ARM: invalidate L1 before enabling coherency

We must invalidate the L1 cache before enabling coherency, otherwise
secondary CPUs can inject invalid cache lines into the coherent CPU
cluster, which could then be migrated to other CPUs.  This fixes a
recent regression with SoCFPGA randomly failing to boot.

Fixes: 02b4e2756e01 ("ARM: v7 setup function should invalidate L1 cache")
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

arch/arm/mm/proc-v7.S

@@ -274,7 +274,10 @@ __v7_ca15mp_setup:
 __v7_b15mp_setup:
 __v7_ca17mp_setup:
 	mov	r10, #0
-1:
+1:	adr	r12, __v7_setup_stack		@ the local stack
+	stmia	r12, {r0-r5, lr}		@ v7_invalidate_l1 touches r0-r6
+	bl      v7_invalidate_l1
+	ldmia	r12, {r0-r5, lr}
 #ifdef CONFIG_SMP
 	ALT_SMP(mrc	p15, 0, r0, c1, c0, 1)
 	ALT_UP(mov	r0, #(1 << 6))		@ fake it for UP
@@ -283,7 +286,7 @@ __v7_ca17mp_setup:
 	orreq	r0, r0, r10			@ Enable CPU-specific SMP bits
 	mcreq	p15, 0, r0, c1, c0, 1
 #endif
-	b	__v7_setup
+	b	__v7_setup_cont
 
 /*
  * Errata:
@@ -413,10 +416,11 @@ __v7_pj4b_setup:
 
 __v7_setup:
 	adr	r12, __v7_setup_stack		@ the local stack
-	stmia	r12, {r0-r5, r7, r9, r11, lr}
+	stmia	r12, {r0-r5, lr}		@ v7_invalidate_l1 touches r0-r6
 	bl      v7_invalidate_l1
-	ldmia	r12, {r0-r5, r7, r9, r11, lr}
+	ldmia	r12, {r0-r5, lr}
 
+__v7_setup_cont:
 	and	r0, r9, #0xff000000		@ ARM?
 	teq	r0, #0x41000000
 	bne	__errata_finish
@@ -480,7 +484,7 @@ ENDPROC(__v7_setup)
 
 	.align	2
 __v7_setup_stack:
-	.space	4 * 11				@ 11 registers
+	.space	4 * 7				@ 12 registers
 
 	__INITDATA