[PATCH v2 0/6] lightnvm: base 2.0 implementation

[PATCH v2 0/6] lightnvm: base 2.0 implementation

[Matias Bjørling]

A couple of patches for 2.0 support for the lightnvm subsystem. They
form the foundation for the integration.

The first two patches is preparation for the 2.0 work. The third patch
implements the 2.0 data structures, the geometry command, and exposes
the sysfs attributes that comes with the 2.0 specification. Note that
the attributes between 1.2 and 2.0 are different, and it is expected
that user-space shall use the version sysfs attribute to know which
attributes will be available.

The next two patches removes max_phys_sect and max_rq_size, as they
not used.

The last patch implements support for using the nvme namespace logical
block and metadata fields and sync it with the internal lightnvm
identify structures.

Changes since v2:

 - Removed blk_queue_block_size() setup in nvm_init and made sure
   to only update csecs and sos in on the late setup path. No reason
   to set it twice. From discussion with Javier.
 - Added two extra patches, that removes max_phys_sect and
   max_rq_size.

Changes since v1:

 - pr_err fix from Randy.
 - Address type fix from Javier.
 - Also CC the nvme mailing list.

Matias Bjørling (6):
  lightnvm: make 1.2 data structures explicit
  lightnvm: flatten nvm_id_group into nvm_id
  lightnvm: add 2.0 geometry identification
  lightnvm: remove max_rq_size
  lightnvm: remove nvm_dev_ops->max_phys_sect
  nvme: lightnvm: add late setup of block size and metadata

 drivers/lightnvm/core.c          |  61 ++---
 drivers/lightnvm/pblk-init.c     |   9 +-
 drivers/lightnvm/pblk-recovery.c |   8 +-
 drivers/nvme/host/core.c         |   2 +
 drivers/nvme/host/lightnvm.c     | 513 ++++++++++++++++++++++++++++-----------
 drivers/nvme/host/nvme.h         |   2 +
 include/linux/lightnvm.h         |  71 +++---
 7 files changed, 442 insertions(+), 224 deletions(-)

-- 
2.11.0

[PATCH v2 1/6] lightnvm: make 1.2 data structures explicit

[Matias Bjørling]

Make the 1.2 data structures explicit, so it will be easy to identify
the 2.0 data structures. Also fix the order of which the nvme_nvm_*
are declared, such that they follow the nvme_nvm_command order.

Signed-off-by: Matias Bjørling <mb@lightnvm.io>
---
 drivers/nvme/host/lightnvm.c | 82 ++++++++++++++++++++++----------------------
 1 file changed, 41 insertions(+), 41 deletions(-)

diff --git a/drivers/nvme/host/lightnvm.c b/drivers/nvme/host/lightnvm.c
index dc0b1335c7c6..60db3f1b59da 100644
--- a/drivers/nvme/host/lightnvm.c
+++ b/drivers/nvme/host/lightnvm.c
@@ -51,6 +51,21 @@ struct nvme_nvm_ph_rw {
 	__le64			resv;
 };
 
+struct nvme_nvm_erase_blk {
+	__u8			opcode;
+	__u8			flags;
+	__u16			command_id;
+	__le32			nsid;
+	__u64			rsvd[2];
+	__le64			prp1;
+	__le64			prp2;
+	__le64			spba;
+	__le16			length;
+	__le16			control;
+	__le32			dsmgmt;
+	__le64			resv;
+};
+
 struct nvme_nvm_identity {
 	__u8			opcode;
 	__u8			flags;
@@ -89,33 +104,18 @@ struct nvme_nvm_setbbtbl {
 	__u32			rsvd4[3];
 };
 
-struct nvme_nvm_erase_blk {
-	__u8			opcode;
-	__u8			flags;
-	__u16			command_id;
-	__le32			nsid;
-	__u64			rsvd[2];
-	__le64			prp1;
-	__le64			prp2;
-	__le64			spba;
-	__le16			length;
-	__le16			control;
-	__le32			dsmgmt;
-	__le64			resv;
-};
-
 struct nvme_nvm_command {
 	union {
 		struct nvme_common_command common;
-		struct nvme_nvm_identity identity;
 		struct nvme_nvm_ph_rw ph_rw;
+		struct nvme_nvm_erase_blk erase;
+		struct nvme_nvm_identity identity;
 		struct nvme_nvm_getbbtbl get_bb;
 		struct nvme_nvm_setbbtbl set_bb;
-		struct nvme_nvm_erase_blk erase;
 	};
 };
 
-struct nvme_nvm_id_group {
+struct nvme_nvm_id12_grp {
 	__u8			mtype;
 	__u8			fmtype;
 	__le16			res16;
@@ -141,7 +141,7 @@ struct nvme_nvm_id_group {
 	__u8			reserved[906];
 } __packed;
 
-struct nvme_nvm_addr_format {
+struct nvme_nvm_id12_addrf {
 	__u8			ch_offset;
 	__u8			ch_len;
 	__u8			lun_offset;
@@ -157,16 +157,16 @@ struct nvme_nvm_addr_format {
 	__u8			res[4];
 } __packed;
 
-struct nvme_nvm_id {
+struct nvme_nvm_id12 {
 	__u8			ver_id;
 	__u8			vmnt;
 	__u8			cgrps;
 	__u8			res;
 	__le32			cap;
 	__le32			dom;
-	struct nvme_nvm_addr_format ppaf;
+	struct nvme_nvm_id12_addrf ppaf;
 	__u8			resv[228];
-	struct nvme_nvm_id_group group;
+	struct nvme_nvm_id12_grp grp;
 	__u8			resv2[2880];
 } __packed;
 
@@ -191,25 +191,25 @@ static inline void _nvme_nvm_check_size(void)
 {
 	BUILD_BUG_ON(sizeof(struct nvme_nvm_identity) != 64);
 	BUILD_BUG_ON(sizeof(struct nvme_nvm_ph_rw) != 64);
+	BUILD_BUG_ON(sizeof(struct nvme_nvm_erase_blk) != 64);
 	BUILD_BUG_ON(sizeof(struct nvme_nvm_getbbtbl) != 64);
 	BUILD_BUG_ON(sizeof(struct nvme_nvm_setbbtbl) != 64);
-	BUILD_BUG_ON(sizeof(struct nvme_nvm_erase_blk) != 64);
-	BUILD_BUG_ON(sizeof(struct nvme_nvm_id_group) != 960);
-	BUILD_BUG_ON(sizeof(struct nvme_nvm_addr_format) != 16);
-	BUILD_BUG_ON(sizeof(struct nvme_nvm_id) != NVME_IDENTIFY_DATA_SIZE);
+	BUILD_BUG_ON(sizeof(struct nvme_nvm_id12_grp) != 960);
+	BUILD_BUG_ON(sizeof(struct nvme_nvm_id12_addrf) != 16);
+	BUILD_BUG_ON(sizeof(struct nvme_nvm_id12) != NVME_IDENTIFY_DATA_SIZE);
 	BUILD_BUG_ON(sizeof(struct nvme_nvm_bb_tbl) != 64);
 }
 
-static int init_grps(struct nvm_id *nvm_id, struct nvme_nvm_id *nvme_nvm_id)
+static int init_grp(struct nvm_id *nvm_id, struct nvme_nvm_id12 *id12)
 {
-	struct nvme_nvm_id_group *src;
+	struct nvme_nvm_id12_grp *src;
 	struct nvm_id_group *grp;
 	int sec_per_pg, sec_per_pl, pg_per_blk;
 
-	if (nvme_nvm_id->cgrps != 1)
+	if (id12->cgrps != 1)
 		return -EINVAL;
 
-	src = &nvme_nvm_id->group;
+	src = &id12->grp;
 	grp = &nvm_id->grp;
 
 	grp->mtype = src->mtype;
@@ -261,34 +261,34 @@ static int init_grps(struct nvm_id *nvm_id, struct nvme_nvm_id *nvme_nvm_id)
 static int nvme_nvm_identity(struct nvm_dev *nvmdev, struct nvm_id *nvm_id)
 {
 	struct nvme_ns *ns = nvmdev->q->queuedata;
-	struct nvme_nvm_id *nvme_nvm_id;
+	struct nvme_nvm_id12 *id;
 	struct nvme_nvm_command c = {};
 	int ret;
 
 	c.identity.opcode = nvme_nvm_admin_identity;
 	c.identity.nsid = cpu_to_le32(ns->head->ns_id);
 
-	nvme_nvm_id = kmalloc(sizeof(struct nvme_nvm_id), GFP_KERNEL);
-	if (!nvme_nvm_id)
+	id = kmalloc(sizeof(struct nvme_nvm_id12), GFP_KERNEL);
+	if (!id)
 		return -ENOMEM;
 
 	ret = nvme_submit_sync_cmd(ns->ctrl->admin_q, (struct nvme_command *)&c,
-				nvme_nvm_id, sizeof(struct nvme_nvm_id));
+				id, sizeof(struct nvme_nvm_id12));
 	if (ret) {
 		ret = -EIO;
 		goto out;
 	}
 
-	nvm_id->ver_id = nvme_nvm_id->ver_id;
-	nvm_id->vmnt = nvme_nvm_id->vmnt;
-	nvm_id->cap = le32_to_cpu(nvme_nvm_id->cap);
-	nvm_id->dom = le32_to_cpu(nvme_nvm_id->dom);
-	memcpy(&nvm_id->ppaf, &nvme_nvm_id->ppaf,
+	nvm_id->ver_id = id->ver_id;
+	nvm_id->vmnt = id->vmnt;
+	nvm_id->cap = le32_to_cpu(id->cap);
+	nvm_id->dom = le32_to_cpu(id->dom);
+	memcpy(&nvm_id->ppaf, &id->ppaf,
 					sizeof(struct nvm_addr_format));
 
-	ret = init_grps(nvm_id, nvme_nvm_id);
+	ret = init_grp(nvm_id, id);
 out:
-	kfree(nvme_nvm_id);
+	kfree(id);
 	return ret;
 }
 
-- 
2.11.0

[PATCH v2 2/6] lightnvm: flatten nvm_id_group into nvm_id

[Matias Bjørling]

There are no groups in the 2.0 specification, make sure that the
nvm_id structure is flattened before 2.0 data structures are added.

Signed-off-by: Matias Bjørling <mb@lightnvm.io>
---
 drivers/lightnvm/core.c      |  25 ++++++-----
 drivers/nvme/host/lightnvm.c | 100 +++++++++++++++++++++----------------------
 include/linux/lightnvm.h     |  53 +++++++++++------------
 3 files changed, 86 insertions(+), 92 deletions(-)

diff --git a/drivers/lightnvm/core.c b/drivers/lightnvm/core.c
index dcc9e621e651..c72863b36439 100644
--- a/drivers/lightnvm/core.c
+++ b/drivers/lightnvm/core.c
@@ -851,33 +851,32 @@ EXPORT_SYMBOL(nvm_get_tgt_bb_tbl);
 static int nvm_core_init(struct nvm_dev *dev)
 {
 	struct nvm_id *id = &dev->identity;
-	struct nvm_id_group *grp = &id->grp;
 	struct nvm_geo *geo = &dev->geo;
 	int ret;
 
 	memcpy(&geo->ppaf, &id->ppaf, sizeof(struct nvm_addr_format));
 
-	if (grp->mtype != 0) {
+	if (id->mtype != 0) {
 		pr_err("nvm: memory type not supported\n");
 		return -EINVAL;
 	}
 
 	/* Whole device values */
-	geo->nr_chnls = grp->num_ch;
-	geo->nr_luns = grp->num_lun;
+	geo->nr_chnls = id->num_ch;
+	geo->nr_luns = id->num_lun;
 
 	/* Generic device geometry values */
-	geo->ws_min = grp->ws_min;
-	geo->ws_opt = grp->ws_opt;
-	geo->ws_seq = grp->ws_seq;
-	geo->ws_per_chk = grp->ws_per_chk;
-	geo->nr_chks = grp->num_chk;
-	geo->sec_size = grp->csecs;
-	geo->oob_size = grp->sos;
-	geo->mccap = grp->mccap;
+	geo->ws_min = id->ws_min;
+	geo->ws_opt = id->ws_opt;
+	geo->ws_seq = id->ws_seq;
+	geo->ws_per_chk = id->ws_per_chk;
+	geo->nr_chks = id->num_chk;
+	geo->sec_size = id->csecs;
+	geo->oob_size = id->sos;
+	geo->mccap = id->mccap;
 	geo->max_rq_size = dev->ops->max_phys_sect * geo->sec_size;
 
-	geo->sec_per_chk = grp->clba;
+	geo->sec_per_chk = id->clba;
 	geo->sec_per_lun = geo->sec_per_chk * geo->nr_chks;
 	geo->all_luns = geo->nr_luns * geo->nr_chnls;
 
diff --git a/drivers/nvme/host/lightnvm.c b/drivers/nvme/host/lightnvm.c
index 60db3f1b59da..6412551ecc65 100644
--- a/drivers/nvme/host/lightnvm.c
+++ b/drivers/nvme/host/lightnvm.c
@@ -203,57 +203,55 @@ static inline void _nvme_nvm_check_size(void)
 static int init_grp(struct nvm_id *nvm_id, struct nvme_nvm_id12 *id12)
 {
 	struct nvme_nvm_id12_grp *src;
-	struct nvm_id_group *grp;
 	int sec_per_pg, sec_per_pl, pg_per_blk;
 
 	if (id12->cgrps != 1)
 		return -EINVAL;
 
 	src = &id12->grp;
-	grp = &nvm_id->grp;
 
-	grp->mtype = src->mtype;
-	grp->fmtype = src->fmtype;
+	nvm_id->mtype = src->mtype;
+	nvm_id->fmtype = src->fmtype;
 
-	grp->num_ch = src->num_ch;
-	grp->num_lun = src->num_lun;
+	nvm_id->num_ch = src->num_ch;
+	nvm_id->num_lun = src->num_lun;
 
-	grp->num_chk = le16_to_cpu(src->num_chk);
-	grp->csecs = le16_to_cpu(src->csecs);
-	grp->sos = le16_to_cpu(src->sos);
+	nvm_id->num_chk = le16_to_cpu(src->num_chk);
+	nvm_id->csecs = le16_to_cpu(src->csecs);
+	nvm_id->sos = le16_to_cpu(src->sos);
 
 	pg_per_blk = le16_to_cpu(src->num_pg);
-	sec_per_pg = le16_to_cpu(src->fpg_sz) / grp->csecs;
+	sec_per_pg = le16_to_cpu(src->fpg_sz) / nvm_id->csecs;
 	sec_per_pl = sec_per_pg * src->num_pln;
-	grp->clba = sec_per_pl * pg_per_blk;
-	grp->ws_per_chk = pg_per_blk;
+	nvm_id->clba = sec_per_pl * pg_per_blk;
+	nvm_id->ws_per_chk = pg_per_blk;
 
-	grp->mpos = le32_to_cpu(src->mpos);
-	grp->cpar = le16_to_cpu(src->cpar);
-	grp->mccap = le32_to_cpu(src->mccap);
+	nvm_id->mpos = le32_to_cpu(src->mpos);
+	nvm_id->cpar = le16_to_cpu(src->cpar);
+	nvm_id->mccap = le32_to_cpu(src->mccap);
 
-	grp->ws_opt = grp->ws_min = sec_per_pg;
-	grp->ws_seq = NVM_IO_SNGL_ACCESS;
+	nvm_id->ws_opt = nvm_id->ws_min = sec_per_pg;
+	nvm_id->ws_seq = NVM_IO_SNGL_ACCESS;
 
-	if (grp->mpos & 0x020202) {
-		grp->ws_seq = NVM_IO_DUAL_ACCESS;
-		grp->ws_opt <<= 1;
-	} else if (grp->mpos & 0x040404) {
-		grp->ws_seq = NVM_IO_QUAD_ACCESS;
-		grp->ws_opt <<= 2;
+	if (nvm_id->mpos & 0x020202) {
+		nvm_id->ws_seq = NVM_IO_DUAL_ACCESS;
+		nvm_id->ws_opt <<= 1;
+	} else if (nvm_id->mpos & 0x040404) {
+		nvm_id->ws_seq = NVM_IO_QUAD_ACCESS;
+		nvm_id->ws_opt <<= 2;
 	}
 
-	grp->trdt = le32_to_cpu(src->trdt);
-	grp->trdm = le32_to_cpu(src->trdm);
-	grp->tprt = le32_to_cpu(src->tprt);
-	grp->tprm = le32_to_cpu(src->tprm);
-	grp->tbet = le32_to_cpu(src->tbet);
-	grp->tbem = le32_to_cpu(src->tbem);
+	nvm_id->trdt = le32_to_cpu(src->trdt);
+	nvm_id->trdm = le32_to_cpu(src->trdm);
+	nvm_id->tprt = le32_to_cpu(src->tprt);
+	nvm_id->tprm = le32_to_cpu(src->tprm);
+	nvm_id->tbet = le32_to_cpu(src->tbet);
+	nvm_id->tbem = le32_to_cpu(src->tbem);
 
 	/* 1.2 compatibility */
-	grp->num_pln = src->num_pln;
-	grp->num_pg = le16_to_cpu(src->num_pg);
-	grp->fpg_sz = le16_to_cpu(src->fpg_sz);
+	nvm_id->num_pln = src->num_pln;
+	nvm_id->num_pg = le16_to_cpu(src->num_pg);
+	nvm_id->fpg_sz = le16_to_cpu(src->fpg_sz);
 
 	return 0;
 }
@@ -740,14 +738,12 @@ static ssize_t nvm_dev_attr_show(struct device *dev,
 	struct nvme_ns *ns = nvme_get_ns_from_dev(dev);
 	struct nvm_dev *ndev = ns->ndev;
 	struct nvm_id *id;
-	struct nvm_id_group *grp;
 	struct attribute *attr;
 
 	if (!ndev)
 		return 0;
 
 	id = &ndev->identity;
-	grp = &id->grp;
 	attr = &dattr->attr;
 
 	if (strcmp(attr->name, "version") == 0) {
@@ -771,41 +767,41 @@ static ssize_t nvm_dev_attr_show(struct device *dev,
 			id->ppaf.pg_offset, id->ppaf.pg_len,
 			id->ppaf.sect_offset, id->ppaf.sect_len);
 	} else if (strcmp(attr->name, "media_type") == 0) {	/* u8 */
-		return scnprintf(page, PAGE_SIZE, "%u\n", grp->mtype);
+		return scnprintf(page, PAGE_SIZE, "%u\n", id->mtype);
 	} else if (strcmp(attr->name, "flash_media_type") == 0) {
-		return scnprintf(page, PAGE_SIZE, "%u\n", grp->fmtype);
+		return scnprintf(page, PAGE_SIZE, "%u\n", id->fmtype);
 	} else if (strcmp(attr->name, "num_channels") == 0) {
-		return scnprintf(page, PAGE_SIZE, "%u\n", grp->num_ch);
+		return scnprintf(page, PAGE_SIZE, "%u\n", id->num_ch);
 	} else if (strcmp(attr->name, "num_luns") == 0) {
-		return scnprintf(page, PAGE_SIZE, "%u\n", grp->num_lun);
+		return scnprintf(page, PAGE_SIZE, "%u\n", id->num_lun);
 	} else if (strcmp(attr->name, "num_planes") == 0) {
-		return scnprintf(page, PAGE_SIZE, "%u\n", grp->num_pln);
+		return scnprintf(page, PAGE_SIZE, "%u\n", id->num_pln);
 	} else if (strcmp(attr->name, "num_blocks") == 0) {	/* u16 */
-		return scnprintf(page, PAGE_SIZE, "%u\n", grp->num_chk);
+		return scnprintf(page, PAGE_SIZE, "%u\n", id->num_chk);
 	} else if (strcmp(attr->name, "num_pages") == 0) {
-		return scnprintf(page, PAGE_SIZE, "%u\n", grp->num_pg);
+		return scnprintf(page, PAGE_SIZE, "%u\n", id->num_pg);
 	} else if (strcmp(attr->name, "page_size") == 0) {
-		return scnprintf(page, PAGE_SIZE, "%u\n", grp->fpg_sz);
+		return scnprintf(page, PAGE_SIZE, "%u\n", id->fpg_sz);
 	} else if (strcmp(attr->name, "hw_sector_size") == 0) {
-		return scnprintf(page, PAGE_SIZE, "%u\n", grp->csecs);
+		return scnprintf(page, PAGE_SIZE, "%u\n", id->csecs);
 	} else if (strcmp(attr->name, "oob_sector_size") == 0) {/* u32 */
-		return scnprintf(page, PAGE_SIZE, "%u\n", grp->sos);
+		return scnprintf(page, PAGE_SIZE, "%u\n", id->sos);
 	} else if (strcmp(attr->name, "read_typ") == 0) {
-		return scnprintf(page, PAGE_SIZE, "%u\n", grp->trdt);
+		return scnprintf(page, PAGE_SIZE, "%u\n", id->trdt);
 	} else if (strcmp(attr->name, "read_max") == 0) {
-		return scnprintf(page, PAGE_SIZE, "%u\n", grp->trdm);
+		return scnprintf(page, PAGE_SIZE, "%u\n", id->trdm);
 	} else if (strcmp(attr->name, "prog_typ") == 0) {
-		return scnprintf(page, PAGE_SIZE, "%u\n", grp->tprt);
+		return scnprintf(page, PAGE_SIZE, "%u\n", id->tprt);
 	} else if (strcmp(attr->name, "prog_max") == 0) {
-		return scnprintf(page, PAGE_SIZE, "%u\n", grp->tprm);
+		return scnprintf(page, PAGE_SIZE, "%u\n", id->tprm);
 	} else if (strcmp(attr->name, "erase_typ") == 0) {
-		return scnprintf(page, PAGE_SIZE, "%u\n", grp->tbet);
+		return scnprintf(page, PAGE_SIZE, "%u\n", id->tbet);
 	} else if (strcmp(attr->name, "erase_max") == 0) {
-		return scnprintf(page, PAGE_SIZE, "%u\n", grp->tbem);
+		return scnprintf(page, PAGE_SIZE, "%u\n", id->tbem);
 	} else if (strcmp(attr->name, "multiplane_modes") == 0) {
-		return scnprintf(page, PAGE_SIZE, "0x%08x\n", grp->mpos);
+		return scnprintf(page, PAGE_SIZE, "0x%08x\n", id->mpos);
 	} else if (strcmp(attr->name, "media_capabilities") == 0) {
-		return scnprintf(page, PAGE_SIZE, "0x%08x\n", grp->mccap);
+		return scnprintf(page, PAGE_SIZE, "0x%08x\n", id->mccap);
 	} else if (strcmp(attr->name, "max_phys_secs") == 0) {
 		return scnprintf(page, PAGE_SIZE, "%u\n",
 				ndev->ops->max_phys_sect);
diff --git a/include/linux/lightnvm.h b/include/linux/lightnvm.h
index 7f4b60abdf27..94b704a8d83d 100644
--- a/include/linux/lightnvm.h
+++ b/include/linux/lightnvm.h
@@ -154,9 +154,29 @@ struct nvm_id_lp_tbl {
 	struct nvm_id_lp_mlc mlc;
 };
 
-struct nvm_id_group {
-	u8	mtype;
-	u8	fmtype;
+struct nvm_addr_format {
+	u8	ch_offset;
+	u8	ch_len;
+	u8	lun_offset;
+	u8	lun_len;
+	u8	pln_offset;
+	u8	pln_len;
+	u8	blk_offset;
+	u8	blk_len;
+	u8	pg_offset;
+	u8	pg_len;
+	u8	sect_offset;
+	u8	sect_len;
+};
+
+struct nvm_id {
+	u8	ver_id;
+	u8	vmnt;
+	u32	cap;
+	u32	dom;
+
+	struct	nvm_addr_format ppaf;
+
 	u8	num_ch;
 	u8	num_lun;
 	u16	num_chk;
@@ -180,33 +200,12 @@ struct nvm_id_group {
 	u16	cpar;
 
 	/* 1.2 compatibility */
+	u8	mtype;
+	u8	fmtype;
+
 	u8	num_pln;
 	u16	num_pg;
 	u16	fpg_sz;
-};
-
-struct nvm_addr_format {
-	u8	ch_offset;
-	u8	ch_len;
-	u8	lun_offset;
-	u8	lun_len;
-	u8	pln_offset;
-	u8	pln_len;
-	u8	blk_offset;
-	u8	blk_len;
-	u8	pg_offset;
-	u8	pg_len;
-	u8	sect_offset;
-	u8	sect_len;
-};
-
-struct nvm_id {
-	u8	ver_id;
-	u8	vmnt;
-	u32	cap;
-	u32	dom;
-	struct nvm_addr_format ppaf;
-	struct nvm_id_group grp;
 } __packed;
 
 struct nvm_target {
-- 
2.11.0

[PATCH v2 3/6] lightnvm: add 2.0 geometry identification

[Matias Bjørling]

Implement the geometry data structures for 2.0 and enable a drive
to be identified as one, including exposing the appropriate 2.0
sysfs entries.

Signed-off-by: Matias Bjørling <mb@lightnvm.io>
---
 drivers/lightnvm/core.c      |   8 +-
 drivers/nvme/host/lightnvm.c | 334 +++++++++++++++++++++++++++++++++++++------
 include/linux/lightnvm.h     |  11 +-
 3 files changed, 297 insertions(+), 56 deletions(-)

diff --git a/drivers/lightnvm/core.c b/drivers/lightnvm/core.c
index c72863b36439..9b1255b3e05e 100644
--- a/drivers/lightnvm/core.c
+++ b/drivers/lightnvm/core.c
@@ -931,11 +931,9 @@ static int nvm_init(struct nvm_dev *dev)
 		goto err;
 	}
 
-	pr_debug("nvm: ver:%x nvm_vendor:%x\n",
-			dev->identity.ver_id, dev->identity.vmnt);
-
-	if (dev->identity.ver_id != 1) {
-		pr_err("nvm: device not supported by kernel.");
+	if (dev->identity.ver_id != 1 && dev->identity.ver_id != 2) {
+		pr_err("nvm: device ver_id %d not supported by kernel.\n",
+				dev->identity.ver_id);
 		goto err;
 	}
 
diff --git a/drivers/nvme/host/lightnvm.c b/drivers/nvme/host/lightnvm.c
index 6412551ecc65..8b243af8a949 100644
--- a/drivers/nvme/host/lightnvm.c
+++ b/drivers/nvme/host/lightnvm.c
@@ -184,6 +184,58 @@ struct nvme_nvm_bb_tbl {
 	__u8	blk[0];
 };
 
+struct nvme_nvm_id20_addrf {
+	__u8			grp_len;
+	__u8			pu_len;
+	__u8			chk_len;
+	__u8			lba_len;
+	__u8			resv[4];
+};
+
+struct nvme_nvm_id20 {
+	__u8			mjr;
+	__u8			mnr;
+	__u8			resv[6];
+
+	struct nvme_nvm_id20_addrf lbaf;
+
+	__le32			mccap;
+	__u8			resv2[12];
+
+	__u8			wit;
+	__u8			resv3[31];
+
+	/* Geometry */
+	__le16			num_grp;
+	__le16			num_pu;
+	__le32			num_chk;
+	__le32			clba;
+	__u8			resv4[52];
+
+	/* Write data requirements */
+	__le32			ws_min;
+	__le32			ws_opt;
+	__le32			mw_cunits;
+	__le32			maxoc;
+	__le32			maxocpu;
+	__u8			resv5[44];
+
+	/* Performance related metrics */
+	__le32			trdt;
+	__le32			trdm;
+	__le32			twrt;
+	__le32			twrm;
+	__le32			tcrst;
+	__le32			tcrsm;
+	__u8			resv6[40];
+
+	/* Reserved area */
+	__u8			resv7[2816];
+
+	/* Vendor specific */
+	__u8			vs[1024];
+};
+
 /*
  * Check we didn't inadvertently grow the command struct
  */
@@ -198,6 +250,8 @@ static inline void _nvme_nvm_check_size(void)
 	BUILD_BUG_ON(sizeof(struct nvme_nvm_id12_addrf) != 16);
 	BUILD_BUG_ON(sizeof(struct nvme_nvm_id12) != NVME_IDENTIFY_DATA_SIZE);
 	BUILD_BUG_ON(sizeof(struct nvme_nvm_bb_tbl) != 64);
+	BUILD_BUG_ON(sizeof(struct nvme_nvm_id20_addrf) != 8);
+	BUILD_BUG_ON(sizeof(struct nvme_nvm_id20) != NVME_IDENTIFY_DATA_SIZE);
 }
 
 static int init_grp(struct nvm_id *nvm_id, struct nvme_nvm_id12 *id12)
@@ -256,6 +310,49 @@ static int init_grp(struct nvm_id *nvm_id, struct nvme_nvm_id12 *id12)
 	return 0;
 }
 
+static int nvme_nvm_setup_12(struct nvm_dev *nvmdev, struct nvm_id *nvm_id,
+		struct nvme_nvm_id12 *id)
+{
+	nvm_id->ver_id = id->ver_id;
+	nvm_id->vmnt = id->vmnt;
+	nvm_id->cap = le32_to_cpu(id->cap);
+	nvm_id->dom = le32_to_cpu(id->dom);
+	memcpy(&nvm_id->ppaf, &id->ppaf,
+					sizeof(struct nvm_addr_format));
+
+	return init_grp(nvm_id, id);
+}
+
+static int nvme_nvm_setup_20(struct nvm_dev *nvmdev, struct nvm_id *nvm_id,
+		struct nvme_nvm_id20 *id)
+{
+	nvm_id->ver_id = id->mjr;
+
+	nvm_id->num_ch = le16_to_cpu(id->num_grp);
+	nvm_id->num_lun = le16_to_cpu(id->num_pu);
+	nvm_id->num_chk = le32_to_cpu(id->num_chk);
+	nvm_id->clba = le32_to_cpu(id->clba);
+
+	nvm_id->ws_min = le32_to_cpu(id->ws_min);
+	nvm_id->ws_opt = le32_to_cpu(id->ws_opt);
+	nvm_id->mw_cunits = le32_to_cpu(id->mw_cunits);
+
+	nvm_id->trdt = le32_to_cpu(id->trdt);
+	nvm_id->trdm = le32_to_cpu(id->trdm);
+	nvm_id->tprt = le32_to_cpu(id->twrt);
+	nvm_id->tprm = le32_to_cpu(id->twrm);
+	nvm_id->tbet = le32_to_cpu(id->tcrst);
+	nvm_id->tbem = le32_to_cpu(id->tcrsm);
+
+	/* calculated values */
+	nvm_id->ws_per_chk = nvm_id->clba / nvm_id->ws_min;
+
+	/* 1.2 compatibility */
+	nvm_id->ws_seq = NVM_IO_SNGL_ACCESS;
+
+	return 0;
+}
+
 static int nvme_nvm_identity(struct nvm_dev *nvmdev, struct nvm_id *nvm_id)
 {
 	struct nvme_ns *ns = nvmdev->q->queuedata;
@@ -277,14 +374,24 @@ static int nvme_nvm_identity(struct nvm_dev *nvmdev, struct nvm_id *nvm_id)
 		goto out;
 	}
 
-	nvm_id->ver_id = id->ver_id;
-	nvm_id->vmnt = id->vmnt;
-	nvm_id->cap = le32_to_cpu(id->cap);
-	nvm_id->dom = le32_to_cpu(id->dom);
-	memcpy(&nvm_id->ppaf, &id->ppaf,
-					sizeof(struct nvm_addr_format));
-
-	ret = init_grp(nvm_id, id);
+	/*
+	 * The 1.2 and 2.0 specifications share the first byte in their geometry
+	 * command to make it possible to know what version a device implements.
+	 */
+	switch (id->ver_id) {
+	case 1:
+		ret = nvme_nvm_setup_12(nvmdev, nvm_id, id);
+		break;
+	case 2:
+		ret = nvme_nvm_setup_20(nvmdev, nvm_id,
+				(struct nvme_nvm_id20 *)id);
+		break;
+	default:
+		dev_err(ns->ctrl->device,
+			"OCSSD revision not supported (%d)\n",
+			nvm_id->ver_id);
+		ret = -EINVAL;
+	}
 out:
 	kfree(id);
 	return ret;
@@ -733,7 +840,7 @@ void nvme_nvm_unregister(struct nvme_ns *ns)
 }
 
 static ssize_t nvm_dev_attr_show(struct device *dev,
-				 struct device_attribute *dattr, char *page)
+		struct device_attribute *dattr, char *page)
 {
 	struct nvme_ns *ns = nvme_get_ns_from_dev(dev);
 	struct nvm_dev *ndev = ns->ndev;
@@ -748,10 +855,36 @@ static ssize_t nvm_dev_attr_show(struct device *dev,
 
 	if (strcmp(attr->name, "version") == 0) {
 		return scnprintf(page, PAGE_SIZE, "%u\n", id->ver_id);
-	} else if (strcmp(attr->name, "vendor_opcode") == 0) {
-		return scnprintf(page, PAGE_SIZE, "%u\n", id->vmnt);
 	} else if (strcmp(attr->name, "capabilities") == 0) {
 		return scnprintf(page, PAGE_SIZE, "%u\n", id->cap);
+	} else if (strcmp(attr->name, "read_typ") == 0) {
+		return scnprintf(page, PAGE_SIZE, "%u\n", id->trdt);
+	} else if (strcmp(attr->name, "read_max") == 0) {
+		return scnprintf(page, PAGE_SIZE, "%u\n", id->trdm);
+	} else {
+		return scnprintf(page,
+				 PAGE_SIZE,
+				 "Unhandled attr(%s) in `nvm_dev_attr_show`\n",
+				 attr->name);
+	}
+}
+
+static ssize_t nvm_dev_attr_show_12(struct device *dev,
+		struct device_attribute *dattr, char *page)
+{
+	struct nvme_ns *ns = nvme_get_ns_from_dev(dev);
+	struct nvm_dev *ndev = ns->ndev;
+	struct nvm_id *id;
+	struct attribute *attr;
+
+	if (!ndev)
+		return 0;
+
+	id = &ndev->identity;
+	attr = &dattr->attr;
+
+	if (strcmp(attr->name, "vendor_opcode") == 0) {
+		return scnprintf(page, PAGE_SIZE, "%u\n", id->vmnt);
 	} else if (strcmp(attr->name, "device_mode") == 0) {
 		return scnprintf(page, PAGE_SIZE, "%u\n", id->dom);
 	/* kept for compatibility */
@@ -786,10 +919,6 @@ static ssize_t nvm_dev_attr_show(struct device *dev,
 		return scnprintf(page, PAGE_SIZE, "%u\n", id->csecs);
 	} else if (strcmp(attr->name, "oob_sector_size") == 0) {/* u32 */
 		return scnprintf(page, PAGE_SIZE, "%u\n", id->sos);
-	} else if (strcmp(attr->name, "read_typ") == 0) {
-		return scnprintf(page, PAGE_SIZE, "%u\n", id->trdt);
-	} else if (strcmp(attr->name, "read_max") == 0) {
-		return scnprintf(page, PAGE_SIZE, "%u\n", id->trdm);
 	} else if (strcmp(attr->name, "prog_typ") == 0) {
 		return scnprintf(page, PAGE_SIZE, "%u\n", id->tprt);
 	} else if (strcmp(attr->name, "prog_max") == 0) {
@@ -808,48 +937,99 @@ static ssize_t nvm_dev_attr_show(struct device *dev,
 	} else {
 		return scnprintf(page,
 				 PAGE_SIZE,
-				 "Unhandled attr(%s) in `nvm_dev_attr_show`\n",
+				 "Unhandled attr(%s) in `nvm_dev_attr_show_12`\n",
 				 attr->name);
 	}
 }
 
-#define NVM_DEV_ATTR_RO(_name)						\
+static ssize_t nvm_dev_attr_show_20(struct device *dev,
+		struct device_attribute *dattr, char *page)
+{
+	struct nvme_ns *ns = nvme_get_ns_from_dev(dev);
+	struct nvm_dev *ndev = ns->ndev;
+	struct nvm_id *id;
+	struct attribute *attr;
+
+	if (!ndev)
+		return 0;
+
+	id = &ndev->identity;
+	attr = &dattr->attr;
+
+	if (strcmp(attr->name, "groups") == 0) {
+		return scnprintf(page, PAGE_SIZE, "%u\n", id->num_ch);
+	} else if (strcmp(attr->name, "punits") == 0) {
+		return scnprintf(page, PAGE_SIZE, "%u\n", id->num_lun);
+	} else if (strcmp(attr->name, "chunks") == 0) {
+		return scnprintf(page, PAGE_SIZE, "%u\n", id->num_chk);
+	} else if (strcmp(attr->name, "clba") == 0) {
+		return scnprintf(page, PAGE_SIZE, "%u\n", id->clba);
+	} else if (strcmp(attr->name, "ws_min") == 0) {
+		return scnprintf(page, PAGE_SIZE, "%u\n", id->ws_min);
+	} else if (strcmp(attr->name, "ws_opt") == 0) {
+		return scnprintf(page, PAGE_SIZE, "%u\n", id->ws_opt);
+	} else if (strcmp(attr->name, "mw_cunits") == 0) {
+		return scnprintf(page, PAGE_SIZE, "%u\n", id->mw_cunits);
+	} else if (strcmp(attr->name, "write_typ") == 0) {
+		return scnprintf(page, PAGE_SIZE, "%u\n", id->tprt);
+	} else if (strcmp(attr->name, "write_max") == 0) {
+		return scnprintf(page, PAGE_SIZE, "%u\n", id->tprm);
+	} else if (strcmp(attr->name, "reset_typ") == 0) {
+		return scnprintf(page, PAGE_SIZE, "%u\n", id->tbet);
+	} else if (strcmp(attr->name, "reset_max") == 0) {
+		return scnprintf(page, PAGE_SIZE, "%u\n", id->tbem);
+	} else {
+		return scnprintf(page,
+				 PAGE_SIZE,
+				 "Unhandled attr(%s) in `nvm_dev_attr_show_20`\n",
+				 attr->name);
+	}
+}
+
+#define NVM_DEV_ATTR_RO(_name)					\
 	DEVICE_ATTR(_name, S_IRUGO, nvm_dev_attr_show, NULL)
+#define NVM_DEV_ATTR_12_RO(_name)					\
+	DEVICE_ATTR(_name, S_IRUGO, nvm_dev_attr_show_12, NULL)
+#define NVM_DEV_ATTR_20_RO(_name)					\
+	DEVICE_ATTR(_name, S_IRUGO, nvm_dev_attr_show_20, NULL)
 
+/* general attributes */
 static NVM_DEV_ATTR_RO(version);
-static NVM_DEV_ATTR_RO(vendor_opcode);
 static NVM_DEV_ATTR_RO(capabilities);
-static NVM_DEV_ATTR_RO(device_mode);
-static NVM_DEV_ATTR_RO(ppa_format);
-static NVM_DEV_ATTR_RO(media_manager);
 
-static NVM_DEV_ATTR_RO(media_type);
-static NVM_DEV_ATTR_RO(flash_media_type);
-static NVM_DEV_ATTR_RO(num_channels);
-static NVM_DEV_ATTR_RO(num_luns);
-static NVM_DEV_ATTR_RO(num_planes);
-static NVM_DEV_ATTR_RO(num_blocks);
-static NVM_DEV_ATTR_RO(num_pages);
-static NVM_DEV_ATTR_RO(page_size);
-static NVM_DEV_ATTR_RO(hw_sector_size);
-static NVM_DEV_ATTR_RO(oob_sector_size);
 static NVM_DEV_ATTR_RO(read_typ);
 static NVM_DEV_ATTR_RO(read_max);
-static NVM_DEV_ATTR_RO(prog_typ);
-static NVM_DEV_ATTR_RO(prog_max);
-static NVM_DEV_ATTR_RO(erase_typ);
-static NVM_DEV_ATTR_RO(erase_max);
-static NVM_DEV_ATTR_RO(multiplane_modes);
-static NVM_DEV_ATTR_RO(media_capabilities);
-static NVM_DEV_ATTR_RO(max_phys_secs);
 
-static struct attribute *nvm_dev_attrs[] = {
+/* 1.2 values */
+static NVM_DEV_ATTR_12_RO(vendor_opcode);
+static NVM_DEV_ATTR_12_RO(device_mode);
+static NVM_DEV_ATTR_12_RO(ppa_format);
+static NVM_DEV_ATTR_12_RO(media_manager);
+static NVM_DEV_ATTR_12_RO(media_type);
+static NVM_DEV_ATTR_12_RO(flash_media_type);
+static NVM_DEV_ATTR_12_RO(num_channels);
+static NVM_DEV_ATTR_12_RO(num_luns);
+static NVM_DEV_ATTR_12_RO(num_planes);
+static NVM_DEV_ATTR_12_RO(num_blocks);
+static NVM_DEV_ATTR_12_RO(num_pages);
+static NVM_DEV_ATTR_12_RO(page_size);
+static NVM_DEV_ATTR_12_RO(hw_sector_size);
+static NVM_DEV_ATTR_12_RO(oob_sector_size);
+static NVM_DEV_ATTR_12_RO(prog_typ);
+static NVM_DEV_ATTR_12_RO(prog_max);
+static NVM_DEV_ATTR_12_RO(erase_typ);
+static NVM_DEV_ATTR_12_RO(erase_max);
+static NVM_DEV_ATTR_12_RO(multiplane_modes);
+static NVM_DEV_ATTR_12_RO(media_capabilities);
+static NVM_DEV_ATTR_12_RO(max_phys_secs);
+
+static struct attribute *nvm_dev_attrs_12[] = {
 	&dev_attr_version.attr,
+	&dev_attr_capabilities.attr,
+
 	&dev_attr_vendor_opcode.attr,
-	&dev_attr_capabilities.attr,
 	&dev_attr_device_mode.attr,
 	&dev_attr_media_manager.attr,
-
 	&dev_attr_ppa_format.attr,
 	&dev_attr_media_type.attr,
 	&dev_attr_flash_media_type.attr,
@@ -870,22 +1050,82 @@ static struct attribute *nvm_dev_attrs[] = {
 	&dev_attr_multiplane_modes.attr,
 	&dev_attr_media_capabilities.attr,
 	&dev_attr_max_phys_secs.attr,
+
+	NULL,
+};
+
+static const struct attribute_group nvm_dev_attr_group_12 = {
+	.name		= "lightnvm",
+	.attrs		= nvm_dev_attrs_12,
+};
+
+/* 2.0 values */
+static NVM_DEV_ATTR_20_RO(groups);
+static NVM_DEV_ATTR_20_RO(punits);
+static NVM_DEV_ATTR_20_RO(chunks);
+static NVM_DEV_ATTR_20_RO(clba);
+static NVM_DEV_ATTR_20_RO(ws_min);
+static NVM_DEV_ATTR_20_RO(ws_opt);
+static NVM_DEV_ATTR_20_RO(mw_cunits);
+static NVM_DEV_ATTR_20_RO(write_typ);
+static NVM_DEV_ATTR_20_RO(write_max);
+static NVM_DEV_ATTR_20_RO(reset_typ);
+static NVM_DEV_ATTR_20_RO(reset_max);
+
+static struct attribute *nvm_dev_attrs_20[] = {
+	&dev_attr_version.attr,
+	&dev_attr_capabilities.attr,
+
+	&dev_attr_groups.attr,
+	&dev_attr_punits.attr,
+	&dev_attr_chunks.attr,
+	&dev_attr_clba.attr,
+	&dev_attr_ws_min.attr,
+	&dev_attr_ws_opt.attr,
+	&dev_attr_mw_cunits.attr,
+
+	&dev_attr_read_typ.attr,
+	&dev_attr_read_max.attr,
+	&dev_attr_write_typ.attr,
+	&dev_attr_write_max.attr,
+	&dev_attr_reset_typ.attr,
+	&dev_attr_reset_max.attr,
+
 	NULL,
 };
 
-static const struct attribute_group nvm_dev_attr_group = {
+static const struct attribute_group nvm_dev_attr_group_20 = {
 	.name		= "lightnvm",
-	.attrs		= nvm_dev_attrs,
+	.attrs		= nvm_dev_attrs_20,
 };
 
 int nvme_nvm_register_sysfs(struct nvme_ns *ns)
 {
-	return sysfs_create_group(&disk_to_dev(ns->disk)->kobj,
-					&nvm_dev_attr_group);
+	if (!ns->ndev)
+		return -EINVAL;
+
+	switch (ns->ndev->identity.ver_id) {
+	case 1:
+		return sysfs_create_group(&disk_to_dev(ns->disk)->kobj,
+					&nvm_dev_attr_group_12);
+	case 2:
+		return sysfs_create_group(&disk_to_dev(ns->disk)->kobj,
+					&nvm_dev_attr_group_20);
+	}
+
+	return -EINVAL;
 }
 
 void nvme_nvm_unregister_sysfs(struct nvme_ns *ns)
 {
-	sysfs_remove_group(&disk_to_dev(ns->disk)->kobj,
-					&nvm_dev_attr_group);
+	switch (ns->ndev->identity.ver_id) {
+	case 1:
+		sysfs_remove_group(&disk_to_dev(ns->disk)->kobj,
+					&nvm_dev_attr_group_12);
+		break;
+	case 2:
+		sysfs_remove_group(&disk_to_dev(ns->disk)->kobj,
+					&nvm_dev_attr_group_20);
+		break;
+	}
 }
diff --git a/include/linux/lightnvm.h b/include/linux/lightnvm.h
index 94b704a8d83d..b717c000b712 100644
--- a/include/linux/lightnvm.h
+++ b/include/linux/lightnvm.h
@@ -184,10 +184,9 @@ struct nvm_id {
 	u16	csecs;
 	u16	sos;
 
-	u16	ws_min;
-	u16	ws_opt;
-	u16	ws_seq;
-	u16	ws_per_chk;
+	u32	ws_min;
+	u32	ws_opt;
+	u32	mw_cunits;
 
 	u32	trdt;
 	u32	trdm;
@@ -199,6 +198,10 @@ struct nvm_id {
 	u32	mccap;
 	u16	cpar;
 
+	/* calculated values */
+	u16	ws_seq;
+	u16	ws_per_chk;
+
 	/* 1.2 compatibility */
 	u8	mtype;
 	u8	fmtype;
-- 
2.11.0

[PATCH v2 4/6] lightnvm: remove max_rq_size

[Matias Bjørling]

The field is no longer used.

Signed-off-by: Matias Bjørling <mb@lightnvm.io>
---
 drivers/lightnvm/core.c  | 1 -
 include/linux/lightnvm.h | 2 --
 2 files changed, 3 deletions(-)

diff --git a/drivers/lightnvm/core.c b/drivers/lightnvm/core.c
index 9b1255b3e05e..511c81c319ee 100644
--- a/drivers/lightnvm/core.c
+++ b/drivers/lightnvm/core.c
@@ -874,7 +874,6 @@ static int nvm_core_init(struct nvm_dev *dev)
 	geo->sec_size = id->csecs;
 	geo->oob_size = id->sos;
 	geo->mccap = id->mccap;
-	geo->max_rq_size = dev->ops->max_phys_sect * geo->sec_size;
 
 	geo->sec_per_chk = id->clba;
 	geo->sec_per_lun = geo->sec_per_chk * geo->nr_chks;
diff --git a/include/linux/lightnvm.h b/include/linux/lightnvm.h
index b717c000b712..67b4fa8e4906 100644
--- a/include/linux/lightnvm.h
+++ b/include/linux/lightnvm.h
@@ -295,8 +295,6 @@ struct nvm_geo {
 	int ws_seq;
 	int ws_per_chk;
 
-	int max_rq_size;
-
 	int op;
 
 	struct nvm_addr_format ppaf;
-- 
2.11.0

[PATCH v2 5/6] lightnvm: remove nvm_dev_ops->max_phys_sect

[Matias Bjørling]

The value of max_phys_sect is always static. Instead of
defining it in the nvm_dev_ops structure, declare it as a global
value.

Signed-off-by: Matias Bjørling <mb@lightnvm.io>
---
 drivers/lightnvm/core.c          | 28 +++++++---------------------
 drivers/lightnvm/pblk-init.c     |  9 ++++-----
 drivers/lightnvm/pblk-recovery.c |  8 ++------
 drivers/nvme/host/lightnvm.c     |  5 +----
 include/linux/lightnvm.h         |  5 ++---
 5 files changed, 16 insertions(+), 39 deletions(-)

diff --git a/drivers/lightnvm/core.c b/drivers/lightnvm/core.c
index 511c81c319ee..782f381e4d61 100644
--- a/drivers/lightnvm/core.c
+++ b/drivers/lightnvm/core.c
@@ -407,7 +407,8 @@ static int nvm_create_tgt(struct nvm_dev *dev, struct nvm_ioctl_create *create)
 	tdisk->private_data = targetdata;
 	tqueue->queuedata = targetdata;
 
-	blk_queue_max_hw_sectors(tqueue, 8 * dev->ops->max_phys_sect);
+	blk_queue_max_hw_sectors(tqueue,
+			(dev->geo.sec_size >> 9) * NVM_MAX_VLBA);
 
 	set_capacity(tdisk, tt->capacity(targetdata));
 	add_disk(tdisk);
@@ -719,7 +720,7 @@ int nvm_set_tgt_bb_tbl(struct nvm_tgt_dev *tgt_dev, struct ppa_addr *ppas,
 	struct nvm_rq rqd;
 	int ret;
 
-	if (nr_ppas > dev->ops->max_phys_sect) {
+	if (nr_ppas > NVM_MAX_VLBA) {
 		pr_err("nvm: unable to update all blocks atomically\n");
 		return -EINVAL;
 	}
@@ -740,14 +741,6 @@ int nvm_set_tgt_bb_tbl(struct nvm_tgt_dev *tgt_dev, struct ppa_addr *ppas,
 }
 EXPORT_SYMBOL(nvm_set_tgt_bb_tbl);
 
-int nvm_max_phys_sects(struct nvm_tgt_dev *tgt_dev)
-{
-	struct nvm_dev *dev = tgt_dev->parent;
-
-	return dev->ops->max_phys_sect;
-}
-EXPORT_SYMBOL(nvm_max_phys_sects);
-
 int nvm_submit_io(struct nvm_tgt_dev *tgt_dev, struct nvm_rq *rqd)
 {
 	struct nvm_dev *dev = tgt_dev->parent;
@@ -965,17 +958,10 @@ int nvm_register(struct nvm_dev *dev)
 	if (!dev->q || !dev->ops)
 		return -EINVAL;
 
-	if (dev->ops->max_phys_sect > 256) {
-		pr_info("nvm: max sectors supported is 256.\n");
-		return -EINVAL;
-	}
-
-	if (dev->ops->max_phys_sect > 1) {
-		dev->dma_pool = dev->ops->create_dma_pool(dev, "ppalist");
-		if (!dev->dma_pool) {
-			pr_err("nvm: could not create dma pool\n");
-			return -ENOMEM;
-		}
+	dev->dma_pool = dev->ops->create_dma_pool(dev, "ppalist");
+	if (!dev->dma_pool) {
+		pr_err("nvm: could not create dma pool\n");
+		return -ENOMEM;
 	}
 
 	ret = nvm_init(dev);
diff --git a/drivers/lightnvm/pblk-init.c b/drivers/lightnvm/pblk-init.c
index 86a94a7faa96..5261702e9ff7 100644
--- a/drivers/lightnvm/pblk-init.c
+++ b/drivers/lightnvm/pblk-init.c
@@ -260,8 +260,7 @@ static int pblk_core_init(struct pblk *pblk)
 		return -ENOMEM;
 
 	/* Internal bios can be at most the sectors signaled by the device. */
-	pblk->page_bio_pool = mempool_create_page_pool(nvm_max_phys_sects(dev),
-									0);
+	pblk->page_bio_pool = mempool_create_page_pool(NVM_MAX_VLBA, 0);
 	if (!pblk->page_bio_pool)
 		goto free_global_caches;
 
@@ -716,12 +715,12 @@ static int pblk_lines_init(struct pblk *pblk)
 
 	pblk->min_write_pgs = geo->sec_per_pl * (geo->sec_size / PAGE_SIZE);
 	max_write_ppas = pblk->min_write_pgs * geo->all_luns;
-	pblk->max_write_pgs = (max_write_ppas < nvm_max_phys_sects(dev)) ?
-				max_write_ppas : nvm_max_phys_sects(dev);
+	pblk->max_write_pgs = min_t(int, max_write_ppas, NVM_MAX_VLBA);
 	pblk_set_sec_per_write(pblk, pblk->min_write_pgs);
 
 	if (pblk->max_write_pgs > PBLK_MAX_REQ_ADDRS) {
-		pr_err("pblk: cannot support device max_phys_sect\n");
+		pr_err("pblk: vector list too big(%u > %u)\n",
+				pblk->max_write_pgs, PBLK_MAX_REQ_ADDRS);
 		return -EINVAL;
 	}
 
diff --git a/drivers/lightnvm/pblk-recovery.c b/drivers/lightnvm/pblk-recovery.c
index e75a1af2eebe..aaab9a5c17cc 100644
--- a/drivers/lightnvm/pblk-recovery.c
+++ b/drivers/lightnvm/pblk-recovery.c
@@ -21,17 +21,15 @@ void pblk_submit_rec(struct work_struct *work)
 	struct pblk_rec_ctx *recovery =
 			container_of(work, struct pblk_rec_ctx, ws_rec);
 	struct pblk *pblk = recovery->pblk;
-	struct nvm_tgt_dev *dev = pblk->dev;
 	struct nvm_rq *rqd = recovery->rqd;
 	struct pblk_c_ctx *c_ctx = nvm_rq_to_pdu(rqd);
-	int max_secs = nvm_max_phys_sects(dev);
 	struct bio *bio;
 	unsigned int nr_rec_secs;
 	unsigned int pgs_read;
 	int ret;
 
 	nr_rec_secs = bitmap_weight((unsigned long int *)&rqd->ppa_status,
-								max_secs);
+								NVM_MAX_VLBA);
 
 	bio = bio_alloc(GFP_KERNEL, nr_rec_secs);
 
@@ -74,8 +72,6 @@ int pblk_recov_setup_rq(struct pblk *pblk, struct pblk_c_ctx *c_ctx,
 			struct pblk_rec_ctx *recovery, u64 *comp_bits,
 			unsigned int comp)
 {
-	struct nvm_tgt_dev *dev = pblk->dev;
-	int max_secs = nvm_max_phys_sects(dev);
 	struct nvm_rq *rec_rqd;
 	struct pblk_c_ctx *rec_ctx;
 	int nr_entries = c_ctx->nr_valid + c_ctx->nr_padded;
@@ -86,7 +82,7 @@ int pblk_recov_setup_rq(struct pblk *pblk, struct pblk_c_ctx *c_ctx,
 	/* Copy completion bitmap, but exclude the first X completed entries */
 	bitmap_shift_right((unsigned long int *)&rec_rqd->ppa_status,
 				(unsigned long int *)comp_bits,
-				comp, max_secs);
+				comp, NVM_MAX_VLBA);
 
 	/* Save the context for the entries that need to be re-written and
 	 * update current context with the completed entries.
diff --git a/drivers/nvme/host/lightnvm.c b/drivers/nvme/host/lightnvm.c
index 8b243af8a949..e38d835b15b5 100644
--- a/drivers/nvme/host/lightnvm.c
+++ b/drivers/nvme/host/lightnvm.c
@@ -612,8 +612,6 @@ static struct nvm_dev_ops nvme_nvm_dev_ops = {
 	.destroy_dma_pool	= nvme_nvm_destroy_dma_pool,
 	.dev_dma_alloc		= nvme_nvm_dev_dma_alloc,
 	.dev_dma_free		= nvme_nvm_dev_dma_free,
-
-	.max_phys_sect		= 64,
 };
 
 static int nvme_nvm_submit_user_cmd(struct request_queue *q,
@@ -932,8 +930,7 @@ static ssize_t nvm_dev_attr_show_12(struct device *dev,
 	} else if (strcmp(attr->name, "media_capabilities") == 0) {
 		return scnprintf(page, PAGE_SIZE, "0x%08x\n", id->mccap);
 	} else if (strcmp(attr->name, "max_phys_secs") == 0) {
-		return scnprintf(page, PAGE_SIZE, "%u\n",
-				ndev->ops->max_phys_sect);
+		return scnprintf(page, PAGE_SIZE, "%u\n", NVM_MAX_VLBA);
 	} else {
 		return scnprintf(page,
 				 PAGE_SIZE,
diff --git a/include/linux/lightnvm.h b/include/linux/lightnvm.h
index 67b4fa8e4906..e55b10573c99 100644
--- a/include/linux/lightnvm.h
+++ b/include/linux/lightnvm.h
@@ -73,8 +73,6 @@ struct nvm_dev_ops {
 	nvm_destroy_dma_pool_fn	*destroy_dma_pool;
 	nvm_dev_dma_alloc_fn	*dev_dma_alloc;
 	nvm_dev_dma_free_fn	*dev_dma_free;
-
-	unsigned int		max_phys_sect;
 };
 
 #ifdef CONFIG_NVM
@@ -228,6 +226,8 @@ struct nvm_target {
 #define NVM_VERSION_MINOR 0
 #define NVM_VERSION_PATCH 0
 
+#define NVM_MAX_VLBA (64) /* max logical blocks in a vector command */
+
 struct nvm_rq;
 typedef void (nvm_end_io_fn)(struct nvm_rq *);
 
@@ -436,7 +436,6 @@ extern void nvm_unregister(struct nvm_dev *);
 
 extern int nvm_set_tgt_bb_tbl(struct nvm_tgt_dev *, struct ppa_addr *,
 			      int, int);
-extern int nvm_max_phys_sects(struct nvm_tgt_dev *);
 extern int nvm_submit_io(struct nvm_tgt_dev *, struct nvm_rq *);
 extern int nvm_submit_io_sync(struct nvm_tgt_dev *, struct nvm_rq *);
 extern void nvm_end_io(struct nvm_rq *);
-- 
2.11.0

[PATCH v2 6/6] nvme: lightnvm: add late setup of block size and metadata

[Matias Bjørling]

The nvme driver sets up the size of the nvme namespace in two steps.
First it initializes the device with standard logical block and
metadata sizes, and then sets the correct logical block and metadata
size. Due to the OCSSD 2.0 specification relies on the namespace to
expose these sizes for correct initialization, let it be updated
appropriately on the LightNVM side as well.

Signed-off-by: Matias Bjørling <mb@lightnvm.io>
---
 drivers/lightnvm/core.c      | 3 ---
 drivers/nvme/host/core.c     | 2 ++
 drivers/nvme/host/lightnvm.c | 8 ++++++++
 drivers/nvme/host/nvme.h     | 2 ++
 4 files changed, 12 insertions(+), 3 deletions(-)

diff --git a/drivers/lightnvm/core.c b/drivers/lightnvm/core.c
index 782f381e4d61..689c97b97775 100644
--- a/drivers/lightnvm/core.c
+++ b/drivers/lightnvm/core.c
@@ -864,8 +864,6 @@ static int nvm_core_init(struct nvm_dev *dev)
 	geo->ws_seq = id->ws_seq;
 	geo->ws_per_chk = id->ws_per_chk;
 	geo->nr_chks = id->num_chk;
-	geo->sec_size = id->csecs;
-	geo->oob_size = id->sos;
 	geo->mccap = id->mccap;
 
 	geo->sec_per_chk = id->clba;
@@ -893,7 +891,6 @@ static int nvm_core_init(struct nvm_dev *dev)
 	if (ret)
 		goto err_fmtype;
 
-	blk_queue_logical_block_size(dev->q, geo->sec_size);
 	return 0;
 err_fmtype:
 	kfree(dev->lun_map);
diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c
index f837d666cbd4..740ceb28067c 100644
--- a/drivers/nvme/host/core.c
+++ b/drivers/nvme/host/core.c
@@ -1379,6 +1379,8 @@ static void __nvme_revalidate_disk(struct gendisk *disk, struct nvme_id_ns *id)
 	if (ns->noiob)
 		nvme_set_chunk_size(ns);
 	nvme_update_disk_info(disk, ns, id);
+	if (ns->ndev)
+		nvme_nvm_update_nvm_info(ns);
 #ifdef CONFIG_NVME_MULTIPATH
 	if (ns->head->disk)
 		nvme_update_disk_info(ns->head->disk, ns, id);
diff --git a/drivers/nvme/host/lightnvm.c b/drivers/nvme/host/lightnvm.c
index e38d835b15b5..839c0b96466a 100644
--- a/drivers/nvme/host/lightnvm.c
+++ b/drivers/nvme/host/lightnvm.c
@@ -812,6 +812,14 @@ int nvme_nvm_ioctl(struct nvme_ns *ns, unsigned int cmd, unsigned long arg)
 	}
 }
 
+void nvme_nvm_update_nvm_info(struct nvme_ns *ns)
+{
+	struct nvm_dev *ndev = ns->ndev;
+
+	ndev->identity.csecs = ndev->geo.sec_size = 1 << ns->lba_shift;
+	ndev->identity.sos = ndev->geo.oob_size = ns->ms;
+}
+
 int nvme_nvm_register(struct nvme_ns *ns, char *disk_name, int node)
 {
 	struct request_queue *q = ns->queue;
diff --git a/drivers/nvme/host/nvme.h b/drivers/nvme/host/nvme.h
index ea1aa5283e8e..1ca08f4993ba 100644
--- a/drivers/nvme/host/nvme.h
+++ b/drivers/nvme/host/nvme.h
@@ -451,12 +451,14 @@ static inline void nvme_mpath_clear_current_path(struct nvme_ns *ns)
 #endif /* CONFIG_NVME_MULTIPATH */
 
 #ifdef CONFIG_NVM
+void nvme_nvm_update_nvm_info(struct nvme_ns *ns);
 int nvme_nvm_register(struct nvme_ns *ns, char *disk_name, int node);
 void nvme_nvm_unregister(struct nvme_ns *ns);
 int nvme_nvm_register_sysfs(struct nvme_ns *ns);
 void nvme_nvm_unregister_sysfs(struct nvme_ns *ns);
 int nvme_nvm_ioctl(struct nvme_ns *ns, unsigned int cmd, unsigned long arg);
 #else
+static inline void nvme_nvm_update_nvm_info(struct nvme_ns *ns) {};
 static inline int nvme_nvm_register(struct nvme_ns *ns, char *disk_name,
 				    int node)
 {
-- 
2.11.0

Re: [PATCH v2 5/6] lightnvm: remove nvm_dev_ops->max_phys_sect

[Javier Gonzalez]

[-- Attachment #1: Type: text/plain, Size: 1029 bytes --]


> On 15 Feb 2018, at 05.11, Matias Bjørling <mb@lightnvm.io> wrote:
> 
> The value of max_phys_sect is always static. Instead of
> defining it in the nvm_dev_ops structure, declare it as a global
> value.
> 
> Signed-off-by: Matias Bjørling <mb@lightnvm.io>
> ---
> drivers/lightnvm/core.c          | 28 +++++++---------------------
> drivers/lightnvm/pblk-init.c     |  9 ++++-----
> drivers/lightnvm/pblk-recovery.c |  8 ++------
> drivers/nvme/host/lightnvm.c     |  5 +----
> include/linux/lightnvm.h         |  5 ++---
> 5 files changed, 16 insertions(+), 39 deletions(-)
> 

The patch looks good, but I have a question. If a target implements the
scalar interface, then it will not be limited to 64 lbas/ppas and it
will not make sense to split the bio base don this value. In fact, it
looks like in time, we will move to a scalar interface in the 2.0 path
to align with the zoned interface, so this value will be dependent on
whether the target is using the scalar or vector interface.

Javier

[-- Attachment #2: Message signed with OpenPGP --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: [PATCH v2 5/6] lightnvm: remove nvm_dev_ops->max_phys_sect

[Matias Bjørling]

On 02/16/2018 07:48 AM, Javier Gonzalez wrote:
> 
>> On 15 Feb 2018, at 05.11, Matias Bjørling <mb@lightnvm.io> wrote:
>>
>> The value of max_phys_sect is always static. Instead of
>> defining it in the nvm_dev_ops structure, declare it as a global
>> value.
>>
>> Signed-off-by: Matias Bjørling <mb@lightnvm.io>
>> ---
>> drivers/lightnvm/core.c          | 28 +++++++---------------------
>> drivers/lightnvm/pblk-init.c     |  9 ++++-----
>> drivers/lightnvm/pblk-recovery.c |  8 ++------
>> drivers/nvme/host/lightnvm.c     |  5 +----
>> include/linux/lightnvm.h         |  5 ++---
>> 5 files changed, 16 insertions(+), 39 deletions(-)
>>
> 
> The patch looks good, but I have a question. If a target implements the
> scalar interface, then it will not be limited to 64 lbas/ppas and it
> will not make sense to split the bio base don this value. In fact, it
> looks like in time, we will move to a scalar interface in the 2.0 path
> to align with the zoned interface, so this value will be dependent on
> whether the target is using the scalar or vector interface.
> 

Both read/write and vector interface will coexist. I am only removing 
what is hardwired into the specification.

The read/write interface has always been able issue more than 64 LBAs, 
it is instead limited by what the hardware reports its max transfer size 
to be.

Re: [PATCH v2 5/6] lightnvm: remove nvm_dev_ops->max_phys_sect

[Javier Gonzalez]

[-- Attachment #1: Type: text/plain, Size: 1803 bytes --]

> On 19 Feb 2018, at 08.31, Matias Bjørling <mb@lightnvm.io> wrote:
> 
> On 02/16/2018 07:48 AM, Javier Gonzalez wrote:
>>> On 15 Feb 2018, at 05.11, Matias Bjørling <mb@lightnvm.io> wrote:
>>> 
>>> The value of max_phys_sect is always static. Instead of
>>> defining it in the nvm_dev_ops structure, declare it as a global
>>> value.
>>> 
>>> Signed-off-by: Matias Bjørling <mb@lightnvm.io>
>>> ---
>>> drivers/lightnvm/core.c          | 28 +++++++---------------------
>>> drivers/lightnvm/pblk-init.c     |  9 ++++-----
>>> drivers/lightnvm/pblk-recovery.c |  8 ++------
>>> drivers/nvme/host/lightnvm.c     |  5 +----
>>> include/linux/lightnvm.h         |  5 ++---
>>> 5 files changed, 16 insertions(+), 39 deletions(-)
>> The patch looks good, but I have a question. If a target implements the
>> scalar interface, then it will not be limited to 64 lbas/ppas and it
>> will not make sense to split the bio base don this value. In fact, it
>> looks like in time, we will move to a scalar interface in the 2.0 path
>> to align with the zoned interface, so this value will be dependent on
>> whether the target is using the scalar or vector interface.
> 
> Both read/write and vector interface will coexist. I am only removing
> what is hardwired into the specification.
> 
> The read/write interface has always been able issue more than 64 LBAs,
> it is instead limited by what the hardware reports its max transfer
> size to be.
> 

Exactly. I was thinking of a similar mechanism for the vector interface
to simplify integration with the scalar interface and avoid having an
if/else for what we now call max_phys_sect.

I guess we can wait and see what the code looks like when we adapt pblk.

Reviewed-by: Javier González <javier@cnexlabs.com>

Javier

[-- Attachment #2: Message signed with OpenPGP --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: [PATCH v2 0/6] lightnvm: base 2.0 implementation

[Javier Gonzalez]

[-- Attachment #1: Type: text/plain, Size: 2187 bytes --]

> On 15 Feb 2018, at 14.11, Matias Bjørling <mb@lightnvm.io> wrote:
> 
> A couple of patches for 2.0 support for the lightnvm subsystem. They
> form the foundation for the integration.
> 
> The first two patches is preparation for the 2.0 work. The third patch
> implements the 2.0 data structures, the geometry command, and exposes
> the sysfs attributes that comes with the 2.0 specification. Note that
> the attributes between 1.2 and 2.0 are different, and it is expected
> that user-space shall use the version sysfs attribute to know which
> attributes will be available.
> 
> The next two patches removes max_phys_sect and max_rq_size, as they
> not used.
> 
> The last patch implements support for using the nvme namespace logical
> block and metadata fields and sync it with the internal lightnvm
> identify structures.
> 
> Changes since v2:
> 
> - Removed blk_queue_block_size() setup in nvm_init and made sure
>   to only update csecs and sos in on the late setup path. No reason
>   to set it twice. From discussion with Javier.
> - Added two extra patches, that removes max_phys_sect and
>   max_rq_size.
> 
> Changes since v1:
> 
> - pr_err fix from Randy.
> - Address type fix from Javier.
> - Also CC the nvme mailing list.
> 
> Matias Bjørling (6):
>  lightnvm: make 1.2 data structures explicit
>  lightnvm: flatten nvm_id_group into nvm_id
>  lightnvm: add 2.0 geometry identification
>  lightnvm: remove max_rq_size
>  lightnvm: remove nvm_dev_ops->max_phys_sect
>  nvme: lightnvm: add late setup of block size and metadata
> 
> drivers/lightnvm/core.c          |  61 ++---
> drivers/lightnvm/pblk-init.c     |   9 +-
> drivers/lightnvm/pblk-recovery.c |   8 +-
> drivers/nvme/host/core.c         |   2 +
> drivers/nvme/host/lightnvm.c     | 513 ++++++++++++++++++++++++++++-----------
> drivers/nvme/host/nvme.h         |   2 +
> include/linux/lightnvm.h         |  71 +++---
> 7 files changed, 442 insertions(+), 224 deletions(-)
> 
> --
> 2.11.0
> 

The patches look good. I tested them together with pblk's 2.0 support
and all works as it should.

Reviewed-by: Javier González <javier@cnexlabs.com>


[-- Attachment #2: Message signed with OpenPGP --]
[-- Type: application/pgp-signature, Size: 833 bytes --]