1 files changed, 38 insertions, 41 deletions

diff --git a/arch/s390/mm/fault.c b/arch/s390/mm/fault.c
index d95265b2719f..19f623f1f21c 100644
--- a/arch/s390/mm/fault.c
+++ b/arch/s390/mm/fault.c
@@ -50,6 +50,7 @@
 #define VM_FAULT_BADMAP		0x020000
 #define VM_FAULT_BADACCESS	0x040000
 #define VM_FAULT_SIGNAL		0x080000
+#define VM_FAULT_PFAULT		0x100000
 
 static unsigned long store_indication __read_mostly;
 
@@ -105,21 +106,24 @@ void bust_spinlocks(int yes)
  * Returns the address space associated with the fault.
  * Returns 0 for kernel space and 1 for user space.
  */
-static inline int user_space_fault(unsigned long trans_exc_code)
+static inline int user_space_fault(struct pt_regs *regs)
 {
+	unsigned long trans_exc_code;
+
 	/*
 	 * The lowest two bits of the translation exception
 	 * identification indicate which paging table was used.
 	 */
-	trans_exc_code &= 3;
-	if (trans_exc_code == 2)
-		/* Access via secondary space, set_fs setting decides */
+	trans_exc_code = regs->int_parm_long & 3;
+	if (trans_exc_code == 3) /* home space -> kernel */
+		return 0;
+	if (user_mode(regs))
+		return 1;
+	if (trans_exc_code == 2) /* secondary space -> set_fs */
 		return current->thread.mm_segment.ar4;
-	/*
-	 * Access via primary space or access register is from user space
-	 * and access via home space is from the kernel.
-	 */
-	return trans_exc_code != 3;
+	if (current->flags & PF_VCPU)
+		return 1;
+	return 0;
 }
 
 static inline void report_user_fault(struct pt_regs *regs, long signr)
@@ -171,7 +175,7 @@ static noinline void do_no_context(struct pt_regs *regs)
 	 * terminate things with extreme prejudice.
 	 */
 	address = regs->int_parm_long & __FAIL_ADDR_MASK;
-	if (!user_space_fault(regs->int_parm_long))
+	if (!user_space_fault(regs))
 		printk(KERN_ALERT "Unable to handle kernel pointer dereference"
 		       " at virtual kernel address %p\n", (void *)address);
 	else
@@ -227,6 +231,7 @@ static noinline void do_fault_error(struct pt_regs *regs, int fault)
 			return;
 		}
 	case VM_FAULT_BADCONTEXT:
+	case VM_FAULT_PFAULT:
 		do_no_context(regs);
 		break;
 	case VM_FAULT_SIGNAL:
@@ -264,6 +269,9 @@ static noinline void do_fault_error(struct pt_regs *regs, int fault)
  */
 static inline int do_exception(struct pt_regs *regs, int access)
 {
+#ifdef CONFIG_PGSTE
+	struct gmap *gmap;
+#endif
 	struct task_struct *tsk;
 	struct mm_struct *mm;
 	struct vm_area_struct *vma;
@@ -291,7 +299,7 @@ static inline int do_exception(struct pt_regs *regs, int access)
 	 * user context.
 	 */
 	fault = VM_FAULT_BADCONTEXT;
-	if (unlikely(!user_space_fault(trans_exc_code) || in_atomic() || !mm))
+	if (unlikely(!user_space_fault(regs) || in_atomic() || !mm))
 		goto out;
 
 	address = trans_exc_code & __FAIL_ADDR_MASK;
@@ -304,9 +312,10 @@ static inline int do_exception(struct pt_regs *regs, int access)
 	down_read(&mm->mmap_sem);
 
 #ifdef CONFIG_PGSTE
-	if ((current->flags & PF_VCPU) && S390_lowcore.gmap) {
-		address = __gmap_fault(address,
-				     (struct gmap *) S390_lowcore.gmap);
+	gmap = (struct gmap *)
+		((current->flags & PF_VCPU) ? S390_lowcore.gmap : 0);
+	if (gmap) {
+		address = __gmap_fault(address, gmap);
 		if (address == -EFAULT) {
 			fault = VM_FAULT_BADMAP;
 			goto out_up;
@@ -315,6 +324,8 @@ static inline int do_exception(struct pt_regs *regs, int access)
 			fault = VM_FAULT_OOM;
 			goto out_up;
 		}
+		if (gmap->pfault_enabled)
+			flags |= FAULT_FLAG_RETRY_NOWAIT;
 	}
 #endif
 
@@ -371,9 +382,19 @@ retry:
 				      regs, address);
 		}
 		if (fault & VM_FAULT_RETRY) {
+#ifdef CONFIG_PGSTE
+			if (gmap && (flags & FAULT_FLAG_RETRY_NOWAIT)) {
+				/* FAULT_FLAG_RETRY_NOWAIT has been set,
+				 * mmap_sem has not been released */
+				current->thread.gmap_pfault = 1;
+				fault = VM_FAULT_PFAULT;
+				goto out_up;
+			}
+#endif
 			/* Clear FAULT_FLAG_ALLOW_RETRY to avoid any risk
 			 * of starvation. */
-			flags &= ~FAULT_FLAG_ALLOW_RETRY;
+			flags &= ~(FAULT_FLAG_ALLOW_RETRY |
+				   FAULT_FLAG_RETRY_NOWAIT);
 			flags |= FAULT_FLAG_TRIED;
 			down_read(&mm->mmap_sem);
 			goto retry;
@@ -423,30 +444,6 @@ void __kprobes do_dat_exception(struct pt_regs *regs)
 		do_fault_error(regs, fault);
 }
 
-int __handle_fault(unsigned long uaddr, unsigned long pgm_int_code, int write)
-{
-	struct pt_regs regs;
-	int access, fault;
-
-	/* Emulate a uaccess fault from kernel mode. */
-	regs.psw.mask = PSW_KERNEL_BITS | PSW_MASK_DAT | PSW_MASK_MCHECK;
-	if (!irqs_disabled())
-		regs.psw.mask |= PSW_MASK_IO | PSW_MASK_EXT;
-	regs.psw.addr = (unsigned long) __builtin_return_address(0);
-	regs.psw.addr |= PSW_ADDR_AMODE;
-	regs.int_code = pgm_int_code;
-	regs.int_parm_long = (uaddr & PAGE_MASK) | 2;
-	access = write ? VM_WRITE : VM_READ;
-	fault = do_exception(&regs, access);
-	/*
-	 * Since the fault happened in kernel mode while performing a uaccess
-	 * all we need to do now is emulating a fixup in case "fault" is not
-	 * zero.
-	 * For the calling uaccess functions this results always in -EFAULT.
-	 */
-	return fault ? -EFAULT : 0;
-}
-
 #ifdef CONFIG_PFAULT 
 /*
  * 'pfault' pseudo page faults routines.
@@ -627,7 +624,7 @@ static int __init pfault_irq_init(void)
 {
 	int rc;
 
-	rc = register_external_interrupt(0x2603, pfault_interrupt);
+	rc = register_external_irq(EXT_IRQ_CP_SERVICE, pfault_interrupt);
 	if (rc)
 		goto out_extint;
 	rc = pfault_init() == 0 ? 0 : -EOPNOTSUPP;
@@ -638,7 +635,7 @@ static int __init pfault_irq_init(void)
 	return 0;
 
 out_pfault:
-	unregister_external_interrupt(0x2603, pfault_interrupt);
+	unregister_external_irq(EXT_IRQ_CP_SERVICE, pfault_interrupt);
 out_extint:
 	pfault_disable = 1;
 	return rc;