Merge branch 'linus' into x86/asm, to pick up fixes

Signed-off-by: Ingo Molnar <mingo@kernel.org>

MAINTAINERS

@@ -3086,6 +3086,7 @@ M:	Stephen Boyd <sboyd@codeaurora.org>
 L:	linux-clk@vger.kernel.org
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux.git
 S:	Maintained
+F:	Documentation/devicetree/bindings/clock/
 F:	drivers/clk/
 X:	drivers/clk/clkdev.c
 F:	include/linux/clk-pr*
@@ -8008,6 +8009,7 @@ Q:	http://patchwork.kernel.org/project/linux-wireless/list/
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/kvalo/wireless-drivers.git
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/kvalo/wireless-drivers-next.git
 S:	Maintained
+F:	Documentation/devicetree/bindings/net/wireless/
 F:	drivers/net/wireless/
 
 NETXEN (1/10) GbE SUPPORT

arch/arc/Kconfig

@@ -61,7 +61,7 @@ config RWSEM_GENERIC_SPINLOCK
 	def_bool y
 
 config ARCH_DISCONTIGMEM_ENABLE
-	def_bool y
+	def_bool n
 
 config ARCH_FLATMEM_ENABLE
 	def_bool y
@@ -186,9 +186,6 @@ if SMP
 config ARC_HAS_COH_CACHES
 	def_bool n
 
-config ARC_HAS_REENTRANT_IRQ_LV2
-	def_bool n
-
 config ARC_MCIP
 	bool "ARConnect Multicore IP (MCIP) Support "
 	depends on ISA_ARCV2
@@ -366,25 +363,10 @@ config NODES_SHIFT
 if ISA_ARCOMPACT
 
 config ARC_COMPACT_IRQ_LEVELS
-	bool "ARCompact IRQ Priorities: High(2)/Low(1)"
+	bool "Setup Timer IRQ as high Priority"
 	default n
-	# Timer HAS to be high priority, for any other high priority config
-	select ARC_IRQ3_LV2
 	# if SMP, LV2 enabled ONLY if ARC implementation has LV2 re-entrancy
-	depends on !SMP || ARC_HAS_REENTRANT_IRQ_LV2
-
-if ARC_COMPACT_IRQ_LEVELS
-
-config ARC_IRQ3_LV2
-	bool
-
-config ARC_IRQ5_LV2
-	bool
-
-config ARC_IRQ6_LV2
-	bool
-
-endif	#ARC_COMPACT_IRQ_LEVELS
+	depends on !SMP
 
 config ARC_FPU_SAVE_RESTORE
 	bool "Enable FPU state persistence across context switch"
@@ -407,11 +389,6 @@ config ARC_HAS_LLSC
 	default y
 	depends on !ARC_CANT_LLSC
 
-config ARC_STAR_9000923308
-	bool "Workaround for llock/scond livelock"
-	default n
-	depends on ISA_ARCV2 && SMP && ARC_HAS_LLSC
-
 config ARC_HAS_SWAPE
 	bool "Insn: SWAPE (endian-swap)"
 	default y
@@ -471,7 +448,7 @@ config LINUX_LINK_BASE
 
 config HIGHMEM
 	bool "High Memory Support"
-	select DISCONTIGMEM
+	select ARCH_DISCONTIGMEM_ENABLE
 	help
 	  With ARC 2G:2G address split, only upper 2G is directly addressable by
 	  kernel. Enable this to potentially allow access to rest of 2G and PAE

arch/arc/Makefile

@@ -127,7 +127,7 @@ libs-y		+= arch/arc/lib/ $(LIBGCC)
 
 boot		:= arch/arc/boot
 
-#default target for make without any arguements.
+#default target for make without any arguments.
 KBUILD_IMAGE	:= bootpImage
 
 all:	$(KBUILD_IMAGE)

arch/arc/boot/dts/abilis_tb100.dtsi

@@ -23,8 +23,6 @@
 
 
 / {
-	clock-frequency		= <500000000>;	/* 500 MHZ */
-
 	soc100 {
 		bus-frequency	= <166666666>;
 

arch/arc/boot/dts/abilis_tb101.dtsi

@@ -23,8 +23,6 @@
 
 
 / {
-	clock-frequency		= <500000000>;	/* 500 MHZ */
-
 	soc100 {
 		bus-frequency	= <166666666>;
 

arch/arc/boot/dts/axc001.dtsi

@@ -15,7 +15,6 @@
 
 / {
 	compatible = "snps,arc";
-	clock-frequency = <750000000>;	/* 750 MHZ */
 	#address-cells = <1>;
 	#size-cells = <1>;
 

arch/arc/boot/dts/axc003.dtsi

@@ -14,7 +14,6 @@
 
 / {
 	compatible = "snps,arc";
-	clock-frequency = <90000000>;
 	#address-cells = <1>;
 	#size-cells = <1>;
 

arch/arc/boot/dts/axc003_idu.dtsi

@@ -14,7 +14,6 @@
 
 / {
 	compatible = "snps,arc";
-	clock-frequency = <90000000>;
 	#address-cells = <1>;
 	#size-cells = <1>;
 

arch/arc/boot/dts/eznps.dts

@@ -18,7 +18,6 @@
 
 / {
 	compatible = "ezchip,arc-nps";
-	clock-frequency = <83333333>;	/* 83.333333 MHZ */
 	#address-cells = <1>;
 	#size-cells = <1>;
 	interrupt-parent = <&intc>;

arch/arc/boot/dts/nsim_700.dts

@@ -11,7 +11,6 @@
 
 / {
 	compatible = "snps,nsim";
-	clock-frequency = <80000000>;	/* 80 MHZ */
 	#address-cells = <1>;
 	#size-cells = <1>;
 	interrupt-parent = <&core_intc>;

arch/arc/boot/dts/nsimosci.dts

@@ -11,7 +11,6 @@
 
 / {
 	compatible = "snps,nsimosci";
-	clock-frequency = <20000000>;	/* 20 MHZ */
 	#address-cells = <1>;
 	#size-cells = <1>;
 	interrupt-parent = <&core_intc>;

arch/arc/boot/dts/nsimosci_hs.dts

@@ -11,7 +11,6 @@
 
 / {
 	compatible = "snps,nsimosci_hs";
-	clock-frequency = <20000000>;	/* 20 MHZ */
 	#address-cells = <1>;
 	#size-cells = <1>;
 	interrupt-parent = <&core_intc>;

arch/arc/boot/dts/nsimosci_hs_idu.dts

@@ -11,7 +11,6 @@
 
 / {
 	compatible = "snps,nsimosci_hs";
-	clock-frequency = <5000000>;	/* 5 MHZ */
 	#address-cells = <1>;
 	#size-cells = <1>;
 	interrupt-parent = <&core_intc>;

arch/arc/boot/dts/skeleton.dtsi

@@ -13,7 +13,6 @@
 
 / {
 	compatible = "snps,arc";
-	clock-frequency = <80000000>;	/* 80 MHZ */
 	#address-cells = <1>;
 	#size-cells = <1>;
 	chosen { };

arch/arc/boot/dts/skeleton_hs.dtsi

@@ -8,7 +8,6 @@
 
 / {
 	compatible = "snps,arc";
-	clock-frequency = <80000000>;	/* 80 MHZ */
 	#address-cells = <1>;
 	#size-cells = <1>;
 	chosen { };

arch/arc/boot/dts/skeleton_hs_idu.dtsi

@@ -8,7 +8,6 @@
 
 / {
 	compatible = "snps,arc";
-	clock-frequency = <80000000>;	/* 80 MHZ */
 	#address-cells = <1>;
 	#size-cells = <1>;
 	chosen { };

arch/arc/boot/dts/vdk_axc003.dtsi

@@ -14,7 +14,6 @@
 
 / {
 	compatible = "snps,arc";
-	clock-frequency = <50000000>;
 	#address-cells = <1>;
 	#size-cells = <1>;
 

arch/arc/boot/dts/vdk_axc003_idu.dtsi

@@ -15,7 +15,6 @@
 
 / {
 	compatible = "snps,arc";
-	clock-frequency = <50000000>;
 	#address-cells = <1>;
 	#size-cells = <1>;
 

arch/arc/include/asm/atomic.h

@@ -25,50 +25,17 @@
 
 #define atomic_set(v, i) WRITE_ONCE(((v)->counter), (i))
 
-#ifdef CONFIG_ARC_STAR_9000923308
-
-#define SCOND_FAIL_RETRY_VAR_DEF						\
-	unsigned int delay = 1, tmp;						\
-
-#define SCOND_FAIL_RETRY_ASM							\
-	"	bz	4f			\n"				\
-	"   ; --- scond fail delay ---		\n"				\
-	"	mov	%[tmp], %[delay]	\n"	/* tmp = delay */	\
-	"2: 	brne.d	%[tmp], 0, 2b		\n"	/* while (tmp != 0) */	\
-	"	sub	%[tmp], %[tmp], 1	\n"	/* tmp-- */		\
-	"	rol	%[delay], %[delay]	\n"	/* delay *= 2 */	\
-	"	b	1b			\n"	/* start over */	\
-	"4: ; --- success ---			\n"				\
-
-#define SCOND_FAIL_RETRY_VARS							\
-	  ,[delay] "+&r" (delay),[tmp] "=&r"	(tmp)				\
-
-#else	/* !CONFIG_ARC_STAR_9000923308 */
-
-#define SCOND_FAIL_RETRY_VAR_DEF
-
-#define SCOND_FAIL_RETRY_ASM							\
-	"	bnz     1b			\n"				\
-
-#define SCOND_FAIL_RETRY_VARS
-
-#endif
-
 #define ATOMIC_OP(op, c_op, asm_op)					\
 static inline void atomic_##op(int i, atomic_t *v)			\
 {									\
-	unsigned int val;				                \
-	SCOND_FAIL_RETRY_VAR_DEF                                        \
+	unsigned int val;						\
 									\
 	__asm__ __volatile__(						\
 	"1:	llock   %[val], [%[ctr]]		\n"		\
 	"	" #asm_op " %[val], %[val], %[i]	\n"		\
 	"	scond   %[val], [%[ctr]]		\n"		\
-	"						\n"		\
-	SCOND_FAIL_RETRY_ASM						\
-									\
+	"	bnz     1b				\n"		\
 	: [val]	"=&r"	(val) /* Early clobber to prevent reg reuse */	\
-	  SCOND_FAIL_RETRY_VARS						\
 	: [ctr]	"r"	(&v->counter), /* Not "m": llock only supports reg direct addr mode */	\
 	  [i]	"ir"	(i)						\
 	: "cc");							\
@@ -77,8 +44,7 @@ static inline void atomic_##op(int i, atomic_t *v)			\
 #define ATOMIC_OP_RETURN(op, c_op, asm_op)				\
 static inline int atomic_##op##_return(int i, atomic_t *v)		\
 {									\
-	unsigned int val;				                \
-	SCOND_FAIL_RETRY_VAR_DEF                                        \
+	unsigned int val;						\
 									\
 	/*								\
 	 * Explicit full memory barrier needed before/after as		\
@@ -90,11 +56,8 @@ static inline int atomic_##op##_return(int i, atomic_t *v)		\
 	"1:	llock   %[val], [%[ctr]]		\n"		\
 	"	" #asm_op " %[val], %[val], %[i]	\n"		\
 	"	scond   %[val], [%[ctr]]		\n"		\
-	"						\n"		\
-	SCOND_FAIL_RETRY_ASM						\
-									\
+	"	bnz     1b				\n"		\
 	: [val]	"=&r"	(val)						\
-	  SCOND_FAIL_RETRY_VARS						\
 	: [ctr]	"r"	(&v->counter),					\
 	  [i]	"ir"	(i)						\
 	: "cc");							\

arch/arc/include/asm/entry-compact.h

@@ -76,8 +76,8 @@
 	 * We need to be a bit more cautious here. What if a kernel bug in
 	 * L1 ISR, caused SP to go whaco (some small value which looks like
 	 * USER stk) and then we take L2 ISR.
-	 * Above brlo alone would treat it as a valid L1-L2 sceanrio
-	 * instead of shouting alound
+	 * Above brlo alone would treat it as a valid L1-L2 scenario
+	 * instead of shouting around
 	 * The only feasible way is to make sure this L2 happened in
 	 * L1 prelogue ONLY i.e. ilink2 is less than a pre-set marker in
 	 * L1 ISR before it switches stack

arch/arc/include/asm/mmu_context.h

@@ -83,7 +83,7 @@ static inline void get_new_mmu_context(struct mm_struct *mm)
 		local_flush_tlb_all();
 
 		/*
-		 * Above checke for rollover of 8 bit ASID in 32 bit container.
+		 * Above check for rollover of 8 bit ASID in 32 bit container.
 		 * If the container itself wrapped around, set it to a non zero
 		 * "generation" to distinguish from no context
 		 */

arch/arc/include/asm/pgtable.h

@@ -47,7 +47,7 @@
  * Page Tables are purely for Linux VM's consumption and the bits below are
  * suited to that (uniqueness). Hence some are not implemented in the TLB and
  * some have different value in TLB.
- * e.g. MMU v2: K_READ bit is 8 and so is GLOBAL (possible becoz they live in
+ * e.g. MMU v2: K_READ bit is 8 and so is GLOBAL (possible because they live in
  *      seperate PD0 and PD1, which combined forms a translation entry)
  *      while for PTE perspective, they are 8 and 9 respectively
  * with MMU v3: Most bits (except SHARED) represent the exact hardware pos

arch/arc/include/asm/processor.h

@@ -78,7 +78,7 @@ struct task_struct;
 #define KSTK_ESP(tsk)   (task_pt_regs(tsk)->sp)
 
 /*
- * Where abouts of Task's sp, fp, blink when it was last seen in kernel mode.
+ * Where about of Task's sp, fp, blink when it was last seen in kernel mode.
  * Look in process.c for details of kernel stack layout
  */
 #define TSK_K_ESP(tsk)		(tsk->thread.ksp)

arch/arc/include/asm/smp.h

@@ -86,7 +86,7 @@ static inline const char *arc_platform_smp_cpuinfo(void)
  * (1) These insn were introduced only in 4.10 release. So for older released
  *	support needed.
  *
- * (2) In a SMP setup, the LLOCK/SCOND atomiticity across CPUs needs to be
+ * (2) In a SMP setup, the LLOCK/SCOND atomicity across CPUs needs to be
  *	gaurantted by the platform (not something which core handles).
  *	Assuming a platform won't, SMP Linux needs to use spinlocks + local IRQ
  *	disabling for atomicity.

arch/arc/include/asm/spinlock.h

@@ -20,11 +20,6 @@
 
 #ifdef CONFIG_ARC_HAS_LLSC
 
-/*
- * A normal LLOCK/SCOND based system, w/o need for livelock workaround
- */
-#ifndef CONFIG_ARC_STAR_9000923308
-
 static inline void arch_spin_lock(arch_spinlock_t *lock)
 {
 	unsigned int val;
@@ -238,293 +233,6 @@ static inline void arch_write_unlock(arch_rwlock_t *rw)
 	smp_mb();
 }
 
-#else	/* CONFIG_ARC_STAR_9000923308 */
-
-/*
- * HS38x4 could get into a LLOCK/SCOND livelock in case of multiple overlapping
- * coherency transactions in the SCU. The exclusive line state keeps rotating
- * among contenting cores leading to a never ending cycle. So break the cycle
- * by deferring the retry of failed exclusive access (SCOND). The actual delay
- * needed is function of number of contending cores as well as the unrelated
- * coherency traffic from other cores. To keep the code simple, start off with
- * small delay of 1 which would suffice most cases and in case of contention
- * double the delay. Eventually the delay is sufficient such that the coherency
- * pipeline is drained, thus a subsequent exclusive access would succeed.
- */
-
-#define SCOND_FAIL_RETRY_VAR_DEF						\
-	unsigned int delay, tmp;						\
-
-#define SCOND_FAIL_RETRY_ASM							\
-	"   ; --- scond fail delay ---		\n"				\
-	"	mov	%[tmp], %[delay]	\n"	/* tmp = delay */	\
-	"2: 	brne.d	%[tmp], 0, 2b		\n"	/* while (tmp != 0) */	\
-	"	sub	%[tmp], %[tmp], 1	\n"	/* tmp-- */		\
-	"	rol	%[delay], %[delay]	\n"	/* delay *= 2 */	\
-	"	b	1b			\n"	/* start over */	\
-	"					\n"				\
-	"4: ; --- done ---			\n"				\
-
-#define SCOND_FAIL_RETRY_VARS							\
-	  ,[delay] "=&r" (delay), [tmp] "=&r"	(tmp)				\
-
-static inline void arch_spin_lock(arch_spinlock_t *lock)
-{
-	unsigned int val;
-	SCOND_FAIL_RETRY_VAR_DEF;
-
-	smp_mb();
-
-	__asm__ __volatile__(
-	"0:	mov	%[delay], 1		\n"
-	"1:	llock	%[val], [%[slock]]	\n"
-	"	breq	%[val], %[LOCKED], 0b	\n"	/* spin while LOCKED */
-	"	scond	%[LOCKED], [%[slock]]	\n"	/* acquire */
-	"	bz	4f			\n"	/* done */
-	"					\n"
-	SCOND_FAIL_RETRY_ASM
-
-	: [val]		"=&r"	(val)
-	  SCOND_FAIL_RETRY_VARS
-	: [slock]	"r"	(&(lock->slock)),
-	  [LOCKED]	"r"	(__ARCH_SPIN_LOCK_LOCKED__)
-	: "memory", "cc");
-
-	smp_mb();
-}
-
-/* 1 - lock taken successfully */
-static inline int arch_spin_trylock(arch_spinlock_t *lock)
-{
-	unsigned int val, got_it = 0;
-	SCOND_FAIL_RETRY_VAR_DEF;
-
-	smp_mb();
-
-	__asm__ __volatile__(
-	"0:	mov	%[delay], 1		\n"
-	"1:	llock	%[val], [%[slock]]	\n"
-	"	breq	%[val], %[LOCKED], 4f	\n"	/* already LOCKED, just bail */
-	"	scond	%[LOCKED], [%[slock]]	\n"	/* acquire */
-	"	bz.d	4f			\n"
-	"	mov.z	%[got_it], 1		\n"	/* got it */
-	"					\n"
-	SCOND_FAIL_RETRY_ASM
-
-	: [val]		"=&r"	(val),
-	  [got_it]	"+&r"	(got_it)
-	  SCOND_FAIL_RETRY_VARS
-	: [slock]	"r"	(&(lock->slock)),
-	  [LOCKED]	"r"	(__ARCH_SPIN_LOCK_LOCKED__)
-	: "memory", "cc");
-
-	smp_mb();
-
-	return got_it;
-}
-
-static inline void arch_spin_unlock(arch_spinlock_t *lock)
-{
-	smp_mb();
-
-	lock->slock = __ARCH_SPIN_LOCK_UNLOCKED__;
-
-	smp_mb();
-}
-
-/*
- * Read-write spinlocks, allowing multiple readers but only one writer.
- * Unfair locking as Writers could be starved indefinitely by Reader(s)
- */
-
-static inline void arch_read_lock(arch_rwlock_t *rw)
-{
-	unsigned int val;
-	SCOND_FAIL_RETRY_VAR_DEF;
-
-	smp_mb();
-
-	/*
-	 * zero means writer holds the lock exclusively, deny Reader.
-	 * Otherwise grant lock to first/subseq reader
-	 *
-	 * 	if (rw->counter > 0) {
-	 *		rw->counter--;
-	 *		ret = 1;
-	 *	}
-	 */
-
-	__asm__ __volatile__(
-	"0:	mov	%[delay], 1		\n"
-	"1:	llock	%[val], [%[rwlock]]	\n"
-	"	brls	%[val], %[WR_LOCKED], 0b\n"	/* <= 0: spin while write locked */
-	"	sub	%[val], %[val], 1	\n"	/* reader lock */
-	"	scond	%[val], [%[rwlock]]	\n"
-	"	bz	4f			\n"	/* done */
-	"					\n"
-	SCOND_FAIL_RETRY_ASM
-
-	: [val]		"=&r"	(val)
-	  SCOND_FAIL_RETRY_VARS
-	: [rwlock]	"r"	(&(rw->counter)),
-	  [WR_LOCKED]	"ir"	(0)
-	: "memory", "cc");
-
-	smp_mb();
-}
-
-/* 1 - lock taken successfully */
-static inline int arch_read_trylock(arch_rwlock_t *rw)
-{
-	unsigned int val, got_it = 0;
-	SCOND_FAIL_RETRY_VAR_DEF;
-
-	smp_mb();
-
-	__asm__ __volatile__(
-	"0:	mov	%[delay], 1		\n"
-	"1:	llock	%[val], [%[rwlock]]	\n"
-	"	brls	%[val], %[WR_LOCKED], 4f\n"	/* <= 0: already write locked, bail */
-	"	sub	%[val], %[val], 1	\n"	/* counter-- */
-	"	scond	%[val], [%[rwlock]]	\n"
-	"	bz.d	4f			\n"
-	"	mov.z	%[got_it], 1		\n"	/* got it */
-	"					\n"
-	SCOND_FAIL_RETRY_ASM
-
-	: [val]		"=&r"	(val),
-	  [got_it]	"+&r"	(got_it)
-	  SCOND_FAIL_RETRY_VARS
-	: [rwlock]	"r"	(&(rw->counter)),
-	  [WR_LOCKED]	"ir"	(0)
-	: "memory", "cc");
-
-	smp_mb();
-
-	return got_it;
-}
-
-static inline void arch_write_lock(arch_rwlock_t *rw)
-{
-	unsigned int val;
-	SCOND_FAIL_RETRY_VAR_DEF;
-
-	smp_mb();
-
-	/*
-	 * If reader(s) hold lock (lock < __ARCH_RW_LOCK_UNLOCKED__),
-	 * deny writer. Otherwise if unlocked grant to writer
-	 * Hence the claim that Linux rwlocks are unfair to writers.
-	 * (can be starved for an indefinite time by readers).
-	 *
-	 *	if (rw->counter == __ARCH_RW_LOCK_UNLOCKED__) {
-	 *		rw->counter = 0;
-	 *		ret = 1;
-	 *	}
-	 */
-
-	__asm__ __volatile__(
-	"0:	mov	%[delay], 1		\n"
-	"1:	llock	%[val], [%[rwlock]]	\n"
-	"	brne	%[val], %[UNLOCKED], 0b	\n"	/* while !UNLOCKED spin */
-	"	mov	%[val], %[WR_LOCKED]	\n"
-	"	scond	%[val], [%[rwlock]]	\n"
-	"	bz	4f			\n"
-	"					\n"
-	SCOND_FAIL_RETRY_ASM
-
-	: [val]		"=&r"	(val)
-	  SCOND_FAIL_RETRY_VARS
-	: [rwlock]	"r"	(&(rw->counter)),
-	  [UNLOCKED]	"ir"	(__ARCH_RW_LOCK_UNLOCKED__),
-	  [WR_LOCKED]	"ir"	(0)
-	: "memory", "cc");
-
-	smp_mb();
-}
-
-/* 1 - lock taken successfully */
-static inline int arch_write_trylock(arch_rwlock_t *rw)
-{
-	unsigned int val, got_it = 0;
-	SCOND_FAIL_RETRY_VAR_DEF;
-
-	smp_mb();
-
-	__asm__ __volatile__(
-	"0:	mov	%[delay], 1		\n"
-	"1:	llock	%[val], [%[rwlock]]	\n"
-	"	brne	%[val], %[UNLOCKED], 4f	\n"	/* !UNLOCKED, bail */
-	"	mov	%[val], %[WR_LOCKED]	\n"
-	"	scond	%[val], [%[rwlock]]	\n"
-	"	bz.d	4f			\n"
-	"	mov.z	%[got_it], 1		\n"	/* got it */
-	"					\n"
-	SCOND_FAIL_RETRY_ASM
-
-	: [val]		"=&r"	(val),
-	  [got_it]	"+&r"	(got_it)
-	  SCOND_FAIL_RETRY_VARS
-	: [rwlock]	"r"	(&(rw->counter)),
-	  [UNLOCKED]	"ir"	(__ARCH_RW_LOCK_UNLOCKED__),
-	  [WR_LOCKED]	"ir"	(0)
-	: "memory", "cc");
-
-	smp_mb();
-
-	return got_it;
-}
-
-static inline void arch_read_unlock(arch_rwlock_t *rw)
-{
-	unsigned int val;
-
-	smp_mb();
-
-	/*
-	 * rw->counter++;
-	 */
-	__asm__ __volatile__(
-	"1:	llock	%[val], [%[rwlock]]	\n"
-	"	add	%[val], %[val], 1	\n"
-	"	scond	%[val], [%[rwlock]]	\n"
-	"	bnz	1b			\n"
-	"					\n"
-	: [val]		"=&r"	(val)
-	: [rwlock]	"r"	(&(rw->counter))
-	: "memory", "cc");
-
-	smp_mb();
-}
-
-static inline void arch_write_unlock(arch_rwlock_t *rw)
-{
-	unsigned int val;
-
-	smp_mb();
-
-	/*
-	 * rw->counter = __ARCH_RW_LOCK_UNLOCKED__;
-	 */
-	__asm__ __volatile__(
-	"1:	llock	%[val], [%[rwlock]]	\n"
-	"	scond	%[UNLOCKED], [%[rwlock]]\n"
-	"	bnz	1b			\n"
-	"					\n"
-	: [val]		"=&r"	(val)
-	: [rwlock]	"r"	(&(rw->counter)),
-	  [UNLOCKED]	"r"	(__ARCH_RW_LOCK_UNLOCKED__)
-	: "memory", "cc");
-
-	smp_mb();
-}
-
-#undef SCOND_FAIL_RETRY_VAR_DEF
-#undef SCOND_FAIL_RETRY_ASM
-#undef SCOND_FAIL_RETRY_VARS
-
-#endif	/* CONFIG_ARC_STAR_9000923308 */
-
 #else	/* !CONFIG_ARC_HAS_LLSC */
 
 static inline void arch_spin_lock(arch_spinlock_t *lock)

arch/arc/include/asm/thread_info.h

@@ -103,7 +103,7 @@ static inline __attribute_const__ struct thread_info *current_thread_info(void)
 
 /*
  * _TIF_ALLWORK_MASK includes SYSCALL_TRACE, but we don't need it.
- * SYSCALL_TRACE is anways seperately/unconditionally tested right after a
+ * SYSCALL_TRACE is anyway seperately/unconditionally tested right after a
  * syscall, so all that reamins to be tested is _TIF_WORK_MASK
  */
 

arch/arc/include/asm/uaccess.h

@@ -32,7 +32,7 @@
 #define __kernel_ok		(segment_eq(get_fs(), KERNEL_DS))
 
 /*
- * Algorthmically, for __user_ok() we want do:
+ * Algorithmically, for __user_ok() we want do:
  * 	(start < TASK_SIZE) && (start+len < TASK_SIZE)
  * where TASK_SIZE could either be retrieved from thread_info->addr_limit or
  * emitted directly in code.

arch/arc/include/uapi/asm/swab.h

@@ -74,7 +74,7 @@
 	__tmp ^ __in;						\
 })
 
-#elif (ARC_BSWAP_TYPE == 2)	/* Custom single cycle bwap instruction */
+#elif (ARC_BSWAP_TYPE == 2)	/* Custom single cycle bswap instruction */
 
 #define __arch_swab32(x)						\
 ({									\

arch/arc/kernel/entry-compact.S

@@ -91,27 +91,13 @@ VECTOR   mem_service             ; 0x8, Mem exception   (0x1)
 VECTOR   instr_service           ; 0x10, Instrn Error   (0x2)
 
 ; ******************** Device ISRs **********************
-#ifdef CONFIG_ARC_IRQ3_LV2
-VECTOR   handle_interrupt_level2
-#else
-VECTOR   handle_interrupt_level1
-#endif
-
-VECTOR   handle_interrupt_level1
-
-#ifdef CONFIG_ARC_IRQ5_LV2
-VECTOR   handle_interrupt_level2
-#else
-VECTOR   handle_interrupt_level1
-#endif
-
-#ifdef CONFIG_ARC_IRQ6_LV2
+#ifdef CONFIG_ARC_COMPACT_IRQ_LEVELS
 VECTOR   handle_interrupt_level2
 #else
 VECTOR   handle_interrupt_level1
 #endif
 
-.rept   25
+.rept   28
 VECTOR   handle_interrupt_level1 ; Other devices
 .endr
 

arch/arc/kernel/intc-compact.c

@@ -28,10 +28,8 @@ void arc_init_IRQ(void)
 {
 	int level_mask = 0;
 
-       /* setup any high priority Interrupts (Level2 in ARCompact jargon) */
-	level_mask |= IS_ENABLED(CONFIG_ARC_IRQ3_LV2) << 3;
-	level_mask |= IS_ENABLED(CONFIG_ARC_IRQ5_LV2) << 5;
-	level_mask |= IS_ENABLED(CONFIG_ARC_IRQ6_LV2) << 6;
+       /* Is timer high priority Interrupt (Level2 in ARCompact jargon) */
+	level_mask |= IS_ENABLED(CONFIG_ARC_COMPACT_IRQ_LEVELS) << TIMER0_IRQ;
 
 	/*
 	 * Write to register, even if no LV2 IRQs configured to reset it

arch/arc/kernel/perf_event.c

@@ -108,7 +108,7 @@ static void arc_perf_event_update(struct perf_event *event,
 	int64_t delta = new_raw_count - prev_raw_count;
 
 	/*
-	 * We don't afaraid of hwc->prev_count changing beneath our feet
+	 * We aren't afraid of hwc->prev_count changing beneath our feet
 	 * because there's no way for us to re-enter this function anytime.
 	 */
 	local64_set(&hwc->prev_count, new_raw_count);

arch/arc/kernel/setup.c

@@ -392,7 +392,7 @@ void __init setup_arch(char **cmdline_p)
 		/*
 		 * If we are here, it is established that @uboot_arg didn't
 		 * point to DT blob. Instead if u-boot says it is cmdline,
-		 * Appent to embedded DT cmdline.
+		 * append to embedded DT cmdline.
 		 * setup_machine_fdt() would have populated @boot_command_line
 		 */
 		if (uboot_tag == 1) {

arch/arc/kernel/signal.c

@@ -34,7 +34,7 @@
  *  -ViXS were still seeing crashes when using insmod to load drivers.
  *   It turned out that the code to change Execute permssions for TLB entries
  *   of user was not guarded for interrupts (mod_tlb_permission)
- *   This was cauing TLB entries to be overwritten on unrelated indexes
+ *   This was causing TLB entries to be overwritten on unrelated indexes
  *
  * Vineetg: July 15th 2008: Bug #94183
  *  -Exception happens in Delay slot of a JMP, and before user space resumes,

arch/arc/kernel/troubleshoot.c

@@ -276,7 +276,7 @@ static int tlb_stats_open(struct inode *inode, struct file *file)
 	return 0;
 }
 
-/* called on user read(): display the couters */
+/* called on user read(): display the counters */
 static ssize_t tlb_stats_output(struct file *file,	/* file descriptor */
 				char __user *user_buf,	/* user buffer */
 				size_t len,		/* length of buffer */

arch/arc/mm/cache.c

@@ -215,7 +215,7 @@ slc_chk:
  * ------------------
  * This ver of MMU supports variable page sizes (1k-16k): although Linux will
  * only support 8k (default), 16k and 4k.
- * However from hardware perspective, smaller page sizes aggrevate aliasing
+ * However from hardware perspective, smaller page sizes aggravate aliasing
  * meaning more vaddr bits needed to disambiguate the cache-line-op ;
  * the existing scheme of piggybacking won't work for certain configurations.
  * Two new registers IC_PTAG and DC_PTAG inttoduced.
@@ -302,7 +302,7 @@ void __cache_line_loop_v3(phys_addr_t paddr, unsigned long vaddr,
 
 	/*
 	 * This is technically for MMU v4, using the MMU v3 programming model
-	 * Special work for HS38 aliasing I-cache configuratino with PAE40
+	 * Special work for HS38 aliasing I-cache configuration with PAE40
 	 *   - upper 8 bits of paddr need to be written into PTAG_HI
 	 *   - (and needs to be written before the lower 32 bits)
 	 * Note that PTAG_HI is hoisted outside the line loop
@@ -936,7 +936,7 @@ void arc_cache_init(void)
 			      ic->ver, CONFIG_ARC_MMU_VER);
 
 		/*
-		 * In MMU v4 (HS38x) the alising icache config uses IVIL/PTAG
+		 * In MMU v4 (HS38x) the aliasing icache config uses IVIL/PTAG
 		 * pair to provide vaddr/paddr respectively, just as in MMU v3
 		 */
 		if (is_isa_arcv2() && ic->alias)

arch/arc/mm/dma.c

@@ -10,7 +10,7 @@
  * DMA Coherent API Notes
  *
  * I/O is inherently non-coherent on ARC. So a coherent DMA buffer is
- * implemented by accessintg it using a kernel virtual address, with
+ * implemented by accessing it using a kernel virtual address, with
  * Cache bit off in the TLB entry.
  *
  * The default DMA address == Phy address which is 0x8000_0000 based.

arch/arm/mach-vexpress/spc.c

@@ -547,7 +547,7 @@ static struct clk *ve_spc_clk_register(struct device *cpu_dev)
 
 	init.name = dev_name(cpu_dev);
 	init.ops = &clk_spc_ops;
-	init.flags = CLK_IS_ROOT | CLK_GET_RATE_NOCACHE;
+	init.flags = CLK_GET_RATE_NOCACHE;
 	init.num_parents = 0;
 
 	return devm_clk_register(cpu_dev, &spc->hw);

arch/powerpc/platforms/512x/clock-commonclk.c

@@ -221,7 +221,7 @@ static bool soc_has_mclk_mux0_canin(void)
 /* convenience wrappers around the common clk API */
 static inline struct clk *mpc512x_clk_fixed(const char *name, int rate)
 {
-	return clk_register_fixed_rate(NULL, name, NULL, CLK_IS_ROOT, rate);
+	return clk_register_fixed_rate(NULL, name, NULL, 0, rate);
 }
 
 static inline struct clk *mpc512x_clk_factor(

arch/powerpc/platforms/cell/spufs/coredump.c

@@ -172,7 +172,7 @@ static int spufs_arch_write_note(struct spu_context *ctx, int i,
 	if (rc < 0)
 		goto out;
 
-	skip = roundup(cprm->file->f_pos - total + sz, 4) - cprm->file->f_pos;
+	skip = roundup(cprm->pos - total + sz, 4) - cprm->pos;
 	if (!dump_skip(cprm, skip))
 		goto Eio;
 out:

arch/x86/boot/Makefile

@@ -162,6 +162,9 @@ isoimage: $(obj)/bzImage
 	for i in lib lib64 share end ; do \
 		if [ -f /usr/$$i/syslinux/isolinux.bin ] ; then \
 			cp /usr/$$i/syslinux/isolinux.bin $(obj)/isoimage ; \
+			if [ -f /usr/$$i/syslinux/ldlinux.c32 ]; then \
+				cp /usr/$$i/syslinux/ldlinux.c32 $(obj)/isoimage ; \
+			fi ; \
 			break ; \
 		fi ; \
 		if [ $$i = end ] ; then exit 1 ; fi ; \

arch/x86/events/intel/rapl.c

@@ -714,7 +714,7 @@ static void cleanup_rapl_pmus(void)
 	int i;
 
 	for (i = 0; i < rapl_pmus->maxpkg; i++)
-		kfree(rapl_pmus->pmus + i);
+		kfree(rapl_pmus->pmus[i]);
 	kfree(rapl_pmus);
 }
 

arch/x86/events/intel/uncore_snbep.c

@@ -2868,27 +2868,10 @@ static struct intel_uncore_type bdx_uncore_cbox = {
 	.format_group		= &hswep_uncore_cbox_format_group,
 };
 
-static struct intel_uncore_type bdx_uncore_sbox = {
-	.name			= "sbox",
-	.num_counters		= 4,
-	.num_boxes		= 4,
-	.perf_ctr_bits		= 48,
-	.event_ctl		= HSWEP_S0_MSR_PMON_CTL0,
-	.perf_ctr		= HSWEP_S0_MSR_PMON_CTR0,
-	.event_mask		= HSWEP_S_MSR_PMON_RAW_EVENT_MASK,
-	.box_ctl		= HSWEP_S0_MSR_PMON_BOX_CTL,
-	.msr_offset		= HSWEP_SBOX_MSR_OFFSET,
-	.ops			= &hswep_uncore_sbox_msr_ops,
-	.format_group		= &hswep_uncore_sbox_format_group,
-};
-
-#define BDX_MSR_UNCORE_SBOX	3
-
 static struct intel_uncore_type *bdx_msr_uncores[] = {
 	&bdx_uncore_ubox,
 	&bdx_uncore_cbox,
 	&hswep_uncore_pcu,
-	&bdx_uncore_sbox,
 	NULL,
 };
 
@@ -2897,10 +2880,6 @@ void bdx_uncore_cpu_init(void)
 	if (bdx_uncore_cbox.num_boxes > boot_cpu_data.x86_max_cores)
 		bdx_uncore_cbox.num_boxes = boot_cpu_data.x86_max_cores;
 	uncore_msr_uncores = bdx_msr_uncores;
-
-	/* BDX-DE doesn't have SBOX */
-	if (boot_cpu_data.x86_model == 86)
-		uncore_msr_uncores[BDX_MSR_UNCORE_SBOX] = NULL;
 }
 
 static struct intel_uncore_type bdx_uncore_ha = {

arch/x86/include/asm/intel-family.h

@@ -0,0 +1,68 @@
+#ifndef _ASM_X86_INTEL_FAMILY_H
+#define _ASM_X86_INTEL_FAMILY_H
+
+/*
+ * "Big Core" Processors (Branded as Core, Xeon, etc...)
+ *
+ * The "_X" parts are generally the EP and EX Xeons, or the
+ * "Extreme" ones, like Broadwell-E.
+ *
+ * Things ending in "2" are usually because we have no better
+ * name for them.  There's no processor called "WESTMERE2".
+ */
+
+#define INTEL_FAM6_CORE_YONAH		0x0E
+#define INTEL_FAM6_CORE2_MEROM		0x0F
+#define INTEL_FAM6_CORE2_MEROM_L	0x16
+#define INTEL_FAM6_CORE2_PENRYN		0x17
+#define INTEL_FAM6_CORE2_DUNNINGTON	0x1D
+
+#define INTEL_FAM6_NEHALEM		0x1E
+#define INTEL_FAM6_NEHALEM_EP		0x1A
+#define INTEL_FAM6_NEHALEM_EX		0x2E
+#define INTEL_FAM6_WESTMERE		0x25
+#define INTEL_FAM6_WESTMERE2		0x1F
+#define INTEL_FAM6_WESTMERE_EP		0x2C
+#define INTEL_FAM6_WESTMERE_EX		0x2F
+
+#define INTEL_FAM6_SANDYBRIDGE		0x2A
+#define INTEL_FAM6_SANDYBRIDGE_X	0x2D
+#define INTEL_FAM6_IVYBRIDGE		0x3A
+#define INTEL_FAM6_IVYBRIDGE_X		0x3E
+
+#define INTEL_FAM6_HASWELL_CORE		0x3C
+#define INTEL_FAM6_HASWELL_X		0x3F
+#define INTEL_FAM6_HASWELL_ULT		0x45
+#define INTEL_FAM6_HASWELL_GT3E		0x46
+
+#define INTEL_FAM6_BROADWELL_CORE	0x3D
+#define INTEL_FAM6_BROADWELL_XEON_D	0x56
+#define INTEL_FAM6_BROADWELL_GT3E	0x47
+#define INTEL_FAM6_BROADWELL_X		0x4F
+
+#define INTEL_FAM6_SKYLAKE_MOBILE	0x4E
+#define INTEL_FAM6_SKYLAKE_DESKTOP	0x5E
+#define INTEL_FAM6_SKYLAKE_X		0x55
+#define INTEL_FAM6_KABYLAKE_MOBILE	0x8E
+#define INTEL_FAM6_KABYLAKE_DESKTOP	0x9E
+
+/* "Small Core" Processors (Atom) */
+
+#define INTEL_FAM6_ATOM_PINEVIEW	0x1C
+#define INTEL_FAM6_ATOM_LINCROFT	0x26
+#define INTEL_FAM6_ATOM_PENWELL		0x27
+#define INTEL_FAM6_ATOM_CLOVERVIEW	0x35
+#define INTEL_FAM6_ATOM_CEDARVIEW	0x36
+#define INTEL_FAM6_ATOM_SILVERMONT1	0x37 /* BayTrail/BYT / Valleyview */
+#define INTEL_FAM6_ATOM_SILVERMONT2	0x4D /* Avaton/Rangely */
+#define INTEL_FAM6_ATOM_AIRMONT		0x4C /* CherryTrail / Braswell */
+#define INTEL_FAM6_ATOM_MERRIFIELD1	0x4A /* Tangier */
+#define INTEL_FAM6_ATOM_MERRIFIELD2	0x5A /* Annidale */
+#define INTEL_FAM6_ATOM_GOLDMONT	0x5C
+#define INTEL_FAM6_ATOM_DENVERTON	0x5F /* Goldmont Microserver */
+
+/* Xeon Phi */
+
+#define INTEL_FAM6_XEON_PHI_KNL		0x57 /* Knights Landing */
+
+#endif /* _ASM_X86_INTEL_FAMILY_H */

arch/x86/kernel/apic/io_apic.c

@@ -2588,8 +2588,8 @@ static struct resource * __init ioapic_setup_resources(void)
 		res[num].flags = IORESOURCE_MEM | IORESOURCE_BUSY;
 		snprintf(mem, IOAPIC_RESOURCE_NAME_SIZE, "IOAPIC %u", i);
 		mem += IOAPIC_RESOURCE_NAME_SIZE;
+		ioapics[i].iomem_res = &res[num];
 		num++;
-		ioapics[i].iomem_res = res;
 	}
 
 	ioapic_resources = res;

arch/x86/kernel/cpu/amd.c

@@ -674,14 +674,14 @@ static void init_amd_bd(struct cpuinfo_x86 *c)
 	u64 value;
 
 	/* re-enable TopologyExtensions if switched off by BIOS */
-	if ((c->x86_model >= 0x10) && (c->x86_model <= 0x1f) &&
+	if ((c->x86_model >= 0x10) && (c->x86_model <= 0x6f) &&
 	    !cpu_has(c, X86_FEATURE_TOPOEXT)) {
 
 		if (msr_set_bit(0xc0011005, 54) > 0) {
 			rdmsrl(0xc0011005, value);
 			if (value & BIT_64(54)) {
 				set_cpu_cap(c, X86_FEATURE_TOPOEXT);
-				pr_info(FW_INFO "CPU: Re-enabling disabled Topology Extensions Support.\n");
+				pr_info_once(FW_INFO "CPU: Re-enabling disabled Topology Extensions Support.\n");
 			}
 		}
 	}

arch/x86/kernel/traps.c

@@ -96,6 +96,12 @@ static inline void cond_local_irq_disable(struct pt_regs *regs)
 		local_irq_disable();
 }
 
+/*
+ * In IST context, we explicitly disable preemption.  This serves two
+ * purposes: it makes it much less likely that we would accidentally
+ * schedule in IST context and it will force a warning if we somehow
+ * manage to schedule by accident.
+ */
 void ist_enter(struct pt_regs *regs)
 {
 	if (user_mode(regs)) {
@@ -110,13 +116,7 @@ void ist_enter(struct pt_regs *regs)
 		rcu_nmi_enter();
 	}
 
-	/*
-	 * We are atomic because we're on the IST stack; or we're on
-	 * x86_32, in which case we still shouldn't schedule; or we're
-	 * on x86_64 and entered from user mode, in which case we're
-	 * still atomic unless ist_begin_non_atomic is called.
-	 */
-	preempt_count_add(HARDIRQ_OFFSET);
+	preempt_disable();
 
 	/* This code is a bit fragile.  Test it. */
 	RCU_LOCKDEP_WARN(!rcu_is_watching(), "ist_enter didn't work");
@@ -124,7 +124,7 @@ void ist_enter(struct pt_regs *regs)
 
 void ist_exit(struct pt_regs *regs)
 {
-	preempt_count_sub(HARDIRQ_OFFSET);
+	preempt_enable_no_resched();
 
 	if (!user_mode(regs))
 		rcu_nmi_exit();
@@ -155,7 +155,7 @@ void ist_begin_non_atomic(struct pt_regs *regs)
 	BUG_ON((unsigned long)(current_top_of_stack() -
 			       current_stack_pointer()) >= THREAD_SIZE);
 
-	preempt_count_sub(HARDIRQ_OFFSET);
+	preempt_enable_no_resched();
 }
 
 /**
@@ -165,7 +165,7 @@ void ist_begin_non_atomic(struct pt_regs *regs)
  */
 void ist_end_non_atomic(void)
 {
-	preempt_count_add(HARDIRQ_OFFSET);
+	preempt_disable();
 }
 
 static nokprobe_inline int

drivers/acpi/bus.c

@@ -1051,7 +1051,7 @@ static int __init acpi_bus_init(void)
 	 * Maybe EC region is required at bus_scan/acpi_get_devices. So it
 	 * is necessary to enable it as early as possible.
 	 */
-	acpi_boot_ec_enable();
+	acpi_ec_dsdt_probe();
 
 	printk(KERN_INFO PREFIX "Interpreter enabled\n");
 

drivers/acpi/ec.c

@@ -1446,10 +1446,30 @@ ec_parse_io_ports(struct acpi_resource *resource, void *context)
 	return AE_OK;
 }
 
-int __init acpi_boot_ec_enable(void)
+static const struct acpi_device_id ec_device_ids[] = {
+	{"PNP0C09", 0},
+	{"", 0},
+};
+
+int __init acpi_ec_dsdt_probe(void)
 {
-	if (!boot_ec)
+	acpi_status status;
+
+	if (boot_ec)
 		return 0;
+
+	/*
+	 * Finding EC from DSDT if there is no ECDT EC available. When this
+	 * function is invoked, ACPI tables have been fully loaded, we can
+	 * walk namespace now.
+	 */
+	boot_ec = make_acpi_ec();
+	if (!boot_ec)
+		return -ENOMEM;
+	status = acpi_get_devices(ec_device_ids[0].id,
+				  ec_parse_device, boot_ec, NULL);
+	if (ACPI_FAILURE(status) || !boot_ec->handle)
+		return -ENODEV;
 	if (!ec_install_handlers(boot_ec)) {
 		first_ec = boot_ec;
 		return 0;
@@ -1457,11 +1477,6 @@ int __init acpi_boot_ec_enable(void)
 	return -EFAULT;
 }
 
-static const struct acpi_device_id ec_device_ids[] = {
-	{"PNP0C09", 0},
-	{"", 0},
-};
-
 #if 0
 /*
  * Some EC firmware variations refuses to respond QR_EC when SCI_EVT is not

drivers/acpi/internal.h

@@ -181,7 +181,7 @@ typedef int (*acpi_ec_query_func) (void *data);
 
 int acpi_ec_init(void);
 int acpi_ec_ecdt_probe(void);
-int acpi_boot_ec_enable(void);
+int acpi_ec_dsdt_probe(void);
 void acpi_ec_block_transactions(void);
 void acpi_ec_unblock_transactions(void);
 void acpi_ec_unblock_transactions_early(void);

drivers/clk/Kconfig

@@ -175,6 +175,7 @@ config COMMON_CLK_KEYSTONE
 config COMMON_CLK_NXP
 	def_bool COMMON_CLK && (ARCH_LPC18XX || ARCH_LPC32XX)
 	select REGMAP_MMIO if ARCH_LPC32XX
+	select MFD_SYSCON if ARCH_LPC18XX
 	---help---
 	  Support for clock providers on NXP platforms.
 

drivers/clk/microchip/clk-pic32mzda.c

@@ -180,15 +180,15 @@ static int pic32mzda_clk_probe(struct platform_device *pdev)
 
 	/* register fixed rate clocks */
 	clks[POSCCLK] = clk_register_fixed_rate(&pdev->dev, "posc_clk", NULL,
-						CLK_IS_ROOT, 24000000);
+						0, 24000000);
 	clks[FRCCLK] =  clk_register_fixed_rate(&pdev->dev, "frc_clk", NULL,
-						CLK_IS_ROOT, 8000000);
+						0, 8000000);
 	clks[BFRCCLK] = clk_register_fixed_rate(&pdev->dev, "bfrc_clk", NULL,
-						CLK_IS_ROOT, 8000000);
+						0, 8000000);
 	clks[LPRCCLK] = clk_register_fixed_rate(&pdev->dev, "lprc_clk", NULL,
-						CLK_IS_ROOT, 32000);
+						0, 32000);
 	clks[UPLLCLK] = clk_register_fixed_rate(&pdev->dev, "usbphy_clk", NULL,
-						CLK_IS_ROOT, 24000000);
+						0, 24000000);
 	/* fixed rate (optional) clock */
 	if (of_find_property(np, "microchip,pic32mzda-sosc", NULL)) {
 		pr_info("pic32-clk: dt requests SOSC.\n");

drivers/cpufreq/intel_pstate.c

@@ -1460,6 +1460,9 @@ static int intel_pstate_set_policy(struct cpufreq_policy *policy)
 
 	intel_pstate_clear_update_util_hook(policy->cpu);
 
+	pr_debug("set_policy cpuinfo.max %u policy->max %u\n",
+		 policy->cpuinfo.max_freq, policy->max);
+
 	cpu = all_cpu_data[0];
 	if (cpu->pstate.max_pstate_physical > cpu->pstate.max_pstate &&
 	    policy->max < policy->cpuinfo.max_freq &&
@@ -1495,13 +1498,13 @@ static int intel_pstate_set_policy(struct cpufreq_policy *policy)
 				   limits->max_sysfs_pct);
 	limits->max_perf_pct = max(limits->min_policy_pct,
 				   limits->max_perf_pct);
-	limits->max_perf = round_up(limits->max_perf, FRAC_BITS);
 
 	/* Make sure min_perf_pct <= max_perf_pct */
 	limits->min_perf_pct = min(limits->max_perf_pct, limits->min_perf_pct);
 
 	limits->min_perf = div_fp(limits->min_perf_pct, 100);
 	limits->max_perf = div_fp(limits->max_perf_pct, 100);
+	limits->max_perf = round_up(limits->max_perf, FRAC_BITS);
 
  out:
 	intel_pstate_set_update_util_hook(policy->cpu);
@@ -1558,8 +1561,11 @@ static int intel_pstate_cpu_init(struct cpufreq_policy *policy)
 
 	/* cpuinfo and default policy values */
 	policy->cpuinfo.min_freq = cpu->pstate.min_pstate * cpu->pstate.scaling;
-	policy->cpuinfo.max_freq =
-		cpu->pstate.turbo_pstate * cpu->pstate.scaling;
+	update_turbo_state();
+	policy->cpuinfo.max_freq = limits->turbo_disabled ?
+			cpu->pstate.max_pstate : cpu->pstate.turbo_pstate;
+	policy->cpuinfo.max_freq *= cpu->pstate.scaling;
+
 	intel_pstate_init_acpi_perf_limits(policy);
 	policy->cpuinfo.transition_latency = CPUFREQ_ETERNAL;
 	cpumask_set_cpu(policy->cpu, policy->cpus);

drivers/edac/edac_mc.c

@@ -565,7 +565,8 @@ void edac_mc_reset_delay_period(unsigned long value)
 	list_for_each(item, &mc_devices) {
 		mci = list_entry(item, struct mem_ctl_info, link);
 
-		edac_mod_work(&mci->work, value);
+		if (mci->op_state == OP_RUNNING_POLL)
+			edac_mod_work(&mci->work, value);
 	}
 	mutex_unlock(&mem_ctls_mutex);
 }

drivers/edac/sb_edac.c

@@ -239,8 +239,11 @@ static const u32 rir_offset[MAX_RIR_RANGES][MAX_RIR_WAY] = {
 	{ 0x1a0, 0x1a4, 0x1a8, 0x1ac, 0x1b0, 0x1b4, 0x1b8, 0x1bc },
 };
 
-#define RIR_RNK_TGT(reg)		GET_BITFIELD(reg, 16, 19)
-#define RIR_OFFSET(reg)		GET_BITFIELD(reg,  2, 14)
+#define RIR_RNK_TGT(type, reg) (((type) == BROADWELL) ? \
+	GET_BITFIELD(reg, 20, 23) : GET_BITFIELD(reg, 16, 19))
+
+#define RIR_OFFSET(type, reg) (((type) == HASWELL || (type) == BROADWELL) ? \
+	GET_BITFIELD(reg,  2, 15) : GET_BITFIELD(reg,  2, 14))
 
 /* Device 16, functions 2-7 */
 
@@ -326,6 +329,7 @@ struct pci_id_descr {
 struct pci_id_table {
 	const struct pci_id_descr	*descr;
 	int				n_devs;
+	enum type			type;
 };
 
 struct sbridge_dev {
@@ -394,9 +398,14 @@ static const struct pci_id_descr pci_dev_descr_sbridge[] = {
 	{ PCI_DESCR(PCI_DEVICE_ID_INTEL_SBRIDGE_BR, 0)		},
 };
 
-#define PCI_ID_TABLE_ENTRY(A) { .descr=A, .n_devs = ARRAY_SIZE(A) }
+#define PCI_ID_TABLE_ENTRY(A, T) {	\
+	.descr = A,			\
+	.n_devs = ARRAY_SIZE(A),	\
+	.type = T			\
+}
+
 static const struct pci_id_table pci_dev_descr_sbridge_table[] = {
-	PCI_ID_TABLE_ENTRY(pci_dev_descr_sbridge),
+	PCI_ID_TABLE_ENTRY(pci_dev_descr_sbridge, SANDY_BRIDGE),
 	{0,}			/* 0 terminated list. */
 };
 
@@ -463,7 +472,7 @@ static const struct pci_id_descr pci_dev_descr_ibridge[] = {
 };
 
 static const struct pci_id_table pci_dev_descr_ibridge_table[] = {
-	PCI_ID_TABLE_ENTRY(pci_dev_descr_ibridge),
+	PCI_ID_TABLE_ENTRY(pci_dev_descr_ibridge, IVY_BRIDGE),
 	{0,}			/* 0 terminated list. */
 };
 
@@ -536,7 +545,7 @@ static const struct pci_id_descr pci_dev_descr_haswell[] = {
 };
 
 static const struct pci_id_table pci_dev_descr_haswell_table[] = {
-	PCI_ID_TABLE_ENTRY(pci_dev_descr_haswell),
+	PCI_ID_TABLE_ENTRY(pci_dev_descr_haswell, HASWELL),
 	{0,}			/* 0 terminated list. */
 };
 
@@ -580,7 +589,7 @@ static const struct pci_id_descr pci_dev_descr_knl[] = {
 };
 
 static const struct pci_id_table pci_dev_descr_knl_table[] = {
-	PCI_ID_TABLE_ENTRY(pci_dev_descr_knl),
+	PCI_ID_TABLE_ENTRY(pci_dev_descr_knl, KNIGHTS_LANDING),
 	{0,}
 };
 
@@ -648,7 +657,7 @@ static const struct pci_id_descr pci_dev_descr_broadwell[] = {
 };
 
 static const struct pci_id_table pci_dev_descr_broadwell_table[] = {
-	PCI_ID_TABLE_ENTRY(pci_dev_descr_broadwell),
+	PCI_ID_TABLE_ENTRY(pci_dev_descr_broadwell, BROADWELL),
 	{0,}			/* 0 terminated list. */
 };
 
@@ -1894,14 +1903,14 @@ static void get_memory_layout(const struct mem_ctl_info *mci)
 				pci_read_config_dword(pvt->pci_tad[i],
 						      rir_offset[j][k],
 						      &reg);
-				tmp_mb = RIR_OFFSET(reg) << 6;
+				tmp_mb = RIR_OFFSET(pvt->info.type, reg) << 6;
 
 				gb = div_u64_rem(tmp_mb, 1024, &mb);
 				edac_dbg(0, "CH#%d RIR#%d INTL#%d, offset %u.%03u GB (0x%016Lx), tgt: %d, reg=0x%08x\n",
 					 i, j, k,
 					 gb, (mb*1000)/1024,
 					 ((u64)tmp_mb) << 20L,
-					 (u32)RIR_RNK_TGT(reg),
+					 (u32)RIR_RNK_TGT(pvt->info.type, reg),
 					 reg);
 			}
 		}
@@ -2234,7 +2243,7 @@ static int get_memory_error_data(struct mem_ctl_info *mci,
 	pci_read_config_dword(pvt->pci_tad[ch_add + base_ch],
 			      rir_offset[n_rir][idx],
 			      &reg);
-	*rank = RIR_RNK_TGT(reg);
+	*rank = RIR_RNK_TGT(pvt->info.type, reg);
 
 	edac_dbg(0, "RIR#%d: channel address 0x%08Lx < 0x%08Lx, RIR interleave %d, index %d\n",
 		 n_rir,
@@ -3357,12 +3366,12 @@ fail0:
 #define ICPU(model, table) \
 	{ X86_VENDOR_INTEL, 6, model, 0, (unsigned long)&table }
 
-/* Order here must match "enum type" */
 static const struct x86_cpu_id sbridge_cpuids[] = {
 	ICPU(0x2d, pci_dev_descr_sbridge_table),	/* SANDY_BRIDGE */
 	ICPU(0x3e, pci_dev_descr_ibridge_table),	/* IVY_BRIDGE */
 	ICPU(0x3f, pci_dev_descr_haswell_table),	/* HASWELL */
 	ICPU(0x4f, pci_dev_descr_broadwell_table),	/* BROADWELL */
+	ICPU(0x56, pci_dev_descr_broadwell_table),	/* BROADWELL-DE */
 	ICPU(0x57, pci_dev_descr_knl_table),		/* KNIGHTS_LANDING */
 	{ }
 };
@@ -3398,7 +3407,7 @@ static int sbridge_probe(const struct x86_cpu_id *id)
 			 mc, mc + 1, num_mc);
 
 		sbridge_dev->mc = mc++;
-		rc = sbridge_register_mci(sbridge_dev, id - sbridge_cpuids);
+		rc = sbridge_register_mci(sbridge_dev, ptable->type);
 		if (unlikely(rc < 0))
 			goto fail1;
 	}

drivers/firmware/efi/arm-init.c

@@ -174,6 +174,7 @@ static __init void reserve_regions(void)
 {
 	efi_memory_desc_t *md;
 	u64 paddr, npages, size;
+	int resv;
 
 	if (efi_enabled(EFI_DBG))
 		pr_info("Processing EFI memory map:\n");
@@ -190,12 +191,14 @@ static __init void reserve_regions(void)
 		paddr = md->phys_addr;
 		npages = md->num_pages;
 
+		resv = is_reserve_region(md);
 		if (efi_enabled(EFI_DBG)) {
 			char buf[64];
 
-			pr_info("  0x%012llx-0x%012llx %s",
+			pr_info("  0x%012llx-0x%012llx %s%s\n",
 				paddr, paddr + (npages << EFI_PAGE_SHIFT) - 1,
-				efi_md_typeattr_format(buf, sizeof(buf), md));
+				efi_md_typeattr_format(buf, sizeof(buf), md),
+				resv ? "*" : "");
 		}
 
 		memrange_efi_to_native(&paddr, &npages);
@@ -204,14 +207,9 @@ static __init void reserve_regions(void)
 		if (is_normal_ram(md))
 			early_init_dt_add_memory_arch(paddr, size);
 
-		if (is_reserve_region(md)) {
+		if (resv)
 			memblock_mark_nomap(paddr, size);
-			if (efi_enabled(EFI_DBG))
-				pr_cont("*");
-		}
 
-		if (efi_enabled(EFI_DBG))
-			pr_cont("\n");
 	}
 
 	set_bit(EFI_MEMMAP, &efi.flags);

drivers/gpu/drm/amd/amdgpu/amdgpu.h

@@ -799,6 +799,7 @@ struct amdgpu_ring {
 	unsigned		cond_exe_offs;
 	u64				cond_exe_gpu_addr;
 	volatile u32	*cond_exe_cpu_addr;
+	int                     vmid;
 };
 
 /*
@@ -936,7 +937,8 @@ int amdgpu_vm_flush(struct amdgpu_ring *ring,
 		    unsigned vm_id, uint64_t pd_addr,
 		    uint32_t gds_base, uint32_t gds_size,
 		    uint32_t gws_base, uint32_t gws_size,
-		    uint32_t oa_base, uint32_t oa_size);
+		    uint32_t oa_base, uint32_t oa_size,
+		    bool vmid_switch);
 void amdgpu_vm_reset_id(struct amdgpu_device *adev, unsigned vm_id);
 uint64_t amdgpu_vm_map_gart(const dma_addr_t *pages_addr, uint64_t addr);
 int amdgpu_vm_update_page_directory(struct amdgpu_device *adev,

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

@@ -696,6 +696,17 @@ static uint32_t fw_type_convert(struct cgs_device *cgs_device, uint32_t fw_type)
 	return result;
 }
 
+static int amdgpu_cgs_rel_firmware(struct cgs_device *cgs_device, enum cgs_ucode_id type)
+{
+	CGS_FUNC_ADEV;
+	if ((CGS_UCODE_ID_SMU == type) || (CGS_UCODE_ID_SMU_SK == type)) {
+		release_firmware(adev->pm.fw);
+		return 0;
+	}
+	/* cannot release other firmware because they are not created by cgs */
+	return -EINVAL;
+}
+
 static int amdgpu_cgs_get_firmware_info(struct cgs_device *cgs_device,
 					enum cgs_ucode_id type,
 					struct cgs_firmware_info *info)
@@ -1125,6 +1136,7 @@ static const struct cgs_ops amdgpu_cgs_ops = {
 	amdgpu_cgs_pm_query_clock_limits,
 	amdgpu_cgs_set_camera_voltages,
 	amdgpu_cgs_get_firmware_info,
+	amdgpu_cgs_rel_firmware,
 	amdgpu_cgs_set_powergating_state,
 	amdgpu_cgs_set_clockgating_state,
 	amdgpu_cgs_get_active_displays_info,

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

@@ -827,8 +827,10 @@ static uint32_t cail_ioreg_read(struct card_info *info, uint32_t reg)
  */
 static void amdgpu_atombios_fini(struct amdgpu_device *adev)
 {
-	if (adev->mode_info.atom_context)
+	if (adev->mode_info.atom_context) {
 		kfree(adev->mode_info.atom_context->scratch);
+		kfree(adev->mode_info.atom_context->iio);
+	}
 	kfree(adev->mode_info.atom_context);
 	adev->mode_info.atom_context = NULL;
 	kfree(adev->mode_info.atom_card_info);
@@ -1325,6 +1327,11 @@ static int amdgpu_fini(struct amdgpu_device *adev)
 		adev->ip_block_status[i].valid = false;
 	}
 
+	for (i = adev->num_ip_blocks - 1; i >= 0; i--) {
+		if (adev->ip_blocks[i].funcs->late_fini)
+			adev->ip_blocks[i].funcs->late_fini((void *)adev);
+	}
+
 	return 0;
 }
 
@@ -1513,8 +1520,7 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 		amdgpu_atombios_has_gpu_virtualization_table(adev);
 
 	/* Post card if necessary */
-	if (!amdgpu_card_posted(adev) ||
-	    adev->virtualization.supports_sr_iov) {
+	if (!amdgpu_card_posted(adev)) {
 		if (!adev->bios) {
 			dev_err(adev->dev, "Card not posted and no BIOS - ignoring\n");
 			return -EINVAL;

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

@@ -122,6 +122,7 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
 	bool skip_preamble, need_ctx_switch;
 	unsigned patch_offset = ~0;
 	struct amdgpu_vm *vm;
+	int vmid = 0, old_vmid = ring->vmid;
 	struct fence *hwf;
 	uint64_t ctx;
 
@@ -135,9 +136,11 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
 	if (job) {
 		vm = job->vm;
 		ctx = job->ctx;
+		vmid = job->vm_id;
 	} else {
 		vm = NULL;
 		ctx = 0;
+		vmid = 0;
 	}
 
 	if (!ring->ready) {
@@ -163,7 +166,8 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
 		r = amdgpu_vm_flush(ring, job->vm_id, job->vm_pd_addr,
 				    job->gds_base, job->gds_size,
 				    job->gws_base, job->gws_size,
-				    job->oa_base, job->oa_size);
+				    job->oa_base, job->oa_size,
+				    (ring->current_ctx == ctx) && (old_vmid != vmid));
 		if (r) {
 			amdgpu_ring_undo(ring);
 			return r;
@@ -180,7 +184,6 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
 	need_ctx_switch = ring->current_ctx != ctx;
 	for (i = 0; i < num_ibs; ++i) {
 		ib = &ibs[i];
-
 		/* drop preamble IBs if we don't have a context switch */
 		if ((ib->flags & AMDGPU_IB_FLAG_PREAMBLE) && skip_preamble)
 			continue;
@@ -188,6 +191,7 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
 		amdgpu_ring_emit_ib(ring, ib, job ? job->vm_id : 0,
 				    need_ctx_switch);
 		need_ctx_switch = false;
+		ring->vmid = vmid;
 	}
 
 	if (ring->funcs->emit_hdp_invalidate)
@@ -198,6 +202,7 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
 		dev_err(adev->dev, "failed to emit fence (%d)\n", r);
 		if (job && job->vm_id)
 			amdgpu_vm_reset_id(adev, job->vm_id);
+		ring->vmid = old_vmid;
 		amdgpu_ring_undo(ring);
 		return r;
 	}

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

@@ -183,13 +183,6 @@ static int amdgpu_pp_sw_fini(void *handle)
 	if (ret)
 		return ret;
 
-#ifdef CONFIG_DRM_AMD_POWERPLAY
-	if (adev->pp_enabled) {
-		amdgpu_pm_sysfs_fini(adev);
-		amd_powerplay_fini(adev->powerplay.pp_handle);
-	}
-#endif
-
 	return ret;
 }
 
@@ -223,6 +216,22 @@ static int amdgpu_pp_hw_fini(void *handle)
 	return ret;
 }
 
+static void amdgpu_pp_late_fini(void *handle)
+{
+#ifdef CONFIG_DRM_AMD_POWERPLAY
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+
+	if (adev->pp_enabled) {
+		amdgpu_pm_sysfs_fini(adev);
+		amd_powerplay_fini(adev->powerplay.pp_handle);
+	}
+
+	if (adev->powerplay.ip_funcs->late_fini)
+		adev->powerplay.ip_funcs->late_fini(
+			  adev->powerplay.pp_handle);
+#endif
+}
+
 static int amdgpu_pp_suspend(void *handle)
 {
 	int ret = 0;
@@ -311,6 +320,7 @@ const struct amd_ip_funcs amdgpu_pp_ip_funcs = {
 	.sw_fini = amdgpu_pp_sw_fini,
 	.hw_init = amdgpu_pp_hw_init,
 	.hw_fini = amdgpu_pp_hw_fini,
+	.late_fini = amdgpu_pp_late_fini,
 	.suspend = amdgpu_pp_suspend,
 	.resume = amdgpu_pp_resume,
 	.is_idle = amdgpu_pp_is_idle,

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

@@ -343,6 +343,7 @@ void amdgpu_ring_fini(struct amdgpu_ring *ring)
 	ring->ring = NULL;
 	ring->ring_obj = NULL;
 
+	amdgpu_wb_free(ring->adev, ring->cond_exe_offs);
 	amdgpu_wb_free(ring->adev, ring->fence_offs);
 	amdgpu_wb_free(ring->adev, ring->rptr_offs);
 	amdgpu_wb_free(ring->adev, ring->wptr_offs);

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

@@ -115,6 +115,7 @@ int amdgpu_sa_bo_manager_start(struct amdgpu_device *adev,
 		return r;
 	}
 	r = amdgpu_bo_kmap(sa_manager->bo, &sa_manager->cpu_ptr);
+	memset(sa_manager->cpu_ptr, 0, sa_manager->size);
 	amdgpu_bo_unreserve(sa_manager->bo);
 	return r;
 }

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

@@ -253,19 +253,20 @@ int amdgpu_uvd_sw_fini(struct amdgpu_device *adev)
 {
 	int r;
 
-	if (adev->uvd.vcpu_bo == NULL)
-		return 0;
+	kfree(adev->uvd.saved_bo);
 
 	amd_sched_entity_fini(&adev->uvd.ring.sched, &adev->uvd.entity);
 
-	r = amdgpu_bo_reserve(adev->uvd.vcpu_bo, false);
-	if (!r) {
-		amdgpu_bo_kunmap(adev->uvd.vcpu_bo);
-		amdgpu_bo_unpin(adev->uvd.vcpu_bo);
-		amdgpu_bo_unreserve(adev->uvd.vcpu_bo);
-	}
+	if (adev->uvd.vcpu_bo) {
+		r = amdgpu_bo_reserve(adev->uvd.vcpu_bo, false);
+		if (!r) {
+			amdgpu_bo_kunmap(adev->uvd.vcpu_bo);
+			amdgpu_bo_unpin(adev->uvd.vcpu_bo);
+			amdgpu_bo_unreserve(adev->uvd.vcpu_bo);
+		}
 
-	amdgpu_bo_unref(&adev->uvd.vcpu_bo);
+		amdgpu_bo_unref(&adev->uvd.vcpu_bo);
+	}
 
 	amdgpu_ring_fini(&adev->uvd.ring);
 

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

@@ -298,7 +298,8 @@ int amdgpu_vm_flush(struct amdgpu_ring *ring,
 		    unsigned vm_id, uint64_t pd_addr,
 		    uint32_t gds_base, uint32_t gds_size,
 		    uint32_t gws_base, uint32_t gws_size,
-		    uint32_t oa_base, uint32_t oa_size)
+		    uint32_t oa_base, uint32_t oa_size,
+		    bool vmid_switch)
 {
 	struct amdgpu_device *adev = ring->adev;
 	struct amdgpu_vm_id *id = &adev->vm_manager.ids[vm_id];
@@ -312,8 +313,7 @@ int amdgpu_vm_flush(struct amdgpu_ring *ring,
 	int r;
 
 	if (ring->funcs->emit_pipeline_sync && (
-	    pd_addr != AMDGPU_VM_NO_FLUSH || gds_switch_needed ||
-		    ring->type == AMDGPU_RING_TYPE_COMPUTE))
+	    pd_addr != AMDGPU_VM_NO_FLUSH || gds_switch_needed || vmid_switch))
 		amdgpu_ring_emit_pipeline_sync(ring);
 
 	if (ring->funcs->emit_vm_flush &&

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

@@ -6221,6 +6221,9 @@ static int ci_dpm_sw_fini(void *handle)
 	ci_dpm_fini(adev);
 	mutex_unlock(&adev->pm.mutex);
 
+	release_firmware(adev->pm.fw);
+	adev->pm.fw = NULL;
+
 	return 0;
 }
 

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

@@ -66,6 +66,16 @@ MODULE_FIRMWARE("radeon/mullins_sdma1.bin");
 
 u32 amdgpu_cik_gpu_check_soft_reset(struct amdgpu_device *adev);
 
+
+static void cik_sdma_free_microcode(struct amdgpu_device *adev)
+{
+	int i;
+	for (i = 0; i < adev->sdma.num_instances; i++) {
+			release_firmware(adev->sdma.instance[i].fw);
+			adev->sdma.instance[i].fw = NULL;
+	}
+}
+
 /*
  * sDMA - System DMA
  * Starting with CIK, the GPU has new asynchronous
@@ -419,6 +429,8 @@ static int cik_sdma_gfx_resume(struct amdgpu_device *adev)
 		/* Initialize the ring buffer's read and write pointers */
 		WREG32(mmSDMA0_GFX_RB_RPTR + sdma_offsets[i], 0);
 		WREG32(mmSDMA0_GFX_RB_WPTR + sdma_offsets[i], 0);
+		WREG32(mmSDMA0_GFX_IB_RPTR + sdma_offsets[i], 0);
+		WREG32(mmSDMA0_GFX_IB_OFFSET + sdma_offsets[i], 0);
 
 		/* set the wb address whether it's enabled or not */
 		WREG32(mmSDMA0_GFX_RB_RPTR_ADDR_HI + sdma_offsets[i],
@@ -446,7 +458,12 @@ static int cik_sdma_gfx_resume(struct amdgpu_device *adev)
 		WREG32(mmSDMA0_GFX_IB_CNTL + sdma_offsets[i], ib_cntl);
 
 		ring->ready = true;
+	}
+
+	cik_sdma_enable(adev, true);
 
+	for (i = 0; i < adev->sdma.num_instances; i++) {
+		ring = &adev->sdma.instance[i].ring;
 		r = amdgpu_ring_test_ring(ring);
 		if (r) {
 			ring->ready = false;
@@ -529,8 +546,8 @@ static int cik_sdma_start(struct amdgpu_device *adev)
 	if (r)
 		return r;
 
-	/* unhalt the MEs */
-	cik_sdma_enable(adev, true);
+	/* halt the engine before programing */
+	cik_sdma_enable(adev, false);
 
 	/* start the gfx rings and rlc compute queues */
 	r = cik_sdma_gfx_resume(adev);
@@ -998,6 +1015,7 @@ static int cik_sdma_sw_fini(void *handle)
 	for (i = 0; i < adev->sdma.num_instances; i++)
 		amdgpu_ring_fini(&adev->sdma.instance[i].ring);
 
+	cik_sdma_free_microcode(adev);
 	return 0;
 }
 

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

@@ -72,6 +72,11 @@ static int fiji_dpm_sw_init(void *handle)
 
 static int fiji_dpm_sw_fini(void *handle)
 {
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+
+	release_firmware(adev->pm.fw);
+	adev->pm.fw = NULL;
+
 	return 0;
 }
 

drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c

@@ -991,6 +991,22 @@ out:
 	return err;
 }
 
+static void gfx_v7_0_free_microcode(struct amdgpu_device *adev)
+{
+	release_firmware(adev->gfx.pfp_fw);
+	adev->gfx.pfp_fw = NULL;
+	release_firmware(adev->gfx.me_fw);
+	adev->gfx.me_fw = NULL;
+	release_firmware(adev->gfx.ce_fw);
+	adev->gfx.ce_fw = NULL;
+	release_firmware(adev->gfx.mec_fw);
+	adev->gfx.mec_fw = NULL;
+	release_firmware(adev->gfx.mec2_fw);
+	adev->gfx.mec2_fw = NULL;
+	release_firmware(adev->gfx.rlc_fw);
+	adev->gfx.rlc_fw = NULL;
+}
+
 /**
  * gfx_v7_0_tiling_mode_table_init - init the hw tiling table
  *
@@ -4489,6 +4505,7 @@ static int gfx_v7_0_sw_fini(void *handle)
 	gfx_v7_0_cp_compute_fini(adev);
 	gfx_v7_0_rlc_fini(adev);
 	gfx_v7_0_mec_fini(adev);
+	gfx_v7_0_free_microcode(adev);
 
 	return 0;
 }

drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c

@@ -836,6 +836,26 @@ err1:
 	return r;
 }
 
+
+static void gfx_v8_0_free_microcode(struct amdgpu_device *adev) {
+	release_firmware(adev->gfx.pfp_fw);
+	adev->gfx.pfp_fw = NULL;
+	release_firmware(adev->gfx.me_fw);
+	adev->gfx.me_fw = NULL;
+	release_firmware(adev->gfx.ce_fw);
+	adev->gfx.ce_fw = NULL;
+	release_firmware(adev->gfx.rlc_fw);
+	adev->gfx.rlc_fw = NULL;
+	release_firmware(adev->gfx.mec_fw);
+	adev->gfx.mec_fw = NULL;
+	if ((adev->asic_type != CHIP_STONEY) &&
+	    (adev->asic_type != CHIP_TOPAZ))
+		release_firmware(adev->gfx.mec2_fw);
+	adev->gfx.mec2_fw = NULL;
+
+	kfree(adev->gfx.rlc.register_list_format);
+}
+
 static int gfx_v8_0_init_microcode(struct amdgpu_device *adev)
 {
 	const char *chip_name;
@@ -1983,7 +2003,7 @@ static int gfx_v8_0_sw_fini(void *handle)
 
 	gfx_v8_0_rlc_fini(adev);
 
-	kfree(adev->gfx.rlc.register_list_format);
+	gfx_v8_0_free_microcode(adev);
 
 	return 0;
 }
@@ -3974,11 +3994,15 @@ static int gfx_v8_0_cp_gfx_start(struct amdgpu_device *adev)
 		amdgpu_ring_write(ring, 0x3a00161a);
 		amdgpu_ring_write(ring, 0x0000002e);
 		break;
-	case CHIP_TOPAZ:
 	case CHIP_CARRIZO:
 		amdgpu_ring_write(ring, 0x00000002);
 		amdgpu_ring_write(ring, 0x00000000);
 		break;
+	case CHIP_TOPAZ:
+		amdgpu_ring_write(ring, adev->gfx.config.num_rbs == 1 ?
+				0x00000000 : 0x00000002);
+		amdgpu_ring_write(ring, 0x00000000);
+		break;
 	case CHIP_STONEY:
 		amdgpu_ring_write(ring, 0x00000000);
 		amdgpu_ring_write(ring, 0x00000000);

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

@@ -72,6 +72,11 @@ static int iceland_dpm_sw_init(void *handle)
 
 static int iceland_dpm_sw_fini(void *handle)
 {
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+
+	release_firmware(adev->pm.fw);
+	adev->pm.fw = NULL;
+
 	return 0;
 }
 

drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c

@@ -105,6 +105,15 @@ static void sdma_v2_4_init_golden_registers(struct amdgpu_device *adev)
 	}
 }
 
+static void sdma_v2_4_free_microcode(struct amdgpu_device *adev)
+{
+	int i;
+	for (i = 0; i < adev->sdma.num_instances; i++) {
+		release_firmware(adev->sdma.instance[i].fw);
+		adev->sdma.instance[i].fw = NULL;
+	}
+}
+
 /**
  * sdma_v2_4_init_microcode - load ucode images from disk
  *
@@ -461,6 +470,8 @@ static int sdma_v2_4_gfx_resume(struct amdgpu_device *adev)
 		/* Initialize the ring buffer's read and write pointers */
 		WREG32(mmSDMA0_GFX_RB_RPTR + sdma_offsets[i], 0);
 		WREG32(mmSDMA0_GFX_RB_WPTR + sdma_offsets[i], 0);
+		WREG32(mmSDMA0_GFX_IB_RPTR + sdma_offsets[i], 0);
+		WREG32(mmSDMA0_GFX_IB_OFFSET + sdma_offsets[i], 0);
 
 		/* set the wb address whether it's enabled or not */
 		WREG32(mmSDMA0_GFX_RB_RPTR_ADDR_HI + sdma_offsets[i],
@@ -489,7 +500,11 @@ static int sdma_v2_4_gfx_resume(struct amdgpu_device *adev)
 		WREG32(mmSDMA0_GFX_IB_CNTL + sdma_offsets[i], ib_cntl);
 
 		ring->ready = true;
+	}
 
+	sdma_v2_4_enable(adev, true);
+	for (i = 0; i < adev->sdma.num_instances; i++) {
+		ring = &adev->sdma.instance[i].ring;
 		r = amdgpu_ring_test_ring(ring);
 		if (r) {
 			ring->ready = false;
@@ -580,8 +595,8 @@ static int sdma_v2_4_start(struct amdgpu_device *adev)
 			return -EINVAL;
 	}
 
-	/* unhalt the MEs */
-	sdma_v2_4_enable(adev, true);
+	/* halt the engine before programing */
+	sdma_v2_4_enable(adev, false);
 
 	/* start the gfx rings and rlc compute queues */
 	r = sdma_v2_4_gfx_resume(adev);
@@ -1012,6 +1027,7 @@ static int sdma_v2_4_sw_fini(void *handle)
 	for (i = 0; i < adev->sdma.num_instances; i++)
 		amdgpu_ring_fini(&adev->sdma.instance[i].ring);
 
+	sdma_v2_4_free_microcode(adev);
 	return 0;
 }
 

drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c

@@ -236,6 +236,15 @@ static void sdma_v3_0_init_golden_registers(struct amdgpu_device *adev)
 	}
 }
 
+static void sdma_v3_0_free_microcode(struct amdgpu_device *adev)
+{
+	int i;
+	for (i = 0; i < adev->sdma.num_instances; i++) {
+		release_firmware(adev->sdma.instance[i].fw);
+		adev->sdma.instance[i].fw = NULL;
+	}
+}
+
 /**
  * sdma_v3_0_init_microcode - load ucode images from disk
  *
@@ -672,6 +681,8 @@ static int sdma_v3_0_gfx_resume(struct amdgpu_device *adev)
 		/* Initialize the ring buffer's read and write pointers */
 		WREG32(mmSDMA0_GFX_RB_RPTR + sdma_offsets[i], 0);
 		WREG32(mmSDMA0_GFX_RB_WPTR + sdma_offsets[i], 0);
+		WREG32(mmSDMA0_GFX_IB_RPTR + sdma_offsets[i], 0);
+		WREG32(mmSDMA0_GFX_IB_OFFSET + sdma_offsets[i], 0);
 
 		/* set the wb address whether it's enabled or not */
 		WREG32(mmSDMA0_GFX_RB_RPTR_ADDR_HI + sdma_offsets[i],
@@ -711,7 +722,15 @@ static int sdma_v3_0_gfx_resume(struct amdgpu_device *adev)
 		WREG32(mmSDMA0_GFX_IB_CNTL + sdma_offsets[i], ib_cntl);
 
 		ring->ready = true;
+	}
+
+	/* unhalt the MEs */
+	sdma_v3_0_enable(adev, true);
+	/* enable sdma ring preemption */
+	sdma_v3_0_ctx_switch_enable(adev, true);
 
+	for (i = 0; i < adev->sdma.num_instances; i++) {
+		ring = &adev->sdma.instance[i].ring;
 		r = amdgpu_ring_test_ring(ring);
 		if (r) {
 			ring->ready = false;
@@ -804,10 +823,9 @@ static int sdma_v3_0_start(struct amdgpu_device *adev)
 		}
 	}
 
-	/* unhalt the MEs */
-	sdma_v3_0_enable(adev, true);
-	/* enable sdma ring preemption */
-	sdma_v3_0_ctx_switch_enable(adev, true);
+	/* disble sdma engine before programing it */
+	sdma_v3_0_ctx_switch_enable(adev, false);
+	sdma_v3_0_enable(adev, false);
 
 	/* start the gfx rings and rlc compute queues */
 	r = sdma_v3_0_gfx_resume(adev);
@@ -1247,6 +1265,7 @@ static int sdma_v3_0_sw_fini(void *handle)
 	for (i = 0; i < adev->sdma.num_instances; i++)
 		amdgpu_ring_fini(&adev->sdma.instance[i].ring);
 
+	sdma_v3_0_free_microcode(adev);
 	return 0;
 }
 

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

@@ -71,6 +71,11 @@ static int tonga_dpm_sw_init(void *handle)
 
 static int tonga_dpm_sw_fini(void *handle)
 {
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+
+	release_firmware(adev->pm.fw);
+	adev->pm.fw = NULL;
+
 	return 0;
 }
 

drivers/gpu/drm/amd/include/amd_shared.h

@@ -157,6 +157,7 @@ struct amd_ip_funcs {
 	int (*hw_init)(void *handle);
 	/* tears down the hw state */
 	int (*hw_fini)(void *handle);
+	void (*late_fini)(void *handle);
 	/* handles IP specific hw/sw changes for suspend */
 	int (*suspend)(void *handle);
 	/* handles IP specific hw/sw changes for resume */

drivers/gpu/drm/amd/include/cgs_common.h

@@ -581,6 +581,9 @@ typedef int (*cgs_get_firmware_info)(struct cgs_device *cgs_device,
 				     enum cgs_ucode_id type,
 				     struct cgs_firmware_info *info);
 
+typedef int (*cgs_rel_firmware)(struct cgs_device *cgs_device,
+					 enum cgs_ucode_id type);
+
 typedef int(*cgs_set_powergating_state)(struct cgs_device *cgs_device,
 				  enum amd_ip_block_type block_type,
 				  enum amd_powergating_state state);
@@ -645,6 +648,7 @@ struct cgs_ops {
 	cgs_set_camera_voltages_t set_camera_voltages;
 	/* Firmware Info */
 	cgs_get_firmware_info get_firmware_info;
+	cgs_rel_firmware rel_firmware;
 	/* cg pg interface*/
 	cgs_set_powergating_state set_powergating_state;
 	cgs_set_clockgating_state set_clockgating_state;
@@ -738,6 +742,8 @@ struct cgs_device
 	CGS_CALL(set_camera_voltages,dev,mask,voltages)
 #define cgs_get_firmware_info(dev, type, info)	\
 	CGS_CALL(get_firmware_info, dev, type, info)
+#define cgs_rel_firmware(dev, type)	\
+	CGS_CALL(rel_firmware, dev, type)
 #define cgs_set_powergating_state(dev, block_type, state)	\
 	CGS_CALL(set_powergating_state, dev, block_type, state)
 #define cgs_set_clockgating_state(dev, block_type, state)	\

drivers/gpu/drm/amd/powerplay/amd_powerplay.c

@@ -73,11 +73,14 @@ static int pp_sw_init(void *handle)
 
 	ret = hwmgr->hwmgr_func->backend_init(hwmgr);
 	if (ret)
-		goto err;
+		goto err1;
 
 	pr_info("amdgpu: powerplay initialized\n");
 
 	return 0;
+err1:
+	if (hwmgr->pptable_func->pptable_fini)
+		hwmgr->pptable_func->pptable_fini(hwmgr);
 err:
 	pr_err("amdgpu: powerplay initialization failed\n");
 	return ret;
@@ -100,6 +103,9 @@ static int pp_sw_fini(void *handle)
 	if (hwmgr->hwmgr_func->backend_fini != NULL)
 		ret = hwmgr->hwmgr_func->backend_fini(hwmgr);
 
+	if (hwmgr->pptable_func->pptable_fini)
+		hwmgr->pptable_func->pptable_fini(hwmgr);
+
 	return ret;
 }
 

drivers/gpu/drm/amd/powerplay/eventmgr/eventmgr.c

@@ -58,9 +58,6 @@ static void pem_fini(struct pp_eventmgr *eventmgr)
 	pem_unregister_interrupts(eventmgr);
 
 	pem_handle_event(eventmgr, AMD_PP_EVENT_UNINITIALIZE, &event_data);
-
-	if (eventmgr != NULL)
-		kfree(eventmgr);
 }
 
 int eventmgr_init(struct pp_instance *handle)

drivers/gpu/drm/amd/powerplay/hwmgr/fiji_hwmgr.c

@@ -1830,7 +1830,7 @@ static uint16_t fiji_find_closest_vddci(struct pp_hwmgr *hwmgr, uint16_t vddci)
 
 	PP_ASSERT_WITH_CODE(false,
 			"VDDCI is larger than max VDDCI in VDDCI Voltage Table!",
-			return vddci_table->entries[i].value);
+			return vddci_table->entries[i-1].value);
 }
 
 static int fiji_get_dependency_volt_by_clk(struct pp_hwmgr *hwmgr,

drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c

@@ -93,6 +93,13 @@ int hwmgr_fini(struct pp_hwmgr *hwmgr)
 	if (hwmgr == NULL || hwmgr->ps == NULL)
 		return -EINVAL;
 
+	/* do hwmgr finish*/
+	kfree(hwmgr->backend);
+
+	kfree(hwmgr->start_thermal_controller.function_list);
+
+	kfree(hwmgr->set_temperature_range.function_list);
+
 	kfree(hwmgr->ps);
 	kfree(hwmgr);
 	return 0;
@@ -462,7 +469,7 @@ uint16_t phm_find_closest_vddci(struct pp_atomctrl_voltage_table *vddci_table, u
 
 	PP_ASSERT_WITH_CODE(false,
 			"VDDCI is larger than max VDDCI in VDDCI Voltage Table!",
-			return vddci_table->entries[i].value);
+			return vddci_table->entries[i-1].value);
 }
 
 int phm_find_boot_level(void *table,

drivers/gpu/drm/amd/powerplay/hwmgr/polaris10_powertune.c

@@ -286,7 +286,7 @@ int polaris10_populate_pm_fuses(struct pp_hwmgr *hwmgr)
 
 		if (polaris10_copy_bytes_to_smc(hwmgr->smumgr, pm_fuse_table_offset,
 				(uint8_t *)&data->power_tune_table,
-				sizeof(struct SMU74_Discrete_PmFuses), data->sram_end))
+				(sizeof(struct SMU74_Discrete_PmFuses) - 92), data->sram_end))
 			PP_ASSERT_WITH_CODE(false,
 					"Attempt to download PmFuseTable Failed!",
 					return -EINVAL);

drivers/gpu/drm/amd/powerplay/hwmgr/tonga_hwmgr.c

@@ -2847,27 +2847,6 @@ static int tonga_setup_default_dpm_tables(struct pp_hwmgr *hwmgr)
 		}
 	}
 
-	/* Initialize Vddc DPM table based on allow Vddc values.  And populate corresponding std values. */
-	for (i = 0; i < allowed_vdd_sclk_table->count; i++) {
-		data->dpm_table.vddc_table.dpm_levels[i].value = allowed_vdd_mclk_table->entries[i].vddc;
-		/* tonga_hwmgr->dpm_table.VddcTable.dpm_levels[i].param1 = stdVoltageTable->entries[i].Leakage; */
-		/* param1 is for corresponding std voltage */
-		data->dpm_table.vddc_table.dpm_levels[i].enabled = 1;
-	}
-	data->dpm_table.vddc_table.count = allowed_vdd_sclk_table->count;
-
-	if (NULL != allowed_vdd_mclk_table) {
-		/* Initialize Vddci DPM table based on allow Mclk values */
-		for (i = 0; i < allowed_vdd_mclk_table->count; i++) {
-			data->dpm_table.vdd_ci_table.dpm_levels[i].value = allowed_vdd_mclk_table->entries[i].vddci;
-			data->dpm_table.vdd_ci_table.dpm_levels[i].enabled = 1;
-			data->dpm_table.mvdd_table.dpm_levels[i].value = allowed_vdd_mclk_table->entries[i].mvdd;
-			data->dpm_table.mvdd_table.dpm_levels[i].enabled = 1;
-		}
-		data->dpm_table.vdd_ci_table.count = allowed_vdd_mclk_table->count;
-		data->dpm_table.mvdd_table.count = allowed_vdd_mclk_table->count;
-	}
-
 	/* setup PCIE gen speed levels*/
 	tonga_setup_default_pcie_tables(hwmgr);
 

drivers/gpu/drm/amd/powerplay/hwmgr/tonga_processpptables.c

@@ -1040,48 +1040,44 @@ int tonga_pp_tables_uninitialize(struct pp_hwmgr *hwmgr)
 	struct phm_ppt_v1_information *pp_table_information =
 		(struct phm_ppt_v1_information *)(hwmgr->pptable);
 
-	if (NULL != hwmgr->soft_pp_table) {
-		kfree(hwmgr->soft_pp_table);
+	if (NULL != hwmgr->soft_pp_table)
 		hwmgr->soft_pp_table = NULL;
-	}
 
-	if (NULL != pp_table_information->vdd_dep_on_sclk)
-		pp_table_information->vdd_dep_on_sclk = NULL;
+	kfree(pp_table_information->vdd_dep_on_sclk);
+	pp_table_information->vdd_dep_on_sclk = NULL;
 
-	if (NULL != pp_table_information->vdd_dep_on_mclk)
-		pp_table_information->vdd_dep_on_mclk = NULL;
+	kfree(pp_table_information->vdd_dep_on_mclk);
+	pp_table_information->vdd_dep_on_mclk = NULL;
 
-	if (NULL != pp_table_information->valid_mclk_values)
-		pp_table_information->valid_mclk_values = NULL;
+	kfree(pp_table_information->valid_mclk_values);
+	pp_table_information->valid_mclk_values = NULL;
 
-	if (NULL != pp_table_information->valid_sclk_values)
-		pp_table_information->valid_sclk_values = NULL;
+	kfree(pp_table_information->valid_sclk_values);
+	pp_table_information->valid_sclk_values = NULL;
 
-	if (NULL != pp_table_information->vddc_lookup_table)
-		pp_table_information->vddc_lookup_table = NULL;
+	kfree(pp_table_information->vddc_lookup_table);
+	pp_table_information->vddc_lookup_table = NULL;
 
-	if (NULL != pp_table_information->vddgfx_lookup_table)
-		pp_table_information->vddgfx_lookup_table = NULL;
+	kfree(pp_table_information->vddgfx_lookup_table);
+	pp_table_information->vddgfx_lookup_table = NULL;
 
-	if (NULL != pp_table_information->mm_dep_table)
-		pp_table_information->mm_dep_table = NULL;
+	kfree(pp_table_information->mm_dep_table);
+	pp_table_information->mm_dep_table = NULL;
 
-	if (NULL != pp_table_information->cac_dtp_table)
-		pp_table_information->cac_dtp_table = NULL;
+	kfree(pp_table_information->cac_dtp_table);
+	pp_table_information->cac_dtp_table = NULL;
 
-	if (NULL != hwmgr->dyn_state.cac_dtp_table)
-		hwmgr->dyn_state.cac_dtp_table = NULL;
+	kfree(hwmgr->dyn_state.cac_dtp_table);
+	hwmgr->dyn_state.cac_dtp_table = NULL;
 
-	if (NULL != pp_table_information->ppm_parameter_table)
-		pp_table_information->ppm_parameter_table = NULL;
+	kfree(pp_table_information->ppm_parameter_table);
+	pp_table_information->ppm_parameter_table = NULL;
 
-	if (NULL != pp_table_information->pcie_table)
-		pp_table_information->pcie_table = NULL;
+	kfree(pp_table_information->pcie_table);
+	pp_table_information->pcie_table = NULL;
 
-	if (NULL != hwmgr->pptable) {
-		kfree(hwmgr->pptable);
-		hwmgr->pptable = NULL;
-	}
+	kfree(hwmgr->pptable);
+	hwmgr->pptable = NULL;
 
 	return result;
 }

drivers/gpu/drm/amd/powerplay/smumgr/fiji_smumgr.c

@@ -1006,10 +1006,16 @@ static int fiji_smu_init(struct pp_smumgr *smumgr)
 
 static int fiji_smu_fini(struct pp_smumgr *smumgr)
 {
+	struct fiji_smumgr *priv = (struct fiji_smumgr *)(smumgr->backend);
+
+	smu_free_memory(smumgr->device, (void *)priv->header_buffer.handle);
+
 	if (smumgr->backend) {
 		kfree(smumgr->backend);
 		smumgr->backend = NULL;
 	}
+
+	cgs_rel_firmware(smumgr->device, CGS_UCODE_ID_SMU);
 	return 0;
 }
 

drivers/gpu/drm/amd/powerplay/smumgr/polaris10_smumgr.c

@@ -469,6 +469,7 @@ int polaris10_smu_fini(struct pp_smumgr *smumgr)
 		kfree(smumgr->backend);
 		smumgr->backend = NULL;
 	}
+	cgs_rel_firmware(smumgr->device, CGS_UCODE_ID_SMU);
 	return 0;
 }
 

drivers/gpu/drm/amd/powerplay/smumgr/smumgr.c

@@ -81,6 +81,7 @@ int smum_init(struct amd_pp_init *pp_init, struct pp_instance *handle)
 
 int smum_fini(struct pp_smumgr *smumgr)
 {
+	kfree(smumgr->device);
 	kfree(smumgr);
 	return 0;
 }

drivers/gpu/drm/amd/powerplay/smumgr/tonga_smumgr.c

@@ -328,10 +328,17 @@ int tonga_write_smc_sram_dword(struct pp_smumgr *smumgr,
 
 static int tonga_smu_fini(struct pp_smumgr *smumgr)
 {
+	struct tonga_smumgr *priv = (struct tonga_smumgr *)(smumgr->backend);
+
+	smu_free_memory(smumgr->device, (void *)priv->smu_buffer.handle);
+	smu_free_memory(smumgr->device, (void *)priv->header_buffer.handle);
+
 	if (smumgr->backend != NULL) {
 		kfree(smumgr->backend);
 		smumgr->backend = NULL;
 	}
+
+	cgs_rel_firmware(smumgr->device, CGS_UCODE_ID_SMU);
 	return 0;
 }
 

drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_drv.c

@@ -42,9 +42,10 @@ static const struct regmap_config fsl_dcu_regmap_config = {
 	.reg_bits = 32,
 	.reg_stride = 4,
 	.val_bits = 32,
-	.cache_type = REGCACHE_RBTREE,
+	.cache_type = REGCACHE_FLAT,
 
 	.volatile_reg = fsl_dcu_drm_is_volatile_reg,
+	.max_register = 0x11fc,
 };
 
 static int fsl_dcu_drm_irq_init(struct drm_device *dev)

drivers/gpu/drm/msm/adreno/adreno_gpu.c

@@ -408,7 +408,7 @@ int adreno_gpu_init(struct drm_device *drm, struct platform_device *pdev,
 	}
 
 	adreno_gpu->memptrs = msm_gem_vaddr(adreno_gpu->memptrs_bo);
-	if (!adreno_gpu->memptrs) {
+	if (IS_ERR(adreno_gpu->memptrs)) {
 		dev_err(drm->dev, "could not vmap memptrs\n");
 		return -ENOMEM;
 	}

drivers/gpu/drm/msm/msm_fbdev.c

@@ -159,6 +159,10 @@ static int msm_fbdev_create(struct drm_fb_helper *helper,
 	dev->mode_config.fb_base = paddr;
 
 	fbi->screen_base = msm_gem_vaddr_locked(fbdev->bo);
+	if (IS_ERR(fbi->screen_base)) {
+		ret = PTR_ERR(fbi->screen_base);
+		goto fail_unlock;
+	}
 	fbi->screen_size = fbdev->bo->size;
 	fbi->fix.smem_start = paddr;
 	fbi->fix.smem_len = fbdev->bo->size;

drivers/gpu/drm/msm/msm_gem.c

@@ -398,6 +398,8 @@ void *msm_gem_vaddr_locked(struct drm_gem_object *obj)
 			return ERR_CAST(pages);
 		msm_obj->vaddr = vmap(pages, obj->size >> PAGE_SHIFT,
 				VM_MAP, pgprot_writecombine(PAGE_KERNEL));
+		if (msm_obj->vaddr == NULL)
+			return ERR_PTR(-ENOMEM);
 	}
 	return msm_obj->vaddr;
 }

drivers/gpu/drm/msm/msm_gem_submit.c

@@ -40,12 +40,14 @@ static struct msm_gem_submit *submit_create(struct drm_device *dev,
 
 	submit->dev = dev;
 	submit->gpu = gpu;
+	submit->fence = NULL;
 	submit->pid = get_pid(task_pid(current));
 
 	/* initially, until copy_from_user() and bo lookup succeeds: */
 	submit->nr_bos = 0;
 	submit->nr_cmds = 0;
 
+	INIT_LIST_HEAD(&submit->node);
 	INIT_LIST_HEAD(&submit->bo_list);
 	ww_acquire_init(&submit->ticket, &reservation_ww_class);
 
@@ -75,6 +77,11 @@ static int submit_lookup_objects(struct msm_gem_submit *submit,
 		void __user *userptr =
 			u64_to_user_ptr(args->bos + (i * sizeof(submit_bo)));
 
+		/* make sure we don't have garbage flags, in case we hit
+		 * error path before flags is initialized:
+		 */
+		submit->bos[i].flags = 0;
+
 		ret = copy_from_user(&submit_bo, userptr, sizeof(submit_bo));
 		if (ret) {
 			ret = -EFAULT;

drivers/gpu/drm/msm/msm_rd.c

@@ -312,6 +312,9 @@ void msm_rd_dump_submit(struct msm_gem_submit *submit)
 		struct msm_gem_object *obj = submit->bos[idx].obj;
 		const char *buf = msm_gem_vaddr_locked(&obj->base);
 
+		if (IS_ERR(buf))
+			continue;
+
 		buf += iova - submit->bos[idx].iova;
 
 		rd_write_section(rd, RD_GPUADDR,

drivers/gpu/drm/msm/msm_ringbuffer.c

@@ -40,6 +40,10 @@ struct msm_ringbuffer *msm_ringbuffer_new(struct msm_gpu *gpu, int size)
 	}
 
 	ring->start = msm_gem_vaddr_locked(ring->bo);
+	if (IS_ERR(ring->start)) {
+		ret = PTR_ERR(ring->start);
+		goto fail;
+	}
 	ring->end   = ring->start + (size / 4);
 	ring->cur   = ring->start;
 

drivers/gpu/drm/nouveau/include/nvkm/core/device.h

@@ -16,9 +16,9 @@ enum nvkm_devidx {
 	NVKM_SUBDEV_MC,
 	NVKM_SUBDEV_BUS,
 	NVKM_SUBDEV_TIMER,
+	NVKM_SUBDEV_INSTMEM,
 	NVKM_SUBDEV_FB,
 	NVKM_SUBDEV_LTC,
-	NVKM_SUBDEV_INSTMEM,
 	NVKM_SUBDEV_MMU,
 	NVKM_SUBDEV_BAR,
 	NVKM_SUBDEV_PMU,

drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/disp.h

@@ -25,7 +25,8 @@ u16 nvbios_outp_match(struct nvkm_bios *, u16 type, u16 mask,
 		      u8 *ver, u8 *hdr, u8 *cnt, u8 *len, struct nvbios_outp *);
 
 struct nvbios_ocfg {
-	u16 match;
+	u8  proto;
+	u8  flags;
 	u16 clkcmp[2];
 };
 
@@ -33,7 +34,7 @@ u16 nvbios_ocfg_entry(struct nvkm_bios *, u16 outp, u8 idx,
 		      u8 *ver, u8 *hdr, u8 *cnt, u8 *len);
 u16 nvbios_ocfg_parse(struct nvkm_bios *, u16 outp, u8 idx,
 		      u8 *ver, u8 *hdr, u8 *cnt, u8 *len, struct nvbios_ocfg *);
-u16 nvbios_ocfg_match(struct nvkm_bios *, u16 outp, u16 type,
+u16 nvbios_ocfg_match(struct nvkm_bios *, u16 outp, u8 proto, u8 flags,
 		      u8 *ver, u8 *hdr, u8 *cnt, u8 *len, struct nvbios_ocfg *);
 u16 nvbios_oclk_match(struct nvkm_bios *, u16 cmp, u32 khz);
 #endif

drivers/gpu/drm/nouveau/nouveau_fbcon.c

@@ -552,6 +552,7 @@ nouveau_fbcon_init(struct drm_device *dev)
 	if (ret)
 		goto fini;
 
+	fbcon->helper.fbdev->pixmap.buf_align = 4;
 	return 0;
 
 fini:

drivers/gpu/drm/nouveau/nv04_fbcon.c

@@ -82,7 +82,6 @@ nv04_fbcon_imageblit(struct fb_info *info, const struct fb_image *image)
 	uint32_t fg;
 	uint32_t bg;
 	uint32_t dsize;
-	uint32_t width;
 	uint32_t *data = (uint32_t *)image->data;
 	int ret;
 
@@ -93,9 +92,6 @@ nv04_fbcon_imageblit(struct fb_info *info, const struct fb_image *image)
 	if (ret)
 		return ret;
 
-	width = ALIGN(image->width, 8);
-	dsize = ALIGN(width * image->height, 32) >> 5;
-
 	if (info->fix.visual == FB_VISUAL_TRUECOLOR ||
 	    info->fix.visual == FB_VISUAL_DIRECTCOLOR) {
 		fg = ((uint32_t *) info->pseudo_palette)[image->fg_color];
@@ -111,10 +107,11 @@ nv04_fbcon_imageblit(struct fb_info *info, const struct fb_image *image)
 			 ((image->dx + image->width) & 0xffff));
 	OUT_RING(chan, bg);
 	OUT_RING(chan, fg);
-	OUT_RING(chan, (image->height << 16) | width);
+	OUT_RING(chan, (image->height << 16) | image->width);
 	OUT_RING(chan, (image->height << 16) | image->width);
 	OUT_RING(chan, (image->dy << 16) | (image->dx & 0xffff));
 
+	dsize = ALIGN(image->width * image->height, 32) >> 5;
 	while (dsize) {
 		int iter_len = dsize > 128 ? 128 : dsize;
 

drivers/gpu/drm/nouveau/nv50_fbcon.c

@@ -95,7 +95,7 @@ nv50_fbcon_imageblit(struct fb_info *info, const struct fb_image *image)
 	struct nouveau_fbdev *nfbdev = info->par;
 	struct nouveau_drm *drm = nouveau_drm(nfbdev->dev);
 	struct nouveau_channel *chan = drm->channel;
-	uint32_t width, dwords, *data = (uint32_t *)image->data;
+	uint32_t dwords, *data = (uint32_t *)image->data;
 	uint32_t mask = ~(~0 >> (32 - info->var.bits_per_pixel));
 	uint32_t *palette = info->pseudo_palette;
 	int ret;
@@ -107,9 +107,6 @@ nv50_fbcon_imageblit(struct fb_info *info, const struct fb_image *image)
 	if (ret)
 		return ret;
 
-	width = ALIGN(image->width, 32);
-	dwords = (width * image->height) >> 5;
-
 	BEGIN_NV04(chan, NvSub2D, 0x0814, 2);
 	if (info->fix.visual == FB_VISUAL_TRUECOLOR ||
 	    info->fix.visual == FB_VISUAL_DIRECTCOLOR) {
@@ -128,6 +125,7 @@ nv50_fbcon_imageblit(struct fb_info *info, const struct fb_image *image)
 	OUT_RING(chan, 0);
 	OUT_RING(chan, image->dy);
 
+	dwords = ALIGN(image->width * image->height, 32) >> 5;
 	while (dwords) {
 		int push = dwords > 2047 ? 2047 : dwords;
 

drivers/gpu/drm/nouveau/nvc0_fbcon.c

@@ -95,7 +95,7 @@ nvc0_fbcon_imageblit(struct fb_info *info, const struct fb_image *image)
 	struct nouveau_fbdev *nfbdev = info->par;
 	struct nouveau_drm *drm = nouveau_drm(nfbdev->dev);
 	struct nouveau_channel *chan = drm->channel;
-	uint32_t width, dwords, *data = (uint32_t *)image->data;
+	uint32_t dwords, *data = (uint32_t *)image->data;
 	uint32_t mask = ~(~0 >> (32 - info->var.bits_per_pixel));
 	uint32_t *palette = info->pseudo_palette;
 	int ret;
@@ -107,9 +107,6 @@ nvc0_fbcon_imageblit(struct fb_info *info, const struct fb_image *image)
 	if (ret)
 		return ret;
 
-	width = ALIGN(image->width, 32);
-	dwords = (width * image->height) >> 5;
-
 	BEGIN_NVC0(chan, NvSub2D, 0x0814, 2);
 	if (info->fix.visual == FB_VISUAL_TRUECOLOR ||
 	    info->fix.visual == FB_VISUAL_DIRECTCOLOR) {
@@ -128,6 +125,7 @@ nvc0_fbcon_imageblit(struct fb_info *info, const struct fb_image *image)
 	OUT_RING  (chan, 0);
 	OUT_RING  (chan, image->dy);
 
+	dwords = ALIGN(image->width * image->height, 32) >> 5;
 	while (dwords) {
 		int push = dwords > 2047 ? 2047 : dwords;
 

drivers/gpu/drm/nouveau/nvkm/engine/disp/Kbuild

@@ -18,6 +18,7 @@ nvkm-y += nvkm/engine/disp/piornv50.o
 nvkm-y += nvkm/engine/disp/sornv50.o
 nvkm-y += nvkm/engine/disp/sorg94.o
 nvkm-y += nvkm/engine/disp/sorgf119.o
+nvkm-y += nvkm/engine/disp/sorgm107.o
 nvkm-y += nvkm/engine/disp/sorgm200.o
 nvkm-y += nvkm/engine/disp/dport.o