNC|linux.RWF_SYNC) != 0 {
-		return 0, syserror.EOPNOTSUPP
+		return 0, linuxerr.EOPNOTSUPP
 	}
 
 	if dst.NumBytes() == 0 {
@@ -416,7 +415,7 @@ func (fd *regularFileFD) pwrite(ctx context.Context, src usermem.IOSequence, off
 	//
 	// TODO(gvisor.dev/issue/2601): Support select preadv2 flags.
 	if opts.Flags&^(linux.RWF_HIPRI|linux.RWF_DSYNC|linux.RWF_SYNC) != 0 {
-		return 0, offset, syserror.EOPNOTSUPP
+		return 0, offset, linuxerr.EOPNOTSUPP
 	}
 
 	srclen := src.NumBytes()
diff --git a/pkg/sentry/fsimpl/tmpfs/tmpfs.go b/pkg/sentry/fsimpl/tmpfs/tmpfs.go
index 4d6731852..79a54eef3 100644
--- a/pkg/sentry/fsimpl/tmpfs/tmpfs.go
+++ b/pkg/sentry/fsimpl/tmpfs/tmpfs.go
@@ -731,7 +731,7 @@ func checkXattrName(name string) error {
 	if strings.HasPrefix(name, linux.XATTR_USER_PREFIX) {
 		return nil
 	}
-	return syserror.EOPNOTSUPP
+	return linuxerr.EOPNOTSUPP
 }
 
 func (i *inode) listXattr(creds *auth.Credentials, size uint64) ([]string, error) {
diff --git a/pkg/sentry/fsimpl/verity/filesystem.go b/pkg/sentry/fsimpl/verity/filesystem.go
index 358a66072..070914a68 100644
--- a/pkg/sentry/fsimpl/verity/filesystem.go
+++ b/pkg/sentry/fsimpl/verity/filesystem.go
@@ -116,7 +116,7 @@ func (fs *filesystem) renameMuUnlockAndCheckDrop(ctx context.Context, ds **[]*de
 // * !rp.Done().
 func (fs *filesystem) stepLocked(ctx context.Context, rp *vfs.ResolvingPath, d *dentry, mayFollowSymlinks bool, ds **[]*dentry) (*dentry, error) {
 	if !d.isDir() {
-		return nil, syserror.ENOTDIR
+		return nil, linuxerr.ENOTDIR
 	}
 
 	if err := d.checkPermissions(rp.Credentials(), vfs.MayExec); err != nil {
@@ -685,7 +685,7 @@ func (fs *filesystem) walkParentDirLocked(ctx context.Context, rp *vfs.Resolving
 		d = next
 	}
 	if !d.isDir() {
-		return nil, syserror.ENOTDIR
+		return nil, linuxerr.ENOTDIR
 	}
 	return d, nil
 }
@@ -705,7 +705,7 @@ func (fs *filesystem) resolveLocked(ctx context.Context, rp *vfs.ResolvingPath,
 		d = next
 	}
 	if rp.MustBeDir() && !d.isDir() {
-		return nil, syserror.ENOTDIR
+		return nil, linuxerr.ENOTDIR
 	}
 	return d, nil
 }
@@ -737,7 +737,7 @@ func (fs *filesystem) GetDentryAt(ctx context.Context, rp *vfs.ResolvingPath, op
 	}
 	if opts.CheckSearchable {
 		if !d.isDir() {
-			return nil, syserror.ENOTDIR
+			return nil, linuxerr.ENOTDIR
 		}
 		if err := d.checkPermissions(rp.Credentials(), vfs.MayExec); err != nil {
 			return nil, err
diff --git a/pkg/sentry/fsimpl/verity/verity.go b/pkg/sentry/fsimpl/verity/verity.go
index 104a28227..c5fa9855b 100644
--- a/pkg/sentry/fsimpl/verity/verity.go
+++ b/pkg/sentry/fsimpl/verity/verity.go
@@ -882,7 +882,7 @@ func (fd *fileDescription) SetStat(ctx context.Context, opts vfs.SetStatOptions)
 // IterDirents implements vfs.FileDescriptionImpl.IterDirents.
 func (fd *fileDescription) IterDirents(ctx context.Context, cb vfs.IterDirentsCallback) error {
 	if !fd.d.isDir() {
-		return syserror.ENOTDIR
+		return linuxerr.ENOTDIR
 	}
 	fd.mu.Lock()
 	defer fd.mu.Unlock()
diff --git a/pkg/sentry/kernel/ptrace.go b/pkg/sentry/kernel/ptrace.go
index 52ea5b44b..21358ec92 100644
--- a/pkg/sentry/kernel/ptrace.go
+++ b/pkg/sentry/kernel/ptrace.go
@@ -940,7 +940,7 @@ func (t *Task) ptraceKill(target *Task) error {
 	t.tg.pidns.owner.mu.Lock()
 	defer t.tg.pidns.owner.mu.Unlock()
 	if target.Tracer() != t {
-		return syserror.ESRCH
+		return linuxerr.ESRCH
 	}
 	target.tg.signalHandlers.mu.Lock()
 	defer target.tg.signalHandlers.mu.Unlock()
@@ -964,7 +964,7 @@ func (t *Task) ptraceInterrupt(target *Task) error {
 	t.tg.pidns.owner.mu.Lock()
 	defer t.tg.pidns.owner.mu.Unlock()
 	if target.Tracer() != t {
-		return syserror.ESRCH
+		return linuxerr.ESRCH
 	}
 	if !target.ptraceSeized {
 		return syserror.EIO
@@ -1022,7 +1022,7 @@ func (t *Task) Ptrace(req int64, pid ThreadID, addr, data hostarch.Addr) error {
 	// specified by pid.
 	target := t.tg.pidns.TaskWithID(pid)
 	if target == nil {
-		return syserror.ESRCH
+		return linuxerr.ESRCH
 	}
 
 	// PTRACE_ATTACH and PTRACE_SEIZE do not require that target is not already
@@ -1047,7 +1047,7 @@ func (t *Task) Ptrace(req int64, pid ThreadID, addr, data hostarch.Addr) error {
 	t.tg.pidns.owner.mu.RLock()
 	if target.Tracer() != t {
 		t.tg.pidns.owner.mu.RUnlock()
-		return syserror.ESRCH
+		return linuxerr.ESRCH
 	}
 	if !target.ptraceFreeze() {
 		t.tg.pidns.owner.mu.RUnlock()
@@ -1055,7 +1055,7 @@ func (t *Task) Ptrace(req int64, pid ThreadID, addr, data hostarch.Addr) error {
 		// PTRACE_TRACEME, PTRACE_INTERRUPT, and PTRACE_KILL) require the
 		// tracee to be in a ptrace-stop, otherwise they fail with ESRCH." -
 		// ptrace(2)
-		return syserror.ESRCH
+		return linuxerr.ESRCH
 	}
 	t.tg.pidns.owner.mu.RUnlock()
 	// Even if the target has a ptrace-stop active, the tracee's task goroutine
diff --git a/pkg/sentry/kernel/rseq.go b/pkg/sentry/kernel/rseq.go
index 5e0618e44..de352f4f2 100644
--- a/pkg/sentry/kernel/rseq.go
+++ b/pkg/sentry/kernel/rseq.go
@@ -21,7 +21,6 @@ import (
 	"gvisor.dev/gvisor/pkg/errors/linuxerr"
 	"gvisor.dev/gvisor/pkg/hostarch"
 	"gvisor.dev/gvisor/pkg/sentry/hostcpu"
-	"gvisor.dev/gvisor/pkg/syserror"
 	"gvisor.dev/gvisor/pkg/usermem"
 )
 
@@ -76,7 +75,7 @@ func (t *Task) SetRSeq(addr hostarch.Addr, length, signature uint32) error {
 		return linuxerr.EINVAL
 	}
 	if _, ok := t.MemoryManager().CheckIORange(addr, linux.SizeOfRSeq); !ok {
-		return syserror.EFAULT
+		return linuxerr.EFAULT
 	}
 
 	t.rseqAddr = addr
@@ -93,7 +92,7 @@ func (t *Task) SetRSeq(addr hostarch.Addr, length, signature uint32) error {
 		t.Debugf("Failed to copy CPU to %#x for rseq: %v", t.rseqAddr, err)
 		t.forceSignal(linux.SIGSEGV, false /* unconditional */)
 		t.SendSignal(SignalInfoPriv(linux.SIGSEGV))
-		return syserror.EFAULT
+		return linuxerr.EFAULT
 	}
 
 	return nil
diff --git a/pkg/sentry/kernel/semaphore/semaphore.go b/pkg/sentry/kernel/semaphore/semaphore.go
index 067cc75c1..485c3a788 100644
--- a/pkg/sentry/kernel/semaphore/semaphore.go
+++ b/pkg/sentry/kernel/semaphore/semaphore.go
@@ -151,7 +151,7 @@ func (r *Registry) FindOrCreate(ctx context.Context, key, nsems int32, mode linu
 				return nil, linuxerr.EINVAL
 			}
 			if create && exclusive {
-				return nil, syserror.EEXIST
+				return nil, linuxerr.EEXIST
 			}
 			return set, nil
 		}
@@ -418,7 +418,7 @@ func (s *Set) semStat(creds *auth.Credentials, permMask fs.PermMask) (*linux.Sem
 // SetVal overrides a semaphore value, waking up waiters as needed.
 func (s *Set) SetVal(ctx context.Context, num int32, val int16, creds *auth.Credentials, pid int32) error {
 	if val < 0 || val > valueMax {
-		return syserror.ERANGE
+		return linuxerr.ERANGE
 	}
 
 	s.mu.Lock()
@@ -431,7 +431,7 @@ func (s *Set) SetVal(ctx context.Context, num int32, val int16, creds *auth.Cred
 
 	sem := s.findSem(num)
 	if sem == nil {
-		return syserror.ERANGE
+		return linuxerr.ERANGE
 	}
 
 	// TODO(gvisor.dev/issue/137): Clear undo entries in all processes.
@@ -453,7 +453,7 @@ func (s *Set) SetValAll(ctx context.Context, vals []uint16, creds *auth.Credenti
 
 	for _, val := range vals {
 		if val > valueMax {
-			return syserror.ERANGE
+			return linuxerr.ERANGE
 		}
 	}
 
@@ -489,7 +489,7 @@ func (s *Set) GetVal(num int32, creds *auth.Credentials) (int16, error) {
 
 	sem := s.findSem(num)
 	if sem == nil {
-		return 0, syserror.ERANGE
+		return 0, linuxerr.ERANGE
 	}
 	return sem.value, nil
 }
@@ -523,7 +523,7 @@ func (s *Set) GetPID(num int32, creds *auth.Credentials) (int32, error) {
 
 	sem := s.findSem(num)
 	if sem == nil {
-		return 0, syserror.ERANGE
+		return 0, linuxerr.ERANGE
 	}
 	return sem.pid, nil
 }
@@ -539,7 +539,7 @@ func (s *Set) countWaiters(num int32, creds *auth.Credentials, pred func(w *wait
 
 	sem := s.findSem(num)
 	if sem == nil {
-		return 0, syserror.ERANGE
+		return 0, linuxerr.ERANGE
 	}
 	var cnt uint16
 	for w := sem.waiters.Front(); w != nil; w = w.Next() {
@@ -625,7 +625,7 @@ func (s *Set) executeOps(ctx context.Context, ops []linux.Sembuf, pid int32) (ch
 			if op.SemOp < 0 {
 				// Handle 'wait' operation.
 				if -op.SemOp > valueMax {
-					return nil, 0, syserror.ERANGE
+					return nil, 0, linuxerr.ERANGE
 				}
 				if -op.SemOp > tmpVals[op.SemNum] {
 					// Not enough resources, must wait.
@@ -640,7 +640,7 @@ func (s *Set) executeOps(ctx context.Context, ops []linux.Sembuf, pid int32) (ch
 			} else {
 				// op.SemOp > 0: Handle 'signal' operation.
 				if tmpVals[op.SemNum] > valueMax-op.SemOp {
-					return nil, 0, syserror.ERANGE
+					return nil, 0, linuxerr.ERANGE
 				}
 			}
 
diff --git a/pkg/sentry/kernel/sessions.go b/pkg/sentry/kernel/sessions.go
index ae21a55da..f9f872522 100644
--- a/pkg/sentry/kernel/sessions.go
+++ b/pkg/sentry/kernel/sessions.go
@@ -17,7 +17,6 @@ package kernel
 import (
 	"gvisor.dev/gvisor/pkg/abi/linux"
 	"gvisor.dev/gvisor/pkg/errors/linuxerr"
-	"gvisor.dev/gvisor/pkg/syserror"
 )
 
 // SessionID is the public identifier.
@@ -373,7 +372,7 @@ func (tg *ThreadGroup) CreateProcessGroup() error {
 
 	// Check whether a process still exists or not.
 	if id == 0 {
-		return syserror.ESRCH
+		return linuxerr.ESRCH
 	}
 
 	// Per above, check for a Session leader or existing group.
diff --git a/pkg/sentry/kernel/shm/shm.go b/pkg/sentry/kernel/shm/shm.go
index 9c66b90ef..f7ac4c2b2 100644
--- a/pkg/sentry/kernel/shm/shm.go
+++ b/pkg/sentry/kernel/shm/shm.go
@@ -183,7 +183,7 @@ func (r *Registry) FindOrCreate(ctx context.Context, pid int32, key Key, size ui
 				// "IPC_CREAT and IPC_EXCL were specified in shmflg, but a
 				// shared memory segment already exists for key."
 				//  - man shmget(2)
-				return nil, syserror.EEXIST
+				return nil, linuxerr.EEXIST
 			}
 
 			shm.IncRef()
@@ -512,7 +512,7 @@ func (*Shm) CopyMapping(context.Context, memmap.MappingSpace, hostarch.AddrRange
 func (s *Shm) Translate(ctx context.Context, required, optional memmap.MappableRange, at hostarch.AccessType) ([]memmap.Translation, error) {
 	var err error
 	if required.End > s.fr.Length() {
-		err = &memmap.BusError{syserror.EFAULT}
+		err = &memmap.BusError{linuxerr.EFAULT}
 	}
 	if source := optional.Intersect(memmap.MappableRange{0, s.fr.Length()}); source.Length() != 0 {
 		return []memmap.Translation{
diff --git a/pkg/sentry/kernel/task_signals.go b/pkg/sentry/kernel/task_signals.go
index 7ce57ec97..7065ac79c 100644
--- a/pkg/sentry/kernel/task_signals.go
+++ b/pkg/sentry/kernel/task_signals.go
@@ -371,7 +371,7 @@ func (t *Task) Sigtimedwait(set linux.SignalSet, timeout time.Duration) (*linux.
 //
 // The following errors may be returned:
 //
-//	syserror.ESRCH - The task has exited.
+//	linuxerr.ESRCH - The task has exited.
 //	linuxerr.EINVAL - The signal is not valid.
 //	linuxerr.EAGAIN - THe signal is realtime, and cannot be queued.
 //
@@ -408,7 +408,7 @@ func (t *Task) sendSignalLocked(info *linux.SignalInfo, group bool) error {
 
 func (t *Task) sendSignalTimerLocked(info *linux.SignalInfo, group bool, timer *IntervalTimer) error {
 	if t.exitState == TaskExitDead {
-		return syserror.ESRCH
+		return linuxerr.ESRCH
 	}
 	sig := linux.Signal(info.Signo)
 	if sig == 0 {
diff --git a/pkg/sentry/kernel/task_usermem.go b/pkg/sentry/kernel/task_usermem.go
index 7935d15a6..8e2c36598 100644
--- a/pkg/sentry/kernel/task_usermem.go
+++ b/pkg/sentry/kernel/task_usermem.go
@@ -133,7 +133,7 @@ func (t *Task) CopyOutIovecs(addr hostarch.Addr, src hostarch.AddrRangeSeq) erro
 	case 8:
 		const itemLen = 16
 		if _, ok := addr.AddLength(uint64(src.NumRanges()) * itemLen); !ok {
-			return syserror.EFAULT
+			return linuxerr.EFAULT
 		}
 
 		b := t.CopyScratchBuffer(itemLen)
@@ -191,7 +191,7 @@ func (t *Task) CopyInIovecs(addr hostarch.Addr, numIovecs int) (hostarch.AddrRan
 	case 8:
 		const itemLen = 16
 		if _, ok := addr.AddLength(uint64(numIovecs) * itemLen); !ok {
-			return hostarch.AddrRangeSeq{}, syserror.EFAULT
+			return hostarch.AddrRangeSeq{}, linuxerr.EFAULT
 		}
 
 		b := t.CopyScratchBuffer(itemLen)
@@ -207,7 +207,7 @@ func (t *Task) CopyInIovecs(addr hostarch.Addr, numIovecs int) (hostarch.AddrRan
 			}
 			ar, ok := t.MemoryManager().CheckIORange(base, int64(length))
 			if !ok {
-				return hostarch.AddrRangeSeq{}, syserror.EFAULT
+				return hostarch.AddrRangeSeq{}, linuxerr.EFAULT
 			}
 
 			if numIovecs == 1 {
@@ -253,7 +253,7 @@ func (t *Task) SingleIOSequence(addr hostarch.Addr, length int, opts usermem.IOO
 	}
 	ar, ok := t.MemoryManager().CheckIORange(addr, int64(length))
 	if !ok {
-		return usermem.IOSequence{}, syserror.EFAULT
+		return usermem.IOSequence{}, linuxerr.EFAULT
 	}
 	return usermem.IOSequence{
 		IO:    t.MemoryManager(),
@@ -313,7 +313,7 @@ func (cc *taskCopyContext) getMemoryManager() (*mm.MemoryManager, error) {
 	tmm := cc.t.MemoryManager()
 	cc.t.mu.Unlock()
 	if !tmm.IncUsers() {
-		return nil, syserror.EFAULT
+		return nil, linuxerr.EFAULT
 	}
 	return tmm, nil
 }
diff --git a/pkg/sentry/kernel/thread_group.go b/pkg/sentry/kernel/thread_group.go
index b61142e53..2eda15303 100644
--- a/pkg/sentry/kernel/thread_group.go
+++ b/pkg/sentry/kernel/thread_group.go
@@ -26,7 +26,6 @@ import (
 	"gvisor.dev/gvisor/pkg/sentry/limits"
 	"gvisor.dev/gvisor/pkg/sentry/usage"
 	"gvisor.dev/gvisor/pkg/sync"
-	"gvisor.dev/gvisor/pkg/syserror"
 )
 
 // A ThreadGroup is a logical grouping of tasks that has widespread
@@ -420,7 +419,7 @@ func (tg *ThreadGroup) ReleaseControllingTTY(tty *TTY) error {
 
 	if tg.tty == nil || tg.tty != tty {
 		tg.signalHandlers.mu.Unlock()
-		return syserror.ENOTTY
+		return linuxerr.ENOTTY
 	}
 
 	// "If the process was session leader, then send SIGHUP and SIGCONT to
@@ -474,7 +473,7 @@ func (tg *ThreadGroup) ForegroundProcessGroup(tty *TTY) (int32, error) {
 	// "When fd does not refer to the controlling terminal of the calling
 	// process, -1 is returned" - tcgetpgrp(3)
 	if tg.tty != tty {
-		return -1, syserror.ENOTTY
+		return -1, linuxerr.ENOTTY
 	}
 
 	return int32(tg.processGroup.session.foreground.id), nil
@@ -497,7 +496,7 @@ func (tg *ThreadGroup) SetForegroundProcessGroup(tty *TTY, pgid ProcessGroupID)
 
 	// tty must be the controlling terminal.
 	if tg.tty != tty {
-		return -1, syserror.ENOTTY
+		return -1, linuxerr.ENOTTY
 	}
 
 	// pgid must be positive.
@@ -509,7 +508,7 @@ func (tg *ThreadGroup) SetForegroundProcessGroup(tty *TTY, pgid ProcessGroupID)
 	// pid namespaces.
 	pg, ok := tg.pidns.processGroups[pgid]
 	if !ok {
-		return -1, syserror.ESRCH
+		return -1, linuxerr.ESRCH
 	}
 
 	// pg must be part of this process's session.
diff --git a/pkg/sentry/mm/aio_context.go b/pkg/sentry/mm/aio_context.go
index 23747daba..b7f765cd7 100644
--- a/pkg/sentry/mm/aio_context.go
+++ b/pkg/sentry/mm/aio_context.go
@@ -23,7 +23,6 @@ import (
 	"gvisor.dev/gvisor/pkg/sentry/pgalloc"
 	"gvisor.dev/gvisor/pkg/sentry/usage"
 	"gvisor.dev/gvisor/pkg/sync"
-	"gvisor.dev/gvisor/pkg/syserror"
 	"gvisor.dev/gvisor/pkg/usermem"
 )
 
@@ -306,7 +305,7 @@ func (m *aioMappable) AddMapping(_ context.Context, _ memmap.MappingSpace, ar ho
 	// Don't allow mappings to be expanded (in Linux, fs/aio.c:aio_ring_mmap()
 	// sets VM_DONTEXPAND).
 	if offset != 0 || uint64(ar.Length()) != aioRingBufferSize {
-		return syserror.EFAULT
+		return linuxerr.EFAULT
 	}
 	return nil
 }
@@ -320,7 +319,7 @@ func (m *aioMappable) CopyMapping(ctx context.Context, ms memmap.MappingSpace, s
 	// Don't allow mappings to be expanded (in Linux, fs/aio.c:aio_ring_mmap()
 	// sets VM_DONTEXPAND).
 	if offset != 0 || uint64(dstAR.Length()) != aioRingBufferSize {
-		return syserror.EFAULT
+		return linuxerr.EFAULT
 	}
 	// Require that the mapping correspond to a live AIOContext. Compare
 	// Linux's fs/aio.c:aio_ring_mremap().
@@ -351,7 +350,7 @@ func (m *aioMappable) CopyMapping(ctx context.Context, ms memmap.MappingSpace, s
 func (m *aioMappable) Translate(ctx context.Context, required, optional memmap.MappableRange, at hostarch.AccessType) ([]memmap.Translation, error) {
 	var err error
 	if required.End > m.fr.Length() {
-		err = &memmap.BusError{syserror.EFAULT}
+		err = &memmap.BusError{linuxerr.EFAULT}
 	}
 	if source := optional.Intersect(memmap.MappableRange{0, m.fr.Length()}); source.Length() != 0 {
 		return []memmap.Translation{
diff --git a/pkg/sentry/mm/io.go b/pkg/sentry/mm/io.go
index 16f318ab3..5fcfeb473 100644
--- a/pkg/sentry/mm/io.go
+++ b/pkg/sentry/mm/io.go
@@ -16,10 +16,10 @@ package mm
 
 import (
 	"gvisor.dev/gvisor/pkg/context"
+	"gvisor.dev/gvisor/pkg/errors/linuxerr"
 	"gvisor.dev/gvisor/pkg/hostarch"
 	"gvisor.dev/gvisor/pkg/safemem"
 	"gvisor.dev/gvisor/pkg/sentry/platform"
-	"gvisor.dev/gvisor/pkg/syserror"
 	"gvisor.dev/gvisor/pkg/usermem"
 )
 
@@ -97,14 +97,14 @@ func translateIOError(ctx context.Context, err error) error {
 	if logIOErrors {
 		ctx.Debugf("MM I/O error: %v", err)
 	}
-	return syserror.EFAULT
+	return linuxerr.EFAULT
 }
 
 // CopyOut implements usermem.IO.CopyOut.
 func (mm *MemoryManager) CopyOut(ctx context.Context, addr hostarch.Addr, src []byte, opts usermem.IOOpts) (int, error) {
 	ar, ok := mm.CheckIORange(addr, int64(len(src)))
 	if !ok {
-		return 0, syserror.EFAULT
+		return 0, linuxerr.EFAULT
 	}
 
 	if len(src) == 0 {
@@ -147,7 +147,7 @@ func (mm *MemoryManager) asCopyOut(ctx context.Context, addr hostarch.Addr, src
 func (mm *MemoryManager) CopyIn(ctx context.Context, addr hostarch.Addr, dst []byte, opts usermem.IOOpts) (int, error) {
 	ar, ok := mm.CheckIORange(addr, int64(len(dst)))
 	if !ok {
-		return 0, syserror.EFAULT
+		return 0, linuxerr.EFAULT
 	}
 
 	if len(dst) == 0 {
@@ -190,7 +190,7 @@ func (mm *MemoryManager) asCopyIn(ctx context.Context, addr hostarch.Addr, dst [
 func (mm *MemoryManager) ZeroOut(ctx context.Context, addr hostarch.Addr, toZero int64, opts usermem.IOOpts) (int64, error) {
 	ar, ok := mm.CheckIORange(addr, toZero)
 	if !ok {
-		return 0, syserror.EFAULT
+		return 0, linuxerr.EFAULT
 	}
 
 	if toZero == 0 {
@@ -231,7 +231,7 @@ func (mm *MemoryManager) asZeroOut(ctx context.Context, addr hostarch.Addr, toZe
 // CopyOutFrom implements usermem.IO.CopyOutFrom.
 func (mm *MemoryManager) CopyOutFrom(ctx context.Context, ars hostarch.AddrRangeSeq, src safemem.Reader, opts usermem.IOOpts) (int64, error) {
 	if !mm.checkIOVec(ars) {
-		return 0, syserror.EFAULT
+		return 0, linuxerr.EFAULT
 	}
 
 	if ars.NumBytes() == 0 {
@@ -276,7 +276,7 @@ func (mm *MemoryManager) CopyOutFrom(ctx context.Context, ars hostarch.AddrRange
 // CopyInTo implements usermem.IO.CopyInTo.
 func (mm *MemoryManager) CopyInTo(ctx context.Context, ars hostarch.AddrRangeSeq, dst safemem.Writer, opts usermem.IOOpts) (int64, error) {
 	if !mm.checkIOVec(ars) {
-		return 0, syserror.EFAULT
+		return 0, linuxerr.EFAULT
 	}
 
 	if ars.NumBytes() == 0 {
@@ -314,7 +314,7 @@ func (mm *MemoryManager) CopyInTo(ctx context.Context, ars hostarch.AddrRangeSeq
 func (mm *MemoryManager) SwapUint32(ctx context.Context, addr hostarch.Addr, new uint32, opts usermem.IOOpts) (uint32, error) {
 	ar, ok := mm.CheckIORange(addr, 4)
 	if !ok {
-		return 0, syserror.EFAULT
+		return 0, linuxerr.EFAULT
 	}
 
 	// Do AddressSpace IO if applicable.
@@ -339,7 +339,7 @@ func (mm *MemoryManager) SwapUint32(ctx context.Context, addr hostarch.Addr, new
 	_, err := mm.withInternalMappings(ctx, ar, hostarch.ReadWrite, opts.IgnorePermissions, func(ims safemem.BlockSeq) (uint64, error) {
 		if ims.NumBlocks() != 1 || ims.NumBytes() != 4 {
 			// Atomicity is unachievable across mappings.
-			return 0, syserror.EFAULT
+			return 0, linuxerr.EFAULT
 		}
 		im := ims.Head()
 		var err error
@@ -357,7 +357,7 @@ func (mm *MemoryManager) SwapUint32(ctx context.Context, addr hostarch.Addr, new
 func (mm *MemoryManager) CompareAndSwapUint32(ctx context.Context, addr hostarch.Addr, old, new uint32, opts usermem.IOOpts) (uint32, error) {
 	ar, ok := mm.CheckIORange(addr, 4)
 	if !ok {
-		return 0, syserror.EFAULT
+		return 0, linuxerr.EFAULT
 	}
 
 	// Do AddressSpace IO if applicable.
@@ -382,7 +382,7 @@ func (mm *MemoryManager) CompareAndSwapUint32(ctx context.Context, addr hostarch
 	_, err := mm.withInternalMappings(ctx, ar, hostarch.ReadWrite, opts.IgnorePermissions, func(ims safemem.BlockSeq) (uint64, error) {
 		if ims.NumBlocks() != 1 || ims.NumBytes() != 4 {
 			// Atomicity is unachievable across mappings.
-			return 0, syserror.EFAULT
+			return 0, linuxerr.EFAULT
 		}
 		im := ims.Head()
 		var err error
@@ -400,7 +400,7 @@ func (mm *MemoryManager) CompareAndSwapUint32(ctx context.Context, addr hostarch
 func (mm *MemoryManager) LoadUint32(ctx context.Context, addr hostarch.Addr, opts usermem.IOOpts) (uint32, error) {
 	ar, ok := mm.CheckIORange(addr, 4)
 	if !ok {
-		return 0, syserror.EFAULT
+		return 0, linuxerr.EFAULT
 	}
 
 	// Do AddressSpace IO if applicable.
@@ -425,7 +425,7 @@ func (mm *MemoryManager) LoadUint32(ctx context.Context, addr hostarch.Addr, opt
 	_, err := mm.withInternalMappings(ctx, ar, hostarch.Read, opts.IgnorePermissions, func(ims safemem.BlockSeq) (uint64, error) {
 		if ims.NumBlocks() != 1 || ims.NumBytes() != 4 {
 			// Atomicity is unachievable across mappings.
-			return 0, syserror.EFAULT
+			return 0, linuxerr.EFAULT
 		}
 		im := ims.Head()
 		var err error
diff --git a/pkg/sentry/mm/pma.go b/pkg/sentry/mm/pma.go
index 5583f62b2..9f4cc238f 100644
--- a/pkg/sentry/mm/pma.go
+++ b/pkg/sentry/mm/pma.go
@@ -18,12 +18,12 @@ import (
 	"fmt"
 
 	"gvisor.dev/gvisor/pkg/context"
+	"gvisor.dev/gvisor/pkg/errors/linuxerr"
 	"gvisor.dev/gvisor/pkg/hostarch"
 	"gvisor.dev/gvisor/pkg/safecopy"
 	"gvisor.dev/gvisor/pkg/safemem"
 	"gvisor.dev/gvisor/pkg/sentry/memmap"
 	"gvisor.dev/gvisor/pkg/sentry/usage"
-	"gvisor.dev/gvisor/pkg/syserror"
 )
 
 // existingPMAsLocked checks that pmas exist for all addresses in ar, and
@@ -116,7 +116,7 @@ func (mm *MemoryManager) getPMAsLocked(ctx context.Context, vseg vmaIterator, ar
 	var alignerr error
 	if !ok {
 		end = ar.End.RoundDown()
-		alignerr = syserror.EFAULT
+		alignerr = linuxerr.EFAULT
 	}
 	ar = hostarch.AddrRange{ar.Start.RoundDown(), end}
 
@@ -162,7 +162,7 @@ func (mm *MemoryManager) getVecPMAsLocked(ctx context.Context, ars hostarch.Addr
 		var alignerr error
 		if !ok {
 			end = ar.End.RoundDown()
-			alignerr = syserror.EFAULT
+			alignerr = linuxerr.EFAULT
 		}
 		ar = hostarch.AddrRange{ar.Start.RoundDown(), end}
 
diff --git a/pkg/sentry/mm/special_mappable.go b/pkg/sentry/mm/special_mappable.go
index feafe76c1..69c6e77a7 100644
--- a/pkg/sentry/mm/special_mappable.go
+++ b/pkg/sentry/mm/special_mappable.go
@@ -21,7 +21,6 @@ import (
 	"gvisor.dev/gvisor/pkg/sentry/memmap"
 	"gvisor.dev/gvisor/pkg/sentry/pgalloc"
 	"gvisor.dev/gvisor/pkg/sentry/usage"
-	"gvisor.dev/gvisor/pkg/syserror"
 )
 
 // SpecialMappable implements memmap.MappingIdentity and memmap.Mappable with
@@ -95,7 +94,7 @@ func (*SpecialMappable) CopyMapping(context.Context, memmap.MappingSpace, hostar
 func (m *SpecialMappable) Translate(ctx context.Context, required, optional memmap.MappableRange, at hostarch.AccessType) ([]memmap.Translation, error) {
 	var err error
 	if required.End > m.fr.Length() {
-		err = &memmap.BusError{syserror.EFAULT}
+		err = &memmap.BusError{linuxerr.EFAULT}
 	}
 	if source := optional.Intersect(memmap.MappableRange{0, m.fr.Length()}); source.Length() != 0 {
 		return []memmap.Translation{
diff --git a/pkg/sentry/mm/syscalls.go b/pkg/sentry/mm/syscalls.go
index b51ec6aa7..256eb4afb 100644
--- a/pkg/sentry/mm/syscalls.go
+++ b/pkg/sentry/mm/syscalls.go
@@ -37,7 +37,7 @@ import (
 func (mm *MemoryManager) HandleUserFault(ctx context.Context, addr hostarch.Addr, at hostarch.AccessType, sp hostarch.Addr) error {
 	ar, ok := addr.RoundDown().ToRange(hostarch.PageSize)
 	if !ok {
-		return syserror.EFAULT
+		return linuxerr.EFAULT
 	}
 
 	// Don't bother trying existingPMAsLocked; in most cases, if we did have
@@ -357,7 +357,7 @@ func (mm *MemoryManager) MRemap(ctx context.Context, oldAddr hostarch.Addr, oldS
 	// All cases require that a vma exists at oldAddr.
 	vseg := mm.vmas.FindSegment(oldAddr)
 	if !vseg.Ok() {
-		return 0, syserror.EFAULT
+		return 0, linuxerr.EFAULT
 	}
 
 	// Behavior matrix:
@@ -404,7 +404,7 @@ func (mm *MemoryManager) MRemap(ctx context.Context, oldAddr hostarch.Addr, oldS
 
 		// Check that oldEnd maps to the same vma as oldAddr.
 		if vseg.End() < oldEnd {
-			return 0, syserror.EFAULT
+			return 0, linuxerr.EFAULT
 		}
 		// "Grow" the existing vma by creating a new mergeable one.
 		vma := vseg.ValuePtr()
@@ -494,7 +494,7 @@ func (mm *MemoryManager) MRemap(ctx context.Context, oldAddr hostarch.Addr, oldS
 
 	// Check that oldEnd maps to the same vma as oldAddr.
 	if vseg.End() < oldEnd {
-		return 0, syserror.EFAULT
+		return 0, linuxerr.EFAULT
 	}
 
 	// Check against RLIMIT_AS.
@@ -732,7 +732,7 @@ func (mm *MemoryManager) Brk(ctx context.Context, addr hostarch.Addr) (hostarch.
 	if !ok {
 		addr = mm.brk.End
 		mm.mappingMu.Unlock()
-		return addr, syserror.EFAULT
+		return addr, linuxerr.EFAULT
 	}
 
 	switch {
@@ -972,7 +972,7 @@ func (mm *MemoryManager) NumaPolicy(addr hostarch.Addr) (linux.NumaPolicy, uint6
 	defer mm.mappingMu.RUnlock()
 	vseg := mm.vmas.FindSegment(addr)
 	if !vseg.Ok() {
-		return 0, 0, syserror.EFAULT
+		return 0, 0, linuxerr.EFAULT
 	}
 	vma := vseg.ValuePtr()
 	return vma.numaPolicy, vma.numaNodemask, nil
@@ -1005,7 +1005,7 @@ func (mm *MemoryManager) SetNumaPolicy(addr hostarch.Addr, length uint64, policy
 		if !vseg.Ok() || lastEnd < vseg.Start() {
 			// "EFAULT: ... there was an unmapped hole in the specified memory
 			// range specified [sic] by addr and len." - mbind(2)
-			return syserror.EFAULT
+			return linuxerr.EFAULT
 		}
 		vseg = mm.vmas.Isolate(vseg, ar)
 		vma := vseg.ValuePtr()
@@ -1193,7 +1193,7 @@ func (mm *MemoryManager) MSync(ctx context.Context, addr hostarch.Addr, length u
 func (mm *MemoryManager) GetSharedFutexKey(ctx context.Context, addr hostarch.Addr) (futex.Key, error) {
 	ar, ok := addr.ToRange(4) // sizeof(int32).
 	if !ok {
-		return futex.Key{}, syserror.EFAULT
+		return futex.Key{}, linuxerr.EFAULT
 	}
 
 	mm.mappingMu.RLock()
diff --git a/pkg/sentry/mm/vma.go b/pkg/sentry/mm/vma.go
index a411eabbf..5f8ab7ca3 100644
--- a/pkg/sentry/mm/vma.go
+++ b/pkg/sentry/mm/vma.go
@@ -289,7 +289,7 @@ func (mm *MemoryManager) getVMAsLocked(ctx context.Context, ar hostarch.AddrRang
 		vma := vseg.ValuePtr()
 		if addr < vseg.Start() {
 			// TODO(jamieliu): Implement vma.growsDown here.
-			return vbegin, vgap, syserror.EFAULT
+			return vbegin, vgap, linuxerr.EFAULT
 		}
 
 		perms := vma.effectivePerms
@@ -309,7 +309,7 @@ func (mm *MemoryManager) getVMAsLocked(ctx context.Context, ar hostarch.AddrRang
 	}
 
 	// Ran out of vmas before ar.End.
-	return vbegin, vgap, syserror.EFAULT
+	return vbegin, vgap, linuxerr.EFAULT
 }
 
 // getVecVMAsLocked ensures that vmas exist for all addresses in ars, and
diff --git a/pkg/sentry/socket/hostinet/socket_unsafe.go b/pkg/sentry/socket/hostinet/socket_unsafe.go
index 86dc879d5..ccf4f534d 100644
--- a/pkg/sentry/socket/hostinet/socket_unsafe.go
+++ b/pkg/sentry/socket/hostinet/socket_unsafe.go
@@ -20,12 +20,12 @@ import (
 	"golang.org/x/sys/unix"
 	"gvisor.dev/gvisor/pkg/abi/linux"
 	"gvisor.dev/gvisor/pkg/context"
+	"gvisor.dev/gvisor/pkg/errors/linuxerr"
 	"gvisor.dev/gvisor/pkg/hostarch"
 	"gvisor.dev/gvisor/pkg/sentry/arch"
 	"gvisor.dev/gvisor/pkg/sentry/kernel"
 	"gvisor.dev/gvisor/pkg/sentry/socket"
 	"gvisor.dev/gvisor/pkg/syserr"
-	"gvisor.dev/gvisor/pkg/syserror"
 	"gvisor.dev/gvisor/pkg/usermem"
 )
 
@@ -85,7 +85,7 @@ func ioctl(ctx context.Context, fd int, io usermem.IO, args arch.SyscallArgument
 		_, err := ifr.CopyOut(cc, args[2].Pointer())
 		return 0, err
 	default:
-		return 0, syserror.ENOTTY
+		return 0, linuxerr.ENOTTY
 	}
 }
 
diff --git a/pkg/sentry/socket/hostinet/socket_vfs2.go b/pkg/sentry/socket/hostinet/socket_vfs2.go
index 2c5c3e2cd..cd6e34ecc 100644
--- a/pkg/sentry/socket/hostinet/socket_vfs2.go
+++ b/pkg/sentry/socket/hostinet/socket_vfs2.go
@@ -27,7 +27,6 @@ import (
 	"gvisor.dev/gvisor/pkg/sentry/socket"
 	"gvisor.dev/gvisor/pkg/sentry/vfs"
 	"gvisor.dev/gvisor/pkg/syserr"
-	"gvisor.dev/gvisor/pkg/syserror"
 	"gvisor.dev/gvisor/pkg/usermem"
 	"gvisor.dev/gvisor/pkg/waiter"
 )
@@ -114,7 +113,7 @@ func (s *socketVFS2) Read(ctx context.Context, dst usermem.IOSequence, opts vfs.
 	// All flags other than RWF_NOWAIT should be ignored.
 	// TODO(gvisor.dev/issue/2601): Support RWF_NOWAIT.
 	if opts.Flags != 0 {
-		return 0, syserror.EOPNOTSUPP
+		return 0, linuxerr.EOPNOTSUPP
 	}
 
 	reader := hostfd.GetReadWriterAt(int32(s.fd), -1, opts.Flags)
@@ -133,7 +132,7 @@ func (s *socketVFS2) Write(ctx context.Context, src usermem.IOSequence, opts vfs
 	// All flags other than RWF_NOWAIT should be ignored.
 	// TODO(gvisor.dev/issue/2601): Support RWF_NOWAIT.
 	if opts.Flags != 0 {
-		return 0, syserror.EOPNOTSUPP
+		return 0, linuxerr.EOPNOTSUPP
 	}
 
 	writer := hostfd.GetReadWriterAt(int32(s.fd), -1, opts.Flags)
diff --git a/pkg/sentry/socket/netlink/socket.go b/pkg/sentry/socket/netlink/socket.go
index d53f23a9a..5c3ae26f8 100644
--- a/pkg/sentry/socket/netlink/socket.go
+++ b/pkg/sentry/socket/netlink/socket.go
@@ -214,7 +214,7 @@ func (s *socketOpsCommon) ConnectedPasscred() bool {
 // Ioctl implements fs.FileOperations.Ioctl.
 func (*Socket) Ioctl(context.Context, *fs.File, usermem.IO, arch.SyscallArguments) (uintptr, error) {
 	// TODO(b/68878065): no ioctls supported.
-	return 0, syserror.ENOTTY
+	return 0, linuxerr.ENOTTY
 }
 
 // ExtractSockAddr extracts the SockAddrNetlink from b.
diff --git a/pkg/sentry/socket/netlink/socket_vfs2.go b/pkg/sentry/socket/netlink/socket_vfs2.go
index 1a98a1373..4d3cdea62 100644
--- a/pkg/sentry/socket/netlink/socket_vfs2.go
+++ b/pkg/sentry/socket/netlink/socket_vfs2.go
@@ -25,7 +25,6 @@ import (
 	"gvisor.dev/gvisor/pkg/sentry/socket/unix/transport"
 	"gvisor.dev/gvisor/pkg/sentry/vfs"
 	"gvisor.dev/gvisor/pkg/syserr"
-	"gvisor.dev/gvisor/pkg/syserror"
 	"gvisor.dev/gvisor/pkg/tcpip"
 	"gvisor.dev/gvisor/pkg/usermem"
 	"gvisor.dev/gvisor/pkg/waiter"
@@ -109,7 +108,7 @@ func (s *SocketVFS2) EventUnregister(e *waiter.Entry) {
 // Ioctl implements vfs.FileDescriptionImpl.
 func (*SocketVFS2) Ioctl(context.Context, usermem.IO, arch.SyscallArguments) (uintptr, error) {
 	// TODO(b/68878065): no ioctls supported.
-	return 0, syserror.ENOTTY
+	return 0, linuxerr.ENOTTY
 }
 
 // PRead implements vfs.FileDescriptionImpl.
@@ -122,7 +121,7 @@ func (s *SocketVFS2) Read(ctx context.Context, dst usermem.IOSequence, opts vfs.
 	// All flags other than RWF_NOWAIT should be ignored.
 	// TODO(gvisor.dev/issue/2601): Support RWF_NOWAIT.
 	if opts.Flags != 0 {
-		return 0, syserror.EOPNOTSUPP
+		return 0, linuxerr.EOPNOTSUPP
 	}
 
 	if dst.NumBytes() == 0 {
@@ -143,7 +142,7 @@ func (s *SocketVFS2) Write(ctx context.Context, src usermem.IOSequence, opts vfs
 	// All flags other than RWF_NOWAIT should be ignored.
 	// TODO(gvisor.dev/issue/2601): Support RWF_NOWAIT.
 	if opts.Flags != 0 {
-		return 0, syserror.EOPNOTSUPP
+		return 0, linuxerr.EOPNOTSUPP
 	}
 
 	n, err := s.sendMsg(ctx, src, nil, 0, socket.ControlMessages{})
diff --git a/pkg/sentry/socket/netstack/netstack.go b/pkg/sentry/socket/netstack/netstack.go
index 11f75628c..5d5c4ebbd 100644
--- a/pkg/sentry/socket/netstack/netstack.go
+++ b/pkg/sentry/socket/netstack/netstack.go
@@ -3016,7 +3016,7 @@ func Ioctl(ctx context.Context, ep commonEndpoint, io usermem.IO, args arch.Sysc
 		unimpl.EmitUnimplementedEvent(ctx)
 	}
 
-	return 0, syserror.ENOTTY
+	return 0, linuxerr.ENOTTY
 }
 
 // interfaceIoctl implements interface requests.
diff --git a/pkg/sentry/socket/netstack/netstack_vfs2.go b/pkg/sentry/socket/netstack/netstack_vfs2.go
index 30f3ad153..edc160b1b 100644
--- a/pkg/sentry/socket/netstack/netstack_vfs2.go
+++ b/pkg/sentry/socket/netstack/netstack_vfs2.go
@@ -17,6 +17,7 @@ package netstack
 import (
 	"gvisor.dev/gvisor/pkg/abi/linux"
 	"gvisor.dev/gvisor/pkg/context"
+	"gvisor.dev/gvisor/pkg/errors/linuxerr"
 	"gvisor.dev/gvisor/pkg/hostarch"
 	"gvisor.dev/gvisor/pkg/marshal"
 	"gvisor.dev/gvisor/pkg/marshal/primitive"
@@ -104,7 +105,7 @@ func (s *SocketVFS2) Read(ctx context.Context, dst usermem.IOSequence, opts vfs.
 	// All flags other than RWF_NOWAIT should be ignored.
 	// TODO(gvisor.dev/issue/2601): Support RWF_NOWAIT.
 	if opts.Flags != 0 {
-		return 0, syserror.EOPNOTSUPP
+		return 0, linuxerr.EOPNOTSUPP
 	}
 
 	if dst.NumBytes() == 0 {
@@ -125,7 +126,7 @@ func (s *SocketVFS2) Write(ctx context.Context, src usermem.IOSequence, opts vfs
 	// All flags other than RWF_NOWAIT should be ignored.
 	// TODO(gvisor.dev/issue/2601): Support RWF_NOWAIT.
 	if opts.Flags != 0 {
-		return 0, syserror.EOPNOTSUPP
+		return 0, linuxerr.EOPNOTSUPP
 	}
 
 	r := src.Reader(ctx)
diff --git a/pkg/sentry/socket/unix/unix_vfs2.go b/pkg/sentry/socket/unix/unix_vfs2.go
index 5401f99e6..8c5075a1c 100644
--- a/pkg/sentry/socket/unix/unix_vfs2.go
+++ b/pkg/sentry/socket/unix/unix_vfs2.go
@@ -30,7 +30,6 @@ import (
 	"gvisor.dev/gvisor/pkg/sentry/socket/unix/transport"
 	"gvisor.dev/gvisor/pkg/sentry/vfs"
 	"gvisor.dev/gvisor/pkg/syserr"
-	"gvisor.dev/gvisor/pkg/syserror"
 	"gvisor.dev/gvisor/pkg/tcpip"
 	"gvisor.dev/gvisor/pkg/usermem"
 	"gvisor.dev/gvisor/pkg/waiter"
@@ -262,7 +261,7 @@ func (s *SocketVFS2) Read(ctx context.Context, dst usermem.IOSequence, opts vfs.
 	// All flags other than RWF_NOWAIT should be ignored.
 	// TODO(gvisor.dev/issue/2601): Support RWF_NOWAIT.
 	if opts.Flags != 0 {
-		return 0, syserror.EOPNOTSUPP
+		return 0, linuxerr.EOPNOTSUPP
 	}
 
 	if dst.NumBytes() == 0 {
@@ -291,7 +290,7 @@ func (s *SocketVFS2) Write(ctx context.Context, src usermem.IOSequence, opts vfs
 	// All flags other than RWF_NOWAIT should be ignored.
 	// TODO(gvisor.dev/issue/2601): Support RWF_NOWAIT.
 	if opts.Flags != 0 {
-		return 0, syserror.EOPNOTSUPP
+		return 0, linuxerr.EOPNOTSUPP
 	}
 
 	t := kernel.TaskFromContext(ctx)
diff --git a/pkg/sentry/syscalls/linux/linux64.go b/pkg/sentry/syscalls/linux/linux64.go
index 8935ea5b8..f1cb5a2c8 100644
--- a/pkg/sentry/syscalls/linux/linux64.go
+++ b/pkg/sentry/syscalls/linux/linux64.go
@@ -356,12 +356,12 @@ var AMD64 = &kernel.SyscallTable{
 		300: syscalls.ErrorWithEvent("fanotify_init", syserror.ENOSYS, "Needs CONFIG_FANOTIFY", nil),
 		301: syscalls.ErrorWithEvent("fanotify_mark", syserror.ENOSYS, "Needs CONFIG_FANOTIFY", nil),
 		302: syscalls.Supported("prlimit64", Prlimit64),
-		303: syscalls.Error("name_to_handle_at", syserror.EOPNOTSUPP, "Not supported by gVisor filesystems", nil),
-		304: syscalls.Error("open_by_handle_at", syserror.EOPNOTSUPP, "Not supported by gVisor filesystems", nil),
+		303: syscalls.Error("name_to_handle_at", linuxerr.EOPNOTSUPP, "Not supported by gVisor filesystems", nil),
+		304: syscalls.Error("open_by_handle_at", linuxerr.EOPNOTSUPP, "Not supported by gVisor filesystems", nil),
 		305: syscalls.CapError("clock_adjtime", linux.CAP_SYS_TIME, "", nil),
 		306: syscalls.PartiallySupported("syncfs", Syncfs, "Depends on backing file system.", nil),
 		307: syscalls.PartiallySupported("sendmmsg", SendMMsg, "Not all flags and control messages are supported.", nil),
-		308: syscalls.ErrorWithEvent("setns", syserror.EOPNOTSUPP, "Needs filesystem support", []string{"gvisor.dev/issue/140"}), // TODO(b/29354995)
+		308: syscalls.ErrorWithEvent("setns", linuxerr.EOPNOTSUPP, "Needs filesystem support", []string{"gvisor.dev/issue/140"}), // TODO(b/29354995)
 		309: syscalls.Supported("getcpu", Getcpu),
 		310: syscalls.ErrorWithEvent("process_vm_readv", syserror.ENOSYS, "", []string{"gvisor.dev/issue/158"}),
 		311: syscalls.ErrorWithEvent("process_vm_writev", syserror.ENOSYS, "", []string{"gvisor.dev/issue/158"}),
@@ -678,11 +678,11 @@ var ARM64 = &kernel.SyscallTable{
 		261: syscalls.Supported("prlimit64", Prlimit64),
 		262: syscalls.ErrorWithEvent("fanotify_init", syserror.ENOSYS, "Needs CONFIG_FANOTIFY", nil),
 		263: syscalls.ErrorWithEvent("fanotify_mark", syserror.ENOSYS, "Needs CONFIG_FANOTIFY", nil),
-		264: syscalls.Error("name_to_handle_at", syserror.EOPNOTSUPP, "Not supported by gVisor filesystems", nil),
-		265: syscalls.Error("open_by_handle_at", syserror.EOPNOTSUPP, "Not supported by gVisor filesystems", nil),
+		264: syscalls.Error("name_to_handle_at", linuxerr.EOPNOTSUPP, "Not supported by gVisor filesystems", nil),
+		265: syscalls.Error("open_by_handle_at", linuxerr.EOPNOTSUPP, "Not supported by gVisor filesystems", nil),
 		266: syscalls.CapError("clock_adjtime", linux.CAP_SYS_TIME, "", nil),
 		267: syscalls.PartiallySupported("syncfs", Syncfs, "Depends on backing file system.", nil),
-		268: syscalls.ErrorWithEvent("setns", syserror.EOPNOTSUPP, "Needs filesystem support", []string{"gvisor.dev/issue/140"}), // TODO(b/29354995)
+		268: syscalls.ErrorWithEvent("setns", linuxerr.EOPNOTSUPP, "Needs filesystem support", []string{"gvisor.dev/issue/140"}), // TODO(b/29354995)
 		269: syscalls.PartiallySupported("sendmmsg", SendMMsg, "Not all flags and control messages are supported.", nil),
 		270: syscalls.ErrorWithEvent("process_vm_readv", syserror.ENOSYS, "", []string{"gvisor.dev/issue/158"}),
 		271: syscalls.ErrorWithEvent("process_vm_writev", syserror.ENOSYS, "", []string{"gvisor.dev/issue/158"}),
diff --git a/pkg/sentry/syscalls/linux/sys_capability.go b/pkg/sentry/syscalls/linux/sys_capability.go
index 8318be001..1e714503c 100644
--- a/pkg/sentry/syscalls/linux/sys_capability.go
+++ b/pkg/sentry/syscalls/linux/sys_capability.go
@@ -20,7 +20,6 @@ import (
 	"gvisor.dev/gvisor/pkg/sentry/arch"
 	"gvisor.dev/gvisor/pkg/sentry/kernel"
 	"gvisor.dev/gvisor/pkg/sentry/kernel/auth"
-	"gvisor.dev/gvisor/pkg/syserror"
 )
 
 func lookupCaps(t *kernel.Task, tid kernel.ThreadID) (permitted, inheritable, effective auth.CapabilitySet, err error) {
@@ -32,7 +31,7 @@ func lookupCaps(t *kernel.Task, tid kernel.ThreadID) (permitted, inheritable, ef
 		t = t.PIDNamespace().TaskWithID(tid)
 	}
 	if t == nil {
-		err = syserror.ESRCH
+		err = linuxerr.ESRCH
 		return
 	}
 	creds := t.Credentials()
diff --git a/pkg/sentry/syscalls/linux/sys_file.go b/pkg/sentry/syscalls/linux/sys_file.go
index a08449824..3528d325f 100644
--- a/pkg/sentry/syscalls/linux/sys_file.go
+++ b/pkg/sentry/syscalls/linux/sys_file.go
@@ -85,7 +85,7 @@ func fileOpOn(t *kernel.Task, dirFD int32, path string, resolve bool, fn func(ro
 		rel = f.Dirent
 		if !fs.IsDir(rel.Inode.StableAttr) {
 			f.DecRef(t)
-			return syserror.ENOTDIR
+			return linuxerr.ENOTDIR
 		}
 	}
 
@@ -167,11 +167,11 @@ func openAt(t *kernel.Task, dirFD int32, addr hostarch.Addr, flags uint) (fd uin
 		} else {
 			// If O_DIRECTORY is set, but the file is not a directory, then fail.
 			if fileFlags.Directory {
-				return syserror.ENOTDIR
+				return linuxerr.ENOTDIR
 			}
 			// If it's a directory, then make sure.
 			if dirPath {
-				return syserror.ENOTDIR
+				return linuxerr.ENOTDIR
 			}
 		}
 
@@ -220,7 +220,7 @@ func mknodAt(t *kernel.Task, dirFD int32, addr hostarch.Addr, mode linux.FileMod
 
 	return fileOpAt(t, dirFD, path, func(root *fs.Dirent, d *fs.Dirent, name string, _ uint) error {
 		if !fs.IsDir(d.Inode.StableAttr) {
-			return syserror.ENOTDIR
+			return linuxerr.ENOTDIR
 		}
 
 		// Do we have the appropriate permissions on the parent?
@@ -261,7 +261,7 @@ func mknodAt(t *kernel.Task, dirFD int32, addr hostarch.Addr, mode linux.FileMod
 			// Instead of emulating this seemingly useless behaviour, we'll
 			// indicate that the filesystem doesn't support the creation of
 			// sockets.
-			return syserror.EOPNOTSUPP
+			return linuxerr.EOPNOTSUPP
 
 		case linux.ModeCharacterDevice:
 			fallthrough
@@ -326,7 +326,7 @@ func createAt(t *kernel.Task, dirFD int32, addr hostarch.Addr, flags uint, mode
 		)
 		for {
 			if !fs.IsDir(parent.Inode.StableAttr) {
-				return syserror.ENOTDIR
+				return linuxerr.ENOTDIR
 			}
 
 			// Start by looking up the dirent at 'name'.
@@ -340,7 +340,7 @@ func createAt(t *kernel.Task, dirFD int32, addr hostarch.Addr, flags uint, mode
 			// O_EXCL flag was passed, then we can immediately
 			// return EEXIST.
 			if flags&linux.O_EXCL != 0 {
-				return syserror.EEXIST
+				return linuxerr.EEXIST
 			}
 
 			// If we have a non-symlink, then we can proceed.
@@ -685,7 +685,7 @@ func Getcwd(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.Syscal
 
 	// Note this is >= because we need a terminator.
 	if uint(len(s)) >= size {
-		return 0, nil, syserror.ERANGE
+		return 0, nil, linuxerr.ERANGE
 	}
 
 	// Copy out the path name for the node.
@@ -715,7 +715,7 @@ func Chroot(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.Syscal
 	return 0, nil, fileOpOn(t, linux.AT_FDCWD, path, true /* resolve */, func(root *fs.Dirent, d *fs.Dirent, _ uint) error {
 		// Is it a directory?
 		if !fs.IsDir(d.Inode.StableAttr) {
-			return syserror.ENOTDIR
+			return linuxerr.ENOTDIR
 		}
 
 		// Does it have execute permissions?
@@ -740,7 +740,7 @@ func Chdir(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.Syscall
 	return 0, nil, fileOpOn(t, linux.AT_FDCWD, path, true /* resolve */, func(root *fs.Dirent, d *fs.Dirent, _ uint) error {
 		// Is it a directory?
 		if !fs.IsDir(d.Inode.StableAttr) {
-			return syserror.ENOTDIR
+			return linuxerr.ENOTDIR
 		}
 
 		// Does it have execute permissions?
@@ -765,7 +765,7 @@ func Fchdir(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.Syscal
 
 	// Is it a directory?
 	if !fs.IsDir(file.Dirent.Inode.StableAttr) {
-		return 0, nil, syserror.ENOTDIR
+		return 0, nil, linuxerr.ENOTDIR
 	}
 
 	// Does it have execute permissions?
@@ -1067,21 +1067,21 @@ func Fcntl(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.Syscall
 		case linux.F_OWNER_TID:
 			task := t.PIDNamespace().TaskWithID(kernel.ThreadID(owner.PID))
 			if task == nil {
-				return 0, nil, syserror.ESRCH
+				return 0, nil, linuxerr.ESRCH
 			}
 			a.SetOwnerTask(t, task)
 			return 0, nil, nil
 		case linux.F_OWNER_PID:
 			tg := t.PIDNamespace().ThreadGroupWithID(kernel.ThreadID(owner.PID))
 			if tg == nil {
-				return 0, nil, syserror.ESRCH
+				return 0, nil, linuxerr.ESRCH
 			}
 			a.SetOwnerThreadGroup(t, tg)
 			return 0, nil, nil
 		case linux.F_OWNER_PGRP:
 			pg := t.PIDNamespace().ProcessGroupWithID(kernel.ProcessGroupID(owner.PID))
 			if pg == nil {
-				return 0, nil, syserror.ESRCH
+				return 0, nil, linuxerr.ESRCH
 			}
 			a.SetOwnerProcessGroup(t, pg)
 			return 0, nil, nil
@@ -1173,7 +1173,7 @@ func mkdirAt(t *kernel.Task, dirFD int32, addr hostarch.Addr, mode linux.FileMod
 
 	return fileOpAt(t, dirFD, path, func(root *fs.Dirent, d *fs.Dirent, name string, _ uint) error {
 		if !fs.IsDir(d.Inode.StableAttr) {
-			return syserror.ENOTDIR
+			return linuxerr.ENOTDIR
 		}
 
 		// Does this directory exist already?
@@ -1183,7 +1183,7 @@ func mkdirAt(t *kernel.Task, dirFD int32, addr hostarch.Addr, mode linux.FileMod
 		case err == nil:
 			// The directory existed.
 			defer f.DecRef(t)
-			return syserror.EEXIST
+			return linuxerr.EEXIST
 		case linuxerr.Equals(linuxerr.EACCES, err):
 			// Permission denied while walking to the directory.
 			return err
@@ -1230,7 +1230,7 @@ func rmdirAt(t *kernel.Task, dirFD int32, addr hostarch.Addr) error {
 
 	return fileOpAt(t, dirFD, path, func(root *fs.Dirent, d *fs.Dirent, name string, _ uint) error {
 		if !fs.IsDir(d.Inode.StableAttr) {
-			return syserror.ENOTDIR
+			return linuxerr.ENOTDIR
 		}
 
 		// Linux returns different ernos when the path ends in single
@@ -1278,7 +1278,7 @@ func symlinkAt(t *kernel.Task, dirFD int32, newAddr hostarch.Addr, oldAddr hosta
 
 	return fileOpAt(t, dirFD, newPath, func(root *fs.Dirent, d *fs.Dirent, name string, _ uint) error {
 		if !fs.IsDir(d.Inode.StableAttr) {
-			return syserror.ENOTDIR
+			return linuxerr.ENOTDIR
 		}
 
 		// Make sure we have write permissions on the parent directory.
@@ -1368,7 +1368,7 @@ func linkAt(t *kernel.Task, oldDirFD int32, oldAddr hostarch.Addr, newDirFD int3
 		// Resolve the target directory.
 		return fileOpAt(t, newDirFD, newPath, func(root *fs.Dirent, newParent *fs.Dirent, newName string, _ uint) error {
 			if !fs.IsDir(newParent.Inode.StableAttr) {
-				return syserror.ENOTDIR
+				return linuxerr.ENOTDIR
 			}
 
 			// Make sure we have write permissions on the parent directory.
@@ -1389,7 +1389,7 @@ func linkAt(t *kernel.Task, oldDirFD int32, oldAddr hostarch.Addr, newDirFD int3
 		// Next resolve newDirFD and newAddr to the parent dirent and name.
 		return fileOpAt(t, newDirFD, newPath, func(root *fs.Dirent, newParent *fs.Dirent, newName string, _ uint) error {
 			if !fs.IsDir(newParent.Inode.StableAttr) {
-				return syserror.ENOTDIR
+				return linuxerr.ENOTDIR
 			}
 
 			// Make sure we have write permissions on the parent directory.
@@ -1520,7 +1520,7 @@ func unlinkAt(t *kernel.Task, dirFD int32, addr hostarch.Addr) error {
 
 	return fileOpAt(t, dirFD, path, func(root *fs.Dirent, d *fs.Dirent, name string, _ uint) error {
 		if !fs.IsDir(d.Inode.StableAttr) {
-			return syserror.ENOTDIR
+			return linuxerr.ENOTDIR
 		}
 
 		if err := d.MayDelete(t, root, name); err != nil {
@@ -2059,7 +2059,7 @@ func renameAt(t *kernel.Task, oldDirFD int32, oldAddr hostarch.Addr, newDirFD in
 
 	return fileOpAt(t, oldDirFD, oldPath, func(root *fs.Dirent, oldParent *fs.Dirent, oldName string, _ uint) error {
 		if !fs.IsDir(oldParent.Inode.StableAttr) {
-			return syserror.ENOTDIR
+			return linuxerr.ENOTDIR
 		}
 
 		// Rename rejects paths that end in ".", "..", or empty (i.e.
@@ -2071,7 +2071,7 @@ func renameAt(t *kernel.Task, oldDirFD int32, oldAddr hostarch.Addr, newDirFD in
 
 		return fileOpAt(t, newDirFD, newPath, func(root *fs.Dirent, newParent *fs.Dirent, newName string, _ uint) error {
 			if !fs.IsDir(newParent.Inode.StableAttr) {
-				return syserror.ENOTDIR
+				return linuxerr.ENOTDIR
 			}
 
 			// Rename rejects paths that end in ".", "..", or empty
diff --git a/pkg/sentry/syscalls/linux/sys_futex.go b/pkg/sentry/syscalls/linux/sys_futex.go
index a66514c7f..717cec04d 100644
--- a/pkg/sentry/syscalls/linux/sys_futex.go
+++ b/pkg/sentry/syscalls/linux/sys_futex.go
@@ -317,7 +317,7 @@ func GetRobustList(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel
 	ot := t
 	if tid != 0 {
 		if ot = t.PIDNamespace().TaskWithID(kernel.ThreadID(tid)); ot == nil {
-			return 0, nil, syserror.ESRCH
+			return 0, nil, linuxerr.ESRCH
 		}
 	}
 
diff --git a/pkg/sentry/syscalls/linux/sys_inotify.go b/pkg/sentry/syscalls/linux/sys_inotify.go
index 3738e4cfb..b7ad1922e 100644
--- a/pkg/sentry/syscalls/linux/sys_inotify.go
+++ b/pkg/sentry/syscalls/linux/sys_inotify.go
@@ -21,7 +21,6 @@ import (
 	"gvisor.dev/gvisor/pkg/sentry/fs"
 	"gvisor.dev/gvisor/pkg/sentry/fs/anon"
 	"gvisor.dev/gvisor/pkg/sentry/kernel"
-	"gvisor.dev/gvisor/pkg/syserror"
 )
 
 const allFlags = int(linux.IN_NONBLOCK | linux.IN_CLOEXEC)
@@ -109,7 +108,7 @@ func InotifyAddWatch(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kern
 	err = fileOpOn(t, linux.AT_FDCWD, path, resolve, func(root *fs.Dirent, dirent *fs.Dirent, _ uint) error {
 		// "IN_ONLYDIR: Only watch pathname if it is a directory." -- inotify(7)
 		if onlyDir := mask&linux.IN_ONLYDIR != 0; onlyDir && !fs.IsDir(dirent.Inode.StableAttr) {
-			return syserror.ENOTDIR
+			return linuxerr.ENOTDIR
 		}
 
 		// Copy out to the return frame.
diff --git a/pkg/sentry/syscalls/linux/sys_mempolicy.go b/pkg/sentry/syscalls/linux/sys_mempolicy.go
index adb513a3d..6e7bcb868 100644
--- a/pkg/sentry/syscalls/linux/sys_mempolicy.go
+++ b/pkg/sentry/syscalls/linux/sys_mempolicy.go
@@ -22,7 +22,6 @@ import (
 	"gvisor.dev/gvisor/pkg/hostarch"
 	"gvisor.dev/gvisor/pkg/sentry/arch"
 	"gvisor.dev/gvisor/pkg/sentry/kernel"
-	"gvisor.dev/gvisor/pkg/syserror"
 	"gvisor.dev/gvisor/pkg/usermem"
 )
 
@@ -90,7 +89,7 @@ func copyOutNodemask(t *kernel.Task, addr hostarch.Addr, maxnode uint32, val uin
 	if bits > 64 {
 		remAddr, ok := addr.AddLength(8)
 		if !ok {
-			return syserror.EFAULT
+			return linuxerr.EFAULT
 		}
 		remUint64 := (bits - 1) / 64
 		if _, err := t.MemoryManager().ZeroOut(t, remAddr, int64(remUint64)*8, usermem.IOOpts{
diff --git a/pkg/sentry/syscalls/linux/sys_random.go b/pkg/sentry/syscalls/linux/sys_random.go
index ec6c80de5..f86e87bc7 100644
--- a/pkg/sentry/syscalls/linux/sys_random.go
+++ b/pkg/sentry/syscalls/linux/sys_random.go
@@ -24,7 +24,6 @@ import (
 	"gvisor.dev/gvisor/pkg/safemem"
 	"gvisor.dev/gvisor/pkg/sentry/arch"
 	"gvisor.dev/gvisor/pkg/sentry/kernel"
-	"gvisor.dev/gvisor/pkg/syserror"
 	"gvisor.dev/gvisor/pkg/usermem"
 )
 
@@ -55,7 +54,7 @@ func GetRandom(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.Sys
 	}
 	ar, ok := addr.ToRange(uint64(length))
 	if !ok {
-		return 0, nil, syserror.EFAULT
+		return 0, nil, linuxerr.EFAULT
 	}
 
 	// "If the urandom source has been initialized, reads of up to 256 bytes
diff --git a/pkg/sentry/syscalls/linux/sys_read.go b/pkg/sentry/syscalls/linux/sys_read.go
index 50f98fc8f..b54a3a11f 100644
--- a/pkg/sentry/syscalls/linux/sys_read.go
+++ b/pkg/sentry/syscalls/linux/sys_read.go
@@ -266,7 +266,7 @@ func Preadv2(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.Sysca
 	// Note: gVisor does not implement the RWF_HIPRI feature, but the flag is
 	// accepted as a valid flag argument for preadv2.
 	if flags&^linux.RWF_VALID != 0 {
-		return 0, nil, syserror.EOPNOTSUPP
+		return 0, nil, linuxerr.EOPNOTSUPP
 	}
 
 	// Read the iovecs that specify the destination of the read.
diff --git a/pkg/sentry/syscalls/linux/sys_rlimit.go b/pkg/sentry/syscalls/linux/sys_rlimit.go
index 06f51f49c..a12e1c915 100644
--- a/pkg/sentry/syscalls/linux/sys_rlimit.go
+++ b/pkg/sentry/syscalls/linux/sys_rlimit.go
@@ -159,7 +159,7 @@ func Setrlimit(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.Sys
 		return 0, nil, err
 	}
 	if _, err := rlim.CopyIn(t, addr); err != nil {
-		return 0, nil, syserror.EFAULT
+		return 0, nil, linuxerr.EFAULT
 	}
 	_, err = prlimit64(t, resource, rlim.toLimit())
 	return 0, nil, err
@@ -180,7 +180,7 @@ func Prlimit64(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.Sys
 	if newRlimAddr != 0 {
 		var nrl rlimit64
 		if err := nrl.copyIn(t, newRlimAddr); err != nil {
-			return 0, nil, syserror.EFAULT
+			return 0, nil, linuxerr.EFAULT
 		}
 		newLim = nrl.toLimit()
 	}
@@ -191,7 +191,7 @@ func Prlimit64(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.Sys
 	ot := t
 	if tid > 0 {
 		if ot = t.PIDNamespace().TaskWithID(tid); ot == nil {
-			return 0, nil, syserror.ESRCH
+			return 0, nil, linuxerr.ESRCH
 		}
 	}
 
@@ -219,7 +219,7 @@ func Prlimit64(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.Sys
 
 	if oldRlimAddr != 0 {
 		if err := makeRlimit64(oldLim).copyOut(t, oldRlimAddr); err != nil {
-			return 0, nil, syserror.EFAULT
+			return 0, nil, linuxerr.EFAULT
 		}
 	}
 
diff --git a/pkg/sentry/syscalls/linux/sys_sched.go b/pkg/sentry/syscalls/linux/sys_sched.go
index b5e7b70b5..59c7a4b22 100644
--- a/pkg/sentry/syscalls/linux/sys_sched.go
+++ b/pkg/sentry/syscalls/linux/sys_sched.go
@@ -19,7 +19,6 @@ import (
 	"gvisor.dev/gvisor/pkg/errors/linuxerr"
 	"gvisor.dev/gvisor/pkg/sentry/arch"
 	"gvisor.dev/gvisor/pkg/sentry/kernel"
-	"gvisor.dev/gvisor/pkg/syserror"
 )
 
 const (
@@ -45,7 +44,7 @@ func SchedGetparam(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel
 		return 0, nil, linuxerr.EINVAL
 	}
 	if pid != 0 && t.PIDNamespace().TaskWithID(kernel.ThreadID(pid)) == nil {
-		return 0, nil, syserror.ESRCH
+		return 0, nil, linuxerr.ESRCH
 	}
 	r := SchedParam{schedPriority: onlyPriority}
 	if _, err := r.CopyOut(t, param); err != nil {
@@ -62,7 +61,7 @@ func SchedGetscheduler(t *kernel.Task, args arch.SyscallArguments) (uintptr, *ke
 		return 0, nil, linuxerr.EINVAL
 	}
 	if pid != 0 && t.PIDNamespace().TaskWithID(kernel.ThreadID(pid)) == nil {
-		return 0, nil, syserror.ESRCH
+		return 0, nil, linuxerr.ESRCH
 	}
 	return onlyScheduler, nil, nil
 }
@@ -79,7 +78,7 @@ func SchedSetscheduler(t *kernel.Task, args arch.SyscallArguments) (uintptr, *ke
 		return 0, nil, linuxerr.EINVAL
 	}
 	if pid != 0 && t.PIDNamespace().TaskWithID(kernel.ThreadID(pid)) == nil {
-		return 0, nil, syserror.ESRCH
+		return 0, nil, linuxerr.ESRCH
 	}
 	var r SchedParam
 	if _, err := r.CopyIn(t, param); err != nil {
diff --git a/pkg/sentry/syscalls/linux/sys_sem.go b/pkg/sentry/syscalls/linux/sys_sem.go
index ef61b8f33..30919eb2f 100644
--- a/pkg/sentry/syscalls/linux/sys_sem.go
+++ b/pkg/sentry/syscalls/linux/sys_sem.go
@@ -26,7 +26,6 @@ import (
 	"gvisor.dev/gvisor/pkg/sentry/fs"
 	"gvisor.dev/gvisor/pkg/sentry/kernel"
 	"gvisor.dev/gvisor/pkg/sentry/kernel/auth"
-	"gvisor.dev/gvisor/pkg/syserror"
 )
 
 const opsMax = 500 // SEMOPM
@@ -140,7 +139,7 @@ func Semctl(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.Syscal
 	case linux.SETVAL:
 		val := args[3].Int()
 		if val > math.MaxInt16 {
-			return 0, nil, syserror.ERANGE
+			return 0, nil, linuxerr.ERANGE
 		}
 		return 0, nil, setVal(t, id, num, int16(val))
 
diff --git a/pkg/sentry/syscalls/linux/sys_signal.go b/pkg/sentry/syscalls/linux/sys_signal.go
index f98d5c903..45608f3fa 100644
--- a/pkg/sentry/syscalls/linux/sys_signal.go
+++ b/pkg/sentry/syscalls/linux/sys_signal.go
@@ -80,7 +80,7 @@ func Kill(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.SyscallC
 		for {
 			target := t.PIDNamespace().TaskWithID(pid)
 			if target == nil {
-				return 0, nil, syserror.ESRCH
+				return 0, nil, linuxerr.ESRCH
 			}
 			if !mayKill(t, target, sig) {
 				return 0, nil, linuxerr.EPERM
@@ -146,7 +146,7 @@ func Kill(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.SyscallC
 		if delivered > 0 {
 			return 0, nil, lastErr
 		}
-		return 0, nil, syserror.ESRCH
+		return 0, nil, linuxerr.ESRCH
 	default:
 		// "If pid equals 0, then sig is sent to every process in the process
 		// group of the calling process."
@@ -160,7 +160,7 @@ func Kill(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.SyscallC
 
 		// If pid != -1 (i.e. signalling a process group), the returned error
 		// is the last error from any call to group_send_sig_info.
-		lastErr := syserror.ESRCH
+		lastErr := error(linuxerr.ESRCH)
 		for _, tg := range t.PIDNamespace().ThreadGroups() {
 			if t.PIDNamespace().IDOfProcessGroup(tg.ProcessGroup()) == pgid {
 				if !mayKill(t, tg.Leader(), sig) {
@@ -208,7 +208,7 @@ func Tkill(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.Syscall
 
 	target := t.PIDNamespace().TaskWithID(tid)
 	if target == nil {
-		return 0, nil, syserror.ESRCH
+		return 0, nil, linuxerr.ESRCH
 	}
 
 	if !mayKill(t, target, sig) {
@@ -232,7 +232,7 @@ func Tgkill(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.Syscal
 	targetTG := t.PIDNamespace().ThreadGroupWithID(tgid)
 	target := t.PIDNamespace().TaskWithID(tid)
 	if targetTG == nil || target == nil || target.ThreadGroup() != targetTG {
-		return 0, nil, syserror.ESRCH
+		return 0, nil, linuxerr.ESRCH
 	}
 
 	if !mayKill(t, target, sig) {
@@ -421,7 +421,7 @@ func RtSigqueueinfo(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kerne
 		// Deliver to the given task's thread group.
 		target := t.PIDNamespace().TaskWithID(pid)
 		if target == nil {
-			return 0, nil, syserror.ESRCH
+			return 0, nil, linuxerr.ESRCH
 		}
 
 		// If the sender is not the receiver, it can't use si_codes used by the
@@ -464,7 +464,7 @@ func RtTgsigqueueinfo(t *kernel.Task, args arch.SyscallArguments) (uintptr, *ker
 	targetTG := t.PIDNamespace().ThreadGroupWithID(tgid)
 	target := t.PIDNamespace().TaskWithID(tid)
 	if targetTG == nil || target == nil || target.ThreadGroup() != targetTG {
-		return 0, nil, syserror.ESRCH
+		return 0, nil, linuxerr.ESRCH
 	}
 
 	// If the sender is not the receiver, it can't use si_codes used by the
diff --git a/pkg/sentry/syscalls/linux/sys_socket.go b/pkg/sentry/syscalls/linux/sys_socket.go
index ea2b2f40c..06eb8f319 100644
--- a/pkg/sentry/syscalls/linux/sys_socket.go
+++ b/pkg/sentry/syscalls/linux/sys_socket.go
@@ -718,7 +718,7 @@ func RecvMMsg(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.Sysc
 	for i := uint64(0); i < uint64(vlen); i++ {
 		mp, ok := msgPtr.AddLength(i * multipleMessageHeader64Len)
 		if !ok {
-			return 0, nil, syserror.EFAULT
+			return 0, nil, linuxerr.EFAULT
 		}
 		var n uintptr
 		if n, err = recvSingleMsg(t, s, mp, flags, haveDeadline, deadline); err != nil {
@@ -728,7 +728,7 @@ func RecvMMsg(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.Sysc
 		// Copy the received length to the caller.
 		lp, ok := mp.AddLength(messageHeader64Len)
 		if !ok {
-			return 0, nil, syserror.EFAULT
+			return 0, nil, linuxerr.EFAULT
 		}
 		if _, err = primitive.CopyUint32Out(t, lp, uint32(n)); err != nil {
 			break
@@ -980,7 +980,7 @@ func SendMMsg(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.Sysc
 	for i := uint64(0); i < uint64(vlen); i++ {
 		mp, ok := msgPtr.AddLength(i * multipleMessageHeader64Len)
 		if !ok {
-			return 0, nil, syserror.EFAULT
+			return 0, nil, linuxerr.EFAULT
 		}
 		var n uintptr
 		if n, err = sendSingleMsg(t, s, file, mp, flags); err != nil {
@@ -990,7 +990,7 @@ func SendMMsg(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.Sysc
 		// Copy the received length to the caller.
 		lp, ok := mp.AddLength(messageHeader64Len)
 		if !ok {
-			return 0, nil, syserror.EFAULT
+			return 0, nil, linuxerr.EFAULT
 		}
 		if _, err = primitive.CopyUint32Out(t, lp, uint32(n)); err != nil {
 			break
diff --git a/pkg/sentry/syscalls/linux/sys_stat.go b/pkg/sentry/syscalls/linux/sys_stat.go
index 46a69a78a..3da385c66 100644
--- a/pkg/sentry/syscalls/linux/sys_stat.go
+++ b/pkg/sentry/syscalls/linux/sys_stat.go
@@ -21,7 +21,6 @@ import (
 	"gvisor.dev/gvisor/pkg/sentry/arch"
 	"gvisor.dev/gvisor/pkg/sentry/fs"
 	"gvisor.dev/gvisor/pkg/sentry/kernel"
-	"gvisor.dev/gvisor/pkg/syserror"
 )
 
 // LINT.IfChange
@@ -109,7 +108,7 @@ func Fstat(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.Syscall
 // stat implements stat from the given *fs.Dirent.
 func stat(t *kernel.Task, d *fs.Dirent, dirPath bool, statAddr hostarch.Addr) error {
 	if dirPath && !fs.IsDir(d.Inode.StableAttr) {
-		return syserror.ENOTDIR
+		return linuxerr.ENOTDIR
 	}
 	uattr, err := d.Inode.UnstableAttr(t)
 	if err != nil {
@@ -171,7 +170,7 @@ func Statx(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.Syscall
 
 	return 0, nil, fileOpOn(t, fd, path, resolve, func(root *fs.Dirent, d *fs.Dirent, _ uint) error {
 		if dirPath && !fs.IsDir(d.Inode.StableAttr) {
-			return syserror.ENOTDIR
+			return linuxerr.ENOTDIR
 		}
 		uattr, err := d.Inode.UnstableAttr(t)
 		if err != nil {
diff --git a/pkg/sentry/syscalls/linux/sys_thread.go b/pkg/sentry/syscalls/linux/sys_thread.go
index 9f0eeb466..46145955e 100644
--- a/pkg/sentry/syscalls/linux/sys_thread.go
+++ b/pkg/sentry/syscalls/linux/sys_thread.go
@@ -154,7 +154,7 @@ func execveat(t *kernel.Task, dirFD int32, pathnameAddr, argvAddr, envvAddr host
 			wd = f.Dirent
 			wd.IncRef()
 			if !fs.IsDir(wd.Inode.StableAttr) {
-				return 0, nil, syserror.ENOTDIR
+				return 0, nil, linuxerr.ENOTDIR
 			}
 		}
 	}
@@ -503,7 +503,7 @@ func SchedSetaffinity(t *kernel.Task, args arch.SyscallArguments) (uintptr, *ker
 	} else {
 		task = t.PIDNamespace().TaskWithID(kernel.ThreadID(tid))
 		if task == nil {
-			return 0, nil, syserror.ESRCH
+			return 0, nil, linuxerr.ESRCH
 		}
 	}
 
@@ -536,7 +536,7 @@ func SchedGetaffinity(t *kernel.Task, args arch.SyscallArguments) (uintptr, *ker
 	} else {
 		task = t.PIDNamespace().TaskWithID(kernel.ThreadID(tid))
 		if task == nil {
-			return 0, nil, syserror.ESRCH
+			return 0, nil, linuxerr.ESRCH
 		}
 	}
 
@@ -589,7 +589,7 @@ func Setpgid(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.Sysca
 	if pid != 0 {
 		ot := t.PIDNamespace().TaskWithID(pid)
 		if ot == nil {
-			return 0, nil, syserror.ESRCH
+			return 0, nil, linuxerr.ESRCH
 		}
 		tg = ot.ThreadGroup()
 		if tg.Leader() != ot {
@@ -598,7 +598,7 @@ func Setpgid(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.Sysca
 
 		// Setpgid only operates on child threadgroups.
 		if tg != t.ThreadGroup() && (tg.Leader().Parent() == nil || tg.Leader().Parent().ThreadGroup() != t.ThreadGroup()) {
-			return 0, nil, syserror.ESRCH
+			return 0, nil, linuxerr.ESRCH
 		}
 	}
 
@@ -653,7 +653,7 @@ func Getpgid(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.Sysca
 
 	target := t.PIDNamespace().TaskWithID(tid)
 	if target == nil {
-		return 0, nil, syserror.ESRCH
+		return 0, nil, linuxerr.ESRCH
 	}
 
 	return uintptr(t.PIDNamespace().IDOfProcessGroup(target.ThreadGroup().ProcessGroup())), nil, nil
@@ -673,7 +673,7 @@ func Getsid(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.Syscal
 
 	target := t.PIDNamespace().TaskWithID(tid)
 	if target == nil {
-		return 0, nil, syserror.ESRCH
+		return 0, nil, linuxerr.ESRCH
 	}
 
 	return uintptr(t.PIDNamespace().IDOfSession(target.ThreadGroup().Session())), nil, nil
@@ -697,7 +697,7 @@ func Getpriority(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.S
 		}
 
 		if task == nil {
-			return 0, nil, syserror.ESRCH
+			return 0, nil, linuxerr.ESRCH
 		}
 
 		// From kernel/sys.c:getpriority:
@@ -743,7 +743,7 @@ func Setpriority(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.S
 		}
 
 		if task == nil {
-			return 0, nil, syserror.ESRCH
+			return 0, nil, linuxerr.ESRCH
 		}
 
 		task.SetNiceness(niceval)
diff --git a/pkg/sentry/syscalls/linux/sys_write.go b/pkg/sentry/syscalls/linux/sys_write.go
index eec917d95..872168606 100644
--- a/pkg/sentry/syscalls/linux/sys_write.go
+++ b/pkg/sentry/syscalls/linux/sys_write.go
@@ -231,7 +231,7 @@ func Pwritev2(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.Sysc
 	// Note: gVisor does not implement the RWF_HIPRI feature, but the flag is
 	// accepted as a valid flag argument for pwritev2.
 	if flags&^linux.RWF_VALID != 0 {
-		return uintptr(flags), nil, syserror.EOPNOTSUPP
+		return uintptr(flags), nil, linuxerr.EOPNOTSUPP
 	}
 
 	// Check that the file is writeable.
diff --git a/pkg/sentry/syscalls/linux/sys_xattr.go b/pkg/sentry/syscalls/linux/sys_xattr.go
index 717e23c59..baaf31191 100644
--- a/pkg/sentry/syscalls/linux/sys_xattr.go
+++ b/pkg/sentry/syscalls/linux/sys_xattr.go
@@ -23,7 +23,6 @@ import (
 	"gvisor.dev/gvisor/pkg/sentry/arch"
 	"gvisor.dev/gvisor/pkg/sentry/fs"
 	"gvisor.dev/gvisor/pkg/sentry/kernel"
-	"gvisor.dev/gvisor/pkg/syserror"
 )
 
 // LINT.IfChange
@@ -74,7 +73,7 @@ func getXattrFromPath(t *kernel.Task, args arch.SyscallArguments, resolveSymlink
 	n := 0
 	err = fileOpOn(t, linux.AT_FDCWD, path, resolveSymlink, func(_ *fs.Dirent, d *fs.Dirent, _ uint) error {
 		if dirPath && !fs.IsDir(d.Inode.StableAttr) {
-			return syserror.ENOTDIR
+			return linuxerr.ENOTDIR
 		}
 
 		n, err = getXattr(t, d, nameAddr, valueAddr, size)
@@ -100,7 +99,7 @@ func getXattr(t *kernel.Task, d *fs.Dirent, nameAddr, valueAddr hostarch.Addr, s
 
 	// TODO(b/148380782): Support xattrs in namespaces other than "user".
 	if !strings.HasPrefix(name, linux.XATTR_USER_PREFIX) {
-		return 0, syserror.EOPNOTSUPP
+		return 0, linuxerr.EOPNOTSUPP
 	}
 
 	// If getxattr(2) is called with size 0, the size of the value will be
@@ -117,7 +116,7 @@ func getXattr(t *kernel.Task, d *fs.Dirent, nameAddr, valueAddr hostarch.Addr, s
 	}
 	n := len(value)
 	if uint64(n) > requestedSize {
-		return 0, syserror.ERANGE
+		return 0, linuxerr.ERANGE
 	}
 
 	// Don't copy out the attribute value if size is 0.
@@ -173,7 +172,7 @@ func setXattrFromPath(t *kernel.Task, args arch.SyscallArguments, resolveSymlink
 
 	return 0, nil, fileOpOn(t, linux.AT_FDCWD, path, resolveSymlink, func(_ *fs.Dirent, d *fs.Dirent, _ uint) error {
 		if dirPath && !fs.IsDir(d.Inode.StableAttr) {
-			return syserror.ENOTDIR
+			return linuxerr.ENOTDIR
 		}
 
 		return setXattr(t, d, nameAddr, valueAddr, uint64(size), flags)
@@ -205,7 +204,7 @@ func setXattr(t *kernel.Task, d *fs.Dirent, nameAddr, valueAddr hostarch.Addr, s
 	value := string(buf)
 
 	if !strings.HasPrefix(name, linux.XATTR_USER_PREFIX) {
-		return syserror.EOPNOTSUPP
+		return linuxerr.EOPNOTSUPP
 	}
 
 	if err := d.Inode.SetXattr(t, d, name, value, flags); err != nil {
@@ -219,12 +218,12 @@ func copyInXattrName(t *kernel.Task, nameAddr hostarch.Addr) (string, error) {
 	name, err := t.CopyInString(nameAddr, linux.XATTR_NAME_MAX+1)
 	if err != nil {
 		if linuxerr.Equals(linuxerr.ENAMETOOLONG, err) {
-			return "", syserror.ERANGE
+			return "", linuxerr.ERANGE
 		}
 		return "", err
 	}
 	if len(name) == 0 {
-		return "", syserror.ERANGE
+		return "", linuxerr.ERANGE
 	}
 	return name, nil
 }
@@ -294,7 +293,7 @@ func listXattrFromPath(t *kernel.Task, args arch.SyscallArguments, resolveSymlin
 	n := 0
 	err = fileOpOn(t, linux.AT_FDCWD, path, resolveSymlink, func(_ *fs.Dirent, d *fs.Dirent, _ uint) error {
 		if dirPath && !fs.IsDir(d.Inode.StableAttr) {
-			return syserror.ENOTDIR
+			return linuxerr.ENOTDIR
 		}
 
 		n, err = listXattr(t, d, listAddr, size)
@@ -337,7 +336,7 @@ func listXattr(t *kernel.Task, d *fs.Dirent, addr hostarch.Addr, size uint64) (i
 		return 0, linuxerr.E2BIG
 	}
 	if uint64(listSize) > requestedSize {
-		return 0, syserror.ERANGE
+		return 0, linuxerr.ERANGE
 	}
 
 	// Don't copy out the attributes if size is 0.
@@ -401,7 +400,7 @@ func removeXattrFromPath(t *kernel.Task, args arch.SyscallArguments, resolveSyml
 
 	return 0, nil, fileOpOn(t, linux.AT_FDCWD, path, resolveSymlink, func(_ *fs.Dirent, d *fs.Dirent, _ uint) error {
 		if dirPath && !fs.IsDir(d.Inode.StableAttr) {
-			return syserror.ENOTDIR
+			return linuxerr.ENOTDIR
 		}
 
 		return removeXattr(t, d, nameAddr)
@@ -420,7 +419,7 @@ func removeXattr(t *kernel.Task, d *fs.Dirent, nameAddr hostarch.Addr) error {
 	}
 
 	if !strings.HasPrefix(name, linux.XATTR_USER_PREFIX) {
-		return syserror.EOPNOTSUPP
+		return linuxerr.EOPNOTSUPP
 	}
 
 	if err := d.Inode.RemoveXattr(t, d, name); err != nil {
diff --git a/pkg/sentry/syscalls/linux/vfs2/fd.go b/pkg/sentry/syscalls/linux/vfs2/fd.go
index 55272fbbc..2cfb12cad 100644
--- a/pkg/sentry/syscalls/linux/vfs2/fd.go
+++ b/pkg/sentry/syscalls/linux/vfs2/fd.go
@@ -283,21 +283,21 @@ func setAsyncOwner(t *kernel.Task, fd int, file *vfs.FileDescription, ownerType,
 	case linux.F_OWNER_TID:
 		task := t.PIDNamespace().TaskWithID(kernel.ThreadID(pid))
 		if task == nil {
-			return syserror.ESRCH
+			return linuxerr.ESRCH
 		}
 		a.SetOwnerTask(t, task)
 		return nil
 	case linux.F_OWNER_PID:
 		tg := t.PIDNamespace().ThreadGroupWithID(kernel.ThreadID(pid))
 		if tg == nil {
-			return syserror.ESRCH
+			return linuxerr.ESRCH
 		}
 		a.SetOwnerThreadGroup(t, tg)
 		return nil
 	case linux.F_OWNER_PGRP:
 		pg := t.PIDNamespace().ProcessGroupWithID(kernel.ProcessGroupID(pid))
 		if pg == nil {
-			return syserror.ESRCH
+			return linuxerr.ESRCH
 		}
 		a.SetOwnerProcessGroup(t, pg)
 		return nil
diff --git a/pkg/sentry/syscalls/linux/vfs2/fscontext.go b/pkg/sentry/syscalls/linux/vfs2/fscontext.go
index f74d6197c..1e36d9c76 100644
--- a/pkg/sentry/syscalls/linux/vfs2/fscontext.go
+++ b/pkg/sentry/syscalls/linux/vfs2/fscontext.go
@@ -21,7 +21,6 @@ import (
 	"gvisor.dev/gvisor/pkg/sentry/arch"
 	"gvisor.dev/gvisor/pkg/sentry/kernel"
 	"gvisor.dev/gvisor/pkg/sentry/vfs"
-	"gvisor.dev/gvisor/pkg/syserror"
 )
 
 // Getcwd implements Linux syscall getcwd(2).
@@ -40,7 +39,7 @@ func Getcwd(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.Syscal
 
 	// Note this is >= because we need a terminator.
 	if uint(len(s)) >= size {
-		return 0, nil, syserror.ERANGE
+		return 0, nil, linuxerr.ERANGE
 	}
 
 	// Construct a byte slice containing a NUL terminator.
diff --git a/pkg/sentry/syscalls/linux/vfs2/socket.go b/pkg/sentry/syscalls/linux/vfs2/socket.go
index b39292f50..0c2e0720b 100644
--- a/pkg/sentry/syscalls/linux/vfs2/socket.go
+++ b/pkg/sentry/syscalls/linux/vfs2/socket.go
@@ -722,7 +722,7 @@ func RecvMMsg(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.Sysc
 	for i := uint64(0); i < uint64(vlen); i++ {
 		mp, ok := msgPtr.AddLength(i * multipleMessageHeader64Len)
 		if !ok {
-			return 0, nil, syserror.EFAULT
+			return 0, nil, linuxerr.EFAULT
 		}
 		var n uintptr
 		if n, err = recvSingleMsg(t, s, mp, flags, haveDeadline, deadline); err != nil {
@@ -732,7 +732,7 @@ func RecvMMsg(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.Sysc
 		// Copy the received length to the caller.
 		lp, ok := mp.AddLength(messageHeader64Len)
 		if !ok {
-			return 0, nil, syserror.EFAULT
+			return 0, nil, linuxerr.EFAULT
 		}
 		if _, err = primitive.CopyUint32Out(t, lp, uint32(n)); err != nil {
 			break
@@ -984,7 +984,7 @@ func SendMMsg(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.Sysc
 	for i := uint64(0); i < uint64(vlen); i++ {
 		mp, ok := msgPtr.AddLength(i * multipleMessageHeader64Len)
 		if !ok {
-			return 0, nil, syserror.EFAULT
+			return 0, nil, linuxerr.EFAULT
 		}
 		var n uintptr
 		if n, err = sendSingleMsg(t, s, file, mp, flags); err != nil {
@@ -994,7 +994,7 @@ func SendMMsg(t *kernel.Task, args arch.SyscallArguments) (uintptr, *kernel.Sysc
 		// Copy the received length to the caller.
 		lp, ok := mp.AddLength(messageHeader64Len)
 		if !ok {
-			return 0, nil, syserror.EFAULT
+			return 0, nil, linuxerr.EFAULT
 		}
 		if _, err = primitive.CopyUint32Out(t, lp, uint32(n)); err != nil {
 			break
diff --git a/pkg/sentry/syscalls/linux/vfs2/xattr.go b/pkg/sentry/syscalls/linux/vfs2/xattr.go
index 0f54529bc..7b2f69c45 100644
--- a/pkg/sentry/syscalls/linux/vfs2/xattr.go
+++ b/pkg/sentry/syscalls/linux/vfs2/xattr.go
@@ -20,12 +20,10 @@ import (
 	"gvisor.dev/gvisor/pkg/abi/linux"
 	"gvisor.dev/gvisor/pkg/errors/linuxerr"
 	"gvisor.dev/gvisor/pkg/gohacks"
+	"gvisor.dev/gvisor/pkg/hostarch"
 	"gvisor.dev/gvisor/pkg/sentry/arch"
 	"gvisor.dev/gvisor/pkg/sentry/kernel"
 	"gvisor.dev/gvisor/pkg/sentry/vfs"
-	"gvisor.dev/gvisor/pkg/syserror"
-
-	"gvisor.dev/gvisor/pkg/hostarch"
 )
 
 // ListXattr implements Linux syscall listxattr(2).
@@ -297,12 +295,12 @@ func copyInXattrName(t *kernel.Task, nameAddr hostarch.Addr) (string, error) {
 	name, err := t.CopyInString(nameAddr, linux.XATTR_NAME_MAX+1)
 	if err != nil {
 		if linuxerr.Equals(linuxerr.ENAMETOOLONG, err) {
-			return "", syserror.ERANGE
+			return "", linuxerr.ERANGE
 		}
 		return "", err
 	}
 	if len(name) == 0 {
-		return "", syserror.ERANGE
+		return "", linuxerr.ERANGE
 	}
 	return name, nil
 }
@@ -324,7 +322,7 @@ func copyOutXattrNameList(t *kernel.Task, listAddr hostarch.Addr, size uint, nam
 		if size >= linux.XATTR_LIST_MAX {
 			return 0, linuxerr.E2BIG
 		}
-		return 0, syserror.ERANGE
+		return 0, linuxerr.ERANGE
 	}
 	return t.CopyOutBytes(listAddr, buf.Bytes())
 }
@@ -352,7 +350,7 @@ func copyOutXattrValue(t *kernel.Task, valueAddr hostarch.Addr, size uint, value
 		if size >= linux.XATTR_SIZE_MAX {
 			return 0, linuxerr.E2BIG
 		}
-		return 0, syserror.ERANGE
+		return 0, linuxerr.ERANGE
 	}
 	return t.CopyOutBytes(valueAddr, gohacks.ImmutableBytesFromString(value))
 }
diff --git a/pkg/sentry/vfs/anonfs.go b/pkg/sentry/vfs/anonfs.go
index 40863cbb6..255d3992e 100644
--- a/pkg/sentry/vfs/anonfs.go
+++ b/pkg/sentry/vfs/anonfs.go
@@ -24,7 +24,6 @@ import (
 	"gvisor.dev/gvisor/pkg/hostarch"
 	"gvisor.dev/gvisor/pkg/sentry/kernel/auth"
 	"gvisor.dev/gvisor/pkg/sentry/socket/unix/transport"
-	"gvisor.dev/gvisor/pkg/syserror"
 )
 
 // NewAnonVirtualDentry returns a VirtualDentry with the given synthetic name,
@@ -102,7 +101,7 @@ func (fs *anonFilesystem) Sync(ctx context.Context) error {
 // AccessAt implements vfs.Filesystem.Impl.AccessAt.
 func (fs *anonFilesystem) AccessAt(ctx context.Context, rp *ResolvingPath, creds *auth.Credentials, ats AccessTypes) error {
 	if !rp.Done() {
-		return syserror.ENOTDIR
+		return linuxerr.ENOTDIR
 	}
 	return GenericCheckPermissions(creds, ats, anonFileMode, anonFileUID, anonFileGID)
 }
@@ -110,10 +109,10 @@ func (fs *anonFilesystem) AccessAt(ctx context.Context, rp *ResolvingPath, creds
 // GetDentryAt implements FilesystemImpl.GetDentryAt.
 func (fs *anonFilesystem) GetDentryAt(ctx context.Context, rp *ResolvingPath, opts GetDentryOptions) (*Dentry, error) {
 	if !rp.Done() {
-		return nil, syserror.ENOTDIR
+		return nil, linuxerr.ENOTDIR
 	}
 	if opts.CheckSearchable {
-		return nil, syserror.ENOTDIR
+		return nil, linuxerr.ENOTDIR
 	}
 	// anonDentry no-ops refcounting.
 	return rp.Start(), nil
@@ -122,7 +121,7 @@ func (fs *anonFilesystem) GetDentryAt(ctx context.Context, rp *ResolvingPath, op
 // GetParentDentryAt implements FilesystemImpl.GetParentDentryAt.
 func (fs *anonFilesystem) GetParentDentryAt(ctx context.Context, rp *ResolvingPath) (*Dentry, error) {
 	if !rp.Final() {
-		return nil, syserror.ENOTDIR
+		return nil, linuxerr.ENOTDIR
 	}
 	// anonDentry no-ops refcounting.
 	return rp.Start(), nil
@@ -131,7 +130,7 @@ func (fs *anonFilesystem) GetParentDentryAt(ctx context.Context, rp *ResolvingPa
 // LinkAt implements FilesystemImpl.LinkAt.
 func (fs *anonFilesystem) LinkAt(ctx context.Context, rp *ResolvingPath, vd VirtualDentry) error {
 	if !rp.Final() {
-		return syserror.ENOTDIR
+		return linuxerr.ENOTDIR
 	}
 	return linuxerr.EPERM
 }
@@ -139,7 +138,7 @@ func (fs *anonFilesystem) LinkAt(ctx context.Context, rp *ResolvingPath, vd Virt
 // MkdirAt implements FilesystemImpl.MkdirAt.
 func (fs *anonFilesystem) MkdirAt(ctx context.Context, rp *ResolvingPath, opts MkdirOptions) error {
 	if !rp.Final() {
-		return syserror.ENOTDIR
+		return linuxerr.ENOTDIR
 	}
 	return linuxerr.EPERM
 }
@@ -147,7 +146,7 @@ func (fs *anonFilesystem) MkdirAt(ctx context.Context, rp *ResolvingPath, opts M
 // MknodAt implements FilesystemImpl.MknodAt.
 func (fs *anonFilesystem) MknodAt(ctx context.Context, rp *ResolvingPath, opts MknodOptions) error {
 	if !rp.Final() {
-		return syserror.ENOTDIR
+		return linuxerr.ENOTDIR
 	}
 	return linuxerr.EPERM
 }
@@ -155,7 +154,7 @@ func (fs *anonFilesystem) MknodAt(ctx context.Context, rp *ResolvingPath, opts M
 // OpenAt implements FilesystemImpl.OpenAt.
 func (fs *anonFilesystem) OpenAt(ctx context.Context, rp *ResolvingPath, opts OpenOptions) (*FileDescription, error) {
 	if !rp.Done() {
-		return nil, syserror.ENOTDIR
+		return nil, linuxerr.ENOTDIR
 	}
 	return nil, linuxerr.ENODEV
 }
@@ -163,7 +162,7 @@ func (fs *anonFilesystem) OpenAt(ctx context.Context, rp *ResolvingPath, opts Op
 // ReadlinkAt implements FilesystemImpl.ReadlinkAt.
 func (fs *anonFilesystem) ReadlinkAt(ctx context.Context, rp *ResolvingPath) (string, error) {
 	if !rp.Done() {
-		return "", syserror.ENOTDIR
+		return "", linuxerr.ENOTDIR
 	}
 	return "", linuxerr.EINVAL
 }
@@ -171,7 +170,7 @@ func (fs *anonFilesystem) ReadlinkAt(ctx context.Context, rp *ResolvingPath) (st
 // RenameAt implements FilesystemImpl.RenameAt.
 func (fs *anonFilesystem) RenameAt(ctx context.Context, rp *ResolvingPath, oldParentVD VirtualDentry, oldName string, opts RenameOptions) error {
 	if !rp.Final() {
-		return syserror.ENOTDIR
+		return linuxerr.ENOTDIR
 	}
 	return linuxerr.EPERM
 }
@@ -179,7 +178,7 @@ func (fs *anonFilesystem) RenameAt(ctx context.Context, rp *ResolvingPath, oldPa
 // RmdirAt implements FilesystemImpl.RmdirAt.
 func (fs *anonFilesystem) RmdirAt(ctx context.Context, rp *ResolvingPath) error {
 	if !rp.Final() {
-		return syserror.ENOTDIR
+		return linuxerr.ENOTDIR
 	}
 	return linuxerr.EPERM
 }
@@ -187,7 +186,7 @@ func (fs *anonFilesystem) RmdirAt(ctx context.Context, rp *ResolvingPath) error
 // SetStatAt implements FilesystemImpl.SetStatAt.
 func (fs *anonFilesystem) SetStatAt(ctx context.Context, rp *ResolvingPath, opts SetStatOptions) error {
 	if !rp.Done() {
-		return syserror.ENOTDIR
+		return linuxerr.ENOTDIR
 	}
 	// Linux actually permits anon_inode_inode's metadata to be set, which is
 	// visible to all users of anon_inode_inode. We just silently ignore
@@ -198,7 +197,7 @@ func (fs *anonFilesystem) SetStatAt(ctx context.Context, rp *ResolvingPath, opts
 // StatAt implements FilesystemImpl.StatAt.
 func (fs *anonFilesystem) StatAt(ctx context.Context, rp *ResolvingPath, opts StatOptions) (linux.Statx, error) {
 	if !rp.Done() {
-		return linux.Statx{}, syserror.ENOTDIR
+		return linux.Statx{}, linuxerr.ENOTDIR
 	}
 	// See fs/anon_inodes.c:anon_inode_init() => fs/libfs.c:alloc_anon_inode().
 	return linux.Statx{
@@ -219,7 +218,7 @@ func (fs *anonFilesystem) StatAt(ctx context.Context, rp *ResolvingPath, opts St
 // StatFSAt implements FilesystemImpl.StatFSAt.
 func (fs *anonFilesystem) StatFSAt(ctx context.Context, rp *ResolvingPath) (linux.Statfs, error) {
 	if !rp.Done() {
-		return linux.Statfs{}, syserror.ENOTDIR
+		return linux.Statfs{}, linuxerr.ENOTDIR
 	}
 	return linux.Statfs{
 		Type:      linux.ANON_INODE_FS_MAGIC,
@@ -230,7 +229,7 @@ func (fs *anonFilesystem) StatFSAt(ctx context.Context, rp *ResolvingPath) (linu
 // SymlinkAt implements FilesystemImpl.SymlinkAt.
 func (fs *anonFilesystem) SymlinkAt(ctx context.Context, rp *ResolvingPath, target string) error {
 	if !rp.Final() {
-		return syserror.ENOTDIR
+		return linuxerr.ENOTDIR
 	}
 	return linuxerr.EPERM
 }
@@ -238,7 +237,7 @@ func (fs *anonFilesystem) SymlinkAt(ctx context.Context, rp *ResolvingPath, targ
 // UnlinkAt implements FilesystemImpl.UnlinkAt.
 func (fs *anonFilesystem) UnlinkAt(ctx context.Context, rp *ResolvingPath) error {
 	if !rp.Final() {
-		return syserror.ENOTDIR
+		return linuxerr.ENOTDIR
 	}
 	return linuxerr.EPERM
 }
@@ -246,7 +245,7 @@ func (fs *anonFilesystem) UnlinkAt(ctx context.Context, rp *ResolvingPath) error
 // BoundEndpointAt implements FilesystemImpl.BoundEndpointAt.
 func (fs *anonFilesystem) BoundEndpointAt(ctx context.Context, rp *ResolvingPath, opts BoundEndpointOptions) (transport.BoundEndpoint, error) {
 	if !rp.Final() {
-		return nil, syserror.ENOTDIR
+		return nil, linuxerr.ENOTDIR
 	}
 	if err := GenericCheckPermissions(rp.Credentials(), MayWrite, anonFileMode, anonFileUID, anonFileGID); err != nil {
 		return nil, err
@@ -257,7 +256,7 @@ func (fs *anonFilesystem) BoundEndpointAt(ctx context.Context, rp *ResolvingPath
 // ListXattrAt implements FilesystemImpl.ListXattrAt.
 func (fs *anonFilesystem) ListXattrAt(ctx context.Context, rp *ResolvingPath, size uint64) ([]string, error) {
 	if !rp.Done() {
-		return nil, syserror.ENOTDIR
+		return nil, linuxerr.ENOTDIR
 	}
 	return nil, nil
 }
@@ -265,7 +264,7 @@ func (fs *anonFilesystem) ListXattrAt(ctx context.Context, rp *ResolvingPath, si
 // GetXattrAt implements FilesystemImpl.GetXattrAt.
 func (fs *anonFilesystem) GetXattrAt(ctx context.Context, rp *ResolvingPath, opts GetXattrOptions) (string, error) {
 	if !rp.Done() {
-		return "", syserror.ENOTDIR
+		return "", linuxerr.ENOTDIR
 	}
 	return "", linuxerr.ENOTSUP
 }
@@ -273,7 +272,7 @@ func (fs *anonFilesystem) GetXattrAt(ctx context.Context, rp *ResolvingPath, opt
 // SetXattrAt implements FilesystemImpl.SetXattrAt.
 func (fs *anonFilesystem) SetXattrAt(ctx context.Context, rp *ResolvingPath, opts SetXattrOptions) error {
 	if !rp.Done() {
-		return syserror.ENOTDIR
+		return linuxerr.ENOTDIR
 	}
 	return linuxerr.EPERM
 }
@@ -281,7 +280,7 @@ func (fs *anonFilesystem) SetXattrAt(ctx context.Context, rp *ResolvingPath, opt
 // RemoveXattrAt implements FilesystemImpl.RemoveXattrAt.
 func (fs *anonFilesystem) RemoveXattrAt(ctx context.Context, rp *ResolvingPath, name string) error {
 	if !rp.Done() {
-		return syserror.ENOTDIR
+		return linuxerr.ENOTDIR
 	}
 	return linuxerr.EPERM
 }
diff --git a/pkg/sentry/vfs/epoll.go b/pkg/sentry/vfs/epoll.go
index a93e757f8..befe3ca25 100644
--- a/pkg/sentry/vfs/epoll.go
+++ b/pkg/sentry/vfs/epoll.go
@@ -188,7 +188,7 @@ func (ep *EpollInstance) AddInterest(file *FileDescription, num int32, event lin
 		num:  num,
 	}
 	if _, ok := ep.interest[key]; ok {
-		return syserror.EEXIST
+		return linuxerr.EEXIST
 	}
 
 	// Register interest in file.
diff --git a/pkg/sentry/vfs/file_description_impl_util.go b/pkg/sentry/vfs/file_description_impl_util.go
index c1ab2f56c..a875fdeca 100644
--- a/pkg/sentry/vfs/file_description_impl_util.go
+++ b/pkg/sentry/vfs/file_description_impl_util.go
@@ -114,7 +114,7 @@ func (FileDescriptionDefaultImpl) Write(ctx context.Context, src usermem.IOSeque
 // file_operations::iterate == file_operations::iterate_shared == NULL in
 // Linux.
 func (FileDescriptionDefaultImpl) IterDirents(ctx context.Context, cb IterDirentsCallback) error {
-	return syserror.ENOTDIR
+	return linuxerr.ENOTDIR
 }
 
 // Seek implements FileDescriptionImpl.Seek analogously to
@@ -138,7 +138,7 @@ func (FileDescriptionDefaultImpl) ConfigureMMap(ctx context.Context, opts *memma
 // Ioctl implements FileDescriptionImpl.Ioctl analogously to
 // file_operations::unlocked_ioctl == NULL in Linux.
 func (FileDescriptionDefaultImpl) Ioctl(ctx context.Context, uio usermem.IO, args arch.SyscallArguments) (uintptr, error) {
-	return 0, syserror.ENOTTY
+	return 0, linuxerr.ENOTTY
 }
 
 // ListXattr implements FileDescriptionImpl.ListXattr analogously to
@@ -358,7 +358,7 @@ func (fd *DynamicBytesFileDescriptionImpl) Seek(ctx context.Context, offset int6
 // Preconditions: fd.mu must be locked.
 func (fd *DynamicBytesFileDescriptionImpl) pwriteLocked(ctx context.Context, src usermem.IOSequence, offset int64, opts WriteOptions) (int64, error) {
 	if opts.Flags&^(linux.RWF_HIPRI|linux.RWF_DSYNC|linux.RWF_SYNC) != 0 {
-		return 0, syserror.EOPNOTSUPP
+		return 0, linuxerr.EOPNOTSUPP
 	}
 	limit, err := CheckLimit(ctx, offset, src.NumBytes())
 	if err != nil {
diff --git a/pkg/sentry/vfs/inotify.go b/pkg/sentry/vfs/inotify.go
index e0624b2fa..088beb8e2 100644
--- a/pkg/sentry/vfs/inotify.go
+++ b/pkg/sentry/vfs/inotify.go
@@ -263,7 +263,7 @@ func (i *Inotify) Ioctl(ctx context.Context, uio usermem.IO, args arch.SyscallAr
 		return 0, err
 
 	default:
-		return 0, syserror.ENOTTY
+		return 0, linuxerr.ENOTTY
 	}
 }
 
diff --git a/pkg/sentry/vfs/memxattr/xattr.go b/pkg/sentry/vfs/memxattr/xattr.go
index efc9c6a5d..f0f82a4d6 100644
--- a/pkg/sentry/vfs/memxattr/xattr.go
+++ b/pkg/sentry/vfs/memxattr/xattr.go
@@ -24,7 +24,6 @@ import (
 	"gvisor.dev/gvisor/pkg/sentry/kernel/auth"
 	"gvisor.dev/gvisor/pkg/sentry/vfs"
 	"gvisor.dev/gvisor/pkg/sync"
-	"gvisor.dev/gvisor/pkg/syserror"
 )
 
 // SimpleExtendedAttributes implements extended attributes using a map of
@@ -55,7 +54,7 @@ func (x *SimpleExtendedAttributes) GetXattr(creds *auth.Credentials, mode linux.
 	// Check that the size of the buffer provided in getxattr(2) is large enough
 	// to contain the value.
 	if opts.Size != 0 && uint64(len(value)) > opts.Size {
-		return "", syserror.ERANGE
+		return "", linuxerr.ERANGE
 	}
 	return value, nil
 }
@@ -77,7 +76,7 @@ func (x *SimpleExtendedAttributes) SetXattr(creds *auth.Credentials, mode linux.
 
 	_, ok := x.xattrs[opts.Name]
 	if ok && opts.Flags&linux.XATTR_CREATE != 0 {
-		return syserror.EEXIST
+		return linuxerr.EEXIST
 	}
 	if !ok && opts.Flags&linux.XATTR_REPLACE != 0 {
 		return linuxerr.ENODATA
@@ -107,7 +106,7 @@ func (x *SimpleExtendedAttributes) ListXattr(creds *auth.Credentials, size uint6
 	}
 	x.mu.RUnlock()
 	if size != 0 && uint64(listSize) > size {
-		return nil, syserror.ERANGE
+		return nil, linuxerr.ERANGE
 	}
 	return names, nil
 }
diff --git a/pkg/sentry/vfs/vfs.go b/pkg/sentry/vfs/vfs.go
index cb9ed81d8..aeca262e3 100644
--- a/pkg/sentry/vfs/vfs.go
+++ b/pkg/sentry/vfs/vfs.go
@@ -279,7 +279,7 @@ func (vfs *VirtualFilesystem) LinkAt(ctx context.Context, creds *auth.Credential
 	if !newpop.Path.Begin.Ok() {
 		oldVD.DecRef(ctx)
 		if newpop.Path.Absolute {
-			return syserror.EEXIST
+			return linuxerr.EEXIST
 		}
 		return syserror.ENOENT
 	}
@@ -316,7 +316,7 @@ func (vfs *VirtualFilesystem) MkdirAt(ctx context.Context, creds *auth.Credentia
 		// pop.Path should not be empty in operations that create/delete files.
 		// This is consistent with mkdirat(dirfd, "", mode).
 		if pop.Path.Absolute {
-			return syserror.EEXIST
+			return linuxerr.EEXIST
 		}
 		return syserror.ENOENT
 	}
@@ -354,7 +354,7 @@ func (vfs *VirtualFilesystem) MknodAt(ctx context.Context, creds *auth.Credentia
 		// pop.Path should not be empty in operations that create/delete files.
 		// This is consistent with mknodat(dirfd, "", mode, dev).
 		if pop.Path.Absolute {
-			return syserror.EEXIST
+			return linuxerr.EEXIST
 		}
 		return syserror.ENOENT
 	}
@@ -639,7 +639,7 @@ func (vfs *VirtualFilesystem) SymlinkAt(ctx context.Context, creds *auth.Credent
 		// pop.Path should not be empty in operations that create/delete files.
 		// This is consistent with symlinkat(oldpath, newdirfd, "").
 		if pop.Path.Absolute {
-			return syserror.EEXIST
+			return linuxerr.EEXIST
 		}
 		return syserror.ENOENT
 	}
@@ -834,7 +834,7 @@ func (vfs *VirtualFilesystem) MkdirAllAt(ctx context.Context, currentPath string
 	switch {
 	case err == nil:
 		if stat.Mask&linux.STATX_TYPE == 0 || stat.Mode&linux.FileTypeMask != linux.ModeDirectory {
-			return syserror.ENOTDIR
+			return linuxerr.ENOTDIR
 		}
 		// Directory already exists.
 		return nil
diff --git a/pkg/syserror/syserror.go b/pkg/syserror/syserror.go
index 7fc6f8a7a..b24edb364 100644
--- a/pkg/syserror/syserror.go
+++ b/pkg/syserror/syserror.go
@@ -26,23 +26,16 @@ import (
 
 // The following variables have the same meaning as their syscall equivalent.
 var (
-	EEXIST     = error(unix.EEXIST)
-	EFAULT     = error(unix.EFAULT)
-	EIDRM      = error(unix.EIDRM)
-	EINTR      = error(unix.EINTR)
-	EIO        = error(unix.EIO)
-	EISDIR     = error(unix.EISDIR)
-	ENOENT     = error(unix.ENOENT)
-	ENOEXEC    = error(unix.ENOEXEC)
-	ENOMEM     = error(unix.ENOMEM)
-	ENOTSOCK   = error(unix.ENOTSOCK)
-	ENOSPC     = error(unix.ENOSPC)
-	ENOSYS     = error(unix.ENOSYS)
-	ENOTDIR    = error(unix.ENOTDIR)
-	ENOTTY     = error(unix.ENOTTY)
-	EOPNOTSUPP = error(unix.EOPNOTSUPP)
-	ERANGE     = error(unix.ERANGE)
-	ESRCH      = error(unix.ESRCH)
+	EIDRM    = error(unix.EIDRM)
+	EINTR    = error(unix.EINTR)
+	EIO      = error(unix.EIO)
+	EISDIR   = error(unix.EISDIR)
+	ENOENT   = error(unix.ENOENT)
+	ENOEXEC  = error(unix.ENOEXEC)
+	ENOMEM   = error(unix.ENOMEM)
+	ENOTSOCK = error(unix.ENOTSOCK)
+	ENOSPC   = error(unix.ENOSPC)
+	ENOSYS   = error(unix.ENOSYS)
 )
 
 var (
diff --git a/pkg/tcpip/link/tun/device.go b/pkg/tcpip/link/tun/device.go
index f3444e8b5..d23210503 100644
--- a/pkg/tcpip/link/tun/device.go
+++ b/pkg/tcpip/link/tun/device.go
@@ -126,7 +126,7 @@ func attachOrCreateNIC(s *stack.Stack, name, prefix string, linkCaps stack.LinkE
 				endpoint, ok := linkEP.(*tunEndpoint)
 				if !ok {
 					// Not a NIC created by tun device.
-					return nil, syserror.EOPNOTSUPP
+					return nil, linuxerr.EOPNOTSUPP
 				}
 				if !endpoint.TryIncRef() {
 					// Race detected: NIC got deleted in between.
diff --git a/pkg/usermem/bytes_io.go b/pkg/usermem/bytes_io.go
index 4c97b9136..777ac59a6 100644
--- a/pkg/usermem/bytes_io.go
+++ b/pkg/usermem/bytes_io.go
@@ -19,7 +19,6 @@ import (
 	"gvisor.dev/gvisor/pkg/errors/linuxerr"
 	"gvisor.dev/gvisor/pkg/hostarch"
 	"gvisor.dev/gvisor/pkg/safemem"
-	"gvisor.dev/gvisor/pkg/syserror"
 )
 
 const maxInt = int(^uint(0) >> 1)
@@ -94,11 +93,11 @@ func (b *BytesIO) rangeCheck(addr hostarch.Addr, length int) (int, error) {
 	}
 	max := hostarch.Addr(len(b.Bytes))
 	if addr >= max {
-		return 0, syserror.EFAULT
+		return 0, linuxerr.EFAULT
 	}
 	end, ok := addr.AddLength(uint64(length))
 	if !ok || end > max {
-		return int(max - addr), syserror.EFAULT
+		return int(max - addr), linuxerr.EFAULT
 	}
 	return length, nil
 }
diff --git a/pkg/usermem/usermem.go b/pkg/usermem/usermem.go
index ca852558e..cde1038ed 100644
--- a/pkg/usermem/usermem.go
+++ b/pkg/usermem/usermem.go
@@ -26,7 +26,6 @@ import (
 	"gvisor.dev/gvisor/pkg/gohacks"
 	"gvisor.dev/gvisor/pkg/hostarch"
 	"gvisor.dev/gvisor/pkg/safemem"
-	"gvisor.dev/gvisor/pkg/syserror"
 )
 
 // IO provides access to the contents of a virtual memory space.
@@ -163,7 +162,7 @@ func (rw *IOReadWriter) Read(dst []byte) (int, error) {
 		// Disallow wraparound.
 		rw.Addr = ^hostarch.Addr(0)
 		if err != nil {
-			err = syserror.EFAULT
+			err = linuxerr.EFAULT
 		}
 	}
 	return n, err
@@ -179,7 +178,7 @@ func (rw *IOReadWriter) Write(src []byte) (int, error) {
 		// Disallow wraparound.
 		rw.Addr = ^hostarch.Addr(0)
 		if err != nil {
-			err = syserror.EFAULT
+			err = linuxerr.EFAULT
 		}
 	}
 	return n, err
@@ -214,7 +213,7 @@ func CopyStringIn(ctx context.Context, uio IO, addr hostarch.Addr, maxlen int, o
 		}
 		end, ok := addr.AddLength(uint64(readlen))
 		if !ok {
-			return gohacks.StringFromImmutableBytes(buf[:done]), syserror.EFAULT
+			return gohacks.StringFromImmutableBytes(buf[:done]), linuxerr.EFAULT
 		}
 		// Shorten the read to avoid crossing page boundaries, since faulting
 		// in a page unnecessarily is expensive. This also ensures that partial