aboutsummaryrefslogtreecommitdiffstats
path: root/drivers/block
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/block')
-rw-r--r--drivers/block/Kconfig3
-rw-r--r--drivers/block/brd.c48
-rw-r--r--drivers/block/drbd/drbd_bitmap.c2
-rw-r--r--drivers/block/drbd/drbd_req.c27
-rw-r--r--drivers/block/loop.c5
-rw-r--r--drivers/block/mtip32xx/mtip32xx.c490
-rw-r--r--drivers/block/mtip32xx/mtip32xx.h10
-rw-r--r--drivers/block/nbd.c14
-rw-r--r--drivers/block/rbd.c366
-rw-r--r--drivers/block/virtio_blk.c10
-rw-r--r--drivers/block/xen-blkback/xenbus.c8
-rw-r--r--drivers/block/zram/zram_drv.c577
-rw-r--r--drivers/block/zram/zram_drv.h6
13 files changed, 738 insertions, 828 deletions
diff --git a/drivers/block/Kconfig b/drivers/block/Kconfig
index 19df4918e37e..8ddc98279c8f 100644
--- a/drivers/block/Kconfig
+++ b/drivers/block/Kconfig
@@ -219,7 +219,7 @@ config BLK_DEV_LOOP
To use the loop device, you need the losetup utility, found in the
util-linux package, see
- <ftp://ftp.kernel.org/pub/linux/utils/util-linux/>.
+ <https://www.kernel.org/pub/linux/utils/util-linux/>.
The loop device driver can also be used to "hide" a file system in
a disk partition, floppy, or regular file, either using encryption
@@ -323,6 +323,7 @@ config BLK_DEV_SX8
config BLK_DEV_RAM
tristate "RAM block device support"
+ select DAX if BLK_DEV_RAM_DAX
---help---
Saying Y here will allow you to use a portion of your RAM memory as
a block device, so that you can make file systems on it, read and
diff --git a/drivers/block/brd.c b/drivers/block/brd.c
index 4ec84d504780..57b574f2f66a 100644
--- a/drivers/block/brd.c
+++ b/drivers/block/brd.c
@@ -21,6 +21,7 @@
#include <linux/slab.h>
#ifdef CONFIG_BLK_DEV_RAM_DAX
#include <linux/pfn_t.h>
+#include <linux/dax.h>
#endif
#include <linux/uaccess.h>
@@ -41,6 +42,9 @@ struct brd_device {
struct request_queue *brd_queue;
struct gendisk *brd_disk;
+#ifdef CONFIG_BLK_DEV_RAM_DAX
+ struct dax_device *dax_dev;
+#endif
struct list_head brd_list;
/*
@@ -326,30 +330,38 @@ static int brd_rw_page(struct block_device *bdev, sector_t sector,
}
#ifdef CONFIG_BLK_DEV_RAM_DAX
-static long brd_direct_access(struct block_device *bdev, sector_t sector,
- void **kaddr, pfn_t *pfn, long size)
+static long __brd_direct_access(struct brd_device *brd, pgoff_t pgoff,
+ long nr_pages, void **kaddr, pfn_t *pfn)
{
- struct brd_device *brd = bdev->bd_disk->private_data;
struct page *page;
if (!brd)
return -ENODEV;
- page = brd_insert_page(brd, sector);
+ page = brd_insert_page(brd, PFN_PHYS(pgoff) / 512);
if (!page)
return -ENOSPC;
*kaddr = page_address(page);
*pfn = page_to_pfn_t(page);
- return PAGE_SIZE;
+ return 1;
}
-#else
-#define brd_direct_access NULL
+
+static long brd_dax_direct_access(struct dax_device *dax_dev,
+ pgoff_t pgoff, long nr_pages, void **kaddr, pfn_t *pfn)
+{
+ struct brd_device *brd = dax_get_private(dax_dev);
+
+ return __brd_direct_access(brd, pgoff, nr_pages, kaddr, pfn);
+}
+
+static const struct dax_operations brd_dax_ops = {
+ .direct_access = brd_dax_direct_access,
+};
#endif
static const struct block_device_operations brd_fops = {
.owner = THIS_MODULE,
.rw_page = brd_rw_page,
- .direct_access = brd_direct_access,
};
/*
@@ -415,9 +427,6 @@ static struct brd_device *brd_alloc(int i)
* is harmless)
*/
blk_queue_physical_block_size(brd->brd_queue, PAGE_SIZE);
-#ifdef CONFIG_BLK_DEV_RAM_DAX
- queue_flag_set_unlocked(QUEUE_FLAG_DAX, brd->brd_queue);
-#endif
disk = brd->brd_disk = alloc_disk(max_part);
if (!disk)
goto out_free_queue;
@@ -430,8 +439,21 @@ static struct brd_device *brd_alloc(int i)
sprintf(disk->disk_name, "ram%d", i);
set_capacity(disk, rd_size * 2);
+#ifdef CONFIG_BLK_DEV_RAM_DAX
+ queue_flag_set_unlocked(QUEUE_FLAG_DAX, brd->brd_queue);
+ brd->dax_dev = alloc_dax(brd, disk->disk_name, &brd_dax_ops);
+ if (!brd->dax_dev)
+ goto out_free_inode;
+#endif
+
+
return brd;
+#ifdef CONFIG_BLK_DEV_RAM_DAX
+out_free_inode:
+ kill_dax(brd->dax_dev);
+ put_dax(brd->dax_dev);
+#endif
out_free_queue:
blk_cleanup_queue(brd->brd_queue);
out_free_dev:
@@ -471,6 +493,10 @@ out:
static void brd_del_one(struct brd_device *brd)
{
list_del(&brd->brd_list);
+#ifdef CONFIG_BLK_DEV_RAM_DAX
+ kill_dax(brd->dax_dev);
+ put_dax(brd->dax_dev);
+#endif
del_gendisk(brd->brd_disk);
brd_free(brd);
}
diff --git a/drivers/block/drbd/drbd_bitmap.c b/drivers/block/drbd/drbd_bitmap.c
index dece26f119d4..a804a4107fbc 100644
--- a/drivers/block/drbd/drbd_bitmap.c
+++ b/drivers/block/drbd/drbd_bitmap.c
@@ -409,7 +409,7 @@ static struct page **bm_realloc_pages(struct drbd_bitmap *b, unsigned long want)
new_pages = kzalloc(bytes, GFP_NOIO | __GFP_NOWARN);
if (!new_pages) {
new_pages = __vmalloc(bytes,
- GFP_NOIO | __GFP_HIGHMEM | __GFP_ZERO,
+ GFP_NOIO | __GFP_ZERO,
PAGE_KERNEL);
if (!new_pages)
return NULL;
diff --git a/drivers/block/drbd/drbd_req.c b/drivers/block/drbd/drbd_req.c
index b5730e17b455..656624314f0d 100644
--- a/drivers/block/drbd/drbd_req.c
+++ b/drivers/block/drbd/drbd_req.c
@@ -315,24 +315,32 @@ void drbd_req_complete(struct drbd_request *req, struct bio_and_error *m)
}
/* still holds resource->req_lock */
-static int drbd_req_put_completion_ref(struct drbd_request *req, struct bio_and_error *m, int put)
+static void drbd_req_put_completion_ref(struct drbd_request *req, struct bio_and_error *m, int put)
{
struct drbd_device *device = req->device;
D_ASSERT(device, m || (req->rq_state & RQ_POSTPONED));
+ if (!put)
+ return;
+
if (!atomic_sub_and_test(put, &req->completion_ref))
- return 0;
+ return;
drbd_req_complete(req, m);
+ /* local completion may still come in later,
+ * we need to keep the req object around. */
+ if (req->rq_state & RQ_LOCAL_ABORTED)
+ return;
+
if (req->rq_state & RQ_POSTPONED) {
/* don't destroy the req object just yet,
* but queue it for retry */
drbd_restart_request(req);
- return 0;
+ return;
}
- return 1;
+ kref_put(&req->kref, drbd_req_destroy);
}
static void set_if_null_req_next(struct drbd_peer_device *peer_device, struct drbd_request *req)
@@ -519,12 +527,8 @@ static void mod_rq_state(struct drbd_request *req, struct bio_and_error *m,
if (req->i.waiting)
wake_up(&device->misc_wait);
- if (c_put) {
- if (drbd_req_put_completion_ref(req, m, c_put))
- kref_put(&req->kref, drbd_req_destroy);
- } else {
- kref_put(&req->kref, drbd_req_destroy);
- }
+ drbd_req_put_completion_ref(req, m, c_put);
+ kref_put(&req->kref, drbd_req_destroy);
}
static void drbd_report_io_error(struct drbd_device *device, struct drbd_request *req)
@@ -1366,8 +1370,7 @@ nodata:
}
out:
- if (drbd_req_put_completion_ref(req, &m, 1))
- kref_put(&req->kref, drbd_req_destroy);
+ drbd_req_put_completion_ref(req, &m, 1);
spin_unlock_irq(&resource->req_lock);
/* Even though above is a kref_put(), this is safe.
diff --git a/drivers/block/loop.c b/drivers/block/loop.c
index 994403efee19..28d932906f24 100644
--- a/drivers/block/loop.c
+++ b/drivers/block/loop.c
@@ -1697,9 +1697,8 @@ static void loop_queue_work(struct kthread_work *work)
loop_handle_cmd(cmd);
}
-static int loop_init_request(void *data, struct request *rq,
- unsigned int hctx_idx, unsigned int request_idx,
- unsigned int numa_node)
+static int loop_init_request(struct blk_mq_tag_set *set, struct request *rq,
+ unsigned int hctx_idx, unsigned int numa_node)
{
struct loop_cmd *cmd = blk_mq_rq_to_pdu(rq);
diff --git a/drivers/block/mtip32xx/mtip32xx.c b/drivers/block/mtip32xx/mtip32xx.c
index 02804cc79d82..3a779a4f5653 100644
--- a/drivers/block/mtip32xx/mtip32xx.c
+++ b/drivers/block/mtip32xx/mtip32xx.c
@@ -195,7 +195,7 @@ static struct mtip_cmd *mtip_get_int_command(struct driver_data *dd)
if (mtip_check_surprise_removal(dd->pdev))
return NULL;
- rq = blk_mq_alloc_request(dd->queue, 0, BLK_MQ_REQ_RESERVED);
+ rq = blk_mq_alloc_request(dd->queue, REQ_OP_DRV_IN, BLK_MQ_REQ_RESERVED);
if (IS_ERR(rq))
return NULL;
@@ -205,66 +205,12 @@ static struct mtip_cmd *mtip_get_int_command(struct driver_data *dd)
return blk_mq_rq_to_pdu(rq);
}
-static void mtip_put_int_command(struct driver_data *dd, struct mtip_cmd *cmd)
-{
- blk_put_request(blk_mq_rq_from_pdu(cmd));
-}
-
-/*
- * Once we add support for one hctx per mtip group, this will change a bit
- */
-static struct request *mtip_rq_from_tag(struct driver_data *dd,
- unsigned int tag)
-{
- struct blk_mq_hw_ctx *hctx = dd->queue->queue_hw_ctx[0];
-
- return blk_mq_tag_to_rq(hctx->tags, tag);
-}
-
static struct mtip_cmd *mtip_cmd_from_tag(struct driver_data *dd,
unsigned int tag)
{
- struct request *rq = mtip_rq_from_tag(dd, tag);
-
- return blk_mq_rq_to_pdu(rq);
-}
-
-/*
- * IO completion function.
- *
- * This completion function is called by the driver ISR when a
- * command that was issued by the kernel completes. It first calls the
- * asynchronous completion function which normally calls back into the block
- * layer passing the asynchronous callback data, then unmaps the
- * scatter list associated with the completed command, and finally
- * clears the allocated bit associated with the completed command.
- *
- * @port Pointer to the port data structure.
- * @tag Tag of the command.
- * @data Pointer to driver_data.
- * @status Completion status.
- *
- * return value
- * None
- */
-static void mtip_async_complete(struct mtip_port *port,
- int tag, struct mtip_cmd *cmd, int status)
-{
- struct driver_data *dd = port->dd;
- struct request *rq;
-
- if (unlikely(!dd) || unlikely(!port))
- return;
-
- if (unlikely(status == PORT_IRQ_TF_ERR)) {
- dev_warn(&port->dd->pdev->dev,
- "Command tag %d failed due to TFE\n", tag);
- }
-
- rq = mtip_rq_from_tag(dd, tag);
+ struct blk_mq_hw_ctx *hctx = dd->queue->queue_hw_ctx[0];
- cmd->status = status;
- blk_mq_complete_request(rq);
+ return blk_mq_rq_to_pdu(blk_mq_tag_to_rq(hctx->tags, tag));
}
/*
@@ -581,43 +527,19 @@ static void print_tags(struct driver_data *dd,
"%d command(s) %s: tagmap [%s]", cnt, msg, tagmap);
}
-/*
- * Internal command completion callback function.
- *
- * This function is normally called by the driver ISR when an internal
- * command completed. This function signals the command completion by
- * calling complete().
- *
- * @port Pointer to the port data structure.
- * @tag Tag of the command that has completed.
- * @data Pointer to a completion structure.
- * @status Completion status.
- *
- * return value
- * None
- */
-static void mtip_completion(struct mtip_port *port,
- int tag, struct mtip_cmd *command, int status)
-{
- struct completion *waiting = command->comp_data;
- if (unlikely(status == PORT_IRQ_TF_ERR))
- dev_warn(&port->dd->pdev->dev,
- "Internal command %d completed with TFE\n", tag);
-
- command->comp_func = NULL;
- command->comp_data = NULL;
- complete(waiting);
-}
-
-static void mtip_null_completion(struct mtip_port *port,
- int tag, struct mtip_cmd *command, int status)
-{
-}
-
static int mtip_read_log_page(struct mtip_port *port, u8 page, u16 *buffer,
dma_addr_t buffer_dma, unsigned int sectors);
static int mtip_get_smart_attr(struct mtip_port *port, unsigned int id,
struct smart_attr *attrib);
+
+static void mtip_complete_command(struct mtip_cmd *cmd, int status)
+{
+ struct request *req = blk_mq_rq_from_pdu(cmd);
+
+ cmd->status = status;
+ blk_mq_complete_request(req);
+}
+
/*
* Handle an error.
*
@@ -646,11 +568,7 @@ static void mtip_handle_tfe(struct driver_data *dd)
if (test_bit(MTIP_PF_IC_ACTIVE_BIT, &port->flags)) {
cmd = mtip_cmd_from_tag(dd, MTIP_TAG_INTERNAL);
dbg_printk(MTIP_DRV_NAME " TFE for the internal command\n");
-
- if (cmd->comp_data && cmd->comp_func) {
- cmd->comp_func(port, MTIP_TAG_INTERNAL,
- cmd, PORT_IRQ_TF_ERR);
- }
+ mtip_complete_command(cmd, -EIO);
return;
}
@@ -677,19 +595,9 @@ static void mtip_handle_tfe(struct driver_data *dd)
continue;
cmd = mtip_cmd_from_tag(dd, tag);
- if (likely(cmd->comp_func)) {
- set_bit(tag, tagaccum);
- cmd_cnt++;
- cmd->comp_func(port, tag, cmd, 0);
- } else {
- dev_err(&port->dd->pdev->dev,
- "Missing completion func for tag %d",
- tag);
- if (mtip_check_surprise_removal(dd->pdev)) {
- /* don't proceed further */
- return;
- }
- }
+ mtip_complete_command(cmd, 0);
+ set_bit(tag, tagaccum);
+ cmd_cnt++;
}
}
@@ -759,10 +667,7 @@ static void mtip_handle_tfe(struct driver_data *dd)
tag,
fail_reason != NULL ?
fail_reason : "unknown");
- if (cmd->comp_func) {
- cmd->comp_func(port, tag,
- cmd, -ENODATA);
- }
+ mtip_complete_command(cmd, -ENODATA);
continue;
}
}
@@ -785,12 +690,7 @@ static void mtip_handle_tfe(struct driver_data *dd)
dev_warn(&port->dd->pdev->dev,
"retiring tag %d\n", tag);
- if (cmd->comp_func)
- cmd->comp_func(port, tag, cmd, PORT_IRQ_TF_ERR);
- else
- dev_warn(&port->dd->pdev->dev,
- "Bad completion for tag %d\n",
- tag);
+ mtip_complete_command(cmd, -EIO);
}
}
print_tags(dd, "reissued (TFE)", tagaccum, cmd_cnt);
@@ -823,18 +723,7 @@ static inline void mtip_workq_sdbfx(struct mtip_port *port, int group,
continue;
command = mtip_cmd_from_tag(dd, tag);
- if (likely(command->comp_func))
- command->comp_func(port, tag, command, 0);
- else {
- dev_dbg(&dd->pdev->dev,
- "Null completion for tag %d",
- tag);
-
- if (mtip_check_surprise_removal(
- dd->pdev)) {
- return;
- }
- }
+ mtip_complete_command(command, 0);
}
completed >>= 1;
}
@@ -852,16 +741,13 @@ static inline void mtip_process_legacy(struct driver_data *dd, u32 port_stat)
struct mtip_port *port = dd->port;
struct mtip_cmd *cmd = mtip_cmd_from_tag(dd, MTIP_TAG_INTERNAL);
- if (test_bit(MTIP_PF_IC_ACTIVE_BIT, &port->flags) &&
- (cmd != NULL) && !(readl(port->cmd_issue[MTIP_TAG_INTERNAL])
- & (1 << MTIP_TAG_INTERNAL))) {
- if (cmd->comp_func) {
- cmd->comp_func(port, MTIP_TAG_INTERNAL, cmd, 0);
- return;
- }
- }
+ if (test_bit(MTIP_PF_IC_ACTIVE_BIT, &port->flags) && cmd) {
+ int group = MTIP_TAG_INDEX(MTIP_TAG_INTERNAL);
+ int status = readl(port->cmd_issue[group]);
- return;
+ if (!(status & (1 << MTIP_TAG_BIT(MTIP_TAG_INTERNAL))))
+ mtip_complete_command(cmd, 0);
+ }
}
/*
@@ -869,7 +755,6 @@ static inline void mtip_process_legacy(struct driver_data *dd, u32 port_stat)
*/
static inline void mtip_process_errors(struct driver_data *dd, u32 port_stat)
{
-
if (unlikely(port_stat & PORT_IRQ_CONNECT)) {
dev_warn(&dd->pdev->dev,
"Clearing PxSERR.DIAG.x\n");
@@ -996,8 +881,7 @@ static irqreturn_t mtip_irq_handler(int irq, void *instance)
static void mtip_issue_non_ncq_command(struct mtip_port *port, int tag)
{
- writel(1 << MTIP_TAG_BIT(tag),
- port->cmd_issue[MTIP_TAG_INDEX(tag)]);
+ writel(1 << MTIP_TAG_BIT(tag), port->cmd_issue[MTIP_TAG_INDEX(tag)]);
}
static bool mtip_pause_ncq(struct mtip_port *port,
@@ -1035,53 +919,53 @@ static bool mtip_pause_ncq(struct mtip_port *port,
return false;
}
+static bool mtip_commands_active(struct mtip_port *port)
+{
+ unsigned int active;
+ unsigned int n;
+
+ /*
+ * Ignore s_active bit 0 of array element 0.
+ * This bit will always be set
+ */
+ active = readl(port->s_active[0]) & 0xFFFFFFFE;
+ for (n = 1; n < port->dd->slot_groups; n++)
+ active |= readl(port->s_active[n]);
+
+ return active != 0;
+}
+
/*
* Wait for port to quiesce
*
* @port Pointer to port data structure
* @timeout Max duration to wait (ms)
- * @atomic gfp_t flag to indicate blockable context or not
*
* return value
* 0 Success
* -EBUSY Commands still active
*/
-static int mtip_quiesce_io(struct mtip_port *port, unsigned long timeout,
- gfp_t atomic)
+static int mtip_quiesce_io(struct mtip_port *port, unsigned long timeout)
{
unsigned long to;
- unsigned int n;
- unsigned int active = 1;
+ bool active = true;
blk_mq_stop_hw_queues(port->dd->queue);
to = jiffies + msecs_to_jiffies(timeout);
do {
if (test_bit(MTIP_PF_SVC_THD_ACTIVE_BIT, &port->flags) &&
- test_bit(MTIP_PF_ISSUE_CMDS_BIT, &port->flags) &&
- atomic == GFP_KERNEL) {
+ test_bit(MTIP_PF_ISSUE_CMDS_BIT, &port->flags)) {
msleep(20);
continue; /* svc thd is actively issuing commands */
}
- if (atomic == GFP_KERNEL)
- msleep(100);
- else {
- cpu_relax();
- udelay(100);
- }
+ msleep(100);
if (mtip_check_surprise_removal(port->dd->pdev))
goto err_fault;
- /*
- * Ignore s_active bit 0 of array element 0.
- * This bit will always be set
- */
- active = readl(port->s_active[0]) & 0xFFFFFFFE;
- for (n = 1; n < port->dd->slot_groups; n++)
- active |= readl(port->s_active[n]);
-
+ active = mtip_commands_active(port);
if (!active)
break;
} while (time_before(jiffies, to));
@@ -1093,6 +977,13 @@ err_fault:
return -EFAULT;
}
+struct mtip_int_cmd {
+ int fis_len;
+ dma_addr_t buffer;
+ int buf_len;
+ u32 opts;
+};
+
/*
* Execute an internal command and wait for the completion.
*
@@ -1117,13 +1008,17 @@ static int mtip_exec_internal_command(struct mtip_port *port,
dma_addr_t buffer,
int buf_len,
u32 opts,
- gfp_t atomic,
unsigned long timeout)
{
- struct mtip_cmd_sg *command_sg;
- DECLARE_COMPLETION_ONSTACK(wait);
struct mtip_cmd *int_cmd;
struct driver_data *dd = port->dd;
+ struct request *rq;
+ struct mtip_int_cmd icmd = {
+ .fis_len = fis_len,
+ .buffer = buffer,
+ .buf_len = buf_len,
+ .opts = opts
+ };
int rv = 0;
unsigned long start;
@@ -1138,6 +1033,8 @@ static int mtip_exec_internal_command(struct mtip_port *port,
dbg_printk(MTIP_DRV_NAME "Unable to allocate tag for PIO cmd\n");
return -EFAULT;
}
+ rq = blk_mq_rq_from_pdu(int_cmd);
+ rq->special = &icmd;
set_bit(MTIP_PF_IC_ACTIVE_BIT, &port->flags);
@@ -1146,135 +1043,60 @@ static int mtip_exec_internal_command(struct mtip_port *port,
clear_bit(MTIP_PF_DM_ACTIVE_BIT, &port->flags);
- if (atomic == GFP_KERNEL) {
- if (fis->command != ATA_CMD_STANDBYNOW1) {
- /* wait for io to complete if non atomic */
- if (mtip_quiesce_io(port,
- MTIP_QUIESCE_IO_TIMEOUT_MS, atomic) < 0) {
- dev_warn(&dd->pdev->dev,
- "Failed to quiesce IO\n");
- mtip_put_int_command(dd, int_cmd);
- clear_bit(MTIP_PF_IC_ACTIVE_BIT, &port->flags);
- wake_up_interruptible(&port->svc_wait);
- return -EBUSY;
- }
+ if (fis->command != ATA_CMD_STANDBYNOW1) {
+ /* wait for io to complete if non atomic */
+ if (mtip_quiesce_io(port, MTIP_QUIESCE_IO_TIMEOUT_MS) < 0) {
+ dev_warn(&dd->pdev->dev, "Failed to quiesce IO\n");
+ blk_mq_free_request(rq);
+ clear_bit(MTIP_PF_IC_ACTIVE_BIT, &port->flags);
+ wake_up_interruptible(&port->svc_wait);
+ return -EBUSY;
}
-
- /* Set the completion function and data for the command. */
- int_cmd->comp_data = &wait;
- int_cmd->comp_func = mtip_completion;
-
- } else {
- /* Clear completion - we're going to poll */
- int_cmd->comp_data = NULL;
- int_cmd->comp_func = mtip_null_completion;
}
/* Copy the command to the command table */
memcpy(int_cmd->command, fis, fis_len*4);
- /* Populate the SG list */
- int_cmd->command_header->opts =
- __force_bit2int cpu_to_le32(opts | fis_len);
- if (buf_len) {
- command_sg = int_cmd->command + AHCI_CMD_TBL_HDR_SZ;
-
- command_sg->info =
- __force_bit2int cpu_to_le32((buf_len-1) & 0x3FFFFF);
- command_sg->dba =
- __force_bit2int cpu_to_le32(buffer & 0xFFFFFFFF);
- command_sg->dba_upper =
- __force_bit2int cpu_to_le32((buffer >> 16) >> 16);
-
- int_cmd->command_header->opts |=
- __force_bit2int cpu_to_le32((1 << 16));
- }
-
- /* Populate the command header */
- int_cmd->command_header->byte_count = 0;
-
start = jiffies;
+ rq->timeout = timeout;
- /* Issue the command to the hardware */
- mtip_issue_non_ncq_command(port, MTIP_TAG_INTERNAL);
-
- if (atomic == GFP_KERNEL) {
- /* Wait for the command to complete or timeout. */
- if ((rv = wait_for_completion_interruptible_timeout(
- &wait,
- msecs_to_jiffies(timeout))) <= 0) {
-
- if (rv == -ERESTARTSYS) { /* interrupted */
- dev_err(&dd->pdev->dev,
- "Internal command [%02X] was interrupted after %u ms\n",
- fis->command,
- jiffies_to_msecs(jiffies - start));
- rv = -EINTR;
- goto exec_ic_exit;
- } else if (rv == 0) /* timeout */
- dev_err(&dd->pdev->dev,
- "Internal command did not complete [%02X] within timeout of %lu ms\n",
- fis->command, timeout);
- else
- dev_err(&dd->pdev->dev,
- "Internal command [%02X] wait returned code [%d] after %lu ms - unhandled\n",
- fis->command, rv, timeout);
-
- if (mtip_check_surprise_removal(dd->pdev) ||
- test_bit(MTIP_DDF_REMOVE_PENDING_BIT,
- &dd->dd_flag)) {
- dev_err(&dd->pdev->dev,
- "Internal command [%02X] wait returned due to SR\n",
- fis->command);
- rv = -ENXIO;
- goto exec_ic_exit;
- }
- mtip_device_reset(dd); /* recover from timeout issue */
- rv = -EAGAIN;
+ /* insert request and run queue */
+ blk_execute_rq(rq->q, NULL, rq, true);
+
+ rv = int_cmd->status;
+ if (rv < 0) {
+ if (rv == -ERESTARTSYS) { /* interrupted */
+ dev_err(&dd->pdev->dev,
+ "Internal command [%02X] was interrupted after %u ms\n",
+ fis->command,
+ jiffies_to_msecs(jiffies - start));
+ rv = -EINTR;
goto exec_ic_exit;
- }
- } else {
- u32 hba_stat, port_stat;
-
- /* Spin for <timeout> checking if command still outstanding */
- timeout = jiffies + msecs_to_jiffies(timeout);
- while ((readl(port->cmd_issue[MTIP_TAG_INTERNAL])
- & (1 << MTIP_TAG_INTERNAL))
- && time_before(jiffies, timeout)) {
- if (mtip_check_surprise_removal(dd->pdev)) {
- rv = -ENXIO;
- goto exec_ic_exit;
- }
- if ((fis->command != ATA_CMD_STANDBYNOW1) &&
- test_bit(MTIP_DDF_REMOVE_PENDING_BIT,
- &dd->dd_flag)) {
- rv = -ENXIO;
- goto exec_ic_exit;
- }
- port_stat = readl(port->mmio + PORT_IRQ_STAT);
- if (!port_stat)
- continue;
+ } else if (rv == 0) /* timeout */
+ dev_err(&dd->pdev->dev,
+ "Internal command did not complete [%02X] within timeout of %lu ms\n",
+ fis->command, timeout);
+ else
+ dev_err(&dd->pdev->dev,
+ "Internal command [%02X] wait returned code [%d] after %lu ms - unhandled\n",
+ fis->command, rv, timeout);
- if (port_stat & PORT_IRQ_ERR) {
- dev_err(&dd->pdev->dev,
- "Internal command [%02X] failed\n",
- fis->command);
- mtip_device_reset(dd);
- rv = -EIO;
- goto exec_ic_exit;
- } else {
- writel(port_stat, port->mmio + PORT_IRQ_STAT);
- hba_stat = readl(dd->mmio + HOST_IRQ_STAT);
- if (hba_stat)
- writel(hba_stat,
- dd->mmio + HOST_IRQ_STAT);
- }
- break;
+ if (mtip_check_surprise_removal(dd->pdev) ||
+ test_bit(MTIP_DDF_REMOVE_PENDING_BIT,
+ &dd->dd_flag)) {
+ dev_err(&dd->pdev->dev,
+ "Internal command [%02X] wait returned due to SR\n",
+ fis->command);
+ rv = -ENXIO;
+ goto exec_ic_exit;
}
+ mtip_device_reset(dd); /* recover from timeout issue */
+ rv = -EAGAIN;
+ goto exec_ic_exit;
}
- if (readl(port->cmd_issue[MTIP_TAG_INTERNAL])
- & (1 << MTIP_TAG_INTERNAL)) {
+ if (readl(port->cmd_issue[MTIP_TAG_INDEX(MTIP_TAG_INTERNAL)])
+ & (1 << MTIP_TAG_BIT(MTIP_TAG_INTERNAL))) {
rv = -ENXIO;
if (!test_bit(MTIP_DDF_REMOVE_PENDING_BIT, &dd->dd_flag)) {
mtip_device_reset(dd);
@@ -1283,7 +1105,7 @@ static int mtip_exec_internal_command(struct mtip_port *port,
}
exec_ic_exit:
/* Clear the allocated and active bits for the internal command. */
- mtip_put_int_command(dd, int_cmd);
+ blk_mq_free_request(rq);
clear_bit(MTIP_PF_IC_ACTIVE_BIT, &port->flags);
if (rv >= 0 && mtip_pause_ncq(port, fis)) {
/* NCQ paused */
@@ -1391,7 +1213,6 @@ static int mtip_get_identify(struct mtip_port *port, void __user *user_buffer)
port->identify_dma,
sizeof(u16) * ATA_ID_WORDS,
0,
- GFP_KERNEL,
MTIP_INT_CMD_TIMEOUT_MS)
< 0) {
rv = -1;
@@ -1477,7 +1298,6 @@ static int mtip_standby_immediate(struct mtip_port *port)
0,
0,
0,
- GFP_ATOMIC,
timeout);
dbg_printk(MTIP_DRV_NAME "Time taken to complete standby cmd: %d ms\n",
jiffies_to_msecs(jiffies - start));
@@ -1523,7 +1343,6 @@ static int mtip_read_log_page(struct mtip_port *port, u8 page, u16 *buffer,
buffer_dma,
sectors * ATA_SECT_SIZE,
0,
- GFP_ATOMIC,
MTIP_INT_CMD_TIMEOUT_MS);
}
@@ -1558,7 +1377,6 @@ static int mtip_get_smart_data(struct mtip_port *port, u8 *buffer,
buffer_dma,
ATA_SECT_SIZE,
0,
- GFP_ATOMIC,
15000);
}
@@ -1686,7 +1504,6 @@ static int mtip_send_trim(struct driver_data *dd, unsigned int lba,
dma_addr,
ATA_SECT_SIZE,
0,
- GFP_KERNEL,
MTIP_TRIM_TIMEOUT_MS) < 0)
rv = -EIO;
@@ -1850,7 +1667,6 @@ static int exec_drive_task(struct mtip_port *port, u8 *command)
0,
0,
0,
- GFP_KERNEL,
to) < 0) {
return -1;
}
@@ -1946,7 +1762,6 @@ static int exec_drive_command(struct mtip_port *port, u8 *command,
(xfer_sz ? dma_addr : 0),
(xfer_sz ? ATA_SECT_SIZE * xfer_sz : 0),
0,
- GFP_KERNEL,
to)
< 0) {
rv = -EFAULT;
@@ -2189,7 +2004,6 @@ static int exec_drive_taskfile(struct driver_data *dd,
dma_buffer,
transfer_size,
0,
- GFP_KERNEL,
timeout) < 0) {
err = -EIO;
goto abort;
@@ -2446,12 +2260,6 @@ static void mtip_hw_submit_io(struct driver_data *dd, struct request *rq,
(nents << 16) | 5 | AHCI_CMD_PREFETCH);
command->command_header->byte_count = 0;
- /*
- * Set the completion function and data for the command
- * within this layer.
- */
- command->comp_data = dd;
- command->comp_func = mtip_async_complete;
command->direction = dma_dir;
/*
@@ -3825,6 +3633,42 @@ static bool mtip_check_unal_depth(struct blk_mq_hw_ctx *hctx,
return false;
}
+static int mtip_issue_reserved_cmd(struct blk_mq_hw_ctx *hctx,
+ struct request *rq)
+{
+ struct driver_data *dd = hctx->queue->queuedata;
+ struct mtip_int_cmd *icmd = rq->special;
+ struct mtip_cmd *cmd = blk_mq_rq_to_pdu(rq);
+ struct mtip_cmd_sg *command_sg;
+
+ if (mtip_commands_active(dd->port))
+ return BLK_MQ_RQ_QUEUE_BUSY;
+
+ /* Populate the SG list */
+ cmd->command_header->opts =
+ __force_bit2int cpu_to_le32(icmd->opts | icmd->fis_len);
+ if (icmd->buf_len) {
+ command_sg = cmd->command + AHCI_CMD_TBL_HDR_SZ;
+
+ command_sg->info =
+ __force_bit2int cpu_to_le32((icmd->buf_len-1) & 0x3FFFFF);
+ command_sg->dba =
+ __force_bit2int cpu_to_le32(icmd->buffer & 0xFFFFFFFF);
+ command_sg->dba_upper =
+ __force_bit2int cpu_to_le32((icmd->buffer >> 16) >> 16);
+
+ cmd->command_header->opts |=
+ __force_bit2int cpu_to_le32((1 << 16));
+ }
+
+ /* Populate the command header */
+ cmd->command_header->byte_count = 0;
+
+ blk_mq_start_request(rq);
+ mtip_issue_non_ncq_command(dd->port, rq->tag);
+ return BLK_MQ_RQ_QUEUE_OK;
+}
+
static int mtip_queue_rq(struct blk_mq_hw_ctx *hctx,
const struct blk_mq_queue_data *bd)
{
@@ -3833,6 +3677,9 @@ static int mtip_queue_rq(struct blk_mq_hw_ctx *hctx,
mtip_init_cmd_header(rq);
+ if (blk_rq_is_passthrough(rq))
+ return mtip_issue_reserved_cmd(hctx, rq);
+
if (unlikely(mtip_check_unal_depth(hctx, rq)))
return BLK_MQ_RQ_QUEUE_BUSY;
@@ -3845,10 +3692,10 @@ static int mtip_queue_rq(struct blk_mq_hw_ctx *hctx,
return BLK_MQ_RQ_QUEUE_ERROR;
}
-static void mtip_free_cmd(void *data, struct request *rq,
- unsigned int hctx_idx, unsigned int request_idx)
+static void mtip_free_cmd(struct blk_mq_tag_set *set, struct request *rq,
+ unsigned int hctx_idx)
{
- struct driver_data *dd = data;
+ struct driver_data *dd = set->driver_data;
struct mtip_cmd *cmd = blk_mq_rq_to_pdu(rq);
if (!cmd->command)
@@ -3858,20 +3705,12 @@ static void mtip_free_cmd(void *data, struct request *rq,
cmd->command, cmd->command_dma);
}
-static int mtip_init_cmd(void *data, struct request *rq, unsigned int hctx_idx,
- unsigned int request_idx, unsigned int numa_node)
+static int mtip_init_cmd(struct blk_mq_tag_set *set, struct request *rq,
+ unsigned int hctx_idx, unsigned int numa_node)
{
- struct driver_data *dd = data;
+ struct driver_data *dd = set->driver_data;
struct mtip_cmd *cmd = blk_mq_rq_to_pdu(rq);
- /*
- * For flush requests, request_idx starts at the end of the
- * tag space. Since we don't support FLUSH/FUA, simply return
- * 0 as there's nothing to be done.
- */
- if (request_idx >= MTIP_MAX_COMMAND_SLOTS)
- return 0;
-
cmd->command = dmam_alloc_coherent(&dd->pdev->dev, CMD_DMA_ALLOC_SZ,
&cmd->command_dma, GFP_KERNEL);
if (!cmd->command)
@@ -3888,8 +3727,12 @@ static enum blk_eh_timer_return mtip_cmd_timeout(struct request *req,
{
struct driver_data *dd = req->q->queuedata;
- if (reserved)
- goto exit_handler;
+ if (reserved) {
+ struct mtip_cmd *cmd = blk_mq_rq_to_pdu(req);
+
+ cmd->status = -ETIME;
+ return BLK_EH_HANDLED;
+ }
if (test_bit(req->tag, dd->port->cmds_to_issue))
goto exit_handler;
@@ -3982,7 +3825,7 @@ static int mtip_block_initialize(struct driver_data *dd)
dd->tags.reserved_tags = 1;
dd->tags.cmd_size = sizeof(struct mtip_cmd);
dd->tags.numa_node = dd->numa_node;
- dd->tags.flags = BLK_MQ_F_SHOULD_MERGE | BLK_MQ_F_NO_SCHED;
+ dd->tags.flags = BLK_MQ_F_SHOULD_MERGE;
dd->tags.driver_data = dd;
dd->tags.timeout = MTIP_NCQ_CMD_TIMEOUT_MS;
@@ -4116,20 +3959,10 @@ protocol_init_error:
static void mtip_no_dev_cleanup(struct request *rq, void *data, bool reserv)
{
- struct driver_data *dd = (struct driver_data *)data;
- struct mtip_cmd *cmd;
-
- if (likely(!reserv)) {
- cmd = blk_mq_rq_to_pdu(rq);
- cmd->status = -ENODEV;
- blk_mq_complete_request(rq);
- } else if (test_bit(MTIP_PF_IC_ACTIVE_BIT, &dd->port->flags)) {
+ struct mtip_cmd *cmd = blk_mq_rq_to_pdu(rq);
- cmd = mtip_cmd_from_tag(dd, MTIP_TAG_INTERNAL);
- if (cmd->comp_func)
- cmd->comp_func(dd->port, MTIP_TAG_INTERNAL,
- cmd, -ENODEV);
- }
+ cmd->status = -ENODEV;
+ blk_mq_complete_request(rq);
}
/*
@@ -4168,8 +4001,7 @@ static int mtip_block_remove(struct driver_data *dd)
* Explicitly wait here for IOs to quiesce,
* as mtip_standby_drive usually won't wait for IOs.
*/
- if (!mtip_quiesce_io(dd->port, MTIP_QUIESCE_IO_TIMEOUT_MS,
- GFP_KERNEL))
+ if (!mtip_quiesce_io(dd->port, MTIP_QUIESCE_IO_TIMEOUT_MS))
mtip_standby_drive(dd);
}
else
diff --git a/drivers/block/mtip32xx/mtip32xx.h b/drivers/block/mtip32xx/mtip32xx.h
index 57b41528a824..37b8e3e0bb78 100644
--- a/drivers/block/mtip32xx/mtip32xx.h
+++ b/drivers/block/mtip32xx/mtip32xx.h
@@ -333,16 +333,6 @@ struct mtip_cmd {
dma_addr_t command_dma; /* corresponding physical address */
- void *comp_data; /* data passed to completion function comp_func() */
- /*
- * Completion function called by the ISR upon completion of
- * a command.
- */
- void (*comp_func)(struct mtip_port *port,
- int tag,
- struct mtip_cmd *cmd,
- int status);
-
int scatter_ents; /* Number of scatter list entries used */
int unaligned; /* command is unaligned on 4k boundary */
diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c
index 9b482baa869e..9a7bb2c29447 100644
--- a/drivers/block/nbd.c
+++ b/drivers/block/nbd.c
@@ -18,6 +18,7 @@
#include <linux/module.h>
#include <linux/init.h>
#include <linux/sched.h>
+#include <linux/sched/mm.h>
#include <linux/fs.h>
#include <linux/bio.h>
#include <linux/stat.h>
@@ -347,7 +348,7 @@ static int sock_xmit(struct nbd_device *nbd, int index, int send,
struct socket *sock = config->socks[index]->sock;
int result;
struct msghdr msg;
- unsigned long pflags = current->flags;
+ unsigned int noreclaim_flag;
if (unlikely(!sock)) {
dev_err_ratelimited(disk_to_dev(nbd->disk),
@@ -358,7 +359,7 @@ static int sock_xmit(struct nbd_device *nbd, int index, int send,
msg.msg_iter = *iter;
- current->flags |= PF_MEMALLOC;
+ noreclaim_flag = memalloc_noreclaim_save();
do {
sock->sk->sk_allocation = GFP_NOIO | __GFP_MEMALLOC;
msg.msg_name = NULL;
@@ -381,7 +382,7 @@ static int sock_xmit(struct nbd_device *nbd, int index, int send,
*sent += result;
} while (msg_data_left(&msg));
- current_restore_flags(pflags, PF_MEMALLOC);
+ memalloc_noreclaim_restore(noreclaim_flag);
return result;
}
@@ -1396,12 +1397,11 @@ static void nbd_dbg_close(void)
#endif
-static int nbd_init_request(void *data, struct request *rq,
- unsigned int hctx_idx, unsigned int request_idx,
- unsigned int numa_node)
+static int nbd_init_request(struct blk_mq_tag_set *set, struct request *rq,
+ unsigned int hctx_idx, unsigned int numa_node)
{
struct nbd_cmd *cmd = blk_mq_rq_to_pdu(rq);
- cmd->nbd = data;
+ cmd->nbd = set->driver_data;
return 0;
}
diff --git a/drivers/block/rbd.c b/drivers/block/rbd.c
index 089ac4179919..454bf9c34882 100644
--- a/drivers/block/rbd.c
+++ b/drivers/block/rbd.c
@@ -387,6 +387,7 @@ struct rbd_device {
struct rw_semaphore lock_rwsem;
enum rbd_lock_state lock_state;
+ char lock_cookie[32];
struct rbd_client_id owner_cid;
struct work_struct acquired_lock_work;
struct work_struct released_lock_work;
@@ -477,13 +478,6 @@ static int minor_to_rbd_dev_id(int minor)
return minor >> RBD_SINGLE_MAJOR_PART_SHIFT;
}
-static bool rbd_is_lock_supported(struct rbd_device *rbd_dev)
-{
- return (rbd_dev->header.features & RBD_FEATURE_EXCLUSIVE_LOCK) &&
- rbd_dev->spec->snap_id == CEPH_NOSNAP &&
- !rbd_dev->mapping.read_only;
-}
-
static bool __rbd_is_lock_owner(struct rbd_device *rbd_dev)
{
return rbd_dev->lock_state == RBD_LOCK_STATE_LOCKED ||
@@ -731,7 +725,7 @@ static struct rbd_client *rbd_client_create(struct ceph_options *ceph_opts)
kref_init(&rbdc->kref);
INIT_LIST_HEAD(&rbdc->node);
- rbdc->client = ceph_create_client(ceph_opts, rbdc, 0, 0);
+ rbdc->client = ceph_create_client(ceph_opts, rbdc);
if (IS_ERR(rbdc->client))
goto out_rbdc;
ceph_opts = NULL; /* Now rbdc->client is responsible for ceph_opts */
@@ -804,6 +798,7 @@ enum {
Opt_read_only,
Opt_read_write,
Opt_lock_on_read,
+ Opt_exclusive,
Opt_err
};
@@ -816,6 +811,7 @@ static match_table_t rbd_opts_tokens = {
{Opt_read_write, "read_write"},
{Opt_read_write, "rw"}, /* Alternate spelling */
{Opt_lock_on_read, "lock_on_read"},
+ {Opt_exclusive, "exclusive"},
{Opt_err, NULL}
};
@@ -823,11 +819,13 @@ struct rbd_options {
int queue_depth;
bool read_only;
bool lock_on_read;
+ bool exclusive;
};
#define RBD_QUEUE_DEPTH_DEFAULT BLKDEV_MAX_RQ
#define RBD_READ_ONLY_DEFAULT false
#define RBD_LOCK_ON_READ_DEFAULT false
+#define RBD_EXCLUSIVE_DEFAULT false
static int parse_rbd_opts_token(char *c, void *private)
{
@@ -866,6 +864,9 @@ static int parse_rbd_opts_token(char *c, void *private)
case Opt_lock_on_read:
rbd_opts->lock_on_read = true;
break;
+ case Opt_exclusive:
+ rbd_opts->exclusive = true;
+ break;
default:
/* libceph prints "bad option" msg */
return -EINVAL;
@@ -1922,7 +1923,7 @@ static void rbd_osd_req_format_write(struct rbd_obj_request *obj_request)
{
struct ceph_osd_request *osd_req = obj_request->osd_req;
- osd_req->r_mtime = CURRENT_TIME;
+ ktime_get_real_ts(&osd_req->r_mtime);
osd_req->r_data_offset = obj_request->offset;
}
@@ -3079,7 +3080,8 @@ static int rbd_lock(struct rbd_device *rbd_dev)
char cookie[32];
int ret;
- WARN_ON(__rbd_is_lock_owner(rbd_dev));
+ WARN_ON(__rbd_is_lock_owner(rbd_dev) ||
+ rbd_dev->lock_cookie[0] != '\0');
format_lock_cookie(rbd_dev, cookie);
ret = ceph_cls_lock(osdc, &rbd_dev->header_oid, &rbd_dev->header_oloc,
@@ -3089,6 +3091,7 @@ static int rbd_lock(struct rbd_device *rbd_dev)
return ret;
rbd_dev->lock_state = RBD_LOCK_STATE_LOCKED;
+ strcpy(rbd_dev->lock_cookie, cookie);
rbd_set_owner_cid(rbd_dev, &cid);
queue_work(rbd_dev->task_wq, &rbd_dev->acquired_lock_work);
return 0;
@@ -3097,27 +3100,24 @@ static int rbd_lock(struct rbd_device *rbd_dev)
/*
* lock_rwsem must be held for write
*/
-static int rbd_unlock(struct rbd_device *rbd_dev)
+static void rbd_unlock(struct rbd_device *rbd_dev)
{
struct ceph_osd_client *osdc = &rbd_dev->rbd_client->client->osdc;
- char cookie[32];
int ret;
- WARN_ON(!__rbd_is_lock_owner(rbd_dev));
-
- rbd_dev->lock_state = RBD_LOCK_STATE_UNLOCKED;
+ WARN_ON(!__rbd_is_lock_owner(rbd_dev) ||
+ rbd_dev->lock_cookie[0] == '\0');
- format_lock_cookie(rbd_dev, cookie);
ret = ceph_cls_unlock(osdc, &rbd_dev->header_oid, &rbd_dev->header_oloc,
- RBD_LOCK_NAME, cookie);
- if (ret && ret != -ENOENT) {
- rbd_warn(rbd_dev, "cls_unlock failed: %d", ret);
- return ret;
- }
+ RBD_LOCK_NAME, rbd_dev->lock_cookie);
+ if (ret && ret != -ENOENT)
+ rbd_warn(rbd_dev, "failed to unlock: %d", ret);
+ /* treat errors as the image is unlocked */
+ rbd_dev->lock_state = RBD_LOCK_STATE_UNLOCKED;
+ rbd_dev->lock_cookie[0] = '\0';
rbd_set_owner_cid(rbd_dev, &rbd_empty_cid);
queue_work(rbd_dev->task_wq, &rbd_dev->released_lock_work);
- return 0;
}
static int __rbd_notify_op_lock(struct rbd_device *rbd_dev,
@@ -3447,6 +3447,18 @@ again:
ret = rbd_request_lock(rbd_dev);
if (ret == -ETIMEDOUT) {
goto again; /* treat this as a dead client */
+ } else if (ret == -EROFS) {
+ rbd_warn(rbd_dev, "peer will not release lock");
+ /*
+ * If this is rbd_add_acquire_lock(), we want to fail
+ * immediately -- reuse BLACKLISTED flag. Otherwise we
+ * want to block.
+ */
+ if (!(rbd_dev->disk->flags & GENHD_FL_UP)) {
+ set_bit(RBD_DEV_FLAG_BLACKLISTED, &rbd_dev->flags);
+ /* wake "rbd map --exclusive" process */
+ wake_requests(rbd_dev, false);
+ }
} else if (ret < 0) {
rbd_warn(rbd_dev, "error requesting lock: %d", ret);
mod_delayed_work(rbd_dev->task_wq, &rbd_dev->lock_dwork,
@@ -3490,16 +3502,15 @@ static bool rbd_release_lock(struct rbd_device *rbd_dev)
if (rbd_dev->lock_state != RBD_LOCK_STATE_RELEASING)
return false;
- if (!rbd_unlock(rbd_dev))
- /*
- * Give others a chance to grab the lock - we would re-acquire
- * almost immediately if we got new IO during ceph_osdc_sync()
- * otherwise. We need to ack our own notifications, so this
- * lock_dwork will be requeued from rbd_wait_state_locked()
- * after wake_requests() in rbd_handle_released_lock().
- */
- cancel_delayed_work(&rbd_dev->lock_dwork);
-
+ rbd_unlock(rbd_dev);
+ /*
+ * Give others a chance to grab the lock - we would re-acquire
+ * almost immediately if we got new IO during ceph_osdc_sync()
+ * otherwise. We need to ack our own notifications, so this
+ * lock_dwork will be requeued from rbd_wait_state_locked()
+ * after wake_requests() in rbd_handle_released_lock().
+ */
+ cancel_delayed_work(&rbd_dev->lock_dwork);
return true;
}
@@ -3580,12 +3591,16 @@ static void rbd_handle_released_lock(struct rbd_device *rbd_dev, u8 struct_v,
up_read(&rbd_dev->lock_rwsem);
}
-static bool rbd_handle_request_lock(struct rbd_device *rbd_dev, u8 struct_v,
- void **p)
+/*
+ * Returns result for ResponseMessage to be encoded (<= 0), or 1 if no
+ * ResponseMessage is needed.
+ */
+static int rbd_handle_request_lock(struct rbd_device *rbd_dev, u8 struct_v,
+ void **p)
{
struct rbd_client_id my_cid = rbd_get_cid(rbd_dev);
struct rbd_client_id cid = { 0 };
- bool need_to_send;
+ int result = 1;
if (struct_v >= 2) {
cid.gid = ceph_decode_64(p);
@@ -3595,19 +3610,36 @@ static bool rbd_handle_request_lock(struct rbd_device *rbd_dev, u8 struct_v,
dout("%s rbd_dev %p cid %llu-%llu\n", __func__, rbd_dev, cid.gid,
cid.handle);
if (rbd_cid_equal(&cid, &my_cid))
- return false;
+ return result;
down_read(&rbd_dev->lock_rwsem);
- need_to_send = __rbd_is_lock_owner(rbd_dev);
- if (rbd_dev->lock_state == RBD_LOCK_STATE_LOCKED) {
- if (!rbd_cid_equal(&rbd_dev->owner_cid, &rbd_empty_cid)) {
- dout("%s rbd_dev %p queueing unlock_work\n", __func__,
- rbd_dev);
- queue_work(rbd_dev->task_wq, &rbd_dev->unlock_work);
+ if (__rbd_is_lock_owner(rbd_dev)) {
+ if (rbd_dev->lock_state == RBD_LOCK_STATE_LOCKED &&
+ rbd_cid_equal(&rbd_dev->owner_cid, &rbd_empty_cid))
+ goto out_unlock;
+
+ /*
+ * encode ResponseMessage(0) so the peer can detect
+ * a missing owner
+ */
+ result = 0;
+
+ if (rbd_dev->lock_state == RBD_LOCK_STATE_LOCKED) {
+ if (!rbd_dev->opts->exclusive) {
+ dout("%s rbd_dev %p queueing unlock_work\n",
+ __func__, rbd_dev);
+ queue_work(rbd_dev->task_wq,
+ &rbd_dev->unlock_work);
+ } else {
+ /* refuse to release the lock */
+ result = -EROFS;
+ }
}
}
+
+out_unlock:
up_read(&rbd_dev->lock_rwsem);
- return need_to_send;
+ return result;
}
static void __rbd_acknowledge_notify(struct rbd_device *rbd_dev,
@@ -3690,13 +3722,10 @@ static void rbd_watch_cb(void *arg, u64 notify_id, u64 cookie,
rbd_acknowledge_notify(rbd_dev, notify_id, cookie);
break;
case RBD_NOTIFY_OP_REQUEST_LOCK:
- if (rbd_handle_request_lock(rbd_dev, struct_v, &p))
- /*
- * send ResponseMessage(0) back so the client
- * can detect a missing owner
- */
+ ret = rbd_handle_request_lock(rbd_dev, struct_v, &p);
+ if (ret <= 0)
rbd_acknowledge_notify_result(rbd_dev, notify_id,
- cookie, 0);
+ cookie, ret);
else
rbd_acknowledge_notify(rbd_dev, notify_id, cookie);
break;
@@ -3821,24 +3850,51 @@ static void rbd_unregister_watch(struct rbd_device *rbd_dev)
ceph_osdc_flush_notifies(&rbd_dev->rbd_client->client->osdc);
}
+/*
+ * lock_rwsem must be held for write
+ */
+static void rbd_reacquire_lock(struct rbd_device *rbd_dev)
+{
+ struct ceph_osd_client *osdc = &rbd_dev->rbd_client->client->osdc;
+ char cookie[32];
+ int ret;
+
+ WARN_ON(rbd_dev->lock_state != RBD_LOCK_STATE_LOCKED);
+
+ format_lock_cookie(rbd_dev, cookie);
+ ret = ceph_cls_set_cookie(osdc, &rbd_dev->header_oid,
+ &rbd_dev->header_oloc, RBD_LOCK_NAME,
+ CEPH_CLS_LOCK_EXCLUSIVE, rbd_dev->lock_cookie,
+ RBD_LOCK_TAG, cookie);
+ if (ret) {
+ if (ret != -EOPNOTSUPP)
+ rbd_warn(rbd_dev, "failed to update lock cookie: %d",
+ ret);
+
+ /*
+ * Lock cookie cannot be updated on older OSDs, so do
+ * a manual release and queue an acquire.
+ */
+ if (rbd_release_lock(rbd_dev))
+ queue_delayed_work(rbd_dev->task_wq,
+ &rbd_dev->lock_dwork, 0);
+ } else {
+ strcpy(rbd_dev->lock_cookie, cookie);
+ }
+}
+
static void rbd_reregister_watch(struct work_struct *work)
{
struct rbd_device *rbd_dev = container_of(to_delayed_work(work),
struct rbd_device, watch_dwork);
- bool was_lock_owner = false;
- bool need_to_wake = false;
int ret;
dout("%s rbd_dev %p\n", __func__, rbd_dev);
- down_write(&rbd_dev->lock_rwsem);
- if (rbd_dev->lock_state == RBD_LOCK_STATE_LOCKED)
- was_lock_owner = rbd_release_lock(rbd_dev);
-
mutex_lock(&rbd_dev->watch_mutex);
if (rbd_dev->watch_state != RBD_WATCH_STATE_ERROR) {
mutex_unlock(&rbd_dev->watch_mutex);
- goto out;
+ return;
}
ret = __rbd_register_watch(rbd_dev);
@@ -3846,36 +3902,28 @@ static void rbd_reregister_watch(struct work_struct *work)
rbd_warn(rbd_dev, "failed to reregister watch: %d", ret);
if (ret == -EBLACKLISTED || ret == -ENOENT) {
set_bit(RBD_DEV_FLAG_BLACKLISTED, &rbd_dev->flags);
- need_to_wake = true;
+ wake_requests(rbd_dev, true);
} else {
queue_delayed_work(rbd_dev->task_wq,
&rbd_dev->watch_dwork,
RBD_RETRY_DELAY);
}
mutex_unlock(&rbd_dev->watch_mutex);
- goto out;
+ return;
}
- need_to_wake = true;
rbd_dev->watch_state = RBD_WATCH_STATE_REGISTERED;
rbd_dev->watch_cookie = rbd_dev->watch_handle->linger_id;
mutex_unlock(&rbd_dev->watch_mutex);
+ down_write(&rbd_dev->lock_rwsem);
+ if (rbd_dev->lock_state == RBD_LOCK_STATE_LOCKED)
+ rbd_reacquire_lock(rbd_dev);
+ up_write(&rbd_dev->lock_rwsem);
+
ret = rbd_dev_refresh(rbd_dev);
if (ret)
rbd_warn(rbd_dev, "reregisteration refresh failed: %d", ret);
-
- if (was_lock_owner) {
- ret = rbd_try_lock(rbd_dev);
- if (ret)
- rbd_warn(rbd_dev, "reregisteration lock failed: %d",
- ret);
- }
-
-out:
- up_write(&rbd_dev->lock_rwsem);
- if (need_to_wake)
- wake_requests(rbd_dev, true);
}
/*
@@ -4034,10 +4082,6 @@ static void rbd_queue_workfn(struct work_struct *work)
if (op_type != OBJ_OP_READ) {
snapc = rbd_dev->header.snapc;
ceph_get_snap_context(snapc);
- must_be_locked = rbd_is_lock_supported(rbd_dev);
- } else {
- must_be_locked = rbd_dev->opts->lock_on_read &&
- rbd_is_lock_supported(rbd_dev);
}
up_read(&rbd_dev->header_rwsem);
@@ -4048,14 +4092,20 @@ static void rbd_queue_workfn(struct work_struct *work)
goto err_rq;
}
+ must_be_locked =
+ (rbd_dev->header.features & RBD_FEATURE_EXCLUSIVE_LOCK) &&
+ (op_type != OBJ_OP_READ || rbd_dev->opts->lock_on_read);
if (must_be_locked) {
down_read(&rbd_dev->lock_rwsem);
if (rbd_dev->lock_state != RBD_LOCK_STATE_LOCKED &&
- !test_bit(RBD_DEV_FLAG_BLACKLISTED, &rbd_dev->flags))
+ !test_bit(RBD_DEV_FLAG_BLACKLISTED, &rbd_dev->flags)) {
+ if (rbd_dev->opts->exclusive) {
+ rbd_warn(rbd_dev, "exclusive lock required");
+ result = -EROFS;
+ goto err_unlock;
+ }
rbd_wait_state_locked(rbd_dev);
-
- WARN_ON((rbd_dev->lock_state == RBD_LOCK_STATE_LOCKED) ^
- !test_bit(RBD_DEV_FLAG_BLACKLISTED, &rbd_dev->flags));
+ }
if (test_bit(RBD_DEV_FLAG_BLACKLISTED, &rbd_dev->flags)) {
result = -EBLACKLISTED;
goto err_unlock;
@@ -4114,19 +4164,10 @@ static int rbd_queue_rq(struct blk_mq_hw_ctx *hctx,
static void rbd_free_disk(struct rbd_device *rbd_dev)
{
- struct gendisk *disk = rbd_dev->disk;
-
- if (!disk)
- return;
-
+ blk_cleanup_queue(rbd_dev->disk->queue);
+ blk_mq_free_tag_set(&rbd_dev->tag_set);
+ put_disk(rbd_dev->disk);
rbd_dev->disk = NULL;
- if (disk->flags & GENHD_FL_UP) {
- del_gendisk(disk);
- if (disk->queue)
- blk_cleanup_queue(disk->queue);
- blk_mq_free_tag_set(&rbd_dev->tag_set);
- }
- put_disk(disk);
}
static int rbd_obj_read_sync(struct rbd_device *rbd_dev,
@@ -4307,9 +4348,8 @@ out:
return ret;
}
-static int rbd_init_request(void *data, struct request *rq,
- unsigned int hctx_idx, unsigned int request_idx,
- unsigned int numa_node)
+static int rbd_init_request(struct blk_mq_tag_set *set, struct request *rq,
+ unsigned int hctx_idx, unsigned int numa_node)
{
struct work_struct *work = blk_mq_rq_to_pdu(rq);
@@ -4384,8 +4424,12 @@ static int rbd_init_disk(struct rbd_device *rbd_dev)
if (!ceph_test_opt(rbd_dev->rbd_client->client, NOCRC))
q->backing_dev_info->capabilities |= BDI_CAP_STABLE_WRITES;
+ /*
+ * disk_release() expects a queue ref from add_disk() and will
+ * put it. Hold an extra ref until add_disk() is called.
+ */
+ WARN_ON(!blk_get_queue(q));
disk->queue = q;
-
q->queuedata = rbd_dev;
rbd_dev->disk = disk;
@@ -5625,6 +5669,7 @@ static int rbd_add_parse_args(const char *buf,
rbd_opts->read_only = RBD_READ_ONLY_DEFAULT;
rbd_opts->queue_depth = RBD_QUEUE_DEPTH_DEFAULT;
rbd_opts->lock_on_read = RBD_LOCK_ON_READ_DEFAULT;
+ rbd_opts->exclusive = RBD_EXCLUSIVE_DEFAULT;
copts = ceph_parse_options(options, mon_addrs,
mon_addrs + mon_addrs_size - 1,
@@ -5683,6 +5728,33 @@ again:
return ret;
}
+static void rbd_dev_image_unlock(struct rbd_device *rbd_dev)
+{
+ down_write(&rbd_dev->lock_rwsem);
+ if (__rbd_is_lock_owner(rbd_dev))
+ rbd_unlock(rbd_dev);
+ up_write(&rbd_dev->lock_rwsem);
+}
+
+static int rbd_add_acquire_lock(struct rbd_device *rbd_dev)
+{
+ if (!(rbd_dev->header.features & RBD_FEATURE_EXCLUSIVE_LOCK)) {
+ rbd_warn(rbd_dev, "exclusive-lock feature is not enabled");
+ return -EINVAL;
+ }
+
+ /* FIXME: "rbd map --exclusive" should be in interruptible */
+ down_read(&rbd_dev->lock_rwsem);
+ rbd_wait_state_locked(rbd_dev);
+ up_read(&rbd_dev->lock_rwsem);
+ if (test_bit(RBD_DEV_FLAG_BLACKLISTED, &rbd_dev->flags)) {
+ rbd_warn(rbd_dev, "failed to acquire exclusive lock");
+ return -EROFS;
+ }
+
+ return 0;
+}
+
/*
* An rbd format 2 image has a unique identifier, distinct from the
* name given to it by the user. Internally, that identifier is
@@ -5874,6 +5946,15 @@ out_err:
return ret;
}
+static void rbd_dev_device_release(struct rbd_device *rbd_dev)
+{
+ clear_bit(RBD_DEV_FLAG_EXISTS, &rbd_dev->flags);
+ rbd_dev_mapping_clear(rbd_dev);
+ rbd_free_disk(rbd_dev);
+ if (!single_major)
+ unregister_blkdev(rbd_dev->major, rbd_dev->name);
+}
+
/*
* rbd_dev->header_rwsem must be locked for write and will be unlocked
* upon return.
@@ -5909,26 +5990,13 @@ static int rbd_dev_device_setup(struct rbd_device *rbd_dev)
set_capacity(rbd_dev->disk, rbd_dev->mapping.size / SECTOR_SIZE);
set_disk_ro(rbd_dev->disk, rbd_dev->mapping.read_only);
- dev_set_name(&rbd_dev->dev, "%d", rbd_dev->dev_id);
- ret = device_add(&rbd_dev->dev);
+ ret = dev_set_name(&rbd_dev->dev, "%d", rbd_dev->dev_id);
if (ret)
goto err_out_mapping;
- /* Everything's ready. Announce the disk to the world. */
-
set_bit(RBD_DEV_FLAG_EXISTS, &rbd_dev->flags);
up_write(&rbd_dev->header_rwsem);
-
- spin_lock(&rbd_dev_list_lock);
- list_add_tail(&rbd_dev->node, &rbd_dev_list);
- spin_unlock(&rbd_dev_list_lock);
-
- add_disk(rbd_dev->disk);
- pr_info("%s: capacity %llu features 0x%llx\n", rbd_dev->disk->disk_name,
- (unsigned long long)get_capacity(rbd_dev->disk) << SECTOR_SHIFT,
- rbd_dev->header.features);
-
- return ret;
+ return 0;
err_out_mapping:
rbd_dev_mapping_clear(rbd_dev);
@@ -5963,11 +6031,11 @@ static int rbd_dev_header_name(struct rbd_device *rbd_dev)
static void rbd_dev_image_release(struct rbd_device *rbd_dev)
{
rbd_dev_unprobe(rbd_dev);
+ if (rbd_dev->opts)
+ rbd_unregister_watch(rbd_dev);
rbd_dev->image_format = 0;
kfree(rbd_dev->spec->image_id);
rbd_dev->spec->image_id = NULL;
-
- rbd_dev_destroy(rbd_dev);
}
/*
@@ -6127,22 +6195,43 @@ static ssize_t do_rbd_add(struct bus_type *bus,
rbd_dev->mapping.read_only = read_only;
rc = rbd_dev_device_setup(rbd_dev);
- if (rc) {
- /*
- * rbd_unregister_watch() can't be moved into
- * rbd_dev_image_release() without refactoring, see
- * commit 1f3ef78861ac.
- */
- rbd_unregister_watch(rbd_dev);
- rbd_dev_image_release(rbd_dev);
- goto out;
+ if (rc)
+ goto err_out_image_probe;
+
+ if (rbd_dev->opts->exclusive) {
+ rc = rbd_add_acquire_lock(rbd_dev);
+ if (rc)
+ goto err_out_device_setup;
}
+ /* Everything's ready. Announce the disk to the world. */
+
+ rc = device_add(&rbd_dev->dev);
+ if (rc)
+ goto err_out_image_lock;
+
+ add_disk(rbd_dev->disk);
+ /* see rbd_init_disk() */
+ blk_put_queue(rbd_dev->disk->queue);
+
+ spin_lock(&rbd_dev_list_lock);
+ list_add_tail(&rbd_dev->node, &rbd_dev_list);
+ spin_unlock(&rbd_dev_list_lock);
+
+ pr_info("%s: capacity %llu features 0x%llx\n", rbd_dev->disk->disk_name,
+ (unsigned long long)get_capacity(rbd_dev->disk) << SECTOR_SHIFT,
+ rbd_dev->header.features);
rc = count;
out:
module_put(THIS_MODULE);
return rc;
+err_out_image_lock:
+ rbd_dev_image_unlock(rbd_dev);
+err_out_device_setup:
+ rbd_dev_device_release(rbd_dev);
+err_out_image_probe:
+ rbd_dev_image_release(rbd_dev);
err_out_rbd_dev:
rbd_dev_destroy(rbd_dev);
err_out_client:
@@ -6170,21 +6259,6 @@ static ssize_t rbd_add_single_major(struct bus_type *bus,
return do_rbd_add(bus, buf, count);
}
-static void rbd_dev_device_release(struct rbd_device *rbd_dev)
-{
- rbd_free_disk(rbd_dev);
-
- spin_lock(&rbd_dev_list_lock);
- list_del_init(&rbd_dev->node);
- spin_unlock(&rbd_dev_list_lock);
-
- clear_bit(RBD_DEV_FLAG_EXISTS, &rbd_dev->flags);
- device_del(&rbd_dev->dev);
- rbd_dev_mapping_clear(rbd_dev);
- if (!single_major)
- unregister_blkdev(rbd_dev->major, rbd_dev->name);
-}
-
static void rbd_dev_remove_parent(struct rbd_device *rbd_dev)
{
while (rbd_dev->parent) {
@@ -6202,6 +6276,7 @@ static void rbd_dev_remove_parent(struct rbd_device *rbd_dev)
}
rbd_assert(second);
rbd_dev_image_release(second);
+ rbd_dev_destroy(second);
first->parent = NULL;
first->parent_overlap = 0;
@@ -6270,21 +6345,16 @@ static ssize_t do_rbd_remove(struct bus_type *bus,
blk_set_queue_dying(rbd_dev->disk->queue);
}
- down_write(&rbd_dev->lock_rwsem);
- if (__rbd_is_lock_owner(rbd_dev))
- rbd_unlock(rbd_dev);
- up_write(&rbd_dev->lock_rwsem);
- rbd_unregister_watch(rbd_dev);
+ del_gendisk(rbd_dev->disk);
+ spin_lock(&rbd_dev_list_lock);
+ list_del_init(&rbd_dev->node);
+ spin_unlock(&rbd_dev_list_lock);
+ device_del(&rbd_dev->dev);
- /*
- * Don't free anything from rbd_dev->disk until after all
- * notifies are completely processed. Otherwise
- * rbd_bus_del_dev() will race with rbd_watch_cb(), resulting
- * in a potential use after free of rbd_dev->disk or rbd_dev.
- */
+ rbd_dev_image_unlock(rbd_dev);
rbd_dev_device_release(rbd_dev);
rbd_dev_image_release(rbd_dev);
-
+ rbd_dev_destroy(rbd_dev);
return count;
}
diff --git a/drivers/block/virtio_blk.c b/drivers/block/virtio_blk.c
index f94614257462..553cc4c542b4 100644
--- a/drivers/block/virtio_blk.c
+++ b/drivers/block/virtio_blk.c
@@ -452,8 +452,7 @@ static int init_vq(struct virtio_blk *vblk)
}
/* Discover virtqueues and write information to configuration. */
- err = vdev->config->find_vqs(vdev, num_vqs, vqs, callbacks, names,
- &desc);
+ err = virtio_find_vqs(vdev, num_vqs, vqs, callbacks, names, &desc);
if (err)
goto out;
@@ -573,11 +572,10 @@ static const struct device_attribute dev_attr_cache_type_rw =
__ATTR(cache_type, S_IRUGO|S_IWUSR,
virtblk_cache_type_show, virtblk_cache_type_store);
-static int virtblk_init_request(void *data, struct request *rq,
- unsigned int hctx_idx, unsigned int request_idx,
- unsigned int numa_node)
+static int virtblk_init_request(struct blk_mq_tag_set *set, struct request *rq,
+ unsigned int hctx_idx, unsigned int numa_node)
{
- struct virtio_blk *vblk = data;
+ struct virtio_blk *vblk = set->driver_data;
struct virtblk_req *vbr = blk_mq_rq_to_pdu(rq);
#ifdef CONFIG_VIRTIO_BLK_SCSI
diff --git a/drivers/block/xen-blkback/xenbus.c b/drivers/block/xen-blkback/xenbus.c
index 8fe61b5dc5a6..1f3dfaa54d87 100644
--- a/drivers/block/xen-blkback/xenbus.c
+++ b/drivers/block/xen-blkback/xenbus.c
@@ -504,11 +504,13 @@ static int xen_blkbk_remove(struct xenbus_device *dev)
dev_set_drvdata(&dev->dev, NULL);
- if (be->blkif)
+ if (be->blkif) {
xen_blkif_disconnect(be->blkif);
- /* Put the reference we set in xen_blkif_alloc(). */
- xen_blkif_put(be->blkif);
+ /* Put the reference we set in xen_blkif_alloc(). */
+ xen_blkif_put(be->blkif);
+ }
+
kfree(be->mode);
kfree(be);
return 0;
diff --git a/drivers/block/zram/zram_drv.c b/drivers/block/zram/zram_drv.c
index 6fac5fedd610..debee952dcc1 100644
--- a/drivers/block/zram/zram_drv.c
+++ b/drivers/block/zram/zram_drv.c
@@ -45,6 +45,8 @@ static const char *default_compressor = "lzo";
/* Module params (documentation at end) */
static unsigned int num_devices = 1;
+static void zram_free_page(struct zram *zram, size_t index);
+
static inline bool init_done(struct zram *zram)
{
return zram->disksize;
@@ -55,53 +57,70 @@ static inline struct zram *dev_to_zram(struct device *dev)
return (struct zram *)dev_to_disk(dev)->private_data;
}
+static unsigned long zram_get_handle(struct zram *zram, u32 index)
+{
+ return zram->table[index].handle;
+}
+
+static void zram_set_handle(struct zram *zram, u32 index, unsigned long handle)
+{
+ zram->table[index].handle = handle;
+}
+
/* flag operations require table entry bit_spin_lock() being held */
-static int zram_test_flag(struct zram_meta *meta, u32 index,
+static int zram_test_flag(struct zram *zram, u32 index,
enum zram_pageflags flag)
{
- return meta->table[index].value & BIT(flag);
+ return zram->table[index].value & BIT(flag);
}
-static void zram_set_flag(struct zram_meta *meta, u32 index,
+static void zram_set_flag(struct zram *zram, u32 index,
enum zram_pageflags flag)
{
- meta->table[index].value |= BIT(flag);
+ zram->table[index].value |= BIT(flag);
}
-static void zram_clear_flag(struct zram_meta *meta, u32 index,
+static void zram_clear_flag(struct zram *zram, u32 index,
enum zram_pageflags flag)
{
- meta->table[index].value &= ~BIT(flag);
+ zram->table[index].value &= ~BIT(flag);
}
-static inline void zram_set_element(struct zram_meta *meta, u32 index,
+static inline void zram_set_element(struct zram *zram, u32 index,
unsigned long element)
{
- meta->table[index].element = element;
+ zram->table[index].element = element;
}
-static inline void zram_clear_element(struct zram_meta *meta, u32 index)
+static unsigned long zram_get_element(struct zram *zram, u32 index)
{
- meta->table[index].element = 0;
+ return zram->table[index].element;
}
-static size_t zram_get_obj_size(struct zram_meta *meta, u32 index)
+static size_t zram_get_obj_size(struct zram *zram, u32 index)
{
- return meta->table[index].value & (BIT(ZRAM_FLAG_SHIFT) - 1);
+ return zram->table[index].value & (BIT(ZRAM_FLAG_SHIFT) - 1);
}
-static void zram_set_obj_size(struct zram_meta *meta,
+static void zram_set_obj_size(struct zram *zram,
u32 index, size_t size)
{
- unsigned long flags = meta->table[index].value >> ZRAM_FLAG_SHIFT;
+ unsigned long flags = zram->table[index].value >> ZRAM_FLAG_SHIFT;
- meta->table[index].value = (flags << ZRAM_FLAG_SHIFT) | size;
+ zram->table[index].value = (flags << ZRAM_FLAG_SHIFT) | size;
}
+#if PAGE_SIZE != 4096
static inline bool is_partial_io(struct bio_vec *bvec)
{
return bvec->bv_len != PAGE_SIZE;
}
+#else
+static inline bool is_partial_io(struct bio_vec *bvec)
+{
+ return false;
+}
+#endif
static void zram_revalidate_disk(struct zram *zram)
{
@@ -137,8 +156,7 @@ static inline bool valid_io_request(struct zram *zram,
static void update_position(u32 *index, int *offset, struct bio_vec *bvec)
{
- if (*offset + bvec->bv_len >= PAGE_SIZE)
- (*index)++;
+ *index += (*offset + bvec->bv_len) / PAGE_SIZE;
*offset = (*offset + bvec->bv_len) % PAGE_SIZE;
}
@@ -177,31 +195,21 @@ static bool page_same_filled(void *ptr, unsigned long *element)
{
unsigned int pos;
unsigned long *page;
+ unsigned long val;
page = (unsigned long *)ptr;
+ val = page[0];
- for (pos = 0; pos < PAGE_SIZE / sizeof(*page) - 1; pos++) {
- if (page[pos] != page[pos + 1])
+ for (pos = 1; pos < PAGE_SIZE / sizeof(*page); pos++) {
+ if (val != page[pos])
return false;
}
- *element = page[pos];
+ *element = val;
return true;
}
-static void handle_same_page(struct bio_vec *bvec, unsigned long element)
-{
- struct page *page = bvec->bv_page;
- void *user_mem;
-
- user_mem = kmap_atomic(page);
- zram_fill_page(user_mem + bvec->bv_offset, bvec->bv_len, element);
- kunmap_atomic(user_mem);
-
- flush_dcache_page(page);
-}
-
static ssize_t initstate_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
@@ -254,9 +262,8 @@ static ssize_t mem_used_max_store(struct device *dev,
down_read(&zram->init_lock);
if (init_done(zram)) {
- struct zram_meta *meta = zram->meta;
atomic_long_set(&zram->stats.max_used_pages,
- zs_get_total_pages(meta->mem_pool));
+ zs_get_total_pages(zram->mem_pool));
}
up_read(&zram->init_lock);
@@ -329,7 +336,6 @@ static ssize_t compact_store(struct device *dev,
struct device_attribute *attr, const char *buf, size_t len)
{
struct zram *zram = dev_to_zram(dev);
- struct zram_meta *meta;
down_read(&zram->init_lock);
if (!init_done(zram)) {
@@ -337,8 +343,7 @@ static ssize_t compact_store(struct device *dev,
return -EINVAL;
}
- meta = zram->meta;
- zs_compact(meta->mem_pool);
+ zs_compact(zram->mem_pool);
up_read(&zram->init_lock);
return len;
@@ -375,8 +380,8 @@ static ssize_t mm_stat_show(struct device *dev,
down_read(&zram->init_lock);
if (init_done(zram)) {
- mem_used = zs_get_total_pages(zram->meta->mem_pool);
- zs_pool_stats(zram->meta->mem_pool, &pool_stats);
+ mem_used = zs_get_total_pages(zram->mem_pool);
+ zs_pool_stats(zram->mem_pool, &pool_stats);
}
orig_size = atomic64_read(&zram->stats.pages_stored);
@@ -417,56 +422,89 @@ static DEVICE_ATTR_RO(io_stat);
static DEVICE_ATTR_RO(mm_stat);
static DEVICE_ATTR_RO(debug_stat);
-static void zram_meta_free(struct zram_meta *meta, u64 disksize)
+static void zram_slot_lock(struct zram *zram, u32 index)
+{
+ bit_spin_lock(ZRAM_ACCESS, &zram->table[index].value);
+}
+
+static void zram_slot_unlock(struct zram *zram, u32 index)
+{
+ bit_spin_unlock(ZRAM_ACCESS, &zram->table[index].value);
+}
+
+static bool zram_same_page_read(struct zram *zram, u32 index,
+ struct page *page,
+ unsigned int offset, unsigned int len)
+{
+ zram_slot_lock(zram, index);
+ if (unlikely(!zram_get_handle(zram, index) ||
+ zram_test_flag(zram, index, ZRAM_SAME))) {
+ void *mem;
+
+ zram_slot_unlock(zram, index);
+ mem = kmap_atomic(page);
+ zram_fill_page(mem + offset, len,
+ zram_get_element(zram, index));
+ kunmap_atomic(mem);
+ return true;
+ }
+ zram_slot_unlock(zram, index);
+
+ return false;
+}
+
+static bool zram_same_page_write(struct zram *zram, u32 index,
+ struct page *page)
+{
+ unsigned long element;
+ void *mem = kmap_atomic(page);
+
+ if (page_same_filled(mem, &element)) {
+ kunmap_atomic(mem);
+ /* Free memory associated with this sector now. */
+ zram_slot_lock(zram, index);
+ zram_free_page(zram, index);
+ zram_set_flag(zram, index, ZRAM_SAME);
+ zram_set_element(zram, index, element);
+ zram_slot_unlock(zram, index);
+
+ atomic64_inc(&zram->stats.same_pages);
+ return true;
+ }
+ kunmap_atomic(mem);
+
+ return false;
+}
+
+static void zram_meta_free(struct zram *zram, u64 disksize)
{
size_t num_pages = disksize >> PAGE_SHIFT;
size_t index;
/* Free all pages that are still in this zram device */
- for (index = 0; index < num_pages; index++) {
- unsigned long handle = meta->table[index].handle;
- /*
- * No memory is allocated for same element filled pages.
- * Simply clear same page flag.
- */
- if (!handle || zram_test_flag(meta, index, ZRAM_SAME))
- continue;
-
- zs_free(meta->mem_pool, handle);
- }
+ for (index = 0; index < num_pages; index++)
+ zram_free_page(zram, index);
- zs_destroy_pool(meta->mem_pool);
- vfree(meta->table);
- kfree(meta);
+ zs_destroy_pool(zram->mem_pool);
+ vfree(zram->table);
}
-static struct zram_meta *zram_meta_alloc(char *pool_name, u64 disksize)
+static bool zram_meta_alloc(struct zram *zram, u64 disksize)
{
size_t num_pages;
- struct zram_meta *meta = kmalloc(sizeof(*meta), GFP_KERNEL);
-
- if (!meta)
- return NULL;
num_pages = disksize >> PAGE_SHIFT;
- meta->table = vzalloc(num_pages * sizeof(*meta->table));
- if (!meta->table) {
- pr_err("Error allocating zram address table\n");
- goto out_error;
- }
+ zram->table = vzalloc(num_pages * sizeof(*zram->table));
+ if (!zram->table)
+ return false;
- meta->mem_pool = zs_create_pool(pool_name);
- if (!meta->mem_pool) {
- pr_err("Error creating memory pool\n");
- goto out_error;
+ zram->mem_pool = zs_create_pool(zram->disk->disk_name);
+ if (!zram->mem_pool) {
+ vfree(zram->table);
+ return false;
}
- return meta;
-
-out_error:
- vfree(meta->table);
- kfree(meta);
- return NULL;
+ return true;
}
/*
@@ -476,16 +514,15 @@ out_error:
*/
static void zram_free_page(struct zram *zram, size_t index)
{
- struct zram_meta *meta = zram->meta;
- unsigned long handle = meta->table[index].handle;
+ unsigned long handle = zram_get_handle(zram, index);
/*
* No memory is allocated for same element filled pages.
* Simply clear same page flag.
*/
- if (zram_test_flag(meta, index, ZRAM_SAME)) {
- zram_clear_flag(meta, index, ZRAM_SAME);
- zram_clear_element(meta, index);
+ if (zram_test_flag(zram, index, ZRAM_SAME)) {
+ zram_clear_flag(zram, index, ZRAM_SAME);
+ zram_set_element(zram, index, 0);
atomic64_dec(&zram->stats.same_pages);
return;
}
@@ -493,179 +530,111 @@ static void zram_free_page(struct zram *zram, size_t index)
if (!handle)
return;
- zs_free(meta->mem_pool, handle);
+ zs_free(zram->mem_pool, handle);
- atomic64_sub(zram_get_obj_size(meta, index),
+ atomic64_sub(zram_get_obj_size(zram, index),
&zram->stats.compr_data_size);
atomic64_dec(&zram->stats.pages_stored);
- meta->table[index].handle = 0;
- zram_set_obj_size(meta, index, 0);
+ zram_set_handle(zram, index, 0);
+ zram_set_obj_size(zram, index, 0);
}
-static int zram_decompress_page(struct zram *zram, char *mem, u32 index)
+static int zram_decompress_page(struct zram *zram, struct page *page, u32 index)
{
- int ret = 0;
- unsigned char *cmem;
- struct zram_meta *meta = zram->meta;
+ int ret;
unsigned long handle;
unsigned int size;
+ void *src, *dst;
- bit_spin_lock(ZRAM_ACCESS, &meta->table[index].value);
- handle = meta->table[index].handle;
- size = zram_get_obj_size(meta, index);
-
- if (!handle || zram_test_flag(meta, index, ZRAM_SAME)) {
- bit_spin_unlock(ZRAM_ACCESS, &meta->table[index].value);
- zram_fill_page(mem, PAGE_SIZE, meta->table[index].element);
+ if (zram_same_page_read(zram, index, page, 0, PAGE_SIZE))
return 0;
- }
- cmem = zs_map_object(meta->mem_pool, handle, ZS_MM_RO);
+ zram_slot_lock(zram, index);
+ handle = zram_get_handle(zram, index);
+ size = zram_get_obj_size(zram, index);
+
+ src = zs_map_object(zram->mem_pool, handle, ZS_MM_RO);
if (size == PAGE_SIZE) {
- memcpy(mem, cmem, PAGE_SIZE);
+ dst = kmap_atomic(page);
+ memcpy(dst, src, PAGE_SIZE);
+ kunmap_atomic(dst);
+ ret = 0;
} else {
struct zcomp_strm *zstrm = zcomp_stream_get(zram->comp);
- ret = zcomp_decompress(zstrm, cmem, size, mem);
+ dst = kmap_atomic(page);
+ ret = zcomp_decompress(zstrm, src, size, dst);
+ kunmap_atomic(dst);
zcomp_stream_put(zram->comp);
}
- zs_unmap_object(meta->mem_pool, handle);
- bit_spin_unlock(ZRAM_ACCESS, &meta->table[index].value);
+ zs_unmap_object(zram->mem_pool, handle);
+ zram_slot_unlock(zram, index);
/* Should NEVER happen. Return bio error if it does. */
- if (unlikely(ret)) {
+ if (unlikely(ret))
pr_err("Decompression failed! err=%d, page=%u\n", ret, index);
- return ret;
- }
- return 0;
+ return ret;
}
static int zram_bvec_read(struct zram *zram, struct bio_vec *bvec,
- u32 index, int offset)
+ u32 index, int offset)
{
int ret;
struct page *page;
- unsigned char *user_mem, *uncmem = NULL;
- struct zram_meta *meta = zram->meta;
- page = bvec->bv_page;
- bit_spin_lock(ZRAM_ACCESS, &meta->table[index].value);
- if (unlikely(!meta->table[index].handle) ||
- zram_test_flag(meta, index, ZRAM_SAME)) {
- bit_spin_unlock(ZRAM_ACCESS, &meta->table[index].value);
- handle_same_page(bvec, meta->table[index].element);
- return 0;
+ page = bvec->bv_page;
+ if (is_partial_io(bvec)) {
+ /* Use a temporary buffer to decompress the page */
+ page = alloc_page(GFP_NOIO|__GFP_HIGHMEM);
+ if (!page)
+ return -ENOMEM;
}
- bit_spin_unlock(ZRAM_ACCESS, &meta->table[index].value);
- if (is_partial_io(bvec))
- /* Use a temporary buffer to decompress the page */
- uncmem = kmalloc(PAGE_SIZE, GFP_NOIO);
+ ret = zram_decompress_page(zram, page, index);
+ if (unlikely(ret))
+ goto out;
- user_mem = kmap_atomic(page);
- if (!is_partial_io(bvec))
- uncmem = user_mem;
+ if (is_partial_io(bvec)) {
+ void *dst = kmap_atomic(bvec->bv_page);
+ void *src = kmap_atomic(page);
- if (!uncmem) {
- pr_err("Unable to allocate temp memory\n");
- ret = -ENOMEM;
- goto out_cleanup;
+ memcpy(dst + bvec->bv_offset, src + offset, bvec->bv_len);
+ kunmap_atomic(src);
+ kunmap_atomic(dst);
}
-
- ret = zram_decompress_page(zram, uncmem, index);
- /* Should NEVER happen. Return bio error if it does. */
- if (unlikely(ret))
- goto out_cleanup;
-
+out:
if (is_partial_io(bvec))
- memcpy(user_mem + bvec->bv_offset, uncmem + offset,
- bvec->bv_len);
+ __free_page(page);
- flush_dcache_page(page);
- ret = 0;
-out_cleanup:
- kunmap_atomic(user_mem);
- if (is_partial_io(bvec))
- kfree(uncmem);
return ret;
}
-static int zram_bvec_write(struct zram *zram, struct bio_vec *bvec, u32 index,
- int offset)
+static int zram_compress(struct zram *zram, struct zcomp_strm **zstrm,
+ struct page *page,
+ unsigned long *out_handle, unsigned int *out_comp_len)
{
- int ret = 0;
- unsigned int clen;
- unsigned long handle = 0;
- struct page *page;
- unsigned char *user_mem, *cmem, *src, *uncmem = NULL;
- struct zram_meta *meta = zram->meta;
- struct zcomp_strm *zstrm = NULL;
+ int ret;
+ unsigned int comp_len;
+ void *src;
unsigned long alloced_pages;
- unsigned long element;
-
- page = bvec->bv_page;
- if (is_partial_io(bvec)) {
- /*
- * This is a partial IO. We need to read the full page
- * before to write the changes.
- */
- uncmem = kmalloc(PAGE_SIZE, GFP_NOIO);
- if (!uncmem) {
- ret = -ENOMEM;
- goto out;
- }
- ret = zram_decompress_page(zram, uncmem, index);
- if (ret)
- goto out;
- }
+ unsigned long handle = 0;
compress_again:
- user_mem = kmap_atomic(page);
- if (is_partial_io(bvec)) {
- memcpy(uncmem + offset, user_mem + bvec->bv_offset,
- bvec->bv_len);
- kunmap_atomic(user_mem);
- user_mem = NULL;
- } else {
- uncmem = user_mem;
- }
-
- if (page_same_filled(uncmem, &element)) {
- if (user_mem)
- kunmap_atomic(user_mem);
- /* Free memory associated with this sector now. */
- bit_spin_lock(ZRAM_ACCESS, &meta->table[index].value);
- zram_free_page(zram, index);
- zram_set_flag(meta, index, ZRAM_SAME);
- zram_set_element(meta, index, element);
- bit_spin_unlock(ZRAM_ACCESS, &meta->table[index].value);
-
- atomic64_inc(&zram->stats.same_pages);
- ret = 0;
- goto out;
- }
-
- zstrm = zcomp_stream_get(zram->comp);
- ret = zcomp_compress(zstrm, uncmem, &clen);
- if (!is_partial_io(bvec)) {
- kunmap_atomic(user_mem);
- user_mem = NULL;
- uncmem = NULL;
- }
+ src = kmap_atomic(page);
+ ret = zcomp_compress(*zstrm, src, &comp_len);
+ kunmap_atomic(src);
if (unlikely(ret)) {
pr_err("Compression failed! err=%d\n", ret);
- goto out;
+ if (handle)
+ zs_free(zram->mem_pool, handle);
+ return ret;
}
- src = zstrm->buffer;
- if (unlikely(clen > max_zpage_size)) {
- clen = PAGE_SIZE;
- if (is_partial_io(bvec))
- src = uncmem;
- }
+ if (unlikely(comp_len > max_zpage_size))
+ comp_len = PAGE_SIZE;
/*
* handle allocation has 2 paths:
@@ -681,71 +650,121 @@ compress_again:
* from the slow path and handle has already been allocated.
*/
if (!handle)
- handle = zs_malloc(meta->mem_pool, clen,
+ handle = zs_malloc(zram->mem_pool, comp_len,
__GFP_KSWAPD_RECLAIM |
__GFP_NOWARN |
__GFP_HIGHMEM |
__GFP_MOVABLE);
if (!handle) {
zcomp_stream_put(zram->comp);
- zstrm = NULL;
-
atomic64_inc(&zram->stats.writestall);
-
- handle = zs_malloc(meta->mem_pool, clen,
+ handle = zs_malloc(zram->mem_pool, comp_len,
GFP_NOIO | __GFP_HIGHMEM |
__GFP_MOVABLE);
+ *zstrm = zcomp_stream_get(zram->comp);
if (handle)
goto compress_again;
-
- pr_err("Error allocating memory for compressed page: %u, size=%u\n",
- index, clen);
- ret = -ENOMEM;
- goto out;
+ return -ENOMEM;
}
- alloced_pages = zs_get_total_pages(meta->mem_pool);
+ alloced_pages = zs_get_total_pages(zram->mem_pool);
update_used_max(zram, alloced_pages);
if (zram->limit_pages && alloced_pages > zram->limit_pages) {
- zs_free(meta->mem_pool, handle);
- ret = -ENOMEM;
- goto out;
+ zs_free(zram->mem_pool, handle);
+ return -ENOMEM;
}
- cmem = zs_map_object(meta->mem_pool, handle, ZS_MM_WO);
+ *out_handle = handle;
+ *out_comp_len = comp_len;
+ return 0;
+}
+
+static int __zram_bvec_write(struct zram *zram, struct bio_vec *bvec, u32 index)
+{
+ int ret;
+ unsigned long handle;
+ unsigned int comp_len;
+ void *src, *dst;
+ struct zcomp_strm *zstrm;
+ struct page *page = bvec->bv_page;
- if ((clen == PAGE_SIZE) && !is_partial_io(bvec)) {
+ if (zram_same_page_write(zram, index, page))
+ return 0;
+
+ zstrm = zcomp_stream_get(zram->comp);
+ ret = zram_compress(zram, &zstrm, page, &handle, &comp_len);
+ if (ret) {
+ zcomp_stream_put(zram->comp);
+ return ret;
+ }
+
+ dst = zs_map_object(zram->mem_pool, handle, ZS_MM_WO);
+
+ src = zstrm->buffer;
+ if (comp_len == PAGE_SIZE)
src = kmap_atomic(page);
- memcpy(cmem, src, PAGE_SIZE);
+ memcpy(dst, src, comp_len);
+ if (comp_len == PAGE_SIZE)
kunmap_atomic(src);
- } else {
- memcpy(cmem, src, clen);
- }
zcomp_stream_put(zram->comp);
- zstrm = NULL;
- zs_unmap_object(meta->mem_pool, handle);
+ zs_unmap_object(zram->mem_pool, handle);
/*
* Free memory associated with this sector
* before overwriting unused sectors.
*/
- bit_spin_lock(ZRAM_ACCESS, &meta->table[index].value);
+ zram_slot_lock(zram, index);
zram_free_page(zram, index);
-
- meta->table[index].handle = handle;
- zram_set_obj_size(meta, index, clen);
- bit_spin_unlock(ZRAM_ACCESS, &meta->table[index].value);
+ zram_set_handle(zram, index, handle);
+ zram_set_obj_size(zram, index, comp_len);
+ zram_slot_unlock(zram, index);
/* Update stats */
- atomic64_add(clen, &zram->stats.compr_data_size);
+ atomic64_add(comp_len, &zram->stats.compr_data_size);
atomic64_inc(&zram->stats.pages_stored);
+ return 0;
+}
+
+static int zram_bvec_write(struct zram *zram, struct bio_vec *bvec,
+ u32 index, int offset)
+{
+ int ret;
+ struct page *page = NULL;
+ void *src;
+ struct bio_vec vec;
+
+ vec = *bvec;
+ if (is_partial_io(bvec)) {
+ void *dst;
+ /*
+ * This is a partial IO. We need to read the full page
+ * before to write the changes.
+ */
+ page = alloc_page(GFP_NOIO|__GFP_HIGHMEM);
+ if (!page)
+ return -ENOMEM;
+
+ ret = zram_decompress_page(zram, page, index);
+ if (ret)
+ goto out;
+
+ src = kmap_atomic(bvec->bv_page);
+ dst = kmap_atomic(page);
+ memcpy(dst + offset, src + bvec->bv_offset, bvec->bv_len);
+ kunmap_atomic(dst);
+ kunmap_atomic(src);
+
+ vec.bv_page = page;
+ vec.bv_len = PAGE_SIZE;
+ vec.bv_offset = 0;
+ }
+
+ ret = __zram_bvec_write(zram, &vec, index);
out:
- if (zstrm)
- zcomp_stream_put(zram->comp);
if (is_partial_io(bvec))
- kfree(uncmem);
+ __free_page(page);
return ret;
}
@@ -758,7 +777,6 @@ static void zram_bio_discard(struct zram *zram, u32 index,
int offset, struct bio *bio)
{
size_t n = bio->bi_iter.bi_size;
- struct zram_meta *meta = zram->meta;
/*
* zram manages data in physical block size units. Because logical block
@@ -779,9 +797,9 @@ static void zram_bio_discard(struct zram *zram, u32 index,
}
while (n >= PAGE_SIZE) {
- bit_spin_lock(ZRAM_ACCESS, &meta->table[index].value);
+ zram_slot_lock(zram, index);
zram_free_page(zram, index);
- bit_spin_unlock(ZRAM_ACCESS, &meta->table[index].value);
+ zram_slot_unlock(zram, index);
atomic64_inc(&zram->stats.notify_free);
index++;
n -= PAGE_SIZE;
@@ -801,6 +819,7 @@ static int zram_bvec_rw(struct zram *zram, struct bio_vec *bvec, u32 index,
if (!is_write) {
atomic64_inc(&zram->stats.num_reads);
ret = zram_bvec_read(zram, bvec, index, offset);
+ flush_dcache_page(bvec->bv_page);
} else {
atomic64_inc(&zram->stats.num_writes);
ret = zram_bvec_write(zram, bvec, index, offset);
@@ -840,34 +859,21 @@ static void __zram_make_request(struct zram *zram, struct bio *bio)
}
bio_for_each_segment(bvec, bio, iter) {
- int max_transfer_size = PAGE_SIZE - offset;
-
- if (bvec.bv_len > max_transfer_size) {
- /*
- * zram_bvec_rw() can only make operation on a single
- * zram page. Split the bio vector.
- */
- struct bio_vec bv;
-
- bv.bv_page = bvec.bv_page;
- bv.bv_len = max_transfer_size;
- bv.bv_offset = bvec.bv_offset;
+ struct bio_vec bv = bvec;
+ unsigned int unwritten = bvec.bv_len;
+ do {
+ bv.bv_len = min_t(unsigned int, PAGE_SIZE - offset,
+ unwritten);
if (zram_bvec_rw(zram, &bv, index, offset,
- op_is_write(bio_op(bio))) < 0)
+ op_is_write(bio_op(bio))) < 0)
goto out;
- bv.bv_len = bvec.bv_len - max_transfer_size;
- bv.bv_offset += max_transfer_size;
- if (zram_bvec_rw(zram, &bv, index + 1, 0,
- op_is_write(bio_op(bio))) < 0)
- goto out;
- } else
- if (zram_bvec_rw(zram, &bvec, index, offset,
- op_is_write(bio_op(bio))) < 0)
- goto out;
+ bv.bv_offset += bv.bv_len;
+ unwritten -= bv.bv_len;
- update_position(&index, &offset, &bvec);
+ update_position(&index, &offset, &bv);
+ } while (unwritten);
}
bio_endio(bio);
@@ -884,8 +890,6 @@ static blk_qc_t zram_make_request(struct request_queue *queue, struct bio *bio)
{
struct zram *zram = queue->queuedata;
- blk_queue_split(queue, &bio, queue->bio_split);
-
if (!valid_io_request(zram, bio->bi_iter.bi_sector,
bio->bi_iter.bi_size)) {
atomic64_inc(&zram->stats.invalid_io);
@@ -904,14 +908,12 @@ static void zram_slot_free_notify(struct block_device *bdev,
unsigned long index)
{
struct zram *zram;
- struct zram_meta *meta;
zram = bdev->bd_disk->private_data;
- meta = zram->meta;
- bit_spin_lock(ZRAM_ACCESS, &meta->table[index].value);
+ zram_slot_lock(zram, index);
zram_free_page(zram, index);
- bit_spin_unlock(ZRAM_ACCESS, &meta->table[index].value);
+ zram_slot_unlock(zram, index);
atomic64_inc(&zram->stats.notify_free);
}
@@ -955,7 +957,6 @@ out:
static void zram_reset_device(struct zram *zram)
{
- struct zram_meta *meta;
struct zcomp *comp;
u64 disksize;
@@ -968,12 +969,8 @@ static void zram_reset_device(struct zram *zram)
return;
}
- meta = zram->meta;
comp = zram->comp;
disksize = zram->disksize;
-
- /* Reset stats */
- memset(&zram->stats, 0, sizeof(zram->stats));
zram->disksize = 0;
set_capacity(zram->disk, 0);
@@ -981,7 +978,8 @@ static void zram_reset_device(struct zram *zram)
up_write(&zram->init_lock);
/* I/O operation under all of CPU are done so let's free */
- zram_meta_free(meta, disksize);
+ zram_meta_free(zram, disksize);
+ memset(&zram->stats, 0, sizeof(zram->stats));
zcomp_destroy(comp);
}
@@ -990,7 +988,6 @@ static ssize_t disksize_store(struct device *dev,
{
u64 disksize;
struct zcomp *comp;
- struct zram_meta *meta;
struct zram *zram = dev_to_zram(dev);
int err;
@@ -998,10 +995,18 @@ static ssize_t disksize_store(struct device *dev,
if (!disksize)
return -EINVAL;
+ down_write(&zram->init_lock);
+ if (init_done(zram)) {
+ pr_info("Cannot change disksize for initialized device\n");
+ err = -EBUSY;
+ goto out_unlock;
+ }
+
disksize = PAGE_ALIGN(disksize);
- meta = zram_meta_alloc(zram->disk->disk_name, disksize);
- if (!meta)
- return -ENOMEM;
+ if (!zram_meta_alloc(zram, disksize)) {
+ err = -ENOMEM;
+ goto out_unlock;
+ }
comp = zcomp_create(zram->compressor);
if (IS_ERR(comp)) {
@@ -1011,14 +1016,6 @@ static ssize_t disksize_store(struct device *dev,
goto out_free_meta;
}
- down_write(&zram->init_lock);
- if (init_done(zram)) {
- pr_info("Cannot change disksize for initialized device\n");
- err = -EBUSY;
- goto out_destroy_comp;
- }
-
- zram->meta = meta;
zram->comp = comp;
zram->disksize = disksize;
set_capacity(zram->disk, zram->disksize >> SECTOR_SHIFT);
@@ -1027,11 +1024,10 @@ static ssize_t disksize_store(struct device *dev,
return len;
-out_destroy_comp:
- up_write(&zram->init_lock);
- zcomp_destroy(comp);
out_free_meta:
- zram_meta_free(meta, disksize);
+ zram_meta_free(zram, disksize);
+out_unlock:
+ up_write(&zram->init_lock);
return err;
}
@@ -1193,8 +1189,6 @@ static int zram_add(void)
blk_queue_io_min(zram->disk->queue, PAGE_SIZE);
blk_queue_io_opt(zram->disk->queue, PAGE_SIZE);
zram->disk->queue->limits.discard_granularity = PAGE_SIZE;
- zram->disk->queue->limits.max_sectors = SECTORS_PER_PAGE;
- zram->disk->queue->limits.chunk_sectors = 0;
blk_queue_max_discard_sectors(zram->disk->queue, UINT_MAX);
queue_flag_set_unlocked(QUEUE_FLAG_DISCARD, zram->disk->queue);
@@ -1219,7 +1213,6 @@ static int zram_add(void)
goto out_free_disk;
}
strlcpy(zram->compressor, default_compressor, sizeof(zram->compressor));
- zram->meta = NULL;
pr_info("Added device: %s\n", zram->disk->disk_name);
return device_id;
diff --git a/drivers/block/zram/zram_drv.h b/drivers/block/zram/zram_drv.h
index caeff51f1571..e34e44d02e3e 100644
--- a/drivers/block/zram/zram_drv.h
+++ b/drivers/block/zram/zram_drv.h
@@ -92,13 +92,9 @@ struct zram_stats {
atomic64_t writestall; /* no. of write slow paths */
};
-struct zram_meta {
+struct zram {
struct zram_table_entry *table;
struct zs_pool *mem_pool;
-};
-
-struct zram {
- struct zram_meta *meta;
struct zcomp *comp;
struct gendisk *disk;
/* Prevent concurrent execution of device init */
Copyright © 1996 – 2023 Jason A. Donenfeld. All Rights Reverse Engineered.