sparc64: Make global reg dumping even more useful.

Record one more level of stack frame program counter.

Particularly when lockdep and all sorts of spinlock debugging is
enabled, figuring out the caller of spin_lock() is difficult when the
cpu is stuck on the lock.

Signed-off-by: David S. Miller <davem@davemloft.net>
diff --git a/arch/sparc/include/asm/ptrace_64.h b/arch/sparc/include/asm/ptrace_64.h
index ec6d45c..390d92a 100644
--- a/arch/sparc/include/asm/ptrace_64.h
+++ b/arch/sparc/include/asm/ptrace_64.h
@@ -134,9 +134,9 @@
 	unsigned long		tnpc;
 	unsigned long		o7;
 	unsigned long		i7;
+	unsigned long		rpc;
 	struct thread_info	*thread;
 	unsigned long		pad1;
-	unsigned long		pad2;
 };
 
 #define __ARCH_WANT_COMPAT_SYS_PTRACE
@@ -315,9 +315,9 @@
 #define GR_SNAP_TNPC	0x10
 #define GR_SNAP_O7	0x18
 #define GR_SNAP_I7	0x20
-#define GR_SNAP_THREAD	0x28
-#define GR_SNAP_PAD1	0x30
-#define GR_SNAP_PAD2	0x38
+#define GR_SNAP_RPC	0x28
+#define GR_SNAP_THREAD	0x30
+#define GR_SNAP_PAD1	0x38
 
 #endif  /*  __KERNEL__  */
 
diff --git a/arch/sparc64/kernel/process.c b/arch/sparc64/kernel/process.c
index 0f60547..fc8137a 100644
--- a/arch/sparc64/kernel/process.c
+++ b/arch/sparc64/kernel/process.c
@@ -304,6 +304,19 @@
 struct global_reg_snapshot global_reg_snapshot[NR_CPUS];
 static DEFINE_SPINLOCK(global_reg_snapshot_lock);
 
+static bool kstack_valid(struct thread_info *tp, struct reg_window *rw)
+{
+	unsigned long thread_base, fp;
+
+	thread_base = (unsigned long) tp;
+	fp = (unsigned long) rw;
+
+	if (fp < (thread_base + sizeof(struct thread_info)) ||
+	    fp >= (thread_base + THREAD_SIZE))
+		return false;
+	return true;
+}
+
 static void __global_reg_self(struct thread_info *tp, struct pt_regs *regs,
 			      int this_cpu)
 {
@@ -315,14 +328,22 @@
 	global_reg_snapshot[this_cpu].o7 = regs->u_regs[UREG_I7];
 
 	if (regs->tstate & TSTATE_PRIV) {
+		struct thread_info *tp = current_thread_info();
 		struct reg_window *rw;
 
 		rw = (struct reg_window *)
 			(regs->u_regs[UREG_FP] + STACK_BIAS);
-		global_reg_snapshot[this_cpu].i7 = rw->ins[7];
-	} else
+		if (kstack_valid(tp, rw)) {
+			global_reg_snapshot[this_cpu].i7 = rw->ins[7];
+			rw = (struct reg_window *)
+				(rw->ins[6] + STACK_BIAS);
+			if (kstack_valid(tp, rw))
+				global_reg_snapshot[this_cpu].rpc = rw->ins[7];
+		}
+	} else {
 		global_reg_snapshot[this_cpu].i7 = 0;
-
+		global_reg_snapshot[this_cpu].rpc = 0;
+	}
 	global_reg_snapshot[this_cpu].thread = tp;
 }
 
@@ -375,13 +396,14 @@
 		       ((tp && tp->task) ? tp->task->pid : -1));
 
 		if (gp->tstate & TSTATE_PRIV) {
-			printk("             TPC[%pS] O7[%pS] I7[%pS]\n",
+			printk("             TPC[%pS] O7[%pS] I7[%pS] RPC[%pS]\n",
 			       (void *) gp->tpc,
 			       (void *) gp->o7,
-			       (void *) gp->i7);
+			       (void *) gp->i7,
+			       (void *) gp->rpc);
 		} else {
-			printk("             TPC[%lx] O7[%lx] I7[%lx]\n",
-			       gp->tpc, gp->o7, gp->i7);
+			printk("             TPC[%lx] O7[%lx] I7[%lx] RPC[%lx]\n",
+			       gp->tpc, gp->o7, gp->i7, gp->rpc);
 		}
 	}
 
diff --git a/arch/sparc64/mm/ultra.S b/arch/sparc64/mm/ultra.S
index 4c8ca13..77ba885 100644
--- a/arch/sparc64/mm/ultra.S
+++ b/arch/sparc64/mm/ultra.S
@@ -531,6 +531,13 @@
 	stx		%g7, [%g1 + GR_SNAP_TNPC]
 	stx		%o7, [%g1 + GR_SNAP_O7]
 	stx		%i7, [%g1 + GR_SNAP_I7]
+	/* Don't try this at home kids... */
+	rdpr		%cwp, %g2
+	sub		%g2, 1, %g7
+	wrpr		%g7, %cwp
+	mov		%i7, %g7
+	wrpr		%g2, %cwp
+	stx		%g7, [%g1 + GR_SNAP_RPC]
 	sethi		%hi(trap_block), %g7
 	or		%g7, %lo(trap_block), %g7
 	sllx		%g2, TRAP_BLOCK_SZ_SHIFT, %g2