Merge branch 'libnvdimm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm

Pull libnvdimm fixes from Dan Williams:
 "Several fixes to the DSM (ACPI device specific method) marshaling
  implementation.

  I consider these urgent enough to send for 4.9 consideration since
  they fix the kernel's handling of ARS (Address Range Scrub) commands.
  Especially for platforms without machine-check-recovery capabilities,
  successful execution of ARS commands enables the platform to
  potentially break out of an infinite reboot problem if a media error
  is present in the boot path. There is also a one line fix for a
  device-dax read-only mapping regression.

  Commits 9a901f5 ("acpi, nfit: fix extended status translations
  for ACPI DSMs") and 325896f ("device-dax: fix private mapping
  restriction, permit read-only") are true regression fixes for changes
  introduced this cycle.

  Commit efda1b5 ("acpi, nfit, libnvdimm: fix / harden ars_status
  output length handling") fixes the kernel's handling of zero-length
  results, this never would have worked in the past, but we only just
  recently discovered a BIOS implementation that emits this arguably
  spec non-compliant result.

  The remaining two commits are additional fall out from thinking
  through the implications of a zero / truncated length result of the
  ARS Status command.

  In order to mitigate the risk that these changes introduce yet more
  regressions they are backstopped by a new unit test in commit
  a7de92d ("tools/testing/nvdimm: unit test acpi_nfit_ctl()") that
  mocks up inputs to acpi_nfit_ctl()"

* 'libnvdimm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm:
  device-dax: fix private mapping restriction, permit read-only
  tools/testing/nvdimm: unit test acpi_nfit_ctl()
  acpi, nfit: fix bus vs dimm confusion in xlat_status
  acpi, nfit: validate ars_status output buffer size
  acpi, nfit, libnvdimm: fix / harden ars_status output length handling
  acpi, nfit: fix extended status translations for ACPI DSMs

Author: torvalds

drivers/acpi/nfit/core.c

@@ -94,7 +94,7 @@ static struct acpi_device *to_acpi_dev(struct acpi_nfit_desc *acpi_desc)
 	return to_acpi_device(acpi_desc->dev);
 }
 
-static int xlat_status(void *buf, unsigned int cmd, u32 status)
+static int xlat_bus_status(void *buf, unsigned int cmd, u32 status)
 {
 	struct nd_cmd_clear_error *clear_err;
 	struct nd_cmd_ars_status *ars_status;
@@ -113,7 +113,7 @@ static int xlat_status(void *buf, unsigned int cmd, u32 status)
 		flags = ND_ARS_PERSISTENT | ND_ARS_VOLATILE;
 		if ((status >> 16 & flags) == 0)
 			return -ENOTTY;
-		break;
+		return 0;
 	case ND_CMD_ARS_START:
 		/* ARS is in progress */
 		if ((status & 0xffff) == NFIT_ARS_START_BUSY)
@@ -122,7 +122,7 @@ static int xlat_status(void *buf, unsigned int cmd, u32 status)
 		/* Command failed */
 		if (status & 0xffff)
 			return -EIO;
-		break;
+		return 0;
 	case ND_CMD_ARS_STATUS:
 		ars_status = buf;
 		/* Command failed */
@@ -146,15 +146,16 @@ static int xlat_status(void *buf, unsigned int cmd, u32 status)
 		 * then just continue with the returned results.
 		 */
 		if (status == NFIT_ARS_STATUS_INTR) {
-			if (ars_status->flags & NFIT_ARS_F_OVERFLOW)
+			if (ars_status->out_length >= 40 && (ars_status->flags
+						& NFIT_ARS_F_OVERFLOW))
 				return -ENOSPC;
 			return 0;
 		}
 
 		/* Unknown status */
 		if (status >> 16)
 			return -EIO;
-		break;
+		return 0;
 	case ND_CMD_CLEAR_ERROR:
 		clear_err = buf;
 		if (status & 0xffff)
@@ -163,7 +164,7 @@ static int xlat_status(void *buf, unsigned int cmd, u32 status)
 			return -EIO;
 		if (clear_err->length > clear_err->cleared)
 			return clear_err->cleared;
-		break;
+		return 0;
 	default:
 		break;
 	}
@@ -174,9 +175,18 @@ static int xlat_status(void *buf, unsigned int cmd, u32 status)
 	return 0;
 }
 
-static int acpi_nfit_ctl(struct nvdimm_bus_descriptor *nd_desc,
-		struct nvdimm *nvdimm, unsigned int cmd, void *buf,
-		unsigned int buf_len, int *cmd_rc)
+static int xlat_status(struct nvdimm *nvdimm, void *buf, unsigned int cmd,
+		u32 status)
+{
+	if (!nvdimm)
+		return xlat_bus_status(buf, cmd, status);
+	if (status)
+		return -EIO;
+	return 0;
+}
+
+int acpi_nfit_ctl(struct nvdimm_bus_descriptor *nd_desc, struct nvdimm *nvdimm,
+		unsigned int cmd, void *buf, unsigned int buf_len, int *cmd_rc)
 {
 	struct acpi_nfit_desc *acpi_desc = to_acpi_nfit_desc(nd_desc);
 	union acpi_object in_obj, in_buf, *out_obj;
@@ -298,7 +308,8 @@ static int acpi_nfit_ctl(struct nvdimm_bus_descriptor *nd_desc,
 
 	for (i = 0, offset = 0; i < desc->out_num; i++) {
 		u32 out_size = nd_cmd_out_size(nvdimm, cmd, desc, i, buf,
-				(u32 *) out_obj->buffer.pointer);
+				(u32 *) out_obj->buffer.pointer,
+				out_obj->buffer.length - offset);
 
 		if (offset + out_size > out_obj->buffer.length) {
 			dev_dbg(dev, "%s:%s output object underflow cmd: %s field: %d\n",
@@ -333,7 +344,8 @@ static int acpi_nfit_ctl(struct nvdimm_bus_descriptor *nd_desc,
 			 */
 			rc = buf_len - offset - in_buf.buffer.length;
 			if (cmd_rc)
-				*cmd_rc = xlat_status(buf, cmd, fw_status);
+				*cmd_rc = xlat_status(nvdimm, buf, cmd,
+						fw_status);
 		} else {
 			dev_err(dev, "%s:%s underrun cmd: %s buf_len: %d out_len: %d\n",
 					__func__, dimm_name, cmd_name, buf_len,
@@ -343,14 +355,15 @@ static int acpi_nfit_ctl(struct nvdimm_bus_descriptor *nd_desc,
 	} else {
 		rc = 0;
 		if (cmd_rc)
-			*cmd_rc = xlat_status(buf, cmd, fw_status);
+			*cmd_rc = xlat_status(nvdimm, buf, cmd, fw_status);
 	}
 
  out:
 	ACPI_FREE(out_obj);
 
 	return rc;
 }
+EXPORT_SYMBOL_GPL(acpi_nfit_ctl);
 
 static const char *spa_type_name(u16 type)
 {
@@ -2001,19 +2014,32 @@ static int ars_get_status(struct acpi_nfit_desc *acpi_desc)
 	return cmd_rc;
 }
 
-static int ars_status_process_records(struct nvdimm_bus *nvdimm_bus,
+static int ars_status_process_records(struct acpi_nfit_desc *acpi_desc,
 		struct nd_cmd_ars_status *ars_status)
 {
+	struct nvdimm_bus *nvdimm_bus = acpi_desc->nvdimm_bus;
 	int rc;
 	u32 i;
 
+	/*
+	 * First record starts at 44 byte offset from the start of the
+	 * payload.
+	 */
+	if (ars_status->out_length < 44)
+		return 0;
 	for (i = 0; i < ars_status->num_records; i++) {
+		/* only process full records */
+		if (ars_status->out_length
+				< 44 + sizeof(struct nd_ars_record) * (i + 1))
+			break;
 		rc = nvdimm_bus_add_poison(nvdimm_bus,
 				ars_status->records[i].err_address,
 				ars_status->records[i].length);
 		if (rc)
 			return rc;
 	}
+	if (i < ars_status->num_records)
+		dev_warn(acpi_desc->dev, "detected truncated ars results\n");
 
 	return 0;
 }
@@ -2266,8 +2292,7 @@ static int acpi_nfit_query_poison(struct acpi_nfit_desc *acpi_desc,
 	if (rc < 0 && rc != -ENOSPC)
 		return rc;
 
-	if (ars_status_process_records(acpi_desc->nvdimm_bus,
-				acpi_desc->ars_status))
+	if (ars_status_process_records(acpi_desc, acpi_desc->ars_status))
 		return -ENOMEM;
 
 	return 0;

drivers/acpi/nfit/nfit.h

@@ -240,5 +240,7 @@ const u8 *to_nfit_uuid(enum nfit_uuids id);
 int acpi_nfit_init(struct acpi_nfit_desc *acpi_desc, void *nfit, acpi_size sz);
 void __acpi_nfit_notify(struct device *dev, acpi_handle handle, u32 event);
 void __acpi_nvdimm_notify(struct device *dev, u32 event);
+int acpi_nfit_ctl(struct nvdimm_bus_descriptor *nd_desc, struct nvdimm *nvdimm,
+		unsigned int cmd, void *buf, unsigned int buf_len, int *cmd_rc);
 void acpi_nfit_desc_init(struct acpi_nfit_desc *acpi_desc, struct device *dev);
 #endif /* __NFIT_H__ */

drivers/dax/dax.c

@@ -271,7 +271,7 @@ static int check_vma(struct dax_dev *dax_dev, struct vm_area_struct *vma,
 		return -ENXIO;
 
 	/* prevent private mappings from being established */
-	if ((vma->vm_flags & VM_SHARED) != VM_SHARED) {
+	if ((vma->vm_flags & VM_MAYSHARE) != VM_MAYSHARE) {
 		dev_info(dev, "%s: %s: fail, attempted private mapping\n",
 				current->comm, func);
 		return -EINVAL;

drivers/nvdimm/bus.c

@@ -715,7 +715,7 @@ EXPORT_SYMBOL_GPL(nd_cmd_in_size);
 
 u32 nd_cmd_out_size(struct nvdimm *nvdimm, int cmd,
 		const struct nd_cmd_desc *desc, int idx, const u32 *in_field,
-		const u32 *out_field)
+		const u32 *out_field, unsigned long remainder)
 {
 	if (idx >= desc->out_num)
 		return UINT_MAX;
@@ -727,9 +727,24 @@ u32 nd_cmd_out_size(struct nvdimm *nvdimm, int cmd,
 		return in_field[1];
 	else if (nvdimm && cmd == ND_CMD_VENDOR && idx == 2)
 		return out_field[1];
-	else if (!nvdimm && cmd == ND_CMD_ARS_STATUS && idx == 2)
-		return out_field[1] - 8;
-	else if (cmd == ND_CMD_CALL) {
+	else if (!nvdimm && cmd == ND_CMD_ARS_STATUS && idx == 2) {
+		/*
+		 * Per table 9-276 ARS Data in ACPI 6.1, out_field[1] is
+		 * "Size of Output Buffer in bytes, including this
+		 * field."
+		 */
+		if (out_field[1] < 4)
+			return 0;
+		/*
+		 * ACPI 6.1 is ambiguous if 'status' is included in the
+		 * output size. If we encounter an output size that
+		 * overshoots the remainder by 4 bytes, assume it was
+		 * including 'status'.
+		 */
+		if (out_field[1] - 8 == remainder)
+			return remainder;
+		return out_field[1] - 4;
+	} else if (cmd == ND_CMD_CALL) {
 		struct nd_cmd_pkg *pkg = (struct nd_cmd_pkg *) in_field;
 
 		return pkg->nd_size_out;
@@ -876,7 +891,7 @@ static int __nd_ioctl(struct nvdimm_bus *nvdimm_bus, struct nvdimm *nvdimm,
 	/* process an output envelope */
 	for (i = 0; i < desc->out_num; i++) {
 		u32 out_size = nd_cmd_out_size(nvdimm, cmd, desc, i,
-				(u32 *) in_env, (u32 *) out_env);
+				(u32 *) in_env, (u32 *) out_env, 0);
 		u32 copy;
 
 		if (out_size == UINT_MAX) {

include/linux/libnvdimm.h

@@ -143,7 +143,7 @@ u32 nd_cmd_in_size(struct nvdimm *nvdimm, int cmd,
 		const struct nd_cmd_desc *desc, int idx, void *buf);
 u32 nd_cmd_out_size(struct nvdimm *nvdimm, int cmd,
 		const struct nd_cmd_desc *desc, int idx, const u32 *in_field,
-		const u32 *out_field);
+		const u32 *out_field, unsigned long remainder);
 int nvdimm_bus_check_dimm_count(struct nvdimm_bus *nvdimm_bus, int dimm_count);
 struct nd_region *nvdimm_pmem_region_create(struct nvdimm_bus *nvdimm_bus,
 		struct nd_region_desc *ndr_desc);

tools/testing/nvdimm/Kbuild

@@ -14,6 +14,7 @@ ldflags-y += --wrap=devm_memremap_pages
 ldflags-y += --wrap=insert_resource
 ldflags-y += --wrap=remove_resource
 ldflags-y += --wrap=acpi_evaluate_object
+ldflags-y += --wrap=acpi_evaluate_dsm
 
 DRIVERS := ../../../drivers
 NVDIMM_SRC := $(DRIVERS)/nvdimm

tools/testing/nvdimm/test/iomap.c

@@ -26,14 +26,17 @@ static LIST_HEAD(iomap_head);
 
 static struct iomap_ops {
 	nfit_test_lookup_fn nfit_test_lookup;
+	nfit_test_evaluate_dsm_fn evaluate_dsm;
 	struct list_head list;
 } iomap_ops = {
 	.list = LIST_HEAD_INIT(iomap_ops.list),
 };
 
-void nfit_test_setup(nfit_test_lookup_fn lookup)
+void nfit_test_setup(nfit_test_lookup_fn lookup,
+		nfit_test_evaluate_dsm_fn evaluate)
 {
 	iomap_ops.nfit_test_lookup = lookup;
+	iomap_ops.evaluate_dsm = evaluate;
 	list_add_rcu(&iomap_ops.list, &iomap_head);
 }
 EXPORT_SYMBOL(nfit_test_setup);
@@ -367,4 +370,22 @@ acpi_status __wrap_acpi_evaluate_object(acpi_handle handle, acpi_string path,
 }
 EXPORT_SYMBOL(__wrap_acpi_evaluate_object);
 
+union acpi_object * __wrap_acpi_evaluate_dsm(acpi_handle handle, const u8 *uuid,
+		u64 rev, u64 func, union acpi_object *argv4)
+{
+	union acpi_object *obj = ERR_PTR(-ENXIO);
+	struct iomap_ops *ops;
+
+	rcu_read_lock();
+	ops = list_first_or_null_rcu(&iomap_head, typeof(*ops), list);
+	if (ops)
+		obj = ops->evaluate_dsm(handle, uuid, rev, func, argv4);
+	rcu_read_unlock();
+
+	if (IS_ERR(obj))
+		return acpi_evaluate_dsm(handle, uuid, rev, func, argv4);
+	return obj;
+}
+EXPORT_SYMBOL(__wrap_acpi_evaluate_dsm);
+
 MODULE_LICENSE("GPL v2");