bdev: add write_zeroes function

Add functionality to the bdev layer to handle the nvme write_zeroes
function.

Change-Id: I0dadad273b28c16db5a2275f7d8d57e98253a8d3
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/372171
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>

CHANGELOG.md

@@ -6,10 +6,16 @@ An [fio](http://github.com/axboe/fio) plugin was added that can route
 I/O to the bdev layer. See the [plugin documentation](https://github.com/spdk/spdk/blob/master/examples/bdev/fio_plugin/README.md)
 for more information.
 
+### Block Device Abstraction Layer (bdev)
+
 spdk_bdev_unmap() was modified to take an offset and a length in bytes as
 arguments instead of requiring the user to provide an array of SCSI
 unmap descriptors. This limits unmaps to a single contiguous range.
 
+spdk_bdev_write_zeroes() was introduced as an alternative to spdk_bdev_unmap().
+It ensures that all unmapped blocks will be zeroed out. This function is
+currently only supported by NVMe block devices.
+
 ### Linux AIO bdev
 
 The AIO bdev now allows the user to override the auto-detected block size.

include/spdk/bdev.h

@@ -88,6 +88,7 @@ enum spdk_bdev_io_type {
 	SPDK_BDEV_IO_TYPE_RESET,
 	SPDK_BDEV_IO_TYPE_NVME_ADMIN,
 	SPDK_BDEV_IO_TYPE_NVME_IO,
+	SPDK_BDEV_IO_TYPE_WRITE_ZEROES,
 };
 
 /**
@@ -329,6 +330,25 @@ int spdk_bdev_writev(struct spdk_bdev_desc *desc, struct spdk_io_channel *ch,
 		     uint64_t offset, uint64_t len,
 		     spdk_bdev_io_completion_cb cb, void *cb_arg);
 
+/**
+ * Submit a write zeroes request to the bdev on the given channel. This command
+ *  ensures that all bytes in the specified range are set to 00h
+ *
+ * \param bdev Block device
+ * \param ch I/O channel. Obtained by calling spdk_bdev_get_io_channel().
+ * \param offset The offset, in bytes, from the start of the block device.
+ * \param nbytes The number of bytes to zero.
+ * \param cb Called when the request is complete.
+ * \param cb_arg Argument passed to cb.
+ *
+ * \return 0 on success. On success, the callback will always
+ * be called (even if the request ultimately failed). Return
+ * negated errno on failure, in which case the callback will not be called.
+ */
+int spdk_bdev_write_zeroes(struct spdk_bdev_desc *desc, struct spdk_io_channel *ch,
+			   uint64_t offset, uint64_t len,
+			   spdk_bdev_io_completion_cb cb, void *cb_arg);
+
 /**
  * Submit an unmap request to the block device. Unmap is sometimes also called trim or
  * deallocate. This notifies the device that the data in the blocks described is no

lib/bdev/bdev.c

@@ -928,6 +928,42 @@ spdk_bdev_writev(struct spdk_bdev_desc *desc, struct spdk_io_channel *ch,
 	return 0;
 }
 
+int
+spdk_bdev_write_zeroes(struct spdk_bdev_desc *desc, struct spdk_io_channel *ch,
+		       uint64_t offset, uint64_t len,
+		       spdk_bdev_io_completion_cb cb, void *cb_arg)
+{
+	int rc;
+	struct spdk_bdev *bdev = desc->bdev;
+	struct spdk_bdev_io *bdev_io;
+	struct spdk_bdev_channel *channel = spdk_io_channel_get_ctx(ch);
+
+	if (!spdk_bdev_io_valid(bdev, offset, len)) {
+		return -EINVAL;
+	}
+
+	bdev_io = spdk_bdev_get_io();
+	if (!bdev_io) {
+		SPDK_ERRLOG("bdev_io memory allocation failed duing write_zeroes\n");
+		return -ENOMEM;
+	}
+
+	bdev_io->ch = channel;
+	bdev_io->u.write.len = len;
+	bdev_io->u.write.offset = offset;
+	bdev_io->type = SPDK_BDEV_IO_TYPE_WRITE_ZEROES;
+
+	spdk_bdev_io_init(bdev_io, bdev, cb_arg, cb);
+
+	rc = spdk_bdev_io_submit(bdev_io);
+	if (rc < 0) {
+		spdk_bdev_put_io(bdev_io);
+		return rc;
+	}
+
+	return 0;
+}
+
 int
 spdk_bdev_unmap(struct spdk_bdev_desc *desc, struct spdk_io_channel *ch,
 		uint64_t offset, uint64_t nbytes,

lib/bdev/rpc/bdev_rpc.c

@@ -82,6 +82,8 @@ spdk_rpc_get_bdevs(struct spdk_jsonrpc_request *request,
 		spdk_json_write_bool(w, spdk_bdev_io_type_supported(bdev, SPDK_BDEV_IO_TYPE_WRITE));
 		spdk_json_write_name(w, "unmap");
 		spdk_json_write_bool(w, spdk_bdev_io_type_supported(bdev, SPDK_BDEV_IO_TYPE_UNMAP));
+		spdk_json_write_name(w, "write_zeroes");
+		spdk_json_write_bool(w, spdk_bdev_io_type_supported(bdev, SPDK_BDEV_IO_TYPE_WRITE_ZEROES));
 		spdk_json_write_name(w, "flush");
 		spdk_json_write_bool(w, spdk_bdev_io_type_supported(bdev, SPDK_BDEV_IO_TYPE_FLUSH));
 		spdk_json_write_name(w, "reset");