summaryrefslogtreecommitdiff
path: root/drivers/block
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/block')
-rw-r--r--drivers/block/drbd/drbd_actlog.c8
-rw-r--r--drivers/block/drbd/drbd_bitmap.c15
-rw-r--r--drivers/block/drbd/drbd_int.h44
-rw-r--r--drivers/block/drbd/drbd_main.c65
-rw-r--r--drivers/block/drbd/drbd_nl.c36
-rw-r--r--drivers/block/drbd/drbd_proc.c3
-rw-r--r--drivers/block/drbd/drbd_receiver.c38
-rw-r--r--drivers/block/drbd/drbd_req.c75
-rw-r--r--drivers/block/drbd/drbd_worker.c12
-rw-r--r--drivers/block/floppy.c1
-rw-r--r--drivers/block/loop.c8
-rw-r--r--drivers/block/mtip32xx/mtip32xx.c390
-rw-r--r--drivers/block/mtip32xx/mtip32xx.h53
-rw-r--r--drivers/block/nbd.c2
-rw-r--r--drivers/block/rbd.c76
-rw-r--r--drivers/block/umem.c17
-rw-r--r--drivers/block/xen-blkback/common.h2
-rw-r--r--drivers/block/xen-blkfront.c58
18 files changed, 598 insertions, 305 deletions
diff --git a/drivers/block/drbd/drbd_actlog.c b/drivers/block/drbd/drbd_actlog.c
index e54e31b02b88..3fbef018ce55 100644
--- a/drivers/block/drbd/drbd_actlog.c
+++ b/drivers/block/drbd/drbd_actlog.c
@@ -411,7 +411,7 @@ w_al_write_transaction(struct drbd_conf *mdev, struct drbd_work *w, int unused)
+ mdev->ldev->md.al_offset + mdev->al_tr_pos;
if (!drbd_md_sync_page_io(mdev, mdev->ldev, sector, WRITE))
- drbd_chk_io_error(mdev, 1, true);
+ drbd_chk_io_error(mdev, 1, DRBD_META_IO_ERROR);
if (++mdev->al_tr_pos >
div_ceil(mdev->act_log->nr_elements, AL_EXTENTS_PT))
@@ -876,7 +876,11 @@ int __drbd_set_out_of_sync(struct drbd_conf *mdev, sector_t sector, int size,
unsigned int enr, count = 0;
struct lc_element *e;
- if (size <= 0 || (size & 0x1ff) != 0 || size > DRBD_MAX_BIO_SIZE) {
+ /* this should be an empty REQ_FLUSH */
+ if (size == 0)
+ return 0;
+
+ if (size < 0 || (size & 0x1ff) != 0 || size > DRBD_MAX_BIO_SIZE) {
dev_err(DEV, "sector: %llus, size: %d\n",
(unsigned long long)sector, size);
return 0;
diff --git a/drivers/block/drbd/drbd_bitmap.c b/drivers/block/drbd/drbd_bitmap.c
index b5c5ff53cb57..ba91b408abad 100644
--- a/drivers/block/drbd/drbd_bitmap.c
+++ b/drivers/block/drbd/drbd_bitmap.c
@@ -1096,7 +1096,7 @@ static int bm_rw(struct drbd_conf *mdev, int rw, unsigned flags, unsigned lazy_w
if (ctx->error) {
dev_alert(DEV, "we had at least one MD IO ERROR during bitmap IO\n");
- drbd_chk_io_error(mdev, 1, true);
+ drbd_chk_io_error(mdev, 1, DRBD_META_IO_ERROR);
err = -EIO; /* ctx->error ? */
}
@@ -1212,7 +1212,7 @@ int drbd_bm_write_page(struct drbd_conf *mdev, unsigned int idx) __must_hold(loc
wait_until_done_or_disk_failure(mdev, mdev->ldev, &ctx->done);
if (ctx->error)
- drbd_chk_io_error(mdev, 1, true);
+ drbd_chk_io_error(mdev, 1, DRBD_META_IO_ERROR);
/* that should force detach, so the in memory bitmap will be
* gone in a moment as well. */
@@ -1475,10 +1475,17 @@ void _drbd_bm_set_bits(struct drbd_conf *mdev, const unsigned long s, const unsi
first_word = 0;
spin_lock_irq(&b->bm_lock);
}
-
/* last page (respectively only page, for first page == last page) */
last_word = MLPP(el >> LN2_BPL);
- bm_set_full_words_within_one_page(mdev->bitmap, last_page, first_word, last_word);
+
+ /* consider bitmap->bm_bits = 32768, bitmap->bm_number_of_pages = 1. (or multiples).
+ * ==> e = 32767, el = 32768, last_page = 2,
+ * and now last_word = 0.
+ * We do not want to touch last_page in this case,
+ * as we did not allocate it, it is not present in bitmap->bm_pages.
+ */
+ if (last_word)
+ bm_set_full_words_within_one_page(mdev->bitmap, last_page, first_word, last_word);
/* possibly trailing bits.
* example: (e & 63) == 63, el will be e+1.
diff --git a/drivers/block/drbd/drbd_int.h b/drivers/block/drbd/drbd_int.h
index 02f013a073a7..b2ca143d0053 100644
--- a/drivers/block/drbd/drbd_int.h
+++ b/drivers/block/drbd/drbd_int.h
@@ -813,7 +813,6 @@ enum {
SIGNAL_ASENDER, /* whether asender wants to be interrupted */
SEND_PING, /* whether asender should send a ping asap */
- UNPLUG_QUEUED, /* only relevant with kernel 2.4 */
UNPLUG_REMOTE, /* sending a "UnplugRemote" could help */
MD_DIRTY, /* current uuids and flags not yet on disk */
DISCARD_CONCURRENT, /* Set on one node, cleared on the peer! */
@@ -824,7 +823,6 @@ enum {
CRASHED_PRIMARY, /* This node was a crashed primary.
* Gets cleared when the state.conn
* goes into C_CONNECTED state. */
- NO_BARRIER_SUPP, /* underlying block device doesn't implement barriers */
CONSIDER_RESYNC,
MD_NO_FUA, /* Users wants us to not use FUA/FLUSH on meta data dev */
@@ -834,6 +832,7 @@ enum {
BITMAP_IO_QUEUED, /* Started bitmap IO */
GO_DISKLESS, /* Disk is being detached, on io-error or admin request. */
WAS_IO_ERROR, /* Local disk failed returned IO error */
+ FORCE_DETACH, /* Force-detach from local disk, aborting any pending local IO */
RESYNC_AFTER_NEG, /* Resync after online grow after the attach&negotiate finished. */
NET_CONGESTED, /* The data socket is congested */
@@ -851,6 +850,13 @@ enum {
AL_SUSPENDED, /* Activity logging is currently suspended. */
AHEAD_TO_SYNC_SOURCE, /* Ahead -> SyncSource queued */
STATE_SENT, /* Do not change state/UUIDs while this is set */
+
+ CALLBACK_PENDING, /* Whether we have a call_usermodehelper(, UMH_WAIT_PROC)
+ * pending, from drbd worker context.
+ * If set, bdi_write_congested() returns true,
+ * so shrink_page_list() would not recurse into,
+ * and potentially deadlock on, this drbd worker.
+ */
};
struct drbd_bitmap; /* opaque for drbd_conf */
@@ -1130,8 +1136,8 @@ struct drbd_conf {
int rs_in_flight; /* resync sectors in flight (to proxy, in proxy and from proxy) */
int rs_planed; /* resync sectors already planned */
atomic_t ap_in_flight; /* App sectors in flight (waiting for ack) */
- int peer_max_bio_size;
- int local_max_bio_size;
+ unsigned int peer_max_bio_size;
+ unsigned int local_max_bio_size;
};
static inline struct drbd_conf *minor_to_mdev(unsigned int minor)
@@ -1435,9 +1441,9 @@ struct bm_extent {
* hash table. */
#define HT_SHIFT 8
#define DRBD_MAX_BIO_SIZE (1U<<(9+HT_SHIFT))
-#define DRBD_MAX_BIO_SIZE_SAFE (1 << 12) /* Works always = 4k */
+#define DRBD_MAX_BIO_SIZE_SAFE (1U << 12) /* Works always = 4k */
-#define DRBD_MAX_SIZE_H80_PACKET (1 << 15) /* The old header only allows packets up to 32Kib data */
+#define DRBD_MAX_SIZE_H80_PACKET (1U << 15) /* The old header only allows packets up to 32Kib data */
/* Number of elements in the app_reads_hash */
#define APP_R_HSIZE 15
@@ -1840,12 +1846,20 @@ static inline int drbd_request_state(struct drbd_conf *mdev,
return _drbd_request_state(mdev, mask, val, CS_VERBOSE + CS_ORDERED);
}
+enum drbd_force_detach_flags {
+ DRBD_IO_ERROR,
+ DRBD_META_IO_ERROR,
+ DRBD_FORCE_DETACH,
+};
+
#define __drbd_chk_io_error(m,f) __drbd_chk_io_error_(m,f, __func__)
-static inline void __drbd_chk_io_error_(struct drbd_conf *mdev, int forcedetach, const char *where)
+static inline void __drbd_chk_io_error_(struct drbd_conf *mdev,
+ enum drbd_force_detach_flags forcedetach,
+ const char *where)
{
switch (mdev->ldev->dc.on_io_error) {
case EP_PASS_ON:
- if (!forcedetach) {
+ if (forcedetach == DRBD_IO_ERROR) {
if (__ratelimit(&drbd_ratelimit_state))
dev_err(DEV, "Local IO failed in %s.\n", where);
if (mdev->state.disk > D_INCONSISTENT)
@@ -1856,6 +1870,8 @@ static inline void __drbd_chk_io_error_(struct drbd_conf *mdev, int forcedetach,
case EP_DETACH:
case EP_CALL_HELPER:
set_bit(WAS_IO_ERROR, &mdev->flags);
+ if (forcedetach == DRBD_FORCE_DETACH)
+ set_bit(FORCE_DETACH, &mdev->flags);
if (mdev->state.disk > D_FAILED) {
_drbd_set_state(_NS(mdev, disk, D_FAILED), CS_HARD, NULL);
dev_err(DEV,
@@ -1875,7 +1891,7 @@ static inline void __drbd_chk_io_error_(struct drbd_conf *mdev, int forcedetach,
*/
#define drbd_chk_io_error(m,e,f) drbd_chk_io_error_(m,e,f, __func__)
static inline void drbd_chk_io_error_(struct drbd_conf *mdev,
- int error, int forcedetach, const char *where)
+ int error, enum drbd_force_detach_flags forcedetach, const char *where)
{
if (error) {
unsigned long flags;
@@ -2405,15 +2421,17 @@ static inline void dec_ap_bio(struct drbd_conf *mdev)
int ap_bio = atomic_dec_return(&mdev->ap_bio_cnt);
D_ASSERT(ap_bio >= 0);
+
+ if (ap_bio == 0 && test_bit(BITMAP_IO, &mdev->flags)) {
+ if (!test_and_set_bit(BITMAP_IO_QUEUED, &mdev->flags))
+ drbd_queue_work(&mdev->data.work, &mdev->bm_io_work.w);
+ }
+
/* this currently does wake_up for every dec_ap_bio!
* maybe rather introduce some type of hysteresis?
* e.g. (ap_bio == mxb/2 || ap_bio == 0) ? */
if (ap_bio < mxb)
wake_up(&mdev->misc_wait);
- if (ap_bio == 0 && test_bit(BITMAP_IO, &mdev->flags)) {
- if (!test_and_set_bit(BITMAP_IO_QUEUED, &mdev->flags))
- drbd_queue_work(&mdev->data.work, &mdev->bm_io_work.w);
- }
}
static inline int drbd_set_ed_uuid(struct drbd_conf *mdev, u64 val)
diff --git a/drivers/block/drbd/drbd_main.c b/drivers/block/drbd/drbd_main.c
index 920ede2829d6..2e0e7fc1dbba 100644
--- a/drivers/block/drbd/drbd_main.c
+++ b/drivers/block/drbd/drbd_main.c
@@ -1514,6 +1514,13 @@ static void after_state_ch(struct drbd_conf *mdev, union drbd_state os,
/* Do not change the order of the if above and the two below... */
if (os.pdsk == D_DISKLESS && ns.pdsk > D_DISKLESS) { /* attach on the peer */
+ /* we probably will start a resync soon.
+ * make sure those things are properly reset. */
+ mdev->rs_total = 0;
+ mdev->rs_failed = 0;
+ atomic_set(&mdev->rs_pending_cnt, 0);
+ drbd_rs_cancel_all(mdev);
+
drbd_send_uuids(mdev);
drbd_send_state(mdev, ns);
}
@@ -1630,9 +1637,24 @@ static void after_state_ch(struct drbd_conf *mdev, union drbd_state os,
eh = mdev->ldev->dc.on_io_error;
was_io_error = test_and_clear_bit(WAS_IO_ERROR, &mdev->flags);
- /* Immediately allow completion of all application IO, that waits
- for completion from the local disk. */
- tl_abort_disk_io(mdev);
+ if (was_io_error && eh == EP_CALL_HELPER)
+ drbd_khelper(mdev, "local-io-error");
+
+ /* Immediately allow completion of all application IO,
+ * that waits for completion from the local disk,
+ * if this was a force-detach due to disk_timeout
+ * or administrator request (drbdsetup detach --force).
+ * Do NOT abort otherwise.
+ * Aborting local requests may cause serious problems,
+ * if requests are completed to upper layers already,
+ * and then later the already submitted local bio completes.
+ * This can cause DMA into former bio pages that meanwhile
+ * have been re-used for other things.
+ * So aborting local requests may cause crashes,
+ * or even worse, silent data corruption.
+ */
+ if (test_and_clear_bit(FORCE_DETACH, &mdev->flags))
+ tl_abort_disk_io(mdev);
/* current state still has to be D_FAILED,
* there is only one way out: to D_DISKLESS,
@@ -1653,9 +1675,6 @@ static void after_state_ch(struct drbd_conf *mdev, union drbd_state os,
drbd_md_sync(mdev);
}
put_ldev(mdev);
-
- if (was_io_error && eh == EP_CALL_HELPER)
- drbd_khelper(mdev, "local-io-error");
}
/* second half of local IO error, failure to attach,
@@ -1669,10 +1688,6 @@ static void after_state_ch(struct drbd_conf *mdev, union drbd_state os,
"ASSERT FAILED: disk is %s while going diskless\n",
drbd_disk_str(mdev->state.disk));
- mdev->rs_total = 0;
- mdev->rs_failed = 0;
- atomic_set(&mdev->rs_pending_cnt, 0);
-
if (ns.conn >= C_CONNECTED)
drbd_send_state(mdev, ns);
@@ -2194,7 +2209,8 @@ int drbd_send_sizes(struct drbd_conf *mdev, int trigger_reply, enum dds_flags fl
{
struct p_sizes p;
sector_t d_size, u_size;
- int q_order_type, max_bio_size;
+ int q_order_type;
+ unsigned int max_bio_size;
int ok;
if (get_ldev_if_state(mdev, D_NEGOTIATING)) {
@@ -2203,7 +2219,7 @@ int drbd_send_sizes(struct drbd_conf *mdev, int trigger_reply, enum dds_flags fl
u_size = mdev->ldev->dc.disk_size;
q_order_type = drbd_queue_order_type(mdev);
max_bio_size = queue_max_hw_sectors(mdev->ldev->backing_bdev->bd_disk->queue) << 9;
- max_bio_size = min_t(int, max_bio_size, DRBD_MAX_BIO_SIZE);
+ max_bio_size = min(max_bio_size, DRBD_MAX_BIO_SIZE);
put_ldev(mdev);
} else {
d_size = 0;
@@ -2214,7 +2230,7 @@ int drbd_send_sizes(struct drbd_conf *mdev, int trigger_reply, enum dds_flags fl
/* Never allow old drbd (up to 8.3.7) to see more than 32KiB */
if (mdev->agreed_pro_version <= 94)
- max_bio_size = min_t(int, max_bio_size, DRBD_MAX_SIZE_H80_PACKET);
+ max_bio_size = min(max_bio_size, DRBD_MAX_SIZE_H80_PACKET);
p.d_size = cpu_to_be64(d_size);
p.u_size = cpu_to_be64(u_size);
@@ -3541,6 +3557,22 @@ static int drbd_congested(void *congested_data, int bdi_bits)
goto out;
}
+ if (test_bit(CALLBACK_PENDING, &mdev->flags)) {
+ r |= (1 << BDI_async_congested);
+ /* Without good local data, we would need to read from remote,
+ * and that would need the worker thread as well, which is
+ * currently blocked waiting for that usermode helper to
+ * finish.
+ */
+ if (!get_ldev_if_state(mdev, D_UP_TO_DATE))
+ r |= (1 << BDI_sync_congested);
+ else
+ put_ldev(mdev);
+ r &= bdi_bits;
+ reason = 'c';
+ goto out;
+ }
+
if (get_ldev(mdev)) {
q = bdev_get_queue(mdev->ldev->backing_bdev);
r = bdi_congested(&q->backing_dev_info, bdi_bits);
@@ -3604,6 +3636,7 @@ struct drbd_conf *drbd_new_device(unsigned int minor)
q->backing_dev_info.congested_data = mdev;
blk_queue_make_request(q, drbd_make_request);
+ blk_queue_flush(q, REQ_FLUSH | REQ_FUA);
/* Setting the max_hw_sectors to an odd value of 8kibyte here
This triggers a max_bio_size message upon first attach or connect */
blk_queue_max_hw_sectors(q, DRBD_MAX_BIO_SIZE_SAFE >> 8);
@@ -3870,7 +3903,7 @@ void drbd_md_sync(struct drbd_conf *mdev)
if (!drbd_md_sync_page_io(mdev, mdev->ldev, sector, WRITE)) {
/* this was a try anyways ... */
dev_err(DEV, "meta data update failed!\n");
- drbd_chk_io_error(mdev, 1, true);
+ drbd_chk_io_error(mdev, 1, DRBD_META_IO_ERROR);
}
/* Update mdev->ldev->md.la_size_sect,
@@ -3950,9 +3983,9 @@ int drbd_md_read(struct drbd_conf *mdev, struct drbd_backing_dev *bdev)
spin_lock_irq(&mdev->req_lock);
if (mdev->state.conn < C_CONNECTED) {
- int peer;
+ unsigned int peer;
peer = be32_to_cpu(buffer->la_peer_max_bio_size);
- peer = max_t(int, peer, DRBD_MAX_BIO_SIZE_SAFE);
+ peer = max(peer, DRBD_MAX_BIO_SIZE_SAFE);
mdev->peer_max_bio_size = peer;
}
spin_unlock_irq(&mdev->req_lock);
diff --git a/drivers/block/drbd/drbd_nl.c b/drivers/block/drbd/drbd_nl.c
index 6d4de6a72e80..fb9dce8daa24 100644
--- a/drivers/block/drbd/drbd_nl.c
+++ b/drivers/block/drbd/drbd_nl.c
@@ -147,6 +147,9 @@ int drbd_khelper(struct drbd_conf *mdev, char *cmd)
char *argv[] = {usermode_helper, cmd, mb, NULL };
int ret;
+ if (current == mdev->worker.task)
+ set_bit(CALLBACK_PENDING, &mdev->flags);
+
snprintf(mb, 12, "minor-%d", mdev_to_minor(mdev));
if (get_net_conf(mdev)) {
@@ -189,6 +192,9 @@ int drbd_khelper(struct drbd_conf *mdev, char *cmd)
usermode_helper, cmd, mb,
(ret >> 8) & 0xff, ret);
+ if (current == mdev->worker.task)
+ clear_bit(CALLBACK_PENDING, &mdev->flags);
+
if (ret < 0) /* Ignore any ERRNOs we got. */
ret = 0;
@@ -795,8 +801,8 @@ static int drbd_check_al_size(struct drbd_conf *mdev)
static void drbd_setup_queue_param(struct drbd_conf *mdev, unsigned int max_bio_size)
{
struct request_queue * const q = mdev->rq_queue;
- int max_hw_sectors = max_bio_size >> 9;
- int max_segments = 0;
+ unsigned int max_hw_sectors = max_bio_size >> 9;
+ unsigned int max_segments = 0;
if (get_ldev_if_state(mdev, D_ATTACHING)) {
struct request_queue * const b = mdev->ldev->backing_bdev->bd_disk->queue;
@@ -829,7 +835,7 @@ static void drbd_setup_queue_param(struct drbd_conf *mdev, unsigned int max_bio_
void drbd_reconsider_max_bio_size(struct drbd_conf *mdev)
{
- int now, new, local, peer;
+ unsigned int now, new, local, peer;
now = queue_max_hw_sectors(mdev->rq_queue) << 9;
local = mdev->local_max_bio_size; /* Eventually last known value, from volatile memory */
@@ -840,13 +846,14 @@ void drbd_reconsider_max_bio_size(struct drbd_conf *mdev)
mdev->local_max_bio_size = local;
put_ldev(mdev);
}
+ local = min(local, DRBD_MAX_BIO_SIZE);
/* We may ignore peer limits if the peer is modern enough.
Because new from 8.3.8 onwards the peer can use multiple
BIOs for a single peer_request */
if (mdev->state.conn >= C_CONNECTED) {
if (mdev->agreed_pro_version < 94) {
- peer = min_t(int, mdev->peer_max_bio_size, DRBD_MAX_SIZE_H80_PACKET);
+ peer = min(mdev->peer_max_bio_size, DRBD_MAX_SIZE_H80_PACKET);
/* Correct old drbd (up to 8.3.7) if it believes it can do more than 32KiB */
} else if (mdev->agreed_pro_version == 94)
peer = DRBD_MAX_SIZE_H80_PACKET;
@@ -854,10 +861,10 @@ void drbd_reconsider_max_bio_size(struct drbd_conf *mdev)
peer = DRBD_MAX_BIO_SIZE;
}
- new = min_t(int, local, peer);
+ new = min(local, peer);
if (mdev->state.role == R_PRIMARY && new < now)
- dev_err(DEV, "ASSERT FAILED new < now; (%d < %d)\n", new, now);
+ dev_err(DEV, "ASSERT FAILED new < now; (%u < %u)\n", new, now);
if (new != now)
dev_info(DEV, "max BIO size = %u\n", new);
@@ -950,6 +957,14 @@ static int drbd_nl_disk_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp
* to realize a "hot spare" feature (not that I'd recommend that) */
wait_event(mdev->misc_wait, !atomic_read(&mdev->local_cnt));
+ /* make sure there is no leftover from previous force-detach attempts */
+ clear_bit(FORCE_DETACH, &mdev->flags);
+
+ /* and no leftover from previously aborted resync or verify, either */
+ mdev->rs_total = 0;
+ mdev->rs_failed = 0;
+ atomic_set(&mdev->rs_pending_cnt, 0);
+
/* allocation not in the IO path, cqueue thread context */
nbc = kzalloc(sizeof(struct drbd_backing_dev), GFP_KERNEL);
if (!nbc) {
@@ -1345,6 +1360,7 @@ static int drbd_nl_detach(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp,
}
if (dt.detach_force) {
+ set_bit(FORCE_DETACH, &mdev->flags);
drbd_force_state(mdev, NS(disk, D_FAILED));
reply->ret_code = SS_SUCCESS;
goto out;
@@ -1962,9 +1978,11 @@ static int drbd_nl_invalidate(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nl
int retcode;
/* If there is still bitmap IO pending, probably because of a previous
- * resync just being finished, wait for it before requesting a new resync. */
+ * resync just being finished, wait for it before requesting a new resync.
+ * Also wait for it's after_state_ch(). */
drbd_suspend_io(mdev);
wait_event(mdev->misc_wait, !test_bit(BITMAP_IO, &mdev->flags));
+ drbd_flush_workqueue(mdev);
retcode = _drbd_request_state(mdev, NS(conn, C_STARTING_SYNC_T), CS_ORDERED);
@@ -2003,9 +2021,11 @@ static int drbd_nl_invalidate_peer(struct drbd_conf *mdev, struct drbd_nl_cfg_re
int retcode;
/* If there is still bitmap IO pending, probably because of a previous
- * resync just being finished, wait for it before requesting a new resync. */
+ * resync just being finished, wait for it before requesting a new resync.
+ * Also wait for it's after_state_ch(). */
drbd_suspend_io(mdev);
wait_event(mdev->misc_wait, !test_bit(BITMAP_IO, &mdev->flags));
+ drbd_flush_workqueue(mdev);
retcode = _drbd_request_state(mdev, NS(conn, C_STARTING_SYNC_S), CS_ORDERED);
diff --git a/drivers/block/drbd/drbd_proc.c b/drivers/block/drbd/drbd_proc.c
index 869bada2ed06..5496104f90b9 100644
--- a/drivers/block/drbd/drbd_proc.c
+++ b/drivers/block/drbd/drbd_proc.c
@@ -245,6 +245,9 @@ static int drbd_seq_show(struct seq_file *seq, void *v)
mdev->state.role == R_SECONDARY) {
seq_printf(seq, "%2d: cs:Unconfigured\n", i);
} else {
+ /* reset mdev->congestion_reason */
+ bdi_rw_congested(&mdev->rq_queue->backing_dev_info);
+
seq_printf(seq,
"%2d: cs:%s ro:%s/%s ds:%s/%s %c %c%c%c%c%c%c\n"
" ns:%u nr:%u dw:%u dr:%u al:%u bm:%u "
diff --git a/drivers/block/drbd/drbd_receiver.c b/drivers/block/drbd/drbd_receiver.c
index ea4836e0ae98..c74ca2df7431 100644
--- a/drivers/block/drbd/drbd_receiver.c
+++ b/drivers/block/drbd/drbd_receiver.c
@@ -277,6 +277,9 @@ static void drbd_pp_free(struct drbd_conf *mdev, struct page *page, int is_net)
atomic_t *a = is_net ? &mdev->pp_in_use_by_net : &mdev->pp_in_use;
int i;
+ if (page == NULL)
+ return;
+
if (drbd_pp_vacant > (DRBD_MAX_BIO_SIZE/PAGE_SIZE)*minor_count)
i = page_chain_free(page);
else {
@@ -316,7 +319,7 @@ struct drbd_epoch_entry *drbd_alloc_ee(struct drbd_conf *mdev,
gfp_t gfp_mask) __must_hold(local)
{
struct drbd_epoch_entry *e;
- struct page *page;
+ struct page *page = NULL;
unsigned nr_pages = (data_size + PAGE_SIZE -1) >> PAGE_SHIFT;
if (drbd_insert_fault(mdev, DRBD_FAULT_AL_EE))
@@ -329,9 +332,11 @@ struct drbd_epoch_entry *drbd_alloc_ee(struct drbd_conf *mdev,
return NULL;
}
- page = drbd_pp_alloc(mdev, nr_pages, (gfp_mask & __GFP_WAIT));
- if (!page)
- goto fail;
+ if (data_size) {
+ page = drbd_pp_alloc(mdev, nr_pages, (gfp_mask & __GFP_WAIT));
+ if (!page)
+ goto fail;
+ }
INIT_HLIST_NODE(&e->collision);
e->epoch = NULL;
@@ -1270,7 +1275,6 @@ read_in_block(struct drbd_conf *mdev, u64 id, sector_t sector, int data_size) __
data_size -= dgs;
- ERR_IF(data_size == 0) return NULL;
ERR_IF(data_size & 0x1ff) return NULL;
ERR_IF(data_size > DRBD_MAX_BIO_SIZE) return NULL;
@@ -1291,6 +1295,9 @@ read_in_block(struct drbd_conf *mdev, u64 id, sector_t sector, int data_size) __
if (!e)
return NULL;
+ if (!data_size)
+ return e;
+
ds = data_size;
page = e->pages;
page_chain_for_each(page) {
@@ -1715,6 +1722,10 @@ static int receive_Data(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned
dp_flags = be32_to_cpu(p->dp_flags);
rw |= wire_flags_to_bio(mdev, dp_flags);
+ if (e->pages == NULL) {
+ D_ASSERT(e->size == 0);
+ D_ASSERT(dp_flags & DP_FLUSH);
+ }
if (dp_flags & DP_MAY_SET_IN_SYNC)
e->flags |= EE_MAY_SET_IN_SYNC;
@@ -3801,11 +3812,18 @@ void drbd_free_tl_hash(struct drbd_conf *mdev)
mdev->ee_hash = NULL;
mdev->ee_hash_s = 0;
- /* paranoia code */
- for (h = mdev->tl_hash; h < mdev->tl_hash + mdev->tl_hash_s; h++)
- if (h->first)
- dev_err(DEV, "ASSERT FAILED tl_hash[%u] == %p, expected NULL\n",
- (int)(h - mdev->tl_hash), h->first);
+ /* We may not have had the chance to wait for all locally pending
+ * application requests. The hlist_add_fake() prevents access after
+ * free on master bio completion. */
+ for (h = mdev->tl_hash; h < mdev->tl_hash + mdev->tl_hash_s; h++) {
+ struct drbd_request *req;
+ struct hlist_node *pos, *n;
+ hlist_for_each_entry_safe(req, pos, n, h, collision) {
+ hlist_del_init(&req->collision);
+ hlist_add_fake(&req->collision);
+ }
+ }
+
kfree(mdev->tl_hash);
mdev->tl_hash = NULL;
mdev->tl_hash_s = 0;
diff --git a/drivers/block/drbd/drbd_req.c b/drivers/block/drbd/drbd_req.c
index 9c5c84946b05..910335c30927 100644
--- a/drivers/block/drbd/drbd_req.c
+++ b/drivers/block/drbd/drbd_req.c
@@ -455,7 +455,7 @@ int __req_mod(struct drbd_request *req, enum drbd_req_event what,
req->rq_state |= RQ_LOCAL_COMPLETED;
req->rq_state &= ~RQ_LOCAL_PENDING;
- __drbd_chk_io_error(mdev, false);
+ __drbd_chk_io_error(mdev, DRBD_IO_ERROR);
_req_may_be_done_not_susp(req, m);
break;
@@ -472,12 +472,17 @@ int __req_mod(struct drbd_request *req, enum drbd_req_event what,
req->rq_state |= RQ_LOCAL_COMPLETED;
req->rq_state &= ~RQ_LOCAL_PENDING;
- D_ASSERT(!(req->rq_state & RQ_NET_MASK));
+ if (req->rq_state & RQ_LOCAL_ABORTED) {
+ _req_may_be_done(req, m);
+ break;
+ }
- __drbd_chk_io_error(mdev, false);
+ __drbd_chk_io_error(mdev, DRBD_IO_ERROR);
goto_queue_for_net_read:
+ D_ASSERT(!(req->rq_state & RQ_NET_MASK));
+
/* no point in retrying if there is no good remote data,
* or we have no connection. */
if (mdev->state.pdsk != D_UP_TO_DATE) {
@@ -765,6 +770,40 @@ static int drbd_may_do_local_read(struct drbd_conf *mdev, sector_t sector, int s
return 0 == drbd_bm_count_bits(mdev, sbnr, ebnr);
}
+static void maybe_pull_ahead(struct drbd_conf *mdev)
+{
+ int congested = 0;
+
+ /* If I don't even have good local storage, we can not reasonably try
+ * to pull ahead of the peer. We also need the local reference to make
+ * sure mdev->act_log is there.
+ * Note: caller has to make sure that net_conf is there.
+ */
+ if (!get_ldev_if_state(mdev, D_UP_TO_DATE))
+ return;
+
+ if (mdev->net_conf->cong_fill &&
+ atomic_read(&mdev->ap_in_flight) >= mdev->net_conf->cong_fill) {
+ dev_info(DEV, "Congestion-fill threshold reached\n");
+ congested = 1;
+ }
+
+ if (mdev->act_log->used >= mdev->net_conf->cong_extents) {
+ dev_info(DEV, "Congestion-extents threshold reached\n");
+ congested = 1;
+ }
+
+ if (congested) {
+ queue_barrier(mdev); /* last barrier, after mirrored writes */
+
+ if (mdev->net_conf->on_congestion == OC_PULL_AHEAD)
+ _drbd_set_state(_NS(mdev, conn, C_AHEAD), 0, NULL);
+ else /*mdev->net_conf->on_congestion == OC_DISCONNECT */
+ _drbd_set_state(_NS(mdev, conn, C_DISCONNECTING), 0, NULL);
+ }
+ put_ldev(mdev);
+}
+
static int drbd_make_request_common(struct drbd_conf *mdev, struct bio *bio, unsigned long start_time)
{
const int rw = bio_rw(bio);
@@ -972,29 +1011,8 @@ allocate_barrier:
_req_mod(req, queue_for_send_oos);
if (remote &&
- mdev->net_conf->on_congestion != OC_BLOCK && mdev->agreed_pro_version >= 96) {
- int congested = 0;
-
- if (mdev->net_conf->cong_fill &&
- atomic_read(&mdev->ap_in_flight) >= mdev->net_conf->cong_fill) {
- dev_info(DEV, "Congestion-fill threshold reached\n");
- congested = 1;
- }
-
- if (mdev->act_log->used >= mdev->net_conf->cong_extents) {
- dev_info(DEV, "Congestion-extents threshold reached\n");
- congested = 1;
- }
-
- if (congested) {
- queue_barrier(mdev); /* last barrier, after mirrored writes */
-
- if (mdev->net_conf->on_congestion == OC_PULL_AHEAD)
- _drbd_set_state(_NS(mdev, conn, C_AHEAD), 0, NULL);
- else /*mdev->net_conf->on_congestion == OC_DISCONNECT */
- _drbd_set_state(_NS(mdev, conn, C_DISCONNECTING), 0, NULL);
- }
- }
+ mdev->net_conf->on_congestion != OC_BLOCK && mdev->agreed_pro_version >= 96)
+ maybe_pull_ahead(mdev);
spin_unlock_irq(&mdev->req_lock);
kfree(b); /* if someone else has beaten us to it... */
@@ -1093,13 +1111,12 @@ void drbd_make_request(struct request_queue *q, struct bio *bio)
/*
* what we "blindly" assume:
*/
- D_ASSERT(bio->bi_size > 0);
D_ASSERT((bio->bi_size & 0x1ff) == 0);
/* to make some things easier, force alignment of requests within the
* granularity of our hash tables */
s_enr = bio->bi_sector >> HT_SHIFT;
- e_enr = (bio->bi_sector+(bio->bi_size>>9)-1) >> HT_SHIFT;
+ e_enr = bio->bi_size ? (bio->bi_sector+(bio->bi_size>>9)-1) >> HT_SHIFT : s_enr;
if (likely(s_enr == e_enr)) {
do {
@@ -1257,7 +1274,7 @@ void request_timer_fn(unsigned long data)
time_after(now, req->start_time + dt) &&
!time_in_range(now, mdev->last_reattach_jif, mdev->last_reattach_jif + dt)) {
dev_warn(DEV, "Local backing device failed to meet the disk-timeout\n");
- __drbd_chk_io_error(mdev, 1);
+ __drbd_chk_io_error(mdev, DRBD_FORCE_DETACH);
}
nt = (time_after(now, req->start_time + et) ? now : req->start_time) + et;
spin_unlock_irq(&mdev->req_lock);
diff --git a/drivers/block/drbd/drbd_worker.c b/drivers/block/drbd/drbd_worker.c
index 620c70ff2231..6bce2cc179d4 100644
--- a/drivers/block/drbd/drbd_worker.c
+++ b/drivers/block/drbd/drbd_worker.c
@@ -111,7 +111,7 @@ void drbd_endio_read_sec_final(struct drbd_epoch_entry *e) __releases(local)
if (list_empty(&mdev->read_ee))
wake_up(&mdev->ee_wait);
if (test_bit(__EE_WAS_ERROR, &e->flags))
- __drbd_chk_io_error(mdev, false);
+ __drbd_chk_io_error(mdev, DRBD_IO_ERROR);
spin_unlock_irqrestore(&mdev->req_lock, flags);
drbd_queue_work(&mdev->data.work, &e->w);
@@ -154,7 +154,7 @@ static void drbd_endio_write_sec_final(struct drbd_epoch_entry *e) __releases(lo
: list_empty(&mdev->active_ee);
if (test_bit(__EE_WAS_ERROR, &e->flags))
- __drbd_chk_io_error(mdev, false);
+ __drbd_chk_io_error(mdev, DRBD_IO_ERROR);
spin_unlock_irqrestore(&mdev->req_lock, flags);
if (is_syncer_req)
@@ -1501,14 +1501,6 @@ void drbd_start_resync(struct drbd_conf *mdev, enum drbd_conns side)
return;
}
- if (mdev->state.conn < C_AHEAD) {
- /* In case a previous resync run was aborted by an IO error/detach on the peer. */
- drbd_rs_cancel_all(mdev);
- /* This should be done when we abort the resync. We definitely do not
- want to have this for connections going back and forth between
- Ahead/Behind and SyncSource/SyncTarget */
- }
-
if (side == C_SYNC_TARGET) {
/* Since application IO was locked out during C_WF_BITMAP_T and
C_WF_SYNC_UUID we are still unmodified. Before going to C_SYNC_TARGET
diff --git a/drivers/block/floppy.c b/drivers/block/floppy.c
index 9d6ef68e2f1b..a7d6347aaa79 100644
--- a/drivers/block/floppy.c
+++ b/drivers/block/floppy.c
@@ -672,6 +672,7 @@ static void __reschedule_timeout(int drive, const char *message)
if (drive == current_reqD)
drive = current_drive;
+ __cancel_delayed_work(&fd_timeout);
if (drive < 0 || drive >= N_DRIVE) {
delay = 20UL * HZ;
diff --git a/drivers/block/loop.c b/drivers/block/loop.c
index bbca966f8f66..3bba65510d23 100644
--- a/drivers/block/loop.c
+++ b/drivers/block/loop.c
@@ -1597,14 +1597,12 @@ static int loop_add(struct loop_device **l, int i)
struct gendisk *disk;
int err;
+ err = -ENOMEM;
lo = kzalloc(sizeof(*lo), GFP_KERNEL);
- if (!lo) {
- err = -ENOMEM;
+ if (!lo)
goto out;
- }
- err = idr_pre_get(&loop_index_idr, GFP_KERNEL);
- if (err < 0)
+ if (!idr_pre_get(&loop_index_idr, GFP_KERNEL))
goto out_free_dev;
if (i >= 0) {
diff --git a/drivers/block/mtip32xx/mtip32xx.c b/drivers/block/mtip32xx/mtip32xx.c
index 304000c3d433..a8fddeb3d638 100644
--- a/drivers/block/mtip32xx/mtip32xx.c
+++ b/drivers/block/mtip32xx/mtip32xx.c
@@ -37,6 +37,7 @@
#include <linux/kthread.h>
#include <../drivers/ata/ahci.h>
#include <linux/export.h>
+#include <linux/debugfs.h>
#include "mtip32xx.h"
#define HW_CMD_SLOT_SZ (MTIP_MAX_COMMAND_SLOTS * 32)
@@ -85,6 +86,7 @@ static int instance;
* allocated in mtip_init().
*/
static int mtip_major;
+static struct dentry *dfs_parent;
static DEFINE_SPINLOCK(rssd_index_lock);
static DEFINE_IDA(rssd_index_ida);
@@ -294,18 +296,16 @@ static int hba_reset_nosleep(struct driver_data *dd)
*/
static inline void mtip_issue_ncq_command(struct mtip_port *port, int tag)
{
- unsigned long flags = 0;
-
atomic_set(&port->commands[tag].active, 1);
- spin_lock_irqsave(&port->cmd_issue_lock, flags);
+ spin_lock(&port->cmd_issue_lock);
writel((1 << MTIP_TAG_BIT(tag)),
port->s_active[MTIP_TAG_INDEX(tag)]);
writel((1 << MTIP_TAG_BIT(tag)),
port->cmd_issue[MTIP_TAG_INDEX(tag)]);
- spin_unlock_irqrestore(&port->cmd_issue_lock, flags);
+ spin_unlock(&port->cmd_issue_lock);
/* Set the command's timeout value.*/
port->commands[tag].comp_time = jiffies + msecs_to_jiffies(
@@ -436,8 +436,7 @@ static void mtip_init_port(struct mtip_port *port)
writel(0xFFFFFFFF, port->completed[i]);
/* Clear any pending interrupts for this port */
- writel(readl(port->dd->mmio + PORT_IRQ_STAT),
- port->dd->mmio + PORT_IRQ_STAT);
+ writel(readl(port->mmio + PORT_IRQ_STAT), port->mmio + PORT_IRQ_STAT);
/* Clear any pending interrupts on the HBA. */
writel(readl(port->dd->mmio + HOST_IRQ_STAT),
@@ -782,13 +781,24 @@ static void mtip_handle_tfe(struct driver_data *dd)
/* Stop the timer to prevent command timeouts. */
del_timer(&port->cmd_timer);
+ set_bit(MTIP_PF_EH_ACTIVE_BIT, &port->flags);
+
+ if (test_bit(MTIP_PF_IC_ACTIVE_BIT, &port->flags) &&
+ test_bit(MTIP_TAG_INTERNAL, port->allocated)) {
+ cmd = &port->commands[MTIP_TAG_INTERNAL];
+ dbg_printk(MTIP_DRV_NAME " TFE for the internal command\n");
+
+ atomic_inc(&cmd->active); /* active > 1 indicates error */
+ if (cmd->comp_data && cmd->comp_func) {
+ cmd->comp_func(port, MTIP_TAG_INTERNAL,
+ cmd->comp_data, PORT_IRQ_TF_ERR);
+ }
+ goto handle_tfe_exit;
+ }
/* clear the tag accumulator */
memset(tagaccum, 0, SLOTBITS_IN_LONGS * sizeof(long));
- /* Set eh_active */
- set_bit(MTIP_PF_EH_ACTIVE_BIT, &port->flags);
-
/* Loop through all the groups */
for (group = 0; group < dd->slot_groups; group++) {
completed = readl(port->completed[group]);
@@ -940,6 +950,7 @@ static void mtip_handle_tfe(struct driver_data *dd)
}
print_tags(dd, "reissued (TFE)", tagaccum, cmd_cnt);
+handle_tfe_exit:
/* clear eh_active */
clear_bit(MTIP_PF_EH_ACTIVE_BIT, &port->flags);
wake_up_interruptible(&port->svc_wait);
@@ -961,6 +972,8 @@ static inline void mtip_process_sdbf(struct driver_data *dd)
/* walk all bits in all slot groups */
for (group = 0; group < dd->slot_groups; group++) {
completed = readl(port->completed[group]);
+ if (!completed)
+ continue;
/* clear completed status register in the hardware.*/
writel(completed, port->completed[group]);
@@ -1329,22 +1342,6 @@ static int mtip_exec_internal_command(struct mtip_port *port,
}
rv = -EAGAIN;
}
-
- if (readl(port->cmd_issue[MTIP_TAG_INTERNAL])
- & (1 << MTIP_TAG_INTERNAL)) {
- dev_warn(&port->dd->pdev->dev,
- "Retiring internal command but CI is 1.\n");
- if (test_bit(MTIP_DDF_REMOVE_PENDING_BIT,
- &port->dd->dd_flag)) {
- hba_reset_nosleep(port->dd);
- rv = -ENXIO;
- } else {
- mtip_restart_port(port);
- rv = -EAGAIN;
- }
- goto exec_ic_exit;
- }
-
} else {
/* Spin for <timeout> checking if command still outstanding */
timeout = jiffies + msecs_to_jiffies(timeout);
@@ -1361,21 +1358,25 @@ static int mtip_exec_internal_command(struct mtip_port *port,
rv = -ENXIO;
goto exec_ic_exit;
}
+ if (readl(port->mmio + PORT_IRQ_STAT) & PORT_IRQ_ERR) {
+ atomic_inc(&int_cmd->active); /* error */
+ break;
+ }
}
+ }
- if (readl(port->cmd_issue[MTIP_TAG_INTERNAL])
+ if (atomic_read(&int_cmd->active) > 1) {
+ dev_err(&port->dd->pdev->dev,
+ "Internal command [%02X] failed\n", fis->command);
+ rv = -EIO;
+ }
+ if (readl(port->cmd_issue[MTIP_TAG_INTERNAL])
& (1 << MTIP_TAG_INTERNAL)) {
- dev_err(&port->dd->pdev->dev,
- "Internal command did not complete [atomic]\n");
+ rv = -ENXIO;
+ if (!test_bit(MTIP_DDF_REMOVE_PENDING_BIT,
+ &port->dd->dd_flag)) {
+ mtip_restart_port(port);
rv = -EAGAIN;
- if (test_bit(MTIP_DDF_REMOVE_PENDING_BIT,
- &port->dd->dd_flag)) {
- hba_reset_nosleep(port->dd);
- rv = -ENXIO;
- } else {
- mtip_restart_port(port);
- rv = -EAGAIN;
- }
}
}
exec_ic_exit:
@@ -1893,13 +1894,33 @@ static int exec_drive_command(struct mtip_port *port, u8 *command,
void __user *user_buffer)
{
struct host_to_dev_fis fis;
- struct host_to_dev_fis *reply = (port->rxfis + RX_FIS_D2H_REG);
+ struct host_to_dev_fis *reply;
+ u8 *buf = NULL;
+ dma_addr_t dma_addr = 0;
+ int rv = 0, xfer_sz = command[3];
+
+ if (xfer_sz) {
+ if (user_buffer)
+ return -EFAULT;
+
+ buf = dmam_alloc_coherent(&port->dd->pdev->dev,
+ ATA_SECT_SIZE * xfer_sz,
+ &dma_addr,
+ GFP_KERNEL);
+ if (!buf) {
+ dev_err(&port->dd->pdev->dev,
+ "Memory allocation failed (%d bytes)\n",
+ ATA_SECT_SIZE * xfer_sz);
+ return -ENOMEM;
+ }
+ memset(buf, 0, ATA_SECT_SIZE * xfer_sz);
+ }
/* Build the FIS. */
memset(&fis, 0, sizeof(struct host_to_dev_fis));
- fis.type = 0x27;
- fis.opts = 1 << 7;
- fis.command = command[0];
+ fis.type = 0x27;
+ fis.opts = 1 << 7;
+ fis.command = command[0];
fis.features = command[2];
fis.sect_count = command[3];
if (fis.command == ATA_CMD_SMART) {
@@ -1908,6 +1929,11 @@ static int exec_drive_command(struct mtip_port *port, u8 *command,
fis.cyl_hi = 0xC2;
}
+ if (xfer_sz)
+ reply = (port->rxfis + RX_FIS_PIO_SETUP);
+ else
+ reply = (port->rxfis + RX_FIS_D2H_REG);
+
dbg_printk(MTIP_DRV_NAME
" %s: User Command: cmd %x, sect %x, "
"feat %x, sectcnt %x\n",
@@ -1917,43 +1943,46 @@ static int exec_drive_command(struct mtip_port *port, u8 *command,
command[2],
command[3]);
- memset(port->sector_buffer, 0x00, ATA_SECT_SIZE);
-
/* Execute the command. */
if (mtip_exec_internal_command(port,
&fis,
5,
- port->sector_buffer_dma,
- (command[3] != 0) ? ATA_SECT_SIZE : 0,
+ (xfer_sz ? dma_addr : 0),
+ (xfer_sz ? ATA_SECT_SIZE * xfer_sz : 0),
0,
GFP_KERNEL,
MTIP_IOCTL_COMMAND_TIMEOUT_MS)
< 0) {
- return -1;
+ rv = -EFAULT;
+ goto exit_drive_command;
}
/* Collect the completion status. */
command[0] = reply->command; /* Status*/
command[1] = reply->features; /* Error*/
- command[2] = command[3];
+ command[2] = reply->sect_count;
dbg_printk(MTIP_DRV_NAME
" %s: Completion Status: stat %x, "
- "err %x, cmd %x\n",
+ "err %x, nsect %x\n",
__func__,
command[0],
command[1],
command[2]);
- if (user_buffer && command[3]) {
+ if (xfer_sz) {
if (copy_to_user(user_buffer,
- port->sector_buffer,
+ buf,
ATA_SECT_SIZE * command[3])) {
- return -EFAULT;
+ rv = -EFAULT;
+ goto exit_drive_command;
}
}
-
- return 0;
+exit_drive_command:
+ if (buf)
+ dmam_free_coherent(&port->dd->pdev->dev,
+ ATA_SECT_SIZE * xfer_sz, buf, dma_addr);
+ return rv;
}
/*
@@ -2003,6 +2032,32 @@ static unsigned int implicit_sector(unsigned char command,
return rv;
}
+static void mtip_set_timeout(struct host_to_dev_fis *fis, unsigned int *timeout)
+{
+ switch (fis->command) {
+ case ATA_CMD_DOWNLOAD_MICRO:
+ *timeout = 120000; /* 2 minutes */
+ break;
+ case ATA_CMD_SEC_ERASE_UNIT:
+ case 0xFC:
+ *timeout = 240000; /* 4 minutes */
+ break;
+ case ATA_CMD_STANDBYNOW1:
+ *timeout = 10000; /* 10 seconds */
+ break;
+ case 0xF7:
+ case 0xFA:
+ *timeout = 60000; /* 60 seconds */
+ break;
+ case ATA_CMD_SMART:
+ *timeout = 15000; /* 15 seconds */
+ break;
+ default:
+ *timeout = MTIP_IOCTL_COMMAND_TIMEOUT_MS;
+ break;
+ }
+}
+
/*
* Executes a taskfile
* See ide_taskfile_ioctl() for derivation
@@ -2023,7 +2078,7 @@ static int exec_drive_taskfile(struct driver_data *dd,
unsigned int taskin = 0;
unsigned int taskout = 0;
u8 nsect = 0;
- unsigned int timeout = MTIP_IOCTL_COMMAND_TIMEOUT_MS;
+ unsigned int timeout;
unsigned int force_single_sector;
unsigned int transfer_size;
unsigned long task_file_data;
@@ -2153,32 +2208,7 @@ static int exec_drive_taskfile(struct driver_data *dd,
fis.lba_hi,
fis.device);
- switch (fis.command) {
- case ATA_CMD_DOWNLOAD_MICRO:
- /* Change timeout for Download Microcode to 2 minutes */
- timeout = 120000;
- break;
- case ATA_CMD_SEC_ERASE_UNIT:
- /* Change timeout for Security Erase Unit to 4 minutes.*/
- timeout = 240000;
- break;
- case ATA_CMD_STANDBYNOW1:
- /* Change timeout for standby immediate to 10 seconds.*/
- timeout = 10000;
- break;
- case 0xF7:
- case 0xFA:
- /* Change timeout for vendor unique command to 10 secs */
- timeout = 10000;
- break;
- case ATA_CMD_SMART:
- /* Change timeout for vendor unique command to 15 secs */
- timeout = 15000;
- break;
- default:
- timeout = MTIP_IOCTL_COMMAND_TIMEOUT_MS;
- break;
- }
+ mtip_set_timeout(&fis, &timeout);
/* Determine the correct transfer size.*/
if (force_single_sector)
@@ -2295,13 +2325,12 @@ static int mtip_hw_ioctl(struct driver_data *dd, unsigned int cmd,
{
switch (cmd) {
case HDIO_GET_IDENTITY:
- if (mtip_get_identify(dd->port, (void __user *) arg) < 0) {
- dev_warn(&dd->pdev->dev,
- "Unable to read identity\n");
- return -EIO;
- }
-
+ {
+ if (copy_to_user((void __user *)arg, dd->port->identify,
+ sizeof(u16) * ATA_ID_WORDS))
+ return -EFAULT;
break;
+ }
case HDIO_DRIVE_CMD:
{
u8 drive_command[4];
@@ -2519,7 +2548,7 @@ static struct scatterlist *mtip_hw_get_scatterlist(struct driver_data *dd,
}
/*
- * Sysfs register/status dump.
+ * Sysfs status dump.
*
* @dev Pointer to the device structure, passed by the kernrel.
* @attr Pointer to the device_attribute structure passed by the kernel.
@@ -2528,72 +2557,138 @@ static struct scatterlist *mtip_hw_get_scatterlist(struct driver_data *dd,
* return value
* The size, in bytes, of the data copied into buf.
*/
-static ssize_t mtip_hw_show_registers(struct device *dev,
+static ssize_t mtip_hw_show_status(struct device *dev,
struct device_attribute *attr,
char *buf)
{
- u32 group_allocated;
struct driver_data *dd = dev_to_disk(dev)->private_data;
int size = 0;
+
+ if (test_bit(MTIP_DDF_OVER_TEMP_BIT, &dd->dd_flag))
+ size += sprintf(buf, "%s", "thermal_shutdown\n");
+ else if (test_bit(MTIP_DDF_WRITE_PROTECT_BIT, &dd->dd_flag))
+ size += sprintf(buf, "%s", "write_protect\n");
+ else
+ size += sprintf(buf, "%s", "online\n");
+
+ return size;
+}
+
+static DEVICE_ATTR(status, S_IRUGO, mtip_hw_show_status, NULL);
+
+static ssize_t mtip_hw_read_registers(struct file *f, char __user *ubuf,
+ size_t len, loff_t *offset)
+{
+ struct driver_data *dd = (struct driver_data *)f->private_data;
+ char buf[MTIP_DFS_MAX_BUF_SIZE];
+ u32 group_allocated;
+ int size = *offset;
int n;
- size += sprintf(&buf[size], "S ACTive:\n");
+ if (!len || size)
+ return 0;
+
+ if (size < 0)
+ return -EINVAL;
+
+ size += sprintf(&buf[size], "H/ S ACTive : [ 0x");
- for (n = 0; n < dd->slot_groups; n++)
- size += sprintf(&buf[size], "0x%08x\n",
+ for (n = dd->slot_groups-1; n >= 0; n--)
+ size += sprintf(&buf[size], "%08X ",
readl(dd->port->s_active[n]));
- size += sprintf(&buf[size], "Command Issue:\n");
+ size += sprintf(&buf[size], "]\n");
+ size += sprintf(&buf[size], "H/ Command Issue : [ 0x");
- for (n = 0; n < dd->slot_groups; n++)
- size += sprintf(&buf[size], "0x%08x\n",
+ for (n = dd->slot_groups-1; n >= 0; n--)
+ size += sprintf(&buf[size], "%08X ",
readl(dd->port->cmd_issue[n]));
- size += sprintf(&buf[size], "Allocated:\n");
+ size += sprintf(&buf[size], "]\n");
+ size += sprintf(&buf[size], "H/ Completed : [ 0x");
+
+ for (n = dd->slot_groups-1; n >= 0; n--)
+ size += sprintf(&buf[size], "%08X ",
+ readl(dd->port->completed[n]));
- for (n = 0; n < dd->slot_groups; n++) {
+ size += sprintf(&buf[size], "]\n");
+ size += sprintf(&buf[size], "H/ PORT IRQ STAT : [ 0x%08X ]\n",
+ readl(dd->port->mmio + PORT_IRQ_STAT));
+ size += sprintf(&buf[size], "H/ HOST IRQ STAT : [ 0x%08X ]\n",
+ readl(dd->mmio + HOST_IRQ_STAT));
+ size += sprintf(&buf[size], "\n");
+
+ size += sprintf(&buf[size], "L/ Allocated : [ 0x");
+
+ for (n = dd->slot_groups-1; n >= 0; n--) {
if (sizeof(long) > sizeof(u32))
group_allocated =
dd->port->allocated[n/2] >> (32*(n&1));
else
group_allocated = dd->port->allocated[n];
- size += sprintf(&buf[size], "0x%08x\n",
- group_allocated);
+ size += sprintf(&buf[size], "%08X ", group_allocated);
}
+ size += sprintf(&buf[size], "]\n");
- size += sprintf(&buf[size], "Completed:\n");
+ size += sprintf(&buf[size], "L/ Commands in Q : [ 0x");
- for (n = 0; n < dd->slot_groups; n++)
- size += sprintf(&buf[size], "0x%08x\n",
- readl(dd->port->completed[n]));
+ for (n = dd->slot_groups-1; n >= 0; n--) {
+ if (sizeof(long) > sizeof(u32))
+ group_allocated =
+ dd->port->cmds_to_issue[n/2] >> (32*(n&1));
+ else
+ group_allocated = dd->port->cmds_to_issue[n];
+ size += sprintf(&buf[size], "%08X ", group_allocated);
+ }
+ size += sprintf(&buf[size], "]\n");
- size += sprintf(&buf[size], "PORT IRQ STAT : 0x%08x\n",
- readl(dd->port->mmio + PORT_IRQ_STAT));
- size += sprintf(&buf[size], "HOST IRQ STAT : 0x%08x\n",
- readl(dd->mmio + HOST_IRQ_STAT));
+ *offset = size <= len ? size : len;
+ size = copy_to_user(ubuf, buf, *offset);
+ if (size)
+ return -EFAULT;
- return size;
+ return *offset;
}
-static ssize_t mtip_hw_show_status(struct device *dev,
- struct device_attribute *attr,
- char *buf)
+static ssize_t mtip_hw_read_flags(struct file *f, char __user *ubuf,
+ size_t len, loff_t *offset)
{
- struct driver_data *dd = dev_to_disk(dev)->private_data;
- int size = 0;
+ struct driver_data *dd = (struct driver_data *)f->private_data;
+ char buf[MTIP_DFS_MAX_BUF_SIZE];
+ int size = *offset;
- if (test_bit(MTIP_DDF_OVER_TEMP_BIT, &dd->dd_flag))
- size += sprintf(buf, "%s", "thermal_shutdown\n");
- else if (test_bit(MTIP_DDF_WRITE_PROTECT_BIT, &dd->dd_flag))
- size += sprintf(buf, "%s", "write_protect\n");
- else
- size += sprintf(buf, "%s", "online\n");
+ if (!len || size)
+ return 0;
- return size;
+ if (size < 0)
+ return -EINVAL;
+
+ size += sprintf(&buf[size], "Flag-port : [ %08lX ]\n",
+ dd->port->flags);
+ size += sprintf(&buf[size], "Flag-dd : [ %08lX ]\n",
+ dd->dd_flag);
+
+ *offset = size <= len ? size : len;
+ size = copy_to_user(ubuf, buf, *offset);
+ if (size)
+ return -EFAULT;
+
+ return *offset;
}
-static DEVICE_ATTR(registers, S_IRUGO, mtip_hw_show_registers, NULL);
-static DEVICE_ATTR(status, S_IRUGO, mtip_hw_show_status, NULL);
+static const struct file_operations mtip_regs_fops = {
+ .owner = THIS_MODULE,
+ .open = simple_open,
+ .read = mtip_hw_read_registers,
+ .llseek = no_llseek,
+};
+
+static const struct file_operations mtip_flags_fops = {
+ .owner = THIS_MODULE,
+ .open = simple_open,
+ .read = mtip_hw_read_flags,
+ .llseek = no_llseek,
+};
/*
* Create the sysfs related attributes.
@@ -2610,9 +2705,6 @@ static int mtip_hw_sysfs_init(struct driver_data *dd, struct kobject *kobj)
if (!kobj || !dd)
return -EINVAL;
- if (sysfs_create_file(kobj, &dev_attr_registers.attr))
- dev_warn(&dd->pdev->dev,
- "Error creating 'registers' sysfs entry\n");
if (sysfs_create_file(kobj, &dev_attr_status.attr))
dev_warn(&dd->pdev->dev,
"Error creating 'status' sysfs entry\n");
@@ -2634,12 +2726,39 @@ static int mtip_hw_sysfs_exit(struct driver_data *dd, struct kobject *kobj)
if (!kobj || !dd)
return -EINVAL;
- sysfs_remove_file(kobj, &dev_attr_registers.attr);
sysfs_remove_file(kobj, &dev_attr_status.attr);
return 0;
}
+static int mtip_hw_debugfs_init(struct driver_data *dd)
+{
+ if (!dfs_parent)
+ return -1;
+
+ dd->dfs_node = debugfs_create_dir(dd->disk->disk_name, dfs_parent);
+ if (IS_ERR_OR_NULL(dd->dfs_node)) {
+ dev_warn(&dd->pdev->dev,
+ "Error creating node %s under debugfs\n",
+ dd->disk->disk_name);
+ dd->dfs_node = NULL;
+ return -1;
+ }
+
+ debugfs_create_file("flags", S_IRUGO, dd->dfs_node, dd,
+ &mtip_flags_fops);
+ debugfs_create_file("registers", S_IRUGO, dd->dfs_node, dd,
+ &mtip_regs_fops);
+
+ return 0;
+}
+
+static void mtip_hw_debugfs_exit(struct driver_data *dd)
+{
+ debugfs_remove_recursive(dd->dfs_node);
+}
+
+
/*
* Perform any init/resume time hardware setup
*
@@ -3634,7 +3753,10 @@ skip_create_disk:
set_bit(QUEUE_FLAG_NONROT, &dd->queue->queue_flags);
blk_queue_max_segments(dd->queue, MTIP_MAX_SG);
blk_queue_physical_block_size(dd->queue, 4096);
+ blk_queue_max_hw_sectors(dd->queue, 0xffff);
+ blk_queue_max_segment_size(dd->queue, 0x400000);
blk_queue_io_min(dd->queue, 4096);
+
/*
* write back cache is not supported in the device. FUA depends on
* write back cache support, hence setting flush support to zero.
@@ -3662,6 +3784,7 @@ skip_create_disk:
mtip_hw_sysfs_init(dd, kobj);
kobject_put(kobj);
}
+ mtip_hw_debugfs_init(dd);
if (dd->mtip_svc_handler) {
set_bit(MTIP_DDF_INIT_DONE_BIT, &dd->dd_flag);
@@ -3687,6 +3810,8 @@ start_service_thread:
return rv;
kthread_run_error:
+ mtip_hw_debugfs_exit(dd);
+
/* Delete our gendisk. This also removes the device from /dev */
del_gendisk(dd->disk);
@@ -3737,6 +3862,7 @@ static int mtip_block_remove(struct driver_data *dd)
kobject_put(kobj);
}
}
+ mtip_hw_debugfs_exit(dd);
/*
* Delete our gendisk structure. This also removes the device
@@ -4084,10 +4210,20 @@ static int __init mtip_init(void)
}
mtip_major = error;
+ if (!dfs_parent) {
+ dfs_parent = debugfs_create_dir("rssd", NULL);
+ if (IS_ERR_OR_NULL(dfs_parent)) {
+ printk(KERN_WARNING "Error creating debugfs parent\n");
+ dfs_parent = NULL;
+ }
+ }
+
/* Register our PCI operations. */
error = pci_register_driver(&mtip_pci_driver);
- if (error)
+ if (error) {
+ debugfs_remove(dfs_parent);
unregister_blkdev(mtip_major, MTIP_DRV_NAME);
+ }
return error;
}
@@ -4104,6 +4240,8 @@ static int __init mtip_init(void)
*/
static void __exit mtip_exit(void)
{
+ debugfs_remove_recursive(dfs_parent);
+
/* Release the allocated major block device number. */
unregister_blkdev(mtip_major, MTIP_DRV_NAME);
diff --git a/drivers/block/mtip32xx/mtip32xx.h b/drivers/block/mtip32xx/mtip32xx.h
index 4ef58336310a..f51fc23d17bb 100644
--- a/drivers/block/mtip32xx/mtip32xx.h
+++ b/drivers/block/mtip32xx/mtip32xx.h
@@ -26,7 +26,6 @@
#include <linux/ata.h>
#include <linux/interrupt.h>
#include <linux/genhd.h>
-#include <linux/version.h>
/* Offset of Subsystem Device ID in pci confoguration space */
#define PCI_SUBSYSTEM_DEVICEID 0x2E
@@ -111,35 +110,39 @@
#define dbg_printk(format, arg...)
#endif
+#define MTIP_DFS_MAX_BUF_SIZE 1024
+
#define __force_bit2int (unsigned int __force)
-/* below are bit numbers in 'flags' defined in mtip_port */
-#define MTIP_PF_IC_ACTIVE_BIT 0 /* pio/ioctl */
-#define MTIP_PF_EH_ACTIVE_BIT 1 /* error handling */
-#define MTIP_PF_SE_ACTIVE_BIT 2 /* secure erase */
-#define MTIP_PF_DM_ACTIVE_BIT 3 /* download microcde */
-#define MTIP_PF_PAUSE_IO ((1 << MTIP_PF_IC_ACTIVE_BIT) | \
+enum {
+ /* below are bit numbers in 'flags' defined in mtip_port */
+ MTIP_PF_IC_ACTIVE_BIT = 0, /* pio/ioctl */
+ MTIP_PF_EH_ACTIVE_BIT = 1, /* error handling */
+ MTIP_PF_SE_ACTIVE_BIT = 2, /* secure erase */
+ MTIP_PF_DM_ACTIVE_BIT = 3, /* download microcde */
+ MTIP_PF_PAUSE_IO = ((1 << MTIP_PF_IC_ACTIVE_BIT) | \
(1 << MTIP_PF_EH_ACTIVE_BIT) | \
(1 << MTIP_PF_SE_ACTIVE_BIT) | \
- (1 << MTIP_PF_DM_ACTIVE_BIT))
-
-#define MTIP_PF_SVC_THD_ACTIVE_BIT 4
-#define MTIP_PF_ISSUE_CMDS_BIT 5
-#define MTIP_PF_REBUILD_BIT 6
-#define MTIP_PF_SVC_THD_STOP_BIT 8
-
-/* below are bit numbers in 'dd_flag' defined in driver_data */
-#define MTIP_DDF_REMOVE_PENDING_BIT 1
-#define MTIP_DDF_OVER_TEMP_BIT 2
-#define MTIP_DDF_WRITE_PROTECT_BIT 3
-#define MTIP_DDF_STOP_IO ((1 << MTIP_DDF_REMOVE_PENDING_BIT) | \
+ (1 << MTIP_PF_DM_ACTIVE_BIT)),
+
+ MTIP_PF_SVC_THD_ACTIVE_BIT = 4,
+ MTIP_PF_ISSUE_CMDS_BIT = 5,
+ MTIP_PF_REBUILD_BIT = 6,
+ MTIP_PF_SVC_THD_STOP_BIT = 8,
+
+ /* below are bit numbers in 'dd_flag' defined in driver_data */
+ MTIP_DDF_REMOVE_PENDING_BIT = 1,
+ MTIP_DDF_OVER_TEMP_BIT = 2,
+ MTIP_DDF_WRITE_PROTECT_BIT = 3,
+ MTIP_DDF_STOP_IO = ((1 << MTIP_DDF_REMOVE_PENDING_BIT) | \
(1 << MTIP_DDF_OVER_TEMP_BIT) | \
- (1 << MTIP_DDF_WRITE_PROTECT_BIT))
+ (1 << MTIP_DDF_WRITE_PROTECT_BIT)),
-#define MTIP_DDF_CLEANUP_BIT 5
-#define MTIP_DDF_RESUME_BIT 6
-#define MTIP_DDF_INIT_DONE_BIT 7
-#define MTIP_DDF_REBUILD_FAILED_BIT 8
+ MTIP_DDF_CLEANUP_BIT = 5,
+ MTIP_DDF_RESUME_BIT = 6,
+ MTIP_DDF_INIT_DONE_BIT = 7,
+ MTIP_DDF_REBUILD_FAILED_BIT = 8,
+};
__packed struct smart_attr{
u8 attr_id;
@@ -445,6 +448,8 @@ struct driver_data {
unsigned long dd_flag; /* NOTE: use atomic bit operations on this */
struct task_struct *mtip_svc_handler; /* task_struct of svc thd */
+
+ struct dentry *dfs_node;
};
#endif
diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c
index 061427a75d37..8957b9f0cfad 100644
--- a/drivers/block/nbd.c
+++ b/drivers/block/nbd.c
@@ -481,7 +481,7 @@ static void nbd_handle_req(struct nbd_device *nbd, struct request *req)
nbd_end_request(req);
} else {
spin_lock(&nbd->queue_lock);
- list_add(&req->queuelist, &nbd->queue_head);
+ list_add_tail(&req->queuelist, &nbd->queue_head);
spin_unlock(&nbd->queue_lock);
}
diff --git a/drivers/block/rbd.c b/drivers/block/rbd.c
index 013c7a549fb6..8f428a8ab003 100644
--- a/drivers/block/rbd.c
+++ b/drivers/block/rbd.c
@@ -141,7 +141,7 @@ struct rbd_request {
struct rbd_snap {
struct device dev;
const char *name;
- size_t size;
+ u64 size;
struct list_head node;
u64 id;
};
@@ -175,8 +175,7 @@ struct rbd_device {
/* protects updating the header */
struct rw_semaphore header_rwsem;
char snap_name[RBD_MAX_SNAP_NAME_LEN];
- u32 cur_snap; /* index+1 of current snapshot within snap context
- 0 - for the head */
+ u64 snap_id; /* current snapshot id */
int read_only;
struct list_head node;
@@ -241,7 +240,7 @@ static void rbd_put_dev(struct rbd_device *rbd_dev)
put_device(&rbd_dev->dev);
}
-static int __rbd_update_snaps(struct rbd_device *rbd_dev);
+static int __rbd_refresh_header(struct rbd_device *rbd_dev);
static int rbd_open(struct block_device *bdev, fmode_t mode)
{
@@ -450,7 +449,9 @@ static void rbd_client_release(struct kref *kref)
struct rbd_client *rbdc = container_of(kref, struct rbd_client, kref);
dout("rbd_release_client %p\n", rbdc);
+ spin_lock(&rbd_client_list_lock);
list_del(&rbdc->node);
+ spin_unlock(&rbd_client_list_lock);
ceph_destroy_client(rbdc->client);
kfree(rbdc->rbd_opts);
@@ -463,9 +464,7 @@ static void rbd_client_release(struct kref *kref)
*/
static void rbd_put_client(struct rbd_device *rbd_dev)
{
- spin_lock(&rbd_client_list_lock);
kref_put(&rbd_dev->rbd_client->kref, rbd_client_release);
- spin_unlock(&rbd_client_list_lock);
rbd_dev->rbd_client = NULL;
}
@@ -487,18 +486,20 @@ static void rbd_coll_release(struct kref *kref)
*/
static int rbd_header_from_disk(struct rbd_image_header *header,
struct rbd_image_header_ondisk *ondisk,
- int allocated_snaps,
+ u32 allocated_snaps,
gfp_t gfp_flags)
{
- int i;
- u32 snap_count;
+ u32 i, snap_count;
if (memcmp(ondisk, RBD_HEADER_TEXT, sizeof(RBD_HEADER_TEXT)))
return -ENXIO;
snap_count = le32_to_cpu(ondisk->snap_count);
+ if (snap_count > (UINT_MAX - sizeof(struct ceph_snap_context))
+ / sizeof (*ondisk))
+ return -EINVAL;
header->snapc = kmalloc(sizeof(struct ceph_snap_context) +
- snap_count * sizeof (*ondisk),
+ snap_count * sizeof(u64),
gfp_flags);
if (!header->snapc)
return -ENOMEM;
@@ -506,11 +507,11 @@ static int rbd_header_from_disk(struct rbd_image_header *header,
header->snap_names_len = le64_to_cpu(ondisk->snap_names_len);
if (snap_count) {
header->snap_names = kmalloc(header->snap_names_len,
- GFP_KERNEL);
+ gfp_flags);
if (!header->snap_names)
goto err_snapc;
header->snap_sizes = kmalloc(snap_count * sizeof(u64),
- GFP_KERNEL);
+ gfp_flags);
if (!header->snap_sizes)
goto err_names;
} else {
@@ -552,21 +553,6 @@ err_snapc:
return -ENOMEM;
}
-static int snap_index(struct rbd_image_header *header, int snap_num)
-{
- return header->total_snaps - snap_num;
-}
-
-static u64 cur_snap_id(struct rbd_device *rbd_dev)
-{
- struct rbd_image_header *header = &rbd_dev->header;
-
- if (!rbd_dev->cur_snap)
- return 0;
-
- return header->snapc->snaps[snap_index(header, rbd_dev->cur_snap)];
-}
-
static int snap_by_name(struct rbd_image_header *header, const char *snap_name,
u64 *seq, u64 *size)
{
@@ -605,7 +591,7 @@ static int rbd_header_set_snap(struct rbd_device *dev, u64 *size)
snapc->seq = header->snap_seq;
else
snapc->seq = 0;
- dev->cur_snap = 0;
+ dev->snap_id = CEPH_NOSNAP;
dev->read_only = 0;
if (size)
*size = header->image_size;
@@ -613,8 +599,7 @@ static int rbd_header_set_snap(struct rbd_device *dev, u64 *size)
ret = snap_by_name(header, dev->snap_name, &snapc->seq, size);
if (ret < 0)
goto done;
-
- dev->cur_snap = header->total_snaps - ret;
+ dev->snap_id = snapc->seq;
dev->read_only = 1;
}
@@ -935,7 +920,6 @@ static int rbd_do_request(struct request *rq,
layout->fl_stripe_unit = cpu_to_le32(1 << RBD_MAX_OBJ_ORDER);
layout->fl_stripe_count = cpu_to_le32(1);
layout->fl_object_size = cpu_to_le32(1 << RBD_MAX_OBJ_ORDER);
- layout->fl_pg_preferred = cpu_to_le32(-1);
layout->fl_pg_pool = cpu_to_le32(dev->poolid);
ceph_calc_raw_layout(osdc, layout, snapid, ofs, &len, &bno,
req, ops);
@@ -993,7 +977,7 @@ static void rbd_req_cb(struct ceph_osd_request *req, struct ceph_msg *msg)
op = (void *)(replyhead + 1);
rc = le32_to_cpu(replyhead->result);
bytes = le64_to_cpu(op->extent.length);
- read_op = (le32_to_cpu(op->op) == CEPH_OSD_OP_READ);
+ read_op = (le16_to_cpu(op->op) == CEPH_OSD_OP_READ);
dout("rbd_req_cb bytes=%lld readop=%d rc=%d\n", bytes, read_op, rc);
@@ -1168,7 +1152,7 @@ static int rbd_req_read(struct request *rq,
int coll_index)
{
return rbd_do_op(rq, rbd_dev, NULL,
- (snapid ? snapid : CEPH_NOSNAP),
+ snapid,
CEPH_OSD_OP_READ,
CEPH_OSD_FLAG_READ,
2,
@@ -1187,7 +1171,7 @@ static int rbd_req_sync_read(struct rbd_device *dev,
u64 *ver)
{
return rbd_req_sync_op(dev, NULL,
- (snapid ? snapid : CEPH_NOSNAP),
+ snapid,
CEPH_OSD_OP_READ,
CEPH_OSD_FLAG_READ,
NULL,
@@ -1238,7 +1222,7 @@ static void rbd_watch_cb(u64 ver, u64 notify_id, u8 opcode, void *data)
dout("rbd_watch_cb %s notify_id=%lld opcode=%d\n", dev->obj_md_name,
notify_id, (int)opcode);
mutex_lock_nested(&ctl_mutex, SINGLE_DEPTH_NESTING);
- rc = __rbd_update_snaps(dev);
+ rc = __rbd_refresh_header(dev);
mutex_unlock(&ctl_mutex);
if (rc)
pr_warning(RBD_DRV_NAME "%d got notification but failed to "
@@ -1521,7 +1505,7 @@ static void rbd_rq_fn(struct request_queue *q)
coll, cur_seg);
else
rbd_req_read(rq, rbd_dev,
- cur_snap_id(rbd_dev),
+ rbd_dev->snap_id,
ofs,
op_size, bio,
coll, cur_seg);
@@ -1592,7 +1576,7 @@ static int rbd_read_header(struct rbd_device *rbd_dev,
{
ssize_t rc;
struct rbd_image_header_ondisk *dh;
- int snap_count = 0;
+ u32 snap_count = 0;
u64 ver;
size_t len;
@@ -1656,7 +1640,7 @@ static int rbd_header_add_snap(struct rbd_device *dev,
struct ceph_mon_client *monc;
/* we should create a snapshot only if we're pointing at the head */
- if (dev->cur_snap)
+ if (dev->snap_id != CEPH_NOSNAP)
return -EINVAL;
monc = &dev->rbd_client->client->monc;
@@ -1683,7 +1667,9 @@ static int rbd_header_add_snap(struct rbd_device *dev,
if (ret < 0)
return ret;
- dev->header.snapc->seq = new_snapid;
+ down_write(&dev->header_rwsem);
+ dev->header.snapc->seq = new_snapid;
+ up_write(&dev->header_rwsem);
return 0;
bad:
@@ -1703,7 +1689,7 @@ static void __rbd_remove_all_snaps(struct rbd_device *rbd_dev)
/*
* only read the first part of the ondisk header, without the snaps info
*/
-static int __rbd_update_snaps(struct rbd_device *rbd_dev)
+static int __rbd_refresh_header(struct rbd_device *rbd_dev)
{
int ret;
struct rbd_image_header h;
@@ -1890,7 +1876,7 @@ static ssize_t rbd_image_refresh(struct device *dev,
mutex_lock_nested(&ctl_mutex, SINGLE_DEPTH_NESTING);
- rc = __rbd_update_snaps(rbd_dev);
+ rc = __rbd_refresh_header(rbd_dev);
if (rc < 0)
ret = rc;
@@ -1949,7 +1935,7 @@ static ssize_t rbd_snap_size_show(struct device *dev,
{
struct rbd_snap *snap = container_of(dev, struct rbd_snap, dev);
- return sprintf(buf, "%zd\n", snap->size);
+ return sprintf(buf, "%llu\n", (unsigned long long)snap->size);
}
static ssize_t rbd_snap_id_show(struct device *dev,
@@ -1958,7 +1944,7 @@ static ssize_t rbd_snap_id_show(struct device *dev,
{
struct rbd_snap *snap = container_of(dev, struct rbd_snap, dev);
- return sprintf(buf, "%llu\n", (unsigned long long) snap->id);
+ return sprintf(buf, "%llu\n", (unsigned long long)snap->id);
}
static DEVICE_ATTR(snap_size, S_IRUGO, rbd_snap_size_show, NULL);
@@ -2173,7 +2159,7 @@ static int rbd_init_watch_dev(struct rbd_device *rbd_dev)
rbd_dev->header.obj_version);
if (ret == -ERANGE) {
mutex_lock_nested(&ctl_mutex, SINGLE_DEPTH_NESTING);
- rc = __rbd_update_snaps(rbd_dev);
+ rc = __rbd_refresh_header(rbd_dev);
mutex_unlock(&ctl_mutex);
if (rc < 0)
return rc;
@@ -2558,7 +2544,7 @@ static ssize_t rbd_snap_add(struct device *dev,
if (ret < 0)
goto err_unlock;
- ret = __rbd_update_snaps(rbd_dev);
+ ret = __rbd_refresh_header(rbd_dev);
if (ret < 0)
goto err_unlock;
diff --git a/drivers/block/umem.c b/drivers/block/umem.c
index aa2712060bfb..eb0d8216f557 100644
--- a/drivers/block/umem.c
+++ b/drivers/block/umem.c
@@ -513,6 +513,21 @@ static void process_page(unsigned long data)
}
}
+static void mm_unplug(struct blk_plug_cb *cb, bool from_schedule)
+{
+ struct cardinfo *card = cb->data;
+
+ spin_lock_irq(&card->lock);
+ activate(card);
+ spin_unlock_irq(&card->lock);
+ kfree(cb);
+}
+
+static int mm_check_plugged(struct cardinfo *card)
+{
+ return !!blk_check_plugged(mm_unplug, card, sizeof(struct blk_plug_cb));
+}
+
static void mm_make_request(struct request_queue *q, struct bio *bio)
{
struct cardinfo *card = q->queuedata;
@@ -523,6 +538,8 @@ static void mm_make_request(struct request_queue *q, struct bio *bio)
*card->biotail = bio;
bio->bi_next = NULL;
card->biotail = &bio->bi_next;
+ if (bio->bi_rw & REQ_SYNC || !mm_check_plugged(card))
+ activate(card);
spin_unlock_irq(&card->lock);
return;
diff --git a/drivers/block/xen-blkback/common.h b/drivers/block/xen-blkback/common.h
index 773cf27dc23f..9ad3b5ec1dc1 100644
--- a/drivers/block/xen-blkback/common.h
+++ b/drivers/block/xen-blkback/common.h
@@ -257,6 +257,7 @@ static inline void blkif_get_x86_32_req(struct blkif_request *dst,
break;
case BLKIF_OP_DISCARD:
dst->u.discard.flag = src->u.discard.flag;
+ dst->u.discard.id = src->u.discard.id;
dst->u.discard.sector_number = src->u.discard.sector_number;
dst->u.discard.nr_sectors = src->u.discard.nr_sectors;
break;
@@ -287,6 +288,7 @@ static inline void blkif_get_x86_64_req(struct blkif_request *dst,
break;
case BLKIF_OP_DISCARD:
dst->u.discard.flag = src->u.discard.flag;
+ dst->u.discard.id = src->u.discard.id;
dst->u.discard.sector_number = src->u.discard.sector_number;
dst->u.discard.nr_sectors = src->u.discard.nr_sectors;
break;
diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c
index 60eed4bdd2e4..e4fb3374dcd2 100644
--- a/drivers/block/xen-blkfront.c
+++ b/drivers/block/xen-blkfront.c
@@ -141,14 +141,36 @@ static int get_id_from_freelist(struct blkfront_info *info)
return free;
}
-static void add_id_to_freelist(struct blkfront_info *info,
+static int add_id_to_freelist(struct blkfront_info *info,
unsigned long id)
{
+ if (info->shadow[id].req.u.rw.id != id)
+ return -EINVAL;
+ if (info->shadow[id].request == NULL)
+ return -EINVAL;
info->shadow[id].req.u.rw.id = info->shadow_free;
info->shadow[id].request = NULL;
info->shadow_free = id;
+ return 0;
}
+static const char *op_name(int op)
+{
+ static const char *const names[] = {
+ [BLKIF_OP_READ] = "read",
+ [BLKIF_OP_WRITE] = "write",
+ [BLKIF_OP_WRITE_BARRIER] = "barrier",
+ [BLKIF_OP_FLUSH_DISKCACHE] = "flush",
+ [BLKIF_OP_DISCARD] = "discard" };
+
+ if (op < 0 || op >= ARRAY_SIZE(names))
+ return "unknown";
+
+ if (!names[op])
+ return "reserved";
+
+ return names[op];
+}
static int xlbd_reserve_minors(unsigned int minor, unsigned int nr)
{
unsigned int end = minor + nr;
@@ -746,20 +768,36 @@ static irqreturn_t blkif_interrupt(int irq, void *dev_id)
bret = RING_GET_RESPONSE(&info->ring, i);
id = bret->id;
+ /*
+ * The backend has messed up and given us an id that we would
+ * never have given to it (we stamp it up to BLK_RING_SIZE -
+ * look in get_id_from_freelist.
+ */
+ if (id >= BLK_RING_SIZE) {
+ WARN(1, "%s: response to %s has incorrect id (%ld)\n",
+ info->gd->disk_name, op_name(bret->operation), id);
+ /* We can't safely get the 'struct request' as
+ * the id is busted. */
+ continue;
+ }
req = info->shadow[id].request;
if (bret->operation != BLKIF_OP_DISCARD)
blkif_completion(&info->shadow[id]);
- add_id_to_freelist(info, id);
+ if (add_id_to_freelist(info, id)) {
+ WARN(1, "%s: response to %s (id %ld) couldn't be recycled!\n",
+ info->gd->disk_name, op_name(bret->operation), id);
+ continue;
+ }
error = (bret->status == BLKIF_RSP_OKAY) ? 0 : -EIO;
switch (bret->operation) {
case BLKIF_OP_DISCARD:
if (unlikely(bret->status == BLKIF_RSP_EOPNOTSUPP)) {
struct request_queue *rq = info->rq;
- printk(KERN_WARNING "blkfront: %s: discard op failed\n",
- info->gd->disk_name);
+ printk(KERN_WARNING "blkfront: %s: %s op failed\n",
+ info->gd->disk_name, op_name(bret->operation));
error = -EOPNOTSUPP;
info->feature_discard = 0;
info->feature_secdiscard = 0;
@@ -771,18 +809,14 @@ static irqreturn_t blkif_interrupt(int irq, void *dev_id)
case BLKIF_OP_FLUSH_DISKCACHE:
case BLKIF_OP_WRITE_BARRIER:
if (unlikely(bret->status == BLKIF_RSP_EOPNOTSUPP)) {
- printk(KERN_WARNING "blkfront: %s: write %s op failed\n",
- info->flush_op == BLKIF_OP_WRITE_BARRIER ?
- "barrier" : "flush disk cache",
- info->gd->disk_name);
+ printk(KERN_WARNING "blkfront: %s: %s op failed\n",
+ info->gd->disk_name, op_name(bret->operation));
error = -EOPNOTSUPP;
}
if (unlikely(bret->status == BLKIF_RSP_ERROR &&
info->shadow[id].req.u.rw.nr_segments == 0)) {
- printk(KERN_WARNING "blkfront: %s: empty write %s op failed\n",
- info->flush_op == BLKIF_OP_WRITE_BARRIER ?
- "barrier" : "flush disk cache",
- info->gd->disk_name);
+ printk(KERN_WARNING "blkfront: %s: empty %s op failed\n",
+ info->gd->disk_name, op_name(bret->operation));
error = -EOPNOTSUPP;
}
if (unlikely(error)) {
generated by cgit v1.2.3 (git 2.39.1) at 2024-07-04 03:03:32 +0000