Merge branch 'entry-macro-cleanup' of git://sources.calxeda.com/kernel/linux into for-armsoc

96 files changed, 219 insertions, 1709 deletions

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index a48aecc17eac..7d809b7e0504 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -186,6 +186,9 @@ config GENERIC_ISA_DMA
 config FIQ
 	bool
 
+config NEED_RET_TO_USER
+	bool
+
 config ARCH_MTD_XIP
 	bool
 
@@ -479,6 +482,7 @@ config ARCH_IOP13XX
 	select ARCH_SUPPORTS_MSI
 	select VMSPLIT_1G
 	select NEED_MACH_MEMORY_H
+	select NEED_RET_TO_USER
 	help
 	  Support for Intel's IOP13XX (XScale) family of processors.
 
@@ -486,6 +490,7 @@ config ARCH_IOP32X
 	bool "IOP32x-based"
 	depends on MMU
 	select CPU_XSCALE
+	select NEED_RET_TO_USER
 	select PLAT_IOP
 	select PCI
 	select ARCH_REQUIRE_GPIOLIB
@@ -497,6 +502,7 @@ config ARCH_IOP33X
 	bool "IOP33x-based"
 	depends on MMU
 	select CPU_XSCALE
+	select NEED_RET_TO_USER
 	select PLAT_IOP
 	select PCI
 	select ARCH_REQUIRE_GPIOLIB
diff --git a/arch/arm/include/asm/hardware/entry-macro-iomd.S b/arch/arm/include/asm/hardware/entry-macro-iomd.S
index e0af4983723f..8c215acd9b57 100644
--- a/arch/arm/include/asm/hardware/entry-macro-iomd.S
+++ b/arch/arm/include/asm/hardware/entry-macro-iomd.S
@@ -11,14 +11,6 @@
 /* IOC / IOMD based hardware */
 #include <asm/hardware/iomd.h>
 
-		.macro	disable_fiq
-		mov	r12, #ioc_base_high
-		.if	ioc_base_low
-		orr	r12, r12, #ioc_base_low
-		.endif
-		strb	r12, [r12, #0x38]	@ Disable FIQ register
-		.endm
-
 		.macro	get_irqnr_and_base, irqnr, irqstat, base, tmp
 		ldrb	\irqstat, [\base, #IOMD_IRQREQB]	@ get high priority first
 		ldr	\tmp, =irq_prio_h
diff --git a/arch/arm/include/asm/tlb.h b/arch/arm/include/asm/tlb.h
index 5d3ed7e38561..314d4664eae7 100644
--- a/arch/arm/include/asm/tlb.h
+++ b/arch/arm/include/asm/tlb.h
@@ -198,7 +198,15 @@ static inline void __pte_free_tlb(struct mmu_gather *tlb, pgtable_t pte,
 	unsigned long addr)
 {
 	pgtable_page_dtor(pte);
-	tlb_add_flush(tlb, addr);
+
+	/*
+	 * With the classic ARM MMU, a pte page has two corresponding pmd
+	 * entries, each covering 1MB.
+	 */
+	addr &= PMD_MASK;
+	tlb_add_flush(tlb, addr + SZ_1M - PAGE_SIZE);
+	tlb_add_flush(tlb, addr + SZ_1M);
+
 	tlb_remove_page(tlb, pte);
 }
 
diff --git a/arch/arm/kernel/entry-armv.S b/arch/arm/kernel/entry-armv.S
index 3a456c6c7005..22f0ed324f37 100644
--- a/arch/arm/kernel/entry-armv.S
+++ b/arch/arm/kernel/entry-armv.S
@@ -19,7 +19,9 @@
 #include <asm/glue-df.h>
 #include <asm/glue-pf.h>
 #include <asm/vfpmacros.h>
+#ifndef CONFIG_MULTI_IRQ_HANDLER
 #include <mach/entry-macro.S>
+#endif
 #include <asm/thread_notify.h>
 #include <asm/unwind.h>
 #include <asm/unistd.h>
@@ -790,7 +792,7 @@ __kuser_cmpxchg64:				@ 0xffff0f60
 	smp_dmb	arm
 	rsbs	r0, r3, #0			@ set returned val and C flag
 	ldmfd	sp!, {r4, r5, r6, r7}
-	bx	lr
+	usr_ret	lr
 
 #elif !defined(CONFIG_SMP)
 
@@ -1101,7 +1103,6 @@ __stubs_start:
  * get out of that mode without clobbering one register.
  */
 vector_fiq:
-	disable_fiq
 	subs	pc, lr, #4
 
 /*=============================================================================
diff --git a/arch/arm/kernel/entry-common.S b/arch/arm/kernel/entry-common.S
index 9fd0ba90c1d2..54ee265dd819 100644
--- a/arch/arm/kernel/entry-common.S
+++ b/arch/arm/kernel/entry-common.S
@@ -10,9 +10,15 @@
 
 #include <asm/unistd.h>
 #include <asm/ftrace.h>
-#include <mach/entry-macro.S>
 #include <asm/unwind.h>
 
+#ifdef CONFIG_NEED_RET_TO_USER
+#include <mach/entry-macro.S>
+#else
+	.macro  arch_ret_to_user, tmp1, tmp2
+	.endm
+#endif
+
 #include "entry-header.S"
 
 
diff --git a/arch/arm/kernel/perf_event_v7.c b/arch/arm/kernel/perf_event_v7.c
index 460bbbb6b885..6933244c68f9 100644
--- a/arch/arm/kernel/perf_event_v7.c
+++ b/arch/arm/kernel/perf_event_v7.c
@@ -469,6 +469,20 @@ static const unsigned armv7_a5_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
 			[C(RESULT_MISS)]	= CACHE_OP_UNSUPPORTED,
 		},
 	},
+	[C(NODE)] = {
+		[C(OP_READ)] = {
+			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
+			[C(RESULT_MISS)]	= CACHE_OP_UNSUPPORTED,
+		},
+		[C(OP_WRITE)] = {
+			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
+			[C(RESULT_MISS)]	= CACHE_OP_UNSUPPORTED,
+		},
+		[C(OP_PREFETCH)] = {
+			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
+			[C(RESULT_MISS)]	= CACHE_OP_UNSUPPORTED,
+		},
+	},
 };
 
 /*
@@ -579,6 +593,20 @@ static const unsigned armv7_a15_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
 			[C(RESULT_MISS)]	= CACHE_OP_UNSUPPORTED,
 		},
 	},
+	[C(NODE)] = {
+		[C(OP_READ)] = {
+			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
+			[C(RESULT_MISS)]	= CACHE_OP_UNSUPPORTED,
+		},
+		[C(OP_WRITE)] = {
+			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
+			[C(RESULT_MISS)]	= CACHE_OP_UNSUPPORTED,
+		},
+		[C(OP_PREFETCH)] = {
+			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
+			[C(RESULT_MISS)]	= CACHE_OP_UNSUPPORTED,
+		},
+	},
 };
 
 /*
diff --git a/arch/arm/kernel/ptrace.c b/arch/arm/kernel/ptrace.c
index e1d5e1929fbd..e33870ff0ac0 100644
--- a/arch/arm/kernel/ptrace.c
+++ b/arch/arm/kernel/ptrace.c
@@ -699,10 +699,13 @@ static int vfp_set(struct task_struct *target,
 {
 	int ret;
 	struct thread_info *thread = task_thread_info(target);
-	struct vfp_hard_struct new_vfp = thread->vfpstate.hard;
+	struct vfp_hard_struct new_vfp;
 	const size_t user_fpregs_offset = offsetof(struct user_vfp, fpregs);
 	const size_t user_fpscr_offset = offsetof(struct user_vfp, fpscr);
 
+	vfp_sync_hwstate(thread);
+	new_vfp = thread->vfpstate.hard;
+
 	ret = user_regset_copyin(&pos, &count, &kbuf, &ubuf,
 				  &new_vfp.fpregs,
 				  user_fpregs_offset,
@@ -723,9 +726,8 @@ static int vfp_set(struct task_struct *target,
 	if (ret)
 		return ret;
 
-	vfp_sync_hwstate(thread);
-	thread->vfpstate.hard = new_vfp;
 	vfp_flush_hwstate(thread);
+	thread->vfpstate.hard = new_vfp;
 
 	return 0;
 }
diff --git a/arch/arm/kernel/signal.c b/arch/arm/kernel/signal.c
index 0340224cf73c..9e617bd4a146 100644
--- a/arch/arm/kernel/signal.c
+++ b/arch/arm/kernel/signal.c
@@ -227,6 +227,8 @@ static int restore_vfp_context(struct vfp_sigframe __user *frame)
 	if (magic != VFP_MAGIC || size != VFP_STORAGE_SIZE)
 		return -EINVAL;
 
+	vfp_flush_hwstate(thread);
+
 	/*
 	 * Copy the floating point registers. There can be unused
 	 * registers see asm/hwcap.h for details.
@@ -251,9 +253,6 @@ static int restore_vfp_context(struct vfp_sigframe __user *frame)
 	__get_user_error(h->fpinst, &frame->ufp_exc.fpinst, err);
 	__get_user_error(h->fpinst2, &frame->ufp_exc.fpinst2, err);
 
-	if (!err)
-		vfp_flush_hwstate(thread);
-
 	return err ? -EFAULT : 0;
 }
 
diff --git a/arch/arm/mach-at91/include/mach/entry-macro.S b/arch/arm/mach-at91/include/mach/entry-macro.S
index 423eea0ed74c..903bf205a333 100644
--- a/arch/arm/mach-at91/include/mach/entry-macro.S
+++ b/arch/arm/mach-at91/include/mach/entry-macro.S
@@ -13,17 +13,11 @@
 #include <mach/hardware.h>
 #include <mach/at91_aic.h>
 
-	.macro	disable_fiq
-	.endm
-
 	.macro  get_irqnr_preamble, base, tmp
 	ldr	\base, =at91_aic_base		@ base virtual address of AIC peripheral
 	ldr	\base, [\base]
 	.endm
 
-	.macro  arch_ret_to_user, tmp1, tmp2
-	.endm
-
 	.macro	get_irqnr_and_base, irqnr, irqstat, base, tmp
 	ldr	\irqnr, [\base, #AT91_AIC_IVR]		@ read IRQ vector register: de-asserts nIRQ to processor (and clears interrupt)
 	ldr	\irqstat, [\base, #AT91_AIC_ISR]	@ read interrupt source number
diff --git a/arch/arm/mach-bcmring/arch.c b/arch/arm/mach-bcmring/arch.c
index 9e5e7552498c..45c97b1ee9b1 100644
--- a/arch/arm/mach-bcmring/arch.c
+++ b/arch/arm/mach-bcmring/arch.c
@@ -194,6 +194,6 @@ MACHINE_START(BCMRING, "BCMRING")
 	.init_early = bcmring_init_early,
 	.init_irq = bcmring_init_irq,
 	.timer = &bcmring_timer,
-	.init_machine = bcmring_init_machine
+	.init_machine = bcmring_init_machine,
 	.restart = bcmring_restart,
 MACHINE_END
diff --git a/arch/arm/mach-bcmring/dma.c b/arch/arm/mach-bcmring/dma.c
index 1a1a27dd5654..1024396797e1 100644
--- a/arch/arm/mach-bcmring/dma.c
+++ b/arch/arm/mach-bcmring/dma.c
@@ -33,17 +33,11 @@
 
 #include <mach/timer.h>
 
-#include <linux/mm.h>
 #include <linux/pfn.h>
 #include <linux/atomic.h>
 #include <linux/sched.h>
 #include <mach/dma.h>
 
-/* I don't quite understand why dc4 fails when this is set to 1 and DMA is enabled */
-/* especially since dc4 doesn't use kmalloc'd memory. */
-
-#define ALLOW_MAP_OF_KMALLOC_MEMORY 0
-
 /* ---- Public Variables ------------------------------------------------- */
 
 /* ---- Private Constants and Types -------------------------------------- */
@@ -53,24 +47,12 @@
 #define CONTROLLER_FROM_HANDLE(handle)    (((handle) >> 4) & 0x0f)
 #define CHANNEL_FROM_HANDLE(handle)       ((handle) & 0x0f)
 
-#define DMA_MAP_DEBUG   0
-
-#if DMA_MAP_DEBUG
-#   define  DMA_MAP_PRINT(fmt, args...)   printk("%s: " fmt, __func__,  ## args)
-#else