Re: [RFC patch 10/15] x86/entry: Move irq tracing to C code

[Peter Zijlstra]

On Mon, Sep 23, 2019 at 01:55:51PM +0200, Peter Zijlstra wrote:
> On Mon, Sep 23, 2019 at 01:49:20PM +0200, Peter Zijlstra wrote:
> > While walking the kids to school I wondered WTH we need to call
> > TRACE_IRQS_OFF in the first place. If this is the return from exception
> > path, interrupts had better be disabled already (in exception enter).
> > 
> > For entry_64.S we have:
> > 
> >   - idtentry_part; which does TRACE_IRQS_OFF at the start and error_exit
> >     at the end.
> > 
> >   - xen_do_hypervisor_callback, xen_failsafe_callback -- which are
> >     confusing.
> > 
> > So in the normal case, it appears we can simply do:
> > 
> > diff --git a/arch/x86/entry/entry_64.S b/arch/x86/entry/entry_64.S
> > index b7c3ea4cb19d..e9cf59ac554e 100644
> > --- a/arch/x86/entry/entry_64.S
> > +++ b/arch/x86/entry/entry_64.S
> > @@ -1368,8 +1368,6 @@ END(error_entry)
> >  
> >  ENTRY(error_exit)
> >  	UNWIND_HINT_REGS
> > -	DISABLE_INTERRUPTS(CLBR_ANY)
> > -	TRACE_IRQS_OFF
> >  	testb	$3, CS(%rsp)
> >  	jz	retint_kernel
> >  	jmp	retint_user
> > 
> > and all should be well. This leaves Xen...
> > 
> > For entry_32.S it looks like nothing uses 'resume_userspace' so that
> > ENTRY can go away. Which then leaves:
> > 
> >  * ret_from_intr:
> >   - common_spurious: TRACE_IRQS_OFF
> >   - common_interrupt: TRACE_IRQS_OFF
> >   - BUILD_INTERRUPT3: TRACE_IRQS_OFF
> >   - xen_do_upcall: ...
> > 
> >  * ret_from_exception:
> >   - xen_failsafe_callback: ...
> >   - common_exception_read_cr2: TRACE_IRQS_OFF
> >   - common_exception: TRACE_IRQS_OFF
> >   - int3: TRACE_IRQS_OFF
> > 
> > Which again suggests:
> > 
> > diff --git a/arch/x86/entry/entry_32.S b/arch/x86/entry/entry_32.S
> > index f83ca5aa8b77..7a19e7413a8e 100644
> > --- a/arch/x86/entry/entry_32.S
> > +++ b/arch/x86/entry/entry_32.S
> > @@ -825,9 +825,6 @@ END(ret_from_fork)
> >  	cmpl	$USER_RPL, %eax
> >  	jb	restore_all_kernel		# not returning to v8086 or userspace
> >  
> > -ENTRY(resume_userspace)
> > -	DISABLE_INTERRUPTS(CLBR_ANY)
> > -	TRACE_IRQS_OFF
> >  	movl	%esp, %eax
> >  	call	prepare_exit_to_usermode
> >  	jmp	restore_all
> > 
> > with the notable exception (oh teh pun!) being Xen... _again_.
> > 
> > With these patchlets on, we'd want prepare_exit_to_usermode() to
> > validate the IRQ state, but AFAICT it _should_ all just 'work' (famous
> > last words).
> > 
> > Cc Juergen because Xen
> 
> Arrgh.. faults!! they do local_irq_enable() but never disable them
> again. Arguably those functions should be symmetric and restore IF when
> they muck with it instead of relying on the exit path fixing things up.

---
diff --git a/arch/x86/entry/entry_32.S b/arch/x86/entry/entry_32.S
index f83ca5aa8b77..7a19e7413a8e 100644
--- a/arch/x86/entry/entry_32.S
+++ b/arch/x86/entry/entry_32.S
@@ -825,9 +825,6 @@ END(ret_from_fork)
 	cmpl	$USER_RPL, %eax
 	jb	restore_all_kernel		# not returning to v8086 or userspace
 
-ENTRY(resume_userspace)
-	DISABLE_INTERRUPTS(CLBR_ANY)
-	TRACE_IRQS_OFF
 	movl	%esp, %eax
 	call	prepare_exit_to_usermode
 	jmp	restore_all
diff --git a/arch/x86/entry/entry_64.S b/arch/x86/entry/entry_64.S
index b7c3ea4cb19d..e9cf59ac554e 100644
--- a/arch/x86/entry/entry_64.S
+++ b/arch/x86/entry/entry_64.S
@@ -1368,8 +1368,6 @@ END(error_entry)
 
 ENTRY(error_exit)
 	UNWIND_HINT_REGS
-	DISABLE_INTERRUPTS(CLBR_ANY)
-	TRACE_IRQS_OFF
 	testb	$3, CS(%rsp)
 	jz	retint_kernel
 	jmp	retint_user
diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c
index 4bb0f8447112..663d44c68f67 100644
--- a/arch/x86/kernel/traps.c
+++ b/arch/x86/kernel/traps.c
@@ -276,6 +276,7 @@ static void do_error_trap(struct pt_regs *regs, long error_code, char *str,
 			NOTIFY_STOP) {
 		cond_local_irq_enable(regs);
 		do_trap(trapnr, signr, str, regs, error_code, sicode, addr);
+		cond_local_irq_disable(regs);
 	}
 }
 
@@ -501,6 +502,7 @@ dotraplinkage void do_bounds(struct pt_regs *regs, long error_code)
 		die("bounds", regs, error_code);
 	}
 
+	cond_local_irq_disable(regs);
 	return;
 
 exit_trap:
@@ -512,6 +514,7 @@ dotraplinkage void do_bounds(struct pt_regs *regs, long error_code)
 	 * time..
 	 */
 	do_trap(X86_TRAP_BR, SIGSEGV, "bounds", regs, error_code, 0, NULL);
+	cond_local_irq_disable(regs);
 }
 
 dotraplinkage void
@@ -525,19 +528,19 @@ do_general_protection(struct pt_regs *regs, long error_code)
 
 	if (static_cpu_has(X86_FEATURE_UMIP)) {
 		if (user_mode(regs) && fixup_umip_exception(regs))
-			return;
+			goto exit_trap;
 	}
 
 	if (v8086_mode(regs)) {
 		local_irq_enable();
 		handle_vm86_fault((struct kernel_vm86_regs *) regs, error_code);
-		return;
+		goto exit_trap;
 	}
 
 	tsk = current;
 	if (!user_mode(regs)) {
 		if (fixup_exception(regs, X86_TRAP_GP, error_code, 0))
-			return;
+			goto exit_trap;
 
 		tsk->thread.error_code = error_code;
 		tsk->thread.trap_nr = X86_TRAP_GP;
@@ -549,12 +552,12 @@ do_general_protection(struct pt_regs *regs, long error_code)
 		 */
 		if (!preemptible() && kprobe_running() &&
 		    kprobe_fault_handler(regs, X86_TRAP_GP))
-			return;
+			goto exit_trap;
 
 		if (notify_die(DIE_GPF, desc, regs, error_code,
 			       X86_TRAP_GP, SIGSEGV) != NOTIFY_STOP)
 			die(desc, regs, error_code);
-		return;
+		goto exit_trap;
 	}
 
 	tsk->thread.error_code = error_code;
@@ -563,6 +566,8 @@ do_general_protection(struct pt_regs *regs, long error_code)
 	show_signal(tsk, SIGSEGV, "", desc, regs, error_code);
 
 	force_sig(SIGSEGV);
+exit_trap:
+	cond_local_irq_disable(regs);
 }
 NOKPROBE_SYMBOL(do_general_protection);
 
@@ -783,9 +788,7 @@ dotraplinkage void do_debug(struct pt_regs *regs, long error_code)
 	if (v8086_mode(regs)) {
 		handle_vm86_trap((struct kernel_vm86_regs *) regs, error_code,
 					X86_TRAP_DB);
-		cond_local_irq_disable(regs);
-		debug_stack_usage_dec();
-		goto exit;
+		goto exit_irq;
 	}
 
 	if (WARN_ON_ONCE((dr6 & DR_STEP) && !user_mode(regs))) {
@@ -802,6 +805,8 @@ dotraplinkage void do_debug(struct pt_regs *regs, long error_code)
 	si_code = get_si_code(tsk->thread.debugreg6);
 	if (tsk->thread.debugreg6 & (DR_STEP | DR_TRAP_BITS) || user_icebp)
 		send_sigtrap(regs, error_code, si_code);
+
+exit_irq:
 	cond_local_irq_disable(regs);
 	debug_stack_usage_dec();
 
@@ -827,7 +832,7 @@ static void math_error(struct pt_regs *regs, int error_code, int trapnr)
 
 	if (!user_mode(regs)) {
 		if (fixup_exception(regs, trapnr, error_code, 0))
-			return;
+			goto exit_trap;
 
 		task->thread.error_code = error_code;
 		task->thread.trap_nr = trapnr;
@@ -835,7 +840,7 @@ static void math_error(struct pt_regs *regs, int error_code, int trapnr)
 		if (notify_die(DIE_TRAP, str, regs, error_code,
 					trapnr, SIGFPE) != NOTIFY_STOP)
 			die(str, regs, error_code);
-		return;
+		goto exit_trap;
 	}
 
 	/*
@@ -849,10 +854,12 @@ static void math_error(struct pt_regs *regs, int error_code, int trapnr)
 	si_code = fpu__exception_code(fpu, trapnr);
 	/* Retry when we get spurious exceptions: */
 	if (!si_code)
-		return;
+		goto exit_trap;
 
 	force_sig_fault(SIGFPE, si_code,
 			(void __user *)uprobe_get_trap_addr(regs));
+exit_trap:
+	cond_local_irq_disable(regs);
 }
 
 dotraplinkage void do_coprocessor_error(struct pt_regs *regs, long error_code)
@@ -889,6 +896,8 @@ do_device_not_available(struct pt_regs *regs, long error_code)
 
 		info.regs = regs;
 		math_emulate(&info);
+
+		cond_local_irq_disable(regs);
 		return;
 	}
 #endif
diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c
index 9ceacd1156db..501cc36a3d6a 100644
--- a/arch/x86/mm/fault.c
+++ b/arch/x86/mm/fault.c
@@ -1500,10 +1500,13 @@ __do_page_fault(struct pt_regs *regs, unsigned long hw_error_code,
 		return;
 
 	/* Was the fault on kernel-controlled part of the address space? */
-	if (unlikely(fault_in_kernel_space(address)))
+	if (unlikely(fault_in_kernel_space(address))) {
 		do_kern_addr_fault(regs, hw_error_code, address);
-	else
+	} else {
 		do_user_addr_fault(regs, hw_error_code, address);
+		if (regs->flags & X86_EFLAGS_IF)
+			local_irq_disable();
+	}
 }
 NOKPROBE_SYMBOL(__do_page_fault);