Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net

A set of overlapping changes in macvlan and the rocker
driver, nothing serious.

Signed-off-by: David S. Miller <davem@davemloft.net>

Documentation/devicetree/bindings/gpio/gpio-mvebu.txt

@@ -41,9 +41,9 @@ Required properties:
 Optional properties:
 
 In order to use the GPIO lines in PWM mode, some additional optional
-properties are required. Only Armada 370 and XP support these properties.
+properties are required.
 
-- compatible: Must contain "marvell,armada-370-xp-gpio"
+- compatible: Must contain "marvell,armada-370-gpio"
 
 - reg: an additional register set is needed, for the GPIO Blink
   Counter on/off registers.
@@ -71,7 +71,7 @@ Example:
 		};
 
 		gpio1: gpio@18140 {
-			compatible = "marvell,armada-370-xp-gpio";
+			compatible = "marvell,armada-370-gpio";
 			reg = <0x18140 0x40>, <0x181c8 0x08>;
 			reg-names = "gpio", "pwm";
 			ngpios = <17>;

Documentation/devicetree/bindings/mfd/stm32-timers.txt

@@ -31,7 +31,7 @@ Example:
 		compatible = "st,stm32-timers";
 		reg = <0x40010000 0x400>;
 		clocks = <&rcc 0 160>;
-		clock-names = "clk_int";
+		clock-names = "int";
 
 		pwm {
 			compatible = "st,stm32-pwm";

MAINTAINERS

@@ -2964,7 +2964,7 @@ F:	sound/pci/oxygen/
 
 C6X ARCHITECTURE
 M:	Mark Salter <msalter@redhat.com>
-M:	Aurelien Jacquiot <a-jacquiot@ti.com>
+M:	Aurelien Jacquiot <jacquiot.aurelien@gmail.com>
 L:	linux-c6x-dev@linux-c6x.org
 W:	http://www.linux-c6x.org/wiki/index.php/Main_Page
 S:	Maintained

Makefile

@@ -1,7 +1,7 @@
 VERSION = 4
 PATCHLEVEL = 12
 SUBLEVEL = 0
-EXTRAVERSION = -rc6
+EXTRAVERSION = -rc7
 NAME = Fearless Coyote
 
 # *DOCUMENTATION*
@@ -1437,7 +1437,7 @@ help:
 	@echo  '  make V=0|1 [targets] 0 => quiet build (default), 1 => verbose build'
 	@echo  '  make V=2   [targets] 2 => give reason for rebuild of target'
 	@echo  '  make O=dir [targets] Locate all output files in "dir", including .config'
-	@echo  '  make C=1   [targets] Check all c source with $$CHECK (sparse by default)'
+	@echo  '  make C=1   [targets] Check re-compiled c source with $$CHECK (sparse by default)'
 	@echo  '  make C=2   [targets] Force check of all c source with $$CHECK'
 	@echo  '  make RECORDMCOUNT_WARN=1 [targets] Warn about ignored mcount sections'
 	@echo  '  make W=n   [targets] Enable extra gcc checks, n=1,2,3 where'

arch/arc/include/asm/processor.h

@@ -86,8 +86,6 @@ struct task_struct;
 #define TSK_K_BLINK(tsk)	TSK_K_REG(tsk, 4)
 #define TSK_K_FP(tsk)		TSK_K_REG(tsk, 0)
 
-#define thread_saved_pc(tsk)	TSK_K_BLINK(tsk)
-
 extern void start_thread(struct pt_regs * regs, unsigned long pc,
 			 unsigned long usp);
 

arch/arm/Kconfig

@@ -1416,6 +1416,7 @@ choice
 	config VMSPLIT_3G
 		bool "3G/1G user/kernel split"
 	config VMSPLIT_3G_OPT
+		depends on !ARM_LPAE
 		bool "3G/1G user/kernel split (for full 1G low memory)"
 	config VMSPLIT_2G
 		bool "2G/2G user/kernel split"

arch/arm/boot/compressed/efi-header.S

@@ -17,7 +17,8 @@
 		@ there.
 		.inst	'M' | ('Z' << 8) | (0x1310 << 16)   @ tstne r0, #0x4d000
 #else
-		W(mov)	r0, r0
+ AR_CLASS(	mov	r0, r0		)
+  M_CLASS(	nop.w			)
 #endif
 		.endm
 

arch/arm/kernel/setup.c

@@ -315,7 +315,7 @@ static void __init cacheid_init(void)
 	if (arch >= CPU_ARCH_ARMv6) {
 		unsigned int cachetype = read_cpuid_cachetype();
 
-		if ((arch == CPU_ARCH_ARMv7M) && !cachetype) {
+		if ((arch == CPU_ARCH_ARMv7M) && !(cachetype & 0xf000f)) {
 			cacheid = 0;
 		} else if ((cachetype & (7 << 29)) == 4 << 29) {
 			/* ARMv7 register format */

arch/arm64/kernel/vdso.c

@@ -221,10 +221,11 @@ void update_vsyscall(struct timekeeper *tk)
 		/* tkr_mono.cycle_last == tkr_raw.cycle_last */
 		vdso_data->cs_cycle_last	= tk->tkr_mono.cycle_last;
 		vdso_data->raw_time_sec		= tk->raw_time.tv_sec;
-		vdso_data->raw_time_nsec	= tk->raw_time.tv_nsec;
+		vdso_data->raw_time_nsec	= (tk->raw_time.tv_nsec <<
+						   tk->tkr_raw.shift) +
+						  tk->tkr_raw.xtime_nsec;
 		vdso_data->xtime_clock_sec	= tk->xtime_sec;
 		vdso_data->xtime_clock_nsec	= tk->tkr_mono.xtime_nsec;
-		/* tkr_raw.xtime_nsec == 0 */
 		vdso_data->cs_mono_mult		= tk->tkr_mono.mult;
 		vdso_data->cs_raw_mult		= tk->tkr_raw.mult;
 		/* tkr_mono.shift == tkr_raw.shift */

arch/arm64/kernel/vdso/gettimeofday.S

@@ -256,7 +256,6 @@ monotonic_raw:
 	seqcnt_check fail=monotonic_raw
 
 	/* All computations are done with left-shifted nsecs. */
-	lsl	x14, x14, x12
 	get_nsec_per_sec res=x9
 	lsl	x9, x9, x12
 

arch/blackfin/include/asm/processor.h

@@ -75,11 +75,6 @@ static inline void release_thread(struct task_struct *dead_task)
 {
 }
 
-/*
- * Return saved PC of a blocked thread.
- */
-#define thread_saved_pc(tsk)	(tsk->thread.pc)
-
 unsigned long get_wchan(struct task_struct *p);
 
 #define	KSTK_EIP(tsk)							\

arch/c6x/include/asm/processor.h

@@ -95,11 +95,6 @@ static inline void release_thread(struct task_struct *dead_task)
 #define copy_segments(tsk, mm)		do { } while (0)
 #define release_segments(mm)		do { } while (0)
 
-/*
- * saved PC of a blocked thread.
- */
-#define thread_saved_pc(tsk) (task_pt_regs(tsk)->pc)
-
 /*
  * saved kernel SP and DP of a blocked thread.
  */

arch/cris/arch-v10/kernel/process.c

@@ -69,14 +69,6 @@ void hard_reset_now (void)
 	while(1) /* waiting for RETRIBUTION! */ ;
 }
 
-/*
- * Return saved PC of a blocked thread.
- */
-unsigned long thread_saved_pc(struct task_struct *t)
-{
-	return task_pt_regs(t)->irp;
-}
-
 /* setup the child's kernel stack with a pt_regs and switch_stack on it.
  * it will be un-nested during _resume and _ret_from_sys_call when the
  * new thread is scheduled.

arch/cris/arch-v32/kernel/process.c

@@ -84,14 +84,6 @@ hard_reset_now(void)
 		; /* Wait for reset. */
 }
 
-/*
- * Return saved PC of a blocked thread.
- */
-unsigned long thread_saved_pc(struct task_struct *t)
-{
-	return task_pt_regs(t)->erp;
-}
-
 /*
  * Setup the child's kernel stack with a pt_regs and call switch_stack() on it.
  * It will be unnested during _resume and _ret_from_sys_call when the new thread

arch/cris/include/asm/processor.h

@@ -52,8 +52,6 @@ unsigned long get_wchan(struct task_struct *p);
 
 #define KSTK_ESP(tsk)   ((tsk) == current ? rdusp() : (tsk)->thread.usp)
 
-extern unsigned long thread_saved_pc(struct task_struct *tsk);
-
 /* Free all resources held by a thread. */
 static inline void release_thread(struct task_struct *dead_task)
 {

arch/frv/include/asm/processor.h

@@ -96,11 +96,6 @@ extern asmlinkage void *restore_user_regs(const struct user_context *target, ...
 #define release_segments(mm)		do { } while (0)
 #define forget_segments()		do { } while (0)
 
-/*
- * Return saved PC of a blocked thread.
- */
-extern unsigned long thread_saved_pc(struct task_struct *tsk);
-
 unsigned long get_wchan(struct task_struct *p);
 
 #define	KSTK_EIP(tsk)	((tsk)->thread.frame0->pc)

arch/frv/kernel/process.c

@@ -198,15 +198,6 @@ unsigned long get_wchan(struct task_struct *p)
 	return 0;
 }
 
-unsigned long thread_saved_pc(struct task_struct *tsk)
-{
-	/* Check whether the thread is blocked in resume() */
-	if (in_sched_functions(tsk->thread.pc))
-		return ((unsigned long *)tsk->thread.fp)[2];
-	else
-		return tsk->thread.pc;
-}
-
 int elf_check_arch(const struct elf32_hdr *hdr)
 {
 	unsigned long hsr0 = __get_HSR(0);

arch/h8300/include/asm/processor.h

@@ -110,10 +110,6 @@ static inline void release_thread(struct task_struct *dead_task)
 {
 }
 
-/*
- * Return saved PC of a blocked thread.
- */
-unsigned long thread_saved_pc(struct task_struct *tsk);
 unsigned long get_wchan(struct task_struct *p);
 
 #define	KSTK_EIP(tsk)	\

arch/h8300/kernel/process.c

@@ -129,11 +129,6 @@ int copy_thread(unsigned long clone_flags,
 	return 0;
 }
 
-unsigned long thread_saved_pc(struct task_struct *tsk)
-{
-	return ((struct pt_regs *)tsk->thread.esp0)->pc;
-}
-
 unsigned long get_wchan(struct task_struct *p)
 {
 	unsigned long fp, pc;

arch/hexagon/include/asm/processor.h

@@ -33,9 +33,6 @@
 /*  task_struct, defined elsewhere, is the "process descriptor" */
 struct task_struct;
 
-/*  this is defined in arch/process.c  */
-extern unsigned long thread_saved_pc(struct task_struct *tsk);
-
 extern void start_thread(struct pt_regs *, unsigned long, unsigned long);
 
 /*

arch/hexagon/kernel/process.c

@@ -60,14 +60,6 @@ void arch_cpu_idle(void)
 	local_irq_enable();
 }
 
-/*
- *  Return saved PC of a blocked thread
- */
-unsigned long thread_saved_pc(struct task_struct *tsk)
-{
-	return 0;
-}
-
 /*
  * Copy architecture-specific thread state
  */

arch/ia64/include/asm/processor.h

@@ -601,23 +601,6 @@ ia64_set_unat (__u64 *unat, void *spill_addr, unsigned long nat)
 	*unat = (*unat & ~mask) | (nat << bit);
 }
 
-/*
- * Return saved PC of a blocked thread.
- * Note that the only way T can block is through a call to schedule() -> switch_to().
- */
-static inline unsigned long
-thread_saved_pc (struct task_struct *t)
-{
-	struct unw_frame_info info;
-	unsigned long ip;
-
-	unw_init_from_blocked_task(&info, t);
-	if (unw_unwind(&info) < 0)
-		return 0;
-	unw_get_ip(&info, &ip);
-	return ip;
-}
-
 /*
  * Get the current instruction/program counter value.
  */

arch/m32r/include/asm/processor.h

@@ -122,8 +122,6 @@ extern void release_thread(struct task_struct *);
 extern void copy_segments(struct task_struct *p, struct mm_struct * mm);
 extern void release_segments(struct mm_struct * mm);
 
-extern unsigned long thread_saved_pc(struct task_struct *);
-
 /* Copy and release all segment info associated with a VM */
 #define copy_segments(p, mm)  do { } while (0)
 #define release_segments(mm)  do { } while (0)

arch/m32r/kernel/process.c

@@ -39,14 +39,6 @@
 
 #include <linux/err.h>
 
-/*
- * Return saved PC of a blocked thread.
- */
-unsigned long thread_saved_pc(struct task_struct *tsk)
-{
-	return tsk->thread.lr;
-}
-
 void (*pm_power_off)(void) = NULL;
 EXPORT_SYMBOL(pm_power_off);
 

arch/m68k/include/asm/processor.h

@@ -130,8 +130,6 @@ static inline void release_thread(struct task_struct *dead_task)
 {
 }
 
-extern unsigned long thread_saved_pc(struct task_struct *tsk);
-
 unsigned long get_wchan(struct task_struct *p);
 
 #define	KSTK_EIP(tsk)	\

arch/m68k/kernel/process.c

@@ -40,20 +40,6 @@
 asmlinkage void ret_from_fork(void);
 asmlinkage void ret_from_kernel_thread(void);
 
-
-/*
- * Return saved PC from a blocked thread
- */
-unsigned long thread_saved_pc(struct task_struct *tsk)
-{
-	struct switch_stack *sw = (struct switch_stack *)tsk->thread.ksp;
-	/* Check whether the thread is blocked in resume() */
-	if (in_sched_functions(sw->retpc))
-		return ((unsigned long *)sw->a6)[1];
-	else
-		return sw->retpc;
-}
-
 void arch_cpu_idle(void)
 {
 #if defined(MACH_ATARI_ONLY)

arch/microblaze/include/asm/processor.h

@@ -69,8 +69,6 @@ static inline void release_thread(struct task_struct *dead_task)
 {
 }
 
-extern unsigned long thread_saved_pc(struct task_struct *t);
-
 extern unsigned long get_wchan(struct task_struct *p);
 
 # define KSTK_EIP(tsk)	(0)
@@ -121,10 +119,6 @@ static inline void release_thread(struct task_struct *dead_task)
 {
 }
 
-/* Return saved (kernel) PC of a blocked thread.  */
-#  define thread_saved_pc(tsk)	\
-	((tsk)->thread.regs ? (tsk)->thread.regs->r15 : 0)
-
 unsigned long get_wchan(struct task_struct *p);
 
 /* The size allocated for kernel stacks. This _must_ be a power of two! */

arch/microblaze/kernel/process.c

@@ -119,23 +119,6 @@ int copy_thread(unsigned long clone_flags, unsigned long usp,
 	return 0;
 }
 
-#ifndef CONFIG_MMU
-/*
- * Return saved PC of a blocked thread.
- */
-unsigned long thread_saved_pc(struct task_struct *tsk)
-{
-	struct cpu_context *ctx =
-		&(((struct thread_info *)(tsk->stack))->cpu_context);
-
-	/* Check whether the thread is blocked in resume() */
-	if (in_sched_functions(ctx->r15))
-		return (unsigned long)ctx->r15;
-	else
-		return ctx->r14;
-}
-#endif
-
 unsigned long get_wchan(struct task_struct *p)
 {
 /* TBD (used by procfs) */

arch/mips/kvm/tlb.c

@@ -166,7 +166,11 @@ static int _kvm_mips_host_tlb_inv(unsigned long entryhi)
 int kvm_mips_host_tlb_inv(struct kvm_vcpu *vcpu, unsigned long va,
 			  bool user, bool kernel)
 {
-	int idx_user, idx_kernel;
+	/*
+	 * Initialize idx_user and idx_kernel to workaround bogus
+	 * maybe-initialized warning when using GCC 6.
+	 */
+	int idx_user = 0, idx_kernel = 0;
 	unsigned long flags, old_entryhi;
 
 	local_irq_save(flags);

arch/mn10300/include/asm/processor.h

@@ -132,11 +132,6 @@ static inline void start_thread(struct pt_regs *regs,
 /* Free all resources held by a thread. */
 extern void release_thread(struct task_struct *);
 
-/*
- * Return saved PC of a blocked thread.
- */
-extern unsigned long thread_saved_pc(struct task_struct *tsk);
-
 unsigned long get_wchan(struct task_struct *p);
 
 #define task_pt_regs(task) ((task)->thread.uregs)

arch/mn10300/kernel/process.c

@@ -39,14 +39,6 @@
 #include <asm/gdb-stub.h>
 #include "internal.h"
 
-/*
- * return saved PC of a blocked thread.
- */
-unsigned long thread_saved_pc(struct task_struct *tsk)
-{
-	return ((unsigned long *) tsk->thread.sp)[3];
-}
-
 /*
  * power off function, if any
  */

arch/nios2/include/asm/processor.h

@@ -75,9 +75,6 @@ static inline void release_thread(struct task_struct *dead_task)
 {
 }
 
-/* Return saved PC of a blocked thread. */
-#define thread_saved_pc(tsk)	((tsk)->thread.kregs->ea)
-
 extern unsigned long get_wchan(struct task_struct *p);
 
 #define task_pt_regs(p) \

arch/openrisc/include/asm/processor.h

@@ -84,11 +84,6 @@ void start_thread(struct pt_regs *regs, unsigned long nip, unsigned long sp);
 void release_thread(struct task_struct *);
 unsigned long get_wchan(struct task_struct *p);
 
-/*
- * Return saved PC of a blocked thread. For now, this is the "user" PC
- */
-extern unsigned long thread_saved_pc(struct task_struct *t);
-
 #define init_stack      (init_thread_union.stack)
 
 #define cpu_relax()     barrier()

arch/openrisc/kernel/process.c

@@ -110,11 +110,6 @@ void show_regs(struct pt_regs *regs)
 	show_registers(regs);
 }
 
-unsigned long thread_saved_pc(struct task_struct *t)
-{
-	return (unsigned long)user_regs(t->stack)->pc;
-}
-
 void release_thread(struct task_struct *dead_task)
 {
 }

arch/parisc/include/asm/processor.h

@@ -163,12 +163,7 @@ struct thread_struct {
 	.flags		= 0 \
 	}
 
-/*
- * Return saved PC of a blocked thread.  This is used by ps mostly.
- */
-
 struct task_struct;
-unsigned long thread_saved_pc(struct task_struct *t);
 void show_trace(struct task_struct *task, unsigned long *stack);
 
 /*

arch/parisc/kernel/process.c

@@ -239,11 +239,6 @@ copy_thread(unsigned long clone_flags, unsigned long usp,
 	return 0;
 }
 
-unsigned long thread_saved_pc(struct task_struct *t)
-{
-	return t->thread.regs.kpc;
-}
-
 unsigned long
 get_wchan(struct task_struct *p)
 {

arch/powerpc/include/asm/kprobes.h

@@ -103,6 +103,7 @@ extern int kprobe_exceptions_notify(struct notifier_block *self,
 extern int kprobe_fault_handler(struct pt_regs *regs, int trapnr);
 extern int kprobe_handler(struct pt_regs *regs);
 extern int kprobe_post_handler(struct pt_regs *regs);
+extern int is_current_kprobe_addr(unsigned long addr);
 #ifdef CONFIG_KPROBES_ON_FTRACE
 extern int skip_singlestep(struct kprobe *p, struct pt_regs *regs,
 			   struct kprobe_ctlblk *kcb);

arch/powerpc/include/asm/processor.h

@@ -378,12 +378,6 @@ struct thread_struct {
 }
 #endif
 
-/*
- * Return saved PC of a blocked thread. For now, this is the "user" PC
- */
-#define thread_saved_pc(tsk)    \
-        ((tsk)->thread.regs? (tsk)->thread.regs->nip: 0)
-
 #define task_pt_regs(tsk)	((struct pt_regs *)(tsk)->thread.regs)
 
 unsigned long get_wchan(struct task_struct *p);

arch/powerpc/kernel/exceptions-64s.S

@@ -1411,10 +1411,8 @@ USE_TEXT_SECTION()
 	.balign	IFETCH_ALIGN_BYTES
 do_hash_page:
 #ifdef CONFIG_PPC_STD_MMU_64
-	andis.	r0,r4,0xa410		/* weird error? */
+	andis.	r0,r4,0xa450		/* weird error? */
 	bne-	handle_page_fault	/* if not, try to insert a HPTE */
-	andis.  r0,r4,DSISR_DABRMATCH@h
-	bne-    handle_dabr_fault
 	CURRENT_THREAD_INFO(r11, r1)
 	lwz	r0,TI_PREEMPT(r11)	/* If we're in an "NMI" */
 	andis.	r0,r0,NMI_MASK@h	/* (i.e. an irq when soft-disabled) */
@@ -1438,11 +1436,16 @@ do_hash_page:
 
 	/* Error */
 	blt-	13f
+
+	/* Reload DSISR into r4 for the DABR check below */
+	ld      r4,_DSISR(r1)
 #endif /* CONFIG_PPC_STD_MMU_64 */
 
 /* Here we have a page fault that hash_page can't handle. */
 handle_page_fault:
-11:	ld	r4,_DAR(r1)
+11:	andis.  r0,r4,DSISR_DABRMATCH@h
+	bne-    handle_dabr_fault
+	ld	r4,_DAR(r1)
 	ld	r5,_DSISR(r1)
 	addi	r3,r1,STACK_FRAME_OVERHEAD
 	bl	do_page_fault

arch/powerpc/kernel/kprobes.c

@@ -43,6 +43,12 @@ DEFINE_PER_CPU(struct kprobe_ctlblk, kprobe_ctlblk);
 
 struct kretprobe_blackpoint kretprobe_blacklist[] = {{NULL, NULL}};
 
+int is_current_kprobe_addr(unsigned long addr)
+{
+	struct kprobe *p = kprobe_running();
+	return (p && (unsigned long)p->addr == addr) ? 1 : 0;
+}
+
 bool arch_within_kprobe_blacklist(unsigned long addr)
 {
 	return  (addr >= (unsigned long)__kprobes_text_start &&
@@ -617,6 +623,15 @@ int setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs)
 	regs->gpr[2] = (unsigned long)(((func_descr_t *)jp->entry)->toc);
 #endif
 
+	/*
+	 * jprobes use jprobe_return() which skips the normal return
+	 * path of the function, and this messes up the accounting of the
+	 * function graph tracer.
+	 *
+	 * Pause function graph tracing while performing the jprobe function.
+	 */
+	pause_graph_tracing();
+
 	return 1;
 }
 NOKPROBE_SYMBOL(setjmp_pre_handler);
@@ -642,6 +657,8 @@ int longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
 	 * saved regs...
 	 */
 	memcpy(regs, &kcb->jprobe_saved_regs, sizeof(struct pt_regs));
+	/* It's OK to start function graph tracing again */
+	unpause_graph_tracing();
 	preempt_enable_no_resched();
 	return 1;
 }

arch/powerpc/kernel/setup_64.c

@@ -615,6 +615,24 @@ void __init exc_lvl_early_init(void)
 }
 #endif
 
+/*
+ * Emergency stacks are used for a range of things, from asynchronous
+ * NMIs (system reset, machine check) to synchronous, process context.
+ * We set preempt_count to zero, even though that isn't necessarily correct. To
+ * get the right value we'd need to copy it from the previous thread_info, but
+ * doing that might fault causing more problems.
+ * TODO: what to do with accounting?
+ */
+static void emerg_stack_init_thread_info(struct thread_info *ti, int cpu)
+{
+	ti->task = NULL;
+	ti->cpu = cpu;
+	ti->preempt_count = 0;
+	ti->local_flags = 0;
+	ti->flags = 0;
+	klp_init_thread_info(ti);
+}
+
 /*
  * Stack space used when we detect a bad kernel stack pointer, and
  * early in SMP boots before relocation is enabled. Exclusive emergency
@@ -633,24 +651,31 @@ void __init emergency_stack_init(void)
 	 * Since we use these as temporary stacks during secondary CPU
 	 * bringup, we need to get at them in real mode. This means they
 	 * must also be within the RMO region.
+	 *
+	 * The IRQ stacks allocated elsewhere in this file are zeroed and
+	 * initialized in kernel/irq.c. These are initialized here in order
+	 * to have emergency stacks available as early as possible.
 	 */
 	limit = min(safe_stack_limit(), ppc64_rma_size);
 
 	for_each_possible_cpu(i) {
 		struct thread_info *ti;
 		ti = __va(memblock_alloc_base(THREAD_SIZE, THREAD_SIZE, limit));
-		klp_init_thread_info(ti);
+		memset(ti, 0, THREAD_SIZE);
+		emerg_stack_init_thread_info(ti, i);
 		paca[i].emergency_sp = (void *)ti + THREAD_SIZE;
 
 #ifdef CONFIG_PPC_BOOK3S_64
 		/* emergency stack for NMI exception handling. */
 		ti = __va(memblock_alloc_base(THREAD_SIZE, THREAD_SIZE, limit));
-		klp_init_thread_info(ti);
+		memset(ti, 0, THREAD_SIZE);
+		emerg_stack_init_thread_info(ti, i);
 		paca[i].nmi_emergency_sp = (void *)ti + THREAD_SIZE;
 
 		/* emergency stack for machine check exception handling. */
 		ti = __va(memblock_alloc_base(THREAD_SIZE, THREAD_SIZE, limit));
-		klp_init_thread_info(ti);
+		memset(ti, 0, THREAD_SIZE);
+		emerg_stack_init_thread_info(ti, i);
 		paca[i].mc_emergency_sp = (void *)ti + THREAD_SIZE;
 #endif
 	}

arch/powerpc/kernel/trace/ftrace_64_mprofile.S

@@ -45,10 +45,14 @@ _GLOBAL(ftrace_caller)
 	stdu	r1,-SWITCH_FRAME_SIZE(r1)
 
 	/* Save all gprs to pt_regs */
-	SAVE_8GPRS(0,r1)
+	SAVE_GPR(0, r1)
-	SAVE_8GPRS(8,r1)
+	SAVE_10GPRS(2, r1)
-	SAVE_8GPRS(16,r1)
+	SAVE_10GPRS(12, r1)
-	SAVE_8GPRS(24,r1)
+	SAVE_10GPRS(22, r1)
+
+	/* Save previous stack pointer (r1) */
+	addi	r8, r1, SWITCH_FRAME_SIZE
+	std	r8, GPR1(r1)
 
 	/* Load special regs for save below */
 	mfmsr   r8
@@ -95,18 +99,44 @@ ftrace_call:
 	bl	ftrace_stub
 	nop
 
-	/* Load ctr with the possibly modified NIP */
+	/* Load the possibly modified NIP */
-	ld	r3, _NIP(r1)
+	ld	r15, _NIP(r1)
-	mtctr	r3
+
 #ifdef CONFIG_LIVEPATCH
-	cmpd	r14,r3		/* has NIP been altered? */
+	cmpd	r14, r15	/* has NIP been altered? */
 #endif
 
+#if defined(CONFIG_LIVEPATCH) && defined(CONFIG_KPROBES_ON_FTRACE)
+	/* NIP has not been altered, skip over further checks */
+	beq	1f
+
+	/* Check if there is an active kprobe on us */
+	subi	r3, r14, 4
+	bl	is_current_kprobe_addr
+	nop
+
+	/*
+	 * If r3 == 1, then this is a kprobe/jprobe.
+	 * else, this is livepatched function.
+	 *
+	 * The conditional branch for livepatch_handler below will use the
+	 * result of this comparison. For kprobe/jprobe, we just need to branch to
+	 * the new NIP, not call livepatch_handler. The branch below is bne, so we
+	 * want CR0[EQ] to be true if this is a kprobe/jprobe. Which means we want
+	 * CR0[EQ] = (r3 == 1).
+	 */
+	cmpdi	r3, 1
+1:
+#endif
+
+	/* Load CTR with the possibly modified NIP */
+	mtctr	r15
+
 	/* Restore gprs */
-	REST_8GPRS(0,r1)
+	REST_GPR(0,r1)
-	REST_8GPRS(8,r1)
+	REST_10GPRS(2,r1)
-	REST_8GPRS(16,r1)
+	REST_10GPRS(12,r1)
-	REST_8GPRS(24,r1)
+	REST_10GPRS(22,r1)
 
 	/* Restore possibly modified LR */
 	ld	r0, _LINK(r1)
@@ -119,7 +149,10 @@ ftrace_call:
 	addi r1, r1, SWITCH_FRAME_SIZE
 
 #ifdef CONFIG_LIVEPATCH
-        /* Based on the cmpd above, if the NIP was altered handle livepatch */
+        /*
+	 * Based on the cmpd or cmpdi above, if the NIP was altered and we're
+	 * not on a kprobe/jprobe, then handle livepatch.
+	 */
 	bne-	livepatch_handler
 #endif
 

arch/powerpc/kvm/book3s_hv.c

@@ -1486,6 +1486,14 @@ static int kvmppc_set_one_reg_hv(struct kvm_vcpu *vcpu, u64 id,
 		r = set_vpa(vcpu, &vcpu->arch.dtl, addr, len);
 		break;
 	case KVM_REG_PPC_TB_OFFSET:
+		/*
+		 * POWER9 DD1 has an erratum where writing TBU40 causes
+		 * the timebase to lose ticks.  So we don't let the
+		 * timebase offset be changed on P9 DD1.  (It is
+		 * initialized to zero.)
+		 */
+		if (cpu_has_feature(CPU_FTR_POWER9_DD1))
+			break;
 		/* round up to multiple of 2^24 */
 		vcpu->arch.vcore->tb_offset =
 			ALIGN(set_reg_val(id, *val), 1UL << 24);
@@ -2907,12 +2915,36 @@ static int kvmppc_vcpu_run_hv(struct kvm_run *run, struct kvm_vcpu *vcpu)
 {
 	int r;
 	int srcu_idx;
+	unsigned long ebb_regs[3] = {};	/* shut up GCC */
+	unsigned long user_tar = 0;
+	unsigned int user_vrsave;
 
 	if (!vcpu->arch.sane) {
 		run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
 		return -EINVAL;
 	}
 
+	/*
+	 * Don't allow entry with a suspended transaction, because
+	 * the guest entry/exit code will lose it.
+	 * If the guest has TM enabled, save away their TM-related SPRs
+	 * (they will get restored by the TM unavailable interrupt).
+	 */
+#ifdef CONFIG_PPC_TRANSACTIONAL_MEM
+	if (cpu_has_feature(CPU_FTR_TM) && current->thread.regs &&
+	    (current->thread.regs->msr & MSR_TM)) {
+		if (MSR_TM_ACTIVE(current->thread.regs->msr)) {
+			run->exit_reason = KVM_EXIT_FAIL_ENTRY;
+			run->fail_entry.hardware_entry_failure_reason = 0;
+			return -EINVAL;
+		}
+		current->thread.tm_tfhar = mfspr(SPRN_TFHAR);
+		current->thread.tm_tfiar = mfspr(SPRN_TFIAR);
+		current->thread.tm_texasr = mfspr(SPRN_TEXASR);
+		current->thread.regs->msr &= ~MSR_TM;
+	}
+#endif
+
 	kvmppc_core_prepare_to_enter(vcpu);
 
 	/* No need to go into the guest when all we'll do is come back out */
@@ -2934,6 +2966,15 @@ static int kvmppc_vcpu_run_hv(struct kvm_run *run, struct kvm_vcpu *vcpu)
 
 	flush_all_to_thread(current);
 
+	/* Save userspace EBB and other register values */
+	if (cpu_has_feature(CPU_FTR_ARCH_207S)) {
+		ebb_regs[0] = mfspr(SPRN_EBBHR);
+		ebb_regs[1] = mfspr(SPRN_EBBRR);
+		ebb_regs[2] = mfspr(SPRN_BESCR);
+		user_tar = mfspr(SPRN_TAR);
+	}
+	user_vrsave = mfspr(SPRN_VRSAVE);
+
 	vcpu->arch.wqp = &vcpu->arch.vcore->wq;
 	vcpu->arch.pgdir = current->mm->pgd;
 	vcpu->arch.state = KVMPPC_VCPU_BUSY_IN_HOST;
@@ -2960,6 +3001,16 @@ static int kvmppc_vcpu_run_hv(struct kvm_run *run, struct kvm_vcpu *vcpu)
 		}
 	} while (is_kvmppc_resume_guest(r));
 
+	/* Restore userspace EBB and other register values */
+	if (cpu_has_feature(CPU_FTR_ARCH_207S)) {
+		mtspr(SPRN_EBBHR, ebb_regs[0]);
+		mtspr(SPRN_EBBRR, ebb_regs[1]);
+		mtspr(SPRN_BESCR, ebb_regs[2]);
+		mtspr(SPRN_TAR, user_tar);
+		mtspr(SPRN_FSCR, current->thread.fscr);
+	}
+	mtspr(SPRN_VRSAVE, user_vrsave);
+
  out:
 	vcpu->arch.state = KVMPPC_VCPU_NOTREADY;
 	atomic_dec(&vcpu->kvm->arch.vcpus_running);

arch/powerpc/kvm/book3s_hv_interrupts.S

@@ -121,10 +121,20 @@ END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S)
 	 * Put whatever is in the decrementer into the
 	 * hypervisor decrementer.
 	 */
+BEGIN_FTR_SECTION
+	ld	r5, HSTATE_KVM_VCORE(r13)
+	ld	r6, VCORE_KVM(r5)
+	ld	r9, KVM_HOST_LPCR(r6)
+	andis.	r9, r9, LPCR_LD@h
+END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
 	mfspr	r8,SPRN_DEC
 	mftb	r7
-	mtspr	SPRN_HDEC,r8
+BEGIN_FTR_SECTION
+	/* On POWER9, don't sign-extend if host LPCR[LD] bit is set */
+	bne	32f
+END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
 	extsw	r8,r8
+32:	mtspr	SPRN_HDEC,r8
 	add	r8,r8,r7
 	std	r8,HSTATE_DECEXP(r13)
 

arch/powerpc/kvm/book3s_hv_rmhandlers.S

@@ -32,12 +32,29 @@
 #include <asm/opal.h>
 #include <asm/xive-regs.h>
 
+/* Sign-extend HDEC if not on POWER9 */
+#define EXTEND_HDEC(reg)			\
+BEGIN_FTR_SECTION;				\
+	extsw	reg, reg;			\
+END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_300)
+
 #define VCPU_GPRS_TM(reg) (((reg) * ULONG_SIZE) + VCPU_GPR_TM)
 
 /* Values in HSTATE_NAPPING(r13) */
 #define NAPPING_CEDE	1
 #define NAPPING_NOVCPU	2
 
+/* Stack frame offsets for kvmppc_hv_entry */
+#define SFS			144
+#define STACK_SLOT_TRAP		(SFS-4)
+#define STACK_SLOT_TID		(SFS-16)
+#define STACK_SLOT_PSSCR	(SFS-24)
+#define STACK_SLOT_PID		(SFS-32)
+#define STACK_SLOT_IAMR		(SFS-40)
+#define STACK_SLOT_CIABR	(SFS-48)
+#define STACK_SLOT_DAWR		(SFS-56)
+#define STACK_SLOT_DAWRX	(SFS-64)
+
 /*
  * Call kvmppc_hv_entry in real mode.
  * Must be called with interrupts hard-disabled.
@@ -214,6 +231,8 @@ END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S)
 kvmppc_primary_no_guest:
 	/* We handle this much like a ceded vcpu */
 	/* put the HDEC into the DEC, since HDEC interrupts don't wake us */
+	/* HDEC may be larger than DEC for arch >= v3.00, but since the */
+	/* HDEC value came from DEC in the first place, it will fit */
 	mfspr	r3, SPRN_HDEC
 	mtspr	SPRN_DEC, r3
 	/*
@@ -295,8 +314,9 @@ kvm_novcpu_wakeup:
 
 	/* See if our timeslice has expired (HDEC is negative) */
 	mfspr	r0, SPRN_HDEC
+	EXTEND_HDEC(r0)
 	li	r12, BOOK3S_INTERRUPT_HV_DECREMENTER
-	cmpwi	r0, 0
+	cmpdi	r0, 0
 	blt	kvm_novcpu_exit
 
 	/* Got an IPI but other vcpus aren't yet exiting, must be a latecomer */
@@ -319,10 +339,10 @@ kvm_novcpu_exit:
 	bl	kvmhv_accumulate_time
 #endif
 13:	mr	r3, r12
-	stw	r12, 112-4(r1)
+	stw	r12, STACK_SLOT_TRAP(r1)
 	bl	kvmhv_commence_exit
 	nop
-	lwz	r12, 112-4(r1)
+	lwz	r12, STACK_SLOT_TRAP(r1)
 	b	kvmhv_switch_to_host
 
 /*
@@ -390,8 +410,8 @@ kvm_secondary_got_guest:
 	lbz	r4, HSTATE_PTID(r13)
 	cmpwi	r4, 0
 	bne	63f
-	lis	r6, 0x7fff
+	LOAD_REG_ADDR(r6, decrementer_max)
-	ori	r6, r6, 0xffff
+	ld	r6, 0(r6)
 	mtspr	SPRN_HDEC, r6
 	/* and set per-LPAR registers, if doing dynamic micro-threading */
 	ld	r6, HSTATE_SPLIT_MODE(r13)
@@ -545,11 +565,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S)
  *                                                                            *
  *****************************************************************************/
 
-/* Stack frame offsets */
-#define STACK_SLOT_TID		(112-16)
-#define STACK_SLOT_PSSCR	(112-24)
-#define STACK_SLOT_PID		(112-32)
-
 .global kvmppc_hv_entry
 kvmppc_hv_entry:
 
@@ -565,7 +580,7 @@ kvmppc_hv_entry:
 	 */
 	mflr	r0
 	std	r0, PPC_LR_STKOFF(r1)
-	stdu	r1, -112(r1)
+	stdu	r1, -SFS(r1)
 
 	/* Save R1 in the PACA */
 	std	r1, HSTATE_HOST_R1(r13)
@@ -749,10 +764,20 @@ BEGIN_FTR_SECTION
 	mfspr	r5, SPRN_TIDR
 	mfspr	r6, SPRN_PSSCR
 	mfspr	r7, SPRN_PID
+	mfspr	r8, SPRN_IAMR
 	std	r5, STACK_SLOT_TID(r1)
 	std	r6, STACK_SLOT_PSSCR(r1)
 	std	r7, STACK_SLOT_PID(r1)
+	std	r8, STACK_SLOT_IAMR(r1)
 END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
+BEGIN_FTR_SECTION
+	mfspr	r5, SPRN_CIABR
+	mfspr	r6, SPRN_DAWR
+	mfspr	r7, SPRN_DAWRX
+	std	r5, STACK_SLOT_CIABR(r1)
+	std	r6, STACK_SLOT_DAWR(r1)
+	std	r7, STACK_SLOT_DAWRX(r1)
+END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S)
 
 BEGIN_FTR_SECTION
 	/* Set partition DABR */
@@ -968,7 +993,8 @@ ALT_FTR_SECTION_END_IFCLR(CPU_FTR_ARCH_300)
 
 	/* Check if HDEC expires soon */
 	mfspr	r3, SPRN_HDEC
-	cmpwi	r3, 512		/* 1 microsecond */
+	EXTEND_HDEC(r3)
+	cmpdi	r3, 512		/* 1 microsecond */
 	blt	hdec_soon
 
 #ifdef CONFIG_KVM_XICS
@@ -1505,11 +1531,10 @@ ALT_FTR_SECTION_END_IFCLR(CPU_FTR_ARCH_300)
 	 * set by the guest could disrupt the host.
 	 */
 	li	r0, 0
-	mtspr	SPRN_IAMR, r0
+	mtspr	SPRN_PSPB, r0
-	mtspr	SPRN_CIABR, r0
-	mtspr	SPRN_DAWRX, r0
 	mtspr	SPRN_WORT, r0
 BEGIN_FTR_SECTION
+	mtspr	SPRN_IAMR, r0
 	mtspr	SPRN_TCSCR, r0
 	/* Set MMCRS to 1<<31 to freeze and disable the SPMC counters */
 	li	r0, 1
@@ -1525,6 +1550,7 @@ END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_300)
 	std	r6,VCPU_UAMOR(r9)
 	li	r6,0
 	mtspr	SPRN_AMR,r6
+	mtspr	SPRN_UAMOR, r6
 
 	/* Switch DSCR back to host value */
 	mfspr	r8, SPRN_DSCR
@@ -1669,13 +1695,23 @@ END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S)
 	ptesync
 
 	/* Restore host values of some registers */
+BEGIN_FTR_SECTION
+	ld	r5, STACK_SLOT_CIABR(r1)
+	ld	r6, STACK_SLOT_DAWR(r1)
+	ld	r7, STACK_SLOT_DAWRX(r1)
+	mtspr	SPRN_CIABR, r5
+	mtspr	SPRN_DAWR, r6
+	mtspr	SPRN_DAWRX, r7
+END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S)
 BEGIN_FTR_SECTION
 	ld	r5, STACK_SLOT_TID(r1)
 	ld	r6, STACK_SLOT_PSSCR(r1)
 	ld	r7, STACK_SLOT_PID(r1)
+	ld	r8, STACK_SLOT_IAMR(r1)
 	mtspr	SPRN_TIDR, r5
 	mtspr	SPRN_PSSCR, r6
 	mtspr	SPRN_PID, r7
+	mtspr	SPRN_IAMR, r8
 END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
 BEGIN_FTR_SECTION
 	PPC_INVALIDATE_ERAT
@@ -1819,8 +1855,8 @@ END_MMU_FTR_SECTION_IFSET(MMU_FTR_TYPE_RADIX)
 	li	r0, KVM_GUEST_MODE_NONE
 	stb	r0, HSTATE_IN_GUEST(r13)
 
-	ld	r0, 112+PPC_LR_STKOFF(r1)
+	ld	r0, SFS+PPC_LR_STKOFF(r1)
-	addi	r1, r1, 112
+	addi	r1, r1, SFS
 	mtlr	r0
 	blr
 
@@ -2366,12 +2402,13 @@ END_FTR_SECTION_IFSET(CPU_FTR_TM)
 	mfspr	r3, SPRN_DEC
 	mfspr	r4, SPRN_HDEC
 	mftb	r5
-	cmpw	r3, r4
+	extsw	r3, r3
+	EXTEND_HDEC(r4)
+	cmpd	r3, r4
 	ble	67f
 	mtspr	SPRN_DEC, r4
 67:
 	/* save expiry time of guest decrementer */
-	extsw	r3, r3
 	add	r3, r3, r5
 	ld	r4, HSTATE_KVM_VCPU(r13)
 	ld	r5, HSTATE_KVM_VCORE(r13)

arch/powerpc/perf/perf_regs.c

@@ -101,5 +101,6 @@ void perf_get_regs_user(struct perf_regs *regs_user,
 			struct pt_regs *regs_user_copy)
 {
 	regs_user->regs = task_pt_regs(current);
-	regs_user->abi  = perf_reg_abi(current);
+	regs_user->abi = (regs_user->regs) ? perf_reg_abi(current) :
+			 PERF_SAMPLE_REGS_ABI_NONE;
 }

arch/powerpc/platforms/powernv/npu-dma.c

@@ -449,7 +449,7 @@ static int mmio_launch_invalidate(struct npu *npu, unsigned long launch,
 	return mmio_atsd_reg;
 }
 
-static int mmio_invalidate_pid(struct npu *npu, unsigned long pid)
+static int mmio_invalidate_pid(struct npu *npu, unsigned long pid, bool flush)
 {
 	unsigned long launch;
 
@@ -465,12 +465,15 @@ static int mmio_invalidate_pid(struct npu *npu, unsigned long pid)
 	/* PID */
 	launch |= pid << PPC_BITLSHIFT(38);
 
+	/* No flush */
+	launch |= !flush << PPC_BITLSHIFT(39);
+
 	/* Invalidating the entire process doesn't use a va */
 	return mmio_launch_invalidate(npu, launch, 0);
 }
 
 static int mmio_invalidate_va(struct npu *npu, unsigned long va,
-			unsigned long pid)
+			unsigned long pid, bool flush)
 {
 	unsigned long launch;
 
@@ -486,26 +489,60 @@ static int mmio_invalidate_va(struct npu *npu, unsigned long va,
 	/* PID */
 	launch |= pid << PPC_BITLSHIFT(38);
 
+	/* No flush */
+	launch |= !flush << PPC_BITLSHIFT(39);
+
 	return mmio_launch_invalidate(npu, launch, va);
 }
 
 #define mn_to_npu_context(x) container_of(x, struct npu_context, mn)
 
+struct mmio_atsd_reg {
+	struct npu *npu;
+	int reg;
+};
+
+static void mmio_invalidate_wait(
+	struct mmio_atsd_reg mmio_atsd_reg[NV_MAX_NPUS], bool flush)
+{
+	struct npu *npu;
+	int i, reg;
+
+	/* Wait for all invalidations to complete */
+	for (i = 0; i <= max_npu2_index; i++) {
+		if (mmio_atsd_reg[i].reg < 0)
+			continue;
+
+		/* Wait for completion */
+		npu = mmio_atsd_reg[i].npu;
+		reg = mmio_atsd_reg[i].reg;
+		while (__raw_readq(npu->mmio_atsd_regs[reg] + XTS_ATSD_STAT))
+			cpu_relax();
+
+		put_mmio_atsd_reg(npu, reg);
+
+		/*
+		 * The GPU requires two flush ATSDs to ensure all entries have
+		 * been flushed. We use PID 0 as it will never be used for a
+		 * process on the GPU.
+		 */
+		if (flush)
+			mmio_invalidate_pid(npu, 0, true);
+	}
+}
+
 /*
  * Invalidate either a single address or an entire PID depending on
  * the value of va.
  */
 static void mmio_invalidate(struct npu_context *npu_context, int va,
-			unsigned long address)
+			unsigned long address, bool flush)
 {
-	int i, j, reg;
+	int i, j;
 	struct npu *npu;
 	struct pnv_phb *nphb;
 	struct pci_dev *npdev;
-	struct {
+	struct mmio_atsd_reg mmio_atsd_reg[NV_MAX_NPUS];
-		struct npu *npu;
-		int reg;
-	} mmio_atsd_reg[NV_MAX_NPUS];
 	unsigned long pid = npu_context->mm->context.id;
 
 	/*
@@ -525,10 +562,11 @@ static void mmio_invalidate(struct npu_context *npu_context, int va,
 
 			if (va)
 				mmio_atsd_reg[i].reg =
-					mmio_invalidate_va(npu, address, pid);
+					mmio_invalidate_va(npu, address, pid,
+							flush);
 			else
 				mmio_atsd_reg[i].reg =
-					mmio_invalidate_pid(npu, pid);
+					mmio_invalidate_pid(npu, pid, flush);
 
 			/*
 			 * The NPU hardware forwards the shootdown to all GPUs
@@ -544,18 +582,10 @@ static void mmio_invalidate(struct npu_context *npu_context, int va,
 	 */
 	flush_tlb_mm(npu_context->mm);
 
-	/* Wait for all invalidations to complete */
+	mmio_invalidate_wait(mmio_atsd_reg, flush);
-	for (i = 0; i <= max_npu2_index; i++) {
+	if (flush)
-		if (mmio_atsd_reg[i].reg < 0)
+		/* Wait for the flush to complete */
-			continue;
+		mmio_invalidate_wait(mmio_atsd_reg, false);
-
-		/* Wait for completion */
-		npu = mmio_atsd_reg[i].npu;
-		reg = mmio_atsd_reg[i].reg;
-		while (__raw_readq(npu->mmio_atsd_regs[reg] + XTS_ATSD_STAT))
-			cpu_relax();
-		put_mmio_atsd_reg(npu, reg);
-	}
 }
 
 static void pnv_npu2_mn_release(struct mmu_notifier *mn,
@@ -571,7 +601,7 @@ static void pnv_npu2_mn_release(struct mmu_notifier *mn,
 	 * There should be no more translation requests for this PID, but we
 	 * need to ensure any entries for it are removed from the TLB.
 	 */
-	mmio_invalidate(npu_context, 0, 0);
+	mmio_invalidate(npu_context, 0, 0, true);
 }
 
 static void pnv_npu2_mn_change_pte(struct mmu_notifier *mn,
@@ -581,7 +611,7 @@ static void pnv_npu2_mn_change_pte(struct mmu_notifier *mn,
 {
 	struct npu_context *npu_context = mn_to_npu_context(mn);
 
-	mmio_invalidate(npu_context, 1, address);
+	mmio_invalidate(npu_context, 1, address, true);
 }
 
 static void pnv_npu2_mn_invalidate_page(struct mmu_notifier *mn,
@@ -590,7 +620,7 @@ static void pnv_npu2_mn_invalidate_page(struct mmu_notifier *mn,
 {
 	struct npu_context *npu_context = mn_to_npu_context(mn);
 
-	mmio_invalidate(npu_context, 1, address);
+	mmio_invalidate(npu_context, 1, address, true);
 }
 
 static void pnv_npu2_mn_invalidate_range(struct mmu_notifier *mn,
@@ -600,8 +630,11 @@ static void pnv_npu2_mn_invalidate_range(struct mmu_notifier *mn,
 	struct npu_context *npu_context = mn_to_npu_context(mn);
 	unsigned long address;
 
-	for (address = start; address <= end; address += PAGE_SIZE)
+	for (address = start; address < end; address += PAGE_SIZE)
-		mmio_invalidate(npu_context, 1, address);
+		mmio_invalidate(npu_context, 1, address, false);
+
+	/* Do the flush only on the final addess == end */
+	mmio_invalidate(npu_context, 1, address, true);
 }
 
 static const struct mmu_notifier_ops nv_nmmu_notifier_ops = {
@@ -651,8 +684,11 @@ struct npu_context *pnv_npu2_init_context(struct pci_dev *gpdev,
 		/* No nvlink associated with this GPU device */
 		return ERR_PTR(-ENODEV);
 
-	if (!mm) {
+	if (!mm || mm->context.id == 0) {
-		/* kernel thread contexts are not supported */
+		/*
+		 * Kernel thread contexts are not supported and context id 0 is
+		 * reserved on the GPU.
+		 */
 		return ERR_PTR(-EINVAL);
 	}
 

arch/s390/include/asm/processor.h

@@ -221,11 +221,6 @@ extern void release_thread(struct task_struct *);
 /* Free guarded storage control block for current */
 void exit_thread_gs(void);
 
-/*
- * Return saved PC of a blocked thread.
- */
-extern unsigned long thread_saved_pc(struct task_struct *t);
-
 unsigned long get_wchan(struct task_struct *p);
 #define task_pt_regs(tsk) ((struct pt_regs *) \
         (task_stack_page(tsk) + THREAD_SIZE) - 1)

arch/s390/kernel/ipl.c

@@ -564,8 +564,6 @@ static struct kset *ipl_kset;
 
 static void __ipl_run(void *unused)
 {
-	if (MACHINE_IS_LPAR && ipl_info.type == IPL_TYPE_CCW)
-		diag308(DIAG308_LOAD_NORMAL_DUMP, NULL);
 	diag308(DIAG308_LOAD_CLEAR, NULL);
 	if (MACHINE_IS_VM)
 		__cpcmd("IPL", NULL, 0, NULL);
@@ -1088,10 +1086,7 @@ static void __reipl_run(void *unused)
 		break;
 	case REIPL_METHOD_CCW_DIAG:
 		diag308(DIAG308_SET, reipl_block_ccw);
-		if (MACHINE_IS_LPAR)
+		diag308(DIAG308_LOAD_CLEAR, NULL);
-			diag308(DIAG308_LOAD_NORMAL_DUMP, NULL);
-		else
-			diag308(DIAG308_LOAD_CLEAR, NULL);
 		break;
 	case REIPL_METHOD_FCP_RW_DIAG:
 		diag308(DIAG308_SET, reipl_block_fcp);

arch/s390/kernel/process.c

@@ -41,31 +41,6 @@
 
 asmlinkage void ret_from_fork(void) asm ("ret_from_fork");
 
-/*
- * Return saved PC of a blocked thread. used in kernel/sched.
- * resume in entry.S does not create a new stack frame, it
- * just stores the registers %r6-%r15 to the frame given by
- * schedule. We want to return the address of the caller of
- * schedule, so we have to walk the backchain one time to
- * find the frame schedule() store its return address.
- */
-unsigned long thread_saved_pc(struct task_struct *tsk)
-{
-	struct stack_frame *sf, *low, *high;
-
-	if (!tsk || !task_stack_page(tsk))
-		return 0;
-	low = task_stack_page(tsk);
-	high = (struct stack_frame *) task_pt_regs(tsk);
-	sf = (struct stack_frame *) tsk->thread.ksp;
-	if (sf <= low || sf > high)
-		return 0;
-	sf = (struct stack_frame *) sf->back_chain;
-	if (sf <= low || sf > high)
-		return 0;
-	return sf->gprs[8];
-}
-
 extern void kernel_thread_starter(void);
 
 /*

arch/s390/kvm/gaccess.c

@@ -977,11 +977,12 @@ static int kvm_s390_shadow_tables(struct gmap *sg, unsigned long saddr,
 	ptr = asce.origin * 4096;
 	if (asce.r) {
 		*fake = 1;
+		ptr = 0;
 		asce.dt = ASCE_TYPE_REGION1;
 	}
 	switch (asce.dt) {
 	case ASCE_TYPE_REGION1:
-		if (vaddr.rfx01 > asce.tl && !asce.r)
+		if (vaddr.rfx01 > asce.tl && !*fake)
 			return PGM_REGION_FIRST_TRANS;
 		break;
 	case ASCE_TYPE_REGION2:
@@ -1009,8 +1010,7 @@ static int kvm_s390_shadow_tables(struct gmap *sg, unsigned long saddr,
 		union region1_table_entry rfte;
 
 		if (*fake) {
-			/* offset in 16EB guest memory block */
+			ptr += (unsigned long) vaddr.rfx << 53;
-			ptr = ptr + ((unsigned long) vaddr.rsx << 53UL);
 			rfte.val = ptr;
 			goto shadow_r2t;
 		}
@@ -1036,8 +1036,7 @@ shadow_r2t:
 		union region2_table_entry rste;
 
 		if (*fake) {
-			/* offset in 8PB guest memory block */
+			ptr += (unsigned long) vaddr.rsx << 42;
-			ptr = ptr + ((unsigned long) vaddr.rtx << 42UL);
 			rste.val = ptr;
 			goto shadow_r3t;
 		}
@@ -1064,8 +1063,7 @@ shadow_r3t:
 		union region3_table_entry rtte;
 
 		if (*fake) {
-			/* offset in 4TB guest memory block */
+			ptr += (unsigned long) vaddr.rtx << 31;
-			ptr = ptr + ((unsigned long) vaddr.sx << 31UL);
 			rtte.val = ptr;
 			goto shadow_sgt;
 		}
@@ -1101,8 +1099,7 @@ shadow_sgt:
 		union segment_table_entry ste;
 
 		if (*fake) {
-			/* offset in 2G guest memory block */
+			ptr += (unsigned long) vaddr.sx << 20;
-			ptr = ptr + ((unsigned long) vaddr.sx << 20UL);
 			ste.val = ptr;
 			goto shadow_pgt;
 		}

arch/score/include/asm/processor.h

@@ -13,7 +13,6 @@ struct task_struct;
  */
 extern void (*cpu_wait)(void);
 
-extern unsigned long thread_saved_pc(struct task_struct *tsk);
 extern void start_thread(struct pt_regs *regs,
 			unsigned long pc, unsigned long sp);
 extern unsigned long get_wchan(struct task_struct *p);

arch/score/kernel/process.c

@@ -101,11 +101,6 @@ int dump_fpu(struct pt_regs *regs, elf_fpregset_t *r)
 	return 1;
 }
 
-unsigned long thread_saved_pc(struct task_struct *tsk)
-{
-	return task_pt_regs(tsk)->cp0_epc;
-}
-
 unsigned long get_wchan(struct task_struct *task)
 {
 	if (!task || task == current || task->state == TASK_RUNNING)

arch/sparc/include/asm/processor_32.h

@@ -67,9 +67,6 @@ struct thread_struct {
 	.current_ds = KERNEL_DS, \
 }
 
-/* Return saved PC of a blocked thread. */
-unsigned long thread_saved_pc(struct task_struct *t);
-
 /* Do necessary setup to start up a newly executed thread. */
 static inline void start_thread(struct pt_regs * regs, unsigned long pc,
 				    unsigned long sp)

arch/sparc/include/asm/processor_64.h

@@ -89,9 +89,7 @@ struct thread_struct {
 #include <linux/types.h>
 #include <asm/fpumacro.h>
 
-/* Return saved PC of a blocked thread. */
 struct task_struct;
-unsigned long thread_saved_pc(struct task_struct *);
 
 /* On Uniprocessor, even in RMO processes see TSO semantics */
 #ifdef CONFIG_SMP

arch/sparc/kernel/process_32.c

@@ -176,14 +176,6 @@ void show_stack(struct task_struct *tsk, unsigned long *_ksp)
 	printk("\n");
 }
 
-/*
- * Note: sparc64 has a pretty intricated thread_saved_pc, check it out.
- */
-unsigned long thread_saved_pc(struct task_struct *tsk)
-{
-	return task_thread_info(tsk)->kpc;
-}
-
 /*
  * Free current thread data structures etc..
  */

arch/sparc/kernel/process_64.c

@@ -400,25 +400,6 @@ core_initcall(sparc_sysrq_init);
 
 #endif
 
-unsigned long thread_saved_pc(struct task_struct *tsk)
-{
-	struct thread_info *ti = task_thread_info(tsk);
-	unsigned long ret = 0xdeadbeefUL;
-	
-	if (ti && ti->ksp) {
-		unsigned long *sp;
-		sp = (unsigned long *)(ti->ksp + STACK_BIAS);
-		if (((unsigned long)sp & (sizeof(long) - 1)) == 0UL &&
-		    sp[14]) {
-			unsigned long *fp;
-			fp = (unsigned long *)(sp[14] + STACK_BIAS);
-			if (((unsigned long)fp & (sizeof(long) - 1)) == 0UL)
-				ret = fp[15];
-		}
-	}
-	return ret;
-}
-
 /* Free current thread data structures etc.. */
 void exit_thread(struct task_struct *tsk)
 {

arch/tile/include/asm/processor.h

@@ -214,13 +214,6 @@ static inline void release_thread(struct task_struct *dead_task)
 
 extern void prepare_exit_to_usermode(struct pt_regs *regs, u32 flags);
 
-
-/*
- * Return saved (kernel) PC of a blocked thread.
- * Only used in a printk() in kernel/sched/core.c, so don't work too hard.
- */
-#define thread_saved_pc(t)   ((t)->thread.pc)
-
 unsigned long get_wchan(struct task_struct *p);
 
 /* Return initial ksp value for given task. */

arch/um/include/asm/processor-generic.h

@@ -58,8 +58,6 @@ static inline void release_thread(struct task_struct *task)
 {
 }
 
-extern unsigned long thread_saved_pc(struct task_struct *t);
-
 static inline void mm_copy_segments(struct mm_struct *from_mm,
 				    struct mm_struct *new_mm)
 {

arch/um/kernel/um_arch.c

@@ -56,12 +56,6 @@ union thread_union cpu0_irqstack
 	__attribute__((__section__(".data..init_irqstack"))) =
 		{ INIT_THREAD_INFO(init_task) };
 
-unsigned long thread_saved_pc(struct task_struct *task)
-{
-	/* FIXME: Need to look up userspace_pid by cpu */
-	return os_process_pc(userspace_pid[0]);
-}
-
 /* Changed in setup_arch, which is called in early boot */
 static char host_info[(__NEW_UTS_LEN + 1) * 5];
 

arch/x86/events/intel/core.c

@@ -431,11 +431,11 @@ static __initconst const u64 skl_hw_cache_event_ids
  [ C(DTLB) ] = {
 	[ C(OP_READ) ] = {
 		[ C(RESULT_ACCESS) ] = 0x81d0,	/* MEM_INST_RETIRED.ALL_LOADS */
-		[ C(RESULT_MISS)   ] = 0x608,	/* DTLB_LOAD_MISSES.WALK_COMPLETED */
+		[ C(RESULT_MISS)   ] = 0xe08,	/* DTLB_LOAD_MISSES.WALK_COMPLETED */
 	},
 	[ C(OP_WRITE) ] = {
 		[ C(RESULT_ACCESS) ] = 0x82d0,	/* MEM_INST_RETIRED.ALL_STORES */
-		[ C(RESULT_MISS)   ] = 0x649,	/* DTLB_STORE_MISSES.WALK_COMPLETED */
+		[ C(RESULT_MISS)   ] = 0xe49,	/* DTLB_STORE_MISSES.WALK_COMPLETED */
 	},
 	[ C(OP_PREFETCH) ] = {
 		[ C(RESULT_ACCESS) ] = 0x0,

arch/x86/include/asm/kvm_emulate.h

@@ -296,6 +296,7 @@ struct x86_emulate_ctxt {
 
 	bool perm_ok; /* do not check permissions if true */
 	bool ud;	/* inject an #UD if host doesn't support insn */
+	bool tf;	/* TF value before instruction (after for syscall/sysret) */
 
 	bool have_exception;
 	struct x86_exception exception;

arch/x86/include/asm/mshyperv.h

@@ -2,8 +2,7 @@
 #define _ASM_X86_MSHYPER_H
 
 #include <linux/types.h>
-#include <linux/interrupt.h>
+#include <linux/atomic.h>
-#include <linux/clocksource.h>
 #include <asm/hyperv.h>
 
 /*

arch/x86/include/asm/processor.h

@@ -860,8 +860,6 @@ extern unsigned long KSTK_ESP(struct task_struct *task);
 
 #endif /* CONFIG_X86_64 */
 
-extern unsigned long thread_saved_pc(struct task_struct *tsk);
-
 extern void start_thread(struct pt_regs *regs, unsigned long new_ip,
 					       unsigned long new_sp);
 

arch/x86/kernel/process.c

@@ -544,17 +544,6 @@ unsigned long arch_randomize_brk(struct mm_struct *mm)
 	return randomize_page(mm->brk, 0x02000000);
 }
 
-/*
- * Return saved PC of a blocked thread.
- * What is this good for? it will be always the scheduler or ret_from_fork.
- */
-unsigned long thread_saved_pc(struct task_struct *tsk)
-{
-	struct inactive_task_frame *frame =
-		(struct inactive_task_frame *) READ_ONCE(tsk->thread.sp);
-	return READ_ONCE_NOCHECK(frame->ret_addr);
-}
-
 /*
  * Called from fs/proc with a reference on @p to find the function
  * which called into schedule(). This needs to be done carefully

arch/x86/kvm/emulate.c

@@ -2742,6 +2742,7 @@ static int em_syscall(struct x86_emulate_ctxt *ctxt)
 		ctxt->eflags &= ~(X86_EFLAGS_VM | X86_EFLAGS_IF);
 	}
 
+	ctxt->tf = (ctxt->eflags & X86_EFLAGS_TF) != 0;
 	return X86EMUL_CONTINUE;
 }
 

arch/x86/kvm/x86.c

@@ -5313,6 +5313,8 @@ static void init_emulate_ctxt(struct kvm_vcpu *vcpu)
 	kvm_x86_ops->get_cs_db_l_bits(vcpu, &cs_db, &cs_l);
 
 	ctxt->eflags = kvm_get_rflags(vcpu);
+	ctxt->tf = (ctxt->eflags & X86_EFLAGS_TF) != 0;
+
 	ctxt->eip = kvm_rip_read(vcpu);
 	ctxt->mode = (!is_protmode(vcpu))		? X86EMUL_MODE_REAL :
 		     (ctxt->eflags & X86_EFLAGS_VM)	? X86EMUL_MODE_VM86 :
@@ -5528,36 +5530,25 @@ static int kvm_vcpu_check_hw_bp(unsigned long addr, u32 type, u32 dr7,
 	return dr6;
 }
 
-static void kvm_vcpu_check_singlestep(struct kvm_vcpu *vcpu, unsigned long rflags, int *r)
+static void kvm_vcpu_do_singlestep(struct kvm_vcpu *vcpu, int *r)
 {
 	struct kvm_run *kvm_run = vcpu->run;
 
-	/*
+	if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP) {
-	 * rflags is the old, "raw" value of the flags.  The new value has
+		kvm_run->debug.arch.dr6 = DR6_BS | DR6_FIXED_1 | DR6_RTM;
-	 * not been saved yet.
+		kvm_run->debug.arch.pc = vcpu->arch.singlestep_rip;
-	 *
+		kvm_run->debug.arch.exception = DB_VECTOR;
-	 * This is correct even for TF set by the guest, because "the
+		kvm_run->exit_reason = KVM_EXIT_DEBUG;
-	 * processor will not generate this exception after the instruction
+		*r = EMULATE_USER_EXIT;
-	 * that sets the TF flag".
+	} else {
-	 */
+		/*
-	if (unlikely(rflags & X86_EFLAGS_TF)) {
+		 * "Certain debug exceptions may clear bit 0-3.  The
-		if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP) {
+		 * remaining contents of the DR6 register are never
-			kvm_run->debug.arch.dr6 = DR6_BS | DR6_FIXED_1 |
+		 * cleared by the processor".
-						  DR6_RTM;
+		 */
-			kvm_run->debug.arch.pc = vcpu->arch.singlestep_rip;
+		vcpu->arch.dr6 &= ~15;
-			kvm_run->debug.arch.exception = DB_VECTOR;
+		vcpu->arch.dr6 |= DR6_BS | DR6_RTM;
-			kvm_run->exit_reason = KVM_EXIT_DEBUG;
+		kvm_queue_exception(vcpu, DB_VECTOR);
-			*r = EMULATE_USER_EXIT;
-		} else {
-			/*
-			 * "Certain debug exceptions may clear bit 0-3.  The
-			 * remaining contents of the DR6 register are never
-			 * cleared by the processor".
-			 */
-			vcpu->arch.dr6 &= ~15;
-			vcpu->arch.dr6 |= DR6_BS | DR6_RTM;
-			kvm_queue_exception(vcpu, DB_VECTOR);
-		}
 	}
 }
 
@@ -5567,7 +5558,17 @@ int kvm_skip_emulated_instruction(struct kvm_vcpu *vcpu)
 	int r = EMULATE_DONE;
 
 	kvm_x86_ops->skip_emulated_instruction(vcpu);
-	kvm_vcpu_check_singlestep(vcpu, rflags, &r);
+
+	/*
+	 * rflags is the old, "raw" value of the flags.  The new value has
+	 * not been saved yet.
+	 *
+	 * This is correct even for TF set by the guest, because "the
+	 * processor will not generate this exception after the instruction
+	 * that sets the TF flag".
+	 */
+	if (unlikely(rflags & X86_EFLAGS_TF))
+		kvm_vcpu_do_singlestep(vcpu, &r);
 	return r == EMULATE_DONE;
 }
 EXPORT_SYMBOL_GPL(kvm_skip_emulated_instruction);
@@ -5726,8 +5727,9 @@ restart:
 		toggle_interruptibility(vcpu, ctxt->interruptibility);
 		vcpu->arch.emulate_regs_need_sync_to_vcpu = false;
 		kvm_rip_write(vcpu, ctxt->eip);
-		if (r == EMULATE_DONE)
+		if (r == EMULATE_DONE &&
-			kvm_vcpu_check_singlestep(vcpu, rflags, &r);
+		    (ctxt->tf || (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP)))
+			kvm_vcpu_do_singlestep(vcpu, &r);
 		if (!ctxt->have_exception ||
 		    exception_type(ctxt->exception.vector) == EXCPT_TRAP)
 			__kvm_set_rflags(vcpu, ctxt->eflags);

arch/xtensa/include/asm/processor.h

@@ -213,8 +213,6 @@ struct mm_struct;
 #define release_segments(mm)	do { } while(0)
 #define forget_segments()	do { } while (0)
 
-#define thread_saved_pc(tsk)	(task_pt_regs(tsk)->pc)
-
 extern unsigned long get_wchan(struct task_struct *p);
 
 #define KSTK_EIP(tsk)		(task_pt_regs(tsk)->pc)

block/bio.c

@@ -240,20 +240,21 @@ fallback:
 	return bvl;
 }
 
-static void __bio_free(struct bio *bio)
+void bio_uninit(struct bio *bio)
 {
 	bio_disassociate_task(bio);
 
 	if (bio_integrity(bio))
 		bio_integrity_free(bio);
 }
+EXPORT_SYMBOL(bio_uninit);
 
 static void bio_free(struct bio *bio)
 {
 	struct bio_set *bs = bio->bi_pool;
 	void *p;
 
-	__bio_free(bio);
+	bio_uninit(bio);
 
 	if (bs) {
 		bvec_free(bs->bvec_pool, bio->bi_io_vec, BVEC_POOL_IDX(bio));
@@ -271,6 +272,11 @@ static void bio_free(struct bio *bio)
 	}
 }
 
+/*
+ * Users of this function have their own bio allocation. Subsequently,
+ * they must remember to pair any call to bio_init() with bio_uninit()
+ * when IO has completed, or when the bio is released.
+ */
 void bio_init(struct bio *bio, struct bio_vec *table,
 	      unsigned short max_vecs)
 {
@@ -297,7 +303,7 @@ void bio_reset(struct bio *bio)
 {
 	unsigned long flags = bio->bi_flags & (~0UL << BIO_RESET_BITS);
 
-	__bio_free(bio);
+	bio_uninit(bio);
 
 	memset(bio, 0, BIO_RESET_BYTES);
 	bio->bi_flags = flags;

block/blk-mq-sched.c

@@ -68,6 +68,45 @@ static void blk_mq_sched_assign_ioc(struct request_queue *q,
 		__blk_mq_sched_assign_ioc(q, rq, bio, ioc);
 }
 
+/*
+ * Mark a hardware queue as needing a restart. For shared queues, maintain
+ * a count of how many hardware queues are marked for restart.
+ */
+static void blk_mq_sched_mark_restart_hctx(struct blk_mq_hw_ctx *hctx)
+{
+	if (test_bit(BLK_MQ_S_SCHED_RESTART, &hctx->state))
+		return;
+
+	if (hctx->flags & BLK_MQ_F_TAG_SHARED) {
+		struct request_queue *q = hctx->queue;
+
+		if (!test_and_set_bit(BLK_MQ_S_SCHED_RESTART, &hctx->state))
+			atomic_inc(&q->shared_hctx_restart);
+	} else
+		set_bit(BLK_MQ_S_SCHED_RESTART, &hctx->state);
+}
+
+static bool blk_mq_sched_restart_hctx(struct blk_mq_hw_ctx *hctx)
+{
+	if (!test_bit(BLK_MQ_S_SCHED_RESTART, &hctx->state))
+		return false;
+
+	if (hctx->flags & BLK_MQ_F_TAG_SHARED) {
+		struct request_queue *q = hctx->queue;
+
+		if (test_and_clear_bit(BLK_MQ_S_SCHED_RESTART, &hctx->state))
+			atomic_dec(&q->shared_hctx_restart);
+	} else
+		clear_bit(BLK_MQ_S_SCHED_RESTART, &hctx->state);
+
+	if (blk_mq_hctx_has_pending(hctx)) {
+		blk_mq_run_hw_queue(hctx, true);
+		return true;
+	}
+
+	return false;
+}
+
 struct request *blk_mq_sched_get_request(struct request_queue *q,
 					 struct bio *bio,
 					 unsigned int op,
@@ -266,18 +305,6 @@ static bool blk_mq_sched_bypass_insert(struct blk_mq_hw_ctx *hctx,
 	return true;
 }
 
-static bool blk_mq_sched_restart_hctx(struct blk_mq_hw_ctx *hctx)
-{
-	if (test_bit(BLK_MQ_S_SCHED_RESTART, &hctx->state)) {
-		clear_bit(BLK_MQ_S_SCHED_RESTART, &hctx->state);
-		if (blk_mq_hctx_has_pending(hctx)) {
-			blk_mq_run_hw_queue(hctx, true);
-			return true;
-		}
-	}
-	return false;
-}
-
 /**
  * list_for_each_entry_rcu_rr - iterate in a round-robin fashion over rcu list
  * @pos:    loop cursor.
@@ -309,6 +336,13 @@ void blk_mq_sched_restart(struct blk_mq_hw_ctx *const hctx)
 	unsigned int i, j;
 
 	if (set->flags & BLK_MQ_F_TAG_SHARED) {
+		/*
+		 * If this is 0, then we know that no hardware queues
+		 * have RESTART marked. We're done.
+		 */
+		if (!atomic_read(&queue->shared_hctx_restart))
+			return;
+
 		rcu_read_lock();
 		list_for_each_entry_rcu_rr(q, queue, &set->tag_list,
 					   tag_set_list) {

block/blk-mq-sched.h

@@ -115,15 +115,6 @@ static inline bool blk_mq_sched_has_work(struct blk_mq_hw_ctx *hctx)
 	return false;
 }
 
-/*
- * Mark a hardware queue as needing a restart.
- */
-static inline void blk_mq_sched_mark_restart_hctx(struct blk_mq_hw_ctx *hctx)
-{
-	if (!test_bit(BLK_MQ_S_SCHED_RESTART, &hctx->state))
-		set_bit(BLK_MQ_S_SCHED_RESTART, &hctx->state);
-}
-
 static inline bool blk_mq_sched_needs_restart(struct blk_mq_hw_ctx *hctx)
 {
 	return test_bit(BLK_MQ_S_SCHED_RESTART, &hctx->state);

block/blk-mq.c

@@ -2103,20 +2103,30 @@ static void blk_mq_map_swqueue(struct request_queue *q,
 	}
 }
 
+/*
+ * Caller needs to ensure that we're either frozen/quiesced, or that
+ * the queue isn't live yet.
+ */
 static void queue_set_hctx_shared(struct request_queue *q, bool shared)
 {
 	struct blk_mq_hw_ctx *hctx;
 	int i;
 
 	queue_for_each_hw_ctx(q, hctx, i) {
-		if (shared)
+		if (shared) {
+			if (test_bit(BLK_MQ_S_SCHED_RESTART, &hctx->state))
+				atomic_inc(&q->shared_hctx_restart);
 			hctx->flags |= BLK_MQ_F_TAG_SHARED;
-		else
+		} else {
+			if (test_bit(BLK_MQ_S_SCHED_RESTART, &hctx->state))
+				atomic_dec(&q->shared_hctx_restart);
 			hctx->flags &= ~BLK_MQ_F_TAG_SHARED;
+		}
 	}
 }
 
-static void blk_mq_update_tag_set_depth(struct blk_mq_tag_set *set, bool shared)
+static void blk_mq_update_tag_set_depth(struct blk_mq_tag_set *set,
+					bool shared)
 {
 	struct request_queue *q;
 

drivers/acpi/scan.c

@@ -1428,6 +1428,37 @@ static void acpi_init_coherency(struct acpi_device *adev)
 	adev->flags.coherent_dma = cca;
 }
 
+static int acpi_check_spi_i2c_slave(struct acpi_resource *ares, void *data)
+{
+	bool *is_spi_i2c_slave_p = data;
+
+	if (ares->type != ACPI_RESOURCE_TYPE_SERIAL_BUS)
+		return 1;
+
+	/*
+	 * devices that are connected to UART still need to be enumerated to
+	 * platform bus
+	 */
+	if (ares->data.common_serial_bus.type != ACPI_RESOURCE_SERIAL_TYPE_UART)
+		*is_spi_i2c_slave_p = true;
+
+	 /* no need to do more checking */
+	return -1;
+}
+
+static bool acpi_is_spi_i2c_slave(struct acpi_device *device)
+{
+	struct list_head resource_list;
+	bool is_spi_i2c_slave = false;
+
+	INIT_LIST_HEAD(&resource_list);
+	acpi_dev_get_resources(device, &resource_list, acpi_check_spi_i2c_slave,
+			       &is_spi_i2c_slave);
+	acpi_dev_free_resource_list(&resource_list);
+
+	return is_spi_i2c_slave;
+}
+
 void acpi_init_device_object(struct acpi_device *device, acpi_handle handle,
 			     int type, unsigned long long sta)
 {
@@ -1443,6 +1474,7 @@ void acpi_init_device_object(struct acpi_device *device, acpi_handle handle,
 	acpi_bus_get_flags(device);
 	device->flags.match_driver = false;
 	device->flags.initialized = true;
+	device->flags.spi_i2c_slave = acpi_is_spi_i2c_slave(device);
 	acpi_device_clear_enumerated(device);
 	device_initialize(&device->dev);
 	dev_set_uevent_suppress(&device->dev, true);
@@ -1727,38 +1759,13 @@ static acpi_status acpi_bus_check_add(acpi_handle handle, u32 lvl_not_used,
 	return AE_OK;
 }
 
-static int acpi_check_spi_i2c_slave(struct acpi_resource *ares, void *data)
-{
-	bool *is_spi_i2c_slave_p = data;
-
-	if (ares->type != ACPI_RESOURCE_TYPE_SERIAL_BUS)
-		return 1;
-
-	/*
-	 * devices that are connected to UART still need to be enumerated to
-	 * platform bus
-	 */
-	if (ares->data.common_serial_bus.type != ACPI_RESOURCE_SERIAL_TYPE_UART)
-		*is_spi_i2c_slave_p = true;
-
-	 /* no need to do more checking */
-	return -1;
-}
-
 static void acpi_default_enumeration(struct acpi_device *device)
 {
-	struct list_head resource_list;
-	bool is_spi_i2c_slave = false;
-
 	/*
 	 * Do not enumerate SPI/I2C slaves as they will be enumerated by their
 	 * respective parents.
 	 */
-	INIT_LIST_HEAD(&resource_list);
+	if (!device->flags.spi_i2c_slave) {
-	acpi_dev_get_resources(device, &resource_list, acpi_check_spi_i2c_slave,
-			       &is_spi_i2c_slave);
-	acpi_dev_free_resource_list(&resource_list);
-	if (!is_spi_i2c_slave) {
 		acpi_create_platform_device(device, NULL);
 		acpi_device_set_enumerated(device);
 	} else {
@@ -1854,7 +1861,7 @@ static void acpi_bus_attach(struct acpi_device *device)
 		return;
 
 	device->flags.match_driver = true;
-	if (ret > 0) {
+	if (ret > 0 && !device->flags.spi_i2c_slave) {
 		acpi_device_set_enumerated(device);
 		goto ok;
 	}
@@ -1863,10 +1870,10 @@ static void acpi_bus_attach(struct acpi_device *device)
 	if (ret < 0)
 		return;
 
-	if (device->pnp.type.platform_id)
+	if (!device->pnp.type.platform_id && !device->flags.spi_i2c_slave)
-		acpi_default_enumeration(device);
-	else
 		acpi_device_set_enumerated(device);
+	else
+		acpi_default_enumeration(device);
 
  ok:
 	list_for_each_entry(child, &device->children, node)

drivers/block/xen-blkback/blkback.c

@@ -609,8 +609,6 @@ int xen_blkif_schedule(void *arg)
 	unsigned long timeout;
 	int ret;
 
-	xen_blkif_get(blkif);
-
 	set_freezable();
 	while (!kthread_should_stop()) {
 		if (try_to_freeze())
@@ -665,7 +663,6 @@ purge_gnt_list:
 		print_stats(ring);
 
 	ring->xenblkd = NULL;
-	xen_blkif_put(blkif);
 
 	return 0;
 }
@@ -1436,34 +1433,35 @@ static int dispatch_rw_block_io(struct xen_blkif_ring *ring,
 static void make_response(struct xen_blkif_ring *ring, u64 id,
 			  unsigned short op, int st)
 {
-	struct blkif_response  resp;
+	struct blkif_response *resp;
 	unsigned long     flags;
 	union blkif_back_rings *blk_rings;
 	int notify;
 
-	resp.id        = id;
-	resp.operation = op;
-	resp.status    = st;
-
 	spin_lock_irqsave(&ring->blk_ring_lock, flags);
 	blk_rings = &ring->blk_rings;
 	/* Place on the response ring for the relevant domain. */
 	switch (ring->blkif->blk_protocol) {
 	case BLKIF_PROTOCOL_NATIVE:
-		memcpy(RING_GET_RESPONSE(&blk_rings->native, blk_rings->native.rsp_prod_pvt),
+		resp = RING_GET_RESPONSE(&blk_rings->native,
-		       &resp, sizeof(resp));
+					 blk_rings->native.rsp_prod_pvt);
 		break;
 	case BLKIF_PROTOCOL_X86_32:
-		memcpy(RING_GET_RESPONSE(&blk_rings->x86_32, blk_rings->x86_32.rsp_prod_pvt),
+		resp = RING_GET_RESPONSE(&blk_rings->x86_32,
-		       &resp, sizeof(resp));
+					 blk_rings->x86_32.rsp_prod_pvt);
 		break;
 	case BLKIF_PROTOCOL_X86_64:
-		memcpy(RING_GET_RESPONSE(&blk_rings->x86_64, blk_rings->x86_64.rsp_prod_pvt),
+		resp = RING_GET_RESPONSE(&blk_rings->x86_64,
-		       &resp, sizeof(resp));
+					 blk_rings->x86_64.rsp_prod_pvt);
 		break;
 	default:
 		BUG();
 	}
+
+	resp->id        = id;
+	resp->operation = op;
+	resp->status    = st;
+
 	blk_rings->common.rsp_prod_pvt++;
 	RING_PUSH_RESPONSES_AND_CHECK_NOTIFY(&blk_rings->common, notify);
 	spin_unlock_irqrestore(&ring->blk_ring_lock, flags);

drivers/block/xen-blkback/common.h

@@ -75,9 +75,8 @@ extern unsigned int xenblk_max_queues;
 struct blkif_common_request {
 	char dummy;
 };
-struct blkif_common_response {
+
-	char dummy;
+/* i386 protocol version */
-};
 
 struct blkif_x86_32_request_rw {
 	uint8_t        nr_segments;  /* number of segments                   */
@@ -129,14 +128,6 @@ struct blkif_x86_32_request {
 	} u;
 } __attribute__((__packed__));
 
-/* i386 protocol version */
-#pragma pack(push, 4)
-struct blkif_x86_32_response {
-	uint64_t        id;              /* copied from request */
-	uint8_t         operation;       /* copied from request */
-	int16_t         status;          /* BLKIF_RSP_???       */
-};
-#pragma pack(pop)
 /* x86_64 protocol version */
 
 struct blkif_x86_64_request_rw {
@@ -193,18 +184,12 @@ struct blkif_x86_64_request {
 	} u;
 } __attribute__((__packed__));
 
-struct blkif_x86_64_response {
-	uint64_t       __attribute__((__aligned__(8))) id;
-	uint8_t         operation;       /* copied from request */
-	int16_t         status;          /* BLKIF_RSP_???       */
-};
-
 DEFINE_RING_TYPES(blkif_common, struct blkif_common_request,
-		  struct blkif_common_response);
+		  struct blkif_response);
 DEFINE_RING_TYPES(blkif_x86_32, struct blkif_x86_32_request,
-		  struct blkif_x86_32_response);
+		  struct blkif_response __packed);
 DEFINE_RING_TYPES(blkif_x86_64, struct blkif_x86_64_request,
-		  struct blkif_x86_64_response);
+		  struct blkif_response);
 
 union blkif_back_rings {
 	struct blkif_back_ring        native;
@@ -281,6 +266,7 @@ struct xen_blkif_ring {
 
 	wait_queue_head_t	wq;
 	atomic_t		inflight;
+	bool			active;
 	/* One thread per blkif ring. */
 	struct task_struct	*xenblkd;
 	unsigned int		waiting_reqs;

drivers/block/xen-blkback/xenbus.c

@@ -159,7 +159,7 @@ static int xen_blkif_alloc_rings(struct xen_blkif *blkif)
 		init_waitqueue_head(&ring->shutdown_wq);
 		ring->blkif = blkif;
 		ring->st_print = jiffies;
-		xen_blkif_get(blkif);
+		ring->active = true;
 	}
 
 	return 0;
@@ -249,10 +249,12 @@ static int xen_blkif_disconnect(struct xen_blkif *blkif)
 		struct xen_blkif_ring *ring = &blkif->rings[r];
 		unsigned int i = 0;
 
+		if (!ring->active)
+			continue;
+
 		if (ring->xenblkd) {
 			kthread_stop(ring->xenblkd);
 			wake_up(&ring->shutdown_wq);
-			ring->xenblkd = NULL;
 		}
 
 		/* The above kthread_stop() guarantees that at this point we
@@ -296,7 +298,7 @@ static int xen_blkif_disconnect(struct xen_blkif *blkif)
 		BUG_ON(ring->free_pages_num != 0);
 		BUG_ON(ring->persistent_gnt_c != 0);
 		WARN_ON(i != (XEN_BLKIF_REQS_PER_PAGE * blkif->nr_ring_pages));
-		xen_blkif_put(blkif);
+		ring->active = false;
 	}
 	blkif->nr_ring_pages = 0;
 	/*
@@ -312,9 +314,10 @@ static int xen_blkif_disconnect(struct xen_blkif *blkif)
 
 static void xen_blkif_free(struct xen_blkif *blkif)
 {
-
+	WARN_ON(xen_blkif_disconnect(blkif));
-	xen_blkif_disconnect(blkif);
 	xen_vbd_free(&blkif->vbd);
+	kfree(blkif->be->mode);
+	kfree(blkif->be);
 
 	/* Make sure everything is drained before shutting down */
 	kmem_cache_free(xen_blkif_cachep, blkif);
@@ -511,8 +514,6 @@ static int xen_blkbk_remove(struct xenbus_device *dev)
 		xen_blkif_put(be->blkif);
 	}
 
-	kfree(be->mode);
-	kfree(be);
 	return 0;
 }
 

drivers/char/random.c

@@ -803,13 +803,13 @@ static int crng_fast_load(const char *cp, size_t len)
 		p[crng_init_cnt % CHACHA20_KEY_SIZE] ^= *cp;
 		cp++; crng_init_cnt++; len--;
 	}
+	spin_unlock_irqrestore(&primary_crng.lock, flags);
 	if (crng_init_cnt >= CRNG_INIT_CNT_THRESH) {
 		invalidate_batched_entropy();
 		crng_init = 1;
 		wake_up_interruptible(&crng_init_wait);
 		pr_notice("random: fast init done\n");
 	}
-	spin_unlock_irqrestore(&primary_crng.lock, flags);
 	return 1;
 }
 
@@ -841,6 +841,7 @@ static void crng_reseed(struct crng_state *crng, struct entropy_store *r)
 	}
 	memzero_explicit(&buf, sizeof(buf));
 	crng->init_time = jiffies;
+	spin_unlock_irqrestore(&primary_crng.lock, flags);
 	if (crng == &primary_crng && crng_init < 2) {
 		invalidate_batched_entropy();
 		crng_init = 2;
@@ -848,7 +849,6 @@ static void crng_reseed(struct crng_state *crng, struct entropy_store *r)
 		wake_up_interruptible(&crng_init_wait);
 		pr_notice("random: crng init done\n");
 	}
-	spin_unlock_irqrestore(&primary_crng.lock, flags);
 }
 
 static inline void crng_wait_ready(void)
@@ -2041,8 +2041,8 @@ static DEFINE_PER_CPU(struct batched_entropy, batched_entropy_u64);
 u64 get_random_u64(void)
 {
 	u64 ret;
-	bool use_lock = crng_init < 2;
+	bool use_lock = READ_ONCE(crng_init) < 2;
-	unsigned long flags;
+	unsigned long flags = 0;
 	struct batched_entropy *batch;
 
 #if BITS_PER_LONG == 64
@@ -2073,8 +2073,8 @@ static DEFINE_PER_CPU(struct batched_entropy, batched_entropy_u32);
 u32 get_random_u32(void)
 {
 	u32 ret;
-	bool use_lock = crng_init < 2;
+	bool use_lock = READ_ONCE(crng_init) < 2;
-	unsigned long flags;
+	unsigned long flags = 0;
 	struct batched_entropy *batch;
 
 	if (arch_get_random_int(&ret))

drivers/clocksource/arm_arch_timer.c

@@ -1209,9 +1209,9 @@ arch_timer_mem_frame_get_cntfrq(struct arch_timer_mem_frame *frame)
 		return 0;
 	}
 
-	rate = readl_relaxed(frame + CNTFRQ);
+	rate = readl_relaxed(base + CNTFRQ);
 
-	iounmap(frame);
+	iounmap(base);
 
 	return rate;
 }

drivers/clocksource/cadence_ttc_timer.c

@@ -18,6 +18,7 @@
 #include <linux/clk.h>
 #include <linux/interrupt.h>
 #include <linux/clockchips.h>
+#include <linux/clocksource.h>
 #include <linux/of_address.h>
 #include <linux/of_irq.h>
 #include <linux/slab.h>

drivers/clocksource/timer-sun5i.c

@@ -12,6 +12,7 @@
 
 #include <linux/clk.h>
 #include <linux/clockchips.h>
+#include <linux/clocksource.h>
 #include <linux/delay.h>
 #include <linux/interrupt.h>
 #include <linux/irq.h>

drivers/gpio/gpio-mvebu.c

@@ -721,7 +721,7 @@ static int mvebu_pwm_probe(struct platform_device *pdev,
 	u32 set;
 
 	if (!of_device_is_compatible(mvchip->chip.of_node,
-				     "marvell,armada-370-xp-gpio"))
+				     "marvell,armada-370-gpio"))
 		return 0;
 
 	if (IS_ERR(mvchip->clk))
@@ -852,7 +852,7 @@ static const struct of_device_id mvebu_gpio_of_match[] = {
 		.data	    = (void *) MVEBU_GPIO_SOC_VARIANT_ARMADAXP,
 	},
 	{
-		.compatible = "marvell,armada-370-xp-gpio",
+		.compatible = "marvell,armada-370-gpio",
 		.data	    = (void *) MVEBU_GPIO_SOC_VARIANT_ORION,
 	},
 	{
@@ -1128,7 +1128,7 @@ static int mvebu_gpio_probe(struct platform_device *pdev)
 						 mvchip);
 	}
 
-	/* Armada 370/XP has simple PWM support for GPIO lines */
+	/* Some MVEBU SoCs have simple PWM support for GPIO lines */
 	if (IS_ENABLED(CONFIG_PWM))
 		return mvebu_pwm_probe(pdev, mvchip, id);
 

drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c

@@ -693,6 +693,10 @@ int amdgpu_atombios_get_clock_info(struct amdgpu_device *adev)
 			DRM_INFO("Changing default dispclk from %dMhz to 600Mhz\n",
 				 adev->clock.default_dispclk / 100);
 			adev->clock.default_dispclk = 60000;
+		} else if (adev->clock.default_dispclk <= 60000) {
+			DRM_INFO("Changing default dispclk from %dMhz to 625Mhz\n",
+				 adev->clock.default_dispclk / 100);
+			adev->clock.default_dispclk = 62500;
 		}
 		adev->clock.dp_extclk =
 			le16_to_cpu(firmware_info->info_21.usUniphyDPModeExtClkFreq);

drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c

@@ -449,6 +449,7 @@ static const struct pci_device_id pciidlist[] = {
 	{0x1002, 0x6986, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_POLARIS12},
 	{0x1002, 0x6987, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_POLARIS12},
 	{0x1002, 0x6995, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_POLARIS12},
+	{0x1002, 0x6997, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_POLARIS12},
 	{0x1002, 0x699F, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_POLARIS12},
 	/* Vega 10 */
 	{0x1002, 0x6860, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA10|AMD_EXP_HW_SUPPORT},

drivers/gpu/drm/amd/amdgpu/atombios_crtc.c

@@ -165,7 +165,7 @@ void amdgpu_atombios_crtc_powergate(struct drm_crtc *crtc, int state)
 	struct drm_device *dev = crtc->dev;
 	struct amdgpu_device *adev = dev->dev_private;
 	int index = GetIndexIntoMasterTable(COMMAND, EnableDispPowerGating);
-	ENABLE_DISP_POWER_GATING_PARAMETERS_V2_1 args;
+	ENABLE_DISP_POWER_GATING_PS_ALLOCATION args;
 
 	memset(&args, 0, sizeof(args));
 
@@ -178,7 +178,7 @@ void amdgpu_atombios_crtc_powergate(struct drm_crtc *crtc, int state)
 void amdgpu_atombios_crtc_powergate_init(struct amdgpu_device *adev)
 {
 	int index = GetIndexIntoMasterTable(COMMAND, EnableDispPowerGating);
-	ENABLE_DISP_POWER_GATING_PARAMETERS_V2_1 args;
+	ENABLE_DISP_POWER_GATING_PS_ALLOCATION args;
 
 	memset(&args, 0, sizeof(args));
 

drivers/gpu/drm/drm_connector.c

@@ -1229,21 +1229,6 @@ int drm_mode_getconnector(struct drm_device *dev, void *data,
 	if (!connector)
 		return -ENOENT;
 
-	drm_modeset_lock(&dev->mode_config.connection_mutex, NULL);
-	encoder = drm_connector_get_encoder(connector);
-	if (encoder)
-		out_resp->encoder_id = encoder->base.id;
-	else
-		out_resp->encoder_id = 0;
-
-	ret = drm_mode_object_get_properties(&connector->base, file_priv->atomic,
-			(uint32_t __user *)(unsigned long)(out_resp->props_ptr),
-			(uint64_t __user *)(unsigned long)(out_resp->prop_values_ptr),
-			&out_resp->count_props);
-	drm_modeset_unlock(&dev->mode_config.connection_mutex);
-	if (ret)
-		goto out_unref;
-
 	for (i = 0; i < DRM_CONNECTOR_MAX_ENCODER; i++)
 		if (connector->encoder_ids[i] != 0)
 			encoders_count++;
@@ -1256,7 +1241,7 @@ int drm_mode_getconnector(struct drm_device *dev, void *data,
 				if (put_user(connector->encoder_ids[i],
 					     encoder_ptr + copied)) {
 					ret = -EFAULT;
-					goto out_unref;
+					goto out;
 				}
 				copied++;
 			}
@@ -1300,15 +1285,32 @@ int drm_mode_getconnector(struct drm_device *dev, void *data,
 			if (copy_to_user(mode_ptr + copied,
 					 &u_mode, sizeof(u_mode))) {
 				ret = -EFAULT;
+				mutex_unlock(&dev->mode_config.mutex);
+
 				goto out;
 			}
 			copied++;
 		}
 	}
 	out_resp->count_modes = mode_count;
-out:
 	mutex_unlock(&dev->mode_config.mutex);
-out_unref:
+
+	drm_modeset_lock(&dev->mode_config.connection_mutex, NULL);
+	encoder = drm_connector_get_encoder(connector);
+	if (encoder)
+		out_resp->encoder_id = encoder->base.id;
+	else
+		out_resp->encoder_id = 0;
+
+	/* Only grab properties after probing, to make sure EDID and other
+	 * properties reflect the latest status. */
+	ret = drm_mode_object_get_properties(&connector->base, file_priv->atomic,
+			(uint32_t __user *)(unsigned long)(out_resp->props_ptr),
+			(uint64_t __user *)(unsigned long)(out_resp->prop_values_ptr),
+			&out_resp->count_props);
+	drm_modeset_unlock(&dev->mode_config.connection_mutex);
+
+out:
 	drm_connector_put(connector);
 
 	return ret;

drivers/gpu/drm/etnaviv/etnaviv_gem.h

@@ -106,9 +106,10 @@ struct etnaviv_gem_submit {
 	struct etnaviv_gpu *gpu;
 	struct ww_acquire_ctx ticket;
 	struct dma_fence *fence;
+	u32 flags;
 	unsigned int nr_bos;
 	struct etnaviv_gem_submit_bo bos[0];
-	u32 flags;
+	/* No new members here, the previous one is variable-length! */
 };
 
 int etnaviv_gem_wait_bo(struct etnaviv_gpu *gpu, struct drm_gem_object *obj,

drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c

@@ -172,7 +172,7 @@ static int submit_fence_sync(const struct etnaviv_gem_submit *submit)
 	for (i = 0; i < submit->nr_bos; i++) {
 		struct etnaviv_gem_object *etnaviv_obj = submit->bos[i].obj;
 		bool write = submit->bos[i].flags & ETNA_SUBMIT_BO_WRITE;
-		bool explicit = !(submit->flags & ETNA_SUBMIT_NO_IMPLICIT);
+		bool explicit = !!(submit->flags & ETNA_SUBMIT_NO_IMPLICIT);
 
 		ret = etnaviv_gpu_fence_sync_obj(etnaviv_obj, context, write,
 						 explicit);

drivers/gpu/drm/i915/i915_debugfs.c

@@ -292,6 +292,8 @@ static int per_file_stats(int id, void *ptr, void *data)
 	struct file_stats *stats = data;
 	struct i915_vma *vma;
 
+	lockdep_assert_held(&obj->base.dev->struct_mutex);
+
 	stats->count++;
 	stats->total += obj->base.size;
 	if (!obj->bind_count)
@@ -476,6 +478,8 @@ static int i915_gem_object_info(struct seq_file *m, void *data)
 		struct drm_i915_gem_request *request;
 		struct task_struct *task;
 
+		mutex_lock(&dev->struct_mutex);
+
 		memset(&stats, 0, sizeof(stats));
 		stats.file_priv = file->driver_priv;
 		spin_lock(&file->table_lock);
@@ -487,7 +491,6 @@ static int i915_gem_object_info(struct seq_file *m, void *data)
 		 * still alive (e.g. get_pid(current) => fork() => exit()).
 		 * Therefore, we need to protect this ->comm access using RCU.
 		 */
-		mutex_lock(&dev->struct_mutex);
 		request = list_first_entry_or_null(&file_priv->mm.request_list,
 						   struct drm_i915_gem_request,
 						   client_link);
@@ -497,6 +500,7 @@ static int i915_gem_object_info(struct seq_file *m, void *data)
 				PIDTYPE_PID);
 		print_file_stats(m, task ? task->comm : "<unknown>", stats);
 		rcu_read_unlock();
+
 		mutex_unlock(&dev->struct_mutex);
 	}
 	mutex_unlock(&dev->filelist_mutex);

drivers/gpu/drm/i915/i915_gem.c

@@ -2285,8 +2285,8 @@ i915_gem_object_get_pages_gtt(struct drm_i915_gem_object *obj)
 	struct page *page;
 	unsigned long last_pfn = 0;	/* suppress gcc warning */
 	unsigned int max_segment;
+	gfp_t noreclaim;
 	int ret;
-	gfp_t gfp;
 
 	/* Assert that the object is not currently in any GPU domain. As it
 	 * wasn't in the GTT, there shouldn't be any way it could have been in
@@ -2315,22 +2315,31 @@ rebuild_st:
 	 * Fail silently without starting the shrinker
 	 */
 	mapping = obj->base.filp->f_mapping;
-	gfp = mapping_gfp_constraint(mapping, ~(__GFP_IO | __GFP_RECLAIM));
+	noreclaim = mapping_gfp_constraint(mapping,
-	gfp |= __GFP_NORETRY | __GFP_NOWARN;
+					   ~(__GFP_IO | __GFP_RECLAIM));
+	noreclaim |= __GFP_NORETRY | __GFP_NOWARN;
+
 	sg = st->sgl;
 	st->nents = 0;
 	for (i = 0; i < page_count; i++) {
-		page = shmem_read_mapping_page_gfp(mapping, i, gfp);
+		const unsigned int shrink[] = {
-		if (unlikely(IS_ERR(page))) {
+			I915_SHRINK_BOUND | I915_SHRINK_UNBOUND | I915_SHRINK_PURGEABLE,
-			i915_gem_shrink(dev_priv,
+			0,
-					page_count,
+		}, *s = shrink;
-					I915_SHRINK_BOUND |
+		gfp_t gfp = noreclaim;
-					I915_SHRINK_UNBOUND |
+
-					I915_SHRINK_PURGEABLE);
+		do {
 			page = shmem_read_mapping_page_gfp(mapping, i, gfp);
-		}
+			if (likely(!IS_ERR(page)))
-		if (unlikely(IS_ERR(page))) {
+				break;
-			gfp_t reclaim;
+
+			if (!*s) {
+				ret = PTR_ERR(page);
+				goto err_sg;
+			}
+
+			i915_gem_shrink(dev_priv, 2 * page_count, *s++);
+			cond_resched();
 
 			/* We've tried hard to allocate the memory by reaping
 			 * our own buffer, now let the real VM do its job and
@@ -2340,15 +2349,26 @@ rebuild_st:
 			 * defer the oom here by reporting the ENOMEM back
 			 * to userspace.
 			 */
-			reclaim = mapping_gfp_mask(mapping);
+			if (!*s) {
-			reclaim |= __GFP_NORETRY; /* reclaim, but no oom */
+				/* reclaim and warn, but no oom */
+				gfp = mapping_gfp_mask(mapping);
 
-			page = shmem_read_mapping_page_gfp(mapping, i, reclaim);
+				/* Our bo are always dirty and so we require
-			if (IS_ERR(page)) {
+				 * kswapd to reclaim our pages (direct reclaim
-				ret = PTR_ERR(page);
+				 * does not effectively begin pageout of our
-				goto err_sg;
+				 * buffers on its own). However, direct reclaim
+				 * only waits for kswapd when under allocation
+				 * congestion. So as a result __GFP_RECLAIM is
+				 * unreliable and fails to actually reclaim our
+				 * dirty pages -- unless you try over and over
+				 * again with !__GFP_NORETRY. However, we still
+				 * want to fail this allocation rather than
+				 * trigger the out-of-memory killer and for
+				 * this we want the future __GFP_MAYFAIL.
+				 */
 			}
-		}
+		} while (1);
+
 		if (!i ||
 		    sg->length >= max_segment ||
 		    page_to_pfn(page) != last_pfn + 1) {
@@ -4222,6 +4242,7 @@ i915_gem_object_create(struct drm_i915_private *dev_priv, u64 size)
 
 	mapping = obj->base.filp->f_mapping;
 	mapping_set_gfp_mask(mapping, mask);
+	GEM_BUG_ON(!(mapping_gfp_mask(mapping) & __GFP_RECLAIM));
 
 	i915_gem_object_init(obj, &i915_gem_object_ops);
 

drivers/gpu/drm/i915/i915_gem_execbuffer.c

@@ -546,11 +546,12 @@ repeat:
 }
 
 static int
-i915_gem_execbuffer_relocate_entry(struct drm_i915_gem_object *obj,
+i915_gem_execbuffer_relocate_entry(struct i915_vma *vma,
 				   struct eb_vmas *eb,
 				   struct drm_i915_gem_relocation_entry *reloc,
 				   struct reloc_cache *cache)
 {
+	struct drm_i915_gem_object *obj = vma->obj;
 	struct drm_i915_private *dev_priv = to_i915(obj->base.dev);
 	struct drm_gem_object *target_obj;
 	struct drm_i915_gem_object *target_i915_obj;
@@ -628,6 +629,16 @@ i915_gem_execbuffer_relocate_entry(struct drm_i915_gem_object *obj,
 		return -EINVAL;
 	}
 
+	/*
+	 * If we write into the object, we need to force the synchronisation
+	 * barrier, either with an asynchronous clflush or if we executed the
+	 * patching using the GPU (though that should be serialised by the
+	 * timeline). To be completely sure, and since we are required to
+	 * do relocations we are already stalling, disable the user's opt
+	 * of our synchronisation.
+	 */
+	vma->exec_entry->flags &= ~EXEC_OBJECT_ASYNC;
+
 	ret = relocate_entry(obj, reloc, cache, target_offset);
 	if (ret)
 		return ret;
@@ -678,7 +689,7 @@ i915_gem_execbuffer_relocate_vma(struct i915_vma *vma,
 		do {
 			u64 offset = r->presumed_offset;
 
-			ret = i915_gem_execbuffer_relocate_entry(vma->obj, eb, r, &cache);
+			ret = i915_gem_execbuffer_relocate_entry(vma, eb, r, &cache);
 			if (ret)
 				goto out;
 
@@ -726,7 +737,7 @@ i915_gem_execbuffer_relocate_vma_slow(struct i915_vma *vma,
 
 	reloc_cache_init(&cache, eb->i915);
 	for (i = 0; i < entry->relocation_count; i++) {
-		ret = i915_gem_execbuffer_relocate_entry(vma->obj, eb, &relocs[i], &cache);
+		ret = i915_gem_execbuffer_relocate_entry(vma, eb, &relocs[i], &cache);
 		if (ret)
 			break;
 	}

drivers/gpu/drm/i915/i915_gem_request.c

@@ -623,7 +623,7 @@ i915_gem_request_alloc(struct intel_engine_cs *engine,
 	 * GPU processing the request, we never over-estimate the
 	 * position of the head.
 	 */
-	req->head = req->ring->tail;
+	req->head = req->ring->emit;
 
 	/* Check that we didn't interrupt ourselves with a new request */
 	GEM_BUG_ON(req->timeline->seqno != req->fence.seqno);

drivers/gpu/drm/i915/i915_guc_submission.c

@@ -480,9 +480,7 @@ static void guc_wq_item_append(struct i915_guc_client *client,
 	GEM_BUG_ON(freespace < wqi_size);
 
 	/* The GuC firmware wants the tail index in QWords, not bytes */
-	tail = rq->tail;
+	tail = intel_ring_set_tail(rq->ring, rq->tail) >> 3;
-	assert_ring_tail_valid(rq->ring, rq->tail);
-	tail >>= 3;
 	GEM_BUG_ON(tail > WQ_RING_TAIL_MAX);
 
 	/* For now workqueue item is 4 DWs; workqueue buffer is 2 pages. So we

drivers/gpu/drm/i915/i915_vma.c

@@ -650,6 +650,11 @@ int i915_vma_unbind(struct i915_vma *vma)
 				break;
 		}
 
+		if (!ret) {
+			ret = i915_gem_active_retire(&vma->last_fence,
+						     &vma->vm->i915->drm.struct_mutex);
+		}
+
 		__i915_vma_unpin(vma);
 		if (ret)
 			return ret;

drivers/gpu/drm/i915/intel_display.c

@@ -120,7 +120,8 @@ static void intel_crtc_init_scalers(struct intel_crtc *crtc,
 static void skylake_pfit_enable(struct intel_crtc *crtc);
 static void ironlake_pfit_disable(struct intel_crtc *crtc, bool force);
 static void ironlake_pfit_enable(struct intel_crtc *crtc);
-static void intel_modeset_setup_hw_state(struct drm_device *dev);
+static void intel_modeset_setup_hw_state(struct drm_device *dev,
+					 struct drm_modeset_acquire_ctx *ctx);
 static void intel_pre_disable_primary_noatomic(struct drm_crtc *crtc);
 
 struct intel_limit {
@@ -3449,7 +3450,7 @@ __intel_display_resume(struct drm_device *dev,
 	struct drm_crtc *crtc;
 	int i, ret;
 
-	intel_modeset_setup_hw_state(dev);
+	intel_modeset_setup_hw_state(dev, ctx);
 	i915_redisable_vga(to_i915(dev));
 
 	if (!state)
@@ -5825,7 +5826,8 @@ static void i9xx_crtc_disable(struct intel_crtc_state *old_crtc_state,
 		intel_update_watermarks(intel_crtc);
 }
 
-static void intel_crtc_disable_noatomic(struct drm_crtc *crtc)
+static void intel_crtc_disable_noatomic(struct drm_crtc *crtc,
+					struct drm_modeset_acquire_ctx *ctx)
 {
 	struct intel_encoder *encoder;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
@@ -5855,7 +5857,7 @@ static void intel_crtc_disable_noatomic(struct drm_crtc *crtc)
 		return;
 	}
 
-	state->acquire_ctx = crtc->dev->mode_config.acquire_ctx;
+	state->acquire_ctx = ctx;
 
 	/* Everything's already locked, -EDEADLK can't happen. */
 	crtc_state = intel_atomic_get_crtc_state(state, intel_crtc);
@@ -15030,7 +15032,7 @@ int intel_modeset_init(struct drm_device *dev)
 	intel_setup_outputs(dev_priv);
 
 	drm_modeset_lock_all(dev);
-	intel_modeset_setup_hw_state(dev);
+	intel_modeset_setup_hw_state(dev, dev->mode_config.acquire_ctx);
 	drm_modeset_unlock_all(dev);
 
 	for_each_intel_crtc(dev, crtc) {
@@ -15067,13 +15069,13 @@ int intel_modeset_init(struct drm_device *dev)
 	return 0;
 }
 
-static void intel_enable_pipe_a(struct drm_device *dev)
+static void intel_enable_pipe_a(struct drm_device *dev,
+				struct drm_modeset_acquire_ctx *ctx)
 {
 	struct intel_connector *connector;
 	struct drm_connector_list_iter conn_iter;
 	struct drm_connector *crt = NULL;
 	struct intel_load_detect_pipe load_detect_temp;
-	struct drm_modeset_acquire_ctx *ctx = dev->mode_config.acquire_ctx;
 	int ret;
 
 	/* We can't just switch on the pipe A, we need to set things up with a
@@ -15145,7 +15147,8 @@ static bool has_pch_trancoder(struct drm_i915_private *dev_priv,
 		(HAS_PCH_LPT_H(dev_priv) && pch_transcoder == TRANSCODER_A);
 }
 
-static void intel_sanitize_crtc(struct intel_crtc *crtc)
+static void intel_sanitize_crtc(struct intel_crtc *crtc,
+				struct drm_modeset_acquire_ctx *ctx)
 {
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = to_i915(dev);
@@ -15191,7 +15194,7 @@ static void intel_sanitize_crtc(struct intel_crtc *crtc)
 		plane = crtc->plane;
 		crtc->base.primary->state->visible = true;
 		crtc->plane = !plane;
-		intel_crtc_disable_noatomic(&crtc->base);
+		intel_crtc_disable_noatomic(&crtc->base, ctx);
 		crtc->plane = plane;
 	}
 
@@ -15201,13 +15204,13 @@ static void intel_sanitize_crtc(struct intel_crtc *crtc)
 		 * resume. Force-enable the pipe to fix this, the update_dpms
 		 * call below we restore the pipe to the right state, but leave
 		 * the required bits on. */
-		intel_enable_pipe_a(dev);
+		intel_enable_pipe_a(dev, ctx);
 	}
 
 	/* Adjust the state of the output pipe according to whether we
 	 * have active connectors/encoders. */
 	if (crtc->active && !intel_crtc_has_encoders(crtc))
-		intel_crtc_disable_noatomic(&crtc->base);
+		intel_crtc_disable_noatomic(&crtc->base, ctx);
 
 	if (crtc->active || HAS_GMCH_DISPLAY(dev_priv)) {
 		/*
@@ -15505,7 +15508,8 @@ get_encoder_power_domains(struct drm_i915_private *dev_priv)
  * and sanitizes it to the current state
  */
 static void
-intel_modeset_setup_hw_state(struct drm_device *dev)
+intel_modeset_setup_hw_state(struct drm_device *dev,
+			     struct drm_modeset_acquire_ctx *ctx)
 {
 	struct drm_i915_private *dev_priv = to_i915(dev);
 	enum pipe pipe;
@@ -15525,7 +15529,7 @@ intel_modeset_setup_hw_state(struct drm_device *dev)
 	for_each_pipe(dev_priv, pipe) {
 		crtc = intel_get_crtc_for_pipe(dev_priv, pipe);
 
-		intel_sanitize_crtc(crtc);
+		intel_sanitize_crtc(crtc, ctx);
 		intel_dump_pipe_config(crtc, crtc->config,
 				       "[setup_hw_state]");
 	}

drivers/gpu/drm/i915/intel_dp_aux_backlight.c

@@ -119,8 +119,6 @@ static int intel_dp_aux_setup_backlight(struct intel_connector *connector,
 	struct intel_dp *intel_dp = enc_to_intel_dp(&connector->encoder->base);
 	struct intel_panel *panel = &connector->panel;
 
-	intel_dp_aux_enable_backlight(connector);
-
 	if (intel_dp->edp_dpcd[2] & DP_EDP_BACKLIGHT_BRIGHTNESS_BYTE_COUNT)
 		panel->backlight.max = 0xFFFF;
 	else

drivers/gpu/drm/i915/intel_lrc.c

@@ -326,8 +326,7 @@ static u64 execlists_update_context(struct drm_i915_gem_request *rq)
 		rq->ctx->ppgtt ?: rq->i915->mm.aliasing_ppgtt;
 	u32 *reg_state = ce->lrc_reg_state;
 
-	assert_ring_tail_valid(rq->ring, rq->tail);
+	reg_state[CTX_RING_TAIL+1] = intel_ring_set_tail(rq->ring, rq->tail);
-	reg_state[CTX_RING_TAIL+1] = rq->tail;
 
 	/* True 32b PPGTT with dynamic page allocation: update PDP
 	 * registers and point the unallocated PDPs to scratch page.
@@ -2036,8 +2035,7 @@ void intel_lr_context_resume(struct drm_i915_private *dev_priv)
 			ce->state->obj->mm.dirty = true;
 			i915_gem_object_unpin_map(ce->state->obj);
 
-			ce->ring->head = ce->ring->tail = 0;
+			intel_ring_reset(ce->ring, 0);
-			intel_ring_update_space(ce->ring);
 		}
 	}
 }

drivers/gpu/drm/i915/intel_ringbuffer.c

@@ -49,7 +49,7 @@ static int __intel_ring_space(int head, int tail, int size)
 
 void intel_ring_update_space(struct intel_ring *ring)
 {
-	ring->space = __intel_ring_space(ring->head, ring->tail, ring->size);
+	ring->space = __intel_ring_space(ring->head, ring->emit, ring->size);
 }
 
 static int
@@ -774,8 +774,8 @@ static void i9xx_submit_request(struct drm_i915_gem_request *request)
 
 	i915_gem_request_submit(request);
 
-	assert_ring_tail_valid(request->ring, request->tail);
+	I915_WRITE_TAIL(request->engine,
-	I915_WRITE_TAIL(request->engine, request->tail);
+			intel_ring_set_tail(request->ring, request->tail));
 }
 
 static void i9xx_emit_breadcrumb(struct drm_i915_gem_request *req, u32 *cs)
@@ -1316,11 +1316,23 @@ err:
 	return PTR_ERR(addr);
 }
 
+void intel_ring_reset(struct intel_ring *ring, u32 tail)
+{
+	GEM_BUG_ON(!list_empty(&ring->request_list));
+	ring->tail = tail;
+	ring->head = tail;
+	ring->emit = tail;
+	intel_ring_update_space(ring);
+}
+
 void intel_ring_unpin(struct intel_ring *ring)
 {
 	GEM_BUG_ON(!ring->vma);
 	GEM_BUG_ON(!ring->vaddr);
 
+	/* Discard any unused bytes beyond that submitted to hw. */
+	intel_ring_reset(ring, ring->tail);
+
 	if (i915_vma_is_map_and_fenceable(ring->vma))
 		i915_vma_unpin_iomap(ring->vma);
 	else
@@ -1562,8 +1574,9 @@ void intel_legacy_submission_resume(struct drm_i915_private *dev_priv)
 	struct intel_engine_cs *engine;
 	enum intel_engine_id id;
 
+	/* Restart from the beginning of the rings for convenience */
 	for_each_engine(engine, dev_priv, id)
-		engine->buffer->head = engine->buffer->tail;
+		intel_ring_reset(engine->buffer, 0);
 }
 
 static int ring_request_alloc(struct drm_i915_gem_request *request)
@@ -1616,7 +1629,7 @@ static int wait_for_space(struct drm_i915_gem_request *req, int bytes)
 		unsigned space;
 
 		/* Would completion of this request free enough space? */
-		space = __intel_ring_space(target->postfix, ring->tail,
+		space = __intel_ring_space(target->postfix, ring->emit,
 					   ring->size);
 		if (space >= bytes)
 			break;
@@ -1641,8 +1654,8 @@ static int wait_for_space(struct drm_i915_gem_request *req, int bytes)
 u32 *intel_ring_begin(struct drm_i915_gem_request *req, int num_dwords)
 {
 	struct intel_ring *ring = req->ring;
-	int remain_actual = ring->size - ring->tail;
+	int remain_actual = ring->size - ring->emit;
-	int remain_usable = ring->effective_size - ring->tail;
+	int remain_usable = ring->effective_size - ring->emit;
 	int bytes = num_dwords * sizeof(u32);
 	int total_bytes, wait_bytes;
 	bool need_wrap = false;
@@ -1678,17 +1691,17 @@ u32 *intel_ring_begin(struct drm_i915_gem_request *req, int num_dwords)
 
 	if (unlikely(need_wrap)) {
 		GEM_BUG_ON(remain_actual > ring->space);
-		GEM_BUG_ON(ring->tail + remain_actual > ring->size);
+		GEM_BUG_ON(ring->emit + remain_actual > ring->size);
 
 		/* Fill the tail with MI_NOOP */
-		memset(ring->vaddr + ring->tail, 0, remain_actual);
+		memset(ring->vaddr + ring->emit, 0, remain_actual);
-		ring->tail = 0;
+		ring->emit = 0;
 		ring->space -= remain_actual;
 	}
 
-	GEM_BUG_ON(ring->tail > ring->size - bytes);
+	GEM_BUG_ON(ring->emit > ring->size - bytes);
-	cs = ring->vaddr + ring->tail;
+	cs = ring->vaddr + ring->emit;
-	ring->tail += bytes;
+	ring->emit += bytes;
 	ring->space -= bytes;
 	GEM_BUG_ON(ring->space < 0);
 
@@ -1699,7 +1712,7 @@ u32 *intel_ring_begin(struct drm_i915_gem_request *req, int num_dwords)
 int intel_ring_cacheline_align(struct drm_i915_gem_request *req)
 {
 	int num_dwords =
-		(req->ring->tail & (CACHELINE_BYTES - 1)) / sizeof(uint32_t);
+		(req->ring->emit & (CACHELINE_BYTES - 1)) / sizeof(uint32_t);
 	u32 *cs;
 
 	if (num_dwords == 0)

drivers/gpu/drm/i915/intel_ringbuffer.h

@@ -145,6 +145,7 @@ struct intel_ring {
 
 	u32 head;
 	u32 tail;
+	u32 emit;
 
 	int space;
 	int size;
@@ -488,6 +489,8 @@ intel_write_status_page(struct intel_engine_cs *engine, int reg, u32 value)
 struct intel_ring *
 intel_engine_create_ring(struct intel_engine_cs *engine, int size);
 int intel_ring_pin(struct intel_ring *ring, unsigned int offset_bias);
+void intel_ring_reset(struct intel_ring *ring, u32 tail);
+void intel_ring_update_space(struct intel_ring *ring);
 void intel_ring_unpin(struct intel_ring *ring);
 void intel_ring_free(struct intel_ring *ring);
 
@@ -511,7 +514,7 @@ intel_ring_advance(struct drm_i915_gem_request *req, u32 *cs)
 	 * reserved for the command packet (i.e. the value passed to
 	 * intel_ring_begin()).
 	 */
-	GEM_BUG_ON((req->ring->vaddr + req->ring->tail) != cs);
+	GEM_BUG_ON((req->ring->vaddr + req->ring->emit) != cs);
 }
 
 static inline u32
@@ -540,7 +543,19 @@ assert_ring_tail_valid(const struct intel_ring *ring, unsigned int tail)
 	GEM_BUG_ON(tail >= ring->size);
 }
 
-void intel_ring_update_space(struct intel_ring *ring);
+static inline unsigned int
+intel_ring_set_tail(struct intel_ring *ring, unsigned int tail)
+{
+	/* Whilst writes to the tail are strictly order, there is no
+	 * serialisation between readers and the writers. The tail may be
+	 * read by i915_gem_request_retire() just as it is being updated
+	 * by execlists, as although the breadcrumb is complete, the context
+	 * switch hasn't been seen.
+	 */
+	assert_ring_tail_valid(ring, tail);
+	ring->tail = tail;
+	return tail;
+}
 
 void intel_engine_init_global_seqno(struct intel_engine_cs *engine, u32 seqno);
 

drivers/gpu/drm/radeon/radeon_combios.c

@@ -3393,6 +3393,13 @@ void radeon_combios_asic_init(struct drm_device *dev)
 	    rdev->pdev->subsystem_vendor == 0x103c &&
 	    rdev->pdev->subsystem_device == 0x280a)
 		return;
+	/* quirk for rs4xx Toshiba Sattellite L20-183 latop to make it resume
+	 * - it hangs on resume inside the dynclk 1 table.
+	 */
+	if (rdev->family == CHIP_RS400 &&
+	    rdev->pdev->subsystem_vendor == 0x1179 &&
+	    rdev->pdev->subsystem_device == 0xff31)
+	        return;
 
 	/* DYN CLK 1 */
 	table = combios_get_table_offset(dev, COMBIOS_DYN_CLK_1_TABLE);

drivers/gpu/drm/radeon/radeon_device.c

@@ -136,6 +136,10 @@ static struct radeon_px_quirk radeon_px_quirk_list[] = {
 	 * https://bugzilla.kernel.org/show_bug.cgi?id=51381
 	 */
 	{ PCI_VENDOR_ID_ATI, 0x6840, 0x1043, 0x2122, RADEON_PX_QUIRK_DISABLE_PX },
+	/* Asus K53TK laptop with AMD A6-3420M APU and Radeon 7670m GPU
+	 * https://bugs.freedesktop.org/show_bug.cgi?id=101491
+	 */
+	{ PCI_VENDOR_ID_ATI, 0x6741, 0x1043, 0x2122, RADEON_PX_QUIRK_DISABLE_PX },
 	/* macbook pro 8.2 */
 	{ PCI_VENDOR_ID_ATI, 0x6741, PCI_VENDOR_ID_APPLE, 0x00e2, RADEON_PX_QUIRK_LONG_WAKEUP },
 	{ 0, 0, 0, 0, 0 },