aboutsummaryrefslogtreecommitdiff
path: root/drivers/block
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/block')
-rw-r--r--drivers/block/DAC960.c18
-rw-r--r--drivers/block/Kconfig2
-rw-r--r--drivers/block/brd.c20
-rw-r--r--drivers/block/drbd/drbd_bitmap.c50
-rw-r--r--drivers/block/drbd/drbd_nl.c6
-rw-r--r--drivers/block/floppy.c56
-rw-r--r--drivers/block/hd.c1
-rw-r--r--drivers/block/loop.c40
-rw-r--r--drivers/block/mtip32xx/mtip32xx.c11
-rw-r--r--drivers/block/mtip32xx/mtip32xx.h5
-rw-r--r--drivers/block/nbd.c296
-rw-r--r--drivers/block/nvme.c2
-rw-r--r--drivers/block/pktcdvd.c8
-rw-r--r--drivers/block/rbd.c730
-rw-r--r--drivers/block/rbd_types.h4
-rw-r--r--drivers/block/sunvdc.c5
-rw-r--r--drivers/block/sx8.c2
-rw-r--r--drivers/block/ub.c39
-rw-r--r--drivers/block/viodasd.c809
-rw-r--r--drivers/block/xd.c1
-rw-r--r--drivers/block/xen-blkfront.c3
21 files changed, 701 insertions, 1407 deletions
diff --git a/drivers/block/DAC960.c b/drivers/block/DAC960.c
index e086fbbbe85..8db9089127c 100644
--- a/drivers/block/DAC960.c
+++ b/drivers/block/DAC960.c
@@ -1177,7 +1177,8 @@ static bool DAC960_V1_EnableMemoryMailboxInterface(DAC960_Controller_T
int TimeoutCounter;
int i;
-
+ memset(&CommandMailbox, 0, sizeof(DAC960_V1_CommandMailbox_T));
+
if (pci_set_dma_mask(Controller->PCIDevice, DMA_BIT_MASK(32)))
return DAC960_Failure(Controller, "DMA mask out of range");
Controller->BounceBufferLimit = DMA_BIT_MASK(32);
@@ -4627,7 +4628,8 @@ static void DAC960_V2_ProcessCompletedCommand(DAC960_Command_T *Command)
DAC960_Controller_T *Controller = Command->Controller;
DAC960_CommandType_T CommandType = Command->CommandType;
DAC960_V2_CommandMailbox_T *CommandMailbox = &Command->V2.CommandMailbox;
- DAC960_V2_IOCTL_Opcode_T CommandOpcode = CommandMailbox->Common.IOCTL_Opcode;
+ DAC960_V2_IOCTL_Opcode_T IOCTLOpcode = CommandMailbox->Common.IOCTL_Opcode;
+ DAC960_V2_CommandOpcode_T CommandOpcode = CommandMailbox->SCSI_10.CommandOpcode;
DAC960_V2_CommandStatus_T CommandStatus = Command->V2.CommandStatus;
if (CommandType == DAC960_ReadCommand ||
@@ -4699,7 +4701,7 @@ static void DAC960_V2_ProcessCompletedCommand(DAC960_Command_T *Command)
{
if (Controller->ShutdownMonitoringTimer)
return;
- if (CommandOpcode == DAC960_V2_GetControllerInfo)
+ if (IOCTLOpcode == DAC960_V2_GetControllerInfo)
{
DAC960_V2_ControllerInfo_T *NewControllerInfo =
Controller->V2.NewControllerInformation;
@@ -4719,14 +4721,14 @@ static void DAC960_V2_ProcessCompletedCommand(DAC960_Command_T *Command)
memcpy(ControllerInfo, NewControllerInfo,
sizeof(DAC960_V2_ControllerInfo_T));
}
- else if (CommandOpcode == DAC960_V2_GetEvent)
+ else if (IOCTLOpcode == DAC960_V2_GetEvent)
{
if (CommandStatus == DAC960_V2_NormalCompletion) {
DAC960_V2_ReportEvent(Controller, Controller->V2.Event);
}
Controller->V2.NextEventSequenceNumber++;
}
- else if (CommandOpcode == DAC960_V2_GetPhysicalDeviceInfoValid &&
+ else if (IOCTLOpcode == DAC960_V2_GetPhysicalDeviceInfoValid &&
CommandStatus == DAC960_V2_NormalCompletion)
{
DAC960_V2_PhysicalDeviceInfo_T *NewPhysicalDeviceInfo =
@@ -4915,7 +4917,7 @@ static void DAC960_V2_ProcessCompletedCommand(DAC960_Command_T *Command)
NewPhysicalDeviceInfo->LogicalUnit++;
Controller->V2.PhysicalDeviceIndex++;
}
- else if (CommandOpcode == DAC960_V2_GetPhysicalDeviceInfoValid)
+ else if (IOCTLOpcode == DAC960_V2_GetPhysicalDeviceInfoValid)
{
unsigned int DeviceIndex;
for (DeviceIndex = Controller->V2.PhysicalDeviceIndex;
@@ -4938,7 +4940,7 @@ static void DAC960_V2_ProcessCompletedCommand(DAC960_Command_T *Command)
}
Controller->V2.NeedPhysicalDeviceInformation = false;
}
- else if (CommandOpcode == DAC960_V2_GetLogicalDeviceInfoValid &&
+ else if (IOCTLOpcode == DAC960_V2_GetLogicalDeviceInfoValid &&
CommandStatus == DAC960_V2_NormalCompletion)
{
DAC960_V2_LogicalDeviceInfo_T *NewLogicalDeviceInfo =
@@ -5065,7 +5067,7 @@ static void DAC960_V2_ProcessCompletedCommand(DAC960_Command_T *Command)
[LogicalDeviceNumber] = true;
NewLogicalDeviceInfo->LogicalDeviceNumber++;
}
- else if (CommandOpcode == DAC960_V2_GetLogicalDeviceInfoValid)
+ else if (IOCTLOpcode == DAC960_V2_GetLogicalDeviceInfoValid)
{
int LogicalDriveNumber;
for (LogicalDriveNumber = 0;
diff --git a/drivers/block/Kconfig b/drivers/block/Kconfig
index 4e4c8a4a5fd..a796407123c 100644
--- a/drivers/block/Kconfig
+++ b/drivers/block/Kconfig
@@ -354,7 +354,7 @@ config BLK_DEV_SX8
Use devices /dev/sx8/$N and /dev/sx8/$Np$M.
config BLK_DEV_UB
- tristate "Low Performance USB Block driver"
+ tristate "Low Performance USB Block driver (deprecated)"
depends on USB
help
This driver supports certain USB attached storage devices
diff --git a/drivers/block/brd.c b/drivers/block/brd.c
index ec246437f5a..531ceb31d0f 100644
--- a/drivers/block/brd.c
+++ b/drivers/block/brd.c
@@ -242,9 +242,9 @@ static void copy_to_brd(struct brd_device *brd, const void *src,
page = brd_lookup_page(brd, sector);
BUG_ON(!page);
- dst = kmap_atomic(page, KM_USER1);
+ dst = kmap_atomic(page);
memcpy(dst + offset, src, copy);
- kunmap_atomic(dst, KM_USER1);
+ kunmap_atomic(dst);
if (copy < n) {
src += copy;
@@ -253,9 +253,9 @@ static void copy_to_brd(struct brd_device *brd, const void *src,
page = brd_lookup_page(brd, sector);
BUG_ON(!page);
- dst = kmap_atomic(page, KM_USER1);
+ dst = kmap_atomic(page);
memcpy(dst, src, copy);
- kunmap_atomic(dst, KM_USER1);
+ kunmap_atomic(dst);
}
}
@@ -273,9 +273,9 @@ static void copy_from_brd(void *dst, struct brd_device *brd,
copy = min_t(size_t, n, PAGE_SIZE - offset);
page = brd_lookup_page(brd, sector);
if (page) {
- src = kmap_atomic(page, KM_USER1);
+ src = kmap_atomic(page);
memcpy(dst, src + offset, copy);
- kunmap_atomic(src, KM_USER1);
+ kunmap_atomic(src);
} else
memset(dst, 0, copy);
@@ -285,9 +285,9 @@ static void copy_from_brd(void *dst, struct brd_device *brd,
copy = n - copy;
page = brd_lookup_page(brd, sector);
if (page) {
- src = kmap_atomic(page, KM_USER1);
+ src = kmap_atomic(page);
memcpy(dst, src, copy);
- kunmap_atomic(src, KM_USER1);
+ kunmap_atomic(src);
} else
memset(dst, 0, copy);
}
@@ -309,7 +309,7 @@ static int brd_do_bvec(struct brd_device *brd, struct page *page,
goto out;
}
- mem = kmap_atomic(page, KM_USER0);
+ mem = kmap_atomic(page);
if (rw == READ) {
copy_from_brd(mem + off, brd, sector, len);
flush_dcache_page(page);
@@ -317,7 +317,7 @@ static int brd_do_bvec(struct brd_device *brd, struct page *page,
flush_dcache_page(page);
copy_to_brd(brd, mem + off, sector, len);
}
- kunmap_atomic(mem, KM_USER0);
+ kunmap_atomic(mem);
out:
return err;
diff --git a/drivers/block/drbd/drbd_bitmap.c b/drivers/block/drbd/drbd_bitmap.c
index 912f585a760..3030201c69d 100644
--- a/drivers/block/drbd/drbd_bitmap.c
+++ b/drivers/block/drbd/drbd_bitmap.c
@@ -289,25 +289,25 @@ static unsigned int bm_bit_to_page_idx(struct drbd_bitmap *b, u64 bitnr)
return page_nr;
}
-static unsigned long *__bm_map_pidx(struct drbd_bitmap *b, unsigned int idx, const enum km_type km)
+static unsigned long *__bm_map_pidx(struct drbd_bitmap *b, unsigned int idx)
{
struct page *page = b->bm_pages[idx];
- return (unsigned long *) kmap_atomic(page, km);
+ return (unsigned long *) kmap_atomic(page);
}
static unsigned long *bm_map_pidx(struct drbd_bitmap *b, unsigned int idx)
{
- return __bm_map_pidx(b, idx, KM_IRQ1);
+ return __bm_map_pidx(b, idx);
}
-static void __bm_unmap(unsigned long *p_addr, const enum km_type km)
+static void __bm_unmap(unsigned long *p_addr)
{
- kunmap_atomic(p_addr, km);
+ kunmap_atomic(p_addr);
};
static void bm_unmap(unsigned long *p_addr)
{
- return __bm_unmap(p_addr, KM_IRQ1);
+ return __bm_unmap(p_addr);
}
/* long word offset of _bitmap_ sector */
@@ -543,15 +543,15 @@ static unsigned long bm_count_bits(struct drbd_bitmap *b)
/* all but last page */
for (idx = 0; idx < b->bm_number_of_pages - 1; idx++) {
- p_addr = __bm_map_pidx(b, idx, KM_USER0);
+ p_addr = __bm_map_pidx(b, idx);
for (i = 0; i < LWPP; i++)
bits += hweight_long(p_addr[i]);
- __bm_unmap(p_addr, KM_USER0);
+ __bm_unmap(p_addr);
cond_resched();
}
/* last (or only) page */
last_word = ((b->bm_bits - 1) & BITS_PER_PAGE_MASK) >> LN2_BPL;
- p_addr = __bm_map_pidx(b, idx, KM_USER0);
+ p_addr = __bm_map_pidx(b, idx);
for (i = 0; i < last_word; i++)
bits += hweight_long(p_addr[i]);
p_addr[last_word] &= cpu_to_lel(mask);
@@ -559,7 +559,7 @@ static unsigned long bm_count_bits(struct drbd_bitmap *b)
/* 32bit arch, may have an unused padding long */
if (BITS_PER_LONG == 32 && (last_word & 1) == 0)
p_addr[last_word+1] = 0;
- __bm_unmap(p_addr, KM_USER0);
+ __bm_unmap(p_addr);
return bits;
}
@@ -970,11 +970,11 @@ static void bm_page_io_async(struct bm_aio_ctx *ctx, int page_nr, int rw) __must
* to use pre-allocated page pool */
void *src, *dest;
page = alloc_page(__GFP_HIGHMEM|__GFP_WAIT);
- dest = kmap_atomic(page, KM_USER0);
- src = kmap_atomic(b->bm_pages[page_nr], KM_USER1);
+ dest = kmap_atomic(page);
+ src = kmap_atomic(b->bm_pages[page_nr]);
memcpy(dest, src, PAGE_SIZE);
- kunmap_atomic(src, KM_USER1);
- kunmap_atomic(dest, KM_USER0);
+ kunmap_atomic(src);
+ kunmap_atomic(dest);
bm_store_page_idx(page, page_nr);
} else
page = b->bm_pages[page_nr];
@@ -1163,7 +1163,7 @@ 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,
- const int find_zero_bit, const enum km_type km)
+ const int find_zero_bit)
{
struct drbd_bitmap *b = mdev->bitmap;
unsigned long *p_addr;
@@ -1178,7 +1178,7 @@ static unsigned long __bm_find_next(struct drbd_conf *mdev, unsigned long bm_fo,
while (bm_fo < b->bm_bits) {
/* bit offset of the first bit in the page */
bit_offset = bm_fo & ~BITS_PER_PAGE_MASK;
- p_addr = __bm_map_pidx(b, bm_bit_to_page_idx(b, bm_fo), km);
+ p_addr = __bm_map_pidx(b, bm_bit_to_page_idx(b, bm_fo));
if (find_zero_bit)
i = find_next_zero_bit_le(p_addr,
@@ -1187,7 +1187,7 @@ static unsigned long __bm_find_next(struct drbd_conf *mdev, unsigned long bm_fo,
i = find_next_bit_le(p_addr,
PAGE_SIZE*8, bm_fo & BITS_PER_PAGE_MASK);
- __bm_unmap(p_addr, km);
+ __bm_unmap(p_addr);
if (i < PAGE_SIZE*8) {
bm_fo = bit_offset + i;
if (bm_fo >= b->bm_bits)
@@ -1215,7 +1215,7 @@ static unsigned long bm_find_next(struct drbd_conf *mdev,
if (BM_DONT_TEST & b->bm_flags)
bm_print_lock_info(mdev);
- i = __bm_find_next(mdev, bm_fo, find_zero_bit, KM_IRQ1);
+ i = __bm_find_next(mdev, bm_fo, find_zero_bit);
spin_unlock_irq(&b->bm_lock);
return i;
@@ -1239,13 +1239,13 @@ unsigned long drbd_bm_find_next_zero(struct drbd_conf *mdev, unsigned long bm_fo
unsigned long _drbd_bm_find_next(struct drbd_conf *mdev, unsigned long bm_fo)
{
/* WARN_ON(!(BM_DONT_SET & mdev->b->bm_flags)); */
- return __bm_find_next(mdev, bm_fo, 0, KM_USER1);
+ return __bm_find_next(mdev, bm_fo, 0);
}
unsigned long _drbd_bm_find_next_zero(struct drbd_conf *mdev, unsigned long bm_fo)
{
/* WARN_ON(!(BM_DONT_SET & mdev->b->bm_flags)); */
- return __bm_find_next(mdev, bm_fo, 1, KM_USER1);
+ return __bm_find_next(mdev, bm_fo, 1);
}
/* returns number of bits actually changed.
@@ -1273,14 +1273,14 @@ static int __bm_change_bits_to(struct drbd_conf *mdev, const unsigned long s,
unsigned int page_nr = bm_bit_to_page_idx(b, bitnr);
if (page_nr != last_page_nr) {
if (p_addr)
- __bm_unmap(p_addr, KM_IRQ1);
+ __bm_unmap(p_addr);
if (c < 0)
bm_set_page_lazy_writeout(b->bm_pages[last_page_nr]);
else if (c > 0)
bm_set_page_need_writeout(b->bm_pages[last_page_nr]);
changed_total += c;
c = 0;
- p_addr = __bm_map_pidx(b, page_nr, KM_IRQ1);
+ p_addr = __bm_map_pidx(b, page_nr);
last_page_nr = page_nr;
}
if (val)
@@ -1289,7 +1289,7 @@ static int __bm_change_bits_to(struct drbd_conf *mdev, const unsigned long s,
c -= (0 != __test_and_clear_bit_le(bitnr & BITS_PER_PAGE_MASK, p_addr));
}
if (p_addr)
- __bm_unmap(p_addr, KM_IRQ1);
+ __bm_unmap(p_addr);
if (c < 0)
bm_set_page_lazy_writeout(b->bm_pages[last_page_nr]);
else if (c > 0)
@@ -1342,13 +1342,13 @@ static inline void bm_set_full_words_within_one_page(struct drbd_bitmap *b,
{
int i;
int bits;
- unsigned long *paddr = kmap_atomic(b->bm_pages[page_nr], KM_IRQ1);
+ unsigned long *paddr = kmap_atomic(b->bm_pages[page_nr]);
for (i = first_word; i < last_word; i++) {
bits = hweight_long(paddr[i]);
paddr[i] = ~0UL;
b->bm_set += BITS_PER_LONG - bits;
}
- kunmap_atomic(paddr, KM_IRQ1);
+ kunmap_atomic(paddr);
}
/* Same thing as drbd_bm_set_bits,
diff --git a/drivers/block/drbd/drbd_nl.c b/drivers/block/drbd/drbd_nl.c
index af2a25049bc..abfaacaaf34 100644
--- a/drivers/block/drbd/drbd_nl.c
+++ b/drivers/block/drbd/drbd_nl.c
@@ -179,7 +179,7 @@ int drbd_khelper(struct drbd_conf *mdev, char *cmd)
dev_info(DEV, "helper command: %s %s %s\n", usermode_helper, cmd, mb);
drbd_bcast_ev_helper(mdev, cmd);
- ret = call_usermodehelper(usermode_helper, argv, envp, 1);
+ ret = call_usermodehelper(usermode_helper, argv, envp, UMH_WAIT_PROC);
if (ret)
dev_warn(DEV, "helper command: %s %s %s exit code %u (0x%x)\n",
usermode_helper, cmd, mb,
@@ -2526,10 +2526,10 @@ void drbd_bcast_ee(struct drbd_conf *mdev,
page = e->pages;
page_chain_for_each(page) {
- void *d = kmap_atomic(page, KM_USER0);
+ void *d = kmap_atomic(page);
unsigned l = min_t(unsigned, len, PAGE_SIZE);
memcpy(tl, d, l);
- kunmap_atomic(d, KM_USER0);
+ kunmap_atomic(d);
tl = (unsigned short*)((char*)tl + l);
len -= l;
if (len == 0)
diff --git a/drivers/block/floppy.c b/drivers/block/floppy.c
index 510fb10ec45..b0b00d70c16 100644
--- a/drivers/block/floppy.c
+++ b/drivers/block/floppy.c
@@ -202,7 +202,6 @@ static int slow_floppy;
#include <asm/dma.h>
#include <asm/irq.h>
-#include <asm/system.h>
static int FLOPPY_IRQ = 6;
static int FLOPPY_DMA = 2;
@@ -1031,37 +1030,6 @@ static int fd_wait_for_completion(unsigned long delay, timeout_fn function)
return 0;
}
-static DEFINE_SPINLOCK(floppy_hlt_lock);
-static int hlt_disabled;
-static void floppy_disable_hlt(void)
-{
- unsigned long flags;
-
- WARN_ONCE(1, "floppy_disable_hlt() scheduled for removal in 2012");
- spin_lock_irqsave(&floppy_hlt_lock, flags);
- if (!hlt_disabled) {
- hlt_disabled = 1;
-#ifdef HAVE_DISABLE_HLT
- disable_hlt();
-#endif
- }
- spin_unlock_irqrestore(&floppy_hlt_lock, flags);
-}
-
-static void floppy_enable_hlt(void)
-{
- unsigned long flags;
-
- spin_lock_irqsave(&floppy_hlt_lock, flags);
- if (hlt_disabled) {
- hlt_disabled = 0;
-#ifdef HAVE_DISABLE_HLT
- enable_hlt();
-#endif
- }
- spin_unlock_irqrestore(&floppy_hlt_lock, flags);
-}
-
static void setup_DMA(void)
{
unsigned long f;
@@ -1106,7 +1074,6 @@ static void setup_DMA(void)
fd_enable_dma();
release_dma_lock(f);
#endif
- floppy_disable_hlt();
}
static void show_floppy(void);
@@ -1708,7 +1675,6 @@ irqreturn_t floppy_interrupt(int irq, void *dev_id)
fd_disable_dma();
release_dma_lock(f);
- floppy_enable_hlt();
do_floppy = NULL;
if (fdc >= N_FDC || FDCS->address == -1) {
/* we don't even know which FDC is the culprit */
@@ -1857,8 +1823,6 @@ static void floppy_shutdown(unsigned long data)
show_floppy();
cancel_activity();
- floppy_enable_hlt();
-
flags = claim_dma_lock();
fd_disable_dma();
release_dma_lock(flags);
@@ -3832,7 +3796,7 @@ static int __floppy_read_block_0(struct block_device *bdev)
bio.bi_size = size;
bio.bi_bdev = bdev;
bio.bi_sector = 0;
- bio.bi_flags = BIO_QUIET;
+ bio.bi_flags = (1 << BIO_QUIET);
init_completion(&complete);
bio.bi_private = &complete;
bio.bi_end_io = floppy_rb0_complete;
@@ -4368,8 +4332,14 @@ out_unreg_blkdev:
out_put_disk:
while (dr--) {
del_timer_sync(&motor_off_timer[dr]);
- if (disks[dr]->queue)
+ if (disks[dr]->queue) {
blk_cleanup_queue(disks[dr]->queue);
+ /*
+ * put_disk() is not paired with add_disk() and
+ * will put queue reference one extra time. fix it.
+ */
+ disks[dr]->queue = NULL;
+ }
put_disk(disks[dr]);
}
return err;
@@ -4503,7 +4473,6 @@ static void floppy_release_irq_and_dma(void)
#if N_FDC > 1
set_dor(1, ~8, 0);
#endif
- floppy_enable_hlt();
if (floppy_track_buffer && max_buffer_sectors) {
tmpsize = max_buffer_sectors * 1024;
@@ -4579,6 +4548,15 @@ static void __exit floppy_module_exit(void)
platform_device_unregister(&floppy_device[drive]);
}
blk_cleanup_queue(disks[drive]->queue);
+
+ /*
+ * These disks have not called add_disk(). Don't put down
+ * queue reference in put_disk().
+ */
+ if (!(allowed_drive_mask & (1 << drive)) ||
+ fdc_state[FDC(drive)].version == FDC_NONE)
+ disks[drive]->queue = NULL;
+
put_disk(disks[drive]);
}
diff --git a/drivers/block/hd.c b/drivers/block/hd.c
index b52c9ca146f..bf397bf108b 100644
--- a/drivers/block/hd.c
+++ b/drivers/block/hd.c
@@ -44,7 +44,6 @@
#define HD_IRQ 14
#define REALLY_SLOW_IO
-#include <asm/system.h>
#include <asm/io.h>
#include <asm/uaccess.h>
diff --git a/drivers/block/loop.c b/drivers/block/loop.c
index f00257782fc..bbca966f8f6 100644
--- a/drivers/block/loop.c
+++ b/drivers/block/loop.c
@@ -93,16 +93,16 @@ static int transfer_none(struct loop_device *lo, int cmd,
struct page *loop_page, unsigned loop_off,
int size, sector_t real_block)
{
- char *raw_buf = kmap_atomic(raw_page, KM_USER0) + raw_off;
- char *loop_buf = kmap_atomic(loop_page, KM_USER1) + loop_off;
+ char *raw_buf = kmap_atomic(raw_page) + raw_off;
+ char *loop_buf = kmap_atomic(loop_page) + loop_off;
if (cmd == READ)
memcpy(loop_buf, raw_buf, size);
else
memcpy(raw_buf, loop_buf, size);
- kunmap_atomic(loop_buf, KM_USER1);
- kunmap_atomic(raw_buf, KM_USER0);
+ kunmap_atomic(loop_buf);
+ kunmap_atomic(raw_buf);
cond_resched();
return 0;
}
@@ -112,8 +112,8 @@ static int transfer_xor(struct loop_device *lo, int cmd,
struct page *loop_page, unsigned loop_off,
int size, sector_t real_block)
{
- char *raw_buf = kmap_atomic(raw_page, KM_USER0) + raw_off;
- char *loop_buf = kmap_atomic(loop_page, KM_USER1) + loop_off;
+ char *raw_buf = kmap_atomic(raw_page) + raw_off;
+ char *loop_buf = kmap_atomic(loop_page) + loop_off;
char *in, *out, *key;
int i, keysize;
@@ -130,8 +130,8 @@ static int transfer_xor(struct loop_device *lo, int cmd,
for (i = 0; i < size; i++)
*out++ = *in++ ^ key[(i & 511) % keysize];
- kunmap_atomic(loop_buf, KM_USER1);
- kunmap_atomic(raw_buf, KM_USER0);
+ kunmap_atomic(loop_buf);
+ kunmap_atomic(raw_buf);
cond_resched();
return 0;
}
@@ -356,14 +356,14 @@ lo_direct_splice_actor(struct pipe_inode_info *pipe, struct splice_desc *sd)
return __splice_from_pipe(pipe, sd, lo_splice_actor);
}
-static int
+static ssize_t
do_lo_receive(struct loop_device *lo,
struct bio_vec *bvec, int bsize, loff_t pos)
{
struct lo_read_data cookie;
struct splice_desc sd;
struct file *file;
- long retval;
+ ssize_t retval;
cookie.lo = lo;
cookie.page = bvec->bv_page;
@@ -379,26 +379,28 @@ do_lo_receive(struct loop_device *lo,
file = lo->lo_backing_file;
retval = splice_direct_to_actor(file, &sd, lo_direct_splice_actor);
- if (retval < 0)
- return retval;
- if (retval != bvec->bv_len)
- return -EIO;
- return 0;
+ return retval;
}
static int
lo_receive(struct loop_device *lo, struct bio *bio, int bsize, loff_t pos)
{
struct bio_vec *bvec;
- int i, ret = 0;
+ ssize_t s;
+ int i;
bio_for_each_segment(bvec, bio, i) {
- ret = do_lo_receive(lo, bvec, bsize, pos);
- if (ret < 0)
+ s = do_lo_receive(lo, bvec, bsize, pos);
+ if (s < 0)
+ return s;
+
+ if (s != bvec->bv_len) {
+ zero_fill_bio(bio);
break;
+ }
pos += bvec->bv_len;
}
- return ret;
+ return 0;
}
static int do_bio_filebacked(struct loop_device *lo, struct bio *bio)
diff --git a/drivers/block/mtip32xx/mtip32xx.c b/drivers/block/mtip32xx/mtip32xx.c
index b74eab70c3d..8eb81c96608 100644
--- a/drivers/block/mtip32xx/mtip32xx.c
+++ b/drivers/block/mtip32xx/mtip32xx.c
@@ -2068,8 +2068,6 @@ static int mtip_hw_ioctl(struct driver_data *dd, unsigned int cmd,
* when the read completes.
* @data Callback data passed to the callback function
* when the read completes.
- * @barrier If non-zero, this command must be completed before
- * issuing any other commands.
* @dir Direction (read or write)
*
* return value
@@ -2077,7 +2075,7 @@ static int mtip_hw_ioctl(struct driver_data *dd, unsigned int cmd,
*/
static void mtip_hw_submit_io(struct driver_data *dd, sector_t start,
int nsect, int nents, int tag, void *callback,
- void *data, int barrier, int dir)
+ void *data, int dir)
{
struct host_to_dev_fis *fis;
struct mtip_port *port = dd->port;
@@ -2108,8 +2106,6 @@ static void mtip_hw_submit_io(struct driver_data *dd, sector_t start,
*((unsigned int *) &fis->lba_low) = (start & 0xFFFFFF);
*((unsigned int *) &fis->lba_low_ex) = ((start >> 24) & 0xFFFFFF);
fis->device = 1 << 6;
- if (barrier)
- fis->device |= FUA_BIT;
fis->features = nsect & 0xFF;
fis->features_ex = (nsect >> 8) & 0xFF;
fis->sect_count = ((tag << 3) | (tag >> 5));
@@ -3087,7 +3083,6 @@ static void mtip_make_request(struct request_queue *queue, struct bio *bio)
tag,
bio_endio,
bio,
- bio->bi_rw & REQ_FUA,
bio_data_dir(bio));
} else
bio_io_error(bio);
@@ -3187,6 +3182,10 @@ skip_create_disk:
blk_queue_max_segments(dd->queue, MTIP_MAX_SG);
blk_queue_physical_block_size(dd->queue, 4096);
blk_queue_io_min(dd->queue, 4096);
+ /*
+ * write back cache is not supported in the device. FUA depends on
+ * write back cache support, hence setting flush support to zero.
+ */
blk_queue_flush(dd->queue, 0);
/* Set the capacity of the device in 512 byte sectors. */
diff --git a/drivers/block/mtip32xx/mtip32xx.h b/drivers/block/mtip32xx/mtip32xx.h
index 723d7c4946d..e0554a8f223 100644
--- a/drivers/block/mtip32xx/mtip32xx.h
+++ b/drivers/block/mtip32xx/mtip32xx.h
@@ -104,9 +104,6 @@
/* BAR number used to access the HBA registers. */
#define MTIP_ABAR 5
-/* Forced Unit Access Bit */
-#define FUA_BIT 0x80
-
#ifdef DEBUG
#define dbg_printk(format, arg...) \
printk(pr_fmt(format), ##arg);
@@ -415,8 +412,6 @@ struct driver_data {
atomic_t resumeflag; /* Atomic variable to track suspend/resume */
- atomic_t eh_active; /* Flag for error handling tracking */
-
struct task_struct *mtip_svc_handler; /* task_struct of svc thd */
};
diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c
index c3f0ee16594..061427a75d3 100644
--- a/drivers/block/nbd.c
+++ b/drivers/block/nbd.c
@@ -34,12 +34,11 @@
#include <linux/kthread.h>
#include <asm/uaccess.h>
-#include <asm/system.h>
#include <asm/types.h>
#include <linux/nbd.h>
-#define LO_MAGIC 0x68797548
+#define NBD_MAGIC 0x68797548
#ifdef NDEBUG
#define dprintk(flags, fmt...)
@@ -116,7 +115,7 @@ static void nbd_end_request(struct request *req)
spin_unlock_irqrestore(q->queue_lock, flags);
}
-static void sock_shutdown(struct nbd_device *lo, int lock)
+static void sock_shutdown(struct nbd_device *nbd, int lock)
{
/* Forcibly shutdown the socket causing all listeners
* to error
@@ -125,14 +124,14 @@ static void sock_shutdown(struct nbd_device *lo, int lock)
* there should be a more generic interface rather than
* calling socket ops directly here */
if (lock)
- mutex_lock(&lo->tx_lock);
- if (lo->sock) {
- dev_warn(disk_to_dev(lo->disk), "shutting down socket\n");
- kernel_sock_shutdown(lo->sock, SHUT_RDWR);
- lo->sock = NULL;
+ mutex_lock(&nbd->tx_lock);
+ if (nbd->sock) {
+ dev_warn(disk_to_dev(nbd->disk), "shutting down socket\n");
+ kernel_sock_shutdown(nbd->sock, SHUT_RDWR);
+ nbd->sock = NULL;
}
if (lock)
- mutex_unlock(&lo->tx_lock);
+ mutex_unlock(&nbd->tx_lock);
}
static void nbd_xmit_timeout(unsigned long arg)
@@ -147,17 +146,17 @@ static void nbd_xmit_timeout(unsigned long arg)
/*
* Send or receive packet.
*/
-static int sock_xmit(struct nbd_device *lo, int send, void *buf, int size,
+static int sock_xmit(struct nbd_device *nbd, int send, void *buf, int size,
int msg_flags)
{
- struct socket *sock = lo->sock;
+ struct socket *sock = nbd->sock;
int result;
struct msghdr msg;
struct kvec iov;
sigset_t blocked, oldset;
if (unlikely(!sock)) {
- dev_err(disk_to_dev(lo->disk),
+ dev_err(disk_to_dev(nbd->disk),
"Attempted %s on closed socket in sock_xmit\n",
(send ? "send" : "recv"));
return -EINVAL;
@@ -181,15 +180,15 @@ static int sock_xmit(struct nbd_device *lo, int send, void *buf, int size,
if (send) {
struct timer_list ti;
- if (lo->xmit_timeout) {
+ if (nbd->xmit_timeout) {
init_timer(&ti);
ti.function = nbd_xmit_timeout;
ti.data = (unsigned long)current;
- ti.expires = jiffies + lo->xmit_timeout;
+ ti.expires = jiffies + nbd->xmit_timeout;
add_timer(&ti);
}
result = kernel_sendmsg(sock, &msg, &iov, 1, size);
- if (lo->xmit_timeout)
+ if (nbd->xmit_timeout)
del_timer_sync(&ti);
} else
result = kernel_recvmsg(sock, &msg, &iov, 1, size,
@@ -201,7 +200,7 @@ static int sock_xmit(struct nbd_device *lo, int send, void *buf, int size,
task_pid_nr(current), current->comm,
dequeue_signal_lock(current, &current->blocked, &info));
result = -EINTR;
- sock_shutdown(lo, !send);
+ sock_shutdown(nbd, !send);
break;
}
@@ -219,18 +218,19 @@ static int sock_xmit(struct nbd_device *lo, int send, void *buf, int size,
return result;
}
-static inline int sock_send_bvec(struct nbd_device *lo, struct bio_vec *bvec,
+static inline int sock_send_bvec(struct nbd_device *nbd, struct bio_vec *bvec,
int flags)
{
int result;
void *kaddr = kmap(bvec->bv_page);
- result = sock_xmit(lo, 1, kaddr + bvec->bv_offset, bvec->bv_len, flags);
+ result = sock_xmit(nbd, 1, kaddr + bvec->bv_offset,
+ bvec->bv_len, flags);
kunmap(bvec->bv_page);
return result;
}
/* always call with the tx_lock held */
-static int nbd_send_req(struct nbd_device *lo, struct request *req)
+static int nbd_send_req(struct nbd_device *nbd, struct request *req)
{
int result, flags;
struct nbd_request request;
@@ -243,14 +243,14 @@ static int nbd_send_req(struct nbd_device *lo, struct request *req)
memcpy(request.handle, &req, sizeof(req));
dprintk(DBG_TX, "%s: request %p: sending control (%s@%llu,%uB)\n",
- lo->disk->disk_name, req,
+ nbd->disk->disk_name, req,
nbdcmd_to_ascii(nbd_cmd(req)),
(unsigned long long)blk_rq_pos(req) << 9,
blk_rq_bytes(req));
- result = sock_xmit(lo, 1, &request, sizeof(request),
+ result = sock_xmit(nbd, 1, &request, sizeof(request),
(nbd_cmd(req) == NBD_CMD_WRITE) ? MSG_MORE : 0);
if (result <= 0) {
- dev_err(disk_to_dev(lo->disk),
+ dev_err(disk_to_dev(nbd->disk),
"Send control failed (result %d)\n", result);
goto error_out;
}
@@ -267,10 +267,10 @@ static int nbd_send_req(struct nbd_device *lo, struct request *req)
if (!rq_iter_last(req, iter))
flags = MSG_MORE;
dprintk(DBG_TX, "%s: request %p: sending %d bytes data\n",
- lo->disk->disk_name, req, bvec->bv_len);
- result = sock_send_bvec(lo, bvec, flags);
+ nbd->disk->disk_name, req, bvec->bv_len);
+ result = sock_send_bvec(nbd, bvec, flags);
if (result <= 0) {
- dev_err(disk_to_dev(lo->disk),
+ dev_err(disk_to_dev(nbd->disk),
"Send data failed (result %d)\n",
result);
goto error_out;
@@ -283,25 +283,25 @@ error_out:
return -EIO;
}
-static struct request *nbd_find_request(struct nbd_device *lo,
+static struct request *nbd_find_request(struct nbd_device *nbd,
struct request *xreq)
{
struct request *req, *tmp;
int err;
- err = wait_event_interruptible(lo->active_wq, lo->active_req != xreq);
+ err = wait_event_interruptible(nbd->active_wq, nbd->active_req != xreq);
if (unlikely(err))
goto out;
- spin_lock(&lo->queue_lock);
- list_for_each_entry_safe(req, tmp, &lo->queue_head, queuelist) {
+ spin_lock(&nbd->queue_lock);
+ list_for_each_entry_safe(req, tmp, &nbd->queue_head, queuelist) {
if (req != xreq)
continue;
list_del_init(&req->queuelist);
- spin_unlock(&lo->queue_lock);
+ spin_unlock(&nbd->queue_lock);
return req;
}
- spin_unlock(&lo->queue_lock);
+ spin_unlock(&nbd->queue_lock);
err = -ENOENT;
@@ -309,78 +309,78 @@ out:
return ERR_PTR(err);
}
-static inline int sock_recv_bvec(struct nbd_device *lo, struct bio_vec *bvec)
+static inline int sock_recv_bvec(struct nbd_device *nbd, struct bio_vec *bvec)
{
int result;
void *kaddr = kmap(bvec->bv_page);
- result = sock_xmit(lo, 0, kaddr + bvec->bv_offset, bvec->bv_len,
+ result = sock_xmit(nbd, 0, kaddr + bvec->bv_offset, bvec->bv_len,
MSG_WAITALL);
kunmap(bvec->bv_page);
return result;
}
/* NULL returned = something went wrong, inform userspace */
-static struct request *nbd_read_stat(struct nbd_device *lo)
+static struct request *nbd_read_stat(struct nbd_device *nbd)
{
int result;
struct nbd_reply reply;
struct request *req;
reply.magic = 0;
- result = sock_xmit(lo, 0, &reply, sizeof(reply), MSG_WAITALL);
+ result = sock_xmit(nbd, 0, &reply, sizeof(reply), MSG_WAITALL);
if (result <= 0) {
- dev_err(disk_to_dev(lo->disk),
+ dev_err(disk_to_dev(nbd->disk),
"Receive control failed (result %d)\n", result);
goto harderror;
}
if (ntohl(reply.magic) != NBD_REPLY_MAGIC) {
- dev_err(disk_to_dev(lo->disk), "Wrong magic (0x%lx)\n",
+ dev_err(disk_to_dev(nbd->disk), "Wrong magic (0x%lx)\n",
(unsigned long)ntohl(reply.magic));
result = -EPROTO;
goto harderror;
}
- req = nbd_find_request(lo, *(struct request **)reply.handle);
+ req = nbd_find_request(nbd, *(struct request **)reply.handle);
if (IS_ERR(req)) {
result = PTR_ERR(req);
if (result != -ENOENT)
goto harderror;
- dev_err(disk_to_dev(lo->disk), "Unexpected reply (%p)\n",
+ dev_err(disk_to_dev(nbd->disk), "Unexpected reply (%p)\n",
reply.handle);
result = -EBADR;
goto harderror;
}
if (ntohl(reply.error)) {
- dev_err(disk_to_dev(lo->disk), "Other side returned error (%d)\n",
+ dev_err(disk_to_dev(nbd->disk), "Other side returned error (%d)\n",
ntohl(reply.error));
req->errors++;
return req;
}
dprintk(DBG_RX, "%s: request %p: got reply\n",
- lo->disk->disk_name, req);
+ nbd->disk->disk_name, req);
if (nbd_cmd(req) == NBD_CMD_READ) {
struct req_iterator iter;
struct bio_vec *bvec;
rq_for_each_segment(bvec, req, iter) {
- result = sock_recv_bvec(lo, bvec);
+ result = sock_recv_bvec(nbd, bvec);
if (result <= 0) {
- dev_err(disk_to_dev(lo->disk), "Receive data failed (result %d)\n",
+ dev_err(disk_to_dev(nbd->disk), "Receive data failed (result %d)\n",
result);
req->errors++;
return req;
}
dprintk(DBG_RX, "%s: request %p: got %d bytes data\n",
- lo->disk->disk_name, req, bvec->bv_len);
+ nbd->disk->disk_name, req, bvec->bv_len);
}
}
return req;
harderror:
- lo->harderror = result;
+ nbd->harderror = result;
return NULL;
}
@@ -398,48 +398,48 @@ static struct device_attribute pid_attr = {
.show = pid_show,
};
-static int nbd_do_it(struct nbd_device *lo)
+static int nbd_do_it(struct nbd_device *nbd)
{
struct request *req;
int ret;
- BUG_ON(lo->magic != LO_MAGIC);
+ BUG_ON(nbd->magic != NBD_MAGIC);
- lo->pid = task_pid_nr(current);
- ret = device_create_file(disk_to_dev(lo->disk), &pid_attr);
+ nbd->pid = task_pid_nr(current);
+ ret = device_create_file(disk_to_dev(nbd->disk), &pid_attr);
if (ret) {
- dev_err(disk_to_dev(lo->disk), "device_create_file failed!\n");
- lo->pid = 0;
+ dev_err(disk_to_dev(nbd->disk), "device_create_file failed!\n");
+ nbd->pid = 0;
return ret;
}
- while ((req = nbd_read_stat(lo)) != NULL)
+ while ((req = nbd_read_stat(nbd)) != NULL)
nbd_end_request(req);
- device_remove_file(disk_to_dev(lo->disk), &pid_attr);
- lo->pid = 0;
+ device_remove_file(disk_to_dev(nbd->disk), &pid_attr);
+ nbd->pid = 0;
return 0;
}
-static void nbd_clear_que(struct nbd_device *lo)
+static void nbd_clear_que(struct nbd_device *nbd)
{
struct request *req;
- BUG_ON(lo->magic != LO_MAGIC);
+ BUG_ON(nbd->magic != NBD_MAGIC);
/*
- * Because we have set lo->sock to NULL under the tx_lock, all
+ * Because we have set nbd->sock to NULL under the tx_lock, all
* modifications to the list must have completed by now. For
* the same reason, the active_req must be NULL.
*
* As a consequence, we don't need to take the spin lock while
* purging the list here.
*/
- BUG_ON(lo->sock);
- BUG_ON(lo->active_req);
+ BUG_ON(nbd->sock);
+ BUG_ON(nbd->active_req);
- while (!list_empty(&lo->queue_head)) {
- req = list_entry(lo->queue_head.next, struct request,
+ while (!list_empty(&nbd->queue_head)) {
+ req = list_entry(nbd->queue_head.next, struct request,
queuelist);
list_del_init(&req->queuelist);
req->errors++;
@@ -448,7 +448,7 @@ static void nbd_clear_que(struct nbd_device *lo)
}
-static void nbd_handle_req(struct nbd_device *lo, struct request *req)
+static void nbd_handle_req(struct nbd_device *nbd, struct request *req)
{
if (req->cmd_type != REQ_TYPE_FS)
goto error_out;
@@ -456,8 +456,8 @@ static void nbd_handle_req(struct nbd_device *lo, struct request *req)
nbd_cmd(req) = NBD_CMD_READ;
if (rq_data_dir(req) == WRITE) {
nbd_cmd(req) = NBD_CMD_WRITE;
- if (lo->flags & NBD_READ_ONLY) {
- dev_err(disk_to_dev(lo->disk),
+ if (nbd->flags & NBD_READ_ONLY) {
+ dev_err(disk_to_dev(nbd->disk),
"Write on read-only\n");
goto error_out;
}
@@ -465,29 +465,29 @@ static void nbd_handle_req(struct nbd_device *lo, struct request *req)
req->errors = 0;
- mutex_lock(&lo->tx_lock);
- if (unlikely(!lo->sock)) {
- mutex_unlock(&lo->tx_lock);
- dev_err(disk_to_dev(lo->disk),
+ mutex_lock(&nbd->tx_lock);
+ if (unlikely(!nbd->sock)) {
+ mutex_unlock(&nbd->tx_lock);
+ dev_err(disk_to_dev(nbd->disk),
"Attempted send on closed socket\n");
goto error_out;
}
- lo->active_req = req;
+ nbd->active_req = req;
- if (nbd_send_req(lo, req) != 0) {
- dev_err(disk_to_dev(lo->disk), "Request send failed\n");
+ if (nbd_send_req(nbd, req) != 0) {
+ dev_err(disk_to_dev(nbd->disk), "Request send failed\n");
req->errors++;
nbd_end_request(req);
} else {
- spin_lock(&lo->queue_lock);
- list_add(&req->queuelist, &lo->queue_head);
- spin_unlock(&lo->queue_lock);
+ spin_lock(&nbd->queue_lock);
+ list_add(&req->queuelist, &nbd->queue_head);
+ spin_unlock(&nbd->queue_lock);
}
- lo->active_req = NULL;
- mutex_unlock(&lo->tx_lock);
- wake_up_all(&lo->active_wq);
+ nbd->active_req = NULL;
+ mutex_unlock(&nbd->tx_lock);
+ wake_up_all(&nbd->active_wq);
return;
@@ -498,28 +498,28 @@ error_out:
static int nbd_thread(void *data)
{
- struct nbd_device *lo = data;
+ struct nbd_device *nbd = data;
struct request *req;
set_user_nice(current, -20);
- while (!kthread_should_stop() || !list_empty(&lo->waiting_queue)) {
+ while (!kthread_should_stop() || !list_empty(&nbd->waiting_queue)) {
/* wait for something to do */
- wait_event_interruptible(lo->waiting_wq,
+ wait_event_interruptible(nbd->waiting_wq,
kthread_should_stop() ||
- !list_empty(&lo->waiting_queue));
+ !list_empty(&nbd->waiting_queue));
/* extract request */
- if (list_empty(&lo->waiting_queue))
+ if (list_empty(&nbd->waiting_queue))
continue;
- spin_lock_irq(&lo->queue_lock);
- req = list_entry(lo->waiting_queue.next, struct request,
+ spin_lock_irq(&nbd->queue_lock);
+ req = list_entry(nbd->waiting_queue.next, struct request,
queuelist);
list_del_init(&req->queuelist);
- spin_unlock_irq(&lo->queue_lock);
+ spin_unlock_irq(&nbd->queue_lock);
/* handle request */
- nbd_handle_req(lo, req);
+ nbd_handle_req(nbd, req);
}
return 0;
}
@@ -527,7 +527,7 @@ static int nbd_thread(void *data)
/*
* We always wait for result of write, for now. It would be nice to make it optional
* in future
- * if ((rq_data_dir(req) == WRITE) && (lo->flags & NBD_WRITE_NOCHK))
+ * if ((rq_data_dir(req) == WRITE) && (nbd->flags & NBD_WRITE_NOCHK))
* { printk( "Warning: Ignoring result!\n"); nbd_end_request( req ); }
*/
@@ -536,19 +536,19 @@ static void do_nbd_request(struct request_queue *q)
struct request *req;
while ((req = blk_fetch_request(q)) != NULL) {
- struct nbd_device *lo;
+ struct nbd_device *nbd;
spin_unlock_irq(q->queue_lock);
dprintk(DBG_BLKDEV, "%s: request %p: dequeued (flags=%x)\n",
req->rq_disk->disk_name, req, req->cmd_type);
- lo = req->rq_disk->private_data;
+ nbd = req->rq_disk->private_data;
- BUG_ON(lo->magic != LO_MAGIC);
+ BUG_ON(nbd->magic != NBD_MAGIC);
- if (unlikely(!lo->sock)) {
- dev_err(disk_to_dev(lo->disk),
+ if (unlikely(!nbd->sock)) {
+ dev_err(disk_to_dev(nbd->disk),
"Attempted send on closed socket\n");
req->errors++;
nbd_end_request(req);
@@ -556,11 +556,11 @@ static void do_nbd_request(struct request_queue *q)
continue;
}
- spin_lock_irq(&lo->queue_lock);
- list_add_tail(&req->queuelist, &lo->waiting_queue);
- spin_unlock_irq(&lo->queue_lock);
+ spin_lock_irq(&nbd->queue_lock);
+ list_add_tail(&req->queuelist, &nbd->waiting_queue);
+ spin_unlock_irq(&nbd->queue_lock);
- wake_up(&lo->waiting_wq);
+ wake_up(&nbd->waiting_wq);
spin_lock_irq(q->queue_lock);
}
@@ -568,32 +568,32 @@ static void do_nbd_request(struct request_queue *q)
/* Must be called with tx_lock held */
-static int __nbd_ioctl(struct block_device *bdev, struct nbd_device *lo,
+static int __nbd_ioctl(struct block_device *bdev, struct nbd_device *nbd,
unsigned int cmd, unsigned long arg)
{
switch (cmd) {
case NBD_DISCONNECT: {
struct request sreq;
- dev_info(disk_to_dev(lo->disk), "NBD_DISCONNECT\n");
+ dev_info(disk_to_dev(nbd->disk), "NBD_DISCONNECT\n");
blk_rq_init(NULL, &sreq);
sreq.cmd_type = REQ_TYPE_SPECIAL;
nbd_cmd(&sreq) = NBD_CMD_DISC;
- if (!lo->sock)
+ if (!nbd->sock)
return -EINVAL;
- nbd_send_req(lo, &sreq);
+ nbd_send_req(nbd, &sreq);
return 0;
}
case NBD_CLEAR_SOCK: {
struct file *file;
- lo->sock = NULL;
- file = lo->file;
- lo->file = NULL;
- nbd_clear_que(lo);
- BUG_ON(!list_empty(&lo->queue_head));
+ nbd->sock = NULL;
+ file = nbd->file;
+ nbd->file = NULL;
+ nbd_clear_que(nbd);
+ BUG_ON(!list_empty(&nbd->queue_head));
if (file)
fput(file);
return 0;
@@ -601,14 +601,14 @@ static int __nbd_ioctl(struct block_device *bdev, struct nbd_device *lo,
case NBD_SET_SOCK: {
struct file *file;
- if (lo->file)
+ if (nbd->file)
return -EBUSY;
file = fget(arg);
if (file) {
struct inode *inode = file->f_path.dentry->d_inode;
if (S_ISSOCK(inode->i_mode)) {
- lo->file = file;
- lo->sock = SOCKET_I(inode);
+ nbd->file = file;
+ nbd->sock = SOCKET_I(inode);
if (max_part > 0)
bdev->bd_invalidated = 1;
return 0;
@@ -620,29 +620,29 @@ static int __nbd_ioctl(struct block_device *bdev, struct nbd_device *lo,
}
case NBD_SET_BLKSIZE:
- lo->blksize = arg;
- lo->bytesize &= ~(lo->blksize-1);
- bdev->bd_inode->i_size = lo->bytesize;
- set_blocksize(bdev, lo->blksize);
- set_capacity(lo->disk, lo->bytesize >> 9);
+ nbd->blksize = arg;
+ nbd->bytesize &= ~(nbd->blksize-1);
+ bdev->bd_inode->i_size = nbd->bytesize;
+ set_blocksize(bdev, nbd->blksize);
+ set_capacity(nbd->disk, nbd->bytesize >> 9);
return 0;
case NBD_SET_SIZE:
- lo->bytesize = arg & ~(lo->blksize-1);
- bdev->bd_inode->i_size = lo->bytesize;
- set_blocksize(bdev, lo->blksize);
- set_capacity(lo->disk, lo->bytesize >> 9);
+ nbd->bytesize = arg & ~(nbd->blksize-1);
+ bdev->bd_inode->i_size = nbd->bytesize;
+ set_blocksize(bdev, nbd->blksize);
+ set_capacity(nbd->disk, nbd->bytesize >> 9);
return 0;
case NBD_SET_TIMEOUT:
- lo->xmit_timeout = arg * HZ;
+ nbd->xmit_timeout = arg * HZ;
return 0;
case NBD_SET_SIZE_BLOCKS:
- lo->bytesize = ((u64) arg) * lo->blksize;
- bdev->bd_inode->i_size = lo->bytesize;
- set_blocksize(bdev, lo->blksize);
- set_capacity(lo->disk, lo->bytesize >> 9);
+ nbd->bytesize = ((u64) arg) * nbd->blksize;
+ bdev->bd_inode->i_size = nbd->bytesize;
+ set_blocksize(bdev, nbd->blksize);
+ set_capacity(nbd->disk, nbd->bytesize >> 9);
return 0;
case NBD_DO_IT: {
@@ -650,38 +650,38 @@ static int __nbd_ioctl(struct block_device *bdev, struct nbd_device *lo,
struct file *file;
int error;
- if (lo->pid)
+ if (nbd->pid)
return -EBUSY;
- if (!lo->file)
+ if (!nbd->file)
return -EINVAL;
- mutex_unlock(&lo->tx_lock);
+ mutex_unlock(&nbd->tx_lock);
- thread = kthread_create(nbd_thread, lo, lo->disk->disk_name);
+ thread = kthread_create(nbd_thread, nbd, nbd->disk->disk_name);
if (IS_ERR(thread)) {
- mutex_lock(&lo->tx_lock);
+ mutex_lock(&nbd->tx_lock);
return PTR_ERR(thread);
}
wake_up_process(thread);
- error = nbd_do_it(lo);
+ error = nbd_do_it(nbd);
kthread_stop(thread);
- mutex_lock(&lo->tx_lock);
+ mutex_lock(&nbd->tx_lock);
if (error)
return error;
- sock_shutdown(lo, 0);
- file = lo->file;
- lo->file = NULL;
- nbd_clear_que(lo);
- dev_warn(disk_to_dev(lo->disk), "queue cleared\n");
+ sock_shutdown(nbd, 0);
+ file = nbd->file;
+ nbd->file = NULL;
+ nbd_clear_que(nbd);
+ dev_warn(disk_to_dev(nbd->disk), "queue cleared\n");
if (file)
fput(file);
- lo->bytesize = 0;
+ nbd->bytesize = 0;
bdev->bd_inode->i_size = 0;
- set_capacity(lo->disk, 0);
+ set_capacity(nbd->disk, 0);
if (max_part > 0)
ioctl_by_bdev(bdev, BLKRRPART, 0);
- return lo->harderror;
+ return nbd->harderror;
}
case NBD_CLEAR_QUE:
@@ -689,14 +689,14 @@ static int __nbd_ioctl(struct block_device *bdev, struct nbd_device *lo,
* This is for compatibility only. The queue is always cleared
* by NBD_DO_IT or NBD_CLEAR_SOCK.
*/
- BUG_ON(!lo->sock && !list_empty(&lo->queue_head));
+ BUG_ON(!nbd->sock && !list_empty(&nbd->queue_head));
return 0;
case NBD_PRINT_DEBUG:
- dev_info(disk_to_dev(lo->disk),
+ dev_info(disk_to_dev(nbd->disk),
"next = %p, prev = %p, head = %p\n",
- lo->queue_head.next, lo->queue_head.prev,
- &lo->queue_head);
+ nbd->queue_head.next, nbd->queue_head.prev,
+ &nbd->queue_head);
return 0;
}
return -ENOTTY;
@@ -705,21 +705,21 @@ static int __nbd_ioctl(struct block_device *bdev, struct nbd_device *lo,
static int nbd_ioctl(struct block_device *bdev, fmode_t mode,
unsigned int cmd, unsigned long arg)
{
- struct nbd_device *lo = bdev->bd_disk->private_data;
+ struct nbd_device *nbd = bdev->bd_disk->private_data;
int error;
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
- BUG_ON(lo->magic != LO_MAGIC);
+ BUG_ON(nbd->magic != NBD_MAGIC);
/* Anyone capable of this syscall can do *real bad* things */
dprintk(DBG_IOCTL, "%s: nbd_ioctl cmd=%s(0x%x) arg=%lu\n",
- lo->disk->disk_name, ioctl_cmd_to_ascii(cmd), cmd, arg);
+ nbd->disk->disk_name, ioctl_cmd_to_ascii(cmd), cmd, arg);
- mutex_lock(&lo->tx_lock);
- error = __nbd_ioctl(bdev, lo, cmd, arg);
- mutex_unlock(&lo->tx_lock);
+ mutex_lock(&nbd->tx_lock);
+ error = __nbd_ioctl(bdev, nbd, cmd, arg);
+ mutex_unlock(&nbd->tx_lock);
return error;
}
@@ -805,7 +805,7 @@ static int __init nbd_init(void)
for (i = 0; i < nbds_max; i++) {
struct gendisk *disk = nbd_dev[i].disk;
nbd_dev[i].file = NULL;
- nbd_dev[i].magic = LO_MAGIC;
+ nbd_dev[i].magic = NBD_MAGIC;
nbd_dev[i].flags = 0;
INIT_LIST_HEAD(&nbd_dev[i].waiting_queue);
spin_lock_init(&nbd_dev[i].queue_lock);
diff --git a/drivers/block/nvme.c b/drivers/block/nvme.c
index 09776daf98e..38a2d063188 100644
--- a/drivers/block/nvme.c
+++ b/drivers/block/nvme.c
@@ -40,6 +40,8 @@
#include <linux/slab.h>
#include <linux/types.h>
+#include <asm-generic/io-64-nonatomic-lo-hi.h>
+
#define NVME_Q_DEPTH 1024
#define SQ_SIZE(depth) (depth * sizeof(struct nvme_command))
#define CQ_SIZE(depth) (depth * sizeof(struct nvme_completion))
diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c
index d59edeabd93..ba66e4445f4 100644
--- a/drivers/block/pktcdvd.c
+++ b/drivers/block/pktcdvd.c
@@ -987,14 +987,14 @@ static void pkt_copy_bio_data(struct bio *src_bio, int seg, int offs, struct pag
while (copy_size > 0) {
struct bio_vec *src_bvl = bio_iovec_idx(src_bio, seg);
- void *vfrom = kmap_atomic(src_bvl->bv_page, KM_USER0) +
+ void *vfrom = kmap_atomic(src_bvl->bv_page) +
src_bvl->bv_offset + offs;
void *vto = page_address(dst_page) + dst_offs;
int len = min_t(int, copy_size, src_bvl->bv_len - offs);
BUG_ON(len < 0);
memcpy(vto, vfrom, len);
- kunmap_atomic(vfrom, KM_USER0);
+ kunmap_atomic(vfrom);
seg++;
offs = 0;
@@ -1019,10 +1019,10 @@ static void pkt_make_local_copy(struct packet_data *pkt, struct bio_vec *bvec)
offs = 0;
for (f = 0; f < pkt->frames; f++) {
if (bvec[f].bv_page != pkt->pages[p]) {
- void *vfrom = kmap_atomic(bvec[f].bv_page, KM_USER0) + bvec[f].bv_offset;
+ void *vfrom = kmap_atomic(bvec[f].bv_page) + bvec[f].bv_offset;
void *vto = page_address(pkt->pages[p]) + offs;
memcpy(vto, vfrom, CD_FRAMESIZE);
- kunmap_atomic(vfrom, KM_USER0);
+ kunmap_atomic(vfrom);
bvec[f].bv_page = pkt->pages[p];
bvec[f].bv_offset = offs;
} else {
diff --git a/drivers/block/rbd.c b/drivers/block/rbd.c
index a6278e7e61a..013c7a549fb 100644
--- a/drivers/block/rbd.c
+++ b/drivers/block/rbd.c
@@ -41,19 +41,35 @@
#include "rbd_types.h"
-#define DRV_NAME "rbd"
-#define DRV_NAME_LONG "rbd (rados block device)"
+/*
+ * The basic unit of block I/O is a sector. It is interpreted in a
+ * number of contexts in Linux (blk, bio, genhd), but the default is
+ * universally 512 bytes. These symbols are just slightly more
+ * meaningful than the bare numbers they represent.
+ */
+#define SECTOR_SHIFT 9
+#define SECTOR_SIZE (1ULL << SECTOR_SHIFT)
+
+#define RBD_DRV_NAME "rbd"
+#define RBD_DRV_NAME_LONG "rbd (rados block device)"
#define RBD_MINORS_PER_MAJOR 256 /* max minors per blkdev */
-#define RBD_MAX_MD_NAME_LEN (96 + sizeof(RBD_SUFFIX))
+#define RBD_MAX_MD_NAME_LEN (RBD_MAX_OBJ_NAME_LEN + sizeof(RBD_SUFFIX))
#define RBD_MAX_POOL_NAME_LEN 64
#define RBD_MAX_SNAP_NAME_LEN 32
#define RBD_MAX_OPT_LEN 1024
#define RBD_SNAP_HEAD_NAME "-"
+/*
+ * An RBD device name will be "rbd#", where the "rbd" comes from
+ * RBD_DRV_NAME above, and # is a unique integer identifier.
+ * MAX_INT_FORMAT_WIDTH is used in ensuring DEV_NAME_LEN is big
+ * enough to hold all possible device names.
+ */
#define DEV_NAME_LEN 32
+#define MAX_INT_FORMAT_WIDTH ((5 * sizeof (int)) / 2 + 1)
#define RBD_NOTIFY_TIMEOUT_DEFAULT 10
@@ -66,7 +82,6 @@ struct rbd_image_header {
__u8 obj_order;
__u8 crypt_type;
__u8 comp_type;
- struct rw_semaphore snap_rwsem;
struct ceph_snap_context *snapc;
size_t snap_names_len;
u64 snap_seq;
@@ -83,7 +98,7 @@ struct rbd_options {
};
/*
- * an instance of the client. multiple devices may share a client.
+ * an instance of the client. multiple devices may share an rbd client.
*/
struct rbd_client {
struct ceph_client *client;
@@ -92,20 +107,9 @@ struct rbd_client {
struct list_head node;
};
-struct rbd_req_coll;
-
/*
- * a single io request
+ * a request completion status
*/
-struct rbd_request {
- struct request *rq; /* blk layer request */
- struct bio *bio; /* cloned bio */
- struct page **pages; /* list of used pages */
- u64 len;
- int coll_index;
- struct rbd_req_coll *coll;
-};
-
struct rbd_req_status {
int done;
int rc;
@@ -122,6 +126,18 @@ struct rbd_req_coll {
struct rbd_req_status status[0];
};
+/*
+ * a single io request
+ */
+struct rbd_request {
+ struct request *rq; /* blk layer request */
+ struct bio *bio; /* cloned bio */
+ struct page **pages; /* list of used pages */
+ u64 len;
+ int coll_index;
+ struct rbd_req_coll *coll;
+};
+
struct rbd_snap {
struct device dev;
const char *name;
@@ -140,7 +156,6 @@ struct rbd_device {
struct gendisk *disk; /* blkdev's gendisk and rq */
struct request_queue *q;
- struct ceph_client *client;
struct rbd_client *rbd_client;
char name[DEV_NAME_LEN]; /* blkdev name, e.g. rbd3 */
@@ -157,6 +172,8 @@ struct rbd_device {
struct ceph_osd_event *watch_event;
struct ceph_osd_request *watch_request;
+ /* protects updating the header */
+ struct rw_semaphore header_rwsem;
char snap_name[RBD_MAX_SNAP_NAME_LEN];
u32 cur_snap; /* index+1 of current snapshot within snap context
0 - for the head */
@@ -171,15 +188,13 @@ struct rbd_device {
struct device dev;
};
-static struct bus_type rbd_bus_type = {
- .name = "rbd",
-};
-
-static spinlock_t node_lock; /* protects client get/put */
-
static DEFINE_MUTEX(ctl_mutex); /* Serialize open/close/setup/teardown */
+
static LIST_HEAD(rbd_dev_list); /* devices */
-static LIST_HEAD(rbd_client_list); /* clients */
+static DEFINE_SPINLOCK(rbd_dev_list_lock);
+
+static LIST_HEAD(rbd_client_list); /* clients */
+static DEFINE_SPINLOCK(rbd_client_list_lock);
static int __rbd_init_snaps_header(struct rbd_device *rbd_dev);
static void rbd_dev_release(struct device *dev);
@@ -190,12 +205,32 @@ static ssize_t rbd_snap_add(struct device *dev,
static void __rbd_remove_snap_dev(struct rbd_device *rbd_dev,
struct rbd_snap *snap);
+static ssize_t rbd_add(struct bus_type *bus, const char *buf,
+ size_t count);
+static ssize_t rbd_remove(struct bus_type *bus, const char *buf,
+ size_t count);
-static struct rbd_device *dev_to_rbd(struct device *dev)
+static struct bus_attribute rbd_bus_attrs[] = {
+ __ATTR(add, S_IWUSR, NULL, rbd_add),
+ __ATTR(remove, S_IWUSR, NULL, rbd_remove),
+ __ATTR_NULL
+};
+
+static struct bus_type rbd_bus_type = {
+ .name = "rbd",
+ .bus_attrs = rbd_bus_attrs,
+};
+
+static void rbd_root_dev_release(struct device *dev)
{
- return container_of(dev, struct rbd_device, dev);
}
+static struct device rbd_root_dev = {
+ .init_name = "rbd",
+ .release = rbd_root_dev_release,
+};
+
+
static struct device *rbd_get_dev(struct rbd_device *rbd_dev)
{
return get_device(&rbd_dev->dev);
@@ -210,8 +245,7 @@ static int __rbd_update_snaps(struct rbd_device *rbd_dev);
static int rbd_open(struct block_device *bdev, fmode_t mode)
{
- struct gendisk *disk = bdev->bd_disk;
- struct rbd_device *rbd_dev = disk->private_data;
+ struct rbd_device *rbd_dev = bdev->bd_disk->private_data;
rbd_get_dev(rbd_dev);
@@ -256,9 +290,11 @@ static struct rbd_client *rbd_client_create(struct ceph_options *opt,
kref_init(&rbdc->kref);
INIT_LIST_HEAD(&rbdc->node);
+ mutex_lock_nested(&ctl_mutex, SINGLE_DEPTH_NESTING);
+
rbdc->client = ceph_create_client(opt, rbdc, 0, 0);
if (IS_ERR(rbdc->client))
- goto out_rbdc;
+ goto out_mutex;
opt = NULL; /* Now rbdc->client is responsible for opt */
ret = ceph_open_session(rbdc->client);
@@ -267,16 +303,19 @@ static struct rbd_client *rbd_client_create(struct ceph_options *opt,
rbdc->rbd_opts = rbd_opts;
- spin_lock(&node_lock);
+ spin_lock(&rbd_client_list_lock);
list_add_tail(&rbdc->node, &rbd_client_list);
- spin_unlock(&node_lock);
+ spin_unlock(&rbd_client_list_lock);
+
+ mutex_unlock(&ctl_mutex);
dout("rbd_client_create created %p\n", rbdc);
return rbdc;
out_err:
ceph_destroy_client(rbdc->client);
-out_rbdc:
+out_mutex:
+ mutex_unlock(&ctl_mutex);
kfree(rbdc);
out_opt:
if (opt)
@@ -324,7 +363,7 @@ static int parse_rbd_opts_token(char *c, void *private)
substring_t argstr[MAX_OPT_ARGS];
int token, intval, ret;
- token = match_token((char *)c, rbdopt_tokens, argstr);
+ token = match_token(c, rbdopt_tokens, argstr);
if (token < 0)
return -EINVAL;
@@ -357,58 +396,54 @@ static int parse_rbd_opts_token(char *c, void *private)
* Get a ceph client with specific addr and configuration, if one does
* not exist create it.
*/
-static int rbd_get_client(struct rbd_device *rbd_dev, const char *mon_addr,
- char *options)
+static struct rbd_client *rbd_get_client(const char *mon_addr,
+ size_t mon_addr_len,
+ char *options)
{
struct rbd_client *rbdc;
struct ceph_options *opt;
- int ret;
struct rbd_options *rbd_opts;
rbd_opts = kzalloc(sizeof(*rbd_opts), GFP_KERNEL);
if (!rbd_opts)
- return -ENOMEM;
+ return ERR_PTR(-ENOMEM);
rbd_opts->notify_timeout = RBD_NOTIFY_TIMEOUT_DEFAULT;
- ret = ceph_parse_options(&opt, options, mon_addr,
- mon_addr + strlen(mon_addr), parse_rbd_opts_token, rbd_opts);
- if (ret < 0)
- goto done_err;
+ opt = ceph_parse_options(options, mon_addr,
+ mon_addr + mon_addr_len,
+ parse_rbd_opts_token, rbd_opts);
+ if (IS_ERR(opt)) {
+ kfree(rbd_opts);
+ return ERR_CAST(opt);
+ }
- spin_lock(&node_lock);
+ spin_lock(&rbd_client_list_lock);
rbdc = __rbd_client_find(opt);
if (rbdc) {
+ /* using an existing client */
+ kref_get(&rbdc->kref);
+ spin_unlock(&rbd_client_list_lock);
+
ceph_destroy_options(opt);
kfree(rbd_opts);
- /* using an existing client */
- kref_get(&rbdc->kref);
- rbd_dev->rbd_client = rbdc;
- rbd_dev->client = rbdc->client;
- spin_unlock(&node_lock);
- return 0;
+ return rbdc;
}
- spin_unlock(&node_lock);
+ spin_unlock(&rbd_client_list_lock);
rbdc = rbd_client_create(opt, rbd_opts);
- if (IS_ERR(rbdc)) {
- ret = PTR_ERR(rbdc);
- goto done_err;
- }
- rbd_dev->rbd_client = rbdc;
- rbd_dev->client = rbdc->client;
- return 0;
-done_err:
- kfree(rbd_opts);
- return ret;
+ if (IS_ERR(rbdc))
+ kfree(rbd_opts);
+
+ return rbdc;
}
/*
* Destroy ceph client
*
- * Caller must hold node_lock.
+ * Caller must hold rbd_client_list_lock.
*/
static void rbd_client_release(struct kref *kref)
{
@@ -428,11 +463,10 @@ static void rbd_client_release(struct kref *kref)
*/
static void rbd_put_client(struct rbd_device *rbd_dev)
{
- spin_lock(&node_lock);
+ spin_lock(&rbd_client_list_lock);
kref_put(&rbd_dev->rbd_client->kref, rbd_client_release);
- spin_unlock(&node_lock);
+ spin_unlock(&rbd_client_list_lock);
rbd_dev->rbd_client = NULL;
- rbd_dev->client = NULL;
}
/*
@@ -457,21 +491,19 @@ static int rbd_header_from_disk(struct rbd_image_header *header,
gfp_t gfp_flags)
{
int i;
- u32 snap_count = le32_to_cpu(ondisk->snap_count);
- int ret = -ENOMEM;
+ u32 snap_count;
- if (memcmp(ondisk, RBD_HEADER_TEXT, sizeof(RBD_HEADER_TEXT))) {
+ if (memcmp(ondisk, RBD_HEADER_TEXT, sizeof(RBD_HEADER_TEXT)))
return -ENXIO;
- }
- init_rwsem(&header->snap_rwsem);
- header->snap_names_len = le64_to_cpu(ondisk->snap_names_len);
+ snap_count = le32_to_cpu(ondisk->snap_count);
header->snapc = kmalloc(sizeof(struct ceph_snap_context) +
- snap_count *
- sizeof(struct rbd_image_snap_ondisk),
+ snap_count * sizeof (*ondisk),
gfp_flags);
if (!header->snapc)
return -ENOMEM;
+
+ header->snap_names_len = le64_to_cpu(ondisk->snap_names_len);
if (snap_count) {
header->snap_names = kmalloc(header->snap_names_len,
GFP_KERNEL);
@@ -498,8 +530,7 @@ static int rbd_header_from_disk(struct rbd_image_header *header,
header->snapc->num_snaps = snap_count;
header->total_snaps = snap_count;
- if (snap_count &&
- allocated_snaps == snap_count) {
+ if (snap_count && allocated_snaps == snap_count) {
for (i = 0; i < snap_count; i++) {
header->snapc->snaps[i] =
le64_to_cpu(ondisk->snaps[i].id);
@@ -518,7 +549,7 @@ err_names:
kfree(header->snap_names);
err_snapc:
kfree(header->snapc);
- return ret;
+ return -ENOMEM;
}
static int snap_index(struct rbd_image_header *header, int snap_num)
@@ -542,35 +573,34 @@ static int snap_by_name(struct rbd_image_header *header, const char *snap_name,
int i;
char *p = header->snap_names;
- for (i = 0; i < header->total_snaps; i++, p += strlen(p) + 1) {
- if (strcmp(snap_name, p) == 0)
- break;
- }
- if (i == header->total_snaps)
- return -ENOENT;
- if (seq)
- *seq = header->snapc->snaps[i];
+ for (i = 0; i < header->total_snaps; i++) {
+ if (!strcmp(snap_name, p)) {
- if (size)
- *size = header->snap_sizes[i];
+ /* Found it. Pass back its id and/or size */
- return i;
+ if (seq)
+ *seq = header->snapc->snaps[i];
+ if (size)
+ *size = header->snap_sizes[i];
+ return i;
+ }
+ p += strlen(p) + 1; /* Skip ahead to the next name */
+ }
+ return -ENOENT;
}
-static int rbd_header_set_snap(struct rbd_device *dev,
- const char *snap_name,
- u64 *size)
+static int rbd_header_set_snap(struct rbd_device *dev, u64 *size)
{
struct rbd_image_header *header = &dev->header;
struct ceph_snap_context *snapc = header->snapc;
int ret = -ENOENT;
- down_write(&header->snap_rwsem);
+ BUILD_BUG_ON(sizeof (dev->snap_name) < sizeof (RBD_SNAP_HEAD_NAME));
- if (!snap_name ||
- !*snap_name ||
- strcmp(snap_name, "-") == 0 ||
- strcmp(snap_name, RBD_SNAP_HEAD_NAME) == 0) {
+ down_write(&dev->header_rwsem);
+
+ if (!memcmp(dev->snap_name, RBD_SNAP_HEAD_NAME,
+ sizeof (RBD_SNAP_HEAD_NAME))) {
if (header->total_snaps)
snapc->seq = header->snap_seq;
else
@@ -580,7 +610,7 @@ static int rbd_header_set_snap(struct rbd_device *dev,
if (size)
*size = header->image_size;
} else {
- ret = snap_by_name(header, snap_name, &snapc->seq, size);
+ ret = snap_by_name(header, dev->snap_name, &snapc->seq, size);
if (ret < 0)
goto done;
@@ -590,7 +620,7 @@ static int rbd_header_set_snap(struct rbd_device *dev,
ret = 0;
done:
- up_write(&header->snap_rwsem);
+ up_write(&dev->header_rwsem);
return ret;
}
@@ -717,7 +747,7 @@ static struct bio *bio_chain_clone(struct bio **old, struct bio **next,
/* split the bio. We'll release it either in the next
call, or it will have to be released outside */
- bp = bio_split(old_chain, (len - total) / 512ULL);
+ bp = bio_split(old_chain, (len - total) / SECTOR_SIZE);
if (!bp)
goto err_out;
@@ -857,7 +887,7 @@ static int rbd_do_request(struct request *rq,
struct timespec mtime = CURRENT_TIME;
struct rbd_request *req_data;
struct ceph_osd_request_head *reqhead;
- struct rbd_image_header *header = &dev->header;
+ struct ceph_osd_client *osdc;
req_data = kzalloc(sizeof(*req_data), GFP_NOIO);
if (!req_data) {
@@ -874,15 +904,13 @@ static int rbd_do_request(struct request *rq,
dout("rbd_do_request obj=%s ofs=%lld len=%lld\n", obj, len, ofs);
- down_read(&header->snap_rwsem);
+ down_read(&dev->header_rwsem);
- req = ceph_osdc_alloc_request(&dev->client->osdc, flags,
- snapc,
- ops,
- false,
- GFP_NOIO, pages, bio);
+ osdc = &dev->rbd_client->client->osdc;
+ req = ceph_osdc_alloc_request(osdc, flags, snapc, ops,
+ false, GFP_NOIO, pages, bio);
if (!req) {
- up_read(&header->snap_rwsem);
+ up_read(&dev->header_rwsem);
ret = -ENOMEM;
goto done_pages;
}
@@ -909,27 +937,27 @@ static int rbd_do_request(struct request *rq,
layout->fl_object_size = cpu_to_le32(1 << RBD_MAX_OBJ_ORDER);
layout->fl_pg_preferred = cpu_to_le32(-1);
layout->fl_pg_pool = cpu_to_le32(dev->poolid);
- ceph_calc_raw_layout(&dev->client->osdc, layout, snapid,
- ofs, &len, &bno, req, ops);
+ ceph_calc_raw_layout(osdc, layout, snapid, ofs, &len, &bno,
+ req, ops);
ceph_osdc_build_request(req, ofs, &len,
ops,
snapc,
&mtime,
req->r_oid, req->r_oid_len);
- up_read(&header->snap_rwsem);
+ up_read(&dev->header_rwsem);
if (linger_req) {
- ceph_osdc_set_request_linger(&dev->client->osdc, req);
+ ceph_osdc_set_request_linger(osdc, req);
*linger_req = req;
}
- ret = ceph_osdc_start_request(&dev->client->osdc, req, false);
+ ret = ceph_osdc_start_request(osdc, req, false);
if (ret < 0)
goto done_err;
if (!rbd_cb) {
- ret = ceph_osdc_wait_request(&dev->client->osdc, req);
+ ret = ceph_osdc_wait_request(osdc, req);
if (ver)
*ver = le64_to_cpu(req->r_reassert_version.version);
dout("reassert_ver=%lld\n",
@@ -1213,8 +1241,8 @@ static void rbd_watch_cb(u64 ver, u64 notify_id, u8 opcode, void *data)
rc = __rbd_update_snaps(dev);
mutex_unlock(&ctl_mutex);
if (rc)
- pr_warning(DRV_NAME "%d got notification but failed to update"
- " snaps: %d\n", dev->major, rc);
+ pr_warning(RBD_DRV_NAME "%d got notification but failed to "
+ " update snaps: %d\n", dev->major, rc);
rbd_req_sync_notify_ack(dev, ver, notify_id, dev->obj_md_name);
}
@@ -1227,7 +1255,7 @@ static int rbd_req_sync_watch(struct rbd_device *dev,
u64 ver)
{
struct ceph_osd_req_op *ops;
- struct ceph_osd_client *osdc = &dev->client->osdc;
+ struct ceph_osd_client *osdc = &dev->rbd_client->client->osdc;
int ret = rbd_create_rw_ops(&ops, 1, CEPH_OSD_OP_WATCH, 0);
if (ret < 0)
@@ -1314,7 +1342,7 @@ static int rbd_req_sync_notify(struct rbd_device *dev,
const char *obj)
{
struct ceph_osd_req_op *ops;
- struct ceph_osd_client *osdc = &dev->client->osdc;
+ struct ceph_osd_client *osdc = &dev->rbd_client->client->osdc;
struct ceph_osd_event *event;
struct rbd_notify_info info;
int payload_len = sizeof(u32) + sizeof(u32);
@@ -1421,9 +1449,7 @@ static void rbd_rq_fn(struct request_queue *q)
struct request *rq;
struct bio_pair *bp = NULL;
- rq = blk_fetch_request(q);
-
- while (1) {
+ while ((rq = blk_fetch_request(q))) {
struct bio *bio;
struct bio *rq_bio, *next_bio = NULL;
bool do_write;
@@ -1441,32 +1467,32 @@ static void rbd_rq_fn(struct request_queue *q)
/* filter out block requests we don't understand */
if ((rq->cmd_type != REQ_TYPE_FS)) {
__blk_end_request_all(rq, 0);
- goto next;
+ continue;
}
/* deduce our operation (read, write) */
do_write = (rq_data_dir(rq) == WRITE);
size = blk_rq_bytes(rq);
- ofs = blk_rq_pos(rq) * 512ULL;
+ ofs = blk_rq_pos(rq) * SECTOR_SIZE;
rq_bio = rq->bio;
if (do_write && rbd_dev->read_only) {
__blk_end_request_all(rq, -EROFS);
- goto next;
+ continue;
}
spin_unlock_irq(q->queue_lock);
dout("%s 0x%x bytes at 0x%llx\n",
do_write ? "write" : "read",
- size, blk_rq_pos(rq) * 512ULL);
+ size, blk_rq_pos(rq) * SECTOR_SIZE);
num_segs = rbd_get_num_segments(&rbd_dev->header, ofs, size);
coll = rbd_alloc_coll(num_segs);
if (!coll) {
spin_lock_irq(q->queue_lock);
__blk_end_request_all(rq, -ENOMEM);
- goto next;
+ continue;
}
do {
@@ -1512,8 +1538,6 @@ next_seg:
if (bp)
bio_pair_release(bp);
spin_lock_irq(q->queue_lock);
-next:
- rq = blk_fetch_request(q);
}
}
@@ -1526,13 +1550,17 @@ static int rbd_merge_bvec(struct request_queue *q, struct bvec_merge_data *bmd,
struct bio_vec *bvec)
{
struct rbd_device *rbd_dev = q->queuedata;
- unsigned int chunk_sectors = 1 << (rbd_dev->header.obj_order - 9);
- sector_t sector = bmd->bi_sector + get_start_sect(bmd->bi_bdev);
- unsigned int bio_sectors = bmd->bi_size >> 9;
+ unsigned int chunk_sectors;
+ sector_t sector;
+ unsigned int bio_sectors;
int max;
+ chunk_sectors = 1 << (rbd_dev->header.obj_order - SECTOR_SHIFT);
+ sector = bmd->bi_sector + get_start_sect(bmd->bi_bdev);
+ bio_sectors = bmd->bi_size >> SECTOR_SHIFT;
+
max = (chunk_sectors - ((sector & (chunk_sectors - 1))
- + bio_sectors)) << 9;
+ + bio_sectors)) << SECTOR_SHIFT;
if (max < 0)
max = 0; /* bio_add cannot handle a negative return */
if (max <= bvec->bv_len && bio_sectors == 0)
@@ -1565,15 +1593,16 @@ static int rbd_read_header(struct rbd_device *rbd_dev,
ssize_t rc;
struct rbd_image_header_ondisk *dh;
int snap_count = 0;
- u64 snap_names_len = 0;
u64 ver;
+ size_t len;
+ /*
+ * First reads the fixed-size header to determine the number
+ * of snapshots, then re-reads it, along with all snapshot
+ * records as well as their stored names.
+ */
+ len = sizeof (*dh);
while (1) {
- int len = sizeof(*dh) +
- snap_count * sizeof(struct rbd_image_snap_ondisk) +
- snap_names_len;
-
- rc = -ENOMEM;
dh = kmalloc(len, GFP_KERNEL);
if (!dh)
return -ENOMEM;
@@ -1588,21 +1617,22 @@ static int rbd_read_header(struct rbd_device *rbd_dev,
rc = rbd_header_from_disk(header, dh, snap_count, GFP_KERNEL);
if (rc < 0) {
- if (rc == -ENXIO) {
+ if (rc == -ENXIO)
pr_warning("unrecognized header format"
" for image %s", rbd_dev->obj);
- }
goto out_dh;
}
- if (snap_count != header->total_snaps) {
- snap_count = header->total_snaps;
- snap_names_len = header->snap_names_len;
- rbd_header_free(header);
- kfree(dh);
- continue;
- }
- break;
+ if (snap_count == header->total_snaps)
+ break;
+
+ snap_count = header->total_snaps;
+ len = sizeof (*dh) +
+ snap_count * sizeof(struct rbd_image_snap_ondisk) +
+ header->snap_names_len;
+
+ rbd_header_free(header);
+ kfree(dh);
}
header->obj_version = ver;
@@ -1623,13 +1653,14 @@ static int rbd_header_add_snap(struct rbd_device *dev,
int ret;
void *data, *p, *e;
u64 ver;
+ struct ceph_mon_client *monc;
/* we should create a snapshot only if we're pointing at the head */
if (dev->cur_snap)
return -EINVAL;
- ret = ceph_monc_create_snapid(&dev->client->monc, dev->poolid,
- &new_snapid);
+ monc = &dev->rbd_client->client->monc;
+ ret = ceph_monc_create_snapid(monc, dev->poolid, &new_snapid);
dout("created snapid=%lld\n", new_snapid);
if (ret < 0)
return ret;
@@ -1684,9 +1715,9 @@ static int __rbd_update_snaps(struct rbd_device *rbd_dev)
return ret;
/* resized? */
- set_capacity(rbd_dev->disk, h.image_size / 512ULL);
+ set_capacity(rbd_dev->disk, h.image_size / SECTOR_SIZE);
- down_write(&rbd_dev->header.snap_rwsem);
+ down_write(&rbd_dev->header_rwsem);
snap_seq = rbd_dev->header.snapc->seq;
if (rbd_dev->header.total_snaps &&
@@ -1711,7 +1742,7 @@ static int __rbd_update_snaps(struct rbd_device *rbd_dev)
ret = __rbd_init_snaps_header(rbd_dev);
- up_write(&rbd_dev->header.snap_rwsem);
+ up_write(&rbd_dev->header_rwsem);
return ret;
}
@@ -1721,6 +1752,7 @@ static int rbd_init_disk(struct rbd_device *rbd_dev)
struct gendisk *disk;
struct request_queue *q;
int rc;
+ u64 segment_size;
u64 total_size = 0;
/* contact OSD, request size info about the object being mapped */
@@ -1733,7 +1765,7 @@ static int rbd_init_disk(struct rbd_device *rbd_dev)
if (rc)
return rc;
- rc = rbd_header_set_snap(rbd_dev, rbd_dev->snap_name, &total_size);
+ rc = rbd_header_set_snap(rbd_dev, &total_size);
if (rc)
return rc;
@@ -1743,7 +1775,7 @@ static int rbd_init_disk(struct rbd_device *rbd_dev)
if (!disk)
goto out;
- snprintf(disk->disk_name, sizeof(disk->disk_name), DRV_NAME "%d",
+ snprintf(disk->disk_name, sizeof(disk->disk_name), RBD_DRV_NAME "%d",
rbd_dev->id);
disk->major = rbd_dev->major;
disk->first_minor = 0;
@@ -1756,11 +1788,15 @@ static int rbd_init_disk(struct rbd_device *rbd_dev)
if (!q)
goto out_disk;
+ /* We use the default size, but let's be explicit about it. */
+ blk_queue_physical_block_size(q, SECTOR_SIZE);
+
/* set io sizes to object size */
- blk_queue_max_hw_sectors(q, rbd_obj_bytes(&rbd_dev->header) / 512ULL);
- blk_queue_max_segment_size(q, rbd_obj_bytes(&rbd_dev->header));
- blk_queue_io_min(q, rbd_obj_bytes(&rbd_dev->header));
- blk_queue_io_opt(q, rbd_obj_bytes(&rbd_dev->header));
+ segment_size = rbd_obj_bytes(&rbd_dev->header);
+ blk_queue_max_hw_sectors(q, segment_size / SECTOR_SIZE);
+ blk_queue_max_segment_size(q, segment_size);
+ blk_queue_io_min(q, segment_size);
+ blk_queue_io_opt(q, segment_size);
blk_queue_merge_bvec(q, rbd_merge_bvec);
disk->queue = q;
@@ -1771,7 +1807,7 @@ static int rbd_init_disk(struct rbd_device *rbd_dev)
rbd_dev->q = q;
/* finally, announce the disk to the world */
- set_capacity(disk, total_size / 512ULL);
+ set_capacity(disk, total_size / SECTOR_SIZE);
add_disk(disk);
pr_info("%s: added with size 0x%llx\n",
@@ -1788,10 +1824,15 @@ out:
sysfs
*/
+static struct rbd_device *dev_to_rbd_dev(struct device *dev)
+{
+ return container_of(dev, struct rbd_device, dev);
+}
+
static ssize_t rbd_size_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
- struct rbd_device *rbd_dev = dev_to_rbd(dev);
+ struct rbd_device *rbd_dev = dev_to_rbd_dev(dev);
return sprintf(buf, "%llu\n", (unsigned long long)rbd_dev->header.image_size);
}
@@ -1799,7 +1840,7 @@ static ssize_t rbd_size_show(struct device *dev,
static ssize_t rbd_major_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
- struct rbd_device *rbd_dev = dev_to_rbd(dev);
+ struct rbd_device *rbd_dev = dev_to_rbd_dev(dev);
return sprintf(buf, "%d\n", rbd_dev->major);
}
@@ -1807,15 +1848,16 @@ static ssize_t rbd_major_show(struct device *dev,
static ssize_t rbd_client_id_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
- struct rbd_device *rbd_dev = dev_to_rbd(dev);
+ struct rbd_device *rbd_dev = dev_to_rbd_dev(dev);
- return sprintf(buf, "client%lld\n", ceph_client_id(rbd_dev->client));
+ return sprintf(buf, "client%lld\n",
+ ceph_client_id(rbd_dev->rbd_client->client));
}
static ssize_t rbd_pool_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
- struct rbd_device *rbd_dev = dev_to_rbd(dev);
+ struct rbd_device *rbd_dev = dev_to_rbd_dev(dev);
return sprintf(buf, "%s\n", rbd_dev->pool_name);
}
@@ -1823,7 +1865,7 @@ static ssize_t rbd_pool_show(struct device *dev,
static ssize_t rbd_name_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
- struct rbd_device *rbd_dev = dev_to_rbd(dev);
+ struct rbd_device *rbd_dev = dev_to_rbd_dev(dev);
return sprintf(buf, "%s\n", rbd_dev->obj);
}
@@ -1832,7 +1874,7 @@ static ssize_t rbd_snap_show(struct device *dev,
struct device_attribute *attr,
char *buf)
{
- struct rbd_device *rbd_dev = dev_to_rbd(dev);
+ struct rbd_device *rbd_dev = dev_to_rbd_dev(dev);
return sprintf(buf, "%s\n", rbd_dev->snap_name);
}
@@ -1842,7 +1884,7 @@ static ssize_t rbd_image_refresh(struct device *dev,
const char *buf,
size_t size)
{
- struct rbd_device *rbd_dev = dev_to_rbd(dev);
+ struct rbd_device *rbd_dev = dev_to_rbd_dev(dev);
int rc;
int ret = size;
@@ -1907,7 +1949,7 @@ static ssize_t rbd_snap_size_show(struct device *dev,
{
struct rbd_snap *snap = container_of(dev, struct rbd_snap, dev);
- return sprintf(buf, "%lld\n", (long long)snap->size);
+ return sprintf(buf, "%zd\n", snap->size);
}
static ssize_t rbd_snap_id_show(struct device *dev,
@@ -1916,7 +1958,7 @@ static ssize_t rbd_snap_id_show(struct device *dev,
{
struct rbd_snap *snap = container_of(dev, struct rbd_snap, dev);
- return sprintf(buf, "%lld\n", (long long)snap->id);
+ return sprintf(buf, "%llu\n", (unsigned long long) snap->id);
}
static DEVICE_ATTR(snap_size, S_IRUGO, rbd_snap_size_show, NULL);
@@ -2088,19 +2130,9 @@ static int __rbd_init_snaps_header(struct rbd_device *rbd_dev)
return 0;
}
-
-static void rbd_root_dev_release(struct device *dev)
-{
-}
-
-static struct device rbd_root_dev = {
- .init_name = "rbd",
- .release = rbd_root_dev_release,
-};
-
static int rbd_bus_add_dev(struct rbd_device *rbd_dev)
{
- int ret = -ENOMEM;
+ int ret;
struct device *dev;
struct rbd_snap *snap;
@@ -2114,7 +2146,7 @@ static int rbd_bus_add_dev(struct rbd_device *rbd_dev)
dev_set_name(dev, "%d", rbd_dev->id);
ret = device_register(dev);
if (ret < 0)
- goto done_free;
+ goto out;
list_for_each_entry(snap, &rbd_dev->snaps, node) {
ret = rbd_register_snap_dev(rbd_dev, snap,
@@ -2122,10 +2154,7 @@ static int rbd_bus_add_dev(struct rbd_device *rbd_dev)
if (ret < 0)
break;
}
-
- mutex_unlock(&ctl_mutex);
- return 0;
-done_free:
+out:
mutex_unlock(&ctl_mutex);
return ret;
}
@@ -2154,104 +2183,250 @@ static int rbd_init_watch_dev(struct rbd_device *rbd_dev)
return ret;
}
+static atomic64_t rbd_id_max = ATOMIC64_INIT(0);
+
+/*
+ * Get a unique rbd identifier for the given new rbd_dev, and add
+ * the rbd_dev to the global list. The minimum rbd id is 1.
+ */
+static void rbd_id_get(struct rbd_device *rbd_dev)
+{
+ rbd_dev->id = atomic64_inc_return(&rbd_id_max);
+
+ spin_lock(&rbd_dev_list_lock);
+ list_add_tail(&rbd_dev->node, &rbd_dev_list);
+ spin_unlock(&rbd_dev_list_lock);
+}
+
+/*
+ * Remove an rbd_dev from the global list, and record that its
+ * identifier is no longer in use.
+ */
+static void rbd_id_put(struct rbd_device *rbd_dev)
+{
+ struct list_head *tmp;
+ int rbd_id = rbd_dev->id;
+ int max_id;
+
+ BUG_ON(rbd_id < 1);
+
+ spin_lock(&rbd_dev_list_lock);
+ list_del_init(&rbd_dev->node);
+
+ /*
+ * If the id being "put" is not the current maximum, there
+ * is nothing special we need to do.
+ */
+ if (rbd_id != atomic64_read(&rbd_id_max)) {
+ spin_unlock(&rbd_dev_list_lock);
+ return;
+ }
+
+ /*
+ * We need to update the current maximum id. Search the
+ * list to find out what it is. We're more likely to find
+ * the maximum at the end, so search the list backward.
+ */
+ max_id = 0;
+ list_for_each_prev(tmp, &rbd_dev_list) {
+ struct rbd_device *rbd_dev;
+
+ rbd_dev = list_entry(tmp, struct rbd_device, node);
+ if (rbd_id > max_id)
+ max_id = rbd_id;
+ }
+ spin_unlock(&rbd_dev_list_lock);
+
+ /*
+ * The max id could have been updated by rbd_id_get(), in
+ * which case it now accurately reflects the new maximum.
+ * Be careful not to overwrite the maximum value in that
+ * case.
+ */
+ atomic64_cmpxchg(&rbd_id_max, rbd_id, max_id);
+}
+
+/*
+ * Skips over white space at *buf, and updates *buf to point to the
+ * first found non-space character (if any). Returns the length of
+ * the token (string of non-white space characters) found. Note
+ * that *buf must be terminated with '\0'.
+ */
+static inline size_t next_token(const char **buf)
+{
+ /*
+ * These are the characters that produce nonzero for
+ * isspace() in the "C" and "POSIX" locales.
+ */
+ const char *spaces = " \f\n\r\t\v";
+
+ *buf += strspn(*buf, spaces); /* Find start of token */
+
+ return strcspn(*buf, spaces); /* Return token length */
+}
+
+/*
+ * Finds the next token in *buf, and if the provided token buffer is
+ * big enough, copies the found token into it. The result, if
+ * copied, is guaranteed to be terminated with '\0'. Note that *buf
+ * must be terminated with '\0' on entry.
+ *
+ * Returns the length of the token found (not including the '\0').
+ * Return value will be 0 if no token is found, and it will be >=
+ * token_size if the token would not fit.
+ *
+ * The *buf pointer will be updated to point beyond the end of the
+ * found token. Note that this occurs even if the token buffer is
+ * too small to hold it.
+ */
+static inline size_t copy_token(const char **buf,
+ char *token,
+ size_t token_size)
+{
+ size_t len;
+
+ len = next_token(buf);
+ if (len < token_size) {
+ memcpy(token, *buf, len);
+ *(token + len) = '\0';
+ }
+ *buf += len;
+
+ return len;
+}
+
+/*
+ * This fills in the pool_name, obj, obj_len, snap_name, obj_len,
+ * rbd_dev, rbd_md_name, and name fields of the given rbd_dev, based
+ * on the list of monitor addresses and other options provided via
+ * /sys/bus/rbd/add.
+ */
+static int rbd_add_parse_args(struct rbd_device *rbd_dev,
+ const char *buf,
+ const char **mon_addrs,
+ size_t *mon_addrs_size,
+ char *options,
+ size_t options_size)
+{
+ size_t len;
+
+ /* The first four tokens are required */
+
+ len = next_token(&buf);
+ if (!len)
+ return -EINVAL;
+ *mon_addrs_size = len + 1;
+ *mon_addrs = buf;
+
+ buf += len;
+
+ len = copy_token(&buf, options, options_size);
+ if (!len || len >= options_size)
+ return -EINVAL;
+
+ len = copy_token(&buf, rbd_dev->pool_name, sizeof (rbd_dev->pool_name));
+ if (!len || len >= sizeof (rbd_dev->pool_name))
+ return -EINVAL;
+
+ len = copy_token(&buf, rbd_dev->obj, sizeof (rbd_dev->obj));
+ if (!len || len >= sizeof (rbd_dev->obj))
+ return -EINVAL;
+
+ /* We have the object length in hand, save it. */
+
+ rbd_dev->obj_len = len;
+
+ BUILD_BUG_ON(RBD_MAX_MD_NAME_LEN
+ < RBD_MAX_OBJ_NAME_LEN + sizeof (RBD_SUFFIX));
+ sprintf(rbd_dev->obj_md_name, "%s%s", rbd_dev->obj, RBD_SUFFIX);
+
+ /*
+ * The snapshot name is optional, but it's an error if it's
+ * too long. If no snapshot is supplied, fill in the default.
+ */
+ len = copy_token(&buf, rbd_dev->snap_name, sizeof (rbd_dev->snap_name));
+ if (!len)
+ memcpy(rbd_dev->snap_name, RBD_SNAP_HEAD_NAME,
+ sizeof (RBD_SNAP_HEAD_NAME));
+ else if (len >= sizeof (rbd_dev->snap_name))
+ return -EINVAL;
+
+ return 0;
+}
+
static ssize_t rbd_add(struct bus_type *bus,
const char *buf,
size_t count)
{
- struct ceph_osd_client *osdc;
struct rbd_device *rbd_dev;
- ssize_t rc = -ENOMEM;
- int irc, new_id = 0;
- struct list_head *tmp;
- char *mon_dev_name;
- char *options;
+ const char *mon_addrs = NULL;
+ size_t mon_addrs_size = 0;
+ char *options = NULL;
+ struct ceph_osd_client *osdc;
+ int rc = -ENOMEM;
if (!try_module_get(THIS_MODULE))
return -ENODEV;
- mon_dev_name = kmalloc(RBD_MAX_OPT_LEN, GFP_KERNEL);
- if (!mon_dev_name)
- goto err_out_mod;
-
- options = kmalloc(RBD_MAX_OPT_LEN, GFP_KERNEL);
- if (!options)
- goto err_mon_dev;
-
- /* new rbd_device object */
rbd_dev = kzalloc(sizeof(*rbd_dev), GFP_KERNEL);
if (!rbd_dev)
- goto err_out_opt;
+ goto err_nomem;
+ options = kmalloc(count, GFP_KERNEL);
+ if (!options)
+ goto err_nomem;
/* static rbd_device initialization */
spin_lock_init(&rbd_dev->lock);
INIT_LIST_HEAD(&rbd_dev->node);
INIT_LIST_HEAD(&rbd_dev->snaps);
+ init_rwsem(&rbd_dev->header_rwsem);
- init_rwsem(&rbd_dev->header.snap_rwsem);
+ init_rwsem(&rbd_dev->header_rwsem);
/* generate unique id: find highest unique id, add one */
- mutex_lock_nested(&ctl_mutex, SINGLE_DEPTH_NESTING);
-
- list_for_each(tmp, &rbd_dev_list) {
- struct rbd_device *rbd_dev;
+ rbd_id_get(rbd_dev);
- rbd_dev = list_entry(tmp, struct rbd_device, node);
- if (rbd_dev->id >= new_id)
- new_id = rbd_dev->id + 1;
- }
-
- rbd_dev->id = new_id;
-
- /* add to global list */
- list_add_tail(&rbd_dev->node, &rbd_dev_list);
+ /* Fill in the device name, now that we have its id. */
+ BUILD_BUG_ON(DEV_NAME_LEN
+ < sizeof (RBD_DRV_NAME) + MAX_INT_FORMAT_WIDTH);
+ sprintf(rbd_dev->name, "%s%d", RBD_DRV_NAME, rbd_dev->id);
/* parse add command */
- if (sscanf(buf, "%" __stringify(RBD_MAX_OPT_LEN) "s "
- "%" __stringify(RBD_MAX_OPT_LEN) "s "
- "%" __stringify(RBD_MAX_POOL_NAME_LEN) "s "
- "%" __stringify(RBD_MAX_OBJ_NAME_LEN) "s"
- "%" __stringify(RBD_MAX_SNAP_NAME_LEN) "s",
- mon_dev_name, options, rbd_dev->pool_name,
- rbd_dev->obj, rbd_dev->snap_name) < 4) {
- rc = -EINVAL;
- goto err_out_slot;
- }
-
- if (rbd_dev->snap_name[0] == 0)
- rbd_dev->snap_name[0] = '-';
-
- rbd_dev->obj_len = strlen(rbd_dev->obj);
- snprintf(rbd_dev->obj_md_name, sizeof(rbd_dev->obj_md_name), "%s%s",
- rbd_dev->obj, RBD_SUFFIX);
-
- /* initialize rest of new object */
- snprintf(rbd_dev->name, DEV_NAME_LEN, DRV_NAME "%d", rbd_dev->id);
- rc = rbd_get_client(rbd_dev, mon_dev_name, options);
- if (rc < 0)
- goto err_out_slot;
+ rc = rbd_add_parse_args(rbd_dev, buf, &mon_addrs, &mon_addrs_size,
+ options, count);
+ if (rc)
+ goto err_put_id;
- mutex_unlock(&ctl_mutex);
+ rbd_dev->rbd_client = rbd_get_client(mon_addrs, mon_addrs_size - 1,
+ options);
+ if (IS_ERR(rbd_dev->rbd_client)) {
+ rc = PTR_ERR(rbd_dev->rbd_client);
+ goto err_put_id;
+ }
/* pick the pool */
- osdc = &rbd_dev->client->osdc;
+ osdc = &rbd_dev->rbd_client->client->osdc;
rc = ceph_pg_poolid_by_name(osdc->osdmap, rbd_dev->pool_name);
if (rc < 0)
goto err_out_client;
rbd_dev->poolid = rc;
/* register our block device */
- irc = register_blkdev(0, rbd_dev->name);
- if (irc < 0) {
- rc = irc;
+ rc = register_blkdev(0, rbd_dev->name);
+ if (rc < 0)
goto err_out_client;
- }
- rbd_dev->major = irc;
+ rbd_dev->major = rc;
rc = rbd_bus_add_dev(rbd_dev);
if (rc)
goto err_out_blkdev;
- /* set up and announce blkdev mapping */
+ /*
+ * At this point cleanup in the event of an error is the job
+ * of the sysfs code (initiated by rbd_bus_del_dev()).
+ *
+ * Set up and announce blkdev mapping.
+ */
rc = rbd_init_disk(rbd_dev);
if (rc)
goto err_out_bus;
@@ -2263,35 +2438,26 @@ static ssize_t rbd_add(struct bus_type *bus,
return count;
err_out_bus:
- mutex_lock_nested(&ctl_mutex, SINGLE_DEPTH_NESTING);
- list_del_init(&rbd_dev->node);
- mutex_unlock(&ctl_mutex);
-
/* this will also clean up rest of rbd_dev stuff */
rbd_bus_del_dev(rbd_dev);
kfree(options);
- kfree(mon_dev_name);
return rc;
err_out_blkdev:
unregister_blkdev(rbd_dev->major, rbd_dev->name);
err_out_client:
rbd_put_client(rbd_dev);
- mutex_lock_nested(&ctl_mutex, SINGLE_DEPTH_NESTING);
-err_out_slot:
- list_del_init(&rbd_dev->node);
- mutex_unlock(&ctl_mutex);
-
- kfree(rbd_dev);
-err_out_opt:
+err_put_id:
+ rbd_id_put(rbd_dev);
+err_nomem:
kfree(options);
-err_mon_dev:
- kfree(mon_dev_name);
-err_out_mod:
+ kfree(rbd_dev);
+
dout("Error adding device %s\n", buf);
module_put(THIS_MODULE);
- return rc;
+
+ return (ssize_t) rc;
}
static struct rbd_device *__rbd_get_dev(unsigned long id)
@@ -2299,22 +2465,28 @@ static struct rbd_device *__rbd_get_dev(unsigned long id)
struct list_head *tmp;
struct rbd_device *rbd_dev;
+ spin_lock(&rbd_dev_list_lock);
list_for_each(tmp, &rbd_dev_list) {
rbd_dev = list_entry(tmp, struct rbd_device, node);
- if (rbd_dev->id == id)
+ if (rbd_dev->id == id) {
+ spin_unlock(&rbd_dev_list_lock);
return rbd_dev;
+ }
}
+ spin_unlock(&rbd_dev_list_lock);
return NULL;
}
static void rbd_dev_release(struct device *dev)
{
- struct rbd_device *rbd_dev =
- container_of(dev, struct rbd_device, dev);
+ struct rbd_device *rbd_dev = dev_to_rbd_dev(dev);
- if (rbd_dev->watch_request)
- ceph_osdc_unregister_linger_request(&rbd_dev->client->osdc,
+ if (rbd_dev->watch_request) {
+ struct ceph_client *client = rbd_dev->rbd_client->client;
+
+ ceph_osdc_unregister_linger_request(&client->osdc,
rbd_dev->watch_request);
+ }
if (rbd_dev->watch_event)
rbd_req_sync_unwatch(rbd_dev, rbd_dev->obj_md_name);
@@ -2323,6 +2495,9 @@ static void rbd_dev_release(struct device *dev)
/* clean up and free blkdev */
rbd_free_disk(rbd_dev);
unregister_blkdev(rbd_dev->major, rbd_dev->name);
+
+ /* done with the id, and with the rbd_dev */
+ rbd_id_put(rbd_dev);
kfree(rbd_dev);
/* release module ref */
@@ -2355,8 +2530,6 @@ static ssize_t rbd_remove(struct bus_type *bus,
goto done;
}
- list_del_init(&rbd_dev->node);
-
__rbd_remove_all_snaps(rbd_dev);
rbd_bus_del_dev(rbd_dev);
@@ -2370,7 +2543,7 @@ static ssize_t rbd_snap_add(struct device *dev,
const char *buf,
size_t count)
{
- struct rbd_device *rbd_dev = dev_to_rbd(dev);
+ struct rbd_device *rbd_dev = dev_to_rbd_dev(dev);
int ret;
char *name = kmalloc(count + 1, GFP_KERNEL);
if (!name)
@@ -2406,12 +2579,6 @@ err_unlock:
return ret;
}
-static struct bus_attribute rbd_bus_attrs[] = {
- __ATTR(add, S_IWUSR, NULL, rbd_add),
- __ATTR(remove, S_IWUSR, NULL, rbd_remove),
- __ATTR_NULL
-};
-
/*
* create control files in sysfs
* /sys/bus/rbd/...
@@ -2420,21 +2587,21 @@ static int rbd_sysfs_init(void)
{
int ret;
- rbd_bus_type.bus_attrs = rbd_bus_attrs;
-
- ret = bus_register(&rbd_bus_type);
- if (ret < 0)
+ ret = device_register(&rbd_root_dev);
+ if (ret < 0)
return ret;
- ret = device_register(&rbd_root_dev);
+ ret = bus_register(&rbd_bus_type);
+ if (ret < 0)
+ device_unregister(&rbd_root_dev);
return ret;
}
static void rbd_sysfs_cleanup(void)
{
- device_unregister(&rbd_root_dev);
bus_unregister(&rbd_bus_type);
+ device_unregister(&rbd_root_dev);
}
int __init rbd_init(void)
@@ -2444,8 +2611,7 @@ int __init rbd_init(void)
rc = rbd_sysfs_init();
if (rc)
return rc;
- spin_lock_init(&node_lock);
- pr_info("loaded " DRV_NAME_LONG "\n");
+ pr_info("loaded " RBD_DRV_NAME_LONG "\n");
return 0;
}
diff --git a/drivers/block/rbd_types.h b/drivers/block/rbd_types.h
index fc6c678aa2c..950708688f1 100644
--- a/drivers/block/rbd_types.h
+++ b/drivers/block/rbd_types.h
@@ -41,10 +41,6 @@
#define RBD_HEADER_SIGNATURE "RBD"
#define RBD_HEADER_VERSION "001.005"
-struct rbd_info {
- __le64 max_id;
-} __attribute__ ((packed));
-
struct rbd_image_snap_ondisk {
__le64 id;
__le64 image_size;
diff --git a/drivers/block/sunvdc.c b/drivers/block/sunvdc.c
index 48e8fee9f2d..9dcf76a10bb 100644
--- a/drivers/block/sunvdc.c
+++ b/drivers/block/sunvdc.c
@@ -839,10 +839,7 @@ static struct vio_driver vdc_port_driver = {
.id_table = vdc_port_match,
.probe = vdc_port_probe,
.remove = vdc_port_remove,
- .driver = {
- .name = "vdc_port",
- .owner = THIS_MODULE,
- }
+ .name = "vdc_port",
};
static int __init vdc_init(void)
diff --git a/drivers/block/sx8.c b/drivers/block/sx8.c
index e7472f567c9..3fb6ab4c8b4 100644
--- a/drivers/block/sx8.c
+++ b/drivers/block/sx8.c
@@ -1120,7 +1120,7 @@ static inline void carm_handle_resp(struct carm_host *host,
break;
case MISC_GET_FW_VER: {
struct carm_fw_ver *ver = (struct carm_fw_ver *)
- mem + sizeof(struct carm_msg_get_fw_ver);
+ (mem + sizeof(struct carm_msg_get_fw_ver));
if (!error) {
host->fw_ver = le32_to_cpu(ver->version);
host->flags |= (ver->features & FL_FW_VER_MASK);
diff --git a/drivers/block/ub.c b/drivers/block/ub.c
index 7333b9e4441..fcec0225ac7 100644
--- a/drivers/block/ub.c
+++ b/drivers/block/ub.c
@@ -119,43 +119,6 @@
/*
*/
-
-/* command block wrapper */
-struct bulk_cb_wrap {
- __le32 Signature; /* contains 'USBC' */
- u32 Tag; /* unique per command id */
- __le32 DataTransferLength; /* size of data */
- u8 Flags; /* direction in bit 0 */
- u8 Lun; /* LUN */
- u8 Length; /* of of the CDB */
- u8 CDB[UB_MAX_CDB_SIZE]; /* max command */
-};
-
-#define US_BULK_CB_WRAP_LEN 31
-#define US_BULK_CB_SIGN 0x43425355 /*spells out USBC */
-#define US_BULK_FLAG_IN 1
-#define US_BULK_FLAG_OUT 0
-
-/* command status wrapper */
-struct bulk_cs_wrap {
- __le32 Signature; /* should = 'USBS' */
- u32 Tag; /* same as original command */
- __le32 Residue; /* amount not transferred */
- u8 Status; /* see below */
-};
-
-#define US_BULK_CS_WRAP_LEN 13
-#define US_BULK_CS_SIGN 0x53425355 /* spells out 'USBS' */
-#define US_BULK_STAT_OK 0
-#define US_BULK_STAT_FAIL 1
-#define US_BULK_STAT_PHASE 2
-
-/* bulk-only class specific requests */
-#define US_BULK_RESET_REQUEST 0xff
-#define US_BULK_GET_MAX_LUN 0xfe
-
-/*
- */
struct ub_dev;
#define UB_MAX_REQ_SG 9 /* cdrecord requires 32KB and maybe a header */
@@ -2477,6 +2440,8 @@ static int __init ub_init(void)
int rc;
int i;
+ pr_info("'Low Performance USB Block' driver is deprecated. "
+ "Please switch to usb-storage\n");
for (i = 0; i < UB_QLOCK_NUM; i++)
spin_lock_init(&ub_qlockv[i]);
diff --git a/drivers/block/viodasd.c b/drivers/block/viodasd.c
deleted file mode 100644
index 9a5b2a2d616..00000000000
--- a/drivers/block/viodasd.c
+++ /dev/null
@@ -1,809 +0,0 @@
-/* -*- linux-c -*-
- * viodasd.c
- * Authors: Dave Boutcher <boutcher@us.ibm.com>
- * Ryan Arnold <ryanarn@us.ibm.com>
- * Colin Devilbiss <devilbis@us.ibm.com>
- * Stephen Rothwell
- *
- * (C) Copyright 2000-2004 IBM Corporation
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License as
- * published by the Free Software Foundation; either version 2 of the
- * License, or (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
- *
- * This routine provides access to disk space (termed "DASD" in historical
- * IBM terms) owned and managed by an OS/400 partition running on the
- * same box as this Linux partition.
- *
- * All disk operations are performed by sending messages back and forth to
- * the OS/400 partition.
- */
-
-#define pr_fmt(fmt) "viod: " fmt
-
-#include <linux/major.h>
-#include <linux/fs.h>
-#include <linux/module.h>
-#include <linux/kernel.h>
-#include <linux/blkdev.h>
-#include <linux/genhd.h>
-#include <linux/hdreg.h>
-#include <linux/errno.h>
-#include <linux/init.h>
-#include <linux/string.h>
-#include <linux/mutex.h>
-#include <linux/dma-mapping.h>
-#include <linux/completion.h>
-#include <linux/device.h>
-#include <linux/scatterlist.h>
-
-#include <asm/uaccess.h>
-#include <asm/vio.h>
-#include <asm/iseries/hv_types.h>
-#include <asm/iseries/hv_lp_event.h>
-#include <asm/iseries/hv_lp_config.h>
-#include <asm/iseries/vio.h>
-#include <asm/firmware.h>
-
-MODULE_DESCRIPTION("iSeries Virtual DASD");
-MODULE_AUTHOR("Dave Boutcher");
-MODULE_LICENSE("GPL");
-
-/*
- * We only support 7 partitions per physical disk....so with minor
- * numbers 0-255 we get a maximum of 32 disks.
- */
-#define VIOD_GENHD_NAME "iseries/vd"
-
-#define VIOD_VERS "1.64"
-
-enum {
- PARTITION_SHIFT = 3,
- MAX_DISKNO = HVMAXARCHITECTEDVIRTUALDISKS,
- MAX_DISK_NAME = FIELD_SIZEOF(struct gendisk, disk_name)
-};
-
-static DEFINE_MUTEX(viodasd_mutex);
-static DEFINE_SPINLOCK(viodasd_spinlock);
-
-#define VIOMAXREQ 16
-
-#define DEVICE_NO(cell) ((struct viodasd_device *)(cell) - &viodasd_devices[0])
-
-struct viodasd_waitevent {
- struct completion com;
- int rc;
- u16 sub_result;
- int max_disk; /* open */
-};
-
-static const struct vio_error_entry viodasd_err_table[] = {
- { 0x0201, EINVAL, "Invalid Range" },
- { 0x0202, EINVAL, "Invalid Token" },
- { 0x0203, EIO, "DMA Error" },
- { 0x0204, EIO, "Use Error" },
- { 0x0205, EIO, "Release Error" },
- { 0x0206, EINVAL, "Invalid Disk" },
- { 0x0207, EBUSY, "Can't Lock" },
- { 0x0208, EIO, "Already Locked" },
- { 0x0209, EIO, "Already Unlocked" },
- { 0x020A, EIO, "Invalid Arg" },
- { 0x020B, EIO, "Bad IFS File" },
- { 0x020C, EROFS, "Read Only Device" },
- { 0x02FF, EIO, "Internal Error" },
- { 0x0000, 0, NULL },
-};
-
-/*
- * Figure out the biggest I/O request (in sectors) we can accept
- */
-#define VIODASD_MAXSECTORS (4096 / 512 * VIOMAXBLOCKDMA)
-
-/*
- * Number of disk I/O requests we've sent to OS/400
- */
-static int num_req_outstanding;
-
-/*
- * This is our internal structure for keeping track of disk devices
- */
-struct viodasd_device {
- u16 cylinders;
- u16 tracks;
- u16 sectors;
- u16 bytes_per_sector;
- u64 size;
- int read_only;
- spinlock_t q_lock;
- struct gendisk *disk;
- struct device *dev;
-} viodasd_devices[MAX_DISKNO];
-
-/*
- * External open entry point.
- */
-static int viodasd_open(struct block_device *bdev, fmode_t mode)
-{
- struct viodasd_device *d = bdev->bd_disk->private_data;
- HvLpEvent_Rc hvrc;
- struct viodasd_waitevent we;
- u16 flags = 0;
-
- if (d->read_only) {
- if (mode & FMODE_WRITE)
- return -EROFS;
- flags = vioblockflags_ro;
- }
-
- init_completion(&we.com);
-
- /* Send the open event to OS/400 */
- hvrc = HvCallEvent_signalLpEventFast(viopath_hostLp,
- HvLpEvent_Type_VirtualIo,
- viomajorsubtype_blockio | vioblockopen,
- HvLpEvent_AckInd_DoAck, HvLpEvent_AckType_ImmediateAck,
- viopath_sourceinst(viopath_hostLp),
- viopath_targetinst(viopath_hostLp),
- (u64)(unsigned long)&we, VIOVERSION << 16,
- ((u64)DEVICE_NO(d) << 48) | ((u64)flags << 32),
- 0, 0, 0);
- if (hvrc != 0) {
- pr_warning("HV open failed %d\n", (int)hvrc);
- return -EIO;
- }
-
- wait_for_completion(&we.com);
-
- /* Check the return code */
- if (we.rc != 0) {
- const struct vio_error_entry *err =
- vio_lookup_rc(viodasd_err_table, we.sub_result);
-
- pr_warning("bad rc opening disk: %d:0x%04x (%s)\n",
- (int)we.rc, we.sub_result, err->msg);
- return -EIO;
- }
-
- return 0;
-}
-
-static int viodasd_unlocked_open(struct block_device *bdev, fmode_t mode)
-{
- int ret;
-
- mutex_lock(&viodasd_mutex);
- ret = viodasd_open(bdev, mode);
- mutex_unlock(&viodasd_mutex);
-
- return ret;
-}
-
-
-/*
- * External release entry point.
- */
-static int viodasd_release(struct gendisk *disk, fmode_t mode)
-{
- struct viodasd_device *d = disk->private_data;
- HvLpEvent_Rc hvrc;
-
- mutex_lock(&viodasd_mutex);
- /* Send the event to OS/400. We DON'T expect a response */
- hvrc = HvCallEvent_signalLpEventFast(viopath_hostLp,
- HvLpEvent_Type_VirtualIo,
- viomajorsubtype_blockio | vioblockclose,
- HvLpEvent_AckInd_NoAck, HvLpEvent_AckType_ImmediateAck,
- viopath_sourceinst(viopath_hostLp),
- viopath_targetinst(viopath_hostLp),
- 0, VIOVERSION << 16,
- ((u64)DEVICE_NO(d) << 48) /* | ((u64)flags << 32) */,
- 0, 0, 0);
- if (hvrc != 0)
- pr_warning("HV close call failed %d\n", (int)hvrc);
-
- mutex_unlock(&viodasd_mutex);
-
- return 0;
-}
-
-
-/* External ioctl entry point.
- */
-static int viodasd_getgeo(struct block_device *bdev, struct hd_geometry *geo)
-{
- struct gendisk *disk = bdev->bd_disk;
- struct viodasd_device *d = disk->private_data;
-
- geo->sectors = d->sectors ? d->sectors : 32;
- geo->heads = d->tracks ? d->tracks : 64;
- geo->cylinders = d->cylinders ? d->cylinders :
- get_capacity(disk) / (geo->sectors * geo->heads);
-
- return 0;
-}
-
-/*
- * Our file operations table
- */
-static const struct block_device_operations viodasd_fops = {
- .owner = THIS_MODULE,
- .open = viodasd_unlocked_open,
- .release = viodasd_release,
- .getgeo = viodasd_getgeo,
-};
-
-/*
- * End a request
- */
-static void viodasd_end_request(struct request *req, int error,
- int num_sectors)
-{
- __blk_end_request(req, error, num_sectors << 9);
-}
-
-/*
- * Send an actual I/O request to OS/400
- */
-static int send_request(struct request *req)
-{
- u64 start;
- int direction;
- int nsg;
- u16 viocmd;
- HvLpEvent_Rc hvrc;
- struct vioblocklpevent *bevent;
- struct HvLpEvent *hev;
- struct scatterlist sg[VIOMAXBLOCKDMA];
- int sgindex;
- struct viodasd_device *d;
- unsigned long flags;
-
- start = (u64)blk_rq_pos(req) << 9;
-
- if (rq_data_dir(req) == READ) {
- direction = DMA_FROM_DEVICE;
- viocmd = viomajorsubtype_blockio | vioblockread;
- } else {
- direction = DMA_TO_DEVICE;
- viocmd = viomajorsubtype_blockio | vioblockwrite;
- }
-
- d = req->rq_disk->private_data;
-
- /* Now build the scatter-gather list */
- sg_init_table(sg, VIOMAXBLOCKDMA);
- nsg = blk_rq_map_sg(req->q, req, sg);
- nsg = dma_map_sg(d->dev, sg, nsg, direction);
-
- spin_lock_irqsave(&viodasd_spinlock, flags);
- num_req_outstanding++;
-
- /* This optimization handles a single DMA block */
- if (nsg == 1)
- hvrc = HvCallEvent_signalLpEventFast(viopath_hostLp,
- HvLpEvent_Type_VirtualIo, viocmd,
- HvLpEvent_AckInd_DoAck,
- HvLpEvent_AckType_ImmediateAck,
- viopath_sourceinst(viopath_hostLp),
- viopath_targetinst(viopath_hostLp),
- (u64)(unsigned long)req, VIOVERSION << 16,
- ((u64)DEVICE_NO(d) << 48), start,
- ((u64)sg_dma_address(&sg[0])) << 32,
- sg_dma_len(&sg[0]));
- else {
- bevent = (struct vioblocklpevent *)
- vio_get_event_buffer(viomajorsubtype_blockio);
- if (bevent == NULL) {
- pr_warning("error allocating disk event buffer\n");
- goto error_ret;
- }
-
- /*
- * Now build up the actual request. Note that we store
- * the pointer to the request in the correlation
- * token so we can match the response up later
- */
- memset(bevent, 0, sizeof(struct vioblocklpevent));
- hev = &bevent->event;
- hev->flags = HV_LP_EVENT_VALID | HV_LP_EVENT_DO_ACK |
- HV_LP_EVENT_INT;
- hev->xType = HvLpEvent_Type_VirtualIo;
- hev->xSubtype = viocmd;
- hev->xSourceLp = HvLpConfig_getLpIndex();
- hev->xTargetLp = viopath_hostLp;
- hev->xSizeMinus1 =
- offsetof(struct vioblocklpevent, u.rw_data.dma_info) +
- (sizeof(bevent->u.rw_data.dma_info[0]) * nsg) - 1;
- hev->xSourceInstanceId = viopath_sourceinst(viopath_hostLp);
- hev->xTargetInstanceId = viopath_targetinst(viopath_hostLp);
- hev->xCorrelationToken = (u64)req;
- bevent->version = VIOVERSION;
- bevent->disk = DEVICE_NO(d);
- bevent->u.rw_data.offset = start;
-
- /*
- * Copy just the dma information from the sg list
- * into the request
- */
- for (sgindex = 0; sgindex < nsg; sgindex++) {
- bevent->u.rw_data.dma_info[sgindex].token =
- sg_dma_address(&sg[sgindex]);
- bevent->u.rw_data.dma_info[sgindex].len =
- sg_dma_len(&sg[sgindex]);
- }
-
- /* Send the request */
- hvrc = HvCallEvent_signalLpEvent(&bevent->event);
- vio_free_event_buffer(viomajorsubtype_blockio, bevent);
- }
-
- if (hvrc != HvLpEvent_Rc_Good) {
- pr_warning("error sending disk event to OS/400 (rc %d)\n",
- (int)hvrc);
- goto error_ret;
- }
- spin_unlock_irqrestore(&viodasd_spinlock, flags);
- return 0;
-
-error_ret:
- num_req_outstanding--;
- spin_unlock_irqrestore(&viodasd_spinlock, flags);
- dma_unmap_sg(d->dev, sg, nsg, direction);
- return -1;
-}
-
-/*
- * This is the external request processing routine
- */
-static void do_viodasd_request(struct request_queue *q)
-{
- struct request *req;
-
- /*
- * If we already have the maximum number of requests
- * outstanding to OS/400 just bail out. We'll come
- * back later.
- */
- while (num_req_outstanding < VIOMAXREQ) {
- req = blk_fetch_request(q);
- if (req == NULL)
- return;
- /* check that request contains a valid command */
- if (req->cmd_type != REQ_TYPE_FS) {
- viodasd_end_request(req, -EIO, blk_rq_sectors(req));
- continue;
- }
- /* Try sending the request */
- if (send_request(req) != 0)
- viodasd_end_request(req, -EIO, blk_rq_sectors(req));
- }
-}
-
-/*
- * Probe a single disk and fill in the viodasd_device structure
- * for it.
- */
-static int probe_disk(struct viodasd_device *d)
-{
- HvLpEvent_Rc hvrc;
- struct viodasd_waitevent we;
- int dev_no = DEVICE_NO(d);
- struct gendisk *g;
- struct request_queue *q;
- u16 flags = 0;
-
-retry:
- init_completion(&we.com);
-
- /* Send the open event to OS/400 */
- hvrc = HvCallEvent_signalLpEventFast(viopath_hostLp,
- HvLpEvent_Type_VirtualIo,
- viomajorsubtype_blockio | vioblockopen,
- HvLpEvent_AckInd_DoAck, HvLpEvent_AckType_ImmediateAck,
- viopath_sourceinst(viopath_hostLp),
- viopath_targetinst(viopath_hostLp),
- (u64)(unsigned long)&we, VIOVERSION << 16,
- ((u64)dev_no << 48) | ((u64)flags<< 32),
- 0, 0, 0);
- if (hvrc != 0) {
- pr_warning("bad rc on HV open %d\n", (int)hvrc);
- return 0;
- }
-
- wait_for_completion(&we.com);
-
- if (we.rc != 0) {
- if (flags != 0)
- return 0;
- /* try again with read only flag set */
- flags = vioblockflags_ro;
- goto retry;
- }
- if (we.max_disk > (MAX_DISKNO - 1)) {
- printk_once(KERN_INFO pr_fmt("Only examining the first %d of %d disks connected\n"),
- MAX_DISKNO, we.max_disk + 1);
- }
-
- /* Send the close event to OS/400. We DON'T expect a response */
- hvrc = HvCallEvent_signalLpEventFast(viopath_hostLp,
- HvLpEvent_Type_VirtualIo,
- viomajorsubtype_blockio | vioblockclose,
- HvLpEvent_AckInd_NoAck, HvLpEvent_AckType_ImmediateAck,
- viopath_sourceinst(viopath_hostLp),
- viopath_targetinst(viopath_hostLp),
- 0, VIOVERSION << 16,
- ((u64)dev_no << 48) | ((u64)flags << 32),
- 0, 0, 0);
- if (hvrc != 0) {
- pr_warning("bad rc sending event to OS/400 %d\n", (int)hvrc);
- return 0;
- }
-
- if (d->dev == NULL) {
- /* this is when we reprobe for new disks */
- if (vio_create_viodasd(dev_no) == NULL) {
- pr_warning("cannot allocate virtual device for disk %d\n",
- dev_no);
- return 0;
- }
- /*
- * The vio_create_viodasd will have recursed into this
- * routine with d->dev set to the new vio device and
- * will finish the setup of the disk below.
- */
- return 1;
- }
-
- /* create the request queue for the disk */
- spin_lock_init(&d->q_lock);
- q = blk_init_queue(do_viodasd_request, &d->q_lock);
- if (q == NULL) {
- pr_warning("cannot allocate queue for disk %d\n", dev_no);
- return 0;
- }
- g = alloc_disk(1 << PARTITION_SHIFT);
- if (g == NULL) {
- pr_warning("cannot allocate disk structure for disk %d\n",
- dev_no);
- blk_cleanup_queue(q);
- return 0;
- }
-
- d->disk = g;
- blk_queue_max_segments(q, VIOMAXBLOCKDMA);
- blk_queue_max_hw_sectors(q, VIODASD_MAXSECTORS);
- g->major = VIODASD_MAJOR;
- g->first_minor = dev_no << PARTITION_SHIFT;
- if (dev_no >= 26)
- snprintf(g->disk_name, sizeof(g->disk_name),
- VIOD_GENHD_NAME "%c%c",
- 'a' + (dev_no / 26) - 1, 'a' + (dev_no % 26));
- else
- snprintf(g->disk_name, sizeof(g->disk_name),
- VIOD_GENHD_NAME "%c", 'a' + (dev_no % 26));
- g->fops = &viodasd_fops;
- g->queue = q;
- g->private_data = d;
- g->driverfs_dev = d->dev;
- set_capacity(g, d->size >> 9);
-
- pr_info("disk %d: %lu sectors (%lu MB) CHS=%d/%d/%d sector size %d%s\n",
- dev_no, (unsigned long)(d->size >> 9),
- (unsigned long)(d->size >> 20),
- (int)d->cylinders, (int)d->tracks,
- (int)d->sectors, (int)d->bytes_per_sector,
- d->read_only ? " (RO)" : "");
-
- /* register us in the global list */
- add_disk(g);
- return 1;
-}
-
-/* returns the total number of scatterlist elements converted */
-static int block_event_to_scatterlist(const struct vioblocklpevent *bevent,
- struct scatterlist *sg, int *total_len)
-{
- int i, numsg;
- const struct rw_data *rw_data = &bevent->u.rw_data;
- static const int offset =
- offsetof(struct vioblocklpevent, u.rw_data.dma_info);
- static const int element_size = sizeof(rw_data->dma_info[0]);
-
- numsg = ((bevent->event.xSizeMinus1 + 1) - offset) / element_size;
- if (numsg > VIOMAXBLOCKDMA)
- numsg = VIOMAXBLOCKDMA;
-
- *total_len = 0;
- sg_init_table(sg, VIOMAXBLOCKDMA);
- for (i = 0; (i < numsg) && (rw_data->dma_info[i].len > 0); ++i) {
- sg_dma_address(&sg[i]) = rw_data->dma_info[i].token;
- sg_dma_len(&sg[i]) = rw_data->dma_info[i].len;
- *total_len += rw_data->dma_info[i].len;
- }
- return i;
-}
-
-/*
- * Restart all queues, starting with the one _after_ the disk given,
- * thus reducing the chance of starvation of higher numbered disks.
- */
-static void viodasd_restart_all_queues_starting_from(int first_index)
-{
- int i;
-
- for (i = first_index + 1; i < MAX_DISKNO; ++i)
- if (viodasd_devices[i].disk)
- blk_run_queue(viodasd_devices[i].disk->queue);
- for (i = 0; i <= first_index; ++i)
- if (viodasd_devices[i].disk)
- blk_run_queue(viodasd_devices[i].disk->queue);
-}
-
-/*
- * For read and write requests, decrement the number of outstanding requests,
- * Free the DMA buffers we allocated.
- */
-static int viodasd_handle_read_write(struct vioblocklpevent *bevent)
-{
- int num_sg, num_sect, pci_direction, total_len;
- struct request *req;
- struct scatterlist sg[VIOMAXBLOCKDMA];
- struct HvLpEvent *event = &bevent->event;
- unsigned long irq_flags;
- struct viodasd_device *d;
- int error;
- spinlock_t *qlock;
-
- num_sg = block_event_to_scatterlist(bevent, sg, &total_len);
- num_sect = total_len >> 9;
- if (event->xSubtype == (viomajorsubtype_blockio | vioblockread))
- pci_direction = DMA_FROM_DEVICE;
- else
- pci_direction = DMA_TO_DEVICE;
- req = (struct request *)bevent->event.xCorrelationToken;
- d = req->rq_disk->private_data;
-
- dma_unmap_sg(d->dev, sg, num_sg, pci_direction);
-
- /*
- * Since this is running in interrupt mode, we need to make sure
- * we're not stepping on any global I/O operations
- */
- spin_lock_irqsave(&viodasd_spinlock, irq_flags);
- num_req_outstanding--;
- spin_unlock_irqrestore(&viodasd_spinlock, irq_flags);
-
- error = (event->xRc == HvLpEvent_Rc_Good) ? 0 : -EIO;
- if (error) {
- const struct vio_error_entry *err;
- err = vio_lookup_rc(viodasd_err_table, bevent->sub_result);
- pr_warning("read/write error %d:0x%04x (%s)\n",
- event->xRc, bevent->sub_result, err->msg);
- num_sect = blk_rq_sectors(req);
- }
- qlock = req->q->queue_lock;
- spin_lock_irqsave(qlock, irq_flags);
- viodasd_end_request(req, error, num_sect);
- spin_unlock_irqrestore(qlock, irq_flags);
-
- /* Finally, try to get more requests off of this device's queue */
- viodasd_restart_all_queues_starting_from(DEVICE_NO(d));
-
- return 0;
-}
-
-/* This routine handles incoming block LP events */
-static void handle_block_event(struct HvLpEvent *event)
-{
- struct vioblocklpevent *bevent = (struct vioblocklpevent *)event;
- struct viodasd_waitevent *pwe;
-
- if (event == NULL)
- /* Notification that a partition went away! */
- return;
- /* First, we should NEVER get an int here...only acks */
- if (hvlpevent_is_int(event)) {
- pr_warning("Yikes! got an int in viodasd event handler!\n");
- if (hvlpevent_need_ack(event)) {
- event->xRc = HvLpEvent_Rc_InvalidSubtype;
- HvCallEvent_ackLpEvent(event);
- }
- }
-
- switch (event->xSubtype & VIOMINOR_SUBTYPE_MASK) {
- case vioblockopen:
- /*
- * Handle a response to an open request. We get all the
- * disk information in the response, so update it. The
- * correlation token contains a pointer to a waitevent
- * structure that has a completion in it. update the
- * return code in the waitevent structure and post the
- * completion to wake up the guy who sent the request
- */
- pwe = (struct viodasd_waitevent *)event->xCorrelationToken;
- pwe->rc = event->xRc;
- pwe->sub_result = bevent->sub_result;
- if (event->xRc == HvLpEvent_Rc_Good) {
- const struct open_data *data = &bevent->u.open_data;
- struct viodasd_device *device =
- &viodasd_devices[bevent->disk];
- device->read_only =
- bevent->flags & vioblockflags_ro;
- device->size = data->disk_size;
- device->cylinders = data->cylinders;
- device->tracks = data->tracks;
- device->sectors = data->sectors;
- device->bytes_per_sector = data->bytes_per_sector;
- pwe->max_disk = data->max_disk;
- }
- complete(&pwe->com);
- break;
- case vioblockclose:
- break;
- case vioblockread:
- case vioblockwrite:
- viodasd_handle_read_write(bevent);
- break;
-
- default:
- pr_warning("invalid subtype!");
- if (hvlpevent_need_ack(event)) {
- event->xRc = HvLpEvent_Rc_InvalidSubtype;
- HvCallEvent_ackLpEvent(event);
- }
- }
-}
-
-/*
- * Get the driver to reprobe for more disks.
- */
-static ssize_t probe_disks(struct device_driver *drv, const char *buf,
- size_t count)
-{
- struct viodasd_device *d;
-
- for (d = viodasd_devices; d < &viodasd_devices[MAX_DISKNO]; d++) {
- if (d->disk == NULL)
- probe_disk(d);
- }
- return count;
-}
-static DRIVER_ATTR(probe, S_IWUSR, NULL, probe_disks);
-
-static int viodasd_probe(struct vio_dev *vdev, const struct vio_device_id *id)
-{
- struct viodasd_device *d = &viodasd_devices[vdev->unit_address];
-
- d->dev = &vdev->dev;
- if (!probe_disk(d))
- return -ENODEV;
- return 0;
-}
-
-static int viodasd_remove(struct vio_dev *vdev)
-{
- struct viodasd_device *d;
-
- d = &viodasd_devices[vdev->unit_address];
- if (d->disk) {
- del_gendisk(d->disk);
- blk_cleanup_queue(d->disk->queue);
- put_disk(d->disk);
- d->disk = NULL;
- }
- d->dev = NULL;
- return 0;
-}
-
-/**
- * viodasd_device_table: Used by vio.c to match devices that we
- * support.
- */
-static struct vio_device_id viodasd_device_table[] __devinitdata = {
- { "block", "IBM,iSeries-viodasd" },
- { "", "" }
-};
-MODULE_DEVICE_TABLE(vio, viodasd_device_table);
-
-static struct vio_driver viodasd_driver = {
- .id_table = viodasd_device_table,
- .probe = viodasd_probe,
- .remove = viodasd_remove,
- .driver = {
- .name = "viodasd",
- .owner = THIS_MODULE,
- }
-};
-
-static int need_delete_probe;
-
-/*
- * Initialize the whole device driver. Handle module and non-module
- * versions
- */
-static int __init viodasd_init(void)
-{
- int rc;
-
- if (!firmware_has_feature(FW_FEATURE_ISERIES)) {
- rc = -ENODEV;
- goto early_fail;
- }
-
- /* Try to open to our host lp */
- if (viopath_hostLp == HvLpIndexInvalid)
- vio_set_hostlp();
-
- if (viopath_hostLp == HvLpIndexInvalid) {
- pr_warning("invalid hosting partition\n");
- rc = -EIO;
- goto early_fail;
- }
-
- pr_info("vers " VIOD_VERS ", hosting partition %d\n", viopath_hostLp);
-
- /* register the block device */
- rc = register_blkdev(VIODASD_MAJOR, VIOD_GENHD_NAME);
- if (rc) {
- pr_warning("Unable to get major number %d for %s\n",
- VIODASD_MAJOR, VIOD_GENHD_NAME);
- goto early_fail;
- }
- /* Actually open the path to the hosting partition */
- rc = viopath_open(viopath_hostLp, viomajorsubtype_blockio,
- VIOMAXREQ + 2);
- if (rc) {
- pr_warning("error opening path to host partition %d\n",
- viopath_hostLp);
- goto unregister_blk;
- }
-
- /* Initialize our request handler */
- vio_setHandler(viomajorsubtype_blockio, handle_block_event);
-
- rc = vio_register_driver(&viodasd_driver);
- if (rc) {
- pr_warning("vio_register_driver failed\n");
- goto unset_handler;
- }
-
- /*
- * If this call fails, it just means that we cannot dynamically
- * add virtual disks, but the driver will still work fine for
- * all existing disk, so ignore the failure.
- */
- if (!driver_create_file(&viodasd_driver.driver, &driver_attr_probe))
- need_delete_probe = 1;
-
- return 0;
-
-unset_handler:
- vio_clearHandler(viomajorsubtype_blockio);
- viopath_close(viopath_hostLp, viomajorsubtype_blockio, VIOMAXREQ + 2);
-unregister_blk:
- unregister_blkdev(VIODASD_MAJOR, VIOD_GENHD_NAME);
-early_fail:
- return rc;
-}
-module_init(viodasd_init);
-
-void __exit viodasd_exit(void)
-{
- if (need_delete_probe)
- driver_remove_file(&viodasd_driver.driver, &driver_attr_probe);
- vio_unregister_driver(&viodasd_driver);
- vio_clearHandler(viomajorsubtype_blockio);
- viopath_close(viopath_hostLp, viomajorsubtype_blockio, VIOMAXREQ + 2);
- unregister_blkdev(VIODASD_MAJOR, VIOD_GENHD_NAME);
-}
-module_exit(viodasd_exit);
diff --git a/drivers/block/xd.c b/drivers/block/xd.c
index 51a972704db..ff540520bad 100644
--- a/drivers/block/xd.c
+++ b/drivers/block/xd.c
@@ -52,7 +52,6 @@
#include <linux/io.h>
#include <linux/gfp.h>
-#include <asm/system.h>
#include <asm/uaccess.h>
#include <asm/dma.h>
diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c
index 2f22874c0a3..98cbeba8cd5 100644
--- a/drivers/block/xen-blkfront.c
+++ b/drivers/block/xen-blkfront.c
@@ -1475,6 +1475,9 @@ static int __init xlblk_init(void)
if (!xen_domain())
return -ENODEV;
+ if (xen_hvm_domain() && !xen_platform_pci_unplug)
+ return -ENODEV;
+
if (register_blkdev(XENVBD_MAJOR, DEV_NAME)) {
printk(KERN_WARNING "xen_blk: can't get major %d with name %s\n",
XENVBD_MAJOR, DEV_NAME);
generated by cgit v1.2.3-18-g5258 (git 2.32.0) at 2025-01-02 11:55:11 +0000