leandrof/linux
1
0
Fork 0
forked from Minki/linux
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Merge branch 'for-3.15/drivers' of git://git.kernel.dk/linux-block

Browse Source 
Pull block driver update from Jens Axboe:
 "On top of the core pull request, here's the pull request for the
  driver related changes for 3.15.  It contains:

   - Improvements for msi-x registration for block drivers (mtip32xx,
     skd, cciss, nvme) from Alexander Gordeev.

   - A round of cleanups and improvements for drbd from Andreas
     Gruenbacher and Rashika Kheria.

   - A round of clanups and improvements for bcache from Kent.

   - Removal of sleep_on() and friends in DAC960, ataflop, swim3 from
     Arnd Bergmann.

   - Bug fix for a bug in the mtip32xx async completion code from Sam
     Bradshaw.

   - Bug fix for accidentally bouncing IO on 32-bit platforms with
     mtip32xx from Felipe Franciosi"

* 'for-3.15/drivers' of git://git.kernel.dk/linux-block: (103 commits)
  bcache: remove nested function usage
  bcache: Kill bucket->gc_gen
  bcache: Kill unused freelist
  bcache: Rework btree cache reserve handling
  bcache: Kill btree_io_wq
  bcache: btree locking rework
  bcache: Fix a race when freeing btree nodes
  bcache: Add a real GC_MARK_RECLAIMABLE
  bcache: Add bch_keylist_init_single()
  bcache: Improve priority_stats
  bcache: Better alloc tracepoints
  bcache: Kill dead cgroup code
  bcache: stop moving_gc marking buckets that can't be moved.
  bcache: Fix moving_pred()
  bcache: Fix moving_gc deadlocking with a foreground write
  bcache: Fix discard granularity
  bcache: Fix another bug recovering from unclean shutdown
  bcache: Fix a bug recovering from unclean shutdown
  bcache: Fix a journalling reclaim after recovery bug
  bcache: Fix a null ptr deref in journal replay
  ...
This commit is contained in:
Linus Torvalds 2014-04-01 19:43:53 -07:00
commit b33ce44299
47 changed files with 6622 additions and 6298 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

38
Documentation/blockdev/drbd/data-structure-v9.txt Normal file
View File

@ -0,0 +1,38 @@
This describes the in kernel data structure for DRBD-9. Starting with
Linux v3.14 we are reorganizing DRBD to use this data structure.
Basic Data Structure
====================
A node has a number of DRBD resources. Each such resource has a number of
devices (aka volumes) and connections to other nodes ("peer nodes"). Each DRBD
device is represented by a block device locally.
The DRBD objects are interconnected to form a matrix as depicted below; a
drbd_peer_device object sits at each intersection between a drbd_device and a
drbd_connection:
/--------------+---------------+.....+---------------\
| resource | device | | device |
+--------------+---------------+.....+---------------+
| connection | peer_device | | peer_device |
+--------------+---------------+.....+---------------+
: : : : :
: : : : :
+--------------+---------------+.....+---------------+
| connection | peer_device | | peer_device |
\--------------+---------------+.....+---------------/
In this table, horizontally, devices can be accessed from resources by their
volume number. Likewise, peer_devices can be accessed from connections by
their volume number. Objects in the vertical direction are connected by double
linked lists. There are back pointers from peer_devices to their connections a
devices, and from connections and devices to their resource.
All resources are in the drbd_resources double-linked list. In addition, all
devices can be accessed by their minor device number via the drbd_devices idr.
The drbd_resource, drbd_connection, and drbd_device objects are reference
counted. The peer_device objects only serve to establish the links between
devices and connections; their lifetime is determined by the lifetime of the
device and connection which they reference.

34
drivers/block/DAC960.c
View File

@ -6411,12 +6411,12 @@ static bool DAC960_V2_ExecuteUserCommand(DAC960_Controller_T *Controller,
.ScatterGatherSegments[0]
.SegmentByteCount =
CommandMailbox->ControllerInfo.DataTransferSize;
DAC960_ExecuteCommand(Command);
while (Controller->V2.NewControllerInformation->PhysicalScanActive)
{
DAC960_ExecuteCommand(Command);
sleep_on_timeout(&Controller->CommandWaitQueue, HZ);
}
while (1) {
DAC960_ExecuteCommand(Command);
if (!Controller->V2.NewControllerInformation->PhysicalScanActive)
break;
msleep(1000);
}
DAC960_UserCritical("Discovery Completed\n", Controller);
}
}
@ -7035,18 +7035,16 @@ static long DAC960_gam_ioctl(struct file *file, unsigned int Request,
ErrorCode = -EFAULT;
break;
}
while (Controller->V2.HealthStatusBuffer->StatusChangeCounter
== HealthStatusBuffer.StatusChangeCounter &&
Controller->V2.HealthStatusBuffer->NextEventSequenceNumber
== HealthStatusBuffer.NextEventSequenceNumber)
{
interruptible_sleep_on_timeout(&Controller->HealthStatusWaitQueue,
DAC960_MonitoringTimerInterval);
if (signal_pending(current)) {
ErrorCode = -EINTR;
break;
}
}
ErrorCode = wait_event_interruptible_timeout(Controller->HealthStatusWaitQueue,
!(Controller->V2.HealthStatusBuffer->StatusChangeCounter
== HealthStatusBuffer.StatusChangeCounter &&
Controller->V2.HealthStatusBuffer->NextEventSequenceNumber
== HealthStatusBuffer.NextEventSequenceNumber),
DAC960_MonitoringTimerInterval);
if (ErrorCode == -ERESTARTSYS) {
ErrorCode = -EINTR;
break;
}
if (copy_to_user(GetHealthStatus.HealthStatusBuffer,
Controller->V2.HealthStatusBuffer,
sizeof(DAC960_V2_HealthStatusBuffer_T)))

16
drivers/block/ataflop.c
View File

@ -68,6 +68,8 @@
#include <linux/init.h>
#include <linux/blkdev.h>
#include <linux/mutex.h>
#include <linux/completion.h>
#include <linux/wait.h>
#include <asm/atafd.h>
#include <asm/atafdreg.h>
@ -301,7 +303,7 @@ module_param_array(UserSteprate, int, NULL, 0);
/* Synchronization of FDC access. */
static volatile int fdc_busy = 0;
static DECLARE_WAIT_QUEUE_HEAD(fdc_wait);
static DECLARE_WAIT_QUEUE_HEAD(format_wait);
static DECLARE_COMPLETION(format_wait);
static unsigned long changed_floppies = 0xff, fake_change = 0;
#define CHECK_CHANGE_DELAY HZ/2
@ -608,7 +610,7 @@ static void fd_error( void )
if (IsFormatting) {
IsFormatting = 0;
FormatError = 1;
wake_up( &format_wait );
complete(&format_wait);
return;
}
@ -650,9 +652,8 @@ static int do_format(int drive, int type, struct atari_format_descr *desc)
DPRINT(("do_format( dr=%d tr=%d he=%d offs=%d )\n",
drive, desc->track, desc->head, desc->sect_offset ));
wait_event(fdc_wait, cmpxchg(&fdc_busy, 0, 1) == 0);
local_irq_save(flags);
while( fdc_busy ) sleep_on( &fdc_wait );
fdc_busy = 1;
stdma_lock(floppy_irq, NULL);
atari_turnon_irq( IRQ_MFP_FDC ); /* should be already, just to be sure */
local_irq_restore(flags);
@ -706,7 +707,7 @@ static int do_format(int drive, int type, struct atari_format_descr *desc)
ReqSide = desc->head;
do_fd_action( drive );
sleep_on( &format_wait );
wait_for_completion(&format_wait);
redo_fd_request();
return( FormatError ? -EIO : 0 );
@ -1229,7 +1230,7 @@ static void fd_writetrack_done( int status )
goto err_end;
}
wake_up( &format_wait );
complete(&format_wait);
return;
err_end:
@ -1497,8 +1498,7 @@ repeat:
void do_fd_request(struct request_queue * q)
{
DPRINT(("do_fd_request for pid %d\n",current->pid));
while( fdc_busy ) sleep_on( &fdc_wait );
fdc_busy = 1;
wait_event(fdc_wait, cmpxchg(&fdc_busy, 0, 1) == 0);
stdma_lock(floppy_irq, NULL);
atari_disable_irq( IRQ_MFP_FDC );

2
drivers/block/cciss.c
View File

@ -4092,11 +4092,9 @@ static void cciss_interrupt_mode(ctlr_info_t *h)
if (err > 0) {
dev_warn(&h->pdev->dev,
"only %d MSI-X vectors available\n", err);
goto default_int_mode;
} else {
dev_warn(&h->pdev->dev,
"MSI-X init failed %d\n", err);
goto default_int_mode;
}
}
if (pci_find_capability(h->pdev, PCI_CAP_ID_MSI)) {

629
drivers/block/drbd/drbd_actlog.c
View File

File diff suppressed because it is too large Load Diff

368
drivers/block/drbd/drbd_bitmap.c
View File

@ -113,54 +113,54 @@ struct drbd_bitmap {
};
#define bm_print_lock_info(m) __bm_print_lock_info(m, __func__)
static void __bm_print_lock_info(struct drbd_conf *mdev, const char *func)
static void __bm_print_lock_info(struct drbd_device *device, const char *func)
{
struct drbd_bitmap *b = mdev->bitmap;
struct drbd_bitmap *b = device->bitmap;
if (!__ratelimit(&drbd_ratelimit_state))
return;
dev_err(DEV, "FIXME %s in %s, bitmap locked for '%s' by %s\n",
drbd_task_to_thread_name(mdev->tconn, current),
func, b->bm_why ?: "?",
drbd_task_to_thread_name(mdev->tconn, b->bm_task));
drbd_err(device, "FIXME %s[%d] in %s, bitmap locked for '%s' by %s[%d]\n",
current->comm, task_pid_nr(current),
func, b->bm_why ?: "?",
b->bm_task->comm, task_pid_nr(b->bm_task));
}
void drbd_bm_lock(struct drbd_conf *mdev, char *why, enum bm_flag flags)
void drbd_bm_lock(struct drbd_device *device, char *why, enum bm_flag flags)
{
struct drbd_bitmap *b = mdev->bitmap;
struct drbd_bitmap *b = device->bitmap;
int trylock_failed;
if (!b) {
dev_err(DEV, "FIXME no bitmap in drbd_bm_lock!?\n");
drbd_err(device, "FIXME no bitmap in drbd_bm_lock!?\n");
return;
}
trylock_failed = !mutex_trylock(&b->bm_change);
if (trylock_failed) {
dev_warn(DEV, "%s going to '%s' but bitmap already locked for '%s' by %s\n",
drbd_task_to_thread_name(mdev->tconn, current),
why, b->bm_why ?: "?",
drbd_task_to_thread_name(mdev->tconn, b->bm_task));
drbd_warn(device, "%s[%d] going to '%s' but bitmap already locked for '%s' by %s[%d]\n",
current->comm, task_pid_nr(current),
why, b->bm_why ?: "?",
b->bm_task->comm, task_pid_nr(b->bm_task));
mutex_lock(&b->bm_change);
}
if (BM_LOCKED_MASK & b->bm_flags)
dev_err(DEV, "FIXME bitmap already locked in bm_lock\n");
drbd_err(device, "FIXME bitmap already locked in bm_lock\n");
b->bm_flags |= flags & BM_LOCKED_MASK;
b->bm_why = why;
b->bm_task = current;
}
void drbd_bm_unlock(struct drbd_conf *mdev)
void drbd_bm_unlock(struct drbd_device *device)
{
struct drbd_bitmap *b = mdev->bitmap;
struct drbd_bitmap *b = device->bitmap;
if (!b) {
dev_err(DEV, "FIXME no bitmap in drbd_bm_unlock!?\n");
drbd_err(device, "FIXME no bitmap in drbd_bm_unlock!?\n");
return;
}
if (!(BM_LOCKED_MASK & mdev->bitmap->bm_flags))
dev_err(DEV, "FIXME bitmap not locked in bm_unlock\n");
if (!(BM_LOCKED_MASK & device->bitmap->bm_flags))
drbd_err(device, "FIXME bitmap not locked in bm_unlock\n");
b->bm_flags &= ~BM_LOCKED_MASK;
b->bm_why = NULL;
@ -211,19 +211,19 @@ static unsigned long bm_page_to_idx(struct page *page)
/* As is very unlikely that the same page is under IO from more than one
* context, we can get away with a bit per page and one wait queue per bitmap.
*/
static void bm_page_lock_io(struct drbd_conf *mdev, int page_nr)
static void bm_page_lock_io(struct drbd_device *device, int page_nr)
{
struct drbd_bitmap *b = mdev->bitmap;
struct drbd_bitmap *b = device->bitmap;
void *addr = &page_private(b->bm_pages[page_nr]);
wait_event(b->bm_io_wait, !test_and_set_bit(BM_PAGE_IO_LOCK, addr));
}
static void bm_page_unlock_io(struct drbd_conf *mdev, int page_nr)
static void bm_page_unlock_io(struct drbd_device *device, int page_nr)
{
struct drbd_bitmap *b = mdev->bitmap;
struct drbd_bitmap *b = device->bitmap;
void *addr = &page_private(b->bm_pages[page_nr]);
clear_bit_unlock(BM_PAGE_IO_LOCK, addr);
wake_up(&mdev->bitmap->bm_io_wait);
wake_up(&device->bitmap->bm_io_wait);
}
/* set _before_ submit_io, so it may be reset due to being changed
@ -242,22 +242,22 @@ static void bm_set_page_need_writeout(struct page *page)
/**
* drbd_bm_mark_for_writeout() - mark a page with a "hint" to be considered for writeout
* @mdev: DRBD device.
* @device: DRBD device.
* @page_nr: the bitmap page to mark with the "hint" flag
*
* From within an activity log transaction, we mark a few pages with these
* hints, then call drbd_bm_write_hinted(), which will only write out changed
* pages which are flagged with this mark.
*/
void drbd_bm_mark_for_writeout(struct drbd_conf *mdev, int page_nr)
void drbd_bm_mark_for_writeout(struct drbd_device *device, int page_nr)
{
struct page *page;
if (page_nr >= mdev->bitmap->bm_number_of_pages) {
dev_warn(DEV, "BAD: page_nr: %u, number_of_pages: %u\n",
page_nr, (int)mdev->bitmap->bm_number_of_pages);
if (page_nr >= device->bitmap->bm_number_of_pages) {
drbd_warn(device, "BAD: page_nr: %u, number_of_pages: %u\n",
page_nr, (int)device->bitmap->bm_number_of_pages);
return;
}
page = mdev->bitmap->bm_pages[page_nr];
page = device->bitmap->bm_pages[page_nr];
set_bit(BM_PAGE_HINT_WRITEOUT, &page_private(page));
}
@ -340,7 +340,7 @@ static void bm_unmap(unsigned long *p_addr)
/*
* actually most functions herein should take a struct drbd_bitmap*, not a
* struct drbd_conf*, but for the debug macros I like to have the mdev around
* struct drbd_device*, but for the debug macros I like to have the device around
* to be able to report device specific.
*/
@ -436,11 +436,11 @@ static struct page **bm_realloc_pages(struct drbd_bitmap *b, unsigned long want)
/*
* called on driver init only. TODO call when a device is created.
* allocates the drbd_bitmap, and stores it in mdev->bitmap.
* allocates the drbd_bitmap, and stores it in device->bitmap.
*/
int drbd_bm_init(struct drbd_conf *mdev)
int drbd_bm_init(struct drbd_device *device)
{
struct drbd_bitmap *b = mdev->bitmap;
struct drbd_bitmap *b = device->bitmap;
WARN_ON(b != NULL);
b = kzalloc(sizeof(struct drbd_bitmap), GFP_KERNEL);
if (!b)
@ -449,28 +449,28 @@ int drbd_bm_init(struct drbd_conf *mdev)
mutex_init(&b->bm_change);
init_waitqueue_head(&b->bm_io_wait);
mdev->bitmap = b;
device->bitmap = b;
return 0;
}
sector_t drbd_bm_capacity(struct drbd_conf *mdev)
sector_t drbd_bm_capacity(struct drbd_device *device)
{
if (!expect(mdev->bitmap))
if (!expect(device->bitmap))
return 0;
return mdev->bitmap->bm_dev_capacity;
return device->bitmap->bm_dev_capacity;
}
/* called on driver unload. TODO: call when a device is destroyed.
*/
void drbd_bm_cleanup(struct drbd_conf *mdev)
void drbd_bm_cleanup(struct drbd_device *device)
{
if (!expect(mdev->bitmap))
if (!expect(device->bitmap))
return;
bm_free_pages(mdev->bitmap->bm_pages, mdev->bitmap->bm_number_of_pages);
bm_vk_free(mdev->bitmap->bm_pages, (BM_P_VMALLOCED & mdev->bitmap->bm_flags));
kfree(mdev->bitmap);
mdev->bitmap = NULL;
bm_free_pages(device->bitmap->bm_pages, device->bitmap->bm_number_of_pages);
bm_vk_free(device->bitmap->bm_pages, (BM_P_VMALLOCED & device->bitmap->bm_flags));
kfree(device->bitmap);
device->bitmap = NULL;
}
/*
@ -631,9 +631,9 @@ static u64 drbd_md_on_disk_bits(struct drbd_backing_dev *ldev)
* In case this is actually a resize, we copy the old bitmap into the new one.
* Otherwise, the bitmap is initialized to all bits set.
*/
int drbd_bm_resize(struct drbd_conf *mdev, sector_t capacity, int set_new_bits)
int drbd_bm_resize(struct drbd_device *device, sector_t capacity, int set_new_bits)
{
struct drbd_bitmap *b = mdev->bitmap;
struct drbd_bitmap *b = device->bitmap;
unsigned long bits, words, owords, obits;
unsigned long want, have, onpages; /* number of pages */
struct page **npages, **opages = NULL;
@ -643,9 +643,9 @@ int drbd_bm_resize(struct drbd_conf *mdev, sector_t capacity, int set_new_bits)
if (!expect(b))
return -ENOMEM;
drbd_bm_lock(mdev, "resize", BM_LOCKED_MASK);
drbd_bm_lock(device, "resize", BM_LOCKED_MASK);
dev_info(DEV, "drbd_bm_resize called with capacity == %llu\n",
drbd_info(device, "drbd_bm_resize called with capacity == %llu\n",
(unsigned long long)capacity);
if (capacity == b->bm_dev_capacity)
@ -678,12 +678,12 @@ int drbd_bm_resize(struct drbd_conf *mdev, sector_t capacity, int set_new_bits)
*/
words = ALIGN(bits, 64) >> LN2_BPL;
if (get_ldev(mdev)) {
u64 bits_on_disk = drbd_md_on_disk_bits(mdev->ldev);
put_ldev(mdev);
if (get_ldev(device)) {
u64 bits_on_disk = drbd_md_on_disk_bits(device->ldev);
put_ldev(device);
if (bits > bits_on_disk) {
dev_info(DEV, "bits = %lu\n", bits);
dev_info(DEV, "bits_on_disk = %llu\n", bits_on_disk);
drbd_info(device, "bits = %lu\n", bits);
drbd_info(device, "bits_on_disk = %llu\n", bits_on_disk);
err = -ENOSPC;
goto out;
}
@ -692,10 +692,10 @@ int drbd_bm_resize(struct drbd_conf *mdev, sector_t capacity, int set_new_bits)
want = ALIGN(words*sizeof(long), PAGE_SIZE) >> PAGE_SHIFT;
have = b->bm_number_of_pages;
if (want == have) {
D_ASSERT(b->bm_pages != NULL);
D_ASSERT(device, b->bm_pages != NULL);
npages = b->bm_pages;
} else {
if (drbd_insert_fault(mdev, DRBD_FAULT_BM_ALLOC))
if (drbd_insert_fault(device, DRBD_FAULT_BM_ALLOC))
npages = NULL;
else
npages = bm_realloc_pages(b, want);
@ -742,10 +742,10 @@ int drbd_bm_resize(struct drbd_conf *mdev, sector_t capacity, int set_new_bits)
bm_vk_free(opages, opages_vmalloced);
if (!growing)
b->bm_set = bm_count_bits(b);
dev_info(DEV, "resync bitmap: bits=%lu words=%lu pages=%lu\n", bits, words, want);
drbd_info(device, "resync bitmap: bits=%lu words=%lu pages=%lu\n", bits, words, want);
out:
drbd_bm_unlock(mdev);
drbd_bm_unlock(device);
return err;
}
@ -757,9 +757,9 @@ int drbd_bm_resize(struct drbd_conf *mdev, sector_t capacity, int set_new_bits)
*
* maybe bm_set should be atomic_t ?
*/
unsigned long _drbd_bm_total_weight(struct drbd_conf *mdev)
unsigned long _drbd_bm_total_weight(struct drbd_device *device)
{
struct drbd_bitmap *b = mdev->bitmap;
struct drbd_bitmap *b = device->bitmap;
unsigned long s;
unsigned long flags;
@ -775,20 +775,20 @@ unsigned long _drbd_bm_total_weight(struct drbd_conf *mdev)
return s;
}
unsigned long drbd_bm_total_weight(struct drbd_conf *mdev)
unsigned long drbd_bm_total_weight(struct drbd_device *device)
{
unsigned long s;
/* if I don't have a disk, I don't know about out-of-sync status */
if (!get_ldev_if_state(mdev, D_NEGOTIATING))
if (!get_ldev_if_state(device, D_NEGOTIATING))
return 0;
s = _drbd_bm_total_weight(mdev);
put_ldev(mdev);
s = _drbd_bm_total_weight(device);
put_ldev(device);
return s;
}
size_t drbd_bm_words(struct drbd_conf *mdev)
size_t drbd_bm_words(struct drbd_device *device)
{
struct drbd_bitmap *b = mdev->bitmap;
struct drbd_bitmap *b = device->bitmap;
if (!expect(b))
return 0;
if (!expect(b->bm_pages))
@ -797,9 +797,9 @@ size_t drbd_bm_words(struct drbd_conf *mdev)
return b->bm_words;
}
unsigned long drbd_bm_bits(struct drbd_conf *mdev)
unsigned long drbd_bm_bits(struct drbd_device *device)
{
struct drbd_bitmap *b = mdev->bitmap;
struct drbd_bitmap *b = device->bitmap;
if (!expect(b))
return 0;
@ -811,10 +811,10 @@ unsigned long drbd_bm_bits(struct drbd_conf *mdev)
* bitmap must be locked by drbd_bm_lock.
* currently only used from receive_bitmap.
*/
void drbd_bm_merge_lel(struct drbd_conf *mdev, size_t offset, size_t number,
void drbd_bm_merge_lel(struct drbd_device *device, size_t offset, size_t number,
unsigned long *buffer)
{
struct drbd_bitmap *b = mdev->bitmap;
struct drbd_bitmap *b = device->bitmap;
unsigned long *p_addr, *bm;
unsigned long word, bits;
unsigned int idx;
@ -860,10 +860,10 @@ void drbd_bm_merge_lel(struct drbd_conf *mdev, size_t offset, size_t number,
/* copy number words from the bitmap starting at offset into the buffer.
* buffer[i] will be little endian unsigned long.
*/
void drbd_bm_get_lel(struct drbd_conf *mdev, size_t offset, size_t number,
void drbd_bm_get_lel(struct drbd_device *device, size_t offset, size_t number,
unsigned long *buffer)
{
struct drbd_bitmap *b = mdev->bitmap;
struct drbd_bitmap *b = device->bitmap;
unsigned long *p_addr, *bm;
size_t end, do_now;
@ -878,7 +878,7 @@ void drbd_bm_get_lel(struct drbd_conf *mdev, size_t offset, size_t number,
if ((offset >= b->bm_words) ||
(end > b->bm_words) ||
(number <= 0))
dev_err(DEV, "offset=%lu number=%lu bm_words=%lu\n",
drbd_err(device, "offset=%lu number=%lu bm_words=%lu\n",
(unsigned long) offset,
(unsigned long) number,
(unsigned long) b->bm_words);
@ -897,9 +897,9 @@ void drbd_bm_get_lel(struct drbd_conf *mdev, size_t offset, size_t number,
}
/* set all bits in the bitmap */
void drbd_bm_set_all(struct drbd_conf *mdev)
void drbd_bm_set_all(struct drbd_device *device)
{
struct drbd_bitmap *b = mdev->bitmap;
struct drbd_bitmap *b = device->bitmap;
if (!expect(b))
return;
if (!expect(b->bm_pages))
@ -913,9 +913,9 @@ void drbd_bm_set_all(struct drbd_conf *mdev)
}
/* clear all bits in the bitmap */
void drbd_bm_clear_all(struct drbd_conf *mdev)
void drbd_bm_clear_all(struct drbd_device *device)
{
struct drbd_bitmap *b = mdev->bitmap;
struct drbd_bitmap *b = device->bitmap;
if (!expect(b))
return;
if (!expect(b->bm_pages))
@ -928,7 +928,7 @@ void drbd_bm_clear_all(struct drbd_conf *mdev)
}
struct bm_aio_ctx {
struct drbd_conf *mdev;
struct drbd_device *device;
atomic_t in_flight;
unsigned int done;
unsigned flags;
@ -943,7 +943,7 @@ static void bm_aio_ctx_destroy(struct kref *kref)
{
struct bm_aio_ctx *ctx = container_of(kref, struct bm_aio_ctx, kref);
put_ldev(ctx->mdev);
put_ldev(ctx->device);
kfree(ctx);
}
@ -951,8 +951,8 @@ static void bm_aio_ctx_destroy(struct kref *kref)
static void bm_async_io_complete(struct bio *bio, int error)
{
struct bm_aio_ctx *ctx = bio->bi_private;
struct drbd_conf *mdev = ctx->mdev;
struct drbd_bitmap *b = mdev->bitmap;
struct drbd_device *device = ctx->device;
struct drbd_bitmap *b = device->bitmap;
unsigned int idx = bm_page_to_idx(bio->bi_io_vec[0].bv_page);
int uptodate = bio_flagged(bio, BIO_UPTODATE);
@ -966,7 +966,7 @@ static void bm_async_io_complete(struct bio *bio, int error)
if ((ctx->flags & BM_AIO_COPY_PAGES) == 0 &&
!bm_test_page_unchanged(b->bm_pages[idx]))
dev_warn(DEV, "bitmap page idx %u changed during IO!\n", idx);
drbd_warn(device, "bitmap page idx %u changed during IO!\n", idx);
if (error) {
/* ctx error will hold the completed-last non-zero error code,
@ -976,14 +976,14 @@ static void bm_async_io_complete(struct bio *bio, int error)
/* Not identical to on disk version of it.
* Is BM_PAGE_IO_ERROR enough? */
if (__ratelimit(&drbd_ratelimit_state))
dev_err(DEV, "IO ERROR %d on bitmap page idx %u\n",
drbd_err(device, "IO ERROR %d on bitmap page idx %u\n",
error, idx);
} else {
bm_clear_page_io_err(b->bm_pages[idx]);
dynamic_dev_dbg(DEV, "bitmap page idx %u completed\n", idx);
dynamic_drbd_dbg(device, "bitmap page idx %u completed\n", idx);
}
bm_page_unlock_io(mdev, idx);
bm_page_unlock_io(device, idx);
if (ctx->flags & BM_AIO_COPY_PAGES)
mempool_free(bio->bi_io_vec[0].bv_page, drbd_md_io_page_pool);
@ -992,7 +992,7 @@ static void bm_async_io_complete(struct bio *bio, int error)
if (atomic_dec_and_test(&ctx->in_flight)) {
ctx->done = 1;
wake_up(&mdev->misc_wait);
wake_up(&device->misc_wait);
kref_put(&ctx->kref, &bm_aio_ctx_destroy);
}
}
@ -1000,23 +1000,23 @@ static void bm_async_io_complete(struct bio *bio, int error)
static void bm_page_io_async(struct bm_aio_ctx *ctx, int page_nr, int rw) __must_hold(local)
{
struct bio *bio = bio_alloc_drbd(GFP_NOIO);
struct drbd_conf *mdev = ctx->mdev;
struct drbd_bitmap *b = mdev->bitmap;
struct drbd_device *device = ctx->device;
struct drbd_bitmap *b = device->bitmap;
struct page *page;
unsigned int len;
sector_t on_disk_sector =
mdev->ldev->md.md_offset + mdev->ldev->md.bm_offset;
device->ldev->md.md_offset + device->ldev->md.bm_offset;
on_disk_sector += ((sector_t)page_nr) << (PAGE_SHIFT-9);
/* this might happen with very small
* flexible external meta data device,
* or with PAGE_SIZE > 4k */
len = min_t(unsigned int, PAGE_SIZE,
(drbd_md_last_sector(mdev->ldev) - on_disk_sector + 1)<<9);
(drbd_md_last_sector(device->ldev) - on_disk_sector + 1)<<9);
/* serialize IO on this page */
bm_page_lock_io(mdev, page_nr);
bm_page_lock_io(device, page_nr);
/* before memcpy and submit,
* so it can be redirtied any time */
bm_set_page_unchanged(b->bm_pages[page_nr]);
@ -1027,7 +1027,7 @@ static void bm_page_io_async(struct bm_aio_ctx *ctx, int page_nr, int rw) __must
bm_store_page_idx(page, page_nr);
} else
page = b->bm_pages[page_nr];
bio->bi_bdev = mdev->ldev->md_bdev;
bio->bi_bdev = device->ldev->md_bdev;
bio->bi_iter.bi_sector = on_disk_sector;
/* bio_add_page of a single page to an empty bio will always succeed,
* according to api. Do we want to assert that? */
@ -1035,24 +1035,24 @@ static void bm_page_io_async(struct bm_aio_ctx *ctx, int page_nr, int rw) __must
bio->bi_private = ctx;
bio->bi_end_io = bm_async_io_complete;
if (drbd_insert_fault(mdev, (rw & WRITE) ? DRBD_FAULT_MD_WR : DRBD_FAULT_MD_RD)) {
if (drbd_insert_fault(device, (rw & WRITE) ? DRBD_FAULT_MD_WR : DRBD_FAULT_MD_RD)) {
bio->bi_rw |= rw;
bio_endio(bio, -EIO);
} else {
submit_bio(rw, bio);
/* this should not count as user activity and cause the
* resync to throttle -- see drbd_rs_should_slow_down(). */
atomic_add(len >> 9, &mdev->rs_sect_ev);
atomic_add(len >> 9, &device->rs_sect_ev);
}
}
/*
* bm_rw: read/write the whole bitmap from/to its on disk location.
*/
static int bm_rw(struct drbd_conf *mdev, int rw, unsigned flags, unsigned lazy_writeout_upper_idx) __must_hold(local)
static int bm_rw(struct drbd_device *device, int rw, unsigned flags, unsigned lazy_writeout_upper_idx) __must_hold(local)
{
struct bm_aio_ctx *ctx;
struct drbd_bitmap *b = mdev->bitmap;
struct drbd_bitmap *b = device->bitmap;
int num_pages, i, count = 0;
unsigned long now;
char ppb[10];
@ -1072,7 +1072,7 @@ static int bm_rw(struct drbd_conf *mdev, int rw, unsigned flags, unsigned lazy_w
return -ENOMEM;
*ctx = (struct bm_aio_ctx) {
.mdev = mdev,
.device = device,
.in_flight = ATOMIC_INIT(1),
.done = 0,
.flags = flags,
@ -1080,8 +1080,8 @@ static int bm_rw(struct drbd_conf *mdev, int rw, unsigned flags, unsigned lazy_w
.kref = { ATOMIC_INIT(2) },
};
if (!get_ldev_if_state(mdev, D_ATTACHING)) { /* put is in bm_aio_ctx_destroy() */
dev_err(DEV, "ASSERT FAILED: get_ldev_if_state() == 1 in bm_rw()\n");
if (!get_ldev_if_state(device, D_ATTACHING)) { /* put is in bm_aio_ctx_destroy() */
drbd_err(device, "ASSERT FAILED: get_ldev_if_state() == 1 in bm_rw()\n");
kfree(ctx);
return -ENODEV;
}
@ -1106,14 +1106,14 @@ static int bm_rw(struct drbd_conf *mdev, int rw, unsigned flags, unsigned lazy_w
if (!(flags & BM_WRITE_ALL_PAGES) &&
bm_test_page_unchanged(b->bm_pages[i])) {
dynamic_dev_dbg(DEV, "skipped bm write for idx %u\n", i);
dynamic_drbd_dbg(device, "skipped bm write for idx %u\n", i);
continue;
}
/* during lazy writeout,
* ignore those pages not marked for lazy writeout. */
if (lazy_writeout_upper_idx &&
!bm_test_page_lazy_writeout(b->bm_pages[i])) {
dynamic_dev_dbg(DEV, "skipped bm lazy write for idx %u\n", i);
dynamic_drbd_dbg(device, "skipped bm lazy write for idx %u\n", i);
continue;
}
}
@ -1132,19 +1132,19 @@ static int bm_rw(struct drbd_conf *mdev, int rw, unsigned flags, unsigned lazy_w
* "in_flight reached zero, all done" event.
*/
if (!atomic_dec_and_test(&ctx->in_flight))
wait_until_done_or_force_detached(mdev, mdev->ldev, &ctx->done);
wait_until_done_or_force_detached(device, device->ldev, &ctx->done);
else
kref_put(&ctx->kref, &bm_aio_ctx_destroy);
/* summary for global bitmap IO */
if (flags == 0)
dev_info(DEV, "bitmap %s of %u pages took %lu jiffies\n",
drbd_info(device, "bitmap %s of %u pages took %lu jiffies\n",
rw == WRITE ? "WRITE" : "READ",
count, jiffies - now);
if (ctx->error) {
dev_alert(DEV, "we had at least one MD IO ERROR during bitmap IO\n");
drbd_chk_io_error(mdev, 1, DRBD_META_IO_ERROR);
drbd_alert(device, "we had at least one MD IO ERROR during bitmap IO\n");
drbd_chk_io_error(device, 1, DRBD_META_IO_ERROR);
err = -EIO; /* ctx->error ? */
}
@ -1153,16 +1153,16 @@ static int bm_rw(struct drbd_conf *mdev, int rw, unsigned flags, unsigned lazy_w
now = jiffies;
if (rw == WRITE) {
drbd_md_flush(mdev);
drbd_md_flush(device);
} else /* rw == READ */ {
b->bm_set = bm_count_bits(b);
dev_info(DEV, "recounting of set bits took additional %lu jiffies\n",
drbd_info(device, "recounting of set bits took additional %lu jiffies\n",
jiffies - now);
}
now = b->bm_set;
if (flags == 0)
dev_info(DEV, "%s (%lu bits) marked out-of-sync by on disk bit-map.\n",
drbd_info(device, "%s (%lu bits) marked out-of-sync by on disk bit-map.\n",
ppsize(ppb, now << (BM_BLOCK_SHIFT-10)), now);
kref_put(&ctx->kref, &bm_aio_ctx_destroy);
@ -1171,48 +1171,38 @@ static int bm_rw(struct drbd_conf *mdev, int rw, unsigned flags, unsigned lazy_w
/**
* drbd_bm_read() - Read the whole bitmap from its on disk location.
* @mdev: DRBD device.
* @device: DRBD device.
*/
int drbd_bm_read(struct drbd_conf *mdev) __must_hold(local)
int drbd_bm_read(struct drbd_device *device) __must_hold(local)
{
return bm_rw(mdev, READ, 0, 0);
return bm_rw(device, READ, 0, 0);
}
/**
* drbd_bm_write() - Write the whole bitmap to its on disk location.
* @mdev: DRBD device.
* @device: DRBD device.
*
* Will only write pages that have changed since last IO.
*/
int drbd_bm_write(struct drbd_conf *mdev) __must_hold(local)
int drbd_bm_write(struct drbd_device *device) __must_hold(local)
{
return bm_rw(mdev, WRITE, 0, 0);
return bm_rw(device, WRITE, 0, 0);
}
/**
* drbd_bm_write_all() - Write the whole bitmap to its on disk location.
* @mdev: DRBD device.
* @device: DRBD device.
*
* Will write all pages.
*/
int drbd_bm_write_all(struct drbd_conf *mdev) __must_hold(local)
int drbd_bm_write_all(struct drbd_device *device) __must_hold(local)
{
return bm_rw(mdev, WRITE, BM_WRITE_ALL_PAGES, 0);
}
/**
* drbd_bm_lazy_write_out() - Write bitmap pages 0 to @upper_idx-1, if they have changed.
* @mdev: DRBD device.
* @upper_idx: 0: write all changed pages; +ve: page index to stop scanning for changed pages
*/
int drbd_bm_write_lazy(struct drbd_conf *mdev, unsigned upper_idx) __must_hold(local)
{
return bm_rw(mdev, WRITE, BM_AIO_COPY_PAGES, upper_idx);
return bm_rw(device, WRITE, BM_WRITE_ALL_PAGES, 0);
}
/**
* drbd_bm_write_copy_pages() - Write the whole bitmap to its on disk location.
* @mdev: DRBD device.
* @device: DRBD device.
*
* Will only write pages that have changed since last IO.
* In contrast to drbd_bm_write(), this will copy the bitmap pages
@ -1221,23 +1211,23 @@ int drbd_bm_write_lazy(struct drbd_conf *mdev, unsigned upper_idx) __must_hold(l
* verify is aborted due to a failed peer disk, while local IO continues, or
* pending resync acks are still being processed.
*/
int drbd_bm_write_copy_pages(struct drbd_conf *mdev) __must_hold(local)
int drbd_bm_write_copy_pages(struct drbd_device *device) __must_hold(local)
{
return bm_rw(mdev, WRITE, BM_AIO_COPY_PAGES, 0);
return bm_rw(device, WRITE, BM_AIO_COPY_PAGES, 0);
}
/**
* drbd_bm_write_hinted() - Write bitmap pages with "hint" marks, if they have changed.
* @mdev: DRBD device.
* @device: DRBD device.
*/
int drbd_bm_write_hinted(struct drbd_conf *mdev) __must_hold(local)
int drbd_bm_write_hinted(struct drbd_device *device) __must_hold(local)
{
return bm_rw(mdev, WRITE, BM_AIO_WRITE_HINTED | BM_AIO_COPY_PAGES, 0);
return bm_rw(device, WRITE, BM_AIO_WRITE_HINTED | BM_AIO_COPY_PAGES, 0);
}
/**
* drbd_bm_write_page() - Writes a PAGE_SIZE aligned piece of bitmap
* @mdev: DRBD device.
* @device: DRBD device.
* @idx: bitmap page index
*
* We don't want to special case on logical_block_size of the backend device,
@ -1247,13 +1237,13 @@ int drbd_bm_write_hinted(struct drbd_conf *mdev) __must_hold(local)
* In case this becomes an issue on systems with larger PAGE_SIZE,
* we may want to change this again to write 4k aligned 4k pieces.
*/
int drbd_bm_write_page(struct drbd_conf *mdev, unsigned int idx) __must_hold(local)
int drbd_bm_write_page(struct drbd_device *device, unsigned int idx) __must_hold(local)
{
struct bm_aio_ctx *ctx;
int err;
if (bm_test_page_unchanged(mdev->bitmap->bm_pages[idx])) {
dynamic_dev_dbg(DEV, "skipped bm page write for idx %u\n", idx);
if (bm_test_page_unchanged(device->bitmap->bm_pages[idx])) {
dynamic_drbd_dbg(device, "skipped bm page write for idx %u\n", idx);
return 0;
}
@ -1262,7 +1252,7 @@ int drbd_bm_write_page(struct drbd_conf *mdev, unsigned int idx) __must_hold(loc
return -ENOMEM;
*ctx = (struct bm_aio_ctx) {
.mdev = mdev,
.device = device,
.in_flight = ATOMIC_INIT(1),
.done = 0,
.flags = BM_AIO_COPY_PAGES,
@ -1270,21 +1260,21 @@ int drbd_bm_write_page(struct drbd_conf *mdev, unsigned int idx) __must_hold(loc
.kref = { ATOMIC_INIT(2) },
};
if (!get_ldev_if_state(mdev, D_ATTACHING)) { /* put is in bm_aio_ctx_destroy() */
dev_err(DEV, "ASSERT FAILED: get_ldev_if_state() == 1 in drbd_bm_write_page()\n");
if (!get_ldev_if_state(device, D_ATTACHING)) { /* put is in bm_aio_ctx_destroy() */
drbd_err(device, "ASSERT FAILED: get_ldev_if_state() == 1 in drbd_bm_write_page()\n");
kfree(ctx);
return -ENODEV;
}
bm_page_io_async(ctx, idx, WRITE_SYNC);
wait_until_done_or_force_detached(mdev, mdev->ldev, &ctx->done);
wait_until_done_or_force_detached(device, device->ldev, &ctx->done);
if (ctx->error)
drbd_chk_io_error(mdev, 1, DRBD_META_IO_ERROR);
drbd_chk_io_error(device, 1, DRBD_META_IO_ERROR);
/* that causes us to detach, so the in memory bitmap will be
* gone in a moment as well. */
mdev->bm_writ_cnt++;
device->bm_writ_cnt++;
err = atomic_read(&ctx->in_flight) ? -EIO : ctx->error;
kref_put(&ctx->kref, &bm_aio_ctx_destroy);
return err;
@ -1298,17 +1288,17 @@ int drbd_bm_write_page(struct drbd_conf *mdev, unsigned int idx) __must_hold(loc
*
* this returns a bit number, NOT a sector!
*/
static unsigned long __bm_find_next(struct drbd_conf *mdev, unsigned long bm_fo,
static unsigned long __bm_find_next(struct drbd_device *device, unsigned long bm_fo,
const int find_zero_bit)
{
struct drbd_bitmap *b = mdev->bitmap;
struct drbd_bitmap *b = device->bitmap;
unsigned long *p_addr;
unsigned long bit_offset;
unsigned i;
if (bm_fo > b->bm_bits) {
dev_err(DEV, "bm_fo=%lu bm_bits=%lu\n", bm_fo, b->bm_bits);
drbd_err(device, "bm_fo=%lu bm_bits=%lu\n", bm_fo, b->bm_bits);
bm_fo = DRBD_END_OF_BITMAP;
} else {
while (bm_fo < b->bm_bits) {
@ -1338,10 +1328,10 @@ static unsigned long __bm_find_next(struct drbd_conf *mdev, unsigned long bm_fo,
return bm_fo;
}
static unsigned long bm_find_next(struct drbd_conf *mdev,
static unsigned long bm_find_next(struct drbd_device *device,
unsigned long bm_fo, const int find_zero_bit)
{
struct drbd_bitmap *b = mdev->bitmap;
struct drbd_bitmap *b = device->bitmap;
unsigned long i = DRBD_END_OF_BITMAP;
if (!expect(b))
@ -1351,39 +1341,39 @@ static unsigned long bm_find_next(struct drbd_conf *mdev,
spin_lock_irq(&b->bm_lock);
if (BM_DONT_TEST & b->bm_flags)
bm_print_lock_info(mdev);
bm_print_lock_info(device);
i = __bm_find_next(mdev, bm_fo, find_zero_bit);
i = __bm_find_next(device, bm_fo, find_zero_bit);
spin_unlock_irq(&b->bm_lock);
return i;
}
unsigned long drbd_bm_find_next(struct drbd_conf *mdev, unsigned long bm_fo)
unsigned long drbd_bm_find_next(struct drbd_device *device, unsigned long bm_fo)
{
return bm_find_next(mdev, bm_fo, 0);
return bm_find_next(device, bm_fo, 0);
}
#if 0
/* not yet needed for anything. */
unsigned long drbd_bm_find_next_zero(struct drbd_conf *mdev, unsigned long bm_fo)
unsigned long drbd_bm_find_next_zero(struct drbd_device *device, unsigned long bm_fo)
{
return bm_find_next(mdev, bm_fo, 1);
return bm_find_next(device, bm_fo, 1);
}
#endif
/* does not spin_lock_irqsave.
* you must take drbd_bm_lock() first */
unsigned long _drbd_bm_find_next(struct drbd_conf *mdev, unsigned long bm_fo)
unsigned long _drbd_bm_find_next(struct drbd_device *device, unsigned long bm_fo)
{
/* WARN_ON(!(BM_DONT_SET & mdev->b->bm_flags)); */
return __bm_find_next(mdev, bm_fo, 0);
/* WARN_ON(!(BM_DONT_SET & device->b->bm_flags)); */
return __bm_find_next(device, bm_fo, 0);
}
unsigned long _drbd_bm_find_next_zero(struct drbd_conf *mdev, unsigned long bm_fo)
unsigned long _drbd_bm_find_next_zero(struct drbd_device *device, unsigned long bm_fo)
{
/* WARN_ON(!(BM_DONT_SET & mdev->b->bm_flags)); */
return __bm_find_next(mdev, bm_fo, 1);
/* WARN_ON(!(BM_DONT_SET & device->b->bm_flags)); */
return __bm_find_next(device, bm_fo, 1);
}
/* returns number of bits actually changed.
@ -1392,10 +1382,10 @@ unsigned long _drbd_bm_find_next_zero(struct drbd_conf *mdev, unsigned long bm_f
* wants bitnr, not sector.
* expected to be called for only a few bits (e - s about BITS_PER_LONG).
* Must hold bitmap lock already. */
static int __bm_change_bits_to(struct drbd_conf *mdev, const unsigned long s,
static int __bm_change_bits_to(struct drbd_device *device, const unsigned long s,
unsigned long e, int val)
{
struct drbd_bitmap *b = mdev->bitmap;
struct drbd_bitmap *b = device->bitmap;
unsigned long *p_addr = NULL;
unsigned long bitnr;
unsigned int last_page_nr = -1U;
@ -1403,7 +1393,7 @@ static int __bm_change_bits_to(struct drbd_conf *mdev, const unsigned long s,
int changed_total = 0;
if (e >= b->bm_bits) {
dev_err(DEV, "ASSERT FAILED: bit_s=%lu bit_e=%lu bm_bits=%lu\n",
drbd_err(device, "ASSERT FAILED: bit_s=%lu bit_e=%lu bm_bits=%lu\n",
s, e, b->bm_bits);
e = b->bm_bits ? b->bm_bits -1 : 0;
}
@ -1441,11 +1431,11 @@ static int __bm_change_bits_to(struct drbd_conf *mdev, const unsigned long s,
* for val != 0, we change 0 -> 1, return code positive
* for val == 0, we change 1 -> 0, return code negative
* wants bitnr, not sector */
static int bm_change_bits_to(struct drbd_conf *mdev, const unsigned long s,
static int bm_change_bits_to(struct drbd_device *device, const unsigned long s,
const unsigned long e, int val)
{
unsigned long flags;
struct drbd_bitmap *b = mdev->bitmap;
struct drbd_bitmap *b = device->bitmap;
int c = 0;
if (!expect(b))
@ -1455,24 +1445,24 @@ static int bm_change_bits_to(struct drbd_conf *mdev, const unsigned long s,
spin_lock_irqsave(&b->bm_lock, flags);
if ((val ? BM_DONT_SET : BM_DONT_CLEAR) & b->bm_flags)
bm_print_lock_info(mdev);
bm_print_lock_info(device);
c = __bm_change_bits_to(mdev, s, e, val);
c = __bm_change_bits_to(device, s, e, val);
spin_unlock_irqrestore(&b->bm_lock, flags);
return c;
}
/* returns number of bits changed 0 -> 1 */
int drbd_bm_set_bits(struct drbd_conf *mdev, const unsigned long s, const unsigned long e)
int drbd_bm_set_bits(struct drbd_device *device, const unsigned long s, const unsigned long e)
{
return bm_change_bits_to(mdev, s, e, 1);
return bm_change_bits_to(device, s, e, 1);
}
/* returns number of bits changed 1 -> 0 */
int drbd_bm_clear_bits(struct drbd_conf *mdev, const unsigned long s, const unsigned long e)
int drbd_bm_clear_bits(struct drbd_device *device, const unsigned long s, const unsigned long e)
{
return -bm_change_bits_to(mdev, s, e, 0);
return -bm_change_bits_to(device, s, e, 0);
}
/* sets all bits in full words,
@ -1504,7 +1494,7 @@ static inline void bm_set_full_words_within_one_page(struct drbd_bitmap *b,
* You must first drbd_bm_lock().
* Can be called to set the whole bitmap in one go.
* Sets bits from s to e _inclusive_. */
void _drbd_bm_set_bits(struct drbd_conf *mdev, const unsigned long s, const unsigned long e)
void _drbd_bm_set_bits(struct drbd_device *device, const unsigned long s, const unsigned long e)
{
/* First set_bit from the first bit (s)
* up to the next long boundary (sl),
@ -1514,7 +1504,7 @@ void _drbd_bm_set_bits(struct drbd_conf *mdev, const unsigned long s, const unsi
* Do not use memset, because we must account for changes,
* so we need to loop over the words with hweight() anyways.
*/
struct drbd_bitmap *b = mdev->bitmap;
struct drbd_bitmap *b = device->bitmap;
unsigned long sl = ALIGN(s,BITS_PER_LONG);
unsigned long el = (e+1) & ~((unsigned long)BITS_PER_LONG-1);
int first_page;
@ -1526,7 +1516,7 @@ void _drbd_bm_set_bits(struct drbd_conf *mdev, const unsigned long s, const unsi
if (e - s <= 3*BITS_PER_LONG) {
/* don't bother; el and sl may even be wrong. */
spin_lock_irq(&b->bm_lock);
__bm_change_bits_to(mdev, s, e, 1);
__bm_change_bits_to(device, s, e, 1);
spin_unlock_irq(&b->bm_lock);
return;
}
@ -1537,7 +1527,7 @@ void _drbd_bm_set_bits(struct drbd_conf *mdev, const unsigned long s, const unsi
/* bits filling the current long */
if (sl)
__bm_change_bits_to(mdev, s, sl-1, 1);
__bm_change_bits_to(device, s, sl-1, 1);
first_page = sl >> (3 + PAGE_SHIFT);
last_page = el >> (3 + PAGE_SHIFT);
@ -1549,7 +1539,7 @@ void _drbd_bm_set_bits(struct drbd_conf *mdev, const unsigned long s, const unsi
/* first and full pages, unless first page == last page */
for (page_nr = first_page; page_nr < last_page; page_nr++) {
bm_set_full_words_within_one_page(mdev->bitmap, page_nr, first_word, last_word);
bm_set_full_words_within_one_page(device->bitmap, page_nr, first_word, last_word);
spin_unlock_irq(&b->bm_lock);
cond_resched();
first_word = 0;
@ -1565,7 +1555,7 @@ void _drbd_bm_set_bits(struct drbd_conf *mdev, const unsigned long s, const unsi
* 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);
bm_set_full_words_within_one_page(device->bitmap, last_page, first_word, last_word);
/* possibly trailing bits.
* example: (e & 63) == 63, el will be e+1.
@ -1573,7 +1563,7 @@ void _drbd_bm_set_bits(struct drbd_conf *mdev, const unsigned long s, const unsi
* it would trigger an assert in __bm_change_bits_to()
*/
if (el <= e)
__bm_change_bits_to(mdev, el, e, 1);
__bm_change_bits_to(device, el, e, 1);
spin_unlock_irq(&b->bm_lock);
}
@ -1584,10 +1574,10 @@ void _drbd_bm_set_bits(struct drbd_conf *mdev, const unsigned long s, const unsi
* 0 ... bit not set
* -1 ... first out of bounds access, stop testing for bits!
*/
int drbd_bm_test_bit(struct drbd_conf *mdev, const unsigned long bitnr)
int drbd_bm_test_bit(struct drbd_device *device, const unsigned long bitnr)
{
unsigned long flags;
struct drbd_bitmap *b = mdev->bitmap;
struct drbd_bitmap *b = device->bitmap;
unsigned long *p_addr;
int i;
@ -1598,7 +1588,7 @@ int drbd_bm_test_bit(struct drbd_conf *mdev, const unsigned long bitnr)
spin_lock_irqsave(&b->bm_lock, flags);
if (BM_DONT_TEST & b->bm_flags)
bm_print_lock_info(mdev);
bm_print_lock_info(device);
if (bitnr < b->bm_bits) {
p_addr = bm_map_pidx(b, bm_bit_to_page_idx(b, bitnr));
i = test_bit_le(bitnr & BITS_PER_PAGE_MASK, p_addr) ? 1 : 0;
@ -1606,7 +1596,7 @@ int drbd_bm_test_bit(struct drbd_conf *mdev, const unsigned long bitnr)
} else if (bitnr == b->bm_bits) {
i = -1;
} else { /* (bitnr > b->bm_bits) */
dev_err(DEV, "bitnr=%lu > bm_bits=%lu\n", bitnr, b->bm_bits);
drbd_err(device, "bitnr=%lu > bm_bits=%lu\n", bitnr, b->bm_bits);
i = 0;
}
@ -1615,10 +1605,10 @@ int drbd_bm_test_bit(struct drbd_conf *mdev, const unsigned long bitnr)
}
/* returns number of bits set in the range [s, e] */
int drbd_bm_count_bits(struct drbd_conf *mdev, const unsigned long s, const unsigned long e)
int drbd_bm_count_bits(struct drbd_device *device, const unsigned long s, const unsigned long e)
{
unsigned long flags;
struct drbd_bitmap *b = mdev->bitmap;
struct drbd_bitmap *b = device->bitmap;
unsigned long *p_addr = NULL;
unsigned long bitnr;
unsigned int page_nr = -1U;
@ -1635,7 +1625,7 @@ int drbd_bm_count_bits(struct drbd_conf *mdev, const unsigned long s, const unsi
spin_lock_irqsave(&b->bm_lock, flags);
if (BM_DONT_TEST & b->bm_flags)
bm_print_lock_info(mdev);
bm_print_lock_info(device);
for (bitnr = s; bitnr <= e; bitnr++) {
unsigned int idx = bm_bit_to_page_idx(b, bitnr);
if (page_nr != idx) {
@ -1647,7 +1637,7 @@ int drbd_bm_count_bits(struct drbd_conf *mdev, const unsigned long s, const unsi
if (expect(bitnr < b->bm_bits))
c += (0 != test_bit_le(bitnr - (page_nr << (PAGE_SHIFT+3)), p_addr));
else
dev_err(DEV, "bitnr=%lu bm_bits=%lu\n", bitnr, b->bm_bits);
drbd_err(device, "bitnr=%lu bm_bits=%lu\n", bitnr, b->bm_bits);
}
if (p_addr)
bm_unmap(p_addr);
@ -1670,9 +1660,9 @@ int drbd_bm_count_bits(struct drbd_conf *mdev, const unsigned long s, const unsi
* reference count of some bitmap extent element from some lru instead...
*
*/
int drbd_bm_e_weight(struct drbd_conf *mdev, unsigned long enr)
int drbd_bm_e_weight(struct drbd_device *device, unsigned long enr)
{
struct drbd_bitmap *b = mdev->bitmap;
struct drbd_bitmap *b = device->bitmap;
int count, s, e;
unsigned long flags;
unsigned long *p_addr, *bm;
@ -1684,7 +1674,7 @@ int drbd_bm_e_weight(struct drbd_conf *mdev, unsigned long enr)
spin_lock_irqsave(&b->bm_lock, flags);
if (BM_DONT_TEST & b->bm_flags)
bm_print_lock_info(mdev);
bm_print_lock_info(device);
s = S2W(enr);
e = min((size_t)S2W(enr+1), b->bm_words);
@ -1697,7 +1687,7 @@ int drbd_bm_e_weight(struct drbd_conf *mdev, unsigned long enr)
count += hweight_long(*bm++);
bm_unmap(p_addr);
} else {
dev_err(DEV, "start offset (%d) too large in drbd_bm_e_weight\n", s);
drbd_err(device, "start offset (%d) too large in drbd_bm_e_weight\n", s);
}
spin_unlock_irqrestore(&b->bm_lock, flags);
return count;

1120
drivers/block/drbd/drbd_int.h
View File

File diff suppressed because it is too large Load Diff

2019
drivers/block/drbd/drbd_main.c
View File

File diff suppressed because it is too large Load Diff

1653
drivers/block/drbd/drbd_nl.c
View File

File diff suppressed because it is too large Load Diff

138
drivers/block/drbd/drbd_proc.c
View File

@ -46,7 +46,7 @@ const struct file_operations drbd_proc_fops = {
.release = drbd_proc_release,
};
void seq_printf_with_thousands_grouping(struct seq_file *seq, long v)
static void seq_printf_with_thousands_grouping(struct seq_file *seq, long v)
{
/* v is in kB/sec. We don't expect TiByte/sec yet. */
if (unlikely(v >= 1000000)) {
@ -66,14 +66,14 @@ void seq_printf_with_thousands_grouping(struct seq_file *seq, long v)
* [=====>..............] 33.5% (23456/123456)
* finish: 2:20:20 speed: 6,345 (6,456) K/sec
*/
static void drbd_syncer_progress(struct drbd_conf *mdev, struct seq_file *seq)
static void drbd_syncer_progress(struct drbd_device *device, struct seq_file *seq)
{
unsigned long db, dt, dbdt, rt, rs_left;
unsigned int res;
int i, x, y;
int stalled = 0;
drbd_get_syncer_progress(mdev, &rs_left, &res);
drbd_get_syncer_progress(device, &rs_left, &res);
x = res/50;
y = 20-x;
@ -85,21 +85,21 @@ static void drbd_syncer_progress(struct drbd_conf *mdev, struct seq_file *seq)
seq_printf(seq, ".");
seq_printf(seq, "] ");
if (mdev->state.conn == C_VERIFY_S || mdev->state.conn == C_VERIFY_T)
if (device->state.conn == C_VERIFY_S || device->state.conn == C_VERIFY_T)
seq_printf(seq, "verified:");
else
seq_printf(seq, "sync'ed:");
seq_printf(seq, "%3u.%u%% ", res / 10, res % 10);
/* if more than a few GB, display in MB */
if (mdev->rs_total > (4UL << (30 - BM_BLOCK_SHIFT)))
if (device->rs_total > (4UL << (30 - BM_BLOCK_SHIFT)))
seq_printf(seq, "(%lu/%lu)M",
(unsigned long) Bit2KB(rs_left >> 10),
(unsigned long) Bit2KB(mdev->rs_total >> 10));
(unsigned long) Bit2KB(device->rs_total >> 10));
else
seq_printf(seq, "(%lu/%lu)K\n\t",
(unsigned long) Bit2KB(rs_left),
(unsigned long) Bit2KB(mdev->rs_total));
(unsigned long) Bit2KB(device->rs_total));
/* see drivers/md/md.c
* We do not want to overflow, so the order of operands and
@ -114,14 +114,14 @@ static void drbd_syncer_progress(struct drbd_conf *mdev, struct seq_file *seq)
* at least (DRBD_SYNC_MARKS-2)*DRBD_SYNC_MARK_STEP old, and has at
* least DRBD_SYNC_MARK_STEP time before it will be modified. */
/* ------------------------ ~18s average ------------------------ */
i = (mdev->rs_last_mark + 2) % DRBD_SYNC_MARKS;
dt = (jiffies - mdev->rs_mark_time[i]) / HZ;
i = (device->rs_last_mark + 2) % DRBD_SYNC_MARKS;
dt = (jiffies - device->rs_mark_time[i]) / HZ;
if (dt > (DRBD_SYNC_MARK_STEP * DRBD_SYNC_MARKS))
stalled = 1;
if (!dt)
dt++;
db = mdev->rs_mark_left[i] - rs_left;
db = device->rs_mark_left[i] - rs_left;
rt = (dt * (rs_left / (db/100+1)))/100; /* seconds */
seq_printf(seq, "finish: %lu:%02lu:%02lu",
@ -134,11 +134,11 @@ static void drbd_syncer_progress(struct drbd_conf *mdev, struct seq_file *seq)
/* ------------------------- ~3s average ------------------------ */
if (proc_details >= 1) {
/* this is what drbd_rs_should_slow_down() uses */
i = (mdev->rs_last_mark + DRBD_SYNC_MARKS-1) % DRBD_SYNC_MARKS;
dt = (jiffies - mdev->rs_mark_time[i]) / HZ;
i = (device->rs_last_mark + DRBD_SYNC_MARKS-1) % DRBD_SYNC_MARKS;
dt = (jiffies - device->rs_mark_time[i]) / HZ;
if (!dt)
dt++;
db = mdev->rs_mark_left[i] - rs_left;
db = device->rs_mark_left[i] - rs_left;
dbdt = Bit2KB(db/dt);
seq_printf_with_thousands_grouping(seq, dbdt);
seq_printf(seq, " -- ");
@ -147,34 +147,34 @@ static void drbd_syncer_progress(struct drbd_conf *mdev, struct seq_file *seq)
/* --------------------- long term average ---------------------- */
/* mean speed since syncer started
* we do account for PausedSync periods */
dt = (jiffies - mdev->rs_start - mdev->rs_paused) / HZ;
dt = (jiffies - device->rs_start - device->rs_paused) / HZ;
if (dt == 0)
dt = 1;
db = mdev->rs_total - rs_left;
db = device->rs_total - rs_left;
dbdt = Bit2KB(db/dt);
seq_printf_with_thousands_grouping(seq, dbdt);
seq_printf(seq, ")");
if (mdev->state.conn == C_SYNC_TARGET ||
mdev->state.conn == C_VERIFY_S) {
if (device->state.conn == C_SYNC_TARGET ||
device->state.conn == C_VERIFY_S) {
seq_printf(seq, " want: ");
seq_printf_with_thousands_grouping(seq, mdev->c_sync_rate);
seq_printf_with_thousands_grouping(seq, device->c_sync_rate);
}
seq_printf(seq, " K/sec%s\n", stalled ? " (stalled)" : "");
if (proc_details >= 1) {
/* 64 bit:
* we convert to sectors in the display below. */
unsigned long bm_bits = drbd_bm_bits(mdev);
unsigned long bm_bits = drbd_bm_bits(device);
unsigned long bit_pos;
unsigned long long stop_sector = 0;
if (mdev->state.conn == C_VERIFY_S ||
mdev->state.conn == C_VERIFY_T) {
bit_pos = bm_bits - mdev->ov_left;
if (verify_can_do_stop_sector(mdev))
stop_sector = mdev->ov_stop_sector;
if (device->state.conn == C_VERIFY_S ||
device->state.conn == C_VERIFY_T) {
bit_pos = bm_bits - device->ov_left;
if (verify_can_do_stop_sector(device))
stop_sector = device->ov_stop_sector;
} else
bit_pos = mdev->bm_resync_fo;
bit_pos = device->bm_resync_fo;
/* Total sectors may be slightly off for oddly
* sized devices. So what. */
seq_printf(seq,
@ -202,7 +202,7 @@ static int drbd_seq_show(struct seq_file *seq, void *v)
{
int i, prev_i = -1;
const char *sn;
struct drbd_conf *mdev;
struct drbd_device *device;
struct net_conf *nc;
char wp;
@ -236,72 +236,72 @@ static int drbd_seq_show(struct seq_file *seq, void *v)
*/
rcu_read_lock();
idr_for_each_entry(&minors, mdev, i) {
idr_for_each_entry(&drbd_devices, device, i) {
if (prev_i != i - 1)
seq_printf(seq, "\n");
prev_i = i;
sn = drbd_conn_str(mdev->state.conn);
sn = drbd_conn_str(device->state.conn);
if (mdev->state.conn == C_STANDALONE &&
mdev->state.disk == D_DISKLESS &&
mdev->state.role == R_SECONDARY) {
if (device->state.conn == C_STANDALONE &&
device->state.disk == D_DISKLESS &&
device->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);
/* reset device->congestion_reason */
bdi_rw_congested(&device->rq_queue->backing_dev_info);
nc = rcu_dereference(mdev->tconn->net_conf);
nc = rcu_dereference(first_peer_device(device)->connection->net_conf);
wp = nc ? nc->wire_protocol - DRBD_PROT_A + 'A' : ' ';
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 "
"lo:%d pe:%d ua:%d ap:%d ep:%d wo:%c",
i, sn,
drbd_role_str(mdev->state.role),
drbd_role_str(mdev->state.peer),
drbd_disk_str(mdev->state.disk),
drbd_disk_str(mdev->state.pdsk),
drbd_role_str(device->state.role),
drbd_role_str(device->state.peer),
drbd_disk_str(device->state.disk),
drbd_disk_str(device->state.pdsk),
wp,
drbd_suspended(mdev) ? 's' : 'r',
mdev->state.aftr_isp ? 'a' : '-',
mdev->state.peer_isp ? 'p' : '-',
mdev->state.user_isp ? 'u' : '-',
mdev->congestion_reason ?: '-',
test_bit(AL_SUSPENDED, &mdev->flags) ? 's' : '-',
mdev->send_cnt/2,
mdev->recv_cnt/2,
mdev->writ_cnt/2,
mdev->read_cnt/2,
mdev->al_writ_cnt,
mdev->bm_writ_cnt,
atomic_read(&mdev->local_cnt),
atomic_read(&mdev->ap_pending_cnt) +
atomic_read(&mdev->rs_pending_cnt),
atomic_read(&mdev->unacked_cnt),
atomic_read(&mdev->ap_bio_cnt),
mdev->tconn->epochs,
write_ordering_chars[mdev->tconn->write_ordering]
drbd_suspended(device) ? 's' : 'r',
device->state.aftr_isp ? 'a' : '-',
device->state.peer_isp ? 'p' : '-',
device->state.user_isp ? 'u' : '-',
device->congestion_reason ?: '-',
test_bit(AL_SUSPENDED, &device->flags) ? 's' : '-',
device->send_cnt/2,
device->recv_cnt/2,
device->writ_cnt/2,
device->read_cnt/2,
device->al_writ_cnt,
device->bm_writ_cnt,
atomic_read(&device->local_cnt),
atomic_read(&device->ap_pending_cnt) +
atomic_read(&device->rs_pending_cnt),
atomic_read(&device->unacked_cnt),
atomic_read(&device->ap_bio_cnt),
first_peer_device(device)->connection->epochs,
write_ordering_chars[first_peer_device(device)->connection->write_ordering]
);
seq_printf(seq, " oos:%llu\n",
Bit2KB((unsigned long long)
drbd_bm_total_weight(mdev)));
drbd_bm_total_weight(device)));
}
if (mdev->state.conn == C_SYNC_SOURCE ||
mdev->state.conn == C_SYNC_TARGET ||
mdev->state.conn == C_VERIFY_S ||
mdev->state.conn == C_VERIFY_T)
drbd_syncer_progress(mdev, seq);
if (device->state.conn == C_SYNC_SOURCE ||
device->state.conn == C_SYNC_TARGET ||
device->state.conn == C_VERIFY_S ||
device->state.conn == C_VERIFY_T)
drbd_syncer_progress(device, seq);
if (proc_details >= 1 && get_ldev_if_state(mdev, D_FAILED)) {
lc_seq_printf_stats(seq, mdev->resync);
lc_seq_printf_stats(seq, mdev->act_log);
put_ldev(mdev);
if (proc_details >= 1 && get_ldev_if_state(device, D_FAILED)) {
lc_seq_printf_stats(seq, device->resync);
lc_seq_printf_stats(seq, device->act_log);
put_ldev(device);
}
if (proc_details >= 2) {
if (mdev->resync) {
lc_seq_dump_details(seq, mdev->resync, "rs_left",
if (device->resync) {
lc_seq_dump_details(seq, device->resync, "rs_left",
resync_dump_detail);
}
}

295
drivers/block/drbd/drbd_protocol.h Normal file
View File

@ -0,0 +1,295 @@
#ifndef __DRBD_PROTOCOL_H
#define __DRBD_PROTOCOL_H
enum drbd_packet {
/* receiver (data socket) */
P_DATA = 0x00,
P_DATA_REPLY = 0x01, /* Response to P_DATA_REQUEST */
P_RS_DATA_REPLY = 0x02, /* Response to P_RS_DATA_REQUEST */
P_BARRIER = 0x03,
P_BITMAP = 0x04,
P_BECOME_SYNC_TARGET = 0x05,
P_BECOME_SYNC_SOURCE = 0x06,
P_UNPLUG_REMOTE = 0x07, /* Used at various times to hint the peer */
P_DATA_REQUEST = 0x08, /* Used to ask for a data block */
P_RS_DATA_REQUEST = 0x09, /* Used to ask for a data block for resync */
P_SYNC_PARAM = 0x0a,
P_PROTOCOL = 0x0b,
P_UUIDS = 0x0c,
P_SIZES = 0x0d,
P_STATE = 0x0e,
P_SYNC_UUID = 0x0f,
P_AUTH_CHALLENGE = 0x10,
P_AUTH_RESPONSE = 0x11,
P_STATE_CHG_REQ = 0x12,
/* asender (meta socket */
P_PING = 0x13,
P_PING_ACK = 0x14,
P_RECV_ACK = 0x15, /* Used in protocol B */
P_WRITE_ACK = 0x16, /* Used in protocol C */
P_RS_WRITE_ACK = 0x17, /* Is a P_WRITE_ACK, additionally call set_in_sync(). */
P_SUPERSEDED = 0x18, /* Used in proto C, two-primaries conflict detection */
P_NEG_ACK = 0x19, /* Sent if local disk is unusable */
P_NEG_DREPLY = 0x1a, /* Local disk is broken... */
P_NEG_RS_DREPLY = 0x1b, /* Local disk is broken... */
P_BARRIER_ACK = 0x1c,
P_STATE_CHG_REPLY = 0x1d,
/* "new" commands, no longer fitting into the ordering scheme above */
P_OV_REQUEST = 0x1e, /* data socket */
P_OV_REPLY = 0x1f,
P_OV_RESULT = 0x20, /* meta socket */
P_CSUM_RS_REQUEST = 0x21, /* data socket */
P_RS_IS_IN_SYNC = 0x22, /* meta socket */
P_SYNC_PARAM89 = 0x23, /* data socket, protocol version 89 replacement for P_SYNC_PARAM */
P_COMPRESSED_BITMAP = 0x24, /* compressed or otherwise encoded bitmap transfer */
/* P_CKPT_FENCE_REQ = 0x25, * currently reserved for protocol D */
/* P_CKPT_DISABLE_REQ = 0x26, * currently reserved for protocol D */
P_DELAY_PROBE = 0x27, /* is used on BOTH sockets */
P_OUT_OF_SYNC = 0x28, /* Mark as out of sync (Outrunning), data socket */
P_RS_CANCEL = 0x29, /* meta: Used to cancel RS_DATA_REQUEST packet by SyncSource */
P_CONN_ST_CHG_REQ = 0x2a, /* data sock: Connection wide state request */
P_CONN_ST_CHG_REPLY = 0x2b, /* meta sock: Connection side state req reply */
P_RETRY_WRITE = 0x2c, /* Protocol C: retry conflicting write request */
P_PROTOCOL_UPDATE = 0x2d, /* data sock: is used in established connections */
P_MAY_IGNORE = 0x100, /* Flag to test if (cmd > P_MAY_IGNORE) ... */
P_MAX_OPT_CMD = 0x101,
/* special command ids for handshake */
P_INITIAL_META = 0xfff1, /* First Packet on the MetaSock */
P_INITIAL_DATA = 0xfff2, /* First Packet on the Socket */
P_CONNECTION_FEATURES = 0xfffe /* FIXED for the next century! */
};
#ifndef __packed
#define __packed __attribute__((packed))
#endif
/* This is the layout for a packet on the wire.
* The byteorder is the network byte order.
* (except block_id and barrier fields.
* these are pointers to local structs
* and have no relevance for the partner,
* which just echoes them as received.)
*
* NOTE that the payload starts at a long aligned offset,
* regardless of 32 or 64 bit arch!
*/
struct p_header80 {
u32 magic;
u16 command;
u16 length; /* bytes of data after this header */
} __packed;
/* Header for big packets, Used for data packets exceeding 64kB */
struct p_header95 {
u16 magic; /* use DRBD_MAGIC_BIG here */
u16 command;
u32 length;
} __packed;
struct p_header100 {
u32 magic;
u16 volume;
u16 command;
u32 length;
u32 pad;
} __packed;
/* these defines must not be changed without changing the protocol version */
#define DP_HARDBARRIER 1 /* depricated */
#define DP_RW_SYNC 2 /* equals REQ_SYNC */
#define DP_MAY_SET_IN_SYNC 4
#define DP_UNPLUG 8 /* not used anymore */
#define DP_FUA 16 /* equals REQ_FUA */
#define DP_FLUSH 32 /* equals REQ_FLUSH */
#define DP_DISCARD 64 /* equals REQ_DISCARD */
#define DP_SEND_RECEIVE_ACK 128 /* This is a proto B write request */
#define DP_SEND_WRITE_ACK 256 /* This is a proto C write request */
struct p_data {
u64 sector; /* 64 bits sector number */
u64 block_id; /* to identify the request in protocol B&C */
u32 seq_num;
u32 dp_flags;
} __packed;
/*
* commands which share a struct:
* p_block_ack:
* P_RECV_ACK (proto B), P_WRITE_ACK (proto C),
* P_SUPERSEDED (proto C, two-primaries conflict detection)
* p_block_req:
* P_DATA_REQUEST, P_RS_DATA_REQUEST
*/
struct p_block_ack {
u64 sector;
u64 block_id;
u32 blksize;
u32 seq_num;
} __packed;
struct p_block_req {
u64 sector;
u64 block_id;
u32 blksize;
u32 pad; /* to multiple of 8 Byte */
} __packed;
/*
* commands with their own struct for additional fields:
* P_CONNECTION_FEATURES
* P_BARRIER
* P_BARRIER_ACK
* P_SYNC_PARAM
* ReportParams
*/
struct p_connection_features {
u32 protocol_min;
u32 feature_flags;
u32 protocol_max;
/* should be more than enough for future enhancements
* for now, feature_flags and the reserved array shall be zero.
*/
u32 _pad;
u64 reserved[7];
} __packed;
struct p_barrier {
u32 barrier; /* barrier number _handle_ only */
u32 pad; /* to multiple of 8 Byte */
} __packed;
struct p_barrier_ack {
u32 barrier;
u32 set_size;
} __packed;
struct p_rs_param {
u32 resync_rate;
/* Since protocol version 88 and higher. */
char verify_alg[0];
} __packed;
struct p_rs_param_89 {
u32 resync_rate;
/* protocol version 89: */
char verify_alg[SHARED_SECRET_MAX];
char csums_alg[SHARED_SECRET_MAX];
} __packed;
struct p_rs_param_95 {
u32 resync_rate;
char verify_alg[SHARED_SECRET_MAX];
char csums_alg[SHARED_SECRET_MAX];
u32 c_plan_ahead;
u32 c_delay_target;
u32 c_fill_target;
u32 c_max_rate;
} __packed;
enum drbd_conn_flags {
CF_DISCARD_MY_DATA = 1,
CF_DRY_RUN = 2,
};
struct p_protocol {
u32 protocol;
u32 after_sb_0p;
u32 after_sb_1p;
u32 after_sb_2p;
u32 conn_flags;
u32 two_primaries;
/* Since protocol version 87 and higher. */
char integrity_alg[0];
} __packed;
struct p_uuids {
u64 uuid[UI_EXTENDED_SIZE];
} __packed;
struct p_rs_uuid {
u64 uuid;
} __packed;
struct p_sizes {
u64 d_size; /* size of disk */
u64 u_size; /* user requested size */
u64 c_size; /* current exported size */
u32 max_bio_size; /* Maximal size of a BIO */
u16 queue_order_type; /* not yet implemented in DRBD*/
u16 dds_flags; /* use enum dds_flags here. */
} __packed;
struct p_state {
u32 state;
} __packed;
struct p_req_state {
u32 mask;
u32 val;
} __packed;
struct p_req_state_reply {
u32 retcode;
} __packed;
struct p_drbd06_param {
u64 size;
u32 state;
u32 blksize;
u32 protocol;
u32 version;
u32 gen_cnt[5];
u32 bit_map_gen[5];
} __packed;
struct p_block_desc {
u64 sector;
u32 blksize;
u32 pad; /* to multiple of 8 Byte */
} __packed;
/* Valid values for the encoding field.
* Bump proto version when changing this. */
enum drbd_bitmap_code {
/* RLE_VLI_Bytes = 0,
* and other bit variants had been defined during
* algorithm evaluation. */
RLE_VLI_Bits = 2,
};
struct p_compressed_bm {
/* (encoding & 0x0f): actual encoding, see enum drbd_bitmap_code
* (encoding & 0x80): polarity (set/unset) of first runlength
* ((encoding >> 4) & 0x07): pad_bits, number of trailing zero bits
* used to pad up to head.length bytes
*/
u8 encoding;
u8 code[0];
} __packed;
struct p_delay_probe93 {
u32 seq_num; /* sequence number to match the two probe packets */
u32 offset; /* usecs the probe got sent after the reference time point */
} __packed;
/*
* Bitmap packets need to fit within a single page on the sender and receiver,
* so we are limited to 4 KiB (and not to PAGE_SIZE, which can be bigger).
*/
#define DRBD_SOCKET_BUFFER_SIZE 4096
#endif /* __DRBD_PROTOCOL_H */

2530
drivers/block/drbd/drbd_receiver.c
View File

File diff suppressed because it is too large Load Diff

464
drivers/block/drbd/drbd_req.c
View File

File diff suppressed because it is too large Load Diff

20
drivers/block/drbd/drbd_req.h
View File

@ -275,17 +275,17 @@ struct bio_and_error {
int error;
};
extern void start_new_tl_epoch(struct drbd_tconn *tconn);
extern void start_new_tl_epoch(struct drbd_connection *connection);
extern void drbd_req_destroy(struct kref *kref);
extern void _req_may_be_done(struct drbd_request *req,
struct bio_and_error *m);
extern int __req_mod(struct drbd_request *req, enum drbd_req_event what,
struct bio_and_error *m);
extern void complete_master_bio(struct drbd_conf *mdev,
extern void complete_master_bio(struct drbd_device *device,
struct bio_and_error *m);
extern void request_timer_fn(unsigned long data);
extern void tl_restart(struct drbd_tconn *tconn, enum drbd_req_event what);
extern void _tl_restart(struct drbd_tconn *tconn, enum drbd_req_event what);
extern void tl_restart(struct drbd_connection *connection, enum drbd_req_event what);
extern void _tl_restart(struct drbd_connection *connection, enum drbd_req_event what);
/* this is in drbd_main.c */
extern void drbd_restart_request(struct drbd_request *req);
@ -294,14 +294,14 @@ extern void drbd_restart_request(struct drbd_request *req);
* outside the spinlock, e.g. when walking some list on cleanup. */
static inline int _req_mod(struct drbd_request *req, enum drbd_req_event what)
{
struct drbd_conf *mdev = req->w.mdev;
struct drbd_device *device = req->device;
struct bio_and_error m;
int rv;
/* __req_mod possibly frees req, do not touch req after that! */
rv = __req_mod(req, what, &m);
if (m.bio)
complete_master_bio(mdev, &m);
complete_master_bio(device, &m);
return rv;
}
@ -314,16 +314,16 @@ static inline int req_mod(struct drbd_request *req,
enum drbd_req_event what)
{
unsigned long flags;
struct drbd_conf *mdev = req->w.mdev;
struct drbd_device *device = req->device;
struct bio_and_error m;
int rv;
spin_lock_irqsave(&mdev->tconn->req_lock, flags);
spin_lock_irqsave(&device->resource->req_lock, flags);
rv = __req_mod(req, what, &m);
spin_unlock_irqrestore(&mdev->tconn->req_lock, flags);
spin_unlock_irqrestore(&device->resource->req_lock, flags);
if (m.bio)
complete_master_bio(mdev, &m);
complete_master_bio(device, &m);
return rv;
}

859
drivers/block/drbd/drbd_state.c
View File

File diff suppressed because it is too large Load Diff

40
drivers/block/drbd/drbd_state.h
View File

@ -1,8 +1,8 @@
#ifndef DRBD_STATE_H
#define DRBD_STATE_H
struct drbd_conf;
struct drbd_tconn;
struct drbd_device;
struct drbd_connection;
/**
* DOC: DRBD State macros
@ -107,36 +107,36 @@ union drbd_dev_state {
unsigned int i;
};
extern enum drbd_state_rv drbd_change_state(struct drbd_conf *mdev,
extern enum drbd_state_rv drbd_change_state(struct drbd_device *device,
enum chg_state_flags f,
union drbd_state mask,
union drbd_state val);
extern void drbd_force_state(struct drbd_conf *, union drbd_state,
extern void drbd_force_state(struct drbd_device *, union drbd_state,
union drbd_state);
extern enum drbd_state_rv _drbd_request_state(struct drbd_conf *,
extern enum drbd_state_rv _drbd_request_state(struct drbd_device *,
union drbd_state,
union drbd_state,
enum chg_state_flags);
extern enum drbd_state_rv __drbd_set_state(struct drbd_conf *, union drbd_state,
extern enum drbd_state_rv __drbd_set_state(struct drbd_device *, union drbd_state,
enum chg_state_flags,
struct completion *done);
extern void print_st_err(struct drbd_conf *, union drbd_state,
extern void print_st_err(struct drbd_device *, union drbd_state,
union drbd_state, int);
enum drbd_state_rv
_conn_request_state(struct drbd_tconn *tconn, union drbd_state mask, union drbd_state val,
_conn_request_state(struct drbd_connection *connection, union drbd_state mask, union drbd_state val,
enum chg_state_flags flags);
enum drbd_state_rv
conn_request_state(struct drbd_tconn *tconn, union drbd_state mask, union drbd_state val,
conn_request_state(struct drbd_connection *connection, union drbd_state mask, union drbd_state val,
enum chg_state_flags flags);
extern void drbd_resume_al(struct drbd_conf *mdev);
extern bool conn_all_vols_unconf(struct drbd_tconn *tconn);
extern void drbd_resume_al(struct drbd_device *device);
extern bool conn_all_vols_unconf(struct drbd_connection *connection);
/**
* drbd_request_state() - Reqest a state change
* @mdev: DRBD device.
* @device: DRBD device.
* @mask: mask of state bits to change.
* @val: value of new state bits.
*
@ -144,18 +144,18 @@ extern bool conn_all_vols_unconf(struct drbd_tconn *tconn);
* quite verbose in case the state change is not possible, and all those
* state changes are globally serialized.
*/
static inline int drbd_request_state(struct drbd_conf *mdev,
static inline int drbd_request_state(struct drbd_device *device,
union drbd_state mask,
union drbd_state val)
{
return _drbd_request_state(mdev, mask, val, CS_VERBOSE + CS_ORDERED);
return _drbd_request_state(device, mask, val, CS_VERBOSE + CS_ORDERED);
}
enum drbd_role conn_highest_role(struct drbd_tconn *tconn);
enum drbd_role conn_highest_peer(struct drbd_tconn *tconn);
enum drbd_disk_state conn_highest_disk(struct drbd_tconn *tconn);
enum drbd_disk_state conn_lowest_disk(struct drbd_tconn *tconn);
enum drbd_disk_state conn_highest_pdsk(struct drbd_tconn *tconn);
enum drbd_conns conn_lowest_conn(struct drbd_tconn *tconn);
enum drbd_role conn_highest_role(struct drbd_connection *connection);
enum drbd_role conn_highest_peer(struct drbd_connection *connection);
enum drbd_disk_state conn_highest_disk(struct drbd_connection *connection);
enum drbd_disk_state conn_lowest_disk(struct drbd_connection *connection);
enum drbd_disk_state conn_highest_pdsk(struct drbd_connection *connection);
enum drbd_conns conn_lowest_conn(struct drbd_connection *connection);
#endif

1
drivers/block/drbd/drbd_strings.c
View File

@ -24,6 +24,7 @@
*/
#include <linux/drbd.h>
#include "drbd_strings.h"
static const char *drbd_conn_s_names[] = {
[C_STANDALONE] = "StandAlone",

9
drivers/block/drbd/drbd_strings.h Normal file
View File

@ -0,0 +1,9 @@
#ifndef __DRBD_STRINGS_H
#define __DRBD_STRINGS_H
extern const char *drbd_conn_str(enum drbd_conns);
extern const char *drbd_role_str(enum drbd_role);
extern const char *drbd_disk_str(enum drbd_disk_state);
extern const char *drbd_set_st_err_str(enum drbd_state_rv);
#endif /* __DRBD_STRINGS_H */

944
drivers/block/drbd/drbd_worker.c
View File

File diff suppressed because it is too large Load Diff

14
drivers/block/drbd/drbd_wrappers.h
View File

@ -9,12 +9,12 @@
extern char *drbd_sec_holder;
/* sets the number of 512 byte sectors of our virtual device */
static inline void drbd_set_my_capacity(struct drbd_conf *mdev,
static inline void drbd_set_my_capacity(struct drbd_device *device,
sector_t size)
{
/* set_capacity(mdev->this_bdev->bd_disk, size); */
set_capacity(mdev->vdisk, size);
mdev->this_bdev->bd_inode->i_size = (loff_t)size << 9;
/* set_capacity(device->this_bdev->bd_disk, size); */
set_capacity(device->vdisk, size);
device->this_bdev->bd_inode->i_size = (loff_t)size << 9;
}
#define drbd_bio_uptodate(bio) bio_flagged(bio, BIO_UPTODATE)
@ -27,20 +27,20 @@ extern void drbd_request_endio(struct bio *bio, int error);
/*
* used to submit our private bio
*/
static inline void drbd_generic_make_request(struct drbd_conf *mdev,
static inline void drbd_generic_make_request(struct drbd_device *device,
int fault_type, struct bio *bio)
{
__release(local);
if (!bio->bi_bdev) {
printk(KERN_ERR "drbd%d: drbd_generic_make_request: "
"bio->bi_bdev == NULL\n",
mdev_to_minor(mdev));
device_to_minor(device));
dump_stack();
bio_endio(bio, -ENODEV);
return;
}
if (drbd_insert_fault(mdev, fault_type))
if (drbd_insert_fault(device, fault_type))
bio_endio(bio, -EIO);
else
generic_make_request(bio);

100
drivers/block/mtip32xx/mtip32xx.c
View File

@ -252,38 +252,45 @@ static void mtip_async_complete(struct mtip_port *port,
void *data,
int status)
{
struct mtip_cmd *command;
struct mtip_cmd *cmd;
struct driver_data *dd = data;
int cb_status = status ? -EIO : 0;
int unaligned, cb_status = status ? -EIO : 0;
void (*func)(void *, int);
if (unlikely(!dd) || unlikely(!port))
return;
command = &port->commands[tag];
cmd = &port->commands[tag];
if (unlikely(status == PORT_IRQ_TF_ERR)) {
dev_warn(&port->dd->pdev->dev,
"Command tag %d failed due to TFE\n", tag);
}
/* Upper layer callback */
if (likely(command->async_callback))
command->async_callback(command->async_data, cb_status);
command->async_callback = NULL;
command->comp_func = NULL;
/* Unmap the DMA scatter list entries */
dma_unmap_sg(&dd->pdev->dev,
command->sg,
command->scatter_ents,
command->direction);
/* Clear the allocated and active bits for the command */
/* Clear the active flag */
atomic_set(&port->commands[tag].active, 0);
release_slot(port, tag);
up(&port->cmd_slot);
/* Upper layer callback */
func = cmd->async_callback;
if (likely(func && cmpxchg(&cmd->async_callback, func, 0) == func)) {
/* Unmap the DMA scatter list entries */
dma_unmap_sg(&dd->pdev->dev,
cmd->sg,
cmd->scatter_ents,
cmd->direction);
func(cmd->async_data, cb_status);
unaligned = cmd->unaligned;
/* Clear the allocated bit for the command */
release_slot(port, tag);
if (unlikely(unaligned))
up(&port->cmd_slot_unal);
else
up(&port->cmd_slot);
}
}
/*
@ -660,11 +667,12 @@ static void mtip_timeout_function(unsigned long int data)
{
struct mtip_port *port = (struct mtip_port *) data;
struct host_to_dev_fis *fis;
struct mtip_cmd *command;
int tag, cmdto_cnt = 0;
struct mtip_cmd *cmd;
int unaligned, tag, cmdto_cnt = 0;
unsigned int bit, group;
unsigned int num_command_slots;
unsigned long to, tagaccum[SLOTBITS_IN_LONGS];
void (*func)(void *, int);
if (unlikely(!port))
return;
@ -694,8 +702,8 @@ static void mtip_timeout_function(unsigned long int data)
group = tag >> 5;
bit = tag & 0x1F;
command = &port->commands[tag];
fis = (struct host_to_dev_fis *) command->command;
cmd = &port->commands[tag];
fis = (struct host_to_dev_fis *) cmd->command;
set_bit(tag, tagaccum);
cmdto_cnt++;
@ -709,27 +717,30 @@ static void mtip_timeout_function(unsigned long int data)
*/
writel(1 << bit, port->completed[group]);
/* Call the async completion callback. */
if (likely(command->async_callback))
command->async_callback(command->async_data,
-EIO);
command->async_callback = NULL;
command->comp_func = NULL;
/* Unmap the DMA scatter list entries */
dma_unmap_sg(&port->dd->pdev->dev,
command->sg,
command->scatter_ents,
command->direction);
/*
* Clear the allocated bit and active tag for the
* command.
*/
/* Clear the active flag for the command */
atomic_set(&port->commands[tag].active, 0);
release_slot(port, tag);
up(&port->cmd_slot);
func = cmd->async_callback;
if (func &&
cmpxchg(&cmd->async_callback, func, 0) == func) {
/* Unmap the DMA scatter list entries */
dma_unmap_sg(&port->dd->pdev->dev,
cmd->sg,
cmd->scatter_ents,
cmd->direction);
func(cmd->async_data, -EIO);
unaligned = cmd->unaligned;
/* Clear the allocated bit for the command. */
release_slot(port, tag);
if (unaligned)
up(&port->cmd_slot_unal);
else
up(&port->cmd_slot);
}
}
}
@ -4213,6 +4224,7 @@ skip_create_disk:
blk_queue_max_hw_sectors(dd->queue, 0xffff);
blk_queue_max_segment_size(dd->queue, 0x400000);
blk_queue_io_min(dd->queue, 4096);
blk_queue_bounce_limit(dd->queue, dd->pdev->dma_mask);
/*
* write back cache is not supported in the device. FUA depends on
@ -4615,7 +4627,7 @@ static int mtip_pci_probe(struct pci_dev *pdev,
if (rv) {
dev_warn(&pdev->dev,
"Unable to enable MSI interrupt.\n");
goto block_initialize_err;
goto msi_initialize_err;
}
/* Initialize the block layer. */
@ -4645,6 +4657,8 @@ static int mtip_pci_probe(struct pci_dev *pdev,
block_initialize_err:
pci_disable_msi(pdev);
msi_initialize_err:
if (dd->isr_workq) {
flush_workqueue(dd->isr_workq);
destroy_workqueue(dd->isr_workq);

2
drivers/block/mtip32xx/mtip32xx.h
View File

@ -92,7 +92,7 @@
/* Driver name and version strings */
#define MTIP_DRV_NAME "mtip32xx"
#define MTIP_DRV_VERSION "1.3.0"
#define MTIP_DRV_VERSION "1.3.1"
/* Maximum number of minor device numbers per device. */
#define MTIP_MAX_MINORS 16

33
drivers/block/nvme-core.c
View File

@ -1836,31 +1836,16 @@ static int nvme_setup_io_queues(struct nvme_dev *dev)
/* Deregister the admin queue's interrupt */
free_irq(dev->entry[0].vector, adminq);
vecs = nr_io_queues;
for (i = 0; i < vecs; i++)
for (i = 0; i < nr_io_queues; i++)
dev->entry[i].entry = i;
for (;;) {
result = pci_enable_msix(pdev, dev->entry, vecs);
if (result <= 0)
break;
vecs = result;
}
if (result < 0) {
vecs = nr_io_queues;
if (vecs > 32)
vecs = 32;
for (;;) {
result = pci_enable_msi_block(pdev, vecs);
if (result == 0) {
for (i = 0; i < vecs; i++)
dev->entry[i].vector = i + pdev->irq;
break;
} else if (result < 0) {
vecs = 1;
break;
}
vecs = result;
vecs = pci_enable_msix_range(pdev, dev->entry, 1, nr_io_queues);
if (vecs < 0) {
vecs = pci_enable_msi_range(pdev, 1, min(nr_io_queues, 32));
if (vecs < 0) {
vecs = 1;
} else {
for (i = 0; i < vecs; i++)
dev->entry[i].vector = i + pdev->irq;
}
}

67
drivers/block/skd_main.c
View File

@ -3910,18 +3910,22 @@ static void skd_release_msix(struct skd_device *skdev)
struct skd_msix_entry *qentry;
int i;
if (skdev->msix_entries == NULL)
return;
for (i = 0; i < skdev->msix_count; i++) {
qentry = &skdev->msix_entries[i];
skdev = qentry->rsp;
if (skdev->msix_entries) {
for (i = 0; i < skdev->msix_count; i++) {
qentry = &skdev->msix_entries[i];
skdev = qentry->rsp;
if (qentry->have_irq)
devm_free_irq(&skdev->pdev->dev,
qentry->vector, qentry->rsp);
if (qentry->have_irq)
devm_free_irq(&skdev->pdev->dev,
qentry->vector, qentry->rsp);
}
kfree(skdev->msix_entries);
}
pci_disable_msix(skdev->pdev);
kfree(skdev->msix_entries);
if (skdev->msix_count)
pci_disable_msix(skdev->pdev);
skdev->msix_count = 0;
skdev->msix_entries = NULL;
}
@ -3929,12 +3933,10 @@ static void skd_release_msix(struct skd_device *skdev)
static int skd_acquire_msix(struct skd_device *skdev)
{
int i, rc;
struct pci_dev *pdev;
struct msix_entry *entries = NULL;
struct pci_dev *pdev = skdev->pdev;
struct msix_entry *entries;
struct skd_msix_entry *qentry;
pdev = skdev->pdev;
skdev->msix_count = SKD_MAX_MSIX_COUNT;
entries = kzalloc(sizeof(struct msix_entry) * SKD_MAX_MSIX_COUNT,
GFP_KERNEL);
if (!entries)
@ -3943,40 +3945,26 @@ static int skd_acquire_msix(struct skd_device *skdev)
for (i = 0; i < SKD_MAX_MSIX_COUNT; i++)
entries[i].entry = i;
rc = pci_enable_msix(pdev, entries, SKD_MAX_MSIX_COUNT);
if (rc < 0)
rc = pci_enable_msix_range(pdev, entries,
SKD_MIN_MSIX_COUNT, SKD_MAX_MSIX_COUNT);
if (rc < 0) {
pr_err("(%s): failed to enable MSI-X %d\n",
skd_name(skdev), rc);
goto msix_out;
if (rc) {
if (rc < SKD_MIN_MSIX_COUNT) {
pr_err("(%s): failed to enable MSI-X %d\n",
skd_name(skdev), rc);
goto msix_out;
}
pr_debug("%s:%s:%d %s: <%s> allocated %d MSI-X vectors\n",
skdev->name, __func__, __LINE__,
pci_name(pdev), skdev->name, rc);
skdev->msix_count = rc;
rc = pci_enable_msix(pdev, entries, skdev->msix_count);
if (rc) {
pr_err("(%s): failed to enable MSI-X "
"support (%d) %d\n",
skd_name(skdev), skdev->msix_count, rc);
goto msix_out;
}
}
skdev->msix_count = rc;
skdev->msix_entries = kzalloc(sizeof(struct skd_msix_entry) *
skdev->msix_count, GFP_KERNEL);
if (!skdev->msix_entries) {
rc = -ENOMEM;
skdev->msix_count = 0;
pr_err("(%s): msix table allocation error\n",
skd_name(skdev));
goto msix_out;
}
qentry = skdev->msix_entries;
for (i = 0; i < skdev->msix_count; i++) {
qentry = &skdev->msix_entries[i];
qentry->vector = entries[i].vector;
qentry->entry = entries[i].entry;
qentry->rsp = NULL;
@ -3985,11 +3973,10 @@ static int skd_acquire_msix(struct skd_device *skdev)
skdev->name, __func__, __LINE__,
pci_name(pdev), skdev->name,
i, qentry->vector, qentry->entry);
qentry++;
}
/* Enable MSI-X vectors for the base queue */
for (i = 0; i < SKD_MAX_MSIX_COUNT; i++) {
for (i = 0; i < skdev->msix_count; i++) {
qentry = &skdev->msix_entries[i];
snprintf(qentry->isr_name, sizeof(qentry->isr_name),
"%s%d-msix %s", DRV_NAME, skdev->devno,
@ -4045,8 +4032,8 @@ RETRY_IRQ_TYPE:
case SKD_IRQ_MSI:
snprintf(skdev->isr_name, sizeof(skdev->isr_name), "%s%d-msi",
DRV_NAME, skdev->devno);
rc = pci_enable_msi(pdev);
if (!rc) {
rc = pci_enable_msi_range(pdev, 1, 1);
if (rc > 0) {
rc = devm_request_irq(&pdev->dev, pdev->irq, skd_isr, 0,
skdev->isr_name, skdev);
if (rc) {

18
drivers/block/swim3.c
View File

@ -30,6 +30,7 @@
#include <linux/mutex.h>
#include <linux/module.h>
#include <linux/spinlock.h>
#include <linux/wait.h>
#include <asm/io.h>
#include <asm/dbdma.h>
#include <asm/prom.h>
@ -840,14 +841,17 @@ static int grab_drive(struct floppy_state *fs, enum swim_state state,
spin_lock_irqsave(&swim3_lock, flags);
if (fs->state != idle && fs->state != available) {
++fs->wanted;
while (fs->state != available) {
/* this will enable irqs in order to sleep */
if (!interruptible)
wait_event_lock_irq(fs->wait,
fs->state == available,
swim3_lock);
else if (wait_event_interruptible_lock_irq(fs->wait,
fs->state == available,
swim3_lock)) {
--fs->wanted;
spin_unlock_irqrestore(&swim3_lock, flags);
if (interruptible && signal_pending(current)) {
--fs->wanted;
return -EINTR;
}
interruptible_sleep_on(&fs->wait);
spin_lock_irqsave(&swim3_lock, flags);
return -EINTR;
}
--fs->wanted;
}

8
drivers/md/bcache/Kconfig
View File

@ -24,11 +24,3 @@ config BCACHE_CLOSURES_DEBUG
Keeps all active closures in a linked list and provides a debugfs
interface to list them, which makes it possible to see asynchronous
operations that get stuck.
# cgroup code needs to be updated:
#
#config CGROUP_BCACHE
# bool "Cgroup controls for bcache"
# depends on BCACHE && BLK_CGROUP
# ---help---
# TODO

173
drivers/md/bcache/alloc.c
View File

@ -78,12 +78,6 @@ uint8_t bch_inc_gen(struct cache *ca, struct bucket *b)
ca->set->need_gc = max(ca->set->need_gc, bucket_gc_gen(b));
WARN_ON_ONCE(ca->set->need_gc > BUCKET_GC_GEN_MAX);
if (CACHE_SYNC(&ca->set->sb)) {
ca->need_save_prio = max(ca->need_save_prio,
bucket_disk_gen(b));
WARN_ON_ONCE(ca->need_save_prio > BUCKET_DISK_GEN_MAX);
}
return ret;
}
@ -120,51 +114,45 @@ void bch_rescale_priorities(struct cache_set *c, int sectors)
mutex_unlock(&c->bucket_lock);
}
/* Allocation */
/*
* Background allocation thread: scans for buckets to be invalidated,
* invalidates them, rewrites prios/gens (marking them as invalidated on disk),
* then optionally issues discard commands to the newly free buckets, then puts
* them on the various freelists.
*/
static inline bool can_inc_bucket_gen(struct bucket *b)
{
return bucket_gc_gen(b) < BUCKET_GC_GEN_MAX &&
bucket_disk_gen(b) < BUCKET_DISK_GEN_MAX;
return bucket_gc_gen(b) < BUCKET_GC_GEN_MAX;
}
bool bch_bucket_add_unused(struct cache *ca, struct bucket *b)
bool bch_can_invalidate_bucket(struct cache *ca, struct bucket *b)
{
BUG_ON(GC_MARK(b) || GC_SECTORS_USED(b));
BUG_ON(!ca->set->gc_mark_valid);
if (CACHE_REPLACEMENT(&ca->sb) == CACHE_REPLACEMENT_FIFO) {
unsigned i;
for (i = 0; i < RESERVE_NONE; i++)
if (!fifo_full(&ca->free[i]))
goto add;
return false;
}
add:
b->prio = 0;
if (can_inc_bucket_gen(b) &&
fifo_push(&ca->unused, b - ca->buckets)) {
atomic_inc(&b->pin);
return true;
}
return false;
}
static bool can_invalidate_bucket(struct cache *ca, struct bucket *b)
{
return GC_MARK(b) == GC_MARK_RECLAIMABLE &&
return (!GC_MARK(b) ||
GC_MARK(b) == GC_MARK_RECLAIMABLE) &&
!atomic_read(&b->pin) &&
can_inc_bucket_gen(b);
}
static void invalidate_one_bucket(struct cache *ca, struct bucket *b)
void __bch_invalidate_one_bucket(struct cache *ca, struct bucket *b)
{
lockdep_assert_held(&ca->set->bucket_lock);
BUG_ON(GC_MARK(b) && GC_MARK(b) != GC_MARK_RECLAIMABLE);
if (GC_SECTORS_USED(b))
trace_bcache_invalidate(ca, b - ca->buckets);
bch_inc_gen(ca, b);
b->prio = INITIAL_PRIO;
atomic_inc(&b->pin);
}
static void bch_invalidate_one_bucket(struct cache *ca, struct bucket *b)
{
__bch_invalidate_one_bucket(ca, b);
fifo_push(&ca->free_inc, b - ca->buckets);
}
@ -195,20 +183,7 @@ static void invalidate_buckets_lru(struct cache *ca)
ca->heap.used = 0;
for_each_bucket(b, ca) {
/*
* If we fill up the unused list, if we then return before
* adding anything to the free_inc list we'll skip writing
* prios/gens and just go back to allocating from the unused
* list:
*/
if (fifo_full(&ca->unused))
return;
if (!can_invalidate_bucket(ca, b))
continue;
if (!GC_SECTORS_USED(b) &&
bch_bucket_add_unused(ca, b))
if (!bch_can_invalidate_bucket(ca, b))
continue;
if (!heap_full(&ca->heap))
@ -233,7 +208,7 @@ static void invalidate_buckets_lru(struct cache *ca)
return;
}
invalidate_one_bucket(ca, b);
bch_invalidate_one_bucket(ca, b);
}
}
@ -249,8 +224,8 @@ static void invalidate_buckets_fifo(struct cache *ca)
b = ca->buckets + ca->fifo_last_bucket++;
if (can_invalidate_bucket(ca, b))
invalidate_one_bucket(ca, b);
if (bch_can_invalidate_bucket(ca, b))
bch_invalidate_one_bucket(ca, b);
if (++checked >= ca->sb.nbuckets) {
ca->invalidate_needs_gc = 1;
@ -274,8 +249,8 @@ static void invalidate_buckets_random(struct cache *ca)
b = ca->buckets + n;
if (can_invalidate_bucket(ca, b))
invalidate_one_bucket(ca, b);
if (bch_can_invalidate_bucket(ca, b))
bch_invalidate_one_bucket(ca, b);
if (++checked >= ca->sb.nbuckets / 2) {
ca->invalidate_needs_gc = 1;
@ -287,8 +262,7 @@ static void invalidate_buckets_random(struct cache *ca)
static void invalidate_buckets(struct cache *ca)
{
if (ca->invalidate_needs_gc)
return;
BUG_ON(ca->invalidate_needs_gc);
switch (CACHE_REPLACEMENT(&ca->sb)) {
case CACHE_REPLACEMENT_LRU:
@ -301,8 +275,6 @@ static void invalidate_buckets(struct cache *ca)
invalidate_buckets_random(ca);
break;
}
trace_bcache_alloc_invalidate(ca);
}
#define allocator_wait(ca, cond) \
@ -350,17 +322,10 @@ static int bch_allocator_thread(void *arg)
* possibly issue discards to them, then we add the bucket to
* the free list:
*/
while (1) {
while (!fifo_empty(&ca->free_inc)) {
long bucket;
if ((!atomic_read(&ca->set->prio_blocked) ||
!CACHE_SYNC(&ca->set->sb)) &&
!fifo_empty(&ca->unused))
fifo_pop(&ca->unused, bucket);
else if (!fifo_empty(&ca->free_inc))
fifo_pop(&ca->free_inc, bucket);
else
break;
fifo_pop(&ca->free_inc, bucket);
if (ca->discard) {
mutex_unlock(&ca->set->bucket_lock);
@ -371,6 +336,7 @@ static int bch_allocator_thread(void *arg)
}
allocator_wait(ca, bch_allocator_push(ca, bucket));
wake_up(&ca->set->btree_cache_wait);
wake_up(&ca->set->bucket_wait);
}
@ -380,9 +346,9 @@ static int bch_allocator_thread(void *arg)
* them to the free_inc list:
*/
retry_invalidate:
allocator_wait(ca, ca->set->gc_mark_valid &&
(ca->need_save_prio > 64 ||
!ca->invalidate_needs_gc));
!ca->invalidate_needs_gc);
invalidate_buckets(ca);
/*
@ -390,13 +356,28 @@ static int bch_allocator_thread(void *arg)
* new stuff to them:
*/
allocator_wait(ca, !atomic_read(&ca->set->prio_blocked));
if (CACHE_SYNC(&ca->set->sb) &&
(!fifo_empty(&ca->free_inc) ||
ca->need_save_prio > 64))
if (CACHE_SYNC(&ca->set->sb)) {
/*
* This could deadlock if an allocation with a btree
* node locked ever blocked - having the btree node
* locked would block garbage collection, but here we're
* waiting on garbage collection before we invalidate
* and free anything.
*
* But this should be safe since the btree code always
* uses btree_check_reserve() before allocating now, and
* if it fails it blocks without btree nodes locked.
*/
if (!fifo_full(&ca->free_inc))
goto retry_invalidate;
bch_prio_write(ca);
}
}
}
/* Allocation */
long bch_bucket_alloc(struct cache *ca, unsigned reserve, bool wait)
{
DEFINE_WAIT(w);
@ -408,8 +389,10 @@ long bch_bucket_alloc(struct cache *ca, unsigned reserve, bool wait)
fifo_pop(&ca->free[reserve], r))
goto out;
if (!wait)
if (!wait) {
trace_bcache_alloc_fail(ca, reserve);
return -1;
}
do {
prepare_to_wait(&ca->set->bucket_wait, &w,
@ -425,6 +408,8 @@ long bch_bucket_alloc(struct cache *ca, unsigned reserve, bool wait)
out:
wake_up_process(ca->alloc_thread);
trace_bcache_alloc(ca, reserve);
if (expensive_debug_checks(ca->set)) {
size_t iter;
long i;
@ -438,8 +423,6 @@ out:
BUG_ON(i == r);
fifo_for_each(i, &ca->free_inc, iter)
BUG_ON(i == r);
fifo_for_each(i, &ca->unused, iter)
BUG_ON(i == r);
}
b = ca->buckets + r;
@ -461,17 +444,19 @@ out:
return r;
}
void __bch_bucket_free(struct cache *ca, struct bucket *b)
{
SET_GC_MARK(b, 0);
SET_GC_SECTORS_USED(b, 0);
}
void bch_bucket_free(struct cache_set *c, struct bkey *k)
{
unsigned i;
for (i = 0; i < KEY_PTRS(k); i++) {
struct bucket *b = PTR_BUCKET(c, k, i);
SET_GC_MARK(b, GC_MARK_RECLAIMABLE);
SET_GC_SECTORS_USED(b, 0);
bch_bucket_add_unused(PTR_CACHE(c, k, i), b);
}
for (i = 0; i < KEY_PTRS(k); i++)
__bch_bucket_free(PTR_CACHE(c, k, i),
PTR_BUCKET(c, k, i));
}
int __bch_bucket_alloc_set(struct cache_set *c, unsigned reserve,
@ -709,25 +694,3 @@ int bch_cache_allocator_start(struct cache *ca)
ca->alloc_thread = k;
return 0;
}
int bch_cache_allocator_init(struct cache *ca)
{
/*
* Reserve:
* Prio/gen writes first
* Then 8 for btree allocations
* Then half for the moving garbage collector
*/
#if 0
ca->watermark[WATERMARK_PRIO] = 0;
ca->watermark[WATERMARK_METADATA] = prio_buckets(ca);
ca->watermark[WATERMARK_MOVINGGC] = 8 +
ca->watermark[WATERMARK_METADATA];
ca->watermark[WATERMARK_NONE] = ca->free.size / 2 +
ca->watermark[WATERMARK_MOVINGGC];
#endif
return 0;
}

56
drivers/md/bcache/bcache.h
View File

@ -195,9 +195,7 @@ struct bucket {
atomic_t pin;
uint16_t prio;
uint8_t gen;
uint8_t disk_gen;
uint8_t last_gc; /* Most out of date gen in the btree */
uint8_t gc_gen;
uint16_t gc_mark; /* Bitfield used by GC. See below for field */
};
@ -207,9 +205,9 @@ struct bucket {
*/
BITMASK(GC_MARK, struct bucket, gc_mark, 0, 2);
#define GC_MARK_RECLAIMABLE 0
#define GC_MARK_DIRTY 1
#define GC_MARK_METADATA 2
#define GC_MARK_RECLAIMABLE 1
#define GC_MARK_DIRTY 2
#define GC_MARK_METADATA 3
#define GC_SECTORS_USED_SIZE 13
#define MAX_GC_SECTORS_USED (~(~0ULL << GC_SECTORS_USED_SIZE))
BITMASK(GC_SECTORS_USED, struct bucket, gc_mark, 2, GC_SECTORS_USED_SIZE);
@ -426,14 +424,9 @@ struct cache {
* their new gen to disk. After prio_write() finishes writing the new
* gens/prios, they'll be moved to the free list (and possibly discarded
* in the process)
*
* unused: GC found nothing pointing into these buckets (possibly
* because all the data they contained was overwritten), so we only
* need to discard them before they can be moved to the free list.
*/
DECLARE_FIFO(long, free)[RESERVE_NR];
DECLARE_FIFO(long, free_inc);
DECLARE_FIFO(long, unused);
size_t fifo_last_bucket;
@ -442,12 +435,6 @@ struct cache {
DECLARE_HEAP(struct bucket *, heap);
/*
* max(gen - disk_gen) for all buckets. When it gets too big we have to
* call prio_write() to keep gens from wrapping.
*/
uint8_t need_save_prio;
/*
* If nonzero, we know we aren't going to find any buckets to invalidate
* until a gc finishes - otherwise we could pointlessly burn a ton of
@ -562,19 +549,16 @@ struct cache_set {
struct list_head btree_cache_freed;
/* Number of elements in btree_cache + btree_cache_freeable lists */
unsigned bucket_cache_used;
unsigned btree_cache_used;
/*
* If we need to allocate memory for a new btree node and that
* allocation fails, we can cannibalize another node in the btree cache
* to satisfy the allocation. However, only one thread can be doing this
* at a time, for obvious reasons - try_harder and try_wait are
* basically a lock for this that we can wait on asynchronously. The
* btree_root() macro releases the lock when it returns.
* to satisfy the allocation - lock to guarantee only one thread does
* this at a time:
*/
struct task_struct *try_harder;
wait_queue_head_t try_wait;
uint64_t try_harder_start;
wait_queue_head_t btree_cache_wait;
struct task_struct *btree_cache_alloc_lock;
/*
* When we free a btree node, we increment the gen of the bucket the
@ -603,7 +587,7 @@ struct cache_set {
uint16_t min_prio;
/*
* max(gen - gc_gen) for all buckets. When it gets too big we have to gc
* max(gen - last_gc) for all buckets. When it gets too big we have to gc
* to keep gens from wrapping around.
*/
uint8_t need_gc;
@ -628,6 +612,8 @@ struct cache_set {
/* Number of moving GC bios in flight */
struct semaphore moving_in_flight;
struct workqueue_struct *moving_gc_wq;
struct btree *root;
#ifdef CONFIG_BCACHE_DEBUG
@ -667,7 +653,6 @@ struct cache_set {
struct time_stats btree_gc_time;
struct time_stats btree_split_time;
struct time_stats btree_read_time;
struct time_stats try_harder_time;
atomic_long_t cache_read_races;
atomic_long_t writeback_keys_done;
@ -850,9 +835,6 @@ static inline bool cached_dev_get(struct cached_dev *dc)
/*
* bucket_gc_gen() returns the difference between the bucket's current gen and
* the oldest gen of any pointer into that bucket in the btree (last_gc).
*
* bucket_disk_gen() returns the difference between the current gen and the gen
* on disk; they're both used to make sure gens don't wrap around.
*/
static inline uint8_t bucket_gc_gen(struct bucket *b)
@ -860,13 +842,7 @@ static inline uint8_t bucket_gc_gen(struct bucket *b)
return b->gen - b->last_gc;
}
static inline uint8_t bucket_disk_gen(struct bucket *b)
{
return b->gen - b->disk_gen;
}
#define BUCKET_GC_GEN_MAX 96U
#define BUCKET_DISK_GEN_MAX 64U
#define kobj_attribute_write(n, fn) \
static struct kobj_attribute ksysfs_##n = __ATTR(n, S_IWUSR, NULL, fn)
@ -899,11 +875,14 @@ void bch_submit_bbio(struct bio *, struct cache_set *, struct bkey *, unsigned);
uint8_t bch_inc_gen(struct cache *, struct bucket *);
void bch_rescale_priorities(struct cache_set *, int);
bool bch_bucket_add_unused(struct cache *, struct bucket *);
long bch_bucket_alloc(struct cache *, unsigned, bool);
bool bch_can_invalidate_bucket(struct cache *, struct bucket *);
void __bch_invalidate_one_bucket(struct cache *, struct bucket *);
void __bch_bucket_free(struct cache *, struct bucket *);
void bch_bucket_free(struct cache_set *, struct bkey *);
long bch_bucket_alloc(struct cache *, unsigned, bool);
int __bch_bucket_alloc_set(struct cache_set *, unsigned,
struct bkey *, int, bool);
int bch_bucket_alloc_set(struct cache_set *, unsigned,
@ -954,13 +933,10 @@ int bch_open_buckets_alloc(struct cache_set *);
void bch_open_buckets_free(struct cache_set *);
int bch_cache_allocator_start(struct cache *ca);
int bch_cache_allocator_init(struct cache *ca);
void bch_debug_exit(void);
int bch_debug_init(struct kobject *);
void bch_request_exit(void);
int bch_request_init(void);
void bch_btree_exit(void);
int bch_btree_init(void);
#endif /* _BCACHE_H */

4
drivers/md/bcache/bset.c
View File

@ -23,8 +23,8 @@ void bch_dump_bset(struct btree_keys *b, struct bset *i, unsigned set)
for (k = i->start; k < bset_bkey_last(i); k = next) {
next = bkey_next(k);
printk(KERN_ERR "block %u key %li/%u: ", set,
(uint64_t *) k - i->d, i->keys);
printk(KERN_ERR "block %u key %u/%u: ", set,
(unsigned) ((u64 *) k - i->d), i->keys);
if (b->ops->key_dump)
b->ops->key_dump(b, k);

6
drivers/md/bcache/bset.h
View File

@ -478,6 +478,12 @@ static inline void bch_keylist_init(struct keylist *l)
l->top_p = l->keys_p = l->inline_keys;
}
static inline void bch_keylist_init_single(struct keylist *l, struct bkey *k)
{
l->keys = k;
l->top = bkey_next(k);
}
static inline void bch_keylist_push(struct keylist *l)
{
l->top = bkey_next(l->top);

592
drivers/md/bcache/btree.c
View File

File diff suppressed because it is too large Load Diff

12
drivers/md/bcache/btree.h
View File

@ -127,6 +127,8 @@ struct btree {
struct cache_set *c;
struct btree *parent;
struct mutex write_lock;
unsigned long flags;
uint16_t written; /* would be nice to kill */
uint8_t level;
@ -236,11 +238,13 @@ static inline void rw_unlock(bool w, struct btree *b)
}
void bch_btree_node_read_done(struct btree *);
void __bch_btree_node_write(struct btree *, struct closure *);
void bch_btree_node_write(struct btree *, struct closure *);
void bch_btree_set_root(struct btree *);
struct btree *bch_btree_node_alloc(struct cache_set *, int, bool);
struct btree *bch_btree_node_get(struct cache_set *, struct bkey *, int, bool);
struct btree *bch_btree_node_alloc(struct cache_set *, struct btree_op *, int);
struct btree *bch_btree_node_get(struct cache_set *, struct btree_op *,
struct bkey *, int, bool);
int bch_btree_insert_check_key(struct btree *, struct btree_op *,
struct bkey *);
@ -248,10 +252,10 @@ int bch_btree_insert(struct cache_set *, struct keylist *,
atomic_t *, struct bkey *);
int bch_gc_thread_start(struct cache_set *);
size_t bch_btree_gc_finish(struct cache_set *);
void bch_initial_gc_finish(struct cache_set *);
void bch_moving_gc(struct cache_set *);
int bch_btree_check(struct cache_set *);
uint8_t __bch_btree_mark_key(struct cache_set *, int, struct bkey *);
void bch_initial_mark_key(struct cache_set *, int, struct bkey *);
static inline void wake_up_gc(struct cache_set *c)
{

36
drivers/md/bcache/extents.c
View File

@ -194,9 +194,9 @@ err:
mutex_unlock(&b->c->bucket_lock);
bch_extent_to_text(buf, sizeof(buf), k);
btree_bug(b,
"inconsistent btree pointer %s: bucket %zi pin %i prio %i gen %i last_gc %i mark %llu gc_gen %i",
"inconsistent btree pointer %s: bucket %zi pin %i prio %i gen %i last_gc %i mark %llu",
buf, PTR_BUCKET_NR(b->c, k, i), atomic_read(&g->pin),
g->prio, g->gen, g->last_gc, GC_MARK(g), g->gc_gen);
g->prio, g->gen, g->last_gc, GC_MARK(g));
return true;
}
@ -308,6 +308,16 @@ static struct bkey *bch_extent_sort_fixup(struct btree_iter *iter,
return NULL;
}
static void bch_subtract_dirty(struct bkey *k,
struct cache_set *c,
uint64_t offset,
int sectors)
{
if (KEY_DIRTY(k))
bcache_dev_sectors_dirty_add(c, KEY_INODE(k),
offset, -sectors);
}
static bool bch_extent_insert_fixup(struct btree_keys *b,
struct bkey *insert,
struct btree_iter *iter,
@ -315,13 +325,6 @@ static bool bch_extent_insert_fixup(struct btree_keys *b,
{
struct cache_set *c = container_of(b, struct btree, keys)->c;
void subtract_dirty(struct bkey *k, uint64_t offset, int sectors)
{
if (KEY_DIRTY(k))
bcache_dev_sectors_dirty_add(c, KEY_INODE(k),
offset, -sectors);
}
uint64_t old_offset;
unsigned old_size, sectors_found = 0;
@ -398,7 +401,8 @@ static bool bch_extent_insert_fixup(struct btree_keys *b,
struct bkey *top;
subtract_dirty(k, KEY_START(insert), KEY_SIZE(insert));
bch_subtract_dirty(k, c, KEY_START(insert),
KEY_SIZE(insert));
if (bkey_written(b, k)) {
/*
@ -448,7 +452,7 @@ static bool bch_extent_insert_fixup(struct btree_keys *b,
}
}
subtract_dirty(k, old_offset, old_size - KEY_SIZE(k));
bch_subtract_dirty(k, c, old_offset, old_size - KEY_SIZE(k));
}
check_failed:
@ -499,9 +503,9 @@ static bool bch_extent_bad_expensive(struct btree *b, const struct bkey *k,
if (mutex_trylock(&b->c->bucket_lock)) {
if (b->c->gc_mark_valid &&
((GC_MARK(g) != GC_MARK_DIRTY &&
KEY_DIRTY(k)) ||
GC_MARK(g) == GC_MARK_METADATA))
(!GC_MARK(g) ||
GC_MARK(g) == GC_MARK_METADATA ||
(GC_MARK(g) != GC_MARK_DIRTY && KEY_DIRTY(k))))
goto err;
if (g->prio == BTREE_PRIO)
@ -515,9 +519,9 @@ err:
mutex_unlock(&b->c->bucket_lock);
bch_extent_to_text(buf, sizeof(buf), k);
btree_bug(b,
"inconsistent extent pointer %s:\nbucket %zu pin %i prio %i gen %i last_gc %i mark %llu gc_gen %i",
"inconsistent extent pointer %s:\nbucket %zu pin %i prio %i gen %i last_gc %i mark %llu",
buf, PTR_BUCKET_NR(b->c, k, ptr), atomic_read(&g->pin),
g->prio, g->gen, g->last_gc, GC_MARK(g), g->gc_gen);
g->prio, g->gen, g->last_gc, GC_MARK(g));
return true;
}

46
drivers/md/bcache/journal.c
View File

@ -237,8 +237,14 @@ bsearch:
for (i = 0; i < ca->sb.njournal_buckets; i++)
if (ja->seq[i] > seq) {
seq = ja->seq[i];
ja->cur_idx = ja->discard_idx =
ja->last_idx = i;
/*
* When journal_reclaim() goes to allocate for
* the first time, it'll use the bucket after
* ja->cur_idx
*/
ja->cur_idx = i;
ja->last_idx = ja->discard_idx = (i + 1) %
ca->sb.njournal_buckets;
}
}
@ -288,16 +294,11 @@ void bch_journal_mark(struct cache_set *c, struct list_head *list)
k = bkey_next(k)) {
unsigned j;
for (j = 0; j < KEY_PTRS(k); j++) {
struct bucket *g = PTR_BUCKET(c, k, j);
atomic_inc(&g->pin);
for (j = 0; j < KEY_PTRS(k); j++)
if (ptr_available(c, k, j))
atomic_inc(&PTR_BUCKET(c, k, j)->pin);
if (g->prio == BTREE_PRIO &&
!ptr_stale(c, k, j))
g->prio = INITIAL_PRIO;
}
__bch_btree_mark_key(c, 0, k);
bch_initial_mark_key(c, 0, k);
}
}
}
@ -312,8 +313,6 @@ int bch_journal_replay(struct cache_set *s, struct list_head *list)
uint64_t start = i->j.last_seq, end = i->j.seq, n = start;
struct keylist keylist;
bch_keylist_init(&keylist);
list_for_each_entry(i, list, list) {
BUG_ON(i->pin && atomic_read(i->pin) != 1);
@ -326,8 +325,7 @@ int bch_journal_replay(struct cache_set *s, struct list_head *list)
k = bkey_next(k)) {
trace_bcache_journal_replay_key(k);
bkey_copy(keylist.top, k);
bch_keylist_push(&keylist);
bch_keylist_init_single(&keylist, k);
ret = bch_btree_insert(s, &keylist, i->pin, NULL);
if (ret)
@ -383,16 +381,15 @@ retry:
b = best;
if (b) {
rw_lock(true, b, b->level);
mutex_lock(&b->write_lock);
if (!btree_current_write(b)->journal) {
rw_unlock(true, b);
mutex_unlock(&b->write_lock);
/* We raced */
goto retry;
}
bch_btree_node_write(b, NULL);
rw_unlock(true, b);
__bch_btree_node_write(b, NULL);
mutex_unlock(&b->write_lock);
}
}
@ -536,6 +533,7 @@ void bch_journal_next(struct journal *j)
atomic_set(&fifo_back(&j->pin), 1);
j->cur->data->seq = ++j->seq;
j->cur->dirty = false;
j->cur->need_write = false;
j->cur->data->keys = 0;
@ -731,7 +729,10 @@ static void journal_write_work(struct work_struct *work)
struct cache_set,
journal.work);
spin_lock(&c->journal.lock);
journal_try_write(c);
if (c->journal.cur->dirty)
journal_try_write(c);
else
spin_unlock(&c->journal.lock);
}
/*
@ -761,7 +762,8 @@ atomic_t *bch_journal(struct cache_set *c,
if (parent) {
closure_wait(&w->wait, parent);
journal_try_write(c);
} else if (!w->need_write) {
} else if (!w->dirty) {
w->dirty = true;
schedule_delayed_work(&c->journal.work,
msecs_to_jiffies(c->journal_delay_ms));
spin_unlock(&c->journal.lock);

1
drivers/md/bcache/journal.h
View File

@ -95,6 +95,7 @@ struct journal_write {
struct cache_set *c;
struct closure_waitlist wait;
bool dirty;
bool need_write;
};

18
drivers/md/bcache/movinggc.c
View File

@ -24,12 +24,10 @@ static bool moving_pred(struct keybuf *buf, struct bkey *k)
moving_gc_keys);
unsigned i;
for (i = 0; i < KEY_PTRS(k); i++) {
struct bucket *g = PTR_BUCKET(c, k, i);
if (GC_MOVE(g))
for (i = 0; i < KEY_PTRS(k); i++)
if (ptr_available(c, k, i) &&
GC_MOVE(PTR_BUCKET(c, k, i)))
return true;
}
return false;
}
@ -115,7 +113,7 @@ static void write_moving(struct closure *cl)
closure_call(&op->cl, bch_data_insert, NULL, cl);
}
continue_at(cl, write_moving_finish, system_wq);
continue_at(cl, write_moving_finish, op->wq);
}
static void read_moving_submit(struct closure *cl)
@ -125,7 +123,7 @@ static void read_moving_submit(struct closure *cl)
bch_submit_bbio(bio, io->op.c, &io->w->key, 0);
continue_at(cl, write_moving, system_wq);
continue_at(cl, write_moving, io->op.wq);
}
static void read_moving(struct cache_set *c)
@ -160,6 +158,7 @@ static void read_moving(struct cache_set *c)
io->w = w;
io->op.inode = KEY_INODE(&w->key);
io->op.c = c;
io->op.wq = c->moving_gc_wq;
moving_init(io);
bio = &io->bio.bio;
@ -216,7 +215,10 @@ void bch_moving_gc(struct cache_set *c)
ca->heap.used = 0;
for_each_bucket(b, ca) {
if (!GC_SECTORS_USED(b))
if (GC_MARK(b) == GC_MARK_METADATA ||
!GC_SECTORS_USED(b) ||
GC_SECTORS_USED(b) == ca->sb.bucket_size ||
atomic_read(&b->pin))
continue;
if (!heap_full(&ca->heap)) {

201
drivers/md/bcache/request.c
View File

@ -12,11 +12,9 @@
#include "request.h"
#include "writeback.h"
#include <linux/cgroup.h>
#include <linux/module.h>
#include <linux/hash.h>
#include <linux/random.h>
#include "blk-cgroup.h"
#include <trace/events/bcache.h>
@ -27,171 +25,13 @@ struct kmem_cache *bch_search_cache;
static void bch_data_insert_start(struct closure *);
/* Cgroup interface */
#ifdef CONFIG_CGROUP_BCACHE
static struct bch_cgroup bcache_default_cgroup = { .cache_mode = -1 };
static struct bch_cgroup *cgroup_to_bcache(struct cgroup *cgroup)
{
struct cgroup_subsys_state *css;
return cgroup &&
(css = cgroup_subsys_state(cgroup, bcache_subsys_id))
? container_of(css, struct bch_cgroup, css)
: &bcache_default_cgroup;
}
struct bch_cgroup *bch_bio_to_cgroup(struct bio *bio)
{
struct cgroup_subsys_state *css = bio->bi_css
? cgroup_subsys_state(bio->bi_css->cgroup, bcache_subsys_id)
: task_subsys_state(current, bcache_subsys_id);
return css
? container_of(css, struct bch_cgroup, css)
: &bcache_default_cgroup;
}
static ssize_t cache_mode_read(struct cgroup *cgrp, struct cftype *cft,
struct file *file,
char __user *buf, size_t nbytes, loff_t *ppos)
{
char tmp[1024];
int len = bch_snprint_string_list(tmp, PAGE_SIZE, bch_cache_modes,
cgroup_to_bcache(cgrp)->cache_mode + 1);
if (len < 0)
return len;
return simple_read_from_buffer(buf, nbytes, ppos, tmp, len);
}
static int cache_mode_write(struct cgroup *cgrp, struct cftype *cft,
const char *buf)
{
int v = bch_read_string_list(buf, bch_cache_modes);
if (v < 0)
return v;
cgroup_to_bcache(cgrp)->cache_mode = v - 1;
return 0;
}
static u64 bch_verify_read(struct cgroup *cgrp, struct cftype *cft)
{
return cgroup_to_bcache(cgrp)->verify;
}
static int bch_verify_write(struct cgroup *cgrp, struct cftype *cft, u64 val)
{
cgroup_to_bcache(cgrp)->verify = val;
return 0;
}
static u64 bch_cache_hits_read(struct cgroup *cgrp, struct cftype *cft)
{
struct bch_cgroup *bcachecg = cgroup_to_bcache(cgrp);
return atomic_read(&bcachecg->stats.cache_hits);
}
static u64 bch_cache_misses_read(struct cgroup *cgrp, struct cftype *cft)
{
struct bch_cgroup *bcachecg = cgroup_to_bcache(cgrp);
return atomic_read(&bcachecg->stats.cache_misses);
}
static u64 bch_cache_bypass_hits_read(struct cgroup *cgrp,
struct cftype *cft)
{
struct bch_cgroup *bcachecg = cgroup_to_bcache(cgrp);
return atomic_read(&bcachecg->stats.cache_bypass_hits);
}
static u64 bch_cache_bypass_misses_read(struct cgroup *cgrp,
struct cftype *cft)
{
struct bch_cgroup *bcachecg = cgroup_to_bcache(cgrp);
return atomic_read(&bcachecg->stats.cache_bypass_misses);
}
static struct cftype bch_files[] = {
{
.name = "cache_mode",
.read = cache_mode_read,
.write_string = cache_mode_write,
},
{
.name = "verify",
.read_u64 = bch_verify_read,
.write_u64 = bch_verify_write,
},
{
.name = "cache_hits",
.read_u64 = bch_cache_hits_read,
},
{
.name = "cache_misses",
.read_u64 = bch_cache_misses_read,
},
{
.name = "cache_bypass_hits",
.read_u64 = bch_cache_bypass_hits_read,
},
{
.name = "cache_bypass_misses",
.read_u64 = bch_cache_bypass_misses_read,
},
{ } /* terminate */
};
static void init_bch_cgroup(struct bch_cgroup *cg)
{
cg->cache_mode = -1;
}
static struct cgroup_subsys_state *bcachecg_create(struct cgroup *cgroup)
{
struct bch_cgroup *cg;
cg = kzalloc(sizeof(*cg), GFP_KERNEL);
if (!cg)
return ERR_PTR(-ENOMEM);
init_bch_cgroup(cg);
return &cg->css;
}
static void bcachecg_destroy(struct cgroup *cgroup)
{
struct bch_cgroup *cg = cgroup_to_bcache(cgroup);
kfree(cg);
}
struct cgroup_subsys bcache_subsys = {
.create = bcachecg_create,
.destroy = bcachecg_destroy,
.subsys_id = bcache_subsys_id,
.name = "bcache",
.module = THIS_MODULE,
};
EXPORT_SYMBOL_GPL(bcache_subsys);
#endif
static unsigned cache_mode(struct cached_dev *dc, struct bio *bio)
{
#ifdef CONFIG_CGROUP_BCACHE
int r = bch_bio_to_cgroup(bio)->cache_mode;
if (r >= 0)
return r;
#endif
return BDEV_CACHE_MODE(&dc->sb);
}
static bool verify(struct cached_dev *dc, struct bio *bio)
{
#ifdef CONFIG_CGROUP_BCACHE
if (bch_bio_to_cgroup(bio)->verify)
return true;
#endif
return dc->verify;
}
@ -248,7 +88,7 @@ static void bch_data_insert_keys(struct closure *cl)
atomic_dec_bug(journal_ref);
if (!op->insert_data_done)
continue_at(cl, bch_data_insert_start, bcache_wq);
continue_at(cl, bch_data_insert_start, op->wq);
bch_keylist_free(&op->insert_keys);
closure_return(cl);
@ -297,7 +137,7 @@ static void bch_data_invalidate(struct closure *cl)
op->insert_data_done = true;
bio_put(bio);
out:
continue_at(cl, bch_data_insert_keys, bcache_wq);
continue_at(cl, bch_data_insert_keys, op->wq);
}
static void bch_data_insert_error(struct closure *cl)
@ -340,7 +180,7 @@ static void bch_data_insert_endio(struct bio *bio, int error)
if (op->writeback)
op->error = error;
else if (!op->replace)
set_closure_fn(cl, bch_data_insert_error, bcache_wq);
set_closure_fn(cl, bch_data_insert_error, op->wq);
else
set_closure_fn(cl, NULL, NULL);
}
@ -376,7 +216,7 @@ static void bch_data_insert_start(struct closure *cl)
if (bch_keylist_realloc(&op->insert_keys,
3 + (op->csum ? 1 : 0),
op->c))
continue_at(cl, bch_data_insert_keys, bcache_wq);
continue_at(cl, bch_data_insert_keys, op->wq);
k = op->insert_keys.top;
bkey_init(k);
@ -413,7 +253,7 @@ static void bch_data_insert_start(struct closure *cl)
} while (n != bio);
op->insert_data_done = true;
continue_at(cl, bch_data_insert_keys, bcache_wq);
continue_at(cl, bch_data_insert_keys, op->wq);
err:
/* bch_alloc_sectors() blocks if s->writeback = true */
BUG_ON(op->writeback);
@ -442,7 +282,7 @@ err:
bio_put(bio);
if (!bch_keylist_empty(&op->insert_keys))
continue_at(cl, bch_data_insert_keys, bcache_wq);
continue_at(cl, bch_data_insert_keys, op->wq);
else
closure_return(cl);
}
@ -824,6 +664,7 @@ static inline struct search *search_alloc(struct bio *bio,
s->iop.error = 0;
s->iop.flags = 0;
s->iop.flush_journal = (bio->bi_rw & (REQ_FLUSH|REQ_FUA)) != 0;
s->iop.wq = bcache_wq;
return s;
}
@ -1203,22 +1044,13 @@ void bch_cached_dev_request_init(struct cached_dev *dc)
static int flash_dev_cache_miss(struct btree *b, struct search *s,
struct bio *bio, unsigned sectors)
{
struct bio_vec bv;
struct bvec_iter iter;
unsigned bytes = min(sectors, bio_sectors(bio)) << 9;
/* Zero fill bio */
swap(bio->bi_iter.bi_size, bytes);
zero_fill_bio(bio);
swap(bio->bi_iter.bi_size, bytes);
bio_for_each_segment(bv, bio, iter) {
unsigned j = min(bv.bv_len >> 9, sectors);
void *p = kmap(bv.bv_page);
memset(p + bv.bv_offset, 0, j << 9);
kunmap(bv.bv_page);
sectors -= j;
}
bio_advance(bio, min(sectors << 9, bio->bi_iter.bi_size));
bio_advance(bio, bytes);
if (!bio->bi_iter.bi_size)
return MAP_DONE;
@ -1313,9 +1145,6 @@ void bch_flash_dev_request_init(struct bcache_device *d)
void bch_request_exit(void)
{
#ifdef CONFIG_CGROUP_BCACHE
cgroup_unload_subsys(&bcache_subsys);
#endif
if (bch_search_cache)
kmem_cache_destroy(bch_search_cache);
}
@ -1326,11 +1155,5 @@ int __init bch_request_init(void)
if (!bch_search_cache)
return -ENOMEM;
#ifdef CONFIG_CGROUP_BCACHE
cgroup_load_subsys(&bcache_subsys);
init_bch_cgroup(&bcache_default_cgroup);
cgroup_add_cftypes(&bcache_subsys, bch_files);
#endif
return 0;
}

19
drivers/md/bcache/request.h
View File

@ -1,12 +1,11 @@
#ifndef _BCACHE_REQUEST_H_
#define _BCACHE_REQUEST_H_
#include <linux/cgroup.h>
struct data_insert_op {
struct closure cl;
struct cache_set *c;
struct bio *bio;
struct workqueue_struct *wq;
unsigned inode;
uint16_t write_point;
@ -41,20 +40,4 @@ void bch_flash_dev_request_init(struct bcache_device *d);
extern struct kmem_cache *bch_search_cache, *bch_passthrough_cache;
struct bch_cgroup {
#ifdef CONFIG_CGROUP_BCACHE
struct cgroup_subsys_state css;
#endif
/*
* We subtract one from the index into bch_cache_modes[], so that
* default == -1; this makes it so the rest match up with d->cache_mode,
* and we use d->cache_mode if cgrp->cache_mode < 0
*/
short cache_mode;
bool verify;
struct cache_stat_collector stats;
};
struct bch_cgroup *bch_bio_to_cgroup(struct bio *bio);
#endif /* _BCACHE_REQUEST_H_ */

3
drivers/md/bcache/stats.c
View File

@ -201,9 +201,6 @@ void bch_mark_cache_accounting(struct cache_set *c, struct bcache_device *d,
struct cached_dev *dc = container_of(d, struct cached_dev, disk);
mark_cache_stats(&dc->accounting.collector, hit, bypass);
mark_cache_stats(&c->accounting.collector, hit, bypass);
#ifdef CONFIG_CGROUP_BCACHE
mark_cache_stats(&(bch_bio_to_cgroup(s->orig_bio)->stats), hit, bypass);
#endif
}
void bch_mark_cache_readahead(struct cache_set *c, struct bcache_device *d)

64
drivers/md/bcache/super.c
View File

@ -541,9 +541,6 @@ static void prio_io(struct cache *ca, uint64_t bucket, unsigned long rw)
closure_sync(cl);
}
#define buckets_free(c) "free %zu, free_inc %zu, unused %zu", \
fifo_used(&c->free), fifo_used(&c->free_inc), fifo_used(&c->unused)
void bch_prio_write(struct cache *ca)
{
int i;
@ -554,10 +551,6 @@ void bch_prio_write(struct cache *ca)
lockdep_assert_held(&ca->set->bucket_lock);
for (b = ca->buckets;
b < ca->buckets + ca->sb.nbuckets; b++)
b->disk_gen = b->gen;
ca->disk_buckets->seq++;
atomic_long_add(ca->sb.bucket_size * prio_buckets(ca),
@ -601,14 +594,17 @@ void bch_prio_write(struct cache *ca)
mutex_lock(&ca->set->bucket_lock);
ca->need_save_prio = 0;
/*
* Don't want the old priorities to get garbage collected until after we
* finish writing the new ones, and they're journalled
*/
for (i = 0; i < prio_buckets(ca); i++)
for (i = 0; i < prio_buckets(ca); i++) {
if (ca->prio_last_buckets[i])
__bch_bucket_free(ca,
&ca->buckets[ca->prio_last_buckets[i]]);
ca->prio_last_buckets[i] = ca->prio_buckets[i];
}
}
static void prio_read(struct cache *ca, uint64_t bucket)
@ -639,7 +635,7 @@ static void prio_read(struct cache *ca, uint64_t bucket)
}
b->prio = le16_to_cpu(d->prio);
b->gen = b->disk_gen = b->last_gc = b->gc_gen = d->gen;
b->gen = b->last_gc = d->gen;
}
}
@ -843,6 +839,7 @@ static int bcache_device_init(struct bcache_device *d, unsigned block_size,
q->limits.max_segment_size = UINT_MAX;
q->limits.max_segments = BIO_MAX_PAGES;
q->limits.max_discard_sectors = UINT_MAX;
q->limits.discard_granularity = 512;
q->limits.io_min = block_size;
q->limits.logical_block_size = block_size;
q->limits.physical_block_size = block_size;
@ -1355,6 +1352,8 @@ static void cache_set_free(struct closure *cl)
bch_bset_sort_state_free(&c->sort);
free_pages((unsigned long) c->uuids, ilog2(bucket_pages(c)));
if (c->moving_gc_wq)
destroy_workqueue(c->moving_gc_wq);
if (c->bio_split)
bioset_free(c->bio_split);
if (c->fill_iter)
@ -1395,14 +1394,21 @@ static void cache_set_flush(struct closure *cl)
list_add(&c->root->list, &c->btree_cache);
/* Should skip this if we're unregistering because of an error */
list_for_each_entry(b, &c->btree_cache, list)
list_for_each_entry(b, &c->btree_cache, list) {
mutex_lock(&b->write_lock);
if (btree_node_dirty(b))
bch_btree_node_write(b, NULL);
__bch_btree_node_write(b, NULL);
mutex_unlock(&b->write_lock);
}
for_each_cache(ca, c, i)
if (ca->alloc_thread)
kthread_stop(ca->alloc_thread);
cancel_delayed_work_sync(&c->journal.work);
/* flush last journal entry if needed */
c->journal.work.work.func(&c->journal.work.work);
closure_return(cl);
}
@ -1485,14 +1491,13 @@ struct cache_set *bch_cache_set_alloc(struct cache_sb *sb)
sema_init(&c->sb_write_mutex, 1);
mutex_init(&c->bucket_lock);
init_waitqueue_head(&c->try_wait);
init_waitqueue_head(&c->btree_cache_wait);
init_waitqueue_head(&c->bucket_wait);
sema_init(&c->uuid_write_mutex, 1);
spin_lock_init(&c->btree_gc_time.lock);
spin_lock_init(&c->btree_split_time.lock);
spin_lock_init(&c->btree_read_time.lock);
spin_lock_init(&c->try_harder_time.lock);
bch_moving_init_cache_set(c);
@ -1517,6 +1522,7 @@ struct cache_set *bch_cache_set_alloc(struct cache_sb *sb)
!(c->fill_iter = mempool_create_kmalloc_pool(1, iter_size)) ||
!(c->bio_split = bioset_create(4, offsetof(struct bbio, bio))) ||
!(c->uuids = alloc_bucket_pages(GFP_KERNEL, c)) ||
!(c->moving_gc_wq = create_workqueue("bcache_gc")) ||
bch_journal_alloc(c) ||
bch_btree_cache_alloc(c) ||
bch_open_buckets_alloc(c) ||
@ -1580,7 +1586,7 @@ static void run_cache_set(struct cache_set *c)
goto err;
err = "error reading btree root";
c->root = bch_btree_node_get(c, k, j->btree_level, true);
c->root = bch_btree_node_get(c, NULL, k, j->btree_level, true);
if (IS_ERR_OR_NULL(c->root))
goto err;
@ -1596,7 +1602,7 @@ static void run_cache_set(struct cache_set *c)
goto err;
bch_journal_mark(c, &journal);
bch_btree_gc_finish(c);
bch_initial_gc_finish(c);
pr_debug("btree_check() done");
/*
@ -1638,7 +1644,7 @@ static void run_cache_set(struct cache_set *c)
ca->sb.d[j] = ca->sb.first_bucket + j;
}
bch_btree_gc_finish(c);
bch_initial_gc_finish(c);
err = "error starting allocator thread";
for_each_cache(ca, c, i)
@ -1655,12 +1661,14 @@ static void run_cache_set(struct cache_set *c)
goto err;
err = "cannot allocate new btree root";
c->root = bch_btree_node_alloc(c, 0, true);
c->root = bch_btree_node_alloc(c, NULL, 0);
if (IS_ERR_OR_NULL(c->root))
goto err;
mutex_lock(&c->root->write_lock);
bkey_copy_key(&c->root->key, &MAX_KEY);
bch_btree_node_write(c->root, &cl);
mutex_unlock(&c->root->write_lock);
bch_btree_set_root(c->root);
rw_unlock(true, c->root);
@ -1782,7 +1790,6 @@ void bch_cache_release(struct kobject *kobj)
vfree(ca->buckets);
free_heap(&ca->heap);
free_fifo(&ca->unused);
free_fifo(&ca->free_inc);
for (i = 0; i < RESERVE_NR; i++)
@ -1819,7 +1826,6 @@ static int cache_alloc(struct cache_sb *sb, struct cache *ca)
!init_fifo(&ca->free[RESERVE_MOVINGGC], free, GFP_KERNEL) ||
!init_fifo(&ca->free[RESERVE_NONE], free, GFP_KERNEL) ||
!init_fifo(&ca->free_inc, free << 2, GFP_KERNEL) ||
!init_fifo(&ca->unused, free << 2, GFP_KERNEL) ||
!init_heap(&ca->heap, free << 3, GFP_KERNEL) ||
!(ca->buckets = vzalloc(sizeof(struct bucket) *
ca->sb.nbuckets)) ||
@ -1834,13 +1840,7 @@ static int cache_alloc(struct cache_sb *sb, struct cache *ca)
for_each_bucket(b, ca)
atomic_set(&b->pin, 0);
if (bch_cache_allocator_init(ca))
goto err;
return 0;
err:
kobject_put(&ca->kobj);
return -ENOMEM;
}
static void register_cache(struct cache_sb *sb, struct page *sb_page,
@ -1869,7 +1869,10 @@ static void register_cache(struct cache_sb *sb, struct page *sb_page,
if (kobject_add(&ca->kobj, &part_to_dev(bdev->bd_part)->kobj, "bcache"))
goto err;
mutex_lock(&bch_register_lock);
err = register_cache_set(ca);
mutex_unlock(&bch_register_lock);
if (err)
goto err;
@ -1931,8 +1934,6 @@ static ssize_t register_bcache(struct kobject *k, struct kobj_attribute *attr,
if (!try_module_get(THIS_MODULE))
return -EBUSY;
mutex_lock(&bch_register_lock);
if (!(path = kstrndup(buffer, size, GFP_KERNEL)) ||
!(sb = kmalloc(sizeof(struct cache_sb), GFP_KERNEL)))
goto err;
@ -1965,7 +1966,9 @@ static ssize_t register_bcache(struct kobject *k, struct kobj_attribute *attr,
if (!dc)
goto err_close;
mutex_lock(&bch_register_lock);
register_bdev(sb, sb_page, bdev, dc);
mutex_unlock(&bch_register_lock);
} else {
struct cache *ca = kzalloc(sizeof(*ca), GFP_KERNEL);
if (!ca)
@ -1978,7 +1981,6 @@ out:
put_page(sb_page);
kfree(sb);
kfree(path);
mutex_unlock(&bch_register_lock);
module_put(THIS_MODULE);
return ret;
@ -2057,7 +2059,6 @@ static void bcache_exit(void)
{
bch_debug_exit();
bch_request_exit();
bch_btree_exit();
if (bcache_kobj)
kobject_put(bcache_kobj);
if (bcache_wq)
@ -2087,7 +2088,6 @@ static int __init bcache_init(void)
if (!(bcache_wq = create_workqueue("bcache")) ||
!(bcache_kobj = kobject_create_and_add("bcache", fs_kobj)) ||
sysfs_create_files(bcache_kobj, files) ||
bch_btree_init() ||
bch_request_init() ||
bch_debug_init(bcache_kobj))
goto err;

189
drivers/md/bcache/sysfs.c
View File

@ -54,7 +54,6 @@ sysfs_time_stats_attribute(btree_gc, sec, ms);
sysfs_time_stats_attribute(btree_split, sec, us);
sysfs_time_stats_attribute(btree_sort, ms, us);
sysfs_time_stats_attribute(btree_read, ms, us);
sysfs_time_stats_attribute(try_harder, ms, us);
read_attribute(btree_nodes);
read_attribute(btree_used_percent);
@ -406,7 +405,7 @@ struct bset_stats_op {
struct bset_stats stats;
};
static int btree_bset_stats(struct btree_op *b_op, struct btree *b)
static int bch_btree_bset_stats(struct btree_op *b_op, struct btree *b)
{
struct bset_stats_op *op = container_of(b_op, struct bset_stats_op, op);
@ -424,7 +423,7 @@ static int bch_bset_print_stats(struct cache_set *c, char *buf)
memset(&op, 0, sizeof(op));
bch_btree_op_init(&op.op, -1);
ret = bch_btree_map_nodes(&op.op, c, &ZERO_KEY, btree_bset_stats);
ret = bch_btree_map_nodes(&op.op, c, &ZERO_KEY, bch_btree_bset_stats);
if (ret < 0)
return ret;
@ -442,81 +441,81 @@ static int bch_bset_print_stats(struct cache_set *c, char *buf)
op.stats.floats, op.stats.failed);
}
static unsigned bch_root_usage(struct cache_set *c)
{
unsigned bytes = 0;
struct bkey *k;
struct btree *b;
struct btree_iter iter;
goto lock_root;
do {
rw_unlock(false, b);
lock_root:
b = c->root;
rw_lock(false, b, b->level);
} while (b != c->root);
for_each_key_filter(&b->keys, k, &iter, bch_ptr_bad)
bytes += bkey_bytes(k);
rw_unlock(false, b);
return (bytes * 100) / btree_bytes(c);
}
static size_t bch_cache_size(struct cache_set *c)
{
size_t ret = 0;
struct btree *b;
mutex_lock(&c->bucket_lock);
list_for_each_entry(b, &c->btree_cache, list)
ret += 1 << (b->keys.page_order + PAGE_SHIFT);
mutex_unlock(&c->bucket_lock);
return ret;
}
static unsigned bch_cache_max_chain(struct cache_set *c)
{
unsigned ret = 0;
struct hlist_head *h;
mutex_lock(&c->bucket_lock);
for (h = c->bucket_hash;
h < c->bucket_hash + (1 << BUCKET_HASH_BITS);
h++) {
unsigned i = 0;
struct hlist_node *p;
hlist_for_each(p, h)
i++;
ret = max(ret, i);
}
mutex_unlock(&c->bucket_lock);
return ret;
}
static unsigned bch_btree_used(struct cache_set *c)
{
return div64_u64(c->gc_stats.key_bytes * 100,
(c->gc_stats.nodes ?: 1) * btree_bytes(c));
}
static unsigned bch_average_key_size(struct cache_set *c)
{
return c->gc_stats.nkeys
? div64_u64(c->gc_stats.data, c->gc_stats.nkeys)
: 0;
}
SHOW(__bch_cache_set)
{
unsigned root_usage(struct cache_set *c)
{
unsigned bytes = 0;
struct bkey *k;
struct btree *b;
struct btree_iter iter;
goto lock_root;
do {
rw_unlock(false, b);
lock_root:
b = c->root;
rw_lock(false, b, b->level);
} while (b != c->root);
for_each_key_filter(&b->keys, k, &iter, bch_ptr_bad)
bytes += bkey_bytes(k);
rw_unlock(false, b);
return (bytes * 100) / btree_bytes(c);
}
size_t cache_size(struct cache_set *c)
{
size_t ret = 0;
struct btree *b;
mutex_lock(&c->bucket_lock);
list_for_each_entry(b, &c->btree_cache, list)
ret += 1 << (b->keys.page_order + PAGE_SHIFT);
mutex_unlock(&c->bucket_lock);
return ret;
}
unsigned cache_max_chain(struct cache_set *c)
{
unsigned ret = 0;
struct hlist_head *h;
mutex_lock(&c->bucket_lock);
for (h = c->bucket_hash;
h < c->bucket_hash + (1 << BUCKET_HASH_BITS);
h++) {
unsigned i = 0;
struct hlist_node *p;
hlist_for_each(p, h)
i++;
ret = max(ret, i);
}
mutex_unlock(&c->bucket_lock);
return ret;
}
unsigned btree_used(struct cache_set *c)
{
return div64_u64(c->gc_stats.key_bytes * 100,
(c->gc_stats.nodes ?: 1) * btree_bytes(c));
}
unsigned average_key_size(struct cache_set *c)
{
return c->gc_stats.nkeys
? div64_u64(c->gc_stats.data, c->gc_stats.nkeys)
: 0;
}
struct cache_set *c = container_of(kobj, struct cache_set, kobj);
sysfs_print(synchronous, CACHE_SYNC(&c->sb));
@ -524,21 +523,20 @@ lock_root:
sysfs_hprint(bucket_size, bucket_bytes(c));
sysfs_hprint(block_size, block_bytes(c));
sysfs_print(tree_depth, c->root->level);
sysfs_print(root_usage_percent, root_usage(c));
sysfs_print(root_usage_percent, bch_root_usage(c));
sysfs_hprint(btree_cache_size, cache_size(c));
sysfs_print(btree_cache_max_chain, cache_max_chain(c));
sysfs_hprint(btree_cache_size, bch_cache_size(c));
sysfs_print(btree_cache_max_chain, bch_cache_max_chain(c));
sysfs_print(cache_available_percent, 100 - c->gc_stats.in_use);
sysfs_print_time_stats(&c->btree_gc_time, btree_gc, sec, ms);
sysfs_print_time_stats(&c->btree_split_time, btree_split, sec, us);
sysfs_print_time_stats(&c->sort.time, btree_sort, ms, us);
sysfs_print_time_stats(&c->btree_read_time, btree_read, ms, us);
sysfs_print_time_stats(&c->try_harder_time, try_harder, ms, us);
sysfs_print(btree_used_percent, btree_used(c));
sysfs_print(btree_used_percent, bch_btree_used(c));
sysfs_print(btree_nodes, c->gc_stats.nodes);
sysfs_hprint(average_key_size, average_key_size(c));
sysfs_hprint(average_key_size, bch_average_key_size(c));
sysfs_print(cache_read_races,
atomic_long_read(&c->cache_read_races));
@ -709,7 +707,6 @@ static struct attribute *bch_cache_set_internal_files[] = {
sysfs_time_stats_attribute_list(btree_split, sec, us)
sysfs_time_stats_attribute_list(btree_sort, ms, us)
sysfs_time_stats_attribute_list(btree_read, ms, us)
sysfs_time_stats_attribute_list(try_harder, ms, us)
&sysfs_btree_nodes,
&sysfs_btree_used_percent,
@ -761,7 +758,9 @@ SHOW(__bch_cache)
int cmp(const void *l, const void *r)
{ return *((uint16_t *) r) - *((uint16_t *) l); }
size_t n = ca->sb.nbuckets, i, unused, btree;
struct bucket *b;
size_t n = ca->sb.nbuckets, i;
size_t unused = 0, available = 0, dirty = 0, meta = 0;
uint64_t sum = 0;
/* Compute 31 quantiles */
uint16_t q[31], *p, *cached;
@ -772,6 +771,17 @@ SHOW(__bch_cache)
return -ENOMEM;
mutex_lock(&ca->set->bucket_lock);
for_each_bucket(b, ca) {
if (!GC_SECTORS_USED(b))
unused++;
if (GC_MARK(b) == GC_MARK_RECLAIMABLE)
available++;
if (GC_MARK(b) == GC_MARK_DIRTY)
dirty++;
if (GC_MARK(b) == GC_MARK_METADATA)
meta++;
}
for (i = ca->sb.first_bucket; i < n; i++)
p[i] = ca->buckets[i].prio;
mutex_unlock(&ca->set->bucket_lock);
@ -786,10 +796,7 @@ SHOW(__bch_cache)
while (cached < p + n &&
*cached == BTREE_PRIO)
cached++;
btree = cached - p;
n -= btree;
cached++, n--;
for (i = 0; i < n; i++)
sum += INITIAL_PRIO - cached[i];
@ -805,12 +812,16 @@ SHOW(__bch_cache)
ret = scnprintf(buf, PAGE_SIZE,
"Unused: %zu%%\n"
"Clean: %zu%%\n"
"Dirty: %zu%%\n"
"Metadata: %zu%%\n"
"Average: %llu\n"
"Sectors per Q: %zu\n"
"Quantiles: [",
unused * 100 / (size_t) ca->sb.nbuckets,
btree * 100 / (size_t) ca->sb.nbuckets, sum,
available * 100 / (size_t) ca->sb.nbuckets,
dirty * 100 / (size_t) ca->sb.nbuckets,
meta * 100 / (size_t) ca->sb.nbuckets, sum,
n * ca->sb.bucket_size / (ARRAY_SIZE(q) + 1));
for (i = 0; i < ARRAY_SIZE(q); i++)

2
drivers/md/bcache/trace.c
View File

@ -45,7 +45,7 @@ EXPORT_TRACEPOINT_SYMBOL_GPL(bcache_btree_node_split);
EXPORT_TRACEPOINT_SYMBOL_GPL(bcache_btree_node_compact);
EXPORT_TRACEPOINT_SYMBOL_GPL(bcache_btree_set_root);
EXPORT_TRACEPOINT_SYMBOL_GPL(bcache_alloc_invalidate);
EXPORT_TRACEPOINT_SYMBOL_GPL(bcache_invalidate);
EXPORT_TRACEPOINT_SYMBOL_GPL(bcache_alloc_fail);
EXPORT_TRACEPOINT_SYMBOL_GPL(bcache_writeback);

8
include/linux/drbd.h
View File

@ -327,12 +327,6 @@ enum drbd_state_rv {
SS_AFTER_LAST_ERROR = -22, /* Keep this at bottom */
};
/* from drbd_strings.c */
extern const char *drbd_conn_str(enum drbd_conns);
extern const char *drbd_role_str(enum drbd_role);
extern const char *drbd_disk_str(enum drbd_disk_state);
extern const char *drbd_set_st_err_str(enum drbd_state_rv);
#define SHARED_SECRET_MAX 64
#define MDF_CONSISTENT (1 << 0)
@ -382,4 +376,6 @@ enum drbd_timeout_flag {
#define DRBD_MD_INDEX_FLEX_EXT -2
#define DRBD_MD_INDEX_FLEX_INT -3
#define DRBD_CPU_MASK_SIZE 32
#endif

6
include/linux/drbd_genl.h
View File

@ -135,7 +135,7 @@ GENL_struct(DRBD_NLA_DISK_CONF, 3, disk_conf,
)
GENL_struct(DRBD_NLA_RESOURCE_OPTS, 4, res_opts,
__str_field_def(1, DRBD_GENLA_F_MANDATORY, cpu_mask, 32)
__str_field_def(1, DRBD_GENLA_F_MANDATORY, cpu_mask, DRBD_CPU_MASK_SIZE)
__u32_field_def(2, DRBD_GENLA_F_MANDATORY, on_no_data, DRBD_ON_NO_DATA_DEF)
)
@ -276,9 +276,9 @@ GENL_op(
)
/* add DRBD minor devices as volumes to resources */
GENL_op(DRBD_ADM_NEW_MINOR, 5, GENL_doit(drbd_adm_add_minor),
GENL_op(DRBD_ADM_NEW_MINOR, 5, GENL_doit(drbd_adm_new_minor),
GENL_tla_expected(DRBD_NLA_CFG_CONTEXT, DRBD_F_REQUIRED))
GENL_op(DRBD_ADM_DEL_MINOR, 6, GENL_doit(drbd_adm_delete_minor),
GENL_op(DRBD_ADM_DEL_MINOR, 6, GENL_doit(drbd_adm_del_minor),
GENL_tla_expected(DRBD_NLA_CFG_CONTEXT, DRBD_F_REQUIRED))
/* add or delete resources */

1
include/linux/idr.h
View File

@ -85,6 +85,7 @@ void idr_remove(struct idr *idp, int id);
void idr_free(struct idr *idp, int id);
void idr_destroy(struct idr *idp);
void idr_init(struct idr *idp);
bool idr_is_empty(struct idr *idp);
/**
* idr_preload_end - end preload section started with idr_preload()

52
include/trace/events/bcache.h
View File

@ -399,26 +399,43 @@ TRACE_EVENT(bcache_keyscan,
/* Allocator */
TRACE_EVENT(bcache_alloc_invalidate,
TP_PROTO(struct cache *ca),
TP_ARGS(ca),
TRACE_EVENT(bcache_invalidate,
TP_PROTO(struct cache *ca, size_t bucket),
TP_ARGS(ca, bucket),
TP_STRUCT__entry(
__field(unsigned, free )
__field(unsigned, free_inc )
__field(unsigned, free_inc_size )
__field(unsigned, unused )
__field(unsigned, sectors )
__field(dev_t, dev )
__field(__u64, offset )
),
TP_fast_assign(
__entry->free = fifo_used(&ca->free[RESERVE_NONE]);
__entry->free_inc = fifo_used(&ca->free_inc);
__entry->free_inc_size = ca->free_inc.size;
__entry->unused = fifo_used(&ca->unused);
__entry->dev = ca->bdev->bd_dev;
__entry->offset = bucket << ca->set->bucket_bits;
__entry->sectors = GC_SECTORS_USED(&ca->buckets[bucket]);
),
TP_printk("free %u free_inc %u/%u unused %u", __entry->free,
__entry->free_inc, __entry->free_inc_size, __entry->unused)
TP_printk("invalidated %u sectors at %d,%d sector=%llu",
__entry->sectors, MAJOR(__entry->dev),
MINOR(__entry->dev), __entry->offset)
);
TRACE_EVENT(bcache_alloc,
TP_PROTO(struct cache *ca, size_t bucket),
TP_ARGS(ca, bucket),
TP_STRUCT__entry(
__field(dev_t, dev )
__field(__u64, offset )
),
TP_fast_assign(
__entry->dev = ca->bdev->bd_dev;
__entry->offset = bucket << ca->set->bucket_bits;
),
TP_printk("allocated %d,%d sector=%llu", MAJOR(__entry->dev),
MINOR(__entry->dev), __entry->offset)
);
TRACE_EVENT(bcache_alloc_fail,
@ -426,21 +443,22 @@ TRACE_EVENT(bcache_alloc_fail,
TP_ARGS(ca, reserve),
TP_STRUCT__entry(
__field(dev_t, dev )
__field(unsigned, free )
__field(unsigned, free_inc )
__field(unsigned, unused )
__field(unsigned, blocked )
),
TP_fast_assign(
__entry->dev = ca->bdev->bd_dev;
__entry->free = fifo_used(&ca->free[reserve]);
__entry->free_inc = fifo_used(&ca->free_inc);
__entry->unused = fifo_used(&ca->unused);
__entry->blocked = atomic_read(&ca->set->prio_blocked);
),
TP_printk("free %u free_inc %u unused %u blocked %u", __entry->free,
__entry->free_inc, __entry->unused, __entry->blocked)
TP_printk("alloc fail %d,%d free %u free_inc %u blocked %u",
MAJOR(__entry->dev), MINOR(__entry->dev), __entry->free,
__entry->free_inc, __entry->blocked)
);
/* Background writeback */

10
lib/idr.c
View File

@ -869,6 +869,16 @@ void idr_init(struct idr *idp)
}
EXPORT_SYMBOL(idr_init);
static int idr_has_entry(int id, void *p, void *data)
{
return 1;
}
bool idr_is_empty(struct idr *idp)
{
return !idr_for_each(idp, idr_has_entry, NULL);
}
EXPORT_SYMBOL(idr_is_empty);
/**
* DOC: IDA description