3 files changed, 304 insertions, 46 deletions

diff --git a/drivers/scsi/osd/Kconfig b/drivers/scsi/osd/Kconfig
index 861b5cebaea..a0703514eb0 100644
--- a/drivers/scsi/osd/Kconfig
+++ b/drivers/scsi/osd/Kconfig
@@ -11,10 +11,6 @@
 # it under the terms of the GNU General Public version 2 License as
 # published by the Free Software Foundation
 #
-# FIXME: SCSI_OSD_INITIATOR should select CONFIG (HMAC) SHA1 somehow.
-#        How is it done properly?
-#
-
 config SCSI_OSD_INITIATOR
 	tristate "OSD-Initiator library"
 	depends on SCSI
diff --git a/drivers/scsi/osd/osd_initiator.c b/drivers/scsi/osd/osd_initiator.c
index 24223473f57..5f4cbf0c475 100644
--- a/drivers/scsi/osd/osd_initiator.c
+++ b/drivers/scsi/osd/osd_initiator.c
@@ -39,6 +39,9 @@
  * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
  */
 
+#include <linux/slab.h>
+#include <linux/module.h>
+
 #include <scsi/osd_initiator.h>
 #include <scsi/osd_sec.h>
 #include <scsi/osd_attributes.h>
@@ -141,6 +144,10 @@ static int _osd_get_print_system_info(struct osd_dev *od,
 	odi->osdname_len = get_attrs[a].len;
 	/* Avoid NULL for memcmp optimization 0-length is good enough */
 	odi->osdname = kzalloc(odi->osdname_len + 1, GFP_KERNEL);
+	if (!odi->osdname) {
+		ret = -ENOMEM;
+		goto out;
+	}
 	if (odi->osdname_len)
 		memcpy(odi->osdname, get_attrs[a].val_ptr, odi->osdname_len);
 	OSD_INFO("OSD_NAME               [%s]\n", odi->osdname);
@@ -450,10 +457,6 @@ void osd_end_request(struct osd_request *or)
 {
 	struct request *rq = or->request;
 
-	_osd_free_seg(or, &or->set_attr);
-	_osd_free_seg(or, &or->enc_get_attr);
-	_osd_free_seg(or, &or->get_attr);
-
 	if (rq) {
 		if (rq->next_rq) {
 			_put_request(rq->next_rq);
@@ -462,6 +465,12 @@ void osd_end_request(struct osd_request *or)
 
 		_put_request(rq);
 	}
+
+	_osd_free_seg(or, &or->get_attr);
+	_osd_free_seg(or, &or->enc_get_attr);
+	_osd_free_seg(or, &or->set_attr);
+	_osd_free_seg(or, &or->cdb_cont);
+
 	_osd_request_free(or);
 }
 EXPORT_SYMBOL(osd_end_request);
@@ -545,6 +554,12 @@ static int _osd_realloc_seg(struct osd_request *or,
 	return 0;
 }
 
+static int _alloc_cdb_cont(struct osd_request *or, unsigned total_bytes)
+{
+	OSD_DEBUG("total_bytes=%d\n", total_bytes);
+	return _osd_realloc_seg(or, &or->cdb_cont, total_bytes);
+}
+
 static int _alloc_set_attr_list(struct osd_request *or,
 	const struct osd_attr *oa, unsigned nelem, unsigned add_bytes)
 {
@@ -714,9 +729,9 @@ static int _osd_req_list_objects(struct osd_request *or,
 		return PTR_ERR(bio);
 	}
 
-	bio->bi_rw &= ~(1 << BIO_RW);
+	bio->bi_rw &= ~REQ_WRITE;
 	or->in.bio = bio;
-	or->in.total_bytes = bio->bi_size;
+	or->in.total_bytes = bio->bi_iter.bi_size;
 	return 0;
 }
 
@@ -812,7 +827,7 @@ void osd_req_write(struct osd_request *or,
 {
 	_osd_req_encode_common(or, OSD_ACT_WRITE, obj, offset, len);
 	WARN_ON(or->out.bio || or->out.total_bytes);
-	WARN_ON(0 ==  bio_rw_flagged(bio, BIO_RW));
+	WARN_ON(0 == (bio->bi_rw & REQ_WRITE));
 	or->out.bio = bio;
 	or->out.total_bytes = len;
 }
@@ -827,7 +842,7 @@ int osd_req_write_kern(struct osd_request *or,
 	if (IS_ERR(bio))
 		return PTR_ERR(bio);
 
-	bio->bi_rw |= (1 << BIO_RW); /* FIXME: bio_set_dir() */
+	bio->bi_rw |= REQ_WRITE; /* FIXME: bio_set_dir() */
 	osd_req_write(or, obj, offset, bio, len);
 	return 0;
 }
@@ -863,7 +878,7 @@ void osd_req_read(struct osd_request *or,
 {
 	_osd_req_encode_common(or, OSD_ACT_READ, obj, offset, len);
 	WARN_ON(or->in.bio || or->in.total_bytes);
-	WARN_ON(1 == bio_rw_flagged(bio, BIO_RW));
+	WARN_ON(bio->bi_rw & REQ_WRITE);
 	or->in.bio = bio;
 	or->in.total_bytes = len;
 }
@@ -883,6 +898,211 @@ int osd_req_read_kern(struct osd_request *or,
 }
 EXPORT_SYMBOL(osd_req_read_kern);
 
+static int _add_sg_continuation_descriptor(struct osd_request *or,
+	const struct osd_sg_entry *sglist, unsigned numentries, u64 *len)
+{
+	struct osd_sg_continuation_descriptor *oscd;
+	u32 oscd_size;
+	unsigned i;
+	int ret;
+
+	oscd_size = sizeof(*oscd) + numentries * sizeof(oscd->entries[0]);
+
+	if (!or->cdb_cont.total_bytes) {
+		/* First time, jump over the header, we will write to:
+		 *	cdb_cont.buff + cdb_cont.total_bytes
+		 */
+		or->cdb_cont.total_bytes =
+				sizeof(struct osd_continuation_segment_header);
+	}
+
+	ret = _alloc_cdb_cont(or, or->cdb_cont.total_bytes + oscd_size);
+	if (unlikely(ret))
+		return ret;
+
+	oscd = or->cdb_cont.buff + or->cdb_cont.total_bytes;
+	oscd->hdr.type = cpu_to_be16(SCATTER_GATHER_LIST);
+	oscd->hdr.pad_length = 0;
+	oscd->hdr.length = cpu_to_be32(oscd_size - sizeof(*oscd));
+
+	*len = 0;
+	/* copy the sg entries and convert to network byte order */
+	for (i = 0; i < numentries; i++) {
+		oscd->entries[i].offset = cpu_to_be64(sglist[i].offset);
+		oscd->entries[i].len    = cpu_to_be64(sglist[i].len);
+		*len += sglist[i].len;
+	}
+
+	or->cdb_cont.total_bytes += oscd_size;
+	OSD_DEBUG("total_bytes=%d oscd_size=%d numentries=%d\n",
+		  or->cdb_cont.total_bytes, oscd_size, numentries);
+	return 0;
+}
+
+static int _osd_req_finalize_cdb_cont(struct osd_request *or, const u8 *cap_key)
+{
+	struct request_queue *req_q = osd_request_queue(or->osd_dev);
+	struct bio *bio;
+	struct osd_cdb_head *cdbh = osd_cdb_head(&or->cdb);
+	struct osd_continuation_segment_header *cont_seg_hdr;
+
+	if (!or->cdb_cont.total_bytes)
+		return 0;
+
+	cont_seg_hdr = or->cdb_cont.buff;
+	cont_seg_hdr->format = CDB_CONTINUATION_FORMAT_V2;
+	cont_seg_hdr->service_action = cdbh->varlen_cdb.service_action;
+
+	/* create a bio for continuation segment */
+	bio = bio_map_kern(req_q, or->cdb_cont.buff, or->cdb_cont.total_bytes,
+			   GFP_KERNEL);
+	if (IS_ERR(bio))
+		return PTR_ERR(bio);
+
+	bio->bi_rw |= REQ_WRITE;
+
+	/* integrity check the continuation before the bio is linked
+	 * with the other data segments since the continuation
+	 * integrity is separate from the other data segments.
+	 */
+	osd_sec_sign_data(cont_seg_hdr->integrity_check, bio, cap_key);
+
+	cdbh->v2.cdb_continuation_length = cpu_to_be32(or->cdb_cont.total_bytes);
+
+	/* we can't use _req_append_segment, because we need to link in the
+	 * continuation bio to the head of the bio list - the
+	 * continuation segment (if it exists) is always the first segment in
+	 * the out data buffer.
+	 */
+	bio->bi_next = or->out.bio;
+	or->out.bio = bio;
+	or->out.total_bytes += or->cdb_cont.total_bytes;
+
+	return 0;
+}
+
+/* osd_req_write_sg: Takes a @bio that points to the data out buffer and an
+ * @sglist that has the scatter gather entries. Scatter-gather enables a write
+ * of multiple none-contiguous areas of an object, in a single call. The extents
+ * may overlap and/or be in any order. The only constrain is that:
+ *	total_bytes(sglist) >= total_bytes(bio)
+ */
+int osd_req_write_sg(struct osd_request *or,
+	const struct osd_obj_id *obj, struct bio *bio,
+	const struct osd_sg_entry *sglist, unsigned numentries)
+{
+	u64 len;
+	int ret = _add_sg_continuation_descriptor(or, sglist, numentries, &len);
+
+	if (ret)
+		return ret;
+	osd_req_write(or, obj, 0, bio, len);
+
+	return 0;
+}
+EXPORT_SYMBOL(osd_req_write_sg);
+
+/* osd_req_read_sg: Read multiple extents of an object into @bio
+ * See osd_req_write_sg
+ */
+int osd_req_read_sg(struct osd_request *or,
+	const struct osd_obj_id *obj, struct bio *bio,
+	const struct osd_sg_entry *sglist, unsigned numentries)
+{
+	u64 len;
+	u64 off;
+	int ret;
+
+	if (numentries > 1) {
+		off = 0;
+		ret = _add_sg_continuation_descriptor(or, sglist, numentries,
+						      &len);
+		if (ret)
+			return ret;
+	} else {
+		/* Optimize the case of single segment, read_sg is a
+		 * bidi operation.
+		 */
+		len = sglist->len;
+		off = sglist->offset;
+	}
+	osd_req_read(or, obj, off, bio, len);
+
+	return 0;
+}
+EXPORT_SYMBOL(osd_req_read_sg);
+
+/* SG-list write/read Kern API
+ *
+ * osd_req_{write,read}_sg_kern takes an array of @buff pointers and an array
+ * of sg_entries. @numentries indicates how many pointers and sg_entries there
+ * are.  By requiring an array of buff pointers. This allows a caller to do a
+ * single write/read and scatter into multiple buffers.
+ * NOTE: Each buffer + len should not cross a page boundary.
+ */
+static struct bio *_create_sg_bios(struct osd_request *or,
+	void **buff, const struct osd_sg_entry *sglist, unsigned numentries)
+{
+	struct request_queue *q = osd_request_queue(or->osd_dev);
+	struct bio *bio;
+	unsigned i;
+
+	bio = bio_kmalloc(GFP_KERNEL, numentries);
+	if (unlikely(!bio)) {
+		OSD_DEBUG("Failed to allocate BIO size=%u\n", numentries);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	for (i = 0; i < numentries; i++) {
+		unsigned offset = offset_in_page(buff[i]);
+		struct page *page = virt_to_page(buff[i]);
+		unsigned len = sglist[i].len;
+		unsigned added_len;
+
+		BUG_ON(offset + len > PAGE_SIZE);
+		added_len = bio_add_pc_page(q, bio, page, len, offset);
+		if (unlikely(len != added_len)) {
+			OSD_DEBUG("bio_add_pc_page len(%d) != added_len(%d)\n",
+				  len, added_len);
+			bio_put(bio);
+			return ERR_PTR(-ENOMEM);
+		}
+	}
+
+	return bio;
+}
+
+int osd_req_write_sg_kern(struct osd_request *or,
+	const struct osd_obj_id *obj, void **buff,
+	const struct osd_sg_entry *sglist, unsigned numentries)
+{
+	struct bio *bio = _create_sg_bios(or, buff, sglist, numentries);
+	if (IS_ERR(bio))
+		return PTR_ERR(bio);
+
+	bio->bi_rw |= REQ_WRITE;
+	osd_req_write_sg(or, obj, bio, sglist, numentries);
+
+	return 0;
+}
+EXPORT_SYMBOL(osd_req_write_sg_kern);
+
+int osd_req_read_sg_kern(struct osd_request *or,
+	const struct osd_obj_id *obj, void **buff,
+	const struct osd_sg_entry *sglist, unsigned numentries)
+{
+	struct bio *bio = _create_sg_bios(or, buff, sglist, numentries);
+	if (IS_ERR(bio))
+		return PTR_ERR(bio);
+
+	osd_req_read_sg(or, obj, bio, sglist, numentries);
+
+	return 0;
+}
+EXPORT_SYMBOL(osd_req_read_sg_kern);
+
+
+
 void osd_req_get_attributes(struct osd_request *or,
 	const struct osd_obj_id *obj)
 {
@@ -1216,17 +1436,18 @@ int osd_req_add_get_attr_page(struct osd_request *or,
 	or->get_attr.buff = attar_page;
 	or->get_attr.total_bytes = max_page_len;
 
-	or->set_attr.buff = set_one_attr->val_ptr;
-	or->set_attr.total_bytes = set_one_attr->len;
-
 	cdbh->attrs_page.get_attr_page = cpu_to_be32(page_id);
 	cdbh->attrs_page.get_attr_alloc_length = cpu_to_be32(max_page_len);
-	/* ocdb->attrs_page.get_attr_offset; */
+
+	if (!set_one_attr || !set_one_attr->attr_page)
+		return 0; /* The set is optional */
+
+	or->set_attr.buff = set_one_attr->val_ptr;
+	or->set_attr.total_bytes = set_one_attr->len;
 
 	cdbh->attrs_page.set_attr_page = cpu_to_be32(set_one_attr->attr_page);
 	cdbh->attrs_page.set_attr_id = cpu_to_be32(set_one_attr->attr_id);
 	cdbh->attrs_page.set_attr_length = cpu_to_be32(set_one_attr->len);
-	/* ocdb->attrs_page.set_attr_offset; */
 	return 0;
 }
 EXPORT_SYMBOL(osd_req_add_get_attr_page);
@@ -1246,11 +1467,14 @@ static int _osd_req_finalize_attr_page(struct osd_request *or)
 	if (ret)
 		return ret;
 
+	if (or->set_attr.total_bytes == 0)
+		return 0;
+
 	/* set one value */
 	cdbh->attrs_page.set_attr_offset =
 		osd_req_encode_offset(or, or->out.total_bytes, &out_padding);
 
-	ret = _req_append_segment(or, out_padding, &or->enc_get_attr, NULL,
+	ret = _req_append_segment(or, out_padding, &or->set_attr, NULL,
 				  &or->out);
 	return ret;
 }
@@ -1274,7 +1498,8 @@ static inline void osd_sec_parms_set_in_offset(bool is_v1,
 }
 
 static int _osd_req_finalize_data_integrity(struct osd_request *or,
-	bool has_in, bool has_out, u64 out_data_bytes, const u8 *cap_key)
+	bool has_in, bool has_out, struct bio *out_data_bio, u64 out_data_bytes,
+	const u8 *cap_key)
 {
 	struct osd_security_parameters *sec_parms = _osd_req_sec_params(or);
 	int ret;
@@ -1305,7 +1530,7 @@ static int _osd_req_finalize_data_integrity(struct osd_request *or,
 		or->out.last_seg = NULL;
 
 		/* they are now all chained to request sign them all together */
-		osd_sec_sign_data(&or->out_data_integ, or->out.req->bio,
+		osd_sec_sign_data(&or->out_data_integ, out_data_bio,
 				  cap_key);
 	}
 
@@ -1345,6 +1570,7 @@ static struct request *_make_request(struct request_queue *q, bool has_write,
 		if (unlikely(!req))
 			return ERR_PTR(-ENOMEM);
 
+		blk_rq_set_block_pc(req);
 		return req;
 	}
 }
@@ -1365,7 +1591,6 @@ static int _init_blk_request(struct osd_request *or,
 	}
 
 	or->request = req;
-	req->cmd_type = REQ_TYPE_BLOCK_PC;
 	req->cmd_flags |= REQ_QUIET;
 
 	req->timeout = or->timeout;
@@ -1383,7 +1608,7 @@ static int _init_blk_request(struct osd_request *or,
 				ret = PTR_ERR(req);
 				goto out;
 			}
-			req->cmd_type = REQ_TYPE_BLOCK_PC;
+			blk_rq_set_block_pc(req);
 			or->in.req = or->request->next_rq = req;
 		}
 	} else if (has_in)
@@ -1401,6 +1626,8 @@ int osd_finalize_request(struct osd_request *or,
 {
 	struct osd_cdb_head *cdbh = osd_cdb_head(&or->cdb);
 	bool has_in, has_out;
+	 /* Save for data_integrity without the cdb_continuation */
+	struct bio *out_data_bio = or->out.bio;
 	u64 out_data_bytes = or->out.total_bytes;
 	int ret;
 
@@ -1416,9 +1643,14 @@ int osd_finalize_request(struct osd_request *or,
 	osd_set_caps(&or->cdb, cap);
 
 	has_in = or->in.bio || or->get_attr.total_bytes;
-	has_out = or->out.bio || or->set_attr.total_bytes ||
-		or->enc_get_attr.total_bytes;
+	has_out = or->out.bio || or->cdb_cont.total_bytes ||
+		or->set_attr.total_bytes || or->enc_get_attr.total_bytes;
 
+	ret = _osd_req_finalize_cdb_cont(or, cap_key);
+	if (ret) {
+		OSD_DEBUG("_osd_req_finalize_cdb_cont failed\n");
+		return ret;
+	}
 	ret = _init_blk_request(or, has_in, has_out);
 	if (ret) {
 		OSD_DEBUG("_init_blk_request failed\n");
@@ -1433,6 +1665,10 @@ int osd_finalize_request(struct osd_request *or,
 	cdbh->command_specific_options |= or->attributes_mode;
 	if (or->attributes_mode == OSD_CDB_GET_ATTR_PAGE_SET_ONE) {
 		ret = _osd_req_finalize_attr_page(or);
+		if (ret) {
+			OSD_DEBUG("_osd_req_finalize_attr_page failed\n");
+			return ret;
+		}
 	} else {
 		/* TODO: I think that for the GET_ATTR command these 2 should
 		 * be reversed to keep them in execution order (for embeded
@@ -1452,7 +1688,8 @@ int osd_finalize_request(struct osd_request *or,
 	}
 
 	ret = _osd_req_finalize_data_integrity(or, has_in, has_out,
-					       out_data_bytes, cap_key);
+					       out_data_bio, out_data_bytes,
+					       cap_key);
 	if (ret)
 		return ret;
 
diff --git a/drivers/scsi/osd/osd_uld.c b/drivers/scsi/osd/osd_uld.c
index 0a90702b3d7..e1d9a4c4c4b 100644
--- a/drivers/scsi/osd/osd_uld.c
+++ b/drivers/scsi/osd/osd_uld.c
@@ -50,6 +50,7 @@
 #include <linux/idr.h>
 #include <linux/major.h>
 #include <linux/file.h>
+#include <linux/slab.h>
 
 #include <scsi/scsi.h>
 #include <scsi/scsi_driver.h>
@@ -68,10 +69,10 @@
 #ifndef SCSI_OSD_MAJOR
 #  define SCSI_OSD_MAJOR 260
 #endif
-#define SCSI_OSD_MAX_MINOR 64
+#define SCSI_OSD_MAX_MINOR MINORMASK
 
 static const char osd_name[] = "osd";
-static const char *osd_version_string = "open-osd 0.2.0";
+static const char *osd_version_string = "open-osd 0.2.1";
 
 MODULE_AUTHOR("Boaz Harrosh <bharrosh@panasas.com>");
 MODULE_DESCRIPTION("open-osd Upper-Layer-Driver osd.ko");
@@ -96,9 +97,40 @@ struct osd_dev_handle {
 
 static DEFINE_IDA(osd_minor_ida);
 
+/*
+ * scsi sysfs attribute operations
+ */
+static ssize_t osdname_show(struct device *dev, struct device_attribute *attr,
+			    char *buf)
+{
+	struct osd_uld_device *ould = container_of(dev, struct osd_uld_device,
+						   class_dev);
+	return sprintf(buf, "%s\n", ould->odi.osdname);
+}
+static DEVICE_ATTR_RO(osdname);
+
+static ssize_t systemid_show(struct device *dev, struct device_attribute *attr,
+			    char *buf)
+{
+	struct osd_uld_device *ould = container_of(dev, struct osd_uld_device,
+						   class_dev);
+
+	memcpy(buf, ould->odi.systemid, ould->odi.systemid_len);
+	return ould->odi.systemid_len;
+}
+static DEVICE_ATTR_RO(systemid);
+
+static struct attribute *osd_uld_attrs[] = {
+	&dev_attr_osdname.attr,
+	&dev_attr_systemid.attr,
+	NULL,
+};
+ATTRIBUTE_GROUPS(osd_uld);
+
 static struct class osd_uld_class = {
 	.owner		= THIS_MODULE,
 	.name		= "scsi_osd",
+	.dev_groups	= osd_uld_groups,
 };
 
 /*
@@ -181,6 +213,7 @@ static const struct file_operations osd_fops = {
 	.open           = osd_uld_open,
 	.release        = osd_uld_release,
 	.unlocked_ioctl = osd_uld_ioctl,
+	.llseek		= noop_llseek,
 };
 
 struct osd_dev *osduld_path_lookup(const char *name)
@@ -238,18 +271,11 @@ static inline bool _the_same_or_null(const u8 *a1, unsigned a1_len,
 	return 0 == memcmp(a1, a2, a1_len);
 }
 
-struct find_oud_t {
-	const struct osd_dev_info *odi;
-	struct device *dev;
-	struct osd_uld_device *oud;
-} ;
-
-int _mach_odi(struct device *dev, void *find_data)
+static int _match_odi(struct device *dev, const void *find_data)
 {
 	struct osd_uld_device *oud = container_of(dev, struct osd_uld_device,
 						  class_dev);
-	struct find_oud_t *fot = find_data;
-	const struct osd_dev_info *odi = fot->odi;
+	const struct osd_dev_info *odi = find_data;
 
 	if (_the_same_or_null(oud->odi.systemid, oud->odi.systemid_len,
 			      odi->systemid, odi->systemid_len) &&
@@ -257,7 +283,6 @@ int _mach_odi(struct device *dev, void *find_data)
 			      odi->osdname, odi->osdname_len)) {
 		OSD_DEBUG("found device sysid_len=%d osdname=%d\n",
 			  odi->systemid_len, odi->osdname_len);
-		fot->oud = oud;
 		return 1;
 	} else {
 		return 0;
@@ -271,19 +296,19 @@ int _mach_odi(struct device *dev, void *find_data)
  */
 struct osd_dev *osduld_info_lookup(const struct osd_dev_info *odi)
 {
-	struct find_oud_t find = {.odi = odi};
-
-	find.dev = class_find_device(&osd_uld_class, NULL, &find, _mach_odi);
-	if (likely(find.dev)) {
+	struct device *dev = class_find_device(&osd_uld_class, NULL, odi, _match_odi);
+	if (likely(dev)) {
 		struct osd_dev_handle *odh = kzalloc(sizeof(*odh), GFP_KERNEL);
+		struct osd_uld_device *oud = container_of(dev,
+			struct osd_uld_device, class_dev);
 
 		if (unlikely(!odh)) {
-			put_device(find.dev);
+			put_device(dev);
 			return ERR_PTR(-ENOMEM);
 		}
 
-		odh->od = find.oud->od;
-		odh->oud = find.oud;
+		odh->od = oud->od;
+		odh->oud = oud;
 
 		return &odh->od;
 	}
@@ -463,7 +488,7 @@ static int osd_probe(struct device *dev)
 	oud->class_dev.class = &osd_uld_class;
 	oud->class_dev.parent = dev;
 	oud->class_dev.release = __remove;
-	error = dev_set_name(&oud->class_dev, disk->disk_name);
+	error = dev_set_name(&oud->class_dev, "%s", disk->disk_name);
 	if (error) {
 		OSD_ERR("dev_set_name failed => %d\n", error);
 		goto err_put_cdev;