Merge branch 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 fix from Ingo Molnar:
"Fix x86 syscall exit code bug that resulted in spurious non-execution
of TIF-driven user-return worklets, causing big trouble for things
like KVM that rely on user notifiers for correctness of their vcpu
model, causing crashes like double faults"
* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/asm/entry: Check for syscall exit work with IRQs disabled
diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S
index 1d74d16..2babb39 100644
--- a/arch/x86/kernel/entry_64.S
+++ b/arch/x86/kernel/entry_64.S
@@ -364,12 +364,21 @@
* Has incomplete stack frame and undefined top of stack.
*/
ret_from_sys_call:
- testl $_TIF_ALLWORK_MASK,TI_flags+THREAD_INFO(%rsp,RIP-ARGOFFSET)
- jnz int_ret_from_sys_call_fixup /* Go the the slow path */
-
LOCKDEP_SYS_EXIT
DISABLE_INTERRUPTS(CLBR_NONE)
TRACE_IRQS_OFF
+
+ /*
+ * We must check ti flags with interrupts (or at least preemption)
+ * off because we must *never* return to userspace without
+ * processing exit work that is enqueued if we're preempted here.
+ * In particular, returning to userspace with any of the one-shot
+ * flags (TIF_NOTIFY_RESUME, TIF_USER_RETURN_NOTIFY, etc) set is
+ * very bad.
+ */
+ testl $_TIF_ALLWORK_MASK,TI_flags+THREAD_INFO(%rsp,RIP-ARGOFFSET)
+ jnz int_ret_from_sys_call_fixup /* Go the the slow path */
+
CFI_REMEMBER_STATE
/*
* sysretq will re-enable interrupts:
@@ -386,7 +395,7 @@
int_ret_from_sys_call_fixup:
FIXUP_TOP_OF_STACK %r11, -ARGOFFSET
- jmp int_ret_from_sys_call
+ jmp int_ret_from_sys_call_irqs_off
/* Do syscall tracing */
tracesys:
@@ -432,6 +441,7 @@
GLOBAL(int_ret_from_sys_call)
DISABLE_INTERRUPTS(CLBR_NONE)
TRACE_IRQS_OFF
+int_ret_from_sys_call_irqs_off:
movl $_TIF_ALLWORK_MASK,%edi
/* edi: mask to check */
GLOBAL(int_with_check)
