Merge tag 'libnvdimm-for-7.2' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm

Pull nvdimm/dax updates from Alison Schofield:

 - Fix a race condition and a couple of static analysis issues in BTT

 - Use sysfs_emit() in preparation for removal of cpumap_print_to_pagebuf()

 - Escalate a dev_dbg to dev_err in a resource conflict message

 - MAINTAINER file updates

* tag 'libnvdimm-for-7.2' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm:
  MAINTAINERS: nvdimm: Include maintainer profile
  MAINTAINERS: Update address for Ira Weiny
  MAINTAINERS: Add maintainer info for libnvdimm and DAX
  nvdimm: Use sysfs_emit() for cpumask show callback
  dax/bus: Upgrade resource conflict message to dev_err() in alloc_dax_region()
  nvdimm/btt: Free arenas on btt_init() error paths
  nvdimm/btt: Free arena sub-allocations on discover_arenas() error path
  nvdimm/btt: Handle preemption in BTT lane acquisition

8 files changed, 50 insertions, 62 deletions

diff --git a/.mailmap b/.mailmap
index 33b03396b18f..7d653f7e693d 100644
--- a/.mailmap
+++ b/.mailmap
@@ -450,6 +450,7 @@ Juha Yrjola <juha.yrjola@nokia.com>
 Juha Yrjola <juha.yrjola@solidboot.com>
 Julien Thierry <julien.thierry.kdev@gmail.com> <julien.thierry@arm.com>
 Justin Iurman <justin.iurman@gmail.com> <justin.iurman@uliege.be>
+Ira Weiny <iweiny@kernel.org> <ira.weiny@intel.com>
 Iskren Chernev <me@iskren.info> <iskren.chernev@gmail.com>
 Kalle Valo <kvalo@kernel.org> <kvalo@codeaurora.org>
 Kalle Valo <kvalo@kernel.org> <quic_kvalo@quicinc.com>
diff --git a/Documentation/driver-api/nvdimm/btt.rst b/Documentation/driver-api/nvdimm/btt.rst
index 2d8269f834bd..d29fab95f149 100644
--- a/Documentation/driver-api/nvdimm/btt.rst
+++ b/Documentation/driver-api/nvdimm/btt.rst
@@ -161,9 +161,8 @@ process::
 	nlanes = min(nfree, num_cpus)
 
 A lane number is obtained at the start of any IO, and is used for indexing into
-all the on-disk and in-memory data structures for the duration of the IO. If
-there are more CPUs than the max number of available lanes, than lanes are
-protected by spinlocks.
+all the on-disk and in-memory data structures for the duration of the IO. Lanes
+are protected by mutexes.
 
 
 d. In-memory data structure: Read Tracking Table (RTT)
diff --git a/MAINTAINERS b/MAINTAINERS
index 3147a467d458..926a17df059e 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -4302,7 +4302,7 @@ M:	Greg Kroah-Hartman <gregkh@linuxfoundation.org>
 M:	"Rafael J. Wysocki" <rafael@kernel.org>
 M:	Danilo Krummrich <dakr@kernel.org>
 R:	Dave Ertman <david.m.ertman@intel.com>
-R:	Ira Weiny <ira.weiny@intel.com>
+R:	Ira Weiny <iweiny@kernel.org>
 R:	Leon Romanovsky <leon@kernel.org>
 L:	driver-core@lists.linux.dev
 S:	Supported
@@ -6454,8 +6454,8 @@ M:	Jonathan Cameron <jic23@kernel.org>
 M:	Dave Jiang <dave.jiang@intel.com>
 M:	Alison Schofield <alison.schofield@intel.com>
 M:	Vishal Verma <vishal.l.verma@intel.com>
-M:	Ira Weiny <ira.weiny@intel.com>
 M:	Dan Williams <djbw@kernel.org>
+R:	Ira Weiny <iweiny@kernel.org>
 L:	linux-cxl@vger.kernel.org
 S:	Maintained
 F:	Documentation/driver-api/cxl
@@ -7320,6 +7320,7 @@ DEVICE DIRECT ACCESS (DAX)
 M:	Dan Williams <djbw@kernel.org>
 M:	Vishal Verma <vishal.l.verma@intel.com>
 M:	Dave Jiang <dave.jiang@intel.com>
+M:	Alison Schofield <alison.schofield@intel.com>
 L:	nvdimm@lists.linux.dev
 L:	linux-cxl@vger.kernel.org
 S:	Supported
@@ -14725,6 +14726,7 @@ LIBNVDIMM BTT: BLOCK TRANSLATION TABLE
 M:	Vishal Verma <vishal.l.verma@intel.com>
 M:	Dan Williams <djbw@kernel.org>
 M:	Dave Jiang <dave.jiang@intel.com>
+M:	Alison Schofield <alison.schofield@intel.com>
 L:	nvdimm@lists.linux.dev
 S:	Supported
 Q:	https://patchwork.kernel.org/project/linux-nvdimm/list/
@@ -14735,6 +14737,7 @@ LIBNVDIMM PMEM: PERSISTENT MEMORY DRIVER
 M:	Dan Williams <djbw@kernel.org>
 M:	Vishal Verma <vishal.l.verma@intel.com>
 M:	Dave Jiang <dave.jiang@intel.com>
+M:	Alison Schofield <alison.schofield@intel.com>
 L:	nvdimm@lists.linux.dev
 S:	Supported
 Q:	https://patchwork.kernel.org/project/linux-nvdimm/list/
@@ -14753,12 +14756,14 @@ LIBNVDIMM: NON-VOLATILE MEMORY DEVICE SUBSYSTEM
 M:	Dan Williams <djbw@kernel.org>
 M:	Vishal Verma <vishal.l.verma@intel.com>
 M:	Dave Jiang <dave.jiang@intel.com>
-M:	Ira Weiny <ira.weiny@intel.com>
+M:	Alison Schofield <alison.schofield@intel.com>
+R:	Ira Weiny <iweiny@kernel.org>
 L:	nvdimm@lists.linux.dev
 S:	Supported
 Q:	https://patchwork.kernel.org/project/linux-nvdimm/list/
 P:	Documentation/nvdimm/maintainer-entry-profile.rst
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm.git
+F:	Documentation/nvdimm/maintainer-entry-profile.rst
 F:	drivers/acpi/nfit/*
 F:	drivers/nvdimm/*
 F:	include/linux/libnvdimm.h
diff --git a/drivers/dax/bus.c b/drivers/dax/bus.c
index ccfe65004888..b809e1a264af 100644
--- a/drivers/dax/bus.c
+++ b/drivers/dax/bus.c
@@ -670,7 +670,7 @@ struct dax_region *alloc_dax_region(struct device *parent, int region_id,
 
 	rc = request_resource(&dax_regions, &dax_region->res);
 	if (rc) {
-		dev_dbg(parent, "dax_region resource conflict for %pR\n",
+		dev_err(parent, "dax_region resource conflict for %pR\n",
 			&dax_region->res);
 		goto err_res;
 	}
diff --git a/drivers/nvdimm/btt.c b/drivers/nvdimm/btt.c
index fdcb080a4314..7e1112960d7f 100644
--- a/drivers/nvdimm/btt.c
+++ b/drivers/nvdimm/btt.c
@@ -919,6 +919,9 @@ static int discover_arenas(struct btt *btt)
 	return ret;
 
  out:
+	kfree(arena->freelist);
+	kfree(arena->rtt);
+	kfree(arena->map_locks);
 	kfree(arena);
 	free_arenas(btt);
 	return ret;
@@ -1589,7 +1592,7 @@ static struct btt *btt_init(struct nd_btt *nd_btt, unsigned long long rawsize,
 	if (btt->init_state != INIT_READY && nd_region->ro) {
 		dev_warn(dev, "%s is read-only, unable to init btt metadata\n",
 				dev_name(&nd_region->dev));
-		return NULL;
+		goto err;
 	} else if (btt->init_state != INIT_READY) {
 		btt->num_arenas = (rawsize / ARENA_MAX_SIZE) +
 			((rawsize % ARENA_MAX_SIZE) ? 1 : 0);
@@ -1599,25 +1602,28 @@ static struct btt *btt_init(struct nd_btt *nd_btt, unsigned long long rawsize,
 		ret = create_arenas(btt);
 		if (ret) {
 			dev_info(dev, "init: create_arenas: %d\n", ret);
-			return NULL;
+			goto err;
 		}
 
 		ret = btt_meta_init(btt);
 		if (ret) {
 			dev_err(dev, "init: error in meta_init: %d\n", ret);
-			return NULL;
+			goto err;
 		}
 	}
 
 	ret = btt_blk_init(btt);
 	if (ret) {
 		dev_err(dev, "init: error in blk_init: %d\n", ret);
-		return NULL;
+		goto err;
 	}
 
 	btt_debugfs_init(btt);
 
 	return btt;
+err:
+	free_arenas(btt);
+	return NULL;
 }
 
 /**
diff --git a/drivers/nvdimm/nd.h b/drivers/nvdimm/nd.h
index 18b64559664b..ed69dea5b75f 100644
--- a/drivers/nvdimm/nd.h
+++ b/drivers/nvdimm/nd.h
@@ -365,11 +365,6 @@ unsigned sizeof_namespace_label(struct nvdimm_drvdata *ndd);
 	for (res = (ndd)->dpa.child, next = res ? res->sibling : NULL; \
 			res; res = next, next = next ? next->sibling : NULL)
 
-struct nd_percpu_lane {
-	int count;
-	spinlock_t lock;
-};
-
 enum nd_label_flags {
 	ND_LABEL_REAP,
 };
@@ -400,6 +395,10 @@ struct nd_mapping {
 	struct nvdimm_drvdata *ndd;
 };
 
+struct nd_lane {
+	struct mutex lock; /* serialize lane access */
+} ____cacheline_aligned_in_smp;
+
 struct nd_region {
 	struct device dev;
 	struct ida ns_ida;
@@ -420,7 +419,7 @@ struct nd_region {
 	struct kernfs_node *bb_state;
 	struct badblocks bb;
 	struct nd_interleave_set *nd_set;
-	struct nd_percpu_lane __percpu *lane;
+	struct nd_lane *lane;
 	int (*flush)(struct nd_region *nd_region, struct bio *bio);
 	struct nd_mapping mapping[] __counted_by(ndr_mappings);
 };
diff --git a/drivers/nvdimm/nd_perf.c b/drivers/nvdimm/nd_perf.c
index e0b51438dc9b..9e497cae65b3 100644
--- a/drivers/nvdimm/nd_perf.c
+++ b/drivers/nvdimm/nd_perf.c
@@ -123,7 +123,7 @@ static ssize_t nvdimm_pmu_cpumask_show(struct device *dev,
 
 	nd_pmu = container_of(pmu, struct nvdimm_pmu, pmu);
 
-	return cpumap_print_to_pagebuf(true, buf, cpumask_of(nd_pmu->cpu));
+	return sysfs_emit(buf, "%*pbl\n", cpumask_pr_args(cpumask_of(nd_pmu->cpu)));
 }
 
 static int nvdimm_pmu_cpu_offline(unsigned int cpu, struct hlist_node *node)
diff --git a/drivers/nvdimm/region_devs.c b/drivers/nvdimm/region_devs.c
index e35c2e18518f..5e079d61cbaa 100644
--- a/drivers/nvdimm/region_devs.c
+++ b/drivers/nvdimm/region_devs.c
@@ -192,7 +192,9 @@ static void nd_region_release(struct device *dev)
 
 		put_device(&nvdimm->dev);
 	}
-	free_percpu(nd_region->lane);
+	for (i = 0; i < nd_region->num_lanes; i++)
+		mutex_destroy(&nd_region->lane[i].lock);
+	kfree(nd_region->lane);
 	if (!test_bit(ND_REGION_CXL, &nd_region->flags))
 		memregion_free(nd_region->id);
 	kfree(nd_region);
@@ -904,52 +906,30 @@ void nd_region_advance_seeds(struct nd_region *nd_region, struct device *dev)
  * nd_region_acquire_lane - allocate and lock a lane
  * @nd_region: region id and number of lanes possible
  *
- * A lane correlates to a BLK-data-window and/or a log slot in the BTT.
- * We optimize for the common case where there are 256 lanes, one
- * per-cpu.  For larger systems we need to lock to share lanes.  For now
- * this implementation assumes the cost of maintaining an allocator for
- * free lanes is on the order of the lock hold time, so it implements a
- * static lane = cpu % num_lanes mapping.
+ * A lane correlates to a log slot in the BTT. Lanes are shared across
+ * CPUs using a static lane = cpu % num_lanes mapping, with a per-lane
+ * mutex to serialize access.
  *
- * In the case of a BTT instance on top of a BLK namespace a lane may be
- * acquired recursively.  We lock on the first instance.
- *
- * In the case of a BTT instance on top of PMEM, we only acquire a lane
- * for the BTT metadata updates.
+ * Callers must be in sleepable context. The only in-tree caller is
+ * BTT's ->submit_bio handler (btt_read_pg / btt_write_pg).
  */
 unsigned int nd_region_acquire_lane(struct nd_region *nd_region)
+	__acquires(&nd_region->lane[lane].lock)
 {
-	unsigned int cpu, lane;
-
-	migrate_disable();
-	cpu = smp_processor_id();
-	if (nd_region->num_lanes < nr_cpu_ids) {
-		struct nd_percpu_lane *ndl_lock, *ndl_count;
+	unsigned int lane;
 
-		lane = cpu % nd_region->num_lanes;
-		ndl_count = per_cpu_ptr(nd_region->lane, cpu);
-		ndl_lock = per_cpu_ptr(nd_region->lane, lane);
-		if (ndl_count->count++ == 0)
-			spin_lock(&ndl_lock->lock);
-	} else
-		lane = cpu;
+	might_sleep();
 
+	lane = raw_smp_processor_id() % nd_region->num_lanes;
+	mutex_lock(&nd_region->lane[lane].lock);
 	return lane;
 }
 EXPORT_SYMBOL(nd_region_acquire_lane);
 
 void nd_region_release_lane(struct nd_region *nd_region, unsigned int lane)
+	__releases(&nd_region->lane[lane].lock)
 {
-	if (nd_region->num_lanes < nr_cpu_ids) {
-		unsigned int cpu = smp_processor_id();
-		struct nd_percpu_lane *ndl_lock, *ndl_count;
-
-		ndl_count = per_cpu_ptr(nd_region->lane, cpu);
-		ndl_lock = per_cpu_ptr(nd_region->lane, lane);
-		if (--ndl_count->count == 0)
-			spin_unlock(&ndl_lock->lock);
-	}
-	migrate_enable();
+	mutex_unlock(&nd_region->lane[lane].lock);
 }
 EXPORT_SYMBOL(nd_region_release_lane);
 
@@ -1019,17 +999,16 @@ static struct nd_region *nd_region_create(struct nvdimm_bus *nvdimm_bus,
 			goto err_id;
 	}
 
-	nd_region->lane = alloc_percpu(struct nd_percpu_lane);
+	nd_region->num_lanes = ndr_desc->num_lanes;
+	if (!nd_region->num_lanes)
+		goto err_percpu;
+	nd_region->lane = kcalloc(nd_region->num_lanes,
+				  sizeof(*nd_region->lane), GFP_KERNEL);
 	if (!nd_region->lane)
 		goto err_percpu;
 
-        for (i = 0; i < nr_cpu_ids; i++) {
-		struct nd_percpu_lane *ndl;
-
-		ndl = per_cpu_ptr(nd_region->lane, i);
-		spin_lock_init(&ndl->lock);
-		ndl->count = 0;
-	}
+	for (i = 0; i < nd_region->num_lanes; i++)
+		mutex_init(&nd_region->lane[i].lock);
 
 	for (i = 0; i < ndr_desc->num_mappings; i++) {
 		struct nd_mapping_desc *mapping = &ndr_desc->mapping[i];
@@ -1046,7 +1025,6 @@ static struct nd_region *nd_region_create(struct nvdimm_bus *nvdimm_bus,
 	}
 	nd_region->provider_data = ndr_desc->provider_data;
 	nd_region->nd_set = ndr_desc->nd_set;
-	nd_region->num_lanes = ndr_desc->num_lanes;
 	nd_region->flags = ndr_desc->flags;
 	nd_region->ro = ro;
 	nd_region->numa_node = ndr_desc->numa_node;