s390: fix race on TIF_MCCK_PENDING

There is a small race window in the __switch_to code in regard to
the transfer of the TIF_MCCK_PENDING bit from the previous to the
next task. The bit is transferred before the task struct pointer
and the thread-info pointer for the next task has been stored to
lowcore. If a machine check sets the TIF_MCCK_PENDING bit between
the transfer code and the store of current/thread_info the bit
is still set for the previous task. And if the previous task has
terminated it can get lost. The effect is that a pending CRW is
not retrieved until the next machine checks sets TIF_MCCK_PENDING.
To fix this reorder __switch_to to first store the task struct
and thread-info pointer and then do the transfer of the bit.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

arch/s390/kernel/entry.S

@@ -145,22 +145,23 @@ STACK_SIZE  = 1 << STACK_SHIFT
  *  gpr2 = prev
  */
 ENTRY(__switch_to)
+	stm	%r6,%r15,__SF_GPRS(%r15)	# store gprs of prev task
+	st	%r15,__THREAD_ksp(%r2)		# store kernel stack of prev
 	l	%r4,__THREAD_info(%r2)		# get thread_info of prev
 	l	%r5,__THREAD_info(%r3)		# get thread_info of next
+	lr	%r15,%r5
+	ahi	%r15,STACK_SIZE			# end of kernel stack of next
+	st	%r3,__LC_CURRENT		# store task struct of next
+	st	%r5,__LC_THREAD_INFO		# store thread info of next
+	st	%r15,__LC_KERNEL_STACK		# store end of kernel stack
+	lctl	%c4,%c4,__TASK_pid(%r3)		# load pid to control reg. 4
+	mvc	__LC_CURRENT_PID(4,%r0),__TASK_pid(%r3)	# store pid of next
+	l	%r15,__THREAD_ksp(%r3)		# load kernel stack of next
 	tm	__TI_flags+3(%r4),_TIF_MCCK_PENDING # machine check pending?
 	jz	0f
 	ni	__TI_flags+3(%r4),255-_TIF_MCCK_PENDING	# clear flag in prev
 	oi	__TI_flags+3(%r5),_TIF_MCCK_PENDING	# set it in next
-0:	stm	%r6,%r15,__SF_GPRS(%r15)	# store gprs of prev task
-	st	%r15,__THREAD_ksp(%r2)		# store kernel stack of prev
-	l	%r15,__THREAD_ksp(%r3)		# load kernel stack of next
-	lctl	%c4,%c4,__TASK_pid(%r3)		# load pid to control reg. 4
-	lm	%r6,%r15,__SF_GPRS(%r15)	# load gprs of next task
-	st	%r3,__LC_CURRENT		# store task struct of next
-	mvc	__LC_CURRENT_PID(4,%r0),__TASK_pid(%r3)	# store pid of next
-	st	%r5,__LC_THREAD_INFO		# store thread info of next
-	ahi	%r5,STACK_SIZE			# end of kernel stack of next
-	st	%r5,__LC_KERNEL_STACK		# store end of kernel stack
+0:	lm	%r6,%r15,__SF_GPRS(%r15)	# load gprs of next task
 	br	%r14
 
 __critical_start:

arch/s390/kernel/entry64.S

@@ -164,22 +164,23 @@ _TIF_EXIT_SIE = (_TIF_SIGPENDING | _TIF_NEED_RESCHED | _TIF_MCCK_PENDING)
  *  gpr2 = prev
  */
 ENTRY(__switch_to)
+	stmg	%r6,%r15,__SF_GPRS(%r15)	# store gprs of prev task
+	stg	%r15,__THREAD_ksp(%r2)		# store kernel stack of prev
 	lg	%r4,__THREAD_info(%r2)		# get thread_info of prev
 	lg	%r5,__THREAD_info(%r3)		# get thread_info of next
+	lgr	%r15,%r5
+	aghi	%r15,STACK_SIZE			# end of kernel stack of next
+	stg	%r3,__LC_CURRENT		# store task struct of next
+	stg	%r5,__LC_THREAD_INFO		# store thread info of next
+	stg	%r15,__LC_KERNEL_STACK		# store end of kernel stack
+	lctl	%c4,%c4,__TASK_pid(%r3)		# load pid to control reg. 4
+	mvc	__LC_CURRENT_PID+4(4,%r0),__TASK_pid(%r3) # store pid of next
+	lg	%r15,__THREAD_ksp(%r3)		# load kernel stack of next
 	tm	__TI_flags+7(%r4),_TIF_MCCK_PENDING # machine check pending?
 	jz	0f
 	ni	__TI_flags+7(%r4),255-_TIF_MCCK_PENDING	# clear flag in prev
 	oi	__TI_flags+7(%r5),_TIF_MCCK_PENDING	# set it in next
-0:	stmg	%r6,%r15,__SF_GPRS(%r15)	# store gprs of prev task
-	stg	%r15,__THREAD_ksp(%r2)		# store kernel stack of prev
-	lg	%r15,__THREAD_ksp(%r3)		# load kernel stack of next
-	lctl	%c4,%c4,__TASK_pid(%r3)		# load pid to control reg. 4
-	lmg	%r6,%r15,__SF_GPRS(%r15)	# load gprs of next task
-	stg	%r3,__LC_CURRENT		# store task struct of next
-	mvc	__LC_CURRENT_PID+4(4,%r0),__TASK_pid(%r3) # store pid of next
-	stg	%r5,__LC_THREAD_INFO		# store thread info of next
-	aghi	%r5,STACK_SIZE			# end of kernel stack of next
-	stg	%r5,__LC_KERNEL_STACK		# store end of kernel stack
+0:	lmg	%r6,%r15,__SF_GPRS(%r15)	# load gprs of next task
 	br	%r14
 
 __critical_start: