Kernel Self Protection Project/Recommended Settings

Sometimes people ask the Kernel Self Protection Project what a secure set of build CONFIGs and runtime settings are. This is a brain-dump of the various options for a particularly paranoid system:

CONFIGs

# Report BUG() conditions and kill the offending process.
CONFIG_BUG=y

# Make sure kernel page tables have safe permissions.
CONFIG_DEBUG_KERNEL=y (prior to v4.11, needed to select CONFIG_DEBUG_RODATA below)
CONFIG_DEBUG_RODATA=y (prior to v4.11)
CONFIG_STRICT_KERNEL_RWX=y (since v4.11)

# Report any dangerous memory permissions (not available on all archs).
CONFIG_DEBUG_WX=y

# Use -fstack-protector-strong (gcc 4.9+) for best stack canary coverage.
# Prior to v4.18, these are:
#  CONFIG_CC_STACKPROTECTOR=y
#  CONFIG_CC_STACKPROTECTOR_STRONG=y
CONFIG_STACKPROTECTOR=y
CONFIG_STACKPROTECTOR_STRONG=y

# Do not allow direct physical memory access (but if you must have it, at least enable STRICT mode...)
# CONFIG_DEVMEM is not set
CONFIG_STRICT_DEVMEM=y
CONFIG_IO_STRICT_DEVMEM=y

# Provides some protections against SYN flooding.
CONFIG_SYN_COOKIES=y

# Perform additional validation of various commonly targeted structures.
CONFIG_DEBUG_CREDENTIALS=y
CONFIG_DEBUG_NOTIFIERS=y
CONFIG_DEBUG_LIST=y
CONFIG_DEBUG_SG=y
CONFIG_BUG_ON_DATA_CORRUPTION=y
CONFIG_SCHED_STACK_END_CHECK=y

# Provide userspace with seccomp BPF API for syscall attack surface reduction.
CONFIG_SECCOMP=y
CONFIG_SECCOMP_FILTER=y

# Provide userspace with ptrace ancestry protections.
CONFIG_SECURITY=y
CONFIG_SECURITY_YAMA=y

# Perform usercopy bounds checking. (And disable fallback to gain full whitelist enforcement.)
CONFIG_HARDENED_USERCOPY=y
# CONFIG_HARDENED_USERCOPY_FALLBACK is not set
# CONFIG_HARDENED_USERCOPY_PAGESPAN is not set

# Randomize allocator freelists, harden metadata.
CONFIG_SLAB_FREELIST_RANDOM=y
CONFIG_SLAB_FREELIST_HARDENED=y

# Randomize high-order page allocation freelist.
CONFIG_SHUFFLE_PAGE_ALLOCATOR=y

# Allow allocator validation checking to be enabled (see "slub_debug=P" below).
CONFIG_SLUB_DEBUG=y

# Wipe higher-level memory allocations when they are freed (needs "page_poison=1" command line below).
# (If you can afford even more performance penalty, leave CONFIG_PAGE_POISONING_NO_SANITY=n)
CONFIG_PAGE_POISONING=y
CONFIG_PAGE_POISONING_NO_SANITY=y
CONFIG_PAGE_POISONING_ZERO=y

# Wipe slab and page allocations (since v5.3)
# Instead of "slub_debug=P" and "page_poison=1", a single place can control memory allocation wiping now.
# The init_on_free is only needed if there is concern about minimizing stale data lifetime.
CONFIG_INIT_ON_ALLOC_DEFAULT_ON=y
CONFIG_INIT_ON_FREE_DEFAULT_ON=y

# Initialize all stack variables on function entry. (Clang builds only. For GCC, see CONFIG_GCC_PLUGIN_STRUCTLEAK_BYREF_ALL=y below)
CONFIG_INIT_STACK_ALL=y

# Adds guard pages to kernel stacks (not all architectures support this yet).
CONFIG_VMAP_STACK=y

# Perform extensive checks on reference counting.
CONFIG_REFCOUNT_FULL=y

# Check for memory copies that might overflow a structure in str*() and mem*() functions both at build-time and run-time.
CONFIG_FORTIFY_SOURCE=y

# Dangerous; enabling this allows direct physical memory writing.
# CONFIG_ACPI_CUSTOM_METHOD is not set

# Dangerous; enabling this disables brk ASLR.
# CONFIG_COMPAT_BRK is not set

# Dangerous; enabling this allows direct kernel memory writing.
# CONFIG_DEVKMEM is not set

# Dangerous; exposes kernel text image layout.
# CONFIG_PROC_KCORE is not set

# Dangerous; enabling this disables VDSO ASLR.
# CONFIG_COMPAT_VDSO is not set

# Dangerous; enabling this allows replacement of running kernel.
# CONFIG_KEXEC is not set

# Dangerous; enabling this allows replacement of running kernel.
# CONFIG_HIBERNATION is not set

# Prior to v4.1, assists heap memory attacks; best to keep interface disabled.
# CONFIG_INET_DIAG is not set

# Easily confused by misconfigured userspace, keep off.
# CONFIG_BINFMT_MISC is not set

# Use the modern PTY interface (devpts) only.
# CONFIG_LEGACY_PTYS is not set

# If SELinux can be disabled at runtime, the LSM structures cannot be read-only; keep off.
# CONFIG_SECURITY_SELINUX_DISABLE is not set

# Reboot devices immediately if kernel experiences an Oops.
CONFIG_PANIC_ON_OOPS=y
CONFIG_PANIC_TIMEOUT=-1

# Keep root from altering kernel memory via loadable modules.
# CONFIG_MODULES is not set

# But if CONFIG_MODULE=y is needed, at least they must be signed with a per-build key.
CONFIG_DEBUG_SET_MODULE_RONX=y (prior to v4.11)
CONFIG_STRICT_MODULE_RWX=y (since v4.11)
CONFIG_MODULE_SIG=y
CONFIG_MODULE_SIG_FORCE=y
CONFIG_MODULE_SIG_ALL=y
CONFIG_MODULE_SIG_SHA512=y
CONFIG_MODULE_SIG_HASH="sha512"
CONFIG_MODULE_SIG_KEY="certs/signing_key.pem"

GCC plugins

# Enable GCC Plugins
CONFIG_GCC_PLUGINS=y

# Gather additional entropy at boot time for systems that may not have appropriate entropy sources.
CONFIG_GCC_PLUGIN_LATENT_ENTROPY=y

# Force all structures to be initialized before they are passed to other functions.
# When building with GCC:
CONFIG_GCC_PLUGIN_STRUCTLEAK=y
CONFIG_GCC_PLUGIN_STRUCTLEAK_BYREF_ALL=y

# Wipe stack contents on syscall exit (reduces stale data lifetime in stack)
CONFIG_GCC_PLUGIN_STACKLEAK=y

# Randomize the layout of system structures. This may have dramatic performance impact, so
# use with caution or also use CONFIG_GCC_PLUGIN_RANDSTRUCT_PERFORMANCE=y
CONFIG_GCC_PLUGIN_RANDSTRUCT=y

x86_32

# On 32-bit kernels, require PAE for NX bit support.
# CONFIG_M486 is not set
# CONFIG_HIGHMEM4G is not set
CONFIG_HIGHMEM64G=y
CONFIG_X86_PAE=y

# Disallow allocating the first 64k of memory.
CONFIG_DEFAULT_MMAP_MIN_ADDR=65536

# Randomize position of kernel.
CONFIG_RANDOMIZE_BASE=y

x86_64

# Full 64-bit means PAE and NX bit.
CONFIG_X86_64=y

# Disallow allocating the first 64k of memory.
CONFIG_DEFAULT_MMAP_MIN_ADDR=65536

# Randomize position of kernel and memory.
CONFIG_RANDOMIZE_BASE=y
CONFIG_RANDOMIZE_MEMORY=y

# Modern libc no longer needs a fixed-position mapping in userspace, remove it as a possible target.
CONFIG_LEGACY_VSYSCALL_NONE=y

# Enable Kernel Page Table Isolation to remove an entire class of cache timing side-channels.
CONFIG_PAGE_TABLE_ISOLATION=y

# Remove additional attack surface, unless you really need them.
# CONFIG_IA32_EMULATION is not set
# CONFIG_X86_X32 is not set
# CONFIG_MODIFY_LDT_SYSCALL is not set

arm

# Disallow allocating the first 32k of memory (cannot be 64k due to ARM loader).
CONFIG_DEFAULT_MMAP_MIN_ADDR=32768

# For maximal userspace memory area (and maximum ASLR).
CONFIG_VMSPLIT_3G=y

# If building an old out-of-tree Qualcomm kernel, this is similar to CONFIG_STRICT_KERNEL_RWX.
CONFIG_STRICT_MEMORY_RWX=y

# Make sure PXN/PAN emulation is enabled.
CONFIG_CPU_SW_DOMAIN_PAN=y

# Dangerous; old interfaces and needless additional attack surface.
# CONFIG_OABI_COMPAT is unset

arm64

# Disallow allocating the first 32k of memory (cannot be 64k due to ARM loader).
CONFIG_DEFAULT_MMAP_MIN_ADDR=32768

# Randomize position of kernel (requires UEFI RNG or bootloader support for /chosen/kaslr-seed DT property).
CONFIG_RANDOMIZE_BASE=y

# Make sure PAN emulation is enabled.
CONFIG_ARM64_SW_TTBR0_PAN=y

# Enable Kernel Page Table Isolation to remove an entire class of cache timing side-channels.
CONFIG_UNMAP_KERNEL_AT_EL0=y

kernel command line options

# Enable slub/slab allocator free poisoning (requires CONFIG_SLUB_DEBUG=y above).
slub_debug=P

# Enable buddy allocator free poisoning (requires CONFIG_PAGE_POISONING=y above).
page_poison=1

# Wipe slab and page allocations (supersedes "slub_debug=P" and "page_poison=1" above, since v5.3)
# See CONFIG_INIT_ON_ALLOC_DEFAULT_ON=1 and CONFIG_INIT_ON_FREE_DEFAULT_ON=1 above.
init_on_alloc=1
init_on_free=1

# Disable slab merging (makes many heap overflow attacks more difficult).
slab_nomerge

# Always enable Kernel Page Table Isolation, even if the CPU claims it is safe from Meltdown.
pti=on

x86_64

# Remove vsyscall entirely to avoid it being a fixed-position ROP target of any kind.
# (Same as CONFIG_LEGACY_VSYSCALL_NONE=y above.)
vsyscall=none

sysctls

# Try to keep kernel address exposures out of various /proc files (kallsyms, modules, etc).
kernel.kptr_restrict = 1

# Avoid kernel memory address exposures via dmesg.
kernel.dmesg_restrict = 1

# Block non-uid-0 profiling (needs distro patch, otherwise this is the same as "= 2")
kernel.perf_event_paranoid = 3

# Turn off kexec, even if it's built in.
kernel.kexec_load_disabled = 1

# Avoid non-ancestor ptrace access to running processes and their credentials.
kernel.yama.ptrace_scope = 1

# Disable User Namespaces, as it opens up a large attack surface to unprivileged users.
user.max_user_namespaces = 0

# Turn off unprivileged eBPF access.
kernel.unprivileged_bpf_disabled = 1

# Turn on BPF JIT hardening, if the JIT is enabled.
net.core.bpf_jit_harden = 2