nvdimm.lists.linux.dev archive mirror
 help / color / mirror / Atom feed
* [PATCH v8 00/12] Adding security support for nvdimm
@ 2018-08-28 22:47 Dave Jiang
  2018-08-28 22:47 ` [PATCH v8 01/12] nfit: add support for Intel DSM 1.7 commands Dave Jiang
                   ` (16 more replies)
  0 siblings, 17 replies; 30+ messages in thread
From: Dave Jiang @ 2018-08-28 22:47 UTC (permalink / raw)
  To: linux-nvdimm; +Cc: alison.schofield, keescook, ebiggers3, dhowells, keyrings

The following series implements security support for nvdimm. Mostly adding
new security DSM support from the Intel NVDIMM DSM spec v1.7, but also
adding generic support libnvdimm for other vendors. The most important
security features are unlocking locked nvdimms, and updating/setting security
passphrase to nvdimms.

Security folks, thanks in advance for taking a look at my key management
implementation and making sure that I'm doing something sane. Mainly you'll
want to review patches 2, 4, and 5 as most relevant ones that need scrutiny.

v8:
- Make the keys retained by the kernel user searchable in order to find the
  key that needs to be updated for key update.

v7:
- Add CONFIG_KEYS depenency for libnvdimm. (Alison)
- Export lookup_user_key(). (David)
- Modified "update" to take two key ids and and use lookup_user_key() in
  order to improve security.  (David)
- Use key ptrs and key_validate() for cached keys. (David)

v6:
- Fix intel DSM data structures to use defined size for passphrase (Robert)
- Fix memcpy size to use sizeof data structure member (Robert)
- Fix defined dimm id length (Robert)
- Making intel_security_ops const (Eric)
- Remove unused var in nvdimm_key_search() (Eric)
- Added wbinvd before secure erase is issued (Robert)
- Removed key_put_sync() usage (David)
- Use init_cred instead of creating own cred (David)
    - Exported init_cred symbol
- Move keyring to dedicated (David)
- Use logon_key_type and friends instead of creating custom (David)
- Use key_lookup() with stored key serial (David)
    - Exported key_lookup() symbol
- Mark passed in key data as const (David)
- Added comment for change_pass_phrase to explain how it works (David)
- Unlink key when it's being removed from keyring. (David)
- Removed request_key() from all security ops except update and unlock.
- Update will now update the existing key's payload with the new key's
  retrieved from userspace when the new payload is accepted by nvdimm.

v5:
- Moved dimm_id initialization (Dan)
- Added a key_put_sync() in order to run key_gc_work and cleanup old key. (Dan)
- Added check to block security state changes while DIMM is active. (Dan)

v4:
- flip payload layout for update passphrase to make it easier on userland.

v3:
- Set x86 wrappers for x86 only bits. (Dan)
- Fixed up some verbiage in commit headers.
- Put in usage of sysfs_streq() for sysfs inputs.
- 0-day build fixes for non-x86 archs.

v2:
- Move inclusion of intel.h to relevant source files and not in nfit.h. (Dan)
- Moved security ring relevant code to dimm_devs.c. (Dan)
- Added dimm_id to nfit_mem to avoid recreate per sysfs show call. (Dan)
- Added routine to return security_ops based on family supplied. (Dan)
- Added nvdimm_key_data struct to wrap raw passphrase string. (Dan)
- Allocate firmware package on stack. (Dan)
- Added missing frozen state detection when retrieving security state.

---

Dave Jiang (12):
      nfit: add support for Intel DSM 1.7 commands
      libnvdimm: create keyring to store security keys
      nfit/libnvdimm: store dimm id as a member to struct nvdimm
      keys: export lookup_user_key to external users
      nfit/libnvdimm: add unlock of nvdimm support for Intel DIMMs
      nfit/libnvdimm: add set passphrase support for Intel nvdimms
      nfit/libnvdimm: add disable passphrase support to Intel nvdimm.
      nfit/libnvdimm: add freeze security support to Intel nvdimm
      nfit/libnvdimm: add support for issue secure erase DSM to Intel nvdimm
      nfit_test: add context to dimm_dev for nfit_test
      nfit_test: add test support for Intel nvdimm security DSMs
      libnvdimm: add documentation for nvdimm security support


 Documentation/nvdimm/security.txt |   82 ++++++
 drivers/acpi/nfit/Makefile        |    1 
 drivers/acpi/nfit/core.c          |   58 +++-
 drivers/acpi/nfit/intel.c         |  382 +++++++++++++++++++++++++++
 drivers/acpi/nfit/intel.h         |   82 ++++++
 drivers/acpi/nfit/nfit.h          |   20 +
 drivers/nvdimm/Kconfig            |    1 
 drivers/nvdimm/bus.c              |    2 
 drivers/nvdimm/core.c             |    7 
 drivers/nvdimm/dimm.c             |    7 
 drivers/nvdimm/dimm_devs.c        |  529 +++++++++++++++++++++++++++++++++++++
 drivers/nvdimm/nd-core.h          |    6 
 drivers/nvdimm/nd.h               |    2 
 include/linux/key.h               |    3 
 include/linux/libnvdimm.h         |   42 +++
 kernel/cred.c                     |    1 
 security/keys/internal.h          |    2 
 security/keys/process_keys.c      |    1 
 tools/testing/nvdimm/Kbuild       |    1 
 tools/testing/nvdimm/test/nfit.c  |  227 +++++++++++++++-
 20 files changed, 1420 insertions(+), 36 deletions(-)
 create mode 100644 Documentation/nvdimm/security.txt
 create mode 100644 drivers/acpi/nfit/intel.c
 create mode 100644 drivers/acpi/nfit/intel.h

--
_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm

^ permalink raw reply	[flat|nested] 30+ messages in thread

* [PATCH v8 01/12] nfit: add support for Intel DSM 1.7 commands
  2018-08-28 22:47 [PATCH v8 00/12] Adding security support for nvdimm Dave Jiang
@ 2018-08-28 22:47 ` Dave Jiang
  2018-08-28 22:47 ` [PATCH v8 02/12] libnvdimm: create keyring to store security keys Dave Jiang
                   ` (15 subsequent siblings)
  16 siblings, 0 replies; 30+ messages in thread
From: Dave Jiang @ 2018-08-28 22:47 UTC (permalink / raw)
  To: linux-nvdimm; +Cc: alison.schofield, keescook, ebiggers3, dhowells, keyrings

Add command definition for security commands defined in Intel DSM
specification v1.7. This includes "get security state", "set passphrase",
"unlock unit", "freeze lock", "secure erase", "ovewrite", and
"overwrite query". Since we are adding a lot of Intel definitions, moving
the relevant bits to its own header. Also, we don't want those commands
to be issued from user space through ioctls. Reason being some of the
security commands require kernel involvement to flush all CPU caches that
can't be done in userspace and the result can cause system crash. So
blocking security commands in the ioctl path.

Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
---
 drivers/acpi/nfit/core.c  |   28 ++++++++++++++++++-
 drivers/acpi/nfit/intel.h |   67 +++++++++++++++++++++++++++++++++++++++++++++
 drivers/acpi/nfit/nfit.h  |   17 +++++++++++
 drivers/nvdimm/bus.c      |    2 +
 include/linux/libnvdimm.h |    2 +
 5 files changed, 112 insertions(+), 4 deletions(-)
 create mode 100644 drivers/acpi/nfit/intel.h

diff --git a/drivers/acpi/nfit/core.c b/drivers/acpi/nfit/core.c
index 0f47917b8ef1..a3a944751e6a 100644
--- a/drivers/acpi/nfit/core.c
+++ b/drivers/acpi/nfit/core.c
@@ -24,6 +24,7 @@
 #include <linux/nd.h>
 #include <asm/cacheflush.h>
 #include <acpi/nfit.h>
+#include "intel.h"
 #include "nfit.h"
 
 /*
@@ -377,6 +378,14 @@ static u8 nfit_dsm_revid(unsigned family, unsigned func)
 			[NVDIMM_INTEL_QUERY_FWUPDATE] = 2,
 			[NVDIMM_INTEL_SET_THRESHOLD] = 2,
 			[NVDIMM_INTEL_INJECT_ERROR] = 2,
+			[NVDIMM_INTEL_GET_SECURITY_STATE] = 2,
+			[NVDIMM_INTEL_SET_PASSPHRASE] = 2,
+			[NVDIMM_INTEL_DISABLE_PASSPHRASE] = 2,
+			[NVDIMM_INTEL_UNLOCK_UNIT] = 2,
+			[NVDIMM_INTEL_FREEZE_LOCK] = 2,
+			[NVDIMM_INTEL_SECURE_ERASE] = 2,
+			[NVDIMM_INTEL_OVERWRITE] = 2,
+			[NVDIMM_INTEL_QUERY_OVERWRITE] = 2,
 		},
 	};
 	u8 id;
@@ -3202,7 +3211,7 @@ static int acpi_nfit_flush_probe(struct nvdimm_bus_descriptor *nd_desc)
 	return 0;
 }
 
-static int acpi_nfit_clear_to_send(struct nvdimm_bus_descriptor *nd_desc,
+static int __acpi_nfit_clear_to_send(struct nvdimm_bus_descriptor *nd_desc,
 		struct nvdimm *nvdimm, unsigned int cmd)
 {
 	struct acpi_nfit_desc *acpi_desc = to_acpi_nfit_desc(nd_desc);
@@ -3224,6 +3233,23 @@ static int acpi_nfit_clear_to_send(struct nvdimm_bus_descriptor *nd_desc,
 	return 0;
 }
 
+/* prevent security commands from being issued via ioctl */
+static int acpi_nfit_clear_to_send(struct nvdimm_bus_descriptor *nd_desc,
+		struct nvdimm *nvdimm, unsigned int cmd, void *buf)
+{
+	struct nd_cmd_pkg *call_pkg = buf;
+	unsigned int func;
+
+	if (nvdimm && cmd == ND_CMD_CALL &&
+			call_pkg->nd_family == NVDIMM_FAMILY_INTEL) {
+		func = call_pkg->nd_command;
+		if ((1 << func) & NVDIMM_INTEL_SECURITY_CMDMASK)
+			return -EOPNOTSUPP;
+	}
+
+	return __acpi_nfit_clear_to_send(nd_desc, nvdimm, cmd);
+}
+
 int acpi_nfit_ars_rescan(struct acpi_nfit_desc *acpi_desc, unsigned long flags)
 {
 	struct device *dev = acpi_desc->dev;
diff --git a/drivers/acpi/nfit/intel.h b/drivers/acpi/nfit/intel.h
new file mode 100644
index 000000000000..a6f8f4bc8ebb
--- /dev/null
+++ b/drivers/acpi/nfit/intel.h
@@ -0,0 +1,67 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 2018 Intel Corporation. All rights reserved. */
+/*
+ * Intel specific definitions for NVDIMM Firmware Interface Table - NFIT
+ */
+#ifndef _NFIT_INTEL_H_
+#define _NFIT_INTEL_H_
+
+#ifdef CONFIG_X86
+
+#define ND_INTEL_STATUS_SIZE		4
+#define ND_INTEL_PASSPHRASE_SIZE	32
+
+#define ND_INTEL_STATUS_RETRY		5
+#define ND_INTEL_STATUS_NOT_READY	9
+#define ND_INTEL_STATUS_INVALID_STATE	10
+#define ND_INTEL_STATUS_INVALID_PASS	11
+
+#define ND_INTEL_SEC_STATE_ENABLED	0x02
+#define ND_INTEL_SEC_STATE_LOCKED	0x04
+#define ND_INTEL_SEC_STATE_FROZEN	0x08
+#define ND_INTEL_SEC_STATE_PLIMIT	0x10
+#define ND_INTEL_SEC_STATE_UNSUPPORTED	0x20
+
+struct nd_intel_get_security_state {
+	u32 status;
+	u32 reserved;
+	u8 state;
+	u8 reserved1[3];
+} __packed;
+
+struct nd_intel_set_passphrase {
+	u8 old_pass[ND_INTEL_PASSPHRASE_SIZE];
+	u8 new_pass[ND_INTEL_PASSPHRASE_SIZE];
+	u32 status;
+} __packed;
+
+struct nd_intel_unlock_unit {
+	u8 passphrase[ND_INTEL_PASSPHRASE_SIZE];
+	u32 status;
+} __packed;
+
+struct nd_intel_disable_passphrase {
+	u8 passphrase[ND_INTEL_PASSPHRASE_SIZE];
+	u32 status;
+} __packed;
+
+struct nd_intel_freeze_lock {
+	u32 status;
+} __packed;
+
+struct nd_intel_secure_erase {
+	u8 passphrase[ND_INTEL_PASSPHRASE_SIZE];
+	u32 status;
+} __packed;
+
+struct nd_intel_overwrite {
+	u8 passphrase[ND_INTEL_PASSPHRASE_SIZE];
+	u32 status;
+} __packed;
+
+struct nd_intel_query_overwrite {
+	u32 status;
+} __packed;
+#endif /* CONFIG_X86 */
+
+#endif
diff --git a/drivers/acpi/nfit/nfit.h b/drivers/acpi/nfit/nfit.h
index 7d15856a739f..40b0003b1805 100644
--- a/drivers/acpi/nfit/nfit.h
+++ b/drivers/acpi/nfit/nfit.h
@@ -60,14 +60,29 @@ enum nvdimm_family_cmds {
 	NVDIMM_INTEL_QUERY_FWUPDATE = 16,
 	NVDIMM_INTEL_SET_THRESHOLD = 17,
 	NVDIMM_INTEL_INJECT_ERROR = 18,
+	NVDIMM_INTEL_GET_SECURITY_STATE = 19,
+	NVDIMM_INTEL_SET_PASSPHRASE = 20,
+	NVDIMM_INTEL_DISABLE_PASSPHRASE = 21,
+	NVDIMM_INTEL_UNLOCK_UNIT = 22,
+	NVDIMM_INTEL_FREEZE_LOCK = 23,
+	NVDIMM_INTEL_SECURE_ERASE = 24,
+	NVDIMM_INTEL_OVERWRITE = 25,
+	NVDIMM_INTEL_QUERY_OVERWRITE = 26,
 };
 
+#define NVDIMM_INTEL_SECURITY_CMDMASK \
+(1 << NVDIMM_INTEL_GET_SECURITY_STATE | 1 << NVDIMM_INTEL_SET_PASSPHRASE \
+| 1 << NVDIMM_INTEL_DISABLE_PASSPHRASE | 1 << NVDIMM_INTEL_UNLOCK_UNIT \
+| 1 << NVDIMM_INTEL_FREEZE_LOCK | 1 << NVDIMM_INTEL_SECURE_ERASE \
+| 1 << NVDIMM_INTEL_OVERWRITE | 1 << NVDIMM_INTEL_QUERY_OVERWRITE)
+
 #define NVDIMM_INTEL_CMDMASK \
 (NVDIMM_STANDARD_CMDMASK | 1 << NVDIMM_INTEL_GET_MODES \
  | 1 << NVDIMM_INTEL_GET_FWINFO | 1 << NVDIMM_INTEL_START_FWUPDATE \
  | 1 << NVDIMM_INTEL_SEND_FWUPDATE | 1 << NVDIMM_INTEL_FINISH_FWUPDATE \
  | 1 << NVDIMM_INTEL_QUERY_FWUPDATE | 1 << NVDIMM_INTEL_SET_THRESHOLD \
- | 1 << NVDIMM_INTEL_INJECT_ERROR | 1 << NVDIMM_INTEL_LATCH_SHUTDOWN)
+ | 1 << NVDIMM_INTEL_INJECT_ERROR | 1 << NVDIMM_INTEL_LATCH_SHUTDOWN \
+ | NVDIMM_INTEL_SECURITY_CMDMASK)
 
 enum nfit_uuids {
 	/* for simplicity alias the uuid index with the family id */
diff --git a/drivers/nvdimm/bus.c b/drivers/nvdimm/bus.c
index 27902a8799b1..de2af49bbf08 100644
--- a/drivers/nvdimm/bus.c
+++ b/drivers/nvdimm/bus.c
@@ -894,7 +894,7 @@ static int nd_cmd_clear_to_send(struct nvdimm_bus *nvdimm_bus,
 
 	/* ask the bus provider if it would like to block this request */
 	if (nd_desc->clear_to_send) {
-		int rc = nd_desc->clear_to_send(nd_desc, nvdimm, cmd);
+		int rc = nd_desc->clear_to_send(nd_desc, nvdimm, cmd, data);
 
 		if (rc)
 			return rc;
diff --git a/include/linux/libnvdimm.h b/include/linux/libnvdimm.h
index 097072c5a852..472171af7f60 100644
--- a/include/linux/libnvdimm.h
+++ b/include/linux/libnvdimm.h
@@ -87,7 +87,7 @@ struct nvdimm_bus_descriptor {
 	ndctl_fn ndctl;
 	int (*flush_probe)(struct nvdimm_bus_descriptor *nd_desc);
 	int (*clear_to_send)(struct nvdimm_bus_descriptor *nd_desc,
-			struct nvdimm *nvdimm, unsigned int cmd);
+			struct nvdimm *nvdimm, unsigned int cmd, void *data);
 };
 
 struct nd_cmd_desc {

_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm

^ permalink raw reply	[flat|nested] 30+ messages in thread

* [PATCH v8 02/12] libnvdimm: create keyring to store security keys
  2018-08-28 22:47 [PATCH v8 00/12] Adding security support for nvdimm Dave Jiang
  2018-08-28 22:47 ` [PATCH v8 01/12] nfit: add support for Intel DSM 1.7 commands Dave Jiang
@ 2018-08-28 22:47 ` Dave Jiang
  2018-09-22  0:19   ` Dan Williams
  2018-09-24 21:04   ` David Howells
  2018-08-28 22:47 ` [PATCH v8 03/12] nfit/libnvdimm: store dimm id as a member to struct nvdimm Dave Jiang
                   ` (14 subsequent siblings)
  16 siblings, 2 replies; 30+ messages in thread
From: Dave Jiang @ 2018-08-28 22:47 UTC (permalink / raw)
  To: linux-nvdimm; +Cc: alison.schofield, keescook, ebiggers3, dhowells, keyrings

Prepping the libnvdimm to support security management by adding a keyring
in order to provide passphrase management through the kernel key management
APIs.

Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
---
 drivers/nvdimm/core.c      |    7 ++++++-
 drivers/nvdimm/dimm_devs.c |   29 ++++++++++++++++++++++++++++-
 drivers/nvdimm/nd-core.h   |    1 +
 include/linux/libnvdimm.h  |    3 +++
 kernel/cred.c              |    1 +
 5 files changed, 39 insertions(+), 2 deletions(-)

diff --git a/drivers/nvdimm/core.c b/drivers/nvdimm/core.c
index acce050856a8..3cd33d5c7cf0 100644
--- a/drivers/nvdimm/core.c
+++ b/drivers/nvdimm/core.c
@@ -437,9 +437,12 @@ static __init int libnvdimm_init(void)
 {
 	int rc;
 
-	rc = nvdimm_bus_init();
+	rc = nvdimm_devs_init();
 	if (rc)
 		return rc;
+	rc = nvdimm_bus_init();
+	if (rc)
+		goto err_bus;
 	rc = nvdimm_init();
 	if (rc)
 		goto err_dimm;
@@ -454,6 +457,8 @@ static __init int libnvdimm_init(void)
 	nvdimm_exit();
  err_dimm:
 	nvdimm_bus_exit();
+ err_bus:
+	nvdimm_devs_exit();
 	return rc;
 }
 
diff --git a/drivers/nvdimm/dimm_devs.c b/drivers/nvdimm/dimm_devs.c
index 8d348b22ba45..586432450c82 100644
--- a/drivers/nvdimm/dimm_devs.c
+++ b/drivers/nvdimm/dimm_devs.c
@@ -18,12 +18,15 @@
 #include <linux/io.h>
 #include <linux/fs.h>
 #include <linux/mm.h>
+#include <linux/key.h>
+#include <linux/init_task.h>
 #include "nd-core.h"
 #include "label.h"
 #include "pmem.h"
 #include "nd.h"
 
 static DEFINE_IDA(dimm_ida);
+static struct key *nvdimm_keyring;
 
 /*
  * Retrieve bus and dimm handle and return if this bus supports
@@ -668,7 +671,31 @@ int nvdimm_bus_check_dimm_count(struct nvdimm_bus *nvdimm_bus, int dimm_count)
 }
 EXPORT_SYMBOL_GPL(nvdimm_bus_check_dimm_count);
 
-void __exit nvdimm_devs_exit(void)
+static int nvdimm_register_keyring(void)
 {
+	nvdimm_keyring = keyring_alloc(".nvdimm",
+			GLOBAL_ROOT_UID, GLOBAL_ROOT_GID, &init_cred,
+			(KEY_POS_ALL & ~KEY_POS_SETATTR) |
+			(KEY_USR_ALL & ~KEY_USR_SETATTR),
+			KEY_ALLOC_NOT_IN_QUOTA, NULL, NULL);
+	if (IS_ERR(nvdimm_keyring))
+		return PTR_ERR(nvdimm_keyring);
+
+	return 0;
+}
+
+static void nvdimm_unregister_keyring(void)
+{
+	key_revoke(nvdimm_keyring);
+}
+
+int __init nvdimm_devs_init(void)
+{
+	return nvdimm_register_keyring();
+}
+
+void  nvdimm_devs_exit(void)
+{
+	nvdimm_unregister_keyring();
 	ida_destroy(&dimm_ida);
 }
diff --git a/drivers/nvdimm/nd-core.h b/drivers/nvdimm/nd-core.h
index 79274ead54fb..2af0f89c4010 100644
--- a/drivers/nvdimm/nd-core.h
+++ b/drivers/nvdimm/nd-core.h
@@ -76,6 +76,7 @@ static inline bool is_memory(struct device *dev)
 struct nvdimm_bus *walk_to_nvdimm_bus(struct device *nd_dev);
 int __init nvdimm_bus_init(void);
 void nvdimm_bus_exit(void);
+int nvdimm_devs_init(void);
 void nvdimm_devs_exit(void);
 void nd_region_devs_exit(void);
 void nd_region_probe_success(struct nvdimm_bus *nvdimm_bus, struct device *dev);
diff --git a/include/linux/libnvdimm.h b/include/linux/libnvdimm.h
index 472171af7f60..51084043915c 100644
--- a/include/linux/libnvdimm.h
+++ b/include/linux/libnvdimm.h
@@ -155,6 +155,9 @@ static inline struct nd_blk_region_desc *to_blk_region_desc(
 
 }
 
+#define NVDIMM_PASSPHRASE_LEN		32
+#define NVDIMM_KEY_DESC_LEN		22
+
 void badrange_init(struct badrange *badrange);
 int badrange_add(struct badrange *badrange, u64 addr, u64 length);
 void badrange_forget(struct badrange *badrange, phys_addr_t start,
diff --git a/kernel/cred.c b/kernel/cred.c
index ecf03657e71c..3a161b78a0a4 100644
--- a/kernel/cred.c
+++ b/kernel/cred.c
@@ -64,6 +64,7 @@ struct cred init_cred = {
 	.user_ns		= &init_user_ns,
 	.group_info		= &init_groups,
 };
+EXPORT_SYMBOL(init_cred);
 
 static inline void set_cred_subscribers(struct cred *cred, int n)
 {

_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm

^ permalink raw reply	[flat|nested] 30+ messages in thread

* [PATCH v8 03/12] nfit/libnvdimm: store dimm id as a member to struct nvdimm
  2018-08-28 22:47 [PATCH v8 00/12] Adding security support for nvdimm Dave Jiang
  2018-08-28 22:47 ` [PATCH v8 01/12] nfit: add support for Intel DSM 1.7 commands Dave Jiang
  2018-08-28 22:47 ` [PATCH v8 02/12] libnvdimm: create keyring to store security keys Dave Jiang
@ 2018-08-28 22:47 ` Dave Jiang
  2018-08-28 22:47 ` [PATCH v8 04/12] keys: export lookup_user_key to external users Dave Jiang
                   ` (13 subsequent siblings)
  16 siblings, 0 replies; 30+ messages in thread
From: Dave Jiang @ 2018-08-28 22:47 UTC (permalink / raw)
  To: linux-nvdimm; +Cc: alison.schofield, keescook, ebiggers3, dhowells, keyrings

The generated dimm id is needed for the sysfs attribute as well as being
used as the identifier/description for the security key. Since it's
constant and should never change, store it as a member of struct nvdimm.

Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
---
 drivers/acpi/nfit/core.c   |   29 +++++++++++++++++------------
 drivers/acpi/nfit/nfit.h   |    3 +++
 drivers/nvdimm/dimm_devs.c |    4 +++-
 drivers/nvdimm/nd-core.h   |    1 +
 include/linux/libnvdimm.h  |    2 +-
 5 files changed, 25 insertions(+), 14 deletions(-)

diff --git a/drivers/acpi/nfit/core.c b/drivers/acpi/nfit/core.c
index a3a944751e6a..21235d555b5f 100644
--- a/drivers/acpi/nfit/core.c
+++ b/drivers/acpi/nfit/core.c
@@ -1572,18 +1572,10 @@ static DEVICE_ATTR_RO(flags);
 static ssize_t id_show(struct device *dev,
 		struct device_attribute *attr, char *buf)
 {
-	struct acpi_nfit_control_region *dcr = to_nfit_dcr(dev);
+	struct nvdimm *nvdimm = to_nvdimm(dev);
+	struct nfit_mem *nfit_mem = nvdimm_provider_data(nvdimm);
 
-	if (dcr->valid_fields & ACPI_NFIT_CONTROL_MFG_INFO_VALID)
-		return sprintf(buf, "%04x-%02x-%04x-%08x\n",
-				be16_to_cpu(dcr->vendor_id),
-				dcr->manufacturing_location,
-				be16_to_cpu(dcr->manufacturing_date),
-				be32_to_cpu(dcr->serial_number));
-	else
-		return sprintf(buf, "%04x-%08x\n",
-				be16_to_cpu(dcr->vendor_id),
-				be32_to_cpu(dcr->serial_number));
+	return sprintf(buf, "%s\n", nfit_mem->id);
 }
 static DEVICE_ATTR_RO(id);
 
@@ -1712,10 +1704,23 @@ static int acpi_nfit_add_dimm(struct acpi_nfit_desc *acpi_desc,
 	const guid_t *guid;
 	int i;
 	int family = -1;
+	struct acpi_nfit_control_region *dcr = nfit_mem->dcr;
 
 	/* nfit test assumes 1:1 relationship between commands and dsms */
 	nfit_mem->dsm_mask = acpi_desc->dimm_cmd_force_en;
 	nfit_mem->family = NVDIMM_FAMILY_INTEL;
+
+	if (dcr->valid_fields & ACPI_NFIT_CONTROL_MFG_INFO_VALID)
+		sprintf(nfit_mem->id, "%04x-%02x-%04x-%08x",
+				be16_to_cpu(dcr->vendor_id),
+				dcr->manufacturing_location,
+				be16_to_cpu(dcr->manufacturing_date),
+				be32_to_cpu(dcr->serial_number));
+	else
+		sprintf(nfit_mem->id, "%04x-%08x",
+				be16_to_cpu(dcr->vendor_id),
+				be32_to_cpu(dcr->serial_number));
+
 	adev = to_acpi_dev(acpi_desc);
 	if (!adev)
 		return 0;
@@ -1899,7 +1904,7 @@ static int acpi_nfit_register_dimms(struct acpi_nfit_desc *acpi_desc)
 		nvdimm = nvdimm_create(acpi_desc->nvdimm_bus, nfit_mem,
 				acpi_nfit_dimm_attribute_groups,
 				flags, cmd_mask, flush ? flush->hint_count : 0,
-				nfit_mem->flush_wpq);
+				nfit_mem->flush_wpq, &nfit_mem->id[0]);
 		if (!nvdimm)
 			return -ENOMEM;
 
diff --git a/drivers/acpi/nfit/nfit.h b/drivers/acpi/nfit/nfit.h
index 40b0003b1805..4048036d6ec2 100644
--- a/drivers/acpi/nfit/nfit.h
+++ b/drivers/acpi/nfit/nfit.h
@@ -173,6 +173,8 @@ struct nfit_memdev {
 	struct acpi_nfit_memory_map memdev[0];
 };
 
+#define NFIT_DIMM_ID_LEN	22
+
 /* assembled tables for a given dimm/memory-device */
 struct nfit_mem {
 	struct nvdimm *nvdimm;
@@ -195,6 +197,7 @@ struct nfit_mem {
 	int family;
 	bool has_lsr;
 	bool has_lsw;
+	char id[NFIT_DIMM_ID_LEN+1];
 };
 
 struct acpi_nfit_desc {
diff --git a/drivers/nvdimm/dimm_devs.c b/drivers/nvdimm/dimm_devs.c
index 586432450c82..e753b2095402 100644
--- a/drivers/nvdimm/dimm_devs.c
+++ b/drivers/nvdimm/dimm_devs.c
@@ -401,7 +401,7 @@ EXPORT_SYMBOL_GPL(nvdimm_attribute_group);
 struct nvdimm *nvdimm_create(struct nvdimm_bus *nvdimm_bus, void *provider_data,
 		const struct attribute_group **groups, unsigned long flags,
 		unsigned long cmd_mask, int num_flush,
-		struct resource *flush_wpq)
+		struct resource *flush_wpq, const char *dimm_id)
 {
 	struct nvdimm *nvdimm = kzalloc(sizeof(*nvdimm), GFP_KERNEL);
 	struct device *dev;
@@ -414,6 +414,8 @@ struct nvdimm *nvdimm_create(struct nvdimm_bus *nvdimm_bus, void *provider_data,
 		kfree(nvdimm);
 		return NULL;
 	}
+
+	nvdimm->dimm_id = dimm_id;
 	nvdimm->provider_data = provider_data;
 	nvdimm->flags = flags;
 	nvdimm->cmd_mask = cmd_mask;
diff --git a/drivers/nvdimm/nd-core.h b/drivers/nvdimm/nd-core.h
index 2af0f89c4010..ed650f1607d8 100644
--- a/drivers/nvdimm/nd-core.h
+++ b/drivers/nvdimm/nd-core.h
@@ -42,6 +42,7 @@ struct nvdimm {
 	atomic_t busy;
 	int id, num_flush;
 	struct resource *flush_wpq;
+	const char *dimm_id;
 };
 
 /**
diff --git a/include/linux/libnvdimm.h b/include/linux/libnvdimm.h
index 51084043915c..3c1f18fdf76f 100644
--- a/include/linux/libnvdimm.h
+++ b/include/linux/libnvdimm.h
@@ -181,7 +181,7 @@ void *nvdimm_provider_data(struct nvdimm *nvdimm);
 struct nvdimm *nvdimm_create(struct nvdimm_bus *nvdimm_bus, void *provider_data,
 		const struct attribute_group **groups, unsigned long flags,
 		unsigned long cmd_mask, int num_flush,
-		struct resource *flush_wpq);
+		struct resource *flush_wpq, const char *dimm_id);
 const struct nd_cmd_desc *nd_cmd_dimm_desc(int cmd);
 const struct nd_cmd_desc *nd_cmd_bus_desc(int cmd);
 u32 nd_cmd_in_size(struct nvdimm *nvdimm, int cmd,

_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm

^ permalink raw reply	[flat|nested] 30+ messages in thread

* [PATCH v8 04/12] keys: export lookup_user_key to external users
  2018-08-28 22:47 [PATCH v8 00/12] Adding security support for nvdimm Dave Jiang
                   ` (2 preceding siblings ...)
  2018-08-28 22:47 ` [PATCH v8 03/12] nfit/libnvdimm: store dimm id as a member to struct nvdimm Dave Jiang
@ 2018-08-28 22:47 ` Dave Jiang
  2018-09-21 21:59   ` Dan Williams
  2018-09-21 23:05   ` David Howells
  2018-08-28 22:47 ` [PATCH v8 05/12] nfit/libnvdimm: add unlock of nvdimm support for Intel DIMMs Dave Jiang
                   ` (12 subsequent siblings)
  16 siblings, 2 replies; 30+ messages in thread
From: Dave Jiang @ 2018-08-28 22:47 UTC (permalink / raw)
  To: linux-nvdimm; +Cc: alison.schofield, keescook, ebiggers3, dhowells, keyrings

Export lookup_user_key() symbol in order to allow nvdimm passphrase
update to retrieve user injected keys.

Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Cc: David Howells <dhowells@redhat.com>
---
 include/linux/key.h          |    3 +++
 security/keys/internal.h     |    2 --
 security/keys/process_keys.c |    1 +
 3 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/include/linux/key.h b/include/linux/key.h
index e58ee10f6e58..7099985e35a9 100644
--- a/include/linux/key.h
+++ b/include/linux/key.h
@@ -346,6 +346,9 @@ static inline key_serial_t key_serial(const struct key *key)
 
 extern void key_set_timeout(struct key *, unsigned);
 
+extern key_ref_t lookup_user_key(key_serial_t id, unsigned long flags,
+				 key_perm_t perm);
+
 /*
  * The permissions required on a key that we're looking up.
  */
diff --git a/security/keys/internal.h b/security/keys/internal.h
index 9f8208dc0e55..9968b21a76dd 100644
--- a/security/keys/internal.h
+++ b/security/keys/internal.h
@@ -158,8 +158,6 @@ extern struct key *request_key_and_link(struct key_type *type,
 
 extern bool lookup_user_key_possessed(const struct key *key,
 				      const struct key_match_data *match_data);
-extern key_ref_t lookup_user_key(key_serial_t id, unsigned long flags,
-				 key_perm_t perm);
 #define KEY_LOOKUP_CREATE	0x01
 #define KEY_LOOKUP_PARTIAL	0x02
 #define KEY_LOOKUP_FOR_UNLINK	0x04
diff --git a/security/keys/process_keys.c b/security/keys/process_keys.c
index d5b25e535d3a..ec4fd4531224 100644
--- a/security/keys/process_keys.c
+++ b/security/keys/process_keys.c
@@ -755,6 +755,7 @@ key_ref_t lookup_user_key(key_serial_t id, unsigned long lflags,
 	put_cred(ctx.cred);
 	goto try_again;
 }
+EXPORT_SYMBOL(lookup_user_key);
 
 /*
  * Join the named keyring as the session keyring if possible else attempt to

_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm

^ permalink raw reply	[flat|nested] 30+ messages in thread

* [PATCH v8 05/12] nfit/libnvdimm: add unlock of nvdimm support for Intel DIMMs
  2018-08-28 22:47 [PATCH v8 00/12] Adding security support for nvdimm Dave Jiang
                   ` (3 preceding siblings ...)
  2018-08-28 22:47 ` [PATCH v8 04/12] keys: export lookup_user_key to external users Dave Jiang
@ 2018-08-28 22:47 ` Dave Jiang
  2018-09-23  0:10   ` Dan Williams
  2018-08-28 22:47 ` [PATCH v8 06/12] nfit/libnvdimm: add set passphrase support for Intel nvdimms Dave Jiang
                   ` (11 subsequent siblings)
  16 siblings, 1 reply; 30+ messages in thread
From: Dave Jiang @ 2018-08-28 22:47 UTC (permalink / raw)
  To: linux-nvdimm; +Cc: alison.schofield, keescook, ebiggers3, dhowells, keyrings

Add support to allow query the security status of the Intel nvdimms and
also unlock the dimm via the kernel key management APIs. The passphrase is
expected to be pulled from userspace through keyutils. Moving the Intel
related bits to its own source file as well.

Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
---
 drivers/acpi/nfit/Makefile |    1 
 drivers/acpi/nfit/core.c   |    3 +
 drivers/acpi/nfit/intel.c  |  152 ++++++++++++++++++++++++++++++++++++++++++++
 drivers/acpi/nfit/intel.h  |   15 ++++
 drivers/nvdimm/Kconfig     |    1 
 drivers/nvdimm/dimm.c      |    7 ++
 drivers/nvdimm/dimm_devs.c |  128 +++++++++++++++++++++++++++++++++++++
 drivers/nvdimm/nd-core.h   |    4 +
 drivers/nvdimm/nd.h        |    2 +
 include/linux/libnvdimm.h  |   24 +++++++
 10 files changed, 334 insertions(+), 3 deletions(-)
 create mode 100644 drivers/acpi/nfit/intel.c

diff --git a/drivers/acpi/nfit/Makefile b/drivers/acpi/nfit/Makefile
index a407e769f103..443c7ef4e6a6 100644
--- a/drivers/acpi/nfit/Makefile
+++ b/drivers/acpi/nfit/Makefile
@@ -1,3 +1,4 @@
 obj-$(CONFIG_ACPI_NFIT) := nfit.o
 nfit-y := core.o
+nfit-$(CONFIG_X86) += intel.o
 nfit-$(CONFIG_X86_MCE) += mce.o
diff --git a/drivers/acpi/nfit/core.c b/drivers/acpi/nfit/core.c
index 21235d555b5f..9cb6a108ecba 100644
--- a/drivers/acpi/nfit/core.c
+++ b/drivers/acpi/nfit/core.c
@@ -1904,7 +1904,8 @@ static int acpi_nfit_register_dimms(struct acpi_nfit_desc *acpi_desc)
 		nvdimm = nvdimm_create(acpi_desc->nvdimm_bus, nfit_mem,
 				acpi_nfit_dimm_attribute_groups,
 				flags, cmd_mask, flush ? flush->hint_count : 0,
-				nfit_mem->flush_wpq, &nfit_mem->id[0]);
+				nfit_mem->flush_wpq, &nfit_mem->id[0],
+				acpi_nfit_get_security_ops(nfit_mem->family));
 		if (!nvdimm)
 			return -ENOMEM;
 
diff --git a/drivers/acpi/nfit/intel.c b/drivers/acpi/nfit/intel.c
new file mode 100644
index 000000000000..4bfc1c1da339
--- /dev/null
+++ b/drivers/acpi/nfit/intel.c
@@ -0,0 +1,152 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2018 Intel Corporation. All rights reserved. */
+/*
+ * Intel specific NFIT ops
+ */
+#include <linux/libnvdimm.h>
+#include <linux/module.h>
+#include <linux/mutex.h>
+#include <linux/ndctl.h>
+#include <linux/sysfs.h>
+#include <linux/delay.h>
+#include <linux/acpi.h>
+#include <linux/io.h>
+#include <linux/nd.h>
+#include <asm/cacheflush.h>
+#include <asm/smp.h>
+#include <acpi/nfit.h>
+#include "intel.h"
+#include "nfit.h"
+
+static int intel_dimm_security_unlock(struct nvdimm_bus *nvdimm_bus,
+		struct nvdimm *nvdimm, const struct nvdimm_key_data *nkey)
+{
+	struct nvdimm_bus_descriptor *nd_desc = to_nd_desc(nvdimm_bus);
+	int cmd_rc, rc = 0;
+	struct nfit_mem *nfit_mem = nvdimm_provider_data(nvdimm);
+	struct {
+		struct nd_cmd_pkg pkg;
+		struct nd_intel_unlock_unit cmd;
+	} nd_cmd = {
+		.pkg = {
+			.nd_command = NVDIMM_INTEL_UNLOCK_UNIT,
+			.nd_family = NVDIMM_FAMILY_INTEL,
+			.nd_size_in = ND_INTEL_PASSPHRASE_SIZE,
+			.nd_size_out = ND_INTEL_STATUS_SIZE,
+			.nd_fw_size = ND_INTEL_STATUS_SIZE,
+		},
+		.cmd = {
+			.status = 0,
+		},
+	};
+
+	if (!test_bit(NVDIMM_INTEL_UNLOCK_UNIT, &nfit_mem->dsm_mask))
+		return -ENOTTY;
+
+	memcpy(nd_cmd.cmd.passphrase, nkey->data,
+			sizeof(nd_cmd.cmd.passphrase));
+	rc = nd_desc->ndctl(nd_desc, nvdimm, ND_CMD_CALL, &nd_cmd,
+			sizeof(nd_cmd), &cmd_rc);
+	if (rc < 0)
+		goto out;
+	if (cmd_rc < 0) {
+		rc = cmd_rc;
+		goto out;
+	}
+
+	switch (nd_cmd.cmd.status) {
+	case 0:
+		break;
+	case ND_INTEL_STATUS_INVALID_PASS:
+		rc = -EINVAL;
+		goto out;
+	case ND_INTEL_STATUS_INVALID_STATE:
+	default:
+		rc = -ENXIO;
+		goto out;
+	}
+
+	/*
+	 * TODO: define a cross arch wbinvd when/if NVDIMM_FAMILY_INTEL
+	 * support arrives on another arch.
+	 */
+	/* DIMM unlocked, invalidate all CPU caches before we read it */
+	wbinvd_on_all_cpus();
+
+ out:
+	return rc;
+}
+
+static int intel_dimm_security_state(struct nvdimm_bus *nvdimm_bus,
+		struct nvdimm *nvdimm, enum nvdimm_security_state *state)
+{
+	struct nvdimm_bus_descriptor *nd_desc = to_nd_desc(nvdimm_bus);
+	int cmd_rc, rc = 0;
+	struct nfit_mem *nfit_mem = nvdimm_provider_data(nvdimm);
+	struct {
+		struct nd_cmd_pkg pkg;
+		struct nd_intel_get_security_state cmd;
+	} nd_cmd = {
+		.pkg = {
+			.nd_command = NVDIMM_INTEL_GET_SECURITY_STATE,
+			.nd_family = NVDIMM_FAMILY_INTEL,
+			.nd_size_in = 0,
+			.nd_size_out =
+				sizeof(struct nd_intel_get_security_state),
+			.nd_fw_size =
+				sizeof(struct nd_intel_get_security_state),
+		},
+		.cmd = {
+			.status = 0,
+			.state = 0,
+		},
+	};
+
+	if (!test_bit(NVDIMM_INTEL_GET_SECURITY_STATE, &nfit_mem->dsm_mask)) {
+		*state = NVDIMM_SECURITY_UNSUPPORTED;
+		return 0;
+	}
+
+	*state = NVDIMM_SECURITY_DISABLED;
+	rc = nd_desc->ndctl(nd_desc, nvdimm, ND_CMD_CALL, &nd_cmd,
+			sizeof(nd_cmd), &cmd_rc);
+	if (rc < 0)
+		goto out;
+	if (cmd_rc < 0) {
+		rc = cmd_rc;
+		goto out;
+	}
+
+	switch (nd_cmd.cmd.status) {
+	case 0:
+		break;
+	case ND_INTEL_STATUS_RETRY:
+		rc = -EAGAIN;
+		goto out;
+	case ND_INTEL_STATUS_NOT_READY:
+	default:
+		rc = -ENXIO;
+		goto out;
+	}
+
+	/* check and see if security is enabled and locked */
+	if (nd_cmd.cmd.state & ND_INTEL_SEC_STATE_UNSUPPORTED)
+		*state = NVDIMM_SECURITY_UNSUPPORTED;
+	else if (nd_cmd.cmd.state & ND_INTEL_SEC_STATE_ENABLED) {
+		if (nd_cmd.cmd.state & ND_INTEL_SEC_STATE_LOCKED)
+			*state = NVDIMM_SECURITY_LOCKED;
+		else
+			*state = NVDIMM_SECURITY_UNLOCKED;
+	} else
+		*state = NVDIMM_SECURITY_DISABLED;
+
+ out:
+	if (rc < 0)
+		*state = NVDIMM_SECURITY_INVALID;
+	return rc;
+}
+
+const struct nvdimm_security_ops intel_security_ops = {
+	.state = intel_dimm_security_state,
+	.unlock = intel_dimm_security_unlock,
+};
diff --git a/drivers/acpi/nfit/intel.h b/drivers/acpi/nfit/intel.h
index a6f8f4bc8ebb..f9ac718776ca 100644
--- a/drivers/acpi/nfit/intel.h
+++ b/drivers/acpi/nfit/intel.h
@@ -8,6 +8,8 @@
 
 #ifdef CONFIG_X86
 
+extern const struct nvdimm_security_ops intel_security_ops;
+
 #define ND_INTEL_STATUS_SIZE		4
 #define ND_INTEL_PASSPHRASE_SIZE	32
 
@@ -64,4 +66,17 @@ struct nd_intel_query_overwrite {
 } __packed;
 #endif /* CONFIG_X86 */
 
+static inline const struct nvdimm_security_ops *
+acpi_nfit_get_security_ops(int family)
+{
+	switch (family) {
+#ifdef CONFIG_X86
+	case NVDIMM_FAMILY_INTEL:
+		return &intel_security_ops;
+#endif
+	default:
+		return NULL;
+	}
+}
+
 #endif
diff --git a/drivers/nvdimm/Kconfig b/drivers/nvdimm/Kconfig
index 9d36473dc2a2..830a7a5342e1 100644
--- a/drivers/nvdimm/Kconfig
+++ b/drivers/nvdimm/Kconfig
@@ -3,6 +3,7 @@ menuconfig LIBNVDIMM
 	depends on PHYS_ADDR_T_64BIT
 	depends on HAS_IOMEM
 	depends on BLK_DEV
+	depends on KEYS
 	help
 	  Generic support for non-volatile memory devices including
 	  ACPI-6-NFIT defined resources.  On platforms that define an
diff --git a/drivers/nvdimm/dimm.c b/drivers/nvdimm/dimm.c
index 6c8fb7590838..b6381ddbd6c1 100644
--- a/drivers/nvdimm/dimm.c
+++ b/drivers/nvdimm/dimm.c
@@ -51,6 +51,13 @@ static int nvdimm_probe(struct device *dev)
 	get_device(dev);
 	kref_init(&ndd->kref);
 
+	nvdimm_security_get_state(dev);
+
+	/* unlock DIMM here before touch label */
+	rc = nvdimm_security_unlock_dimm(dev);
+	if (rc < 0)
+		dev_warn(dev, "failed to unlock dimm %s\n", dev_name(dev));
+
 	/*
 	 * EACCES failures reading the namespace label-area-properties
 	 * are interpreted as the DIMM capacity being locked but the
diff --git a/drivers/nvdimm/dimm_devs.c b/drivers/nvdimm/dimm_devs.c
index e753b2095402..fe61f2a2ad5d 100644
--- a/drivers/nvdimm/dimm_devs.c
+++ b/drivers/nvdimm/dimm_devs.c
@@ -20,6 +20,7 @@
 #include <linux/mm.h>
 #include <linux/key.h>
 #include <linux/init_task.h>
+#include <keys/user-type.h>
 #include "nd-core.h"
 #include "label.h"
 #include "pmem.h"
@@ -28,6 +29,129 @@
 static DEFINE_IDA(dimm_ida);
 static struct key *nvdimm_keyring;
 
+#define NVDIMM_PREFIX          "nvdimm:"
+
+/*
+ * Find key in kernel keyring
+ */
+static struct key *nvdimm_get_key(struct device *dev)
+{
+	struct nvdimm *nvdimm = to_nvdimm(dev);
+
+	if (!nvdimm->key)
+		return NULL;
+
+	if (key_validate(nvdimm->key) < 0)
+		return NULL;
+
+	key_get(nvdimm->key);
+
+	dev_dbg(dev, "%s: key found: %d\n", __func__,
+			key_serial(nvdimm->key));
+	return nvdimm->key;
+}
+
+/*
+ * Retrieve kernel key for DIMM and request from user space if necessary.
+ */
+static struct key *nvdimm_request_key(struct device *dev)
+{
+	struct nvdimm *nvdimm = to_nvdimm(dev);
+	struct key *key = NULL;
+	char desc[NVDIMM_KEY_DESC_LEN + strlen(NVDIMM_PREFIX)];
+
+	sprintf(desc, "%s%s", NVDIMM_PREFIX, nvdimm->dimm_id);
+	key = request_key(&key_type_logon, desc, desc);
+	if (IS_ERR(key))
+		key = NULL;
+
+	return key;
+}
+
+static int nvdimm_check_key_len(unsigned short len, bool update)
+{
+	if (len == (NVDIMM_PASSPHRASE_LEN * 2) && update)
+		return 0;
+
+	if (len == NVDIMM_PASSPHRASE_LEN)
+		return 0;
+
+	return -EINVAL;
+}
+
+int nvdimm_security_get_state(struct device *dev)
+{
+	struct nvdimm *nvdimm = to_nvdimm(dev);
+	struct nvdimm_bus *nvdimm_bus = walk_to_nvdimm_bus(dev);
+
+	if (!nvdimm->security_ops)
+		return 0;
+
+	return nvdimm->security_ops->state(nvdimm_bus, nvdimm,
+			&nvdimm->state);
+}
+
+int nvdimm_security_unlock_dimm(struct device *dev)
+{
+	struct nvdimm *nvdimm = to_nvdimm(dev);
+	struct nvdimm_bus *nvdimm_bus = walk_to_nvdimm_bus(dev);
+	struct key *key;
+	struct user_key_payload *payload;
+	int rc;
+	bool cached_key = false;
+
+	if (!nvdimm->security_ops)
+		return 0;
+
+	if (nvdimm->state == NVDIMM_SECURITY_UNLOCKED ||
+			nvdimm->state == NVDIMM_SECURITY_UNSUPPORTED ||
+			nvdimm->state == NVDIMM_SECURITY_DISABLED)
+		return 0;
+
+	key = nvdimm_get_key(dev);
+	if (!key)
+		key = nvdimm_request_key(dev);
+	else
+		cached_key = true;
+	if (!key)
+		return -ENXIO;
+
+	if (!cached_key) {
+		rc = nvdimm_check_key_len(key->datalen, false);
+		if (rc < 0) {
+			key_invalidate(key);
+			key_put(key);
+			return rc;
+		}
+	}
+
+	dev_dbg(dev, "%s: key: %#x\n", __func__, key_serial(key));
+	down_read(&key->sem);
+	payload = key->payload.data[0];
+	rc = nvdimm->security_ops->unlock(nvdimm_bus, nvdimm,
+			(const void *)payload->data);
+	up_read(&key->sem);
+
+	if (rc == 0) {
+		if (!cached_key) {
+			key_link(nvdimm_keyring, key);
+			nvdimm->key = key;
+			key->perm |= KEY_USR_SEARCH;
+		}
+		nvdimm->state = NVDIMM_SECURITY_UNLOCKED;
+		dev_info(dev, "DIMM %s unlocked\n", dev_name(dev));
+	} else {
+		key_unlink(nvdimm_keyring, key);
+		key_invalidate(key);
+		nvdimm->key = NULL;
+		dev_warn(dev, "Failed to unlock dimm: %s\n", dev_name(dev));
+	}
+
+	key_put(key);
+	nvdimm_security_get_state(dev);
+	return rc;
+}
+
 /*
  * Retrieve bus and dimm handle and return if this bus supports
  * get_config_data commands
@@ -401,7 +525,8 @@ EXPORT_SYMBOL_GPL(nvdimm_attribute_group);
 struct nvdimm *nvdimm_create(struct nvdimm_bus *nvdimm_bus, void *provider_data,
 		const struct attribute_group **groups, unsigned long flags,
 		unsigned long cmd_mask, int num_flush,
-		struct resource *flush_wpq, const char *dimm_id)
+		struct resource *flush_wpq, const char *dimm_id,
+		const struct nvdimm_security_ops *sec_ops)
 {
 	struct nvdimm *nvdimm = kzalloc(sizeof(*nvdimm), GFP_KERNEL);
 	struct device *dev;
@@ -416,6 +541,7 @@ struct nvdimm *nvdimm_create(struct nvdimm_bus *nvdimm_bus, void *provider_data,
 	}
 
 	nvdimm->dimm_id = dimm_id;
+	nvdimm->security_ops = sec_ops;
 	nvdimm->provider_data = provider_data;
 	nvdimm->flags = flags;
 	nvdimm->cmd_mask = cmd_mask;
diff --git a/drivers/nvdimm/nd-core.h b/drivers/nvdimm/nd-core.h
index ed650f1607d8..47ad63e4d522 100644
--- a/drivers/nvdimm/nd-core.h
+++ b/drivers/nvdimm/nd-core.h
@@ -18,6 +18,7 @@
 #include <linux/sizes.h>
 #include <linux/mutex.h>
 #include <linux/nd.h>
+#include <linux/key.h>
 
 extern struct list_head nvdimm_bus_list;
 extern struct mutex nvdimm_bus_list_mutex;
@@ -43,6 +44,9 @@ struct nvdimm {
 	int id, num_flush;
 	struct resource *flush_wpq;
 	const char *dimm_id;
+	const struct nvdimm_security_ops *security_ops;
+	enum nvdimm_security_state state;
+	struct key *key;
 };
 
 /**
diff --git a/drivers/nvdimm/nd.h b/drivers/nvdimm/nd.h
index 9d17abd9f8d0..9c80e0f8c327 100644
--- a/drivers/nvdimm/nd.h
+++ b/drivers/nvdimm/nd.h
@@ -424,4 +424,6 @@ static inline bool is_bad_pmem(struct badblocks *bb, sector_t sector,
 resource_size_t nd_namespace_blk_validate(struct nd_namespace_blk *nsblk);
 const u8 *nd_dev_to_uuid(struct device *dev);
 bool pmem_should_map_pages(struct device *dev);
+int nvdimm_security_unlock_dimm(struct device *dev);
+int nvdimm_security_get_state(struct device *dev);
 #endif /* __ND_H__ */
diff --git a/include/linux/libnvdimm.h b/include/linux/libnvdimm.h
index 3c1f18fdf76f..257ff2637ce1 100644
--- a/include/linux/libnvdimm.h
+++ b/include/linux/libnvdimm.h
@@ -155,9 +155,30 @@ static inline struct nd_blk_region_desc *to_blk_region_desc(
 
 }
 
+enum nvdimm_security_state {
+	NVDIMM_SECURITY_INVALID = 0,
+	NVDIMM_SECURITY_DISABLED,
+	NVDIMM_SECURITY_UNLOCKED,
+	NVDIMM_SECURITY_LOCKED,
+	NVDIMM_SECURITY_UNSUPPORTED,
+};
+
 #define NVDIMM_PASSPHRASE_LEN		32
 #define NVDIMM_KEY_DESC_LEN		22
 
+struct nvdimm_key_data {
+	u8 data[NVDIMM_PASSPHRASE_LEN];
+};
+
+struct nvdimm_security_ops {
+	int (*state)(struct nvdimm_bus *nvdimm_bus,
+			struct nvdimm *nvdimm,
+			enum nvdimm_security_state *state);
+	int (*unlock)(struct nvdimm_bus *nvdimm_bus,
+			struct nvdimm *nvdimm,
+			const struct nvdimm_key_data *nkey);
+};
+
 void badrange_init(struct badrange *badrange);
 int badrange_add(struct badrange *badrange, u64 addr, u64 length);
 void badrange_forget(struct badrange *badrange, phys_addr_t start,
@@ -181,7 +202,8 @@ void *nvdimm_provider_data(struct nvdimm *nvdimm);
 struct nvdimm *nvdimm_create(struct nvdimm_bus *nvdimm_bus, void *provider_data,
 		const struct attribute_group **groups, unsigned long flags,
 		unsigned long cmd_mask, int num_flush,
-		struct resource *flush_wpq, const char *dimm_id);
+		struct resource *flush_wpq, const char *dimm_id,
+		const struct nvdimm_security_ops *sec_ops);
 const struct nd_cmd_desc *nd_cmd_dimm_desc(int cmd);
 const struct nd_cmd_desc *nd_cmd_bus_desc(int cmd);
 u32 nd_cmd_in_size(struct nvdimm *nvdimm, int cmd,

_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm

^ permalink raw reply	[flat|nested] 30+ messages in thread

* [PATCH v8 06/12] nfit/libnvdimm: add set passphrase support for Intel nvdimms
  2018-08-28 22:47 [PATCH v8 00/12] Adding security support for nvdimm Dave Jiang
                   ` (4 preceding siblings ...)
  2018-08-28 22:47 ` [PATCH v8 05/12] nfit/libnvdimm: add unlock of nvdimm support for Intel DIMMs Dave Jiang
@ 2018-08-28 22:47 ` Dave Jiang
  2018-08-28 22:47 ` [PATCH v8 07/12] nfit/libnvdimm: add disable passphrase support to Intel nvdimm Dave Jiang
                   ` (10 subsequent siblings)
  16 siblings, 0 replies; 30+ messages in thread
From: Dave Jiang @ 2018-08-28 22:47 UTC (permalink / raw)
  To: linux-nvdimm; +Cc: alison.schofield, keescook, ebiggers3, dhowells, keyrings

Add support for setting and/or updating passphrase on the Intel nvdimms.
The passphrase is pulled from userspace through the kernel key management.
We trigger the update via writing "update" to the sysfs attribute
"security". The state of the security can also be read via the "security"
attribute. libnvdimm will generically support the key_change API call.

Signed-off-by: Dave Jiang <dave.jiang@intel.com>
---
 drivers/acpi/nfit/intel.c  |   68 ++++++++++++
 drivers/nvdimm/dimm_devs.c |  256 ++++++++++++++++++++++++++++++++++++++++++++
 include/linux/libnvdimm.h  |    5 +
 3 files changed, 329 insertions(+)

diff --git a/drivers/acpi/nfit/intel.c b/drivers/acpi/nfit/intel.c
index 4bfc1c1da339..314eae7e02d7 100644
--- a/drivers/acpi/nfit/intel.c
+++ b/drivers/acpi/nfit/intel.c
@@ -18,6 +18,73 @@
 #include "intel.h"
 #include "nfit.h"
 
+/*
+ * The update passphrase takes the old passphrase and the new passphrase
+ * and send those to the nvdimm. The nvdimm will verify the old
+ * passphrase and then update it with the new passphrase if pending
+ * verification. The function will pass in a zeroed passphrase field
+ * if the old passphrase is NULL. This typically happens when we are
+ * enabling security from the disabled state.
+ */
+static int intel_dimm_security_update_passphrase(
+		struct nvdimm_bus *nvdimm_bus, struct nvdimm *nvdimm,
+		const struct nvdimm_key_data *old_data,
+		const struct nvdimm_key_data *new_data)
+{
+	struct nvdimm_bus_descriptor *nd_desc = to_nd_desc(nvdimm_bus);
+	int cmd_rc, rc = 0;
+	struct nfit_mem *nfit_mem = nvdimm_provider_data(nvdimm);
+	struct {
+		struct nd_cmd_pkg pkg;
+		struct nd_intel_set_passphrase cmd;
+	} nd_cmd = {
+		.pkg = {
+			.nd_command = NVDIMM_INTEL_SET_PASSPHRASE,
+			.nd_family = NVDIMM_FAMILY_INTEL,
+			.nd_size_in = ND_INTEL_PASSPHRASE_SIZE * 2,
+			.nd_size_out = ND_INTEL_STATUS_SIZE,
+			.nd_fw_size = ND_INTEL_STATUS_SIZE,
+		},
+		.cmd = {
+			.status = 0,
+		},
+	};
+
+	if (!test_bit(NVDIMM_INTEL_SET_PASSPHRASE, &nfit_mem->dsm_mask))
+		return -ENOTTY;
+
+	if (old_data)
+		memcpy(nd_cmd.cmd.old_pass, old_data->data,
+				sizeof(nd_cmd.cmd.old_pass));
+	else
+		memset(nd_cmd.cmd.old_pass, 0, sizeof(nd_cmd.cmd.old_pass));
+	memcpy(nd_cmd.cmd.new_pass, new_data->data,
+			sizeof(nd_cmd.cmd.new_pass));
+	rc = nd_desc->ndctl(nd_desc, nvdimm, ND_CMD_CALL, &nd_cmd,
+			sizeof(nd_cmd), &cmd_rc);
+	if (rc < 0)
+		goto out;
+	if (cmd_rc < 0) {
+		rc = cmd_rc;
+		goto out;
+	}
+
+	switch (nd_cmd.cmd.status) {
+	case 0:
+		break;
+	case ND_INTEL_STATUS_INVALID_PASS:
+		rc = -EINVAL;
+		goto out;
+	case ND_INTEL_STATUS_INVALID_STATE:
+	default:
+		rc = -ENXIO;
+		goto out;
+	}
+
+ out:
+	return rc;
+}
+
 static int intel_dimm_security_unlock(struct nvdimm_bus *nvdimm_bus,
 		struct nvdimm *nvdimm, const struct nvdimm_key_data *nkey)
 {
@@ -149,4 +216,5 @@ static int intel_dimm_security_state(struct nvdimm_bus *nvdimm_bus,
 const struct nvdimm_security_ops intel_security_ops = {
 	.state = intel_dimm_security_state,
 	.unlock = intel_dimm_security_unlock,
+	.change_key = intel_dimm_security_update_passphrase,
 };
diff --git a/drivers/nvdimm/dimm_devs.c b/drivers/nvdimm/dimm_devs.c
index fe61f2a2ad5d..e7129998a5f3 100644
--- a/drivers/nvdimm/dimm_devs.c
+++ b/drivers/nvdimm/dimm_devs.c
@@ -20,6 +20,7 @@
 #include <linux/mm.h>
 #include <linux/key.h>
 #include <linux/init_task.h>
+#include <linux/key-type.h>
 #include <keys/user-type.h>
 #include "nd-core.h"
 #include "label.h"
@@ -51,6 +52,57 @@ static struct key *nvdimm_get_key(struct device *dev)
 	return nvdimm->key;
 }
 
+/*
+ * Replacing the user key with a kernel key
+ */
+static struct key *nvdimm_replace_key(struct key *key)
+{
+	struct key *new_key;
+	struct user_key_payload *payload;
+	int rc;
+
+	new_key = key_alloc(&key_type_logon, key->description,
+			GLOBAL_ROOT_UID, GLOBAL_ROOT_GID, &init_cred, 0,
+			KEY_ALLOC_NOT_IN_QUOTA, NULL);
+	if (IS_ERR(new_key))
+		return NULL;
+
+	down_read(&key->sem);
+	payload = key->payload.data[0];
+	rc = key_instantiate_and_link(new_key, payload->data, key->datalen,
+			nvdimm_keyring, NULL);
+	up_read(&key->sem);
+	if (rc < 0) {
+		key_revoke(new_key);
+		key_put(new_key);
+		return NULL;
+	}
+
+	key_invalidate(key);
+	key_put(key);
+
+	return new_key;
+}
+
+/*
+ * Retrieve user injected key
+ */
+static struct key *nvdimm_lookup_user_key(struct device *dev,
+		key_serial_t id)
+{
+	key_ref_t keyref;
+	struct key *key;
+
+	keyref = lookup_user_key(id, 0, 0);
+	if (IS_ERR(keyref))
+		return NULL;
+
+	key = key_ref_to_ptr(keyref);
+	dev_dbg(dev, "%s: key found: %d\n", __func__, key_serial(key));
+
+	return key;
+}
+
 /*
  * Retrieve kernel key for DIMM and request from user space if necessary.
  */
@@ -152,6 +204,96 @@ int nvdimm_security_unlock_dimm(struct device *dev)
 	return rc;
 }
 
+static int nvdimm_security_change_key(struct device *dev,
+		unsigned int old_keyid, unsigned int new_keyid)
+{
+	struct nvdimm *nvdimm = to_nvdimm(dev);
+	struct nvdimm_bus *nvdimm_bus = walk_to_nvdimm_bus(dev);
+	struct key *key = NULL, *old_key = NULL;
+	int rc;
+	void *old_data, *new_data;
+	bool update = false;
+	struct user_key_payload *payload;
+
+	if (!nvdimm->security_ops)
+		return 0;
+
+	if (nvdimm->state == NVDIMM_SECURITY_FROZEN)
+		return -EBUSY;
+
+	/* look for a key from keyring if exists and remove */
+	if (old_keyid != 0) {
+		old_key = nvdimm_get_key(dev);
+		if (old_key) {
+			if (key_serial(old_key) != old_keyid) {
+				dev_warn(dev, "incorrect key id: %u\n",
+						old_keyid);
+				return -EINVAL;
+			}
+			dev_dbg(dev, "%s: old key: %#x\n",
+					__func__, key_serial(old_key));
+			update = true;
+		}
+	}
+
+	/* request new key from userspace */
+	key = nvdimm_lookup_user_key(dev, new_keyid);
+	if (!key) {
+		dev_dbg(dev, "%s: failed to acquire new key\n", __func__);
+		rc = -ENXIO;
+		goto out;
+	}
+
+	dev_dbg(dev, "%s: new key: %#x\n", __func__, key_serial(key));
+
+	rc = nvdimm_check_key_len(key->datalen, update);
+	if (rc < 0)
+		goto out;
+
+	if (!update)
+		key = nvdimm_replace_key(key);
+
+	if (!key)
+		goto out;
+
+	payload = key->payload.data[0];
+	if (!update) {
+		old_data = NULL;
+		new_data = payload->data;
+	} else {
+		new_data = payload->data;
+		old_data = new_data + NVDIMM_PASSPHRASE_LEN;
+	}
+
+	down_read(&key->sem);
+	rc = nvdimm->security_ops->change_key(nvdimm_bus, nvdimm, old_data,
+			new_data);
+	/* copy new payload to old payload */
+	if (rc == 0) {
+		if (update)
+			key_update(make_key_ref(old_key, 1), new_data,
+					old_key->datalen);
+	} else
+		dev_warn(dev, "key update failed\n");
+	up_read(&key->sem);
+
+	if (update)
+		key_invalidate(key);
+	else {
+		key_link(nvdimm_keyring, key);
+		nvdimm->key = key;
+		key->perm |= KEY_USR_SEARCH;
+	}
+	nvdimm_security_get_state(dev);
+
+ out:
+	if (old_key)
+		key_put(old_key);
+	if (key)
+		key_put(key);
+	return rc;
+}
+
 /*
  * Retrieve bus and dimm handle and return if this bus supports
  * get_config_data commands
@@ -509,11 +651,125 @@ static ssize_t available_slots_show(struct device *dev,
 }
 static DEVICE_ATTR_RO(available_slots);
 
+static ssize_t security_show(struct device *dev,
+		struct device_attribute *attr, char *buf)
+{
+	struct nvdimm *nvdimm = to_nvdimm(dev);
+
+	switch (nvdimm->state) {
+	case NVDIMM_SECURITY_DISABLED:
+		return sprintf(buf, "disabled\n");
+	case NVDIMM_SECURITY_UNLOCKED:
+		return sprintf(buf, "unlocked\n");
+	case NVDIMM_SECURITY_LOCKED:
+		return sprintf(buf, "locked\n");
+	case NVDIMM_SECURITY_FROZEN:
+		return sprintf(buf, "frozen\n");
+	case NVDIMM_SECURITY_UNSUPPORTED:
+	default:
+		return sprintf(buf, "unsupported\n");
+	}
+
+	return -ENOTTY;
+}
+
+/*
+ * The update command:
+ *	update:<old_id>:<new_id>
+ */
+#define STR_UPDATE_LEN		6
+/* ":0:0" */
+#define NVDIMM_UPDATE_MIN_LEN	STR_UPDATE_LEN + 4
+
+static int __parse_update(const char *buf, size_t len, unsigned int *old_id,
+		unsigned int *new_id)
+{
+	int i, rc = 0;
+	const char *tmp, *old_idstr;
+	char *desc;
+	size_t buf_len;
+
+	if (len < NVDIMM_UPDATE_MIN_LEN)
+		return -EINVAL;
+
+	desc = kmemdup_nul(buf, STR_UPDATE_LEN, GFP_KERNEL);
+	if (!desc)
+		return -EINVAL;
+
+	if (strcmp(desc, "update") != 0) {
+		rc = -ENOMEM;
+		goto free_desc;
+	}
+
+	tmp = &buf[STR_UPDATE_LEN];
+	if (*tmp != ':') {
+		rc = -EINVAL;
+		goto free_desc;
+	}
+
+	tmp++;
+
+	buf_len = len - STR_UPDATE_LEN - 1;
+	for (i = 0; i < buf_len; i++) {
+		if (tmp[i] == ':')
+			break;
+	}
+
+	old_idstr = kmemdup_nul(tmp, i, GFP_KERNEL);
+	if (!old_idstr) {
+		rc = -ENOMEM;
+		goto free_desc;
+	}
+
+	rc = kstrtouint(old_idstr, 0, old_id);
+	if (rc < 0)
+		goto free_old;
+
+	tmp += i + 1;
+
+	rc = kstrtouint(tmp, 0, new_id);
+	if (rc < 0)
+		goto free_old;
+
+free_old:
+	kfree(old_idstr);
+free_desc:
+	kfree(desc);
+	return rc;
+}
+
+static ssize_t security_store(struct device *dev,
+		struct device_attribute *attr, const char *buf, size_t len)
+
+{
+	struct nvdimm *nvdimm = to_nvdimm(dev);
+	struct nvdimm_bus *nvdimm_bus = walk_to_nvdimm_bus(dev);
+	ssize_t rc = -EINVAL;
+	unsigned int new_id, old_id;
+
+        wait_nvdimm_bus_probe_idle(&nvdimm_bus->dev);
+        if (atomic_read(&nvdimm->busy))
+                return -EBUSY;
+
+	if (__parse_update(buf, len, &old_id, &new_id) == 0) {
+		dev_dbg(dev, "update: old: %u new: %u\n", old_id, new_id);
+		rc = nvdimm_security_change_key(dev, old_id, new_id);
+	} else
+		return -EINVAL;
+
+	if (rc == 0)
+		rc = len;
+
+	return rc;
+}
+static DEVICE_ATTR_RW(security);
+
 static struct attribute *nvdimm_attributes[] = {
 	&dev_attr_state.attr,
 	&dev_attr_flags.attr,
 	&dev_attr_commands.attr,
 	&dev_attr_available_slots.attr,
+	&dev_attr_security.attr,
 	NULL,
 };
 
diff --git a/include/linux/libnvdimm.h b/include/linux/libnvdimm.h
index 257ff2637ce1..bd6a413164ee 100644
--- a/include/linux/libnvdimm.h
+++ b/include/linux/libnvdimm.h
@@ -160,6 +160,7 @@ enum nvdimm_security_state {
 	NVDIMM_SECURITY_DISABLED,
 	NVDIMM_SECURITY_UNLOCKED,
 	NVDIMM_SECURITY_LOCKED,
+	NVDIMM_SECURITY_FROZEN,
 	NVDIMM_SECURITY_UNSUPPORTED,
 };
 
@@ -177,6 +178,10 @@ struct nvdimm_security_ops {
 	int (*unlock)(struct nvdimm_bus *nvdimm_bus,
 			struct nvdimm *nvdimm,
 			const struct nvdimm_key_data *nkey);
+	int (*change_key)(struct nvdimm_bus *nvdimm_bus,
+			struct nvdimm *nvdimm,
+			const struct nvdimm_key_data *old_data,
+			const struct nvdimm_key_data *new_data);
 };
 
 void badrange_init(struct badrange *badrange);

_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm

^ permalink raw reply	[flat|nested] 30+ messages in thread

* [PATCH v8 07/12] nfit/libnvdimm: add disable passphrase support to Intel nvdimm.
  2018-08-28 22:47 [PATCH v8 00/12] Adding security support for nvdimm Dave Jiang
                   ` (5 preceding siblings ...)
  2018-08-28 22:47 ` [PATCH v8 06/12] nfit/libnvdimm: add set passphrase support for Intel nvdimms Dave Jiang
@ 2018-08-28 22:47 ` Dave Jiang
  2018-08-28 22:48 ` [PATCH v8 08/12] nfit/libnvdimm: add freeze security " Dave Jiang
                   ` (9 subsequent siblings)
  16 siblings, 0 replies; 30+ messages in thread
From: Dave Jiang @ 2018-08-28 22:47 UTC (permalink / raw)
  To: linux-nvdimm; +Cc: alison.schofield, keescook, ebiggers3, dhowells, keyrings

Add support to disable passphrase (security) for the Intel nvdimm. The
passphrase used for disabling is pulled from userspace via the kernel
key management. The action is triggered by writing "disable" to the sysfs
attribute "security". libnvdimm will support the generic disable API call.

Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
---
 drivers/acpi/nfit/intel.c  |   53 ++++++++++++++++++++++++++++++++++++++++++++
 drivers/nvdimm/dimm_devs.c |   45 +++++++++++++++++++++++++++++++++++++
 include/linux/libnvdimm.h  |    3 ++
 3 files changed, 100 insertions(+), 1 deletion(-)

diff --git a/drivers/acpi/nfit/intel.c b/drivers/acpi/nfit/intel.c
index 314eae7e02d7..21d30222371f 100644
--- a/drivers/acpi/nfit/intel.c
+++ b/drivers/acpi/nfit/intel.c
@@ -18,6 +18,58 @@
 #include "intel.h"
 #include "nfit.h"
 
+static int intel_dimm_security_disable(struct nvdimm_bus *nvdimm_bus,
+		struct nvdimm *nvdimm, const struct nvdimm_key_data *nkey)
+{
+	struct nvdimm_bus_descriptor *nd_desc = to_nd_desc(nvdimm_bus);
+	int cmd_rc, rc = 0;
+	struct nfit_mem *nfit_mem = nvdimm_provider_data(nvdimm);
+	struct {
+		struct nd_cmd_pkg pkg;
+		struct nd_intel_disable_passphrase cmd;
+	} nd_cmd = {
+		.pkg = {
+			.nd_command = NVDIMM_INTEL_DISABLE_PASSPHRASE,
+			.nd_family = NVDIMM_FAMILY_INTEL,
+			.nd_size_in = ND_INTEL_PASSPHRASE_SIZE,
+			.nd_size_out = ND_INTEL_STATUS_SIZE,
+			.nd_fw_size = ND_INTEL_STATUS_SIZE,
+		},
+		.cmd = {
+			.status = 0,
+		},
+	};
+
+	if (!test_bit(NVDIMM_INTEL_DISABLE_PASSPHRASE, &nfit_mem->dsm_mask))
+		return -ENOTTY;
+
+	memcpy(nd_cmd.cmd.passphrase, nkey->data,
+			sizeof(nd_cmd.cmd.passphrase));
+	rc = nd_desc->ndctl(nd_desc, nvdimm, ND_CMD_CALL, &nd_cmd,
+			sizeof(nd_cmd), &cmd_rc);
+	if (rc < 0)
+		goto out;
+	if (cmd_rc < 0) {
+		rc = cmd_rc;
+		goto out;
+	}
+
+	switch (nd_cmd.cmd.status) {
+	case 0:
+		break;
+	case ND_INTEL_STATUS_INVALID_PASS:
+		rc = -EINVAL;
+		goto out;
+	case ND_INTEL_STATUS_INVALID_STATE:
+	default:
+		rc = -ENXIO;
+		goto out;
+	}
+
+ out:
+	return rc;
+}
+
 /*
  * The update passphrase takes the old passphrase and the new passphrase
  * and send those to the nvdimm. The nvdimm will verify the old
@@ -217,4 +269,5 @@ const struct nvdimm_security_ops intel_security_ops = {
 	.state = intel_dimm_security_state,
 	.unlock = intel_dimm_security_unlock,
 	.change_key = intel_dimm_security_update_passphrase,
+	.disable = intel_dimm_security_disable,
 };
diff --git a/drivers/nvdimm/dimm_devs.c b/drivers/nvdimm/dimm_devs.c
index e7129998a5f3..3bae8b4b613a 100644
--- a/drivers/nvdimm/dimm_devs.c
+++ b/drivers/nvdimm/dimm_devs.c
@@ -143,6 +143,47 @@ int nvdimm_security_get_state(struct device *dev)
 			&nvdimm->state);
 }
 
+static int nvdimm_security_disable(struct device *dev)
+{
+	struct nvdimm *nvdimm = to_nvdimm(dev);
+	struct nvdimm_bus *nvdimm_bus = walk_to_nvdimm_bus(dev);
+	struct key *key;
+	int rc;
+	struct user_key_payload *payload;
+
+	if (!nvdimm->security_ops)
+		return 0;
+
+	if (nvdimm->state == NVDIMM_SECURITY_UNSUPPORTED)
+		return 0;
+
+	key = nvdimm_get_key(dev);
+	if (!key) {
+		dev_warn(dev, "no key found\n");
+		return -ENOKEY;
+	}
+
+	down_read(&key->sem);
+	payload = key->payload.data[0];
+	rc = nvdimm->security_ops->disable(nvdimm_bus, nvdimm,
+			(void *)payload->data);
+	up_read(&key->sem);
+	if (rc < 0) {
+		dev_warn(dev, "unlock failed\n");
+		goto out;
+	}
+
+	/* If we succeed then remove the key */
+	key_unlink(nvdimm_keyring, key);
+	key_invalidate(key);
+	nvdimm->key = NULL;
+
+ out:
+	key_put(key);
+	nvdimm_security_get_state(dev);
+	return rc;
+}
+
 int nvdimm_security_unlock_dimm(struct device *dev)
 {
 	struct nvdimm *nvdimm = to_nvdimm(dev);
@@ -754,7 +795,9 @@ static ssize_t security_store(struct device *dev,
 	if (__parse_update(buf, len, &old_id, &new_id) == 0) {
 		dev_dbg(dev, "update: old: %u new: %u\n", old_id, new_id);
 		rc = nvdimm_security_change_key(dev, old_id, new_id);
-	} else
+	} else if (sysfs_streq(buf, "disable"))
+		rc = nvdimm_security_disable(dev);
+	else
 		return -EINVAL;
 
 	if (rc == 0)
diff --git a/include/linux/libnvdimm.h b/include/linux/libnvdimm.h
index bd6a413164ee..c60ab4b238f3 100644
--- a/include/linux/libnvdimm.h
+++ b/include/linux/libnvdimm.h
@@ -182,6 +182,9 @@ struct nvdimm_security_ops {
 			struct nvdimm *nvdimm,
 			const struct nvdimm_key_data *old_data,
 			const struct nvdimm_key_data *new_data);
+	int (*disable)(struct nvdimm_bus *nvdimm_bus,
+			struct nvdimm *nvdimm,
+			const struct nvdimm_key_data *nkey);
 };
 
 void badrange_init(struct badrange *badrange);

_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm

^ permalink raw reply	[flat|nested] 30+ messages in thread

* [PATCH v8 08/12] nfit/libnvdimm: add freeze security support to Intel nvdimm
  2018-08-28 22:47 [PATCH v8 00/12] Adding security support for nvdimm Dave Jiang
                   ` (6 preceding siblings ...)
  2018-08-28 22:47 ` [PATCH v8 07/12] nfit/libnvdimm: add disable passphrase support to Intel nvdimm Dave Jiang
@ 2018-08-28 22:48 ` Dave Jiang
  2018-08-28 22:48 ` [PATCH v8 09/12] nfit/libnvdimm: add support for issue secure erase DSM " Dave Jiang
                   ` (8 subsequent siblings)
  16 siblings, 0 replies; 30+ messages in thread
From: Dave Jiang @ 2018-08-28 22:48 UTC (permalink / raw)
  To: linux-nvdimm; +Cc: alison.schofield, keescook, ebiggers3, dhowells, keyrings

Add support for freeze security on Intel nvdimm. This locks out any
changes to security for the DIMM unless a reboot is done. This is triggered
by writing "freeze" to the "security" sysfs attribute. libnvdimm will
support the generic freeze_lock API call.

Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
---
 drivers/acpi/nfit/intel.c  |   51 ++++++++++++++++++++++++++++++++++++++++++++
 drivers/nvdimm/dimm_devs.c |   22 +++++++++++++++++++
 include/linux/libnvdimm.h  |    2 ++
 3 files changed, 75 insertions(+)

diff --git a/drivers/acpi/nfit/intel.c b/drivers/acpi/nfit/intel.c
index 21d30222371f..ba886f1f5399 100644
--- a/drivers/acpi/nfit/intel.c
+++ b/drivers/acpi/nfit/intel.c
@@ -18,6 +18,53 @@
 #include "intel.h"
 #include "nfit.h"
 
+static int intel_dimm_security_freeze_lock(struct nvdimm_bus *nvdimm_bus,
+		struct nvdimm *nvdimm)
+{
+	struct nvdimm_bus_descriptor *nd_desc = to_nd_desc(nvdimm_bus);
+	int cmd_rc, rc = 0;
+	struct nfit_mem *nfit_mem = nvdimm_provider_data(nvdimm);
+	struct {
+		struct nd_cmd_pkg pkg;
+		struct nd_intel_freeze_lock cmd;
+	} nd_cmd = {
+		.pkg = {
+			.nd_command = NVDIMM_INTEL_FREEZE_LOCK,
+			.nd_family = NVDIMM_FAMILY_INTEL,
+			.nd_size_in = 0,
+			.nd_size_out = ND_INTEL_STATUS_SIZE,
+			.nd_fw_size = ND_INTEL_STATUS_SIZE,
+		},
+		.cmd = {
+			.status = 0,
+		},
+	};
+
+	if (!test_bit(NVDIMM_INTEL_FREEZE_LOCK, &nfit_mem->dsm_mask))
+		return -ENOTTY;
+
+	rc = nd_desc->ndctl(nd_desc, nvdimm, ND_CMD_CALL, &nd_cmd,
+			sizeof(nd_cmd), &cmd_rc);
+	if (rc < 0)
+		goto out;
+	if (cmd_rc < 0) {
+		rc = cmd_rc;
+		goto out;
+	}
+
+	switch (nd_cmd.cmd.status) {
+	case 0:
+		break;
+	case ND_INTEL_STATUS_INVALID_STATE:
+	default:
+		rc = -ENXIO;
+		goto out;
+	}
+
+ out:
+	return rc;
+}
+
 static int intel_dimm_security_disable(struct nvdimm_bus *nvdimm_bus,
 		struct nvdimm *nvdimm, const struct nvdimm_key_data *nkey)
 {
@@ -254,6 +301,9 @@ static int intel_dimm_security_state(struct nvdimm_bus *nvdimm_bus,
 	else if (nd_cmd.cmd.state & ND_INTEL_SEC_STATE_ENABLED) {
 		if (nd_cmd.cmd.state & ND_INTEL_SEC_STATE_LOCKED)
 			*state = NVDIMM_SECURITY_LOCKED;
+		else if (nd_cmd.cmd.state & ND_INTEL_SEC_STATE_FROZEN ||
+				nd_cmd.cmd.state & ND_INTEL_SEC_STATE_PLIMIT)
+			*state = NVDIMM_SECURITY_FROZEN;
 		else
 			*state = NVDIMM_SECURITY_UNLOCKED;
 	} else
@@ -270,4 +320,5 @@ const struct nvdimm_security_ops intel_security_ops = {
 	.unlock = intel_dimm_security_unlock,
 	.change_key = intel_dimm_security_update_passphrase,
 	.disable = intel_dimm_security_disable,
+	.freeze_lock = intel_dimm_security_freeze_lock,
 };
diff --git a/drivers/nvdimm/dimm_devs.c b/drivers/nvdimm/dimm_devs.c
index 3bae8b4b613a..338c3fedfdf3 100644
--- a/drivers/nvdimm/dimm_devs.c
+++ b/drivers/nvdimm/dimm_devs.c
@@ -143,6 +143,26 @@ int nvdimm_security_get_state(struct device *dev)
 			&nvdimm->state);
 }
 
+static int nvdimm_security_freeze_lock(struct device *dev)
+{
+	struct nvdimm *nvdimm = to_nvdimm(dev);
+	struct nvdimm_bus *nvdimm_bus = walk_to_nvdimm_bus(dev);
+	int rc;
+
+	if (!nvdimm->security_ops)
+		return 0;
+
+	if (nvdimm->state == NVDIMM_SECURITY_UNSUPPORTED)
+		return 0;
+
+	rc = nvdimm->security_ops->freeze_lock(nvdimm_bus, nvdimm);
+	if (rc < 0)
+		return rc;
+
+	nvdimm_security_get_state(dev);
+	return 0;
+}
+
 static int nvdimm_security_disable(struct device *dev)
 {
 	struct nvdimm *nvdimm = to_nvdimm(dev);
@@ -797,6 +817,8 @@ static ssize_t security_store(struct device *dev,
 		rc = nvdimm_security_change_key(dev, old_id, new_id);
 	} else if (sysfs_streq(buf, "disable"))
 		rc = nvdimm_security_disable(dev);
+	else if (sysfs_streq(buf, "freeze"))
+		rc = nvdimm_security_freeze_lock(dev);
 	else
 		return -EINVAL;
 
diff --git a/include/linux/libnvdimm.h b/include/linux/libnvdimm.h
index c60ab4b238f3..bcab42caa948 100644
--- a/include/linux/libnvdimm.h
+++ b/include/linux/libnvdimm.h
@@ -185,6 +185,8 @@ struct nvdimm_security_ops {
 	int (*disable)(struct nvdimm_bus *nvdimm_bus,
 			struct nvdimm *nvdimm,
 			const struct nvdimm_key_data *nkey);
+	int (*freeze_lock)(struct nvdimm_bus *nvdimm_bus,
+			struct nvdimm *nvdimm);
 };
 
 void badrange_init(struct badrange *badrange);

_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm

^ permalink raw reply	[flat|nested] 30+ messages in thread

* [PATCH v8 09/12] nfit/libnvdimm: add support for issue secure erase DSM to Intel nvdimm
  2018-08-28 22:47 [PATCH v8 00/12] Adding security support for nvdimm Dave Jiang
                   ` (7 preceding siblings ...)
  2018-08-28 22:48 ` [PATCH v8 08/12] nfit/libnvdimm: add freeze security " Dave Jiang
@ 2018-08-28 22:48 ` Dave Jiang
  2018-08-28 22:48 ` [PATCH v8 10/12] nfit_test: add context to dimm_dev for nfit_test Dave Jiang
                   ` (7 subsequent siblings)
  16 siblings, 0 replies; 30+ messages in thread
From: Dave Jiang @ 2018-08-28 22:48 UTC (permalink / raw)
  To: linux-nvdimm; +Cc: alison.schofield, keescook, ebiggers3, dhowells, keyrings

Add support to issue a secure erase DSM to the Intel nvdimm. The
required passphrase is acquired from userspace through the kernel key
management. To trigger the action, "erase" is written to the "security"
sysfs attribute. libnvdimm will support the erase generic API call.

Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
---
 drivers/acpi/nfit/intel.c  |   58 ++++++++++++++++++++++++++++++++++++++++++++
 drivers/nvdimm/dimm_devs.c |   49 +++++++++++++++++++++++++++++++++++++
 include/linux/libnvdimm.h  |    3 ++
 3 files changed, 110 insertions(+)

diff --git a/drivers/acpi/nfit/intel.c b/drivers/acpi/nfit/intel.c
index ba886f1f5399..419a7d54d4e8 100644
--- a/drivers/acpi/nfit/intel.c
+++ b/drivers/acpi/nfit/intel.c
@@ -18,6 +18,63 @@
 #include "intel.h"
 #include "nfit.h"
 
+static int intel_dimm_security_erase(struct nvdimm_bus *nvdimm_bus,
+		struct nvdimm *nvdimm, const struct nvdimm_key_data *nkey)
+{
+	struct nvdimm_bus_descriptor *nd_desc = to_nd_desc(nvdimm_bus);
+	int cmd_rc, rc = 0;
+	struct nfit_mem *nfit_mem = nvdimm_provider_data(nvdimm);
+	struct {
+		struct nd_cmd_pkg pkg;
+		struct nd_intel_secure_erase cmd;
+	} nd_cmd = {
+		.pkg = {
+			.nd_command = NVDIMM_INTEL_SECURE_ERASE,
+			.nd_family = NVDIMM_FAMILY_INTEL,
+			.nd_size_in = ND_INTEL_PASSPHRASE_SIZE,
+			.nd_size_out = ND_INTEL_STATUS_SIZE,
+			.nd_fw_size = ND_INTEL_STATUS_SIZE,
+		},
+		.cmd = {
+			.status = 0,
+		},
+	};
+
+	if (!test_bit(NVDIMM_INTEL_SECURE_ERASE, &nfit_mem->dsm_mask))
+		return -ENOTTY;
+
+	/* flush all cache before we erase DIMM */
+	wbinvd_on_all_cpus();
+	memcpy(nd_cmd.cmd.passphrase, nkey->data,
+			sizeof(nd_cmd.cmd.passphrase));
+	rc = nd_desc->ndctl(nd_desc, nvdimm, ND_CMD_CALL, &nd_cmd,
+			sizeof(nd_cmd), &cmd_rc);
+	if (rc < 0)
+		goto out;
+	if (cmd_rc < 0) {
+		rc = cmd_rc;
+		goto out;
+	}
+
+	switch (nd_cmd.cmd.status) {
+	case 0:
+		break;
+	case ND_INTEL_STATUS_INVALID_PASS:
+		rc = -EINVAL;
+		goto out;
+	case ND_INTEL_STATUS_INVALID_STATE:
+	default:
+		rc = -ENXIO;
+		goto out;
+	}
+
+	/* DIMM erased, invalidate all CPU caches before we read it */
+	wbinvd_on_all_cpus();
+
+ out:
+	return rc;
+}
+
 static int intel_dimm_security_freeze_lock(struct nvdimm_bus *nvdimm_bus,
 		struct nvdimm *nvdimm)
 {
@@ -321,4 +378,5 @@ const struct nvdimm_security_ops intel_security_ops = {
 	.change_key = intel_dimm_security_update_passphrase,
 	.disable = intel_dimm_security_disable,
 	.freeze_lock = intel_dimm_security_freeze_lock,
+	.erase = intel_dimm_security_erase,
 };
diff --git a/drivers/nvdimm/dimm_devs.c b/drivers/nvdimm/dimm_devs.c
index 338c3fedfdf3..102accb68893 100644
--- a/drivers/nvdimm/dimm_devs.c
+++ b/drivers/nvdimm/dimm_devs.c
@@ -143,6 +143,53 @@ int nvdimm_security_get_state(struct device *dev)
 			&nvdimm->state);
 }
 
+static int nvdimm_security_erase(struct device *dev)
+{
+	struct nvdimm *nvdimm = to_nvdimm(dev);
+	struct nvdimm_bus *nvdimm_bus = walk_to_nvdimm_bus(dev);
+	struct key *key;
+	struct user_key_payload *payload;
+	int rc = 0;
+
+	if (!nvdimm->security_ops)
+		return 0;
+
+	/* lock the device and disallow driver bind */
+	device_lock(dev);
+	/* No driver data means dimm is disabled. Proceed if so. */
+	if (dev_get_drvdata(dev)) {
+		dev_warn(dev, "Unable to secure erase while DIMM active.\n");
+		rc = -EINVAL;
+		goto out;
+	}
+
+	if (nvdimm->state == NVDIMM_SECURITY_UNSUPPORTED)
+		goto out;
+
+	key = nvdimm_get_key(dev);
+	if (!key) {
+		dev_warn(dev, "no key found\n");
+		rc = -ENOKEY;
+		goto out;
+	}
+
+	down_read(&key->sem);
+	payload = key->payload.data[0];
+	rc = nvdimm->security_ops->erase(nvdimm_bus, nvdimm,
+			(void *)payload->data);
+	up_read(&key->sem);
+	/* remove key since secure erase kills the passphrase */
+	key_unlink(nvdimm_keyring, key);
+	key_invalidate(key);
+	nvdimm->key = NULL;
+	key_put(key);
+
+ out:
+	device_unlock(dev);
+	nvdimm_security_get_state(dev);
+	return rc;
+}
+
 static int nvdimm_security_freeze_lock(struct device *dev)
 {
 	struct nvdimm *nvdimm = to_nvdimm(dev);
@@ -819,6 +866,8 @@ static ssize_t security_store(struct device *dev,
 		rc = nvdimm_security_disable(dev);
 	else if (sysfs_streq(buf, "freeze"))
 		rc = nvdimm_security_freeze_lock(dev);
+	else if (sysfs_streq(buf, "erase"))
+		rc = nvdimm_security_erase(dev);
 	else
 		return -EINVAL;
 
diff --git a/include/linux/libnvdimm.h b/include/linux/libnvdimm.h
index bcab42caa948..0d85e092a6dd 100644
--- a/include/linux/libnvdimm.h
+++ b/include/linux/libnvdimm.h
@@ -187,6 +187,9 @@ struct nvdimm_security_ops {
 			const struct nvdimm_key_data *nkey);
 	int (*freeze_lock)(struct nvdimm_bus *nvdimm_bus,
 			struct nvdimm *nvdimm);
+	int (*erase)(struct nvdimm_bus *nvdimm_bus,
+			struct nvdimm *nvdimm,
+			const struct nvdimm_key_data *nkey);
 };
 
 void badrange_init(struct badrange *badrange);

_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm

^ permalink raw reply	[flat|nested] 30+ messages in thread

* [PATCH v8 10/12] nfit_test: add context to dimm_dev for nfit_test
  2018-08-28 22:47 [PATCH v8 00/12] Adding security support for nvdimm Dave Jiang
                   ` (8 preceding siblings ...)
  2018-08-28 22:48 ` [PATCH v8 09/12] nfit/libnvdimm: add support for issue secure erase DSM " Dave Jiang
@ 2018-08-28 22:48 ` Dave Jiang
  2018-08-28 22:48 ` [PATCH v8 11/12] nfit_test: add test support for Intel nvdimm security DSMs Dave Jiang
                   ` (6 subsequent siblings)
  16 siblings, 0 replies; 30+ messages in thread
From: Dave Jiang @ 2018-08-28 22:48 UTC (permalink / raw)
  To: linux-nvdimm; +Cc: alison.schofield, keescook, ebiggers3, dhowells, keyrings

In order to access the nfit_test context via sideband sysfs knobs, the
dimm_dev needs to be more than struct device in order to point back to
struct nfit_test. Wrapping the original struct device with a struct
nfit_dimm_dev and saving the nfit_test as private driver data. Also
changing the nfit_mem to be a member of struct nfit_dimm_dev instead of
saving as private driver data of that device. This is in preparation for
adding security DSM support and allowing the locking of DIMMs for testing
via sideband.

Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
---
 tools/testing/nvdimm/test/nfit.c |   41 +++++++++++++++++++++++++-------------
 1 file changed, 27 insertions(+), 14 deletions(-)

diff --git a/tools/testing/nvdimm/test/nfit.c b/tools/testing/nvdimm/test/nfit.c
index 18157b0c0d0d..7b3ce2194889 100644
--- a/tools/testing/nvdimm/test/nfit.c
+++ b/tools/testing/nvdimm/test/nfit.c
@@ -150,6 +150,11 @@ struct nfit_test_fw {
 	u64 end_time;
 };
 
+struct nfit_dimm_dev {
+	struct device dev;
+	struct nfit_mem *nfit_mem;
+};
+
 struct nfit_test {
 	struct acpi_nfit_desc acpi_desc;
 	struct platform_device pdev;
@@ -181,7 +186,7 @@ struct nfit_test {
 		unsigned long deadline;
 		spinlock_t lock;
 	} ars_state;
-	struct device *dimm_dev[NUM_DCR];
+	struct nfit_dimm_dev dimm_dev[NUM_DCR];
 	struct nd_intel_smart *smart;
 	struct nd_intel_smart_threshold *smart_threshold;
 	struct badrange badrange;
@@ -981,7 +986,8 @@ static int nfit_test_ctl(struct nvdimm_bus_descriptor *nd_desc,
 						&t->smart_threshold[i -
 							t->dcr_idx],
 						&t->smart[i - t->dcr_idx],
-						&t->pdev.dev, t->dimm_dev[i]);
+						&t->pdev.dev,
+						&t->dimm_dev[i].dev);
 				break;
 			case ND_INTEL_SMART_INJECT:
 				rc = nfit_test_cmd_smart_inject(buf,
@@ -989,7 +995,8 @@ static int nfit_test_ctl(struct nvdimm_bus_descriptor *nd_desc,
 						&t->smart_threshold[i -
 							t->dcr_idx],
 						&t->smart[i - t->dcr_idx],
-						&t->pdev.dev, t->dimm_dev[i]);
+						&t->pdev.dev,
+						&t->dimm_dev[i].dev);
 				break;
 			default:
 				return -ENOTTY;
@@ -1187,8 +1194,7 @@ static void put_dimms(void *data)
 	int i;
 
 	for (i = 0; i < t->num_dcr; i++)
-		if (t->dimm_dev[i])
-			device_unregister(t->dimm_dev[i]);
+		device_unregister(&t->dimm_dev[i].dev);
 }
 
 static struct class *nfit_test_dimm;
@@ -1290,19 +1296,27 @@ static const struct attribute_group *nfit_test_dimm_attribute_groups[] = {
 	NULL,
 };
 
+static void dimm_dev_release(struct device *dev)
+{
+}
+
 static int nfit_test_dimm_init(struct nfit_test *t)
 {
-	int i;
+	int i, rc;
 
 	if (devm_add_action_or_reset(&t->pdev.dev, put_dimms, t))
 		return -ENOMEM;
 	for (i = 0; i < t->num_dcr; i++) {
-		t->dimm_dev[i] = device_create_with_groups(nfit_test_dimm,
-				&t->pdev.dev, 0, NULL,
-				nfit_test_dimm_attribute_groups,
-				"test_dimm%d", i + t->dcr_idx);
-		if (!t->dimm_dev[i])
-			return -ENOMEM;
+		t->dimm_dev[i].dev.parent = &t->pdev.dev;
+		dev_set_name(&t->dimm_dev[i].dev, "test_dimm%d",
+				i + t->dcr_idx);
+		t->dimm_dev[i].dev.class = nfit_test_dimm;
+		t->dimm_dev[i].dev.groups = nfit_test_dimm_attribute_groups;
+		t->dimm_dev[i].dev.release = dimm_dev_release;
+		rc = device_register(&t->dimm_dev[i].dev);
+		if (rc < 0)
+			return rc;
+		dev_set_drvdata(&t->dimm_dev[i].dev, t);
 	}
 	return 0;
 }
@@ -2665,8 +2679,7 @@ static int nfit_test_probe(struct platform_device *pdev)
 
 		for (i = 0; i < NUM_DCR; i++)
 			if (nfit_handle == handle[i])
-				dev_set_drvdata(nfit_test->dimm_dev[i],
-						nfit_mem);
+				nfit_test->dimm_dev[i].nfit_mem = nfit_mem;
 	}
 	mutex_unlock(&acpi_desc->init_mutex);
 

_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm

^ permalink raw reply	[flat|nested] 30+ messages in thread

* [PATCH v8 11/12] nfit_test: add test support for Intel nvdimm security DSMs
  2018-08-28 22:47 [PATCH v8 00/12] Adding security support for nvdimm Dave Jiang
                   ` (9 preceding siblings ...)
  2018-08-28 22:48 ` [PATCH v8 10/12] nfit_test: add context to dimm_dev for nfit_test Dave Jiang
@ 2018-08-28 22:48 ` Dave Jiang
  2018-08-28 22:48 ` [PATCH v8 12/12] libnvdimm: add documentation for nvdimm security support Dave Jiang
                   ` (5 subsequent siblings)
  16 siblings, 0 replies; 30+ messages in thread
From: Dave Jiang @ 2018-08-28 22:48 UTC (permalink / raw)
  To: linux-nvdimm; +Cc: alison.schofield, keescook, ebiggers3, dhowells, keyrings

Add nfit_test support for DSM functions "Get Security State",
"Set Passphrase", "Disable Passphrase", "Unlock Unit", "Freeze Lock",
and "Secure Erase" for the fake DIMMs.

Also adding a sysfs knob in order to put the DIMMs in "locked" state. The
order of testing DIMM unlocking would be.
1a. Disable DIMM X.
1b. Set Passphrase to DIMM X.
2. Write to
/sys/devices/platform/nfit_test.0/nfit_test_dimm/test_dimmX/lock_dimm
3. Renable DIMM X
4. Check DIMM X state via sysfs "security" attribute for nmemX.

Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
---
 tools/testing/nvdimm/Kbuild      |    1 
 tools/testing/nvdimm/test/nfit.c |  186 ++++++++++++++++++++++++++++++++++++++
 2 files changed, 187 insertions(+)

diff --git a/tools/testing/nvdimm/Kbuild b/tools/testing/nvdimm/Kbuild
index 0392153a0009..a13670d3b389 100644
--- a/tools/testing/nvdimm/Kbuild
+++ b/tools/testing/nvdimm/Kbuild
@@ -36,6 +36,7 @@ obj-$(CONFIG_DEV_DAX) += device_dax.o
 obj-$(CONFIG_DEV_DAX_PMEM) += dax_pmem.o
 
 nfit-y := $(ACPI_SRC)/core.o
+nfit-$(CONFIG_X86) += $(ACPI_SRC)/intel.o
 nfit-$(CONFIG_X86_MCE) += $(ACPI_SRC)/mce.o
 nfit-y += acpi_nfit_test.o
 nfit-y += config_check.o
diff --git a/tools/testing/nvdimm/test/nfit.c b/tools/testing/nvdimm/test/nfit.c
index 7b3ce2194889..e5b62396c41f 100644
--- a/tools/testing/nvdimm/test/nfit.c
+++ b/tools/testing/nvdimm/test/nfit.c
@@ -24,6 +24,7 @@
 #include <linux/list.h>
 #include <linux/slab.h>
 #include <nd-core.h>
+#include <intel.h>
 #include <nfit.h>
 #include <nd.h>
 #include "nfit_test.h"
@@ -155,6 +156,11 @@ struct nfit_dimm_dev {
 	struct nfit_mem *nfit_mem;
 };
 
+struct nfit_test_sec {
+	u8 state;
+	u8 passphrase[32];
+};
+
 struct nfit_test {
 	struct acpi_nfit_desc acpi_desc;
 	struct platform_device pdev;
@@ -192,6 +198,7 @@ struct nfit_test {
 	struct badrange badrange;
 	struct work_struct work;
 	struct nfit_test_fw *fw;
+	struct nfit_test_sec *sec;
 };
 
 static struct workqueue_struct *nfit_wq;
@@ -899,6 +906,138 @@ static int override_return_code(int dimm, unsigned int func, int rc)
 	return rc;
 }
 
+static int nd_intel_test_cmd_security_status(struct nfit_test *t,
+		struct nd_intel_get_security_state *nd_cmd,
+		unsigned int buf_len, int idx)
+{
+	struct device *dev = &t->pdev.dev;
+	struct nfit_test_sec *sec = &t->sec[idx];
+
+	nd_cmd->status = 0;
+	nd_cmd->state = sec->state;
+	dev_dbg(dev, "security state (%#x) returned\n", nd_cmd->state);
+
+	return 0;
+}
+
+static int nd_intel_test_cmd_unlock_unit(struct nfit_test *t,
+		struct nd_intel_unlock_unit *nd_cmd,
+		unsigned int buf_len, int idx)
+{
+	struct device *dev = &t->pdev.dev;
+	struct nfit_test_sec *sec = &t->sec[idx];
+
+	if (!(sec->state & ND_INTEL_SEC_STATE_LOCKED) ||
+			(sec->state & ND_INTEL_SEC_STATE_FROZEN)) {
+		nd_cmd->status = ND_INTEL_STATUS_INVALID_STATE;
+		dev_dbg(dev, "unlock unit: invalid state: %#x\n",
+				sec->state);
+	} else if (memcmp(nd_cmd->passphrase, sec->passphrase,
+				ND_INTEL_PASSPHRASE_SIZE) != 0) {
+		nd_cmd->status = ND_INTEL_STATUS_INVALID_PASS;
+		dev_dbg(dev, "unlock unit: invalid passphrase\n");
+	} else {
+		nd_cmd->status = 0;
+		sec->state = ND_INTEL_SEC_STATE_ENABLED;
+		dev_dbg(dev, "Unit unlocked\n");
+	}
+
+	dev_dbg(dev, "unlocking status returned: %#x\n", nd_cmd->status);
+	return 0;
+}
+
+static int nd_intel_test_cmd_set_pass(struct nfit_test *t,
+		struct nd_intel_set_passphrase *nd_cmd,
+		unsigned int buf_len, int idx)
+{
+	struct device *dev = &t->pdev.dev;
+	struct nfit_test_sec *sec = &t->sec[idx];
+
+	if (sec->state & ND_INTEL_SEC_STATE_FROZEN) {
+		nd_cmd->status = ND_INTEL_STATUS_INVALID_STATE;
+		dev_dbg(dev, "set passphrase: wrong security state\n");
+	} else if (memcmp(nd_cmd->old_pass, sec->passphrase,
+				ND_INTEL_PASSPHRASE_SIZE) != 0) {
+		nd_cmd->status = ND_INTEL_STATUS_INVALID_PASS;
+		dev_dbg(dev, "set passphrase: wrong passphrase\n");
+	} else {
+		memcpy(sec->passphrase, nd_cmd->new_pass,
+				ND_INTEL_PASSPHRASE_SIZE);
+		sec->state |= ND_INTEL_SEC_STATE_ENABLED;
+		nd_cmd->status = 0;
+		dev_dbg(dev, "passphrase updated\n");
+	}
+
+	return 0;
+}
+
+static int nd_intel_test_cmd_freeze_lock(struct nfit_test *t,
+		struct nd_intel_freeze_lock *nd_cmd,
+		unsigned int buf_len, int idx)
+{
+	struct device *dev = &t->pdev.dev;
+	struct nfit_test_sec *sec = &t->sec[idx];
+
+	if (!(sec->state & ND_INTEL_SEC_STATE_ENABLED)) {
+		nd_cmd->status = ND_INTEL_STATUS_INVALID_STATE;
+		dev_dbg(dev, "freeze lock: wrong security state\n");
+	} else {
+		sec->state |= ND_INTEL_SEC_STATE_FROZEN;
+		nd_cmd->status = 0;
+		dev_dbg(dev, "security frozen\n");
+	}
+
+	return 0;
+}
+
+static int nd_intel_test_cmd_disable_pass(struct nfit_test *t,
+		struct nd_intel_disable_passphrase *nd_cmd,
+		unsigned int buf_len, int idx)
+{
+	struct device *dev = &t->pdev.dev;
+	struct nfit_test_sec *sec = &t->sec[idx];
+
+	if (!(sec->state & ND_INTEL_SEC_STATE_ENABLED) ||
+			(sec->state & ND_INTEL_SEC_STATE_FROZEN)) {
+		nd_cmd->status = ND_INTEL_STATUS_INVALID_STATE;
+		dev_dbg(dev, "disable passphrase: wrong security state\n");
+	} else if (memcmp(nd_cmd->passphrase, sec->passphrase,
+				ND_INTEL_PASSPHRASE_SIZE) != 0) {
+		nd_cmd->status = ND_INTEL_STATUS_INVALID_PASS;
+		dev_dbg(dev, "disable passphrase: wrong passphrase\n");
+	} else {
+		memset(sec->passphrase, 0, ND_INTEL_PASSPHRASE_SIZE);
+		sec->state = 0;
+		dev_dbg(dev, "disable passphrase: done\n");
+	}
+
+	return 0;
+}
+
+static int nd_intel_test_cmd_secure_erase(struct nfit_test *t,
+		struct nd_intel_secure_erase *nd_cmd,
+		unsigned int buf_len, int idx)
+{
+	struct device *dev = &t->pdev.dev;
+	struct nfit_test_sec *sec = &t->sec[idx];
+
+	if (!(sec->state & ND_INTEL_SEC_STATE_ENABLED) ||
+			(sec->state & ND_INTEL_SEC_STATE_FROZEN)) {
+		nd_cmd->status = ND_INTEL_STATUS_INVALID_STATE;
+		dev_dbg(dev, "secure erase: wrong security state\n");
+	} else if (memcmp(nd_cmd->passphrase, sec->passphrase,
+				ND_INTEL_PASSPHRASE_SIZE) != 0) {
+		nd_cmd->status = ND_INTEL_STATUS_INVALID_PASS;
+		dev_dbg(dev, "secure erase: wrong passphrase\n");
+	} else {
+		memset(sec->passphrase, 0, ND_INTEL_PASSPHRASE_SIZE);
+		sec->state = 0;
+		dev_dbg(dev, "secure erase: done\n");
+	}
+
+	return 0;
+}
+
 static int get_dimm(struct nfit_mem *nfit_mem, unsigned int func)
 {
 	int i;
@@ -946,6 +1085,30 @@ static int nfit_test_ctl(struct nvdimm_bus_descriptor *nd_desc,
 				return i;
 
 			switch (func) {
+			case NVDIMM_INTEL_GET_SECURITY_STATE:
+				rc = nd_intel_test_cmd_security_status(t,
+						buf, buf_len, i - t->dcr_idx);
+				break;
+			case NVDIMM_INTEL_UNLOCK_UNIT:
+				rc = nd_intel_test_cmd_unlock_unit(t,
+						buf, buf_len, i - t->dcr_idx);
+				break;
+			case NVDIMM_INTEL_SET_PASSPHRASE:
+				rc = nd_intel_test_cmd_set_pass(t,
+						buf, buf_len, i - t->dcr_idx);
+				break;
+			case NVDIMM_INTEL_DISABLE_PASSPHRASE:
+				rc = nd_intel_test_cmd_disable_pass(t,
+						buf, buf_len, i - t->dcr_idx);
+				break;
+			case NVDIMM_INTEL_FREEZE_LOCK:
+				rc = nd_intel_test_cmd_freeze_lock(t,
+						buf, buf_len, i - t->dcr_idx);
+				break;
+			case NVDIMM_INTEL_SECURE_ERASE:
+				rc = nd_intel_test_cmd_secure_erase(t,
+						buf, buf_len, i - t->dcr_idx);
+				break;
 			case ND_INTEL_ENABLE_LSS_STATUS:
 				rc = nd_intel_test_cmd_set_lss_status(t,
 						buf, buf_len);
@@ -1280,10 +1443,23 @@ static ssize_t fail_cmd_code_store(struct device *dev, struct device_attribute *
 }
 static DEVICE_ATTR_RW(fail_cmd_code);
 
+static ssize_t lock_dimm_store(struct device *dev,
+		struct device_attribute *attr, const char *buf, size_t size)
+{
+	int dimm = dimm_name_to_id(dev);
+	struct nfit_test *t = dev_get_drvdata(dev);
+	struct nfit_test_sec *sec = &t->sec[dimm];
+
+	sec->state = ND_INTEL_SEC_STATE_ENABLED | ND_INTEL_SEC_STATE_LOCKED;
+	return size;
+}
+static DEVICE_ATTR_WO(lock_dimm);
+
 static struct attribute *nfit_test_dimm_attributes[] = {
 	&dev_attr_fail_cmd.attr,
 	&dev_attr_fail_cmd_code.attr,
 	&dev_attr_handle.attr,
+	&dev_attr_lock_dimm.attr,
 	NULL,
 };
 
@@ -2192,6 +2368,14 @@ static void nfit_test0_setup(struct nfit_test *t)
 	set_bit(ND_INTEL_FW_FINISH_UPDATE, &acpi_desc->dimm_cmd_force_en);
 	set_bit(ND_INTEL_FW_FINISH_QUERY, &acpi_desc->dimm_cmd_force_en);
 	set_bit(ND_INTEL_ENABLE_LSS_STATUS, &acpi_desc->dimm_cmd_force_en);
+	set_bit(NVDIMM_INTEL_GET_SECURITY_STATE,
+			&acpi_desc->dimm_cmd_force_en);
+	set_bit(NVDIMM_INTEL_SET_PASSPHRASE, &acpi_desc->dimm_cmd_force_en);
+	set_bit(NVDIMM_INTEL_DISABLE_PASSPHRASE,
+			&acpi_desc->dimm_cmd_force_en);
+	set_bit(NVDIMM_INTEL_UNLOCK_UNIT, &acpi_desc->dimm_cmd_force_en);
+	set_bit(NVDIMM_INTEL_FREEZE_LOCK, &acpi_desc->dimm_cmd_force_en);
+	set_bit(NVDIMM_INTEL_SECURE_ERASE, &acpi_desc->dimm_cmd_force_en);
 }
 
 static void nfit_test1_setup(struct nfit_test *t)
@@ -2611,6 +2795,8 @@ static int nfit_test_probe(struct platform_device *pdev)
 				GFP_KERNEL);
 		nfit_test->fw = devm_kcalloc(dev, num,
 				sizeof(struct nfit_test_fw), GFP_KERNEL);
+		nfit_test->sec = devm_kcalloc(dev, num,
+				sizeof(struct nfit_test_sec), GFP_KERNEL);
 		if (nfit_test->dimm && nfit_test->dimm_dma && nfit_test->label
 				&& nfit_test->label_dma && nfit_test->dcr
 				&& nfit_test->dcr_dma && nfit_test->flush

_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm

^ permalink raw reply	[flat|nested] 30+ messages in thread

* [PATCH v8 12/12] libnvdimm: add documentation for nvdimm security support
  2018-08-28 22:47 [PATCH v8 00/12] Adding security support for nvdimm Dave Jiang
                   ` (10 preceding siblings ...)
  2018-08-28 22:48 ` [PATCH v8 11/12] nfit_test: add test support for Intel nvdimm security DSMs Dave Jiang
@ 2018-08-28 22:48 ` Dave Jiang
  2018-09-21 23:07 ` [PATCH v8 02/12] libnvdimm: create keyring to store security keys David Howells
                   ` (4 subsequent siblings)
  16 siblings, 0 replies; 30+ messages in thread
From: Dave Jiang @ 2018-08-28 22:48 UTC (permalink / raw)
  To: linux-nvdimm; +Cc: alison.schofield, keescook, ebiggers3, dhowells, keyrings

Add theory of operation for the security support that's going into
libnvdimm.

Signed-off-by: Dave Jiang <dave.jiang@intel.com>
---
 Documentation/nvdimm/security.txt |   82 +++++++++++++++++++++++++++++++++++++
 1 file changed, 82 insertions(+)
 create mode 100644 Documentation/nvdimm/security.txt

diff --git a/Documentation/nvdimm/security.txt b/Documentation/nvdimm/security.txt
new file mode 100644
index 000000000000..b230447bf237
--- /dev/null
+++ b/Documentation/nvdimm/security.txt
@@ -0,0 +1,82 @@
+NVDIMM SECURITY
+===============
+
+1. Introduction
+---------------
+
+With the introduction of Intel DSM v1.7 specification [1], security DSMs are
+introduced. The spec added the following security DSMs: "get security state",
+"set passphrase", "disable passphrase", "unlock unit", "freeze lock",
+"secure erase", and "overwrite". A security_ops data structure has been
+added to struct dimm in order to support the security operations and generic
+APIs are exposed to allow vendor neutral operations.
+
+2. Sysfs Interface
+------------------
+The "security" sysfs attribute is provided in the nvdimm sysfs directory. For
+example:
+/sys/devices/LNXSYSTM:00/LNXSYBUS:00/ACPI0012:00/ndbus0/nmem0/security
+
+The "show" function of that attribute will display the security state for
+that DIMM. The following states are available: disabled, unlocked, locked,
+frozen, and unsupported.
+
+The "store" function takes several commands when the attribute is written to
+in order to support some of the security functionalities:
+update - enable security. Add or update current key.
+disable - disable enabled security and remove key.
+freeze - freeze changing of security states.
+erase - generate new ecryption key for DIMM and crypto-scrambles all existing
+	user data.
+
+3. Key Management
+-----------------
+
+The key is associted to the payload by the DIMM id. For example:
+# cat /sys/devices/LNXSYSTM:00/LNXSYBUS:00/ACPI0012:00/ndbus0/nmem0/nfit/id
+8089-a2-1740-00000133
+The DIMM id would be provided along with the key payload (passphrase) to
+the kernel.
+
+The security keys are managed on the basis of a single key per DIMM. The
+key "passphrase" is expected to be 32bytes long or padded to 32bytes. This is
+similar to the ATA security specification [2]. A key is initially acquired
+via the request_key() kernel API call and retrieved from userspace. It is up to
+the user to provide an upcall application to retrieve the key in whatever
+fashion meets their security requirements.
+
+The payload provided to the key can be a 32bytes payload or 64bytes payload
+when doing an "update". The payload is viewed as 64 bytes in the following
+format:
+[32 bytes new key data zero padded][32 bytes current key data zero padded]
+However, a 32bytes payload can be provided and will be assumed as the old
+key to be 32 bytes of 0s and the provided 32bytes payload is the new key.
+It is up to the user upcall function how that's presented as the key payload
+to the kernel.
+
+All the other security functions that require a provided key can accept a
+32bytes payload or 64bytes. If the payload is 64bytes, then second 32bytes
+will be ignored and the first 32bytes contains the expected "passphrase".
+
+4. Unlocking
+------------
+When the DIMMs are being enumerated by the kernel, the kernel will attempt to
+retrieve the key from its keyring. If that fails, it will attempt to
+acquire the key from the userspace upcall function. This is the only time
+a locked DIMM can be unlocked. Once unlocked, the DIMM will remain unlocked
+until reboot.
+
+5. Update
+---------
+When doing an update, it is expected that the new key with the 64bit payload of
+format described above is added via the keyutils API or utility. The update
+command written to the sysfs attribute will be with the format:
+update:<old id>:<new id>
+If there is no old ID due to a security enabling, then a 0 should be passed in.
+It is expected that a user logon key has been injected via keyutils to provide
+the payload for the update operation. The kernel will take the new user key,
+attempt the update operation with the nvdimm, and replace the existing key's
+payload with the new passphrase.
+
+[1]: http://pmem.io/documents/NVDIMM_DSM_Interface-V1.7.pdf
+[2]: http://www.t13.org/documents/UploadedDocuments/docs2006/e05179r4-ACS-SecurityClarifications.pdf

_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm

^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [PATCH v8 04/12] keys: export lookup_user_key to external users
  2018-08-28 22:47 ` [PATCH v8 04/12] keys: export lookup_user_key to external users Dave Jiang
@ 2018-09-21 21:59   ` Dan Williams
  2018-09-21 22:02     ` Dave Jiang
  2018-09-21 23:05   ` David Howells
  1 sibling, 1 reply; 30+ messages in thread
From: Dan Williams @ 2018-09-21 21:59 UTC (permalink / raw)
  To: Dave Jiang
  Cc: Schofield, Alison, Kees Cook, linux-nvdimm, ebiggers3,
	David Howells, keyrings

On Tue, Aug 28, 2018 at 3:47 PM Dave Jiang <dave.jiang@intel.com> wrote:
>
> Export lookup_user_key() symbol in order to allow nvdimm passphrase
> update to retrieve user injected keys.
>
> Signed-off-by: Dave Jiang <dave.jiang@intel.com>
> Cc: David Howells <dhowells@redhat.com>
> ---
>  include/linux/key.h          |    3 +++
>  security/keys/internal.h     |    2 --
>  security/keys/process_keys.c |    1 +
>  3 files changed, 4 insertions(+), 2 deletions(-)
>
> diff --git a/include/linux/key.h b/include/linux/key.h
> index e58ee10f6e58..7099985e35a9 100644
> --- a/include/linux/key.h
> +++ b/include/linux/key.h
> @@ -346,6 +346,9 @@ static inline key_serial_t key_serial(const struct key *key)
>
>  extern void key_set_timeout(struct key *, unsigned);
>
> +extern key_ref_t lookup_user_key(key_serial_t id, unsigned long flags,
> +                                key_perm_t perm);
> +
>  /*
>   * The permissions required on a key that we're looking up.
>   */
> diff --git a/security/keys/internal.h b/security/keys/internal.h
> index 9f8208dc0e55..9968b21a76dd 100644
> --- a/security/keys/internal.h
> +++ b/security/keys/internal.h
> @@ -158,8 +158,6 @@ extern struct key *request_key_and_link(struct key_type *type,
>
>  extern bool lookup_user_key_possessed(const struct key *key,
>                                       const struct key_match_data *match_data);
> -extern key_ref_t lookup_user_key(key_serial_t id, unsigned long flags,
> -                                key_perm_t perm);
>  #define KEY_LOOKUP_CREATE      0x01
>  #define KEY_LOOKUP_PARTIAL     0x02
>  #define KEY_LOOKUP_FOR_UNLINK  0x04
> diff --git a/security/keys/process_keys.c b/security/keys/process_keys.c
> index d5b25e535d3a..ec4fd4531224 100644
> --- a/security/keys/process_keys.c
> +++ b/security/keys/process_keys.c
> @@ -755,6 +755,7 @@ key_ref_t lookup_user_key(key_serial_t id, unsigned long lflags,
>         put_cred(ctx.cred);
>         goto try_again;
>  }
> +EXPORT_SYMBOL(lookup_user_key);

This looks like a core api for affecting keyctl api internals, should
it be EXPORT_SYMBOL_GPL?

At a minimum this needs David's ack.
_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm

^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [PATCH v8 04/12] keys: export lookup_user_key to external users
  2018-09-21 21:59   ` Dan Williams
@ 2018-09-21 22:02     ` Dave Jiang
  0 siblings, 0 replies; 30+ messages in thread
From: Dave Jiang @ 2018-09-21 22:02 UTC (permalink / raw)
  To: Dan Williams
  Cc: Schofield, Alison, Kees Cook, linux-nvdimm, ebiggers3,
	David Howells, keyrings



On 09/21/2018 02:59 PM, Dan Williams wrote:
> On Tue, Aug 28, 2018 at 3:47 PM Dave Jiang <dave.jiang@intel.com> wrote:
>>
>> Export lookup_user_key() symbol in order to allow nvdimm passphrase
>> update to retrieve user injected keys.
>>
>> Signed-off-by: Dave Jiang <dave.jiang@intel.com>
>> Cc: David Howells <dhowells@redhat.com>
>> ---
>>  include/linux/key.h          |    3 +++
>>  security/keys/internal.h     |    2 --
>>  security/keys/process_keys.c |    1 +
>>  3 files changed, 4 insertions(+), 2 deletions(-)
>>
>> diff --git a/include/linux/key.h b/include/linux/key.h
>> index e58ee10f6e58..7099985e35a9 100644
>> --- a/include/linux/key.h
>> +++ b/include/linux/key.h
>> @@ -346,6 +346,9 @@ static inline key_serial_t key_serial(const struct key *key)
>>
>>  extern void key_set_timeout(struct key *, unsigned);
>>
>> +extern key_ref_t lookup_user_key(key_serial_t id, unsigned long flags,
>> +                                key_perm_t perm);
>> +
>>  /*
>>   * The permissions required on a key that we're looking up.
>>   */
>> diff --git a/security/keys/internal.h b/security/keys/internal.h
>> index 9f8208dc0e55..9968b21a76dd 100644
>> --- a/security/keys/internal.h
>> +++ b/security/keys/internal.h
>> @@ -158,8 +158,6 @@ extern struct key *request_key_and_link(struct key_type *type,
>>
>>  extern bool lookup_user_key_possessed(const struct key *key,
>>                                       const struct key_match_data *match_data);
>> -extern key_ref_t lookup_user_key(key_serial_t id, unsigned long flags,
>> -                                key_perm_t perm);
>>  #define KEY_LOOKUP_CREATE      0x01
>>  #define KEY_LOOKUP_PARTIAL     0x02
>>  #define KEY_LOOKUP_FOR_UNLINK  0x04
>> diff --git a/security/keys/process_keys.c b/security/keys/process_keys.c
>> index d5b25e535d3a..ec4fd4531224 100644
>> --- a/security/keys/process_keys.c
>> +++ b/security/keys/process_keys.c
>> @@ -755,6 +755,7 @@ key_ref_t lookup_user_key(key_serial_t id, unsigned long lflags,
>>         put_cred(ctx.cred);
>>         goto try_again;
>>  }
>> +EXPORT_SYMBOL(lookup_user_key);
> 
> This looks like a core api for affecting keyctl api internals, should
> it be EXPORT_SYMBOL_GPL?

Yes something I overlooked.

> 
> At a minimum this needs David's ack.
> 

David, can you please ACK? Thanks!
_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm

^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [PATCH v8 04/12] keys: export lookup_user_key to external users
  2018-08-28 22:47 ` [PATCH v8 04/12] keys: export lookup_user_key to external users Dave Jiang
  2018-09-21 21:59   ` Dan Williams
@ 2018-09-21 23:05   ` David Howells
  1 sibling, 0 replies; 30+ messages in thread
From: David Howells @ 2018-09-21 23:05 UTC (permalink / raw)
  To: Dan Williams
  Cc: Schofield, Alison  <alison.schofield@intel.com>,
	Kees Cook <keescook@chromium.org>,
	linux-nvdimm, ebiggers3, dhowells, keyrings

Dan Williams <dan.j.williams@intel.com> wrote:

> This looks like a core api for affecting keyctl api internals, should
> it be EXPORT_SYMBOL_GPL?

No, it should be EXPORT_SYMBOL().  Let me look through the other patches.

David
_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm

^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [PATCH v8 02/12] libnvdimm: create keyring to store security keys
  2018-08-28 22:47 [PATCH v8 00/12] Adding security support for nvdimm Dave Jiang
                   ` (11 preceding siblings ...)
  2018-08-28 22:48 ` [PATCH v8 12/12] libnvdimm: add documentation for nvdimm security support Dave Jiang
@ 2018-09-21 23:07 ` David Howells
  2018-09-21 23:20 ` [PATCH v8 05/12] nfit/libnvdimm: add unlock of nvdimm support for Intel DIMMs David Howells
                   ` (3 subsequent siblings)
  16 siblings, 0 replies; 30+ messages in thread
From: David Howells @ 2018-09-21 23:07 UTC (permalink / raw)
  To: Dave Jiang
  Cc: alison.schofield, keescook, linux-nvdimm, ebiggers3, dhowells, keyrings

Dave Jiang <dave.jiang@intel.com> wrote:

> +static void nvdimm_unregister_keyring(void)
> +{
> +	key_revoke(nvdimm_keyring);
> +}

You need key_put() as well otherwise you'll leak a ref.

David
_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm

^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [PATCH v8 05/12] nfit/libnvdimm: add unlock of nvdimm support for Intel DIMMs
  2018-08-28 22:47 [PATCH v8 00/12] Adding security support for nvdimm Dave Jiang
                   ` (12 preceding siblings ...)
  2018-09-21 23:07 ` [PATCH v8 02/12] libnvdimm: create keyring to store security keys David Howells
@ 2018-09-21 23:20 ` David Howells
  2018-09-21 23:27   ` Dave Jiang
  2018-09-21 23:57 ` [PATCH v8 06/12] nfit/libnvdimm: add set passphrase support for Intel nvdimms David Howells
                   ` (2 subsequent siblings)
  16 siblings, 1 reply; 30+ messages in thread
From: David Howells @ 2018-09-21 23:20 UTC (permalink / raw)
  To: Dave Jiang
  Cc: alison.schofield, keescook, linux-nvdimm, ebiggers3, dhowells, keyrings

Dave Jiang <dave.jiang@intel.com> wrote:

> +	depends on KEYS

That needs to be in patch 2 where you create a keyring.

> +	char desc[NVDIMM_KEY_DESC_LEN + strlen(NVDIMM_PREFIX)];

You should be using sizeof() not strlen() and do you need + 1 for the NUL
char?

> +	sprintf(desc, "%s%s", NVDIMM_PREFIX, nvdimm->dimm_id);

NVDIMM_PREFIX is a constant string.  I would recommend either declaring it as
a const char[] or just sticking it in the format string in place of the %s:

	sprintf(desc, NVDIMM_PREFIX "%s", nvdimm->dimm_id);

> +		if (!cached_key) {
> +			key_link(nvdimm_keyring, key);
> +			nvdimm->key = key;
> +			key->perm |= KEY_USR_SEARCH;
> +		}

Ummm...  You're altering the key permission?  That's not really yours to
change.

> +static int nvdimm_check_key_len(unsigned short len, bool update)
> +{
> +	if (len == (NVDIMM_PASSPHRASE_LEN * 2) && update)
> +		return 0;

In the cover you said:

	- Modified "update" to take two key ids and and use lookup_user_key()
	  in order to improve security.  (David)

is this a holdover?

David
_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm

^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [PATCH v8 05/12] nfit/libnvdimm: add unlock of nvdimm support for Intel DIMMs
  2018-09-21 23:20 ` [PATCH v8 05/12] nfit/libnvdimm: add unlock of nvdimm support for Intel DIMMs David Howells
@ 2018-09-21 23:27   ` Dave Jiang
  2018-09-21 23:51     ` Dan Williams
  0 siblings, 1 reply; 30+ messages in thread
From: Dave Jiang @ 2018-09-21 23:27 UTC (permalink / raw)
  To: David Howells
  Cc: alison.schofield, keescook, linux-nvdimm, ebiggers3, keyrings



On 09/21/2018 04:20 PM, David Howells wrote:
> Dave Jiang <dave.jiang@intel.com> wrote:
> 
>> +	depends on KEYS
> 
> That needs to be in patch 2 where you create a keyring.
> 
>> +	char desc[NVDIMM_KEY_DESC_LEN + strlen(NVDIMM_PREFIX)];
> 
> You should be using sizeof() not strlen() and do you need + 1 for the NUL
> char?
> 
>> +	sprintf(desc, "%s%s", NVDIMM_PREFIX, nvdimm->dimm_id);
> 
> NVDIMM_PREFIX is a constant string.  I would recommend either declaring it as
> a const char[] or just sticking it in the format string in place of the %s:
> 
> 	sprintf(desc, NVDIMM_PREFIX "%s", nvdimm->dimm_id);
> 
>> +		if (!cached_key) {
>> +			key_link(nvdimm_keyring, key);
>> +			nvdimm->key = key;
>> +			key->perm |= KEY_USR_SEARCH;
>> +		}
> 
> Ummm...  You're altering the key permission?  That's not really yours to
> change.

What can I do to allow the user app to look up the right key in order to
pass the key id to sysfs? Without the KEY_USR_SEARCH I am not able to
search for that key in the keyring.

> 
>> +static int nvdimm_check_key_len(unsigned short len, bool update)
>> +{
>> +	if (len == (NVDIMM_PASSPHRASE_LEN * 2) && update)
>> +		return 0;
> 
> In the cover you said:
> 
> 	- Modified "update" to take two key ids and and use lookup_user_key()
> 	  in order to improve security.  (David)
> 
> is this a holdover?

No. We are passing in two key ids to sysfs. The new one and the existing
one. I think that was what you suggested to do. With the above issue,
should I only just pass in the new key id since the specific sysfs node
should know which key to update already?


> 
> David
> 
_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm

^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [PATCH v8 05/12] nfit/libnvdimm: add unlock of nvdimm support for Intel DIMMs
  2018-09-21 23:27   ` Dave Jiang
@ 2018-09-21 23:51     ` Dan Williams
  0 siblings, 0 replies; 30+ messages in thread
From: Dan Williams @ 2018-09-21 23:51 UTC (permalink / raw)
  To: Dave Jiang
  Cc: Schofield, Alison, Kees Cook, linux-nvdimm, Eric Biggers,
	David Howells, keyrings

On Fri, Sep 21, 2018 at 4:27 PM Dave Jiang <dave.jiang@intel.com> wrote:
>
>
>
> On 09/21/2018 04:20 PM, David Howells wrote:
> > Dave Jiang <dave.jiang@intel.com> wrote:
> >
> >> +    depends on KEYS
> >
> > That needs to be in patch 2 where you create a keyring.
> >
> >> +    char desc[NVDIMM_KEY_DESC_LEN + strlen(NVDIMM_PREFIX)];
> >
> > You should be using sizeof() not strlen() and do you need + 1 for the NUL
> > char?
> >
> >> +    sprintf(desc, "%s%s", NVDIMM_PREFIX, nvdimm->dimm_id);
> >
> > NVDIMM_PREFIX is a constant string.  I would recommend either declaring it as
> > a const char[] or just sticking it in the format string in place of the %s:
> >
> >       sprintf(desc, NVDIMM_PREFIX "%s", nvdimm->dimm_id);
> >
> >> +            if (!cached_key) {
> >> +                    key_link(nvdimm_keyring, key);
> >> +                    nvdimm->key = key;
> >> +                    key->perm |= KEY_USR_SEARCH;
> >> +            }
> >
> > Ummm...  You're altering the key permission?  That's not really yours to
> > change.
>
> What can I do to allow the user app to look up the right key in order to
> pass the key id to sysfs? Without the KEY_USR_SEARCH I am not able to
> search for that key in the keyring.

I don't think you need to search. I would but NVDIMM_PREFIX into
include/uapi/linux/ndctl.h and userspace would need to know by
convention that for NFIT described DIMMs the piece after the prefix is
the ACPI NFIT unique-id. Other buses will need to have other
conventions, but I think it's fine to just document the convention as
"<prefix><bus-provider-specific-unique-id>", and then write up a
document of the known <bus-provider-specific-unique-id> formats.
_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm

^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [PATCH v8 06/12] nfit/libnvdimm: add set passphrase support for Intel nvdimms
  2018-08-28 22:47 [PATCH v8 00/12] Adding security support for nvdimm Dave Jiang
                   ` (13 preceding siblings ...)
  2018-09-21 23:20 ` [PATCH v8 05/12] nfit/libnvdimm: add unlock of nvdimm support for Intel DIMMs David Howells
@ 2018-09-21 23:57 ` David Howells
  2018-09-22  0:25   ` Dave Jiang
  2018-09-22  0:01 ` [PATCH v8 04/12] keys: export lookup_user_key to external users David Howells
  2018-09-24 21:02 ` [PATCH v8 02/12] libnvdimm: create keyring to store security keys David Howells
  16 siblings, 1 reply; 30+ messages in thread
From: David Howells @ 2018-09-21 23:57 UTC (permalink / raw)
  To: Dave Jiang
  Cc: alison.schofield, keescook, linux-nvdimm, ebiggers3, dhowells, keyrings

Dave Jiang <dave.jiang@intel.com> wrote:

> +	new_key = key_alloc(&key_type_logon, key->description,
> +			GLOBAL_ROOT_UID, GLOBAL_ROOT_GID, &init_cred, 0,

KEY_POS_SEARCH?  KEY_USR_VIEW?

> +			KEY_ALLOC_NOT_IN_QUOTA, NULL);
> + ...
> +	down_read(&key->sem);
> +	payload = key->payload.data[0];
> +	rc = key_instantiate_and_link(new_key, payload->data, key->datalen,

payload->datalen, not key->datalen.

> +			nvdimm_keyring, NULL);

Okay, that's a weird way of going about things.  I presume you don't want to
add key to nvdimm_keyring - maybe in case it gets updated whilst you're using
it and your private key isn't in quota?

> +	up_read(&key->sem);
> +	if (rc < 0) {
> +		key_revoke(new_key);
> +		key_put(new_key);
> +		return NULL;
> +	}

Just putting it here should work since it didn't get linked to the keyring if
any errors occurred.  Revoking it too shouldn't be necessary.

> +	key_invalidate(key);
> +	key_put(key);

Why are you invalidating the user's key?

> +	keyref = lookup_user_key(id, 0, 0);

KEY_NEED_SEARCH?  Though I suppose it's not strictly necessary as it's a key
that's private to the kernel.

> +	if (old_keyid != 0) {
> +		old_key = nvdimm_get_key(dev);
> +		if (old_key) {
> +			if (key_serial(old_key) != old_keyid) {

Ummm...  That's not what I meant.  Given the permissions you've set on your
private key, userspace shouldn't be able to find it, let alone give you the
key ID.

What I meant here was to use, say, nvdimm_lookup_user_key() to get a key from
userspace that contains the old password.  You can use the description of the
key to search nvdimm_keyring for the private key and then compare the
passwords.

Then you don't need to passphrases in the new key.

> +	rc = nvdimm_check_key_len(key->datalen, update);

payload->datalen.

> +	down_read(&key->sem);

This needs to be earlier.  The payload attached to the new key can be replaced
by keyctl_update() at any time whilst you're not holding the lock, so you
cannot use key->payload[*] without holding the lock or the RCU read lock.

> +	if (update)
> +		key_invalidate(key);

The key doesn't belong to you - should you really be invalidating it?

> +	else {
> +		key_link(nvdimm_keyring, key);
> +		nvdimm->key = key;
> +		key->perm |= KEY_USR_SEARCH;
> +	}

Um - do you really want to be taking the key into your internal keyring?  Why
aren't you calling nvdimm_replace_key()?  Also, you shouldn't alter the
permission - it's not your key.

> +static int __parse_update(const char *buf, size_t len, unsigned int *old_id,
> +		unsigned int *new_id)
> +{

Try using sscanf()?

David
_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm

^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [PATCH v8 04/12] keys: export lookup_user_key to external users
  2018-08-28 22:47 [PATCH v8 00/12] Adding security support for nvdimm Dave Jiang
                   ` (14 preceding siblings ...)
  2018-09-21 23:57 ` [PATCH v8 06/12] nfit/libnvdimm: add set passphrase support for Intel nvdimms David Howells
@ 2018-09-22  0:01 ` David Howells
  2018-09-24 21:02 ` [PATCH v8 02/12] libnvdimm: create keyring to store security keys David Howells
  16 siblings, 0 replies; 30+ messages in thread
From: David Howells @ 2018-09-22  0:01 UTC (permalink / raw)
  To: Dave Jiang
  Cc: alison.schofield, keescook, linux-nvdimm, ebiggers3, dhowells, keyrings

Dave Jiang <dave.jiang@intel.com> wrote:

> Export lookup_user_key() symbol in order to allow nvdimm passphrase
> update to retrieve user injected keys.
> 
> Signed-off-by: Dave Jiang <dave.jiang@intel.com>

Acked-by: David Howells <dhowells@redhat.com>
_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm

^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [PATCH v8 02/12] libnvdimm: create keyring to store security keys
  2018-08-28 22:47 ` [PATCH v8 02/12] libnvdimm: create keyring to store security keys Dave Jiang
@ 2018-09-22  0:19   ` Dan Williams
  2018-09-24 21:04   ` David Howells
  1 sibling, 0 replies; 30+ messages in thread
From: Dan Williams @ 2018-09-22  0:19 UTC (permalink / raw)
  To: Dave Jiang
  Cc: Schofield, Alison, Kees Cook, linux-nvdimm, Eric Biggers,
	David Howells, keyrings

On Tue, Aug 28, 2018 at 3:47 PM Dave Jiang <dave.jiang@intel.com> wrote:
>
> Prepping the libnvdimm to support security management by adding a keyring
> in order to provide passphrase management through the kernel key management
> APIs.
>
> Signed-off-by: Dave Jiang <dave.jiang@intel.com>
> Reviewed-by: Dan Williams <dan.j.williams@intel.com>
> ---
>  drivers/nvdimm/core.c      |    7 ++++++-
>  drivers/nvdimm/dimm_devs.c |   29 ++++++++++++++++++++++++++++-
>  drivers/nvdimm/nd-core.h   |    1 +
>  include/linux/libnvdimm.h  |    3 +++
>  kernel/cred.c              |    1 +
>  5 files changed, 39 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/nvdimm/core.c b/drivers/nvdimm/core.c
> index acce050856a8..3cd33d5c7cf0 100644
> --- a/drivers/nvdimm/core.c
> +++ b/drivers/nvdimm/core.c
> @@ -437,9 +437,12 @@ static __init int libnvdimm_init(void)
>  {
>         int rc;
>
> -       rc = nvdimm_bus_init();
> +       rc = nvdimm_devs_init();
>         if (rc)
>                 return rc;
> +       rc = nvdimm_bus_init();
> +       if (rc)
> +               goto err_bus;
>         rc = nvdimm_init();
>         if (rc)
>                 goto err_dimm;
> @@ -454,6 +457,8 @@ static __init int libnvdimm_init(void)
>         nvdimm_exit();
>   err_dimm:
>         nvdimm_bus_exit();
> + err_bus:
> +       nvdimm_devs_exit();
>         return rc;
>  }
>
> diff --git a/drivers/nvdimm/dimm_devs.c b/drivers/nvdimm/dimm_devs.c
> index 8d348b22ba45..586432450c82 100644
> --- a/drivers/nvdimm/dimm_devs.c
> +++ b/drivers/nvdimm/dimm_devs.c
> @@ -18,12 +18,15 @@
>  #include <linux/io.h>
>  #include <linux/fs.h>
>  #include <linux/mm.h>
> +#include <linux/key.h>
> +#include <linux/init_task.h>
>  #include "nd-core.h"
>  #include "label.h"
>  #include "pmem.h"
>  #include "nd.h"
>
>  static DEFINE_IDA(dimm_ida);
> +static struct key *nvdimm_keyring;
>
>  /*
>   * Retrieve bus and dimm handle and return if this bus supports
> @@ -668,7 +671,31 @@ int nvdimm_bus_check_dimm_count(struct nvdimm_bus *nvdimm_bus, int dimm_count)
>  }
>  EXPORT_SYMBOL_GPL(nvdimm_bus_check_dimm_count);
>
> -void __exit nvdimm_devs_exit(void)
> +static int nvdimm_register_keyring(void)
>  {
> +       nvdimm_keyring = keyring_alloc(".nvdimm",
> +                       GLOBAL_ROOT_UID, GLOBAL_ROOT_GID, &init_cred,
[..]
> diff --git a/kernel/cred.c b/kernel/cred.c
> index ecf03657e71c..3a161b78a0a4 100644
> --- a/kernel/cred.c
> +++ b/kernel/cred.c
> @@ -64,6 +64,7 @@ struct cred init_cred = {
>         .user_ns                = &init_user_ns,
>         .group_info             = &init_groups,
>  };
> +EXPORT_SYMBOL(init_cred);

I think you want to use prepare_kernel_cred(), not export init_cred.
_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm

^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [PATCH v8 06/12] nfit/libnvdimm: add set passphrase support for Intel nvdimms
  2018-09-21 23:57 ` [PATCH v8 06/12] nfit/libnvdimm: add set passphrase support for Intel nvdimms David Howells
@ 2018-09-22  0:25   ` Dave Jiang
  2018-09-22  1:26     ` Dan Williams
  0 siblings, 1 reply; 30+ messages in thread
From: Dave Jiang @ 2018-09-22  0:25 UTC (permalink / raw)
  To: David Howells
  Cc: alison.schofield, keescook, linux-nvdimm, ebiggers3, keyrings



On 09/21/2018 04:57 PM, David Howells wrote:
> Dave Jiang <dave.jiang@intel.com> wrote:
> 
>> +	new_key = key_alloc(&key_type_logon, key->description,
>> +			GLOBAL_ROOT_UID, GLOBAL_ROOT_GID, &init_cred, 0,
> 
> KEY_POS_SEARCH?  KEY_USR_VIEW?
> 
>> +			KEY_ALLOC_NOT_IN_QUOTA, NULL);
>> + ...
>> +	down_read(&key->sem);
>> +	payload = key->payload.data[0];
>> +	rc = key_instantiate_and_link(new_key, payload->data, key->datalen,
> 
> payload->datalen, not key->datalen.
> 
>> +			nvdimm_keyring, NULL);
> 
> Okay, that's a weird way of going about things.  I presume you don't want to
> add key to nvdimm_keyring - maybe in case it gets updated whilst you're using
> it and your private key isn't in quota?
> 
>> +	up_read(&key->sem);
>> +	if (rc < 0) {
>> +		key_revoke(new_key);
>> +		key_put(new_key);
>> +		return NULL;
>> +	}
> 
> Just putting it here should work since it didn't get linked to the keyring if
> any errors occurred.  Revoking it too shouldn't be necessary.
> 
>> +	key_invalidate(key);
>> +	key_put(key);
> 
> Why are you invalidating the user's key?
> 
>> +	keyref = lookup_user_key(id, 0, 0);
> 
> KEY_NEED_SEARCH?  Though I suppose it's not strictly necessary as it's a key
> that's private to the kernel.
> 
>> +	if (old_keyid != 0) {
>> +		old_key = nvdimm_get_key(dev);
>> +		if (old_key) {
>> +			if (key_serial(old_key) != old_keyid) {
> 
> Ummm...  That's not what I meant.  Given the permissions you've set on your
> private key, userspace shouldn't be able to find it, let alone give you the
> key ID.
> 
> What I meant here was to use, say, nvdimm_lookup_user_key() to get a key from
> userspace that contains the old password.  You can use the description of the
> key to search nvdimm_keyring for the private key and then compare the
> passwords.

Ok I have a bit of confusion here. When the user injects a new key with
the same description and new passphrase, would that not replace the
existing user key with the old passphrase? Also, if I'm calling
lookup_user_key, where would the key_id come from for the old user key?
I suppose I can cache it.... Maybe I'm not quite understanding the exact
flow of how things you are suggesting.

> 
> Then you don't need to passphrases in the new key.
> 
>> +	rc = nvdimm_check_key_len(key->datalen, update);
> 
> payload->datalen.
> 
>> +	down_read(&key->sem);
> 
> This needs to be earlier.  The payload attached to the new key can be replaced
> by keyctl_update() at any time whilst you're not holding the lock, so you
> cannot use key->payload[*] without holding the lock or the RCU read lock.
> 
>> +	if (update)
>> +		key_invalidate(key);
> 
> The key doesn't belong to you - should you really be invalidating it?
> 
>> +	else {
>> +		key_link(nvdimm_keyring, key);
>> +		nvdimm->key = key;
>> +		key->perm |= KEY_USR_SEARCH;
>> +	}
> 
> Um - do you really want to be taking the key into your internal keyring?  Why
> aren't you calling nvdimm_replace_key()?  Also, you shouldn't alter the
> permission - it's not your key.
> 
>> +static int __parse_update(const char *buf, size_t len, unsigned int *old_id,
>> +		unsigned int *new_id)
>> +{
> 
> Try using sscanf()?
> 
> David
> 
_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm

^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [PATCH v8 06/12] nfit/libnvdimm: add set passphrase support for Intel nvdimms
  2018-09-22  0:25   ` Dave Jiang
@ 2018-09-22  1:26     ` Dan Williams
  0 siblings, 0 replies; 30+ messages in thread
From: Dan Williams @ 2018-09-22  1:26 UTC (permalink / raw)
  To: Dave Jiang
  Cc: Schofield, Alison, Kees Cook, linux-nvdimm, Eric Biggers,
	David Howells, keyrings

On Fri, Sep 21, 2018 at 5:29 PM Dave Jiang <dave.jiang@intel.com> wrote:
[..]
> > What I meant here was to use, say, nvdimm_lookup_user_key() to get a key from
> > userspace that contains the old password.  You can use the description of the
> > key to search nvdimm_keyring for the private key and then compare the
> > passwords.
>
> Ok I have a bit of confusion here. When the user injects a new key with
> the same description and new passphrase, would that not replace the
> existing user key with the old passphrase? Also, if I'm calling
> lookup_user_key, where would the key_id come from for the old user key?
> I suppose I can cache it.... Maybe I'm not quite understanding the exact
> flow of how things you are suggesting.

If the DIMM is in the in the unlocked state the kernel should already
have the old passphrase cached, right? Could we just have the kernel
fail ->change_key() requests if the kernel does not already have a
valid old passphrase?

It does mean we can't support going from the locked state directly to
changing the password, but I would expect requiring an initial unlock
is fine.
_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm

^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [PATCH v8 05/12] nfit/libnvdimm: add unlock of nvdimm support for Intel DIMMs
  2018-08-28 22:47 ` [PATCH v8 05/12] nfit/libnvdimm: add unlock of nvdimm support for Intel DIMMs Dave Jiang
@ 2018-09-23  0:10   ` Dan Williams
  0 siblings, 0 replies; 30+ messages in thread
From: Dan Williams @ 2018-09-23  0:10 UTC (permalink / raw)
  To: Dave Jiang
  Cc: Schofield, Alison, Kees Cook, linux-nvdimm, Eric Biggers,
	David Howells, keyrings

On Tue, Aug 28, 2018 at 3:47 PM Dave Jiang <dave.jiang@intel.com> wrote:
>
> Add support to allow query the security status of the Intel nvdimms and
> also unlock the dimm via the kernel key management APIs. The passphrase is
> expected to be pulled from userspace through keyutils. Moving the Intel
> related bits to its own source file as well.
>
> Signed-off-by: Dave Jiang <dave.jiang@intel.com>
> Reviewed-by: Dan Williams <dan.j.williams@intel.com>
> ---
>  drivers/acpi/nfit/Makefile |    1
>  drivers/acpi/nfit/core.c   |    3 +
>  drivers/acpi/nfit/intel.c  |  152 ++++++++++++++++++++++++++++++++++++++++++++
>  drivers/acpi/nfit/intel.h  |   15 ++++
>  drivers/nvdimm/Kconfig     |    1
>  drivers/nvdimm/dimm.c      |    7 ++
>  drivers/nvdimm/dimm_devs.c |  128 +++++++++++++++++++++++++++++++++++++
>  drivers/nvdimm/nd-core.h   |    4 +
>  drivers/nvdimm/nd.h        |    2 +
>  include/linux/libnvdimm.h  |   24 +++++++
>  10 files changed, 334 insertions(+), 3 deletions(-)
>  create mode 100644 drivers/acpi/nfit/intel.c
>
> diff --git a/drivers/acpi/nfit/Makefile b/drivers/acpi/nfit/Makefile
> index a407e769f103..443c7ef4e6a6 100644
> --- a/drivers/acpi/nfit/Makefile
> +++ b/drivers/acpi/nfit/Makefile
> @@ -1,3 +1,4 @@
>  obj-$(CONFIG_ACPI_NFIT) := nfit.o
>  nfit-y := core.o
> +nfit-$(CONFIG_X86) += intel.o
>  nfit-$(CONFIG_X86_MCE) += mce.o
> diff --git a/drivers/acpi/nfit/core.c b/drivers/acpi/nfit/core.c
> index 21235d555b5f..9cb6a108ecba 100644
> --- a/drivers/acpi/nfit/core.c
> +++ b/drivers/acpi/nfit/core.c
> @@ -1904,7 +1904,8 @@ static int acpi_nfit_register_dimms(struct acpi_nfit_desc *acpi_desc)
>                 nvdimm = nvdimm_create(acpi_desc->nvdimm_bus, nfit_mem,
>                                 acpi_nfit_dimm_attribute_groups,
>                                 flags, cmd_mask, flush ? flush->hint_count : 0,
> -                               nfit_mem->flush_wpq, &nfit_mem->id[0]);
> +                               nfit_mem->flush_wpq, &nfit_mem->id[0],
> +                               acpi_nfit_get_security_ops(nfit_mem->family));
>                 if (!nvdimm)
>                         return -ENOMEM;
>
> diff --git a/drivers/acpi/nfit/intel.c b/drivers/acpi/nfit/intel.c
> new file mode 100644
> index 000000000000..4bfc1c1da339
> --- /dev/null
> +++ b/drivers/acpi/nfit/intel.c
> @@ -0,0 +1,152 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/* Copyright(c) 2018 Intel Corporation. All rights reserved. */
> +/*
> + * Intel specific NFIT ops
> + */
> +#include <linux/libnvdimm.h>
> +#include <linux/module.h>
> +#include <linux/mutex.h>
> +#include <linux/ndctl.h>
> +#include <linux/sysfs.h>
> +#include <linux/delay.h>
> +#include <linux/acpi.h>
> +#include <linux/io.h>
> +#include <linux/nd.h>
> +#include <asm/cacheflush.h>
> +#include <asm/smp.h>
> +#include <acpi/nfit.h>
> +#include "intel.h"
> +#include "nfit.h"
> +
> +static int intel_dimm_security_unlock(struct nvdimm_bus *nvdimm_bus,
> +               struct nvdimm *nvdimm, const struct nvdimm_key_data *nkey)
> +{
> +       struct nvdimm_bus_descriptor *nd_desc = to_nd_desc(nvdimm_bus);
> +       int cmd_rc, rc = 0;
> +       struct nfit_mem *nfit_mem = nvdimm_provider_data(nvdimm);
> +       struct {
> +               struct nd_cmd_pkg pkg;
> +               struct nd_intel_unlock_unit cmd;
> +       } nd_cmd = {
> +               .pkg = {
> +                       .nd_command = NVDIMM_INTEL_UNLOCK_UNIT,
> +                       .nd_family = NVDIMM_FAMILY_INTEL,
> +                       .nd_size_in = ND_INTEL_PASSPHRASE_SIZE,
> +                       .nd_size_out = ND_INTEL_STATUS_SIZE,
> +                       .nd_fw_size = ND_INTEL_STATUS_SIZE,
> +               },
> +               .cmd = {
> +                       .status = 0,
> +               },
> +       };
> +
> +       if (!test_bit(NVDIMM_INTEL_UNLOCK_UNIT, &nfit_mem->dsm_mask))
> +               return -ENOTTY;
> +
> +       memcpy(nd_cmd.cmd.passphrase, nkey->data,
> +                       sizeof(nd_cmd.cmd.passphrase));
> +       rc = nd_desc->ndctl(nd_desc, nvdimm, ND_CMD_CALL, &nd_cmd,
> +                       sizeof(nd_cmd), &cmd_rc);
> +       if (rc < 0)
> +               goto out;
> +       if (cmd_rc < 0) {
> +               rc = cmd_rc;
> +               goto out;
> +       }
> +
> +       switch (nd_cmd.cmd.status) {
> +       case 0:
> +               break;
> +       case ND_INTEL_STATUS_INVALID_PASS:
> +               rc = -EINVAL;
> +               goto out;
> +       case ND_INTEL_STATUS_INVALID_STATE:
> +       default:
> +               rc = -ENXIO;
> +               goto out;
> +       }
> +
> +       /*
> +        * TODO: define a cross arch wbinvd when/if NVDIMM_FAMILY_INTEL
> +        * support arrives on another arch.
> +        */
> +       /* DIMM unlocked, invalidate all CPU caches before we read it */
> +       wbinvd_on_all_cpus();
> +
> + out:
> +       return rc;
> +}
> +
> +static int intel_dimm_security_state(struct nvdimm_bus *nvdimm_bus,
> +               struct nvdimm *nvdimm, enum nvdimm_security_state *state)
> +{
> +       struct nvdimm_bus_descriptor *nd_desc = to_nd_desc(nvdimm_bus);
> +       int cmd_rc, rc = 0;
> +       struct nfit_mem *nfit_mem = nvdimm_provider_data(nvdimm);
> +       struct {
> +               struct nd_cmd_pkg pkg;
> +               struct nd_intel_get_security_state cmd;
> +       } nd_cmd = {
> +               .pkg = {
> +                       .nd_command = NVDIMM_INTEL_GET_SECURITY_STATE,
> +                       .nd_family = NVDIMM_FAMILY_INTEL,
> +                       .nd_size_in = 0,
> +                       .nd_size_out =
> +                               sizeof(struct nd_intel_get_security_state),
> +                       .nd_fw_size =
> +                               sizeof(struct nd_intel_get_security_state),
> +               },
> +               .cmd = {
> +                       .status = 0,
> +                       .state = 0,
> +               },
> +       };
> +
> +       if (!test_bit(NVDIMM_INTEL_GET_SECURITY_STATE, &nfit_mem->dsm_mask)) {
> +               *state = NVDIMM_SECURITY_UNSUPPORTED;
> +               return 0;
> +       }
> +
> +       *state = NVDIMM_SECURITY_DISABLED;
> +       rc = nd_desc->ndctl(nd_desc, nvdimm, ND_CMD_CALL, &nd_cmd,
> +                       sizeof(nd_cmd), &cmd_rc);
> +       if (rc < 0)
> +               goto out;
> +       if (cmd_rc < 0) {
> +               rc = cmd_rc;
> +               goto out;
> +       }
> +
> +       switch (nd_cmd.cmd.status) {
> +       case 0:
> +               break;
> +       case ND_INTEL_STATUS_RETRY:
> +               rc = -EAGAIN;
> +               goto out;
> +       case ND_INTEL_STATUS_NOT_READY:
> +       default:
> +               rc = -ENXIO;
> +               goto out;
> +       }
> +
> +       /* check and see if security is enabled and locked */
> +       if (nd_cmd.cmd.state & ND_INTEL_SEC_STATE_UNSUPPORTED)
> +               *state = NVDIMM_SECURITY_UNSUPPORTED;
> +       else if (nd_cmd.cmd.state & ND_INTEL_SEC_STATE_ENABLED) {
> +               if (nd_cmd.cmd.state & ND_INTEL_SEC_STATE_LOCKED)
> +                       *state = NVDIMM_SECURITY_LOCKED;
> +               else
> +                       *state = NVDIMM_SECURITY_UNLOCKED;
> +       } else
> +               *state = NVDIMM_SECURITY_DISABLED;
> +
> + out:
> +       if (rc < 0)
> +               *state = NVDIMM_SECURITY_INVALID;
> +       return rc;
> +}
> +
> +const struct nvdimm_security_ops intel_security_ops = {
> +       .state = intel_dimm_security_state,
> +       .unlock = intel_dimm_security_unlock,
> +};
> diff --git a/drivers/acpi/nfit/intel.h b/drivers/acpi/nfit/intel.h
> index a6f8f4bc8ebb..f9ac718776ca 100644
> --- a/drivers/acpi/nfit/intel.h
> +++ b/drivers/acpi/nfit/intel.h
> @@ -8,6 +8,8 @@
>
>  #ifdef CONFIG_X86
>
> +extern const struct nvdimm_security_ops intel_security_ops;
> +
>  #define ND_INTEL_STATUS_SIZE           4
>  #define ND_INTEL_PASSPHRASE_SIZE       32
>
> @@ -64,4 +66,17 @@ struct nd_intel_query_overwrite {
>  } __packed;
>  #endif /* CONFIG_X86 */
>
> +static inline const struct nvdimm_security_ops *
> +acpi_nfit_get_security_ops(int family)
> +{
> +       switch (family) {
> +#ifdef CONFIG_X86
> +       case NVDIMM_FAMILY_INTEL:
> +               return &intel_security_ops;
> +#endif
> +       default:
> +               return NULL;
> +       }
> +}
> +
>  #endif
> diff --git a/drivers/nvdimm/Kconfig b/drivers/nvdimm/Kconfig
> index 9d36473dc2a2..830a7a5342e1 100644
> --- a/drivers/nvdimm/Kconfig
> +++ b/drivers/nvdimm/Kconfig
> @@ -3,6 +3,7 @@ menuconfig LIBNVDIMM
>         depends on PHYS_ADDR_T_64BIT
>         depends on HAS_IOMEM
>         depends on BLK_DEV
> +       depends on KEYS
>         help
>           Generic support for non-volatile memory devices including
>           ACPI-6-NFIT defined resources.  On platforms that define an
> diff --git a/drivers/nvdimm/dimm.c b/drivers/nvdimm/dimm.c
> index 6c8fb7590838..b6381ddbd6c1 100644
> --- a/drivers/nvdimm/dimm.c
> +++ b/drivers/nvdimm/dimm.c
> @@ -51,6 +51,13 @@ static int nvdimm_probe(struct device *dev)
>         get_device(dev);
>         kref_init(&ndd->kref);
>
> +       nvdimm_security_get_state(dev);
> +
> +       /* unlock DIMM here before touch label */
> +       rc = nvdimm_security_unlock_dimm(dev);
> +       if (rc < 0)
> +               dev_warn(dev, "failed to unlock dimm %s\n", dev_name(dev));
> +
>         /*
>          * EACCES failures reading the namespace label-area-properties
>          * are interpreted as the DIMM capacity being locked but the
> diff --git a/drivers/nvdimm/dimm_devs.c b/drivers/nvdimm/dimm_devs.c
> index e753b2095402..fe61f2a2ad5d 100644
> --- a/drivers/nvdimm/dimm_devs.c
> +++ b/drivers/nvdimm/dimm_devs.c
> @@ -20,6 +20,7 @@
>  #include <linux/mm.h>
>  #include <linux/key.h>
>  #include <linux/init_task.h>
> +#include <keys/user-type.h>
>  #include "nd-core.h"
>  #include "label.h"
>  #include "pmem.h"
> @@ -28,6 +29,129 @@
>  static DEFINE_IDA(dimm_ida);
>  static struct key *nvdimm_keyring;
>
> +#define NVDIMM_PREFIX          "nvdimm:"
> +
> +/*
> + * Find key in kernel keyring
> + */
> +static struct key *nvdimm_get_key(struct device *dev)
> +{
> +       struct nvdimm *nvdimm = to_nvdimm(dev);
> +
> +       if (!nvdimm->key)
> +               return NULL;
> +
> +       if (key_validate(nvdimm->key) < 0)
> +               return NULL;
> +
> +       key_get(nvdimm->key);
> +
> +       dev_dbg(dev, "%s: key found: %d\n", __func__,
> +                       key_serial(nvdimm->key));
> +       return nvdimm->key;
> +}
> +
> +/*
> + * Retrieve kernel key for DIMM and request from user space if necessary.
> + */
> +static struct key *nvdimm_request_key(struct device *dev)
> +{
> +       struct nvdimm *nvdimm = to_nvdimm(dev);
> +       struct key *key = NULL;
> +       char desc[NVDIMM_KEY_DESC_LEN + strlen(NVDIMM_PREFIX)];
> +
> +       sprintf(desc, "%s%s", NVDIMM_PREFIX, nvdimm->dimm_id);
> +       key = request_key(&key_type_logon, desc, desc);
> +       if (IS_ERR(key))
> +               key = NULL;
> +
> +       return key;
> +}
> +
> +static int nvdimm_check_key_len(unsigned short len, bool update)
> +{
> +       if (len == (NVDIMM_PASSPHRASE_LEN * 2) && update)
> +               return 0;
> +
> +       if (len == NVDIMM_PASSPHRASE_LEN)
> +               return 0;
> +
> +       return -EINVAL;
> +}
> +
> +int nvdimm_security_get_state(struct device *dev)
> +{
> +       struct nvdimm *nvdimm = to_nvdimm(dev);
> +       struct nvdimm_bus *nvdimm_bus = walk_to_nvdimm_bus(dev);
> +
> +       if (!nvdimm->security_ops)
> +               return 0;
> +
> +       return nvdimm->security_ops->state(nvdimm_bus, nvdimm,
> +                       &nvdimm->state);
> +}
> +
> +int nvdimm_security_unlock_dimm(struct device *dev)
> +{
> +       struct nvdimm *nvdimm = to_nvdimm(dev);
> +       struct nvdimm_bus *nvdimm_bus = walk_to_nvdimm_bus(dev);
> +       struct key *key;
> +       struct user_key_payload *payload;
> +       int rc;
> +       bool cached_key = false;
> +
> +       if (!nvdimm->security_ops)
> +               return 0;
> +
> +       if (nvdimm->state == NVDIMM_SECURITY_UNLOCKED ||

Hmm, if the DIMM is unlocked before the driver loads that means we'll
never ask for the current key which means we can't count on having it
around for change key requests. I think there needs to be a sysfs
interface to trigger loading the current key into the keyring cache so
it is available for future operations that need it.
_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm

^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [PATCH v8 02/12] libnvdimm: create keyring to store security keys
  2018-08-28 22:47 [PATCH v8 00/12] Adding security support for nvdimm Dave Jiang
                   ` (15 preceding siblings ...)
  2018-09-22  0:01 ` [PATCH v8 04/12] keys: export lookup_user_key to external users David Howells
@ 2018-09-24 21:02 ` David Howells
  2018-09-24 21:15   ` Dave Jiang
  16 siblings, 1 reply; 30+ messages in thread
From: David Howells @ 2018-09-24 21:02 UTC (permalink / raw)
  To: Dave Jiang
  Cc: alison.schofield, keescook, linux-nvdimm, ebiggers3, dhowells, keyrings

Dave Jiang <dave.jiang@intel.com> wrote:

> +			GLOBAL_ROOT_UID, GLOBAL_ROOT_GID, &init_cred,

Hmmm...  I wonder if current_cred() would suffice since you must be called
from something that has the ability to load modules.  Further, I wonder if
module loading should be wrapped in an override with init_cred in the core.

> +			(KEY_USR_ALL & ~KEY_USR_SETATTR),

Did you really want to give the user write access, btw?

David
_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm

^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [PATCH v8 02/12] libnvdimm: create keyring to store security keys
  2018-08-28 22:47 ` [PATCH v8 02/12] libnvdimm: create keyring to store security keys Dave Jiang
  2018-09-22  0:19   ` Dan Williams
@ 2018-09-24 21:04   ` David Howells
  2018-09-24 21:12     ` Dave Jiang
  1 sibling, 1 reply; 30+ messages in thread
From: David Howells @ 2018-09-24 21:04 UTC (permalink / raw)
  To: Dan Williams
  Cc: Schofield, Alison  <alison.schofield@intel.com>,
	Kees Cook <keescook@chromium.org>,
	linux-nvdimm, Eric Biggers, dhowells, keyrings

Dan Williams <dan.j.williams@intel.com> wrote:

> I think you want to use prepare_kernel_cred(), not export init_cred.

That only works if the searching is done with the creds generated by
prepare_kernel_cred().  He probably does want init_cred, or at least
current_cred() if module loading overrides the creds with init_cred whilst
initialising the module.

David
_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm

^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [PATCH v8 02/12] libnvdimm: create keyring to store security keys
  2018-09-24 21:04   ` David Howells
@ 2018-09-24 21:12     ` Dave Jiang
  0 siblings, 0 replies; 30+ messages in thread
From: Dave Jiang @ 2018-09-24 21:12 UTC (permalink / raw)
  To: David Howells, Dan Williams
  Cc: Schofield, Alison, Kees Cook, linux-nvdimm, Eric Biggers, keyrings



On 09/24/2018 02:04 PM, David Howells wrote:
> Dan Williams <dan.j.williams@intel.com> wrote:
> 
>> I think you want to use prepare_kernel_cred(), not export init_cred.
> 
> That only works if the searching is done with the creds generated by
> prepare_kernel_cred().  He probably does want init_cred, or at least
> current_cred() if module loading overrides the creds with init_cred whilst
> initialising the module.
> 
> David
> 

Can I pass in NULL or is some sort of cred is necessary?
_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm

^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [PATCH v8 02/12] libnvdimm: create keyring to store security keys
  2018-09-24 21:02 ` [PATCH v8 02/12] libnvdimm: create keyring to store security keys David Howells
@ 2018-09-24 21:15   ` Dave Jiang
  0 siblings, 0 replies; 30+ messages in thread
From: Dave Jiang @ 2018-09-24 21:15 UTC (permalink / raw)
  To: David Howells
  Cc: alison.schofield, keescook, linux-nvdimm, ebiggers3, keyrings

On 09/24/2018 02:02 PM, David Howells wrote:
> Dave Jiang <dave.jiang@intel.com> wrote:
> 
>> +			GLOBAL_ROOT_UID, GLOBAL_ROOT_GID, &init_cred,
> 
> Hmmm...  I wonder if current_cred() would suffice since you must be called
> from something that has the ability to load modules.  Further, I wonder if
> module loading should be wrapped in an override with init_cred in the core.
> 
>> +			(KEY_USR_ALL & ~KEY_USR_SETATTR),
> 
> Did you really want to give the user write access, btw?

Hmmm....maybe I don't want user access at all since this is a kernel key
ring?

> 
> David
> 
_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm

^ permalink raw reply	[flat|nested] 30+ messages in thread

end of thread, other threads:[~2018-09-24 21:15 UTC | newest]

Thread overview: 30+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-08-28 22:47 [PATCH v8 00/12] Adding security support for nvdimm Dave Jiang
2018-08-28 22:47 ` [PATCH v8 01/12] nfit: add support for Intel DSM 1.7 commands Dave Jiang
2018-08-28 22:47 ` [PATCH v8 02/12] libnvdimm: create keyring to store security keys Dave Jiang
2018-09-22  0:19   ` Dan Williams
2018-09-24 21:04   ` David Howells
2018-09-24 21:12     ` Dave Jiang
2018-08-28 22:47 ` [PATCH v8 03/12] nfit/libnvdimm: store dimm id as a member to struct nvdimm Dave Jiang
2018-08-28 22:47 ` [PATCH v8 04/12] keys: export lookup_user_key to external users Dave Jiang
2018-09-21 21:59   ` Dan Williams
2018-09-21 22:02     ` Dave Jiang
2018-09-21 23:05   ` David Howells
2018-08-28 22:47 ` [PATCH v8 05/12] nfit/libnvdimm: add unlock of nvdimm support for Intel DIMMs Dave Jiang
2018-09-23  0:10   ` Dan Williams
2018-08-28 22:47 ` [PATCH v8 06/12] nfit/libnvdimm: add set passphrase support for Intel nvdimms Dave Jiang
2018-08-28 22:47 ` [PATCH v8 07/12] nfit/libnvdimm: add disable passphrase support to Intel nvdimm Dave Jiang
2018-08-28 22:48 ` [PATCH v8 08/12] nfit/libnvdimm: add freeze security " Dave Jiang
2018-08-28 22:48 ` [PATCH v8 09/12] nfit/libnvdimm: add support for issue secure erase DSM " Dave Jiang
2018-08-28 22:48 ` [PATCH v8 10/12] nfit_test: add context to dimm_dev for nfit_test Dave Jiang
2018-08-28 22:48 ` [PATCH v8 11/12] nfit_test: add test support for Intel nvdimm security DSMs Dave Jiang
2018-08-28 22:48 ` [PATCH v8 12/12] libnvdimm: add documentation for nvdimm security support Dave Jiang
2018-09-21 23:07 ` [PATCH v8 02/12] libnvdimm: create keyring to store security keys David Howells
2018-09-21 23:20 ` [PATCH v8 05/12] nfit/libnvdimm: add unlock of nvdimm support for Intel DIMMs David Howells
2018-09-21 23:27   ` Dave Jiang
2018-09-21 23:51     ` Dan Williams
2018-09-21 23:57 ` [PATCH v8 06/12] nfit/libnvdimm: add set passphrase support for Intel nvdimms David Howells
2018-09-22  0:25   ` Dave Jiang
2018-09-22  1:26     ` Dan Williams
2018-09-22  0:01 ` [PATCH v8 04/12] keys: export lookup_user_key to external users David Howells
2018-09-24 21:02 ` [PATCH v8 02/12] libnvdimm: create keyring to store security keys David Howells
2018-09-24 21:15   ` Dave Jiang

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).