Merge tag 'libnvdimm-for-5.1' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm

Pull libnvdimm updates from Dan Williams:
 "The bulk of this has been in -next since before the merge window
  opened, with no known collisions / issues reported.

  The only detail worth noting, outside the summary below, is that the
  "libnvdimm-start-pad" topic has been truncated to just cleanups and
  small fixes. The full topic branch would have doubled down on hacks
  around the "section alignment" limitation of the core-mm, instead
  effort is now being spent to address that root issue in the memory
  hotplug implementation for v5.2.

   - Fix nfit-bus command submission regression

   - Support retrieval of short-ARS results if the ARS state is
     "requires continuation", and even if the "no_init_ars" module
     parameter is specified

   - Allow busy-polling of the kernel ARS state by allowing root to
     reset the exponential back-off timer

   - Filter potentially stale ARS results by tracking query-ARS relative
     to the previous start-ARS

   - Enhance dax_device alignment checks

   - Add support for the Hyper-V family of device-specific-methods
     (DSMs)

   - Add several fixes and workarounds for Hyper-V compatibility

   - Fix support to cache the dirty-shutdown-count at init"

* tag 'libnvdimm-for-5.1' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm: (25 commits)
  libnvdimm/namespace: Clean up holder_class_store()
  libnvdimm/of_pmem: Fix platform_no_drv_owner.cocci warnings
  acpi/nfit: Update NFIT flags error message
  libnvdimm/btt: Fix LBA masking during 'free list' population
  libnvdimm/btt: Remove unnecessary code in btt_freelist_init
  libnvdimm/pfn: Remove dax_label_reserve
  dax: Check the end of the block-device capacity with dax_direct_access()
  nfit/ars: Avoid stale ARS results
  nfit/ars: Allow root to busy-poll the ARS state machine
  nfit/ars: Introduce scrub_flags
  nfit/ars: Remove ars_start_flags
  nfit/ars: Attempt short-ARS even in the no_init_ars case
  nfit/ars: Attempt a short-ARS whenever the ARS state is idle at boot
  acpi/nfit: Require opt-in for read-only label configurations
  libnvdimm/pmem: Honor force_raw for legacy pmem regions
  libnvdimm/pfn: Account for PAGE_SIZE > info-block-size in nd_pfn_init()
  libnvdimm: Fix altmap reservation size calculation
  libnvdimm, pfn: Fix over-trim in trim_pfn_device()
  acpi/nfit: Fix bus command validation
  libnvdimm/dimm: Add a no-BLK quirk based on NVDIMM family
  ...

drivers/nvdimm/btt.c

@@ -541,9 +541,9 @@ static int arena_clear_freelist_error(struct arena_info *arena, u32 lane)
 
 static int btt_freelist_init(struct arena_info *arena)
 {
-	int old, new, ret;
-	u32 i, map_entry;
-	struct log_entry log_new, log_old;
+	int new, ret;
+	struct log_entry log_new;
+	u32 i, map_entry, log_oldmap, log_newmap;
 
 	arena->freelist = kcalloc(arena->nfree, sizeof(struct free_entry),
 					GFP_KERNEL);
@@ -551,24 +551,26 @@ static int btt_freelist_init(struct arena_info *arena)
 		return -ENOMEM;
 
 	for (i = 0; i < arena->nfree; i++) {
-		old = btt_log_read(arena, i, &log_old, LOG_OLD_ENT);
-		if (old < 0)
-			return old;
-
 		new = btt_log_read(arena, i, &log_new, LOG_NEW_ENT);
 		if (new < 0)
 			return new;
 
+		/* old and new map entries with any flags stripped out */
+		log_oldmap = ent_lba(le32_to_cpu(log_new.old_map));
+		log_newmap = ent_lba(le32_to_cpu(log_new.new_map));
+
 		/* sub points to the next one to be overwritten */
 		arena->freelist[i].sub = 1 - new;
 		arena->freelist[i].seq = nd_inc_seq(le32_to_cpu(log_new.seq));
-		arena->freelist[i].block = le32_to_cpu(log_new.old_map);
+		arena->freelist[i].block = log_oldmap;
 
 		/*
 		 * FIXME: if error clearing fails during init, we want to make
 		 * the BTT read-only
 		 */
-		if (ent_e_flag(log_new.old_map)) {
+		if (ent_e_flag(log_new.old_map) &&
+				!ent_normal(log_new.old_map)) {
+			arena->freelist[i].has_err = 1;
 			ret = arena_clear_freelist_error(arena, i);
 			if (ret)
 				dev_err_ratelimited(to_dev(arena),
@@ -576,7 +578,7 @@ static int btt_freelist_init(struct arena_info *arena)
 		}
 
 		/* This implies a newly created or untouched flog entry */
-		if (log_new.old_map == log_new.new_map)
+		if (log_oldmap == log_newmap)
 			continue;
 
 		/* Check if map recovery is needed */
@@ -584,8 +586,15 @@ static int btt_freelist_init(struct arena_info *arena)
 				NULL, NULL, 0);
 		if (ret)
 			return ret;
-		if ((le32_to_cpu(log_new.new_map) != map_entry) &&
-				(le32_to_cpu(log_new.old_map) == map_entry)) {
+
+		/*
+		 * The map_entry from btt_read_map is stripped of any flag bits,
+		 * so use the stripped out versions from the log as well for
+		 * testing whether recovery is needed. For restoration, use the
+		 * 'raw' version of the log entries as that captured what we
+		 * were going to write originally.
+		 */
+		if ((log_newmap != map_entry) && (log_oldmap == map_entry)) {
 			/*
 			 * Last transaction wrote the flog, but wasn't able
 			 * to complete the map write. So fix up the map.

drivers/nvdimm/btt.h

@@ -44,6 +44,8 @@
 #define ent_e_flag(ent) (!!(ent & MAP_ERR_MASK))
 #define ent_z_flag(ent) (!!(ent & MAP_TRIM_MASK))
 #define set_e_flag(ent) (ent |= MAP_ERR_MASK)
+/* 'normal' is both e and z flags set */
+#define ent_normal(ent) (ent_e_flag(ent) && ent_z_flag(ent))
 
 enum btt_init_state {
 	INIT_UNCHECKED = 0,

drivers/nvdimm/btt_devs.c

@@ -159,11 +159,19 @@ static ssize_t size_show(struct device *dev,
 }
 static DEVICE_ATTR_RO(size);
 
+static ssize_t log_zero_flags_show(struct device *dev,
+		struct device_attribute *attr, char *buf)
+{
+	return sprintf(buf, "Y\n");
+}
+static DEVICE_ATTR_RO(log_zero_flags);
+
 static struct attribute *nd_btt_attributes[] = {
 	&dev_attr_sector_size.attr,
 	&dev_attr_namespace.attr,
 	&dev_attr_uuid.attr,
 	&dev_attr_size.attr,
+	&dev_attr_log_zero_flags.attr,
 	NULL,
 };
 

drivers/nvdimm/dimm_devs.c

@@ -11,6 +11,7 @@
  * General Public License for more details.
  */
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+#include <linux/moduleparam.h>
 #include <linux/vmalloc.h>
 #include <linux/device.h>
 #include <linux/ndctl.h>
@@ -25,6 +26,10 @@
 
 static DEFINE_IDA(dimm_ida);
 
+static bool noblk;
+module_param(noblk, bool, 0444);
+MODULE_PARM_DESC(noblk, "force disable BLK / local alias support");
+
 /*
  * Retrieve bus and dimm handle and return if this bus supports
  * get_config_data commands
@@ -551,6 +556,8 @@ struct nvdimm *__nvdimm_create(struct nvdimm_bus *nvdimm_bus,
 
 	nvdimm->dimm_id = dimm_id;
 	nvdimm->provider_data = provider_data;
+	if (noblk)
+		flags |= 1 << NDD_NOBLK;
 	nvdimm->flags = flags;
 	nvdimm->cmd_mask = cmd_mask;
 	nvdimm->num_flush = num_flush;

drivers/nvdimm/label.c

@@ -392,6 +392,7 @@ int nd_label_reserve_dpa(struct nvdimm_drvdata *ndd)
 		return 0; /* no label, nothing to reserve */
 
 	for_each_clear_bit_le(slot, free, nslot) {
+		struct nvdimm *nvdimm = to_nvdimm(ndd->dev);
 		struct nd_namespace_label *nd_label;
 		struct nd_region *nd_region = NULL;
 		u8 label_uuid[NSLABEL_UUID_LEN];
@@ -406,6 +407,8 @@ int nd_label_reserve_dpa(struct nvdimm_drvdata *ndd)
 
 		memcpy(label_uuid, nd_label->uuid, NSLABEL_UUID_LEN);
 		flags = __le32_to_cpu(nd_label->flags);
+		if (test_bit(NDD_NOBLK, &nvdimm->flags))
+			flags &= ~NSLABEL_FLAG_LOCAL;
 		nd_label_gen_id(&label_id, label_uuid, flags);
 		res = nvdimm_allocate_dpa(ndd, &label_id,
 				__le64_to_cpu(nd_label->dpa),
@@ -755,7 +758,7 @@ static const guid_t *to_abstraction_guid(enum nvdimm_claim_class claim_class,
 
 static int __pmem_label_update(struct nd_region *nd_region,
 		struct nd_mapping *nd_mapping, struct nd_namespace_pmem *nspm,
-		int pos)
+		int pos, unsigned long flags)
 {
 	struct nd_namespace_common *ndns = &nspm->nsio.common;
 	struct nd_interleave_set *nd_set = nd_region->nd_set;
@@ -796,7 +799,7 @@ static int __pmem_label_update(struct nd_region *nd_region,
 	memcpy(nd_label->uuid, nspm->uuid, NSLABEL_UUID_LEN);
 	if (nspm->alt_name)
 		memcpy(nd_label->name, nspm->alt_name, NSLABEL_NAME_LEN);
-	nd_label->flags = __cpu_to_le32(NSLABEL_FLAG_UPDATING);
+	nd_label->flags = __cpu_to_le32(flags);
 	nd_label->nlabel = __cpu_to_le16(nd_region->ndr_mappings);
 	nd_label->position = __cpu_to_le16(pos);
 	nd_label->isetcookie = __cpu_to_le64(cookie);
@@ -1249,13 +1252,13 @@ static int del_labels(struct nd_mapping *nd_mapping, u8 *uuid)
 int nd_pmem_namespace_label_update(struct nd_region *nd_region,
 		struct nd_namespace_pmem *nspm, resource_size_t size)
 {
-	int i;
+	int i, rc;
 
 	for (i = 0; i < nd_region->ndr_mappings; i++) {
 		struct nd_mapping *nd_mapping = &nd_region->mapping[i];
 		struct nvdimm_drvdata *ndd = to_ndd(nd_mapping);
 		struct resource *res;
-		int rc, count = 0;
+		int count = 0;
 
 		if (size == 0) {
 			rc = del_labels(nd_mapping, nspm->uuid);
@@ -1273,7 +1276,20 @@ int nd_pmem_namespace_label_update(struct nd_region *nd_region,
 		if (rc < 0)
 			return rc;
 
-		rc = __pmem_label_update(nd_region, nd_mapping, nspm, i);
+		rc = __pmem_label_update(nd_region, nd_mapping, nspm, i,
+				NSLABEL_FLAG_UPDATING);
+		if (rc)
+			return rc;
+	}
+
+	if (size == 0)
+		return 0;
+
+	/* Clear the UPDATING flag per UEFI 2.7 expectations */
+	for (i = 0; i < nd_region->ndr_mappings; i++) {
+		struct nd_mapping *nd_mapping = &nd_region->mapping[i];
+
+		rc = __pmem_label_update(nd_region, nd_mapping, nspm, i, 0);
 		if (rc)
 			return rc;
 	}

drivers/nvdimm/namespace_devs.c

@@ -138,6 +138,7 @@ bool nd_is_uuid_unique(struct device *dev, u8 *uuid)
 bool pmem_should_map_pages(struct device *dev)
 {
 	struct nd_region *nd_region = to_nd_region(dev->parent);
+	struct nd_namespace_common *ndns = to_ndns(dev);
 	struct nd_namespace_io *nsio;
 
 	if (!IS_ENABLED(CONFIG_ZONE_DEVICE))
@@ -149,6 +150,9 @@ bool pmem_should_map_pages(struct device *dev)
 	if (is_nd_pfn(dev) || is_nd_btt(dev))
 		return false;
 
+	if (ndns->force_raw)
+		return false;
+
 	nsio = to_nd_namespace_io(dev);
 	if (region_intersects(nsio->res.start, resource_size(&nsio->res),
 				IORESOURCE_SYSTEM_RAM,
@@ -1506,13 +1510,13 @@ static ssize_t __holder_class_store(struct device *dev, const char *buf)
 	if (dev->driver || ndns->claim)
 		return -EBUSY;
 
-	if (strcmp(buf, "btt") == 0 || strcmp(buf, "btt\n") == 0)
+	if (sysfs_streq(buf, "btt"))
 		ndns->claim_class = btt_claim_class(dev);
-	else if (strcmp(buf, "pfn") == 0 || strcmp(buf, "pfn\n") == 0)
+	else if (sysfs_streq(buf, "pfn"))
 		ndns->claim_class = NVDIMM_CCLASS_PFN;
-	else if (strcmp(buf, "dax") == 0 || strcmp(buf, "dax\n") == 0)
+	else if (sysfs_streq(buf, "dax"))
 		ndns->claim_class = NVDIMM_CCLASS_DAX;
-	else if (strcmp(buf, "") == 0 || strcmp(buf, "\n") == 0)
+	else if (sysfs_streq(buf, ""))
 		ndns->claim_class = NVDIMM_CCLASS_NONE;
 	else
 		return -EINVAL;
@@ -2492,6 +2496,12 @@ static int init_active_labels(struct nd_region *nd_region)
 			if (!label_ent)
 				break;
 			label = nd_label_active(ndd, j);
+			if (test_bit(NDD_NOBLK, &nvdimm->flags)) {
+				u32 flags = __le32_to_cpu(label->flags);
+
+				flags &= ~NSLABEL_FLAG_LOCAL;
+				label->flags = __cpu_to_le32(flags);
+			}
 			label_ent->label = label;
 
 			mutex_lock(&nd_mapping->lock);

drivers/nvdimm/of_pmem.c

@@ -108,7 +108,6 @@ static struct platform_driver of_pmem_region_driver = {
 	.remove = of_pmem_region_remove,
 	.driver = {
 		.name = "of_pmem",
-		.owner = THIS_MODULE,
 		.of_match_table = of_pmem_region_match,
 	},
 };

drivers/nvdimm/pfn_devs.c

@@ -580,6 +580,11 @@ int nd_pfn_probe(struct device *dev, struct nd_namespace_common *ndns)
 }
 EXPORT_SYMBOL(nd_pfn_probe);
 
+static u32 info_block_reserve(void)
+{
+	return ALIGN(SZ_8K, PAGE_SIZE);
+}
+
 /*
  * We hotplug memory at section granularity, pad the reserved area from
  * the previous section base to the namespace base address.
@@ -593,7 +598,7 @@ static unsigned long init_altmap_base(resource_size_t base)
 
 static unsigned long init_altmap_reserve(resource_size_t base)
 {
-	unsigned long reserve = PHYS_PFN(SZ_8K);
+	unsigned long reserve = info_block_reserve() >> PAGE_SHIFT;
 	unsigned long base_pfn = PHYS_PFN(base);
 
 	reserve += base_pfn - PFN_SECTION_ALIGN_DOWN(base_pfn);
@@ -608,6 +613,7 @@ static int __nvdimm_setup_pfn(struct nd_pfn *nd_pfn, struct dev_pagemap *pgmap)
 	u64 offset = le64_to_cpu(pfn_sb->dataoff);
 	u32 start_pad = __le32_to_cpu(pfn_sb->start_pad);
 	u32 end_trunc = __le32_to_cpu(pfn_sb->end_trunc);
+	u32 reserve = info_block_reserve();
 	struct nd_namespace_common *ndns = nd_pfn->ndns;
 	struct nd_namespace_io *nsio = to_nd_namespace_io(&ndns->dev);
 	resource_size_t base = nsio->res.start + start_pad;
@@ -621,7 +627,7 @@ static int __nvdimm_setup_pfn(struct nd_pfn *nd_pfn, struct dev_pagemap *pgmap)
 	res->end -= end_trunc;
 
 	if (nd_pfn->mode == PFN_MODE_RAM) {
-		if (offset < SZ_8K)
+		if (offset < reserve)
 			return -EINVAL;
 		nd_pfn->npfns = le64_to_cpu(pfn_sb->npfns);
 		pgmap->altmap_valid = false;
@@ -634,7 +640,7 @@ static int __nvdimm_setup_pfn(struct nd_pfn *nd_pfn, struct dev_pagemap *pgmap)
 					le64_to_cpu(nd_pfn->pfn_sb->npfns),
 					nd_pfn->npfns);
 		memcpy(altmap, &__altmap, sizeof(*altmap));
-		altmap->free = PHYS_PFN(offset - SZ_8K);
+		altmap->free = PHYS_PFN(offset - reserve);
 		altmap->alloc = 0;
 		pgmap->altmap_valid = true;
 	} else
@@ -678,18 +684,17 @@ static void trim_pfn_device(struct nd_pfn *nd_pfn, u32 *start_pad, u32 *end_trun
 	if (region_intersects(start, size, IORESOURCE_SYSTEM_RAM,
 				IORES_DESC_NONE) == REGION_MIXED
 			|| !IS_ALIGNED(end, nd_pfn->align)
-			|| nd_region_conflict(nd_region, start, size + adjust))
+			|| nd_region_conflict(nd_region, start, size))
 		*end_trunc = end - phys_pmem_align_down(nd_pfn, end);
 }
 
 static int nd_pfn_init(struct nd_pfn *nd_pfn)
 {
-	u32 dax_label_reserve = is_nd_dax(&nd_pfn->dev) ? SZ_128K : 0;
 	struct nd_namespace_common *ndns = nd_pfn->ndns;
 	struct nd_namespace_io *nsio = to_nd_namespace_io(&ndns->dev);
+	u32 start_pad, end_trunc, reserve = info_block_reserve();
 	resource_size_t start, size;
 	struct nd_region *nd_region;
-	u32 start_pad, end_trunc;
 	struct nd_pfn_sb *pfn_sb;
 	unsigned long npfns;
 	phys_addr_t offset;
@@ -734,7 +739,7 @@ static int nd_pfn_init(struct nd_pfn *nd_pfn)
 	 */
 	start = nsio->res.start + start_pad;
 	size = resource_size(&nsio->res);
-	npfns = PFN_SECTION_ALIGN_UP((size - start_pad - end_trunc - SZ_8K)
+	npfns = PFN_SECTION_ALIGN_UP((size - start_pad - end_trunc - reserve)
 			/ PAGE_SIZE);
 	if (nd_pfn->mode == PFN_MODE_PMEM) {
 		/*
@@ -742,11 +747,10 @@ 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 + SZ_8K + 64 * npfns + dax_label_reserve,
+		offset = ALIGN(start + reserve + 64 * npfns,
 				max(nd_pfn->align, PMD_SIZE)) - start;
 	} else if (nd_pfn->mode == PFN_MODE_RAM)
-		offset = ALIGN(start + SZ_8K + dax_label_reserve,
-				nd_pfn->align) - start;
+		offset = ALIGN(start + reserve, nd_pfn->align) - start;
 	else
 		return -ENXIO;
 

drivers/nvdimm/region_devs.c

@@ -1003,6 +1003,13 @@ static struct nd_region *nd_region_create(struct nvdimm_bus *nvdimm_bus,
 
 		if (test_bit(NDD_UNARMED, &nvdimm->flags))
 			ro = 1;
+
+		if (test_bit(NDD_NOBLK, &nvdimm->flags)
+				&& dev_type == &nd_blk_device_type) {
+			dev_err(&nvdimm_bus->dev, "%s: %s mapping%d is not BLK capable\n",
+					caller, dev_name(&nvdimm->dev), i);
+			return NULL;
+		}
 	}
 
 	if (dev_type == &nd_blk_device_type) {