Skip to contentSkip to conversation

mm/nvdimm: Add PFN_MIN_VERSION support

RFC PATCH In review Aneesh Kumar K.V 2 participants 5 messages May 22, 2019 – May 30, 2019

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

RFC PATCH v2 1/3 mm/nvdimm: Add PFN_MIN_VERSION support

This allows us to make changes in a backward incompatible way. I have kept the PFN_MIN_VERSION in this patch '0' because we are not introducing any incompatible changes in this patch. We also may want to backport this to older kernels.

The error looks like

dax0.1: init failed, superblock min version 1, kernel support version 0

and the namespace is marked disabled

$ndctl list -Ni
[

  {
    "dev":"namespace0.0",
    "mode":"fsdax",
    "map":"mem",
    "size":10737418240,
    "uuid":"9605de6d-cefa-4a87-99cd-dec28b02cffe",
    "state":"disabled"

}
]

---
 drivers/nvdimm/pfn.h      |  9 ++++++++-
 drivers/nvdimm/pfn_devs.c |  8 ++++++++
 drivers/nvdimm/pmem.c     | 26 ++++++++++++++++++++++----
 3 files changed, 38 insertions(+), 5 deletions(-)
drivers/nvdimm/pfn.h +8 -1
index dde9853453d3..5fd29242745a 100644
@@ -20,6 +20,12 @@
 #define PFN_SIG_LEN 16
 #define PFN_SIG "NVDIMM_PFN_INFO\0"
 #define DAX_SIG "NVDIMM_DAX_INFO\0"
+/*
+ * increment this when we are making changes such that older
+ * kernel should fail to initialize that namespace.
+ */
+
+#define PFN_MIN_VERSION 0
 
 struct nd_pfn_sb {
 	u8 signature[PFN_SIG_LEN];
@@ -36,7 +42,8 @@ struct nd_pfn_sb {
 	__le32 end_trunc;
 	/* minor-version-2 record the base alignment of the mapping */
 	__le32 align;
-	u8 padding[4000];
+	__le16 min_version;
+	u8 padding[3998];
 	__le64 checksum;
 };
 
drivers/nvdimm/pfn_devs.c +8 -0
index 01f40672507f..a2268cf262f5 100644
@@ -439,6 +439,13 @@ int nd_pfn_validate(struct nd_pfn *nd_pfn, const char *sig)
 	if (nvdimm_read_bytes(ndns, SZ_4K, pfn_sb, sizeof(*pfn_sb), 0))
 		return -ENXIO;
 
+	if (le16_to_cpu(pfn_sb->min_version) > PFN_MIN_VERSION) {
+		dev_err(&nd_pfn->dev,
+			"init failed, superblock min version %ld kernel support version %ld\n",
+			le16_to_cpu(pfn_sb->min_version), PFN_MIN_VERSION);
+		return -EOPNOTSUPP;
+	}
+
 	if (memcmp(pfn_sb->signature, sig, PFN_SIG_LEN) != 0)
 		return -ENODEV;
 
@@ -769,6 +776,7 @@ static int nd_pfn_init(struct nd_pfn *nd_pfn)
 	memcpy(pfn_sb->parent_uuid, nd_dev_to_uuid(&ndns->dev), 16);
 	pfn_sb->version_major = cpu_to_le16(1);
 	pfn_sb->version_minor = cpu_to_le16(2);
+	pfn_sb->min_version = cpu_to_le16(PFN_MIN_VERSION);
 	pfn_sb->start_pad = cpu_to_le32(start_pad);
 	pfn_sb->end_trunc = cpu_to_le32(end_trunc);
 	pfn_sb->align = cpu_to_le32(nd_pfn->align);
drivers/nvdimm/pmem.c +22 -4
index 845c5b430cdd..406427c064d9 100644
@@ -490,6 +490,7 @@ static int pmem_attach_disk(struct device *dev,
 
 static int nd_pmem_probe(struct device *dev)
 {
+	int ret;
 	struct nd_namespace_common *ndns;
 
 	ndns = nvdimm_namespace_common_probe(dev);
@@ -505,12 +506,29 @@ static int nd_pmem_probe(struct device *dev)
 	if (is_nd_pfn(dev))
 		return pmem_attach_disk(dev, ndns);
 
-	/* if we find a valid info-block we'll come back as that personality */
-	if (nd_btt_probe(dev, ndns) == 0 || nd_pfn_probe(dev, ndns) == 0
-			|| nd_dax_probe(dev, ndns) == 0)
+	ret = nd_btt_probe(dev, ndns);
+	if (ret == 0)
 		return -ENXIO;
+	else if (ret == -EOPNOTSUPP)
+		return ret;
 
-	/* ...otherwise we're just a raw pmem device */
+	ret = nd_pfn_probe(dev, ndns);
+	if (ret == 0)
+		return -ENXIO;
+	else if (ret == -EOPNOTSUPP)
+		return ret;
+
+	ret = nd_dax_probe(dev, ndns);
+	if (ret == 0)
+		return -ENXIO;
+	else if (ret == -EOPNOTSUPP)
+		return ret;
+	/*
+	 * We have two failure conditions here, there is no
+	 * info reserver block or we found a valid info reserve block
+	 * but failed to initialize the pfn superblock.
+	 * Don't create a raw pmem disk for the second case.
+	 */
 	return pmem_attach_disk(dev, ndns);
 }
 
Signature · 1 line
2.21.0
#1 lore raw

RFC PATCH v2 2/3 mm/nvdimm: Add page size and struct page size to pfn superblock

This is needed so that we don't wrongly initialize a namespace which doesn't have enough space reserved for holding struct pages with the current kernel.

We also increment PFN_MIN_VERSION to make sure that older kernel won't initialize namespace created with newer kernel.

---
 drivers/nvdimm/pfn.h      |  7 +++++--
 drivers/nvdimm/pfn_devs.c | 19 ++++++++++++++++++-
 2 files changed, 23 insertions(+), 3 deletions(-)
drivers/nvdimm/pfn.h +5 -2
index 5fd29242745a..ba11738ca8a2 100644
@@ -25,7 +25,7 @@
  * kernel should fail to initialize that namespace.
  */
 
-#define PFN_MIN_VERSION 0
+#define PFN_MIN_VERSION 1
 
 struct nd_pfn_sb {
 	u8 signature[PFN_SIG_LEN];
@@ -43,7 +43,10 @@ struct nd_pfn_sb {
 	/* minor-version-2 record the base alignment of the mapping */
 	__le32 align;
 	__le16 min_version;
-	u8 padding[3998];
+	/* minor-version-3 record the page size and struct page size */
+	__le16 page_struct_size;
+	__le32 page_size;
+	u8 padding[3992];
 	__le64 checksum;
 };
 
drivers/nvdimm/pfn_devs.c +18 -1
index a2268cf262f5..39fa8cf8ef58 100644
@@ -466,6 +466,15 @@ int nd_pfn_validate(struct nd_pfn *nd_pfn, const char *sig)
 	if (__le16_to_cpu(pfn_sb->version_minor) < 2)
 		pfn_sb->align = 0;
 
+	if (__le16_to_cpu(pfn_sb->version_minor) < 3) {
+		/*
+		 * For a large part we use PAGE_SIZE. But we
+		 * do have some accounting code using SZ_4K.
+		 */
+		pfn_sb->page_struct_size = cpu_to_le16(64);
+		pfn_sb->page_size = cpu_to_le32(SZ_4K);
+	}
+
 	switch (le32_to_cpu(pfn_sb->mode)) {
 	case PFN_MODE_RAM:
 	case PFN_MODE_PMEM:
@@ -481,6 +490,12 @@ int nd_pfn_validate(struct nd_pfn *nd_pfn, const char *sig)
 		align = 1UL << ilog2(offset);
 	mode = le32_to_cpu(pfn_sb->mode);
 
+	if (le32_to_cpu(pfn_sb->page_size) != PAGE_SIZE)
+		return -EOPNOTSUPP;
+
+	if (le16_to_cpu(pfn_sb->page_struct_size) != sizeof(struct page))
+		return -EOPNOTSUPP;
+
 	if (!nd_pfn->uuid) {
 		/*
 		 * When probing a namepace via nd_pfn_probe() the uuid
@@ -775,11 +790,13 @@ static int nd_pfn_init(struct nd_pfn *nd_pfn)
 	memcpy(pfn_sb->uuid, nd_pfn->uuid, 16);
 	memcpy(pfn_sb->parent_uuid, nd_dev_to_uuid(&ndns->dev), 16);
 	pfn_sb->version_major = cpu_to_le16(1);
-	pfn_sb->version_minor = cpu_to_le16(2);
+	pfn_sb->version_minor = cpu_to_le16(3);
 	pfn_sb->min_version = cpu_to_le16(PFN_MIN_VERSION);
 	pfn_sb->start_pad = cpu_to_le32(start_pad);
 	pfn_sb->end_trunc = cpu_to_le32(end_trunc);
 	pfn_sb->align = cpu_to_le32(nd_pfn->align);
+	pfn_sb->page_struct_size = cpu_to_le16(sizeof(struct page));
+	pfn_sb->page_size = cpu_to_le32(PAGE_SIZE);
 	checksum = nd_sb_checksum((struct nd_gen_sb *) pfn_sb);
 	pfn_sb->checksum = cpu_to_le64(checksum);
 
Signature · 1 line
2.21.0
#2 lore raw

RFC PATCH v2 3/3 mm/nvdimm: Use correct #defines instead of opencoding

In reply to Aneesh Kumar K.V

The nfpn related change is needed to fix the kernel message

"number of pfns truncated from 2617344 to 163584"

The change makes sure the nfpns stored in the superblock is right value.

---
 drivers/nvdimm/label.c       | 2 +-
 drivers/nvdimm/pfn_devs.c    | 6 +++---
 drivers/nvdimm/region_devs.c | 8 ++++----
 3 files changed, 8 insertions(+), 8 deletions(-)
drivers/nvdimm/label.c +1 -1
index f3d753d3169c..bc6de8fb0153 100644
@@ -361,7 +361,7 @@ static bool slot_valid(struct nvdimm_drvdata *ndd,
 
 	/* check that DPA allocations are page aligned */
 	if ((__le64_to_cpu(nd_label->dpa)
-				| __le64_to_cpu(nd_label->rawsize)) % SZ_4K)
+				| __le64_to_cpu(nd_label->rawsize)) % PAGE_SIZE)
 		return false;
 
 	/* check checksum */
drivers/nvdimm/pfn_devs.c +3 -3
index 39fa8cf8ef58..9fc2e514e28a 100644
@@ -769,8 +769,8 @@ static int nd_pfn_init(struct nd_pfn *nd_pfn)
 		 * when populating the vmemmap. This *should* be equal to
 		 * PMD_SIZE for most architectures.
 		 */
-		offset = ALIGN(start + reserve + 64 * npfns,
-				max(nd_pfn->align, PMD_SIZE)) - start;
+		offset = ALIGN(start + reserve + sizeof(struct page) * npfns,
+			       max(nd_pfn->align, PMD_SIZE)) - start;
 	} else if (nd_pfn->mode == PFN_MODE_RAM)
 		offset = ALIGN(start + reserve, nd_pfn->align) - start;
 	else
@@ -782,7 +782,7 @@ static int nd_pfn_init(struct nd_pfn *nd_pfn)
 		return -ENXIO;
 	}
 
-	npfns = (size - offset - start_pad - end_trunc) / SZ_4K;
+	npfns = (size - offset - start_pad - end_trunc) / PAGE_SIZE;
 	pfn_sb->mode = cpu_to_le32(nd_pfn->mode);
 	pfn_sb->dataoff = cpu_to_le64(offset);
 	pfn_sb->npfns = cpu_to_le64(npfns);
drivers/nvdimm/region_devs.c +4 -4
index b4ef7d9ff22e..2d8facea5a03 100644
@@ -994,10 +994,10 @@ static struct nd_region *nd_region_create(struct nvdimm_bus *nvdimm_bus,
 		struct nd_mapping_desc *mapping = &ndr_desc->mapping[i];
 		struct nvdimm *nvdimm = mapping->nvdimm;
 
-		if ((mapping->start | mapping->size) % SZ_4K) {
-			dev_err(&nvdimm_bus->dev, "%s: %s mapping%d is not 4K aligned\n",
-					caller, dev_name(&nvdimm->dev), i);
-
+		if ((mapping->start | mapping->size) % PAGE_SIZE) {
+			dev_err(&nvdimm_bus->dev,
+				"%s: %s mapping%d is not 4K aligned\n",
+				caller, dev_name(&nvdimm->dev), i);
 			return NULL;
 		}
 
Signature · 1 line
2.21.0
#3 lore raw

Re: [RFC PATCH V2 3/3] mm/nvdimm: Use correct #defines instead of opencoding

Quoted 55 lines · Aneesh Kumar K.V

The nfpn related change is needed to fix the kernel message

"number of pfns truncated from 2617344 to 163584"

The change makes sure the nfpns stored in the superblock is right value.

---
 drivers/nvdimm/label.c       | 2 +-
 drivers/nvdimm/pfn_devs.c    | 6 +++---
 drivers/nvdimm/region_devs.c | 8 ++++----
 3 files changed, 8 insertions(+), 8 deletions(-)
drivers/nvdimm/label.c +1 -1
index f3d753d3169c..bc6de8fb0153 100644
@@ -361,7 +361,7 @@ static bool slot_valid(struct nvdimm_drvdata *ndd,
 
 	/* check that DPA allocations are page aligned */
 	if ((__le64_to_cpu(nd_label->dpa)
-				| __le64_to_cpu(nd_label->rawsize)) % SZ_4K)
+				| __le64_to_cpu(nd_label->rawsize)) % PAGE_SIZE)
 		return false;
 
 	/* check checksum */
drivers/nvdimm/pfn_devs.c +3 -3
index 39fa8cf8ef58..9fc2e514e28a 100644
@@ -769,8 +769,8 @@ static int nd_pfn_init(struct nd_pfn *nd_pfn)
 		 * when populating the vmemmap. This *should* be equal to
 		 * PMD_SIZE for most architectures.
 		 */
-		offset = ALIGN(start + reserve + 64 * npfns,
-				max(nd_pfn->align, PMD_SIZE)) - start;
+		offset = ALIGN(start + reserve + sizeof(struct page) * npfns,
+			       max(nd_pfn->align, PMD_SIZE)) - start;
 	} else if (nd_pfn->mode == PFN_MODE_RAM)
 		offset = ALIGN(start + reserve, nd_pfn->align) - start;
 	else
@@ -782,7 +782,7 @@ static int nd_pfn_init(struct nd_pfn *nd_pfn)
 		return -ENXIO;
 	}
 
-	npfns = (size - offset - start_pad - end_trunc) / SZ_4K;
+	npfns = (size - offset - start_pad - end_trunc) / PAGE_SIZE;
 	pfn_sb->mode = cpu_to_le32(nd_pfn->mode);
 	pfn_sb->dataoff = cpu_to_le64(offset);
 	pfn_sb->npfns = cpu_to_le64(npfns);
drivers/nvdimm/region_devs.c +0 -0
index b4ef7d9ff22e..2d8facea5a03 100644

drivers/nvdimm/region_devs.c static struct nd_region *nd_region_create(struct nvdimm_bus *nvdimm_bus, · line 999

Quoted 10 lines
drivers/nvdimm/region_devs.c +3 -4
@@ -994,10 +994,10 @@ static struct nd_region *nd_region_create(struct nvdimm_bus *nvdimm_bus,
 		struct nd_mapping_desc *mapping = &ndr_desc->mapping[i];
 		struct nvdimm *nvdimm = mapping->nvdimm;
 
-		if ((mapping->start | mapping->size) % SZ_4K) {
-			dev_err(&nvdimm_bus->dev, "%s: %s mapping%d is not 4K aligned\n",
-					caller, dev_name(&nvdimm->dev), i);
-
+		if ((mapping->start | mapping->size) % PAGE_SIZE) {
+			dev_err(&nvdimm_bus->dev,
+				"%s: %s mapping%d is not 4K aligned\n",

s/not 4K aligned/not PAGE_SIZE aligned ?

hope the error msg need to be changed as well..

Regards,
-Satheesh.

Quoted 7 lines
drivers/nvdimm/region_devs.c +1 -0
+				caller, dev_name(&nvdimm->dev), i);
 			return NULL;
 		}
 
Signature · 1 line
2.21.0

_______________________________________________
Linux-nvdimm mailing list
[email protected]
https://lists.01.org/mailman/listinfo/linux-nvdimm

#4 lore raw
In reply to Aneesh Kumar K.V

Hi Dan,

Are you ok with this patch series? If yes I can send a non-RFC version for this series. Since we are now marking all previously created pfn_sb on ppc64 as not supported, (pfn_sb->page_size = SZ_4K) I would like to get this merged early.

-aneesh

Quoted 131 lines · Aneesh Kumar K.V

This allows us to make changes in a backward incompatible way. I have kept the PFN_MIN_VERSION in this patch '0' because we are not introducing any incompatible changes in this patch. We also may want to backport this to older kernels.

The error looks like

dax0.1: init failed, superblock min version 1, kernel support version 0

and the namespace is marked disabled

$ndctl list -Ni
[

  {
    "dev":"namespace0.0",
    "mode":"fsdax",
    "map":"mem",
    "size":10737418240,
    "uuid":"9605de6d-cefa-4a87-99cd-dec28b02cffe",
    "state":"disabled"

}
]

---
 drivers/nvdimm/pfn.h      |  9 ++++++++-
 drivers/nvdimm/pfn_devs.c |  8 ++++++++
 drivers/nvdimm/pmem.c     | 26 ++++++++++++++++++++++----
 3 files changed, 38 insertions(+), 5 deletions(-)
drivers/nvdimm/pfn.h +8 -1
index dde9853453d3..5fd29242745a 100644
@@ -20,6 +20,12 @@
 #define PFN_SIG_LEN 16
 #define PFN_SIG "NVDIMM_PFN_INFO\0"
 #define DAX_SIG "NVDIMM_DAX_INFO\0"
+/*
+ * increment this when we are making changes such that older
+ * kernel should fail to initialize that namespace.
+ */
+
+#define PFN_MIN_VERSION 0
 
 struct nd_pfn_sb {
 	u8 signature[PFN_SIG_LEN];
@@ -36,7 +42,8 @@ struct nd_pfn_sb {
 	__le32 end_trunc;
 	/* minor-version-2 record the base alignment of the mapping */
 	__le32 align;
-	u8 padding[4000];
+	__le16 min_version;
+	u8 padding[3998];
 	__le64 checksum;
 };
 
drivers/nvdimm/pfn_devs.c +8 -0
index 01f40672507f..a2268cf262f5 100644
@@ -439,6 +439,13 @@ int nd_pfn_validate(struct nd_pfn *nd_pfn, const char *sig)
 	if (nvdimm_read_bytes(ndns, SZ_4K, pfn_sb, sizeof(*pfn_sb), 0))
 		return -ENXIO;
 
+	if (le16_to_cpu(pfn_sb->min_version) > PFN_MIN_VERSION) {
+		dev_err(&nd_pfn->dev,
+			"init failed, superblock min version %ld kernel support version %ld\n",
+			le16_to_cpu(pfn_sb->min_version), PFN_MIN_VERSION);
+		return -EOPNOTSUPP;
+	}
+
 	if (memcmp(pfn_sb->signature, sig, PFN_SIG_LEN) != 0)
 		return -ENODEV;
 
@@ -769,6 +776,7 @@ static int nd_pfn_init(struct nd_pfn *nd_pfn)
 	memcpy(pfn_sb->parent_uuid, nd_dev_to_uuid(&ndns->dev), 16);
 	pfn_sb->version_major = cpu_to_le16(1);
 	pfn_sb->version_minor = cpu_to_le16(2);
+	pfn_sb->min_version = cpu_to_le16(PFN_MIN_VERSION);
 	pfn_sb->start_pad = cpu_to_le32(start_pad);
 	pfn_sb->end_trunc = cpu_to_le32(end_trunc);
 	pfn_sb->align = cpu_to_le32(nd_pfn->align);
drivers/nvdimm/pmem.c +22 -4
index 845c5b430cdd..406427c064d9 100644
@@ -490,6 +490,7 @@ static int pmem_attach_disk(struct device *dev,
 
 static int nd_pmem_probe(struct device *dev)
 {
+	int ret;
 	struct nd_namespace_common *ndns;
 
 	ndns = nvdimm_namespace_common_probe(dev);
@@ -505,12 +506,29 @@ static int nd_pmem_probe(struct device *dev)
 	if (is_nd_pfn(dev))
 		return pmem_attach_disk(dev, ndns);
 
-	/* if we find a valid info-block we'll come back as that personality */
-	if (nd_btt_probe(dev, ndns) == 0 || nd_pfn_probe(dev, ndns) == 0
-			|| nd_dax_probe(dev, ndns) == 0)
+	ret = nd_btt_probe(dev, ndns);
+	if (ret == 0)
 		return -ENXIO;
+	else if (ret == -EOPNOTSUPP)
+		return ret;
 
-	/* ...otherwise we're just a raw pmem device */
+	ret = nd_pfn_probe(dev, ndns);
+	if (ret == 0)
+		return -ENXIO;
+	else if (ret == -EOPNOTSUPP)
+		return ret;
+
+	ret = nd_dax_probe(dev, ndns);
+	if (ret == 0)
+		return -ENXIO;
+	else if (ret == -EOPNOTSUPP)
+		return ret;
+	/*
+	 * We have two failure conditions here, there is no
+	 * info reserver block or we found a valid info reserve block
+	 * but failed to initialize the pfn superblock.
+	 * Don't create a raw pmem disk for the second case.
+	 */
 	return pmem_attach_disk(dev, ndns);
 }
 
Signature · 1 line
2.21.0

_______________________________________________
Linux-nvdimm mailing list
[email protected]
https://lists.01.org/mailman/listinfo/linux-nvdimm