[PATCH 4.4 11/36] x86/syscall: Sanitize syscall table de-references under speculation fix

[Greg Kroah-Hartman]

4.4-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Jiri Slaby <jslaby@xxxxxxx>

In 4.4.118, we have commit c8961332d6da (x86/syscall: Sanitize syscall
table de-references under speculation), which is a backport of upstream
commit 2fbd7af5af86. But it fixed only the C part of the upstream patch
-- the IA32 sysentry. So it ommitted completely the assembly part -- the
64bit sysentry.

Fix that in this patch by explicit array_index_mask_nospec written in
assembly. The same was used in lib/getuser.S.

However, to have "sbb" working properly, we have to switch from "cmp"
against (NR_syscalls-1) to (NR_syscalls), otherwise the last syscall
number would be "and"ed by 0. It is because the original "ja" relies on
"CF" or "ZF", but we rely only on "CF" in "sbb". That means: switch to
"jae" conditional jump too.

Final note: use rcx for mask as this is exactly what is overwritten by
the 4th syscall argument (r10) right after.

Reported-by: Jan Beulich <JBeulich@xxxxxxxx>
Cc: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
Cc: Dan Williams <dan.j.williams@xxxxxxxxx>
Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
Cc: linux-arch@xxxxxxxxxxxxxxx
Cc: kernel-hardening@xxxxxxxxxxxxxxxxxx
Cc: gregkh@xxxxxxxxxxxxxxxxxxx
Cc: Andy Lutomirski <luto@xxxxxxxxxx>
Cc: alan@xxxxxxxxxxxxxxx
Cc: Jinpu Wang <jinpu.wang@xxxxxxxxxxxxxxxx>
Signed-off-by: Jiri Slaby <jslaby@xxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
---
 arch/x86/entry/entry_64.S |   16 ++++++++++------
 1 file changed, 10 insertions(+), 6 deletions(-)

--- a/arch/x86/entry/entry_64.S
+++ b/arch/x86/entry/entry_64.S
@@ -178,12 +178,14 @@ GLOBAL(entry_SYSCALL_64_after_swapgs)
 	jnz	tracesys
 entry_SYSCALL_64_fastpath:
 #if __SYSCALL_MASK == ~0
-	cmpq	$__NR_syscall_max, %rax
+	cmpq	$NR_syscalls, %rax
 #else
 	andl	$__SYSCALL_MASK, %eax
-	cmpl	$__NR_syscall_max, %eax
+	cmpl	$NR_syscalls, %eax
 #endif
-	ja	1f				/* return -ENOSYS (already in pt_regs->ax) */
+	jae	1f				/* return -ENOSYS (already in pt_regs->ax) */
+	sbb	%rcx, %rcx			/* array_index_mask_nospec() */
+	and	%rcx, %rax
 	movq	%r10, %rcx
 #ifdef CONFIG_RETPOLINE
 	movq	sys_call_table(, %rax, 8), %rax
@@ -276,12 +278,14 @@ tracesys_phase2:
 	RESTORE_C_REGS_EXCEPT_RAX
 	RESTORE_EXTRA_REGS
 #if __SYSCALL_MASK == ~0
-	cmpq	$__NR_syscall_max, %rax
+	cmpq	$NR_syscalls, %rax
 #else
 	andl	$__SYSCALL_MASK, %eax
-	cmpl	$__NR_syscall_max, %eax
+	cmpl	$NR_syscalls, %eax
 #endif
-	ja	1f				/* return -ENOSYS (already in pt_regs->ax) */
+	jae	1f				/* return -ENOSYS (already in pt_regs->ax) */
+	sbb	%rcx, %rcx			/* array_index_mask_nospec() */
+	and	%rcx, %rax
 	movq	%r10, %rcx			/* fixup for C */
 #ifdef CONFIG_RETPOLINE
 	movq	sys_call_table(, %rax, 8), %rax