From 141eba2e006dd8145bed2e49fae3de5af65ab9b0 Mon Sep 17 00:00:00 2001 From: Stephen Warren Date: Thu, 24 May 2012 10:47:26 -0600 Subject: regmap: allow busses to request formatting with specific endianness Add a field to struct regmap_bus that allows bus drivers to request that register addresses and values be formatted with a specific endianness. The default endianness is unchanged from current operation: Big. Implement native endian formatting/parsing for 16- and 32-bit values. This will be enough to support regmap-mmio.c. Signed-off-by: Stephen Warren Signed-off-by: Mark Brown --- drivers/base/regmap/regmap.c | 101 ++++++++++++++++++++++++++++++++++++++----- 1 file changed, 90 insertions(+), 11 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/regmap/regmap.c b/drivers/base/regmap/regmap.c index 0bcda488f11c..1cf721421bec 100644 --- a/drivers/base/regmap/regmap.c +++ b/drivers/base/regmap/regmap.c @@ -119,13 +119,19 @@ static void regmap_format_8(void *buf, unsigned int val, unsigned int shift) b[0] = val << shift; } -static void regmap_format_16(void *buf, unsigned int val, unsigned int shift) +static void regmap_format_16_be(void *buf, unsigned int val, unsigned int shift) { __be16 *b = buf; b[0] = cpu_to_be16(val << shift); } +static void regmap_format_16_native(void *buf, unsigned int val, + unsigned int shift) +{ + *(u16 *)buf = val << shift; +} + static void regmap_format_24(void *buf, unsigned int val, unsigned int shift) { u8 *b = buf; @@ -137,13 +143,19 @@ static void regmap_format_24(void *buf, unsigned int val, unsigned int shift) b[2] = val; } -static void regmap_format_32(void *buf, unsigned int val, unsigned int shift) +static void regmap_format_32_be(void *buf, unsigned int val, unsigned int shift) { __be32 *b = buf; b[0] = cpu_to_be32(val << shift); } +static void regmap_format_32_native(void *buf, unsigned int val, + unsigned int shift) +{ + *(u32 *)buf = val << shift; +} + static unsigned int regmap_parse_8(void *buf) { u8 *b = buf; @@ -151,7 +163,7 @@ static unsigned int regmap_parse_8(void *buf) return b[0]; } -static unsigned int regmap_parse_16(void *buf) +static unsigned int regmap_parse_16_be(void *buf) { __be16 *b = buf; @@ -160,6 +172,11 @@ static unsigned int regmap_parse_16(void *buf) return b[0]; } +static unsigned int regmap_parse_16_native(void *buf) +{ + return *(u16 *)buf; +} + static unsigned int regmap_parse_24(void *buf) { u8 *b = buf; @@ -170,7 +187,7 @@ static unsigned int regmap_parse_24(void *buf) return ret; } -static unsigned int regmap_parse_32(void *buf) +static unsigned int regmap_parse_32_be(void *buf) { __be32 *b = buf; @@ -179,6 +196,11 @@ static unsigned int regmap_parse_32(void *buf) return b[0]; } +static unsigned int regmap_parse_32_native(void *buf) +{ + return *(u32 *)buf; +} + static void regmap_lock_mutex(struct regmap *map) { mutex_lock(&map->mutex); @@ -227,6 +249,7 @@ struct regmap *regmap_init(struct device *dev, { struct regmap *map, **m; int ret = -EINVAL; + enum regmap_endian reg_endian, val_endian; if (!bus || !config) goto err; @@ -275,6 +298,18 @@ struct regmap *regmap_init(struct device *dev, map->read_flag_mask = bus->read_flag_mask; } + reg_endian = config->reg_format_endian; + if (reg_endian == REGMAP_ENDIAN_DEFAULT) + reg_endian = bus->reg_format_endian_default; + if (reg_endian == REGMAP_ENDIAN_DEFAULT) + reg_endian = REGMAP_ENDIAN_BIG; + + val_endian = config->val_format_endian; + if (val_endian == REGMAP_ENDIAN_DEFAULT) + val_endian = bus->val_format_endian_default; + if (val_endian == REGMAP_ENDIAN_DEFAULT) + val_endian = REGMAP_ENDIAN_BIG; + switch (config->reg_bits + map->reg_shift) { case 2: switch (config->val_bits) { @@ -321,11 +356,29 @@ struct regmap *regmap_init(struct device *dev, break; case 16: - map->format.format_reg = regmap_format_16; + switch (reg_endian) { + case REGMAP_ENDIAN_BIG: + map->format.format_reg = regmap_format_16_be; + break; + case REGMAP_ENDIAN_NATIVE: + map->format.format_reg = regmap_format_16_native; + break; + default: + goto err_map; + } break; case 32: - map->format.format_reg = regmap_format_32; + switch (reg_endian) { + case REGMAP_ENDIAN_BIG: + map->format.format_reg = regmap_format_32_be; + break; + case REGMAP_ENDIAN_NATIVE: + map->format.format_reg = regmap_format_32_native; + break; + default: + goto err_map; + } break; default: @@ -338,21 +391,47 @@ struct regmap *regmap_init(struct device *dev, map->format.parse_val = regmap_parse_8; break; case 16: - map->format.format_val = regmap_format_16; - map->format.parse_val = regmap_parse_16; + switch (val_endian) { + case REGMAP_ENDIAN_BIG: + map->format.format_val = regmap_format_16_be; + map->format.parse_val = regmap_parse_16_be; + break; + case REGMAP_ENDIAN_NATIVE: + map->format.format_val = regmap_format_16_native; + map->format.parse_val = regmap_parse_16_native; + break; + default: + goto err_map; + } break; case 24: + if (val_endian != REGMAP_ENDIAN_BIG) + goto err_map; map->format.format_val = regmap_format_24; map->format.parse_val = regmap_parse_24; break; case 32: - map->format.format_val = regmap_format_32; - map->format.parse_val = regmap_parse_32; + switch (val_endian) { + case REGMAP_ENDIAN_BIG: + map->format.format_val = regmap_format_32_be; + map->format.parse_val = regmap_parse_32_be; + break; + case REGMAP_ENDIAN_NATIVE: + map->format.format_val = regmap_format_32_native; + map->format.parse_val = regmap_parse_32_native; + break; + default: + goto err_map; + } break; } - if (map->format.format_write) + if (map->format.format_write) { + if ((reg_endian != REGMAP_ENDIAN_BIG) || + (val_endian != REGMAP_ENDIAN_BIG)) + goto err_map; map->use_single_rw = true; + } if (!map->format.format_write && !(map->format.format_reg && map->format.format_val)) -- cgit v1.2.3 From 6a55244e897d32952832a67cb35cfbfa3f722c50 Mon Sep 17 00:00:00 2001 From: Stephen Warren Date: Thu, 24 May 2012 10:47:27 -0600 Subject: regmap: mmio: request native endian formatting This will avoid the regmap core converting all addresses and values into big endian, only for the mmio bus driver to have to convert them back to native endian. Signed-off-by: Stephen Warren Signed-off-by: Mark Brown --- drivers/base/regmap/regmap-mmio.c | 26 ++++++++++++++++++-------- 1 file changed, 18 insertions(+), 8 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/regmap/regmap-mmio.c b/drivers/base/regmap/regmap-mmio.c index febd6de6c8ac..eec86639cac4 100644 --- a/drivers/base/regmap/regmap-mmio.c +++ b/drivers/base/regmap/regmap-mmio.c @@ -37,7 +37,7 @@ static int regmap_mmio_gather_write(void *context, BUG_ON(reg_size != 4); - offset = be32_to_cpup(reg); + offset = *(u32 *)reg; while (val_size) { switch (ctx->val_bytes) { @@ -45,14 +45,14 @@ static int regmap_mmio_gather_write(void *context, writeb(*(u8 *)val, ctx->regs + offset); break; case 2: - writew(be16_to_cpup(val), ctx->regs + offset); + writew(*(u16 *)val, ctx->regs + offset); break; case 4: - writel(be32_to_cpup(val), ctx->regs + offset); + writel(*(u32 *)val, ctx->regs + offset); break; #ifdef CONFIG_64BIT case 8: - writeq(be64_to_cpup(val), ctx->regs + offset); + writeq(*(u64 *)val, ctx->regs + offset); break; #endif default: @@ -83,7 +83,7 @@ static int regmap_mmio_read(void *context, BUG_ON(reg_size != 4); - offset = be32_to_cpup(reg); + offset = *(u32 *)reg; while (val_size) { switch (ctx->val_bytes) { @@ -91,14 +91,14 @@ static int regmap_mmio_read(void *context, *(u8 *)val = readb(ctx->regs + offset); break; case 2: - *(u16 *)val = cpu_to_be16(readw(ctx->regs + offset)); + *(u16 *)val = readw(ctx->regs + offset); break; case 4: - *(u32 *)val = cpu_to_be32(readl(ctx->regs + offset)); + *(u32 *)val = readl(ctx->regs + offset); break; #ifdef CONFIG_64BIT case 8: - *(u64 *)val = cpu_to_be32(readq(ctx->regs + offset)); + *(u64 *)val = readq(ctx->regs + offset); break; #endif default: @@ -124,6 +124,8 @@ static struct regmap_bus regmap_mmio = { .gather_write = regmap_mmio_gather_write, .read = regmap_mmio_read, .free_context = regmap_mmio_free_context, + .reg_format_endian_default = REGMAP_ENDIAN_NATIVE, + .val_format_endian_default = REGMAP_ENDIAN_NATIVE, }; struct regmap_mmio_context *regmap_mmio_gen_context(void __iomem *regs, @@ -162,6 +164,14 @@ struct regmap_mmio_context *regmap_mmio_gen_context(void __iomem *regs, if (config->reg_stride < min_stride) return ERR_PTR(-EINVAL); + switch (config->reg_format_endian) { + case REGMAP_ENDIAN_DEFAULT: + case REGMAP_ENDIAN_NATIVE: + break; + default: + return ERR_PTR(-EINVAL); + } + ctx = kzalloc(GFP_KERNEL, sizeof(*ctx)); if (!ctx) return ERR_PTR(-ENOMEM); -- cgit v1.2.3 From b026ddbbd25e3560c8d69beb96a5980d96c59b43 Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Thu, 31 May 2012 21:01:46 +0100 Subject: regmap: Constify regmap_irq_chip We should never be modifying it and it lets drivers declare it const. Signed-off-by: Mark Brown --- drivers/base/regmap/regmap-irq.c | 6 +++--- include/linux/regmap.h | 2 +- 2 files changed, 4 insertions(+), 4 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/regmap/regmap-irq.c b/drivers/base/regmap/regmap-irq.c index 4fac4b9be88f..c190229daa59 100644 --- a/drivers/base/regmap/regmap-irq.c +++ b/drivers/base/regmap/regmap-irq.c @@ -24,7 +24,7 @@ struct regmap_irq_chip_data { struct mutex lock; struct regmap *map; - struct regmap_irq_chip *chip; + const struct regmap_irq_chip *chip; int irq_base; struct irq_domain *domain; @@ -103,7 +103,7 @@ static struct irq_chip regmap_irq_chip = { static irqreturn_t regmap_irq_thread(int irq, void *d) { struct regmap_irq_chip_data *data = d; - struct regmap_irq_chip *chip = data->chip; + const struct regmap_irq_chip *chip = data->chip; struct regmap *map = data->map; int ret, i; bool handled = false; @@ -195,7 +195,7 @@ static struct irq_domain_ops regmap_domain_ops = { * register values used by the IRQ controller over suspend and resume. */ int regmap_add_irq_chip(struct regmap *map, int irq, int irq_flags, - int irq_base, struct regmap_irq_chip *chip, + int irq_base, const struct regmap_irq_chip *chip, struct regmap_irq_chip_data **data) { struct regmap_irq_chip_data *d; diff --git a/include/linux/regmap.h b/include/linux/regmap.h index 26136b577009..f9b624c83464 100644 --- a/include/linux/regmap.h +++ b/include/linux/regmap.h @@ -268,7 +268,7 @@ struct regmap_irq_chip { struct regmap_irq_chip_data; int regmap_add_irq_chip(struct regmap *map, int irq, int irq_flags, - int irq_base, struct regmap_irq_chip *chip, + int irq_base, const struct regmap_irq_chip *chip, struct regmap_irq_chip_data **data); void regmap_del_irq_chip(int irq, struct regmap_irq_chip_data *data); int regmap_irq_chip_get_base(struct regmap_irq_chip_data *data); -- cgit v1.2.3 From bfd6185ddecc6e6f6bd654c053c307c9e49ca391 Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Tue, 5 Jun 2012 14:29:36 +0100 Subject: regmap: Don't try to map non-existant IRQs If the driver supplied an empty entry in the array of IRQs then return an error rather than trying to do the mapping. This is intended for use with handling chip variants and similar situations. Signed-off-by: Mark Brown --- drivers/base/regmap/regmap-irq.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'drivers/base') diff --git a/drivers/base/regmap/regmap-irq.c b/drivers/base/regmap/regmap-irq.c index 4fac4b9be88f..b74e14c4dff4 100644 --- a/drivers/base/regmap/regmap-irq.c +++ b/drivers/base/regmap/regmap-irq.c @@ -346,6 +346,10 @@ EXPORT_SYMBOL_GPL(regmap_irq_chip_get_base); */ int regmap_irq_get_virq(struct regmap_irq_chip_data *data, int irq) { + /* Handle holes in the IRQ list */ + if (!data->chip->irqs[irq].mask) + return -EINVAL; + return irq_create_mapping(data->domain, irq); } EXPORT_SYMBOL_GPL(regmap_irq_get_virq); -- cgit v1.2.3 From a43fd50dc99a5f65505f174eca5a421707d73b4c Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Tue, 5 Jun 2012 14:34:03 +0100 Subject: regmap: Implement support for wake IRQs Allow chips to provide a bank of registers for controlling the wake state in a similar fashion to the masks and propagate the wake count to the parent interrupt controller. Signed-off-by: Mark Brown --- drivers/base/regmap/regmap-irq.c | 47 ++++++++++++++++++++++++++++++++++++++++ include/linux/regmap.h | 2 ++ 2 files changed, 49 insertions(+) (limited to 'drivers/base') diff --git a/drivers/base/regmap/regmap-irq.c b/drivers/base/regmap/regmap-irq.c index b74e14c4dff4..b480b529f020 100644 --- a/drivers/base/regmap/regmap-irq.c +++ b/drivers/base/regmap/regmap-irq.c @@ -29,9 +29,13 @@ struct regmap_irq_chip_data { int irq_base; struct irq_domain *domain; + int irq; + int wake_count; + unsigned int *status_buf; unsigned int *mask_buf; unsigned int *mask_buf_def; + unsigned int *wake_buf; unsigned int irq_reg_stride; }; @@ -71,6 +75,16 @@ static void regmap_irq_sync_unlock(struct irq_data *data) d->chip->mask_base + (i * map->reg_stride)); } + /* If we've changed our wakeup count propagate it to the parent */ + if (d->wake_count < 0) + for (i = d->wake_count; i < 0; i++) + irq_set_irq_wake(d->irq, 0); + else if (d->wake_count > 0) + for (i = 0; i < d->wake_count; i++) + irq_set_irq_wake(d->irq, 1); + + d->wake_count = 0; + mutex_unlock(&d->lock); } @@ -92,12 +106,35 @@ static void regmap_irq_disable(struct irq_data *data) d->mask_buf[irq_data->reg_offset / map->reg_stride] |= irq_data->mask; } +static int regmap_irq_set_wake(struct irq_data *data, unsigned int on) +{ + struct regmap_irq_chip_data *d = irq_data_get_irq_chip_data(data); + struct regmap *map = d->map; + const struct regmap_irq *irq_data = irq_to_regmap_irq(d, data->hwirq); + + if (!d->chip->wake_base) + return -EINVAL; + + if (on) { + d->wake_buf[irq_data->reg_offset / map->reg_stride] + &= ~irq_data->mask; + d->wake_count++; + } else { + d->wake_buf[irq_data->reg_offset / map->reg_stride] + |= irq_data->mask; + d->wake_count--; + } + + return 0; +} + static struct irq_chip regmap_irq_chip = { .name = "regmap", .irq_bus_lock = regmap_irq_lock, .irq_bus_sync_unlock = regmap_irq_sync_unlock, .irq_disable = regmap_irq_disable, .irq_enable = regmap_irq_enable, + .irq_set_wake = regmap_irq_set_wake, }; static irqreturn_t regmap_irq_thread(int irq, void *d) @@ -240,6 +277,14 @@ int regmap_add_irq_chip(struct regmap *map, int irq, int irq_flags, if (!d->mask_buf_def) goto err_alloc; + if (chip->wake_base) { + d->wake_buf = kzalloc(sizeof(unsigned int) * chip->num_regs, + GFP_KERNEL); + if (!d->wake_buf) + goto err_alloc; + } + + d->irq = irq; d->map = map; d->chip = chip; d->irq_base = irq_base; @@ -294,6 +339,7 @@ int regmap_add_irq_chip(struct regmap *map, int irq, int irq_flags, err_domain: /* Should really dispose of the domain but... */ err_alloc: + kfree(d->wake_buf); kfree(d->mask_buf_def); kfree(d->mask_buf); kfree(d->status_buf); @@ -315,6 +361,7 @@ void regmap_del_irq_chip(int irq, struct regmap_irq_chip_data *d) free_irq(irq, d); /* We should unmap the domain but... */ + kfree(d->wake_buf); kfree(d->mask_buf_def); kfree(d->mask_buf); kfree(d->status_buf); diff --git a/include/linux/regmap.h b/include/linux/regmap.h index 56af22ec9aba..58ec0cba0ae6 100644 --- a/include/linux/regmap.h +++ b/include/linux/regmap.h @@ -219,6 +219,7 @@ struct regmap_irq { * @status_base: Base status register address. * @mask_base: Base mask register address. * @ack_base: Base ack address. If zero then the chip is clear on read. + * @wake_base: Base address for wake enables. If zero unsupported. * @irq_reg_stride: Stride to use for chips where registers are not contiguous. * * @num_regs: Number of registers in each control bank. @@ -232,6 +233,7 @@ struct regmap_irq_chip { unsigned int status_base; unsigned int mask_base; unsigned int ack_base; + unsigned int wake_base; unsigned int irq_reg_stride; int num_regs; -- cgit v1.2.3 From bdd4034df8b37841eeaf7b05f86e732ab8e0b08a Mon Sep 17 00:00:00 2001 From: Rabin Vincent Date: Mon, 23 Apr 2012 09:16:36 +0200 Subject: driver core: always handle dpm_order If !dev->class, device_move() does not respect the dpm_order. Fix it to do so. Acked-by: Rafael J. Wysocki Signed-off-by: Rabin Vincent Reviewed-by: Srinidhi Kasagar [Fixed a small dangling label compile warning] Signed-off-by: Linus Walleij Signed-off-by: Greg Kroah-Hartman --- drivers/base/core.c | 36 ++++++++++++++++++------------------ 1 file changed, 18 insertions(+), 18 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/core.c b/drivers/base/core.c index 346be8b78b24..846d12b5d579 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -1754,25 +1754,25 @@ int device_move(struct device *dev, struct device *new_parent, set_dev_node(dev, dev_to_node(new_parent)); } - if (!dev->class) - goto out_put; - error = device_move_class_links(dev, old_parent, new_parent); - if (error) { - /* We ignore errors on cleanup since we're hosed anyway... */ - device_move_class_links(dev, new_parent, old_parent); - if (!kobject_move(&dev->kobj, &old_parent->kobj)) { - if (new_parent) - klist_remove(&dev->p->knode_parent); - dev->parent = old_parent; - if (old_parent) { - klist_add_tail(&dev->p->knode_parent, - &old_parent->p->klist_children); - set_dev_node(dev, dev_to_node(old_parent)); + if (dev->class) { + error = device_move_class_links(dev, old_parent, new_parent); + if (error) { + /* We ignore errors on cleanup since we're hosed anyway... */ + device_move_class_links(dev, new_parent, old_parent); + if (!kobject_move(&dev->kobj, &old_parent->kobj)) { + if (new_parent) + klist_remove(&dev->p->knode_parent); + dev->parent = old_parent; + if (old_parent) { + klist_add_tail(&dev->p->knode_parent, + &old_parent->p->klist_children); + set_dev_node(dev, dev_to_node(old_parent)); + } } + cleanup_glue_dir(dev, new_parent_kobj); + put_device(new_parent); + goto out; } - cleanup_glue_dir(dev, new_parent_kobj); - put_device(new_parent); - goto out; } switch (dpm_order) { case DPM_ORDER_NONE: @@ -1787,7 +1787,7 @@ int device_move(struct device *dev, struct device *new_parent, device_pm_move_last(dev); break; } -out_put: + put_device(old_parent); out: device_pm_unlock(); -- cgit v1.2.3 From e8790ab4ce2a46505a8e1479645414425dbc81f0 Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Wed, 13 Jun 2012 22:57:24 +0800 Subject: regmap: mmio: Staticize regmap_mmio_gen_context() regmap_mmio_gen_context() is only used in regmap-mmio.c. Thus make it static. Signed-off-by: Axel Lin Acked-by: Stephen Warren Signed-off-by: Mark Brown --- drivers/base/regmap/regmap-mmio.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/base') diff --git a/drivers/base/regmap/regmap-mmio.c b/drivers/base/regmap/regmap-mmio.c index eec86639cac4..d64a7fc71df8 100644 --- a/drivers/base/regmap/regmap-mmio.c +++ b/drivers/base/regmap/regmap-mmio.c @@ -128,7 +128,7 @@ static struct regmap_bus regmap_mmio = { .val_format_endian_default = REGMAP_ENDIAN_NATIVE, }; -struct regmap_mmio_context *regmap_mmio_gen_context(void __iomem *regs, +static struct regmap_mmio_context *regmap_mmio_gen_context(void __iomem *regs, const struct regmap_config *config) { struct regmap_mmio_context *ctx; -- cgit v1.2.3 From 0998d0631001288a5974afc0b2a5f568bcdecb4d Mon Sep 17 00:00:00 2001 From: Hans de Goede Date: Wed, 23 May 2012 00:09:34 +0200 Subject: device-core: Ensure drvdata = NULL when no driver is bound 1) drvdata is for a driver to store a pointer to driver specific data 2) If no driver is bound, there is no driver specific data associated with the device 3) Thus logically drvdata should be NULL if no driver is bound. But many drivers don't clear drvdata on device_release, or set drvdata early on in probe and leave it set on probe error. Both of which results in a dangling pointer in drvdata. This patch enforce for drvdata to be NULL after device_release or on probe failure. Signed-off-by: Hans de Goede Signed-off-by: Greg Kroah-Hartman --- drivers/base/dd.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'drivers/base') diff --git a/drivers/base/dd.c b/drivers/base/dd.c index 1b1cbb571d38..9a1e9704d782 100644 --- a/drivers/base/dd.c +++ b/drivers/base/dd.c @@ -283,6 +283,7 @@ probe_failed: devres_release_all(dev); driver_sysfs_remove(dev); dev->driver = NULL; + dev_set_drvdata(dev, NULL); if (ret == -EPROBE_DEFER) { /* Driver requested deferred probing */ @@ -487,6 +488,7 @@ static void __device_release_driver(struct device *dev) drv->remove(dev); devres_release_all(dev); dev->driver = NULL; + dev_set_drvdata(dev, NULL); klist_remove(&dev->p->knode_driver); if (dev->bus) blocking_notifier_call_chain(&dev->bus->p->bus_notifier, -- cgit v1.2.3 From 6e7b4a59b3d7bb2dcd11c019354bf0c91037dadd Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sat, 9 Jun 2012 15:02:59 -0700 Subject: driver core: fix some kernel-doc warnings in dma*.c Fix kernel-doc warnings in drivers/base/dma*.c: Warning(drivers/base/dma-buf.c:498): No description found for parameter 'vaddr' Warning(drivers/base/dma-coherent.c:199): No description found for parameter 'ret' Signed-off-by: Randy Dunlap Signed-off-by: Greg Kroah-Hartman --- drivers/base/dma-buf.c | 1 + drivers/base/dma-coherent.c | 1 + 2 files changed, 2 insertions(+) (limited to 'drivers/base') diff --git a/drivers/base/dma-buf.c b/drivers/base/dma-buf.c index 24e88fe29ec1..c30f3e1d0efc 100644 --- a/drivers/base/dma-buf.c +++ b/drivers/base/dma-buf.c @@ -493,6 +493,7 @@ EXPORT_SYMBOL_GPL(dma_buf_vmap); /** * dma_buf_vunmap - Unmap a vmap obtained by dma_buf_vmap. * @dmabuf: [in] buffer to vunmap + * @vaddr: [in] vmap to vunmap */ void dma_buf_vunmap(struct dma_buf *dmabuf, void *vaddr) { diff --git a/drivers/base/dma-coherent.c b/drivers/base/dma-coherent.c index 1b85949e3d2f..560a7173f810 100644 --- a/drivers/base/dma-coherent.c +++ b/drivers/base/dma-coherent.c @@ -186,6 +186,7 @@ EXPORT_SYMBOL(dma_release_from_coherent); * @vma: vm_area for the userspace memory * @vaddr: cpu address returned by dma_alloc_from_coherent * @size: size of the memory buffer allocated by dma_alloc_from_coherent + * @ret: result from remap_pfn_range() * * This checks whether the memory was allocated from the per-device * coherent memory pool and if so, maps that memory to the provided vma. -- cgit v1.2.3 From fc3ebd788e894b4dd6c9524cb3874eeeb1e862d6 Mon Sep 17 00:00:00 2001 From: Krystian Garbaciak Date: Fri, 15 Jun 2012 11:23:56 +0100 Subject: regmap: Move lock out from internal function _regmap_update_bits(). Locks are moved to regmap_update_bits(), which allows to reenter internal function _regmap_update_bits() from inside of regmap read/write routines. Signed-off-by: Krystian Garbaciak Signed-off-by: Mark Brown --- drivers/base/regmap/regmap.c | 22 ++++++++++++++-------- 1 file changed, 14 insertions(+), 8 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/regmap/regmap.c b/drivers/base/regmap/regmap.c index c89aa01fb1de..652017991da6 100644 --- a/drivers/base/regmap/regmap.c +++ b/drivers/base/regmap/regmap.c @@ -982,11 +982,9 @@ static int _regmap_update_bits(struct regmap *map, unsigned int reg, int ret; unsigned int tmp, orig; - map->lock(map); - ret = _regmap_read(map, reg, &orig); if (ret != 0) - goto out; + return ret; tmp = orig & ~mask; tmp |= val & mask; @@ -998,9 +996,6 @@ static int _regmap_update_bits(struct regmap *map, unsigned int reg, *change = false; } -out: - map->unlock(map); - return ret; } @@ -1018,7 +1013,13 @@ int regmap_update_bits(struct regmap *map, unsigned int reg, unsigned int mask, unsigned int val) { bool change; - return _regmap_update_bits(map, reg, mask, val, &change); + int ret; + + map->lock(map); + ret = _regmap_update_bits(map, reg, mask, val, &change); + map->unlock(map); + + return ret; } EXPORT_SYMBOL_GPL(regmap_update_bits); @@ -1038,7 +1039,12 @@ int regmap_update_bits_check(struct regmap *map, unsigned int reg, unsigned int mask, unsigned int val, bool *change) { - return _regmap_update_bits(map, reg, mask, val, change); + int ret; + + map->lock(map); + ret = _regmap_update_bits(map, reg, mask, val, change); + map->unlock(map); + return ret; } EXPORT_SYMBOL_GPL(regmap_update_bits_check); -- cgit v1.2.3 From 6863ca6227598d15c372f1e03449bbb4cfbcca7f Mon Sep 17 00:00:00 2001 From: Krystian Garbaciak Date: Fri, 15 Jun 2012 11:23:56 +0100 Subject: regmap: Add support for register indirect addressing. Devices with register paging or indirectly accessed registers can configure register mapping to map those on virtual address range. During access to virtually mapped register range, indirect addressing is processed automatically, in following steps: 1. selector for page or indirect register is updated (when needed); 2. register in data window is accessed. Configuration should provide minimum and maximum register for virtual range, details of selector field for page selection, minimum and maximum register of data window for indirect access. Virtual range registers are managed by cache as well as direct access registers. In order to make indirect access more efficient, selector register should be declared as non-volatile, if possible. struct regmap_config is extended with the following: struct regmap_range_cfg *ranges; unsigned int n_ranges; [Also reordered debugfs init to later on since the cleanup code was conflicting with the new cleanup code for ranges anyway -- broonie] Signed-off-by: Krystian Garbaciak Signed-off-by: Mark Brown --- drivers/base/regmap/internal.h | 17 ++++ drivers/base/regmap/regmap.c | 201 ++++++++++++++++++++++++++++++++++++++++- include/linux/regmap.h | 39 ++++++++ 3 files changed, 252 insertions(+), 5 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/regmap/internal.h b/drivers/base/regmap/internal.h index b986b8660b0c..80f9ab9c3aa4 100644 --- a/drivers/base/regmap/internal.h +++ b/drivers/base/regmap/internal.h @@ -95,6 +95,9 @@ struct regmap { /* if set, converts bulk rw to single rw */ bool use_single_rw; + + struct rb_root range_tree; + void *selector_work_buf; /* Scratch buffer used for selector */ }; struct regcache_ops { @@ -115,6 +118,20 @@ bool regmap_precious(struct regmap *map, unsigned int reg); int _regmap_write(struct regmap *map, unsigned int reg, unsigned int val); +struct regmap_range_node { + struct rb_node node; + + unsigned int range_min; + unsigned int range_max; + + unsigned int selector_reg; + unsigned int selector_mask; + int selector_shift; + + unsigned int window_start; + unsigned int window_len; +}; + #ifdef CONFIG_DEBUG_FS extern void regmap_debugfs_initcall(void); extern void regmap_debugfs_init(struct regmap *map, const char *name); diff --git a/drivers/base/regmap/regmap.c b/drivers/base/regmap/regmap.c index 652017991da6..83a0166420a4 100644 --- a/drivers/base/regmap/regmap.c +++ b/drivers/base/regmap/regmap.c @@ -15,12 +15,17 @@ #include #include #include +#include #define CREATE_TRACE_POINTS #include #include "internal.h" +static int _regmap_update_bits(struct regmap *map, unsigned int reg, + unsigned int mask, unsigned int val, + bool *change); + bool regmap_writeable(struct regmap *map, unsigned int reg) { if (map->max_register && reg > map->max_register) @@ -208,6 +213,67 @@ static void dev_get_regmap_release(struct device *dev, void *res) */ } +static bool _regmap_range_add(struct regmap *map, + struct regmap_range_node *data) +{ + struct rb_root *root = &map->range_tree; + struct rb_node **new = &(root->rb_node), *parent = NULL; + + while (*new) { + struct regmap_range_node *this = + container_of(*new, struct regmap_range_node, node); + + parent = *new; + if (data->range_max < this->range_min) + new = &((*new)->rb_left); + else if (data->range_min > this->range_max) + new = &((*new)->rb_right); + else + return false; + } + + rb_link_node(&data->node, parent, new); + rb_insert_color(&data->node, root); + + return true; +} + +static struct regmap_range_node *_regmap_range_lookup(struct regmap *map, + unsigned int reg) +{ + struct rb_node *node = map->range_tree.rb_node; + + while (node) { + struct regmap_range_node *this = + container_of(node, struct regmap_range_node, node); + + if (reg < this->range_min) + node = node->rb_left; + else if (reg > this->range_max) + node = node->rb_right; + else + return this; + } + + return NULL; +} + +static void regmap_range_exit(struct regmap *map) +{ + struct rb_node *next; + struct regmap_range_node *range_node; + + next = rb_first(&map->range_tree); + while (next) { + range_node = rb_entry(next, struct regmap_range_node, node); + next = rb_next(&range_node->node); + rb_erase(&range_node->node, &map->range_tree); + kfree(range_node); + } + + kfree(map->selector_work_buf); +} + /** * regmap_init(): Initialise register map * @@ -227,6 +293,7 @@ struct regmap *regmap_init(struct device *dev, { struct regmap *map, **m; int ret = -EINVAL; + int i, j; if (!bus || !config) goto err; @@ -364,27 +431,88 @@ struct regmap *regmap_init(struct device *dev, goto err_map; } - regmap_debugfs_init(map, config->name); + map->range_tree = RB_ROOT; + for (i = 0; i < config->n_ranges; i++) { + const struct regmap_range_cfg *range_cfg = &config->ranges[i]; + struct regmap_range_node *new; + + /* Sanity check */ + if (range_cfg->range_max < range_cfg->range_min || + range_cfg->range_max > map->max_register || + range_cfg->selector_reg > map->max_register || + range_cfg->window_len == 0) + goto err_range; + + /* Make sure, that this register range has no selector + or data window within its boundary */ + for (j = 0; j < config->n_ranges; j++) { + unsigned sel_reg = config->ranges[j].selector_reg; + unsigned win_min = config->ranges[j].window_start; + unsigned win_max = win_min + + config->ranges[j].window_len - 1; + + if (range_cfg->range_min <= sel_reg && + sel_reg <= range_cfg->range_max) { + goto err_range; + } + + if (!(win_max < range_cfg->range_min || + win_min > range_cfg->range_max)) { + goto err_range; + } + } + + new = kzalloc(sizeof(*new), GFP_KERNEL); + if (new == NULL) { + ret = -ENOMEM; + goto err_range; + } + + new->range_min = range_cfg->range_min; + new->range_max = range_cfg->range_max; + new->selector_reg = range_cfg->selector_reg; + new->selector_mask = range_cfg->selector_mask; + new->selector_shift = range_cfg->selector_shift; + new->window_start = range_cfg->window_start; + new->window_len = range_cfg->window_len; + + if (_regmap_range_add(map, new) == false) { + kfree(new); + goto err_range; + } + + if (map->selector_work_buf == NULL) { + map->selector_work_buf = + kzalloc(map->format.buf_size, GFP_KERNEL); + if (map->selector_work_buf == NULL) { + ret = -ENOMEM; + goto err_range; + } + } + } ret = regcache_init(map, config); if (ret < 0) - goto err_debugfs; + goto err_range; + + regmap_debugfs_init(map, config->name); /* Add a devres resource for dev_get_regmap() */ m = devres_alloc(dev_get_regmap_release, sizeof(*m), GFP_KERNEL); if (!m) { ret = -ENOMEM; - goto err_cache; + goto err_debugfs; } *m = map; devres_add(dev, m); return map; -err_cache: - regcache_exit(map); err_debugfs: regmap_debugfs_exit(map); + regcache_exit(map); +err_range: + regmap_range_exit(map); kfree(map->work_buf); err_map: kfree(map); @@ -481,6 +609,7 @@ void regmap_exit(struct regmap *map) { regcache_exit(map); regmap_debugfs_exit(map); + regmap_range_exit(map); if (map->bus->free_context) map->bus->free_context(map->bus_context); kfree(map->work_buf); @@ -526,6 +655,56 @@ struct regmap *dev_get_regmap(struct device *dev, const char *name) } EXPORT_SYMBOL_GPL(dev_get_regmap); +static int _regmap_select_page(struct regmap *map, unsigned int *reg, + unsigned int val_num) +{ + struct regmap_range_node *range; + void *orig_work_buf; + unsigned int win_offset; + unsigned int win_page; + bool page_chg; + int ret; + + range = _regmap_range_lookup(map, *reg); + if (range) { + win_offset = (*reg - range->range_min) % range->window_len; + win_page = (*reg - range->range_min) / range->window_len; + + if (val_num > 1) { + /* Bulk write shouldn't cross range boundary */ + if (*reg + val_num - 1 > range->range_max) + return -EINVAL; + + /* ... or single page boundary */ + if (val_num > range->window_len - win_offset) + return -EINVAL; + } + + /* It is possible to have selector register inside data window. + In that case, selector register is located on every page and + it needs no page switching, when accessed alone. */ + if (val_num > 1 || + range->window_start + win_offset != range->selector_reg) { + /* Use separate work_buf during page switching */ + orig_work_buf = map->work_buf; + map->work_buf = map->selector_work_buf; + + ret = _regmap_update_bits(map, range->selector_reg, + range->selector_mask, + win_page << range->selector_shift, + &page_chg); + if (ret < 0) + return ret; + + map->work_buf = orig_work_buf; + } + + *reg = range->window_start + win_offset; + } + + return 0; +} + static int _regmap_raw_write(struct regmap *map, unsigned int reg, const void *val, size_t val_len) { @@ -563,6 +742,10 @@ static int _regmap_raw_write(struct regmap *map, unsigned int reg, } } + ret = _regmap_select_page(map, ®, val_len / map->format.val_bytes); + if (ret < 0) + return ret; + map->format.format_reg(map->work_buf, reg, map->reg_shift); u8[0] |= map->write_flag_mask; @@ -626,6 +809,10 @@ int _regmap_write(struct regmap *map, unsigned int reg, trace_regmap_reg_write(map->dev, reg, val); if (map->format.format_write) { + ret = _regmap_select_page(map, ®, 1); + if (ret < 0) + return ret; + map->format.format_write(map, reg, val); trace_regmap_hw_write_start(map->dev, reg, 1); @@ -783,6 +970,10 @@ static int _regmap_raw_read(struct regmap *map, unsigned int reg, void *val, u8 *u8 = map->work_buf; int ret; + ret = _regmap_select_page(map, ®, val_len / map->format.val_bytes); + if (ret < 0) + return ret; + map->format.format_reg(map->work_buf, reg, map->reg_shift); /* diff --git a/include/linux/regmap.h b/include/linux/regmap.h index 56af22ec9aba..5f69d4ad3eb1 100644 --- a/include/linux/regmap.h +++ b/include/linux/regmap.h @@ -14,12 +14,14 @@ */ #include +#include struct module; struct device; struct i2c_client; struct spi_device; struct regmap; +struct regmap_range_cfg; /* An enum of all the supported cache types */ enum regcache_type { @@ -84,6 +86,9 @@ struct reg_default { * @reg_defaults_raw: Power on reset values for registers (for use with * register cache support). * @num_reg_defaults_raw: Number of elements in reg_defaults_raw. + * + * @ranges: Array of configuration entries for virtual address ranges. + * @num_ranges: Number of range configuration entries. */ struct regmap_config { const char *name; @@ -109,6 +114,40 @@ struct regmap_config { u8 write_flag_mask; bool use_single_rw; + + const struct regmap_range_cfg *ranges; + unsigned int n_ranges; +}; + +/** + * Configuration for indirectly accessed or paged registers. + * Registers, mapped to this virtual range, are accessed in two steps: + * 1. page selector register update; + * 2. access through data window registers. + * + * @range_min: Address of the lowest register address in virtual range. + * @range_max: Address of the highest register in virtual range. + * + * @page_sel_reg: Register with selector field. + * @page_sel_mask: Bit shift for selector value. + * @page_sel_shift: Bit mask for selector value. + * + * @window_start: Address of first (lowest) register in data window. + * @window_len: Number of registers in data window. + */ +struct regmap_range_cfg { + /* Registers of virtual address range */ + unsigned int range_min; + unsigned int range_max; + + /* Page selector for indirect addressing */ + unsigned int selector_reg; + unsigned int selector_mask; + int selector_shift; + + /* Data window (per each page) */ + unsigned int window_start; + unsigned int window_len; }; typedef int (*regmap_hw_write)(void *context, const void *data, -- cgit v1.2.3 From 632a5b01db8cd4068a06f8a67720ea4f9b00d9b3 Mon Sep 17 00:00:00 2001 From: Krystian Garbaciak Date: Mon, 18 Jun 2012 13:04:29 +0100 Subject: regmap: Fix work_buf switching for page update during virtual range access. After page update, orginal work_buf has to be restored regardless of the result. Signed-off-by: Krystian Garbaciak Signed-off-by: Mark Brown --- drivers/base/regmap/regmap.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/regmap/regmap.c b/drivers/base/regmap/regmap.c index 83a0166420a4..d912eb2d19ce 100644 --- a/drivers/base/regmap/regmap.c +++ b/drivers/base/regmap/regmap.c @@ -693,10 +693,11 @@ static int _regmap_select_page(struct regmap *map, unsigned int *reg, range->selector_mask, win_page << range->selector_shift, &page_chg); - if (ret < 0) - return ret; map->work_buf = orig_work_buf; + + if (ret < 0) + return ret; } *reg = range->window_start + win_offset; -- cgit v1.2.3 From 1f758b23177d588a71b96ad02990e715949bb82f Mon Sep 17 00:00:00 2001 From: Mandeep Singh Baines Date: Sun, 24 Jun 2012 23:31:09 +0200 Subject: PM / Sleep: Prevent waiting forever on asynchronous suspend after abort __device_suspend() must always send a completion. Otherwise, parent devices will wait forever. Commit 1e2ef05b, "PM: Limit race conditions between runtime PM and system sleep (v2)", introduced a regression by short-circuiting the complete_all() for certain error cases. This patch fixes the bug by always signalling a completion. Addresses http://crosbug.com/31972 Tested by injecting an abort. Signed-off-by: Mandeep Singh Baines Cc: stable@vger.kernel.org Signed-off-by: Rafael J. Wysocki --- drivers/base/power/main.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index e0fb5b0435a3..9cb845e49334 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -1031,7 +1031,7 @@ static int __device_suspend(struct device *dev, pm_message_t state, bool async) dpm_wait_for_children(dev, async); if (async_error) - return 0; + goto Complete; pm_runtime_get_noresume(dev); if (pm_runtime_barrier(dev) && device_may_wakeup(dev)) @@ -1040,7 +1040,7 @@ static int __device_suspend(struct device *dev, pm_message_t state, bool async) if (pm_wakeup_pending()) { pm_runtime_put_sync(dev); async_error = -EBUSY; - return 0; + goto Complete; } device_lock(dev); @@ -1097,6 +1097,8 @@ static int __device_suspend(struct device *dev, pm_message_t state, bool async) } device_unlock(dev); + + Complete: complete_all(&dev->power.completion); if (error) { -- cgit v1.2.3 From 4b7760ba0dd3319f66886ab2335a0fbecdbc808a Mon Sep 17 00:00:00 2001 From: Sameer Nanda Date: Tue, 19 Jun 2012 22:23:33 +0200 Subject: PM / Sleep: add knob for printing device resume times Added a new knob called /sys/power/pm_print_times. Setting it to 1 enables printing of time taken by devices to suspend and resume. Setting it to 0 disables this printing (unless overridden by initcall_debug kernel command line option). Signed-off-by: Sameer Nanda Acked-by: Greg KH Signed-off-by: Rafael J. Wysocki --- Documentation/ABI/testing/sysfs-power | 13 +++++++++++++ drivers/base/power/main.c | 4 ++-- drivers/base/power/power.h | 11 +++++++++++ kernel/power/main.c | 33 +++++++++++++++++++++++++++++++++ 4 files changed, 59 insertions(+), 2 deletions(-) (limited to 'drivers/base') diff --git a/Documentation/ABI/testing/sysfs-power b/Documentation/ABI/testing/sysfs-power index 31725ffeeb3a..217772615d02 100644 --- a/Documentation/ABI/testing/sysfs-power +++ b/Documentation/ABI/testing/sysfs-power @@ -231,3 +231,16 @@ Description: Reads from this file return a string consisting of the names of wakeup sources created with the help of /sys/power/wake_lock that are inactive at the moment, separated with spaces. + +What: /sys/power/pm_print_times +Date: May 2012 +Contact: Sameer Nanda +Description: + The /sys/power/pm_print_times file allows user space to + control whether the time taken by devices to suspend and + resume is printed. These prints are useful for hunting down + devices that take too long to suspend or resume. + + Writing a "1" enables this printing while writing a "0" + disables it. The default value is "0". Reading from this file + will display the current value. diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index 9cb845e49334..6e4db96958d1 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -166,7 +166,7 @@ static ktime_t initcall_debug_start(struct device *dev) { ktime_t calltime = ktime_set(0, 0); - if (initcall_debug) { + if (pm_print_times) { pr_info("calling %s+ @ %i, parent: %s\n", dev_name(dev), task_pid_nr(current), dev->parent ? dev_name(dev->parent) : "none"); @@ -181,7 +181,7 @@ static void initcall_debug_report(struct device *dev, ktime_t calltime, { ktime_t delta, rettime; - if (initcall_debug) { + if (pm_print_times) { rettime = ktime_get(); delta = ktime_sub(rettime, calltime); pr_info("call %s+ returned %d after %Ld usecs\n", dev_name(dev), diff --git a/drivers/base/power/power.h b/drivers/base/power/power.h index eeb4bff9505c..12c77b7ff8e8 100644 --- a/drivers/base/power/power.h +++ b/drivers/base/power/power.h @@ -85,3 +85,14 @@ static inline int pm_qos_sysfs_add(struct device *dev) { return 0; } static inline void pm_qos_sysfs_remove(struct device *dev) {} #endif + +#ifdef CONFIG_PM_DEBUG + +extern int pm_print_times_enabled; +#define pm_print_times (initcall_debug || pm_print_times_enabled) + +#else /* CONFIG_PM_DEBUG */ + +#define pm_print_times initcall_debug + +#endif /* CONFIG_PM_DEBUG */ diff --git a/kernel/power/main.c b/kernel/power/main.c index 428f8a034e96..7beb3fb3670b 100644 --- a/kernel/power/main.c +++ b/kernel/power/main.c @@ -132,6 +132,38 @@ static ssize_t pm_test_store(struct kobject *kobj, struct kobj_attribute *attr, } power_attr(pm_test); + +/* + * pm_print_times: print time taken by devices to suspend and resume. + * + * show() returns whether printing of suspend and resume times is enabled. + * store() accepts 0 or 1. 0 disables printing and 1 enables it. + */ +int pm_print_times_enabled; + +static ssize_t pm_print_times_show(struct kobject *kobj, + struct kobj_attribute *attr, char *buf) +{ + return sprintf(buf, "%d\n", pm_print_times_enabled); +} + +static ssize_t pm_print_times_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t n) +{ + unsigned long val; + + if (kstrtoul(buf, 10, &val)) + return -EINVAL; + + if (val > 1) + return -EINVAL; + + pm_print_times_enabled = val; + return n; +} + +power_attr(pm_print_times); #endif /* CONFIG_PM_DEBUG */ #ifdef CONFIG_DEBUG_FS @@ -530,6 +562,7 @@ static struct attribute * g[] = { #endif #ifdef CONFIG_PM_DEBUG &pm_test_attr.attr, + &pm_print_times_attr.attr, #endif #endif NULL, -- cgit v1.2.3 From b2df1d4f8b95d9d1e3f064cef02fc5c5116b05cf Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Thu, 21 Jun 2012 00:19:33 +0200 Subject: PM / Sleep: Separate printing suspend times from initcall_debug Change the behavior of the newly introduced /sys/power/pm_print_times attribute so that its initial value depends on initcall_debug, but setting it to 0 will cause device suspend/resume times not to be printed, even if initcall_debug has been set. This way, the people who use initcall_debug for reasons other than PM debugging will be able to switch the suspend/resume times printing off, if need be. Signed-off-by: Rafael J. Wysocki Reviewed-by: Srivatsa S. Bhat Acked-by: Greg Kroah-Hartman --- drivers/base/power/main.c | 4 +-- drivers/base/power/power.h | 11 ------- include/linux/suspend.h | 6 ++++ kernel/power/Kconfig | 4 +-- kernel/power/main.c | 76 +++++++++++++++++++++++++++------------------- 5 files changed, 54 insertions(+), 47 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index 6e4db96958d1..df5f41d2ec95 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -166,7 +166,7 @@ static ktime_t initcall_debug_start(struct device *dev) { ktime_t calltime = ktime_set(0, 0); - if (pm_print_times) { + if (pm_print_times_enabled) { pr_info("calling %s+ @ %i, parent: %s\n", dev_name(dev), task_pid_nr(current), dev->parent ? dev_name(dev->parent) : "none"); @@ -181,7 +181,7 @@ static void initcall_debug_report(struct device *dev, ktime_t calltime, { ktime_t delta, rettime; - if (pm_print_times) { + if (pm_print_times_enabled) { rettime = ktime_get(); delta = ktime_sub(rettime, calltime); pr_info("call %s+ returned %d after %Ld usecs\n", dev_name(dev), diff --git a/drivers/base/power/power.h b/drivers/base/power/power.h index 12c77b7ff8e8..eeb4bff9505c 100644 --- a/drivers/base/power/power.h +++ b/drivers/base/power/power.h @@ -85,14 +85,3 @@ static inline int pm_qos_sysfs_add(struct device *dev) { return 0; } static inline void pm_qos_sysfs_remove(struct device *dev) {} #endif - -#ifdef CONFIG_PM_DEBUG - -extern int pm_print_times_enabled; -#define pm_print_times (initcall_debug || pm_print_times_enabled) - -#else /* CONFIG_PM_DEBUG */ - -#define pm_print_times initcall_debug - -#endif /* CONFIG_PM_DEBUG */ diff --git a/include/linux/suspend.h b/include/linux/suspend.h index cd83059fb592..0c808d7fa579 100644 --- a/include/linux/suspend.h +++ b/include/linux/suspend.h @@ -408,6 +408,12 @@ static inline void unlock_system_sleep(void) {} #endif /* !CONFIG_PM_SLEEP */ +#ifdef CONFIG_PM_SLEEP_DEBUG +extern bool pm_print_times_enabled; +#else +#define pm_print_times_enabled (false) +#endif + #ifdef CONFIG_PM_AUTOSLEEP /* kernel/power/autosleep.c */ diff --git a/kernel/power/Kconfig b/kernel/power/Kconfig index 8f9b4eb974e0..a70518c9d82f 100644 --- a/kernel/power/Kconfig +++ b/kernel/power/Kconfig @@ -175,7 +175,7 @@ config PM_TEST_SUSPEND You probably want to have your system's RTC driver statically linked, ensuring that it's available when this test runs. -config CAN_PM_TRACE +config PM_SLEEP_DEBUG def_bool y depends on PM_DEBUG && PM_SLEEP @@ -196,7 +196,7 @@ config PM_TRACE config PM_TRACE_RTC bool "Suspend/resume event tracing" - depends on CAN_PM_TRACE + depends on PM_SLEEP_DEBUG depends on X86 select PM_TRACE ---help--- diff --git a/kernel/power/main.c b/kernel/power/main.c index 7beb3fb3670b..f458238109cc 100644 --- a/kernel/power/main.c +++ b/kernel/power/main.c @@ -132,38 +132,6 @@ static ssize_t pm_test_store(struct kobject *kobj, struct kobj_attribute *attr, } power_attr(pm_test); - -/* - * pm_print_times: print time taken by devices to suspend and resume. - * - * show() returns whether printing of suspend and resume times is enabled. - * store() accepts 0 or 1. 0 disables printing and 1 enables it. - */ -int pm_print_times_enabled; - -static ssize_t pm_print_times_show(struct kobject *kobj, - struct kobj_attribute *attr, char *buf) -{ - return sprintf(buf, "%d\n", pm_print_times_enabled); -} - -static ssize_t pm_print_times_store(struct kobject *kobj, - struct kobj_attribute *attr, - const char *buf, size_t n) -{ - unsigned long val; - - if (kstrtoul(buf, 10, &val)) - return -EINVAL; - - if (val > 1) - return -EINVAL; - - pm_print_times_enabled = val; - return n; -} - -power_attr(pm_print_times); #endif /* CONFIG_PM_DEBUG */ #ifdef CONFIG_DEBUG_FS @@ -267,6 +235,47 @@ late_initcall(pm_debugfs_init); #endif /* CONFIG_PM_SLEEP */ +#ifdef CONFIG_PM_SLEEP_DEBUG +/* + * pm_print_times: print time taken by devices to suspend and resume. + * + * show() returns whether printing of suspend and resume times is enabled. + * store() accepts 0 or 1. 0 disables printing and 1 enables it. + */ +bool pm_print_times_enabled; + +static ssize_t pm_print_times_show(struct kobject *kobj, + struct kobj_attribute *attr, char *buf) +{ + return sprintf(buf, "%d\n", pm_print_times_enabled); +} + +static ssize_t pm_print_times_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t n) +{ + unsigned long val; + + if (kstrtoul(buf, 10, &val)) + return -EINVAL; + + if (val > 1) + return -EINVAL; + + pm_print_times_enabled = !!val; + return n; +} + +power_attr(pm_print_times); + +static inline void pm_print_times_init(void) +{ + pm_print_times_enabled = !!initcall_debug; +} +#else /* !CONFIG_PP_SLEEP_DEBUG */ +static inline void pm_print_times_init(void) {} +#endif /* CONFIG_PM_SLEEP_DEBUG */ + struct kobject *power_kobj; /** @@ -562,6 +571,8 @@ static struct attribute * g[] = { #endif #ifdef CONFIG_PM_DEBUG &pm_test_attr.attr, +#endif +#ifdef CONFIG_PM_SLEEP_DEBUG &pm_print_times_attr.attr, #endif #endif @@ -599,6 +610,7 @@ static int __init pm_init(void) error = sysfs_create_group(power_kobj, &attr_group); if (error) return error; + pm_print_times_init(); return pm_autosleep_init(); } -- cgit v1.2.3 From 0b589741b8b83a33e25fb32bb714ea1b972af5aa Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sat, 16 Jun 2012 00:02:22 +0200 Subject: PM / Domains: Use subsystem runtime suspend/resume callbacks by default Currently, the default "save state" and "restore state" routines for generic PM domains, pm_genpd_default_save_state() and pm_genpd_default_restore_state(), respectively, only use runtime PM callbacks provided by device drivers, but in general those callbacks need not provide the entire necessary functionality. Namely, in general it may be necessary to execute subsystem (i.e. device type, device class or bus type) callbacks that will carry out all of the necessary operations. For this reason, modify pm_genpd_default_save_state() and pm_genpd_default_restore_state() to execute subsystem callbacks, if they are provided, and fall back to driver callbacks otherwise. Signed-off-by: Rafael J. Wysocki --- drivers/base/power/domain.c | 32 ++++++++++++++++++++++++-------- 1 file changed, 24 insertions(+), 8 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/power/domain.c b/drivers/base/power/domain.c index 83aa694a8efe..49fdd4002b9f 100644 --- a/drivers/base/power/domain.c +++ b/drivers/base/power/domain.c @@ -1615,16 +1615,24 @@ EXPORT_SYMBOL_GPL(__pm_genpd_remove_callbacks); static int pm_genpd_default_save_state(struct device *dev) { int (*cb)(struct device *__dev); - struct device_driver *drv = dev->driver; cb = dev_gpd_data(dev)->ops.save_state; if (cb) return cb(dev); - if (drv && drv->pm && drv->pm->runtime_suspend) - return drv->pm->runtime_suspend(dev); + if (dev->type && dev->type->pm) + cb = dev->type->pm->runtime_suspend; + else if (dev->class && dev->class->pm) + cb = dev->class->pm->runtime_suspend; + else if (dev->bus && dev->bus->pm) + cb = dev->bus->pm->runtime_suspend; + else + cb = NULL; - return 0; + if (!cb && dev->driver && dev->driver->pm) + cb = dev->driver->pm->runtime_suspend; + + return cb ? cb(dev) : 0; } /** @@ -1634,16 +1642,24 @@ static int pm_genpd_default_save_state(struct device *dev) static int pm_genpd_default_restore_state(struct device *dev) { int (*cb)(struct device *__dev); - struct device_driver *drv = dev->driver; cb = dev_gpd_data(dev)->ops.restore_state; if (cb) return cb(dev); - if (drv && drv->pm && drv->pm->runtime_resume) - return drv->pm->runtime_resume(dev); + if (dev->type && dev->type->pm) + cb = dev->type->pm->runtime_resume; + else if (dev->class && dev->class->pm) + cb = dev->class->pm->runtime_resume; + else if (dev->bus && dev->bus->pm) + cb = dev->bus->pm->runtime_resume; + else + cb = NULL; - return 0; + if (!cb && dev->driver && dev->driver->pm) + cb = dev->driver->pm->runtime_resume; + + return cb ? cb(dev) : 0; } #ifdef CONFIG_PM_SLEEP -- cgit v1.2.3 From 80de3d7f416f1accd03f2e519ead32d6fde4fcf4 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sat, 16 Jun 2012 00:02:34 +0200 Subject: PM / Domains: Do not stop devices after restoring their states While resuming a device belonging to a PM domain, pm_genpd_runtime_resume() calls __pm_genpd_restore_device() to restore its state, if necessary. The latter starts the device, using genpd_start_dev(), restores its state, using genpd_restore_dev(), and then stops it, using genpd_stop_dev(). However, this last operation is not necessary, because the device is supposed to be operational after pm_genpd_runtime_resume() has returned and because of it pm_genpd_runtime_resume() has to call genpd_start_dev() once again for the "restored" device, which is inefficient. To make things more efficient, remove the call to genpd_stop_dev() from __pm_genpd_restore_device() and the direct call to genpd_start_dev() from pm_genpd_runtime_resume(). [Of course, genpd_start_dev() still has to be called by it for devices with the power.irq_safe flag set, because __pm_genpd_restore_device() is not executed for them.] This change has been tested on the SH7372 Mackerel board. Signed-off-by: Rafael J. Wysocki --- drivers/base/power/domain.c | 16 +++++----------- 1 file changed, 5 insertions(+), 11 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/power/domain.c b/drivers/base/power/domain.c index 49fdd4002b9f..da1d52576ec9 100644 --- a/drivers/base/power/domain.c +++ b/drivers/base/power/domain.c @@ -339,19 +339,16 @@ static void __pm_genpd_restore_device(struct pm_domain_data *pdd, { struct generic_pm_domain_data *gpd_data = to_gpd_data(pdd); struct device *dev = pdd->dev; + bool need_restore = gpd_data->need_restore; - if (!gpd_data->need_restore) - return; - + gpd_data->need_restore = false; mutex_unlock(&genpd->lock); genpd_start_dev(genpd, dev); - genpd_restore_dev(genpd, dev); - genpd_stop_dev(genpd, dev); + if (need_restore) + genpd_restore_dev(genpd, dev); mutex_lock(&genpd->lock); - - gpd_data->need_restore = false; } /** @@ -595,7 +592,7 @@ static int pm_genpd_runtime_resume(struct device *dev) /* If power.irq_safe, the PM domain is never powered off. */ if (dev->power.irq_safe) - goto out; + return genpd_start_dev(genpd, dev); mutex_lock(&genpd->lock); ret = __pm_genpd_poweron(genpd); @@ -628,9 +625,6 @@ static int pm_genpd_runtime_resume(struct device *dev) wake_up_all(&genpd->status_wait_queue); mutex_unlock(&genpd->lock); - out: - genpd_start_dev(genpd, dev); - return 0; } -- cgit v1.2.3 From cbc9ef0287ab764d3da0129efa673808df641fe3 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Tue, 3 Jul 2012 19:07:42 +0200 Subject: PM / Domains: Add preliminary support for cpuidle, v2 On some systems there are CPU cores located in the same power domains as I/O devices. Then, power can only be removed from the domain if all I/O devices in it are not in use and the CPU core is idle. Add preliminary support for that to the generic PM domains framework. First, the platform is expected to provide a cpuidle driver with one extra state designated for use with the generic PM domains code. This state should be initially disabled and its exit_latency value should be set to whatever time is needed to bring up the CPU core itself after restoring power to it, not including the domain's power on latency. Its .enter() callback should point to a procedure that will remove power from the domain containing the CPU core at the end of the CPU power transition. The remaining characteristics of the extra cpuidle state, referred to as the "domain" cpuidle state below, (e.g. power usage, target residency) should be populated in accordance with the properties of the hardware. Next, the platform should execute genpd_attach_cpuidle() on the PM domain containing the CPU core. That will cause the generic PM domains framework to treat that domain in a special way such that: * When all devices in the domain have been suspended and it is about to be turned off, the states of the devices will be saved, but power will not be removed from the domain. Instead, the "domain" cpuidle state will be enabled so that power can be removed from the domain when the CPU core is idle and the state has been chosen as the target by the cpuidle governor. * When the first I/O device in the domain is resumed and __pm_genpd_poweron(() is called for the first time after power has been removed from the domain, the "domain" cpuidle state will be disabled to avoid subsequent surprise power removals via cpuidle. The effective exit_latency value of the "domain" cpuidle state depends on the time needed to bring up the CPU core itself after restoring power to it as well as on the power on latency of the domain containing the CPU core. Thus the "domain" cpuidle state's exit_latency has to be recomputed every time the domain's power on latency is updated, which may happen every time power is restored to the domain, if the measured power on latency is greater than the latency stored in the corresponding generic_pm_domain structure. Signed-off-by: Rafael J. Wysocki Reviewed-by: Kevin Hilman --- drivers/base/power/domain.c | 117 +++++++++++++++++++++++++++++++++++++++ drivers/cpuidle/cpuidle.c | 1 + drivers/cpuidle/governors/menu.c | 3 +- include/linux/cpuidle.h | 1 + include/linux/pm_domain.h | 17 ++++++ 5 files changed, 138 insertions(+), 1 deletion(-) (limited to 'drivers/base') diff --git a/drivers/base/power/domain.c b/drivers/base/power/domain.c index da1d52576ec9..4b5f090fccb6 100644 --- a/drivers/base/power/domain.c +++ b/drivers/base/power/domain.c @@ -139,6 +139,19 @@ static void genpd_set_active(struct generic_pm_domain *genpd) genpd->status = GPD_STATE_ACTIVE; } +static void genpd_recalc_cpu_exit_latency(struct generic_pm_domain *genpd) +{ + s64 usecs64; + + if (!genpd->cpu_data) + return; + + usecs64 = genpd->power_on_latency_ns; + do_div(usecs64, NSEC_PER_USEC); + usecs64 += genpd->cpu_data->saved_exit_latency; + genpd->cpu_data->idle_state->exit_latency = usecs64; +} + /** * __pm_genpd_poweron - Restore power to a given PM domain and its masters. * @genpd: PM domain to power up. @@ -176,6 +189,13 @@ int __pm_genpd_poweron(struct generic_pm_domain *genpd) return 0; } + if (genpd->cpu_data) { + cpuidle_pause_and_lock(); + genpd->cpu_data->idle_state->disabled = true; + cpuidle_resume_and_unlock(); + goto out; + } + /* * The list is guaranteed not to change while the loop below is being * executed, unless one of the masters' .power_on() callbacks fiddles @@ -215,6 +235,7 @@ int __pm_genpd_poweron(struct generic_pm_domain *genpd) if (elapsed_ns > genpd->power_on_latency_ns) { genpd->power_on_latency_ns = elapsed_ns; genpd->max_off_time_changed = true; + genpd_recalc_cpu_exit_latency(genpd); if (genpd->name) pr_warning("%s: Power-on latency exceeded, " "new value %lld ns\n", genpd->name, @@ -222,6 +243,7 @@ int __pm_genpd_poweron(struct generic_pm_domain *genpd) } } + out: genpd_set_active(genpd); return 0; @@ -455,6 +477,21 @@ static int pm_genpd_poweroff(struct generic_pm_domain *genpd) } } + if (genpd->cpu_data) { + /* + * If cpu_data is set, cpuidle should turn the domain off when + * the CPU in it is idle. In that case we don't decrement the + * subdomain counts of the master domains, so that power is not + * removed from the current domain prematurely as a result of + * cutting off the masters' power. + */ + genpd->status = GPD_STATE_POWER_OFF; + cpuidle_pause_and_lock(); + genpd->cpu_data->idle_state->disabled = false; + cpuidle_resume_and_unlock(); + goto out; + } + if (genpd->power_off) { ktime_t time_start; s64 elapsed_ns; @@ -1600,6 +1637,86 @@ int __pm_genpd_remove_callbacks(struct device *dev, bool clear_td) } EXPORT_SYMBOL_GPL(__pm_genpd_remove_callbacks); +int genpd_attach_cpuidle(struct generic_pm_domain *genpd, int state) +{ + struct cpuidle_driver *cpuidle_drv; + struct gpd_cpu_data *cpu_data; + struct cpuidle_state *idle_state; + int ret = 0; + + if (IS_ERR_OR_NULL(genpd) || state < 0) + return -EINVAL; + + genpd_acquire_lock(genpd); + + if (genpd->cpu_data) { + ret = -EEXIST; + goto out; + } + cpu_data = kzalloc(sizeof(*cpu_data), GFP_KERNEL); + if (!cpu_data) { + ret = -ENOMEM; + goto out; + } + cpuidle_drv = cpuidle_driver_ref(); + if (!cpuidle_drv) { + ret = -ENODEV; + goto out; + } + if (cpuidle_drv->state_count <= state) { + ret = -EINVAL; + goto err; + } + idle_state = &cpuidle_drv->states[state]; + if (!idle_state->disabled) { + ret = -EAGAIN; + goto err; + } + cpu_data->idle_state = idle_state; + cpu_data->saved_exit_latency = idle_state->exit_latency; + genpd->cpu_data = cpu_data; + genpd_recalc_cpu_exit_latency(genpd); + + out: + genpd_release_lock(genpd); + return ret; + + err: + cpuidle_driver_unref(); + goto out; +} + +int genpd_detach_cpuidle(struct generic_pm_domain *genpd) +{ + struct gpd_cpu_data *cpu_data; + struct cpuidle_state *idle_state; + int ret = 0; + + if (IS_ERR_OR_NULL(genpd)) + return -EINVAL; + + genpd_acquire_lock(genpd); + + cpu_data = genpd->cpu_data; + if (!cpu_data) { + ret = -ENODEV; + goto out; + } + idle_state = cpu_data->idle_state; + if (!idle_state->disabled) { + ret = -EAGAIN; + goto out; + } + idle_state->exit_latency = cpu_data->saved_exit_latency; + cpuidle_driver_unref(); + genpd->cpu_data = NULL; + kfree(cpu_data); + + out: + genpd_release_lock(genpd); + return ret; +} + /* Default device callbacks for generic PM domains. */ /** diff --git a/drivers/cpuidle/cpuidle.c b/drivers/cpuidle/cpuidle.c index 04e4b7674a47..0132706251df 100644 --- a/drivers/cpuidle/cpuidle.c +++ b/drivers/cpuidle/cpuidle.c @@ -265,6 +265,7 @@ static void poll_idle_init(struct cpuidle_driver *drv) state->power_usage = -1; state->flags = 0; state->enter = poll_idle; + state->disabled = false; } #else static void poll_idle_init(struct cpuidle_driver *drv) {} diff --git a/drivers/cpuidle/governors/menu.c b/drivers/cpuidle/governors/menu.c index 8391d93f57d5..5b1f2c372c1f 100644 --- a/drivers/cpuidle/governors/menu.c +++ b/drivers/cpuidle/governors/menu.c @@ -281,6 +281,7 @@ static int menu_select(struct cpuidle_driver *drv, struct cpuidle_device *dev) * unless the timer is happening really really soon. */ if (data->expected_us > 5 && + !drv->states[CPUIDLE_DRIVER_STATE_START].disabled && dev->states_usage[CPUIDLE_DRIVER_STATE_START].disable == 0) data->last_state_idx = CPUIDLE_DRIVER_STATE_START; @@ -292,7 +293,7 @@ static int menu_select(struct cpuidle_driver *drv, struct cpuidle_device *dev) struct cpuidle_state *s = &drv->states[i]; struct cpuidle_state_usage *su = &dev->states_usage[i]; - if (su->disable) + if (s->disabled || su->disable) continue; if (s->target_residency > data->predicted_us) continue; diff --git a/include/linux/cpuidle.h b/include/linux/cpuidle.h index 27cfced7b57b..8684a0d07b87 100644 --- a/include/linux/cpuidle.h +++ b/include/linux/cpuidle.h @@ -47,6 +47,7 @@ struct cpuidle_state { unsigned int exit_latency; /* in US */ int power_usage; /* in mW */ unsigned int target_residency; /* in US */ + bool disabled; /* disabled on all CPUs */ int (*enter) (struct cpuidle_device *dev, struct cpuidle_driver *drv, diff --git a/include/linux/pm_domain.h b/include/linux/pm_domain.h index 30f794eb3826..2febe31d2675 100644 --- a/include/linux/pm_domain.h +++ b/include/linux/pm_domain.h @@ -15,6 +15,7 @@ #include #include #include +#include enum gpd_status { GPD_STATE_ACTIVE = 0, /* PM domain is active */ @@ -45,6 +46,11 @@ struct gpd_dev_ops { bool (*active_wakeup)(struct device *dev); }; +struct gpd_cpu_data { + unsigned int saved_exit_latency; + struct cpuidle_state *idle_state; +}; + struct generic_pm_domain { struct dev_pm_domain domain; /* PM domain operations */ struct list_head gpd_list_node; /* Node in the global PM domains list */ @@ -75,6 +81,7 @@ struct generic_pm_domain { bool max_off_time_changed; bool cached_power_down_ok; struct device_node *of_node; /* Node in device tree */ + struct gpd_cpu_data *cpu_data; }; static inline struct generic_pm_domain *pd_to_genpd(struct dev_pm_domain *pd) @@ -155,6 +162,8 @@ extern int pm_genpd_add_callbacks(struct device *dev, struct gpd_dev_ops *ops, struct gpd_timing_data *td); extern int __pm_genpd_remove_callbacks(struct device *dev, bool clear_td); +extern int genpd_attach_cpuidle(struct generic_pm_domain *genpd, int state); +extern int genpd_detach_cpuidle(struct generic_pm_domain *genpd); extern void pm_genpd_init(struct generic_pm_domain *genpd, struct dev_power_governor *gov, bool is_off); @@ -211,6 +220,14 @@ static inline int __pm_genpd_remove_callbacks(struct device *dev, bool clear_td) { return -ENOSYS; } +static inline int genpd_attach_cpuidle(struct generic_pm_domain *genpd, int st) +{ + return -ENOSYS; +} +static inline int genpd_detach_cpuidle(struct generic_pm_domain *genpd) +{ + return -ENOSYS; +} static inline void pm_genpd_init(struct generic_pm_domain *genpd, struct dev_power_governor *gov, bool is_off) { -- cgit v1.2.3 From 1d5fcfec22ce5f69db0d29284d2b65ff8ab1bfaa Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Thu, 5 Jul 2012 22:12:32 +0200 Subject: PM / Domains: Add device domain data reference counter Add a mechanism for counting references to the struct generic_pm_domain_data object pointed to by dev->power.subsys_data->domain_data if the device in question belongs to a generic PM domain. This change is necessary for a subsequent patch making it possible to allocate that object from within pm_genpd_add_callbacks(), so that drivers can attach their PM domain device callbacks to devices before those devices are added to PM domains. This patch has been tested on the SH7372 Mackerel board. Signed-off-by: Rafael J. Wysocki --- drivers/base/power/domain.c | 83 +++++++++++++++++++++++++++++++-------------- include/linux/pm_domain.h | 1 + 2 files changed, 58 insertions(+), 26 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/power/domain.c b/drivers/base/power/domain.c index 4b5f090fccb6..45eb3b155b6d 100644 --- a/drivers/base/power/domain.c +++ b/drivers/base/power/domain.c @@ -297,7 +297,7 @@ static int genpd_dev_pm_qos_notifier(struct notifier_block *nb, pdd = dev->power.subsys_data ? dev->power.subsys_data->domain_data : NULL; - if (pdd) { + if (pdd && pdd->dev) { to_gpd_data(pdd)->td.constraint_changed = true; genpd = dev_to_genpd(dev); } else { @@ -1266,6 +1266,27 @@ static void pm_genpd_complete(struct device *dev) #endif /* CONFIG_PM_SLEEP */ +static struct generic_pm_domain_data *__pm_genpd_alloc_dev_data(struct device *dev) +{ + struct generic_pm_domain_data *gpd_data; + + gpd_data = kzalloc(sizeof(*gpd_data), GFP_KERNEL); + if (!gpd_data) + return NULL; + + mutex_init(&gpd_data->lock); + gpd_data->nb.notifier_call = genpd_dev_pm_qos_notifier; + dev_pm_qos_add_notifier(dev, &gpd_data->nb); + return gpd_data; +} + +static void __pm_genpd_free_dev_data(struct device *dev, + struct generic_pm_domain_data *gpd_data) +{ + dev_pm_qos_remove_notifier(dev, &gpd_data->nb); + kfree(gpd_data); +} + /** * __pm_genpd_add_device - Add a device to an I/O PM domain. * @genpd: PM domain to add the device to. @@ -1275,7 +1296,7 @@ static void pm_genpd_complete(struct device *dev) int __pm_genpd_add_device(struct generic_pm_domain *genpd, struct device *dev, struct gpd_timing_data *td) { - struct generic_pm_domain_data *gpd_data; + struct generic_pm_domain_data *gpd_data_new, *gpd_data = NULL; struct pm_domain_data *pdd; int ret = 0; @@ -1284,14 +1305,10 @@ int __pm_genpd_add_device(struct generic_pm_domain *genpd, struct device *dev, if (IS_ERR_OR_NULL(genpd) || IS_ERR_OR_NULL(dev)) return -EINVAL; - gpd_data = kzalloc(sizeof(*gpd_data), GFP_KERNEL); - if (!gpd_data) + gpd_data_new = __pm_genpd_alloc_dev_data(dev); + if (!gpd_data_new) return -ENOMEM; - mutex_init(&gpd_data->lock); - gpd_data->nb.notifier_call = genpd_dev_pm_qos_notifier; - dev_pm_qos_add_notifier(dev, &gpd_data->nb); - genpd_acquire_lock(genpd); if (genpd->prepared_count > 0) { @@ -1305,35 +1322,42 @@ int __pm_genpd_add_device(struct generic_pm_domain *genpd, struct device *dev, goto out; } + ret = dev_pm_get_subsys_data(dev); + if (ret) + goto out; + genpd->device_count++; genpd->max_off_time_changed = true; - dev_pm_get_subsys_data(dev); - - mutex_lock(&gpd_data->lock); spin_lock_irq(&dev->power.lock); + dev->pm_domain = &genpd->domain; - dev->power.subsys_data->domain_data = &gpd_data->base; - gpd_data->base.dev = dev; - list_add_tail(&gpd_data->base.list_node, &genpd->dev_list); - gpd_data->need_restore = genpd->status == GPD_STATE_POWER_OFF; + if (dev->power.subsys_data->domain_data) { + gpd_data = to_gpd_data(dev->power.subsys_data->domain_data); + } else { + gpd_data = gpd_data_new; + dev->power.subsys_data->domain_data = &gpd_data->base; + } + gpd_data->refcount++; if (td) gpd_data->td = *td; + spin_unlock_irq(&dev->power.lock); + + mutex_lock(&gpd_data->lock); + gpd_data->base.dev = dev; + list_add_tail(&gpd_data->base.list_node, &genpd->dev_list); + gpd_data->need_restore = genpd->status == GPD_STATE_POWER_OFF; gpd_data->td.constraint_changed = true; gpd_data->td.effective_constraint_ns = -1; - spin_unlock_irq(&dev->power.lock); mutex_unlock(&gpd_data->lock); - genpd_release_lock(genpd); - - return 0; - out: genpd_release_lock(genpd); - dev_pm_qos_remove_notifier(dev, &gpd_data->nb); - kfree(gpd_data); + if (gpd_data != gpd_data_new) + __pm_genpd_free_dev_data(dev, gpd_data_new); + return ret; } @@ -1379,6 +1403,7 @@ int pm_genpd_remove_device(struct generic_pm_domain *genpd, { struct generic_pm_domain_data *gpd_data; struct pm_domain_data *pdd; + bool remove = false; int ret = 0; dev_dbg(dev, "%s()\n", __func__); @@ -1399,22 +1424,28 @@ int pm_genpd_remove_device(struct generic_pm_domain *genpd, genpd->max_off_time_changed = true; spin_lock_irq(&dev->power.lock); + dev->pm_domain = NULL; pdd = dev->power.subsys_data->domain_data; list_del_init(&pdd->list_node); - dev->power.subsys_data->domain_data = NULL; + gpd_data = to_gpd_data(pdd); + if (--gpd_data->refcount == 0) { + dev->power.subsys_data->domain_data = NULL; + remove = true; + } + spin_unlock_irq(&dev->power.lock); - gpd_data = to_gpd_data(pdd); mutex_lock(&gpd_data->lock); pdd->dev = NULL; mutex_unlock(&gpd_data->lock); genpd_release_lock(genpd); - dev_pm_qos_remove_notifier(dev, &gpd_data->nb); - kfree(gpd_data); dev_pm_put_subsys_data(dev); + if (remove) + __pm_genpd_free_dev_data(dev, gpd_data); + return 0; out: diff --git a/include/linux/pm_domain.h b/include/linux/pm_domain.h index 2febe31d2675..a7d6172922d4 100644 --- a/include/linux/pm_domain.h +++ b/include/linux/pm_domain.h @@ -112,6 +112,7 @@ struct generic_pm_domain_data { struct gpd_timing_data td; struct notifier_block nb; struct mutex lock; + unsigned int refcount; bool need_restore; bool always_on; }; -- cgit v1.2.3 From 62d4490294e3de7015f715c00af992d7b6a6e22c Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Thu, 5 Jul 2012 22:12:54 +0200 Subject: PM / Domains: Allow device callbacks to be added at any time Make it possible to modify device callbacks used by the generic PM domains core code at any time, not only after the device has been added to a domain. This will allow device drivers to provide their own device PM domain callbacks even if they are registered before adding the devices to PM domains. For this purpose, use the observation that the struct generic_pm_domain_data object containing the relevant callback pointers may be allocated by pm_genpd_add_callbacks() and the callbacks may be set before __pm_genpd_add_device() is run for the given device. This object will then be used by __pm_genpd_add_device(), but it has to be protected from premature removal by reference counting. Signed-off-by: Rafael J. Wysocki --- drivers/base/power/domain.c | 64 +++++++++++++++++++++++++++++++++++---------- 1 file changed, 50 insertions(+), 14 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/power/domain.c b/drivers/base/power/domain.c index 45eb3b155b6d..9727bc56320a 100644 --- a/drivers/base/power/domain.c +++ b/drivers/base/power/domain.c @@ -1603,33 +1603,52 @@ int pm_genpd_remove_subdomain(struct generic_pm_domain *genpd, * @dev: Device to add the callbacks to. * @ops: Set of callbacks to add. * @td: Timing data to add to the device along with the callbacks (optional). + * + * Every call to this routine should be balanced with a call to + * __pm_genpd_remove_callbacks() and they must not be nested. */ int pm_genpd_add_callbacks(struct device *dev, struct gpd_dev_ops *ops, struct gpd_timing_data *td) { - struct pm_domain_data *pdd; + struct generic_pm_domain_data *gpd_data_new, *gpd_data = NULL; int ret = 0; - if (!(dev && dev->power.subsys_data && ops)) + if (!(dev && ops)) return -EINVAL; + gpd_data_new = __pm_genpd_alloc_dev_data(dev); + if (!gpd_data_new) + return -ENOMEM; + pm_runtime_disable(dev); device_pm_lock(); - pdd = dev->power.subsys_data->domain_data; - if (pdd) { - struct generic_pm_domain_data *gpd_data = to_gpd_data(pdd); + ret = dev_pm_get_subsys_data(dev); + if (ret) + goto out; + + spin_lock_irq(&dev->power.lock); - gpd_data->ops = *ops; - if (td) - gpd_data->td = *td; + if (dev->power.subsys_data->domain_data) { + gpd_data = to_gpd_data(dev->power.subsys_data->domain_data); } else { - ret = -EINVAL; + gpd_data = gpd_data_new; + dev->power.subsys_data->domain_data = &gpd_data->base; } + gpd_data->refcount++; + gpd_data->ops = *ops; + if (td) + gpd_data->td = *td; + spin_unlock_irq(&dev->power.lock); + + out: device_pm_unlock(); pm_runtime_enable(dev); + if (gpd_data != gpd_data_new) + __pm_genpd_free_dev_data(dev, gpd_data_new); + return ret; } EXPORT_SYMBOL_GPL(pm_genpd_add_callbacks); @@ -1638,10 +1657,13 @@ EXPORT_SYMBOL_GPL(pm_genpd_add_callbacks); * __pm_genpd_remove_callbacks - Remove PM domain callbacks from a given device. * @dev: Device to remove the callbacks from. * @clear_td: If set, clear the device's timing data too. + * + * This routine can only be called after pm_genpd_add_callbacks(). */ int __pm_genpd_remove_callbacks(struct device *dev, bool clear_td) { - struct pm_domain_data *pdd; + struct generic_pm_domain_data *gpd_data = NULL; + bool remove = false; int ret = 0; if (!(dev && dev->power.subsys_data)) @@ -1650,21 +1672,35 @@ int __pm_genpd_remove_callbacks(struct device *dev, bool clear_td) pm_runtime_disable(dev); device_pm_lock(); - pdd = dev->power.subsys_data->domain_data; - if (pdd) { - struct generic_pm_domain_data *gpd_data = to_gpd_data(pdd); + spin_lock_irq(&dev->power.lock); + if (dev->power.subsys_data->domain_data) { + gpd_data = to_gpd_data(dev->power.subsys_data->domain_data); gpd_data->ops = (struct gpd_dev_ops){ 0 }; if (clear_td) gpd_data->td = (struct gpd_timing_data){ 0 }; + + if (--gpd_data->refcount == 0) { + dev->power.subsys_data->domain_data = NULL; + remove = true; + } } else { ret = -EINVAL; } + spin_unlock_irq(&dev->power.lock); + device_pm_unlock(); pm_runtime_enable(dev); - return ret; + if (ret) + return ret; + + dev_pm_put_subsys_data(dev); + if (remove) + __pm_genpd_free_dev_data(dev, gpd_data); + + return 0; } EXPORT_SYMBOL_GPL(__pm_genpd_remove_callbacks); -- cgit v1.2.3 From 1044c180de4ba426aa5fb1b4b13b5f219ddb7105 Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Fri, 6 Jul 2012 14:10:23 +0100 Subject: regmap: Add hook for printk logging for debugging during early init Sometimes for failures during very early init the trace infrastructure isn't available early enough to be used. For this sort of problem defining LOG_DEVICE will add printks for basic register I/O on a specific device, allowing trace to be extracted when the trace system doesn't come up early enough to work with. Signed-off-by: Mark Brown --- drivers/base/regmap/regmap.c | 23 +++++++++++++++++++++++ 1 file changed, 23 insertions(+) (limited to 'drivers/base') diff --git a/drivers/base/regmap/regmap.c b/drivers/base/regmap/regmap.c index 1cf721421bec..f8a937654af2 100644 --- a/drivers/base/regmap/regmap.c +++ b/drivers/base/regmap/regmap.c @@ -21,6 +21,18 @@ #include "internal.h" +/* + * Sometimes for failures during very early init the trace + * infrastructure isn't available early enough to be used. For this + * sort of problem defining LOG_DEVICE will add printks for basic + * register I/O on a specific device. + */ +#undef LOG_DEVICE + +static int _regmap_update_bits(struct regmap *map, unsigned int reg, + unsigned int mask, unsigned int val, + bool *change); + bool regmap_writeable(struct regmap *map, unsigned int reg) { if (map->max_register && reg > map->max_register) @@ -700,6 +712,11 @@ int _regmap_write(struct regmap *map, unsigned int reg, } } +#ifdef LOG_DEVICE + if (strcmp(dev_name(map->dev), LOG_DEVICE) == 0) + dev_info(map->dev, "%x <= %x\n", reg, val); +#endif + trace_regmap_reg_write(map->dev, reg, val); if (map->format.format_write) { @@ -903,6 +920,12 @@ static int _regmap_read(struct regmap *map, unsigned int reg, ret = _regmap_raw_read(map, reg, map->work_buf, map->format.val_bytes); if (ret == 0) { *val = map->format.parse_val(map->work_buf); + +#ifdef LOG_DEVICE + if (strcmp(dev_name(map->dev), LOG_DEVICE) == 0) + dev_info(map->dev, "%x => %x\n", reg, *val); +#endif + trace_regmap_reg_read(map->dev, reg, *val); } -- cgit v1.2.3 From 8651f97bd951d0bb1c10fa24e3fa3455193f3548 Mon Sep 17 00:00:00 2001 From: Preeti U Murthy Date: Mon, 9 Jul 2012 10:12:56 +0200 Subject: PM / cpuidle: System resume hang fix with cpuidle On certain bios, resume hangs if cpus are allowed to enter idle states during suspend [1]. This was fixed in apci idle driver [2].But intel_idle driver does not have this fix. Thus instead of replicating the fix in both the idle drivers, or in more platform specific idle drivers if needed, the more general cpuidle infrastructure could handle this. A suspend callback in cpuidle_driver could handle this fix. But a cpuidle_driver provides only basic functionalities like platform idle state detection capability and mechanisms to support entry and exit into CPU idle states. All other cpuidle functions are found in the cpuidle generic infrastructure for good reason that all cpuidle drivers, irrepective of their platforms will support these functions. One option therefore would be to register a suspend callback in cpuidle which handles this fix. This could be called through a PM_SUSPEND_PREPARE notifier. But this is too generic a notfier for a driver to handle. Also, ideally the job of cpuidle is not to handle side effects of suspend. It should expose the interfaces which "handle cpuidle 'during' suspend" or any other operation, which the subsystems call during that respective operation. The fix demands that during suspend, no cpus should be allowed to enter deep C-states. The interface cpuidle_uninstall_idle_handler() in cpuidle ensures that. Not just that it also kicks all the cpus which are already in idle out of their idle states which was being done during cpu hotplug through a CPU_DYING_FROZEN callbacks. Now the question arises about when during suspend should cpuidle_uninstall_idle_handler() be called. Since we are dealing with drivers it seems best to call this function during dpm_suspend(). Delaying the call till dpm_suspend_noirq() does no harm, as long as it is before cpu_hotplug_begin() to avoid race conditions with cpu hotpulg operations. In dpm_suspend_noirq(), it would be wise to place this call before suspend_device_irqs() to avoid ugly interactions with the same. Ananlogously, during resume. References: [1] https://bugs.launchpad.net/ubuntu/+source/linux/+bug/674075. [2] http://marc.info/?l=linux-pm&m=133958534231884&w=2 Reported-and-tested-by: Dave Hansen Signed-off-by: Preeti U Murthy Reviewed-by: Srivatsa S. Bhat Signed-off-by: Rafael J. Wysocki --- drivers/acpi/processor_idle.c | 30 +----------------------------- drivers/base/power/main.c | 4 +++- drivers/cpuidle/cpuidle.c | 16 ++++++++++++++++ include/linux/cpuidle.h | 4 ++++ 4 files changed, 24 insertions(+), 30 deletions(-) (limited to 'drivers/base') diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c index 47a8caa89dbe..d8366ee75716 100644 --- a/drivers/acpi/processor_idle.c +++ b/drivers/acpi/processor_idle.c @@ -221,10 +221,6 @@ static void lapic_timer_state_broadcast(struct acpi_processor *pr, #endif -/* - * Suspend / resume control - */ -static int acpi_idle_suspend; static u32 saved_bm_rld; static void acpi_idle_bm_rld_save(void) @@ -243,21 +239,13 @@ static void acpi_idle_bm_rld_restore(void) int acpi_processor_suspend(struct acpi_device * device, pm_message_t state) { - if (acpi_idle_suspend == 1) - return 0; - acpi_idle_bm_rld_save(); - acpi_idle_suspend = 1; return 0; } int acpi_processor_resume(struct acpi_device * device) { - if (acpi_idle_suspend == 0) - return 0; - acpi_idle_bm_rld_restore(); - acpi_idle_suspend = 0; return 0; } @@ -763,11 +751,6 @@ static int acpi_idle_enter_c1(struct cpuidle_device *dev, local_irq_disable(); - if (acpi_idle_suspend) { - local_irq_enable(); - cpu_relax(); - return -EBUSY; - } lapic_timer_state_broadcast(pr, cx, 1); kt1 = ktime_get_real(); @@ -838,11 +821,6 @@ static int acpi_idle_enter_simple(struct cpuidle_device *dev, local_irq_disable(); - if (acpi_idle_suspend) { - local_irq_enable(); - cpu_relax(); - return -EBUSY; - } if (cx->entry_method != ACPI_CSTATE_FFH) { current_thread_info()->status &= ~TS_POLLING; @@ -928,8 +906,7 @@ static int acpi_idle_enter_bm(struct cpuidle_device *dev, drv, drv->safe_state_index); } else { local_irq_disable(); - if (!acpi_idle_suspend) - acpi_safe_halt(); + acpi_safe_halt(); local_irq_enable(); return -EBUSY; } @@ -937,11 +914,6 @@ static int acpi_idle_enter_bm(struct cpuidle_device *dev, local_irq_disable(); - if (acpi_idle_suspend) { - local_irq_enable(); - cpu_relax(); - return -EBUSY; - } if (cx->entry_method != ACPI_CSTATE_FFH) { current_thread_info()->status &= ~TS_POLLING; diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index 9cb845e49334..63048f79de5f 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -28,7 +28,7 @@ #include #include #include - +#include #include "../base.h" #include "power.h" @@ -467,6 +467,7 @@ static void dpm_resume_noirq(pm_message_t state) mutex_unlock(&dpm_list_mtx); dpm_show_time(starttime, state, "noirq"); resume_device_irqs(); + cpuidle_resume(); } /** @@ -867,6 +868,7 @@ static int dpm_suspend_noirq(pm_message_t state) ktime_t starttime = ktime_get(); int error = 0; + cpuidle_pause(); suspend_device_irqs(); mutex_lock(&dpm_list_mtx); while (!list_empty(&dpm_late_early_list)) { diff --git a/drivers/cpuidle/cpuidle.c b/drivers/cpuidle/cpuidle.c index 04e4b7674a47..efa9a2ca30e7 100644 --- a/drivers/cpuidle/cpuidle.c +++ b/drivers/cpuidle/cpuidle.c @@ -201,6 +201,22 @@ void cpuidle_resume_and_unlock(void) EXPORT_SYMBOL_GPL(cpuidle_resume_and_unlock); +/* Currently used in suspend/resume path to suspend cpuidle */ +void cpuidle_pause(void) +{ + mutex_lock(&cpuidle_lock); + cpuidle_uninstall_idle_handler(); + mutex_unlock(&cpuidle_lock); +} + +/* Currently used in suspend/resume path to resume cpuidle */ +void cpuidle_resume(void) +{ + mutex_lock(&cpuidle_lock); + cpuidle_install_idle_handler(); + mutex_unlock(&cpuidle_lock); +} + /** * cpuidle_wrap_enter - performs timekeeping and irqen around enter function * @dev: pointer to a valid cpuidle_device object diff --git a/include/linux/cpuidle.h b/include/linux/cpuidle.h index 524bb6f3b6c4..ca6cdf55eb18 100644 --- a/include/linux/cpuidle.h +++ b/include/linux/cpuidle.h @@ -145,6 +145,8 @@ extern void cpuidle_unregister_device(struct cpuidle_device *dev); extern void cpuidle_pause_and_lock(void); extern void cpuidle_resume_and_unlock(void); +extern void cpuidle_pause(void); +extern void cpuidle_resume(void); extern int cpuidle_enable_device(struct cpuidle_device *dev); extern void cpuidle_disable_device(struct cpuidle_device *dev); extern int cpuidle_wrap_enter(struct cpuidle_device *dev, @@ -168,6 +170,8 @@ static inline void cpuidle_unregister_device(struct cpuidle_device *dev) { } static inline void cpuidle_pause_and_lock(void) { } static inline void cpuidle_resume_and_unlock(void) { } +static inline void cpuidle_pause(void) { } +static inline void cpuidle_resume(void) { } static inline int cpuidle_enable_device(struct cpuidle_device *dev) {return -ENODEV; } static inline void cpuidle_disable_device(struct cpuidle_device *dev) { } -- cgit v1.2.3 From 8951ef021904752a843a81fee785f1b180f31c07 Mon Sep 17 00:00:00 2001 From: Sachin Kamat Date: Tue, 10 Jul 2012 21:47:07 +0200 Subject: PM / Domains: Add missing static storage class specifier in domain.c file Fixes the folloiwng sparse warning: drivers/base/power/domain.c:149:5: warning: symbol '__pm_genpd_poweron' was not declared. Should it be static? Signed-off-by: Sachin Kamat Signed-off-by: Rafael J. Wysocki --- drivers/base/power/domain.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/base') diff --git a/drivers/base/power/domain.c b/drivers/base/power/domain.c index 9727bc56320a..4f8e0d2e7780 100644 --- a/drivers/base/power/domain.c +++ b/drivers/base/power/domain.c @@ -159,7 +159,7 @@ static void genpd_recalc_cpu_exit_latency(struct generic_pm_domain *genpd) * Restore power to @genpd and all of its masters so that it is possible to * resume a device belonging to it. */ -int __pm_genpd_poweron(struct generic_pm_domain *genpd) +static int __pm_genpd_poweron(struct generic_pm_domain *genpd) __releases(&genpd->lock) __acquires(&genpd->lock) { struct gpd_link *link; -- cgit v1.2.3 From db79e53dd5957321ca78efe6c0cda578034e381b Mon Sep 17 00:00:00 2001 From: Sachin Kamat Date: Wed, 11 Jul 2012 12:25:49 +0200 Subject: PM / Domains: Replace plain integer with NULL pointer in domain.c file Fixes the following sparse warning: drivers/base/power/domain.c:1679:55: warning: Using plain integer as NULL pointer Signed-off-by: Sachin Kamat Signed-off-by: Rafael J. Wysocki --- drivers/base/power/domain.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/base') diff --git a/drivers/base/power/domain.c b/drivers/base/power/domain.c index 4f8e0d2e7780..8d6741780b0f 100644 --- a/drivers/base/power/domain.c +++ b/drivers/base/power/domain.c @@ -1676,7 +1676,7 @@ int __pm_genpd_remove_callbacks(struct device *dev, bool clear_td) if (dev->power.subsys_data->domain_data) { gpd_data = to_gpd_data(dev->power.subsys_data->domain_data); - gpd_data->ops = (struct gpd_dev_ops){ 0 }; + gpd_data->ops = (struct gpd_dev_ops){ NULL }; if (clear_td) gpd_data->td = (struct gpd_timing_data){ 0 }; -- cgit v1.2.3 From 8e9afafdad59f5973a5e72e05db9802f82091398 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Wed, 11 Jul 2012 22:42:52 +0200 Subject: PM / Domains: Fix build warning for CONFIG_PM_RUNTIME unset The functions genpd_save_dev() and genpd_restore_dev() are not used for CONFIG_PM_RUNTIME unset, so move them under an appropriate #ifdef. Signed-off-by: Rafael J. Wysocki --- drivers/base/power/domain.c | 26 +++++++++++++------------- 1 file changed, 13 insertions(+), 13 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/power/domain.c b/drivers/base/power/domain.c index 8d6741780b0f..ba3487c9835b 100644 --- a/drivers/base/power/domain.c +++ b/drivers/base/power/domain.c @@ -75,19 +75,6 @@ static int genpd_start_dev(struct generic_pm_domain *genpd, struct device *dev) start_latency_ns, "start"); } -static int genpd_save_dev(struct generic_pm_domain *genpd, struct device *dev) -{ - return GENPD_DEV_TIMED_CALLBACK(genpd, int, save_state, dev, - save_state_latency_ns, "state save"); -} - -static int genpd_restore_dev(struct generic_pm_domain *genpd, struct device *dev) -{ - return GENPD_DEV_TIMED_CALLBACK(genpd, int, restore_state, dev, - restore_state_latency_ns, - "state restore"); -} - static bool genpd_sd_counter_dec(struct generic_pm_domain *genpd) { bool ret = false; @@ -273,6 +260,19 @@ int pm_genpd_poweron(struct generic_pm_domain *genpd) #ifdef CONFIG_PM_RUNTIME +static int genpd_save_dev(struct generic_pm_domain *genpd, struct device *dev) +{ + return GENPD_DEV_TIMED_CALLBACK(genpd, int, save_state, dev, + save_state_latency_ns, "state save"); +} + +static int genpd_restore_dev(struct generic_pm_domain *genpd, struct device *dev) +{ + return GENPD_DEV_TIMED_CALLBACK(genpd, int, restore_state, dev, + restore_state_latency_ns, + "state restore"); +} + static int genpd_dev_pm_qos_notifier(struct notifier_block *nb, unsigned long val, void *ptr) { -- cgit v1.2.3 From d181b49eb3d76ed6a134cb599456176f466047c0 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Wed, 11 Jul 2012 22:42:58 +0200 Subject: PM / Sleep: Fix build warning in sysfs.c for CONFIG_PM_SLEEP unset The power/async device sysfs attribute is only used if both CONFIG_PM_ADVANCED_DEBUG and CONFIG_PM_SLEEP are set, but the code implementing it doesn't depend on CONFIG_PM_SLEEP. As a result, a build warning appears if CONFIG_PM_ADVANCED_DEBUG is set and CONFIG_PM_SLEEP is not set. Fix it by adding a #ifdef CONFIG_PM_SLEEP around the code in question. Signed-off-by: Rafael J. Wysocki --- drivers/base/power/sysfs.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'drivers/base') diff --git a/drivers/base/power/sysfs.c b/drivers/base/power/sysfs.c index 48be2ad4dd2c..b91dc6f1e914 100644 --- a/drivers/base/power/sysfs.c +++ b/drivers/base/power/sysfs.c @@ -474,6 +474,8 @@ static DEVICE_ATTR(runtime_enabled, 0444, rtpm_enabled_show, NULL); #endif +#ifdef CONFIG_PM_SLEEP + static ssize_t async_show(struct device *dev, struct device_attribute *attr, char *buf) { @@ -500,6 +502,8 @@ static ssize_t async_store(struct device *dev, struct device_attribute *attr, } static DEVICE_ATTR(async, 0644, async_show, async_store); + +#endif #endif /* CONFIG_PM_ADVANCED_DEBUG */ static struct attribute *power_attrs[] = { -- cgit v1.2.3 From 79714f72d3b964611997de512cb29198c9f2dbbb Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 15 Jun 2012 03:01:42 +0400 Subject: get rid of kern_path_parent() all callers want the same thing, actually - a kinda-sorta analog of kern_path_create(). I.e. they want parent vfsmount/dentry (with ->i_mutex held, to make sure the child dentry is still their child) + the child dentry. Signed-off-by Al Viro --- drivers/base/devtmpfs.c | 98 +++++++++++++++++++++---------------------------- fs/namei.c | 22 ++++++++++- include/linux/namei.h | 2 +- kernel/audit_watch.c | 25 ++----------- 4 files changed, 65 insertions(+), 82 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/devtmpfs.c b/drivers/base/devtmpfs.c index 765c3a28077a..d91a3a0b2325 100644 --- a/drivers/base/devtmpfs.c +++ b/drivers/base/devtmpfs.c @@ -227,33 +227,24 @@ static int handle_create(const char *nodename, umode_t mode, struct device *dev) static int dev_rmdir(const char *name) { - struct nameidata nd; + struct path parent; struct dentry *dentry; int err; - err = kern_path_parent(name, &nd); - if (err) - return err; - - mutex_lock_nested(&nd.path.dentry->d_inode->i_mutex, I_MUTEX_PARENT); - dentry = lookup_one_len(nd.last.name, nd.path.dentry, nd.last.len); - if (!IS_ERR(dentry)) { - if (dentry->d_inode) { - if (dentry->d_inode->i_private == &thread) - err = vfs_rmdir(nd.path.dentry->d_inode, - dentry); - else - err = -EPERM; - } else { - err = -ENOENT; - } - dput(dentry); + dentry = kern_path_locked(name, &parent); + if (IS_ERR(dentry)) + return PTR_ERR(dentry); + if (dentry->d_inode) { + if (dentry->d_inode->i_private == &thread) + err = vfs_rmdir(parent.dentry->d_inode, dentry); + else + err = -EPERM; } else { - err = PTR_ERR(dentry); + err = -ENOENT; } - - mutex_unlock(&nd.path.dentry->d_inode->i_mutex); - path_put(&nd.path); + dput(dentry); + mutex_unlock(&parent.dentry->d_inode->i_mutex); + path_put(&parent); return err; } @@ -305,50 +296,43 @@ static int dev_mynode(struct device *dev, struct inode *inode, struct kstat *sta static int handle_remove(const char *nodename, struct device *dev) { - struct nameidata nd; + struct path parent; struct dentry *dentry; - struct kstat stat; int deleted = 1; int err; - err = kern_path_parent(nodename, &nd); - if (err) - return err; + dentry = kern_path_locked(nodename, &parent); + if (IS_ERR(dentry)) + return PTR_ERR(dentry); - mutex_lock_nested(&nd.path.dentry->d_inode->i_mutex, I_MUTEX_PARENT); - dentry = lookup_one_len(nd.last.name, nd.path.dentry, nd.last.len); - if (!IS_ERR(dentry)) { - if (dentry->d_inode) { - err = vfs_getattr(nd.path.mnt, dentry, &stat); - if (!err && dev_mynode(dev, dentry->d_inode, &stat)) { - struct iattr newattrs; - /* - * before unlinking this node, reset permissions - * of possible references like hardlinks - */ - newattrs.ia_uid = 0; - newattrs.ia_gid = 0; - newattrs.ia_mode = stat.mode & ~0777; - newattrs.ia_valid = - ATTR_UID|ATTR_GID|ATTR_MODE; - mutex_lock(&dentry->d_inode->i_mutex); - notify_change(dentry, &newattrs); - mutex_unlock(&dentry->d_inode->i_mutex); - err = vfs_unlink(nd.path.dentry->d_inode, - dentry); - if (!err || err == -ENOENT) - deleted = 1; - } - } else { - err = -ENOENT; + if (dentry->d_inode) { + struct kstat stat; + err = vfs_getattr(parent.mnt, dentry, &stat); + if (!err && dev_mynode(dev, dentry->d_inode, &stat)) { + struct iattr newattrs; + /* + * before unlinking this node, reset permissions + * of possible references like hardlinks + */ + newattrs.ia_uid = 0; + newattrs.ia_gid = 0; + newattrs.ia_mode = stat.mode & ~0777; + newattrs.ia_valid = + ATTR_UID|ATTR_GID|ATTR_MODE; + mutex_lock(&dentry->d_inode->i_mutex); + notify_change(dentry, &newattrs); + mutex_unlock(&dentry->d_inode->i_mutex); + err = vfs_unlink(parent.dentry->d_inode, dentry); + if (!err || err == -ENOENT) + deleted = 1; } - dput(dentry); } else { - err = PTR_ERR(dentry); + err = -ENOENT; } - mutex_unlock(&nd.path.dentry->d_inode->i_mutex); + dput(dentry); + mutex_unlock(&parent.dentry->d_inode->i_mutex); - path_put(&nd.path); + path_put(&parent); if (deleted && strchr(nodename, '/')) delete_path(nodename); return err; diff --git a/fs/namei.c b/fs/namei.c index 5abab9176903..6b29a51bef5d 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -1814,9 +1814,27 @@ static int do_path_lookup(int dfd, const char *name, return retval; } -int kern_path_parent(const char *name, struct nameidata *nd) +/* does lookup, returns the object with parent locked */ +struct dentry *kern_path_locked(const char *name, struct path *path) { - return do_path_lookup(AT_FDCWD, name, LOOKUP_PARENT, nd); + struct nameidata nd; + struct dentry *d; + int err = do_path_lookup(AT_FDCWD, name, LOOKUP_PARENT, &nd); + if (err) + return ERR_PTR(err); + if (nd.last_type != LAST_NORM) { + path_put(&nd.path); + return ERR_PTR(-EINVAL); + } + mutex_lock_nested(&nd.path.dentry->d_inode->i_mutex, I_MUTEX_PARENT); + d = lookup_one_len(nd.last.name, nd.path.dentry, nd.last.len); + if (IS_ERR(d)) { + mutex_unlock(&nd.path.dentry->d_inode->i_mutex); + path_put(&nd.path); + return d; + } + *path = nd.path; + return d; } int kern_path(const char *name, unsigned int flags, struct path *path) diff --git a/include/linux/namei.h b/include/linux/namei.h index 23d859879210..f5931489e150 100644 --- a/include/linux/namei.h +++ b/include/linux/namei.h @@ -67,7 +67,7 @@ extern int kern_path(const char *, unsigned, struct path *); extern struct dentry *kern_path_create(int, const char *, struct path *, int); extern struct dentry *user_path_create(int, const char __user *, struct path *, int); -extern int kern_path_parent(const char *, struct nameidata *); +extern struct dentry *kern_path_locked(const char *, struct path *); extern int vfs_path_lookup(struct dentry *, struct vfsmount *, const char *, unsigned int, struct path *); diff --git a/kernel/audit_watch.c b/kernel/audit_watch.c index e683869365d9..3823281401b5 100644 --- a/kernel/audit_watch.c +++ b/kernel/audit_watch.c @@ -355,34 +355,15 @@ static void audit_remove_parent_watches(struct audit_parent *parent) /* Get path information necessary for adding watches. */ static int audit_get_nd(struct audit_watch *watch, struct path *parent) { - struct nameidata nd; - struct dentry *d; - int err; - - err = kern_path_parent(watch->path, &nd); - if (err) - return err; - - if (nd.last_type != LAST_NORM) { - path_put(&nd.path); - return -EINVAL; - } - - mutex_lock_nested(&nd.path.dentry->d_inode->i_mutex, I_MUTEX_PARENT); - d = lookup_one_len(nd.last.name, nd.path.dentry, nd.last.len); - if (IS_ERR(d)) { - mutex_unlock(&nd.path.dentry->d_inode->i_mutex); - path_put(&nd.path); + struct dentry *d = kern_path_locked(watch->path, parent); + if (IS_ERR(d)) return PTR_ERR(d); - } + mutex_unlock(&parent->dentry->d_inode->i_mutex); if (d->d_inode) { /* update watch filter fields */ watch->dev = d->d_inode->i_sb->s_dev; watch->ino = d->d_inode->i_ino; } - mutex_unlock(&nd.path.dentry->d_inode->i_mutex); - - *parent = nd.path; dput(d); return 0; } -- cgit v1.2.3 From d1c6c030fcec6f860d9bb6c632a3ebe62e28440b Mon Sep 17 00:00:00 2001 From: Ming Lei Date: Fri, 22 Jun 2012 18:01:40 +0800 Subject: driver core: fix shutdown races with probe/remove(v3) Firstly, .shutdown callback may touch a uninitialized hardware if dev->driver is set and .probe is not completed. Secondly, device_shutdown() may dereference a null pointer to cause oops when dev->driver is cleared after it has been checked in device_shutdown(). So just hold device lock and its parent lock(if it has) to fix the races. Cc: Alan Stern Signed-off-by: Ming Lei Signed-off-by: Greg Kroah-Hartman --- drivers/base/core.c | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) (limited to 'drivers/base') diff --git a/drivers/base/core.c b/drivers/base/core.c index 846d12b5d579..65849a9aeece 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -1812,6 +1812,13 @@ void device_shutdown(void) while (!list_empty(&devices_kset->list)) { dev = list_entry(devices_kset->list.prev, struct device, kobj.entry); + + /* + * hold reference count of device's parent to + * prevent it from being freed because parent's + * lock is to be held + */ + get_device(dev->parent); get_device(dev); /* * Make sure the device is off the kset list, in the @@ -1820,6 +1827,11 @@ void device_shutdown(void) list_del_init(&dev->kobj.entry); spin_unlock(&devices_kset->list_lock); + /* hold lock to avoid race with probe/release */ + if (dev->parent) + device_lock(dev->parent); + device_lock(dev); + /* Don't allow any more runtime suspends */ pm_runtime_get_noresume(dev); pm_runtime_barrier(dev); @@ -1831,7 +1843,13 @@ void device_shutdown(void) dev_dbg(dev, "shutdown\n"); dev->driver->shutdown(dev); } + + device_unlock(dev); + if (dev->parent) + device_unlock(dev->parent); + put_device(dev); + put_device(dev->parent); spin_lock(&devices_kset->list_lock); } -- cgit v1.2.3 From 5a7689fd5b4f2094e7a32beae67f290f8619b042 Mon Sep 17 00:00:00 2001 From: Sebastian Ott Date: Mon, 2 Jul 2012 19:08:15 +0200 Subject: driver core: move uevent call to driver_register Device driver attribute groups are created after userspace is notified via an add event. Fix this by moving the kobject_uevent call to driver_register after the attribute groups are added. Signed-off-by: Sebastian Ott Signed-off-by: Greg Kroah-Hartman --- drivers/base/bus.c | 1 - drivers/base/driver.c | 3 +++ 2 files changed, 3 insertions(+), 1 deletion(-) (limited to 'drivers/base') diff --git a/drivers/base/bus.c b/drivers/base/bus.c index 2bcef657a60c..181ed2660b33 100644 --- a/drivers/base/bus.c +++ b/drivers/base/bus.c @@ -743,7 +743,6 @@ int bus_add_driver(struct device_driver *drv) } } - kobject_uevent(&priv->kobj, KOBJ_ADD); return 0; out_unregister: diff --git a/drivers/base/driver.c b/drivers/base/driver.c index 207c27ddf828..1b500d6fcc2e 100644 --- a/drivers/base/driver.c +++ b/drivers/base/driver.c @@ -187,6 +187,9 @@ int driver_register(struct device_driver *drv) ret = driver_add_groups(drv, drv->groups); if (ret) bus_remove_driver(drv); + + kobject_uevent(&drv->p->kobj, KOBJ_ADD); + return ret; } EXPORT_SYMBOL_GPL(driver_register); -- cgit v1.2.3 From 8153584e3fdf78753bf653d5f583b6ecb86e5e70 Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Thu, 5 Jul 2012 14:04:44 +0100 Subject: driver core: Move deferred devices to the end of dpm_list before probing When deferred probe was originally added the idea was that devices which defer their probes would move themselves to the end of dpm_list in order to try to keep the assumptions that we're making about the list being in roughly the order things should be suspended correct. However this hasn't been what's been happening and doing it requires a lot of duplicated code to do the moves. Instead take a simple, brute force solution and have the deferred probe code push devices to the end of dpm_list before it retries the probe. This does mean we lock the dpm_list a bit more often but it's very simple and the code shouldn't be a fast path. We do the move with the deferred mutex dropped since doing things with fewer locks held simultaneously seems like a good idea. This approach was most recently suggested by Grant Likely. Signed-off-by: Mark Brown Acked-by: Rafael J. Wysocki Cc: Grant Likely , Signed-off-by: Greg Kroah-Hartman --- drivers/base/dd.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) (limited to 'drivers/base') diff --git a/drivers/base/dd.c b/drivers/base/dd.c index 6cd2c6ca9b0d..9b0aca479580 100644 --- a/drivers/base/dd.c +++ b/drivers/base/dd.c @@ -85,8 +85,20 @@ static void deferred_probe_work_func(struct work_struct *work) * manipulate the deferred list */ mutex_unlock(&deferred_probe_mutex); + + /* + * Force the device to the end of the dpm_list since + * the PM code assumes that the order we add things to + * the list is a good order for suspend but deferred + * probe makes that very unsafe. + */ + device_pm_lock(); + device_pm_move_last(dev); + device_pm_unlock(); + dev_dbg(dev, "Retrying from deferred list\n"); bus_probe_device(dev); + mutex_lock(&deferred_probe_mutex); put_device(dev); -- cgit v1.2.3 From b0d1f807f340e91b27aa721f6474956af11571da Mon Sep 17 00:00:00 2001 From: Lars-Peter Clausen Date: Tue, 3 Jul 2012 18:49:36 +0200 Subject: driver-core: Use kobj_to_dev instead of re-implementing it Signed-off-by: Lars-Peter Clausen Signed-off-by: Greg Kroah-Hartman --- drivers/base/core.c | 17 ++++++++--------- drivers/base/firmware_class.c | 6 ++---- 2 files changed, 10 insertions(+), 13 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/core.c b/drivers/base/core.c index 65849a9aeece..f338037a4f3d 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -85,14 +85,13 @@ const char *dev_driver_string(const struct device *dev) } EXPORT_SYMBOL(dev_driver_string); -#define to_dev(obj) container_of(obj, struct device, kobj) #define to_dev_attr(_attr) container_of(_attr, struct device_attribute, attr) static ssize_t dev_attr_show(struct kobject *kobj, struct attribute *attr, char *buf) { struct device_attribute *dev_attr = to_dev_attr(attr); - struct device *dev = to_dev(kobj); + struct device *dev = kobj_to_dev(kobj); ssize_t ret = -EIO; if (dev_attr->show) @@ -108,7 +107,7 @@ static ssize_t dev_attr_store(struct kobject *kobj, struct attribute *attr, const char *buf, size_t count) { struct device_attribute *dev_attr = to_dev_attr(attr); - struct device *dev = to_dev(kobj); + struct device *dev = kobj_to_dev(kobj); ssize_t ret = -EIO; if (dev_attr->store) @@ -182,7 +181,7 @@ EXPORT_SYMBOL_GPL(device_show_int); */ static void device_release(struct kobject *kobj) { - struct device *dev = to_dev(kobj); + struct device *dev = kobj_to_dev(kobj); struct device_private *p = dev->p; if (dev->release) @@ -200,7 +199,7 @@ static void device_release(struct kobject *kobj) static const void *device_namespace(struct kobject *kobj) { - struct device *dev = to_dev(kobj); + struct device *dev = kobj_to_dev(kobj); const void *ns = NULL; if (dev->class && dev->class->ns_type) @@ -221,7 +220,7 @@ static int dev_uevent_filter(struct kset *kset, struct kobject *kobj) struct kobj_type *ktype = get_ktype(kobj); if (ktype == &device_ktype) { - struct device *dev = to_dev(kobj); + struct device *dev = kobj_to_dev(kobj); if (dev->bus) return 1; if (dev->class) @@ -232,7 +231,7 @@ static int dev_uevent_filter(struct kset *kset, struct kobject *kobj) static const char *dev_uevent_name(struct kset *kset, struct kobject *kobj) { - struct device *dev = to_dev(kobj); + struct device *dev = kobj_to_dev(kobj); if (dev->bus) return dev->bus->name; @@ -244,7 +243,7 @@ static const char *dev_uevent_name(struct kset *kset, struct kobject *kobj) static int dev_uevent(struct kset *kset, struct kobject *kobj, struct kobj_uevent_env *env) { - struct device *dev = to_dev(kobj); + struct device *dev = kobj_to_dev(kobj); int retval = 0; /* add device node properties if present */ @@ -1132,7 +1131,7 @@ int device_register(struct device *dev) */ struct device *get_device(struct device *dev) { - return dev ? to_dev(kobject_get(&dev->kobj)) : NULL; + return dev ? kobj_to_dev(kobject_get(&dev->kobj)) : NULL; } /** diff --git a/drivers/base/firmware_class.c b/drivers/base/firmware_class.c index 5401814c874d..803cfc1597a9 100644 --- a/drivers/base/firmware_class.c +++ b/drivers/base/firmware_class.c @@ -22,8 +22,6 @@ #include #include -#define to_dev(obj) container_of(obj, struct device, kobj) - MODULE_AUTHOR("Manuel Estrada Sainz"); MODULE_DESCRIPTION("Multi purpose firmware loading support"); MODULE_LICENSE("GPL"); @@ -290,7 +288,7 @@ static ssize_t firmware_data_read(struct file *filp, struct kobject *kobj, struct bin_attribute *bin_attr, char *buffer, loff_t offset, size_t count) { - struct device *dev = to_dev(kobj); + struct device *dev = kobj_to_dev(kobj); struct firmware_priv *fw_priv = to_firmware_priv(dev); struct firmware *fw; ssize_t ret_count; @@ -384,7 +382,7 @@ static ssize_t firmware_data_write(struct file *filp, struct kobject *kobj, struct bin_attribute *bin_attr, char *buffer, loff_t offset, size_t count) { - struct device *dev = to_dev(kobj); + struct device *dev = kobj_to_dev(kobj); struct firmware_priv *fw_priv = to_firmware_priv(dev); struct firmware *fw; ssize_t retval; -- cgit v1.2.3 From eed5d2150752bd08b22333d739f3120151773d28 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Thu, 12 Jul 2012 11:51:48 +0200 Subject: PM / Runtime: Do not increment device usage counts before probing The pm_runtime_get_noresume() calls before really_probe() and before executing __device_attach() for each driver on the device's bus cause problems to happen if probing fails and if the driver has enabled runtime PM for the device in its .probe() callback. Namely, in that case, if the device has been resumed by the driver after enabling its runtime PM and if it turns out that .probe() should return an error, the driver is supposed to suspend the device and disable its runtime PM before exiting .probe(). However, because the device's runtime PM usage counter was incremented by the core before calling .probe(), the driver's attempt to suspend the device will not succeed and the device will remain in the full-power state after the failing .probe() has returned. To fix this issue, remove the pm_runtime_get_noresume() calls from driver_probe_device() and from device_attach() and replace the corresponding pm_runtime_put_sync() calls with pm_runtime_idle() to preserve the existing behavior (which is to check if the device is idle and to suspend it eventually in that case after probing). Reported-and-tested-by: Kevin Hilman Reviewed-by: Kevin Hilman Signed-off-by: Rafael J. Wysocki Signed-off-by: Greg Kroah-Hartman --- drivers/base/dd.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/dd.c b/drivers/base/dd.c index 9b0aca479580..e3bbed8a617c 100644 --- a/drivers/base/dd.c +++ b/drivers/base/dd.c @@ -369,10 +369,9 @@ int driver_probe_device(struct device_driver *drv, struct device *dev) pr_debug("bus: '%s': %s: matched device %s with driver %s\n", drv->bus->name, __func__, dev_name(dev), drv->name); - pm_runtime_get_noresume(dev); pm_runtime_barrier(dev); ret = really_probe(dev, drv); - pm_runtime_put_sync(dev); + pm_runtime_idle(dev); return ret; } @@ -419,9 +418,8 @@ int device_attach(struct device *dev) ret = 0; } } else { - pm_runtime_get_noresume(dev); ret = bus_for_each_drv(dev->bus, NULL, dev, __device_attach); - pm_runtime_put_sync(dev); + pm_runtime_idle(dev); } out_unlock: device_unlock(dev); -- cgit v1.2.3 From a14af325641794d1ce8e676e9c4967342349195c Mon Sep 17 00:00:00 2001 From: Sebastian Ott Date: Tue, 17 Jul 2012 10:39:10 +0200 Subject: driver core: don't trigger uevent after failure Do not send the uevent if driver_add_groups failed. Reported-by: Ming Lei Signed-off-by: Sebastian Ott Signed-off-by: Greg Kroah-Hartman --- drivers/base/driver.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/driver.c b/drivers/base/driver.c index 1b500d6fcc2e..974e301a1ef0 100644 --- a/drivers/base/driver.c +++ b/drivers/base/driver.c @@ -185,9 +185,10 @@ int driver_register(struct device_driver *drv) if (ret) return ret; ret = driver_add_groups(drv, drv->groups); - if (ret) + if (ret) { bus_remove_driver(drv); - + return ret; + } kobject_uevent(&drv->p->kobj, KOBJ_ADD); return ret; -- cgit v1.2.3 From 463351194d826af7ae97fff0ed6f0fe2a4fed615 Mon Sep 17 00:00:00 2001 From: Dimitris Papastamos Date: Wed, 18 Jul 2012 14:17:23 +0100 Subject: regmap: Fix incorrect arguments to kzalloc() call Signed-off-by: Dimitris Papastamos Signed-off-by: Mark Brown --- drivers/base/regmap/regmap-mmio.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/base') diff --git a/drivers/base/regmap/regmap-mmio.c b/drivers/base/regmap/regmap-mmio.c index d64a7fc71df8..f05fc74dd84a 100644 --- a/drivers/base/regmap/regmap-mmio.c +++ b/drivers/base/regmap/regmap-mmio.c @@ -172,7 +172,7 @@ static struct regmap_mmio_context *regmap_mmio_gen_context(void __iomem *regs, return ERR_PTR(-EINVAL); } - ctx = kzalloc(GFP_KERNEL, sizeof(*ctx)); + ctx = kzalloc(sizeof(*ctx), GFP_KERNEL); if (!ctx) return ERR_PTR(-ENOMEM); -- cgit v1.2.3 From 7664e969359917ba3212c5d6efdec494eaf8586e Mon Sep 17 00:00:00 2001 From: Sachin Kamat Date: Tue, 17 Jul 2012 22:38:08 +0200 Subject: PM / Sleep: Add missing static storage class specifiers in main.c Fix the following sparse warnings: drivers/base/power/main.c:48:1: warning: symbol 'dpm_prepared_list' was not declared. Should it be static? drivers/base/power/main.c:49:1: warning: symbol 'dpm_suspended_list' was not declared. Should it be static? drivers/base/power/main.c:50:1: warning: symbol 'dpm_late_early_list' was not declared. Should it be static? drivers/base/power/main.c:51:1: warning: symbol 'dpm_noirq_list' was not declared. Should it be static? Signed-off-by: Sachin Kamat Signed-off-by: Rafael J. Wysocki --- drivers/base/power/main.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index df5f41d2ec95..dbc3dbea5b25 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -45,10 +45,10 @@ typedef int (*pm_callback_t)(struct device *); */ LIST_HEAD(dpm_list); -LIST_HEAD(dpm_prepared_list); -LIST_HEAD(dpm_suspended_list); -LIST_HEAD(dpm_late_early_list); -LIST_HEAD(dpm_noirq_list); +static LIST_HEAD(dpm_prepared_list); +static LIST_HEAD(dpm_suspended_list); +static LIST_HEAD(dpm_late_early_list); +static LIST_HEAD(dpm_noirq_list); struct suspend_stats suspend_stats; static DEFINE_MUTEX(dpm_list_mtx); -- cgit v1.2.3 From ad0446eb11b6886a73e077bd087f98544b8c0403 Mon Sep 17 00:00:00 2001 From: Sachin Kamat Date: Tue, 17 Jul 2012 22:38:26 +0200 Subject: PM / QoS: Use NULL pointer instead of plain integer in qos.c Fix the following sparse warning: drivers/base/power/qos.c:465:29: warning: Using plain integer as NULL pointer Signed-off-by: Sachin Kamat Signed-off-by: Rafael J. Wysocki --- drivers/base/power/qos.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/base') diff --git a/drivers/base/power/qos.c b/drivers/base/power/qos.c index fd849a2c4fa8..74a67e0019a2 100644 --- a/drivers/base/power/qos.c +++ b/drivers/base/power/qos.c @@ -462,7 +462,7 @@ EXPORT_SYMBOL_GPL(dev_pm_qos_add_ancestor_request); static void __dev_pm_qos_drop_user_request(struct device *dev) { dev_pm_qos_remove_request(dev->power.pq_req); - dev->power.pq_req = 0; + dev->power.pq_req = NULL; } /** -- cgit v1.2.3 From eea03c20ae38a55405c0865ed9adfccc400e4c8e Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Wed, 18 Jul 2012 18:15:46 -0700 Subject: Make wait_for_device_probe() also do scsi_complete_async_scans() Commit a7a20d103994 ("sd: limit the scope of the async probe domain") make the SCSI device probing run device discovery in it's own async domain. However, as a result, the partition detection was no longer synchronized by async_synchronize_full() (which, despite the name, only synchronizes the global async space, not all of them). Which in turn meant that "wait_for_device_probe()" would not wait for the SCSI partitions to be parsed. And "wait_for_device_probe()" was what the boot time init code relied on for mounting the root filesystem. Now, most people never noticed this, because not only is it timing-dependent, but modern distributions all use initrd. So the root filesystem isn't actually on a disk at all. And then before they actually mount the final disk filesystem, they will have loaded the scsi-wait-scan module, which not only does the expected wait_for_device_probe(), but also does scsi_complete_async_scans(). [ Side note: scsi_complete_async_scans() had also been partially broken, but that was fixed in commit 43a8d39d0137 ("fix async probe regression"), so that same commit a7a20d103994 had actually broken setups even if you used scsi-wait-scan explicitly ] Solve this problem by just moving the scsi_complete_async_scans() call into wait_for_device_probe(). Everybody who wants to wait for device probing to finish really wants the SCSI probing to complete, so there's no reason not to do this. So now "wait_for_device_probe()" really does what the name implies, and properly waits for device probing to finish. This also removes the now unnecessary extra calls to scsi_complete_async_scans(). Reported-and-tested-by: Artem S. Tashkinov Cc: Dan Williams Cc: Alan Stern Cc: James Bottomley Cc: Borislav Petkov Cc: linux-scsi Signed-off-by: Linus Torvalds --- drivers/base/dd.c | 2 ++ drivers/scsi/scsi_wait_scan.c | 5 ----- include/linux/device.h | 2 -- kernel/power/hibernate.c | 8 -------- kernel/power/user.c | 2 -- 5 files changed, 2 insertions(+), 17 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/dd.c b/drivers/base/dd.c index dcb8a6e48692..4b01ab3d2c24 100644 --- a/drivers/base/dd.c +++ b/drivers/base/dd.c @@ -24,6 +24,7 @@ #include #include #include +#include #include "base.h" #include "power/power.h" @@ -332,6 +333,7 @@ void wait_for_device_probe(void) /* wait for the known devices to complete their probing */ wait_event(probe_waitqueue, atomic_read(&probe_count) == 0); async_synchronize_full(); + scsi_complete_async_scans(); } EXPORT_SYMBOL_GPL(wait_for_device_probe); diff --git a/drivers/scsi/scsi_wait_scan.c b/drivers/scsi/scsi_wait_scan.c index ae7814874618..072734538876 100644 --- a/drivers/scsi/scsi_wait_scan.c +++ b/drivers/scsi/scsi_wait_scan.c @@ -22,11 +22,6 @@ static int __init wait_scan_init(void) * and might not yet have reached the scsi async scanning */ wait_for_device_probe(); - /* - * and then we wait for the actual asynchronous scsi scan - * to finish. - */ - scsi_complete_async_scans(); return 0; } diff --git a/include/linux/device.h b/include/linux/device.h index 161d96241b1b..6de94151ff6f 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -865,8 +865,6 @@ extern int (*platform_notify_remove)(struct device *dev); extern struct device *get_device(struct device *dev); extern void put_device(struct device *dev); -extern void wait_for_device_probe(void); - #ifdef CONFIG_DEVTMPFS extern int devtmpfs_create_node(struct device *dev); extern int devtmpfs_delete_node(struct device *dev); diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c index 8b53db38a279..238025f5472e 100644 --- a/kernel/power/hibernate.c +++ b/kernel/power/hibernate.c @@ -27,7 +27,6 @@ #include #include #include -#include #include "power.h" @@ -748,13 +747,6 @@ static int software_resume(void) async_synchronize_full(); } - /* - * We can't depend on SCSI devices being available after loading - * one of their modules until scsi_complete_async_scans() is - * called and the resume device usually is a SCSI one. - */ - scsi_complete_async_scans(); - swsusp_resume_device = name_to_dev_t(resume_file); if (!swsusp_resume_device) { error = -ENODEV; diff --git a/kernel/power/user.c b/kernel/power/user.c index 91b0fd021a95..4ed81e74f86f 100644 --- a/kernel/power/user.c +++ b/kernel/power/user.c @@ -24,7 +24,6 @@ #include #include #include -#include #include @@ -84,7 +83,6 @@ static int snapshot_open(struct inode *inode, struct file *filp) * appear. */ wait_for_device_probe(); - scsi_complete_async_scans(); data->swap = -1; data->mode = O_WRONLY; -- cgit v1.2.3 From 064b021fbe470ecc9ca10f9f87af48c0fc0865fb Mon Sep 17 00:00:00 2001 From: Colin Cross Date: Thu, 19 Jul 2012 10:38:06 +0200 Subject: PM / Sleep: call early resume handlers when suspend_noirq fails Commit cf579dfb82550e34de7ccf3ef090d8b834ccd3a9 (PM / Sleep: Introduce "late suspend" and "early resume" of devices) introduced a bug where suspend_late handlers would be called, but if dpm_suspend_noirq returned an error the early_resume handlers would never be called. All devices would end up on the dpm_late_early_list, and would never be resumed again. Fix it by calling dpm_resume_early when dpm_suspend_noirq returns an error. Signed-off-by: Colin Cross Cc: stable@vger.kernel.org Signed-off-by: Rafael J. Wysocki --- drivers/base/power/main.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) (limited to 'drivers/base') diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index dbc3dbea5b25..c642f8dab1df 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -989,8 +989,16 @@ static int dpm_suspend_late(pm_message_t state) int dpm_suspend_end(pm_message_t state) { int error = dpm_suspend_late(state); + if (error) + return error; - return error ? : dpm_suspend_noirq(state); + error = dpm_suspend_noirq(state); + if (error) { + dpm_resume_early(state); + return error; + } + + return 0; } EXPORT_SYMBOL_GPL(dpm_suspend_end); -- cgit v1.2.3 From 492d542273a4859f8bf8cc7744cdf71ef50b39ea Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Mon, 9 Jul 2012 19:33:40 -0700 Subject: [SCSI] cleanup usages of scsi_complete_async_scans Now that scsi registers its async scan work with the async subsystem, wait_for_device_probe() is sufficient for ensuring all scanning is complete. [jejb: fix merge problems with eea03c20ae38 Make wait_for_device_probe() also do scsi_complete_async_scans()] Signed-off-by: Dan Williams Tested-by: Eldad Zack Signed-off-by: James Bottomley --- drivers/base/dd.c | 2 -- drivers/scsi/scsi_scan.c | 12 ------------ include/scsi/scsi_scan.h | 11 ----------- 3 files changed, 25 deletions(-) delete mode 100644 include/scsi/scsi_scan.h (limited to 'drivers/base') diff --git a/drivers/base/dd.c b/drivers/base/dd.c index 4b01ab3d2c24..dcb8a6e48692 100644 --- a/drivers/base/dd.c +++ b/drivers/base/dd.c @@ -24,7 +24,6 @@ #include #include #include -#include #include "base.h" #include "power/power.h" @@ -333,7 +332,6 @@ void wait_for_device_probe(void) /* wait for the known devices to complete their probing */ wait_event(probe_waitqueue, atomic_read(&probe_count) == 0); async_synchronize_full(); - scsi_complete_async_scans(); } EXPORT_SYMBOL_GPL(wait_for_device_probe); diff --git a/drivers/scsi/scsi_scan.c b/drivers/scsi/scsi_scan.c index dff17c198a72..a0bc66372654 100644 --- a/drivers/scsi/scsi_scan.c +++ b/drivers/scsi/scsi_scan.c @@ -187,18 +187,6 @@ int scsi_complete_async_scans(void) return 0; } -/* Only exported for the benefit of scsi_wait_scan */ -EXPORT_SYMBOL_GPL(scsi_complete_async_scans); - -#ifndef MODULE -/* - * For async scanning we need to wait for all the scans to complete before - * trying to mount the root fs. Otherwise non-modular drivers may not be ready - * yet. - */ -late_initcall(scsi_complete_async_scans); -#endif - /** * scsi_unlock_floptical - unlock device via a special MODE SENSE command * @sdev: scsi device to send command to diff --git a/include/scsi/scsi_scan.h b/include/scsi/scsi_scan.h deleted file mode 100644 index 78898889243d..000000000000 --- a/include/scsi/scsi_scan.h +++ /dev/null @@ -1,11 +0,0 @@ -#ifndef _SCSI_SCSI_SCAN_H -#define _SCSI_SCSI_SCAN_H - -#ifdef CONFIG_SCSI -/* drivers/scsi/scsi_scan.c */ -extern int scsi_complete_async_scans(void); -#else -static inline int scsi_complete_async_scans(void) { return 0; } -#endif - -#endif /* _SCSI_SCSI_SCAN_H */ -- cgit v1.2.3 From 921a1650de9eed40dd64d681aba4a4d98856f289 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 20 Jul 2012 01:15:31 +0400 Subject: new helper: done_path_create() releases what needs to be released after {kern,user}_path_create() Signed-off-by: Al Viro --- arch/powerpc/platforms/cell/spufs/syscalls.c | 4 +--- drivers/base/devtmpfs.c | 9 ++------- fs/namei.c | 24 ++++++++++++------------ fs/ocfs2/refcounttree.c | 4 +--- include/linux/namei.h | 1 + net/unix/af_unix.c | 9 ++++----- 6 files changed, 21 insertions(+), 30 deletions(-) (limited to 'drivers/base') diff --git a/arch/powerpc/platforms/cell/spufs/syscalls.c b/arch/powerpc/platforms/cell/spufs/syscalls.c index 8591bb62d7fc..5b7d8ffbf890 100644 --- a/arch/powerpc/platforms/cell/spufs/syscalls.c +++ b/arch/powerpc/platforms/cell/spufs/syscalls.c @@ -70,9 +70,7 @@ static long do_spu_create(const char __user *pathname, unsigned int flags, ret = PTR_ERR(dentry); if (!IS_ERR(dentry)) { ret = spufs_create(&path, dentry, flags, mode, neighbor); - mutex_unlock(&path.dentry->d_inode->i_mutex); - dput(dentry); - path_put(&path); + done_path_create(&path, dentry); } return ret; diff --git a/drivers/base/devtmpfs.c b/drivers/base/devtmpfs.c index d91a3a0b2325..deb4a456cf83 100644 --- a/drivers/base/devtmpfs.c +++ b/drivers/base/devtmpfs.c @@ -156,9 +156,7 @@ static int dev_mkdir(const char *name, umode_t mode) if (!err) /* mark as kernel-created inode */ dentry->d_inode->i_private = &thread; - dput(dentry); - mutex_unlock(&path.dentry->d_inode->i_mutex); - path_put(&path); + done_path_create(&path, dentry); return err; } @@ -218,10 +216,7 @@ static int handle_create(const char *nodename, umode_t mode, struct device *dev) /* mark as kernel-created inode */ dentry->d_inode->i_private = &thread; } - dput(dentry); - - mutex_unlock(&path.dentry->d_inode->i_mutex); - path_put(&path); + done_path_create(&path, dentry); return err; } diff --git a/fs/namei.c b/fs/namei.c index 2ccc35c4dc24..5bc6f3d1dc8a 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -2893,6 +2893,14 @@ out: } EXPORT_SYMBOL(kern_path_create); +void done_path_create(struct path *path, struct dentry *dentry) +{ + dput(dentry); + mutex_unlock(&path->dentry->d_inode->i_mutex); + path_put(path); +} +EXPORT_SYMBOL(done_path_create); + struct dentry *user_path_create(int dfd, const char __user *pathname, struct path *path, int is_dir) { char *tmp = getname(pathname); @@ -2989,9 +2997,7 @@ SYSCALL_DEFINE4(mknodat, int, dfd, const char __user *, filename, umode_t, mode, out_drop_write: mnt_drop_write(path.mnt); out_dput: - dput(dentry); - mutex_unlock(&path.dentry->d_inode->i_mutex); - path_put(&path); + done_path_create(&path, dentry); return error; } @@ -3048,9 +3054,7 @@ SYSCALL_DEFINE3(mkdirat, int, dfd, const char __user *, pathname, umode_t, mode) out_drop_write: mnt_drop_write(path.mnt); out_dput: - dput(dentry); - mutex_unlock(&path.dentry->d_inode->i_mutex); - path_put(&path); + done_path_create(&path, dentry); return error; } @@ -3334,9 +3338,7 @@ SYSCALL_DEFINE3(symlinkat, const char __user *, oldname, out_drop_write: mnt_drop_write(path.mnt); out_dput: - dput(dentry); - mutex_unlock(&path.dentry->d_inode->i_mutex); - path_put(&path); + done_path_create(&path, dentry); out_putname: putname(from); return error; @@ -3446,9 +3448,7 @@ SYSCALL_DEFINE5(linkat, int, olddfd, const char __user *, oldname, out_drop_write: mnt_drop_write(new_path.mnt); out_dput: - dput(new_dentry); - mutex_unlock(&new_path.dentry->d_inode->i_mutex); - path_put(&new_path); + done_path_create(&new_path, new_dentry); out: path_put(&old_path); diff --git a/fs/ocfs2/refcounttree.c b/fs/ocfs2/refcounttree.c index 9f32d7cbb7a3..23cf78f68503 100644 --- a/fs/ocfs2/refcounttree.c +++ b/fs/ocfs2/refcounttree.c @@ -4477,9 +4477,7 @@ int ocfs2_reflink_ioctl(struct inode *inode, new_dentry, preserve); mnt_drop_write(new_path.mnt); out_dput: - dput(new_dentry); - mutex_unlock(&new_path.dentry->d_inode->i_mutex); - path_put(&new_path); + done_path_create(&new_path, new_dentry); out: path_put(&old_path); diff --git a/include/linux/namei.h b/include/linux/namei.h index d2ef8b34b967..4bf19d8174ed 100644 --- a/include/linux/namei.h +++ b/include/linux/namei.h @@ -67,6 +67,7 @@ extern int kern_path(const char *, unsigned, struct path *); extern struct dentry *kern_path_create(int, const char *, struct path *, int); extern struct dentry *user_path_create(int, const char __user *, struct path *, int); +extern void done_path_create(struct path *, struct dentry *); extern struct dentry *kern_path_locked(const char *, struct path *); extern int vfs_path_lookup(struct dentry *, struct vfsmount *, const char *, unsigned int, struct path *); diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 641f2e47f165..e8239540683a 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -887,8 +887,9 @@ out_mknod_drop_write: mnt_drop_write(path.mnt); if (err) goto out_mknod_dput; - mutex_unlock(&path.dentry->d_inode->i_mutex); - dput(path.dentry); + mntget(path.mnt); + dget(dentry); + done_path_create(&path, dentry); path.dentry = dentry; addr->hash = UNIX_HASH_SIZE; @@ -923,9 +924,7 @@ out: return err; out_mknod_dput: - dput(dentry); - mutex_unlock(&path.dentry->d_inode->i_mutex); - path_put(&path); + done_path_create(&path, dentry); out_mknod_parent: if (err == -EEXIST) err = -EADDRINUSE; -- cgit v1.2.3 From 64ccc9c033c6089b2d426dad3c56477ab066c999 Mon Sep 17 00:00:00 2001 From: Marek Szyprowski Date: Thu, 14 Jun 2012 13:03:04 +0200 Subject: common: dma-mapping: add support for generic dma_mmap_* calls Commit 9adc5374 ('common: dma-mapping: introduce mmap method') added a generic method for implementing mmap user call to dma_map_ops structure. This patch converts ARM and PowerPC architectures (the only providers of dma_mmap_coherent/dma_mmap_writecombine calls) to use this generic dma_map_ops based call and adds a generic cross architecture definition for dma_mmap_attrs, dma_mmap_coherent, dma_mmap_writecombine functions. The generic mmap virt_to_page-based fallback implementation is provided for architectures which don't provide their own implementation for mmap method. Signed-off-by: Marek Szyprowski Reviewed-by: Kyungmin Park --- arch/arm/include/asm/dma-mapping.h | 19 ---------------- arch/powerpc/include/asm/dma-mapping.h | 8 +++---- arch/powerpc/kernel/dma-iommu.c | 1 + arch/powerpc/kernel/dma-swiotlb.c | 1 + arch/powerpc/kernel/dma.c | 36 ++++++++++++++++--------------- arch/powerpc/kernel/vio.c | 1 + drivers/base/dma-mapping.c | 31 ++++++++++++++++++++++++++ include/asm-generic/dma-coherent.h | 1 + include/asm-generic/dma-mapping-common.h | 37 ++++++++++++++++++++++++++++++++ 9 files changed, 95 insertions(+), 40 deletions(-) (limited to 'drivers/base') diff --git a/arch/arm/include/asm/dma-mapping.h b/arch/arm/include/asm/dma-mapping.h index 80777d871422..a04803331144 100644 --- a/arch/arm/include/asm/dma-mapping.h +++ b/arch/arm/include/asm/dma-mapping.h @@ -186,17 +186,6 @@ extern int arm_dma_mmap(struct device *dev, struct vm_area_struct *vma, void *cpu_addr, dma_addr_t dma_addr, size_t size, struct dma_attrs *attrs); -#define dma_mmap_coherent(d, v, c, h, s) dma_mmap_attrs(d, v, c, h, s, NULL) - -static inline int dma_mmap_attrs(struct device *dev, struct vm_area_struct *vma, - void *cpu_addr, dma_addr_t dma_addr, - size_t size, struct dma_attrs *attrs) -{ - struct dma_map_ops *ops = get_dma_ops(dev); - BUG_ON(!ops); - return ops->mmap(dev, vma, cpu_addr, dma_addr, size, attrs); -} - static inline void *dma_alloc_writecombine(struct device *dev, size_t size, dma_addr_t *dma_handle, gfp_t flag) { @@ -213,14 +202,6 @@ static inline void dma_free_writecombine(struct device *dev, size_t size, return dma_free_attrs(dev, size, cpu_addr, dma_handle, &attrs); } -static inline int dma_mmap_writecombine(struct device *dev, struct vm_area_struct *vma, - void *cpu_addr, dma_addr_t dma_addr, size_t size) -{ - DEFINE_DMA_ATTRS(attrs); - dma_set_attr(DMA_ATTR_WRITE_COMBINE, &attrs); - return dma_mmap_attrs(dev, vma, cpu_addr, dma_addr, size, &attrs); -} - /* * This can be called during boot to increase the size of the consistent * DMA region above it's default value of 2MB. It must be called before the diff --git a/arch/powerpc/include/asm/dma-mapping.h b/arch/powerpc/include/asm/dma-mapping.h index 62678e365ca0..78160874809a 100644 --- a/arch/powerpc/include/asm/dma-mapping.h +++ b/arch/powerpc/include/asm/dma-mapping.h @@ -27,7 +27,10 @@ extern void *dma_direct_alloc_coherent(struct device *dev, size_t size, extern void dma_direct_free_coherent(struct device *dev, size_t size, void *vaddr, dma_addr_t dma_handle, struct dma_attrs *attrs); - +extern int dma_direct_mmap_coherent(struct device *dev, + struct vm_area_struct *vma, + void *cpu_addr, dma_addr_t handle, + size_t size, struct dma_attrs *attrs); #ifdef CONFIG_NOT_COHERENT_CACHE /* @@ -207,11 +210,8 @@ static inline phys_addr_t dma_to_phys(struct device *dev, dma_addr_t daddr) #define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f) #define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h) -extern int dma_mmap_coherent(struct device *, struct vm_area_struct *, - void *, dma_addr_t, size_t); #define ARCH_HAS_DMA_MMAP_COHERENT - static inline void dma_cache_sync(struct device *dev, void *vaddr, size_t size, enum dma_data_direction direction) { diff --git a/arch/powerpc/kernel/dma-iommu.c b/arch/powerpc/kernel/dma-iommu.c index bcfdcd22c766..2d7bb8ced136 100644 --- a/arch/powerpc/kernel/dma-iommu.c +++ b/arch/powerpc/kernel/dma-iommu.c @@ -109,6 +109,7 @@ static u64 dma_iommu_get_required_mask(struct device *dev) struct dma_map_ops dma_iommu_ops = { .alloc = dma_iommu_alloc_coherent, .free = dma_iommu_free_coherent, + .mmap = dma_direct_mmap_coherent, .map_sg = dma_iommu_map_sg, .unmap_sg = dma_iommu_unmap_sg, .dma_supported = dma_iommu_dma_supported, diff --git a/arch/powerpc/kernel/dma-swiotlb.c b/arch/powerpc/kernel/dma-swiotlb.c index 4ab88dafb235..46943651da23 100644 --- a/arch/powerpc/kernel/dma-swiotlb.c +++ b/arch/powerpc/kernel/dma-swiotlb.c @@ -49,6 +49,7 @@ static u64 swiotlb_powerpc_get_required(struct device *dev) struct dma_map_ops swiotlb_dma_ops = { .alloc = dma_direct_alloc_coherent, .free = dma_direct_free_coherent, + .mmap = dma_direct_mmap_coherent, .map_sg = swiotlb_map_sg_attrs, .unmap_sg = swiotlb_unmap_sg_attrs, .dma_supported = swiotlb_dma_supported, diff --git a/arch/powerpc/kernel/dma.c b/arch/powerpc/kernel/dma.c index b1ec983dcec8..062bf20e6dd4 100644 --- a/arch/powerpc/kernel/dma.c +++ b/arch/powerpc/kernel/dma.c @@ -65,6 +65,24 @@ void dma_direct_free_coherent(struct device *dev, size_t size, #endif } +int dma_direct_mmap_coherent(struct device *dev, struct vm_area_struct *vma, + void *cpu_addr, dma_addr_t handle, size_t size, + struct dma_attrs *attrs) +{ + unsigned long pfn; + +#ifdef CONFIG_NOT_COHERENT_CACHE + vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot); + pfn = __dma_get_coherent_pfn((unsigned long)cpu_addr); +#else + pfn = page_to_pfn(virt_to_page(cpu_addr)); +#endif + return remap_pfn_range(vma, vma->vm_start, + pfn + vma->vm_pgoff, + vma->vm_end - vma->vm_start, + vma->vm_page_prot); +} + static int dma_direct_map_sg(struct device *dev, struct scatterlist *sgl, int nents, enum dma_data_direction direction, struct dma_attrs *attrs) @@ -154,6 +172,7 @@ static inline void dma_direct_sync_single(struct device *dev, struct dma_map_ops dma_direct_ops = { .alloc = dma_direct_alloc_coherent, .free = dma_direct_free_coherent, + .mmap = dma_direct_mmap_coherent, .map_sg = dma_direct_map_sg, .unmap_sg = dma_direct_unmap_sg, .dma_supported = dma_direct_dma_supported, @@ -211,20 +230,3 @@ static int __init dma_init(void) } fs_initcall(dma_init); -int dma_mmap_coherent(struct device *dev, struct vm_area_struct *vma, - void *cpu_addr, dma_addr_t handle, size_t size) -{ - unsigned long pfn; - -#ifdef CONFIG_NOT_COHERENT_CACHE - vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot); - pfn = __dma_get_coherent_pfn((unsigned long)cpu_addr); -#else - pfn = page_to_pfn(virt_to_page(cpu_addr)); -#endif - return remap_pfn_range(vma, vma->vm_start, - pfn + vma->vm_pgoff, - vma->vm_end - vma->vm_start, - vma->vm_page_prot); -} -EXPORT_SYMBOL_GPL(dma_mmap_coherent); diff --git a/arch/powerpc/kernel/vio.c b/arch/powerpc/kernel/vio.c index cb87301ccd55..dda3d9ad1094 100644 --- a/arch/powerpc/kernel/vio.c +++ b/arch/powerpc/kernel/vio.c @@ -613,6 +613,7 @@ static u64 vio_dma_get_required_mask(struct device *dev) struct dma_map_ops vio_dma_mapping_ops = { .alloc = vio_dma_iommu_alloc_coherent, .free = vio_dma_iommu_free_coherent, + .mmap = dma_direct_mmap_coherent, .map_sg = vio_dma_iommu_map_sg, .unmap_sg = vio_dma_iommu_unmap_sg, .map_page = vio_dma_iommu_map_page, diff --git a/drivers/base/dma-mapping.c b/drivers/base/dma-mapping.c index 6f3676f1559f..db5db02e885f 100644 --- a/drivers/base/dma-mapping.c +++ b/drivers/base/dma-mapping.c @@ -10,6 +10,7 @@ #include #include #include +#include /* * Managed DMA API @@ -218,3 +219,33 @@ void dmam_release_declared_memory(struct device *dev) EXPORT_SYMBOL(dmam_release_declared_memory); #endif + +/* + * Create userspace mapping for the DMA-coherent memory. + */ +int dma_common_mmap(struct device *dev, struct vm_area_struct *vma, + void *cpu_addr, dma_addr_t dma_addr, size_t size) +{ + int ret = -ENXIO; +#ifdef CONFIG_MMU + unsigned long user_count = (vma->vm_end - vma->vm_start) >> PAGE_SHIFT; + unsigned long count = PAGE_ALIGN(size) >> PAGE_SHIFT; + unsigned long pfn = page_to_pfn(virt_to_page(cpu_addr)); + unsigned long off = vma->vm_pgoff; + + vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot); + + if (dma_mmap_from_coherent(dev, vma, cpu_addr, size, &ret)) + return ret; + + if (off < count && user_count <= (count - off)) { + ret = remap_pfn_range(vma, vma->vm_start, + pfn + off, + user_count << PAGE_SHIFT, + vma->vm_page_prot); + } +#endif /* CONFIG_MMU */ + + return ret; +} +EXPORT_SYMBOL(dma_common_mmap); diff --git a/include/asm-generic/dma-coherent.h b/include/asm-generic/dma-coherent.h index abfb2682de7f..2be8a2dbc868 100644 --- a/include/asm-generic/dma-coherent.h +++ b/include/asm-generic/dma-coherent.h @@ -29,6 +29,7 @@ dma_mark_declared_memory_occupied(struct device *dev, #else #define dma_alloc_from_coherent(dev, size, handle, ret) (0) #define dma_release_from_coherent(dev, order, vaddr) (0) +#define dma_mmap_from_coherent(dev, vma, vaddr, order, ret) (0) #endif #endif diff --git a/include/asm-generic/dma-mapping-common.h b/include/asm-generic/dma-mapping-common.h index 2e248d8924dc..9073aeb3bb1a 100644 --- a/include/asm-generic/dma-mapping-common.h +++ b/include/asm-generic/dma-mapping-common.h @@ -176,4 +176,41 @@ dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg, #define dma_map_sg(d, s, n, r) dma_map_sg_attrs(d, s, n, r, NULL) #define dma_unmap_sg(d, s, n, r) dma_unmap_sg_attrs(d, s, n, r, NULL) +extern int dma_common_mmap(struct device *dev, struct vm_area_struct *vma, + void *cpu_addr, dma_addr_t dma_addr, size_t size); + +/** + * dma_mmap_attrs - map a coherent DMA allocation into user space + * @dev: valid struct device pointer, or NULL for ISA and EISA-like devices + * @vma: vm_area_struct describing requested user mapping + * @cpu_addr: kernel CPU-view address returned from dma_alloc_attrs + * @handle: device-view address returned from dma_alloc_attrs + * @size: size of memory originally requested in dma_alloc_attrs + * @attrs: attributes of mapping properties requested in dma_alloc_attrs + * + * Map a coherent DMA buffer previously allocated by dma_alloc_attrs + * into user space. The coherent DMA buffer must not be freed by the + * driver until the user space mapping has been released. + */ +static inline int +dma_mmap_attrs(struct device *dev, struct vm_area_struct *vma, void *cpu_addr, + dma_addr_t dma_addr, size_t size, struct dma_attrs *attrs) +{ + struct dma_map_ops *ops = get_dma_ops(dev); + BUG_ON(!ops); + if (ops->mmap) + return ops->mmap(dev, vma, cpu_addr, dma_addr, size, attrs); + return dma_common_mmap(dev, vma, cpu_addr, dma_addr, size); +} + +#define dma_mmap_coherent(d, v, c, h, s) dma_mmap_attrs(d, v, c, h, s, NULL) + +static inline int dma_mmap_writecombine(struct device *dev, struct vm_area_struct *vma, + void *cpu_addr, dma_addr_t dma_addr, size_t size) +{ + DEFINE_DMA_ATTRS(attrs); + dma_set_attr(DMA_ATTR_WRITE_COMBINE, &attrs); + return dma_mmap_attrs(dev, vma, cpu_addr, dma_addr, size, &attrs); +} + #endif -- cgit v1.2.3 From d2b7428eb0caa7c66e34b6ac869a43915b294123 Mon Sep 17 00:00:00 2001 From: Marek Szyprowski Date: Wed, 13 Jun 2012 10:05:52 +0200 Subject: common: dma-mapping: introduce dma_get_sgtable() function This patch adds dma_get_sgtable() function which is required to let drivers to share the buffers allocated by DMA-mapping subsystem. Right now the driver gets a dma address of the allocated buffer and the kernel virtual mapping for it. If it wants to share it with other device (= map into its dma address space) it usually hacks around kernel virtual addresses to get pointers to pages or assumes that both devices share the DMA address space. Both solutions are just hacks for the special cases, which should be avoided in the final version of buffer sharing. To solve this issue in a generic way, a new call to DMA mapping has been introduced - dma_get_sgtable(). It allocates a scatter-list which describes the allocated buffer and lets the driver(s) to use it with other device(s) by calling dma_map_sg() on it. This patch provides a generic implementation based on virt_to_page() call. Architectures which require more sophisticated translation might provide their own get_sgtable() methods. Signed-off-by: Marek Szyprowski Reviewed-by: Kyungmin Park Reviewed-by: Daniel Vetter --- drivers/base/dma-mapping.c | 18 ++++++++++++++++++ include/asm-generic/dma-mapping-common.h | 18 ++++++++++++++++++ include/linux/dma-mapping.h | 3 +++ 3 files changed, 39 insertions(+) (limited to 'drivers/base') diff --git a/drivers/base/dma-mapping.c b/drivers/base/dma-mapping.c index db5db02e885f..3fbedc75e7c5 100644 --- a/drivers/base/dma-mapping.c +++ b/drivers/base/dma-mapping.c @@ -218,6 +218,24 @@ void dmam_release_declared_memory(struct device *dev) } EXPORT_SYMBOL(dmam_release_declared_memory); +/* + * Create scatter-list for the already allocated DMA buffer. + */ +int dma_common_get_sgtable(struct device *dev, struct sg_table *sgt, + void *cpu_addr, dma_addr_t handle, size_t size) +{ + struct page *page = virt_to_page(cpu_addr); + int ret; + + ret = sg_alloc_table(sgt, 1, GFP_KERNEL); + if (unlikely(ret)) + return ret; + + sg_set_page(sgt->sgl, page, PAGE_ALIGN(size), 0); + return 0; +} +EXPORT_SYMBOL(dma_common_get_sgtable); + #endif /* diff --git a/include/asm-generic/dma-mapping-common.h b/include/asm-generic/dma-mapping-common.h index 9073aeb3bb1a..de8bf89940f8 100644 --- a/include/asm-generic/dma-mapping-common.h +++ b/include/asm-generic/dma-mapping-common.h @@ -213,4 +213,22 @@ static inline int dma_mmap_writecombine(struct device *dev, struct vm_area_struc return dma_mmap_attrs(dev, vma, cpu_addr, dma_addr, size, &attrs); } +int +dma_common_get_sgtable(struct device *dev, struct sg_table *sgt, + void *cpu_addr, dma_addr_t dma_addr, size_t size); + +static inline int +dma_get_sgtable_attrs(struct device *dev, struct sg_table *sgt, void *cpu_addr, + dma_addr_t dma_addr, size_t size, struct dma_attrs *attrs) +{ + struct dma_map_ops *ops = get_dma_ops(dev); + BUG_ON(!ops); + if (ops->get_sgtable) + return ops->get_sgtable(dev, sgt, cpu_addr, dma_addr, size, + attrs); + return dma_common_get_sgtable(dev, sgt, cpu_addr, dma_addr, size); +} + +#define dma_get_sgtable(d, t, v, h, s) dma_get_sgtable_attrs(d, t, v, h, s, NULL) + #endif diff --git a/include/linux/dma-mapping.h b/include/linux/dma-mapping.h index dfc099e56a66..94af41858513 100644 --- a/include/linux/dma-mapping.h +++ b/include/linux/dma-mapping.h @@ -18,6 +18,9 @@ struct dma_map_ops { int (*mmap)(struct device *, struct vm_area_struct *, void *, dma_addr_t, size_t, struct dma_attrs *attrs); + int (*get_sgtable)(struct device *dev, struct sg_table *sgt, void *, + dma_addr_t, size_t, struct dma_attrs *attrs); + dma_addr_t (*map_page)(struct device *dev, struct page *page, unsigned long offset, size_t size, enum dma_data_direction dir, -- cgit v1.2.3 From ee6f509c3274014d1f52e7a7a10aee9f85393c5e Mon Sep 17 00:00:00 2001 From: Minchan Kim Date: Tue, 31 Jul 2012 16:43:50 -0700 Subject: mm: factor out memory isolate functions mm/page_alloc.c has some memory isolation functions but they are used only when we enable CONFIG_{CMA|MEMORY_HOTPLUG|MEMORY_FAILURE}. So let's make it configurable by new CONFIG_MEMORY_ISOLATION so that it can reduce binary size and we can check it simple by CONFIG_MEMORY_ISOLATION, not if defined CONFIG_{CMA|MEMORY_HOTPLUG|MEMORY_FAILURE}. Signed-off-by: Minchan Kim Cc: Andi Kleen Cc: Marek Szyprowski Acked-by: KAMEZAWA Hiroyuki Cc: KOSAKI Motohiro Cc: Mel Gorman Cc: Michal Hocko Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/base/Kconfig | 1 + include/linux/page-isolation.h | 13 ++++--- mm/Kconfig | 5 +++ mm/Makefile | 5 +-- mm/page_alloc.c | 80 +++--------------------------------------- mm/page_isolation.c | 71 +++++++++++++++++++++++++++++++++++++ 6 files changed, 93 insertions(+), 82 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/Kconfig b/drivers/base/Kconfig index 9b21469482ae..08b4c5209384 100644 --- a/drivers/base/Kconfig +++ b/drivers/base/Kconfig @@ -196,6 +196,7 @@ config CMA bool "Contiguous Memory Allocator (EXPERIMENTAL)" depends on HAVE_DMA_CONTIGUOUS && HAVE_MEMBLOCK && EXPERIMENTAL select MIGRATION + select MEMORY_ISOLATION help This enables the Contiguous Memory Allocator which allows drivers to allocate big physically-contiguous blocks of memory for use with diff --git a/include/linux/page-isolation.h b/include/linux/page-isolation.h index 3bdcab30ca41..105077aa7685 100644 --- a/include/linux/page-isolation.h +++ b/include/linux/page-isolation.h @@ -1,6 +1,11 @@ #ifndef __LINUX_PAGEISOLATION_H #define __LINUX_PAGEISOLATION_H + +bool has_unmovable_pages(struct zone *zone, struct page *page, int count); +void set_pageblock_migratetype(struct page *page, int migratetype); +int move_freepages_block(struct zone *zone, struct page *page, + int migratetype); /* * Changes migrate type in [start_pfn, end_pfn) to be MIGRATE_ISOLATE. * If specified range includes migrate types other than MOVABLE or CMA, @@ -10,7 +15,7 @@ * free all pages in the range. test_page_isolated() can be used for * test it. */ -extern int +int start_isolate_page_range(unsigned long start_pfn, unsigned long end_pfn, unsigned migratetype); @@ -18,7 +23,7 @@ start_isolate_page_range(unsigned long start_pfn, unsigned long end_pfn, * Changes MIGRATE_ISOLATE to MIGRATE_MOVABLE. * target range is [start_pfn, end_pfn) */ -extern int +int undo_isolate_page_range(unsigned long start_pfn, unsigned long end_pfn, unsigned migratetype); @@ -30,8 +35,8 @@ int test_pages_isolated(unsigned long start_pfn, unsigned long end_pfn); /* * Internal functions. Changes pageblock's migrate type. */ -extern int set_migratetype_isolate(struct page *page); -extern void unset_migratetype_isolate(struct page *page, unsigned migratetype); +int set_migratetype_isolate(struct page *page); +void unset_migratetype_isolate(struct page *page, unsigned migratetype); #endif diff --git a/mm/Kconfig b/mm/Kconfig index 82fed4eb2b6f..d5c8019c6627 100644 --- a/mm/Kconfig +++ b/mm/Kconfig @@ -140,9 +140,13 @@ config ARCH_DISCARD_MEMBLOCK config NO_BOOTMEM boolean +config MEMORY_ISOLATION + boolean + # eventually, we can have this option just 'select SPARSEMEM' config MEMORY_HOTPLUG bool "Allow for memory hot-add" + select MEMORY_ISOLATION depends on SPARSEMEM || X86_64_ACPI_NUMA depends on HOTPLUG && ARCH_ENABLE_MEMORY_HOTPLUG depends on (IA64 || X86 || PPC_BOOK3S_64 || SUPERH || S390) @@ -272,6 +276,7 @@ config MEMORY_FAILURE depends on MMU depends on ARCH_SUPPORTS_MEMORY_FAILURE bool "Enable recovery from hardware memory errors" + select MEMORY_ISOLATION help Enables code to recover from some memory failures on systems with MCA recovery. This allows a system to continue running diff --git a/mm/Makefile b/mm/Makefile index 290bbfe33698..92753e2d82da 100644 --- a/mm/Makefile +++ b/mm/Makefile @@ -15,8 +15,8 @@ obj-y := filemap.o mempool.o oom_kill.o fadvise.o \ maccess.o page_alloc.o page-writeback.o \ readahead.o swap.o truncate.o vmscan.o shmem.o \ prio_tree.o util.o mmzone.o vmstat.o backing-dev.o \ - page_isolation.o mm_init.o mmu_context.o percpu.o \ - compaction.o slab_common.o $(mmu-y) + mm_init.o mmu_context.o percpu.o slab_common.o \ + compaction.o $(mmu-y) obj-y += init-mm.o @@ -56,3 +56,4 @@ obj-$(CONFIG_HWPOISON_INJECT) += hwpoison-inject.o obj-$(CONFIG_DEBUG_KMEMLEAK) += kmemleak.o obj-$(CONFIG_DEBUG_KMEMLEAK_TEST) += kmemleak-test.o obj-$(CONFIG_CLEANCACHE) += cleancache.o +obj-$(CONFIG_MEMORY_ISOLATION) += page_isolation.o diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 667338e80e94..228194728ccd 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -51,7 +51,6 @@ #include #include #include -#include #include #include #include @@ -219,7 +218,7 @@ EXPORT_SYMBOL(nr_online_nodes); int page_group_by_mobility_disabled __read_mostly; -static void set_pageblock_migratetype(struct page *page, int migratetype) +void set_pageblock_migratetype(struct page *page, int migratetype) { if (unlikely(page_group_by_mobility_disabled)) @@ -954,7 +953,7 @@ static int move_freepages(struct zone *zone, return pages_moved; } -static int move_freepages_block(struct zone *zone, struct page *page, +int move_freepages_block(struct zone *zone, struct page *page, int migratetype) { unsigned long start_pfn, end_pfn; @@ -5463,8 +5462,7 @@ void set_pageblock_flags_group(struct page *page, unsigned long flags, * MIGRATE_MOVABLE block might include unmovable pages. It means you can't * expect this function should be exact. */ -static bool -__has_unmovable_pages(struct zone *zone, struct page *page, int count) +bool has_unmovable_pages(struct zone *zone, struct page *page, int count) { unsigned long pfn, iter, found; int mt; @@ -5541,77 +5539,7 @@ bool is_pageblock_removable_nolock(struct page *page) zone->zone_start_pfn + zone->spanned_pages <= pfn) return false; - return !__has_unmovable_pages(zone, page, 0); -} - -int set_migratetype_isolate(struct page *page) -{ - struct zone *zone; - unsigned long flags, pfn; - struct memory_isolate_notify arg; - int notifier_ret; - int ret = -EBUSY; - - zone = page_zone(page); - - spin_lock_irqsave(&zone->lock, flags); - - pfn = page_to_pfn(page); - arg.start_pfn = pfn; - arg.nr_pages = pageblock_nr_pages; - arg.pages_found = 0; - - /* - * It may be possible to isolate a pageblock even if the - * migratetype is not MIGRATE_MOVABLE. The memory isolation - * notifier chain is used by balloon drivers to return the - * number of pages in a range that are held by the balloon - * driver to shrink memory. If all the pages are accounted for - * by balloons, are free, or on the LRU, isolation can continue. - * Later, for example, when memory hotplug notifier runs, these - * pages reported as "can be isolated" should be isolated(freed) - * by the balloon driver through the memory notifier chain. - */ - notifier_ret = memory_isolate_notify(MEM_ISOLATE_COUNT, &arg); - notifier_ret = notifier_to_errno(notifier_ret); - if (notifier_ret) - goto out; - /* - * FIXME: Now, memory hotplug doesn't call shrink_slab() by itself. - * We just check MOVABLE pages. - */ - if (!__has_unmovable_pages(zone, page, arg.pages_found)) - ret = 0; - /* - * Unmovable means "not-on-lru" pages. If Unmovable pages are - * larger than removable-by-driver pages reported by notifier, - * we'll fail. - */ - -out: - if (!ret) { - set_pageblock_migratetype(page, MIGRATE_ISOLATE); - move_freepages_block(zone, page, MIGRATE_ISOLATE); - } - - spin_unlock_irqrestore(&zone->lock, flags); - if (!ret) - drain_all_pages(); - return ret; -} - -void unset_migratetype_isolate(struct page *page, unsigned migratetype) -{ - struct zone *zone; - unsigned long flags; - zone = page_zone(page); - spin_lock_irqsave(&zone->lock, flags); - if (get_pageblock_migratetype(page) != MIGRATE_ISOLATE) - goto out; - set_pageblock_migratetype(page, migratetype); - move_freepages_block(zone, page, migratetype); -out: - spin_unlock_irqrestore(&zone->lock, flags); + return !has_unmovable_pages(zone, page, 0); } #ifdef CONFIG_CMA diff --git a/mm/page_isolation.c b/mm/page_isolation.c index c9f04774f2b8..fb482cf438da 100644 --- a/mm/page_isolation.c +++ b/mm/page_isolation.c @@ -5,8 +5,79 @@ #include #include #include +#include #include "internal.h" +int set_migratetype_isolate(struct page *page) +{ + struct zone *zone; + unsigned long flags, pfn; + struct memory_isolate_notify arg; + int notifier_ret; + int ret = -EBUSY; + + zone = page_zone(page); + + spin_lock_irqsave(&zone->lock, flags); + + pfn = page_to_pfn(page); + arg.start_pfn = pfn; + arg.nr_pages = pageblock_nr_pages; + arg.pages_found = 0; + + /* + * It may be possible to isolate a pageblock even if the + * migratetype is not MIGRATE_MOVABLE. The memory isolation + * notifier chain is used by balloon drivers to return the + * number of pages in a range that are held by the balloon + * driver to shrink memory. If all the pages are accounted for + * by balloons, are free, or on the LRU, isolation can continue. + * Later, for example, when memory hotplug notifier runs, these + * pages reported as "can be isolated" should be isolated(freed) + * by the balloon driver through the memory notifier chain. + */ + notifier_ret = memory_isolate_notify(MEM_ISOLATE_COUNT, &arg); + notifier_ret = notifier_to_errno(notifier_ret); + if (notifier_ret) + goto out; + /* + * FIXME: Now, memory hotplug doesn't call shrink_slab() by itself. + * We just check MOVABLE pages. + */ + if (!has_unmovable_pages(zone, page, arg.pages_found)) + ret = 0; + + /* + * immobile means "not-on-lru" paes. If immobile is larger than + * removable-by-driver pages reported by notifier, we'll fail. + */ + +out: + if (!ret) { + set_pageblock_migratetype(page, MIGRATE_ISOLATE); + move_freepages_block(zone, page, MIGRATE_ISOLATE); + } + + spin_unlock_irqrestore(&zone->lock, flags); + if (!ret) + drain_all_pages(); + return ret; +} + +void unset_migratetype_isolate(struct page *page, unsigned migratetype) +{ + struct zone *zone; + unsigned long flags; + zone = page_zone(page); + spin_lock_irqsave(&zone->lock, flags); + if (get_pageblock_migratetype(page) != MIGRATE_ISOLATE) + goto out; + set_pageblock_migratetype(page, migratetype); + move_freepages_block(zone, page, migratetype); +out: + spin_unlock_irqrestore(&zone->lock, flags); +} + static inline struct page * __first_valid_page(unsigned long pfn, unsigned long nr_pages) { -- cgit v1.2.3 From 77254950bb174d637cd88d1a74e137ecf3185a65 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Tue, 7 Aug 2012 13:50:14 +0200 Subject: PM: Make dev_pm_get_subsys_data() always return 0 on success MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Commits 1d5fcfec22 (PM / Domains: Add device domain data reference counter) and 62d4490294 (PM / Domains: Allow device callbacks to be added at any time) added checks for the return value of dev_pm_get_subsys_data(), but those checks were incorrect, because that function returned 1 on success in some cases. Since all of the existing users of dev_pm_get_subsys_data() don't use the positive value returned by it on success, change its definition so that it always returns 0 when successful. Reported-by: Heiko Stübner Reported-by: Tushar Behera Signed-off-by: Rafael J. Wysocki --- drivers/base/power/clock_ops.c | 3 +-- drivers/base/power/common.c | 4 +--- 2 files changed, 2 insertions(+), 5 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/power/clock_ops.c b/drivers/base/power/clock_ops.c index 869d7ff2227f..eb78e9640c4a 100644 --- a/drivers/base/power/clock_ops.c +++ b/drivers/base/power/clock_ops.c @@ -169,8 +169,7 @@ void pm_clk_init(struct device *dev) */ int pm_clk_create(struct device *dev) { - int ret = dev_pm_get_subsys_data(dev); - return ret < 0 ? ret : 0; + return dev_pm_get_subsys_data(dev); } /** diff --git a/drivers/base/power/common.c b/drivers/base/power/common.c index a14085cc613f..39c32529b833 100644 --- a/drivers/base/power/common.c +++ b/drivers/base/power/common.c @@ -24,7 +24,6 @@ int dev_pm_get_subsys_data(struct device *dev) { struct pm_subsys_data *psd; - int ret = 0; psd = kzalloc(sizeof(*psd), GFP_KERNEL); if (!psd) @@ -40,7 +39,6 @@ int dev_pm_get_subsys_data(struct device *dev) dev->power.subsys_data = psd; pm_clk_init(dev); psd = NULL; - ret = 1; } spin_unlock_irq(&dev->power.lock); @@ -48,7 +46,7 @@ int dev_pm_get_subsys_data(struct device *dev) /* kfree() verifies that its argument is nonzero. */ kfree(psd); - return ret; + return 0; } EXPORT_SYMBOL_GPL(dev_pm_get_subsys_data); -- cgit v1.2.3