Merge tag 'libnvdimm-for-5.7' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm

Pull libnvdimm and dax updates from Dan Williams:
 "There were multiple touches outside of drivers/nvdimm/ this round to
  add cross arch compatibility to the devm_memremap_pages() interface,
  enhance numa information for persistent memory ranges, and add a
  zero_page_range() dax operation.

  This cycle I switched from the patchwork api to Konstantin's b4 script
  for collecting tags (from x86, PowerPC, filesystem, and device-mapper
  folks), and everything looks to have gone ok there. This has all
  appeared in -next with no reported issues.

  Summary:

   - Add support for region alignment configuration and enforcement to
     fix compatibility across architectures and PowerPC page size
     configurations.

   - Introduce 'zero_page_range' as a dax operation. This facilitates
     filesystem-dax operation without a block-device.

   - Introduce phys_to_target_node() to facilitate drivers that want to
     know resulting numa node if a given reserved address range was
     onlined.

   - Advertise a persistence-domain for of_pmem and papr_scm. The
     persistence domain indicates where cpu-store cycles need to reach
     in the platform-memory subsystem before the platform will consider
     them power-fail protected.

   - Promote numa_map_to_online_node() to a cross-kernel generic
     facility.

   - Save x86 numa information to allow for node-id lookups for reserved
     memory ranges, deploy that capability for the e820-pmem driver.

   - Pick up some miscellaneous minor fixes, that missed v5.6-final,
     including a some smatch reports in the ioctl path and some unit
     test compilation fixups.

   - Fixup some flexible-array declarations"

* tag 'libnvdimm-for-5.7' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm: (29 commits)
  dax: Move mandatory ->zero_page_range() check in alloc_dax()
  dax,iomap: Add helper dax_iomap_zero() to zero a range
  dax: Use new dax zero page method for zeroing a page
  dm,dax: Add dax zero_page_range operation
  s390,dcssblk,dax: Add dax zero_page_range operation to dcssblk driver
  dax, pmem: Add a dax operation zero_page_range
  pmem: Add functions for reading/writing page to/from pmem
  libnvdimm: Update persistence domain value for of_pmem and papr_scm device
  tools/test/nvdimm: Fix out of tree build
  libnvdimm/region: Fix build error
  libnvdimm/region: Replace zero-length array with flexible-array member
  libnvdimm/label: Replace zero-length array with flexible-array member
  ACPI: NFIT: Replace zero-length array with flexible-array member
  libnvdimm/region: Introduce an 'align' attribute
  libnvdimm/region: Introduce NDD_LABELING
  libnvdimm/namespace: Enforce memremap_compat_align()
  libnvdimm/pfn: Prevent raw mode fallback if pfn-infoblock valid
  libnvdimm: Out of bounds read in __nd_ioctl()
  acpi/nfit: improve bounds checking for 'func'
  mm/memremap_pages: Introduce memremap_compat_align()
  ...

5 files changed, 80 insertions, 37 deletions

diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index 5fc45364e86e..a1a1c15a5023 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -122,6 +122,7 @@ config PPC
 	select ARCH_HAS_GCOV_PROFILE_ALL
 	select ARCH_HAS_KCOV
 	select ARCH_HAS_HUGEPD			if HUGETLB_PAGE
+	select ARCH_HAS_MEMREMAP_COMPAT_ALIGN
 	select ARCH_HAS_MMIOWB			if PPC64
 	select ARCH_HAS_PHYS_TO_DMA
 	select ARCH_HAS_PMEM_API
diff --git a/arch/powerpc/mm/ioremap.c b/arch/powerpc/mm/ioremap.c
index fc669643ce6a..b1a0aebe8c48 100644
--- a/arch/powerpc/mm/ioremap.c
+++ b/arch/powerpc/mm/ioremap.c
@@ -2,6 +2,7 @@
 
 #include <linux/io.h>
 #include <linux/slab.h>
+#include <linux/mmzone.h>
 #include <linux/vmalloc.h>
 #include <asm/io-workarounds.h>
 
@@ -97,3 +98,23 @@ void __iomem *do_ioremap(phys_addr_t pa, phys_addr_t offset, unsigned long size,
 
 	return NULL;
 }
+
+#ifdef CONFIG_ZONE_DEVICE
+/*
+ * Override the generic version in mm/memremap.c.
+ *
+ * With hash translation, the direct-map range is mapped with just one
+ * page size selected by htab_init_page_sizes(). Consult
+ * mmu_psize_defs[] to determine the minimum page size alignment.
+*/
+unsigned long memremap_compat_align(void)
+{
+	unsigned int shift = mmu_psize_defs[mmu_linear_psize].shift;
+
+	if (radix_enabled())
+		return SUBSECTION_SIZE;
+	return max(SUBSECTION_SIZE, 1UL << shift);
+
+}
+EXPORT_SYMBOL_GPL(memremap_compat_align);
+#endif
diff --git a/arch/powerpc/platforms/pseries/papr_scm.c b/arch/powerpc/platforms/pseries/papr_scm.c
index e4606100e286..f35592423380 100644
--- a/arch/powerpc/platforms/pseries/papr_scm.c
+++ b/arch/powerpc/platforms/pseries/papr_scm.c
@@ -286,25 +286,6 @@ static int papr_scm_ndctl(struct nvdimm_bus_descriptor *nd_desc,
 	return 0;
 }
 
-static inline int papr_scm_node(int node)
-{
-	int min_dist = INT_MAX, dist;
-	int nid, min_node;
-
-	if ((node == NUMA_NO_NODE) || node_online(node))
-		return node;
-
-	min_node = first_online_node;
-	for_each_online_node(nid) {
-		dist = node_distance(node, nid);
-		if (dist < min_dist) {
-			min_dist = dist;
-			min_node = nid;
-		}
-	}
-	return min_node;
-}
-
 static int papr_scm_nvdimm_init(struct papr_scm_priv *p)
 {
 	struct device *dev = &p->pdev->dev;
@@ -329,7 +310,7 @@ static int papr_scm_nvdimm_init(struct papr_scm_priv *p)
 	}
 
 	dimm_flags = 0;
-	set_bit(NDD_ALIASING, &dimm_flags);
+	set_bit(NDD_LABELING, &dimm_flags);
 
 	p->nvdimm = nvdimm_create(p->bus, p, NULL, dimm_flags,
 				  PAPR_SCM_DIMM_CMD_MASK, 0, NULL);
@@ -350,7 +331,7 @@ static int papr_scm_nvdimm_init(struct papr_scm_priv *p)
 
 	memset(&ndr_desc, 0, sizeof(ndr_desc));
 	target_nid = dev_to_node(&p->pdev->dev);
-	online_nid = papr_scm_node(target_nid);
+	online_nid = numa_map_to_online_node(target_nid);
 	ndr_desc.numa_node = online_nid;
 	ndr_desc.target_node = target_nid;
 	ndr_desc.res = &p->res;
@@ -362,8 +343,10 @@ static int papr_scm_nvdimm_init(struct papr_scm_priv *p)
 
 	if (p->is_volatile)
 		p->region = nvdimm_volatile_region_create(p->bus, &ndr_desc);
-	else
+	else {
+		set_bit(ND_REGION_PERSIST_MEMCTRL, &ndr_desc.flags);
 		p->region = nvdimm_pmem_region_create(p->bus, &ndr_desc);
+	}
 	if (!p->region) {
 		dev_err(dev, "Error registering region %pR from %pOF\n",
 				ndr_desc.res, p->dn);
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 8d078642b4be..0163d764ade4 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -1661,6 +1661,7 @@ config X86_PMEM_LEGACY
 	depends on PHYS_ADDR_T_64BIT
 	depends on BLK_DEV
 	select X86_PMEM_LEGACY_DEVICE
+	select NUMA_KEEP_MEMINFO if NUMA
 	select LIBNVDIMM
 	help
 	  Treat memory marked using the non-standard e820 type of 12 as used
diff --git a/arch/x86/mm/numa.c b/arch/x86/mm/numa.c
index 99f7a68738f0..59ba008504dc 100644
--- a/arch/x86/mm/numa.c
+++ b/arch/x86/mm/numa.c
@@ -25,11 +25,8 @@ nodemask_t numa_nodes_parsed __initdata;
 struct pglist_data *node_data[MAX_NUMNODES] __read_mostly;
 EXPORT_SYMBOL(node_data);
 
-static struct numa_meminfo numa_meminfo
-#ifndef CONFIG_MEMORY_HOTPLUG
-__initdata
-#endif
-;
+static struct numa_meminfo numa_meminfo __initdata_or_meminfo;
+static struct numa_meminfo numa_reserved_meminfo __initdata_or_meminfo;
 
 static int numa_distance_cnt;
 static u8 *numa_distance;
@@ -169,6 +166,19 @@ void __init numa_remove_memblk_from(int idx, struct numa_meminfo *mi)
 }
 
 /**
+ * numa_move_tail_memblk - Move a numa_memblk from one numa_meminfo to another
+ * @dst: numa_meminfo to append block to
+ * @idx: Index of memblk to remove
+ * @src: numa_meminfo to remove memblk from
+ */
+static void __init numa_move_tail_memblk(struct numa_meminfo *dst, int idx,
+					 struct numa_meminfo *src)
+{
+	dst->blk[dst->nr_blks++] = src->blk[idx];
+	numa_remove_memblk_from(idx, src);
+}
+
+/**
  * numa_add_memblk - Add one numa_memblk to numa_meminfo
  * @nid: NUMA node ID of the new memblk
  * @start: Start address of the new memblk
@@ -237,14 +247,19 @@ int __init numa_cleanup_meminfo(struct numa_meminfo *mi)
 	for (i = 0; i < mi->nr_blks; i++) {
 		struct numa_memblk *bi = &mi->blk[i];
 
-		/* make sure all blocks are inside the limits */
+		/* move / save reserved memory ranges */
+		if (!memblock_overlaps_region(&memblock.memory,
+					bi->start, bi->end - bi->start)) {
+			numa_move_tail_memblk(&numa_reserved_meminfo, i--, mi);
+			continue;
+		}
+
+		/* make sure all non-reserved blocks are inside the limits */
 		bi->start = max(bi->start, low);
 		bi->end = min(bi->end, high);
 
-		/* and there's no empty or non-exist block */
-		if (bi->start >= bi->end ||
-		    !memblock_overlaps_region(&memblock.memory,
-			bi->start, bi->end - bi->start))
+		/* and there's no empty block */
+		if (bi->start >= bi->end)
 			numa_remove_memblk_from(i--, mi);
 	}
 
@@ -881,16 +896,38 @@ EXPORT_SYMBOL(cpumask_of_node);
 
 #endif	/* !CONFIG_DEBUG_PER_CPU_MAPS */
 
-#ifdef CONFIG_MEMORY_HOTPLUG
-int memory_add_physaddr_to_nid(u64 start)
+#ifdef CONFIG_NUMA_KEEP_MEMINFO
+static int meminfo_to_nid(struct numa_meminfo *mi, u64 start)
 {
-	struct numa_meminfo *mi = &numa_meminfo;
-	int nid = mi->blk[0].nid;
 	int i;
 
 	for (i = 0; i < mi->nr_blks; i++)
 		if (mi->blk[i].start <= start && mi->blk[i].end > start)
-			nid = mi->blk[i].nid;
+			return mi->blk[i].nid;
+	return NUMA_NO_NODE;
+}
+
+int phys_to_target_node(phys_addr_t start)
+{
+	int nid = meminfo_to_nid(&numa_meminfo, start);
+
+	/*
+	 * Prefer online nodes, but if reserved memory might be
+	 * hot-added continue the search with reserved ranges.
+	 */
+	if (nid != NUMA_NO_NODE)
+		return nid;
+
+	return meminfo_to_nid(&numa_reserved_meminfo, start);
+}
+EXPORT_SYMBOL_GPL(phys_to_target_node);
+
+int memory_add_physaddr_to_nid(u64 start)
+{
+	int nid = meminfo_to_nid(&numa_meminfo, start);
+
+	if (nid == NUMA_NO_NODE)
+		nid = numa_meminfo.blk[0].nid;
 	return nid;
 }
 EXPORT_SYMBOL_GPL(memory_add_physaddr_to_nid);