diff --git a/arch/powerpc/include/asm/mmu-book3e.h b/arch/powerpc/include/asm/mmu-book3e.h
index 89b785d16846..901dac6b6cb7 100644
--- a/arch/powerpc/include/asm/mmu-book3e.h
+++ b/arch/powerpc/include/asm/mmu-book3e.h
@@ -287,11 +287,14 @@ extern int mmu_linear_psize;
 extern int mmu_vmemmap_psize;
 
 struct tlb_core_data {
+	/*
+	 * Per-core spinlock for e6500 TLB handlers (no tlbsrx.)
+	 * Must be the first struct element.
+	 */
+	u8 lock;
+
 	/* For software way selection, as on Freescale TLB1 */
 	u8 esel_next, esel_max, esel_first;
-
-	/* Per-core spinlock for e6500 TLB handlers (no tlbsrx.) */
-	u8 lock;
 };
 
 #ifdef CONFIG_PPC64
diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c
index da9c42f53bb1..4933909cc5c0 100644
--- a/arch/powerpc/kernel/setup_64.c
+++ b/arch/powerpc/kernel/setup_64.c
@@ -102,6 +102,8 @@ static void setup_tlb_core_data(void)
 {
 	int cpu;
 
+	BUILD_BUG_ON(offsetof(struct tlb_core_data, lock) != 0);
+
 	for_each_possible_cpu(cpu) {
 		int first = cpu_first_thread_sibling(cpu);
 
diff --git a/arch/powerpc/mm/tlb_low_64e.S b/arch/powerpc/mm/tlb_low_64e.S
index 6bf50507a4b5..1e50249e900b 100644
--- a/arch/powerpc/mm/tlb_low_64e.S
+++ b/arch/powerpc/mm/tlb_low_64e.S
@@ -284,7 +284,7 @@ itlb_miss_fault_bolted:
  * r14 = page table base
  * r13 = PACA
  * r11 = tlb_per_core ptr
- * r10 = crap (free to use)
+ * r10 = cpu number
  */
 tlb_miss_common_e6500:
 	/*
@@ -293,15 +293,18 @@ tlb_miss_common_e6500:
 	 *
 	 * MAS6:IND should be already set based on MAS4
 	 */
-	addi	r10,r11,TCD_LOCK
-1:	lbarx	r15,0,r10
+1:	lbarx	r15,0,r11
+	lhz	r10,PACAPACAINDEX(r13)
 	cmpdi	r15,0
+	cmpdi	cr1,r15,1	/* set cr1.eq = 0 for non-recursive */
 	bne	2f
-	li	r15,1
-	stbcx.	r15,0,r10
+	stbcx.	r10,0,r11
 	bne	1b
+3:
 	.subsection 1
-2:	lbz	r15,0(r10)
+2:	cmpd	cr1,r15,r10	/* recursive lock due to mcheck/crit/etc? */
+	beq	cr1,3b		/* unlock will happen if cr1.eq = 0 */
+	lbz	r15,0(r11)
 	cmpdi	r15,0
 	bne	2b
 	b	1b
@@ -379,9 +382,11 @@ tlb_miss_common_e6500:
 
 tlb_miss_done_e6500:
 	.macro	tlb_unlock_e6500
+	beq	cr1,1f		/* no unlock if lock was recursively grabbed */
 	li	r15,0
 	isync
-	stb	r15,TCD_LOCK(r11)
+	stb	r15,0(r11)
+1:
 	.endm
 
 	tlb_unlock_e6500