Re: [PATCH v23 28/28] mm: Introduce PROT_SHSTK for shadow stack
On Tue, Mar 16, 2021 at 08:10:54AM -0700, Yu-cheng Yu wrote:
> There are three possible options to create a shadow stack allocation API:
> an arch_prctl, a new syscall, or adding PROT_SHSTK to mmap()/mprotect().
> Each has its advantages and compromises.
>
> An arch_prctl() is the least intrusive. However, the existing x86
> arch_prctl() takes only two parameters. Multiple parameters must be
> passed in a memory buffer. There is a proposal to pass more parameters in
> registers [1], but no active discussion on that.
>
> A new syscall minimizes compatibility issues and offers an extensible frame
> work to other architectures, but this will likely result in some overlap of
> mmap()/mprotect().
>
> The introduction of PROT_SHSTK to mmap()/mprotect() takes advantage of
> existing APIs. The x86-specific PROT_SHSTK is translated to VM_SHSTK and
> a shadow stack mapping is created without reinventing the wheel. There are
> potential pitfalls though. The most obvious one would be using this as a
> bypass to shadow stack protection. However, the attacker would have to get
> to the syscall first.
>
> [1] https://lore.kernel.org/lkml/20200828121624.108243-1-hjl.to...@gmail.com/
>
> Signed-off-by: Yu-cheng Yu
> Reviewed-by: Kees Cook
> ---
> arch/x86/include/asm/mman.h | 57 +++-
> arch/x86/include/uapi/asm/mman.h | 1 +
> include/linux/mm.h | 1 +
> mm/mmap.c| 8 -
> 4 files changed, 65 insertions(+), 2 deletions(-)
>
> diff --git a/arch/x86/include/asm/mman.h b/arch/x86/include/asm/mman.h
> index 629f6c81263a..bd94e30b5d34 100644
> --- a/arch/x86/include/asm/mman.h
> +++ b/arch/x86/include/asm/mman.h
> @@ -20,11 +20,66 @@
> ((vm_flags) & VM_PKEY_BIT2 ? _PAGE_PKEY_BIT2 : 0) | \
> ((vm_flags) & VM_PKEY_BIT3 ? _PAGE_PKEY_BIT3 : 0))
>
> -#define arch_calc_vm_prot_bits(prot, key) ( \
> +#define pkey_vm_prot_bits(prot, key) ( \
> ((key) & 0x1 ? VM_PKEY_BIT0 : 0) | \
> ((key) & 0x2 ? VM_PKEY_BIT1 : 0) | \
> ((key) & 0x4 ? VM_PKEY_BIT2 : 0) | \
> ((key) & 0x8 ? VM_PKEY_BIT3 : 0))
> +#else
> +#define pkey_vm_prot_bits(prot, key) (0)
> +#endif
> +
> +static inline unsigned long arch_calc_vm_prot_bits(unsigned long prot,
> +unsigned long pkey)
> +{
> + unsigned long vm_prot_bits = pkey_vm_prot_bits(prot, pkey);
> +
> + if (!(prot & PROT_WRITE) && (prot & PROT_SHSTK))
> + vm_prot_bits |= VM_SHSTK;
> +
> + return vm_prot_bits;
> +}
> +
> +#define arch_calc_vm_prot_bits(prot, pkey) arch_calc_vm_prot_bits(prot, pkey)
> +
> +#ifdef CONFIG_X86_CET
> +static inline bool arch_validate_prot(unsigned long prot, unsigned long addr)
> +{
> + unsigned long valid = PROT_READ | PROT_WRITE | PROT_EXEC | PROT_SEM;
> +
> + if (prot & ~(valid | PROT_SHSTK))
Why PROT_SHSTK is not part of valid?
> + return false;
> +
> + if (prot & PROT_SHSTK) {
> + struct vm_area_struct *vma;
> +
> + if (!current->thread.cet.shstk_size)
> + return false;
> +
> + /*
> + * A shadow stack mapping is indirectly writable by only
> + * the CALL and WRUSS instructions, but not other write
> + * instructions). PROT_SHSTK and PROT_WRITE are mutually
> + * exclusive.
> + */
> + if (prot & PROT_WRITE)
> + return false;
> +
> + vma = find_vma(current->mm, addr);
> + if (!vma)
> + return false;
NAK.
This is racy. arch_validate_prot() called outside of mmap_lock and the vma
may be freed or modified under us.
do_mprotect_pkey() already calls find_vma() with the right locking. Maybe
re-strucure do_mprotect_pkey() to call arch_validate_prot() after
find_vma() and pass down the vma?
> +
> + /*
> + * Shadow stack cannot be backed by a file or shared.
> + */
> + if (vma->vm_file || (vma->vm_flags & VM_SHARED))
> + return false;
> + }
> +
> + return true;
> +}
> +
> +#define arch_validate_prot arch_validate_prot
> #endif
>
> #endif /* _ASM_X86_MMAN_H */
> diff --git a/arch/x86/include/uapi/asm/mman.h
> b/arch/x86/include/uapi/asm/mman.h
> index 3ce1923e6ed9..39bb7db344a6 100644
> --- a/arch/x86/include/uapi/asm/mman.h
> +++ b/arch/x86/include/uapi/asm/mman.h
> @@ -4,6 +4,7 @@
>
> #define MAP_32BIT0x40/* only give out 32bit addresses */
>
> +#define PROT_SHSTK 0x10/* shadow stack pages */
>
> #include
>
> diff --git a/include/linux/mm.h b/include/linux/mm.h
> index e178be052419..40c4b0fe7cc4 100644
> --- a/include/linux/mm.h
> +++ b/include/linux/mm.h
> @@ -342,6 +342,7 @@ extern unsigned int kobjsize(const void *objp);
>
> #if defined(CONF
[PATCH v23 28/28] mm: Introduce PROT_SHSTK for shadow stack
There are three possible options to create a shadow stack allocation API:
an arch_prctl, a new syscall, or adding PROT_SHSTK to mmap()/mprotect().
Each has its advantages and compromises.
An arch_prctl() is the least intrusive. However, the existing x86
arch_prctl() takes only two parameters. Multiple parameters must be
passed in a memory buffer. There is a proposal to pass more parameters in
registers [1], but no active discussion on that.
A new syscall minimizes compatibility issues and offers an extensible frame
work to other architectures, but this will likely result in some overlap of
mmap()/mprotect().
The introduction of PROT_SHSTK to mmap()/mprotect() takes advantage of
existing APIs. The x86-specific PROT_SHSTK is translated to VM_SHSTK and
a shadow stack mapping is created without reinventing the wheel. There are
potential pitfalls though. The most obvious one would be using this as a
bypass to shadow stack protection. However, the attacker would have to get
to the syscall first.
[1] https://lore.kernel.org/lkml/20200828121624.108243-1-hjl.to...@gmail.com/
Signed-off-by: Yu-cheng Yu
Reviewed-by: Kees Cook
---
arch/x86/include/asm/mman.h | 57 +++-
arch/x86/include/uapi/asm/mman.h | 1 +
include/linux/mm.h | 1 +
mm/mmap.c| 8 -
4 files changed, 65 insertions(+), 2 deletions(-)
diff --git a/arch/x86/include/asm/mman.h b/arch/x86/include/asm/mman.h
index 629f6c81263a..bd94e30b5d34 100644
--- a/arch/x86/include/asm/mman.h
+++ b/arch/x86/include/asm/mman.h
@@ -20,11 +20,66 @@
((vm_flags) & VM_PKEY_BIT2 ? _PAGE_PKEY_BIT2 : 0) | \
((vm_flags) & VM_PKEY_BIT3 ? _PAGE_PKEY_BIT3 : 0))
-#define arch_calc_vm_prot_bits(prot, key) (\
+#define pkey_vm_prot_bits(prot, key) ( \
((key) & 0x1 ? VM_PKEY_BIT0 : 0) | \
((key) & 0x2 ? VM_PKEY_BIT1 : 0) | \
((key) & 0x4 ? VM_PKEY_BIT2 : 0) | \
((key) & 0x8 ? VM_PKEY_BIT3 : 0))
+#else
+#define pkey_vm_prot_bits(prot, key) (0)
+#endif
+
+static inline unsigned long arch_calc_vm_prot_bits(unsigned long prot,
+ unsigned long pkey)
+{
+ unsigned long vm_prot_bits = pkey_vm_prot_bits(prot, pkey);
+
+ if (!(prot & PROT_WRITE) && (prot & PROT_SHSTK))
+ vm_prot_bits |= VM_SHSTK;
+
+ return vm_prot_bits;
+}
+
+#define arch_calc_vm_prot_bits(prot, pkey) arch_calc_vm_prot_bits(prot, pkey)
+
+#ifdef CONFIG_X86_CET
+static inline bool arch_validate_prot(unsigned long prot, unsigned long addr)
+{
+ unsigned long valid = PROT_READ | PROT_WRITE | PROT_EXEC | PROT_SEM;
+
+ if (prot & ~(valid | PROT_SHSTK))
+ return false;
+
+ if (prot & PROT_SHSTK) {
+ struct vm_area_struct *vma;
+
+ if (!current->thread.cet.shstk_size)
+ return false;
+
+ /*
+* A shadow stack mapping is indirectly writable by only
+* the CALL and WRUSS instructions, but not other write
+* instructions). PROT_SHSTK and PROT_WRITE are mutually
+* exclusive.
+*/
+ if (prot & PROT_WRITE)
+ return false;
+
+ vma = find_vma(current->mm, addr);
+ if (!vma)
+ return false;
+
+ /*
+* Shadow stack cannot be backed by a file or shared.
+*/
+ if (vma->vm_file || (vma->vm_flags & VM_SHARED))
+ return false;
+ }
+
+ return true;
+}
+
+#define arch_validate_prot arch_validate_prot
#endif
#endif /* _ASM_X86_MMAN_H */
diff --git a/arch/x86/include/uapi/asm/mman.h b/arch/x86/include/uapi/asm/mman.h
index 3ce1923e6ed9..39bb7db344a6 100644
--- a/arch/x86/include/uapi/asm/mman.h
+++ b/arch/x86/include/uapi/asm/mman.h
@@ -4,6 +4,7 @@
#define MAP_32BIT 0x40/* only give out 32bit addresses */
+#define PROT_SHSTK 0x10/* shadow stack pages */
#include
diff --git a/include/linux/mm.h b/include/linux/mm.h
index e178be052419..40c4b0fe7cc4 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -342,6 +342,7 @@ extern unsigned int kobjsize(const void *objp);
#if defined(CONFIG_X86)
# define VM_PATVM_ARCH_1 /* PAT reserves whole VMA at
once (x86) */
+# define VM_ARCH_CLEAR VM_SHSTK
#elif defined(CONFIG_PPC)
# define VM_SAOVM_ARCH_1 /* Strong Access Ordering
(powerpc) */
#elif defined(CONFIG_PARISC)
diff --git a/mm/mmap.c b/mm/mmap.c
index 99077171010b..934cb3cbe952 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -1481,6 +1481,12 @@ unsigned long do_mmap(struct file *file, unsigned long
addr,
struct inode *inode = file_inode(file);
unsigned long flags_mask;
