Merge remote-tracking branch 'origin/master' into perf/core

Further perf/core patches will depend on:

  d3f7b1bb2040 ("mm/gup: fix gup_fast with dynamic page table folding")

which is already in Linus' tree.

46 files changed, 363 insertions, 202 deletions

diff --git a/arch/arm/include/asm/cache.h b/arch/arm/include/asm/cache.h
index 1d65ed3a2755..e3ea34558ada 100644
--- a/arch/arm/include/asm/cache.h
+++ b/arch/arm/include/asm/cache.h
@@ -24,6 +24,6 @@
 #define ARCH_SLAB_MINALIGN 8
 #endif
 
-#define __read_mostly __attribute__((__section__(".data..read_mostly")))
+#define __read_mostly __section(".data..read_mostly")
 
 #endif
diff --git a/arch/arm/include/asm/checksum.h b/arch/arm/include/asm/checksum.h
index ed6073fee338..f0f54aef3724 100644
--- a/arch/arm/include/asm/checksum.h
+++ b/arch/arm/include/asm/checksum.h
@@ -35,23 +35,20 @@ __wsum csum_partial(const void *buff, int len, __wsum sum);
  */
 
 __wsum
-csum_partial_copy_nocheck(const void *src, void *dst, int len, __wsum sum);
+csum_partial_copy_nocheck(const void *src, void *dst, int len);
 
 __wsum
-csum_partial_copy_from_user(const void __user *src, void *dst, int len, __wsum sum, int *err_ptr);
+csum_partial_copy_from_user(const void __user *src, void *dst, int len);
 
 #define _HAVE_ARCH_COPY_AND_CSUM_FROM_USER
+#define _HAVE_ARCH_CSUM_AND_COPY
 static inline
-__wsum csum_and_copy_from_user (const void __user *src, void *dst,
-				      int len, __wsum sum, int *err_ptr)
+__wsum csum_and_copy_from_user(const void __user *src, void *dst, int len)
 {
-	if (access_ok(src, len))
-		return csum_partial_copy_from_user(src, dst, len, sum, err_ptr);
+	if (!access_ok(src, len))
+		return 0;
 
-	if (len)
-		*err_ptr = -EFAULT;
-
-	return sum;
+	return csum_partial_copy_from_user(src, dst, len);
 }
 
 /*
diff --git a/arch/arm/include/asm/cpuidle.h b/arch/arm/include/asm/cpuidle.h
index 6b2ff7243b4b..0d67ed682e07 100644
--- a/arch/arm/include/asm/cpuidle.h
+++ b/arch/arm/include/asm/cpuidle.h
@@ -42,7 +42,7 @@ struct of_cpuidle_method {
 
 #define CPUIDLE_METHOD_OF_DECLARE(name, _method, _ops)			\
 	static const struct of_cpuidle_method __cpuidle_method_of_table_##name \
-	__used __section(__cpuidle_method_of_table)			\
+	__used __section("__cpuidle_method_of_table")			\
 	= { .method = _method, .ops = _ops }
 
 extern int arm_cpuidle_suspend(int index);
diff --git a/arch/arm/include/asm/dma-contiguous.h b/arch/arm/include/asm/dma-contiguous.h
deleted file mode 100644
index d785187a6f8a..000000000000
--- a/arch/arm/include/asm/dma-contiguous.h
+++ /dev/null
@@ -1,15 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-#ifndef ASMARM_DMA_CONTIGUOUS_H
-#define ASMARM_DMA_CONTIGUOUS_H
-
-#ifdef __KERNEL__
-#ifdef CONFIG_DMA_CMA
-
-#include <linux/types.h>
-
-void dma_contiguous_early_fixup(phys_addr_t base, unsigned long size);
-
-#endif
-#endif
-
-#endif
diff --git a/arch/arm/include/asm/dma-direct.h b/arch/arm/include/asm/dma-direct.h
index 7c3001a6a775..77fcb7ee5ec9 100644
--- a/arch/arm/include/asm/dma-direct.h
+++ b/arch/arm/include/asm/dma-direct.h
@@ -2,13 +2,44 @@
 #ifndef ASM_ARM_DMA_DIRECT_H
 #define ASM_ARM_DMA_DIRECT_H 1
 
-static inline dma_addr_t __phys_to_dma(struct device *dev, phys_addr_t paddr)
+#include <asm/memory.h>
+
+/*
+ * dma_to_pfn/pfn_to_dma/virt_to_dma are architecture private
+ * functions used internally by the DMA-mapping API to provide DMA
+ * addresses. They must not be used by drivers.
+ */
+static inline dma_addr_t pfn_to_dma(struct device *dev, unsigned long pfn)
+{
+	if (dev && dev->dma_range_map)
+		pfn = PFN_DOWN(translate_phys_to_dma(dev, PFN_PHYS(pfn)));
+	return (dma_addr_t)__pfn_to_bus(pfn);
+}
+
+static inline unsigned long dma_to_pfn(struct device *dev, dma_addr_t addr)
+{
+	unsigned long pfn = __bus_to_pfn(addr);
+
+	if (dev && dev->dma_range_map)
+		pfn = PFN_DOWN(translate_dma_to_phys(dev, PFN_PHYS(pfn)));
+	return pfn;
+}
+
+static inline dma_addr_t virt_to_dma(struct device *dev, void *addr)
+{
+	if (dev)
+		return pfn_to_dma(dev, virt_to_pfn(addr));
+
+	return (dma_addr_t)__virt_to_bus((unsigned long)(addr));
+}
+
+static inline dma_addr_t phys_to_dma(struct device *dev, phys_addr_t paddr)
 {
 	unsigned int offset = paddr & ~PAGE_MASK;
 	return pfn_to_dma(dev, __phys_to_pfn(paddr)) + offset;
 }
 
-static inline phys_addr_t __dma_to_phys(struct device *dev, dma_addr_t dev_addr)
+static inline phys_addr_t dma_to_phys(struct device *dev, dma_addr_t dev_addr)
 {
 	unsigned int offset = dev_addr & ~PAGE_MASK;
 	return __pfn_to_phys(dma_to_pfn(dev, dev_addr)) + offset;
diff --git a/arch/arm/include/asm/dma-iommu.h b/arch/arm/include/asm/dma-iommu.h
index 86405cc81385..fe9ef6f79e9c 100644
--- a/arch/arm/include/asm/dma-iommu.h
+++ b/arch/arm/include/asm/dma-iommu.h
@@ -6,7 +6,6 @@
 
 #include <linux/mm_types.h>
 #include <linux/scatterlist.h>
-#include <linux/dma-debug.h>
 #include <linux/kref.h>
 
 struct dma_iommu_mapping {
diff --git a/arch/arm/include/asm/dma-mapping.h b/arch/arm/include/asm/dma-mapping.h
index bdd80ddbca34..77082246a5e1 100644
--- a/arch/arm/include/asm/dma-mapping.h
+++ b/arch/arm/include/asm/dma-mapping.h
@@ -6,9 +6,6 @@
 
 #include <linux/mm_types.h>
 #include <linux/scatterlist.h>
-#include <linux/dma-debug.h>
-
-#include <asm/memory.h>
 
 #include <xen/xen.h>
 #include <asm/xen/hypervisor.h>
@@ -23,74 +20,6 @@ static inline const struct dma_map_ops *get_arch_dma_ops(struct bus_type *bus)
 	return NULL;
 }
 
-#ifdef __arch_page_to_dma
-#error Please update to __arch_pfn_to_dma
-#endif
-
-/*
- * dma_to_pfn/pfn_to_dma/dma_to_virt/virt_to_dma are architecture private
- * functions used internally by the DMA-mapping API to provide DMA
- * addresses. They must not be used by drivers.
- */
-#ifndef __arch_pfn_to_dma
-static inline dma_addr_t pfn_to_dma(struct device *dev, unsigned long pfn)
-{
-	if (dev)
-		pfn -= dev->dma_pfn_offset;
-	return (dma_addr_t)__pfn_to_bus(pfn);
-}
-
-static inline unsigned long dma_to_pfn(struct device *dev, dma_addr_t addr)
-{
-	unsigned long pfn = __bus_to_pfn(addr);
-
-	if (dev)
-		pfn += dev->dma_pfn_offset;
-
-	return pfn;
-}
-
-static inline void *dma_to_virt(struct device *dev, dma_addr_t addr)
-{
-	if (dev) {
-		unsigned long pfn = dma_to_pfn(dev, addr);
-
-		return phys_to_virt(__pfn_to_phys(pfn));
-	}
-
-	return (void *)__bus_to_virt((unsigned long)addr);
-}
-
-static inline dma_addr_t virt_to_dma(struct device *dev, void *addr)
-{
-	if (dev)
-		return pfn_to_dma(dev, virt_to_pfn(addr));
-
-	return (dma_addr_t)__virt_to_bus((unsigned long)(addr));
-}
-
-#else
-static inline dma_addr_t pfn_to_dma(struct device *dev, unsigned long pfn)
-{
-	return __arch_pfn_to_dma(dev, pfn);
-}
-
-static inline unsigned long dma_to_pfn(struct device *dev, dma_addr_t addr)
-{
-	return __arch_dma_to_pfn(dev, addr);
-}
-
-static inline void *dma_to_virt(struct device *dev, dma_addr_t addr)
-{
-	return __arch_dma_to_virt(dev, addr);
-}
-
-static inline dma_addr_t virt_to_dma(struct device *dev, void *addr)
-{
-	return __arch_virt_to_dma(dev, addr);
-}
-#endif
-
 /**
  * arm_dma_alloc - allocate consistent memory for DMA
  * @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
diff --git a/arch/arm/include/asm/efi.h b/arch/arm/include/asm/efi.h
index 5dcf3c6011b7..3ee4f4381985 100644
--- a/arch/arm/include/asm/efi.h
+++ b/arch/arm/include/asm/efi.h
@@ -66,25 +66,24 @@ static inline void efifb_setup_from_dmi(struct screen_info *si, const char *opt)
 #define MAX_UNCOMP_KERNEL_SIZE	SZ_32M
 
 /*
- * The kernel zImage should preferably be located between 32 MB and 128 MB
- * from the base of DRAM. The min address leaves space for a maximal size
- * uncompressed image, and the max address is due to how the zImage decompressor
- * picks a destination address.
+ * phys-to-virt patching requires that the physical to virtual offset fits
+ * into the immediate field of an add/sub instruction, which comes down to the
+ * 24 least significant bits being zero, and so the offset should be a multiple
+ * of 16 MB. Since PAGE_OFFSET itself is a multiple of 16 MB, the physical
+ * base should be aligned to 16 MB as well.
  */
-#define ZIMAGE_OFFSET_LIMIT	SZ_128M
-#define MIN_ZIMAGE_OFFSET	MAX_UNCOMP_KERNEL_SIZE
+#define EFI_PHYS_ALIGN		SZ_16M
 
-/* on ARM, the FDT should be located in the first 128 MB of RAM */
-static inline unsigned long efi_get_max_fdt_addr(unsigned long dram_base)
+/* on ARM, the FDT should be located in a lowmem region */
+static inline unsigned long efi_get_max_fdt_addr(unsigned long image_addr)
 {
-	return dram_base + ZIMAGE_OFFSET_LIMIT;
+	return round_down(image_addr, EFI_PHYS_ALIGN) + SZ_512M;
 }
 
 /* on ARM, the initrd should be loaded in a lowmem region */
-static inline unsigned long efi_get_max_initrd_addr(unsigned long dram_base,
-						    unsigned long image_addr)
+static inline unsigned long efi_get_max_initrd_addr(unsigned long image_addr)
 {
-	return dram_base + SZ_512M;
+	return round_down(image_addr, EFI_PHYS_ALIGN) + SZ_512M;
 }
 
 struct efi_arm_entry_state {
diff --git a/arch/arm/include/asm/hardirq.h b/arch/arm/include/asm/hardirq.h
index 7a88f160b1fb..b95848ed2bc7 100644
--- a/arch/arm/include/asm/hardirq.h
+++ b/arch/arm/include/asm/hardirq.h
@@ -6,29 +6,12 @@
 #include <linux/threads.h>
 #include <asm/irq.h>
 
-/* number of IPIS _not_ including IPI_CPU_BACKTRACE */
-#define NR_IPI	7
-
 typedef struct {
 	unsigned int __softirq_pending;
-#ifdef CONFIG_SMP
-	unsigned int ipi_irqs[NR_IPI];
-#endif
 } ____cacheline_aligned irq_cpustat_t;
 
 #include <linux/irq_cpustat.h>	/* Standard mappings for irq_cpustat_t above */
 
-#define __inc_irq_stat(cpu, member)	__IRQ_STAT(cpu, member)++
-#define __get_irq_stat(cpu, member)	__IRQ_STAT(cpu, member)
-
-#ifdef CONFIG_SMP
-u64 smp_irq_stat_cpu(unsigned int cpu);
-#else
-#define smp_irq_stat_cpu(cpu)	0
-#endif
-
-#define arch_irq_stat_cpu	smp_irq_stat_cpu
-
 #define __ARCH_IRQ_EXIT_IRQS_DISABLED	1
 
 #endif /* __ASM_HARDIRQ_H */
diff --git a/arch/arm/include/asm/idmap.h b/arch/arm/include/asm/idmap.h
index aab7e8358e6a..baebb67b3512 100644
--- a/arch/arm/include/asm/idmap.h
+++ b/arch/arm/include/asm/idmap.h
@@ -6,7 +6,7 @@
 #include <linux/pgtable.h>
 
 /* Tag a function as requiring to be executed via an identity mapping. */
-#define __idmap __section(.idmap.text) noinline notrace
+#define __idmap __section(".idmap.text") noinline notrace
 
 extern pgd_t *idmap_pgd;
 
diff --git a/arch/arm/include/asm/kprobes.h b/arch/arm/include/asm/kprobes.h
index 213607a1f45c..e26a278d301a 100644
--- a/arch/arm/include/asm/kprobes.h
+++ b/arch/arm/include/asm/kprobes.h
@@ -44,20 +44,20 @@ int kprobe_exceptions_notify(struct notifier_block *self,
 			     unsigned long val, void *data);
 
 /* optinsn template addresses */
-extern __visible kprobe_opcode_t optprobe_template_entry;
-extern __visible kprobe_opcode_t optprobe_template_val;
-extern __visible kprobe_opcode_t optprobe_template_call;
-extern __visible kprobe_opcode_t optprobe_template_end;
-extern __visible kprobe_opcode_t optprobe_template_sub_sp;
-extern __visible kprobe_opcode_t optprobe_template_add_sp;
-extern __visible kprobe_opcode_t optprobe_template_restore_begin;
-extern __visible kprobe_opcode_t optprobe_template_restore_orig_insn;
-extern __visible kprobe_opcode_t optprobe_template_restore_end;
+extern __visible kprobe_opcode_t optprobe_template_entry[];
+extern __visible kprobe_opcode_t optprobe_template_val[];
+extern __visible kprobe_opcode_t optprobe_template_call[];
+extern __visible kprobe_opcode_t optprobe_template_end[];
+extern __visible kprobe_opcode_t optprobe_template_sub_sp[];
+extern __visible kprobe_opcode_t optprobe_template_add_sp[];
+extern __visible kprobe_opcode_t optprobe_template_restore_begin[];
+extern __visible kprobe_opcode_t optprobe_template_restore_orig_insn[];
+extern __visible kprobe_opcode_t optprobe_template_restore_end[];
 
 #define MAX_OPTIMIZED_LENGTH	4
 #define MAX_OPTINSN_SIZE				\
-	((unsigned long)&optprobe_template_end -	\
-	 (unsigned long)&optprobe_template_entry)
+	((unsigned long)optprobe_template_end -	\
+	 (unsigned long)optprobe_template_entry)
 #define RELATIVEJUMP_SIZE	4
 
 struct arch_optimized_insn {
diff --git a/arch/arm/include/asm/mach/arch.h b/arch/arm/include/asm/mach/arch.h
index e7df5a822cab..eec0c0bda766 100644
--- a/arch/arm/include/asm/mach/arch.h
+++ b/arch/arm/include/asm/mach/arch.h
@@ -81,7 +81,7 @@ extern const struct machine_desc __arch_info_begin[], __arch_info_end[];
 #define MACHINE_START(_type,_name)			\
 static const struct machine_desc __mach_desc_##_type	\
  __used							\
- __attribute__((__section__(".arch.info.init"))) = {	\
+ __section(".arch.info.init") = {			\
 	.nr		= MACH_TYPE_##_type,		\
 	.name		= _name,
 
@@ -91,7 +91,7 @@ static const struct machine_desc __mach_desc_##_type	\
 #define DT_MACHINE_START(_name, _namestr)		\
 static const struct machine_desc __mach_desc_##_name	\
  __used							\
- __attribute__((__section__(".arch.info.init"))) = {	\
+ __section(".arch.info.init") = {			\
 	.nr		= ~0,				\
 	.name		= _namestr,
 
diff --git a/arch/arm/include/asm/mach/pci.h b/arch/arm/include/asm/mach/pci.h
index 83d340702680..ea9bd08895b7 100644
--- a/arch/arm/include/asm/mach/pci.h
+++ b/arch/arm/include/asm/mach/pci.h
@@ -17,10 +17,8 @@ struct pci_host_bridge;
 struct device;
 
 struct hw_pci {
-	struct msi_controller *msi_ctrl;
 	struct pci_ops	*ops;
 	int		nr_controllers;
-	unsigned int	io_optional:1;
 	void		**private_data;
 	int		(*setup)(int nr, struct pci_sys_data *);
 	int		(*scan)(int nr, struct pci_host_bridge *);
@@ -28,11 +26,6 @@ struct hw_pci {
 	void		(*postinit)(void);
 	u8		(*swizzle)(struct pci_dev *dev, u8 *pin);
 	int		(*map_irq)(const struct pci_dev *dev, u8 slot, u8 pin);
-	resource_size_t (*align_resource)(struct pci_dev *dev,
-					  const struct resource *res,
-					  resource_size_t start,
-					  resource_size_t size,
-					  resource_size_t align);
 };
 
 /*
diff --git a/arch/arm/include/asm/module.lds.h b/arch/arm/include/asm/module.lds.h
new file mode 100644
index 000000000000..0e7cb4e314b4
--- /dev/null
+++ b/arch/arm/include/asm/module.lds.h
@@ -0,0 +1,7 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifdef CONFIG_ARM_MODULE_PLTS
+SECTIONS {
+	.plt : { BYTE(0) }
+	.init.plt : { BYTE(0) }
+}
+#endif
diff --git a/arch/arm/include/asm/setup.h b/arch/arm/include/asm/setup.h
index 67d20712cb48..3ae68a1b3de6 100644
--- a/arch/arm/include/asm/setup.h
+++ b/arch/arm/include/asm/setup.h
@@ -14,7 +14,7 @@
 #include <uapi/asm/setup.h>
 
 
-#define __tag __used __attribute__((__section__(".taglist.init")))
+#define __tag __used __section(".taglist.init")
 #define __tagtable(tag, fn) \
 static const struct tagtable __tagtable_##fn __tag = { tag, fn }
 
diff --git a/arch/arm/include/asm/smp.h b/arch/arm/include/asm/smp.h
index a91f21e3c5b5..5d508f5d56c4 100644
--- a/arch/arm/include/asm/smp.h
+++ b/arch/arm/include/asm/smp.h
@@ -39,11 +39,10 @@ void handle_IPI(int ipinr, struct pt_regs *regs);
  */
 extern void smp_init_cpus(void);
 
-
 /*
- * Provide a function to raise an IPI cross call on CPUs in callmap.
+ * Register IPI interrupts with the arch SMP code
  */
-extern void set_smp_cross_call(void (*)(const struct cpumask *, unsigned int));
+extern void set_smp_ipi_range(int ipi_base, int nr_ipi);
 
 /*
  * Called from platform specific assembly code, this is the
@@ -113,7 +112,7 @@ struct of_cpu_method {
 
 #define CPU_METHOD_OF_DECLARE(name, _method, _ops)			\
 	static const struct of_cpu_method __cpu_method_of_table_##name	\
-		__used __section(__cpu_method_of_table)			\
+		__used __section("__cpu_method_of_table")		\
 		= { .method = _method, .ops = _ops }
 /*
  * set platform specific SMP operations
diff --git a/arch/arm/include/asm/tcm.h b/arch/arm/include/asm/tcm.h
index b845b10fe29a..d8bd8a4b0ede 100644
--- a/arch/arm/include/asm/tcm.h
+++ b/arch/arm/include/asm/tcm.h
@@ -16,13 +16,13 @@
 #include <linux/compiler.h>
 
 /* Tag variables with this */
-#define __tcmdata __section(.tcm.data)
+#define __tcmdata __section(".tcm.data")
 /* Tag constants with this */
-#define __tcmconst __section(.tcm.rodata)
+#define __tcmconst __section(".tcm.rodata")
 /* Tag functions inside TCM called from outside TCM with this */
-#define __tcmfunc __attribute__((long_call)) __section(.tcm.text) noinline
+#define __tcmfunc __attribute__((long_call)) __section(".tcm.text") noinline
 /* Tag function inside TCM called from inside TCM  with this */
-#define __tcmlocalfunc __section(.tcm.text)
+#define __tcmlocalfunc __section(".tcm.text")
 
 void *tcm_alloc(size_t len);
 void tcm_free(void *addr, size_t len);
diff --git a/arch/arm/include/asm/tlb.h b/arch/arm/include/asm/tlb.h
index 9415222b49ad..b8cbe03ad260 100644
--- a/arch/arm/include/asm/tlb.h
+++ b/arch/arm/include/asm/tlb.h
@@ -59,6 +59,7 @@ __pmd_free_tlb(struct mmu_gather *tlb, pmd_t *pmdp, unsigned long addr)
 #ifdef CONFIG_ARM_LPAE
 	struct page *page = virt_to_page(pmdp);
 
+	pgtable_pmd_page_dtor(page);
 	tlb_remove_table(tlb, page);
 #endif
 }
diff --git a/arch/arm/include/asm/topology.h b/arch/arm/include/asm/topology.h
index e0593cf095d0..470299ee2fba 100644
--- a/arch/arm/include/asm/topology.h
+++ b/arch/arm/include/asm/topology.h
@@ -7,8 +7,13 @@
 #include <linux/cpumask.h>
 #include <linux/arch_topology.h>
 
+/* big.LITTLE switcher is incompatible with frequency invariance */
+#ifndef CONFIG_BL_SWITCHER
 /* Replace task scheduler's default frequency-invariant accounting */
+#define arch_set_freq_scale topology_set_freq_scale
 #define arch_scale_freq_capacity topology_get_freq_scale
+#define arch_scale_freq_invariant topology_scale_freq_invariant
+#endif
 
 /* Replace task scheduler's default cpu-invariant accounting */
 #define arch_scale_cpu_capacity topology_get_cpu_scale
diff --git a/arch/arm/include/asm/vmlinux.lds.h b/arch/arm/include/asm/vmlinux.lds.h
new file mode 100644
index 000000000000..4a91428c324d
--- /dev/null
+++ b/arch/arm/include/asm/vmlinux.lds.h
@@ -0,0 +1,147 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#include <asm-generic/vmlinux.lds.h>
+
+#ifdef CONFIG_HOTPLUG_CPU
+#define ARM_CPU_DISCARD(x)
+#define ARM_CPU_KEEP(x)		x
+#else
+#define ARM_CPU_DISCARD(x)	x
+#define ARM_CPU_KEEP(x)
+#endif
+
+#if (defined(CONFIG_SMP_ON_UP) && !defined(CONFIG_DEBUG_SPINLOCK)) || \
+	defined(CONFIG_GENERIC_BUG) || defined(CONFIG_JUMP_LABEL)
+#define ARM_EXIT_KEEP(x)	x
+#define ARM_EXIT_DISCARD(x)
+#else
+#define ARM_EXIT_KEEP(x)
+#define ARM_EXIT_DISCARD(x)	x
+#endif
+
+#ifdef CONFIG_MMU
+#define ARM_MMU_KEEP(x)		x
+#define ARM_MMU_DISCARD(x)
+#else
+#define ARM_MMU_KEEP(x)
+#define ARM_MMU_DISCARD(x)	x
+#endif
+
+#define PROC_INFO							\
+		. = ALIGN(4);						\
+		__proc_info_begin = .;					\
+		*(.proc.info.init)					\
+		__proc_info_end = .;
+
+#define IDMAP_TEXT							\
+		ALIGN_FUNCTION();					\
+		__idmap_text_start = .;					\
+		*(.idmap.text)						\
+		__idmap_text_end = .;					\
+
+#define ARM_DISCARD							\
+		*(.ARM.exidx.exit.text)					\
+		*(.ARM.extab.exit.text)					\
+		*(.ARM.exidx.text.exit)					\
+		*(.ARM.extab.text.exit)					\
+		ARM_CPU_DISCARD(*(.ARM.exidx.cpuexit.text))		\
+		ARM_CPU_DISCARD(*(.ARM.extab.cpuexit.text))		\
+		ARM_EXIT_DISCARD(EXIT_TEXT)				\
+		ARM_EXIT_DISCARD(EXIT_DATA)				\
+		EXIT_CALL						\
+		ARM_MMU_DISCARD(*(.text.fixup))				\
+		ARM_MMU_DISCARD(*(__ex_table))				\
+		COMMON_DISCARDS
+
+/*
+ * Sections that should stay zero sized, which is safer to explicitly
+ * check instead of blindly discarding.
+ */
+#define ARM_ASSERTS							\
+	.plt : {							\
+		*(.iplt) *(.rel.iplt) *(.iplt) *(.igot.plt)		\
+	}								\
+	ASSERT(SIZEOF(.plt) == 0,					\
+	       "Unexpected run-time procedure linkages detected!")
+
+#define ARM_DETAILS							\
+		ELF_DETAILS						\
+		.ARM.attributes 0 : { *(.ARM.attributes) }
+
+#define ARM_STUBS_TEXT							\
+		*(.gnu.warning)						\
+		*(.glue_7)						\
+		*(.glue_7t)						\
+		*(.vfp11_veneer)                                        \
+		*(.v4_bx)
+
+#define ARM_TEXT							\
+		IDMAP_TEXT						\
+		__entry_text_start = .;					\
+		*(.entry.text)						\
+		__entry_text_end = .;					\
+		IRQENTRY_TEXT						\
+		SOFTIRQENTRY_TEXT					\
+		TEXT_TEXT						\
+		SCHED_TEXT						\
+		CPUIDLE_TEXT						\
+		LOCK_TEXT						\
+		KPROBES_TEXT						\
+		ARM_STUBS_TEXT						\
+		. = ALIGN(4);						\
+		*(.got)			/* Global offset table */	\
+		ARM_CPU_KEEP(PROC_INFO)
+
+/* Stack unwinding tables */
+#define ARM_UNWIND_SECTIONS						\
+	. = ALIGN(8);							\
+	.ARM.unwind_idx : {						\
+		__start_unwind_idx = .;					\
+		*(.ARM.exidx*)						\
+		__stop_unwind_idx = .;					\
+	}								\
+	.ARM.unwind_tab : {						\
+		__start_unwind_tab = .;					\
+		*(.ARM.extab*)						\
+		__stop_unwind_tab = .;					\
+	}
+
+/*
+ * The vectors and stubs are relocatable code, and the
+ * only thing that matters is their relative offsets
+ */
+#define ARM_VECTORS							\
+	__vectors_start = .;						\
+	.vectors 0xffff0000 : AT(__vectors_start) {			\
+		*(.vectors)						\
+	}								\
+	. = __vectors_start + SIZEOF(.vectors);				\
+	__vectors_end = .;						\
+									\
+	__stubs_start = .;						\
+	.stubs ADDR(.vectors) + 0x1000 : AT(__stubs_start) {		\
+		*(.stubs)						\
+	}								\
+	. = __stubs_start + SIZEOF(.stubs);				\
+	__stubs_end = .;						\
+									\
+	PROVIDE(vector_fiq_offset = vector_fiq - ADDR(.vectors));
+
+#define ARM_TCM								\
+	__itcm_start = ALIGN(4);					\
+	.text_itcm ITCM_OFFSET : AT(__itcm_start - LOAD_OFFSET) {	\
+		__sitcm_text = .;					\
+		*(.tcm.text)						\
+		*(.tcm.rodata)						\
+		. = ALIGN(4);						\
+		__eitcm_text = .;					\
+	}								\
+	. = __itcm_start + SIZEOF(.text_itcm);				\
+									\
+	__dtcm_start = .;						\
+	.data_dtcm DTCM_OFFSET : AT(__dtcm_start - LOAD_OFFSET) {	\
+		__sdtcm_data = .;					\
+		*(.tcm.data)						\
+		. = ALIGN(4);						\
+		__edtcm_data = .;					\
+	}								\
+	. = __dtcm_start + SIZEOF(.data_dtcm);
diff --git a/arch/arm/include/asm/xen/page.h b/arch/arm/include/asm/xen/page.h
index 31bbc803cecb..dc7f6e91aafa 100644
--- a/arch/arm/include/asm/xen/page.h
+++ b/arch/arm/include/asm/xen/page.h
@@ -1 +1,6 @@
 #include <xen/arm/page.h>
+
+static inline bool xen_kernel_unmapped_at_usr(void)
+{
+	return false;
+}
diff --git a/arch/arm/include/debug/8250.S b/arch/arm/include/debug/8250.S
index e4a036f082c2..e3692a37cede 100644
--- a/arch/arm/include/debug/8250.S
+++ b/arch/arm/include/debug/8250.S
@@ -45,10 +45,11 @@
 		bne	1002b
 		.endm
 
-		.macro	waituart,rd,rx
-#ifdef CONFIG_DEBUG_UART_8250_FLOW_CONTROL
+		.macro	waituarttxrdy,rd,rx
+		.endm
+
+		.macro	waituartcts,rd,rx
 1001:		load	\rd, [\rx, #UART_MSR << UART_SHIFT]
 		tst	\rd, #UART_MSR_CTS
 		beq	1001b
-#endif
 		.endm
diff --git a/arch/arm/include/debug/asm9260.S b/arch/arm/include/debug/asm9260.S
index 0da1eb625331..5a0ce145c44a 100644
--- a/arch/arm/include/debug/asm9260.S
+++ b/arch/arm/include/debug/asm9260.S
@@ -11,7 +11,10 @@
 		ldr	\rv, = CONFIG_DEBUG_UART_VIRT
 		.endm
 
-		.macro	waituart,rd,rx
+		.macro	waituarttxrdy,rd,rx
+		.endm
+
+		.macro	waituartcts,rd,rx
 		.endm
 
 		.macro	senduart,rd,rx
diff --git a/arch/arm/include/debug/at91.S b/arch/arm/include/debug/at91.S
index 6c91cbaaa20b..17722824e2f2 100644
--- a/arch/arm/include/debug/at91.S
+++ b/arch/arm/include/debug/at91.S
@@ -19,12 +19,15 @@
 	strb	\rd, [\rx, #(AT91_DBGU_THR)]		@ Write to Transmitter Holding Register
 	.endm
 
-	.macro	waituart,rd,rx
+	.macro	waituarttxrdy,rd,rx
 1001:	ldr	\rd, [\rx, #(AT91_DBGU_SR)]		@ Read Status Register
 	tst	\rd, #AT91_DBGU_TXRDY			@ DBGU_TXRDY = 1 when ready to transmit
 	beq	1001b
 	.endm
 
+	.macro	waituartcts,rd,rx
+	.endm
+
 	.macro	busyuart,rd,rx
 1001:	ldr	\rd, [\rx, #(AT91_DBGU_SR)]		@ Read Status Register
 	tst	\rd, #AT91_DBGU_TXEMPTY			@ DBGU_TXEMPTY = 1 when transmission complete
diff --git a/arch/arm/include/debug/bcm63xx.S b/arch/arm/include/debug/bcm63xx.S
index 06a896227396..da65abb6738d 100644
--- a/arch/arm/include/debug/bcm63xx.S
+++ b/arch/arm/include/debug/bcm63xx.S
@@ -17,12 +17,15 @@
 	strb	\rd, [\rx, #UART_FIFO_REG]
 	.endm
 
-	.macro	waituart, rd, rx
+	.macro	waituarttxrdy, rd, rx
 1001:	ldr	\rd, [\rx, #UART_IR_REG]
 	tst	\rd, #(1 << UART_IR_TXEMPTY)
 	beq	1001b
 	.endm
 
+	.macro	waituartcts, rd, rx
+	.endm
+
 	.macro	busyuart, rd, rx
 1002:	ldr	\rd, [\rx, #UART_IR_REG]
 	tst	\rd, #(1 << UART_IR_TXTRESH)
diff --git a/arch/arm/include/debug/brcmstb.S b/arch/arm/include/debug/brcmstb.S
index 132a20c4a676..0ff32ffc610c 100644
--- a/arch/arm/include/debug/brcmstb.S
+++ b/arch/arm/include/debug/brcmstb.S
@@ -32,6 +32,8 @@
 #define UARTA_7271		UARTA_7268
 #define UARTA_7278		REG_PHYS_ADDR_V7(0x40c000)
 #define UARTA_7216		UARTA_7278
+#define UARTA_72164		UARTA_7278
+#define UARTA_72165		UARTA_7278
 #define UARTA_7364		REG_PHYS_ADDR(0x40b000)
 #define UARTA_7366		UARTA_7364
 #define UARTA_74371		REG_PHYS_ADDR(0x406b00)
@@ -84,17 +86,19 @@ ARM_BE8(	rev	\rv, \rv )
 		/* Chip specific detection starts here */
 20:		checkuart(\rp, \rv, 0x33900000, 3390)
 21:		checkuart(\rp, \rv, 0x72160000, 7216)
-22:		checkuart(\rp, \rv, 0x72500000, 7250)
-23:		checkuart(\rp, \rv, 0x72550000, 7255)
-24:		checkuart(\rp, \rv, 0x72600000, 7260)
-25:		checkuart(\rp, \rv, 0x72680000, 7268)
-26:		checkuart(\rp, \rv, 0x72710000, 7271)
-27:		checkuart(\rp, \rv, 0x72780000, 7278)
-28:		checkuart(\rp, \rv, 0x73640000, 7364)
-29:		checkuart(\rp, \rv, 0x73660000, 7366)
-30:		checkuart(\rp, \rv, 0x07437100, 74371)
-31:		checkuart(\rp, \rv, 0x74390000, 7439)
-32:		checkuart(\rp, \rv, 0x74450000, 7445)
+22:		checkuart(\rp, \rv, 0x07216400, 72164)
+23:		checkuart(\rp, \rv, 0x07216500, 72165)
+24:		checkuart(\rp, \rv, 0x72500000, 7250)
+25:		checkuart(\rp, \rv, 0x72550000, 7255)
+26:		checkuart(\rp, \rv, 0x72600000, 7260)
+27:		checkuart(\rp, \rv, 0x72680000, 7268)
+28:		checkuart(\rp, \rv, 0x72710000, 7271)
+29:		checkuart(\rp, \rv, 0x72780000, 7278)
+30:		checkuart(\rp, \rv, 0x73640000, 7364)
+31:		checkuart(\rp, \rv, 0x73660000, 7366)
+32:		checkuart(\rp, \rv, 0x07437100, 74371)
+33:		checkuart(\rp, \rv, 0x74390000, 7439)
+34:		checkuart(\rp, \rv, 0x74450000, 7445)
 
 		/* No valid UART found */
 90:		mov	\rp, #0
@@ -142,7 +146,10 @@ ARM_BE8(	rev	\rd, \rd )
 		bne	1002b
 		.endm
 
-		.macro	waituart,rd,rx
+		.macro	waituarttxrdy,rd,rx
+		.endm
+
+		.macro	waituartcts,rd,rx
 		.endm
 
 /*
diff --git a/arch/arm/include/debug/clps711x.S b/arch/arm/include/debug/clps711x.S
index 774a67ac3877..a983d12a6515 100644
--- a/arch/arm/include/debug/clps711x.S
+++ b/arch/arm/include/debug/clps711x.S
@@ -20,7 +20,10 @@
 	ldr	\rp, =CLPS711X_UART_PADDR
 	.endm
 
-	.macro	waituart,rd,rx
+	.macro	waituartcts,rd,rx
+	.endm
+
+	.macro	waituarttxrdy,rd,rx
 	.endm
 
 	.macro	senduart,rd,rx
diff --git a/arch/arm/include/debug/dc21285.S b/arch/arm/include/debug/dc21285.S
index d7e8c71706ab..4ec0e5e31704 100644
--- a/arch/arm/include/debug/dc21285.S
+++ b/arch/arm/include/debug/dc21285.S
@@ -34,5 +34,8 @@
 		bne	1001b
 		.endm
 
-		.macro	waituart,rd,rx
+		.macro	waituartcts,rd,rx
+		.endm
+
+		.macro	waituarttxrdy,rd,rx
 		.endm
diff --git a/arch/arm/include/debug/digicolor.S b/arch/arm/include/debug/digicolor.S
index 256f5f4da275..443674cad76a 100644
--- a/arch/arm/include/debug/digicolor.S
+++ b/arch/arm/include/debug/digicolor.S
@@ -21,7 +21,10 @@
 		strb	\rd, [\rx, #UA0_EMI_REC]
 		.endm
 
-		.macro	waituart,rd,rx
+		.macro	waituartcts,rd,rx
+		.endm
+
+		.macro	waituarttxrdy,rd,rx
 		.endm
 
 	.macro	busyuart,rd,rx
diff --git a/arch/arm/include/debug/efm32.S b/arch/arm/include/debug/efm32.S
index 5ed5028306f4..b0083d6e31e8 100644
--- a/arch/arm/include/debug/efm32.S
+++ b/arch/arm/include/debug/efm32.S
@@ -29,7 +29,10 @@
 		strb	\rd, [\rx, #UARTn_TXDATA]
 		.endm
 
-		.macro	waituart,rd,rx
+		.macro	waituartcts,rd,rx
+		.endm
+
+		.macro	waituarttxrdy,rd,rx
 1001:		ldr	\rd, [\rx, #UARTn_STATUS]
 		tst	\rd, #UARTn_STATUS_TXBL
 		beq	1001b
diff --git a/arch/arm/include/debug/icedcc.S b/arch/arm/include/debug/icedcc.S
index 74a0dd036a17..d5e65da8a687 100644
--- a/arch/arm/include/debug/icedcc.S
+++ b/arch/arm/include/debug/icedcc.S
@@ -23,7 +23,10 @@
 		beq	1001b
 		.endm
 
-		.macro	waituart, rd, rx
+		.macro	waituartcts, rd, rx
+		.endm
+
+		.macro	waituarttxrdy, rd, rx
 		mov	\rd, #0x2000000
 1001:
 		subs	\rd, \rd, #1
@@ -47,7 +50,10 @@
 		beq	1001b
 		.endm
 
-		.macro	waituart, rd, rx
+		.macro	waituartcts, rd, rx
+		.endm
+
+		.macro	waituarttxrdy, rd, rx
 		mov	\rd, #0x10000000
 1001:
 		subs	\rd, \rd, #1
@@ -72,7 +78,10 @@
 
 		.endm
 
-		.macro	waituart, rd, rx
+		.macro	waituartcts, rd, rx
+		.endm
+
+		.macro	waituarttxrdy, rd, rx
 		mov	\rd, #0x2000000
 1001:
 		subs	\rd, \rd, #1
diff --git a/arch/arm/include/debug/imx.S b/arch/arm/include/debug/imx.S
index 1c1b9d1da4c8..bb7b9550580c 100644
--- a/arch/arm/include/debug/imx.S
+++ b/arch/arm/include/debug/imx.S
@@ -35,7 +35,10 @@
 		str	\rd, [\rx, #0x40]	@ TXDATA
 		.endm
 
-		.macro	waituart,rd,rx
+		.macro	waituartcts,rd,rx
+		.endm
+
+		.macro	waituarttxrdy,rd,rx
 		.endm
 
 		.macro	busyuart,rd,rx
diff --git a/arch/arm/include/debug/meson.S b/arch/arm/include/debug/meson.S
index 1e501a0054ae..7b60e4401225 100644
--- a/arch/arm/include/debug/meson.S
+++ b/arch/arm/include/debug/meson.S
@@ -25,7 +25,10 @@
 	beq	1002b
 	.endm
 
-	.macro	waituart,rd,rx
+	.macro	waituartcts,rd,rx
+	.endm
+
+	.macro	waituarttxrdy,rd,rx
 1001:	ldr	\rd, [\rx, #MESON_AO_UART_STATUS]
 	tst	\rd, #MESON_AO_UART_TX_FIFO_FULL
 	bne	1001b
diff --git a/arch/arm/include/debug/msm.S b/arch/arm/include/debug/msm.S
index 9405b71461da..530edc74f9a3 100644
--- a/arch/arm/include/debug/msm.S
+++ b/arch/arm/include/debug/msm.S
@@ -17,7 +17,10 @@ ARM_BE8(rev	\rd, \rd )
 	str	\rd, [\rx, #0x70]
 	.endm
 
-	.macro	waituart, rd, rx
+	.macro	waituartcts,rd,rx
+	.endm
+
+	.macro	waituarttxrdy, rd, rx
 	@ check for TX_EMT in UARTDM_SR
 	ldr	\rd, [\rx, #0x08]
 ARM_BE8(rev     \rd, \rd )
diff --git a/arch/arm/include/debug/omap2plus.S b/arch/arm/include/debug/omap2plus.S
index b5696a33ba0f..0680be6c79d3 100644
--- a/arch/arm/include/debug/omap2plus.S
+++ b/arch/arm/include/debug/omap2plus.S
@@ -75,5 +75,8 @@ omap_uart_lsr:	.word	0
 		bne	1001b
 		.endm
 
-		.macro	waituart,rd,rx
+		.macro	waituartcts,rd,rx
+		.endm
+
+		.macro	waituarttxrdy,rd,rx
 		.endm
diff --git a/arch/arm/include/debug/pl01x.S b/arch/arm/include/debug/pl01x.S
index a2a553afe7b8..0c7bfa4c10db 100644
--- a/arch/arm/include/debug/pl01x.S
+++ b/arch/arm/include/debug/pl01x.S
@@ -26,7 +26,10 @@
 		strb	\rd, [\rx, #UART01x_DR]
 		.endm
 
-		.macro	waituart,rd,rx
+		.macro	waituartcts,rd,rx
+		.endm
+
+		.macro	waituarttxrdy,rd,rx
 1001:		ldr	\rd, [\rx, #UART01x_FR]
  ARM_BE8(	rev	\rd, \rd )
 		tst	\rd, #UART01x_FR_TXFF
diff --git a/arch/arm/include/debug/renesas-scif.S b/arch/arm/include/debug/renesas-scif.S
index 25f06663a9a4..8e433e981bbe 100644
--- a/arch/arm/include/debug/renesas-scif.S
+++ b/arch/arm/include/debug/renesas-scif.S
@@ -33,7 +33,10 @@
 	ldr	\rv, =SCIF_VIRT
 	.endm
 
-	.macro	waituart, rd, rx
+	.macro	waituartcts,rd,rx
+	.endm
+
+	.macro	waituarttxrdy, rd, rx
 1001:	ldrh	\rd, [\rx, #FSR]
 	tst	\rd, #TDFE
 	beq	1001b
diff --git a/arch/arm/include/debug/sa1100.S b/arch/arm/include/debug/sa1100.S
index 6109e6058e5b..7968ea52df3d 100644
--- a/arch/arm/include/debug/sa1100.S
+++ b/arch/arm/include/debug/sa1100.S
@@ -51,7 +51,10 @@
 		str	\rd, [\rx, #UTDR]
 		.endm
 
-		.macro	waituart,rd,rx
+		.macro	waituartcts,rd,rx
+		.endm
+
+		.macro	waituarttxrdy,rd,rx
 1001:		ldr	\rd, [\rx, #UTSR1]
 		tst	\rd, #UTSR1_TNF
 		beq	1001b
diff --git a/arch/arm/include/debug/samsung.S b/arch/arm/include/debug/samsung.S
index 69201d7fb48f..ab474d564a90 100644
--- a/arch/arm/include/debug/samsung.S
+++ b/arch/arm/include/debug/samsung.S
@@ -69,7 +69,10 @@ ARM_BE8(rev \rd, \rd)
 1002:		@ exit busyuart
 	.endm
 
-	.macro	waituart,rd,rx
+	.macro	waituartcts,rd,rx
+	.endm
+
+	.macro	waituarttxrdy,rd,rx
 		ldr	\rd, [\rx, # S3C2410_UFCON]
 ARM_BE8(rev \rd, \rd)
 		tst	\rd, #S3C2410_UFCON_FIFOMODE	@ fifo enabled?
diff --git a/arch/arm/include/debug/sirf.S b/arch/arm/include/debug/sirf.S
index e73e4de0a015..3612c7b9cbe7 100644
--- a/arch/arm/include/debug/sirf.S
+++ b/arch/arm/include/debug/sirf.S
@@ -29,7 +29,10 @@
 	.macro	busyuart,rd,rx
 	.endm
 
-	.macro	waituart,rd,rx
+	.macro	waituartcts,rd,rx
+	.endm
+
+	.macro	waituarttxrdy,rd,rx
 1001:	ldr	\rd, [\rx, #SIRF_LLUART_TXFIFO_STATUS]
 	tst	\rd, #SIRF_LLUART_TXFIFO_EMPTY
 	beq	1001b
diff --git a/arch/arm/include/debug/sti.S b/arch/arm/include/debug/sti.S
index 6b42c91f217d..72d052511890 100644
--- a/arch/arm/include/debug/sti.S
+++ b/arch/arm/include/debug/sti.S
@@ -45,7 +45,10 @@
                 strb    \rd, [\rx, #ASC_TX_BUF_OFF]
                 .endm
 
-                .macro  waituart,rd,rx
+		.macro	waituartcts,rd,rx
+		.endm
+
+                .macro  waituarttxrdy,rd,rx
 1001:           ldr     \rd, [\rx, #ASC_STA_OFF]
                 tst     \rd, #ASC_STA_TX_FULL
                 bne     1001b
diff --git a/arch/arm/include/debug/stm32.S b/arch/arm/include/debug/stm32.S
index f3c4a37210ed..b6d9df30e37d 100644
--- a/arch/arm/include/debug/stm32.S
+++ b/arch/arm/include/debug/stm32.S
@@ -27,7 +27,10 @@
 	strb    \rd, [\rx, #STM32_USART_TDR_OFF]
 .endm
 
-.macro  waituart,rd,rx
+.macro	waituartcts,rd,rx
+.endm
+
+.macro  waituarttxrdy,rd,rx
 1001:	ldr	\rd, [\rx, #(STM32_USART_SR_OFF)]	@ Read Status Register
 	tst	\rd, #STM32_USART_TXE			@ TXE = 1 = tx empty
 	beq	1001b
diff --git a/arch/arm/include/debug/tegra.S b/arch/arm/include/debug/tegra.S
index 2148d0f88591..98daa7f48314 100644
--- a/arch/arm/include/debug/tegra.S
+++ b/arch/arm/include/debug/tegra.S
@@ -178,15 +178,16 @@
 1002:
 		.endm
 
-		.macro	waituart, rd, rx
-#ifdef FLOW_CONTROL
+		.macro	waituartcts, rd, rx
 		cmp	\rx, #0
 		beq	1002f
 1001:		ldrb	\rd, [\rx, #UART_MSR << UART_SHIFT]
 		tst	\rd, #UART_MSR_CTS
 		beq	1001b
 1002:
-#endif
+		.endm
+
+		.macro	waituarttxrdy,rd,rx
 		.endm
 
 /*
diff --git a/arch/arm/include/debug/vf.S b/arch/arm/include/debug/vf.S
index 854d9bd82770..035bcbf117ab 100644
--- a/arch/arm/include/debug/vf.S
+++ b/arch/arm/include/debug/vf.S
@@ -29,5 +29,8 @@
 	beq	1001b			@ wait until transmit done
 	.endm
 
-	.macro	waituart,rd,rx
+	.macro	waituartcts,rd,rx
+	.endm
+
+	.macro	waituarttxrdy,rd,rx
 	.endm
diff --git a/arch/arm/include/debug/vt8500.S b/arch/arm/include/debug/vt8500.S
index 8dc1df2d91b8..d01094fdbc8c 100644
--- a/arch/arm/include/debug/vt8500.S
+++ b/arch/arm/include/debug/vt8500.S
@@ -28,7 +28,10 @@
 	bne	1001b
 	.endm
 
-	.macro	waituart,rd,rx
+	.macro	waituartcts,rd,rx
+	.endm
+
+	.macro	waituarttxrdy,rd,rx
 	.endm
 
 #endif
diff --git a/arch/arm/include/debug/zynq.S b/arch/arm/include/debug/zynq.S
index 58d77c972fd6..5d42cc35ecf3 100644
--- a/arch/arm/include/debug/zynq.S
+++ b/arch/arm/include/debug/zynq.S
@@ -33,7 +33,10 @@
 		strb	\rd, [\rx, #UART_FIFO_OFFSET]	@ TXDATA
 		.endm
 
-		.macro	waituart,rd,rx
+		.macro	waituartcts,rd,rx
+		.endm
+
+		.macro	waituarttxrdy,rd,rx
 1001:		ldr	\rd, [\rx, #UART_SR_OFFSET]
 ARM_BE8(	rev	\rd, \rd )
 		tst	\rd, #UART_SR_TXEMPTY