* 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.,
- * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
*/
#include <linux/nvme.h>
#include <scsi/sg.h>
#include <asm-generic/io-64-nonatomic-lo-hi.h>
+#include <trace/events/block.h>
+
#define NVME_Q_DEPTH 1024
#define SQ_SIZE(depth) (depth * sizeof(struct nvme_command))
#define CQ_SIZE(depth) (depth * sizeof(struct nvme_completion))
#define ADMIN_TIMEOUT (60 * HZ)
-#define IOD_TIMEOUT (4 * NVME_IO_TIMEOUT)
+#define IOD_TIMEOUT (retry_time * HZ)
unsigned char io_timeout = 30;
module_param(io_timeout, byte, 0644);
MODULE_PARM_DESC(io_timeout, "timeout in seconds for I/O");
+static unsigned char retry_time = 30;
+module_param(retry_time, byte, 0644);
+MODULE_PARM_DESC(retry_time, "time in seconds to retry failed I/O");
+
static int nvme_major;
module_param(nvme_major, int, 0);
#define CMD_CTX_CANCELLED (0x30C + CMD_CTX_BASE)
#define CMD_CTX_COMPLETED (0x310 + CMD_CTX_BASE)
#define CMD_CTX_INVALID (0x314 + CMD_CTX_BASE)
-#define CMD_CTX_FLUSH (0x318 + CMD_CTX_BASE)
-#define CMD_CTX_ABORT (0x31C + CMD_CTX_BASE)
+#define CMD_CTX_ABORT (0x318 + CMD_CTX_BASE)
static void special_completion(struct nvme_queue *nvmeq, void *ctx,
struct nvme_completion *cqe)
{
if (ctx == CMD_CTX_CANCELLED)
return;
- if (ctx == CMD_CTX_FLUSH)
- return;
if (ctx == CMD_CTX_ABORT) {
++nvmeq->dev->abort_limit;
return;
void *ctx;
struct nvme_cmd_info *info = nvme_cmd_info(nvmeq);
- if (cmdid >= nvmeq->q_depth) {
- *fn = special_completion;
+ if (cmdid >= nvmeq->q_depth || !info[cmdid].fn) {
+ if (fn)
+ *fn = special_completion;
return CMD_CTX_INVALID;
}
if (fn)
struct nvme_iod *iod = ctx;
struct bio *bio = iod->private;
u16 status = le16_to_cpup(&cqe->status) >> 1;
+ int error = 0;
if (unlikely(status)) {
if (!(status & NVME_SC_DNR ||
wake_up(&nvmeq->sq_full);
return;
}
+ error = -EIO;
}
if (iod->nents) {
dma_unmap_sg(nvmeq->q_dmadev, iod->sg, iod->nents,
nvme_end_io_acct(bio, iod->start_time);
}
nvme_free_iod(nvmeq->dev, iod);
- if (status)
- bio_endio(bio, -EIO);
- else
- bio_endio(bio, 0);
+
+ trace_block_bio_complete(bdev_get_queue(bio->bi_bdev), bio, error);
+ bio_endio(bio, error);
}
/* length is in bytes. gfp flags indicates whether we may sleep. */
if (!split)
return -ENOMEM;
+ trace_block_split(bdev_get_queue(bio->bi_bdev), bio,
+ split->bi_iter.bi_sector);
bio_chain(split, bio);
if (!waitqueue_active(&nvmeq->sq_full))
return 0;
}
-int nvme_submit_flush_data(struct nvme_queue *nvmeq, struct nvme_ns *ns)
-{
- int cmdid = alloc_cmdid(nvmeq, (void *)CMD_CTX_FLUSH,
- special_completion, NVME_IO_TIMEOUT);
- if (unlikely(cmdid < 0))
- return cmdid;
-
- return nvme_submit_flush(nvmeq, ns, cmdid);
-}
-
static int nvme_submit_iod(struct nvme_queue *nvmeq, struct nvme_iod *iod)
{
struct bio *bio = iod->private;
if (bio->bi_rw & REQ_DISCARD)
return nvme_submit_discard(nvmeq, ns, bio, iod, cmdid);
- if ((bio->bi_rw & REQ_FLUSH) && !iod->nents)
+ if (bio->bi_rw & REQ_FLUSH)
return nvme_submit_flush(nvmeq, ns, cmdid);
control = 0;
return 0;
}
+static int nvme_split_flush_data(struct nvme_queue *nvmeq, struct bio *bio)
+{
+ struct bio *split = bio_clone(bio, GFP_ATOMIC);
+ if (!split)
+ return -ENOMEM;
+
+ split->bi_iter.bi_size = 0;
+ split->bi_phys_segments = 0;
+ bio->bi_rw &= ~REQ_FLUSH;
+ bio_chain(split, bio);
+
+ if (!waitqueue_active(&nvmeq->sq_full))
+ add_wait_queue(&nvmeq->sq_full, &nvmeq->sq_cong_wait);
+ bio_list_add(&nvmeq->sq_cong, split);
+ bio_list_add(&nvmeq->sq_cong, bio);
+ wake_up_process(nvme_thread);
+
+ return 0;
+}
+
/*
* Called with local interrupts disabled and the q_lock held. May not sleep.
*/
int psegs = bio_phys_segments(ns->queue, bio);
int result;
- if ((bio->bi_rw & REQ_FLUSH) && psegs) {
- result = nvme_submit_flush_data(nvmeq, ns);
- if (result)
- return result;
- }
+ if ((bio->bi_rw & REQ_FLUSH) && psegs)
+ return nvme_split_flush_data(nvmeq, bio);
iod = nvme_alloc_iod(psegs, bio->bi_iter.bi_size, GFP_ATOMIC);
if (!iod)
return -EINTR;
if (time_after(jiffies, timeout)) {
dev_err(&dev->pci_dev->dev,
- "Device not ready; aborting initialisation\n");
+ "Device not ready; aborting %s\n", enabled ?
+ "initialisation" : "reset");
return -ENODEV;
}
}
goto put_pages;
}
+ err = -ENOMEM;
iod = nvme_alloc_iod(count, length, GFP_KERNEL);
+ if (!iod)
+ goto put_pages;
+
sg = iod->sg;
sg_init_table(sg, count);
for (i = 0; i < count; i++) {
sg_mark_end(&sg[i - 1]);
iod->nents = count;
- err = -ENOMEM;
nents = dma_map_sg(&dev->pci_dev->dev, sg, count,
write ? DMA_TO_DEVICE : DMA_FROM_DEVICE);
if (!nents)
blk_queue_logical_block_size(ns->queue, 1 << ns->lba_shift);
if (dev->max_hw_sectors)
blk_queue_max_hw_sectors(ns->queue, dev->max_hw_sectors);
+ if (dev->vwc & NVME_CTRL_VWC_PRESENT)
+ blk_queue_flush(ns->queue, REQ_FLUSH | REQ_FUA);
disk->major = nvme_major;
disk->first_minor = 0;
status = nvme_set_features(dev, NVME_FEAT_NUM_QUEUES, q_count, 0,
&result);
- if (status)
- return status < 0 ? -EIO : -EBUSY;
+ if (status < 0)
+ return status;
+ if (status > 0) {
+ dev_err(&dev->pci_dev->dev, "Could not set queue count (%d)\n",
+ status);
+ return -EBUSY;
+ }
return min(result & 0xffff, result >> 16) + 1;
}
res = nvme_identify(dev, 0, 1, dma_addr);
if (res) {
+ dev_err(&pdev->dev, "Identify Controller failed (%d)\n", res);
res = -EIO;
goto out;
}
nn = le32_to_cpup(&ctrl->nn);
dev->oncs = le16_to_cpup(&ctrl->oncs);
dev->abort_limit = ctrl->acl + 1;
+ dev->vwc = ctrl->vwc;
memcpy(dev->serial, ctrl->sn, sizeof(ctrl->sn));
memcpy(dev->model, ctrl->mn, sizeof(ctrl->mn));
memcpy(dev->firmware_rev, ctrl->fr, sizeof(ctrl->fr));
unregister_blkdev(nvme_major, "nvme");
destroy_workqueue(nvme_workq);
BUG_ON(nvme_thread && !IS_ERR(nvme_thread));
+ _nvme_check_size();
}
MODULE_AUTHOR("Matthew Wilcox <willy@linux.intel.com>");