diff options
Diffstat (limited to 'drivers/acpi/nfit')
| -rw-r--r-- | drivers/acpi/nfit/Kconfig | 26 | ||||
| -rw-r--r-- | drivers/acpi/nfit/Makefile | 5 | ||||
| -rw-r--r-- | drivers/acpi/nfit/core.c | 3541 | ||||
| -rw-r--r-- | drivers/acpi/nfit/intel.c | 749 | ||||
| -rw-r--r-- | drivers/acpi/nfit/intel.h | 175 | ||||
| -rw-r--r-- | drivers/acpi/nfit/mce.c | 100 | ||||
| -rw-r--r-- | drivers/acpi/nfit/nfit.h | 352 |
7 files changed, 4948 insertions, 0 deletions
diff --git a/drivers/acpi/nfit/Kconfig b/drivers/acpi/nfit/Kconfig new file mode 100644 index 0000000000..52eefd732c --- /dev/null +++ b/drivers/acpi/nfit/Kconfig @@ -0,0 +1,26 @@ +# SPDX-License-Identifier: GPL-2.0 +config ACPI_NFIT + tristate "ACPI NVDIMM Firmware Interface Table (NFIT)" + depends on PHYS_ADDR_T_64BIT + depends on BLK_DEV + depends on ARCH_HAS_PMEM_API + select LIBNVDIMM + help + Infrastructure to probe ACPI 6 compliant platforms for + NVDIMMs (NFIT) and register a libnvdimm device tree. In + addition to storage devices this also enables libnvdimm to pass + ACPI._DSM messages for platform/dimm configuration. + + To compile this driver as a module, choose M here: + the module will be called nfit. + +config NFIT_SECURITY_DEBUG + bool "Enable debug for NVDIMM security commands" + depends on ACPI_NFIT + help + Some NVDIMM devices and controllers support encryption and + other security features. The payloads for the commands that + enable those features may contain sensitive clear-text + security material. Disable debug of those command payloads + by default. If you are a kernel developer actively working + on NVDIMM security enabling say Y, otherwise say N. diff --git a/drivers/acpi/nfit/Makefile b/drivers/acpi/nfit/Makefile new file mode 100644 index 0000000000..07f53c4e85 --- /dev/null +++ b/drivers/acpi/nfit/Makefile @@ -0,0 +1,5 @@ +# SPDX-License-Identifier: GPL-2.0-only +obj-$(CONFIG_ACPI_NFIT) := nfit.o +nfit-y := core.o +nfit-y += intel.o +nfit-$(CONFIG_X86_MCE) += mce.o diff --git a/drivers/acpi/nfit/core.c b/drivers/acpi/nfit/core.c new file mode 100644 index 0000000000..7d88db451c --- /dev/null +++ b/drivers/acpi/nfit/core.c @@ -0,0 +1,3541 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright(c) 2013-2015 Intel Corporation. All rights reserved. + */ +#include <linux/list_sort.h> +#include <linux/libnvdimm.h> +#include <linux/module.h> +#include <linux/nospec.h> +#include <linux/mutex.h> +#include <linux/ndctl.h> +#include <linux/sysfs.h> +#include <linux/delay.h> +#include <linux/list.h> +#include <linux/acpi.h> +#include <linux/sort.h> +#include <linux/io.h> +#include <linux/nd.h> +#include <asm/cacheflush.h> +#include <acpi/nfit.h> +#include "intel.h" +#include "nfit.h" + +/* + * For readq() and writeq() on 32-bit builds, the hi-lo, lo-hi order is + * irrelevant. + */ +#include <linux/io-64-nonatomic-hi-lo.h> + +static bool force_enable_dimms; +module_param(force_enable_dimms, bool, S_IRUGO|S_IWUSR); +MODULE_PARM_DESC(force_enable_dimms, "Ignore _STA (ACPI DIMM device) status"); + +static bool disable_vendor_specific; +module_param(disable_vendor_specific, bool, S_IRUGO); +MODULE_PARM_DESC(disable_vendor_specific, + "Limit commands to the publicly specified set"); + +static unsigned long override_dsm_mask; +module_param(override_dsm_mask, ulong, S_IRUGO); +MODULE_PARM_DESC(override_dsm_mask, "Bitmask of allowed NVDIMM DSM functions"); + +static int default_dsm_family = -1; +module_param(default_dsm_family, int, S_IRUGO); +MODULE_PARM_DESC(default_dsm_family, + "Try this DSM type first when identifying NVDIMM family"); + +static bool no_init_ars; +module_param(no_init_ars, bool, 0644); +MODULE_PARM_DESC(no_init_ars, "Skip ARS run at nfit init time"); + +static bool force_labels; +module_param(force_labels, bool, 0444); +MODULE_PARM_DESC(force_labels, "Opt-in to labels despite missing methods"); + +LIST_HEAD(acpi_descs); +DEFINE_MUTEX(acpi_desc_lock); + +static struct workqueue_struct *nfit_wq; + +struct nfit_table_prev { + struct list_head spas; + struct list_head memdevs; + struct list_head dcrs; + struct list_head bdws; + struct list_head idts; + struct list_head flushes; +}; + +static guid_t nfit_uuid[NFIT_UUID_MAX]; + +const guid_t *to_nfit_uuid(enum nfit_uuids id) +{ + return &nfit_uuid[id]; +} +EXPORT_SYMBOL(to_nfit_uuid); + +static const guid_t *to_nfit_bus_uuid(int family) +{ + if (WARN_ONCE(family == NVDIMM_BUS_FAMILY_NFIT, + "only secondary bus families can be translated\n")) + return NULL; + /* + * The index of bus UUIDs starts immediately following the last + * NVDIMM/leaf family. + */ + return to_nfit_uuid(family + NVDIMM_FAMILY_MAX); +} + +static struct acpi_device *to_acpi_dev(struct acpi_nfit_desc *acpi_desc) +{ + struct nvdimm_bus_descriptor *nd_desc = &acpi_desc->nd_desc; + + /* + * If provider == 'ACPI.NFIT' we can assume 'dev' is a struct + * acpi_device. + */ + if (!nd_desc->provider_name + || strcmp(nd_desc->provider_name, "ACPI.NFIT") != 0) + return NULL; + + return to_acpi_device(acpi_desc->dev); +} + +static int xlat_bus_status(void *buf, unsigned int cmd, u32 status) +{ + struct nd_cmd_clear_error *clear_err; + struct nd_cmd_ars_status *ars_status; + u16 flags; + + switch (cmd) { + case ND_CMD_ARS_CAP: + if ((status & 0xffff) == NFIT_ARS_CAP_NONE) + return -ENOTTY; + + /* Command failed */ + if (status & 0xffff) + return -EIO; + + /* No supported scan types for this range */ + flags = ND_ARS_PERSISTENT | ND_ARS_VOLATILE; + if ((status >> 16 & flags) == 0) + return -ENOTTY; + return 0; + case ND_CMD_ARS_START: + /* ARS is in progress */ + if ((status & 0xffff) == NFIT_ARS_START_BUSY) + return -EBUSY; + + /* Command failed */ + if (status & 0xffff) + return -EIO; + return 0; + case ND_CMD_ARS_STATUS: + ars_status = buf; + /* Command failed */ + if (status & 0xffff) + return -EIO; + /* Check extended status (Upper two bytes) */ + if (status == NFIT_ARS_STATUS_DONE) + return 0; + + /* ARS is in progress */ + if (status == NFIT_ARS_STATUS_BUSY) + return -EBUSY; + + /* No ARS performed for the current boot */ + if (status == NFIT_ARS_STATUS_NONE) + return -EAGAIN; + + /* + * ARS interrupted, either we overflowed or some other + * agent wants the scan to stop. If we didn't overflow + * then just continue with the returned results. + */ + if (status == NFIT_ARS_STATUS_INTR) { + if (ars_status->out_length >= 40 && (ars_status->flags + & NFIT_ARS_F_OVERFLOW)) + return -ENOSPC; + return 0; + } + + /* Unknown status */ + if (status >> 16) + return -EIO; + return 0; + case ND_CMD_CLEAR_ERROR: + clear_err = buf; + if (status & 0xffff) + return -EIO; + if (!clear_err->cleared) + return -EIO; + if (clear_err->length > clear_err->cleared) + return clear_err->cleared; + return 0; + default: + break; + } + + /* all other non-zero status results in an error */ + if (status) + return -EIO; + return 0; +} + +#define ACPI_LABELS_LOCKED 3 + +static int xlat_nvdimm_status(struct nvdimm *nvdimm, void *buf, unsigned int cmd, + u32 status) +{ + struct nfit_mem *nfit_mem = nvdimm_provider_data(nvdimm); + + switch (cmd) { + case ND_CMD_GET_CONFIG_SIZE: + /* + * In the _LSI, _LSR, _LSW case the locked status is + * communicated via the read/write commands + */ + if (test_bit(NFIT_MEM_LSR, &nfit_mem->flags)) + break; + + if (status >> 16 & ND_CONFIG_LOCKED) + return -EACCES; + break; + case ND_CMD_GET_CONFIG_DATA: + if (test_bit(NFIT_MEM_LSR, &nfit_mem->flags) + && status == ACPI_LABELS_LOCKED) + return -EACCES; + break; + case ND_CMD_SET_CONFIG_DATA: + if (test_bit(NFIT_MEM_LSW, &nfit_mem->flags) + && status == ACPI_LABELS_LOCKED) + return -EACCES; + break; + default: + break; + } + + /* all other non-zero status results in an error */ + if (status) + return -EIO; + return 0; +} + +static int xlat_status(struct nvdimm *nvdimm, void *buf, unsigned int cmd, + u32 status) +{ + if (!nvdimm) + return xlat_bus_status(buf, cmd, status); + return xlat_nvdimm_status(nvdimm, buf, cmd, status); +} + +/* convert _LS{I,R} packages to the buffer object acpi_nfit_ctl expects */ +static union acpi_object *pkg_to_buf(union acpi_object *pkg) +{ + int i; + void *dst; + size_t size = 0; + union acpi_object *buf = NULL; + + if (pkg->type != ACPI_TYPE_PACKAGE) { + WARN_ONCE(1, "BIOS bug, unexpected element type: %d\n", + pkg->type); + goto err; + } + + for (i = 0; i < pkg->package.count; i++) { + union acpi_object *obj = &pkg->package.elements[i]; + + if (obj->type == ACPI_TYPE_INTEGER) + size += 4; + else if (obj->type == ACPI_TYPE_BUFFER) + size += obj->buffer.length; + else { + WARN_ONCE(1, "BIOS bug, unexpected element type: %d\n", + obj->type); + goto err; + } + } + + buf = ACPI_ALLOCATE(sizeof(*buf) + size); + if (!buf) + goto err; + + dst = buf + 1; + buf->type = ACPI_TYPE_BUFFER; + buf->buffer.length = size; + buf->buffer.pointer = dst; + for (i = 0; i < pkg->package.count; i++) { + union acpi_object *obj = &pkg->package.elements[i]; + + if (obj->type == ACPI_TYPE_INTEGER) { + memcpy(dst, &obj->integer.value, 4); + dst += 4; + } else if (obj->type == ACPI_TYPE_BUFFER) { + memcpy(dst, obj->buffer.pointer, obj->buffer.length); + dst += obj->buffer.length; + } + } +err: + ACPI_FREE(pkg); + return buf; +} + +static union acpi_object *int_to_buf(union acpi_object *integer) +{ + union acpi_object *buf = NULL; + void *dst = NULL; + + if (integer->type != ACPI_TYPE_INTEGER) { + WARN_ONCE(1, "BIOS bug, unexpected element type: %d\n", + integer->type); + goto err; + } + + buf = ACPI_ALLOCATE(sizeof(*buf) + 4); + if (!buf) + goto err; + + dst = buf + 1; + buf->type = ACPI_TYPE_BUFFER; + buf->buffer.length = 4; + buf->buffer.pointer = dst; + memcpy(dst, &integer->integer.value, 4); +err: + ACPI_FREE(integer); + return buf; +} + +static union acpi_object *acpi_label_write(acpi_handle handle, u32 offset, + u32 len, void *data) +{ + acpi_status rc; + struct acpi_buffer buf = { ACPI_ALLOCATE_BUFFER, NULL }; + struct acpi_object_list input = { + .count = 3, + .pointer = (union acpi_object []) { + [0] = { + .integer.type = ACPI_TYPE_INTEGER, + .integer.value = offset, + }, + [1] = { + .integer.type = ACPI_TYPE_INTEGER, + .integer.value = len, + }, + [2] = { + .buffer.type = ACPI_TYPE_BUFFER, + .buffer.pointer = data, + .buffer.length = len, + }, + }, + }; + + rc = acpi_evaluate_object(handle, "_LSW", &input, &buf); + if (ACPI_FAILURE(rc)) + return NULL; + return int_to_buf(buf.pointer); +} + +static union acpi_object *acpi_label_read(acpi_handle handle, u32 offset, + u32 len) +{ + acpi_status rc; + struct acpi_buffer buf = { ACPI_ALLOCATE_BUFFER, NULL }; + struct acpi_object_list input = { + .count = 2, + .pointer = (union acpi_object []) { + [0] = { + .integer.type = ACPI_TYPE_INTEGER, + .integer.value = offset, + }, + [1] = { + .integer.type = ACPI_TYPE_INTEGER, + .integer.value = len, + }, + }, + }; + + rc = acpi_evaluate_object(handle, "_LSR", &input, &buf); + if (ACPI_FAILURE(rc)) + return NULL; + return pkg_to_buf(buf.pointer); +} + +static union acpi_object *acpi_label_info(acpi_handle handle) +{ + acpi_status rc; + struct acpi_buffer buf = { ACPI_ALLOCATE_BUFFER, NULL }; + + rc = acpi_evaluate_object(handle, "_LSI", NULL, &buf); + if (ACPI_FAILURE(rc)) + return NULL; + return pkg_to_buf(buf.pointer); +} + +static u8 nfit_dsm_revid(unsigned family, unsigned func) +{ + static const u8 revid_table[NVDIMM_FAMILY_MAX+1][NVDIMM_CMD_MAX+1] = { + [NVDIMM_FAMILY_INTEL] = { + [NVDIMM_INTEL_GET_MODES ... + NVDIMM_INTEL_FW_ACTIVATE_ARM] = 2, + }, + }; + u8 id; + + if (family > NVDIMM_FAMILY_MAX) + return 0; + if (func > NVDIMM_CMD_MAX) + return 0; + id = revid_table[family][func]; + if (id == 0) + return 1; /* default */ + return id; +} + +static bool payload_dumpable(struct nvdimm *nvdimm, unsigned int func) +{ + struct nfit_mem *nfit_mem = nvdimm_provider_data(nvdimm); + + if (nfit_mem && nfit_mem->family == NVDIMM_FAMILY_INTEL + && func >= NVDIMM_INTEL_GET_SECURITY_STATE + && func <= NVDIMM_INTEL_MASTER_SECURE_ERASE) + return IS_ENABLED(CONFIG_NFIT_SECURITY_DEBUG); + return true; +} + +static int cmd_to_func(struct nfit_mem *nfit_mem, unsigned int cmd, + struct nd_cmd_pkg *call_pkg, int *family) +{ + if (call_pkg) { + int i; + + if (nfit_mem && nfit_mem->family != call_pkg->nd_family) + return -ENOTTY; + + for (i = 0; i < ARRAY_SIZE(call_pkg->nd_reserved2); i++) + if (call_pkg->nd_reserved2[i]) + return -EINVAL; + *family = call_pkg->nd_family; + return call_pkg->nd_command; + } + + /* In the !call_pkg case, bus commands == bus functions */ + if (!nfit_mem) + return cmd; + + /* Linux ND commands == NVDIMM_FAMILY_INTEL function numbers */ + if (nfit_mem->family == NVDIMM_FAMILY_INTEL) + return cmd; + + /* + * Force function number validation to fail since 0 is never + * published as a valid function in dsm_mask. + */ + return 0; +} + +int acpi_nfit_ctl(struct nvdimm_bus_descriptor *nd_desc, struct nvdimm *nvdimm, + unsigned int cmd, void *buf, unsigned int buf_len, int *cmd_rc) +{ + struct acpi_nfit_desc *acpi_desc = to_acpi_desc(nd_desc); + struct nfit_mem *nfit_mem = nvdimm_provider_data(nvdimm); + union acpi_object in_obj, in_buf, *out_obj; + const struct nd_cmd_desc *desc = NULL; + struct device *dev = acpi_desc->dev; + struct nd_cmd_pkg *call_pkg = NULL; + const char *cmd_name, *dimm_name; + unsigned long cmd_mask, dsm_mask; + u32 offset, fw_status = 0; + acpi_handle handle; + const guid_t *guid; + int func, rc, i; + int family = 0; + + if (cmd_rc) + *cmd_rc = -EINVAL; + + if (cmd == ND_CMD_CALL) + call_pkg = buf; + func = cmd_to_func(nfit_mem, cmd, call_pkg, &family); + if (func < 0) + return func; + + if (nvdimm) { + struct acpi_device *adev = nfit_mem->adev; + + if (!adev) + return -ENOTTY; + + dimm_name = nvdimm_name(nvdimm); + cmd_name = nvdimm_cmd_name(cmd); + cmd_mask = nvdimm_cmd_mask(nvdimm); + dsm_mask = nfit_mem->dsm_mask; + desc = nd_cmd_dimm_desc(cmd); + guid = to_nfit_uuid(nfit_mem->family); + handle = adev->handle; + } else { + struct acpi_device *adev = to_acpi_dev(acpi_desc); + + cmd_name = nvdimm_bus_cmd_name(cmd); + cmd_mask = nd_desc->cmd_mask; + if (cmd == ND_CMD_CALL && call_pkg->nd_family) { + family = call_pkg->nd_family; + if (family > NVDIMM_BUS_FAMILY_MAX || + !test_bit(family, &nd_desc->bus_family_mask)) + return -EINVAL; + family = array_index_nospec(family, + NVDIMM_BUS_FAMILY_MAX + 1); + dsm_mask = acpi_desc->family_dsm_mask[family]; + guid = to_nfit_bus_uuid(family); + } else { + dsm_mask = acpi_desc->bus_dsm_mask; + guid = to_nfit_uuid(NFIT_DEV_BUS); + } + desc = nd_cmd_bus_desc(cmd); + handle = adev->handle; + dimm_name = "bus"; + } + + if (!desc || (cmd && (desc->out_num + desc->in_num == 0))) + return -ENOTTY; + + /* + * Check for a valid command. For ND_CMD_CALL, we also have to + * make sure that the DSM function is supported. + */ + if (cmd == ND_CMD_CALL && + (func > NVDIMM_CMD_MAX || !test_bit(func, &dsm_mask))) + return -ENOTTY; + else if (!test_bit(cmd, &cmd_mask)) + return -ENOTTY; + + in_obj.type = ACPI_TYPE_PACKAGE; + in_obj.package.count = 1; + in_obj.package.elements = &in_buf; + in_buf.type = ACPI_TYPE_BUFFER; + in_buf.buffer.pointer = buf; + in_buf.buffer.length = 0; + + /* libnvdimm has already validated the input envelope */ + for (i = 0; i < desc->in_num; i++) + in_buf.buffer.length += nd_cmd_in_size(nvdimm, cmd, desc, + i, buf); + + if (call_pkg) { + /* skip over package wrapper */ + in_buf.buffer.pointer = (void *) &call_pkg->nd_payload; + in_buf.buffer.length = call_pkg->nd_size_in; + } + + dev_dbg(dev, "%s cmd: %d: family: %d func: %d input length: %d\n", + dimm_name, cmd, family, func, in_buf.buffer.length); + if (payload_dumpable(nvdimm, func)) + print_hex_dump_debug("nvdimm in ", DUMP_PREFIX_OFFSET, 4, 4, + in_buf.buffer.pointer, + min_t(u32, 256, in_buf.buffer.length), true); + + /* call the BIOS, prefer the named methods over _DSM if available */ + if (nvdimm && cmd == ND_CMD_GET_CONFIG_SIZE + && test_bit(NFIT_MEM_LSR, &nfit_mem->flags)) + out_obj = acpi_label_info(handle); + else if (nvdimm && cmd == ND_CMD_GET_CONFIG_DATA + && test_bit(NFIT_MEM_LSR, &nfit_mem->flags)) { + struct nd_cmd_get_config_data_hdr *p = buf; + + out_obj = acpi_label_read(handle, p->in_offset, p->in_length); + } else if (nvdimm && cmd == ND_CMD_SET_CONFIG_DATA + && test_bit(NFIT_MEM_LSW, &nfit_mem->flags)) { + struct nd_cmd_set_config_hdr *p = buf; + + out_obj = acpi_label_write(handle, p->in_offset, p->in_length, + p->in_buf); + } else { + u8 revid; + + if (nvdimm) + revid = nfit_dsm_revid(nfit_mem->family, func); + else + revid = 1; + out_obj = acpi_evaluate_dsm(handle, guid, revid, func, &in_obj); + } + + if (!out_obj) { + dev_dbg(dev, "%s _DSM failed cmd: %s\n", dimm_name, cmd_name); + return -EINVAL; + } + + if (out_obj->type != ACPI_TYPE_BUFFER) { + dev_dbg(dev, "%s unexpected output object type cmd: %s type: %d\n", + dimm_name, cmd_name, out_obj->type); + rc = -EINVAL; + goto out; + } + + dev_dbg(dev, "%s cmd: %s output length: %d\n", dimm_name, + cmd_name, out_obj->buffer.length); + print_hex_dump_debug(cmd_name, DUMP_PREFIX_OFFSET, 4, 4, + out_obj->buffer.pointer, + min_t(u32, 128, out_obj->buffer.length), true); + + if (call_pkg) { + call_pkg->nd_fw_size = out_obj->buffer.length; + memcpy(call_pkg->nd_payload + call_pkg->nd_size_in, + out_obj->buffer.pointer, + min(call_pkg->nd_fw_size, call_pkg->nd_size_out)); + + ACPI_FREE(out_obj); + /* + * Need to support FW function w/o known size in advance. + * Caller can determine required size based upon nd_fw_size. + * If we return an error (like elsewhere) then caller wouldn't + * be able to rely upon data returned to make calculation. + */ + if (cmd_rc) + *cmd_rc = 0; + return 0; + } + + for (i = 0, offset = 0; i < desc->out_num; i++) { + u32 out_size = nd_cmd_out_size(nvdimm, cmd, desc, i, buf, + (u32 *) out_obj->buffer.pointer, + out_obj->buffer.length - offset); + + if (offset + out_size > out_obj->buffer.length) { + dev_dbg(dev, "%s output object underflow cmd: %s field: %d\n", + dimm_name, cmd_name, i); + break; + } + + if (in_buf.buffer.length + offset + out_size > buf_len) { + dev_dbg(dev, "%s output overrun cmd: %s field: %d\n", + dimm_name, cmd_name, i); + rc = -ENXIO; + goto out; + } + memcpy(buf + in_buf.buffer.length + offset, + out_obj->buffer.pointer + offset, out_size); + offset += out_size; + } + + /* + * Set fw_status for all the commands with a known format to be + * later interpreted by xlat_status(). + */ + if (i >= 1 && ((!nvdimm && cmd >= ND_CMD_ARS_CAP + && cmd <= ND_CMD_CLEAR_ERROR) + || (nvdimm && cmd >= ND_CMD_SMART + && cmd <= ND_CMD_VENDOR))) + fw_status = *(u32 *) out_obj->buffer.pointer; + + if (offset + in_buf.buffer.length < buf_len) { + if (i >= 1) { + /* + * status valid, return the number of bytes left + * unfilled in the output buffer + */ + rc = buf_len - offset - in_buf.buffer.length; + if (cmd_rc) + *cmd_rc = xlat_status(nvdimm, buf, cmd, + fw_status); + } else { + dev_err(dev, "%s:%s underrun cmd: %s buf_len: %d out_len: %d\n", + __func__, dimm_name, cmd_name, buf_len, + offset); + rc = -ENXIO; + } + } else { + rc = 0; + if (cmd_rc) + *cmd_rc = xlat_status(nvdimm, buf, cmd, fw_status); + } + + out: + ACPI_FREE(out_obj); + + return rc; +} +EXPORT_SYMBOL_GPL(acpi_nfit_ctl); + +static const char *spa_type_name(u16 type) +{ + static const char *to_name[] = { + [NFIT_SPA_VOLATILE] = "volatile", + [NFIT_SPA_PM] = "pmem", + [NFIT_SPA_DCR] = "dimm-control-region", + [NFIT_SPA_BDW] = "block-data-window", + [NFIT_SPA_VDISK] = "volatile-disk", + [NFIT_SPA_VCD] = "volatile-cd", + [NFIT_SPA_PDISK] = "persistent-disk", + [NFIT_SPA_PCD] = "persistent-cd", + + }; + + if (type > NFIT_SPA_PCD) + return "unknown"; + + return to_name[type]; +} + +int nfit_spa_type(struct acpi_nfit_system_address *spa) +{ + guid_t guid; + int i; + + import_guid(&guid, spa->range_guid); + for (i = 0; i < NFIT_UUID_MAX; i++) + if (guid_equal(to_nfit_uuid(i), &guid)) + return i; + return -1; +} + +static size_t sizeof_spa(struct acpi_nfit_system_address *spa) +{ + if (spa->flags & ACPI_NFIT_LOCATION_COOKIE_VALID) + return sizeof(*spa); + return sizeof(*spa) - 8; +} + +static bool add_spa(struct acpi_nfit_desc *acpi_desc, + struct nfit_table_prev *prev, + struct acpi_nfit_system_address *spa) +{ + struct device *dev = acpi_desc->dev; + struct nfit_spa *nfit_spa; + + if (spa->header.length != sizeof_spa(spa)) + return false; + + list_for_each_entry(nfit_spa, &prev->spas, list) { + if (memcmp(nfit_spa->spa, spa, sizeof_spa(spa)) == 0) { + list_move_tail(&nfit_spa->list, &acpi_desc->spas); + return true; + } + } + + nfit_spa = devm_kzalloc(dev, sizeof(*nfit_spa) + sizeof_spa(spa), + GFP_KERNEL); + if (!nfit_spa) + return false; + INIT_LIST_HEAD(&nfit_spa->list); + memcpy(nfit_spa->spa, spa, sizeof_spa(spa)); + list_add_tail(&nfit_spa->list, &acpi_desc->spas); + dev_dbg(dev, "spa index: %d type: %s\n", + spa->range_index, + spa_type_name(nfit_spa_type(spa))); + return true; +} + +static bool add_memdev(struct acpi_nfit_desc *acpi_desc, + struct nfit_table_prev *prev, + struct acpi_nfit_memory_map *memdev) +{ + struct device *dev = acpi_desc->dev; + struct nfit_memdev *nfit_memdev; + + if (memdev->header.length != sizeof(*memdev)) + return false; + + list_for_each_entry(nfit_memdev, &prev->memdevs, list) + if (memcmp(nfit_memdev->memdev, memdev, sizeof(*memdev)) == 0) { + list_move_tail(&nfit_memdev->list, &acpi_desc->memdevs); + return true; + } + + nfit_memdev = devm_kzalloc(dev, sizeof(*nfit_memdev) + sizeof(*memdev), + GFP_KERNEL); + if (!nfit_memdev) + return false; + INIT_LIST_HEAD(&nfit_memdev->list); + memcpy(nfit_memdev->memdev, memdev, sizeof(*memdev)); + list_add_tail(&nfit_memdev->list, &acpi_desc->memdevs); + dev_dbg(dev, "memdev handle: %#x spa: %d dcr: %d flags: %#x\n", + memdev->device_handle, memdev->range_index, + memdev->region_index, memdev->flags); + return true; +} + +int nfit_get_smbios_id(u32 device_handle, u16 *flags) +{ + struct acpi_nfit_memory_map *memdev; + struct acpi_nfit_desc *acpi_desc; + struct nfit_mem *nfit_mem; + u16 physical_id; + + mutex_lock(&acpi_desc_lock); + list_for_each_entry(acpi_desc, &acpi_descs, list) { + mutex_lock(&acpi_desc->init_mutex); + list_for_each_entry(nfit_mem, &acpi_desc->dimms, list) { + memdev = __to_nfit_memdev(nfit_mem); + if (memdev->device_handle == device_handle) { + *flags = memdev->flags; + physical_id = memdev->physical_id; + mutex_unlock(&acpi_desc->init_mutex); + mutex_unlock(&acpi_desc_lock); + return physical_id; + } + } + mutex_unlock(&acpi_desc->init_mutex); + } + mutex_unlock(&acpi_desc_lock); + + return -ENODEV; +} +EXPORT_SYMBOL_GPL(nfit_get_smbios_id); + +/* + * An implementation may provide a truncated control region if no block windows + * are defined. + */ +static size_t sizeof_dcr(struct acpi_nfit_control_region *dcr) +{ + if (dcr->header.length < offsetof(struct acpi_nfit_control_region, + window_size)) + return 0; + if (dcr->windows) + return sizeof(*dcr); + return offsetof(struct acpi_nfit_control_region, window_size); +} + +static bool add_dcr(struct acpi_nfit_desc *acpi_desc, + struct nfit_table_prev *prev, + struct acpi_nfit_control_region *dcr) +{ + struct device *dev = acpi_desc->dev; + struct nfit_dcr *nfit_dcr; + + if (!sizeof_dcr(dcr)) + return false; + + list_for_each_entry(nfit_dcr, &prev->dcrs, list) + if (memcmp(nfit_dcr->dcr, dcr, sizeof_dcr(dcr)) == 0) { + list_move_tail(&nfit_dcr->list, &acpi_desc->dcrs); + return true; + } + + nfit_dcr = devm_kzalloc(dev, sizeof(*nfit_dcr) + sizeof(*dcr), + GFP_KERNEL); + if (!nfit_dcr) + return false; + INIT_LIST_HEAD(&nfit_dcr->list); + memcpy(nfit_dcr->dcr, dcr, sizeof_dcr(dcr)); + list_add_tail(&nfit_dcr->list, &acpi_desc->dcrs); + dev_dbg(dev, "dcr index: %d windows: %d\n", + dcr->region_index, dcr->windows); + return true; +} + +static bool add_bdw(struct acpi_nfit_desc *acpi_desc, + struct nfit_table_prev *prev, + struct acpi_nfit_data_region *bdw) +{ + struct device *dev = acpi_desc->dev; + struct nfit_bdw *nfit_bdw; + + if (bdw->header.length != sizeof(*bdw)) + return false; + list_for_each_entry(nfit_bdw, &prev->bdws, list) + if (memcmp(nfit_bdw->bdw, bdw, sizeof(*bdw)) == 0) { + list_move_tail(&nfit_bdw->list, &acpi_desc->bdws); + return true; + } + + nfit_bdw = devm_kzalloc(dev, sizeof(*nfit_bdw) + sizeof(*bdw), + GFP_KERNEL); + if (!nfit_bdw) + return false; + INIT_LIST_HEAD(&nfit_bdw->list); + memcpy(nfit_bdw->bdw, bdw, sizeof(*bdw)); + list_add_tail(&nfit_bdw->list, &acpi_desc->bdws); + dev_dbg(dev, "bdw dcr: %d windows: %d\n", + bdw->region_index, bdw->windows); + return true; +} + +static size_t sizeof_idt(struct acpi_nfit_interleave *idt) +{ + if (idt->header.length < sizeof(*idt)) + return 0; + return sizeof(*idt) + sizeof(u32) * idt->line_count; +} + +static bool add_idt(struct acpi_nfit_desc *acpi_desc, + struct nfit_table_prev *prev, + struct acpi_nfit_interleave *idt) +{ + struct device *dev = acpi_desc->dev; + struct nfit_idt *nfit_idt; + + if (!sizeof_idt(idt)) + return false; + + list_for_each_entry(nfit_idt, &prev->idts, list) { + if (sizeof_idt(nfit_idt->idt) != sizeof_idt(idt)) + continue; + + if (memcmp(nfit_idt->idt, idt, sizeof_idt(idt)) == 0) { + list_move_tail(&nfit_idt->list, &acpi_desc->idts); + return true; + } + } + + nfit_idt = devm_kzalloc(dev, sizeof(*nfit_idt) + sizeof_idt(idt), + GFP_KERNEL); + if (!nfit_idt) + return false; + INIT_LIST_HEAD(&nfit_idt->list); + memcpy(nfit_idt->idt, idt, sizeof_idt(idt)); + list_add_tail(&nfit_idt->list, &acpi_desc->idts); + dev_dbg(dev, "idt index: %d num_lines: %d\n", + idt->interleave_index, idt->line_count); + return true; +} + +static size_t sizeof_flush(struct acpi_nfit_flush_address *flush) +{ + if (flush->header.length < sizeof(*flush)) + return 0; + return struct_size(flush, hint_address, flush->hint_count); +} + +static bool add_flush(struct acpi_nfit_desc *acpi_desc, + struct nfit_table_prev *prev, + struct acpi_nfit_flush_address *flush) +{ + struct device *dev = acpi_desc->dev; + struct nfit_flush *nfit_flush; + + if (!sizeof_flush(flush)) + return false; + + list_for_each_entry(nfit_flush, &prev->flushes, list) { + if (sizeof_flush(nfit_flush->flush) != sizeof_flush(flush)) + continue; + + if (memcmp(nfit_flush->flush, flush, + sizeof_flush(flush)) == 0) { + list_move_tail(&nfit_flush->list, &acpi_desc->flushes); + return true; + } + } + + nfit_flush = devm_kzalloc(dev, sizeof(*nfit_flush) + + sizeof_flush(flush), GFP_KERNEL); + if (!nfit_flush) + return false; + INIT_LIST_HEAD(&nfit_flush->list); + memcpy(nfit_flush->flush, flush, sizeof_flush(flush)); + list_add_tail(&nfit_flush->list, &acpi_desc->flushes); + dev_dbg(dev, "nfit_flush handle: %d hint_count: %d\n", + flush->device_handle, flush->hint_count); + return true; +} + +static bool add_platform_cap(struct acpi_nfit_desc *acpi_desc, + struct acpi_nfit_capabilities *pcap) +{ + struct device *dev = acpi_desc->dev; + u32 mask; + + mask = (1 << (pcap->highest_capability + 1)) - 1; + acpi_desc->platform_cap = pcap->capabilities & mask; + dev_dbg(dev, "cap: %#x\n", acpi_desc->platform_cap); + return true; +} + +static void *add_table(struct acpi_nfit_desc *acpi_desc, + struct nfit_table_prev *prev, void *table, const void *end) +{ + struct device *dev = acpi_desc->dev; + struct acpi_nfit_header *hdr; + void *err = ERR_PTR(-ENOMEM); + + if (table >= end) + return NULL; + + hdr = table; + if (!hdr->length) { + dev_warn(dev, "found a zero length table '%d' parsing nfit\n", + hdr->type); + return NULL; + } + + switch (hdr->type) { + case ACPI_NFIT_TYPE_SYSTEM_ADDRESS: + if (!add_spa(acpi_desc, prev, table)) + return err; + break; + case ACPI_NFIT_TYPE_MEMORY_MAP: + if (!add_memdev(acpi_desc, prev, table)) + return err; + break; + case ACPI_NFIT_TYPE_CONTROL_REGION: + if (!add_dcr(acpi_desc, prev, table)) + return err; + break; + case ACPI_NFIT_TYPE_DATA_REGION: + if (!add_bdw(acpi_desc, prev, table)) + return err; + break; + case ACPI_NFIT_TYPE_INTERLEAVE: + if (!add_idt(acpi_desc, prev, table)) + return err; + break; + case ACPI_NFIT_TYPE_FLUSH_ADDRESS: + if (!add_flush(acpi_desc, prev, table)) + return err; + break; + case ACPI_NFIT_TYPE_SMBIOS: + dev_dbg(dev, "smbios\n"); + break; + case ACPI_NFIT_TYPE_CAPABILITIES: + if (!add_platform_cap(acpi_desc, table)) + return err; + break; + default: + dev_err(dev, "unknown table '%d' parsing nfit\n", hdr->type); + break; + } + + return table + hdr->length; +} + +static int __nfit_mem_init(struct acpi_nfit_desc *acpi_desc, + struct acpi_nfit_system_address *spa) +{ + struct nfit_mem *nfit_mem, *found; + struct nfit_memdev *nfit_memdev; + int type = spa ? nfit_spa_type(spa) : 0; + + switch (type) { + case NFIT_SPA_DCR: + case NFIT_SPA_PM: + break; + default: + if (spa) + return 0; + } + + /* + * This loop runs in two modes, when a dimm is mapped the loop + * adds memdev associations to an existing dimm, or creates a + * dimm. In the unmapped dimm case this loop sweeps for memdev + * instances with an invalid / zero range_index and adds those + * dimms without spa associations. + */ + list_for_each_entry(nfit_memdev, &acpi_desc->memdevs, list) { + struct nfit_flush *nfit_flush; + struct nfit_dcr *nfit_dcr; + u32 device_handle; + u16 dcr; + + if (spa && nfit_memdev->memdev->range_index != spa->range_index) + continue; + if (!spa && nfit_memdev->memdev->range_index) + continue; + found = NULL; + dcr = nfit_memdev->memdev->region_index; + device_handle = nfit_memdev->memdev->device_handle; + list_for_each_entry(nfit_mem, &acpi_desc->dimms, list) + if (__to_nfit_memdev(nfit_mem)->device_handle + == device_handle) { + found = nfit_mem; + break; + } + + if (found) + nfit_mem = found; + else { + nfit_mem = devm_kzalloc(acpi_desc->dev, + sizeof(*nfit_mem), GFP_KERNEL); + if (!nfit_mem) + return -ENOMEM; + INIT_LIST_HEAD(&nfit_mem->list); + nfit_mem->acpi_desc = acpi_desc; + list_add(&nfit_mem->list, &acpi_desc->dimms); + } + + list_for_each_entry(nfit_dcr, &acpi_desc->dcrs, list) { + if (nfit_dcr->dcr->region_index != dcr) + continue; + /* + * Record the control region for the dimm. For + * the ACPI 6.1 case, where there are separate + * control regions for the pmem vs blk + * interfaces, be sure to record the extended + * blk details. + */ + if (!nfit_mem->dcr) + nfit_mem->dcr = nfit_dcr->dcr; + else if (nfit_mem->dcr->windows == 0 + && nfit_dcr->dcr->windows) + nfit_mem->dcr = nfit_dcr->dcr; + break; + } + + list_for_each_entry(nfit_flush, &acpi_desc->flushes, list) { + struct acpi_nfit_flush_address *flush; + u16 i; + + if (nfit_flush->flush->device_handle != device_handle) + continue; + nfit_mem->nfit_flush = nfit_flush; + flush = nfit_flush->flush; + nfit_mem->flush_wpq = devm_kcalloc(acpi_desc->dev, + flush->hint_count, + sizeof(struct resource), + GFP_KERNEL); + if (!nfit_mem->flush_wpq) + return -ENOMEM; + for (i = 0; i < flush->hint_count; i++) { + struct resource *res = &nfit_mem->flush_wpq[i]; + + res->start = flush->hint_address[i]; + res->end = res->start + 8 - 1; + } + break; + } + + if (dcr && !nfit_mem->dcr) { + dev_err(acpi_desc->dev, "SPA %d missing DCR %d\n", + spa->range_index, dcr); + return -ENODEV; + } + + if (type == NFIT_SPA_DCR) { + struct nfit_idt *nfit_idt; + u16 idt_idx; + + /* multiple dimms may share a SPA when interleaved */ + nfit_mem->spa_dcr = spa; + nfit_mem->memdev_dcr = nfit_memdev->memdev; + idt_idx = nfit_memdev->memdev->interleave_index; + list_for_each_entry(nfit_idt, &acpi_desc->idts, list) { + if (nfit_idt->idt->interleave_index != idt_idx) + continue; + nfit_mem->idt_dcr = nfit_idt->idt; + break; + } + } else if (type == NFIT_SPA_PM) { + /* + * A single dimm may belong to multiple SPA-PM + * ranges, record at least one in addition to + * any SPA-DCR range. + */ + nfit_mem->memdev_pmem = nfit_memdev->memdev; + } else + nfit_mem->memdev_dcr = nfit_memdev->memdev; + } + + return 0; +} + +static int nfit_mem_cmp(void *priv, const struct list_head *_a, + const struct list_head *_b) +{ + struct nfit_mem *a = container_of(_a, typeof(*a), list); + struct nfit_mem *b = container_of(_b, typeof(*b), list); + u32 handleA, handleB; + + handleA = __to_nfit_memdev(a)->device_handle; + handleB = __to_nfit_memdev(b)->device_handle; + if (handleA < handleB) + return -1; + else if (handleA > handleB) + return 1; + return 0; +} + +static int nfit_mem_init(struct acpi_nfit_desc *acpi_desc) +{ + struct nfit_spa *nfit_spa; + int rc; + + + /* + * For each SPA-DCR or SPA-PMEM address range find its + * corresponding MEMDEV(s). From each MEMDEV find the + * corresponding DCR. Then, if we're operating on a SPA-DCR, + * try to find a SPA-BDW and a corresponding BDW that references + * the DCR. Throw it all into an nfit_mem object. Note, that + * BDWs are optional. + */ + list_for_each_entry(nfit_spa, &acpi_desc->spas, list) { + rc = __nfit_mem_init(acpi_desc, nfit_spa->spa); + if (rc) + return rc; + } + + /* + * If a DIMM has failed to be mapped into SPA there will be no + * SPA entries above. Find and register all the unmapped DIMMs + * for reporting and recovery purposes. + */ + rc = __nfit_mem_init(acpi_desc, NULL); + if (rc) + return rc; + + list_sort(NULL, &acpi_desc->dimms, nfit_mem_cmp); + + return 0; +} + +static ssize_t bus_dsm_mask_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct nvdimm_bus *nvdimm_bus = to_nvdimm_bus(dev); + struct nvdimm_bus_descriptor *nd_desc = to_nd_desc(nvdimm_bus); + struct acpi_nfit_desc *acpi_desc = to_acpi_desc(nd_desc); + + return sprintf(buf, "%#lx\n", acpi_desc->bus_dsm_mask); +} +static struct device_attribute dev_attr_bus_dsm_mask = + __ATTR(dsm_mask, 0444, bus_dsm_mask_show, NULL); + +static ssize_t revision_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct nvdimm_bus *nvdimm_bus = to_nvdimm_bus(dev); + struct nvdimm_bus_descriptor *nd_desc = to_nd_desc(nvdimm_bus); + struct acpi_nfit_desc *acpi_desc = to_acpi_desc(nd_desc); + + return sprintf(buf, "%d\n", acpi_desc->acpi_header.revision); +} +static DEVICE_ATTR_RO(revision); + +static ssize_t hw_error_scrub_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct nvdimm_bus *nvdimm_bus = to_nvdimm_bus(dev); + struct nvdimm_bus_descriptor *nd_desc = to_nd_desc(nvdimm_bus); + struct acpi_nfit_desc *acpi_desc = to_acpi_desc(nd_desc); + + return sprintf(buf, "%d\n", acpi_desc->scrub_mode); +} + +/* + * The 'hw_error_scrub' attribute can have the following values written to it: + * '0': Switch to the default mode where an exception will only insert + * the address of the memory error into the poison and badblocks lists. + * '1': Enable a full scrub to happen if an exception for a memory error is + * received. + */ +static ssize_t hw_error_scrub_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t size) +{ + struct nvdimm_bus_descriptor *nd_desc; + ssize_t rc; + long val; + + rc = kstrtol(buf, 0, &val); + if (rc) + return rc; + + device_lock(dev); + nd_desc = dev_get_drvdata(dev); + if (nd_desc) { + struct acpi_nfit_desc *acpi_desc = to_acpi_desc(nd_desc); + + switch (val) { + case HW_ERROR_SCRUB_ON: + acpi_desc->scrub_mode = HW_ERROR_SCRUB_ON; + break; + case HW_ERROR_SCRUB_OFF: + acpi_desc->scrub_mode = HW_ERROR_SCRUB_OFF; + break; + default: + rc = -EINVAL; + break; + } + } + device_unlock(dev); + if (rc) + return rc; + return size; +} +static DEVICE_ATTR_RW(hw_error_scrub); + +/* + * This shows the number of full Address Range Scrubs that have been + * completed since driver load time. Userspace can wait on this using + * select/poll etc. A '+' at the end indicates an ARS is in progress + */ +static ssize_t scrub_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct nvdimm_bus_descriptor *nd_desc; + struct acpi_nfit_desc *acpi_desc; + ssize_t rc = -ENXIO; + bool busy; + + device_lock(dev); + nd_desc = dev_get_drvdata(dev); + if (!nd_desc) { + device_unlock(dev); + return rc; + } + acpi_desc = to_acpi_desc(nd_desc); + + mutex_lock(&acpi_desc->init_mutex); + busy = test_bit(ARS_BUSY, &acpi_desc->scrub_flags) + && !test_bit(ARS_CANCEL, &acpi_desc->scrub_flags); + rc = sprintf(buf, "%d%s", acpi_desc->scrub_count, busy ? "+\n" : "\n"); + /* Allow an admin to poll the busy state at a higher rate */ + if (busy && capable(CAP_SYS_RAWIO) && !test_and_set_bit(ARS_POLL, + &acpi_desc->scrub_flags)) { + acpi_desc->scrub_tmo = 1; + mod_delayed_work(nfit_wq, &acpi_desc->dwork, HZ); + } + + mutex_unlock(&acpi_desc->init_mutex); + device_unlock(dev); + return rc; +} + +static ssize_t scrub_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t size) +{ + struct nvdimm_bus_descriptor *nd_desc; + ssize_t rc; + long val; + + rc = kstrtol(buf, 0, &val); + if (rc) + return rc; + if (val != 1) + return -EINVAL; + + device_lock(dev); + nd_desc = dev_get_drvdata(dev); + if (nd_desc) { + struct acpi_nfit_desc *acpi_desc = to_acpi_desc(nd_desc); + + rc = acpi_nfit_ars_rescan(acpi_desc, ARS_REQ_LONG); + } + device_unlock(dev); + if (rc) + return rc; + return size; +} +static DEVICE_ATTR_RW(scrub); + +static bool ars_supported(struct nvdimm_bus *nvdimm_bus) +{ + struct nvdimm_bus_descriptor *nd_desc = to_nd_desc(nvdimm_bus); + const unsigned long mask = 1 << ND_CMD_ARS_CAP | 1 << ND_CMD_ARS_START + | 1 << ND_CMD_ARS_STATUS; + + return (nd_desc->cmd_mask & mask) == mask; +} + +static umode_t nfit_visible(struct kobject *kobj, struct attribute *a, int n) +{ + struct device *dev = kobj_to_dev(kobj); + struct nvdimm_bus *nvdimm_bus = to_nvdimm_bus(dev); + + if (a == &dev_attr_scrub.attr) + return ars_supported(nvdimm_bus) ? a->mode : 0; + + if (a == &dev_attr_firmware_activate_noidle.attr) + return intel_fwa_supported(nvdimm_bus) ? a->mode : 0; + + return a->mode; +} + +static struct attribute *acpi_nfit_attributes[] = { + &dev_attr_revision.attr, + &dev_attr_scrub.attr, + &dev_attr_hw_error_scrub.attr, + &dev_attr_bus_dsm_mask.attr, + &dev_attr_firmware_activate_noidle.attr, + NULL, +}; + +static const struct attribute_group acpi_nfit_attribute_group = { + .name = "nfit", + .attrs = acpi_nfit_attributes, + .is_visible = nfit_visible, +}; + +static const struct attribute_group *acpi_nfit_attribute_groups[] = { + &acpi_nfit_attribute_group, + NULL, +}; + +static struct acpi_nfit_memory_map *to_nfit_memdev(struct device *dev) +{ + struct nvdimm *nvdimm = to_nvdimm(dev); + struct nfit_mem *nfit_mem = nvdimm_provider_data(nvdimm); + + return __to_nfit_memdev(nfit_mem); +} + +static struct acpi_nfit_control_region *to_nfit_dcr(struct device *dev) +{ + struct nvdimm *nvdimm = to_nvdimm(dev); + struct nfit_mem *nfit_mem = nvdimm_provider_data(nvdimm); + + return nfit_mem->dcr; +} + +static ssize_t handle_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct acpi_nfit_memory_map *memdev = to_nfit_memdev(dev); + + return sprintf(buf, "%#x\n", memdev->device_handle); +} +static DEVICE_ATTR_RO(handle); + +static ssize_t phys_id_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct acpi_nfit_memory_map *memdev = to_nfit_memdev(dev); + + return sprintf(buf, "%#x\n", memdev->physical_id); +} +static DEVICE_ATTR_RO(phys_id); + +static ssize_t vendor_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct acpi_nfit_control_region *dcr = to_nfit_dcr(dev); + + return sprintf(buf, "0x%04x\n", be16_to_cpu(dcr->vendor_id)); +} +static DEVICE_ATTR_RO(vendor); + +static ssize_t rev_id_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct acpi_nfit_control_region *dcr = to_nfit_dcr(dev); + + return sprintf(buf, "0x%04x\n", be16_to_cpu(dcr->revision_id)); +} +static DEVICE_ATTR_RO(rev_id); + +static ssize_t device_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct acpi_nfit_control_region *dcr = to_nfit_dcr(dev); + + return sprintf(buf, "0x%04x\n", be16_to_cpu(dcr->device_id)); +} +static DEVICE_ATTR_RO(device); + +static ssize_t subsystem_vendor_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct acpi_nfit_control_region *dcr = to_nfit_dcr(dev); + + return sprintf(buf, "0x%04x\n", be16_to_cpu(dcr->subsystem_vendor_id)); +} +static DEVICE_ATTR_RO(subsystem_vendor); + +static ssize_t subsystem_rev_id_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct acpi_nfit_control_region *dcr = to_nfit_dcr(dev); + + return sprintf(buf, "0x%04x\n", + be16_to_cpu(dcr->subsystem_revision_id)); +} +static DEVICE_ATTR_RO(subsystem_rev_id); + +static ssize_t subsystem_device_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct acpi_nfit_control_region *dcr = to_nfit_dcr(dev); + + return sprintf(buf, "0x%04x\n", be16_to_cpu(dcr->subsystem_device_id)); +} +static DEVICE_ATTR_RO(subsystem_device); + +static int num_nvdimm_formats(struct nvdimm *nvdimm) +{ + struct nfit_mem *nfit_mem = nvdimm_provider_data(nvdimm); + int formats = 0; + + if (nfit_mem->memdev_pmem) + formats++; + return formats; +} + +static ssize_t format_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct acpi_nfit_control_region *dcr = to_nfit_dcr(dev); + + return sprintf(buf, "0x%04x\n", le16_to_cpu(dcr->code)); +} +static DEVICE_ATTR_RO(format); + +static ssize_t format1_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + u32 handle; + ssize_t rc = -ENXIO; + struct nfit_mem *nfit_mem; + struct nfit_memdev *nfit_memdev; + struct acpi_nfit_desc *acpi_desc; + struct nvdimm *nvdimm = to_nvdimm(dev); + struct acpi_nfit_control_region *dcr = to_nfit_dcr(dev); + + nfit_mem = nvdimm_provider_data(nvdimm); + acpi_desc = nfit_mem->acpi_desc; + handle = to_nfit_memdev(dev)->device_handle; + + /* assumes DIMMs have at most 2 published interface codes */ + mutex_lock(&acpi_desc->init_mutex); + list_for_each_entry(nfit_memdev, &acpi_desc->memdevs, list) { + struct acpi_nfit_memory_map *memdev = nfit_memdev->memdev; + struct nfit_dcr *nfit_dcr; + + if (memdev->device_handle != handle) + continue; + + list_for_each_entry(nfit_dcr, &acpi_desc->dcrs, list) { + if (nfit_dcr->dcr->region_index != memdev->region_index) + continue; + if (nfit_dcr->dcr->code == dcr->code) + continue; + rc = sprintf(buf, "0x%04x\n", + le16_to_cpu(nfit_dcr->dcr->code)); + break; + } + if (rc != -ENXIO) + break; + } + mutex_unlock(&acpi_desc->init_mutex); + return rc; +} +static DEVICE_ATTR_RO(format1); + +static ssize_t formats_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct nvdimm *nvdimm = to_nvdimm(dev); + + return sprintf(buf, "%d\n", num_nvdimm_formats(nvdimm)); +} +static DEVICE_ATTR_RO(formats); + +static ssize_t serial_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct acpi_nfit_control_region *dcr = to_nfit_dcr(dev); + + return sprintf(buf, "0x%08x\n", be32_to_cpu(dcr->serial_number)); +} +static DEVICE_ATTR_RO(serial); + +static ssize_t family_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct nvdimm *nvdimm = to_nvdimm(dev); + struct nfit_mem *nfit_mem = nvdimm_provider_data(nvdimm); + + if (nfit_mem->family < 0) + return -ENXIO; + return sprintf(buf, "%d\n", nfit_mem->family); +} +static DEVICE_ATTR_RO(family); + +static ssize_t dsm_mask_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct nvdimm *nvdimm = to_nvdimm(dev); + struct nfit_mem *nfit_mem = nvdimm_provider_data(nvdimm); + + if (nfit_mem->family < 0) + return -ENXIO; + return sprintf(buf, "%#lx\n", nfit_mem->dsm_mask); +} +static DEVICE_ATTR_RO(dsm_mask); + +static ssize_t flags_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct nvdimm *nvdimm = to_nvdimm(dev); + struct nfit_mem *nfit_mem = nvdimm_provider_data(nvdimm); + u16 flags = __to_nfit_memdev(nfit_mem)->flags; + + if (test_bit(NFIT_MEM_DIRTY, &nfit_mem->flags)) + flags |= ACPI_NFIT_MEM_FLUSH_FAILED; + + return sprintf(buf, "%s%s%s%s%s%s%s\n", + flags & ACPI_NFIT_MEM_SAVE_FAILED ? "save_fail " : "", + flags & ACPI_NFIT_MEM_RESTORE_FAILED ? "restore_fail " : "", + flags & ACPI_NFIT_MEM_FLUSH_FAILED ? "flush_fail " : "", + flags & ACPI_NFIT_MEM_NOT_ARMED ? "not_armed " : "", + flags & ACPI_NFIT_MEM_HEALTH_OBSERVED ? "smart_event " : "", + flags & ACPI_NFIT_MEM_MAP_FAILED ? "map_fail " : "", + flags & ACPI_NFIT_MEM_HEALTH_ENABLED ? "smart_notify " : ""); +} +static DEVICE_ATTR_RO(flags); + +static ssize_t id_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct nvdimm *nvdimm = to_nvdimm(dev); + struct nfit_mem *nfit_mem = nvdimm_provider_data(nvdimm); + + return sprintf(buf, "%s\n", nfit_mem->id); +} +static DEVICE_ATTR_RO(id); + +static ssize_t dirty_shutdown_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct nvdimm *nvdimm = to_nvdimm(dev); + struct nfit_mem *nfit_mem = nvdimm_provider_data(nvdimm); + + return sprintf(buf, "%d\n", nfit_mem->dirty_shutdown); +} +static DEVICE_ATTR_RO(dirty_shutdown); + +static struct attribute *acpi_nfit_dimm_attributes[] = { + &dev_attr_handle.attr, + &dev_attr_phys_id.attr, + &dev_attr_vendor.attr, + &dev_attr_device.attr, + &dev_attr_rev_id.attr, + &dev_attr_subsystem_vendor.attr, + &dev_attr_subsystem_device.attr, + &dev_attr_subsystem_rev_id.attr, + &dev_attr_format.attr, + &dev_attr_formats.attr, + &dev_attr_format1.attr, + &dev_attr_serial.attr, + &dev_attr_flags.attr, + &dev_attr_id.attr, + &dev_attr_family.attr, + &dev_attr_dsm_mask.attr, + &dev_attr_dirty_shutdown.attr, + NULL, +}; + +static umode_t acpi_nfit_dimm_attr_visible(struct kobject *kobj, + struct attribute *a, int n) +{ + struct device *dev = kobj_to_dev(kobj); + struct nvdimm *nvdimm = to_nvdimm(dev); + struct nfit_mem *nfit_mem = nvdimm_provider_data(nvdimm); + + if (!to_nfit_dcr(dev)) { + /* Without a dcr only the memdev attributes can be surfaced */ + if (a == &dev_attr_handle.attr || a == &dev_attr_phys_id.attr + || a == &dev_attr_flags.attr + || a == &dev_attr_family.attr + || a == &dev_attr_dsm_mask.attr) + return a->mode; + return 0; + } + + if (a == &dev_attr_format1.attr && num_nvdimm_formats(nvdimm) <= 1) + return 0; + + if (!test_bit(NFIT_MEM_DIRTY_COUNT, &nfit_mem->flags) + && a == &dev_attr_dirty_shutdown.attr) + return 0; + + return a->mode; +} + +static const struct attribute_group acpi_nfit_dimm_attribute_group = { + .name = "nfit", + .attrs = acpi_nfit_dimm_attributes, + .is_visible = acpi_nfit_dimm_attr_visible, +}; + +static const struct attribute_group *acpi_nfit_dimm_attribute_groups[] = { + &acpi_nfit_dimm_attribute_group, + NULL, +}; + +static struct nvdimm *acpi_nfit_dimm_by_handle(struct acpi_nfit_desc *acpi_desc, + u32 device_handle) +{ + struct nfit_mem *nfit_mem; + + list_for_each_entry(nfit_mem, &acpi_desc->dimms, list) + if (__to_nfit_memdev(nfit_mem)->device_handle == device_handle) + return nfit_mem->nvdimm; + + return NULL; +} + +void __acpi_nvdimm_notify(struct device *dev, u32 event) +{ + struct nfit_mem *nfit_mem; + struct acpi_nfit_desc *acpi_desc; + + dev_dbg(dev->parent, "%s: event: %d\n", dev_name(dev), + event); + + if (event != NFIT_NOTIFY_DIMM_HEALTH) { + dev_dbg(dev->parent, "%s: unknown event: %d\n", dev_name(dev), + event); + return; + } + + acpi_desc = dev_get_drvdata(dev->parent); + if (!acpi_desc) + return; + + /* + * If we successfully retrieved acpi_desc, then we know nfit_mem data + * is still valid. + */ + nfit_mem = dev_get_drvdata(dev); + if (nfit_mem && nfit_mem->flags_attr) + sysfs_notify_dirent(nfit_mem->flags_attr); +} +EXPORT_SYMBOL_GPL(__acpi_nvdimm_notify); + +static void acpi_nvdimm_notify(acpi_handle handle, u32 event, void *data) +{ + struct acpi_device *adev = data; + struct device *dev = &adev->dev; + + device_lock(dev->parent); + __acpi_nvdimm_notify(dev, event); + device_unlock(dev->parent); +} + +static bool acpi_nvdimm_has_method(struct acpi_device *adev, char *method) +{ + acpi_handle handle; + acpi_status status; + + status = acpi_get_handle(adev->handle, method, &handle); + + if (ACPI_SUCCESS(status)) + return true; + return false; +} + +__weak void nfit_intel_shutdown_status(struct nfit_mem *nfit_mem) +{ + struct device *dev = &nfit_mem->adev->dev; + struct nd_intel_smart smart = { 0 }; + union acpi_object in_buf = { + .buffer.type = ACPI_TYPE_BUFFER, + .buffer.length = 0, + }; + union acpi_object in_obj = { + .package.type = ACPI_TYPE_PACKAGE, + .package.count = 1, + .package.elements = &in_buf, + }; + const u8 func = ND_INTEL_SMART; + const guid_t *guid = to_nfit_uuid(nfit_mem->family); + u8 revid = nfit_dsm_revid(nfit_mem->family, func); + struct acpi_device *adev = nfit_mem->adev; + acpi_handle handle = adev->handle; + union acpi_object *out_obj; + + if ((nfit_mem->dsm_mask & (1 << func)) == 0) + return; + + out_obj = acpi_evaluate_dsm(handle, guid, revid, func, &in_obj); + if (!out_obj || out_obj->type != ACPI_TYPE_BUFFER + || out_obj->buffer.length < sizeof(smart)) { + dev_dbg(dev->parent, "%s: failed to retrieve initial health\n", + dev_name(dev)); + ACPI_FREE(out_obj); + return; + } + memcpy(&smart, out_obj->buffer.pointer, sizeof(smart)); + ACPI_FREE(out_obj); + + if (smart.flags & ND_INTEL_SMART_SHUTDOWN_VALID) { + if (smart.shutdown_state) + set_bit(NFIT_MEM_DIRTY, &nfit_mem->flags); + } + + if (smart.flags & ND_INTEL_SMART_SHUTDOWN_COUNT_VALID) { + set_bit(NFIT_MEM_DIRTY_COUNT, &nfit_mem->flags); + nfit_mem->dirty_shutdown = smart.shutdown_count; + } +} + +static void populate_shutdown_status(struct nfit_mem *nfit_mem) +{ + /* + * For DIMMs that provide a dynamic facility to retrieve a + * dirty-shutdown status and/or a dirty-shutdown count, cache + * these values in nfit_mem. + */ + if (nfit_mem->family == NVDIMM_FAMILY_INTEL) + nfit_intel_shutdown_status(nfit_mem); +} + +static int acpi_nfit_add_dimm(struct acpi_nfit_desc *acpi_desc, + struct nfit_mem *nfit_mem, u32 device_handle) +{ + struct nvdimm_bus_descriptor *nd_desc = &acpi_desc->nd_desc; + struct acpi_device *adev, *adev_dimm; + struct device *dev = acpi_desc->dev; + unsigned long dsm_mask, label_mask; + 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; + set_bit(NVDIMM_FAMILY_INTEL, &nd_desc->dimm_family_mask); + + 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) { + /* unit test case */ + populate_shutdown_status(nfit_mem); + return 0; + } + + adev_dimm = acpi_find_child_device(adev, device_handle, false); + nfit_mem->adev = adev_dimm; + if (!adev_dimm) { + dev_err(dev, "no ACPI.NFIT device with _ADR %#x, disabling...\n", + device_handle); + return force_enable_dimms ? 0 : -ENODEV; + } + + if (ACPI_FAILURE(acpi_install_notify_handler(adev_dimm->handle, + ACPI_DEVICE_NOTIFY, acpi_nvdimm_notify, adev_dimm))) { + dev_err(dev, "%s: notification registration failed\n", + dev_name(&adev_dimm->dev)); + return -ENXIO; + } + /* + * Record nfit_mem for the notification path to track back to + * the nfit sysfs attributes for this dimm device object. + */ + dev_set_drvdata(&adev_dimm->dev, nfit_mem); + + /* + * There are 4 "legacy" NVDIMM command sets + * (NVDIMM_FAMILY_{INTEL,MSFT,HPE1,HPE2}) that were created before + * an EFI working group was established to constrain this + * proliferation. The nfit driver probes for the supported command + * set by GUID. Note, if you're a platform developer looking to add + * a new command set to this probe, consider using an existing set, + * or otherwise seek approval to publish the command set at + * http://www.uefi.org/RFIC_LIST. + * + * Note, that checking for function0 (bit0) tells us if any commands + * are reachable through this GUID. + */ + clear_bit(NVDIMM_FAMILY_INTEL, &nd_desc->dimm_family_mask); + for (i = 0; i <= NVDIMM_FAMILY_MAX; i++) + if (acpi_check_dsm(adev_dimm->handle, to_nfit_uuid(i), 1, 1)) { + set_bit(i, &nd_desc->dimm_family_mask); + if (family < 0 || i == default_dsm_family) + family = i; + } + + /* limit the supported commands to those that are publicly documented */ + nfit_mem->family = family; + if (override_dsm_mask && !disable_vendor_specific) + dsm_mask = override_dsm_mask; + else if (nfit_mem->family == NVDIMM_FAMILY_INTEL) { + dsm_mask = NVDIMM_INTEL_CMDMASK; + if (disable_vendor_specific) + dsm_mask &= ~(1 << ND_CMD_VENDOR); + } else if (nfit_mem->family == NVDIMM_FAMILY_HPE1) { + dsm_mask = 0x1c3c76; + } else if (nfit_mem->family == NVDIMM_FAMILY_HPE2) { + dsm_mask = 0x1fe; + if (disable_vendor_specific) + dsm_mask &= ~(1 << 8); + } else if (nfit_mem->family == NVDIMM_FAMILY_MSFT) { + dsm_mask = 0xffffffff; + } else if (nfit_mem->family == NVDIMM_FAMILY_HYPERV) { + dsm_mask = 0x1f; + } else { + dev_dbg(dev, "unknown dimm command family\n"); + nfit_mem->family = -1; + /* DSMs are optional, continue loading the driver... */ + return 0; + } + + /* + * Function 0 is the command interrogation function, don't + * export it to potential userspace use, and enable it to be + * used as an error value in acpi_nfit_ctl(). + */ + dsm_mask &= ~1UL; + + guid = to_nfit_uuid(nfit_mem->family); + for_each_set_bit(i, &dsm_mask, BITS_PER_LONG) + if (acpi_check_dsm(adev_dimm->handle, guid, + nfit_dsm_revid(nfit_mem->family, i), + 1ULL << i)) + set_bit(i, &nfit_mem->dsm_mask); + + /* + * Prefer the NVDIMM_FAMILY_INTEL label read commands if present + * due to their better semantics handling locked capacity. + */ + label_mask = 1 << ND_CMD_GET_CONFIG_SIZE | 1 << ND_CMD_GET_CONFIG_DATA + | 1 << ND_CMD_SET_CONFIG_DATA; + if (family == NVDIMM_FAMILY_INTEL + && (dsm_mask & label_mask) == label_mask) + /* skip _LS{I,R,W} enabling */; + else { + if (acpi_nvdimm_has_method(adev_dimm, "_LSI") + && acpi_nvdimm_has_method(adev_dimm, "_LSR")) { + dev_dbg(dev, "%s: has _LSR\n", dev_name(&adev_dimm->dev)); + set_bit(NFIT_MEM_LSR, &nfit_mem->flags); + } + + if (test_bit(NFIT_MEM_LSR, &nfit_mem->flags) + && acpi_nvdimm_has_method(adev_dimm, "_LSW")) { + dev_dbg(dev, "%s: has _LSW\n", dev_name(&adev_dimm->dev)); + set_bit(NFIT_MEM_LSW, &nfit_mem->flags); + } + + /* + * Quirk read-only label configurations to preserve + * access to label-less namespaces by default. + */ + if (!test_bit(NFIT_MEM_LSW, &nfit_mem->flags) + && !force_labels) { + dev_dbg(dev, "%s: No _LSW, disable labels\n", + dev_name(&adev_dimm->dev)); + clear_bit(NFIT_MEM_LSR, &nfit_mem->flags); + } else + dev_dbg(dev, "%s: Force enable labels\n", + dev_name(&adev_dimm->dev)); + } + + populate_shutdown_status(nfit_mem); + + return 0; +} + +static void shutdown_dimm_notify(void *data) +{ + struct acpi_nfit_desc *acpi_desc = data; + struct nfit_mem *nfit_mem; + + mutex_lock(&acpi_desc->init_mutex); + /* + * Clear out the nfit_mem->flags_attr and shut down dimm event + * notifications. + */ + list_for_each_entry(nfit_mem, &acpi_desc->dimms, list) { + struct acpi_device *adev_dimm = nfit_mem->adev; + + if (nfit_mem->flags_attr) { + sysfs_put(nfit_mem->flags_attr); + nfit_mem->flags_attr = NULL; + } + if (adev_dimm) { + acpi_remove_notify_handler(adev_dimm->handle, + ACPI_DEVICE_NOTIFY, acpi_nvdimm_notify); + dev_set_drvdata(&adev_dimm->dev, NULL); + } + } + mutex_unlock(&acpi_desc->init_mutex); +} + +static const struct nvdimm_security_ops *acpi_nfit_get_security_ops(int family) +{ + switch (family) { + case NVDIMM_FAMILY_INTEL: + return intel_security_ops; + default: + return NULL; + } +} + +static const struct nvdimm_fw_ops *acpi_nfit_get_fw_ops( + struct nfit_mem *nfit_mem) +{ + unsigned long mask; + struct acpi_nfit_desc *acpi_desc = nfit_mem->acpi_desc; + struct nvdimm_bus_descriptor *nd_desc = &acpi_desc->nd_desc; + + if (!nd_desc->fw_ops) + return NULL; + + if (nfit_mem->family != NVDIMM_FAMILY_INTEL) + return NULL; + + mask = nfit_mem->dsm_mask & NVDIMM_INTEL_FW_ACTIVATE_CMDMASK; + if (mask != NVDIMM_INTEL_FW_ACTIVATE_CMDMASK) + return NULL; + + return intel_fw_ops; +} + +static int acpi_nfit_register_dimms(struct acpi_nfit_desc *acpi_desc) +{ + struct nfit_mem *nfit_mem; + int dimm_count = 0, rc; + struct nvdimm *nvdimm; + + list_for_each_entry(nfit_mem, &acpi_desc->dimms, list) { + struct acpi_nfit_flush_address *flush; + unsigned long flags = 0, cmd_mask; + struct nfit_memdev *nfit_memdev; + u32 device_handle; + u16 mem_flags; + + device_handle = __to_nfit_memdev(nfit_mem)->device_handle; + nvdimm = acpi_nfit_dimm_by_handle(acpi_desc, device_handle); + if (nvdimm) { + dimm_count++; + continue; + } + + /* collate flags across all memdevs for this dimm */ + list_for_each_entry(nfit_memdev, &acpi_desc->memdevs, list) { + struct acpi_nfit_memory_map *dimm_memdev; + + dimm_memdev = __to_nfit_memdev(nfit_mem); + if (dimm_memdev->device_handle + != nfit_memdev->memdev->device_handle) + continue; + dimm_memdev->flags |= nfit_memdev->memdev->flags; + } + + mem_flags = __to_nfit_memdev(nfit_mem)->flags; + if (mem_flags & ACPI_NFIT_MEM_NOT_ARMED) + set_bit(NDD_UNARMED, &flags); + + rc = acpi_nfit_add_dimm(acpi_desc, nfit_mem, device_handle); + if (rc) + continue; + + /* + * TODO: provide translation for non-NVDIMM_FAMILY_INTEL + * devices (i.e. from nd_cmd to acpi_dsm) to standardize the + * userspace interface. + */ + cmd_mask = 1UL << ND_CMD_CALL; + if (nfit_mem->family == NVDIMM_FAMILY_INTEL) { + /* + * These commands have a 1:1 correspondence + * between DSM payload and libnvdimm ioctl + * payload format. + */ + cmd_mask |= nfit_mem->dsm_mask & NVDIMM_STANDARD_CMDMASK; + } + + if (test_bit(NFIT_MEM_LSR, &nfit_mem->flags)) { + set_bit(ND_CMD_GET_CONFIG_SIZE, &cmd_mask); + set_bit(ND_CMD_GET_CONFIG_DATA, &cmd_mask); + } + if (test_bit(NFIT_MEM_LSW, &nfit_mem->flags)) + set_bit(ND_CMD_SET_CONFIG_DATA, &cmd_mask); + + flush = nfit_mem->nfit_flush ? nfit_mem->nfit_flush->flush + : NULL; + 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], + acpi_nfit_get_security_ops(nfit_mem->family), + acpi_nfit_get_fw_ops(nfit_mem)); + if (!nvdimm) + return -ENOMEM; + + nfit_mem->nvdimm = nvdimm; + dimm_count++; + + if ((mem_flags & ACPI_NFIT_MEM_FAILED_MASK) == 0) + continue; + + dev_err(acpi_desc->dev, "Error found in NVDIMM %s flags:%s%s%s%s%s\n", + nvdimm_name(nvdimm), + mem_flags & ACPI_NFIT_MEM_SAVE_FAILED ? " save_fail" : "", + mem_flags & ACPI_NFIT_MEM_RESTORE_FAILED ? " restore_fail":"", + mem_flags & ACPI_NFIT_MEM_FLUSH_FAILED ? " flush_fail" : "", + mem_flags & ACPI_NFIT_MEM_NOT_ARMED ? " not_armed" : "", + mem_flags & ACPI_NFIT_MEM_MAP_FAILED ? " map_fail" : ""); + + } + + rc = nvdimm_bus_check_dimm_count(acpi_desc->nvdimm_bus, dimm_count); + if (rc) + return rc; + + /* + * Now that dimms are successfully registered, and async registration + * is flushed, attempt to enable event notification. + */ + list_for_each_entry(nfit_mem, &acpi_desc->dimms, list) { + struct kernfs_node *nfit_kernfs; + + nvdimm = nfit_mem->nvdimm; + if (!nvdimm) + continue; + + nfit_kernfs = sysfs_get_dirent(nvdimm_kobj(nvdimm)->sd, "nfit"); + if (nfit_kernfs) + nfit_mem->flags_attr = sysfs_get_dirent(nfit_kernfs, + "flags"); + sysfs_put(nfit_kernfs); + if (!nfit_mem->flags_attr) + dev_warn(acpi_desc->dev, "%s: notifications disabled\n", + nvdimm_name(nvdimm)); + } + + return devm_add_action_or_reset(acpi_desc->dev, shutdown_dimm_notify, + acpi_desc); +} + +/* + * These constants are private because there are no kernel consumers of + * these commands. + */ +enum nfit_aux_cmds { + NFIT_CMD_TRANSLATE_SPA = 5, + NFIT_CMD_ARS_INJECT_SET = 7, + NFIT_CMD_ARS_INJECT_CLEAR = 8, + NFIT_CMD_ARS_INJECT_GET = 9, +}; + +static void acpi_nfit_init_dsms(struct acpi_nfit_desc *acpi_desc) +{ + struct nvdimm_bus_descriptor *nd_desc = &acpi_desc->nd_desc; + const guid_t *guid = to_nfit_uuid(NFIT_DEV_BUS); + unsigned long dsm_mask, *mask; + struct acpi_device *adev; + int i; + + set_bit(ND_CMD_CALL, &nd_desc->cmd_mask); + set_bit(NVDIMM_BUS_FAMILY_NFIT, &nd_desc->bus_family_mask); + + /* enable nfit_test to inject bus command emulation */ + if (acpi_desc->bus_cmd_force_en) { + nd_desc->cmd_mask = acpi_desc->bus_cmd_force_en; + mask = &nd_desc->bus_family_mask; + if (acpi_desc->family_dsm_mask[NVDIMM_BUS_FAMILY_INTEL]) { + set_bit(NVDIMM_BUS_FAMILY_INTEL, mask); + nd_desc->fw_ops = intel_bus_fw_ops; + } + } + + adev = to_acpi_dev(acpi_desc); + if (!adev) + return; + + for (i = ND_CMD_ARS_CAP; i <= ND_CMD_CLEAR_ERROR; i++) + if (acpi_check_dsm(adev->handle, guid, 1, 1ULL << i)) + set_bit(i, &nd_desc->cmd_mask); + + dsm_mask = + (1 << ND_CMD_ARS_CAP) | + (1 << ND_CMD_ARS_START) | + (1 << ND_CMD_ARS_STATUS) | + (1 << ND_CMD_CLEAR_ERROR) | + (1 << NFIT_CMD_TRANSLATE_SPA) | + (1 << NFIT_CMD_ARS_INJECT_SET) | + (1 << NFIT_CMD_ARS_INJECT_CLEAR) | + (1 << NFIT_CMD_ARS_INJECT_GET); + for_each_set_bit(i, &dsm_mask, BITS_PER_LONG) + if (acpi_check_dsm(adev->handle, guid, 1, 1ULL << i)) + set_bit(i, &acpi_desc->bus_dsm_mask); + + /* Enumerate allowed NVDIMM_BUS_FAMILY_INTEL commands */ + dsm_mask = NVDIMM_BUS_INTEL_FW_ACTIVATE_CMDMASK; + guid = to_nfit_bus_uuid(NVDIMM_BUS_FAMILY_INTEL); + mask = &acpi_desc->family_dsm_mask[NVDIMM_BUS_FAMILY_INTEL]; + for_each_set_bit(i, &dsm_mask, BITS_PER_LONG) + if (acpi_check_dsm(adev->handle, guid, 1, 1ULL << i)) + set_bit(i, mask); + + if (*mask == dsm_mask) { + set_bit(NVDIMM_BUS_FAMILY_INTEL, &nd_desc->bus_family_mask); + nd_desc->fw_ops = intel_bus_fw_ops; + } +} + +static ssize_t range_index_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct nd_region *nd_region = to_nd_region(dev); + struct nfit_spa *nfit_spa = nd_region_provider_data(nd_region); + + return sprintf(buf, "%d\n", nfit_spa->spa->range_index); +} +static DEVICE_ATTR_RO(range_index); + +static struct attribute *acpi_nfit_region_attributes[] = { + &dev_attr_range_index.attr, + NULL, +}; + +static const struct attribute_group acpi_nfit_region_attribute_group = { + .name = "nfit", + .attrs = acpi_nfit_region_attributes, +}; + +static const struct attribute_group *acpi_nfit_region_attribute_groups[] = { + &acpi_nfit_region_attribute_group, + NULL, +}; + +/* enough info to uniquely specify an interleave set */ +struct nfit_set_info { + u64 region_offset; + u32 serial_number; + u32 pad; +}; + +struct nfit_set_info2 { + u64 region_offset; + u32 serial_number; + u16 vendor_id; + u16 manufacturing_date; + u8 manufacturing_location; + u8 reserved[31]; +}; + +static int cmp_map_compat(const void *m0, const void *m1) +{ + const struct nfit_set_info *map0 = m0; + const struct nfit_set_info *map1 = m1; + + return memcmp(&map0->region_offset, &map1->region_offset, + sizeof(u64)); +} + +static int cmp_map(const void *m0, const void *m1) +{ + const struct nfit_set_info *map0 = m0; + const struct nfit_set_info *map1 = m1; + + if (map0->region_offset < map1->region_offset) + return -1; + else if (map0->region_offset > map1->region_offset) + return 1; + return 0; +} + +static int cmp_map2(const void *m0, const void *m1) +{ + const struct nfit_set_info2 *map0 = m0; + const struct nfit_set_info2 *map1 = m1; + + if (map0->region_offset < map1->region_offset) + return -1; + else if (map0->region_offset > map1->region_offset) + return 1; + return 0; +} + +/* Retrieve the nth entry referencing this spa */ +static struct acpi_nfit_memory_map *memdev_from_spa( + struct acpi_nfit_desc *acpi_desc, u16 range_index, int n) +{ + struct nfit_memdev *nfit_memdev; + + list_for_each_entry(nfit_memdev, &acpi_desc->memdevs, list) + if (nfit_memdev->memdev->range_index == range_index) + if (n-- == 0) + return nfit_memdev->memdev; + return NULL; +} + +static int acpi_nfit_init_interleave_set(struct acpi_nfit_desc *acpi_desc, + struct nd_region_desc *ndr_desc, + struct acpi_nfit_system_address *spa) +{ + struct device *dev = acpi_desc->dev; + struct nd_interleave_set *nd_set; + u16 nr = ndr_desc->num_mappings; + struct nfit_set_info2 *info2; + struct nfit_set_info *info; + int i; + + nd_set = devm_kzalloc(dev, sizeof(*nd_set), GFP_KERNEL); + if (!nd_set) + return -ENOMEM; + import_guid(&nd_set->type_guid, spa->range_guid); + + info = devm_kcalloc(dev, nr, sizeof(*info), GFP_KERNEL); + if (!info) + return -ENOMEM; + + info2 = devm_kcalloc(dev, nr, sizeof(*info2), GFP_KERNEL); + if (!info2) + return -ENOMEM; + + for (i = 0; i < nr; i++) { + struct nd_mapping_desc *mapping = &ndr_desc->mapping[i]; + struct nvdimm *nvdimm = mapping->nvdimm; + struct nfit_mem *nfit_mem = nvdimm_provider_data(nvdimm); + struct nfit_set_info *map = &info[i]; + struct nfit_set_info2 *map2 = &info2[i]; + struct acpi_nfit_memory_map *memdev = + memdev_from_spa(acpi_desc, spa->range_index, i); + struct acpi_nfit_control_region *dcr = nfit_mem->dcr; + + if (!memdev || !nfit_mem->dcr) { + dev_err(dev, "%s: failed to find DCR\n", __func__); + return -ENODEV; + } + + map->region_offset = memdev->region_offset; + map->serial_number = dcr->serial_number; + + map2->region_offset = memdev->region_offset; + map2->serial_number = dcr->serial_number; + map2->vendor_id = dcr->vendor_id; + map2->manufacturing_date = dcr->manufacturing_date; + map2->manufacturing_location = dcr->manufacturing_location; + } + + /* v1.1 namespaces */ + sort(info, nr, sizeof(*info), cmp_map, NULL); + nd_set->cookie1 = nd_fletcher64(info, sizeof(*info) * nr, 0); + + /* v1.2 namespaces */ + sort(info2, nr, sizeof(*info2), cmp_map2, NULL); + nd_set->cookie2 = nd_fletcher64(info2, sizeof(*info2) * nr, 0); + + /* support v1.1 namespaces created with the wrong sort order */ + sort(info, nr, sizeof(*info), cmp_map_compat, NULL); + nd_set->altcookie = nd_fletcher64(info, sizeof(*info) * nr, 0); + + /* record the result of the sort for the mapping position */ + for (i = 0; i < nr; i++) { + struct nfit_set_info2 *map2 = &info2[i]; + int j; + + for (j = 0; j < nr; j++) { + struct nd_mapping_desc *mapping = &ndr_desc->mapping[j]; + struct nvdimm *nvdimm = mapping->nvdimm; + struct nfit_mem *nfit_mem = nvdimm_provider_data(nvdimm); + struct acpi_nfit_control_region *dcr = nfit_mem->dcr; + + if (map2->serial_number == dcr->serial_number && + map2->vendor_id == dcr->vendor_id && + map2->manufacturing_date == dcr->manufacturing_date && + map2->manufacturing_location + == dcr->manufacturing_location) { + mapping->position = i; + break; + } + } + } + + ndr_desc->nd_set = nd_set; + devm_kfree(dev, info); + devm_kfree(dev, info2); + + return 0; +} + +static int ars_get_cap(struct acpi_nfit_desc *acpi_desc, + struct nd_cmd_ars_cap *cmd, struct nfit_spa *nfit_spa) +{ + struct nvdimm_bus_descriptor *nd_desc = &acpi_desc->nd_desc; + struct acpi_nfit_system_address *spa = nfit_spa->spa; + int cmd_rc, rc; + + cmd->address = spa->address; + cmd->length = spa->length; + rc = nd_desc->ndctl(nd_desc, NULL, ND_CMD_ARS_CAP, cmd, + sizeof(*cmd), &cmd_rc); + if (rc < 0) + return rc; + return cmd_rc; +} + +static int ars_start(struct acpi_nfit_desc *acpi_desc, + struct nfit_spa *nfit_spa, enum nfit_ars_state req_type) +{ + int rc; + int cmd_rc; + struct nd_cmd_ars_start ars_start; + struct acpi_nfit_system_address *spa = nfit_spa->spa; + struct nvdimm_bus_descriptor *nd_desc = &acpi_desc->nd_desc; + + memset(&ars_start, 0, sizeof(ars_start)); + ars_start.address = spa->address; + ars_start.length = spa->length; + if (req_type == ARS_REQ_SHORT) + ars_start.flags = ND_ARS_RETURN_PREV_DATA; + if (nfit_spa_type(spa) == NFIT_SPA_PM) + ars_start.type = ND_ARS_PERSISTENT; + else if (nfit_spa_type(spa) == NFIT_SPA_VOLATILE) + ars_start.type = ND_ARS_VOLATILE; + else + return -ENOTTY; + + rc = nd_desc->ndctl(nd_desc, NULL, ND_CMD_ARS_START, &ars_start, + sizeof(ars_start), &cmd_rc); + + if (rc < 0) + return rc; + if (cmd_rc < 0) + return cmd_rc; + set_bit(ARS_VALID, &acpi_desc->scrub_flags); + return 0; +} + +static int ars_continue(struct acpi_nfit_desc *acpi_desc) +{ + int rc, cmd_rc; + struct nd_cmd_ars_start ars_start; + struct nvdimm_bus_descriptor *nd_desc = &acpi_desc->nd_desc; + struct nd_cmd_ars_status *ars_status = acpi_desc->ars_status; + + ars_start = (struct nd_cmd_ars_start) { + .address = ars_status->restart_address, + .length = ars_status->restart_length, + .type = ars_status->type, + }; + rc = nd_desc->ndctl(nd_desc, NULL, ND_CMD_ARS_START, &ars_start, + sizeof(ars_start), &cmd_rc); + if (rc < 0) + return rc; + return cmd_rc; +} + +static int ars_get_status(struct acpi_nfit_desc *acpi_desc) +{ + struct nvdimm_bus_descriptor *nd_desc = &acpi_desc->nd_desc; + struct nd_cmd_ars_status *ars_status = acpi_desc->ars_status; + int rc, cmd_rc; + + rc = nd_desc->ndctl(nd_desc, NULL, ND_CMD_ARS_STATUS, ars_status, + acpi_desc->max_ars, &cmd_rc); + if (rc < 0) + return rc; + return cmd_rc; +} + +static void ars_complete(struct acpi_nfit_desc *acpi_desc, + struct nfit_spa *nfit_spa) +{ + struct nd_cmd_ars_status *ars_status = acpi_desc->ars_status; + struct acpi_nfit_system_address *spa = nfit_spa->spa; + struct nd_region *nd_region = nfit_spa->nd_region; + struct device *dev; + + lockdep_assert_held(&acpi_desc->init_mutex); + /* + * Only advance the ARS state for ARS runs initiated by the + * kernel, ignore ARS results from BIOS initiated runs for scrub + * completion tracking. + */ + if (acpi_desc->scrub_spa != nfit_spa) + return; + + if ((ars_status->address >= spa->address && ars_status->address + < spa->address + spa->length) + || (ars_status->address < spa->address)) { + /* + * Assume that if a scrub starts at an offset from the + * start of nfit_spa that we are in the continuation + * case. + * + * Otherwise, if the scrub covers the spa range, mark + * any pending request complete. + */ + if (ars_status->address + ars_status->length + >= spa->address + spa->length) + /* complete */; + else + return; + } else + return; + + acpi_desc->scrub_spa = NULL; + if (nd_region) { + dev = nd_region_dev(nd_region); + nvdimm_region_notify(nd_region, NVDIMM_REVALIDATE_POISON); + } else + dev = acpi_desc->dev; + dev_dbg(dev, "ARS: range %d complete\n", spa->range_index); +} + +static int ars_status_process_records(struct acpi_nfit_desc *acpi_desc) +{ + struct nvdimm_bus *nvdimm_bus = acpi_desc->nvdimm_bus; + struct nd_cmd_ars_status *ars_status = acpi_desc->ars_status; + int rc; + u32 i; + + /* + * First record starts at 44 byte offset from the start of the + * payload. + */ + if (ars_status->out_length < 44) + return 0; + + /* + * Ignore potentially stale results that are only refreshed + * after a start-ARS event. + */ + if (!test_and_clear_bit(ARS_VALID, &acpi_desc->scrub_flags)) { + dev_dbg(acpi_desc->dev, "skip %d stale records\n", + ars_status->num_records); + return 0; + } + + for (i = 0; i < ars_status->num_records; i++) { + /* only process full records */ + if (ars_status->out_length + < 44 + sizeof(struct nd_ars_record) * (i + 1)) + break; + rc = nvdimm_bus_add_badrange(nvdimm_bus, + ars_status->records[i].err_address, + ars_status->records[i].length); + if (rc) + return rc; + } + if (i < ars_status->num_records) + dev_warn(acpi_desc->dev, "detected truncated ars results\n"); + + return 0; +} + +static void acpi_nfit_remove_resource(void *data) +{ + struct resource *res = data; + + remove_resource(res); +} + +static int acpi_nfit_insert_resource(struct acpi_nfit_desc *acpi_desc, + struct nd_region_desc *ndr_desc) +{ + struct resource *res, *nd_res = ndr_desc->res; + int is_pmem, ret; + + /* No operation if the region is already registered as PMEM */ + is_pmem = region_intersects(nd_res->start, resource_size(nd_res), + IORESOURCE_MEM, IORES_DESC_PERSISTENT_MEMORY); + if (is_pmem == REGION_INTERSECTS) + return 0; + + res = devm_kzalloc(acpi_desc->dev, sizeof(*res), GFP_KERNEL); + if (!res) + return -ENOMEM; + + res->name = "Persistent Memory"; + res->start = nd_res->start; + res->end = nd_res->end; + res->flags = IORESOURCE_MEM; + res->desc = IORES_DESC_PERSISTENT_MEMORY; + + ret = insert_resource(&iomem_resource, res); + if (ret) + return ret; + + ret = devm_add_action_or_reset(acpi_desc->dev, + acpi_nfit_remove_resource, + res); + if (ret) + return ret; + + return 0; +} + +static int acpi_nfit_init_mapping(struct acpi_nfit_desc *acpi_desc, + struct nd_mapping_desc *mapping, struct nd_region_desc *ndr_desc, + struct acpi_nfit_memory_map *memdev, + struct nfit_spa *nfit_spa) +{ + struct nvdimm *nvdimm = acpi_nfit_dimm_by_handle(acpi_desc, + memdev->device_handle); + struct acpi_nfit_system_address *spa = nfit_spa->spa; + + if (!nvdimm) { + dev_err(acpi_desc->dev, "spa%d dimm: %#x not found\n", + spa->range_index, memdev->device_handle); + return -ENODEV; + } + + mapping->nvdimm = nvdimm; + switch (nfit_spa_type(spa)) { + case NFIT_SPA_PM: + case NFIT_SPA_VOLATILE: + mapping->start = memdev->address; + mapping->size = memdev->region_size; + break; + } + + return 0; +} + +static bool nfit_spa_is_virtual(struct acpi_nfit_system_address *spa) +{ + return (nfit_spa_type(spa) == NFIT_SPA_VDISK || + nfit_spa_type(spa) == NFIT_SPA_VCD || + nfit_spa_type(spa) == NFIT_SPA_PDISK || + nfit_spa_type(spa) == NFIT_SPA_PCD); +} + +static bool nfit_spa_is_volatile(struct acpi_nfit_system_address *spa) +{ + return (nfit_spa_type(spa) == NFIT_SPA_VDISK || + nfit_spa_type(spa) == NFIT_SPA_VCD || + nfit_spa_type(spa) == NFIT_SPA_VOLATILE); +} + +static int acpi_nfit_register_region(struct acpi_nfit_desc *acpi_desc, + struct nfit_spa *nfit_spa) +{ + static struct nd_mapping_desc mappings[ND_MAX_MAPPINGS]; + struct acpi_nfit_system_address *spa = nfit_spa->spa; + struct nd_region_desc *ndr_desc, _ndr_desc; + struct nfit_memdev *nfit_memdev; + struct nvdimm_bus *nvdimm_bus; + struct resource res; + int count = 0, rc; + + if (nfit_spa->nd_region) + return 0; + + if (spa->range_index == 0 && !nfit_spa_is_virtual(spa)) { + dev_dbg(acpi_desc->dev, "detected invalid spa index\n"); + return 0; + } + + memset(&res, 0, sizeof(res)); + memset(&mappings, 0, sizeof(mappings)); + memset(&_ndr_desc, 0, sizeof(_ndr_desc)); + res.start = spa->address; + res.end = res.start + spa->length - 1; + ndr_desc = &_ndr_desc; + ndr_desc->res = &res; + ndr_desc->provider_data = nfit_spa; + ndr_desc->attr_groups = acpi_nfit_region_attribute_groups; + if (spa->flags & ACPI_NFIT_PROXIMITY_VALID) { + ndr_desc->numa_node = pxm_to_online_node(spa->proximity_domain); + ndr_desc->target_node = pxm_to_node(spa->proximity_domain); + } else { + ndr_desc->numa_node = NUMA_NO_NODE; + ndr_desc->target_node = NUMA_NO_NODE; + } + + /* Fallback to address based numa information if node lookup failed */ + if (ndr_desc->numa_node == NUMA_NO_NODE) { + ndr_desc->numa_node = memory_add_physaddr_to_nid(spa->address); + dev_info(acpi_desc->dev, "changing numa node from %d to %d for nfit region [%pa-%pa]", + NUMA_NO_NODE, ndr_desc->numa_node, &res.start, &res.end); + } + if (ndr_desc->target_node == NUMA_NO_NODE) { + ndr_desc->target_node = phys_to_target_node(spa->address); + dev_info(acpi_desc->dev, "changing target node from %d to %d for nfit region [%pa-%pa]", + NUMA_NO_NODE, ndr_desc->numa_node, &res.start, &res.end); + } + + /* + * Persistence domain bits are hierarchical, if + * ACPI_NFIT_CAPABILITY_CACHE_FLUSH is set then + * ACPI_NFIT_CAPABILITY_MEM_FLUSH is implied. + */ + if (acpi_desc->platform_cap & ACPI_NFIT_CAPABILITY_CACHE_FLUSH) + set_bit(ND_REGION_PERSIST_CACHE, &ndr_desc->flags); + else if (acpi_desc->platform_cap & ACPI_NFIT_CAPABILITY_MEM_FLUSH) + set_bit(ND_REGION_PERSIST_MEMCTRL, &ndr_desc->flags); + + list_for_each_entry(nfit_memdev, &acpi_desc->memdevs, list) { + struct acpi_nfit_memory_map *memdev = nfit_memdev->memdev; + struct nd_mapping_desc *mapping; + + /* range index 0 == unmapped in SPA or invalid-SPA */ + if (memdev->range_index == 0 || spa->range_index == 0) + continue; + if (memdev->range_index != spa->range_index) + continue; + if (count >= ND_MAX_MAPPINGS) { + dev_err(acpi_desc->dev, "spa%d exceeds max mappings %d\n", + spa->range_index, ND_MAX_MAPPINGS); + return -ENXIO; + } + mapping = &mappings[count++]; + rc = acpi_nfit_init_mapping(acpi_desc, mapping, ndr_desc, + memdev, nfit_spa); + if (rc) + goto out; + } + + ndr_desc->mapping = mappings; + ndr_desc->num_mappings = count; + rc = acpi_nfit_init_interleave_set(acpi_desc, ndr_desc, spa); + if (rc) + goto out; + + nvdimm_bus = acpi_desc->nvdimm_bus; + if (nfit_spa_type(spa) == NFIT_SPA_PM) { + rc = acpi_nfit_insert_resource(acpi_desc, ndr_desc); + if (rc) { + dev_warn(acpi_desc->dev, + "failed to insert pmem resource to iomem: %d\n", + rc); + goto out; + } + + nfit_spa->nd_region = nvdimm_pmem_region_create(nvdimm_bus, + ndr_desc); + if (!nfit_spa->nd_region) + rc = -ENOMEM; + } else if (nfit_spa_is_volatile(spa)) { + nfit_spa->nd_region = nvdimm_volatile_region_create(nvdimm_bus, + ndr_desc); + if (!nfit_spa->nd_region) + rc = -ENOMEM; + } else if (nfit_spa_is_virtual(spa)) { + nfit_spa->nd_region = nvdimm_pmem_region_create(nvdimm_bus, + ndr_desc); + if (!nfit_spa->nd_region) + rc = -ENOMEM; + } + + out: + if (rc) + dev_err(acpi_desc->dev, "failed to register spa range %d\n", + nfit_spa->spa->range_index); + return rc; +} + +static int ars_status_alloc(struct acpi_nfit_desc *acpi_desc) +{ + struct device *dev = acpi_desc->dev; + struct nd_cmd_ars_status *ars_status; + + if (acpi_desc->ars_status) { + memset(acpi_desc->ars_status, 0, acpi_desc->max_ars); + return 0; + } + + ars_status = devm_kzalloc(dev, acpi_desc->max_ars, GFP_KERNEL); + if (!ars_status) + return -ENOMEM; + acpi_desc->ars_status = ars_status; + return 0; +} + +static int acpi_nfit_query_poison(struct acpi_nfit_desc *acpi_desc) +{ + int rc; + + if (ars_status_alloc(acpi_desc)) + return -ENOMEM; + + rc = ars_get_status(acpi_desc); + + if (rc < 0 && rc != -ENOSPC) + return rc; + + if (ars_status_process_records(acpi_desc)) + dev_err(acpi_desc->dev, "Failed to process ARS records\n"); + + return rc; +} + +static int ars_register(struct acpi_nfit_desc *acpi_desc, + struct nfit_spa *nfit_spa) +{ + int rc; + + if (test_bit(ARS_FAILED, &nfit_spa->ars_state)) + return acpi_nfit_register_region(acpi_desc, nfit_spa); + + set_bit(ARS_REQ_SHORT, &nfit_spa->ars_state); + if (!no_init_ars) + set_bit(ARS_REQ_LONG, &nfit_spa->ars_state); + + switch (acpi_nfit_query_poison(acpi_desc)) { + case 0: + case -ENOSPC: + case -EAGAIN: + rc = ars_start(acpi_desc, nfit_spa, ARS_REQ_SHORT); + /* shouldn't happen, try again later */ + if (rc == -EBUSY) + break; + if (rc) { + set_bit(ARS_FAILED, &nfit_spa->ars_state); + break; + } + clear_bit(ARS_REQ_SHORT, &nfit_spa->ars_state); + rc = acpi_nfit_query_poison(acpi_desc); + if (rc) + break; + acpi_desc->scrub_spa = nfit_spa; + ars_complete(acpi_desc, nfit_spa); + /* + * If ars_complete() says we didn't complete the + * short scrub, we'll try again with a long + * request. + */ + acpi_desc->scrub_spa = NULL; + break; + case -EBUSY: + case -ENOMEM: + /* + * BIOS was using ARS, wait for it to complete (or + * resources to become available) and then perform our + * own scrubs. + */ + break; + default: + set_bit(ARS_FAILED, &nfit_spa->ars_state); + break; + } + + return acpi_nfit_register_region(acpi_desc, nfit_spa); +} + +static void ars_complete_all(struct acpi_nfit_desc *acpi_desc) +{ + struct nfit_spa *nfit_spa; + + list_for_each_entry(nfit_spa, &acpi_desc->spas, list) { + if (test_bit(ARS_FAILED, &nfit_spa->ars_state)) + continue; + ars_complete(acpi_desc, nfit_spa); + } +} + +static unsigned int __acpi_nfit_scrub(struct acpi_nfit_desc *acpi_desc, + int query_rc) +{ + unsigned int tmo = acpi_desc->scrub_tmo; + struct device *dev = acpi_desc->dev; + struct nfit_spa *nfit_spa; + + lockdep_assert_held(&acpi_desc->init_mutex); + + if (test_bit(ARS_CANCEL, &acpi_desc->scrub_flags)) + return 0; + + if (query_rc == -EBUSY) { + dev_dbg(dev, "ARS: ARS busy\n"); + return min(30U * 60U, tmo * 2); + } + if (query_rc == -ENOSPC) { + dev_dbg(dev, "ARS: ARS continue\n"); + ars_continue(acpi_desc); + return 1; + } + if (query_rc && query_rc != -EAGAIN) { + unsigned long long addr, end; + + addr = acpi_desc->ars_status->address; + end = addr + acpi_desc->ars_status->length; + dev_dbg(dev, "ARS: %llx-%llx failed (%d)\n", addr, end, + query_rc); + } + + ars_complete_all(acpi_desc); + list_for_each_entry(nfit_spa, &acpi_desc->spas, list) { + enum nfit_ars_state req_type; + int rc; + + if (test_bit(ARS_FAILED, &nfit_spa->ars_state)) + continue; + + /* prefer short ARS requests first */ + if (test_bit(ARS_REQ_SHORT, &nfit_spa->ars_state)) + req_type = ARS_REQ_SHORT; + else if (test_bit(ARS_REQ_LONG, &nfit_spa->ars_state)) + req_type = ARS_REQ_LONG; + else + continue; + rc = ars_start(acpi_desc, nfit_spa, req_type); + + dev = nd_region_dev(nfit_spa->nd_region); + dev_dbg(dev, "ARS: range %d ARS start %s (%d)\n", + nfit_spa->spa->range_index, + req_type == ARS_REQ_SHORT ? "short" : "long", + rc); + /* + * Hmm, we raced someone else starting ARS? Try again in + * a bit. + */ + if (rc == -EBUSY) + return 1; + if (rc == 0) { + dev_WARN_ONCE(dev, acpi_desc->scrub_spa, + "scrub start while range %d active\n", + acpi_desc->scrub_spa->spa->range_index); + clear_bit(req_type, &nfit_spa->ars_state); + acpi_desc->scrub_spa = nfit_spa; + /* + * Consider this spa last for future scrub + * requests + */ + list_move_tail(&nfit_spa->list, &acpi_desc->spas); + return 1; + } + + dev_err(dev, "ARS: range %d ARS failed (%d)\n", + nfit_spa->spa->range_index, rc); + set_bit(ARS_FAILED, &nfit_spa->ars_state); + } + return 0; +} + +static void __sched_ars(struct acpi_nfit_desc *acpi_desc, unsigned int tmo) +{ + lockdep_assert_held(&acpi_desc->init_mutex); + + set_bit(ARS_BUSY, &acpi_desc->scrub_flags); + /* note this should only be set from within the workqueue */ + if (tmo) + acpi_desc->scrub_tmo = tmo; + queue_delayed_work(nfit_wq, &acpi_desc->dwork, tmo * HZ); +} + +static void sched_ars(struct acpi_nfit_desc *acpi_desc) +{ + __sched_ars(acpi_desc, 0); +} + +static void notify_ars_done(struct acpi_nfit_desc *acpi_desc) +{ + lockdep_assert_held(&acpi_desc->init_mutex); + + clear_bit(ARS_BUSY, &acpi_desc->scrub_flags); + acpi_desc->scrub_count++; + if (acpi_desc->scrub_count_state) + sysfs_notify_dirent(acpi_desc->scrub_count_state); +} + +static void acpi_nfit_scrub(struct work_struct *work) +{ + struct acpi_nfit_desc *acpi_desc; + unsigned int tmo; + int query_rc; + + acpi_desc = container_of(work, typeof(*acpi_desc), dwork.work); + mutex_lock(&acpi_desc->init_mutex); + query_rc = acpi_nfit_query_poison(acpi_desc); + tmo = __acpi_nfit_scrub(acpi_desc, query_rc); + if (tmo) + __sched_ars(acpi_desc, tmo); + else + notify_ars_done(acpi_desc); + memset(acpi_desc->ars_status, 0, acpi_desc->max_ars); + clear_bit(ARS_POLL, &acpi_desc->scrub_flags); + mutex_unlock(&acpi_desc->init_mutex); +} + +static void acpi_nfit_init_ars(struct acpi_nfit_desc *acpi_desc, + struct nfit_spa *nfit_spa) +{ + int type = nfit_spa_type(nfit_spa->spa); + struct nd_cmd_ars_cap ars_cap; + int rc; + + set_bit(ARS_FAILED, &nfit_spa->ars_state); + memset(&ars_cap, 0, sizeof(ars_cap)); + rc = ars_get_cap(acpi_desc, &ars_cap, nfit_spa); + if (rc < 0) + return; + /* check that the supported scrub types match the spa type */ + if (type == NFIT_SPA_VOLATILE && ((ars_cap.status >> 16) + & ND_ARS_VOLATILE) == 0) + return; + if (type == NFIT_SPA_PM && ((ars_cap.status >> 16) + & ND_ARS_PERSISTENT) == 0) + return; + + nfit_spa->max_ars = ars_cap.max_ars_out; + nfit_spa->clear_err_unit = ars_cap.clear_err_unit; + acpi_desc->max_ars = max(nfit_spa->max_ars, acpi_desc->max_ars); + clear_bit(ARS_FAILED, &nfit_spa->ars_state); +} + +static int acpi_nfit_register_regions(struct acpi_nfit_desc *acpi_desc) +{ + struct nfit_spa *nfit_spa; + int rc, do_sched_ars = 0; + + set_bit(ARS_VALID, &acpi_desc->scrub_flags); + list_for_each_entry(nfit_spa, &acpi_desc->spas, list) { + switch (nfit_spa_type(nfit_spa->spa)) { + case NFIT_SPA_VOLATILE: + case NFIT_SPA_PM: + acpi_nfit_init_ars(acpi_desc, nfit_spa); + break; + } + } + + list_for_each_entry(nfit_spa, &acpi_desc->spas, list) { + switch (nfit_spa_type(nfit_spa->spa)) { + case NFIT_SPA_VOLATILE: + case NFIT_SPA_PM: + /* register regions and kick off initial ARS run */ + rc = ars_register(acpi_desc, nfit_spa); + if (rc) + return rc; + + /* + * Kick off background ARS if at least one + * region successfully registered ARS + */ + if (!test_bit(ARS_FAILED, &nfit_spa->ars_state)) + do_sched_ars++; + break; + case NFIT_SPA_BDW: + /* nothing to register */ + break; + case NFIT_SPA_DCR: + case NFIT_SPA_VDISK: + case NFIT_SPA_VCD: + case NFIT_SPA_PDISK: + case NFIT_SPA_PCD: + /* register known regions that don't support ARS */ + rc = acpi_nfit_register_region(acpi_desc, nfit_spa); + if (rc) + return rc; + break; + default: + /* don't register unknown regions */ + break; + } + } + + if (do_sched_ars) + sched_ars(acpi_desc); + return 0; +} + +static int acpi_nfit_check_deletions(struct acpi_nfit_desc *acpi_desc, + struct nfit_table_prev *prev) +{ + struct device *dev = acpi_desc->dev; + + if (!list_empty(&prev->spas) || + !list_empty(&prev->memdevs) || + !list_empty(&prev->dcrs) || + !list_empty(&prev->bdws) || + !list_empty(&prev->idts) || + !list_empty(&prev->flushes)) { + dev_err(dev, "new nfit deletes entries (unsupported)\n"); + return -ENXIO; + } + return 0; +} + +static int acpi_nfit_desc_init_scrub_attr(struct acpi_nfit_desc *acpi_desc) +{ + struct device *dev = acpi_desc->dev; + struct kernfs_node *nfit; + struct device *bus_dev; + + if (!ars_supported(acpi_desc->nvdimm_bus)) + return 0; + + bus_dev = to_nvdimm_bus_dev(acpi_desc->nvdimm_bus); + nfit = sysfs_get_dirent(bus_dev->kobj.sd, "nfit"); + if (!nfit) { + dev_err(dev, "sysfs_get_dirent 'nfit' failed\n"); + return -ENODEV; + } + acpi_desc->scrub_count_state = sysfs_get_dirent(nfit, "scrub"); + sysfs_put(nfit); + if (!acpi_desc->scrub_count_state) { + dev_err(dev, "sysfs_get_dirent 'scrub' failed\n"); + return -ENODEV; + } + + return 0; +} + +static void acpi_nfit_unregister(void *data) +{ + struct acpi_nfit_desc *acpi_desc = data; + + nvdimm_bus_unregister(acpi_desc->nvdimm_bus); +} + +int acpi_nfit_init(struct acpi_nfit_desc *acpi_desc, void *data, acpi_size sz) +{ + struct device *dev = acpi_desc->dev; + struct nfit_table_prev prev; + const void *end; + int rc; + + if (!acpi_desc->nvdimm_bus) { + acpi_nfit_init_dsms(acpi_desc); + + acpi_desc->nvdimm_bus = nvdimm_bus_register(dev, + &acpi_desc->nd_desc); + if (!acpi_desc->nvdimm_bus) + return -ENOMEM; + + rc = devm_add_action_or_reset(dev, acpi_nfit_unregister, + acpi_desc); + if (rc) + return rc; + + rc = acpi_nfit_desc_init_scrub_attr(acpi_desc); + if (rc) + return rc; + + /* register this acpi_desc for mce notifications */ + mutex_lock(&acpi_desc_lock); + list_add_tail(&acpi_desc->list, &acpi_descs); + mutex_unlock(&acpi_desc_lock); + } + + mutex_lock(&acpi_desc->init_mutex); + + INIT_LIST_HEAD(&prev.spas); + INIT_LIST_HEAD(&prev.memdevs); + INIT_LIST_HEAD(&prev.dcrs); + INIT_LIST_HEAD(&prev.bdws); + INIT_LIST_HEAD(&prev.idts); + INIT_LIST_HEAD(&prev.flushes); + + list_cut_position(&prev.spas, &acpi_desc->spas, + acpi_desc->spas.prev); + list_cut_position(&prev.memdevs, &acpi_desc->memdevs, + acpi_desc->memdevs.prev); + list_cut_position(&prev.dcrs, &acpi_desc->dcrs, + acpi_desc->dcrs.prev); + list_cut_position(&prev.bdws, &acpi_desc->bdws, + acpi_desc->bdws.prev); + list_cut_position(&prev.idts, &acpi_desc->idts, + acpi_desc->idts.prev); + list_cut_position(&prev.flushes, &acpi_desc->flushes, + acpi_desc->flushes.prev); + + end = data + sz; + while (!IS_ERR_OR_NULL(data)) + data = add_table(acpi_desc, &prev, data, end); + + if (IS_ERR(data)) { + dev_dbg(dev, "nfit table parsing error: %ld\n", PTR_ERR(data)); + rc = PTR_ERR(data); + goto out_unlock; + } + + rc = acpi_nfit_check_deletions(acpi_desc, &prev); + if (rc) + goto out_unlock; + + rc = nfit_mem_init(acpi_desc); + if (rc) + goto out_unlock; + + rc = acpi_nfit_register_dimms(acpi_desc); + if (rc) + goto out_unlock; + + rc = acpi_nfit_register_regions(acpi_desc); + + out_unlock: + mutex_unlock(&acpi_desc->init_mutex); + return rc; +} +EXPORT_SYMBOL_GPL(acpi_nfit_init); + +static int acpi_nfit_flush_probe(struct nvdimm_bus_descriptor *nd_desc) +{ + struct acpi_nfit_desc *acpi_desc = to_acpi_desc(nd_desc); + struct device *dev = acpi_desc->dev; + + /* Bounce the device lock to flush acpi_nfit_add / acpi_nfit_notify */ + device_lock(dev); + device_unlock(dev); + + /* Bounce the init_mutex to complete initial registration */ + mutex_lock(&acpi_desc->init_mutex); + mutex_unlock(&acpi_desc->init_mutex); + + return 0; +} + +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_desc(nd_desc); + + if (nvdimm) + return 0; + if (cmd != ND_CMD_ARS_START) + return 0; + + /* + * The kernel and userspace may race to initiate a scrub, but + * the scrub thread is prepared to lose that initial race. It + * just needs guarantees that any ARS it initiates are not + * interrupted by any intervening start requests from userspace. + */ + if (work_busy(&acpi_desc->dwork.work)) + return -EBUSY; + + return 0; +} + +/* + * Prevent security and firmware activate 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 (func > NVDIMM_CMD_MAX || + (1 << func) & NVDIMM_INTEL_DENY_CMDMASK) + return -EOPNOTSUPP; + } + + /* block all non-nfit bus commands */ + if (!nvdimm && cmd == ND_CMD_CALL && + call_pkg->nd_family != NVDIMM_BUS_FAMILY_NFIT) + return -EOPNOTSUPP; + + return __acpi_nfit_clear_to_send(nd_desc, nvdimm, cmd); +} + +int acpi_nfit_ars_rescan(struct acpi_nfit_desc *acpi_desc, + enum nfit_ars_state req_type) +{ + struct device *dev = acpi_desc->dev; + int scheduled = 0, busy = 0; + struct nfit_spa *nfit_spa; + + mutex_lock(&acpi_desc->init_mutex); + if (test_bit(ARS_CANCEL, &acpi_desc->scrub_flags)) { + mutex_unlock(&acpi_desc->init_mutex); + return 0; + } + + list_for_each_entry(nfit_spa, &acpi_desc->spas, list) { + int type = nfit_spa_type(nfit_spa->spa); + + if (type != NFIT_SPA_PM && type != NFIT_SPA_VOLATILE) + continue; + if (test_bit(ARS_FAILED, &nfit_spa->ars_state)) + continue; + + if (test_and_set_bit(req_type, &nfit_spa->ars_state)) + busy++; + else + scheduled++; + } + if (scheduled) { + sched_ars(acpi_desc); + dev_dbg(dev, "ars_scan triggered\n"); + } + mutex_unlock(&acpi_desc->init_mutex); + + if (scheduled) + return 0; + if (busy) + return -EBUSY; + return -ENOTTY; +} + +void acpi_nfit_desc_init(struct acpi_nfit_desc *acpi_desc, struct device *dev) +{ + struct nvdimm_bus_descriptor *nd_desc; + + dev_set_drvdata(dev, acpi_desc); + acpi_desc->dev = dev; + nd_desc = &acpi_desc->nd_desc; + nd_desc->provider_name = "ACPI.NFIT"; + nd_desc->module = THIS_MODULE; + nd_desc->ndctl = acpi_nfit_ctl; + nd_desc->flush_probe = acpi_nfit_flush_probe; + nd_desc->clear_to_send = acpi_nfit_clear_to_send; + nd_desc->attr_groups = acpi_nfit_attribute_groups; + + INIT_LIST_HEAD(&acpi_desc->spas); + INIT_LIST_HEAD(&acpi_desc->dcrs); + INIT_LIST_HEAD(&acpi_desc->bdws); + INIT_LIST_HEAD(&acpi_desc->idts); + INIT_LIST_HEAD(&acpi_desc->flushes); + INIT_LIST_HEAD(&acpi_desc->memdevs); + INIT_LIST_HEAD(&acpi_desc->dimms); + INIT_LIST_HEAD(&acpi_desc->list); + mutex_init(&acpi_desc->init_mutex); + acpi_desc->scrub_tmo = 1; + INIT_DELAYED_WORK(&acpi_desc->dwork, acpi_nfit_scrub); +} +EXPORT_SYMBOL_GPL(acpi_nfit_desc_init); + +static void acpi_nfit_put_table(void *table) +{ + acpi_put_table(table); +} + +static void acpi_nfit_notify(acpi_handle handle, u32 event, void *data) +{ + struct acpi_device *adev = data; + + device_lock(&adev->dev); + __acpi_nfit_notify(&adev->dev, handle, event); + device_unlock(&adev->dev); +} + +static void acpi_nfit_remove_notify_handler(void *data) +{ + struct acpi_device *adev = data; + + acpi_dev_remove_notify_handler(adev, ACPI_DEVICE_NOTIFY, + acpi_nfit_notify); +} + +void acpi_nfit_shutdown(void *data) +{ + struct acpi_nfit_desc *acpi_desc = data; + struct device *bus_dev = to_nvdimm_bus_dev(acpi_desc->nvdimm_bus); + + /* + * Destruct under acpi_desc_lock so that nfit_handle_mce does not + * race teardown + */ + mutex_lock(&acpi_desc_lock); + list_del(&acpi_desc->list); + mutex_unlock(&acpi_desc_lock); + + mutex_lock(&acpi_desc->init_mutex); + set_bit(ARS_CANCEL, &acpi_desc->scrub_flags); + mutex_unlock(&acpi_desc->init_mutex); + cancel_delayed_work_sync(&acpi_desc->dwork); + + /* + * Bounce the nvdimm bus lock to make sure any in-flight + * acpi_nfit_ars_rescan() submissions have had a chance to + * either submit or see ->cancel set. + */ + device_lock(bus_dev); + device_unlock(bus_dev); + + flush_workqueue(nfit_wq); +} +EXPORT_SYMBOL_GPL(acpi_nfit_shutdown); + +static int acpi_nfit_add(struct acpi_device *adev) +{ + struct acpi_buffer buf = { ACPI_ALLOCATE_BUFFER, NULL }; + struct acpi_nfit_desc *acpi_desc; + struct device *dev = &adev->dev; + struct acpi_table_header *tbl; + acpi_status status = AE_OK; + acpi_size sz; + int rc = 0; + + rc = acpi_dev_install_notify_handler(adev, ACPI_DEVICE_NOTIFY, + acpi_nfit_notify); + if (rc) + return rc; + + rc = devm_add_action_or_reset(dev, acpi_nfit_remove_notify_handler, + adev); + if (rc) + return rc; + + status = acpi_get_table(ACPI_SIG_NFIT, 0, &tbl); + if (ACPI_FAILURE(status)) { + /* The NVDIMM root device allows OS to trigger enumeration of + * NVDIMMs through NFIT at boot time and re-enumeration at + * root level via the _FIT method during runtime. + * This is ok to return 0 here, we could have an nvdimm + * hotplugged later and evaluate _FIT method which returns + * data in the format of a series of NFIT Structures. + */ + dev_dbg(dev, "failed to find NFIT at startup\n"); + return 0; + } + + rc = devm_add_action_or_reset(dev, acpi_nfit_put_table, tbl); + if (rc) + return rc; + sz = tbl->length; + + acpi_desc = devm_kzalloc(dev, sizeof(*acpi_desc), GFP_KERNEL); + if (!acpi_desc) + return -ENOMEM; + acpi_nfit_desc_init(acpi_desc, &adev->dev); + + /* Save the acpi header for exporting the revision via sysfs */ + acpi_desc->acpi_header = *tbl; + + /* Evaluate _FIT and override with that if present */ + status = acpi_evaluate_object(adev->handle, "_FIT", NULL, &buf); + if (ACPI_SUCCESS(status) && buf.length > 0) { + union acpi_object *obj = buf.pointer; + + if (obj->type == ACPI_TYPE_BUFFER) + rc = acpi_nfit_init(acpi_desc, obj->buffer.pointer, + obj->buffer.length); + else + dev_dbg(dev, "invalid type %d, ignoring _FIT\n", + (int) obj->type); + kfree(buf.pointer); + } else + /* skip over the lead-in header table */ + rc = acpi_nfit_init(acpi_desc, (void *) tbl + + sizeof(struct acpi_table_nfit), + sz - sizeof(struct acpi_table_nfit)); + + if (rc) + return rc; + + return devm_add_action_or_reset(dev, acpi_nfit_shutdown, acpi_desc); +} + +static void acpi_nfit_update_notify(struct device *dev, acpi_handle handle) +{ + struct acpi_nfit_desc *acpi_desc = dev_get_drvdata(dev); + struct acpi_buffer buf = { ACPI_ALLOCATE_BUFFER, NULL }; + union acpi_object *obj; + acpi_status status; + int ret; + + if (!dev->driver) { + /* dev->driver may be null if we're being removed */ + dev_dbg(dev, "no driver found for dev\n"); + return; + } + + if (!acpi_desc) { + acpi_desc = devm_kzalloc(dev, sizeof(*acpi_desc), GFP_KERNEL); + if (!acpi_desc) + return; + acpi_nfit_desc_init(acpi_desc, dev); + } else { + /* + * Finish previous registration before considering new + * regions. + */ + flush_workqueue(nfit_wq); + } + + /* Evaluate _FIT */ + status = acpi_evaluate_object(handle, "_FIT", NULL, &buf); + if (ACPI_FAILURE(status)) { + dev_err(dev, "failed to evaluate _FIT\n"); + return; + } + + obj = buf.pointer; + if (obj->type == ACPI_TYPE_BUFFER) { + ret = acpi_nfit_init(acpi_desc, obj->buffer.pointer, + obj->buffer.length); + if (ret) + dev_err(dev, "failed to merge updated NFIT\n"); + } else + dev_err(dev, "Invalid _FIT\n"); + kfree(buf.pointer); +} + +static void acpi_nfit_uc_error_notify(struct device *dev, acpi_handle handle) +{ + struct acpi_nfit_desc *acpi_desc = dev_get_drvdata(dev); + + if (acpi_desc->scrub_mode == HW_ERROR_SCRUB_ON) + acpi_nfit_ars_rescan(acpi_desc, ARS_REQ_LONG); + else + acpi_nfit_ars_rescan(acpi_desc, ARS_REQ_SHORT); +} + +void __acpi_nfit_notify(struct device *dev, acpi_handle handle, u32 event) +{ + dev_dbg(dev, "event: 0x%x\n", event); + + switch (event) { + case NFIT_NOTIFY_UPDATE: + return acpi_nfit_update_notify(dev, handle); + case NFIT_NOTIFY_UC_MEMORY_ERROR: + return acpi_nfit_uc_error_notify(dev, handle); + default: + return; + } +} +EXPORT_SYMBOL_GPL(__acpi_nfit_notify); + +static const struct acpi_device_id acpi_nfit_ids[] = { + { "ACPI0012", 0 }, + { "", 0 }, +}; +MODULE_DEVICE_TABLE(acpi, acpi_nfit_ids); + +static struct acpi_driver acpi_nfit_driver = { + .name = KBUILD_MODNAME, + .ids = acpi_nfit_ids, + .ops = { + .add = acpi_nfit_add, + }, +}; + +static __init int nfit_init(void) +{ + int ret; + + BUILD_BUG_ON(sizeof(struct acpi_table_nfit) != 40); + BUILD_BUG_ON(sizeof(struct acpi_nfit_system_address) != 64); + BUILD_BUG_ON(sizeof(struct acpi_nfit_memory_map) != 48); + BUILD_BUG_ON(sizeof(struct acpi_nfit_interleave) != 16); + BUILD_BUG_ON(sizeof(struct acpi_nfit_smbios) != 8); + BUILD_BUG_ON(sizeof(struct acpi_nfit_control_region) != 80); + BUILD_BUG_ON(sizeof(struct acpi_nfit_data_region) != 40); + BUILD_BUG_ON(sizeof(struct acpi_nfit_capabilities) != 16); + + guid_parse(UUID_VOLATILE_MEMORY, &nfit_uuid[NFIT_SPA_VOLATILE]); + guid_parse(UUID_PERSISTENT_MEMORY, &nfit_uuid[NFIT_SPA_PM]); + guid_parse(UUID_CONTROL_REGION, &nfit_uuid[NFIT_SPA_DCR]); + guid_parse(UUID_DATA_REGION, &nfit_uuid[NFIT_SPA_BDW]); + guid_parse(UUID_VOLATILE_VIRTUAL_DISK, &nfit_uuid[NFIT_SPA_VDISK]); + guid_parse(UUID_VOLATILE_VIRTUAL_CD, &nfit_uuid[NFIT_SPA_VCD]); + guid_parse(UUID_PERSISTENT_VIRTUAL_DISK, &nfit_uuid[NFIT_SPA_PDISK]); + guid_parse(UUID_PERSISTENT_VIRTUAL_CD, &nfit_uuid[NFIT_SPA_PCD]); + guid_parse(UUID_NFIT_BUS, &nfit_uuid[NFIT_DEV_BUS]); + guid_parse(UUID_NFIT_DIMM, &nfit_uuid[NFIT_DEV_DIMM]); + guid_parse(UUID_NFIT_DIMM_N_HPE1, &nfit_uuid[NFIT_DEV_DIMM_N_HPE1]); + guid_parse(UUID_NFIT_DIMM_N_HPE2, &nfit_uuid[NFIT_DEV_DIMM_N_HPE2]); + guid_parse(UUID_NFIT_DIMM_N_MSFT, &nfit_uuid[NFIT_DEV_DIMM_N_MSFT]); + guid_parse(UUID_NFIT_DIMM_N_HYPERV, &nfit_uuid[NFIT_DEV_DIMM_N_HYPERV]); + guid_parse(UUID_INTEL_BUS, &nfit_uuid[NFIT_BUS_INTEL]); + + nfit_wq = create_singlethread_workqueue("nfit"); + if (!nfit_wq) + return -ENOMEM; + + nfit_mce_register(); + ret = acpi_bus_register_driver(&acpi_nfit_driver); + if (ret) { + nfit_mce_unregister(); + destroy_workqueue(nfit_wq); + } + + return ret; + +} + +static __exit void nfit_exit(void) +{ + nfit_mce_unregister(); + acpi_bus_unregister_driver(&acpi_nfit_driver); + destroy_workqueue(nfit_wq); + WARN_ON(!list_empty(&acpi_descs)); +} + +module_init(nfit_init); +module_exit(nfit_exit); +MODULE_LICENSE("GPL v2"); +MODULE_AUTHOR("Intel Corporation"); diff --git a/drivers/acpi/nfit/intel.c b/drivers/acpi/nfit/intel.c new file mode 100644 index 0000000000..3902759abc --- /dev/null +++ b/drivers/acpi/nfit/intel.c @@ -0,0 +1,749 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright(c) 2018 Intel Corporation. All rights reserved. */ +#include <linux/libnvdimm.h> +#include <linux/ndctl.h> +#include <linux/acpi.h> +#include <linux/memregion.h> +#include <asm/smp.h> +#include "intel.h" +#include "nfit.h" + +static ssize_t firmware_activate_noidle_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct nvdimm_bus *nvdimm_bus = to_nvdimm_bus(dev); + struct nvdimm_bus_descriptor *nd_desc = to_nd_desc(nvdimm_bus); + struct acpi_nfit_desc *acpi_desc = to_acpi_desc(nd_desc); + + return sprintf(buf, "%s\n", acpi_desc->fwa_noidle ? "Y" : "N"); +} + +static ssize_t firmware_activate_noidle_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t size) +{ + struct nvdimm_bus *nvdimm_bus = to_nvdimm_bus(dev); + struct nvdimm_bus_descriptor *nd_desc = to_nd_desc(nvdimm_bus); + struct acpi_nfit_desc *acpi_desc = to_acpi_desc(nd_desc); + ssize_t rc; + bool val; + + rc = kstrtobool(buf, &val); + if (rc) + return rc; + if (val != acpi_desc->fwa_noidle) + acpi_desc->fwa_cap = NVDIMM_FWA_CAP_INVALID; + acpi_desc->fwa_noidle = val; + return size; +} +DEVICE_ATTR_RW(firmware_activate_noidle); + +bool intel_fwa_supported(struct nvdimm_bus *nvdimm_bus) +{ + struct nvdimm_bus_descriptor *nd_desc = to_nd_desc(nvdimm_bus); + struct acpi_nfit_desc *acpi_desc = to_acpi_desc(nd_desc); + unsigned long *mask; + + if (!test_bit(NVDIMM_BUS_FAMILY_INTEL, &nd_desc->bus_family_mask)) + return false; + + mask = &acpi_desc->family_dsm_mask[NVDIMM_BUS_FAMILY_INTEL]; + return *mask == NVDIMM_BUS_INTEL_FW_ACTIVATE_CMDMASK; +} + +static unsigned long intel_security_flags(struct nvdimm *nvdimm, + enum nvdimm_passphrase_type ptype) +{ + struct nfit_mem *nfit_mem = nvdimm_provider_data(nvdimm); + unsigned long security_flags = 0; + 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_out = + sizeof(struct nd_intel_get_security_state), + .nd_fw_size = + sizeof(struct nd_intel_get_security_state), + }, + }; + int rc; + + if (!test_bit(NVDIMM_INTEL_GET_SECURITY_STATE, &nfit_mem->dsm_mask)) + return 0; + + /* + * Short circuit the state retrieval while we are doing overwrite. + * The DSM spec states that the security state is indeterminate + * until the overwrite DSM completes. + */ + if (nvdimm_in_overwrite(nvdimm) && ptype == NVDIMM_USER) + return BIT(NVDIMM_SECURITY_OVERWRITE); + + rc = nvdimm_ctl(nvdimm, ND_CMD_CALL, &nd_cmd, sizeof(nd_cmd), NULL); + if (rc < 0 || nd_cmd.cmd.status) { + pr_err("%s: security state retrieval failed (%d:%#x)\n", + nvdimm_name(nvdimm), rc, nd_cmd.cmd.status); + return 0; + } + + /* check and see if security is enabled and locked */ + if (ptype == NVDIMM_MASTER) { + if (nd_cmd.cmd.extended_state & ND_INTEL_SEC_ESTATE_ENABLED) + set_bit(NVDIMM_SECURITY_UNLOCKED, &security_flags); + else + set_bit(NVDIMM_SECURITY_DISABLED, &security_flags); + if (nd_cmd.cmd.extended_state & ND_INTEL_SEC_ESTATE_PLIMIT) + set_bit(NVDIMM_SECURITY_FROZEN, &security_flags); + return security_flags; + } + + if (nd_cmd.cmd.state & ND_INTEL_SEC_STATE_UNSUPPORTED) + return 0; + + if (nd_cmd.cmd.state & ND_INTEL_SEC_STATE_ENABLED) { + if (nd_cmd.cmd.state & ND_INTEL_SEC_STATE_FROZEN || + nd_cmd.cmd.state & ND_INTEL_SEC_STATE_PLIMIT) + set_bit(NVDIMM_SECURITY_FROZEN, &security_flags); + + if (nd_cmd.cmd.state & ND_INTEL_SEC_STATE_LOCKED) + set_bit(NVDIMM_SECURITY_LOCKED, &security_flags); + else + set_bit(NVDIMM_SECURITY_UNLOCKED, &security_flags); + } else + set_bit(NVDIMM_SECURITY_DISABLED, &security_flags); + + return security_flags; +} + +static int intel_security_freeze(struct nvdimm *nvdimm) +{ + 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_out = ND_INTEL_STATUS_SIZE, + .nd_fw_size = ND_INTEL_STATUS_SIZE, + }, + }; + int rc; + + if (!test_bit(NVDIMM_INTEL_FREEZE_LOCK, &nfit_mem->dsm_mask)) + return -ENOTTY; + + rc = nvdimm_ctl(nvdimm, ND_CMD_CALL, &nd_cmd, sizeof(nd_cmd), NULL); + if (rc < 0) + return rc; + if (nd_cmd.cmd.status) + return -EIO; + return 0; +} + +static int intel_security_change_key(struct nvdimm *nvdimm, + const struct nvdimm_key_data *old_data, + const struct nvdimm_key_data *new_data, + enum nvdimm_passphrase_type ptype) +{ + struct nfit_mem *nfit_mem = nvdimm_provider_data(nvdimm); + unsigned int cmd = ptype == NVDIMM_MASTER ? + NVDIMM_INTEL_SET_MASTER_PASSPHRASE : + NVDIMM_INTEL_SET_PASSPHRASE; + struct { + struct nd_cmd_pkg pkg; + struct nd_intel_set_passphrase cmd; + } nd_cmd = { + .pkg = { + .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, + .nd_command = cmd, + }, + }; + int rc; + + if (!test_bit(cmd, &nfit_mem->dsm_mask)) + return -ENOTTY; + + memcpy(nd_cmd.cmd.old_pass, old_data->data, + sizeof(nd_cmd.cmd.old_pass)); + memcpy(nd_cmd.cmd.new_pass, new_data->data, + sizeof(nd_cmd.cmd.new_pass)); + rc = nvdimm_ctl(nvdimm, ND_CMD_CALL, &nd_cmd, sizeof(nd_cmd), NULL); + if (rc < 0) + return rc; + + switch (nd_cmd.cmd.status) { + case 0: + return 0; + case ND_INTEL_STATUS_INVALID_PASS: + return -EINVAL; + case ND_INTEL_STATUS_NOT_SUPPORTED: + return -EOPNOTSUPP; + case ND_INTEL_STATUS_INVALID_STATE: + default: + return -EIO; + } +} + +static int __maybe_unused intel_security_unlock(struct nvdimm *nvdimm, + const struct nvdimm_key_data *key_data) +{ + 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, + }, + }; + int rc; + + if (!test_bit(NVDIMM_INTEL_UNLOCK_UNIT, &nfit_mem->dsm_mask)) + return -ENOTTY; + + memcpy(nd_cmd.cmd.passphrase, key_data->data, + sizeof(nd_cmd.cmd.passphrase)); + rc = nvdimm_ctl(nvdimm, ND_CMD_CALL, &nd_cmd, sizeof(nd_cmd), NULL); + if (rc < 0) + return rc; + switch (nd_cmd.cmd.status) { + case 0: + break; + case ND_INTEL_STATUS_INVALID_PASS: + return -EINVAL; + default: + return -EIO; + } + + return 0; +} + +static int intel_security_disable(struct nvdimm *nvdimm, + const struct nvdimm_key_data *key_data) +{ + int rc; + 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, + }, + }; + + if (!test_bit(NVDIMM_INTEL_DISABLE_PASSPHRASE, &nfit_mem->dsm_mask)) + return -ENOTTY; + + memcpy(nd_cmd.cmd.passphrase, key_data->data, + sizeof(nd_cmd.cmd.passphrase)); + rc = nvdimm_ctl(nvdimm, ND_CMD_CALL, &nd_cmd, sizeof(nd_cmd), NULL); + if (rc < 0) + return rc; + + switch (nd_cmd.cmd.status) { + case 0: + break; + case ND_INTEL_STATUS_INVALID_PASS: + return -EINVAL; + case ND_INTEL_STATUS_INVALID_STATE: + default: + return -ENXIO; + } + + return 0; +} + +static int __maybe_unused intel_security_erase(struct nvdimm *nvdimm, + const struct nvdimm_key_data *key, + enum nvdimm_passphrase_type ptype) +{ + int rc; + struct nfit_mem *nfit_mem = nvdimm_provider_data(nvdimm); + unsigned int cmd = ptype == NVDIMM_MASTER ? + NVDIMM_INTEL_MASTER_SECURE_ERASE : NVDIMM_INTEL_SECURE_ERASE; + struct { + struct nd_cmd_pkg pkg; + struct nd_intel_secure_erase cmd; + } nd_cmd = { + .pkg = { + .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, + .nd_command = cmd, + }, + }; + + if (!test_bit(cmd, &nfit_mem->dsm_mask)) + return -ENOTTY; + + memcpy(nd_cmd.cmd.passphrase, key->data, + sizeof(nd_cmd.cmd.passphrase)); + rc = nvdimm_ctl(nvdimm, ND_CMD_CALL, &nd_cmd, sizeof(nd_cmd), NULL); + if (rc < 0) + return rc; + + switch (nd_cmd.cmd.status) { + case 0: + break; + case ND_INTEL_STATUS_NOT_SUPPORTED: + return -EOPNOTSUPP; + case ND_INTEL_STATUS_INVALID_PASS: + return -EINVAL; + case ND_INTEL_STATUS_INVALID_STATE: + default: + return -ENXIO; + } + + return 0; +} + +static int __maybe_unused intel_security_query_overwrite(struct nvdimm *nvdimm) +{ + int rc; + struct nfit_mem *nfit_mem = nvdimm_provider_data(nvdimm); + struct { + struct nd_cmd_pkg pkg; + struct nd_intel_query_overwrite cmd; + } nd_cmd = { + .pkg = { + .nd_command = NVDIMM_INTEL_QUERY_OVERWRITE, + .nd_family = NVDIMM_FAMILY_INTEL, + .nd_size_out = ND_INTEL_STATUS_SIZE, + .nd_fw_size = ND_INTEL_STATUS_SIZE, + }, + }; + + if (!test_bit(NVDIMM_INTEL_QUERY_OVERWRITE, &nfit_mem->dsm_mask)) + return -ENOTTY; + + rc = nvdimm_ctl(nvdimm, ND_CMD_CALL, &nd_cmd, sizeof(nd_cmd), NULL); + if (rc < 0) + return rc; + + switch (nd_cmd.cmd.status) { + case 0: + break; + case ND_INTEL_STATUS_OQUERY_INPROGRESS: + return -EBUSY; + default: + return -ENXIO; + } + + return 0; +} + +static int __maybe_unused intel_security_overwrite(struct nvdimm *nvdimm, + const struct nvdimm_key_data *nkey) +{ + int rc; + struct nfit_mem *nfit_mem = nvdimm_provider_data(nvdimm); + struct { + struct nd_cmd_pkg pkg; + struct nd_intel_overwrite cmd; + } nd_cmd = { + .pkg = { + .nd_command = NVDIMM_INTEL_OVERWRITE, + .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, + }, + }; + + if (!test_bit(NVDIMM_INTEL_OVERWRITE, &nfit_mem->dsm_mask)) + return -ENOTTY; + + memcpy(nd_cmd.cmd.passphrase, nkey->data, + sizeof(nd_cmd.cmd.passphrase)); + rc = nvdimm_ctl(nvdimm, ND_CMD_CALL, &nd_cmd, sizeof(nd_cmd), NULL); + if (rc < 0) + return rc; + + switch (nd_cmd.cmd.status) { + case 0: + return 0; + case ND_INTEL_STATUS_OVERWRITE_UNSUPPORTED: + return -ENOTSUPP; + case ND_INTEL_STATUS_INVALID_PASS: + return -EINVAL; + case ND_INTEL_STATUS_INVALID_STATE: + default: + return -ENXIO; + } +} + +static const struct nvdimm_security_ops __intel_security_ops = { + .get_flags = intel_security_flags, + .freeze = intel_security_freeze, + .change_key = intel_security_change_key, + .disable = intel_security_disable, +#ifdef CONFIG_X86 + .unlock = intel_security_unlock, + .erase = intel_security_erase, + .overwrite = intel_security_overwrite, + .query_overwrite = intel_security_query_overwrite, +#endif +}; + +const struct nvdimm_security_ops *intel_security_ops = &__intel_security_ops; + +static int intel_bus_fwa_businfo(struct nvdimm_bus_descriptor *nd_desc, + struct nd_intel_bus_fw_activate_businfo *info) +{ + struct { + struct nd_cmd_pkg pkg; + struct nd_intel_bus_fw_activate_businfo cmd; + } nd_cmd = { + .pkg = { + .nd_command = NVDIMM_BUS_INTEL_FW_ACTIVATE_BUSINFO, + .nd_family = NVDIMM_BUS_FAMILY_INTEL, + .nd_size_out = + sizeof(struct nd_intel_bus_fw_activate_businfo), + .nd_fw_size = + sizeof(struct nd_intel_bus_fw_activate_businfo), + }, + }; + int rc; + + rc = nd_desc->ndctl(nd_desc, NULL, ND_CMD_CALL, &nd_cmd, sizeof(nd_cmd), + NULL); + *info = nd_cmd.cmd; + return rc; +} + +/* The fw_ops expect to be called with the nvdimm_bus_lock() held */ +static enum nvdimm_fwa_state intel_bus_fwa_state( + struct nvdimm_bus_descriptor *nd_desc) +{ + struct acpi_nfit_desc *acpi_desc = to_acpi_desc(nd_desc); + struct nd_intel_bus_fw_activate_businfo info; + struct device *dev = acpi_desc->dev; + enum nvdimm_fwa_state state; + int rc; + + /* + * It should not be possible for platform firmware to return + * busy because activate is a synchronous operation. Treat it + * similar to invalid, i.e. always refresh / poll the status. + */ + switch (acpi_desc->fwa_state) { + case NVDIMM_FWA_INVALID: + case NVDIMM_FWA_BUSY: + break; + default: + /* check if capability needs to be refreshed */ + if (acpi_desc->fwa_cap == NVDIMM_FWA_CAP_INVALID) + break; + return acpi_desc->fwa_state; + } + + /* Refresh with platform firmware */ + rc = intel_bus_fwa_businfo(nd_desc, &info); + if (rc) + return NVDIMM_FWA_INVALID; + + switch (info.state) { + case ND_INTEL_FWA_IDLE: + state = NVDIMM_FWA_IDLE; + break; + case ND_INTEL_FWA_BUSY: + state = NVDIMM_FWA_BUSY; + break; + case ND_INTEL_FWA_ARMED: + if (info.activate_tmo > info.max_quiesce_tmo) + state = NVDIMM_FWA_ARM_OVERFLOW; + else + state = NVDIMM_FWA_ARMED; + break; + default: + dev_err_once(dev, "invalid firmware activate state %d\n", + info.state); + return NVDIMM_FWA_INVALID; + } + + /* + * Capability data is available in the same payload as state. It + * is expected to be static. + */ + if (acpi_desc->fwa_cap == NVDIMM_FWA_CAP_INVALID) { + if (info.capability & ND_INTEL_BUS_FWA_CAP_FWQUIESCE) + acpi_desc->fwa_cap = NVDIMM_FWA_CAP_QUIESCE; + else if (info.capability & ND_INTEL_BUS_FWA_CAP_OSQUIESCE) { + /* + * Skip hibernate cycle by default if platform + * indicates that it does not need devices to be + * quiesced. + */ + acpi_desc->fwa_cap = NVDIMM_FWA_CAP_LIVE; + } else + acpi_desc->fwa_cap = NVDIMM_FWA_CAP_NONE; + } + + acpi_desc->fwa_state = state; + + return state; +} + +static enum nvdimm_fwa_capability intel_bus_fwa_capability( + struct nvdimm_bus_descriptor *nd_desc) +{ + struct acpi_nfit_desc *acpi_desc = to_acpi_desc(nd_desc); + + if (acpi_desc->fwa_cap > NVDIMM_FWA_CAP_INVALID) + return acpi_desc->fwa_cap; + + if (intel_bus_fwa_state(nd_desc) > NVDIMM_FWA_INVALID) + return acpi_desc->fwa_cap; + + return NVDIMM_FWA_CAP_INVALID; +} + +static int intel_bus_fwa_activate(struct nvdimm_bus_descriptor *nd_desc) +{ + struct acpi_nfit_desc *acpi_desc = to_acpi_desc(nd_desc); + struct { + struct nd_cmd_pkg pkg; + struct nd_intel_bus_fw_activate cmd; + } nd_cmd = { + .pkg = { + .nd_command = NVDIMM_BUS_INTEL_FW_ACTIVATE, + .nd_family = NVDIMM_BUS_FAMILY_INTEL, + .nd_size_in = sizeof(nd_cmd.cmd.iodev_state), + .nd_size_out = + sizeof(struct nd_intel_bus_fw_activate), + .nd_fw_size = + sizeof(struct nd_intel_bus_fw_activate), + }, + /* + * Even though activate is run from a suspended context, + * for safety, still ask platform firmware to force + * quiesce devices by default. Let a module + * parameter override that policy. + */ + .cmd = { + .iodev_state = acpi_desc->fwa_noidle + ? ND_INTEL_BUS_FWA_IODEV_OS_IDLE + : ND_INTEL_BUS_FWA_IODEV_FORCE_IDLE, + }, + }; + int rc; + + switch (intel_bus_fwa_state(nd_desc)) { + case NVDIMM_FWA_ARMED: + case NVDIMM_FWA_ARM_OVERFLOW: + break; + default: + return -ENXIO; + } + + rc = nd_desc->ndctl(nd_desc, NULL, ND_CMD_CALL, &nd_cmd, sizeof(nd_cmd), + NULL); + + /* + * Whether the command succeeded, or failed, the agent checking + * for the result needs to query the DIMMs individually. + * Increment the activation count to invalidate all the DIMM + * states at once (it's otherwise not possible to take + * acpi_desc->init_mutex in this context) + */ + acpi_desc->fwa_state = NVDIMM_FWA_INVALID; + acpi_desc->fwa_count++; + + dev_dbg(acpi_desc->dev, "result: %d\n", rc); + + return rc; +} + +static const struct nvdimm_bus_fw_ops __intel_bus_fw_ops = { + .activate_state = intel_bus_fwa_state, + .capability = intel_bus_fwa_capability, + .activate = intel_bus_fwa_activate, +}; + +const struct nvdimm_bus_fw_ops *intel_bus_fw_ops = &__intel_bus_fw_ops; + +static int intel_fwa_dimminfo(struct nvdimm *nvdimm, + struct nd_intel_fw_activate_dimminfo *info) +{ + struct { + struct nd_cmd_pkg pkg; + struct nd_intel_fw_activate_dimminfo cmd; + } nd_cmd = { + .pkg = { + .nd_command = NVDIMM_INTEL_FW_ACTIVATE_DIMMINFO, + .nd_family = NVDIMM_FAMILY_INTEL, + .nd_size_out = + sizeof(struct nd_intel_fw_activate_dimminfo), + .nd_fw_size = + sizeof(struct nd_intel_fw_activate_dimminfo), + }, + }; + int rc; + + rc = nvdimm_ctl(nvdimm, ND_CMD_CALL, &nd_cmd, sizeof(nd_cmd), NULL); + *info = nd_cmd.cmd; + return rc; +} + +static enum nvdimm_fwa_state intel_fwa_state(struct nvdimm *nvdimm) +{ + struct nfit_mem *nfit_mem = nvdimm_provider_data(nvdimm); + struct acpi_nfit_desc *acpi_desc = nfit_mem->acpi_desc; + struct nd_intel_fw_activate_dimminfo info; + int rc; + + /* + * Similar to the bus state, since activate is synchronous the + * busy state should resolve within the context of 'activate'. + */ + switch (nfit_mem->fwa_state) { + case NVDIMM_FWA_INVALID: + case NVDIMM_FWA_BUSY: + break; + default: + /* If no activations occurred the old state is still valid */ + if (nfit_mem->fwa_count == acpi_desc->fwa_count) + return nfit_mem->fwa_state; + } + + rc = intel_fwa_dimminfo(nvdimm, &info); + if (rc) + return NVDIMM_FWA_INVALID; + + switch (info.state) { + case ND_INTEL_FWA_IDLE: + nfit_mem->fwa_state = NVDIMM_FWA_IDLE; + break; + case ND_INTEL_FWA_BUSY: + nfit_mem->fwa_state = NVDIMM_FWA_BUSY; + break; + case ND_INTEL_FWA_ARMED: + nfit_mem->fwa_state = NVDIMM_FWA_ARMED; + break; + default: + nfit_mem->fwa_state = NVDIMM_FWA_INVALID; + break; + } + + switch (info.result) { + case ND_INTEL_DIMM_FWA_NONE: + nfit_mem->fwa_result = NVDIMM_FWA_RESULT_NONE; + break; + case ND_INTEL_DIMM_FWA_SUCCESS: + nfit_mem->fwa_result = NVDIMM_FWA_RESULT_SUCCESS; + break; + case ND_INTEL_DIMM_FWA_NOTSTAGED: + nfit_mem->fwa_result = NVDIMM_FWA_RESULT_NOTSTAGED; + break; + case ND_INTEL_DIMM_FWA_NEEDRESET: + nfit_mem->fwa_result = NVDIMM_FWA_RESULT_NEEDRESET; + break; + case ND_INTEL_DIMM_FWA_MEDIAFAILED: + case ND_INTEL_DIMM_FWA_ABORT: + case ND_INTEL_DIMM_FWA_NOTSUPP: + case ND_INTEL_DIMM_FWA_ERROR: + default: + nfit_mem->fwa_result = NVDIMM_FWA_RESULT_FAIL; + break; + } + + nfit_mem->fwa_count = acpi_desc->fwa_count; + + return nfit_mem->fwa_state; +} + +static enum nvdimm_fwa_result intel_fwa_result(struct nvdimm *nvdimm) +{ + struct nfit_mem *nfit_mem = nvdimm_provider_data(nvdimm); + struct acpi_nfit_desc *acpi_desc = nfit_mem->acpi_desc; + + if (nfit_mem->fwa_count == acpi_desc->fwa_count + && nfit_mem->fwa_result > NVDIMM_FWA_RESULT_INVALID) + return nfit_mem->fwa_result; + + if (intel_fwa_state(nvdimm) > NVDIMM_FWA_INVALID) + return nfit_mem->fwa_result; + + return NVDIMM_FWA_RESULT_INVALID; +} + +static int intel_fwa_arm(struct nvdimm *nvdimm, enum nvdimm_fwa_trigger arm) +{ + struct nfit_mem *nfit_mem = nvdimm_provider_data(nvdimm); + struct acpi_nfit_desc *acpi_desc = nfit_mem->acpi_desc; + struct { + struct nd_cmd_pkg pkg; + struct nd_intel_fw_activate_arm cmd; + } nd_cmd = { + .pkg = { + .nd_command = NVDIMM_INTEL_FW_ACTIVATE_ARM, + .nd_family = NVDIMM_FAMILY_INTEL, + .nd_size_in = sizeof(nd_cmd.cmd.activate_arm), + .nd_size_out = + sizeof(struct nd_intel_fw_activate_arm), + .nd_fw_size = + sizeof(struct nd_intel_fw_activate_arm), + }, + .cmd = { + .activate_arm = arm == NVDIMM_FWA_ARM + ? ND_INTEL_DIMM_FWA_ARM + : ND_INTEL_DIMM_FWA_DISARM, + }, + }; + int rc; + + switch (intel_fwa_state(nvdimm)) { + case NVDIMM_FWA_INVALID: + return -ENXIO; + case NVDIMM_FWA_BUSY: + return -EBUSY; + case NVDIMM_FWA_IDLE: + if (arm == NVDIMM_FWA_DISARM) + return 0; + break; + case NVDIMM_FWA_ARMED: + if (arm == NVDIMM_FWA_ARM) + return 0; + break; + default: + return -ENXIO; + } + + /* + * Invalidate the bus-level state, now that we're committed to + * changing the 'arm' state. + */ + acpi_desc->fwa_state = NVDIMM_FWA_INVALID; + nfit_mem->fwa_state = NVDIMM_FWA_INVALID; + + rc = nvdimm_ctl(nvdimm, ND_CMD_CALL, &nd_cmd, sizeof(nd_cmd), NULL); + + dev_dbg(acpi_desc->dev, "%s result: %d\n", arm == NVDIMM_FWA_ARM + ? "arm" : "disarm", rc); + return rc; +} + +static const struct nvdimm_fw_ops __intel_fw_ops = { + .activate_state = intel_fwa_state, + .activate_result = intel_fwa_result, + .arm = intel_fwa_arm, +}; + +const struct nvdimm_fw_ops *intel_fw_ops = &__intel_fw_ops; diff --git a/drivers/acpi/nfit/intel.h b/drivers/acpi/nfit/intel.h new file mode 100644 index 0000000000..b768234cce --- /dev/null +++ b/drivers/acpi/nfit/intel.h @@ -0,0 +1,175 @@ +// 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_ + +#define ND_INTEL_SMART 1 + +#define ND_INTEL_SMART_SHUTDOWN_COUNT_VALID (1 << 5) +#define ND_INTEL_SMART_SHUTDOWN_VALID (1 << 10) + +struct nd_intel_smart { + u32 status; + union { + struct { + u32 flags; + u8 reserved0[4]; + u8 health; + u8 spares; + u8 life_used; + u8 alarm_flags; + u16 media_temperature; + u16 ctrl_temperature; + u32 shutdown_count; + u8 ait_status; + u16 pmic_temperature; + u8 reserved1[8]; + u8 shutdown_state; + u32 vendor_size; + u8 vendor_data[92]; + } __packed; + u8 data[128]; + }; +} __packed; + +extern const struct nvdimm_security_ops *intel_security_ops; + +#define ND_INTEL_STATUS_SIZE 4 +#define ND_INTEL_PASSPHRASE_SIZE 32 + +#define ND_INTEL_STATUS_NOT_SUPPORTED 1 +#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_STATUS_OVERWRITE_UNSUPPORTED 0x10007 +#define ND_INTEL_STATUS_OQUERY_INPROGRESS 0x10007 +#define ND_INTEL_STATUS_OQUERY_SEQUENCE_ERR 0x20007 + +#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 +#define ND_INTEL_SEC_STATE_OVERWRITE 0x40 + +#define ND_INTEL_SEC_ESTATE_ENABLED 0x01 +#define ND_INTEL_SEC_ESTATE_PLIMIT 0x02 + +struct nd_intel_get_security_state { + u32 status; + u8 extended_state; + u8 reserved[3]; + 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; + +struct nd_intel_set_master_passphrase { + u8 old_pass[ND_INTEL_PASSPHRASE_SIZE]; + u8 new_pass[ND_INTEL_PASSPHRASE_SIZE]; + u32 status; +} __packed; + +struct nd_intel_master_secure_erase { + u8 passphrase[ND_INTEL_PASSPHRASE_SIZE]; + u32 status; +} __packed; + +#define ND_INTEL_FWA_IDLE 0 +#define ND_INTEL_FWA_ARMED 1 +#define ND_INTEL_FWA_BUSY 2 + +#define ND_INTEL_DIMM_FWA_NONE 0 +#define ND_INTEL_DIMM_FWA_NOTSTAGED 1 +#define ND_INTEL_DIMM_FWA_SUCCESS 2 +#define ND_INTEL_DIMM_FWA_NEEDRESET 3 +#define ND_INTEL_DIMM_FWA_MEDIAFAILED 4 +#define ND_INTEL_DIMM_FWA_ABORT 5 +#define ND_INTEL_DIMM_FWA_NOTSUPP 6 +#define ND_INTEL_DIMM_FWA_ERROR 7 + +struct nd_intel_fw_activate_dimminfo { + u32 status; + u16 result; + u8 state; + u8 reserved[7]; +} __packed; + +#define ND_INTEL_DIMM_FWA_ARM 1 +#define ND_INTEL_DIMM_FWA_DISARM 0 + +struct nd_intel_fw_activate_arm { + u8 activate_arm; + u32 status; +} __packed; + +/* Root device command payloads */ +#define ND_INTEL_BUS_FWA_CAP_FWQUIESCE (1 << 0) +#define ND_INTEL_BUS_FWA_CAP_OSQUIESCE (1 << 1) +#define ND_INTEL_BUS_FWA_CAP_RESET (1 << 2) + +struct nd_intel_bus_fw_activate_businfo { + u32 status; + u16 reserved; + u8 state; + u8 capability; + u64 activate_tmo; + u64 cpu_quiesce_tmo; + u64 io_quiesce_tmo; + u64 max_quiesce_tmo; +} __packed; + +#define ND_INTEL_BUS_FWA_STATUS_NOARM (6 | 1 << 16) +#define ND_INTEL_BUS_FWA_STATUS_BUSY (6 | 2 << 16) +#define ND_INTEL_BUS_FWA_STATUS_NOFW (6 | 3 << 16) +#define ND_INTEL_BUS_FWA_STATUS_TMO (6 | 4 << 16) +#define ND_INTEL_BUS_FWA_STATUS_NOIDLE (6 | 5 << 16) +#define ND_INTEL_BUS_FWA_STATUS_ABORT (6 | 6 << 16) + +#define ND_INTEL_BUS_FWA_IODEV_FORCE_IDLE (0) +#define ND_INTEL_BUS_FWA_IODEV_OS_IDLE (1) +struct nd_intel_bus_fw_activate { + u8 iodev_state; + u32 status; +} __packed; + +extern const struct nvdimm_fw_ops *intel_fw_ops; +extern const struct nvdimm_bus_fw_ops *intel_bus_fw_ops; +#endif diff --git a/drivers/acpi/nfit/mce.c b/drivers/acpi/nfit/mce.c new file mode 100644 index 0000000000..d48a388b79 --- /dev/null +++ b/drivers/acpi/nfit/mce.c @@ -0,0 +1,100 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * NFIT - Machine Check Handler + * + * Copyright(c) 2013-2016 Intel Corporation. All rights reserved. + */ +#include <linux/notifier.h> +#include <linux/acpi.h> +#include <linux/nd.h> +#include <asm/mce.h> +#include "nfit.h" + +static int nfit_handle_mce(struct notifier_block *nb, unsigned long val, + void *data) +{ + struct mce *mce = (struct mce *)data; + struct acpi_nfit_desc *acpi_desc; + struct nfit_spa *nfit_spa; + + /* We only care about uncorrectable memory errors */ + if (!mce_is_memory_error(mce) || mce_is_correctable(mce)) + return NOTIFY_DONE; + + /* Verify the address reported in the MCE is valid. */ + if (!mce_usable_address(mce)) + return NOTIFY_DONE; + + /* + * mce->addr contains the physical addr accessed that caused the + * machine check. We need to walk through the list of NFITs, and see + * if any of them matches that address, and only then start a scrub. + */ + mutex_lock(&acpi_desc_lock); + list_for_each_entry(acpi_desc, &acpi_descs, list) { + unsigned int align = 1UL << MCI_MISC_ADDR_LSB(mce->misc); + struct device *dev = acpi_desc->dev; + int found_match = 0; + + mutex_lock(&acpi_desc->init_mutex); + list_for_each_entry(nfit_spa, &acpi_desc->spas, list) { + struct acpi_nfit_system_address *spa = nfit_spa->spa; + + if (nfit_spa_type(spa) != NFIT_SPA_PM) + continue; + /* find the spa that covers the mce addr */ + if (spa->address > mce->addr) + continue; + if ((spa->address + spa->length - 1) < mce->addr) + continue; + found_match = 1; + dev_dbg(dev, "addr in SPA %d (0x%llx, 0x%llx)\n", + spa->range_index, spa->address, spa->length); + /* + * We can break at the first match because we're going + * to rescan all the SPA ranges. There shouldn't be any + * aliasing anyway. + */ + break; + } + mutex_unlock(&acpi_desc->init_mutex); + + if (!found_match) + continue; + + /* If this fails due to an -ENOMEM, there is little we can do */ + nvdimm_bus_add_badrange(acpi_desc->nvdimm_bus, + ALIGN_DOWN(mce->addr, align), align); + nvdimm_region_notify(nfit_spa->nd_region, + NVDIMM_REVALIDATE_POISON); + + if (acpi_desc->scrub_mode == HW_ERROR_SCRUB_ON) { + /* + * We can ignore an -EBUSY here because if an ARS is + * already in progress, just let that be the last + * authoritative one + */ + acpi_nfit_ars_rescan(acpi_desc, 0); + } + mce->kflags |= MCE_HANDLED_NFIT; + break; + } + + mutex_unlock(&acpi_desc_lock); + return NOTIFY_DONE; +} + +static struct notifier_block nfit_mce_dec = { + .notifier_call = nfit_handle_mce, + .priority = MCE_PRIO_NFIT, +}; + +void nfit_mce_register(void) +{ + mce_register_decode_chain(&nfit_mce_dec); +} + +void nfit_mce_unregister(void) +{ + mce_unregister_decode_chain(&nfit_mce_dec); +} diff --git a/drivers/acpi/nfit/nfit.h b/drivers/acpi/nfit/nfit.h new file mode 100644 index 0000000000..573bc0de29 --- /dev/null +++ b/drivers/acpi/nfit/nfit.h @@ -0,0 +1,352 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * NVDIMM Firmware Interface Table - NFIT + * + * Copyright(c) 2013-2015 Intel Corporation. All rights reserved. + */ +#ifndef __NFIT_H__ +#define __NFIT_H__ +#include <linux/workqueue.h> +#include <linux/libnvdimm.h> +#include <linux/ndctl.h> +#include <linux/types.h> +#include <linux/acpi.h> +#include <acpi/acuuid.h> + +/* ACPI 6.1 */ +#define UUID_NFIT_BUS "2f10e7a4-9e91-11e4-89d3-123b93f75cba" + +/* https://pmem.io/documents/NVDIMM_DSM_Interface-V1.6.pdf */ +#define UUID_NFIT_DIMM "4309ac30-0d11-11e4-9191-0800200c9a66" +#define UUID_INTEL_BUS "c7d8acd4-2df8-4b82-9f65-a325335af149" + +/* https://github.com/HewlettPackard/hpe-nvm/blob/master/Documentation/ */ +#define UUID_NFIT_DIMM_N_HPE1 "9002c334-acf3-4c0e-9642-a235f0d53bc6" +#define UUID_NFIT_DIMM_N_HPE2 "5008664b-b758-41a0-a03c-27c2f2d04f7e" + +/* https://msdn.microsoft.com/library/windows/hardware/mt604741 */ +#define UUID_NFIT_DIMM_N_MSFT "1ee68b36-d4bd-4a1a-9a16-4f8e53d46e05" + +/* http://www.uefi.org/RFIC_LIST (see "Virtual NVDIMM 0x1901") */ +#define UUID_NFIT_DIMM_N_HYPERV "5746c5f2-a9a2-4264-ad0e-e4ddc9e09e80" + +#define ACPI_NFIT_MEM_FAILED_MASK (ACPI_NFIT_MEM_SAVE_FAILED \ + | ACPI_NFIT_MEM_RESTORE_FAILED | ACPI_NFIT_MEM_FLUSH_FAILED \ + | ACPI_NFIT_MEM_NOT_ARMED | ACPI_NFIT_MEM_MAP_FAILED) + +#define NVDIMM_CMD_MAX 31 + +#define NVDIMM_STANDARD_CMDMASK \ +(1 << ND_CMD_SMART | 1 << ND_CMD_SMART_THRESHOLD | 1 << ND_CMD_DIMM_FLAGS \ + | 1 << ND_CMD_GET_CONFIG_SIZE | 1 << ND_CMD_GET_CONFIG_DATA \ + | 1 << ND_CMD_SET_CONFIG_DATA | 1 << ND_CMD_VENDOR_EFFECT_LOG_SIZE \ + | 1 << ND_CMD_VENDOR_EFFECT_LOG | 1 << ND_CMD_VENDOR) + +/* + * Command numbers that the kernel needs to know about to handle + * non-default DSM revision ids + */ +enum nvdimm_family_cmds { + NVDIMM_INTEL_LATCH_SHUTDOWN = 10, + NVDIMM_INTEL_GET_MODES = 11, + NVDIMM_INTEL_GET_FWINFO = 12, + NVDIMM_INTEL_START_FWUPDATE = 13, + NVDIMM_INTEL_SEND_FWUPDATE = 14, + NVDIMM_INTEL_FINISH_FWUPDATE = 15, + 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, + NVDIMM_INTEL_SET_MASTER_PASSPHRASE = 27, + NVDIMM_INTEL_MASTER_SECURE_ERASE = 28, + NVDIMM_INTEL_FW_ACTIVATE_DIMMINFO = 29, + NVDIMM_INTEL_FW_ACTIVATE_ARM = 30, +}; + +enum nvdimm_bus_family_cmds { + NVDIMM_BUS_INTEL_FW_ACTIVATE_BUSINFO = 1, + NVDIMM_BUS_INTEL_FW_ACTIVATE = 2, +}; + +#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 \ +| 1 << NVDIMM_INTEL_SET_MASTER_PASSPHRASE \ +| 1 << NVDIMM_INTEL_MASTER_SECURE_ERASE) + +#define NVDIMM_INTEL_FW_ACTIVATE_CMDMASK \ +(1 << NVDIMM_INTEL_FW_ACTIVATE_DIMMINFO | 1 << NVDIMM_INTEL_FW_ACTIVATE_ARM) + +#define NVDIMM_BUS_INTEL_FW_ACTIVATE_CMDMASK \ +(1 << NVDIMM_BUS_INTEL_FW_ACTIVATE_BUSINFO | 1 << NVDIMM_BUS_INTEL_FW_ACTIVATE) + +#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 \ + | NVDIMM_INTEL_SECURITY_CMDMASK | NVDIMM_INTEL_FW_ACTIVATE_CMDMASK) + +#define NVDIMM_INTEL_DENY_CMDMASK \ +(NVDIMM_INTEL_SECURITY_CMDMASK | NVDIMM_INTEL_FW_ACTIVATE_CMDMASK) + +enum nfit_uuids { + /* for simplicity alias the uuid index with the family id */ + NFIT_DEV_DIMM = NVDIMM_FAMILY_INTEL, + NFIT_DEV_DIMM_N_HPE1 = NVDIMM_FAMILY_HPE1, + NFIT_DEV_DIMM_N_HPE2 = NVDIMM_FAMILY_HPE2, + NFIT_DEV_DIMM_N_MSFT = NVDIMM_FAMILY_MSFT, + NFIT_DEV_DIMM_N_HYPERV = NVDIMM_FAMILY_HYPERV, + /* + * to_nfit_bus_uuid() expects to translate bus uuid family ids + * to a UUID index using NVDIMM_FAMILY_MAX as an offset + */ + NFIT_BUS_INTEL = NVDIMM_FAMILY_MAX + NVDIMM_BUS_FAMILY_INTEL, + NFIT_SPA_VOLATILE, + NFIT_SPA_PM, + NFIT_SPA_DCR, + NFIT_SPA_BDW, + NFIT_SPA_VDISK, + NFIT_SPA_VCD, + NFIT_SPA_PDISK, + NFIT_SPA_PCD, + NFIT_DEV_BUS, + NFIT_UUID_MAX, +}; + +/* + * Region format interface codes are stored with the interface as the + * LSB and the function as the MSB. + */ +#define NFIT_FIC_BYTE cpu_to_le16(0x101) /* byte-addressable energy backed */ +#define NFIT_FIC_BLK cpu_to_le16(0x201) /* block-addressable non-energy backed */ +#define NFIT_FIC_BYTEN cpu_to_le16(0x301) /* byte-addressable non-energy backed */ + +enum { + NFIT_BLK_READ_FLUSH = 1, + NFIT_BLK_DCR_LATCH = 2, + NFIT_ARS_STATUS_DONE = 0, + NFIT_ARS_STATUS_BUSY = 1 << 16, + NFIT_ARS_STATUS_NONE = 2 << 16, + NFIT_ARS_STATUS_INTR = 3 << 16, + NFIT_ARS_START_BUSY = 6, + NFIT_ARS_CAP_NONE = 1, + NFIT_ARS_F_OVERFLOW = 1, + NFIT_ARS_TIMEOUT = 90, +}; + +enum nfit_root_notifiers { + NFIT_NOTIFY_UPDATE = 0x80, + NFIT_NOTIFY_UC_MEMORY_ERROR = 0x81, +}; + +enum nfit_dimm_notifiers { + NFIT_NOTIFY_DIMM_HEALTH = 0x81, +}; + +enum nfit_ars_state { + ARS_REQ_SHORT, + ARS_REQ_LONG, + ARS_FAILED, +}; + +struct nfit_spa { + struct list_head list; + struct nd_region *nd_region; + unsigned long ars_state; + u32 clear_err_unit; + u32 max_ars; + struct acpi_nfit_system_address spa[]; +}; + +struct nfit_dcr { + struct list_head list; + struct acpi_nfit_control_region dcr[]; +}; + +struct nfit_bdw { + struct list_head list; + struct acpi_nfit_data_region bdw[]; +}; + +struct nfit_idt { + struct list_head list; + struct acpi_nfit_interleave idt[]; +}; + +struct nfit_flush { + struct list_head list; + struct acpi_nfit_flush_address flush[]; +}; + +struct nfit_memdev { + struct list_head list; + struct acpi_nfit_memory_map memdev[]; +}; + +enum nfit_mem_flags { + NFIT_MEM_LSR, + NFIT_MEM_LSW, + NFIT_MEM_DIRTY, + NFIT_MEM_DIRTY_COUNT, +}; + +#define NFIT_DIMM_ID_LEN 22 + +/* assembled tables for a given dimm/memory-device */ +struct nfit_mem { + struct nvdimm *nvdimm; + struct acpi_nfit_memory_map *memdev_dcr; + struct acpi_nfit_memory_map *memdev_pmem; + struct acpi_nfit_control_region *dcr; + struct acpi_nfit_system_address *spa_dcr; + struct acpi_nfit_interleave *idt_dcr; + struct kernfs_node *flags_attr; + struct nfit_flush *nfit_flush; + struct list_head list; + struct acpi_device *adev; + struct acpi_nfit_desc *acpi_desc; + enum nvdimm_fwa_state fwa_state; + enum nvdimm_fwa_result fwa_result; + int fwa_count; + char id[NFIT_DIMM_ID_LEN+1]; + struct resource *flush_wpq; + unsigned long dsm_mask; + unsigned long flags; + u32 dirty_shutdown; + int family; +}; + +enum scrub_flags { + ARS_BUSY, + ARS_CANCEL, + ARS_VALID, + ARS_POLL, +}; + +struct acpi_nfit_desc { + struct nvdimm_bus_descriptor nd_desc; + struct acpi_table_header acpi_header; + struct mutex init_mutex; + struct list_head memdevs; + struct list_head flushes; + struct list_head dimms; + struct list_head spas; + struct list_head dcrs; + struct list_head bdws; + struct list_head idts; + struct nvdimm_bus *nvdimm_bus; + struct device *dev; + struct nd_cmd_ars_status *ars_status; + struct nfit_spa *scrub_spa; + struct delayed_work dwork; + struct list_head list; + struct kernfs_node *scrub_count_state; + unsigned int max_ars; + unsigned int scrub_count; + unsigned int scrub_mode; + unsigned long scrub_flags; + unsigned long dimm_cmd_force_en; + unsigned long bus_cmd_force_en; + unsigned long bus_dsm_mask; + unsigned long family_dsm_mask[NVDIMM_BUS_FAMILY_MAX + 1]; + unsigned int platform_cap; + unsigned int scrub_tmo; + enum nvdimm_fwa_state fwa_state; + enum nvdimm_fwa_capability fwa_cap; + int fwa_count; + bool fwa_noidle; + bool fwa_nosuspend; +}; + +enum scrub_mode { + HW_ERROR_SCRUB_OFF, + HW_ERROR_SCRUB_ON, +}; + +enum nd_blk_mmio_selector { + BDW, + DCR, +}; + +struct nd_blk_addr { + union { + void __iomem *base; + void *aperture; + }; +}; + +struct nfit_blk { + struct nfit_blk_mmio { + struct nd_blk_addr addr; + u64 size; + u64 base_offset; + u32 line_size; + u32 num_lines; + u32 table_size; + struct acpi_nfit_interleave *idt; + struct acpi_nfit_system_address *spa; + } mmio[2]; + struct nd_region *nd_region; + u64 bdw_offset; /* post interleave offset */ + u64 stat_offset; + u64 cmd_offset; + u32 dimm_flags; +}; + +extern struct list_head acpi_descs; +extern struct mutex acpi_desc_lock; +int acpi_nfit_ars_rescan(struct acpi_nfit_desc *acpi_desc, + enum nfit_ars_state req_type); + +#ifdef CONFIG_X86_MCE +void nfit_mce_register(void); +void nfit_mce_unregister(void); +#else +static inline void nfit_mce_register(void) +{ +} +static inline void nfit_mce_unregister(void) +{ +} +#endif + +int nfit_spa_type(struct acpi_nfit_system_address *spa); + +static inline struct acpi_nfit_memory_map *__to_nfit_memdev( + struct nfit_mem *nfit_mem) +{ + if (nfit_mem->memdev_dcr) + return nfit_mem->memdev_dcr; + return nfit_mem->memdev_pmem; +} + +static inline struct acpi_nfit_desc *to_acpi_desc( + struct nvdimm_bus_descriptor *nd_desc) +{ + return container_of(nd_desc, struct acpi_nfit_desc, nd_desc); +} + +const guid_t *to_nfit_uuid(enum nfit_uuids id); +int acpi_nfit_init(struct acpi_nfit_desc *acpi_desc, void *nfit, acpi_size sz); +void acpi_nfit_shutdown(void *data); +void __acpi_nfit_notify(struct device *dev, acpi_handle handle, u32 event); +void __acpi_nvdimm_notify(struct device *dev, u32 event); +int acpi_nfit_ctl(struct nvdimm_bus_descriptor *nd_desc, struct nvdimm *nvdimm, + unsigned int cmd, void *buf, unsigned int buf_len, int *cmd_rc); +void acpi_nfit_desc_init(struct acpi_nfit_desc *acpi_desc, struct device *dev); +bool intel_fwa_supported(struct nvdimm_bus *nvdimm_bus); +extern struct device_attribute dev_attr_firmware_activate_noidle; +void nfit_intel_shutdown_status(struct nfit_mem *nfit_mem); + +#endif /* __NFIT_H__ */ |