powerpc/64s: avoid reloading (H)SRR registers if they are still valid

When an interrupt is taken, the SRR registers are set to return to where
it left off. Unless they are modified in the meantime, or the return
address or MSR are modified, there is no need to reload these registers
when returning from interrupt.

Introduce per-CPU flags that track the validity of SRR and HSRR
registers. These are cleared when returning from interrupt, when
using the registers for something else (e.g., OPAL calls), when
adjusting the return address or MSR of a context, and when context
switching (which changes the return address and MSR).

This improves the performance of interrupt returns.

Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
[mpe: Fold in fixup patch from Nick]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20210617155116.2167984-5-npiggin@gmail.com

23 files changed, 273 insertions, 134 deletions

diff --git a/arch/powerpc/kernel/asm-offsets.c b/arch/powerpc/kernel/asm-offsets.c
index 31881ef3641d..69a2885adfe9 100644
--- a/arch/powerpc/kernel/asm-offsets.c
+++ b/arch/powerpc/kernel/asm-offsets.c
@@ -190,6 +190,10 @@ int main(void)
 	OFFSET(PACATOC, paca_struct, kernel_toc);
 	OFFSET(PACAKBASE, paca_struct, kernelbase);
 	OFFSET(PACAKMSR, paca_struct, kernel_msr);
+#ifdef CONFIG_PPC_BOOK3S_64
+	OFFSET(PACAHSRR_VALID, paca_struct, hsrr_valid);
+	OFFSET(PACASRR_VALID, paca_struct, srr_valid);
+#endif
 	OFFSET(PACAIRQSOFTMASK, paca_struct, irq_soft_mask);
 	OFFSET(PACAIRQHAPPENED, paca_struct, irq_happened);
 	OFFSET(PACA_FTRACE_ENABLED, paca_struct, ftrace_enabled);
diff --git a/arch/powerpc/kernel/entry_64.S b/arch/powerpc/kernel/entry_64.S
index 03a45a88b4b8..9a1d5e5599d3 100644
--- a/arch/powerpc/kernel/entry_64.S
+++ b/arch/powerpc/kernel/entry_64.S
@@ -64,6 +64,30 @@ exception_marker:
 	.section	".text"
 	.align 7
 
+.macro DEBUG_SRR_VALID srr
+#ifdef CONFIG_PPC_RFI_SRR_DEBUG
+	.ifc \srr,srr
+	mfspr	r11,SPRN_SRR0
+	ld	r12,_NIP(r1)
+100:	tdne	r11,r12
+	EMIT_BUG_ENTRY 100b,__FILE__,__LINE__,(BUGFLAG_WARNING | BUGFLAG_ONCE)
+	mfspr	r11,SPRN_SRR1
+	ld	r12,_MSR(r1)
+100:	tdne	r11,r12
+	EMIT_BUG_ENTRY 100b,__FILE__,__LINE__,(BUGFLAG_WARNING | BUGFLAG_ONCE)
+	.else
+	mfspr	r11,SPRN_HSRR0
+	ld	r12,_NIP(r1)
+100:	tdne	r11,r12
+	EMIT_BUG_ENTRY 100b,__FILE__,__LINE__,(BUGFLAG_WARNING | BUGFLAG_ONCE)
+	mfspr	r11,SPRN_HSRR1
+	ld	r12,_MSR(r1)
+100:	tdne	r11,r12
+	EMIT_BUG_ENTRY 100b,__FILE__,__LINE__,(BUGFLAG_WARNING | BUGFLAG_ONCE)
+	.endif
+#endif
+.endm
+
 #ifdef CONFIG_PPC_BOOK3S
 .macro system_call_vectored name trapnr
 	.globl system_call_vectored_\name
@@ -286,6 +310,11 @@ END_BTB_FLUSH_SECTION
 	ld	r11,exception_marker@toc(r2)
 	std	r11,-16(r10)		/* "regshere" marker */
 
+#ifdef CONFIG_PPC_BOOK3S
+	li	r11,1
+	stb	r11,PACASRR_VALID(r13)
+#endif
+
 	/*
 	 * We always enter kernel from userspace with irq soft-mask enabled and
 	 * nothing pending. system_call_exception() will call
@@ -306,18 +335,27 @@ END_BTB_FLUSH_SECTION
 	bl	syscall_exit_prepare
 
 	ld	r2,_CCR(r1)
+	ld	r6,_LINK(r1)
+	mtlr	r6
+
+#ifdef CONFIG_PPC_BOOK3S
+	lbz	r4,PACASRR_VALID(r13)
+	cmpdi	r4,0
+	bne	1f
+	li	r4,0
+	stb	r4,PACASRR_VALID(r13)
+#endif
 	ld	r4,_NIP(r1)
 	ld	r5,_MSR(r1)
-	ld	r6,_LINK(r1)
+	mtspr	SPRN_SRR0,r4
+	mtspr	SPRN_SRR1,r5
+1:
+	DEBUG_SRR_VALID srr
 
 BEGIN_FTR_SECTION
 	stdcx.	r0,0,r1			/* to clear the reservation */
 END_FTR_SECTION_IFCLR(CPU_FTR_STCX_CHECKS_ADDRESS)
 
-	mtspr	SPRN_SRR0,r4
-	mtspr	SPRN_SRR1,r5
-	mtlr	r6
-
 	cmpdi	r3,0
 	bne	.Lsyscall_restore_regs
 	/* Zero volatile regs that may contain sensitive kernel data */
@@ -673,19 +711,40 @@ _ASM_NOKPROBE_SYMBOL(interrupt_return_\srr\())
 	kuap_user_restore r3, r4
 #endif
 .Lfast_user_interrupt_return_\srr\():
-	ld	r11,_NIP(r1)
-	ld	r12,_MSR(r1)
+
 BEGIN_FTR_SECTION
 	ld	r10,_PPR(r1)
 	mtspr	SPRN_PPR,r10
 END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
+
+#ifdef CONFIG_PPC_BOOK3S
+	.ifc \srr,srr
+	lbz	r4,PACASRR_VALID(r13)
+	.else
+	lbz	r4,PACAHSRR_VALID(r13)
+	.endif
+	cmpdi	r4,0
+	li	r4,0
+	bne	1f
+#endif
+	ld	r11,_NIP(r1)
+	ld	r12,_MSR(r1)
 	.ifc \srr,srr
 	mtspr	SPRN_SRR0,r11
 	mtspr	SPRN_SRR1,r12
+1:
+#ifdef CONFIG_PPC_BOOK3S
+	stb	r4,PACASRR_VALID(r13)
+#endif
 	.else
 	mtspr	SPRN_HSRR0,r11
 	mtspr	SPRN_HSRR1,r12
+1:
+#ifdef CONFIG_PPC_BOOK3S
+	stb	r4,PACAHSRR_VALID(r13)
+#endif
 	.endif
+	DEBUG_SRR_VALID \srr
 
 BEGIN_FTR_SECTION
 	stdcx.	r0,0,r1		/* to clear the reservation */
@@ -730,15 +789,34 @@ ALT_FTR_SECTION_END_IFCLR(CPU_FTR_STCX_CHECKS_ADDRESS)
 
 .Lfast_kernel_interrupt_return_\srr\():
 	cmpdi	cr1,r3,0
+#ifdef CONFIG_PPC_BOOK3S
+	.ifc \srr,srr
+	lbz	r4,PACASRR_VALID(r13)
+	.else
+	lbz	r4,PACAHSRR_VALID(r13)
+	.endif
+	cmpdi	r4,0
+	li	r4,0
+	bne	1f
+#endif
 	ld	r11,_NIP(r1)
 	ld	r12,_MSR(r1)
 	.ifc \srr,srr
 	mtspr	SPRN_SRR0,r11
 	mtspr	SPRN_SRR1,r12
+1:
+#ifdef CONFIG_PPC_BOOK3S
+	stb	r4,PACASRR_VALID(r13)
+#endif
 	.else
 	mtspr	SPRN_HSRR0,r11
 	mtspr	SPRN_HSRR1,r12
+1:
+#ifdef CONFIG_PPC_BOOK3S
+	stb	r4,PACAHSRR_VALID(r13)
+#endif
 	.endif
+	DEBUG_SRR_VALID \srr
 
 BEGIN_FTR_SECTION
 	stdcx.	r0,0,r1		/* to clear the reservation */
diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
index 1bc27af1b425..3d238a3b2a24 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
@@ -485,6 +485,20 @@ DEFINE_FIXED_SYMBOL(\name\()_common_real)
 	std	r0,GPR0(r1)		/* save r0 in stackframe	*/
 	std	r10,GPR1(r1)		/* save r1 in stackframe	*/
 
+	/* Mark our [H]SRRs valid for return */
+	li	r10,1
+	.if IHSRR_IF_HVMODE
+	BEGIN_FTR_SECTION
+	stb	r10,PACAHSRR_VALID(r13)
+	FTR_SECTION_ELSE
+	stb	r10,PACASRR_VALID(r13)
+	ALT_FTR_SECTION_END_IFSET(CPU_FTR_HVMODE | CPU_FTR_ARCH_206)
+	.elseif IHSRR
+	stb	r10,PACAHSRR_VALID(r13)
+	.else
+	stb	r10,PACASRR_VALID(r13)
+	.endif
+
 	.if ISET_RI
 	li	r10,MSR_RI
 	mtmsrd	r10,1			/* Set MSR_RI */
@@ -584,10 +598,13 @@ END_FTR_SECTION_IFSET(CPU_FTR_CFAR)
 .macro EXCEPTION_RESTORE_REGS hsrr=0
 	/* Move original SRR0 and SRR1 into the respective regs */
 	ld	r9,_MSR(r1)
+	li	r10,0
 	.if \hsrr
 	mtspr	SPRN_HSRR1,r9
+	stb	r10,PACAHSRR_VALID(r13)
 	.else
 	mtspr	SPRN_SRR1,r9
+	stb	r10,PACASRR_VALID(r13)
 	.endif
 	ld	r9,_NIP(r1)
 	.if \hsrr
@@ -1718,6 +1735,8 @@ EXC_COMMON_BEGIN(hdecrementer_common)
 	 *
 	 * Be careful to avoid touching the kernel stack.
 	 */
+	li	r10,0
+	stb	r10,PACAHSRR_VALID(r13)
 	ld	r10,PACA_EXGEN+EX_CTR(r13)
 	mtctr	r10
 	mtcrf	0x80,r9
@@ -2513,6 +2532,8 @@ BEGIN_FTR_SECTION
 	ld	r10,PACA_EXGEN+EX_CFAR(r13)
 	mtspr	SPRN_CFAR,r10
 END_FTR_SECTION_IFSET(CPU_FTR_CFAR)
+	li	r10,0
+	stb	r10,PACAHSRR_VALID(r13)
 	ld	r10,PACA_EXGEN+EX_R10(r13)
 	ld	r11,PACA_EXGEN+EX_R11(r13)
 	ld	r12,PACA_EXGEN+EX_R12(r13)
@@ -2673,6 +2694,12 @@ masked_interrupt:
 	ori	r11,r11,PACA_IRQ_HARD_DIS
 	stb	r11,PACAIRQHAPPENED(r13)
 2:	/* done */
+	li	r10,0
+	.if \hsrr
+	stb	r10,PACAHSRR_VALID(r13)
+	.else
+	stb	r10,PACASRR_VALID(r13)
+	.endif
 	ld	r10,PACA_EXGEN+EX_CTR(r13)
 	mtctr	r10
 	mtcrf	0x80,r9
diff --git a/arch/powerpc/kernel/fpu.S b/arch/powerpc/kernel/fpu.S
index 2c57ece6671c..6010adcee16e 100644
--- a/arch/powerpc/kernel/fpu.S
+++ b/arch/powerpc/kernel/fpu.S
@@ -103,6 +103,10 @@ END_FTR_SECTION_IFSET(CPU_FTR_VSX)
 	ori	r12,r12,MSR_FP
 	or	r12,r12,r4
 	std	r12,_MSR(r1)
+#ifdef CONFIG_PPC_BOOK3S_64
+	li	r4,0
+	stb	r4,PACASRR_VALID(r13)
+#endif
 #endif
 	li	r4,1
 	stb	r4,THREAD_LOAD_FP(r5)
diff --git a/arch/powerpc/kernel/hw_breakpoint.c b/arch/powerpc/kernel/hw_breakpoint.c
index 8fc7a14e4d71..21a638aff72f 100644
--- a/arch/powerpc/kernel/hw_breakpoint.c
+++ b/arch/powerpc/kernel/hw_breakpoint.c
@@ -486,7 +486,7 @@ void thread_change_pc(struct task_struct *tsk, struct pt_regs *regs)
 	return;
 
 reset:
-	regs->msr &= ~MSR_SE;
+	regs_set_return_msr(regs, regs->msr & ~MSR_SE);
 	for (i = 0; i < nr_wp_slots(); i++) {
 		info = counter_arch_bp(__this_cpu_read(bp_per_reg[i]));
 		__set_breakpoint(i, info);
@@ -537,7 +537,7 @@ static bool stepping_handler(struct pt_regs *regs, struct perf_event **bp,
 			current->thread.last_hit_ubp[i] = bp[i];
 			info[i] = NULL;
 		}
-		regs->msr |= MSR_SE;
+		regs_set_return_msr(regs, regs->msr | MSR_SE);
 		return false;
 	}
 
diff --git a/arch/powerpc/kernel/kgdb.c b/arch/powerpc/kernel/kgdb.c
index c48565762698..bdee7262c080 100644
--- a/arch/powerpc/kernel/kgdb.c
+++ b/arch/powerpc/kernel/kgdb.c
@@ -147,7 +147,7 @@ static int kgdb_handle_breakpoint(struct pt_regs *regs)
 		return 0;
 
 	if (*(u32 *)regs->nip == BREAK_INSTR)
-		regs->nip += BREAK_INSTR_SIZE;
+		regs_add_return_ip(regs, BREAK_INSTR_SIZE);
 
 	return 1;
 }
@@ -372,7 +372,7 @@ int dbg_set_reg(int regno, void *mem, struct pt_regs *regs)
 
 void kgdb_arch_set_pc(struct pt_regs *regs, unsigned long pc)
 {
-	regs->nip = pc;
+	regs_set_return_ip(regs, pc);
 }
 
 /*
@@ -394,7 +394,7 @@ int kgdb_arch_handle_exception(int vector, int signo, int err_code,
 	case 'c':
 		/* handle the optional parameter */
 		if (kgdb_hex2long(&ptr, &addr))
-			linux_regs->nip = addr;
+			regs_set_return_ip(linux_regs, addr);
 
 		atomic_set(&kgdb_cpu_doing_single_step, -1);
 		/* set the trace bit if we're stepping */
@@ -402,9 +402,9 @@ int kgdb_arch_handle_exception(int vector, int signo, int err_code,
 #ifdef CONFIG_PPC_ADV_DEBUG_REGS
 			mtspr(SPRN_DBCR0,
 			      mfspr(SPRN_DBCR0) | DBCR0_IC | DBCR0_IDM);
-			linux_regs->msr |= MSR_DE;
+			regs_set_return_msr(linux_regs, linux_regs->msr | MSR_DE);
 #else
-			linux_regs->msr |= MSR_SE;
+			regs_set_return_msr(linux_regs, linux_regs->msr | MSR_SE);
 #endif
 			atomic_set(&kgdb_cpu_doing_single_step,
 				   raw_smp_processor_id());
diff --git a/arch/powerpc/kernel/kprobes-ftrace.c b/arch/powerpc/kernel/kprobes-ftrace.c
index 660138f6c4b2..7154d58338cc 100644
--- a/arch/powerpc/kernel/kprobes-ftrace.c
+++ b/arch/powerpc/kernel/kprobes-ftrace.c
@@ -39,7 +39,7 @@ void kprobe_ftrace_handler(unsigned long nip, unsigned long parent_nip,
 		 * On powerpc, NIP is *before* this instruction for the
 		 * pre handler
 		 */
-		regs->nip -= MCOUNT_INSN_SIZE;
+		regs_add_return_ip(regs, -MCOUNT_INSN_SIZE);
 
 		__this_cpu_write(current_kprobe, p);
 		kcb->kprobe_status = KPROBE_HIT_ACTIVE;
@@ -48,7 +48,7 @@ void kprobe_ftrace_handler(unsigned long nip, unsigned long parent_nip,
 			 * Emulate singlestep (and also recover regs->nip)
 			 * as if there is a nop
 			 */
-			regs->nip += MCOUNT_INSN_SIZE;
+			regs_add_return_ip(regs, MCOUNT_INSN_SIZE);
 			if (unlikely(p->post_handler)) {
 				kcb->kprobe_status = KPROBE_HIT_SSDONE;
 				p->post_handler(p, regs, 0);
diff --git a/arch/powerpc/kernel/kprobes.c b/arch/powerpc/kernel/kprobes.c
index 8ac248cb518a..04714895d3f9 100644
--- a/arch/powerpc/kernel/kprobes.c
+++ b/arch/powerpc/kernel/kprobes.c
@@ -194,7 +194,7 @@ static nokprobe_inline void prepare_singlestep(struct kprobe *p, struct pt_regs
 	 * variant as values in regs could play a part in
 	 * if the trap is taken or not
 	 */
-	regs->nip = (unsigned long)p->ainsn.insn;
+	regs_set_return_ip(regs, (unsigned long)p->ainsn.insn);
 }
 
 static nokprobe_inline void save_previous_kprobe(struct kprobe_ctlblk *kcb)
@@ -335,8 +335,9 @@ int kprobe_handler(struct pt_regs *regs)
 		kprobe_opcode_t insn = *p->ainsn.insn;
 		if (kcb->kprobe_status == KPROBE_HIT_SS && is_trap(insn)) {
 			/* Turn off 'trace' bits */
-			regs->msr &= ~MSR_SINGLESTEP;
-			regs->msr |= kcb->kprobe_saved_msr;
+			regs_set_return_msr(regs,
+				(regs->msr & ~MSR_SINGLESTEP) |
+				kcb->kprobe_saved_msr);
 			goto no_kprobe;
 		}
 
@@ -431,7 +432,7 @@ static int trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
 	 * we end up emulating it in kprobe_handler(), which increments the nip
 	 * again.
 	 */
-	regs->nip = orig_ret_address - 4;
+	regs_set_return_ip(regs, orig_ret_address - 4);
 	regs->link = orig_ret_address;
 
 	return 0;
@@ -466,8 +467,8 @@ int kprobe_post_handler(struct pt_regs *regs)
 	}
 
 	/* Adjust nip to after the single-stepped instruction */
-	regs->nip = (unsigned long)cur->addr + len;
-	regs->msr |= kcb->kprobe_saved_msr;
+	regs_set_return_ip(regs, (unsigned long)cur->addr + len);
+	regs_set_return_msr(regs, regs->msr | kcb->kprobe_saved_msr);
 
 	/*Restore back the original saved kprobes variables and continue. */
 	if (kcb->kprobe_status == KPROBE_REENTER) {
@@ -506,9 +507,11 @@ int kprobe_fault_handler(struct pt_regs *regs, int trapnr)
 		 * and allow the page fault handler to continue as a
 		 * normal page fault.
 		 */
-		regs->nip = (unsigned long)cur->addr;
-		regs->msr &= ~MSR_SINGLESTEP; /* Turn off 'trace' bits */
-		regs->msr |= kcb->kprobe_saved_msr;
+		regs_set_return_ip(regs, (unsigned long)cur->addr);
+		/* Turn off 'trace' bits */
+		regs_set_return_msr(regs,
+			(regs->msr & ~MSR_SINGLESTEP) |
+			kcb->kprobe_saved_msr);
 		if (kcb->kprobe_status == KPROBE_REENTER)
 			restore_previous_kprobe(kcb);
 		else
@@ -539,7 +542,7 @@ int kprobe_fault_handler(struct pt_regs *regs, int trapnr)
 		 * zero, try to fix up.
 		 */
 		if ((entry = search_exception_tables(regs->nip)) != NULL) {
-			regs->nip = extable_fixup(entry);
+			regs_set_return_ip(regs, extable_fixup(entry));
 			return 1;
 		}
 
diff --git a/arch/powerpc/kernel/mce.c b/arch/powerpc/kernel/mce.c
index 9a3c2a84a2ac..f4b922a1e6ad 100644
--- a/arch/powerpc/kernel/mce.c
+++ b/arch/powerpc/kernel/mce.c
@@ -273,7 +273,7 @@ void mce_common_process_ue(struct pt_regs *regs,
 	entry = search_kernel_exception_table(regs->nip);
 	if (entry) {
 		mce_err->ignore_event = true;
-		regs->nip = extable_fixup(entry);
+		regs_set_return_ip(regs, extable_fixup(entry));
 	}
 }
 
diff --git a/arch/powerpc/kernel/optprobes.c b/arch/powerpc/kernel/optprobes.c
index 2b8fe40069ad..8b9f82dc6ece 100644
--- a/arch/powerpc/kernel/optprobes.c
+++ b/arch/powerpc/kernel/optprobes.c
@@ -106,7 +106,7 @@ static void optimized_callback(struct optimized_kprobe *op,
 		kprobes_inc_nmissed_count(&op->kp);
 	} else {
 		__this_cpu_write(current_kprobe, &op->kp);
-		regs->nip = (unsigned long)op->kp.addr;
+		regs_set_return_ip(regs, (unsigned long)op->kp.addr);
 		get_kprobe_ctlblk()->kprobe_status = KPROBE_HIT_ACTIVE;
 		opt_pre_handler(&op->kp, regs);
 		__this_cpu_write(current_kprobe, NULL);
diff --git a/arch/powerpc/kernel/process.c b/arch/powerpc/kernel/process.c
index 4e593fc2c66d..8ab1bfdfd31a 100644
--- a/arch/powerpc/kernel/process.c
+++ b/arch/powerpc/kernel/process.c
@@ -96,7 +96,8 @@ static void check_if_tm_restore_required(struct task_struct *tsk)
 	if (tsk == current && tsk->thread.regs &&
 	    MSR_TM_ACTIVE(tsk->thread.regs->msr) &&
 	    !test_thread_flag(TIF_RESTORE_TM)) {
-		tsk->thread.ckpt_regs.msr = tsk->thread.regs->msr;
+		regs_set_return_msr(&tsk->thread.ckpt_regs,
+						tsk->thread.regs->msr);
 		set_thread_flag(TIF_RESTORE_TM);
 	}
 }
@@ -161,7 +162,7 @@ static void __giveup_fpu(struct task_struct *tsk)
 	msr &= ~(MSR_FP|MSR_FE0|MSR_FE1);
 	if (cpu_has_feature(CPU_FTR_VSX))
 		msr &= ~MSR_VSX;
-	tsk->thread.regs->msr = msr;
+	regs_set_return_msr(tsk->thread.regs, msr);
 }
 
 void giveup_fpu(struct task_struct *tsk)
@@ -244,7 +245,7 @@ static void __giveup_altivec(struct task_struct *tsk)
 	msr &= ~MSR_VEC;
 	if (cpu_has_feature(CPU_FTR_VSX))
 		msr &= ~MSR_VSX;
-	tsk->thread.regs->msr = msr;
+	regs_set_return_msr(tsk->thread.regs, msr);
 }
 
 void giveup_altivec(struct task_struct *tsk)
@@ -559,7 +560,7 @@ void notrace restore_math(struct pt_regs *regs)
 
 		msr_check_and_clear(new_msr);
 
-		regs->msr |= new_msr | fpexc_mode;
+		regs_set_return_msr(regs, regs->msr | new_msr | fpexc_mode);
 	}
 }
 #endif /* CONFIG_PPC_BOOK3S_64 */
@@ -1114,7 +1115,7 @@ void restore_tm_state(struct pt_regs *regs)
 #endif
 	restore_math(regs);
 
-	regs->msr |= msr_diff;
+	regs_set_return_msr(regs, regs->msr | msr_diff);
 }
 
 #else /* !CONFIG_PPC_TRANSACTIONAL_MEM */
@@ -1257,14 +1258,16 @@ struct task_struct *__switch_to(struct task_struct *prev,
 	}
 
 	/*
-	 * Call restore_sprs() before calling _switch(). If we move it after
-	 * _switch() then we miss out on calling it for new tasks. The reason
-	 * for this is we manually create a stack frame for new tasks that
-	 * directly returns through ret_from_fork() or
+	 * Call restore_sprs() and set_return_regs_changed() before calling
+	 * _switch(). If we move it after _switch() then we miss out on calling
+	 * it for new tasks. The reason for this is we manually create a stack
+	 * frame for new tasks that directly returns through ret_from_fork() or
 	 * ret_from_kernel_thread(). See copy_thread() for details.
 	 */
 	restore_sprs(old_thread, new_thread);
 
+	set_return_regs_changed(); /* _switch changes stack (and regs) */
+
 #ifdef CONFIG_PPC32
 	kuap_assert_locked();
 #endif
@@ -1850,13 +1853,14 @@ void start_thread(struct pt_regs *regs, unsigned long start, unsigned long sp)
 			}
 			regs->gpr[2] = toc;
 		}
-		regs->nip = entry;
-		regs->msr = MSR_USER64;
+		regs_set_return_ip(regs, entry);
+		regs_set_return_msr(regs, MSR_USER64);
 	} else {
-		regs->nip = start;
 		regs->gpr[2] = 0;
-		regs->msr = MSR_USER32;
+		regs_set_return_ip(regs, start);
+		regs_set_return_msr(regs, MSR_USER32);
 	}
+
 #endif
 #ifdef CONFIG_VSX
 	current->thread.used_vsr = 0;
@@ -1887,7 +1891,6 @@ void start_thread(struct pt_regs *regs, unsigned long start, unsigned long sp)
 	current->thread.tm_tfiar = 0;
 	current->thread.load_tm = 0;
 #endif /* CONFIG_PPC_TRANSACTIONAL_MEM */
-
 }
 EXPORT_SYMBOL(start_thread);
 
@@ -1935,9 +1938,10 @@ int set_fpexc_mode(struct task_struct *tsk, unsigned int val)
 	if (val > PR_FP_EXC_PRECISE)
 		return -EINVAL;
 	tsk->thread.fpexc_mode = __pack_fe01(val);
-	if (regs != NULL && (regs->msr & MSR_FP) != 0)
-		regs->msr = (regs->msr & ~(MSR_FE0|MSR_FE1))
-			| tsk->thread.fpexc_mode;
+	if (regs != NULL && (regs->msr & MSR_FP) != 0) {
+		regs_set_return_msr(regs, (regs->msr & ~(MSR_FE0|MSR_FE1))
+						| tsk->thread.fpexc_mode);
+	}
 	return 0;
 }
 
@@ -1983,9 +1987,9 @@ int set_endian(struct task_struct *tsk, unsigned int val)
 		return -EINVAL;
 
 	if (val == PR_ENDIAN_BIG)
-		regs->msr &= ~MSR_LE;
+		regs_set_return_msr(regs, regs->msr & ~MSR_LE);
 	else if (val == PR_ENDIAN_LITTLE || val == PR_ENDIAN_PPC_LITTLE)
-		regs->msr |= MSR_LE;
+		regs_set_return_msr(regs, regs->msr | MSR_LE);
 	else
 		return -EINVAL;
 
diff --git a/arch/powerpc/kernel/prom_init.c b/arch/powerpc/kernel/prom_init.c
index cb5be081e499..05ce15b854e2 100644
--- a/arch/powerpc/kernel/prom_init.c
+++ b/arch/powerpc/kernel/prom_init.c
@@ -32,6 +32,7 @@
 #include <asm/rtas.h>
 #include <asm/page.h>
 #include <asm/processor.h>
+#include <asm/interrupt.h>
 #include <asm/irq.h>
 #include <asm/io.h>
 #include <asm/smp.h>
@@ -1792,6 +1793,8 @@ static int prom_rtas_hcall(uint64_t args)
 	asm volatile("sc 1\n" : "=r" (arg1) :
 			"r" (arg1),
 			"r" (arg2) :);
+	srr_regs_clobbered();
+
 	return arg1;
 }
 
diff --git a/arch/powerpc/kernel/ptrace/ptrace-adv.c b/arch/powerpc/kernel/ptrace/ptrace-adv.c
index 3990c01ef8cf..399f5d94a3df 100644
--- a/arch/powerpc/kernel/ptrace/ptrace-adv.c
+++ b/arch/powerpc/kernel/ptrace/ptrace-adv.c
@@ -12,7 +12,7 @@ void user_enable_single_step(struct task_struct *task)
 	if (regs != NULL) {
 		task->thread.debug.dbcr0 &= ~DBCR0_BT;
 		task->thread.debug.dbcr0 |= DBCR0_IDM | DBCR0_IC;
-		regs->msr |= MSR_DE;
+		regs_set_return_msr(regs, regs->msr | MSR_DE);
 	}
 	set_tsk_thread_flag(task, TIF_SINGLESTEP);
 }
@@ -24,7 +24,7 @@ void user_enable_block_step(struct task_struct *task)
 	if (regs != NULL) {
 		task->thread.debug.dbcr0 &= ~DBCR0_IC;
 		task->thread.debug.dbcr0 = DBCR0_IDM | DBCR0_BT;
-		regs->msr |= MSR_DE;
+		regs_set_return_msr(regs, regs->msr | MSR_DE);
 	}
 	set_tsk_thread_flag(task, TIF_SINGLESTEP);
 }
@@ -50,7 +50,7 @@ void user_disable_single_step(struct task_struct *task)
 			 * All debug events were off.....
 			 */
 			task->thread.debug.dbcr0 &= ~DBCR0_IDM;
-			regs->msr &= ~MSR_DE;
+			regs_set_return_msr(regs, regs->msr & ~MSR_DE);
 		}
 	}
 	clear_tsk_thread_flag(task, TIF_SINGLESTEP);
@@ -82,6 +82,7 @@ int ptrace_get_debugreg(struct task_struct *child, unsigned long addr,
 
 int ptrace_set_debugreg(struct task_struct *task, unsigned long addr, unsigned long data)
 {
+	struct pt_regs *regs = task->thread.regs;
 #ifdef CONFIG_HAVE_HW_BREAKPOINT
 	int ret;
 	struct thread_struct *thread = &task->thread;
@@ -112,7 +113,7 @@ int ptrace_set_debugreg(struct task_struct *task, unsigned long addr, unsigned l
 		dbcr_dac(task) &= ~(DBCR_DAC1R | DBCR_DAC1W);
 		if (!DBCR_ACTIVE_EVENTS(task->thread.debug.dbcr0,
 					task->thread.debug.dbcr1)) {
-			task->thread.regs->msr &= ~MSR_DE;
+			regs_set_return_msr(regs, regs->msr & ~MSR_DE);
 			task->thread.debug.dbcr0 &= ~DBCR0_IDM;
 		}
 		return 0;
@@ -132,7 +133,7 @@ int ptrace_set_debugreg(struct task_struct *task, unsigned long addr, unsigned l
 		dbcr_dac(task) |= DBCR_DAC1R;
 	if (data & 0x2UL)
 		dbcr_dac(task) |= DBCR_DAC1W;
-	task->thread.regs->msr |= MSR_DE;
+	regs_set_return_msr(regs, regs->msr | MSR_DE);
 	return 0;
 }
 
@@ -220,7 +221,7 @@ static long set_instruction_bp(struct task_struct *child,
 	}
 out:
 	child->thread.debug.dbcr0 |= DBCR0_IDM;
-	child->thread.regs->msr |= MSR_DE;
+	regs_set_return_msr(child->thread.regs, child->thread.regs->msr | MSR_DE);
 
 	return slot;
 }
@@ -336,7 +337,7 @@ static int set_dac(struct task_struct *child, struct ppc_hw_breakpoint *bp_info)
 		return -ENOSPC;
 	}
 	child->thread.debug.dbcr0 |= DBCR0_IDM;
-	child->thread.regs->msr |= MSR_DE;
+	regs_set_return_msr(child->thread.regs, child->thread.regs->msr | MSR_DE);
 
 	return slot + 4;
 }
@@ -430,7 +431,7 @@ static int set_dac_range(struct task_struct *child,
 		child->thread.debug.dbcr2  |= DBCR2_DAC12MX;
 	else	/* PPC_BREAKPOINT_MODE_MASK */
 		child->thread.debug.dbcr2  |= DBCR2_DAC12MM;
-	child->thread.regs->msr |= MSR_DE;
+	regs_set_return_msr(child->thread.regs, child->thread.regs->msr | MSR_DE);
 
 	return 5;
 }
@@ -485,7 +486,8 @@ long ppc_del_hwdebug(struct task_struct *child, long data)
 		if (!DBCR_ACTIVE_EVENTS(child->thread.debug.dbcr0,
 					child->thread.debug.dbcr1)) {
 			child->thread.debug.dbcr0 &= ~DBCR0_IDM;
-			child->thread.regs->msr &= ~MSR_DE;
+			regs_set_return_msr(child->thread.regs,
+					child->thread.regs->msr & ~MSR_DE);
 		}
 	}
 	return rc;
diff --git a/arch/powerpc/kernel/ptrace/ptrace-noadv.c b/arch/powerpc/kernel/ptrace/ptrace-noadv.c
index aa36fcad36cd..a5dd7d2e2c9e 100644
--- a/arch/powerpc/kernel/ptrace/ptrace-noadv.c
+++ b/arch/powerpc/kernel/ptrace/ptrace-noadv.c
@@ -11,10 +11,8 @@ void user_enable_single_step(struct task_struct *task)
 {
 	struct pt_regs *regs = task->thread.regs;
 
-	if (regs != NULL) {
-		regs->msr &= ~MSR_BE;
-		regs->msr |= MSR_SE;
-	}
+	if (regs != NULL)
+		regs_set_return_msr(regs, (regs->msr & ~MSR_BE) | MSR_SE);
 	set_tsk_thread_flag(task, TIF_SINGLESTEP);
 }
 
@@ -22,10 +20,8 @@ void user_enable_block_step(struct task_struct *task)
 {
 	struct pt_regs *regs = task->thread.regs;
 
-	if (regs != NULL) {
-		regs->msr &= ~MSR_SE;
-		regs->msr |= MSR_BE;
-	}
+	if (regs != NULL)
+		regs_set_return_msr(regs, (regs->msr & ~MSR_SE) | MSR_BE);
 	set_tsk_thread_flag(task, TIF_SINGLESTEP);
 }
 
@@ -34,7 +30,7 @@ void user_disable_single_step(struct task_struct *task)
 	struct pt_regs *regs = task->thread.regs;
 
 	if (regs != NULL)
-		regs->msr &= ~(MSR_SE | MSR_BE);
+		regs_set_return_msr(regs, regs->msr & ~(MSR_SE | MSR_BE));
 
 	clear_tsk_thread_flag(task, TIF_SINGLESTEP);
 }
diff --git a/arch/powerpc/kernel/ptrace/ptrace-view.c b/arch/powerpc/kernel/ptrace/ptrace-view.c
index 773bcc4ca843..b8be1d6668b5 100644
--- a/arch/powerpc/kernel/ptrace/ptrace-view.c
+++ b/arch/powerpc/kernel/ptrace/ptrace-view.c
@@ -113,8 +113,9 @@ static unsigned long get_user_msr(struct task_struct *task)
 
 static __always_inline int set_user_msr(struct task_struct *task, unsigned long msr)
 {
-	task->thread.regs->msr &= ~MSR_DEBUGCHANGE;
-	task->thread.regs->msr |= msr & MSR_DEBUGCHANGE;
+	unsigned long newmsr = (task->thread.regs->msr & ~MSR_DEBUGCHANGE) |
+				(msr & MSR_DEBUGCHANGE);
+	regs_set_return_msr(task->thread.regs, newmsr);
 	return 0;
 }
 
diff --git a/arch/powerpc/kernel/rtas.c b/arch/powerpc/kernel/rtas.c
index 6bada744402b..99f2cce635fb 100644
--- a/arch/powerpc/kernel/rtas.c
+++ b/arch/powerpc/kernel/rtas.c
@@ -25,6 +25,7 @@
 #include <linux/reboot.h>
 #include <linux/syscalls.h>
 
+#include <asm/interrupt.h>
 #include <asm/prom.h>
 #include <asm/rtas.h>
 #include <asm/hvcall.h>
@@ -46,6 +47,13 @@
 /* This is here deliberately so it's only used in this file */
 void enter_rtas(unsigned long);
 
+static inline void do_enter_rtas(unsigned long args)
+{
+	enter_rtas(args);
+
+	srr_regs_clobbered(); /* rtas uses SRRs, invalidate */
+}
+
 struct rtas_t rtas = {
 	.lock = __ARCH_SPIN_LOCK_UNLOCKED
 };
@@ -384,7 +392,7 @@ static char *__fetch_rtas_last_error(char *altbuf)
 	save_args = rtas.args;
 	rtas.args = err_args;
 
-	enter_rtas(__pa(&rtas.args));
+	do_enter_rtas(__pa(&rtas.args));
 
 	err_args = rtas.args;
 	rtas.args = save_args;
@@ -430,7 +438,7 @@ va_rtas_call_unlocked(struct rtas_args *args, int token, int nargs, int nret,
 	for (i = 0; i < nret; ++i)
 		args->rets[i] = 0;
 
-	enter_rtas(__pa(args));
+	do_enter_rtas(__pa(args));
 }
 
 void rtas_call_unlocked(struct rtas_args *args, int token, int nargs, int nret, ...)
@@ -1138,7 +1146,7 @@ SYSCALL_DEFINE1(rtas, struct rtas_args __user *, uargs)
 	flags = lock_rtas();
 
 	rtas.args = args;
-	enter_rtas(__pa(&rtas.args));
+	do_enter_rtas(__pa(&rtas.args));
 	args = rtas.args;
 
 	/* A -1 return code indicates that the last command couldn't
diff --git a/arch/powerpc/kernel/signal.c b/arch/powerpc/kernel/signal.c
index 9ded046edb0e..e600764a926c 100644
--- a/arch/powerpc/kernel/signal.c
+++ b/arch/powerpc/kernel/signal.c
@@ -214,7 +214,7 @@ static void check_syscall_restart(struct pt_regs *regs, struct k_sigaction *ka,
 			regs->gpr[0] = __NR_restart_syscall;
 		else
 			regs->gpr[3] = regs->orig_gpr3;
-		regs->nip -= 4;
+		regs_add_return_ip(regs, -4);
 		regs->result = 0;
 	} else {
 		if (trap_is_scv(regs)) {
@@ -322,16 +322,16 @@ static unsigned long get_tm_stackpointer(struct task_struct *tsk)
 	 * For signals taken in non-TM or suspended mode, we use the
 	 * normal/non-checkpointed stack pointer.
 	 */
-
-	unsigned long ret = tsk->thread.regs->gpr[1];
+	struct pt_regs *regs = tsk->thread.regs;
+	unsigned long ret = regs->gpr[1];
 
 #ifdef CONFIG_PPC_TRANSACTIONAL_MEM
 	BUG_ON(tsk != current);
 
-	if (MSR_TM_ACTIVE(tsk->thread.regs->msr)) {
+	if (MSR_TM_ACTIVE(regs->msr)) {
 		preempt_disable();
 		tm_reclaim_current(TM_CAUSE_SIGNAL);
-		if (MSR_TM_TRANSACTIONAL(tsk->thread.regs->msr))
+		if (MSR_TM_TRANSACTIONAL(regs->msr))
 			ret = tsk->thread.ckpt_regs.gpr[1];
 
 		/*
@@ -341,7 +341,7 @@ static unsigned long get_tm_stackpointer(struct task_struct *tsk)
 		 * (tm_recheckpoint_new_task() would recheckpoint). Besides, we
 		 * enter the signal handler in non-transactional state.
 		 */
-		tsk->thread.regs->msr &= ~MSR_TS_MASK;
+		regs_set_return_msr(regs, regs->msr & ~MSR_TS_MASK);
 		preempt_enable();
 	}
 #endif
diff --git a/arch/powerpc/kernel/signal_32.c b/arch/powerpc/kernel/signal_32.c
index f55ac65c7492..0608581967f0 100644
--- a/arch/powerpc/kernel/signal_32.c
+++ b/arch/powerpc/kernel/signal_32.c
@@ -479,14 +479,14 @@ static long restore_user_regs(struct pt_regs *regs,
 
 	/* if doing signal return, restore the previous little-endian mode */
 	if (sig)
-		regs->msr = (regs->msr & ~MSR_LE) | (msr & MSR_LE);
+		regs_set_return_msr(regs, (regs->msr & ~MSR_LE) | (msr & MSR_LE));
 
 #ifdef CONFIG_ALTIVEC
 	/*
 	 * Force the process to reload the altivec registers from
 	 * current->thread when it next does altivec instructions
 	 */
-	regs->msr &= ~MSR_VEC;
+	regs_set_return_msr(regs, regs->msr & ~MSR_VEC);
 	if (msr & MSR_VEC) {
 		/* restore altivec registers from the stack */
 		unsafe_copy_from_user(&current->thread.vr_state, &sr->mc_vregs,
@@ -508,7 +508,7 @@ static long restore_user_regs(struct pt_regs *regs,
 	 * Force the process to reload the VSX registers from
 	 * current->thread when it next does VSX instruction.
 	 */
-	regs->msr &= ~MSR_VSX;
+	regs_set_return_msr(regs, regs->msr & ~MSR_VSX);
 	if (msr & MSR_VSX) {
 		/*
 		 * Restore altivec registers from the stack to a local
@@ -524,12 +524,12 @@ static long restore_user_regs(struct pt_regs *regs,
 	 * force the process to reload the FP registers from
 	 * current->thread when it next does FP instructions
 	 */
-	regs->msr &= ~(MSR_FP | MSR_FE0 | MSR_FE1);
+	regs_set_return_msr(regs, regs->msr & ~(MSR_FP | MSR_FE0 | MSR_FE1));
 
 #ifdef CONFIG_SPE
 	/* force the process to reload the spe registers from
 	   current->thread when it next does spe instructions */
-	regs->msr &= ~MSR_SPE;
+	regs_set_return_msr(regs, regs->msr & ~MSR_SPE);
 	if (msr & MSR_SPE) {
 		/* restore spe registers from the stack */
 		unsafe_copy_from_user(current->thread.evr, &sr->mc_vregs,
@@ -580,9 +580,9 @@ static long restore_tm_user_regs(struct pt_regs *regs,
 	unsafe_get_user(msr, &sr->mc_gregs[PT_MSR], failed);
 
 	/* Restore the previous little-endian mode */
-	regs->msr = (regs->msr & ~MSR_LE) | (msr & MSR_LE);
+	regs_set_return_msr(regs, (regs->msr & ~MSR_LE) | (msr & MSR_LE));
 
-	regs->msr &= ~MSR_VEC;
+	regs_set_return_msr(regs, regs->msr & ~MSR_VEC);
 	if (msr & MSR_VEC) {
 		/* restore altivec registers from the stack */
 		unsafe_copy_from_user(&current->thread.ckvr_state, &sr->mc_vregs,
@@ -601,11 +601,11 @@ static long restore_tm_user_regs(struct pt_regs *regs,
 	if (cpu_has_feature(CPU_FTR_ALTIVEC))
 		mtspr(SPRN_VRSAVE, current->thread.ckvrsave);
 
-	regs->msr &= ~(MSR_FP | MSR_FE0 | MSR_FE1);
+	regs_set_return_msr(regs, regs->msr & ~(MSR_FP | MSR_FE0 | MSR_FE1));
 
 	unsafe_copy_fpr_from_user(current, &sr->mc_fregs, failed);
 
-	regs->msr &= ~MSR_VSX;
+	regs_set_return_msr(regs, regs->msr & ~MSR_VSX);
 	if (msr & MSR_VSX) {
 		/*
 		 * Restore altivec registers from the stack to a local
@@ -672,7 +672,7 @@ static long restore_tm_user_regs(struct pt_regs *regs,
 	 *
 	 * Pull in the MSR TM bits from the user context
 	 */
-	regs->msr = (regs->msr & ~MSR_TS_MASK) | (msr_hi & MSR_TS_MASK);
+	regs_set_return_msr(regs, (regs->msr & ~MSR_TS_MASK) | (msr_hi & MSR_TS_MASK));
 	/* Now, recheckpoint.  This loads up all of the checkpointed (older)
 	 * registers, including FP and V[S]Rs.  After recheckpointing, the
 	 * transactional versions should be loaded.
@@ -687,11 +687,11 @@ static long restore_tm_user_regs(struct pt_regs *regs,
 	msr_check_and_set(msr & (MSR_FP | MSR_VEC));
 	if (msr & MSR_FP) {
 		load_fp_state(&current->thread.fp_state);
-		regs->msr |= (MSR_FP | current->thread.fpexc_mode);
+		regs_set_return_msr(regs, regs->msr | (MSR_FP | current->thread.fpexc_mode));
 	}
 	if (msr & MSR_VEC) {
 		load_vr_state(&current->thread.vr_state);
-		regs->msr |= MSR_VEC;
+		regs_set_return_msr(regs, regs->msr | MSR_VEC);
 	}
 
 	preempt_enable();
@@ -801,10 +801,10 @@ int handle_rt_signal32(struct ksignal *ksig, sigset_t *oldset,
 	regs->gpr[4] = (unsigned long)&frame->info;
 	regs->gpr[5] = (unsigned long)&frame->uc;
 	regs->gpr[6] = (unsigned long)frame;
-	regs->nip = (unsigned long) ksig->ka.sa.sa_handler;
+	regs_set_return_ip(regs, (unsigned long) ksig->ka.sa.sa_handler);
 	/* enter the signal handler in native-endian mode */
-	regs->msr &= ~MSR_LE;
-	regs->msr |= (MSR_KERNEL & MSR_LE);
+	regs_set_return_msr(regs, (regs->msr & ~MSR_LE) | (MSR_KERNEL & MSR_LE));
+
 	return 0;
 
 failed:
@@ -889,10 +889,10 @@ int handle_signal32(struct ksignal *ksig, sigset_t *oldset,
 	regs->gpr[1] = newsp;
 	regs->gpr[3] = ksig->sig;
 	regs->gpr[4] = (unsigned long) sc;
-	regs->nip = (unsigned long)ksig->ka.sa.sa_handler;
+	regs_set_return_ip(regs, (unsigned long) ksig->ka.sa.sa_handler);
 	/* enter the signal handler in native-endian mode */
-	regs->msr &= ~MSR_LE;
-	regs->msr |= (MSR_KERNEL & MSR_LE);
+	regs_set_return_msr(regs, (regs->msr & ~MSR_LE) | (MSR_KERNEL & MSR_LE));
+
 	return 0;
 
 failed:
@@ -1142,7 +1142,7 @@ SYSCALL_DEFINE0(rt_sigreturn)
 		 * set, and recheckpoint was not called. This avoid
 		 * hitting a TM Bad thing at RFID
 		 */
-		regs->msr &= ~MSR_TS_MASK;
+		regs_set_return_msr(regs, regs->msr & ~MSR_TS_MASK);
 	}
 	/* Fall through, for non-TM restore */
 #endif
@@ -1231,7 +1231,7 @@ SYSCALL_DEFINE3(debug_setcontext, struct ucontext __user *, ctx,
 	   affect the contents of these registers.  After this point,
 	   failure is a problem, anyway, and it's very unlikely unless
 	   the user is really doing something wrong. */
-	regs->msr = new_msr;
+	regs_set_return_msr(regs, new_msr);
 #ifdef CONFIG_PPC_ADV_DEBUG_REGS
 	current->thread.debug.dbcr0 = new_dbcr0;
 #endif
diff --git a/arch/powerpc/kernel/signal_64.c b/arch/powerpc/kernel/signal_64.c
index 16b41470ec92..1831bba0582e 100644
--- a/arch/powerpc/kernel/signal_64.c
+++ b/arch/powerpc/kernel/signal_64.c
@@ -354,7 +354,7 @@ static long notrace __unsafe_restore_sigcontext(struct task_struct *tsk, sigset_
 	/* get MSR separately, transfer the LE bit if doing signal return */
 	unsafe_get_user(msr, &sc->gp_regs[PT_MSR], efault_out);
 	if (sig)
-		regs->msr = (regs->msr & ~MSR_LE) | (msr & MSR_LE);
+		regs_set_return_msr(regs, (regs->msr & ~MSR_LE) | (msr & MSR_LE));
 	unsafe_get_user(regs->orig_gpr3, &sc->gp_regs[PT_ORIG_R3], efault_out);
 	unsafe_get_user(regs->ctr, &sc->gp_regs[PT_CTR], efault_out);
 	unsafe_get_user(regs->link, &sc->gp_regs[PT_LNK], efault_out);
@@ -376,7 +376,7 @@ static long notrace __unsafe_restore_sigcontext(struct task_struct *tsk, sigset_
 	 * This has to be done before copying stuff into tsk->thread.fpr/vr
 	 * for the reasons explained in the previous comment.
 	 */
-	regs->msr &= ~(MSR_FP | MSR_FE0 | MSR_FE1 | MSR_VEC | MSR_VSX);
+	regs_set_return_msr(regs, regs->msr & ~(MSR_FP | MSR_FE0 | MSR_FE1 | MSR_VEC | MSR_VSX));
 
 #ifdef CONFIG_ALTIVEC
 	unsafe_get_user(v_regs, &sc->v_regs, efault_out);
@@ -468,7 +468,7 @@ static long restore_tm_sigcontexts(struct task_struct *tsk,
 		return -EINVAL;
 
 	/* pull in MSR LE from user context */
-	regs->msr = (regs->msr & ~MSR_LE) | (msr & MSR_LE);
+	regs_set_return_msr(regs, (regs->msr & ~MSR_LE) | (msr & MSR_LE));
 
 	/* The following non-GPR non-FPR non-VR state is also checkpointed: */
 	err |= __get_user(regs->ctr, &tm_sc->gp_regs[PT_CTR]);
@@ -495,7 +495,7 @@ static long restore_tm_sigcontexts(struct task_struct *tsk,
 	 * This has to be done before copying stuff into tsk->thread.fpr/vr
 	 * for the reasons explained in the previous comment.
 	 */
-	regs->msr &= ~(MSR_FP | MSR_FE0 | MSR_FE1 | MSR_VEC | MSR_VSX);
+	regs_set_return_msr(regs, regs->msr & ~(MSR_FP | MSR_FE0 | MSR_FE1 | MSR_VEC | MSR_VSX));
 
 #ifdef CONFIG_ALTIVEC
 	err |= __get_user(v_regs, &sc->v_regs);
@@ -565,7 +565,7 @@ static long restore_tm_sigcontexts(struct task_struct *tsk,
 	preempt_disable();
 
 	/* pull in MSR TS bits from user context */
-	regs->msr |= msr & MSR_TS_MASK;
+	regs_set_return_msr(regs, regs->msr | (msr & MSR_TS_MASK));
 
 	/*
 	 * Ensure that TM is enabled in regs->msr before we leave the signal
@@ -583,7 +583,7 @@ static long restore_tm_sigcontexts(struct task_struct *tsk,
 	 * to be de-scheduled with MSR[TS] set but without calling
 	 * tm_recheckpoint(). This can cause a bug.
 	 */
-	regs->msr |= MSR_TM;
+	regs_set_return_msr(regs, regs->msr | MSR_TM);
 
 	/* This loads the checkpointed FP/VEC state, if used */
 	tm_recheckpoint(&tsk->thread);
@@ -591,11 +591,11 @@ static long restore_tm_sigcontexts(struct task_struct *tsk,
 	msr_check_and_set(msr & (MSR_FP | MSR_VEC));
 	if (msr & MSR_FP) {
 		load_fp_state(&tsk->thread.fp_state);
-		regs->msr |= (MSR_FP | tsk->thread.fpexc_mode);
+		regs_set_return_msr(regs, regs->msr | (MSR_FP | tsk->thread.fpexc_mode));
 	}
 	if (msr & MSR_VEC) {
 		load_vr_state(&tsk->thread.vr_state);
-		regs->msr |= MSR_VEC;
+		regs_set_return_msr(regs, regs->msr | MSR_VEC);
 	}
 
 	preempt_enable();
@@ -717,6 +717,7 @@ SYSCALL_DEFINE3(swapcontext, struct ucontext __user *, old_ctx,
 
 	/* This returns like rt_sigreturn */
 	set_thread_flag(TIF_RESTOREALL);
+
 	return 0;
 
 efault_out:
@@ -783,7 +784,7 @@ SYSCALL_DEFINE0(rt_sigreturn)
 		 * the MSR[TS] that came from user context later, at
 		 * restore_tm_sigcontexts.
 		 */
-		regs->msr &= ~MSR_TS_MASK;
+		regs_set_return_msr(regs, regs->msr & ~MSR_TS_MASK);
 
 		if (__get_user(msr, &uc->uc_mcontext.gp_regs[PT_MSR]))
 			goto badframe;
@@ -815,7 +816,8 @@ SYSCALL_DEFINE0(rt_sigreturn)
 		 * MSR[TS] set, but without CPU in the proper state,
 		 * causing a TM bad thing.
 		 */
-		current->thread.regs->msr &= ~MSR_TS_MASK;
+		regs_set_return_msr(current->thread.regs,
+				current->thread.regs->msr & ~MSR_TS_MASK);
 		if (!user_read_access_begin(&uc->uc_mcontext, sizeof(uc->uc_mcontext)))
 			goto badframe;
 
@@ -829,6 +831,7 @@ SYSCALL_DEFINE0(rt_sigreturn)
 		goto badframe;
 
 	set_thread_flag(TIF_RESTOREALL);
+
 	return 0;
 
 badframe_block:
@@ -908,12 +911,12 @@ int handle_rt_signal64(struct ksignal *ksig, sigset_t *set,
 
 	/* Set up to return from userspace. */
 	if (tsk->mm->context.vdso) {
-		regs->nip = VDSO64_SYMBOL(tsk->mm->context.vdso, sigtramp_rt64);
+		regs_set_return_ip(regs, VDSO64_SYMBOL(tsk->mm->context.vdso, sigtramp_rt64));
 	} else {
 		err |= setup_trampoline(__NR_rt_sigreturn, &frame->tramp[0]);
 		if (err)
 			goto badframe;
-		regs->nip = (unsigned long) &frame->tramp[0];
+		regs_set_return_ip(regs, (unsigned long) &frame->tramp[0]);
 	}
 
 	/* Allocate a dummy caller frame for the signal handler. */
@@ -938,8 +941,7 @@ int handle_rt_signal64(struct ksignal *ksig, sigset_t *set,
 	}
 
 	/* enter the signal handler in native-endian mode */
-	regs->msr &= ~MSR_LE;
-	regs->msr |= (MSR_KERNEL & MSR_LE);
+	regs_set_return_msr(regs, (regs->msr & ~MSR_LE) | (MSR_KERNEL & MSR_LE));
 	regs->gpr[1] = newsp;
 	regs->gpr[3] = ksig->sig;
 	regs->result = 0;
diff --git a/arch/powerpc/kernel/syscalls.c b/arch/powerpc/kernel/syscalls.c
index a552c9e68d7e..bf4ae0f0e36c 100644
--- a/arch/powerpc/kernel/syscalls.c
+++ b/arch/powerpc/kernel/syscalls.c
@@ -114,7 +114,8 @@ SYSCALL_DEFINE0(switch_endian)
 {
 	struct thread_info *ti;
 
-	current->thread.regs->msr ^= MSR_LE;
+	regs_set_return_msr(current->thread.regs,
+				current->thread.regs->msr ^ MSR_LE);
 
 	/*
 	 * Set TIF_RESTOREALL so that r3 isn't clobbered on return to
diff --git a/arch/powerpc/kernel/traps.c b/arch/powerpc/kernel/traps.c
index c929d93c35d0..dfbce527c98e 100644
--- a/arch/powerpc/kernel/traps.c
+++ b/arch/powerpc/kernel/traps.c
@@ -428,7 +428,7 @@ void hv_nmi_check_nonrecoverable(struct pt_regs *regs)
 	return;
 
 nonrecoverable:
-	regs->msr &= ~MSR_RI;
+	regs_set_return_msr(regs, regs->msr & ~MSR_RI);
 #endif
 }
 DEFINE_INTERRUPT_HANDLER_NMI(system_reset_exception)
@@ -550,8 +550,8 @@ static inline int check_io_access(struct pt_regs *regs)
 			printk(KERN_DEBUG "%s bad port %lx at %p\n",
 			       (*nip & 0x100)? "OUT to": "IN from",
 			       regs->gpr[rb] - _IO_BASE, nip);
-			regs->msr |= MSR_RI;
-			regs->nip = extable_fixup(entry);
+			regs_set_return_msr(regs, regs->msr | MSR_RI);
+			regs_set_return_ip(regs, extable_fixup(entry));
 			return 1;
 		}
 	}
@@ -587,8 +587,8 @@ static inline int check_io_access(struct pt_regs *regs)
 #define REASON_BOUNDARY		SRR1_BOUNDARY
 
 #define single_stepping(regs)	((regs)->msr & MSR_SE)
-#define clear_single_step(regs)	((regs)->msr &= ~MSR_SE)
-#define clear_br_trace(regs)	((regs)->msr &= ~MSR_BE)
+#define clear_single_step(regs)	(regs_set_return_msr((regs), (regs)->msr & ~MSR_SE))
+#define clear_br_trace(regs)	(regs_set_return_msr((regs), (regs)->msr & ~MSR_BE))
 #endif
 
 #define inst_length(reason)	(((reason) & REASON_PREFIXED) ? 8 : 4)
@@ -1032,7 +1032,7 @@ static void p9_hmi_special_emu(struct pt_regs *regs)
 #endif /* !__LITTLE_ENDIAN__ */
 
 	/* Go to next instruction */
-	regs->nip += 4;
+	regs_add_return_ip(regs, 4);
 }
 #endif /* CONFIG_VSX */
 
@@ -1477,7 +1477,7 @@ static void do_program_check(struct pt_regs *regs)
 
 		if (!(regs->msr & MSR_PR) &&  /* not user-mode */
 		    report_bug(bugaddr, regs) == BUG_TRAP_TYPE_WARN) {
-			regs->nip += 4;
+			regs_add_return_ip(regs, 4);
 			return;
 		}
 		_exception(SIGTRAP, regs, TRAP_BRKPT, regs->nip);
@@ -1539,7 +1539,7 @@ static void do_program_check(struct pt_regs *regs)
 	if (reason & (REASON_ILLEGAL | REASON_PRIVILEGED)) {
 		switch (emulate_instruction(regs)) {
 		case 0:
-			regs->nip += 4;
+			regs_add_return_ip(regs, 4);
 			emulate_single_step(regs);
 			return;
 		case -EFAULT:
@@ -1567,7 +1567,7 @@ DEFINE_INTERRUPT_HANDLER(program_check_exception)
  */
 DEFINE_INTERRUPT_HANDLER(emulation_assist_interrupt)
 {
-	regs->msr |= REASON_ILLEGAL;
+	regs_set_return_msr(regs, regs->msr | REASON_ILLEGAL);
 	do_program_check(regs);
 }
 
@@ -1594,7 +1594,7 @@ DEFINE_INTERRUPT_HANDLER(alignment_exception)
 
 	if (fixed == 1) {
 		/* skip over emulated instruction */
-		regs->nip += inst_length(reason);
+		regs_add_return_ip(regs, inst_length(reason));
 		emulate_single_step(regs);
 		return;
 	}
@@ -1660,7 +1660,7 @@ static void tm_unavailable(struct pt_regs *regs)
 #ifdef CONFIG_PPC_TRANSACTIONAL_MEM
 	if (user_mode(regs)) {
 		current->thread.load_tm++;
-		regs->msr |= MSR_TM;
+		regs_set_return_msr(regs, regs->msr | MSR_TM);
 		tm_enable();
 		tm_restore_sprs(&current->thread);
 		return;
@@ -1752,7 +1752,7 @@ DEFINE_INTERRUPT_HANDLER(facility_unavailable_exception)
 				pr_err("DSCR based mfspr emulation failed\n");
 				return;
 			}
-			regs->nip += 4;
+			regs_add_return_ip(regs, 4);
 			emulate_single_step(regs);
 		}
 		return;
@@ -1949,7 +1949,7 @@ static void handle_debug(struct pt_regs *regs, unsigned long debug_status)
 	 */
 	if (DBCR_ACTIVE_EVENTS(current->thread.debug.dbcr0,
 			       current->thread.debug.dbcr1))
-		regs->msr |= MSR_DE;
+		regs_set_return_msr(regs, regs->msr | MSR_DE);
 	else
 		/* Make sure the IDM flag is off */
 		current->thread.debug.dbcr0 &= ~DBCR0_IDM;
@@ -1970,7 +1970,7 @@ DEFINE_INTERRUPT_HANDLER(DebugException)
 	 * instead of stopping here when hitting a BT
 	 */
 	if (debug_status & DBSR_BT) {
-		regs->msr &= ~MSR_DE;
+		regs_set_return_msr(regs, regs->msr & ~MSR_DE);
 
 		/* Disable BT */
 		mtspr(SPRN_DBCR0, mfspr(SPRN_DBCR0) & ~DBCR0_BT);
@@ -1981,7 +1981,7 @@ DEFINE_INTERRUPT_HANDLER(DebugException)
 		if (user_mode(regs)) {
 			current->thread.debug.dbcr0 &= ~DBCR0_BT;
 			current->thread.debug.dbcr0 |= DBCR0_IDM | DBCR0_IC;
-			regs->msr |= MSR_DE;
+			regs_set_return_msr(regs, regs->msr | MSR_DE);
 			return;
 		}
 
@@ -1995,7 +1995,7 @@ DEFINE_INTERRUPT_HANDLER(DebugException)
 		if (debugger_sstep(regs))
 			return;
 	} else if (debug_status & DBSR_IC) { 	/* Instruction complete */
-		regs->msr &= ~MSR_DE;
+		regs_set_return_msr(regs, regs->msr & ~MSR_DE);
 
 		/* Disable instruction completion */
 		mtspr(SPRN_DBCR0, mfspr(SPRN_DBCR0) & ~DBCR0_IC);
@@ -2017,7 +2017,7 @@ DEFINE_INTERRUPT_HANDLER(DebugException)
 			current->thread.debug.dbcr0 &= ~DBCR0_IC;
 			if (DBCR_ACTIVE_EVENTS(current->thread.debug.dbcr0,
 					       current->thread.debug.dbcr1))
-				regs->msr |= MSR_DE;
+				regs_set_return_msr(regs, regs->msr | MSR_DE);
 			else
 				/* Make sure the IDM bit is off */
 				current->thread.debug.dbcr0 &= ~DBCR0_IDM;
@@ -2045,7 +2045,7 @@ DEFINE_INTERRUPT_HANDLER(altivec_assist_exception)
 	PPC_WARN_EMULATED(altivec, regs);
 	err = emulate_altivec(regs);
 	if (err == 0) {
-		regs->nip += 4;		/* skip emulated instruction */
+		regs_add_return_ip(regs, 4); /* skip emulated instruction */
 		emulate_single_step(regs);
 		return;
 	}
@@ -2110,7 +2110,7 @@ DEFINE_INTERRUPT_HANDLER(SPEFloatingPointException)
 
 	err = do_spe_mathemu(regs);
 	if (err == 0) {
-		regs->nip += 4;		/* skip emulated instruction */
+		regs_add_return_ip(regs, 4); /* skip emulated instruction */
 		emulate_single_step(regs);
 		return;
 	}
@@ -2141,10 +2141,10 @@ DEFINE_INTERRUPT_HANDLER(SPEFloatingPointRoundException)
 		giveup_spe(current);
 	preempt_enable();
 
-	regs->nip -= 4;
+	regs_add_return_ip(regs, -4);
 	err = speround_handler(regs);
 	if (err == 0) {
-		regs->nip += 4;		/* skip emulated instruction */
+		regs_add_return_ip(regs, 4); /* skip emulated instruction */
 		emulate_single_step(regs);
 		return;
 	}
diff --git a/arch/powerpc/kernel/uprobes.c b/arch/powerpc/kernel/uprobes.c
index 1aefd2ecbb8a..c6975467d9ff 100644
--- a/arch/powerpc/kernel/uprobes.c
+++ b/arch/powerpc/kernel/uprobes.c
@@ -62,7 +62,7 @@ int arch_uprobe_pre_xol(struct arch_uprobe *auprobe, struct pt_regs *regs)
 
 	autask->saved_trap_nr = current->thread.trap_nr;
 	current->thread.trap_nr = UPROBE_TRAP_NR;
-	regs->nip = current->utask->xol_vaddr;
+	regs_set_return_ip(regs, current->utask->xol_vaddr);
 
 	user_enable_single_step(current);
 	return 0;
@@ -119,7 +119,7 @@ int arch_uprobe_post_xol(struct arch_uprobe *auprobe, struct pt_regs *regs)
 	 * support doesn't exist and have to fix-up the next instruction
 	 * to be executed.
 	 */
-	regs->nip = (unsigned long)ppc_inst_next((void *)utask->vaddr, auprobe->insn);
+	regs_set_return_ip(regs, (unsigned long)ppc_inst_next((void *)utask->vaddr, auprobe->insn));
 
 	user_disable_single_step(current);
 	return 0;
diff --git a/arch/powerpc/kernel/vector.S b/arch/powerpc/kernel/vector.S
index 54dbefcb4cde..fc120fac1910 100644
--- a/arch/powerpc/kernel/vector.S
+++ b/arch/powerpc/kernel/vector.S
@@ -73,6 +73,10 @@ _GLOBAL(load_up_altivec)
 	addi	r5,r4,THREAD		/* Get THREAD */
 	oris	r12,r12,MSR_VEC@h
 	std	r12,_MSR(r1)
+#ifdef CONFIG_PPC_BOOK3S_64
+	li	r4,0
+	stb	r4,PACASRR_VALID(r13)
+#endif
 #endif
 	li	r4,1
 	stb	r4,THREAD_LOAD_VEC(r5)
@@ -131,6 +135,8 @@ _GLOBAL(load_up_vsx)
 	/* enable use of VSX after return */
 	oris	r12,r12,MSR_VSX@h
 	std	r12,_MSR(r1)
+	li	r4,0
+	stb	r4,PACASRR_VALID(r13)
 	b	fast_interrupt_return_srr
 
 #endif /* CONFIG_VSX */