summaryrefslogtreecommitdiffstats
path: root/drivers/md
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/md')
-rw-r--r--drivers/md/Kconfig8
-rw-r--r--drivers/md/Makefile6
-rw-r--r--drivers/md/dm-bio-list.h4
-rw-r--r--drivers/md/dm-crypt.c33
-rw-r--r--drivers/md/dm-delay.c23
-rw-r--r--drivers/md/dm-exception-store.c76
-rw-r--r--drivers/md/dm-io.c5
-rw-r--r--drivers/md/dm-mpath-rdac.c700
-rw-r--r--drivers/md/dm-mpath.c34
-rw-r--r--drivers/md/dm-raid1.c75
-rw-r--r--drivers/md/dm-round-robin.c2
-rw-r--r--drivers/md/dm-snap.c116
-rw-r--r--drivers/md/dm-snap.h6
-rw-r--r--drivers/md/dm.c33
-rw-r--r--drivers/md/dm.h40
-rw-r--r--drivers/md/kcopyd.c11
-rw-r--r--drivers/md/md.c2
-rw-r--r--drivers/md/raid5.c2727
-rw-r--r--drivers/md/xor.c154
19 files changed, 2669 insertions, 1386 deletions
diff --git a/drivers/md/Kconfig b/drivers/md/Kconfig
index 7df934d6913..64bf3a81db9 100644
--- a/drivers/md/Kconfig
+++ b/drivers/md/Kconfig
@@ -109,6 +109,8 @@ config MD_RAID10
config MD_RAID456
tristate "RAID-4/RAID-5/RAID-6 mode"
depends on BLK_DEV_MD
+ select ASYNC_MEMCPY
+ select ASYNC_XOR
---help---
A RAID-5 set of N drives with a capacity of C MB per drive provides
the capacity of C * (N - 1) MB, and protects against a failure
@@ -262,6 +264,12 @@ config DM_MULTIPATH_EMC
---help---
Multipath support for EMC CX/AX series hardware.
+config DM_MULTIPATH_RDAC
+ tristate "LSI/Engenio RDAC multipath support (EXPERIMENTAL)"
+ depends on DM_MULTIPATH && BLK_DEV_DM && EXPERIMENTAL
+ ---help---
+ Multipath support for LSI/Engenio RDAC.
+
config DM_DELAY
tristate "I/O delaying target (EXPERIMENTAL)"
depends on BLK_DEV_DM && EXPERIMENTAL
diff --git a/drivers/md/Makefile b/drivers/md/Makefile
index 38754084eac..c49366cdc05 100644
--- a/drivers/md/Makefile
+++ b/drivers/md/Makefile
@@ -7,6 +7,7 @@ dm-mod-objs := dm.o dm-table.o dm-target.o dm-linear.o dm-stripe.o \
dm-multipath-objs := dm-hw-handler.o dm-path-selector.o dm-mpath.o
dm-snapshot-objs := dm-snap.o dm-exception-store.o
dm-mirror-objs := dm-log.o dm-raid1.o
+dm-rdac-objs := dm-mpath-rdac.o
md-mod-objs := md.o bitmap.o
raid456-objs := raid5.o raid6algos.o raid6recov.o raid6tables.o \
raid6int1.o raid6int2.o raid6int4.o \
@@ -17,7 +18,7 @@ raid456-objs := raid5.o raid6algos.o raid6recov.o raid6tables.o \
hostprogs-y := mktables
# Note: link order is important. All raid personalities
-# and xor.o must come before md.o, as they each initialise
+# and must come before md.o, as they each initialise
# themselves, and md.o may use the personalities when it
# auto-initialised.
@@ -25,7 +26,7 @@ obj-$(CONFIG_MD_LINEAR) += linear.o
obj-$(CONFIG_MD_RAID0) += raid0.o
obj-$(CONFIG_MD_RAID1) += raid1.o
obj-$(CONFIG_MD_RAID10) += raid10.o
-obj-$(CONFIG_MD_RAID456) += raid456.o xor.o
+obj-$(CONFIG_MD_RAID456) += raid456.o
obj-$(CONFIG_MD_MULTIPATH) += multipath.o
obj-$(CONFIG_MD_FAULTY) += faulty.o
obj-$(CONFIG_BLK_DEV_MD) += md-mod.o
@@ -34,6 +35,7 @@ obj-$(CONFIG_DM_CRYPT) += dm-crypt.o
obj-$(CONFIG_DM_DELAY) += dm-delay.o
obj-$(CONFIG_DM_MULTIPATH) += dm-multipath.o dm-round-robin.o
obj-$(CONFIG_DM_MULTIPATH_EMC) += dm-emc.o
+obj-$(CONFIG_DM_MULTIPATH_RDAC) += dm-rdac.o
obj-$(CONFIG_DM_SNAPSHOT) += dm-snapshot.o
obj-$(CONFIG_DM_MIRROR) += dm-mirror.o
obj-$(CONFIG_DM_ZERO) += dm-zero.o
diff --git a/drivers/md/dm-bio-list.h b/drivers/md/dm-bio-list.h
index c6be88826fa..16ee3b018b3 100644
--- a/drivers/md/dm-bio-list.h
+++ b/drivers/md/dm-bio-list.h
@@ -8,7 +8,6 @@
#define DM_BIO_LIST_H
#include <linux/bio.h>
-#include <linux/prefetch.h>
struct bio_list {
struct bio *head;
@@ -31,8 +30,7 @@ static inline void bio_list_init(struct bio_list *bl)
}
#define bio_list_for_each(bio, bl) \
- for (bio = (bl)->head; bio && ({ prefetch(bio->bi_next); 1; }); \
- bio = bio->bi_next)
+ for (bio = (bl)->head; bio; bio = bio->bi_next)
static inline unsigned bio_list_size(const struct bio_list *bl)
{
diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c
index 7b0fcfc9eaa..ba952a03259 100644
--- a/drivers/md/dm-crypt.c
+++ b/drivers/md/dm-crypt.c
@@ -30,7 +30,7 @@
/*
* per bio private data
*/
-struct crypt_io {
+struct dm_crypt_io {
struct dm_target *target;
struct bio *base_bio;
struct work_struct work;
@@ -106,7 +106,7 @@ struct crypt_config {
static struct kmem_cache *_crypt_io_pool;
-static void clone_init(struct crypt_io *, struct bio *);
+static void clone_init(struct dm_crypt_io *, struct bio *);
/*
* Different IV generation algorithms:
@@ -382,7 +382,7 @@ static int crypt_convert(struct crypt_config *cc,
static void dm_crypt_bio_destructor(struct bio *bio)
{
- struct crypt_io *io = bio->bi_private;
+ struct dm_crypt_io *io = bio->bi_private;
struct crypt_config *cc = io->target->private;
bio_free(bio, cc->bs);
@@ -393,7 +393,7 @@ static int crypt_convert(struct crypt_config *cc,
* This should never violate the device limitations
* May return a smaller bio when running out of pages
*/
-static struct bio *crypt_alloc_buffer(struct crypt_io *io, unsigned int size)
+static struct bio *crypt_alloc_buffer(struct dm_crypt_io *io, unsigned size)
{
struct crypt_config *cc = io->target->private;
struct bio *clone;
@@ -479,7 +479,7 @@ static void crypt_free_buffer_pages(struct crypt_config *cc,
* One of the bios was finished. Check for completion of
* the whole request and correctly clean up the buffer.
*/
-static void dec_pending(struct crypt_io *io, int error)
+static void dec_pending(struct dm_crypt_io *io, int error)
{
struct crypt_config *cc = (struct crypt_config *) io->target->private;
@@ -503,7 +503,7 @@ static void dec_pending(struct crypt_io *io, int error)
static struct workqueue_struct *_kcryptd_workqueue;
static void kcryptd_do_work(struct work_struct *work);
-static void kcryptd_queue_io(struct crypt_io *io)
+static void kcryptd_queue_io(struct dm_crypt_io *io)
{
INIT_WORK(&io->work, kcryptd_do_work);
queue_work(_kcryptd_workqueue, &io->work);
@@ -511,7 +511,7 @@ static void kcryptd_queue_io(struct crypt_io *io)
static int crypt_endio(struct bio *clone, unsigned int done, int error)
{
- struct crypt_io *io = clone->bi_private;
+ struct dm_crypt_io *io = clone->bi_private;
struct crypt_config *cc = io->target->private;
unsigned read_io = bio_data_dir(clone) == READ;
@@ -545,7 +545,7 @@ out:
return error;
}
-static void clone_init(struct crypt_io *io, struct bio *clone)
+static void clone_init(struct dm_crypt_io *io, struct bio *clone)
{
struct crypt_config *cc = io->target->private;
@@ -556,7 +556,7 @@ static void clone_init(struct crypt_io *io, struct bio *clone)
clone->bi_destructor = dm_crypt_bio_destructor;
}
-static void process_read(struct crypt_io *io)
+static void process_read(struct dm_crypt_io *io)
{
struct crypt_config *cc = io->target->private;
struct bio *base_bio = io->base_bio;
@@ -587,7 +587,7 @@ static void process_read(struct crypt_io *io)
generic_make_request(clone);
}
-static void process_write(struct crypt_io *io)
+static void process_write(struct dm_crypt_io *io)
{
struct crypt_config *cc = io->target->private;
struct bio *base_bio = io->base_bio;
@@ -644,7 +644,7 @@ static void process_write(struct crypt_io *io)
}
}
-static void process_read_endio(struct crypt_io *io)
+static void process_read_endio(struct dm_crypt_io *io)
{
struct crypt_config *cc = io->target->private;
struct convert_context ctx;
@@ -657,7 +657,7 @@ static void process_read_endio(struct crypt_io *io)
static void kcryptd_do_work(struct work_struct *work)
{
- struct crypt_io *io = container_of(work, struct crypt_io, work);
+ struct dm_crypt_io *io = container_of(work, struct dm_crypt_io, work);
if (io->post_process)
process_read_endio(io);
@@ -939,10 +939,7 @@ static int crypt_map(struct dm_target *ti, struct bio *bio,
union map_info *map_context)
{
struct crypt_config *cc = ti->private;
- struct crypt_io *io;
-
- if (bio_barrier(bio))
- return -EOPNOTSUPP;
+ struct dm_crypt_io *io;
io = mempool_alloc(cc->io_pool, GFP_NOIO);
io->target = ti;
@@ -1062,9 +1059,7 @@ static int __init dm_crypt_init(void)
{
int r;
- _crypt_io_pool = kmem_cache_create("dm-crypt_io",
- sizeof(struct crypt_io),
- 0, 0, NULL, NULL);
+ _crypt_io_pool = KMEM_CACHE(dm_crypt_io, 0);
if (!_crypt_io_pool)
return -ENOMEM;
diff --git a/drivers/md/dm-delay.c b/drivers/md/dm-delay.c
index 52c7cf9e580..6928c136d3c 100644
--- a/drivers/md/dm-delay.c
+++ b/drivers/md/dm-delay.c
@@ -20,7 +20,7 @@
struct delay_c {
struct timer_list delay_timer;
- struct semaphore timer_lock;
+ struct mutex timer_lock;
struct work_struct flush_expired_bios;
struct list_head delayed_bios;
atomic_t may_delay;
@@ -37,7 +37,7 @@ struct delay_c {
unsigned writes;
};
-struct delay_info {
+struct dm_delay_info {
struct delay_c *context;
struct list_head list;
struct bio *bio;
@@ -58,12 +58,12 @@ static void handle_delayed_timer(unsigned long data)
static void queue_timeout(struct delay_c *dc, unsigned long expires)
{
- down(&dc->timer_lock);
+ mutex_lock(&dc->timer_lock);
if (!timer_pending(&dc->delay_timer) || expires < dc->delay_timer.expires)
mod_timer(&dc->delay_timer, expires);
- up(&dc->timer_lock);
+ mutex_unlock(&dc->timer_lock);
}
static void flush_bios(struct bio *bio)
@@ -80,7 +80,7 @@ static void flush_bios(struct bio *bio)
static struct bio *flush_delayed_bios(struct delay_c *dc, int flush_all)
{
- struct delay_info *delayed, *next;
+ struct dm_delay_info *delayed, *next;
unsigned long next_expires = 0;
int start_timer = 0;
BIO_LIST(flush_bios);
@@ -193,13 +193,11 @@ out:
goto bad;
}
- init_timer(&dc->delay_timer);
- dc->delay_timer.function = handle_delayed_timer;
- dc->delay_timer.data = (unsigned long)dc;
+ setup_timer(&dc->delay_timer, handle_delayed_timer, (unsigned long)dc);
INIT_WORK(&dc->flush_expired_bios, flush_expired_bios);
INIT_LIST_HEAD(&dc->delayed_bios);
- init_MUTEX(&dc->timer_lock);
+ mutex_init(&dc->timer_lock);
atomic_set(&dc->may_delay, 1);
ti->private = dc;
@@ -227,7 +225,7 @@ static void delay_dtr(struct dm_target *ti)
static int delay_bio(struct delay_c *dc, int delay, struct bio *bio)
{
- struct delay_info *delayed;
+ struct dm_delay_info *delayed;
unsigned long expires = 0;
if (!delay || !atomic_read(&dc->may_delay))
@@ -338,10 +336,7 @@ static int __init dm_delay_init(void)
goto bad_queue;
}
- delayed_cache = kmem_cache_create("dm-delay",
- sizeof(struct delay_info),
- __alignof__(struct delay_info),
- 0, NULL, NULL);
+ delayed_cache = KMEM_CACHE(dm_delay_info, 0);
if (!delayed_cache) {
DMERR("Couldn't create delayed bio cache.");
goto bad_memcache;
diff --git a/drivers/md/dm-exception-store.c b/drivers/md/dm-exception-store.c
index 07e0a0c84f6..3d65917a1bb 100644
--- a/drivers/md/dm-exception-store.c
+++ b/drivers/md/dm-exception-store.c
@@ -125,9 +125,11 @@ struct pstore {
uint32_t callback_count;
struct commit_callback *callbacks;
struct dm_io_client *io_client;
+
+ struct workqueue_struct *metadata_wq;
};
-static inline unsigned int sectors_to_pages(unsigned int sectors)
+static unsigned sectors_to_pages(unsigned sectors)
{
return sectors / (PAGE_SIZE >> 9);
}
@@ -156,10 +158,24 @@ static void free_area(struct pstore *ps)
ps->area = NULL;
}
+struct mdata_req {
+ struct io_region *where;
+ struct dm_io_request *io_req;
+ struct work_struct work;
+ int result;
+};
+
+static void do_metadata(struct work_struct *work)
+{
+ struct mdata_req *req = container_of(work, struct mdata_req, work);
+
+ req->result = dm_io(req->io_req, 1, req->where, NULL);
+}
+
/*
* Read or write a chunk aligned and sized block of data from a device.
*/
-static int chunk_io(struct pstore *ps, uint32_t chunk, int rw)
+static int chunk_io(struct pstore *ps, uint32_t chunk, int rw, int metadata)
{
struct io_region where = {
.bdev = ps->snap->cow->bdev,
@@ -173,8 +189,23 @@ static int chunk_io(struct pstore *ps, uint32_t chunk, int rw)
.client = ps->io_client,
.notify.fn = NULL,
};
+ struct mdata_req req;
+
+ if (!metadata)
+ return dm_io(&io_req, 1, &where, NULL);
+
+ req.where = &where;
+ req.io_req = &io_req;
- return dm_io(&io_req, 1, &where, NULL);
+ /*
+ * Issue the synchronous I/O from a different thread
+ * to avoid generic_make_request recursion.
+ */
+ INIT_WORK(&req.work, do_metadata);
+ queue_work(ps->metadata_wq, &req.work);
+ flush_workqueue(ps->metadata_wq);
+
+ return req.result;
}
/*
@@ -189,7 +220,7 @@ static int area_io(struct pstore *ps, uint32_t area, int rw)
/* convert a metadata area index to a chunk index */
chunk = 1 + ((ps->exceptions_per_area + 1) * area);
- r = chunk_io(ps, chunk, rw);
+ r = chunk_io(ps, chunk, rw, 0);
if (r)
return r;
@@ -230,7 +261,7 @@ static int read_header(struct pstore *ps, int *new_snapshot)
if (r)
return r;
- r = chunk_io(ps, 0, READ);
+ r = chunk_io(ps, 0, READ, 1);
if (r)
goto bad;
@@ -292,7 +323,7 @@ static int write_header(struct pstore *ps)
dh->version = cpu_to_le32(ps->version);
dh->chunk_size = cpu_to_le32(ps->snap->chunk_size);
- return chunk_io(ps, 0, WRITE);
+ return chunk_io(ps, 0, WRITE, 1);
}
/*
@@ -393,7 +424,7 @@ static int read_exceptions(struct pstore *ps)
return 0;
}
-static inline struct pstore *get_info(struct exception_store *store)
+static struct pstore *get_info(struct exception_store *store)
{
return (struct pstore *) store->context;
}
@@ -409,6 +440,7 @@ static void persistent_destroy(struct exception_store *store)
{
struct pstore *ps = get_info(store);
+ destroy_workqueue(ps->metadata_wq);
dm_io_client_destroy(ps->io_client);
vfree(ps->callbacks);
free_area(ps);
@@ -457,11 +489,6 @@ static int persistent_read_metadata(struct exception_store *store)
/*
* Sanity checks.
*/
- if (!ps->valid) {
- DMWARN("snapshot is marked invalid");
- return -EINVAL;
- }
-
if (ps->version != SNAPSHOT_DISK_VERSION) {
DMWARN("unable to handle snapshot disk version %d",
ps->version);
@@ -469,6 +496,12 @@ static int persistent_read_metadata(struct exception_store *store)
}
/*
+ * Metadata are valid, but snapshot is invalidated
+ */
+ if (!ps->valid)
+ return 1;
+
+ /*
* Read the metadata.
*/
r = read_exceptions(ps);
@@ -480,7 +513,7 @@ static int persistent_read_metadata(struct exception_store *store)
}
static int persistent_prepare(struct exception_store *store,
- struct exception *e)
+ struct dm_snap_exception *e)
{
struct pstore *ps = get_info(store);
uint32_t stride;
@@ -505,7 +538,7 @@ static int persistent_prepare(struct exception_store *store,
}
static void persistent_commit(struct exception_store *store,
- struct exception *e,
+ struct dm_snap_exception *e,
void (*callback) (void *, int success),
void *callback_context)
{
@@ -588,6 +621,12 @@ int dm_create_persistent(struct exception_store *store)
atomic_set(&ps->pending_count, 0);
ps->callbacks = NULL;
+ ps->metadata_wq = create_singlethread_workqueue("ksnaphd");
+ if (!ps->metadata_wq) {
+ DMERR("couldn't start header metadata update thread");
+ return -ENOMEM;
+ }
+
store->destroy = persistent_destroy;
store->read_metadata = persistent_read_metadata;
store->prepare_exception = persistent_prepare;
@@ -616,7 +655,8 @@ static int transient_read_metadata(struct exception_store *store)
return 0;
}
-static int transient_prepare(struct exception_store *store, struct exception *e)
+static int transient_prepare(struct exception_store *store,
+ struct dm_snap_exception *e)
{
struct transient_c *tc = (struct transient_c *) store->context;
sector_t size = get_dev_size(store->snap->cow->bdev);
@@ -631,9 +671,9 @@ static int transient_prepare(struct exception_store *store, struct exception *e)
}
static void transient_commit(struct exception_store *store,
- struct exception *e,
- void (*callback) (void *, int success),
- void *callback_context)
+ struct dm_snap_exception *e,
+ void (*callback) (void *, int success),
+ void *callback_context)
{
/* Just succeed */
callback(callback_context, 1);
diff --git a/drivers/md/dm-io.c b/drivers/md/dm-io.c
index 352c6fbeac5..f3a77248643 100644
--- a/drivers/md/dm-io.c
+++ b/drivers/md/dm-io.c
@@ -293,7 +293,10 @@ static void do_region(int rw, unsigned int region, struct io_region *where,
* bvec for bio_get/set_region() and decrement bi_max_vecs
* to hide it from bio_add_page().
*/
- num_bvecs = (remaining / (PAGE_SIZE >> SECTOR_SHIFT)) + 2;
+ num_bvecs = dm_sector_div_up(remaining,
+ (PAGE_SIZE >> SECTOR_SHIFT));
+ num_bvecs = 1 + min_t(int, bio_get_nr_vecs(where->bdev),
+ num_bvecs);
bio = bio_alloc_bioset(GFP_NOIO, num_bvecs, io->client->bios);
bio->bi_sector = where->sector + (where->count - remaining);
bio->bi_bdev = where->bdev;
diff --git a/drivers/md/dm-mpath-rdac.c b/drivers/md/dm-mpath-rdac.c
new file mode 100644
index 00000000000..8b776b8cb7f
--- /dev/null
+++ b/drivers/md/dm-mpath-rdac.c
@@ -0,0 +1,700 @@
+/*
+ * Engenio/LSI RDAC DM HW handler
+ *
+ * Copyright (C) 2005 Mike Christie. All rights reserved.
+ * Copyright (C) Chandra Seetharaman, IBM Corp. 2007
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ */
+#include <scsi/scsi.h>
+#include <scsi/scsi_cmnd.h>
+#include <scsi/scsi_eh.h>
+
+#define DM_MSG_PREFIX "multipath rdac"
+
+#include "dm.h"
+#include "dm-hw-handler.h"
+
+#define RDAC_DM_HWH_NAME "rdac"
+#define RDAC_DM_HWH_VER "0.4"
+
+/*
+ * LSI mode page stuff
+ *
+ * These struct definitions and the forming of the
+ * mode page were taken from the LSI RDAC 2.4 GPL'd
+ * driver, and then converted to Linux conventions.
+ */
+#define RDAC_QUIESCENCE_TIME 20;
+/*
+ * Page Codes
+ */
+#define RDAC_PAGE_CODE_REDUNDANT_CONTROLLER 0x2c
+
+/*
+ * Controller modes definitions
+ */
+#define RDAC_MODE_TRANSFER_ALL_LUNS 0x01
+#define RDAC_MODE_TRANSFER_SPECIFIED_LUNS 0x02
+
+/*
+ * RDAC Options field
+ */
+#define RDAC_FORCED_QUIESENCE 0x02
+
+#define RDAC_FAILOVER_TIMEOUT (60 * HZ)
+
+struct rdac_mode_6_hdr {
+ u8 data_len;
+ u8 medium_type;
+ u8 device_params;
+ u8 block_desc_len;
+};
+
+struct rdac_mode_10_hdr {
+ u16 data_len;
+ u8 medium_type;
+ u8 device_params;
+ u16 reserved;
+ u16 block_desc_len;
+};
+
+struct rdac_mode_common {
+ u8 controller_serial[16];
+ u8 alt_controller_serial[16];
+ u8 rdac_mode[2];
+ u8 alt_rdac_mode[2];
+ u8 quiescence_timeout;
+ u8 rdac_options;
+};
+
+struct rdac_pg_legacy {
+ struct rdac_mode_6_hdr hdr;
+ u8 page_code;
+ u8 page_len;
+ struct rdac_mode_common common;
+#define MODE6_MAX_LUN 32
+ u8 lun_table[MODE6_MAX_LUN];
+ u8 reserved2[32];
+ u8 reserved3;
+ u8 reserved4;
+};
+
+struct rdac_pg_expanded {
+ struct rdac_mode_10_hdr hdr;
+ u8 page_code;
+ u8 subpage_code;
+ u8 page_len[2];
+ struct rdac_mode_common common;
+ u8 lun_table[256];
+ u8 reserved3;
+ u8 reserved4;
+};
+
+struct c9_inquiry {
+ u8 peripheral_info;
+ u8 page_code; /* 0xC9 */
+ u8 reserved1;
+ u8 page_len;
+ u8 page_id[4]; /* "vace" */
+ u8 avte_cvp;
+ u8 path_prio;
+ u8 reserved2[38];
+};
+
+#define SUBSYS_ID_LEN 16
+#define SLOT_ID_LEN 2
+
+struct c4_inquiry {
+ u8 peripheral_info;
+ u8 page_code; /* 0xC4 */
+ u8 reserved1;
+ u8 page_len;
+ u8 page_id[4]; /* "subs" */
+ u8 subsys_id[SUBSYS_ID_LEN];
+ u8 revision[4];
+ u8 slot_id[SLOT_ID_LEN];
+ u8 reserved[2];
+};
+
+struct rdac_controller {
+ u8 subsys_id[SUBSYS_ID_LEN];
+ u8 slot_id[SLOT_ID_LEN];
+ int use_10_ms;
+ struct kref kref;
+ struct list_head node; /* list of all controllers */
+ spinlock_t lock;
+ int submitted;
+ struct list_head cmd_list; /* list of commands to be submitted */
+ union {
+ struct rdac_pg_legacy legacy;
+ struct rdac_pg_expanded expanded;
+ } mode_select;
+};
+struct c8_inquiry {
+ u8 peripheral_info;
+ u8 page_code; /* 0xC8 */
+ u8 reserved1;
+ u8 page_len;
+ u8 page_id[4]; /* "edid" */
+ u8 reserved2[3];
+ u8 vol_uniq_id_len;
+ u8 vol_uniq_id[16];
+ u8 vol_user_label_len;
+ u8 vol_user_label[60];
+ u8 array_uniq_id_len;
+ u8 array_unique_id[16];
+ u8 array_user_label_len;
+ u8 array_user_label[60];
+ u8 lun[8];
+};
+
+struct c2_inquiry {
+ u8 peripheral_info;
+ u8 page_code; /* 0xC2 */
+ u8 reserved1;
+ u8 page_len;
+ u8 page_id[4]; /* "swr4" */
+ u8 sw_version[3];
+ u8 sw_date[3];
+ u8 features_enabled;
+ u8 max_lun_supported;
+ u8 partitions[239]; /* Total allocation length should be 0xFF */
+};
+
+struct rdac_handler {
+ struct list_head entry; /* list waiting to submit MODE SELECT */
+ unsigned timeout;
+ struct rdac_controller *ctlr;
+#define UNINITIALIZED_LUN (1 << 8)
+ unsigned lun;
+ unsigned char sense[SCSI_SENSE_BUFFERSIZE];
+ struct dm_path *path;
+ struct work_struct work;
+#define SEND_C2_INQUIRY 1
+#define SEND_C4_INQUIRY 2
+#define SEND_C8_INQUIRY 3
+#define SEND_C9_INQUIRY 4
+#define SEND_MODE_SELECT 5
+ int cmd_to_send;
+ union {
+ struct c2_inquiry c2;
+ struct c4_inquiry c4;
+ struct c8_inquiry c8;
+ struct c9_inquiry c9;
+ } inq;
+};
+
+static LIST_HEAD(ctlr_list);
+static DEFINE_SPINLOCK(list_lock);
+static struct workqueue_struct *rdac_wkqd;
+
+static inline int had_failures(struct request *req, int error)
+{
+ return (error || host_byte(req->errors) != DID_OK ||
+ msg_byte(req->errors) != COMMAND_COMPLETE);
+}
+
+static void rdac_resubmit_all(struct rdac_handler *h)
+{
+ struct rdac_controller *ctlr = h->ctlr;
+ struct rdac_handler *tmp, *h1;
+
+ spin_lock(&ctlr->lock);
+ list_for_each_entry_safe(h1, tmp, &ctlr->cmd_list, entry) {
+ h1->cmd_to_send = SEND_C9_INQUIRY;
+ queue_work(rdac_wkqd, &h1->work);
+ list_del(&h1->entry);
+ }
+ ctlr->submitted = 0;
+ spin_unlock(&ctlr->lock);
+}
+
+static void mode_select_endio(struct request *req, int error)
+{
+ struct rdac_handler *h = req->end_io_data;
+ struct scsi_sense_hdr sense_hdr;
+ int sense = 0, fail = 0;
+
+ if (had_failures(req, error)) {
+ fail = 1;
+ goto failed;
+ }
+
+ if (status_byte(req->errors) == CHECK_CONDITION) {
+ scsi_normalize_sense(req->sense, SCSI_SENSE_BUFFERSIZE,
+ &sense_hdr);
+ sense = (sense_hdr.sense_key << 16) | (sense_hdr.asc << 8) |
+ sense_hdr.ascq;
+ /* If it is retryable failure, submit the c9 inquiry again */
+ if (sense == 0x59136 || sense == 0x68b02 || sense == 0xb8b02 ||
+ sense == 0x62900) {
+ /* 0x59136 - Command lock contention
+ * 0x[6b]8b02 - Quiesense in progress or achieved
+ * 0x62900 - Power On, Reset, or Bus Device Reset
+ */
+ h->cmd_to_send = SEND_C9_INQUIRY;
+ queue_work(rdac_wkqd, &h->work);
+ goto done;
+ }
+ if (sense)
+ DMINFO("MODE_SELECT failed on %s with sense 0x%x",
+ h->path->dev->name, sense);
+ }
+failed:
+ if (fail || sense)
+ dm_pg_init_complete(h->path, MP_FAIL_PATH);
+ else
+ dm_pg_init_complete(h->path, 0);
+
+done:
+ rdac_resubmit_all(h);
+ __blk_put_request(req->q, req);
+}
+
+static struct request *get_rdac_req(struct rdac_handler *h,
+ void *buffer, unsigned buflen, int rw)
+{
+ struct request *rq;
+ struct request_queue *q = bdev_get_queue(h->path->dev->bdev);
+
+ rq = blk_get_request(q, rw, GFP_KERNEL);
+
+ if (!rq) {
+ DMINFO("get_rdac_req: blk_get_request failed");
+ return NULL;
+ }
+
+ if (buflen && blk_rq_map_kern(q, rq, buffer, buflen, GFP_KERNEL)) {
+ blk_put_request(rq);
+ DMINFO("get_rdac_req: blk_rq_map_kern failed");
+ return NULL;
+ }
+
+ memset(&rq->cmd, 0, BLK_MAX_CDB);
+ rq->sense = h->sense;
+ memset(rq->sense, 0, SCSI_SENSE_BUFFERSIZE);
+ rq->sense_len = 0;
+
+ rq->end_io_data = h;
+ rq->timeout = h->timeout;
+ rq->cmd_type = REQ_TYPE_BLOCK_PC;
+ rq->cmd_flags = REQ_FAILFAST | REQ_NOMERGE;
+ return rq;
+}
+
+static struct request *rdac_failover_get(struct rdac_handler *h)
+{
+ struct request *rq;
+ struct rdac_mode_common *common;
+ unsigned data_size;
+
+ if (h->ctlr->use_10_ms) {
+ struct rdac_pg_expanded *rdac_pg;
+
+ data_size = sizeof(struct rdac_pg_expanded);
+ rdac_pg = &h->ctlr->mode_select.expanded;
+ memset(rdac_pg, 0, data_size);
+ common = &rdac_pg->common;
+ rdac_pg->page_code = RDAC_PAGE_CODE_REDUNDANT_CONTROLLER + 0x40;
+ rdac_pg->subpage_code = 0x1;
+ rdac_pg->page_len[0] = 0x01;
+ rdac_pg->page_len[1] = 0x28;
+ rdac_pg->lun_table[h->lun] = 0x81;
+ } else {
+ struct rdac_pg_legacy *rdac_pg;
+
+ data_size = sizeof(struct rdac_pg_legacy);
+ rdac_pg = &h->ctlr->mode_select.legacy;
+ memset(rdac_pg, 0, data_size);
+ common = &rdac_pg->common;
+ rdac_pg->page_code = RDAC_PAGE_CODE_REDUNDANT_CONTROLLER;
+ rdac_pg->page_len = 0x68;
+ rdac_pg->lun_table[h->lun] = 0x81;
+ }
+ common->rdac_mode[1] = RDAC_MODE_TRANSFER_SPECIFIED_LUNS;
+ common->quiescence_timeout = RDAC_QUIESCENCE_TIME;
+ common->rdac_options = RDAC_FORCED_QUIESENCE;
+
+ /* get request for block layer packet command */
+ rq = get_rdac_req(h, &h->ctlr->mode_select, data_size, WRITE);
+ if (!rq) {
+ DMERR("rdac_failover_get: no rq");
+ return NULL;
+ }
+
+ /* Prepare the command. */
+ if (h->ctlr->use_10_ms) {
+ rq->cmd[0] = MODE_SELECT_10;
+ rq->cmd[7] = data_size >> 8;
+ rq->cmd[8] = data_size & 0xff;
+ } else {
+ rq->cmd[0] = MODE_SELECT;
+ rq->cmd[4] = data_size;
+ }
+ rq->cmd_len = COMMAND_SIZE(rq->cmd[0]);
+
+ return rq;
+}
+
+/* Acquires h->ctlr->lock */
+static void submit_mode_select(struct rdac_handler *h)
+{
+ struct request *rq;
+ struct request_queue *q = bdev_get_queue(h->path->dev->bdev);
+
+ spin_lock(&h->ctlr->lock);
+ if (h->ctlr->submitted) {
+ list_add(&h->entry, &h->ctlr->cmd_list);
+ goto drop_lock;
+ }
+
+ if (!q) {
+ DMINFO("submit_mode_select: no queue");
+ goto fail_path;
+ }
+
+ rq = rdac_failover_get(h);
+ if (!rq) {
+ DMERR("submit_mode_select: no rq");
+ goto fail_path;
+ }
+
+ DMINFO("queueing MODE_SELECT command on %s", h->path->dev->name);
+
+ blk_execute_rq_nowait(q, NULL, rq, 1, mode_select_endio);
+ h->ctlr->submitted = 1;
+ goto drop_lock;
+fail_path:
+ dm_pg_init_complete(h->path, MP_FAIL_PATH);
+drop_lock:
+ spin_unlock(&h->ctlr->lock);
+}
+
+static void release_ctlr(struct kref *kref)
+{
+ struct rdac_controller *ctlr;
+ ctlr = container_of(kref, struct rdac_controller, kref);
+
+ spin_lock(&list_lock);
+ list_del(&ctlr->node);
+ spin_unlock(&list_lock);
+ kfree(ctlr);
+}
+
+static struct rdac_controller *get_controller(u8 *subsys_id, u8 *slot_id)
+{
+ struct rdac_controller *ctlr, *tmp;
+
+ spin_lock(&list_lock);
+
+ list_for_each_entry(tmp, &ctlr_list, node) {
+ if ((memcmp(tmp->subsys_id, subsys_id, SUBSYS_ID_LEN) == 0) &&
+ (memcmp(tmp->slot_id, slot_id, SLOT_ID_LEN) == 0)) {
+ kref_get(&tmp->kref);
+ spin_unlock(&list_lock);
+ return tmp;
+ }
+ }
+ ctlr = kmalloc(sizeof(*ctlr), GFP_ATOMIC);
+ if (!ctlr)
+ goto done;
+
+ /* initialize fields of controller */
+ memcpy(ctlr->subsys_id, subsys_id, SUBSYS_ID_LEN);
+ memcpy(ctlr->slot_id, slot_id, SLOT_ID_LEN);
+ kref_init(&ctlr->kref);
+ spin_lock_init(&ctlr->lock);
+ ctlr->submitted = 0;
+ ctlr->use_10_ms = -1;
+ INIT_LIST_HEAD(&ctlr->cmd_list);
+ list_add(&ctlr->node, &ctlr_list);
+done:
+ spin_unlock(&list_lock);
+ return ctlr;
+}
+
+static void c4_endio(struct request *req, int error)
+{
+ struct rdac_handler *h = req->end_io_data;
+ struct c4_inquiry *sp;
+
+ if (had_failures(req, error)) {
+ dm_pg_init_complete(h->path, MP_FAIL_PATH);
+ goto done;
+ }
+
+ sp = &h->inq.c4;
+
+ h->ctlr = get_controller(sp->subsys_id, sp->slot_id);
+
+ if (h->ctlr) {
+ h->cmd_to_send = SEND_C9_INQUIRY;
+ queue_work(rdac_wkqd, &h->work);
+ } else
+ dm_pg_init_complete(h->path, MP_FAIL_PATH);
+done:
+ __blk_put_request(req->q, req);
+}
+
+static void c2_endio(struct request *req, int error)
+{
+ struct rdac_handler *h = req->end_io_data;
+ struct c2_inquiry *sp;
+
+ if (had_failures(req, error)) {
+ dm_pg_init_complete(h->path, MP_FAIL_PATH);
+ goto done;
+ }
+
+ sp = &h->inq.c2;
+
+ /* If more than MODE6_MAX_LUN luns are supported, use mode select 10 */
+ if (sp->max_lun_supported >= MODE6_MAX_LUN)
+ h->ctlr->use_10_ms = 1;
+ else
+ h->ctlr->use_10_ms = 0;
+
+ h->cmd_to_send = SEND_MODE_SELECT;
+ queue_work(rdac_wkqd, &h->work);
+done:
+ __blk_put_request(req->q, req);
+}
+
+static void c9_endio(struct request *req, int error)
+{
+ struct rdac_handler *h = req->end_io_data;
+ struct c9_inquiry *sp;
+
+ if (had_failures(req, error)) {
+ dm_pg_init_complete(h->path, MP_FAIL_PATH);
+ goto done;
+ }
+
+ /* We need to look at the sense keys here to take clear action.
+ * For now simple logic: If the host is in AVT mode or if controller
+ * owns the lun, return dm_pg_init_complete(), otherwise submit
+ * MODE SELECT.
+ */
+ sp = &h->inq.c9;
+
+ /* If in AVT mode, return success */
+ if ((sp->avte_cvp >> 7) == 0x1) {
+ dm_pg_init_complete(h->path, 0);
+ goto done;
+ }
+
+ /* If the controller on this path owns the LUN, return success */
+ if (sp->avte_cvp & 0x1) {
+ dm_pg_init_complete(h->path, 0);
+ goto done;
+ }
+
+ if (h->ctlr) {
+ if (h->ctlr->use_10_ms == -1)
+ h->cmd_to_send = SEND_C2_INQUIRY;
+ else
+ h->cmd_to_send = SEND_MODE_SELECT;
+ } else
+ h->cmd_to_send = SEND_C4_INQUIRY;
+ queue_work(rdac_wkqd, &h->work);
+done:
+ __blk_put_request(req->q, req);
+}
+
+static void c8_endio(struct request *req, int error)
+{
+ struct rdac_handler *h = req->end_io_data;
+ struct c8_inquiry *sp;
+
+ if (had_failures(req, error)) {
+ dm_pg_init_complete(h->path, MP_FAIL_PATH);
+ goto done;
+ }
+
+ /* We need to look at the sense keys here to take clear action.
+ * For now simple logic: Get the lun from the inquiry page.
+ */
+ sp = &h->inq.c8;
+ h->lun = sp->lun[7]; /* currently it uses only one byte */
+ h->cmd_to_send = SEND_C9_INQUIRY;
+ queue_work(rdac_wkqd, &h->work);
+done:
+ __blk_put_request(req->q, req);
+}
+
+static void submit_inquiry(struct rdac_handler *h, int page_code,
+ unsigned int len, rq_end_io_fn endio)
+{
+ struct request *rq;
+ struct request_queue *q = bdev_get_queue(h->path->dev->bdev);
+
+ if (!q)
+ goto fail_path;
+
+ rq = get_rdac_req(h, &h->inq, len, READ);
+ if (!rq)
+ goto fail_path;
+
+ /* Prepare the command. */
+ rq->cmd[0] = INQUIRY;
+ rq->cmd[1] = 1;
+ rq->cmd[2] = page_code;
+ rq->cmd[4] = len;
+ rq->cmd_len = COMMAND_SIZE(INQUIRY);
+ blk_execute_rq_nowait(q, NULL, rq, 1, endio);
+ return;
+
+fail_path:
+ dm_pg_init_complete(h->path, MP_FAIL_PATH);
+}
+
+static void service_wkq(struct work_struct *work)
+{
+ struct rdac_handler *h = container_of(work, struct rdac_handler, work);
+
+ switch (h->cmd_to_send) {
+ case SEND_C2_INQUIRY:
+ submit_inquiry(h, 0xC2, sizeof(struct c2_inquiry), c2_endio);
+ break;
+ case SEND_C4_INQUIRY:
+ submit_inquiry(h, 0xC4, sizeof(struct c4_inquiry), c4_endio);
+ break;
+ case SEND_C8_INQUIRY:
+ submit_inquiry(h, 0xC8, sizeof(struct c8_inquiry), c8_endio);
+ break;
+ case SEND_C9_INQUIRY:
+ submit_inquiry(h, 0xC9, sizeof(struct c9_inquiry), c9_endio);
+ break;
+ case SEND_MODE_SELECT:
+ submit_mode_select(h);
+ break;
+ default:
+ BUG();
+ }
+}
+/*
+ * only support subpage2c until we confirm that this is just a matter of
+ * of updating firmware or not, and RDAC (basic AVT works already) for now
+ * but we can add these in in when we get time and testers
+ */
+static int rdac_create(struct hw_handler *hwh, unsigned argc, char **argv)
+{
+ struct rdac_handler *h;
+ unsigned timeout;
+
+ if (argc == 0) {
+ /* No arguments: use defaults */
+ timeout = RDAC_FAILOVER_TIMEOUT;
+ } else if (argc != 1) {
+ DMWARN("incorrect number of arguments");
+ return -EINVAL;
+ } else {
+ if (sscanf(argv[1], "%u", &timeout) != 1) {
+ DMWARN("invalid timeout value");
+ return -EINVAL;
+ }
+ }
+
+ h = kzalloc(sizeof(*h), GFP_KERNEL);
+ if (!h)
+ return -ENOMEM;
+
+ hwh->context = h;
+ h->timeout = timeout;
+ h->lun = UNINITIALIZED_LUN;
+ INIT_WORK(&h->work, service_wkq);
+ DMWARN("using RDAC command with timeout %u", h->timeout);
+
+ return 0;
+}
+
+static void rdac_destroy(struct hw_handler *hwh)
+{
+ struct rdac_handler *h = hwh->context;
+
+ if (h->ctlr)
+ kref_put(&h->ctlr->kref, release_ctlr);
+ kfree(h);
+ hwh->context = NULL;
+}
+
+static unsigned rdac_error(struct hw_handler *hwh, struct bio *bio)
+{
+ /* Try default handler */
+ return dm_scsi_err_handler(hwh, bio);
+}
+
+static void rdac_pg_init(struct hw_handler *hwh, unsigned bypassed,
+ struct dm_path *path)
+{
+ struct rdac_handler *h = hwh->context;
+
+ h->path = path;
+ switch (h->lun) {
+ case UNINITIALIZED_LUN:
+ submit_inquiry(h, 0xC8, sizeof(struct c8_inquiry), c8_endio);
+ break;
+ default:
+ submit_inquiry(h, 0xC9, sizeof(struct c9_inquiry), c9_endio);
+ }
+}
+
+static struct hw_handler_type rdac_handler = {
+ .name = RDAC_DM_HWH_NAME,
+ .module = THIS_MODULE,
+ .create = rdac_create,
+ .destroy = rdac_destroy,
+ .pg_init = rdac_pg_init,
+ .error = rdac_error,
+};
+
+static int __init rdac_init(void)
+{
+ int r = dm_register_hw_handler(&rdac_handler);
+
+ if (r < 0) {
+ DMERR("%s: register failed %d", RDAC_DM_HWH_NAME, r);
+ return r;
+ }
+
+ rdac_wkqd = create_singlethread_workqueue("rdac_wkqd");
+ if (!rdac_wkqd) {
+ DMERR("Failed to create workqueue rdac_wkqd.");
+ dm_unregister_hw_handler(&rdac_handler);
+ return -ENOMEM;
+ }
+
+ DMINFO("%s: version %s loaded", RDAC_DM_HWH_NAME, RDAC_DM_HWH_VER);
+ return 0;
+}
+
+static void __exit rdac_exit(void)
+{
+ int r = dm_unregister_hw_handler(&rdac_handler);
+
+ destroy_workqueue(rdac_wkqd);
+ if (r < 0)
+ DMERR("%s: unregister failed %d", RDAC_DM_HWH_NAME, r);
+}
+
+module_init(rdac_init);
+module_exit(rdac_exit);
+
+MODULE_DESCRIPTION("DM Multipath LSI/Engenio RDAC support");
+MODULE_AUTHOR("Mike Christie, Chandra Seetharaman");
+MODULE_LICENSE("GPL");
+MODULE_VERSION(RDAC_DM_HWH_VER);
diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c
index de54b39e6ff..d6ca9d0a6fd 100644
--- a/drivers/md/dm-mpath.c
+++ b/drivers/md/dm-mpath.c
@@ -83,7 +83,7 @@ struct multipath {
struct work_struct trigger_event;
/*
- * We must use a mempool of mpath_io structs so that we
+ * We must use a mempool of dm_mpath_io structs so that we
* can resubmit bios on error.
*/
mempool_t *mpio_pool;
@@ -92,7 +92,7 @@ struct multipath {
/*
* Context information attached to each bio we process.
*/
-struct mpath_io {
+struct dm_mpath_io {
struct pgpath *pgpath;
struct dm_bio_details details;
};
@@ -122,7 +122,7 @@ static struct pgpath *alloc_pgpath(void)
return pgpath;
}
-static inline void free_pgpath(struct pgpath *pgpath)
+static void free_pgpath(struct pgpath *pgpath)
{
kfree(pgpath);
}
@@ -299,8 +299,8 @@ static int __must_push_back(struct multipath *m)
dm_noflush_suspending(m->ti));
}
-static int map_io(struct multipath *m, struct bio *bio, struct mpath_io *mpio,
- unsigned was_queued)
+static int map_io(struct multipath *m, struct bio *bio,
+ struct dm_mpath_io *mpio, unsigned was_queued)
{
int r = DM_MAPIO_REMAPPED;
unsigned long flags;
@@ -374,7 +374,7 @@ static void dispatch_queued_ios(struct multipath *m)
int r;
unsigned long flags;
struct bio *bio = NULL, *next;
- struct mpath_io *mpio;
+ struct dm_mpath_io *mpio;
union map_info *info;
spin_lock_irqsave(&m->lock, flags);
@@ -795,12 +795,9 @@ static int multipath_map(struct dm_target *ti, struct bio *bio,
union map_info *map_context)
{
int r;
- struct mpath_io *mpio;
+ struct dm_mpath_io *mpio;
struct multipath *m = (struct multipath *) ti->private;
- if (bio_barrier(bio))
- return -EOPNOTSUPP;
-
mpio = mempool_alloc(m->mpio_pool, GFP_NOIO);
dm_bio_record(&mpio->details, bio);
@@ -1014,7 +1011,7 @@ void dm_pg_init_complete(struct dm_path *path, unsigned err_flags)
* end_io handling
*/
static int do_end_io(struct multipath *m, struct bio *bio,
- int error, struct mpath_io *mpio)
+ int error, struct dm_mpath_io *mpio)
{
struct hw_handler *hwh = &m->hw_handler;
unsigned err_flags = MP_FAIL_PATH; /* Default behavior */
@@ -1075,8 +1072,8 @@ static int do_end_io(struct multipath *m, struct bio *bio,
static int multipath_end_io(struct dm_target *ti, struct bio *bio,
int error, union map_info *map_context)
{
- struct multipath *m = (struct multipath *) ti->private;
- struct mpath_io *mpio = (struct mpath_io *) map_context->ptr;
+ struct multipath *m = ti->private;
+ struct dm_mpath_io *mpio = map_context->ptr;
struct pgpath *pgpath = mpio->pgpath;
struct path_selector *ps;
int r;
@@ -1346,22 +1343,20 @@ static int __init dm_multipath_init(void)
int r;
/* allocate a slab for the dm_ios */
- _mpio_cache = kmem_cache_create("dm_mpath", sizeof(struct mpath_io),
- 0, 0, NULL, NULL);
+ _mpio_cache = KMEM_CACHE(dm_mpath_io, 0);
if (!_mpio_cache)
return -ENOMEM;
r = dm_register_target(&multipath_target);
if (r < 0) {
- DMERR("%s: register failed %d", multipath_target.name, r);
+ DMERR("register failed %d", r);
kmem_cache_destroy(_mpio_cache);
return -EINVAL;
}
kmultipathd = create_workqueue("kmpathd");
if (!kmultipathd) {
- DMERR("%s: failed to create workqueue kmpathd",
- multipath_target.name);
+ DMERR("failed to create workqueue kmpathd");
dm_unregister_target(&multipath_target);
kmem_cache_destroy(_mpio_cache);
return -ENOMEM;
@@ -1382,8 +1377,7 @@ static void __exit dm_multipath_exit(void)
r = dm_unregister_target(&multipath_target);
if (r < 0)
- DMERR("%s: target unregister failed %d",
- multipath_target.name, r);
+ DMERR("target unregister failed %d", r);
kmem_cache_destroy(_mpio_cache);
}
diff --git a/drivers/md/dm-raid1.c b/drivers/md/dm-raid1.c
index ef124b71ccc..1a876f9965e 100644
--- a/drivers/md/dm-raid1.c
+++ b/drivers/md/dm-raid1.c
@@ -24,6 +24,7 @@
#define DM_IO_PAGES 64
#define DM_RAID1_HANDLE_ERRORS 0x01
+#define errors_handled(p) ((p)->features & DM_RAID1_HANDLE_ERRORS)
static DECLARE_WAIT_QUEUE_HEAD(_kmirrord_recovery_stopped);
@@ -85,6 +86,7 @@ struct region_hash {
struct list_head clean_regions;
struct list_head quiesced_regions;
struct list_head recovered_regions;
+ struct list_head failed_recovered_regions;
};
enum {
@@ -132,6 +134,7 @@ struct mirror_set {
/* recovery */
region_t nr_regions;
int in_sync;
+ int log_failure;
struct mirror *default_mirror; /* Default mirror */
@@ -204,6 +207,7 @@ static int rh_init(struct region_hash *rh, struct mirror_set *ms,
INIT_LIST_HEAD(&rh->clean_regions);
INIT_LIST_HEAD(&rh->quiesced_regions);
INIT_LIST_HEAD(&rh->recovered_regions);
+ INIT_LIST_HEAD(&rh->failed_recovered_regions);
rh->region_pool = mempool_create_kmalloc_pool(MIN_REGIONS,
sizeof(struct region));
@@ -368,6 +372,7 @@ static void rh_update_states(struct region_hash *rh)
LIST_HEAD(clean);
LIST_HEAD(recovered);
+ LIST_HEAD(failed_recovered);
/*
* Quickly grab the lists.
@@ -378,10 +383,8 @@ static void rh_update_states(struct region_hash *rh)
list_splice(&rh->clean_regions, &clean);
INIT_LIST_HEAD(&rh->clean_regions);
- list_for_each_entry (reg, &clean, list) {
- rh->log->type->clear_region(rh->log, reg->key);
+ list_for_each_entry(reg, &clean, list)
list_del(&reg->hash_list);
- }
}
if (!list_empty(&rh->recovered_regions)) {
@@ -391,6 +394,15 @@ static void rh_update_states(struct region_hash *rh)
list_for_each_entry (reg, &recovered, list)
list_del(&reg->hash_list);
}
+
+ if (!list_empty(&rh->failed_recovered_regions)) {
+ list_splice(&rh->failed_recovered_regions, &failed_recovered);
+ INIT_LIST_HEAD(&rh->failed_recovered_regions);
+
+ list_for_each_entry(reg, &failed_recovered, list)
+ list_del(&reg->hash_list);
+ }
+
spin_unlock(&rh->region_lock);
write_unlock_irq(&rh->hash_lock);
@@ -405,10 +417,17 @@ static void rh_update_states(struct region_hash *rh)
mempool_free(reg, rh->region_pool);
}
- rh->log->type->flush(rh->log);
+ list_for_each_entry_safe(reg, next, &failed_recovered, list) {
+ complete_resync_work(reg, errors_handled(rh->ms) ? 0 : 1);
+ mempool_free(reg, rh->region_pool);
+ }
- list_for_each_entry_safe (reg, next, &clean, list)
+ list_for_each_entry_safe(reg, next, &clean, list) {
+ rh->log->type->clear_region(rh->log, reg->key);
mempool_free(reg, rh->region_pool);
+ }
+
+ rh->log->type->flush(rh->log);
}
static void rh_inc(struct region_hash *rh, region_t region)
@@ -555,21 +574,25 @@ static struct region *rh_recovery_start(struct region_hash *rh)
return reg;
}
-/* FIXME: success ignored for now */
static void rh_recovery_end(struct region *reg, int success)
{
struct region_hash *rh = reg->rh;
spin_lock_irq(&rh->region_lock);
- list_add(&reg->list, &reg->rh->recovered_regions);
+ if (success)
+ list_add(&reg->list, &reg->rh->recovered_regions);
+ else {
+ reg->state = RH_NOSYNC;
+ list_add(&reg->list, &reg->rh->failed_recovered_regions);
+ }
spin_unlock_irq(&rh->region_lock);
wake(rh->ms);
}
-static void rh_flush(struct region_hash *rh)
+static int rh_flush(struct region_hash *rh)
{
- rh->log->type->flush(rh->log);
+ return rh->log->type->flush(rh->log);
}
static void rh_delay(struct region_hash *rh, struct bio *bio)
@@ -633,7 +656,14 @@ static void recovery_complete(int read_err, unsigned int write_err,
{
struct region *reg = (struct region *) context;
- /* FIXME: better error handling */
+ if (read_err)
+ /* Read error means the failure of default mirror. */
+ DMERR_LIMIT("Unable to read primary mirror during recovery");
+
+ if (write_err)
+ DMERR_LIMIT("Write error during recovery (error = 0x%x)",
+ write_err);
+
rh_recovery_end(reg, !(read_err || write_err));
}
@@ -863,12 +893,15 @@ static void do_writes(struct mirror_set *ms, struct bio_list *writes)
*/
rh_inc_pending(&ms->rh, &sync);
rh_inc_pending(&ms->rh, &nosync);
- rh_flush(&ms->rh);
+ ms->log_failure = rh_flush(&ms->rh) ? 1 : 0;
/*
* Dispatch io.
*/
- while ((bio = bio_list_pop(&sync)))
+ if (unlikely(ms->log_failure))
+ while ((bio = bio_list_pop(&sync)))
+ bio_endio(bio, bio->bi_size, -EIO);
+ else while ((bio = bio_list_pop(&sync)))
do_write(ms, bio);
while ((bio = bio_list_pop(&recover)))
@@ -1145,6 +1178,15 @@ static int mirror_ctr(struct dm_target *ti, unsigned int argc, char **argv)
argv += args_used;
argc -= args_used;
+ /*
+ * Any read-balancing addition depends on the
+ * DM_RAID1_HANDLE_ERRORS flag being present.
+ * This is because the decision to balance depends
+ * on the sync state of a region. If the above
+ * flag is not present, we ignore errors; and
+ * the sync state may be inaccurate.
+ */
+
if (argc) {
ti->error = "Too many mirror arguments";
free_context(ms, ti, ms->nr_mirrors);
@@ -1288,12 +1330,12 @@ static int mirror_status(struct dm_target *ti, status_type_t type,
for (m = 0; m < ms->nr_mirrors; m++)
DMEMIT("%s ", ms->mirror[m].dev->name);
- DMEMIT("%llu/%llu",
+ DMEMIT("%llu/%llu 0 ",
(unsigned long long)ms->rh.log->type->
get_sync_count(ms->rh.log),
(unsigned long long)ms->nr_regions);
- sz = ms->rh.log->type->status(ms->rh.log, type, result, maxlen);
+ sz += ms->rh.log->type->status(ms->rh.log, type, result+sz, maxlen-sz);
break;
@@ -1335,8 +1377,7 @@ static int __init dm_mirror_init(void)
r = dm_register_target(&mirror_target);
if (r < 0) {
- DMERR("%s: Failed to register mirror target",
- mirror_target.name);
+ DMERR("Failed to register mirror target");
dm_dirty_log_exit();
}
@@ -1349,7 +1390,7 @@ static void __exit dm_mirror_exit(void)
r = dm_unregister_target(&mirror_target);
if (r < 0)
- DMERR("%s: unregister failed %d", mirror_target.name, r);
+ DMERR("unregister failed %d", r);
dm_dirty_log_exit();
}
diff --git a/drivers/md/dm-round-robin.c b/drivers/md/dm-round-robin.c
index a348a97b65a..391dfa2ad43 100644
--- a/drivers/md/dm-round-robin.c
+++ b/drivers/md/dm-round-robin.c
@@ -205,7 +205,7 @@ static void __exit dm_rr_exit(void)
int r = dm_unregister_path_selector(&rr_ps);
if (r < 0)
- DMERR("round-robin: unregister failed %d", r);
+ DMERR("unregister failed %d", r);
}
module_init(dm_rr_init);
diff --git a/drivers/md/dm-snap.c b/drivers/md/dm-snap.c
index 0821a2b68a7..83ddbfe6b8a 100644
--- a/drivers/md/dm-snap.c
+++ b/drivers/md/dm-snap.c
@@ -42,8 +42,8 @@
static struct workqueue_struct *ksnapd;
static void flush_queued_bios(struct work_struct *work);
-struct pending_exception {
- struct exception e;
+struct dm_snap_pending_exception {
+ struct dm_snap_exception e;
/*
* Origin buffers waiting for this to complete are held
@@ -63,7 +63,7 @@ struct pending_exception {
* group of pending_exceptions. It is always last to get freed.
* These fields get set up when writing to the origin.
*/
- struct pending_exception *primary_pe;
+ struct dm_snap_pending_exception *primary_pe;
/*
* Number of pending_exceptions processing this chunk.
@@ -137,7 +137,7 @@ static void exit_origin_hash(void)
kfree(_origins);
}
-static inline unsigned int origin_hash(struct block_device *bdev)
+static unsigned origin_hash(struct block_device *bdev)
{
return bdev->bd_dev & ORIGIN_MASK;
}
@@ -231,7 +231,7 @@ static int init_exception_table(struct exception_table *et, uint32_t size)
static void exit_exception_table(struct exception_table *et, struct kmem_cache *mem)
{
struct list_head *slot;
- struct exception *ex, *next;
+ struct dm_snap_exception *ex, *next;
int i, size;
size = et->hash_mask + 1;
@@ -245,18 +245,19 @@ static void exit_exception_table(struct exception_table *et, struct kmem_cache *
vfree(et->table);
}
-static inline uint32_t exception_hash(struct exception_table *et, chunk_t chunk)
+static uint32_t exception_hash(struct exception_table *et, chunk_t chunk)
{
return chunk & et->hash_mask;
}
-static void insert_exception(struct exception_table *eh, struct exception *e)
+static void insert_exception(struct exception_table *eh,
+ struct dm_snap_exception *e)
{
struct list_head *l = &eh->table[exception_hash(eh, e->old_chunk)];
list_add(&e->hash_list, l);
}
-static inline void remove_exception(struct exception *e)
+static void remove_exception(struct dm_snap_exception *e)
{
list_del(&e->hash_list);
}
@@ -265,11 +266,11 @@ static inline void remove_exception(struct exception *e)
* Return the exception data for a sector, or NULL if not
* remapped.
*/
-static struct exception *lookup_exception(struct exception_table *et,
- chunk_t chunk)
+static struct dm_snap_exception *lookup_exception(struct exception_table *et,
+ chunk_t chunk)
{
struct list_head *slot;
- struct exception *e;
+ struct dm_snap_exception *e;
slot = &et->table[exception_hash(et, chunk)];
list_for_each_entry (e, slot, hash_list)
@@ -279,9 +280,9 @@ static struct exception *lookup_exception(struct exception_table *et,
return NULL;
}
-static inline struct exception *alloc_exception(void)
+static struct dm_snap_exception *alloc_exception(void)
{
- struct exception *e;
+ struct dm_snap_exception *e;
e = kmem_cache_alloc(exception_cache, GFP_NOIO);
if (!e)
@@ -290,24 +291,24 @@ static inline struct exception *alloc_exception(void)
return e;
}
-static inline void free_exception(struct exception *e)
+static void free_exception(struct dm_snap_exception *e)
{
kmem_cache_free(exception_cache, e);
}
-static inline struct pending_exception *alloc_pending_exception(void)
+static struct dm_snap_pending_exception *alloc_pending_exception(void)
{
return mempool_alloc(pending_pool, GFP_NOIO);
}
-static inline void free_pending_exception(struct pending_exception *pe)
+static void free_pending_exception(struct dm_snap_pending_exception *pe)
{
mempool_free(pe, pending_pool);
}
int dm_add_exception(struct dm_snapshot *s, chunk_t old, chunk_t new)
{
- struct exception *e;
+ struct dm_snap_exception *e;
e = alloc_exception();
if (!e)
@@ -334,7 +335,7 @@ static int calc_max_buckets(void)
/*
* Rounds a number down to a power of 2.
*/
-static inline uint32_t round_down(uint32_t n)
+static uint32_t round_down(uint32_t n)
{
while (n & (n - 1))
n &= (n - 1);
@@ -384,7 +385,7 @@ static int init_hash_tables(struct dm_snapshot *s)
* Round a number up to the nearest 'size' boundary. size must
* be a power of 2.
*/
-static inline ulong round_up(ulong n, ulong size)
+static ulong round_up(ulong n, ulong size)
{
size--;
return (n + size) & ~size;
@@ -522,9 +523,12 @@ static int snapshot_ctr(struct dm_target *ti, unsigned int argc, char **argv)
/* Metadata must only be loaded into one table at once */
r = s->store.read_metadata(&s->store);
- if (r) {
+ if (r < 0) {
ti->error = "Failed to read snapshot metadata";
goto bad6;
+ } else if (r > 0) {
+ s->valid = 0;
+ DMWARN("Snapshot is marked invalid.");
}
bio_list_init(&s->queued_bios);
@@ -577,7 +581,7 @@ static void __free_exceptions(struct dm_snapshot *s)
static void snapshot_dtr(struct dm_target *ti)
{
- struct dm_snapshot *s = (struct dm_snapshot *) ti->private;
+ struct dm_snapshot *s = ti->private;
flush_workqueue(ksnapd);
@@ -655,14 +659,14 @@ static void __invalidate_snapshot(struct dm_snapshot *s, int err)
dm_table_event(s->table);
}
-static void get_pending_exception(struct pending_exception *pe)
+static void get_pending_exception(struct dm_snap_pending_exception *pe)
{
atomic_inc(&pe->ref_count);
}
-static struct bio *put_pending_exception(struct pending_exception *pe)
+static struct bio *put_pending_exception(struct dm_snap_pending_exception *pe)
{
- struct pending_exception *primary_pe;
+ struct dm_snap_pending_exception *primary_pe;
struct bio *origin_bios = NULL;
primary_pe = pe->primary_pe;
@@ -692,9 +696,9 @@ static struct bio *put_pending_exception(struct pending_exception *pe)
return origin_bios;
}
-static void pending_complete(struct pending_exception *pe, int success)
+static void pending_complete(struct dm_snap_pending_exception *pe, int success)
{
- struct exception *e;
+ struct dm_snap_exception *e;
struct dm_snapshot *s = pe->snap;
struct bio *origin_bios = NULL;
struct bio *snapshot_bios = NULL;
@@ -748,7 +752,8 @@ static void pending_complete(struct pending_exception *pe, int success)
static void commit_callback(void *context, int success)
{
- struct pending_exception *pe = (struct pending_exception *) context;
+ struct dm_snap_pending_exception *pe = context;
+
pending_complete(pe, success);
}
@@ -758,7 +763,7 @@ static void commit_callback(void *context, int success)
*/
static void copy_callback(int read_err, unsigned int write_err, void *context)
{
- struct pending_exception *pe = (struct pending_exception *) context;
+ struct dm_snap_pending_exception *pe = context;
struct dm_snapshot *s = pe->snap;
if (read_err || write_err)
@@ -773,7 +778,7 @@ static void copy_callback(int read_err, unsigned int write_err, void *context)
/*
* Dispatches the copy operation to kcopyd.
*/
-static void start_copy(struct pending_exception *pe)
+static void start_copy(struct dm_snap_pending_exception *pe)
{
struct dm_snapshot *s = pe->snap;
struct io_region src, dest;
@@ -803,11 +808,11 @@ static void start_copy(struct pending_exception *pe)
* NOTE: a write lock must be held on snap->lock before calling
* this.
*/
-static struct pending_exception *
+static struct dm_snap_pending_exception *
__find_pending_exception(struct dm_snapshot *s, struct bio *bio)
{
- struct exception *e;
- struct pending_exception *pe;
+ struct dm_snap_exception *e;
+ struct dm_snap_pending_exception *pe;
chunk_t chunk = sector_to_chunk(s, bio->bi_sector);
/*
@@ -816,7 +821,7 @@ __find_pending_exception(struct dm_snapshot *s, struct bio *bio)
e = lookup_exception(&s->pending, chunk);
if (e) {
/* cast the exception to a pending exception */
- pe = container_of(e, struct pending_exception, e);
+ pe = container_of(e, struct dm_snap_pending_exception, e);
goto out;
}
@@ -836,7 +841,7 @@ __find_pending_exception(struct dm_snapshot *s, struct bio *bio)
e = lookup_exception(&s->pending, chunk);
if (e) {
free_pending_exception(pe);
- pe = container_of(e, struct pending_exception, e);
+ pe = container_of(e, struct dm_snap_pending_exception, e);
goto out;
}
@@ -860,8 +865,8 @@ __find_pending_exception(struct dm_snapshot *s, struct bio *bio)
return pe;
}
-static inline void remap_exception(struct dm_snapshot *s, struct exception *e,
- struct bio *bio)
+static void remap_exception(struct dm_snapshot *s, struct dm_snap_exception *e,
+ struct bio *bio)
{
bio->bi_bdev = s->cow->bdev;
bio->bi_sector = chunk_to_sector(s, e->new_chunk) +
@@ -871,11 +876,11 @@ static inline void remap_exception(struct dm_snapshot *s, struct exception *e,
static int snapshot_map(struct dm_target *ti, struct bio *bio,
union map_info *map_context)
{
- struct exception *e;
- struct dm_snapshot *s = (struct dm_snapshot *) ti->private;
+ struct dm_snap_exception *e;
+ struct dm_snapshot *s = ti->private;
int r = DM_MAPIO_REMAPPED;
chunk_t chunk;
- struct pending_exception *pe = NULL;
+ struct dm_snap_pending_exception *pe = NULL;
chunk = sector_to_chunk(s, bio->bi_sector);
@@ -884,9 +889,6 @@ static int snapshot_map(struct dm_target *ti, struct bio *bio,
if (!s->valid)
return -EIO;
- if (unlikely(bio_barrier(bio)))
- return -EOPNOTSUPP;
-
/* FIXME: should only take write lock if we need
* to copy an exception */
down_write(&s->lock);
@@ -945,7 +947,7 @@ static int snapshot_map(struct dm_target *ti, struct bio *bio,
static void snapshot_resume(struct dm_target *ti)
{
- struct dm_snapshot *s = (struct dm_snapshot *) ti->private;
+ struct dm_snapshot *s = ti->private;
down_write(&s->lock);
s->active = 1;
@@ -955,7 +957,7 @@ static void snapshot_resume(struct dm_target *ti)
static int snapshot_status(struct dm_target *ti, status_type_t type,
char *result, unsigned int maxlen)
{
- struct dm_snapshot *snap = (struct dm_snapshot *) ti->private;
+ struct dm_snapshot *snap = ti->private;
switch (type) {
case STATUSTYPE_INFO:
@@ -999,8 +1001,8 @@ static int __origin_write(struct list_head *snapshots, struct bio *bio)
{
int r = DM_MAPIO_REMAPPED, first = 0;
struct dm_snapshot *snap;
- struct exception *e;
- struct pending_exception *pe, *next_pe, *primary_pe = NULL;
+ struct dm_snap_exception *e;
+ struct dm_snap_pending_exception *pe, *next_pe, *primary_pe = NULL;
chunk_t chunk;
LIST_HEAD(pe_queue);
@@ -1147,19 +1149,16 @@ static int origin_ctr(struct dm_target *ti, unsigned int argc, char **argv)
static void origin_dtr(struct dm_target *ti)
{
- struct dm_dev *dev = (struct dm_dev *) ti->private;
+ struct dm_dev *dev = ti->private;
dm_put_device(ti, dev);
}
static int origin_map(struct dm_target *ti, struct bio *bio,
union map_info *map_context)
{
- struct dm_dev *dev = (struct dm_dev *) ti->private;
+ struct dm_dev *dev = ti->private;
bio->bi_bdev = dev->bdev;
- if (unlikely(bio_barrier(bio)))
- return -EOPNOTSUPP;
-
/* Only tell snapshots if this is a write */
return (bio_rw(bio) == WRITE) ? do_origin(dev, bio) : DM_MAPIO_REMAPPED;
}
@@ -1172,7 +1171,7 @@ static int origin_map(struct dm_target *ti, struct bio *bio,
*/
static void origin_resume(struct dm_target *ti)
{
- struct dm_dev *dev = (struct dm_dev *) ti->private;
+ struct dm_dev *dev = ti->private;
struct dm_snapshot *snap;
struct origin *o;
chunk_t chunk_size = 0;
@@ -1190,7 +1189,7 @@ static void origin_resume(struct dm_target *ti)
static int origin_status(struct dm_target *ti, status_type_t type, char *result,
unsigned int maxlen)
{
- struct dm_dev *dev = (struct dm_dev *) ti->private;
+ struct dm_dev *dev = ti->private;
switch (type) {
case STATUSTYPE_INFO:
@@ -1249,21 +1248,14 @@ static int __init dm_snapshot_init(void)
goto bad2;
}
- exception_cache = kmem_cache_create("dm-snapshot-ex",
- sizeof(struct exception),
- __alignof__(struct exception),
- 0, NULL, NULL);
+ exception_cache = KMEM_CACHE(dm_snap_exception, 0);
if (!exception_cache) {
DMERR("Couldn't create exception cache.");
r = -ENOMEM;
goto bad3;
}
- pending_cache =
- kmem_cache_create("dm-snapshot-in",
- sizeof(struct pending_exception),
- __alignof__(struct pending_exception),
- 0, NULL, NULL);
+ pending_cache = KMEM_CACHE(dm_snap_pending_exception, 0);
if (!pending_cache) {
DMERR("Couldn't create pending cache.");
r = -ENOMEM;
diff --git a/drivers/md/dm-snap.h b/drivers/md/dm-snap.h
index 15fa2ae6cdc..650e0f1f51d 100644
--- a/drivers/md/dm-snap.h
+++ b/drivers/md/dm-snap.h
@@ -30,7 +30,7 @@ typedef sector_t chunk_t;
* An exception is used where an old chunk of data has been
* replaced by a new one.
*/
-struct exception {
+struct dm_snap_exception {
struct list_head hash_list;
chunk_t old_chunk;
@@ -58,13 +58,13 @@ struct exception_store {
* Find somewhere to store the next exception.
*/
int (*prepare_exception) (struct exception_store *store,
- struct exception *e);
+ struct dm_snap_exception *e);
/*
* Update the metadata with this exception.
*/
void (*commit_exception) (struct exception_store *store,
- struct exception *e,
+ struct dm_snap_exception *e,
void (*callback) (void *, int success),
void *callback_context);
diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index 2717a355dc5..f4f7d35561a 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -45,7 +45,7 @@ struct dm_io {
* One of these is allocated per target within a bio. Hopefully
* this will be simplified out one day.
*/
-struct target_io {
+struct dm_target_io {
struct dm_io *io;
struct dm_target *ti;
union map_info info;
@@ -54,7 +54,7 @@ struct target_io {
union map_info *dm_get_mapinfo(struct bio *bio)
{
if (bio && bio->bi_private)
- return &((struct target_io *)bio->bi_private)->info;
+ return &((struct dm_target_io *)bio->bi_private)->info;
return NULL;
}
@@ -132,14 +132,12 @@ static int __init local_init(void)
int r;
/* allocate a slab for the dm_ios */
- _io_cache = kmem_cache_create("dm_io",
- sizeof(struct dm_io), 0, 0, NULL, NULL);
+ _io_cache = KMEM_CACHE(dm_io, 0);
if (!_io_cache)
return -ENOMEM;
/* allocate a slab for the target ios */
- _tio_cache = kmem_cache_create("dm_tio", sizeof(struct target_io),
- 0, 0, NULL, NULL);
+ _tio_cache = KMEM_CACHE(dm_target_io, 0);
if (!_tio_cache) {
kmem_cache_destroy(_io_cache);
return -ENOMEM;
@@ -325,22 +323,22 @@ out:
return r;
}
-static inline struct dm_io *alloc_io(struct mapped_device *md)
+static struct dm_io *alloc_io(struct mapped_device *md)
{
return mempool_alloc(md->io_pool, GFP_NOIO);
}
-static inline void free_io(struct mapped_device *md, struct dm_io *io)
+static void free_io(struct mapped_device *md, struct dm_io *io)
{
mempool_free(io, md->io_pool);
}
-static inline struct target_io *alloc_tio(struct mapped_device *md)
+static struct dm_target_io *alloc_tio(struct mapped_device *md)
{
return mempool_alloc(md->tio_pool, GFP_NOIO);
}
-static inline void free_tio(struct mapped_device *md, struct target_io *tio)
+static void free_tio(struct mapped_device *md, struct dm_target_io *tio)
{
mempool_free(tio, md->tio_pool);
}
@@ -498,7 +496,7 @@ static void dec_pending(struct dm_io *io, int error)
static int clone_endio(struct bio *bio, unsigned int done, int error)
{
int r = 0;
- struct target_io *tio = bio->bi_private;
+ struct dm_target_io *tio = bio->bi_private;
struct mapped_device *md = tio->io->md;
dm_endio_fn endio = tio->ti->type->end_io;
@@ -558,7 +556,7 @@ static sector_t max_io_len(struct mapped_device *md,
}
static void __map_bio(struct dm_target *ti, struct bio *clone,
- struct target_io *tio)
+ struct dm_target_io *tio)
{
int r;
sector_t sector;
@@ -672,7 +670,7 @@ static void __clone_and_map(struct clone_info *ci)
struct bio *clone, *bio = ci->bio;
struct dm_target *ti = dm_table_find_target(ci->map, ci->sector);
sector_t len = 0, max = max_io_len(ci->md, ci->sector, ti);
- struct target_io *tio;
+ struct dm_target_io *tio;
/*
* Allocate a target io object.
@@ -802,6 +800,15 @@ static int dm_request(request_queue_t *q, struct bio *bio)
int rw = bio_data_dir(bio);
struct mapped_device *md = q->queuedata;
+ /*
+ * There is no use in forwarding any barrier request since we can't
+ * guarantee it is (or can be) handled by the targets correctly.
+ */
+ if (unlikely(bio_barrier(bio))) {
+ bio_endio(bio, bio->bi_size, -EOPNOTSUPP);
+ return 0;
+ }
+
down_read(&md->io_lock);
disk_stat_inc(dm_disk(md), ios[rw]);
diff --git a/drivers/md/dm.h b/drivers/md/dm.h
index 2f796b1436b..462ee652a89 100644
--- a/drivers/md/dm.h
+++ b/drivers/md/dm.h
@@ -18,13 +18,45 @@
#define DM_NAME "device-mapper"
-#define DMERR(f, arg...) printk(KERN_ERR DM_NAME ": " DM_MSG_PREFIX ": " f "\n", ## arg)
-#define DMWARN(f, arg...) printk(KERN_WARNING DM_NAME ": " DM_MSG_PREFIX ": " f "\n", ## arg)
-#define DMINFO(f, arg...) printk(KERN_INFO DM_NAME ": " DM_MSG_PREFIX ": " f "\n", ## arg)
+#define DMERR(f, arg...) \
+ printk(KERN_ERR DM_NAME ": " DM_MSG_PREFIX ": " f "\n", ## arg)
+#define DMERR_LIMIT(f, arg...) \
+ do { \
+ if (printk_ratelimit()) \
+ printk(KERN_ERR DM_NAME ": " DM_MSG_PREFIX ": " \
+ f "\n", ## arg); \
+ } while (0)
+
+#define DMWARN(f, arg...) \
+ printk(KERN_WARNING DM_NAME ": " DM_MSG_PREFIX ": " f "\n", ## arg)
+#define DMWARN_LIMIT(f, arg...) \
+ do { \
+ if (printk_ratelimit()) \
+ printk(KERN_WARNING DM_NAME ": " DM_MSG_PREFIX ": " \
+ f "\n", ## arg); \
+ } while (0)
+
+#define DMINFO(f, arg...) \
+ printk(KERN_INFO DM_NAME ": " DM_MSG_PREFIX ": " f "\n", ## arg)
+#define DMINFO_LIMIT(f, arg...) \
+ do { \
+ if (printk_ratelimit()) \
+ printk(KERN_INFO DM_NAME ": " DM_MSG_PREFIX ": " f \
+ "\n", ## arg); \
+ } while (0)
+
#ifdef CONFIG_DM_DEBUG
-# define DMDEBUG(f, arg...) printk(KERN_DEBUG DM_NAME ": " DM_MSG_PREFIX " DEBUG: " f "\n", ## arg)
+# define DMDEBUG(f, arg...) \
+ printk(KERN_DEBUG DM_NAME ": " DM_MSG_PREFIX " DEBUG: " f "\n", ## arg)
+# define DMDEBUG_LIMIT(f, arg...) \
+ do { \
+ if (printk_ratelimit()) \
+ printk(KERN_DEBUG DM_NAME ": " DM_MSG_PREFIX ": " f \
+ "\n", ## arg); \
+ } while (0)
#else
# define DMDEBUG(f, arg...) do {} while (0)
+# define DMDEBUG_LIMIT(f, arg...) do {} while (0)
#endif
#define DMEMIT(x...) sz += ((sz >= maxlen) ? \
diff --git a/drivers/md/kcopyd.c b/drivers/md/kcopyd.c
index dbc234e3c69..7e052378c47 100644
--- a/drivers/md/kcopyd.c
+++ b/drivers/md/kcopyd.c
@@ -29,7 +29,7 @@
static struct workqueue_struct *_kcopyd_wq;
static struct work_struct _kcopyd_work;
-static inline void wake(void)
+static void wake(void)
{
queue_work(_kcopyd_wq, &_kcopyd_work);
}
@@ -226,10 +226,7 @@ static LIST_HEAD(_pages_jobs);
static int jobs_init(void)
{
- _job_cache = kmem_cache_create("kcopyd-jobs",
- sizeof(struct kcopyd_job),
- __alignof__(struct kcopyd_job),
- 0, NULL, NULL);
+ _job_cache = KMEM_CACHE(kcopyd_job, 0);
if (!_job_cache)
return -ENOMEM;
@@ -258,7 +255,7 @@ static void jobs_exit(void)
* Functions to push and pop a job onto the head of a given job
* list.
*/
-static inline struct kcopyd_job *pop(struct list_head *jobs)
+static struct kcopyd_job *pop(struct list_head *jobs)
{
struct kcopyd_job *job = NULL;
unsigned long flags;
@@ -274,7 +271,7 @@ static inline struct kcopyd_job *pop(struct list_head *jobs)
return job;
}
-static inline void push(struct list_head *jobs, struct kcopyd_job *job)
+static void push(struct list_head *jobs, struct kcopyd_job *job)
{
unsigned long flags;
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 1c54f3c1cca..33beaa7da08 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -5814,7 +5814,7 @@ static __exit void md_exit(void)
}
}
-module_init(md_init)
+subsys_initcall(md_init);
module_exit(md_exit)
static int get_ro(char *buffer, struct kernel_param *kp)
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index 061375ee659..0b66afef2d8 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -52,6 +52,7 @@
#include "raid6.h"
#include <linux/raid/bitmap.h>
+#include <linux/async_tx.h>
/*
* Stripe cache
@@ -80,7 +81,6 @@
/*
* The following can be used to debug the driver
*/
-#define RAID5_DEBUG 0
#define RAID5_PARANOIA 1
#if RAID5_PARANOIA && defined(CONFIG_SMP)
# define CHECK_DEVLOCK() assert_spin_locked(&conf->device_lock)
@@ -88,8 +88,7 @@
# define CHECK_DEVLOCK()
#endif
-#define PRINTK(x...) ((void)(RAID5_DEBUG && printk(x)))
-#if RAID5_DEBUG
+#ifdef DEBUG
#define inline
#define __inline__
#endif
@@ -104,6 +103,23 @@ static inline int raid6_next_disk(int disk, int raid_disks)
disk++;
return (disk < raid_disks) ? disk : 0;
}
+
+static void return_io(struct bio *return_bi)
+{
+ struct bio *bi = return_bi;
+ while (bi) {
+ int bytes = bi->bi_size;
+
+ return_bi = bi->bi_next;
+ bi->bi_next = NULL;
+ bi->bi_size = 0;
+ bi->bi_end_io(bi, bytes,
+ test_bit(BIO_UPTODATE, &bi->bi_flags)
+ ? 0 : -EIO);
+ bi = return_bi;
+ }
+}
+
static void print_raid5_conf (raid5_conf_t *conf);
static void __release_stripe(raid5_conf_t *conf, struct stripe_head *sh)
@@ -125,6 +141,7 @@ static void __release_stripe(raid5_conf_t *conf, struct stripe_head *sh)
}
md_wakeup_thread(conf->mddev->thread);
} else {
+ BUG_ON(sh->ops.pending);
if (test_and_clear_bit(STRIPE_PREREAD_ACTIVE, &sh->state)) {
atomic_dec(&conf->preread_active_stripes);
if (atomic_read(&conf->preread_active_stripes) < IO_THRESHOLD)
@@ -152,7 +169,8 @@ static void release_stripe(struct stripe_head *sh)
static inline void remove_hash(struct stripe_head *sh)
{
- PRINTK("remove_hash(), stripe %llu\n", (unsigned long long)sh->sector);
+ pr_debug("remove_hash(), stripe %llu\n",
+ (unsigned long long)sh->sector);
hlist_del_init(&sh->hash);
}
@@ -161,7 +179,8 @@ static inline void insert_hash(raid5_conf_t *conf, struct stripe_head *sh)
{
struct hlist_head *hp = stripe_hash(conf, sh->sector);
- PRINTK("insert_hash(), stripe %llu\n", (unsigned long long)sh->sector);
+ pr_debug("insert_hash(), stripe %llu\n",
+ (unsigned long long)sh->sector);
CHECK_DEVLOCK();
hlist_add_head(&sh->hash, hp);
@@ -224,9 +243,10 @@ static void init_stripe(struct stripe_head *sh, sector_t sector, int pd_idx, int
BUG_ON(atomic_read(&sh->count) != 0);
BUG_ON(test_bit(STRIPE_HANDLE, &sh->state));
-
+ BUG_ON(sh->ops.pending || sh->ops.ack || sh->ops.complete);
+
CHECK_DEVLOCK();
- PRINTK("init_stripe called, stripe %llu\n",
+ pr_debug("init_stripe called, stripe %llu\n",
(unsigned long long)sh->sector);
remove_hash(sh);
@@ -240,11 +260,11 @@ static void init_stripe(struct stripe_head *sh, sector_t sector, int pd_idx, int
for (i = sh->disks; i--; ) {
struct r5dev *dev = &sh->dev[i];
- if (dev->toread || dev->towrite || dev->written ||
+ if (dev->toread || dev->read || dev->towrite || dev->written ||
test_bit(R5_LOCKED, &dev->flags)) {
- printk("sector=%llx i=%d %p %p %p %d\n",
+ printk(KERN_ERR "sector=%llx i=%d %p %p %p %p %d\n",
(unsigned long long)sh->sector, i, dev->toread,
- dev->towrite, dev->written,
+ dev->read, dev->towrite, dev->written,
test_bit(R5_LOCKED, &dev->flags));
BUG();
}
@@ -260,11 +280,11 @@ static struct stripe_head *__find_stripe(raid5_conf_t *conf, sector_t sector, in
struct hlist_node *hn;
CHECK_DEVLOCK();
- PRINTK("__find_stripe, sector %llu\n", (unsigned long long)sector);
+ pr_debug("__find_stripe, sector %llu\n", (unsigned long long)sector);
hlist_for_each_entry(sh, hn, stripe_hash(conf, sector), hash)
if (sh->sector == sector && sh->disks == disks)
return sh;
- PRINTK("__stripe %llu not in cache\n", (unsigned long long)sector);
+ pr_debug("__stripe %llu not in cache\n", (unsigned long long)sector);
return NULL;
}
@@ -276,7 +296,7 @@ static struct stripe_head *get_active_stripe(raid5_conf_t *conf, sector_t sector
{
struct stripe_head *sh;
- PRINTK("get_stripe, sector %llu\n", (unsigned long long)sector);
+ pr_debug("get_stripe, sector %llu\n", (unsigned long long)sector);
spin_lock_irq(&conf->device_lock);
@@ -324,6 +344,579 @@ static struct stripe_head *get_active_stripe(raid5_conf_t *conf, sector_t sector
return sh;
}
+/* test_and_ack_op() ensures that we only dequeue an operation once */
+#define test_and_ack_op(op, pend) \
+do { \
+ if (test_bit(op, &sh->ops.pending) && \
+ !test_bit(op, &sh->ops.complete)) { \
+ if (test_and_set_bit(op, &sh->ops.ack)) \
+ clear_bit(op, &pend); \
+ else \
+ ack++; \
+ } else \
+ clear_bit(op, &pend); \
+} while (0)
+
+/* find new work to run, do not resubmit work that is already
+ * in flight
+ */
+static unsigned long get_stripe_work(struct stripe_head *sh)
+{
+ unsigned long pending;
+ int ack = 0;
+
+ pending = sh->ops.pending;
+
+ test_and_ack_op(STRIPE_OP_BIOFILL, pending);
+ test_and_ack_op(STRIPE_OP_COMPUTE_BLK, pending);
+ test_and_ack_op(STRIPE_OP_PREXOR, pending);
+ test_and_ack_op(STRIPE_OP_BIODRAIN, pending);
+ test_and_ack_op(STRIPE_OP_POSTXOR, pending);
+ test_and_ack_op(STRIPE_OP_CHECK, pending);
+ if (test_and_clear_bit(STRIPE_OP_IO, &sh->ops.pending))
+ ack++;
+
+ sh->ops.count -= ack;
+ BUG_ON(sh->ops.count < 0);
+
+ return pending;
+}
+
+static int
+raid5_end_read_request(struct bio *bi, unsigned int bytes_done, int error);
+static int
+raid5_end_write_request (struct bio *bi, unsigned int bytes_done, int error);
+
+static void ops_run_io(struct stripe_head *sh)
+{
+ raid5_conf_t *conf = sh->raid_conf;
+ int i, disks = sh->disks;
+
+ might_sleep();
+
+ for (i = disks; i--; ) {
+ int rw;
+ struct bio *bi;
+ mdk_rdev_t *rdev;
+ if (test_and_clear_bit(R5_Wantwrite, &sh->dev[i].flags))
+ rw = WRITE;
+ else if (test_and_clear_bit(R5_Wantread, &sh->dev[i].flags))
+ rw = READ;
+ else
+ continue;
+
+ bi = &sh->dev[i].req;
+
+ bi->bi_rw = rw;
+ if (rw == WRITE)
+ bi->bi_end_io = raid5_end_write_request;
+ else
+ bi->bi_end_io = raid5_end_read_request;
+
+ rcu_read_lock();
+ rdev = rcu_dereference(conf->disks[i].rdev);
+ if (rdev && test_bit(Faulty, &rdev->flags))
+ rdev = NULL;
+ if (rdev)
+ atomic_inc(&rdev->nr_pending);
+ rcu_read_unlock();
+
+ if (rdev) {
+ if (test_bit(STRIPE_SYNCING, &sh->state) ||
+ test_bit(STRIPE_EXPAND_SOURCE, &sh->state) ||
+ test_bit(STRIPE_EXPAND_READY, &sh->state))
+ md_sync_acct(rdev->bdev, STRIPE_SECTORS);
+
+ bi->bi_bdev = rdev->bdev;
+ pr_debug("%s: for %llu schedule op %ld on disc %d\n",
+ __FUNCTION__, (unsigned long long)sh->sector,
+ bi->bi_rw, i);
+ atomic_inc(&sh->count);
+ bi->bi_sector = sh->sector + rdev->data_offset;
+ bi->bi_flags = 1 << BIO_UPTODATE;
+ bi->bi_vcnt = 1;
+ bi->bi_max_vecs = 1;
+ bi->bi_idx = 0;
+ bi->bi_io_vec = &sh->dev[i].vec;
+ bi->bi_io_vec[0].bv_len = STRIPE_SIZE;
+ bi->bi_io_vec[0].bv_offset = 0;
+ bi->bi_size = STRIPE_SIZE;
+ bi->bi_next = NULL;
+ if (rw == WRITE &&
+ test_bit(R5_ReWrite, &sh->dev[i].flags))
+ atomic_add(STRIPE_SECTORS,
+ &rdev->corrected_errors);
+ generic_make_request(bi);
+ } else {
+ if (rw == WRITE)
+ set_bit(STRIPE_DEGRADED, &sh->state);
+ pr_debug("skip op %ld on disc %d for sector %llu\n",
+ bi->bi_rw, i, (unsigned long long)sh->sector);
+ clear_bit(R5_LOCKED, &sh->dev[i].flags);
+ set_bit(STRIPE_HANDLE, &sh->state);
+ }
+ }
+}
+
+static struct dma_async_tx_descriptor *
+async_copy_data(int frombio, struct bio *bio, struct page *page,
+ sector_t sector, struct dma_async_tx_descriptor *tx)
+{
+ struct bio_vec *bvl;
+ struct page *bio_page;
+ int i;
+ int page_offset;
+
+ if (bio->bi_sector >= sector)
+ page_offset = (signed)(bio->bi_sector - sector) * 512;
+ else
+ page_offset = (signed)(sector - bio->bi_sector) * -512;
+ bio_for_each_segment(bvl, bio, i) {
+ int len = bio_iovec_idx(bio, i)->bv_len;
+ int clen;
+ int b_offset = 0;
+
+ if (page_offset < 0) {
+ b_offset = -page_offset;
+ page_offset += b_offset;
+ len -= b_offset;
+ }
+
+ if (len > 0 && page_offset + len > STRIPE_SIZE)
+ clen = STRIPE_SIZE - page_offset;
+ else
+ clen = len;
+
+ if (clen > 0) {
+ b_offset += bio_iovec_idx(bio, i)->bv_offset;
+ bio_page = bio_iovec_idx(bio, i)->bv_page;
+ if (frombio)
+ tx = async_memcpy(page, bio_page, page_offset,
+ b_offset, clen,
+ ASYNC_TX_DEP_ACK | ASYNC_TX_KMAP_SRC,
+ tx, NULL, NULL);
+ else
+ tx = async_memcpy(bio_page, page, b_offset,
+ page_offset, clen,
+ ASYNC_TX_DEP_ACK | ASYNC_TX_KMAP_DST,
+ tx, NULL, NULL);
+ }
+ if (clen < len) /* hit end of page */
+ break;
+ page_offset += len;
+ }
+
+ return tx;
+}
+
+static void ops_complete_biofill(void *stripe_head_ref)
+{
+ struct stripe_head *sh = stripe_head_ref;
+ struct bio *return_bi = NULL;
+ raid5_conf_t *conf = sh->raid_conf;
+ int i, more_to_read = 0;
+
+ pr_debug("%s: stripe %llu\n", __FUNCTION__,
+ (unsigned long long)sh->sector);
+
+ /* clear completed biofills */
+ for (i = sh->disks; i--; ) {
+ struct r5dev *dev = &sh->dev[i];
+ /* check if this stripe has new incoming reads */
+ if (dev->toread)
+ more_to_read++;
+
+ /* acknowledge completion of a biofill operation */
+ /* and check if we need to reply to a read request
+ */
+ if (test_bit(R5_Wantfill, &dev->flags) && !dev->toread) {
+ struct bio *rbi, *rbi2;
+ clear_bit(R5_Wantfill, &dev->flags);
+
+ /* The access to dev->read is outside of the
+ * spin_lock_irq(&conf->device_lock), but is protected
+ * by the STRIPE_OP_BIOFILL pending bit
+ */
+ BUG_ON(!dev->read);
+ rbi = dev->read;
+ dev->read = NULL;
+ while (rbi && rbi->bi_sector <
+ dev->sector + STRIPE_SECTORS) {
+ rbi2 = r5_next_bio(rbi, dev->sector);
+ spin_lock_irq(&conf->device_lock);
+ if (--rbi->bi_phys_segments == 0) {
+ rbi->bi_next = return_bi;
+ return_bi = rbi;
+ }
+ spin_unlock_irq(&conf->device_lock);
+ rbi = rbi2;
+ }
+ }
+ }
+ clear_bit(STRIPE_OP_BIOFILL, &sh->ops.ack);
+ clear_bit(STRIPE_OP_BIOFILL, &sh->ops.pending);
+
+ return_io(return_bi);
+
+ if (more_to_read)
+ set_bit(STRIPE_HANDLE, &sh->state);
+ release_stripe(sh);
+}
+
+static void ops_run_biofill(struct stripe_head *sh)
+{
+ struct dma_async_tx_descriptor *tx = NULL;
+ raid5_conf_t *conf = sh->raid_conf;
+ int i;
+
+ pr_debug("%s: stripe %llu\n", __FUNCTION__,
+ (unsigned long long)sh->sector);
+
+ for (i = sh->disks; i--; ) {
+ struct r5dev *dev = &sh->dev[i];
+ if (test_bit(R5_Wantfill, &dev->flags)) {
+ struct bio *rbi;
+ spin_lock_irq(&conf->device_lock);
+ dev->read = rbi = dev->toread;
+ dev->toread = NULL;
+ spin_unlock_irq(&conf->device_lock);
+ while (rbi && rbi->bi_sector <
+ dev->sector + STRIPE_SECTORS) {
+ tx = async_copy_data(0, rbi, dev->page,
+ dev->sector, tx);
+ rbi = r5_next_bio(rbi, dev->sector);
+ }
+ }
+ }
+
+ atomic_inc(&sh->count);
+ async_trigger_callback(ASYNC_TX_DEP_ACK | ASYNC_TX_ACK, tx,
+ ops_complete_biofill, sh);
+}
+
+static void ops_complete_compute5(void *stripe_head_ref)
+{
+ struct stripe_head *sh = stripe_head_ref;
+ int target = sh->ops.target;
+ struct r5dev *tgt = &sh->dev[target];
+
+ pr_debug("%s: stripe %llu\n", __FUNCTION__,
+ (unsigned long long)sh->sector);
+
+ set_bit(R5_UPTODATE, &tgt->flags);
+ BUG_ON(!test_bit(R5_Wantcompute, &tgt->flags));
+ clear_bit(R5_Wantcompute, &tgt->flags);
+ set_bit(STRIPE_OP_COMPUTE_BLK, &sh->ops.complete);
+ set_bit(STRIPE_HANDLE, &sh->state);
+ release_stripe(sh);
+}
+
+static struct dma_async_tx_descriptor *
+ops_run_compute5(struct stripe_head *sh, unsigned long pending)
+{
+ /* kernel stack size limits the total number of disks */
+ int disks = sh->disks;
+ struct page *xor_srcs[disks];
+ int target = sh->ops.target;
+ struct r5dev *tgt = &sh->dev[target];
+ struct page *xor_dest = tgt->page;
+ int count = 0;
+ struct dma_async_tx_descriptor *tx;
+ int i;
+
+ pr_debug("%s: stripe %llu block: %d\n",
+ __FUNCTION__, (unsigned long long)sh->sector, target);
+ BUG_ON(!test_bit(R5_Wantcompute, &tgt->flags));
+
+ for (i = disks; i--; )
+ if (i != target)
+ xor_srcs[count++] = sh->dev[i].page;
+
+ atomic_inc(&sh->count);
+
+ if (unlikely(count == 1))
+ tx = async_memcpy(xor_dest, xor_srcs[0], 0, 0, STRIPE_SIZE,
+ 0, NULL, ops_complete_compute5, sh);
+ else
+ tx = async_xor(xor_dest, xor_srcs, 0, count, STRIPE_SIZE,
+ ASYNC_TX_XOR_ZERO_DST, NULL,
+ ops_complete_compute5, sh);
+
+ /* ack now if postxor is not set to be run */
+ if (tx && !test_bit(STRIPE_OP_POSTXOR, &pending))
+ async_tx_ack(tx);
+
+ return tx;
+}
+
+static void ops_complete_prexor(void *stripe_head_ref)
+{
+ struct stripe_head *sh = stripe_head_ref;
+
+ pr_debug("%s: stripe %llu\n", __FUNCTION__,
+ (unsigned long long)sh->sector);
+
+ set_bit(STRIPE_OP_PREXOR, &sh->ops.complete);
+}
+
+static struct dma_async_tx_descriptor *
+ops_run_prexor(struct stripe_head *sh, struct dma_async_tx_descriptor *tx)
+{
+ /* kernel stack size limits the total number of disks */
+ int disks = sh->disks;
+ struct page *xor_srcs[disks];
+ int count = 0, pd_idx = sh->pd_idx, i;
+
+ /* existing parity data subtracted */
+ struct page *xor_dest = xor_srcs[count++] = sh->dev[pd_idx].page;
+
+ pr_debug("%s: stripe %llu\n", __FUNCTION__,
+ (unsigned long long)sh->sector);
+
+ for (i = disks; i--; ) {
+ struct r5dev *dev = &sh->dev[i];
+ /* Only process blocks that are known to be uptodate */
+ if (dev->towrite && test_bit(R5_Wantprexor, &dev->flags))
+ xor_srcs[count++] = dev->page;
+ }
+
+ tx = async_xor(xor_dest, xor_srcs, 0, count, STRIPE_SIZE,
+ ASYNC_TX_DEP_ACK | ASYNC_TX_XOR_DROP_DST, tx,
+ ops_complete_prexor, sh);
+
+ return tx;
+}
+
+static struct dma_async_tx_descriptor *
+ops_run_biodrain(struct stripe_head *sh, struct dma_async_tx_descriptor *tx)
+{
+ int disks = sh->disks;
+ int pd_idx = sh->pd_idx, i;
+
+ /* check if prexor is active which means only process blocks
+ * that are part of a read-modify-write (Wantprexor)
+ */
+ int prexor = test_bit(STRIPE_OP_PREXOR, &sh->ops.pending);
+
+ pr_debug("%s: stripe %llu\n", __FUNCTION__,
+ (unsigned long long)sh->sector);
+
+ for (i = disks; i--; ) {
+ struct r5dev *dev = &sh->dev[i];
+ struct bio *chosen;
+ int towrite;
+
+ towrite = 0;
+ if (prexor) { /* rmw */
+ if (dev->towrite &&
+ test_bit(R5_Wantprexor, &dev->flags))
+ towrite = 1;
+ } else { /* rcw */
+ if (i != pd_idx && dev->towrite &&
+ test_bit(R5_LOCKED, &dev->flags))
+ towrite = 1;
+ }
+
+ if (towrite) {
+ struct bio *wbi;
+
+ spin_lock(&sh->lock);
+ chosen = dev->towrite;
+ dev->towrite = NULL;
+ BUG_ON(dev->written);
+ wbi = dev->written = chosen;
+ spin_unlock(&sh->lock);
+
+ while (wbi && wbi->bi_sector <
+ dev->sector + STRIPE_SECTORS) {
+ tx = async_copy_data(1, wbi, dev->page,
+ dev->sector, tx);
+ wbi = r5_next_bio(wbi, dev->sector);
+ }
+ }
+ }
+
+ return tx;
+}
+
+static void ops_complete_postxor(void *stripe_head_ref)
+{
+ struct stripe_head *sh = stripe_head_ref;
+
+ pr_debug("%s: stripe %llu\n", __FUNCTION__,
+ (unsigned long long)sh->sector);
+
+ set_bit(STRIPE_OP_POSTXOR, &sh->ops.complete);
+ set_bit(STRIPE_HANDLE, &sh->state);
+ release_stripe(sh);
+}
+
+static void ops_complete_write(void *stripe_head_ref)
+{
+ struct stripe_head *sh = stripe_head_ref;
+ int disks = sh->disks, i, pd_idx = sh->pd_idx;
+
+ pr_debug("%s: stripe %llu\n", __FUNCTION__,
+ (unsigned long long)sh->sector);
+
+ for (i = disks; i--; ) {
+ struct r5dev *dev = &sh->dev[i];
+ if (dev->written || i == pd_idx)
+ set_bit(R5_UPTODATE, &dev->flags);
+ }
+
+ set_bit(STRIPE_OP_BIODRAIN, &sh->ops.complete);
+ set_bit(STRIPE_OP_POSTXOR, &sh->ops.complete);
+
+ set_bit(STRIPE_HANDLE, &sh->state);
+ release_stripe(sh);
+}
+
+static void
+ops_run_postxor(struct stripe_head *sh, struct dma_async_tx_descriptor *tx)
+{
+ /* kernel stack size limits the total number of disks */
+ int disks = sh->disks;
+ struct page *xor_srcs[disks];
+
+ int count = 0, pd_idx = sh->pd_idx, i;
+ struct page *xor_dest;
+ int prexor = test_bit(STRIPE_OP_PREXOR, &sh->ops.pending);
+ unsigned long flags;
+ dma_async_tx_callback callback;
+
+ pr_debug("%s: stripe %llu\n", __FUNCTION__,
+ (unsigned long long)sh->sector);
+
+ /* check if prexor is active which means only process blocks
+ * that are part of a read-modify-write (written)
+ */
+ if (prexor) {
+ xor_dest = xor_srcs[count++] = sh->dev[pd_idx].page;
+ for (i = disks; i--; ) {
+ struct r5dev *dev = &sh->dev[i];
+ if (dev->written)
+ xor_srcs[count++] = dev->page;
+ }
+ } else {
+ xor_dest = sh->dev[pd_idx].page;
+ for (i = disks; i--; ) {
+ struct r5dev *dev = &sh->dev[i];
+ if (i != pd_idx)
+ xor_srcs[count++] = dev->page;
+ }
+ }
+
+ /* check whether this postxor is part of a write */
+ callback = test_bit(STRIPE_OP_BIODRAIN, &sh->ops.pending) ?
+ ops_complete_write : ops_complete_postxor;
+
+ /* 1/ if we prexor'd then the dest is reused as a source
+ * 2/ if we did not prexor then we are redoing the parity
+ * set ASYNC_TX_XOR_DROP_DST and ASYNC_TX_XOR_ZERO_DST
+ * for the synchronous xor case
+ */
+ flags = ASYNC_TX_DEP_ACK | ASYNC_TX_ACK |
+ (prexor ? ASYNC_TX_XOR_DROP_DST : ASYNC_TX_XOR_ZERO_DST);
+
+ atomic_inc(&sh->count);
+
+ if (unlikely(count == 1)) {
+ flags &= ~(ASYNC_TX_XOR_DROP_DST | ASYNC_TX_XOR_ZERO_DST);
+ tx = async_memcpy(xor_dest, xor_srcs[0], 0, 0, STRIPE_SIZE,
+ flags, tx, callback, sh);
+ } else
+ tx = async_xor(xor_dest, xor_srcs, 0, count, STRIPE_SIZE,
+ flags, tx, callback, sh);
+}
+
+static void ops_complete_check(void *stripe_head_ref)
+{
+ struct stripe_head *sh = stripe_head_ref;
+ int pd_idx = sh->pd_idx;
+
+ pr_debug("%s: stripe %llu\n", __FUNCTION__,
+ (unsigned long long)sh->sector);
+
+ if (test_and_clear_bit(STRIPE_OP_MOD_DMA_CHECK, &sh->ops.pending) &&
+ sh->ops.zero_sum_result == 0)
+ set_bit(R5_UPTODATE, &sh->dev[pd_idx].flags);
+
+ set_bit(STRIPE_OP_CHECK, &sh->ops.complete);
+ set_bit(STRIPE_HANDLE, &sh->state);
+ release_stripe(sh);
+}
+
+static void ops_run_check(struct stripe_head *sh)
+{
+ /* kernel stack size limits the total number of disks */
+ int disks = sh->disks;
+ struct page *xor_srcs[disks];
+ struct dma_async_tx_descriptor *tx;
+
+ int count = 0, pd_idx = sh->pd_idx, i;
+ struct page *xor_dest = xor_srcs[count++] = sh->dev[pd_idx].page;
+
+ pr_debug("%s: stripe %llu\n", __FUNCTION__,
+ (unsigned long long)sh->sector);
+
+ for (i = disks; i--; ) {
+ struct r5dev *dev = &sh->dev[i];
+ if (i != pd_idx)
+ xor_srcs[count++] = dev->page;
+ }
+
+ tx = async_xor_zero_sum(xor_dest, xor_srcs, 0, count, STRIPE_SIZE,
+ &sh->ops.zero_sum_result, 0, NULL, NULL, NULL);
+
+ if (tx)
+ set_bit(STRIPE_OP_MOD_DMA_CHECK, &sh->ops.pending);
+ else
+ clear_bit(STRIPE_OP_MOD_DMA_CHECK, &sh->ops.pending);
+
+ atomic_inc(&sh->count);
+ tx = async_trigger_callback(ASYNC_TX_DEP_ACK | ASYNC_TX_ACK, tx,
+ ops_complete_check, sh);
+}
+
+static void raid5_run_ops(struct stripe_head *sh, unsigned long pending)
+{
+ int overlap_clear = 0, i, disks = sh->disks;
+ struct dma_async_tx_descriptor *tx = NULL;
+
+ if (test_bit(STRIPE_OP_BIOFILL, &pending)) {
+ ops_run_biofill(sh);
+ overlap_clear++;
+ }
+
+ if (test_bit(STRIPE_OP_COMPUTE_BLK, &pending))
+ tx = ops_run_compute5(sh, pending);
+
+ if (test_bit(STRIPE_OP_PREXOR, &pending))
+ tx = ops_run_prexor(sh, tx);
+
+ if (test_bit(STRIPE_OP_BIODRAIN, &pending)) {
+ tx = ops_run_biodrain(sh, tx);
+ overlap_clear++;
+ }
+
+ if (test_bit(STRIPE_OP_POSTXOR, &pending))
+ ops_run_postxor(sh, tx);
+
+ if (test_bit(STRIPE_OP_CHECK, &pending))
+ ops_run_check(sh);
+
+ if (test_bit(STRIPE_OP_IO, &pending))
+ ops_run_io(sh);
+
+ if (overlap_clear)
+ for (i = disks; i--; ) {
+ struct r5dev *dev = &sh->dev[i];
+ if (test_and_clear_bit(R5_Overlap, &dev->flags))
+ wake_up(&sh->raid_conf->wait_for_overlap);
+ }
+}
+
static int grow_one_stripe(raid5_conf_t *conf)
{
struct stripe_head *sh;
@@ -537,8 +1130,8 @@ static int raid5_end_read_request(struct bio * bi, unsigned int bytes_done,
if (bi == &sh->dev[i].req)
break;
- PRINTK("end_read_request %llu/%d, count: %d, uptodate %d.\n",
- (unsigned long long)sh->sector, i, atomic_read(&sh->count),
+ pr_debug("end_read_request %llu/%d, count: %d, uptodate %d.\n",
+ (unsigned long long)sh->sector, i, atomic_read(&sh->count),
uptodate);
if (i == disks) {
BUG();
@@ -613,7 +1206,7 @@ static int raid5_end_write_request (struct bio *bi, unsigned int bytes_done,
if (bi == &sh->dev[i].req)
break;
- PRINTK("end_write_request %llu/%d, count %d, uptodate: %d.\n",
+ pr_debug("end_write_request %llu/%d, count %d, uptodate: %d.\n",
(unsigned long long)sh->sector, i, atomic_read(&sh->count),
uptodate);
if (i == disks) {
@@ -658,7 +1251,7 @@ static void error(mddev_t *mddev, mdk_rdev_t *rdev)
{
char b[BDEVNAME_SIZE];
raid5_conf_t *conf = (raid5_conf_t *) mddev->private;
- PRINTK("raid5: error called\n");
+ pr_debug("raid5: error called\n");
if (!test_bit(Faulty, &rdev->flags)) {
set_bit(MD_CHANGE_DEVS, &mddev->flags);
@@ -916,137 +1509,13 @@ static void copy_data(int frombio, struct bio *bio,
}
}
-#define check_xor() do { \
- if (count == MAX_XOR_BLOCKS) { \
- xor_block(count, STRIPE_SIZE, ptr); \
- count = 1; \
- } \
+#define check_xor() do { \
+ if (count == MAX_XOR_BLOCKS) { \
+ xor_blocks(count, STRIPE_SIZE, dest, ptr);\
+ count = 0; \
+ } \
} while(0)
-
-static void compute_block(struct stripe_head *sh, int dd_idx)
-{
- int i, count, disks = sh->disks;
- void *ptr[MAX_XOR_BLOCKS], *p;
-
- PRINTK("compute_block, stripe %llu, idx %d\n",
- (unsigned long long)sh->sector, dd_idx);
-
- ptr[0] = page_address(sh->dev[dd_idx].page);
- memset(ptr[0], 0, STRIPE_SIZE);
- count = 1;
- for (i = disks ; i--; ) {
- if (i == dd_idx)
- continue;
- p = page_address(sh->dev[i].page);
- if (test_bit(R5_UPTODATE, &sh->dev[i].flags))
- ptr[count++] = p;
- else
- printk(KERN_ERR "compute_block() %d, stripe %llu, %d"
- " not present\n", dd_idx,
- (unsigned long long)sh->sector, i);
-
- check_xor();
- }
- if (count != 1)
- xor_block(count, STRIPE_SIZE, ptr);
- set_bit(R5_UPTODATE, &sh->dev[dd_idx].flags);
-}
-
-static void compute_parity5(struct stripe_head *sh, int method)
-{
- raid5_conf_t *conf = sh->raid_conf;
- int i, pd_idx = sh->pd_idx, disks = sh->disks, count;
- void *ptr[MAX_XOR_BLOCKS];
- struct bio *chosen;
-
- PRINTK("compute_parity5, stripe %llu, method %d\n",
- (unsigned long long)sh->sector, method);
-
- count = 1;
- ptr[0] = page_address(sh->dev[pd_idx].page);
- switch(method) {
- case READ_MODIFY_WRITE:
- BUG_ON(!test_bit(R5_UPTODATE, &sh->dev[pd_idx].flags));
- for (i=disks ; i-- ;) {
- if (i==pd_idx)
- continue;
- if (sh->dev[i].towrite &&
- test_bit(R5_UPTODATE, &sh->dev[i].flags)) {
- ptr[count++] = page_address(sh->dev[i].page);
- chosen = sh->dev[i].towrite;
- sh->dev[i].towrite = NULL;
-
- if (test_and_clear_bit(R5_Overlap, &sh->dev[i].flags))
- wake_up(&conf->wait_for_overlap);
-
- BUG_ON(sh->dev[i].written);
- sh->dev[i].written = chosen;
- check_xor();
- }
- }
- break;
- case RECONSTRUCT_WRITE:
- memset(ptr[0], 0, STRIPE_SIZE);
- for (i= disks; i-- ;)
- if (i!=pd_idx && sh->dev[i].towrite) {
- chosen = sh->dev[i].towrite;
- sh->dev[i].towrite = NULL;
-
- if (test_and_clear_bit(R5_Overlap, &sh->dev[i].flags))
- wake_up(&conf->wait_for_overlap);
-
- BUG_ON(sh->dev[i].written);
- sh->dev[i].written = chosen;
- }
- break;
- case CHECK_PARITY:
- break;
- }
- if (count>1) {
- xor_block(count, STRIPE_SIZE, ptr);
- count = 1;
- }
-
- for (i = disks; i--;)
- if (sh->dev[i].written) {
- sector_t sector = sh->dev[i].sector;
- struct bio *wbi = sh->dev[i].written;
- while (wbi && wbi->bi_sector < sector + STRIPE_SECTORS) {
- copy_data(1, wbi, sh->dev[i].page, sector);
- wbi = r5_next_bio(wbi, sector);
- }
-
- set_bit(R5_LOCKED, &sh->dev[i].flags);
- set_bit(R5_UPTODATE, &sh->dev[i].flags);
- }
-
- switch(method) {
- case RECONSTRUCT_WRITE:
- case CHECK_PARITY:
- for (i=disks; i--;)
- if (i != pd_idx) {
- ptr[count++] = page_address(sh->dev[i].page);
- check_xor();
- }
- break;
- case READ_MODIFY_WRITE:
- for (i = disks; i--;)
- if (sh->dev[i].written) {
- ptr[count++] = page_address(sh->dev[i].page);
- check_xor();
- }
- }
- if (count != 1)
- xor_block(count, STRIPE_SIZE, ptr);
-
- if (method != CHECK_PARITY) {
- set_bit(R5_UPTODATE, &sh->dev[pd_idx].flags);
- set_bit(R5_LOCKED, &sh->dev[pd_idx].flags);
- } else
- clear_bit(R5_UPTODATE, &sh->dev[pd_idx].flags);
-}
-
static void compute_parity6(struct stripe_head *sh, int method)
{
raid6_conf_t *conf = sh->raid_conf;
@@ -1058,7 +1527,7 @@ static void compute_parity6(struct stripe_head *sh, int method)
qd_idx = raid6_next_disk(pd_idx, disks);
d0_idx = raid6_next_disk(qd_idx, disks);
- PRINTK("compute_parity, stripe %llu, method %d\n",
+ pr_debug("compute_parity, stripe %llu, method %d\n",
(unsigned long long)sh->sector, method);
switch(method) {
@@ -1132,20 +1601,20 @@ static void compute_parity6(struct stripe_head *sh, int method)
static void compute_block_1(struct stripe_head *sh, int dd_idx, int nozero)
{
int i, count, disks = sh->disks;
- void *ptr[MAX_XOR_BLOCKS], *p;
+ void *ptr[MAX_XOR_BLOCKS], *dest, *p;
int pd_idx = sh->pd_idx;
int qd_idx = raid6_next_disk(pd_idx, disks);
- PRINTK("compute_block_1, stripe %llu, idx %d\n",
+ pr_debug("compute_block_1, stripe %llu, idx %d\n",
(unsigned long long)sh->sector, dd_idx);
if ( dd_idx == qd_idx ) {
/* We're actually computing the Q drive */
compute_parity6(sh, UPDATE_PARITY);
} else {
- ptr[0] = page_address(sh->dev[dd_idx].page);
- if (!nozero) memset(ptr[0], 0, STRIPE_SIZE);
- count = 1;
+ dest = page_address(sh->dev[dd_idx].page);
+ if (!nozero) memset(dest, 0, STRIPE_SIZE);
+ count = 0;
for (i = disks ; i--; ) {
if (i == dd_idx || i == qd_idx)
continue;
@@ -1159,8 +1628,8 @@ static void compute_block_1(struct stripe_head *sh, int dd_idx, int nozero)
check_xor();
}
- if (count != 1)
- xor_block(count, STRIPE_SIZE, ptr);
+ if (count)
+ xor_blocks(count, STRIPE_SIZE, dest, ptr);
if (!nozero) set_bit(R5_UPTODATE, &sh->dev[dd_idx].flags);
else clear_bit(R5_UPTODATE, &sh->dev[dd_idx].flags);
}
@@ -1183,7 +1652,7 @@ static void compute_block_2(struct stripe_head *sh, int dd_idx1, int dd_idx2)
BUG_ON(faila == failb);
if ( failb < faila ) { int tmp = faila; faila = failb; failb = tmp; }
- PRINTK("compute_block_2, stripe %llu, idx %d,%d (%d,%d)\n",
+ pr_debug("compute_block_2, stripe %llu, idx %d,%d (%d,%d)\n",
(unsigned long long)sh->sector, dd_idx1, dd_idx2, faila, failb);
if ( failb == disks-1 ) {
@@ -1229,7 +1698,79 @@ static void compute_block_2(struct stripe_head *sh, int dd_idx1, int dd_idx2)
}
}
+static int
+handle_write_operations5(struct stripe_head *sh, int rcw, int expand)
+{
+ int i, pd_idx = sh->pd_idx, disks = sh->disks;
+ int locked = 0;
+ if (rcw) {
+ /* if we are not expanding this is a proper write request, and
+ * there will be bios with new data to be drained into the
+ * stripe cache
+ */
+ if (!expand) {
+ set_bit(STRIPE_OP_BIODRAIN, &sh->ops.pending);
+ sh->ops.count++;
+ }
+
+ set_bit(STRIPE_OP_POSTXOR, &sh->ops.pending);
+ sh->ops.count++;
+
+ for (i = disks; i--; ) {
+ struct r5dev *dev = &sh->dev[i];
+
+ if (dev->towrite) {
+ set_bit(R5_LOCKED, &dev->flags);
+ if (!expand)
+ clear_bit(R5_UPTODATE, &dev->flags);
+ locked++;
+ }
+ }
+ } else {
+ BUG_ON(!(test_bit(R5_UPTODATE, &sh->dev[pd_idx].flags) ||
+ test_bit(R5_Wantcompute, &sh->dev[pd_idx].flags)));
+
+ set_bit(STRIPE_OP_PREXOR, &sh->ops.pending);
+ set_bit(STRIPE_OP_BIODRAIN, &sh->ops.pending);
+ set_bit(STRIPE_OP_POSTXOR, &sh->ops.pending);
+
+ sh->ops.count += 3;
+
+ for (i = disks; i--; ) {
+ struct r5dev *dev = &sh->dev[i];
+ if (i == pd_idx)
+ continue;
+
+ /* For a read-modify write there may be blocks that are
+ * locked for reading while others are ready to be
+ * written so we distinguish these blocks by the
+ * R5_Wantprexor bit
+ */
+ if (dev->towrite &&
+ (test_bit(R5_UPTODATE, &dev->flags) ||
+ test_bit(R5_Wantcompute, &dev->flags))) {
+ set_bit(R5_Wantprexor, &dev->flags);
+ set_bit(R5_LOCKED, &dev->flags);
+ clear_bit(R5_UPTODATE, &dev->flags);
+ locked++;
+ }
+ }
+ }
+
+ /* keep the parity disk locked while asynchronous operations
+ * are in flight
+ */
+ set_bit(R5_LOCKED, &sh->dev[pd_idx].flags);
+ clear_bit(R5_UPTODATE, &sh->dev[pd_idx].flags);
+ locked++;
+
+ pr_debug("%s: stripe %llu locked: %d pending: %lx\n",
+ __FUNCTION__, (unsigned long long)sh->sector,
+ locked, sh->ops.pending);
+
+ return locked;
+}
/*
* Each stripe/dev can have one or more bion attached.
@@ -1242,7 +1783,7 @@ static int add_stripe_bio(struct stripe_head *sh, struct bio *bi, int dd_idx, in
raid5_conf_t *conf = sh->raid_conf;
int firstwrite=0;
- PRINTK("adding bh b#%llu to stripe s#%llu\n",
+ pr_debug("adding bh b#%llu to stripe s#%llu\n",
(unsigned long long)bi->bi_sector,
(unsigned long long)sh->sector);
@@ -1271,7 +1812,7 @@ static int add_stripe_bio(struct stripe_head *sh, struct bio *bi, int dd_idx, in
spin_unlock_irq(&conf->device_lock);
spin_unlock(&sh->lock);
- PRINTK("added bi b#%llu to stripe s#%llu, disk %d.\n",
+ pr_debug("added bi b#%llu to stripe s#%llu, disk %d.\n",
(unsigned long long)bi->bi_sector,
(unsigned long long)sh->sector, dd_idx);
@@ -1326,6 +1867,729 @@ static int stripe_to_pdidx(sector_t stripe, raid5_conf_t *conf, int disks)
return pd_idx;
}
+static void
+handle_requests_to_failed_array(raid5_conf_t *conf, struct stripe_head *sh,
+ struct stripe_head_state *s, int disks,
+ struct bio **return_bi)
+{
+ int i;
+ for (i = disks; i--; ) {
+ struct bio *bi;
+ int bitmap_end = 0;
+
+ if (test_bit(R5_ReadError, &sh->dev[i].flags)) {
+ mdk_rdev_t *rdev;
+ rcu_read_lock();
+ rdev = rcu_dereference(conf->disks[i].rdev);
+ if (rdev && test_bit(In_sync, &rdev->flags))
+ /* multiple read failures in one stripe */
+ md_error(conf->mddev, rdev);
+ rcu_read_unlock();
+ }
+ spin_lock_irq(&conf->device_lock);
+ /* fail all writes first */
+ bi = sh->dev[i].towrite;
+ sh->dev[i].towrite = NULL;
+ if (bi) {
+ s->to_write--;
+ bitmap_end = 1;
+ }
+
+ if (test_and_clear_bit(R5_Overlap, &sh->dev[i].flags))
+ wake_up(&conf->wait_for_overlap);
+
+ while (bi && bi->bi_sector <
+ sh->dev[i].sector + STRIPE_SECTORS) {
+ struct bio *nextbi = r5_next_bio(bi, sh->dev[i].sector);
+ clear_bit(BIO_UPTODATE, &bi->bi_flags);
+ if (--bi->bi_phys_segments == 0) {
+ md_write_end(conf->mddev);
+ bi->bi_next = *return_bi;
+ *return_bi = bi;
+ }
+ bi = nextbi;
+ }
+ /* and fail all 'written' */
+ bi = sh->dev[i].written;
+ sh->dev[i].written = NULL;
+ if (bi) bitmap_end = 1;
+ while (bi && bi->bi_sector <
+ sh->dev[i].sector + STRIPE_SECTORS) {
+ struct bio *bi2 = r5_next_bio(bi, sh->dev[i].sector);
+ clear_bit(BIO_UPTODATE, &bi->bi_flags);
+ if (--bi->bi_phys_segments == 0) {
+ md_write_end(conf->mddev);
+ bi->bi_next = *return_bi;
+ *return_bi = bi;
+ }
+ bi = bi2;
+ }
+
+ /* fail any reads if this device is non-operational and
+ * the data has not reached the cache yet.
+ */
+ if (!test_bit(R5_Wantfill, &sh->dev[i].flags) &&
+ (!test_bit(R5_Insync, &sh->dev[i].flags) ||
+ test_bit(R5_ReadError, &sh->dev[i].flags))) {
+ bi = sh->dev[i].toread;
+ sh->dev[i].toread = NULL;
+ if (test_and_clear_bit(R5_Overlap, &sh->dev[i].flags))
+ wake_up(&conf->wait_for_overlap);
+ if (bi) s->to_read--;
+ while (bi && bi->bi_sector <
+ sh->dev[i].sector + STRIPE_SECTORS) {
+ struct bio *nextbi =
+ r5_next_bio(bi, sh->dev[i].sector);
+ clear_bit(BIO_UPTODATE, &bi->bi_flags);
+ if (--bi->bi_phys_segments == 0) {
+ bi->bi_next = *return_bi;
+ *return_bi = bi;
+ }
+ bi = nextbi;
+ }
+ }
+ spin_unlock_irq(&conf->device_lock);
+ if (bitmap_end)
+ bitmap_endwrite(conf->mddev->bitmap, sh->sector,
+ STRIPE_SECTORS, 0, 0);
+ }
+
+}
+
+/* __handle_issuing_new_read_requests5 - returns 0 if there are no more disks
+ * to process
+ */
+static int __handle_issuing_new_read_requests5(struct stripe_head *sh,
+ struct stripe_head_state *s, int disk_idx, int disks)
+{
+ struct r5dev *dev = &sh->dev[disk_idx];
+ struct r5dev *failed_dev = &sh->dev[s->failed_num];
+
+ /* don't schedule compute operations or reads on the parity block while
+ * a check is in flight
+ */
+ if ((disk_idx == sh->pd_idx) &&
+ test_bit(STRIPE_OP_CHECK, &sh->ops.pending))
+ return ~0;
+
+ /* is the data in this block needed, and can we get it? */
+ if (!test_bit(R5_LOCKED, &dev->flags) &&
+ !test_bit(R5_UPTODATE, &dev->flags) && (dev->toread ||
+ (dev->towrite && !test_bit(R5_OVERWRITE, &dev->flags)) ||
+ s->syncing || s->expanding || (s->failed &&
+ (failed_dev->toread || (failed_dev->towrite &&
+ !test_bit(R5_OVERWRITE, &failed_dev->flags)
+ ))))) {
+ /* 1/ We would like to get this block, possibly by computing it,
+ * but we might not be able to.
+ *
+ * 2/ Since parity check operations potentially make the parity
+ * block !uptodate it will need to be refreshed before any
+ * compute operations on data disks are scheduled.
+ *
+ * 3/ We hold off parity block re-reads until check operations
+ * have quiesced.
+ */
+ if ((s->uptodate == disks - 1) &&
+ !test_bit(STRIPE_OP_CHECK, &sh->ops.pending)) {
+ set_bit(STRIPE_OP_COMPUTE_BLK, &sh->ops.pending);
+ set_bit(R5_Wantcompute, &dev->flags);
+ sh->ops.target = disk_idx;
+ s->req_compute = 1;
+ sh->ops.count++;
+ /* Careful: from this point on 'uptodate' is in the eye
+ * of raid5_run_ops which services 'compute' operations
+ * before writes. R5_Wantcompute flags a block that will
+ * be R5_UPTODATE by the time it is needed for a
+ * subsequent operation.
+ */
+ s->uptodate++;
+ return 0; /* uptodate + compute == disks */
+ } else if ((s->uptodate < disks - 1) &&
+ test_bit(R5_Insync, &dev->flags)) {
+ /* Note: we hold off compute operations while checks are
+ * in flight, but we still prefer 'compute' over 'read'
+ * hence we only read if (uptodate < * disks-1)
+ */
+ set_bit(R5_LOCKED, &dev->flags);
+ set_bit(R5_Wantread, &dev->flags);
+ if (!test_and_set_bit(STRIPE_OP_IO, &sh->ops.pending))
+ sh->ops.count++;
+ s->locked++;
+ pr_debug("Reading block %d (sync=%d)\n", disk_idx,
+ s->syncing);
+ }
+ }
+
+ return ~0;
+}
+
+static void handle_issuing_new_read_requests5(struct stripe_head *sh,
+ struct stripe_head_state *s, int disks)
+{
+ int i;
+
+ /* Clear completed compute operations. Parity recovery
+ * (STRIPE_OP_MOD_REPAIR_PD) implies a write-back which is handled
+ * later on in this routine
+ */
+ if (test_bit(STRIPE_OP_COMPUTE_BLK, &sh->ops.complete) &&
+ !test_bit(STRIPE_OP_MOD_REPAIR_PD, &sh->ops.pending)) {
+ clear_bit(STRIPE_OP_COMPUTE_BLK, &sh->ops.complete);
+ clear_bit(STRIPE_OP_COMPUTE_BLK, &sh->ops.ack);
+ clear_bit(STRIPE_OP_COMPUTE_BLK, &sh->ops.pending);
+ }
+
+ /* look for blocks to read/compute, skip this if a compute
+ * is already in flight, or if the stripe contents are in the
+ * midst of changing due to a write
+ */
+ if (!test_bit(STRIPE_OP_COMPUTE_BLK, &sh->ops.pending) &&
+ !test_bit(STRIPE_OP_PREXOR, &sh->ops.pending) &&
+ !test_bit(STRIPE_OP_POSTXOR, &sh->ops.pending)) {
+ for (i = disks; i--; )
+ if (__handle_issuing_new_read_requests5(
+ sh, s, i, disks) == 0)
+ break;
+ }
+ set_bit(STRIPE_HANDLE, &sh->state);
+}
+
+static void handle_issuing_new_read_requests6(struct stripe_head *sh,
+ struct stripe_head_state *s, struct r6_state *r6s,
+ int disks)
+{
+ int i;
+ for (i = disks; i--; ) {
+ struct r5dev *dev = &sh->dev[i];
+ if (!test_bit(R5_LOCKED, &dev->flags) &&
+ !test_bit(R5_UPTODATE, &dev->flags) &&
+ (dev->toread || (dev->towrite &&
+ !test_bit(R5_OVERWRITE, &dev->flags)) ||
+ s->syncing || s->expanding ||
+ (s->failed >= 1 &&
+ (sh->dev[r6s->failed_num[0]].toread ||
+ s->to_write)) ||
+ (s->failed >= 2 &&
+ (sh->dev[r6s->failed_num[1]].toread ||
+ s->to_write)))) {
+ /* we would like to get this block, possibly
+ * by computing it, but we might not be able to
+ */
+ if (s->uptodate == disks-1) {
+ pr_debug("Computing stripe %llu block %d\n",
+ (unsigned long long)sh->sector, i);
+ compute_block_1(sh, i, 0);
+ s->uptodate++;
+ } else if ( s->uptodate == disks-2 && s->failed >= 2 ) {
+ /* Computing 2-failure is *very* expensive; only
+ * do it if failed >= 2
+ */
+ int other;
+ for (other = disks; other--; ) {
+ if (other == i)
+ continue;
+ if (!test_bit(R5_UPTODATE,
+ &sh->dev[other].flags))
+ break;
+ }
+ BUG_ON(other < 0);
+ pr_debug("Computing stripe %llu blocks %d,%d\n",
+ (unsigned long long)sh->sector,
+ i, other);
+ compute_block_2(sh, i, other);
+ s->uptodate += 2;
+ } else if (test_bit(R5_Insync, &dev->flags)) {
+ set_bit(R5_LOCKED, &dev->flags);
+ set_bit(R5_Wantread, &dev->flags);
+ s->locked++;
+ pr_debug("Reading block %d (sync=%d)\n",
+ i, s->syncing);
+ }
+ }
+ }
+ set_bit(STRIPE_HANDLE, &sh->state);
+}
+
+
+/* handle_completed_write_requests
+ * any written block on an uptodate or failed drive can be returned.
+ * Note that if we 'wrote' to a failed drive, it will be UPTODATE, but
+ * never LOCKED, so we don't need to test 'failed' directly.
+ */
+static void handle_completed_write_requests(raid5_conf_t *conf,
+ struct stripe_head *sh, int disks, struct bio **return_bi)
+{
+ int i;
+ struct r5dev *dev;
+
+ for (i = disks; i--; )
+ if (sh->dev[i].written) {
+ dev = &sh->dev[i];
+ if (!test_bit(R5_LOCKED, &dev->flags) &&
+ test_bit(R5_UPTODATE, &dev->flags)) {
+ /* We can return any write requests */
+ struct bio *wbi, *wbi2;
+ int bitmap_end = 0;
+ pr_debug("Return write for disc %d\n", i);
+ spin_lock_irq(&conf->device_lock);
+ wbi = dev->written;
+ dev->written = NULL;
+ while (wbi && wbi->bi_sector <
+ dev->sector + STRIPE_SECTORS) {
+ wbi2 = r5_next_bio(wbi, dev->sector);
+ if (--wbi->bi_phys_segments == 0) {
+ md_write_end(conf->mddev);
+ wbi->bi_next = *return_bi;
+ *return_bi = wbi;
+ }
+ wbi = wbi2;
+ }
+ if (dev->towrite == NULL)
+ bitmap_end = 1;
+ spin_unlock_irq(&conf->device_lock);
+ if (bitmap_end)
+ bitmap_endwrite(conf->mddev->bitmap,
+ sh->sector,
+ STRIPE_SECTORS,
+ !test_bit(STRIPE_DEGRADED, &sh->state),
+ 0);
+ }
+ }
+}
+
+static void handle_issuing_new_write_requests5(raid5_conf_t *conf,
+ struct stripe_head *sh, struct stripe_head_state *s, int disks)
+{
+ int rmw = 0, rcw = 0, i;
+ for (i = disks; i--; ) {
+ /* would I have to read this buffer for read_modify_write */
+ struct r5dev *dev = &sh->dev[i];
+ if ((dev->towrite || i == sh->pd_idx) &&
+ !test_bit(R5_LOCKED, &dev->flags) &&
+ !(test_bit(R5_UPTODATE, &dev->flags) ||
+ test_bit(R5_Wantcompute, &dev->flags))) {
+ if (test_bit(R5_Insync, &dev->flags))
+ rmw++;
+ else
+ rmw += 2*disks; /* cannot read it */
+ }
+ /* Would I have to read this buffer for reconstruct_write */
+ if (!test_bit(R5_OVERWRITE, &dev->flags) && i != sh->pd_idx &&
+ !test_bit(R5_LOCKED, &dev->flags) &&
+ !(test_bit(R5_UPTODATE, &dev->flags) ||
+ test_bit(R5_Wantcompute, &dev->flags))) {
+ if (test_bit(R5_Insync, &dev->flags)) rcw++;
+ else
+ rcw += 2*disks;
+ }
+ }
+ pr_debug("for sector %llu, rmw=%d rcw=%d\n",
+ (unsigned long long)sh->sector, rmw, rcw);
+ set_bit(STRIPE_HANDLE, &sh->state);
+ if (rmw < rcw && rmw > 0)
+ /* prefer read-modify-write, but need to get some data */
+ for (i = disks; i--; ) {
+ struct r5dev *dev = &sh->dev[i];
+ if ((dev->towrite || i == sh->pd_idx) &&
+ !test_bit(R5_LOCKED, &dev->flags) &&
+ !(test_bit(R5_UPTODATE, &dev->flags) ||
+ test_bit(R5_Wantcompute, &dev->flags)) &&
+ test_bit(R5_Insync, &dev->flags)) {
+ if (
+ test_bit(STRIPE_PREREAD_ACTIVE, &sh->state)) {
+ pr_debug("Read_old block "
+ "%d for r-m-w\n", i);
+ set_bit(R5_LOCKED, &dev->flags);
+ set_bit(R5_Wantread, &dev->flags);
+ if (!test_and_set_bit(
+ STRIPE_OP_IO, &sh->ops.pending))
+ sh->ops.count++;
+ s->locked++;
+ } else {
+ set_bit(STRIPE_DELAYED, &sh->state);
+ set_bit(STRIPE_HANDLE, &sh->state);
+ }
+ }
+ }
+ if (rcw <= rmw && rcw > 0)
+ /* want reconstruct write, but need to get some data */
+ for (i = disks; i--; ) {
+ struct r5dev *dev = &sh->dev[i];
+ if (!test_bit(R5_OVERWRITE, &dev->flags) &&
+ i != sh->pd_idx &&
+ !test_bit(R5_LOCKED, &dev->flags) &&
+ !(test_bit(R5_UPTODATE, &dev->flags) ||
+ test_bit(R5_Wantcompute, &dev->flags)) &&
+ test_bit(R5_Insync, &dev->flags)) {
+ if (
+ test_bit(STRIPE_PREREAD_ACTIVE, &sh->state)) {
+ pr_debug("Read_old block "
+ "%d for Reconstruct\n", i);
+ set_bit(R5_LOCKED, &dev->flags);
+ set_bit(R5_Wantread, &dev->flags);
+ if (!test_and_set_bit(
+ STRIPE_OP_IO, &sh->ops.pending))
+ sh->ops.count++;
+ s->locked++;
+ } else {
+ set_bit(STRIPE_DELAYED, &sh->state);
+ set_bit(STRIPE_HANDLE, &sh->state);
+ }
+ }
+ }
+ /* now if nothing is locked, and if we have enough data,
+ * we can start a write request
+ */
+ /* since handle_stripe can be called at any time we need to handle the
+ * case where a compute block operation has been submitted and then a
+ * subsequent call wants to start a write request. raid5_run_ops only
+ * handles the case where compute block and postxor are requested
+ * simultaneously. If this is not the case then new writes need to be
+ * held off until the compute completes.
+ */
+ if ((s->req_compute ||
+ !test_bit(STRIPE_OP_COMPUTE_BLK, &sh->ops.pending)) &&
+ (s->locked == 0 && (rcw == 0 || rmw == 0) &&
+ !test_bit(STRIPE_BIT_DELAY, &sh->state)))
+ s->locked += handle_write_operations5(sh, rcw == 0, 0);
+}
+
+static void handle_issuing_new_write_requests6(raid5_conf_t *conf,
+ struct stripe_head *sh, struct stripe_head_state *s,
+ struct r6_state *r6s, int disks)
+{
+ int rcw = 0, must_compute = 0, pd_idx = sh->pd_idx, i;
+ int qd_idx = r6s->qd_idx;
+ for (i = disks; i--; ) {
+ struct r5dev *dev = &sh->dev[i];
+ /* Would I have to read this buffer for reconstruct_write */
+ if (!test_bit(R5_OVERWRITE, &dev->flags)
+ && i != pd_idx && i != qd_idx
+ && (!test_bit(R5_LOCKED, &dev->flags)
+ ) &&
+ !test_bit(R5_UPTODATE, &dev->flags)) {
+ if (test_bit(R5_Insync, &dev->flags)) rcw++;
+ else {
+ pr_debug("raid6: must_compute: "
+ "disk %d flags=%#lx\n", i, dev->flags);
+ must_compute++;
+ }
+ }
+ }
+ pr_debug("for sector %llu, rcw=%d, must_compute=%d\n",
+ (unsigned long long)sh->sector, rcw, must_compute);
+ set_bit(STRIPE_HANDLE, &sh->state);
+
+ if (rcw > 0)
+ /* want reconstruct write, but need to get some data */
+ for (i = disks; i--; ) {
+ struct r5dev *dev = &sh->dev[i];
+ if (!test_bit(R5_OVERWRITE, &dev->flags)
+ && !(s->failed == 0 && (i == pd_idx || i == qd_idx))
+ && !test_bit(R5_LOCKED, &dev->flags) &&
+ !test_bit(R5_UPTODATE, &dev->flags) &&
+ test_bit(R5_Insync, &dev->flags)) {
+ if (
+ test_bit(STRIPE_PREREAD_ACTIVE, &sh->state)) {
+ pr_debug("Read_old stripe %llu "
+ "block %d for Reconstruct\n",
+ (unsigned long long)sh->sector, i);
+ set_bit(R5_LOCKED, &dev->flags);
+ set_bit(R5_Wantread, &dev->flags);
+ s->locked++;
+ } else {
+ pr_debug("Request delayed stripe %llu "
+ "block %d for Reconstruct\n",
+ (unsigned long long)sh->sector, i);
+ set_bit(STRIPE_DELAYED, &sh->state);
+ set_bit(STRIPE_HANDLE, &sh->state);
+ }
+ }
+ }
+ /* now if nothing is locked, and if we have enough data, we can start a
+ * write request
+ */
+ if (s->locked == 0 && rcw == 0 &&
+ !test_bit(STRIPE_BIT_DELAY, &sh->state)) {
+ if (must_compute > 0) {
+ /* We have failed blocks and need to compute them */
+ switch (s->failed) {
+ case 0:
+ BUG();
+ case 1:
+ compute_block_1(sh, r6s->failed_num[0], 0);
+ break;
+ case 2:
+ compute_block_2(sh, r6s->failed_num[0],
+ r6s->failed_num[1]);
+ break;
+ default: /* This request should have been failed? */
+ BUG();
+ }
+ }
+
+ pr_debug("Computing parity for stripe %llu\n",
+ (unsigned long long)sh->sector);
+ compute_parity6(sh, RECONSTRUCT_WRITE);
+ /* now every locked buffer is ready to be written */
+ for (i = disks; i--; )
+ if (test_bit(R5_LOCKED, &sh->dev[i].flags)) {
+ pr_debug("Writing stripe %llu block %d\n",
+ (unsigned long long)sh->sector, i);
+ s->locked++;
+ set_bit(R5_Wantwrite, &sh->dev[i].flags);
+ }
+ /* after a RECONSTRUCT_WRITE, the stripe MUST be in-sync */
+ set_bit(STRIPE_INSYNC, &sh->state);
+
+ if (test_and_clear_bit(STRIPE_PREREAD_ACTIVE, &sh->state)) {
+ atomic_dec(&conf->preread_active_stripes);
+ if (atomic_read(&conf->preread_active_stripes) <
+ IO_THRESHOLD)
+ md_wakeup_thread(conf->mddev->thread);
+ }
+ }
+}
+
+static void handle_parity_checks5(raid5_conf_t *conf, struct stripe_head *sh,
+ struct stripe_head_state *s, int disks)
+{
+ set_bit(STRIPE_HANDLE, &sh->state);
+ /* Take one of the following actions:
+ * 1/ start a check parity operation if (uptodate == disks)
+ * 2/ finish a check parity operation and act on the result
+ * 3/ skip to the writeback section if we previously
+ * initiated a recovery operation
+ */
+ if (s->failed == 0 &&
+ !test_bit(STRIPE_OP_MOD_REPAIR_PD, &sh->ops.pending)) {
+ if (!test_and_set_bit(STRIPE_OP_CHECK, &sh->ops.pending)) {
+ BUG_ON(s->uptodate != disks);
+ clear_bit(R5_UPTODATE, &sh->dev[sh->pd_idx].flags);
+ sh->ops.count++;
+ s->uptodate--;
+ } else if (
+ test_and_clear_bit(STRIPE_OP_CHECK, &sh->ops.complete)) {
+ clear_bit(STRIPE_OP_CHECK, &sh->ops.ack);
+ clear_bit(STRIPE_OP_CHECK, &sh->ops.pending);
+
+ if (sh->ops.zero_sum_result == 0)
+ /* parity is correct (on disc,
+ * not in buffer any more)
+ */
+ set_bit(STRIPE_INSYNC, &sh->state);
+ else {
+ conf->mddev->resync_mismatches +=
+ STRIPE_SECTORS;
+ if (test_bit(
+ MD_RECOVERY_CHECK, &conf->mddev->recovery))
+ /* don't try to repair!! */
+ set_bit(STRIPE_INSYNC, &sh->state);
+ else {
+ set_bit(STRIPE_OP_COMPUTE_BLK,
+ &sh->ops.pending);
+ set_bit(STRIPE_OP_MOD_REPAIR_PD,
+ &sh->ops.pending);
+ set_bit(R5_Wantcompute,
+ &sh->dev[sh->pd_idx].flags);
+ sh->ops.target = sh->pd_idx;
+ sh->ops.count++;
+ s->uptodate++;
+ }
+ }
+ }
+ }
+
+ /* check if we can clear a parity disk reconstruct */
+ if (test_bit(STRIPE_OP_COMPUTE_BLK, &sh->ops.complete) &&
+ test_bit(STRIPE_OP_MOD_REPAIR_PD, &sh->ops.pending)) {
+
+ clear_bit(STRIPE_OP_MOD_REPAIR_PD, &sh->ops.pending);
+ clear_bit(STRIPE_OP_COMPUTE_BLK, &sh->ops.complete);
+ clear_bit(STRIPE_OP_COMPUTE_BLK, &sh->ops.ack);
+ clear_bit(STRIPE_OP_COMPUTE_BLK, &sh->ops.pending);
+ }
+
+ /* Wait for check parity and compute block operations to complete
+ * before write-back
+ */
+ if (!test_bit(STRIPE_INSYNC, &sh->state) &&
+ !test_bit(STRIPE_OP_CHECK, &sh->ops.pending) &&
+ !test_bit(STRIPE_OP_COMPUTE_BLK, &sh->ops.pending)) {
+ struct r5dev *dev;
+ /* either failed parity check, or recovery is happening */
+ if (s->failed == 0)
+ s->failed_num = sh->pd_idx;
+ dev = &sh->dev[s->failed_num];
+ BUG_ON(!test_bit(R5_UPTODATE, &dev->flags));
+ BUG_ON(s->uptodate != disks);
+
+ set_bit(R5_LOCKED, &dev->flags);
+ set_bit(R5_Wantwrite, &dev->flags);
+ if (!test_and_set_bit(STRIPE_OP_IO, &sh->ops.pending))
+ sh->ops.count++;
+
+ clear_bit(STRIPE_DEGRADED, &sh->state);
+ s->locked++;
+ set_bit(STRIPE_INSYNC, &sh->state);
+ }
+}
+
+
+static void handle_parity_checks6(raid5_conf_t *conf, struct stripe_head *sh,
+ struct stripe_head_state *s,
+ struct r6_state *r6s, struct page *tmp_page,
+ int disks)
+{
+ int update_p = 0, update_q = 0;
+ struct r5dev *dev;
+ int pd_idx = sh->pd_idx;
+ int qd_idx = r6s->qd_idx;
+
+ set_bit(STRIPE_HANDLE, &sh->state);
+
+ BUG_ON(s->failed > 2);
+ BUG_ON(s->uptodate < disks);
+ /* Want to check and possibly repair P and Q.
+ * However there could be one 'failed' device, in which
+ * case we can only check one of them, possibly using the
+ * other to generate missing data
+ */
+
+ /* If !tmp_page, we cannot do the calculations,
+ * but as we have set STRIPE_HANDLE, we will soon be called
+ * by stripe_handle with a tmp_page - just wait until then.
+ */
+ if (tmp_page) {
+ if (s->failed == r6s->q_failed) {
+ /* The only possible failed device holds 'Q', so it
+ * makes sense to check P (If anything else were failed,
+ * we would have used P to recreate it).
+ */
+ compute_block_1(sh, pd_idx, 1);
+ if (!page_is_zero(sh->dev[pd_idx].page)) {
+ compute_block_1(sh, pd_idx, 0);
+ update_p = 1;
+ }
+ }
+ if (!r6s->q_failed && s->failed < 2) {
+ /* q is not failed, and we didn't use it to generate
+ * anything, so it makes sense to check it
+ */
+ memcpy(page_address(tmp_page),
+ page_address(sh->dev[qd_idx].page),
+ STRIPE_SIZE);
+ compute_parity6(sh, UPDATE_PARITY);
+ if (memcmp(page_address(tmp_page),
+ page_address(sh->dev[qd_idx].page),
+ STRIPE_SIZE) != 0) {
+ clear_bit(STRIPE_INSYNC, &sh->state);
+ update_q = 1;
+ }
+ }
+ if (update_p || update_q) {
+ conf->mddev->resync_mismatches += STRIPE_SECTORS;
+ if (test_bit(MD_RECOVERY_CHECK, &conf->mddev->recovery))
+ /* don't try to repair!! */
+ update_p = update_q = 0;
+ }
+
+ /* now write out any block on a failed drive,
+ * or P or Q if they need it
+ */
+
+ if (s->failed == 2) {
+ dev = &sh->dev[r6s->failed_num[1]];
+ s->locked++;
+ set_bit(R5_LOCKED, &dev->flags);
+ set_bit(R5_Wantwrite, &dev->flags);
+ }
+ if (s->failed >= 1) {
+ dev = &sh->dev[r6s->failed_num[0]];
+ s->locked++;
+ set_bit(R5_LOCKED, &dev->flags);
+ set_bit(R5_Wantwrite, &dev->flags);
+ }
+
+ if (update_p) {
+ dev = &sh->dev[pd_idx];
+ s->locked++;
+ set_bit(R5_LOCKED, &dev->flags);
+ set_bit(R5_Wantwrite, &dev->flags);
+ }
+ if (update_q) {
+ dev = &sh->dev[qd_idx];
+ s->locked++;
+ set_bit(R5_LOCKED, &dev->flags);
+ set_bit(R5_Wantwrite, &dev->flags);
+ }
+ clear_bit(STRIPE_DEGRADED, &sh->state);
+
+ set_bit(STRIPE_INSYNC, &sh->state);
+ }
+}
+
+static void handle_stripe_expansion(raid5_conf_t *conf, struct stripe_head *sh,
+ struct r6_state *r6s)
+{
+ int i;
+
+ /* We have read all the blocks in this stripe and now we need to
+ * copy some of them into a target stripe for expand.
+ */
+ struct dma_async_tx_descriptor *tx = NULL;
+ clear_bit(STRIPE_EXPAND_SOURCE, &sh->state);
+ for (i = 0; i < sh->disks; i++)
+ if (i != sh->pd_idx && (r6s && i != r6s->qd_idx)) {
+ int dd_idx, pd_idx, j;
+ struct stripe_head *sh2;
+
+ sector_t bn = compute_blocknr(sh, i);
+ sector_t s = raid5_compute_sector(bn, conf->raid_disks,
+ conf->raid_disks -
+ conf->max_degraded, &dd_idx,
+ &pd_idx, conf);
+ sh2 = get_active_stripe(conf, s, conf->raid_disks,
+ pd_idx, 1);
+ if (sh2 == NULL)
+ /* so far only the early blocks of this stripe
+ * have been requested. When later blocks
+ * get requested, we will try again
+ */
+ continue;
+ if (!test_bit(STRIPE_EXPANDING, &sh2->state) ||
+ test_bit(R5_Expanded, &sh2->dev[dd_idx].flags)) {
+ /* must have already done this block */
+ release_stripe(sh2);
+ continue;
+ }
+
+ /* place all the copies on one channel */
+ tx = async_memcpy(sh2->dev[dd_idx].page,
+ sh->dev[i].page, 0, 0, STRIPE_SIZE,
+ ASYNC_TX_DEP_ACK, tx, NULL, NULL);
+
+ set_bit(R5_Expanded, &sh2->dev[dd_idx].flags);
+ set_bit(R5_UPTODATE, &sh2->dev[dd_idx].flags);
+ for (j = 0; j < conf->raid_disks; j++)
+ if (j != sh2->pd_idx &&
+ (r6s && j != r6s->qd_idx) &&
+ !test_bit(R5_Expanded, &sh2->dev[j].flags))
+ break;
+ if (j == conf->raid_disks) {
+ set_bit(STRIPE_EXPAND_READY, &sh2->state);
+ set_bit(STRIPE_HANDLE, &sh2->state);
+ }
+ release_stripe(sh2);
+
+ /* done submitting copies, wait for them to complete */
+ if (i + 1 >= sh->disks) {
+ async_tx_ack(tx);
+ dma_wait_for_async_tx(tx);
+ }
+ }
+}
/*
* handle_stripe - do things to a stripe.
@@ -1339,81 +2603,70 @@ static int stripe_to_pdidx(sector_t stripe, raid5_conf_t *conf, int disks)
* schedule a write of some buffers
* return confirmation of parity correctness
*
- * Parity calculations are done inside the stripe lock
* buffers are taken off read_list or write_list, and bh_cache buffers
* get BH_Lock set before the stripe lock is released.
*
*/
-
+
static void handle_stripe5(struct stripe_head *sh)
{
raid5_conf_t *conf = sh->raid_conf;
- int disks = sh->disks;
- struct bio *return_bi= NULL;
- struct bio *bi;
- int i;
- int syncing, expanding, expanded;
- int locked=0, uptodate=0, to_read=0, to_write=0, failed=0, written=0;
- int non_overwrite = 0;
- int failed_num=0;
+ int disks = sh->disks, i;
+ struct bio *return_bi = NULL;
+ struct stripe_head_state s;
struct r5dev *dev;
+ unsigned long pending = 0;
- PRINTK("handling stripe %llu, cnt=%d, pd_idx=%d\n",
- (unsigned long long)sh->sector, atomic_read(&sh->count),
- sh->pd_idx);
+ memset(&s, 0, sizeof(s));
+ pr_debug("handling stripe %llu, state=%#lx cnt=%d, pd_idx=%d "
+ "ops=%lx:%lx:%lx\n", (unsigned long long)sh->sector, sh->state,
+ atomic_read(&sh->count), sh->pd_idx,
+ sh->ops.pending, sh->ops.ack, sh->ops.complete);
spin_lock(&sh->lock);
clear_bit(STRIPE_HANDLE, &sh->state);
clear_bit(STRIPE_DELAYED, &sh->state);
- syncing = test_bit(STRIPE_SYNCING, &sh->state);
- expanding = test_bit(STRIPE_EXPAND_SOURCE, &sh->state);
- expanded = test_bit(STRIPE_EXPAND_READY, &sh->state);
+ s.syncing = test_bit(STRIPE_SYNCING, &sh->state);
+ s.expanding = test_bit(STRIPE_EXPAND_SOURCE, &sh->state);
+ s.expanded = test_bit(STRIPE_EXPAND_READY, &sh->state);
/* Now to look around and see what can be done */
rcu_read_lock();
for (i=disks; i--; ) {
mdk_rdev_t *rdev;
- dev = &sh->dev[i];
+ struct r5dev *dev = &sh->dev[i];
clear_bit(R5_Insync, &dev->flags);
- PRINTK("check %d: state 0x%lx read %p write %p written %p\n",
- i, dev->flags, dev->toread, dev->towrite, dev->written);
- /* maybe we can reply to a read */
- if (test_bit(R5_UPTODATE, &dev->flags) && dev->toread) {
- struct bio *rbi, *rbi2;
- PRINTK("Return read for disc %d\n", i);
- spin_lock_irq(&conf->device_lock);
- rbi = dev->toread;
- dev->toread = NULL;
- if (test_and_clear_bit(R5_Overlap, &dev->flags))
- wake_up(&conf->wait_for_overlap);
- spin_unlock_irq(&conf->device_lock);
- while (rbi && rbi->bi_sector < dev->sector + STRIPE_SECTORS) {
- copy_data(0, rbi, dev->page, dev->sector);
- rbi2 = r5_next_bio(rbi, dev->sector);
- spin_lock_irq(&conf->device_lock);
- if (--rbi->bi_phys_segments == 0) {
- rbi->bi_next = return_bi;
- return_bi = rbi;
- }
- spin_unlock_irq(&conf->device_lock);
- rbi = rbi2;
- }
- }
+ pr_debug("check %d: state 0x%lx toread %p read %p write %p "
+ "written %p\n", i, dev->flags, dev->toread, dev->read,
+ dev->towrite, dev->written);
- /* now count some things */
- if (test_bit(R5_LOCKED, &dev->flags)) locked++;
- if (test_bit(R5_UPTODATE, &dev->flags)) uptodate++;
+ /* maybe we can request a biofill operation
+ *
+ * new wantfill requests are only permitted while
+ * STRIPE_OP_BIOFILL is clear
+ */
+ if (test_bit(R5_UPTODATE, &dev->flags) && dev->toread &&
+ !test_bit(STRIPE_OP_BIOFILL, &sh->ops.pending))
+ set_bit(R5_Wantfill, &dev->flags);
-
- if (dev->toread) to_read++;
+ /* now count some things */
+ if (test_bit(R5_LOCKED, &dev->flags)) s.locked++;
+ if (test_bit(R5_UPTODATE, &dev->flags)) s.uptodate++;
+ if (test_bit(R5_Wantcompute, &dev->flags)) s.compute++;
+
+ if (test_bit(R5_Wantfill, &dev->flags))
+ s.to_fill++;
+ else if (dev->toread)
+ s.to_read++;
if (dev->towrite) {
- to_write++;
+ s.to_write++;
if (!test_bit(R5_OVERWRITE, &dev->flags))
- non_overwrite++;
+ s.non_overwrite++;
}
- if (dev->written) written++;
+ if (dev->written)
+ s.written++;
rdev = rcu_dereference(conf->disks[i].rdev);
if (!rdev || !test_bit(In_sync, &rdev->flags)) {
/* The ReadError flag will just be confusing now */
@@ -1422,306 +2675,131 @@ static void handle_stripe5(struct stripe_head *sh)
}
if (!rdev || !test_bit(In_sync, &rdev->flags)
|| test_bit(R5_ReadError, &dev->flags)) {
- failed++;
- failed_num = i;
+ s.failed++;
+ s.failed_num = i;
} else
set_bit(R5_Insync, &dev->flags);
}
rcu_read_unlock();
- PRINTK("locked=%d uptodate=%d to_read=%d"
+
+ if (s.to_fill && !test_and_set_bit(STRIPE_OP_BIOFILL, &sh->ops.pending))
+ sh->ops.count++;
+
+ pr_debug("locked=%d uptodate=%d to_read=%d"
" to_write=%d failed=%d failed_num=%d\n",
- locked, uptodate, to_read, to_write, failed, failed_num);
+ s.locked, s.uptodate, s.to_read, s.to_write,
+ s.failed, s.failed_num);
/* check if the array has lost two devices and, if so, some requests might
* need to be failed
*/
- if (failed > 1 && to_read+to_write+written) {
- for (i=disks; i--; ) {
- int bitmap_end = 0;
-
- if (test_bit(R5_ReadError, &sh->dev[i].flags)) {
- mdk_rdev_t *rdev;
- rcu_read_lock();
- rdev = rcu_dereference(conf->disks[i].rdev);
- if (rdev && test_bit(In_sync, &rdev->flags))
- /* multiple read failures in one stripe */
- md_error(conf->mddev, rdev);
- rcu_read_unlock();
- }
-
- spin_lock_irq(&conf->device_lock);
- /* fail all writes first */
- bi = sh->dev[i].towrite;
- sh->dev[i].towrite = NULL;
- if (bi) { to_write--; bitmap_end = 1; }
-
- if (test_and_clear_bit(R5_Overlap, &sh->dev[i].flags))
- wake_up(&conf->wait_for_overlap);
-
- while (bi && bi->bi_sector < sh->dev[i].sector + STRIPE_SECTORS){
- struct bio *nextbi = r5_next_bio(bi, sh->dev[i].sector);
- clear_bit(BIO_UPTODATE, &bi->bi_flags);
- if (--bi->bi_phys_segments == 0) {
- md_write_end(conf->mddev);
- bi->bi_next = return_bi;
- return_bi = bi;
- }
- bi = nextbi;
- }
- /* and fail all 'written' */
- bi = sh->dev[i].written;
- sh->dev[i].written = NULL;
- if (bi) bitmap_end = 1;
- while (bi && bi->bi_sector < sh->dev[i].sector + STRIPE_SECTORS) {
- struct bio *bi2 = r5_next_bio(bi, sh->dev[i].sector);
- clear_bit(BIO_UPTODATE, &bi->bi_flags);
- if (--bi->bi_phys_segments == 0) {
- md_write_end(conf->mddev);
- bi->bi_next = return_bi;
- return_bi = bi;
- }
- bi = bi2;
- }
-
- /* fail any reads if this device is non-operational */
- if (!test_bit(R5_Insync, &sh->dev[i].flags) ||
- test_bit(R5_ReadError, &sh->dev[i].flags)) {
- bi = sh->dev[i].toread;
- sh->dev[i].toread = NULL;
- if (test_and_clear_bit(R5_Overlap, &sh->dev[i].flags))
- wake_up(&conf->wait_for_overlap);
- if (bi) to_read--;
- while (bi && bi->bi_sector < sh->dev[i].sector + STRIPE_SECTORS){
- struct bio *nextbi = r5_next_bio(bi, sh->dev[i].sector);
- clear_bit(BIO_UPTODATE, &bi->bi_flags);
- if (--bi->bi_phys_segments == 0) {
- bi->bi_next = return_bi;
- return_bi = bi;
- }
- bi = nextbi;
- }
- }
- spin_unlock_irq(&conf->device_lock);
- if (bitmap_end)
- bitmap_endwrite(conf->mddev->bitmap, sh->sector,
- STRIPE_SECTORS, 0, 0);
- }
- }
- if (failed > 1 && syncing) {
+ if (s.failed > 1 && s.to_read+s.to_write+s.written)
+ handle_requests_to_failed_array(conf, sh, &s, disks,
+ &return_bi);
+ if (s.failed > 1 && s.syncing) {
md_done_sync(conf->mddev, STRIPE_SECTORS,0);
clear_bit(STRIPE_SYNCING, &sh->state);
- syncing = 0;
+ s.syncing = 0;
}
/* might be able to return some write requests if the parity block
* is safe, or on a failed drive
*/
dev = &sh->dev[sh->pd_idx];
- if ( written &&
- ( (test_bit(R5_Insync, &dev->flags) && !test_bit(R5_LOCKED, &dev->flags) &&
- test_bit(R5_UPTODATE, &dev->flags))
- || (failed == 1 && failed_num == sh->pd_idx))
- ) {
- /* any written block on an uptodate or failed drive can be returned.
- * Note that if we 'wrote' to a failed drive, it will be UPTODATE, but
- * never LOCKED, so we don't need to test 'failed' directly.
- */
- for (i=disks; i--; )
- if (sh->dev[i].written) {
- dev = &sh->dev[i];
- if (!test_bit(R5_LOCKED, &dev->flags) &&
- test_bit(R5_UPTODATE, &dev->flags) ) {
- /* We can return any write requests */
- struct bio *wbi, *wbi2;
- int bitmap_end = 0;
- PRINTK("Return write for disc %d\n", i);
- spin_lock_irq(&conf->device_lock);
- wbi = dev->written;
- dev->written = NULL;
- while (wbi && wbi->bi_sector < dev->sector + STRIPE_SECTORS) {
- wbi2 = r5_next_bio(wbi, dev->sector);
- if (--wbi->bi_phys_segments == 0) {
- md_write_end(conf->mddev);
- wbi->bi_next = return_bi;
- return_bi = wbi;
- }
- wbi = wbi2;
- }
- if (dev->towrite == NULL)
- bitmap_end = 1;
- spin_unlock_irq(&conf->device_lock);
- if (bitmap_end)
- bitmap_endwrite(conf->mddev->bitmap, sh->sector,
- STRIPE_SECTORS,
- !test_bit(STRIPE_DEGRADED, &sh->state), 0);
- }
- }
- }
+ if ( s.written &&
+ ((test_bit(R5_Insync, &dev->flags) &&
+ !test_bit(R5_LOCKED, &dev->flags) &&
+ test_bit(R5_UPTODATE, &dev->flags)) ||
+ (s.failed == 1 && s.failed_num == sh->pd_idx)))
+ handle_completed_write_requests(conf, sh, disks, &return_bi);
/* Now we might consider reading some blocks, either to check/generate
* parity, or to satisfy requests
* or to load a block that is being partially written.
*/
- if (to_read || non_overwrite || (syncing && (uptodate < disks)) || expanding) {
- for (i=disks; i--;) {
- dev = &sh->dev[i];
- if (!test_bit(R5_LOCKED, &dev->flags) && !test_bit(R5_UPTODATE, &dev->flags) &&
- (dev->toread ||
- (dev->towrite && !test_bit(R5_OVERWRITE, &dev->flags)) ||
- syncing ||
- expanding ||
- (failed && (sh->dev[failed_num].toread ||
- (sh->dev[failed_num].towrite && !test_bit(R5_OVERWRITE, &sh->dev[failed_num].flags))))
- )
- ) {
- /* we would like to get this block, possibly
- * by computing it, but we might not be able to
- */
- if (uptodate == disks-1) {
- PRINTK("Computing block %d\n", i);
- compute_block(sh, i);
- uptodate++;
- } else if (test_bit(R5_Insync, &dev->flags)) {
- set_bit(R5_LOCKED, &dev->flags);
- set_bit(R5_Wantread, &dev->flags);
- locked++;
- PRINTK("Reading block %d (sync=%d)\n",
- i, syncing);
- }
- }
- }
- set_bit(STRIPE_HANDLE, &sh->state);
+ if (s.to_read || s.non_overwrite ||
+ (s.syncing && (s.uptodate + s.compute < disks)) || s.expanding ||
+ test_bit(STRIPE_OP_COMPUTE_BLK, &sh->ops.pending))
+ handle_issuing_new_read_requests5(sh, &s, disks);
+
+ /* Now we check to see if any write operations have recently
+ * completed
+ */
+
+ /* leave prexor set until postxor is done, allows us to distinguish
+ * a rmw from a rcw during biodrain
+ */
+ if (test_bit(STRIPE_OP_PREXOR, &sh->ops.complete) &&
+ test_bit(STRIPE_OP_POSTXOR, &sh->ops.complete)) {
+
+ clear_bit(STRIPE_OP_PREXOR, &sh->ops.complete);
+ clear_bit(STRIPE_OP_PREXOR, &sh->ops.ack);
+ clear_bit(STRIPE_OP_PREXOR, &sh->ops.pending);
+
+ for (i = disks; i--; )
+ clear_bit(R5_Wantprexor, &sh->dev[i].flags);
}
- /* now to consider writing and what else, if anything should be read */
- if (to_write) {
- int rmw=0, rcw=0;
- for (i=disks ; i--;) {
- /* would I have to read this buffer for read_modify_write */
+ /* if only POSTXOR is set then this is an 'expand' postxor */
+ if (test_bit(STRIPE_OP_BIODRAIN, &sh->ops.complete) &&
+ test_bit(STRIPE_OP_POSTXOR, &sh->ops.complete)) {
+
+ clear_bit(STRIPE_OP_BIODRAIN, &sh->ops.complete);
+ clear_bit(STRIPE_OP_BIODRAIN, &sh->ops.ack);
+ clear_bit(STRIPE_OP_BIODRAIN, &sh->ops.pending);
+
+ clear_bit(STRIPE_OP_POSTXOR, &sh->ops.complete);
+ clear_bit(STRIPE_OP_POSTXOR, &sh->ops.ack);
+ clear_bit(STRIPE_OP_POSTXOR, &sh->ops.pending);
+
+ /* All the 'written' buffers and the parity block are ready to
+ * be written back to disk
+ */
+ BUG_ON(!test_bit(R5_UPTODATE, &sh->dev[sh->pd_idx].flags));
+ for (i = disks; i--; ) {
dev = &sh->dev[i];
- if ((dev->towrite || i == sh->pd_idx) &&
- (!test_bit(R5_LOCKED, &dev->flags)
- ) &&
- !test_bit(R5_UPTODATE, &dev->flags)) {
- if (test_bit(R5_Insync, &dev->flags)
-/* && !(!mddev->insync && i == sh->pd_idx) */
- )
- rmw++;
- else rmw += 2*disks; /* cannot read it */
- }
- /* Would I have to read this buffer for reconstruct_write */
- if (!test_bit(R5_OVERWRITE, &dev->flags) && i != sh->pd_idx &&
- (!test_bit(R5_LOCKED, &dev->flags)
- ) &&
- !test_bit(R5_UPTODATE, &dev->flags)) {
- if (test_bit(R5_Insync, &dev->flags)) rcw++;
- else rcw += 2*disks;
+ if (test_bit(R5_LOCKED, &dev->flags) &&
+ (i == sh->pd_idx || dev->written)) {
+ pr_debug("Writing block %d\n", i);
+ set_bit(R5_Wantwrite, &dev->flags);
+ if (!test_and_set_bit(
+ STRIPE_OP_IO, &sh->ops.pending))
+ sh->ops.count++;
+ if (!test_bit(R5_Insync, &dev->flags) ||
+ (i == sh->pd_idx && s.failed == 0))
+ set_bit(STRIPE_INSYNC, &sh->state);
}
}
- PRINTK("for sector %llu, rmw=%d rcw=%d\n",
- (unsigned long long)sh->sector, rmw, rcw);
- set_bit(STRIPE_HANDLE, &sh->state);
- if (rmw < rcw && rmw > 0)
- /* prefer read-modify-write, but need to get some data */
- for (i=disks; i--;) {
- dev = &sh->dev[i];
- if ((dev->towrite || i == sh->pd_idx) &&
- !test_bit(R5_LOCKED, &dev->flags) && !test_bit(R5_UPTODATE, &dev->flags) &&
- test_bit(R5_Insync, &dev->flags)) {
- if (test_bit(STRIPE_PREREAD_ACTIVE, &sh->state))
- {
- PRINTK("Read_old block %d for r-m-w\n", i);
- set_bit(R5_LOCKED, &dev->flags);
- set_bit(R5_Wantread, &dev->flags);
- locked++;
- } else {
- set_bit(STRIPE_DELAYED, &sh->state);
- set_bit(STRIPE_HANDLE, &sh->state);
- }
- }
- }
- if (rcw <= rmw && rcw > 0)
- /* want reconstruct write, but need to get some data */
- for (i=disks; i--;) {
- dev = &sh->dev[i];
- if (!test_bit(R5_OVERWRITE, &dev->flags) && i != sh->pd_idx &&
- !test_bit(R5_LOCKED, &dev->flags) && !test_bit(R5_UPTODATE, &dev->flags) &&
- test_bit(R5_Insync, &dev->flags)) {
- if (test_bit(STRIPE_PREREAD_ACTIVE, &sh->state))
- {
- PRINTK("Read_old block %d for Reconstruct\n", i);
- set_bit(R5_LOCKED, &dev->flags);
- set_bit(R5_Wantread, &dev->flags);
- locked++;
- } else {
- set_bit(STRIPE_DELAYED, &sh->state);
- set_bit(STRIPE_HANDLE, &sh->state);
- }
- }
- }
- /* now if nothing is locked, and if we have enough data, we can start a write request */
- if (locked == 0 && (rcw == 0 ||rmw == 0) &&
- !test_bit(STRIPE_BIT_DELAY, &sh->state)) {
- PRINTK("Computing parity...\n");
- compute_parity5(sh, rcw==0 ? RECONSTRUCT_WRITE : READ_MODIFY_WRITE);
- /* now every locked buffer is ready to be written */
- for (i=disks; i--;)
- if (test_bit(R5_LOCKED, &sh->dev[i].flags)) {
- PRINTK("Writing block %d\n", i);
- locked++;
- set_bit(R5_Wantwrite, &sh->dev[i].flags);
- if (!test_bit(R5_Insync, &sh->dev[i].flags)
- || (i==sh->pd_idx && failed == 0))
- set_bit(STRIPE_INSYNC, &sh->state);
- }
- if (test_and_clear_bit(STRIPE_PREREAD_ACTIVE, &sh->state)) {
- atomic_dec(&conf->preread_active_stripes);
- if (atomic_read(&conf->preread_active_stripes) < IO_THRESHOLD)
- md_wakeup_thread(conf->mddev->thread);
- }
+ if (test_and_clear_bit(STRIPE_PREREAD_ACTIVE, &sh->state)) {
+ atomic_dec(&conf->preread_active_stripes);
+ if (atomic_read(&conf->preread_active_stripes) <
+ IO_THRESHOLD)
+ md_wakeup_thread(conf->mddev->thread);
}
}
- /* maybe we need to check and possibly fix the parity for this stripe
- * Any reads will already have been scheduled, so we just see if enough data
- * is available
+ /* Now to consider new write requests and what else, if anything
+ * should be read. We do not handle new writes when:
+ * 1/ A 'write' operation (copy+xor) is already in flight.
+ * 2/ A 'check' operation is in flight, as it may clobber the parity
+ * block.
*/
- if (syncing && locked == 0 &&
- !test_bit(STRIPE_INSYNC, &sh->state)) {
- set_bit(STRIPE_HANDLE, &sh->state);
- if (failed == 0) {
- BUG_ON(uptodate != disks);
- compute_parity5(sh, CHECK_PARITY);
- uptodate--;
- if (page_is_zero(sh->dev[sh->pd_idx].page)) {
- /* parity is correct (on disc, not in buffer any more) */
- set_bit(STRIPE_INSYNC, &sh->state);
- } else {
- conf->mddev->resync_mismatches += STRIPE_SECTORS;
- if (test_bit(MD_RECOVERY_CHECK, &conf->mddev->recovery))
- /* don't try to repair!! */
- set_bit(STRIPE_INSYNC, &sh->state);
- else {
- compute_block(sh, sh->pd_idx);
- uptodate++;
- }
- }
- }
- if (!test_bit(STRIPE_INSYNC, &sh->state)) {
- /* either failed parity check, or recovery is happening */
- if (failed==0)
- failed_num = sh->pd_idx;
- dev = &sh->dev[failed_num];
- BUG_ON(!test_bit(R5_UPTODATE, &dev->flags));
- BUG_ON(uptodate != disks);
+ if (s.to_write && !test_bit(STRIPE_OP_POSTXOR, &sh->ops.pending) &&
+ !test_bit(STRIPE_OP_CHECK, &sh->ops.pending))
+ handle_issuing_new_write_requests5(conf, sh, &s, disks);
- set_bit(R5_LOCKED, &dev->flags);
- set_bit(R5_Wantwrite, &dev->flags);
- clear_bit(STRIPE_DEGRADED, &sh->state);
- locked++;
- set_bit(STRIPE_INSYNC, &sh->state);
- }
- }
- if (syncing && locked == 0 && test_bit(STRIPE_INSYNC, &sh->state)) {
+ /* maybe we need to check and possibly fix the parity for this stripe
+ * Any reads will already have been scheduled, so we just see if enough
+ * data is available. The parity check is held off while parity
+ * dependent operations are in flight.
+ */
+ if ((s.syncing && s.locked == 0 &&
+ !test_bit(STRIPE_OP_COMPUTE_BLK, &sh->ops.pending) &&
+ !test_bit(STRIPE_INSYNC, &sh->state)) ||
+ test_bit(STRIPE_OP_CHECK, &sh->ops.pending) ||
+ test_bit(STRIPE_OP_MOD_REPAIR_PD, &sh->ops.pending))
+ handle_parity_checks5(conf, sh, &s, disks);
+
+ if (s.syncing && s.locked == 0 && test_bit(STRIPE_INSYNC, &sh->state)) {
md_done_sync(conf->mddev, STRIPE_SECTORS,1);
clear_bit(STRIPE_SYNCING, &sh->state);
}
@@ -1729,186 +2807,102 @@ static void handle_stripe5(struct stripe_head *sh)
/* If the failed drive is just a ReadError, then we might need to progress
* the repair/check process
*/
- if (failed == 1 && ! conf->mddev->ro &&
- test_bit(R5_ReadError, &sh->dev[failed_num].flags)
- && !test_bit(R5_LOCKED, &sh->dev[failed_num].flags)
- && test_bit(R5_UPTODATE, &sh->dev[failed_num].flags)
+ if (s.failed == 1 && !conf->mddev->ro &&
+ test_bit(R5_ReadError, &sh->dev[s.failed_num].flags)
+ && !test_bit(R5_LOCKED, &sh->dev[s.failed_num].flags)
+ && test_bit(R5_UPTODATE, &sh->dev[s.failed_num].flags)
) {
- dev = &sh->dev[failed_num];
+ dev = &sh->dev[s.failed_num];
if (!test_bit(R5_ReWrite, &dev->flags)) {
set_bit(R5_Wantwrite, &dev->flags);
+ if (!test_and_set_bit(STRIPE_OP_IO, &sh->ops.pending))
+ sh->ops.count++;
set_bit(R5_ReWrite, &dev->flags);
set_bit(R5_LOCKED, &dev->flags);
- locked++;
+ s.locked++;
} else {
/* let's read it back */
set_bit(R5_Wantread, &dev->flags);
+ if (!test_and_set_bit(STRIPE_OP_IO, &sh->ops.pending))
+ sh->ops.count++;
set_bit(R5_LOCKED, &dev->flags);
- locked++;
+ s.locked++;
}
}
- if (expanded && test_bit(STRIPE_EXPANDING, &sh->state)) {
- /* Need to write out all blocks after computing parity */
- sh->disks = conf->raid_disks;
- sh->pd_idx = stripe_to_pdidx(sh->sector, conf, conf->raid_disks);
- compute_parity5(sh, RECONSTRUCT_WRITE);
- for (i= conf->raid_disks; i--;) {
- set_bit(R5_LOCKED, &sh->dev[i].flags);
- locked++;
+ /* Finish postxor operations initiated by the expansion
+ * process
+ */
+ if (test_bit(STRIPE_OP_POSTXOR, &sh->ops.complete) &&
+ !test_bit(STRIPE_OP_BIODRAIN, &sh->ops.pending)) {
+
+ clear_bit(STRIPE_EXPANDING, &sh->state);
+
+ clear_bit(STRIPE_OP_POSTXOR, &sh->ops.pending);
+ clear_bit(STRIPE_OP_POSTXOR, &sh->ops.ack);
+ clear_bit(STRIPE_OP_POSTXOR, &sh->ops.complete);
+
+ for (i = conf->raid_disks; i--; ) {
set_bit(R5_Wantwrite, &sh->dev[i].flags);
+ if (!test_and_set_bit(STRIPE_OP_IO, &sh->ops.pending))
+ sh->ops.count++;
}
- clear_bit(STRIPE_EXPANDING, &sh->state);
- } else if (expanded) {
+ }
+
+ if (s.expanded && test_bit(STRIPE_EXPANDING, &sh->state) &&
+ !test_bit(STRIPE_OP_POSTXOR, &sh->ops.pending)) {
+ /* Need to write out all blocks after computing parity */
+ sh->disks = conf->raid_disks;
+ sh->pd_idx = stripe_to_pdidx(sh->sector, conf,
+ conf->raid_disks);
+ s.locked += handle_write_operations5(sh, 0, 1);
+ } else if (s.expanded &&
+ !test_bit(STRIPE_OP_POSTXOR, &sh->ops.pending)) {
clear_bit(STRIPE_EXPAND_READY, &sh->state);
atomic_dec(&conf->reshape_stripes);
wake_up(&conf->wait_for_overlap);
md_done_sync(conf->mddev, STRIPE_SECTORS, 1);
}
- if (expanding && locked == 0) {
- /* We have read all the blocks in this stripe and now we need to
- * copy some of them into a target stripe for expand.
- */
- clear_bit(STRIPE_EXPAND_SOURCE, &sh->state);
- for (i=0; i< sh->disks; i++)
- if (i != sh->pd_idx) {
- int dd_idx, pd_idx, j;
- struct stripe_head *sh2;
-
- sector_t bn = compute_blocknr(sh, i);
- sector_t s = raid5_compute_sector(bn, conf->raid_disks,
- conf->raid_disks-1,
- &dd_idx, &pd_idx, conf);
- sh2 = get_active_stripe(conf, s, conf->raid_disks, pd_idx, 1);
- if (sh2 == NULL)
- /* so far only the early blocks of this stripe
- * have been requested. When later blocks
- * get requested, we will try again
- */
- continue;
- if(!test_bit(STRIPE_EXPANDING, &sh2->state) ||
- test_bit(R5_Expanded, &sh2->dev[dd_idx].flags)) {
- /* must have already done this block */
- release_stripe(sh2);
- continue;
- }
- memcpy(page_address(sh2->dev[dd_idx].page),
- page_address(sh->dev[i].page),
- STRIPE_SIZE);
- set_bit(R5_Expanded, &sh2->dev[dd_idx].flags);
- set_bit(R5_UPTODATE, &sh2->dev[dd_idx].flags);
- for (j=0; j<conf->raid_disks; j++)
- if (j != sh2->pd_idx &&
- !test_bit(R5_Expanded, &sh2->dev[j].flags))
- break;
- if (j == conf->raid_disks) {
- set_bit(STRIPE_EXPAND_READY, &sh2->state);
- set_bit(STRIPE_HANDLE, &sh2->state);
- }
- release_stripe(sh2);
- }
- }
+ if (s.expanding && s.locked == 0)
+ handle_stripe_expansion(conf, sh, NULL);
+
+ if (sh->ops.count)
+ pending = get_stripe_work(sh);
spin_unlock(&sh->lock);
- while ((bi=return_bi)) {
- int bytes = bi->bi_size;
+ if (pending)
+ raid5_run_ops(sh, pending);
- return_bi = bi->bi_next;
- bi->bi_next = NULL;
- bi->bi_size = 0;
- bi->bi_end_io(bi, bytes,
- test_bit(BIO_UPTODATE, &bi->bi_flags)
- ? 0 : -EIO);
- }
- for (i=disks; i-- ;) {
- int rw;
- struct bio *bi;
- mdk_rdev_t *rdev;
- if (test_and_clear_bit(R5_Wantwrite, &sh->dev[i].flags))
- rw = WRITE;
- else if (test_and_clear_bit(R5_Wantread, &sh->dev[i].flags))
- rw = READ;
- else
- continue;
-
- bi = &sh->dev[i].req;
-
- bi->bi_rw = rw;
- if (rw == WRITE)
- bi->bi_end_io = raid5_end_write_request;
- else
- bi->bi_end_io = raid5_end_read_request;
-
- rcu_read_lock();
- rdev = rcu_dereference(conf->disks[i].rdev);
- if (rdev && test_bit(Faulty, &rdev->flags))
- rdev = NULL;
- if (rdev)
- atomic_inc(&rdev->nr_pending);
- rcu_read_unlock();
-
- if (rdev) {
- if (syncing || expanding || expanded)
- md_sync_acct(rdev->bdev, STRIPE_SECTORS);
+ return_io(return_bi);
- bi->bi_bdev = rdev->bdev;
- PRINTK("for %llu schedule op %ld on disc %d\n",
- (unsigned long long)sh->sector, bi->bi_rw, i);
- atomic_inc(&sh->count);
- bi->bi_sector = sh->sector + rdev->data_offset;
- bi->bi_flags = 1 << BIO_UPTODATE;
- bi->bi_vcnt = 1;
- bi->bi_max_vecs = 1;
- bi->bi_idx = 0;
- bi->bi_io_vec = &sh->dev[i].vec;
- bi->bi_io_vec[0].bv_len = STRIPE_SIZE;
- bi->bi_io_vec[0].bv_offset = 0;
- bi->bi_size = STRIPE_SIZE;
- bi->bi_next = NULL;
- if (rw == WRITE &&
- test_bit(R5_ReWrite, &sh->dev[i].flags))
- atomic_add(STRIPE_SECTORS, &rdev->corrected_errors);
- generic_make_request(bi);
- } else {
- if (rw == WRITE)
- set_bit(STRIPE_DEGRADED, &sh->state);
- PRINTK("skip op %ld on disc %d for sector %llu\n",
- bi->bi_rw, i, (unsigned long long)sh->sector);
- clear_bit(R5_LOCKED, &sh->dev[i].flags);
- set_bit(STRIPE_HANDLE, &sh->state);
- }
- }
}
static void handle_stripe6(struct stripe_head *sh, struct page *tmp_page)
{
raid6_conf_t *conf = sh->raid_conf;
int disks = sh->disks;
- struct bio *return_bi= NULL;
- struct bio *bi;
- int i;
- int syncing, expanding, expanded;
- int locked=0, uptodate=0, to_read=0, to_write=0, failed=0, written=0;
- int non_overwrite = 0;
- int failed_num[2] = {0, 0};
+ struct bio *return_bi = NULL;
+ int i, pd_idx = sh->pd_idx;
+ struct stripe_head_state s;
+ struct r6_state r6s;
struct r5dev *dev, *pdev, *qdev;
- int pd_idx = sh->pd_idx;
- int qd_idx = raid6_next_disk(pd_idx, disks);
- int p_failed, q_failed;
- PRINTK("handling stripe %llu, state=%#lx cnt=%d, pd_idx=%d, qd_idx=%d\n",
- (unsigned long long)sh->sector, sh->state, atomic_read(&sh->count),
- pd_idx, qd_idx);
+ r6s.qd_idx = raid6_next_disk(pd_idx, disks);
+ pr_debug("handling stripe %llu, state=%#lx cnt=%d, "
+ "pd_idx=%d, qd_idx=%d\n",
+ (unsigned long long)sh->sector, sh->state,
+ atomic_read(&sh->count), pd_idx, r6s.qd_idx);
+ memset(&s, 0, sizeof(s));
spin_lock(&sh->lock);
clear_bit(STRIPE_HANDLE, &sh->state);
clear_bit(STRIPE_DELAYED, &sh->state);
- syncing = test_bit(STRIPE_SYNCING, &sh->state);
- expanding = test_bit(STRIPE_EXPAND_SOURCE, &sh->state);
- expanded = test_bit(STRIPE_EXPAND_READY, &sh->state);
+ s.syncing = test_bit(STRIPE_SYNCING, &sh->state);
+ s.expanding = test_bit(STRIPE_EXPAND_SOURCE, &sh->state);
+ s.expanded = test_bit(STRIPE_EXPAND_READY, &sh->state);
/* Now to look around and see what can be done */
rcu_read_lock();
@@ -1917,12 +2911,12 @@ static void handle_stripe6(struct stripe_head *sh, struct page *tmp_page)
dev = &sh->dev[i];
clear_bit(R5_Insync, &dev->flags);
- PRINTK("check %d: state 0x%lx read %p write %p written %p\n",
+ pr_debug("check %d: state 0x%lx read %p write %p written %p\n",
i, dev->flags, dev->toread, dev->towrite, dev->written);
/* maybe we can reply to a read */
if (test_bit(R5_UPTODATE, &dev->flags) && dev->toread) {
struct bio *rbi, *rbi2;
- PRINTK("Return read for disc %d\n", i);
+ pr_debug("Return read for disc %d\n", i);
spin_lock_irq(&conf->device_lock);
rbi = dev->toread;
dev->toread = NULL;
@@ -1943,17 +2937,19 @@ static void handle_stripe6(struct stripe_head *sh, struct page *tmp_page)
}
/* now count some things */
- if (test_bit(R5_LOCKED, &dev->flags)) locked++;
- if (test_bit(R5_UPTODATE, &dev->flags)) uptodate++;
+ if (test_bit(R5_LOCKED, &dev->flags)) s.locked++;
+ if (test_bit(R5_UPTODATE, &dev->flags)) s.uptodate++;
- if (dev->toread) to_read++;
+ if (dev->toread)
+ s.to_read++;
if (dev->towrite) {
- to_write++;
+ s.to_write++;
if (!test_bit(R5_OVERWRITE, &dev->flags))
- non_overwrite++;
+ s.non_overwrite++;
}
- if (dev->written) written++;
+ if (dev->written)
+ s.written++;
rdev = rcu_dereference(conf->disks[i].rdev);
if (!rdev || !test_bit(In_sync, &rdev->flags)) {
/* The ReadError flag will just be confusing now */
@@ -1962,96 +2958,27 @@ static void handle_stripe6(struct stripe_head *sh, struct page *tmp_page)
}
if (!rdev || !test_bit(In_sync, &rdev->flags)
|| test_bit(R5_ReadError, &dev->flags)) {
- if ( failed < 2 )
- failed_num[failed] = i;
- failed++;
+ if (s.failed < 2)
+ r6s.failed_num[s.failed] = i;
+ s.failed++;
} else
set_bit(R5_Insync, &dev->flags);
}
rcu_read_unlock();
- PRINTK("locked=%d uptodate=%d to_read=%d"
+ pr_debug("locked=%d uptodate=%d to_read=%d"
" to_write=%d failed=%d failed_num=%d,%d\n",
- locked, uptodate, to_read, to_write, failed,
- failed_num[0], failed_num[1]);
- /* check if the array has lost >2 devices and, if so, some requests might
- * need to be failed
+ s.locked, s.uptodate, s.to_read, s.to_write, s.failed,
+ r6s.failed_num[0], r6s.failed_num[1]);
+ /* check if the array has lost >2 devices and, if so, some requests
+ * might need to be failed
*/
- if (failed > 2 && to_read+to_write+written) {
- for (i=disks; i--; ) {
- int bitmap_end = 0;
-
- if (test_bit(R5_ReadError, &sh->dev[i].flags)) {
- mdk_rdev_t *rdev;
- rcu_read_lock();
- rdev = rcu_dereference(conf->disks[i].rdev);
- if (rdev && test_bit(In_sync, &rdev->flags))
- /* multiple read failures in one stripe */
- md_error(conf->mddev, rdev);
- rcu_read_unlock();
- }
-
- spin_lock_irq(&conf->device_lock);
- /* fail all writes first */
- bi = sh->dev[i].towrite;
- sh->dev[i].towrite = NULL;
- if (bi) { to_write--; bitmap_end = 1; }
-
- if (test_and_clear_bit(R5_Overlap, &sh->dev[i].flags))
- wake_up(&conf->wait_for_overlap);
-
- while (bi && bi->bi_sector < sh->dev[i].sector + STRIPE_SECTORS){
- struct bio *nextbi = r5_next_bio(bi, sh->dev[i].sector);
- clear_bit(BIO_UPTODATE, &bi->bi_flags);
- if (--bi->bi_phys_segments == 0) {
- md_write_end(conf->mddev);
- bi->bi_next = return_bi;
- return_bi = bi;
- }
- bi = nextbi;
- }
- /* and fail all 'written' */
- bi = sh->dev[i].written;
- sh->dev[i].written = NULL;
- if (bi) bitmap_end = 1;
- while (bi && bi->bi_sector < sh->dev[i].sector + STRIPE_SECTORS) {
- struct bio *bi2 = r5_next_bio(bi, sh->dev[i].sector);
- clear_bit(BIO_UPTODATE, &bi->bi_flags);
- if (--bi->bi_phys_segments == 0) {
- md_write_end(conf->mddev);
- bi->bi_next = return_bi;
- return_bi = bi;
- }
- bi = bi2;
- }
-
- /* fail any reads if this device is non-operational */
- if (!test_bit(R5_Insync, &sh->dev[i].flags) ||
- test_bit(R5_ReadError, &sh->dev[i].flags)) {
- bi = sh->dev[i].toread;
- sh->dev[i].toread = NULL;
- if (test_and_clear_bit(R5_Overlap, &sh->dev[i].flags))
- wake_up(&conf->wait_for_overlap);
- if (bi) to_read--;
- while (bi && bi->bi_sector < sh->dev[i].sector + STRIPE_SECTORS){
- struct bio *nextbi = r5_next_bio(bi, sh->dev[i].sector);
- clear_bit(BIO_UPTODATE, &bi->bi_flags);
- if (--bi->bi_phys_segments == 0) {
- bi->bi_next = return_bi;
- return_bi = bi;
- }
- bi = nextbi;
- }
- }
- spin_unlock_irq(&conf->device_lock);
- if (bitmap_end)
- bitmap_endwrite(conf->mddev->bitmap, sh->sector,
- STRIPE_SECTORS, 0, 0);
- }
- }
- if (failed > 2 && syncing) {
+ if (s.failed > 2 && s.to_read+s.to_write+s.written)
+ handle_requests_to_failed_array(conf, sh, &s, disks,
+ &return_bi);
+ if (s.failed > 2 && s.syncing) {
md_done_sync(conf->mddev, STRIPE_SECTORS,0);
clear_bit(STRIPE_SYNCING, &sh->state);
- syncing = 0;
+ s.syncing = 0;
}
/*
@@ -2059,279 +2986,41 @@ static void handle_stripe6(struct stripe_head *sh, struct page *tmp_page)
* are safe, or on a failed drive
*/
pdev = &sh->dev[pd_idx];
- p_failed = (failed >= 1 && failed_num[0] == pd_idx)
- || (failed >= 2 && failed_num[1] == pd_idx);
- qdev = &sh->dev[qd_idx];
- q_failed = (failed >= 1 && failed_num[0] == qd_idx)
- || (failed >= 2 && failed_num[1] == qd_idx);
-
- if ( written &&
- ( p_failed || ((test_bit(R5_Insync, &pdev->flags)
+ r6s.p_failed = (s.failed >= 1 && r6s.failed_num[0] == pd_idx)
+ || (s.failed >= 2 && r6s.failed_num[1] == pd_idx);
+ qdev = &sh->dev[r6s.qd_idx];
+ r6s.q_failed = (s.failed >= 1 && r6s.failed_num[0] == r6s.qd_idx)
+ || (s.failed >= 2 && r6s.failed_num[1] == r6s.qd_idx);
+
+ if ( s.written &&
+ ( r6s.p_failed || ((test_bit(R5_Insync, &pdev->flags)
&& !test_bit(R5_LOCKED, &pdev->flags)
- && test_bit(R5_UPTODATE, &pdev->flags))) ) &&
- ( q_failed || ((test_bit(R5_Insync, &qdev->flags)
+ && test_bit(R5_UPTODATE, &pdev->flags)))) &&
+ ( r6s.q_failed || ((test_bit(R5_Insync, &qdev->flags)
&& !test_bit(R5_LOCKED, &qdev->flags)
- && test_bit(R5_UPTODATE, &qdev->flags))) ) ) {
- /* any written block on an uptodate or failed drive can be
- * returned. Note that if we 'wrote' to a failed drive,
- * it will be UPTODATE, but never LOCKED, so we don't need
- * to test 'failed' directly.
- */
- for (i=disks; i--; )
- if (sh->dev[i].written) {
- dev = &sh->dev[i];
- if (!test_bit(R5_LOCKED, &dev->flags) &&
- test_bit(R5_UPTODATE, &dev->flags) ) {
- /* We can return any write requests */
- int bitmap_end = 0;
- struct bio *wbi, *wbi2;
- PRINTK("Return write for stripe %llu disc %d\n",
- (unsigned long long)sh->sector, i);
- spin_lock_irq(&conf->device_lock);
- wbi = dev->written;
- dev->written = NULL;
- while (wbi && wbi->bi_sector < dev->sector + STRIPE_SECTORS) {
- wbi2 = r5_next_bio(wbi, dev->sector);
- if (--wbi->bi_phys_segments == 0) {
- md_write_end(conf->mddev);
- wbi->bi_next = return_bi;
- return_bi = wbi;
- }
- wbi = wbi2;
- }
- if (dev->towrite == NULL)
- bitmap_end = 1;
- spin_unlock_irq(&conf->device_lock);
- if (bitmap_end)
- bitmap_endwrite(conf->mddev->bitmap, sh->sector,
- STRIPE_SECTORS,
- !test_bit(STRIPE_DEGRADED, &sh->state), 0);
- }
- }
- }
+ && test_bit(R5_UPTODATE, &qdev->flags)))))
+ handle_completed_write_requests(conf, sh, disks, &return_bi);
/* Now we might consider reading some blocks, either to check/generate
* parity, or to satisfy requests
* or to load a block that is being partially written.
*/
- if (to_read || non_overwrite || (to_write && failed) ||
- (syncing && (uptodate < disks)) || expanding) {
- for (i=disks; i--;) {
- dev = &sh->dev[i];
- if (!test_bit(R5_LOCKED, &dev->flags) && !test_bit(R5_UPTODATE, &dev->flags) &&
- (dev->toread ||
- (dev->towrite && !test_bit(R5_OVERWRITE, &dev->flags)) ||
- syncing ||
- expanding ||
- (failed >= 1 && (sh->dev[failed_num[0]].toread || to_write)) ||
- (failed >= 2 && (sh->dev[failed_num[1]].toread || to_write))
- )
- ) {
- /* we would like to get this block, possibly
- * by computing it, but we might not be able to
- */
- if (uptodate == disks-1) {
- PRINTK("Computing stripe %llu block %d\n",
- (unsigned long long)sh->sector, i);
- compute_block_1(sh, i, 0);
- uptodate++;
- } else if ( uptodate == disks-2 && failed >= 2 ) {
- /* Computing 2-failure is *very* expensive; only do it if failed >= 2 */
- int other;
- for (other=disks; other--;) {
- if ( other == i )
- continue;
- if ( !test_bit(R5_UPTODATE, &sh->dev[other].flags) )
- break;
- }
- BUG_ON(other < 0);
- PRINTK("Computing stripe %llu blocks %d,%d\n",
- (unsigned long long)sh->sector, i, other);
- compute_block_2(sh, i, other);
- uptodate += 2;
- } else if (test_bit(R5_Insync, &dev->flags)) {
- set_bit(R5_LOCKED, &dev->flags);
- set_bit(R5_Wantread, &dev->flags);
- locked++;
- PRINTK("Reading block %d (sync=%d)\n",
- i, syncing);
- }
- }
- }
- set_bit(STRIPE_HANDLE, &sh->state);
- }
+ if (s.to_read || s.non_overwrite || (s.to_write && s.failed) ||
+ (s.syncing && (s.uptodate < disks)) || s.expanding)
+ handle_issuing_new_read_requests6(sh, &s, &r6s, disks);
/* now to consider writing and what else, if anything should be read */
- if (to_write) {
- int rcw=0, must_compute=0;
- for (i=disks ; i--;) {
- dev = &sh->dev[i];
- /* Would I have to read this buffer for reconstruct_write */
- if (!test_bit(R5_OVERWRITE, &dev->flags)
- && i != pd_idx && i != qd_idx
- && (!test_bit(R5_LOCKED, &dev->flags)
- ) &&
- !test_bit(R5_UPTODATE, &dev->flags)) {
- if (test_bit(R5_Insync, &dev->flags)) rcw++;
- else {
- PRINTK("raid6: must_compute: disk %d flags=%#lx\n", i, dev->flags);
- must_compute++;
- }
- }
- }
- PRINTK("for sector %llu, rcw=%d, must_compute=%d\n",
- (unsigned long long)sh->sector, rcw, must_compute);
- set_bit(STRIPE_HANDLE, &sh->state);
-
- if (rcw > 0)
- /* want reconstruct write, but need to get some data */
- for (i=disks; i--;) {
- dev = &sh->dev[i];
- if (!test_bit(R5_OVERWRITE, &dev->flags)
- && !(failed == 0 && (i == pd_idx || i == qd_idx))
- && !test_bit(R5_LOCKED, &dev->flags) && !test_bit(R5_UPTODATE, &dev->flags) &&
- test_bit(R5_Insync, &dev->flags)) {
- if (test_bit(STRIPE_PREREAD_ACTIVE, &sh->state))
- {
- PRINTK("Read_old stripe %llu block %d for Reconstruct\n",
- (unsigned long long)sh->sector, i);
- set_bit(R5_LOCKED, &dev->flags);
- set_bit(R5_Wantread, &dev->flags);
- locked++;
- } else {
- PRINTK("Request delayed stripe %llu block %d for Reconstruct\n",
- (unsigned long long)sh->sector, i);
- set_bit(STRIPE_DELAYED, &sh->state);
- set_bit(STRIPE_HANDLE, &sh->state);
- }
- }
- }
- /* now if nothing is locked, and if we have enough data, we can start a write request */
- if (locked == 0 && rcw == 0 &&
- !test_bit(STRIPE_BIT_DELAY, &sh->state)) {
- if ( must_compute > 0 ) {
- /* We have failed blocks and need to compute them */
- switch ( failed ) {
- case 0: BUG();
- case 1: compute_block_1(sh, failed_num[0], 0); break;
- case 2: compute_block_2(sh, failed_num[0], failed_num[1]); break;
- default: BUG(); /* This request should have been failed? */
- }
- }
-
- PRINTK("Computing parity for stripe %llu\n", (unsigned long long)sh->sector);
- compute_parity6(sh, RECONSTRUCT_WRITE);
- /* now every locked buffer is ready to be written */
- for (i=disks; i--;)
- if (test_bit(R5_LOCKED, &sh->dev[i].flags)) {
- PRINTK("Writing stripe %llu block %d\n",
- (unsigned long long)sh->sector, i);
- locked++;
- set_bit(R5_Wantwrite, &sh->dev[i].flags);
- }
- /* after a RECONSTRUCT_WRITE, the stripe MUST be in-sync */
- set_bit(STRIPE_INSYNC, &sh->state);
-
- if (test_and_clear_bit(STRIPE_PREREAD_ACTIVE, &sh->state)) {
- atomic_dec(&conf->preread_active_stripes);
- if (atomic_read(&conf->preread_active_stripes) < IO_THRESHOLD)
- md_wakeup_thread(conf->mddev->thread);
- }
- }
- }
+ if (s.to_write)
+ handle_issuing_new_write_requests6(conf, sh, &s, &r6s, disks);
/* maybe we need to check and possibly fix the parity for this stripe
- * Any reads will already have been scheduled, so we just see if enough data
- * is available
+ * Any reads will already have been scheduled, so we just see if enough
+ * data is available
*/
- if (syncing && locked == 0 && !test_bit(STRIPE_INSYNC, &sh->state)) {
- int update_p = 0, update_q = 0;
- struct r5dev *dev;
-
- set_bit(STRIPE_HANDLE, &sh->state);
-
- BUG_ON(failed>2);
- BUG_ON(uptodate < disks);
- /* Want to check and possibly repair P and Q.
- * However there could be one 'failed' device, in which
- * case we can only check one of them, possibly using the
- * other to generate missing data
- */
-
- /* If !tmp_page, we cannot do the calculations,
- * but as we have set STRIPE_HANDLE, we will soon be called
- * by stripe_handle with a tmp_page - just wait until then.
- */
- if (tmp_page) {
- if (failed == q_failed) {
- /* The only possible failed device holds 'Q', so it makes
- * sense to check P (If anything else were failed, we would
- * have used P to recreate it).
- */
- compute_block_1(sh, pd_idx, 1);
- if (!page_is_zero(sh->dev[pd_idx].page)) {
- compute_block_1(sh,pd_idx,0);
- update_p = 1;
- }
- }
- if (!q_failed && failed < 2) {
- /* q is not failed, and we didn't use it to generate
- * anything, so it makes sense to check it
- */
- memcpy(page_address(tmp_page),
- page_address(sh->dev[qd_idx].page),
- STRIPE_SIZE);
- compute_parity6(sh, UPDATE_PARITY);
- if (memcmp(page_address(tmp_page),
- page_address(sh->dev[qd_idx].page),
- STRIPE_SIZE)!= 0) {
- clear_bit(STRIPE_INSYNC, &sh->state);
- update_q = 1;
- }
- }
- if (update_p || update_q) {
- conf->mddev->resync_mismatches += STRIPE_SECTORS;
- if (test_bit(MD_RECOVERY_CHECK, &conf->mddev->recovery))
- /* don't try to repair!! */
- update_p = update_q = 0;
- }
-
- /* now write out any block on a failed drive,
- * or P or Q if they need it
- */
+ if (s.syncing && s.locked == 0 && !test_bit(STRIPE_INSYNC, &sh->state))
+ handle_parity_checks6(conf, sh, &s, &r6s, tmp_page, disks);
- if (failed == 2) {
- dev = &sh->dev[failed_num[1]];
- locked++;
- set_bit(R5_LOCKED, &dev->flags);
- set_bit(R5_Wantwrite, &dev->flags);
- }
- if (failed >= 1) {
- dev = &sh->dev[failed_num[0]];
- locked++;
- set_bit(R5_LOCKED, &dev->flags);
- set_bit(R5_Wantwrite, &dev->flags);
- }
-
- if (update_p) {
- dev = &sh->dev[pd_idx];
- locked ++;
- set_bit(R5_LOCKED, &dev->flags);
- set_bit(R5_Wantwrite, &dev->flags);
- }
- if (update_q) {
- dev = &sh->dev[qd_idx];
- locked++;
- set_bit(R5_LOCKED, &dev->flags);
- set_bit(R5_Wantwrite, &dev->flags);
- }
- clear_bit(STRIPE_DEGRADED, &sh->state);
-
- set_bit(STRIPE_INSYNC, &sh->state);
- }
- }
-
- if (syncing && locked == 0 && test_bit(STRIPE_INSYNC, &sh->state)) {
+ if (s.syncing && s.locked == 0 && test_bit(STRIPE_INSYNC, &sh->state)) {
md_done_sync(conf->mddev, STRIPE_SECTORS,1);
clear_bit(STRIPE_SYNCING, &sh->state);
}
@@ -2339,9 +3028,9 @@ static void handle_stripe6(struct stripe_head *sh, struct page *tmp_page)
/* If the failed drives are just a ReadError, then we might need
* to progress the repair/check process
*/
- if (failed <= 2 && ! conf->mddev->ro)
- for (i=0; i<failed;i++) {
- dev = &sh->dev[failed_num[i]];
+ if (s.failed <= 2 && !conf->mddev->ro)
+ for (i = 0; i < s.failed; i++) {
+ dev = &sh->dev[r6s.failed_num[i]];
if (test_bit(R5_ReadError, &dev->flags)
&& !test_bit(R5_LOCKED, &dev->flags)
&& test_bit(R5_UPTODATE, &dev->flags)
@@ -2358,7 +3047,7 @@ static void handle_stripe6(struct stripe_head *sh, struct page *tmp_page)
}
}
- if (expanded && test_bit(STRIPE_EXPANDING, &sh->state)) {
+ if (s.expanded && test_bit(STRIPE_EXPANDING, &sh->state)) {
/* Need to write out all blocks after computing P&Q */
sh->disks = conf->raid_disks;
sh->pd_idx = stripe_to_pdidx(sh->sector, conf,
@@ -2366,82 +3055,24 @@ static void handle_stripe6(struct stripe_head *sh, struct page *tmp_page)
compute_parity6(sh, RECONSTRUCT_WRITE);
for (i = conf->raid_disks ; i-- ; ) {
set_bit(R5_LOCKED, &sh->dev[i].flags);
- locked++;
+ s.locked++;
set_bit(R5_Wantwrite, &sh->dev[i].flags);
}
clear_bit(STRIPE_EXPANDING, &sh->state);
- } else if (expanded) {
+ } else if (s.expanded) {
clear_bit(STRIPE_EXPAND_READY, &sh->state);
atomic_dec(&conf->reshape_stripes);
wake_up(&conf->wait_for_overlap);
md_done_sync(conf->mddev, STRIPE_SECTORS, 1);
}
- if (expanding && locked == 0) {
- /* We have read all the blocks in this stripe and now we need to
- * copy some of them into a target stripe for expand.
- */
- clear_bit(STRIPE_EXPAND_SOURCE, &sh->state);
- for (i = 0; i < sh->disks ; i++)
- if (i != pd_idx && i != qd_idx) {
- int dd_idx2, pd_idx2, j;
- struct stripe_head *sh2;
-
- sector_t bn = compute_blocknr(sh, i);
- sector_t s = raid5_compute_sector(
- bn, conf->raid_disks,
- conf->raid_disks - conf->max_degraded,
- &dd_idx2, &pd_idx2, conf);
- sh2 = get_active_stripe(conf, s,
- conf->raid_disks,
- pd_idx2, 1);
- if (sh2 == NULL)
- /* so for only the early blocks of
- * this stripe have been requests.
- * When later blocks get requests, we
- * will try again
- */
- continue;
- if (!test_bit(STRIPE_EXPANDING, &sh2->state) ||
- test_bit(R5_Expanded,
- &sh2->dev[dd_idx2].flags)) {
- /* must have already done this block */
- release_stripe(sh2);
- continue;
- }
- memcpy(page_address(sh2->dev[dd_idx2].page),
- page_address(sh->dev[i].page),
- STRIPE_SIZE);
- set_bit(R5_Expanded, &sh2->dev[dd_idx2].flags);
- set_bit(R5_UPTODATE, &sh2->dev[dd_idx2].flags);
- for (j = 0 ; j < conf->raid_disks ; j++)
- if (j != sh2->pd_idx &&
- j != raid6_next_disk(sh2->pd_idx,
- sh2->disks) &&
- !test_bit(R5_Expanded,
- &sh2->dev[j].flags))
- break;
- if (j == conf->raid_disks) {
- set_bit(STRIPE_EXPAND_READY,
- &sh2->state);
- set_bit(STRIPE_HANDLE, &sh2->state);
- }
- release_stripe(sh2);
- }
- }
+ if (s.expanding && s.locked == 0)
+ handle_stripe_expansion(conf, sh, &r6s);
spin_unlock(&sh->lock);
- while ((bi=return_bi)) {
- int bytes = bi->bi_size;
+ return_io(return_bi);
- return_bi = bi->bi_next;
- bi->bi_next = NULL;
- bi->bi_size = 0;
- bi->bi_end_io(bi, bytes,
- test_bit(BIO_UPTODATE, &bi->bi_flags)
- ? 0 : -EIO);
- }
for (i=disks; i-- ;) {
int rw;
struct bio *bi;
@@ -2470,11 +3101,11 @@ static void handle_stripe6(struct stripe_head *sh, struct page *tmp_page)
rcu_read_unlock();
if (rdev) {
- if (syncing || expanding || expanded)
+ if (s.syncing || s.expanding || s.expanded)
md_sync_acct(rdev->bdev, STRIPE_SECTORS);
bi->bi_bdev = rdev->bdev;
- PRINTK("for %llu schedule op %ld on disc %d\n",
+ pr_debug("for %llu schedule op %ld on disc %d\n",
(unsigned long long)sh->sector, bi->bi_rw, i);
atomic_inc(&sh->count);
bi->bi_sector = sh->sector + rdev->data_offset;
@@ -2494,7 +3125,7 @@ static void handle_stripe6(struct stripe_head *sh, struct page *tmp_page)
} else {
if (rw == WRITE)
set_bit(STRIPE_DEGRADED, &sh->state);
- PRINTK("skip op %ld on disc %d for sector %llu\n",
+ pr_debug("skip op %ld on disc %d for sector %llu\n",
bi->bi_rw, i, (unsigned long long)sh->sector);
clear_bit(R5_LOCKED, &sh->dev[i].flags);
set_bit(STRIPE_HANDLE, &sh->state);
@@ -2738,7 +3369,7 @@ static int raid5_align_endio(struct bio *bi, unsigned int bytes, int error)
}
- PRINTK("raid5_align_endio : io error...handing IO for a retry\n");
+ pr_debug("raid5_align_endio : io error...handing IO for a retry\n");
add_bio_to_retry(raid_bi, conf);
return 0;
@@ -2776,7 +3407,7 @@ static int chunk_aligned_read(request_queue_t *q, struct bio * raid_bio)
mdk_rdev_t *rdev;
if (!in_chunk_boundary(mddev, raid_bio)) {
- PRINTK("chunk_aligned_read : non aligned\n");
+ pr_debug("chunk_aligned_read : non aligned\n");
return 0;
}
/*
@@ -2900,7 +3531,7 @@ static int make_request(request_queue_t *q, struct bio * bi)
new_sector = raid5_compute_sector(logical_sector, disks, data_disks,
&dd_idx, &pd_idx, conf);
- PRINTK("raid5: make_request, sector %llu logical %llu\n",
+ pr_debug("raid5: make_request, sector %llu logical %llu\n",
(unsigned long long)new_sector,
(unsigned long long)logical_sector);
@@ -3273,7 +3904,7 @@ static void raid5d (mddev_t *mddev)
raid5_conf_t *conf = mddev_to_conf(mddev);
int handled;
- PRINTK("+++ raid5d active\n");
+ pr_debug("+++ raid5d active\n");
md_check_recovery(mddev);
@@ -3308,8 +3939,10 @@ static void raid5d (mddev_t *mddev)
handled++;
}
- if (list_empty(&conf->handle_list))
+ if (list_empty(&conf->handle_list)) {
+ async_tx_issue_pending_all();
break;
+ }
first = conf->handle_list.next;
sh = list_entry(first, struct stripe_head, lru);
@@ -3325,13 +3958,13 @@ static void raid5d (mddev_t *mddev)
spin_lock_irq(&conf->device_lock);
}
- PRINTK("%d stripes handled\n", handled);
+ pr_debug("%d stripes handled\n", handled);
spin_unlock_irq(&conf->device_lock);
unplug_slaves(mddev);
- PRINTK("--- raid5d inactive\n");
+ pr_debug("--- raid5d inactive\n");
}
static ssize_t
@@ -3507,7 +4140,7 @@ static int run(mddev_t *mddev)
atomic_set(&conf->preread_active_stripes, 0);
atomic_set(&conf->active_aligned_reads, 0);
- PRINTK("raid5: run(%s) called.\n", mdname(mddev));
+ pr_debug("raid5: run(%s) called.\n", mdname(mddev));
ITERATE_RDEV(mddev,rdev,tmp) {
raid_disk = rdev->raid_disk;
@@ -3690,7 +4323,7 @@ static int stop(mddev_t *mddev)
return 0;
}
-#if RAID5_DEBUG
+#ifdef DEBUG
static void print_sh (struct seq_file *seq, struct stripe_head *sh)
{
int i;
@@ -3737,7 +4370,7 @@ static void status (struct seq_file *seq, mddev_t *mddev)
conf->disks[i].rdev &&
test_bit(In_sync, &conf->disks[i].rdev->flags) ? "U" : "_");
seq_printf (seq, "]");
-#if RAID5_DEBUG
+#ifdef DEBUG
seq_printf (seq, "\n");
printall(seq, conf);
#endif
diff --git a/drivers/md/xor.c b/drivers/md/xor.c
deleted file mode 100644
index 324897c4be4..00000000000
--- a/drivers/md/xor.c
+++ /dev/null
@@ -1,154 +0,0 @@
-/*
- * xor.c : Multiple Devices driver for Linux
- *
- * Copyright (C) 1996, 1997, 1998, 1999, 2000,
- * Ingo Molnar, Matti Aarnio, Jakub Jelinek, Richard Henderson.
- *
- * Dispatch optimized RAID-5 checksumming functions.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2, or (at your option)
- * any later version.
- *
- * You should have received a copy of the GNU General Public License
- * (for example /usr/src/linux/COPYING); if not, write to the Free
- * Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
- */
-
-#define BH_TRACE 0
-#include <linux/module.h>
-#include <linux/raid/md.h>
-#include <linux/raid/xor.h>
-#include <asm/xor.h>
-
-/* The xor routines to use. */
-static struct xor_block_template *active_template;
-
-void
-xor_block(unsigned int count, unsigned int bytes, void **ptr)
-{
- unsigned long *p0, *p1, *p2, *p3, *p4;
-
- p0 = (unsigned long *) ptr[0];
- p1 = (unsigned long *) ptr[1];
- if (count == 2) {
- active_template->do_2(bytes, p0, p1);
- return;
- }
-
- p2 = (unsigned long *) ptr[2];
- if (count == 3) {
- active_template->do_3(bytes, p0, p1, p2);
- return;
- }
-
- p3 = (unsigned long *) ptr[3];
- if (count == 4) {
- active_template->do_4(bytes, p0, p1, p2, p3);
- return;
- }
-
- p4 = (unsigned long *) ptr[4];
- active_template->do_5(bytes, p0, p1, p2, p3, p4);
-}
-
-/* Set of all registered templates. */
-static struct xor_block_template *template_list;
-
-#define BENCH_SIZE (PAGE_SIZE)
-
-static void
-do_xor_speed(struct xor_block_template *tmpl, void *b1, void *b2)
-{
- int speed;
- unsigned long now;
- int i, count, max;
-
- tmpl->next = template_list;
- template_list = tmpl;
-
- /*
- * Count the number of XORs done during a whole jiffy, and use
- * this to calculate the speed of checksumming. We use a 2-page
- * allocation to have guaranteed color L1-cache layout.
- */
- max = 0;
- for (i = 0; i < 5; i++) {
- now = jiffies;
- count = 0;
- while (jiffies == now) {
- mb();
- tmpl->do_2(BENCH_SIZE, b1, b2);
- mb();
- count++;
- mb();
- }
- if (count > max)
- max = count;
- }
-
- speed = max * (HZ * BENCH_SIZE / 1024);
- tmpl->speed = speed;
-
- printk(" %-10s: %5d.%03d MB/sec\n", tmpl->name,
- speed / 1000, speed % 1000);
-}
-
-static int
-calibrate_xor_block(void)
-{
- void *b1, *b2;
- struct xor_block_template *f, *fastest;
-
- b1 = (void *) __get_free_pages(GFP_KERNEL, 2);
- if (! b1) {
- printk("raid5: Yikes! No memory available.\n");
- return -ENOMEM;
- }
- b2 = b1 + 2*PAGE_SIZE + BENCH_SIZE;
-
- /*
- * If this arch/cpu has a short-circuited selection, don't loop through all
- * the possible functions, just test the best one
- */
-
- fastest = NULL;
-
-#ifdef XOR_SELECT_TEMPLATE
- fastest = XOR_SELECT_TEMPLATE(fastest);
-#endif
-
-#define xor_speed(templ) do_xor_speed((templ), b1, b2)
-
- if (fastest) {
- printk(KERN_INFO "raid5: automatically using best checksumming function: %s\n",
- fastest->name);
- xor_speed(fastest);
- } else {
- printk(KERN_INFO "raid5: measuring checksumming speed\n");
- XOR_TRY_TEMPLATES;
- fastest = template_list;
- for (f = fastest; f; f = f->next)
- if (f->speed > fastest->speed)
- fastest = f;
- }
-
- printk("raid5: using function: %s (%d.%03d MB/sec)\n",
- fastest->name, fastest->speed / 1000, fastest->speed % 1000);
-
-#undef xor_speed
-
- free_pages((unsigned long)b1, 2);
-
- active_template = fastest;
- return 0;
-}
-
-static __exit void xor_exit(void) { }
-
-EXPORT_SYMBOL(xor_block);
-MODULE_LICENSE("GPL");
-
-module_init(calibrate_xor_block);
-module_exit(xor_exit);
generated by cgit v1.2.3-70-g09d2 (git 2.46.0) at 2024-11-22 02:33:30 +0000