summaryrefslogtreecommitdiff
path: root/fs/zonefs
diff options
context:
space:
mode:
Diffstat (limited to 'fs/zonefs')
-rw-r--r--fs/zonefs/file.c49
-rw-r--r--fs/zonefs/super.c274
-rw-r--r--fs/zonefs/sysfs.c1
3 files changed, 181 insertions, 143 deletions
diff --git a/fs/zonefs/file.c b/fs/zonefs/file.c
index 6ab2318a9c8e..42e2c0065bb3 100644
--- a/fs/zonefs/file.c
+++ b/fs/zonefs/file.c
@@ -125,7 +125,8 @@ static void zonefs_readahead(struct readahead_control *rac)
* which implies that the page range can only be within the fixed inode size.
*/
static int zonefs_write_map_blocks(struct iomap_writepage_ctx *wpc,
- struct inode *inode, loff_t offset)
+ struct inode *inode, loff_t offset,
+ unsigned int len)
{
struct zonefs_zone *z = zonefs_inode_zone(inode);
@@ -298,7 +299,7 @@ static vm_fault_t zonefs_filemap_page_mkwrite(struct vm_fault *vmf)
/* Serialize against truncates */
filemap_invalidate_lock_shared(inode->i_mapping);
- ret = iomap_page_mkwrite(vmf, &zonefs_write_iomap_ops);
+ ret = iomap_page_mkwrite(vmf, &zonefs_write_iomap_ops, NULL);
filemap_invalidate_unlock_shared(inode->i_mapping);
sb_end_pagefault(inode->i_sb);
@@ -348,7 +349,12 @@ static int zonefs_file_write_dio_end_io(struct kiocb *iocb, ssize_t size,
struct zonefs_inode_info *zi = ZONEFS_I(inode);
if (error) {
- zonefs_io_error(inode, true);
+ /*
+ * For Sync IOs, error recovery is called from
+ * zonefs_file_dio_write().
+ */
+ if (!is_sync_kiocb(iocb))
+ zonefs_io_error(inode, true);
return error;
}
@@ -491,6 +497,14 @@ static ssize_t zonefs_file_dio_write(struct kiocb *iocb, struct iov_iter *from)
ret = -EINVAL;
goto inode_unlock;
}
+ /*
+ * Advance the zone write pointer offset. This assumes that the
+ * IO will succeed, which is OK to do because we do not allow
+ * partial writes (IOMAP_DIO_PARTIAL is not set) and if the IO
+ * fails, the error path will correct the write pointer offset.
+ */
+ z->z_wpoffset += count;
+ zonefs_inode_account_active(inode);
mutex_unlock(&zi->i_truncate_mutex);
}
@@ -504,20 +518,19 @@ static ssize_t zonefs_file_dio_write(struct kiocb *iocb, struct iov_iter *from)
if (ret == -ENOTBLK)
ret = -EBUSY;
- if (zonefs_zone_is_seq(z) &&
- (ret > 0 || ret == -EIOCBQUEUED)) {
- if (ret > 0)
- count = ret;
-
- /*
- * Update the zone write pointer offset assuming the write
- * operation succeeded. If it did not, the error recovery path
- * will correct it. Also do active seq file accounting.
- */
- mutex_lock(&zi->i_truncate_mutex);
- z->z_wpoffset += count;
- zonefs_inode_account_active(inode);
- mutex_unlock(&zi->i_truncate_mutex);
+ /*
+ * For a failed IO or partial completion, trigger error recovery
+ * to update the zone write pointer offset to a correct value.
+ * For asynchronous IOs, zonefs_file_write_dio_end_io() may already
+ * have executed error recovery if the IO already completed when we
+ * reach here. However, we cannot know that and execute error recovery
+ * again (that will not change anything).
+ */
+ if (zonefs_zone_is_seq(z)) {
+ if (ret > 0 && ret != count)
+ ret = -EIO;
+ if (ret < 0 && ret != -EIOCBQUEUED)
+ zonefs_io_error(inode, true);
}
inode_unlock:
@@ -550,7 +563,7 @@ static ssize_t zonefs_file_buffered_write(struct kiocb *iocb,
if (ret <= 0)
goto inode_unlock;
- ret = iomap_file_buffered_write(iocb, from, &zonefs_write_iomap_ops);
+ ret = iomap_file_buffered_write(iocb, from, &zonefs_write_iomap_ops, NULL);
if (ret == -EIO)
zonefs_io_error(inode, true);
diff --git a/fs/zonefs/super.c b/fs/zonefs/super.c
index 93971742613a..d165eb979f21 100644
--- a/fs/zonefs/super.c
+++ b/fs/zonefs/super.c
@@ -15,12 +15,13 @@
#include <linux/writeback.h>
#include <linux/quotaops.h>
#include <linux/seq_file.h>
-#include <linux/parser.h>
#include <linux/uio.h>
#include <linux/mman.h>
#include <linux/sched/mm.h>
#include <linux/crc32.h>
#include <linux/task_io_accounting_ops.h>
+#include <linux/fs_parser.h>
+#include <linux/fs_context.h>
#include "zonefs.h"
@@ -113,7 +114,7 @@ static int zonefs_zone_mgmt(struct super_block *sb,
trace_zonefs_zone_mgmt(sb, z, op);
ret = blkdev_zone_mgmt(sb->s_bdev, op, z->z_sector,
- z->z_size >> SECTOR_SHIFT, GFP_NOFS);
+ z->z_size >> SECTOR_SHIFT);
if (ret) {
zonefs_err(sb,
"Zone management operation %s at %llu failed %d\n",
@@ -246,16 +247,18 @@ static void zonefs_inode_update_mode(struct inode *inode)
z->z_mode = inode->i_mode;
}
-struct zonefs_ioerr_data {
- struct inode *inode;
- bool write;
-};
-
static int zonefs_io_error_cb(struct blk_zone *zone, unsigned int idx,
void *data)
{
- struct zonefs_ioerr_data *err = data;
- struct inode *inode = err->inode;
+ struct blk_zone *z = data;
+
+ *z = *zone;
+ return 0;
+}
+
+static void zonefs_handle_io_error(struct inode *inode, struct blk_zone *zone,
+ bool write)
+{
struct zonefs_zone *z = zonefs_inode_zone(inode);
struct super_block *sb = inode->i_sb;
struct zonefs_sb_info *sbi = ZONEFS_SB(sb);
@@ -270,8 +273,8 @@ static int zonefs_io_error_cb(struct blk_zone *zone, unsigned int idx,
data_size = zonefs_check_zone_condition(sb, z, zone);
isize = i_size_read(inode);
if (!(z->z_flags & (ZONEFS_ZONE_READONLY | ZONEFS_ZONE_OFFLINE)) &&
- !err->write && isize == data_size)
- return 0;
+ !write && isize == data_size)
+ return;
/*
* At this point, we detected either a bad zone or an inconsistency
@@ -292,7 +295,7 @@ static int zonefs_io_error_cb(struct blk_zone *zone, unsigned int idx,
* In all cases, warn about inode size inconsistency and handle the
* IO error according to the zone condition and to the mount options.
*/
- if (zonefs_zone_is_seq(z) && isize != data_size)
+ if (isize != data_size)
zonefs_warn(sb,
"inode %lu: invalid size %lld (should be %lld)\n",
inode->i_ino, isize, data_size);
@@ -352,8 +355,6 @@ static int zonefs_io_error_cb(struct blk_zone *zone, unsigned int idx,
zonefs_i_size_write(inode, data_size);
z->z_wpoffset = data_size;
zonefs_inode_account_active(inode);
-
- return 0;
}
/*
@@ -367,23 +368,25 @@ void __zonefs_io_error(struct inode *inode, bool write)
{
struct zonefs_zone *z = zonefs_inode_zone(inode);
struct super_block *sb = inode->i_sb;
- struct zonefs_sb_info *sbi = ZONEFS_SB(sb);
unsigned int noio_flag;
- unsigned int nr_zones = 1;
- struct zonefs_ioerr_data err = {
- .inode = inode,
- .write = write,
- };
+ struct blk_zone zone;
int ret;
/*
- * The only files that have more than one zone are conventional zone
- * files with aggregated conventional zones, for which the inode zone
- * size is always larger than the device zone size.
+ * Conventional zone have no write pointer and cannot become read-only
+ * or offline. So simply fake a report for a single or aggregated zone
+ * and let zonefs_handle_io_error() correct the zone inode information
+ * according to the mount options.
*/
- if (z->z_size > bdev_zone_sectors(sb->s_bdev))
- nr_zones = z->z_size >>
- (sbi->s_zone_sectors_shift + SECTOR_SHIFT);
+ if (!zonefs_zone_is_seq(z)) {
+ zone.start = z->z_sector;
+ zone.len = z->z_size >> SECTOR_SHIFT;
+ zone.wp = zone.start + zone.len;
+ zone.type = BLK_ZONE_TYPE_CONVENTIONAL;
+ zone.cond = BLK_ZONE_COND_NOT_WP;
+ zone.capacity = zone.len;
+ goto handle_io_error;
+ }
/*
* Memory allocations in blkdev_report_zones() can trigger a memory
@@ -394,12 +397,20 @@ void __zonefs_io_error(struct inode *inode, bool write)
* the GFP_NOIO context avoids both problems.
*/
noio_flag = memalloc_noio_save();
- ret = blkdev_report_zones(sb->s_bdev, z->z_sector, nr_zones,
- zonefs_io_error_cb, &err);
- if (ret != nr_zones)
+ ret = blkdev_report_zones(sb->s_bdev, z->z_sector, 1,
+ zonefs_io_error_cb, &zone);
+ memalloc_noio_restore(noio_flag);
+
+ if (ret != 1) {
zonefs_err(sb, "Get inode %lu zone information failed %d\n",
inode->i_ino, ret);
- memalloc_noio_restore(noio_flag);
+ zonefs_warn(sb, "remounting filesystem read-only\n");
+ sb->s_flags |= SB_RDONLY;
+ return;
+ }
+
+handle_io_error:
+ zonefs_handle_io_error(inode, &zone, write);
}
static struct kmem_cache *zonefs_inode_cachep;
@@ -460,58 +471,47 @@ static int zonefs_statfs(struct dentry *dentry, struct kstatfs *buf)
}
enum {
- Opt_errors_ro, Opt_errors_zro, Opt_errors_zol, Opt_errors_repair,
- Opt_explicit_open, Opt_err,
+ Opt_errors, Opt_explicit_open,
};
-static const match_table_t tokens = {
- { Opt_errors_ro, "errors=remount-ro"},
- { Opt_errors_zro, "errors=zone-ro"},
- { Opt_errors_zol, "errors=zone-offline"},
- { Opt_errors_repair, "errors=repair"},
- { Opt_explicit_open, "explicit-open" },
- { Opt_err, NULL}
+struct zonefs_context {
+ unsigned long s_mount_opts;
};
-static int zonefs_parse_options(struct super_block *sb, char *options)
-{
- struct zonefs_sb_info *sbi = ZONEFS_SB(sb);
- substring_t args[MAX_OPT_ARGS];
- char *p;
-
- if (!options)
- return 0;
-
- while ((p = strsep(&options, ",")) != NULL) {
- int token;
+static const struct constant_table zonefs_param_errors[] = {
+ {"remount-ro", ZONEFS_MNTOPT_ERRORS_RO},
+ {"zone-ro", ZONEFS_MNTOPT_ERRORS_ZRO},
+ {"zone-offline", ZONEFS_MNTOPT_ERRORS_ZOL},
+ {"repair", ZONEFS_MNTOPT_ERRORS_REPAIR},
+ {}
+};
- if (!*p)
- continue;
+static const struct fs_parameter_spec zonefs_param_spec[] = {
+ fsparam_enum ("errors", Opt_errors, zonefs_param_errors),
+ fsparam_flag ("explicit-open", Opt_explicit_open),
+ {}
+};
- token = match_token(p, tokens, args);
- switch (token) {
- case Opt_errors_ro:
- sbi->s_mount_opts &= ~ZONEFS_MNTOPT_ERRORS_MASK;
- sbi->s_mount_opts |= ZONEFS_MNTOPT_ERRORS_RO;
- break;
- case Opt_errors_zro:
- sbi->s_mount_opts &= ~ZONEFS_MNTOPT_ERRORS_MASK;
- sbi->s_mount_opts |= ZONEFS_MNTOPT_ERRORS_ZRO;
- break;
- case Opt_errors_zol:
- sbi->s_mount_opts &= ~ZONEFS_MNTOPT_ERRORS_MASK;
- sbi->s_mount_opts |= ZONEFS_MNTOPT_ERRORS_ZOL;
- break;
- case Opt_errors_repair:
- sbi->s_mount_opts &= ~ZONEFS_MNTOPT_ERRORS_MASK;
- sbi->s_mount_opts |= ZONEFS_MNTOPT_ERRORS_REPAIR;
- break;
- case Opt_explicit_open:
- sbi->s_mount_opts |= ZONEFS_MNTOPT_EXPLICIT_OPEN;
- break;
- default:
- return -EINVAL;
- }
+static int zonefs_parse_param(struct fs_context *fc, struct fs_parameter *param)
+{
+ struct zonefs_context *ctx = fc->fs_private;
+ struct fs_parse_result result;
+ int opt;
+
+ opt = fs_parse(fc, zonefs_param_spec, param, &result);
+ if (opt < 0)
+ return opt;
+
+ switch (opt) {
+ case Opt_errors:
+ ctx->s_mount_opts &= ~ZONEFS_MNTOPT_ERRORS_MASK;
+ ctx->s_mount_opts |= result.uint_32;
+ break;
+ case Opt_explicit_open:
+ ctx->s_mount_opts |= ZONEFS_MNTOPT_EXPLICIT_OPEN;
+ break;
+ default:
+ return -EINVAL;
}
return 0;
@@ -533,13 +533,6 @@ static int zonefs_show_options(struct seq_file *seq, struct dentry *root)
return 0;
}
-static int zonefs_remount(struct super_block *sb, int *flags, char *data)
-{
- sync_filesystem(sb);
-
- return zonefs_parse_options(sb, data);
-}
-
static int zonefs_inode_setattr(struct mnt_idmap *idmap,
struct dentry *dentry, struct iattr *iattr)
{
@@ -669,6 +662,7 @@ static struct inode *zonefs_get_file_inode(struct inode *dir,
inode->i_op = &zonefs_file_inode_operations;
inode->i_fop = &zonefs_file_operations;
inode->i_mapping->a_ops = &zonefs_file_aops;
+ mapping_set_large_folios(inode->i_mapping);
/* Update the inode access rights depending on the zone condition */
zonefs_inode_update_mode(inode);
@@ -1055,7 +1049,7 @@ static int zonefs_init_zgroup(struct super_block *sb,
zonefs_info(sb, "Zone group \"%s\" has %u file%s\n",
zonefs_zgroup_name(ztype),
zgroup->g_nr_zones,
- zgroup->g_nr_zones > 1 ? "s" : "");
+ str_plural(zgroup->g_nr_zones));
return 0;
}
@@ -1117,28 +1111,19 @@ static int zonefs_read_super(struct super_block *sb)
struct zonefs_sb_info *sbi = ZONEFS_SB(sb);
struct zonefs_super *super;
u32 crc, stored_crc;
- struct page *page;
- struct bio_vec bio_vec;
- struct bio bio;
int ret;
- page = alloc_page(GFP_KERNEL);
- if (!page)
+ super = kmalloc(PAGE_SIZE, GFP_KERNEL);
+ if (!super)
return -ENOMEM;
- bio_init(&bio, sb->s_bdev, &bio_vec, 1, REQ_OP_READ);
- bio.bi_iter.bi_sector = 0;
- __bio_add_page(&bio, page, PAGE_SIZE, 0);
-
- ret = submit_bio_wait(&bio);
+ ret = bdev_rw_virt(sb->s_bdev, 0, super, PAGE_SIZE, REQ_OP_READ);
if (ret)
- goto free_page;
-
- super = page_address(page);
+ goto free_super;
ret = -EINVAL;
if (le32_to_cpu(super->s_magic) != ZONEFS_MAGIC)
- goto free_page;
+ goto free_super;
stored_crc = le32_to_cpu(super->s_crc);
super->s_crc = 0;
@@ -1146,14 +1131,14 @@ static int zonefs_read_super(struct super_block *sb)
if (crc != stored_crc) {
zonefs_err(sb, "Invalid checksum (Expected 0x%08x, got 0x%08x)",
crc, stored_crc);
- goto free_page;
+ goto free_super;
}
sbi->s_features = le64_to_cpu(super->s_features);
if (sbi->s_features & ~ZONEFS_F_DEFINED_FEATURES) {
zonefs_err(sb, "Unknown features set 0x%llx\n",
sbi->s_features);
- goto free_page;
+ goto free_super;
}
if (sbi->s_features & ZONEFS_F_UID) {
@@ -1161,7 +1146,7 @@ static int zonefs_read_super(struct super_block *sb)
le32_to_cpu(super->s_uid));
if (!uid_valid(sbi->s_uid)) {
zonefs_err(sb, "Invalid UID feature\n");
- goto free_page;
+ goto free_super;
}
}
@@ -1170,7 +1155,7 @@ static int zonefs_read_super(struct super_block *sb)
le32_to_cpu(super->s_gid));
if (!gid_valid(sbi->s_gid)) {
zonefs_err(sb, "Invalid GID feature\n");
- goto free_page;
+ goto free_super;
}
}
@@ -1179,15 +1164,14 @@ static int zonefs_read_super(struct super_block *sb)
if (memchr_inv(super->s_reserved, 0, sizeof(super->s_reserved))) {
zonefs_err(sb, "Reserved area is being used\n");
- goto free_page;
+ goto free_super;
}
import_uuid(&sbi->s_uuid, super->s_uuid);
ret = 0;
-free_page:
- __free_page(page);
-
+free_super:
+ kfree(super);
return ret;
}
@@ -1195,7 +1179,6 @@ static const struct super_operations zonefs_sops = {
.alloc_inode = zonefs_alloc_inode,
.free_inode = zonefs_free_inode,
.statfs = zonefs_statfs,
- .remount_fs = zonefs_remount,
.show_options = zonefs_show_options,
};
@@ -1240,9 +1223,10 @@ static void zonefs_release_zgroup_inodes(struct super_block *sb)
* sub-directories and files according to the device zone configuration and
* format options.
*/
-static int zonefs_fill_super(struct super_block *sb, void *data, int silent)
+static int zonefs_fill_super(struct super_block *sb, struct fs_context *fc)
{
struct zonefs_sb_info *sbi;
+ struct zonefs_context *ctx = fc->fs_private;
struct inode *inode;
enum zonefs_ztype ztype;
int ret;
@@ -1279,7 +1263,7 @@ static int zonefs_fill_super(struct super_block *sb, void *data, int silent)
sbi->s_uid = GLOBAL_ROOT_UID;
sbi->s_gid = GLOBAL_ROOT_GID;
sbi->s_perm = 0640;
- sbi->s_mount_opts = ZONEFS_MNTOPT_ERRORS_RO;
+ sbi->s_mount_opts = ctx->s_mount_opts;
atomic_set(&sbi->s_wro_seq_files, 0);
sbi->s_max_wro_seq_files = bdev_max_open_zones(sb->s_bdev);
@@ -1290,10 +1274,6 @@ static int zonefs_fill_super(struct super_block *sb, void *data, int silent)
if (ret)
return ret;
- ret = zonefs_parse_options(sb, data);
- if (ret)
- return ret;
-
zonefs_info(sb, "Mounting %u zones", bdev_nr_zones(sb->s_bdev));
if (!sbi->s_max_wro_seq_files &&
@@ -1354,12 +1334,6 @@ cleanup:
return ret;
}
-static struct dentry *zonefs_mount(struct file_system_type *fs_type,
- int flags, const char *dev_name, void *data)
-{
- return mount_bdev(fs_type, flags, dev_name, data, zonefs_fill_super);
-}
-
static void zonefs_kill_super(struct super_block *sb)
{
struct zonefs_sb_info *sbi = ZONEFS_SB(sb);
@@ -1374,22 +1348,72 @@ static void zonefs_kill_super(struct super_block *sb)
kfree(sbi);
}
+static void zonefs_free_fc(struct fs_context *fc)
+{
+ struct zonefs_context *ctx = fc->fs_private;
+
+ kfree(ctx);
+}
+
+static int zonefs_get_tree(struct fs_context *fc)
+{
+ return get_tree_bdev(fc, zonefs_fill_super);
+}
+
+static int zonefs_reconfigure(struct fs_context *fc)
+{
+ struct zonefs_context *ctx = fc->fs_private;
+ struct super_block *sb = fc->root->d_sb;
+ struct zonefs_sb_info *sbi = sb->s_fs_info;
+
+ sync_filesystem(fc->root->d_sb);
+ /* Copy new options from ctx into sbi. */
+ sbi->s_mount_opts = ctx->s_mount_opts;
+
+ return 0;
+}
+
+static const struct fs_context_operations zonefs_context_ops = {
+ .parse_param = zonefs_parse_param,
+ .get_tree = zonefs_get_tree,
+ .reconfigure = zonefs_reconfigure,
+ .free = zonefs_free_fc,
+};
+
+/*
+ * Set up the filesystem mount context.
+ */
+static int zonefs_init_fs_context(struct fs_context *fc)
+{
+ struct zonefs_context *ctx;
+
+ ctx = kzalloc(sizeof(struct zonefs_context), GFP_KERNEL);
+ if (!ctx)
+ return -ENOMEM;
+ ctx->s_mount_opts = ZONEFS_MNTOPT_ERRORS_RO;
+ fc->ops = &zonefs_context_ops;
+ fc->fs_private = ctx;
+
+ return 0;
+}
+
/*
* File system definition and registration.
*/
static struct file_system_type zonefs_type = {
- .owner = THIS_MODULE,
- .name = "zonefs",
- .mount = zonefs_mount,
- .kill_sb = zonefs_kill_super,
- .fs_flags = FS_REQUIRES_DEV,
+ .owner = THIS_MODULE,
+ .name = "zonefs",
+ .kill_sb = zonefs_kill_super,
+ .fs_flags = FS_REQUIRES_DEV,
+ .init_fs_context = zonefs_init_fs_context,
+ .parameters = zonefs_param_spec,
};
static int __init zonefs_init_inodecache(void)
{
zonefs_inode_cachep = kmem_cache_create("zonefs_inode_cache",
sizeof(struct zonefs_inode_info), 0,
- (SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD | SLAB_ACCOUNT),
+ SLAB_RECLAIM_ACCOUNT | SLAB_ACCOUNT,
NULL);
if (zonefs_inode_cachep == NULL)
return -ENOMEM;
diff --git a/fs/zonefs/sysfs.c b/fs/zonefs/sysfs.c
index 8ccb65c2b419..ff9a688f1f9c 100644
--- a/fs/zonefs/sysfs.c
+++ b/fs/zonefs/sysfs.c
@@ -92,6 +92,7 @@ int zonefs_sysfs_register(struct super_block *sb)
struct zonefs_sb_info *sbi = ZONEFS_SB(sb);
int ret;
+ super_set_sysfs_name_id(sb);
init_completion(&sbi->s_kobj_unregister);
ret = kobject_init_and_add(&sbi->s_kobj, &zonefs_sb_ktype,
zonefs_sysfs_root, "%s", sb->s_id);
generated by cgit (git 2.34.1) at 2025-07-30 03:57:59 +0000