Merge commit 'v2.6.29-rc4' into tracing/core

.mailmap

@@ -92,6 +92,7 @@ Rudolf Marek <R.Marek@sh.cvut.cz>
 Rui Saraiva <rmps@joel.ist.utl.pt>
 Sachin P Sant <ssant@in.ibm.com>
 Sam Ravnborg <sam@mars.ravnborg.org>
+Sascha Hauer <s.hauer@pengutronix.de>
 S.Çağlar Onur <caglar@pardus.org.tr>
 Simon Kelley <simon@thekelleys.org.uk>
 Stéphane Witzmann <stephane.witzmann@ubpmes.univ-bpclermont.fr>
@@ -100,6 +101,7 @@ Tejun Heo <htejun@gmail.com>
 Thomas Graf <tgraf@suug.ch>
 Tony Luck <tony.luck@intel.com>
 Tsuneo Yoshioka <Tsuneo.Yoshioka@f-secure.com>
-Uwe Kleine-König <Uwe.Kleine-Koenig@digi.com>
 Uwe Kleine-König <ukleinek@informatik.uni-freiburg.de>
+Uwe Kleine-König <ukl@pengutronix.de>
+Uwe Kleine-König <Uwe.Kleine-Koenig@digi.com>
 Valdis Kletnieks <Valdis.Kletnieks@vt.edu>

Documentation/cpu-freq/user-guide.txt

@@ -195,19 +195,3 @@ scaling_setspeed.		By "echoing" a new frequency into this
 				you can change the speed of the CPU,
 				but only within the limits of
 				scaling_min_freq and scaling_max_freq.
-				
-
-3.2 Deprecated Interfaces
--------------------------
-
-Depending on your kernel configuration, you might find the following 
-cpufreq-related files:
-/proc/cpufreq
-/proc/sys/cpu/*/speed
-/proc/sys/cpu/*/speed-min
-/proc/sys/cpu/*/speed-max
-
-These are files for deprecated interfaces to cpufreq, which offer far
-less functionality. Because of this, these interfaces aren't described
-here.
-

Documentation/filesystems/sysfs-pci.txt

@@ -9,6 +9,7 @@ that support it.  For example, a given bus might look like this:
      |   |-- class
      |   |-- config
      |   |-- device
+     |   |-- enable
      |   |-- irq
      |   |-- local_cpus
      |   |-- resource
@@ -32,6 +33,7 @@ files, each with their own function.
        class		   PCI class (ascii, ro)
        config		   PCI config space (binary, rw)
        device		   PCI device (ascii, ro)
+       enable	           Whether the device is enabled (ascii, rw)
        irq		   IRQ number (ascii, ro)
        local_cpus	   nearby CPU mask (cpumask, ro)
        resource		   PCI resource host addresses (ascii, ro)
@@ -57,10 +59,19 @@ used to do actual device programming from userspace.  Note that some platforms
 don't support mmapping of certain resources, so be sure to check the return
 value from any attempted mmap.
 
+The 'enable' file provides a counter that indicates how many times the device 
+has been enabled.  If the 'enable' file currently returns '4', and a '1' is
+echoed into it, it will then return '5'.  Echoing a '0' into it will decrease
+the count.  Even when it returns to 0, though, some of the initialisation
+may not be reversed.  
+
 The 'rom' file is special in that it provides read-only access to the device's
 ROM file, if available.  It's disabled by default, however, so applications
 should write the string "1" to the file to enable it before attempting a read
-call, and disable it following the access by writing "0" to the file.
+call, and disable it following the access by writing "0" to the file.  Note
+that the device must be enabled for a rom read to return data succesfully.
+In the event a driver is not bound to the device, it can be enabled using the
+'enable' file, documented above.
 
 Accessing legacy resources through sysfs
 ----------------------------------------

MAINTAINERS

@@ -1021,6 +1021,14 @@ M:	mb@bu3sch.de
 W:	http://bu3sch.de/btgpio.php
 S:	Maintained
 
+BTRFS FILE SYSTEM
+P:	Chris Mason
+M:	chris.mason@oracle.com
+L:	linux-btrfs@vger.kernel.org
+W:	http://btrfs.wiki.kernel.org/
+T:	git kernel.org:/pub/scm/linux/kernel/git/mason/btrfs-unstable.git
+S:	Maintained
+
 BTTV VIDEO4LINUX DRIVER
 P:	Mauro Carvalho Chehab
 M:	mchehab@infradead.org
@@ -2212,7 +2220,7 @@ P:	Sean Hefty
 M:	sean.hefty@intel.com
 P:	Hal Rosenstock
 M:	hal.rosenstock@gmail.com
-L:	general@lists.openfabrics.org
+L:	general@lists.openfabrics.org (moderated for non-subscribers)
 W:	http://www.openib.org/
 T:	git kernel.org:/pub/scm/linux/kernel/git/roland/infiniband.git
 S:	Supported
@@ -4847,6 +4855,7 @@ P:	Ingo Molnar
 M:	mingo@redhat.com
 P:	H. Peter Anvin
 M:	hpa@zytor.com
+M:	x86@kernel.org
 L:	linux-kernel@vger.kernel.org
 T:	git://git.kernel.org/pub/scm/linux/kernel/git/x86/linux-2.6-x86.git
 S:	Maintained

Makefile

@@ -1,7 +1,7 @@
 VERSION = 2
 PATCHLEVEL = 6
 SUBLEVEL = 29
-EXTRAVERSION = -rc3
+EXTRAVERSION = -rc4
 NAME = Erotic Pickled Herring
 
 # *DOCUMENTATION*

arch/alpha/include/asm/bug.h

@@ -8,12 +8,12 @@
 
 /* ??? Would be nice to use .gprel32 here, but we can't be sure that the
    function loaded the GP, so this could fail in modules.  */
-#define BUG()	{							\
+#define BUG()	do {							\
 	__asm__ __volatile__(						\
 		"call_pal %0  # bugchk\n\t"				\
 		".long %1\n\t.8byte %2"					\
 		: : "i"(PAL_bugchk), "i"(__LINE__), "i"(__FILE__));	\
-	for ( ; ; ); }
+	for ( ; ; ); } while (0)
 
 #define HAVE_ARCH_BUG
 #endif

arch/ia64/sn/kernel/io_acpi_init.c

@@ -443,7 +443,7 @@ sn_acpi_slot_fixup(struct pci_dev *dev)
 		size = pci_resource_len(dev, PCI_ROM_RESOURCE);
 		addr = ioremap(pcidev_info->pdi_pio_mapped_addr[PCI_ROM_RESOURCE],
 			       size);
-		image_size = pci_get_rom_size(addr, size);
+		image_size = pci_get_rom_size(dev, addr, size);
 		dev->resource[PCI_ROM_RESOURCE].start = (unsigned long) addr;
 		dev->resource[PCI_ROM_RESOURCE].end =
 					(unsigned long) addr + image_size - 1;

arch/ia64/sn/kernel/io_init.c

@@ -269,7 +269,7 @@ sn_io_slot_fixup(struct pci_dev *dev)
 
 			rom = ioremap(pci_resource_start(dev, PCI_ROM_RESOURCE),
 				      size + 1);
-			image_size = pci_get_rom_size(rom, size + 1);
+			image_size = pci_get_rom_size(dev, rom, size + 1);
 			dev->resource[PCI_ROM_RESOURCE].end =
 				dev->resource[PCI_ROM_RESOURCE].start +
 				image_size - 1;

arch/sh/boards/board-ap325rxa.c

@@ -22,6 +22,7 @@
 #include <linux/gpio.h>
 #include <linux/spi/spi.h>
 #include <linux/spi/spi_gpio.h>
+#include <media/ov772x.h>
 #include <media/soc_camera_platform.h>
 #include <media/sh_mobile_ceu.h>
 #include <video/sh_mobile_lcdc.h>
@@ -216,7 +217,14 @@ static struct platform_device lcdc_device = {
 	},
 };
 
+static void camera_power(int val)
+{
+	gpio_set_value(GPIO_PTZ5, val); /* RST_CAM/RSTB */
+	mdelay(10);
+}
+
 #ifdef CONFIG_I2C
+/* support for the old ncm03j camera */
 static unsigned char camera_ncm03j_magic[] =
 {
 	0x87, 0x00, 0x88, 0x08, 0x89, 0x01, 0x8A, 0xE8,
@@ -237,6 +245,23 @@ static unsigned char camera_ncm03j_magic[] =
 	0x63, 0xD4, 0x64, 0xEA, 0xD6, 0x0F,
 };
 
+static int camera_probe(void)
+{
+	struct i2c_adapter *a = i2c_get_adapter(0);
+	struct i2c_msg msg;
+	int ret;
+
+	camera_power(1);
+	msg.addr = 0x6e;
+	msg.buf = camera_ncm03j_magic;
+	msg.len = 2;
+	msg.flags = 0;
+	ret = i2c_transfer(a, &msg, 1);
+	camera_power(0);
+
+	return ret;
+}
+
 static int camera_set_capture(struct soc_camera_platform_info *info,
 			      int enable)
 {
@@ -245,9 +270,11 @@ static int camera_set_capture(struct soc_camera_platform_info *info,
 	int ret = 0;
 	int i;
 
+	camera_power(0);
 	if (!enable)
 		return 0; /* no disable for now */
 
+	camera_power(1);
 	for (i = 0; i < ARRAY_SIZE(camera_ncm03j_magic); i += 2) {
 		u_int8_t buf[8];
 
@@ -286,8 +313,35 @@ static struct platform_device camera_device = {
 		.platform_data	= &camera_info,
 	},
 };
+
+static int __init camera_setup(void)
+{
+	if (camera_probe() > 0)
+		platform_device_register(&camera_device);
+
+	return 0;
+}
+late_initcall(camera_setup);
+
 #endif /* CONFIG_I2C */
 
+static int ov7725_power(struct device *dev, int mode)
+{
+	camera_power(0);
+	if (mode)
+		camera_power(1);
+
+	return 0;
+}
+
+static struct ov772x_camera_info ov7725_info = {
+	.buswidth  = SOCAM_DATAWIDTH_8,
+	.flags = OV772X_FLAG_VFLIP | OV772X_FLAG_HFLIP,
+	.link = {
+		.power  = ov7725_power,
+	},
+};
+
 static struct sh_mobile_ceu_info sh_mobile_ceu_info = {
 	.flags = SOCAM_PCLK_SAMPLE_RISING | SOCAM_HSYNC_ACTIVE_HIGH |
 	SOCAM_VSYNC_ACTIVE_HIGH | SOCAM_MASTER | SOCAM_DATAWIDTH_8,
@@ -338,9 +392,6 @@ static struct platform_device *ap325rxa_devices[] __initdata = {
 	&ap325rxa_nor_flash_device,
 	&lcdc_device,
 	&ceu_device,
-#ifdef CONFIG_I2C
-	&camera_device,
-#endif
 	&nand_flash_device,
 	&sdcard_cn3_device,
 };
@@ -349,6 +400,10 @@ static struct i2c_board_info __initdata ap325rxa_i2c_devices[] = {
 	{
 		I2C_BOARD_INFO("pcf8563", 0x51),
 	},
+	{
+		I2C_BOARD_INFO("ov772x", 0x21),
+		.platform_data = &ov7725_info,
+	},
 };
 
 static struct spi_board_info ap325rxa_spi_devices[] = {
@@ -426,7 +481,7 @@ static int __init ap325rxa_devices_setup(void)
 	gpio_request(GPIO_PTZ6, NULL);
 	gpio_direction_output(GPIO_PTZ6, 0); /* STBY_CAM */
 	gpio_request(GPIO_PTZ5, NULL);
-	gpio_direction_output(GPIO_PTZ5, 1); /* RST_CAM */
+	gpio_direction_output(GPIO_PTZ5, 0); /* RST_CAM */
 	gpio_request(GPIO_PTZ4, NULL);
 	gpio_direction_output(GPIO_PTZ4, 0); /* SADDR */
 

arch/sh/configs/ap325rxa_defconfig

@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.28
-# Fri Jan  9 16:54:19 2009
+# Linux kernel version: 2.6.29-rc2
+# Tue Jan 27 11:45:08 2009
 #
 CONFIG_SUPERH=y
 CONFIG_SUPERH32=y
@@ -45,12 +45,12 @@ CONFIG_BSD_PROCESS_ACCT=y
 # CONFIG_AUDIT is not set
 # CONFIG_IKCONFIG is not set
 CONFIG_LOG_BUF_SHIFT=14
-# CONFIG_CGROUPS is not set
 CONFIG_GROUP_SCHED=y
 CONFIG_FAIR_GROUP_SCHED=y
 # CONFIG_RT_GROUP_SCHED is not set
 CONFIG_USER_SCHED=y
 # CONFIG_CGROUP_SCHED is not set
+# CONFIG_CGROUPS is not set
 CONFIG_SYSFS_DEPRECATED=y
 CONFIG_SYSFS_DEPRECATED_V2=y
 # CONFIG_RELAY is not set
@@ -378,6 +378,7 @@ CONFIG_WIRELESS=y
 # CONFIG_WIRELESS_EXT is not set
 # CONFIG_LIB80211 is not set
 # CONFIG_MAC80211 is not set
+# CONFIG_WIMAX is not set
 # CONFIG_RFKILL is not set
 # CONFIG_NET_9P is not set
 
@@ -400,6 +401,7 @@ CONFIG_MTD=y
 # CONFIG_MTD_DEBUG is not set
 CONFIG_MTD_CONCAT=y
 CONFIG_MTD_PARTITIONS=y
+# CONFIG_MTD_TESTS is not set
 # CONFIG_MTD_REDBOOT_PARTS is not set
 CONFIG_MTD_CMDLINE_PARTS=y
 # CONFIG_MTD_AR7_PARTS is not set
@@ -447,9 +449,7 @@ CONFIG_MTD_CFI_UTIL=y
 #
 # CONFIG_MTD_COMPLEX_MAPPINGS is not set
 CONFIG_MTD_PHYSMAP=y
-CONFIG_MTD_PHYSMAP_START=0xffffffff
-CONFIG_MTD_PHYSMAP_LEN=0
-CONFIG_MTD_PHYSMAP_BANKWIDTH=0
+# CONFIG_MTD_PHYSMAP_COMPAT is not set
 # CONFIG_MTD_PLATRAM is not set
 
 #
@@ -479,6 +479,12 @@ CONFIG_MTD_NAND_IDS=y
 CONFIG_MTD_NAND_SH_FLCTL=y
 # CONFIG_MTD_ONENAND is not set
 
+#
+# LPDDR flash memory drivers
+#
+# CONFIG_MTD_LPDDR is not set
+# CONFIG_MTD_QINFO_PROBE is not set
+
 #
 # UBI - Unsorted block images
 #
@@ -607,6 +613,10 @@ CONFIG_SMSC911X=y
 # CONFIG_WLAN_PRE80211 is not set
 # CONFIG_WLAN_80211 is not set
 # CONFIG_IWLWIFI_LEDS is not set
+
+#
+# Enable WiMAX (Networking options) to see the WiMAX drivers
+#
 # CONFIG_WAN is not set
 # CONFIG_PPP is not set
 # CONFIG_SLIP is not set
@@ -790,6 +800,7 @@ CONFIG_SSB_POSSIBLE=y
 # CONFIG_PMIC_DA903X is not set
 # CONFIG_MFD_WM8400 is not set
 # CONFIG_MFD_WM8350_I2C is not set
+# CONFIG_MFD_PCF50633 is not set
 # CONFIG_REGULATOR is not set
 
 #
@@ -837,7 +848,7 @@ CONFIG_SOC_CAMERA=y
 # CONFIG_SOC_CAMERA_MT9V022 is not set
 # CONFIG_SOC_CAMERA_TW9910 is not set
 CONFIG_SOC_CAMERA_PLATFORM=y
-# CONFIG_SOC_CAMERA_OV772X is not set
+CONFIG_SOC_CAMERA_OV772X=y
 CONFIG_VIDEO_SH_MOBILE_CEU=y
 # CONFIG_RADIO_ADAPTERS is not set
 # CONFIG_DAB is not set
@@ -1012,6 +1023,7 @@ CONFIG_FS_POSIX_ACL=y
 CONFIG_FILE_LOCKING=y
 # CONFIG_XFS_FS is not set
 # CONFIG_OCFS2_FS is not set
+# CONFIG_BTRFS_FS is not set
 CONFIG_DNOTIFY=y
 CONFIG_INOTIFY=y
 CONFIG_INOTIFY_USER=y
@@ -1060,6 +1072,7 @@ CONFIG_MISC_FILESYSTEMS=y
 # CONFIG_JFFS2_FS is not set
 # CONFIG_UBIFS_FS is not set
 # CONFIG_CRAMFS is not set
+# CONFIG_SQUASHFS is not set
 # CONFIG_VXFS_FS is not set
 # CONFIG_MINIX_FS is not set
 # CONFIG_OMFS_FS is not set

arch/sh/configs/migor_defconfig

@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.28
-# Fri Jan  9 17:09:35 2009
+# Linux kernel version: 2.6.29-rc1
+# Thu Jan 22 09:16:16 2009
 #
 CONFIG_SUPERH=y
 CONFIG_SUPERH32=y
@@ -45,8 +45,12 @@ CONFIG_SYSVIPC_SYSCTL=y
 CONFIG_IKCONFIG=y
 CONFIG_IKCONFIG_PROC=y
 CONFIG_LOG_BUF_SHIFT=14
-# CONFIG_CGROUPS is not set
 # CONFIG_GROUP_SCHED is not set
+
+#
+# Control Group support
+#
+# CONFIG_CGROUPS is not set
 CONFIG_SYSFS_DEPRECATED=y
 CONFIG_SYSFS_DEPRECATED_V2=y
 # CONFIG_RELAY is not set
@@ -389,6 +393,7 @@ CONFIG_WIRELESS_EXT=y
 CONFIG_WIRELESS_EXT_SYSFS=y
 # CONFIG_LIB80211 is not set
 # CONFIG_MAC80211 is not set
+# CONFIG_WIMAX is not set
 # CONFIG_RFKILL is not set
 # CONFIG_NET_9P is not set
 
@@ -411,6 +416,7 @@ CONFIG_MTD=y
 # CONFIG_MTD_DEBUG is not set
 CONFIG_MTD_CONCAT=y
 CONFIG_MTD_PARTITIONS=y
+# CONFIG_MTD_TESTS is not set
 # CONFIG_MTD_REDBOOT_PARTS is not set
 CONFIG_MTD_CMDLINE_PARTS=y
 # CONFIG_MTD_AR7_PARTS is not set
@@ -458,9 +464,7 @@ CONFIG_MTD_CFI_UTIL=y
 #
 # CONFIG_MTD_COMPLEX_MAPPINGS is not set
 CONFIG_MTD_PHYSMAP=y
-CONFIG_MTD_PHYSMAP_START=0xffffffff
-CONFIG_MTD_PHYSMAP_LEN=0
-CONFIG_MTD_PHYSMAP_BANKWIDTH=0
+# CONFIG_MTD_PHYSMAP_COMPAT is not set
 # CONFIG_MTD_PLATRAM is not set
 
 #
@@ -487,6 +491,12 @@ CONFIG_MTD_NAND_IDS=y
 CONFIG_MTD_NAND_PLATFORM=y
 # CONFIG_MTD_ONENAND is not set
 
+#
+# LPDDR flash memory drivers
+#
+# CONFIG_MTD_LPDDR is not set
+# CONFIG_MTD_QINFO_PROBE is not set
+
 #
 # UBI - Unsorted block images
 #
@@ -587,6 +597,10 @@ CONFIG_SMC91X=y
 # CONFIG_WLAN_PRE80211 is not set
 # CONFIG_WLAN_80211 is not set
 # CONFIG_IWLWIFI_LEDS is not set
+
+#
+# Enable WiMAX (Networking options) to see the WiMAX drivers
+#
 # CONFIG_WAN is not set
 # CONFIG_PPP is not set
 # CONFIG_SLIP is not set
@@ -761,6 +775,7 @@ CONFIG_SSB_POSSIBLE=y
 # CONFIG_PMIC_DA903X is not set
 # CONFIG_MFD_WM8400 is not set
 # CONFIG_MFD_WM8350_I2C is not set
+# CONFIG_MFD_PCF50633 is not set
 # CONFIG_REGULATOR is not set
 
 #
@@ -806,9 +821,9 @@ CONFIG_SOC_CAMERA=y
 # CONFIG_SOC_CAMERA_MT9M111 is not set
 # CONFIG_SOC_CAMERA_MT9T031 is not set
 # CONFIG_SOC_CAMERA_MT9V022 is not set
-# CONFIG_SOC_CAMERA_TW9910 is not set
-CONFIG_SOC_CAMERA_PLATFORM=y
-# CONFIG_SOC_CAMERA_OV772X is not set
+CONFIG_SOC_CAMERA_TW9910=y
+# CONFIG_SOC_CAMERA_PLATFORM is not set
+CONFIG_SOC_CAMERA_OV772X=y
 CONFIG_VIDEO_SH_MOBILE_CEU=y
 # CONFIG_RADIO_ADAPTERS is not set
 # CONFIG_DAB is not set
@@ -866,11 +881,13 @@ CONFIG_USB_GADGET_SELECTED=y
 # CONFIG_USB_GADGET_PXA25X is not set
 # CONFIG_USB_GADGET_PXA27X is not set
 # CONFIG_USB_GADGET_S3C2410 is not set
+# CONFIG_USB_GADGET_IMX is not set
 CONFIG_USB_GADGET_M66592=y
 CONFIG_USB_M66592=y
 CONFIG_SUPERH_BUILT_IN_M66592=y
 # CONFIG_USB_GADGET_AMD5536UDC is not set
 # CONFIG_USB_GADGET_FSL_QE is not set
+# CONFIG_USB_GADGET_CI13XXX is not set
 # CONFIG_USB_GADGET_NET2280 is not set
 # CONFIG_USB_GADGET_GOKU is not set
 # CONFIG_USB_GADGET_DUMMY_HCD is not set
@@ -883,6 +900,11 @@ CONFIG_USB_G_SERIAL=y
 # CONFIG_USB_MIDI_GADGET is not set
 # CONFIG_USB_G_PRINTER is not set
 # CONFIG_USB_CDC_COMPOSITE is not set
+
+#
+# OTG and related infrastructure
+#
+# CONFIG_USB_GPIO_VBUS is not set
 # CONFIG_MMC is not set
 # CONFIG_MEMSTICK is not set
 # CONFIG_NEW_LEDS is not set
@@ -961,6 +983,7 @@ CONFIG_UIO_PDRV_GENIRQ=y
 CONFIG_FILE_LOCKING=y
 # CONFIG_XFS_FS is not set
 # CONFIG_OCFS2_FS is not set
+# CONFIG_BTRFS_FS is not set
 # CONFIG_DNOTIFY is not set
 # CONFIG_INOTIFY is not set
 # CONFIG_QUOTA is not set
@@ -1004,6 +1027,7 @@ CONFIG_MISC_FILESYSTEMS=y
 # CONFIG_EFS_FS is not set
 # CONFIG_JFFS2_FS is not set
 # CONFIG_CRAMFS is not set
+# CONFIG_SQUASHFS is not set
 # CONFIG_VXFS_FS is not set
 # CONFIG_MINIX_FS is not set
 # CONFIG_OMFS_FS is not set

arch/sh/include/asm/mutex-llsc.h

@@ -21,38 +21,36 @@
 static inline void
 __mutex_fastpath_lock(atomic_t *count, void (*fail_fn)(atomic_t *))
 {
-	int __ex_flag, __res;
+	int __done, __res;
 
 	__asm__ __volatile__ (
 		"movli.l	@%2, %0	\n"
 		"add		#-1, %0	\n"
 		"movco.l	%0, @%2	\n"
 		"movt		%1	\n"
-		: "=&z" (__res), "=&r" (__ex_flag)
+		: "=&z" (__res), "=&r" (__done)
 		: "r" (&(count)->counter)
 		: "t");
 
-	__res |= !__ex_flag;
-	if (unlikely(__res != 0))
+	if (unlikely(!__done || __res != 0))
 		fail_fn(count);
 }
 
 static inline int
 __mutex_fastpath_lock_retval(atomic_t *count, int (*fail_fn)(atomic_t *))
 {
-	int __ex_flag, __res;
+	int __done, __res;
 
 	__asm__ __volatile__ (
 		"movli.l	@%2, %0	\n"
 		"add		#-1, %0	\n"
 		"movco.l	%0, @%2	\n"
 		"movt		%1	\n"
-		: "=&z" (__res), "=&r" (__ex_flag)
+		: "=&z" (__res), "=&r" (__done)
 		: "r" (&(count)->counter)
 		: "t");
 
-	__res |= !__ex_flag;
-	if (unlikely(__res != 0))
+	if (unlikely(!__done || __res != 0))
 		__res = fail_fn(count);
 
 	return __res;
@@ -61,19 +59,18 @@ __mutex_fastpath_lock_retval(atomic_t *count, int (*fail_fn)(atomic_t *))
 static inline void
 __mutex_fastpath_unlock(atomic_t *count, void (*fail_fn)(atomic_t *))
 {
-	int __ex_flag, __res;
+	int __done, __res;
 
 	__asm__ __volatile__ (
 		"movli.l	@%2, %0	\n\t"
 		"add		#1, %0	\n\t"
 		"movco.l	%0, @%2 \n\t"
 		"movt		%1	\n\t"
-		: "=&z" (__res), "=&r" (__ex_flag)
+		: "=&z" (__res), "=&r" (__done)
 		: "r" (&(count)->counter)
 		: "t");
 
-	__res |= !__ex_flag;
-	if (unlikely(__res <= 0))
+	if (unlikely(!__done || __res <= 0))
 		fail_fn(count);
 }
 

arch/sh/include/asm/syscall_32.h

@@ -21,23 +21,10 @@ static inline void syscall_rollback(struct task_struct *task,
 	 */
 }
 
-static inline bool syscall_has_error(struct pt_regs *regs)
-{
-	return (regs->sr & 0x1) ? true : false;
-}
-static inline void syscall_set_error(struct pt_regs *regs)
-{
-	regs->sr |= 0x1;
-}
-static inline void syscall_clear_error(struct pt_regs *regs)
-{
-	regs->sr &= ~0x1;
-}
-
 static inline long syscall_get_error(struct task_struct *task,
 				     struct pt_regs *regs)
 {
-	return syscall_has_error(regs) ? regs->regs[0] : 0;
+	return IS_ERR_VALUE(regs->regs[0]) ? regs->regs[0] : 0;
 }
 
 static inline long syscall_get_return_value(struct task_struct *task,
@@ -50,13 +37,10 @@ static inline void syscall_set_return_value(struct task_struct *task,
 					    struct pt_regs *regs,
 					    int error, long val)
 {
-	if (error) {
-		syscall_set_error(regs);
+	if (error)
 		regs->regs[0] = -error;
-	} else {
-		syscall_clear_error(regs);
+	else
 		regs->regs[0] = val;
-	}
 }
 
 static inline void syscall_get_arguments(struct task_struct *task,

arch/sh/include/asm/syscall_64.h

@@ -21,23 +21,10 @@ static inline void syscall_rollback(struct task_struct *task,
 	 */
 }
 
-static inline bool syscall_has_error(struct pt_regs *regs)
-{
-	return (regs->sr & 0x1) ? true : false;
-}
-static inline void syscall_set_error(struct pt_regs *regs)
-{
-	regs->sr |= 0x1;
-}
-static inline void syscall_clear_error(struct pt_regs *regs)
-{
-	regs->sr &= ~0x1;
-}
-
 static inline long syscall_get_error(struct task_struct *task,
 				     struct pt_regs *regs)
 {
-	return syscall_has_error(regs) ? regs->regs[9] : 0;
+	return IS_ERR_VALUE(regs->regs[9]) ? regs->regs[9] : 0;
 }
 
 static inline long syscall_get_return_value(struct task_struct *task,
@@ -50,13 +37,10 @@ static inline void syscall_set_return_value(struct task_struct *task,
 					    struct pt_regs *regs,
 					    int error, long val)
 {
-	if (error) {
-		syscall_set_error(regs);
+	if (error)
 		regs->regs[9] = -error;
-	} else {
-		syscall_clear_error(regs);
+	else
 		regs->regs[9] = val;
-	}
 }
 
 static inline void syscall_get_arguments(struct task_struct *task,

arch/sh/kernel/cpu/sh4/fpu.c

@@ -423,7 +423,7 @@ static int ieee_fpe_handler(struct pt_regs *regs)
 		int m;
 		unsigned int hx;
 
-		m = (finsn >> 9) & 0x7;
+		m = (finsn >> 8) & 0x7;
 		hx = tsk->thread.fpu.hard.fp_regs[m];
 
 		if ((tsk->thread.fpu.hard.fpscr & FPSCR_CAUSE_ERROR)

arch/sh/kernel/setup.c

@@ -262,11 +262,11 @@ void __init setup_bootmem_allocator(unsigned long free_pfn)
 			BOOTMEM_DEFAULT);
 
 	/*
-	 * reserve physical page 0 - it's a special BIOS page on many boxes,
-	 * enabling clean reboots, SMP operation, laptop functions.
+	 * Reserve physical pages below CONFIG_ZERO_PAGE_OFFSET.
 	 */
-	reserve_bootmem(__MEMORY_START, CONFIG_ZERO_PAGE_OFFSET,
-			BOOTMEM_DEFAULT);
+	if (CONFIG_ZERO_PAGE_OFFSET != 0)
+		reserve_bootmem(__MEMORY_START, CONFIG_ZERO_PAGE_OFFSET,
+				BOOTMEM_DEFAULT);
 
 	sparse_memory_present_with_active_regions(0);
 

arch/sh/kernel/signal_32.c

@@ -510,7 +510,6 @@ handle_syscall_restart(unsigned long save_r0, struct pt_regs *regs,
 		case -ERESTARTNOHAND:
 		no_system_call_restart:
 			regs->regs[0] = -EINTR;
-			regs->sr |= 1;
 			break;
 
 		case -ERESTARTSYS:
@@ -589,8 +588,7 @@ static void do_signal(struct pt_regs *regs, unsigned int save_r0)
 
 	signr = get_signal_to_deliver(&info, &ka, regs, NULL);
 	if (signr > 0) {
-		if (regs->sr & 1)
-			handle_syscall_restart(save_r0, regs, &ka.sa);
+		handle_syscall_restart(save_r0, regs, &ka.sa);
 
 		/* Whee!  Actually deliver the signal.  */
 		if (handle_signal(signr, &ka, &info, oldset,

arch/sh/kernel/signal_64.c

@@ -60,7 +60,6 @@ handle_syscall_restart(struct pt_regs *regs, struct sigaction *sa)
 		case -ERESTARTNOHAND:
 		no_system_call_restart:
 			regs->regs[REG_RET] = -EINTR;
-			regs->sr |= 1;
 			break;
 
 		case -ERESTARTSYS:
@@ -109,8 +108,7 @@ static int do_signal(struct pt_regs *regs, sigset_t *oldset)
 
 	signr = get_signal_to_deliver(&info, &ka, regs, 0);
 	if (signr > 0) {
-		if (regs->sr & 1)
-			handle_syscall_restart(regs, &ka.sa);
+		handle_syscall_restart(regs, &ka.sa);
 
 		/* Whee!  Actually deliver the signal.  */
 		if (handle_signal(signr, &info, &ka, oldset, regs) == 0) {

arch/sh/lib/checksum.S

@@ -36,8 +36,7 @@
  */
 
 /*	
- * unsigned int csum_partial(const unsigned char *buf, int len,
- *                           unsigned int sum);
+ * asmlinkage __wsum csum_partial(const void *buf, int len, __wsum sum);
  */
 
 .text
@@ -49,11 +48,31 @@ ENTRY(csum_partial)
 	   * Fortunately, it is easy to convert 2-byte alignment to 4-byte
 	   * alignment for the unrolled loop.
 	   */
-	mov	r5, r1
 	mov	r4, r0
-	tst	#2, r0		! Check alignment.
-	bt	2f		! Jump if alignment is ok.
+	tst	#3, r0		! Check alignment.
+	bt/s	2f		! Jump if alignment is ok.
+	 mov	r4, r7		! Keep a copy to check for alignment
 	!
+	tst	#1, r0		! Check alignment.
+	bt	21f		! Jump if alignment is boundary of 2bytes.
+
+	! buf is odd
+	tst	r5, r5
+	add	#-1, r5
+	bt	9f
+	mov.b	@r4+, r0
+	extu.b	r0, r0
+	addc	r0, r6		! t=0 from previous tst
+	mov	r6, r0
+	shll8	r6
+	shlr16	r0
+	shlr8	r0
+	or	r0, r6
+	mov	r4, r0
+	tst	#2, r0
+	bt	2f
+21:
+	! buf is 2 byte aligned (len could be 0)
 	add	#-2, r5		! Alignment uses up two bytes.
 	cmp/pz	r5		!
 	bt/s	1f		! Jump if we had at least two bytes.
@@ -61,16 +80,17 @@ ENTRY(csum_partial)
 	bra	6f
 	 add	#2, r5		! r5 was < 2.  Deal with it.
 1:
-	mov	r5, r1		! Save new len for later use.
 	mov.w	@r4+, r0
 	extu.w	r0, r0
 	addc	r0, r6
 	bf	2f
 	add	#1, r6
 2:
+	! buf is 4 byte aligned (len could be 0)
+	mov	r5, r1
 	mov	#-5, r0
-	shld	r0, r5
-	tst	r5, r5
+	shld	r0, r1
+	tst	r1, r1
 	bt/s	4f		! if it's =0, go to 4f
 	 clrt
 	.align	2
@@ -92,30 +112,31 @@ ENTRY(csum_partial)
 	addc	r0, r6
 	addc	r2, r6
 	movt	r0
-	dt	r5
+	dt	r1
 	bf/s	3b
 	 cmp/eq	#1, r0
-	! here, we know r5==0
-	addc	r5, r6			! add carry to r6
+	! here, we know r1==0
+	addc	r1, r6			! add carry to r6
 4:
-	mov	r1, r0
+	mov	r5, r0
 	and	#0x1c, r0
 	tst	r0, r0
-	bt/s	6f
-	 mov	r0, r5
-	shlr2	r5
+	bt	6f
+	! 4 bytes or more remaining
+	mov	r0, r1
+	shlr2	r1
 	mov	#0, r2
 5:
 	addc	r2, r6
 	mov.l	@r4+, r2
 	movt	r0
-	dt	r5
+	dt	r1
 	bf/s	5b
 	 cmp/eq	#1, r0
 	addc	r2, r6
-	addc	r5, r6		! r5==0 here, so it means add carry-bit
+	addc	r1, r6		! r1==0 here, so it means add carry-bit
 6:
-	mov	r1, r5
+	! 3 bytes or less remaining
 	mov	#3, r0
 	and	r0, r5
 	tst	r5, r5
@@ -139,8 +160,18 @@ ENTRY(csum_partial)
 8:
 	addc	r0, r6
 	mov	#0, r0
-	addc	r0, r6 
+	addc	r0, r6
 9:
+	! Check if the buffer was misaligned, if so realign sum
+	mov	r7, r0
+	tst	#1, r0
+	bt	10f
+	mov	r6, r0
+	shll8	r6
+	shlr16	r0
+	shlr8	r0
+	or	r0, r6
+10:
 	rts
 	 mov	r6, r0
 

arch/x86/ia32/ia32entry.S

@@ -418,9 +418,9 @@ ENTRY(ia32_syscall)
 	orl   $TS_COMPAT,TI_status(%r10)
 	testl $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%r10)
 	jnz ia32_tracesys
-ia32_do_syscall:	
 	cmpl $(IA32_NR_syscalls-1),%eax
-	ja  int_ret_from_sys_call	/* ia32_tracesys has set RAX(%rsp) */
+	ja ia32_badsys
+ia32_do_call:
 	IA32_ARG_FIXUP
 	call *ia32_sys_call_table(,%rax,8) # xxx: rip relative
 ia32_sysret:
@@ -435,7 +435,9 @@ ia32_tracesys:
 	call syscall_trace_enter
 	LOAD_ARGS32 ARGOFFSET  /* reload args from stack in case ptrace changed it */
 	RESTORE_REST
-	jmp ia32_do_syscall
+	cmpl $(IA32_NR_syscalls-1),%eax
+	ja  int_ret_from_sys_call	/* ia32_tracesys has set RAX(%rsp) */
+	jmp ia32_do_call
 END(ia32_syscall)
 
 ia32_badsys:

arch/x86/kernel/acpi/sleep.c

@@ -156,11 +156,11 @@ static int __init acpi_sleep_setup(char *str)
 #ifdef CONFIG_HIBERNATION
 		if (strncmp(str, "s4_nohwsig", 10) == 0)
 			acpi_no_s4_hw_signature();
+		if (strncmp(str, "s4_nonvs", 8) == 0)
+			acpi_s4_no_nvs();
 #endif
 		if (strncmp(str, "old_ordering", 12) == 0)
 			acpi_old_suspend_ordering();
-		if (strncmp(str, "s4_nonvs", 8) == 0)
-			acpi_s4_no_nvs();
 		str = strchr(str, ',');
 		if (str != NULL)
 			str += strspn(str, ", \t");

arch/x86/kernel/apic.c

@@ -1436,7 +1436,7 @@ static int __init detect_init_APIC(void)
 	switch (boot_cpu_data.x86_vendor) {
 	case X86_VENDOR_AMD:
 		if ((boot_cpu_data.x86 == 6 && boot_cpu_data.x86_model > 1) ||
-		    (boot_cpu_data.x86 == 15))
+		    (boot_cpu_data.x86 >= 15))
 			break;
 		goto no_apic;
 	case X86_VENDOR_INTEL:

arch/x86/kernel/cpu/cpufreq/Kconfig

@@ -245,17 +245,6 @@ config X86_E_POWERSAVER
 
 comment "shared options"
 
-config X86_ACPI_CPUFREQ_PROC_INTF
-	bool "/proc/acpi/processor/../performance interface (deprecated)"
-	depends on PROC_FS
-	depends on X86_ACPI_CPUFREQ || X86_POWERNOW_K7_ACPI || X86_POWERNOW_K8_ACPI
-	help
-	  This enables the deprecated /proc/acpi/processor/../performance
-	  interface. While it is helpful for debugging, the generic,
-	  cross-architecture cpufreq interfaces should be used.
-
-	  If in doubt, say N.
-
 config X86_SPEEDSTEP_LIB
 	tristate
 	default (X86_SPEEDSTEP_ICH || X86_SPEEDSTEP_SMI || X86_P4_CLOCKMOD)

arch/x86/kernel/cpu/intel_cacheinfo.c

@@ -36,8 +36,11 @@ static struct _cache_table cache_table[] __cpuinitdata =
 {
 	{ 0x06, LVL_1_INST, 8 },	/* 4-way set assoc, 32 byte line size */
 	{ 0x08, LVL_1_INST, 16 },	/* 4-way set assoc, 32 byte line size */
+	{ 0x09, LVL_1_INST, 32 },	/* 4-way set assoc, 64 byte line size */
 	{ 0x0a, LVL_1_DATA, 8 },	/* 2 way set assoc, 32 byte line size */
 	{ 0x0c, LVL_1_DATA, 16 },	/* 4-way set assoc, 32 byte line size */
+	{ 0x0d, LVL_1_DATA, 16 },	/* 4-way set assoc, 64 byte line size */
+	{ 0x21, LVL_2,      256 },	/* 8-way set assoc, 64 byte line size */
 	{ 0x22, LVL_3,      512 },	/* 4-way set assoc, sectored cache, 64 byte line size */
 	{ 0x23, LVL_3,      1024 },	/* 8-way set assoc, sectored cache, 64 byte line size */
 	{ 0x25, LVL_3,      2048 },	/* 8-way set assoc, sectored cache, 64 byte line size */
@@ -85,6 +88,18 @@ static struct _cache_table cache_table[] __cpuinitdata =
 	{ 0x85, LVL_2,    2048 },	/* 8-way set assoc, 32 byte line size */
 	{ 0x86, LVL_2,     512 },	/* 4-way set assoc, 64 byte line size */
 	{ 0x87, LVL_2,    1024 },	/* 8-way set assoc, 64 byte line size */
+	{ 0xd0, LVL_3,     512 },	/* 4-way set assoc, 64 byte line size */
+	{ 0xd1, LVL_3,    1024 },	/* 4-way set assoc, 64 byte line size */
+	{ 0xd2, LVL_3,    2048 },	/* 4-way set assoc, 64 byte line size */
+	{ 0xd6, LVL_3,    1024 },	/* 8-way set assoc, 64 byte line size */
+	{ 0xd7, LVL_3,    2038 },	/* 8-way set assoc, 64 byte line size */
+	{ 0xd8, LVL_3,    4096 },	/* 12-way set assoc, 64 byte line size */
+	{ 0xdc, LVL_3,    2048 },	/* 12-way set assoc, 64 byte line size */
+	{ 0xdd, LVL_3,    4096 },	/* 12-way set assoc, 64 byte line size */
+	{ 0xde, LVL_3,    8192 },	/* 12-way set assoc, 64 byte line size */
+	{ 0xe2, LVL_3,    2048 },	/* 16-way set assoc, 64 byte line size */
+	{ 0xe3, LVL_3,    4096 },	/* 16-way set assoc, 64 byte line size */
+	{ 0xe4, LVL_3,    8192 },	/* 16-way set assoc, 64 byte line size */
 	{ 0x00, 0, 0}
 };
 

arch/x86/kernel/entry_64.S

@@ -346,6 +346,7 @@ ENTRY(save_args)
 	popq_cfi %rax			/* move return address... */
 	mov %gs:pda_irqstackptr,%rsp
 	EMPTY_FRAME 0
+	pushq_cfi %rbp			/* backlink for unwinder */
 	pushq_cfi %rax			/* ... to the new stack */
 	/*
 	 * We entered an interrupt context - irqs are off:

arch/x86/kernel/io_apic.c

@@ -2528,14 +2528,15 @@ static void irq_complete_move(struct irq_desc **descp)
 
 	vector = ~get_irq_regs()->orig_ax;
 	me = smp_processor_id();
+
+	if (vector == cfg->vector && cpumask_test_cpu(me, cfg->domain)) {
 #ifdef CONFIG_NUMA_MIGRATE_IRQ_DESC
 		*descp = desc = move_irq_desc(desc, me);
 		/* get the new one */
 		cfg = desc->chip_data;
 #endif
-
-	if (vector == cfg->vector && cpumask_test_cpu(me, cfg->domain))
 		send_cleanup_vector(cfg);
+	}
 }
 #else
 static inline void irq_complete_move(struct irq_desc **descp) {}

arch/x86/kernel/irqinit_32.c

@@ -78,15 +78,6 @@ void __init init_ISA_irqs(void)
 	}
 }
 
-/*
- * IRQ2 is cascade interrupt to second interrupt controller
- */
-static struct irqaction irq2 = {
-	.handler = no_action,
-	.mask = CPU_MASK_NONE,
-	.name = "cascade",
-};
-
 DEFINE_PER_CPU(vector_irq_t, vector_irq) = {
 	[0 ... IRQ0_VECTOR - 1] = -1,
 	[IRQ0_VECTOR] = 0,
@@ -178,9 +169,6 @@ void __init native_init_IRQ(void)
 	alloc_intr_gate(THERMAL_APIC_VECTOR, thermal_interrupt);
 #endif
 
-	if (!acpi_ioapic)
-		setup_irq(2, &irq2);
-
 	/* setup after call gates are initialised (usually add in
 	 * the architecture specific gates)
 	 */

arch/x86/mach-default/setup.c

@@ -38,6 +38,15 @@ void __init pre_intr_init_hook(void)
 	init_ISA_irqs();
 }
 
+/*
+ * IRQ2 is cascade interrupt to second interrupt controller
+ */
+static struct irqaction irq2 = {
+	.handler = no_action,
+	.mask = CPU_MASK_NONE,
+	.name = "cascade",
+};
+
 /**
  * intr_init_hook - post gate setup interrupt initialisation
  *
@@ -53,6 +62,9 @@ void __init intr_init_hook(void)
 		if (x86_quirks->arch_intr_init())
 			return;
 	}
+	if (!acpi_ioapic)
+		setup_irq(2, &irq2);
+
 }
 
 /**

arch/x86/mach-voyager/setup.c

@@ -33,13 +33,23 @@ void __init intr_init_hook(void)
 	setup_irq(2, &irq2);
 }
 
-void __init pre_setup_arch_hook(void)
+static void voyager_disable_tsc(void)
 {
 	/* Voyagers run their CPUs from independent clocks, so disable
 	 * the TSC code because we can't sync them */
 	setup_clear_cpu_cap(X86_FEATURE_TSC);
 }
 
+void __init pre_setup_arch_hook(void)
+{
+	voyager_disable_tsc();
+}
+
+void __init pre_time_init_hook(void)
+{
+	voyager_disable_tsc();
+}
+
 void __init trap_init_hook(void)
 {
 }

arch/x86/mach-voyager/voyager_smp.c

@@ -81,7 +81,7 @@ static void enable_local_vic_irq(unsigned int irq);
 static void disable_local_vic_irq(unsigned int irq);
 static void before_handle_vic_irq(unsigned int irq);
 static void after_handle_vic_irq(unsigned int irq);
-static void set_vic_irq_affinity(unsigned int irq, cpumask_t mask);
+static void set_vic_irq_affinity(unsigned int irq, const struct cpumask *mask);
 static void ack_vic_irq(unsigned int irq);
 static void vic_enable_cpi(void);
 static void do_boot_cpu(__u8 cpuid);
@@ -211,8 +211,6 @@ static __u32 cpu_booted_map;
 static cpumask_t smp_commenced_mask = CPU_MASK_NONE;
 
 /* This is for the new dynamic CPU boot code */
-cpumask_t cpu_callin_map = CPU_MASK_NONE;
-cpumask_t cpu_callout_map = CPU_MASK_NONE;
 
 /* The per processor IRQ masks (these are usually kept in sync) */
 static __u16 vic_irq_mask[NR_CPUS] __cacheline_aligned;
@@ -378,7 +376,7 @@ void __init find_smp_config(void)
 	cpus_addr(phys_cpu_present_map)[0] |=
 	    voyager_extended_cmos_read(VOYAGER_PROCESSOR_PRESENT_MASK +
 				       3) << 24;
-	cpu_possible_map = phys_cpu_present_map;
+	init_cpu_possible(&phys_cpu_present_map);
 	printk("VOYAGER SMP: phys_cpu_present_map = 0x%lx\n",
 	       cpus_addr(phys_cpu_present_map)[0]);
 	/* Here we set up the VIC to enable SMP */
@@ -1599,16 +1597,16 @@ static void after_handle_vic_irq(unsigned int irq)
  * change the mask and then do an interrupt enable CPI to re-enable on
  * the selected processors */
 
-void set_vic_irq_affinity(unsigned int irq, cpumask_t mask)
+void set_vic_irq_affinity(unsigned int irq, const struct cpumask *mask)
 {
 	/* Only extended processors handle interrupts */
 	unsigned long real_mask;
 	unsigned long irq_mask = 1 << irq;
 	int cpu;
 
-	real_mask = cpus_addr(mask)[0] & voyager_extended_vic_processors;
+	real_mask = cpus_addr(*mask)[0] & voyager_extended_vic_processors;
 
-	if (cpus_addr(mask)[0] == 0)
+	if (cpus_addr(*mask)[0] == 0)
 		/* can't have no CPUs to accept the interrupt -- extremely
 		 * bad things will happen */
 		return;
@@ -1750,10 +1748,11 @@ static void __cpuinit voyager_smp_prepare_boot_cpu(void)
 	init_gdt(smp_processor_id());
 	switch_to_new_gdt();
 
-	cpu_set(smp_processor_id(), cpu_online_map);
-	cpu_set(smp_processor_id(), cpu_callout_map);
-	cpu_set(smp_processor_id(), cpu_possible_map);
-	cpu_set(smp_processor_id(), cpu_present_map);
+	cpu_online_map = cpumask_of_cpu(smp_processor_id());
+	cpu_callout_map = cpumask_of_cpu(smp_processor_id());
+	cpu_callin_map = CPU_MASK_NONE;
+	cpu_present_map = cpumask_of_cpu(smp_processor_id());
+
 }
 
 static int __cpuinit voyager_cpu_up(unsigned int cpu)
@@ -1783,9 +1782,9 @@ void __init smp_setup_processor_id(void)
 	x86_write_percpu(cpu_number, hard_smp_processor_id());
 }
 
-static void voyager_send_call_func(cpumask_t callmask)
+static void voyager_send_call_func(const struct cpumask *callmask)
 {
-	__u32 mask = cpus_addr(callmask)[0] & ~(1 << smp_processor_id());
+	__u32 mask = cpus_addr(*callmask)[0] & ~(1 << smp_processor_id());
 	send_CPI(mask, VIC_CALL_FUNCTION_CPI);
 }
 

arch/x86/mm/fault.c

@@ -603,8 +603,6 @@ void __kprobes do_page_fault(struct pt_regs *regs, unsigned long error_code)
 
 	si_code = SEGV_MAPERR;
 
-	if (notify_page_fault(regs))
-		return;
 	if (unlikely(kmmio_fault(regs, address)))
 		return;
 
@@ -634,6 +632,9 @@ void __kprobes do_page_fault(struct pt_regs *regs, unsigned long error_code)
 		if (spurious_fault(address, error_code))
 			return;
 
+		/* kprobes don't want to hook the spurious faults. */
+		if (notify_page_fault(regs))
+			return;
 		/*
 		 * Don't take the mm semaphore here. If we fixup a prefetch
 		 * fault we could otherwise deadlock.
@@ -641,6 +642,9 @@ void __kprobes do_page_fault(struct pt_regs *regs, unsigned long error_code)
 		goto bad_area_nosemaphore;
 	}
 
+	/* kprobes don't want to hook the spurious faults. */
+	if (notify_page_fault(regs))
+		return;
 
 	/*
 	 * It's safe to allow irq's after cr2 has been saved and the

arch/x86/xen/multicalls.h

@@ -19,8 +19,10 @@ DECLARE_PER_CPU(unsigned long, xen_mc_irq_flags);
    paired with xen_mc_issue() */
 static inline void xen_mc_batch(void)
 {
+	unsigned long flags;
 	/* need to disable interrupts until this entry is complete */
-	local_irq_save(__get_cpu_var(xen_mc_irq_flags));
+	local_irq_save(flags);
+	__get_cpu_var(xen_mc_irq_flags) = flags;
 }
 
 static inline struct multicall_space xen_mc_entry(size_t args)

drivers/acpi/Kconfig

@@ -9,6 +9,7 @@ menuconfig ACPI
 	depends on PCI
 	depends on PM
 	select PNP
+	select CPU_IDLE
 	default y
 	---help---
 	  Advanced Configuration and Power Interface (ACPI) support for 
@@ -287,7 +288,7 @@ config ACPI_CONTAINER
 	  support physical cpu/memory hot-plug.
 
 	  If one selects "m", this driver can be loaded with
-	  "modprobe acpi_container".
+	  "modprobe container".
 
 config ACPI_HOTPLUG_MEMORY
 	tristate "Memory Hotplug"

drivers/acpi/acpica/tbutils.c

@@ -538,10 +538,9 @@ acpi_tb_parse_root_table(acpi_physical_address rsdp_address, u8 flags)
 			if (ACPI_FAILURE(status)) {
 				ACPI_WARNING((AE_INFO,
 					      "Truncating %u table entries!",
-					      (unsigned)
-					      (acpi_gbl_root_table_list.size -
-					       acpi_gbl_root_table_list.
-					       count)));
+					      (unsigned) (table_count -
+					       (acpi_gbl_root_table_list.
+					       count - 2))));
 				break;
 			}
 		}

drivers/acpi/acpica/uteval.c

@@ -116,9 +116,9 @@ acpi_status acpi_ut_osi_implementation(struct acpi_walk_state *walk_state)
 		return_ACPI_STATUS(AE_NO_MEMORY);
 	}
 
-	/* Default return value is SUPPORTED */
+	/* Default return value is 0, NOT-SUPPORTED */
 
-	return_desc->integer.value = ACPI_UINT32_MAX;
+	return_desc->integer.value = 0;
 	walk_state->return_desc = return_desc;
 
 	/* Compare input string to static table of supported interfaces */
@@ -127,10 +127,8 @@ acpi_status acpi_ut_osi_implementation(struct acpi_walk_state *walk_state)
 		if (!ACPI_STRCMP
 		    (string_desc->string.pointer,
 		     acpi_interfaces_supported[i])) {
-
-			/* The interface is supported */
-
-			return_ACPI_STATUS(AE_OK);
+			return_desc->integer.value = ACPI_UINT32_MAX;
+			goto done;
 		}
 	}
 
@@ -141,15 +139,14 @@ acpi_status acpi_ut_osi_implementation(struct acpi_walk_state *walk_state)
 	 */
 	status = acpi_os_validate_interface(string_desc->string.pointer);
 	if (ACPI_SUCCESS(status)) {
-
-		/* The interface is supported */
-
-		return_ACPI_STATUS(AE_OK);
+		return_desc->integer.value = ACPI_UINT32_MAX;
 	}
 
-	/* The interface is not supported */
+done:
+	ACPI_DEBUG_PRINT_RAW((ACPI_DB_INFO, "ACPI: BIOS _OSI(%s) %ssupported\n",
+		string_desc->string.pointer,
+		return_desc->integer.value == 0 ? "not-" : ""));
 
-	return_desc->integer.value = 0;
 	return_ACPI_STATUS(AE_OK);
 }
 

drivers/acpi/container.c

@@ -163,7 +163,7 @@ static void container_notify_cb(acpi_handle handle, u32 type, void *context)
 	case ACPI_NOTIFY_BUS_CHECK:
 		/* Fall through */
 	case ACPI_NOTIFY_DEVICE_CHECK:
-		printk("Container driver received %s event\n",
+		printk(KERN_WARNING "Container driver received %s event\n",
 		       (type == ACPI_NOTIFY_BUS_CHECK) ?
 		       "ACPI_NOTIFY_BUS_CHECK" : "ACPI_NOTIFY_DEVICE_CHECK");
 		status = acpi_bus_get_device(handle, &device);
@@ -174,7 +174,8 @@ static void container_notify_cb(acpi_handle handle, u32 type, void *context)
 					kobject_uevent(&device->dev.kobj,
 						       KOBJ_ONLINE);
 				else
-					printk("Failed to add container\n");
+					printk(KERN_WARNING
+					       "Failed to add container\n");
 			}
 		} else {
 			if (ACPI_SUCCESS(status)) {

drivers/acpi/dock.c

@@ -855,10 +855,14 @@ fdd_out:
 static ssize_t show_docked(struct device *dev,
 			   struct device_attribute *attr, char *buf)
 {
+	struct acpi_device *tmp;
+
 	struct dock_station *dock_station = *((struct dock_station **)
 		dev->platform_data);
-	return snprintf(buf, PAGE_SIZE, "%d\n", dock_present(dock_station));
 
+	if (ACPI_SUCCESS(acpi_bus_get_device(dock_station->handle, &tmp)))
+		return snprintf(buf, PAGE_SIZE, "1\n");
+	return snprintf(buf, PAGE_SIZE, "0\n");
 }
 static DEVICE_ATTR(docked, S_IRUGO, show_docked, NULL);
 
@@ -984,7 +988,7 @@ static int dock_add(acpi_handle handle)
 
 	ret = device_create_file(&dock_device->dev, &dev_attr_docked);
 	if (ret) {
-		printk("Error %d adding sysfs file\n", ret);
+		printk(KERN_ERR "Error %d adding sysfs file\n", ret);
 		platform_device_unregister(dock_device);
 		kfree(dock_station);
 		dock_station = NULL;
@@ -992,7 +996,7 @@ static int dock_add(acpi_handle handle)
 	}
 	ret = device_create_file(&dock_device->dev, &dev_attr_undock);
 	if (ret) {
-		printk("Error %d adding sysfs file\n", ret);
+		printk(KERN_ERR "Error %d adding sysfs file\n", ret);
 		device_remove_file(&dock_device->dev, &dev_attr_docked);
 		platform_device_unregister(dock_device);
 		kfree(dock_station);
@@ -1001,7 +1005,7 @@ static int dock_add(acpi_handle handle)
 	}
 	ret = device_create_file(&dock_device->dev, &dev_attr_uid);
 	if (ret) {
-		printk("Error %d adding sysfs file\n", ret);
+		printk(KERN_ERR "Error %d adding sysfs file\n", ret);
 		device_remove_file(&dock_device->dev, &dev_attr_docked);
 		device_remove_file(&dock_device->dev, &dev_attr_undock);
 		platform_device_unregister(dock_device);
@@ -1011,7 +1015,7 @@ static int dock_add(acpi_handle handle)
 	}
 	ret = device_create_file(&dock_device->dev, &dev_attr_flags);
 	if (ret) {
-		printk("Error %d adding sysfs file\n", ret);
+		printk(KERN_ERR "Error %d adding sysfs file\n", ret);
 		device_remove_file(&dock_device->dev, &dev_attr_docked);
 		device_remove_file(&dock_device->dev, &dev_attr_undock);
 		device_remove_file(&dock_device->dev, &dev_attr_uid);

drivers/acpi/ec.c

@@ -982,7 +982,7 @@ int __init acpi_ec_ecdt_probe(void)
 		saved_ec = kmalloc(sizeof(struct acpi_ec), GFP_KERNEL);
 		if (!saved_ec)
 			return -ENOMEM;
-		memcpy(&saved_ec, boot_ec, sizeof(saved_ec));
+		memcpy(saved_ec, boot_ec, sizeof(*saved_ec));
 	/* fall through */
 	}
 	/* This workaround is needed only on some broken machines,

drivers/acpi/glue.c

@@ -255,12 +255,12 @@ static int acpi_platform_notify(struct device *dev)
 	}
 	type = acpi_get_bus_type(dev->bus);
 	if (!type) {
-		DBG("No ACPI bus support for %s\n", dev->bus_id);
+		DBG("No ACPI bus support for %s\n", dev_name(dev));
 		ret = -EINVAL;
 		goto end;
 	}
 	if ((ret = type->find_device(dev, &handle)) != 0)
-		DBG("Can't get handler for %s\n", dev->bus_id);
+		DBG("Can't get handler for %s\n", dev_name(dev));
       end:
 	if (!ret)
 		acpi_bind_one(dev, handle);
@@ -271,10 +271,10 @@ static int acpi_platform_notify(struct device *dev)
 
 		acpi_get_name(dev->archdata.acpi_handle,
 			      ACPI_FULL_PATHNAME, &buffer);
-		DBG("Device %s -> %s\n", dev->bus_id, (char *)buffer.pointer);
+		DBG("Device %s -> %s\n", dev_name(dev), (char *)buffer.pointer);
 		kfree(buffer.pointer);
 	} else
-		DBG("Device %s -> No ACPI support\n", dev->bus_id);
+		DBG("Device %s -> No ACPI support\n", dev_name(dev));
 #endif
 
 	return ret;

drivers/acpi/osl.c

@@ -228,10 +228,10 @@ void acpi_os_vprintf(const char *fmt, va_list args)
 	if (acpi_in_debugger) {
 		kdb_printf("%s", buffer);
 	} else {
-		printk("%s", buffer);
+		printk(KERN_CONT "%s", buffer);
 	}
 #else
-	printk("%s", buffer);
+	printk(KERN_CONT "%s", buffer);
 #endif
 }
 
@@ -1317,6 +1317,54 @@ acpi_os_validate_interface (char *interface)
 	return AE_SUPPORT;
 }
 
+#ifdef	CONFIG_X86
+
+struct aml_port_desc {
+	uint	start;
+	uint	end;
+	char*   name;
+	char	warned;
+};
+
+static struct aml_port_desc aml_invalid_port_list[] = {
+	{0x20, 0x21, "PIC0", 0},
+	{0xA0, 0xA1, "PIC1", 0},
+	{0x4D0, 0x4D1, "ELCR", 0}
+};
+
+/*
+ * valid_aml_io_address()
+ *
+ * if valid, return true
+ * else invalid, warn once, return false
+ */
+static bool valid_aml_io_address(uint address, uint length)
+{
+	int i;
+	int entries = sizeof(aml_invalid_port_list) / sizeof(struct aml_port_desc);
+
+	for (i = 0; i < entries; ++i) {
+		if ((address >= aml_invalid_port_list[i].start &&
+			address <= aml_invalid_port_list[i].end) ||
+			(address + length >= aml_invalid_port_list[i].start &&
+			address  + length <= aml_invalid_port_list[i].end))
+		{
+			if (!aml_invalid_port_list[i].warned)
+			{
+				printk(KERN_ERR "ACPI: Denied BIOS AML access"
+					" to invalid port 0x%x+0x%x (%s)\n",
+					address, length,
+					aml_invalid_port_list[i].name);
+				aml_invalid_port_list[i].warned = 1;
+			}
+			return false;	/* invalid */
+		}
+	}
+	return true;	/* valid */
+}
+#else
+static inline bool valid_aml_io_address(uint address, uint length) { return true; }
+#endif
 /******************************************************************************
  *
  * FUNCTION:    acpi_os_validate_address
@@ -1346,6 +1394,8 @@ acpi_os_validate_address (
 
 	switch (space_id) {
 	case ACPI_ADR_SPACE_SYSTEM_IO:
+		if (!valid_aml_io_address(address, length))
+			return AE_AML_ILLEGAL_ADDRESS;
 	case ACPI_ADR_SPACE_SYSTEM_MEMORY:
 		/* Only interference checks against SystemIO and SytemMemory
 		   are needed */

drivers/acpi/pci_link.c

@@ -593,7 +593,7 @@ static int acpi_pci_link_allocate(struct acpi_pci_link *link)
 		return -ENODEV;
 	} else {
 		acpi_irq_penalty[link->irq.active] += PIRQ_PENALTY_PCI_USING;
-		printk(PREFIX "%s [%s] enabled at IRQ %d\n",
+		printk(KERN_WARNING PREFIX "%s [%s] enabled at IRQ %d\n",
 		       acpi_device_name(link->device),
 		       acpi_device_bid(link->device), link->irq.active);
 	}

drivers/acpi/processor_idle.c

@@ -66,43 +66,17 @@ ACPI_MODULE_NAME("processor_idle");
 #define ACPI_PROCESSOR_FILE_POWER	"power"
 #define US_TO_PM_TIMER_TICKS(t)		((t * (PM_TIMER_FREQUENCY/1000)) / 1000)
 #define PM_TIMER_TICK_NS		(1000000000ULL/PM_TIMER_FREQUENCY)
-#ifndef CONFIG_CPU_IDLE
-#define C2_OVERHEAD			4	/* 1us (3.579 ticks per us) */
-#define C3_OVERHEAD			4	/* 1us (3.579 ticks per us) */
-static void (*pm_idle_save) (void) __read_mostly;
-#else
 #define C2_OVERHEAD			1	/* 1us */
 #define C3_OVERHEAD			1	/* 1us */
-#endif
 #define PM_TIMER_TICKS_TO_US(p)		(((p) * 1000)/(PM_TIMER_FREQUENCY/1000))
 
 static unsigned int max_cstate __read_mostly = ACPI_PROCESSOR_MAX_POWER;
-#ifdef CONFIG_CPU_IDLE
 module_param(max_cstate, uint, 0000);
-#else
-module_param(max_cstate, uint, 0644);
-#endif
 static unsigned int nocst __read_mostly;
 module_param(nocst, uint, 0000);
 
-#ifndef CONFIG_CPU_IDLE
-/*
- * bm_history -- bit-mask with a bit per jiffy of bus-master activity
- * 1000 HZ: 0xFFFFFFFF: 32 jiffies = 32ms
- * 800 HZ: 0xFFFFFFFF: 32 jiffies = 40ms
- * 100 HZ: 0x0000000F: 4 jiffies = 40ms
- * reduce history for more aggressive entry into C3
- */
-static unsigned int bm_history __read_mostly =
-    (HZ >= 800 ? 0xFFFFFFFF : ((1U << (HZ / 25)) - 1));
-module_param(bm_history, uint, 0644);
-
-static int acpi_processor_set_power_policy(struct acpi_processor *pr);
-
-#else	/* CONFIG_CPU_IDLE */
 static unsigned int latency_factor __read_mostly = 2;
 module_param(latency_factor, uint, 0644);
-#endif
 
 /*
  * IBM ThinkPad R40e crashes mysteriously when going into C2 or C3.
@@ -224,71 +198,6 @@ static void acpi_safe_halt(void)
 	current_thread_info()->status |= TS_POLLING;
 }
 
-#ifndef CONFIG_CPU_IDLE
-
-static void
-acpi_processor_power_activate(struct acpi_processor *pr,
-			      struct acpi_processor_cx *new)
-{
-	struct acpi_processor_cx *old;
-
-	if (!pr || !new)
-		return;
-
-	old = pr->power.state;
-
-	if (old)
-		old->promotion.count = 0;
-	new->demotion.count = 0;
-
-	/* Cleanup from old state. */
-	if (old) {
-		switch (old->type) {
-		case ACPI_STATE_C3:
-			/* Disable bus master reload */
-			if (new->type != ACPI_STATE_C3 && pr->flags.bm_check)
-				acpi_set_register(ACPI_BITREG_BUS_MASTER_RLD, 0);
-			break;
-		}
-	}
-
-	/* Prepare to use new state. */
-	switch (new->type) {
-	case ACPI_STATE_C3:
-		/* Enable bus master reload */
-		if (old->type != ACPI_STATE_C3 && pr->flags.bm_check)
-			acpi_set_register(ACPI_BITREG_BUS_MASTER_RLD, 1);
-		break;
-	}
-
-	pr->power.state = new;
-
-	return;
-}
-
-static atomic_t c3_cpu_count;
-
-/* Common C-state entry for C2, C3, .. */
-static void acpi_cstate_enter(struct acpi_processor_cx *cstate)
-{
-	/* Don't trace irqs off for idle */
-	stop_critical_timings();
-	if (cstate->entry_method == ACPI_CSTATE_FFH) {
-		/* Call into architectural FFH based C-state */
-		acpi_processor_ffh_cstate_enter(cstate);
-	} else {
-		int unused;
-		/* IO port based C-state */
-		inb(cstate->address);
-		/* Dummy wait op - must do something useless after P_LVL2 read
-		   because chipsets cannot guarantee that STPCLK# signal
-		   gets asserted in time to freeze execution properly. */
-		unused = inl(acpi_gbl_FADT.xpm_timer_block.address);
-	}
-	start_critical_timings();
-}
-#endif /* !CONFIG_CPU_IDLE */
-
 #ifdef ARCH_APICTIMER_STOPS_ON_C3
 
 /*
@@ -390,421 +299,6 @@ static int tsc_halts_in_c(int state)
 }
 #endif
 
-#ifndef CONFIG_CPU_IDLE
-static void acpi_processor_idle(void)
-{
-	struct acpi_processor *pr = NULL;
-	struct acpi_processor_cx *cx = NULL;
-	struct acpi_processor_cx *next_state = NULL;
-	int sleep_ticks = 0;
-	u32 t1, t2 = 0;
-
-	/*
-	 * Interrupts must be disabled during bus mastering calculations and
-	 * for C2/C3 transitions.
-	 */
-	local_irq_disable();
-
-	pr = __get_cpu_var(processors);
-	if (!pr) {
-		local_irq_enable();
-		return;
-	}
-
-	/*
-	 * Check whether we truly need to go idle, or should
-	 * reschedule:
-	 */
-	if (unlikely(need_resched())) {
-		local_irq_enable();
-		return;
-	}
-
-	cx = pr->power.state;
-	if (!cx || acpi_idle_suspend) {
-		if (pm_idle_save) {
-			pm_idle_save(); /* enables IRQs */
-		} else {
-			acpi_safe_halt();
-			local_irq_enable();
-		}
-
-		return;
-	}
-
-	/*
-	 * Check BM Activity
-	 * -----------------
-	 * Check for bus mastering activity (if required), record, and check
-	 * for demotion.
-	 */
-	if (pr->flags.bm_check) {
-		u32 bm_status = 0;
-		unsigned long diff = jiffies - pr->power.bm_check_timestamp;
-
-		if (diff > 31)
-			diff = 31;
-
-		pr->power.bm_activity <<= diff;
-
-		acpi_get_register(ACPI_BITREG_BUS_MASTER_STATUS, &bm_status);
-		if (bm_status) {
-			pr->power.bm_activity |= 0x1;
-			acpi_set_register(ACPI_BITREG_BUS_MASTER_STATUS, 1);
-		}
-		/*
-		 * PIIX4 Erratum #18: Note that BM_STS doesn't always reflect
-		 * the true state of bus mastering activity; forcing us to
-		 * manually check the BMIDEA bit of each IDE channel.
-		 */
-		else if (errata.piix4.bmisx) {
-			if ((inb_p(errata.piix4.bmisx + 0x02) & 0x01)
-			    || (inb_p(errata.piix4.bmisx + 0x0A) & 0x01))
-				pr->power.bm_activity |= 0x1;
-		}
-
-		pr->power.bm_check_timestamp = jiffies;
-
-		/*
-		 * If bus mastering is or was active this jiffy, demote
-		 * to avoid a faulty transition.  Note that the processor
-		 * won't enter a low-power state during this call (to this
-		 * function) but should upon the next.
-		 *
-		 * TBD: A better policy might be to fallback to the demotion
-		 *      state (use it for this quantum only) istead of
-		 *      demoting -- and rely on duration as our sole demotion
-		 *      qualification.  This may, however, introduce DMA
-		 *      issues (e.g. floppy DMA transfer overrun/underrun).
-		 */
-		if ((pr->power.bm_activity & 0x1) &&
-		    cx->demotion.threshold.bm) {
-			local_irq_enable();
-			next_state = cx->demotion.state;
-			goto end;
-		}
-	}
-
-#ifdef CONFIG_HOTPLUG_CPU
-	/*
-	 * Check for P_LVL2_UP flag before entering C2 and above on
-	 * an SMP system. We do it here instead of doing it at _CST/P_LVL
-	 * detection phase, to work cleanly with logical CPU hotplug.
-	 */
-	if ((cx->type != ACPI_STATE_C1) && (num_online_cpus() > 1) &&
-	    !pr->flags.has_cst && !(acpi_gbl_FADT.flags & ACPI_FADT_C2_MP_SUPPORTED))
-		cx = &pr->power.states[ACPI_STATE_C1];
-#endif
-
-	/*
-	 * Sleep:
-	 * ------
-	 * Invoke the current Cx state to put the processor to sleep.
-	 */
-	if (cx->type == ACPI_STATE_C2 || cx->type == ACPI_STATE_C3) {
-		current_thread_info()->status &= ~TS_POLLING;
-		/*
-		 * TS_POLLING-cleared state must be visible before we
-		 * test NEED_RESCHED:
-		 */
-		smp_mb();
-		if (need_resched()) {
-			current_thread_info()->status |= TS_POLLING;
-			local_irq_enable();
-			return;
-		}
-	}
-
-	switch (cx->type) {
-
-	case ACPI_STATE_C1:
-		/*
-		 * Invoke C1.
-		 * Use the appropriate idle routine, the one that would
-		 * be used without acpi C-states.
-		 */
-		if (pm_idle_save) {
-			pm_idle_save(); /* enables IRQs */
-		} else {
-			acpi_safe_halt();
-			local_irq_enable();
-		}
-
-		/*
-		 * TBD: Can't get time duration while in C1, as resumes
-		 *      go to an ISR rather than here.  Need to instrument
-		 *      base interrupt handler.
-		 *
-		 * Note: the TSC better not stop in C1, sched_clock() will
-		 *       skew otherwise.
-		 */
-		sleep_ticks = 0xFFFFFFFF;
-
-		break;
-
-	case ACPI_STATE_C2:
-		/* Get start time (ticks) */
-		t1 = inl(acpi_gbl_FADT.xpm_timer_block.address);
-		/* Tell the scheduler that we are going deep-idle: */
-		sched_clock_idle_sleep_event();
-		/* Invoke C2 */
-		acpi_state_timer_broadcast(pr, cx, 1);
-		acpi_cstate_enter(cx);
-		/* Get end time (ticks) */
-		t2 = inl(acpi_gbl_FADT.xpm_timer_block.address);
-
-#if defined (CONFIG_GENERIC_TIME) && defined (CONFIG_X86)
-		/* TSC halts in C2, so notify users */
-		if (tsc_halts_in_c(ACPI_STATE_C2))
-			mark_tsc_unstable("possible TSC halt in C2");
-#endif
-		/* Compute time (ticks) that we were actually asleep */
-		sleep_ticks = ticks_elapsed(t1, t2);
-
-		/* Tell the scheduler how much we idled: */
-		sched_clock_idle_wakeup_event(sleep_ticks*PM_TIMER_TICK_NS);
-
-		/* Re-enable interrupts */
-		local_irq_enable();
-		/* Do not account our idle-switching overhead: */
-		sleep_ticks -= cx->latency_ticks + C2_OVERHEAD;
-
-		current_thread_info()->status |= TS_POLLING;
-		acpi_state_timer_broadcast(pr, cx, 0);
-		break;
-
-	case ACPI_STATE_C3:
-		acpi_unlazy_tlb(smp_processor_id());
-		/*
-		 * Must be done before busmaster disable as we might
-		 * need to access HPET !
-		 */
-		acpi_state_timer_broadcast(pr, cx, 1);
-		/*
-		 * disable bus master
-		 * bm_check implies we need ARB_DIS
-		 * !bm_check implies we need cache flush
-		 * bm_control implies whether we can do ARB_DIS
-		 *
-		 * That leaves a case where bm_check is set and bm_control is
-		 * not set. In that case we cannot do much, we enter C3
-		 * without doing anything.
-		 */
-		if (pr->flags.bm_check && pr->flags.bm_control) {
-			if (atomic_inc_return(&c3_cpu_count) ==
-			    num_online_cpus()) {
-				/*
-				 * All CPUs are trying to go to C3
-				 * Disable bus master arbitration
-				 */
-				acpi_set_register(ACPI_BITREG_ARB_DISABLE, 1);
-			}
-		} else if (!pr->flags.bm_check) {
-			/* SMP with no shared cache... Invalidate cache  */
-			ACPI_FLUSH_CPU_CACHE();
-		}
-
-		/* Get start time (ticks) */
-		t1 = inl(acpi_gbl_FADT.xpm_timer_block.address);
-		/* Invoke C3 */
-		/* Tell the scheduler that we are going deep-idle: */
-		sched_clock_idle_sleep_event();
-		acpi_cstate_enter(cx);
-		/* Get end time (ticks) */
-		t2 = inl(acpi_gbl_FADT.xpm_timer_block.address);
-		if (pr->flags.bm_check && pr->flags.bm_control) {
-			/* Enable bus master arbitration */
-			atomic_dec(&c3_cpu_count);
-			acpi_set_register(ACPI_BITREG_ARB_DISABLE, 0);
-		}
-
-#if defined (CONFIG_GENERIC_TIME) && defined (CONFIG_X86)
-		/* TSC halts in C3, so notify users */
-		if (tsc_halts_in_c(ACPI_STATE_C3))
-			mark_tsc_unstable("TSC halts in C3");
-#endif
-		/* Compute time (ticks) that we were actually asleep */
-		sleep_ticks = ticks_elapsed(t1, t2);
-		/* Tell the scheduler how much we idled: */
-		sched_clock_idle_wakeup_event(sleep_ticks*PM_TIMER_TICK_NS);
-
-		/* Re-enable interrupts */
-		local_irq_enable();
-		/* Do not account our idle-switching overhead: */
-		sleep_ticks -= cx->latency_ticks + C3_OVERHEAD;
-
-		current_thread_info()->status |= TS_POLLING;
-		acpi_state_timer_broadcast(pr, cx, 0);
-		break;
-
-	default:
-		local_irq_enable();
-		return;
-	}
-	cx->usage++;
-	if ((cx->type != ACPI_STATE_C1) && (sleep_ticks > 0))
-		cx->time += sleep_ticks;
-
-	next_state = pr->power.state;
-
-#ifdef CONFIG_HOTPLUG_CPU
-	/* Don't do promotion/demotion */
-	if ((cx->type == ACPI_STATE_C1) && (num_online_cpus() > 1) &&
-	    !pr->flags.has_cst && !(acpi_gbl_FADT.flags & ACPI_FADT_C2_MP_SUPPORTED)) {
-		next_state = cx;
-		goto end;
-	}
-#endif
-
-	/*
-	 * Promotion?
-	 * ----------
-	 * Track the number of longs (time asleep is greater than threshold)
-	 * and promote when the count threshold is reached.  Note that bus
-	 * mastering activity may prevent promotions.
-	 * Do not promote above max_cstate.
-	 */
-	if (cx->promotion.state &&
-	    ((cx->promotion.state - pr->power.states) <= max_cstate)) {
-		if (sleep_ticks > cx->promotion.threshold.ticks &&
-		  cx->promotion.state->latency <=
-				pm_qos_requirement(PM_QOS_CPU_DMA_LATENCY)) {
-			cx->promotion.count++;
-			cx->demotion.count = 0;
-			if (cx->promotion.count >=
-			    cx->promotion.threshold.count) {
-				if (pr->flags.bm_check) {
-					if (!
-					    (pr->power.bm_activity & cx->
-					     promotion.threshold.bm)) {
-						next_state =
-						    cx->promotion.state;
-						goto end;
-					}
-				} else {
-					next_state = cx->promotion.state;
-					goto end;
-				}
-			}
-		}
-	}
-
-	/*
-	 * Demotion?
-	 * ---------
-	 * Track the number of shorts (time asleep is less than time threshold)
-	 * and demote when the usage threshold is reached.
-	 */
-	if (cx->demotion.state) {
-		if (sleep_ticks < cx->demotion.threshold.ticks) {
-			cx->demotion.count++;
-			cx->promotion.count = 0;
-			if (cx->demotion.count >= cx->demotion.threshold.count) {
-				next_state = cx->demotion.state;
-				goto end;
-			}
-		}
-	}
-
-      end:
-	/*
-	 * Demote if current state exceeds max_cstate
-	 * or if the latency of the current state is unacceptable
-	 */
-	if ((pr->power.state - pr->power.states) > max_cstate ||
-		pr->power.state->latency >
-				pm_qos_requirement(PM_QOS_CPU_DMA_LATENCY)) {
-		if (cx->demotion.state)
-			next_state = cx->demotion.state;
-	}
-
-	/*
-	 * New Cx State?
-	 * -------------
-	 * If we're going to start using a new Cx state we must clean up
-	 * from the previous and prepare to use the new.
-	 */
-	if (next_state != pr->power.state)
-		acpi_processor_power_activate(pr, next_state);
-}
-
-static int acpi_processor_set_power_policy(struct acpi_processor *pr)
-{
-	unsigned int i;
-	unsigned int state_is_set = 0;
-	struct acpi_processor_cx *lower = NULL;
-	struct acpi_processor_cx *higher = NULL;
-	struct acpi_processor_cx *cx;
-
-
-	if (!pr)
-		return -EINVAL;
-
-	/*
-	 * This function sets the default Cx state policy (OS idle handler).
-	 * Our scheme is to promote quickly to C2 but more conservatively
-	 * to C3.  We're favoring C2  for its characteristics of low latency
-	 * (quick response), good power savings, and ability to allow bus
-	 * mastering activity.  Note that the Cx state policy is completely
-	 * customizable and can be altered dynamically.
-	 */
-
-	/* startup state */
-	for (i = 1; i < ACPI_PROCESSOR_MAX_POWER; i++) {
-		cx = &pr->power.states[i];
-		if (!cx->valid)
-			continue;
-
-		if (!state_is_set)
-			pr->power.state = cx;
-		state_is_set++;
-		break;
-	}
-
-	if (!state_is_set)
-		return -ENODEV;
-
-	/* demotion */
-	for (i = 1; i < ACPI_PROCESSOR_MAX_POWER; i++) {
-		cx = &pr->power.states[i];
-		if (!cx->valid)
-			continue;
-
-		if (lower) {
-			cx->demotion.state = lower;
-			cx->demotion.threshold.ticks = cx->latency_ticks;
-			cx->demotion.threshold.count = 1;
-			if (cx->type == ACPI_STATE_C3)
-				cx->demotion.threshold.bm = bm_history;
-		}
-
-		lower = cx;
-	}
-
-	/* promotion */
-	for (i = (ACPI_PROCESSOR_MAX_POWER - 1); i > 0; i--) {
-		cx = &pr->power.states[i];
-		if (!cx->valid)
-			continue;
-
-		if (higher) {
-			cx->promotion.state = higher;
-			cx->promotion.threshold.ticks = cx->latency_ticks;
-			if (cx->type >= ACPI_STATE_C2)
-				cx->promotion.threshold.count = 4;
-			else
-				cx->promotion.threshold.count = 10;
-			if (higher->type == ACPI_STATE_C3)
-				cx->promotion.threshold.bm = bm_history;
-		}
-
-		higher = cx;
-	}
-
-	return 0;
-}
-#endif /* !CONFIG_CPU_IDLE */
-
 static int acpi_processor_get_power_info_fadt(struct acpi_processor *pr)
 {
 
@@ -1047,11 +541,7 @@ static void acpi_processor_power_verify_c2(struct acpi_processor_cx *cx)
 	 */
 	cx->valid = 1;
 
-#ifndef CONFIG_CPU_IDLE
-	cx->latency_ticks = US_TO_PM_TIMER_TICKS(cx->latency);
-#else
 	cx->latency_ticks = cx->latency;
-#endif
 
 	return;
 }
@@ -1121,7 +611,6 @@ static void acpi_processor_power_verify_c3(struct acpi_processor *pr,
 					  " for C3 to be enabled on SMP systems\n"));
 			return;
 		}
-		acpi_set_register(ACPI_BITREG_BUS_MASTER_RLD, 0);
 	}
 
 	/*
@@ -1132,11 +621,16 @@ static void acpi_processor_power_verify_c3(struct acpi_processor *pr,
 	 */
 	cx->valid = 1;
 
-#ifndef CONFIG_CPU_IDLE
-	cx->latency_ticks = US_TO_PM_TIMER_TICKS(cx->latency);
-#else
 	cx->latency_ticks = cx->latency;
-#endif
+	/*
+	 * On older chipsets, BM_RLD needs to be set
+	 * in order for Bus Master activity to wake the
+	 * system from C3.  Newer chipsets handle DMA
+	 * during C3 automatically and BM_RLD is a NOP.
+	 * In either case, the proper way to
+	 * handle BM_RLD is to set it and leave it set.
+	 */
+	acpi_set_register(ACPI_BITREG_BUS_MASTER_RLD, 1);
 
 	return;
 }
@@ -1201,20 +695,6 @@ static int acpi_processor_get_power_info(struct acpi_processor *pr)
 
 	pr->power.count = acpi_processor_power_verify(pr);
 
-#ifndef CONFIG_CPU_IDLE
-	/*
-	 * Set Default Policy
-	 * ------------------
-	 * Now that we know which states are supported, set the default
-	 * policy.  Note that this policy can be changed dynamically
-	 * (e.g. encourage deeper sleeps to conserve battery life when
-	 * not on AC).
-	 */
-	result = acpi_processor_set_power_policy(pr);
-	if (result)
-		return result;
-#endif
-
 	/*
 	 * if one state of type C2 or C3 is available, mark this
 	 * CPU as being "idle manageable"
@@ -1312,69 +792,6 @@ static const struct file_operations acpi_processor_power_fops = {
 	.release = single_release,
 };
 
-#ifndef CONFIG_CPU_IDLE
-
-int acpi_processor_cst_has_changed(struct acpi_processor *pr)
-{
-	int result = 0;
-
-	if (boot_option_idle_override)
-		return 0;
-
-	if (!pr)
-		return -EINVAL;
-
-	if (nocst) {
-		return -ENODEV;
-	}
-
-	if (!pr->flags.power_setup_done)
-		return -ENODEV;
-
-	/*
-	 * Fall back to the default idle loop, when pm_idle_save had
-	 * been initialized.
-	 */
-	if (pm_idle_save) {
-		pm_idle = pm_idle_save;
-		/* Relies on interrupts forcing exit from idle. */
-		synchronize_sched();
-	}
-
-	pr->flags.power = 0;
-	result = acpi_processor_get_power_info(pr);
-	if ((pr->flags.power == 1) && (pr->flags.power_setup_done))
-		pm_idle = acpi_processor_idle;
-
-	return result;
-}
-
-#ifdef CONFIG_SMP
-static void smp_callback(void *v)
-{
-	/* we already woke the CPU up, nothing more to do */
-}
-
-/*
- * This function gets called when a part of the kernel has a new latency
- * requirement.  This means we need to get all processors out of their C-state,
- * and then recalculate a new suitable C-state. Just do a cross-cpu IPI; that
- * wakes them all right up.
- */
-static int acpi_processor_latency_notify(struct notifier_block *b,
-		unsigned long l, void *v)
-{
-	smp_call_function(smp_callback, NULL, 1);
-	return NOTIFY_OK;
-}
-
-static struct notifier_block acpi_processor_latency_notifier = {
-	.notifier_call = acpi_processor_latency_notify,
-};
-
-#endif
-
-#else /* CONFIG_CPU_IDLE */
 
 /**
  * acpi_idle_bm_check - checks if bus master activity was detected
@@ -1383,7 +800,7 @@ static int acpi_idle_bm_check(void)
 {
 	u32 bm_status = 0;
 
-	acpi_get_register(ACPI_BITREG_BUS_MASTER_STATUS, &bm_status);
+	acpi_get_register_unlocked(ACPI_BITREG_BUS_MASTER_STATUS, &bm_status);
 	if (bm_status)
 		acpi_set_register(ACPI_BITREG_BUS_MASTER_STATUS, 1);
 	/*
@@ -1399,25 +816,6 @@ static int acpi_idle_bm_check(void)
 	return bm_status;
 }
 
-/**
- * acpi_idle_update_bm_rld - updates the BM_RLD bit depending on target state
- * @pr: the processor
- * @target: the new target state
- */
-static inline void acpi_idle_update_bm_rld(struct acpi_processor *pr,
-					   struct acpi_processor_cx *target)
-{
-	if (pr->flags.bm_rld_set && target->type != ACPI_STATE_C3) {
-		acpi_set_register(ACPI_BITREG_BUS_MASTER_RLD, 0);
-		pr->flags.bm_rld_set = 0;
-	}
-
-	if (!pr->flags.bm_rld_set && target->type == ACPI_STATE_C3) {
-		acpi_set_register(ACPI_BITREG_BUS_MASTER_RLD, 1);
-		pr->flags.bm_rld_set = 1;
-	}
-}
-
 /**
  * acpi_idle_do_entry - a helper function that does C2 and C3 type entry
  * @cx: cstate data
@@ -1473,9 +871,6 @@ static int acpi_idle_enter_c1(struct cpuidle_device *dev,
 		return 0;
 	}
 
-	if (pr->flags.bm_check)
-		acpi_idle_update_bm_rld(pr, cx);
-
 	t1 = inl(acpi_gbl_FADT.xpm_timer_block.address);
 	acpi_idle_do_entry(cx);
 	t2 = inl(acpi_gbl_FADT.xpm_timer_block.address);
@@ -1527,9 +922,6 @@ static int acpi_idle_enter_simple(struct cpuidle_device *dev,
 	 */
 	acpi_state_timer_broadcast(pr, cx, 1);
 
-	if (pr->flags.bm_check)
-		acpi_idle_update_bm_rld(pr, cx);
-
 	if (cx->type == ACPI_STATE_C3)
 		ACPI_FLUSH_CPU_CACHE();
 
@@ -1621,8 +1013,6 @@ static int acpi_idle_enter_bm(struct cpuidle_device *dev,
 	 */
 	acpi_state_timer_broadcast(pr, cx, 1);
 
-	acpi_idle_update_bm_rld(pr, cx);
-
 	/*
 	 * disable bus master
 	 * bm_check implies we need ARB_DIS
@@ -1795,8 +1185,6 @@ int acpi_processor_cst_has_changed(struct acpi_processor *pr)
 	return ret;
 }
 
-#endif /* CONFIG_CPU_IDLE */
-
 int __cpuinit acpi_processor_power_init(struct acpi_processor *pr,
 			      struct acpi_device *device)
 {
@@ -1825,10 +1213,6 @@ int __cpuinit acpi_processor_power_init(struct acpi_processor *pr,
 			       "ACPI: processor limited to max C-state %d\n",
 			       max_cstate);
 		first_run++;
-#if !defined(CONFIG_CPU_IDLE) && defined(CONFIG_SMP)
-		pm_qos_add_notifier(PM_QOS_CPU_DMA_LATENCY,
-				&acpi_processor_latency_notifier);
-#endif
 	}
 
 	if (!pr)
@@ -1852,11 +1236,9 @@ int __cpuinit acpi_processor_power_init(struct acpi_processor *pr,
 	 * platforms that only support C1.
 	 */
 	if (pr->flags.power) {
-#ifdef CONFIG_CPU_IDLE
 		acpi_processor_setup_cpuidle(pr);
 		if (cpuidle_register_device(&pr->power.dev))
 			return -EIO;
-#endif
 
 		printk(KERN_INFO PREFIX "CPU%d (power states:", pr->id);
 		for (i = 1; i <= pr->power.count; i++)
@@ -1864,13 +1246,6 @@ int __cpuinit acpi_processor_power_init(struct acpi_processor *pr,
 				printk(" C%d[C%d]", i,
 				       pr->power.states[i].type);
 		printk(")\n");
-
-#ifndef CONFIG_CPU_IDLE
-		if (pr->id == 0) {
-			pm_idle_save = pm_idle;
-			pm_idle = acpi_processor_idle;
-		}
-#endif
 	}
 
 	/* 'power' [R] */
@@ -1889,34 +1264,12 @@ int acpi_processor_power_exit(struct acpi_processor *pr,
 	if (boot_option_idle_override)
 		return 0;
 
-#ifdef CONFIG_CPU_IDLE
 	cpuidle_unregister_device(&pr->power.dev);
-#endif
 	pr->flags.power_setup_done = 0;
 
 	if (acpi_device_dir(device))
 		remove_proc_entry(ACPI_PROCESSOR_FILE_POWER,
 				  acpi_device_dir(device));
 
-#ifndef CONFIG_CPU_IDLE
-
-	/* Unregister the idle handler when processor #0 is removed. */
-	if (pr->id == 0) {
-		if (pm_idle_save)
-			pm_idle = pm_idle_save;
-
-		/*
-		 * We are about to unload the current idle thread pm callback
-		 * (pm_idle), Wait for all processors to update cached/local
-		 * copies of pm_idle before proceeding.
-		 */
-		cpu_idle_wait();
-#ifdef CONFIG_SMP
-		pm_qos_remove_notifier(PM_QOS_CPU_DMA_LATENCY,
-				&acpi_processor_latency_notifier);
-#endif
-	}
-#endif
-
 	return 0;
 }

drivers/acpi/processor_perflib.c

@@ -31,14 +31,6 @@
 #include <linux/init.h>
 #include <linux/cpufreq.h>
 
-#ifdef CONFIG_X86_ACPI_CPUFREQ_PROC_INTF
-#include <linux/proc_fs.h>
-#include <linux/seq_file.h>
-#include <linux/mutex.h>
-
-#include <asm/uaccess.h>
-#endif
-
 #ifdef CONFIG_X86
 #include <asm/cpufeature.h>
 #endif
@@ -434,96 +426,6 @@ int acpi_processor_notify_smm(struct module *calling_module)
 
 EXPORT_SYMBOL(acpi_processor_notify_smm);
 
-#ifdef CONFIG_X86_ACPI_CPUFREQ_PROC_INTF
-/* /proc/acpi/processor/../performance interface (DEPRECATED) */
-
-static int acpi_processor_perf_open_fs(struct inode *inode, struct file *file);
-static struct file_operations acpi_processor_perf_fops = {
-	.owner = THIS_MODULE,
-	.open = acpi_processor_perf_open_fs,
-	.read = seq_read,
-	.llseek = seq_lseek,
-	.release = single_release,
-};
-
-static int acpi_processor_perf_seq_show(struct seq_file *seq, void *offset)
-{
-	struct acpi_processor *pr = seq->private;
-	int i;
-
-
-	if (!pr)
-		goto end;
-
-	if (!pr->performance) {
-		seq_puts(seq, "<not supported>\n");
-		goto end;
-	}
-
-	seq_printf(seq, "state count:             %d\n"
-		   "active state:            P%d\n",
-		   pr->performance->state_count, pr->performance->state);
-
-	seq_puts(seq, "states:\n");
-	for (i = 0; i < pr->performance->state_count; i++)
-		seq_printf(seq,
-			   "   %cP%d:                  %d MHz, %d mW, %d uS\n",
-			   (i == pr->performance->state ? '*' : ' '), i,
-			   (u32) pr->performance->states[i].core_frequency,
-			   (u32) pr->performance->states[i].power,
-			   (u32) pr->performance->states[i].transition_latency);
-
-      end:
-	return 0;
-}
-
-static int acpi_processor_perf_open_fs(struct inode *inode, struct file *file)
-{
-	return single_open(file, acpi_processor_perf_seq_show,
-			   PDE(inode)->data);
-}
-
-static void acpi_cpufreq_add_file(struct acpi_processor *pr)
-{
-	struct acpi_device *device = NULL;
-
-
-	if (acpi_bus_get_device(pr->handle, &device))
-		return;
-
-	/* add file 'performance' [R/W] */
-	proc_create_data(ACPI_PROCESSOR_FILE_PERFORMANCE, S_IFREG | S_IRUGO,
-			 acpi_device_dir(device),
-			 &acpi_processor_perf_fops, acpi_driver_data(device));
-	return;
-}
-
-static void acpi_cpufreq_remove_file(struct acpi_processor *pr)
-{
-	struct acpi_device *device = NULL;
-
-
-	if (acpi_bus_get_device(pr->handle, &device))
-		return;
-
-	/* remove file 'performance' */
-	remove_proc_entry(ACPI_PROCESSOR_FILE_PERFORMANCE,
-			  acpi_device_dir(device));
-
-	return;
-}
-
-#else
-static void acpi_cpufreq_add_file(struct acpi_processor *pr)
-{
-	return;
-}
-static void acpi_cpufreq_remove_file(struct acpi_processor *pr)
-{
-	return;
-}
-#endif				/* CONFIG_X86_ACPI_CPUFREQ_PROC_INTF */
-
 static int acpi_processor_get_psd(struct acpi_processor	*pr)
 {
 	int result = 0;
@@ -747,14 +649,12 @@ err_ret:
 }
 EXPORT_SYMBOL(acpi_processor_preregister_performance);
 
-
 int
 acpi_processor_register_performance(struct acpi_processor_performance
 				    *performance, unsigned int cpu)
 {
 	struct acpi_processor *pr;
 
-
 	if (!(acpi_processor_ppc_status & PPC_REGISTERED))
 		return -EINVAL;
 
@@ -781,8 +681,6 @@ acpi_processor_register_performance(struct acpi_processor_performance
 		return -EIO;
 	}
 
-	acpi_cpufreq_add_file(pr);
-
 	mutex_unlock(&performance_mutex);
 	return 0;
 }
@@ -795,7 +693,6 @@ acpi_processor_unregister_performance(struct acpi_processor_performance
 {
 	struct acpi_processor *pr;
 
-
 	mutex_lock(&performance_mutex);
 
 	pr = per_cpu(processors, cpu);
@@ -808,8 +705,6 @@ acpi_processor_unregister_performance(struct acpi_processor_performance
 		kfree(pr->performance->states);
 	pr->performance = NULL;
 
-	acpi_cpufreq_remove_file(pr);
-
 	mutex_unlock(&performance_mutex);
 
 	return;

drivers/acpi/sleep.c

@@ -90,31 +90,6 @@ void __init acpi_old_suspend_ordering(void)
 	old_suspend_ordering = true;
 }
 
-/*
- * According to the ACPI specification the BIOS should make sure that ACPI is
- * enabled and SCI_EN bit is set on wake-up from S1 - S3 sleep states.  Still,
- * some BIOSes don't do that and therefore we use acpi_enable() to enable ACPI
- * on such systems during resume.  Unfortunately that doesn't help in
- * particularly pathological cases in which SCI_EN has to be set directly on
- * resume, although the specification states very clearly that this flag is
- * owned by the hardware.  The set_sci_en_on_resume variable will be set in such
- * cases.
- */
-static bool set_sci_en_on_resume;
-/*
- * The ACPI specification wants us to save NVS memory regions during hibernation
- * and to restore them during the subsequent resume.  However, it is not certain
- * if this mechanism is going to work on all machines, so we allow the user to
- * disable this mechanism using the 'acpi_sleep=s4_nonvs' kernel command line
- * option.
- */
-static bool s4_no_nvs;
-
-void __init acpi_s4_no_nvs(void)
-{
-	s4_no_nvs = true;
-}
-
 /**
  *	acpi_pm_disable_gpes - Disable the GPEs.
  */
@@ -193,6 +168,18 @@ static void acpi_pm_end(void)
 #endif /* CONFIG_ACPI_SLEEP */
 
 #ifdef CONFIG_SUSPEND
+/*
+ * According to the ACPI specification the BIOS should make sure that ACPI is
+ * enabled and SCI_EN bit is set on wake-up from S1 - S3 sleep states.  Still,
+ * some BIOSes don't do that and therefore we use acpi_enable() to enable ACPI
+ * on such systems during resume.  Unfortunately that doesn't help in
+ * particularly pathological cases in which SCI_EN has to be set directly on
+ * resume, although the specification states very clearly that this flag is
+ * owned by the hardware.  The set_sci_en_on_resume variable will be set in such
+ * cases.
+ */
+static bool set_sci_en_on_resume;
+
 extern void do_suspend_lowlevel(void);
 
 static u32 acpi_suspend_states[] = {
@@ -396,6 +383,20 @@ static struct dmi_system_id __initdata acpisleep_dmi_table[] = {
 #endif /* CONFIG_SUSPEND */
 
 #ifdef CONFIG_HIBERNATION
+/*
+ * The ACPI specification wants us to save NVS memory regions during hibernation
+ * and to restore them during the subsequent resume.  However, it is not certain
+ * if this mechanism is going to work on all machines, so we allow the user to
+ * disable this mechanism using the 'acpi_sleep=s4_nonvs' kernel command line
+ * option.
+ */
+static bool s4_no_nvs;
+
+void __init acpi_s4_no_nvs(void)
+{
+	s4_no_nvs = true;
+}
+
 static unsigned long s4_hardware_signature;
 static struct acpi_table_facs *facs;
 static bool nosigcheck;
@@ -679,7 +680,7 @@ static void acpi_power_off_prepare(void)
 static void acpi_power_off(void)
 {
 	/* acpi_sleep_prepare(ACPI_STATE_S5) should have already been called */
-	printk("%s called\n", __func__);
+	printk(KERN_DEBUG "%s called\n", __func__);
 	local_irq_disable();
 	acpi_enable_wakeup_device(ACPI_STATE_S5);
 	acpi_enter_sleep_state(ACPI_STATE_S5);

drivers/acpi/tables.c

@@ -293,7 +293,12 @@ static void __init check_multiple_madt(void)
 
 int __init acpi_table_init(void)
 {
-	acpi_initialize_tables(initial_tables, ACPI_MAX_TABLES, 0);
+	acpi_status status;
+
+	status = acpi_initialize_tables(initial_tables, ACPI_MAX_TABLES, 0);
+	if (ACPI_FAILURE(status))
+		return 1;
+
 	check_multiple_madt();
 	return 0;
 }

drivers/acpi/video.c

@@ -1020,7 +1020,7 @@ acpi_video_device_brightness_seq_show(struct seq_file *seq, void *offset)
 	}
 
 	seq_printf(seq, "levels: ");
-	for (i = 0; i < dev->brightness->count; i++)
+	for (i = 2; i < dev->brightness->count; i++)
 		seq_printf(seq, " %d", dev->brightness->levels[i]);
 	seq_printf(seq, "\ncurrent: %d\n", dev->brightness->curr);
 
@@ -1059,7 +1059,7 @@ acpi_video_device_write_brightness(struct file *file,
 		return -EFAULT;
 
 	/* validate through the list of available levels */
-	for (i = 0; i < dev->brightness->count; i++)
+	for (i = 2; i < dev->brightness->count; i++)
 		if (level == dev->brightness->levels[i]) {
 			if (ACPI_SUCCESS
 			    (acpi_video_device_lcd_set_level(dev, level)))
@@ -1260,7 +1260,7 @@ static int acpi_video_bus_POST_info_seq_show(struct seq_file *seq, void *offset)
 			printk(KERN_WARNING PREFIX
 			       "This indicates a BIOS bug. Please contact the manufacturer.\n");
 		}
-		printk("%llx\n", options);
+		printk(KERN_WARNING "%llx\n", options);
 		seq_printf(seq, "can POST: <integrated video>");
 		if (options & 2)
 			seq_printf(seq, " <PCI video>");
@@ -1712,7 +1712,7 @@ acpi_video_get_next_level(struct acpi_video_device *device,
 	max = max_below = 0;
 	min = min_above = 255;
 	/* Find closest level to level_current */
-	for (i = 0; i < device->brightness->count; i++) {
+	for (i = 2; i < device->brightness->count; i++) {
 		l = device->brightness->levels[i];
 		if (abs(l - level_current) < abs(delta)) {
 			delta = l - level_current;
@@ -1722,7 +1722,7 @@ acpi_video_get_next_level(struct acpi_video_device *device,
 	}
 	/* Ajust level_current to closest available level */
 	level_current += delta;
-	for (i = 0; i < device->brightness->count; i++) {
+	for (i = 2; i < device->brightness->count; i++) {
 		l = device->brightness->levels[i];
 		if (l < min)
 			min = l;
@@ -2006,6 +2006,12 @@ static int acpi_video_bus_add(struct acpi_device *device)
 			device->pnp.bus_id[3] = '0' + instance;
 		instance ++;
 	}
+	/* a hack to fix the duplicate name "VGA" problem on Pa 3553 */
+	if (!strcmp(device->pnp.bus_id, "VGA")) {
+		if (instance)
+			device->pnp.bus_id[3] = '0' + instance;
+		instance++;
+	}
 
 	video->device = device;
 	strcpy(acpi_device_name(device), ACPI_VIDEO_BUS_NAME);

drivers/char/sx.c

@@ -1713,8 +1713,8 @@ static long sx_fw_ioctl(struct file *filp, unsigned int cmd,
 		for (i = 0; i < SX_NBOARDS; i++)
 			sx_dprintk(SX_DEBUG_FIRMWARE, "<%x> ", boards[i].flags);
 		sx_dprintk(SX_DEBUG_FIRMWARE, "\n");
-		unlock_kernel();
-		return -EIO;
+		rc = -EIO;
+		goto out;
 	}
 
 	switch (cmd) {
@@ -1747,7 +1747,8 @@ static long sx_fw_ioctl(struct file *filp, unsigned int cmd,
 		break;
 	case SXIO_DO_RAMTEST:
 		if (sx_initialized)	/* Already initialized: better not ramtest the board.  */
-			return -EPERM;
+			rc = -EPERM;
+			break;
 		if (IS_SX_BOARD(board)) {
 			rc = do_memtest(board, 0, 0x7000);
 			if (!rc)
@@ -1844,6 +1845,7 @@ static long sx_fw_ioctl(struct file *filp, unsigned int cmd,
 		rc = -ENOTTY;
 		break;
 	}
+out:
 	unlock_kernel();
 	func_exit();
 	return rc;

drivers/firewire/fw-card.c

@@ -412,6 +412,7 @@ fw_card_add(struct fw_card *card,
 {
 	u32 *config_rom;
 	size_t length;
+	int err;
 
 	card->max_receive = max_receive;
 	card->link_speed = link_speed;
@@ -422,7 +423,13 @@ fw_card_add(struct fw_card *card,
 	list_add_tail(&card->link, &card_list);
 	mutex_unlock(&card_mutex);
 
-	return card->driver->enable(card, config_rom, length);
+	err = card->driver->enable(card, config_rom, length);
+	if (err < 0) {
+		mutex_lock(&card_mutex);
+		list_del(&card->link);
+		mutex_unlock(&card_mutex);
+	}
+	return err;
 }
 EXPORT_SYMBOL(fw_card_add);
 

drivers/hwmon/hp_accel.c

@@ -153,7 +153,10 @@ static struct axis_conversion lis3lv02d_axis_y_inverted = {1, -2, 3};
 static struct axis_conversion lis3lv02d_axis_x_inverted = {-1, 2, 3};
 static struct axis_conversion lis3lv02d_axis_z_inverted = {1, 2, -3};
 static struct axis_conversion lis3lv02d_axis_xy_rotated_left = {-2, 1, 3};
+static struct axis_conversion lis3lv02d_axis_xy_rotated_left_usd = {-2, 1, -3};
 static struct axis_conversion lis3lv02d_axis_xy_swap_inverted = {-2, -1, 3};
+static struct axis_conversion lis3lv02d_axis_xy_rotated_right = {2, -1, 3};
+static struct axis_conversion lis3lv02d_axis_xy_swap_yz_inverted = {2, -1, -3};
 
 #define AXIS_DMI_MATCH(_ident, _name, _axis) {		\
 	.ident = _ident,				\
@@ -172,10 +175,12 @@ static struct dmi_system_id lis3lv02d_dmi_ids[] = {
 	AXIS_DMI_MATCH("NC2510", "HP Compaq 2510", y_inverted),
 	AXIS_DMI_MATCH("NC8510", "HP Compaq 8510", xy_swap_inverted),
 	AXIS_DMI_MATCH("HP2133", "HP 2133", xy_rotated_left),
+	AXIS_DMI_MATCH("NC653x", "HP Compaq 653", xy_rotated_left_usd),
+	AXIS_DMI_MATCH("NC673x", "HP Compaq 673", xy_rotated_left_usd),
+	AXIS_DMI_MATCH("NC651xx", "HP Compaq 651", xy_rotated_right),
+	AXIS_DMI_MATCH("NC671xx", "HP Compaq 671", xy_swap_yz_inverted),
 	{ NULL, }
 /* Laptop models without axis info (yet):
- * "NC651xx" "HP Compaq 651"
- * "NC671xx" "HP Compaq 671"
  * "NC6910" "HP Compaq 6910"
  * HP Compaq 8710x Notebook PC / Mobile Workstation
  * "NC2400" "HP Compaq nc2400"

drivers/ieee1394/dv1394.c

@@ -1823,6 +1823,10 @@ static int dv1394_open(struct inode *inode, struct file *file)
 
 #endif
 
+	printk(KERN_INFO "%s: NOTE, the dv1394 interface is unsupported "
+	       "and will not be available in the new firewire driver stack. "
+	       "Try libraw1394 based programs instead.\n", current->comm);
+
 	return 0;
 }
 
@@ -2567,10 +2571,6 @@ static int __init dv1394_init_module(void)
 {
 	int ret;
 
-	printk(KERN_WARNING
-	       "NOTE: The dv1394 driver is unsupported and may be removed in a "
-	       "future Linux release. Use raw1394 instead.\n");
-
 	cdev_init(&dv1394_cdev, &dv1394_fops);
 	dv1394_cdev.owner = THIS_MODULE;
 	ret = cdev_add(&dv1394_cdev, IEEE1394_DV1394_DEV, 16);

drivers/md/linear.c

@@ -25,13 +25,13 @@ static inline dev_info_t *which_dev(mddev_t *mddev, sector_t sector)
 {
 	dev_info_t *hash;
 	linear_conf_t *conf = mddev_to_conf(mddev);
+	sector_t idx = sector >> conf->sector_shift;
 
 	/*
 	 * sector_div(a,b) returns the remainer and sets a to a/b
 	 */
-	sector >>= conf->sector_shift;
-	(void)sector_div(sector, conf->spacing);
-	hash = conf->hash_table[sector];
+	(void)sector_div(idx, conf->spacing);
+	hash = conf->hash_table[idx];
 
 	while (sector >= hash->num_sectors + hash->start_sector)
 		hash++;

drivers/md/md.c

@@ -1481,6 +1481,11 @@ static int bind_rdev_to_array(mdk_rdev_t * rdev, mddev_t * mddev)
 		if (find_rdev_nr(mddev, rdev->desc_nr))
 			return -EBUSY;
 	}
+	if (mddev->max_disks && rdev->desc_nr >= mddev->max_disks) {
+		printk(KERN_WARNING "md: %s: array is limited to %d devices\n",
+		       mdname(mddev), mddev->max_disks);
+		return -EBUSY;
+	}
 	bdevname(rdev->bdev,b);
 	while ( (s=strchr(b, '/')) != NULL)
 		*s = '!';
@@ -2441,6 +2446,15 @@ static void analyze_sbs(mddev_t * mddev)
 
 	i = 0;
 	rdev_for_each(rdev, tmp, mddev) {
+		if (rdev->desc_nr >= mddev->max_disks ||
+		    i > mddev->max_disks) {
+			printk(KERN_WARNING
+			       "md: %s: %s: only %d devices permitted\n",
+			       mdname(mddev), bdevname(rdev->bdev, b),
+			       mddev->max_disks);
+			kick_rdev_from_array(rdev);
+			continue;
+		}
 		if (rdev != freshest)
 			if (super_types[mddev->major_version].
 			    validate_super(mddev, rdev)) {
@@ -4614,13 +4628,6 @@ static int hot_add_disk(mddev_t * mddev, dev_t dev)
 	 * noticed in interrupt contexts ...
 	 */
 
-	if (rdev->desc_nr == mddev->max_disks) {
-		printk(KERN_WARNING "%s: can not hot-add to full array!\n",
-			mdname(mddev));
-		err = -EBUSY;
-		goto abort_unbind_export;
-	}
-
 	rdev->raid_disk = -1;
 
 	md_update_sb(mddev, 1);
@@ -4634,9 +4641,6 @@ static int hot_add_disk(mddev_t * mddev, dev_t dev)
 	md_new_event(mddev);
 	return 0;
 
-abort_unbind_export:
-	unbind_rdev_from_array(rdev);
-
 abort_export:
 	export_rdev(rdev);
 	return err;

drivers/md/raid1.c

@@ -1640,7 +1640,8 @@ static void raid1d(mddev_t *mddev)
 			}
 
 			bio = r1_bio->bios[r1_bio->read_disk];
-			if ((disk=read_balance(conf, r1_bio)) == -1) {
+			if ((disk=read_balance(conf, r1_bio)) == -1 ||
+			    disk == r1_bio->read_disk) {
 				printk(KERN_ALERT "raid1: %s: unrecoverable I/O"
 				       " read error for block %llu\n",
 				       bdevname(bio->bi_bdev,b),

drivers/misc/Kconfig

@@ -217,6 +217,7 @@ config DELL_LAPTOP
 	depends on EXPERIMENTAL
 	depends on BACKLIGHT_CLASS_DEVICE
 	depends on RFKILL
+	depends on POWER_SUPPLY
 	default n
 	---help---
 	This driver adds support for rfkill and backlight control to Dell

drivers/misc/atmel-ssc.c

@@ -35,7 +35,7 @@ struct ssc_device *ssc_request(unsigned int ssc_num)
 
 	if (!ssc_valid) {
 		spin_unlock(&user_lock);
-		dev_dbg(&ssc->pdev->dev, "could not find requested device\n");
+		pr_err("ssc: ssc%d platform device is missing\n", ssc_num);
 		return ERR_PTR(-ENODEV);
 	}
 

drivers/misc/hpilo.c

@@ -207,7 +207,7 @@ static void ilo_ccb_close(struct pci_dev *pdev, struct ccb_data *data)
 		  &device_ccb->recv_ctrl);
 
 	/* give iLO some time to process stop request */
-	for (retries = 1000; retries > 0; retries--) {
+	for (retries = MAX_WAIT; retries > 0; retries--) {
 		doorbell_set(driver_ccb);
 		udelay(1);
 		if (!(ioread32(&device_ccb->send_ctrl) & (1 << CTRL_BITPOS_A))
@@ -309,7 +309,7 @@ static int ilo_ccb_open(struct ilo_hwinfo *hw, struct ccb_data *data, int slot)
 	doorbell_clr(driver_ccb);
 
 	/* make sure iLO is really handling requests */
-	for (i = 1000; i > 0; i--) {
+	for (i = MAX_WAIT; i > 0; i--) {
 		if (ilo_pkt_dequeue(hw, driver_ccb, SENDQ, &pkt_id, NULL, NULL))
 			break;
 		udelay(1);
@@ -326,7 +326,7 @@ static int ilo_ccb_open(struct ilo_hwinfo *hw, struct ccb_data *data, int slot)
 
 	return 0;
 free:
-	pci_free_consistent(pdev, data->dma_size, data->dma_va, data->dma_pa);
+	ilo_ccb_close(pdev, data);
 out:
 	return error;
 }

drivers/misc/hpilo.h

@@ -19,6 +19,8 @@
 #define MAX_ILO_DEV	1
 /* max number of files */
 #define MAX_OPEN	(MAX_CCB * MAX_ILO_DEV)
+/* spin counter for open/close delay */
+#define MAX_WAIT	10000
 
 /*
  * Per device, used to track global memory allocations.

drivers/misc/sgi-xp/xpc.h

@@ -3,7 +3,7 @@
  * License.  See the file "COPYING" in the main directory of this archive
  * for more details.
  *
- * Copyright (c) 2004-2008 Silicon Graphics, Inc.  All Rights Reserved.
+ * Copyright (c) 2004-2009 Silicon Graphics, Inc.  All Rights Reserved.
  */
 
 /*
@@ -514,7 +514,8 @@ struct xpc_channel_uv {
 						/* partition's notify mq */
 
 	struct xpc_send_msg_slot_uv *send_msg_slots;
-	struct xpc_notify_mq_msg_uv *recv_msg_slots;
+	void *recv_msg_slots;	/* each slot will hold a xpc_notify_mq_msg_uv */
+				/* structure plus the user's payload */
 
 	struct xpc_fifo_head_uv msg_slot_free_list;
 	struct xpc_fifo_head_uv recv_msg_list;	/* deliverable payloads */

drivers/misc/sgi-xp/xpc_uv.c

@@ -3,7 +3,7 @@
  * License.  See the file "COPYING" in the main directory of this archive
  * for more details.
  *
- * Copyright (c) 2008 Silicon Graphics, Inc.  All Rights Reserved.
+ * Copyright (c) 2008-2009 Silicon Graphics, Inc.  All Rights Reserved.
  */
 
 /*
@@ -1010,8 +1010,8 @@ xpc_allocate_recv_msg_slot_uv(struct xpc_channel *ch)
 			continue;
 
 		for (entry = 0; entry < nentries; entry++) {
-			msg_slot = ch_uv->recv_msg_slots + entry *
-			    ch->entry_size;
+			msg_slot = ch_uv->recv_msg_slots +
+			    entry * ch->entry_size;
 
 			msg_slot->hdr.msg_slot_number = entry;
 		}
@@ -1308,9 +1308,8 @@ xpc_handle_notify_mq_msg_uv(struct xpc_partition *part,
 	/* we're dealing with a normal message sent via the notify_mq */
 	ch_uv = &ch->sn.uv;
 
-	msg_slot = (struct xpc_notify_mq_msg_uv *)((u64)ch_uv->recv_msg_slots +
-		    (msg->hdr.msg_slot_number % ch->remote_nentries) *
-		    ch->entry_size);
+	msg_slot = ch_uv->recv_msg_slots +
+	    (msg->hdr.msg_slot_number % ch->remote_nentries) * ch->entry_size;
 
 	BUG_ON(msg->hdr.msg_slot_number != msg_slot->hdr.msg_slot_number);
 	BUG_ON(msg_slot->hdr.size != 0);

drivers/net/cxgb3/sge.c

@@ -2276,8 +2276,7 @@ no_mem:
 		} else if ((len = ntohl(r->len_cq)) != 0) {
 			struct sge_fl *fl;
 
-			if (eth)
-				lro = qs->lro_enabled && is_eth_tcp(rss_hi);
+			lro &= eth && is_eth_tcp(rss_hi);
 
 			fl = (len & F_RSPD_FLQ) ? &qs->fl[1] : &qs->fl[0];
 			if (fl->use_pages) {

drivers/net/gianfar.c

@@ -351,6 +351,9 @@ static int gfar_probe(struct of_device *ofdev,
 	/* Reset MAC layer */
 	gfar_write(&priv->regs->maccfg1, MACCFG1_SOFT_RESET);
 
+	/* We need to delay at least 3 TX clocks */
+	udelay(2);
+
 	tempval = (MACCFG1_TX_FLOW | MACCFG1_RX_FLOW);
 	gfar_write(&priv->regs->maccfg1, tempval);
 

drivers/net/gianfar.h

@@ -312,7 +312,7 @@ extern const char gfar_driver_version[];
 #define ATTRELI_EI(x) (x)
 
 #define BD_LFLAG(flags) ((flags) << 16)
-#define BD_LENGTH_MASK		0x00ff
+#define BD_LENGTH_MASK		0x0000ffff
 
 /* TxBD status field bits */
 #define TXBD_READY		0x8000

drivers/net/wireless/iwlwifi/iwl-agn.c

@@ -4042,6 +4042,7 @@ static int iwl_pci_suspend(struct pci_dev *pdev, pm_message_t state)
 		priv->is_open = 1;
 	}
 
+	pci_save_state(pdev);
 	pci_set_power_state(pdev, PCI_D3hot);
 
 	return 0;
@@ -4052,6 +4053,7 @@ static int iwl_pci_resume(struct pci_dev *pdev)
 	struct iwl_priv *priv = pci_get_drvdata(pdev);
 
 	pci_set_power_state(pdev, PCI_D0);
+	pci_restore_state(pdev);
 
 	if (priv->is_open)
 		iwl_mac_start(priv->hw);

drivers/net/wireless/iwlwifi/iwl-sta.c

@@ -480,6 +480,9 @@ void iwl_clear_stations_table(struct iwl_priv *priv)
 	priv->num_stations = 0;
 	memset(priv->stations, 0, sizeof(priv->stations));
 
+	/* clean ucode key table bit map */
+	priv->ucode_key_table = 0;
+
 	spin_unlock_irqrestore(&priv->sta_lock, flags);
 }
 EXPORT_SYMBOL(iwl_clear_stations_table);

drivers/net/wireless/iwlwifi/iwl3945-base.c

@@ -8143,6 +8143,7 @@ static int iwl3945_pci_suspend(struct pci_dev *pdev, pm_message_t state)
 		priv->is_open = 1;
 	}
 
+	pci_save_state(pdev);
 	pci_set_power_state(pdev, PCI_D3hot);
 
 	return 0;
@@ -8153,6 +8154,7 @@ static int iwl3945_pci_resume(struct pci_dev *pdev)
 	struct iwl3945_priv *priv = pci_get_drvdata(pdev);
 
 	pci_set_power_state(pdev, PCI_D0);
+	pci_restore_state(pdev);
 
 	if (priv->is_open)
 		iwl3945_mac_start(priv->hw);

drivers/pci/pci-driver.c

@@ -355,6 +355,8 @@ static int pci_legacy_suspend(struct device *dev, pm_message_t state)
 	int i = 0;
 
 	if (drv && drv->suspend) {
+		pci_power_t prev = pci_dev->current_state;
+
 		pci_dev->state_saved = false;
 
 		i = drv->suspend(pci_dev, state);
@@ -365,12 +367,16 @@ static int pci_legacy_suspend(struct device *dev, pm_message_t state)
 		if (pci_dev->state_saved)
 			goto Fixup;
 
-		if (WARN_ON_ONCE(pci_dev->current_state != PCI_D0))
+		if (pci_dev->current_state != PCI_D0
+		    && pci_dev->current_state != PCI_UNKNOWN) {
+			WARN_ONCE(pci_dev->current_state != prev,
+				"PCI PM: Device state not saved by %pF\n",
+				drv->suspend);
 			goto Fixup;
+		}
 	}
 
 	pci_save_state(pci_dev);
-	pci_dev->state_saved = true;
 	/*
 	 * This is for compatibility with existing code with legacy PM support.
 	 */
@@ -424,35 +430,20 @@ static void pci_pm_default_resume_noirq(struct pci_dev *pci_dev)
 	pci_fixup_device(pci_fixup_resume_early, pci_dev);
 }
 
-static int pci_pm_default_resume(struct pci_dev *pci_dev)
+static void pci_pm_default_resume(struct pci_dev *pci_dev)
 {
 	pci_fixup_device(pci_fixup_resume, pci_dev);
 
 	if (!pci_is_bridge(pci_dev))
 		pci_enable_wake(pci_dev, PCI_D0, false);
-
-	return pci_pm_reenable_device(pci_dev);
-}
-
-static void pci_pm_default_suspend_generic(struct pci_dev *pci_dev)
-{
-	/* If device is enabled at this point, disable it */
-	pci_disable_enabled_device(pci_dev);
-	/*
-	 * Save state with interrupts enabled, because in principle the bus the
-	 * device is on may be put into a low power state after this code runs.
-	 */
-	pci_save_state(pci_dev);
 }
 
 static void pci_pm_default_suspend(struct pci_dev *pci_dev)
 {
-	pci_pm_default_suspend_generic(pci_dev);
-
+	/* Disable non-bridge devices without PM support */
 	if (!pci_is_bridge(pci_dev))
-		pci_prepare_to_sleep(pci_dev);
-
-	pci_fixup_device(pci_fixup_suspend, pci_dev);
+		pci_disable_enabled_device(pci_dev);
+	pci_save_state(pci_dev);
 }
 
 static bool pci_has_legacy_pm_support(struct pci_dev *pci_dev)
@@ -497,21 +488,49 @@ static void pci_pm_complete(struct device *dev)
 static int pci_pm_suspend(struct device *dev)
 {
 	struct pci_dev *pci_dev = to_pci_dev(dev);
-	struct device_driver *drv = dev->driver;
-	int error = 0;
+	struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL;
 
 	if (pci_has_legacy_pm_support(pci_dev))
 		return pci_legacy_suspend(dev, PMSG_SUSPEND);
 
-	if (drv && drv->pm && drv->pm->suspend) {
-		error = drv->pm->suspend(dev);
-		suspend_report_result(drv->pm->suspend, error);
+	if (!pm) {
+		pci_pm_default_suspend(pci_dev);
+		goto Fixup;
 	}
 
-	if (!error)
-		pci_pm_default_suspend(pci_dev);
+	pci_dev->state_saved = false;
 
-	return error;
+	if (pm->suspend) {
+		pci_power_t prev = pci_dev->current_state;
+		int error;
+
+		error = pm->suspend(dev);
+		suspend_report_result(pm->suspend, error);
+		if (error)
+			return error;
+
+		if (pci_dev->state_saved)
+			goto Fixup;
+
+		if (pci_dev->current_state != PCI_D0
+		    && pci_dev->current_state != PCI_UNKNOWN) {
+			WARN_ONCE(pci_dev->current_state != prev,
+				"PCI PM: State of device not saved by %pF\n",
+				pm->suspend);
+			goto Fixup;
+		}
+	}
+
+	if (!pci_dev->state_saved) {
+		pci_save_state(pci_dev);
+		if (!pci_is_bridge(pci_dev))
+			pci_prepare_to_sleep(pci_dev);
+	}
+
+ Fixup:
+	pci_fixup_device(pci_fixup_suspend, pci_dev);
+
+	return 0;
 }
 
 static int pci_pm_suspend_noirq(struct device *dev)
@@ -554,7 +573,7 @@ static int pci_pm_resume_noirq(struct device *dev)
 static int pci_pm_resume(struct device *dev)
 {
 	struct pci_dev *pci_dev = to_pci_dev(dev);
-	struct device_driver *drv = dev->driver;
+	struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL;
 	int error = 0;
 
 	/*
@@ -567,12 +586,16 @@ static int pci_pm_resume(struct device *dev)
 	if (pci_has_legacy_pm_support(pci_dev))
 		return pci_legacy_resume(dev);
 
-	error = pci_pm_default_resume(pci_dev);
+	pci_pm_default_resume(pci_dev);
 
-	if (!error && drv && drv->pm && drv->pm->resume)
-		error = drv->pm->resume(dev);
+	if (pm) {
+		if (pm->resume)
+			error = pm->resume(dev);
+	} else {
+		pci_pm_reenable_device(pci_dev);
+	}
 
-	return error;
+	return 0;
 }
 
 #else /* !CONFIG_SUSPEND */
@@ -589,21 +612,31 @@ static int pci_pm_resume(struct device *dev)
 static int pci_pm_freeze(struct device *dev)
 {
 	struct pci_dev *pci_dev = to_pci_dev(dev);
-	struct device_driver *drv = dev->driver;
-	int error = 0;
+	struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL;
 
 	if (pci_has_legacy_pm_support(pci_dev))
 		return pci_legacy_suspend(dev, PMSG_FREEZE);
 
-	if (drv && drv->pm && drv->pm->freeze) {
-		error = drv->pm->freeze(dev);
-		suspend_report_result(drv->pm->freeze, error);
+	if (!pm) {
+		pci_pm_default_suspend(pci_dev);
+		return 0;
 	}
 
-	if (!error)
-		pci_pm_default_suspend_generic(pci_dev);
+	pci_dev->state_saved = false;
 
-	return error;
+	if (pm->freeze) {
+		int error;
+
+		error = pm->freeze(dev);
+		suspend_report_result(pm->freeze, error);
+		if (error)
+			return error;
+	}
+
+	if (!pci_dev->state_saved)
+		pci_save_state(pci_dev);
+
+	return 0;
 }
 
 static int pci_pm_freeze_noirq(struct device *dev)
@@ -646,16 +679,18 @@ static int pci_pm_thaw_noirq(struct device *dev)
 static int pci_pm_thaw(struct device *dev)
 {
 	struct pci_dev *pci_dev = to_pci_dev(dev);
-	struct device_driver *drv = dev->driver;
+	struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL;
 	int error = 0;
 
 	if (pci_has_legacy_pm_support(pci_dev))
 		return pci_legacy_resume(dev);
 
-	pci_pm_reenable_device(pci_dev);
-
-	if (drv && drv->pm && drv->pm->thaw)
-		error =  drv->pm->thaw(dev);
+	if (pm) {
+		if (pm->thaw)
+			error = pm->thaw(dev);
+	} else {
+		pci_pm_reenable_device(pci_dev);
+	}
 
 	return error;
 }
@@ -663,22 +698,29 @@ static int pci_pm_thaw(struct device *dev)
 static int pci_pm_poweroff(struct device *dev)
 {
 	struct pci_dev *pci_dev = to_pci_dev(dev);
-	struct device_driver *drv = dev->driver;
+	struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL;
 	int error = 0;
 
 	if (pci_has_legacy_pm_support(pci_dev))
 		return pci_legacy_suspend(dev, PMSG_HIBERNATE);
 
-	if (!drv || !drv->pm)
-		return 0;
+	if (!pm) {
+		pci_pm_default_suspend(pci_dev);
+		goto Fixup;
+	}
+
+	pci_dev->state_saved = false;
 
-	if (drv->pm->poweroff) {
-		error = drv->pm->poweroff(dev);
-		suspend_report_result(drv->pm->poweroff, error);
+	if (pm->poweroff) {
+		error = pm->poweroff(dev);
+		suspend_report_result(pm->poweroff, error);
 	}
 
-	if (!error)
-		pci_pm_default_suspend(pci_dev);
+	if (!pci_dev->state_saved && !pci_is_bridge(pci_dev))
+		pci_prepare_to_sleep(pci_dev);
+
+ Fixup:
+	pci_fixup_device(pci_fixup_suspend, pci_dev);
 
 	return error;
 }
@@ -719,7 +761,7 @@ static int pci_pm_restore_noirq(struct device *dev)
 static int pci_pm_restore(struct device *dev)
 {
 	struct pci_dev *pci_dev = to_pci_dev(dev);
-	struct device_driver *drv = dev->driver;
+	struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL;
 	int error = 0;
 
 	/*
@@ -732,10 +774,14 @@ static int pci_pm_restore(struct device *dev)
 	if (pci_has_legacy_pm_support(pci_dev))
 		return pci_legacy_resume(dev);
 
-	error = pci_pm_default_resume(pci_dev);
+	pci_pm_default_resume(pci_dev);
 
-	if (!error && drv && drv->pm && drv->pm->restore)
-		error = drv->pm->restore(dev);
+	if (pm) {
+		if (pm->restore)
+			error = pm->restore(dev);
+	} else {
+		pci_pm_reenable_device(pci_dev);
+	}
 
 	return error;
 }

drivers/pci/pci-sysfs.c

@@ -768,8 +768,8 @@ pci_read_rom(struct kobject *kobj, struct bin_attribute *bin_attr,
 		return -EINVAL;
 	
 	rom = pci_map_rom(pdev, &size);	/* size starts out as PCI window size */
-	if (!rom)
-		return 0;
+	if (!rom || !size)
+		return -EIO;
 		
 	if (off >= size)
 		count = 0;

drivers/pci/pci.c

@@ -1418,10 +1418,10 @@ int pci_restore_standard_config(struct pci_dev *dev)
 		break;
 	}
 
-	dev->current_state = PCI_D0;
+	pci_update_current_state(dev, PCI_D0);
 
  Restore:
-	return pci_restore_state(dev);
+	return dev->state_saved ? pci_restore_state(dev) : 0;
 }
 
 /**

drivers/pci/pcie/aspm.c

@@ -718,9 +718,9 @@ void pcie_aspm_exit_link_state(struct pci_dev *pdev)
 
 	/*
 	 * All PCIe functions are in one slot, remove one function will remove
-	 * the the whole slot, so just wait
+	 * the whole slot, so just wait until we are the last function left.
 	 */
-	if (!list_empty(&parent->subordinate->devices))
+	if (!list_is_last(&pdev->bus_list, &parent->subordinate->devices))
 		goto out;
 
 	/* All functions are removed, so just disable ASPM for the link */

drivers/pci/pcie/portdrv_pci.c

@@ -55,25 +55,13 @@ static int pcie_portdrv_suspend(struct pci_dev *dev, pm_message_t state)
 
 }
 
-static int pcie_portdrv_suspend_late(struct pci_dev *dev, pm_message_t state)
-{
-	return pci_save_state(dev);
-}
-
-static int pcie_portdrv_resume_early(struct pci_dev *dev)
-{
-	return pci_restore_state(dev);
-}
-
 static int pcie_portdrv_resume(struct pci_dev *dev)
 {
-	pcie_portdrv_restore_config(dev);
+	pci_set_master(dev);
 	return pcie_port_device_resume(dev);
 }
 #else
 #define pcie_portdrv_suspend NULL
-#define pcie_portdrv_suspend_late NULL
-#define pcie_portdrv_resume_early NULL
 #define pcie_portdrv_resume NULL
 #endif
 
@@ -292,8 +280,6 @@ static struct pci_driver pcie_portdriver = {
 	.remove		= pcie_portdrv_remove,
 
 	.suspend	= pcie_portdrv_suspend,
-	.suspend_late	= pcie_portdrv_suspend_late,
-	.resume_early	= pcie_portdrv_resume_early,
 	.resume		= pcie_portdrv_resume,
 
 	.err_handler 	= &pcie_portdrv_err_handler,

drivers/pci/rom.c

@@ -63,7 +63,7 @@ void pci_disable_rom(struct pci_dev *pdev)
  * The PCI window size could be much larger than the
  * actual image size.
  */
-size_t pci_get_rom_size(void __iomem *rom, size_t size)
+size_t pci_get_rom_size(struct pci_dev *pdev, void __iomem *rom, size_t size)
 {
 	void __iomem *image;
 	int last_image;
@@ -72,8 +72,10 @@ size_t pci_get_rom_size(void __iomem *rom, size_t size)
 	do {
 		void __iomem *pds;
 		/* Standard PCI ROMs start out with these bytes 55 AA */
-		if (readb(image) != 0x55)
+		if (readb(image) != 0x55) {
+			dev_err(&pdev->dev, "Invalid ROM contents\n");
 			break;
+		}
 		if (readb(image + 1) != 0xAA)
 			break;
 		/* get the PCI data structure and check its signature */
@@ -159,7 +161,7 @@ void __iomem *pci_map_rom(struct pci_dev *pdev, size_t *size)
 	 * size is much larger than the actual size of the ROM.
 	 * True size is important if the ROM is going to be copied.
 	 */
-	*size = pci_get_rom_size(rom, *size);
+	*size = pci_get_rom_size(pdev, rom, *size);
 	return rom;
 }
 

drivers/platform/x86/Kconfig

@@ -42,6 +42,7 @@ config ASUS_LAPTOP
 	depends on LEDS_CLASS
 	depends on NEW_LEDS
 	depends on BACKLIGHT_CLASS_DEVICE
+	depends on INPUT
 	---help---
 	  This is the new Linux driver for Asus laptops. It may also support some
 	  MEDION, JVC or VICTOR laptops. It makes all the extra buttons generate

drivers/platform/x86/asus-laptop.c

@@ -46,6 +46,7 @@
 #include <acpi/acpi_drivers.h>
 #include <acpi/acpi_bus.h>
 #include <asm/uaccess.h>
+#include <linux/input.h>
 
 #define ASUS_LAPTOP_VERSION "0.42"
 
@@ -181,6 +182,8 @@ struct asus_hotk {
 	u8 light_level;		//light sensor level
 	u8 light_switch;	//light sensor switch value
 	u16 event_count[128];	//count for each event TODO make this better
+	struct input_dev *inputdev;
+	u16 *keycode_map;
 };
 
 /*
@@ -250,6 +253,37 @@ ASUS_LED(rled, "record");
 ASUS_LED(pled, "phone");
 ASUS_LED(gled, "gaming");
 
+struct key_entry {
+	char type;
+	u8 code;
+	u16 keycode;
+};
+
+enum { KE_KEY, KE_END };
+
+static struct key_entry asus_keymap[] = {
+	{KE_KEY, 0x30, KEY_VOLUMEUP},
+	{KE_KEY, 0x31, KEY_VOLUMEDOWN},
+	{KE_KEY, 0x32, KEY_MUTE},
+	{KE_KEY, 0x33, KEY_SWITCHVIDEOMODE},
+	{KE_KEY, 0x34, KEY_SWITCHVIDEOMODE},
+	{KE_KEY, 0x40, KEY_PREVIOUSSONG},
+	{KE_KEY, 0x41, KEY_NEXTSONG},
+	{KE_KEY, 0x43, KEY_STOP},
+	{KE_KEY, 0x45, KEY_PLAYPAUSE},
+	{KE_KEY, 0x50, KEY_EMAIL},
+	{KE_KEY, 0x51, KEY_WWW},
+	{KE_KEY, 0x5C, BTN_EXTRA},  /* Performance */
+	{KE_KEY, 0x5D, KEY_WLAN},
+	{KE_KEY, 0x61, KEY_SWITCHVIDEOMODE},
+	{KE_KEY, 0x6B, BTN_TOUCH}, /* Lock Mouse */
+	{KE_KEY, 0x82, KEY_CAMERA},
+	{KE_KEY, 0x8A, KEY_TV},
+	{KE_KEY, 0x95, KEY_MEDIA},
+	{KE_KEY, 0x99, KEY_PHONE},
+	{KE_END, 0},
+};
+
 /*
  * This function evaluates an ACPI method, given an int as parameter, the
  * method is searched within the scope of the handle, can be NULL. The output
@@ -720,8 +754,68 @@ static ssize_t store_gps(struct device *dev, struct device_attribute *attr,
 	return store_status(buf, count, NULL, GPS_ON);
 }
 
+/*
+ * Hotkey functions
+ */
+static struct key_entry *asus_get_entry_by_scancode(int code)
+{
+	struct key_entry *key;
+
+	for (key = asus_keymap; key->type != KE_END; key++)
+		if (code == key->code)
+			return key;
+
+	return NULL;
+}
+
+static struct key_entry *asus_get_entry_by_keycode(int code)
+{
+	struct key_entry *key;
+
+	for (key = asus_keymap; key->type != KE_END; key++)
+		if (code == key->keycode && key->type == KE_KEY)
+			return key;
+
+	return NULL;
+}
+
+static int asus_getkeycode(struct input_dev *dev, int scancode, int *keycode)
+{
+	struct key_entry *key = asus_get_entry_by_scancode(scancode);
+
+	if (key && key->type == KE_KEY) {
+		*keycode = key->keycode;
+		return 0;
+	}
+
+	return -EINVAL;
+}
+
+static int asus_setkeycode(struct input_dev *dev, int scancode, int keycode)
+{
+	struct key_entry *key;
+	int old_keycode;
+
+	if (keycode < 0 || keycode > KEY_MAX)
+		return -EINVAL;
+
+	key = asus_get_entry_by_scancode(scancode);
+	if (key && key->type == KE_KEY) {
+		old_keycode = key->keycode;
+		key->keycode = keycode;
+		set_bit(keycode, dev->keybit);
+		if (!asus_get_entry_by_keycode(old_keycode))
+			clear_bit(old_keycode, dev->keybit);
+		return 0;
+	}
+
+	return -EINVAL;
+}
+
 static void asus_hotk_notify(acpi_handle handle, u32 event, void *data)
 {
+	static struct key_entry *key;
+
 	/* TODO Find a better way to handle events count. */
 	if (!hotk)
 		return;
@@ -738,10 +832,24 @@ static void asus_hotk_notify(acpi_handle handle, u32 event, void *data)
 		lcd_blank(FB_BLANK_POWERDOWN);
 	}
 
-	acpi_bus_generate_proc_event(hotk->device, event,
-				hotk->event_count[event % 128]++);
-
-	return;
+	acpi_bus_generate_netlink_event(hotk->device->pnp.device_class,
+					dev_name(&hotk->device->dev), event,
+					hotk->event_count[event % 128]++);
+
+	if (hotk->inputdev) {
+		key = asus_get_entry_by_scancode(event);
+		if (!key)
+			return ;
+
+		switch (key->type) {
+		case KE_KEY:
+			input_report_key(hotk->inputdev, key->keycode, 1);
+			input_sync(hotk->inputdev);
+			input_report_key(hotk->inputdev, key->keycode, 0);
+			input_sync(hotk->inputdev);
+			break;
+		}
+	}
 }
 
 #define ASUS_CREATE_DEVICE_ATTR(_name)					\
@@ -959,6 +1067,38 @@ static int asus_hotk_get_info(void)
 	return AE_OK;
 }
 
+static int asus_input_init(void)
+{
+	const struct key_entry *key;
+	int result;
+
+	hotk->inputdev = input_allocate_device();
+	if (!hotk->inputdev) {
+		printk(ASUS_INFO "Unable to allocate input device\n");
+		return 0;
+	}
+	hotk->inputdev->name = "Asus Laptop extra buttons";
+	hotk->inputdev->phys = ASUS_HOTK_FILE "/input0";
+	hotk->inputdev->id.bustype = BUS_HOST;
+	hotk->inputdev->getkeycode = asus_getkeycode;
+	hotk->inputdev->setkeycode = asus_setkeycode;
+
+	for (key = asus_keymap; key->type != KE_END; key++) {
+		switch (key->type) {
+		case KE_KEY:
+			set_bit(EV_KEY, hotk->inputdev->evbit);
+			set_bit(key->keycode, hotk->inputdev->keybit);
+			break;
+		}
+	}
+	result = input_register_device(hotk->inputdev);
+	if (result) {
+		printk(ASUS_INFO "Unable to register input device\n");
+		input_free_device(hotk->inputdev);
+	}
+	return result;
+}
+
 static int asus_hotk_check(void)
 {
 	int result = 0;
@@ -1044,7 +1184,7 @@ static int asus_hotk_add(struct acpi_device *device)
 	/* GPS is on by default */
 	write_status(NULL, 1, GPS_ON);
 
-      end:
+end:
 	if (result) {
 		kfree(hotk->name);
 		kfree(hotk);
@@ -1091,10 +1231,17 @@ static void asus_led_exit(void)
 	ASUS_LED_UNREGISTER(gled);
 }
 
+static void asus_input_exit(void)
+{
+	if (hotk->inputdev)
+		input_unregister_device(hotk->inputdev);
+}
+
 static void __exit asus_laptop_exit(void)
 {
 	asus_backlight_exit();
 	asus_led_exit();
+	asus_input_exit();
 
 	acpi_bus_unregister_driver(&asus_hotk_driver);
 	sysfs_remove_group(&asuspf_device->dev.kobj, &asuspf_attribute_group);
@@ -1216,6 +1363,10 @@ static int __init asus_laptop_init(void)
 		printk(ASUS_INFO "Brightness ignored, must be controlled by "
 		       "ACPI video driver\n");
 
+	result = asus_input_init();
+	if (result)
+		goto fail_input;
+
 	result = asus_led_init(dev);
 	if (result)
 		goto fail_led;
@@ -1242,22 +1393,25 @@ static int __init asus_laptop_init(void)
 
 	return 0;
 
-      fail_sysfs:
+fail_sysfs:
 	platform_device_del(asuspf_device);
 
-      fail_platform_device2:
+fail_platform_device2:
 	platform_device_put(asuspf_device);
 
-      fail_platform_device1:
+fail_platform_device1:
 	platform_driver_unregister(&asuspf_driver);
 
-      fail_platform_driver:
+fail_platform_driver:
 	asus_led_exit();
 
-      fail_led:
+fail_led:
+	asus_input_exit();
+
+fail_input:
 	asus_backlight_exit();
 
-      fail_backlight:
+fail_backlight:
 
 	return result;
 }

drivers/platform/x86/asus_acpi.c

@@ -143,6 +143,7 @@ struct asus_hotk {
 							 S1300N, S5200N*/
 		A4S,            /* Z81sp */
 		F3Sa,		/* (Centrino) */
+		R1F,
 		END_MODEL
 	} model;		/* Models currently supported */
 	u16 event_count[128];	/* Count for each event TODO make this better */
@@ -420,7 +421,18 @@ static struct model_data model_conf[END_MODEL] = {
 		.display_get	= "\\ADVG",
 		.display_set	= "SDSP",
 	},
-
+	{
+		.name = "R1F",
+		.mt_bt_switch = "BLED",
+		.mt_mled = "MLED",
+		.mt_wled = "WLED",
+		.mt_lcd_switch = "\\Q10",
+		.lcd_status = "\\GP06",
+		.brightness_set = "SPLV",
+		.brightness_get = "GPLV",
+		.display_set = "SDSP",
+		.display_get = "\\INFB"
+	}
 };
 
 /* procdir we use */
@@ -1165,6 +1177,8 @@ static int asus_model_match(char *model)
 		return W3V;
 	else if (strncmp(model, "W5A", 3) == 0)
 		return W5A;
+	else if (strncmp(model, "R1F", 3) == 0)
+		return R1F;
 	else if (strncmp(model, "A4S", 3) == 0)
 		return A4S;
 	else if (strncmp(model, "F3Sa", 4) == 0)

drivers/platform/x86/eeepc-laptop.c

@@ -30,6 +30,7 @@
 #include <linux/uaccess.h>
 #include <linux/input.h>
 #include <linux/rfkill.h>
+#include <linux/pci.h>
 
 #define EEEPC_LAPTOP_VERSION	"0.1"
 
@@ -161,6 +162,10 @@ static struct key_entry eeepc_keymap[] = {
 	{KE_KEY, 0x13, KEY_MUTE },
 	{KE_KEY, 0x14, KEY_VOLUMEDOWN },
 	{KE_KEY, 0x15, KEY_VOLUMEUP },
+	{KE_KEY, 0x1a, KEY_COFFEE },
+	{KE_KEY, 0x1b, KEY_ZOOM },
+	{KE_KEY, 0x1c, KEY_PROG2 },
+	{KE_KEY, 0x1d, KEY_PROG3 },
 	{KE_KEY, 0x30, KEY_SWITCHVIDEOMODE },
 	{KE_KEY, 0x31, KEY_SWITCHVIDEOMODE },
 	{KE_KEY, 0x32, KEY_SWITCHVIDEOMODE },
@@ -510,7 +515,43 @@ static int eeepc_hotk_check(void)
 static void notify_brn(void)
 {
 	struct backlight_device *bd = eeepc_backlight_device;
-	bd->props.brightness = read_brightness(bd);
+	if (bd)
+		bd->props.brightness = read_brightness(bd);
+}
+
+static void eeepc_rfkill_notify(acpi_handle handle, u32 event, void *data)
+{
+	struct pci_dev *dev;
+	struct pci_bus *bus = pci_find_bus(0, 1);
+
+	if (event != ACPI_NOTIFY_BUS_CHECK)
+		return;
+
+	if (!bus) {
+		printk(EEEPC_WARNING "Unable to find PCI bus 1?\n");
+		return;
+	}
+
+	if (get_acpi(CM_ASL_WLAN) == 1) {
+		dev = pci_get_slot(bus, 0);
+		if (dev) {
+			/* Device already present */
+			pci_dev_put(dev);
+			return;
+		}
+		dev = pci_scan_single_device(bus, 0);
+		if (dev) {
+			pci_bus_assign_resources(bus);
+			if (pci_bus_add_device(dev))
+				printk(EEEPC_ERR "Unable to hotplug wifi\n");
+		}
+	} else {
+		dev = pci_get_slot(bus, 0);
+		if (dev) {
+			pci_remove_bus_device(dev);
+			pci_dev_put(dev);
+		}
+	}
 }
 
 static void eeepc_hotk_notify(acpi_handle handle, u32 event, void *data)
@@ -520,8 +561,9 @@ static void eeepc_hotk_notify(acpi_handle handle, u32 event, void *data)
 		return;
 	if (event >= NOTIFY_BRN_MIN && event <= NOTIFY_BRN_MAX)
 		notify_brn();
-	acpi_bus_generate_proc_event(ehotk->device, event,
-				     ehotk->event_count[event % 128]++);
+	acpi_bus_generate_netlink_event(ehotk->device->pnp.device_class,
+					dev_name(&ehotk->device->dev), event,
+					ehotk->event_count[event % 128]++);
 	if (ehotk->inputdev) {
 		key = eepc_get_entry_by_scancode(event);
 		if (key) {
@@ -539,6 +581,45 @@ static void eeepc_hotk_notify(acpi_handle handle, u32 event, void *data)
 	}
 }
 
+static int eeepc_register_rfkill_notifier(char *node)
+{
+	acpi_status status = AE_OK;
+	acpi_handle handle;
+
+	status = acpi_get_handle(NULL, node, &handle);
+
+	if (ACPI_SUCCESS(status)) {
+		status = acpi_install_notify_handler(handle,
+						     ACPI_SYSTEM_NOTIFY,
+						     eeepc_rfkill_notify,
+						     NULL);
+		if (ACPI_FAILURE(status))
+			printk(EEEPC_WARNING
+			       "Failed to register notify on %s\n", node);
+	} else
+		return -ENODEV;
+
+	return 0;
+}
+
+static void eeepc_unregister_rfkill_notifier(char *node)
+{
+	acpi_status status = AE_OK;
+	acpi_handle handle;
+
+	status = acpi_get_handle(NULL, node, &handle);
+
+	if (ACPI_SUCCESS(status)) {
+		status = acpi_remove_notify_handler(handle,
+						     ACPI_SYSTEM_NOTIFY,
+						     eeepc_rfkill_notify);
+		if (ACPI_FAILURE(status))
+			printk(EEEPC_ERR
+			       "Error removing rfkill notify handler %s\n",
+				node);
+	}
+}
+
 static int eeepc_hotk_add(struct acpi_device *device)
 {
 	acpi_status status = AE_OK;
@@ -558,7 +639,7 @@ static int eeepc_hotk_add(struct acpi_device *device)
 	ehotk->device = device;
 	result = eeepc_hotk_check();
 	if (result)
-		goto end;
+		goto ehotk_fail;
 	status = acpi_install_notify_handler(ehotk->handle, ACPI_SYSTEM_NOTIFY,
 					     eeepc_hotk_notify, ehotk);
 	if (ACPI_FAILURE(status))
@@ -569,18 +650,25 @@ static int eeepc_hotk_add(struct acpi_device *device)
 							   RFKILL_TYPE_WLAN);
 
 		if (!ehotk->eeepc_wlan_rfkill)
-			goto end;
+			goto wlan_fail;
 
 		ehotk->eeepc_wlan_rfkill->name = "eeepc-wlan";
 		ehotk->eeepc_wlan_rfkill->toggle_radio = eeepc_wlan_rfkill_set;
 		ehotk->eeepc_wlan_rfkill->get_state = eeepc_wlan_rfkill_state;
-		if (get_acpi(CM_ASL_WLAN) == 1)
+		if (get_acpi(CM_ASL_WLAN) == 1) {
 			ehotk->eeepc_wlan_rfkill->state =
 				RFKILL_STATE_UNBLOCKED;
-		else
+			rfkill_set_default(RFKILL_TYPE_WLAN,
+					   RFKILL_STATE_UNBLOCKED);
+		} else {
 			ehotk->eeepc_wlan_rfkill->state =
 				RFKILL_STATE_SOFT_BLOCKED;
-		rfkill_register(ehotk->eeepc_wlan_rfkill);
+			rfkill_set_default(RFKILL_TYPE_WLAN,
+					   RFKILL_STATE_SOFT_BLOCKED);
+		}
+		result = rfkill_register(ehotk->eeepc_wlan_rfkill);
+		if (result)
+			goto wlan_fail;
 	}
 
 	if (get_acpi(CM_ASL_BLUETOOTH) != -1) {
@@ -588,27 +676,47 @@ static int eeepc_hotk_add(struct acpi_device *device)
 			rfkill_allocate(&device->dev, RFKILL_TYPE_BLUETOOTH);
 
 		if (!ehotk->eeepc_bluetooth_rfkill)
-			goto end;
+			goto bluetooth_fail;
 
 		ehotk->eeepc_bluetooth_rfkill->name = "eeepc-bluetooth";
 		ehotk->eeepc_bluetooth_rfkill->toggle_radio =
 			eeepc_bluetooth_rfkill_set;
 		ehotk->eeepc_bluetooth_rfkill->get_state =
 			eeepc_bluetooth_rfkill_state;
-		if (get_acpi(CM_ASL_BLUETOOTH) == 1)
+		if (get_acpi(CM_ASL_BLUETOOTH) == 1) {
 			ehotk->eeepc_bluetooth_rfkill->state =
 				RFKILL_STATE_UNBLOCKED;
-		else
+			rfkill_set_default(RFKILL_TYPE_BLUETOOTH,
+					   RFKILL_STATE_UNBLOCKED);
+		} else {
 			ehotk->eeepc_bluetooth_rfkill->state =
 				RFKILL_STATE_SOFT_BLOCKED;
-		rfkill_register(ehotk->eeepc_bluetooth_rfkill);
-	}
+			rfkill_set_default(RFKILL_TYPE_BLUETOOTH,
+					   RFKILL_STATE_SOFT_BLOCKED);
+		}
 
- end:
-	if (result) {
-		kfree(ehotk);
-		ehotk = NULL;
+		result = rfkill_register(ehotk->eeepc_bluetooth_rfkill);
+		if (result)
+			goto bluetooth_fail;
 	}
+
+	eeepc_register_rfkill_notifier("\\_SB.PCI0.P0P6");
+	eeepc_register_rfkill_notifier("\\_SB.PCI0.P0P7");
+
+	return 0;
+
+ bluetooth_fail:
+	if (ehotk->eeepc_bluetooth_rfkill)
+		rfkill_free(ehotk->eeepc_bluetooth_rfkill);
+	rfkill_unregister(ehotk->eeepc_wlan_rfkill);
+	ehotk->eeepc_wlan_rfkill = NULL;
+ wlan_fail:
+	if (ehotk->eeepc_wlan_rfkill)
+		rfkill_free(ehotk->eeepc_wlan_rfkill);
+ ehotk_fail:
+	kfree(ehotk);
+	ehotk = NULL;
+
 	return result;
 }
 
@@ -622,6 +730,10 @@ static int eeepc_hotk_remove(struct acpi_device *device, int type)
 					    eeepc_hotk_notify);
 	if (ACPI_FAILURE(status))
 		printk(EEEPC_ERR "Error removing notify handler\n");
+
+	eeepc_unregister_rfkill_notifier("\\_SB.PCI0.P0P6");
+	eeepc_unregister_rfkill_notifier("\\_SB.PCI0.P0P7");
+
 	kfree(ehotk);
 	return 0;
 }
@@ -737,13 +849,21 @@ static void eeepc_backlight_exit(void)
 {
 	if (eeepc_backlight_device)
 		backlight_device_unregister(eeepc_backlight_device);
-	if (ehotk->inputdev)
-		input_unregister_device(ehotk->inputdev);
+	eeepc_backlight_device = NULL;
+}
+
+static void eeepc_rfkill_exit(void)
+{
 	if (ehotk->eeepc_wlan_rfkill)
 		rfkill_unregister(ehotk->eeepc_wlan_rfkill);
 	if (ehotk->eeepc_bluetooth_rfkill)
 		rfkill_unregister(ehotk->eeepc_bluetooth_rfkill);
-	eeepc_backlight_device = NULL;
+}
+
+static void eeepc_input_exit(void)
+{
+	if (ehotk->inputdev)
+		input_unregister_device(ehotk->inputdev);
 }
 
 static void eeepc_hwmon_exit(void)
@@ -762,6 +882,8 @@ static void eeepc_hwmon_exit(void)
 static void __exit eeepc_laptop_exit(void)
 {
 	eeepc_backlight_exit();
+	eeepc_rfkill_exit();
+	eeepc_input_exit();
 	eeepc_hwmon_exit();
 	acpi_bus_unregister_driver(&eeepc_hotk_driver);
 	sysfs_remove_group(&platform_device->dev.kobj,
@@ -865,6 +987,8 @@ fail_platform_driver:
 fail_hwmon:
 	eeepc_backlight_exit();
 fail_backlight:
+	eeepc_input_exit();
+	eeepc_rfkill_exit();
 	return result;
 }
 

drivers/platform/x86/hp-wmi.c

@@ -463,9 +463,11 @@ static int __init hp_wmi_bios_setup(struct platform_device *device)
 
 	return 0;
 register_wwan_err:
-	rfkill_unregister(bluetooth_rfkill);
+	if (bluetooth_rfkill)
+		rfkill_unregister(bluetooth_rfkill);
 register_bluetooth_error:
-	rfkill_unregister(wifi_rfkill);
+	if (wifi_rfkill)
+		rfkill_unregister(wifi_rfkill);
 add_sysfs_error:
 	cleanup_sysfs(device);
 	return err;

drivers/platform/x86/panasonic-laptop.c

@@ -507,7 +507,7 @@ static void acpi_pcc_generate_keyinput(struct pcc_acpi *pcc)
 
 	hkey_num = result & 0xf;
 
-	if (hkey_num < 0 || hkey_num > ARRAY_SIZE(pcc->keymap)) {
+	if (hkey_num < 0 || hkey_num >= ARRAY_SIZE(pcc->keymap)) {
 		ACPI_DEBUG_PRINT((ACPI_DB_ERROR,
 				  "hotkey number out of range: %d\n",
 				  hkey_num));

drivers/rtc/Kconfig

@@ -241,6 +241,12 @@ config RTC_DRV_M41T80_WDT
 	  If you say Y here you will get support for the
 	  watchdog timer in the ST M41T60 and M41T80 RTC chips series.
 
+config RTC_DRV_DM355EVM
+	tristate "TI DaVinci DM355 EVM RTC"
+	depends on MFD_DM355EVM_MSP
+	help
+	  Supports the RTC firmware in the MSP430 on the DM355 EVM.
+
 config RTC_DRV_TWL92330
 	boolean "TI TWL92330/Menelaus"
 	depends on MENELAUS

drivers/rtc/Makefile

@@ -23,6 +23,7 @@ obj-$(CONFIG_RTC_DRV_AT91SAM9)	+= rtc-at91sam9.o
 obj-$(CONFIG_RTC_DRV_AU1XXX)	+= rtc-au1xxx.o
 obj-$(CONFIG_RTC_DRV_BFIN)	+= rtc-bfin.o
 obj-$(CONFIG_RTC_DRV_CMOS)	+= rtc-cmos.o
+obj-$(CONFIG_RTC_DRV_DM355EVM)	+= rtc-dm355evm.o
 obj-$(CONFIG_RTC_DRV_DS1216)	+= rtc-ds1216.o
 obj-$(CONFIG_RTC_DRV_DS1286)	+= rtc-ds1286.o
 obj-$(CONFIG_RTC_DRV_DS1302)	+= rtc-ds1302.o

drivers/rtc/rtc-dm355evm.c

@@ -0,0 +1,175 @@
+/*
+ * rtc-dm355evm.c - access battery-backed counter in MSP430 firmware
+ *
+ * Copyright (c) 2008 by David Brownell
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/rtc.h>
+#include <linux/platform_device.h>
+
+#include <linux/i2c/dm355evm_msp.h>
+
+
+/*
+ * The MSP430 firmware on the DM355 EVM uses a watch crystal to feed
+ * a 1 Hz counter.  When a backup battery is supplied, that makes a
+ * reasonable RTC for applications where alarms and non-NTP drift
+ * compensation aren't important.
+ *
+ * The only real glitch is the inability to read or write all four
+ * counter bytes atomically:  the count may increment in the middle
+ * of an operation, causing trouble when the LSB rolls over.
+ *
+ * This driver was tested with firmware revision A4.
+ */
+union evm_time {
+	u8	bytes[4];
+	u32	value;
+};
+
+static int dm355evm_rtc_read_time(struct device *dev, struct rtc_time *tm)
+{
+	union evm_time	time;
+	int		status;
+	int		tries = 0;
+
+	do {
+		/*
+		 * Read LSB(0) to MSB(3) bytes.  Defend against the counter
+		 * rolling over by re-reading until the value is stable,
+		 * and assuming the four reads take at most a few seconds.
+		 */
+		status = dm355evm_msp_read(DM355EVM_MSP_RTC_0);
+		if (status < 0)
+			return status;
+		if (tries && time.bytes[0] == status)
+			break;
+		time.bytes[0] = status;
+
+		status = dm355evm_msp_read(DM355EVM_MSP_RTC_1);
+		if (status < 0)
+			return status;
+		if (tries && time.bytes[1] == status)
+			break;
+		time.bytes[1] = status;
+
+		status = dm355evm_msp_read(DM355EVM_MSP_RTC_2);
+		if (status < 0)
+			return status;
+		if (tries && time.bytes[2] == status)
+			break;
+		time.bytes[2] = status;
+
+		status = dm355evm_msp_read(DM355EVM_MSP_RTC_3);
+		if (status < 0)
+			return status;
+		if (tries && time.bytes[3] == status)
+			break;
+		time.bytes[3] = status;
+
+	} while (++tries < 5);
+
+	dev_dbg(dev, "read timestamp %08x\n", time.value);
+
+	rtc_time_to_tm(le32_to_cpu(time.value), tm);
+	return 0;
+}
+
+static int dm355evm_rtc_set_time(struct device *dev, struct rtc_time *tm)
+{
+	union evm_time	time;
+	unsigned long	value;
+	int		status;
+
+	rtc_tm_to_time(tm, &value);
+	time.value = cpu_to_le32(value);
+
+	dev_dbg(dev, "write timestamp %08x\n", time.value);
+
+	/*
+	 * REVISIT handle non-atomic writes ... maybe just retry until
+	 * byte[1] sticks (no rollover)?
+	 */
+	status = dm355evm_msp_write(time.bytes[0], DM355EVM_MSP_RTC_0);
+	if (status < 0)
+		return status;
+
+	status = dm355evm_msp_write(time.bytes[1], DM355EVM_MSP_RTC_1);
+	if (status < 0)
+		return status;
+
+	status = dm355evm_msp_write(time.bytes[2], DM355EVM_MSP_RTC_2);
+	if (status < 0)
+		return status;
+
+	status = dm355evm_msp_write(time.bytes[3], DM355EVM_MSP_RTC_3);
+	if (status < 0)
+		return status;
+
+	return 0;
+}
+
+static struct rtc_class_ops dm355evm_rtc_ops = {
+	.read_time	= dm355evm_rtc_read_time,
+	.set_time	= dm355evm_rtc_set_time,
+};
+
+/*----------------------------------------------------------------------*/
+
+static int __devinit dm355evm_rtc_probe(struct platform_device *pdev)
+{
+	struct rtc_device *rtc;
+
+	rtc = rtc_device_register(pdev->name,
+				  &pdev->dev, &dm355evm_rtc_ops, THIS_MODULE);
+	if (IS_ERR(rtc)) {
+		dev_err(&pdev->dev, "can't register RTC device, err %ld\n",
+			PTR_ERR(rtc));
+		return PTR_ERR(rtc);
+	}
+	platform_set_drvdata(pdev, rtc);
+
+	return 0;
+}
+
+static int __devexit dm355evm_rtc_remove(struct platform_device *pdev)
+{
+	struct rtc_device *rtc = platform_get_drvdata(pdev);
+
+	rtc_device_unregister(rtc);
+	platform_set_drvdata(pdev, NULL);
+	return 0;
+}
+
+/*
+ * I2C is used to talk to the MSP430, but this platform device is
+ * exposed by an MFD driver that manages I2C communications.
+ */
+static struct platform_driver rtc_dm355evm_driver = {
+	.probe		= dm355evm_rtc_probe,
+	.remove		= __devexit_p(dm355evm_rtc_remove),
+	.driver		= {
+		.owner	= THIS_MODULE,
+		.name	= "rtc-dm355evm",
+	},
+};
+
+static int __init dm355evm_rtc_init(void)
+{
+	return platform_driver_register(&rtc_dm355evm_driver);
+}
+module_init(dm355evm_rtc_init);
+
+static void __exit dm355evm_rtc_exit(void)
+{
+	platform_driver_unregister(&rtc_dm355evm_driver);
+}
+module_exit(dm355evm_rtc_exit);
+
+MODULE_LICENSE("GPL");

drivers/rtc/rtc-ds1390.c

@@ -122,7 +122,6 @@ static const struct rtc_class_ops ds1390_rtc_ops = {
 
 static int __devinit ds1390_probe(struct spi_device *spi)
 {
-	struct rtc_device *rtc;
 	unsigned char tmp;
 	struct ds1390 *chip;
 	int res;

drivers/video/aty/aty128fb.c

@@ -1475,7 +1475,7 @@ static int aty128fb_set_par(struct fb_info *info)
 	aty128_set_pll(&par->pll, par);
 	aty128_set_fifo(&par->fifo_reg, par);
 
-	config = aty_ld_le32(CONFIG_CNTL) & ~3;
+	config = aty_ld_le32(CNFG_CNTL) & ~3;
 
 #if defined(__BIG_ENDIAN)
 	if (par->crtc.bpp == 32)
@@ -1484,7 +1484,7 @@ static int aty128fb_set_par(struct fb_info *info)
 		config |= 1;	/* make aperture do 16 bit swapping */
 #endif
 
-	aty_st_le32(CONFIG_CNTL, config);
+	aty_st_le32(CNFG_CNTL, config);
 	aty_st_8(CRTC_EXT_CNTL + 1, 0);	/* turn the video back on */
 
 	info->fix.line_length = (par->crtc.vxres * par->crtc.bpp) >> 3;
@@ -1875,7 +1875,7 @@ static int __devinit aty128_init(struct pci_dev *pdev, const struct pci_device_i
 	u32 dac;
 
 	/* Get the chip revision */
-	chip_rev = (aty_ld_le32(CONFIG_CNTL) >> 16) & 0x1F;
+	chip_rev = (aty_ld_le32(CNFG_CNTL) >> 16) & 0x1F;
 
 	strcpy(video_card, "Rage128 XX ");
 	video_card[8] = ent->device >> 8;
@@ -2057,7 +2057,7 @@ static int __devinit aty128_probe(struct pci_dev *pdev, const struct pci_device_
 
 	/* Grab memory size from the card */
 	// How does this relate to the resource length from the PCI hardware?
-	par->vram_size = aty_ld_le32(CONFIG_MEMSIZE) & 0x03FFFFFF;
+	par->vram_size = aty_ld_le32(CNFG_MEMSIZE) & 0x03FFFFFF;
 
 	/* Virtualize the framebuffer */
 	info->screen_base = ioremap(fb_addr, par->vram_size);
@@ -2374,6 +2374,8 @@ static void aty128_set_suspend(struct aty128fb_par *par, int suspend)
 	/* Set the chip into the appropriate suspend mode (we use D2,
 	 * D3 would require a complete re-initialisation of the chip,
 	 * including PCI config registers, clocks, AGP configuration, ...)
+	 *
+	 * For resume, the core will have already brought us back to D0
 	 */
 	if (suspend) {
 		/* Make sure CRTC2 is reset. Remove that the day we decide to
@@ -2391,17 +2393,9 @@ static void aty128_set_suspend(struct aty128fb_par *par, int suspend)
 		aty_st_le32(BUS_CNTL1, 0x00000010);
 		aty_st_le32(MEM_POWER_MISC, 0x0c830000);
 		mdelay(100);
-		pci_read_config_word(pdev, par->pm_reg+PCI_PM_CTRL, &pwr_command);
+
 		/* Switch PCI power management to D2 */
-		pci_write_config_word(pdev, par->pm_reg+PCI_PM_CTRL,
-			(pwr_command & ~PCI_PM_CTRL_STATE_MASK) | 2);
-		pci_read_config_word(pdev, par->pm_reg+PCI_PM_CTRL, &pwr_command);
-	} else {
-		/* Switch back PCI power management to D0 */
-		mdelay(100);
-		pci_write_config_word(pdev, par->pm_reg+PCI_PM_CTRL, 0);
-		pci_read_config_word(pdev, par->pm_reg+PCI_PM_CTRL, &pwr_command);
-		mdelay(100);
+		pci_set_power_state(pdev, PCI_D2);
 	}
 }
 
@@ -2410,6 +2404,12 @@ static int aty128_pci_suspend(struct pci_dev *pdev, pm_message_t state)
 	struct fb_info *info = pci_get_drvdata(pdev);
 	struct aty128fb_par *par = info->par;
 
+	/* Because we may change PCI D state ourselves, we need to
+	 * first save the config space content so the core can
+	 * restore it properly on resume.
+	 */
+	pci_save_state(pdev);
+
 	/* We don't do anything but D2, for now we return 0, but
 	 * we may want to change that. How do we know if the BIOS
 	 * can properly take care of D3 ? Also, with swsusp, we
@@ -2476,6 +2476,11 @@ static int aty128_do_resume(struct pci_dev *pdev)
 	if (pdev->dev.power.power_state.event == PM_EVENT_ON)
 		return 0;
 
+	/* PCI state will have been restored by the core, so
+	 * we should be in D0 now with our config space fully
+	 * restored
+	 */
+
 	/* Wakeup chip */
 	aty128_set_suspend(par, 0);
 	par->asleep = 0;

drivers/video/aty/atyfb_base.c

@@ -135,7 +135,7 @@
 #if defined(CONFIG_PM) || defined(CONFIG_PMAC_BACKLIGHT) || \
 defined (CONFIG_FB_ATY_GENERIC_LCD) || defined(CONFIG_FB_ATY_BACKLIGHT)
 static const u32 lt_lcd_regs[] = {
-	CONFIG_PANEL_LG,
+	CNFG_PANEL_LG,
 	LCD_GEN_CNTL_LG,
 	DSTN_CONTROL_LG,
 	HFB_PITCH_ADDR_LG,
@@ -446,7 +446,7 @@ static int __devinit correct_chipset(struct atyfb_par *par)
 	par->pll_limits.ecp_max = aty_chips[i].ecp_max;
 	par->features = aty_chips[i].features;
 
-	chip_id = aty_ld_le32(CONFIG_CHIP_ID, par);
+	chip_id = aty_ld_le32(CNFG_CHIP_ID, par);
 	type = chip_id & CFG_CHIP_TYPE;
 	rev = (chip_id & CFG_CHIP_REV) >> 24;
 
@@ -629,7 +629,7 @@ static void aty_get_crtc(const struct atyfb_par *par, struct crtc *crtc)
 		    crtc->lcd_index = aty_ld_le32(LCD_INDEX, par);
 		    aty_st_le32(LCD_INDEX, crtc->lcd_index, par);
 		}
-		crtc->lcd_config_panel = aty_ld_lcd(CONFIG_PANEL, par);
+		crtc->lcd_config_panel = aty_ld_lcd(CNFG_PANEL, par);
 		crtc->lcd_gen_cntl = aty_ld_lcd(LCD_GEN_CNTL, par);
 
 
@@ -676,7 +676,7 @@ static void aty_set_crtc(const struct atyfb_par *par, const struct crtc *crtc)
 		aty_st_le32(CRTC_GEN_CNTL, crtc->gen_cntl & ~(CRTC_EXT_DISP_EN | CRTC_EN), par);
 
 		/* update non-shadow registers first */
-		aty_st_lcd(CONFIG_PANEL, crtc->lcd_config_panel, par);
+		aty_st_lcd(CNFG_PANEL, crtc->lcd_config_panel, par);
 		aty_st_lcd(LCD_GEN_CNTL, crtc->lcd_gen_cntl &
 			~(CRTC_RW_SELECT | SHADOW_EN | SHADOW_RW_EN), par);
 
@@ -858,7 +858,7 @@ static int aty_var_to_crtc(const struct fb_info *info,
 		if (!M64_HAS(MOBIL_BUS))
 			crtc->lcd_index |= CRTC2_DISPLAY_DIS;
 
-		crtc->lcd_config_panel = aty_ld_lcd(CONFIG_PANEL, par) | 0x4000;
+		crtc->lcd_config_panel = aty_ld_lcd(CNFG_PANEL, par) | 0x4000;
 		crtc->lcd_gen_cntl = aty_ld_lcd(LCD_GEN_CNTL, par) & ~CRTC_RW_SELECT;
 
 		crtc->lcd_gen_cntl &=
@@ -1978,7 +1978,7 @@ static int aty_power_mgmt(int sleep, struct atyfb_par *par)
 
 	return timeout ? 0 : -EIO;
 }
-#endif
+#endif /* CONFIG_PPC_PMAC */
 
 static int atyfb_pci_suspend(struct pci_dev *pdev, pm_message_t state)
 {
@@ -2002,9 +2002,15 @@ static int atyfb_pci_suspend(struct pci_dev *pdev, pm_message_t state)
 	par->asleep = 1;
 	par->lock_blank = 1;
 
+	/* Because we may change PCI D state ourselves, we need to
+	 * first save the config space content so the core can
+	 * restore it properly on resume.
+	 */
+	pci_save_state(pdev);
+
 #ifdef CONFIG_PPC_PMAC
 	/* Set chip to "suspend" mode */
-	if (aty_power_mgmt(1, par)) {
+	if (machine_is(powermac) && aty_power_mgmt(1, par)) {
 		par->asleep = 0;
 		par->lock_blank = 0;
 		atyfb_blank(FB_BLANK_UNBLANK, info);
@@ -2047,11 +2053,15 @@ static int atyfb_pci_resume(struct pci_dev *pdev)
 
 	acquire_console_sem();
 
+	/* PCI state will have been restored by the core, so
+	 * we should be in D0 now with our config space fully
+	 * restored
+	 */
+
 #ifdef CONFIG_PPC_PMAC
-	if (pdev->dev.power.power_state.event == 2)
+	if (machine_is(powermac) &&
+	    pdev->dev.power.power_state.event == PM_EVENT_SUSPEND)
 		aty_power_mgmt(0, par);
-#else
-	pci_set_power_state(pdev, PCI_D0);
 #endif
 
 	aty_resume_chip(info);
@@ -2254,7 +2264,7 @@ static int __devinit aty_init(struct fb_info *info)
 	if (!M64_HAS(INTEGRATED)) {
 		u32 stat0;
 		u8 dac_type, dac_subtype, clk_type;
-		stat0 = aty_ld_le32(CONFIG_STAT0, par);
+		stat0 = aty_ld_le32(CNFG_STAT0, par);
 		par->bus_type = (stat0 >> 0) & 0x07;
 		par->ram_type = (stat0 >> 3) & 0x07;
 		ramname = aty_gx_ram[par->ram_type];
@@ -2324,7 +2334,7 @@ static int __devinit aty_init(struct fb_info *info)
 		par->dac_ops = &aty_dac_ct;
 		par->pll_ops = &aty_pll_ct;
 		par->bus_type = PCI;
-		par->ram_type = (aty_ld_le32(CONFIG_STAT0, par) & 0x07);
+		par->ram_type = (aty_ld_le32(CNFG_STAT0, par) & 0x07);
 		ramname = aty_ct_ram[par->ram_type];
 		/* for many chips, the mclk is 67 MHz for SDRAM, 63 MHz otherwise */
 		if (par->pll_limits.mclk == 67 && par->ram_type < SDRAM)
@@ -2433,7 +2443,7 @@ static int __devinit aty_init(struct fb_info *info)
 		}
 
 	if (M64_HAS(MAGIC_VRAM_SIZE)) {
-		if (aty_ld_le32(CONFIG_STAT1, par) & 0x40000000)
+		if (aty_ld_le32(CNFG_STAT1, par) & 0x40000000)
 			info->fix.smem_len += 0x400000;
 	}
 
@@ -2946,7 +2956,7 @@ static int __devinit atyfb_setup_sparc(struct pci_dev *pdev,
 		 * Fix PROMs idea of MEM_CNTL settings...
 		 */
 		mem = aty_ld_le32(MEM_CNTL, par);
-		chip_id = aty_ld_le32(CONFIG_CHIP_ID, par);
+		chip_id = aty_ld_le32(CNFG_CHIP_ID, par);
 		if (((chip_id & CFG_CHIP_TYPE) == VT_CHIP_ID) && !((chip_id >> 24) & 1)) {
 			switch (mem & 0x0f) {
 			case 3:
@@ -2964,7 +2974,7 @@ static int __devinit atyfb_setup_sparc(struct pci_dev *pdev,
 			default:
 				break;
 			}
-			if ((aty_ld_le32(CONFIG_STAT0, par) & 7) >= SDRAM)
+			if ((aty_ld_le32(CNFG_STAT0, par) & 7) >= SDRAM)
 				mem &= ~(0x00700000);
 		}
 		mem &= ~(0xcf80e000);	/* Turn off all undocumented bits. */
@@ -3572,7 +3582,7 @@ static int __init atyfb_atari_probe(void)
 		}
 
 		/* Fake pci_id for correct_chipset() */
-		switch (aty_ld_le32(CONFIG_CHIP_ID, par) & CFG_CHIP_TYPE) {
+		switch (aty_ld_le32(CNFG_CHIP_ID, par) & CFG_CHIP_TYPE) {
 		case 0x00d7:
 			par->pci_id = PCI_CHIP_MACH64GX;
 			break;

drivers/video/aty/radeon_base.c

@@ -1936,8 +1936,8 @@ static void fixup_memory_mappings(struct radeonfb_info *rinfo)
 	OUTREG(CRTC_GEN_CNTL, save_crtc_gen_cntl | CRTC_DISP_REQ_EN_B);
 	mdelay(100);
 
-	aper_base = INREG(CONFIG_APER_0_BASE);
-	aper_size = INREG(CONFIG_APER_SIZE);
+	aper_base = INREG(CNFG_APER_0_BASE);
+	aper_size = INREG(CNFG_APER_SIZE);
 
 #ifdef SET_MC_FB_FROM_APERTURE
 	/* Set framebuffer to be at the same address as set in PCI BAR */
@@ -2024,11 +2024,11 @@ static void radeon_identify_vram(struct radeonfb_info *rinfo)
                      ~CRTC_H_CUTOFF_ACTIVE_EN);
           }
         } else {
-          tmp = INREG(CONFIG_MEMSIZE);
+          tmp = INREG(CNFG_MEMSIZE);
         }
 
 	/* mem size is bits [28:0], mask off the rest */
-	rinfo->video_ram = tmp & CONFIG_MEMSIZE_MASK;
+	rinfo->video_ram = tmp & CNFG_MEMSIZE_MASK;
 
 	/*
 	 * Hack to get around some busted production M6's
@@ -2228,7 +2228,7 @@ static int __devinit radeonfb_pci_register (struct pci_dev *pdev,
 	 */
 	rinfo->errata = 0;
 	if (rinfo->family == CHIP_FAMILY_R300 &&
-	    (INREG(CONFIG_CNTL) & CFG_ATI_REV_ID_MASK)
+	    (INREG(CNFG_CNTL) & CFG_ATI_REV_ID_MASK)
 	    == CFG_ATI_REV_A11)
 		rinfo->errata |= CHIP_ERRATA_R300_CG;
 

drivers/video/aty/radeon_pm.c

@@ -333,7 +333,7 @@ static void radeon_pm_enable_dynamic_mode(struct radeonfb_info *rinfo)
 	if (!rinfo->has_CRTC2) {
                 tmp = INPLL(pllSCLK_CNTL);
 
-		if ((INREG(CONFIG_CNTL) & CFG_ATI_REV_ID_MASK) > CFG_ATI_REV_A13)
+		if ((INREG(CNFG_CNTL) & CFG_ATI_REV_ID_MASK) > CFG_ATI_REV_A13)
                     tmp &= ~(SCLK_CNTL__FORCE_CP	| SCLK_CNTL__FORCE_RB);
                 tmp &= ~(SCLK_CNTL__FORCE_HDP		| SCLK_CNTL__FORCE_DISP1 |
 			 SCLK_CNTL__FORCE_TOP		| SCLK_CNTL__FORCE_SE   |
@@ -468,9 +468,9 @@ static void radeon_pm_enable_dynamic_mode(struct radeonfb_info *rinfo)
 
 	/*RAGE_6::A11 A12 A12N1 A13, RV250::A11 A12, R300*/
 	if ((rinfo->family == CHIP_FAMILY_RV250 &&
-	     ((INREG(CONFIG_CNTL) & CFG_ATI_REV_ID_MASK) < CFG_ATI_REV_A13)) ||
+	     ((INREG(CNFG_CNTL) & CFG_ATI_REV_ID_MASK) < CFG_ATI_REV_A13)) ||
 	    ((rinfo->family == CHIP_FAMILY_RV100) &&
-	     ((INREG(CONFIG_CNTL) & CFG_ATI_REV_ID_MASK) <= CFG_ATI_REV_A13))) {
+	     ((INREG(CNFG_CNTL) & CFG_ATI_REV_ID_MASK) <= CFG_ATI_REV_A13))) {
 		tmp |= SCLK_CNTL__FORCE_CP;
 		tmp |= SCLK_CNTL__FORCE_VIP;
 	}
@@ -486,7 +486,7 @@ static void radeon_pm_enable_dynamic_mode(struct radeonfb_info *rinfo)
 		/* RV200::A11 A12 RV250::A11 A12 */
 		if (((rinfo->family == CHIP_FAMILY_RV200) ||
 		     (rinfo->family == CHIP_FAMILY_RV250)) &&
-		    ((INREG(CONFIG_CNTL) & CFG_ATI_REV_ID_MASK) < CFG_ATI_REV_A13))
+		    ((INREG(CNFG_CNTL) & CFG_ATI_REV_ID_MASK) < CFG_ATI_REV_A13))
 			tmp |= SCLK_MORE_CNTL__FORCEON;
 
 		OUTPLL(pllSCLK_MORE_CNTL, tmp);
@@ -497,7 +497,7 @@ static void radeon_pm_enable_dynamic_mode(struct radeonfb_info *rinfo)
 	/* RV200::A11 A12, RV250::A11 A12 */
 	if (((rinfo->family == CHIP_FAMILY_RV200) ||
 	     (rinfo->family == CHIP_FAMILY_RV250)) &&
-	    ((INREG(CONFIG_CNTL) & CFG_ATI_REV_ID_MASK) < CFG_ATI_REV_A13)) {
+	    ((INREG(CNFG_CNTL) & CFG_ATI_REV_ID_MASK) < CFG_ATI_REV_A13)) {
 		tmp = INPLL(pllPLL_PWRMGT_CNTL);
 		tmp |= PLL_PWRMGT_CNTL__TCL_BYPASS_DISABLE;
 		OUTPLL(pllPLL_PWRMGT_CNTL, tmp);
@@ -702,7 +702,7 @@ static void radeon_pm_restore_regs(struct radeonfb_info *rinfo)
 	OUTREG(DISPLAY_BASE_ADDR, rinfo->save_regs[31]);
 	OUTREG(MC_AGP_LOCATION, rinfo->save_regs[32]);
 	OUTREG(CRTC2_DISPLAY_BASE_ADDR, rinfo->save_regs[33]);
-	OUTREG(CONFIG_MEMSIZE, rinfo->video_ram);
+	OUTREG(CNFG_MEMSIZE, rinfo->video_ram);
 
 	OUTREG(DISP_MISC_CNTL, rinfo->save_regs[9]);
 	OUTREG(DISP_PWR_MAN, rinfo->save_regs[10]);
@@ -1723,7 +1723,7 @@ static void radeon_reinitialize_M10(struct radeonfb_info *rinfo)
 	OUTREG(CRTC2_DISPLAY_BASE_ADDR, rinfo->save_regs[33]);
 	OUTREG(MC_FB_LOCATION, rinfo->save_regs[30]);
 	OUTREG(OV0_BASE_ADDR, rinfo->save_regs[80]);
-	OUTREG(CONFIG_MEMSIZE, rinfo->video_ram);
+	OUTREG(CNFG_MEMSIZE, rinfo->video_ram);
 	OUTREG(BUS_CNTL, rinfo->save_regs[36]);
 	OUTREG(BUS_CNTL1, rinfo->save_regs[14]);
 	OUTREG(MPP_TB_CONFIG, rinfo->save_regs[37]);
@@ -1961,7 +1961,7 @@ static void radeon_pm_m9p_reconfigure_mc(struct radeonfb_info *rinfo)
 	OUTMC(rinfo, ixMC_CHP_IO_CNTL_B1, rinfo->save_regs[68] /*0x141555ff*/);
 	OUTMC(rinfo, ixMC_IMP_CNTL_0, rinfo->save_regs[71] /*0x00009249*/);
 	OUTREG(MC_IND_INDEX, 0);
-	OUTREG(CONFIG_MEMSIZE, rinfo->video_ram);
+	OUTREG(CNFG_MEMSIZE, rinfo->video_ram);
 
 	mdelay(20);
 }
@@ -2361,7 +2361,7 @@ static void radeon_reinitialize_QW(struct radeonfb_info *rinfo)
 	OUTMC(rinfo, ixMC_IMP_CNTL_0, 0x00009249);
 	OUTREG(MC_IND_INDEX, 0);
 
-	OUTREG(CONFIG_MEMSIZE, rinfo->video_ram);
+	OUTREG(CNFG_MEMSIZE, rinfo->video_ram);
 
 	radeon_pm_full_reset_sdram(rinfo);
 
@@ -2509,9 +2509,7 @@ static void radeon_reinitialize_QW(struct radeonfb_info *rinfo)
 
 static void radeon_set_suspend(struct radeonfb_info *rinfo, int suspend)
 {
-	u16 pwr_cmd;
 	u32 tmp;
-	int i;
 
 	if (!rinfo->pm_reg)
 		return;
@@ -2557,32 +2555,14 @@ static void radeon_set_suspend(struct radeonfb_info *rinfo, int suspend)
 			}
 		}
 
-		for (i = 0; i < 64; ++i)
-			pci_read_config_dword(rinfo->pdev, i * 4,
-					      &rinfo->cfg_save[i]);
-
 		/* Switch PCI power management to D2. */
 		pci_disable_device(rinfo->pdev);
-		for (;;) {
-			pci_read_config_word(
-				rinfo->pdev, rinfo->pm_reg+PCI_PM_CTRL,
-				&pwr_cmd);
-			if (pwr_cmd & 2)
-				break;			
-			pci_write_config_word(
-				rinfo->pdev, rinfo->pm_reg+PCI_PM_CTRL,
-				(pwr_cmd & ~PCI_PM_CTRL_STATE_MASK) | 2);
-			mdelay(500);
-		}
+		pci_save_state(rinfo->pdev);
+		pci_set_power_state(rinfo->pdev, PCI_D2);
 	} else {
 		printk(KERN_DEBUG "radeonfb (%s): switching to D0 state...\n",
 		       pci_name(rinfo->pdev));
 
-		/* Switch back PCI powermanagment to D0 */
-		mdelay(200);
-		pci_write_config_word(rinfo->pdev, rinfo->pm_reg+PCI_PM_CTRL, 0);
-		mdelay(500);
-
 		if (rinfo->family <= CHIP_FAMILY_RV250) {
 			/* Reset the SDRAM controller  */
 			radeon_pm_full_reset_sdram(rinfo);
@@ -2598,37 +2578,10 @@ static void radeon_set_suspend(struct radeonfb_info *rinfo, int suspend)
 	}
 }
 
-static int radeon_restore_pci_cfg(struct radeonfb_info *rinfo)
-{
-	int i;
-	static u32 radeon_cfg_after_resume[64];
-
-	for (i = 0; i < 64; ++i)
-		pci_read_config_dword(rinfo->pdev, i * 4,
-				      &radeon_cfg_after_resume[i]);
-
-	if (radeon_cfg_after_resume[PCI_BASE_ADDRESS_0/4]
-	    == rinfo->cfg_save[PCI_BASE_ADDRESS_0/4])
-		return 0;	/* assume everything is ok */
-
-	for (i = PCI_BASE_ADDRESS_0/4; i < 64; ++i) {
-		if (radeon_cfg_after_resume[i] != rinfo->cfg_save[i])
-			pci_write_config_dword(rinfo->pdev, i * 4,
-					       rinfo->cfg_save[i]);
-	}
-	pci_write_config_word(rinfo->pdev, PCI_CACHE_LINE_SIZE,
-			      rinfo->cfg_save[PCI_CACHE_LINE_SIZE/4]);
-	pci_write_config_word(rinfo->pdev, PCI_COMMAND,
-			      rinfo->cfg_save[PCI_COMMAND/4]);
-	return 1;
-}
-
-
 int radeonfb_pci_suspend(struct pci_dev *pdev, pm_message_t mesg)
 {
         struct fb_info *info = pci_get_drvdata(pdev);
         struct radeonfb_info *rinfo = info->par;
-	int i;
 
 	if (mesg.event == pdev->dev.power.power_state.event)
 		return 0;
@@ -2674,6 +2627,11 @@ int radeonfb_pci_suspend(struct pci_dev *pdev, pm_message_t mesg)
 	pmac_suspend_agp_for_card(pdev);
 #endif /* CONFIG_PPC_PMAC */
 
+	/* It's unclear whether or when the generic code will do that, so let's
+	 * do it ourselves. We save state before we do any power management
+	 */
+	pci_save_state(pdev);
+
 	/* If we support wakeup from poweroff, we save all regs we can including cfg
 	 * space
 	 */
@@ -2698,9 +2656,6 @@ int radeonfb_pci_suspend(struct pci_dev *pdev, pm_message_t mesg)
 			mdelay(20);
 			OUTREG(LVDS_GEN_CNTL, INREG(LVDS_GEN_CNTL) & ~(LVDS_DIGON));
 		}
-		// FIXME: Use PCI layer
-		for (i = 0; i < 64; ++i)
-			pci_read_config_dword(pdev, i * 4, &rinfo->cfg_save[i]);
 		pci_disable_device(pdev);
 	}
 	/* If we support D2, we go to it (should be fixed later with a flag forcing
@@ -2717,6 +2672,13 @@ int radeonfb_pci_suspend(struct pci_dev *pdev, pm_message_t mesg)
 	return 0;
 }
 
+static int radeon_check_power_loss(struct radeonfb_info *rinfo)
+{
+	return rinfo->save_regs[4] != INPLL(CLK_PIN_CNTL) ||
+	       rinfo->save_regs[2] != INPLL(MCLK_CNTL) ||
+	       rinfo->save_regs[3] != INPLL(SCLK_CNTL);
+}
+
 int radeonfb_pci_resume(struct pci_dev *pdev)
 {
         struct fb_info *info = pci_get_drvdata(pdev);
@@ -2735,20 +2697,13 @@ int radeonfb_pci_resume(struct pci_dev *pdev)
 	printk(KERN_DEBUG "radeonfb (%s): resuming from state: %d...\n",
 	       pci_name(pdev), pdev->dev.power.power_state.event);
 
-
-	if (pci_enable_device(pdev)) {
-		rc = -ENODEV;
-		printk(KERN_ERR "radeonfb (%s): can't enable PCI device !\n",
-		       pci_name(pdev));
-		goto bail;
-	}
-	pci_set_master(pdev);
-
+	/* PCI state will have been restored by the core, so
+	 * we should be in D0 now with our config space fully
+	 * restored
+	 */
 	if (pdev->dev.power.power_state.event == PM_EVENT_SUSPEND) {
-		/* Wakeup chip. Check from config space if we were powered off
-		 * (todo: additionally, check CLK_PIN_CNTL too)
-		 */
-		if ((rinfo->pm_mode & radeon_pm_off) && radeon_restore_pci_cfg(rinfo)) {
+		/* Wakeup chip */
+		if ((rinfo->pm_mode & radeon_pm_off) && radeon_check_power_loss(rinfo)) {
 			if (rinfo->reinit_func != NULL)
 				rinfo->reinit_func(rinfo);
 			else {

drivers/video/aty/radeonfb.h

@@ -361,8 +361,6 @@ struct radeonfb_info {
 #ifdef CONFIG_FB_RADEON_I2C
 	struct radeon_i2c_chan 	i2c[4];
 #endif
-
-	u32			cfg_save[64];
 };
 
 

drivers/video/backlight/Makefile

@@ -18,7 +18,7 @@ obj-$(CONFIG_BACKLIGHT_OMAP1)	+= omap1_bl.o
 obj-$(CONFIG_BACKLIGHT_PROGEAR) += progear_bl.o
 obj-$(CONFIG_BACKLIGHT_CARILLO_RANCH) += cr_bllcd.o
 obj-$(CONFIG_BACKLIGHT_PWM)	+= pwm_bl.o
-obj-$(CONFIG_BACKLIGHT_DA903X)	+= da903x.o
+obj-$(CONFIG_BACKLIGHT_DA903X)	+= da903x_bl.o
 obj-$(CONFIG_BACKLIGHT_MBP_NVIDIA) += mbp_nvidia_bl.o
 obj-$(CONFIG_BACKLIGHT_TOSA)	+= tosa_bl.o
 obj-$(CONFIG_BACKLIGHT_SAHARA)	+= kb3886_bl.o

drivers/video/fbcmap.c

@@ -250,10 +250,6 @@ int fb_set_user_cmap(struct fb_cmap_user *cmap, struct fb_info *info)
 	int rc, size = cmap->len * sizeof(u16);
 	struct fb_cmap umap;
 
-	if (cmap->start < 0 || (!info->fbops->fb_setcolreg &&
-			        !info->fbops->fb_setcmap))
-		return -EINVAL;
-
 	memset(&umap, 0, sizeof(struct fb_cmap));
 	rc = fb_alloc_cmap(&umap, cmap->len, cmap->transp != NULL);
 	if (rc)
@@ -262,11 +258,23 @@ int fb_set_user_cmap(struct fb_cmap_user *cmap, struct fb_info *info)
 	    copy_from_user(umap.green, cmap->green, size) ||
 	    copy_from_user(umap.blue, cmap->blue, size) ||
 	    (cmap->transp && copy_from_user(umap.transp, cmap->transp, size))) {
-		fb_dealloc_cmap(&umap);
-		return -EFAULT;
+		rc = -EFAULT;
+		goto out;
 	}
 	umap.start = cmap->start;
+	if (!lock_fb_info(info)) {
+		rc = -ENODEV;
+		goto out;
+	}
+	if (cmap->start < 0 || (!info->fbops->fb_setcolreg &&
+				!info->fbops->fb_setcmap)) {
+		rc = -EINVAL;
+		goto out1;
+	}
 	rc = fb_set_cmap(&umap, info);
+out1:
+	unlock_fb_info(info);
+out:
 	fb_dealloc_cmap(&umap);
 	return rc;
 }

drivers/video/fbmem.c

@@ -1013,132 +1013,139 @@ static long do_fb_ioctl(struct fb_info *info, unsigned int cmd,
 	struct fb_var_screeninfo var;
 	struct fb_fix_screeninfo fix;
 	struct fb_con2fbmap con2fb;
+	struct fb_cmap cmap_from;
 	struct fb_cmap_user cmap;
 	struct fb_event event;
 	void __user *argp = (void __user *)arg;
 	long ret = 0;
 
-	fb = info->fbops;
-	if (!fb)
-		return -ENODEV;
-
 	switch (cmd) {
 	case FBIOGET_VSCREENINFO:
-		ret = copy_to_user(argp, &info->var,
-				    sizeof(var)) ? -EFAULT : 0;
+		if (!lock_fb_info(info))
+			return -ENODEV;
+		var = info->var;
+		unlock_fb_info(info);
+
+		ret = copy_to_user(argp, &var, sizeof(var)) ? -EFAULT : 0;
 		break;
 	case FBIOPUT_VSCREENINFO:
-		if (copy_from_user(&var, argp, sizeof(var))) {
-			ret =  -EFAULT;
-			break;
-		}
+		if (copy_from_user(&var, argp, sizeof(var)))
+			return -EFAULT;
+		if (!lock_fb_info(info))
+			return -ENODEV;
 		acquire_console_sem();
 		info->flags |= FBINFO_MISC_USEREVENT;
 		ret = fb_set_var(info, &var);
 		info->flags &= ~FBINFO_MISC_USEREVENT;
 		release_console_sem();
-		if (ret == 0 && copy_to_user(argp, &var, sizeof(var)))
+		unlock_fb_info(info);
+		if (!ret && copy_to_user(argp, &var, sizeof(var)))
 			ret = -EFAULT;
 		break;
 	case FBIOGET_FSCREENINFO:
-		ret = copy_to_user(argp, &info->fix,
-				    sizeof(fix)) ? -EFAULT : 0;
+		if (!lock_fb_info(info))
+			return -ENODEV;
+		fix = info->fix;
+		unlock_fb_info(info);
+
+		ret = copy_to_user(argp, &fix, sizeof(fix)) ? -EFAULT : 0;
 		break;
 	case FBIOPUTCMAP:
 		if (copy_from_user(&cmap, argp, sizeof(cmap)))
-			ret = -EFAULT;
-		else
-			ret = fb_set_user_cmap(&cmap, info);
+			return -EFAULT;
+		ret = fb_set_user_cmap(&cmap, info);
 		break;
 	case FBIOGETCMAP:
 		if (copy_from_user(&cmap, argp, sizeof(cmap)))
-			ret = -EFAULT;
-		else
-			ret = fb_cmap_to_user(&info->cmap, &cmap);
+			return -EFAULT;
+		if (!lock_fb_info(info))
+			return -ENODEV;
+		cmap_from = info->cmap;
+		unlock_fb_info(info);
+		ret = fb_cmap_to_user(&cmap_from, &cmap);
 		break;
 	case FBIOPAN_DISPLAY:
-		if (copy_from_user(&var, argp, sizeof(var))) {
-			ret = -EFAULT;
-			break;
-		}
+		if (copy_from_user(&var, argp, sizeof(var)))
+			return -EFAULT;
+		if (!lock_fb_info(info))
+			return -ENODEV;
 		acquire_console_sem();
 		ret = fb_pan_display(info, &var);
 		release_console_sem();
+		unlock_fb_info(info);
 		if (ret == 0 && copy_to_user(argp, &var, sizeof(var)))
-			ret = -EFAULT;
+			return -EFAULT;
 		break;
 	case FBIO_CURSOR:
 		ret = -EINVAL;
 		break;
 	case FBIOGET_CON2FBMAP:
 		if (copy_from_user(&con2fb, argp, sizeof(con2fb)))
-			ret = -EFAULT;
-		else if (con2fb.console < 1 || con2fb.console > MAX_NR_CONSOLES)
-			ret = -EINVAL;
-		else {
-			con2fb.framebuffer = -1;
-			event.info = info;
-			event.data = &con2fb;
-			fb_notifier_call_chain(FB_EVENT_GET_CONSOLE_MAP,
-								&event);
-			ret = copy_to_user(argp, &con2fb,
-				    sizeof(con2fb)) ? -EFAULT : 0;
-		}
+			return -EFAULT;
+		if (con2fb.console < 1 || con2fb.console > MAX_NR_CONSOLES)
+			return -EINVAL;
+		con2fb.framebuffer = -1;
+		event.data = &con2fb;
+
+		if (!lock_fb_info(info))
+			return -ENODEV;
+		event.info = info;
+		fb_notifier_call_chain(FB_EVENT_GET_CONSOLE_MAP, &event);
+		unlock_fb_info(info);
+
+		ret = copy_to_user(argp, &con2fb, sizeof(con2fb)) ? -EFAULT : 0;
 		break;
 	case FBIOPUT_CON2FBMAP:
-		if (copy_from_user(&con2fb, argp, sizeof(con2fb))) {
-			ret = -EFAULT;
-			break;
-		}
-		if (con2fb.console < 1 || con2fb.console > MAX_NR_CONSOLES) {
-			ret = -EINVAL;
-			break;
-		}
-		if (con2fb.framebuffer < 0 || con2fb.framebuffer >= FB_MAX) {
-			ret = -EINVAL;
-			break;
-		}
+		if (copy_from_user(&con2fb, argp, sizeof(con2fb)))
+			return -EFAULT;
+		if (con2fb.console < 1 || con2fb.console > MAX_NR_CONSOLES)
+			return -EINVAL;
+		if (con2fb.framebuffer < 0 || con2fb.framebuffer >= FB_MAX)
+			return -EINVAL;
 		if (!registered_fb[con2fb.framebuffer])
 			request_module("fb%d", con2fb.framebuffer);
 		if (!registered_fb[con2fb.framebuffer]) {
 			ret = -EINVAL;
 			break;
 		}
-		event.info = info;
 		event.data = &con2fb;
+		if (!lock_fb_info(info))
+			return -ENODEV;
+		event.info = info;
 		ret = fb_notifier_call_chain(FB_EVENT_SET_CONSOLE_MAP,
 					      &event);
+		unlock_fb_info(info);
 		break;
 	case FBIOBLANK:
+		if (!lock_fb_info(info))
+			return -ENODEV;
 		acquire_console_sem();
 		info->flags |= FBINFO_MISC_USEREVENT;
 		ret = fb_blank(info, arg);
 		info->flags &= ~FBINFO_MISC_USEREVENT;
 		release_console_sem();
-		break;;
+		unlock_fb_info(info);
+		break;
 	default:
-		if (fb->fb_ioctl == NULL)
-			ret = -ENOTTY;
-		else
+		if (!lock_fb_info(info))
+			return -ENODEV;
+		fb = info->fbops;
+		if (fb->fb_ioctl)
 			ret = fb->fb_ioctl(info, cmd, arg);
+		else
+			ret = -ENOTTY;
+		unlock_fb_info(info);
 	}
 	return ret;
 }
 
 static long fb_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
-__acquires(&info->lock)
-__releases(&info->lock)
 {
 	struct inode *inode = file->f_path.dentry->d_inode;
 	int fbidx = iminor(inode);
-	struct fb_info *info;
-	long ret;
+	struct fb_info *info = registered_fb[fbidx];
 
-	info = registered_fb[fbidx];
-	mutex_lock(&info->lock);
-	ret = do_fb_ioctl(info, cmd, arg);
-	mutex_unlock(&info->lock);
-	return ret;
+	return do_fb_ioctl(info, cmd, arg);
 }
 
 #ifdef CONFIG_COMPAT
@@ -1257,8 +1264,6 @@ static int fb_get_fscreeninfo(struct fb_info *info, unsigned int cmd,
 
 static long fb_compat_ioctl(struct file *file, unsigned int cmd,
 			    unsigned long arg)
-__acquires(&info->lock)
-__releases(&info->lock)
 {
 	struct inode *inode = file->f_path.dentry->d_inode;
 	int fbidx = iminor(inode);
@@ -1266,7 +1271,6 @@ __releases(&info->lock)
 	struct fb_ops *fb = info->fbops;
 	long ret = -ENOIOCTLCMD;
 
-	mutex_lock(&info->lock);
 	switch(cmd) {
 	case FBIOGET_VSCREENINFO:
 	case FBIOPUT_VSCREENINFO:
@@ -1292,7 +1296,6 @@ __releases(&info->lock)
 			ret = fb->fb_compat_ioctl(info, cmd, arg);
 		break;
 	}
-	mutex_unlock(&info->lock);
 	return ret;
 }
 #endif

fs/binfmt_elf.c

@@ -1208,9 +1208,11 @@ static unsigned long vma_dump_size(struct vm_area_struct *vma,
 	 * check for an ELF header.  If we find one, dump the first page to
 	 * aid in determining what was mapped here.
 	 */
-	if (FILTER(ELF_HEADERS) && vma->vm_file != NULL && vma->vm_pgoff == 0) {
+	if (FILTER(ELF_HEADERS) &&
+	    vma->vm_pgoff == 0 && (vma->vm_flags & VM_READ)) {
 		u32 __user *header = (u32 __user *) vma->vm_start;
 		u32 word;
+		mm_segment_t fs = get_fs();
 		/*
 		 * Doing it this way gets the constant folded by GCC.
 		 */
@@ -1223,7 +1225,15 @@ static unsigned long vma_dump_size(struct vm_area_struct *vma,
 		magic.elfmag[EI_MAG1] = ELFMAG1;
 		magic.elfmag[EI_MAG2] = ELFMAG2;
 		magic.elfmag[EI_MAG3] = ELFMAG3;
-		if (get_user(word, header) == 0 && word == magic.cmp)
+		/*
+		 * Switch to the user "segment" for get_user(),
+		 * then put back what elf_core_dump() had in place.
+		 */
+		set_fs(USER_DS);
+		if (unlikely(get_user(word, header)))
+			word = 0;
+		set_fs(fs);
+		if (word == magic.cmp)
 			return PAGE_SIZE;
 	}
 

fs/btrfs/Kconfig

@@ -16,3 +16,16 @@ config BTRFS_FS
 	  module will be called btrfs.
 
 	  If unsure, say N.
+
+config BTRFS_FS_POSIX_ACL
+	bool "Btrfs POSIX Access Control Lists"
+	depends on BTRFS_FS
+	select FS_POSIX_ACL
+	help
+	  POSIX Access Control Lists (ACLs) support permissions for users and
+	  groups beyond the owner/group/world scheme.
+
+	  To learn more about Access Control Lists, visit the POSIX ACLs for
+	  Linux website <http://acl.bestbits.at/>.
+
+	  If you don't know what Access Control Lists are, say N

fs/btrfs/async-thread.c

@@ -16,11 +16,11 @@
  * Boston, MA 021110-1307, USA.
  */
 
-#include <linux/version.h>
 #include <linux/kthread.h>
 #include <linux/list.h>
 #include <linux/spinlock.h>
-# include <linux/freezer.h>
+#include <linux/freezer.h>
+#include <linux/ftrace.h>
 #include "async-thread.h"
 
 #define WORK_QUEUED_BIT 0
@@ -143,6 +143,7 @@ static int worker_loop(void *arg)
 	struct btrfs_work *work;
 	do {
 		spin_lock_irq(&worker->lock);
+again_locked:
 		while (!list_empty(&worker->pending)) {
 			cur = worker->pending.next;
 			work = list_entry(cur, struct btrfs_work, list);
@@ -165,14 +166,50 @@ static int worker_loop(void *arg)
 			check_idle_worker(worker);
 
 		}
-		worker->working = 0;
 		if (freezing(current)) {
+			worker->working = 0;
+			spin_unlock_irq(&worker->lock);
 			refrigerator();
 		} else {
-			set_current_state(TASK_INTERRUPTIBLE);
 			spin_unlock_irq(&worker->lock);
-			if (!kthread_should_stop())
+			if (!kthread_should_stop()) {
+				cpu_relax();
+				/*
+				 * we've dropped the lock, did someone else
+				 * jump_in?
+				 */
+				smp_mb();
+				if (!list_empty(&worker->pending))
+					continue;
+
+				/*
+				 * this short schedule allows more work to
+				 * come in without the queue functions
+				 * needing to go through wake_up_process()
+				 *
+				 * worker->working is still 1, so nobody
+				 * is going to try and wake us up
+				 */
+				schedule_timeout(1);
+				smp_mb();
+				if (!list_empty(&worker->pending))
+					continue;
+
+				/* still no more work?, sleep for real */
+				spin_lock_irq(&worker->lock);
+				set_current_state(TASK_INTERRUPTIBLE);
+				if (!list_empty(&worker->pending))
+					goto again_locked;
+
+				/*
+				 * this makes sure we get a wakeup when someone
+				 * adds something new to the queue
+				 */
+				worker->working = 0;
+				spin_unlock_irq(&worker->lock);
+
 				schedule();
+			}
 			__set_current_state(TASK_RUNNING);
 		}
 	} while (!kthread_should_stop());
@@ -350,13 +387,14 @@ int btrfs_requeue_work(struct btrfs_work *work)
 {
 	struct btrfs_worker_thread *worker = work->worker;
 	unsigned long flags;
+	int wake = 0;
 
 	if (test_and_set_bit(WORK_QUEUED_BIT, &work->flags))
 		goto out;
 
 	spin_lock_irqsave(&worker->lock, flags);
-	atomic_inc(&worker->num_pending);
 	list_add_tail(&work->list, &worker->pending);
+	atomic_inc(&worker->num_pending);
 
 	/* by definition we're busy, take ourselves off the idle
 	 * list
@@ -368,10 +406,16 @@ int btrfs_requeue_work(struct btrfs_work *work)
 			       &worker->workers->worker_list);
 		spin_unlock_irqrestore(&worker->workers->lock, flags);
 	}
+	if (!worker->working) {
+		wake = 1;
+		worker->working = 1;
+	}
 
 	spin_unlock_irqrestore(&worker->lock, flags);
-
+	if (wake)
+		wake_up_process(worker->task);
 out:
+
 	return 0;
 }
 
@@ -398,9 +442,10 @@ int btrfs_queue_worker(struct btrfs_workers *workers, struct btrfs_work *work)
 	}
 
 	spin_lock_irqsave(&worker->lock, flags);
+
+	list_add_tail(&work->list, &worker->pending);
 	atomic_inc(&worker->num_pending);
 	check_busy_worker(worker);
-	list_add_tail(&work->list, &worker->pending);
 
 	/*
 	 * avoid calling into wake_up_process if this thread has already

fs/btrfs/compression.c

@@ -32,7 +32,6 @@
 #include <linux/swap.h>
 #include <linux/writeback.h>
 #include <linux/bit_spinlock.h>
-#include <linux/version.h>
 #include <linux/pagevec.h>
 #include "compat.h"
 #include "ctree.h"

fs/btrfs/ctree.c

@@ -54,6 +54,31 @@ struct btrfs_path *btrfs_alloc_path(void)
 	return path;
 }
 
+/*
+ * set all locked nodes in the path to blocking locks.  This should
+ * be done before scheduling
+ */
+noinline void btrfs_set_path_blocking(struct btrfs_path *p)
+{
+	int i;
+	for (i = 0; i < BTRFS_MAX_LEVEL; i++) {
+		if (p->nodes[i] && p->locks[i])
+			btrfs_set_lock_blocking(p->nodes[i]);
+	}
+}
+
+/*
+ * reset all the locked nodes in the patch to spinning locks.
+ */
+noinline void btrfs_clear_path_blocking(struct btrfs_path *p)
+{
+	int i;
+	for (i = 0; i < BTRFS_MAX_LEVEL; i++) {
+		if (p->nodes[i] && p->locks[i])
+			btrfs_clear_lock_blocking(p->nodes[i]);
+	}
+}
+
 /* this also releases the path */
 void btrfs_free_path(struct btrfs_path *p)
 {
@@ -272,6 +297,8 @@ static noinline int __btrfs_cow_block(struct btrfs_trans_handle *trans,
 	if (IS_ERR(cow))
 		return PTR_ERR(cow);
 
+	/* cow is set to blocking by btrfs_init_new_buffer */
+
 	copy_extent_buffer(cow, buf, 0, 0, cow->len);
 	btrfs_set_header_bytenr(cow, cow->start);
 	btrfs_set_header_generation(cow, trans->transid);
@@ -388,17 +415,20 @@ noinline int btrfs_cow_block(struct btrfs_trans_handle *trans,
 		WARN_ON(1);
 	}
 
-	spin_lock(&root->fs_info->hash_lock);
 	if (btrfs_header_generation(buf) == trans->transid &&
 	    btrfs_header_owner(buf) == root->root_key.objectid &&
 	    !btrfs_header_flag(buf, BTRFS_HEADER_FLAG_WRITTEN)) {
 		*cow_ret = buf;
-		spin_unlock(&root->fs_info->hash_lock);
 		WARN_ON(prealloc_dest);
 		return 0;
 	}
-	spin_unlock(&root->fs_info->hash_lock);
+
 	search_start = buf->start & ~((u64)(1024 * 1024 * 1024) - 1);
+
+	if (parent)
+		btrfs_set_lock_blocking(parent);
+	btrfs_set_lock_blocking(buf);
+
 	ret = __btrfs_cow_block(trans, root, buf, parent,
 				 parent_slot, cow_ret, search_start, 0,
 				 prealloc_dest);
@@ -504,6 +534,8 @@ int btrfs_realloc_node(struct btrfs_trans_handle *trans,
 	if (parent_nritems == 1)
 		return 0;
 
+	btrfs_set_lock_blocking(parent);
+
 	for (i = start_slot; i < end_slot; i++) {
 		int close = 1;
 
@@ -564,6 +596,7 @@ int btrfs_realloc_node(struct btrfs_trans_handle *trans,
 			search_start = last_block;
 
 		btrfs_tree_lock(cur);
+		btrfs_set_lock_blocking(cur);
 		err = __btrfs_cow_block(trans, root, cur, parent, i,
 					&cur, search_start,
 					min(16 * blocksize,
@@ -862,6 +895,7 @@ static noinline int balance_level(struct btrfs_trans_handle *trans,
 		return 0;
 
 	mid = path->nodes[level];
+
 	WARN_ON(!path->locks[level]);
 	WARN_ON(btrfs_header_generation(mid) != trans->transid);
 
@@ -884,6 +918,7 @@ static noinline int balance_level(struct btrfs_trans_handle *trans,
 		/* promote the child to a root */
 		child = read_node_slot(root, mid, 0);
 		btrfs_tree_lock(child);
+		btrfs_set_lock_blocking(child);
 		BUG_ON(!child);
 		ret = btrfs_cow_block(trans, root, child, mid, 0, &child, 0);
 		BUG_ON(ret);
@@ -900,6 +935,7 @@ static noinline int balance_level(struct btrfs_trans_handle *trans,
 
 		add_root_to_dirty_list(root);
 		btrfs_tree_unlock(child);
+
 		path->locks[level] = 0;
 		path->nodes[level] = NULL;
 		clean_tree_block(trans, root, mid);
@@ -924,6 +960,7 @@ static noinline int balance_level(struct btrfs_trans_handle *trans,
 	left = read_node_slot(root, parent, pslot - 1);
 	if (left) {
 		btrfs_tree_lock(left);
+		btrfs_set_lock_blocking(left);
 		wret = btrfs_cow_block(trans, root, left,
 				       parent, pslot - 1, &left, 0);
 		if (wret) {
@@ -934,6 +971,7 @@ static noinline int balance_level(struct btrfs_trans_handle *trans,
 	right = read_node_slot(root, parent, pslot + 1);
 	if (right) {
 		btrfs_tree_lock(right);
+		btrfs_set_lock_blocking(right);
 		wret = btrfs_cow_block(trans, root, right,
 				       parent, pslot + 1, &right, 0);
 		if (wret) {
@@ -1109,6 +1147,8 @@ static noinline int push_nodes_for_insert(struct btrfs_trans_handle *trans,
 		u32 left_nr;
 
 		btrfs_tree_lock(left);
+		btrfs_set_lock_blocking(left);
+
 		left_nr = btrfs_header_nritems(left);
 		if (left_nr >= BTRFS_NODEPTRS_PER_BLOCK(root) - 1) {
 			wret = 1;
@@ -1155,7 +1195,10 @@ static noinline int push_nodes_for_insert(struct btrfs_trans_handle *trans,
 	 */
 	if (right) {
 		u32 right_nr;
+
 		btrfs_tree_lock(right);
+		btrfs_set_lock_blocking(right);
+
 		right_nr = btrfs_header_nritems(right);
 		if (right_nr >= BTRFS_NODEPTRS_PER_BLOCK(root) - 1) {
 			wret = 1;
@@ -1210,8 +1253,7 @@ static noinline void reada_for_search(struct btrfs_root *root,
 	struct btrfs_disk_key disk_key;
 	u32 nritems;
 	u64 search;
-	u64 lowest_read;
-	u64 highest_read;
+	u64 target;
 	u64 nread = 0;
 	int direction = path->reada;
 	struct extent_buffer *eb;
@@ -1235,8 +1277,7 @@ static noinline void reada_for_search(struct btrfs_root *root,
 		return;
 	}
 
-	highest_read = search;
-	lowest_read = search;
+	target = search;
 
 	nritems = btrfs_header_nritems(node);
 	nr = slot;
@@ -1256,27 +1297,80 @@ static noinline void reada_for_search(struct btrfs_root *root,
 				break;
 		}
 		search = btrfs_node_blockptr(node, nr);
-		if ((search >= lowest_read && search <= highest_read) ||
-		    (search < lowest_read && lowest_read - search <= 16384) ||
-		    (search > highest_read && search - highest_read <= 16384)) {
+		if ((search <= target && target - search <= 65536) ||
+		    (search > target && search - target <= 65536)) {
 			readahead_tree_block(root, search, blocksize,
 				     btrfs_node_ptr_generation(node, nr));
 			nread += blocksize;
 		}
 		nscan++;
-		if (path->reada < 2 && (nread > (64 * 1024) || nscan > 32))
+		if ((nread > 65536 || nscan > 32))
 			break;
+	}
+}
 
-		if (nread > (256 * 1024) || nscan > 128)
-			break;
+/*
+ * returns -EAGAIN if it had to drop the path, or zero if everything was in
+ * cache
+ */
+static noinline int reada_for_balance(struct btrfs_root *root,
+				      struct btrfs_path *path, int level)
+{
+	int slot;
+	int nritems;
+	struct extent_buffer *parent;
+	struct extent_buffer *eb;
+	u64 gen;
+	u64 block1 = 0;
+	u64 block2 = 0;
+	int ret = 0;
+	int blocksize;
 
-		if (search < lowest_read)
-			lowest_read = search;
-		if (search > highest_read)
-			highest_read = search;
+	parent = path->nodes[level - 1];
+	if (!parent)
+		return 0;
+
+	nritems = btrfs_header_nritems(parent);
+	slot = path->slots[level];
+	blocksize = btrfs_level_size(root, level);
+
+	if (slot > 0) {
+		block1 = btrfs_node_blockptr(parent, slot - 1);
+		gen = btrfs_node_ptr_generation(parent, slot - 1);
+		eb = btrfs_find_tree_block(root, block1, blocksize);
+		if (eb && btrfs_buffer_uptodate(eb, gen))
+			block1 = 0;
+		free_extent_buffer(eb);
+	}
+	if (slot < nritems) {
+		block2 = btrfs_node_blockptr(parent, slot + 1);
+		gen = btrfs_node_ptr_generation(parent, slot + 1);
+		eb = btrfs_find_tree_block(root, block2, blocksize);
+		if (eb && btrfs_buffer_uptodate(eb, gen))
+			block2 = 0;
+		free_extent_buffer(eb);
+	}
+	if (block1 || block2) {
+		ret = -EAGAIN;
+		btrfs_release_path(root, path);
+		if (block1)
+			readahead_tree_block(root, block1, blocksize, 0);
+		if (block2)
+			readahead_tree_block(root, block2, blocksize, 0);
+
+		if (block1) {
+			eb = read_tree_block(root, block1, blocksize, 0);
+			free_extent_buffer(eb);
+		}
+		if (block1) {
+			eb = read_tree_block(root, block2, blocksize, 0);
+			free_extent_buffer(eb);
+		}
 	}
+	return ret;
 }
 
+
 /*
  * when we walk down the tree, it is usually safe to unlock the higher layers
  * in the tree.  The exceptions are when our path goes through slot 0, because
@@ -1327,6 +1421,32 @@ static noinline void unlock_up(struct btrfs_path *path, int level,
 	}
 }
 
+/*
+ * This releases any locks held in the path starting at level and
+ * going all the way up to the root.
+ *
+ * btrfs_search_slot will keep the lock held on higher nodes in a few
+ * corner cases, such as COW of the block at slot zero in the node.  This
+ * ignores those rules, and it should only be called when there are no
+ * more updates to be done higher up in the tree.
+ */
+noinline void btrfs_unlock_up_safe(struct btrfs_path *path, int level)
+{
+	int i;
+
+	if (path->keep_locks || path->lowest_level)
+		return;
+
+	for (i = level; i < BTRFS_MAX_LEVEL; i++) {
+		if (!path->nodes[i])
+			continue;
+		if (!path->locks[i])
+			continue;
+		btrfs_tree_unlock(path->nodes[i]);
+		path->locks[i] = 0;
+	}
+}
+
 /*
  * look for key in the tree.  path is filled in with nodes along the way
  * if key is found, we return zero and you can find the item in the leaf
@@ -1387,31 +1507,30 @@ again:
 			int wret;
 
 			/* is a cow on this block not required */
-			spin_lock(&root->fs_info->hash_lock);
 			if (btrfs_header_generation(b) == trans->transid &&
 			    btrfs_header_owner(b) == root->root_key.objectid &&
 			    !btrfs_header_flag(b, BTRFS_HEADER_FLAG_WRITTEN)) {
-				spin_unlock(&root->fs_info->hash_lock);
 				goto cow_done;
 			}
-			spin_unlock(&root->fs_info->hash_lock);
 
 			/* ok, we have to cow, is our old prealloc the right
 			 * size?
 			 */
 			if (prealloc_block.objectid &&
 			    prealloc_block.offset != b->len) {
+				btrfs_release_path(root, p);
 				btrfs_free_reserved_extent(root,
 					   prealloc_block.objectid,
 					   prealloc_block.offset);
 				prealloc_block.objectid = 0;
+				goto again;
 			}
 
 			/*
 			 * for higher level blocks, try not to allocate blocks
 			 * with the block and the parent locks held.
 			 */
-			if (level > 1 && !prealloc_block.objectid &&
+			if (level > 0 && !prealloc_block.objectid &&
 			    btrfs_path_lock_waiting(p, level)) {
 				u32 size = b->len;
 				u64 hint = b->start;
@@ -1425,6 +1544,8 @@ again:
 				goto again;
 			}
 
+			btrfs_set_path_blocking(p);
+
 			wret = btrfs_cow_block(trans, root, b,
 					       p->nodes[level + 1],
 					       p->slots[level + 1],
@@ -1446,6 +1567,22 @@ cow_done:
 		if (!p->skip_locking)
 			p->locks[level] = 1;
 
+		btrfs_clear_path_blocking(p);
+
+		/*
+		 * we have a lock on b and as long as we aren't changing
+		 * the tree, there is no way to for the items in b to change.
+		 * It is safe to drop the lock on our parent before we
+		 * go through the expensive btree search on b.
+		 *
+		 * If cow is true, then we might be changing slot zero,
+		 * which may require changing the parent.  So, we can't
+		 * drop the lock until after we know which slot we're
+		 * operating on.
+		 */
+		if (!cow)
+			btrfs_unlock_up_safe(p, level + 1);
+
 		ret = check_block(root, p, level);
 		if (ret) {
 			ret = -1;
@@ -1453,6 +1590,7 @@ cow_done:
 		}
 
 		ret = bin_search(b, key, level, &slot);
+
 		if (level != 0) {
 			if (ret && slot > 0)
 				slot -= 1;
@@ -1460,7 +1598,16 @@ cow_done:
 			if ((p->search_for_split || ins_len > 0) &&
 			    btrfs_header_nritems(b) >=
 			    BTRFS_NODEPTRS_PER_BLOCK(root) - 3) {
-				int sret = split_node(trans, root, p, level);
+				int sret;
+
+				sret = reada_for_balance(root, p, level);
+				if (sret)
+					goto again;
+
+				btrfs_set_path_blocking(p);
+				sret = split_node(trans, root, p, level);
+				btrfs_clear_path_blocking(p);
+
 				BUG_ON(sret > 0);
 				if (sret) {
 					ret = sret;
@@ -1468,9 +1615,19 @@ cow_done:
 				}
 				b = p->nodes[level];
 				slot = p->slots[level];
-			} else if (ins_len < 0) {
-				int sret = balance_level(trans, root, p,
-							 level);
+			} else if (ins_len < 0 &&
+				   btrfs_header_nritems(b) <
+				   BTRFS_NODEPTRS_PER_BLOCK(root) / 4) {
+				int sret;
+
+				sret = reada_for_balance(root, p, level);
+				if (sret)
+					goto again;
+
+				btrfs_set_path_blocking(p);
+				sret = balance_level(trans, root, p, level);
+				btrfs_clear_path_blocking(p);
+
 				if (sret) {
 					ret = sret;
 					goto done;
@@ -1504,7 +1661,7 @@ cow_done:
 				 * of the btree by dropping locks before
 				 * we read.
 				 */
-				if (level > 1) {
+				if (level > 0) {
 					btrfs_release_path(NULL, p);
 					if (tmp)
 						free_extent_buffer(tmp);
@@ -1519,6 +1676,7 @@ cow_done:
 						free_extent_buffer(tmp);
 					goto again;
 				} else {
+					btrfs_set_path_blocking(p);
 					if (tmp)
 						free_extent_buffer(tmp);
 					if (should_reada)
@@ -1528,14 +1686,29 @@ cow_done:
 					b = read_node_slot(root, b, slot);
 				}
 			}
-			if (!p->skip_locking)
-				btrfs_tree_lock(b);
+			if (!p->skip_locking) {
+				int lret;
+
+				btrfs_clear_path_blocking(p);
+				lret = btrfs_try_spin_lock(b);
+
+				if (!lret) {
+					btrfs_set_path_blocking(p);
+					btrfs_tree_lock(b);
+					btrfs_clear_path_blocking(p);
+				}
+			}
 		} else {
 			p->slots[level] = slot;
 			if (ins_len > 0 &&
 			    btrfs_leaf_free_space(root, b) < ins_len) {
-				int sret = split_leaf(trans, root, key,
+				int sret;
+
+				btrfs_set_path_blocking(p);
+				sret = split_leaf(trans, root, key,
 						      p, ins_len, ret == 0);
+				btrfs_clear_path_blocking(p);
+
 				BUG_ON(sret > 0);
 				if (sret) {
 					ret = sret;
@@ -1549,12 +1722,16 @@ cow_done:
 	}
 	ret = 1;
 done:
+	/*
+	 * we don't really know what they plan on doing with the path
+	 * from here on, so for now just mark it as blocking
+	 */
+	btrfs_set_path_blocking(p);
 	if (prealloc_block.objectid) {
 		btrfs_free_reserved_extent(root,
 			   prealloc_block.objectid,
 			   prealloc_block.offset);
 	}
-
 	return ret;
 }
 
@@ -1578,6 +1755,8 @@ int btrfs_merge_path(struct btrfs_trans_handle *trans,
 	ret = btrfs_cow_block(trans, root, eb, NULL, 0, &eb, 0);
 	BUG_ON(ret);
 
+	btrfs_set_lock_blocking(eb);
+
 	parent = eb;
 	while (1) {
 		level = btrfs_header_level(parent);
@@ -1602,6 +1781,7 @@ int btrfs_merge_path(struct btrfs_trans_handle *trans,
 			eb = read_tree_block(root, bytenr, blocksize,
 					     generation);
 			btrfs_tree_lock(eb);
+			btrfs_set_lock_blocking(eb);
 		}
 
 		/*
@@ -1626,6 +1806,7 @@ int btrfs_merge_path(struct btrfs_trans_handle *trans,
 				eb = read_tree_block(root, bytenr, blocksize,
 						generation);
 				btrfs_tree_lock(eb);
+				btrfs_set_lock_blocking(eb);
 			}
 
 			ret = btrfs_cow_block(trans, root, eb, parent, slot,
@@ -2172,6 +2353,8 @@ static int push_leaf_right(struct btrfs_trans_handle *trans, struct btrfs_root
 
 	right = read_node_slot(root, upper, slot + 1);
 	btrfs_tree_lock(right);
+	btrfs_set_lock_blocking(right);
+
 	free_space = btrfs_leaf_free_space(root, right);
 	if (free_space < data_size)
 		goto out_unlock;
@@ -2367,6 +2550,8 @@ static int push_leaf_left(struct btrfs_trans_handle *trans, struct btrfs_root
 
 	left = read_node_slot(root, path->nodes[1], slot - 1);
 	btrfs_tree_lock(left);
+	btrfs_set_lock_blocking(left);
+
 	free_space = btrfs_leaf_free_space(root, left);
 	if (free_space < data_size) {
 		ret = 1;
@@ -2825,6 +3010,12 @@ int btrfs_split_item(struct btrfs_trans_handle *trans,
 	path->keep_locks = 0;
 	BUG_ON(ret);
 
+	/*
+	 * make sure any changes to the path from split_leaf leave it
+	 * in a blocking state
+	 */
+	btrfs_set_path_blocking(path);
+
 	leaf = path->nodes[0];
 	BUG_ON(btrfs_leaf_free_space(root, leaf) < sizeof(struct btrfs_item));
 
@@ -3354,6 +3545,7 @@ int btrfs_insert_empty_items(struct btrfs_trans_handle *trans,
 		BUG();
 	}
 out:
+	btrfs_unlock_up_safe(path, 1);
 	return ret;
 }
 
@@ -3441,15 +3633,22 @@ noinline int btrfs_del_leaf(struct btrfs_trans_handle *trans,
 {
 	int ret;
 	u64 root_gen = btrfs_header_generation(path->nodes[1]);
+	u64 parent_start = path->nodes[1]->start;
+	u64 parent_owner = btrfs_header_owner(path->nodes[1]);
 
 	ret = del_ptr(trans, root, path, 1, path->slots[1]);
 	if (ret)
 		return ret;
 
+	/*
+	 * btrfs_free_extent is expensive, we want to make sure we
+	 * aren't holding any locks when we call it
+	 */
+	btrfs_unlock_up_safe(path, 0);
+
 	ret = btrfs_free_extent(trans, root, bytenr,
 				btrfs_level_size(root, 0),
-				path->nodes[1]->start,
-				btrfs_header_owner(path->nodes[1]),
+				parent_start, parent_owner,
 				root_gen, 0, 1);
 	return ret;
 }
@@ -3721,12 +3920,14 @@ find_next_key:
 		 */
 		if (slot >= nritems) {
 			path->slots[level] = slot;
+			btrfs_set_path_blocking(path);
 			sret = btrfs_find_next_key(root, path, min_key, level,
 						  cache_only, min_trans);
 			if (sret == 0) {
 				btrfs_release_path(root, path);
 				goto again;
 			} else {
+				btrfs_clear_path_blocking(path);
 				goto out;
 			}
 		}
@@ -3738,16 +3939,20 @@ find_next_key:
 			unlock_up(path, level, 1);
 			goto out;
 		}
+		btrfs_set_path_blocking(path);
 		cur = read_node_slot(root, cur, slot);
 
 		btrfs_tree_lock(cur);
+
 		path->locks[level - 1] = 1;
 		path->nodes[level - 1] = cur;
 		unlock_up(path, level, 1);
+		btrfs_clear_path_blocking(path);
 	}
 out:
 	if (ret == 0)
 		memcpy(min_key, &found_key, sizeof(found_key));
+	btrfs_set_path_blocking(path);
 	return ret;
 }
 
@@ -3843,6 +4048,7 @@ int btrfs_next_leaf(struct btrfs_root *root, struct btrfs_path *path)
 	if (ret < 0)
 		return ret;
 
+	btrfs_set_path_blocking(path);
 	nritems = btrfs_header_nritems(path->nodes[0]);
 	/*
 	 * by releasing the path above we dropped all our locks.  A balance
@@ -3873,6 +4079,7 @@ int btrfs_next_leaf(struct btrfs_root *root, struct btrfs_path *path)
 			free_extent_buffer(next);
 		}
 
+		/* the path was set to blocking above */
 		if (level == 1 && (path->locks[1] || path->skip_locking) &&
 		    path->reada)
 			reada_for_search(root, path, level, slot, 0);
@@ -3881,6 +4088,7 @@ int btrfs_next_leaf(struct btrfs_root *root, struct btrfs_path *path)
 		if (!path->skip_locking) {
 			WARN_ON(!btrfs_tree_locked(c));
 			btrfs_tree_lock(next);
+			btrfs_set_lock_blocking(next);
 		}
 		break;
 	}
@@ -3897,12 +4105,15 @@ int btrfs_next_leaf(struct btrfs_root *root, struct btrfs_path *path)
 			path->locks[level] = 1;
 		if (!level)
 			break;
+
+		btrfs_set_path_blocking(path);
 		if (level == 1 && path->locks[1] && path->reada)
 			reada_for_search(root, path, level, slot, 0);
 		next = read_node_slot(root, next, 0);
 		if (!path->skip_locking) {
 			WARN_ON(!btrfs_tree_locked(path->nodes[level]));
 			btrfs_tree_lock(next);
+			btrfs_set_lock_blocking(next);
 		}
 	}
 done:
@@ -3927,6 +4138,7 @@ int btrfs_previous_item(struct btrfs_root *root,
 
 	while (1) {
 		if (path->slots[0] == 0) {
+			btrfs_set_path_blocking(path);
 			ret = btrfs_prev_leaf(root, path);
 			if (ret != 0)
 				return ret;

fs/btrfs/ctree.h

@@ -454,17 +454,11 @@ struct btrfs_timespec {
 	__le32 nsec;
 } __attribute__ ((__packed__));
 
-typedef enum {
+enum btrfs_compression_type {
 	BTRFS_COMPRESS_NONE = 0,
 	BTRFS_COMPRESS_ZLIB = 1,
 	BTRFS_COMPRESS_LAST = 2,
-} btrfs_compression_type;
-
-/* we don't understand any encryption methods right now */
-typedef enum {
-	BTRFS_ENCRYPTION_NONE = 0,
-	BTRFS_ENCRYPTION_LAST = 1,
-} btrfs_encryption_type;
+};
 
 struct btrfs_inode_item {
 	/* nfs style generation number */
@@ -701,9 +695,7 @@ struct btrfs_fs_info {
 	struct btrfs_transaction *running_transaction;
 	wait_queue_head_t transaction_throttle;
 	wait_queue_head_t transaction_wait;
-
 	wait_queue_head_t async_submit_wait;
-	wait_queue_head_t tree_log_wait;
 
 	struct btrfs_super_block super_copy;
 	struct btrfs_super_block super_for_commit;
@@ -711,7 +703,6 @@ struct btrfs_fs_info {
 	struct super_block *sb;
 	struct inode *btree_inode;
 	struct backing_dev_info bdi;
-	spinlock_t hash_lock;
 	struct mutex trans_mutex;
 	struct mutex tree_log_mutex;
 	struct mutex transaction_kthread_mutex;
@@ -730,10 +721,6 @@ struct btrfs_fs_info {
 	atomic_t async_submit_draining;
 	atomic_t nr_async_bios;
 	atomic_t async_delalloc_pages;
-	atomic_t tree_log_writers;
-	atomic_t tree_log_commit;
-	unsigned long tree_log_batch;
-	u64 tree_log_transid;
 
 	/*
 	 * this is used by the balancing code to wait for all the pending
@@ -833,7 +820,14 @@ struct btrfs_root {
 	struct kobject root_kobj;
 	struct completion kobj_unregister;
 	struct mutex objectid_mutex;
+
 	struct mutex log_mutex;
+	wait_queue_head_t log_writer_wait;
+	wait_queue_head_t log_commit_wait[2];
+	atomic_t log_writers;
+	atomic_t log_commit[2];
+	unsigned long log_transid;
+	unsigned long log_batch;
 
 	u64 objectid;
 	u64 last_trans;
@@ -1841,6 +1835,10 @@ void btrfs_release_path(struct btrfs_root *root, struct btrfs_path *p);
 struct btrfs_path *btrfs_alloc_path(void);
 void btrfs_free_path(struct btrfs_path *p);
 void btrfs_init_path(struct btrfs_path *p);
+void btrfs_set_path_blocking(struct btrfs_path *p);
+void btrfs_clear_path_blocking(struct btrfs_path *p);
+void btrfs_unlock_up_safe(struct btrfs_path *p, int level);
+
 int btrfs_del_items(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 		   struct btrfs_path *path, int slot, int nr);
 int btrfs_del_leaf(struct btrfs_trans_handle *trans,

fs/btrfs/disk-io.c

@@ -16,7 +16,6 @@
  * Boston, MA 021110-1307, USA.
  */
 
-#include <linux/version.h>
 #include <linux/fs.h>
 #include <linux/blkdev.h>
 #include <linux/scatterlist.h>
@@ -800,7 +799,7 @@ struct extent_buffer *read_tree_block(struct btrfs_root *root, u64 bytenr,
 	ret = btree_read_extent_buffer_pages(root, buf, 0, parent_transid);
 
 	if (ret == 0)
-		buf->flags |= EXTENT_UPTODATE;
+		set_bit(EXTENT_BUFFER_UPTODATE, &buf->bflags);
 	else
 		WARN_ON(1);
 	return buf;
@@ -814,6 +813,10 @@ int clean_tree_block(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 	if (btrfs_header_generation(buf) ==
 	    root->fs_info->running_transaction->transid) {
 		WARN_ON(!btrfs_tree_locked(buf));
+
+		/* ugh, clear_extent_buffer_dirty can be expensive */
+		btrfs_set_lock_blocking(buf);
+
 		clear_extent_buffer_dirty(&BTRFS_I(btree_inode)->io_tree,
 					  buf);
 	}
@@ -850,6 +853,14 @@ static int __setup_root(u32 nodesize, u32 leafsize, u32 sectorsize,
 	spin_lock_init(&root->list_lock);
 	mutex_init(&root->objectid_mutex);
 	mutex_init(&root->log_mutex);
+	init_waitqueue_head(&root->log_writer_wait);
+	init_waitqueue_head(&root->log_commit_wait[0]);
+	init_waitqueue_head(&root->log_commit_wait[1]);
+	atomic_set(&root->log_commit[0], 0);
+	atomic_set(&root->log_commit[1], 0);
+	atomic_set(&root->log_writers, 0);
+	root->log_batch = 0;
+	root->log_transid = 0;
 	extent_io_tree_init(&root->dirty_log_pages,
 			     fs_info->btree_inode->i_mapping, GFP_NOFS);
 
@@ -934,15 +945,16 @@ int btrfs_free_log_root_tree(struct btrfs_trans_handle *trans,
 	return 0;
 }
 
-int btrfs_init_log_root_tree(struct btrfs_trans_handle *trans,
-			     struct btrfs_fs_info *fs_info)
+static struct btrfs_root *alloc_log_tree(struct btrfs_trans_handle *trans,
+					 struct btrfs_fs_info *fs_info)
 {
 	struct btrfs_root *root;
 	struct btrfs_root *tree_root = fs_info->tree_root;
+	struct extent_buffer *leaf;
 
 	root = kzalloc(sizeof(*root), GFP_NOFS);
 	if (!root)
-		return -ENOMEM;
+		return ERR_PTR(-ENOMEM);
 
 	__setup_root(tree_root->nodesize, tree_root->leafsize,
 		     tree_root->sectorsize, tree_root->stripesize,
@@ -951,12 +963,23 @@ int btrfs_init_log_root_tree(struct btrfs_trans_handle *trans,
 	root->root_key.objectid = BTRFS_TREE_LOG_OBJECTID;
 	root->root_key.type = BTRFS_ROOT_ITEM_KEY;
 	root->root_key.offset = BTRFS_TREE_LOG_OBJECTID;
+	/*
+	 * log trees do not get reference counted because they go away
+	 * before a real commit is actually done.  They do store pointers
+	 * to file data extents, and those reference counts still get
+	 * updated (along with back refs to the log tree).
+	 */
 	root->ref_cows = 0;
 
-	root->node = btrfs_alloc_free_block(trans, root, root->leafsize,
-					    0, BTRFS_TREE_LOG_OBJECTID,
-					    trans->transid, 0, 0, 0);
+	leaf = btrfs_alloc_free_block(trans, root, root->leafsize,
+				      0, BTRFS_TREE_LOG_OBJECTID,
+				      trans->transid, 0, 0, 0);
+	if (IS_ERR(leaf)) {
+		kfree(root);
+		return ERR_CAST(leaf);
+	}
 
+	root->node = leaf;
 	btrfs_set_header_nritems(root->node, 0);
 	btrfs_set_header_level(root->node, 0);
 	btrfs_set_header_bytenr(root->node, root->node->start);
@@ -968,7 +991,48 @@ int btrfs_init_log_root_tree(struct btrfs_trans_handle *trans,
 			    BTRFS_FSID_SIZE);
 	btrfs_mark_buffer_dirty(root->node);
 	btrfs_tree_unlock(root->node);
-	fs_info->log_root_tree = root;
+	return root;
+}
+
+int btrfs_init_log_root_tree(struct btrfs_trans_handle *trans,
+			     struct btrfs_fs_info *fs_info)
+{
+	struct btrfs_root *log_root;
+
+	log_root = alloc_log_tree(trans, fs_info);
+	if (IS_ERR(log_root))
+		return PTR_ERR(log_root);
+	WARN_ON(fs_info->log_root_tree);
+	fs_info->log_root_tree = log_root;
+	return 0;
+}
+
+int btrfs_add_log_tree(struct btrfs_trans_handle *trans,
+		       struct btrfs_root *root)
+{
+	struct btrfs_root *log_root;
+	struct btrfs_inode_item *inode_item;
+
+	log_root = alloc_log_tree(trans, root->fs_info);
+	if (IS_ERR(log_root))
+		return PTR_ERR(log_root);
+
+	log_root->last_trans = trans->transid;
+	log_root->root_key.offset = root->root_key.objectid;
+
+	inode_item = &log_root->root_item.inode;
+	inode_item->generation = cpu_to_le64(1);
+	inode_item->size = cpu_to_le64(3);
+	inode_item->nlink = cpu_to_le32(1);
+	inode_item->nbytes = cpu_to_le64(root->leafsize);
+	inode_item->mode = cpu_to_le32(S_IFDIR | 0755);
+
+	btrfs_set_root_bytenr(&log_root->root_item, log_root->node->start);
+	btrfs_set_root_generation(&log_root->root_item, trans->transid);
+
+	WARN_ON(root->log_root);
+	root->log_root = log_root;
+	root->log_transid = 0;
 	return 0;
 }
 
@@ -1136,7 +1200,6 @@ static int btrfs_congested_fn(void *congested_data, int bdi_bits)
 {
 	struct btrfs_fs_info *info = (struct btrfs_fs_info *)congested_data;
 	int ret = 0;
-	struct list_head *cur;
 	struct btrfs_device *device;
 	struct backing_dev_info *bdi;
 #if 0
@@ -1144,8 +1207,7 @@ static int btrfs_congested_fn(void *congested_data, int bdi_bits)
 	    btrfs_congested_async(info, 0))
 		return 1;
 #endif
-	list_for_each(cur, &info->fs_devices->devices) {
-		device = list_entry(cur, struct btrfs_device, dev_list);
+	list_for_each_entry(device, &info->fs_devices->devices, dev_list) {
 		if (!device->bdev)
 			continue;
 		bdi = blk_get_backing_dev_info(device->bdev);
@@ -1163,13 +1225,11 @@ static int btrfs_congested_fn(void *congested_data, int bdi_bits)
  */
 static void __unplug_io_fn(struct backing_dev_info *bdi, struct page *page)
 {
-	struct list_head *cur;
 	struct btrfs_device *device;
 	struct btrfs_fs_info *info;
 
 	info = (struct btrfs_fs_info *)bdi->unplug_io_data;
-	list_for_each(cur, &info->fs_devices->devices) {
-		device = list_entry(cur, struct btrfs_device, dev_list);
+	list_for_each_entry(device, &info->fs_devices->devices, dev_list) {
 		if (!device->bdev)
 			continue;
 
@@ -1447,7 +1507,6 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 	INIT_LIST_HEAD(&fs_info->dead_roots);
 	INIT_LIST_HEAD(&fs_info->hashers);
 	INIT_LIST_HEAD(&fs_info->delalloc_inodes);
-	spin_lock_init(&fs_info->hash_lock);
 	spin_lock_init(&fs_info->delalloc_lock);
 	spin_lock_init(&fs_info->new_trans_lock);
 	spin_lock_init(&fs_info->ref_cache_lock);
@@ -1535,10 +1594,6 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 	init_waitqueue_head(&fs_info->transaction_throttle);
 	init_waitqueue_head(&fs_info->transaction_wait);
 	init_waitqueue_head(&fs_info->async_submit_wait);
-	init_waitqueue_head(&fs_info->tree_log_wait);
-	atomic_set(&fs_info->tree_log_commit, 0);
-	atomic_set(&fs_info->tree_log_writers, 0);
-	fs_info->tree_log_transid = 0;
 
 	__setup_root(4096, 4096, 4096, 4096, tree_root,
 		     fs_info, BTRFS_ROOT_TREE_OBJECTID);
@@ -1627,6 +1682,8 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 	 * low idle thresh
 	 */
 	fs_info->endio_workers.idle_thresh = 4;
+	fs_info->endio_meta_workers.idle_thresh = 4;
+
 	fs_info->endio_write_workers.idle_thresh = 64;
 	fs_info->endio_meta_write_workers.idle_thresh = 64;
 
@@ -1740,13 +1797,13 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 	fs_info->system_alloc_profile = fs_info->metadata_alloc_profile;
 	fs_info->cleaner_kthread = kthread_run(cleaner_kthread, tree_root,
 					       "btrfs-cleaner");
-	if (!fs_info->cleaner_kthread)
+	if (IS_ERR(fs_info->cleaner_kthread))
 		goto fail_csum_root;
 
 	fs_info->transaction_kthread = kthread_run(transaction_kthread,
 						   tree_root,
 						   "btrfs-transaction");
-	if (!fs_info->transaction_kthread)
+	if (IS_ERR(fs_info->transaction_kthread))
 		goto fail_cleaner;
 
 	if (btrfs_super_log_root(disk_super) != 0) {
@@ -1828,13 +1885,14 @@ fail_sb_buffer:
 fail_iput:
 	invalidate_inode_pages2(fs_info->btree_inode->i_mapping);
 	iput(fs_info->btree_inode);
-fail:
+
 	btrfs_close_devices(fs_info->fs_devices);
 	btrfs_mapping_tree_free(&fs_info->mapping_tree);
+	bdi_destroy(&fs_info->bdi);
 
+fail:
 	kfree(extent_root);
 	kfree(tree_root);
-	bdi_destroy(&fs_info->bdi);
 	kfree(fs_info);
 	kfree(chunk_root);
 	kfree(dev_root);
@@ -1995,7 +2053,6 @@ static int write_dev_supers(struct btrfs_device *device,
 
 int write_all_supers(struct btrfs_root *root, int max_mirrors)
 {
-	struct list_head *cur;
 	struct list_head *head = &root->fs_info->fs_devices->devices;
 	struct btrfs_device *dev;
 	struct btrfs_super_block *sb;
@@ -2011,8 +2068,7 @@ int write_all_supers(struct btrfs_root *root, int max_mirrors)
 
 	sb = &root->fs_info->super_for_commit;
 	dev_item = &sb->dev_item;
-	list_for_each(cur, head) {
-		dev = list_entry(cur, struct btrfs_device, dev_list);
+	list_for_each_entry(dev, head, dev_list) {
 		if (!dev->bdev) {
 			total_errors++;
 			continue;
@@ -2045,8 +2101,7 @@ int write_all_supers(struct btrfs_root *root, int max_mirrors)
 	}
 
 	total_errors = 0;
-	list_for_each(cur, head) {
-		dev = list_entry(cur, struct btrfs_device, dev_list);
+	list_for_each_entry(dev, head, dev_list) {
 		if (!dev->bdev)
 			continue;
 		if (!dev->in_fs_metadata || !dev->writeable)
@@ -2260,6 +2315,8 @@ void btrfs_mark_buffer_dirty(struct extent_buffer *buf)
 	u64 transid = btrfs_header_generation(buf);
 	struct inode *btree_inode = root->fs_info->btree_inode;
 
+	btrfs_set_lock_blocking(buf);
+
 	WARN_ON(!btrfs_tree_locked(buf));
 	if (transid != root->fs_info->generation) {
 		printk(KERN_CRIT "btrfs transid mismatch buffer %llu, "
@@ -2302,14 +2359,13 @@ int btrfs_read_buffer(struct extent_buffer *buf, u64 parent_transid)
 	int ret;
 	ret = btree_read_extent_buffer_pages(root, buf, 0, parent_transid);
 	if (ret == 0)
-		buf->flags |= EXTENT_UPTODATE;
+		set_bit(EXTENT_BUFFER_UPTODATE, &buf->bflags);
 	return ret;
 }
 
 int btree_lock_page_hook(struct page *page)
 {
 	struct inode *inode = page->mapping->host;
-	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
 	struct extent_buffer *eb;
 	unsigned long len;
@@ -2324,9 +2380,7 @@ int btree_lock_page_hook(struct page *page)
 		goto out;
 
 	btrfs_tree_lock(eb);
-	spin_lock(&root->fs_info->hash_lock);
 	btrfs_set_header_flag(eb, BTRFS_HEADER_FLAG_WRITTEN);
-	spin_unlock(&root->fs_info->hash_lock);
 	btrfs_tree_unlock(eb);
 	free_extent_buffer(eb);
 out:

fs/btrfs/disk-io.h

@@ -98,5 +98,7 @@ int btrfs_free_log_root_tree(struct btrfs_trans_handle *trans,
 			     struct btrfs_fs_info *fs_info);
 int btrfs_init_log_root_tree(struct btrfs_trans_handle *trans,
 			     struct btrfs_fs_info *fs_info);
+int btrfs_add_log_tree(struct btrfs_trans_handle *trans,
+		       struct btrfs_root *root);
 int btree_lock_page_hook(struct page *page);
 #endif

fs/btrfs/extent-tree.c

@@ -19,7 +19,7 @@
 #include <linux/pagemap.h>
 #include <linux/writeback.h>
 #include <linux/blkdev.h>
-#include <linux/version.h>
+#include <linux/sort.h>
 #include "compat.h"
 #include "hash.h"
 #include "crc32c.h"
@@ -30,7 +30,6 @@
 #include "volumes.h"
 #include "locking.h"
 #include "ref-cache.h"
-#include "compat.h"
 
 #define PENDING_EXTENT_INSERT 0
 #define PENDING_EXTENT_DELETE 1
@@ -326,10 +325,8 @@ static struct btrfs_space_info *__find_space_info(struct btrfs_fs_info *info,
 						  u64 flags)
 {
 	struct list_head *head = &info->space_info;
-	struct list_head *cur;
 	struct btrfs_space_info *found;
-	list_for_each(cur, head) {
-		found = list_entry(cur, struct btrfs_space_info, list);
+	list_for_each_entry(found, head, list) {
 		if (found->flags == flags)
 			return found;
 	}
@@ -1525,15 +1522,55 @@ out:
 	return ret;
 }
 
-int btrfs_inc_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root,
-		  struct extent_buffer *orig_buf, struct extent_buffer *buf,
-		  u32 *nr_extents)
+/* when a block goes through cow, we update the reference counts of
+ * everything that block points to.  The internal pointers of the block
+ * can be in just about any order, and it is likely to have clusters of
+ * things that are close together and clusters of things that are not.
+ *
+ * To help reduce the seeks that come with updating all of these reference
+ * counts, sort them by byte number before actual updates are done.
+ *
+ * struct refsort is used to match byte number to slot in the btree block.
+ * we sort based on the byte number and then use the slot to actually
+ * find the item.
+ *
+ * struct refsort is smaller than strcut btrfs_item and smaller than
+ * struct btrfs_key_ptr.  Since we're currently limited to the page size
+ * for a btree block, there's no way for a kmalloc of refsorts for a
+ * single node to be bigger than a page.
+ */
+struct refsort {
+	u64 bytenr;
+	u32 slot;
+};
+
+/*
+ * for passing into sort()
+ */
+static int refsort_cmp(const void *a_void, const void *b_void)
+{
+	const struct refsort *a = a_void;
+	const struct refsort *b = b_void;
+
+	if (a->bytenr < b->bytenr)
+		return -1;
+	if (a->bytenr > b->bytenr)
+		return 1;
+	return 0;
+}
+
+
+noinline int btrfs_inc_ref(struct btrfs_trans_handle *trans,
+			   struct btrfs_root *root,
+			   struct extent_buffer *orig_buf,
+			   struct extent_buffer *buf, u32 *nr_extents)
 {
 	u64 bytenr;
 	u64 ref_root;
 	u64 orig_root;
 	u64 ref_generation;
 	u64 orig_generation;
+	struct refsort *sorted;
 	u32 nritems;
 	u32 nr_file_extents = 0;
 	struct btrfs_key key;
@@ -1542,6 +1579,8 @@ int btrfs_inc_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 	int level;
 	int ret = 0;
 	int faili = 0;
+	int refi = 0;
+	int slot;
 	int (*process_func)(struct btrfs_trans_handle *, struct btrfs_root *,
 			    u64, u64, u64, u64, u64, u64, u64, u64);
 
@@ -1553,6 +1592,9 @@ int btrfs_inc_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 	nritems = btrfs_header_nritems(buf);
 	level = btrfs_header_level(buf);
 
+	sorted = kmalloc(sizeof(struct refsort) * nritems, GFP_NOFS);
+	BUG_ON(!sorted);
+
 	if (root->ref_cows) {
 		process_func = __btrfs_inc_extent_ref;
 	} else {
@@ -1565,6 +1607,11 @@ int btrfs_inc_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 		process_func = __btrfs_update_extent_ref;
 	}
 
+	/*
+	 * we make two passes through the items.  In the first pass we
+	 * only record the byte number and slot.  Then we sort based on
+	 * byte number and do the actual work based on the sorted results
+	 */
 	for (i = 0; i < nritems; i++) {
 		cond_resched();
 		if (level == 0) {
@@ -1581,6 +1628,32 @@ int btrfs_inc_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 				continue;
 
 			nr_file_extents++;
+			sorted[refi].bytenr = bytenr;
+			sorted[refi].slot = i;
+			refi++;
+		} else {
+			bytenr = btrfs_node_blockptr(buf, i);
+			sorted[refi].bytenr = bytenr;
+			sorted[refi].slot = i;
+			refi++;
+		}
+	}
+	/*
+	 * if refi == 0, we didn't actually put anything into the sorted
+	 * array and we're done
+	 */
+	if (refi == 0)
+		goto out;
+
+	sort(sorted, refi, sizeof(struct refsort), refsort_cmp, NULL);
+
+	for (i = 0; i < refi; i++) {
+		cond_resched();
+		slot = sorted[i].slot;
+		bytenr = sorted[i].bytenr;
+
+		if (level == 0) {
+			btrfs_item_key_to_cpu(buf, &key, slot);
 
 			ret = process_func(trans, root, bytenr,
 					   orig_buf->start, buf->start,
@@ -1589,25 +1662,25 @@ int btrfs_inc_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 					   key.objectid);
 
 			if (ret) {
-				faili = i;
+				faili = slot;
 				WARN_ON(1);
 				goto fail;
 			}
 		} else {
-			bytenr = btrfs_node_blockptr(buf, i);
 			ret = process_func(trans, root, bytenr,
 					   orig_buf->start, buf->start,
 					   orig_root, ref_root,
 					   orig_generation, ref_generation,
 					   level - 1);
 			if (ret) {
-				faili = i;
+				faili = slot;
 				WARN_ON(1);
 				goto fail;
 			}
 		}
 	}
 out:
+	kfree(sorted);
 	if (nr_extents) {
 		if (level == 0)
 			*nr_extents = nr_file_extents;
@@ -1616,6 +1689,7 @@ out:
 	}
 	return 0;
 fail:
+	kfree(sorted);
 	WARN_ON(1);
 	return ret;
 }
@@ -2159,7 +2233,8 @@ again:
 		ret = find_first_extent_bit(&info->extent_ins, search, &start,
 					    &end, EXTENT_WRITEBACK);
 		if (ret) {
-			if (skipped && all && !num_inserts) {
+			if (skipped && all && !num_inserts &&
+			    list_empty(&update_list)) {
 				skipped = 0;
 				search = 0;
 				continue;
@@ -2547,6 +2622,7 @@ again:
 		if (ret) {
 			if (all && skipped && !nr) {
 				search = 0;
+				skipped = 0;
 				continue;
 			}
 			mutex_unlock(&info->extent_ins_mutex);
@@ -2700,13 +2776,9 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
 	/* if metadata always pin */
 	if (owner_objectid < BTRFS_FIRST_FREE_OBJECTID) {
 		if (root->root_key.objectid == BTRFS_TREE_LOG_OBJECTID) {
-			struct btrfs_block_group_cache *cache;
-
-			/* btrfs_free_reserved_extent */
-			cache = btrfs_lookup_block_group(root->fs_info, bytenr);
-			BUG_ON(!cache);
-			btrfs_add_free_space(cache, bytenr, num_bytes);
-			put_block_group(cache);
+			mutex_lock(&root->fs_info->pinned_mutex);
+			btrfs_update_pinned_extents(root, bytenr, num_bytes, 1);
+			mutex_unlock(&root->fs_info->pinned_mutex);
 			update_reserved_extents(root, bytenr, num_bytes, 0);
 			return 0;
 		}
@@ -3014,7 +3086,6 @@ loop_check:
 static void dump_space_info(struct btrfs_space_info *info, u64 bytes)
 {
 	struct btrfs_block_group_cache *cache;
-	struct list_head *l;
 
 	printk(KERN_INFO "space_info has %llu free, is %sfull\n",
 	       (unsigned long long)(info->total_bytes - info->bytes_used -
@@ -3022,8 +3093,7 @@ static void dump_space_info(struct btrfs_space_info *info, u64 bytes)
 	       (info->full) ? "" : "not ");
 
 	down_read(&info->groups_sem);
-	list_for_each(l, &info->block_groups) {
-		cache = list_entry(l, struct btrfs_block_group_cache, list);
+	list_for_each_entry(cache, &info->block_groups, list) {
 		spin_lock(&cache->lock);
 		printk(KERN_INFO "block group %llu has %llu bytes, %llu used "
 		       "%llu pinned %llu reserved\n",
@@ -3342,7 +3412,10 @@ struct extent_buffer *btrfs_init_new_buffer(struct btrfs_trans_handle *trans,
 	btrfs_set_header_generation(buf, trans->transid);
 	btrfs_tree_lock(buf);
 	clean_tree_block(trans, root, buf);
+
+	btrfs_set_lock_blocking(buf);
 	btrfs_set_buffer_uptodate(buf);
+
 	if (root->root_key.objectid == BTRFS_TREE_LOG_OBJECTID) {
 		set_extent_dirty(&root->dirty_log_pages, buf->start,
 			 buf->start + buf->len - 1, GFP_NOFS);
@@ -3351,6 +3424,7 @@ struct extent_buffer *btrfs_init_new_buffer(struct btrfs_trans_handle *trans,
 			 buf->start + buf->len - 1, GFP_NOFS);
 	}
 	trans->blocks_used++;
+	/* this returns a buffer locked for blocking */
 	return buf;
 }
 
@@ -3388,36 +3462,73 @@ int btrfs_drop_leaf_ref(struct btrfs_trans_handle *trans,
 {
 	u64 leaf_owner;
 	u64 leaf_generation;
+	struct refsort *sorted;
 	struct btrfs_key key;
 	struct btrfs_file_extent_item *fi;
 	int i;
 	int nritems;
 	int ret;
+	int refi = 0;
+	int slot;
 
 	BUG_ON(!btrfs_is_leaf(leaf));
 	nritems = btrfs_header_nritems(leaf);
 	leaf_owner = btrfs_header_owner(leaf);
 	leaf_generation = btrfs_header_generation(leaf);
 
+	sorted = kmalloc(sizeof(*sorted) * nritems, GFP_NOFS);
+	/* we do this loop twice.  The first time we build a list
+	 * of the extents we have a reference on, then we sort the list
+	 * by bytenr.  The second time around we actually do the
+	 * extent freeing.
+	 */
 	for (i = 0; i < nritems; i++) {
 		u64 disk_bytenr;
 		cond_resched();
 
 		btrfs_item_key_to_cpu(leaf, &key, i);
+
+		/* only extents have references, skip everything else */
 		if (btrfs_key_type(&key) != BTRFS_EXTENT_DATA_KEY)
 			continue;
+
 		fi = btrfs_item_ptr(leaf, i, struct btrfs_file_extent_item);
+
+		/* inline extents live in the btree, they don't have refs */
 		if (btrfs_file_extent_type(leaf, fi) ==
 		    BTRFS_FILE_EXTENT_INLINE)
 			continue;
-		/*
-		 * FIXME make sure to insert a trans record that
-		 * repeats the snapshot del on crash
-		 */
+
 		disk_bytenr = btrfs_file_extent_disk_bytenr(leaf, fi);
+
+		/* holes don't have refs */
 		if (disk_bytenr == 0)
 			continue;
 
+		sorted[refi].bytenr = disk_bytenr;
+		sorted[refi].slot = i;
+		refi++;
+	}
+
+	if (refi == 0)
+		goto out;
+
+	sort(sorted, refi, sizeof(struct refsort), refsort_cmp, NULL);
+
+	for (i = 0; i < refi; i++) {
+		u64 disk_bytenr;
+
+		disk_bytenr = sorted[i].bytenr;
+		slot = sorted[i].slot;
+
+		cond_resched();
+
+		btrfs_item_key_to_cpu(leaf, &key, slot);
+		if (btrfs_key_type(&key) != BTRFS_EXTENT_DATA_KEY)
+			continue;
+
+		fi = btrfs_item_ptr(leaf, slot, struct btrfs_file_extent_item);
+
 		ret = __btrfs_free_extent(trans, root, disk_bytenr,
 				btrfs_file_extent_disk_num_bytes(leaf, fi),
 				leaf->start, leaf_owner, leaf_generation,
@@ -3428,6 +3539,8 @@ int btrfs_drop_leaf_ref(struct btrfs_trans_handle *trans,
 		wake_up(&root->fs_info->transaction_throttle);
 		cond_resched();
 	}
+out:
+	kfree(sorted);
 	return 0;
 }
 
@@ -3437,9 +3550,25 @@ static noinline int cache_drop_leaf_ref(struct btrfs_trans_handle *trans,
 {
 	int i;
 	int ret;
-	struct btrfs_extent_info *info = ref->extents;
+	struct btrfs_extent_info *info;
+	struct refsort *sorted;
+
+	if (ref->nritems == 0)
+		return 0;
 
+	sorted = kmalloc(sizeof(*sorted) * ref->nritems, GFP_NOFS);
 	for (i = 0; i < ref->nritems; i++) {
+		sorted[i].bytenr = ref->extents[i].bytenr;
+		sorted[i].slot = i;
+	}
+	sort(sorted, ref->nritems, sizeof(struct refsort), refsort_cmp, NULL);
+
+	/*
+	 * the items in the ref were sorted when the ref was inserted
+	 * into the ref cache, so this is already in order
+	 */
+	for (i = 0; i < ref->nritems; i++) {
+		info = ref->extents + sorted[i].slot;
 		ret = __btrfs_free_extent(trans, root, info->bytenr,
 					  info->num_bytes, ref->bytenr,
 					  ref->owner, ref->generation,
@@ -3453,6 +3582,7 @@ static noinline int cache_drop_leaf_ref(struct btrfs_trans_handle *trans,
 		info++;
 	}
 
+	kfree(sorted);
 	return 0;
 }
 
@@ -3496,6 +3626,152 @@ static int drop_snap_lookup_refcount(struct btrfs_root *root, u64 start,
 	return ret;
 }
 
+/*
+ * this is used while deleting old snapshots, and it drops the refs
+ * on a whole subtree starting from a level 1 node.
+ *
+ * The idea is to sort all the leaf pointers, and then drop the
+ * ref on all the leaves in order.  Most of the time the leaves
+ * will have ref cache entries, so no leaf IOs will be required to
+ * find the extents they have references on.
+ *
+ * For each leaf, any references it has are also dropped in order
+ *
+ * This ends up dropping the references in something close to optimal
+ * order for reading and modifying the extent allocation tree.
+ */
+static noinline int drop_level_one_refs(struct btrfs_trans_handle *trans,
+					struct btrfs_root *root,
+					struct btrfs_path *path)
+{
+	u64 bytenr;
+	u64 root_owner;
+	u64 root_gen;
+	struct extent_buffer *eb = path->nodes[1];
+	struct extent_buffer *leaf;
+	struct btrfs_leaf_ref *ref;
+	struct refsort *sorted = NULL;
+	int nritems = btrfs_header_nritems(eb);
+	int ret;
+	int i;
+	int refi = 0;
+	int slot = path->slots[1];
+	u32 blocksize = btrfs_level_size(root, 0);
+	u32 refs;
+
+	if (nritems == 0)
+		goto out;
+
+	root_owner = btrfs_header_owner(eb);
+	root_gen = btrfs_header_generation(eb);
+	sorted = kmalloc(sizeof(*sorted) * nritems, GFP_NOFS);
+
+	/*
+	 * step one, sort all the leaf pointers so we don't scribble
+	 * randomly into the extent allocation tree
+	 */
+	for (i = slot; i < nritems; i++) {
+		sorted[refi].bytenr = btrfs_node_blockptr(eb, i);
+		sorted[refi].slot = i;
+		refi++;
+	}
+
+	/*
+	 * nritems won't be zero, but if we're picking up drop_snapshot
+	 * after a crash, slot might be > 0, so double check things
+	 * just in case.
+	 */
+	if (refi == 0)
+		goto out;
+
+	sort(sorted, refi, sizeof(struct refsort), refsort_cmp, NULL);
+
+	/*
+	 * the first loop frees everything the leaves point to
+	 */
+	for (i = 0; i < refi; i++) {
+		u64 ptr_gen;
+
+		bytenr = sorted[i].bytenr;
+
+		/*
+		 * check the reference count on this leaf.  If it is > 1
+		 * we just decrement it below and don't update any
+		 * of the refs the leaf points to.
+		 */
+		ret = drop_snap_lookup_refcount(root, bytenr, blocksize, &refs);
+		BUG_ON(ret);
+		if (refs != 1)
+			continue;
+
+		ptr_gen = btrfs_node_ptr_generation(eb, sorted[i].slot);
+
+		/*
+		 * the leaf only had one reference, which means the
+		 * only thing pointing to this leaf is the snapshot
+		 * we're deleting.  It isn't possible for the reference
+		 * count to increase again later
+		 *
+		 * The reference cache is checked for the leaf,
+		 * and if found we'll be able to drop any refs held by
+		 * the leaf without needing to read it in.
+		 */
+		ref = btrfs_lookup_leaf_ref(root, bytenr);
+		if (ref && ref->generation != ptr_gen) {
+			btrfs_free_leaf_ref(root, ref);
+			ref = NULL;
+		}
+		if (ref) {
+			ret = cache_drop_leaf_ref(trans, root, ref);
+			BUG_ON(ret);
+			btrfs_remove_leaf_ref(root, ref);
+			btrfs_free_leaf_ref(root, ref);
+		} else {
+			/*
+			 * the leaf wasn't in the reference cache, so
+			 * we have to read it.
+			 */
+			leaf = read_tree_block(root, bytenr, blocksize,
+					       ptr_gen);
+			ret = btrfs_drop_leaf_ref(trans, root, leaf);
+			BUG_ON(ret);
+			free_extent_buffer(leaf);
+		}
+		atomic_inc(&root->fs_info->throttle_gen);
+		wake_up(&root->fs_info->transaction_throttle);
+		cond_resched();
+	}
+
+	/*
+	 * run through the loop again to free the refs on the leaves.
+	 * This is faster than doing it in the loop above because
+	 * the leaves are likely to be clustered together.  We end up
+	 * working in nice chunks on the extent allocation tree.
+	 */
+	for (i = 0; i < refi; i++) {
+		bytenr = sorted[i].bytenr;
+		ret = __btrfs_free_extent(trans, root, bytenr,
+					blocksize, eb->start,
+					root_owner, root_gen, 0, 1);
+		BUG_ON(ret);
+
+		atomic_inc(&root->fs_info->throttle_gen);
+		wake_up(&root->fs_info->transaction_throttle);
+		cond_resched();
+	}
+out:
+	kfree(sorted);
+
+	/*
+	 * update the path to show we've processed the entire level 1
+	 * node.  This will get saved into the root's drop_snapshot_progress
+	 * field so these drops are not repeated again if this transaction
+	 * commits.
+	 */
+	path->slots[1] = nritems;
+	return 0;
+}
+
 /*
  * helper function for drop_snapshot, this walks down the tree dropping ref
  * counts as it goes.
@@ -3511,7 +3787,6 @@ static noinline int walk_down_tree(struct btrfs_trans_handle *trans,
 	struct extent_buffer *next;
 	struct extent_buffer *cur;
 	struct extent_buffer *parent;
-	struct btrfs_leaf_ref *ref;
 	u32 blocksize;
 	int ret;
 	u32 refs;
@@ -3538,17 +3813,46 @@ static noinline int walk_down_tree(struct btrfs_trans_handle *trans,
 		if (path->slots[*level] >=
 		    btrfs_header_nritems(cur))
 			break;
+
+		/* the new code goes down to level 1 and does all the
+		 * leaves pointed to that node in bulk.  So, this check
+		 * for level 0 will always be false.
+		 *
+		 * But, the disk format allows the drop_snapshot_progress
+		 * field in the root to leave things in a state where
+		 * a leaf will need cleaning up here.  If someone crashes
+		 * with the old code and then boots with the new code,
+		 * we might find a leaf here.
+		 */
 		if (*level == 0) {
 			ret = btrfs_drop_leaf_ref(trans, root, cur);
 			BUG_ON(ret);
 			break;
 		}
+
+		/*
+		 * once we get to level one, process the whole node
+		 * at once, including everything below it.
+		 */
+		if (*level == 1) {
+			ret = drop_level_one_refs(trans, root, path);
+			BUG_ON(ret);
+			break;
+		}
+
 		bytenr = btrfs_node_blockptr(cur, path->slots[*level]);
 		ptr_gen = btrfs_node_ptr_generation(cur, path->slots[*level]);
 		blocksize = btrfs_level_size(root, *level - 1);
 
 		ret = drop_snap_lookup_refcount(root, bytenr, blocksize, &refs);
 		BUG_ON(ret);
+
+		/*
+		 * if there is more than one reference, we don't need
+		 * to read that node to drop any references it has.  We
+		 * just drop the ref we hold on that node and move on to the
+		 * next slot in this level.
+		 */
 		if (refs != 1) {
 			parent = path->nodes[*level];
 			root_owner = btrfs_header_owner(parent);
@@ -3567,46 +3871,12 @@ static noinline int walk_down_tree(struct btrfs_trans_handle *trans,
 
 			continue;
 		}
+
 		/*
-		 * at this point, we have a single ref, and since the
-		 * only place referencing this extent is a dead root
-		 * the reference count should never go higher.
-		 * So, we don't need to check it again
+		 * we need to keep freeing things in the next level down.
+		 * read the block and loop around to process it
 		 */
-		if (*level == 1) {
-			ref = btrfs_lookup_leaf_ref(root, bytenr);
-			if (ref && ref->generation != ptr_gen) {
-				btrfs_free_leaf_ref(root, ref);
-				ref = NULL;
-			}
-			if (ref) {
-				ret = cache_drop_leaf_ref(trans, root, ref);
-				BUG_ON(ret);
-				btrfs_remove_leaf_ref(root, ref);
-				btrfs_free_leaf_ref(root, ref);
-				*level = 0;
-				break;
-			}
-		}
-		next = btrfs_find_tree_block(root, bytenr, blocksize);
-		if (!next || !btrfs_buffer_uptodate(next, ptr_gen)) {
-			free_extent_buffer(next);
-
-			next = read_tree_block(root, bytenr, blocksize,
-					       ptr_gen);
-			cond_resched();
-#if 0
-			/*
-			 * this is a debugging check and can go away
-			 * the ref should never go all the way down to 1
-			 * at this point
-			 */
-			ret = lookup_extent_ref(NULL, root, bytenr, blocksize,
-						&refs);
-			BUG_ON(ret);
-			WARN_ON(refs != 1);
-#endif
-		}
+		next = read_tree_block(root, bytenr, blocksize, ptr_gen);
 		WARN_ON(*level <= 0);
 		if (path->nodes[*level-1])
 			free_extent_buffer(path->nodes[*level-1]);
@@ -3631,11 +3901,16 @@ out:
 	root_owner = btrfs_header_owner(parent);
 	root_gen = btrfs_header_generation(parent);
 
+	/*
+	 * cleanup and free the reference on the last node
+	 * we processed
+	 */
 	ret = __btrfs_free_extent(trans, root, bytenr, blocksize,
 				  parent->start, root_owner, root_gen,
 				  *level, 1);
 	free_extent_buffer(path->nodes[*level]);
 	path->nodes[*level] = NULL;
+
 	*level += 1;
 	BUG_ON(ret);
 
@@ -3687,6 +3962,7 @@ static noinline int walk_down_subtree(struct btrfs_trans_handle *trans,
 
 		next = read_tree_block(root, bytenr, blocksize, ptr_gen);
 		btrfs_tree_lock(next);
+		btrfs_set_lock_blocking(next);
 
 		ret = btrfs_lookup_extent_ref(trans, root, bytenr, blocksize,
 					      &refs);
@@ -3754,6 +4030,13 @@ static noinline int walk_up_tree(struct btrfs_trans_handle *trans,
 		if (slot < btrfs_header_nritems(path->nodes[i]) - 1) {
 			struct extent_buffer *node;
 			struct btrfs_disk_key disk_key;
+
+			/*
+			 * there is more work to do in this level.
+			 * Update the drop_progress marker to reflect
+			 * the work we've done so far, and then bump
+			 * the slot number
+			 */
 			node = path->nodes[i];
 			path->slots[i]++;
 			*level = i;
@@ -3765,6 +4048,11 @@ static noinline int walk_up_tree(struct btrfs_trans_handle *trans,
 			return 0;
 		} else {
 			struct extent_buffer *parent;
+
+			/*
+			 * this whole node is done, free our reference
+			 * on it and go up one level
+			 */
 			if (path->nodes[*level] == root->node)
 				parent = path->nodes[*level];
 			else
@@ -4444,7 +4732,7 @@ static noinline int replace_one_extent(struct btrfs_trans_handle *trans,
 	u64 lock_end = 0;
 	u64 num_bytes;
 	u64 ext_offset;
-	u64 first_pos;
+	u64 search_end = (u64)-1;
 	u32 nritems;
 	int nr_scaned = 0;
 	int extent_locked = 0;
@@ -4452,7 +4740,6 @@ static noinline int replace_one_extent(struct btrfs_trans_handle *trans,
 	int ret;
 
 	memcpy(&key, leaf_key, sizeof(key));
-	first_pos = INT_LIMIT(loff_t) - extent_key->offset;
 	if (ref_path->owner_objectid != BTRFS_MULTIPLE_OBJECTIDS) {
 		if (key.objectid < ref_path->owner_objectid ||
 		    (key.objectid == ref_path->owner_objectid &&
@@ -4501,7 +4788,7 @@ next:
 			if ((key.objectid > ref_path->owner_objectid) ||
 			    (key.objectid == ref_path->owner_objectid &&
 			     key.type > BTRFS_EXTENT_DATA_KEY) ||
-			    (key.offset >= first_pos + extent_key->offset))
+			    key.offset >= search_end)
 				break;
 		}
 
@@ -4534,8 +4821,10 @@ next:
 		num_bytes = btrfs_file_extent_num_bytes(leaf, fi);
 		ext_offset = btrfs_file_extent_offset(leaf, fi);
 
-		if (first_pos > key.offset - ext_offset)
-			first_pos = key.offset - ext_offset;
+		if (search_end == (u64)-1) {
+			search_end = key.offset - ext_offset +
+				btrfs_file_extent_ram_bytes(leaf, fi);
+		}
 
 		if (!extent_locked) {
 			lock_start = key.offset;
@@ -4724,7 +5013,7 @@ next:
 		}
 skip:
 		if (ref_path->owner_objectid != BTRFS_MULTIPLE_OBJECTIDS &&
-		    key.offset >= first_pos + extent_key->offset)
+		    key.offset >= search_end)
 			break;
 
 		cond_resched();
@@ -4778,6 +5067,7 @@ int btrfs_reloc_tree_cache_ref(struct btrfs_trans_handle *trans,
 		ref->bytenr = buf->start;
 		ref->owner = btrfs_header_owner(buf);
 		ref->generation = btrfs_header_generation(buf);
+
 		ret = btrfs_add_leaf_ref(root, ref, 0);
 		WARN_ON(ret);
 		btrfs_free_leaf_ref(root, ref);
@@ -5957,9 +6247,11 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
 	path = btrfs_alloc_path();
 	BUG_ON(!path);
 
-	btrfs_remove_free_space_cache(block_group);
+	spin_lock(&root->fs_info->block_group_cache_lock);
 	rb_erase(&block_group->cache_node,
 		 &root->fs_info->block_group_cache_tree);
+	spin_unlock(&root->fs_info->block_group_cache_lock);
+	btrfs_remove_free_space_cache(block_group);
 	down_write(&block_group->space_info->groups_sem);
 	list_del(&block_group->list);
 	up_write(&block_group->space_info->groups_sem);

fs/btrfs/extent_io.c

@@ -9,7 +9,6 @@
 #include <linux/spinlock.h>
 #include <linux/blkdev.h>
 #include <linux/swap.h>
-#include <linux/version.h>
 #include <linux/writeback.h>
 #include <linux/pagevec.h>
 #include "extent_io.h"
@@ -31,7 +30,7 @@ static LIST_HEAD(buffers);
 static LIST_HEAD(states);
 
 #define LEAK_DEBUG 0
-#ifdef LEAK_DEBUG
+#if LEAK_DEBUG
 static DEFINE_SPINLOCK(leak_lock);
 #endif
 
@@ -120,7 +119,7 @@ void extent_io_tree_init(struct extent_io_tree *tree,
 static struct extent_state *alloc_extent_state(gfp_t mask)
 {
 	struct extent_state *state;
-#ifdef LEAK_DEBUG
+#if LEAK_DEBUG
 	unsigned long flags;
 #endif
 
@@ -130,7 +129,7 @@ static struct extent_state *alloc_extent_state(gfp_t mask)
 	state->state = 0;
 	state->private = 0;
 	state->tree = NULL;
-#ifdef LEAK_DEBUG
+#if LEAK_DEBUG
 	spin_lock_irqsave(&leak_lock, flags);
 	list_add(&state->leak_list, &states);
 	spin_unlock_irqrestore(&leak_lock, flags);
@@ -145,11 +144,11 @@ static void free_extent_state(struct extent_state *state)
 	if (!state)
 		return;
 	if (atomic_dec_and_test(&state->refs)) {
-#ifdef LEAK_DEBUG
+#if LEAK_DEBUG
 		unsigned long flags;
 #endif
 		WARN_ON(state->tree);
-#ifdef LEAK_DEBUG
+#if LEAK_DEBUG
 		spin_lock_irqsave(&leak_lock, flags);
 		list_del(&state->leak_list);
 		spin_unlock_irqrestore(&leak_lock, flags);
@@ -2378,11 +2377,6 @@ static int extent_write_cache_pages(struct extent_io_tree *tree,
 	int scanned = 0;
 	int range_whole = 0;
 
-	if (wbc->nonblocking && bdi_write_congested(bdi)) {
-		wbc->encountered_congestion = 1;
-		return 0;
-	}
-
 	pagevec_init(&pvec, 0);
 	if (wbc->range_cyclic) {
 		index = mapping->writeback_index; /* Start from prev offset */
@@ -2855,6 +2849,98 @@ out:
 	return sector;
 }
 
+int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
+		__u64 start, __u64 len, get_extent_t *get_extent)
+{
+	int ret;
+	u64 off = start;
+	u64 max = start + len;
+	u32 flags = 0;
+	u64 disko = 0;
+	struct extent_map *em = NULL;
+	int end = 0;
+	u64 em_start = 0, em_len = 0;
+	unsigned long emflags;
+	ret = 0;
+
+	if (len == 0)
+		return -EINVAL;
+
+	lock_extent(&BTRFS_I(inode)->io_tree, start, start + len,
+		GFP_NOFS);
+	em = get_extent(inode, NULL, 0, off, max - off, 0);
+	if (!em)
+		goto out;
+	if (IS_ERR(em)) {
+		ret = PTR_ERR(em);
+		goto out;
+	}
+	while (!end) {
+		off = em->start + em->len;
+		if (off >= max)
+			end = 1;
+
+		em_start = em->start;
+		em_len = em->len;
+
+		disko = 0;
+		flags = 0;
+
+		switch (em->block_start) {
+		case EXTENT_MAP_LAST_BYTE:
+			end = 1;
+			flags |= FIEMAP_EXTENT_LAST;
+			break;
+		case EXTENT_MAP_HOLE:
+			flags |= FIEMAP_EXTENT_UNWRITTEN;
+			break;
+		case EXTENT_MAP_INLINE:
+			flags |= (FIEMAP_EXTENT_DATA_INLINE |
+				  FIEMAP_EXTENT_NOT_ALIGNED);
+			break;
+		case EXTENT_MAP_DELALLOC:
+			flags |= (FIEMAP_EXTENT_DELALLOC |
+				  FIEMAP_EXTENT_UNKNOWN);
+			break;
+		default:
+			disko = em->block_start;
+			break;
+		}
+		if (test_bit(EXTENT_FLAG_COMPRESSED, &em->flags))
+			flags |= FIEMAP_EXTENT_ENCODED;
+
+		emflags = em->flags;
+		free_extent_map(em);
+		em = NULL;
+
+		if (!end) {
+			em = get_extent(inode, NULL, 0, off, max - off, 0);
+			if (!em)
+				goto out;
+			if (IS_ERR(em)) {
+				ret = PTR_ERR(em);
+				goto out;
+			}
+			emflags = em->flags;
+		}
+		if (test_bit(EXTENT_FLAG_VACANCY, &emflags)) {
+			flags |= FIEMAP_EXTENT_LAST;
+			end = 1;
+		}
+
+		ret = fiemap_fill_next_extent(fieinfo, em_start, disko,
+					em_len, flags);
+		if (ret)
+			goto out_free;
+	}
+out_free:
+	free_extent_map(em);
+out:
+	unlock_extent(&BTRFS_I(inode)->io_tree, start, start + len,
+			GFP_NOFS);
+	return ret;
+}
+
 static inline struct page *extent_buffer_page(struct extent_buffer *eb,
 					      unsigned long i)
 {
@@ -2892,15 +2978,17 @@ static struct extent_buffer *__alloc_extent_buffer(struct extent_io_tree *tree,
 						   gfp_t mask)
 {
 	struct extent_buffer *eb = NULL;
-#ifdef LEAK_DEBUG
+#if LEAK_DEBUG
 	unsigned long flags;
 #endif
 
 	eb = kmem_cache_zalloc(extent_buffer_cache, mask);
 	eb->start = start;
 	eb->len = len;
-	mutex_init(&eb->mutex);
-#ifdef LEAK_DEBUG
+	spin_lock_init(&eb->lock);
+	init_waitqueue_head(&eb->lock_wq);
+
+#if LEAK_DEBUG
 	spin_lock_irqsave(&leak_lock, flags);
 	list_add(&eb->leak_list, &buffers);
 	spin_unlock_irqrestore(&leak_lock, flags);
@@ -2912,7 +3000,7 @@ static struct extent_buffer *__alloc_extent_buffer(struct extent_io_tree *tree,
 
 static void __free_extent_buffer(struct extent_buffer *eb)
 {
-#ifdef LEAK_DEBUG
+#if LEAK_DEBUG
 	unsigned long flags;
 	spin_lock_irqsave(&leak_lock, flags);
 	list_del(&eb->leak_list);
@@ -2980,8 +3068,7 @@ struct extent_buffer *alloc_extent_buffer(struct extent_io_tree *tree,
 		unlock_page(p);
 	}
 	if (uptodate)
-		eb->flags |= EXTENT_UPTODATE;
-	eb->flags |= EXTENT_BUFFER_FILLED;
+		set_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags);
 
 	spin_lock(&tree->buffer_lock);
 	exists = buffer_tree_insert(tree, start, &eb->rb_node);
@@ -3135,7 +3222,7 @@ int clear_extent_buffer_uptodate(struct extent_io_tree *tree,
 	unsigned long num_pages;
 
 	num_pages = num_extent_pages(eb->start, eb->len);
-	eb->flags &= ~EXTENT_UPTODATE;
+	clear_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags);
 
 	clear_extent_uptodate(tree, eb->start, eb->start + eb->len - 1,
 			      GFP_NOFS);
@@ -3206,7 +3293,7 @@ int extent_buffer_uptodate(struct extent_io_tree *tree,
 	struct page *page;
 	int pg_uptodate = 1;
 
-	if (eb->flags & EXTENT_UPTODATE)
+	if (test_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags))
 		return 1;
 
 	ret = test_range_bit(tree, eb->start, eb->start + eb->len - 1,
@@ -3242,7 +3329,7 @@ int read_extent_buffer_pages(struct extent_io_tree *tree,
 	struct bio *bio = NULL;
 	unsigned long bio_flags = 0;
 
-	if (eb->flags & EXTENT_UPTODATE)
+	if (test_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags))
 		return 0;
 
 	if (test_range_bit(tree, eb->start, eb->start + eb->len - 1,
@@ -3273,7 +3360,7 @@ int read_extent_buffer_pages(struct extent_io_tree *tree,
 	}
 	if (all_uptodate) {
 		if (start_i == 0)
-			eb->flags |= EXTENT_UPTODATE;
+			set_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags);
 		goto unlock_exit;
 	}
 
@@ -3309,7 +3396,7 @@ int read_extent_buffer_pages(struct extent_io_tree *tree,
 	}
 
 	if (!ret)
-		eb->flags |= EXTENT_UPTODATE;
+		set_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags);
 	return ret;
 
 unlock_exit:
@@ -3406,7 +3493,6 @@ int map_extent_buffer(struct extent_buffer *eb, unsigned long start,
 		unmap_extent_buffer(eb, eb->map_token, km);
 		eb->map_token = NULL;
 		save = 1;
-		WARN_ON(!mutex_is_locked(&eb->mutex));
 	}
 	err = map_private_extent_buffer(eb, start, min_len, token, map,
 				       map_start, map_len, km);