diff options
Diffstat (limited to 'kernel/cpuset.c')
| -rw-r--r-- | kernel/cpuset.c | 124 |
1 files changed, 85 insertions, 39 deletions
diff --git a/kernel/cpuset.c b/kernel/cpuset.c index 6313c38c930..6b05dc69c95 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -413,8 +413,8 @@ static struct file_system_type cpuset_fs_type = { * * * When reading/writing to a file: - * - the cpuset to use in file->f_dentry->d_parent->d_fsdata - * - the 'cftype' of the file is file->f_dentry->d_fsdata + * - the cpuset to use in file->f_path.dentry->d_parent->d_fsdata + * - the 'cftype' of the file is file->f_path.dentry->d_fsdata */ struct cftype { @@ -729,9 +729,11 @@ static int validate_change(const struct cpuset *cur, const struct cpuset *trial) } /* Remaining checks don't apply to root cpuset */ - if ((par = cur->parent) == NULL) + if (cur == &top_cpuset) return 0; + par = cur->parent; + /* We must be a subset of our parent cpuset */ if (!is_cpuset_subset(trial, par)) return -EACCES; @@ -1060,10 +1062,7 @@ static int update_flag(cpuset_flagbits_t bit, struct cpuset *cs, char *buf) cpu_exclusive_changed = (is_cpu_exclusive(cs) != is_cpu_exclusive(&trialcs)); mutex_lock(&callback_mutex); - if (turning_on) - set_bit(bit, &cs->flags); - else - clear_bit(bit, &cs->flags); + cs->flags = trialcs.flags; mutex_unlock(&callback_mutex); if (cpu_exclusive_changed) @@ -1281,18 +1280,19 @@ typedef enum { FILE_TASKLIST, } cpuset_filetype_t; -static ssize_t cpuset_common_file_write(struct file *file, const char __user *userbuf, +static ssize_t cpuset_common_file_write(struct file *file, + const char __user *userbuf, size_t nbytes, loff_t *unused_ppos) { - struct cpuset *cs = __d_cs(file->f_dentry->d_parent); - struct cftype *cft = __d_cft(file->f_dentry); + struct cpuset *cs = __d_cs(file->f_path.dentry->d_parent); + struct cftype *cft = __d_cft(file->f_path.dentry); cpuset_filetype_t type = cft->private; char *buffer; char *pathbuf = NULL; int retval = 0; /* Crude upper limit on largest legitimate cpulist user might write. */ - if (nbytes > 100 + 6 * NR_CPUS) + if (nbytes > 100 + 6 * max(NR_CPUS, MAX_NUMNODES)) return -E2BIG; /* +1 for nul-terminator */ @@ -1367,7 +1367,7 @@ static ssize_t cpuset_file_write(struct file *file, const char __user *buf, size_t nbytes, loff_t *ppos) { ssize_t retval = 0; - struct cftype *cft = __d_cft(file->f_dentry); + struct cftype *cft = __d_cft(file->f_path.dentry); if (!cft) return -ENODEV; @@ -1417,8 +1417,8 @@ static int cpuset_sprintf_memlist(char *page, struct cpuset *cs) static ssize_t cpuset_common_file_read(struct file *file, char __user *buf, size_t nbytes, loff_t *ppos) { - struct cftype *cft = __d_cft(file->f_dentry); - struct cpuset *cs = __d_cs(file->f_dentry->d_parent); + struct cftype *cft = __d_cft(file->f_path.dentry); + struct cpuset *cs = __d_cs(file->f_path.dentry->d_parent); cpuset_filetype_t type = cft->private; char *page; ssize_t retval = 0; @@ -1476,7 +1476,7 @@ static ssize_t cpuset_file_read(struct file *file, char __user *buf, size_t nbyt loff_t *ppos) { ssize_t retval = 0; - struct cftype *cft = __d_cft(file->f_dentry); + struct cftype *cft = __d_cft(file->f_path.dentry); if (!cft) return -ENODEV; @@ -1498,7 +1498,7 @@ static int cpuset_file_open(struct inode *inode, struct file *file) if (err) return err; - cft = __d_cft(file->f_dentry); + cft = __d_cft(file->f_path.dentry); if (!cft) return -ENODEV; if (cft->open) @@ -1511,7 +1511,7 @@ static int cpuset_file_open(struct inode *inode, struct file *file) static int cpuset_file_release(struct inode *inode, struct file *file) { - struct cftype *cft = __d_cft(file->f_dentry); + struct cftype *cft = __d_cft(file->f_path.dentry); if (cft->release) return cft->release(inode, file); return 0; @@ -1532,7 +1532,7 @@ static int cpuset_rename(struct inode *old_dir, struct dentry *old_dentry, return simple_rename(old_dir, old_dentry, new_dir, new_dentry); } -static struct file_operations cpuset_file_operations = { +static const struct file_operations cpuset_file_operations = { .read = cpuset_file_read, .write = cpuset_file_write, .llseek = generic_file_llseek, @@ -1700,7 +1700,7 @@ static int pid_array_to_buf(char *buf, int sz, pid_t *a, int npids) */ static int cpuset_tasks_open(struct inode *unused, struct file *file) { - struct cpuset *cs = __d_cs(file->f_dentry->d_parent); + struct cpuset *cs = __d_cs(file->f_path.dentry->d_parent); struct ctr_struct *ctr; pid_t *pidarray; int npids; @@ -2045,7 +2045,6 @@ out: return err; } -#if defined(CONFIG_HOTPLUG_CPU) || defined(CONFIG_MEMORY_HOTPLUG) /* * If common_cpu_mem_hotplug_unplug(), below, unplugs any CPUs * or memory nodes, we need to walk over the cpuset hierarchy, @@ -2109,9 +2108,7 @@ static void common_cpu_mem_hotplug_unplug(void) mutex_unlock(&callback_mutex); mutex_unlock(&manage_mutex); } -#endif -#ifdef CONFIG_HOTPLUG_CPU /* * The top_cpuset tracks what CPUs and Memory Nodes are online, * period. This is necessary in order to make cpusets transparent @@ -2128,7 +2125,6 @@ static int cpuset_handle_cpuhp(struct notifier_block *nb, common_cpu_mem_hotplug_unplug(); return 0; } -#endif #ifdef CONFIG_MEMORY_HOTPLUG /* @@ -2346,32 +2342,48 @@ static const struct cpuset *nearest_exclusive_ancestor(const struct cpuset *cs) } /** - * cpuset_zone_allowed - Can we allocate memory on zone z's memory node? + * cpuset_zone_allowed_softwall - Can we allocate on zone z's memory node? * @z: is this zone on an allowed node? - * @gfp_mask: memory allocation flags (we use __GFP_HARDWALL) + * @gfp_mask: memory allocation flags * - * If we're in interrupt, yes, we can always allocate. If zone + * If we're in interrupt, yes, we can always allocate. If + * __GFP_THISNODE is set, yes, we can always allocate. If zone * z's node is in our tasks mems_allowed, yes. If it's not a * __GFP_HARDWALL request and this zone's nodes is in the nearest * mem_exclusive cpuset ancestor to this tasks cpuset, yes. * Otherwise, no. * + * If __GFP_HARDWALL is set, cpuset_zone_allowed_softwall() + * reduces to cpuset_zone_allowed_hardwall(). Otherwise, + * cpuset_zone_allowed_softwall() might sleep, and might allow a zone + * from an enclosing cpuset. + * + * cpuset_zone_allowed_hardwall() only handles the simpler case of + * hardwall cpusets, and never sleeps. + * + * The __GFP_THISNODE placement logic is really handled elsewhere, + * by forcibly using a zonelist starting at a specified node, and by + * (in get_page_from_freelist()) refusing to consider the zones for + * any node on the zonelist except the first. By the time any such + * calls get to this routine, we should just shut up and say 'yes'. + * * GFP_USER allocations are marked with the __GFP_HARDWALL bit, * and do not allow allocations outside the current tasks cpuset. * GFP_KERNEL allocations are not so marked, so can escape to the - * nearest mem_exclusive ancestor cpuset. + * nearest enclosing mem_exclusive ancestor cpuset. * - * Scanning up parent cpusets requires callback_mutex. The __alloc_pages() - * routine only calls here with __GFP_HARDWALL bit _not_ set if - * it's a GFP_KERNEL allocation, and all nodes in the current tasks - * mems_allowed came up empty on the first pass over the zonelist. - * So only GFP_KERNEL allocations, if all nodes in the cpuset are - * short of memory, might require taking the callback_mutex mutex. + * Scanning up parent cpusets requires callback_mutex. The + * __alloc_pages() routine only calls here with __GFP_HARDWALL bit + * _not_ set if it's a GFP_KERNEL allocation, and all nodes in the + * current tasks mems_allowed came up empty on the first pass over + * the zonelist. So only GFP_KERNEL allocations, if all nodes in the + * cpuset are short of memory, might require taking the callback_mutex + * mutex. * * The first call here from mm/page_alloc:get_page_from_freelist() - * has __GFP_HARDWALL set in gfp_mask, enforcing hardwall cpusets, so - * no allocation on a node outside the cpuset is allowed (unless in - * interrupt, of course). + * has __GFP_HARDWALL set in gfp_mask, enforcing hardwall cpusets, + * so no allocation on a node outside the cpuset is allowed (unless + * in interrupt, of course). * * The second pass through get_page_from_freelist() doesn't even call * here for GFP_ATOMIC calls. For those calls, the __alloc_pages() @@ -2384,12 +2396,12 @@ static const struct cpuset *nearest_exclusive_ancestor(const struct cpuset *cs) * GFP_USER - only nodes in current tasks mems allowed ok. * * Rule: - * Don't call cpuset_zone_allowed() if you can't sleep, unless you + * Don't call cpuset_zone_allowed_softwall if you can't sleep, unless you * pass in the __GFP_HARDWALL flag set in gfp_flag, which disables * the code that might scan up ancestor cpusets and sleep. - **/ + */ -int __cpuset_zone_allowed(struct zone *z, gfp_t gfp_mask) +int __cpuset_zone_allowed_softwall(struct zone *z, gfp_t gfp_mask) { int node; /* node that zone z is on */ const struct cpuset *cs; /* current cpuset ancestors */ @@ -2419,6 +2431,40 @@ int __cpuset_zone_allowed(struct zone *z, gfp_t gfp_mask) return allowed; } +/* + * cpuset_zone_allowed_hardwall - Can we allocate on zone z's memory node? + * @z: is this zone on an allowed node? + * @gfp_mask: memory allocation flags + * + * If we're in interrupt, yes, we can always allocate. + * If __GFP_THISNODE is set, yes, we can always allocate. If zone + * z's node is in our tasks mems_allowed, yes. Otherwise, no. + * + * The __GFP_THISNODE placement logic is really handled elsewhere, + * by forcibly using a zonelist starting at a specified node, and by + * (in get_page_from_freelist()) refusing to consider the zones for + * any node on the zonelist except the first. By the time any such + * calls get to this routine, we should just shut up and say 'yes'. + * + * Unlike the cpuset_zone_allowed_softwall() variant, above, + * this variant requires that the zone be in the current tasks + * mems_allowed or that we're in interrupt. It does not scan up the + * cpuset hierarchy for the nearest enclosing mem_exclusive cpuset. + * It never sleeps. + */ + +int __cpuset_zone_allowed_hardwall(struct zone *z, gfp_t gfp_mask) +{ + int node; /* node that zone z is on */ + + if (in_interrupt() || (gfp_mask & __GFP_THISNODE)) + return 1; + node = zone_to_nid(z); + if (node_isset(node, current->mems_allowed)) + return 1; + return 0; +} + /** * cpuset_lock - lock out any changes to cpuset structures * |