Merge branch 's3c-fixes' of git://aeryn.fluff.org.uk/bjdooks/linux

Documentation/ioctl/ioctl-number.txt

@@ -139,6 +139,7 @@ Code	Seq#	Include File		Comments
 'm'	all	linux/synclink.h	conflict!
 'm'	00-1F	net/irda/irmod.h	conflict!
 'n'	00-7F	linux/ncp_fs.h
+'n'	80-8F	linux/nilfs2_fs.h	NILFS2
 'n'	E0-FF	video/matrox.h          matroxfb
 'o'	00-1F	fs/ocfs2/ocfs2_fs.h	OCFS2
 'o'     00-03   include/mtd/ubi-user.h  conflict! (OCFS2 and UBI overlaps)

Documentation/kernel-parameters.txt

@@ -1115,6 +1115,10 @@ and is between 256 and 4096 characters. It is defined in the file
 			libata.dma=4	  Compact Flash DMA only 
 			Combinations also work, so libata.dma=3 enables DMA
 			for disks and CDROMs, but not CFs.
+	
+	libata.ignore_hpa=	[LIBATA] Ignore HPA limit
+			libata.ignore_hpa=0	  keep BIOS limits (default)
+			libata.ignore_hpa=1	  ignore limits, using full disk
 
 	libata.noacpi	[LIBATA] Disables use of ACPI in libata suspend/resume
 			when set.

arch/arm/plat-s3c24xx/clock-dclk.c

@@ -129,7 +129,7 @@ static int s3c24xx_clkout_setparent(struct clk *clk, struct clk *parent)
 
 	/* calculate the MISCCR setting for the clock */
 
-	if (parent == &clk_xtal)
+	if (parent == &clk_mpll)
 		source = S3C2410_MISCCR_CLK0_MPLL;
 	else if (parent == &clk_upll)
 		source = S3C2410_MISCCR_CLK0_UPLL;

arch/ia64/Makefile

@@ -41,11 +41,6 @@ $(error Sorry, you need a newer version of the assember, one that is built from
 		ftp://ftp.hpl.hp.com/pub/linux-ia64/gas-030124.tar.gz)
 endif
 
-ifeq ($(call cc-version),0304)
-	cflags-$(CONFIG_ITANIUM)	+= -mtune=merced
-	cflags-$(CONFIG_MCKINLEY)	+= -mtune=mckinley
-endif
-
 KBUILD_CFLAGS += $(cflags-y)
 head-y := arch/ia64/kernel/head.o arch/ia64/kernel/init_task.o
 

arch/ia64/include/asm/bitops.h

@@ -286,7 +286,7 @@ __test_and_clear_bit(int nr, volatile void * addr)
 {
 	__u32 *p = (__u32 *) addr + (nr >> 5);
 	__u32 m = 1 << (nr & 31);
-	int oldbitset = *p & m;
+	int oldbitset = (*p & m) != 0;
 
 	*p &= ~m;
 	return oldbitset;

arch/ia64/include/asm/pgtable.h

@@ -155,7 +155,6 @@
 #include <linux/bitops.h>
 #include <asm/cacheflush.h>
 #include <asm/mmu_context.h>
-#include <asm/processor.h>
 
 /*
  * Next come the mappings that determine how mmap() protection bits

arch/ia64/kernel/ia64_ksyms.c

@@ -21,6 +21,7 @@ EXPORT_SYMBOL(csum_ipv6_magic);
 
 #include <asm/page.h>
 EXPORT_SYMBOL(clear_page);
+EXPORT_SYMBOL(copy_page);
 
 #ifdef CONFIG_VIRTUAL_MEM_MAP
 #include <linux/bootmem.h>
@@ -60,9 +61,6 @@ EXPORT_SYMBOL(__udivdi3);
 EXPORT_SYMBOL(__moddi3);
 EXPORT_SYMBOL(__umoddi3);
 
-#include <asm/page.h>
-EXPORT_SYMBOL(copy_page);
-
 #if defined(CONFIG_MD_RAID456) || defined(CONFIG_MD_RAID456_MODULE)
 extern void xor_ia64_2(void);
 extern void xor_ia64_3(void);

arch/ia64/kernel/iosapic.c

@@ -1072,6 +1072,10 @@ iosapic_init (unsigned long phys_addr, unsigned int gsi_base)
 	}
 
 	addr = ioremap(phys_addr, 0);
+	if (addr == NULL) {
+		spin_unlock_irqrestore(&iosapic_lock, flags);
+		return -ENOMEM;
+	}
 	ver = iosapic_version(addr);
 	if ((err = iosapic_check_gsi_range(gsi_base, ver))) {
 		iounmap(addr);

arch/ia64/kernel/pci-dma.c

@@ -69,11 +69,6 @@ iommu_dma_init(void)
 
 int iommu_dma_supported(struct device *dev, u64 mask)
 {
-	struct dma_map_ops *ops = platform_dma_get_ops(dev);
-
-	if (ops->dma_supported)
-		return ops->dma_supported(dev, mask);
-
 	/* Copied from i386. Doesn't make much sense, because it will
 	   only work for pci_alloc_coherent.
 	   The caller just has to use GFP_DMA in this case. */

arch/ia64/kernel/topology.c

@@ -372,6 +372,10 @@ static int __cpuinit cache_add_dev(struct sys_device * sys_dev)
 	retval = kobject_init_and_add(&all_cpu_cache_info[cpu].kobj,
 				      &cache_ktype_percpu_entry, &sys_dev->kobj,
 				      "%s", "cache");
+	if (unlikely(retval < 0)) {
+		cpu_cache_sysfs_exit(cpu);
+		return retval;
+	}
 
 	for (i = 0; i < all_cpu_cache_info[cpu].num_cache_leaves; i++) {
 		this_object = LEAF_KOBJECT_PTR(cpu,i);
@@ -385,7 +389,7 @@ static int __cpuinit cache_add_dev(struct sys_device * sys_dev)
 			}
 			kobject_put(&all_cpu_cache_info[cpu].kobj);
 			cpu_cache_sysfs_exit(cpu);
-			break;
+			return retval;
 		}
 		kobject_uevent(&(this_object->kobj), KOBJ_ADD);
 	}

arch/mn10300/include/asm/pci.h

@@ -59,7 +59,6 @@ void pcibios_penalize_isa_irq(int irq);
 #include <linux/slab.h>
 #include <asm/scatterlist.h>
 #include <linux/string.h>
-#include <linux/mm.h>
 #include <asm/io.h>
 
 struct pci_dev;

arch/powerpc/kernel/dma.c

@@ -7,6 +7,7 @@
 
 #include <linux/device.h>
 #include <linux/dma-mapping.h>
+#include <linux/lmb.h>
 #include <asm/bug.h>
 #include <asm/abs_addr.h>
 
@@ -90,11 +91,10 @@ static void dma_direct_unmap_sg(struct device *dev, struct scatterlist *sg,
 static int dma_direct_dma_supported(struct device *dev, u64 mask)
 {
 #ifdef CONFIG_PPC64
-	/* Could be improved to check for memory though it better be
-	 * done via some global so platforms can set the limit in case
+	/* Could be improved so platforms can set the limit in case
 	 * they have limited DMA windows
 	 */
-	return mask >= DMA_BIT_MASK(32);
+	return mask >= (lmb_end_of_DRAM() - 1);
 #else
 	return 1;
 #endif

arch/powerpc/kernel/perf_counter.c

@@ -518,6 +518,8 @@ void hw_perf_disable(void)
 	struct cpu_hw_counters *cpuhw;
 	unsigned long flags;
 
+	if (!ppmu)
+		return;
 	local_irq_save(flags);
 	cpuhw = &__get_cpu_var(cpu_hw_counters);
 
@@ -572,6 +574,8 @@ void hw_perf_enable(void)
 	int n_lim;
 	int idx;
 
+	if (!ppmu)
+		return;
 	local_irq_save(flags);
 	cpuhw = &__get_cpu_var(cpu_hw_counters);
 	if (!cpuhw->disabled) {
@@ -737,6 +741,8 @@ int hw_perf_group_sched_in(struct perf_counter *group_leader,
 	long i, n, n0;
 	struct perf_counter *sub;
 
+	if (!ppmu)
+		return 0;
 	cpuhw = &__get_cpu_var(cpu_hw_counters);
 	n0 = cpuhw->n_counters;
 	n = collect_events(group_leader, ppmu->n_counter - n0,
@@ -1281,6 +1287,8 @@ void hw_perf_counter_setup(int cpu)
 {
 	struct cpu_hw_counters *cpuhw = &per_cpu(cpu_hw_counters, cpu);
 
+	if (!ppmu)
+		return;
 	memset(cpuhw, 0, sizeof(*cpuhw));
 	cpuhw->mmcr[0] = MMCR0_FC;
 }

arch/x86/kernel/tsc.c

@@ -275,15 +275,20 @@ static unsigned long pit_calibrate_tsc(u32 latch, unsigned long ms, int loopmin)
  * use the TSC value at the transitions to calculate a pretty
  * good value for the TSC frequencty.
  */
+static inline int pit_verify_msb(unsigned char val)
+{
+	/* Ignore LSB */
+	inb(0x42);
+	return inb(0x42) == val;
+}
+
 static inline int pit_expect_msb(unsigned char val, u64 *tscp, unsigned long *deltap)
 {
 	int count;
 	u64 tsc = 0;
 
 	for (count = 0; count < 50000; count++) {
-		/* Ignore LSB */
-		inb(0x42);
-		if (inb(0x42) != val)
+		if (!pit_verify_msb(val))
 			break;
 		tsc = get_cycles();
 	}
@@ -336,8 +341,7 @@ static unsigned long quick_pit_calibrate(void)
 	 * to do that is to just read back the 16-bit counter
 	 * once from the PIT.
 	 */
-	inb(0x42);
-	inb(0x42);
+	pit_verify_msb(0);
 
 	if (pit_expect_msb(0xff, &tsc, &d1)) {
 		for (i = 1; i <= MAX_QUICK_PIT_ITERATIONS; i++) {
@@ -348,8 +352,19 @@ static unsigned long quick_pit_calibrate(void)
 			 * Iterate until the error is less than 500 ppm
 			 */
 			delta -= tsc;
-			if (d1+d2 < delta >> 11)
-				goto success;
+			if (d1+d2 >= delta >> 11)
+				continue;
+
+			/*
+			 * Check the PIT one more time to verify that
+			 * all TSC reads were stable wrt the PIT.
+			 *
+			 * This also guarantees serialization of the
+			 * last cycle read ('d2') in pit_expect_msb.
+			 */
+			if (!pit_verify_msb(0xfe - i))
+				break;
+			goto success;
 		}
 	}
 	printk("Fast TSC calibration failed\n");

drivers/ata/ahci.c

@@ -219,6 +219,8 @@ enum {
 	AHCI_HFLAG_SECT255		= (1 << 8), /* max 255 sectors */
 	AHCI_HFLAG_YES_NCQ		= (1 << 9), /* force NCQ cap on */
 	AHCI_HFLAG_NO_SUSPEND		= (1 << 10), /* don't suspend */
+	AHCI_HFLAG_SRST_TOUT_IS_OFFLINE	= (1 << 11), /* treat SRST timeout as
+							link offline */
 
 	/* ap->flags bits */
 
@@ -1663,6 +1665,7 @@ static int ahci_do_softreset(struct ata_link *link, unsigned int *class,
 			     int (*check_ready)(struct ata_link *link))
 {
 	struct ata_port *ap = link->ap;
+	struct ahci_host_priv *hpriv = ap->host->private_data;
 	const char *reason = NULL;
 	unsigned long now, msecs;
 	struct ata_taskfile tf;
@@ -1701,12 +1704,21 @@ static int ahci_do_softreset(struct ata_link *link, unsigned int *class,
 
 	/* wait for link to become ready */
 	rc = ata_wait_after_reset(link, deadline, check_ready);
-	/* link occupied, -ENODEV too is an error */
-	if (rc) {
+	if (rc == -EBUSY && hpriv->flags & AHCI_HFLAG_SRST_TOUT_IS_OFFLINE) {
+		/*
+		 * Workaround for cases where link online status can't
+		 * be trusted.  Treat device readiness timeout as link
+		 * offline.
+		 */
+		ata_link_printk(link, KERN_INFO,
+				"device not ready, treating as offline\n");
+		*class = ATA_DEV_NONE;
+	} else if (rc) {
+		/* link occupied, -ENODEV too is an error */
 		reason = "device not ready";
 		goto fail;
-	}
-	*class = ahci_dev_classify(ap);
+	} else
+		*class = ahci_dev_classify(ap);
 
 	DPRINTK("EXIT, class=%u\n", *class);
 	return 0;
@@ -1773,7 +1785,8 @@ static int ahci_sb600_softreset(struct ata_link *link, unsigned int *class,
 		irq_sts = readl(port_mmio + PORT_IRQ_STAT);
 		if (irq_sts & PORT_IRQ_BAD_PMP) {
 			ata_link_printk(link, KERN_WARNING,
-					"failed due to HW bug, retry pmp=0\n");
+					"applying SB600 PMP SRST workaround "
+					"and retrying\n");
 			rc = ahci_do_softreset(link, class, 0, deadline,
 					       ahci_check_ready);
 		}
@@ -2726,6 +2739,56 @@ static bool ahci_broken_suspend(struct pci_dev *pdev)
 	return !ver || strcmp(ver, dmi->driver_data) < 0;
 }
 
+static bool ahci_broken_online(struct pci_dev *pdev)
+{
+#define ENCODE_BUSDEVFN(bus, slot, func)			\
+	(void *)(unsigned long)(((bus) << 8) | PCI_DEVFN((slot), (func)))
+	static const struct dmi_system_id sysids[] = {
+		/*
+		 * There are several gigabyte boards which use
+		 * SIMG5723s configured as hardware RAID.  Certain
+		 * 5723 firmware revisions shipped there keep the link
+		 * online but fail to answer properly to SRST or
+		 * IDENTIFY when no device is attached downstream
+		 * causing libata to retry quite a few times leading
+		 * to excessive detection delay.
+		 *
+		 * As these firmwares respond to the second reset try
+		 * with invalid device signature, considering unknown
+		 * sig as offline works around the problem acceptably.
+		 */
+		{
+			.ident = "EP45-DQ6",
+			.matches = {
+				DMI_MATCH(DMI_BOARD_VENDOR,
+					  "Gigabyte Technology Co., Ltd."),
+				DMI_MATCH(DMI_BOARD_NAME, "EP45-DQ6"),
+			},
+			.driver_data = ENCODE_BUSDEVFN(0x0a, 0x00, 0),
+		},
+		{
+			.ident = "EP45-DS5",
+			.matches = {
+				DMI_MATCH(DMI_BOARD_VENDOR,
+					  "Gigabyte Technology Co., Ltd."),
+				DMI_MATCH(DMI_BOARD_NAME, "EP45-DS5"),
+			},
+			.driver_data = ENCODE_BUSDEVFN(0x03, 0x00, 0),
+		},
+		{ }	/* terminate list */
+	};
+#undef ENCODE_BUSDEVFN
+	const struct dmi_system_id *dmi = dmi_first_match(sysids);
+	unsigned int val;
+
+	if (!dmi)
+		return false;
+
+	val = (unsigned long)dmi->driver_data;
+
+	return pdev->bus->number == (val >> 8) && pdev->devfn == (val & 0xff);
+}
+
 static int ahci_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 {
 	static int printed_version;
@@ -2841,6 +2904,12 @@ static int ahci_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 			   "BIOS update required for suspend/resume\n");
 	}
 
+	if (ahci_broken_online(pdev)) {
+		hpriv->flags |= AHCI_HFLAG_SRST_TOUT_IS_OFFLINE;
+		dev_info(&pdev->dev,
+			 "online status unreliable, applying workaround\n");
+	}
+
 	/* CAP.NP sometimes indicate the index of the last enabled
 	 * port, at other times, that of the last possible port, so
 	 * determining the maximum port number requires looking at

drivers/ata/libata-core.c

@@ -4302,6 +4302,9 @@ static const struct ata_blacklist_entry ata_device_blacklist [] = {
 	{ "WDC WD2500JD-00HBB0", "WD-WMAL71490727", ATA_HORKAGE_BROKEN_HPA },
 	{ "MAXTOR 6L080L4",	"A93.0500",	ATA_HORKAGE_BROKEN_HPA },
 
+	/* this one allows HPA unlocking but fails IOs on the area */
+	{ "OCZ-VERTEX",		    "1.30",	ATA_HORKAGE_BROKEN_HPA },
+
 	/* Devices which report 1 sector over size HPA */
 	{ "ST340823A",		NULL,		ATA_HORKAGE_HPA_SIZE, },
 	{ "ST320413A",		NULL,		ATA_HORKAGE_HPA_SIZE, },

drivers/ata/pata_at91.c

@@ -250,7 +250,7 @@ static int __devinit pata_at91_probe(struct platform_device *pdev)
 		ata_port_desc(ap, "no IRQ, using PIO polling");
 	}
 
-	info = kzalloc(sizeof(*info), GFP_KERNEL);
+	info = devm_kzalloc(dev, sizeof(*info), GFP_KERNEL);
 
 	if (!info) {
 		dev_err(dev, "failed to allocate memory for private data\n");
@@ -275,7 +275,7 @@ static int __devinit pata_at91_probe(struct platform_device *pdev)
 	if (!info->ide_addr) {
 		dev_err(dev, "failed to map IO base\n");
 		ret = -ENOMEM;
-		goto err_ide_ioremap;
+		goto err_put;
 	}
 
 	info->alt_addr = devm_ioremap(dev,
@@ -284,7 +284,7 @@ static int __devinit pata_at91_probe(struct platform_device *pdev)
 	if (!info->alt_addr) {
 		dev_err(dev, "failed to map CTL base\n");
 		ret = -ENOMEM;
-		goto err_alt_ioremap;
+		goto err_put;
 	}
 
 	ap->ioaddr.cmd_addr = info->ide_addr;
@@ -303,13 +303,8 @@ static int __devinit pata_at91_probe(struct platform_device *pdev)
 			irq ? ata_sff_interrupt : NULL,
 			irq_flags, &pata_at91_sht);
 
-err_alt_ioremap:
-	devm_iounmap(dev, info->ide_addr);
-
-err_ide_ioremap:
+err_put:
 	clk_put(info->mck);
-	kfree(info);
-
 	return ret;
 }
 
@@ -317,7 +312,6 @@ static int __devexit pata_at91_remove(struct platform_device *pdev)
 {
 	struct ata_host *host = dev_get_drvdata(&pdev->dev);
 	struct at91_ide_info *info;
-	struct device *dev = &pdev->dev;
 
 	if (!host)
 		return 0;
@@ -328,11 +322,8 @@ static int __devexit pata_at91_remove(struct platform_device *pdev)
 	if (!info)
 		return 0;
 
-	devm_iounmap(dev, info->ide_addr);
-	devm_iounmap(dev, info->alt_addr);
 	clk_put(info->mck);
 
-	kfree(info);
 	return 0;
 }
 

drivers/ata/pata_atiixp.c

 /*
  * pata_atiixp.c 	- ATI PATA for new ATA layer
  *			  (C) 2005 Red Hat Inc
+ *			  (C) 2009 Bartlomiej Zolnierkiewicz
  *
  * Based on
  *
@@ -61,20 +62,19 @@ static void atiixp_set_pio_timing(struct ata_port *ap, struct ata_device *adev,
 
 	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
 	int dn = 2 * ap->port_no + adev->devno;
-
-	/* Check this is correct - the order is odd in both drivers */
 	int timing_shift = (16 * ap->port_no) + 8 * (adev->devno ^ 1);
-	u16 pio_mode_data, pio_timing_data;
+	u32 pio_timing_data;
+	u16 pio_mode_data;
 
 	pci_read_config_word(pdev, ATIIXP_IDE_PIO_MODE, &pio_mode_data);
 	pio_mode_data &= ~(0x7 << (4 * dn));
 	pio_mode_data |= pio << (4 * dn);
 	pci_write_config_word(pdev, ATIIXP_IDE_PIO_MODE, pio_mode_data);
 
-	pci_read_config_word(pdev, ATIIXP_IDE_PIO_TIMING, &pio_timing_data);
+	pci_read_config_dword(pdev, ATIIXP_IDE_PIO_TIMING, &pio_timing_data);
 	pio_timing_data &= ~(0xFF << timing_shift);
 	pio_timing_data |= (pio_timings[pio] << timing_shift);
-	pci_write_config_word(pdev, ATIIXP_IDE_PIO_TIMING, pio_timing_data);
+	pci_write_config_dword(pdev, ATIIXP_IDE_PIO_TIMING, pio_timing_data);
 }
 
 /**
@@ -119,16 +119,17 @@ static void atiixp_set_dmamode(struct ata_port *ap, struct ata_device *adev)
 		udma_mode_data |= dma << (4 * dn);
 		pci_write_config_word(pdev, ATIIXP_IDE_UDMA_MODE, udma_mode_data);
 	} else {
-		u16 mwdma_timing_data;
-		/* Check this is correct - the order is odd in both drivers */
 		int timing_shift = (16 * ap->port_no) + 8 * (adev->devno ^ 1);
+		u32 mwdma_timing_data;
 
 		dma -= XFER_MW_DMA_0;
 
-		pci_read_config_word(pdev, ATIIXP_IDE_MWDMA_TIMING, &mwdma_timing_data);
+		pci_read_config_dword(pdev, ATIIXP_IDE_MWDMA_TIMING,
+				      &mwdma_timing_data);
 		mwdma_timing_data &= ~(0xFF << timing_shift);
 		mwdma_timing_data |= (mwdma_timings[dma] << timing_shift);
-		pci_write_config_word(pdev, ATIIXP_IDE_MWDMA_TIMING, mwdma_timing_data);
+		pci_write_config_dword(pdev, ATIIXP_IDE_MWDMA_TIMING,
+				       mwdma_timing_data);
 	}
 	/*
 	 *	We must now look at the PIO mode situation. We may need to

drivers/ata/sata_nv.c

@@ -602,6 +602,7 @@ MODULE_VERSION(DRV_VERSION);
 
 static int adma_enabled;
 static int swncq_enabled = 1;
+static int msi_enabled;
 
 static void nv_adma_register_mode(struct ata_port *ap)
 {
@@ -2459,6 +2460,11 @@ static int nv_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	} else if (type == SWNCQ)
 		nv_swncq_host_init(host);
 
+	if (msi_enabled) {
+		dev_printk(KERN_NOTICE, &pdev->dev, "Using MSI\n");
+		pci_enable_msi(pdev);
+	}
+
 	pci_set_master(pdev);
 	return ata_host_activate(host, pdev->irq, ipriv->irq_handler,
 				 IRQF_SHARED, ipriv->sht);
@@ -2558,4 +2564,6 @@ module_param_named(adma, adma_enabled, bool, 0444);
 MODULE_PARM_DESC(adma, "Enable use of ADMA (Default: false)");
 module_param_named(swncq, swncq_enabled, bool, 0444);
 MODULE_PARM_DESC(swncq, "Enable use of SWNCQ (Default: true)");
+module_param_named(msi, msi_enabled, bool, 0444);
+MODULE_PARM_DESC(msi, "Enable use of MSI (Default: false)");
 

drivers/base/platform.c

@@ -483,9 +483,6 @@ int platform_driver_register(struct platform_driver *drv)
 		drv->driver.remove = platform_drv_remove;
 	if (drv->shutdown)
 		drv->driver.shutdown = platform_drv_shutdown;
-	if (drv->suspend || drv->resume)
-		pr_warning("Platform driver '%s' needs updating - please use "
-			"dev_pm_ops\n", drv->driver.name);
 
 	return driver_register(&drv->driver);
 }

drivers/char/pty.c

@@ -144,6 +144,8 @@ static int pty_write(struct tty_struct *tty, const unsigned char *buf,
 
 static int pty_write_room(struct tty_struct *tty)
 {
+	if (tty->stopped)
+		return 0;
 	return pty_space(tty->link);
 }
 

drivers/pci/hotplug/sgi_hotplug.c

@@ -90,11 +90,10 @@ static struct hotplug_slot_ops sn_hotplug_slot_ops = {
 
 static DEFINE_MUTEX(sn_hotplug_mutex);
 
-static ssize_t path_show (struct hotplug_slot *bss_hotplug_slot,
-	       		  char *buf)
+static ssize_t path_show(struct pci_slot *pci_slot, char *buf)
 {
 	int retval = -ENOENT;
-	struct slot *slot = bss_hotplug_slot->private;
+	struct slot *slot = pci_slot->hotplug->private;
 
 	if (!slot)
 		return retval;
@@ -103,7 +102,7 @@ static ssize_t path_show (struct hotplug_slot *bss_hotplug_slot,
 	return retval;
 }
 
-static struct hotplug_slot_attribute sn_slot_path_attr = __ATTR_RO(path);
+static struct pci_slot_attribute sn_slot_path_attr = __ATTR_RO(path);
 
 static int sn_pci_slot_valid(struct pci_bus *pci_bus, int device)
 {

drivers/serial/Kconfig

@@ -527,7 +527,7 @@ config SERIAL_S3C24A0
 
 config SERIAL_S3C6400
 	tristate "Samsung S3C6400/S3C6410 Serial port support"
-	depends on SERIAL_SAMSUNG && (CPU_S3C600 || CPU_S3C6410)
+	depends on SERIAL_SAMSUNG && (CPU_S3C6400 || CPU_S3C6410)
 	default y
 	help
 	  Serial port support for the Samsung S3C6400 and S3C6410

fs/nfs/direct.c

@@ -255,7 +255,7 @@ static void nfs_direct_read_release(void *calldata)
 
 	if (put_dreq(dreq))
 		nfs_direct_complete(dreq);
-	nfs_readdata_release(calldata);
+	nfs_readdata_free(data);
 }
 
 static const struct rpc_call_ops nfs_read_direct_ops = {
@@ -314,14 +314,14 @@ static ssize_t nfs_direct_read_schedule_segment(struct nfs_direct_req *dreq,
 					data->npages, 1, 0, data->pagevec, NULL);
 		up_read(&current->mm->mmap_sem);
 		if (result < 0) {
-			nfs_readdata_release(data);
+			nfs_readdata_free(data);
 			break;
 		}
 		if ((unsigned)result < data->npages) {
 			bytes = result * PAGE_SIZE;
 			if (bytes <= pgbase) {
 				nfs_direct_release_pages(data->pagevec, result);
-				nfs_readdata_release(data);
+				nfs_readdata_free(data);
 				break;
 			}
 			bytes -= pgbase;
@@ -334,7 +334,7 @@ static ssize_t nfs_direct_read_schedule_segment(struct nfs_direct_req *dreq,
 		data->inode = inode;
 		data->cred = msg.rpc_cred;
 		data->args.fh = NFS_FH(inode);
-		data->args.context = get_nfs_open_context(ctx);
+		data->args.context = ctx;
 		data->args.offset = pos;
 		data->args.pgbase = pgbase;
 		data->args.pages = data->pagevec;
@@ -441,7 +441,7 @@ static void nfs_direct_free_writedata(struct nfs_direct_req *dreq)
 		struct nfs_write_data *data = list_entry(dreq->rewrite_list.next, struct nfs_write_data, pages);
 		list_del(&data->pages);
 		nfs_direct_release_pages(data->pagevec, data->npages);
-		nfs_writedata_release(data);
+		nfs_writedata_free(data);
 	}
 }
 
@@ -534,7 +534,7 @@ static void nfs_direct_commit_release(void *calldata)
 
 	dprintk("NFS: %5u commit returned %d\n", data->task.tk_pid, status);
 	nfs_direct_write_complete(dreq, data->inode);
-	nfs_commitdata_release(calldata);
+	nfs_commit_free(data);
 }
 
 static const struct rpc_call_ops nfs_commit_direct_ops = {
@@ -570,7 +570,7 @@ static void nfs_direct_commit_schedule(struct nfs_direct_req *dreq)
 	data->args.fh = NFS_FH(data->inode);
 	data->args.offset = 0;
 	data->args.count = 0;
-	data->args.context = get_nfs_open_context(dreq->ctx);
+	data->args.context = dreq->ctx;
 	data->res.count = 0;
 	data->res.fattr = &data->fattr;
 	data->res.verf = &data->verf;
@@ -734,14 +734,14 @@ static ssize_t nfs_direct_write_schedule_segment(struct nfs_direct_req *dreq,
 					data->npages, 0, 0, data->pagevec, NULL);
 		up_read(&current->mm->mmap_sem);
 		if (result < 0) {
-			nfs_writedata_release(data);
+			nfs_writedata_free(data);
 			break;
 		}
 		if ((unsigned)result < data->npages) {
 			bytes = result * PAGE_SIZE;
 			if (bytes <= pgbase) {
 				nfs_direct_release_pages(data->pagevec, result);
-				nfs_writedata_release(data);
+				nfs_writedata_free(data);
 				break;
 			}
 			bytes -= pgbase;
@@ -756,7 +756,7 @@ static ssize_t nfs_direct_write_schedule_segment(struct nfs_direct_req *dreq,
 		data->inode = inode;
 		data->cred = msg.rpc_cred;
 		data->args.fh = NFS_FH(inode);
-		data->args.context = get_nfs_open_context(ctx);
+		data->args.context = ctx;
 		data->args.offset = pos;
 		data->args.pgbase = pgbase;
 		data->args.pages = data->pagevec;

fs/nfs/read.c

@@ -60,17 +60,15 @@ struct nfs_read_data *nfs_readdata_alloc(unsigned int pagecount)
 	return p;
 }
 
-static void nfs_readdata_free(struct nfs_read_data *p)
+void nfs_readdata_free(struct nfs_read_data *p)
 {
 	if (p && (p->pagevec != &p->page_array[0]))
 		kfree(p->pagevec);
 	mempool_free(p, nfs_rdata_mempool);
 }
 
-void nfs_readdata_release(void *data)
+static void nfs_readdata_release(struct nfs_read_data *rdata)
 {
-	struct nfs_read_data *rdata = data;
-
 	put_nfs_open_context(rdata->args.context);
 	nfs_readdata_free(rdata);
 }

fs/nfs/write.c

@@ -87,17 +87,15 @@ struct nfs_write_data *nfs_writedata_alloc(unsigned int pagecount)
 	return p;
 }
 
-static void nfs_writedata_free(struct nfs_write_data *p)
+void nfs_writedata_free(struct nfs_write_data *p)
 {
 	if (p && (p->pagevec != &p->page_array[0]))
 		kfree(p->pagevec);
 	mempool_free(p, nfs_wdata_mempool);
 }
 
-void nfs_writedata_release(void *data)
+static void nfs_writedata_release(struct nfs_write_data *wdata)
 {
-	struct nfs_write_data *wdata = data;
-
 	put_nfs_open_context(wdata->args.context);
 	nfs_writedata_free(wdata);
 }

fs/proc/base.c

@@ -234,23 +234,20 @@ static int check_mem_permission(struct task_struct *task)
 
 struct mm_struct *mm_for_maps(struct task_struct *task)
 {
-	struct mm_struct *mm = get_task_mm(task);
-	if (!mm)
+	struct mm_struct *mm;
+
+	if (mutex_lock_killable(&task->cred_guard_mutex))
 		return NULL;
-	down_read(&mm->mmap_sem);
-	task_lock(task);
-	if (task->mm != mm)
-		goto out;
-	if (task->mm != current->mm &&
-	    __ptrace_may_access(task, PTRACE_MODE_READ) < 0)
-		goto out;
-	task_unlock(task);
+
+	mm = get_task_mm(task);
+	if (mm && mm != current->mm &&
+			!ptrace_may_access(task, PTRACE_MODE_READ)) {
+		mmput(mm);
+		mm = NULL;
+	}
+	mutex_unlock(&task->cred_guard_mutex);
+
 	return mm;
-out:
-	task_unlock(task);
-	up_read(&mm->mmap_sem);
-	mmput(mm);
-	return NULL;
 }
 
 static int proc_pid_cmdline(struct task_struct *task, char * buffer)

fs/proc/task_mmu.c

@@ -119,6 +119,7 @@ static void *m_start(struct seq_file *m, loff_t *pos)
 	mm = mm_for_maps(priv->task);
 	if (!mm)
 		return NULL;
+	down_read(&mm->mmap_sem);
 
 	tail_vma = get_gate_vma(priv->task);
 	priv->tail_vma = tail_vma;

fs/proc/task_nommu.c

@@ -189,6 +189,7 @@ static void *m_start(struct seq_file *m, loff_t *pos)
 		priv->task = NULL;
 		return NULL;
 	}
+	down_read(&mm->mmap_sem);
 
 	/* start from the Nth VMA */
 	for (p = rb_first(&mm->mm_rb); p; p = rb_next(p))

fs/xfs/linux-2.6/xfs_buf.c

@@ -770,7 +770,7 @@ xfs_buf_associate_memory(
 	bp->b_pages = NULL;
 	bp->b_addr = mem;
 
-	rval = _xfs_buf_get_pages(bp, page_count, 0);
+	rval = _xfs_buf_get_pages(bp, page_count, XBF_DONT_BLOCK);
 	if (rval)
 		return rval;
 

fs/xfs/xfs_attr.c

@@ -2010,7 +2010,9 @@ xfs_attr_rmtval_get(xfs_da_args_t *args)
 			dblkno = XFS_FSB_TO_DADDR(mp, map[i].br_startblock);
 			blkcnt = XFS_FSB_TO_BB(mp, map[i].br_blockcount);
 			error = xfs_read_buf(mp, mp->m_ddev_targp, dblkno,
-					     blkcnt, XFS_BUF_LOCK, &bp);
+					     blkcnt,
+					     XFS_BUF_LOCK | XBF_DONT_BLOCK,
+					     &bp);
 			if (error)
 				return(error);
 
@@ -2141,8 +2143,8 @@ xfs_attr_rmtval_set(xfs_da_args_t *args)
 		dblkno = XFS_FSB_TO_DADDR(mp, map.br_startblock),
 		blkcnt = XFS_FSB_TO_BB(mp, map.br_blockcount);
 
-		bp = xfs_buf_get_flags(mp->m_ddev_targp, dblkno,
-							blkcnt, XFS_BUF_LOCK);
+		bp = xfs_buf_get_flags(mp->m_ddev_targp, dblkno, blkcnt,
+				       XFS_BUF_LOCK | XBF_DONT_BLOCK);
 		ASSERT(bp);
 		ASSERT(!XFS_BUF_GETERROR(bp));
 

fs/xfs/xfs_bmap.c

@@ -6009,7 +6009,7 @@ xfs_getbmap(
 	 */
 	error = ENOMEM;
 	subnex = 16;
-	map = kmem_alloc(subnex * sizeof(*map), KM_MAYFAIL);
+	map = kmem_alloc(subnex * sizeof(*map), KM_MAYFAIL | KM_NOFS);
 	if (!map)
 		goto out_unlock_ilock;
 

fs/xfs/xfs_btree.c

@@ -120,8 +120,8 @@ xfs_btree_check_sblock(
 			XFS_RANDOM_BTREE_CHECK_SBLOCK))) {
 		if (bp)
 			xfs_buftrace("SBTREE ERROR", bp);
-		XFS_ERROR_REPORT("xfs_btree_check_sblock", XFS_ERRLEVEL_LOW,
-				 cur->bc_mp);
+		XFS_CORRUPTION_ERROR("xfs_btree_check_sblock",
+			XFS_ERRLEVEL_LOW, cur->bc_mp, block);
 		return XFS_ERROR(EFSCORRUPTED);
 	}
 	return 0;

fs/xfs/xfs_da_btree.c

@@ -2201,7 +2201,7 @@ kmem_zone_t *xfs_dabuf_zone;		/* dabuf zone */
 xfs_da_state_t *
 xfs_da_state_alloc(void)
 {
-	return kmem_zone_zalloc(xfs_da_state_zone, KM_SLEEP);
+	return kmem_zone_zalloc(xfs_da_state_zone, KM_NOFS);
 }
 
 /*
@@ -2261,9 +2261,9 @@ xfs_da_buf_make(int nbuf, xfs_buf_t **bps, inst_t *ra)
 	int		off;
 
 	if (nbuf == 1)
-		dabuf = kmem_zone_alloc(xfs_dabuf_zone, KM_SLEEP);
+		dabuf = kmem_zone_alloc(xfs_dabuf_zone, KM_NOFS);
 	else
-		dabuf = kmem_alloc(XFS_DA_BUF_SIZE(nbuf), KM_SLEEP);
+		dabuf = kmem_alloc(XFS_DA_BUF_SIZE(nbuf), KM_NOFS);
 	dabuf->dirty = 0;
 #ifdef XFS_DABUF_DEBUG
 	dabuf->ra = ra;

fs/xfs/xfs_dir2.c

@@ -256,7 +256,7 @@ xfs_dir_cilookup_result(
 					!(args->op_flags & XFS_DA_OP_CILOOKUP))
 		return EEXIST;
 
-	args->value = kmem_alloc(len, KM_MAYFAIL);
+	args->value = kmem_alloc(len, KM_NOFS | KM_MAYFAIL);
 	if (!args->value)
 		return ENOMEM;
 

fs/xfs/xfs_fsops.c

@@ -167,17 +167,25 @@ xfs_growfs_data_private(
 	new = nb - mp->m_sb.sb_dblocks;
 	oagcount = mp->m_sb.sb_agcount;
 	if (nagcount > oagcount) {
+		void *new_perag, *old_perag;
+
 		xfs_filestream_flush(mp);
+
+		new_perag = kmem_zalloc(sizeof(xfs_perag_t) * nagcount,
+					KM_MAYFAIL);
+		if (!new_perag)
+			return XFS_ERROR(ENOMEM);
+
 		down_write(&mp->m_peraglock);
-		mp->m_perag = kmem_realloc(mp->m_perag,
-			sizeof(xfs_perag_t) * nagcount,
-			sizeof(xfs_perag_t) * oagcount,
-			KM_SLEEP);
-		memset(&mp->m_perag[oagcount], 0,
-			(nagcount - oagcount) * sizeof(xfs_perag_t));
+		memcpy(new_perag, mp->m_perag, sizeof(xfs_perag_t) * oagcount);
+		old_perag = mp->m_perag;
+		mp->m_perag = new_perag;
+
 		mp->m_flags |= XFS_MOUNT_32BITINODES;
 		nagimax = xfs_initialize_perag(mp, nagcount);
 		up_write(&mp->m_peraglock);
+
+		kmem_free(old_perag);
 	}
 	tp = xfs_trans_alloc(mp, XFS_TRANS_GROWFS);
 	tp->t_flags |= XFS_TRANS_RESERVE;

fs/xfs/xfs_inode.c

@@ -343,6 +343,16 @@ xfs_iformat(
 		return XFS_ERROR(EFSCORRUPTED);
 	}
 
+	if (unlikely((ip->i_d.di_flags & XFS_DIFLAG_REALTIME) &&
+		     !ip->i_mount->m_rtdev_targp)) {
+		xfs_fs_repair_cmn_err(CE_WARN, ip->i_mount,
+			"corrupt dinode %Lu, has realtime flag set.",
+			ip->i_ino);
+		XFS_CORRUPTION_ERROR("xfs_iformat(realtime)",
+				     XFS_ERRLEVEL_LOW, ip->i_mount, dip);
+		return XFS_ERROR(EFSCORRUPTED);
+	}
+
 	switch (ip->i_d.di_mode & S_IFMT) {
 	case S_IFIFO:
 	case S_IFCHR:

fs/xfs/xfs_log.c

@@ -3180,7 +3180,7 @@ xlog_state_sync(xlog_t	  *log,
 STATIC void
 xlog_state_want_sync(xlog_t *log, xlog_in_core_t *iclog)
 {
-	ASSERT(spin_is_locked(&log->l_icloglock));
+	assert_spin_locked(&log->l_icloglock);
 
 	if (iclog->ic_state == XLOG_STATE_ACTIVE) {
 		xlog_state_switch_iclogs(log, iclog, 0);

fs/xfs/xfs_vnodeops.c

@@ -538,7 +538,9 @@ xfs_readlink_bmap(
 		d = XFS_FSB_TO_DADDR(mp, mval[n].br_startblock);
 		byte_cnt = XFS_FSB_TO_B(mp, mval[n].br_blockcount);
 
-		bp = xfs_buf_read(mp->m_ddev_targp, d, BTOBB(byte_cnt), 0);
+		bp = xfs_buf_read_flags(mp->m_ddev_targp, d, BTOBB(byte_cnt),
+					XBF_LOCK | XBF_MAPPED |
+					XBF_DONT_BLOCK);
 		error = XFS_BUF_GETERROR(bp);
 		if (error) {
 			xfs_ioerror_alert("xfs_readlink",

include/linux/nfs_fs.h

@@ -473,7 +473,6 @@ extern int  nfs_writepages(struct address_space *, struct writeback_control *);
 extern int  nfs_flush_incompatible(struct file *file, struct page *page);
 extern int  nfs_updatepage(struct file *, struct page *, unsigned int, unsigned int);
 extern int nfs_writeback_done(struct rpc_task *, struct nfs_write_data *);
-extern void nfs_writedata_release(void *);
 
 /*
  * Try to write back everything synchronously (but check the
@@ -488,7 +487,6 @@ extern int nfs_wb_page_cancel(struct inode *inode, struct page* page);
 extern int  nfs_commit_inode(struct inode *, int);
 extern struct nfs_write_data *nfs_commitdata_alloc(void);
 extern void nfs_commit_free(struct nfs_write_data *wdata);
-extern void nfs_commitdata_release(void *wdata);
 #else
 static inline int
 nfs_commit_inode(struct inode *inode, int how)
@@ -507,6 +505,7 @@ nfs_have_writebacks(struct inode *inode)
  * Allocate nfs_write_data structures
  */
 extern struct nfs_write_data *nfs_writedata_alloc(unsigned int npages);
+extern void nfs_writedata_free(struct nfs_write_data *);
 
 /*
  * linux/fs/nfs/read.c
@@ -515,7 +514,6 @@ extern int  nfs_readpage(struct file *, struct page *);
 extern int  nfs_readpages(struct file *, struct address_space *,
 		struct list_head *, unsigned);
 extern int  nfs_readpage_result(struct rpc_task *, struct nfs_read_data *);
-extern void nfs_readdata_release(void *data);
 extern int  nfs_readpage_async(struct nfs_open_context *, struct inode *,
 			       struct page *);
 
@@ -523,6 +521,7 @@ extern int  nfs_readpage_async(struct nfs_open_context *, struct inode *,
  * Allocate nfs_read_data structures
  */
 extern struct nfs_read_data *nfs_readdata_alloc(unsigned int npages);
+extern void nfs_readdata_free(struct nfs_read_data *);
 
 /*
  * linux/fs/nfs3proc.c

include/linux/perf_counter.h

@@ -121,7 +121,7 @@ enum perf_counter_sample_format {
 	PERF_SAMPLE_CPU				= 1U << 7,
 	PERF_SAMPLE_PERIOD			= 1U << 8,
 	PERF_SAMPLE_STREAM_ID			= 1U << 9,
-	PERF_SAMPLE_TP_RECORD			= 1U << 10,
+	PERF_SAMPLE_RAW				= 1U << 10,
 
 	PERF_SAMPLE_MAX = 1U << 11,		/* non-ABI */
 };
@@ -369,6 +369,8 @@ enum perf_event_type {
 	 *
 	 *	{ u64			nr,
 	 *	  u64			ips[nr];  } && PERF_SAMPLE_CALLCHAIN
+	 *	{ u32			size;
+	 *	  char                  data[size];}&& PERF_SAMPLE_RAW
 	 * };
 	 */
 	PERF_EVENT_SAMPLE		= 9,
@@ -414,9 +416,9 @@ struct perf_callchain_entry {
 	__u64				ip[PERF_MAX_STACK_DEPTH];
 };
 
-struct perf_tracepoint_record {
-	int				size;
-	char				*record;
+struct perf_raw_record {
+	u32				size;
+	void				*data;
 };
 
 struct task_struct;
@@ -687,7 +689,7 @@ struct perf_sample_data {
 	struct pt_regs			*regs;
 	u64				addr;
 	u64				period;
-	void				*private;
+	struct perf_raw_record		*raw;
 };
 
 extern int perf_counter_overflow(struct perf_counter *counter, int nmi,

include/trace/ftrace.h

@@ -637,12 +637,20 @@ __attribute__((section("_ftrace_events"))) event_##call = {		\
  *	pc = preempt_count();
  *
  *	__data_size = ftrace_get_offsets_<call>(&__data_offsets, args);
- *	__entry_size = __data_size + sizeof(*entry);
+ *
+ *	// Below we want to get the aligned size by taking into account
+ *	// the u32 field that will later store the buffer size
+ *	__entry_size = ALIGN(__data_size + sizeof(*entry) + sizeof(u32),
+ *			     sizeof(u64));
+ *	__entry_size -= sizeof(u32);
  *
  *	do {
  *		char raw_data[__entry_size]; <- allocate our sample in the stack
  *		struct trace_entry *ent;
  *
+ *		zero dead bytes from alignment to avoid stack leak to userspace:
+ *
+ *		*(u64 *)(&raw_data[__entry_size - sizeof(u64)]) = 0ULL;
  *		entry = (struct ftrace_raw_<call> *)raw_data;
  *		ent = &entry->ent;
  *		tracing_generic_entry_update(ent, irq_flags, pc);
@@ -685,12 +693,15 @@ static void ftrace_profile_##call(proto)				\
 	pc = preempt_count();						\
 									\
 	__data_size = ftrace_get_offsets_##call(&__data_offsets, args); \
-	__entry_size = ALIGN(__data_size + sizeof(*entry), sizeof(u64));\
+	__entry_size = ALIGN(__data_size + sizeof(*entry) + sizeof(u32),\
+			     sizeof(u64));				\
+	__entry_size -= sizeof(u32);					\
 									\
 	do {								\
 		char raw_data[__entry_size];				\
 		struct trace_entry *ent;				\
 									\
+		*(u64 *)(&raw_data[__entry_size - sizeof(u64)]) = 0ULL;	\
 		entry = (struct ftrace_raw_##call *)raw_data;		\
 		ent = &entry->ent;					\
 		tracing_generic_entry_update(ent, irq_flags, pc);	\

kernel/irq/numa_migrate.c

@@ -107,8 +107,8 @@ static struct irq_desc *__real_move_irq_desc(struct irq_desc *old_desc,
 
 struct irq_desc *move_irq_desc(struct irq_desc *desc, int node)
 {
-	/* those all static, do move them */
-	if (desc->irq < NR_IRQS_LEGACY)
+	/* those static or target node is -1, do not move them */
+	if (desc->irq < NR_IRQS_LEGACY || node == -1)
 		return desc;
 
 	if (desc->node != node)

kernel/perf_counter.c

@@ -2646,7 +2646,6 @@ static void perf_counter_output(struct perf_counter *counter, int nmi,
 		u64 counter;
 	} group_entry;
 	struct perf_callchain_entry *callchain = NULL;
-	struct perf_tracepoint_record *tp;
 	int callchain_size = 0;
 	u64 time;
 	struct {
@@ -2715,9 +2714,16 @@ static void perf_counter_output(struct perf_counter *counter, int nmi,
 			header.size += sizeof(u64);
 	}
 
-	if (sample_type & PERF_SAMPLE_TP_RECORD) {
-		tp = data->private;
-		header.size += tp->size;
+	if (sample_type & PERF_SAMPLE_RAW) {
+		int size = sizeof(u32);
+
+		if (data->raw)
+			size += data->raw->size;
+		else
+			size += sizeof(u32);
+
+		WARN_ON_ONCE(size & (sizeof(u64)-1));
+		header.size += size;
 	}
 
 	ret = perf_output_begin(&handle, counter, header.size, nmi, 1);
@@ -2783,8 +2789,21 @@ static void perf_counter_output(struct perf_counter *counter, int nmi,
 		}
 	}
 
-	if (sample_type & PERF_SAMPLE_TP_RECORD)
-		perf_output_copy(&handle, tp->record, tp->size);
+	if (sample_type & PERF_SAMPLE_RAW) {
+		if (data->raw) {
+			perf_output_put(&handle, data->raw->size);
+			perf_output_copy(&handle, data->raw->data, data->raw->size);
+		} else {
+			struct {
+				u32	size;
+				u32	data;
+			} raw = {
+				.size = sizeof(u32),
+				.data = 0,
+			};
+			perf_output_put(&handle, raw);
+		}
+	}
 
 	perf_output_end(&handle);
 }
@@ -2849,7 +2868,8 @@ perf_counter_read_event(struct perf_counter *counter,
  */
 
 struct perf_task_event {
-	struct task_struct	*task;
+	struct task_struct		*task;
+	struct perf_counter_context	*task_ctx;
 
 	struct {
 		struct perf_event_header	header;
@@ -2909,24 +2929,23 @@ static void perf_counter_task_ctx(struct perf_counter_context *ctx,
 static void perf_counter_task_event(struct perf_task_event *task_event)
 {
 	struct perf_cpu_context *cpuctx;
-	struct perf_counter_context *ctx;
+	struct perf_counter_context *ctx = task_event->task_ctx;
 
 	cpuctx = &get_cpu_var(perf_cpu_context);
 	perf_counter_task_ctx(&cpuctx->ctx, task_event);
 	put_cpu_var(perf_cpu_context);
 
 	rcu_read_lock();
-	/*
-	 * doesn't really matter which of the child contexts the
-	 * events ends up in.
-	 */
-	ctx = rcu_dereference(current->perf_counter_ctxp);
+	if (!ctx)
+		ctx = rcu_dereference(task_event->task->perf_counter_ctxp);
 	if (ctx)
 		perf_counter_task_ctx(ctx, task_event);
 	rcu_read_unlock();
 }
 
-static void perf_counter_task(struct task_struct *task, int new)
+static void perf_counter_task(struct task_struct *task,
+			      struct perf_counter_context *task_ctx,
+			      int new)
 {
 	struct perf_task_event task_event;
 
@@ -2936,8 +2955,9 @@ static void perf_counter_task(struct task_struct *task, int new)
 		return;
 
 	task_event = (struct perf_task_event){
-		.task	= task,
-		.event  = {
+		.task	  = task,
+		.task_ctx = task_ctx,
+		.event    = {
 			.header = {
 				.type = new ? PERF_EVENT_FORK : PERF_EVENT_EXIT,
 				.misc = 0,
@@ -2955,7 +2975,7 @@ static void perf_counter_task(struct task_struct *task, int new)
 
 void perf_counter_fork(struct task_struct *task)
 {
-	perf_counter_task(task, 1);
+	perf_counter_task(task, NULL, 1);
 }
 
 /*
@@ -3344,87 +3364,81 @@ int perf_counter_overflow(struct perf_counter *counter, int nmi,
  * Generic software counter infrastructure
  */
 
-static void perf_swcounter_update(struct perf_counter *counter)
+/*
+ * We directly increment counter->count and keep a second value in
+ * counter->hw.period_left to count intervals. This period counter
+ * is kept in the range [-sample_period, 0] so that we can use the
+ * sign as trigger.
+ */
+
+static u64 perf_swcounter_set_period(struct perf_counter *counter)
 {
 	struct hw_perf_counter *hwc = &counter->hw;
-	u64 prev, now;
-	s64 delta;
+	u64 period = hwc->last_period;
+	u64 nr, offset;
+	s64 old, val;
+
+	hwc->last_period = hwc->sample_period;
 
 again:
-	prev = atomic64_read(&hwc->prev_count);
-	now = atomic64_read(&hwc->count);
-	if (atomic64_cmpxchg(&hwc->prev_count, prev, now) != prev)
-		goto again;
+	old = val = atomic64_read(&hwc->period_left);
+	if (val < 0)
+		return 0;
 
-	delta = now - prev;
+	nr = div64_u64(period + val, period);
+	offset = nr * period;
+	val -= offset;
+	if (atomic64_cmpxchg(&hwc->period_left, old, val) != old)
+		goto again;
 
-	atomic64_add(delta, &counter->count);
-	atomic64_sub(delta, &hwc->period_left);
+	return nr;
 }
 
-static void perf_swcounter_set_period(struct perf_counter *counter)
+static void perf_swcounter_overflow(struct perf_counter *counter,
+				    int nmi, struct perf_sample_data *data)
 {
 	struct hw_perf_counter *hwc = &counter->hw;
-	s64 left = atomic64_read(&hwc->period_left);
-	s64 period = hwc->sample_period;
+	u64 overflow;
 
-	if (unlikely(left <= -period)) {
-		left = period;
-		atomic64_set(&hwc->period_left, left);
-		hwc->last_period = period;
-	}
+	data->period = counter->hw.last_period;
+	overflow = perf_swcounter_set_period(counter);
 
-	if (unlikely(left <= 0)) {
-		left += period;
-		atomic64_add(period, &hwc->period_left);
-		hwc->last_period = period;
-	}
+	if (hwc->interrupts == MAX_INTERRUPTS)
+		return;
 
-	atomic64_set(&hwc->prev_count, -left);
-	atomic64_set(&hwc->count, -left);
+	for (; overflow; overflow--) {
+		if (perf_counter_overflow(counter, nmi, data)) {
+			/*
+			 * We inhibit the overflow from happening when
+			 * hwc->interrupts == MAX_INTERRUPTS.
+			 */
+			break;
+		}
+	}
 }
 
-static enum hrtimer_restart perf_swcounter_hrtimer(struct hrtimer *hrtimer)
+static void perf_swcounter_unthrottle(struct perf_counter *counter)
 {
-	enum hrtimer_restart ret = HRTIMER_RESTART;
-	struct perf_sample_data data;
-	struct perf_counter *counter;
-	u64 period;
-
-	counter	= container_of(hrtimer, struct perf_counter, hw.hrtimer);
-	counter->pmu->read(counter);
-
-	data.addr = 0;
-	data.regs = get_irq_regs();
 	/*
-	 * In case we exclude kernel IPs or are somehow not in interrupt
-	 * context, provide the next best thing, the user IP.
+	 * Nothing to do, we already reset hwc->interrupts.
 	 */
-	if ((counter->attr.exclude_kernel || !data.regs) &&
-			!counter->attr.exclude_user)
-		data.regs = task_pt_regs(current);
+}
 
-	if (data.regs) {
-		if (perf_counter_overflow(counter, 0, &data))
-			ret = HRTIMER_NORESTART;
-	}
+static void perf_swcounter_add(struct perf_counter *counter, u64 nr,
+			       int nmi, struct perf_sample_data *data)
+{
+	struct hw_perf_counter *hwc = &counter->hw;
 
-	period = max_t(u64, 10000, counter->hw.sample_period);
-	hrtimer_forward_now(hrtimer, ns_to_ktime(period));
+	atomic64_add(nr, &counter->count);
 
-	return ret;
-}
+	if (!hwc->sample_period)
+		return;
 
-static void perf_swcounter_overflow(struct perf_counter *counter,
-				    int nmi, struct perf_sample_data *data)
-{
-	data->period = counter->hw.last_period;
+	if (!data->regs)
+		return;
 
-	perf_swcounter_update(counter);
-	perf_swcounter_set_period(counter);
-	if (perf_counter_overflow(counter, nmi, data))
-		/* soft-disable the counter */
-		;
+	if (!atomic64_add_negative(nr, &hwc->period_left))
+		perf_swcounter_overflow(counter, nmi, data);
 }
 
 static int perf_swcounter_is_counting(struct perf_counter *counter)
@@ -3488,15 +3502,6 @@ static int perf_swcounter_match(struct perf_counter *counter,
 	return 1;
 }
 
-static void perf_swcounter_add(struct perf_counter *counter, u64 nr,
-			       int nmi, struct perf_sample_data *data)
-{
-	int neg = atomic64_add_negative(nr, &counter->hw.count);
-
-	if (counter->hw.sample_period && !neg && data->regs)
-		perf_swcounter_overflow(counter, nmi, data);
-}
-
 static void perf_swcounter_ctx_event(struct perf_counter_context *ctx,
 				     enum perf_type_id type,
 				     u32 event, u64 nr, int nmi,
@@ -3575,26 +3580,65 @@ void __perf_swcounter_event(u32 event, u64 nr, int nmi,
 
 static void perf_swcounter_read(struct perf_counter *counter)
 {
-	perf_swcounter_update(counter);
 }
 
 static int perf_swcounter_enable(struct perf_counter *counter)
 {
-	perf_swcounter_set_period(counter);
+	struct hw_perf_counter *hwc = &counter->hw;
+
+	if (hwc->sample_period) {
+		hwc->last_period = hwc->sample_period;
+		perf_swcounter_set_period(counter);
+	}
 	return 0;
 }
 
 static void perf_swcounter_disable(struct perf_counter *counter)
 {
-	perf_swcounter_update(counter);
 }
 
 static const struct pmu perf_ops_generic = {
 	.enable		= perf_swcounter_enable,
 	.disable	= perf_swcounter_disable,
 	.read		= perf_swcounter_read,
+	.unthrottle	= perf_swcounter_unthrottle,
 };
 
+/*
+ * hrtimer based swcounter callback
+ */
+
+static enum hrtimer_restart perf_swcounter_hrtimer(struct hrtimer *hrtimer)
+{
+	enum hrtimer_restart ret = HRTIMER_RESTART;
+	struct perf_sample_data data;
+	struct perf_counter *counter;
+	u64 period;
+
+	counter	= container_of(hrtimer, struct perf_counter, hw.hrtimer);
+	counter->pmu->read(counter);
+
+	data.addr = 0;
+	data.regs = get_irq_regs();
+	/*
+	 * In case we exclude kernel IPs or are somehow not in interrupt
+	 * context, provide the next best thing, the user IP.
+	 */
+	if ((counter->attr.exclude_kernel || !data.regs) &&
+			!counter->attr.exclude_user)
+		data.regs = task_pt_regs(current);
+
+	if (data.regs) {
+		if (perf_counter_overflow(counter, 0, &data))
+			ret = HRTIMER_NORESTART;
+	}
+
+	period = max_t(u64, 10000, counter->hw.sample_period);
+	hrtimer_forward_now(hrtimer, ns_to_ktime(period));
+
+	return ret;
+}
+
 /*
  * Software counter: cpu wall time clock
  */
@@ -3715,15 +3759,15 @@ static const struct pmu perf_ops_task_clock = {
 void perf_tpcounter_event(int event_id, u64 addr, u64 count, void *record,
 			  int entry_size)
 {
-	struct perf_tracepoint_record tp = {
+	struct perf_raw_record raw = {
 		.size = entry_size,
-		.record = record,
+		.data = record,
 	};
 
 	struct perf_sample_data data = {
 		.regs = get_irq_regs(),
 		.addr = addr,
-		.private = &tp,
+		.raw = &raw,
 	};
 
 	if (!data.regs)
@@ -3743,6 +3787,14 @@ static void tp_perf_counter_destroy(struct perf_counter *counter)
 
 static const struct pmu *tp_perf_counter_init(struct perf_counter *counter)
 {
+	/*
+	 * Raw tracepoint data is a severe data leak, only allow root to
+	 * have these.
+	 */
+	if ((counter->attr.sample_type & PERF_SAMPLE_RAW) &&
+			!capable(CAP_SYS_ADMIN))
+		return ERR_PTR(-EPERM);
+
 	if (ftrace_profile_enable(counter->attr.config))
 		return NULL;
 
@@ -4285,7 +4337,7 @@ void perf_counter_exit_task(struct task_struct *child)
 	unsigned long flags;
 
 	if (likely(!child->perf_counter_ctxp)) {
-		perf_counter_task(child, 0);
+		perf_counter_task(child, NULL, 0);
 		return;
 	}
 
@@ -4305,6 +4357,7 @@ void perf_counter_exit_task(struct task_struct *child)
 	 * incremented the context's refcount before we do put_ctx below.
 	 */
 	spin_lock(&child_ctx->lock);
+	child->perf_counter_ctxp = NULL;
 	/*
 	 * If this context is a clone; unclone it so it can't get
 	 * swapped to another process while we're removing all
@@ -4318,9 +4371,7 @@ void perf_counter_exit_task(struct task_struct *child)
 	 * won't get any samples after PERF_EVENT_EXIT. We can however still
 	 * get a few PERF_EVENT_READ events.
 	 */
-	perf_counter_task(child, 0);
-
-	child->perf_counter_ctxp = NULL;
+	perf_counter_task(child, child_ctx, 0);
 
 	/*
 	 * We can recurse on the same lock type through:

kernel/trace/blktrace.c

@@ -267,8 +267,8 @@ static void blk_trace_free(struct blk_trace *bt)
 {
 	debugfs_remove(bt->msg_file);
 	debugfs_remove(bt->dropped_file);
-	debugfs_remove(bt->dir);
 	relay_close(bt->rchan);
+	debugfs_remove(bt->dir);
 	free_percpu(bt->sequence);
 	free_percpu(bt->msg_data);
 	kfree(bt);
@@ -378,18 +378,8 @@ static int blk_subbuf_start_callback(struct rchan_buf *buf, void *subbuf,
 
 static int blk_remove_buf_file_callback(struct dentry *dentry)
 {
-	struct dentry *parent = dentry->d_parent;
 	debugfs_remove(dentry);
 
-	/*
-	* this will fail for all but the last file, but that is ok. what we
-	* care about is the top level buts->name directory going away, when
-	* the last trace file is gone. Then we don't have to rmdir() that
-	* manually on trace stop, so it nicely solves the issue with
-	* force killing of running traces.
-	*/
-
-	debugfs_remove(parent);
 	return 0;
 }
 

mm/mempool.c

@@ -303,14 +303,14 @@ EXPORT_SYMBOL(mempool_free_slab);
  */
 void *mempool_kmalloc(gfp_t gfp_mask, void *pool_data)
 {
-	size_t size = (size_t)(long)pool_data;
+	size_t size = (size_t)pool_data;
 	return kmalloc(size, gfp_mask);
 }
 EXPORT_SYMBOL(mempool_kmalloc);
 
 void *mempool_kzalloc(gfp_t gfp_mask, void *pool_data)
 {
-	size_t size = (size_t) pool_data;
+	size_t size = (size_t)pool_data;
 	return kzalloc(size, gfp_mask);
 }
 EXPORT_SYMBOL(mempool_kzalloc);

security/selinux/hooks.c

@@ -1285,6 +1285,8 @@ static int inode_doinit_with_dentry(struct inode *inode, struct dentry *opt_dent
 		rc = inode->i_op->getxattr(dentry, XATTR_NAME_SELINUX,
 					   context, len);
 		if (rc == -ERANGE) {
+			kfree(context);
+
 			/* Need a larger buffer.  Query for the right size. */
 			rc = inode->i_op->getxattr(dentry, XATTR_NAME_SELINUX,
 						   NULL, 0);
@@ -1292,7 +1294,6 @@ static int inode_doinit_with_dentry(struct inode *inode, struct dentry *opt_dent
 				dput(dentry);
 				goto out_unlock;
 			}
-			kfree(context);
 			len = rc;
 			context = kmalloc(len+1, GFP_NOFS);
 			if (!context) {

sound/pci/hda/patch_realtek.c

@@ -13563,6 +13563,8 @@ static int patch_alc269(struct hda_codec *codec)
 		set_capture_mixer(spec);
 	set_beep_amp(spec, 0x0b, 0x04, HDA_INPUT);
 
+	spec->vmaster_nid = 0x02;
+
 	codec->patch_ops = alc_patch_ops;
 	if (board_config == ALC269_AUTO)
 		spec->init_hook = alc269_auto_init;
@@ -15577,9 +15579,12 @@ static int patch_alc861vd(struct hda_codec *codec)
 	spec->stream_digital_playback = &alc861vd_pcm_digital_playback;
 	spec->stream_digital_capture = &alc861vd_pcm_digital_capture;
 
-	spec->adc_nids = alc861vd_adc_nids;
-	spec->num_adc_nids = ARRAY_SIZE(alc861vd_adc_nids);
-	spec->capsrc_nids = alc861vd_capsrc_nids;
+	if (!spec->adc_nids) {
+		spec->adc_nids = alc861vd_adc_nids;
+		spec->num_adc_nids = ARRAY_SIZE(alc861vd_adc_nids);
+	}
+	if (!spec->capsrc_nids)
+		spec->capsrc_nids = alc861vd_capsrc_nids;
 
 	set_capture_mixer(spec);
 	set_beep_amp(spec, 0x0b, 0x05, HDA_INPUT);
@@ -17496,9 +17501,12 @@ static int patch_alc662(struct hda_codec *codec)
 	spec->stream_digital_playback = &alc662_pcm_digital_playback;
 	spec->stream_digital_capture = &alc662_pcm_digital_capture;
 
-	spec->adc_nids = alc662_adc_nids;
-	spec->num_adc_nids = ARRAY_SIZE(alc662_adc_nids);
-	spec->capsrc_nids = alc662_capsrc_nids;
+	if (!spec->adc_nids) {
+		spec->adc_nids = alc662_adc_nids;
+		spec->num_adc_nids = ARRAY_SIZE(alc662_adc_nids);
+	}
+	if (!spec->capsrc_nids)
+		spec->capsrc_nids = alc662_capsrc_nids;
 
 	if (!spec->cap_mixer)
 		set_capture_mixer(spec);

sound/soc/fsl/efika-audio-fabric.c

@@ -30,6 +30,8 @@
 #include "mpc5200_psc_ac97.h"
 #include "../codecs/stac9766.h"
 
+#define DRV_NAME "efika-audio-fabric"
+
 static struct snd_soc_device device;
 static struct snd_soc_card card;
 

sound/soc/fsl/pcm030-audio-fabric.c

@@ -30,6 +30,8 @@
 #include "mpc5200_psc_ac97.h"
 #include "../codecs/wm9712.h"
 
+#define DRV_NAME "pcm030-audio-fabric"
+
 static struct snd_soc_device device;
 static struct snd_soc_card card;
 

tools/perf/Documentation/perf-examples.txt

+
+		------------------------------
+		****** perf by examples ******
+		------------------------------
+
+[ From an e-mail by Ingo Molnar, http://lkml.org/lkml/2009/8/4/346 ]
+
+
+First, discovery/enumeration of available counters can be done via
+'perf list':
+
+titan:~> perf list
+  [...]
+  kmem:kmalloc                             [Tracepoint event]
+  kmem:kmem_cache_alloc                    [Tracepoint event]
+  kmem:kmalloc_node                        [Tracepoint event]
+  kmem:kmem_cache_alloc_node               [Tracepoint event]
+  kmem:kfree                               [Tracepoint event]
+  kmem:kmem_cache_free                     [Tracepoint event]
+  kmem:mm_page_free_direct                 [Tracepoint event]
+  kmem:mm_pagevec_free                     [Tracepoint event]
+  kmem:mm_page_alloc                       [Tracepoint event]
+  kmem:mm_page_alloc_zone_locked           [Tracepoint event]
+  kmem:mm_page_pcpu_drain                  [Tracepoint event]
+  kmem:mm_page_alloc_extfrag               [Tracepoint event]
+
+Then any (or all) of the above event sources can be activated and
+measured. For example the page alloc/free properties of a 'hackbench
+run' are:
+
+ titan:~> perf stat -e kmem:mm_page_pcpu_drain -e kmem:mm_page_alloc
+ -e kmem:mm_pagevec_free -e kmem:mm_page_free_direct ./hackbench 10
+ Time: 0.575
+
+ Performance counter stats for './hackbench 10':
+
+          13857  kmem:mm_page_pcpu_drain
+          27576  kmem:mm_page_alloc
+           6025  kmem:mm_pagevec_free
+          20934  kmem:mm_page_free_direct
+
+    0.613972165  seconds time elapsed
+
+You can observe the statistical properties as well, by using the
+'repeat the workload N times' feature of perf stat:
+
+ titan:~> perf stat --repeat 5 -e kmem:mm_page_pcpu_drain -e
+   kmem:mm_page_alloc -e kmem:mm_pagevec_free -e
+   kmem:mm_page_free_direct ./hackbench 10
+ Time: 0.627
+ Time: 0.644
+ Time: 0.564
+ Time: 0.559
+ Time: 0.626
+
+ Performance counter stats for './hackbench 10' (5 runs):
+
+          12920  kmem:mm_page_pcpu_drain    ( +-   3.359% )
+          25035  kmem:mm_page_alloc         ( +-   3.783% )
+           6104  kmem:mm_pagevec_free       ( +-   0.934% )
+          18376  kmem:mm_page_free_direct   ( +-   4.941% )
+
+    0.643954516  seconds time elapsed   ( +-   2.363% )
+
+Furthermore, these tracepoints can be used to sample the workload as
+well. For example the page allocations done by a 'git gc' can be
+captured the following way:
+
+ titan:~/git> perf record -f -e kmem:mm_page_alloc -c 1 ./git gc
+ Counting objects: 1148, done.
+ Delta compression using up to 2 threads.
+ Compressing objects: 100% (450/450), done.
+ Writing objects: 100% (1148/1148), done.
+ Total 1148 (delta 690), reused 1148 (delta 690)
+ [ perf record: Captured and wrote 0.267 MB perf.data (~11679 samples) ]
+
+To check which functions generated page allocations:
+
+ titan:~/git> perf report
+ # Samples: 10646
+ #
+ # Overhead          Command               Shared Object
+ # ........  ...............  ..........................
+ #
+    23.57%       git-repack  /lib64/libc-2.5.so
+    21.81%              git  /lib64/libc-2.5.so
+    14.59%              git  ./git
+    11.79%       git-repack  ./git
+     7.12%              git  /lib64/ld-2.5.so
+     3.16%       git-repack  /lib64/libpthread-2.5.so
+     2.09%       git-repack  /bin/bash
+     1.97%               rm  /lib64/libc-2.5.so
+     1.39%               mv  /lib64/ld-2.5.so
+     1.37%               mv  /lib64/libc-2.5.so
+     1.12%       git-repack  /lib64/ld-2.5.so
+     0.95%               rm  /lib64/ld-2.5.so
+     0.90%  git-update-serv  /lib64/libc-2.5.so
+     0.73%  git-update-serv  /lib64/ld-2.5.so
+     0.68%             perf  /lib64/libpthread-2.5.so
+     0.64%       git-repack  /usr/lib64/libz.so.1.2.3
+
+Or to see it on a more finegrained level:
+
+titan:~/git> perf report --sort comm,dso,symbol
+# Samples: 10646
+#
+# Overhead          Command               Shared Object  Symbol
+# ........  ...............  ..........................  ......
+#
+     9.35%       git-repack  ./git                       [.] insert_obj_hash
+     9.12%              git  ./git                       [.] insert_obj_hash
+     7.31%              git  /lib64/libc-2.5.so          [.] memcpy
+     6.34%       git-repack  /lib64/libc-2.5.so          [.] _int_malloc
+     6.24%       git-repack  /lib64/libc-2.5.so          [.] memcpy
+     5.82%       git-repack  /lib64/libc-2.5.so          [.] __GI___fork
+     5.47%              git  /lib64/libc-2.5.so          [.] _int_malloc
+     2.99%              git  /lib64/libc-2.5.so          [.] memset
+
+Furthermore, call-graph sampling can be done too, of page
+allocations - to see precisely what kind of page allocations there
+are:
+
+ titan:~/git> perf record -f -g -e kmem:mm_page_alloc -c 1 ./git gc
+ Counting objects: 1148, done.
+ Delta compression using up to 2 threads.
+ Compressing objects: 100% (450/450), done.
+ Writing objects: 100% (1148/1148), done.
+ Total 1148 (delta 690), reused 1148 (delta 690)
+ [ perf record: Captured and wrote 0.963 MB perf.data (~42069 samples) ]
+
+ titan:~/git> perf report -g
+ # Samples: 10686
+ #
+ # Overhead          Command               Shared Object
+ # ........  ...............  ..........................
+ #
+    23.25%       git-repack  /lib64/libc-2.5.so
+                |
+                |--50.00%-- _int_free
+                |
+                |--37.50%-- __GI___fork
+                |          make_child
+                |
+                |--12.50%-- ptmalloc_unlock_all2
+                |          make_child
+                |
+                 --6.25%-- __GI_strcpy
+    21.61%              git  /lib64/libc-2.5.so
+                |
+                |--30.00%-- __GI_read
+                |          |
+                |           --83.33%-- git_config_from_file
+                |                     git_config
+                |                     |
+   [...]
+
+Or you can observe the whole system's page allocations for 10
+seconds:
+
+titan:~/git> perf stat -a -e kmem:mm_page_pcpu_drain -e
+kmem:mm_page_alloc -e kmem:mm_pagevec_free -e
+kmem:mm_page_free_direct sleep 10
+
+ Performance counter stats for 'sleep 10':
+
+         171585  kmem:mm_page_pcpu_drain
+         322114  kmem:mm_page_alloc
+          73623  kmem:mm_pagevec_free
+         254115  kmem:mm_page_free_direct
+
+   10.000591410  seconds time elapsed
+
+Or observe how fluctuating the page allocations are, via statistical
+analysis done over ten 1-second intervals:
+
+ titan:~/git> perf stat --repeat 10 -a -e kmem:mm_page_pcpu_drain -e
+   kmem:mm_page_alloc -e kmem:mm_pagevec_free -e
+   kmem:mm_page_free_direct sleep 1
+
+ Performance counter stats for 'sleep 1' (10 runs):
+
+          17254  kmem:mm_page_pcpu_drain    ( +-   3.709% )
+          34394  kmem:mm_page_alloc         ( +-   4.617% )
+           7509  kmem:mm_pagevec_free       ( +-   4.820% )
+          25653  kmem:mm_page_free_direct   ( +-   3.672% )
+
+    1.058135029  seconds time elapsed   ( +-   3.089% )
+
+Or you can annotate the recorded 'git gc' run on a per symbol basis
+and check which instructions/source-code generated page allocations:
+
+ titan:~/git> perf annotate __GI___fork
+ ------------------------------------------------
+  Percent |      Source code & Disassembly of libc-2.5.so
+ ------------------------------------------------
+          :
+          :
+          :      Disassembly of section .plt:
+          :      Disassembly of section .text:
+          :
+          :      00000031a2e95560 <__fork>:
+ [...]
+     0.00 :        31a2e95602:   b8 38 00 00 00          mov    $0x38,%eax
+     0.00 :        31a2e95607:   0f 05                   syscall
+    83.42 :        31a2e95609:   48 3d 00 f0 ff ff       cmp    $0xfffffffffffff000,%rax
+     0.00 :        31a2e9560f:   0f 87 4d 01 00 00       ja     31a2e95762 <__fork+0x202>
+     0.00 :        31a2e95615:   85 c0                   test   %eax,%eax
+
+( this shows that 83.42% of __GI___fork's page allocations come from
+  the 0x38 system call it performs. )
+
+etc. etc. - a lot more is possible. I could list a dozen of
+other different usecases straight away - neither of which is
+possible via /proc/vmstat.
+
+/proc/vmstat is not in the same league really, in terms of
+expressive power of system analysis and performance
+analysis.
+
+All that the above results needed were those new tracepoints
+in include/tracing/events/kmem.h.
+
+	Ingo
+
+

tools/perf/Documentation/perf-stat.txt

@@ -40,7 +40,7 @@ OPTIONS
 -a::
         system-wide collection
 
--S::
+-c::
         scale counter values
 
 EXAMPLES

tools/perf/Documentation/perf-top.txt

@@ -3,36 +3,122 @@ perf-top(1)
 
 NAME
 ----
-perf-top - Run a command and profile it
+perf-top - System profiling tool.
 
 SYNOPSIS
 --------
 [verse]
-'perf top' [-e <EVENT> | --event=EVENT] [-l] [-a] <command>
+'perf top' [-e <EVENT> | --event=EVENT] [<options>]
 
 DESCRIPTION
 -----------
-This command runs a command and gathers a performance counter profile
-from it.
+This command generates and displays a performance counter profile in realtime.
 
 
 OPTIONS
 -------
-<command>...::
-	Any command you can specify in a shell.
+-a::
+--all-cpus::
+        System-wide collection.  (default)
+
+-c <count>::
+--count=<count>::
+	Event period to sample.
+
+-C <cpu>::
+--CPU=<cpu>::
+	CPU to profile.
+
+-d <seconds>::
+--delay=<seconds>::
+	Number of seconds to delay between refreshes.
 
--e::
---event=::
+-e <event>::
+--event=<event>::
 	Select the PMU event. Selection can be a symbolic event name
 	(use 'perf list' to list all events) or a raw PMU
 	event (eventsel+umask) in the form of rNNN where NNN is a
-	 hexadecimal event descriptor.
+	hexadecimal event descriptor.
 
--a::
-        system-wide collection
+-E <entries>::
+--entries=<entries>::
+	Display this many functions.
+
+-f <count>::
+--count-filter=<count>::
+	Only display functions with more events than this.
+
+-F <freq>::
+--freq=<freq>::
+	Profile at this frequency.
+
+-i::
+--inherit::
+	Child tasks inherit counters, only makes sens with -p option.
+
+-k <path>::
+--vmlinux=<path>::
+	Path to vmlinux.  Required for annotation functionality.
+
+-m <pages>::
+--mmap-pages=<pages>::
+	Number of mmapped data pages.
+
+-p <pid>::
+--pid=<pid>::
+	Profile events on existing pid.
+
+-r <priority>::
+--realtime=<priority>::
+	Collect data with this RT SCHED_FIFO priority.
+
+-s <symbol>::
+--sym-annotate=<symbol>::
+        Annotate this symbol.  Requires -k option.
+
+-v::
+--verbose::
+	Be more verbose (show counter open errors, etc).
+
+-z::
+--zero::
+	Zero history across display updates.
+
+INTERACTIVE PROMPTING KEYS
+--------------------------
+
+[d]::
+	Display refresh delay.
+
+[e]::
+	Number of entries to display.
+
+[E]::
+	Event to display when multiple counters are active.
+
+[f]::
+	Profile display filter (>= hit count).
+
+[F]::
+	Annotation display filter (>= % of total).
+
+[s]::
+	Annotate symbol.
+
+[S]::
+	Stop annotation, return to full profile display.
+
+[w]::
+	Toggle between weighted sum and individual count[E]r profile.
+
+[z]::
+	Toggle event count zeroing across display updates.
+
+[qQ]::
+	Quit.
+
+Pressing any unmapped key displays a menu, and prompts for input.
 
--l::
-        scale counter values
 
 SEE ALSO
 --------

tools/perf/Makefile

@@ -387,10 +387,14 @@ else
 
 	has_bfd_iberty := $(shell sh -c "(echo '\#include <bfd.h>'; echo 'int main(void) { bfd_demangle(0, 0, 0); return 0; }') | $(CC) -x c - $(ALL_CFLAGS) -o /dev/null $(ALL_LDFLAGS) -lbfd -liberty > /dev/null 2>&1 && echo y")
 
+	has_bfd_iberty_z := $(shell sh -c "(echo '\#include <bfd.h>'; echo 'int main(void) { bfd_demangle(0, 0, 0); return 0; }') | $(CC) -x c - $(ALL_CFLAGS) -o /dev/null $(ALL_LDFLAGS) -lbfd -liberty -lz > /dev/null 2>&1 && echo y")
+
 	ifeq ($(has_bfd),y)
 		EXTLIBS += -lbfd
 	else ifeq ($(has_bfd_iberty),y)
 		EXTLIBS += -lbfd -liberty
+	else ifeq ($(has_bfd_iberty_z),y)
+		EXTLIBS += -lbfd -liberty -lz
 	else
 		msg := $(warning No bfd.h/libbfd found, install binutils-dev[el] to gain symbol demangling)
 		BASIC_CFLAGS += -DNO_DEMANGLE

tools/perf/builtin-record.c

@@ -525,10 +525,14 @@ static int __cmd_record(int argc, const char **argv)
 	signal(SIGCHLD, sig_handler);
 	signal(SIGINT, sig_handler);
 
-	if (!stat(output_name, &st) && !force && !append_file) {
-		fprintf(stderr, "Error, output file %s exists, use -A to append or -f to overwrite.\n",
-				output_name);
-		exit(-1);
+	if (!stat(output_name, &st) && st.st_size) {
+		if (!force && !append_file) {
+			fprintf(stderr, "Error, output file %s exists, use -A to append or -f to overwrite.\n",
+					output_name);
+			exit(-1);
+		}
+	} else {
+		append_file = 0;
 	}
 
 	flags = O_CREAT|O_RDWR;

tools/perf/builtin-report.c

@@ -68,7 +68,7 @@ static int		callchain;
 
 static
 struct callchain_param	callchain_param = {
-	.mode	= CHAIN_GRAPH_ABS,
+	.mode	= CHAIN_GRAPH_REL,
 	.min_percent = 0.5
 };
 
@@ -112,7 +112,9 @@ struct read_event {
 	struct perf_event_header header;
 	u32 pid,tid;
 	u64 value;
-	u64 format[3];
+	u64 time_enabled;
+	u64 time_running;
+	u64 id;
 };
 
 typedef union event_union {
@@ -698,7 +700,8 @@ sort__sym_print(FILE *fp, struct hist_entry *self, unsigned int width __used)
 	size_t ret = 0;
 
 	if (verbose)
-		ret += repsep_fprintf(fp, "%#018llx  ", (u64)self->ip);
+		ret += repsep_fprintf(fp, "%#018llx %c ", (u64)self->ip,
+				      dso__symtab_origin(self->dso));
 
 	ret += repsep_fprintf(fp, "[%c] ", self->level);
 	if (self->sym) {
@@ -888,6 +891,21 @@ ipchain__fprintf_graph(FILE *fp, struct callchain_list *chain, int depth,
 	return ret;
 }
 
+static struct symbol *rem_sq_bracket;
+static struct callchain_list rem_hits;
+
+static void init_rem_hits(void)
+{
+	rem_sq_bracket = malloc(sizeof(*rem_sq_bracket) + 6);
+	if (!rem_sq_bracket) {
+		fprintf(stderr, "Not enough memory to display remaining hits\n");
+		return;
+	}
+
+	strcpy(rem_sq_bracket->name, "[...]");
+	rem_hits.sym = rem_sq_bracket;
+}
+
 static size_t
 callchain__fprintf_graph(FILE *fp, struct callchain_node *self,
 			u64 total_samples, int depth, int depth_mask)
@@ -897,25 +915,34 @@ callchain__fprintf_graph(FILE *fp, struct callchain_node *self,
 	struct callchain_list *chain;
 	int new_depth_mask = depth_mask;
 	u64 new_total;
+	u64 remaining;
 	size_t ret = 0;
 	int i;
 
 	if (callchain_param.mode == CHAIN_GRAPH_REL)
-		new_total = self->cumul_hit;
+		new_total = self->children_hit;
 	else
 		new_total = total_samples;
 
+	remaining = new_total;
+
 	node = rb_first(&self->rb_root);
 	while (node) {
+		u64 cumul;
+
 		child = rb_entry(node, struct callchain_node, rb_node);
+		cumul = cumul_hits(child);
+		remaining -= cumul;
 
 		/*
 		 * The depth mask manages the output of pipes that show
 		 * the depth. We don't want to keep the pipes of the current
-		 * level for the last child of this depth
+		 * level for the last child of this depth.
+		 * Except if we have remaining filtered hits. They will
+		 * supersede the last child
 		 */
 		next = rb_next(node);
-		if (!next)
+		if (!next && (callchain_param.mode != CHAIN_GRAPH_REL || !remaining))
 			new_depth_mask &= ~(1 << (depth - 1));
 
 		/*
@@ -930,7 +957,7 @@ callchain__fprintf_graph(FILE *fp, struct callchain_node *self,
 			ret += ipchain__fprintf_graph(fp, chain, depth,
 						      new_depth_mask, i++,
 						      new_total,
-						      child->cumul_hit);
+						      cumul);
 		}
 		ret += callchain__fprintf_graph(fp, child, new_total,
 						depth + 1,
@@ -938,6 +965,19 @@ callchain__fprintf_graph(FILE *fp, struct callchain_node *self,
 		node = next;
 	}
 
+	if (callchain_param.mode == CHAIN_GRAPH_REL &&
+		remaining && remaining != new_total) {
+
+		if (!rem_sq_bracket)
+			return ret;
+
+		new_depth_mask &= ~(1 << (depth - 1));
+
+		ret += ipchain__fprintf_graph(fp, &rem_hits, depth,
+					      new_depth_mask, 0, new_total,
+					      remaining);
+	}
+
 	return ret;
 }
 
@@ -1358,6 +1398,8 @@ static size_t output__fprintf(FILE *fp, u64 total_samples)
 	unsigned int width;
 	char *col_width = col_width_list_str;
 
+	init_rem_hits();
+
 	fprintf(fp, "# Samples: %Ld\n", (u64)total_samples);
 	fprintf(fp, "#\n");
 
@@ -1429,6 +1471,8 @@ static size_t output__fprintf(FILE *fp, u64 total_samples)
 	}
 	fprintf(fp, "\n");
 
+	free(rem_sq_bracket);
+
 	return ret;
 }
 
@@ -1690,14 +1734,37 @@ static void trace_event(event_t *event)
 	dprintf(".\n");
 }
 
+static struct perf_header	*header;
+
+static struct perf_counter_attr *perf_header__find_attr(u64 id)
+{
+	int i;
+
+	for (i = 0; i < header->attrs; i++) {
+		struct perf_header_attr *attr = header->attr[i];
+		int j;
+
+		for (j = 0; j < attr->ids; j++) {
+			if (attr->id[j] == id)
+				return &attr->attr;
+		}
+	}
+
+	return NULL;
+}
+
 static int
 process_read_event(event_t *event, unsigned long offset, unsigned long head)
 {
-	dprintf("%p [%p]: PERF_EVENT_READ: %d %d %Lu\n",
+	struct perf_counter_attr *attr = perf_header__find_attr(event->read.id);
+
+	dprintf("%p [%p]: PERF_EVENT_READ: %d %d %s %Lu\n",
 			(void *)(offset + head),
 			(void *)(long)(event->header.size),
 			event->read.pid,
 			event->read.tid,
+			attr ? __event_name(attr->type, attr->config)
+			     : "FAIL",
 			event->read.value);
 
 	return 0;
@@ -1743,8 +1810,6 @@ process_event(event_t *event, unsigned long offset, unsigned long head)
 	return 0;
 }
 
-static struct perf_header	*header;
-
 static u64 perf_header__sample_type(void)
 {
 	u64 sample_type = 0;
@@ -1812,6 +1877,13 @@ static int __cmd_report(void)
 					" -g?\n");
 			exit(-1);
 		}
+	} else if (callchain_param.mode != CHAIN_NONE && !callchain) {
+			callchain = 1;
+			if (register_callchain_param(&callchain_param) < 0) {
+				fprintf(stderr, "Can't register callchain"
+						" params\n");
+				exit(-1);
+			}
 	}
 
 	if (load_kernel() < 0) {
@@ -1950,6 +2022,13 @@ parse_callchain_opt(const struct option *opt __used, const char *arg,
 	else if (!strncmp(tok, "fractal", strlen(arg)))
 		callchain_param.mode = CHAIN_GRAPH_REL;
 
+	else if (!strncmp(tok, "none", strlen(arg))) {
+		callchain_param.mode = CHAIN_NONE;
+		callchain = 0;
+
+		return 0;
+	}
+
 	else
 		return -1;
 

tools/perf/builtin-stat.c

@@ -496,7 +496,7 @@ static const struct option options[] = {
 		    "stat events on existing pid"),
 	OPT_BOOLEAN('a', "all-cpus", &system_wide,
 		    "system-wide collection from all CPUs"),
-	OPT_BOOLEAN('S', "scale", &scale,
+	OPT_BOOLEAN('c', "scale", &scale,
 		    "scale/normalize counters"),
 	OPT_BOOLEAN('v', "verbose", &verbose,
 		    "be more verbose (show counter open errors, etc)"),

tools/perf/builtin-top.c

@@ -31,6 +31,8 @@
 #include <fcntl.h>
 
 #include <stdio.h>
+#include <termios.h>
+#include <unistd.h>
 
 #include <errno.h>
 #include <time.h>
@@ -54,7 +56,7 @@ static int			system_wide			=  0;
 
 static int			default_interval		= 100000;
 
-static u64			count_filter			=  5;
+static int			count_filter			=  5;
 static int			print_entries			= 15;
 
 static int			target_pid			= -1;
@@ -69,14 +71,27 @@ static int			freq				=  0;
 static int			verbose				=  0;
 static char			*vmlinux			=  NULL;
 
-static char			*sym_filter;
-static unsigned long		filter_start;
-static unsigned long		filter_end;
-
 static int			delay_secs			=  2;
 static int			zero;
 static int			dump_symtab;
 
+/*
+ * Source
+ */
+
+struct source_line {
+	u64			eip;
+	unsigned long		count[MAX_COUNTERS];
+	char			*line;
+	struct source_line	*next;
+};
+
+static char			*sym_filter			=  NULL;
+struct sym_entry		*sym_filter_entry		=  NULL;
+static int			sym_pcnt_filter			=  5;
+static int			sym_counter			=  0;
+static int			display_weighted		= -1;
+
 /*
  * Symbols
  */
@@ -91,9 +106,237 @@ struct sym_entry {
 	unsigned long		snap_count;
 	double			weight;
 	int			skip;
+	struct source_line	*source;
+	struct source_line	*lines;
+	struct source_line	**lines_tail;
+	pthread_mutex_t		source_lock;
 };
 
-struct sym_entry		*sym_filter_entry;
+/*
+ * Source functions
+ */
+
+static void parse_source(struct sym_entry *syme)
+{
+	struct symbol *sym;
+	struct module *module;
+	struct section *section = NULL;
+	FILE *file;
+	char command[PATH_MAX*2], *path = vmlinux;
+	u64 start, end, len;
+
+	if (!syme)
+		return;
+
+	if (syme->lines) {
+		pthread_mutex_lock(&syme->source_lock);
+		goto out_assign;
+	}
+
+	sym = (struct symbol *)(syme + 1);
+	module = sym->module;
+
+	if (module)
+		path = module->path;
+	if (!path)
+		return;
+
+	start = sym->obj_start;
+	if (!start)
+		start = sym->start;
+
+	if (module) {
+		section = module->sections->find_section(module->sections, ".text");
+		if (section)
+			start -= section->vma;
+	}
+
+	end = start + sym->end - sym->start + 1;
+	len = sym->end - sym->start;
+
+	sprintf(command, "objdump --start-address=0x%016Lx --stop-address=0x%016Lx -dS %s", start, end, path);
+
+	file = popen(command, "r");
+	if (!file)
+		return;
+
+	pthread_mutex_lock(&syme->source_lock);
+	syme->lines_tail = &syme->lines;
+	while (!feof(file)) {
+		struct source_line *src;
+		size_t dummy = 0;
+		char *c;
+
+		src = malloc(sizeof(struct source_line));
+		assert(src != NULL);
+		memset(src, 0, sizeof(struct source_line));
+
+		if (getline(&src->line, &dummy, file) < 0)
+			break;
+		if (!src->line)
+			break;
+
+		c = strchr(src->line, '\n');
+		if (c)
+			*c = 0;
+
+		src->next = NULL;
+		*syme->lines_tail = src;
+		syme->lines_tail = &src->next;
+
+		if (strlen(src->line)>8 && src->line[8] == ':') {
+			src->eip = strtoull(src->line, NULL, 16);
+			if (section)
+				src->eip += section->vma;
+		}
+		if (strlen(src->line)>8 && src->line[16] == ':') {
+			src->eip = strtoull(src->line, NULL, 16);
+			if (section)
+				src->eip += section->vma;
+		}
+	}
+	pclose(file);
+out_assign:
+	sym_filter_entry = syme;
+	pthread_mutex_unlock(&syme->source_lock);
+}
+
+static void __zero_source_counters(struct sym_entry *syme)
+{
+	int i;
+	struct source_line *line;
+
+	line = syme->lines;
+	while (line) {
+		for (i = 0; i < nr_counters; i++)
+			line->count[i] = 0;
+		line = line->next;
+	}
+}
+
+static void record_precise_ip(struct sym_entry *syme, int counter, u64 ip)
+{
+	struct source_line *line;
+
+	if (syme != sym_filter_entry)
+		return;
+
+	if (pthread_mutex_trylock(&syme->source_lock))
+		return;
+
+	if (!syme->source)
+		goto out_unlock;
+
+	for (line = syme->lines; line; line = line->next) {
+		if (line->eip == ip) {
+			line->count[counter]++;
+			break;
+		}
+		if (line->eip > ip)
+			break;
+	}
+out_unlock:
+	pthread_mutex_unlock(&syme->source_lock);
+}
+
+static void lookup_sym_source(struct sym_entry *syme)
+{
+	struct symbol *symbol = (struct symbol *)(syme + 1);
+	struct source_line *line;
+	char pattern[PATH_MAX];
+	char *idx;
+
+	sprintf(pattern, "<%s>:", symbol->name);
+
+	if (symbol->module) {
+		idx = strstr(pattern, "\t");
+		if (idx)
+			*idx = 0;
+	}
+
+	pthread_mutex_lock(&syme->source_lock);
+	for (line = syme->lines; line; line = line->next) {
+		if (strstr(line->line, pattern)) {
+			syme->source = line;
+			break;
+		}
+	}
+	pthread_mutex_unlock(&syme->source_lock);
+}
+
+static void show_lines(struct source_line *queue, int count, int total)
+{
+	int i;
+	struct source_line *line;
+
+	line = queue;
+	for (i = 0; i < count; i++) {
+		float pcnt = 100.0*(float)line->count[sym_counter]/(float)total;
+
+		printf("%8li %4.1f%%\t%s\n", line->count[sym_counter], pcnt, line->line);
+		line = line->next;
+	}
+}
+
+#define TRACE_COUNT     3
+
+static void show_details(struct sym_entry *syme)
+{
+	struct symbol *symbol;
+	struct source_line *line;
+	struct source_line *line_queue = NULL;
+	int displayed = 0;
+	int line_queue_count = 0, total = 0, more = 0;
+
+	if (!syme)
+		return;
+
+	if (!syme->source)
+		lookup_sym_source(syme);
+
+	if (!syme->source)
+		return;
+
+	symbol = (struct symbol *)(syme + 1);
+	printf("Showing %s for %s\n", event_name(sym_counter), symbol->name);
+	printf("  Events  Pcnt (>=%d%%)\n", sym_pcnt_filter);
+
+	pthread_mutex_lock(&syme->source_lock);
+	line = syme->source;
+	while (line) {
+		total += line->count[sym_counter];
+		line = line->next;
+	}
+
+	line = syme->source;
+	while (line) {
+		float pcnt = 0.0;
+
+		if (!line_queue_count)
+			line_queue = line;
+		line_queue_count++;
+
+		if (line->count[sym_counter])
+			pcnt = 100.0 * line->count[sym_counter] / (float)total;
+		if (pcnt >= (float)sym_pcnt_filter) {
+			if (displayed <= print_entries)
+				show_lines(line_queue, line_queue_count, total);
+			else more++;
+			displayed += line_queue_count;
+			line_queue_count = 0;
+			line_queue = NULL;
+		} else if (line_queue_count > TRACE_COUNT) {
+			line_queue = line_queue->next;
+			line_queue_count--;
+		}
+
+		line->count[sym_counter] = zero ? 0 : line->count[sym_counter] * 7 / 8;
+		line = line->next;
+	}
+	pthread_mutex_unlock(&syme->source_lock);
+	if (more)
+		printf("%d lines not displayed, maybe increase display entries [e]\n", more);
+}
 
 struct dso			*kernel_dso;
 
@@ -112,6 +355,9 @@ static double sym_weight(const struct sym_entry *sym)
 	double weight = sym->snap_count;
 	int counter;
 
+	if (!display_weighted)
+		return weight;
+
 	for (counter = 1; counter < nr_counters-1; counter++)
 		weight *= sym->count[counter];
 
@@ -159,7 +405,7 @@ static void rb_insert_active_sym(struct rb_root *tree, struct sym_entry *se)
 static void print_sym_table(void)
 {
 	int printed = 0, j;
-	int counter;
+	int counter, snap = !display_weighted ? sym_counter : 0;
 	float samples_per_sec = samples/delay_secs;
 	float ksamples_per_sec = (samples-userspace_samples)/delay_secs;
 	float sum_ksamples = 0.0;
@@ -175,7 +421,7 @@ static void print_sym_table(void)
 	pthread_mutex_unlock(&active_symbols_lock);
 
 	list_for_each_entry_safe_from(syme, n, &active_symbols, node) {
-		syme->snap_count = syme->count[0];
+		syme->snap_count = syme->count[snap];
 		if (syme->snap_count != 0) {
 			syme->weight = sym_weight(syme);
 			rb_insert_active_sym(&tmp, syme);
@@ -195,7 +441,7 @@ static void print_sym_table(void)
 		samples_per_sec,
 		100.0 - (100.0*((samples_per_sec-ksamples_per_sec)/samples_per_sec)));
 
-	if (nr_counters == 1) {
+	if (nr_counters == 1 || !display_weighted) {
 		printf("%Ld", (u64)attrs[0].sample_period);
 		if (freq)
 			printf("Hz ");
@@ -203,7 +449,9 @@ static void print_sym_table(void)
 			printf(" ");
 	}
 
-	for (counter = 0; counter < nr_counters; counter++) {
+	if (!display_weighted)
+		printf("%s", event_name(sym_counter));
+	else for (counter = 0; counter < nr_counters; counter++) {
 		if (counter)
 			printf("/");
 
@@ -228,6 +476,11 @@ static void print_sym_table(void)
 
 	printf("------------------------------------------------------------------------------\n\n");
 
+	if (sym_filter_entry) {
+		show_details(sym_filter_entry);
+		return;
+	}
+
 	if (nr_counters == 1)
 		printf("             samples    pcnt");
 	else
@@ -242,13 +495,13 @@ static void print_sym_table(void)
 		struct symbol *sym = (struct symbol *)(syme + 1);
 		double pcnt;
 
-		if (++printed > print_entries || syme->snap_count < count_filter)
+		if (++printed > print_entries || (int)syme->snap_count < count_filter)
 			continue;
 
 		pcnt = 100.0 - (100.0 * ((sum_ksamples - syme->snap_count) /
 					 sum_ksamples));
 
-		if (nr_counters == 1)
+		if (nr_counters == 1 || !display_weighted)
 			printf("%20.2f - ", syme->weight);
 		else
 			printf("%9.1f %10ld - ", syme->weight, syme->snap_count);
@@ -261,19 +514,250 @@ static void print_sym_table(void)
 	}
 }
 
+static void prompt_integer(int *target, const char *msg)
+{
+	char *buf = malloc(0), *p;
+	size_t dummy = 0;
+	int tmp;
+
+	fprintf(stdout, "\n%s: ", msg);
+	if (getline(&buf, &dummy, stdin) < 0)
+		return;
+
+	p = strchr(buf, '\n');
+	if (p)
+		*p = 0;
+
+	p = buf;
+	while(*p) {
+		if (!isdigit(*p))
+			goto out_free;
+		p++;
+	}
+	tmp = strtoul(buf, NULL, 10);
+	*target = tmp;
+out_free:
+	free(buf);
+}
+
+static void prompt_percent(int *target, const char *msg)
+{
+	int tmp = 0;
+
+	prompt_integer(&tmp, msg);
+	if (tmp >= 0 && tmp <= 100)
+		*target = tmp;
+}
+
+static void prompt_symbol(struct sym_entry **target, const char *msg)
+{
+	char *buf = malloc(0), *p;
+	struct sym_entry *syme = *target, *n, *found = NULL;
+	size_t dummy = 0;
+
+	/* zero counters of active symbol */
+	if (syme) {
+		pthread_mutex_lock(&syme->source_lock);
+		__zero_source_counters(syme);
+		*target = NULL;
+		pthread_mutex_unlock(&syme->source_lock);
+	}
+
+	fprintf(stdout, "\n%s: ", msg);
+	if (getline(&buf, &dummy, stdin) < 0)
+		goto out_free;
+
+	p = strchr(buf, '\n');
+	if (p)
+		*p = 0;
+
+	pthread_mutex_lock(&active_symbols_lock);
+	syme = list_entry(active_symbols.next, struct sym_entry, node);
+	pthread_mutex_unlock(&active_symbols_lock);
+
+	list_for_each_entry_safe_from(syme, n, &active_symbols, node) {
+		struct symbol *sym = (struct symbol *)(syme + 1);
+
+		if (!strcmp(buf, sym->name)) {
+			found = syme;
+			break;
+		}
+	}
+
+	if (!found) {
+		fprintf(stderr, "Sorry, %s is not active.\n", sym_filter);
+		sleep(1);
+		return;
+	} else
+		parse_source(found);
+
+out_free:
+	free(buf);
+}
+
+static void print_mapped_keys(void)
+{
+	char *name = NULL;
+
+	if (sym_filter_entry) {
+		struct symbol *sym = (struct symbol *)(sym_filter_entry+1);
+		name = sym->name;
+	}
+
+	fprintf(stdout, "\nMapped keys:\n");
+	fprintf(stdout, "\t[d]     display refresh delay.             \t(%d)\n", delay_secs);
+	fprintf(stdout, "\t[e]     display entries (lines).           \t(%d)\n", print_entries);
+
+	if (nr_counters > 1)
+		fprintf(stdout, "\t[E]     active event counter.              \t(%s)\n", event_name(sym_counter));
+
+	fprintf(stdout, "\t[f]     profile display filter (count).    \t(%d)\n", count_filter);
+
+	if (vmlinux) {
+		fprintf(stdout, "\t[F]     annotate display filter (percent). \t(%d%%)\n", sym_pcnt_filter);
+		fprintf(stdout, "\t[s]     annotate symbol.                   \t(%s)\n", name?: "NULL");
+		fprintf(stdout, "\t[S]     stop annotation.\n");
+	}
+
+	if (nr_counters > 1)
+		fprintf(stdout, "\t[w]     toggle display weighted/count[E]r. \t(%d)\n", display_weighted ? 1 : 0);
+
+	fprintf(stdout, "\t[z]     toggle sample zeroing.             \t(%d)\n", zero ? 1 : 0);
+	fprintf(stdout, "\t[qQ]    quit.\n");
+}
+
+static int key_mapped(int c)
+{
+	switch (c) {
+		case 'd':
+		case 'e':
+		case 'f':
+		case 'z':
+		case 'q':
+		case 'Q':
+			return 1;
+		case 'E':
+		case 'w':
+			return nr_counters > 1 ? 1 : 0;
+		case 'F':
+		case 's':
+		case 'S':
+			return vmlinux ? 1 : 0;
+	}
+
+	return 0;
+}
+
+static void handle_keypress(int c)
+{
+	if (!key_mapped(c)) {
+		struct pollfd stdin_poll = { .fd = 0, .events = POLLIN };
+		struct termios tc, save;
+
+		print_mapped_keys();
+		fprintf(stdout, "\nEnter selection, or unmapped key to continue: ");
+		fflush(stdout);
+
+		tcgetattr(0, &save);
+		tc = save;
+		tc.c_lflag &= ~(ICANON | ECHO);
+		tc.c_cc[VMIN] = 0;
+		tc.c_cc[VTIME] = 0;
+		tcsetattr(0, TCSANOW, &tc);
+
+		poll(&stdin_poll, 1, -1);
+		c = getc(stdin);
+
+		tcsetattr(0, TCSAFLUSH, &save);
+		if (!key_mapped(c))
+			return;
+	}
+
+	switch (c) {
+		case 'd':
+			prompt_integer(&delay_secs, "Enter display delay");
+			break;
+		case 'e':
+			prompt_integer(&print_entries, "Enter display entries (lines)");
+			break;
+		case 'E':
+			if (nr_counters > 1) {
+				int i;
+
+				fprintf(stderr, "\nAvailable events:");
+				for (i = 0; i < nr_counters; i++)
+					fprintf(stderr, "\n\t%d %s", i, event_name(i));
+
+				prompt_integer(&sym_counter, "Enter details event counter");
+
+				if (sym_counter >= nr_counters) {
+					fprintf(stderr, "Sorry, no such event, using %s.\n", event_name(0));
+					sym_counter = 0;
+					sleep(1);
+				}
+			} else sym_counter = 0;
+			break;
+		case 'f':
+			prompt_integer(&count_filter, "Enter display event count filter");
+			break;
+		case 'F':
+			prompt_percent(&sym_pcnt_filter, "Enter details display event filter (percent)");
+			break;
+		case 'q':
+		case 'Q':
+			printf("exiting.\n");
+			exit(0);
+		case 's':
+			prompt_symbol(&sym_filter_entry, "Enter details symbol");
+			break;
+		case 'S':
+			if (!sym_filter_entry)
+				break;
+			else {
+				struct sym_entry *syme = sym_filter_entry;
+
+				pthread_mutex_lock(&syme->source_lock);
+				sym_filter_entry = NULL;
+				__zero_source_counters(syme);
+				pthread_mutex_unlock(&syme->source_lock);
+			}
+			break;
+		case 'w':
+			display_weighted = ~display_weighted;
+			break;
+		case 'z':
+			zero = ~zero;
+			break;
+	}
+}
+
 static void *display_thread(void *arg __used)
 {
 	struct pollfd stdin_poll = { .fd = 0, .events = POLLIN };
-	int delay_msecs = delay_secs * 1000;
+	struct termios tc, save;
+	int delay_msecs, c;
+
+	tcgetattr(0, &save);
+	tc = save;
+	tc.c_lflag &= ~(ICANON | ECHO);
+	tc.c_cc[VMIN] = 0;
+	tc.c_cc[VTIME] = 0;
 
-	printf("PerfTop refresh period: %d seconds\n", delay_secs);
+repeat:
+	delay_msecs = delay_secs * 1000;
+	tcsetattr(0, TCSANOW, &tc);
+	/* trash return*/
+	getc(stdin);
 
 	do {
 		print_sym_table();
 	} while (!poll(&stdin_poll, 1, delay_msecs) == 1);
 
-	printf("key pressed - exiting.\n");
-	exit(0);
+	c = getc(stdin);
+	tcsetattr(0, TCSAFLUSH, &save);
+
+	handle_keypress(c);
+	goto repeat;
 
 	return NULL;
 }
@@ -293,7 +777,6 @@ static const char *skip_symbols[] = {
 
 static int symbol_filter(struct dso *self, struct symbol *sym)
 {
-	static int filter_match;
 	struct sym_entry *syme;
 	const char *name = sym->name;
 	int i;
@@ -315,6 +798,10 @@ static int symbol_filter(struct dso *self, struct symbol *sym)
 		return 1;
 
 	syme = dso__sym_priv(self, sym);
+	pthread_mutex_init(&syme->source_lock, NULL);
+	if (!sym_filter_entry && sym_filter && !strcmp(name, sym_filter))
+		sym_filter_entry = syme;
+
 	for (i = 0; skip_symbols[i]; i++) {
 		if (!strcmp(skip_symbols[i], name)) {
 			syme->skip = 1;
@@ -322,29 +809,6 @@ static int symbol_filter(struct dso *self, struct symbol *sym)
 		}
 	}
 
-	if (filter_match == 1) {
-		filter_end = sym->start;
-		filter_match = -1;
-		if (filter_end - filter_start > 10000) {
-			fprintf(stderr,
-				"hm, too large filter symbol <%s> - skipping.\n",
-				sym_filter);
-			fprintf(stderr, "symbol filter start: %016lx\n",
-				filter_start);
-			fprintf(stderr, "                end: %016lx\n",
-				filter_end);
-			filter_end = filter_start = 0;
-			sym_filter = NULL;
-			sleep(1);
-		}
-	}
-
-	if (filter_match == 0 && sym_filter && !strcmp(name, sym_filter)) {
-		filter_match = 1;
-		filter_start = sym->start;
-	}
-
-
 	return 0;
 }
 
@@ -380,8 +844,6 @@ static int parse_symbols(void)
 	return -1;
 }
 
-#define TRACE_COUNT     3
-
 /*
  * Binary search in the histogram table and record the hit:
  */
@@ -394,6 +856,7 @@ static void record_ip(u64 ip, int counter)
 
 		if (!syme->skip) {
 			syme->count[counter]++;
+			record_precise_ip(syme, counter, ip);
 			pthread_mutex_lock(&active_symbols_lock);
 			if (list_empty(&syme->node) || !syme->node.next)
 				__list_insert_active_sym(syme);
@@ -690,8 +1153,8 @@ static const struct option options[] = {
 			    "put the counters into a counter group"),
 	OPT_BOOLEAN('i', "inherit", &inherit,
 		    "child tasks inherit counters"),
-	OPT_STRING('s', "sym-filter", &sym_filter, "pattern",
-		    "only display symbols matchig this pattern"),
+	OPT_STRING('s', "sym-annotate", &sym_filter, "symbol name",
+		    "symbol to annotate - requires -k option"),
 	OPT_BOOLEAN('z', "zero", &zero,
 		    "zero history across updates"),
 	OPT_INTEGER('F', "freq", &freq,
@@ -734,6 +1197,7 @@ int cmd_top(int argc, const char **argv, const char *prefix __used)
 		delay_secs = 1;
 
 	parse_symbols();
+	parse_source(sym_filter_entry);
 
 	/*
 	 * Fill in the ones not specifically initialized via -c:

tools/perf/util/callchain.c

@@ -13,6 +13,7 @@
 #include <stdio.h>
 #include <stdbool.h>
 #include <errno.h>
+#include <math.h>
 
 #include "callchain.h"
 
@@ -26,10 +27,14 @@ rb_insert_callchain(struct rb_root *root, struct callchain_node *chain,
 	struct rb_node **p = &root->rb_node;
 	struct rb_node *parent = NULL;
 	struct callchain_node *rnode;
+	u64 chain_cumul = cumul_hits(chain);
 
 	while (*p) {
+		u64 rnode_cumul;
+
 		parent = *p;
 		rnode = rb_entry(parent, struct callchain_node, rb_node);
+		rnode_cumul = cumul_hits(rnode);
 
 		switch (mode) {
 		case CHAIN_FLAT:
@@ -40,7 +45,7 @@ rb_insert_callchain(struct rb_root *root, struct callchain_node *chain,
 			break;
 		case CHAIN_GRAPH_ABS: /* Falldown */
 		case CHAIN_GRAPH_REL:
-			if (rnode->cumul_hit < chain->cumul_hit)
+			if (rnode_cumul < chain_cumul)
 				p = &(*p)->rb_left;
 			else
 				p = &(*p)->rb_right;
@@ -87,7 +92,7 @@ static void __sort_chain_graph_abs(struct callchain_node *node,
 
 	chain_for_each_child(child, node) {
 		__sort_chain_graph_abs(child, min_hit);
-		if (child->cumul_hit >= min_hit)
+		if (cumul_hits(child) >= min_hit)
 			rb_insert_callchain(&node->rb_root, child,
 					    CHAIN_GRAPH_ABS);
 	}
@@ -108,11 +113,11 @@ static void __sort_chain_graph_rel(struct callchain_node *node,
 	u64 min_hit;
 
 	node->rb_root = RB_ROOT;
-	min_hit = node->cumul_hit * min_percent / 100.0;
+	min_hit = ceil(node->children_hit * min_percent);
 
 	chain_for_each_child(child, node) {
 		__sort_chain_graph_rel(child, min_percent);
-		if (child->cumul_hit >= min_hit)
+		if (cumul_hits(child) >= min_hit)
 			rb_insert_callchain(&node->rb_root, child,
 					    CHAIN_GRAPH_REL);
 	}
@@ -122,7 +127,7 @@ static void
 sort_chain_graph_rel(struct rb_root *rb_root, struct callchain_node *chain_root,
 		     u64 min_hit __used, struct callchain_param *param)
 {
-	__sort_chain_graph_rel(chain_root, param->min_percent);
+	__sort_chain_graph_rel(chain_root, param->min_percent / 100.0);
 	rb_root->rb_node = chain_root->rb_root.rb_node;
 }
 
@@ -211,7 +216,8 @@ add_child(struct callchain_node *parent, struct ip_callchain *chain,
 	new = create_child(parent, false);
 	fill_node(new, chain, start, syms);
 
-	new->cumul_hit = new->hit = 1;
+	new->children_hit = 0;
+	new->hit = 1;
 }
 
 /*
@@ -241,7 +247,8 @@ split_add_child(struct callchain_node *parent, struct ip_callchain *chain,
 
 	/* split the hits */
 	new->hit = parent->hit;
-	new->cumul_hit = parent->cumul_hit;
+	new->children_hit = parent->children_hit;
+	parent->children_hit = cumul_hits(new);
 	new->val_nr = parent->val_nr - idx_local;
 	parent->val_nr = idx_local;
 
@@ -249,6 +256,7 @@ split_add_child(struct callchain_node *parent, struct ip_callchain *chain,
 	if (idx_total < chain->nr) {
 		parent->hit = 0;
 		add_child(parent, chain, idx_total, syms);
+		parent->children_hit++;
 	} else {
 		parent->hit = 1;
 	}
@@ -269,13 +277,13 @@ __append_chain_children(struct callchain_node *root, struct ip_callchain *chain,
 		unsigned int ret = __append_chain(rnode, chain, start, syms);
 
 		if (!ret)
-			goto cumul;
+			goto inc_children_hit;
 	}
 	/* nothing in children, add to the current node */
 	add_child(root, chain, start, syms);
 
-cumul:
-	root->cumul_hit++;
+inc_children_hit:
+	root->children_hit++;
 }
 
 static int
@@ -317,8 +325,6 @@ __append_chain(struct callchain_node *root, struct ip_callchain *chain,
 	/* we match 100% of the path, increment the hit */
 	if (i - start == root->val_nr && i == chain->nr) {
 		root->hit++;
-		root->cumul_hit++;
-
 		return 0;
 	}
 
@@ -331,5 +337,7 @@ __append_chain(struct callchain_node *root, struct ip_callchain *chain,
 void append_chain(struct callchain_node *root, struct ip_callchain *chain,
 		  struct symbol **syms)
 {
+	if (!chain->nr)
+		return;
 	__append_chain_children(root, chain, syms, 0);
 }

tools/perf/util/callchain.h

@@ -7,6 +7,7 @@
 #include "symbol.h"
 
 enum chain_mode {
+	CHAIN_NONE,
 	CHAIN_FLAT,
 	CHAIN_GRAPH_ABS,
 	CHAIN_GRAPH_REL
@@ -21,7 +22,7 @@ struct callchain_node {
 	struct rb_root		rb_root; /* sorted tree of children */
 	unsigned int		val_nr;
 	u64			hit;
-	u64			cumul_hit; /* hit + hits of children */
+	u64			children_hit;
 };
 
 struct callchain_param;
@@ -48,6 +49,11 @@ static inline void callchain_init(struct callchain_node *node)
 	INIT_LIST_HEAD(&node->val);
 }
 
+static inline u64 cumul_hits(struct callchain_node *node)
+{
+	return node->hit + node->children_hit;
+}
+
 int register_callchain_param(struct callchain_param *param);
 void append_chain(struct callchain_node *root, struct ip_callchain *chain,
 		  struct symbol **syms);

tools/perf/util/header.c

@@ -185,6 +185,8 @@ static void do_read(int fd, void *buf, size_t size)
 
 		if (ret < 0)
 			die("failed to read");
+		if (ret == 0)
+			die("failed to read: missing data");
 
 		size -= ret;
 		buf += ret;
@@ -213,9 +215,10 @@ struct perf_header *perf_header__read(int fd)
 
 	for (i = 0; i < nr_attrs; i++) {
 		struct perf_header_attr *attr;
-		off_t tmp = lseek(fd, 0, SEEK_CUR);
+		off_t tmp;
 
 		do_read(fd, &f_attr, sizeof(f_attr));
+		tmp = lseek(fd, 0, SEEK_CUR);
 
 		attr = perf_header_attr__new(&f_attr.attr);
 

tools/perf/util/parse-events.c

@@ -121,13 +121,29 @@ static unsigned long hw_cache_stat[C(MAX)] = {
 	   (strcmp(sys_dirent.d_name, ".")) &&				       \
 	   (strcmp(sys_dirent.d_name, "..")))
 
+static int tp_event_has_id(struct dirent *sys_dir, struct dirent *evt_dir)
+{
+	char evt_path[MAXPATHLEN];
+	int fd;
+
+	snprintf(evt_path, MAXPATHLEN, "%s/%s/%s/id", debugfs_path,
+			sys_dir->d_name, evt_dir->d_name);
+	fd = open(evt_path, O_RDONLY);
+	if (fd < 0)
+		return -EINVAL;
+	close(fd);
+
+	return 0;
+}
+
 #define for_each_event(sys_dirent, evt_dir, evt_dirent, evt_next, file, st)    \
 	while (!readdir_r(evt_dir, &evt_dirent, &evt_next) && evt_next)        \
 	if (snprintf(file, MAXPATHLEN, "%s/%s/%s", debugfs_path,	       \
 		     sys_dirent.d_name, evt_dirent.d_name) &&		       \
 	   (!stat(file, &st)) && (S_ISDIR(st.st_mode)) &&		       \
 	   (strcmp(evt_dirent.d_name, ".")) &&				       \
-	   (strcmp(evt_dirent.d_name, "..")))
+	   (strcmp(evt_dirent.d_name, "..")) &&				       \
+	   (!tp_event_has_id(&sys_dirent, &evt_dirent)))
 
 #define MAX_EVENT_LENGTH 30
 
@@ -223,9 +239,15 @@ char *event_name(int counter)
 {
 	u64 config = attrs[counter].config;
 	int type = attrs[counter].type;
+
+	return __event_name(type, config);
+}
+
+char *__event_name(int type, u64 config)
+{
 	static char buf[32];
 
-	if (attrs[counter].type == PERF_TYPE_RAW) {
+	if (type == PERF_TYPE_RAW) {
 		sprintf(buf, "raw 0x%llx", config);
 		return buf;
 	}

tools/perf/util/parse-events.h

@@ -10,6 +10,7 @@ extern int			nr_counters;
 extern struct perf_counter_attr attrs[MAX_COUNTERS];
 
 extern char *event_name(int ctr);
+extern char *__event_name(int type, u64 config);
 
 extern int parse_events(const struct option *opt, const char *str, int unset);
 

tools/perf/util/symbol.c

@@ -24,6 +24,16 @@ const char *sym_hist_filter;
 #define DMGL_ANSI        (1 << 1)       /* Include const, volatile, etc */
 #endif
 
+enum dso_origin {
+	DSO__ORIG_KERNEL = 0,
+	DSO__ORIG_JAVA_JIT,
+	DSO__ORIG_FEDORA,
+	DSO__ORIG_UBUNTU,
+	DSO__ORIG_BUILDID,
+	DSO__ORIG_DSO,
+	DSO__ORIG_NOT_FOUND,
+};
+
 static struct symbol *symbol__new(u64 start, u64 len,
 				  const char *name, unsigned int priv_size,
 				  u64 obj_start, int verbose)
@@ -81,6 +91,7 @@ struct dso *dso__new(const char *name, unsigned int sym_priv_size)
 		self->sym_priv_size = sym_priv_size;
 		self->find_symbol = dso__find_symbol;
 		self->slen_calculated = 0;
+		self->origin = DSO__ORIG_NOT_FOUND;
 	}
 
 	return self;
@@ -710,7 +721,7 @@ static char *dso__read_build_id(struct dso *self, int verbose)
 		++raw;
 		bid += 2;
 	}
-	if (verbose)
+	if (verbose >= 2)
 		printf("%s(%s): %s\n", __func__, self->name, build_id);
 out_elf_end:
 	elf_end(elf);
@@ -720,11 +731,26 @@ static char *dso__read_build_id(struct dso *self, int verbose)
 	return build_id;
 }
 
+char dso__symtab_origin(const struct dso *self)
+{
+	static const char origin[] = {
+		[DSO__ORIG_KERNEL] =   'k',
+		[DSO__ORIG_JAVA_JIT] = 'j',
+		[DSO__ORIG_FEDORA] =   'f',
+		[DSO__ORIG_UBUNTU] =   'u',
+		[DSO__ORIG_BUILDID] =  'b',
+		[DSO__ORIG_DSO] =      'd',
+	};
+
+	if (self == NULL || self->origin == DSO__ORIG_NOT_FOUND)
+		return '!';
+	return origin[self->origin];
+}
+
 int dso__load(struct dso *self, symbol_filter_t filter, int verbose)
 {
 	int size = PATH_MAX;
 	char *name = malloc(size), *build_id = NULL;
-	int variant = 0;
 	int ret = -1;
 	int fd;
 
@@ -733,19 +759,26 @@ int dso__load(struct dso *self, symbol_filter_t filter, int verbose)
 
 	self->adjust_symbols = 0;
 
-	if (strncmp(self->name, "/tmp/perf-", 10) == 0)
-		return dso__load_perf_map(self, filter, verbose);
+	if (strncmp(self->name, "/tmp/perf-", 10) == 0) {
+		ret = dso__load_perf_map(self, filter, verbose);
+		self->origin = ret > 0 ? DSO__ORIG_JAVA_JIT :
+					 DSO__ORIG_NOT_FOUND;
+		return ret;
+	}
+
+	self->origin = DSO__ORIG_FEDORA - 1;
 
 more:
 	do {
-		switch (variant) {
-		case 0: /* Fedora */
+		self->origin++;
+		switch (self->origin) {
+		case DSO__ORIG_FEDORA:
 			snprintf(name, size, "/usr/lib/debug%s.debug", self->name);
 			break;
-		case 1: /* Ubuntu */
+		case DSO__ORIG_UBUNTU:
 			snprintf(name, size, "/usr/lib/debug%s", self->name);
 			break;
-		case 2:
+		case DSO__ORIG_BUILDID:
 			build_id = dso__read_build_id(self, verbose);
 			if (build_id != NULL) {
 				snprintf(name, size,
@@ -754,16 +787,15 @@ int dso__load(struct dso *self, symbol_filter_t filter, int verbose)
 				free(build_id);
 				break;
 			}
-			variant++;
+			self->origin++;
 			/* Fall thru */
-		case 3: /* Sane people */
+		case DSO__ORIG_DSO:
 			snprintf(name, size, "%s", self->name);
 			break;
 
 		default:
 			goto out;
 		}
-		variant++;
 
 		fd = open(name, O_RDONLY);
 	} while (fd < 0);
@@ -899,6 +931,9 @@ int dso__load_kernel(struct dso *self, const char *vmlinux,
 	if (err <= 0)
 		err = dso__load_kallsyms(self, filter, verbose);
 
+	if (err > 0)
+		self->origin = DSO__ORIG_KERNEL;
+
 	return err;
 }
 

tools/perf/util/symbol.h

@@ -26,6 +26,7 @@ struct dso {
 	unsigned int	 sym_priv_size;
 	unsigned char	 adjust_symbols;
 	unsigned char	 slen_calculated;
+	unsigned char	 origin;
 	char		 name[0];
 };
 
@@ -49,6 +50,7 @@ int dso__load_modules(struct dso *self, symbol_filter_t filter, int verbose);
 int dso__load(struct dso *self, symbol_filter_t filter, int verbose);
 
 size_t dso__fprintf(struct dso *self, FILE *fp);
+char dso__symtab_origin(const struct dso *self);
 
 void symbol__init(void);
 #endif /* _PERF_SYMBOL_ */