From 3b034b0d084221596bf35c8d893e1d4d5477b9cc Mon Sep 17 00:00:00 2001 From: Vivek Goyal Date: Tue, 24 Nov 2009 15:50:03 +0900 Subject: percpu: Fix kdump failure if booted with percpu_alloc=page o kdump functionality reserves a per cpu area at boot time and exports the physical address of that area to user space through sys interface. This area stores some dump related information like cpu register states etc at the time of crash. o We were assuming that per cpu area always come from linearly mapped meory region and using __pa() to determine physical address. With percpu_alloc=page, per cpu area can come from vmalloc region also and __pa() breaks. o This patch implments a new function to convert per cpu address to physical address. Before the patch, crash_notes addresses looked as follows. cpu0 60fffff49800 cpu1 60fffff60800 cpu2 60fffff77800 These are bogus phsyical addresses. After the patch, address are following. cpu0 13eb44000 cpu1 13eb43000 cpu2 13eb42000 cpu3 13eb41000 These look fine. I got 4G of memory and /proc/iomem tell me following. 100000000-13fffffff : System RAM tj: * added missing asm/io.h include reported by Stephen Rothwell * repositioned per_cpu_ptr_phys() in percpu.c and added comment. Signed-off-by: Vivek Goyal Signed-off-by: Tejun Heo Cc: Stephen Rothwell --- drivers/base/cpu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/base') diff --git a/drivers/base/cpu.c b/drivers/base/cpu.c index e62a4ccea54d..69ee5b7517ec 100644 --- a/drivers/base/cpu.c +++ b/drivers/base/cpu.c @@ -97,7 +97,7 @@ static ssize_t show_crash_notes(struct sys_device *dev, struct sysdev_attribute * boot up and this data does not change there after. Hence this * operation should be safe. No locking required. */ - addr = __pa(per_cpu_ptr(crash_notes, cpunum)); + addr = per_cpu_ptr_to_phys(per_cpu_ptr(crash_notes, cpunum)); rc = sprintf(buf, "%Lx\n", addr); return rc; } -- cgit v1.2.3 From 12633e803a2a556f6469e0933d08233d0844a2d9 Mon Sep 17 00:00:00 2001 From: Nathan Fontenot Date: Wed, 25 Nov 2009 17:23:25 +0000 Subject: sysfs/cpu: Add probe/release files Version 3 of this patch is updated with documentation added to Documentation/ABI. There are no changes to any of the C code from v2 of the patch. In order to support kernel DLPAR of CPU resources we need to provide an interface to add (probe) and remove (release) the resource from the system. This patch Creates new generic probe and release sysfs files to facilitate cpu probe/release. The probe/release interface provides for allowing each arch to supply their own routines for implementing the backend of adding and removing cpus to/from the system. This also creates the powerpc specific stubs to handle the arch callouts from writes to the sysfs files. The creation and use of these files is regulated by the CONFIG_ARCH_CPU_PROBE_RELEASE option so that only architectures that need the capability will have the files created. Signed-off-by: Nathan Fontenot Signed-off-by: Benjamin Herrenschmidt --- drivers/base/cpu.c | 32 ++++++++++++++++++++++++++++++++ 1 file changed, 32 insertions(+) (limited to 'drivers/base') diff --git a/drivers/base/cpu.c b/drivers/base/cpu.c index e62a4ccea54d..7c03af7b84a9 100644 --- a/drivers/base/cpu.c +++ b/drivers/base/cpu.c @@ -72,6 +72,38 @@ void unregister_cpu(struct cpu *cpu) per_cpu(cpu_sys_devices, logical_cpu) = NULL; return; } + +#ifdef CONFIG_ARCH_CPU_PROBE_RELEASE +static ssize_t cpu_probe_store(struct class *class, const char *buf, + size_t count) +{ + return arch_cpu_probe(buf, count); +} + +static ssize_t cpu_release_store(struct class *class, const char *buf, + size_t count) +{ + return arch_cpu_release(buf, count); +} + +static CLASS_ATTR(probe, S_IWUSR, NULL, cpu_probe_store); +static CLASS_ATTR(release, S_IWUSR, NULL, cpu_release_store); + +int __init cpu_probe_release_init(void) +{ + int rc; + + rc = sysfs_create_file(&cpu_sysdev_class.kset.kobj, + &class_attr_probe.attr); + if (!rc) + rc = sysfs_create_file(&cpu_sysdev_class.kset.kobj, + &class_attr_release.attr); + + return rc; +} +device_initcall(cpu_probe_release_init); +#endif /* CONFIG_ARCH_CPU_PROBE_RELEASE */ + #else /* ... !CONFIG_HOTPLUG_CPU */ static inline void register_cpu_control(struct cpu *cpu) { -- cgit v1.2.3 From 51badebdcf394cc5fd574a524b55b3f6085e5e9c Mon Sep 17 00:00:00 2001 From: Gautham R Shenoy Date: Thu, 26 Nov 2009 09:59:05 +0000 Subject: powerpc/pseries: Serialize cpu hotplug operations during deactivate Vs deallocate Currently the cpu-allocation/deallocation process comprises of two steps: - Set the indicators and to update the device tree with DLPAR node information. - Online/offline the allocated/deallocated CPU. This is achieved by writing to the sysfs tunables "probe" during allocation and "release" during deallocation. At the sametime, the userspace can independently online/offline the CPUs of the system using the sysfs tunable "online". It is quite possible that when a userspace tool offlines a CPU for the purpose of deallocation and is in the process of updating the device tree, some other userspace tool could bring the CPU back online by writing to the "online" sysfs tunable thereby causing the deallocate process to fail. The solution to this is to serialize writes to the "probe/release" sysfs tunable with the writes to the "online" sysfs tunable. This patch employs a mutex to provide this serialization, which is a no-op on all architectures except PPC_PSERIES Signed-off-by: Gautham R Shenoy Acked-by: Vaidyanathan Srinivasan Signed-off-by: Benjamin Herrenschmidt --- drivers/base/cpu.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'drivers/base') diff --git a/drivers/base/cpu.c b/drivers/base/cpu.c index 7c03af7b84a9..27fd775375b0 100644 --- a/drivers/base/cpu.c +++ b/drivers/base/cpu.c @@ -35,6 +35,7 @@ static ssize_t __ref store_online(struct sys_device *dev, struct sysdev_attribut struct cpu *cpu = container_of(dev, struct cpu, sysdev); ssize_t ret; + cpu_hotplug_driver_lock(); switch (buf[0]) { case '0': ret = cpu_down(cpu->sysdev.id); @@ -49,6 +50,7 @@ static ssize_t __ref store_online(struct sys_device *dev, struct sysdev_attribut default: ret = -EINVAL; } + cpu_hotplug_driver_unlock(); if (ret >= 0) ret = count; -- cgit v1.2.3 From 0092699643703aefca6af0aa758a73f1624d53be Mon Sep 17 00:00:00 2001 From: Kay Sievers Date: Wed, 28 Oct 2009 19:50:57 +0100 Subject: Driver Core: devtmpfs: ignore umask while setting file mode Signed-off-by: Kay Sievers Signed-off-by: Greg Kroah-Hartman --- drivers/base/devtmpfs.c | 26 ++++++++++++++++---------- 1 file changed, 16 insertions(+), 10 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/devtmpfs.c b/drivers/base/devtmpfs.c index a1cb5afe6801..48526b935682 100644 --- a/drivers/base/devtmpfs.c +++ b/drivers/base/devtmpfs.c @@ -156,34 +156,40 @@ int devtmpfs_create_node(struct device *dev) mode |= S_IFCHR; curr_cred = override_creds(&init_cred); + err = vfs_path_lookup(dev_mnt->mnt_root, dev_mnt, nodename, LOOKUP_PARENT, &nd); if (err == -ENOENT) { - /* create missing parent directories */ create_path(nodename); err = vfs_path_lookup(dev_mnt->mnt_root, dev_mnt, nodename, LOOKUP_PARENT, &nd); - if (err) - goto out; } + if (err) + goto out; dentry = lookup_create(&nd, 0); if (!IS_ERR(dentry)) { - int umask; - - umask = sys_umask(0000); err = vfs_mknod(nd.path.dentry->d_inode, dentry, mode, dev->devt); - sys_umask(umask); - /* mark as kernel created inode */ - if (!err) + if (!err) { + struct iattr newattrs; + + /* fixup possibly umasked mode */ + newattrs.ia_mode = mode; + newattrs.ia_valid = ATTR_MODE; + mutex_lock(&dentry->d_inode->i_mutex); + notify_change(dentry, &newattrs); + mutex_unlock(&dentry->d_inode->i_mutex); + + /* mark as kernel-created inode */ dentry->d_inode->i_private = &dev_mnt; + } dput(dentry); } else { err = PTR_ERR(dentry); } - mutex_unlock(&nd.path.dentry->d_inode->i_mutex); + mutex_unlock(&nd.path.dentry->d_inode->i_mutex); path_put(&nd.path); out: kfree(tmp); -- cgit v1.2.3 From ed413ae6e7813d3227eef43bc6d84ca4f4fe6b21 Mon Sep 17 00:00:00 2001 From: Kay Sievers Date: Wed, 28 Oct 2009 19:51:06 +0100 Subject: Driver core: devtmpfs: prevent concurrent subdirectory creation and removal Signed-off-by: Kay Sievers Signed-off-by: Greg Kroah-Hartman --- drivers/base/devtmpfs.c | 26 ++++++++++++++++---------- 1 file changed, 16 insertions(+), 10 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/devtmpfs.c b/drivers/base/devtmpfs.c index 48526b935682..1cf498fd2b52 100644 --- a/drivers/base/devtmpfs.c +++ b/drivers/base/devtmpfs.c @@ -32,6 +32,8 @@ static int dev_mount = 1; static int dev_mount; #endif +static rwlock_t dirlock; + static int __init mount_param(char *str) { dev_mount = simple_strtoul(str, NULL, 0); @@ -86,16 +88,12 @@ static int dev_mkdir(const char *name, mode_t mode) static int create_path(const char *nodepath) { - char *path; struct nameidata nd; int err = 0; - path = kstrdup(nodepath, GFP_KERNEL); - if (!path) - return -ENOMEM; - + read_lock(&dirlock); err = vfs_path_lookup(dev_mnt->mnt_root, dev_mnt, - path, LOOKUP_PARENT, &nd); + nodepath, LOOKUP_PARENT, &nd); if (err == 0) { struct dentry *dentry; @@ -107,14 +105,17 @@ static int create_path(const char *nodepath) dput(dentry); } mutex_unlock(&nd.path.dentry->d_inode->i_mutex); - path_put(&nd.path); } else if (err == -ENOENT) { + char *path; char *s; /* parent directories do not exist, create them */ + path = kstrdup(nodepath, GFP_KERNEL); + if (!path) + return -ENOMEM; s = path; - while (1) { + for (;;) { s = strchr(s, '/'); if (!s) break; @@ -125,9 +126,10 @@ static int create_path(const char *nodepath) s[0] = '/'; s++; } + kfree(path); } + read_unlock(&dirlock); - kfree(path); return err; } @@ -234,7 +236,8 @@ static int delete_path(const char *nodepath) if (!path) return -ENOMEM; - while (1) { + write_lock(&dirlock); + for (;;) { char *base; base = strrchr(path, '/'); @@ -245,6 +248,7 @@ static int delete_path(const char *nodepath) if (err) break; } + write_unlock(&dirlock); kfree(path); return err; @@ -360,6 +364,8 @@ int __init devtmpfs_init(void) int err; struct vfsmount *mnt; + rwlock_init(&dirlock); + err = register_filesystem(&dev_fs_type); if (err) { printk(KERN_ERR "devtmpfs: unable to register devtmpfs " -- cgit v1.2.3 From 073120cc28ad9f6003452c8bb9d15a87b1820201 Mon Sep 17 00:00:00 2001 From: Kay Sievers Date: Wed, 28 Oct 2009 19:51:17 +0100 Subject: Driver Core: devtmpfs: use sys_mount() Signed-off-by: Kay Sievers Signed-off-by: Greg Kroah-Hartman --- drivers/base/devtmpfs.c | 9 ++------- 1 file changed, 2 insertions(+), 7 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/devtmpfs.c b/drivers/base/devtmpfs.c index 1cf498fd2b52..880a203b6688 100644 --- a/drivers/base/devtmpfs.c +++ b/drivers/base/devtmpfs.c @@ -332,9 +332,8 @@ out: * If configured, or requested by the commandline, devtmpfs will be * auto-mounted after the kernel mounted the root filesystem. */ -int devtmpfs_mount(const char *mountpoint) +int devtmpfs_mount(const char *mntdir) { - struct path path; int err; if (!dev_mount) @@ -343,15 +342,11 @@ int devtmpfs_mount(const char *mountpoint) if (!dev_mnt) return 0; - err = kern_path(mountpoint, LOOKUP_FOLLOW, &path); - if (err) - return err; - err = do_add_mount(dev_mnt, &path, 0, NULL); + err = sys_mount("devtmpfs", (char *)mntdir, "devtmpfs", MS_SILENT, NULL); if (err) printk(KERN_INFO "devtmpfs: error mounting %i\n", err); else printk(KERN_INFO "devtmpfs: mounted\n"); - path_put(&path); return err; } -- cgit v1.2.3 From 015bf43b07158668c2f38af463939afcc6d19403 Mon Sep 17 00:00:00 2001 From: Kay Sievers Date: Wed, 28 Oct 2009 19:51:26 +0100 Subject: Driver Core: devtmpfs: do not remove non-kernel-created directories Signed-off-by: Kay Sievers Signed-off-by: Greg Kroah-Hartman --- drivers/base/devtmpfs.c | 39 ++++++++++++++++----------------------- 1 file changed, 16 insertions(+), 23 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/devtmpfs.c b/drivers/base/devtmpfs.c index 880a203b6688..c7f5c08f8790 100644 --- a/drivers/base/devtmpfs.c +++ b/drivers/base/devtmpfs.c @@ -76,37 +76,26 @@ static int dev_mkdir(const char *name, mode_t mode) dentry = lookup_create(&nd, 1); if (!IS_ERR(dentry)) { err = vfs_mkdir(nd.path.dentry->d_inode, dentry, mode); + if (!err) + /* mark as kernel-created inode */ + dentry->d_inode->i_private = &dev_mnt; dput(dentry); } else { err = PTR_ERR(dentry); } - mutex_unlock(&nd.path.dentry->d_inode->i_mutex); + mutex_unlock(&nd.path.dentry->d_inode->i_mutex); path_put(&nd.path); return err; } static int create_path(const char *nodepath) { - struct nameidata nd; - int err = 0; + int err; read_lock(&dirlock); - err = vfs_path_lookup(dev_mnt->mnt_root, dev_mnt, - nodepath, LOOKUP_PARENT, &nd); - if (err == 0) { - struct dentry *dentry; - - /* create directory right away */ - dentry = lookup_create(&nd, 1); - if (!IS_ERR(dentry)) { - err = vfs_mkdir(nd.path.dentry->d_inode, - dentry, 0755); - dput(dentry); - } - mutex_unlock(&nd.path.dentry->d_inode->i_mutex); - path_put(&nd.path); - } else if (err == -ENOENT) { + err = dev_mkdir(nodepath, 0755); + if (err == -ENOENT) { char *path; char *s; @@ -129,7 +118,6 @@ static int create_path(const char *nodepath) kfree(path); } read_unlock(&dirlock); - return err; } @@ -213,16 +201,21 @@ static int dev_rmdir(const char *name) 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_rmdir(nd.path.dentry->d_inode, dentry); - else + if (dentry->d_inode) { + if (dentry->d_inode->i_private == &dev_mnt) + err = vfs_rmdir(nd.path.dentry->d_inode, + dentry); + else + err = -EPERM; + } else { err = -ENOENT; + } dput(dentry); } else { err = PTR_ERR(dentry); } - mutex_unlock(&nd.path.dentry->d_inode->i_mutex); + mutex_unlock(&nd.path.dentry->d_inode->i_mutex); path_put(&nd.path); return err; } -- cgit v1.2.3 From ad72956df2ce83f58be1dc4e503c78c28e414c2c Mon Sep 17 00:00:00 2001 From: Kay Sievers Date: Wed, 28 Oct 2009 19:51:37 +0100 Subject: Driver Core: devtmpfs: cleanup node on device creation error Signed-off-by: Kay Sievers Signed-off-by: Greg Kroah-Hartman --- drivers/base/core.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'drivers/base') diff --git a/drivers/base/core.c b/drivers/base/core.c index 6bee6af8d8e1..0d3c29d72215 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -986,6 +986,8 @@ done: AttrsError: device_remove_class_symlinks(dev); SymlinkError: + if (MAJOR(dev->devt)) + devtmpfs_delete_node(dev); if (MAJOR(dev->devt)) device_remove_sys_dev_entry(dev); devtattrError: -- cgit v1.2.3 From 03d673e6af6490371aaf64dfe1f84c658c48b71d Mon Sep 17 00:00:00 2001 From: Kay Sievers Date: Fri, 30 Oct 2009 12:48:32 +0100 Subject: Driver-Core: devtmpfs - set root directory mode to 0755 Signed-off-by: Kay Sievers Cc: Mark Rosenstand Signed-off-by: Greg Kroah-Hartman --- drivers/base/devtmpfs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/base') diff --git a/drivers/base/devtmpfs.c b/drivers/base/devtmpfs.c index c7f5c08f8790..50375bb8e51d 100644 --- a/drivers/base/devtmpfs.c +++ b/drivers/base/devtmpfs.c @@ -361,7 +361,7 @@ int __init devtmpfs_init(void) return err; } - mnt = kern_mount(&dev_fs_type); + mnt = kern_mount_data(&dev_fs_type, "mode=0755"); if (IS_ERR(mnt)) { err = PTR_ERR(mnt); printk(KERN_ERR "devtmpfs: unable to create devtmpfs %i\n", err); -- cgit v1.2.3 From 9ebfbd45f9d4ee9cd72529cf99e5f300eb398e67 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Thu, 29 Oct 2009 12:36:02 +0100 Subject: firmware_class: make request_firmware_nowait more useful Unfortunately, one cannot hold on to the struct firmware that request_firmware_nowait() hands off, which is needed in some cases. Allow this by requiring the callback to free it (via release_firmware). Additionally, give it a gfp_t parameter -- all the current users call it from a GFP_KERNEL context so the GFP_ATOMIC isn't necessary. This also marks an API break which is useful in a sense, although that is obviously not the primary purpose of this change. Signed-off-by: Johannes Berg Acked-by: Marcel Holtmann Cc: Ming Lei Cc: Catalin Marinas Cc: David Woodhouse Cc: Pavel Roskin Cc: Abhay Salunke Signed-off-by: Greg Kroah-Hartman --- drivers/base/firmware_class.c | 14 ++++++-------- 1 file changed, 6 insertions(+), 8 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/firmware_class.c b/drivers/base/firmware_class.c index 7376367bcb80..a95024166b66 100644 --- a/drivers/base/firmware_class.c +++ b/drivers/base/firmware_class.c @@ -601,12 +601,9 @@ request_firmware_work_func(void *arg) } ret = _request_firmware(&fw, fw_work->name, fw_work->device, fw_work->uevent); - if (ret < 0) - fw_work->cont(NULL, fw_work->context); - else { - fw_work->cont(fw, fw_work->context); - release_firmware(fw); - } + + fw_work->cont(fw, fw_work->context); + module_put(fw_work->module); kfree(fw_work); return ret; @@ -619,6 +616,7 @@ request_firmware_work_func(void *arg) * is non-zero else the firmware copy must be done manually. * @name: name of firmware file * @device: device for which firmware is being loaded + * @gfp: allocation flags * @context: will be passed over to @cont, and * @fw may be %NULL if firmware request fails. * @cont: function will be called asynchronously when the firmware @@ -631,12 +629,12 @@ request_firmware_work_func(void *arg) int request_firmware_nowait( struct module *module, int uevent, - const char *name, struct device *device, void *context, + const char *name, struct device *device, gfp_t gfp, void *context, void (*cont)(const struct firmware *fw, void *context)) { struct task_struct *task; struct firmware_work *fw_work = kmalloc(sizeof (struct firmware_work), - GFP_ATOMIC); + gfp); if (!fw_work) return -ENOMEM; -- cgit v1.2.3 From 18ef545e47c126abd87c013b762b5fbb574858ce Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Wed, 4 Nov 2009 02:50:28 -0800 Subject: Driver core: Don't remove kobjects in device_shutdown. device_shutdown is defined to just shutdown the hardware and to not clean up any kernel data structures. Therefore don't put the kobjects for /sys/dev and /sys/dev/block and /sys/dev/char. This ensures we don't remove /sys/dev/block and /sys/dev/char while we still have symlinks from there to the actual devices. Acked-by: Kay Sievers Signed-off-by: Eric W. Biederman Signed-off-by: Greg Kroah-Hartman --- drivers/base/core.c | 3 --- 1 file changed, 3 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/core.c b/drivers/base/core.c index 0d3c29d72215..353b13782165 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -1730,8 +1730,5 @@ void device_shutdown(void) dev->driver->shutdown(dev); } } - kobject_put(sysfs_dev_char_kobj); - kobject_put(sysfs_dev_block_kobj); - kobject_put(dev_kobj); async_synchronize_full(); } -- cgit v1.2.3 From c60e0504c8e4fa14179d0687d80ef25148dd6dd4 Mon Sep 17 00:00:00 2001 From: Magnus Damm Date: Fri, 27 Nov 2009 17:38:51 +0900 Subject: Driver Core: Early platform driver buffer Add early_platform_init_buffer() support and update the early platform driver code to allow passing parameters to the driver on the kernel command line. early_platform_init_buffer() simply allows early platform drivers to provide a pointer and length to a memory area where the remaining part of the kernel command line option will be stored. Needed to pass baud rate and other serial port options to the reworked early serial console code on SuperH. Signed-off-by: Magnus Damm Signed-off-by: Greg Kroah-Hartman --- drivers/base/platform.c | 29 ++++++++++++++++++++++------- 1 file changed, 22 insertions(+), 7 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/platform.c b/drivers/base/platform.c index 4fa954b07ac4..9d2ee25deaf5 100644 --- a/drivers/base/platform.c +++ b/drivers/base/platform.c @@ -1000,7 +1000,7 @@ static __initdata LIST_HEAD(early_platform_device_list); int __init early_platform_driver_register(struct early_platform_driver *epdrv, char *buf) { - unsigned long index; + char *tmp; int n; /* Simply add the driver to the end of the global list. @@ -1019,13 +1019,28 @@ int __init early_platform_driver_register(struct early_platform_driver *epdrv, if (buf && !strncmp(buf, epdrv->pdrv->driver.name, n)) { list_move(&epdrv->list, &early_platform_driver_list); - if (!strcmp(buf, epdrv->pdrv->driver.name)) + /* Allow passing parameters after device name */ + if (buf[n] == '\0' || buf[n] == ',') epdrv->requested_id = -1; - else if (buf[n] == '.' && strict_strtoul(&buf[n + 1], 10, - &index) == 0) - epdrv->requested_id = index; - else - epdrv->requested_id = EARLY_PLATFORM_ID_ERROR; + else { + epdrv->requested_id = simple_strtoul(&buf[n + 1], + &tmp, 10); + + if (buf[n] != '.' || (tmp == &buf[n + 1])) { + epdrv->requested_id = EARLY_PLATFORM_ID_ERROR; + n = 0; + } else + n += strcspn(&buf[n + 1], ",") + 1; + } + + if (buf[n] == ',') + n++; + + if (epdrv->bufsize) { + memcpy(epdrv->buffer, &buf[n], + min_t(int, epdrv->bufsize, strlen(&buf[n]) + 1)); + epdrv->buffer[epdrv->bufsize - 1] = '\0'; + } } return 0; -- cgit v1.2.3 From 3589972e51fac1e02d0aaa576fa47f568cb94d40 Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Fri, 4 Dec 2009 11:06:57 -0500 Subject: Driver core: fix race in dev_driver_string This patch (as1310) works around a race in dev_driver_string(). If the device is unbound while the function is running, dev->driver might become NULL after we test it and before we dereference it. Signed-off-by: Alan Stern Cc: stable Cc: Oliver Neukum Signed-off-by: Greg Kroah-Hartman --- drivers/base/core.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) (limited to 'drivers/base') diff --git a/drivers/base/core.c b/drivers/base/core.c index 353b13782165..f1290cbd1350 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -56,7 +56,14 @@ static inline int device_is_not_partition(struct device *dev) */ const char *dev_driver_string(const struct device *dev) { - return dev->driver ? dev->driver->name : + struct device_driver *drv; + + /* dev->driver can change to NULL underneath us because of unbinding, + * so be careful about accessing it. dev->bus and dev->class should + * never change once they are set, so they don't need special care. + */ + drv = ACCESS_ONCE(dev->driver); + return drv ? drv->name : (dev->bus ? dev->bus->name : (dev->class ? dev->class->name : "")); } -- cgit v1.2.3 From 9a30523066cde73c1442b76224bb540de9f9b0b0 Mon Sep 17 00:00:00 2001 From: Lee Schermerhorn Date: Mon, 14 Dec 2009 17:58:25 -0800 Subject: hugetlb: add per node hstate attributes Add the per huge page size control/query attributes to the per node sysdevs: /sys/devices/system/node/node/hugepages/hugepages-/ nr_hugepages - r/w free_huge_pages - r/o surplus_huge_pages - r/o The patch attempts to re-use/share as much of the existing global hstate attribute initialization and handling, and the "nodes_allowed" constraint processing as possible. Calling set_max_huge_pages() with no node indicates a change to global hstate parameters. In this case, any non-default task mempolicy will be used to generate the nodes_allowed mask. A valid node id indicates an update to that node's hstate parameters, and the count argument specifies the target count for the specified node. From this info, we compute the target global count for the hstate and construct a nodes_allowed node mask contain only the specified node. Setting the node specific nr_hugepages via the per node attribute effectively ignores any task mempolicy or cpuset constraints. With this patch: (me):ls /sys/devices/system/node/node0/hugepages/hugepages-2048kB ./ ../ free_hugepages nr_hugepages surplus_hugepages Starting from: Node 0 HugePages_Total: 0 Node 0 HugePages_Free: 0 Node 0 HugePages_Surp: 0 Node 1 HugePages_Total: 0 Node 1 HugePages_Free: 0 Node 1 HugePages_Surp: 0 Node 2 HugePages_Total: 0 Node 2 HugePages_Free: 0 Node 2 HugePages_Surp: 0 Node 3 HugePages_Total: 0 Node 3 HugePages_Free: 0 Node 3 HugePages_Surp: 0 vm.nr_hugepages = 0 Allocate 16 persistent huge pages on node 2: (me):echo 16 >/sys/devices/system/node/node2/hugepages/hugepages-2048kB/nr_hugepages [Note that this is equivalent to: numactl -m 2 hugeadmin --pool-pages-min 2M:+16 ] Yields: Node 0 HugePages_Total: 0 Node 0 HugePages_Free: 0 Node 0 HugePages_Surp: 0 Node 1 HugePages_Total: 0 Node 1 HugePages_Free: 0 Node 1 HugePages_Surp: 0 Node 2 HugePages_Total: 16 Node 2 HugePages_Free: 16 Node 2 HugePages_Surp: 0 Node 3 HugePages_Total: 0 Node 3 HugePages_Free: 0 Node 3 HugePages_Surp: 0 vm.nr_hugepages = 16 Global controls work as expected--reduce pool to 8 persistent huge pages: (me):echo 8 >/sys/kernel/mm/hugepages/hugepages-2048kB/nr_hugepages Node 0 HugePages_Total: 0 Node 0 HugePages_Free: 0 Node 0 HugePages_Surp: 0 Node 1 HugePages_Total: 0 Node 1 HugePages_Free: 0 Node 1 HugePages_Surp: 0 Node 2 HugePages_Total: 8 Node 2 HugePages_Free: 8 Node 2 HugePages_Surp: 0 Node 3 HugePages_Total: 0 Node 3 HugePages_Free: 0 Node 3 HugePages_Surp: 0 Signed-off-by: Lee Schermerhorn Acked-by: Mel Gorman Reviewed-by: Andi Kleen Cc: KAMEZAWA Hiroyuki Cc: Randy Dunlap Cc: Nishanth Aravamudan Cc: David Rientjes Cc: Adam Litke Cc: Andy Whitcroft Cc: Eric Whitney Cc: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/base/node.c | 39 +++++++++++++++++++++++++++++++++++++++ 1 file changed, 39 insertions(+) (limited to 'drivers/base') diff --git a/drivers/base/node.c b/drivers/base/node.c index 1fe5536d404f..f502711d28db 100644 --- a/drivers/base/node.c +++ b/drivers/base/node.c @@ -173,6 +173,43 @@ static ssize_t node_read_distance(struct sys_device * dev, } static SYSDEV_ATTR(distance, S_IRUGO, node_read_distance, NULL); +#ifdef CONFIG_HUGETLBFS +/* + * hugetlbfs per node attributes registration interface: + * When/if hugetlb[fs] subsystem initializes [sometime after this module], + * it will register its per node attributes for all nodes online at that + * time. It will also call register_hugetlbfs_with_node(), below, to + * register its attribute registration functions with this node driver. + * Once these hooks have been initialized, the node driver will call into + * the hugetlb module to [un]register attributes for hot-plugged nodes. + */ +static node_registration_func_t __hugetlb_register_node; +static node_registration_func_t __hugetlb_unregister_node; + +static inline void hugetlb_register_node(struct node *node) +{ + if (__hugetlb_register_node) + __hugetlb_register_node(node); +} + +static inline void hugetlb_unregister_node(struct node *node) +{ + if (__hugetlb_unregister_node) + __hugetlb_unregister_node(node); +} + +void register_hugetlbfs_with_node(node_registration_func_t doregister, + node_registration_func_t unregister) +{ + __hugetlb_register_node = doregister; + __hugetlb_unregister_node = unregister; +} +#else +static inline void hugetlb_register_node(struct node *node) {} + +static inline void hugetlb_unregister_node(struct node *node) {} +#endif + /* * register_node - Setup a sysfs device for a node. @@ -196,6 +233,7 @@ int register_node(struct node *node, int num, struct node *parent) sysdev_create_file(&node->sysdev, &attr_distance); scan_unevictable_register_node(node); + hugetlb_register_node(node); } return error; } @@ -216,6 +254,7 @@ void unregister_node(struct node *node) sysdev_remove_file(&node->sysdev, &attr_distance); scan_unevictable_unregister_node(node); + hugetlb_unregister_node(node); sysdev_unregister(&node->sysdev); } -- cgit v1.2.3 From 4faf8d950ec438c49ae4526b897c30f8a2cad741 Mon Sep 17 00:00:00 2001 From: Lee Schermerhorn Date: Mon, 14 Dec 2009 17:58:35 -0800 Subject: hugetlb: handle memory hot-plug events Register per node hstate attributes only for nodes with memory. As suggested by David Rientjes. With Memory Hotplug, memory can be added to a memoryless node and a node with memory can become memoryless. Therefore, add a memory on/off-line notifier callback to [un]register a node's attributes on transition to/from memoryless state. N.B., Only tested build, boot, libhugetlbfs regression. i.e., no memory hotplug testing. Signed-off-by: Lee Schermerhorn Reviewed-by: Andi Kleen Acked-by: David Rientjes Cc: KAMEZAWA Hiroyuki Cc: Lee Schermerhorn Cc: Mel Gorman Cc: Randy Dunlap Cc: Nishanth Aravamudan Cc: Adam Litke Cc: Andy Whitcroft Cc: Eric Whitney Cc: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/base/node.c | 53 ++++++++++++++++++++++++++++++++++++++++++++++++----- 1 file changed, 48 insertions(+), 5 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/node.c b/drivers/base/node.c index f502711d28db..9e218a6d4a5b 100644 --- a/drivers/base/node.c +++ b/drivers/base/node.c @@ -177,8 +177,8 @@ static SYSDEV_ATTR(distance, S_IRUGO, node_read_distance, NULL); /* * hugetlbfs per node attributes registration interface: * When/if hugetlb[fs] subsystem initializes [sometime after this module], - * it will register its per node attributes for all nodes online at that - * time. It will also call register_hugetlbfs_with_node(), below, to + * it will register its per node attributes for all online nodes with + * memory. It will also call register_hugetlbfs_with_node(), below, to * register its attribute registration functions with this node driver. * Once these hooks have been initialized, the node driver will call into * the hugetlb module to [un]register attributes for hot-plugged nodes. @@ -188,7 +188,8 @@ static node_registration_func_t __hugetlb_unregister_node; static inline void hugetlb_register_node(struct node *node) { - if (__hugetlb_register_node) + if (__hugetlb_register_node && + node_state(node->sysdev.id, N_HIGH_MEMORY)) __hugetlb_register_node(node); } @@ -233,6 +234,7 @@ int register_node(struct node *node, int num, struct node *parent) sysdev_create_file(&node->sysdev, &attr_distance); scan_unevictable_register_node(node); + hugetlb_register_node(node); } return error; @@ -254,7 +256,7 @@ void unregister_node(struct node *node) sysdev_remove_file(&node->sysdev, &attr_distance); scan_unevictable_unregister_node(node); - hugetlb_unregister_node(node); + hugetlb_unregister_node(node); /* no-op, if memoryless node */ sysdev_unregister(&node->sysdev); } @@ -384,8 +386,45 @@ static int link_mem_sections(int nid) } return err; } + +/* + * Handle per node hstate attribute [un]registration on transistions + * to/from memoryless state. + */ + +static int node_memory_callback(struct notifier_block *self, + unsigned long action, void *arg) +{ + struct memory_notify *mnb = arg; + int nid = mnb->status_change_nid; + + switch (action) { + case MEM_ONLINE: /* memory successfully brought online */ + if (nid != NUMA_NO_NODE) + hugetlb_register_node(&node_devices[nid]); + break; + case MEM_OFFLINE: /* or offline */ + if (nid != NUMA_NO_NODE) + hugetlb_unregister_node(&node_devices[nid]); + break; + case MEM_GOING_ONLINE: + case MEM_GOING_OFFLINE: + case MEM_CANCEL_ONLINE: + case MEM_CANCEL_OFFLINE: + default: + break; + } + + return NOTIFY_OK; +} #else static int link_mem_sections(int nid) { return 0; } + +static inline int node_memory_callback(struct notifier_block *self, + unsigned long action, void *arg) +{ + return NOTIFY_OK; +} #endif /* CONFIG_MEMORY_HOTPLUG_SPARSE */ int register_one_node(int nid) @@ -499,13 +538,17 @@ static int node_states_init(void) return err; } +#define NODE_CALLBACK_PRI 2 /* lower than SLAB */ static int __init register_node_type(void) { int ret; ret = sysdev_class_register(&node_class); - if (!ret) + if (!ret) { ret = node_states_init(); + hotplug_memory_notifier(node_memory_callback, + NODE_CALLBACK_PRI); + } /* * Note: we're not going to unregister the node class if we fail -- cgit v1.2.3 From 39da08cb074cf19cb249832a2a955dfb28837e65 Mon Sep 17 00:00:00 2001 From: Lee Schermerhorn Date: Mon, 14 Dec 2009 17:58:36 -0800 Subject: hugetlb: offload per node attribute registrations Offload the registration and unregistration of per node hstate sysfs attributes to a worker thread rather than attempt the allocation/attachment or detachment/freeing of the attributes in the context of the memory hotplug handler. I don't know that this is absolutely required, but the registration can sleep in allocations and other mem hot plug handlers do it this way. If it turns out this is NOT required, we can drop this patch. N.B., Only tested build, boot, libhugetlbfs regression. i.e., no memory hotplug testing. Signed-off-by: Lee Schermerhorn Reviewed-by: Andi Kleen Cc: KAMEZAWA Hiroyuki Cc: Lee Schermerhorn Cc: Mel Gorman Cc: Randy Dunlap Cc: Nishanth Aravamudan Cc: David Rientjes Cc: Adam Litke Cc: Andy Whitcroft Cc: Eric Whitney Cc: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/base/node.c | 57 +++++++++++++++++++++++++++++++++++++++++++---------- 1 file changed, 47 insertions(+), 10 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/node.c b/drivers/base/node.c index 9e218a6d4a5b..54e5d8eaf70e 100644 --- a/drivers/base/node.c +++ b/drivers/base/node.c @@ -186,11 +186,14 @@ static SYSDEV_ATTR(distance, S_IRUGO, node_read_distance, NULL); static node_registration_func_t __hugetlb_register_node; static node_registration_func_t __hugetlb_unregister_node; -static inline void hugetlb_register_node(struct node *node) +static inline bool hugetlb_register_node(struct node *node) { if (__hugetlb_register_node && - node_state(node->sysdev.id, N_HIGH_MEMORY)) + node_state(node->sysdev.id, N_HIGH_MEMORY)) { __hugetlb_register_node(node); + return true; + } + return false; } static inline void hugetlb_unregister_node(struct node *node) @@ -387,10 +390,31 @@ static int link_mem_sections(int nid) return err; } +#ifdef CONFIG_HUGETLBFS /* * Handle per node hstate attribute [un]registration on transistions * to/from memoryless state. */ +static void node_hugetlb_work(struct work_struct *work) +{ + struct node *node = container_of(work, struct node, node_work); + + /* + * We only get here when a node transitions to/from memoryless state. + * We can detect which transition occurred by examining whether the + * node has memory now. hugetlb_register_node() already check this + * so we try to register the attributes. If that fails, then the + * node has transitioned to memoryless, try to unregister the + * attributes. + */ + if (!hugetlb_register_node(node)) + hugetlb_unregister_node(node); +} + +static void init_node_hugetlb_work(int nid) +{ + INIT_WORK(&node_devices[nid].node_work, node_hugetlb_work); +} static int node_memory_callback(struct notifier_block *self, unsigned long action, void *arg) @@ -399,14 +423,16 @@ static int node_memory_callback(struct notifier_block *self, int nid = mnb->status_change_nid; switch (action) { - case MEM_ONLINE: /* memory successfully brought online */ + case MEM_ONLINE: + case MEM_OFFLINE: + /* + * offload per node hstate [un]registration to a work thread + * when transitioning to/from memoryless state. + */ if (nid != NUMA_NO_NODE) - hugetlb_register_node(&node_devices[nid]); - break; - case MEM_OFFLINE: /* or offline */ - if (nid != NUMA_NO_NODE) - hugetlb_unregister_node(&node_devices[nid]); + schedule_work(&node_devices[nid].node_work); break; + case MEM_GOING_ONLINE: case MEM_GOING_OFFLINE: case MEM_CANCEL_ONLINE: @@ -417,15 +443,23 @@ static int node_memory_callback(struct notifier_block *self, return NOTIFY_OK; } -#else +#endif /* CONFIG_HUGETLBFS */ +#else /* !CONFIG_MEMORY_HOTPLUG_SPARSE */ + static int link_mem_sections(int nid) { return 0; } +#endif /* CONFIG_MEMORY_HOTPLUG_SPARSE */ +#if !defined(CONFIG_MEMORY_HOTPLUG_SPARSE) || \ + !defined(CONFIG_HUGETLBFS) static inline int node_memory_callback(struct notifier_block *self, unsigned long action, void *arg) { return NOTIFY_OK; } -#endif /* CONFIG_MEMORY_HOTPLUG_SPARSE */ + +static void init_node_hugetlb_work(int nid) { } + +#endif int register_one_node(int nid) { @@ -449,6 +483,9 @@ int register_one_node(int nid) /* link memory sections under this node */ error = link_mem_sections(nid); + + /* initialize work queue for memory hot plug */ + init_node_hugetlb_work(nid); } return error; -- cgit v1.2.3 From dee5d0d518defd0337a41f1a504428c9acc87be5 Mon Sep 17 00:00:00 2001 From: Alex Chiang Date: Mon, 14 Dec 2009 17:59:05 -0800 Subject: mm: add numa node symlink for memory section in sysfs Commit c04fc586c (mm: show node to memory section relationship with symlinks in sysfs) created symlinks from nodes to memory sections, e.g. /sys/devices/system/node/node1/memory135 -> ../../memory/memory135 If you're examining the memory section though and are wondering what node it might belong to, you can find it by grovelling around in sysfs, but it's a little cumbersome. Add a reverse symlink for each memory section that points back to the node to which it belongs. Signed-off-by: Alex Chiang Cc: Gary Hade Cc: Badari Pulavarty Cc: Ingo Molnar Acked-by: David Rientjes Cc: Greg KH Cc: Randy Dunlap Cc: David Rientjes Cc: KOSAKI Motohiro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/base/node.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) (limited to 'drivers/base') diff --git a/drivers/base/node.c b/drivers/base/node.c index 54e5d8eaf70e..44eed11bbdf3 100644 --- a/drivers/base/node.c +++ b/drivers/base/node.c @@ -312,6 +312,7 @@ static int get_nid_for_pfn(unsigned long pfn) /* register memory section under specified node if it spans that node */ int register_mem_sect_under_node(struct memory_block *mem_blk, int nid) { + int ret; unsigned long pfn, sect_start_pfn, sect_end_pfn; if (!mem_blk) @@ -328,9 +329,15 @@ int register_mem_sect_under_node(struct memory_block *mem_blk, int nid) continue; if (page_nid != nid) continue; - return sysfs_create_link_nowarn(&node_devices[nid].sysdev.kobj, + ret = sysfs_create_link_nowarn(&node_devices[nid].sysdev.kobj, &mem_blk->sysdev.kobj, kobject_name(&mem_blk->sysdev.kobj)); + if (ret) + return ret; + + return sysfs_create_link_nowarn(&mem_blk->sysdev.kobj, + &node_devices[nid].sysdev.kobj, + kobject_name(&node_devices[nid].sysdev.kobj)); } /* mem section does not span the specified node */ return 0; @@ -359,6 +366,8 @@ int unregister_mem_sect_under_nodes(struct memory_block *mem_blk) continue; sysfs_remove_link(&node_devices[nid].sysdev.kobj, kobject_name(&mem_blk->sysdev.kobj)); + sysfs_remove_link(&mem_blk->sysdev.kobj, + kobject_name(&node_devices[nid].sysdev.kobj)); } return 0; } -- cgit v1.2.3 From f8246f3159dfdf97b8b40f9e03e715bafedd22fc Mon Sep 17 00:00:00 2001 From: Alex Chiang Date: Mon, 14 Dec 2009 17:59:06 -0800 Subject: mm: refactor register_cpu_under_node() By returning early if the node is not online, we can unindent the interesting code by one level. No functional change. Signed-off-by: Alex Chiang Cc: Gary Hade Cc: Badari Pulavarty Cc: Ingo Molnar Cc: David Rientjes Cc: Greg KH Cc: Randy Dunlap Cc: David Rientjes Cc: KOSAKI Motohiro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/base/node.c | 20 +++++++++++--------- 1 file changed, 11 insertions(+), 9 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/node.c b/drivers/base/node.c index 44eed11bbdf3..eeae035dadc3 100644 --- a/drivers/base/node.c +++ b/drivers/base/node.c @@ -271,16 +271,18 @@ struct node node_devices[MAX_NUMNODES]; */ int register_cpu_under_node(unsigned int cpu, unsigned int nid) { - if (node_online(nid)) { - struct sys_device *obj = get_cpu_sysdev(cpu); - if (!obj) - return 0; - return sysfs_create_link(&node_devices[nid].sysdev.kobj, - &obj->kobj, - kobject_name(&obj->kobj)); - } + struct sys_device *obj; - return 0; + if (!node_online(nid)) + return 0; + + obj = get_cpu_sysdev(cpu); + if (!obj) + return 0; + + return sysfs_create_link(&node_devices[nid].sysdev.kobj, + &obj->kobj, + kobject_name(&obj->kobj)); } int unregister_cpu_under_node(unsigned int cpu, unsigned int nid) -- cgit v1.2.3 From b9d52dad9447d0db4b52d67d5e9e9d339b5e8302 Mon Sep 17 00:00:00 2001 From: Alex Chiang Date: Mon, 14 Dec 2009 17:59:07 -0800 Subject: mm: refactor unregister_cpu_under_node() By returning early if the node is not online, we can unindent the interesting code by two levels. No functional change. Signed-off-by: Alex Chiang Cc: Gary Hade Cc: Badari Pulavarty Cc: Ingo Molnar Cc: David Rientjes Cc: Greg KH Cc: Randy Dunlap Cc: David Rientjes Cc: KOSAKI Motohiro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/base/node.c | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/node.c b/drivers/base/node.c index eeae035dadc3..9b9acc39a1eb 100644 --- a/drivers/base/node.c +++ b/drivers/base/node.c @@ -287,12 +287,18 @@ int register_cpu_under_node(unsigned int cpu, unsigned int nid) int unregister_cpu_under_node(unsigned int cpu, unsigned int nid) { - if (node_online(nid)) { - struct sys_device *obj = get_cpu_sysdev(cpu); - if (obj) - sysfs_remove_link(&node_devices[nid].sysdev.kobj, - kobject_name(&obj->kobj)); - } + struct sys_device *obj; + + if (!node_online(nid)) + return 0; + + obj = get_cpu_sysdev(cpu); + if (!obj) + return 0; + + sysfs_remove_link(&node_devices[nid].sysdev.kobj, + kobject_name(&obj->kobj)); + return 0; } -- cgit v1.2.3 From 1830794ae6392ce12d36dbcc5ff52f11298ddab6 Mon Sep 17 00:00:00 2001 From: Alex Chiang Date: Mon, 14 Dec 2009 17:59:08 -0800 Subject: mm: add numa node symlink for cpu devices in sysfs You can discover which CPUs belong to a NUMA node by examining /sys/devices/system/node/node#/ However, it's not convenient to go in the other direction, when looking at /sys/devices/system/cpu/cpu#/ Yes, you can muck about in sysfs, but adding these symlinks makes life a lot more convenient. Signed-off-by: Alex Chiang Acked-by: David Rientjes Cc: Gary Hade Cc: Badari Pulavarty Cc: Ingo Molnar Cc: David Rientjes Cc: Greg KH Cc: Randy Dunlap Cc: David Rientjes Cc: KOSAKI Motohiro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/base/node.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) (limited to 'drivers/base') diff --git a/drivers/base/node.c b/drivers/base/node.c index 9b9acc39a1eb..41414113b9f0 100644 --- a/drivers/base/node.c +++ b/drivers/base/node.c @@ -271,6 +271,7 @@ struct node node_devices[MAX_NUMNODES]; */ int register_cpu_under_node(unsigned int cpu, unsigned int nid) { + int ret; struct sys_device *obj; if (!node_online(nid)) @@ -280,9 +281,15 @@ int register_cpu_under_node(unsigned int cpu, unsigned int nid) if (!obj) return 0; - return sysfs_create_link(&node_devices[nid].sysdev.kobj, + ret = sysfs_create_link(&node_devices[nid].sysdev.kobj, &obj->kobj, kobject_name(&obj->kobj)); + if (ret) + return ret; + + return sysfs_create_link(&obj->kobj, + &node_devices[nid].sysdev.kobj, + kobject_name(&node_devices[nid].sysdev.kobj)); } int unregister_cpu_under_node(unsigned int cpu, unsigned int nid) @@ -298,6 +305,8 @@ int unregister_cpu_under_node(unsigned int cpu, unsigned int nid) sysfs_remove_link(&node_devices[nid].sysdev.kobj, kobject_name(&obj->kobj)); + sysfs_remove_link(&obj->kobj, + kobject_name(&node_devices[nid].sysdev.kobj)); return 0; } -- cgit v1.2.3 From 9ae49fab239fb49de92a657c7426271e0793c4e1 Mon Sep 17 00:00:00 2001 From: David Rientjes Date: Mon, 14 Dec 2009 17:59:46 -0800 Subject: mm: slab-allocate memory section nodemask for large systems Nodemasks should not be allocated on the stack for large systems (when it is larger than 256 bytes) since there is a threat of overflow. This patch causes the unregister_mem_sect_under_nodes() nodemask to be allocated on the stack for smaller systems and be allocated by slab for larger systems. GFP_KERNEL is used since remove_memory_block() can block. Cc: Gary Hade Cc: Badari Pulavarty Cc: Alex Chiang Signed-off-by: David Rientjes Cc: Greg Kroah-Hartman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/base/node.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/node.c b/drivers/base/node.c index 41414113b9f0..70122791683d 100644 --- a/drivers/base/node.c +++ b/drivers/base/node.c @@ -363,12 +363,16 @@ int register_mem_sect_under_node(struct memory_block *mem_blk, int nid) /* unregister memory section under all nodes that it spans */ int unregister_mem_sect_under_nodes(struct memory_block *mem_blk) { - nodemask_t unlinked_nodes; + NODEMASK_ALLOC(nodemask_t, unlinked_nodes, GFP_KERNEL); unsigned long pfn, sect_start_pfn, sect_end_pfn; - if (!mem_blk) + if (!mem_blk) { + NODEMASK_FREE(unlinked_nodes); return -EFAULT; - nodes_clear(unlinked_nodes); + } + if (!unlinked_nodes) + return -ENOMEM; + nodes_clear(*unlinked_nodes); sect_start_pfn = section_nr_to_pfn(mem_blk->phys_index); sect_end_pfn = sect_start_pfn + PAGES_PER_SECTION - 1; for (pfn = sect_start_pfn; pfn <= sect_end_pfn; pfn++) { @@ -379,13 +383,14 @@ int unregister_mem_sect_under_nodes(struct memory_block *mem_blk) continue; if (!node_online(nid)) continue; - if (node_test_and_set(nid, unlinked_nodes)) + if (node_test_and_set(nid, *unlinked_nodes)) continue; sysfs_remove_link(&node_devices[nid].sysdev.kobj, kobject_name(&mem_blk->sysdev.kobj)); sysfs_remove_link(&mem_blk->sysdev.kobj, kobject_name(&node_devices[nid].sysdev.kobj)); } + NODEMASK_FREE(unlinked_nodes); return 0; } -- cgit v1.2.3 From 1d531c14d2ed4b24472a4d773f00ed6d1cd34ee7 Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Sun, 13 Dec 2009 20:28:30 +0100 Subject: PM: allow for usage_count > 0 in pm_runtime_get() This patch (as1308c) fixes __pm_runtime_get(). Currently the routine will resume a device if the prior usage count was 0. But this isn't right; thanks to pm_runtime_get_noresume() the usage count can be positive even while the device is suspended. Signed-off-by: Alan Stern Signed-off-by: Rafael J. Wysocki --- drivers/base/power/runtime.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/power/runtime.c b/drivers/base/power/runtime.c index 5a01ecef4af3..40d7720a4b21 100644 --- a/drivers/base/power/runtime.c +++ b/drivers/base/power/runtime.c @@ -701,15 +701,15 @@ EXPORT_SYMBOL_GPL(pm_request_resume); * @dev: Device to handle. * @sync: If set and the device is suspended, resume it synchronously. * - * Increment the usage count of the device and if it was zero previously, - * resume it or submit a resume request for it, depending on the value of @sync. + * Increment the usage count of the device and resume it or submit a resume + * request for it, depending on the value of @sync. */ int __pm_runtime_get(struct device *dev, bool sync) { - int retval = 1; + int retval; - if (atomic_add_return(1, &dev->power.usage_count) == 1) - retval = sync ? pm_runtime_resume(dev) : pm_request_resume(dev); + atomic_inc(&dev->power.usage_count); + retval = sync ? pm_runtime_resume(dev) : pm_request_resume(dev); return retval; } -- cgit v1.2.3 From f2511774863487e61b56a97da07ebf8dd61d7836 Mon Sep 17 00:00:00 2001 From: Arjan van de Ven Date: Sun, 13 Dec 2009 20:29:01 +0100 Subject: PM: Add initcall_debug style timing for suspend/resume In order to diagnose overall suspend/resume times, we need basic instrumentation to break down the total time into per device timing, similar to initcall_debug. This patch adds the basic timing instrumentation, needed for a scritps/bootgraph.pl equivalent or humans. The bootgraph.pl program is still a work in progress, but is far enough along to know that this patch is sufficient. Signed-off-by: Arjan van de Ven Signed-off-by: Rafael J. Wysocki --- drivers/base/power/main.c | 31 +++++++++++++++++++++++++++++++ 1 file changed, 31 insertions(+) (limited to 'drivers/base') diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index 8aa2443182d5..30f0ceebd36c 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -25,6 +25,7 @@ #include #include #include +#include #include "../base.h" #include "power.h" @@ -172,6 +173,13 @@ static int pm_op(struct device *dev, pm_message_t state) { int error = 0; + ktime_t calltime, delta, rettime; + + if (initcall_debug) { + pr_info("calling %s+ @ %i\n", + dev_name(dev), task_pid_nr(current)); + calltime = ktime_get(); + } switch (state.event) { #ifdef CONFIG_SUSPEND @@ -219,6 +227,14 @@ static int pm_op(struct device *dev, default: error = -EINVAL; } + + if (initcall_debug) { + rettime = ktime_get(); + delta = ktime_sub(rettime, calltime); + pr_info("call %s+ returned %d after %Ld usecs\n", dev_name(dev), + error, (unsigned long long)ktime_to_ns(delta) >> 10); + } + return error; } @@ -236,6 +252,13 @@ static int pm_noirq_op(struct device *dev, pm_message_t state) { int error = 0; + ktime_t calltime, delta, rettime; + + if (initcall_debug) { + pr_info("calling %s_i+ @ %i\n", + dev_name(dev), task_pid_nr(current)); + calltime = ktime_get(); + } switch (state.event) { #ifdef CONFIG_SUSPEND @@ -283,6 +306,14 @@ static int pm_noirq_op(struct device *dev, default: error = -EINVAL; } + + if (initcall_debug) { + rettime = ktime_get(); + delta = ktime_sub(rettime, calltime); + printk("initcall %s_i+ returned %d after %Ld usecs\n", dev_name(dev), + error, (unsigned long long)ktime_to_ns(delta) >> 10); + } + return error; } -- cgit v1.2.3 From 33c3374031facf7599c30a1548dfa4c83da87da3 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sun, 13 Dec 2009 20:31:12 +0100 Subject: PM: Remove unnecessary goto from device_resume_noirq() In device_resume_noirq() there is the 'End' label and the associated goto statement that aren't strictly necessary, so rework the code to get rid of them. Also modify device_suspend_noirq() so that it looks completely analogous to device_resume_noirq(). Signed-off-by: Rafael J. Wysocki --- drivers/base/power/main.c | 12 +++--------- 1 file changed, 3 insertions(+), 9 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index 30f0ceebd36c..df04cb4a3611 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -372,14 +372,11 @@ static int device_resume_noirq(struct device *dev, pm_message_t state) TRACE_DEVICE(dev); TRACE_RESUME(0); - if (!dev->bus) - goto End; - - if (dev->bus->pm) { + if (dev->bus && dev->bus->pm) { pm_dev_dbg(dev, state, "EARLY "); error = pm_noirq_op(dev, dev->bus->pm, state); } - End: + TRACE_RESUME(error); return error; } @@ -615,10 +612,7 @@ static int device_suspend_noirq(struct device *dev, pm_message_t state) { int error = 0; - if (!dev->bus) - return 0; - - if (dev->bus->pm) { + if (dev->bus && dev->bus->pm) { pm_dev_dbg(dev, state, "LATE "); error = pm_noirq_op(dev, dev->bus->pm, state); } -- cgit v1.2.3 From d8bed5a4f343d1826153ecf8e7932126c757a21d Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sun, 13 Dec 2009 20:48:54 +0100 Subject: PM: rwsem.h need not be included into main.c It is not necessary to include into drivers/base/power/main.c, so don't do that. Signed-off-by: Rafael J. Wysocki --- drivers/base/power/main.c | 1 - 1 file changed, 1 deletion(-) (limited to 'drivers/base') diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index df04cb4a3611..1a216c114a0f 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -23,7 +23,6 @@ #include #include #include -#include #include #include -- cgit v1.2.3 From facb6011f3993947283fa15d039dacb4ad140230 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Wed, 16 Dec 2009 12:20:00 +0100 Subject: HWPOISON: Add soft page offline support This is a simpler, gentler variant of memory_failure() for soft page offlining controlled from user space. It doesn't kill anything, just tries to invalidate and if that doesn't work migrate the page away. This is useful for predictive failure analysis, where a page has a high rate of corrected errors, but hasn't gone bad yet. Instead it can be offlined early and avoided. The offlining is controlled from sysfs, including a new generic entry point for hard page offlining for symmetry too. We use the page isolate facility to prevent re-allocation race. Normally this is only used by memory hotplug. To avoid races with memory allocation I am using lock_system_sleep(). This avoids the situation where memory hotplug is about to isolate a page range and then hwpoison undoes that work. This is a big hammer currently, but the simplest solution currently. When the page is not free or LRU we try to free pages from slab and other caches. The slab freeing is currently quite dumb and does not try to focus on the specific slab cache which might own the page. This could be potentially improved later. Thanks to Fengguang Wu and Haicheng Li for some fixes. [Added fix from Andrew Morton to adapt to new migrate_pages prototype] Signed-off-by: Andi Kleen --- drivers/base/memory.c | 61 +++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 61 insertions(+) (limited to 'drivers/base') diff --git a/drivers/base/memory.c b/drivers/base/memory.c index 989429cfed88..c4c8f2e1dd15 100644 --- a/drivers/base/memory.c +++ b/drivers/base/memory.c @@ -341,6 +341,64 @@ static inline int memory_probe_init(void) } #endif +#ifdef CONFIG_MEMORY_FAILURE +/* + * Support for offlining pages of memory + */ + +/* Soft offline a page */ +static ssize_t +store_soft_offline_page(struct class *class, const char *buf, size_t count) +{ + int ret; + u64 pfn; + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + if (strict_strtoull(buf, 0, &pfn) < 0) + return -EINVAL; + pfn >>= PAGE_SHIFT; + if (!pfn_valid(pfn)) + return -ENXIO; + ret = soft_offline_page(pfn_to_page(pfn), 0); + return ret == 0 ? count : ret; +} + +/* Forcibly offline a page, including killing processes. */ +static ssize_t +store_hard_offline_page(struct class *class, const char *buf, size_t count) +{ + int ret; + u64 pfn; + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + if (strict_strtoull(buf, 0, &pfn) < 0) + return -EINVAL; + pfn >>= PAGE_SHIFT; + ret = __memory_failure(pfn, 0, 0); + return ret ? ret : count; +} + +static CLASS_ATTR(soft_offline_page, 0644, NULL, store_soft_offline_page); +static CLASS_ATTR(hard_offline_page, 0644, NULL, store_hard_offline_page); + +static __init int memory_fail_init(void) +{ + int err; + + err = sysfs_create_file(&memory_sysdev_class.kset.kobj, + &class_attr_soft_offline_page.attr); + if (!err) + err = sysfs_create_file(&memory_sysdev_class.kset.kobj, + &class_attr_hard_offline_page.attr); + return err; +} +#else +static inline int memory_fail_init(void) +{ + return 0; +} +#endif + /* * Note that phys_device is optional. It is here to allow for * differentiation between which *physical* devices each @@ -471,6 +529,9 @@ int __init memory_dev_init(void) } err = memory_probe_init(); + if (!ret) + ret = err; + err = memory_fail_init(); if (!ret) ret = err; err = block_size_init(); -- cgit v1.2.3