X-Git-Url: http://ftp.safe.ca/?a=blobdiff_plain;f=mm%2Fshmem.c;h=e2a6ae1a44e9a5e8b5e6594d11f76b11a02ea49f;hb=f910a9dc7c865896815e2a95fe33363e9522f277;hp=ee9024483f6094bac4d6bdac8b798d11e8846fe7;hpb=1b1b32f2c6f6bb32535d2da62075b51c980880eb;p=safe%2Fjmp%2Flinux-2.6 diff --git a/mm/shmem.c b/mm/shmem.c index ee90244..e2a6ae1 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -49,6 +49,7 @@ #include #include #include +#include #include #include @@ -84,6 +85,18 @@ enum sgp_type { SGP_WRITE, /* may exceed i_size, may allocate page */ }; +#ifdef CONFIG_TMPFS +static unsigned long shmem_default_max_blocks(void) +{ + return totalram_pages / 2; +} + +static unsigned long shmem_default_max_inodes(void) +{ + return min(totalram_pages - totalhigh_pages, totalram_pages / 2); +} +#endif + static int shmem_getpage(struct inode *inode, unsigned long idx, struct page **pagep, enum sgp_type sgp, int *type); @@ -188,7 +201,7 @@ static struct vm_operations_struct shmem_vm_ops; static struct backing_dev_info shmem_backing_dev_info __read_mostly = { .ra_pages = 0, /* No readahead */ - .capabilities = BDI_CAP_NO_ACCT_DIRTY | BDI_CAP_NO_WRITEBACK, + .capabilities = BDI_CAP_NO_ACCT_AND_WRITEBACK, .unplug_io_fn = default_unplug_io_fn, }; @@ -231,9 +244,8 @@ static void shmem_free_inode(struct super_block *sb) } } -/* +/** * shmem_recalc_inode - recalculate the size of an inode - * * @inode: inode to recalc * * We have to calculate the free blocks since the mm can drop @@ -257,9 +269,8 @@ static void shmem_recalc_inode(struct inode *inode) } } -/* +/** * shmem_swp_entry - find the swap vector position in the info structure - * * @info: info structure for the inode * @index: index of the page to find * @page: optional page to add to the structure. Has to be preset to @@ -361,13 +372,13 @@ static void shmem_swp_set(struct shmem_inode_info *info, swp_entry_t *entry, uns } } -/* +/** * shmem_swp_alloc - get the position of the swap entry for the page. - * If it does not exist allocate the entry. - * * @info: info structure for the inode * @index: index of the page to find * @sgp: check and recheck i_size? skip allocation? + * + * If the entry does not exist, allocate it. */ static swp_entry_t *shmem_swp_alloc(struct shmem_inode_info *info, unsigned long index, enum sgp_type sgp) { @@ -427,9 +438,8 @@ static swp_entry_t *shmem_swp_alloc(struct shmem_inode_info *info, unsigned long return entry; } -/* +/** * shmem_free_swp - free some swap entries in a directory - * * @dir: pointer to the directory * @edir: pointer after last entry of the directory * @punch_lock: pointer to spinlock when needed for the holepunch case @@ -912,9 +922,13 @@ found: error = 1; if (!inode) goto out; - error = radix_tree_preload(GFP_KERNEL); + /* Precharge page while we can wait, compensate afterwards */ + error = mem_cgroup_cache_charge(page, current->mm, GFP_KERNEL); if (error) goto out; + error = radix_tree_preload(GFP_KERNEL); + if (error) + goto uncharge; error = 1; spin_lock(&info->lock); @@ -947,6 +961,8 @@ found: shmem_swp_unmap(ptr); spin_unlock(&info->lock); radix_tree_preload_end(); +uncharge: + mem_cgroup_uncharge_page(page); out: unlock_page(page); page_cache_release(page); @@ -1062,68 +1078,48 @@ redirty: } #ifdef CONFIG_NUMA -static inline int shmem_parse_mpol(char *value, int *policy, nodemask_t *policy_nodes) +#ifdef CONFIG_TMPFS +static void shmem_show_mpol(struct seq_file *seq, struct mempolicy *mpol) { - char *nodelist = strchr(value, ':'); - int err = 1; + char buffer[64]; - if (nodelist) { - /* NUL-terminate policy string */ - *nodelist++ = '\0'; - if (nodelist_parse(nodelist, *policy_nodes)) - goto out; - if (!nodes_subset(*policy_nodes, node_states[N_HIGH_MEMORY])) - goto out; - } - if (!strcmp(value, "default")) { - *policy = MPOL_DEFAULT; - /* Don't allow a nodelist */ - if (!nodelist) - err = 0; - } else if (!strcmp(value, "prefer")) { - *policy = MPOL_PREFERRED; - /* Insist on a nodelist of one node only */ - if (nodelist) { - char *rest = nodelist; - while (isdigit(*rest)) - rest++; - if (!*rest) - err = 0; - } - } else if (!strcmp(value, "bind")) { - *policy = MPOL_BIND; - /* Insist on a nodelist */ - if (nodelist) - err = 0; - } else if (!strcmp(value, "interleave")) { - *policy = MPOL_INTERLEAVE; - /* - * Default to online nodes with memory if no nodelist - */ - if (!nodelist) - *policy_nodes = node_states[N_HIGH_MEMORY]; - err = 0; + if (!mpol || mpol->mode == MPOL_DEFAULT) + return; /* show nothing */ + + mpol_to_str(buffer, sizeof(buffer), mpol, 1); + + seq_printf(seq, ",mpol=%s", buffer); +} + +static struct mempolicy *shmem_get_sbmpol(struct shmem_sb_info *sbinfo) +{ + struct mempolicy *mpol = NULL; + if (sbinfo->mpol) { + spin_lock(&sbinfo->stat_lock); /* prevent replace/use races */ + mpol = sbinfo->mpol; + mpol_get(mpol); + spin_unlock(&sbinfo->stat_lock); } -out: - /* Restore string for error message */ - if (nodelist) - *--nodelist = ':'; - return err; + return mpol; } +#endif /* CONFIG_TMPFS */ static struct page *shmem_swapin(swp_entry_t entry, gfp_t gfp, struct shmem_inode_info *info, unsigned long idx) { + struct mempolicy mpol, *spol; struct vm_area_struct pvma; struct page *page; + spol = mpol_cond_copy(&mpol, + mpol_shared_policy_lookup(&info->policy, idx)); + /* Create a pseudo vma that just contains the policy */ pvma.vm_start = 0; pvma.vm_pgoff = idx; pvma.vm_ops = NULL; - pvma.vm_policy = mpol_shared_policy_lookup(&info->policy, idx); + pvma.vm_policy = spol; page = swapin_readahead(entry, gfp, &pvma, 0); - mpol_free(pvma.vm_policy); return page; } @@ -1131,23 +1127,24 @@ static struct page *shmem_alloc_page(gfp_t gfp, struct shmem_inode_info *info, unsigned long idx) { struct vm_area_struct pvma; - struct page *page; /* Create a pseudo vma that just contains the policy */ pvma.vm_start = 0; pvma.vm_pgoff = idx; pvma.vm_ops = NULL; pvma.vm_policy = mpol_shared_policy_lookup(&info->policy, idx); - page = alloc_page_vma(gfp, &pvma, 0); - mpol_free(pvma.vm_policy); - return page; + + /* + * alloc_page_vma() will drop the shared policy reference + */ + return alloc_page_vma(gfp, &pvma, 0); } -#else -static inline int shmem_parse_mpol(char *value, int *policy, - nodemask_t *policy_nodes) +#else /* !CONFIG_NUMA */ +#ifdef CONFIG_TMPFS +static inline void shmem_show_mpol(struct seq_file *seq, struct mempolicy *p) { - return 1; } +#endif /* CONFIG_TMPFS */ static inline struct page *shmem_swapin(swp_entry_t entry, gfp_t gfp, struct shmem_inode_info *info, unsigned long idx) @@ -1160,6 +1157,13 @@ static inline struct page *shmem_alloc_page(gfp_t gfp, { return alloc_page(gfp); } +#endif /* CONFIG_NUMA */ + +#if !defined(CONFIG_NUMA) || !defined(CONFIG_TMPFS) +static inline struct mempolicy *shmem_get_sbmpol(struct shmem_sb_info *sbinfo) +{ + return NULL; +} #endif /* @@ -1307,6 +1311,16 @@ repeat: shmem_swp_unmap(entry); spin_unlock(&info->lock); unlock_page(swappage); + if (error == -ENOMEM) { + /* allow reclaim from this memory cgroup */ + error = mem_cgroup_cache_charge(swappage, + current->mm, gfp & ~__GFP_HIGHMEM); + if (error) { + page_cache_release(swappage); + goto failed; + } + mem_cgroup_uncharge_page(swappage); + } page_cache_release(swappage); goto repeat; } @@ -1353,6 +1367,17 @@ repeat: goto failed; } + /* Precharge page while we can wait, compensate after */ + error = mem_cgroup_cache_charge(filepage, current->mm, + gfp & ~__GFP_HIGHMEM); + if (error) { + page_cache_release(filepage); + shmem_unacct_blocks(info->flags, 1); + shmem_free_blocks(inode, 1); + filepage = NULL; + goto failed; + } + spin_lock(&info->lock); entry = shmem_swp_alloc(info, idx, sgp); if (IS_ERR(entry)) @@ -1364,6 +1389,7 @@ repeat: if (error || swap.val || 0 != add_to_page_cache_lru( filepage, mapping, idx, GFP_NOWAIT)) { spin_unlock(&info->lock); + mem_cgroup_uncharge_page(filepage); page_cache_release(filepage); shmem_unacct_blocks(info->flags, 1); shmem_free_blocks(inode, 1); @@ -1372,6 +1398,7 @@ repeat: goto failed; goto repeat; } + mem_cgroup_uncharge_page(filepage); info->flags |= SHMEM_PAGEIN; } @@ -1493,8 +1520,8 @@ shmem_get_inode(struct super_block *sb, int mode, dev_t dev) case S_IFREG: inode->i_op = &shmem_inode_operations; inode->i_fop = &shmem_file_operations; - mpol_shared_policy_init(&info->policy, sbinfo->policy, - &sbinfo->policy_nodes); + mpol_shared_policy_init(&info->policy, + shmem_get_sbmpol(sbinfo)); break; case S_IFDIR: inc_nlink(inode); @@ -1508,8 +1535,7 @@ shmem_get_inode(struct super_block *sb, int mode, dev_t dev) * Must not load anything in the rbtree, * mpol_free_shared_policy will not be called. */ - mpol_shared_policy_init(&info->policy, MPOL_DEFAULT, - NULL); + mpol_shared_policy_init(&info->policy, NULL); break; } } else @@ -1936,7 +1962,7 @@ static const struct inode_operations shmem_symlink_inode_operations = { }; #ifdef CONFIG_TMPFS_POSIX_ACL -/** +/* * Superblocks without xattr inode operations will get security.* xattr * support from the VFS "for free". As soon as we have any other xattrs * like ACLs, we also need to implement the security.* handlers at @@ -1955,8 +1981,7 @@ static int shmem_xattr_security_get(struct inode *inode, const char *name, { if (strcmp(name, "") == 0) return -EINVAL; - return security_inode_getsecurity(inode, name, buffer, size, - -EOPNOTSUPP); + return xattr_getsecurity(inode, name, buffer, size); } static int shmem_xattr_security_set(struct inode *inode, const char *name, @@ -2052,9 +2077,8 @@ static const struct export_operations shmem_export_ops = { .fh_to_dentry = shmem_fh_to_dentry, }; -static int shmem_parse_options(char *options, int *mode, uid_t *uid, - gid_t *gid, unsigned long *blocks, unsigned long *inodes, - int *policy, nodemask_t *policy_nodes) +static int shmem_parse_options(char *options, struct shmem_sb_info *sbinfo, + bool remount) { char *this_char, *value, *rest; @@ -2097,35 +2121,36 @@ static int shmem_parse_options(char *options, int *mode, uid_t *uid, } if (*rest) goto bad_val; - *blocks = DIV_ROUND_UP(size, PAGE_CACHE_SIZE); + sbinfo->max_blocks = + DIV_ROUND_UP(size, PAGE_CACHE_SIZE); } else if (!strcmp(this_char,"nr_blocks")) { - *blocks = memparse(value,&rest); + sbinfo->max_blocks = memparse(value, &rest); if (*rest) goto bad_val; } else if (!strcmp(this_char,"nr_inodes")) { - *inodes = memparse(value,&rest); + sbinfo->max_inodes = memparse(value, &rest); if (*rest) goto bad_val; } else if (!strcmp(this_char,"mode")) { - if (!mode) + if (remount) continue; - *mode = simple_strtoul(value,&rest,8); + sbinfo->mode = simple_strtoul(value, &rest, 8) & 07777; if (*rest) goto bad_val; } else if (!strcmp(this_char,"uid")) { - if (!uid) + if (remount) continue; - *uid = simple_strtoul(value,&rest,0); + sbinfo->uid = simple_strtoul(value, &rest, 0); if (*rest) goto bad_val; } else if (!strcmp(this_char,"gid")) { - if (!gid) + if (remount) continue; - *gid = simple_strtoul(value,&rest,0); + sbinfo->gid = simple_strtoul(value, &rest, 0); if (*rest) goto bad_val; } else if (!strcmp(this_char,"mpol")) { - if (shmem_parse_mpol(value,policy,policy_nodes)) + if (mpol_parse_str(value, &sbinfo->mpol, 1)) goto bad_val; } else { printk(KERN_ERR "tmpfs: Bad mount option %s\n", @@ -2145,24 +2170,20 @@ bad_val: static int shmem_remount_fs(struct super_block *sb, int *flags, char *data) { struct shmem_sb_info *sbinfo = SHMEM_SB(sb); - unsigned long max_blocks = sbinfo->max_blocks; - unsigned long max_inodes = sbinfo->max_inodes; - int policy = sbinfo->policy; - nodemask_t policy_nodes = sbinfo->policy_nodes; + struct shmem_sb_info config = *sbinfo; unsigned long blocks; unsigned long inodes; int error = -EINVAL; - if (shmem_parse_options(data, NULL, NULL, NULL, &max_blocks, - &max_inodes, &policy, &policy_nodes)) + if (shmem_parse_options(data, &config, true)) return error; spin_lock(&sbinfo->stat_lock); blocks = sbinfo->max_blocks - sbinfo->free_blocks; inodes = sbinfo->max_inodes - sbinfo->free_inodes; - if (max_blocks < blocks) + if (config.max_blocks < blocks) goto out; - if (max_inodes < inodes) + if (config.max_inodes < inodes) goto out; /* * Those tests also disallow limited->unlimited while any are in @@ -2170,23 +2191,43 @@ static int shmem_remount_fs(struct super_block *sb, int *flags, char *data) * but we must separately disallow unlimited->limited, because * in that case we have no record of how much is already in use. */ - if (max_blocks && !sbinfo->max_blocks) + if (config.max_blocks && !sbinfo->max_blocks) goto out; - if (max_inodes && !sbinfo->max_inodes) + if (config.max_inodes && !sbinfo->max_inodes) goto out; error = 0; - sbinfo->max_blocks = max_blocks; - sbinfo->free_blocks = max_blocks - blocks; - sbinfo->max_inodes = max_inodes; - sbinfo->free_inodes = max_inodes - inodes; - sbinfo->policy = policy; - sbinfo->policy_nodes = policy_nodes; + sbinfo->max_blocks = config.max_blocks; + sbinfo->free_blocks = config.max_blocks - blocks; + sbinfo->max_inodes = config.max_inodes; + sbinfo->free_inodes = config.max_inodes - inodes; + + mpol_put(sbinfo->mpol); + sbinfo->mpol = config.mpol; /* transfers initial ref */ out: spin_unlock(&sbinfo->stat_lock); return error; } -#endif + +static int shmem_show_options(struct seq_file *seq, struct vfsmount *vfs) +{ + struct shmem_sb_info *sbinfo = SHMEM_SB(vfs->mnt_sb); + + if (sbinfo->max_blocks != shmem_default_max_blocks()) + seq_printf(seq, ",size=%luk", + sbinfo->max_blocks << (PAGE_CACHE_SHIFT - 10)); + if (sbinfo->max_inodes != shmem_default_max_inodes()) + seq_printf(seq, ",nr_inodes=%lu", sbinfo->max_inodes); + if (sbinfo->mode != (S_IRWXUGO | S_ISVTX)) + seq_printf(seq, ",mode=%03o", sbinfo->mode); + if (sbinfo->uid != 0) + seq_printf(seq, ",uid=%u", sbinfo->uid); + if (sbinfo->gid != 0) + seq_printf(seq, ",gid=%u", sbinfo->gid); + shmem_show_mpol(seq, sbinfo->mpol); + return 0; +} +#endif /* CONFIG_TMPFS */ static void shmem_put_super(struct super_block *sb) { @@ -2199,15 +2240,22 @@ static int shmem_fill_super(struct super_block *sb, { struct inode *inode; struct dentry *root; - int mode = S_IRWXUGO | S_ISVTX; - uid_t uid = current->fsuid; - gid_t gid = current->fsgid; - int err = -ENOMEM; struct shmem_sb_info *sbinfo; - unsigned long blocks = 0; - unsigned long inodes = 0; - int policy = MPOL_DEFAULT; - nodemask_t policy_nodes = node_states[N_HIGH_MEMORY]; + int err = -ENOMEM; + + /* Round up to L1_CACHE_BYTES to resist false sharing */ + sbinfo = kmalloc(max((int)sizeof(struct shmem_sb_info), + L1_CACHE_BYTES), GFP_KERNEL); + if (!sbinfo) + return -ENOMEM; + + sbinfo->max_blocks = 0; + sbinfo->max_inodes = 0; + sbinfo->mode = S_IRWXUGO | S_ISVTX; + sbinfo->uid = current->fsuid; + sbinfo->gid = current->fsgid; + sbinfo->mpol = NULL; + sb->s_fs_info = sbinfo; #ifdef CONFIG_TMPFS /* @@ -2216,34 +2264,22 @@ static int shmem_fill_super(struct super_block *sb, * but the internal instance is left unlimited. */ if (!(sb->s_flags & MS_NOUSER)) { - blocks = totalram_pages / 2; - inodes = totalram_pages - totalhigh_pages; - if (inodes > blocks) - inodes = blocks; - if (shmem_parse_options(data, &mode, &uid, &gid, &blocks, - &inodes, &policy, &policy_nodes)) - return -EINVAL; + sbinfo->max_blocks = shmem_default_max_blocks(); + sbinfo->max_inodes = shmem_default_max_inodes(); + if (shmem_parse_options(data, sbinfo, false)) { + err = -EINVAL; + goto failed; + } } sb->s_export_op = &shmem_export_ops; #else sb->s_flags |= MS_NOUSER; #endif - /* Round up to L1_CACHE_BYTES to resist false sharing */ - sbinfo = kmalloc(max((int)sizeof(struct shmem_sb_info), - L1_CACHE_BYTES), GFP_KERNEL); - if (!sbinfo) - return -ENOMEM; - spin_lock_init(&sbinfo->stat_lock); - sbinfo->max_blocks = blocks; - sbinfo->free_blocks = blocks; - sbinfo->max_inodes = inodes; - sbinfo->free_inodes = inodes; - sbinfo->policy = policy; - sbinfo->policy_nodes = policy_nodes; + sbinfo->free_blocks = sbinfo->max_blocks; + sbinfo->free_inodes = sbinfo->max_inodes; - sb->s_fs_info = sbinfo; sb->s_maxbytes = SHMEM_MAX_BYTES; sb->s_blocksize = PAGE_CACHE_SIZE; sb->s_blocksize_bits = PAGE_CACHE_SHIFT; @@ -2255,11 +2291,11 @@ static int shmem_fill_super(struct super_block *sb, sb->s_flags |= MS_POSIXACL; #endif - inode = shmem_get_inode(sb, S_IFDIR | mode, 0); + inode = shmem_get_inode(sb, S_IFDIR | sbinfo->mode, 0); if (!inode) goto failed; - inode->i_uid = uid; - inode->i_gid = gid; + inode->i_uid = sbinfo->uid; + inode->i_gid = sbinfo->gid; root = d_alloc_root(inode); if (!root) goto failed_iput; @@ -2395,6 +2431,7 @@ static const struct super_operations shmem_ops = { #ifdef CONFIG_TMPFS .statfs = shmem_statfs, .remount_fs = shmem_remount_fs, + .show_options = shmem_show_options, #endif .delete_inode = shmem_delete_inode, .drop_inode = generic_delete_inode, @@ -2463,12 +2500,11 @@ out4: } module_init(init_tmpfs) -/* +/** * shmem_file_setup - get an unlinked file living in tmpfs - * * @name: name for dentry (to be seen in /proc//maps * @size: size to be set for the file - * + * @flags: vm_flags */ struct file *shmem_file_setup(char *name, loff_t size, unsigned long flags) { @@ -2523,9 +2559,8 @@ put_memory: return ERR_PTR(error); } -/* +/** * shmem_zero_setup - setup a shared anonymous mapping - * * @vma: the vma to be mmapped is prepared by do_mmap_pgoff */ int shmem_zero_setup(struct vm_area_struct *vma)