diff options
| author | Dmitry Shmidt <dimitrysh@google.com> | 2016-08-12 14:20:42 -0700 |
|---|---|---|
| committer | Dmitry Shmidt <dimitrysh@google.com> | 2016-08-12 14:20:42 -0700 |
| commit | 8e4f993802d3d853d5377b9aaa8df28926140077 (patch) | |
| tree | 3e7f0636ce97b7f5006ff1f2c774b034d5b630cc | |
| parent | 1f369b24e2987c1b29c88ffcbd0ca99cddabe01c (diff) | |
| parent | aa3cda16a57e73e9fc269e1de1eb3c35f3f0ed20 (diff) | |
Merge remote-tracking branch 'common/android-4.4' into android-4.4.y
| -rw-r--r-- | Documentation/device-mapper/boot.txt | 42 | ||||
| -rw-r--r-- | Documentation/devicetree/bindings/misc/memory-state-time.txt | 8 | ||||
| -rw-r--r-- | Documentation/kernel-parameters.txt | 6 | ||||
| -rw-r--r-- | android/configs/android-recommended.cfg | 1 | ||||
| -rw-r--r-- | drivers/md/Kconfig | 16 | ||||
| -rw-r--r-- | drivers/md/Makefile | 4 | ||||
| -rw-r--r-- | drivers/md/dm-android-verity.c | 928 | ||||
| -rw-r--r-- | drivers/md/dm-android-verity.h | 121 | ||||
| -rw-r--r-- | drivers/md/dm-ioctl.c | 39 | ||||
| -rw-r--r-- | drivers/md/dm-linear.c | 24 | ||||
| -rw-r--r-- | drivers/md/dm-table.c | 1 | ||||
| -rw-r--r-- | drivers/md/dm-verity-target.c | 14 | ||||
| -rw-r--r-- | drivers/md/dm-verity.h | 10 | ||||
| -rw-r--r-- | drivers/misc/Kconfig | 6 | ||||
| -rw-r--r-- | drivers/misc/Makefile | 1 | ||||
| -rw-r--r-- | drivers/misc/memory_state_time.c | 454 | ||||
| -rw-r--r-- | fs/ecryptfs/file.c | 71 | ||||
| -rw-r--r-- | fs/proc/base.c | 34 | ||||
| -rw-r--r-- | include/linux/device-mapper.h | 6 | ||||
| -rw-r--r-- | include/linux/memory-state-time.h | 42 | ||||
| -rw-r--r-- | include/net/fib_rules.h | 1 | ||||
| -rw-r--r-- | include/uapi/linux/rtnetlink.h | 2 | ||||
| -rw-r--r-- | init/Makefile | 1 | ||||
| -rw-r--r-- | init/do_mounts.c | 1 | ||||
| -rw-r--r-- | init/do_mounts.h | 10 | ||||
| -rw-r--r-- | init/do_mounts_dm.c | 426 |
26 files changed, 2217 insertions, 52 deletions
diff --git a/Documentation/device-mapper/boot.txt b/Documentation/device-mapper/boot.txt new file mode 100644 index 000000000000..adcaad5e5e32 --- /dev/null +++ b/Documentation/device-mapper/boot.txt @@ -0,0 +1,42 @@ +Boot time creation of mapped devices +=================================== + +It is possible to configure a device mapper device to act as the root +device for your system in two ways. + +The first is to build an initial ramdisk which boots to a minimal +userspace which configures the device, then pivot_root(8) in to it. + +For simple device mapper configurations, it is possible to boot directly +using the following kernel command line: + +dm="<name> <uuid> <ro>,table line 1,...,table line n" + +name = the name to associate with the device + after boot, udev, if used, will use that name to label + the device node. +uuid = may be 'none' or the UUID desired for the device. +ro = may be "ro" or "rw". If "ro", the device and device table will be + marked read-only. + +Each table line may be as normal when using the dmsetup tool except for +two variations: +1. Any use of commas will be interpreted as a newline +2. Quotation marks cannot be escaped and cannot be used without + terminating the dm= argument. + +Unless renamed by udev, the device node created will be dm-0 as the +first minor number for the device-mapper is used during early creation. + +Example +======= + +- Booting to a linear array made up of user-mode linux block devices: + + dm="lroot none 0, 0 4096 linear 98:16 0, 4096 4096 linear 98:32 0" \ + root=/dev/dm-0 + +Will boot to a rw dm-linear target of 8192 sectors split across two +block devices identified by their major:minor numbers. After boot, udev +will rename this target to /dev/mapper/lroot (depending on the rules). +No uuid was assigned. diff --git a/Documentation/devicetree/bindings/misc/memory-state-time.txt b/Documentation/devicetree/bindings/misc/memory-state-time.txt new file mode 100644 index 000000000000..c99a506c030d --- /dev/null +++ b/Documentation/devicetree/bindings/misc/memory-state-time.txt @@ -0,0 +1,8 @@ +Memory bandwidth and frequency state tracking + +Required properties: +- compatible : should be: + "memory-state-time" +- freq-tbl: Should contain entries with each frequency in Hz. +- bw-buckets: Should contain upper-bound limits for each bandwidth bucket in Mbps. + Must match the framework power_profile.xml for the device. diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 0e4102ae1a61..dfdb38df7a03 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -56,6 +56,7 @@ parameter is applicable: BLACKFIN Blackfin architecture is enabled. CLK Common clock infrastructure is enabled. CMA Contiguous Memory Area support is enabled. + DM Device mapper support is enabled. DRM Direct Rendering Management support is enabled. DYNAMIC_DEBUG Build in debug messages and enable them at runtime EDD BIOS Enhanced Disk Drive Services (EDD) is enabled @@ -915,6 +916,11 @@ bytes respectively. Such letter suffixes can also be entirely omitted. dis_ucode_ldr [X86] Disable the microcode loader. + dm= [DM] Allows early creation of a device-mapper device. + See Documentation/device-mapper/boot.txt. + + dmasound= [HW,OSS] Sound subsystem buff + dma_debug=off If the kernel is compiled with DMA_API_DEBUG support, this option disables the debugging code at boot. diff --git a/android/configs/android-recommended.cfg b/android/configs/android-recommended.cfg index 2f1ef077aa9e..3465a848d74d 100644 --- a/android/configs/android-recommended.cfg +++ b/android/configs/android-recommended.cfg @@ -119,6 +119,7 @@ CONFIG_TIMER_STATS=y CONFIG_TMPFS=y CONFIG_TMPFS_POSIX_ACL=y CONFIG_UHID=y +CONFIG_MEMORY_STATE_TIME=y CONFIG_USB_ANNOUNCE_NEW_DEVICES=y CONFIG_USB_EHCI_HCD=y CONFIG_USB_HIDDEV=y diff --git a/drivers/md/Kconfig b/drivers/md/Kconfig index d8b0ab6f3753..96b419b544ed 100644 --- a/drivers/md/Kconfig +++ b/drivers/md/Kconfig @@ -500,4 +500,20 @@ config DM_LOG_WRITES If unsure, say N. +config DM_ANDROID_VERITY + bool "Android verity target support" + depends on DM_VERITY + depends on X509_CERTIFICATE_PARSER + depends on SYSTEM_TRUSTED_KEYRING + depends on PUBLIC_KEY_ALGO_RSA + depends on KEYS + depends on ASYMMETRIC_KEY_TYPE + depends on ASYMMETRIC_PUBLIC_KEY_SUBTYPE + ---help--- + This device-mapper target is virtually a VERITY target. This + target is setup by reading the metadata contents piggybacked + to the actual data blocks in the block device. The signature + of the metadata contents are verified against the key included + in the system keyring. Upon success, the underlying verity + target is setup. endif # MD diff --git a/drivers/md/Makefile b/drivers/md/Makefile index 62a65764e8e0..c8fb00d8cc36 100644 --- a/drivers/md/Makefile +++ b/drivers/md/Makefile @@ -68,3 +68,7 @@ endif ifeq ($(CONFIG_DM_VERITY_FEC),y) dm-verity-objs += dm-verity-fec.o endif + +ifeq ($(CONFIG_DM_ANDROID_VERITY),y) +dm-verity-objs += dm-android-verity.o +endif diff --git a/drivers/md/dm-android-verity.c b/drivers/md/dm-android-verity.c new file mode 100644 index 000000000000..15ce2a81c1f4 --- /dev/null +++ b/drivers/md/dm-android-verity.c @@ -0,0 +1,928 @@ +/* + * Copyright (C) 2015 Google, Inc. + * + * This software is licensed under the terms of the GNU General Public + * License version 2, as published by the Free Software Foundation, and + * may be copied, distributed, and modified under those terms. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + */ + +#include <linux/buffer_head.h> +#include <linux/debugfs.h> +#include <linux/delay.h> +#include <linux/device.h> +#include <linux/device-mapper.h> +#include <linux/errno.h> +#include <linux/fs.h> +#include <linux/fcntl.h> +#include <linux/init.h> +#include <linux/kernel.h> +#include <linux/key.h> +#include <linux/module.h> +#include <linux/mount.h> +#include <linux/namei.h> +#include <linux/of.h> +#include <linux/reboot.h> +#include <linux/string.h> +#include <linux/vmalloc.h> + +#include <asm/setup.h> +#include <crypto/hash.h> +#include <crypto/public_key.h> +#include <crypto/sha.h> +#include <keys/asymmetric-type.h> +#include <keys/system_keyring.h> + +#include "dm-verity.h" +#include "dm-android-verity.h" + +static char verifiedbootstate[VERITY_COMMANDLINE_PARAM_LENGTH]; +static char veritymode[VERITY_COMMANDLINE_PARAM_LENGTH]; +static char veritykeyid[VERITY_DEFAULT_KEY_ID_LENGTH]; +static char buildvariant[BUILD_VARIANT]; + +static bool target_added; +static bool verity_enabled = true; +struct dentry *debug_dir; +static int android_verity_ctr(struct dm_target *ti, unsigned argc, char **argv); + +static struct target_type android_verity_target = { + .name = "android-verity", + .version = {1, 0, 0}, + .module = THIS_MODULE, + .ctr = android_verity_ctr, + .dtr = verity_dtr, + .map = verity_map, + .status = verity_status, + .prepare_ioctl = verity_prepare_ioctl, + .iterate_devices = verity_iterate_devices, + .io_hints = verity_io_hints, +}; + +static int __init verified_boot_state_param(char *line) +{ + strlcpy(verifiedbootstate, line, sizeof(verifiedbootstate)); + return 1; +} + +__setup("androidboot.verifiedbootstate=", verified_boot_state_param); + +static int __init verity_mode_param(char *line) +{ + strlcpy(veritymode, line, sizeof(veritymode)); + return 1; +} + +__setup("androidboot.veritymode=", verity_mode_param); + +static int __init verity_keyid_param(char *line) +{ + strlcpy(veritykeyid, line, sizeof(veritykeyid)); + return 1; +} + +__setup("veritykeyid=", verity_keyid_param); + +static int __init verity_buildvariant(char *line) +{ + strlcpy(buildvariant, line, sizeof(buildvariant)); + return 1; +} + +__setup("buildvariant=", verity_buildvariant); + +static inline bool default_verity_key_id(void) +{ + return veritykeyid[0] != '\0'; +} + +static inline bool is_eng(void) +{ + static const char typeeng[] = "eng"; + + return !strncmp(buildvariant, typeeng, sizeof(typeeng)); +} + +static inline bool is_userdebug(void) +{ + static const char typeuserdebug[] = "userdebug"; + + return !strncmp(buildvariant, typeuserdebug, sizeof(typeuserdebug)); +} + + +static int table_extract_mpi_array(struct public_key_signature *pks, + const void *data, size_t len) +{ + MPI mpi = mpi_read_raw_data(data, len); + + if (!mpi) { + DMERR("Error while allocating mpi array"); + return -ENOMEM; + } + + pks->mpi[0] = mpi; + pks->nr_mpi = 1; + return 0; +} + +static struct public_key_signature *table_make_digest( + enum hash_algo hash, + const void *table, + unsigned long table_len) +{ + struct public_key_signature *pks = NULL; + struct crypto_shash *tfm; + struct shash_desc *desc; + size_t digest_size, desc_size; + int ret; + + /* Allocate the hashing algorithm we're going to need and find out how + * big the hash operational data will be. + */ + tfm = crypto_alloc_shash(hash_algo_name[hash], 0, 0); + if (IS_ERR(tfm)) + return ERR_CAST(tfm); + + desc_size = crypto_shash_descsize(tfm) + sizeof(*desc); + digest_size = crypto_shash_digestsize(tfm); + + /* We allocate the hash operational data storage on the end of out + * context data and the digest output buffer on the end of that. + */ + ret = -ENOMEM; + pks = kzalloc(digest_size + sizeof(*pks) + desc_size, GFP_KERNEL); + if (!pks) + goto error; + + pks->pkey_hash_algo = hash; + pks->digest = (u8 *)pks + sizeof(*pks) + desc_size; + pks->digest_size = digest_size; + + desc = (struct shash_desc *)(pks + 1); + desc->tfm = tfm; + desc->flags = CRYPTO_TFM_REQ_MAY_SLEEP; + + ret = crypto_shash_init(desc); + if (ret < 0) + goto error; + + ret = crypto_shash_finup(desc, table, table_len, pks->digest); + if (ret < 0) + goto error; + + crypto_free_shash(tfm); + return pks; + +error: + kfree(pks); + crypto_free_shash(tfm); + return ERR_PTR(ret); +} + +static int read_block_dev(struct bio_read *payload, struct block_device *bdev, + sector_t offset, int length) +{ + struct bio *bio; + int err = 0, i; + + payload->number_of_pages = DIV_ROUND_UP(length, PAGE_SIZE); + + bio = bio_alloc(GFP_KERNEL, payload->number_of_pages); + if (!bio) { + DMERR("Error while allocating bio"); + return -ENOMEM; + } + + bio->bi_bdev = bdev; + bio->bi_iter.bi_sector = offset; + + payload->page_io = kzalloc(sizeof(struct page *) * + payload->number_of_pages, GFP_KERNEL); + if (!payload->page_io) { + DMERR("page_io array alloc failed"); + err = -ENOMEM; + goto free_bio; + } + + for (i = 0; i < payload->number_of_pages; i++) { + payload->page_io[i] = alloc_page(GFP_KERNEL); + if (!payload->page_io[i]) { + DMERR("alloc_page failed"); + err = -ENOMEM; + goto free_pages; + } + if (!bio_add_page(bio, payload->page_io[i], PAGE_SIZE, 0)) { + DMERR("bio_add_page error"); + err = -EIO; + goto free_pages; + } + } + + if (!submit_bio_wait(READ, bio)) + /* success */ + goto free_bio; + DMERR("bio read failed"); + err = -EIO; + +free_pages: + for (i = 0; i < payload->number_of_pages; i++) + if (payload->page_io[i]) + __free_page(payload->page_io[i]); + kfree(payload->page_io); +free_bio: + bio_put(bio); + return err; +} + +static inline u64 fec_div_round_up(u64 x, u64 y) +{ + u64 remainder; + + return div64_u64_rem(x, y, &remainder) + + (remainder > 0 ? 1 : 0); +} + +static inline void populate_fec_metadata(struct fec_header *header, + struct fec_ecc_metadata *ecc) +{ + ecc->blocks = fec_div_round_up(le64_to_cpu(header->inp_size), + FEC_BLOCK_SIZE); + ecc->roots = le32_to_cpu(header->roots); + ecc->start = le64_to_cpu(header->inp_size); +} + +static inline int validate_fec_header(struct fec_header *header, u64 offset) +{ + /* move offset to make the sanity check work for backup header + * as well. */ + offset -= offset % FEC_BLOCK_SIZE; + if (le32_to_cpu(header->magic) != FEC_MAGIC || + le32_to_cpu(header->version) != FEC_VERSION || + le32_to_cpu(header->size) != sizeof(struct fec_header) || + le32_to_cpu(header->roots) == 0 || + le32_to_cpu(header->roots) >= FEC_RSM || + offset < le32_to_cpu(header->fec_size) || + offset - le32_to_cpu(header->fec_size) != + le64_to_cpu(header->inp_size)) + return -EINVAL; + + return 0; +} + +static int extract_fec_header(dev_t dev, struct fec_header *fec, + struct fec_ecc_metadata *ecc) +{ + u64 device_size; + struct bio_read payload; + int i, err = 0; + struct block_device *bdev; + + bdev = blkdev_get_by_dev(dev, FMODE_READ, NULL); + + if (IS_ERR_OR_NULL(bdev)) { + DMERR("bdev get error"); + return PTR_ERR(bdev); + } + + device_size = i_size_read(bdev->bd_inode); + + /* fec metadata size is a power of 2 and PAGE_SIZE + * is a power of 2 as well. + */ + BUG_ON(FEC_BLOCK_SIZE > PAGE_SIZE); + /* 512 byte sector alignment */ + BUG_ON(((device_size - FEC_BLOCK_SIZE) % (1 << SECTOR_SHIFT)) != 0); + + err = read_block_dev(&payload, bdev, (device_size - + FEC_BLOCK_SIZE) / (1 << SECTOR_SHIFT), FEC_BLOCK_SIZE); + if (err) { + DMERR("Error while reading verity metadata"); + goto error; + } + + BUG_ON(sizeof(struct fec_header) > PAGE_SIZE); + memcpy(fec, page_address(payload.page_io[0]), + sizeof(*fec)); + + ecc->valid = true; + if (validate_fec_header(fec, device_size - FEC_BLOCK_SIZE)) { + /* Try the backup header */ + memcpy(fec, page_address(payload.page_io[0]) + FEC_BLOCK_SIZE + - sizeof(*fec) , + sizeof(*fec)); + if (validate_fec_header(fec, device_size - + sizeof(struct fec_header))) + ecc->valid = false; + } + + if (ecc->valid) + populate_fec_metadata(fec, ecc); + + for (i = 0; i < payload.number_of_pages; i++) + __free_page(payload.page_io[i]); + kfree(payload.page_io); + +error: + blkdev_put(bdev, FMODE_READ); + return err; +} +static void find_metadata_offset(struct fec_header *fec, + struct block_device *bdev, u64 *metadata_offset) +{ + u64 device_size; + + device_size = i_size_read(bdev->bd_inode); + + if (le32_to_cpu(fec->magic) == FEC_MAGIC) + *metadata_offset = le64_to_cpu(fec->inp_size) - + VERITY_METADATA_SIZE; + else + *metadata_offset = device_size - VERITY_METADATA_SIZE; +} + +static int find_size(dev_t dev, u64 *device_size) +{ + struct block_device *bdev; + + bdev = blkdev_get_by_dev(dev, FMODE_READ, NULL); + if (IS_ERR_OR_NULL(bdev)) { + DMERR("blkdev_get_by_dev failed"); + return PTR_ERR(bdev); + } + + *device_size = i_size_read(bdev->bd_inode); + *device_size >>= SECTOR_SHIFT; + + DMINFO("blkdev size in sectors: %llu", *device_size); + blkdev_put(bdev, FMODE_READ); + return 0; +} + +static int verify_header(struct android_metadata_header *header) +{ + int retval = -EINVAL; + + if (is_userdebug() && le32_to_cpu(header->magic_number) == + VERITY_METADATA_MAGIC_DISABLE) + return VERITY_STATE_DISABLE; + + if (!(le32_to_cpu(header->magic_number) == + VERITY_METADATA_MAGIC_NUMBER) || + (le32_to_cpu(header->magic_number) == + VERITY_METADATA_MAGIC_DISABLE)) { + DMERR("Incorrect magic number"); + return retval; + } + + if (le32_to_cpu(header->protocol_version) != + VERITY_METADATA_VERSION) { + DMERR("Unsupported version %u", + le32_to_cpu(header->protocol_version)); + return retval; + } + + return 0; +} + +static int extract_metadata(dev_t dev, struct fec_header *fec, + struct android_metadata **metadata, + bool *verity_enabled) +{ + struct block_device *bdev; + struct android_metadata_header *header; + int i; + u32 table_length, copy_length, offset; + u64 metadata_offset; + struct bio_read payload; + int err = 0; + + bdev = blkdev_get_by_dev(dev, FMODE_READ, NULL); + + if (IS_ERR_OR_NULL(bdev)) { + DMERR("blkdev_get_by_dev failed"); + return -ENODEV; + } + + find_metadata_offset(fec, bdev, &metadata_offset); + + /* Verity metadata size is a power of 2 and PAGE_SIZE + * is a power of 2 as well. + * PAGE_SIZE is also a multiple of 512 bytes. + */ + if (VERITY_METADATA_SIZE > PAGE_SIZE) + BUG_ON(VERITY_METADATA_SIZE % PAGE_SIZE != 0); + /* 512 byte sector alignment */ + BUG_ON(metadata_offset % (1 << SECTOR_SHIFT) != 0); + + err = read_block_dev(&payload, bdev, metadata_offset / + (1 << SECTOR_SHIFT), VERITY_METADATA_SIZE); + if (err) { + DMERR("Error while reading verity metadata"); + goto blkdev_release; + } + + header = kzalloc(sizeof(*header), GFP_KERNEL); + if (!header) { + DMERR("kzalloc failed for header"); + err = -ENOMEM; + goto free_payload; + } + + memcpy(header, page_address(payload.page_io[0]), + sizeof(*header)); + + DMINFO("bio magic_number:%u protocol_version:%d table_length:%u", + le32_to_cpu(header->magic_number), + le32_to_cpu(header->protocol_version), + le32_to_cpu(header->table_length)); + + err = verify_header(header); + + if (err == VERITY_STATE_DISABLE) { + DMERR("Mounting root with verity disabled"); + *verity_enabled = false; + /* we would still have to read the metadata to figure out + * the data blocks size. Or may be could map the entire + * partition similar to mounting the device. + * + * Reset error as well as the verity_enabled flag is changed. + */ + err = 0; + } else if (err) + goto free_header; + + *metadata = kzalloc(sizeof(**metadata), GFP_KERNEL); + if (!*metadata) { + DMERR("kzalloc for metadata failed"); + err = -ENOMEM; + goto free_header; + } + + (*metadata)->header = header; + table_length = le32_to_cpu(header->table_length); + + if (table_length == 0 || + table_length > (VERITY_METADATA_SIZE - + sizeof(struct android_metadata_header))) { + DMERR("table_length too long"); + err = -EINVAL; + goto free_metadata; + } + + (*metadata)->verity_table = kzalloc(table_length + 1, GFP_KERNEL); + + if (!(*metadata)->verity_table) { + DMERR("kzalloc verity_table failed"); + err = -ENOMEM; + goto free_metadata; + } + + if (sizeof(struct android_metadata_header) + + table_length <= PAGE_SIZE) { + memcpy((*metadata)->verity_table, + page_address(payload.page_io[0]) + + sizeof(struct android_metadata_header), + table_length); + } else { + copy_length = PAGE_SIZE - + sizeof(struct android_metadata_header); + memcpy((*metadata)->verity_table, + page_address(payload.page_io[0]) + + sizeof(struct android_metadata_header), + copy_length); + table_length -= copy_length; + offset = copy_length; + i = 1; + while (table_length != 0) { + if (table_length > PAGE_SIZE) { + memcpy((*metadata)->verity_table + offset, + page_address(payload.page_io[i]), + PAGE_SIZE); + offset += PAGE_SIZE; + table_length -= PAGE_SIZE; + } else { + memcpy((*metadata)->verity_table + offset, + page_address(payload.page_io[i]), + table_length); + table_length = 0; + } + i++; + } + } + (*metadata)->verity_table[table_length] = '\0'; + + DMINFO("verity_table: %s", (*metadata)->verity_table); + goto free_payload; + +free_metadata: + kfree(*metadata); +free_header: + kfree(header); +free_payload: + for (i = 0; i < payload.number_of_pages; i++) + if (payload.page_io[i]) + __free_page(payload.page_io[i]); + kfree(payload.page_io); +blkdev_release: + blkdev_put(bdev, FMODE_READ); + return err; +} + +/* helper functions to extract properties from dts */ +const char *find_dt_value(const char *name) +{ + struct device_node *firmware; + const char *value; + + firmware = of_find_node_by_path("/firmware/android"); + if (!firmware) + return NULL; + value = of_get_property(firmware, name, NULL); + of_node_put(firmware); + + return value; +} + +static int verity_mode(void) +{ + static const char enforcing[] = "enforcing"; + static const char verified_mode_prop[] = "veritymode"; + const char *value; + + value = find_dt_value(verified_mode_prop); + if (!value) + value = veritymode; + if (!strncmp(value, enforcing, sizeof(enforcing) - 1)) + return DM_VERITY_MODE_RESTART; + + return DM_VERITY_MODE_EIO; +} + +static int verify_verity_signature(char *key_id, + struct android_metadata *metadata) +{ + key_ref_t key_ref; + struct key *key; + struct public_key_signature *pks = NULL; + int retval = -EINVAL; + + key_ref = keyring_search(make_key_ref(system_trusted_keyring, 1), + &key_type_asymmetric, key_id); + + if (IS_ERR(key_ref)) { + DMERR("keyring: key not found"); + return -ENOKEY; + } + + key = key_ref_to_ptr(key_ref); + + pks = table_make_digest(HASH_ALGO_SHA256, + (const void *)metadata->verity_table, + le32_to_cpu(metadata->header->table_length)); + + if (IS_ERR(pks)) { + DMERR("hashing failed"); + goto error; + } + + retval = table_extract_mpi_array(pks, &metadata->header->signature[0], + RSANUMBYTES); + if (retval < 0) { + DMERR("Error extracting mpi %d", retval); + goto error; + } + + retval = verify_signature(key, pks); + mpi_free(pks->rsa.s); +error: + kfree(pks); + key_put(key); + + return retval; +} + +static void handle_error(void) +{ + int mode = verity_mode(); + if (mode == DM_VERITY_MODE_RESTART) { + DMERR("triggering restart"); + kernel_restart("dm-verity device corrupted"); + } else { + DMERR("Mounting verity root failed"); + } +} + +static inline bool test_mult_overflow(sector_t a, u32 b) +{ + sector_t r = (sector_t)~0ULL; + + sector_div(r, b); + return a > r; +} + +static int add_as_linear_device(struct dm_target *ti, char *dev) +{ + /*Move to linear mapping defines*/ + char *linear_table_args[DM_LINEAR_ARGS] = {dev, + DM_LINEAR_TARGET_OFFSET}; + int err = 0; + + android_verity_target.dtr = dm_linear_dtr, + android_verity_target.map = dm_linear_map, + android_verity_target.status = dm_linear_status, + android_verity_target.prepare_ioctl = dm_linear_prepare_ioctl, + android_verity_target.iterate_devices = dm_linear_iterate_devices, + android_verity_target.io_hints = NULL; + + err = dm_linear_ctr(ti, DM_LINEAR_ARGS, linear_table_args); + + if (!err) { + DMINFO("Added android-verity as a linear target"); + target_added = true; + } else + DMERR("Failed to add android-verity as linear target"); + + return err; +} + +/* + * Target parameters: + * <key id> Key id of the public key in the system keyring. + * Verity metadata's signature would be verified against + * this. If the key id contains spaces, replace them + * with '#'. + * <block device> The block device for which dm-verity is being setup. + */ +static int android_verity_ctr(struct dm_target *ti, unsigned argc, char **argv) +{ + dev_t uninitialized_var(dev); + struct android_metadata *metadata = NULL; + int err = 0, i, mode; + char *key_id, *table_ptr, dummy, *target_device, + *verity_table_args[VERITY_TABLE_ARGS + 2 + VERITY_TABLE_OPT_FEC_ARGS]; + /* One for specifying number of opt args and one for mode */ + sector_t data_sectors; + u32 data_block_size; + unsigned int no_of_args = VERITY_TABLE_ARGS + 2 + VERITY_TABLE_OPT_FEC_ARGS; + struct fec_header uninitialized_var(fec); + struct fec_ecc_metadata uninitialized_var(ecc); + char buf[FEC_ARG_LENGTH], *buf_ptr; + unsigned long long tmpll; + u64 uninitialized_var(device_size); + + if (argc == 1) { + /* Use the default keyid */ + if (default_verity_key_id()) + key_id = veritykeyid; + else if (!is_eng()) { + DMERR("veritykeyid= is not set"); + handle_error(); + return -EINVAL; + } + } else if (argc == 2) + key_id = argv[1]; + else { + DMERR("Incorrect number of arguments"); + handle_error(); + return -EINVAL; + } + + target_device = argv[0]; + + dev = name_to_dev_t(target_device); + if (!dev) { + DMERR("no dev found for %s", target_device); + handle_error(); + return -EINVAL; + } + + if (is_eng()) { + err = find_size(dev, &device_size); + if (err) { + DMERR("error finding bdev size"); + handle_error(); + return err; + } + + ti->len = device_size; + err = add_as_linear_device(ti, target_device); + if (err) { + handle_error(); + return err; + } + verity_enabled = false; + return 0; + } + + strreplace(key_id, '#', ' '); + + DMINFO("key:%s dev:%s", key_id, target_device); + + if (extract_fec_header(dev, &fec, &ecc)) { + DMERR("Error while extracting fec header"); + handle_error(); + return -EINVAL; + } + + err = extract_metadata(dev, &fec, &metadata, &verity_enabled); + + if (err) { + DMERR("Error while extracting metadata"); + handle_error(); + goto free_metadata; + } + + if (verity_enabled) { + err = verify_verity_signature(key_id, metadata); + + if (err) { + DMERR("Signature verification failed"); + handle_error(); + goto free_metadata; + } else + DMINFO("Signature verification success"); + } + + table_ptr = metadata->verity_table; + + for (i = 0; i < VERITY_TABLE_ARGS; i++) { + verity_table_args[i] = strsep(&table_ptr, " "); + if (verity_table_args[i] == NULL) + break; + } + + if (i != VERITY_TABLE_ARGS) { + DMERR("Verity table not in the expected format"); + err = -EINVAL; + handle_error(); + goto free_metadata; + } + + if (sscanf(verity_table_args[5], "%llu%c", &tmpll, &dummy) + != 1) { + DMERR("Verity table not in the expected format"); + handle_error(); + err = -EINVAL; + goto free_metadata; + } + + if (tmpll > ULONG_MAX) { + DMERR("<num_data_blocks> too large. Forgot to turn on CONFIG_LBDAF?"); + handle_error(); + err = -EINVAL; + goto free_metadata; + } + + data_sectors = tmpll; + + if (sscanf(verity_table_args[3], "%u%c", &data_block_size, &dummy) + != 1) { + DMERR("Verity table not in the expected format"); + handle_error(); + err = -EINVAL; + goto free_metadata; + } + + if (test_mult_overflow(data_sectors, data_block_size >> + SECTOR_SHIFT)) { + DMERR("data_sectors too large"); + handle_error(); + err = -EOVERFLOW; + goto free_metadata; + } + + data_sectors *= data_block_size >> SECTOR_SHIFT; + DMINFO("Data sectors %llu", (unsigned long long)data_sectors); + + /* update target length */ + ti->len = data_sectors; + + /* Setup linear target and free */ + if (!verity_enabled) { + err = add_as_linear_device(ti, target_device); + goto free_metadata; + } + + /*substitute data_dev and hash_dev*/ + verity_table_args[1] = target_device; + verity_table_args[2] = target_device; + + mode = verity_mode(); + + if (ecc.valid && IS_BUILTIN(CONFIG_DM_VERITY_FEC)) { + if (mode) { + err = snprintf(buf, FEC_ARG_LENGTH, + "%u %s " VERITY_TABLE_OPT_FEC_FORMAT, + 1 + VERITY_TABLE_OPT_FEC_ARGS, + mode == DM_VERITY_MODE_RESTART ? + VERITY_TABLE_OPT_RESTART : + VERITY_TABLE_OPT_LOGGING, + target_device, + ecc.start / FEC_BLOCK_SIZE, ecc.blocks, + ecc.roots); + } else { + err = snprintf(buf, FEC_ARG_LENGTH, + "%u " VERITY_TABLE_OPT_FEC_FORMAT, + VERITY_TABLE_OPT_FEC_ARGS, target_device, + ecc.start / FEC_BLOCK_SIZE, ecc.blocks, + ecc.roots); + } + } else if (mode) { + err = snprintf(buf, FEC_ARG_LENGTH, + "2 " VERITY_TABLE_OPT_IGNZERO " %s", + mode == DM_VERITY_MODE_RESTART ? + VERITY_TABLE_OPT_RESTART : VERITY_TABLE_OPT_LOGGING); + } else { + err = snprintf(buf, FEC_ARG_LENGTH, "1 %s", + "ignore_zero_blocks"); + } + + if (err < 0 || err >= FEC_ARG_LENGTH) + goto free_metadata; + + buf_ptr = buf; + + for (i = VERITY_TABLE_ARGS; i < (VERITY_TABLE_ARGS + + VERITY_TABLE_OPT_FEC_ARGS + 2); i++) { + verity_table_args[i] = strsep(&buf_ptr, " "); + if (verity_table_args[i] == NULL) { + no_of_args = i; + break; + } + } + + err = verity_ctr(ti, no_of_args, verity_table_args); + + if (err) + DMERR("android-verity failed to mount as verity target"); + else { + target_added = true; + DMINFO("android-verity mounted as verity target"); + } + +free_metadata: + if (metadata) { + kfree(metadata->header); + kfree(metadata->verity_table); + } + kfree(metadata); + return err; +} + +static int __init dm_android_verity_init(void) +{ + int r; + struct dentry *file; + + r = dm_register_target(&android_verity_target); + if (r < 0) + DMERR("register failed %d", r); + + /* Tracks the status of the last added target */ + debug_dir = debugfs_create_dir("android_verity", NULL); + + if (IS_ERR_OR_NULL(debug_dir)) { + DMERR("Cannot create android_verity debugfs directory: %ld", + PTR_ERR(debug_dir)); + goto end; + } + + file = debugfs_create_bool("target_added", S_IRUGO, debug_dir, + &target_added); + + if (IS_ERR_OR_NULL(file)) { + DMERR("Cannot create android_verity debugfs directory: %ld", + PTR_ERR(debug_dir)); + debugfs_remove_recursive(debug_dir); + goto end; + } + + file = debugfs_create_bool("verity_enabled", S_IRUGO, debug_dir, + &verity_enabled); + + if (IS_ERR_OR_NULL(file)) { + DMERR("Cannot create android_verity debugfs directory: %ld", + PTR_ERR(debug_dir)); + debugfs_remove_recursive(debug_dir); + } + +end: + return r; +} + +static void __exit dm_android_verity_exit(void) +{ + if (!IS_ERR_OR_NULL(debug_dir)) + debugfs_remove_recursive(debug_dir); + + dm_unregister_target(&android_verity_target); +} + +module_init(dm_android_verity_init); +module_exit(dm_android_verity_exit); diff --git a/drivers/md/dm-android-verity.h b/drivers/md/dm-android-verity.h new file mode 100644 index 000000000000..0c7ff6afec69 --- /dev/null +++ b/drivers/md/dm-android-verity.h @@ -0,0 +1,121 @@ +/* + * Copyright (C) 2015 Google, Inc. + * + * This software is licensed under the terms of the GNU General Public + * License version 2, as published by the Free Software Foundation, and + * may be copied, distributed, and modified under those terms. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + */ + +#ifndef DM_ANDROID_VERITY_H +#define DM_ANDROID_VERITY_H + +#include <crypto/sha.h> + +#define RSANUMBYTES 256 +#define VERITY_METADATA_MAGIC_NUMBER 0xb001b001 +#define VERITY_METADATA_MAGIC_DISABLE 0x46464f56 +#define VERITY_METADATA_VERSION 0 +#define VERITY_STATE_DISABLE 1 +#define DATA_BLOCK_SIZE (4 * 1024) +#define VERITY_METADATA_SIZE (8 * DATA_BLOCK_SIZE) +#define VERITY_TABLE_ARGS 10 +#define VERITY_COMMANDLINE_PARAM_LENGTH 20 +#define BUILD_VARIANT 20 + +/* + * <subject>:<sha1-id> is the format for the identifier. + * subject can either be the Common Name(CN) + Organization Name(O) or + * just the CN if the it is prefixed with O + * From https://tools.ietf.org/html/rfc5280#appendix-A + * ub-organization-name-length INTEGER ::= 64 + * ub-common-name-length INTEGER ::= 64 + * + * http://lxr.free-electrons.com/source/crypto/asymmetric_keys/x509_cert_parser.c?v=3.9#L278 + * ctx->o_size + 2 + ctx->cn_size + 1 + * + 41 characters for ":" and sha1 id + * 64 + 2 + 64 + 1 + 1 + 40 (172) + * setting VERITY_DEFAULT_KEY_ID_LENGTH to 200 characters. + */ +#define VERITY_DEFAULT_KEY_ID_LENGTH 200 + +#define FEC_MAGIC 0xFECFECFE +#define FEC_BLOCK_SIZE (4 * 1024) +#define FEC_VERSION 0 +#define FEC_RSM 255 +#define FEC_ARG_LENGTH 300 + +#define VERITY_TABLE_OPT_RESTART "restart_on_corruption" +#define VERITY_TABLE_OPT_LOGGING "ignore_corruption" +#define VERITY_TABLE_OPT_IGNZERO "ignore_zero_blocks" + +#define VERITY_TABLE_OPT_FEC_FORMAT \ + "use_fec_from_device %s fec_start %llu fec_blocks %llu fec_roots %u ignore_zero_blocks" +#define VERITY_TABLE_OPT_FEC_ARGS 9 + +#define VERITY_DEBUG 0 + +#define DM_MSG_PREFIX "android-verity" + +#define DM_LINEAR_ARGS 2 +#define DM_LINEAR_TARGET_OFFSET "0" + +/* + * There can be two formats. + * if fec is present + * <data_blocks> <verity_tree> <verity_metdata_32K><fec_data><fec_data_4K> + * if fec is not present + * <data_blocks> <verity_tree> <verity_metdata_32K> + */ +struct fec_header { + __le32 magic; + __le32 version; + __le32 size; + __le32 roots; + __le32 fec_size; + __le64 inp_size; + u8 hash[SHA256_DIGEST_SIZE]; +} __attribute__((packed)); + +struct android_metadata_header { + __le32 magic_number; + __le32 protocol_version; + char signature[RSANUMBYTES]; + __le32 table_length; +}; + +struct android_metadata { + struct android_metadata_header *header; + char *verity_table; +}; + +struct fec_ecc_metadata { + bool valid; + u32 roots; + u64 blocks; + u64 rounds; + u64 start; +}; + +struct bio_read { + struct page **page_io; + int number_of_pages; +}; + +extern struct target_type linear_target; + +extern void dm_linear_dtr(struct dm_target *ti); +extern int dm_linear_map(struct dm_target *ti, struct bio *bio); +extern void dm_linear_status(struct dm_target *ti, status_type_t type, + unsigned status_flags, char *result, unsigned maxlen); +extern int dm_linear_prepare_ioctl(struct dm_target *ti, + struct block_device **bdev, fmode_t *mode); +extern int dm_linear_iterate_devices(struct dm_target *ti, + iterate_devices_callout_fn fn, void *data); +extern int dm_linear_ctr(struct dm_target *ti, unsigned int argc, char **argv); +#endif /* DM_ANDROID_VERITY_H */ diff --git a/drivers/md/dm-ioctl.c b/drivers/md/dm-ioctl.c index 80a439543259..bc5e9a5b1f30 100644 --- a/drivers/md/dm-ioctl.c +++ b/drivers/md/dm-ioctl.c @@ -1923,6 +1923,45 @@ void dm_interface_exit(void) dm_hash_exit(); } + +/** + * dm_ioctl_export - Permanently export a mapped device via the ioctl interface + * @md: Pointer to mapped_device + * @name: Buffer (size DM_NAME_LEN) for name + * @uuid: Buffer (size DM_UUID_LEN) for uuid or NULL if not desired + */ +int dm_ioctl_export(struct mapped_device *md, const char *name, + const char *uuid) +{ + int r = 0; + struct hash_cell *hc; + + if (!md) { + r = -ENXIO; + goto out; + } + + /* The name and uuid can only be set once. */ + mutex_lock(&dm_hash_cells_mutex); + hc = dm_get_mdptr(md); + mutex_unlock(&dm_hash_cells_mutex); + if (hc) { + DMERR("%s: already exported", dm_device_name(md)); + r = -ENXIO; + goto out; + } + + r = dm_hash_insert(name, uuid, md); + if (r) { + DMERR("%s: could not bind to '%s'", dm_device_name(md), name); + goto out; + } + + /* Let udev know we've changed. */ + dm_kobject_uevent(md, KOBJ_CHANGE, dm_get_event_nr(md)); +out: + return r; +} /** * dm_copy_name_and_uuid - Copy mapped device name & uuid into supplied buffers * @md: Pointer to mapped_device diff --git a/drivers/md/dm-linear.c b/drivers/md/dm-linear.c index 05c35aacb3aa..8505a771de42 100644 --- a/drivers/md/dm-linear.c +++ b/drivers/md/dm-linear.c @@ -25,7 +25,7 @@ struct linear_c { /* * Construct a linear mapping: <dev_path> <offset> */ -static int linear_ctr(struct dm_target *ti, unsigned int argc, char **argv) +int dm_linear_ctr(struct dm_target *ti, unsigned int argc, char **argv) { struct linear_c *lc; unsigned long long tmp; @@ -67,7 +67,7 @@ static int linear_ctr(struct dm_target *ti, unsigned int argc, char **argv) return ret; } -static void linear_dtr(struct dm_target *ti) +void dm_linear_dtr(struct dm_target *ti) { struct linear_c *lc = (struct linear_c *) ti->private; @@ -92,14 +92,14 @@ static void linear_map_bio(struct dm_target *ti, struct bio *bio) linear_map_sector(ti, bio->bi_iter.bi_sector); } -static int linear_map(struct dm_target *ti, struct bio *bio) +int dm_linear_map(struct dm_target *ti, struct bio *bio) { linear_map_bio(ti, bio); return DM_MAPIO_REMAPPED; } -static void linear_status(struct dm_target *ti, status_type_t type, +void dm_linear_status(struct dm_target *ti, status_type_t type, unsigned status_flags, char *result, unsigned maxlen) { struct linear_c *lc = (struct linear_c *) ti->private; @@ -116,7 +116,7 @@ static void linear_status(struct dm_target *ti, status_type_t type, } } -static int linear_prepare_ioctl(struct dm_target *ti, +int dm_linear_prepare_ioctl(struct dm_target *ti, struct block_device **bdev, fmode_t *mode) { struct linear_c *lc = (struct linear_c *) ti->private; @@ -133,7 +133,7 @@ static int linear_prepare_ioctl(struct dm_target *ti, return 0; } -static int linear_iterate_devices(struct dm_target *ti, +int dm_linear_iterate_devices(struct dm_target *ti, iterate_devices_callout_fn fn, void *data) { struct linear_c *lc = ti->private; @@ -145,12 +145,12 @@ static struct target_type linear_target = { .name = "linear", .version = {1, 2, 1}, .module = THIS_MODULE, - .ctr = linear_ctr, - .dtr = linear_dtr, - .map = linear_map, - .status = linear_status, - .prepare_ioctl = linear_prepare_ioctl, - .iterate_devices = linear_iterate_devices, + .ctr = dm_linear_ctr, + .dtr = dm_linear_dtr, + .map = dm_linear_map, + .status = dm_linear_status, + .prepare_ioctl = dm_linear_prepare_ioctl, + .iterate_devices = dm_linear_iterate_devices, }; int __init dm_linear_init(void) diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c index cb5d0daf53bb..b3d78bba3a79 100644 --- a/drivers/md/dm-table.c +++ b/drivers/md/dm-table.c @@ -11,6 +11,7 @@ #include <linux/vmalloc.h> #include <linux/blkdev.h> #include <linux/namei.h> +#include <linux/mount.h> #include <linux/ctype.h> #include <linux/string.h> #include <linux/slab.h> diff --git a/drivers/md/dm-verity-target.c b/drivers/md/dm-verity-target.c index 5c5d30cb6ec5..5214ed2c7507 100644 --- a/drivers/md/dm-verity-target.c +++ b/drivers/md/dm-verity-target.c @@ -551,7 +551,7 @@ static void verity_submit_prefetch(struct dm_verity *v, struct dm_verity_io *io) * Bio map function. It allocates dm_verity_io structure and bio vector and * fills them. Then it issues prefetches and the I/O. */ -static int verity_map(struct dm_target *ti, struct bio *bio) +int verity_map(struct dm_target *ti, struct bio *bio) { struct dm_verity *v = ti->private; struct dm_verity_io *io; @@ -596,7 +596,7 @@ static int verity_map(struct dm_target *ti, struct bio *bio) /* * Status: V (valid) or C (corruption found) */ -static void verity_status(struct dm_target *ti, status_type_t type, +void verity_status(struct dm_target *ti, status_type_t type, unsigned status_flags, char *result, unsigned maxlen) { struct dm_verity *v = ti->private; @@ -656,7 +656,7 @@ static void verity_status(struct dm_target *ti, status_type_t type, } } -static int verity_prepare_ioctl(struct dm_target *ti, +int verity_prepare_ioctl(struct dm_target *ti, struct block_device **bdev, fmode_t *mode) { struct dm_verity *v = ti->private; @@ -669,7 +669,7 @@ static int verity_prepare_ioctl(struct dm_target *ti, return 0; } -static int verity_iterate_devices(struct dm_target *ti, +int verity_iterate_devices(struct dm_target *ti, iterate_devices_callout_fn fn, void *data) { struct dm_verity *v = ti->private; @@ -677,7 +677,7 @@ static int verity_iterate_devices(struct dm_target *ti, return fn(ti, v->data_dev, v->data_start, ti->len, data); } -static void verity_io_hints(struct dm_target *ti, struct queue_limits *limits) +void verity_io_hints(struct dm_target *ti, struct queue_limits *limits) { struct dm_verity *v = ti->private; @@ -690,7 +690,7 @@ static void verity_io_hints(struct dm_target *ti, struct queue_limits *limits) blk_limits_io_min(limits, limits->logical_block_size); } -static void verity_dtr(struct dm_target *ti) +void verity_dtr(struct dm_target *ti) { struct dm_verity *v = ti->private; @@ -817,7 +817,7 @@ static int verity_parse_opt_args(struct dm_arg_set *as, struct dm_verity *v) * <digest> * <salt> Hex string or "-" if no salt. */ -static int verity_ctr(struct dm_target *ti, unsigned argc, char **argv) +int verity_ctr(struct dm_target *ti, unsigned argc, char **argv) { struct dm_verity *v; struct dm_arg_set as; diff --git a/drivers/md/dm-verity.h b/drivers/md/dm-verity.h index fb419f422d73..75effca400a3 100644 --- a/drivers/md/dm-verity.h +++ b/drivers/md/dm-verity.h @@ -126,4 +126,14 @@ extern int verity_hash(struct dm_verity *v, struct shash_desc *desc, extern int verity_hash_for_block(struct dm_verity *v, struct dm_verity_io *io, sector_t block, u8 *digest, bool *is_zero); +extern void verity_status(struct dm_target *ti, status_type_t type, + unsigned status_flags, char *result, unsigned maxlen); +extern int verity_prepare_ioctl(struct dm_target *ti, + struct block_device **bdev, fmode_t *mode); +extern int verity_iterate_devices(struct dm_target *ti, + iterate_devices_callout_fn fn, void *data); +extern void verity_io_hints(struct dm_target *ti, struct queue_limits *limits); +extern void verity_dtr(struct dm_target *ti); +extern int verity_ctr(struct dm_target *ti, unsigned argc, char **argv); +extern int verity_map(struct dm_target *ti, struct bio *bio); #endif /* DM_VERITY_H */ diff --git a/drivers/misc/Kconfig b/drivers/misc/Kconfig index ca7463544c72..06eddc0cb24f 100644 --- a/drivers/misc/Kconfig +++ b/drivers/misc/Kconfig @@ -531,6 +531,12 @@ config UID_CPUTIME help Per UID based cpu time statistics exported to /proc/uid_cputime +config MEMORY_STATE_TIME + tristate "Memory freq/bandwidth time statistics" + depends on PROFILING + help + Memory time statistics exported to /sys/kernel/memory_state_time + source "drivers/misc/c2port/Kconfig" source "drivers/misc/eeprom/Kconfig" source "drivers/misc/cb710/Kconfig" diff --git a/drivers/misc/Makefile b/drivers/misc/Makefile index e5142b836aee..b76b4c9fe104 100644 --- a/drivers/misc/Makefile +++ b/drivers/misc/Makefile @@ -57,3 +57,4 @@ obj-$(CONFIG_ECHO) += echo/ obj-$(CONFIG_VEXPRESS_SYSCFG) += vexpress-syscfg.o obj-$(CONFIG_CXL_BASE) += cxl/ obj-$(CONFIG_UID_CPUTIME) += uid_cputime.o +obj-$(CONFIG_MEMORY_STATE_TIME) += memory_state_time.o diff --git a/drivers/misc/memory_state_time.c b/drivers/misc/memory_state_time.c new file mode 100644 index 000000000000..34c797a06a31 --- /dev/null +++ b/drivers/misc/memory_state_time.c @@ -0,0 +1,454 @@ +/* drivers/misc/memory_state_time.c + * + * Copyright (C) 2016 Google, Inc. + * + * This software is licensed under the terms of the GNU General Public + * License version 2, as published by the Free Software Foundation, and + * may be copied, distributed, and modified under those terms. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + */ + +#include <linux/device.h> +#include <linux/err.h> +#include <linux/errno.h> +#include <linux/hashtable.h> +#include <linux/kconfig.h> +#include <linux/kernel.h> +#include <linux/kobject.h> +#include <linux/memory-state-time.h> +#include <linux/module.h> +#include <linux/mutex.h> +#include <linux/of_platform.h> +#include <linux/slab.h> +#include <linux/sysfs.h> +#include <linux/time.h> +#include <linux/timekeeping.h> +#include <linux/workqueue.h> + +#define KERNEL_ATTR_RO(_name) \ +static struct kobj_attribute _name##_attr = __ATTR_RO(_name) + +#define KERNEL_ATTR_RW(_name) \ +static struct kobj_attribute _name##_attr = \ + __ATTR(_name, 0644, _name##_show, _name##_store) + +#define FREQ_HASH_BITS 4 +DECLARE_HASHTABLE(freq_hash_table, FREQ_HASH_BITS); + +static DEFINE_MUTEX(mem_lock); + +#define TAG "memory_state_time" +#define BW_NODE "/soc/memory-state-time" +#define FREQ_TBL "freq-tbl" +#define BW_TBL "bw-buckets" +#define NUM_SOURCES "num-sources" + +#define LOWEST_FREQ 2 + +static int curr_bw; +static int curr_freq; +static u32 *bw_buckets; +static u32 *freq_buckets; +static int num_freqs; +static int num_buckets; +static int registered_bw_sources; +static u64 last_update; +static bool init_success; +static struct workqueue_struct *memory_wq; +static u32 num_sources = 10; +static int *bandwidths; + +struct freq_entry { + int freq; + u64 *buckets; /* Bandwidth buckets. */ + struct hlist_node hash; +}; + +struct queue_container { + struct work_struct update_state; + int value; + u64 time_now; + int id; + struct mutex *lock; +}; + +static int find_bucket(int bw) +{ + int i; + + if (bw_buckets != NULL) { + for (i = 0; i < num_buckets; i++) { + if (bw_buckets[i] > bw) { + pr_debug("Found bucket %d for bandwidth %d\n", + i, bw); + return i; + } + } + return num_buckets - 1; + } + return 0; +} + +static u64 get_time_diff(u64 time_now) +{ + u64 ms; + + ms = time_now - last_update; + last_update = time_now; + return ms; +} + +static ssize_t show_stat_show(struct kobject *kobj, + struct kobj_attribute *attr, char *buf) +{ + int i, j; + int len = 0; + struct freq_entry *freq_entry; + + for (i = 0; i < num_freqs; i++) { + hash_for_each_possible(freq_hash_table, freq_entry, hash, + freq_buckets[i]) { + if (freq_entry->freq == freq_buckets[i]) { + len += scnprintf(buf + len, PAGE_SIZE - len, + "%d ", freq_buckets[i]); + if (len >= PAGE_SIZE) + break; + for (j = 0; j < num_buckets; j++) { + len += scnprintf(buf + len, + PAGE_SIZE - len, + "%llu ", + freq_entry->buckets[j]); + } + len += scnprintf(buf + len, PAGE_SIZE - len, + "\n"); + } + } + } + pr_debug("Current Time: %llu\n", ktime_get_boot_ns()); + return len; +} +KERNEL_ATTR_RO(show_stat); + +static void update_table(u64 time_now) +{ + struct freq_entry *freq_entry; + + pr_debug("Last known bw %d freq %d\n", curr_bw, curr_freq); + hash_for_each_possible(freq_hash_table, freq_entry, hash, curr_freq) { + if (curr_freq == freq_entry->freq) { + freq_entry->buckets[find_bucket(curr_bw)] + += get_time_diff(time_now); + break; + } + } +} + +static bool freq_exists(int freq) +{ + int i; + + for (i = 0; i < num_freqs; i++) { + if (freq == freq_buckets[i]) + return true; + } + return false; +} + +static int calculate_total_bw(int bw, int index) +{ + int i; + int total_bw = 0; + + pr_debug("memory_state_time New bw %d for id %d\n", bw, index); + bandwidths[index] = bw; + for (i = 0; i < registered_bw_sources; i++) + total_bw += bandwidths[i]; + return total_bw; +} + +static void freq_update_do_work(struct work_struct *work) +{ + struct queue_container *freq_state_update + = container_of(work, struct queue_container, + update_state); + if (freq_state_update) { + mutex_lock(&mem_lock); + update_table(freq_state_update->time_now); + curr_freq = freq_state_update->value; + mutex_unlock(&mem_lock); + kfree(freq_state_update); + } +} + +static void bw_update_do_work(struct work_struct *work) +{ + struct queue_container *bw_state_update + = container_of(work, struct queue_container, + update_state); + if (bw_state_update) { + mutex_lock(&mem_lock); + update_table(bw_state_update->time_now); + curr_bw = calculate_total_bw(bw_state_update->value, + bw_state_update->id); + mutex_unlock(&mem_lock); + kfree(bw_state_update); + } +} + +static void memory_state_freq_update(struct memory_state_update_block *ub, + int value) +{ + if (IS_ENABLED(CONFIG_MEMORY_STATE_TIME)) { + if (freq_exists(value) && init_success) { + struct queue_container *freq_container + = kmalloc(sizeof(struct queue_container), + GFP_KERNEL); + if (!freq_container) + return; + INIT_WORK(&freq_container->update_state, + freq_update_do_work); + freq_container->time_now = ktime_get_boot_ns(); + freq_container->value = value; + pr_debug("Scheduling freq update in work queue\n"); + queue_work(memory_wq, &freq_container->update_state); + } else { + pr_debug("Freq does not exist.\n"); + } + } +} + +static void memory_state_bw_update(struct memory_state_update_block *ub, + int value) +{ + if (IS_ENABLED(CONFIG_MEMORY_STATE_TIME)) { + if (init_success) { + struct queue_container *bw_container + = kmalloc(sizeof(struct queue_container), + GFP_KERNEL); + if (!bw_container) + return; + INIT_WORK(&bw_container->update_state, + bw_update_do_work); + bw_container->time_now = ktime_get_boot_ns(); + bw_container->value = value; + bw_container->id = ub->id; + pr_debug("Scheduling bandwidth update in work queue\n"); + queue_work(memory_wq, &bw_container->update_state); + } + } +} + +struct memory_state_update_block *memory_state_register_frequency_source(void) +{ + struct memory_state_update_block *block; + + if (IS_ENABLED(CONFIG_MEMORY_STATE_TIME)) { + pr_debug("Allocating frequency source\n"); + block = kmalloc(sizeof(struct memory_state_update_block), + GFP_KERNEL); + if (!block) + return NULL; + block->update_call = memory_state_freq_update; + return block; + } + pr_err("Config option disabled.\n"); + return NULL; +} +EXPORT_SYMBOL_GPL(memory_state_register_frequency_source); + +struct memory_state_update_block *memory_state_register_bandwidth_source(void) +{ + struct memory_state_update_block *block; + + if (IS_ENABLED(CONFIG_MEMORY_STATE_TIME)) { + pr_debug("Allocating bandwidth source %d\n", + registered_bw_sources); + block = kmalloc(sizeof(struct memory_state_update_block), + GFP_KERNEL); + if (!block) + return NULL; + block->update_call = memory_state_bw_update; + if (registered_bw_sources < num_sources) { + block->id = registered_bw_sources++; + } else { + pr_err("Unable to allocate source; max number reached\n"); + kfree(block); + return NULL; + } + return block; + } + pr_err("Config option disabled.\n"); + return NULL; +} +EXPORT_SYMBOL_GPL(memory_state_register_bandwidth_source); + +/* Buckets are designated by their maximum. + * Returns the buckets decided by the capability of the device. + */ +static int get_bw_buckets(struct device *dev) +{ + int ret, lenb; + struct device_node *node = dev->of_node; + + of_property_read_u32(node, NUM_SOURCES, &num_sources); + if (of_find_property(node, BW_TBL, &lenb)) { + bandwidths = devm_kzalloc(dev, + sizeof(*bandwidths) * num_sources, GFP_KERNEL); + if (!bandwidths) + return -ENOMEM; + lenb /= sizeof(*bw_buckets); + bw_buckets = devm_kzalloc(dev, lenb * sizeof(*bw_buckets), + GFP_KERNEL); + if (!bw_buckets) { + devm_kfree(dev, bandwidths); + return -ENOMEM; + } + ret = of_property_read_u32_array(node, BW_TBL, bw_buckets, + lenb); + if (ret < 0) { + devm_kfree(dev, bandwidths); + devm_kfree(dev, bw_buckets); + pr_err("Unable to read bandwidth table from device tree.\n"); + return ret; + } + } + curr_bw = 0; + num_buckets = lenb; + return 0; +} + +/* Adds struct freq_entry nodes to the hashtable for each compatible frequency. + * Returns the supported number of frequencies. + */ +static int freq_buckets_init(struct device *dev) +{ + struct freq_entry *freq_entry; + int i; + int ret, lenf; + struct device_node *node = dev->of_node; + + if (of_find_property(node, FREQ_TBL, &lenf)) { + lenf /= sizeof(*freq_buckets); + freq_buckets = devm_kzalloc(dev, lenf * sizeof(*freq_buckets), + GFP_KERNEL); + if (!freq_buckets) + return -ENOMEM; + pr_debug("freqs found len %d\n", lenf); + ret = of_property_read_u32_array(node, FREQ_TBL, freq_buckets, + lenf); + if (ret < 0) { + devm_kfree(dev, freq_buckets); + pr_err("Unable to read frequency table from device tree.\n"); + return ret; + } + pr_debug("ret freq %d\n", ret); + } + num_freqs = lenf; + curr_freq = freq_buckets[LOWEST_FREQ]; + + for (i = 0; i < num_freqs; i++) { + freq_entry = devm_kzalloc(dev, sizeof(struct freq_entry), + GFP_KERNEL); + if (!freq_entry) + return -ENOMEM; + freq_entry->buckets = devm_kzalloc(dev, sizeof(u64)*num_buckets, + GFP_KERNEL); + if (!freq_entry->buckets) { + devm_kfree(dev, freq_entry); + return -ENOMEM; + } + pr_debug("memory_state_time Adding freq to ht %d\n", + freq_buckets[i]); + freq_entry->freq = freq_buckets[i]; + hash_add(freq_hash_table, &freq_entry->hash, freq_buckets[i]); + } + return 0; +} + +struct kobject *memory_kobj; +EXPORT_SYMBOL_GPL(memory_kobj); + +static struct attribute *memory_attrs[] = { + &show_stat_attr.attr, + NULL +}; + +static struct attribute_group memory_attr_group = { + .attrs = memory_attrs, +}; + +static int memory_state_time_probe(struct platform_device *pdev) +{ + int error; + + error = get_bw_buckets(&pdev->dev); + if (error) + return error; + error = freq_buckets_init(&pdev->dev); + if (error) + return error; + last_update = ktime_get_boot_ns(); + init_success = true; + + pr_debug("memory_state_time initialized with num_freqs %d\n", + num_freqs); + return 0; +} + +static const struct of_device_id match_table[] = { + { .compatible = "memory-state-time" }, + {} +}; + +static struct platform_driver memory_state_time_driver = { + .probe = memory_state_time_probe, + .driver = { + .name = "memory-state-time", + .of_match_table = match_table, + .owner = THIS_MODULE, + }, +}; + +static int __init memory_state_time_init(void) +{ + int error; + + hash_init(freq_hash_table); + memory_wq = create_singlethread_workqueue("memory_wq"); + if (!memory_wq) { + pr_err("Unable to create workqueue.\n"); + return -EINVAL; + } + /* + * Create sys/kernel directory for memory_state_time. + */ + memory_kobj = kobject_create_and_add(TAG, kernel_kobj); + if (!memory_kobj) { + pr_err("Unable to allocate memory_kobj for sysfs directory.\n"); + error = -ENOMEM; + goto wq; + } + error = sysfs_create_group(memory_kobj, &memory_attr_group); + if (error) { + pr_err("Unable to create sysfs folder.\n"); + goto kobj; + } + + error = platform_driver_register(&memory_state_time_driver); + if (error) { + pr_err("Unable to register memory_state_time platform driver.\n"); + goto group; + } + return 0; + +group: sysfs_remove_group(memory_kobj, &memory_attr_group); +kobj: kobject_put(memory_kobj); +wq: destroy_workqueue(memory_wq); + return error; +} +module_init(memory_state_time_init); diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c index 11309683d65f..27794b137b24 100644 --- a/fs/ecryptfs/file.c +++ b/fs/ecryptfs/file.c @@ -112,7 +112,6 @@ static int ecryptfs_readdir(struct file *file, struct dir_context *ctx) .sb = inode->i_sb, }; lower_file = ecryptfs_file_to_lower(file); - lower_file->f_pos = ctx->pos; rc = iterate_dir(lower_file, &buf.ctx); ctx->pos = buf.ctx.pos; if (rc < 0) @@ -236,14 +235,6 @@ static int ecryptfs_open(struct inode *inode, struct file *file) } ecryptfs_set_file_lower( file, ecryptfs_inode_to_private(inode)->lower_file); - if (d_is_dir(ecryptfs_dentry)) { - ecryptfs_printk(KERN_DEBUG, "This is a directory\n"); - mutex_lock(&crypt_stat->cs_mutex); - crypt_stat->flags &= ~(ECRYPTFS_ENCRYPTED); - mutex_unlock(&crypt_stat->cs_mutex); - rc = 0; - goto out; - } rc = read_or_initialize_metadata(ecryptfs_dentry); if (rc) goto out_put; @@ -260,6 +251,45 @@ out: return rc; } +/** + * ecryptfs_dir_open + * @inode: inode speciying file to open + * @file: Structure to return filled in + * + * Opens the file specified by inode. + * + * Returns zero on success; non-zero otherwise + */ +static int ecryptfs_dir_open(struct inode *inode, struct file *file) +{ + struct dentry *ecryptfs_dentry = file->f_path.dentry; + /* Private value of ecryptfs_dentry allocated in + * ecryptfs_lookup() */ + struct ecryptfs_file_info *file_info; + struct file *lower_file; + + /* Released in ecryptfs_release or end of function if failure */ + file_info = kmem_cache_zalloc(ecryptfs_file_info_cache, GFP_KERNEL); + ecryptfs_set_file_private(file, file_info); + if (unlikely(!file_info)) { + ecryptfs_printk(KERN_ERR, + "Error attempting to allocate memory\n"); + return -ENOMEM; + } + lower_file = dentry_open(ecryptfs_dentry_to_lower_path(ecryptfs_dentry), + file->f_flags, current_cred()); + if (IS_ERR(lower_file)) { + printk(KERN_ERR "%s: Error attempting to initialize " + "the lower file for the dentry with name " + "[%pd]; rc = [%ld]\n", __func__, + ecryptfs_dentry, PTR_ERR(lower_file)); + kmem_cache_free(ecryptfs_file_info_cache, file_info); + return PTR_ERR(lower_file); + } + ecryptfs_set_file_lower(file, lower_file); + return 0; +} + static int ecryptfs_flush(struct file *file, fl_owner_t td) { struct file *lower_file = ecryptfs_file_to_lower(file); @@ -280,6 +310,19 @@ static int ecryptfs_release(struct inode *inode, struct file *file) return 0; } +static int ecryptfs_dir_release(struct inode *inode, struct file *file) +{ + fput(ecryptfs_file_to_lower(file)); + kmem_cache_free(ecryptfs_file_info_cache, + ecryptfs_file_to_private(file)); + return 0; +} + +static loff_t ecryptfs_dir_llseek(struct file *file, loff_t offset, int whence) +{ + return vfs_llseek(ecryptfs_file_to_lower(file), offset, whence); +} + static int ecryptfs_fsync(struct file *file, loff_t start, loff_t end, int datasync) { @@ -359,20 +402,16 @@ const struct file_operations ecryptfs_dir_fops = { #ifdef CONFIG_COMPAT .compat_ioctl = ecryptfs_compat_ioctl, #endif - .open = ecryptfs_open, - .flush = ecryptfs_flush, - .release = ecryptfs_release, + .open = ecryptfs_dir_open, + .release = ecryptfs_dir_release, .fsync = ecryptfs_fsync, - .fasync = ecryptfs_fasync, - .splice_read = generic_file_splice_read, - .llseek = default_llseek, + .llseek = ecryptfs_dir_llseek, }; const struct file_operations ecryptfs_main_fops = { .llseek = generic_file_llseek, .read_iter = ecryptfs_read_update_atime, .write_iter = generic_file_write_iter, - .iterate = ecryptfs_readdir, .unlocked_ioctl = ecryptfs_unlocked_ioctl, #ifdef CONFIG_COMPAT .compat_ioctl = ecryptfs_compat_ioctl, diff --git a/fs/proc/base.c b/fs/proc/base.c index 308d4a321874..df715a095328 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -2261,15 +2261,17 @@ static ssize_t timerslack_ns_write(struct file *file, const char __user *buf, if (!p) return -ESRCH; - if (!capable(CAP_SYS_NICE)) { - count = -EPERM; - goto out; - } + if (p != current) { + if (!capable(CAP_SYS_NICE)) { + count = -EPERM; + goto out; + } - err = security_task_setscheduler(p); - if (err) { - count = err; - goto out; + err = security_task_setscheduler(p); + if (err) { + count = err; + goto out; + } } task_lock(p); @@ -2295,14 +2297,16 @@ static int timerslack_ns_show(struct seq_file *m, void *v) if (!p) return -ESRCH; - if (!capable(CAP_SYS_NICE)) { - err = -EPERM; - goto out; - } + if (p != current) { - err = security_task_getscheduler(p); - if (err) - goto out; + if (!capable(CAP_SYS_NICE)) { + err = -EPERM; + goto out; + } + err = security_task_getscheduler(p); + if (err) + goto out; + } task_lock(p); seq_printf(m, "%llu\n", p->timer_slack_ns); diff --git a/include/linux/device-mapper.h b/include/linux/device-mapper.h index 899ab9f8549e..b874d5b61ffc 100644 --- a/include/linux/device-mapper.h +++ b/include/linux/device-mapper.h @@ -383,6 +383,12 @@ void dm_set_mdptr(struct mapped_device *md, void *ptr); void *dm_get_mdptr(struct mapped_device *md); /* + * Export the device via the ioctl interface (uses mdptr). + */ +int dm_ioctl_export(struct mapped_device *md, const char *name, + const char *uuid); + +/* * A device can still be used while suspended, but I/O is deferred. */ int dm_suspend(struct mapped_device *md, unsigned suspend_flags); diff --git a/include/linux/memory-state-time.h b/include/linux/memory-state-time.h new file mode 100644 index 000000000000..d2212b027866 --- /dev/null +++ b/include/linux/memory-state-time.h @@ -0,0 +1,42 @@ +/* include/linux/memory-state-time.h + * + * Copyright (C) 2016 Google, Inc. + * + * This software is licensed under the terms of the GNU General Public + * License version 2, as published by the Free Software Foundation, and + * may be copied, distributed, and modified under those terms. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + */ + +#include <linux/workqueue.h> + +#define UPDATE_MEMORY_STATE(BLOCK, VALUE) BLOCK->update_call(BLOCK, VALUE) + +struct memory_state_update_block; + +typedef void (*memory_state_update_fn_t)(struct memory_state_update_block *ub, + int value); + +/* This struct is populated when you pass it to a memory_state_register* + * function. The update_call function is used for an update and defined in the + * typedef memory_state_update_fn_t + */ +struct memory_state_update_block { + memory_state_update_fn_t update_call; + int id; +}; + +/* Register a frequency struct memory_state_update_block to provide updates to + * memory_state_time about frequency changes using its update_call function. + */ +struct memory_state_update_block *memory_state_register_frequency_source(void); + +/* Register a bandwidth struct memory_state_update_block to provide updates to + * memory_state_time about bandwidth changes using its update_call function. + */ +struct memory_state_update_block *memory_state_register_bandwidth_source(void); diff --git a/include/net/fib_rules.h b/include/net/fib_rules.h index 55b5419cb6a7..bdd985f41022 100644 --- a/include/net/fib_rules.h +++ b/include/net/fib_rules.h @@ -89,7 +89,6 @@ struct fib_rules_ops { [FRA_FWMARK] = { .type = NLA_U32 }, \ [FRA_FWMASK] = { .type = NLA_U32 }, \ [FRA_TABLE] = { .type = NLA_U32 }, \ - [FRA_GOTO] = { .type = NLA_U32 }, \ [FRA_UID_START] = { .type = NLA_U32 }, \ [FRA_UID_END] = { .type = NLA_U32 }, \ [FRA_SUPPRESS_PREFIXLEN] = { .type = NLA_U32 }, \ diff --git a/include/uapi/linux/rtnetlink.h b/include/uapi/linux/rtnetlink.h index 355eea225dd9..3eb02a1d6d8c 100644 --- a/include/uapi/linux/rtnetlink.h +++ b/include/uapi/linux/rtnetlink.h @@ -306,12 +306,12 @@ enum rtattr_type_t { RTA_TABLE, RTA_MARK, RTA_MFC_STATS, + RTA_UID, RTA_VIA, RTA_NEWDST, RTA_PREF, RTA_ENCAP_TYPE, RTA_ENCAP, - RTA_UID, __RTA_MAX }; diff --git a/init/Makefile b/init/Makefile index 692b91f1c1d4..243f61de2cba 100644 --- a/init/Makefile +++ b/init/Makefile @@ -15,6 +15,7 @@ mounts-y := do_mounts.o mounts-$(CONFIG_BLK_DEV_RAM) += do_mounts_rd.o mounts-$(CONFIG_BLK_DEV_INITRD) += do_mounts_initrd.o mounts-$(CONFIG_BLK_DEV_MD) += do_mounts_md.o +mounts-$(CONFIG_BLK_DEV_DM) += do_mounts_dm.o # dependencies on generated files need to be listed explicitly $(obj)/version.o: include/generated/compile.h diff --git a/init/do_mounts.c b/init/do_mounts.c index dea5de95c2dd..1902a1c80831 100644 --- a/init/do_mounts.c +++ b/init/do_mounts.c @@ -566,6 +566,7 @@ void __init prepare_namespace(void) wait_for_device_probe(); md_run_setup(); + dm_run_setup(); if (saved_root_name[0]) { root_device_name = saved_root_name; diff --git a/init/do_mounts.h b/init/do_mounts.h index f5b978a9bb92..09d22862e8c3 100644 --- a/init/do_mounts.h +++ b/init/do_mounts.h @@ -74,3 +74,13 @@ void md_run_setup(void); static inline void md_run_setup(void) {} #endif + +#ifdef CONFIG_BLK_DEV_DM + +void dm_run_setup(void); + +#else + +static inline void dm_run_setup(void) {} + +#endif diff --git a/init/do_mounts_dm.c b/init/do_mounts_dm.c new file mode 100644 index 000000000000..ecda58df9a19 --- /dev/null +++ b/init/do_mounts_dm.c @@ -0,0 +1,426 @@ +/* do_mounts_dm.c + * Copyright (C) 2010 The Chromium OS Authors <chromium-os-dev@chromium.org> + * All Rights Reserved. + * Based on do_mounts_md.c + * + * This file is released under the GPL. + */ +#include <linux/device-mapper.h> +#include <linux/fs.h> +#include <linux/string.h> + +#include "do_mounts.h" +#include "../drivers/md/dm.h" + +#define DM_MAX_NAME 32 +#define DM_MAX_UUID 129 +#define DM_NO_UUID "none" + +#define DM_MSG_PREFIX "init" + +/* Separators used for parsing the dm= argument. */ +#define DM_FIELD_SEP ' ' +#define DM_LINE_SEP ',' + +/* + * When the device-mapper and any targets are compiled into the kernel + * (not a module), one target may be created and used as the root device at + * boot time with the parameters given with the boot line dm=... + * The code for that is here. + */ + +struct dm_setup_target { + sector_t begin; + sector_t length; + char *type; + char *params; + /* simple singly linked list */ + struct dm_setup_target *next; +}; + +static struct { + int minor; + int ro; + char name[DM_MAX_NAME]; + char uuid[DM_MAX_UUID]; + char *targets; + struct dm_setup_target *target; + int target_count; +} dm_setup_args __initdata; + +static __initdata int dm_early_setup; + +static size_t __init get_dm_option(char *str, char **next, char sep) +{ + size_t len = 0; + char *endp = NULL; + + if (!str) + return 0; + + endp = strchr(str, sep); + if (!endp) { /* act like strchrnul */ + len = strlen(str); + endp = str + len; + } else { + len = endp - str; + } + + if (endp == str) + return 0; + + if (!next) + return len; + + if (*endp == 0) { + /* Don't advance past the nul. */ + *next = endp; + } else { + *next = endp + 1; + } + return len; +} + +static int __init dm_setup_args_init(void) +{ + dm_setup_args.minor = 0; + dm_setup_args.ro = 0; + dm_setup_args.target = NULL; + dm_setup_args.target_count = 0; + return 0; +} + +static int __init dm_setup_cleanup(void) +{ + struct dm_setup_target *target = dm_setup_args.target; + struct dm_setup_target *old_target = NULL; + while (target) { + kfree(target->type); + kfree(target->params); + old_target = target; + target = target->next; + kfree(old_target); + dm_setup_args.target_count--; + } + BUG_ON(dm_setup_args.target_count); + return 0; +} + +static char * __init dm_setup_parse_device_args(char *str) +{ + char *next = NULL; + size_t len = 0; + + /* Grab the logical name of the device to be exported to udev */ + len = get_dm_option(str, &next, DM_FIELD_SEP); + if (!len) { + DMERR("failed to parse device name"); + goto parse_fail; + } + len = min(len + 1, sizeof(dm_setup_args.name)); + strlcpy(dm_setup_args.name, str, len); /* includes nul */ + str = skip_spaces(next); + + /* Grab the UUID value or "none" */ + len = get_dm_option(str, &next, DM_FIELD_SEP); + if (!len) { + DMERR("failed to parse device uuid"); + goto parse_fail; + } + len = min(len + 1, sizeof(dm_setup_args.uuid)); + strlcpy(dm_setup_args.uuid, str, len); + str = skip_spaces(next); + + /* Determine if the table/device will be read only or read-write */ + if (!strncmp("ro,", str, 3)) { + dm_setup_args.ro = 1; + } else if (!strncmp("rw,", str, 3)) { + dm_setup_args.ro = 0; + } else { + DMERR("failed to parse table mode"); + goto parse_fail; + } + str = skip_spaces(str + 3); + + return str; + +parse_fail: + return NULL; +} + +static void __init dm_substitute_devices(char *str, size_t str_len) +{ + char *candidate = str; + char *candidate_end = str; + char old_char; + size_t len = 0; + dev_t dev; + + if (str_len < 3) + return; + + while (str && *str) { + candidate = strchr(str, '/'); + if (!candidate) + break; + + /* Avoid embedded slashes */ + if (candidate != str && *(candidate - 1) != DM_FIELD_SEP) { + str = strchr(candidate, DM_FIELD_SEP); + continue; + } + + len = get_dm_option(candidate, &candidate_end, DM_FIELD_SEP); + str = skip_spaces(candidate_end); + if (len < 3 || len > 37) /* name_to_dev_t max; maj:mix min */ + continue; + + /* Temporarily terminate with a nul */ + if (*candidate_end) + candidate_end--; + old_char = *candidate_end; + *candidate_end = '\0'; + + DMDEBUG("converting candidate device '%s' to dev_t", candidate); + /* Use the boot-time specific device naming */ + dev = name_to_dev_t(candidate); + *candidate_end = old_char; + + DMDEBUG(" -> %u", dev); + /* No suitable replacement found */ + if (!dev) + continue; + + /* Rewrite the /dev/path as a major:minor */ + len = snprintf(candidate, len, "%u:%u", MAJOR(dev), MINOR(dev)); + if (!len) { + DMERR("error substituting device major/minor."); + break; + } + candidate += len; + /* Pad out with spaces (fixing our nul) */ + while (candidate < candidate_end) + *(candidate++) = DM_FIELD_SEP; + } +} + +static int __init dm_setup_parse_targets(char *str) +{ + char *next = NULL; + size_t len = 0; + struct dm_setup_target **target = NULL; + + /* Targets are defined as per the table format but with a + * comma as a newline separator. */ + target = &dm_setup_args.target; + while (str && *str) { + *target = kzalloc(sizeof(struct dm_setup_target), GFP_KERNEL); + if (!*target) { + DMERR("failed to allocate memory for target %d", + dm_setup_args.target_count); + goto parse_fail; + } + dm_setup_args.target_count++; + + (*target)->begin = simple_strtoull(str, &next, 10); + if (!next || *next != DM_FIELD_SEP) { + DMERR("failed to parse starting sector for target %d", + dm_setup_args.target_count - 1); + goto parse_fail; + } + str = skip_spaces(next + 1); + + (*target)->length = simple_strtoull(str, &next, 10); + if (!next || *next != DM_FIELD_SEP) { + DMERR("failed to parse length for target %d", + dm_setup_args.target_count - 1); + goto parse_fail; + } + str = skip_spaces(next + 1); + + len = get_dm_option(str, &next, DM_FIELD_SEP); + if (!len || + !((*target)->type = kstrndup(str, len, GFP_KERNEL))) { + DMERR("failed to parse type for target %d", + dm_setup_args.target_count - 1); + goto parse_fail; + } + str = skip_spaces(next); + + len = get_dm_option(str, &next, DM_LINE_SEP); + if (!len || + !((*target)->params = kstrndup(str, len, GFP_KERNEL))) { + DMERR("failed to parse params for target %d", + dm_setup_args.target_count - 1); + goto parse_fail; + } + str = skip_spaces(next); + + /* Before moving on, walk through the copied target and + * attempt to replace all /dev/xxx with the major:minor number. + * It may not be possible to resolve them traditionally at + * boot-time. */ + dm_substitute_devices((*target)->params, len); + + target = &((*target)->next); + } + DMDEBUG("parsed %d targets", dm_setup_args.target_count); + + return 0; + +parse_fail: + return 1; +} + +/* + * Parse the command-line parameters given our kernel, but do not + * actually try to invoke the DM device now; that is handled by + * dm_setup_drive after the low-level disk drivers have initialised. + * dm format is as follows: + * dm="name uuid fmode,[table line 1],[table line 2],..." + * May be used with root=/dev/dm-0 as it always uses the first dm minor. + */ + +static int __init dm_setup(char *str) +{ + dm_setup_args_init(); + + str = dm_setup_parse_device_args(str); + if (!str) { + DMDEBUG("str is NULL"); + goto parse_fail; + } + + /* Target parsing is delayed until we have dynamic memory */ + dm_setup_args.targets = str; + + printk(KERN_INFO "dm: will configure '%s' on dm-%d\n", + dm_setup_args.name, dm_setup_args.minor); + + dm_early_setup = 1; + return 1; + +parse_fail: + printk(KERN_WARNING "dm: Invalid arguments supplied to dm=.\n"); + return 0; +} + + +static void __init dm_setup_drive(void) +{ + struct mapped_device *md = NULL; + struct dm_table *table = NULL; + struct dm_setup_target *target; + char *uuid = dm_setup_args.uuid; + fmode_t fmode = FMODE_READ; + + /* Finish parsing the targets. */ + if (dm_setup_parse_targets(dm_setup_args.targets)) + goto parse_fail; + + if (dm_create(dm_setup_args.minor, &md)) { + DMDEBUG("failed to create the device"); + goto dm_create_fail; + } + DMDEBUG("created device '%s'", dm_device_name(md)); + + /* In addition to flagging the table below, the disk must be + * set explicitly ro/rw. */ + set_disk_ro(dm_disk(md), dm_setup_args.ro); + + if (!dm_setup_args.ro) + fmode |= FMODE_WRITE; + if (dm_table_create(&table, fmode, dm_setup_args.target_count, md)) { + DMDEBUG("failed to create the table"); + goto dm_table_create_fail; + } + + dm_lock_md_type(md); + target = dm_setup_args.target; + while (target) { + DMINFO("adding target '%llu %llu %s %s'", + (unsigned long long) target->begin, + (unsigned long long) target->length, target->type, + target->params); + if (dm_table_add_target(table, target->type, target->begin, + target->length, target->params)) { + DMDEBUG("failed to add the target to the table"); + goto add_target_fail; + } + target = target->next; + } + + if (dm_table_complete(table)) { + DMDEBUG("failed to complete the table"); + goto table_complete_fail; + } + + if (dm_get_md_type(md) == DM_TYPE_NONE) { + dm_set_md_type(md, dm_table_get_type(table)); + if (dm_setup_md_queue(md)) { + DMWARN("unable to set up device queue for new table."); + goto setup_md_queue_fail; + } + } else if (dm_get_md_type(md) != dm_table_get_type(table)) { + DMWARN("can't change device type after initial table load."); + goto setup_md_queue_fail; + } + + /* Suspend the device so that we can bind it to the table. */ + if (dm_suspend(md, 0)) { + DMDEBUG("failed to suspend the device pre-bind"); + goto suspend_fail; + } + + /* Bind the table to the device. This is the only way to associate + * md->map with the table and set the disk capacity directly. */ + if (dm_swap_table(md, table)) { /* should return NULL. */ + DMDEBUG("failed to bind the device to the table"); + goto table_bind_fail; + } + + /* Finally, resume and the device should be ready. */ + if (dm_resume(md)) { + DMDEBUG("failed to resume the device"); + goto resume_fail; + } + + /* Export the dm device via the ioctl interface */ + if (!strcmp(DM_NO_UUID, dm_setup_args.uuid)) + uuid = NULL; + if (dm_ioctl_export(md, dm_setup_args.name, uuid)) { + DMDEBUG("failed to export device with given name and uuid"); + goto export_fail; + } + printk(KERN_INFO "dm: dm-%d is ready\n", dm_setup_args.minor); + + dm_unlock_md_type(md); + dm_setup_cleanup(); + return; + +export_fail: +resume_fail: +table_bind_fail: +suspend_fail: +setup_md_queue_fail: +table_complete_fail: +add_target_fail: + dm_unlock_md_type(md); +dm_table_create_fail: + dm_put(md); +dm_create_fail: + dm_setup_cleanup(); +parse_fail: + printk(KERN_WARNING "dm: starting dm-%d (%s) failed\n", + dm_setup_args.minor, dm_setup_args.name); +} + +__setup("dm=", dm_setup); + +void __init dm_run_setup(void) +{ + if (!dm_early_setup) + return; + printk(KERN_INFO "dm: attempting early device configuration.\n"); + dm_setup_drive(); +} |
