Merge branches 'x86/mmio', 'x86/delay', 'x86/idle', 'x86/oprofile',...

Merge branches 'x86/mmio', 'x86/delay', 'x86/idle', 'x86/oprofile', 'x86/debug', 'x86/ptrace' and 'x86/amd-iommu' into x86/devel

Documentation/kernel-parameters.txt

@@ -271,6 +271,17 @@ and is between 256 and 4096 characters. It is defined in the file
 	aic79xx=	[HW,SCSI]
 			See Documentation/scsi/aic79xx.txt.
 
+	amd_iommu=	[HW,X86-84]
+			Pass parameters to the AMD IOMMU driver in the system.
+			Possible values are:
+			isolate - enable device isolation (each device, as far
+			          as possible, will get its own protection
+			          domain)
+	amd_iommu_size= [HW,X86-64]
+			Define the size of the aperture for the AMD IOMMU
+			driver. Possible values are:
+			'32M', '64M' (default), '128M', '256M', '512M', '1G'
+
 	amijoy.map=	[HW,JOY] Amiga joystick support
 			Map of devices attached to JOY0DAT and JOY1DAT
 			Format: <a>,<b>

MAINTAINERS

@@ -376,6 +376,12 @@ L:	linux-geode@lists.infradead.org (moderated for non-subscribers)
 W:	http://www.amd.com/us-en/ConnectivitySolutions/TechnicalResources/0,,50_2334_2452_11363,00.html
 S:	Supported
 
+AMD IOMMU (AMD-VI)
+P:	Joerg Roedel
+M:	joerg.roedel@amd.com
+L:	iommu@lists.linux-foundation.org
+S:	Supported
+
 AMS (Apple Motion Sensor) DRIVER
 P:	Stelian Pop
 M:	stelian@popies.net

arch/x86/Kconfig

@@ -533,6 +533,21 @@ config CALGARY_IOMMU_ENABLED_BY_DEFAULT
 	  Calgary anyway, pass 'iommu=calgary' on the kernel command line.
 	  If unsure, say Y.
 
+config AMD_IOMMU
+	bool "AMD IOMMU support"
+	select SWIOTLB
+	depends on X86_64 && PCI && ACPI
+	help
+	  With this option you can enable support for AMD IOMMU hardware in
+	  your system. An IOMMU is a hardware component which provides
+	  remapping of DMA memory accesses from devices. With an AMD IOMMU you
+	  can isolate the the DMA memory of different devices and protect the
+	  system from misbehaving device drivers or hardware.
+
+	  You can find out if your system has an AMD IOMMU if you look into
+	  your BIOS for an option to enable it or if you have an IVRS ACPI
+	  table.
+
 # need this always selected by IOMMU for the VIA workaround
 config SWIOTLB
 	bool

arch/x86/kernel/Makefile

@@ -99,6 +99,7 @@ ifeq ($(CONFIG_X86_64),y)
 
         obj-$(CONFIG_GART_IOMMU)	+= pci-gart_64.o aperture_64.o
         obj-$(CONFIG_CALGARY_IOMMU)	+= pci-calgary_64.o tce_64.o
+        obj-$(CONFIG_AMD_IOMMU)		+= amd_iommu_init.o amd_iommu.o
         obj-$(CONFIG_SWIOTLB)		+= pci-swiotlb_64.o
 
         obj-$(CONFIG_PCI_MMCONFIG)	+= mmconf-fam10h_64.o

arch/x86/kernel/entry_64.S

@@ -925,11 +925,11 @@ error_kernelspace:
 	   iret run with kernel gs again, so don't set the user space flag.
 	   B stepping K8s sometimes report an truncated RIP for IRET 
 	   exceptions returning to compat mode. Check for these here too. */
-	leaq irq_return(%rip),%rbp
-	cmpq %rbp,RIP(%rsp) 
+	leaq irq_return(%rip),%rcx
+	cmpq %rcx,RIP(%rsp)
 	je   error_swapgs
-	movl %ebp,%ebp	/* zero extend */
-	cmpq %rbp,RIP(%rsp) 
+	movl %ecx,%ecx	/* zero extend */
+	cmpq %rcx,RIP(%rsp)
 	je   error_swapgs
 	cmpq $gs_change,RIP(%rsp)
         je   error_swapgs

arch/x86/kernel/i387.c

@@ -162,7 +162,7 @@ int xfpregs_get(struct task_struct *target, const struct user_regset *regset,
 	int ret;
 
 	if (!cpu_has_fxsr)
-		return -EIO;
+		return -ENODEV;
 
 	ret = init_fpu(target);
 	if (ret)
@@ -179,7 +179,7 @@ int xfpregs_set(struct task_struct *target, const struct user_regset *regset,
 	int ret;
 
 	if (!cpu_has_fxsr)
-		return -EIO;
+		return -ENODEV;
 
 	ret = init_fpu(target);
 	if (ret)

arch/x86/kernel/pci-dma.c

@@ -7,6 +7,7 @@
 #include <asm/dma.h>
 #include <asm/gart.h>
 #include <asm/calgary.h>
+#include <asm/amd_iommu.h>
 
 int forbid_dac __read_mostly;
 EXPORT_SYMBOL(forbid_dac);
@@ -123,6 +124,8 @@ void __init pci_iommu_alloc(void)
 
 	detect_intel_iommu();
 
+	amd_iommu_detect();
+
 #ifdef CONFIG_SWIOTLB
 	pci_swiotlb_init();
 #endif
@@ -503,6 +506,8 @@ static int __init pci_iommu_init(void)
 
 	intel_iommu_init();
 
+	amd_iommu_init();
+
 #ifdef CONFIG_GART_IOMMU
 	gart_iommu_init();
 #endif

arch/x86/kernel/ptrace.c

@@ -943,13 +943,13 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 		return copy_regset_to_user(child, &user_x86_32_view,
 					   REGSET_XFP,
 					   0, sizeof(struct user_fxsr_struct),
-					   datap);
+					   datap) ? -EIO : 0;
 
 	case PTRACE_SETFPXREGS:	/* Set the child extended FPU state. */
 		return copy_regset_from_user(child, &user_x86_32_view,
 					     REGSET_XFP,
 					     0, sizeof(struct user_fxsr_struct),
-					     datap);
+					     datap) ? -EIO : 0;
 #endif
 
 #if defined CONFIG_X86_32 || defined CONFIG_IA32_EMULATION

arch/x86/kernel/time_64.c

@@ -123,6 +123,8 @@ void __init time_init(void)
 		(boot_cpu_data.x86_vendor == X86_VENDOR_AMD))
 		cpu_khz = calculate_cpu_khz();
 
+	lpj_fine = ((unsigned long)tsc_khz * 1000)/HZ;
+
 	if (unsynchronized_tsc())
 		mark_tsc_unstable("TSCs unsynchronized");
 

arch/x86/kernel/tsc_32.c

 #include <linux/sched.h>
 #include <linux/clocksource.h>
 #include <linux/workqueue.h>
+#include <linux/delay.h>
 #include <linux/cpufreq.h>
 #include <linux/jiffies.h>
 #include <linux/init.h>
@@ -403,6 +404,7 @@ static inline void check_geode_tsc_reliable(void) { }
 void __init tsc_init(void)
 {
 	int cpu;
+	u64 lpj;
 
 	if (!cpu_has_tsc || tsc_disabled > 0)
 		return;
@@ -415,6 +417,10 @@ void __init tsc_init(void)
 		return;
 	}
 
+	lpj = ((u64)tsc_khz * 1000);
+	do_div(lpj, HZ);
+	lpj_fine = lpj;
+
 	/* now allow native_sched_clock() to use rdtsc */
 	tsc_disabled = 0;
 

arch/x86/lib/delay_32.c

@@ -3,6 +3,7 @@
  *
  *	Copyright (C) 1993 Linus Torvalds
  *	Copyright (C) 1997 Martin Mares <mj@atrey.karlin.mff.cuni.cz>
+ *	Copyright (C) 2008 Jiri Hladky <hladky _dot_ jiri _at_ gmail _dot_ com>
  *
  *	The __delay function must _NOT_ be inlined as its execution time
  *	depends wildly on alignment on many x86 processors. The additional
@@ -28,16 +29,22 @@
 /* simple loop based delay: */
 static void delay_loop(unsigned long loops)
 {
-	int d0;
-
 	__asm__ __volatile__(
-		"\tjmp 1f\n"
-		".align 16\n"
-		"1:\tjmp 2f\n"
-		".align 16\n"
-		"2:\tdecl %0\n\tjns 2b"
-		:"=&a" (d0)
-		:"0" (loops));
+		"	test %0,%0	\n"
+		"	jz 3f		\n"
+		"	jmp 1f		\n"
+
+		".align 16		\n"
+		"1:	jmp 2f		\n"
+
+		".align 16		\n"
+		"2:	decl %0		\n"
+		"	jnz 2b		\n"
+		"3:	decl %0		\n"
+
+		: /* we don't need output */
+		:"a" (loops)
+	);
 }
 
 /* TSC based delay: */

arch/x86/mm/fault.c

@@ -392,11 +392,7 @@ static void show_fault_oops(struct pt_regs *regs, unsigned long error_code,
 		printk(KERN_CONT "NULL pointer dereference");
 	else
 		printk(KERN_CONT "paging request");
-#ifdef CONFIG_X86_32
-	printk(KERN_CONT " at %08lx\n", address);
-#else
-	printk(KERN_CONT " at %016lx\n", address);
-#endif
+	printk(KERN_CONT " at %p\n", (void *) address);
 	printk(KERN_ALERT "IP:");
 	printk_address(regs->ip, 1);
 	dump_pagetable(address);
@@ -796,14 +792,10 @@ void __kprobes do_page_fault(struct pt_regs *regs, unsigned long error_code)
 		if (show_unhandled_signals && unhandled_signal(tsk, SIGSEGV) &&
 		    printk_ratelimit()) {
 			printk(
-#ifdef CONFIG_X86_32
-			"%s%s[%d]: segfault at %lx ip %08lx sp %08lx error %lx",
-#else
-			"%s%s[%d]: segfault at %lx ip %lx sp %lx error %lx",
-#endif
+			"%s%s[%d]: segfault at %lx ip %p sp %p error %lx",
 			task_pid_nr(tsk) > 1 ? KERN_INFO : KERN_EMERG,
-			tsk->comm, task_pid_nr(tsk), address, regs->ip,
-			regs->sp, error_code);
+			tsk->comm, task_pid_nr(tsk), address,
+			(void *) regs->ip, (void *) regs->sp, error_code);
 			print_vma_addr(" in ", regs->ip);
 			printk("\n");
 		}

arch/x86/oprofile/nmi_int.c

@@ -269,12 +269,13 @@ static void nmi_cpu_shutdown(void *dummy)
 
 static void nmi_shutdown(void)
 {
-	struct op_msrs *msrs = &__get_cpu_var(cpu_msrs);
+	struct op_msrs *msrs = &get_cpu_var(cpu_msrs);
 	nmi_enabled = 0;
 	on_each_cpu(nmi_cpu_shutdown, NULL, 0, 1);
 	unregister_die_notifier(&profile_exceptions_nb);
 	model->shutdown(msrs);
 	free_msrs();
+	put_cpu_var(cpu_msrs);
 }
 
 static void nmi_cpu_start(void *dummy)

include/asm-x86/amd_iommu.h

+/*
+ * Copyright (C) 2007-2008 Advanced Micro Devices, Inc.
+ * Author: Joerg Roedel <joerg.roedel@amd.com>
+ *         Leo Duran <leo.duran@amd.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
+ */
+
+#ifndef _ASM_X86_AMD_IOMMU_H
+#define _ASM_X86_AMD_IOMMU_H
+
+#ifdef CONFIG_AMD_IOMMU
+extern int amd_iommu_init(void);
+extern int amd_iommu_init_dma_ops(void);
+extern void amd_iommu_detect(void);
+#else
+static inline int amd_iommu_init(void) { return -ENODEV; }
+static inline void amd_iommu_detect(void) { }
+#endif
+
+#endif

include/asm-x86/amd_iommu_types.h

+/*
+ * Copyright (C) 2007-2008 Advanced Micro Devices, Inc.
+ * Author: Joerg Roedel <joerg.roedel@amd.com>
+ *         Leo Duran <leo.duran@amd.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
+ */
+
+#ifndef __AMD_IOMMU_TYPES_H__
+#define __AMD_IOMMU_TYPES_H__
+
+#include <linux/types.h>
+#include <linux/list.h>
+#include <linux/spinlock.h>
+
+/*
+ * some size calculation constants
+ */
+#define DEV_TABLE_ENTRY_SIZE		256
+#define ALIAS_TABLE_ENTRY_SIZE		2
+#define RLOOKUP_TABLE_ENTRY_SIZE	(sizeof(void *))
+
+/* helper macros */
+#define LOW_U32(x) ((x) & ((1ULL << 32)-1))
+#define HIGH_U32(x) (LOW_U32((x) >> 32))
+
+/* Length of the MMIO region for the AMD IOMMU */
+#define MMIO_REGION_LENGTH       0x4000
+
+/* Capability offsets used by the driver */
+#define MMIO_CAP_HDR_OFFSET	0x00
+#define MMIO_RANGE_OFFSET	0x0c
+
+/* Masks, shifts and macros to parse the device range capability */
+#define MMIO_RANGE_LD_MASK	0xff000000
+#define MMIO_RANGE_FD_MASK	0x00ff0000
+#define MMIO_RANGE_BUS_MASK	0x0000ff00
+#define MMIO_RANGE_LD_SHIFT	24
+#define MMIO_RANGE_FD_SHIFT	16
+#define MMIO_RANGE_BUS_SHIFT	8
+#define MMIO_GET_LD(x)  (((x) & MMIO_RANGE_LD_MASK) >> MMIO_RANGE_LD_SHIFT)
+#define MMIO_GET_FD(x)  (((x) & MMIO_RANGE_FD_MASK) >> MMIO_RANGE_FD_SHIFT)
+#define MMIO_GET_BUS(x) (((x) & MMIO_RANGE_BUS_MASK) >> MMIO_RANGE_BUS_SHIFT)
+
+/* Flag masks for the AMD IOMMU exclusion range */
+#define MMIO_EXCL_ENABLE_MASK 0x01ULL
+#define MMIO_EXCL_ALLOW_MASK  0x02ULL
+
+/* Used offsets into the MMIO space */
+#define MMIO_DEV_TABLE_OFFSET   0x0000
+#define MMIO_CMD_BUF_OFFSET     0x0008
+#define MMIO_EVT_BUF_OFFSET     0x0010
+#define MMIO_CONTROL_OFFSET     0x0018
+#define MMIO_EXCL_BASE_OFFSET   0x0020
+#define MMIO_EXCL_LIMIT_OFFSET  0x0028
+#define MMIO_CMD_HEAD_OFFSET	0x2000
+#define MMIO_CMD_TAIL_OFFSET	0x2008
+#define MMIO_EVT_HEAD_OFFSET	0x2010
+#define MMIO_EVT_TAIL_OFFSET	0x2018
+#define MMIO_STATUS_OFFSET	0x2020
+
+/* feature control bits */
+#define CONTROL_IOMMU_EN        0x00ULL
+#define CONTROL_HT_TUN_EN       0x01ULL
+#define CONTROL_EVT_LOG_EN      0x02ULL
+#define CONTROL_EVT_INT_EN      0x03ULL
+#define CONTROL_COMWAIT_EN      0x04ULL
+#define CONTROL_PASSPW_EN       0x08ULL
+#define CONTROL_RESPASSPW_EN    0x09ULL
+#define CONTROL_COHERENT_EN     0x0aULL
+#define CONTROL_ISOC_EN         0x0bULL
+#define CONTROL_CMDBUF_EN       0x0cULL
+#define CONTROL_PPFLOG_EN       0x0dULL
+#define CONTROL_PPFINT_EN       0x0eULL
+
+/* command specific defines */
+#define CMD_COMPL_WAIT          0x01
+#define CMD_INV_DEV_ENTRY       0x02
+#define CMD_INV_IOMMU_PAGES     0x03
+
+#define CMD_COMPL_WAIT_STORE_MASK	0x01
+#define CMD_INV_IOMMU_PAGES_SIZE_MASK	0x01
+#define CMD_INV_IOMMU_PAGES_PDE_MASK	0x02
+
+#define CMD_INV_IOMMU_ALL_PAGES_ADDRESS	0x7fffffffffffffffULL
+
+/* macros and definitions for device table entries */
+#define DEV_ENTRY_VALID         0x00
+#define DEV_ENTRY_TRANSLATION   0x01
+#define DEV_ENTRY_IR            0x3d
+#define DEV_ENTRY_IW            0x3e
+#define DEV_ENTRY_EX            0x67
+#define DEV_ENTRY_SYSMGT1       0x68
+#define DEV_ENTRY_SYSMGT2       0x69
+#define DEV_ENTRY_INIT_PASS     0xb8
+#define DEV_ENTRY_EINT_PASS     0xb9
+#define DEV_ENTRY_NMI_PASS      0xba
+#define DEV_ENTRY_LINT0_PASS    0xbe
+#define DEV_ENTRY_LINT1_PASS    0xbf
+
+/* constants to configure the command buffer */
+#define CMD_BUFFER_SIZE    8192
+#define CMD_BUFFER_ENTRIES 512
+#define MMIO_CMD_SIZE_SHIFT 56
+#define MMIO_CMD_SIZE_512 (0x9ULL << MMIO_CMD_SIZE_SHIFT)
+
+#define PAGE_MODE_1_LEVEL 0x01
+#define PAGE_MODE_2_LEVEL 0x02
+#define PAGE_MODE_3_LEVEL 0x03
+
+#define IOMMU_PDE_NL_0   0x000ULL
+#define IOMMU_PDE_NL_1   0x200ULL
+#define IOMMU_PDE_NL_2   0x400ULL
+#define IOMMU_PDE_NL_3   0x600ULL
+
+#define IOMMU_PTE_L2_INDEX(address) (((address) >> 30) & 0x1ffULL)
+#define IOMMU_PTE_L1_INDEX(address) (((address) >> 21) & 0x1ffULL)
+#define IOMMU_PTE_L0_INDEX(address) (((address) >> 12) & 0x1ffULL)
+
+#define IOMMU_MAP_SIZE_L1 (1ULL << 21)
+#define IOMMU_MAP_SIZE_L2 (1ULL << 30)
+#define IOMMU_MAP_SIZE_L3 (1ULL << 39)
+
+#define IOMMU_PTE_P  (1ULL << 0)
+#define IOMMU_PTE_U  (1ULL << 59)
+#define IOMMU_PTE_FC (1ULL << 60)
+#define IOMMU_PTE_IR (1ULL << 61)
+#define IOMMU_PTE_IW (1ULL << 62)
+
+#define IOMMU_L1_PDE(address) \
+	((address) | IOMMU_PDE_NL_1 | IOMMU_PTE_P | IOMMU_PTE_IR | IOMMU_PTE_IW)
+#define IOMMU_L2_PDE(address) \
+	((address) | IOMMU_PDE_NL_2 | IOMMU_PTE_P | IOMMU_PTE_IR | IOMMU_PTE_IW)
+
+#define IOMMU_PAGE_MASK (((1ULL << 52) - 1) & ~0xfffULL)
+#define IOMMU_PTE_PRESENT(pte) ((pte) & IOMMU_PTE_P)
+#define IOMMU_PTE_PAGE(pte) (phys_to_virt((pte) & IOMMU_PAGE_MASK))
+#define IOMMU_PTE_MODE(pte) (((pte) >> 9) & 0x07)
+
+#define IOMMU_PROT_MASK 0x03
+#define IOMMU_PROT_IR 0x01
+#define IOMMU_PROT_IW 0x02
+
+/* IOMMU capabilities */
+#define IOMMU_CAP_IOTLB   24
+#define IOMMU_CAP_NPCACHE 26
+
+#define MAX_DOMAIN_ID 65536
+
+struct protection_domain {
+	spinlock_t lock;
+	u16 id;
+	int mode;
+	u64 *pt_root;
+	void *priv;
+};
+
+struct dma_ops_domain {
+	struct list_head list;
+	struct protection_domain domain;
+	unsigned long aperture_size;
+	unsigned long next_bit;
+	unsigned long *bitmap;
+	u64 **pte_pages;
+};
+
+struct amd_iommu {
+	struct list_head list;
+	spinlock_t lock;
+
+	u16 devid;
+	u16 cap_ptr;
+
+	u64 mmio_phys;
+	u8 *mmio_base;
+	u32 cap;
+	u16 first_device;
+	u16 last_device;
+	u64 exclusion_start;
+	u64 exclusion_length;
+
+	u8 *cmd_buf;
+	u32 cmd_buf_size;
+
+	int need_sync;
+
+	struct dma_ops_domain *default_dom;
+};
+
+extern struct list_head amd_iommu_list;
+
+struct dev_table_entry {
+	u32 data[8];
+};
+
+struct unity_map_entry {
+	struct list_head list;
+	u16 devid_start;
+	u16 devid_end;
+	u64 address_start;
+	u64 address_end;
+	int prot;
+};
+
+extern struct list_head amd_iommu_unity_map;
+
+/* data structures for device handling */
+extern struct dev_table_entry *amd_iommu_dev_table;
+extern u16 *amd_iommu_alias_table;
+extern struct amd_iommu **amd_iommu_rlookup_table;
+
+extern unsigned amd_iommu_aperture_order;
+
+extern u16 amd_iommu_last_bdf;
+
+/* data structures for protection domain handling */
+extern struct protection_domain **amd_iommu_pd_table;
+extern unsigned long *amd_iommu_pd_alloc_bitmap;
+
+extern int amd_iommu_isolate;
+
+static inline void print_devid(u16 devid, int nl)
+{
+	int bus = devid >> 8;
+	int dev = devid >> 3 & 0x1f;
+	int fn  = devid & 0x07;
+
+	printk("%02x:%02x.%x", bus, dev, fn);
+	if (nl)
+		printk("\n");
+}
+
+#endif

include/asm-x86/cpufeature.h

@@ -106,6 +106,7 @@
 /* More extended AMD flags: CPUID level 0x80000001, ecx, word 6 */
 #define X86_FEATURE_LAHF_LM	(6*32+ 0) /* LAHF/SAHF in long mode */
 #define X86_FEATURE_CMP_LEGACY	(6*32+ 1) /* If yes HyperThreading not valid */
+#define X86_FEATURE_IBS		(6*32+ 10) /* Instruction Based Sampling */
 
 /*
  * Auxiliary flags: Linux defined - For features scattered in various

include/asm-x86/gart.h

@@ -22,8 +22,9 @@ extern int gart_iommu_aperture_allowed;
 extern int gart_iommu_aperture_disabled;
 extern int fix_aperture;
 #else
-#define gart_iommu_aperture 0
-#define gart_iommu_aperture_allowed 0
+#define gart_iommu_aperture		0
+#define gart_iommu_aperture_allowed	0
+#define gart_iommu_aperture_disabled	1
 
 static inline void early_gart_iommu_check(void)
 {

include/asm-x86/io.h

@@ -3,6 +3,62 @@
 
 #define ARCH_HAS_IOREMAP_WC
 
+#include <linux/compiler.h>
+
+#define build_mmio_read(name, size, type, reg, barrier) \
+static inline type name(const volatile void __iomem *addr) \
+{ type ret; asm volatile("mov" size " %1,%0":"=" reg (ret) \
+:"m" (*(volatile type __force *)addr) barrier); return ret; }
+
+#define build_mmio_write(name, size, type, reg, barrier) \
+static inline void name(type val, volatile void __iomem *addr) \
+{ asm volatile("mov" size " %0,%1": :reg (val), \
+"m" (*(volatile type __force *)addr) barrier); }
+
+build_mmio_read(readb, "b", unsigned char, "q", :"memory")
+build_mmio_read(readw, "w", unsigned short, "r", :"memory")
+build_mmio_read(readl, "l", unsigned int, "r", :"memory")
+
+build_mmio_read(__readb, "b", unsigned char, "q", )
+build_mmio_read(__readw, "w", unsigned short, "r", )
+build_mmio_read(__readl, "l", unsigned int, "r", )
+
+build_mmio_write(writeb, "b", unsigned char, "q", :"memory")
+build_mmio_write(writew, "w", unsigned short, "r", :"memory")
+build_mmio_write(writel, "l", unsigned int, "r", :"memory")
+
+build_mmio_write(__writeb, "b", unsigned char, "q", )
+build_mmio_write(__writew, "w", unsigned short, "r", )
+build_mmio_write(__writel, "l", unsigned int, "r", )
+
+#define readb_relaxed(a) __readb(a)
+#define readw_relaxed(a) __readw(a)
+#define readl_relaxed(a) __readl(a)
+#define __raw_readb __readb
+#define __raw_readw __readw
+#define __raw_readl __readl
+
+#define __raw_writeb __writeb
+#define __raw_writew __writew
+#define __raw_writel __writel
+
+#define mmiowb() barrier()
+
+#ifdef CONFIG_X86_64
+build_mmio_read(readq, "q", unsigned long, "r", :"memory")
+build_mmio_read(__readq, "q", unsigned long, "r", )
+build_mmio_write(writeq, "q", unsigned long, "r", :"memory")
+build_mmio_write(__writeq, "q", unsigned long, "r", )
+
+#define readq_relaxed(a) __readq(a)
+#define __raw_readq __readq
+#define __raw_writeq writeq
+
+/* Let people know we have them */
+#define readq readq
+#define writeq writeq
+#endif
+
 #ifdef CONFIG_X86_32
 # include "io_32.h"
 #else

include/asm-x86/io_32.h

@@ -149,55 +149,6 @@ extern void __iomem *fix_ioremap(unsigned idx, unsigned long phys);
 #define virt_to_bus virt_to_phys
 #define bus_to_virt phys_to_virt
 
-/*
- * readX/writeX() are used to access memory mapped devices. On some
- * architectures the memory mapped IO stuff needs to be accessed
- * differently. On the x86 architecture, we just read/write the
- * memory location directly.
- */
-
-static inline unsigned char readb(const volatile void __iomem *addr)
-{
-	return *(volatile unsigned char __force *)addr;
-}
-
-static inline unsigned short readw(const volatile void __iomem *addr)
-{
-	return *(volatile unsigned short __force *)addr;
-}
-
-static inline unsigned int readl(const volatile void __iomem *addr)
-{
-	return *(volatile unsigned int __force *) addr;
-}
-
-#define readb_relaxed(addr) readb(addr)
-#define readw_relaxed(addr) readw(addr)
-#define readl_relaxed(addr) readl(addr)
-#define __raw_readb readb
-#define __raw_readw readw
-#define __raw_readl readl
-
-static inline void writeb(unsigned char b, volatile void __iomem *addr)
-{
-	*(volatile unsigned char __force *)addr = b;
-}
-
-static inline void writew(unsigned short b, volatile void __iomem *addr)
-{
-	*(volatile unsigned short __force *)addr = b;
-}
-
-static inline void writel(unsigned int b, volatile void __iomem *addr)
-{
-	*(volatile unsigned int __force *)addr = b;
-}
-#define __raw_writeb writeb
-#define __raw_writew writew
-#define __raw_writel writel
-
-#define mmiowb()
-
 static inline void
 memset_io(volatile void __iomem *addr, unsigned char val, int count)
 {

include/asm-x86/io_64.h

@@ -204,77 +204,6 @@ extern void __iomem *fix_ioremap(unsigned idx, unsigned long phys);
 #define virt_to_bus virt_to_phys
 #define bus_to_virt phys_to_virt
 
-/*
- * readX/writeX() are used to access memory mapped devices. On some
- * architectures the memory mapped IO stuff needs to be accessed
- * differently. On the x86 architecture, we just read/write the
- * memory location directly.
- */
-
-static inline __u8 __readb(const volatile void __iomem *addr)
-{
-	return *(__force volatile __u8 *)addr;
-}
-
-static inline __u16 __readw(const volatile void __iomem *addr)
-{
-	return *(__force volatile __u16 *)addr;
-}
-
-static __always_inline __u32 __readl(const volatile void __iomem *addr)
-{
-	return *(__force volatile __u32 *)addr;
-}
-
-static inline __u64 __readq(const volatile void __iomem *addr)
-{
-	return *(__force volatile __u64 *)addr;
-}
-
-#define readb(x) __readb(x)
-#define readw(x) __readw(x)
-#define readl(x) __readl(x)
-#define readq(x) __readq(x)
-#define readb_relaxed(a) readb(a)
-#define readw_relaxed(a) readw(a)
-#define readl_relaxed(a) readl(a)
-#define readq_relaxed(a) readq(a)
-#define __raw_readb readb
-#define __raw_readw readw
-#define __raw_readl readl
-#define __raw_readq readq
-
-#define mmiowb()
-
-static inline void __writel(__u32 b, volatile void __iomem *addr)
-{
-	*(__force volatile __u32 *)addr = b;
-}
-
-static inline void __writeq(__u64 b, volatile void __iomem *addr)
-{
-	*(__force volatile __u64 *)addr = b;
-}
-
-static inline void __writeb(__u8 b, volatile void __iomem *addr)
-{
-	*(__force volatile __u8 *)addr = b;
-}
-
-static inline void __writew(__u16 b, volatile void __iomem *addr)
-{
-	*(__force volatile __u16 *)addr = b;
-}
-
-#define writeq(val, addr) __writeq((val), (addr))
-#define writel(val, addr) __writel((val), (addr))
-#define writew(val, addr) __writew((val), (addr))
-#define writeb(val, addr) __writeb((val), (addr))
-#define __raw_writeb writeb
-#define __raw_writew writew
-#define __raw_writel writel
-#define __raw_writeq writeq
-
 void __memcpy_fromio(void *, unsigned long, unsigned);
 void __memcpy_toio(unsigned long, const void *, unsigned);
 

include/linux/delay.h

@@ -41,6 +41,7 @@ static inline void ndelay(unsigned long x)
 #define ndelay(x) ndelay(x)
 #endif
 
+extern unsigned long lpj_fine;
 void calibrate_delay(void);
 void msleep(unsigned int msecs);
 unsigned long msleep_interruptible(unsigned int msecs);

init/calibrate.c

@@ -8,7 +8,9 @@
 #include <linux/delay.h>
 #include <linux/init.h>
 #include <linux/timex.h>
+#include <linux/smp.h>
 
+unsigned long lpj_fine;
 unsigned long preset_lpj;
 static int __init lpj_setup(char *str)
 {
@@ -33,9 +35,9 @@ static unsigned long __cpuinit calibrate_delay_direct(void)
 	unsigned long pre_start, start, post_start;
 	unsigned long pre_end, end, post_end;
 	unsigned long start_jiffies;
-	unsigned long tsc_rate_min, tsc_rate_max;
-	unsigned long good_tsc_sum = 0;
-	unsigned long good_tsc_count = 0;
+	unsigned long timer_rate_min, timer_rate_max;
+	unsigned long good_timer_sum = 0;
+	unsigned long good_timer_count = 0;
 	int i;
 
 	if (read_current_timer(&pre_start) < 0 )
@@ -79,22 +81,24 @@ static unsigned long __cpuinit calibrate_delay_direct(void)
 		}
 		read_current_timer(&post_end);
 
-		tsc_rate_max = (post_end - pre_start) / DELAY_CALIBRATION_TICKS;
-		tsc_rate_min = (pre_end - post_start) / DELAY_CALIBRATION_TICKS;
+		timer_rate_max = (post_end - pre_start) /
+					DELAY_CALIBRATION_TICKS;
+		timer_rate_min = (pre_end - post_start) /
+					DELAY_CALIBRATION_TICKS;
 
 		/*
-	 	 * If the upper limit and lower limit of the tsc_rate is
+		 * If the upper limit and lower limit of the timer_rate is
 		 * >= 12.5% apart, redo calibration.
 		 */
 		if (pre_start != 0 && pre_end != 0 &&
-		    (tsc_rate_max - tsc_rate_min) < (tsc_rate_max >> 3)) {
-			good_tsc_count++;
-			good_tsc_sum += tsc_rate_max;
+		    (timer_rate_max - timer_rate_min) < (timer_rate_max >> 3)) {
+			good_timer_count++;
+			good_timer_sum += timer_rate_max;
 		}
 	}
 
-	if (good_tsc_count)
-		return (good_tsc_sum/good_tsc_count);
+	if (good_timer_count)
+		return (good_timer_sum/good_timer_count);
 
 	printk(KERN_WARNING "calibrate_delay_direct() failed to get a good "
 	       "estimate for loops_per_jiffy.\nProbably due to long platform interrupts. Consider using \"lpj=\" boot option.\n");
@@ -108,6 +112,10 @@ static unsigned long __cpuinit calibrate_delay_direct(void) {return 0;}
  * This is the number of bits of precision for the loops_per_jiffy.  Each
  * bit takes on average 1.5/HZ seconds.  This (like the original) is a little
  * better than 1%
+ * For the boot cpu we can skip the delay calibration and assign it a value
+ * calculated based on the timer frequency.
+ * For the rest of the CPUs we cannot assume that the timer frequency is same as
+ * the cpu frequency, hence do the calibration for those.
  */
 #define LPS_PREC 8
 
@@ -118,20 +126,20 @@ void __cpuinit calibrate_delay(void)
 
 	if (preset_lpj) {
 		loops_per_jiffy = preset_lpj;
-		printk("Calibrating delay loop (skipped)... "
-			"%lu.%02lu BogoMIPS preset\n",
-			loops_per_jiffy/(500000/HZ),
-			(loops_per_jiffy/(5000/HZ)) % 100);
+		printk(KERN_INFO
+			"Calibrating delay loop (skipped) preset value.. ");
+	} else if ((smp_processor_id() == 0) && lpj_fine) {
+		loops_per_jiffy = lpj_fine;
+		printk(KERN_INFO
+			"Calibrating delay loop (skipped), "
+			"value calculated using timer frequency.. ");
 	} else if ((loops_per_jiffy = calibrate_delay_direct()) != 0) {
-		printk("Calibrating delay using timer specific routine.. ");
-		printk("%lu.%02lu BogoMIPS (lpj=%lu)\n",
-			loops_per_jiffy/(500000/HZ),
-			(loops_per_jiffy/(5000/HZ)) % 100,
-			loops_per_jiffy);
+		printk(KERN_INFO
+			"Calibrating delay using timer specific routine.. ");
 	} else {
 		loops_per_jiffy = (1<<12);
 
-		printk(KERN_DEBUG "Calibrating delay loop... ");
+		printk(KERN_INFO "Calibrating delay loop... ");
 		while ((loops_per_jiffy <<= 1) != 0) {
 			/* wait for "start of" clock tick */
 			ticks = jiffies;
@@ -161,12 +169,8 @@ void __cpuinit calibrate_delay(void)
 			if (jiffies != ticks)	/* longer than 1 tick */
 				loops_per_jiffy &= ~loopbit;
 		}
-
-		/* Round the value and print it */
-		printk("%lu.%02lu BogoMIPS (lpj=%lu)\n",
-			loops_per_jiffy/(500000/HZ),
-			(loops_per_jiffy/(5000/HZ)) % 100,
-			loops_per_jiffy);
 	}
-
+	printk(KERN_INFO "%lu.%02lu BogoMIPS (lpj=%lu)\n",
+			loops_per_jiffy/(500000/HZ),
+			(loops_per_jiffy/(5000/HZ)) % 100, loops_per_jiffy);
 }