diff options
Diffstat (limited to 'drivers/nvme')
52 files changed, 15439 insertions, 4803 deletions
diff --git a/drivers/nvme/Makefile b/drivers/nvme/Makefile index eedca8c72098..74f59ceed3d5 100644 --- a/drivers/nvme/Makefile +++ b/drivers/nvme/Makefile @@ -1,5 +1,5 @@ # SPDX-License-Identifier: GPL-2.0-only -obj-$(CONFIG_NVME_COMMON) += common/ +obj-y += common/ obj-y += host/ obj-y += target/ diff --git a/drivers/nvme/common/Kconfig b/drivers/nvme/common/Kconfig index 4514f44362dd..da963e4f3f1f 100644 --- a/drivers/nvme/common/Kconfig +++ b/drivers/nvme/common/Kconfig @@ -1,4 +1,15 @@ # SPDX-License-Identifier: GPL-2.0-only -config NVME_COMMON +config NVME_KEYRING tristate + select KEYS + +config NVME_AUTH + tristate + select CRYPTO + select CRYPTO_HMAC + select CRYPTO_SHA256 + select CRYPTO_SHA512 + select CRYPTO_DH + select CRYPTO_DH_RFC7919_GROUPS + select CRYPTO_HKDF diff --git a/drivers/nvme/common/Makefile b/drivers/nvme/common/Makefile index 720c625b8a52..681514cf2e2f 100644 --- a/drivers/nvme/common/Makefile +++ b/drivers/nvme/common/Makefile @@ -2,6 +2,8 @@ ccflags-y += -I$(src) -obj-$(CONFIG_NVME_COMMON) += nvme-common.o +obj-$(CONFIG_NVME_AUTH) += nvme-auth.o +obj-$(CONFIG_NVME_KEYRING) += nvme-keyring.o -nvme-common-y += auth.o +nvme-auth-y += auth.o +nvme-keyring-y += keyring.o diff --git a/drivers/nvme/common/auth.c b/drivers/nvme/common/auth.c index d90e4f0c08b7..e07e7d4bf8b6 100644 --- a/drivers/nvme/common/auth.c +++ b/drivers/nvme/common/auth.c @@ -8,12 +8,15 @@ #include <linux/base64.h> #include <linux/prandom.h> #include <linux/scatterlist.h> -#include <asm/unaligned.h> +#include <linux/unaligned.h> #include <crypto/hash.h> #include <crypto/dh.h> +#include <crypto/hkdf.h> #include <linux/nvme.h> #include <linux/nvme-auth.h> +#define HKDF_MAX_HASHLEN 64 + static u32 nvme_dhchap_seqnum; static DEFINE_MUTEX(nvme_dhchap_mutex); @@ -150,6 +153,14 @@ size_t nvme_auth_hmac_hash_len(u8 hmac_id) } EXPORT_SYMBOL_GPL(nvme_auth_hmac_hash_len); +u32 nvme_auth_key_struct_size(u32 key_len) +{ + struct nvme_dhchap_key key; + + return struct_size(&key, key, key_len); +} +EXPORT_SYMBOL_GPL(nvme_auth_key_struct_size); + struct nvme_dhchap_key *nvme_auth_extract_key(unsigned char *secret, u8 key_hash) { @@ -163,16 +174,11 @@ struct nvme_dhchap_key *nvme_auth_extract_key(unsigned char *secret, p = strrchr(secret, ':'); if (p) allocated_len = p - secret; - key = kzalloc(sizeof(*key), GFP_KERNEL); + key = nvme_auth_alloc_key(allocated_len, 0); if (!key) return ERR_PTR(-ENOMEM); - key->key = kzalloc(allocated_len, GFP_KERNEL); - if (!key->key) { - ret = -ENOMEM; - goto out_free_key; - } - key_len = base64_decode(secret, allocated_len, key->key); + key_len = base64_decode(secret, allocated_len, key->key, true, BASE64_STD); if (key_len < 0) { pr_debug("base64 key decoding error %d\n", key_len); @@ -187,14 +193,6 @@ struct nvme_dhchap_key *nvme_auth_extract_key(unsigned char *secret, goto out_free_secret; } - if (key_hash > 0 && - (key_len - 4) != nvme_auth_hmac_hash_len(key_hash)) { - pr_err("Mismatched key len %d for %s\n", key_len, - nvme_auth_hmac_name(key_hash)); - ret = -EINVAL; - goto out_free_secret; - } - /* The last four bytes is the CRC in little-endian format */ key_len -= 4; /* @@ -213,37 +211,51 @@ struct nvme_dhchap_key *nvme_auth_extract_key(unsigned char *secret, key->hash = key_hash; return key; out_free_secret: - kfree_sensitive(key->key); -out_free_key: - kfree(key); + nvme_auth_free_key(key); return ERR_PTR(ret); } EXPORT_SYMBOL_GPL(nvme_auth_extract_key); +struct nvme_dhchap_key *nvme_auth_alloc_key(u32 len, u8 hash) +{ + u32 num_bytes = nvme_auth_key_struct_size(len); + struct nvme_dhchap_key *key = kzalloc(num_bytes, GFP_KERNEL); + + if (key) { + key->len = len; + key->hash = hash; + } + return key; +} +EXPORT_SYMBOL_GPL(nvme_auth_alloc_key); + void nvme_auth_free_key(struct nvme_dhchap_key *key) { if (!key) return; - kfree_sensitive(key->key); - kfree(key); + kfree_sensitive(key); } EXPORT_SYMBOL_GPL(nvme_auth_free_key); -u8 *nvme_auth_transform_key(struct nvme_dhchap_key *key, char *nqn) +struct nvme_dhchap_key *nvme_auth_transform_key( + struct nvme_dhchap_key *key, char *nqn) { const char *hmac_name; struct crypto_shash *key_tfm; - struct shash_desc *shash; - u8 *transformed_key; - int ret; + SHASH_DESC_ON_STACK(shash, key_tfm); + struct nvme_dhchap_key *transformed_key; + int ret, key_len; - if (!key || !key->key) { + if (!key) { pr_warn("No key specified\n"); return ERR_PTR(-ENOKEY); } if (key->hash == 0) { - transformed_key = kmemdup(key->key, key->len, GFP_KERNEL); - return transformed_key ? transformed_key : ERR_PTR(-ENOMEM); + key_len = nvme_auth_key_struct_size(key->len); + transformed_key = kmemdup(key, key_len, GFP_KERNEL); + if (!transformed_key) + return ERR_PTR(-ENOMEM); + return transformed_key; } hmac_name = nvme_auth_hmac_name(key->hash); if (!hmac_name) { @@ -253,20 +265,13 @@ u8 *nvme_auth_transform_key(struct nvme_dhchap_key *key, char *nqn) key_tfm = crypto_alloc_shash(hmac_name, 0, 0); if (IS_ERR(key_tfm)) - return (u8 *)key_tfm; - - shash = kmalloc(sizeof(struct shash_desc) + - crypto_shash_descsize(key_tfm), - GFP_KERNEL); - if (!shash) { - ret = -ENOMEM; - goto out_free_key; - } + return ERR_CAST(key_tfm); - transformed_key = kzalloc(crypto_shash_digestsize(key_tfm), GFP_KERNEL); + key_len = crypto_shash_digestsize(key_tfm); + transformed_key = nvme_auth_alloc_key(key_len, key->hash); if (!transformed_key) { ret = -ENOMEM; - goto out_free_shash; + goto out_free_key; } shash->tfm = key_tfm; @@ -282,19 +287,16 @@ u8 *nvme_auth_transform_key(struct nvme_dhchap_key *key, char *nqn) ret = crypto_shash_update(shash, "NVMe-over-Fabrics", 17); if (ret < 0) goto out_free_transformed_key; - ret = crypto_shash_final(shash, transformed_key); + ret = crypto_shash_final(shash, transformed_key->key); if (ret < 0) goto out_free_transformed_key; - kfree(shash); crypto_free_shash(key_tfm); return transformed_key; out_free_transformed_key: - kfree_sensitive(transformed_key); -out_free_shash: - kfree(shash); + nvme_auth_free_key(transformed_key); out_free_key: crypto_free_shash(key_tfm); @@ -331,7 +333,6 @@ int nvme_auth_augmented_challenge(u8 hmac_id, u8 *skey, size_t skey_len, u8 *challenge, u8 *aug, size_t hlen) { struct crypto_shash *tfm; - struct shash_desc *desc; u8 *hashed_key; const char *hmac_name; int ret; @@ -359,29 +360,11 @@ int nvme_auth_augmented_challenge(u8 hmac_id, u8 *skey, size_t skey_len, goto out_free_key; } - desc = kmalloc(sizeof(struct shash_desc) + crypto_shash_descsize(tfm), - GFP_KERNEL); - if (!desc) { - ret = -ENOMEM; - goto out_free_hash; - } - desc->tfm = tfm; - ret = crypto_shash_setkey(tfm, hashed_key, hlen); if (ret) - goto out_free_desc; - - ret = crypto_shash_init(desc); - if (ret) - goto out_free_desc; - - ret = crypto_shash_update(desc, challenge, hlen); - if (ret) - goto out_free_desc; + goto out_free_hash; - ret = crypto_shash_final(desc, aug); -out_free_desc: - kfree_sensitive(desc); + ret = crypto_shash_tfm_digest(tfm, challenge, hlen, aug); out_free_hash: crypto_free_shash(tfm); out_free_key: @@ -480,4 +463,385 @@ int nvme_auth_generate_key(u8 *secret, struct nvme_dhchap_key **ret_key) } EXPORT_SYMBOL_GPL(nvme_auth_generate_key); +/** + * nvme_auth_generate_psk - Generate a PSK for TLS + * @hmac_id: Hash function identifier + * @skey: Session key + * @skey_len: Length of @skey + * @c1: Value of challenge C1 + * @c2: Value of challenge C2 + * @hash_len: Hash length of the hash algorithm + * @ret_psk: Pointer to the resulting generated PSK + * @ret_len: length of @ret_psk + * + * Generate a PSK for TLS as specified in NVMe base specification, section + * 8.13.5.9: Generated PSK for TLS + * + * The generated PSK for TLS shall be computed applying the HMAC function + * using the hash function H( ) selected by the HashID parameter in the + * DH-HMAC-CHAP_Challenge message with the session key KS as key to the + * concatenation of the two challenges C1 and C2 (i.e., generated + * PSK = HMAC(KS, C1 || C2)). + * + * Returns 0 on success with a valid generated PSK pointer in @ret_psk and + * the length of @ret_psk in @ret_len, or a negative error number otherwise. + */ +int nvme_auth_generate_psk(u8 hmac_id, u8 *skey, size_t skey_len, + u8 *c1, u8 *c2, size_t hash_len, u8 **ret_psk, size_t *ret_len) +{ + struct crypto_shash *tfm; + SHASH_DESC_ON_STACK(shash, tfm); + u8 *psk; + const char *hmac_name; + int ret, psk_len; + + if (!c1 || !c2) + return -EINVAL; + + hmac_name = nvme_auth_hmac_name(hmac_id); + if (!hmac_name) { + pr_warn("%s: invalid hash algorithm %d\n", + __func__, hmac_id); + return -EINVAL; + } + + tfm = crypto_alloc_shash(hmac_name, 0, 0); + if (IS_ERR(tfm)) + return PTR_ERR(tfm); + + psk_len = crypto_shash_digestsize(tfm); + psk = kzalloc(psk_len, GFP_KERNEL); + if (!psk) { + ret = -ENOMEM; + goto out_free_tfm; + } + + shash->tfm = tfm; + ret = crypto_shash_setkey(tfm, skey, skey_len); + if (ret) + goto out_free_psk; + + ret = crypto_shash_init(shash); + if (ret) + goto out_free_psk; + + ret = crypto_shash_update(shash, c1, hash_len); + if (ret) + goto out_free_psk; + + ret = crypto_shash_update(shash, c2, hash_len); + if (ret) + goto out_free_psk; + + ret = crypto_shash_final(shash, psk); + if (!ret) { + *ret_psk = psk; + *ret_len = psk_len; + } + +out_free_psk: + if (ret) + kfree_sensitive(psk); +out_free_tfm: + crypto_free_shash(tfm); + + return ret; +} +EXPORT_SYMBOL_GPL(nvme_auth_generate_psk); + +/** + * nvme_auth_generate_digest - Generate TLS PSK digest + * @hmac_id: Hash function identifier + * @psk: Generated input PSK + * @psk_len: Length of @psk + * @subsysnqn: NQN of the subsystem + * @hostnqn: NQN of the host + * @ret_digest: Pointer to the returned digest + * + * Generate a TLS PSK digest as specified in TP8018 Section 3.6.1.3: + * TLS PSK and PSK identity Derivation + * + * The PSK digest shall be computed by encoding in Base64 (refer to RFC + * 4648) the result of the application of the HMAC function using the hash + * function specified in item 4 above (ie the hash function of the cipher + * suite associated with the PSK identity) with the PSK as HMAC key to the + * concatenation of: + * - the NQN of the host (i.e., NQNh) not including the null terminator; + * - a space character; + * - the NQN of the NVM subsystem (i.e., NQNc) not including the null + * terminator; + * - a space character; and + * - the seventeen ASCII characters "NVMe-over-Fabrics" + * (i.e., <PSK digest> = Base64(HMAC(PSK, NQNh || " " || NQNc || " " || + * "NVMe-over-Fabrics"))). + * The length of the PSK digest depends on the hash function used to compute + * it as follows: + * - If the SHA-256 hash function is used, the resulting PSK digest is 44 + * characters long; or + * - If the SHA-384 hash function is used, the resulting PSK digest is 64 + * characters long. + * + * Returns 0 on success with a valid digest pointer in @ret_digest, or a + * negative error number on failure. + */ +int nvme_auth_generate_digest(u8 hmac_id, u8 *psk, size_t psk_len, + char *subsysnqn, char *hostnqn, u8 **ret_digest) +{ + struct crypto_shash *tfm; + SHASH_DESC_ON_STACK(shash, tfm); + u8 *digest, *enc; + const char *hmac_name; + size_t digest_len, hmac_len; + int ret; + + if (WARN_ON(!subsysnqn || !hostnqn)) + return -EINVAL; + + hmac_name = nvme_auth_hmac_name(hmac_id); + if (!hmac_name) { + pr_warn("%s: invalid hash algorithm %d\n", + __func__, hmac_id); + return -EINVAL; + } + + switch (nvme_auth_hmac_hash_len(hmac_id)) { + case 32: + hmac_len = 44; + break; + case 48: + hmac_len = 64; + break; + default: + pr_warn("%s: invalid hash algorithm '%s'\n", + __func__, hmac_name); + return -EINVAL; + } + + enc = kzalloc(hmac_len + 1, GFP_KERNEL); + if (!enc) + return -ENOMEM; + + tfm = crypto_alloc_shash(hmac_name, 0, 0); + if (IS_ERR(tfm)) { + ret = PTR_ERR(tfm); + goto out_free_enc; + } + + digest_len = crypto_shash_digestsize(tfm); + digest = kzalloc(digest_len, GFP_KERNEL); + if (!digest) { + ret = -ENOMEM; + goto out_free_tfm; + } + + shash->tfm = tfm; + ret = crypto_shash_setkey(tfm, psk, psk_len); + if (ret) + goto out_free_digest; + + ret = crypto_shash_init(shash); + if (ret) + goto out_free_digest; + + ret = crypto_shash_update(shash, hostnqn, strlen(hostnqn)); + if (ret) + goto out_free_digest; + + ret = crypto_shash_update(shash, " ", 1); + if (ret) + goto out_free_digest; + + ret = crypto_shash_update(shash, subsysnqn, strlen(subsysnqn)); + if (ret) + goto out_free_digest; + + ret = crypto_shash_update(shash, " NVMe-over-Fabrics", 18); + if (ret) + goto out_free_digest; + + ret = crypto_shash_final(shash, digest); + if (ret) + goto out_free_digest; + + ret = base64_encode(digest, digest_len, enc, true, BASE64_STD); + if (ret < hmac_len) { + ret = -ENOKEY; + goto out_free_digest; + } + *ret_digest = enc; + ret = 0; + +out_free_digest: + kfree_sensitive(digest); +out_free_tfm: + crypto_free_shash(tfm); +out_free_enc: + if (ret) + kfree_sensitive(enc); + + return ret; +} +EXPORT_SYMBOL_GPL(nvme_auth_generate_digest); + +/** + * hkdf_expand_label - HKDF-Expand-Label (RFC 8846 section 7.1) + * @hmac_tfm: hash context keyed with pseudorandom key + * @label: ASCII label without "tls13 " prefix + * @labellen: length of @label + * @context: context bytes + * @contextlen: length of @context + * @okm: output keying material + * @okmlen: length of @okm + * + * Build the TLS 1.3 HkdfLabel structure and invoke hkdf_expand(). + * + * Returns 0 on success with output keying material stored in @okm, + * or a negative errno value otherwise. + */ +static int hkdf_expand_label(struct crypto_shash *hmac_tfm, + const u8 *label, unsigned int labellen, + const u8 *context, unsigned int contextlen, + u8 *okm, unsigned int okmlen) +{ + int err; + u8 *info; + unsigned int infolen; + const char *tls13_prefix = "tls13 "; + unsigned int prefixlen = strlen(tls13_prefix); + + if (WARN_ON(labellen > (255 - prefixlen))) + return -EINVAL; + if (WARN_ON(contextlen > 255)) + return -EINVAL; + + infolen = 2 + (1 + prefixlen + labellen) + (1 + contextlen); + info = kzalloc(infolen, GFP_KERNEL); + if (!info) + return -ENOMEM; + + /* HkdfLabel.Length */ + put_unaligned_be16(okmlen, info); + + /* HkdfLabel.Label */ + info[2] = prefixlen + labellen; + memcpy(info + 3, tls13_prefix, prefixlen); + memcpy(info + 3 + prefixlen, label, labellen); + + /* HkdfLabel.Context */ + info[3 + prefixlen + labellen] = contextlen; + memcpy(info + 4 + prefixlen + labellen, context, contextlen); + + err = hkdf_expand(hmac_tfm, info, infolen, okm, okmlen); + kfree_sensitive(info); + return err; +} + +/** + * nvme_auth_derive_tls_psk - Derive TLS PSK + * @hmac_id: Hash function identifier + * @psk: generated input PSK + * @psk_len: size of @psk + * @psk_digest: TLS PSK digest + * @ret_psk: Pointer to the resulting TLS PSK + * + * Derive a TLS PSK as specified in TP8018 Section 3.6.1.3: + * TLS PSK and PSK identity Derivation + * + * The TLS PSK shall be derived as follows from an input PSK + * (i.e., either a retained PSK or a generated PSK) and a PSK + * identity using the HKDF-Extract and HKDF-Expand-Label operations + * (refer to RFC 5869 and RFC 8446) where the hash function is the + * one specified by the hash specifier of the PSK identity: + * 1. PRK = HKDF-Extract(0, Input PSK); and + * 2. TLS PSK = HKDF-Expand-Label(PRK, "nvme-tls-psk", PskIdentityContext, L), + * where PskIdentityContext is the hash identifier indicated in + * the PSK identity concatenated to a space character and to the + * Base64 PSK digest (i.e., "<hash> <PSK digest>") and L is the + * output size in bytes of the hash function (i.e., 32 for SHA-256 + * and 48 for SHA-384). + * + * Returns 0 on success with a valid psk pointer in @ret_psk or a negative + * error number otherwise. + */ +int nvme_auth_derive_tls_psk(int hmac_id, u8 *psk, size_t psk_len, + u8 *psk_digest, u8 **ret_psk) +{ + struct crypto_shash *hmac_tfm; + const char *hmac_name; + const char *label = "nvme-tls-psk"; + static const char default_salt[HKDF_MAX_HASHLEN]; + size_t prk_len; + const char *ctx; + unsigned char *prk, *tls_key; + int ret; + + hmac_name = nvme_auth_hmac_name(hmac_id); + if (!hmac_name) { + pr_warn("%s: invalid hash algorithm %d\n", + __func__, hmac_id); + return -EINVAL; + } + if (hmac_id == NVME_AUTH_HASH_SHA512) { + pr_warn("%s: unsupported hash algorithm %s\n", + __func__, hmac_name); + return -EINVAL; + } + + hmac_tfm = crypto_alloc_shash(hmac_name, 0, 0); + if (IS_ERR(hmac_tfm)) + return PTR_ERR(hmac_tfm); + + prk_len = crypto_shash_digestsize(hmac_tfm); + prk = kzalloc(prk_len, GFP_KERNEL); + if (!prk) { + ret = -ENOMEM; + goto out_free_shash; + } + + if (WARN_ON(prk_len > HKDF_MAX_HASHLEN)) { + ret = -EINVAL; + goto out_free_prk; + } + ret = hkdf_extract(hmac_tfm, psk, psk_len, + default_salt, prk_len, prk); + if (ret) + goto out_free_prk; + + ret = crypto_shash_setkey(hmac_tfm, prk, prk_len); + if (ret) + goto out_free_prk; + + ctx = kasprintf(GFP_KERNEL, "%02d %s", hmac_id, psk_digest); + if (!ctx) { + ret = -ENOMEM; + goto out_free_prk; + } + + tls_key = kzalloc(psk_len, GFP_KERNEL); + if (!tls_key) { + ret = -ENOMEM; + goto out_free_ctx; + } + ret = hkdf_expand_label(hmac_tfm, + label, strlen(label), + ctx, strlen(ctx), + tls_key, psk_len); + if (ret) { + kfree(tls_key); + goto out_free_ctx; + } + *ret_psk = tls_key; + +out_free_ctx: + kfree(ctx); +out_free_prk: + kfree(prk); +out_free_shash: + crypto_free_shash(hmac_tfm); + + return ret; +} +EXPORT_SYMBOL_GPL(nvme_auth_derive_tls_psk); + +MODULE_DESCRIPTION("NVMe Authentication framework"); MODULE_LICENSE("GPL v2"); diff --git a/drivers/nvme/common/keyring.c b/drivers/nvme/common/keyring.c new file mode 100644 index 000000000000..32d16c53133b --- /dev/null +++ b/drivers/nvme/common/keyring.c @@ -0,0 +1,287 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (c) 2023 Hannes Reinecke, SUSE Labs + */ + +#include <linux/module.h> +#include <linux/seq_file.h> +#include <linux/key-type.h> +#include <keys/user-type.h> +#include <linux/nvme.h> +#include <linux/nvme-tcp.h> +#include <linux/nvme-keyring.h> + +static struct key *nvme_keyring; + +key_serial_t nvme_keyring_id(void) +{ + return nvme_keyring->serial; +} +EXPORT_SYMBOL_GPL(nvme_keyring_id); + +static bool nvme_tls_psk_revoked(struct key *psk) +{ + return test_bit(KEY_FLAG_REVOKED, &psk->flags) || + test_bit(KEY_FLAG_INVALIDATED, &psk->flags); +} + +struct key *nvme_tls_key_lookup(key_serial_t key_id) +{ + struct key *key = key_lookup(key_id); + + if (IS_ERR(key)) { + pr_err("key id %08x not found\n", key_id); + return key; + } + if (nvme_tls_psk_revoked(key)) { + pr_err("key id %08x revoked\n", key_id); + return ERR_PTR(-EKEYREVOKED); + } + return key; +} +EXPORT_SYMBOL_GPL(nvme_tls_key_lookup); + +static void nvme_tls_psk_describe(const struct key *key, struct seq_file *m) +{ + seq_puts(m, key->description); + seq_printf(m, ": %u", key->datalen); +} + +static bool nvme_tls_psk_match(const struct key *key, + const struct key_match_data *match_data) +{ + const char *match_id; + size_t match_len; + + if (!key->description) { + pr_debug("%s: no key description\n", __func__); + return false; + } + if (!match_data->raw_data) { + pr_debug("%s: no match data\n", __func__); + return false; + } + match_id = match_data->raw_data; + match_len = strlen(match_id); + pr_debug("%s: match '%s' '%s' len %zd\n", + __func__, match_id, key->description, match_len); + return !memcmp(key->description, match_id, match_len); +} + +static int nvme_tls_psk_match_preparse(struct key_match_data *match_data) +{ + match_data->lookup_type = KEYRING_SEARCH_LOOKUP_ITERATE; + match_data->cmp = nvme_tls_psk_match; + return 0; +} + +static struct key_type nvme_tls_psk_key_type = { + .name = "psk", + .flags = KEY_TYPE_NET_DOMAIN, + .preparse = user_preparse, + .free_preparse = user_free_preparse, + .match_preparse = nvme_tls_psk_match_preparse, + .instantiate = generic_key_instantiate, + .revoke = user_revoke, + .destroy = user_destroy, + .describe = nvme_tls_psk_describe, + .read = user_read, +}; + +static struct key *nvme_tls_psk_lookup(struct key *keyring, + const char *hostnqn, const char *subnqn, + u8 hmac, u8 psk_ver, bool generated) +{ + char *identity; + size_t identity_len = (NVMF_NQN_SIZE) * 2 + 11; + key_ref_t keyref; + key_serial_t keyring_id; + + identity = kzalloc(identity_len, GFP_KERNEL); + if (!identity) + return ERR_PTR(-ENOMEM); + + snprintf(identity, identity_len, "NVMe%u%c%02u %s %s", + psk_ver, generated ? 'G' : 'R', hmac, hostnqn, subnqn); + + if (!keyring) + keyring = nvme_keyring; + keyring_id = key_serial(keyring); + pr_debug("keyring %x lookup tls psk '%s'\n", + keyring_id, identity); + keyref = keyring_search(make_key_ref(keyring, true), + &nvme_tls_psk_key_type, + identity, false); + if (IS_ERR(keyref)) { + pr_debug("lookup tls psk '%s' failed, error %ld\n", + identity, PTR_ERR(keyref)); + kfree(identity); + return ERR_PTR(-ENOKEY); + } + kfree(identity); + + return key_ref_to_ptr(keyref); +} + +/** + * nvme_tls_psk_refresh - Refresh TLS PSK + * @keyring: Keyring holding the TLS PSK + * @hostnqn: Host NQN to use + * @subnqn: Subsystem NQN to use + * @hmac_id: Hash function identifier + * @data: TLS PSK key material + * @data_len: Length of @data + * @digest: TLS PSK digest + * + * Refresh a generated version 1 TLS PSK with the identity generated + * from @hmac_id, @hostnqn, @subnqn, and @digest in the keyring given + * by @keyring. + * + * Returns the updated key success or an error pointer otherwise. + */ +struct key *nvme_tls_psk_refresh(struct key *keyring, + const char *hostnqn, const char *subnqn, u8 hmac_id, + u8 *data, size_t data_len, const char *digest) +{ + key_perm_t keyperm = + KEY_POS_SEARCH | KEY_POS_VIEW | KEY_POS_READ | + KEY_POS_WRITE | KEY_POS_LINK | KEY_POS_SETATTR | + KEY_USR_SEARCH | KEY_USR_VIEW | KEY_USR_READ; + char *identity; + key_ref_t keyref; + key_serial_t keyring_id; + struct key *key; + + if (!hostnqn || !subnqn || !data || !data_len) + return ERR_PTR(-EINVAL); + + identity = kasprintf(GFP_KERNEL, "NVMe1G%02d %s %s %s", + hmac_id, hostnqn, subnqn, digest); + if (!identity) + return ERR_PTR(-ENOMEM); + + if (!keyring) + keyring = nvme_keyring; + keyring_id = key_serial(keyring); + pr_debug("keyring %x refresh tls psk '%s'\n", + keyring_id, identity); + keyref = key_create_or_update(make_key_ref(keyring, true), + "psk", identity, data, data_len, + keyperm, KEY_ALLOC_NOT_IN_QUOTA | + KEY_ALLOC_BUILT_IN | + KEY_ALLOC_BYPASS_RESTRICTION); + if (IS_ERR(keyref)) { + pr_debug("refresh tls psk '%s' failed, error %ld\n", + identity, PTR_ERR(keyref)); + kfree(identity); + return ERR_PTR(-ENOKEY); + } + kfree(identity); + /* + * Set the default timeout to 1 hour + * as suggested in TP8018. + */ + key = key_ref_to_ptr(keyref); + key_set_timeout(key, 3600); + return key; +} +EXPORT_SYMBOL_GPL(nvme_tls_psk_refresh); + +/* + * NVMe PSK priority list + * + * 'Retained' PSKs (ie 'generated == false') should be preferred to 'generated' + * PSKs, PSKs with hash (psk_ver 1) should be preferred to PSKs without hash + * (psk_ver 0), and SHA-384 should be preferred to SHA-256. + */ +static struct nvme_tls_psk_priority_list { + bool generated; + u8 psk_ver; + enum nvme_tcp_tls_cipher cipher; +} nvme_tls_psk_prio[] = { + { .generated = false, + .psk_ver = 1, + .cipher = NVME_TCP_TLS_CIPHER_SHA384, }, + { .generated = false, + .psk_ver = 1, + .cipher = NVME_TCP_TLS_CIPHER_SHA256, }, + { .generated = false, + .psk_ver = 0, + .cipher = NVME_TCP_TLS_CIPHER_SHA384, }, + { .generated = false, + .psk_ver = 0, + .cipher = NVME_TCP_TLS_CIPHER_SHA256, }, + { .generated = true, + .psk_ver = 1, + .cipher = NVME_TCP_TLS_CIPHER_SHA384, }, + { .generated = true, + .psk_ver = 1, + .cipher = NVME_TCP_TLS_CIPHER_SHA256, }, + { .generated = true, + .psk_ver = 0, + .cipher = NVME_TCP_TLS_CIPHER_SHA384, }, + { .generated = true, + .psk_ver = 0, + .cipher = NVME_TCP_TLS_CIPHER_SHA256, }, +}; + +/* + * nvme_tls_psk_default - Return the preferred PSK to use for TLS ClientHello + */ +key_serial_t nvme_tls_psk_default(struct key *keyring, + const char *hostnqn, const char *subnqn) +{ + struct key *tls_key; + key_serial_t tls_key_id; + int prio; + + for (prio = 0; prio < ARRAY_SIZE(nvme_tls_psk_prio); prio++) { + bool generated = nvme_tls_psk_prio[prio].generated; + u8 ver = nvme_tls_psk_prio[prio].psk_ver; + enum nvme_tcp_tls_cipher cipher = nvme_tls_psk_prio[prio].cipher; + + tls_key = nvme_tls_psk_lookup(keyring, hostnqn, subnqn, + cipher, ver, generated); + if (!IS_ERR(tls_key)) { + tls_key_id = tls_key->serial; + key_put(tls_key); + return tls_key_id; + } + } + return 0; +} +EXPORT_SYMBOL_GPL(nvme_tls_psk_default); + +static int __init nvme_keyring_init(void) +{ + int err; + + nvme_keyring = keyring_alloc(".nvme", + GLOBAL_ROOT_UID, GLOBAL_ROOT_GID, + current_cred(), + (KEY_POS_ALL & ~KEY_POS_SETATTR) | + (KEY_USR_ALL & ~KEY_USR_SETATTR), + KEY_ALLOC_NOT_IN_QUOTA, NULL, NULL); + if (IS_ERR(nvme_keyring)) + return PTR_ERR(nvme_keyring); + + err = register_key_type(&nvme_tls_psk_key_type); + if (err) { + key_put(nvme_keyring); + return err; + } + return 0; +} + +static void __exit nvme_keyring_exit(void) +{ + unregister_key_type(&nvme_tls_psk_key_type); + key_revoke(nvme_keyring); + key_put(nvme_keyring); +} + +MODULE_LICENSE("GPL v2"); +MODULE_AUTHOR("Hannes Reinecke <hare@suse.de>"); +MODULE_DESCRIPTION("NVMe Keyring implementation"); +module_init(nvme_keyring_init); +module_exit(nvme_keyring_exit); diff --git a/drivers/nvme/host/Kconfig b/drivers/nvme/host/Kconfig index 2f6a7f8c94e8..31974c7dd20c 100644 --- a/drivers/nvme/host/Kconfig +++ b/drivers/nvme/host/Kconfig @@ -1,7 +1,6 @@ # SPDX-License-Identifier: GPL-2.0-only config NVME_CORE tristate - select BLK_DEV_INTEGRITY_T10 if BLK_DEV_INTEGRITY config BLK_DEV_NVME tristate "NVM Express block device" @@ -19,10 +18,15 @@ config NVME_MULTIPATH bool "NVMe multipath support" depends on NVME_CORE help - This option enables support for multipath access to NVMe - subsystems. If this option is enabled only a single - /dev/nvmeXnY device will show up for each NVMe namespace, - even if it is accessible through multiple controllers. + This option controls support for multipath access to NVMe + subsystems. If this option is enabled support for NVMe multipath + access is included in the kernel. If this option is disabled support + for NVMe multipath access is excluded from the kernel. When this + option is disabled each controller/namespace receives its + own /dev/nvmeXnY device entry and NVMe multipath access is + not supported. + + If unsure, say Y. config NVME_VERBOSE_ERRORS bool "NVMe verbose error reporting" @@ -42,6 +46,7 @@ config NVME_HWMON config NVME_FABRICS select NVME_CORE + select NVME_KEYRING if NVME_TCP_TLS tristate config NVME_RDMA @@ -79,9 +84,9 @@ config NVME_TCP tristate "NVM Express over Fabrics TCP host driver" depends on INET depends on BLOCK + select CRC32 + select NET_CRC32C select NVME_FABRICS - select CRYPTO - select CRYPTO_CRC32C help This provides support for the NVMe over Fabrics protocol using the TCP transport. This allows you to use remote block devices @@ -92,18 +97,28 @@ config NVME_TCP If unsure, say N. -config NVME_AUTH - bool "NVM Express over Fabrics In-Band Authentication" +config NVME_TCP_TLS + bool "NVMe over Fabrics TCP TLS encryption support" + depends on NVME_TCP + select NET_HANDSHAKE + select KEYS + select TLS + help + Enables TLS encryption for NVMe TCP using the netlink handshake API. + + The TLS handshake daemon is available at + https://github.com/oracle/ktls-utils. + + If unsure, say N. + +config NVME_HOST_AUTH + bool "NVMe over Fabrics In-Band Authentication in host side" depends on NVME_CORE - select NVME_COMMON - select CRYPTO - select CRYPTO_HMAC - select CRYPTO_SHA256 - select CRYPTO_SHA512 - select CRYPTO_DH - select CRYPTO_DH_RFC7919_GROUPS + select NVME_AUTH + select NVME_KEYRING help - This provides support for NVMe over Fabrics In-Band Authentication. + This provides support for NVMe over Fabrics In-Band Authentication in + host side. If unsure, say N. diff --git a/drivers/nvme/host/Makefile b/drivers/nvme/host/Makefile index e27202d22c7d..6414ec968f99 100644 --- a/drivers/nvme/host/Makefile +++ b/drivers/nvme/host/Makefile @@ -10,14 +10,14 @@ obj-$(CONFIG_NVME_FC) += nvme-fc.o obj-$(CONFIG_NVME_TCP) += nvme-tcp.o obj-$(CONFIG_NVME_APPLE) += nvme-apple.o -nvme-core-y += core.o ioctl.o +nvme-core-y += core.o ioctl.o sysfs.o pr.o nvme-core-$(CONFIG_NVME_VERBOSE_ERRORS) += constants.o nvme-core-$(CONFIG_TRACING) += trace.o nvme-core-$(CONFIG_NVME_MULTIPATH) += multipath.o nvme-core-$(CONFIG_BLK_DEV_ZONED) += zns.o nvme-core-$(CONFIG_FAULT_INJECTION_DEBUG_FS) += fault_inject.o nvme-core-$(CONFIG_NVME_HWMON) += hwmon.o -nvme-core-$(CONFIG_NVME_AUTH) += auth.o +nvme-core-$(CONFIG_NVME_HOST_AUTH) += auth.o nvme-y += pci.o diff --git a/drivers/nvme/host/apple.c b/drivers/nvme/host/apple.c index b317ce6c4ec3..15b3d07f8ccd 100644 --- a/drivers/nvme/host/apple.c +++ b/drivers/nvme/host/apple.c @@ -35,7 +35,6 @@ #include "nvme.h" #define APPLE_ANS_BOOT_TIMEOUT USEC_PER_SEC -#define APPLE_ANS_MAX_QUEUE_DEPTH 64 #define APPLE_ANS_COPROC_CPU_CONTROL 0x44 #define APPLE_ANS_COPROC_CPU_CONTROL_RUN BIT(4) @@ -75,6 +74,8 @@ #define APPLE_NVME_AQ_DEPTH 2 #define APPLE_NVME_AQ_MQ_TAG_DEPTH (APPLE_NVME_AQ_DEPTH - 1) +#define APPLE_NVME_IOSQES 7 + /* * These can be higher, but we need to ensure that any command doesn't * require an sg allocation that needs more than a page of data. @@ -142,6 +143,7 @@ struct apple_nvme_queue { u32 __iomem *sq_db; u32 __iomem *cq_db; + u16 sq_tail; u16 cq_head; u8 cq_phase; @@ -166,11 +168,17 @@ struct apple_nvme_iod { struct scatterlist *sg; }; +struct apple_nvme_hw { + bool has_lsq_nvmmu; + u32 max_queue_depth; +}; + struct apple_nvme { struct device *dev; void __iomem *mmio_coproc; void __iomem *mmio_nvme; + const struct apple_nvme_hw *hw; struct device **pd_dev; struct device_link **pd_link; @@ -209,19 +217,21 @@ static inline struct apple_nvme *queue_to_apple_nvme(struct apple_nvme_queue *q) { if (q->is_adminq) return container_of(q, struct apple_nvme, adminq); - else - return container_of(q, struct apple_nvme, ioq); + + return container_of(q, struct apple_nvme, ioq); } static unsigned int apple_nvme_queue_depth(struct apple_nvme_queue *q) { - if (q->is_adminq) + struct apple_nvme *anv = queue_to_apple_nvme(q); + + if (q->is_adminq && anv->hw->has_lsq_nvmmu) return APPLE_NVME_AQ_DEPTH; - else - return APPLE_ANS_MAX_QUEUE_DEPTH; + + return anv->hw->max_queue_depth; } -static void apple_nvme_rtkit_crashed(void *cookie) +static void apple_nvme_rtkit_crashed(void *cookie, const void *crashlog, size_t crashlog_size) { struct apple_nvme *anv = cookie; @@ -280,7 +290,28 @@ static void apple_nvmmu_inval(struct apple_nvme_queue *q, unsigned int tag) "NVMMU TCB invalidation failed\n"); } -static void apple_nvme_submit_cmd(struct apple_nvme_queue *q, +static void apple_nvme_submit_cmd_t8015(struct apple_nvme_queue *q, + struct nvme_command *cmd) +{ + struct apple_nvme *anv = queue_to_apple_nvme(q); + + spin_lock_irq(&anv->lock); + + if (q->is_adminq) + memcpy(&q->sqes[q->sq_tail], cmd, sizeof(*cmd)); + else + memcpy((void *)q->sqes + (q->sq_tail << APPLE_NVME_IOSQES), + cmd, sizeof(*cmd)); + + if (++q->sq_tail == anv->hw->max_queue_depth) + q->sq_tail = 0; + + writel(q->sq_tail, q->sq_db); + spin_unlock_irq(&anv->lock); +} + + +static void apple_nvme_submit_cmd_t8103(struct apple_nvme_queue *q, struct nvme_command *cmd) { struct apple_nvme *anv = queue_to_apple_nvme(q); @@ -301,8 +332,8 @@ static void apple_nvme_submit_cmd(struct apple_nvme_queue *q, memcpy(&q->sqes[tag], cmd, sizeof(*cmd)); /* - * This lock here doesn't make much sense at a first glace but - * removing it will result in occasional missed completetion + * This lock here doesn't make much sense at a first glance but + * removing it will result in occasional missed completion * interrupts even though the commands still appear on the CQ. * It's unclear why this happens but our best guess is that * there is a bug in the firmware triggered when a new command @@ -525,7 +556,7 @@ static blk_status_t apple_nvme_map_data(struct apple_nvme *anv, if (!iod->sg) return BLK_STS_RESOURCE; sg_init_table(iod->sg, blk_rq_nr_phys_segments(req)); - iod->nents = blk_rq_map_sg(req->q, req, iod->sg); + iod->nents = blk_rq_map_sg(req, iod->sg); if (!iod->nents) goto out_free_sg; @@ -590,7 +621,8 @@ static inline void apple_nvme_handle_cqe(struct apple_nvme_queue *q, __u16 command_id = READ_ONCE(cqe->command_id); struct request *req; - apple_nvmmu_inval(q, command_id); + if (anv->hw->has_lsq_nvmmu) + apple_nvmmu_inval(q, command_id); req = nvme_find_rq(apple_nvme_queue_tagset(anv, q), command_id); if (unlikely(!req)) { @@ -599,7 +631,8 @@ static inline void apple_nvme_handle_cqe(struct apple_nvme_queue *q, } if (!nvme_try_complete_req(req, cqe->status, cqe->result) && - !blk_mq_add_to_batch(req, iob, nvme_req(req)->status, + !blk_mq_add_to_batch(req, iob, + nvme_req(req)->status != NVME_SC_SUCCESS, apple_nvme_complete_batch)) apple_nvme_complete_rq(req); } @@ -649,7 +682,7 @@ static bool apple_nvme_handle_cq(struct apple_nvme_queue *q, bool force) found = apple_nvme_poll_cq(q, &iob); - if (!rq_list_empty(iob.req_list)) + if (!rq_list_empty(&iob.req_list)) apple_nvme_complete_batch(&iob); return found; @@ -684,7 +717,7 @@ static int apple_nvme_create_cq(struct apple_nvme *anv) c.create_cq.opcode = nvme_admin_create_cq; c.create_cq.prp1 = cpu_to_le64(anv->ioq.cq_dma_addr); c.create_cq.cqid = cpu_to_le16(1); - c.create_cq.qsize = cpu_to_le16(APPLE_ANS_MAX_QUEUE_DEPTH - 1); + c.create_cq.qsize = cpu_to_le16(anv->hw->max_queue_depth - 1); c.create_cq.cq_flags = cpu_to_le16(NVME_QUEUE_PHYS_CONTIG | NVME_CQ_IRQ_ENABLED); c.create_cq.irq_vector = cpu_to_le16(0); @@ -712,7 +745,7 @@ static int apple_nvme_create_sq(struct apple_nvme *anv) c.create_sq.opcode = nvme_admin_create_sq; c.create_sq.prp1 = cpu_to_le64(anv->ioq.sq_dma_addr); c.create_sq.sqid = cpu_to_le16(1); - c.create_sq.qsize = cpu_to_le16(APPLE_ANS_MAX_QUEUE_DEPTH - 1); + c.create_sq.qsize = cpu_to_le16(anv->hw->max_queue_depth - 1); c.create_sq.sq_flags = cpu_to_le16(NVME_QUEUE_PHYS_CONTIG); c.create_sq.cqid = cpu_to_le16(1); @@ -764,7 +797,12 @@ static blk_status_t apple_nvme_queue_rq(struct blk_mq_hw_ctx *hctx, } nvme_start_request(req); - apple_nvme_submit_cmd(q, cmnd); + + if (anv->hw->has_lsq_nvmmu) + apple_nvme_submit_cmd_t8103(q, cmnd); + else + apple_nvme_submit_cmd_t8015(q, cmnd); + return BLK_STS_OK; out_free_cmd: @@ -797,6 +835,7 @@ static int apple_nvme_init_request(struct blk_mq_tag_set *set, static void apple_nvme_disable(struct apple_nvme *anv, bool shutdown) { + enum nvme_ctrl_state state = nvme_ctrl_state(&anv->ctrl); u32 csts = readl(anv->mmio_nvme + NVME_REG_CSTS); bool dead = false, freeze = false; unsigned long flags; @@ -808,8 +847,8 @@ static void apple_nvme_disable(struct apple_nvme *anv, bool shutdown) if (csts & NVME_CSTS_CFS) dead = true; - if (anv->ctrl.state == NVME_CTRL_LIVE || - anv->ctrl.state == NVME_CTRL_RESETTING) { + if (state == NVME_CTRL_LIVE || + state == NVME_CTRL_RESETTING) { freeze = true; nvme_start_freeze(&anv->ctrl); } @@ -881,7 +920,7 @@ static enum blk_eh_timer_return apple_nvme_timeout(struct request *req) unsigned long flags; u32 csts = readl(anv->mmio_nvme + NVME_REG_CSTS); - if (anv->ctrl.state != NVME_CTRL_LIVE) { + if (nvme_ctrl_state(&anv->ctrl) != NVME_CTRL_LIVE) { /* * From rdma.c: * If we are resetting, connecting or deleting we should @@ -968,11 +1007,13 @@ static const struct blk_mq_ops apple_nvme_mq_ops = { static void apple_nvme_init_queue(struct apple_nvme_queue *q) { unsigned int depth = apple_nvme_queue_depth(q); + struct apple_nvme *anv = queue_to_apple_nvme(q); q->cq_head = 0; q->cq_phase = 1; - memset(q->tcbs, 0, - APPLE_ANS_MAX_QUEUE_DEPTH * sizeof(struct apple_nvmmu_tcb)); + if (anv->hw->has_lsq_nvmmu) + memset(q->tcbs, 0, anv->hw->max_queue_depth + * sizeof(struct apple_nvmmu_tcb)); memset(q->cqes, 0, depth * sizeof(struct nvme_completion)); WRITE_ONCE(q->enabled, true); wmb(); /* ensure the first interrupt sees the initialization */ @@ -985,10 +1026,10 @@ static void apple_nvme_reset_work(struct work_struct *work) u32 boot_status, aqa; struct apple_nvme *anv = container_of(work, struct apple_nvme, ctrl.reset_work); + enum nvme_ctrl_state state = nvme_ctrl_state(&anv->ctrl); - if (anv->ctrl.state != NVME_CTRL_RESETTING) { - dev_warn(anv->dev, "ctrl state %d is not RESETTING\n", - anv->ctrl.state); + if (state != NVME_CTRL_RESETTING) { + dev_warn(anv->dev, "ctrl state %d is not RESETTING\n", state); ret = -ENODEV; goto out; } @@ -1010,25 +1051,37 @@ static void apple_nvme_reset_work(struct work_struct *work) ret = apple_rtkit_shutdown(anv->rtk); if (ret) goto out; + + writel(0, anv->mmio_coproc + APPLE_ANS_COPROC_CPU_CONTROL); } - writel(0, anv->mmio_coproc + APPLE_ANS_COPROC_CPU_CONTROL); + /* + * Only do the soft-reset if the CPU is not running, which means either we + * or the previous stage shut it down cleanly. + */ + if (!(readl(anv->mmio_coproc + APPLE_ANS_COPROC_CPU_CONTROL) & + APPLE_ANS_COPROC_CPU_CONTROL_RUN)) { - ret = reset_control_assert(anv->reset); - if (ret) - goto out; + ret = reset_control_assert(anv->reset); + if (ret) + goto out; - ret = apple_rtkit_reinit(anv->rtk); - if (ret) - goto out; + ret = apple_rtkit_reinit(anv->rtk); + if (ret) + goto out; - ret = reset_control_deassert(anv->reset); - if (ret) - goto out; + ret = reset_control_deassert(anv->reset); + if (ret) + goto out; + + writel(APPLE_ANS_COPROC_CPU_CONTROL_RUN, + anv->mmio_coproc + APPLE_ANS_COPROC_CPU_CONTROL); + + ret = apple_rtkit_boot(anv->rtk); + } else { + ret = apple_rtkit_wake(anv->rtk); + } - writel(APPLE_ANS_COPROC_CPU_CONTROL_RUN, - anv->mmio_coproc + APPLE_ANS_COPROC_CPU_CONTROL); - ret = apple_rtkit_boot(anv->rtk); if (ret) { dev_err(anv->dev, "ANS did not boot"); goto out; @@ -1055,49 +1108,55 @@ static void apple_nvme_reset_work(struct work_struct *work) dma_set_max_seg_size(anv->dev, 0xffffffff); - /* - * Enable NVMMU and linear submission queues. - * While we could keep those disabled and pretend this is slightly - * more common NVMe controller we'd still need some quirks (e.g. - * sq entries will be 128 bytes) and Apple might drop support for - * that mode in the future. - */ - writel(APPLE_ANS_LINEAR_SQ_EN, - anv->mmio_nvme + APPLE_ANS_LINEAR_SQ_CTRL); + if (anv->hw->has_lsq_nvmmu) { + /* + * Enable NVMMU and linear submission queues which is required + * since T6000. + */ + writel(APPLE_ANS_LINEAR_SQ_EN, + anv->mmio_nvme + APPLE_ANS_LINEAR_SQ_CTRL); - /* Allow as many pending command as possible for both queues */ - writel(APPLE_ANS_MAX_QUEUE_DEPTH | (APPLE_ANS_MAX_QUEUE_DEPTH << 16), - anv->mmio_nvme + APPLE_ANS_MAX_PEND_CMDS_CTRL); + /* Allow as many pending command as possible for both queues */ + writel(anv->hw->max_queue_depth + | (anv->hw->max_queue_depth << 16), anv->mmio_nvme + + APPLE_ANS_MAX_PEND_CMDS_CTRL); - /* Setup the NVMMU for the maximum admin and IO queue depth */ - writel(APPLE_ANS_MAX_QUEUE_DEPTH - 1, - anv->mmio_nvme + APPLE_NVMMU_NUM_TCBS); + /* Setup the NVMMU for the maximum admin and IO queue depth */ + writel(anv->hw->max_queue_depth - 1, + anv->mmio_nvme + APPLE_NVMMU_NUM_TCBS); - /* - * This is probably a chicken bit: without it all commands where any PRP - * is set to zero (including those that don't use that field) fail and - * the co-processor complains about "completed with err BAD_CMD-" or - * a "NULL_PRP_PTR_ERR" in the syslog - */ - writel(readl(anv->mmio_nvme + APPLE_ANS_UNKNOWN_CTRL) & - ~APPLE_ANS_PRP_NULL_CHECK, - anv->mmio_nvme + APPLE_ANS_UNKNOWN_CTRL); + /* + * This is probably a chicken bit: without it all commands + * where any PRP is set to zero (including those that don't use + * that field) fail and the co-processor complains about + * "completed with err BAD_CMD-" or a "NULL_PRP_PTR_ERR" in the + * syslog + */ + writel(readl(anv->mmio_nvme + APPLE_ANS_UNKNOWN_CTRL) & + ~APPLE_ANS_PRP_NULL_CHECK, + anv->mmio_nvme + APPLE_ANS_UNKNOWN_CTRL); + } /* Setup the admin queue */ - aqa = APPLE_NVME_AQ_DEPTH - 1; + if (anv->hw->has_lsq_nvmmu) + aqa = APPLE_NVME_AQ_DEPTH - 1; + else + aqa = anv->hw->max_queue_depth - 1; aqa |= aqa << 16; writel(aqa, anv->mmio_nvme + NVME_REG_AQA); writeq(anv->adminq.sq_dma_addr, anv->mmio_nvme + NVME_REG_ASQ); writeq(anv->adminq.cq_dma_addr, anv->mmio_nvme + NVME_REG_ACQ); - /* Setup NVMMU for both queues */ - writeq(anv->adminq.tcb_dma_addr, - anv->mmio_nvme + APPLE_NVMMU_ASQ_TCB_BASE); - writeq(anv->ioq.tcb_dma_addr, - anv->mmio_nvme + APPLE_NVMMU_IOSQ_TCB_BASE); + if (anv->hw->has_lsq_nvmmu) { + /* Setup NVMMU for both queues */ + writeq(anv->adminq.tcb_dma_addr, + anv->mmio_nvme + APPLE_NVMMU_ASQ_TCB_BASE); + writeq(anv->ioq.tcb_dma_addr, + anv->mmio_nvme + APPLE_NVMMU_IOSQ_TCB_BASE); + } anv->ctrl.sqsize = - APPLE_ANS_MAX_QUEUE_DEPTH - 1; /* 0's based queue depth */ + anv->hw->max_queue_depth - 1; /* 0's based queue depth */ anv->ctrl.cap = readq(anv->mmio_nvme + NVME_REG_CAP); dev_dbg(anv->dev, "Enabling controller now"); @@ -1224,6 +1283,7 @@ static const struct nvme_ctrl_ops nvme_ctrl_ops = { .reg_read64 = apple_nvme_reg_read64, .free_ctrl = apple_nvme_free_ctrl, .get_address = apple_nvme_get_address, + .get_virt_boundary = nvme_get_virt_boundary, }; static void apple_nvme_async_probe(void *data, async_cookie_t cookie) @@ -1250,7 +1310,6 @@ static int apple_nvme_alloc_tagsets(struct apple_nvme *anv) anv->admin_tagset.timeout = NVME_ADMIN_TIMEOUT; anv->admin_tagset.numa_node = NUMA_NO_NODE; anv->admin_tagset.cmd_size = sizeof(struct apple_nvme_iod); - anv->admin_tagset.flags = BLK_MQ_F_NO_SCHED; anv->admin_tagset.driver_data = &anv->adminq; ret = blk_mq_alloc_tag_set(&anv->admin_tagset); @@ -1269,12 +1328,12 @@ static int apple_nvme_alloc_tagsets(struct apple_nvme *anv) * both queues. The admin queue gets the first APPLE_NVME_AQ_DEPTH which * must be marked as reserved in the IO queue. */ - anv->tagset.reserved_tags = APPLE_NVME_AQ_DEPTH; - anv->tagset.queue_depth = APPLE_ANS_MAX_QUEUE_DEPTH - 1; + if (anv->hw->has_lsq_nvmmu) + anv->tagset.reserved_tags = APPLE_NVME_AQ_DEPTH; + anv->tagset.queue_depth = anv->hw->max_queue_depth - 1; anv->tagset.timeout = NVME_IO_TIMEOUT; anv->tagset.numa_node = NUMA_NO_NODE; anv->tagset.cmd_size = sizeof(struct apple_nvme_iod); - anv->tagset.flags = BLK_MQ_F_SHOULD_MERGE; anv->tagset.driver_data = &anv->ioq; ret = blk_mq_alloc_tag_set(&anv->tagset); @@ -1295,6 +1354,7 @@ static int apple_nvme_queue_alloc(struct apple_nvme *anv, struct apple_nvme_queue *q) { unsigned int depth = apple_nvme_queue_depth(q); + size_t iosq_size; q->cqes = dmam_alloc_coherent(anv->dev, depth * sizeof(struct nvme_completion), @@ -1302,22 +1362,28 @@ static int apple_nvme_queue_alloc(struct apple_nvme *anv, if (!q->cqes) return -ENOMEM; - q->sqes = dmam_alloc_coherent(anv->dev, - depth * sizeof(struct nvme_command), + if (anv->hw->has_lsq_nvmmu) + iosq_size = depth * sizeof(struct nvme_command); + else + iosq_size = depth << APPLE_NVME_IOSQES; + + q->sqes = dmam_alloc_coherent(anv->dev, iosq_size, &q->sq_dma_addr, GFP_KERNEL); if (!q->sqes) return -ENOMEM; - /* - * We need the maximum queue depth here because the NVMMU only has a - * single depth configuration shared between both queues. - */ - q->tcbs = dmam_alloc_coherent(anv->dev, - APPLE_ANS_MAX_QUEUE_DEPTH * - sizeof(struct apple_nvmmu_tcb), - &q->tcb_dma_addr, GFP_KERNEL); - if (!q->tcbs) - return -ENOMEM; + if (anv->hw->has_lsq_nvmmu) { + /* + * We need the maximum queue depth here because the NVMMU only + * has a single depth configuration shared between both queues. + */ + q->tcbs = dmam_alloc_coherent(anv->dev, + anv->hw->max_queue_depth * + sizeof(struct apple_nvmmu_tcb), + &q->tcb_dma_addr, GFP_KERNEL); + if (!q->tcbs) + return -ENOMEM; + } /* * initialize phase to make sure the allocated and empty memory @@ -1387,7 +1453,7 @@ static void devm_apple_nvme_mempool_destroy(void *data) mempool_destroy(data); } -static int apple_nvme_probe(struct platform_device *pdev) +static struct apple_nvme *apple_nvme_alloc(struct platform_device *pdev) { struct device *dev = &pdev->dev; struct apple_nvme *anv; @@ -1395,12 +1461,18 @@ static int apple_nvme_probe(struct platform_device *pdev) anv = devm_kzalloc(dev, sizeof(*anv), GFP_KERNEL); if (!anv) - return -ENOMEM; + return ERR_PTR(-ENOMEM); anv->dev = get_device(dev); anv->adminq.is_adminq = true; platform_set_drvdata(pdev, anv); + anv->hw = of_device_get_match_data(&pdev->dev); + if (!anv->hw) { + ret = -ENODEV; + goto put_dev; + } + ret = apple_nvme_attach_genpd(anv); if (ret < 0) { dev_err_probe(dev, ret, "Failed to attach power domains"); @@ -1432,10 +1504,17 @@ static int apple_nvme_probe(struct platform_device *pdev) goto put_dev; } - anv->adminq.sq_db = anv->mmio_nvme + APPLE_ANS_LINEAR_ASQ_DB; - anv->adminq.cq_db = anv->mmio_nvme + APPLE_ANS_ACQ_DB; - anv->ioq.sq_db = anv->mmio_nvme + APPLE_ANS_LINEAR_IOSQ_DB; - anv->ioq.cq_db = anv->mmio_nvme + APPLE_ANS_IOCQ_DB; + if (anv->hw->has_lsq_nvmmu) { + anv->adminq.sq_db = anv->mmio_nvme + APPLE_ANS_LINEAR_ASQ_DB; + anv->adminq.cq_db = anv->mmio_nvme + APPLE_ANS_ACQ_DB; + anv->ioq.sq_db = anv->mmio_nvme + APPLE_ANS_LINEAR_IOSQ_DB; + anv->ioq.cq_db = anv->mmio_nvme + APPLE_ANS_IOCQ_DB; + } else { + anv->adminq.sq_db = anv->mmio_nvme + NVME_REG_DBS; + anv->adminq.cq_db = anv->mmio_nvme + APPLE_ANS_ACQ_DB; + anv->ioq.sq_db = anv->mmio_nvme + NVME_REG_DBS + 8; + anv->ioq.cq_db = anv->mmio_nvme + APPLE_ANS_IOCQ_DB; + } anv->sart = devm_apple_sart_get(dev); if (IS_ERR(anv->sart)) { @@ -1515,10 +1594,31 @@ static int apple_nvme_probe(struct platform_device *pdev) goto put_dev; } - anv->ctrl.admin_q = blk_mq_init_queue(&anv->admin_tagset); + return anv; +put_dev: + apple_nvme_detach_genpd(anv); + put_device(anv->dev); + return ERR_PTR(ret); +} + +static int apple_nvme_probe(struct platform_device *pdev) +{ + struct apple_nvme *anv; + int ret; + + anv = apple_nvme_alloc(pdev); + if (IS_ERR(anv)) + return PTR_ERR(anv); + + ret = nvme_add_ctrl(&anv->ctrl); + if (ret) + goto out_put_ctrl; + + anv->ctrl.admin_q = blk_mq_alloc_queue(&anv->admin_tagset, NULL, NULL); if (IS_ERR(anv->ctrl.admin_q)) { ret = -ENOMEM; - goto put_dev; + anv->ctrl.admin_q = NULL; + goto out_uninit_ctrl; } nvme_reset_ctrl(&anv->ctrl); @@ -1526,12 +1626,15 @@ static int apple_nvme_probe(struct platform_device *pdev) return 0; -put_dev: - put_device(anv->dev); +out_uninit_ctrl: + nvme_uninit_ctrl(&anv->ctrl); +out_put_ctrl: + nvme_put_ctrl(&anv->ctrl); + apple_nvme_detach_genpd(anv); return ret; } -static int apple_nvme_remove(struct platform_device *pdev) +static void apple_nvme_remove(struct platform_device *pdev) { struct apple_nvme *anv = platform_get_drvdata(pdev); @@ -1542,12 +1645,13 @@ static int apple_nvme_remove(struct platform_device *pdev) apple_nvme_disable(anv, true); nvme_uninit_ctrl(&anv->ctrl); - if (apple_rtkit_is_running(anv->rtk)) + if (apple_rtkit_is_running(anv->rtk)) { apple_rtkit_shutdown(anv->rtk); - apple_nvme_detach_genpd(anv); + writel(0, anv->mmio_coproc + APPLE_ANS_COPROC_CPU_CONTROL); + } - return 0; + apple_nvme_detach_genpd(anv); } static void apple_nvme_shutdown(struct platform_device *pdev) @@ -1555,8 +1659,11 @@ static void apple_nvme_shutdown(struct platform_device *pdev) struct apple_nvme *anv = platform_get_drvdata(pdev); apple_nvme_disable(anv, true); - if (apple_rtkit_is_running(anv->rtk)) + if (apple_rtkit_is_running(anv->rtk)) { apple_rtkit_shutdown(anv->rtk); + + writel(0, anv->mmio_coproc + APPLE_ANS_COPROC_CPU_CONTROL); + } } static int apple_nvme_resume(struct device *dev) @@ -1573,10 +1680,11 @@ static int apple_nvme_suspend(struct device *dev) apple_nvme_disable(anv, true); - if (apple_rtkit_is_running(anv->rtk)) + if (apple_rtkit_is_running(anv->rtk)) { ret = apple_rtkit_shutdown(anv->rtk); - writel(0, anv->mmio_coproc + APPLE_ANS_COPROC_CPU_CONTROL); + writel(0, anv->mmio_coproc + APPLE_ANS_COPROC_CPU_CONTROL); + } return ret; } @@ -1584,8 +1692,19 @@ static int apple_nvme_suspend(struct device *dev) static DEFINE_SIMPLE_DEV_PM_OPS(apple_nvme_pm_ops, apple_nvme_suspend, apple_nvme_resume); +static const struct apple_nvme_hw apple_nvme_t8015_hw = { + .has_lsq_nvmmu = false, + .max_queue_depth = 16, +}; + +static const struct apple_nvme_hw apple_nvme_t8103_hw = { + .has_lsq_nvmmu = true, + .max_queue_depth = 64, +}; + static const struct of_device_id apple_nvme_of_match[] = { - { .compatible = "apple,nvme-ans2" }, + { .compatible = "apple,t8015-nvme-ans2", .data = &apple_nvme_t8015_hw }, + { .compatible = "apple,nvme-ans2", .data = &apple_nvme_t8103_hw }, {}, }; MODULE_DEVICE_TABLE(of, apple_nvme_of_match); @@ -1603,4 +1722,5 @@ static struct platform_driver apple_nvme_driver = { module_platform_driver(apple_nvme_driver); MODULE_AUTHOR("Sven Peter <sven@svenpeter.dev>"); +MODULE_DESCRIPTION("Apple ANS NVM Express device driver"); MODULE_LICENSE("GPL"); diff --git a/drivers/nvme/host/auth.c b/drivers/nvme/host/auth.c index 4424f53a8a0a..8f3ccb317e4d 100644 --- a/drivers/nvme/host/auth.c +++ b/drivers/nvme/host/auth.c @@ -6,12 +6,13 @@ #include <linux/crc32.h> #include <linux/base64.h> #include <linux/prandom.h> -#include <asm/unaligned.h> +#include <linux/unaligned.h> #include <crypto/hash.h> #include <crypto/dh.h> #include "nvme.h" #include "fabrics.h" #include <linux/nvme-auth.h> +#include <linux/nvme-keyring.h> #define CHAP_BUF_SIZE 4096 static struct kmem_cache *nvme_chap_buf_cache; @@ -23,32 +24,32 @@ struct nvme_dhchap_queue_context { struct nvme_ctrl *ctrl; struct crypto_shash *shash_tfm; struct crypto_kpp *dh_tfm; + struct nvme_dhchap_key *transformed_key; void *buf; int qid; int error; u32 s1; u32 s2; + bool bi_directional; + bool authenticated; u16 transaction; u8 status; + u8 dhgroup_id; u8 hash_id; + u8 sc_c; size_t hash_len; - u8 dhgroup_id; u8 c1[64]; u8 c2[64]; u8 response[64]; - u8 *host_response; u8 *ctrl_key; - int ctrl_key_len; u8 *host_key; - int host_key_len; u8 *sess_key; + int ctrl_key_len; + int host_key_len; int sess_key_len; }; -#define nvme_auth_flags_from_qid(qid) \ - (qid == 0) ? 0 : BLK_MQ_REQ_NOWAIT | BLK_MQ_REQ_RESERVED -#define nvme_auth_queue_from_qid(ctrl, qid) \ - (qid == 0) ? (ctrl)->fabrics_q : (ctrl)->connect_q +static struct workqueue_struct *nvme_auth_wq; static inline int ctrl_max_dhchaps(struct nvme_ctrl *ctrl) { @@ -60,10 +61,15 @@ static int nvme_auth_submit(struct nvme_ctrl *ctrl, int qid, void *data, size_t data_len, bool auth_send) { struct nvme_command cmd = {}; - blk_mq_req_flags_t flags = nvme_auth_flags_from_qid(qid); - struct request_queue *q = nvme_auth_queue_from_qid(ctrl, qid); + nvme_submit_flags_t flags = NVME_SUBMIT_RETRY; + struct request_queue *q = ctrl->fabrics_q; int ret; + if (qid != 0) { + flags |= NVME_SUBMIT_NOWAIT | NVME_SUBMIT_RESERVED; + q = ctrl->connect_q; + } + cmd.auth_common.opcode = nvme_fabrics_command; cmd.auth_common.secp = NVME_AUTH_DHCHAP_PROTOCOL_IDENTIFIER; cmd.auth_common.spsp0 = 0x01; @@ -77,8 +83,7 @@ static int nvme_auth_submit(struct nvme_ctrl *ctrl, int qid, } ret = __nvme_submit_sync_cmd(q, &cmd, NULL, data, data_len, - qid == 0 ? NVME_QID_ANY : qid, - 0, flags); + qid == 0 ? NVME_QID_ANY : qid, flags); if (ret > 0) dev_warn(ctrl->device, "qid %d auth_send failed with status %d\n", qid, ret); @@ -129,7 +134,13 @@ static int nvme_auth_set_dhchap_negotiate_data(struct nvme_ctrl *ctrl, data->auth_type = NVME_AUTH_COMMON_MESSAGES; data->auth_id = NVME_AUTH_DHCHAP_MESSAGE_NEGOTIATE; data->t_id = cpu_to_le16(chap->transaction); - data->sc_c = 0; /* No secure channel concatenation */ + if (ctrl->opts->concat && chap->qid == 0) { + if (ctrl->opts->tls_key) + data->sc_c = NVME_AUTH_SECP_REPLACETLSPSK; + else + data->sc_c = NVME_AUTH_SECP_NEWTLSPSK; + } else + data->sc_c = NVME_AUTH_SECP_NOSC; data->napd = 1; data->auth_protocol[0].dhchap.authid = NVME_AUTH_DHCHAP_AUTH_ID; data->auth_protocol[0].dhchap.halen = 3; @@ -144,6 +155,8 @@ static int nvme_auth_set_dhchap_negotiate_data(struct nvme_ctrl *ctrl, data->auth_protocol[0].dhchap.idlist[34] = NVME_AUTH_DHGROUP_6144; data->auth_protocol[0].dhchap.idlist[35] = NVME_AUTH_DHGROUP_8192; + chap->sc_c = data->sc_c; + return size; } @@ -158,7 +171,7 @@ static int nvme_auth_process_dhchap_challenge(struct nvme_ctrl *ctrl, if (size > CHAP_BUF_SIZE) { chap->status = NVME_AUTH_DHCHAP_FAILURE_INCORRECT_PAYLOAD; - return NVME_SC_INVALID_FIELD; + return -EINVAL; } hmac_name = nvme_auth_hmac_name(data->hashid); @@ -167,7 +180,7 @@ static int nvme_auth_process_dhchap_challenge(struct nvme_ctrl *ctrl, "qid %d: invalid HASH ID %d\n", chap->qid, data->hashid); chap->status = NVME_AUTH_DHCHAP_FAILURE_HASH_UNUSABLE; - return NVME_SC_INVALID_FIELD; + return -EPROTO; } if (chap->hash_id == data->hashid && chap->shash_tfm && @@ -193,7 +206,7 @@ static int nvme_auth_process_dhchap_challenge(struct nvme_ctrl *ctrl, chap->qid, hmac_name, PTR_ERR(chap->shash_tfm)); chap->shash_tfm = NULL; chap->status = NVME_AUTH_DHCHAP_FAILURE_FAILED; - return NVME_SC_AUTH_REQUIRED; + return -ENOMEM; } if (crypto_shash_digestsize(chap->shash_tfm) != data->hl) { @@ -203,7 +216,7 @@ static int nvme_auth_process_dhchap_challenge(struct nvme_ctrl *ctrl, crypto_free_shash(chap->shash_tfm); chap->shash_tfm = NULL; chap->status = NVME_AUTH_DHCHAP_FAILURE_HASH_UNUSABLE; - return NVME_SC_AUTH_REQUIRED; + return -EPROTO; } chap->hash_id = data->hashid; @@ -219,7 +232,7 @@ select_kpp: chap->qid, data->dhgid); chap->status = NVME_AUTH_DHCHAP_FAILURE_DHGROUP_UNUSABLE; /* Leave previous dh_tfm intact */ - return NVME_SC_AUTH_REQUIRED; + return -EPROTO; } if (chap->dhgroup_id == data->dhgid && @@ -242,7 +255,7 @@ select_kpp: "qid %d: empty DH value\n", chap->qid); chap->status = NVME_AUTH_DHCHAP_FAILURE_DHGROUP_UNUSABLE; - return NVME_SC_INVALID_FIELD; + return -EPROTO; } chap->dh_tfm = crypto_alloc_kpp(kpp_name, 0, 0); @@ -254,7 +267,7 @@ select_kpp: chap->qid, ret, gid_name); chap->status = NVME_AUTH_DHCHAP_FAILURE_DHGROUP_UNUSABLE; chap->dh_tfm = NULL; - return NVME_SC_AUTH_REQUIRED; + return ret; } dev_dbg(ctrl->device, "qid %d: selected DH group %s\n", chap->qid, gid_name); @@ -263,7 +276,7 @@ select_kpp: "qid %d: invalid DH value for NULL DH\n", chap->qid); chap->status = NVME_AUTH_DHCHAP_FAILURE_INCORRECT_PAYLOAD; - return NVME_SC_INVALID_FIELD; + return -EPROTO; } chap->dhgroup_id = data->dhgid; @@ -274,7 +287,7 @@ skip_kpp: chap->ctrl_key = kmalloc(dhvlen, GFP_KERNEL); if (!chap->ctrl_key) { chap->status = NVME_AUTH_DHCHAP_FAILURE_FAILED; - return NVME_SC_AUTH_REQUIRED; + return -ENOMEM; } chap->ctrl_key_len = dhvlen; memcpy(chap->ctrl_key, data->cval + chap->hash_len, @@ -309,18 +322,23 @@ static int nvme_auth_set_dhchap_reply_data(struct nvme_ctrl *ctrl, data->hl = chap->hash_len; data->dhvlen = cpu_to_le16(chap->host_key_len); memcpy(data->rval, chap->response, chap->hash_len); - if (ctrl->ctrl_key) { + if (ctrl->ctrl_key) + chap->bi_directional = true; + if (ctrl->ctrl_key || ctrl->opts->concat) { get_random_bytes(chap->c2, chap->hash_len); data->cvalid = 1; - chap->s2 = nvme_auth_get_seqnum(); memcpy(data->rval + chap->hash_len, chap->c2, chap->hash_len); dev_dbg(ctrl->device, "%s: qid %d ctrl challenge %*ph\n", __func__, chap->qid, (int)chap->hash_len, chap->c2); } else { memset(chap->c2, 0, chap->hash_len); - chap->s2 = 0; } + if (ctrl->opts->concat) { + chap->s2 = 0; + chap->bi_directional = false; + } else + chap->s2 = nvme_auth_get_seqnum(); data->seqnum = cpu_to_le32(chap->s2); if (chap->host_key_len) { dev_dbg(ctrl->device, "%s: qid %d host public key %*ph\n", @@ -337,14 +355,11 @@ static int nvme_auth_process_dhchap_success1(struct nvme_ctrl *ctrl, struct nvme_dhchap_queue_context *chap) { struct nvmf_auth_dhchap_success1_data *data = chap->buf; - size_t size = sizeof(*data); - - if (chap->ctrl_key) - size += chap->hash_len; + size_t size = sizeof(*data) + chap->hash_len; if (size > CHAP_BUF_SIZE) { chap->status = NVME_AUTH_DHCHAP_FAILURE_INCORRECT_PAYLOAD; - return NVME_SC_INVALID_FIELD; + return -EINVAL; } if (data->hl != chap->hash_len) { @@ -352,7 +367,7 @@ static int nvme_auth_process_dhchap_success1(struct nvme_ctrl *ctrl, "qid %d: invalid hash length %u\n", chap->qid, data->hl); chap->status = NVME_AUTH_DHCHAP_FAILURE_HASH_UNUSABLE; - return NVME_SC_INVALID_FIELD; + return -EPROTO; } /* Just print out information for the admin queue */ @@ -376,7 +391,7 @@ static int nvme_auth_process_dhchap_success1(struct nvme_ctrl *ctrl, "qid %d: controller authentication failed\n", chap->qid); chap->status = NVME_AUTH_DHCHAP_FAILURE_FAILED; - return NVME_SC_AUTH_REQUIRED; + return -ECONNREFUSED; } /* Just print out information for the admin queue */ @@ -426,12 +441,12 @@ static int nvme_auth_dhchap_setup_host_response(struct nvme_ctrl *ctrl, dev_dbg(ctrl->device, "%s: qid %d host response seq %u transaction %d\n", __func__, chap->qid, chap->s1, chap->transaction); - if (!chap->host_response) { - chap->host_response = nvme_auth_transform_key(ctrl->host_key, + if (!chap->transformed_key) { + chap->transformed_key = nvme_auth_transform_key(ctrl->host_key, ctrl->opts->host->nqn); - if (IS_ERR(chap->host_response)) { - ret = PTR_ERR(chap->host_response); - chap->host_response = NULL; + if (IS_ERR(chap->transformed_key)) { + ret = PTR_ERR(chap->transformed_key); + chap->transformed_key = NULL; return ret; } } else { @@ -440,7 +455,7 @@ static int nvme_auth_dhchap_setup_host_response(struct nvme_ctrl *ctrl, } ret = crypto_shash_setkey(chap->shash_tfm, - chap->host_response, ctrl->host_key->len); + chap->transformed_key->key, chap->transformed_key->len); if (ret) { dev_warn(ctrl->device, "qid %d: failed to set key, error %d\n", chap->qid, ret); @@ -477,7 +492,7 @@ static int nvme_auth_dhchap_setup_host_response(struct nvme_ctrl *ctrl, ret = crypto_shash_update(shash, buf, 2); if (ret) goto out; - memset(buf, 0, sizeof(buf)); + *buf = chap->sc_c; ret = crypto_shash_update(shash, buf, 1); if (ret) goto out; @@ -488,6 +503,7 @@ static int nvme_auth_dhchap_setup_host_response(struct nvme_ctrl *ctrl, strlen(ctrl->opts->host->nqn)); if (ret) goto out; + memset(buf, 0, sizeof(buf)); ret = crypto_shash_update(shash, buf, 1); if (ret) goto out; @@ -506,19 +522,19 @@ static int nvme_auth_dhchap_setup_ctrl_response(struct nvme_ctrl *ctrl, struct nvme_dhchap_queue_context *chap) { SHASH_DESC_ON_STACK(shash, chap->shash_tfm); - u8 *ctrl_response; + struct nvme_dhchap_key *transformed_key; u8 buf[4], *challenge = chap->c2; int ret; - ctrl_response = nvme_auth_transform_key(ctrl->ctrl_key, + transformed_key = nvme_auth_transform_key(ctrl->ctrl_key, ctrl->opts->subsysnqn); - if (IS_ERR(ctrl_response)) { - ret = PTR_ERR(ctrl_response); + if (IS_ERR(transformed_key)) { + ret = PTR_ERR(transformed_key); return ret; } ret = crypto_shash_setkey(chap->shash_tfm, - ctrl_response, ctrl->ctrl_key->len); + transformed_key->key, transformed_key->len); if (ret) { dev_warn(ctrl->device, "qid %d: failed to set key, error %d\n", chap->qid, ret); @@ -584,7 +600,7 @@ static int nvme_auth_dhchap_setup_ctrl_response(struct nvme_ctrl *ctrl, out: if (challenge != chap->c2) kfree(challenge); - kfree(ctrl_response); + nvme_auth_free_key(transformed_key); return ret; } @@ -646,8 +662,8 @@ gen_sesskey: static void nvme_auth_reset_dhchap(struct nvme_dhchap_queue_context *chap) { - kfree_sensitive(chap->host_response); - chap->host_response = NULL; + nvme_auth_free_key(chap->transformed_key); + chap->transformed_key = NULL; kfree_sensitive(chap->host_key); chap->host_key = NULL; chap->host_key_len = 0; @@ -661,6 +677,7 @@ static void nvme_auth_reset_dhchap(struct nvme_dhchap_queue_context *chap) chap->error = 0; chap->s1 = 0; chap->s2 = 0; + chap->bi_directional = false; chap->transaction = 0; memset(chap->c1, 0, sizeof(chap->c1)); memset(chap->c2, 0, sizeof(chap->c2)); @@ -671,12 +688,99 @@ static void nvme_auth_reset_dhchap(struct nvme_dhchap_queue_context *chap) static void nvme_auth_free_dhchap(struct nvme_dhchap_queue_context *chap) { nvme_auth_reset_dhchap(chap); + chap->authenticated = false; if (chap->shash_tfm) crypto_free_shash(chap->shash_tfm); if (chap->dh_tfm) crypto_free_kpp(chap->dh_tfm); } +void nvme_auth_revoke_tls_key(struct nvme_ctrl *ctrl) +{ + dev_dbg(ctrl->device, "Wipe generated TLS PSK %08x\n", + key_serial(ctrl->opts->tls_key)); + key_revoke(ctrl->opts->tls_key); + key_put(ctrl->opts->tls_key); + ctrl->opts->tls_key = NULL; +} +EXPORT_SYMBOL_GPL(nvme_auth_revoke_tls_key); + +static int nvme_auth_secure_concat(struct nvme_ctrl *ctrl, + struct nvme_dhchap_queue_context *chap) +{ + u8 *psk, *digest, *tls_psk; + struct key *tls_key; + size_t psk_len; + int ret = 0; + + if (!chap->sess_key) { + dev_warn(ctrl->device, + "%s: qid %d no session key negotiated\n", + __func__, chap->qid); + return -ENOKEY; + } + + if (chap->qid) { + dev_warn(ctrl->device, + "qid %d: secure concatenation not supported on I/O queues\n", + chap->qid); + return -EINVAL; + } + ret = nvme_auth_generate_psk(chap->hash_id, chap->sess_key, + chap->sess_key_len, + chap->c1, chap->c2, + chap->hash_len, &psk, &psk_len); + if (ret) { + dev_warn(ctrl->device, + "%s: qid %d failed to generate PSK, error %d\n", + __func__, chap->qid, ret); + return ret; + } + dev_dbg(ctrl->device, + "%s: generated psk %*ph\n", __func__, (int)psk_len, psk); + + ret = nvme_auth_generate_digest(chap->hash_id, psk, psk_len, + ctrl->opts->subsysnqn, + ctrl->opts->host->nqn, &digest); + if (ret) { + dev_warn(ctrl->device, + "%s: qid %d failed to generate digest, error %d\n", + __func__, chap->qid, ret); + goto out_free_psk; + } + dev_dbg(ctrl->device, "%s: generated digest %s\n", + __func__, digest); + ret = nvme_auth_derive_tls_psk(chap->hash_id, psk, psk_len, + digest, &tls_psk); + if (ret) { + dev_warn(ctrl->device, + "%s: qid %d failed to derive TLS psk, error %d\n", + __func__, chap->qid, ret); + goto out_free_digest; + } + + tls_key = nvme_tls_psk_refresh(ctrl->opts->keyring, + ctrl->opts->host->nqn, + ctrl->opts->subsysnqn, chap->hash_id, + tls_psk, psk_len, digest); + if (IS_ERR(tls_key)) { + ret = PTR_ERR(tls_key); + dev_warn(ctrl->device, + "%s: qid %d failed to insert generated key, error %d\n", + __func__, chap->qid, ret); + tls_key = NULL; + } + kfree_sensitive(tls_psk); + if (ctrl->opts->tls_key) + nvme_auth_revoke_tls_key(ctrl); + ctrl->opts->tls_key = tls_key; +out_free_digest: + kfree_sensitive(digest); +out_free_psk: + kfree_sensitive(psk); + return ret; +} + static void nvme_queue_auth_work(struct work_struct *work) { struct nvme_dhchap_queue_context *chap = @@ -730,7 +834,7 @@ static void nvme_queue_auth_work(struct work_struct *work) NVME_AUTH_DHCHAP_MESSAGE_CHALLENGE); if (ret) { chap->status = ret; - chap->error = NVME_SC_AUTH_REQUIRED; + chap->error = -EKEYREJECTED; return; } @@ -756,12 +860,11 @@ static void nvme_queue_auth_work(struct work_struct *work) __func__, chap->qid); mutex_lock(&ctrl->dhchap_auth_mutex); ret = nvme_auth_dhchap_setup_host_response(ctrl, chap); + mutex_unlock(&ctrl->dhchap_auth_mutex); if (ret) { - mutex_unlock(&ctrl->dhchap_auth_mutex); chap->error = ret; goto fail2; } - mutex_unlock(&ctrl->dhchap_auth_mutex); /* DH-HMAC-CHAP Step 3: send reply */ dev_dbg(ctrl->device, "%s: qid %d send reply\n", @@ -798,7 +901,7 @@ static void nvme_queue_auth_work(struct work_struct *work) NVME_AUTH_DHCHAP_MESSAGE_SUCCESS1); if (ret) { chap->status = ret; - chap->error = NVME_SC_AUTH_REQUIRED; + chap->error = -EKEYREJECTED; return; } @@ -819,11 +922,11 @@ static void nvme_queue_auth_work(struct work_struct *work) ret = nvme_auth_process_dhchap_success1(ctrl, chap); if (ret) { /* Controller authentication failed */ - chap->error = NVME_SC_AUTH_REQUIRED; + chap->error = -EKEYREJECTED; goto fail2; } - if (chap->ctrl_key) { + if (chap->bi_directional) { /* DH-HMAC-CHAP Step 5: send success2 */ dev_dbg(ctrl->device, "%s: qid %d send success2\n", __func__, chap->qid); @@ -834,10 +937,21 @@ static void nvme_queue_auth_work(struct work_struct *work) } if (!ret) { chap->error = 0; + chap->authenticated = true; + if (ctrl->opts->concat && + (ret = nvme_auth_secure_concat(ctrl, chap))) { + dev_warn(ctrl->device, + "%s: qid %d failed to enable secure concatenation\n", + __func__, chap->qid); + chap->error = ret; + chap->authenticated = false; + } return; } fail2: + if (chap->status == 0) + chap->status = NVME_AUTH_DHCHAP_FAILURE_FAILED; dev_dbg(ctrl->device, "%s: qid %d send failure2, status %x\n", __func__, chap->qid, chap->status); tl = nvme_auth_set_dhchap_failure2_data(ctrl, chap); @@ -866,7 +980,7 @@ int nvme_auth_negotiate(struct nvme_ctrl *ctrl, int qid) chap = &ctrl->dhchap_ctxs[qid]; cancel_work_sync(&chap->auth_work); - queue_work(nvme_wq, &chap->auth_work); + queue_work(nvme_auth_wq, &chap->auth_work); return 0; } EXPORT_SYMBOL_GPL(nvme_auth_negotiate); @@ -895,7 +1009,7 @@ static void nvme_ctrl_auth_work(struct work_struct *work) * If the ctrl is no connected, bail as reconnect will handle * authentication. */ - if (ctrl->state != NVME_CTRL_LIVE) + if (nvme_ctrl_state(ctrl) != NVME_CTRL_LIVE) return; /* Authenticate admin queue first */ @@ -911,15 +1025,23 @@ static void nvme_ctrl_auth_work(struct work_struct *work) "qid 0: authentication failed\n"); return; } + /* + * Only run authentication on the admin queue for secure concatenation. + */ + if (ctrl->opts->concat) + return; for (q = 1; q < ctrl->queue_count; q++) { - ret = nvme_auth_negotiate(ctrl, q); - if (ret) { - dev_warn(ctrl->device, - "qid %d: error %d setting up authentication\n", - q, ret); - break; - } + struct nvme_dhchap_queue_context *chap = + &ctrl->dhchap_ctxs[q]; + /* + * Skip re-authentication if the queue had + * not been authenticated initially. + */ + if (!chap->authenticated) + continue; + cancel_work_sync(&chap->auth_work); + queue_work(nvme_auth_wq, &chap->auth_work); } /* @@ -927,7 +1049,13 @@ static void nvme_ctrl_auth_work(struct work_struct *work) * the controller terminates the connection. */ for (q = 1; q < ctrl->queue_count; q++) { - ret = nvme_auth_wait(ctrl, q); + struct nvme_dhchap_queue_context *chap = + &ctrl->dhchap_ctxs[q]; + if (!chap->authenticated) + continue; + flush_work(&chap->auth_work); + ret = chap->error; + nvme_auth_reset_dhchap(chap); if (ret) dev_warn(ctrl->device, "qid %d: authentication failed\n", q); @@ -966,6 +1094,7 @@ int nvme_auth_init_ctrl(struct nvme_ctrl *ctrl) chap = &ctrl->dhchap_ctxs[i]; chap->qid = i; chap->ctrl = ctrl; + chap->authenticated = false; INIT_WORK(&chap->auth_work, nvme_queue_auth_work); } @@ -993,7 +1122,7 @@ void nvme_auth_free(struct nvme_ctrl *ctrl) if (ctrl->dhchap_ctxs) { for (i = 0; i < ctrl_max_dhchaps(ctrl); i++) nvme_auth_free_dhchap(&ctrl->dhchap_ctxs[i]); - kfree(ctrl->dhchap_ctxs); + kvfree(ctrl->dhchap_ctxs); } if (ctrl->host_key) { nvme_auth_free_key(ctrl->host_key); @@ -1008,10 +1137,15 @@ EXPORT_SYMBOL_GPL(nvme_auth_free); int __init nvme_init_auth(void) { + nvme_auth_wq = alloc_workqueue("nvme-auth-wq", + WQ_UNBOUND | WQ_MEM_RECLAIM | WQ_SYSFS, 0); + if (!nvme_auth_wq) + return -ENOMEM; + nvme_chap_buf_cache = kmem_cache_create("nvme-chap-buf-cache", CHAP_BUF_SIZE, 0, SLAB_HWCACHE_ALIGN, NULL); if (!nvme_chap_buf_cache) - return -ENOMEM; + goto err_destroy_workqueue; nvme_chap_buf_pool = mempool_create(16, mempool_alloc_slab, mempool_free_slab, nvme_chap_buf_cache); @@ -1021,6 +1155,8 @@ int __init nvme_init_auth(void) return 0; err_destroy_chap_buf_cache: kmem_cache_destroy(nvme_chap_buf_cache); +err_destroy_workqueue: + destroy_workqueue(nvme_auth_wq); return -ENOMEM; } @@ -1028,4 +1164,5 @@ void __exit nvme_exit_auth(void) { mempool_destroy(nvme_chap_buf_pool); kmem_cache_destroy(nvme_chap_buf_cache); + destroy_workqueue(nvme_auth_wq); } diff --git a/drivers/nvme/host/constants.c b/drivers/nvme/host/constants.c index e958d5015585..dc90df9e13a2 100644 --- a/drivers/nvme/host/constants.c +++ b/drivers/nvme/host/constants.c @@ -12,7 +12,7 @@ static const char * const nvme_ops[] = { [nvme_cmd_read] = "Read", [nvme_cmd_write_uncor] = "Write Uncorrectable", [nvme_cmd_compare] = "Compare", - [nvme_cmd_write_zeroes] = "Write Zeros", + [nvme_cmd_write_zeroes] = "Write Zeroes", [nvme_cmd_dsm] = "Dataset Management", [nvme_cmd_verify] = "Verify", [nvme_cmd_resv_register] = "Reservation Register", @@ -21,7 +21,7 @@ static const char * const nvme_ops[] = { [nvme_cmd_resv_release] = "Reservation Release", [nvme_cmd_zone_mgmt_send] = "Zone Management Send", [nvme_cmd_zone_mgmt_recv] = "Zone Management Receive", - [nvme_cmd_zone_append] = "Zone Management Append", + [nvme_cmd_zone_append] = "Zone Append", }; static const char * const nvme_admin_ops[] = { @@ -54,6 +54,14 @@ static const char * const nvme_admin_ops[] = { [nvme_admin_get_lba_status] = "Get LBA Status", }; +static const char * const nvme_fabrics_ops[] = { + [nvme_fabrics_type_property_set] = "Property Set", + [nvme_fabrics_type_property_get] = "Property Get", + [nvme_fabrics_type_connect] = "Connect", + [nvme_fabrics_type_auth_send] = "Authentication Send", + [nvme_fabrics_type_auth_receive] = "Authentication Receive", +}; + static const char * const nvme_statuses[] = { [NVME_SC_SUCCESS] = "Success", [NVME_SC_INVALID_OPCODE] = "Invalid Command Opcode", @@ -125,7 +133,7 @@ static const char * const nvme_statuses[] = { [NVME_SC_NS_NOT_ATTACHED] = "Namespace Not Attached", [NVME_SC_THIN_PROV_NOT_SUPP] = "Thin Provisioning Not Supported", [NVME_SC_CTRL_LIST_INVALID] = "Controller List Invalid", - [NVME_SC_SELT_TEST_IN_PROGRESS] = "Device Self-test In Progress", + [NVME_SC_SELF_TEST_IN_PROGRESS] = "Device Self-test In Progress", [NVME_SC_BP_WRITE_PROHIBITED] = "Boot Partition Write Prohibited", [NVME_SC_CTRL_ID_INVALID] = "Invalid Controller Identifier", [NVME_SC_SEC_CTRL_STATE_INVALID] = "Invalid Secondary Controller State", @@ -137,7 +145,7 @@ static const char * const nvme_statuses[] = { [NVME_SC_BAD_ATTRIBUTES] = "Conflicting Attributes", [NVME_SC_INVALID_PI] = "Invalid Protection Information", [NVME_SC_READ_ONLY] = "Attempted Write to Read Only Range", - [NVME_SC_ONCS_NOT_SUPPORTED] = "ONCS Not Supported", + [NVME_SC_CMD_SIZE_LIM_EXCEEDED] = "Command Size Limits Exceeded", [NVME_SC_ZONE_BOUNDARY_ERROR] = "Zoned Boundary Error", [NVME_SC_ZONE_FULL] = "Zone Is Full", [NVME_SC_ZONE_READ_ONLY] = "Zone Is Read Only", @@ -163,15 +171,15 @@ static const char * const nvme_statuses[] = { [NVME_SC_HOST_ABORTED_CMD] = "Host Aborted Command", }; -const unsigned char *nvme_get_error_status_str(u16 status) +const char *nvme_get_error_status_str(u16 status) { - status &= 0x7ff; + status &= NVME_SCT_SC_MASK; if (status < ARRAY_SIZE(nvme_statuses) && nvme_statuses[status]) - return nvme_statuses[status & 0x7ff]; + return nvme_statuses[status]; return "Unknown"; } -const unsigned char *nvme_get_opcode_str(u8 opcode) +const char *nvme_get_opcode_str(u8 opcode) { if (opcode < ARRAY_SIZE(nvme_ops) && nvme_ops[opcode]) return nvme_ops[opcode]; @@ -179,9 +187,17 @@ const unsigned char *nvme_get_opcode_str(u8 opcode) } EXPORT_SYMBOL_GPL(nvme_get_opcode_str); -const unsigned char *nvme_get_admin_opcode_str(u8 opcode) +const char *nvme_get_admin_opcode_str(u8 opcode) { if (opcode < ARRAY_SIZE(nvme_admin_ops) && nvme_admin_ops[opcode]) return nvme_admin_ops[opcode]; return "Unknown"; } +EXPORT_SYMBOL_GPL(nvme_get_admin_opcode_str); + +const char *nvme_get_fabrics_opcode_str(u8 opcode) { + if (opcode < ARRAY_SIZE(nvme_fabrics_ops) && nvme_fabrics_ops[opcode]) + return nvme_fabrics_ops[opcode]; + return "Unknown"; +} +EXPORT_SYMBOL_GPL(nvme_get_fabrics_opcode_str); diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c index 505e16f20e57..7bf228df6001 100644 --- a/drivers/nvme/host/core.c +++ b/drivers/nvme/host/core.c @@ -4,6 +4,7 @@ * Copyright (c) 2011-2014, Intel Corporation. */ +#include <linux/async.h> #include <linux/blkdev.h> #include <linux/blk-mq.h> #include <linux/blk-integrity.h> @@ -20,7 +21,8 @@ #include <linux/ptrace.h> #include <linux/nvme_ioctl.h> #include <linux/pm_qos.h> -#include <asm/unaligned.h> +#include <linux/ratelimit.h> +#include <linux/unaligned.h> #include "nvme.h" #include "fabrics.h" @@ -35,9 +37,15 @@ struct nvme_ns_info { struct nvme_ns_ids ids; u32 nsid; __le32 anagrpid; + u8 pi_offset; + u16 endgid; + u64 runs; bool is_shared; bool is_readonly; bool is_ready; + bool is_removed; + bool is_rotational; + bool no_vwc; }; unsigned int admin_timeout = 60; @@ -88,6 +96,17 @@ MODULE_PARM_DESC(apst_secondary_latency_tol_us, "secondary APST latency tolerance in us"); /* + * Older kernels didn't enable protection information if it was at an offset. + * Newer kernels do, so it breaks reads on the upgrade if such formats were + * used in prior kernels since the metadata written did not contain a valid + * checksum. + */ +static bool disable_pi_offsets = false; +module_param(disable_pi_offsets, bool, 0444); +MODULE_PARM_DESC(disable_pi_offsets, + "disable protection information if it has an offset"); + +/* * nvme_wq - hosts nvme related works that are not reset or delete * nvme_reset_wq - hosts nvme reset works * nvme_delete_wq - hosts nvme delete works @@ -108,29 +127,40 @@ struct workqueue_struct *nvme_delete_wq; EXPORT_SYMBOL_GPL(nvme_delete_wq); static LIST_HEAD(nvme_subsystems); -static DEFINE_MUTEX(nvme_subsystems_lock); +DEFINE_MUTEX(nvme_subsystems_lock); static DEFINE_IDA(nvme_instance_ida); static dev_t nvme_ctrl_base_chr_devt; -static struct class *nvme_class; -static struct class *nvme_subsys_class; +static int nvme_class_uevent(const struct device *dev, struct kobj_uevent_env *env); +static const struct class nvme_class = { + .name = "nvme", + .dev_uevent = nvme_class_uevent, +}; + +static const struct class nvme_subsys_class = { + .name = "nvme-subsystem", +}; static DEFINE_IDA(nvme_ns_chr_minor_ida); static dev_t nvme_ns_chr_devt; -static struct class *nvme_ns_chr_class; +static const struct class nvme_ns_chr_class = { + .name = "nvme-generic", +}; static void nvme_put_subsystem(struct nvme_subsystem *subsys); static void nvme_remove_invalid_namespaces(struct nvme_ctrl *ctrl, unsigned nsid); static void nvme_update_keep_alive(struct nvme_ctrl *ctrl, struct nvme_command *cmd); +static int nvme_get_log_lsi(struct nvme_ctrl *ctrl, u32 nsid, u8 log_page, + u8 lsp, u8 csi, void *log, size_t size, u64 offset, u16 lsi); void nvme_queue_scan(struct nvme_ctrl *ctrl) { /* * Only new queue scan work when admin and IO queues are both alive */ - if (ctrl->state == NVME_CTRL_LIVE && ctrl->tagset) + if (nvme_ctrl_state(ctrl) == NVME_CTRL_LIVE && ctrl->tagset) queue_work(nvme_wq, &ctrl->scan_work); } @@ -142,7 +172,7 @@ void nvme_queue_scan(struct nvme_ctrl *ctrl) */ int nvme_try_sched_reset(struct nvme_ctrl *ctrl) { - if (ctrl->state != NVME_CTRL_RESETTING) + if (nvme_ctrl_state(ctrl) != NVME_CTRL_RESETTING) return -EBUSY; if (!queue_work(nvme_reset_wq, &ctrl->reset_work)) return -EBUSY; @@ -155,7 +185,7 @@ static void nvme_failfast_work(struct work_struct *work) struct nvme_ctrl *ctrl = container_of(to_delayed_work(work), struct nvme_ctrl, failfast_work); - if (ctrl->state != NVME_CTRL_CONNECTING) + if (nvme_ctrl_state(ctrl) != NVME_CTRL_CONNECTING) return; set_bit(NVME_CTRL_FAILFAST_EXPIRED, &ctrl->flags); @@ -199,7 +229,7 @@ int nvme_reset_ctrl_sync(struct nvme_ctrl *ctrl) ret = nvme_reset_ctrl(ctrl); if (!ret) { flush_work(&ctrl->reset_work); - if (ctrl->state != NVME_CTRL_LIVE) + if (nvme_ctrl_state(ctrl) != NVME_CTRL_LIVE) ret = -ENETRESET; } @@ -236,7 +266,7 @@ int nvme_delete_ctrl(struct nvme_ctrl *ctrl) } EXPORT_SYMBOL_GPL(nvme_delete_ctrl); -static void nvme_delete_ctrl_sync(struct nvme_ctrl *ctrl) +void nvme_delete_ctrl_sync(struct nvme_ctrl *ctrl) { /* * Keep a reference until nvme_do_delete_ctrl() complete, @@ -250,7 +280,7 @@ static void nvme_delete_ctrl_sync(struct nvme_ctrl *ctrl) static blk_status_t nvme_error_status(u16 status) { - switch (status & 0x7ff) { + switch (status & NVME_SCT_SC_MASK) { case NVME_SC_SUCCESS: return BLK_STS_OK; case NVME_SC_CAP_EXCEEDED: @@ -260,7 +290,6 @@ static blk_status_t nvme_error_status(u16 status) case NVME_SC_NS_NOT_READY: return BLK_STS_TARGET; case NVME_SC_BAD_ATTRIBUTES: - case NVME_SC_ONCS_NOT_SUPPORTED: case NVME_SC_INVALID_OPCODE: case NVME_SC_INVALID_FIELD: case NVME_SC_INVALID_NS: @@ -278,7 +307,7 @@ static blk_status_t nvme_error_status(u16 status) case NVME_SC_INVALID_PI: return BLK_STS_PROTECTION; case NVME_SC_RESERVATION_CONFLICT: - return BLK_STS_NEXUS; + return BLK_STS_RESV_CONFLICT; case NVME_SC_HOST_PATH_ERROR: return BLK_STS_TRANSPORT; case NVME_SC_ZONE_TOO_MANY_ACTIVE: @@ -296,7 +325,7 @@ static void nvme_retry_req(struct request *req) u16 crd; /* The mask and shift result must be <= 3 */ - crd = (nvme_req(req)->status & NVME_SC_CRD) >> 11; + crd = (nvme_req(req)->status & NVME_STATUS_CRD) >> 11; if (crd) delay = nvme_req(req)->ctrl->crdt[crd - 1] * 100; @@ -311,17 +340,17 @@ static void nvme_log_error(struct request *req) struct nvme_request *nr = nvme_req(req); if (ns) { - pr_err_ratelimited("%s: %s(0x%x) @ LBA %llu, %llu blocks, %s (sct 0x%x / sc 0x%x) %s%s\n", + pr_err_ratelimited("%s: %s(0x%x) @ LBA %llu, %u blocks, %s (sct 0x%x / sc 0x%x) %s%s\n", ns->disk ? ns->disk->disk_name : "?", nvme_get_opcode_str(nr->cmd->common.opcode), nr->cmd->common.opcode, - (unsigned long long)nvme_sect_to_lba(ns, blk_rq_pos(req)), - (unsigned long long)blk_rq_bytes(req) >> ns->lba_shift, + nvme_sect_to_lba(ns->head, blk_rq_pos(req)), + blk_rq_bytes(req) >> ns->head->lba_shift, nvme_get_error_status_str(nr->status), - nr->status >> 8 & 7, /* Status Code Type */ - nr->status & 0xff, /* Status Code */ - nr->status & NVME_SC_MORE ? "MORE " : "", - nr->status & NVME_SC_DNR ? "DNR " : ""); + NVME_SCT(nr->status), /* Status Code Type */ + nr->status & NVME_SC_MASK, /* Status Code */ + nr->status & NVME_STATUS_MORE ? "MORE " : "", + nr->status & NVME_STATUS_DNR ? "DNR " : ""); return; } @@ -330,10 +359,34 @@ static void nvme_log_error(struct request *req) nvme_get_admin_opcode_str(nr->cmd->common.opcode), nr->cmd->common.opcode, nvme_get_error_status_str(nr->status), - nr->status >> 8 & 7, /* Status Code Type */ - nr->status & 0xff, /* Status Code */ - nr->status & NVME_SC_MORE ? "MORE " : "", - nr->status & NVME_SC_DNR ? "DNR " : ""); + NVME_SCT(nr->status), /* Status Code Type */ + nr->status & NVME_SC_MASK, /* Status Code */ + nr->status & NVME_STATUS_MORE ? "MORE " : "", + nr->status & NVME_STATUS_DNR ? "DNR " : ""); +} + +static void nvme_log_err_passthru(struct request *req) +{ + struct nvme_ns *ns = req->q->queuedata; + struct nvme_request *nr = nvme_req(req); + + pr_err_ratelimited("%s: %s(0x%x), %s (sct 0x%x / sc 0x%x) %s%s" + "cdw10=0x%x cdw11=0x%x cdw12=0x%x cdw13=0x%x cdw14=0x%x cdw15=0x%x\n", + ns ? ns->disk->disk_name : dev_name(nr->ctrl->device), + ns ? nvme_get_opcode_str(nr->cmd->common.opcode) : + nvme_get_admin_opcode_str(nr->cmd->common.opcode), + nr->cmd->common.opcode, + nvme_get_error_status_str(nr->status), + NVME_SCT(nr->status), /* Status Code Type */ + nr->status & NVME_SC_MASK, /* Status Code */ + nr->status & NVME_STATUS_MORE ? "MORE " : "", + nr->status & NVME_STATUS_DNR ? "DNR " : "", + le32_to_cpu(nr->cmd->common.cdw10), + le32_to_cpu(nr->cmd->common.cdw11), + le32_to_cpu(nr->cmd->common.cdw12), + le32_to_cpu(nr->cmd->common.cdw13), + le32_to_cpu(nr->cmd->common.cdw14), + le32_to_cpu(nr->cmd->common.cdw15)); } enum nvme_disposition { @@ -348,14 +401,14 @@ static inline enum nvme_disposition nvme_decide_disposition(struct request *req) if (likely(nvme_req(req)->status == 0)) return COMPLETE; - if ((nvme_req(req)->status & 0x7ff) == NVME_SC_AUTH_REQUIRED) - return AUTHENTICATE; - if (blk_noretry_request(req) || - (nvme_req(req)->status & NVME_SC_DNR) || + (nvme_req(req)->status & NVME_STATUS_DNR) || nvme_req(req)->retries >= nvme_max_retries) return COMPLETE; + if ((nvme_req(req)->status & NVME_SCT_SC_MASK) == NVME_SC_AUTH_REQUIRED) + return AUTHENTICATE; + if (req->cmd_flags & REQ_NVME_MPATH) { if (nvme_is_path_error(nvme_req(req)->status) || blk_queue_dying(req->q)) @@ -371,21 +424,33 @@ static inline enum nvme_disposition nvme_decide_disposition(struct request *req) static inline void nvme_end_req_zoned(struct request *req) { if (IS_ENABLED(CONFIG_BLK_DEV_ZONED) && - req_op(req) == REQ_OP_ZONE_APPEND) - req->__sector = nvme_lba_to_sect(req->q->queuedata, + req_op(req) == REQ_OP_ZONE_APPEND) { + struct nvme_ns *ns = req->q->queuedata; + + req->__sector = nvme_lba_to_sect(ns->head, le64_to_cpu(nvme_req(req)->result.u64)); + } } -static inline void nvme_end_req(struct request *req) +static inline void __nvme_end_req(struct request *req) { - blk_status_t status = nvme_error_status(nvme_req(req)->status); - - if (unlikely(nvme_req(req)->status && !(req->rq_flags & RQF_QUIET))) - nvme_log_error(req); + if (unlikely(nvme_req(req)->status && !(req->rq_flags & RQF_QUIET))) { + if (blk_rq_is_passthrough(req)) + nvme_log_err_passthru(req); + else + nvme_log_error(req); + } nvme_end_req_zoned(req); nvme_trace_bio_complete(req); if (req->cmd_flags & REQ_NVME_MPATH) nvme_mpath_end_request(req); +} + +void nvme_end_req(struct request *req) +{ + blk_status_t status = nvme_error_status(nvme_req(req)->status); + + __nvme_end_req(req); blk_mq_end_request(req, status); } @@ -396,7 +461,16 @@ void nvme_complete_rq(struct request *req) trace_nvme_complete_rq(req); nvme_cleanup_cmd(req); - if (ctrl->kas) + /* + * Completions of long-running commands should not be able to + * defer sending of periodic keep alives, since the controller + * may have completed processing such commands a long time ago + * (arbitrarily close to command submission time). + * req->deadline - req->timeout is the command submission time + * in jiffies. + */ + if (ctrl->kas && + req->deadline - req->timeout >= ctrl->ka_last_check_time) ctrl->comp_seen = true; switch (nvme_decide_disposition(req)) { @@ -410,7 +484,7 @@ void nvme_complete_rq(struct request *req) nvme_failover_req(req); return; case AUTHENTICATE: -#ifdef CONFIG_NVME_AUTH +#ifdef CONFIG_NVME_HOST_AUTH queue_work(nvme_wq, &ctrl->dhchap_auth_work); nvme_retry_req(req); #else @@ -425,7 +499,7 @@ void nvme_complete_batch_req(struct request *req) { trace_nvme_complete_rq(req); nvme_cleanup_cmd(req); - nvme_end_req_zoned(req); + __nvme_end_req(req); } EXPORT_SYMBOL_GPL(nvme_complete_batch_req); @@ -449,8 +523,8 @@ bool nvme_cancel_request(struct request *req, void *data) dev_dbg_ratelimited(((struct nvme_ctrl *) data)->device, "Cancelling I/O %d", req->tag); - /* don't abort one completed request */ - if (blk_mq_request_completed(req)) + /* don't abort one completed or idle request */ + if (blk_mq_rq_state(req) != MQ_RQ_IN_FLIGHT) return true; nvme_req(req)->status = NVME_SC_HOST_ABORTED_CMD; @@ -489,12 +563,10 @@ bool nvme_change_ctrl_state(struct nvme_ctrl *ctrl, spin_lock_irqsave(&ctrl->lock, flags); - old_state = ctrl->state; + old_state = nvme_ctrl_state(ctrl); switch (new_state) { case NVME_CTRL_LIVE: switch (old_state) { - case NVME_CTRL_NEW: - case NVME_CTRL_RESETTING: case NVME_CTRL_CONNECTING: changed = true; fallthrough; @@ -557,7 +629,7 @@ bool nvme_change_ctrl_state(struct nvme_ctrl *ctrl, } if (changed) { - ctrl->state = new_state; + WRITE_ONCE(ctrl->state, new_state); wake_up_all(&ctrl->state_wq); } @@ -565,11 +637,11 @@ bool nvme_change_ctrl_state(struct nvme_ctrl *ctrl, if (!changed) return false; - if (ctrl->state == NVME_CTRL_LIVE) { + if (new_state == NVME_CTRL_LIVE) { if (old_state == NVME_CTRL_CONNECTING) nvme_stop_failfast_work(ctrl); nvme_kick_requeue_lists(ctrl); - } else if (ctrl->state == NVME_CTRL_CONNECTING && + } else if (new_state == NVME_CTRL_CONNECTING && old_state == NVME_CTRL_RESETTING) { nvme_start_failfast_work(ctrl); } @@ -578,27 +650,6 @@ bool nvme_change_ctrl_state(struct nvme_ctrl *ctrl, EXPORT_SYMBOL_GPL(nvme_change_ctrl_state); /* - * Returns true for sink states that can't ever transition back to live. - */ -static bool nvme_state_terminal(struct nvme_ctrl *ctrl) -{ - switch (ctrl->state) { - case NVME_CTRL_NEW: - case NVME_CTRL_LIVE: - case NVME_CTRL_RESETTING: - case NVME_CTRL_CONNECTING: - return false; - case NVME_CTRL_DELETING: - case NVME_CTRL_DELETING_NOIO: - case NVME_CTRL_DEAD: - return true; - default: - WARN_ONCE(1, "Unhandled ctrl state:%d", ctrl->state); - return true; - } -} - -/* * Waits for the controller state to be resetting, or returns false if it is * not possible to ever transition to that state. */ @@ -607,7 +658,7 @@ bool nvme_wait_reset(struct nvme_ctrl *ctrl) wait_event(ctrl->state_wq, nvme_change_ctrl_state(ctrl, NVME_CTRL_RESETTING) || nvme_state_terminal(ctrl)); - return ctrl->state == NVME_CTRL_RESETTING; + return nvme_ctrl_state(ctrl) == NVME_CTRL_RESETTING; } EXPORT_SYMBOL_GPL(nvme_wait_reset); @@ -616,10 +667,11 @@ static void nvme_free_ns_head(struct kref *ref) struct nvme_ns_head *head = container_of(ref, struct nvme_ns_head, ref); - nvme_mpath_remove_disk(head); + nvme_mpath_put_disk(head); ida_free(&head->subsys->ns_ida, head->instance); cleanup_srcu_struct(&head->srcu); nvme_put_subsystem(head->subsys); + kfree(head->plids); kfree(head); } @@ -643,7 +695,7 @@ static void nvme_free_ns(struct kref *kref) kfree(ns); } -static inline bool nvme_get_ns(struct nvme_ns *ns) +bool nvme_get_ns(struct nvme_ns *ns) { return kref_get_unless_zero(&ns->kref); } @@ -652,7 +704,7 @@ void nvme_put_ns(struct nvme_ns *ns) { kref_put(&ns->kref, nvme_free_ns); } -EXPORT_SYMBOL_NS_GPL(nvme_put_ns, NVME_TARGET_PASSTHRU); +EXPORT_SYMBOL_NS_GPL(nvme_put_ns, "NVME_TARGET_PASSTHRU"); static inline void nvme_clear_nvme_request(struct request *req) { @@ -665,10 +717,21 @@ static inline void nvme_clear_nvme_request(struct request *req) /* initialize a passthrough request */ void nvme_init_request(struct request *req, struct nvme_command *cmd) { - if (req->q->queuedata) + struct nvme_request *nr = nvme_req(req); + bool logging_enabled; + + if (req->q->queuedata) { + struct nvme_ns *ns = req->q->disk->private_data; + + logging_enabled = ns->head->passthru_err_log_enabled; req->timeout = NVME_IO_TIMEOUT; - else /* no queuedata implies admin queue */ + } else { /* no queuedata implies admin queue */ + logging_enabled = nr->ctrl->passthru_err_log_enabled; req->timeout = NVME_ADMIN_TIMEOUT; + } + + if (!logging_enabled) + req->rq_flags |= RQF_QUIET; /* passthru commands should let the driver set the SGL flags */ cmd->common.flags &= ~NVME_CMD_SGL_ALL; @@ -677,8 +740,7 @@ void nvme_init_request(struct request *req, struct nvme_command *cmd) if (req->mq_hctx->type == HCTX_TYPE_POLL) req->cmd_flags |= REQ_POLLED; nvme_clear_nvme_request(req); - req->rq_flags |= RQF_QUIET; - memcpy(nvme_req(req)->cmd, cmd, sizeof(*cmd)); + memcpy(nr->cmd, cmd, sizeof(*cmd)); } EXPORT_SYMBOL_GPL(nvme_init_request); @@ -694,18 +756,24 @@ EXPORT_SYMBOL_GPL(nvme_init_request); blk_status_t nvme_fail_nonready_command(struct nvme_ctrl *ctrl, struct request *rq) { - if (ctrl->state != NVME_CTRL_DELETING_NOIO && - ctrl->state != NVME_CTRL_DELETING && - ctrl->state != NVME_CTRL_DEAD && + enum nvme_ctrl_state state = nvme_ctrl_state(ctrl); + + if (state != NVME_CTRL_DELETING_NOIO && + state != NVME_CTRL_DELETING && + state != NVME_CTRL_DEAD && !test_bit(NVME_CTRL_FAILFAST_EXPIRED, &ctrl->flags) && !blk_noretry_request(rq) && !(rq->cmd_flags & REQ_NVME_MPATH)) return BLK_STS_RESOURCE; + + if (!(rq->rq_flags & RQF_DONTPREP)) + nvme_clear_nvme_request(rq); + return nvme_host_path_error(rq); } EXPORT_SYMBOL_GPL(nvme_fail_nonready_command); bool __nvme_check_ready(struct nvme_ctrl *ctrl, struct request *rq, - bool queue_live) + bool queue_live, enum nvme_ctrl_state state) { struct nvme_request *req = nvme_req(rq); @@ -726,7 +794,7 @@ bool __nvme_check_ready(struct nvme_ctrl *ctrl, struct request *rq, * command, which is require to set the queue live in the * appropinquate states. */ - switch (ctrl->state) { + switch (state) { case NVME_CTRL_CONNECTING: if (blk_rq_is_passthrough(rq) && nvme_is_fabrics(req->cmd) && (req->cmd->fabrics.fctype == nvme_fabrics_type_connect || @@ -780,16 +848,27 @@ static blk_status_t nvme_setup_discard(struct nvme_ns *ns, struct request *req, range = page_address(ns->ctrl->discard_page); } - __rq_for_each_bio(bio, req) { - u64 slba = nvme_sect_to_lba(ns, bio->bi_iter.bi_sector); - u32 nlb = bio->bi_iter.bi_size >> ns->lba_shift; + if (queue_max_discard_segments(req->q) == 1) { + u64 slba = nvme_sect_to_lba(ns->head, blk_rq_pos(req)); + u32 nlb = blk_rq_sectors(req) >> (ns->head->lba_shift - 9); - if (n < segments) { - range[n].cattr = cpu_to_le32(0); - range[n].nlb = cpu_to_le32(nlb); - range[n].slba = cpu_to_le64(slba); + range[0].cattr = cpu_to_le32(0); + range[0].nlb = cpu_to_le32(nlb); + range[0].slba = cpu_to_le64(slba); + n = 1; + } else { + __rq_for_each_bio(bio, req) { + u64 slba = nvme_sect_to_lba(ns->head, + bio->bi_iter.bi_sector); + u32 nlb = bio->bi_iter.bi_size >> ns->head->lba_shift; + + if (n < segments) { + range[n].cattr = cpu_to_le32(0); + range[n].nlb = cpu_to_le32(nlb); + range[n].slba = cpu_to_le64(slba); + } + n++; } - n++; } if (WARN_ON_ONCE(n != segments)) { @@ -806,22 +885,35 @@ static blk_status_t nvme_setup_discard(struct nvme_ns *ns, struct request *req, cmnd->dsm.nr = cpu_to_le32(segments - 1); cmnd->dsm.attributes = cpu_to_le32(NVME_DSMGMT_AD); - req->special_vec.bv_page = virt_to_page(range); - req->special_vec.bv_offset = offset_in_page(range); - req->special_vec.bv_len = alloc_size; + bvec_set_virt(&req->special_vec, range, alloc_size); req->rq_flags |= RQF_SPECIAL_PAYLOAD; return BLK_STS_OK; } +static void nvme_set_app_tag(struct request *req, struct nvme_command *cmnd) +{ + cmnd->rw.lbat = cpu_to_le16(bio_integrity(req->bio)->app_tag); + cmnd->rw.lbatm = cpu_to_le16(0xffff); +} + static void nvme_set_ref_tag(struct nvme_ns *ns, struct nvme_command *cmnd, struct request *req) { u32 upper, lower; u64 ref48; + /* only type1 and type 2 PI formats have a reftag */ + switch (ns->head->pi_type) { + case NVME_NS_DPS_PI_TYPE1: + case NVME_NS_DPS_PI_TYPE2: + break; + default: + return; + } + /* both rw and write zeroes share the same reftag format */ - switch (ns->guard_type) { + switch (ns->head->guard_type) { case NVME_NVM_NS_16B_GUARD: cmnd->rw.reftag = cpu_to_le32(t10_pi_ref_tag(req)); break; @@ -849,27 +941,52 @@ static inline blk_status_t nvme_setup_write_zeroes(struct nvme_ns *ns, cmnd->write_zeroes.opcode = nvme_cmd_write_zeroes; cmnd->write_zeroes.nsid = cpu_to_le32(ns->head->ns_id); cmnd->write_zeroes.slba = - cpu_to_le64(nvme_sect_to_lba(ns, blk_rq_pos(req))); + cpu_to_le64(nvme_sect_to_lba(ns->head, blk_rq_pos(req))); cmnd->write_zeroes.length = - cpu_to_le16((blk_rq_bytes(req) >> ns->lba_shift) - 1); + cpu_to_le16((blk_rq_bytes(req) >> ns->head->lba_shift) - 1); - if (!(req->cmd_flags & REQ_NOUNMAP) && (ns->features & NVME_NS_DEAC)) + if (!(req->cmd_flags & REQ_NOUNMAP) && + (ns->head->features & NVME_NS_DEAC)) cmnd->write_zeroes.control |= cpu_to_le16(NVME_WZ_DEAC); - if (nvme_ns_has_pi(ns)) { + if (nvme_ns_has_pi(ns->head)) { cmnd->write_zeroes.control |= cpu_to_le16(NVME_RW_PRINFO_PRACT); - - switch (ns->pi_type) { - case NVME_NS_DPS_PI_TYPE1: - case NVME_NS_DPS_PI_TYPE2: - nvme_set_ref_tag(ns, cmnd, req); - break; - } + nvme_set_ref_tag(ns, cmnd, req); } return BLK_STS_OK; } +/* + * NVMe does not support a dedicated command to issue an atomic write. A write + * which does adhere to the device atomic limits will silently be executed + * non-atomically. The request issuer should ensure that the write is within + * the queue atomic writes limits, but just validate this in case it is not. + */ +static bool nvme_valid_atomic_write(struct request *req) +{ + struct request_queue *q = req->q; + u32 boundary_bytes = queue_atomic_write_boundary_bytes(q); + + if (blk_rq_bytes(req) > queue_atomic_write_unit_max_bytes(q)) + return false; + + if (boundary_bytes) { + u64 mask = boundary_bytes - 1, imask = ~mask; + u64 start = blk_rq_pos(req) << SECTOR_SHIFT; + u64 end = start + blk_rq_bytes(req) - 1; + + /* If greater then must be crossing a boundary */ + if (blk_rq_bytes(req) > boundary_bytes) + return false; + + if ((start & imask) != (end & imask)) + return false; + } + + return true; +} + static inline blk_status_t nvme_setup_rw(struct nvme_ns *ns, struct request *req, struct nvme_command *cmnd, enum nvme_opcode op) @@ -885,43 +1002,60 @@ static inline blk_status_t nvme_setup_rw(struct nvme_ns *ns, if (req->cmd_flags & REQ_RAHEAD) dsmgmt |= NVME_RW_DSM_FREQ_PREFETCH; + if (op == nvme_cmd_write && ns->head->nr_plids) { + u16 write_stream = req->bio->bi_write_stream; + + if (WARN_ON_ONCE(write_stream > ns->head->nr_plids)) + return BLK_STS_INVAL; + + if (write_stream) { + dsmgmt |= ns->head->plids[write_stream - 1] << 16; + control |= NVME_RW_DTYPE_DPLCMT; + } + } + + if (req->cmd_flags & REQ_ATOMIC && !nvme_valid_atomic_write(req)) + return BLK_STS_INVAL; + cmnd->rw.opcode = op; cmnd->rw.flags = 0; cmnd->rw.nsid = cpu_to_le32(ns->head->ns_id); cmnd->rw.cdw2 = 0; cmnd->rw.cdw3 = 0; cmnd->rw.metadata = 0; - cmnd->rw.slba = cpu_to_le64(nvme_sect_to_lba(ns, blk_rq_pos(req))); - cmnd->rw.length = cpu_to_le16((blk_rq_bytes(req) >> ns->lba_shift) - 1); + cmnd->rw.slba = + cpu_to_le64(nvme_sect_to_lba(ns->head, blk_rq_pos(req))); + cmnd->rw.length = + cpu_to_le16((blk_rq_bytes(req) >> ns->head->lba_shift) - 1); cmnd->rw.reftag = 0; - cmnd->rw.apptag = 0; - cmnd->rw.appmask = 0; + cmnd->rw.lbat = 0; + cmnd->rw.lbatm = 0; - if (ns->ms) { + if (ns->head->ms) { /* - * If formated with metadata, the block layer always provides a + * If formatted with metadata, the block layer always provides a * metadata buffer if CONFIG_BLK_DEV_INTEGRITY is enabled. Else * we enable the PRACT bit for protection information or set the * namespace capacity to zero to prevent any I/O. */ if (!blk_integrity_rq(req)) { - if (WARN_ON_ONCE(!nvme_ns_has_pi(ns))) + if (WARN_ON_ONCE(!nvme_ns_has_pi(ns->head))) return BLK_STS_NOTSUPP; control |= NVME_RW_PRINFO_PRACT; + nvme_set_ref_tag(ns, cmnd, req); } - switch (ns->pi_type) { - case NVME_NS_DPS_PI_TYPE3: + if (bio_integrity_flagged(req->bio, BIP_CHECK_GUARD)) control |= NVME_RW_PRINFO_PRCHK_GUARD; - break; - case NVME_NS_DPS_PI_TYPE1: - case NVME_NS_DPS_PI_TYPE2: - control |= NVME_RW_PRINFO_PRCHK_GUARD | - NVME_RW_PRINFO_PRCHK_REF; + if (bio_integrity_flagged(req->bio, BIP_CHECK_REFTAG)) { + control |= NVME_RW_PRINFO_PRCHK_REF; if (op == nvme_cmd_zone_append) control |= NVME_RW_APPEND_PIREMAP; nvme_set_ref_tag(ns, cmnd, req); - break; + } + if (bio_integrity_flagged(req->bio, BIP_CHECK_APPTAG)) { + control |= NVME_RW_PRINFO_PRCHK_APP; + nvme_set_app_tag(req, cmnd); } } @@ -939,6 +1073,7 @@ void nvme_cleanup_cmd(struct request *req) clear_bit_unlock(0, &ctrl->discard_page_busy); else kfree(bvec_virt(&req->special_vec)); + req->rq_flags &= ~RQF_SPECIAL_PAYLOAD; } } EXPORT_SYMBOL_GPL(nvme_cleanup_cmd); @@ -1004,7 +1139,7 @@ EXPORT_SYMBOL_GPL(nvme_setup_cmd); * >0: nvme controller's cqe status response * <0: kernel error in lieu of controller response */ -static int nvme_execute_rq(struct request *rq, bool at_head) +int nvme_execute_rq(struct request *rq, bool at_head) { blk_status_t status; @@ -1015,6 +1150,7 @@ static int nvme_execute_rq(struct request *rq, bool at_head) return nvme_req(rq)->status; return blk_status_to_errno(status); } +EXPORT_SYMBOL_NS_GPL(nvme_execute_rq, "NVME_TARGET_PASSTHRU"); /* * Returns 0 on success. If the result is negative, it's a Linux error code; @@ -1022,28 +1158,35 @@ static int nvme_execute_rq(struct request *rq, bool at_head) */ int __nvme_submit_sync_cmd(struct request_queue *q, struct nvme_command *cmd, union nvme_result *result, void *buffer, unsigned bufflen, - int qid, int at_head, blk_mq_req_flags_t flags) + int qid, nvme_submit_flags_t flags) { struct request *req; int ret; + blk_mq_req_flags_t blk_flags = 0; + if (flags & NVME_SUBMIT_NOWAIT) + blk_flags |= BLK_MQ_REQ_NOWAIT; + if (flags & NVME_SUBMIT_RESERVED) + blk_flags |= BLK_MQ_REQ_RESERVED; if (qid == NVME_QID_ANY) - req = blk_mq_alloc_request(q, nvme_req_op(cmd), flags); + req = blk_mq_alloc_request(q, nvme_req_op(cmd), blk_flags); else - req = blk_mq_alloc_request_hctx(q, nvme_req_op(cmd), flags, + req = blk_mq_alloc_request_hctx(q, nvme_req_op(cmd), blk_flags, qid - 1); if (IS_ERR(req)) return PTR_ERR(req); nvme_init_request(req, cmd); + if (flags & NVME_SUBMIT_RETRY) + req->cmd_flags &= ~REQ_FAILFAST_DRIVER; if (buffer && bufflen) { - ret = blk_rq_map_kern(q, req, buffer, bufflen, GFP_KERNEL); + ret = blk_rq_map_kern(req, buffer, bufflen, GFP_KERNEL); if (ret) goto out; } - ret = nvme_execute_rq(req, at_head); + ret = nvme_execute_rq(req, flags & NVME_SUBMIT_AT_HEAD); if (result && ret >= 0) *result = nvme_req(req)->result; out: @@ -1056,45 +1199,16 @@ int nvme_submit_sync_cmd(struct request_queue *q, struct nvme_command *cmd, void *buffer, unsigned bufflen) { return __nvme_submit_sync_cmd(q, cmd, NULL, buffer, bufflen, - NVME_QID_ANY, 0, 0); + NVME_QID_ANY, 0); } EXPORT_SYMBOL_GPL(nvme_submit_sync_cmd); -static u32 nvme_known_admin_effects(u8 opcode) -{ - switch (opcode) { - case nvme_admin_format_nvm: - return NVME_CMD_EFFECTS_LBCC | NVME_CMD_EFFECTS_NCC | - NVME_CMD_EFFECTS_CSE_MASK; - case nvme_admin_sanitize_nvm: - return NVME_CMD_EFFECTS_LBCC | NVME_CMD_EFFECTS_CSE_MASK; - default: - break; - } - return 0; -} - -static u32 nvme_known_nvm_effects(u8 opcode) -{ - switch (opcode) { - case nvme_cmd_write: - case nvme_cmd_write_zeroes: - case nvme_cmd_write_uncor: - return NVME_CMD_EFFECTS_LBCC; - default: - return 0; - } -} - u32 nvme_command_effects(struct nvme_ctrl *ctrl, struct nvme_ns *ns, u8 opcode) { u32 effects = 0; if (ns) { - if (ns->head->effects) - effects = le32_to_cpu(ns->head->effects->iocs[opcode]); - if (ns->head->ids.csi == NVME_CSI_NVM) - effects |= nvme_known_nvm_effects(opcode); + effects = le32_to_cpu(ns->head->effects->iocs[opcode]); if (effects & ~(NVME_CMD_EFFECTS_CSUPP | NVME_CMD_EFFECTS_LBCC)) dev_warn_once(ctrl->device, "IO command:%02x has unusual effects:%08x\n", @@ -1107,17 +1221,18 @@ u32 nvme_command_effects(struct nvme_ctrl *ctrl, struct nvme_ns *ns, u8 opcode) */ effects &= ~NVME_CMD_EFFECTS_CSE_MASK; } else { - if (ctrl->effects) - effects = le32_to_cpu(ctrl->effects->acs[opcode]); - effects |= nvme_known_admin_effects(opcode); + effects = le32_to_cpu(ctrl->effects->acs[opcode]); + + /* Ignore execution restrictions if any relaxation bits are set */ + if (effects & NVME_CMD_EFFECTS_CSER_MASK) + effects &= ~NVME_CMD_EFFECTS_CSE_MASK; } return effects; } -EXPORT_SYMBOL_NS_GPL(nvme_command_effects, NVME_TARGET_PASSTHRU); +EXPORT_SYMBOL_NS_GPL(nvme_command_effects, "NVME_TARGET_PASSTHRU"); -static u32 nvme_passthru_start(struct nvme_ctrl *ctrl, struct nvme_ns *ns, - u8 opcode) +u32 nvme_passthru_start(struct nvme_ctrl *ctrl, struct nvme_ns *ns, u8 opcode) { u32 effects = nvme_command_effects(ctrl, ns, opcode); @@ -1135,8 +1250,9 @@ static u32 nvme_passthru_start(struct nvme_ctrl *ctrl, struct nvme_ns *ns, } return effects; } +EXPORT_SYMBOL_NS_GPL(nvme_passthru_start, "NVME_TARGET_PASSTHRU"); -void nvme_passthru_end(struct nvme_ctrl *ctrl, u32 effects, +void nvme_passthru_end(struct nvme_ctrl *ctrl, struct nvme_ns *ns, u32 effects, struct nvme_command *cmd, int status) { if (effects & NVME_CMD_EFFECTS_CSE_MASK) { @@ -1146,13 +1262,18 @@ void nvme_passthru_end(struct nvme_ctrl *ctrl, u32 effects, mutex_unlock(&ctrl->scan_lock); } if (effects & NVME_CMD_EFFECTS_CCC) { - dev_info(ctrl->device, + if (!test_and_set_bit(NVME_CTRL_DIRTY_CAPABILITY, + &ctrl->flags)) { + dev_info(ctrl->device, "controller capabilities changed, reset may be required to take effect.\n"); + } } if (effects & (NVME_CMD_EFFECTS_NIC | NVME_CMD_EFFECTS_NCC)) { nvme_queue_scan(ctrl); flush_work(&ctrl->scan_work); } + if (ns) + return; switch (cmd->common.opcode) { case nvme_admin_set_features: @@ -1174,36 +1295,62 @@ void nvme_passthru_end(struct nvme_ctrl *ctrl, u32 effects, break; } } -EXPORT_SYMBOL_NS_GPL(nvme_passthru_end, NVME_TARGET_PASSTHRU); - -int nvme_execute_passthru_rq(struct request *rq, u32 *effects) -{ - struct nvme_command *cmd = nvme_req(rq)->cmd; - struct nvme_ctrl *ctrl = nvme_req(rq)->ctrl; - struct nvme_ns *ns = rq->q->queuedata; - - *effects = nvme_passthru_start(ctrl, ns, cmd->common.opcode); - return nvme_execute_rq(rq, false); -} -EXPORT_SYMBOL_NS_GPL(nvme_execute_passthru_rq, NVME_TARGET_PASSTHRU); +EXPORT_SYMBOL_NS_GPL(nvme_passthru_end, "NVME_TARGET_PASSTHRU"); /* * Recommended frequency for KATO commands per NVMe 1.4 section 7.12.1: - * + * * The host should send Keep Alive commands at half of the Keep Alive Timeout * accounting for transport roundtrip times [..]. */ +static unsigned long nvme_keep_alive_work_period(struct nvme_ctrl *ctrl) +{ + unsigned long delay = ctrl->kato * HZ / 2; + + /* + * When using Traffic Based Keep Alive, we need to run + * nvme_keep_alive_work at twice the normal frequency, as one + * command completion can postpone sending a keep alive command + * by up to twice the delay between runs. + */ + if (ctrl->ctratt & NVME_CTRL_ATTR_TBKAS) + delay /= 2; + return delay; +} + static void nvme_queue_keep_alive_work(struct nvme_ctrl *ctrl) { - queue_delayed_work(nvme_wq, &ctrl->ka_work, ctrl->kato * HZ / 2); + unsigned long now = jiffies; + unsigned long delay = nvme_keep_alive_work_period(ctrl); + unsigned long ka_next_check_tm = ctrl->ka_last_check_time + delay; + + if (time_after(now, ka_next_check_tm)) + delay = 0; + else + delay = ka_next_check_tm - now; + + queue_delayed_work(nvme_wq, &ctrl->ka_work, delay); } static enum rq_end_io_ret nvme_keep_alive_end_io(struct request *rq, blk_status_t status) { struct nvme_ctrl *ctrl = rq->end_io_data; - unsigned long flags; - bool startka = false; + unsigned long rtt = jiffies - (rq->deadline - rq->timeout); + unsigned long delay = nvme_keep_alive_work_period(ctrl); + enum nvme_ctrl_state state = nvme_ctrl_state(ctrl); + + /* + * Subtract off the keepalive RTT so nvme_keep_alive_work runs + * at the desired frequency. + */ + if (rtt <= delay) { + delay -= rtt; + } else { + dev_warn(ctrl->device, "long keepalive RTT (%u ms)\n", + jiffies_to_msecs(rtt)); + delay = 0; + } blk_mq_free_request(rq); @@ -1214,14 +1361,10 @@ static enum rq_end_io_ret nvme_keep_alive_end_io(struct request *rq, return RQ_END_IO_NONE; } + ctrl->ka_last_check_time = jiffies; ctrl->comp_seen = false; - spin_lock_irqsave(&ctrl->lock, flags); - if (ctrl->state == NVME_CTRL_LIVE || - ctrl->state == NVME_CTRL_CONNECTING) - startka = true; - spin_unlock_irqrestore(&ctrl->lock, flags); - if (startka) - nvme_queue_keep_alive_work(ctrl); + if (state == NVME_CTRL_LIVE || state == NVME_CTRL_CONNECTING) + queue_delayed_work(nvme_wq, &ctrl->ka_work, delay); return RQ_END_IO_NONE; } @@ -1232,6 +1375,8 @@ static void nvme_keep_alive_work(struct work_struct *work) bool comp_seen = ctrl->comp_seen; struct request *rq; + ctrl->ka_last_check_time = jiffies; + if ((ctrl->ctratt & NVME_CTRL_ATTR_TBKAS) && comp_seen) { dev_dbg(ctrl->device, "reschedule traffic based keep-alive timer\n"); @@ -1288,17 +1433,30 @@ static void nvme_update_keep_alive(struct nvme_ctrl *ctrl, nvme_start_keep_alive(ctrl); } -/* - * In NVMe 1.0 the CNS field was just a binary controller or namespace - * flag, thus sending any new CNS opcodes has a big chance of not working. - * Qemu unfortunately had that bug after reporting a 1.1 version compliance - * (but not for any later version). - */ -static bool nvme_ctrl_limited_cns(struct nvme_ctrl *ctrl) +static bool nvme_id_cns_ok(struct nvme_ctrl *ctrl, u8 cns) { - if (ctrl->quirks & NVME_QUIRK_IDENTIFY_CNS) - return ctrl->vs < NVME_VS(1, 2, 0); - return ctrl->vs < NVME_VS(1, 1, 0); + /* + * The CNS field occupies a full byte starting with NVMe 1.2 + */ + if (ctrl->vs >= NVME_VS(1, 2, 0)) + return true; + + /* + * NVMe 1.1 expanded the CNS value to two bits, which means values + * larger than that could get truncated and treated as an incorrect + * value. + * + * Qemu implemented 1.0 behavior for controllers claiming 1.1 + * compliance, so they need to be quirked here. + */ + if (ctrl->vs >= NVME_VS(1, 1, 0) && + !(ctrl->quirks & NVME_QUIRK_IDENTIFY_CNS)) + return cns <= 3; + + /* + * NVMe 1.0 used a single bit for the CNS value. + */ + return cns <= 1; } static int nvme_identify_ctrl(struct nvme_ctrl *dev, struct nvme_id_ctrl **id) @@ -1316,8 +1474,10 @@ static int nvme_identify_ctrl(struct nvme_ctrl *dev, struct nvme_id_ctrl **id) error = nvme_submit_sync_cmd(dev->admin_q, &c, *id, sizeof(struct nvme_id_ctrl)); - if (error) + if (error) { kfree(*id); + *id = NULL; + } return error; } @@ -1427,7 +1587,7 @@ free_data: return status; } -static int nvme_identify_ns(struct nvme_ctrl *ctrl, unsigned nsid, +int nvme_identify_ns(struct nvme_ctrl *ctrl, unsigned nsid, struct nvme_id_ns **id) { struct nvme_command c = { }; @@ -1445,16 +1605,9 @@ static int nvme_identify_ns(struct nvme_ctrl *ctrl, unsigned nsid, error = nvme_submit_sync_cmd(ctrl->admin_q, &c, *id, sizeof(**id)); if (error) { dev_warn(ctrl->device, "Identify namespace failed (%d)\n", error); - goto out_free_id; + kfree(*id); + *id = NULL; } - - error = NVME_SC_INVALID_NS | NVME_SC_DNR; - if ((*id)->ncap == 0) /* namespace not allocated or attached */ - goto out_free_id; - return 0; - -out_free_id: - kfree(*id); return error; } @@ -1468,10 +1621,19 @@ static int nvme_ns_info_from_identify(struct nvme_ctrl *ctrl, ret = nvme_identify_ns(ctrl, info->nsid, &id); if (ret) return ret; + + if (id->ncap == 0) { + /* namespace not allocated or attached */ + info->is_removed = true; + ret = -ENODEV; + goto error; + } + info->anagrpid = id->anagrpid; info->is_shared = id->nmic & NVME_NS_NMIC_SHARED; info->is_readonly = id->nsattr & NVME_NS_ATTR_RO; info->is_ready = true; + info->endgid = le16_to_cpu(id->endgid); if (ctrl->quirks & NVME_QUIRK_BOGUS_NID) { dev_info(ctrl->device, "Ignoring bogus Namespace Identifiers\n"); @@ -1483,8 +1645,10 @@ static int nvme_ns_info_from_identify(struct nvme_ctrl *ctrl, !memchr_inv(ids->nguid, 0, sizeof(ids->nguid))) memcpy(ids->nguid, id->nguid, sizeof(ids->nguid)); } + +error: kfree(id); - return 0; + return ret; } static int nvme_ns_info_from_id_cs_indep(struct nvme_ctrl *ctrl, @@ -1508,6 +1672,9 @@ static int nvme_ns_info_from_id_cs_indep(struct nvme_ctrl *ctrl, info->is_shared = id->nmic & NVME_NS_NMIC_SHARED; info->is_readonly = id->nsattr & NVME_NS_ATTR_RO; info->is_ready = id->nstat & NVME_NSTAT_NRDY; + info->is_rotational = id->nsfeat & NVME_NS_ROTATIONAL; + info->no_vwc = id->nsfeat & NVME_NS_VWC_NOT_PRESENT; + info->endgid = le16_to_cpu(id->endgid); } kfree(id); return ret; @@ -1525,7 +1692,7 @@ static int nvme_features(struct nvme_ctrl *dev, u8 op, unsigned int fid, c.features.dword11 = cpu_to_le32(dword11); ret = __nvme_submit_sync_cmd(dev->admin_q, &c, &res, - buffer, buflen, NVME_QID_ANY, 0, 0); + buffer, buflen, NVME_QID_ANY, 0); if (ret >= 0 && result) *result = le32_to_cpu(res.u32); return ret; @@ -1533,7 +1700,7 @@ static int nvme_features(struct nvme_ctrl *dev, u8 op, unsigned int fid, int nvme_set_features(struct nvme_ctrl *dev, unsigned int fid, unsigned int dword11, void *buffer, size_t buflen, - u32 *result) + void *result) { return nvme_features(dev, nvme_admin_set_features, fid, dword11, buffer, buflen, result); @@ -1542,7 +1709,7 @@ EXPORT_SYMBOL_GPL(nvme_set_features); int nvme_get_features(struct nvme_ctrl *dev, unsigned int fid, unsigned int dword11, void *buffer, size_t buflen, - u32 *result) + void *result) { return nvme_features(dev, nvme_admin_get_features, fid, dword11, buffer, buflen, result); @@ -1557,7 +1724,13 @@ int nvme_set_queue_count(struct nvme_ctrl *ctrl, int *count) status = nvme_set_features(ctrl, NVME_FEAT_NUM_QUEUES, q_count, NULL, 0, &result); - if (status < 0) + + /* + * It's either a kernel error or the host observed a connection + * lost. In either case it's not possible communicate with the + * controller and thus enter the error code path. + */ + if (status < 0 || status == NVME_SC_HOST_PATH_ERROR) return status; /* @@ -1624,111 +1797,109 @@ static void nvme_ns_release(struct nvme_ns *ns) nvme_put_ns(ns); } -static int nvme_open(struct block_device *bdev, fmode_t mode) +static int nvme_open(struct gendisk *disk, blk_mode_t mode) { - return nvme_ns_open(bdev->bd_disk->private_data); + return nvme_ns_open(disk->private_data); } -static void nvme_release(struct gendisk *disk, fmode_t mode) +static void nvme_release(struct gendisk *disk) { nvme_ns_release(disk->private_data); } -int nvme_getgeo(struct block_device *bdev, struct hd_geometry *geo) +int nvme_getgeo(struct gendisk *disk, struct hd_geometry *geo) { /* some standard values */ geo->heads = 1 << 6; geo->sectors = 1 << 5; - geo->cylinders = get_capacity(bdev->bd_disk) >> 11; + geo->cylinders = get_capacity(disk) >> 11; return 0; } -#ifdef CONFIG_BLK_DEV_INTEGRITY -static void nvme_init_integrity(struct gendisk *disk, struct nvme_ns *ns, - u32 max_integrity_segments) +static bool nvme_init_integrity(struct nvme_ns_head *head, + struct queue_limits *lim, struct nvme_ns_info *info) { - struct blk_integrity integrity = { }; + struct blk_integrity *bi = &lim->integrity; + + memset(bi, 0, sizeof(*bi)); + + if (!head->ms) + return true; + + /* + * PI can always be supported as we can ask the controller to simply + * insert/strip it, which is not possible for other kinds of metadata. + */ + if (!IS_ENABLED(CONFIG_BLK_DEV_INTEGRITY) || + !(head->features & NVME_NS_METADATA_SUPPORTED)) + return nvme_ns_has_pi(head); - switch (ns->pi_type) { + switch (head->pi_type) { case NVME_NS_DPS_PI_TYPE3: - switch (ns->guard_type) { + switch (head->guard_type) { case NVME_NVM_NS_16B_GUARD: - integrity.profile = &t10_pi_type3_crc; - integrity.tag_size = sizeof(u16) + sizeof(u32); - integrity.flags |= BLK_INTEGRITY_DEVICE_CAPABLE; + bi->csum_type = BLK_INTEGRITY_CSUM_CRC; + bi->tag_size = sizeof(u16) + sizeof(u32); + bi->flags |= BLK_INTEGRITY_DEVICE_CAPABLE; break; case NVME_NVM_NS_64B_GUARD: - integrity.profile = &ext_pi_type3_crc64; - integrity.tag_size = sizeof(u16) + 6; - integrity.flags |= BLK_INTEGRITY_DEVICE_CAPABLE; + bi->csum_type = BLK_INTEGRITY_CSUM_CRC64; + bi->tag_size = sizeof(u16) + 6; + bi->flags |= BLK_INTEGRITY_DEVICE_CAPABLE; break; default: - integrity.profile = NULL; break; } break; case NVME_NS_DPS_PI_TYPE1: case NVME_NS_DPS_PI_TYPE2: - switch (ns->guard_type) { + switch (head->guard_type) { case NVME_NVM_NS_16B_GUARD: - integrity.profile = &t10_pi_type1_crc; - integrity.tag_size = sizeof(u16); - integrity.flags |= BLK_INTEGRITY_DEVICE_CAPABLE; + bi->csum_type = BLK_INTEGRITY_CSUM_CRC; + bi->tag_size = sizeof(u16); + bi->flags |= BLK_INTEGRITY_DEVICE_CAPABLE | + BLK_INTEGRITY_REF_TAG; break; case NVME_NVM_NS_64B_GUARD: - integrity.profile = &ext_pi_type1_crc64; - integrity.tag_size = sizeof(u16); - integrity.flags |= BLK_INTEGRITY_DEVICE_CAPABLE; + bi->csum_type = BLK_INTEGRITY_CSUM_CRC64; + bi->tag_size = sizeof(u16); + bi->flags |= BLK_INTEGRITY_DEVICE_CAPABLE | + BLK_INTEGRITY_REF_TAG; break; default: - integrity.profile = NULL; break; } break; default: - integrity.profile = NULL; break; } - integrity.tuple_size = ns->ms; - blk_integrity_register(disk, &integrity); - blk_queue_max_integrity_segments(disk->queue, max_integrity_segments); -} -#else -static void nvme_init_integrity(struct gendisk *disk, struct nvme_ns *ns, - u32 max_integrity_segments) -{ + bi->metadata_size = head->ms; + if (bi->csum_type) { + bi->pi_tuple_size = head->pi_size; + bi->pi_offset = info->pi_offset; + } + return true; } -#endif /* CONFIG_BLK_DEV_INTEGRITY */ -static void nvme_config_discard(struct gendisk *disk, struct nvme_ns *ns) +static void nvme_config_discard(struct nvme_ns *ns, struct queue_limits *lim) { struct nvme_ctrl *ctrl = ns->ctrl; - struct request_queue *queue = disk->queue; - u32 size = queue_logical_block_size(queue); - - if (ctrl->max_discard_sectors == 0) { - blk_queue_max_discard_sectors(queue, 0); - return; - } - - BUILD_BUG_ON(PAGE_SIZE / sizeof(struct nvme_dsm_range) < - NVME_DSM_MAX_RANGES); - - queue->limits.discard_granularity = size; - - /* If discard is already enabled, don't reset queue limits */ - if (queue->limits.max_discard_sectors) - return; - if (ctrl->dmrsl && ctrl->dmrsl <= nvme_sect_to_lba(ns, UINT_MAX)) - ctrl->max_discard_sectors = nvme_lba_to_sect(ns, ctrl->dmrsl); + if (ctrl->dmrsl && ctrl->dmrsl <= nvme_sect_to_lba(ns->head, UINT_MAX)) + lim->max_hw_discard_sectors = + nvme_lba_to_sect(ns->head, ctrl->dmrsl); + else if (ctrl->oncs & NVME_CTRL_ONCS_DSM) + lim->max_hw_discard_sectors = UINT_MAX; + else + lim->max_hw_discard_sectors = 0; - blk_queue_max_discard_sectors(queue, ctrl->max_discard_sectors); - blk_queue_max_discard_segments(queue, ctrl->max_discard_segments); + lim->discard_granularity = lim->logical_block_size; - if (ctrl->quirks & NVME_QUIRK_DEALLOCATE_ZEROES) - blk_queue_max_write_zeroes_sectors(queue, UINT_MAX); + if (ctrl->dmrl) + lim->max_discard_segments = ctrl->dmrl; + else + lim->max_discard_segments = NVME_DSM_MAX_RANGES; } static bool nvme_ns_ids_equal(struct nvme_ns_ids *a, struct nvme_ns_ids *b) @@ -1739,76 +1910,84 @@ static bool nvme_ns_ids_equal(struct nvme_ns_ids *a, struct nvme_ns_ids *b) a->csi == b->csi; } -static int nvme_init_ms(struct nvme_ns *ns, struct nvme_id_ns *id) +static int nvme_identify_ns_nvm(struct nvme_ctrl *ctrl, unsigned int nsid, + struct nvme_id_ns_nvm **nvmp) { - bool first = id->dps & NVME_NS_DPS_PI_FIRST; - unsigned lbaf = nvme_lbaf_index(id->flbas); - struct nvme_ctrl *ctrl = ns->ctrl; - struct nvme_command c = { }; + struct nvme_command c = { + .identify.opcode = nvme_admin_identify, + .identify.nsid = cpu_to_le32(nsid), + .identify.cns = NVME_ID_CNS_CS_NS, + .identify.csi = NVME_CSI_NVM, + }; struct nvme_id_ns_nvm *nvm; - int ret = 0; - u32 elbaf; - - ns->pi_size = 0; - ns->ms = le16_to_cpu(id->lbaf[lbaf].ms); - if (!(ctrl->ctratt & NVME_CTRL_ATTR_ELBAS)) { - ns->pi_size = sizeof(struct t10_pi_tuple); - ns->guard_type = NVME_NVM_NS_16B_GUARD; - goto set_pi; - } + int ret; nvm = kzalloc(sizeof(*nvm), GFP_KERNEL); if (!nvm) return -ENOMEM; - c.identify.opcode = nvme_admin_identify; - c.identify.nsid = cpu_to_le32(ns->head->ns_id); - c.identify.cns = NVME_ID_CNS_CS_NS; - c.identify.csi = NVME_CSI_NVM; - - ret = nvme_submit_sync_cmd(ns->ctrl->admin_q, &c, nvm, sizeof(*nvm)); + ret = nvme_submit_sync_cmd(ctrl->admin_q, &c, nvm, sizeof(*nvm)); if (ret) - goto free_data; + kfree(nvm); + else + *nvmp = nvm; + return ret; +} - elbaf = le32_to_cpu(nvm->elbaf[lbaf]); +static void nvme_configure_pi_elbas(struct nvme_ns_head *head, + struct nvme_id_ns *id, struct nvme_id_ns_nvm *nvm) +{ + u32 elbaf = le32_to_cpu(nvm->elbaf[nvme_lbaf_index(id->flbas)]); + u8 guard_type; /* no support for storage tag formats right now */ if (nvme_elbaf_sts(elbaf)) - goto free_data; + return; - ns->guard_type = nvme_elbaf_guard_type(elbaf); - switch (ns->guard_type) { + guard_type = nvme_elbaf_guard_type(elbaf); + if ((nvm->pic & NVME_ID_NS_NVM_QPIFS) && + guard_type == NVME_NVM_NS_QTYPE_GUARD) + guard_type = nvme_elbaf_qualified_guard_type(elbaf); + + head->guard_type = guard_type; + switch (head->guard_type) { case NVME_NVM_NS_64B_GUARD: - ns->pi_size = sizeof(struct crc64_pi_tuple); + head->pi_size = sizeof(struct crc64_pi_tuple); break; case NVME_NVM_NS_16B_GUARD: - ns->pi_size = sizeof(struct t10_pi_tuple); + head->pi_size = sizeof(struct t10_pi_tuple); break; default: break; } - -free_data: - kfree(nvm); -set_pi: - if (ns->pi_size && (first || ns->ms == ns->pi_size)) - ns->pi_type = id->dps & NVME_NS_DPS_PI_MASK; - else - ns->pi_type = 0; - - return ret; } -static void nvme_configure_metadata(struct nvme_ns *ns, struct nvme_id_ns *id) +static void nvme_configure_metadata(struct nvme_ctrl *ctrl, + struct nvme_ns_head *head, struct nvme_id_ns *id, + struct nvme_id_ns_nvm *nvm, struct nvme_ns_info *info) { - struct nvme_ctrl *ctrl = ns->ctrl; - - if (nvme_init_ms(ns, id)) + head->features &= ~(NVME_NS_METADATA_SUPPORTED | NVME_NS_EXT_LBAS); + head->pi_type = 0; + head->pi_size = 0; + head->ms = le16_to_cpu(id->lbaf[nvme_lbaf_index(id->flbas)].ms); + if (!head->ms || !(ctrl->ops->flags & NVME_F_METADATA_SUPPORTED)) return; - ns->features &= ~(NVME_NS_METADATA_SUPPORTED | NVME_NS_EXT_LBAS); - if (!ns->ms || !(ctrl->ops->flags & NVME_F_METADATA_SUPPORTED)) - return; + if (nvm && (ctrl->ctratt & NVME_CTRL_ATTR_ELBAS)) { + nvme_configure_pi_elbas(head, id, nvm); + } else { + head->pi_size = sizeof(struct t10_pi_tuple); + head->guard_type = NVME_NVM_NS_16B_GUARD; + } + + if (head->pi_size && head->ms >= head->pi_size) + head->pi_type = id->dps & NVME_NS_DPS_PI_MASK; + if (!(id->dps & NVME_NS_DPS_PI_FIRST)) { + if (disable_pi_offsets) + head->pi_type = 0; + else + info->pi_offset = head->ms - head->pi_size; + } if (ctrl->ops->flags & NVME_F_FABRICS) { /* @@ -1819,7 +1998,7 @@ static void nvme_configure_metadata(struct nvme_ns *ns, struct nvme_id_ns *id) if (WARN_ON_ONCE(!(id->flbas & NVME_NS_FLBAS_META_EXT))) return; - ns->features |= NVME_NS_EXT_LBAS; + head->features |= NVME_NS_EXT_LBAS; /* * The current fabrics transport drivers support namespace @@ -1830,8 +2009,8 @@ static void nvme_configure_metadata(struct nvme_ns *ns, struct nvme_id_ns *id) * Note, this check will need to be modified if any drivers * gain the ability to use other metadata formats. */ - if (ctrl->max_integrity_segments && nvme_ns_has_pi(ns)) - ns->features |= NVME_NS_METADATA_SUPPORTED; + if (ctrl->max_integrity_segments && nvme_ns_has_pi(head)) + head->features |= NVME_NS_METADATA_SUPPORTED; } else { /* * For PCIe controllers, we can't easily remap the separate @@ -1840,98 +2019,111 @@ static void nvme_configure_metadata(struct nvme_ns *ns, struct nvme_id_ns *id) * We allow extended LBAs for the passthrough interface, though. */ if (id->flbas & NVME_NS_FLBAS_META_EXT) - ns->features |= NVME_NS_EXT_LBAS; + head->features |= NVME_NS_EXT_LBAS; else - ns->features |= NVME_NS_METADATA_SUPPORTED; + head->features |= NVME_NS_METADATA_SUPPORTED; } } -static void nvme_set_queue_limits(struct nvme_ctrl *ctrl, - struct request_queue *q) + +static u32 nvme_configure_atomic_write(struct nvme_ns *ns, + struct nvme_id_ns *id, struct queue_limits *lim, u32 bs) { - bool vwc = ctrl->vwc & NVME_CTRL_VWC_PRESENT; + u32 atomic_bs, boundary = 0; - if (ctrl->max_hw_sectors) { - u32 max_segments = - (ctrl->max_hw_sectors / (NVME_CTRL_PAGE_SIZE >> 9)) + 1; + /* + * We do not support an offset for the atomic boundaries. + */ + if (id->nabo) + return bs; - max_segments = min_not_zero(max_segments, ctrl->max_segments); - blk_queue_max_hw_sectors(q, ctrl->max_hw_sectors); - blk_queue_max_segments(q, min_t(u32, max_segments, USHRT_MAX)); + if ((id->nsfeat & NVME_NS_FEAT_ATOMICS) && id->nawupf) { + /* + * Use the per-namespace atomic write unit when available. + */ + atomic_bs = (1 + le16_to_cpu(id->nawupf)) * bs; + if (id->nabspf) + boundary = (le16_to_cpu(id->nabspf) + 1) * bs; + } else { + /* + * Use the controller wide atomic write unit. This sucks + * because the limit is defined in terms of logical blocks while + * namespaces can have different formats, and because there is + * no clear language in the specification prohibiting different + * values for different controllers in the subsystem. + */ + atomic_bs = (1 + ns->ctrl->subsys->awupf) * bs; } - blk_queue_virt_boundary(q, NVME_CTRL_PAGE_SIZE - 1); - blk_queue_dma_alignment(q, 3); - blk_queue_write_cache(q, vwc, vwc); + + lim->atomic_write_hw_max = atomic_bs; + lim->atomic_write_hw_boundary = boundary; + lim->atomic_write_hw_unit_min = bs; + lim->atomic_write_hw_unit_max = rounddown_pow_of_two(atomic_bs); + lim->features |= BLK_FEAT_ATOMIC_WRITES; + return atomic_bs; } -static void nvme_update_disk_info(struct gendisk *disk, - struct nvme_ns *ns, struct nvme_id_ns *id) +static u32 nvme_max_drv_segments(struct nvme_ctrl *ctrl) { - sector_t capacity = nvme_lba_to_sect(ns, le64_to_cpu(id->nsze)); - unsigned short bs = 1 << ns->lba_shift; + return ctrl->max_hw_sectors / (NVME_CTRL_PAGE_SIZE >> SECTOR_SHIFT) + 1; +} + +static void nvme_set_ctrl_limits(struct nvme_ctrl *ctrl, + struct queue_limits *lim, bool is_admin) +{ + lim->max_hw_sectors = ctrl->max_hw_sectors; + lim->max_segments = min_t(u32, USHRT_MAX, + min_not_zero(nvme_max_drv_segments(ctrl), ctrl->max_segments)); + lim->max_integrity_segments = ctrl->max_integrity_segments; + lim->virt_boundary_mask = ctrl->ops->get_virt_boundary(ctrl, is_admin); + lim->max_segment_size = UINT_MAX; + lim->dma_alignment = 3; +} + +static bool nvme_update_disk_info(struct nvme_ns *ns, struct nvme_id_ns *id, + struct queue_limits *lim) +{ + struct nvme_ns_head *head = ns->head; + u32 bs = 1U << head->lba_shift; u32 atomic_bs, phys_bs, io_opt = 0; + bool valid = true; /* * The block layer can't support LBA sizes larger than the page size - * yet, so catch this early and don't allow block I/O. + * or smaller than a sector size yet, so catch this early and don't + * allow block I/O. */ - if (ns->lba_shift > PAGE_SHIFT) { - capacity = 0; + if (blk_validate_block_size(bs)) { bs = (1 << 9); + valid = false; } - blk_integrity_unregister(disk); - - atomic_bs = phys_bs = bs; - if (id->nabo == 0) { - /* - * Bit 1 indicates whether NAWUPF is defined for this namespace - * and whether it should be used instead of AWUPF. If NAWUPF == - * 0 then AWUPF must be used instead. - */ - if (id->nsfeat & NVME_NS_FEAT_ATOMICS && id->nawupf) - atomic_bs = (1 + le16_to_cpu(id->nawupf)) * bs; - else - atomic_bs = (1 + ns->ctrl->subsys->awupf) * bs; - } + phys_bs = bs; + atomic_bs = nvme_configure_atomic_write(ns, id, lim, bs); if (id->nsfeat & NVME_NS_FEAT_IO_OPT) { /* NPWG = Namespace Preferred Write Granularity */ phys_bs = bs * (1 + le16_to_cpu(id->npwg)); /* NOWS = Namespace Optimal Write Size */ - io_opt = bs * (1 + le16_to_cpu(id->nows)); + if (id->nows) + io_opt = bs * (1 + le16_to_cpu(id->nows)); } - blk_queue_logical_block_size(disk->queue, bs); /* * Linux filesystems assume writing a single physical block is * an atomic operation. Hence limit the physical block size to the * value of the Atomic Write Unit Power Fail parameter. */ - blk_queue_physical_block_size(disk->queue, min(phys_bs, atomic_bs)); - blk_queue_io_min(disk->queue, phys_bs); - blk_queue_io_opt(disk->queue, io_opt); - - /* - * Register a metadata profile for PI, or the plain non-integrity NVMe - * metadata masquerading as Type 0 if supported, otherwise reject block - * I/O to namespaces with metadata except when the namespace supports - * PI, as it can strip/insert in that case. - */ - if (ns->ms) { - if (IS_ENABLED(CONFIG_BLK_DEV_INTEGRITY) && - (ns->features & NVME_NS_METADATA_SUPPORTED)) - nvme_init_integrity(disk, ns, - ns->ctrl->max_integrity_segments); - else if (!nvme_ns_has_pi(ns)) - capacity = 0; - } - - set_capacity_and_notify(disk, capacity); - - nvme_config_discard(disk, ns); - blk_queue_max_write_zeroes_sectors(disk->queue, - ns->ctrl->max_zeroes_sectors); + lim->logical_block_size = bs; + lim->physical_block_size = min(phys_bs, atomic_bs); + lim->io_min = phys_bs; + lim->io_opt = io_opt; + if ((ns->ctrl->quirks & NVME_QUIRK_DEALLOCATE_ZEROES) && + (ns->ctrl->oncs & NVME_CTRL_ONCS_DSM)) + lim->max_write_zeroes_sectors = UINT_MAX; + else + lim->max_write_zeroes_sectors = ns->ctrl->max_zeroes_sectors; + return valid; } static bool nvme_ns_is_readonly(struct nvme_ns *ns, struct nvme_ns_info *info) @@ -1945,7 +2137,8 @@ static inline bool nvme_first_scan(struct gendisk *disk) return !disk_live(disk); } -static void nvme_set_chunk_sectors(struct nvme_ns *ns, struct nvme_id_ns *id) +static void nvme_set_chunk_sectors(struct nvme_ns *ns, struct nvme_id_ns *id, + struct queue_limits *lim) { struct nvme_ctrl *ctrl = ns->ctrl; u32 iob; @@ -1954,7 +2147,7 @@ static void nvme_set_chunk_sectors(struct nvme_ns *ns, struct nvme_id_ns *id) is_power_of_2(ctrl->max_hw_sectors)) iob = ctrl->max_hw_sectors; else - iob = nvme_lba_to_sect(ns, le16_to_cpu(id->noiob)); + iob = nvme_lba_to_sect(ns->head, le16_to_cpu(id->noiob)); if (!iob) return; @@ -1973,38 +2166,181 @@ static void nvme_set_chunk_sectors(struct nvme_ns *ns, struct nvme_id_ns *id) return; } - blk_queue_chunk_sectors(ns->queue, iob); + lim->chunk_sectors = iob; } static int nvme_update_ns_info_generic(struct nvme_ns *ns, struct nvme_ns_info *info) { - blk_mq_freeze_queue(ns->disk->queue); - nvme_set_queue_limits(ns->ctrl, ns->queue); + struct queue_limits lim; + unsigned int memflags; + int ret; + + lim = queue_limits_start_update(ns->disk->queue); + nvme_set_ctrl_limits(ns->ctrl, &lim, false); + + memflags = blk_mq_freeze_queue(ns->disk->queue); + ret = queue_limits_commit_update(ns->disk->queue, &lim); set_disk_ro(ns->disk, nvme_ns_is_readonly(ns, info)); - blk_mq_unfreeze_queue(ns->disk->queue); + blk_mq_unfreeze_queue(ns->disk->queue, memflags); - if (nvme_ns_head_multipath(ns->head)) { - blk_mq_freeze_queue(ns->head->disk->queue); - set_disk_ro(ns->head->disk, nvme_ns_is_readonly(ns, info)); - nvme_mpath_revalidate_paths(ns); - blk_stack_limits(&ns->head->disk->queue->limits, - &ns->queue->limits, 0); - ns->head->disk->flags |= GENHD_FL_HIDDEN; - blk_mq_unfreeze_queue(ns->head->disk->queue); + /* Hide the block-interface for these devices */ + if (!ret) + ret = -ENODEV; + return ret; +} + +static int nvme_query_fdp_granularity(struct nvme_ctrl *ctrl, + struct nvme_ns_info *info, u8 fdp_idx) +{ + struct nvme_fdp_config_log hdr, *h; + struct nvme_fdp_config_desc *desc; + size_t size = sizeof(hdr); + void *log, *end; + int i, n, ret; + + ret = nvme_get_log_lsi(ctrl, 0, NVME_LOG_FDP_CONFIGS, 0, + NVME_CSI_NVM, &hdr, size, 0, info->endgid); + if (ret) { + dev_warn(ctrl->device, + "FDP configs log header status:0x%x endgid:%d\n", ret, + info->endgid); + return ret; } - /* Hide the block-interface for these devices */ - ns->disk->flags |= GENHD_FL_HIDDEN; - set_bit(NVME_NS_READY, &ns->flags); + size = le32_to_cpu(hdr.sze); + if (size > PAGE_SIZE * MAX_ORDER_NR_PAGES) { + dev_warn(ctrl->device, "FDP config size too large:%zu\n", + size); + return 0; + } - return 0; + h = kvmalloc(size, GFP_KERNEL); + if (!h) + return -ENOMEM; + + ret = nvme_get_log_lsi(ctrl, 0, NVME_LOG_FDP_CONFIGS, 0, + NVME_CSI_NVM, h, size, 0, info->endgid); + if (ret) { + dev_warn(ctrl->device, + "FDP configs log status:0x%x endgid:%d\n", ret, + info->endgid); + goto out; + } + + n = le16_to_cpu(h->numfdpc) + 1; + if (fdp_idx > n) { + dev_warn(ctrl->device, "FDP index:%d out of range:%d\n", + fdp_idx, n); + /* Proceed without registering FDP streams */ + ret = 0; + goto out; + } + + log = h + 1; + desc = log; + end = log + size - sizeof(*h); + for (i = 0; i < fdp_idx; i++) { + log += le16_to_cpu(desc->dsze); + desc = log; + if (log >= end) { + dev_warn(ctrl->device, + "FDP invalid config descriptor list\n"); + ret = 0; + goto out; + } + } + + if (le32_to_cpu(desc->nrg) > 1) { + dev_warn(ctrl->device, "FDP NRG > 1 not supported\n"); + ret = 0; + goto out; + } + + info->runs = le64_to_cpu(desc->runs); +out: + kvfree(h); + return ret; +} + +static int nvme_query_fdp_info(struct nvme_ns *ns, struct nvme_ns_info *info) +{ + struct nvme_ns_head *head = ns->head; + struct nvme_ctrl *ctrl = ns->ctrl; + struct nvme_fdp_ruh_status *ruhs; + struct nvme_fdp_config fdp; + struct nvme_command c = {}; + size_t size; + int i, ret; + + /* + * The FDP configuration is static for the lifetime of the namespace, + * so return immediately if we've already registered this namespace's + * streams. + */ + if (head->nr_plids) + return 0; + + ret = nvme_get_features(ctrl, NVME_FEAT_FDP, info->endgid, NULL, 0, + &fdp); + if (ret) { + dev_warn(ctrl->device, "FDP get feature status:0x%x\n", ret); + return ret; + } + + if (!(fdp.flags & FDPCFG_FDPE)) + return 0; + + ret = nvme_query_fdp_granularity(ctrl, info, fdp.fdpcidx); + if (!info->runs) + return ret; + + size = struct_size(ruhs, ruhsd, S8_MAX - 1); + ruhs = kzalloc(size, GFP_KERNEL); + if (!ruhs) + return -ENOMEM; + + c.imr.opcode = nvme_cmd_io_mgmt_recv; + c.imr.nsid = cpu_to_le32(head->ns_id); + c.imr.mo = NVME_IO_MGMT_RECV_MO_RUHS; + c.imr.numd = cpu_to_le32(nvme_bytes_to_numd(size)); + ret = nvme_submit_sync_cmd(ns->queue, &c, ruhs, size); + if (ret) { + dev_warn(ctrl->device, "FDP io-mgmt status:0x%x\n", ret); + goto free; + } + + head->nr_plids = le16_to_cpu(ruhs->nruhsd); + if (!head->nr_plids) + goto free; + + head->plids = kcalloc(head->nr_plids, sizeof(*head->plids), + GFP_KERNEL); + if (!head->plids) { + dev_warn(ctrl->device, + "failed to allocate %u FDP placement IDs\n", + head->nr_plids); + head->nr_plids = 0; + ret = -ENOMEM; + goto free; + } + + for (i = 0; i < head->nr_plids; i++) + head->plids[i] = le16_to_cpu(ruhs->ruhsd[i].pid); +free: + kfree(ruhs); + return ret; } static int nvme_update_ns_info_block(struct nvme_ns *ns, struct nvme_ns_info *info) { + struct queue_limits lim; + struct nvme_id_ns_nvm *nvm = NULL; + struct nvme_zone_info zi = {}; struct nvme_id_ns *id; + unsigned int memflags; + sector_t capacity; unsigned lbaf; int ret; @@ -2012,235 +2348,218 @@ static int nvme_update_ns_info_block(struct nvme_ns *ns, if (ret) return ret; - blk_mq_freeze_queue(ns->disk->queue); + if (id->ncap == 0) { + /* namespace not allocated or attached */ + info->is_removed = true; + ret = -ENXIO; + goto out; + } lbaf = nvme_lbaf_index(id->flbas); - ns->lba_shift = id->lbaf[lbaf].ds; - nvme_set_queue_limits(ns->ctrl, ns->queue); - nvme_configure_metadata(ns, id); - nvme_set_chunk_sectors(ns, id); - nvme_update_disk_info(ns->disk, ns, id); + if (ns->ctrl->ctratt & NVME_CTRL_ATTR_ELBAS) { + ret = nvme_identify_ns_nvm(ns->ctrl, info->nsid, &nvm); + if (ret < 0) + goto out; + } - if (ns->head->ids.csi == NVME_CSI_ZNS) { - ret = nvme_update_zone_info(ns, lbaf); - if (ret) { - blk_mq_unfreeze_queue(ns->disk->queue); + if (IS_ENABLED(CONFIG_BLK_DEV_ZONED) && + ns->head->ids.csi == NVME_CSI_ZNS) { + ret = nvme_query_zone_info(ns, lbaf, &zi); + if (ret < 0) goto out; - } } + if (ns->ctrl->ctratt & NVME_CTRL_ATTR_FDPS) { + ret = nvme_query_fdp_info(ns, info); + if (ret < 0) + goto out; + } + + lim = queue_limits_start_update(ns->disk->queue); + + memflags = blk_mq_freeze_queue(ns->disk->queue); + ns->head->lba_shift = id->lbaf[lbaf].ds; + ns->head->nuse = le64_to_cpu(id->nuse); + capacity = nvme_lba_to_sect(ns->head, le64_to_cpu(id->nsze)); + nvme_set_ctrl_limits(ns->ctrl, &lim, false); + nvme_configure_metadata(ns->ctrl, ns->head, id, nvm, info); + nvme_set_chunk_sectors(ns, id, &lim); + if (!nvme_update_disk_info(ns, id, &lim)) + capacity = 0; + + nvme_config_discard(ns, &lim); + if (IS_ENABLED(CONFIG_BLK_DEV_ZONED) && + ns->head->ids.csi == NVME_CSI_ZNS) + nvme_update_zone_info(ns, &lim, &zi); + + if ((ns->ctrl->vwc & NVME_CTRL_VWC_PRESENT) && !info->no_vwc) + lim.features |= BLK_FEAT_WRITE_CACHE | BLK_FEAT_FUA; + else + lim.features &= ~(BLK_FEAT_WRITE_CACHE | BLK_FEAT_FUA); + + if (info->is_rotational) + lim.features |= BLK_FEAT_ROTATIONAL; + + /* + * Register a metadata profile for PI, or the plain non-integrity NVMe + * metadata masquerading as Type 0 if supported, otherwise reject block + * I/O to namespaces with metadata except when the namespace supports + * PI, as it can strip/insert in that case. + */ + if (!nvme_init_integrity(ns->head, &lim, info)) + capacity = 0; + + lim.max_write_streams = ns->head->nr_plids; + if (lim.max_write_streams) + lim.write_stream_granularity = min(info->runs, U32_MAX); + else + lim.write_stream_granularity = 0; + /* * Only set the DEAC bit if the device guarantees that reads from * deallocated data return zeroes. While the DEAC bit does not * require that, it must be a no-op if reads from deallocated data * do not return zeroes. */ - if ((id->dlfeat & 0x7) == 0x1 && (id->dlfeat & (1 << 3))) - ns->features |= NVME_NS_DEAC; + if ((id->dlfeat & 0x7) == 0x1 && (id->dlfeat & (1 << 3))) { + ns->head->features |= NVME_NS_DEAC; + lim.max_hw_wzeroes_unmap_sectors = lim.max_write_zeroes_sectors; + } + + ret = queue_limits_commit_update(ns->disk->queue, &lim); + if (ret) { + blk_mq_unfreeze_queue(ns->disk->queue, memflags); + goto out; + } + + set_capacity_and_notify(ns->disk, capacity); set_disk_ro(ns->disk, nvme_ns_is_readonly(ns, info)); set_bit(NVME_NS_READY, &ns->flags); - blk_mq_unfreeze_queue(ns->disk->queue); + blk_mq_unfreeze_queue(ns->disk->queue, memflags); if (blk_queue_is_zoned(ns->queue)) { - ret = nvme_revalidate_zones(ns); + ret = blk_revalidate_disk_zones(ns->disk); if (ret && !nvme_first_scan(ns->disk)) goto out; } - if (nvme_ns_head_multipath(ns->head)) { - blk_mq_freeze_queue(ns->head->disk->queue); - nvme_update_disk_info(ns->head->disk, ns, id); - set_disk_ro(ns->head->disk, nvme_ns_is_readonly(ns, info)); - nvme_mpath_revalidate_paths(ns); - blk_stack_limits(&ns->head->disk->queue->limits, - &ns->queue->limits, 0); - disk_update_readahead(ns->head->disk); - blk_mq_unfreeze_queue(ns->head->disk->queue); - } - ret = 0; out: - /* - * If probing fails due an unsupported feature, hide the block device, - * but still allow other access. - */ - if (ret == -ENODEV) { - ns->disk->flags |= GENHD_FL_HIDDEN; - set_bit(NVME_NS_READY, &ns->flags); - ret = 0; - } + kfree(nvm); kfree(id); return ret; } static int nvme_update_ns_info(struct nvme_ns *ns, struct nvme_ns_info *info) { + bool unsupported = false; + int ret; + switch (info->ids.csi) { case NVME_CSI_ZNS: if (!IS_ENABLED(CONFIG_BLK_DEV_ZONED)) { dev_info(ns->ctrl->device, "block device for nsid %u not supported without CONFIG_BLK_DEV_ZONED\n", info->nsid); - return nvme_update_ns_info_generic(ns, info); + ret = nvme_update_ns_info_generic(ns, info); + break; } - return nvme_update_ns_info_block(ns, info); + ret = nvme_update_ns_info_block(ns, info); + break; case NVME_CSI_NVM: - return nvme_update_ns_info_block(ns, info); + ret = nvme_update_ns_info_block(ns, info); + break; default: dev_info(ns->ctrl->device, "block device for nsid %u not supported (csi %u)\n", info->nsid, info->ids.csi); - return nvme_update_ns_info_generic(ns, info); + ret = nvme_update_ns_info_generic(ns, info); + break; } -} -static char nvme_pr_type(enum pr_type type) -{ - switch (type) { - case PR_WRITE_EXCLUSIVE: - return 1; - case PR_EXCLUSIVE_ACCESS: - return 2; - case PR_WRITE_EXCLUSIVE_REG_ONLY: - return 3; - case PR_EXCLUSIVE_ACCESS_REG_ONLY: - return 4; - case PR_WRITE_EXCLUSIVE_ALL_REGS: - return 5; - case PR_EXCLUSIVE_ACCESS_ALL_REGS: - return 6; - default: - return 0; + /* + * If probing fails due an unsupported feature, hide the block device, + * but still allow other access. + */ + if (ret == -ENODEV) { + ns->disk->flags |= GENHD_FL_HIDDEN; + set_bit(NVME_NS_READY, &ns->flags); + unsupported = true; + ret = 0; } -} -static int nvme_send_ns_head_pr_command(struct block_device *bdev, - struct nvme_command *c, u8 data[16]) -{ - struct nvme_ns_head *head = bdev->bd_disk->private_data; - int srcu_idx = srcu_read_lock(&head->srcu); - struct nvme_ns *ns = nvme_find_path(head); - int ret = -EWOULDBLOCK; + if (!ret && nvme_ns_head_multipath(ns->head)) { + struct queue_limits *ns_lim = &ns->disk->queue->limits; + struct queue_limits lim; + unsigned int memflags; - if (ns) { - c->common.nsid = cpu_to_le32(ns->head->ns_id); - ret = nvme_submit_sync_cmd(ns->queue, c, data, 16); - } - srcu_read_unlock(&head->srcu, srcu_idx); - return ret; -} - -static int nvme_send_ns_pr_command(struct nvme_ns *ns, struct nvme_command *c, - u8 data[16]) -{ - c->common.nsid = cpu_to_le32(ns->head->ns_id); - return nvme_submit_sync_cmd(ns->queue, c, data, 16); -} + lim = queue_limits_start_update(ns->head->disk->queue); + memflags = blk_mq_freeze_queue(ns->head->disk->queue); + /* + * queue_limits mixes values that are the hardware limitations + * for bio splitting with what is the device configuration. + * + * For NVMe the device configuration can change after e.g. a + * Format command, and we really want to pick up the new format + * value here. But we must still stack the queue limits to the + * least common denominator for multipathing to split the bios + * properly. + * + * To work around this, we explicitly set the device + * configuration to those that we just queried, but only stack + * the splitting limits in to make sure we still obey possibly + * lower limitations of other controllers. + */ + lim.logical_block_size = ns_lim->logical_block_size; + lim.physical_block_size = ns_lim->physical_block_size; + lim.io_min = ns_lim->io_min; + lim.io_opt = ns_lim->io_opt; + queue_limits_stack_bdev(&lim, ns->disk->part0, 0, + ns->head->disk->disk_name); + if (unsupported) + ns->head->disk->flags |= GENHD_FL_HIDDEN; + else + nvme_init_integrity(ns->head, &lim, info); + lim.max_write_streams = ns_lim->max_write_streams; + lim.write_stream_granularity = ns_lim->write_stream_granularity; + ret = queue_limits_commit_update(ns->head->disk->queue, &lim); -static int nvme_sc_to_pr_err(int nvme_sc) -{ - if (nvme_is_path_error(nvme_sc)) - return PR_STS_PATH_FAILED; + set_capacity_and_notify(ns->head->disk, get_capacity(ns->disk)); + set_disk_ro(ns->head->disk, nvme_ns_is_readonly(ns, info)); + nvme_mpath_revalidate_paths(ns); - switch (nvme_sc) { - case NVME_SC_SUCCESS: - return PR_STS_SUCCESS; - case NVME_SC_RESERVATION_CONFLICT: - return PR_STS_RESERVATION_CONFLICT; - case NVME_SC_ONCS_NOT_SUPPORTED: - return -EOPNOTSUPP; - case NVME_SC_BAD_ATTRIBUTES: - case NVME_SC_INVALID_OPCODE: - case NVME_SC_INVALID_FIELD: - case NVME_SC_INVALID_NS: - return -EINVAL; - default: - return PR_STS_IOERR; + blk_mq_unfreeze_queue(ns->head->disk->queue, memflags); } -} - -static int nvme_pr_command(struct block_device *bdev, u32 cdw10, - u64 key, u64 sa_key, u8 op) -{ - struct nvme_command c = { }; - u8 data[16] = { 0, }; - int ret; - - put_unaligned_le64(key, &data[0]); - put_unaligned_le64(sa_key, &data[8]); - - c.common.opcode = op; - c.common.cdw10 = cpu_to_le32(cdw10); - - if (IS_ENABLED(CONFIG_NVME_MULTIPATH) && - bdev->bd_disk->fops == &nvme_ns_head_ops) - ret = nvme_send_ns_head_pr_command(bdev, &c, data); - else - ret = nvme_send_ns_pr_command(bdev->bd_disk->private_data, &c, - data); - if (ret < 0) - return ret; - return nvme_sc_to_pr_err(ret); -} - -static int nvme_pr_register(struct block_device *bdev, u64 old, - u64 new, unsigned flags) -{ - u32 cdw10; - - if (flags & ~PR_FL_IGNORE_KEY) - return -EOPNOTSUPP; - - cdw10 = old ? 2 : 0; - cdw10 |= (flags & PR_FL_IGNORE_KEY) ? 1 << 3 : 0; - cdw10 |= (1 << 30) | (1 << 31); /* PTPL=1 */ - return nvme_pr_command(bdev, cdw10, old, new, nvme_cmd_resv_register); + return ret; } -static int nvme_pr_reserve(struct block_device *bdev, u64 key, - enum pr_type type, unsigned flags) +int nvme_ns_get_unique_id(struct nvme_ns *ns, u8 id[16], + enum blk_unique_id type) { - u32 cdw10; + struct nvme_ns_ids *ids = &ns->head->ids; - if (flags & ~PR_FL_IGNORE_KEY) - return -EOPNOTSUPP; - - cdw10 = nvme_pr_type(type) << 8; - cdw10 |= ((flags & PR_FL_IGNORE_KEY) ? 1 << 3 : 0); - return nvme_pr_command(bdev, cdw10, key, 0, nvme_cmd_resv_acquire); -} + if (type != BLK_UID_EUI64) + return -EINVAL; -static int nvme_pr_preempt(struct block_device *bdev, u64 old, u64 new, - enum pr_type type, bool abort) -{ - u32 cdw10 = nvme_pr_type(type) << 8 | (abort ? 2 : 1); + if (memchr_inv(ids->nguid, 0, sizeof(ids->nguid))) { + memcpy(id, &ids->nguid, sizeof(ids->nguid)); + return sizeof(ids->nguid); + } + if (memchr_inv(ids->eui64, 0, sizeof(ids->eui64))) { + memcpy(id, &ids->eui64, sizeof(ids->eui64)); + return sizeof(ids->eui64); + } - return nvme_pr_command(bdev, cdw10, old, new, nvme_cmd_resv_acquire); + return -EINVAL; } -static int nvme_pr_clear(struct block_device *bdev, u64 key) +static int nvme_get_unique_id(struct gendisk *disk, u8 id[16], + enum blk_unique_id type) { - u32 cdw10 = 1 | (key ? 0 : 1 << 3); - - return nvme_pr_command(bdev, cdw10, key, 0, nvme_cmd_resv_release); + return nvme_ns_get_unique_id(disk->private_data, id, type); } -static int nvme_pr_release(struct block_device *bdev, u64 key, enum pr_type type) -{ - u32 cdw10 = nvme_pr_type(type) << 8 | (key ? 0 : 1 << 3); - - return nvme_pr_command(bdev, cdw10, key, 0, nvme_cmd_resv_release); -} - -const struct pr_ops nvme_pr_ops = { - .pr_register = nvme_pr_register, - .pr_reserve = nvme_pr_reserve, - .pr_release = nvme_pr_release, - .pr_preempt = nvme_pr_preempt, - .pr_clear = nvme_pr_clear, -}; - #ifdef CONFIG_BLK_SED_OPAL static int nvme_sec_submit(void *data, u16 spsp, u8 secp, void *buffer, size_t len, bool send) @@ -2257,7 +2576,7 @@ static int nvme_sec_submit(void *data, u16 spsp, u8 secp, void *buffer, size_t l cmd.common.cdw11 = cpu_to_le32(len); return __nvme_submit_sync_cmd(ctrl->admin_q, &cmd, NULL, buffer, len, - NVME_QID_ANY, 1, 0); + NVME_QID_ANY, NVME_SUBMIT_AT_HEAD); } static void nvme_configure_opal(struct nvme_ctrl *ctrl, bool was_suspended) @@ -2280,22 +2599,22 @@ static void nvme_configure_opal(struct nvme_ctrl *ctrl, bool was_suspended) #ifdef CONFIG_BLK_DEV_ZONED static int nvme_report_zones(struct gendisk *disk, sector_t sector, - unsigned int nr_zones, report_zones_cb cb, void *data) + unsigned int nr_zones, struct blk_report_zones_args *args) { - return nvme_ns_report_zones(disk->private_data, sector, nr_zones, cb, - data); + return nvme_ns_report_zones(disk->private_data, sector, nr_zones, args); } #else #define nvme_report_zones NULL #endif /* CONFIG_BLK_DEV_ZONED */ -static const struct block_device_operations nvme_bdev_ops = { +const struct block_device_operations nvme_bdev_ops = { .owner = THIS_MODULE, .ioctl = nvme_ioctl, .compat_ioctl = blkdev_compat_ptr_ioctl, .open = nvme_open, .release = nvme_release, .getgeo = nvme_getgeo, + .get_unique_id = nvme_get_unique_id, .report_zones = nvme_report_zones, .pr_ops = &nvme_pr_ops, }; @@ -2378,25 +2697,13 @@ int nvme_enable_ctrl(struct nvme_ctrl *ctrl) else ctrl->ctrl_config = NVME_CC_CSS_NVM; - if (ctrl->cap & NVME_CAP_CRMS_CRWMS) { - u32 crto; - - ret = ctrl->ops->reg_read32(ctrl, NVME_REG_CRTO, &crto); - if (ret) { - dev_err(ctrl->device, "Reading CRTO failed (%d)\n", - ret); - return ret; - } - - if (ctrl->cap & NVME_CAP_CRMS_CRIMS) { - ctrl->ctrl_config |= NVME_CC_CRIME; - timeout = NVME_CRTO_CRIMT(crto); - } else { - timeout = NVME_CRTO_CRWMT(crto); - } - } else { - timeout = NVME_CAP_TIMEOUT(ctrl->cap); - } + /* + * Setting CRIME results in CSTS.RDY before the media is ready. This + * makes it possible for media related commands to return the error + * NVME_SC_ADMIN_COMMAND_MEDIA_NOT_READY. Until the driver is + * restructured to handle retries, disable CC.CRIME. + */ + ctrl->ctrl_config &= ~NVME_CC_CRIME; ctrl->ctrl_config |= (NVME_CTRL_PAGE_SHIFT - 12) << NVME_CC_MPS_SHIFT; ctrl->ctrl_config |= NVME_CC_AMS_RR | NVME_CC_SHN_NONE; @@ -2405,11 +2712,36 @@ int nvme_enable_ctrl(struct nvme_ctrl *ctrl) if (ret) return ret; - /* Flush write to device (required if transport is PCI) */ - ret = ctrl->ops->reg_read32(ctrl, NVME_REG_CC, &ctrl->ctrl_config); + /* CAP value may change after initial CC write */ + ret = ctrl->ops->reg_read64(ctrl, NVME_REG_CAP, &ctrl->cap); if (ret) return ret; + timeout = NVME_CAP_TIMEOUT(ctrl->cap); + if (ctrl->cap & NVME_CAP_CRMS_CRWMS) { + u32 crto, ready_timeout; + + ret = ctrl->ops->reg_read32(ctrl, NVME_REG_CRTO, &crto); + if (ret) { + dev_err(ctrl->device, "Reading CRTO failed (%d)\n", + ret); + return ret; + } + + /* + * CRTO should always be greater or equal to CAP.TO, but some + * devices are known to get this wrong. Use the larger of the + * two values. + */ + ready_timeout = NVME_CRTO_CRWMT(crto); + + if (ready_timeout < timeout) + dev_warn_once(ctrl->device, "bad crto:%x cap:%llx\n", + crto, ctrl->cap); + else + timeout = ready_timeout; + } + ctrl->ctrl_config |= NVME_CC_ENABLE; ret = ctrl->ops->reg_write32(ctrl, NVME_REG_CC, ctrl->ctrl_config); if (ret) @@ -2639,7 +2971,7 @@ static void nvme_set_latency_tolerance(struct device *dev, s32 val) if (ctrl->ps_max_latency_us != latency) { ctrl->ps_max_latency_us = latency; - if (ctrl->state == NVME_CTRL_LIVE) + if (nvme_ctrl_state(ctrl) == NVME_CTRL_LIVE) nvme_configure_apst(ctrl); } } @@ -2824,78 +3156,19 @@ static struct nvme_subsystem *__nvme_find_get_subsystem(const char *subsysnqn) return NULL; } -#define SUBSYS_ATTR_RO(_name, _mode, _show) \ - struct device_attribute subsys_attr_##_name = \ - __ATTR(_name, _mode, _show, NULL) - -static ssize_t nvme_subsys_show_nqn(struct device *dev, - struct device_attribute *attr, - char *buf) +static inline bool nvme_discovery_ctrl(struct nvme_ctrl *ctrl) { - struct nvme_subsystem *subsys = - container_of(dev, struct nvme_subsystem, dev); - - return sysfs_emit(buf, "%s\n", subsys->subnqn); + return ctrl->opts && ctrl->opts->discovery_nqn; } -static SUBSYS_ATTR_RO(subsysnqn, S_IRUGO, nvme_subsys_show_nqn); -static ssize_t nvme_subsys_show_type(struct device *dev, - struct device_attribute *attr, - char *buf) +static inline bool nvme_admin_ctrl(struct nvme_ctrl *ctrl) { - struct nvme_subsystem *subsys = - container_of(dev, struct nvme_subsystem, dev); - - switch (subsys->subtype) { - case NVME_NQN_DISC: - return sysfs_emit(buf, "discovery\n"); - case NVME_NQN_NVME: - return sysfs_emit(buf, "nvm\n"); - default: - return sysfs_emit(buf, "reserved\n"); - } -} -static SUBSYS_ATTR_RO(subsystype, S_IRUGO, nvme_subsys_show_type); - -#define nvme_subsys_show_str_function(field) \ -static ssize_t subsys_##field##_show(struct device *dev, \ - struct device_attribute *attr, char *buf) \ -{ \ - struct nvme_subsystem *subsys = \ - container_of(dev, struct nvme_subsystem, dev); \ - return sysfs_emit(buf, "%.*s\n", \ - (int)sizeof(subsys->field), subsys->field); \ -} \ -static SUBSYS_ATTR_RO(field, S_IRUGO, subsys_##field##_show); - -nvme_subsys_show_str_function(model); -nvme_subsys_show_str_function(serial); -nvme_subsys_show_str_function(firmware_rev); - -static struct attribute *nvme_subsys_attrs[] = { - &subsys_attr_model.attr, - &subsys_attr_serial.attr, - &subsys_attr_firmware_rev.attr, - &subsys_attr_subsysnqn.attr, - &subsys_attr_subsystype.attr, -#ifdef CONFIG_NVME_MULTIPATH - &subsys_attr_iopolicy.attr, -#endif - NULL, -}; - -static const struct attribute_group nvme_subsys_attrs_group = { - .attrs = nvme_subsys_attrs, -}; - -static const struct attribute_group *nvme_subsys_attrs_groups[] = { - &nvme_subsys_attrs_group, - NULL, -}; + return ctrl->cntrltype == NVME_CTRL_ADMIN; +} -static inline bool nvme_discovery_ctrl(struct nvme_ctrl *ctrl) +static inline bool nvme_is_io_ctrl(struct nvme_ctrl *ctrl) { - return ctrl->opts && ctrl->opts->discovery_nqn; + return !nvme_discovery_ctrl(ctrl) && !nvme_admin_ctrl(ctrl); } static bool nvme_validate_cntlid(struct nvme_subsystem *subsys, @@ -2948,6 +3221,7 @@ static int nvme_init_subsystem(struct nvme_ctrl *ctrl, struct nvme_id_ctrl *id) memcpy(subsys->model, id->mn, sizeof(subsys->model)); subsys->vendor_id = le16_to_cpu(id->vid); subsys->cmic = id->cmic; + subsys->awupf = le16_to_cpu(id->awupf); /* Versions prior to 1.4 don't necessarily report a valid type */ if (id->cntrltype == NVME_CTRL_DISC || @@ -2963,10 +3237,9 @@ static int nvme_init_subsystem(struct nvme_ctrl *ctrl, struct nvme_id_ctrl *id) kfree(subsys); return -EINVAL; } - subsys->awupf = le16_to_cpu(id->awupf); nvme_mpath_default_iopolicy(subsys); - subsys->dev.class = nvme_subsys_class; + subsys->dev.class = &nvme_subsys_class; subsys->dev.release = nvme_release_subsystem; subsys->dev.groups = nvme_subsys_attrs_groups; dev_set_name(&subsys->dev, "nvme-subsys%d", ctrl->instance); @@ -3016,8 +3289,8 @@ out_unlock: return ret; } -int nvme_get_log(struct nvme_ctrl *ctrl, u32 nsid, u8 log_page, u8 lsp, u8 csi, - void *log, size_t size, u64 offset) +static int nvme_get_log_lsi(struct nvme_ctrl *ctrl, u32 nsid, u8 log_page, + u8 lsp, u8 csi, void *log, size_t size, u64 offset, u16 lsi) { struct nvme_command c = { }; u32 dwlen = nvme_bytes_to_numd(size); @@ -3031,14 +3304,22 @@ int nvme_get_log(struct nvme_ctrl *ctrl, u32 nsid, u8 log_page, u8 lsp, u8 csi, c.get_log_page.lpol = cpu_to_le32(lower_32_bits(offset)); c.get_log_page.lpou = cpu_to_le32(upper_32_bits(offset)); c.get_log_page.csi = csi; + c.get_log_page.lsi = cpu_to_le16(lsi); return nvme_submit_sync_cmd(ctrl->admin_q, &c, log, size); } +int nvme_get_log(struct nvme_ctrl *ctrl, u32 nsid, u8 log_page, u8 lsp, u8 csi, + void *log, size_t size, u64 offset) +{ + return nvme_get_log_lsi(ctrl, nsid, log_page, lsp, csi, log, size, + offset, 0); +} + static int nvme_get_effects_log(struct nvme_ctrl *ctrl, u8 csi, struct nvme_effects_log **log) { - struct nvme_effects_log *cel = xa_load(&ctrl->cels, csi); + struct nvme_effects_log *old, *cel = xa_load(&ctrl->cels, csi); int ret; if (cel) @@ -3055,7 +3336,11 @@ static int nvme_get_effects_log(struct nvme_ctrl *ctrl, u8 csi, return ret; } - xa_store(&ctrl->cels, csi, cel, GFP_KERNEL); + old = xa_store(&ctrl->cels, csi, cel, GFP_KERNEL); + if (xa_is_err(old)) { + kfree(cel); + return xa_err(old); + } out: *log = cel; return 0; @@ -3076,14 +3361,6 @@ static int nvme_init_non_mdts_limits(struct nvme_ctrl *ctrl) struct nvme_id_ctrl_nvm *id; int ret; - if (ctrl->oncs & NVME_CTRL_ONCS_DSM) { - ctrl->max_discard_sectors = UINT_MAX; - ctrl->max_discard_segments = NVME_DSM_MAX_RANGES; - } else { - ctrl->max_discard_sectors = 0; - ctrl->max_discard_segments = 0; - } - /* * Even though NVMe spec explicitly states that MDTS is not applicable * to the write-zeroes, we are cautious and limit the size to the @@ -3096,7 +3373,9 @@ static int nvme_init_non_mdts_limits(struct nvme_ctrl *ctrl) else ctrl->max_zeroes_sectors = 0; - if (nvme_ctrl_limited_cns(ctrl)) + if (!nvme_is_io_ctrl(ctrl) || + !nvme_id_cns_ok(ctrl, NVME_ID_CNS_CS_CTRL) || + test_bit(NVME_CTRL_SKIP_ID_CNS_CS, &ctrl->flags)) return 0; id = kzalloc(sizeof(*id), GFP_KERNEL); @@ -3111,19 +3390,137 @@ static int nvme_init_non_mdts_limits(struct nvme_ctrl *ctrl) if (ret) goto free_data; - if (id->dmrl) - ctrl->max_discard_segments = id->dmrl; + ctrl->dmrl = id->dmrl; ctrl->dmrsl = le32_to_cpu(id->dmrsl); if (id->wzsl) ctrl->max_zeroes_sectors = nvme_mps_to_sectors(ctrl, id->wzsl); free_data: + if (ret > 0) + set_bit(NVME_CTRL_SKIP_ID_CNS_CS, &ctrl->flags); kfree(id); return ret; } +static int nvme_init_effects_log(struct nvme_ctrl *ctrl, + u8 csi, struct nvme_effects_log **log) +{ + struct nvme_effects_log *effects, *old; + + effects = kzalloc(sizeof(*effects), GFP_KERNEL); + if (!effects) + return -ENOMEM; + + old = xa_store(&ctrl->cels, csi, effects, GFP_KERNEL); + if (xa_is_err(old)) { + kfree(effects); + return xa_err(old); + } + + *log = effects; + return 0; +} + +static void nvme_init_known_nvm_effects(struct nvme_ctrl *ctrl) +{ + struct nvme_effects_log *log = ctrl->effects; + + log->acs[nvme_admin_format_nvm] |= cpu_to_le32(NVME_CMD_EFFECTS_LBCC | + NVME_CMD_EFFECTS_NCC | + NVME_CMD_EFFECTS_CSE_MASK); + log->acs[nvme_admin_sanitize_nvm] |= cpu_to_le32(NVME_CMD_EFFECTS_LBCC | + NVME_CMD_EFFECTS_CSE_MASK); + + /* + * The spec says the result of a security receive command depends on + * the previous security send command. As such, many vendors log this + * command as one to submitted only when no other commands to the same + * namespace are outstanding. The intention is to tell the host to + * prevent mixing security send and receive. + * + * This driver can only enforce such exclusive access against IO + * queues, though. We are not readily able to enforce such a rule for + * two commands to the admin queue, which is the only queue that + * matters for this command. + * + * Rather than blindly freezing the IO queues for this effect that + * doesn't even apply to IO, mask it off. + */ + log->acs[nvme_admin_security_recv] &= cpu_to_le32(~NVME_CMD_EFFECTS_CSE_MASK); + + log->iocs[nvme_cmd_write] |= cpu_to_le32(NVME_CMD_EFFECTS_LBCC); + log->iocs[nvme_cmd_write_zeroes] |= cpu_to_le32(NVME_CMD_EFFECTS_LBCC); + log->iocs[nvme_cmd_write_uncor] |= cpu_to_le32(NVME_CMD_EFFECTS_LBCC); +} + +static int nvme_init_effects(struct nvme_ctrl *ctrl, struct nvme_id_ctrl *id) +{ + int ret = 0; + + if (ctrl->effects) + return 0; + + if (id->lpa & NVME_CTRL_LPA_CMD_EFFECTS_LOG) { + ret = nvme_get_effects_log(ctrl, NVME_CSI_NVM, &ctrl->effects); + if (ret < 0) + return ret; + } + + if (!ctrl->effects) { + ret = nvme_init_effects_log(ctrl, NVME_CSI_NVM, &ctrl->effects); + if (ret < 0) + return ret; + } + + nvme_init_known_nvm_effects(ctrl); + return 0; +} + +static int nvme_check_ctrl_fabric_info(struct nvme_ctrl *ctrl, struct nvme_id_ctrl *id) +{ + /* + * In fabrics we need to verify the cntlid matches the + * admin connect + */ + if (ctrl->cntlid != le16_to_cpu(id->cntlid)) { + dev_err(ctrl->device, + "Mismatching cntlid: Connect %u vs Identify %u, rejecting\n", + ctrl->cntlid, le16_to_cpu(id->cntlid)); + return -EINVAL; + } + + if (!nvme_discovery_ctrl(ctrl) && !ctrl->kas) { + dev_err(ctrl->device, + "keep-alive support is mandatory for fabrics\n"); + return -EINVAL; + } + + if (nvme_is_io_ctrl(ctrl) && ctrl->ioccsz < 4) { + dev_err(ctrl->device, + "I/O queue command capsule supported size %d < 4\n", + ctrl->ioccsz); + return -EINVAL; + } + + if (nvme_is_io_ctrl(ctrl) && ctrl->iorcsz < 1) { + dev_err(ctrl->device, + "I/O queue response capsule supported size %d < 1\n", + ctrl->iorcsz); + return -EINVAL; + } + + if (!ctrl->maxcmd) { + dev_warn(ctrl->device, + "Firmware bug: maximum outstanding commands is 0\n"); + ctrl->maxcmd = ctrl->sqsize + 1; + } + + return 0; +} + static int nvme_init_identify(struct nvme_ctrl *ctrl) { + struct queue_limits lim; struct nvme_id_ctrl *id; u32 max_hw_sectors; bool prev_apst_enabled; @@ -3135,12 +3532,6 @@ static int nvme_init_identify(struct nvme_ctrl *ctrl) return -EIO; } - if (id->lpa & NVME_CTRL_LPA_CMD_EFFECTS_LOG) { - ret = nvme_get_effects_log(ctrl, NVME_CSI_NVM, &ctrl->effects); - if (ret < 0) - goto out_free; - } - if (!(ctrl->ops->flags & NVME_F_FABRICS)) ctrl->cntlid = le16_to_cpu(id->cntlid); @@ -3163,6 +3554,10 @@ static int nvme_init_identify(struct nvme_ctrl *ctrl) ret = nvme_init_subsystem(ctrl, id); if (ret) goto out_free; + + ret = nvme_init_effects(ctrl, id); + if (ret) + goto out_free; } memcpy(ctrl->subsys->firmware_rev, id->fr, sizeof(ctrl->subsys->firmware_rev)); @@ -3192,7 +3587,12 @@ static int nvme_init_identify(struct nvme_ctrl *ctrl) ctrl->max_hw_sectors = min_not_zero(ctrl->max_hw_sectors, max_hw_sectors); - nvme_set_queue_limits(ctrl, ctrl->admin_q); + lim = queue_limits_start_update(ctrl->admin_q); + nvme_set_ctrl_limits(ctrl, &lim, true); + ret = queue_limits_commit_update(ctrl->admin_q, &lim); + if (ret) + goto out_free; + ctrl->sgls = le32_to_cpu(id->sgls); ctrl->kas = le16_to_cpu(id->kas); ctrl->max_namespaces = le32_to_cpu(id->mnan); @@ -3210,7 +3610,7 @@ static int nvme_init_identify(struct nvme_ctrl *ctrl) if (ctrl->shutdown_timeout != shutdown_timeout) dev_info(ctrl->device, - "Shutdown timeout set to %u seconds\n", + "D3 entry latency set to %u seconds\n", ctrl->shutdown_timeout); } else ctrl->shutdown_timeout = shutdown_timeout; @@ -3236,25 +3636,9 @@ static int nvme_init_identify(struct nvme_ctrl *ctrl) ctrl->iorcsz = le32_to_cpu(id->iorcsz); ctrl->maxcmd = le16_to_cpu(id->maxcmd); - /* - * In fabrics we need to verify the cntlid matches the - * admin connect - */ - if (ctrl->cntlid != le16_to_cpu(id->cntlid)) { - dev_err(ctrl->device, - "Mismatching cntlid: Connect %u vs Identify " - "%u, rejecting\n", - ctrl->cntlid, le16_to_cpu(id->cntlid)); - ret = -EINVAL; - goto out_free; - } - - if (!nvme_discovery_ctrl(ctrl) && !ctrl->kas) { - dev_err(ctrl->device, - "keep-alive support is mandatory for fabrics\n"); - ret = -EINVAL; + ret = nvme_check_ctrl_fabric_info(ctrl, id); + if (ret) goto out_free; - } } else { ctrl->hmpre = le32_to_cpu(id->hmpre); ctrl->hmmin = le32_to_cpu(id->hmmin); @@ -3270,7 +3654,6 @@ static int nvme_init_identify(struct nvme_ctrl *ctrl) dev_pm_qos_expose_latency_tolerance(ctrl->device); else if (!ctrl->apst_enabled && prev_apst_enabled) dev_pm_qos_hide_latency_tolerance(ctrl->device); - out_free: kfree(id); return ret; @@ -3300,6 +3683,17 @@ int nvme_init_ctrl_finish(struct nvme_ctrl *ctrl, bool was_suspended) if (ret) return ret; + if (nvme_admin_ctrl(ctrl)) { + /* + * An admin controller has one admin queue, but no I/O queues. + * Override queue_count so it only creates an admin queue. + */ + dev_dbg(ctrl->device, + "Subsystem %s is an administrative controller", + ctrl->subsys->subnqn); + ctrl->queue_count = 1; + } + ret = nvme_configure_apst(ctrl); if (ret < 0) return ret; @@ -3324,8 +3718,11 @@ int nvme_init_ctrl_finish(struct nvme_ctrl *ctrl, bool was_suspended) return ret; } + clear_bit(NVME_CTRL_DIRTY_CAPABILITY, &ctrl->flags); ctrl->identified = true; + nvme_start_keep_alive(ctrl); + return 0; } EXPORT_SYMBOL_GPL(nvme_init_ctrl_finish); @@ -3335,7 +3732,7 @@ static int nvme_dev_open(struct inode *inode, struct file *file) struct nvme_ctrl *ctrl = container_of(inode->i_cdev, struct nvme_ctrl, cdev); - switch (ctrl->state) { + switch (nvme_ctrl_state(ctrl)) { case NVME_CTRL_LIVE: break; default: @@ -3371,583 +3768,6 @@ static const struct file_operations nvme_dev_fops = { .uring_cmd = nvme_dev_uring_cmd, }; -static ssize_t nvme_sysfs_reset(struct device *dev, - struct device_attribute *attr, const char *buf, - size_t count) -{ - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); - int ret; - - ret = nvme_reset_ctrl_sync(ctrl); - if (ret < 0) - return ret; - return count; -} -static DEVICE_ATTR(reset_controller, S_IWUSR, NULL, nvme_sysfs_reset); - -static ssize_t nvme_sysfs_rescan(struct device *dev, - struct device_attribute *attr, const char *buf, - size_t count) -{ - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); - - nvme_queue_scan(ctrl); - return count; -} -static DEVICE_ATTR(rescan_controller, S_IWUSR, NULL, nvme_sysfs_rescan); - -static inline struct nvme_ns_head *dev_to_ns_head(struct device *dev) -{ - struct gendisk *disk = dev_to_disk(dev); - - if (disk->fops == &nvme_bdev_ops) - return nvme_get_ns_from_dev(dev)->head; - else - return disk->private_data; -} - -static ssize_t wwid_show(struct device *dev, struct device_attribute *attr, - char *buf) -{ - struct nvme_ns_head *head = dev_to_ns_head(dev); - struct nvme_ns_ids *ids = &head->ids; - struct nvme_subsystem *subsys = head->subsys; - int serial_len = sizeof(subsys->serial); - int model_len = sizeof(subsys->model); - - if (!uuid_is_null(&ids->uuid)) - return sysfs_emit(buf, "uuid.%pU\n", &ids->uuid); - - if (memchr_inv(ids->nguid, 0, sizeof(ids->nguid))) - return sysfs_emit(buf, "eui.%16phN\n", ids->nguid); - - if (memchr_inv(ids->eui64, 0, sizeof(ids->eui64))) - return sysfs_emit(buf, "eui.%8phN\n", ids->eui64); - - while (serial_len > 0 && (subsys->serial[serial_len - 1] == ' ' || - subsys->serial[serial_len - 1] == '\0')) - serial_len--; - while (model_len > 0 && (subsys->model[model_len - 1] == ' ' || - subsys->model[model_len - 1] == '\0')) - model_len--; - - return sysfs_emit(buf, "nvme.%04x-%*phN-%*phN-%08x\n", subsys->vendor_id, - serial_len, subsys->serial, model_len, subsys->model, - head->ns_id); -} -static DEVICE_ATTR_RO(wwid); - -static ssize_t nguid_show(struct device *dev, struct device_attribute *attr, - char *buf) -{ - return sysfs_emit(buf, "%pU\n", dev_to_ns_head(dev)->ids.nguid); -} -static DEVICE_ATTR_RO(nguid); - -static ssize_t uuid_show(struct device *dev, struct device_attribute *attr, - char *buf) -{ - struct nvme_ns_ids *ids = &dev_to_ns_head(dev)->ids; - - /* For backward compatibility expose the NGUID to userspace if - * we have no UUID set - */ - if (uuid_is_null(&ids->uuid)) { - dev_warn_ratelimited(dev, - "No UUID available providing old NGUID\n"); - return sysfs_emit(buf, "%pU\n", ids->nguid); - } - return sysfs_emit(buf, "%pU\n", &ids->uuid); -} -static DEVICE_ATTR_RO(uuid); - -static ssize_t eui_show(struct device *dev, struct device_attribute *attr, - char *buf) -{ - return sysfs_emit(buf, "%8ph\n", dev_to_ns_head(dev)->ids.eui64); -} -static DEVICE_ATTR_RO(eui); - -static ssize_t nsid_show(struct device *dev, struct device_attribute *attr, - char *buf) -{ - return sysfs_emit(buf, "%d\n", dev_to_ns_head(dev)->ns_id); -} -static DEVICE_ATTR_RO(nsid); - -static struct attribute *nvme_ns_id_attrs[] = { - &dev_attr_wwid.attr, - &dev_attr_uuid.attr, - &dev_attr_nguid.attr, - &dev_attr_eui.attr, - &dev_attr_nsid.attr, -#ifdef CONFIG_NVME_MULTIPATH - &dev_attr_ana_grpid.attr, - &dev_attr_ana_state.attr, -#endif - NULL, -}; - -static umode_t nvme_ns_id_attrs_are_visible(struct kobject *kobj, - struct attribute *a, int n) -{ - struct device *dev = container_of(kobj, struct device, kobj); - struct nvme_ns_ids *ids = &dev_to_ns_head(dev)->ids; - - if (a == &dev_attr_uuid.attr) { - if (uuid_is_null(&ids->uuid) && - !memchr_inv(ids->nguid, 0, sizeof(ids->nguid))) - return 0; - } - if (a == &dev_attr_nguid.attr) { - if (!memchr_inv(ids->nguid, 0, sizeof(ids->nguid))) - return 0; - } - if (a == &dev_attr_eui.attr) { - if (!memchr_inv(ids->eui64, 0, sizeof(ids->eui64))) - return 0; - } -#ifdef CONFIG_NVME_MULTIPATH - if (a == &dev_attr_ana_grpid.attr || a == &dev_attr_ana_state.attr) { - if (dev_to_disk(dev)->fops != &nvme_bdev_ops) /* per-path attr */ - return 0; - if (!nvme_ctrl_use_ana(nvme_get_ns_from_dev(dev)->ctrl)) - return 0; - } -#endif - return a->mode; -} - -static const struct attribute_group nvme_ns_id_attr_group = { - .attrs = nvme_ns_id_attrs, - .is_visible = nvme_ns_id_attrs_are_visible, -}; - -const struct attribute_group *nvme_ns_id_attr_groups[] = { - &nvme_ns_id_attr_group, - NULL, -}; - -#define nvme_show_str_function(field) \ -static ssize_t field##_show(struct device *dev, \ - struct device_attribute *attr, char *buf) \ -{ \ - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); \ - return sysfs_emit(buf, "%.*s\n", \ - (int)sizeof(ctrl->subsys->field), ctrl->subsys->field); \ -} \ -static DEVICE_ATTR(field, S_IRUGO, field##_show, NULL); - -nvme_show_str_function(model); -nvme_show_str_function(serial); -nvme_show_str_function(firmware_rev); - -#define nvme_show_int_function(field) \ -static ssize_t field##_show(struct device *dev, \ - struct device_attribute *attr, char *buf) \ -{ \ - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); \ - return sysfs_emit(buf, "%d\n", ctrl->field); \ -} \ -static DEVICE_ATTR(field, S_IRUGO, field##_show, NULL); - -nvme_show_int_function(cntlid); -nvme_show_int_function(numa_node); -nvme_show_int_function(queue_count); -nvme_show_int_function(sqsize); -nvme_show_int_function(kato); - -static ssize_t nvme_sysfs_delete(struct device *dev, - struct device_attribute *attr, const char *buf, - size_t count) -{ - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); - - if (device_remove_file_self(dev, attr)) - nvme_delete_ctrl_sync(ctrl); - return count; -} -static DEVICE_ATTR(delete_controller, S_IWUSR, NULL, nvme_sysfs_delete); - -static ssize_t nvme_sysfs_show_transport(struct device *dev, - struct device_attribute *attr, - char *buf) -{ - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); - - return sysfs_emit(buf, "%s\n", ctrl->ops->name); -} -static DEVICE_ATTR(transport, S_IRUGO, nvme_sysfs_show_transport, NULL); - -static ssize_t nvme_sysfs_show_state(struct device *dev, - struct device_attribute *attr, - char *buf) -{ - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); - static const char *const state_name[] = { - [NVME_CTRL_NEW] = "new", - [NVME_CTRL_LIVE] = "live", - [NVME_CTRL_RESETTING] = "resetting", - [NVME_CTRL_CONNECTING] = "connecting", - [NVME_CTRL_DELETING] = "deleting", - [NVME_CTRL_DELETING_NOIO]= "deleting (no IO)", - [NVME_CTRL_DEAD] = "dead", - }; - - if ((unsigned)ctrl->state < ARRAY_SIZE(state_name) && - state_name[ctrl->state]) - return sysfs_emit(buf, "%s\n", state_name[ctrl->state]); - - return sysfs_emit(buf, "unknown state\n"); -} - -static DEVICE_ATTR(state, S_IRUGO, nvme_sysfs_show_state, NULL); - -static ssize_t nvme_sysfs_show_subsysnqn(struct device *dev, - struct device_attribute *attr, - char *buf) -{ - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); - - return sysfs_emit(buf, "%s\n", ctrl->subsys->subnqn); -} -static DEVICE_ATTR(subsysnqn, S_IRUGO, nvme_sysfs_show_subsysnqn, NULL); - -static ssize_t nvme_sysfs_show_hostnqn(struct device *dev, - struct device_attribute *attr, - char *buf) -{ - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); - - return sysfs_emit(buf, "%s\n", ctrl->opts->host->nqn); -} -static DEVICE_ATTR(hostnqn, S_IRUGO, nvme_sysfs_show_hostnqn, NULL); - -static ssize_t nvme_sysfs_show_hostid(struct device *dev, - struct device_attribute *attr, - char *buf) -{ - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); - - return sysfs_emit(buf, "%pU\n", &ctrl->opts->host->id); -} -static DEVICE_ATTR(hostid, S_IRUGO, nvme_sysfs_show_hostid, NULL); - -static ssize_t nvme_sysfs_show_address(struct device *dev, - struct device_attribute *attr, - char *buf) -{ - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); - - return ctrl->ops->get_address(ctrl, buf, PAGE_SIZE); -} -static DEVICE_ATTR(address, S_IRUGO, nvme_sysfs_show_address, NULL); - -static ssize_t nvme_ctrl_loss_tmo_show(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); - struct nvmf_ctrl_options *opts = ctrl->opts; - - if (ctrl->opts->max_reconnects == -1) - return sysfs_emit(buf, "off\n"); - return sysfs_emit(buf, "%d\n", - opts->max_reconnects * opts->reconnect_delay); -} - -static ssize_t nvme_ctrl_loss_tmo_store(struct device *dev, - struct device_attribute *attr, const char *buf, size_t count) -{ - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); - struct nvmf_ctrl_options *opts = ctrl->opts; - int ctrl_loss_tmo, err; - - err = kstrtoint(buf, 10, &ctrl_loss_tmo); - if (err) - return -EINVAL; - - if (ctrl_loss_tmo < 0) - opts->max_reconnects = -1; - else - opts->max_reconnects = DIV_ROUND_UP(ctrl_loss_tmo, - opts->reconnect_delay); - return count; -} -static DEVICE_ATTR(ctrl_loss_tmo, S_IRUGO | S_IWUSR, - nvme_ctrl_loss_tmo_show, nvme_ctrl_loss_tmo_store); - -static ssize_t nvme_ctrl_reconnect_delay_show(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); - - if (ctrl->opts->reconnect_delay == -1) - return sysfs_emit(buf, "off\n"); - return sysfs_emit(buf, "%d\n", ctrl->opts->reconnect_delay); -} - -static ssize_t nvme_ctrl_reconnect_delay_store(struct device *dev, - struct device_attribute *attr, const char *buf, size_t count) -{ - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); - unsigned int v; - int err; - - err = kstrtou32(buf, 10, &v); - if (err) - return err; - - ctrl->opts->reconnect_delay = v; - return count; -} -static DEVICE_ATTR(reconnect_delay, S_IRUGO | S_IWUSR, - nvme_ctrl_reconnect_delay_show, nvme_ctrl_reconnect_delay_store); - -static ssize_t nvme_ctrl_fast_io_fail_tmo_show(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); - - if (ctrl->opts->fast_io_fail_tmo == -1) - return sysfs_emit(buf, "off\n"); - return sysfs_emit(buf, "%d\n", ctrl->opts->fast_io_fail_tmo); -} - -static ssize_t nvme_ctrl_fast_io_fail_tmo_store(struct device *dev, - struct device_attribute *attr, const char *buf, size_t count) -{ - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); - struct nvmf_ctrl_options *opts = ctrl->opts; - int fast_io_fail_tmo, err; - - err = kstrtoint(buf, 10, &fast_io_fail_tmo); - if (err) - return -EINVAL; - - if (fast_io_fail_tmo < 0) - opts->fast_io_fail_tmo = -1; - else - opts->fast_io_fail_tmo = fast_io_fail_tmo; - return count; -} -static DEVICE_ATTR(fast_io_fail_tmo, S_IRUGO | S_IWUSR, - nvme_ctrl_fast_io_fail_tmo_show, nvme_ctrl_fast_io_fail_tmo_store); - -static ssize_t cntrltype_show(struct device *dev, - struct device_attribute *attr, char *buf) -{ - static const char * const type[] = { - [NVME_CTRL_IO] = "io\n", - [NVME_CTRL_DISC] = "discovery\n", - [NVME_CTRL_ADMIN] = "admin\n", - }; - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); - - if (ctrl->cntrltype > NVME_CTRL_ADMIN || !type[ctrl->cntrltype]) - return sysfs_emit(buf, "reserved\n"); - - return sysfs_emit(buf, type[ctrl->cntrltype]); -} -static DEVICE_ATTR_RO(cntrltype); - -static ssize_t dctype_show(struct device *dev, - struct device_attribute *attr, char *buf) -{ - static const char * const type[] = { - [NVME_DCTYPE_NOT_REPORTED] = "none\n", - [NVME_DCTYPE_DDC] = "ddc\n", - [NVME_DCTYPE_CDC] = "cdc\n", - }; - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); - - if (ctrl->dctype > NVME_DCTYPE_CDC || !type[ctrl->dctype]) - return sysfs_emit(buf, "reserved\n"); - - return sysfs_emit(buf, type[ctrl->dctype]); -} -static DEVICE_ATTR_RO(dctype); - -#ifdef CONFIG_NVME_AUTH -static ssize_t nvme_ctrl_dhchap_secret_show(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); - struct nvmf_ctrl_options *opts = ctrl->opts; - - if (!opts->dhchap_secret) - return sysfs_emit(buf, "none\n"); - return sysfs_emit(buf, "%s\n", opts->dhchap_secret); -} - -static ssize_t nvme_ctrl_dhchap_secret_store(struct device *dev, - struct device_attribute *attr, const char *buf, size_t count) -{ - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); - struct nvmf_ctrl_options *opts = ctrl->opts; - char *dhchap_secret; - - if (!ctrl->opts->dhchap_secret) - return -EINVAL; - if (count < 7) - return -EINVAL; - if (memcmp(buf, "DHHC-1:", 7)) - return -EINVAL; - - dhchap_secret = kzalloc(count + 1, GFP_KERNEL); - if (!dhchap_secret) - return -ENOMEM; - memcpy(dhchap_secret, buf, count); - nvme_auth_stop(ctrl); - if (strcmp(dhchap_secret, opts->dhchap_secret)) { - struct nvme_dhchap_key *key, *host_key; - int ret; - - ret = nvme_auth_generate_key(dhchap_secret, &key); - if (ret) - return ret; - kfree(opts->dhchap_secret); - opts->dhchap_secret = dhchap_secret; - host_key = ctrl->host_key; - mutex_lock(&ctrl->dhchap_auth_mutex); - ctrl->host_key = key; - mutex_unlock(&ctrl->dhchap_auth_mutex); - nvme_auth_free_key(host_key); - } - /* Start re-authentication */ - dev_info(ctrl->device, "re-authenticating controller\n"); - queue_work(nvme_wq, &ctrl->dhchap_auth_work); - - return count; -} -static DEVICE_ATTR(dhchap_secret, S_IRUGO | S_IWUSR, - nvme_ctrl_dhchap_secret_show, nvme_ctrl_dhchap_secret_store); - -static ssize_t nvme_ctrl_dhchap_ctrl_secret_show(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); - struct nvmf_ctrl_options *opts = ctrl->opts; - - if (!opts->dhchap_ctrl_secret) - return sysfs_emit(buf, "none\n"); - return sysfs_emit(buf, "%s\n", opts->dhchap_ctrl_secret); -} - -static ssize_t nvme_ctrl_dhchap_ctrl_secret_store(struct device *dev, - struct device_attribute *attr, const char *buf, size_t count) -{ - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); - struct nvmf_ctrl_options *opts = ctrl->opts; - char *dhchap_secret; - - if (!ctrl->opts->dhchap_ctrl_secret) - return -EINVAL; - if (count < 7) - return -EINVAL; - if (memcmp(buf, "DHHC-1:", 7)) - return -EINVAL; - - dhchap_secret = kzalloc(count + 1, GFP_KERNEL); - if (!dhchap_secret) - return -ENOMEM; - memcpy(dhchap_secret, buf, count); - nvme_auth_stop(ctrl); - if (strcmp(dhchap_secret, opts->dhchap_ctrl_secret)) { - struct nvme_dhchap_key *key, *ctrl_key; - int ret; - - ret = nvme_auth_generate_key(dhchap_secret, &key); - if (ret) - return ret; - kfree(opts->dhchap_ctrl_secret); - opts->dhchap_ctrl_secret = dhchap_secret; - ctrl_key = ctrl->ctrl_key; - mutex_lock(&ctrl->dhchap_auth_mutex); - ctrl->ctrl_key = key; - mutex_unlock(&ctrl->dhchap_auth_mutex); - nvme_auth_free_key(ctrl_key); - } - /* Start re-authentication */ - dev_info(ctrl->device, "re-authenticating controller\n"); - queue_work(nvme_wq, &ctrl->dhchap_auth_work); - - return count; -} -static DEVICE_ATTR(dhchap_ctrl_secret, S_IRUGO | S_IWUSR, - nvme_ctrl_dhchap_ctrl_secret_show, nvme_ctrl_dhchap_ctrl_secret_store); -#endif - -static struct attribute *nvme_dev_attrs[] = { - &dev_attr_reset_controller.attr, - &dev_attr_rescan_controller.attr, - &dev_attr_model.attr, - &dev_attr_serial.attr, - &dev_attr_firmware_rev.attr, - &dev_attr_cntlid.attr, - &dev_attr_delete_controller.attr, - &dev_attr_transport.attr, - &dev_attr_subsysnqn.attr, - &dev_attr_address.attr, - &dev_attr_state.attr, - &dev_attr_numa_node.attr, - &dev_attr_queue_count.attr, - &dev_attr_sqsize.attr, - &dev_attr_hostnqn.attr, - &dev_attr_hostid.attr, - &dev_attr_ctrl_loss_tmo.attr, - &dev_attr_reconnect_delay.attr, - &dev_attr_fast_io_fail_tmo.attr, - &dev_attr_kato.attr, - &dev_attr_cntrltype.attr, - &dev_attr_dctype.attr, -#ifdef CONFIG_NVME_AUTH - &dev_attr_dhchap_secret.attr, - &dev_attr_dhchap_ctrl_secret.attr, -#endif - NULL -}; - -static umode_t nvme_dev_attrs_are_visible(struct kobject *kobj, - struct attribute *a, int n) -{ - struct device *dev = container_of(kobj, struct device, kobj); - struct nvme_ctrl *ctrl = dev_get_drvdata(dev); - - if (a == &dev_attr_delete_controller.attr && !ctrl->ops->delete_ctrl) - return 0; - if (a == &dev_attr_address.attr && !ctrl->ops->get_address) - return 0; - if (a == &dev_attr_hostnqn.attr && !ctrl->opts) - return 0; - if (a == &dev_attr_hostid.attr && !ctrl->opts) - return 0; - if (a == &dev_attr_ctrl_loss_tmo.attr && !ctrl->opts) - return 0; - if (a == &dev_attr_reconnect_delay.attr && !ctrl->opts) - return 0; - if (a == &dev_attr_fast_io_fail_tmo.attr && !ctrl->opts) - return 0; -#ifdef CONFIG_NVME_AUTH - if (a == &dev_attr_dhchap_secret.attr && !ctrl->opts) - return 0; - if (a == &dev_attr_dhchap_ctrl_secret.attr && !ctrl->opts) - return 0; -#endif - - return a->mode; -} - -const struct attribute_group nvme_dev_attrs_group = { - .attrs = nvme_dev_attrs, - .is_visible = nvme_dev_attrs_are_visible, -}; -EXPORT_SYMBOL_GPL(nvme_dev_attrs_group); - -static const struct attribute_group *nvme_dev_attr_groups[] = { - &nvme_dev_attrs_group, - NULL, -}; - static struct nvme_ns_head *nvme_find_ns_head(struct nvme_ctrl *ctrl, unsigned nsid) { @@ -3963,7 +3783,7 @@ static struct nvme_ns_head *nvme_find_ns_head(struct nvme_ctrl *ctrl, */ if (h->ns_id != nsid || !nvme_is_unique_nsid(ctrl, h)) continue; - if (!list_empty(&h->list) && nvme_tryget_ns_head(h)) + if (nvme_tryget_ns_head(h)) return h; } @@ -4014,7 +3834,7 @@ int nvme_cdev_add(struct cdev *cdev, struct device *cdev_device, if (minor < 0) return minor; cdev_device->devt = MKDEV(MAJOR(nvme_ns_chr_devt), minor); - cdev_device->class = nvme_ns_chr_class; + cdev_device->class = &nvme_ns_chr_class; cdev_device->release = nvme_cdev_rel; device_initialize(cdev_device); cdev_init(cdev, fops); @@ -4087,6 +3907,9 @@ static struct nvme_ns_head *nvme_alloc_ns_head(struct nvme_ctrl *ctrl, head->ns_id = info->nsid; head->ids = info->ids; head->shared = info->is_shared; + head->rotational = info->is_rotational; + ratelimit_state_init(&head->rs_nuse, 5 * HZ, 1); + ratelimit_set_flags(&head->rs_nuse, RATELIMIT_MSG_ON_RELEASE); kref_init(&head->ref); if (head->ids.csi) { @@ -4151,10 +3974,40 @@ static int nvme_init_ns_head(struct nvme_ns *ns, struct nvme_ns_info *info) ret = nvme_global_check_duplicate_ids(ctrl->subsys, &info->ids); if (ret) { - dev_err(ctrl->device, - "globally duplicate IDs for nsid %d\n", info->nsid); + /* + * We've found two different namespaces on two different + * subsystems that report the same ID. This is pretty nasty + * for anything that actually requires unique device + * identification. In the kernel we need this for multipathing, + * and in user space the /dev/disk/by-id/ links rely on it. + * + * If the device also claims to be multi-path capable back off + * here now and refuse the probe the second device as this is a + * recipe for data corruption. If not this is probably a + * cheap consumer device if on the PCIe bus, so let the user + * proceed and use the shiny toy, but warn that with changing + * probing order (which due to our async probing could just be + * device taking longer to startup) the other device could show + * up at any time. + */ nvme_print_device_info(ctrl); - return ret; + if ((ns->ctrl->ops->flags & NVME_F_FABRICS) || /* !PCIe */ + ((ns->ctrl->subsys->cmic & NVME_CTRL_CMIC_MULTI_CTRL) && + info->is_shared)) { + dev_err(ctrl->device, + "ignoring nsid %d because of duplicate IDs\n", + info->nsid); + return ret; + } + + dev_err(ctrl->device, + "clearing duplicate IDs for nsid %d\n", info->nsid); + dev_err(ctrl->device, + "use of /dev/disk/by-id/ may cause data corruption\n"); + memset(&info->ids.nguid, 0, sizeof(info->ids.nguid)); + memset(&info->ids.uuid, 0, sizeof(info->ids.uuid)); + memset(&info->ids.eui64, 0, sizeof(info->ids.eui64)); + ctrl->quirks |= NVME_QUIRK_BOGUS_NID; } mutex_lock(&ctrl->subsys->lock); @@ -4174,7 +4027,8 @@ static int nvme_init_ns_head(struct nvme_ns *ns, struct nvme_ns_info *info) } } else { ret = -EINVAL; - if (!info->is_shared || !head->shared) { + if ((!info->is_shared || !head->shared) && + !list_empty(&head->list)) { dev_err(ctrl->device, "Duplicate unshared namespace %d\n", info->nsid); @@ -4187,18 +4041,22 @@ static int nvme_init_ns_head(struct nvme_ns *ns, struct nvme_ns_info *info) goto out_put_ns_head; } - if (!multipath && !list_empty(&head->list)) { + if (!multipath) { dev_warn(ctrl->device, "Found shared namespace %d, but multipathing not supported.\n", info->nsid); dev_warn_once(ctrl->device, - "Support for shared namespaces without CONFIG_NVME_MULTIPATH is deprecated and will be removed in Linux 6.0\n."); + "Shared namespace support requires core_nvme.multipath=Y.\n"); } } list_add_tail_rcu(&ns->siblings, &head->list); ns->head = head; mutex_unlock(&ctrl->subsys->lock); + +#ifdef CONFIG_NVME_MULTIPATH + cancel_delayed_work(&head->remove_work); +#endif return 0; out_put_ns_head: @@ -4211,9 +4069,11 @@ out_unlock: struct nvme_ns *nvme_find_get_ns(struct nvme_ctrl *ctrl, unsigned nsid) { struct nvme_ns *ns, *ret = NULL; + int srcu_idx; - down_read(&ctrl->namespaces_rwsem); - list_for_each_entry(ns, &ctrl->namespaces, list) { + srcu_idx = srcu_read_lock(&ctrl->srcu); + list_for_each_entry_srcu(ns, &ctrl->namespaces, list, + srcu_read_lock_held(&ctrl->srcu)) { if (ns->head->ns_id == nsid) { if (!nvme_get_ns(ns)) continue; @@ -4223,10 +4083,10 @@ struct nvme_ns *nvme_find_get_ns(struct nvme_ctrl *ctrl, unsigned nsid) if (ns->head->ns_id > nsid) break; } - up_read(&ctrl->namespaces_rwsem); + srcu_read_unlock(&ctrl->srcu, srcu_idx); return ret; } -EXPORT_SYMBOL_NS_GPL(nvme_find_get_ns, NVME_TARGET_PASSTHRU); +EXPORT_SYMBOL_NS_GPL(nvme_find_get_ns, "NVME_TARGET_PASSTHRU"); /* * Add the namespace to the controller list while keeping the list ordered. @@ -4237,24 +4097,32 @@ static void nvme_ns_add_to_ctrl_list(struct nvme_ns *ns) list_for_each_entry_reverse(tmp, &ns->ctrl->namespaces, list) { if (tmp->head->ns_id < ns->head->ns_id) { - list_add(&ns->list, &tmp->list); + list_add_rcu(&ns->list, &tmp->list); return; } } - list_add(&ns->list, &ns->ctrl->namespaces); + list_add_rcu(&ns->list, &ns->ctrl->namespaces); } static void nvme_alloc_ns(struct nvme_ctrl *ctrl, struct nvme_ns_info *info) { + struct queue_limits lim = { }; struct nvme_ns *ns; struct gendisk *disk; int node = ctrl->numa_node; + bool last_path = false; ns = kzalloc_node(sizeof(*ns), GFP_KERNEL, node); if (!ns) return; - disk = blk_mq_alloc_disk(ctrl->tagset, ns); + if (ctrl->opts && ctrl->opts->data_digest) + lim.features |= BLK_FEAT_STABLE_WRITES; + if (ctrl->ops->supports_pci_p2pdma && + ctrl->ops->supports_pci_p2pdma(ctrl)) + lim.features |= BLK_FEAT_PCI_P2PDMA; + + disk = blk_mq_alloc_disk(ctrl->tagset, &lim, ns); if (IS_ERR(disk)) goto out_free_ns; disk->fops = &nvme_bdev_ops; @@ -4262,15 +4130,6 @@ static void nvme_alloc_ns(struct nvme_ctrl *ctrl, struct nvme_ns_info *info) ns->disk = disk; ns->queue = disk->queue; - - if (ctrl->opts && ctrl->opts->data_digest) - blk_queue_flag_set(QUEUE_FLAG_STABLE_WRITES, ns->queue); - - blk_queue_flag_set(QUEUE_FLAG_NONROT, ns->queue); - if (ctrl->ops->supports_pci_p2pdma && - ctrl->ops->supports_pci_p2pdma(ctrl)) - blk_queue_flag_set(QUEUE_FLAG_PCI_P2PDMA, ns->queue); - ns->ctrl = ctrl; kref_init(&ns->kref); @@ -4288,7 +4147,7 @@ static void nvme_alloc_ns(struct nvme_ctrl *ctrl, struct nvme_ns_info *info) * instance as shared namespaces will show up as multiple block * devices. */ - if (ns->head->disk) { + if (nvme_ns_head_multipath(ns->head)) { sprintf(disk->disk_name, "nvme%dc%dn%d", ctrl->subsys->instance, ctrl->instance, ns->head->instance); disk->flags |= GENHD_FL_HIDDEN; @@ -4303,12 +4162,21 @@ static void nvme_alloc_ns(struct nvme_ctrl *ctrl, struct nvme_ns_info *info) if (nvme_update_ns_info(ns, info)) goto out_unlink_ns; - down_write(&ctrl->namespaces_rwsem); + mutex_lock(&ctrl->namespaces_lock); + /* + * Ensure that no namespaces are added to the ctrl list after the queues + * are frozen, thereby avoiding a deadlock between scan and reset. + */ + if (test_bit(NVME_CTRL_FROZEN, &ctrl->flags)) { + mutex_unlock(&ctrl->namespaces_lock); + goto out_unlink_ns; + } nvme_ns_add_to_ctrl_list(ns); - up_write(&ctrl->namespaces_rwsem); + mutex_unlock(&ctrl->namespaces_lock); + synchronize_srcu(&ctrl->srcu); nvme_get_ctrl(ctrl); - if (device_add_disk(ctrl->device, ns->disk, nvme_ns_id_attr_groups)) + if (device_add_disk(ctrl->device, ns->disk, nvme_ns_attr_groups)) goto out_cleanup_ns_from_list; if (!nvme_ns_head_multipath(ns->head)) @@ -4317,19 +4185,40 @@ static void nvme_alloc_ns(struct nvme_ctrl *ctrl, struct nvme_ns_info *info) nvme_mpath_add_disk(ns, info->anagrpid); nvme_fault_inject_init(&ns->fault_inject, ns->disk->disk_name); + /* + * Set ns->disk->device->driver_data to ns so we can access + * ns->head->passthru_err_log_enabled in + * nvme_io_passthru_err_log_enabled_[store | show](). + */ + dev_set_drvdata(disk_to_dev(ns->disk), ns); + return; out_cleanup_ns_from_list: nvme_put_ctrl(ctrl); - down_write(&ctrl->namespaces_rwsem); - list_del_init(&ns->list); - up_write(&ctrl->namespaces_rwsem); + mutex_lock(&ctrl->namespaces_lock); + list_del_rcu(&ns->list); + mutex_unlock(&ctrl->namespaces_lock); + synchronize_srcu(&ctrl->srcu); out_unlink_ns: mutex_lock(&ctrl->subsys->lock); list_del_rcu(&ns->siblings); - if (list_empty(&ns->head->list)) + if (list_empty(&ns->head->list)) { list_del_init(&ns->head->entry); + /* + * If multipath is not configured, we still create a namespace + * head (nshead), but head->disk is not initialized in that + * case. As a result, only a single reference to nshead is held + * (via kref_init()) when it is created. Therefore, ensure that + * we do not release the reference to nshead twice if head->disk + * is not present. + */ + if (ns->head->disk) + last_path = true; + } mutex_unlock(&ctrl->subsys->lock); + if (last_path) + nvme_put_ns_head(ns->head); nvme_put_ns_head(ns->head); out_cleanup_disk: put_disk(disk); @@ -4361,7 +4250,8 @@ static void nvme_ns_remove(struct nvme_ns *ns) mutex_lock(&ns->ctrl->subsys->lock); list_del_rcu(&ns->siblings); if (list_empty(&ns->head->list)) { - list_del_init(&ns->head->entry); + if (!nvme_mpath_queue_if_no_path(ns->head)) + list_del_init(&ns->head->entry); last_path = true; } mutex_unlock(&ns->ctrl->subsys->lock); @@ -4371,14 +4261,18 @@ static void nvme_ns_remove(struct nvme_ns *ns) if (!nvme_ns_head_multipath(ns->head)) nvme_cdev_del(&ns->cdev, &ns->cdev_device); + + nvme_mpath_remove_sysfs_link(ns); + del_gendisk(ns->disk); - down_write(&ns->ctrl->namespaces_rwsem); - list_del_init(&ns->list); - up_write(&ns->ctrl->namespaces_rwsem); + mutex_lock(&ns->ctrl->namespaces_lock); + list_del_rcu(&ns->list); + mutex_unlock(&ns->ctrl->namespaces_lock); + synchronize_srcu(&ns->ctrl->srcu); if (last_path) - nvme_mpath_shutdown_disk(ns->head); + nvme_mpath_remove_disk(ns->head); nvme_put_ns(ns); } @@ -4394,7 +4288,7 @@ static void nvme_ns_remove_by_nsid(struct nvme_ctrl *ctrl, u32 nsid) static void nvme_validate_ns(struct nvme_ns *ns, struct nvme_ns_info *info) { - int ret = NVME_SC_INVALID_NS | NVME_SC_DNR; + int ret = NVME_SC_INVALID_NS | NVME_STATUS_DNR; if (!nvme_ns_ids_equal(&ns->head->ids, &info->ids)) { dev_err(ns->ctrl->device, @@ -4410,7 +4304,7 @@ out: * * TODO: we should probably schedule a delayed retry here. */ - if (ret > 0 && (ret & NVME_SC_DNR)) + if (ret > 0 && (ret & NVME_STATUS_DNR)) nvme_ns_remove(ns); } @@ -4418,6 +4312,7 @@ static void nvme_scan_ns(struct nvme_ctrl *ctrl, unsigned nsid) { struct nvme_ns_info info = { .nsid = nsid }; struct nvme_ns *ns; + int ret = 1; if (nvme_identify_ns_descs(ctrl, &info)) return; @@ -4429,24 +4324,25 @@ static void nvme_scan_ns(struct nvme_ctrl *ctrl, unsigned nsid) } /* - * If available try to use the Command Set Idependent Identify Namespace + * If available try to use the Command Set Independent Identify Namespace * data structure to find all the generic information that is needed to * set up a namespace. If not fall back to the legacy version. */ if ((ctrl->cap & NVME_CAP_CRMS_CRIMS) || - (info.ids.csi != NVME_CSI_NVM && info.ids.csi != NVME_CSI_ZNS)) { - if (nvme_ns_info_from_id_cs_indep(ctrl, &info)) - return; - } else { - if (nvme_ns_info_from_identify(ctrl, &info)) - return; - } + (info.ids.csi != NVME_CSI_NVM && info.ids.csi != NVME_CSI_ZNS) || + ctrl->vs >= NVME_VS(2, 0, 0)) + ret = nvme_ns_info_from_id_cs_indep(ctrl, &info); + if (ret > 0) + ret = nvme_ns_info_from_identify(ctrl, &info); + + if (info.is_removed) + nvme_ns_remove_by_nsid(ctrl, nsid); /* * Ignore the namespace if it is not ready. We will get an AEN once it * becomes ready and restart the scan. */ - if (!info.is_ready) + if (ret || !info.is_ready) return; ns = nvme_find_get_ns(ctrl, nsid); @@ -4458,22 +4354,53 @@ static void nvme_scan_ns(struct nvme_ctrl *ctrl, unsigned nsid) } } +/** + * struct async_scan_info - keeps track of controller & NSIDs to scan + * @ctrl: Controller on which namespaces are being scanned + * @next_nsid: Index of next NSID to scan in ns_list + * @ns_list: Pointer to list of NSIDs to scan + * + * Note: There is a single async_scan_info structure shared by all instances + * of nvme_scan_ns_async() scanning a given controller, so the atomic + * operations on next_nsid are critical to ensure each instance scans a unique + * NSID. + */ +struct async_scan_info { + struct nvme_ctrl *ctrl; + atomic_t next_nsid; + __le32 *ns_list; +}; + +static void nvme_scan_ns_async(void *data, async_cookie_t cookie) +{ + struct async_scan_info *scan_info = data; + int idx; + u32 nsid; + + idx = (u32)atomic_fetch_inc(&scan_info->next_nsid); + nsid = le32_to_cpu(scan_info->ns_list[idx]); + + nvme_scan_ns(scan_info->ctrl, nsid); +} + static void nvme_remove_invalid_namespaces(struct nvme_ctrl *ctrl, unsigned nsid) { struct nvme_ns *ns, *next; LIST_HEAD(rm_list); - down_write(&ctrl->namespaces_rwsem); + mutex_lock(&ctrl->namespaces_lock); list_for_each_entry_safe(ns, next, &ctrl->namespaces, list) { - if (ns->head->ns_id > nsid) - list_move_tail(&ns->list, &rm_list); + if (ns->head->ns_id > nsid) { + list_del_rcu(&ns->list); + synchronize_srcu(&ctrl->srcu); + list_add_tail_rcu(&ns->list, &rm_list); + } } - up_write(&ctrl->namespaces_rwsem); + mutex_unlock(&ctrl->namespaces_lock); list_for_each_entry_safe(ns, next, &rm_list, list) nvme_ns_remove(ns); - } static int nvme_scan_ns_list(struct nvme_ctrl *ctrl) @@ -4482,11 +4409,15 @@ static int nvme_scan_ns_list(struct nvme_ctrl *ctrl) __le32 *ns_list; u32 prev = 0; int ret = 0, i; + ASYNC_DOMAIN(domain); + struct async_scan_info scan_info; ns_list = kzalloc(NVME_IDENTIFY_DATA_SIZE, GFP_KERNEL); if (!ns_list) return -ENOMEM; + scan_info.ctrl = ctrl; + scan_info.ns_list = ns_list; for (;;) { struct nvme_command cmd = { .identify.opcode = nvme_admin_identify, @@ -4502,19 +4433,23 @@ static int nvme_scan_ns_list(struct nvme_ctrl *ctrl) goto free; } + atomic_set(&scan_info.next_nsid, 0); for (i = 0; i < nr_entries; i++) { u32 nsid = le32_to_cpu(ns_list[i]); if (!nsid) /* end of the list? */ goto out; - nvme_scan_ns(ctrl, nsid); + async_schedule_domain(nvme_scan_ns_async, &scan_info, + &domain); while (++prev < nsid) nvme_ns_remove_by_nsid(ctrl, prev); } + async_synchronize_full_domain(&domain); } out: nvme_remove_invalid_namespaces(ctrl, prev); free: + async_synchronize_full_domain(&domain); kfree(ns_list); return ret; } @@ -4567,7 +4502,7 @@ static void nvme_scan_work(struct work_struct *work) int ret; /* No tagset on a live ctrl means IO queues could not created */ - if (ctrl->state != NVME_CTRL_LIVE || !ctrl->tagset) + if (nvme_ctrl_state(ctrl) != NVME_CTRL_LIVE || !ctrl->tagset) return; /* @@ -4590,7 +4525,7 @@ static void nvme_scan_work(struct work_struct *work) } mutex_lock(&ctrl->scan_lock); - if (nvme_ctrl_limited_cns(ctrl)) { + if (!nvme_id_cns_ok(ctrl, NVME_ID_CNS_NS_ACTIVE_LIST)) { nvme_scan_ns_sequential(ctrl); } else { /* @@ -4599,10 +4534,19 @@ static void nvme_scan_work(struct work_struct *work) * they report) but don't actually support it. */ ret = nvme_scan_ns_list(ctrl); - if (ret > 0 && ret & NVME_SC_DNR) + if (ret > 0 && ret & NVME_STATUS_DNR) nvme_scan_ns_sequential(ctrl); } mutex_unlock(&ctrl->scan_lock); + + /* Requeue if we have missed AENs */ + if (test_bit(NVME_AER_NOTICE_NS_CHANGED, &ctrl->events)) + nvme_queue_scan(ctrl); +#ifdef CONFIG_NVME_MULTIPATH + else if (ctrl->ana_log_buf) + /* Re-read the ANA log page to not miss updates */ + queue_work(nvme_wq, &ctrl->ana_work); +#endif } /* @@ -4622,6 +4566,12 @@ void nvme_remove_namespaces(struct nvme_ctrl *ctrl) */ nvme_mpath_clear_ctrl_paths(ctrl); + /* + * Unquiesce io queues so any pending IO won't hang, especially + * those submitted from scan work + */ + nvme_unquiesce_io_queues(ctrl); + /* prevent racing with ns scanning */ flush_work(&ctrl->scan_work); @@ -4631,17 +4581,16 @@ void nvme_remove_namespaces(struct nvme_ctrl *ctrl) * removing the namespaces' disks; fail all the queues now to avoid * potentially having to clean up the failed sync later. */ - if (ctrl->state == NVME_CTRL_DEAD) { + if (nvme_ctrl_state(ctrl) == NVME_CTRL_DEAD) nvme_mark_namespaces_dead(ctrl); - nvme_unquiesce_io_queues(ctrl); - } /* this is a no-op when called from the controller reset handler */ nvme_change_ctrl_state(ctrl, NVME_CTRL_DELETING_NOIO); - down_write(&ctrl->namespaces_rwsem); - list_splice_init(&ctrl->namespaces, &ns_list); - up_write(&ctrl->namespaces_rwsem); + mutex_lock(&ctrl->namespaces_lock); + list_splice_init_rcu(&ctrl->namespaces, &ns_list, synchronize_rcu); + mutex_unlock(&ctrl->namespaces_lock); + synchronize_srcu(&ctrl->srcu); list_for_each_entry_safe(ns, next, &ns_list, list) nvme_ns_remove(ns); @@ -4715,7 +4664,7 @@ static void nvme_async_event_work(struct work_struct *work) * flushing ctrl async_event_work after changing the controller state * from LIVE and before freeing the admin queue. */ - if (ctrl->state == NVME_CTRL_LIVE) + if (nvme_ctrl_state(ctrl) == NVME_CTRL_LIVE) ctrl->ops->submit_async_event(ctrl); } @@ -4736,14 +4685,30 @@ static bool nvme_ctrl_pp_status(struct nvme_ctrl *ctrl) static void nvme_get_fw_slot_info(struct nvme_ctrl *ctrl) { struct nvme_fw_slot_info_log *log; + u8 next_fw_slot, cur_fw_slot; log = kmalloc(sizeof(*log), GFP_KERNEL); if (!log) return; if (nvme_get_log(ctrl, NVME_NSID_ALL, NVME_LOG_FW_SLOT, 0, NVME_CSI_NVM, - log, sizeof(*log), 0)) + log, sizeof(*log), 0)) { dev_warn(ctrl->device, "Get FW SLOT INFO log error\n"); + goto out_free_log; + } + + cur_fw_slot = log->afi & 0x7; + next_fw_slot = (log->afi & 0x70) >> 4; + if (!cur_fw_slot || (next_fw_slot && (cur_fw_slot != next_fw_slot))) { + dev_info(ctrl->device, + "Firmware is activated after next Controller Level Reset\n"); + goto out_free_log; + } + + memcpy(ctrl->subsys->firmware_rev, &log->frs[cur_fw_slot - 1], + sizeof(ctrl->subsys->firmware_rev)); + +out_free_log: kfree(log); } @@ -4753,12 +4718,12 @@ static void nvme_fw_act_work(struct work_struct *work) struct nvme_ctrl, fw_act_work); unsigned long fw_act_timeout; + nvme_auth_stop(ctrl); + if (ctrl->mtfa) - fw_act_timeout = jiffies + - msecs_to_jiffies(ctrl->mtfa * 100); + fw_act_timeout = jiffies + msecs_to_jiffies(ctrl->mtfa * 100); else - fw_act_timeout = jiffies + - msecs_to_jiffies(admin_timeout * 1000); + fw_act_timeout = jiffies + secs_to_jiffies(admin_timeout); nvme_quiesce_io_queues(ctrl); while (nvme_ctrl_pp_status(ctrl)) { @@ -4771,7 +4736,8 @@ static void nvme_fw_act_work(struct work_struct *work) msleep(100); } - if (!nvme_change_ctrl_state(ctrl, NVME_CTRL_LIVE)) + if (!nvme_change_ctrl_state(ctrl, NVME_CTRL_CONNECTING) || + !nvme_change_ctrl_state(ctrl, NVME_CTRL_LIVE)) return; nvme_unquiesce_io_queues(ctrl); @@ -4796,8 +4762,6 @@ static bool nvme_handle_aen_notice(struct nvme_ctrl *ctrl, u32 result) u32 aer_notice_type = nvme_aer_subtype(result); bool requeue = true; - trace_nvme_async_event(ctrl, aer_notice_type); - switch (aer_notice_type) { case NVME_AER_NOTICE_NS_CHANGED: set_bit(NVME_AER_NOTICE_NS_CHANGED, &ctrl->events); @@ -4810,7 +4774,6 @@ static bool nvme_handle_aen_notice(struct nvme_ctrl *ctrl, u32 result) * firmware activation. */ if (nvme_change_ctrl_state(ctrl, NVME_CTRL_RESETTING)) { - nvme_auth_stop(ctrl); requeue = false; queue_work(nvme_wq, &ctrl->fw_act_work); } @@ -4833,8 +4796,8 @@ static bool nvme_handle_aen_notice(struct nvme_ctrl *ctrl, u32 result) static void nvme_handle_aer_persistent_error(struct nvme_ctrl *ctrl) { - trace_nvme_async_event(ctrl, NVME_AER_ERROR); - dev_warn(ctrl->device, "resetting controller due to AER\n"); + dev_warn(ctrl->device, + "resetting controller due to persistent internal error\n"); nvme_reset_ctrl(ctrl); } @@ -4849,6 +4812,7 @@ void nvme_complete_async_event(struct nvme_ctrl *ctrl, __le16 status, if (le16_to_cpu(status) >> 1 != NVME_SC_SUCCESS) return; + trace_nvme_async_event(ctrl, result); switch (aer_type) { case NVME_AER_NOTICE: requeue = nvme_handle_aen_notice(ctrl, result); @@ -4866,7 +4830,6 @@ void nvme_complete_async_event(struct nvme_ctrl *ctrl, __le16 status, case NVME_AER_SMART: case NVME_AER_CSS: case NVME_AER_VS: - trace_nvme_async_event(ctrl, aer_type); ctrl->aen_result = result; break; default: @@ -4881,15 +4844,16 @@ EXPORT_SYMBOL_GPL(nvme_complete_async_event); int nvme_alloc_admin_tag_set(struct nvme_ctrl *ctrl, struct blk_mq_tag_set *set, const struct blk_mq_ops *ops, unsigned int cmd_size) { + struct queue_limits lim = {}; int ret; memset(set, 0, sizeof(*set)); set->ops = ops; set->queue_depth = NVME_AQ_MQ_TAG_DEPTH; if (ctrl->ops->flags & NVME_F_FABRICS) - set->reserved_tags = NVMF_RESERVED_TAGS; + /* Reserved for fabric connect and keep alive */ + set->reserved_tags = 2; set->numa_node = ctrl->numa_node; - set->flags = BLK_MQ_F_NO_SCHED; if (ctrl->ops->flags & NVME_F_BLOCKING) set->flags |= BLK_MQ_F_BLOCKING; set->cmd_size = cmd_size; @@ -4900,14 +4864,14 @@ int nvme_alloc_admin_tag_set(struct nvme_ctrl *ctrl, struct blk_mq_tag_set *set, if (ret) return ret; - ctrl->admin_q = blk_mq_init_queue(set); + ctrl->admin_q = blk_mq_alloc_queue(set, &lim, NULL); if (IS_ERR(ctrl->admin_q)) { ret = PTR_ERR(ctrl->admin_q); goto out_free_tagset; } if (ctrl->ops->flags & NVME_F_FABRICS) { - ctrl->fabrics_q = blk_mq_init_queue(set); + ctrl->fabrics_q = blk_mq_alloc_queue(set, NULL, NULL); if (IS_ERR(ctrl->fabrics_q)) { ret = PTR_ERR(ctrl->fabrics_q); goto out_cleanup_admin_q; @@ -4921,15 +4885,21 @@ out_cleanup_admin_q: blk_mq_destroy_queue(ctrl->admin_q); blk_put_queue(ctrl->admin_q); out_free_tagset: - blk_mq_free_tag_set(ctrl->admin_tagset); + blk_mq_free_tag_set(set); + ctrl->admin_q = NULL; + ctrl->fabrics_q = NULL; return ret; } EXPORT_SYMBOL_GPL(nvme_alloc_admin_tag_set); void nvme_remove_admin_tag_set(struct nvme_ctrl *ctrl) { + /* + * As we're about to destroy the queue and free tagset + * we can not have keep-alive work running. + */ + nvme_stop_keep_alive(ctrl); blk_mq_destroy_queue(ctrl->admin_q); - blk_put_queue(ctrl->admin_q); if (ctrl->ops->flags & NVME_F_FABRICS) { blk_mq_destroy_queue(ctrl->fabrics_q); blk_put_queue(ctrl->fabrics_q); @@ -4954,12 +4924,12 @@ int nvme_alloc_io_tag_set(struct nvme_ctrl *ctrl, struct blk_mq_tag_set *set, if (ctrl->quirks & NVME_QUIRK_SHARED_TAGS) set->reserved_tags = NVME_AQ_DEPTH; else if (ctrl->ops->flags & NVME_F_FABRICS) - set->reserved_tags = NVMF_RESERVED_TAGS; + /* Reserved for fabric connect */ + set->reserved_tags = 1; set->numa_node = ctrl->numa_node; - set->flags = BLK_MQ_F_SHOULD_MERGE; if (ctrl->ops->flags & NVME_F_BLOCKING) set->flags |= BLK_MQ_F_BLOCKING; - set->cmd_size = cmd_size, + set->cmd_size = cmd_size; set->driver_data = ctrl; set->nr_hw_queues = ctrl->queue_count - 1; set->timeout = NVME_IO_TIMEOUT; @@ -4969,13 +4939,15 @@ int nvme_alloc_io_tag_set(struct nvme_ctrl *ctrl, struct blk_mq_tag_set *set, return ret; if (ctrl->ops->flags & NVME_F_FABRICS) { - ctrl->connect_q = blk_mq_init_queue(set); + struct queue_limits lim = { + .features = BLK_FEAT_SKIP_TAGSET_QUIESCE, + }; + + ctrl->connect_q = blk_mq_alloc_queue(set, &lim, NULL); if (IS_ERR(ctrl->connect_q)) { ret = PTR_ERR(ctrl->connect_q); goto out_free_tag_set; } - blk_queue_flag_set(QUEUE_FLAG_SKIP_TAGSET_QUIESCE, - ctrl->connect_q); } ctrl->tagset = set; @@ -4983,6 +4955,7 @@ int nvme_alloc_io_tag_set(struct nvme_ctrl *ctrl, struct blk_mq_tag_set *set, out_free_tag_set: blk_mq_free_tag_set(set); + ctrl->connect_q = NULL; return ret; } EXPORT_SYMBOL_GPL(nvme_alloc_io_tag_set); @@ -5001,7 +4974,6 @@ void nvme_stop_ctrl(struct nvme_ctrl *ctrl) { nvme_mpath_stop(ctrl); nvme_auth_stop(ctrl); - nvme_stop_keep_alive(ctrl); nvme_stop_failfast_work(ctrl); flush_work(&ctrl->async_event_work); cancel_work_sync(&ctrl->fw_act_work); @@ -5012,8 +4984,6 @@ EXPORT_SYMBOL_GPL(nvme_stop_ctrl); void nvme_start_ctrl(struct nvme_ctrl *ctrl) { - nvme_start_keep_alive(ctrl); - nvme_enable_aen(ctrl); /* @@ -5022,9 +4992,15 @@ void nvme_start_ctrl(struct nvme_ctrl *ctrl) * that were missed. We identify persistent discovery controllers by * checking that they started once before, hence are reconnecting back. */ - if (test_and_set_bit(NVME_CTRL_STARTED_ONCE, &ctrl->flags) && - nvme_discovery_ctrl(ctrl)) + if (test_bit(NVME_CTRL_STARTED_ONCE, &ctrl->flags) && + nvme_discovery_ctrl(ctrl)) { + if (!ctrl->kato) { + nvme_stop_keep_alive(ctrl); + ctrl->kato = NVME_DEFAULT_KATO; + nvme_start_keep_alive(ctrl); + } nvme_change_uevent(ctrl, "NVME_EVENT=rediscover"); + } if (ctrl->queue_count > 1) { nvme_queue_scan(ctrl); @@ -5033,11 +5009,13 @@ void nvme_start_ctrl(struct nvme_ctrl *ctrl) } nvme_change_uevent(ctrl, "NVME_EVENT=connected"); + set_bit(NVME_CTRL_STARTED_ONCE, &ctrl->flags); } EXPORT_SYMBOL_GPL(nvme_start_ctrl); void nvme_uninit_ctrl(struct nvme_ctrl *ctrl) { + nvme_stop_keep_alive(ctrl); nvme_hwmon_exit(ctrl); nvme_fault_inject_fini(&ctrl->fault_inject); dev_pm_qos_hide_latency_tolerance(ctrl->device); @@ -5065,11 +5043,13 @@ static void nvme_free_ctrl(struct device *dev) container_of(dev, struct nvme_ctrl, ctrl_device); struct nvme_subsystem *subsys = ctrl->subsys; + if (ctrl->admin_q) + blk_put_queue(ctrl->admin_q); if (!subsys || ctrl->instance != subsys->instance) ida_free(&nvme_instance_ida, ctrl->instance); - nvme_free_cels(ctrl); nvme_mpath_uninit(ctrl); + cleanup_srcu_struct(&ctrl->srcu); nvme_auth_stop(ctrl); nvme_auth_free(ctrl); __free_page(ctrl->discard_page); @@ -5092,19 +5072,28 @@ static void nvme_free_ctrl(struct device *dev) * Initialize a NVMe controller structures. This needs to be called during * earliest initialization so that we have the initialized structured around * during probing. + * + * On success, the caller must use the nvme_put_ctrl() to release this when + * needed, which also invokes the ops->free_ctrl() callback. */ int nvme_init_ctrl(struct nvme_ctrl *ctrl, struct device *dev, const struct nvme_ctrl_ops *ops, unsigned long quirks) { int ret; - ctrl->state = NVME_CTRL_NEW; + WRITE_ONCE(ctrl->state, NVME_CTRL_NEW); + ctrl->passthru_err_log_enabled = false; clear_bit(NVME_CTRL_FAILFAST_EXPIRED, &ctrl->flags); spin_lock_init(&ctrl->lock); + mutex_init(&ctrl->namespaces_lock); + + ret = init_srcu_struct(&ctrl->srcu); + if (ret) + return ret; + mutex_init(&ctrl->scan_lock); INIT_LIST_HEAD(&ctrl->namespaces); xa_init(&ctrl->cels); - init_rwsem(&ctrl->namespaces_rwsem); ctrl->dev = dev; ctrl->ops = ops; ctrl->quirks = quirks; @@ -5119,6 +5108,7 @@ int nvme_init_ctrl(struct nvme_ctrl *ctrl, struct device *dev, INIT_DELAYED_WORK(&ctrl->failfast_work, nvme_failfast_work); memset(&ctrl->ka_cmd, 0, sizeof(ctrl->ka_cmd)); ctrl->ka_cmd.common.opcode = nvme_admin_keep_alive; + ctrl->ka_last_check_time = jiffies; BUILD_BUG_ON(NVME_DSM_MAX_RANGES * sizeof(struct nvme_dsm_range) > PAGE_SIZE); @@ -5133,11 +5123,17 @@ int nvme_init_ctrl(struct nvme_ctrl *ctrl, struct device *dev, goto out; ctrl->instance = ret; + ret = nvme_auth_init_ctrl(ctrl); + if (ret) + goto out_release_instance; + + nvme_mpath_init_ctrl(ctrl); + device_initialize(&ctrl->ctrl_device); ctrl->device = &ctrl->ctrl_device; ctrl->device->devt = MKDEV(MAJOR(nvme_ctrl_base_chr_devt), ctrl->instance); - ctrl->device->class = nvme_class; + ctrl->device->class = &nvme_class; ctrl->device->parent = ctrl->dev; if (ops->dev_attr_groups) ctrl->device->groups = ops->dev_attr_groups; @@ -5145,16 +5141,36 @@ int nvme_init_ctrl(struct nvme_ctrl *ctrl, struct device *dev, ctrl->device->groups = nvme_dev_attr_groups; ctrl->device->release = nvme_free_ctrl; dev_set_drvdata(ctrl->device, ctrl); + + return ret; + +out_release_instance: + ida_free(&nvme_instance_ida, ctrl->instance); +out: + if (ctrl->discard_page) + __free_page(ctrl->discard_page); + cleanup_srcu_struct(&ctrl->srcu); + return ret; +} +EXPORT_SYMBOL_GPL(nvme_init_ctrl); + +/* + * On success, returns with an elevated controller reference and caller must + * use nvme_uninit_ctrl() to properly free resources associated with the ctrl. + */ +int nvme_add_ctrl(struct nvme_ctrl *ctrl) +{ + int ret; + ret = dev_set_name(ctrl->device, "nvme%d", ctrl->instance); if (ret) - goto out_release_instance; + return ret; - nvme_get_ctrl(ctrl); cdev_init(&ctrl->cdev, &nvme_dev_fops); - ctrl->cdev.owner = ops->module; + ctrl->cdev.owner = ctrl->ops->module; ret = cdev_device_add(&ctrl->cdev, ctrl->device); if (ret) - goto out_free_name; + return ret; /* * Initialize latency tolerance controls. The sysfs files won't @@ -5165,60 +5181,53 @@ int nvme_init_ctrl(struct nvme_ctrl *ctrl, struct device *dev, min(default_ps_max_latency_us, (unsigned long)S32_MAX)); nvme_fault_inject_init(&ctrl->fault_inject, dev_name(ctrl->device)); - nvme_mpath_init_ctrl(ctrl); - ret = nvme_auth_init_ctrl(ctrl); - if (ret) - goto out_free_cdev; + nvme_get_ctrl(ctrl); return 0; -out_free_cdev: - cdev_device_del(&ctrl->cdev, ctrl->device); -out_free_name: - nvme_put_ctrl(ctrl); - kfree_const(ctrl->device->kobj.name); -out_release_instance: - ida_free(&nvme_instance_ida, ctrl->instance); -out: - if (ctrl->discard_page) - __free_page(ctrl->discard_page); - return ret; } -EXPORT_SYMBOL_GPL(nvme_init_ctrl); +EXPORT_SYMBOL_GPL(nvme_add_ctrl); /* let I/O to all namespaces fail in preparation for surprise removal */ void nvme_mark_namespaces_dead(struct nvme_ctrl *ctrl) { struct nvme_ns *ns; + int srcu_idx; - down_read(&ctrl->namespaces_rwsem); - list_for_each_entry(ns, &ctrl->namespaces, list) + srcu_idx = srcu_read_lock(&ctrl->srcu); + list_for_each_entry_srcu(ns, &ctrl->namespaces, list, + srcu_read_lock_held(&ctrl->srcu)) blk_mark_disk_dead(ns->disk); - up_read(&ctrl->namespaces_rwsem); + srcu_read_unlock(&ctrl->srcu, srcu_idx); } EXPORT_SYMBOL_GPL(nvme_mark_namespaces_dead); void nvme_unfreeze(struct nvme_ctrl *ctrl) { struct nvme_ns *ns; + int srcu_idx; - down_read(&ctrl->namespaces_rwsem); - list_for_each_entry(ns, &ctrl->namespaces, list) - blk_mq_unfreeze_queue(ns->queue); - up_read(&ctrl->namespaces_rwsem); + srcu_idx = srcu_read_lock(&ctrl->srcu); + list_for_each_entry_srcu(ns, &ctrl->namespaces, list, + srcu_read_lock_held(&ctrl->srcu)) + blk_mq_unfreeze_queue_non_owner(ns->queue); + srcu_read_unlock(&ctrl->srcu, srcu_idx); + clear_bit(NVME_CTRL_FROZEN, &ctrl->flags); } EXPORT_SYMBOL_GPL(nvme_unfreeze); int nvme_wait_freeze_timeout(struct nvme_ctrl *ctrl, long timeout) { struct nvme_ns *ns; + int srcu_idx; - down_read(&ctrl->namespaces_rwsem); - list_for_each_entry(ns, &ctrl->namespaces, list) { + srcu_idx = srcu_read_lock(&ctrl->srcu); + list_for_each_entry_srcu(ns, &ctrl->namespaces, list, + srcu_read_lock_held(&ctrl->srcu)) { timeout = blk_mq_freeze_queue_wait_timeout(ns->queue, timeout); if (timeout <= 0) break; } - up_read(&ctrl->namespaces_rwsem); + srcu_read_unlock(&ctrl->srcu, srcu_idx); return timeout; } EXPORT_SYMBOL_GPL(nvme_wait_freeze_timeout); @@ -5226,22 +5235,32 @@ EXPORT_SYMBOL_GPL(nvme_wait_freeze_timeout); void nvme_wait_freeze(struct nvme_ctrl *ctrl) { struct nvme_ns *ns; + int srcu_idx; - down_read(&ctrl->namespaces_rwsem); - list_for_each_entry(ns, &ctrl->namespaces, list) + srcu_idx = srcu_read_lock(&ctrl->srcu); + list_for_each_entry_srcu(ns, &ctrl->namespaces, list, + srcu_read_lock_held(&ctrl->srcu)) blk_mq_freeze_queue_wait(ns->queue); - up_read(&ctrl->namespaces_rwsem); + srcu_read_unlock(&ctrl->srcu, srcu_idx); } EXPORT_SYMBOL_GPL(nvme_wait_freeze); void nvme_start_freeze(struct nvme_ctrl *ctrl) { struct nvme_ns *ns; + int srcu_idx; - down_read(&ctrl->namespaces_rwsem); - list_for_each_entry(ns, &ctrl->namespaces, list) - blk_freeze_queue_start(ns->queue); - up_read(&ctrl->namespaces_rwsem); + set_bit(NVME_CTRL_FROZEN, &ctrl->flags); + srcu_idx = srcu_read_lock(&ctrl->srcu); + list_for_each_entry_srcu(ns, &ctrl->namespaces, list, + srcu_read_lock_held(&ctrl->srcu)) + /* + * Typical non_owner use case is from pci driver, in which + * start_freeze is called from timeout work function, but + * unfreeze is done in reset work context + */ + blk_freeze_queue_start_non_owner(ns->queue); + srcu_read_unlock(&ctrl->srcu, srcu_idx); } EXPORT_SYMBOL_GPL(nvme_start_freeze); @@ -5284,11 +5303,13 @@ EXPORT_SYMBOL_GPL(nvme_unquiesce_admin_queue); void nvme_sync_io_queues(struct nvme_ctrl *ctrl) { struct nvme_ns *ns; + int srcu_idx; - down_read(&ctrl->namespaces_rwsem); - list_for_each_entry(ns, &ctrl->namespaces, list) + srcu_idx = srcu_read_lock(&ctrl->srcu); + list_for_each_entry_srcu(ns, &ctrl->namespaces, list, + srcu_read_lock_held(&ctrl->srcu)) blk_sync_queue(ns->queue); - up_read(&ctrl->namespaces_rwsem); + srcu_read_unlock(&ctrl->srcu, srcu_idx); } EXPORT_SYMBOL_GPL(nvme_sync_io_queues); @@ -5306,7 +5327,7 @@ struct nvme_ctrl *nvme_ctrl_from_file(struct file *file) return NULL; return file->private_data; } -EXPORT_SYMBOL_NS_GPL(nvme_ctrl_from_file, NVME_TARGET_PASSTHRU); +EXPORT_SYMBOL_NS_GPL(nvme_ctrl_from_file, "NVME_TARGET_PASSTHRU"); /* * Check we didn't inadvertently grow the command structure sizes: @@ -5334,6 +5355,8 @@ static inline void _nvme_check_size(void) BUILD_BUG_ON(sizeof(struct nvme_id_ctrl_nvm) != NVME_IDENTIFY_DATA_SIZE); BUILD_BUG_ON(sizeof(struct nvme_lba_range_type) != 64); BUILD_BUG_ON(sizeof(struct nvme_smart_log) != 512); + BUILD_BUG_ON(sizeof(struct nvme_endurance_group_log) != 512); + BUILD_BUG_ON(sizeof(struct nvme_rotational_media_log) != 512); BUILD_BUG_ON(sizeof(struct nvme_dbbuf) != 64); BUILD_BUG_ON(sizeof(struct nvme_directive_cmd) != 64); BUILD_BUG_ON(sizeof(struct nvme_feat_host_behavior) != 512); @@ -5342,22 +5365,20 @@ static inline void _nvme_check_size(void) static int __init nvme_core_init(void) { + unsigned int wq_flags = WQ_UNBOUND | WQ_MEM_RECLAIM | WQ_SYSFS; int result = -ENOMEM; _nvme_check_size(); - nvme_wq = alloc_workqueue("nvme-wq", - WQ_UNBOUND | WQ_MEM_RECLAIM | WQ_SYSFS, 0); + nvme_wq = alloc_workqueue("nvme-wq", wq_flags, 0); if (!nvme_wq) goto out; - nvme_reset_wq = alloc_workqueue("nvme-reset-wq", - WQ_UNBOUND | WQ_MEM_RECLAIM | WQ_SYSFS, 0); + nvme_reset_wq = alloc_workqueue("nvme-reset-wq", wq_flags, 0); if (!nvme_reset_wq) goto destroy_wq; - nvme_delete_wq = alloc_workqueue("nvme-delete-wq", - WQ_UNBOUND | WQ_MEM_RECLAIM | WQ_SYSFS, 0); + nvme_delete_wq = alloc_workqueue("nvme-delete-wq", wq_flags, 0); if (!nvme_delete_wq) goto destroy_reset_wq; @@ -5366,29 +5387,22 @@ static int __init nvme_core_init(void) if (result < 0) goto destroy_delete_wq; - nvme_class = class_create(THIS_MODULE, "nvme"); - if (IS_ERR(nvme_class)) { - result = PTR_ERR(nvme_class); + result = class_register(&nvme_class); + if (result) goto unregister_chrdev; - } - nvme_class->dev_uevent = nvme_class_uevent; - nvme_subsys_class = class_create(THIS_MODULE, "nvme-subsystem"); - if (IS_ERR(nvme_subsys_class)) { - result = PTR_ERR(nvme_subsys_class); + result = class_register(&nvme_subsys_class); + if (result) goto destroy_class; - } result = alloc_chrdev_region(&nvme_ns_chr_devt, 0, NVME_MINORS, "nvme-generic"); if (result < 0) goto destroy_subsys_class; - nvme_ns_chr_class = class_create(THIS_MODULE, "nvme-generic"); - if (IS_ERR(nvme_ns_chr_class)) { - result = PTR_ERR(nvme_ns_chr_class); + result = class_register(&nvme_ns_chr_class); + if (result) goto unregister_generic_ns; - } result = nvme_init_auth(); if (result) @@ -5396,13 +5410,13 @@ static int __init nvme_core_init(void) return 0; destroy_ns_chr: - class_destroy(nvme_ns_chr_class); + class_unregister(&nvme_ns_chr_class); unregister_generic_ns: unregister_chrdev_region(nvme_ns_chr_devt, NVME_MINORS); destroy_subsys_class: - class_destroy(nvme_subsys_class); + class_unregister(&nvme_subsys_class); destroy_class: - class_destroy(nvme_class); + class_unregister(&nvme_class); unregister_chrdev: unregister_chrdev_region(nvme_ctrl_base_chr_devt, NVME_MINORS); destroy_delete_wq: @@ -5418,9 +5432,9 @@ out: static void __exit nvme_core_exit(void) { nvme_exit_auth(); - class_destroy(nvme_ns_chr_class); - class_destroy(nvme_subsys_class); - class_destroy(nvme_class); + class_unregister(&nvme_ns_chr_class); + class_unregister(&nvme_subsys_class); + class_unregister(&nvme_class); unregister_chrdev_region(nvme_ns_chr_devt, NVME_MINORS); unregister_chrdev_region(nvme_ctrl_base_chr_devt, NVME_MINORS); destroy_workqueue(nvme_delete_wq); @@ -5432,5 +5446,6 @@ static void __exit nvme_core_exit(void) MODULE_LICENSE("GPL"); MODULE_VERSION("1.0"); +MODULE_DESCRIPTION("NVMe host core framework"); module_init(nvme_core_init); module_exit(nvme_core_exit); diff --git a/drivers/nvme/host/fabrics.c b/drivers/nvme/host/fabrics.c index ce27276f552d..55a8afd2efd5 100644 --- a/drivers/nvme/host/fabrics.c +++ b/drivers/nvme/host/fabrics.c @@ -12,6 +12,7 @@ #include <linux/seq_file.h> #include "nvme.h" #include "fabrics.h" +#include <linux/nvme-keyring.h> static LIST_HEAD(nvmf_transports); static DECLARE_RWSEM(nvmf_transports_rwsem); @@ -21,35 +22,60 @@ static DEFINE_MUTEX(nvmf_hosts_mutex); static struct nvmf_host *nvmf_default_host; -static struct nvmf_host *__nvmf_host_find(const char *hostnqn) +static struct nvmf_host *nvmf_host_alloc(const char *hostnqn, uuid_t *id) { struct nvmf_host *host; - list_for_each_entry(host, &nvmf_hosts, list) { - if (!strcmp(host->nqn, hostnqn)) - return host; - } + host = kmalloc(sizeof(*host), GFP_KERNEL); + if (!host) + return NULL; - return NULL; + kref_init(&host->ref); + uuid_copy(&host->id, id); + strscpy(host->nqn, hostnqn, NVMF_NQN_SIZE); + + return host; } -static struct nvmf_host *nvmf_host_add(const char *hostnqn) +static struct nvmf_host *nvmf_host_add(const char *hostnqn, uuid_t *id) { struct nvmf_host *host; mutex_lock(&nvmf_hosts_mutex); - host = __nvmf_host_find(hostnqn); - if (host) { - kref_get(&host->ref); - goto out_unlock; + + /* + * We have defined a host as how it is perceived by the target. + * Therefore, we don't allow different Host NQNs with the same Host ID. + * Similarly, we do not allow the usage of the same Host NQN with + * different Host IDs. This'll maintain unambiguous host identification. + */ + list_for_each_entry(host, &nvmf_hosts, list) { + bool same_hostnqn = !strcmp(host->nqn, hostnqn); + bool same_hostid = uuid_equal(&host->id, id); + + if (same_hostnqn && same_hostid) { + kref_get(&host->ref); + goto out_unlock; + } + if (same_hostnqn) { + pr_err("found same hostnqn %s but different hostid %pUb\n", + hostnqn, id); + host = ERR_PTR(-EINVAL); + goto out_unlock; + } + if (same_hostid) { + pr_err("found same hostid %pUb but different hostnqn %s\n", + id, hostnqn); + host = ERR_PTR(-EINVAL); + goto out_unlock; + } } - host = kmalloc(sizeof(*host), GFP_KERNEL); - if (!host) + host = nvmf_host_alloc(hostnqn, id); + if (!host) { + host = ERR_PTR(-ENOMEM); goto out_unlock; - - kref_init(&host->ref); - strscpy(host->nqn, hostnqn, NVMF_NQN_SIZE); + } list_add_tail(&host->list, &nvmf_hosts); out_unlock: @@ -60,16 +86,17 @@ out_unlock: static struct nvmf_host *nvmf_host_default(void) { struct nvmf_host *host; + char nqn[NVMF_NQN_SIZE]; + uuid_t id; - host = kmalloc(sizeof(*host), GFP_KERNEL); + uuid_gen(&id); + snprintf(nqn, NVMF_NQN_SIZE, + "nqn.2014-08.org.nvmexpress:uuid:%pUb", &id); + + host = nvmf_host_alloc(nqn, &id); if (!host) return NULL; - kref_init(&host->ref); - uuid_gen(&host->id); - snprintf(host->nqn, NVMF_NQN_SIZE, - "nqn.2014-08.org.nvmexpress:uuid:%pUb", &host->id); - mutex_lock(&nvmf_hosts_mutex); list_add_tail(&host->list, &nvmf_hosts); mutex_unlock(&nvmf_hosts_mutex); @@ -153,14 +180,14 @@ int nvmf_reg_read32(struct nvme_ctrl *ctrl, u32 off, u32 *val) cmd.prop_get.offset = cpu_to_le32(off); ret = __nvme_submit_sync_cmd(ctrl->fabrics_q, &cmd, &res, NULL, 0, - NVME_QID_ANY, 0, 0); + NVME_QID_ANY, NVME_SUBMIT_RESERVED); if (ret >= 0) *val = le64_to_cpu(res.u64); if (unlikely(ret != 0)) dev_err(ctrl->device, "Property Get error: %d, offset %#x\n", - ret > 0 ? ret & ~NVME_SC_DNR : ret, off); + ret > 0 ? ret & ~NVME_STATUS_DNR : ret, off); return ret; } @@ -199,14 +226,14 @@ int nvmf_reg_read64(struct nvme_ctrl *ctrl, u32 off, u64 *val) cmd.prop_get.offset = cpu_to_le32(off); ret = __nvme_submit_sync_cmd(ctrl->fabrics_q, &cmd, &res, NULL, 0, - NVME_QID_ANY, 0, 0); + NVME_QID_ANY, NVME_SUBMIT_RESERVED); if (ret >= 0) *val = le64_to_cpu(res.u64); if (unlikely(ret != 0)) dev_err(ctrl->device, "Property Get error: %d, offset %#x\n", - ret > 0 ? ret & ~NVME_SC_DNR : ret, off); + ret > 0 ? ret & ~NVME_STATUS_DNR : ret, off); return ret; } EXPORT_SYMBOL_GPL(nvmf_reg_read64); @@ -244,15 +271,30 @@ int nvmf_reg_write32(struct nvme_ctrl *ctrl, u32 off, u32 val) cmd.prop_set.value = cpu_to_le64(val); ret = __nvme_submit_sync_cmd(ctrl->fabrics_q, &cmd, NULL, NULL, 0, - NVME_QID_ANY, 0, 0); + NVME_QID_ANY, NVME_SUBMIT_RESERVED); if (unlikely(ret)) dev_err(ctrl->device, "Property Set error: %d, offset %#x\n", - ret > 0 ? ret & ~NVME_SC_DNR : ret, off); + ret > 0 ? ret & ~NVME_STATUS_DNR : ret, off); return ret; } EXPORT_SYMBOL_GPL(nvmf_reg_write32); +int nvmf_subsystem_reset(struct nvme_ctrl *ctrl) +{ + int ret; + + if (!nvme_wait_reset(ctrl)) + return -EBUSY; + + ret = ctrl->ops->reg_write32(ctrl, NVME_REG_NSSR, NVME_SUBSYS_RESET); + if (ret) + return ret; + + return nvme_try_sched_reset(ctrl); +} +EXPORT_SYMBOL_GPL(nvmf_subsystem_reset); + /** * nvmf_log_connect_error() - Error-parsing-diagnostic print out function for * connect() errors. @@ -268,7 +310,7 @@ static void nvmf_log_connect_error(struct nvme_ctrl *ctrl, int errval, int offset, struct nvme_command *cmd, struct nvmf_connect_data *data) { - int err_sctype = errval & ~NVME_SC_DNR; + int err_sctype = errval & ~NVME_STATUS_DNR; if (errval < 0) { dev_err(ctrl->device, @@ -349,6 +391,45 @@ static void nvmf_log_connect_error(struct nvme_ctrl *ctrl, } } +static struct nvmf_connect_data *nvmf_connect_data_prep(struct nvme_ctrl *ctrl, + u16 cntlid) +{ + struct nvmf_connect_data *data; + + data = kzalloc(sizeof(*data), GFP_KERNEL); + if (!data) + return NULL; + + uuid_copy(&data->hostid, &ctrl->opts->host->id); + data->cntlid = cpu_to_le16(cntlid); + strscpy(data->subsysnqn, ctrl->opts->subsysnqn, NVMF_NQN_SIZE); + strscpy(data->hostnqn, ctrl->opts->host->nqn, NVMF_NQN_SIZE); + + return data; +} + +static void nvmf_connect_cmd_prep(struct nvme_ctrl *ctrl, u16 qid, + struct nvme_command *cmd) +{ + cmd->connect.opcode = nvme_fabrics_command; + cmd->connect.fctype = nvme_fabrics_type_connect; + cmd->connect.qid = cpu_to_le16(qid); + + if (qid) { + cmd->connect.sqsize = cpu_to_le16(ctrl->sqsize); + } else { + cmd->connect.sqsize = cpu_to_le16(NVME_AQ_DEPTH - 1); + + /* + * set keep-alive timeout in seconds granularity (ms * 1000) + */ + cmd->connect.kato = cpu_to_le32(ctrl->kato * 1000); + } + + if (ctrl->opts->disable_sqflow) + cmd->connect.cattr |= NVME_CONNECT_DISABLE_SQFLOW; +} + /** * nvmf_connect_admin_queue() - NVMe Fabrics Admin Queue "Connect" * API function. @@ -362,12 +443,6 @@ static void nvmf_log_connect_error(struct nvme_ctrl *ctrl, * fabrics-protocol connection of the NVMe Admin queue between the * host system device and the allocated NVMe controller on the * target system via a NVMe Fabrics "Connect" command. - * - * Return: - * 0: success - * > 0: NVMe error status code - * < 0: Linux errno error code - * */ int nvmf_connect_admin_queue(struct nvme_ctrl *ctrl) { @@ -377,31 +452,17 @@ int nvmf_connect_admin_queue(struct nvme_ctrl *ctrl) int ret; u32 result; - cmd.connect.opcode = nvme_fabrics_command; - cmd.connect.fctype = nvme_fabrics_type_connect; - cmd.connect.qid = 0; - cmd.connect.sqsize = cpu_to_le16(NVME_AQ_DEPTH - 1); - - /* - * Set keep-alive timeout in seconds granularity (ms * 1000) - */ - cmd.connect.kato = cpu_to_le32(ctrl->kato * 1000); - - if (ctrl->opts->disable_sqflow) - cmd.connect.cattr |= NVME_CONNECT_DISABLE_SQFLOW; + nvmf_connect_cmd_prep(ctrl, 0, &cmd); - data = kzalloc(sizeof(*data), GFP_KERNEL); + data = nvmf_connect_data_prep(ctrl, 0xffff); if (!data) return -ENOMEM; - uuid_copy(&data->hostid, &ctrl->opts->host->id); - data->cntlid = cpu_to_le16(0xffff); - strncpy(data->subsysnqn, ctrl->opts->subsysnqn, NVMF_NQN_SIZE); - strncpy(data->hostnqn, ctrl->opts->host->nqn, NVMF_NQN_SIZE); - ret = __nvme_submit_sync_cmd(ctrl->fabrics_q, &cmd, &res, - data, sizeof(*data), NVME_QID_ANY, 1, - BLK_MQ_REQ_RESERVED | BLK_MQ_REQ_NOWAIT); + data, sizeof(*data), NVME_QID_ANY, + NVME_SUBMIT_AT_HEAD | + NVME_SUBMIT_NOWAIT | + NVME_SUBMIT_RESERVED); if (ret) { nvmf_log_connect_error(ctrl, ret, le32_to_cpu(res.u32), &cmd, data); @@ -410,20 +471,28 @@ int nvmf_connect_admin_queue(struct nvme_ctrl *ctrl) result = le32_to_cpu(res.u32); ctrl->cntlid = result & 0xFFFF; - if ((result >> 16) & 0x3) { + if (result & (NVME_CONNECT_AUTHREQ_ATR | NVME_CONNECT_AUTHREQ_ASCR)) { + /* Check for secure concatenation */ + if ((result & NVME_CONNECT_AUTHREQ_ASCR) && + !ctrl->opts->concat) { + dev_warn(ctrl->device, + "qid 0: secure concatenation is not supported\n"); + ret = -EOPNOTSUPP; + goto out_free_data; + } /* Authentication required */ ret = nvme_auth_negotiate(ctrl, 0); if (ret) { dev_warn(ctrl->device, "qid 0: authentication setup failed\n"); - ret = NVME_SC_AUTH_REQUIRED; goto out_free_data; } ret = nvme_auth_wait(ctrl, 0); - if (ret) + if (ret) { dev_warn(ctrl->device, - "qid 0: authentication failed\n"); - else + "qid 0: authentication failed, error %d\n", + ret); + } else dev_info(ctrl->device, "qid 0: authenticated\n"); } @@ -461,52 +530,71 @@ int nvmf_connect_io_queue(struct nvme_ctrl *ctrl, u16 qid) int ret; u32 result; - cmd.connect.opcode = nvme_fabrics_command; - cmd.connect.fctype = nvme_fabrics_type_connect; - cmd.connect.qid = cpu_to_le16(qid); - cmd.connect.sqsize = cpu_to_le16(ctrl->sqsize); + nvmf_connect_cmd_prep(ctrl, qid, &cmd); - if (ctrl->opts->disable_sqflow) - cmd.connect.cattr |= NVME_CONNECT_DISABLE_SQFLOW; - - data = kzalloc(sizeof(*data), GFP_KERNEL); + data = nvmf_connect_data_prep(ctrl, ctrl->cntlid); if (!data) return -ENOMEM; - uuid_copy(&data->hostid, &ctrl->opts->host->id); - data->cntlid = cpu_to_le16(ctrl->cntlid); - strncpy(data->subsysnqn, ctrl->opts->subsysnqn, NVMF_NQN_SIZE); - strncpy(data->hostnqn, ctrl->opts->host->nqn, NVMF_NQN_SIZE); - ret = __nvme_submit_sync_cmd(ctrl->connect_q, &cmd, &res, - data, sizeof(*data), qid, 1, - BLK_MQ_REQ_RESERVED | BLK_MQ_REQ_NOWAIT); + data, sizeof(*data), qid, + NVME_SUBMIT_AT_HEAD | + NVME_SUBMIT_RESERVED | + NVME_SUBMIT_NOWAIT); if (ret) { nvmf_log_connect_error(ctrl, ret, le32_to_cpu(res.u32), &cmd, data); + goto out_free_data; } result = le32_to_cpu(res.u32); - if ((result >> 16) & 2) { + if (result & (NVME_CONNECT_AUTHREQ_ATR | NVME_CONNECT_AUTHREQ_ASCR)) { + /* Secure concatenation is not implemented */ + if (result & NVME_CONNECT_AUTHREQ_ASCR) { + dev_warn(ctrl->device, + "qid %d: secure concatenation is not supported\n", qid); + ret = -EOPNOTSUPP; + goto out_free_data; + } /* Authentication required */ ret = nvme_auth_negotiate(ctrl, qid); if (ret) { dev_warn(ctrl->device, "qid %d: authentication setup failed\n", qid); - ret = NVME_SC_AUTH_REQUIRED; - } else { - ret = nvme_auth_wait(ctrl, qid); - if (ret) - dev_warn(ctrl->device, - "qid %u: authentication failed\n", qid); + goto out_free_data; + } + ret = nvme_auth_wait(ctrl, qid); + if (ret) { + dev_warn(ctrl->device, + "qid %u: authentication failed, error %d\n", + qid, ret); } } +out_free_data: kfree(data); return ret; } EXPORT_SYMBOL_GPL(nvmf_connect_io_queue); -bool nvmf_should_reconnect(struct nvme_ctrl *ctrl) +/* + * Evaluate the status information returned by the transport in order to decided + * if a reconnect attempt should be scheduled. + * + * Do not retry when: + * + * - the DNR bit is set and the specification states no further connect + * attempts with the same set of parameters should be attempted. + * + * - when the authentication attempt fails, because the key was invalid. + * This error code is set on the host side. + */ +bool nvmf_should_reconnect(struct nvme_ctrl *ctrl, int status) { + if (status > 0 && (status & NVME_STATUS_DNR)) + return false; + + if (status == -EKEYREJECTED || status == -ENOKEY) + return false; + if (ctrl->opts->max_reconnects == -1 || ctrl->nr_reconnects < ctrl->opts->max_reconnects) return true; @@ -569,6 +657,23 @@ static struct nvmf_transport_ops *nvmf_lookup_transport( return NULL; } +static struct key *nvmf_parse_key(int key_id) +{ + struct key *key; + + if (!IS_ENABLED(CONFIG_NVME_TCP_TLS)) { + pr_err("TLS is not supported\n"); + return ERR_PTR(-EINVAL); + } + + key = nvme_tls_key_lookup(key_id); + if (IS_ERR(key)) + pr_err("key id %08x not found\n", key_id); + else + pr_debug("Using key id %08x\n", key_id); + return key; +} + static const match_table_t opt_tokens = { { NVMF_OPT_TRANSPORT, "transport=%s" }, { NVMF_OPT_TRADDR, "traddr=%s" }, @@ -590,10 +695,20 @@ static const match_table_t opt_tokens = { { NVMF_OPT_NR_WRITE_QUEUES, "nr_write_queues=%d" }, { NVMF_OPT_NR_POLL_QUEUES, "nr_poll_queues=%d" }, { NVMF_OPT_TOS, "tos=%d" }, +#ifdef CONFIG_NVME_TCP_TLS + { NVMF_OPT_KEYRING, "keyring=%d" }, + { NVMF_OPT_TLS_KEY, "tls_key=%d" }, +#endif { NVMF_OPT_FAIL_FAST_TMO, "fast_io_fail_tmo=%d" }, { NVMF_OPT_DISCOVERY, "discovery" }, +#ifdef CONFIG_NVME_HOST_AUTH { NVMF_OPT_DHCHAP_SECRET, "dhchap_secret=%s" }, { NVMF_OPT_DHCHAP_CTRL_SECRET, "dhchap_ctrl_secret=%s" }, +#endif +#ifdef CONFIG_NVME_TCP_TLS + { NVMF_OPT_TLS, "tls" }, + { NVMF_OPT_CONCAT, "concat" }, +#endif { NVMF_OPT_ERR, NULL } }; @@ -604,8 +719,10 @@ static int nvmf_parse_options(struct nvmf_ctrl_options *opts, char *options, *o, *p; int token, ret = 0; size_t nqnlen = 0; - int ctrl_loss_tmo = NVMF_DEF_CTRL_LOSS_TMO; + int ctrl_loss_tmo = NVMF_DEF_CTRL_LOSS_TMO, key_id; uuid_t hostid; + char hostnqn[NVMF_NQN_SIZE]; + struct key *key; /* Set defaults */ opts->queue_size = NVMF_DEF_QUEUE_SIZE; @@ -617,12 +734,18 @@ static int nvmf_parse_options(struct nvmf_ctrl_options *opts, opts->hdr_digest = false; opts->data_digest = false; opts->tos = -1; /* < 0 == use transport default */ + opts->tls = false; + opts->tls_key = NULL; + opts->keyring = NULL; + opts->concat = false; options = o = kstrdup(buf, GFP_KERNEL); if (!options) return -ENOMEM; - uuid_gen(&hostid); + /* use default host if not given by user space */ + uuid_copy(&hostid, &nvmf_default_host->id); + strscpy(hostnqn, nvmf_default_host->nqn, NVMF_NQN_SIZE); while ((p = strsep(&o, ",\n")) != NULL) { if (!*p) @@ -768,12 +891,8 @@ static int nvmf_parse_options(struct nvmf_ctrl_options *opts, ret = -EINVAL; goto out; } - opts->host = nvmf_host_add(p); + strscpy(hostnqn, p, NVMF_NQN_SIZE); kfree(p); - if (!opts->host) { - ret = -ENOMEM; - goto out; - } break; case NVMF_OPT_RECONNECT_DELAY: if (match_int(args, &token)) { @@ -872,6 +991,32 @@ static int nvmf_parse_options(struct nvmf_ctrl_options *opts, } opts->tos = token; break; + case NVMF_OPT_KEYRING: + if (match_int(args, &key_id) || key_id <= 0) { + ret = -EINVAL; + goto out; + } + key = nvmf_parse_key(key_id); + if (IS_ERR(key)) { + ret = PTR_ERR(key); + goto out; + } + key_put(opts->keyring); + opts->keyring = key; + break; + case NVMF_OPT_TLS_KEY: + if (match_int(args, &key_id) || key_id <= 0) { + ret = -EINVAL; + goto out; + } + key = nvmf_parse_key(key_id); + if (IS_ERR(key)) { + ret = PTR_ERR(key); + goto out; + } + key_put(opts->tls_key); + opts->tls_key = key; + break; case NVMF_OPT_DISCOVERY: opts->discovery_nqn = true; break; @@ -903,6 +1048,22 @@ static int nvmf_parse_options(struct nvmf_ctrl_options *opts, kfree(opts->dhchap_ctrl_secret); opts->dhchap_ctrl_secret = p; break; + case NVMF_OPT_TLS: + if (!IS_ENABLED(CONFIG_NVME_TCP_TLS)) { + pr_err("TLS is not supported\n"); + ret = -EINVAL; + goto out; + } + opts->tls = true; + break; + case NVMF_OPT_CONCAT: + if (!IS_ENABLED(CONFIG_NVME_TCP_TLS)) { + pr_err("TLS is not supported\n"); + ret = -EINVAL; + goto out; + } + opts->concat = true; + break; default: pr_warn("unknown parameter or missing value '%s' in ctrl creation request\n", p); @@ -929,19 +1090,112 @@ static int nvmf_parse_options(struct nvmf_ctrl_options *opts, pr_warn("failfast tmo (%d) larger than controller loss tmo (%d)\n", opts->fast_io_fail_tmo, ctrl_loss_tmo); } - - if (!opts->host) { - kref_get(&nvmf_default_host->ref); - opts->host = nvmf_default_host; + if (opts->concat) { + if (opts->tls) { + pr_err("Secure concatenation over TLS is not supported\n"); + ret = -EINVAL; + goto out; + } + if (opts->tls_key) { + pr_err("Cannot specify a TLS key for secure concatenation\n"); + ret = -EINVAL; + goto out; + } + if (!opts->dhchap_secret) { + pr_err("Need to enable DH-CHAP for secure concatenation\n"); + ret = -EINVAL; + goto out; + } } - uuid_copy(&opts->host->id, &hostid); + opts->host = nvmf_host_add(hostnqn, &hostid); + if (IS_ERR(opts->host)) { + ret = PTR_ERR(opts->host); + opts->host = NULL; + goto out; + } out: kfree(options); return ret; } +void nvmf_set_io_queues(struct nvmf_ctrl_options *opts, u32 nr_io_queues, + u32 io_queues[HCTX_MAX_TYPES]) +{ + if (opts->nr_write_queues && opts->nr_io_queues < nr_io_queues) { + /* + * separate read/write queues + * hand out dedicated default queues only after we have + * sufficient read queues. + */ + io_queues[HCTX_TYPE_READ] = opts->nr_io_queues; + nr_io_queues -= io_queues[HCTX_TYPE_READ]; + io_queues[HCTX_TYPE_DEFAULT] = + min(opts->nr_write_queues, nr_io_queues); + nr_io_queues -= io_queues[HCTX_TYPE_DEFAULT]; + } else { + /* + * shared read/write queues + * either no write queues were requested, or we don't have + * sufficient queue count to have dedicated default queues. + */ + io_queues[HCTX_TYPE_DEFAULT] = + min(opts->nr_io_queues, nr_io_queues); + nr_io_queues -= io_queues[HCTX_TYPE_DEFAULT]; + } + + if (opts->nr_poll_queues && nr_io_queues) { + /* map dedicated poll queues only if we have queues left */ + io_queues[HCTX_TYPE_POLL] = + min(opts->nr_poll_queues, nr_io_queues); + } +} +EXPORT_SYMBOL_GPL(nvmf_set_io_queues); + +void nvmf_map_queues(struct blk_mq_tag_set *set, struct nvme_ctrl *ctrl, + u32 io_queues[HCTX_MAX_TYPES]) +{ + struct nvmf_ctrl_options *opts = ctrl->opts; + + if (opts->nr_write_queues && io_queues[HCTX_TYPE_READ]) { + /* separate read/write queues */ + set->map[HCTX_TYPE_DEFAULT].nr_queues = + io_queues[HCTX_TYPE_DEFAULT]; + set->map[HCTX_TYPE_DEFAULT].queue_offset = 0; + set->map[HCTX_TYPE_READ].nr_queues = + io_queues[HCTX_TYPE_READ]; + set->map[HCTX_TYPE_READ].queue_offset = + io_queues[HCTX_TYPE_DEFAULT]; + } else { + /* shared read/write queues */ + set->map[HCTX_TYPE_DEFAULT].nr_queues = + io_queues[HCTX_TYPE_DEFAULT]; + set->map[HCTX_TYPE_DEFAULT].queue_offset = 0; + set->map[HCTX_TYPE_READ].nr_queues = + io_queues[HCTX_TYPE_DEFAULT]; + set->map[HCTX_TYPE_READ].queue_offset = 0; + } + + blk_mq_map_queues(&set->map[HCTX_TYPE_DEFAULT]); + blk_mq_map_queues(&set->map[HCTX_TYPE_READ]); + if (opts->nr_poll_queues && io_queues[HCTX_TYPE_POLL]) { + /* map dedicated poll queues only if we have queues left */ + set->map[HCTX_TYPE_POLL].nr_queues = io_queues[HCTX_TYPE_POLL]; + set->map[HCTX_TYPE_POLL].queue_offset = + io_queues[HCTX_TYPE_DEFAULT] + + io_queues[HCTX_TYPE_READ]; + blk_mq_map_queues(&set->map[HCTX_TYPE_POLL]); + } + + dev_info(ctrl->device, + "mapped %d/%d/%d default/read/poll queues.\n", + io_queues[HCTX_TYPE_DEFAULT], + io_queues[HCTX_TYPE_READ], + io_queues[HCTX_TYPE_POLL]); +} +EXPORT_SYMBOL_GPL(nvmf_map_queues); + static int nvmf_check_required_opts(struct nvmf_ctrl_options *opts, unsigned int required_opts) { @@ -1028,6 +1282,8 @@ static int nvmf_check_allowed_opts(struct nvmf_ctrl_options *opts, void nvmf_free_options(struct nvmf_ctrl_options *opts) { nvmf_host_put(opts->host); + key_put(opts->keyring); + key_put(opts->tls_key); kfree(opts->transport); kfree(opts->traddr); kfree(opts->trsvcid); @@ -1119,7 +1375,10 @@ out_free_opts: return ERR_PTR(ret); } -static struct class *nvmf_class; +static const struct class nvmf_class = { + .name = "nvme-fabrics", +}; + static struct device *nvmf_device; static DEFINE_MUTEX(nvmf_dev_mutex); @@ -1172,10 +1431,10 @@ static void __nvmf_concat_opt_tokens(struct seq_file *seq_file) tok = &opt_tokens[idx]; if (tok->token == NVMF_OPT_ERR) continue; - seq_puts(seq_file, ","); + seq_putc(seq_file, ','); seq_puts(seq_file, tok->pattern); } - seq_puts(seq_file, "\n"); + seq_putc(seq_file, '\n'); } static int nvmf_dev_show(struct seq_file *seq_file, void *private) @@ -1239,15 +1498,14 @@ static int __init nvmf_init(void) if (!nvmf_default_host) return -ENOMEM; - nvmf_class = class_create(THIS_MODULE, "nvme-fabrics"); - if (IS_ERR(nvmf_class)) { + ret = class_register(&nvmf_class); + if (ret) { pr_err("couldn't register class nvme-fabrics\n"); - ret = PTR_ERR(nvmf_class); goto out_free_host; } nvmf_device = - device_create(nvmf_class, NULL, MKDEV(0, 0), NULL, "ctl"); + device_create(&nvmf_class, NULL, MKDEV(0, 0), NULL, "ctl"); if (IS_ERR(nvmf_device)) { pr_err("couldn't create nvme-fabrics device!\n"); ret = PTR_ERR(nvmf_device); @@ -1263,9 +1521,9 @@ static int __init nvmf_init(void) return 0; out_destroy_device: - device_destroy(nvmf_class, MKDEV(0, 0)); + device_destroy(&nvmf_class, MKDEV(0, 0)); out_destroy_class: - class_destroy(nvmf_class); + class_unregister(&nvmf_class); out_free_host: nvmf_host_put(nvmf_default_host); return ret; @@ -1274,8 +1532,8 @@ out_free_host: static void __exit nvmf_exit(void) { misc_deregister(&nvmf_misc); - device_destroy(nvmf_class, MKDEV(0, 0)); - class_destroy(nvmf_class); + device_destroy(&nvmf_class, MKDEV(0, 0)); + class_unregister(&nvmf_class); nvmf_host_put(nvmf_default_host); BUILD_BUG_ON(sizeof(struct nvmf_common_command) != 64); @@ -1293,6 +1551,7 @@ static void __exit nvmf_exit(void) } MODULE_LICENSE("GPL v2"); +MODULE_DESCRIPTION("NVMe host fabrics library"); module_init(nvmf_init); module_exit(nvmf_exit); diff --git a/drivers/nvme/host/fabrics.h b/drivers/nvme/host/fabrics.h index a6e22116e139..caf5503d0833 100644 --- a/drivers/nvme/host/fabrics.h +++ b/drivers/nvme/host/fabrics.h @@ -19,13 +19,6 @@ #define NVMF_DEF_FAIL_FAST_TMO -1 /* - * Reserved one command for internal usage. This command is used for sending - * the connect command, as well as for the keep alive command on the admin - * queue once live. - */ -#define NVMF_RESERVED_TAGS 1 - -/* * Define a host as seen by the target. We allocate one at boot, but also * allow the override it when creating controllers. This is both to provide * persistence of the Host NQN over multiple boots, and to allow using @@ -70,6 +63,10 @@ enum { NVMF_OPT_DISCOVERY = 1 << 22, NVMF_OPT_DHCHAP_SECRET = 1 << 23, NVMF_OPT_DHCHAP_CTRL_SECRET = 1 << 24, + NVMF_OPT_TLS = 1 << 25, + NVMF_OPT_KEYRING = 1 << 26, + NVMF_OPT_TLS_KEY = 1 << 27, + NVMF_OPT_CONCAT = 1 << 28, }; /** @@ -77,10 +74,13 @@ enum { * with the parsing opts enum. * @mask: Used by the fabrics library to parse through sysfs options * on adding a NVMe controller. + * @max_reconnects: maximum number of allowed reconnect attempts before removing + * the controller, (-1) means reconnect forever, zero means remove + * immediately; * @transport: Holds the fabric transport "technology name" (for a lack of * better description) that will be used by an NVMe controller * being added. - * @subsysnqn: Hold the fully qualified NQN subystem name (format defined + * @subsysnqn: Hold the fully qualified NQN subsystem name (format defined * in the NVMe specification, "NVMe Qualified Names"). * @traddr: The transport-specific TRADDR field for a port on the * subsystem which is adding a controller. @@ -96,12 +96,13 @@ enum { * @discovery_nqn: indicates if the subsysnqn is the well-known discovery NQN. * @kato: Keep-alive timeout. * @host: Virtual NVMe host, contains the NQN and Host ID. - * @max_reconnects: maximum number of allowed reconnect attempts before removing - * the controller, (-1) means reconnect forever, zero means remove - * immediately; * @dhchap_secret: DH-HMAC-CHAP secret * @dhchap_ctrl_secret: DH-HMAC-CHAP controller secret for bi-directional * authentication + * @keyring: Keyring to use for key lookups + * @tls_key: TLS key for encrypted connections (TCP) + * @tls: Start TLS encrypted connections (TCP) + * @concat: Enabled Secure channel concatenation (TCP) * @disable_sqflow: disable controller sq flow control * @hdr_digest: generate/verify header digest (TCP) * @data_digest: generate/verify data digest (TCP) @@ -112,6 +113,7 @@ enum { */ struct nvmf_ctrl_options { unsigned mask; + int max_reconnects; char *transport; char *subsysnqn; char *traddr; @@ -125,9 +127,12 @@ struct nvmf_ctrl_options { bool duplicate_connect; unsigned int kato; struct nvmf_host *host; - int max_reconnects; char *dhchap_secret; char *dhchap_ctrl_secret; + struct key *keyring; + struct key *tls_key; + bool tls; + bool concat; bool disable_sqflow; bool hdr_digest; bool data_digest; @@ -151,7 +156,7 @@ struct nvmf_ctrl_options { * @create_ctrl(): function pointer that points to a non-NVMe * implementation-specific fabric technology * that would go into starting up that fabric - * for the purpose of conneciton to an NVMe controller + * for the purpose of connection to an NVMe controller * using that fabric technology. * * Notes: @@ -160,7 +165,7 @@ struct nvmf_ctrl_options { * 2. create_ctrl() must be defined (even if it does nothing) * 3. struct nvmf_transport_ops must be statically allocated in the * modules .bss section so that a pure module_get on @module - * prevents the memory from beeing freed. + * prevents the memory from being freed. */ struct nvmf_transport_ops { struct list_head entry; @@ -176,12 +181,14 @@ static inline bool nvmf_ctlr_matches_baseopts(struct nvme_ctrl *ctrl, struct nvmf_ctrl_options *opts) { - if (ctrl->state == NVME_CTRL_DELETING || - ctrl->state == NVME_CTRL_DELETING_NOIO || - ctrl->state == NVME_CTRL_DEAD || + enum nvme_ctrl_state state = nvme_ctrl_state(ctrl); + + if (state == NVME_CTRL_DELETING || + state == NVME_CTRL_DELETING_NOIO || + state == NVME_CTRL_DEAD || strcmp(opts->subsysnqn, ctrl->opts->subsysnqn) || strcmp(opts->host->nqn, ctrl->opts->host->nqn) || - memcmp(&opts->host->id, &ctrl->opts->host->id, sizeof(uuid_t))) + !uuid_equal(&opts->host->id, &ctrl->opts->host->id)) return false; return true; @@ -189,7 +196,8 @@ nvmf_ctlr_matches_baseopts(struct nvme_ctrl *ctrl, static inline char *nvmf_ctrl_subsysnqn(struct nvme_ctrl *ctrl) { - if (!ctrl->subsys) + if (!ctrl->subsys || + !strcmp(ctrl->opts->subsysnqn, NVME_DISC_SUBSYS_NAME)) return ctrl->opts->subsysnqn; return ctrl->subsys->subnqn; } @@ -202,17 +210,35 @@ static inline void nvmf_complete_timed_out_request(struct request *rq) } } +static inline unsigned int nvmf_nr_io_queues(struct nvmf_ctrl_options *opts) +{ + return min(opts->nr_io_queues, num_online_cpus()) + + min(opts->nr_write_queues, num_online_cpus()) + + min(opts->nr_poll_queues, num_online_cpus()); +} + +static inline unsigned long nvmf_get_virt_boundary(struct nvme_ctrl *ctrl, + bool is_admin) +{ + return 0; +} + int nvmf_reg_read32(struct nvme_ctrl *ctrl, u32 off, u32 *val); int nvmf_reg_read64(struct nvme_ctrl *ctrl, u32 off, u64 *val); int nvmf_reg_write32(struct nvme_ctrl *ctrl, u32 off, u32 val); +int nvmf_subsystem_reset(struct nvme_ctrl *ctrl); int nvmf_connect_admin_queue(struct nvme_ctrl *ctrl); int nvmf_connect_io_queue(struct nvme_ctrl *ctrl, u16 qid); int nvmf_register_transport(struct nvmf_transport_ops *ops); void nvmf_unregister_transport(struct nvmf_transport_ops *ops); void nvmf_free_options(struct nvmf_ctrl_options *opts); int nvmf_get_address(struct nvme_ctrl *ctrl, char *buf, int size); -bool nvmf_should_reconnect(struct nvme_ctrl *ctrl); +bool nvmf_should_reconnect(struct nvme_ctrl *ctrl, int status); bool nvmf_ip_options_match(struct nvme_ctrl *ctrl, struct nvmf_ctrl_options *opts); +void nvmf_set_io_queues(struct nvmf_ctrl_options *opts, u32 nr_io_queues, + u32 io_queues[HCTX_MAX_TYPES]); +void nvmf_map_queues(struct blk_mq_tag_set *set, struct nvme_ctrl *ctrl, + u32 io_queues[HCTX_MAX_TYPES]); #endif /* _NVME_FABRICS_H */ diff --git a/drivers/nvme/host/fault_inject.c b/drivers/nvme/host/fault_inject.c index 83d2e6860d38..105d6cb41c72 100644 --- a/drivers/nvme/host/fault_inject.c +++ b/drivers/nvme/host/fault_inject.c @@ -6,6 +6,7 @@ */ #include <linux/moduleparam.h> +#include <linux/debugfs.h> #include "nvme.h" static DECLARE_FAULT_ATTR(fail_default_attr); @@ -27,7 +28,7 @@ void nvme_fault_inject_init(struct nvme_fault_inject *fault_inj, /* create debugfs directory and attribute */ parent = debugfs_create_dir(dev_name, NULL); - if (!parent) { + if (IS_ERR(parent)) { pr_warn("%s: failed to create debugfs directory\n", dev_name); return; } @@ -75,7 +76,7 @@ void nvme_should_fail(struct request *req) /* inject status code and DNR bit */ status = fault_inject->status; if (fault_inject->dont_retry) - status |= NVME_SC_DNR; + status |= NVME_STATUS_DNR; nvme_req(req)->status = status; } } diff --git a/drivers/nvme/host/fc.c b/drivers/nvme/host/fc.c index 456ee42a6133..bc455fa98246 100644 --- a/drivers/nvme/host/fc.c +++ b/drivers/nvme/host/fc.c @@ -16,7 +16,6 @@ #include <linux/nvme-fc.h> #include "fc.h" #include <scsi/scsi_transport_fc.h> -#include <linux/blk-mq-pci.h> /* *************************** Data Structures/Defines ****************** */ @@ -221,11 +220,6 @@ static LIST_HEAD(nvme_fc_lport_list); static DEFINE_IDA(nvme_fc_local_port_cnt); static DEFINE_IDA(nvme_fc_ctrl_cnt); -static struct workqueue_struct *nvme_fc_wq; - -static bool nvme_fc_waiting_to_unload; -static DECLARE_COMPLETION(nvme_fc_unload_proceed); - /* * These items are short-term. They will eventually be moved into * a generic FC class. See comments in module init. @@ -255,8 +249,6 @@ nvme_fc_free_lport(struct kref *ref) /* remove from transport list */ spin_lock_irqsave(&nvme_fc_lock, flags); list_del(&lport->port_list); - if (nvme_fc_waiting_to_unload && list_empty(&nvme_fc_lport_list)) - complete(&nvme_fc_unload_proceed); spin_unlock_irqrestore(&nvme_fc_lock, flags); ida_free(&nvme_fc_local_port_cnt, lport->localport.port_num); @@ -528,6 +520,8 @@ nvme_fc_free_rport(struct kref *ref) WARN_ON(rport->remoteport.port_state != FC_OBJSTATE_DELETED); WARN_ON(!list_empty(&rport->ctrl_list)); + WARN_ON(!list_empty(&rport->ls_req_list)); + WARN_ON(!list_empty(&rport->ls_rcv_list)); /* remove from lport list */ spin_lock_irqsave(&nvme_fc_lock, flags); @@ -557,7 +551,7 @@ nvme_fc_rport_get(struct nvme_fc_rport *rport) static void nvme_fc_resume_controller(struct nvme_fc_ctrl *ctrl) { - switch (ctrl->ctrl.state) { + switch (nvme_ctrl_state(&ctrl->ctrl)) { case NVME_CTRL_NEW: case NVME_CTRL_CONNECTING: /* @@ -793,49 +787,8 @@ nvme_fc_ctrl_connectivity_loss(struct nvme_fc_ctrl *ctrl) "NVME-FC{%d}: controller connectivity lost. Awaiting " "Reconnect", ctrl->cnum); - switch (ctrl->ctrl.state) { - case NVME_CTRL_NEW: - case NVME_CTRL_LIVE: - /* - * Schedule a controller reset. The reset will terminate the - * association and schedule the reconnect timer. Reconnects - * will be attempted until either the ctlr_loss_tmo - * (max_retries * connect_delay) expires or the remoteport's - * dev_loss_tmo expires. - */ - if (nvme_reset_ctrl(&ctrl->ctrl)) { - dev_warn(ctrl->ctrl.device, - "NVME-FC{%d}: Couldn't schedule reset.\n", - ctrl->cnum); - nvme_delete_ctrl(&ctrl->ctrl); - } - break; - - case NVME_CTRL_CONNECTING: - /* - * The association has already been terminated and the - * controller is attempting reconnects. No need to do anything - * futher. Reconnects will be attempted until either the - * ctlr_loss_tmo (max_retries * connect_delay) expires or the - * remoteport's dev_loss_tmo expires. - */ - break; - - case NVME_CTRL_RESETTING: - /* - * Controller is already in the process of terminating the - * association. No need to do anything further. The reconnect - * step will kick in naturally after the association is - * terminated. - */ - break; - - case NVME_CTRL_DELETING: - case NVME_CTRL_DELETING_NOIO: - default: - /* no action to take - let it delete */ - break; - } + set_bit(ASSOC_FAILED, &ctrl->flags); + nvme_reset_ctrl(&ctrl->ctrl); } /** @@ -948,7 +901,7 @@ EXPORT_SYMBOL_GPL(nvme_fc_set_remoteport_devloss); * may crash. * * As such: - * Wrapper all the dma routines and check the dev pointer. + * Wrap all the dma routines and check the dev pointer. * * If simple mappings (return just a dma address, we'll noop them, * returning a dma address of 0. @@ -1218,10 +1171,10 @@ nvme_fc_connect_admin_queue(struct nvme_fc_ctrl *ctrl, /* Linux supports only Dynamic controllers */ assoc_rqst->assoc_cmd.cntlid = cpu_to_be16(0xffff); uuid_copy(&assoc_rqst->assoc_cmd.hostid, &ctrl->ctrl.opts->host->id); - strncpy(assoc_rqst->assoc_cmd.hostnqn, ctrl->ctrl.opts->host->nqn, - min(FCNVME_ASSOC_HOSTNQN_LEN, NVMF_NQN_SIZE)); - strncpy(assoc_rqst->assoc_cmd.subnqn, ctrl->ctrl.opts->subsysnqn, - min(FCNVME_ASSOC_SUBNQN_LEN, NVMF_NQN_SIZE)); + strscpy(assoc_rqst->assoc_cmd.hostnqn, ctrl->ctrl.opts->host->nqn, + sizeof(assoc_rqst->assoc_cmd.hostnqn)); + strscpy(assoc_rqst->assoc_cmd.subnqn, ctrl->ctrl.opts->subsysnqn, + sizeof(assoc_rqst->assoc_cmd.subnqn)); lsop->queue = queue; lsreq->rqstaddr = assoc_rqst; @@ -1412,7 +1365,7 @@ nvme_fc_disconnect_assoc_done(struct nvmefc_ls_req *lsreq, int status) * down, and the related FC-NVME Association ID and Connection IDs * become invalid. * - * The behavior of the fc-nvme initiator is such that it's + * The behavior of the fc-nvme initiator is such that its * understanding of the association and connections will implicitly * be torn down. The action is implicit as it may be due to a loss of * connectivity with the fc-nvme target, so you may never get a @@ -1459,9 +1412,8 @@ nvme_fc_xmt_disconnect_assoc(struct nvme_fc_ctrl *ctrl) } static void -nvme_fc_xmt_ls_rsp_done(struct nvmefc_ls_rsp *lsrsp) +nvme_fc_xmt_ls_rsp_free(struct nvmefc_ls_rcv_op *lsop) { - struct nvmefc_ls_rcv_op *lsop = lsrsp->nvme_fc_private; struct nvme_fc_rport *rport = lsop->rport; struct nvme_fc_lport *lport = rport->lport; unsigned long flags; @@ -1483,6 +1435,14 @@ nvme_fc_xmt_ls_rsp_done(struct nvmefc_ls_rsp *lsrsp) } static void +nvme_fc_xmt_ls_rsp_done(struct nvmefc_ls_rsp *lsrsp) +{ + struct nvmefc_ls_rcv_op *lsop = lsrsp->nvme_fc_private; + + nvme_fc_xmt_ls_rsp_free(lsop); +} + +static void nvme_fc_xmt_ls_rsp(struct nvmefc_ls_rcv_op *lsop) { struct nvme_fc_rport *rport = lsop->rport; @@ -1499,7 +1459,7 @@ nvme_fc_xmt_ls_rsp(struct nvmefc_ls_rcv_op *lsop) dev_warn(lport->dev, "LLDD rejected LS RSP xmt: LS %d status %d\n", w0->ls_cmd, ret); - nvme_fc_xmt_ls_rsp_done(lsop->lsrsp); + nvme_fc_xmt_ls_rsp_free(lsop); return; } } @@ -1510,14 +1470,14 @@ nvme_fc_match_disconn_ls(struct nvme_fc_rport *rport, { struct fcnvme_ls_disconnect_assoc_rqst *rqst = &lsop->rqstbuf->rq_dis_assoc; - struct nvme_fc_ctrl *ctrl, *ret = NULL; + struct nvme_fc_ctrl *ctrl, *tmp, *ret = NULL; struct nvmefc_ls_rcv_op *oldls = NULL; u64 association_id = be64_to_cpu(rqst->associd.association_id); unsigned long flags; spin_lock_irqsave(&rport->lock, flags); - list_for_each_entry(ctrl, &rport->ctrl_list, ctrl_list) { + list_for_each_entry_safe(ctrl, tmp, &rport->ctrl_list, ctrl_list) { if (!nvme_fc_ctrl_get(ctrl)) continue; spin_lock(&ctrl->lock); @@ -1530,7 +1490,9 @@ nvme_fc_match_disconn_ls(struct nvme_fc_rport *rport, if (ret) /* leave the ctrl get reference */ break; + spin_unlock_irqrestore(&rport->lock, flags); nvme_fc_ctrl_put(ctrl); + spin_lock_irqsave(&rport->lock, flags); } spin_unlock_irqrestore(&rport->lock, flags); @@ -1924,7 +1886,7 @@ char *nvme_fc_io_getuuid(struct nvmefc_fcp_req *req) struct nvme_fc_fcp_op *op = fcp_req_to_fcp_op(req); struct request *rq = op->rq; - if (!IS_ENABLED(CONFIG_BLK_CGROUP_FC_APPID) || !rq->bio) + if (!IS_ENABLED(CONFIG_BLK_CGROUP_FC_APPID) || !rq || !rq->bio) return NULL; return blkcg_get_fc_appid(rq->bio); } @@ -1997,8 +1959,8 @@ nvme_fc_fcpio_done(struct nvmefc_fcp_req *req) } /* - * For the linux implementation, if we have an unsuccesful - * status, they blk-mq layer can typically be called with the + * For the linux implementation, if we have an unsuccessful + * status, the blk-mq layer can typically be called with the * non-zero status and the content of the cqe isn't important. */ if (status) @@ -2087,7 +2049,8 @@ done: nvme_fc_complete_rq(rq); check_error: - if (terminate_assoc && ctrl->ctrl.state != NVME_CTRL_RESETTING) + if (terminate_assoc && + nvme_ctrl_state(&ctrl->ctrl) != NVME_CTRL_RESETTING) queue_work(nvme_reset_wq, &ctrl->ioerr_work); } @@ -2396,17 +2359,11 @@ nvme_fc_ctrl_free(struct kref *ref) container_of(ref, struct nvme_fc_ctrl, ref); unsigned long flags; - if (ctrl->ctrl.tagset) - nvme_remove_io_tag_set(&ctrl->ctrl); - /* remove from rport list */ spin_lock_irqsave(&ctrl->rport->lock, flags); list_del(&ctrl->ctrl_list); spin_unlock_irqrestore(&ctrl->rport->lock, flags); - nvme_unquiesce_admin_queue(&ctrl->ctrl); - nvme_remove_admin_tag_set(&ctrl->ctrl); - kfree(ctrl->queues); put_device(ctrl->dev); @@ -2435,7 +2392,7 @@ nvme_fc_ctrl_get(struct nvme_fc_ctrl *ctrl) * controller. Called after last nvme_put_ctrl() call */ static void -nvme_fc_nvme_ctrl_freed(struct nvme_ctrl *nctrl) +nvme_fc_free_ctrl(struct nvme_ctrl *nctrl) { struct nvme_fc_ctrl *ctrl = to_fc_ctrl(nctrl); @@ -2470,7 +2427,7 @@ static bool nvme_fc_terminate_exchange(struct request *req, void *data) /* * This routine runs through all outstanding commands on the association - * and aborts them. This routine is typically be called by the + * and aborts them. This routine is typically called by the * delete_association routine. It is also called due to an error during * reconnect. In that scenario, it is most likely a command that initializes * the controller, including fabric Connect commands on io queues, that @@ -2520,7 +2477,7 @@ __nvme_fc_abort_outstanding_ios(struct nvme_fc_ctrl *ctrl, bool start_queues) * writing the registers for shutdown and polling (call * nvme_disable_ctrl()). Given a bunch of i/o was potentially * just aborted and we will wait on those contexts, and given - * there was no indication of how live the controlelr is on the + * there was no indication of how live the controller is on the * link, don't send more io to create more contexts for the * shutdown. Let the controller fail via keepalive failure if * its still present. @@ -2541,6 +2498,8 @@ __nvme_fc_abort_outstanding_ios(struct nvme_fc_ctrl *ctrl, bool start_queues) static void nvme_fc_error_recovery(struct nvme_fc_ctrl *ctrl, char *errmsg) { + enum nvme_ctrl_state state = nvme_ctrl_state(&ctrl->ctrl); + /* * if an error (io timeout, etc) while (re)connecting, the remote * port requested terminating of the association (disconnect_ls) @@ -2548,14 +2507,16 @@ nvme_fc_error_recovery(struct nvme_fc_ctrl *ctrl, char *errmsg) * the controller. Abort any ios on the association and let the * create_association error path resolve things. */ - if (ctrl->ctrl.state == NVME_CTRL_CONNECTING) { + if (state == NVME_CTRL_CONNECTING) { __nvme_fc_abort_outstanding_ios(ctrl, true); - set_bit(ASSOC_FAILED, &ctrl->flags); + dev_warn(ctrl->ctrl.device, + "NVME-FC{%d}: transport error during (re)connect\n", + ctrl->cnum); return; } /* Otherwise, only proceed if in LIVE state - e.g. on first error */ - if (ctrl->ctrl.state != NVME_CTRL_LIVE) + if (state != NVME_CTRL_LIVE) return; dev_warn(ctrl->ctrl.device, @@ -2571,6 +2532,7 @@ static enum blk_eh_timer_return nvme_fc_timeout(struct request *rq) { struct nvme_fc_fcp_op *op = blk_mq_rq_to_pdu(rq); struct nvme_fc_ctrl *ctrl = op->ctrl; + u16 qnum = op->queue->qnum; struct nvme_fc_cmd_iu *cmdiu = &op->cmd_iu; struct nvme_command *sqe = &cmdiu->sqe; @@ -2579,10 +2541,11 @@ static enum blk_eh_timer_return nvme_fc_timeout(struct request *rq) * will detect the aborted io and will fail the connection. */ dev_info(ctrl->ctrl.device, - "NVME-FC{%d.%d}: io timeout: opcode %d fctype %d w10/11: " + "NVME-FC{%d.%d}: io timeout: opcode %d fctype %d (%s) w10/11: " "x%08x/x%08x\n", - ctrl->cnum, op->queue->qnum, sqe->common.opcode, - sqe->connect.fctype, sqe->common.cdw10, sqe->common.cdw11); + ctrl->cnum, qnum, sqe->common.opcode, sqe->fabrics.fctype, + nvme_fabrics_opcode_str(qnum, sqe), + sqe->common.cdw10, sqe->common.cdw11); if (__nvme_fc_abort_op(ctrl, op)) nvme_fc_error_recovery(ctrl, "io timeout abort failed"); @@ -2613,7 +2576,7 @@ nvme_fc_map_data(struct nvme_fc_ctrl *ctrl, struct request *rq, if (ret) return -ENOMEM; - op->nents = blk_rq_map_sg(rq->q, rq, freq->sg_table.sgl); + op->nents = blk_rq_map_sg(rq, freq->sg_table.sgl); WARN_ON(op->nents > blk_rq_nr_phys_segments(rq)); freq->sg_cnt = fc_dma_map_sg(ctrl->lport->dev, freq->sg_table.sgl, op->nents, rq_dma_dir(rq)); @@ -2657,7 +2620,7 @@ nvme_fc_unmap_data(struct nvme_fc_ctrl *ctrl, struct request *rq, * as part of the exchange. The CQE is the last thing for the io, * which is transferred (explicitly or implicitly) with the RSP IU * sent on the exchange. After the CQE is received, the FC exchange is - * terminaed and the Exchange may be used on a different io. + * terminated and the Exchange may be used on a different io. * * The transport to LLDD api has the transport making a request for a * new fcp io request to the LLDD. The LLDD then allocates a FC exchange @@ -2812,7 +2775,7 @@ nvme_fc_queue_rq(struct blk_mq_hw_ctx *hctx, * as WRITE ZEROES will return a non-zero rq payload_bytes yet * there is no actual payload to be transferred. * To get it right, key data transmission on there being 1 or - * more physical segments in the sg list. If there is no + * more physical segments in the sg list. If there are no * physical segments, there is no payload. */ if (blk_rq_nr_phys_segments(rq)) { @@ -2900,7 +2863,7 @@ nvme_fc_create_io_queues(struct nvme_fc_ctrl *ctrl) unsigned int nr_io_queues; int ret; - nr_io_queues = min(min(opts->nr_io_queues, num_online_cpus()), + nr_io_queues = min3(opts->nr_io_queues, num_online_cpus(), ctrl->lport->ops->max_hw_queues); ret = nvme_set_queue_count(&ctrl->ctrl, &nr_io_queues); if (ret) { @@ -2917,8 +2880,8 @@ nvme_fc_create_io_queues(struct nvme_fc_ctrl *ctrl) ret = nvme_alloc_io_tag_set(&ctrl->ctrl, &ctrl->tag_set, &nvme_fc_mq_ops, 1, - struct_size((struct nvme_fcp_op_w_sgl *)NULL, priv, - ctrl->lport->ops->fcprqst_priv_sz)); + struct_size_t(struct nvme_fcp_op_w_sgl, priv, + ctrl->lport->ops->fcprqst_priv_sz)); if (ret) return ret; @@ -2954,7 +2917,7 @@ nvme_fc_recreate_io_queues(struct nvme_fc_ctrl *ctrl) unsigned int nr_io_queues; int ret; - nr_io_queues = min(min(opts->nr_io_queues, num_online_cpus()), + nr_io_queues = min3(opts->nr_io_queues, num_online_cpus(), ctrl->lport->ops->max_hw_queues); ret = nvme_set_queue_count(&ctrl->ctrl, &nr_io_queues); if (ret) { @@ -3064,15 +3027,20 @@ nvme_fc_create_association(struct nvme_fc_ctrl *ctrl) struct nvmefc_ls_rcv_op *disls = NULL; unsigned long flags; int ret; - bool changed; ++ctrl->ctrl.nr_reconnects; - if (ctrl->rport->remoteport.port_state != FC_OBJSTATE_ONLINE) + spin_lock_irqsave(&ctrl->rport->lock, flags); + if (ctrl->rport->remoteport.port_state != FC_OBJSTATE_ONLINE) { + spin_unlock_irqrestore(&ctrl->rport->lock, flags); return -ENODEV; + } - if (nvme_fc_ctlr_active_on_rport(ctrl)) + if (nvme_fc_ctlr_active_on_rport(ctrl)) { + spin_unlock_irqrestore(&ctrl->rport->lock, flags); return -ENOTUNIQ; + } + spin_unlock_irqrestore(&ctrl->rport->lock, flags); dev_info(ctrl->ctrl.device, "NVME-FC{%d}: create association : host wwpn 0x%016llx " @@ -3110,7 +3078,9 @@ nvme_fc_create_association(struct nvme_fc_ctrl *ctrl) */ ret = nvme_enable_ctrl(&ctrl->ctrl); - if (ret || test_bit(ASSOC_FAILED, &ctrl->flags)) + if (!ret && test_bit(ASSOC_FAILED, &ctrl->flags)) + ret = -EIO; + if (ret) goto out_disconnect_admin_queue; ctrl->ctrl.max_segments = ctrl->lport->ops->max_sgl_segments; @@ -3120,25 +3090,28 @@ nvme_fc_create_association(struct nvme_fc_ctrl *ctrl) nvme_unquiesce_admin_queue(&ctrl->ctrl); ret = nvme_init_ctrl_finish(&ctrl->ctrl, false); - if (ret || test_bit(ASSOC_FAILED, &ctrl->flags)) + if (ret) goto out_disconnect_admin_queue; - + if (test_bit(ASSOC_FAILED, &ctrl->flags)) { + ret = -EIO; + goto out_stop_keep_alive; + } /* sanity checks */ /* FC-NVME does not have other data in the capsule */ if (ctrl->ctrl.icdoff) { dev_err(ctrl->ctrl.device, "icdoff %d is not supported!\n", ctrl->ctrl.icdoff); - ret = NVME_SC_INVALID_FIELD | NVME_SC_DNR; - goto out_disconnect_admin_queue; + ret = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; + goto out_stop_keep_alive; } /* FC-NVME supports normal SGL Data Block Descriptors */ if (!nvme_ctrl_sgl_supported(&ctrl->ctrl)) { dev_err(ctrl->ctrl.device, "Mandatory sgls are not supported!\n"); - ret = NVME_SC_INVALID_FIELD | NVME_SC_DNR; - goto out_disconnect_admin_queue; + ret = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; + goto out_stop_keep_alive; } if (opts->queue_size > ctrl->ctrl.maxcmd) { @@ -3165,21 +3138,29 @@ nvme_fc_create_association(struct nvme_fc_ctrl *ctrl) else ret = nvme_fc_recreate_io_queues(ctrl); } - if (ret || test_bit(ASSOC_FAILED, &ctrl->flags)) + if (!ret && test_bit(ASSOC_FAILED, &ctrl->flags)) + ret = -EIO; + if (ret) goto out_term_aen_ops; - changed = nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_LIVE); + if (!nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_LIVE)) { + ret = -EIO; + goto out_term_aen_ops; + } ctrl->ctrl.nr_reconnects = 0; - - if (changed) - nvme_start_ctrl(&ctrl->ctrl); + nvme_start_ctrl(&ctrl->ctrl); return 0; /* Success */ out_term_aen_ops: nvme_fc_term_aen_ops(ctrl); +out_stop_keep_alive: + nvme_stop_keep_alive(&ctrl->ctrl); out_disconnect_admin_queue: + dev_warn(ctrl->ctrl.device, + "NVME-FC{%d}: create_assoc failed, assoc_id %llx ret %d\n", + ctrl->cnum, ctrl->association_id, ret); /* send a Disconnect(association) LS to fc-nvme target */ nvme_fc_xmt_disconnect_assoc(ctrl); spin_lock_irqsave(&ctrl->lock, flags); @@ -3276,13 +3257,20 @@ nvme_fc_delete_ctrl(struct nvme_ctrl *nctrl) { struct nvme_fc_ctrl *ctrl = to_fc_ctrl(nctrl); - cancel_work_sync(&ctrl->ioerr_work); cancel_delayed_work_sync(&ctrl->connect_work); + /* * kill the association on the link side. this will block * waiting for io to terminate */ nvme_fc_delete_association(ctrl); + cancel_work_sync(&ctrl->ioerr_work); + + if (ctrl->ctrl.tagset) + nvme_remove_io_tag_set(&ctrl->ctrl); + + nvme_unquiesce_admin_queue(&ctrl->ctrl); + nvme_remove_admin_tag_set(&ctrl->ctrl); } static void @@ -3293,19 +3281,17 @@ nvme_fc_reconnect_or_delete(struct nvme_fc_ctrl *ctrl, int status) unsigned long recon_delay = ctrl->ctrl.opts->reconnect_delay * HZ; bool recon = true; - if (ctrl->ctrl.state != NVME_CTRL_CONNECTING) + if (nvme_ctrl_state(&ctrl->ctrl) != NVME_CTRL_CONNECTING) return; if (portptr->port_state == FC_OBJSTATE_ONLINE) { dev_info(ctrl->ctrl.device, "NVME-FC{%d}: reset: Reconnect attempt failed (%d)\n", ctrl->cnum, status); - if (status > 0 && (status & NVME_SC_DNR)) - recon = false; } else if (time_after_eq(jiffies, rport->dev_loss_end)) recon = false; - if (recon && nvmf_should_reconnect(&ctrl->ctrl)) { + if (recon && nvmf_should_reconnect(&ctrl->ctrl, status)) { if (portptr->port_state == FC_OBJSTATE_ONLINE) dev_info(ctrl->ctrl.device, "NVME-FC{%d}: Reconnect attempt in %ld " @@ -3317,7 +3303,7 @@ nvme_fc_reconnect_or_delete(struct nvme_fc_ctrl *ctrl, int status) queue_delayed_work(nvme_wq, &ctrl->connect_work, recon_delay); } else { if (portptr->port_state == FC_OBJSTATE_ONLINE) { - if (status > 0 && (status & NVME_SC_DNR)) + if (status > 0 && (status & NVME_STATUS_DNR)) dev_warn(ctrl->ctrl.device, "NVME-FC{%d}: reconnect failure\n", ctrl->cnum); @@ -3374,10 +3360,12 @@ static const struct nvme_ctrl_ops nvme_fc_ctrl_ops = { .reg_read32 = nvmf_reg_read32, .reg_read64 = nvmf_reg_read64, .reg_write32 = nvmf_reg_write32, - .free_ctrl = nvme_fc_nvme_ctrl_freed, + .subsystem_reset = nvmf_subsystem_reset, + .free_ctrl = nvme_fc_free_ctrl, .submit_async_event = nvme_fc_submit_async_event, .delete_ctrl = nvme_fc_delete_ctrl, .get_address = nvmf_get_address, + .get_virt_boundary = nvmf_get_virt_boundary, }; static void @@ -3436,12 +3424,11 @@ nvme_fc_existing_controller(struct nvme_fc_rport *rport, return found; } -static struct nvme_ctrl * -nvme_fc_init_ctrl(struct device *dev, struct nvmf_ctrl_options *opts, +static struct nvme_fc_ctrl * +nvme_fc_alloc_ctrl(struct device *dev, struct nvmf_ctrl_options *opts, struct nvme_fc_lport *lport, struct nvme_fc_rport *rport) { struct nvme_fc_ctrl *ctrl; - unsigned long flags; int ret, idx, ctrl_loss_tmo; if (!(rport->remoteport.port_role & @@ -3483,10 +3470,6 @@ nvme_fc_init_ctrl(struct device *dev, struct nvmf_ctrl_options *opts, ctrl->ctrl.opts = opts; ctrl->ctrl.nr_reconnects = 0; - if (lport->dev) - ctrl->ctrl.numa_node = dev_to_node(lport->dev); - else - ctrl->ctrl.numa_node = NUMA_NO_NODE; INIT_LIST_HEAD(&ctrl->ctrl_list); ctrl->lport = lport; ctrl->rport = rport; @@ -3531,13 +3514,43 @@ nvme_fc_init_ctrl(struct device *dev, struct nvmf_ctrl_options *opts, ret = nvme_init_ctrl(&ctrl->ctrl, dev, &nvme_fc_ctrl_ops, 0); if (ret) goto out_free_queues; + if (lport->dev) + ctrl->ctrl.numa_node = dev_to_node(lport->dev); + + return ctrl; + +out_free_queues: + kfree(ctrl->queues); +out_free_ida: + put_device(ctrl->dev); + ida_free(&nvme_fc_ctrl_cnt, ctrl->cnum); +out_free_ctrl: + kfree(ctrl); +out_fail: + /* exit via here doesn't follow ctlr ref points */ + return ERR_PTR(ret); +} + +static struct nvme_ctrl * +nvme_fc_init_ctrl(struct device *dev, struct nvmf_ctrl_options *opts, + struct nvme_fc_lport *lport, struct nvme_fc_rport *rport) +{ + struct nvme_fc_ctrl *ctrl; + unsigned long flags; + int ret; + + ctrl = nvme_fc_alloc_ctrl(dev, opts, lport, rport); + if (IS_ERR(ctrl)) + return ERR_CAST(ctrl); - /* at this point, teardown path changes to ref counting on nvme ctrl */ + ret = nvme_add_ctrl(&ctrl->ctrl); + if (ret) + goto out_put_ctrl; ret = nvme_alloc_admin_tag_set(&ctrl->ctrl, &ctrl->admin_tag_set, &nvme_fc_admin_mq_ops, - struct_size((struct nvme_fcp_op_w_sgl *)NULL, priv, - ctrl->lport->ops->fcprqst_priv_sz)); + struct_size_t(struct nvme_fcp_op_w_sgl, priv, + ctrl->lport->ops->fcprqst_priv_sz)); if (ret) goto fail_ctrl; @@ -3545,8 +3558,7 @@ nvme_fc_init_ctrl(struct device *dev, struct nvmf_ctrl_options *opts, list_add_tail(&ctrl->ctrl_list, &rport->ctrl_list); spin_unlock_irqrestore(&rport->lock, flags); - if (!nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_RESETTING) || - !nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_CONNECTING)) { + if (!nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_CONNECTING)) { dev_err(ctrl->ctrl.device, "NVME-FC{%d}: failed to init ctrl state\n", ctrl->cnum); goto fail_ctrl; @@ -3562,8 +3574,8 @@ nvme_fc_init_ctrl(struct device *dev, struct nvmf_ctrl_options *opts, flush_delayed_work(&ctrl->connect_work); dev_info(ctrl->ctrl.device, - "NVME-FC{%d}: new ctrl: NQN \"%s\"\n", - ctrl->cnum, nvmf_ctrl_subsysnqn(&ctrl->ctrl)); + "NVME-FC{%d}: new ctrl: NQN \"%s\", hostnqn: %s\n", + ctrl->cnum, nvmf_ctrl_subsysnqn(&ctrl->ctrl), opts->host->nqn); return &ctrl->ctrl; @@ -3578,6 +3590,7 @@ fail_ctrl: /* initiate nvme ctrl ref counting teardown */ nvme_uninit_ctrl(&ctrl->ctrl); +out_put_ctrl: /* Remove core ctrl ref. */ nvme_put_ctrl(&ctrl->ctrl); @@ -3591,20 +3604,8 @@ fail_ctrl: nvme_fc_rport_get(rport); return ERR_PTR(-EIO); - -out_free_queues: - kfree(ctrl->queues); -out_free_ida: - put_device(ctrl->dev); - ida_free(&nvme_fc_ctrl_cnt, ctrl->cnum); -out_free_ctrl: - kfree(ctrl); -out_fail: - /* exit via here doesn't follow ctlr ref points */ - return ERR_PTR(ret); } - struct nvmet_fc_traddr { u64 nn; u64 pn; @@ -3875,17 +3876,12 @@ static const struct attribute_group *nvme_fc_attr_groups[] = { static struct class fc_class = { .name = "fc", .dev_groups = nvme_fc_attr_groups, - .owner = THIS_MODULE, }; static int __init nvme_fc_init_module(void) { int ret; - nvme_fc_wq = alloc_workqueue("nvme_fc_wq", WQ_MEM_RECLAIM, 0); - if (!nvme_fc_wq) - return -ENOMEM; - /* * NOTE: * It is expected that in the future the kernel will combine @@ -3903,7 +3899,7 @@ static int __init nvme_fc_init_module(void) ret = class_register(&fc_class); if (ret) { pr_err("couldn't register class fc\n"); - goto out_destroy_wq; + return ret; } /* @@ -3927,8 +3923,6 @@ out_destroy_device: device_destroy(&fc_class, MKDEV(0, 0)); out_destroy_class: class_unregister(&fc_class); -out_destroy_wq: - destroy_workqueue(nvme_fc_wq); return ret; } @@ -3948,48 +3942,27 @@ nvme_fc_delete_controllers(struct nvme_fc_rport *rport) spin_unlock(&rport->lock); } -static void -nvme_fc_cleanup_for_unload(void) +static void __exit nvme_fc_exit_module(void) { struct nvme_fc_lport *lport; struct nvme_fc_rport *rport; - - list_for_each_entry(lport, &nvme_fc_lport_list, port_list) { - list_for_each_entry(rport, &lport->endp_list, endp_list) { - nvme_fc_delete_controllers(rport); - } - } -} - -static void __exit nvme_fc_exit_module(void) -{ unsigned long flags; - bool need_cleanup = false; spin_lock_irqsave(&nvme_fc_lock, flags); - nvme_fc_waiting_to_unload = true; - if (!list_empty(&nvme_fc_lport_list)) { - need_cleanup = true; - nvme_fc_cleanup_for_unload(); - } + list_for_each_entry(lport, &nvme_fc_lport_list, port_list) + list_for_each_entry(rport, &lport->endp_list, endp_list) + nvme_fc_delete_controllers(rport); spin_unlock_irqrestore(&nvme_fc_lock, flags); - if (need_cleanup) { - pr_info("%s: waiting for ctlr deletes\n", __func__); - wait_for_completion(&nvme_fc_unload_proceed); - pr_info("%s: ctrl deletes complete\n", __func__); - } + flush_workqueue(nvme_delete_wq); nvmf_unregister_transport(&nvme_fc_transport); - ida_destroy(&nvme_fc_local_port_cnt); - ida_destroy(&nvme_fc_ctrl_cnt); - device_destroy(&fc_class, MKDEV(0, 0)); class_unregister(&fc_class); - destroy_workqueue(nvme_fc_wq); } module_init(nvme_fc_init_module); module_exit(nvme_fc_exit_module); +MODULE_DESCRIPTION("NVMe host FC transport driver"); MODULE_LICENSE("GPL v2"); diff --git a/drivers/nvme/host/hwmon.c b/drivers/nvme/host/hwmon.c index 9e6e56c20ec9..89a1a1043d63 100644 --- a/drivers/nvme/host/hwmon.c +++ b/drivers/nvme/host/hwmon.c @@ -6,7 +6,7 @@ #include <linux/hwmon.h> #include <linux/units.h> -#include <asm/unaligned.h> +#include <linux/unaligned.h> #include "nvme.h" @@ -163,7 +163,9 @@ static umode_t nvme_hwmon_is_visible(const void *_data, case hwmon_temp_max: case hwmon_temp_min: if ((!channel && data->ctrl->wctemp) || - (channel && data->log->temp_sensor[channel - 1])) { + (channel && data->log->temp_sensor[channel - 1] && + !(data->ctrl->quirks & + NVME_QUIRK_NO_SECONDARY_TEMP_THRESH))) { if (data->ctrl->quirks & NVME_QUIRK_NO_TEMP_THRESH_CHANGE) return 0444; @@ -185,7 +187,7 @@ static umode_t nvme_hwmon_is_visible(const void *_data, return 0; } -static const struct hwmon_channel_info *nvme_hwmon_info[] = { +static const struct hwmon_channel_info *const nvme_hwmon_info[] = { HWMON_CHANNEL_INFO(chip, HWMON_C_REGISTER_TZ), HWMON_CHANNEL_INFO(temp, HWMON_T_INPUT | HWMON_T_MAX | HWMON_T_MIN | diff --git a/drivers/nvme/host/ioctl.c b/drivers/nvme/host/ioctl.c index 06f52db34be9..a9c097dacad6 100644 --- a/drivers/nvme/host/ioctl.c +++ b/drivers/nvme/host/ioctl.c @@ -3,9 +3,10 @@ * Copyright (c) 2011-2014, Intel Corporation. * Copyright (c) 2017-2021 Christoph Hellwig. */ +#include <linux/blk-integrity.h> #include <linux/ptrace.h> /* for force_successful_syscall_return */ #include <linux/nvme_ioctl.h> -#include <linux/io_uring.h> +#include <linux/io_uring/cmd.h> #include "nvme.h" enum { @@ -14,19 +15,16 @@ enum { }; static bool nvme_cmd_allowed(struct nvme_ns *ns, struct nvme_command *c, - unsigned int flags, fmode_t mode) + unsigned int flags, bool open_for_write) { u32 effects; - if (capable(CAP_SYS_ADMIN)) - return true; - /* * Do not allow unprivileged passthrough on partitions, as that allows an * escape from the containment of the partition. */ if (flags & NVME_IOCTL_PARTITION) - return false; + goto admin; /* * Do not allow unprivileged processes to send vendor specific or fabrics @@ -34,7 +32,7 @@ static bool nvme_cmd_allowed(struct nvme_ns *ns, struct nvme_command *c, */ if (c->common.opcode >= nvme_cmd_vendor_start || c->common.opcode == nvme_fabrics_command) - return false; + goto admin; /* * Do not allow unprivileged passthrough of admin commands except @@ -53,7 +51,7 @@ static bool nvme_cmd_allowed(struct nvme_ns *ns, struct nvme_command *c, return true; } } - return false; + goto admin; } /* @@ -63,7 +61,7 @@ static bool nvme_cmd_allowed(struct nvme_ns *ns, struct nvme_command *c, */ effects = nvme_command_effects(ns->ctrl, ns, c->common.opcode); if (!(effects & NVME_CMD_EFFECTS_CSUPP)) - return false; + goto admin; /* * Don't allow passthrough for command that have intrusive (or unknown) @@ -72,16 +70,20 @@ static bool nvme_cmd_allowed(struct nvme_ns *ns, struct nvme_command *c, if (effects & ~(NVME_CMD_EFFECTS_CSUPP | NVME_CMD_EFFECTS_LBCC | NVME_CMD_EFFECTS_UUID_SEL | NVME_CMD_EFFECTS_SCOPE_MASK)) - return false; + goto admin; /* * Only allow I/O commands that transfer data to the controller or that * change the logical block contents if the file descriptor is open for * writing. */ - if (nvme_is_write(c) || (effects & NVME_CMD_EFFECTS_LBCC)) - return mode & FMODE_WRITE; + if ((nvme_is_write(c) || (effects & NVME_CMD_EFFECTS_LBCC)) && + !open_for_write) + goto admin; + return true; +admin: + return capable(CAP_SYS_ADMIN); } /* @@ -96,55 +98,6 @@ static void __user *nvme_to_user_ptr(uintptr_t ptrval) return (void __user *)ptrval; } -static void *nvme_add_user_metadata(struct request *req, void __user *ubuf, - unsigned len, u32 seed) -{ - struct bio_integrity_payload *bip; - int ret = -ENOMEM; - void *buf; - struct bio *bio = req->bio; - - buf = kmalloc(len, GFP_KERNEL); - if (!buf) - goto out; - - ret = -EFAULT; - if ((req_op(req) == REQ_OP_DRV_OUT) && copy_from_user(buf, ubuf, len)) - goto out_free_meta; - - bip = bio_integrity_alloc(bio, GFP_KERNEL, 1); - if (IS_ERR(bip)) { - ret = PTR_ERR(bip); - goto out_free_meta; - } - - bip->bip_iter.bi_size = len; - bip->bip_iter.bi_sector = seed; - ret = bio_integrity_add_page(bio, virt_to_page(buf), len, - offset_in_page(buf)); - if (ret != len) { - ret = -ENOMEM; - goto out_free_meta; - } - - req->cmd_flags |= REQ_INTEGRITY; - return buf; -out_free_meta: - kfree(buf); -out: - return ERR_PTR(ret); -} - -static int nvme_finish_user_metadata(struct request *req, void __user *ubuf, - void *meta, unsigned len, int ret) -{ - if (!ret && req_op(req) == REQ_OP_DRV_IN && - copy_to_user(ubuf, meta, len)) - ret = -EFAULT; - kfree(meta); - return ret; -} - static struct request *nvme_alloc_user_request(struct request_queue *q, struct nvme_command *cmd, blk_opf_t rq_flags, blk_mq_req_flags_t blk_flags) @@ -161,47 +114,41 @@ static struct request *nvme_alloc_user_request(struct request_queue *q, static int nvme_map_user_request(struct request *req, u64 ubuffer, unsigned bufflen, void __user *meta_buffer, unsigned meta_len, - u32 meta_seed, void **metap, struct io_uring_cmd *ioucmd, - unsigned int flags) + struct iov_iter *iter, unsigned int flags) { struct request_queue *q = req->q; struct nvme_ns *ns = q->queuedata; struct block_device *bdev = ns ? ns->disk->part0 : NULL; + bool supports_metadata = bdev && blk_get_integrity(bdev->bd_disk); + struct nvme_ctrl *ctrl = nvme_req(req)->ctrl; + bool has_metadata = meta_buffer && meta_len; struct bio *bio = NULL; - void *meta = NULL; int ret; - if (ioucmd && (ioucmd->flags & IORING_URING_CMD_FIXED)) { - struct iov_iter iter; - - /* fixedbufs is only for non-vectored io */ - if (WARN_ON_ONCE(flags & NVME_IOCTL_VEC)) + if (!nvme_ctrl_sgl_supported(ctrl)) + dev_warn_once(ctrl->device, "using unchecked data buffer\n"); + if (has_metadata) { + if (!supports_metadata) return -EINVAL; - ret = io_uring_cmd_import_fixed(ubuffer, bufflen, - rq_data_dir(req), &iter, ioucmd); - if (ret < 0) - goto out; - ret = blk_rq_map_user_iov(q, req, NULL, &iter, GFP_KERNEL); - } else { + + if (!nvme_ctrl_meta_sgl_supported(ctrl)) + dev_warn_once(ctrl->device, + "using unchecked metadata buffer\n"); + } + + if (iter) + ret = blk_rq_map_user_iov(q, req, NULL, iter, GFP_KERNEL); + else ret = blk_rq_map_user_io(req, NULL, nvme_to_user_ptr(ubuffer), bufflen, GFP_KERNEL, flags & NVME_IOCTL_VEC, 0, 0, rq_data_dir(req)); - } - if (ret) - goto out; - bio = req->bio; - if (bdev) - bio_set_dev(bio, bdev); - - if (bdev && meta_buffer && meta_len) { - meta = nvme_add_user_metadata(req, meta_buffer, meta_len, - meta_seed); - if (IS_ERR(meta)) { - ret = PTR_ERR(meta); + return ret; + + if (has_metadata) { + ret = blk_rq_integrity_map_user(req, meta_buffer, meta_len); + if (ret) goto out_unmap; - } - *metap = meta; } return ret; @@ -209,19 +156,17 @@ static int nvme_map_user_request(struct request *req, u64 ubuffer, out_unmap: if (bio) blk_rq_unmap_user(bio); -out: - blk_mq_free_request(req); return ret; } static int nvme_submit_user_cmd(struct request_queue *q, struct nvme_command *cmd, u64 ubuffer, unsigned bufflen, - void __user *meta_buffer, unsigned meta_len, u32 meta_seed, + void __user *meta_buffer, unsigned meta_len, u64 *result, unsigned timeout, unsigned int flags) { + struct nvme_ns *ns = q->queuedata; struct nvme_ctrl *ctrl; struct request *req; - void *meta = NULL; struct bio *bio; u32 effects; int ret; @@ -233,28 +178,28 @@ static int nvme_submit_user_cmd(struct request_queue *q, req->timeout = timeout; if (ubuffer && bufflen) { ret = nvme_map_user_request(req, ubuffer, bufflen, meta_buffer, - meta_len, meta_seed, &meta, NULL, flags); + meta_len, NULL, flags); if (ret) - return ret; + goto out_free_req; } bio = req->bio; ctrl = nvme_req(req)->ctrl; - ret = nvme_execute_passthru_rq(req, &effects); - + effects = nvme_passthru_start(ctrl, ns, cmd->common.opcode); + ret = nvme_execute_rq(req, false); if (result) *result = le64_to_cpu(nvme_req(req)->result.u64); - if (meta) - ret = nvme_finish_user_metadata(req, meta_buffer, meta, - meta_len, ret); if (bio) blk_rq_unmap_user(bio); blk_mq_free_request(req); if (effects) - nvme_passthru_end(ctrl, effects, cmd, ret); + nvme_passthru_end(ctrl, ns, effects, cmd, ret); + return ret; +out_free_req: + blk_mq_free_request(req); return ret; } @@ -279,10 +224,10 @@ static int nvme_submit_io(struct nvme_ns *ns, struct nvme_user_io __user *uio) return -EINVAL; } - length = (io.nblocks + 1) << ns->lba_shift; + length = (io.nblocks + 1) << ns->head->lba_shift; if ((io.control & NVME_RW_PRINFO_PRACT) && - ns->ms == sizeof(struct t10_pi_tuple)) { + (ns->head->ms == ns->head->pi_size)) { /* * Protection information is stripped/inserted by the * controller. @@ -292,11 +237,11 @@ static int nvme_submit_io(struct nvme_ns *ns, struct nvme_user_io __user *uio) meta_len = 0; metadata = NULL; } else { - meta_len = (io.nblocks + 1) * ns->ms; + meta_len = (io.nblocks + 1) * ns->head->ms; metadata = nvme_to_user_ptr(io.metadata); } - if (ns->features & NVME_NS_EXT_LBAS) { + if (ns->head->features & NVME_NS_EXT_LBAS) { length += meta_len; meta_len = 0; } else if (meta_len) { @@ -313,11 +258,11 @@ static int nvme_submit_io(struct nvme_ns *ns, struct nvme_user_io __user *uio) c.rw.control = cpu_to_le16(io.control); c.rw.dsmgmt = cpu_to_le32(io.dsmgmt); c.rw.reftag = cpu_to_le32(io.reftag); - c.rw.apptag = cpu_to_le16(io.apptag); - c.rw.appmask = cpu_to_le16(io.appmask); + c.rw.lbat = cpu_to_le16(io.apptag); + c.rw.lbatm = cpu_to_le16(io.appmask); return nvme_submit_user_cmd(ns->queue, &c, io.addr, length, metadata, - meta_len, lower_32_bits(io.slba), NULL, 0, 0); + meta_len, NULL, 0, 0); } static bool nvme_validate_passthru_nsid(struct nvme_ctrl *ctrl, @@ -325,8 +270,7 @@ static bool nvme_validate_passthru_nsid(struct nvme_ctrl *ctrl, { if (ns && nsid != ns->head->ns_id) { dev_err(ctrl->device, - "%s: nsid (%u) in cmd does not match nsid (%u)" - "of namespace\n", + "%s: nsid (%u) in cmd does not match nsid (%u) of namespace\n", current->comm, nsid, ns->head->ns_id); return false; } @@ -336,7 +280,7 @@ static bool nvme_validate_passthru_nsid(struct nvme_ctrl *ctrl, static int nvme_user_cmd(struct nvme_ctrl *ctrl, struct nvme_ns *ns, struct nvme_passthru_cmd __user *ucmd, unsigned int flags, - fmode_t mode) + bool open_for_write) { struct nvme_passthru_cmd cmd; struct nvme_command c; @@ -364,7 +308,7 @@ static int nvme_user_cmd(struct nvme_ctrl *ctrl, struct nvme_ns *ns, c.common.cdw14 = cpu_to_le32(cmd.cdw14); c.common.cdw15 = cpu_to_le32(cmd.cdw15); - if (!nvme_cmd_allowed(ns, &c, 0, mode)) + if (!nvme_cmd_allowed(ns, &c, 0, open_for_write)) return -EACCES; if (cmd.timeout_ms) @@ -372,7 +316,7 @@ static int nvme_user_cmd(struct nvme_ctrl *ctrl, struct nvme_ns *ns, status = nvme_submit_user_cmd(ns ? ns->queue : ctrl->admin_q, &c, cmd.addr, cmd.data_len, nvme_to_user_ptr(cmd.metadata), - cmd.metadata_len, 0, &result, timeout, 0); + cmd.metadata_len, &result, timeout, 0); if (status >= 0) { if (put_user(result, &ucmd->result)) @@ -384,7 +328,7 @@ static int nvme_user_cmd(struct nvme_ctrl *ctrl, struct nvme_ns *ns, static int nvme_user_cmd64(struct nvme_ctrl *ctrl, struct nvme_ns *ns, struct nvme_passthru_cmd64 __user *ucmd, unsigned int flags, - fmode_t mode) + bool open_for_write) { struct nvme_passthru_cmd64 cmd; struct nvme_command c; @@ -411,7 +355,7 @@ static int nvme_user_cmd64(struct nvme_ctrl *ctrl, struct nvme_ns *ns, c.common.cdw14 = cpu_to_le32(cmd.cdw14); c.common.cdw15 = cpu_to_le32(cmd.cdw15); - if (!nvme_cmd_allowed(ns, &c, flags, mode)) + if (!nvme_cmd_allowed(ns, &c, flags, open_for_write)) return -EACCES; if (cmd.timeout_ms) @@ -419,7 +363,7 @@ static int nvme_user_cmd64(struct nvme_ctrl *ctrl, struct nvme_ns *ns, status = nvme_submit_user_cmd(ns ? ns->queue : ctrl->admin_q, &c, cmd.addr, cmd.data_len, nvme_to_user_ptr(cmd.metadata), - cmd.metadata_len, 0, &cmd.result, timeout, flags); + cmd.metadata_len, &cmd.result, timeout, flags); if (status >= 0) { if (put_user(cmd.result, &ucmd->result)) @@ -442,59 +386,27 @@ struct nvme_uring_data { * Expect build errors if this grows larger than that. */ struct nvme_uring_cmd_pdu { - union { - struct bio *bio; - struct request *req; - }; - u32 meta_len; - u32 nvme_status; - union { - struct { - void *meta; /* kernel-resident buffer */ - void __user *meta_buffer; - }; - u64 result; - } u; + struct request *req; + struct bio *bio; + u64 result; + int status; }; static inline struct nvme_uring_cmd_pdu *nvme_uring_cmd_pdu( struct io_uring_cmd *ioucmd) { - return (struct nvme_uring_cmd_pdu *)&ioucmd->pdu; -} - -static void nvme_uring_task_meta_cb(struct io_uring_cmd *ioucmd) -{ - struct nvme_uring_cmd_pdu *pdu = nvme_uring_cmd_pdu(ioucmd); - struct request *req = pdu->req; - int status; - u64 result; - - if (nvme_req(req)->flags & NVME_REQ_CANCELLED) - status = -EINTR; - else - status = nvme_req(req)->status; - - result = le64_to_cpu(nvme_req(req)->result.u64); - - if (pdu->meta_len) - status = nvme_finish_user_metadata(req, pdu->u.meta_buffer, - pdu->u.meta, pdu->meta_len, status); - if (req->bio) - blk_rq_unmap_user(req->bio); - blk_mq_free_request(req); - - io_uring_cmd_done(ioucmd, status, result); + return io_uring_cmd_to_pdu(ioucmd, struct nvme_uring_cmd_pdu); } -static void nvme_uring_task_cb(struct io_uring_cmd *ioucmd) +static void nvme_uring_task_cb(struct io_tw_req tw_req, io_tw_token_t tw) { + struct io_uring_cmd *ioucmd = io_uring_cmd_from_tw(tw_req); struct nvme_uring_cmd_pdu *pdu = nvme_uring_cmd_pdu(ioucmd); if (pdu->bio) blk_rq_unmap_user(pdu->bio); - - io_uring_cmd_done(ioucmd, pdu->nvme_status, pdu->u.result); + io_uring_cmd_done32(ioucmd, pdu->status, pdu->result, + IO_URING_CMD_TASK_WORK_ISSUE_FLAGS); } static enum rq_end_io_ret nvme_uring_cmd_end_io(struct request *req, @@ -502,61 +414,41 @@ static enum rq_end_io_ret nvme_uring_cmd_end_io(struct request *req, { struct io_uring_cmd *ioucmd = req->end_io_data; struct nvme_uring_cmd_pdu *pdu = nvme_uring_cmd_pdu(ioucmd); - void *cookie = READ_ONCE(ioucmd->cookie); - req->bio = pdu->bio; - if (nvme_req(req)->flags & NVME_REQ_CANCELLED) - pdu->nvme_status = -EINTR; - else - pdu->nvme_status = nvme_req(req)->status; - pdu->u.result = le64_to_cpu(nvme_req(req)->result.u64); + if (nvme_req(req)->flags & NVME_REQ_CANCELLED) { + pdu->status = -EINTR; + } else { + pdu->status = nvme_req(req)->status; + if (!pdu->status) + pdu->status = blk_status_to_errno(err); + } + pdu->result = le64_to_cpu(nvme_req(req)->result.u64); /* - * For iopoll, complete it directly. - * Otherwise, move the completion to task work. + * IOPOLL could potentially complete this request directly, but + * if multiple rings are polling on the same queue, then it's possible + * for one ring to find completions for another ring. Punting the + * completion via task_work will always direct it to the right + * location, rather than potentially complete requests for ringA + * under iopoll invocations from ringB. */ - if (cookie != NULL && blk_rq_is_poll(req)) - nvme_uring_task_cb(ioucmd); - else - io_uring_cmd_complete_in_task(ioucmd, nvme_uring_task_cb); - + io_uring_cmd_do_in_task_lazy(ioucmd, nvme_uring_task_cb); return RQ_END_IO_FREE; } -static enum rq_end_io_ret nvme_uring_cmd_end_io_meta(struct request *req, - blk_status_t err) -{ - struct io_uring_cmd *ioucmd = req->end_io_data; - struct nvme_uring_cmd_pdu *pdu = nvme_uring_cmd_pdu(ioucmd); - void *cookie = READ_ONCE(ioucmd->cookie); - - req->bio = pdu->bio; - pdu->req = req; - - /* - * For iopoll, complete it directly. - * Otherwise, move the completion to task work. - */ - if (cookie != NULL && blk_rq_is_poll(req)) - nvme_uring_task_meta_cb(ioucmd); - else - io_uring_cmd_complete_in_task(ioucmd, nvme_uring_task_meta_cb); - - return RQ_END_IO_NONE; -} - static int nvme_uring_cmd_io(struct nvme_ctrl *ctrl, struct nvme_ns *ns, struct io_uring_cmd *ioucmd, unsigned int issue_flags, bool vec) { struct nvme_uring_cmd_pdu *pdu = nvme_uring_cmd_pdu(ioucmd); - const struct nvme_uring_cmd *cmd = ioucmd->cmd; + const struct nvme_uring_cmd *cmd = io_uring_sqe_cmd(ioucmd->sqe); struct request_queue *q = ns ? ns->queue : ctrl->admin_q; struct nvme_uring_data d; struct nvme_command c; + struct iov_iter iter; + struct iov_iter *map_iter = NULL; struct request *req; blk_opf_t rq_flags = 0; blk_mq_req_flags_t blk_flags = 0; - void *meta = NULL; int ret; c.common.opcode = READ_ONCE(cmd->opcode); @@ -580,7 +472,7 @@ static int nvme_uring_cmd_io(struct nvme_ctrl *ctrl, struct nvme_ns *ns, c.common.cdw14 = cpu_to_le32(READ_ONCE(cmd->cdw14)); c.common.cdw15 = cpu_to_le32(READ_ONCE(cmd->cdw15)); - if (!nvme_cmd_allowed(ns, &c, 0, ioucmd->file->f_mode)) + if (!nvme_cmd_allowed(ns, &c, 0, ioucmd->file->f_mode & FMODE_WRITE)) return -EACCES; d.metadata = READ_ONCE(cmd->metadata); @@ -589,51 +481,53 @@ static int nvme_uring_cmd_io(struct nvme_ctrl *ctrl, struct nvme_ns *ns, d.metadata_len = READ_ONCE(cmd->metadata_len); d.timeout_ms = READ_ONCE(cmd->timeout_ms); + if (d.data_len && (ioucmd->flags & IORING_URING_CMD_FIXED)) { + int ddir = nvme_is_write(&c) ? WRITE : READ; + + if (vec) + ret = io_uring_cmd_import_fixed_vec(ioucmd, + u64_to_user_ptr(d.addr), d.data_len, + ddir, &iter, issue_flags); + else + ret = io_uring_cmd_import_fixed(d.addr, d.data_len, + ddir, &iter, ioucmd, issue_flags); + if (ret < 0) + return ret; + + map_iter = &iter; + } + if (issue_flags & IO_URING_F_NONBLOCK) { - rq_flags = REQ_NOWAIT; + rq_flags |= REQ_NOWAIT; blk_flags = BLK_MQ_REQ_NOWAIT; } if (issue_flags & IO_URING_F_IOPOLL) rq_flags |= REQ_POLLED; -retry: req = nvme_alloc_user_request(q, &c, rq_flags, blk_flags); if (IS_ERR(req)) return PTR_ERR(req); req->timeout = d.timeout_ms ? msecs_to_jiffies(d.timeout_ms) : 0; - if (d.addr && d.data_len) { - ret = nvme_map_user_request(req, d.addr, - d.data_len, nvme_to_user_ptr(d.metadata), - d.metadata_len, 0, &meta, ioucmd, vec); + if (d.data_len) { + ret = nvme_map_user_request(req, d.addr, d.data_len, + nvme_to_user_ptr(d.metadata), d.metadata_len, + map_iter, vec ? NVME_IOCTL_VEC : 0); if (ret) - return ret; + goto out_free_req; } - if (issue_flags & IO_URING_F_IOPOLL && rq_flags & REQ_POLLED) { - if (unlikely(!req->bio)) { - /* we can't poll this, so alloc regular req instead */ - blk_mq_free_request(req); - rq_flags &= ~REQ_POLLED; - goto retry; - } else { - WRITE_ONCE(ioucmd->cookie, req->bio); - req->bio->bi_opf |= REQ_POLLED; - } - } /* to free bio on completion, as req->bio will be null at that time */ pdu->bio = req->bio; - pdu->meta_len = d.metadata_len; + pdu->req = req; req->end_io_data = ioucmd; - if (pdu->meta_len) { - pdu->u.meta = meta; - pdu->u.meta_buffer = nvme_to_user_ptr(d.metadata); - req->end_io = nvme_uring_cmd_end_io_meta; - } else { - req->end_io = nvme_uring_cmd_end_io; - } + req->end_io = nvme_uring_cmd_end_io; blk_execute_rq_nowait(req, false); return -EIOCBQUEUED; + +out_free_req: + blk_mq_free_request(req); + return ret; } static bool is_ctrl_ioctl(unsigned int cmd) @@ -646,13 +540,13 @@ static bool is_ctrl_ioctl(unsigned int cmd) } static int nvme_ctrl_ioctl(struct nvme_ctrl *ctrl, unsigned int cmd, - void __user *argp, fmode_t mode) + void __user *argp, bool open_for_write) { switch (cmd) { case NVME_IOCTL_ADMIN_CMD: - return nvme_user_cmd(ctrl, NULL, argp, 0, mode); + return nvme_user_cmd(ctrl, NULL, argp, 0, open_for_write); case NVME_IOCTL_ADMIN64_CMD: - return nvme_user_cmd64(ctrl, NULL, argp, 0, mode); + return nvme_user_cmd64(ctrl, NULL, argp, 0, open_for_write); default: return sed_ioctl(ctrl->opal_dev, cmd, argp); } @@ -677,14 +571,14 @@ struct nvme_user_io32 { #endif /* COMPAT_FOR_U64_ALIGNMENT */ static int nvme_ns_ioctl(struct nvme_ns *ns, unsigned int cmd, - void __user *argp, unsigned int flags, fmode_t mode) + void __user *argp, unsigned int flags, bool open_for_write) { switch (cmd) { case NVME_IOCTL_ID: force_successful_syscall_return(); return ns->head->ns_id; case NVME_IOCTL_IO_CMD: - return nvme_user_cmd(ns->ctrl, ns, argp, flags, mode); + return nvme_user_cmd(ns->ctrl, ns, argp, flags, open_for_write); /* * struct nvme_user_io can have different padding on some 32-bit ABIs. * Just accept the compat version as all fields that are used are the @@ -699,16 +593,18 @@ static int nvme_ns_ioctl(struct nvme_ns *ns, unsigned int cmd, flags |= NVME_IOCTL_VEC; fallthrough; case NVME_IOCTL_IO64_CMD: - return nvme_user_cmd64(ns->ctrl, ns, argp, flags, mode); + return nvme_user_cmd64(ns->ctrl, ns, argp, flags, + open_for_write); default: return -ENOTTY; } } -int nvme_ioctl(struct block_device *bdev, fmode_t mode, +int nvme_ioctl(struct block_device *bdev, blk_mode_t mode, unsigned int cmd, unsigned long arg) { struct nvme_ns *ns = bdev->bd_disk->private_data; + bool open_for_write = mode & BLK_OPEN_WRITE; void __user *argp = (void __user *)arg; unsigned int flags = 0; @@ -716,19 +612,20 @@ int nvme_ioctl(struct block_device *bdev, fmode_t mode, flags |= NVME_IOCTL_PARTITION; if (is_ctrl_ioctl(cmd)) - return nvme_ctrl_ioctl(ns->ctrl, cmd, argp, mode); - return nvme_ns_ioctl(ns, cmd, argp, flags, mode); + return nvme_ctrl_ioctl(ns->ctrl, cmd, argp, open_for_write); + return nvme_ns_ioctl(ns, cmd, argp, flags, open_for_write); } long nvme_ns_chr_ioctl(struct file *file, unsigned int cmd, unsigned long arg) { struct nvme_ns *ns = container_of(file_inode(file)->i_cdev, struct nvme_ns, cdev); + bool open_for_write = file->f_mode & FMODE_WRITE; void __user *argp = (void __user *)arg; if (is_ctrl_ioctl(cmd)) - return nvme_ctrl_ioctl(ns->ctrl, cmd, argp, file->f_mode); - return nvme_ns_ioctl(ns, cmd, argp, 0, file->f_mode); + return nvme_ctrl_ioctl(ns->ctrl, cmd, argp, open_for_write); + return nvme_ns_ioctl(ns, cmd, argp, 0, open_for_write); } static int nvme_uring_cmd_checks(unsigned int issue_flags) @@ -747,8 +644,6 @@ static int nvme_ns_uring_cmd(struct nvme_ns *ns, struct io_uring_cmd *ioucmd, struct nvme_ctrl *ctrl = ns->ctrl; int ret; - BUILD_BUG_ON(sizeof(struct nvme_uring_cmd_pdu) > sizeof(ioucmd->pdu)); - ret = nvme_uring_cmd_checks(issue_flags); if (ret) return ret; @@ -779,25 +674,17 @@ int nvme_ns_chr_uring_cmd_iopoll(struct io_uring_cmd *ioucmd, struct io_comp_batch *iob, unsigned int poll_flags) { - struct bio *bio; - int ret = 0; - struct nvme_ns *ns; - struct request_queue *q; + struct nvme_uring_cmd_pdu *pdu = nvme_uring_cmd_pdu(ioucmd); + struct request *req = pdu->req; - rcu_read_lock(); - bio = READ_ONCE(ioucmd->cookie); - ns = container_of(file_inode(ioucmd->file)->i_cdev, - struct nvme_ns, cdev); - q = ns->queue; - if (test_bit(QUEUE_FLAG_POLL, &q->queue_flags) && bio && bio->bi_bdev) - ret = bio_poll(bio, iob, poll_flags); - rcu_read_unlock(); - return ret; + if (req && blk_rq_is_poll(req)) + return blk_rq_poll(req, iob, poll_flags); + return 0; } #ifdef CONFIG_NVME_MULTIPATH static int nvme_ns_head_ctrl_ioctl(struct nvme_ns *ns, unsigned int cmd, void __user *argp, struct nvme_ns_head *head, int srcu_idx, - fmode_t mode) + bool open_for_write) __releases(&head->srcu) { struct nvme_ctrl *ctrl = ns->ctrl; @@ -805,16 +692,17 @@ static int nvme_ns_head_ctrl_ioctl(struct nvme_ns *ns, unsigned int cmd, nvme_get_ctrl(ns->ctrl); srcu_read_unlock(&head->srcu, srcu_idx); - ret = nvme_ctrl_ioctl(ns->ctrl, cmd, argp, mode); + ret = nvme_ctrl_ioctl(ns->ctrl, cmd, argp, open_for_write); nvme_put_ctrl(ctrl); return ret; } -int nvme_ns_head_ioctl(struct block_device *bdev, fmode_t mode, +int nvme_ns_head_ioctl(struct block_device *bdev, blk_mode_t mode, unsigned int cmd, unsigned long arg) { struct nvme_ns_head *head = bdev->bd_disk->private_data; + bool open_for_write = mode & BLK_OPEN_WRITE; void __user *argp = (void __user *)arg; struct nvme_ns *ns; int srcu_idx, ret = -EWOULDBLOCK; @@ -830,14 +718,14 @@ int nvme_ns_head_ioctl(struct block_device *bdev, fmode_t mode, /* * Handle ioctls that apply to the controller instead of the namespace - * seperately and drop the ns SRCU reference early. This avoids a + * separately and drop the ns SRCU reference early. This avoids a * deadlock when deleting namespaces using the passthrough interface. */ if (is_ctrl_ioctl(cmd)) return nvme_ns_head_ctrl_ioctl(ns, cmd, argp, head, srcu_idx, - mode); + open_for_write); - ret = nvme_ns_ioctl(ns, cmd, argp, flags, mode); + ret = nvme_ns_ioctl(ns, cmd, argp, flags, open_for_write); out_unlock: srcu_read_unlock(&head->srcu, srcu_idx); return ret; @@ -846,6 +734,7 @@ out_unlock: long nvme_ns_head_chr_ioctl(struct file *file, unsigned int cmd, unsigned long arg) { + bool open_for_write = file->f_mode & FMODE_WRITE; struct cdev *cdev = file_inode(file)->i_cdev; struct nvme_ns_head *head = container_of(cdev, struct nvme_ns_head, cdev); @@ -860,9 +749,9 @@ long nvme_ns_head_chr_ioctl(struct file *file, unsigned int cmd, if (is_ctrl_ioctl(cmd)) return nvme_ns_head_ctrl_ioctl(ns, cmd, argp, head, srcu_idx, - file->f_mode); + open_for_write); - ret = nvme_ns_ioctl(ns, cmd, argp, 0, file->f_mode); + ret = nvme_ns_ioctl(ns, cmd, argp, 0, open_for_write); out_unlock: srcu_read_unlock(&head->srcu, srcu_idx); return ret; @@ -882,31 +771,6 @@ int nvme_ns_head_chr_uring_cmd(struct io_uring_cmd *ioucmd, srcu_read_unlock(&head->srcu, srcu_idx); return ret; } - -int nvme_ns_head_chr_uring_cmd_iopoll(struct io_uring_cmd *ioucmd, - struct io_comp_batch *iob, - unsigned int poll_flags) -{ - struct cdev *cdev = file_inode(ioucmd->file)->i_cdev; - struct nvme_ns_head *head = container_of(cdev, struct nvme_ns_head, cdev); - int srcu_idx = srcu_read_lock(&head->srcu); - struct nvme_ns *ns = nvme_find_path(head); - struct bio *bio; - int ret = 0; - struct request_queue *q; - - if (ns) { - rcu_read_lock(); - bio = READ_ONCE(ioucmd->cookie); - q = ns->queue; - if (test_bit(QUEUE_FLAG_POLL, &q->queue_flags) && bio - && bio->bi_bdev) - ret = bio_poll(bio, iob, poll_flags); - rcu_read_unlock(); - } - srcu_read_unlock(&head->srcu, srcu_idx); - return ret; -} #endif /* CONFIG_NVME_MULTIPATH */ int nvme_dev_uring_cmd(struct io_uring_cmd *ioucmd, unsigned int issue_flags) @@ -937,18 +801,18 @@ int nvme_dev_uring_cmd(struct io_uring_cmd *ioucmd, unsigned int issue_flags) } static int nvme_dev_user_cmd(struct nvme_ctrl *ctrl, void __user *argp, - fmode_t mode) + bool open_for_write) { struct nvme_ns *ns; - int ret; + int ret, srcu_idx; - down_read(&ctrl->namespaces_rwsem); + srcu_idx = srcu_read_lock(&ctrl->srcu); if (list_empty(&ctrl->namespaces)) { ret = -ENOTTY; goto out_unlock; } - ns = list_first_entry(&ctrl->namespaces, struct nvme_ns, list); + ns = list_first_or_null_rcu(&ctrl->namespaces, struct nvme_ns, list); if (ns != list_last_entry(&ctrl->namespaces, struct nvme_ns, list)) { dev_warn(ctrl->device, "NVME_IOCTL_IO_CMD not supported when multiple namespaces present!\n"); @@ -958,31 +822,35 @@ static int nvme_dev_user_cmd(struct nvme_ctrl *ctrl, void __user *argp, dev_warn(ctrl->device, "using deprecated NVME_IOCTL_IO_CMD ioctl on the char device!\n"); - kref_get(&ns->kref); - up_read(&ctrl->namespaces_rwsem); + if (!nvme_get_ns(ns)) { + ret = -ENXIO; + goto out_unlock; + } + srcu_read_unlock(&ctrl->srcu, srcu_idx); - ret = nvme_user_cmd(ctrl, ns, argp, 0, mode); + ret = nvme_user_cmd(ctrl, ns, argp, 0, open_for_write); nvme_put_ns(ns); return ret; out_unlock: - up_read(&ctrl->namespaces_rwsem); + srcu_read_unlock(&ctrl->srcu, srcu_idx); return ret; } long nvme_dev_ioctl(struct file *file, unsigned int cmd, unsigned long arg) { + bool open_for_write = file->f_mode & FMODE_WRITE; struct nvme_ctrl *ctrl = file->private_data; void __user *argp = (void __user *)arg; switch (cmd) { case NVME_IOCTL_ADMIN_CMD: - return nvme_user_cmd(ctrl, NULL, argp, 0, file->f_mode); + return nvme_user_cmd(ctrl, NULL, argp, 0, open_for_write); case NVME_IOCTL_ADMIN64_CMD: - return nvme_user_cmd64(ctrl, NULL, argp, 0, file->f_mode); + return nvme_user_cmd64(ctrl, NULL, argp, 0, open_for_write); case NVME_IOCTL_IO_CMD: - return nvme_dev_user_cmd(ctrl, argp, file->f_mode); + return nvme_dev_user_cmd(ctrl, argp, open_for_write); case NVME_IOCTL_RESET: if (!capable(CAP_SYS_ADMIN)) return -EACCES; diff --git a/drivers/nvme/host/multipath.c b/drivers/nvme/host/multipath.c index fc39d01e7b63..174027d1cc19 100644 --- a/drivers/nvme/host/multipath.c +++ b/drivers/nvme/host/multipath.c @@ -10,13 +10,65 @@ #include "nvme.h" bool multipath = true; -module_param(multipath, bool, 0444); +static bool multipath_always_on; + +static int multipath_param_set(const char *val, const struct kernel_param *kp) +{ + int ret; + bool *arg = kp->arg; + + ret = param_set_bool(val, kp); + if (ret) + return ret; + + if (multipath_always_on && !*arg) { + pr_err("Can't disable multipath when multipath_always_on is configured.\n"); + *arg = true; + return -EINVAL; + } + + return 0; +} + +static const struct kernel_param_ops multipath_param_ops = { + .set = multipath_param_set, + .get = param_get_bool, +}; + +module_param_cb(multipath, &multipath_param_ops, &multipath, 0444); MODULE_PARM_DESC(multipath, "turn on native support for multiple controllers per subsystem"); +static int multipath_always_on_set(const char *val, + const struct kernel_param *kp) +{ + int ret; + bool *arg = kp->arg; + + ret = param_set_bool(val, kp); + if (ret < 0) + return ret; + + if (*arg) + multipath = true; + + return 0; +} + +static const struct kernel_param_ops multipath_always_on_ops = { + .set = multipath_always_on_set, + .get = param_get_bool, +}; + +module_param_cb(multipath_always_on, &multipath_always_on_ops, + &multipath_always_on, 0444); +MODULE_PARM_DESC(multipath_always_on, + "create multipath node always except for private namespace with non-unique nsid; note that this also implicitly enables native multipath support"); + static const char *nvme_iopolicy_names[] = { [NVME_IOPOLICY_NUMA] = "numa", [NVME_IOPOLICY_RR] = "round-robin", + [NVME_IOPOLICY_QD] = "queue-depth", }; static int iopolicy = NVME_IOPOLICY_NUMA; @@ -29,6 +81,8 @@ static int nvme_set_iopolicy(const char *val, const struct kernel_param *kp) iopolicy = NVME_IOPOLICY_NUMA; else if (!strncmp(val, "round-robin", 11)) iopolicy = NVME_IOPOLICY_RR; + else if (!strncmp(val, "queue-depth", 11)) + iopolicy = NVME_IOPOLICY_QD; else return -EINVAL; @@ -43,7 +97,7 @@ static int nvme_get_iopolicy(char *buf, const struct kernel_param *kp) module_param_call(iopolicy, nvme_set_iopolicy, nvme_get_iopolicy, &iopolicy, 0644); MODULE_PARM_DESC(iopolicy, - "Default multipath I/O policy; 'numa' (default) or 'round-robin'"); + "Default multipath I/O policy; 'numa' (default), 'round-robin' or 'queue-depth'"); void nvme_mpath_default_iopolicy(struct nvme_subsystem *subsys) { @@ -57,7 +111,7 @@ void nvme_mpath_unfreeze(struct nvme_subsystem *subsys) lockdep_assert_held(&subsys->lock); list_for_each_entry(h, &subsys->nsheads, entry) if (h->disk) - blk_mq_unfreeze_queue(h->disk->queue); + blk_mq_unfreeze_queue_nomemrestore(h->disk->queue); } void nvme_mpath_wait_freeze(struct nvme_subsystem *subsys) @@ -83,7 +137,7 @@ void nvme_mpath_start_freeze(struct nvme_subsystem *subsys) void nvme_failover_req(struct request *req) { struct nvme_ns *ns = req->q->queuedata; - u16 status = nvme_req(req)->status & 0x7ff; + u16 status = nvme_req(req)->status & NVME_SCT_SC_MASK; unsigned long flags; struct bio *bio; @@ -106,11 +160,20 @@ void nvme_failover_req(struct request *req) bio->bi_opf &= ~REQ_POLLED; bio->bi_cookie = BLK_QC_T_NONE; } + /* + * The alternate request queue that we may end up submitting + * the bio to may be frozen temporarily, in this case REQ_NOWAIT + * will fail the I/O immediately with EAGAIN to the issuer. + * We are not in the issuer context which cannot block. Clear + * the flag to avoid spurious EAGAIN I/O failures. + */ + bio->bi_opf &= ~REQ_NOWAIT; } blk_steal_bios(&ns->head->requeue_list, req); spin_unlock_irqrestore(&ns->head->requeue_lock, flags); - blk_mq_end_request(req, 0); + nvme_req(req)->status = 0; + nvme_end_req(req); kblockd_schedule_work(&ns->head->requeue_work); } @@ -119,13 +182,19 @@ void nvme_mpath_start_request(struct request *rq) struct nvme_ns *ns = rq->q->queuedata; struct gendisk *disk = ns->head->disk; - if (!blk_queue_io_stat(disk->queue) || blk_rq_is_passthrough(rq)) + if ((READ_ONCE(ns->head->subsys->iopolicy) == NVME_IOPOLICY_QD) && + !(nvme_req(rq)->flags & NVME_MPATH_CNT_ACTIVE)) { + atomic_inc(&ns->ctrl->nr_active); + nvme_req(rq)->flags |= NVME_MPATH_CNT_ACTIVE; + } + + if (!blk_queue_io_stat(disk->queue) || blk_rq_is_passthrough(rq) || + (nvme_req(rq)->flags & NVME_MPATH_IO_STATS)) return; nvme_req(rq)->flags |= NVME_MPATH_IO_STATS; - nvme_req(rq)->start_time = bdev_start_io_acct(disk->part0, - blk_rq_bytes(rq) >> SECTOR_SHIFT, - req_op(rq), jiffies); + nvme_req(rq)->start_time = bdev_start_io_acct(disk->part0, req_op(rq), + jiffies); } EXPORT_SYMBOL_GPL(nvme_mpath_start_request); @@ -133,25 +202,31 @@ void nvme_mpath_end_request(struct request *rq) { struct nvme_ns *ns = rq->q->queuedata; + if (nvme_req(rq)->flags & NVME_MPATH_CNT_ACTIVE) + atomic_dec_if_positive(&ns->ctrl->nr_active); + if (!(nvme_req(rq)->flags & NVME_MPATH_IO_STATS)) return; bdev_end_io_acct(ns->head->disk->part0, req_op(rq), - nvme_req(rq)->start_time); + blk_rq_bytes(rq) >> SECTOR_SHIFT, + nvme_req(rq)->start_time); } void nvme_kick_requeue_lists(struct nvme_ctrl *ctrl) { struct nvme_ns *ns; + int srcu_idx; - down_read(&ctrl->namespaces_rwsem); - list_for_each_entry(ns, &ctrl->namespaces, list) { + srcu_idx = srcu_read_lock(&ctrl->srcu); + list_for_each_entry_srcu(ns, &ctrl->namespaces, list, + srcu_read_lock_held(&ctrl->srcu)) { if (!ns->head->disk) continue; kblockd_schedule_work(&ns->head->requeue_work); - if (ctrl->state == NVME_CTRL_LIVE) + if (nvme_ctrl_state(ns->ctrl) == NVME_CTRL_LIVE) disk_uevent(ns->head->disk, KOBJ_CHANGE); } - up_read(&ctrl->namespaces_rwsem); + srcu_read_unlock(&ctrl->srcu, srcu_idx); } static const char *nvme_ana_state_names[] = { @@ -185,13 +260,15 @@ out: void nvme_mpath_clear_ctrl_paths(struct nvme_ctrl *ctrl) { struct nvme_ns *ns; + int srcu_idx; - down_read(&ctrl->namespaces_rwsem); - list_for_each_entry(ns, &ctrl->namespaces, list) { + srcu_idx = srcu_read_lock(&ctrl->srcu); + list_for_each_entry_srcu(ns, &ctrl->namespaces, list, + srcu_read_lock_held(&ctrl->srcu)) { nvme_mpath_clear_current_path(ns); kblockd_schedule_work(&ns->head->requeue_work); } - up_read(&ctrl->namespaces_rwsem); + srcu_read_unlock(&ctrl->srcu, srcu_idx); } void nvme_mpath_revalidate_paths(struct nvme_ns *ns) @@ -202,7 +279,8 @@ void nvme_mpath_revalidate_paths(struct nvme_ns *ns) int srcu_idx; srcu_idx = srcu_read_lock(&head->srcu); - list_for_each_entry_rcu(ns, &head->list, siblings) { + list_for_each_entry_srcu(ns, &head->list, siblings, + srcu_read_lock_held(&head->srcu)) { if (capacity != get_capacity(ns->disk)) clear_bit(NVME_NS_READY, &ns->flags); } @@ -215,13 +293,14 @@ void nvme_mpath_revalidate_paths(struct nvme_ns *ns) static bool nvme_path_is_disabled(struct nvme_ns *ns) { + enum nvme_ctrl_state state = nvme_ctrl_state(ns->ctrl); + /* * We don't treat NVME_CTRL_DELETING as a disabled path as I/O should * still be able to complete assuming that the controller is connected. * Otherwise it will fail immediately and return to the requeue list. */ - if (ns->ctrl->state != NVME_CTRL_LIVE && - ns->ctrl->state != NVME_CTRL_DELETING) + if (state != NVME_CTRL_LIVE && state != NVME_CTRL_DELETING) return true; if (test_bit(NVME_NS_ANA_PENDING, &ns->flags) || !test_bit(NVME_NS_READY, &ns->flags)) @@ -234,11 +313,13 @@ static struct nvme_ns *__nvme_find_path(struct nvme_ns_head *head, int node) int found_distance = INT_MAX, fallback_distance = INT_MAX, distance; struct nvme_ns *found = NULL, *fallback = NULL, *ns; - list_for_each_entry_rcu(ns, &head->list, siblings) { + list_for_each_entry_srcu(ns, &head->list, siblings, + srcu_read_lock_held(&head->srcu)) { if (nvme_path_is_disabled(ns)) continue; - if (READ_ONCE(head->subsys->iopolicy) == NVME_IOPOLICY_NUMA) + if (ns->ctrl->numa_node != NUMA_NO_NODE && + READ_ONCE(head->subsys->iopolicy) == NVME_IOPOLICY_NUMA) distance = node_distance(node, ns->ctrl->numa_node); else distance = LOCAL_DISTANCE; @@ -278,10 +359,15 @@ static struct nvme_ns *nvme_next_ns(struct nvme_ns_head *head, return list_first_or_null_rcu(&head->list, struct nvme_ns, siblings); } -static struct nvme_ns *nvme_round_robin_path(struct nvme_ns_head *head, - int node, struct nvme_ns *old) +static struct nvme_ns *nvme_round_robin_path(struct nvme_ns_head *head) { struct nvme_ns *ns, *found = NULL; + int node = numa_node_id(); + struct nvme_ns *old = srcu_dereference(head->current_path[node], + &head->srcu); + + if (unlikely(!old)) + return __nvme_find_path(head, node); if (list_is_singular(&head->list)) { if (nvme_path_is_disabled(old)) @@ -321,13 +407,50 @@ out: return found; } +static struct nvme_ns *nvme_queue_depth_path(struct nvme_ns_head *head) +{ + struct nvme_ns *best_opt = NULL, *best_nonopt = NULL, *ns; + unsigned int min_depth_opt = UINT_MAX, min_depth_nonopt = UINT_MAX; + unsigned int depth; + + list_for_each_entry_srcu(ns, &head->list, siblings, + srcu_read_lock_held(&head->srcu)) { + if (nvme_path_is_disabled(ns)) + continue; + + depth = atomic_read(&ns->ctrl->nr_active); + + switch (ns->ana_state) { + case NVME_ANA_OPTIMIZED: + if (depth < min_depth_opt) { + min_depth_opt = depth; + best_opt = ns; + } + break; + case NVME_ANA_NONOPTIMIZED: + if (depth < min_depth_nonopt) { + min_depth_nonopt = depth; + best_nonopt = ns; + } + break; + default: + break; + } + + if (min_depth_opt == 0) + return best_opt; + } + + return best_opt ? best_opt : best_nonopt; +} + static inline bool nvme_path_is_optimized(struct nvme_ns *ns) { - return ns->ctrl->state == NVME_CTRL_LIVE && + return nvme_ctrl_state(ns->ctrl) == NVME_CTRL_LIVE && ns->ana_state == NVME_ANA_OPTIMIZED; } -inline struct nvme_ns *nvme_find_path(struct nvme_ns_head *head) +static struct nvme_ns *nvme_numa_path(struct nvme_ns_head *head) { int node = numa_node_id(); struct nvme_ns *ns; @@ -335,32 +458,54 @@ inline struct nvme_ns *nvme_find_path(struct nvme_ns_head *head) ns = srcu_dereference(head->current_path[node], &head->srcu); if (unlikely(!ns)) return __nvme_find_path(head, node); - - if (READ_ONCE(head->subsys->iopolicy) == NVME_IOPOLICY_RR) - return nvme_round_robin_path(head, node, ns); if (unlikely(!nvme_path_is_optimized(ns))) return __nvme_find_path(head, node); return ns; } +inline struct nvme_ns *nvme_find_path(struct nvme_ns_head *head) +{ + switch (READ_ONCE(head->subsys->iopolicy)) { + case NVME_IOPOLICY_QD: + return nvme_queue_depth_path(head); + case NVME_IOPOLICY_RR: + return nvme_round_robin_path(head); + default: + return nvme_numa_path(head); + } +} + static bool nvme_available_path(struct nvme_ns_head *head) { struct nvme_ns *ns; - list_for_each_entry_rcu(ns, &head->list, siblings) { + if (!test_bit(NVME_NSHEAD_DISK_LIVE, &head->flags)) + return false; + + list_for_each_entry_srcu(ns, &head->list, siblings, + srcu_read_lock_held(&head->srcu)) { if (test_bit(NVME_CTRL_FAILFAST_EXPIRED, &ns->ctrl->flags)) continue; - switch (ns->ctrl->state) { + switch (nvme_ctrl_state(ns->ctrl)) { case NVME_CTRL_LIVE: case NVME_CTRL_RESETTING: case NVME_CTRL_CONNECTING: - /* fallthru */ return true; default: break; } } - return false; + + /* + * If "head->delayed_removal_secs" is configured (i.e., non-zero), do + * not immediately fail I/O. Instead, requeue the I/O for the configured + * duration, anticipating that if there's a transient link failure then + * it may recover within this time window. This parameter is exported to + * userspace via sysfs, and its default value is zero. It is internally + * mapped to NVME_NSHEAD_QUEUE_IF_NO_PATH. When delayed_removal_secs is + * non-zero, this flag is set to true. When zero, the flag is cleared. + */ + return nvme_mpath_queue_if_no_path(head); } static void nvme_ns_head_submit_bio(struct bio *bio) @@ -402,21 +547,36 @@ static void nvme_ns_head_submit_bio(struct bio *bio) srcu_read_unlock(&head->srcu, srcu_idx); } -static int nvme_ns_head_open(struct block_device *bdev, fmode_t mode) +static int nvme_ns_head_open(struct gendisk *disk, blk_mode_t mode) { - if (!nvme_tryget_ns_head(bdev->bd_disk->private_data)) + if (!nvme_tryget_ns_head(disk->private_data)) return -ENXIO; return 0; } -static void nvme_ns_head_release(struct gendisk *disk, fmode_t mode) +static void nvme_ns_head_release(struct gendisk *disk) { nvme_put_ns_head(disk->private_data); } +static int nvme_ns_head_get_unique_id(struct gendisk *disk, u8 id[16], + enum blk_unique_id type) +{ + struct nvme_ns_head *head = disk->private_data; + struct nvme_ns *ns; + int srcu_idx, ret = -EWOULDBLOCK; + + srcu_idx = srcu_read_lock(&head->srcu); + ns = nvme_find_path(head); + if (ns) + ret = nvme_ns_get_unique_id(ns, id, type); + srcu_read_unlock(&head->srcu, srcu_idx); + return ret; +} + #ifdef CONFIG_BLK_DEV_ZONED static int nvme_ns_head_report_zones(struct gendisk *disk, sector_t sector, - unsigned int nr_zones, report_zones_cb cb, void *data) + unsigned int nr_zones, struct blk_report_zones_args *args) { struct nvme_ns_head *head = disk->private_data; struct nvme_ns *ns; @@ -425,7 +585,7 @@ static int nvme_ns_head_report_zones(struct gendisk *disk, sector_t sector, srcu_idx = srcu_read_lock(&head->srcu); ns = nvme_find_path(head); if (ns) - ret = nvme_ns_report_zones(ns, sector, nr_zones, cb, data); + ret = nvme_ns_report_zones(ns, sector, nr_zones, args); srcu_read_unlock(&head->srcu, srcu_idx); return ret; } @@ -441,6 +601,7 @@ const struct block_device_operations nvme_ns_head_ops = { .ioctl = nvme_ns_head_ioctl, .compat_ioctl = blkdev_compat_ptr_ioctl, .getgeo = nvme_getgeo, + .get_unique_id = nvme_ns_head_get_unique_id, .report_zones = nvme_ns_head_report_zones, .pr_ops = &nvme_pr_ops, }; @@ -470,7 +631,7 @@ static const struct file_operations nvme_ns_head_chr_fops = { .unlocked_ioctl = nvme_ns_head_chr_ioctl, .compat_ioctl = compat_ptr_ioctl, .uring_cmd = nvme_ns_head_chr_uring_cmd, - .uring_cmd_iopoll = nvme_ns_head_chr_uring_cmd_iopoll, + .uring_cmd_iopoll = nvme_ns_chr_uring_cmd_iopoll, }; static int nvme_add_ns_head_cdev(struct nvme_ns_head *head) @@ -487,6 +648,20 @@ static int nvme_add_ns_head_cdev(struct nvme_ns_head *head) return ret; } +static void nvme_partition_scan_work(struct work_struct *work) +{ + struct nvme_ns_head *head = + container_of(work, struct nvme_ns_head, partition_scan_work); + + if (WARN_ON_ONCE(!test_and_clear_bit(GD_SUPPRESS_PART_SCAN, + &head->disk->state))) + return; + + mutex_lock(&head->disk->open_mutex); + bdev_disk_changed(head->disk, false); + mutex_unlock(&head->disk->open_mutex); +} + static void nvme_requeue_work(struct work_struct *work) { struct nvme_ns_head *head = @@ -505,54 +680,95 @@ static void nvme_requeue_work(struct work_struct *work) } } +static void nvme_remove_head(struct nvme_ns_head *head) +{ + if (test_and_clear_bit(NVME_NSHEAD_DISK_LIVE, &head->flags)) { + /* + * requeue I/O after NVME_NSHEAD_DISK_LIVE has been cleared + * to allow multipath to fail all I/O. + */ + kblockd_schedule_work(&head->requeue_work); + + nvme_cdev_del(&head->cdev, &head->cdev_device); + synchronize_srcu(&head->srcu); + del_gendisk(head->disk); + } + nvme_put_ns_head(head); +} + +static void nvme_remove_head_work(struct work_struct *work) +{ + struct nvme_ns_head *head = container_of(to_delayed_work(work), + struct nvme_ns_head, remove_work); + bool remove = false; + + mutex_lock(&head->subsys->lock); + if (list_empty(&head->list)) { + list_del_init(&head->entry); + remove = true; + } + mutex_unlock(&head->subsys->lock); + if (remove) + nvme_remove_head(head); + + module_put(THIS_MODULE); +} + int nvme_mpath_alloc_disk(struct nvme_ctrl *ctrl, struct nvme_ns_head *head) { - bool vwc = false; + struct queue_limits lim; mutex_init(&head->lock); bio_list_init(&head->requeue_list); spin_lock_init(&head->requeue_lock); INIT_WORK(&head->requeue_work, nvme_requeue_work); + INIT_WORK(&head->partition_scan_work, nvme_partition_scan_work); + INIT_DELAYED_WORK(&head->remove_work, nvme_remove_head_work); + head->delayed_removal_secs = 0; /* - * Add a multipath node if the subsystems supports multiple controllers. - * We also do this for private namespaces as the namespace sharing flag - * could change after a rescan. + * If "multipath_always_on" is enabled, a multipath node is added + * regardless of whether the disk is single/multi ported, and whether + * the namespace is shared or private. If "multipath_always_on" is not + * enabled, a multipath node is added only if the subsystem supports + * multiple controllers and the "multipath" option is configured. In + * either case, for private namespaces, we ensure that the NSID is + * unique. */ - if (!(ctrl->subsys->cmic & NVME_CTRL_CMIC_MULTI_CTRL) || - !nvme_is_unique_nsid(ctrl, head) || !multipath) + if (!multipath_always_on) { + if (!(ctrl->subsys->cmic & NVME_CTRL_CMIC_MULTI_CTRL) || + !multipath) + return 0; + } + + if (!nvme_is_unique_nsid(ctrl, head)) return 0; - head->disk = blk_alloc_disk(ctrl->numa_node); - if (!head->disk) - return -ENOMEM; + blk_set_stacking_limits(&lim); + lim.dma_alignment = 3; + lim.features |= BLK_FEAT_IO_STAT | BLK_FEAT_NOWAIT | + BLK_FEAT_POLL | BLK_FEAT_ATOMIC_WRITES; + if (head->ids.csi == NVME_CSI_ZNS) + lim.features |= BLK_FEAT_ZONED; + + head->disk = blk_alloc_disk(&lim, ctrl->numa_node); + if (IS_ERR(head->disk)) + return PTR_ERR(head->disk); head->disk->fops = &nvme_ns_head_ops; head->disk->private_data = head; - sprintf(head->disk->disk_name, "nvme%dn%d", - ctrl->subsys->instance, head->instance); - blk_queue_flag_set(QUEUE_FLAG_NONROT, head->disk->queue); - blk_queue_flag_set(QUEUE_FLAG_NOWAIT, head->disk->queue); - blk_queue_flag_set(QUEUE_FLAG_IO_STAT, head->disk->queue); /* - * This assumes all controllers that refer to a namespace either - * support poll queues or not. That is not a strict guarantee, - * but if the assumption is wrong the effect is only suboptimal - * performance but not correctness problem. + * We need to suppress the partition scan from occuring within the + * controller's scan_work context. If a path error occurs here, the IO + * will wait until a path becomes available or all paths are torn down, + * but that action also occurs within scan_work, so it would deadlock. + * Defer the partition scan to a different context that does not block + * scan_work. */ - if (ctrl->tagset->nr_maps > HCTX_TYPE_POLL && - ctrl->tagset->map[HCTX_TYPE_POLL].nr_queues) - blk_queue_flag_set(QUEUE_FLAG_POLL, head->disk->queue); - - /* set to a default value of 512 until the disk is validated */ - blk_queue_logical_block_size(head->disk->queue, 512); - blk_set_stacking_limits(&head->disk->queue->limits); - blk_queue_dma_alignment(head->disk->queue, 3); - - /* we need to propagate up the VMC settings */ - if (ctrl->vwc & NVME_CTRL_VWC_PRESENT) - vwc = true; - blk_queue_write_cache(head->disk->queue, vwc, vwc); + set_bit(GD_SUPPRESS_PART_SCAN, &head->disk->state); + sprintf(head->disk->disk_name, "nvme%dn%d", + ctrl->subsys->instance, head->instance); + nvme_tryget_ns_head(head); return 0; } @@ -571,20 +787,23 @@ static void nvme_mpath_set_live(struct nvme_ns *ns) */ if (!test_and_set_bit(NVME_NSHEAD_DISK_LIVE, &head->flags)) { rc = device_add_disk(&head->subsys->dev, head->disk, - nvme_ns_id_attr_groups); + nvme_ns_attr_groups); if (rc) { - clear_bit(NVME_NSHEAD_DISK_LIVE, &ns->flags); + clear_bit(NVME_NSHEAD_DISK_LIVE, &head->flags); return; } nvme_add_ns_head_cdev(head); + queue_work(nvme_wq, &head->partition_scan_work); } + nvme_mpath_add_sysfs_link(ns->head); + mutex_lock(&head->lock); if (nvme_path_is_optimized(ns)) { int node, srcu_idx; srcu_idx = srcu_read_lock(&head->srcu); - for_each_node(node) + for_each_online_node(node) __nvme_find_path(head, node); srcu_read_unlock(&head->srcu, srcu_idx); } @@ -659,8 +878,27 @@ static void nvme_update_ns_ana_state(struct nvme_ana_group_desc *desc, * controller is ready. */ if (nvme_state_is_live(ns->ana_state) && - ns->ctrl->state == NVME_CTRL_LIVE) + nvme_ctrl_state(ns->ctrl) == NVME_CTRL_LIVE) nvme_mpath_set_live(ns); + else { + /* + * Add sysfs link from multipath head gendisk node to path + * device gendisk node. + * If path's ana state is live (i.e. state is either optimized + * or non-optimized) while we alloc the ns then sysfs link would + * be created from nvme_mpath_set_live(). In that case we would + * not fallthrough this code path. However for the path's ana + * state other than live, we call nvme_mpath_set_live() only + * after ana state transitioned to the live state. But we still + * want to create the sysfs link from head node to a path device + * irrespctive of the path's ana state. + * If we reach through here then it means that path's ana state + * is not live but still create the sysfs link to this path from + * head node if head node of the path has already come alive. + */ + if (test_bit(NVME_NSHEAD_DISK_LIVE, &ns->head->flags)) + nvme_mpath_add_sysfs_link(ns->head); + } } static int nvme_update_ana_state(struct nvme_ctrl *ctrl, @@ -669,6 +907,7 @@ static int nvme_update_ana_state(struct nvme_ctrl *ctrl, u32 nr_nsids = le32_to_cpu(desc->nnsids), n = 0; unsigned *nr_change_groups = data; struct nvme_ns *ns; + int srcu_idx; dev_dbg(ctrl->device, "ANA group %d: %s.\n", le32_to_cpu(desc->grpid), @@ -680,8 +919,9 @@ static int nvme_update_ana_state(struct nvme_ctrl *ctrl, if (!nr_nsids) return 0; - down_read(&ctrl->namespaces_rwsem); - list_for_each_entry(ns, &ctrl->namespaces, list) { + srcu_idx = srcu_read_lock(&ctrl->srcu); + list_for_each_entry_srcu(ns, &ctrl->namespaces, list, + srcu_read_lock_held(&ctrl->srcu)) { unsigned nsid; again: nsid = le32_to_cpu(desc->nsids[n]); @@ -694,7 +934,7 @@ again: if (ns->head->ns_id > nsid) goto again; } - up_read(&ctrl->namespaces_rwsem); + srcu_read_unlock(&ctrl->srcu, srcu_idx); return 0; } @@ -730,7 +970,7 @@ static int nvme_read_ana_log(struct nvme_ctrl *ctrl) if (nr_change_groups) mod_timer(&ctrl->anatt_timer, ctrl->anatt * HZ * 2 + jiffies); else - del_timer_sync(&ctrl->anatt_timer); + timer_delete_sync(&ctrl->anatt_timer); out_unlock: mutex_unlock(&ctrl->ana_lock); return error; @@ -740,7 +980,7 @@ static void nvme_ana_work(struct work_struct *work) { struct nvme_ctrl *ctrl = container_of(work, struct nvme_ctrl, ana_work); - if (ctrl->state != NVME_CTRL_LIVE) + if (nvme_ctrl_state(ctrl) != NVME_CTRL_LIVE) return; nvme_read_ana_log(ctrl); @@ -760,7 +1000,7 @@ void nvme_mpath_update(struct nvme_ctrl *ctrl) static void nvme_anatt_timeout(struct timer_list *t) { - struct nvme_ctrl *ctrl = from_timer(ctrl, t, anatt_timer); + struct nvme_ctrl *ctrl = timer_container_of(ctrl, t, anatt_timer); dev_info(ctrl->device, "ANATT timeout, resetting controller.\n"); nvme_reset_ctrl(ctrl); @@ -770,7 +1010,7 @@ void nvme_mpath_stop(struct nvme_ctrl *ctrl) { if (!nvme_ctrl_use_ana(ctrl)) return; - del_timer_sync(&ctrl->anatt_timer); + timer_delete_sync(&ctrl->anatt_timer); cancel_work_sync(&ctrl->ana_work); } @@ -788,6 +1028,29 @@ static ssize_t nvme_subsys_iopolicy_show(struct device *dev, nvme_iopolicy_names[READ_ONCE(subsys->iopolicy)]); } +static void nvme_subsys_iopolicy_update(struct nvme_subsystem *subsys, + int iopolicy) +{ + struct nvme_ctrl *ctrl; + int old_iopolicy = READ_ONCE(subsys->iopolicy); + + if (old_iopolicy == iopolicy) + return; + + WRITE_ONCE(subsys->iopolicy, iopolicy); + + /* iopolicy changes clear the mpath by design */ + mutex_lock(&nvme_subsystems_lock); + list_for_each_entry(ctrl, &subsys->ctrls, subsys_entry) + nvme_mpath_clear_ctrl_paths(ctrl); + mutex_unlock(&nvme_subsystems_lock); + + pr_notice("subsysnqn %s iopolicy changed from %s to %s\n", + subsys->subnqn, + nvme_iopolicy_names[old_iopolicy], + nvme_iopolicy_names[iopolicy]); +} + static ssize_t nvme_subsys_iopolicy_store(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { @@ -797,7 +1060,7 @@ static ssize_t nvme_subsys_iopolicy_store(struct device *dev, for (i = 0; i < ARRAY_SIZE(nvme_iopolicy_names); i++) { if (sysfs_streq(buf, nvme_iopolicy_names[i])) { - WRITE_ONCE(subsys->iopolicy, i); + nvme_subsys_iopolicy_update(subsys, i); return count; } } @@ -823,6 +1086,88 @@ static ssize_t ana_state_show(struct device *dev, struct device_attribute *attr, } DEVICE_ATTR_RO(ana_state); +static ssize_t queue_depth_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct nvme_ns *ns = nvme_get_ns_from_dev(dev); + + if (ns->head->subsys->iopolicy != NVME_IOPOLICY_QD) + return 0; + + return sysfs_emit(buf, "%d\n", atomic_read(&ns->ctrl->nr_active)); +} +DEVICE_ATTR_RO(queue_depth); + +static ssize_t numa_nodes_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + int node, srcu_idx; + nodemask_t numa_nodes; + struct nvme_ns *current_ns; + struct nvme_ns *ns = nvme_get_ns_from_dev(dev); + struct nvme_ns_head *head = ns->head; + + if (head->subsys->iopolicy != NVME_IOPOLICY_NUMA) + return 0; + + nodes_clear(numa_nodes); + + srcu_idx = srcu_read_lock(&head->srcu); + for_each_node(node) { + current_ns = srcu_dereference(head->current_path[node], + &head->srcu); + if (ns == current_ns) + node_set(node, numa_nodes); + } + srcu_read_unlock(&head->srcu, srcu_idx); + + return sysfs_emit(buf, "%*pbl\n", nodemask_pr_args(&numa_nodes)); +} +DEVICE_ATTR_RO(numa_nodes); + +static ssize_t delayed_removal_secs_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct gendisk *disk = dev_to_disk(dev); + struct nvme_ns_head *head = disk->private_data; + int ret; + + mutex_lock(&head->subsys->lock); + ret = sysfs_emit(buf, "%u\n", head->delayed_removal_secs); + mutex_unlock(&head->subsys->lock); + return ret; +} + +static ssize_t delayed_removal_secs_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t count) +{ + struct gendisk *disk = dev_to_disk(dev); + struct nvme_ns_head *head = disk->private_data; + unsigned int sec; + int ret; + + ret = kstrtouint(buf, 0, &sec); + if (ret < 0) + return ret; + + mutex_lock(&head->subsys->lock); + head->delayed_removal_secs = sec; + if (sec) + set_bit(NVME_NSHEAD_QUEUE_IF_NO_PATH, &head->flags); + else + clear_bit(NVME_NSHEAD_QUEUE_IF_NO_PATH, &head->flags); + mutex_unlock(&head->subsys->lock); + /* + * Ensure that update to NVME_NSHEAD_QUEUE_IF_NO_PATH is seen + * by its reader. + */ + synchronize_srcu(&head->srcu); + + return count; +} + +DEVICE_ATTR_RW(delayed_removal_secs); + static int nvme_lookup_ana_group_desc(struct nvme_ctrl *ctrl, struct nvme_ana_group_desc *desc, void *data) { @@ -835,6 +1180,85 @@ static int nvme_lookup_ana_group_desc(struct nvme_ctrl *ctrl, return -ENXIO; /* just break out of the loop */ } +void nvme_mpath_add_sysfs_link(struct nvme_ns_head *head) +{ + struct device *target; + int rc, srcu_idx; + struct nvme_ns *ns; + struct kobject *kobj; + + /* + * Ensure head disk node is already added otherwise we may get invalid + * kobj for head disk node + */ + if (!test_bit(GD_ADDED, &head->disk->state)) + return; + + kobj = &disk_to_dev(head->disk)->kobj; + + /* + * loop through each ns chained through the head->list and create the + * sysfs link from head node to the ns path node + */ + srcu_idx = srcu_read_lock(&head->srcu); + + list_for_each_entry_srcu(ns, &head->list, siblings, + srcu_read_lock_held(&head->srcu)) { + /* + * Ensure that ns path disk node is already added otherwise we + * may get invalid kobj name for target + */ + if (!test_bit(GD_ADDED, &ns->disk->state)) + continue; + + /* + * Avoid creating link if it already exists for the given path. + * When path ana state transitions from optimized to non- + * optimized or vice-versa, the nvme_mpath_set_live() is + * invoked which in truns call this function. Now if the sysfs + * link already exists for the given path and we attempt to re- + * create the link then sysfs code would warn about it loudly. + * So we evaluate NVME_NS_SYSFS_ATTR_LINK flag here to ensure + * that we're not creating duplicate link. + * The test_and_set_bit() is used because it is protecting + * against multiple nvme paths being simultaneously added. + */ + if (test_and_set_bit(NVME_NS_SYSFS_ATTR_LINK, &ns->flags)) + continue; + + target = disk_to_dev(ns->disk); + /* + * Create sysfs link from head gendisk kobject @kobj to the + * ns path gendisk kobject @target->kobj. + */ + rc = sysfs_add_link_to_group(kobj, nvme_ns_mpath_attr_group.name, + &target->kobj, dev_name(target)); + if (unlikely(rc)) { + dev_err(disk_to_dev(ns->head->disk), + "failed to create link to %s\n", + dev_name(target)); + clear_bit(NVME_NS_SYSFS_ATTR_LINK, &ns->flags); + } + } + + srcu_read_unlock(&head->srcu, srcu_idx); +} + +void nvme_mpath_remove_sysfs_link(struct nvme_ns *ns) +{ + struct device *target; + struct kobject *kobj; + + if (!test_bit(NVME_NS_SYSFS_ATTR_LINK, &ns->flags)) + return; + + target = disk_to_dev(ns->disk); + kobj = &disk_to_dev(ns->head->disk)->kobj; + sysfs_remove_link_from_group(kobj, nvme_ns_mpath_attr_group.name, + dev_name(target)); + clear_bit(NVME_NS_SYSFS_ATTR_LINK, &ns->flags); +} + void nvme_mpath_add_disk(struct nvme_ns *ns, __le32 anagrpid) { if (nvme_ctrl_use_ana(ns->ctrl)) { @@ -860,34 +1284,59 @@ void nvme_mpath_add_disk(struct nvme_ns *ns, __le32 anagrpid) nvme_mpath_set_live(ns); } - if (blk_queue_stable_writes(ns->queue) && ns->head->disk) - blk_queue_flag_set(QUEUE_FLAG_STABLE_WRITES, - ns->head->disk->queue); #ifdef CONFIG_BLK_DEV_ZONED if (blk_queue_is_zoned(ns->queue) && ns->head->disk) ns->head->disk->nr_zones = ns->disk->nr_zones; #endif } -void nvme_mpath_shutdown_disk(struct nvme_ns_head *head) +void nvme_mpath_remove_disk(struct nvme_ns_head *head) { + bool remove = false; + if (!head->disk) return; - kblockd_schedule_work(&head->requeue_work); - if (test_bit(NVME_NSHEAD_DISK_LIVE, &head->flags)) { - nvme_cdev_del(&head->cdev, &head->cdev_device); - del_gendisk(head->disk); + + mutex_lock(&head->subsys->lock); + /* + * We are called when all paths have been removed, and at that point + * head->list is expected to be empty. However, nvme_remove_ns() and + * nvme_init_ns_head() can run concurrently and so if head->delayed_ + * removal_secs is configured, it is possible that by the time we reach + * this point, head->list may no longer be empty. Therefore, we recheck + * head->list here. If it is no longer empty then we skip enqueuing the + * delayed head removal work. + */ + if (!list_empty(&head->list)) + goto out; + + if (head->delayed_removal_secs) { + /* + * Ensure that no one could remove this module while the head + * remove work is pending. + */ + if (!try_module_get(THIS_MODULE)) + goto out; + mod_delayed_work(nvme_wq, &head->remove_work, + head->delayed_removal_secs * HZ); + } else { + list_del_init(&head->entry); + remove = true; } +out: + mutex_unlock(&head->subsys->lock); + if (remove) + nvme_remove_head(head); } -void nvme_mpath_remove_disk(struct nvme_ns_head *head) +void nvme_mpath_put_disk(struct nvme_ns_head *head) { if (!head->disk) return; - blk_mark_disk_dead(head->disk); /* make sure all pending bios are cleaned up */ kblockd_schedule_work(&head->requeue_work); flush_work(&head->requeue_work); + flush_work(&head->partition_scan_work); put_disk(head->disk); } @@ -909,6 +1358,9 @@ int nvme_mpath_init_identify(struct nvme_ctrl *ctrl, struct nvme_id_ctrl *id) !(ctrl->subsys->cmic & NVME_CTRL_CMIC_ANA)) return 0; + /* initialize this in the identify path to cover controller resets */ + atomic_set(&ctrl->nr_active, 0); + if (!ctrl->max_namespaces || ctrl->max_namespaces > le32_to_cpu(id->nn)) { dev_err(ctrl->device, diff --git a/drivers/nvme/host/nvme.h b/drivers/nvme/host/nvme.h index 424c8a467a0c..9a5f28c5103c 100644 --- a/drivers/nvme/host/nvme.h +++ b/drivers/nvme/host/nvme.h @@ -16,9 +16,12 @@ #include <linux/rcupdate.h> #include <linux/wait.h> #include <linux/t10-pi.h> +#include <linux/ratelimit_types.h> #include <trace/events/block.h> +extern const struct pr_ops nvme_pr_ops; + extern unsigned int nvme_io_timeout; #define NVME_IO_TIMEOUT (nvme_io_timeout * HZ) @@ -46,6 +49,7 @@ extern unsigned int admin_timeout; extern struct workqueue_struct *nvme_wq; extern struct workqueue_struct *nvme_reset_wq; extern struct workqueue_struct *nvme_delete_wq; +extern struct mutex nvme_subsystems_lock; /* * List of workarounds for devices that required behavior not specified in @@ -65,7 +69,7 @@ enum nvme_quirks { NVME_QUIRK_IDENTIFY_CNS = (1 << 1), /* - * The controller deterministically returns O's on reads to + * The controller deterministically returns 0's on reads to * logical blocks that deallocate was called on. */ NVME_QUIRK_DEALLOCATE_ZEROES = (1 << 2), @@ -87,6 +91,11 @@ enum nvme_quirks { NVME_QUIRK_NO_DEEPEST_PS = (1 << 5), /* + * Problems seen with concurrent commands + */ + NVME_QUIRK_QDEPTH_ONE = (1 << 6), + + /* * Set MEDIUM priority on SQ creation */ NVME_QUIRK_MEDIUM_PRIO_SQ = (1 << 7), @@ -149,6 +158,26 @@ enum nvme_quirks { * Reports garbage in the namespace identifiers (eui64, nguid, uuid). */ NVME_QUIRK_BOGUS_NID = (1 << 18), + + /* + * No temperature thresholds for channels other than 0 (Composite). + */ + NVME_QUIRK_NO_SECONDARY_TEMP_THRESH = (1 << 19), + + /* + * Disables simple suspend/resume path. + */ + NVME_QUIRK_FORCE_NO_SIMPLE_SUSPEND = (1 << 20), + + /* + * MSI (but not MSI-X) interrupts are broken and never fire. + */ + NVME_QUIRK_BROKEN_MSI = (1 << 21), + + /* + * Align dma pool segment size to 512 bytes + */ + NVME_QUIRK_DMAPOOL_ALIGN_512 = (1 << 22), }; /* @@ -177,6 +206,7 @@ enum { NVME_REQ_CANCELLED = (1 << 0), NVME_REQ_USERCMD = (1 << 1), NVME_MPATH_IO_STATS = (1 << 2), + NVME_MPATH_CNT_ACTIVE = (1 << 3), }; static inline struct nvme_request *nvme_req(struct request *req) @@ -242,12 +272,16 @@ enum nvme_ctrl_flags { NVME_CTRL_ADMIN_Q_STOPPED = 1, NVME_CTRL_STARTED_ONCE = 2, NVME_CTRL_STOPPED = 3, + NVME_CTRL_SKIP_ID_CNS_CS = 4, + NVME_CTRL_DIRTY_CAPABILITY = 5, + NVME_CTRL_FROZEN = 6, }; struct nvme_ctrl { bool comp_seen; - enum nvme_ctrl_state state; bool identified; + bool passthru_err_log_enabled; + enum nvme_ctrl_state state; spinlock_t lock; struct mutex scan_lock; const struct nvme_ctrl_ops *ops; @@ -260,7 +294,8 @@ struct nvme_ctrl { struct blk_mq_tag_set *tagset; struct blk_mq_tag_set *admin_tagset; struct list_head namespaces; - struct rw_semaphore namespaces_rwsem; + struct mutex namespaces_lock; + struct srcu_struct srcu; struct device ctrl_device; struct device *device; /* char device */ #ifdef CONFIG_NVME_HWMON @@ -276,25 +311,23 @@ struct nvme_ctrl { struct opal_dev *opal_dev; - char name[12]; u16 cntlid; - u32 ctrl_config; u16 mtfa; + u32 ctrl_config; u32 queue_count; u64 cap; u32 max_hw_sectors; u32 max_segments; u32 max_integrity_segments; - u32 max_discard_sectors; - u32 max_discard_segments; u32 max_zeroes_sectors; #ifdef CONFIG_BLK_DEV_ZONED u32 max_zone_append; #endif u16 crdt[3]; u16 oncs; + u8 dmrl; u32 dmrsl; u16 oacs; u16 sqsize; @@ -323,6 +356,7 @@ struct nvme_ctrl { struct delayed_work ka_work; struct delayed_work failfast_work; struct nvme_command ka_cmd; + unsigned long ka_last_check_time; struct work_struct fw_act_work; unsigned long events; @@ -337,9 +371,10 @@ struct nvme_ctrl { size_t ana_log_size; struct timer_list anatt_timer; struct work_struct ana_work; + atomic_t nr_active; #endif -#ifdef CONFIG_NVME_AUTH +#ifdef CONFIG_NVME_HOST_AUTH struct work_struct dhchap_auth_work; struct mutex dhchap_auth_mutex; struct nvme_dhchap_queue_context *dhchap_ctxs; @@ -347,16 +382,17 @@ struct nvme_ctrl { struct nvme_dhchap_key *ctrl_key; u16 transaction; #endif + key_serial_t tls_pskid; /* Power saving configuration */ u64 ps_max_latency_us; bool apst_enabled; /* PCIe only: */ + u16 hmmaxd; u32 hmpre; u32 hmmin; u32 hmminds; - u16 hmmaxd; /* Fabrics only */ u32 ioccsz; @@ -376,9 +412,15 @@ struct nvme_ctrl { enum nvme_dctype dctype; }; +static inline enum nvme_ctrl_state nvme_ctrl_state(struct nvme_ctrl *ctrl) +{ + return READ_ONCE(ctrl->state); +} + enum nvme_iopolicy { NVME_IOPOLICY_NUMA, NVME_IOPOLICY_RR, + NVME_IOPOLICY_QD, }; struct nvme_subsystem { @@ -400,7 +442,7 @@ struct nvme_subsystem { u8 cmic; enum nvme_subsys_type subtype; u16 vendor_id; - u16 awupf; /* 0's based awupf value. */ + u16 awupf; /* 0's based value. */ struct ida ns_ida; #ifdef CONFIG_NVME_MULTIPATH enum nvme_iopolicy iopolicy; @@ -428,25 +470,46 @@ struct nvme_ns_head { struct list_head list; struct srcu_struct srcu; struct nvme_subsystem *subsys; - unsigned ns_id; struct nvme_ns_ids ids; + u8 lba_shift; + u16 ms; + u16 pi_size; + u8 pi_type; + u8 guard_type; struct list_head entry; struct kref ref; bool shared; - int instance; + bool rotational; + bool passthru_err_log_enabled; struct nvme_effects_log *effects; + u64 nuse; + unsigned ns_id; + int instance; +#ifdef CONFIG_BLK_DEV_ZONED + u64 zsze; +#endif + unsigned long features; + + struct ratelimit_state rs_nuse; struct cdev cdev; struct device cdev_device; struct gendisk *disk; + + u16 nr_plids; + u16 *plids; #ifdef CONFIG_NVME_MULTIPATH struct bio_list requeue_list; spinlock_t requeue_lock; struct work_struct requeue_work; + struct work_struct partition_scan_work; struct mutex lock; unsigned long flags; -#define NVME_NSHEAD_DISK_LIVE 0 + struct delayed_work remove_work; + unsigned int delayed_removal_secs; +#define NVME_NSHEAD_DISK_LIVE 0 +#define NVME_NSHEAD_QUEUE_IF_NO_PATH 1 struct nvme_ns __rcu *current_path[]; #endif }; @@ -459,7 +522,7 @@ static inline bool nvme_ns_head_multipath(struct nvme_ns_head *head) enum nvme_ns_features { NVME_NS_EXT_LBAS = 1 << 0, /* support extended LBA format */ NVME_NS_METADATA_SUPPORTED = 1 << 1, /* support getting generated md */ - NVME_NS_DEAC, /* DEAC bit in Write Zeores supported */ + NVME_NS_DEAC = 1 << 2, /* DEAC bit in Write Zeroes supported */ }; struct nvme_ns { @@ -476,34 +539,29 @@ struct nvme_ns { struct kref kref; struct nvme_ns_head *head; - int lba_shift; - u16 ms; - u16 pi_size; - u16 sgs; - u32 sws; - u8 pi_type; - u8 guard_type; -#ifdef CONFIG_BLK_DEV_ZONED - u64 zsze; -#endif - unsigned long features; unsigned long flags; -#define NVME_NS_REMOVING 0 -#define NVME_NS_ANA_PENDING 2 -#define NVME_NS_FORCE_RO 3 -#define NVME_NS_READY 4 +#define NVME_NS_REMOVING 0 +#define NVME_NS_ANA_PENDING 2 +#define NVME_NS_FORCE_RO 3 +#define NVME_NS_READY 4 +#define NVME_NS_SYSFS_ATTR_LINK 5 struct cdev cdev; struct device cdev_device; struct nvme_fault_inject fault_inject; - }; /* NVMe ns supports metadata actions by the controller (generate/strip) */ -static inline bool nvme_ns_has_pi(struct nvme_ns *ns) +static inline bool nvme_ns_has_pi(struct nvme_ns_head *head) { - return ns->pi_type && ns->ms == ns->pi_size; + return head->pi_type && head->ms == head->pi_size; +} + +static inline unsigned long nvme_get_virt_boundary(struct nvme_ctrl *ctrl, + bool is_admin) +{ + return NVME_CTRL_PAGE_SIZE - 1; } struct nvme_ctrl_ops { @@ -520,11 +578,13 @@ struct nvme_ctrl_ops { int (*reg_read64)(struct nvme_ctrl *ctrl, u32 off, u64 *val); void (*free_ctrl)(struct nvme_ctrl *ctrl); void (*submit_async_event)(struct nvme_ctrl *ctrl); + int (*subsystem_reset)(struct nvme_ctrl *ctrl); void (*delete_ctrl)(struct nvme_ctrl *ctrl); void (*stop_ctrl)(struct nvme_ctrl *ctrl); int (*get_address)(struct nvme_ctrl *ctrl, char *buf, int size); void (*print_device_info)(struct nvme_ctrl *ctrl); bool (*supports_pci_p2pdma)(struct nvme_ctrl *ctrl); + unsigned long (*get_virt_boundary)(struct nvme_ctrl *ctrl, bool is_admin); }; /* @@ -618,34 +678,25 @@ int nvme_try_sched_reset(struct nvme_ctrl *ctrl); static inline int nvme_reset_subsystem(struct nvme_ctrl *ctrl) { - int ret; - - if (!ctrl->subsystem) + if (!ctrl->subsystem || !ctrl->ops->subsystem_reset) return -ENOTTY; - if (!nvme_wait_reset(ctrl)) - return -EBUSY; - - ret = ctrl->ops->reg_write32(ctrl, NVME_REG_NSSR, 0x4E564D65); - if (ret) - return ret; - - return nvme_try_sched_reset(ctrl); + return ctrl->ops->subsystem_reset(ctrl); } /* * Convert a 512B sector number to a device logical block number. */ -static inline u64 nvme_sect_to_lba(struct nvme_ns *ns, sector_t sector) +static inline u64 nvme_sect_to_lba(struct nvme_ns_head *head, sector_t sector) { - return sector >> (ns->lba_shift - SECTOR_SHIFT); + return sector >> (head->lba_shift - SECTOR_SHIFT); } /* * Convert a device logical block number to a 512B sector number. */ -static inline sector_t nvme_lba_to_sect(struct nvme_ns *ns, u64 lba) +static inline sector_t nvme_lba_to_sect(struct nvme_ns_head *head, u64 lba) { - return lba << (ns->lba_shift - SECTOR_SHIFT); + return lba << (head->lba_shift - SECTOR_SHIFT); } /* @@ -658,7 +709,7 @@ static inline u32 nvme_bytes_to_numd(size_t len) static inline bool nvme_is_ana_error(u16 status) { - switch (status & 0x7ff) { + switch (status & NVME_SCT_SC_MASK) { case NVME_SC_ANA_TRANSITION: case NVME_SC_ANA_INACCESSIBLE: case NVME_SC_ANA_PERSISTENT_LOSS: @@ -671,7 +722,7 @@ static inline bool nvme_is_ana_error(u16 status) static inline bool nvme_is_path_error(u16 status) { /* check for a status code type of 'path related status' */ - return (status & 0x700) == 0x300; + return (status & NVME_SCT_MASK) == NVME_SCT_PATH; } /* @@ -714,6 +765,28 @@ static inline bool nvme_is_aen_req(u16 qid, __u16 command_id) nvme_tag_from_cid(command_id) >= NVME_AQ_BLK_MQ_DEPTH; } +/* + * Returns true for sink states that can't ever transition back to live. + */ +static inline bool nvme_state_terminal(struct nvme_ctrl *ctrl) +{ + switch (nvme_ctrl_state(ctrl)) { + case NVME_CTRL_NEW: + case NVME_CTRL_LIVE: + case NVME_CTRL_RESETTING: + case NVME_CTRL_CONNECTING: + return false; + case NVME_CTRL_DELETING: + case NVME_CTRL_DELETING_NOIO: + case NVME_CTRL_DEAD: + return true; + default: + WARN_ONCE(1, "Unhandled ctrl state:%d", ctrl->state); + return true; + } +} + +void nvme_end_req(struct request *req); void nvme_complete_rq(struct request *req); void nvme_complete_batch_req(struct request *req); @@ -739,6 +812,7 @@ int nvme_disable_ctrl(struct nvme_ctrl *ctrl, bool shutdown); int nvme_enable_ctrl(struct nvme_ctrl *ctrl); int nvme_init_ctrl(struct nvme_ctrl *ctrl, struct device *dev, const struct nvme_ctrl_ops *ops, unsigned long quirks); +int nvme_add_ctrl(struct nvme_ctrl *ctrl); void nvme_uninit_ctrl(struct nvme_ctrl *ctrl); void nvme_start_ctrl(struct nvme_ctrl *ctrl); void nvme_stop_ctrl(struct nvme_ctrl *ctrl); @@ -780,17 +854,18 @@ blk_status_t nvme_setup_cmd(struct nvme_ns *ns, struct request *req); blk_status_t nvme_fail_nonready_command(struct nvme_ctrl *ctrl, struct request *req); bool __nvme_check_ready(struct nvme_ctrl *ctrl, struct request *rq, - bool queue_live); + bool queue_live, enum nvme_ctrl_state state); static inline bool nvme_check_ready(struct nvme_ctrl *ctrl, struct request *rq, bool queue_live) { - if (likely(ctrl->state == NVME_CTRL_LIVE)) + enum nvme_ctrl_state state = nvme_ctrl_state(ctrl); + + if (likely(state == NVME_CTRL_LIVE)) return true; - if (ctrl->ops->flags & NVME_F_FABRICS && - ctrl->state == NVME_CTRL_DELETING) + if (ctrl->ops->flags & NVME_F_FABRICS && state == NVME_CTRL_DELETING) return queue_live; - return __nvme_check_ready(ctrl, rq, queue_live); + return __nvme_check_ready(ctrl, rq, queue_live, state); } /* @@ -811,18 +886,33 @@ static inline bool nvme_is_unique_nsid(struct nvme_ctrl *ctrl, (ctrl->ctratt & NVME_CTRL_CTRATT_NVM_SETS); } +/* + * Flags for __nvme_submit_sync_cmd() + */ +typedef __u32 __bitwise nvme_submit_flags_t; + +enum { + /* Insert request at the head of the queue */ + NVME_SUBMIT_AT_HEAD = (__force nvme_submit_flags_t)(1 << 0), + /* Set BLK_MQ_REQ_NOWAIT when allocating request */ + NVME_SUBMIT_NOWAIT = (__force nvme_submit_flags_t)(1 << 1), + /* Set BLK_MQ_REQ_RESERVED when allocating request */ + NVME_SUBMIT_RESERVED = (__force nvme_submit_flags_t)(1 << 2), + /* Retry command when NVME_STATUS_DNR is not set in the result */ + NVME_SUBMIT_RETRY = (__force nvme_submit_flags_t)(1 << 3), +}; + int nvme_submit_sync_cmd(struct request_queue *q, struct nvme_command *cmd, void *buf, unsigned bufflen); int __nvme_submit_sync_cmd(struct request_queue *q, struct nvme_command *cmd, union nvme_result *result, void *buffer, unsigned bufflen, - int qid, int at_head, - blk_mq_req_flags_t flags); + int qid, nvme_submit_flags_t flags); int nvme_set_features(struct nvme_ctrl *dev, unsigned int fid, unsigned int dword11, void *buffer, size_t buflen, - u32 *result); + void *result); int nvme_get_features(struct nvme_ctrl *dev, unsigned int fid, unsigned int dword11, void *buffer, size_t buflen, - u32 *result); + void *result); int nvme_set_queue_count(struct nvme_ctrl *ctrl, int *count); void nvme_stop_keep_alive(struct nvme_ctrl *ctrl); int nvme_reset_ctrl(struct nvme_ctrl *ctrl); @@ -836,10 +926,10 @@ void nvme_put_ns_head(struct nvme_ns_head *head); int nvme_cdev_add(struct cdev *cdev, struct device *cdev_device, const struct file_operations *fops, struct module *owner); void nvme_cdev_del(struct cdev *cdev, struct device *cdev_device); -int nvme_ioctl(struct block_device *bdev, fmode_t mode, +int nvme_ioctl(struct block_device *bdev, blk_mode_t mode, unsigned int cmd, unsigned long arg); long nvme_ns_chr_ioctl(struct file *file, unsigned int cmd, unsigned long arg); -int nvme_ns_head_ioctl(struct block_device *bdev, fmode_t mode, +int nvme_ns_head_ioctl(struct block_device *bdev, blk_mode_t mode, unsigned int cmd, unsigned long arg); long nvme_ns_head_chr_ioctl(struct file *file, unsigned int cmd, unsigned long arg); @@ -847,20 +937,25 @@ long nvme_dev_ioctl(struct file *file, unsigned int cmd, unsigned long arg); int nvme_ns_chr_uring_cmd_iopoll(struct io_uring_cmd *ioucmd, struct io_comp_batch *iob, unsigned int poll_flags); -int nvme_ns_head_chr_uring_cmd_iopoll(struct io_uring_cmd *ioucmd, - struct io_comp_batch *iob, unsigned int poll_flags); int nvme_ns_chr_uring_cmd(struct io_uring_cmd *ioucmd, unsigned int issue_flags); int nvme_ns_head_chr_uring_cmd(struct io_uring_cmd *ioucmd, unsigned int issue_flags); -int nvme_getgeo(struct block_device *bdev, struct hd_geometry *geo); +int nvme_identify_ns(struct nvme_ctrl *ctrl, unsigned nsid, + struct nvme_id_ns **id); +int nvme_getgeo(struct gendisk *disk, struct hd_geometry *geo); int nvme_dev_uring_cmd(struct io_uring_cmd *ioucmd, unsigned int issue_flags); -extern const struct attribute_group *nvme_ns_id_attr_groups[]; +extern const struct attribute_group *nvme_ns_attr_groups[]; +extern const struct attribute_group nvme_ns_mpath_attr_group; extern const struct pr_ops nvme_pr_ops; extern const struct block_device_operations nvme_ns_head_ops; extern const struct attribute_group nvme_dev_attrs_group; +extern const struct attribute_group *nvme_subsys_attrs_groups[]; +extern const struct attribute_group *nvme_dev_attr_groups[]; +extern const struct block_device_operations nvme_bdev_ops; +void nvme_delete_ctrl_sync(struct nvme_ctrl *ctrl); struct nvme_ns *nvme_find_path(struct nvme_ns_head *head); #ifdef CONFIG_NVME_MULTIPATH static inline bool nvme_ctrl_use_ana(struct nvme_ctrl *ctrl) @@ -875,8 +970,10 @@ void nvme_mpath_default_iopolicy(struct nvme_subsystem *subsys); void nvme_failover_req(struct request *req); void nvme_kick_requeue_lists(struct nvme_ctrl *ctrl); int nvme_mpath_alloc_disk(struct nvme_ctrl *ctrl,struct nvme_ns_head *head); +void nvme_mpath_add_sysfs_link(struct nvme_ns_head *ns); +void nvme_mpath_remove_sysfs_link(struct nvme_ns *ns); void nvme_mpath_add_disk(struct nvme_ns *ns, __le32 anagrpid); -void nvme_mpath_remove_disk(struct nvme_ns_head *head); +void nvme_mpath_put_disk(struct nvme_ns_head *head); int nvme_mpath_init_identify(struct nvme_ctrl *ctrl, struct nvme_id_ctrl *id); void nvme_mpath_init_ctrl(struct nvme_ctrl *ctrl); void nvme_mpath_update(struct nvme_ctrl *ctrl); @@ -885,7 +982,7 @@ void nvme_mpath_stop(struct nvme_ctrl *ctrl); bool nvme_mpath_clear_current_path(struct nvme_ns *ns); void nvme_mpath_revalidate_paths(struct nvme_ns *ns); void nvme_mpath_clear_ctrl_paths(struct nvme_ctrl *ctrl); -void nvme_mpath_shutdown_disk(struct nvme_ns_head *head); +void nvme_mpath_remove_disk(struct nvme_ns_head *head); void nvme_mpath_start_request(struct request *rq); void nvme_mpath_end_request(struct request *rq); @@ -900,8 +997,21 @@ static inline void nvme_trace_bio_complete(struct request *req) extern bool multipath; extern struct device_attribute dev_attr_ana_grpid; extern struct device_attribute dev_attr_ana_state; +extern struct device_attribute dev_attr_queue_depth; +extern struct device_attribute dev_attr_numa_nodes; +extern struct device_attribute dev_attr_delayed_removal_secs; extern struct device_attribute subsys_attr_iopolicy; +static inline bool nvme_disk_is_ns_head(struct gendisk *disk) +{ + return disk->fops == &nvme_ns_head_ops; +} +static inline bool nvme_mpath_queue_if_no_path(struct nvme_ns_head *head) +{ + if (test_bit(NVME_NSHEAD_QUEUE_IF_NO_PATH, &head->flags)) + return true; + return false; +} #else #define multipath false static inline bool nvme_ctrl_use_ana(struct nvme_ctrl *ctrl) @@ -922,7 +1032,13 @@ static inline int nvme_mpath_alloc_disk(struct nvme_ctrl *ctrl, static inline void nvme_mpath_add_disk(struct nvme_ns *ns, __le32 anagrpid) { } -static inline void nvme_mpath_remove_disk(struct nvme_ns_head *head) +static inline void nvme_mpath_put_disk(struct nvme_ns_head *head) +{ +} +static inline void nvme_mpath_add_sysfs_link(struct nvme_ns *ns) +{ +} +static inline void nvme_mpath_remove_sysfs_link(struct nvme_ns *ns) { } static inline bool nvme_mpath_clear_current_path(struct nvme_ns *ns) @@ -935,7 +1051,7 @@ static inline void nvme_mpath_revalidate_paths(struct nvme_ns *ns) static inline void nvme_mpath_clear_ctrl_paths(struct nvme_ctrl *ctrl) { } -static inline void nvme_mpath_shutdown_disk(struct nvme_ns_head *head) +static inline void nvme_mpath_remove_disk(struct nvme_ns_head *head) { } static inline void nvme_trace_bio_complete(struct request *req) @@ -979,13 +1095,32 @@ static inline void nvme_mpath_start_request(struct request *rq) static inline void nvme_mpath_end_request(struct request *rq) { } +static inline bool nvme_disk_is_ns_head(struct gendisk *disk) +{ + return false; +} +static inline bool nvme_mpath_queue_if_no_path(struct nvme_ns_head *head) +{ + return false; +} #endif /* CONFIG_NVME_MULTIPATH */ -int nvme_revalidate_zones(struct nvme_ns *ns); +int nvme_ns_get_unique_id(struct nvme_ns *ns, u8 id[16], + enum blk_unique_id type); + +struct nvme_zone_info { + u64 zone_size; + unsigned int max_open_zones; + unsigned int max_active_zones; +}; + int nvme_ns_report_zones(struct nvme_ns *ns, sector_t sector, - unsigned int nr_zones, report_zones_cb cb, void *data); + unsigned int nr_zones, struct blk_report_zones_args *args); +int nvme_query_zone_info(struct nvme_ns *ns, unsigned lbaf, + struct nvme_zone_info *zi); +void nvme_update_zone_info(struct nvme_ns *ns, struct queue_limits *lim, + struct nvme_zone_info *zi); #ifdef CONFIG_BLK_DEV_ZONED -int nvme_update_zone_info(struct nvme_ns *ns, unsigned lbaf); blk_status_t nvme_setup_zone_mgmt_send(struct nvme_ns *ns, struct request *req, struct nvme_command *cmnd, enum nvme_zone_mgmt_action action); @@ -996,18 +1131,14 @@ static inline blk_status_t nvme_setup_zone_mgmt_send(struct nvme_ns *ns, { return BLK_STS_NOTSUPP; } - -static inline int nvme_update_zone_info(struct nvme_ns *ns, unsigned lbaf) -{ - dev_warn(ns->ctrl->device, - "Please enable CONFIG_BLK_DEV_ZONED to support ZNS devices\n"); - return -EPROTONOSUPPORT; -} #endif static inline struct nvme_ns *nvme_get_ns_from_dev(struct device *dev) { - return dev_to_disk(dev)->private_data; + struct gendisk *disk = dev_to_disk(dev); + + WARN_ON(nvme_disk_is_ns_head(disk)); + return disk->private_data; } #ifdef CONFIG_NVME_HWMON @@ -1033,10 +1164,18 @@ static inline void nvme_start_request(struct request *rq) static inline bool nvme_ctrl_sgl_supported(struct nvme_ctrl *ctrl) { - return ctrl->sgls & ((1 << 0) | (1 << 1)); + return ctrl->sgls & (NVME_CTRL_SGLS_BYTE_ALIGNED | + NVME_CTRL_SGLS_DWORD_ALIGNED); } -#ifdef CONFIG_NVME_AUTH +static inline bool nvme_ctrl_meta_sgl_supported(struct nvme_ctrl *ctrl) +{ + if (ctrl->ops->flags & NVME_F_FABRICS) + return true; + return ctrl->sgls & NVME_CTRL_SGLS_MSDS; +} + +#ifdef CONFIG_NVME_HOST_AUTH int __init nvme_init_auth(void); void __exit nvme_exit_auth(void); int nvme_auth_init_ctrl(struct nvme_ctrl *ctrl); @@ -1044,6 +1183,7 @@ void nvme_auth_stop(struct nvme_ctrl *ctrl); int nvme_auth_negotiate(struct nvme_ctrl *ctrl, int qid); int nvme_auth_wait(struct nvme_ctrl *ctrl, int qid); void nvme_auth_free(struct nvme_ctrl *ctrl); +void nvme_auth_revoke_tls_key(struct nvme_ctrl *ctrl); #else static inline int nvme_auth_init_ctrl(struct nvme_ctrl *ctrl) { @@ -1063,18 +1203,21 @@ static inline int nvme_auth_negotiate(struct nvme_ctrl *ctrl, int qid) } static inline int nvme_auth_wait(struct nvme_ctrl *ctrl, int qid) { - return NVME_SC_AUTH_REQUIRED; + return -EPROTONOSUPPORT; } static inline void nvme_auth_free(struct nvme_ctrl *ctrl) {}; +static inline void nvme_auth_revoke_tls_key(struct nvme_ctrl *ctrl) {}; #endif u32 nvme_command_effects(struct nvme_ctrl *ctrl, struct nvme_ns *ns, u8 opcode); -int nvme_execute_passthru_rq(struct request *rq, u32 *effects); -void nvme_passthru_end(struct nvme_ctrl *ctrl, u32 effects, +u32 nvme_passthru_start(struct nvme_ctrl *ctrl, struct nvme_ns *ns, u8 opcode); +int nvme_execute_rq(struct request *rq, bool at_head); +void nvme_passthru_end(struct nvme_ctrl *ctrl, struct nvme_ns *ns, u32 effects, struct nvme_command *cmd, int status); struct nvme_ctrl *nvme_ctrl_from_file(struct file *file); struct nvme_ns *nvme_find_get_ns(struct nvme_ctrl *ctrl, unsigned nsid); +bool nvme_get_ns(struct nvme_ns *ns); void nvme_put_ns(struct nvme_ns *ns); static inline bool nvme_multi_css(struct nvme_ctrl *ctrl) @@ -1082,23 +1225,4 @@ static inline bool nvme_multi_css(struct nvme_ctrl *ctrl) return (ctrl->ctrl_config & NVME_CC_CSS_MASK) == NVME_CC_CSS_CSI; } -#ifdef CONFIG_NVME_VERBOSE_ERRORS -const unsigned char *nvme_get_error_status_str(u16 status); -const unsigned char *nvme_get_opcode_str(u8 opcode); -const unsigned char *nvme_get_admin_opcode_str(u8 opcode); -#else /* CONFIG_NVME_VERBOSE_ERRORS */ -static inline const unsigned char *nvme_get_error_status_str(u16 status) -{ - return "I/O Error"; -} -static inline const unsigned char *nvme_get_opcode_str(u8 opcode) -{ - return "I/O Cmd"; -} -static inline const unsigned char *nvme_get_admin_opcode_str(u8 opcode) -{ - return "Admin Cmd"; -} -#endif /* CONFIG_NVME_VERBOSE_ERRORS */ - #endif /* _NVME_H */ diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c index c734934c407c..0e4caeab739c 100644 --- a/drivers/nvme/host/pci.c +++ b/drivers/nvme/host/pci.c @@ -5,11 +5,9 @@ */ #include <linux/acpi.h> -#include <linux/aer.h> #include <linux/async.h> #include <linux/blkdev.h> -#include <linux/blk-mq.h> -#include <linux/blk-mq-pci.h> +#include <linux/blk-mq-dma.h> #include <linux/blk-integrity.h> #include <linux/dmi.h> #include <linux/init.h> @@ -20,6 +18,7 @@ #include <linux/mm.h> #include <linux/module.h> #include <linux/mutex.h> +#include <linux/nodemask.h> #include <linux/once.h> #include <linux/pci.h> #include <linux/suspend.h> @@ -28,7 +27,6 @@ #include <linux/io-64-nonatomic-lo-hi.h> #include <linux/io-64-nonatomic-hi-lo.h> #include <linux/sed-opal.h> -#include <linux/pci-p2pdma.h> #include "trace.h" #include "nvme.h" @@ -36,14 +34,43 @@ #define SQ_SIZE(q) ((q)->q_depth << (q)->sqes) #define CQ_SIZE(q) ((q)->q_depth * sizeof(struct nvme_completion)) -#define SGES_PER_PAGE (NVME_CTRL_PAGE_SIZE / sizeof(struct nvme_sgl_desc)) +/* Optimisation for I/Os between 4k and 128k */ +#define NVME_SMALL_POOL_SIZE 256 /* - * These can be higher, but we need to ensure that any command doesn't - * require an sg allocation that needs more than a page of data. + * Arbitrary upper bound. */ -#define NVME_MAX_KB_SZ 4096 -#define NVME_MAX_SEGS 127 +#define NVME_MAX_BYTES SZ_8M +#define NVME_MAX_NR_DESCRIPTORS 5 + +/* + * For data SGLs we support a single descriptors worth of SGL entries. + * For PRPs, segments don't matter at all. + */ +#define NVME_MAX_SEGS \ + (NVME_CTRL_PAGE_SIZE / sizeof(struct nvme_sgl_desc)) + +/* + * For metadata SGLs, only the small descriptor is supported, and the first + * entry is the segment descriptor, which for the data pointer sits in the SQE. + */ +#define NVME_MAX_META_SEGS \ + ((NVME_SMALL_POOL_SIZE / sizeof(struct nvme_sgl_desc)) - 1) + +/* + * The last entry is used to link to the next descriptor. + */ +#define PRPS_PER_PAGE \ + (((NVME_CTRL_PAGE_SIZE / sizeof(__le64))) - 1) + +/* + * I/O could be non-aligned both at the beginning and end. + */ +#define MAX_PRP_RANGE \ + (NVME_MAX_BYTES + 2 * (NVME_CTRL_PAGE_SIZE - 1)) + +static_assert(MAX_PRP_RANGE / NVME_CTRL_PAGE_SIZE <= + (1 /* prp1 */ + NVME_MAX_NR_DESCRIPTORS * PRPS_PER_PAGE)); static int use_threaded_interrupts; module_param(use_threaded_interrupts, int, 0444); @@ -81,7 +108,7 @@ static int io_queue_count_set(const char *val, const struct kernel_param *kp) int ret; ret = kstrtouint(val, 10, &n); - if (ret != 0 || n > num_possible_cpus()) + if (ret != 0 || n > blk_mq_num_possible_queues(0)) return -EINVAL; return param_set_uint(val, kp); } @@ -110,6 +137,12 @@ struct nvme_queue; static void nvme_dev_disable(struct nvme_dev *dev, bool shutdown); static void nvme_delete_io_queues(struct nvme_dev *dev); +static void nvme_update_attrs(struct nvme_dev *dev); + +struct nvme_descriptor_pools { + struct dma_pool *large; + struct dma_pool *small; +}; /* * Represents an NVM Express device. Each nvme_dev is a PCI function. @@ -120,8 +153,6 @@ struct nvme_dev { struct blk_mq_tag_set admin_tagset; u32 __iomem *dbs; struct device *dev; - struct dma_pool *prp_page_pool; - struct dma_pool *prp_small_pool; unsigned online_queues; unsigned max_qid; unsigned io_queues[HCTX_MAX_TYPES]; @@ -140,8 +171,8 @@ struct nvme_dev { struct nvme_ctrl ctrl; u32 last_ps; bool hmb; - - mempool_t *iod_mempool; + struct sg_table *hmb_sgt; + mempool_t *dmavec_mempool; /* shadow doorbell buffer support: */ __le32 *dbbuf_dbs; @@ -152,12 +183,14 @@ struct nvme_dev { /* host memory buffer support: */ u64 host_mem_size; u32 nr_host_mem_descs; + u32 host_mem_descs_size; dma_addr_t host_mem_descs_dma; struct nvme_host_mem_buf_desc *host_mem_descs; void **host_mem_desc_bufs; unsigned int nr_allocated_queues; unsigned int nr_write_queues; unsigned int nr_poll_queues; + struct nvme_descriptor_pools descriptor_pools[]; }; static int io_queue_depth_set(const char *val, const struct kernel_param *kp) @@ -187,6 +220,7 @@ static inline struct nvme_dev *to_nvme_dev(struct nvme_ctrl *ctrl) */ struct nvme_queue { struct nvme_dev *dev; + struct nvme_descriptor_pools descriptor_pools; spinlock_t sq_lock; void *sq_cmds; /* only used for poll queues: */ @@ -215,23 +249,57 @@ struct nvme_queue { struct completion delete_done; }; +/* bits for iod->flags */ +enum nvme_iod_flags { + /* this command has been aborted by the timeout handler */ + IOD_ABORTED = 1U << 0, + + /* uses the small descriptor pool */ + IOD_SMALL_DESCRIPTOR = 1U << 1, + + /* single segment dma mapping */ + IOD_SINGLE_SEGMENT = 1U << 2, + + /* Data payload contains p2p memory */ + IOD_DATA_P2P = 1U << 3, + + /* Metadata contains p2p memory */ + IOD_META_P2P = 1U << 4, + + /* Data payload contains MMIO memory */ + IOD_DATA_MMIO = 1U << 5, + + /* Metadata contains MMIO memory */ + IOD_META_MMIO = 1U << 6, + + /* Metadata using non-coalesced MPTR */ + IOD_SINGLE_META_SEGMENT = 1U << 7, +}; + +struct nvme_dma_vec { + dma_addr_t addr; + unsigned int len; +}; + /* * The nvme_iod describes the data in an I/O. - * - * The sg pointer contains the list of PRP/SGL chunk allocations in addition - * to the actual struct scatterlist. */ struct nvme_iod { struct nvme_request req; struct nvme_command cmd; - bool use_sgl; - bool aborted; - s8 nr_allocations; /* PRP list pool allocations. 0 means small - pool in use */ - unsigned int dma_len; /* length of single DMA segment mapping */ - dma_addr_t first_dma; + u8 flags; + u8 nr_descriptors; + + unsigned int total_len; + struct dma_iova_state dma_state; + void *descriptors[NVME_MAX_NR_DESCRIPTORS]; + struct nvme_dma_vec *dma_vecs; + unsigned int nr_dma_vecs; + dma_addr_t meta_dma; - struct sg_table sgt; + unsigned int meta_total_len; + struct dma_iova_state meta_dma_state; + struct nvme_sgl_desc *meta_descriptor; }; static inline unsigned int nvme_dbbuf_size(struct nvme_dev *dev) @@ -361,7 +429,7 @@ static bool nvme_dbbuf_update_and_check_event(u16 value, __le32 *dbbuf_db, /* * Ensure that the doorbell is updated before reading the event * index from memory. The controller needs to provide similar - * ordering to ensure the envent index is updated before reading + * ordering to ensure the event index is updated before reading * the doorbell. */ mb(); @@ -374,60 +442,85 @@ static bool nvme_dbbuf_update_and_check_event(u16 value, __le32 *dbbuf_db, return true; } -/* - * Will slightly overestimate the number of pages needed. This is OK - * as it only leads to a small amount of wasted memory for the lifetime of - * the I/O. - */ -static int nvme_pci_npages_prp(void) +static struct nvme_descriptor_pools * +nvme_setup_descriptor_pools(struct nvme_dev *dev, unsigned numa_node) { - unsigned max_bytes = (NVME_MAX_KB_SZ * 1024) + NVME_CTRL_PAGE_SIZE; - unsigned nprps = DIV_ROUND_UP(max_bytes, NVME_CTRL_PAGE_SIZE); - return DIV_ROUND_UP(8 * nprps, NVME_CTRL_PAGE_SIZE - 8); + struct nvme_descriptor_pools *pools = &dev->descriptor_pools[numa_node]; + size_t small_align = NVME_SMALL_POOL_SIZE; + + if (pools->small) + return pools; /* already initialized */ + + pools->large = dma_pool_create_node("nvme descriptor page", dev->dev, + NVME_CTRL_PAGE_SIZE, NVME_CTRL_PAGE_SIZE, 0, numa_node); + if (!pools->large) + return ERR_PTR(-ENOMEM); + + if (dev->ctrl.quirks & NVME_QUIRK_DMAPOOL_ALIGN_512) + small_align = 512; + + pools->small = dma_pool_create_node("nvme descriptor small", dev->dev, + NVME_SMALL_POOL_SIZE, small_align, 0, numa_node); + if (!pools->small) { + dma_pool_destroy(pools->large); + pools->large = NULL; + return ERR_PTR(-ENOMEM); + } + + return pools; } -/* - * Calculates the number of pages needed for the SGL segments. For example a 4k - * page can accommodate 256 SGL descriptors. - */ -static int nvme_pci_npages_sgl(void) +static void nvme_release_descriptor_pools(struct nvme_dev *dev) { - return DIV_ROUND_UP(NVME_MAX_SEGS * sizeof(struct nvme_sgl_desc), - NVME_CTRL_PAGE_SIZE); + unsigned i; + + for (i = 0; i < nr_node_ids; i++) { + struct nvme_descriptor_pools *pools = &dev->descriptor_pools[i]; + + dma_pool_destroy(pools->large); + dma_pool_destroy(pools->small); + } } -static int nvme_admin_init_hctx(struct blk_mq_hw_ctx *hctx, void *data, - unsigned int hctx_idx) +static int nvme_init_hctx_common(struct blk_mq_hw_ctx *hctx, void *data, + unsigned qid) { struct nvme_dev *dev = to_nvme_dev(data); - struct nvme_queue *nvmeq = &dev->queues[0]; + struct nvme_queue *nvmeq = &dev->queues[qid]; + struct nvme_descriptor_pools *pools; + struct blk_mq_tags *tags; - WARN_ON(hctx_idx != 0); - WARN_ON(dev->admin_tagset.tags[0] != hctx->tags); + tags = qid ? dev->tagset.tags[qid - 1] : dev->admin_tagset.tags[0]; + WARN_ON(tags != hctx->tags); + pools = nvme_setup_descriptor_pools(dev, hctx->numa_node); + if (IS_ERR(pools)) + return PTR_ERR(pools); + nvmeq->descriptor_pools = *pools; hctx->driver_data = nvmeq; return 0; } -static int nvme_init_hctx(struct blk_mq_hw_ctx *hctx, void *data, - unsigned int hctx_idx) +static int nvme_admin_init_hctx(struct blk_mq_hw_ctx *hctx, void *data, + unsigned int hctx_idx) { - struct nvme_dev *dev = to_nvme_dev(data); - struct nvme_queue *nvmeq = &dev->queues[hctx_idx + 1]; + WARN_ON(hctx_idx != 0); + return nvme_init_hctx_common(hctx, data, 0); +} - WARN_ON(dev->tagset.tags[hctx_idx] != hctx->tags); - hctx->driver_data = nvmeq; - return 0; +static int nvme_init_hctx(struct blk_mq_hw_ctx *hctx, void *data, + unsigned int hctx_idx) +{ + return nvme_init_hctx_common(hctx, data, hctx_idx + 1); } static int nvme_pci_init_request(struct blk_mq_tag_set *set, struct request *req, unsigned int hctx_idx, unsigned int numa_node) { - struct nvme_dev *dev = to_nvme_dev(set->driver_data); struct nvme_iod *iod = blk_mq_rq_to_pdu(req); - nvme_req(req)->ctrl = &dev->ctrl; + nvme_req(req)->ctrl = set->driver_data; nvme_req(req)->cmd = &iod->cmd; return 0; } @@ -462,7 +555,7 @@ static void nvme_pci_map_queues(struct blk_mq_tag_set *set) */ map->queue_offset = qoff; if (i != HCTX_TYPE_POLL && offset) - blk_mq_pci_map_queues(map, to_pci_dev(dev->dev), offset); + blk_mq_map_hw_queues(map, dev->dev, offset); else blk_mq_map_queues(map); qoff += map->nr_queues; @@ -509,195 +602,368 @@ static void nvme_commit_rqs(struct blk_mq_hw_ctx *hctx) spin_unlock(&nvmeq->sq_lock); } -static void **nvme_pci_iod_list(struct request *req) +enum nvme_use_sgl { + SGL_UNSUPPORTED, + SGL_SUPPORTED, + SGL_FORCED, +}; + +static inline bool nvme_pci_metadata_use_sgls(struct request *req) { - struct nvme_iod *iod = blk_mq_rq_to_pdu(req); - return (void **)(iod->sgt.sgl + blk_rq_nr_phys_segments(req)); + struct nvme_queue *nvmeq = req->mq_hctx->driver_data; + struct nvme_dev *dev = nvmeq->dev; + + if (!nvme_ctrl_meta_sgl_supported(&dev->ctrl)) + return false; + return req->nr_integrity_segments > 1 || + nvme_req(req)->flags & NVME_REQ_USERCMD; } -static inline bool nvme_pci_use_sgls(struct nvme_dev *dev, struct request *req) +static inline enum nvme_use_sgl nvme_pci_use_sgls(struct nvme_dev *dev, + struct request *req) { struct nvme_queue *nvmeq = req->mq_hctx->driver_data; - int nseg = blk_rq_nr_phys_segments(req); - unsigned int avg_seg_size; - avg_seg_size = DIV_ROUND_UP(blk_rq_payload_bytes(req), nseg); + if (nvmeq->qid && nvme_ctrl_sgl_supported(&dev->ctrl)) { + /* + * When the controller is capable of using SGL, there are + * several conditions that we force to use it: + * + * 1. A request containing page gaps within the controller's + * mask can not use the PRP format. + * + * 2. User commands use SGL because that lets the device + * validate the requested transfer lengths. + * + * 3. Multiple integrity segments must use SGL as that's the + * only way to describe such a command in NVMe. + */ + if (req_phys_gap_mask(req) & (NVME_CTRL_PAGE_SIZE - 1) || + nvme_req(req)->flags & NVME_REQ_USERCMD || + req->nr_integrity_segments > 1) + return SGL_FORCED; + return SGL_SUPPORTED; + } - if (!nvme_ctrl_sgl_supported(&dev->ctrl)) - return false; - if (!nvmeq->qid) - return false; - if (!sgl_threshold || avg_seg_size < sgl_threshold) - return false; - return true; + return SGL_UNSUPPORTED; +} + +static unsigned int nvme_pci_avg_seg_size(struct request *req) +{ + struct nvme_iod *iod = blk_mq_rq_to_pdu(req); + unsigned int nseg; + + if (blk_rq_dma_map_coalesce(&iod->dma_state)) + nseg = 1; + else + nseg = blk_rq_nr_phys_segments(req); + return DIV_ROUND_UP(blk_rq_payload_bytes(req), nseg); } -static void nvme_free_prps(struct nvme_dev *dev, struct request *req) +static inline struct dma_pool *nvme_dma_pool(struct nvme_queue *nvmeq, + struct nvme_iod *iod) { + if (iod->flags & IOD_SMALL_DESCRIPTOR) + return nvmeq->descriptor_pools.small; + return nvmeq->descriptor_pools.large; +} + +static inline bool nvme_pci_cmd_use_meta_sgl(struct nvme_command *cmd) +{ + return (cmd->common.flags & NVME_CMD_SGL_ALL) == NVME_CMD_SGL_METASEG; +} + +static inline bool nvme_pci_cmd_use_sgl(struct nvme_command *cmd) +{ + return cmd->common.flags & + (NVME_CMD_SGL_METABUF | NVME_CMD_SGL_METASEG); +} + +static inline dma_addr_t nvme_pci_first_desc_dma_addr(struct nvme_command *cmd) +{ + if (nvme_pci_cmd_use_sgl(cmd)) + return le64_to_cpu(cmd->common.dptr.sgl.addr); + return le64_to_cpu(cmd->common.dptr.prp2); +} + +static void nvme_free_descriptors(struct request *req) +{ + struct nvme_queue *nvmeq = req->mq_hctx->driver_data; const int last_prp = NVME_CTRL_PAGE_SIZE / sizeof(__le64) - 1; struct nvme_iod *iod = blk_mq_rq_to_pdu(req); - dma_addr_t dma_addr = iod->first_dma; + dma_addr_t dma_addr = nvme_pci_first_desc_dma_addr(&iod->cmd); int i; - for (i = 0; i < iod->nr_allocations; i++) { - __le64 *prp_list = nvme_pci_iod_list(req)[i]; + if (iod->nr_descriptors == 1) { + dma_pool_free(nvme_dma_pool(nvmeq, iod), iod->descriptors[0], + dma_addr); + return; + } + + for (i = 0; i < iod->nr_descriptors; i++) { + __le64 *prp_list = iod->descriptors[i]; dma_addr_t next_dma_addr = le64_to_cpu(prp_list[last_prp]); - dma_pool_free(dev->prp_page_pool, prp_list, dma_addr); + dma_pool_free(nvmeq->descriptor_pools.large, prp_list, + dma_addr); dma_addr = next_dma_addr; } } -static void nvme_free_sgls(struct nvme_dev *dev, struct request *req) +static void nvme_free_prps(struct request *req, unsigned int attrs) { - const int last_sg = SGES_PER_PAGE - 1; struct nvme_iod *iod = blk_mq_rq_to_pdu(req); - dma_addr_t dma_addr = iod->first_dma; - int i; + struct nvme_queue *nvmeq = req->mq_hctx->driver_data; + unsigned int i; - for (i = 0; i < iod->nr_allocations; i++) { - struct nvme_sgl_desc *sg_list = nvme_pci_iod_list(req)[i]; - dma_addr_t next_dma_addr = le64_to_cpu((sg_list[last_sg]).addr); + for (i = 0; i < iod->nr_dma_vecs; i++) + dma_unmap_phys(nvmeq->dev->dev, iod->dma_vecs[i].addr, + iod->dma_vecs[i].len, rq_dma_dir(req), attrs); + mempool_free(iod->dma_vecs, nvmeq->dev->dmavec_mempool); +} - dma_pool_free(dev->prp_page_pool, sg_list, dma_addr); - dma_addr = next_dma_addr; +static void nvme_free_sgls(struct request *req, struct nvme_sgl_desc *sge, + struct nvme_sgl_desc *sg_list, unsigned int attrs) +{ + struct nvme_queue *nvmeq = req->mq_hctx->driver_data; + enum dma_data_direction dir = rq_dma_dir(req); + unsigned int len = le32_to_cpu(sge->length); + struct device *dma_dev = nvmeq->dev->dev; + unsigned int i; + + if (sge->type == (NVME_SGL_FMT_DATA_DESC << 4)) { + dma_unmap_phys(dma_dev, le64_to_cpu(sge->addr), len, dir, + attrs); + return; } + + for (i = 0; i < len / sizeof(*sg_list); i++) + dma_unmap_phys(dma_dev, le64_to_cpu(sg_list[i].addr), + le32_to_cpu(sg_list[i].length), dir, attrs); } -static void nvme_unmap_data(struct nvme_dev *dev, struct request *req) +static void nvme_unmap_metadata(struct request *req) { + struct nvme_queue *nvmeq = req->mq_hctx->driver_data; + enum pci_p2pdma_map_type map = PCI_P2PDMA_MAP_NONE; + enum dma_data_direction dir = rq_dma_dir(req); struct nvme_iod *iod = blk_mq_rq_to_pdu(req); + struct device *dma_dev = nvmeq->dev->dev; + struct nvme_sgl_desc *sge = iod->meta_descriptor; + unsigned int attrs = 0; - if (iod->dma_len) { - dma_unmap_page(dev->dev, iod->first_dma, iod->dma_len, + if (iod->flags & IOD_SINGLE_META_SEGMENT) { + dma_unmap_page(dma_dev, iod->meta_dma, + rq_integrity_vec(req).bv_len, rq_dma_dir(req)); return; } - WARN_ON_ONCE(!iod->sgt.nents); + if (iod->flags & IOD_META_P2P) + map = PCI_P2PDMA_MAP_BUS_ADDR; + else if (iod->flags & IOD_META_MMIO) { + map = PCI_P2PDMA_MAP_THRU_HOST_BRIDGE; + attrs |= DMA_ATTR_MMIO; + } + + if (!blk_rq_dma_unmap(req, dma_dev, &iod->meta_dma_state, + iod->meta_total_len, map)) { + if (nvme_pci_cmd_use_meta_sgl(&iod->cmd)) + nvme_free_sgls(req, sge, &sge[1], attrs); + else + dma_unmap_phys(dma_dev, iod->meta_dma, + iod->meta_total_len, dir, attrs); + } + + if (iod->meta_descriptor) + dma_pool_free(nvmeq->descriptor_pools.small, + iod->meta_descriptor, iod->meta_dma); +} + +static void nvme_unmap_data(struct request *req) +{ + enum pci_p2pdma_map_type map = PCI_P2PDMA_MAP_NONE; + struct nvme_iod *iod = blk_mq_rq_to_pdu(req); + struct nvme_queue *nvmeq = req->mq_hctx->driver_data; + struct device *dma_dev = nvmeq->dev->dev; + unsigned int attrs = 0; + + if (iod->flags & IOD_SINGLE_SEGMENT) { + static_assert(offsetof(union nvme_data_ptr, prp1) == + offsetof(union nvme_data_ptr, sgl.addr)); + dma_unmap_page(dma_dev, le64_to_cpu(iod->cmd.common.dptr.prp1), + iod->total_len, rq_dma_dir(req)); + return; + } + + if (iod->flags & IOD_DATA_P2P) + map = PCI_P2PDMA_MAP_BUS_ADDR; + else if (iod->flags & IOD_DATA_MMIO) { + map = PCI_P2PDMA_MAP_THRU_HOST_BRIDGE; + attrs |= DMA_ATTR_MMIO; + } - dma_unmap_sgtable(dev->dev, &iod->sgt, rq_dma_dir(req), 0); + if (!blk_rq_dma_unmap(req, dma_dev, &iod->dma_state, iod->total_len, + map)) { + if (nvme_pci_cmd_use_sgl(&iod->cmd)) + nvme_free_sgls(req, iod->descriptors[0], + &iod->cmd.common.dptr.sgl, attrs); + else + nvme_free_prps(req, attrs); + } - if (iod->nr_allocations == 0) - dma_pool_free(dev->prp_small_pool, nvme_pci_iod_list(req)[0], - iod->first_dma); - else if (iod->use_sgl) - nvme_free_sgls(dev, req); - else - nvme_free_prps(dev, req); - mempool_free(iod->sgt.sgl, dev->iod_mempool); + if (iod->nr_descriptors) + nvme_free_descriptors(req); } -static void nvme_print_sgl(struct scatterlist *sgl, int nents) +static bool nvme_pci_prp_iter_next(struct request *req, struct device *dma_dev, + struct blk_dma_iter *iter) { - int i; - struct scatterlist *sg; + struct nvme_iod *iod = blk_mq_rq_to_pdu(req); - for_each_sg(sgl, sg, nents, i) { - dma_addr_t phys = sg_phys(sg); - pr_warn("sg[%d] phys_addr:%pad offset:%d length:%d " - "dma_address:%pad dma_length:%d\n", - i, &phys, sg->offset, sg->length, &sg_dma_address(sg), - sg_dma_len(sg)); + if (iter->len) + return true; + if (!blk_rq_dma_map_iter_next(req, dma_dev, &iod->dma_state, iter)) + return false; + if (!dma_use_iova(&iod->dma_state) && dma_need_unmap(dma_dev)) { + iod->dma_vecs[iod->nr_dma_vecs].addr = iter->addr; + iod->dma_vecs[iod->nr_dma_vecs].len = iter->len; + iod->nr_dma_vecs++; } + return true; } -static blk_status_t nvme_pci_setup_prps(struct nvme_dev *dev, - struct request *req, struct nvme_rw_command *cmnd) +static blk_status_t nvme_pci_setup_data_prp(struct request *req, + struct blk_dma_iter *iter) { struct nvme_iod *iod = blk_mq_rq_to_pdu(req); - struct dma_pool *pool; - int length = blk_rq_payload_bytes(req); - struct scatterlist *sg = iod->sgt.sgl; - int dma_len = sg_dma_len(sg); - u64 dma_addr = sg_dma_address(sg); - int offset = dma_addr & (NVME_CTRL_PAGE_SIZE - 1); + struct nvme_queue *nvmeq = req->mq_hctx->driver_data; + unsigned int length = blk_rq_payload_bytes(req); + dma_addr_t prp1_dma, prp2_dma = 0; + unsigned int prp_len, i; __le64 *prp_list; - void **list = nvme_pci_iod_list(req); - dma_addr_t prp_dma; - int nprps, i; - length -= (NVME_CTRL_PAGE_SIZE - offset); - if (length <= 0) { - iod->first_dma = 0; - goto done; + if (!dma_use_iova(&iod->dma_state) && dma_need_unmap(nvmeq->dev->dev)) { + iod->dma_vecs = mempool_alloc(nvmeq->dev->dmavec_mempool, + GFP_ATOMIC); + if (!iod->dma_vecs) + return BLK_STS_RESOURCE; + iod->dma_vecs[0].addr = iter->addr; + iod->dma_vecs[0].len = iter->len; + iod->nr_dma_vecs = 1; } - dma_len -= (NVME_CTRL_PAGE_SIZE - offset); - if (dma_len) { - dma_addr += (NVME_CTRL_PAGE_SIZE - offset); - } else { - sg = sg_next(sg); - dma_addr = sg_dma_address(sg); - dma_len = sg_dma_len(sg); + /* + * PRP1 always points to the start of the DMA transfers. + * + * This is the only PRP (except for the list entries) that could be + * non-aligned. + */ + prp1_dma = iter->addr; + prp_len = min(length, NVME_CTRL_PAGE_SIZE - + (iter->addr & (NVME_CTRL_PAGE_SIZE - 1))); + iod->total_len += prp_len; + iter->addr += prp_len; + iter->len -= prp_len; + length -= prp_len; + if (!length) + goto done; + + if (!nvme_pci_prp_iter_next(req, nvmeq->dev->dev, iter)) { + if (WARN_ON_ONCE(!iter->status)) + goto bad_sgl; + goto done; } + /* + * PRP2 is usually a list, but can point to data if all data to be + * transferred fits into PRP1 + PRP2: + */ if (length <= NVME_CTRL_PAGE_SIZE) { - iod->first_dma = dma_addr; + prp2_dma = iter->addr; + iod->total_len += length; goto done; } - nprps = DIV_ROUND_UP(length, NVME_CTRL_PAGE_SIZE); - if (nprps <= (256 / 8)) { - pool = dev->prp_small_pool; - iod->nr_allocations = 0; - } else { - pool = dev->prp_page_pool; - iod->nr_allocations = 1; - } + if (DIV_ROUND_UP(length, NVME_CTRL_PAGE_SIZE) <= + NVME_SMALL_POOL_SIZE / sizeof(__le64)) + iod->flags |= IOD_SMALL_DESCRIPTOR; - prp_list = dma_pool_alloc(pool, GFP_ATOMIC, &prp_dma); + prp_list = dma_pool_alloc(nvme_dma_pool(nvmeq, iod), GFP_ATOMIC, + &prp2_dma); if (!prp_list) { - iod->nr_allocations = -1; - return BLK_STS_RESOURCE; + iter->status = BLK_STS_RESOURCE; + goto done; } - list[0] = prp_list; - iod->first_dma = prp_dma; + iod->descriptors[iod->nr_descriptors++] = prp_list; + i = 0; for (;;) { + prp_list[i++] = cpu_to_le64(iter->addr); + prp_len = min(length, NVME_CTRL_PAGE_SIZE); + if (WARN_ON_ONCE(iter->len < prp_len)) + goto bad_sgl; + + iod->total_len += prp_len; + iter->addr += prp_len; + iter->len -= prp_len; + length -= prp_len; + if (!length) + break; + + if (!nvme_pci_prp_iter_next(req, nvmeq->dev->dev, iter)) { + if (WARN_ON_ONCE(!iter->status)) + goto bad_sgl; + goto done; + } + + /* + * If we've filled the entire descriptor, allocate a new that is + * pointed to be the last entry in the previous PRP list. To + * accommodate for that move the last actual entry to the new + * descriptor. + */ if (i == NVME_CTRL_PAGE_SIZE >> 3) { __le64 *old_prp_list = prp_list; - prp_list = dma_pool_alloc(pool, GFP_ATOMIC, &prp_dma); - if (!prp_list) - goto free_prps; - list[iod->nr_allocations++] = prp_list; + dma_addr_t prp_list_dma; + + prp_list = dma_pool_alloc(nvmeq->descriptor_pools.large, + GFP_ATOMIC, &prp_list_dma); + if (!prp_list) { + iter->status = BLK_STS_RESOURCE; + goto done; + } + iod->descriptors[iod->nr_descriptors++] = prp_list; + prp_list[0] = old_prp_list[i - 1]; - old_prp_list[i - 1] = cpu_to_le64(prp_dma); + old_prp_list[i - 1] = cpu_to_le64(prp_list_dma); i = 1; } - prp_list[i++] = cpu_to_le64(dma_addr); - dma_len -= NVME_CTRL_PAGE_SIZE; - dma_addr += NVME_CTRL_PAGE_SIZE; - length -= NVME_CTRL_PAGE_SIZE; - if (length <= 0) - break; - if (dma_len > 0) - continue; - if (unlikely(dma_len < 0)) - goto bad_sgl; - sg = sg_next(sg); - dma_addr = sg_dma_address(sg); - dma_len = sg_dma_len(sg); } + done: - cmnd->dptr.prp1 = cpu_to_le64(sg_dma_address(iod->sgt.sgl)); - cmnd->dptr.prp2 = cpu_to_le64(iod->first_dma); - return BLK_STS_OK; -free_prps: - nvme_free_prps(dev, req); - return BLK_STS_RESOURCE; + /* + * nvme_unmap_data uses the DPT field in the SQE to tear down the + * mapping, so initialize it even for failures. + */ + iod->cmd.common.dptr.prp1 = cpu_to_le64(prp1_dma); + iod->cmd.common.dptr.prp2 = cpu_to_le64(prp2_dma); + if (unlikely(iter->status)) + nvme_unmap_data(req); + return iter->status; + bad_sgl: - WARN(DO_ONCE(nvme_print_sgl, iod->sgt.sgl, iod->sgt.nents), - "Invalid SGL for payload:%d nents:%d\n", - blk_rq_payload_bytes(req), iod->sgt.nents); + dev_err_once(nvmeq->dev->dev, + "Incorrectly formed request for payload:%d nents:%d\n", + blk_rq_payload_bytes(req), blk_rq_nr_phys_segments(req)); return BLK_STS_IOERR; } static void nvme_pci_sgl_set_data(struct nvme_sgl_desc *sge, - struct scatterlist *sg) + struct blk_dma_iter *iter) { - sge->addr = cpu_to_le64(sg_dma_address(sg)); - sge->length = cpu_to_le32(sg_dma_len(sg)); + sge->addr = cpu_to_le64(iter->addr); + sge->length = cpu_to_le32(iter->len); sge->type = NVME_SGL_FMT_DATA_DESC << 4; } @@ -705,207 +971,267 @@ static void nvme_pci_sgl_set_seg(struct nvme_sgl_desc *sge, dma_addr_t dma_addr, int entries) { sge->addr = cpu_to_le64(dma_addr); - if (entries < SGES_PER_PAGE) { - sge->length = cpu_to_le32(entries * sizeof(*sge)); - sge->type = NVME_SGL_FMT_LAST_SEG_DESC << 4; - } else { - sge->length = cpu_to_le32(NVME_CTRL_PAGE_SIZE); - sge->type = NVME_SGL_FMT_SEG_DESC << 4; - } + sge->length = cpu_to_le32(entries * sizeof(*sge)); + sge->type = NVME_SGL_FMT_LAST_SEG_DESC << 4; } -static blk_status_t nvme_pci_setup_sgls(struct nvme_dev *dev, - struct request *req, struct nvme_rw_command *cmd) +static blk_status_t nvme_pci_setup_data_sgl(struct request *req, + struct blk_dma_iter *iter) { struct nvme_iod *iod = blk_mq_rq_to_pdu(req); - struct dma_pool *pool; + struct nvme_queue *nvmeq = req->mq_hctx->driver_data; + unsigned int entries = blk_rq_nr_phys_segments(req); struct nvme_sgl_desc *sg_list; - struct scatterlist *sg = iod->sgt.sgl; - unsigned int entries = iod->sgt.nents; dma_addr_t sgl_dma; - int i = 0; + unsigned int mapped = 0; - /* setting the transfer type as SGL */ - cmd->flags = NVME_CMD_SGL_METABUF; + /* set the transfer type as SGL */ + iod->cmd.common.flags = NVME_CMD_SGL_METABUF; - if (entries == 1) { - nvme_pci_sgl_set_data(&cmd->dptr.sgl, sg); + if (entries == 1 || blk_rq_dma_map_coalesce(&iod->dma_state)) { + nvme_pci_sgl_set_data(&iod->cmd.common.dptr.sgl, iter); + iod->total_len += iter->len; return BLK_STS_OK; } - if (entries <= (256 / sizeof(struct nvme_sgl_desc))) { - pool = dev->prp_small_pool; - iod->nr_allocations = 0; - } else { - pool = dev->prp_page_pool; - iod->nr_allocations = 1; - } + if (entries <= NVME_SMALL_POOL_SIZE / sizeof(*sg_list)) + iod->flags |= IOD_SMALL_DESCRIPTOR; - sg_list = dma_pool_alloc(pool, GFP_ATOMIC, &sgl_dma); - if (!sg_list) { - iod->nr_allocations = -1; + sg_list = dma_pool_alloc(nvme_dma_pool(nvmeq, iod), GFP_ATOMIC, + &sgl_dma); + if (!sg_list) return BLK_STS_RESOURCE; - } - - nvme_pci_iod_list(req)[0] = sg_list; - iod->first_dma = sgl_dma; - - nvme_pci_sgl_set_seg(&cmd->dptr.sgl, sgl_dma, entries); + iod->descriptors[iod->nr_descriptors++] = sg_list; do { - if (i == SGES_PER_PAGE) { - struct nvme_sgl_desc *old_sg_desc = sg_list; - struct nvme_sgl_desc *link = &old_sg_desc[i - 1]; - - sg_list = dma_pool_alloc(pool, GFP_ATOMIC, &sgl_dma); - if (!sg_list) - goto free_sgls; - - i = 0; - nvme_pci_iod_list(req)[iod->nr_allocations++] = sg_list; - sg_list[i++] = *link; - nvme_pci_sgl_set_seg(link, sgl_dma, entries); + if (WARN_ON_ONCE(mapped == entries)) { + iter->status = BLK_STS_IOERR; + break; } + nvme_pci_sgl_set_data(&sg_list[mapped++], iter); + iod->total_len += iter->len; + } while (blk_rq_dma_map_iter_next(req, nvmeq->dev->dev, &iod->dma_state, + iter)); - nvme_pci_sgl_set_data(&sg_list[i++], sg); - sg = sg_next(sg); - } while (--entries > 0); - - return BLK_STS_OK; -free_sgls: - nvme_free_sgls(dev, req); - return BLK_STS_RESOURCE; + nvme_pci_sgl_set_seg(&iod->cmd.common.dptr.sgl, sgl_dma, mapped); + if (unlikely(iter->status)) + nvme_unmap_data(req); + return iter->status; } -static blk_status_t nvme_setup_prp_simple(struct nvme_dev *dev, - struct request *req, struct nvme_rw_command *cmnd, - struct bio_vec *bv) +static blk_status_t nvme_pci_setup_data_simple(struct request *req, + enum nvme_use_sgl use_sgl) { struct nvme_iod *iod = blk_mq_rq_to_pdu(req); - unsigned int offset = bv->bv_offset & (NVME_CTRL_PAGE_SIZE - 1); - unsigned int first_prp_len = NVME_CTRL_PAGE_SIZE - offset; - - iod->first_dma = dma_map_bvec(dev->dev, bv, rq_dma_dir(req), 0); - if (dma_mapping_error(dev->dev, iod->first_dma)) + struct nvme_queue *nvmeq = req->mq_hctx->driver_data; + struct bio_vec bv = req_bvec(req); + unsigned int prp1_offset = bv.bv_offset & (NVME_CTRL_PAGE_SIZE - 1); + bool prp_possible = prp1_offset + bv.bv_len <= NVME_CTRL_PAGE_SIZE * 2; + dma_addr_t dma_addr; + + if (!use_sgl && !prp_possible) + return BLK_STS_AGAIN; + if (is_pci_p2pdma_page(bv.bv_page)) + return BLK_STS_AGAIN; + + dma_addr = dma_map_bvec(nvmeq->dev->dev, &bv, rq_dma_dir(req), 0); + if (dma_mapping_error(nvmeq->dev->dev, dma_addr)) return BLK_STS_RESOURCE; - iod->dma_len = bv->bv_len; + iod->total_len = bv.bv_len; + iod->flags |= IOD_SINGLE_SEGMENT; + + if (use_sgl == SGL_FORCED || !prp_possible) { + iod->cmd.common.flags = NVME_CMD_SGL_METABUF; + iod->cmd.common.dptr.sgl.addr = cpu_to_le64(dma_addr); + iod->cmd.common.dptr.sgl.length = cpu_to_le32(bv.bv_len); + iod->cmd.common.dptr.sgl.type = NVME_SGL_FMT_DATA_DESC << 4; + } else { + unsigned int first_prp_len = NVME_CTRL_PAGE_SIZE - prp1_offset; + + iod->cmd.common.dptr.prp1 = cpu_to_le64(dma_addr); + iod->cmd.common.dptr.prp2 = 0; + if (bv.bv_len > first_prp_len) + iod->cmd.common.dptr.prp2 = + cpu_to_le64(dma_addr + first_prp_len); + } - cmnd->dptr.prp1 = cpu_to_le64(iod->first_dma); - if (bv->bv_len > first_prp_len) - cmnd->dptr.prp2 = cpu_to_le64(iod->first_dma + first_prp_len); - else - cmnd->dptr.prp2 = 0; return BLK_STS_OK; } -static blk_status_t nvme_setup_sgl_simple(struct nvme_dev *dev, - struct request *req, struct nvme_rw_command *cmnd, - struct bio_vec *bv) +static blk_status_t nvme_map_data(struct request *req) { struct nvme_iod *iod = blk_mq_rq_to_pdu(req); + struct nvme_queue *nvmeq = req->mq_hctx->driver_data; + struct nvme_dev *dev = nvmeq->dev; + enum nvme_use_sgl use_sgl = nvme_pci_use_sgls(dev, req); + struct blk_dma_iter iter; + blk_status_t ret; + + /* + * Try to skip the DMA iterator for single segment requests, as that + * significantly improves performances for small I/O sizes. + */ + if (blk_rq_nr_phys_segments(req) == 1) { + ret = nvme_pci_setup_data_simple(req, use_sgl); + if (ret != BLK_STS_AGAIN) + return ret; + } + + if (!blk_rq_dma_map_iter_start(req, dev->dev, &iod->dma_state, &iter)) + return iter.status; - iod->first_dma = dma_map_bvec(dev->dev, bv, rq_dma_dir(req), 0); - if (dma_mapping_error(dev->dev, iod->first_dma)) + switch (iter.p2pdma.map) { + case PCI_P2PDMA_MAP_BUS_ADDR: + iod->flags |= IOD_DATA_P2P; + break; + case PCI_P2PDMA_MAP_THRU_HOST_BRIDGE: + iod->flags |= IOD_DATA_MMIO; + break; + case PCI_P2PDMA_MAP_NONE: + break; + default: return BLK_STS_RESOURCE; - iod->dma_len = bv->bv_len; + } - cmnd->flags = NVME_CMD_SGL_METABUF; - cmnd->dptr.sgl.addr = cpu_to_le64(iod->first_dma); - cmnd->dptr.sgl.length = cpu_to_le32(iod->dma_len); - cmnd->dptr.sgl.type = NVME_SGL_FMT_DATA_DESC << 4; - return BLK_STS_OK; + if (use_sgl == SGL_FORCED || + (use_sgl == SGL_SUPPORTED && + (sgl_threshold && nvme_pci_avg_seg_size(req) >= sgl_threshold))) + return nvme_pci_setup_data_sgl(req, &iter); + return nvme_pci_setup_data_prp(req, &iter); } -static blk_status_t nvme_map_data(struct nvme_dev *dev, struct request *req, - struct nvme_command *cmnd) +static blk_status_t nvme_pci_setup_meta_iter(struct request *req) { + struct nvme_queue *nvmeq = req->mq_hctx->driver_data; + unsigned int entries = req->nr_integrity_segments; struct nvme_iod *iod = blk_mq_rq_to_pdu(req); - blk_status_t ret = BLK_STS_RESOURCE; - int rc; + struct nvme_dev *dev = nvmeq->dev; + struct nvme_sgl_desc *sg_list; + struct blk_dma_iter iter; + dma_addr_t sgl_dma; + int i = 0; - if (blk_rq_nr_phys_segments(req) == 1) { - struct nvme_queue *nvmeq = req->mq_hctx->driver_data; - struct bio_vec bv = req_bvec(req); - - if (!is_pci_p2pdma_page(bv.bv_page)) { - if (bv.bv_offset + bv.bv_len <= NVME_CTRL_PAGE_SIZE * 2) - return nvme_setup_prp_simple(dev, req, - &cmnd->rw, &bv); - - if (nvmeq->qid && sgl_threshold && - nvme_ctrl_sgl_supported(&dev->ctrl)) - return nvme_setup_sgl_simple(dev, req, - &cmnd->rw, &bv); - } + if (!blk_rq_integrity_dma_map_iter_start(req, dev->dev, + &iod->meta_dma_state, &iter)) + return iter.status; + + switch (iter.p2pdma.map) { + case PCI_P2PDMA_MAP_BUS_ADDR: + iod->flags |= IOD_META_P2P; + break; + case PCI_P2PDMA_MAP_THRU_HOST_BRIDGE: + iod->flags |= IOD_META_MMIO; + break; + case PCI_P2PDMA_MAP_NONE: + break; + default: + return BLK_STS_RESOURCE; + } + + if (blk_rq_dma_map_coalesce(&iod->meta_dma_state)) + entries = 1; + + /* + * The NVMe MPTR descriptor has an implicit length that the host and + * device must agree on to avoid data/memory corruption. We trust the + * kernel allocated correctly based on the format's parameters, so use + * the more efficient MPTR to avoid extra dma pool allocations for the + * SGL indirection. + * + * But for user commands, we don't necessarily know what they do, so + * the driver can't validate the metadata buffer size. The SGL + * descriptor provides an explicit length, so we're relying on that + * mechanism to catch any misunderstandings between the application and + * device. + * + * P2P DMA also needs to use the blk_dma_iter method, so mptr setup + * leverages this routine when that happens. + */ + if (!nvme_ctrl_meta_sgl_supported(&dev->ctrl) || + (entries == 1 && !(nvme_req(req)->flags & NVME_REQ_USERCMD))) { + iod->cmd.common.metadata = cpu_to_le64(iter.addr); + iod->meta_total_len = iter.len; + iod->meta_dma = iter.addr; + iod->meta_descriptor = NULL; + return BLK_STS_OK; } - iod->dma_len = 0; - iod->sgt.sgl = mempool_alloc(dev->iod_mempool, GFP_ATOMIC); - if (!iod->sgt.sgl) + sg_list = dma_pool_alloc(nvmeq->descriptor_pools.small, GFP_ATOMIC, + &sgl_dma); + if (!sg_list) return BLK_STS_RESOURCE; - sg_init_table(iod->sgt.sgl, blk_rq_nr_phys_segments(req)); - iod->sgt.orig_nents = blk_rq_map_sg(req->q, req, iod->sgt.sgl); - if (!iod->sgt.orig_nents) - goto out_free_sg; - - rc = dma_map_sgtable(dev->dev, &iod->sgt, rq_dma_dir(req), - DMA_ATTR_NO_WARN); - if (rc) { - if (rc == -EREMOTEIO) - ret = BLK_STS_TARGET; - goto out_free_sg; - } - - iod->use_sgl = nvme_pci_use_sgls(dev, req); - if (iod->use_sgl) - ret = nvme_pci_setup_sgls(dev, req, &cmnd->rw); - else - ret = nvme_pci_setup_prps(dev, req, &cmnd->rw); - if (ret != BLK_STS_OK) - goto out_unmap_sg; - return BLK_STS_OK; -out_unmap_sg: - dma_unmap_sgtable(dev->dev, &iod->sgt, rq_dma_dir(req), 0); -out_free_sg: - mempool_free(iod->sgt.sgl, dev->iod_mempool); - return ret; + iod->meta_descriptor = sg_list; + iod->meta_dma = sgl_dma; + iod->cmd.common.flags = NVME_CMD_SGL_METASEG; + iod->cmd.common.metadata = cpu_to_le64(sgl_dma); + if (entries == 1) { + iod->meta_total_len = iter.len; + nvme_pci_sgl_set_data(sg_list, &iter); + return BLK_STS_OK; + } + + sgl_dma += sizeof(*sg_list); + do { + nvme_pci_sgl_set_data(&sg_list[++i], &iter); + iod->meta_total_len += iter.len; + } while (blk_rq_integrity_dma_map_iter_next(req, dev->dev, &iter)); + + nvme_pci_sgl_set_seg(sg_list, sgl_dma, i); + if (unlikely(iter.status)) + nvme_unmap_metadata(req); + return iter.status; } -static blk_status_t nvme_map_metadata(struct nvme_dev *dev, struct request *req, - struct nvme_command *cmnd) +static blk_status_t nvme_pci_setup_meta_mptr(struct request *req) { struct nvme_iod *iod = blk_mq_rq_to_pdu(req); + struct nvme_queue *nvmeq = req->mq_hctx->driver_data; + struct bio_vec bv = rq_integrity_vec(req); + + if (is_pci_p2pdma_page(bv.bv_page)) + return nvme_pci_setup_meta_iter(req); - iod->meta_dma = dma_map_bvec(dev->dev, rq_integrity_vec(req), - rq_dma_dir(req), 0); - if (dma_mapping_error(dev->dev, iod->meta_dma)) + iod->meta_dma = dma_map_bvec(nvmeq->dev->dev, &bv, rq_dma_dir(req), 0); + if (dma_mapping_error(nvmeq->dev->dev, iod->meta_dma)) return BLK_STS_IOERR; - cmnd->rw.metadata = cpu_to_le64(iod->meta_dma); + iod->cmd.common.metadata = cpu_to_le64(iod->meta_dma); + iod->flags |= IOD_SINGLE_META_SEGMENT; return BLK_STS_OK; } -static blk_status_t nvme_prep_rq(struct nvme_dev *dev, struct request *req) +static blk_status_t nvme_map_metadata(struct request *req) +{ + struct nvme_iod *iod = blk_mq_rq_to_pdu(req); + + if ((iod->cmd.common.flags & NVME_CMD_SGL_METABUF) && + nvme_pci_metadata_use_sgls(req)) + return nvme_pci_setup_meta_iter(req); + return nvme_pci_setup_meta_mptr(req); +} + +static blk_status_t nvme_prep_rq(struct request *req) { struct nvme_iod *iod = blk_mq_rq_to_pdu(req); blk_status_t ret; - iod->aborted = false; - iod->nr_allocations = -1; - iod->sgt.nents = 0; + iod->flags = 0; + iod->nr_descriptors = 0; + iod->total_len = 0; + iod->meta_total_len = 0; ret = nvme_setup_cmd(req->q->queuedata, req); if (ret) return ret; if (blk_rq_nr_phys_segments(req)) { - ret = nvme_map_data(dev, req, &iod->cmd); + ret = nvme_map_data(req); if (ret) goto out_free_cmd; } if (blk_integrity_rq(req)) { - ret = nvme_map_metadata(dev, req, &iod->cmd); + ret = nvme_map_metadata(req); if (ret) goto out_unmap_data; } @@ -913,15 +1239,13 @@ static blk_status_t nvme_prep_rq(struct nvme_dev *dev, struct request *req) nvme_start_request(req); return BLK_STS_OK; out_unmap_data: - nvme_unmap_data(dev, req); + if (blk_rq_nr_phys_segments(req)) + nvme_unmap_data(req); out_free_cmd: nvme_cleanup_cmd(req); return ret; } -/* - * NOTE: ns is NULL when called on the admin queue. - */ static blk_status_t nvme_queue_rq(struct blk_mq_hw_ctx *hctx, const struct blk_mq_queue_data *bd) { @@ -941,7 +1265,7 @@ static blk_status_t nvme_queue_rq(struct blk_mq_hw_ctx *hctx, if (unlikely(!nvme_check_ready(&dev->ctrl, req, true))) return nvme_fail_nonready_command(&dev->ctrl, req); - ret = nvme_prep_rq(dev, req); + ret = nvme_prep_rq(req); if (unlikely(ret)) return ret; spin_lock(&nvmeq->sq_lock); @@ -951,11 +1275,15 @@ static blk_status_t nvme_queue_rq(struct blk_mq_hw_ctx *hctx, return BLK_STS_OK; } -static void nvme_submit_cmds(struct nvme_queue *nvmeq, struct request **rqlist) +static void nvme_submit_cmds(struct nvme_queue *nvmeq, struct rq_list *rqlist) { + struct request *req; + + if (rq_list_empty(rqlist)) + return; + spin_lock(&nvmeq->sq_lock); - while (!rq_list_empty(*rqlist)) { - struct request *req = rq_list_pop(rqlist); + while ((req = rq_list_pop(rqlist))) { struct nvme_iod *iod = blk_mq_rq_to_pdu(req); nvme_sq_copy_cmd(nvmeq, &iod->cmd); @@ -975,54 +1303,38 @@ static bool nvme_prep_rq_batch(struct nvme_queue *nvmeq, struct request *req) if (unlikely(!nvme_check_ready(&nvmeq->dev->ctrl, req, true))) return false; - req->mq_hctx->tags->rqs[req->tag] = req; - return nvme_prep_rq(nvmeq->dev, req) == BLK_STS_OK; + return nvme_prep_rq(req) == BLK_STS_OK; } -static void nvme_queue_rqs(struct request **rqlist) +static void nvme_queue_rqs(struct rq_list *rqlist) { - struct request *req, *next, *prev = NULL; - struct request *requeue_list = NULL; - - rq_list_for_each_safe(rqlist, req, next) { - struct nvme_queue *nvmeq = req->mq_hctx->driver_data; - - if (!nvme_prep_rq_batch(nvmeq, req)) { - /* detach 'req' and add to remainder list */ - rq_list_move(rqlist, &requeue_list, req, prev); + struct rq_list submit_list = { }; + struct rq_list requeue_list = { }; + struct nvme_queue *nvmeq = NULL; + struct request *req; - req = prev; - if (!req) - continue; - } + while ((req = rq_list_pop(rqlist))) { + if (nvmeq && nvmeq != req->mq_hctx->driver_data) + nvme_submit_cmds(nvmeq, &submit_list); + nvmeq = req->mq_hctx->driver_data; - if (!next || req->mq_hctx != next->mq_hctx) { - /* detach rest of list, and submit */ - req->rq_next = NULL; - nvme_submit_cmds(nvmeq, rqlist); - *rqlist = next; - prev = NULL; - } else - prev = req; + if (nvme_prep_rq_batch(nvmeq, req)) + rq_list_add_tail(&submit_list, req); + else + rq_list_add_tail(&requeue_list, req); } + if (nvmeq) + nvme_submit_cmds(nvmeq, &submit_list); *rqlist = requeue_list; } static __always_inline void nvme_pci_unmap_rq(struct request *req) { - struct nvme_queue *nvmeq = req->mq_hctx->driver_data; - struct nvme_dev *dev = nvmeq->dev; - - if (blk_integrity_rq(req)) { - struct nvme_iod *iod = blk_mq_rq_to_pdu(req); - - dma_unmap_page(dev->dev, iod->meta_dma, - rq_integrity_vec(req)->bv_len, rq_data_dir(req)); - } - + if (blk_integrity_rq(req)) + nvme_unmap_metadata(req); if (blk_rq_nr_phys_segments(req)) - nvme_unmap_data(dev, req); + nvme_unmap_data(req); } static void nvme_pci_complete_rq(struct request *req) @@ -1089,8 +1401,9 @@ static inline void nvme_handle_cqe(struct nvme_queue *nvmeq, trace_nvme_sq(req, cqe->sq_head, nvmeq->sq_tail); if (!nvme_try_complete_req(req, cqe->status, cqe->result) && - !blk_mq_add_to_batch(req, iob, nvme_req(req)->status, - nvme_pci_complete_batch)) + !blk_mq_add_to_batch(req, iob, + nvme_req(req)->status != NVME_SC_SUCCESS, + nvme_pci_complete_batch)) nvme_pci_complete_rq(req); } @@ -1106,13 +1419,13 @@ static inline void nvme_update_cq_head(struct nvme_queue *nvmeq) } } -static inline int nvme_poll_cq(struct nvme_queue *nvmeq, - struct io_comp_batch *iob) +static inline bool nvme_poll_cq(struct nvme_queue *nvmeq, + struct io_comp_batch *iob) { - int found = 0; + bool found = false; while (nvme_cqe_pending(nvmeq)) { - found++; + found = true; /* * load-load control dependency between phase and the rest of * the cqe requires a full read memory barrier @@ -1133,7 +1446,7 @@ static irqreturn_t nvme_irq(int irq, void *data) DEFINE_IO_COMP_BATCH(iob); if (nvme_poll_cq(nvmeq, &iob)) { - if (!rq_list_empty(iob.req_list)) + if (!rq_list_empty(&iob.req_list)) nvme_pci_complete_batch(&iob); return IRQ_HANDLED; } @@ -1160,7 +1473,9 @@ static void nvme_poll_irqdisable(struct nvme_queue *nvmeq) WARN_ON_ONCE(test_bit(NVMEQ_POLLED, &nvmeq->flags)); disable_irq(pci_irq_vector(pdev, nvmeq->cq_vector)); + spin_lock(&nvmeq->cq_poll_lock); nvme_poll_cq(nvmeq, NULL); + spin_unlock(&nvmeq->cq_poll_lock); enable_irq(pci_irq_vector(pdev, nvmeq->cq_vector)); } @@ -1194,6 +1509,41 @@ static void nvme_pci_submit_async_event(struct nvme_ctrl *ctrl) spin_unlock(&nvmeq->sq_lock); } +static int nvme_pci_subsystem_reset(struct nvme_ctrl *ctrl) +{ + struct nvme_dev *dev = to_nvme_dev(ctrl); + int ret = 0; + + /* + * Taking the shutdown_lock ensures the BAR mapping is not being + * altered by reset_work. Holding this lock before the RESETTING state + * change, if successful, also ensures nvme_remove won't be able to + * proceed to iounmap until we're done. + */ + mutex_lock(&dev->shutdown_lock); + if (!dev->bar_mapped_size) { + ret = -ENODEV; + goto unlock; + } + + if (!nvme_change_ctrl_state(ctrl, NVME_CTRL_RESETTING)) { + ret = -EBUSY; + goto unlock; + } + + writel(NVME_SUBSYS_RESET, dev->bar + NVME_REG_NSSR); + nvme_change_ctrl_state(ctrl, NVME_CTRL_LIVE); + + /* + * Read controller status to flush the previous write and trigger a + * pcie read error. + */ + readl(dev->bar + NVME_REG_CSTS); +unlock: + mutex_unlock(&dev->shutdown_lock); + return ret; +} + static int adapter_delete_queue(struct nvme_dev *dev, u8 opcode, u16 id) { struct nvme_command c = { }; @@ -1285,7 +1635,7 @@ static bool nvme_should_reset(struct nvme_dev *dev, u32 csts) bool nssro = dev->subsystem && (csts & NVME_CSTS_NSSRO); /* If there is a reset/reinit ongoing, we shouldn't reset again. */ - switch (dev->ctrl.state) { + switch (nvme_ctrl_state(&dev->ctrl)) { case NVME_CTRL_RESETTING: case NVME_CTRL_CONNECTING: return false; @@ -1325,7 +1675,7 @@ static void nvme_warn_reset(struct nvme_dev *dev, u32 csts) dev_warn(dev->ctrl.device, "Does your device have a faulty power saving mode enabled?\n"); dev_warn(dev->ctrl.device, - "Try \"nvme_core.default_ps_max_latency_us=0 pcie_aspm=off\" and report a bug\n"); + "Try \"nvme_core.default_ps_max_latency_us=0 pcie_aspm=off pcie_port_pm=off\" and report a bug\n"); } static enum blk_eh_timer_return nvme_timeout(struct request *req) @@ -1335,13 +1685,28 @@ static enum blk_eh_timer_return nvme_timeout(struct request *req) struct nvme_dev *dev = nvmeq->dev; struct request *abort_req; struct nvme_command cmd = { }; + struct pci_dev *pdev = to_pci_dev(dev->dev); u32 csts = readl(dev->bar + NVME_REG_CSTS); + u8 opcode; + + /* + * Shutdown the device immediately if we see it is disconnected. This + * unblocks PCIe error handling if the nvme driver is waiting in + * error_resume for a device that has been removed. We can't unbind the + * driver while the driver's error callback is waiting to complete, so + * we're relying on a timeout to break that deadlock if a removal + * occurs while reset work is running. + */ + if (pci_dev_is_disconnected(pdev)) + nvme_change_ctrl_state(&dev->ctrl, NVME_CTRL_DELETING); + if (nvme_state_terminal(&dev->ctrl)) + goto disable; /* If PCI error recovery process is happening, we cannot reset or * the recovery mechanism will surely fail. */ mb(); - if (pci_channel_offline(to_pci_dev(dev->dev))) + if (pci_channel_offline(pdev)) return BLK_EH_RESET_TIMER; /* @@ -1349,9 +1714,7 @@ static enum blk_eh_timer_return nvme_timeout(struct request *req) */ if (nvme_should_reset(dev, csts)) { nvme_warn_reset(dev, csts); - nvme_dev_disable(dev, false); - nvme_reset_ctrl(&dev->ctrl); - return BLK_EH_DONE; + goto disable; } /* @@ -1364,8 +1727,8 @@ static enum blk_eh_timer_return nvme_timeout(struct request *req) if (blk_mq_rq_state(req) != MQ_RQ_IN_FLIGHT) { dev_warn(dev->ctrl.device, - "I/O %d QID %d timeout, completion polled\n", - req->tag, nvmeq->qid); + "I/O tag %d (%04x) QID %d timeout, completion polled\n", + req->tag, nvme_cid(req), nvmeq->qid); return BLK_EH_DONE; } @@ -1375,14 +1738,14 @@ static enum blk_eh_timer_return nvme_timeout(struct request *req) * cancellation error. All outstanding requests are completed on * shutdown, so we return BLK_EH_DONE. */ - switch (dev->ctrl.state) { + switch (nvme_ctrl_state(&dev->ctrl)) { case NVME_CTRL_CONNECTING: nvme_change_ctrl_state(&dev->ctrl, NVME_CTRL_DELETING); fallthrough; case NVME_CTRL_DELETING: dev_warn_ratelimited(dev->ctrl.device, - "I/O %d QID %d timeout, disable controller\n", - req->tag, nvmeq->qid); + "I/O tag %d (%04x) QID %d timeout, disable controller\n", + req->tag, nvme_cid(req), nvmeq->qid); nvme_req(req)->flags |= NVME_REQ_CANCELLED; nvme_dev_disable(dev, true); return BLK_EH_DONE; @@ -1397,32 +1760,31 @@ static enum blk_eh_timer_return nvme_timeout(struct request *req) * command was already aborted once before and still hasn't been * returned to the driver, or if this is the admin queue. */ - if (!nvmeq->qid || iod->aborted) { + opcode = nvme_req(req)->cmd->common.opcode; + if (!nvmeq->qid || (iod->flags & IOD_ABORTED)) { dev_warn(dev->ctrl.device, - "I/O %d QID %d timeout, reset controller\n", - req->tag, nvmeq->qid); + "I/O tag %d (%04x) opcode %#x (%s) QID %d timeout, reset controller\n", + req->tag, nvme_cid(req), opcode, + nvme_opcode_str(nvmeq->qid, opcode), nvmeq->qid); nvme_req(req)->flags |= NVME_REQ_CANCELLED; - nvme_dev_disable(dev, false); - nvme_reset_ctrl(&dev->ctrl); - - return BLK_EH_DONE; + goto disable; } if (atomic_dec_return(&dev->ctrl.abort_limit) < 0) { atomic_inc(&dev->ctrl.abort_limit); return BLK_EH_RESET_TIMER; } - iod->aborted = true; + iod->flags |= IOD_ABORTED; cmd.abort.opcode = nvme_admin_abort_cmd; cmd.abort.cid = nvme_cid(req); cmd.abort.sqid = cpu_to_le16(nvmeq->qid); dev_warn(nvmeq->dev->ctrl.device, - "I/O %d (%s) QID %d timeout, aborting\n", - req->tag, - nvme_get_opcode_str(nvme_req(req)->cmd->common.opcode), - nvmeq->qid); + "I/O tag %d (%04x) opcode %#x (%s) QID %d timeout, aborting req_op:%s(%u) size:%u\n", + req->tag, nvme_cid(req), opcode, nvme_get_opcode_str(opcode), + nvmeq->qid, blk_op_str(req_op(req)), req_op(req), + blk_rq_bytes(req)); abort_req = blk_mq_alloc_request(dev->ctrl.admin_q, nvme_req_op(&cmd), BLK_MQ_REQ_NOWAIT); @@ -1442,6 +1804,18 @@ static enum blk_eh_timer_return nvme_timeout(struct request *req) * as the device then is in a faulty state. */ return BLK_EH_RESET_TIMER; + +disable: + if (!nvme_change_ctrl_state(&dev->ctrl, NVME_CTRL_RESETTING)) { + if (nvme_state_terminal(&dev->ctrl)) + nvme_dev_disable(dev, true); + return BLK_EH_DONE; + } + + nvme_dev_disable(dev, false); + if (nvme_try_sched_reset(&dev->ctrl)) + nvme_unquiesce_io_queues(&dev->ctrl); + return BLK_EH_DONE; } static void nvme_free_queue(struct nvme_queue *nvmeq) @@ -1641,7 +2015,7 @@ static int nvme_setup_io_queues_trylock(struct nvme_dev *dev) /* * Controller is in wrong state, fail early. */ - if (dev->ctrl.state != NVME_CTRL_CONNECTING) { + if (nvme_ctrl_state(&dev->ctrl) != NVME_CTRL_CONNECTING) { mutex_unlock(&dev->shutdown_lock); return -ENODEV; } @@ -1785,8 +2159,28 @@ static int nvme_pci_configure_admin_queue(struct nvme_dev *dev) * might be pointing at! */ result = nvme_disable_ctrl(&dev->ctrl, false); - if (result < 0) - return result; + if (result < 0) { + struct pci_dev *pdev = to_pci_dev(dev->dev); + + /* + * The NVMe Controller Reset method did not get an expected + * CSTS.RDY transition, so something with the device appears to + * be stuck. Use the lower level and bigger hammer PCIe + * Function Level Reset to attempt restoring the device to its + * initial state, and try again. + */ + result = pcie_reset_flr(pdev, false); + if (result < 0) + return result; + + pci_restore_state(pdev); + result = nvme_disable_ctrl(&dev->ctrl, false); + if (result < 0) + return result; + + dev_info(dev->ctrl.device, + "controller reset completed after pcie flr\n"); + } result = nvme_alloc_queue(dev, 0, NVME_AQ_DEPTH); if (result) @@ -1894,6 +2288,18 @@ static void nvme_map_cmb(struct nvme_dev *dev) return; /* + * Controllers may support a CMB size larger than their BAR, for + * example, due to being behind a bridge. Reduce the CMB to the + * reported size of the BAR + */ + size = min(size, bar_size - offset); + + if (!IS_ALIGNED(size, memremap_compat_align()) || + !IS_ALIGNED(pci_resource_start(pdev, bar), + memremap_compat_align())) + return; + + /* * Tell the controller about the host side address mapping the CMB, * and enable CMB decoding for the NVMe 1.4+ scheme: */ @@ -1903,17 +2309,10 @@ static void nvme_map_cmb(struct nvme_dev *dev) dev->bar + NVME_REG_CMBMSC); } - /* - * Controllers may support a CMB size larger than their BAR, - * for example, due to being behind a bridge. Reduce the CMB to - * the reported size of the BAR - */ - if (size > bar_size - offset) - size = bar_size - offset; - if (pci_p2pdma_add_resource(pdev, bar, size, offset)) { dev_warn(dev->ctrl.device, "failed to register the CMB\n"); + hi_lo_writeq(0, dev->bar + NVME_REG_CMBMSC); return; } @@ -1951,7 +2350,7 @@ static int nvme_set_host_mem(struct nvme_dev *dev, u32 bits) return ret; } -static void nvme_free_host_mem(struct nvme_dev *dev) +static void nvme_free_host_mem_multi(struct nvme_dev *dev) { int i; @@ -1966,18 +2365,54 @@ static void nvme_free_host_mem(struct nvme_dev *dev) kfree(dev->host_mem_desc_bufs); dev->host_mem_desc_bufs = NULL; - dma_free_coherent(dev->dev, - dev->nr_host_mem_descs * sizeof(*dev->host_mem_descs), +} + +static void nvme_free_host_mem(struct nvme_dev *dev) +{ + if (dev->hmb_sgt) + dma_free_noncontiguous(dev->dev, dev->host_mem_size, + dev->hmb_sgt, DMA_BIDIRECTIONAL); + else + nvme_free_host_mem_multi(dev); + + dma_free_coherent(dev->dev, dev->host_mem_descs_size, dev->host_mem_descs, dev->host_mem_descs_dma); dev->host_mem_descs = NULL; + dev->host_mem_descs_size = 0; dev->nr_host_mem_descs = 0; } -static int __nvme_alloc_host_mem(struct nvme_dev *dev, u64 preferred, +static int nvme_alloc_host_mem_single(struct nvme_dev *dev, u64 size) +{ + dev->hmb_sgt = dma_alloc_noncontiguous(dev->dev, size, + DMA_BIDIRECTIONAL, GFP_KERNEL, 0); + if (!dev->hmb_sgt) + return -ENOMEM; + + dev->host_mem_descs = dma_alloc_coherent(dev->dev, + sizeof(*dev->host_mem_descs), &dev->host_mem_descs_dma, + GFP_KERNEL); + if (!dev->host_mem_descs) { + dma_free_noncontiguous(dev->dev, size, dev->hmb_sgt, + DMA_BIDIRECTIONAL); + dev->hmb_sgt = NULL; + return -ENOMEM; + } + dev->host_mem_size = size; + dev->host_mem_descs_size = sizeof(*dev->host_mem_descs); + dev->nr_host_mem_descs = 1; + + dev->host_mem_descs[0].addr = + cpu_to_le64(dev->hmb_sgt->sgl->dma_address); + dev->host_mem_descs[0].size = cpu_to_le32(size / NVME_CTRL_PAGE_SIZE); + return 0; +} + +static int nvme_alloc_host_mem_multi(struct nvme_dev *dev, u64 preferred, u32 chunk_size) { struct nvme_host_mem_buf_desc *descs; - u32 max_entries, len; + u32 max_entries, len, descs_size; dma_addr_t descs_dma; int i = 0; void **bufs; @@ -1990,8 +2425,9 @@ static int __nvme_alloc_host_mem(struct nvme_dev *dev, u64 preferred, if (dev->ctrl.hmmaxd && dev->ctrl.hmmaxd < max_entries) max_entries = dev->ctrl.hmmaxd; - descs = dma_alloc_coherent(dev->dev, max_entries * sizeof(*descs), - &descs_dma, GFP_KERNEL); + descs_size = max_entries * sizeof(*descs); + descs = dma_alloc_coherent(dev->dev, descs_size, &descs_dma, + GFP_KERNEL); if (!descs) goto out; @@ -2020,22 +2456,14 @@ static int __nvme_alloc_host_mem(struct nvme_dev *dev, u64 preferred, dev->host_mem_size = size; dev->host_mem_descs = descs; dev->host_mem_descs_dma = descs_dma; + dev->host_mem_descs_size = descs_size; dev->host_mem_desc_bufs = bufs; return 0; out_free_bufs: - while (--i >= 0) { - size_t size = le32_to_cpu(descs[i].size) * NVME_CTRL_PAGE_SIZE; - - dma_free_attrs(dev->dev, size, bufs[i], - le64_to_cpu(descs[i].addr), - DMA_ATTR_NO_KERNEL_MAPPING | DMA_ATTR_NO_WARN); - } - kfree(bufs); out_free_descs: - dma_free_coherent(dev->dev, max_entries * sizeof(*descs), descs, - descs_dma); + dma_free_coherent(dev->dev, descs_size, descs, descs_dma); out: dev->host_mem_descs = NULL; return -ENOMEM; @@ -2043,13 +2471,23 @@ out: static int nvme_alloc_host_mem(struct nvme_dev *dev, u64 min, u64 preferred) { + unsigned long dma_merge_boundary = dma_get_merge_boundary(dev->dev); u64 min_chunk = min_t(u64, preferred, PAGE_SIZE * MAX_ORDER_NR_PAGES); u64 hmminds = max_t(u32, dev->ctrl.hmminds * 4096, PAGE_SIZE * 2); u64 chunk_size; + /* + * If there is an IOMMU that can merge pages, try a virtually + * non-contiguous allocation for a single segment first. + */ + if (dma_merge_boundary && (PAGE_SIZE & dma_merge_boundary) == 0) { + if (!nvme_alloc_host_mem_single(dev, preferred)) + return 0; + } + /* start big and work our way down */ for (chunk_size = min_chunk; chunk_size >= hmminds; chunk_size /= 2) { - if (!__nvme_alloc_host_mem(dev, preferred, chunk_size)) { + if (!nvme_alloc_host_mem_multi(dev, preferred, chunk_size)) { if (!min || dev->host_mem_size >= min) return 0; nvme_free_host_mem(dev); @@ -2097,8 +2535,10 @@ static int nvme_setup_host_mem(struct nvme_dev *dev) } dev_info(dev->ctrl.device, - "allocated %lld MiB host memory buffer.\n", - dev->host_mem_size >> ilog2(SZ_1M)); + "allocated %lld MiB host memory buffer (%u segment%s).\n", + dev->host_mem_size >> ilog2(SZ_1M), + dev->nr_host_mem_descs, + str_plural(dev->nr_host_mem_descs)); } ret = nvme_set_host_mem(dev, enable_bits); @@ -2112,7 +2552,7 @@ static ssize_t cmb_show(struct device *dev, struct device_attribute *attr, { struct nvme_dev *ndev = to_nvme_dev(dev_get_drvdata(dev)); - return sysfs_emit(buf, "cmbloc : x%08x\ncmbsz : x%08x\n", + return sysfs_emit(buf, "cmbloc : 0x%08x\ncmbsz : 0x%08x\n", ndev->cmbloc, ndev->cmbsz); } static DEVICE_ATTR_RO(cmb); @@ -2209,6 +2649,11 @@ static const struct attribute_group *nvme_pci_dev_attr_groups[] = { NULL, }; +static void nvme_update_attrs(struct nvme_dev *dev) +{ + sysfs_update_group(&dev->ctrl.device->kobj, &nvme_pci_dev_attrs_group); +} + /* * nirqs is the number of interrupts available for write and read * queues. The core already reserved an interrupt for the admin queue. @@ -2256,6 +2701,7 @@ static int nvme_setup_irqs(struct nvme_dev *dev, unsigned int nr_io_queues) .priv = dev, }; unsigned int irq_queues, poll_queues; + unsigned int flags = PCI_IRQ_ALL_TYPES | PCI_IRQ_AFFINITY; /* * Poll queues don't need interrupts, but we need at least one I/O queue @@ -2279,8 +2725,10 @@ static int nvme_setup_irqs(struct nvme_dev *dev, unsigned int nr_io_queues) irq_queues = 1; if (!(dev->ctrl.quirks & NVME_QUIRK_SINGLE_VECTOR)) irq_queues += (nr_io_queues - poll_queues); - return pci_alloc_irq_vectors_affinity(pdev, 1, irq_queues, - PCI_IRQ_ALL_TYPES | PCI_IRQ_AFFINITY, &affd); + if (dev->ctrl.quirks & NVME_QUIRK_BROKEN_MSI) + flags &= ~PCI_IRQ_MSI; + return pci_alloc_irq_vectors_affinity(pdev, 1, irq_queues, flags, + &affd); } static unsigned int nvme_max_io_queues(struct nvme_dev *dev) @@ -2291,7 +2739,8 @@ static unsigned int nvme_max_io_queues(struct nvme_dev *dev) */ if (dev->ctrl.quirks & NVME_QUIRK_SHARED_TAGS) return 1; - return num_possible_cpus() + dev->nr_write_queues + dev->nr_poll_queues; + return blk_mq_num_possible_queues(0) + dev->nr_write_queues + + dev->nr_poll_queues; } static int nvme_setup_io_queues(struct nvme_dev *dev) @@ -2498,30 +2947,44 @@ static unsigned int nvme_pci_nr_maps(struct nvme_dev *dev) return 1; } -static void nvme_pci_update_nr_queues(struct nvme_dev *dev) +static bool nvme_pci_update_nr_queues(struct nvme_dev *dev) { + if (!dev->ctrl.tagset) { + nvme_alloc_io_tag_set(&dev->ctrl, &dev->tagset, &nvme_mq_ops, + nvme_pci_nr_maps(dev), sizeof(struct nvme_iod)); + return true; + } + + /* Give up if we are racing with nvme_dev_disable() */ + if (!mutex_trylock(&dev->shutdown_lock)) + return false; + + /* Check if nvme_dev_disable() has been executed already */ + if (!dev->online_queues) { + mutex_unlock(&dev->shutdown_lock); + return false; + } + blk_mq_update_nr_hw_queues(&dev->tagset, dev->online_queues - 1); /* free previously allocated queues that are no longer usable */ nvme_free_queues(dev, dev->online_queues); + mutex_unlock(&dev->shutdown_lock); + return true; } static int nvme_pci_enable(struct nvme_dev *dev) { int result = -ENOMEM; struct pci_dev *pdev = to_pci_dev(dev->dev); - int dma_address_bits = 64; + unsigned int flags = PCI_IRQ_ALL_TYPES; if (pci_enable_device_mem(pdev)) return result; pci_set_master(pdev); - if (dev->ctrl.quirks & NVME_QUIRK_DMA_ADDRESS_BITS_48) - dma_address_bits = 48; - if (dma_set_mask_and_coherent(dev->dev, DMA_BIT_MASK(dma_address_bits))) - goto disable; - if (readl(dev->bar + NVME_REG_CSTS) == -1) { + dev_dbg(dev->ctrl.device, "reading CSTS register failed\n"); result = -ENODEV; goto disable; } @@ -2531,7 +2994,9 @@ static int nvme_pci_enable(struct nvme_dev *dev) * interrupts. Pre-enable a single MSIX or MSI vec for setup. We'll * adjust this later. */ - result = pci_alloc_irq_vectors(pdev, 1, 1, PCI_IRQ_ALL_TYPES); + if (dev->ctrl.quirks & NVME_QUIRK_BROKEN_MSI) + flags &= ~PCI_IRQ_MSI; + result = pci_alloc_irq_vectors(pdev, 1, 1, flags); if (result < 0) goto disable; @@ -2552,15 +3017,8 @@ static int nvme_pci_enable(struct nvme_dev *dev) else dev->io_sqes = NVME_NVM_IOSQES; - /* - * Temporary fix for the Apple controller found in the MacBook8,1 and - * some MacBook7,1 to avoid controller resets and data loss. - */ - if (pdev->vendor == PCI_VENDOR_ID_APPLE && pdev->device == 0x2001) { + if (dev->ctrl.quirks & NVME_QUIRK_QDEPTH_ONE) { dev->q_depth = 2; - dev_warn(dev->ctrl.device, "detected Apple NVMe controller, " - "set queue depth=%u to work around controller resets\n", - dev->q_depth); } else if (pdev->vendor == PCI_VENDOR_ID_SAMSUNG && (pdev->device == 0xa821 || pdev->device == 0xa822) && NVME_CAP_MQES(dev->ctrl.cap) == 0) { @@ -2583,7 +3041,6 @@ static int nvme_pci_enable(struct nvme_dev *dev) nvme_map_cmb(dev); - pci_enable_pcie_error_reporting(pdev); pci_save_state(pdev); result = nvme_pci_configure_admin_queue(dev); @@ -2621,13 +3078,13 @@ static bool nvme_pci_ctrl_is_dead(struct nvme_dev *dev) static void nvme_dev_disable(struct nvme_dev *dev, bool shutdown) { + enum nvme_ctrl_state state = nvme_ctrl_state(&dev->ctrl); struct pci_dev *pdev = to_pci_dev(dev->dev); bool dead; mutex_lock(&dev->shutdown_lock); dead = nvme_pci_ctrl_is_dead(dev); - if (dev->ctrl.state == NVME_CTRL_LIVE || - dev->ctrl.state == NVME_CTRL_RESETTING) { + if (state == NVME_CTRL_LIVE || state == NVME_CTRL_RESETTING) { if (pci_is_enabled(pdev)) nvme_start_freeze(&dev->ctrl); /* @@ -2648,10 +3105,8 @@ static void nvme_dev_disable(struct nvme_dev *dev, bool shutdown) nvme_suspend_io_queues(dev); nvme_suspend_queue(dev, 0); pci_free_irq_vectors(pdev); - if (pci_is_enabled(pdev)) { - pci_disable_pcie_error_reporting(pdev); + if (pci_is_enabled(pdev)) pci_disable_device(pdev); - } nvme_reap_pending_cqes(dev); nvme_cancel_tagset(&dev->ctrl); @@ -2678,42 +3133,15 @@ static int nvme_disable_prepare_reset(struct nvme_dev *dev, bool shutdown) return 0; } -static int nvme_setup_prp_pools(struct nvme_dev *dev) -{ - dev->prp_page_pool = dma_pool_create("prp list page", dev->dev, - NVME_CTRL_PAGE_SIZE, - NVME_CTRL_PAGE_SIZE, 0); - if (!dev->prp_page_pool) - return -ENOMEM; - - /* Optimisation for I/Os between 4k and 128k */ - dev->prp_small_pool = dma_pool_create("prp list 256", dev->dev, - 256, 256, 0); - if (!dev->prp_small_pool) { - dma_pool_destroy(dev->prp_page_pool); - return -ENOMEM; - } - return 0; -} - -static void nvme_release_prp_pools(struct nvme_dev *dev) -{ - dma_pool_destroy(dev->prp_page_pool); - dma_pool_destroy(dev->prp_small_pool); -} - static int nvme_pci_alloc_iod_mempool(struct nvme_dev *dev) { - size_t npages = max(nvme_pci_npages_prp(), nvme_pci_npages_sgl()); - size_t alloc_size = sizeof(__le64 *) * npages + - sizeof(struct scatterlist) * NVME_MAX_SEGS; + size_t alloc_size = sizeof(struct nvme_dma_vec) * NVME_MAX_SEGS; - WARN_ON_ONCE(alloc_size > PAGE_SIZE); - dev->iod_mempool = mempool_create_node(1, + dev->dmavec_mempool = mempool_create_node(1, mempool_kmalloc, mempool_kfree, (void *)alloc_size, GFP_KERNEL, dev_to_node(dev->dev)); - if (!dev->iod_mempool) + if (!dev->dmavec_mempool) return -ENOMEM; return 0; } @@ -2743,10 +3171,11 @@ static void nvme_reset_work(struct work_struct *work) bool was_suspend = !!(dev->ctrl.ctrl_config & NVME_CC_SHN_NORMAL); int result; - if (dev->ctrl.state != NVME_CTRL_RESETTING) { + if (nvme_ctrl_state(&dev->ctrl) != NVME_CTRL_RESETTING) { dev_warn(dev->ctrl.device, "ctrl state %d is not RESETTING\n", dev->ctrl.state); - return; + result = -ENODEV; + goto out; } /* @@ -2779,26 +3208,34 @@ static void nvme_reset_work(struct work_struct *work) if (result) goto out; + if (nvme_ctrl_meta_sgl_supported(&dev->ctrl)) + dev->ctrl.max_integrity_segments = NVME_MAX_META_SEGS; + else + dev->ctrl.max_integrity_segments = 1; + nvme_dbbuf_dma_alloc(dev); result = nvme_setup_host_mem(dev); if (result < 0) goto out; + nvme_update_attrs(dev); + result = nvme_setup_io_queues(dev); if (result) goto out; /* - * Freeze and update the number of I/O queues as thos might have + * Freeze and update the number of I/O queues as those might have * changed. If there are no I/O queues left after this reset, keep the * controller around but remove all namespaces. */ if (dev->online_queues > 1) { + nvme_dbbuf_set(dev); nvme_unquiesce_io_queues(&dev->ctrl); nvme_wait_freeze(&dev->ctrl); - nvme_pci_update_nr_queues(dev); - nvme_dbbuf_set(dev); + if (!nvme_pci_update_nr_queues(dev)) + goto out; nvme_unfreeze(&dev->ctrl); } else { dev_warn(dev->ctrl.device, "IO queues lost\n"); @@ -2833,7 +3270,9 @@ static void nvme_reset_work(struct work_struct *work) result); nvme_change_ctrl_state(&dev->ctrl, NVME_CTRL_DELETING); nvme_dev_disable(dev, true); + nvme_sync_queues(&dev->ctrl); nvme_mark_namespaces_dead(&dev->ctrl); + nvme_unquiesce_io_queues(&dev->ctrl); nvme_change_ctrl_state(&dev->ctrl, NVME_CTRL_DEAD); } @@ -2883,6 +3322,14 @@ static bool nvme_pci_supports_pci_p2pdma(struct nvme_ctrl *ctrl) return dma_pci_p2pdma_supported(dev->dev); } +static unsigned long nvme_pci_get_virt_boundary(struct nvme_ctrl *ctrl, + bool is_admin) +{ + if (!nvme_ctrl_sgl_supported(ctrl) || is_admin) + return NVME_CTRL_PAGE_SIZE - 1; + return 0; +} + static const struct nvme_ctrl_ops nvme_pci_ctrl_ops = { .name = "pcie", .module = THIS_MODULE, @@ -2893,9 +3340,11 @@ static const struct nvme_ctrl_ops nvme_pci_ctrl_ops = { .reg_read64 = nvme_pci_reg_read64, .free_ctrl = nvme_pci_free_ctrl, .submit_async_event = nvme_pci_submit_async_event, + .subsystem_reset = nvme_pci_subsystem_reset, .get_address = nvme_pci_get_address, .print_device_info = nvme_pci_print_device_info, .supports_pci_p2pdma = nvme_pci_supports_pci_p2pdma, + .get_virt_boundary = nvme_pci_get_virt_boundary, }; static int nvme_dev_map(struct nvme_dev *dev) @@ -2952,8 +3401,42 @@ static unsigned long check_vendor_combination_bug(struct pci_dev *pdev) if ((dmi_match(DMI_BOARD_VENDOR, "LENOVO")) && dmi_match(DMI_BOARD_NAME, "LNVNB161216")) return NVME_QUIRK_SIMPLE_SUSPEND; + } else if (pdev->vendor == 0x2646 && (pdev->device == 0x2263 || + pdev->device == 0x500f)) { + /* + * Exclude some Kingston NV1 and A2000 devices from + * NVME_QUIRK_SIMPLE_SUSPEND. Do a full suspend to save a + * lot of energy with s2idle sleep on some TUXEDO platforms. + */ + if (dmi_match(DMI_BOARD_NAME, "NS5X_NS7XAU") || + dmi_match(DMI_BOARD_NAME, "NS5x_7xAU") || + dmi_match(DMI_BOARD_NAME, "NS5x_7xPU") || + dmi_match(DMI_BOARD_NAME, "PH4PRX1_PH6PRX1")) + return NVME_QUIRK_FORCE_NO_SIMPLE_SUSPEND; + } else if (pdev->vendor == 0x144d && pdev->device == 0xa80d) { + /* + * Exclude Samsung 990 Evo from NVME_QUIRK_SIMPLE_SUSPEND + * because of high power consumption (> 2 Watt) in s2idle + * sleep. Only some boards with Intel CPU are affected. + * (Note for testing: Samsung 990 Evo Plus has same PCI ID) + */ + if (dmi_match(DMI_BOARD_NAME, "DN50Z-140HC-YD") || + dmi_match(DMI_BOARD_NAME, "GMxPXxx") || + dmi_match(DMI_BOARD_NAME, "GXxMRXx") || + dmi_match(DMI_BOARD_NAME, "NS5X_NS7XAU") || + dmi_match(DMI_BOARD_NAME, "PH4PG31") || + dmi_match(DMI_BOARD_NAME, "PH4PRX1_PH6PRX1") || + dmi_match(DMI_BOARD_NAME, "PH6PG01_PH6PG71")) + return NVME_QUIRK_FORCE_NO_SIMPLE_SUSPEND; } + /* + * NVMe SSD drops off the PCIe bus after system idle + * for 10 hours on a Lenovo N60z board. + */ + if (dmi_match(DMI_BOARD_NAME, "LXKT-ZXEG-N6")) + return NVME_QUIRK_NO_APST; + return 0; } @@ -2965,12 +3448,10 @@ static struct nvme_dev *nvme_pci_alloc_dev(struct pci_dev *pdev, struct nvme_dev *dev; int ret = -ENOMEM; - if (node == NUMA_NO_NODE) - set_dev_node(&pdev->dev, first_memory_node); - - dev = kzalloc_node(sizeof(*dev), GFP_KERNEL, node); + dev = kzalloc_node(struct_size(dev, descriptor_pools, nr_node_ids), + GFP_KERNEL, node); if (!dev) - return NULL; + return ERR_PTR(-ENOMEM); INIT_WORK(&dev->ctrl.reset_work, nvme_reset_work); mutex_init(&dev->shutdown_lock); @@ -2985,7 +3466,9 @@ static struct nvme_dev *nvme_pci_alloc_dev(struct pci_dev *pdev, dev->dev = get_device(&pdev->dev); quirks |= check_vendor_combination_bug(pdev); - if (!noacpi && acpi_storage_d3(&pdev->dev)) { + if (!noacpi && + !(quirks & NVME_QUIRK_FORCE_NO_SIMPLE_SUSPEND) && + acpi_storage_d3(&pdev->dev)) { /* * Some systems use a bios work around to ask for D3 on * platforms that support kernel managed suspend. @@ -2998,7 +3481,11 @@ static struct nvme_dev *nvme_pci_alloc_dev(struct pci_dev *pdev, quirks); if (ret) goto out_put_device; - + + if (dev->ctrl.quirks & NVME_QUIRK_DMA_ADDRESS_BITS_48) + dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(48)); + else + dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64)); dma_set_min_align_mask(&pdev->dev, NVME_CTRL_PAGE_SIZE - 1); dma_set_max_seg_size(&pdev->dev, 0xffffffff); @@ -3007,13 +3494,9 @@ static struct nvme_dev *nvme_pci_alloc_dev(struct pci_dev *pdev, * over a single page. */ dev->ctrl.max_hw_sectors = min_t(u32, - NVME_MAX_KB_SZ << 1, dma_max_mapping_size(&pdev->dev) >> 9); + NVME_MAX_BYTES >> SECTOR_SHIFT, + dma_opt_mapping_size(&pdev->dev) >> 9); dev->ctrl.max_segments = NVME_MAX_SEGS; - - /* - * There is no support for SGLs for metadata (yet), so we are limited to - * a single integrity segment for the separate metadata pointer. - */ dev->ctrl.max_integrity_segments = 1; return dev; @@ -3031,20 +3514,20 @@ static int nvme_probe(struct pci_dev *pdev, const struct pci_device_id *id) int result = -ENOMEM; dev = nvme_pci_alloc_dev(pdev, id); - if (!dev) - return -ENOMEM; + if (IS_ERR(dev)) + return PTR_ERR(dev); - result = nvme_dev_map(dev); + result = nvme_add_ctrl(&dev->ctrl); if (result) - goto out_uninit_ctrl; + goto out_put_ctrl; - result = nvme_setup_prp_pools(dev); + result = nvme_dev_map(dev); if (result) - goto out_dev_unmap; + goto out_uninit_ctrl; result = nvme_pci_alloc_iod_mempool(dev); if (result) - goto out_release_prp_pools; + goto out_dev_unmap; dev_info(dev->ctrl.device, "pci function %s\n", dev_name(&pdev->dev)); @@ -3072,12 +3555,19 @@ static int nvme_probe(struct pci_dev *pdev, const struct pci_device_id *id) if (result) goto out_disable; + if (nvme_ctrl_meta_sgl_supported(&dev->ctrl)) + dev->ctrl.max_integrity_segments = NVME_MAX_META_SEGS; + else + dev->ctrl.max_integrity_segments = 1; + nvme_dbbuf_dma_alloc(dev); result = nvme_setup_host_mem(dev); if (result < 0) goto out_disable; + nvme_update_attrs(dev); + result = nvme_setup_io_queues(dev); if (result) goto out_disable; @@ -3113,13 +3603,14 @@ out_disable: nvme_dbbuf_dma_free(dev); nvme_free_queues(dev, 0); out_release_iod_mempool: - mempool_destroy(dev->iod_mempool); -out_release_prp_pools: - nvme_release_prp_pools(dev); + mempool_destroy(dev->dmavec_mempool); out_dev_unmap: nvme_dev_unmap(dev); out_uninit_ctrl: nvme_uninit_ctrl(&dev->ctrl); +out_put_ctrl: + nvme_put_ctrl(&dev->ctrl); + dev_err_probe(&pdev->dev, result, "probe failed\n"); return result; } @@ -3176,8 +3667,8 @@ static void nvme_remove(struct pci_dev *pdev) nvme_dev_remove_admin(dev); nvme_dbbuf_dma_free(dev); nvme_free_queues(dev, 0); - mempool_destroy(dev->iod_mempool); - nvme_release_prp_pools(dev); + mempool_destroy(dev->dmavec_mempool); + nvme_release_descriptor_pools(dev); nvme_dev_unmap(dev); nvme_uninit_ctrl(&dev->ctrl); } @@ -3240,7 +3731,7 @@ static int nvme_suspend(struct device *dev) nvme_wait_freeze(ctrl); nvme_sync_queues(ctrl); - if (ctrl->state != NVME_CTRL_LIVE) + if (nvme_ctrl_state(ctrl) != NVME_CTRL_LIVE) goto unfreeze; /* @@ -3326,6 +3817,10 @@ static pci_ers_result_t nvme_error_detected(struct pci_dev *pdev, case pci_channel_io_frozen: dev_warn(dev->ctrl.device, "frozen state error detected, reset controller\n"); + if (!nvme_change_ctrl_state(&dev->ctrl, NVME_CTRL_RESETTING)) { + nvme_dev_disable(dev, true); + return PCI_ERS_RESULT_DISCONNECT; + } nvme_dev_disable(dev, false); return PCI_ERS_RESULT_NEED_RESET; case pci_channel_io_perm_failure: @@ -3342,7 +3837,8 @@ static pci_ers_result_t nvme_slot_reset(struct pci_dev *pdev) dev_info(dev->ctrl.device, "restart after slot reset\n"); pci_restore_state(pdev); - nvme_reset_ctrl(&dev->ctrl); + if (nvme_try_sched_reset(&dev->ctrl)) + nvme_unquiesce_io_queues(&dev->ctrl); return PCI_ERS_RESULT_RECOVERED; } @@ -3370,11 +3866,10 @@ static const struct pci_device_id nvme_id_table[] = { NVME_QUIRK_DEALLOCATE_ZEROES, }, { PCI_VDEVICE(INTEL, 0x0a54), /* Intel P4500/P4600 */ .driver_data = NVME_QUIRK_STRIPE_SIZE | - NVME_QUIRK_DEALLOCATE_ZEROES | - NVME_QUIRK_IGNORE_DEV_SUBNQN, }, + NVME_QUIRK_IGNORE_DEV_SUBNQN | + NVME_QUIRK_BOGUS_NID, }, { PCI_VDEVICE(INTEL, 0x0a55), /* Dell Express Flash P4600 */ - .driver_data = NVME_QUIRK_STRIPE_SIZE | - NVME_QUIRK_DEALLOCATE_ZEROES, }, + .driver_data = NVME_QUIRK_STRIPE_SIZE, }, { PCI_VDEVICE(INTEL, 0xf1a5), /* Intel 600P/P3100 */ .driver_data = NVME_QUIRK_NO_DEEPEST_PS | NVME_QUIRK_MEDIUM_PRIO_SQ | @@ -3388,6 +3883,14 @@ static const struct pci_device_id nvme_id_table[] = { NVME_QUIRK_BOGUS_NID, }, { PCI_VDEVICE(REDHAT, 0x0010), /* Qemu emulated controller */ .driver_data = NVME_QUIRK_BOGUS_NID, }, + { PCI_DEVICE(0x1217, 0x8760), /* O2 Micro 64GB Steam Deck */ + .driver_data = NVME_QUIRK_DMAPOOL_ALIGN_512, }, + { PCI_DEVICE(0x126f, 0x1001), /* Silicon Motion generic */ + .driver_data = NVME_QUIRK_NO_DEEPEST_PS | + NVME_QUIRK_IGNORE_DEV_SUBNQN, }, + { PCI_DEVICE(0x126f, 0x2262), /* Silicon Motion generic */ + .driver_data = NVME_QUIRK_NO_DEEPEST_PS | + NVME_QUIRK_BOGUS_NID, }, { PCI_DEVICE(0x126f, 0x2263), /* Silicon Motion unidentified */ .driver_data = NVME_QUIRK_NO_NS_DESC_LIST | NVME_QUIRK_BOGUS_NID, }, @@ -3406,6 +3909,11 @@ static const struct pci_device_id nvme_id_table[] = { .driver_data = NVME_QUIRK_DELAY_BEFORE_CHK_RDY | NVME_QUIRK_DISABLE_WRITE_ZEROES| NVME_QUIRK_IGNORE_DEV_SUBNQN, }, + { PCI_DEVICE(0x15b7, 0x5008), /* Sandisk SN530 */ + .driver_data = NVME_QUIRK_BROKEN_MSI }, + { PCI_DEVICE(0x15b7, 0x5009), /* Sandisk SN550 */ + .driver_data = NVME_QUIRK_BROKEN_MSI | + NVME_QUIRK_NO_DEEPEST_PS }, { PCI_DEVICE(0x1987, 0x5012), /* Phison E12 */ .driver_data = NVME_QUIRK_BOGUS_NID, }, { PCI_DEVICE(0x1987, 0x5016), /* Phison E16 */ @@ -3423,6 +3931,8 @@ static const struct pci_device_id nvme_id_table[] = { { PCI_DEVICE(0x10ec, 0x5762), /* ADATA SX6000LNP */ .driver_data = NVME_QUIRK_IGNORE_DEV_SUBNQN | NVME_QUIRK_BOGUS_NID, }, + { PCI_DEVICE(0x10ec, 0x5763), /* ADATA SX6000PNP */ + .driver_data = NVME_QUIRK_BOGUS_NID, }, { PCI_DEVICE(0x1cc1, 0x8201), /* ADATA SX8200PNP 512GB */ .driver_data = NVME_QUIRK_NO_DEEPEST_PS | NVME_QUIRK_IGNORE_DEV_SUBNQN, }, @@ -3434,14 +3944,19 @@ static const struct pci_device_id nvme_id_table[] = { .driver_data = NVME_QUIRK_DISABLE_WRITE_ZEROES, }, { PCI_DEVICE(0x1c5c, 0x174a), /* SK Hynix P31 SSD */ .driver_data = NVME_QUIRK_BOGUS_NID, }, + { PCI_DEVICE(0x1c5c, 0x1D59), /* SK Hynix BC901 */ + .driver_data = NVME_QUIRK_DISABLE_WRITE_ZEROES, }, { PCI_DEVICE(0x15b7, 0x2001), /* Sandisk Skyhawk */ .driver_data = NVME_QUIRK_DISABLE_WRITE_ZEROES, }, { PCI_DEVICE(0x1d97, 0x2263), /* SPCC */ .driver_data = NVME_QUIRK_DISABLE_WRITE_ZEROES, }, { PCI_DEVICE(0x144d, 0xa80b), /* Samsung PM9B1 256G and 512G */ - .driver_data = NVME_QUIRK_DISABLE_WRITE_ZEROES, }, + .driver_data = NVME_QUIRK_DISABLE_WRITE_ZEROES | + NVME_QUIRK_BOGUS_NID, }, { PCI_DEVICE(0x144d, 0xa809), /* Samsung MZALQ256HBJD 256G */ .driver_data = NVME_QUIRK_DISABLE_WRITE_ZEROES, }, + { PCI_DEVICE(0x144d, 0xa802), /* Samsung SM953 */ + .driver_data = NVME_QUIRK_BOGUS_NID, }, { PCI_DEVICE(0x1cc4, 0x6303), /* UMIS RPJTJ512MGE1QDY 512G */ .driver_data = NVME_QUIRK_DISABLE_WRITE_ZEROES, }, { PCI_DEVICE(0x1cc4, 0x6302), /* UMIS RPJTJ256MGE1QDY 256G */ @@ -3450,6 +3965,8 @@ static const struct pci_device_id nvme_id_table[] = { .driver_data = NVME_QUIRK_NO_DEEPEST_PS, }, { PCI_DEVICE(0x2646, 0x2263), /* KINGSTON A2000 NVMe SSD */ .driver_data = NVME_QUIRK_NO_DEEPEST_PS, }, + { PCI_DEVICE(0x2646, 0x5013), /* Kingston KC3000, Kingston FURY Renegade */ + .driver_data = NVME_QUIRK_NO_SECONDARY_TEMP_THRESH, }, { PCI_DEVICE(0x2646, 0x5018), /* KINGSTON OM8SFP4xxxxP OS21012 NVMe SSD */ .driver_data = NVME_QUIRK_DISABLE_WRITE_ZEROES, }, { PCI_DEVICE(0x2646, 0x5016), /* KINGSTON OM3PGP4xxxxP OS21011 NVMe SSD */ @@ -3460,6 +3977,8 @@ static const struct pci_device_id nvme_id_table[] = { .driver_data = NVME_QUIRK_DISABLE_WRITE_ZEROES, }, { PCI_DEVICE(0x2646, 0x501E), /* KINGSTON OM3PGP4xxxxQ OS21011 NVMe SSD */ .driver_data = NVME_QUIRK_DISABLE_WRITE_ZEROES, }, + { PCI_DEVICE(0x1f40, 0x1202), /* Netac Technologies Co. NV3000 NVMe SSD */ + .driver_data = NVME_QUIRK_BOGUS_NID, }, { PCI_DEVICE(0x1f40, 0x5236), /* Netac Technologies Co. NV7000 NVMe SSD */ .driver_data = NVME_QUIRK_BOGUS_NID, }, { PCI_DEVICE(0x1e4B, 0x1001), /* MAXIO MAP1001 */ @@ -3468,19 +3987,34 @@ static const struct pci_device_id nvme_id_table[] = { .driver_data = NVME_QUIRK_BOGUS_NID, }, { PCI_DEVICE(0x1e4B, 0x1202), /* MAXIO MAP1202 */ .driver_data = NVME_QUIRK_BOGUS_NID, }, + { PCI_DEVICE(0x1e4B, 0x1602), /* MAXIO MAP1602 */ + .driver_data = NVME_QUIRK_BOGUS_NID, }, { PCI_DEVICE(0x1cc1, 0x5350), /* ADATA XPG GAMMIX S50 */ .driver_data = NVME_QUIRK_BOGUS_NID, }, + { PCI_DEVICE(0x1dbe, 0x5216), /* Acer/INNOGRIT FA100/5216 NVMe SSD */ + .driver_data = NVME_QUIRK_BOGUS_NID, }, { PCI_DEVICE(0x1dbe, 0x5236), /* ADATA XPG GAMMIX S70 */ .driver_data = NVME_QUIRK_BOGUS_NID, }, { PCI_DEVICE(0x1e49, 0x0021), /* ZHITAI TiPro5000 NVMe SSD */ .driver_data = NVME_QUIRK_NO_DEEPEST_PS, }, { PCI_DEVICE(0x1e49, 0x0041), /* ZHITAI TiPro7000 NVMe SSD */ .driver_data = NVME_QUIRK_NO_DEEPEST_PS, }, + { PCI_DEVICE(0x025e, 0xf1ac), /* SOLIDIGM P44 pro SSDPFKKW020X7 */ + .driver_data = NVME_QUIRK_NO_DEEPEST_PS, }, { PCI_DEVICE(0xc0a9, 0x540a), /* Crucial P2 */ .driver_data = NVME_QUIRK_BOGUS_NID, }, { PCI_DEVICE(0x1d97, 0x2263), /* Lexar NM610 */ .driver_data = NVME_QUIRK_BOGUS_NID, }, + { PCI_DEVICE(0x1d97, 0x1d97), /* Lexar NM620 */ + .driver_data = NVME_QUIRK_BOGUS_NID, }, { PCI_DEVICE(0x1d97, 0x2269), /* Lexar NM760 */ + .driver_data = NVME_QUIRK_BOGUS_NID | + NVME_QUIRK_IGNORE_DEV_SUBNQN, }, + { PCI_DEVICE(0x10ec, 0x5763), /* TEAMGROUP T-FORCE CARDEA ZERO Z330 SSD */ + .driver_data = NVME_QUIRK_BOGUS_NID, }, + { PCI_DEVICE(0x1e4b, 0x1602), /* HS-SSD-FUTURE 2048G */ + .driver_data = NVME_QUIRK_BOGUS_NID, }, + { PCI_DEVICE(0x10ec, 0x5765), /* TEAMGROUP MP33 2TB SSD */ .driver_data = NVME_QUIRK_BOGUS_NID, }, { PCI_DEVICE(PCI_VENDOR_ID_AMAZON, 0x0061), .driver_data = NVME_QUIRK_DMA_ADDRESS_BITS_48, }, @@ -3495,7 +4029,12 @@ static const struct pci_device_id nvme_id_table[] = { { PCI_DEVICE(PCI_VENDOR_ID_AMAZON, 0xcd02), .driver_data = NVME_QUIRK_DMA_ADDRESS_BITS_48, }, { PCI_DEVICE(PCI_VENDOR_ID_APPLE, 0x2001), - .driver_data = NVME_QUIRK_SINGLE_VECTOR }, + /* + * Fix for the Apple controller found in the MacBook8,1 and + * some MacBook7,1 to avoid controller resets and data loss. + */ + .driver_data = NVME_QUIRK_SINGLE_VECTOR | + NVME_QUIRK_QDEPTH_ONE }, { PCI_DEVICE(PCI_VENDOR_ID_APPLE, 0x2003) }, { PCI_DEVICE(PCI_VENDOR_ID_APPLE, 0x2005), .driver_data = NVME_QUIRK_SINGLE_VECTOR | @@ -3530,8 +4069,6 @@ static int __init nvme_init(void) BUILD_BUG_ON(sizeof(struct nvme_create_sq) != 64); BUILD_BUG_ON(sizeof(struct nvme_delete_queue) != 64); BUILD_BUG_ON(IRQ_AFFINITY_MAX_SETS < 2); - BUILD_BUG_ON(DIV_ROUND_UP(nvme_pci_npages_prp(), NVME_CTRL_PAGE_SIZE) > - S8_MAX); return pci_register_driver(&nvme_driver); } @@ -3545,5 +4082,6 @@ static void __exit nvme_exit(void) MODULE_AUTHOR("Matthew Wilcox <willy@linux.intel.com>"); MODULE_LICENSE("GPL"); MODULE_VERSION("1.0"); +MODULE_DESCRIPTION("NVMe host PCIe transport driver"); module_init(nvme_init); module_exit(nvme_exit); diff --git a/drivers/nvme/host/pr.c b/drivers/nvme/host/pr.c new file mode 100644 index 000000000000..ad2ecc2f49a9 --- /dev/null +++ b/drivers/nvme/host/pr.c @@ -0,0 +1,344 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (c) 2015 Intel Corporation + * Keith Busch <kbusch@kernel.org> + */ +#include <linux/blkdev.h> +#include <linux/pr.h> +#include <linux/unaligned.h> + +#include "nvme.h" + +static enum nvme_pr_type nvme_pr_type_from_blk(enum pr_type type) +{ + switch (type) { + case PR_WRITE_EXCLUSIVE: + return NVME_PR_WRITE_EXCLUSIVE; + case PR_EXCLUSIVE_ACCESS: + return NVME_PR_EXCLUSIVE_ACCESS; + case PR_WRITE_EXCLUSIVE_REG_ONLY: + return NVME_PR_WRITE_EXCLUSIVE_REG_ONLY; + case PR_EXCLUSIVE_ACCESS_REG_ONLY: + return NVME_PR_EXCLUSIVE_ACCESS_REG_ONLY; + case PR_WRITE_EXCLUSIVE_ALL_REGS: + return NVME_PR_WRITE_EXCLUSIVE_ALL_REGS; + case PR_EXCLUSIVE_ACCESS_ALL_REGS: + return NVME_PR_EXCLUSIVE_ACCESS_ALL_REGS; + } + + return 0; +} + +static enum pr_type block_pr_type_from_nvme(enum nvme_pr_type type) +{ + switch (type) { + case NVME_PR_WRITE_EXCLUSIVE: + return PR_WRITE_EXCLUSIVE; + case NVME_PR_EXCLUSIVE_ACCESS: + return PR_EXCLUSIVE_ACCESS; + case NVME_PR_WRITE_EXCLUSIVE_REG_ONLY: + return PR_WRITE_EXCLUSIVE_REG_ONLY; + case NVME_PR_EXCLUSIVE_ACCESS_REG_ONLY: + return PR_EXCLUSIVE_ACCESS_REG_ONLY; + case NVME_PR_WRITE_EXCLUSIVE_ALL_REGS: + return PR_WRITE_EXCLUSIVE_ALL_REGS; + case NVME_PR_EXCLUSIVE_ACCESS_ALL_REGS: + return PR_EXCLUSIVE_ACCESS_ALL_REGS; + } + + return 0; +} + +static int nvme_send_ns_head_pr_command(struct block_device *bdev, + struct nvme_command *c, void *data, unsigned int data_len) +{ + struct nvme_ns_head *head = bdev->bd_disk->private_data; + int srcu_idx = srcu_read_lock(&head->srcu); + struct nvme_ns *ns = nvme_find_path(head); + int ret = -EWOULDBLOCK; + + if (ns) { + c->common.nsid = cpu_to_le32(ns->head->ns_id); + ret = nvme_submit_sync_cmd(ns->queue, c, data, data_len); + } + srcu_read_unlock(&head->srcu, srcu_idx); + return ret; +} + +static int nvme_send_ns_pr_command(struct nvme_ns *ns, struct nvme_command *c, + void *data, unsigned int data_len) +{ + c->common.nsid = cpu_to_le32(ns->head->ns_id); + return nvme_submit_sync_cmd(ns->queue, c, data, data_len); +} + +static int nvme_status_to_pr_err(int status) +{ + if (nvme_is_path_error(status)) + return PR_STS_PATH_FAILED; + + switch (status & NVME_SCT_SC_MASK) { + case NVME_SC_SUCCESS: + return PR_STS_SUCCESS; + case NVME_SC_RESERVATION_CONFLICT: + return PR_STS_RESERVATION_CONFLICT; + case NVME_SC_BAD_ATTRIBUTES: + case NVME_SC_INVALID_OPCODE: + case NVME_SC_INVALID_FIELD: + case NVME_SC_INVALID_NS: + return -EINVAL; + default: + return PR_STS_IOERR; + } +} + +static int __nvme_send_pr_command(struct block_device *bdev, u32 cdw10, + u32 cdw11, u8 op, void *data, unsigned int data_len) +{ + struct nvme_command c = { 0 }; + + c.common.opcode = op; + c.common.cdw10 = cpu_to_le32(cdw10); + c.common.cdw11 = cpu_to_le32(cdw11); + + if (nvme_disk_is_ns_head(bdev->bd_disk)) + return nvme_send_ns_head_pr_command(bdev, &c, data, data_len); + return nvme_send_ns_pr_command(bdev->bd_disk->private_data, &c, + data, data_len); +} + +static int nvme_send_pr_command(struct block_device *bdev, u32 cdw10, u32 cdw11, + u8 op, void *data, unsigned int data_len) +{ + int ret; + + ret = __nvme_send_pr_command(bdev, cdw10, cdw11, op, data, data_len); + return ret < 0 ? ret : nvme_status_to_pr_err(ret); +} + +static int nvme_pr_register(struct block_device *bdev, u64 old_key, u64 new_key, + unsigned int flags) +{ + struct nvmet_pr_register_data data = { 0 }; + u32 cdw10; + + if (flags & ~PR_FL_IGNORE_KEY) + return -EOPNOTSUPP; + + data.crkey = cpu_to_le64(old_key); + data.nrkey = cpu_to_le64(new_key); + + cdw10 = old_key ? NVME_PR_REGISTER_ACT_REPLACE : + NVME_PR_REGISTER_ACT_REG; + cdw10 |= (flags & PR_FL_IGNORE_KEY) ? NVME_PR_IGNORE_KEY : 0; + cdw10 |= NVME_PR_CPTPL_PERSIST; + + return nvme_send_pr_command(bdev, cdw10, 0, nvme_cmd_resv_register, + &data, sizeof(data)); +} + +static int nvme_pr_reserve(struct block_device *bdev, u64 key, + enum pr_type type, unsigned flags) +{ + struct nvmet_pr_acquire_data data = { 0 }; + u32 cdw10; + + if (flags & ~PR_FL_IGNORE_KEY) + return -EOPNOTSUPP; + + data.crkey = cpu_to_le64(key); + + cdw10 = NVME_PR_ACQUIRE_ACT_ACQUIRE; + cdw10 |= nvme_pr_type_from_blk(type) << 8; + cdw10 |= (flags & PR_FL_IGNORE_KEY) ? NVME_PR_IGNORE_KEY : 0; + + return nvme_send_pr_command(bdev, cdw10, 0, nvme_cmd_resv_acquire, + &data, sizeof(data)); +} + +static int nvme_pr_preempt(struct block_device *bdev, u64 old, u64 new, + enum pr_type type, bool abort) +{ + struct nvmet_pr_acquire_data data = { 0 }; + u32 cdw10; + + data.crkey = cpu_to_le64(old); + data.prkey = cpu_to_le64(new); + + cdw10 = abort ? NVME_PR_ACQUIRE_ACT_PREEMPT_AND_ABORT : + NVME_PR_ACQUIRE_ACT_PREEMPT; + cdw10 |= nvme_pr_type_from_blk(type) << 8; + + return nvme_send_pr_command(bdev, cdw10, 0, nvme_cmd_resv_acquire, + &data, sizeof(data)); +} + +static int nvme_pr_clear(struct block_device *bdev, u64 key) +{ + struct nvmet_pr_release_data data = { 0 }; + u32 cdw10; + + data.crkey = cpu_to_le64(key); + + cdw10 = NVME_PR_RELEASE_ACT_CLEAR; + cdw10 |= key ? 0 : NVME_PR_IGNORE_KEY; + + return nvme_send_pr_command(bdev, cdw10, 0, nvme_cmd_resv_release, + &data, sizeof(data)); +} + +static int nvme_pr_release(struct block_device *bdev, u64 key, enum pr_type type) +{ + struct nvmet_pr_release_data data = { 0 }; + u32 cdw10; + + data.crkey = cpu_to_le64(key); + + cdw10 = NVME_PR_RELEASE_ACT_RELEASE; + cdw10 |= nvme_pr_type_from_blk(type) << 8; + cdw10 |= key ? 0 : NVME_PR_IGNORE_KEY; + + return nvme_send_pr_command(bdev, cdw10, 0, nvme_cmd_resv_release, + &data, sizeof(data)); +} + +static int nvme_pr_resv_report(struct block_device *bdev, void *data, + u32 data_len, bool *eds) +{ + u32 cdw10, cdw11; + int ret; + + cdw10 = nvme_bytes_to_numd(data_len); + cdw11 = NVME_EXTENDED_DATA_STRUCT; + *eds = true; + +retry: + ret = __nvme_send_pr_command(bdev, cdw10, cdw11, nvme_cmd_resv_report, + data, data_len); + if (ret == NVME_SC_HOST_ID_INCONSIST && + cdw11 == NVME_EXTENDED_DATA_STRUCT) { + cdw11 = 0; + *eds = false; + goto retry; + } + + return ret < 0 ? ret : nvme_status_to_pr_err(ret); +} + +static int nvme_pr_read_keys(struct block_device *bdev, + struct pr_keys *keys_info) +{ + size_t rse_len; + u32 num_keys = keys_info->num_keys; + struct nvme_reservation_status_ext *rse; + int ret, i; + bool eds; + + /* + * Assume we are using 128-bit host IDs and allocate a buffer large + * enough to get enough keys to fill the return keys buffer. + */ + rse_len = struct_size(rse, regctl_eds, num_keys); + if (rse_len > U32_MAX) + return -EINVAL; + + rse = kzalloc(rse_len, GFP_KERNEL); + if (!rse) + return -ENOMEM; + + ret = nvme_pr_resv_report(bdev, rse, rse_len, &eds); + if (ret) + goto free_rse; + + keys_info->generation = le32_to_cpu(rse->gen); + keys_info->num_keys = get_unaligned_le16(&rse->regctl); + + num_keys = min(num_keys, keys_info->num_keys); + for (i = 0; i < num_keys; i++) { + if (eds) { + keys_info->keys[i] = + le64_to_cpu(rse->regctl_eds[i].rkey); + } else { + struct nvme_reservation_status *rs; + + rs = (struct nvme_reservation_status *)rse; + keys_info->keys[i] = le64_to_cpu(rs->regctl_ds[i].rkey); + } + } + +free_rse: + kfree(rse); + return ret; +} + +static int nvme_pr_read_reservation(struct block_device *bdev, + struct pr_held_reservation *resv) +{ + struct nvme_reservation_status_ext tmp_rse, *rse; + int ret, i, num_regs; + u32 rse_len; + bool eds; + +get_num_regs: + /* + * Get the number of registrations so we know how big to allocate + * the response buffer. + */ + ret = nvme_pr_resv_report(bdev, &tmp_rse, sizeof(tmp_rse), &eds); + if (ret) + return ret; + + num_regs = get_unaligned_le16(&tmp_rse.regctl); + if (!num_regs) { + resv->generation = le32_to_cpu(tmp_rse.gen); + return 0; + } + + rse_len = struct_size(rse, regctl_eds, num_regs); + rse = kzalloc(rse_len, GFP_KERNEL); + if (!rse) + return -ENOMEM; + + ret = nvme_pr_resv_report(bdev, rse, rse_len, &eds); + if (ret) + goto free_rse; + + if (num_regs != get_unaligned_le16(&rse->regctl)) { + kfree(rse); + goto get_num_regs; + } + + resv->generation = le32_to_cpu(rse->gen); + resv->type = block_pr_type_from_nvme(rse->rtype); + + for (i = 0; i < num_regs; i++) { + if (eds) { + if (rse->regctl_eds[i].rcsts) { + resv->key = le64_to_cpu(rse->regctl_eds[i].rkey); + break; + } + } else { + struct nvme_reservation_status *rs; + + rs = (struct nvme_reservation_status *)rse; + if (rs->regctl_ds[i].rcsts) { + resv->key = le64_to_cpu(rs->regctl_ds[i].rkey); + break; + } + } + } + +free_rse: + kfree(rse); + return ret; +} + +const struct pr_ops nvme_pr_ops = { + .pr_register = nvme_pr_register, + .pr_reserve = nvme_pr_reserve, + .pr_release = nvme_pr_release, + .pr_preempt = nvme_pr_preempt, + .pr_clear = nvme_pr_clear, + .pr_read_keys = nvme_pr_read_keys, + .pr_read_reservation = nvme_pr_read_reservation, +}; diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c index bbad26b82b56..35c0822edb2d 100644 --- a/drivers/nvme/host/rdma.c +++ b/drivers/nvme/host/rdma.c @@ -12,14 +12,13 @@ #include <linux/string.h> #include <linux/atomic.h> #include <linux/blk-mq.h> -#include <linux/blk-mq-rdma.h> #include <linux/blk-integrity.h> #include <linux/types.h> #include <linux/list.h> #include <linux/mutex.h> #include <linux/scatterlist.h> #include <linux/nvme.h> -#include <asm/unaligned.h> +#include <linux/unaligned.h> #include <rdma/ib_verbs.h> #include <rdma/rdma_cm.h> @@ -222,7 +221,7 @@ static struct nvme_rdma_qe *nvme_rdma_alloc_ring(struct ib_device *ibdev, /* * Bind the CQEs (post recv buffers) DMA mapping to the RDMA queue - * lifetime. It's safe, since any chage in the underlying RDMA device + * lifetime. It's safe, since any change in the underlying RDMA device * will issue error recovery and queue re-creation. */ for (i = 0; i < ib_queue_size; i++) { @@ -464,7 +463,6 @@ static int nvme_rdma_create_cq(struct ib_device *ibdev, struct nvme_rdma_queue *queue) { int ret, comp_vector, idx = nvme_rdma_queue_idx(queue); - enum ib_poll_context poll_ctx; /* * Spread I/O queues completion vectors according their queue index. @@ -473,15 +471,12 @@ static int nvme_rdma_create_cq(struct ib_device *ibdev, comp_vector = (idx == 0 ? idx : idx - 1) % ibdev->num_comp_vectors; /* Polling queues need direct cq polling context */ - if (nvme_rdma_poll_queue(queue)) { - poll_ctx = IB_POLL_DIRECT; + if (nvme_rdma_poll_queue(queue)) queue->ib_cq = ib_alloc_cq(ibdev, queue, queue->cq_size, - comp_vector, poll_ctx); - } else { - poll_ctx = IB_POLL_SOFTIRQ; + comp_vector, IB_POLL_DIRECT); + else queue->ib_cq = ib_cq_pool_get(ibdev, queue->cq_size, - comp_vector, poll_ctx); - } + comp_vector, IB_POLL_SOFTIRQ); if (IS_ERR(queue->ib_cq)) { ret = PTR_ERR(queue->ib_cq); @@ -506,7 +501,7 @@ static int nvme_rdma_create_queue_ib(struct nvme_rdma_queue *queue) } ibdev = queue->device->dev; - /* +1 for ib_stop_cq */ + /* +1 for ib_drain_qp */ queue->cq_size = cq_factor * queue->queue_size + 1; ret = nvme_rdma_create_cq(ibdev, queue); @@ -643,6 +638,9 @@ static void __nvme_rdma_stop_queue(struct nvme_rdma_queue *queue) static void nvme_rdma_stop_queue(struct nvme_rdma_queue *queue) { + if (!test_bit(NVME_RDMA_Q_ALLOCATED, &queue->flags)) + return; + mutex_lock(&queue->queue_lock); if (test_and_clear_bit(NVME_RDMA_Q_LIVE, &queue->flags)) __nvme_rdma_stop_queue(queue); @@ -718,18 +716,10 @@ out_stop_queues: static int nvme_rdma_alloc_io_queues(struct nvme_rdma_ctrl *ctrl) { struct nvmf_ctrl_options *opts = ctrl->ctrl.opts; - struct ib_device *ibdev = ctrl->device->dev; - unsigned int nr_io_queues, nr_default_queues; - unsigned int nr_read_queues, nr_poll_queues; + unsigned int nr_io_queues; int i, ret; - nr_read_queues = min_t(unsigned int, ibdev->num_comp_vectors, - min(opts->nr_io_queues, num_online_cpus())); - nr_default_queues = min_t(unsigned int, ibdev->num_comp_vectors, - min(opts->nr_write_queues, num_online_cpus())); - nr_poll_queues = min(opts->nr_poll_queues, num_online_cpus()); - nr_io_queues = nr_read_queues + nr_default_queues + nr_poll_queues; - + nr_io_queues = nvmf_nr_io_queues(opts); ret = nvme_set_queue_count(&ctrl->ctrl, &nr_io_queues); if (ret) return ret; @@ -744,34 +734,7 @@ static int nvme_rdma_alloc_io_queues(struct nvme_rdma_ctrl *ctrl) dev_info(ctrl->ctrl.device, "creating %d I/O queues.\n", nr_io_queues); - if (opts->nr_write_queues && nr_read_queues < nr_io_queues) { - /* - * separate read/write queues - * hand out dedicated default queues only after we have - * sufficient read queues. - */ - ctrl->io_queues[HCTX_TYPE_READ] = nr_read_queues; - nr_io_queues -= ctrl->io_queues[HCTX_TYPE_READ]; - ctrl->io_queues[HCTX_TYPE_DEFAULT] = - min(nr_default_queues, nr_io_queues); - nr_io_queues -= ctrl->io_queues[HCTX_TYPE_DEFAULT]; - } else { - /* - * shared read/write queues - * either no write queues were requested, or we don't have - * sufficient queue count to have dedicated default queues. - */ - ctrl->io_queues[HCTX_TYPE_DEFAULT] = - min(nr_read_queues, nr_io_queues); - nr_io_queues -= ctrl->io_queues[HCTX_TYPE_DEFAULT]; - } - - if (opts->nr_poll_queues && nr_io_queues) { - /* map dedicated poll queues only if we have queues left */ - ctrl->io_queues[HCTX_TYPE_POLL] = - min(nr_poll_queues, nr_io_queues); - } - + nvmf_set_io_queues(opts, nr_io_queues, ctrl->io_queues); for (i = 1; i < ctrl->ctrl.queue_count; i++) { ret = nvme_rdma_alloc_queue(ctrl, i, ctrl->ctrl.sqsize + 1); @@ -837,7 +800,7 @@ static int nvme_rdma_configure_admin_queue(struct nvme_rdma_ctrl *ctrl, /* * Bind the async event SQE DMA mapping to the admin queue lifetime. - * It's safe, since any chage in the underlying RDMA device will issue + * It's safe, since any change in the underlying RDMA device will issue * error recovery and queue re-creation. */ error = nvme_rdma_alloc_qe(ctrl->device->dev, &ctrl->async_event_sqe, @@ -914,7 +877,7 @@ static int nvme_rdma_configure_io_queues(struct nvme_rdma_ctrl *ctrl, bool new) /* * Only start IO queues for which we have allocated the tagset - * and limitted it to the available queues. On reconnects, the + * and limited it to the available queues. On reconnects, the * queue number might have changed. */ nr_queues = min(ctrl->tag_set.nr_hw_queues + 1, ctrl->ctrl.queue_count); @@ -923,6 +886,7 @@ static int nvme_rdma_configure_io_queues(struct nvme_rdma_ctrl *ctrl, bool new) goto out_cleanup_tagset; if (!new) { + nvme_start_freeze(&ctrl->ctrl); nvme_unquiesce_io_queues(&ctrl->ctrl); if (!nvme_wait_freeze_timeout(&ctrl->ctrl, NVME_IO_TIMEOUT)) { /* @@ -931,6 +895,7 @@ static int nvme_rdma_configure_io_queues(struct nvme_rdma_ctrl *ctrl, bool new) * to be safe. */ ret = -ENODEV; + nvme_unfreeze(&ctrl->ctrl); goto out_wait_freeze_timed_out; } blk_mq_update_nr_hw_queues(ctrl->ctrl.tagset, @@ -980,7 +945,6 @@ static void nvme_rdma_teardown_io_queues(struct nvme_rdma_ctrl *ctrl, bool remove) { if (ctrl->ctrl.queue_count > 1) { - nvme_start_freeze(&ctrl->ctrl); nvme_quiesce_io_queues(&ctrl->ctrl); nvme_sync_io_queues(&ctrl->ctrl); nvme_rdma_stop_io_queues(ctrl); @@ -1018,16 +982,18 @@ free_ctrl: kfree(ctrl); } -static void nvme_rdma_reconnect_or_remove(struct nvme_rdma_ctrl *ctrl) +static void nvme_rdma_reconnect_or_remove(struct nvme_rdma_ctrl *ctrl, + int status) { + enum nvme_ctrl_state state = nvme_ctrl_state(&ctrl->ctrl); + /* If we are resetting/deleting then do nothing */ - if (ctrl->ctrl.state != NVME_CTRL_CONNECTING) { - WARN_ON_ONCE(ctrl->ctrl.state == NVME_CTRL_NEW || - ctrl->ctrl.state == NVME_CTRL_LIVE); + if (state != NVME_CTRL_CONNECTING) { + WARN_ON_ONCE(state == NVME_CTRL_NEW || state == NVME_CTRL_LIVE); return; } - if (nvmf_should_reconnect(&ctrl->ctrl)) { + if (nvmf_should_reconnect(&ctrl->ctrl, status)) { dev_info(ctrl->ctrl.device, "Reconnecting in %d seconds...\n", ctrl->ctrl.opts->reconnect_delay); queue_delayed_work(nvme_wq, &ctrl->reconnect_work, @@ -1041,6 +1007,7 @@ static int nvme_rdma_setup_ctrl(struct nvme_rdma_ctrl *ctrl, bool new) { int ret; bool changed; + u16 max_queue_size; ret = nvme_rdma_configure_admin_queue(ctrl, new); if (ret) @@ -1052,7 +1019,7 @@ static int nvme_rdma_setup_ctrl(struct nvme_rdma_ctrl *ctrl, bool new) goto destroy_admin; } - if (!(ctrl->ctrl.sgls & (1 << 2))) { + if (!(ctrl->ctrl.sgls & NVME_CTRL_SGLS_KSDBDS)) { ret = -EOPNOTSUPP; dev_err(ctrl->ctrl.device, "Mandatory keyed sgls are not supported!\n"); @@ -1065,11 +1032,16 @@ static int nvme_rdma_setup_ctrl(struct nvme_rdma_ctrl *ctrl, bool new) ctrl->ctrl.opts->queue_size, ctrl->ctrl.sqsize + 1); } - if (ctrl->ctrl.sqsize + 1 > NVME_RDMA_MAX_QUEUE_SIZE) { + if (ctrl->ctrl.max_integrity_segments) + max_queue_size = NVME_RDMA_MAX_METADATA_QUEUE_SIZE; + else + max_queue_size = NVME_RDMA_MAX_QUEUE_SIZE; + + if (ctrl->ctrl.sqsize + 1 > max_queue_size) { dev_warn(ctrl->ctrl.device, - "ctrl sqsize %u > max queue size %u, clamping down\n", - ctrl->ctrl.sqsize + 1, NVME_RDMA_MAX_QUEUE_SIZE); - ctrl->ctrl.sqsize = NVME_RDMA_MAX_QUEUE_SIZE - 1; + "ctrl sqsize %u > max queue size %u, clamping down\n", + ctrl->ctrl.sqsize + 1, max_queue_size); + ctrl->ctrl.sqsize = max_queue_size - 1; } if (ctrl->ctrl.sqsize + 1 > ctrl->ctrl.maxcmd) { @@ -1079,7 +1051,7 @@ static int nvme_rdma_setup_ctrl(struct nvme_rdma_ctrl *ctrl, bool new) ctrl->ctrl.sqsize = ctrl->ctrl.maxcmd - 1; } - if (ctrl->ctrl.sgls & (1 << 20)) + if (ctrl->ctrl.sgls & NVME_CTRL_SGLS_SAOS) ctrl->use_inline_data = true; if (ctrl->ctrl.queue_count > 1) { @@ -1095,8 +1067,10 @@ static int nvme_rdma_setup_ctrl(struct nvme_rdma_ctrl *ctrl, bool new) * unless we're during creation of a new controller to * avoid races with teardown flow. */ - WARN_ON_ONCE(ctrl->ctrl.state != NVME_CTRL_DELETING && - ctrl->ctrl.state != NVME_CTRL_DELETING_NOIO); + enum nvme_ctrl_state state = nvme_ctrl_state(&ctrl->ctrl); + + WARN_ON_ONCE(state != NVME_CTRL_DELETING && + state != NVME_CTRL_DELETING_NOIO); WARN_ON_ONCE(new); ret = -EINVAL; goto destroy_io; @@ -1116,13 +1090,8 @@ destroy_io: nvme_rdma_free_io_queues(ctrl); } destroy_admin: - nvme_quiesce_admin_queue(&ctrl->ctrl); - blk_sync_queue(ctrl->ctrl.admin_q); - nvme_rdma_stop_queue(&ctrl->queues[0]); - nvme_cancel_admin_tagset(&ctrl->ctrl); - if (new) - nvme_remove_admin_tag_set(&ctrl->ctrl); - nvme_rdma_destroy_admin_queue(ctrl); + nvme_stop_keep_alive(&ctrl->ctrl); + nvme_rdma_teardown_admin_queue(ctrl, new); return ret; } @@ -1130,10 +1099,12 @@ static void nvme_rdma_reconnect_ctrl_work(struct work_struct *work) { struct nvme_rdma_ctrl *ctrl = container_of(to_delayed_work(work), struct nvme_rdma_ctrl, reconnect_work); + int ret; ++ctrl->ctrl.nr_reconnects; - if (nvme_rdma_setup_ctrl(ctrl, false)) + ret = nvme_rdma_setup_ctrl(ctrl, false); + if (ret) goto requeue; dev_info(ctrl->ctrl.device, "Successfully reconnected (%d attempts)\n", @@ -1144,9 +1115,9 @@ static void nvme_rdma_reconnect_ctrl_work(struct work_struct *work) return; requeue: - dev_info(ctrl->ctrl.device, "Failed reconnect attempt %d\n", - ctrl->ctrl.nr_reconnects); - nvme_rdma_reconnect_or_remove(ctrl); + dev_info(ctrl->ctrl.device, "Failed reconnect attempt %d/%d\n", + ctrl->ctrl.nr_reconnects, ctrl->ctrl.opts->max_reconnects); + nvme_rdma_reconnect_or_remove(ctrl, ret); } static void nvme_rdma_error_recovery_work(struct work_struct *work) @@ -1164,12 +1135,14 @@ static void nvme_rdma_error_recovery_work(struct work_struct *work) if (!nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_CONNECTING)) { /* state change failure is ok if we started ctrl delete */ - WARN_ON_ONCE(ctrl->ctrl.state != NVME_CTRL_DELETING && - ctrl->ctrl.state != NVME_CTRL_DELETING_NOIO); + enum nvme_ctrl_state state = nvme_ctrl_state(&ctrl->ctrl); + + WARN_ON_ONCE(state != NVME_CTRL_DELETING && + state != NVME_CTRL_DELETING_NOIO); return; } - nvme_rdma_reconnect_or_remove(ctrl); + nvme_rdma_reconnect_or_remove(ctrl, 0); } static void nvme_rdma_error_recovery(struct nvme_rdma_ctrl *ctrl) @@ -1197,7 +1170,7 @@ static void nvme_rdma_wr_error(struct ib_cq *cq, struct ib_wc *wc, struct nvme_rdma_queue *queue = wc->qp->qp_context; struct nvme_rdma_ctrl *ctrl = queue->ctrl; - if (ctrl->ctrl.state == NVME_CTRL_LIVE) + if (nvme_ctrl_state(&ctrl->ctrl) == NVME_CTRL_LIVE) dev_info(ctrl->ctrl.device, "%s for CQE 0x%p failed with status %s (%d)\n", op, wc->wr_cqe, @@ -1384,8 +1357,8 @@ static void nvme_rdma_set_sig_domain(struct blk_integrity *bi, if (control & NVME_RW_PRINFO_PRCHK_REF) domain->sig.dif.ref_remap = true; - domain->sig.dif.app_tag = le16_to_cpu(cmd->rw.apptag); - domain->sig.dif.apptag_check_mask = le16_to_cpu(cmd->rw.appmask); + domain->sig.dif.app_tag = le16_to_cpu(cmd->rw.lbat); + domain->sig.dif.apptag_check_mask = le16_to_cpu(cmd->rw.lbatm); domain->sig.dif.app_escape = true; if (pi_type == NVME_NS_DPS_PI_TYPE3) domain->sig.dif.ref_escape = true; @@ -1440,6 +1413,8 @@ static int nvme_rdma_map_sg_pi(struct nvme_rdma_queue *queue, struct nvme_ns *ns = rq->q->queuedata; struct bio *bio = rq->bio; struct nvme_keyed_sgl_desc *sg = &c->common.dptr.ksgl; + struct blk_integrity *bi = blk_get_integrity(bio->bi_bdev->bd_disk); + u32 xfer_len; int nr; req->mr = ib_mr_pool_get(queue->qp, &queue->qp->sig_mrs); @@ -1452,8 +1427,7 @@ static int nvme_rdma_map_sg_pi(struct nvme_rdma_queue *queue, if (unlikely(nr)) goto mr_put; - nvme_rdma_set_sig_attrs(blk_get_integrity(bio->bi_bdev->bd_disk), c, - req->mr->sig_attrs, ns->pi_type); + nvme_rdma_set_sig_attrs(bi, c, req->mr->sig_attrs, ns->head->pi_type); nvme_rdma_set_prot_checks(c, &req->mr->sig_attrs->check_mask); ib_update_fast_reg_key(req->mr, ib_inc_rkey(req->mr->rkey)); @@ -1471,7 +1445,11 @@ static int nvme_rdma_map_sg_pi(struct nvme_rdma_queue *queue, IB_ACCESS_REMOTE_WRITE; sg->addr = cpu_to_le64(req->mr->iova); - put_unaligned_le24(req->mr->length, sg->length); + xfer_len = req->mr->length; + /* Check if PI is added by the HW */ + if (!pi_count) + xfer_len += (xfer_len >> bi->interval_exp) * ns->head->pi_size; + put_unaligned_le24(xfer_len, sg->length); put_unaligned_le32(req->mr->rkey, sg->key); sg->type = NVME_KEY_SGL_FMT_DATA_DESC << 4; @@ -1498,8 +1476,7 @@ static int nvme_rdma_dma_map_req(struct ib_device *ibdev, struct request *rq, if (ret) return -ENOMEM; - req->data_sgl.nents = blk_rq_map_sg(rq->q, rq, - req->data_sgl.sg_table.sgl); + req->data_sgl.nents = blk_rq_map_sg(rq, req->data_sgl.sg_table.sgl); *count = ib_dma_map_sg(ibdev, req->data_sgl.sg_table.sgl, req->data_sgl.nents, rq_dma_dir(rq)); @@ -1512,7 +1489,7 @@ static int nvme_rdma_dma_map_req(struct ib_device *ibdev, struct request *rq, req->metadata_sgl->sg_table.sgl = (struct scatterlist *)(req->metadata_sgl + 1); ret = sg_alloc_table_chained(&req->metadata_sgl->sg_table, - blk_rq_count_integrity_sg(rq->q, rq->bio), + rq->nr_integrity_segments, req->metadata_sgl->sg_table.sgl, NVME_INLINE_METADATA_SG_CNT); if (unlikely(ret)) { @@ -1520,8 +1497,8 @@ static int nvme_rdma_dma_map_req(struct ib_device *ibdev, struct request *rq, goto out_unmap_sg; } - req->metadata_sgl->nents = blk_rq_map_integrity_sg(rq->q, - rq->bio, req->metadata_sgl->sg_table.sgl); + req->metadata_sgl->nents = blk_rq_map_integrity_sg(rq, + req->metadata_sgl->sg_table.sgl); *pi_count = ib_dma_map_sg(ibdev, req->metadata_sgl->sg_table.sgl, req->metadata_sgl->nents, @@ -1892,6 +1869,8 @@ static int nvme_rdma_route_resolved(struct nvme_rdma_queue *queue) */ priv.hrqsize = cpu_to_le16(queue->queue_size); priv.hsqsize = cpu_to_le16(queue->ctrl->ctrl.sqsize); + /* cntlid should only be set when creating an I/O queue */ + priv.cntlid = cpu_to_le16(ctrl->ctrl.cntlid); } ret = rdma_connect_locked(queue->cm_id, ¶m); @@ -1976,11 +1955,15 @@ static enum blk_eh_timer_return nvme_rdma_timeout(struct request *rq) struct nvme_rdma_request *req = blk_mq_rq_to_pdu(rq); struct nvme_rdma_queue *queue = req->queue; struct nvme_rdma_ctrl *ctrl = queue->ctrl; + struct nvme_command *cmd = req->req.cmd; + int qid = nvme_rdma_queue_idx(queue); - dev_warn(ctrl->ctrl.device, "I/O %d QID %d timeout\n", - rq->tag, nvme_rdma_queue_idx(queue)); + dev_warn(ctrl->ctrl.device, + "I/O tag %d (%04x) opcode %#x (%s) QID %d timeout\n", + rq->tag, nvme_cid(rq), cmd->common.opcode, + nvme_fabrics_opcode_str(qid, cmd), qid); - if (ctrl->ctrl.state != NVME_CTRL_LIVE) { + if (nvme_ctrl_state(&ctrl->ctrl) != NVME_CTRL_LIVE) { /* * If we are resetting, connecting or deleting we should * complete immediately because we may block controller @@ -2047,7 +2030,7 @@ static blk_status_t nvme_rdma_queue_rq(struct blk_mq_hw_ctx *hctx, queue->pi_support && (c->common.opcode == nvme_cmd_write || c->common.opcode == nvme_cmd_read) && - nvme_ns_has_pi(ns)) + nvme_ns_has_pi(ns->head)) req->use_sig_mr = true; else req->use_sig_mr = false; @@ -2143,46 +2126,8 @@ static void nvme_rdma_complete_rq(struct request *rq) static void nvme_rdma_map_queues(struct blk_mq_tag_set *set) { struct nvme_rdma_ctrl *ctrl = to_rdma_ctrl(set->driver_data); - struct nvmf_ctrl_options *opts = ctrl->ctrl.opts; - - if (opts->nr_write_queues && ctrl->io_queues[HCTX_TYPE_READ]) { - /* separate read/write queues */ - set->map[HCTX_TYPE_DEFAULT].nr_queues = - ctrl->io_queues[HCTX_TYPE_DEFAULT]; - set->map[HCTX_TYPE_DEFAULT].queue_offset = 0; - set->map[HCTX_TYPE_READ].nr_queues = - ctrl->io_queues[HCTX_TYPE_READ]; - set->map[HCTX_TYPE_READ].queue_offset = - ctrl->io_queues[HCTX_TYPE_DEFAULT]; - } else { - /* shared read/write queues */ - set->map[HCTX_TYPE_DEFAULT].nr_queues = - ctrl->io_queues[HCTX_TYPE_DEFAULT]; - set->map[HCTX_TYPE_DEFAULT].queue_offset = 0; - set->map[HCTX_TYPE_READ].nr_queues = - ctrl->io_queues[HCTX_TYPE_DEFAULT]; - set->map[HCTX_TYPE_READ].queue_offset = 0; - } - blk_mq_rdma_map_queues(&set->map[HCTX_TYPE_DEFAULT], - ctrl->device->dev, 0); - blk_mq_rdma_map_queues(&set->map[HCTX_TYPE_READ], - ctrl->device->dev, 0); - - if (opts->nr_poll_queues && ctrl->io_queues[HCTX_TYPE_POLL]) { - /* map dedicated poll queues only if we have queues left */ - set->map[HCTX_TYPE_POLL].nr_queues = - ctrl->io_queues[HCTX_TYPE_POLL]; - set->map[HCTX_TYPE_POLL].queue_offset = - ctrl->io_queues[HCTX_TYPE_DEFAULT] + - ctrl->io_queues[HCTX_TYPE_READ]; - blk_mq_map_queues(&set->map[HCTX_TYPE_POLL]); - } - dev_info(ctrl->ctrl.device, - "mapped %d/%d/%d default/read/poll queues.\n", - ctrl->io_queues[HCTX_TYPE_DEFAULT], - ctrl->io_queues[HCTX_TYPE_READ], - ctrl->io_queues[HCTX_TYPE_POLL]); + nvmf_map_queues(set, &ctrl->ctrl, ctrl->io_queues); } static const struct blk_mq_ops nvme_rdma_mq_ops = { @@ -2222,6 +2167,7 @@ static void nvme_rdma_reset_ctrl_work(struct work_struct *work) { struct nvme_rdma_ctrl *ctrl = container_of(work, struct nvme_rdma_ctrl, ctrl.reset_work); + int ret; nvme_stop_ctrl(&ctrl->ctrl); nvme_rdma_shutdown_ctrl(ctrl, false); @@ -2232,14 +2178,15 @@ static void nvme_rdma_reset_ctrl_work(struct work_struct *work) return; } - if (nvme_rdma_setup_ctrl(ctrl, false)) + ret = nvme_rdma_setup_ctrl(ctrl, false); + if (ret) goto out_fail; return; out_fail: ++ctrl->ctrl.nr_reconnects; - nvme_rdma_reconnect_or_remove(ctrl); + nvme_rdma_reconnect_or_remove(ctrl, ret); } static const struct nvme_ctrl_ops nvme_rdma_ctrl_ops = { @@ -2249,11 +2196,13 @@ static const struct nvme_ctrl_ops nvme_rdma_ctrl_ops = { .reg_read32 = nvmf_reg_read32, .reg_read64 = nvmf_reg_read64, .reg_write32 = nvmf_reg_write32, + .subsystem_reset = nvmf_subsystem_reset, .free_ctrl = nvme_rdma_free_ctrl, .submit_async_event = nvme_rdma_submit_async_event, .delete_ctrl = nvme_rdma_delete_ctrl, .get_address = nvmf_get_address, .stop_ctrl = nvme_rdma_stop_ctrl, + .get_virt_boundary = nvme_get_virt_boundary, }; /* @@ -2285,12 +2234,11 @@ nvme_rdma_existing_controller(struct nvmf_ctrl_options *opts) return found; } -static struct nvme_ctrl *nvme_rdma_create_ctrl(struct device *dev, +static struct nvme_rdma_ctrl *nvme_rdma_alloc_ctrl(struct device *dev, struct nvmf_ctrl_options *opts) { struct nvme_rdma_ctrl *ctrl; int ret; - bool changed; ctrl = kzalloc(sizeof(*ctrl), GFP_KERNEL); if (!ctrl) @@ -2352,6 +2300,30 @@ static struct nvme_ctrl *nvme_rdma_create_ctrl(struct device *dev, if (ret) goto out_kfree_queues; + return ctrl; + +out_kfree_queues: + kfree(ctrl->queues); +out_free_ctrl: + kfree(ctrl); + return ERR_PTR(ret); +} + +static struct nvme_ctrl *nvme_rdma_create_ctrl(struct device *dev, + struct nvmf_ctrl_options *opts) +{ + struct nvme_rdma_ctrl *ctrl; + bool changed; + int ret; + + ctrl = nvme_rdma_alloc_ctrl(dev, opts); + if (IS_ERR(ctrl)) + return ERR_CAST(ctrl); + + ret = nvme_add_ctrl(&ctrl->ctrl); + if (ret) + goto out_put_ctrl; + changed = nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_CONNECTING); WARN_ON_ONCE(!changed); @@ -2359,8 +2331,8 @@ static struct nvme_ctrl *nvme_rdma_create_ctrl(struct device *dev, if (ret) goto out_uninit_ctrl; - dev_info(ctrl->ctrl.device, "new ctrl: NQN \"%s\", addr %pISpcs\n", - nvmf_ctrl_subsysnqn(&ctrl->ctrl), &ctrl->addr); + dev_info(ctrl->ctrl.device, "new ctrl: NQN \"%s\", addr %pISpcs, hostnqn: %s\n", + nvmf_ctrl_subsysnqn(&ctrl->ctrl), &ctrl->addr, opts->host->nqn); mutex_lock(&nvme_rdma_ctrl_mutex); list_add_tail(&ctrl->list, &nvme_rdma_ctrl_list); @@ -2370,15 +2342,11 @@ static struct nvme_ctrl *nvme_rdma_create_ctrl(struct device *dev, out_uninit_ctrl: nvme_uninit_ctrl(&ctrl->ctrl); +out_put_ctrl: nvme_put_ctrl(&ctrl->ctrl); if (ret > 0) ret = -EIO; return ERR_PTR(ret); -out_kfree_queues: - kfree(ctrl->queues); -out_free_ctrl: - kfree(ctrl); - return ERR_PTR(ret); } static struct nvmf_transport_ops nvme_rdma_transport = { @@ -2463,4 +2431,5 @@ static void __exit nvme_rdma_cleanup_module(void) module_init(nvme_rdma_init_module); module_exit(nvme_rdma_cleanup_module); +MODULE_DESCRIPTION("NVMe host RDMA transport driver"); MODULE_LICENSE("GPL v2"); diff --git a/drivers/nvme/host/sysfs.c b/drivers/nvme/host/sysfs.c new file mode 100644 index 000000000000..29430949ce2f --- /dev/null +++ b/drivers/nvme/host/sysfs.c @@ -0,0 +1,931 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Sysfs interface for the NVMe core driver. + * + * Copyright (c) 2011-2014, Intel Corporation. + */ + +#include <linux/nvme-auth.h> + +#include "nvme.h" +#include "fabrics.h" + +static ssize_t nvme_sysfs_reset(struct device *dev, + struct device_attribute *attr, const char *buf, + size_t count) +{ + struct nvme_ctrl *ctrl = dev_get_drvdata(dev); + int ret; + + ret = nvme_reset_ctrl_sync(ctrl); + if (ret < 0) + return ret; + return count; +} +static DEVICE_ATTR(reset_controller, S_IWUSR, NULL, nvme_sysfs_reset); + +static ssize_t nvme_sysfs_rescan(struct device *dev, + struct device_attribute *attr, const char *buf, + size_t count) +{ + struct nvme_ctrl *ctrl = dev_get_drvdata(dev); + + nvme_queue_scan(ctrl); + return count; +} +static DEVICE_ATTR(rescan_controller, S_IWUSR, NULL, nvme_sysfs_rescan); + +static ssize_t nvme_adm_passthru_err_log_enabled_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct nvme_ctrl *ctrl = dev_get_drvdata(dev); + + return sysfs_emit(buf, + ctrl->passthru_err_log_enabled ? "on\n" : "off\n"); +} + +static ssize_t nvme_adm_passthru_err_log_enabled_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t count) +{ + struct nvme_ctrl *ctrl = dev_get_drvdata(dev); + bool passthru_err_log_enabled; + int err; + + err = kstrtobool(buf, &passthru_err_log_enabled); + if (err) + return -EINVAL; + + ctrl->passthru_err_log_enabled = passthru_err_log_enabled; + + return count; +} + +static inline struct nvme_ns_head *dev_to_ns_head(struct device *dev) +{ + struct gendisk *disk = dev_to_disk(dev); + + if (nvme_disk_is_ns_head(disk)) + return disk->private_data; + return nvme_get_ns_from_dev(dev)->head; +} + +static ssize_t nvme_io_passthru_err_log_enabled_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct nvme_ns_head *head = dev_to_ns_head(dev); + + return sysfs_emit(buf, head->passthru_err_log_enabled ? "on\n" : "off\n"); +} + +static ssize_t nvme_io_passthru_err_log_enabled_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t count) +{ + struct nvme_ns_head *head = dev_to_ns_head(dev); + bool passthru_err_log_enabled; + int err; + + err = kstrtobool(buf, &passthru_err_log_enabled); + if (err) + return -EINVAL; + head->passthru_err_log_enabled = passthru_err_log_enabled; + + return count; +} + +static struct device_attribute dev_attr_adm_passthru_err_log_enabled = \ + __ATTR(passthru_err_log_enabled, S_IRUGO | S_IWUSR, \ + nvme_adm_passthru_err_log_enabled_show, nvme_adm_passthru_err_log_enabled_store); + +static struct device_attribute dev_attr_io_passthru_err_log_enabled = \ + __ATTR(passthru_err_log_enabled, S_IRUGO | S_IWUSR, \ + nvme_io_passthru_err_log_enabled_show, nvme_io_passthru_err_log_enabled_store); + +static ssize_t wwid_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct nvme_ns_head *head = dev_to_ns_head(dev); + struct nvme_ns_ids *ids = &head->ids; + struct nvme_subsystem *subsys = head->subsys; + int serial_len = sizeof(subsys->serial); + int model_len = sizeof(subsys->model); + + if (!uuid_is_null(&ids->uuid)) + return sysfs_emit(buf, "uuid.%pU\n", &ids->uuid); + + if (memchr_inv(ids->nguid, 0, sizeof(ids->nguid))) + return sysfs_emit(buf, "eui.%16phN\n", ids->nguid); + + if (memchr_inv(ids->eui64, 0, sizeof(ids->eui64))) + return sysfs_emit(buf, "eui.%8phN\n", ids->eui64); + + while (serial_len > 0 && (subsys->serial[serial_len - 1] == ' ' || + subsys->serial[serial_len - 1] == '\0')) + serial_len--; + while (model_len > 0 && (subsys->model[model_len - 1] == ' ' || + subsys->model[model_len - 1] == '\0')) + model_len--; + + return sysfs_emit(buf, "nvme.%04x-%*phN-%*phN-%08x\n", subsys->vendor_id, + serial_len, subsys->serial, model_len, subsys->model, + head->ns_id); +} +static DEVICE_ATTR_RO(wwid); + +static ssize_t nguid_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + return sysfs_emit(buf, "%pU\n", dev_to_ns_head(dev)->ids.nguid); +} +static DEVICE_ATTR_RO(nguid); + +static ssize_t uuid_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct nvme_ns_ids *ids = &dev_to_ns_head(dev)->ids; + + /* For backward compatibility expose the NGUID to userspace if + * we have no UUID set + */ + if (uuid_is_null(&ids->uuid)) { + dev_warn_once(dev, + "No UUID available providing old NGUID\n"); + return sysfs_emit(buf, "%pU\n", ids->nguid); + } + return sysfs_emit(buf, "%pU\n", &ids->uuid); +} +static DEVICE_ATTR_RO(uuid); + +static ssize_t eui_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + return sysfs_emit(buf, "%8ph\n", dev_to_ns_head(dev)->ids.eui64); +} +static DEVICE_ATTR_RO(eui); + +static ssize_t nsid_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + return sysfs_emit(buf, "%d\n", dev_to_ns_head(dev)->ns_id); +} +static DEVICE_ATTR_RO(nsid); + +static ssize_t csi_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + return sysfs_emit(buf, "%u\n", dev_to_ns_head(dev)->ids.csi); +} +static DEVICE_ATTR_RO(csi); + +static ssize_t metadata_bytes_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + return sysfs_emit(buf, "%u\n", dev_to_ns_head(dev)->ms); +} +static DEVICE_ATTR_RO(metadata_bytes); + +static int ns_head_update_nuse(struct nvme_ns_head *head) +{ + struct nvme_id_ns *id; + struct nvme_ns *ns; + int srcu_idx, ret = -EWOULDBLOCK; + + /* Avoid issuing commands too often by rate limiting the update */ + if (!__ratelimit(&head->rs_nuse)) + return 0; + + srcu_idx = srcu_read_lock(&head->srcu); + ns = nvme_find_path(head); + if (!ns) + goto out_unlock; + + ret = nvme_identify_ns(ns->ctrl, head->ns_id, &id); + if (ret) + goto out_unlock; + + head->nuse = le64_to_cpu(id->nuse); + kfree(id); + +out_unlock: + srcu_read_unlock(&head->srcu, srcu_idx); + return ret; +} + +static int ns_update_nuse(struct nvme_ns *ns) +{ + struct nvme_id_ns *id; + int ret; + + /* Avoid issuing commands too often by rate limiting the update. */ + if (!__ratelimit(&ns->head->rs_nuse)) + return 0; + + ret = nvme_identify_ns(ns->ctrl, ns->head->ns_id, &id); + if (ret) + return ret; + + ns->head->nuse = le64_to_cpu(id->nuse); + kfree(id); + return 0; +} + +static ssize_t nuse_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct nvme_ns_head *head = dev_to_ns_head(dev); + struct gendisk *disk = dev_to_disk(dev); + int ret; + + if (nvme_disk_is_ns_head(disk)) + ret = ns_head_update_nuse(head); + else + ret = ns_update_nuse(disk->private_data); + if (ret) + return ret; + + return sysfs_emit(buf, "%llu\n", head->nuse); +} +static DEVICE_ATTR_RO(nuse); + +static struct attribute *nvme_ns_attrs[] = { + &dev_attr_wwid.attr, + &dev_attr_uuid.attr, + &dev_attr_nguid.attr, + &dev_attr_eui.attr, + &dev_attr_csi.attr, + &dev_attr_nsid.attr, + &dev_attr_metadata_bytes.attr, + &dev_attr_nuse.attr, +#ifdef CONFIG_NVME_MULTIPATH + &dev_attr_ana_grpid.attr, + &dev_attr_ana_state.attr, + &dev_attr_queue_depth.attr, + &dev_attr_numa_nodes.attr, + &dev_attr_delayed_removal_secs.attr, +#endif + &dev_attr_io_passthru_err_log_enabled.attr, + NULL, +}; + +static umode_t nvme_ns_attrs_are_visible(struct kobject *kobj, + struct attribute *a, int n) +{ + struct device *dev = container_of(kobj, struct device, kobj); + struct nvme_ns_ids *ids = &dev_to_ns_head(dev)->ids; + + if (a == &dev_attr_uuid.attr) { + if (uuid_is_null(&ids->uuid) && + !memchr_inv(ids->nguid, 0, sizeof(ids->nguid))) + return 0; + } + if (a == &dev_attr_nguid.attr) { + if (!memchr_inv(ids->nguid, 0, sizeof(ids->nguid))) + return 0; + } + if (a == &dev_attr_eui.attr) { + if (!memchr_inv(ids->eui64, 0, sizeof(ids->eui64))) + return 0; + } +#ifdef CONFIG_NVME_MULTIPATH + if (a == &dev_attr_ana_grpid.attr || a == &dev_attr_ana_state.attr) { + /* per-path attr */ + if (nvme_disk_is_ns_head(dev_to_disk(dev))) + return 0; + if (!nvme_ctrl_use_ana(nvme_get_ns_from_dev(dev)->ctrl)) + return 0; + } + if (a == &dev_attr_queue_depth.attr || a == &dev_attr_numa_nodes.attr) { + if (nvme_disk_is_ns_head(dev_to_disk(dev))) + return 0; + } + if (a == &dev_attr_delayed_removal_secs.attr) { + struct gendisk *disk = dev_to_disk(dev); + + if (!nvme_disk_is_ns_head(disk)) + return 0; + } +#endif + return a->mode; +} + +static const struct attribute_group nvme_ns_attr_group = { + .attrs = nvme_ns_attrs, + .is_visible = nvme_ns_attrs_are_visible, +}; + +#ifdef CONFIG_NVME_MULTIPATH +/* + * NOTE: The dummy attribute does not appear in sysfs. It exists solely to allow + * control over the visibility of the multipath sysfs node. Without at least one + * attribute defined in nvme_ns_mpath_attrs[], the sysfs implementation does not + * invoke the multipath_sysfs_group_visible() method. As a result, we would not + * be able to control the visibility of the multipath sysfs node. + */ +static struct attribute dummy_attr = { + .name = "dummy", +}; + +static struct attribute *nvme_ns_mpath_attrs[] = { + &dummy_attr, + NULL, +}; + +static bool multipath_sysfs_group_visible(struct kobject *kobj) +{ + struct device *dev = container_of(kobj, struct device, kobj); + + return nvme_disk_is_ns_head(dev_to_disk(dev)); +} + +static bool multipath_sysfs_attr_visible(struct kobject *kobj, + struct attribute *attr, int n) +{ + return false; +} + +DEFINE_SYSFS_GROUP_VISIBLE(multipath_sysfs) + +const struct attribute_group nvme_ns_mpath_attr_group = { + .name = "multipath", + .attrs = nvme_ns_mpath_attrs, + .is_visible = SYSFS_GROUP_VISIBLE(multipath_sysfs), +}; +#endif + +const struct attribute_group *nvme_ns_attr_groups[] = { + &nvme_ns_attr_group, +#ifdef CONFIG_NVME_MULTIPATH + &nvme_ns_mpath_attr_group, +#endif + NULL, +}; + +#define nvme_show_str_function(field) \ +static ssize_t field##_show(struct device *dev, \ + struct device_attribute *attr, char *buf) \ +{ \ + struct nvme_ctrl *ctrl = dev_get_drvdata(dev); \ + return sysfs_emit(buf, "%.*s\n", \ + (int)sizeof(ctrl->subsys->field), ctrl->subsys->field); \ +} \ +static DEVICE_ATTR(field, S_IRUGO, field##_show, NULL); + +nvme_show_str_function(model); +nvme_show_str_function(serial); +nvme_show_str_function(firmware_rev); + +#define nvme_show_int_function(field) \ +static ssize_t field##_show(struct device *dev, \ + struct device_attribute *attr, char *buf) \ +{ \ + struct nvme_ctrl *ctrl = dev_get_drvdata(dev); \ + return sysfs_emit(buf, "%d\n", ctrl->field); \ +} \ +static DEVICE_ATTR(field, S_IRUGO, field##_show, NULL); + +nvme_show_int_function(cntlid); +nvme_show_int_function(numa_node); +nvme_show_int_function(queue_count); +nvme_show_int_function(sqsize); +nvme_show_int_function(kato); + +static ssize_t nvme_sysfs_delete(struct device *dev, + struct device_attribute *attr, const char *buf, + size_t count) +{ + struct nvme_ctrl *ctrl = dev_get_drvdata(dev); + + if (!test_bit(NVME_CTRL_STARTED_ONCE, &ctrl->flags)) + return -EBUSY; + + if (device_remove_file_self(dev, attr)) + nvme_delete_ctrl_sync(ctrl); + return count; +} +static DEVICE_ATTR(delete_controller, S_IWUSR, NULL, nvme_sysfs_delete); + +static ssize_t nvme_sysfs_show_transport(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + struct nvme_ctrl *ctrl = dev_get_drvdata(dev); + + return sysfs_emit(buf, "%s\n", ctrl->ops->name); +} +static DEVICE_ATTR(transport, S_IRUGO, nvme_sysfs_show_transport, NULL); + +static ssize_t nvme_sysfs_show_state(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + struct nvme_ctrl *ctrl = dev_get_drvdata(dev); + unsigned state = (unsigned)nvme_ctrl_state(ctrl); + static const char *const state_name[] = { + [NVME_CTRL_NEW] = "new", + [NVME_CTRL_LIVE] = "live", + [NVME_CTRL_RESETTING] = "resetting", + [NVME_CTRL_CONNECTING] = "connecting", + [NVME_CTRL_DELETING] = "deleting", + [NVME_CTRL_DELETING_NOIO]= "deleting (no IO)", + [NVME_CTRL_DEAD] = "dead", + }; + + if (state < ARRAY_SIZE(state_name) && state_name[state]) + return sysfs_emit(buf, "%s\n", state_name[state]); + + return sysfs_emit(buf, "unknown state\n"); +} + +static DEVICE_ATTR(state, S_IRUGO, nvme_sysfs_show_state, NULL); + +static ssize_t nvme_sysfs_show_subsysnqn(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + struct nvme_ctrl *ctrl = dev_get_drvdata(dev); + + return sysfs_emit(buf, "%s\n", ctrl->subsys->subnqn); +} +static DEVICE_ATTR(subsysnqn, S_IRUGO, nvme_sysfs_show_subsysnqn, NULL); + +static ssize_t nvme_sysfs_show_hostnqn(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + struct nvme_ctrl *ctrl = dev_get_drvdata(dev); + + return sysfs_emit(buf, "%s\n", ctrl->opts->host->nqn); +} +static DEVICE_ATTR(hostnqn, S_IRUGO, nvme_sysfs_show_hostnqn, NULL); + +static ssize_t nvme_sysfs_show_hostid(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + struct nvme_ctrl *ctrl = dev_get_drvdata(dev); + + return sysfs_emit(buf, "%pU\n", &ctrl->opts->host->id); +} +static DEVICE_ATTR(hostid, S_IRUGO, nvme_sysfs_show_hostid, NULL); + +static ssize_t nvme_sysfs_show_address(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + struct nvme_ctrl *ctrl = dev_get_drvdata(dev); + + return ctrl->ops->get_address(ctrl, buf, PAGE_SIZE); +} +static DEVICE_ATTR(address, S_IRUGO, nvme_sysfs_show_address, NULL); + +static ssize_t nvme_ctrl_loss_tmo_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct nvme_ctrl *ctrl = dev_get_drvdata(dev); + struct nvmf_ctrl_options *opts = ctrl->opts; + + if (ctrl->opts->max_reconnects == -1) + return sysfs_emit(buf, "off\n"); + return sysfs_emit(buf, "%d\n", + opts->max_reconnects * opts->reconnect_delay); +} + +static ssize_t nvme_ctrl_loss_tmo_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t count) +{ + struct nvme_ctrl *ctrl = dev_get_drvdata(dev); + struct nvmf_ctrl_options *opts = ctrl->opts; + int ctrl_loss_tmo, err; + + err = kstrtoint(buf, 10, &ctrl_loss_tmo); + if (err) + return -EINVAL; + + if (ctrl_loss_tmo < 0) + opts->max_reconnects = -1; + else + opts->max_reconnects = DIV_ROUND_UP(ctrl_loss_tmo, + opts->reconnect_delay); + return count; +} +static DEVICE_ATTR(ctrl_loss_tmo, S_IRUGO | S_IWUSR, + nvme_ctrl_loss_tmo_show, nvme_ctrl_loss_tmo_store); + +static ssize_t nvme_ctrl_reconnect_delay_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct nvme_ctrl *ctrl = dev_get_drvdata(dev); + + if (ctrl->opts->reconnect_delay == -1) + return sysfs_emit(buf, "off\n"); + return sysfs_emit(buf, "%d\n", ctrl->opts->reconnect_delay); +} + +static ssize_t nvme_ctrl_reconnect_delay_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t count) +{ + struct nvme_ctrl *ctrl = dev_get_drvdata(dev); + unsigned int v; + int err; + + err = kstrtou32(buf, 10, &v); + if (err) + return err; + + ctrl->opts->reconnect_delay = v; + return count; +} +static DEVICE_ATTR(reconnect_delay, S_IRUGO | S_IWUSR, + nvme_ctrl_reconnect_delay_show, nvme_ctrl_reconnect_delay_store); + +static ssize_t nvme_ctrl_fast_io_fail_tmo_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct nvme_ctrl *ctrl = dev_get_drvdata(dev); + + if (ctrl->opts->fast_io_fail_tmo == -1) + return sysfs_emit(buf, "off\n"); + return sysfs_emit(buf, "%d\n", ctrl->opts->fast_io_fail_tmo); +} + +static ssize_t nvme_ctrl_fast_io_fail_tmo_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t count) +{ + struct nvme_ctrl *ctrl = dev_get_drvdata(dev); + struct nvmf_ctrl_options *opts = ctrl->opts; + int fast_io_fail_tmo, err; + + err = kstrtoint(buf, 10, &fast_io_fail_tmo); + if (err) + return -EINVAL; + + if (fast_io_fail_tmo < 0) + opts->fast_io_fail_tmo = -1; + else + opts->fast_io_fail_tmo = fast_io_fail_tmo; + return count; +} +static DEVICE_ATTR(fast_io_fail_tmo, S_IRUGO | S_IWUSR, + nvme_ctrl_fast_io_fail_tmo_show, nvme_ctrl_fast_io_fail_tmo_store); + +static ssize_t cntrltype_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + static const char * const type[] = { + [NVME_CTRL_IO] = "io\n", + [NVME_CTRL_DISC] = "discovery\n", + [NVME_CTRL_ADMIN] = "admin\n", + }; + struct nvme_ctrl *ctrl = dev_get_drvdata(dev); + + if (ctrl->cntrltype > NVME_CTRL_ADMIN || !type[ctrl->cntrltype]) + return sysfs_emit(buf, "reserved\n"); + + return sysfs_emit(buf, type[ctrl->cntrltype]); +} +static DEVICE_ATTR_RO(cntrltype); + +static ssize_t dctype_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + static const char * const type[] = { + [NVME_DCTYPE_NOT_REPORTED] = "none\n", + [NVME_DCTYPE_DDC] = "ddc\n", + [NVME_DCTYPE_CDC] = "cdc\n", + }; + struct nvme_ctrl *ctrl = dev_get_drvdata(dev); + + if (ctrl->dctype > NVME_DCTYPE_CDC || !type[ctrl->dctype]) + return sysfs_emit(buf, "reserved\n"); + + return sysfs_emit(buf, type[ctrl->dctype]); +} +static DEVICE_ATTR_RO(dctype); + +#ifdef CONFIG_NVME_HOST_AUTH +static ssize_t nvme_ctrl_dhchap_secret_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct nvme_ctrl *ctrl = dev_get_drvdata(dev); + struct nvmf_ctrl_options *opts = ctrl->opts; + + if (!opts->dhchap_secret) + return sysfs_emit(buf, "none\n"); + return sysfs_emit(buf, "%s\n", opts->dhchap_secret); +} + +static ssize_t nvme_ctrl_dhchap_secret_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t count) +{ + struct nvme_ctrl *ctrl = dev_get_drvdata(dev); + struct nvmf_ctrl_options *opts = ctrl->opts; + char *dhchap_secret; + + if (!ctrl->opts->dhchap_secret) + return -EINVAL; + if (count < 7) + return -EINVAL; + if (memcmp(buf, "DHHC-1:", 7)) + return -EINVAL; + + dhchap_secret = kzalloc(count + 1, GFP_KERNEL); + if (!dhchap_secret) + return -ENOMEM; + memcpy(dhchap_secret, buf, count); + nvme_auth_stop(ctrl); + if (strcmp(dhchap_secret, opts->dhchap_secret)) { + struct nvme_dhchap_key *key, *host_key; + int ret; + + ret = nvme_auth_generate_key(dhchap_secret, &key); + if (ret) { + kfree(dhchap_secret); + return ret; + } + kfree(opts->dhchap_secret); + opts->dhchap_secret = dhchap_secret; + host_key = ctrl->host_key; + mutex_lock(&ctrl->dhchap_auth_mutex); + ctrl->host_key = key; + mutex_unlock(&ctrl->dhchap_auth_mutex); + nvme_auth_free_key(host_key); + } else + kfree(dhchap_secret); + /* Start re-authentication */ + dev_info(ctrl->device, "re-authenticating controller\n"); + queue_work(nvme_wq, &ctrl->dhchap_auth_work); + + return count; +} + +static DEVICE_ATTR(dhchap_secret, S_IRUGO | S_IWUSR, + nvme_ctrl_dhchap_secret_show, nvme_ctrl_dhchap_secret_store); + +static ssize_t nvme_ctrl_dhchap_ctrl_secret_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct nvme_ctrl *ctrl = dev_get_drvdata(dev); + struct nvmf_ctrl_options *opts = ctrl->opts; + + if (!opts->dhchap_ctrl_secret) + return sysfs_emit(buf, "none\n"); + return sysfs_emit(buf, "%s\n", opts->dhchap_ctrl_secret); +} + +static ssize_t nvme_ctrl_dhchap_ctrl_secret_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t count) +{ + struct nvme_ctrl *ctrl = dev_get_drvdata(dev); + struct nvmf_ctrl_options *opts = ctrl->opts; + char *dhchap_secret; + + if (!ctrl->opts->dhchap_ctrl_secret) + return -EINVAL; + if (count < 7) + return -EINVAL; + if (memcmp(buf, "DHHC-1:", 7)) + return -EINVAL; + + dhchap_secret = kzalloc(count + 1, GFP_KERNEL); + if (!dhchap_secret) + return -ENOMEM; + memcpy(dhchap_secret, buf, count); + nvme_auth_stop(ctrl); + if (strcmp(dhchap_secret, opts->dhchap_ctrl_secret)) { + struct nvme_dhchap_key *key, *ctrl_key; + int ret; + + ret = nvme_auth_generate_key(dhchap_secret, &key); + if (ret) { + kfree(dhchap_secret); + return ret; + } + kfree(opts->dhchap_ctrl_secret); + opts->dhchap_ctrl_secret = dhchap_secret; + ctrl_key = ctrl->ctrl_key; + mutex_lock(&ctrl->dhchap_auth_mutex); + ctrl->ctrl_key = key; + mutex_unlock(&ctrl->dhchap_auth_mutex); + nvme_auth_free_key(ctrl_key); + } else + kfree(dhchap_secret); + /* Start re-authentication */ + dev_info(ctrl->device, "re-authenticating controller\n"); + queue_work(nvme_wq, &ctrl->dhchap_auth_work); + + return count; +} + +static DEVICE_ATTR(dhchap_ctrl_secret, S_IRUGO | S_IWUSR, + nvme_ctrl_dhchap_ctrl_secret_show, nvme_ctrl_dhchap_ctrl_secret_store); +#endif + +static struct attribute *nvme_dev_attrs[] = { + &dev_attr_reset_controller.attr, + &dev_attr_rescan_controller.attr, + &dev_attr_model.attr, + &dev_attr_serial.attr, + &dev_attr_firmware_rev.attr, + &dev_attr_cntlid.attr, + &dev_attr_delete_controller.attr, + &dev_attr_transport.attr, + &dev_attr_subsysnqn.attr, + &dev_attr_address.attr, + &dev_attr_state.attr, + &dev_attr_numa_node.attr, + &dev_attr_queue_count.attr, + &dev_attr_sqsize.attr, + &dev_attr_hostnqn.attr, + &dev_attr_hostid.attr, + &dev_attr_ctrl_loss_tmo.attr, + &dev_attr_reconnect_delay.attr, + &dev_attr_fast_io_fail_tmo.attr, + &dev_attr_kato.attr, + &dev_attr_cntrltype.attr, + &dev_attr_dctype.attr, +#ifdef CONFIG_NVME_HOST_AUTH + &dev_attr_dhchap_secret.attr, + &dev_attr_dhchap_ctrl_secret.attr, +#endif + &dev_attr_adm_passthru_err_log_enabled.attr, + NULL +}; + +static umode_t nvme_dev_attrs_are_visible(struct kobject *kobj, + struct attribute *a, int n) +{ + struct device *dev = container_of(kobj, struct device, kobj); + struct nvme_ctrl *ctrl = dev_get_drvdata(dev); + + if (a == &dev_attr_delete_controller.attr && !ctrl->ops->delete_ctrl) + return 0; + if (a == &dev_attr_address.attr && !ctrl->ops->get_address) + return 0; + if (a == &dev_attr_hostnqn.attr && !ctrl->opts) + return 0; + if (a == &dev_attr_hostid.attr && !ctrl->opts) + return 0; + if (a == &dev_attr_ctrl_loss_tmo.attr && !ctrl->opts) + return 0; + if (a == &dev_attr_reconnect_delay.attr && !ctrl->opts) + return 0; + if (a == &dev_attr_fast_io_fail_tmo.attr && !ctrl->opts) + return 0; +#ifdef CONFIG_NVME_HOST_AUTH + if (a == &dev_attr_dhchap_secret.attr && !ctrl->opts) + return 0; + if (a == &dev_attr_dhchap_ctrl_secret.attr && !ctrl->opts) + return 0; +#endif + + return a->mode; +} + +const struct attribute_group nvme_dev_attrs_group = { + .attrs = nvme_dev_attrs, + .is_visible = nvme_dev_attrs_are_visible, +}; +EXPORT_SYMBOL_GPL(nvme_dev_attrs_group); + +#ifdef CONFIG_NVME_TCP_TLS +static ssize_t tls_key_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct nvme_ctrl *ctrl = dev_get_drvdata(dev); + + if (!ctrl->tls_pskid) + return 0; + return sysfs_emit(buf, "%08x\n", ctrl->tls_pskid); +} +static DEVICE_ATTR_RO(tls_key); + +static ssize_t tls_configured_key_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct nvme_ctrl *ctrl = dev_get_drvdata(dev); + struct key *key = ctrl->opts->tls_key; + + return sysfs_emit(buf, "%08x\n", key_serial(key)); +} +static DEVICE_ATTR_RO(tls_configured_key); + +static ssize_t tls_keyring_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct nvme_ctrl *ctrl = dev_get_drvdata(dev); + struct key *keyring = ctrl->opts->keyring; + + return sysfs_emit(buf, "%s\n", keyring->description); +} +static DEVICE_ATTR_RO(tls_keyring); + +static struct attribute *nvme_tls_attrs[] = { + &dev_attr_tls_key.attr, + &dev_attr_tls_configured_key.attr, + &dev_attr_tls_keyring.attr, + NULL, +}; + +static umode_t nvme_tls_attrs_are_visible(struct kobject *kobj, + struct attribute *a, int n) +{ + struct device *dev = container_of(kobj, struct device, kobj); + struct nvme_ctrl *ctrl = dev_get_drvdata(dev); + + if (!ctrl->opts || strcmp(ctrl->opts->transport, "tcp")) + return 0; + + if (a == &dev_attr_tls_key.attr && + !ctrl->opts->tls && !ctrl->opts->concat) + return 0; + if (a == &dev_attr_tls_configured_key.attr && + (!ctrl->opts->tls_key || ctrl->opts->concat)) + return 0; + if (a == &dev_attr_tls_keyring.attr && + !ctrl->opts->keyring) + return 0; + + return a->mode; +} + +static const struct attribute_group nvme_tls_attrs_group = { + .attrs = nvme_tls_attrs, + .is_visible = nvme_tls_attrs_are_visible, +}; +#endif + +const struct attribute_group *nvme_dev_attr_groups[] = { + &nvme_dev_attrs_group, +#ifdef CONFIG_NVME_TCP_TLS + &nvme_tls_attrs_group, +#endif + NULL, +}; + +#define SUBSYS_ATTR_RO(_name, _mode, _show) \ + struct device_attribute subsys_attr_##_name = \ + __ATTR(_name, _mode, _show, NULL) + +static ssize_t nvme_subsys_show_nqn(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + struct nvme_subsystem *subsys = + container_of(dev, struct nvme_subsystem, dev); + + return sysfs_emit(buf, "%s\n", subsys->subnqn); +} +static SUBSYS_ATTR_RO(subsysnqn, S_IRUGO, nvme_subsys_show_nqn); + +static ssize_t nvme_subsys_show_type(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + struct nvme_subsystem *subsys = + container_of(dev, struct nvme_subsystem, dev); + + switch (subsys->subtype) { + case NVME_NQN_DISC: + return sysfs_emit(buf, "discovery\n"); + case NVME_NQN_NVME: + return sysfs_emit(buf, "nvm\n"); + default: + return sysfs_emit(buf, "reserved\n"); + } +} +static SUBSYS_ATTR_RO(subsystype, S_IRUGO, nvme_subsys_show_type); + +#define nvme_subsys_show_str_function(field) \ +static ssize_t subsys_##field##_show(struct device *dev, \ + struct device_attribute *attr, char *buf) \ +{ \ + struct nvme_subsystem *subsys = \ + container_of(dev, struct nvme_subsystem, dev); \ + return sysfs_emit(buf, "%.*s\n", \ + (int)sizeof(subsys->field), subsys->field); \ +} \ +static SUBSYS_ATTR_RO(field, S_IRUGO, subsys_##field##_show); + +nvme_subsys_show_str_function(model); +nvme_subsys_show_str_function(serial); +nvme_subsys_show_str_function(firmware_rev); + +static struct attribute *nvme_subsys_attrs[] = { + &subsys_attr_model.attr, + &subsys_attr_serial.attr, + &subsys_attr_firmware_rev.attr, + &subsys_attr_subsysnqn.attr, + &subsys_attr_subsystype.attr, +#ifdef CONFIG_NVME_MULTIPATH + &subsys_attr_iopolicy.attr, +#endif + NULL, +}; + +static const struct attribute_group nvme_subsys_attrs_group = { + .attrs = nvme_subsys_attrs, +}; + +const struct attribute_group *nvme_subsys_attrs_groups[] = { + &nvme_subsys_attrs_group, + NULL, +}; diff --git a/drivers/nvme/host/tcp.c b/drivers/nvme/host/tcp.c index 8cedc1ef496c..69cb04406b47 100644 --- a/drivers/nvme/host/tcp.c +++ b/drivers/nvme/host/tcp.c @@ -8,12 +8,17 @@ #include <linux/init.h> #include <linux/slab.h> #include <linux/err.h> +#include <linux/crc32.h> #include <linux/nvme-tcp.h> +#include <linux/nvme-keyring.h> #include <net/sock.h> #include <net/tcp.h> +#include <net/tls.h> +#include <net/tls_prot.h> +#include <net/handshake.h> #include <linux/blk-mq.h> -#include <crypto/hash.h> #include <net/busy_poll.h> +#include <trace/events/sock.h> #include "nvme.h" #include "fabrics.h" @@ -30,6 +35,26 @@ static int so_priority; module_param(so_priority, int, 0644); MODULE_PARM_DESC(so_priority, "nvme tcp socket optimize priority"); +/* + * Use the unbound workqueue for nvme_tcp_wq, then we can set the cpu affinity + * from sysfs. + */ +static bool wq_unbound; +module_param(wq_unbound, bool, 0644); +MODULE_PARM_DESC(wq_unbound, "Use unbound workqueue for nvme-tcp IO context (default false)"); + +/* + * TLS handshake timeout + */ +static int tls_handshake_timeout = 10; +#ifdef CONFIG_NVME_TCP_TLS +module_param(tls_handshake_timeout, int, 0644); +MODULE_PARM_DESC(tls_handshake_timeout, + "nvme TLS handshake timeout in seconds (default 10)"); +#endif + +static atomic_t nvme_tcp_cpu_queues[NR_CPUS]; + #ifdef CONFIG_DEBUG_LOCK_ALLOC /* lockdep can detect a circular dependency of the form * sk_lock -> mmap_lock (page fault) -> fs locks -> sk_lock @@ -103,6 +128,7 @@ enum nvme_tcp_queue_flags { NVME_TCP_Q_ALLOCATED = 0, NVME_TCP_Q_LIVE = 1, NVME_TCP_Q_POLLING = 2, + NVME_TCP_Q_IO_CPU_SET = 3, }; enum nvme_tcp_recv_state { @@ -141,11 +167,13 @@ struct nvme_tcp_queue { bool hdr_digest; bool data_digest; - struct ahash_request *rcv_hash; - struct ahash_request *snd_hash; + bool tls_enabled; + u32 rcv_crc; + u32 snd_crc; __le32 exp_ddgst; __le32 recv_ddgst; - + struct completion tls_complete; + int tls_err; struct page_frag_cache pf_cache; void (*state_change)(struct sock *); @@ -188,6 +216,41 @@ static inline int nvme_tcp_queue_id(struct nvme_tcp_queue *queue) return queue - queue->ctrl->queues; } +static inline bool nvme_tcp_recv_pdu_supported(enum nvme_tcp_pdu_type type) +{ + switch (type) { + case nvme_tcp_c2h_term: + case nvme_tcp_c2h_data: + case nvme_tcp_r2t: + case nvme_tcp_rsp: + return true; + default: + return false; + } +} + +/* + * Check if the queue is TLS encrypted + */ +static inline bool nvme_tcp_queue_tls(struct nvme_tcp_queue *queue) +{ + if (!IS_ENABLED(CONFIG_NVME_TCP_TLS)) + return 0; + + return queue->tls_enabled; +} + +/* + * Check if TLS is configured for the controller. + */ +static inline bool nvme_tcp_tls_configured(struct nvme_ctrl *ctrl) +{ + if (!IS_ENABLED(CONFIG_NVME_TCP_TLS)) + return 0; + + return ctrl->opts->tls || ctrl->opts->concat; +} + static inline struct blk_mq_tags *nvme_tcp_tagset(struct nvme_tcp_queue *queue) { u32 queue_idx = nvme_tcp_queue_id(queue); @@ -207,6 +270,18 @@ static inline u8 nvme_tcp_ddgst_len(struct nvme_tcp_queue *queue) return queue->data_digest ? NVME_TCP_DIGEST_LENGTH : 0; } +static inline void *nvme_tcp_req_cmd_pdu(struct nvme_tcp_request *req) +{ + return req->pdu; +} + +static inline void *nvme_tcp_req_data_pdu(struct nvme_tcp_request *req) +{ + /* use the pdu space in the back for the data pdu */ + return req->pdu + sizeof(struct nvme_tcp_cmd_pdu) - + sizeof(struct nvme_tcp_data_pdu); +} + static inline size_t nvme_tcp_inline_data_size(struct nvme_tcp_request *req) { if (nvme_is_fabrics(req->req.cmd)) @@ -315,14 +390,20 @@ static inline void nvme_tcp_send_all(struct nvme_tcp_queue *queue) } while (ret > 0); } -static inline bool nvme_tcp_queue_more(struct nvme_tcp_queue *queue) +static inline bool nvme_tcp_queue_has_pending(struct nvme_tcp_queue *queue) { return !list_empty(&queue->send_list) || !llist_empty(&queue->req_list); } +static inline bool nvme_tcp_queue_more(struct nvme_tcp_queue *queue) +{ + return !nvme_tcp_queue_tls(queue) && + nvme_tcp_queue_has_pending(queue); +} + static inline void nvme_tcp_queue_request(struct nvme_tcp_request *req, - bool sync, bool last) + bool last) { struct nvme_tcp_queue *queue = req->queue; bool empty; @@ -336,12 +417,12 @@ static inline void nvme_tcp_queue_request(struct nvme_tcp_request *req, * are on the same cpu, so we don't introduce contention. */ if (queue->io_cpu == raw_smp_processor_id() && - sync && empty && mutex_trylock(&queue->send_mutex)) { + empty && mutex_trylock(&queue->send_mutex)) { nvme_tcp_send_all(queue); mutex_unlock(&queue->send_mutex); } - if (last && nvme_tcp_queue_more(queue)) + if (last && nvme_tcp_queue_has_pending(queue)) queue_work_on(queue->io_cpu, nvme_tcp_wq, &queue->io_work); } @@ -371,36 +452,43 @@ nvme_tcp_fetch_request(struct nvme_tcp_queue *queue) return NULL; } - list_del(&req->entry); + list_del_init(&req->entry); + init_llist_node(&req->lentry); return req; } -static inline void nvme_tcp_ddgst_final(struct ahash_request *hash, - __le32 *dgst) +#define NVME_TCP_CRC_SEED (~0) + +static inline void nvme_tcp_ddgst_update(u32 *crcp, + struct page *page, size_t off, size_t len) { - ahash_request_set_crypt(hash, NULL, (u8 *)dgst, 0); - crypto_ahash_final(hash); + page += off / PAGE_SIZE; + off %= PAGE_SIZE; + while (len) { + const void *vaddr = kmap_local_page(page); + size_t n = min(len, (size_t)PAGE_SIZE - off); + + *crcp = crc32c(*crcp, vaddr + off, n); + kunmap_local(vaddr); + page++; + off = 0; + len -= n; + } } -static inline void nvme_tcp_ddgst_update(struct ahash_request *hash, - struct page *page, off_t off, size_t len) +static inline __le32 nvme_tcp_ddgst_final(u32 crc) { - struct scatterlist sg; - - sg_init_table(&sg, 1); - sg_set_page(&sg, page, len, off); - ahash_request_set_crypt(hash, &sg, NULL, len); - crypto_ahash_update(hash); + return cpu_to_le32(~crc); } -static inline void nvme_tcp_hdgst(struct ahash_request *hash, - void *pdu, size_t len) +static inline __le32 nvme_tcp_hdgst(const void *pdu, size_t len) { - struct scatterlist sg; + return cpu_to_le32(~crc32c(NVME_TCP_CRC_SEED, pdu, len)); +} - sg_init_one(&sg, pdu, len); - ahash_request_set_crypt(hash, &sg, pdu + len, len); - crypto_ahash_digest(hash); +static inline void nvme_tcp_set_hdgst(void *pdu, size_t len) +{ + *(__le32 *)(pdu + len) = nvme_tcp_hdgst(pdu, len); } static int nvme_tcp_verify_hdgst(struct nvme_tcp_queue *queue, @@ -418,8 +506,7 @@ static int nvme_tcp_verify_hdgst(struct nvme_tcp_queue *queue, } recv_digest = *(__le32 *)(pdu + hdr->hlen); - nvme_tcp_hdgst(queue->rcv_hash, pdu, pdu_len); - exp_digest = *(__le32 *)(pdu + hdr->hlen); + exp_digest = nvme_tcp_hdgst(pdu, pdu_len); if (recv_digest != exp_digest) { dev_err(queue->ctrl->ctrl.device, "header digest error: recv %#x expected %#x\n", @@ -445,7 +532,7 @@ static int nvme_tcp_check_ddgst(struct nvme_tcp_queue *queue, void *pdu) nvme_tcp_queue_id(queue)); return -EPROTO; } - crypto_ahash_init(queue->rcv_hash); + queue->rcv_crc = NVME_TCP_CRC_SEED; return 0; } @@ -479,6 +566,8 @@ static int nvme_tcp_init_request(struct blk_mq_tag_set *set, req->queue = queue; nvme_req(rq)->ctrl = &ctrl->ctrl; nvme_req(rq)->cmd = &pdu->cmd; + init_llist_node(&req->lentry); + INIT_LIST_HEAD(&req->entry); return 0; } @@ -613,7 +702,7 @@ static int nvme_tcp_handle_comp(struct nvme_tcp_queue *queue, static void nvme_tcp_setup_h2c_data_pdu(struct nvme_tcp_request *req) { - struct nvme_tcp_data_pdu *data = req->pdu; + struct nvme_tcp_data_pdu *data = nvme_tcp_req_data_pdu(req); struct nvme_tcp_queue *queue = req->queue; struct request *rq = blk_mq_rq_from_pdu(req); u32 h2cdata_sent = req->pdu_len; @@ -683,17 +772,61 @@ static int nvme_tcp_handle_r2t(struct nvme_tcp_queue *queue, return -EPROTO; } + if (llist_on_list(&req->lentry) || + !list_empty(&req->entry)) { + dev_err(queue->ctrl->ctrl.device, + "req %d unexpected r2t while processing request\n", + rq->tag); + return -EPROTO; + } + req->pdu_len = 0; req->h2cdata_left = r2t_length; req->h2cdata_offset = r2t_offset; req->ttag = pdu->ttag; nvme_tcp_setup_h2c_data_pdu(req); - nvme_tcp_queue_request(req, false, true); + + llist_add(&req->lentry, &queue->req_list); + queue_work_on(queue->io_cpu, nvme_tcp_wq, &queue->io_work); return 0; } +static void nvme_tcp_handle_c2h_term(struct nvme_tcp_queue *queue, + struct nvme_tcp_term_pdu *pdu) +{ + u16 fes; + const char *msg; + u32 plen = le32_to_cpu(pdu->hdr.plen); + + static const char * const msg_table[] = { + [NVME_TCP_FES_INVALID_PDU_HDR] = "Invalid PDU Header Field", + [NVME_TCP_FES_PDU_SEQ_ERR] = "PDU Sequence Error", + [NVME_TCP_FES_HDR_DIGEST_ERR] = "Header Digest Error", + [NVME_TCP_FES_DATA_OUT_OF_RANGE] = "Data Transfer Out Of Range", + [NVME_TCP_FES_DATA_LIMIT_EXCEEDED] = "Data Transfer Limit Exceeded", + [NVME_TCP_FES_UNSUPPORTED_PARAM] = "Unsupported Parameter", + }; + + if (plen < NVME_TCP_MIN_C2HTERM_PLEN || + plen > NVME_TCP_MAX_C2HTERM_PLEN) { + dev_err(queue->ctrl->ctrl.device, + "Received a malformed C2HTermReq PDU (plen = %u)\n", + plen); + return; + } + + fes = le16_to_cpu(pdu->fes); + if (fes && fes < ARRAY_SIZE(msg_table)) + msg = msg_table[fes]; + else + msg = "Unknown"; + + dev_err(queue->ctrl->ctrl.device, + "Received C2HTermReq (FES = %s)\n", msg); +} + static int nvme_tcp_recv_pdu(struct nvme_tcp_queue *queue, struct sk_buff *skb, unsigned int *offset, size_t *len) { @@ -715,6 +848,25 @@ static int nvme_tcp_recv_pdu(struct nvme_tcp_queue *queue, struct sk_buff *skb, return 0; hdr = queue->pdu; + if (unlikely(hdr->hlen != sizeof(struct nvme_tcp_rsp_pdu))) { + if (!nvme_tcp_recv_pdu_supported(hdr->type)) + goto unsupported_pdu; + + dev_err(queue->ctrl->ctrl.device, + "pdu type %d has unexpected header length (%d)\n", + hdr->type, hdr->hlen); + return -EPROTO; + } + + if (unlikely(hdr->type == nvme_tcp_c2h_term)) { + /* + * C2HTermReq never includes Header or Data digests. + * Skip the checks. + */ + nvme_tcp_handle_c2h_term(queue, (void *)queue->pdu); + return -EINVAL; + } + if (queue->hdr_digest) { ret = nvme_tcp_verify_hdgst(queue, queue->pdu, hdr->hlen); if (unlikely(ret)) @@ -738,10 +890,13 @@ static int nvme_tcp_recv_pdu(struct nvme_tcp_queue *queue, struct sk_buff *skb, nvme_tcp_init_recv_ctx(queue); return nvme_tcp_handle_r2t(queue, (void *)queue->pdu); default: - dev_err(queue->ctrl->ctrl.device, - "unsupported pdu type (%d)\n", hdr->type); - return -EINVAL; + goto unsupported_pdu; } + +unsupported_pdu: + dev_err(queue->ctrl->ctrl.device, + "unsupported pdu type (%d)\n", hdr->type); + return -EINVAL; } static inline void nvme_tcp_end_request(struct request *rq, u16 status) @@ -789,8 +944,8 @@ static int nvme_tcp_recv_data(struct nvme_tcp_queue *queue, struct sk_buff *skb, iov_iter_count(&req->iter)); if (queue->data_digest) - ret = skb_copy_and_hash_datagram_iter(skb, *offset, - &req->iter, recv_len, queue->rcv_hash); + ret = skb_copy_and_crc32c_datagram_iter(skb, *offset, + &req->iter, recv_len, &queue->rcv_crc); else ret = skb_copy_datagram_iter(skb, *offset, &req->iter, recv_len); @@ -808,7 +963,7 @@ static int nvme_tcp_recv_data(struct nvme_tcp_queue *queue, struct sk_buff *skb, if (!queue->data_remaining) { if (queue->data_digest) { - nvme_tcp_ddgst_final(queue->rcv_hash, &queue->exp_ddgst); + queue->exp_ddgst = nvme_tcp_ddgst_final(queue->rcv_crc); queue->ddgst_remaining = NVME_TCP_DIGEST_LENGTH; } else { if (pdu->hdr.flags & NVME_TCP_F_DATA_SUCCESS) { @@ -875,6 +1030,9 @@ static int nvme_tcp_recv_skb(read_descriptor_t *desc, struct sk_buff *skb, size_t consumed = len; int result; + if (unlikely(!queue->rd_enabled)) + return -EFAULT; + while (len) { switch (nvme_tcp_recv_state(queue)) { case NVME_TCP_RECV_PDU: @@ -905,6 +1063,8 @@ static void nvme_tcp_data_ready(struct sock *sk) { struct nvme_tcp_queue *queue; + trace_sk_data_ready(sk); + read_lock_bh(&sk->sk_callback_lock); queue = sk->sk_user_data; if (likely(queue && queue->rd_enabled) && @@ -921,6 +1081,9 @@ static void nvme_tcp_write_space(struct sock *sk) queue = sk->sk_user_data; if (likely(queue && sk_stream_is_writeable(sk))) { clear_bit(SOCK_NOSPACE, &sk->sk_socket->flags); + /* Ensure pending TLS partial records are retried */ + if (nvme_tcp_queue_tls(queue)) + queue->write_space(sk); queue_work_on(queue->io_cpu, nvme_tcp_wq, &queue->io_work); } read_unlock_bh(&sk->sk_callback_lock); @@ -979,30 +1142,33 @@ static int nvme_tcp_try_send_data(struct nvme_tcp_request *req) u32 h2cdata_left = req->h2cdata_left; while (true) { + struct bio_vec bvec; + struct msghdr msg = { + .msg_flags = MSG_DONTWAIT | MSG_SPLICE_PAGES, + }; struct page *page = nvme_tcp_req_cur_page(req); size_t offset = nvme_tcp_req_cur_offset(req); size_t len = nvme_tcp_req_cur_length(req); bool last = nvme_tcp_pdu_last_send(req, len); int req_data_sent = req->data_sent; - int ret, flags = MSG_DONTWAIT; + int ret; if (last && !queue->data_digest && !nvme_tcp_queue_more(queue)) - flags |= MSG_EOR; + msg.msg_flags |= MSG_EOR; else - flags |= MSG_MORE | MSG_SENDPAGE_NOTLAST; + msg.msg_flags |= MSG_MORE; - if (sendpage_ok(page)) { - ret = kernel_sendpage(queue->sock, page, offset, len, - flags); - } else { - ret = sock_no_sendpage(queue->sock, page, offset, len, - flags); - } + if (!sendpages_ok(page, len, offset)) + msg.msg_flags &= ~MSG_SPLICE_PAGES; + + bvec_set_page(&bvec, page, len, offset); + iov_iter_bvec(&msg.msg_iter, ITER_SOURCE, &bvec, 1, len); + ret = sock_sendmsg(queue->sock, &msg); if (ret <= 0) return ret; if (queue->data_digest) - nvme_tcp_ddgst_update(queue->snd_hash, page, + nvme_tcp_ddgst_update(&queue->snd_crc, page, offset, ret); /* @@ -1016,8 +1182,8 @@ static int nvme_tcp_try_send_data(struct nvme_tcp_request *req) /* fully successful last send in current PDU */ if (last && ret == len) { if (queue->data_digest) { - nvme_tcp_ddgst_final(queue->snd_hash, - &req->ddgst); + req->ddgst = + nvme_tcp_ddgst_final(queue->snd_crc); req->state = NVME_TCP_SEND_DDGST; req->offset = 0; } else { @@ -1035,23 +1201,25 @@ static int nvme_tcp_try_send_data(struct nvme_tcp_request *req) static int nvme_tcp_try_send_cmd_pdu(struct nvme_tcp_request *req) { struct nvme_tcp_queue *queue = req->queue; - struct nvme_tcp_cmd_pdu *pdu = req->pdu; + struct nvme_tcp_cmd_pdu *pdu = nvme_tcp_req_cmd_pdu(req); + struct bio_vec bvec; + struct msghdr msg = { .msg_flags = MSG_DONTWAIT | MSG_SPLICE_PAGES, }; bool inline_data = nvme_tcp_has_inline_data(req); u8 hdgst = nvme_tcp_hdgst_len(queue); int len = sizeof(*pdu) + hdgst - req->offset; - int flags = MSG_DONTWAIT; int ret; if (inline_data || nvme_tcp_queue_more(queue)) - flags |= MSG_MORE | MSG_SENDPAGE_NOTLAST; + msg.msg_flags |= MSG_MORE; else - flags |= MSG_EOR; + msg.msg_flags |= MSG_EOR; if (queue->hdr_digest && !req->offset) - nvme_tcp_hdgst(queue->snd_hash, pdu, sizeof(*pdu)); + nvme_tcp_set_hdgst(pdu, sizeof(*pdu)); - ret = kernel_sendpage(queue->sock, virt_to_page(pdu), - offset_in_page(pdu) + req->offset, len, flags); + bvec_set_virt(&bvec, (void *)pdu + req->offset, len); + iov_iter_bvec(&msg.msg_iter, ITER_SOURCE, &bvec, 1, len); + ret = sock_sendmsg(queue->sock, &msg); if (unlikely(ret <= 0)) return ret; @@ -1060,7 +1228,7 @@ static int nvme_tcp_try_send_cmd_pdu(struct nvme_tcp_request *req) if (inline_data) { req->state = NVME_TCP_SEND_DATA; if (queue->data_digest) - crypto_ahash_init(queue->snd_hash); + queue->snd_crc = NVME_TCP_CRC_SEED; } else { nvme_tcp_done_send_req(queue); } @@ -1074,22 +1242,22 @@ static int nvme_tcp_try_send_cmd_pdu(struct nvme_tcp_request *req) static int nvme_tcp_try_send_data_pdu(struct nvme_tcp_request *req) { struct nvme_tcp_queue *queue = req->queue; - struct nvme_tcp_data_pdu *pdu = req->pdu; + struct nvme_tcp_data_pdu *pdu = nvme_tcp_req_data_pdu(req); + struct bio_vec bvec; + struct msghdr msg = { .msg_flags = MSG_DONTWAIT | MSG_MORE, }; u8 hdgst = nvme_tcp_hdgst_len(queue); int len = sizeof(*pdu) - req->offset + hdgst; int ret; if (queue->hdr_digest && !req->offset) - nvme_tcp_hdgst(queue->snd_hash, pdu, sizeof(*pdu)); + nvme_tcp_set_hdgst(pdu, sizeof(*pdu)); if (!req->h2cdata_left) - ret = kernel_sendpage(queue->sock, virt_to_page(pdu), - offset_in_page(pdu) + req->offset, len, - MSG_DONTWAIT | MSG_MORE | MSG_SENDPAGE_NOTLAST); - else - ret = sock_no_sendpage(queue->sock, virt_to_page(pdu), - offset_in_page(pdu) + req->offset, len, - MSG_DONTWAIT | MSG_MORE); + msg.msg_flags |= MSG_SPLICE_PAGES; + + bvec_set_virt(&bvec, (void *)pdu + req->offset, len); + iov_iter_bvec(&msg.msg_iter, ITER_SOURCE, &bvec, 1, len); + ret = sock_sendmsg(queue->sock, &msg); if (unlikely(ret <= 0)) return ret; @@ -1097,7 +1265,7 @@ static int nvme_tcp_try_send_data_pdu(struct nvme_tcp_request *req) if (!len) { req->state = NVME_TCP_SEND_DATA; if (queue->data_digest) - crypto_ahash_init(queue->snd_hash); + queue->snd_crc = NVME_TCP_CRC_SEED; return 1; } req->offset += ret; @@ -1201,7 +1369,7 @@ static int nvme_tcp_try_recv(struct nvme_tcp_queue *queue) queue->nr_cqe = 0; consumed = sock->ops->read_sock(sk, &rd_desc, nvme_tcp_recv_skb); release_sock(sk); - return consumed; + return consumed == -EAGAIN ? 0 : consumed; } static void nvme_tcp_io_work(struct work_struct *w) @@ -1229,6 +1397,11 @@ static void nvme_tcp_io_work(struct work_struct *w) else if (unlikely(result < 0)) return; + /* did we get some space after spending time in recv? */ + if (nvme_tcp_queue_has_pending(queue) && + sk_stream_is_writeable(queue->sock->sk)) + pending = true; + if (!pending || !queue->rd_enabled) return; @@ -1237,41 +1410,6 @@ static void nvme_tcp_io_work(struct work_struct *w) queue_work_on(queue->io_cpu, nvme_tcp_wq, &queue->io_work); } -static void nvme_tcp_free_crypto(struct nvme_tcp_queue *queue) -{ - struct crypto_ahash *tfm = crypto_ahash_reqtfm(queue->rcv_hash); - - ahash_request_free(queue->rcv_hash); - ahash_request_free(queue->snd_hash); - crypto_free_ahash(tfm); -} - -static int nvme_tcp_alloc_crypto(struct nvme_tcp_queue *queue) -{ - struct crypto_ahash *tfm; - - tfm = crypto_alloc_ahash("crc32c", 0, CRYPTO_ALG_ASYNC); - if (IS_ERR(tfm)) - return PTR_ERR(tfm); - - queue->snd_hash = ahash_request_alloc(tfm, GFP_KERNEL); - if (!queue->snd_hash) - goto free_tfm; - ahash_request_set_callback(queue->snd_hash, 0, NULL, NULL); - - queue->rcv_hash = ahash_request_alloc(tfm, GFP_KERNEL); - if (!queue->rcv_hash) - goto free_snd_hash; - ahash_request_set_callback(queue->rcv_hash, 0, NULL, NULL); - - return 0; -free_snd_hash: - ahash_request_free(queue->snd_hash); -free_tfm: - crypto_free_ahash(tfm); - return -ENOMEM; -} - static void nvme_tcp_free_async_req(struct nvme_tcp_ctrl *ctrl) { struct nvme_tcp_request *async = &ctrl->async_req; @@ -1297,7 +1435,6 @@ static int nvme_tcp_alloc_async_req(struct nvme_tcp_ctrl *ctrl) static void nvme_tcp_free_queue(struct nvme_ctrl *nctrl, int qid) { - struct page *page; struct nvme_tcp_ctrl *ctrl = to_tcp_ctrl(nctrl); struct nvme_tcp_queue *queue = &ctrl->queues[qid]; unsigned int noreclaim_flag; @@ -1305,17 +1442,12 @@ static void nvme_tcp_free_queue(struct nvme_ctrl *nctrl, int qid) if (!test_and_clear_bit(NVME_TCP_Q_ALLOCATED, &queue->flags)) return; - if (queue->hdr_digest || queue->data_digest) - nvme_tcp_free_crypto(queue); - - if (queue->pf_cache.va) { - page = virt_to_head_page(queue->pf_cache.va); - __page_frag_cache_drain(page, queue->pf_cache.pagecnt_bias); - queue->pf_cache.va = NULL; - } + page_frag_cache_drain(&queue->pf_cache); noreclaim_flag = memalloc_noreclaim_save(); - sock_release(queue->sock); + /* ->sock will be released by fput() */ + fput(queue->sock->file); + queue->sock = NULL; memalloc_noreclaim_restore(noreclaim_flag); kfree(queue->pdu); @@ -1327,6 +1459,8 @@ static int nvme_tcp_init_connection(struct nvme_tcp_queue *queue) { struct nvme_tcp_icreq_pdu *icreq; struct nvme_tcp_icresp_pdu *icresp; + char cbuf[CMSG_LEN(sizeof(char))] = {}; + u8 ctype; struct msghdr msg = {}; struct kvec iov; bool ctrl_hdgst, ctrl_ddgst; @@ -1358,17 +1492,39 @@ static int nvme_tcp_init_connection(struct nvme_tcp_queue *queue) iov.iov_base = icreq; iov.iov_len = sizeof(*icreq); ret = kernel_sendmsg(queue->sock, &msg, &iov, 1, iov.iov_len); - if (ret < 0) + if (ret < 0) { + pr_warn("queue %d: failed to send icreq, error %d\n", + nvme_tcp_queue_id(queue), ret); goto free_icresp; + } memset(&msg, 0, sizeof(msg)); iov.iov_base = icresp; iov.iov_len = sizeof(*icresp); + if (nvme_tcp_queue_tls(queue)) { + msg.msg_control = cbuf; + msg.msg_controllen = sizeof(cbuf); + } + msg.msg_flags = MSG_WAITALL; ret = kernel_recvmsg(queue->sock, &msg, &iov, 1, iov.iov_len, msg.msg_flags); - if (ret < 0) + if (ret >= 0 && ret < sizeof(*icresp)) + ret = -ECONNRESET; + if (ret < 0) { + pr_warn("queue %d: failed to receive icresp, error %d\n", + nvme_tcp_queue_id(queue), ret); goto free_icresp; - + } + ret = -ENOTCONN; + if (nvme_tcp_queue_tls(queue)) { + ctype = tls_get_record_type(queue->sock->sk, + (struct cmsghdr *)cbuf); + if (ctype != TLS_RECORD_TYPE_DATA) { + pr_err("queue %d: unhandled TLS record %d\n", + nvme_tcp_queue_id(queue), ctype); + goto free_icresp; + } + } ret = -EINVAL; if (icresp->hdr.type != nvme_tcp_icresp) { pr_err("queue %d: bad type returned %d\n", @@ -1468,27 +1624,150 @@ static bool nvme_tcp_poll_queue(struct nvme_tcp_queue *queue) ctrl->io_queues[HCTX_TYPE_POLL]; } +/* + * Track the number of queues assigned to each cpu using a global per-cpu + * counter and select the least used cpu from the mq_map. Our goal is to spread + * different controllers I/O threads across different cpu cores. + * + * Note that the accounting is not 100% perfect, but we don't need to be, we're + * simply putting our best effort to select the best candidate cpu core that we + * find at any given point. + */ static void nvme_tcp_set_queue_io_cpu(struct nvme_tcp_queue *queue) { struct nvme_tcp_ctrl *ctrl = queue->ctrl; - int qid = nvme_tcp_queue_id(queue); - int n = 0; + struct blk_mq_tag_set *set = &ctrl->tag_set; + int qid = nvme_tcp_queue_id(queue) - 1; + unsigned int *mq_map = NULL; + int cpu, min_queues = INT_MAX, io_cpu; + + if (wq_unbound) + goto out; if (nvme_tcp_default_queue(queue)) - n = qid - 1; + mq_map = set->map[HCTX_TYPE_DEFAULT].mq_map; else if (nvme_tcp_read_queue(queue)) - n = qid - ctrl->io_queues[HCTX_TYPE_DEFAULT] - 1; + mq_map = set->map[HCTX_TYPE_READ].mq_map; else if (nvme_tcp_poll_queue(queue)) - n = qid - ctrl->io_queues[HCTX_TYPE_DEFAULT] - - ctrl->io_queues[HCTX_TYPE_READ] - 1; - queue->io_cpu = cpumask_next_wrap(n - 1, cpu_online_mask, -1, false); + mq_map = set->map[HCTX_TYPE_POLL].mq_map; + + if (WARN_ON(!mq_map)) + goto out; + + /* Search for the least used cpu from the mq_map */ + io_cpu = WORK_CPU_UNBOUND; + for_each_online_cpu(cpu) { + int num_queues = atomic_read(&nvme_tcp_cpu_queues[cpu]); + + if (mq_map[cpu] != qid) + continue; + if (num_queues < min_queues) { + io_cpu = cpu; + min_queues = num_queues; + } + } + if (io_cpu != WORK_CPU_UNBOUND) { + queue->io_cpu = io_cpu; + atomic_inc(&nvme_tcp_cpu_queues[io_cpu]); + set_bit(NVME_TCP_Q_IO_CPU_SET, &queue->flags); + } +out: + dev_dbg(ctrl->ctrl.device, "queue %d: using cpu %d\n", + qid, queue->io_cpu); +} + +static void nvme_tcp_tls_done(void *data, int status, key_serial_t pskid) +{ + struct nvme_tcp_queue *queue = data; + struct nvme_tcp_ctrl *ctrl = queue->ctrl; + int qid = nvme_tcp_queue_id(queue); + struct key *tls_key; + + dev_dbg(ctrl->ctrl.device, "queue %d: TLS handshake done, key %x, status %d\n", + qid, pskid, status); + + if (status) { + queue->tls_err = -status; + goto out_complete; + } + + tls_key = nvme_tls_key_lookup(pskid); + if (IS_ERR(tls_key)) { + dev_warn(ctrl->ctrl.device, "queue %d: Invalid key %x\n", + qid, pskid); + queue->tls_err = -ENOKEY; + } else { + queue->tls_enabled = true; + if (qid == 0) + ctrl->ctrl.tls_pskid = key_serial(tls_key); + key_put(tls_key); + queue->tls_err = 0; + } + +out_complete: + complete(&queue->tls_complete); +} + +static int nvme_tcp_start_tls(struct nvme_ctrl *nctrl, + struct nvme_tcp_queue *queue, + key_serial_t pskid) +{ + int qid = nvme_tcp_queue_id(queue); + int ret; + struct tls_handshake_args args; + unsigned long tmo = tls_handshake_timeout * HZ; + key_serial_t keyring = nvme_keyring_id(); + + dev_dbg(nctrl->device, "queue %d: start TLS with key %x\n", + qid, pskid); + memset(&args, 0, sizeof(args)); + args.ta_sock = queue->sock; + args.ta_done = nvme_tcp_tls_done; + args.ta_data = queue; + args.ta_my_peerids[0] = pskid; + args.ta_num_peerids = 1; + if (nctrl->opts->keyring) + keyring = key_serial(nctrl->opts->keyring); + args.ta_keyring = keyring; + args.ta_timeout_ms = tls_handshake_timeout * 1000; + queue->tls_err = -EOPNOTSUPP; + init_completion(&queue->tls_complete); + ret = tls_client_hello_psk(&args, GFP_KERNEL); + if (ret) { + dev_err(nctrl->device, "queue %d: failed to start TLS: %d\n", + qid, ret); + return ret; + } + ret = wait_for_completion_interruptible_timeout(&queue->tls_complete, tmo); + if (ret <= 0) { + if (ret == 0) + ret = -ETIMEDOUT; + + dev_err(nctrl->device, + "queue %d: TLS handshake failed, error %d\n", + qid, ret); + tls_handshake_cancel(queue->sock->sk); + } else { + if (queue->tls_err) { + dev_err(nctrl->device, + "queue %d: TLS handshake complete, error %d\n", + qid, queue->tls_err); + } else { + dev_dbg(nctrl->device, + "queue %d: TLS handshake complete\n", qid); + } + ret = queue->tls_err; + } + return ret; } -static int nvme_tcp_alloc_queue(struct nvme_ctrl *nctrl, int qid) +static int nvme_tcp_alloc_queue(struct nvme_ctrl *nctrl, int qid, + key_serial_t pskid) { struct nvme_tcp_ctrl *ctrl = to_tcp_ctrl(nctrl); struct nvme_tcp_queue *queue = &ctrl->queues[qid]; int ret, rcv_pdu_size; + struct file *sock_file; mutex_init(&queue->queue_lock); queue->ctrl = ctrl; @@ -1503,7 +1782,8 @@ static int nvme_tcp_alloc_queue(struct nvme_ctrl *nctrl, int qid) queue->cmnd_capsule_len = sizeof(struct nvme_command) + NVME_TCP_ADMIN_CCSZ; - ret = sock_create(ctrl->addr.ss_family, SOCK_STREAM, + ret = sock_create_kern(current->nsproxy->net_ns, + ctrl->addr.ss_family, SOCK_STREAM, IPPROTO_TCP, &queue->sock); if (ret) { dev_err(nctrl->device, @@ -1511,6 +1791,13 @@ static int nvme_tcp_alloc_queue(struct nvme_ctrl *nctrl, int qid) goto err_destroy_mutex; } + sock_file = sock_alloc_file(queue->sock, O_CLOEXEC, NULL); + if (IS_ERR(sock_file)) { + ret = PTR_ERR(sock_file); + goto err_destroy_mutex; + } + + sk_net_refcnt_upgrade(queue->sock->sk); nvme_tcp_reclassify_socket(queue->sock); /* Single syn retry */ @@ -1538,7 +1825,7 @@ static int nvme_tcp_alloc_queue(struct nvme_ctrl *nctrl, int qid) queue->sock->sk->sk_allocation = GFP_ATOMIC; queue->sock->sk->sk_use_task_frag = false; - nvme_tcp_set_queue_io_cpu(queue); + queue->io_cpu = WORK_CPU_UNBOUND; queue->request = NULL; queue->data_remaining = 0; queue->ddgst_remaining = 0; @@ -1547,7 +1834,7 @@ static int nvme_tcp_alloc_queue(struct nvme_ctrl *nctrl, int qid) sk_set_memalloc(queue->sock->sk); if (nctrl->opts->mask & NVMF_OPT_HOST_TRADDR) { - ret = kernel_bind(queue->sock, (struct sockaddr *)&ctrl->src_addr, + ret = kernel_bind(queue->sock, (struct sockaddr_unsized *)&ctrl->src_addr, sizeof(ctrl->src_addr)); if (ret) { dev_err(nctrl->device, @@ -1573,27 +1860,19 @@ static int nvme_tcp_alloc_queue(struct nvme_ctrl *nctrl, int qid) queue->hdr_digest = nctrl->opts->hdr_digest; queue->data_digest = nctrl->opts->data_digest; - if (queue->hdr_digest || queue->data_digest) { - ret = nvme_tcp_alloc_crypto(queue); - if (ret) { - dev_err(nctrl->device, - "failed to allocate queue %d crypto\n", qid); - goto err_sock; - } - } rcv_pdu_size = sizeof(struct nvme_tcp_rsp_pdu) + nvme_tcp_hdgst_len(queue); queue->pdu = kmalloc(rcv_pdu_size, GFP_KERNEL); if (!queue->pdu) { ret = -ENOMEM; - goto err_crypto; + goto err_sock; } dev_dbg(nctrl->device, "connecting queue %d\n", nvme_tcp_queue_id(queue)); - ret = kernel_connect(queue->sock, (struct sockaddr *)&ctrl->addr, + ret = kernel_connect(queue->sock, (struct sockaddr_unsized *)&ctrl->addr, sizeof(ctrl->addr), 0); if (ret) { dev_err(nctrl->device, @@ -1601,26 +1880,18 @@ static int nvme_tcp_alloc_queue(struct nvme_ctrl *nctrl, int qid) goto err_rcv_pdu; } + /* If PSKs are configured try to start TLS */ + if (nvme_tcp_tls_configured(nctrl) && pskid) { + ret = nvme_tcp_start_tls(nctrl, queue, pskid); + if (ret) + goto err_init_connect; + } + ret = nvme_tcp_init_connection(queue); if (ret) goto err_init_connect; - queue->rd_enabled = true; set_bit(NVME_TCP_Q_ALLOCATED, &queue->flags); - nvme_tcp_init_recv_ctx(queue); - - write_lock_bh(&queue->sock->sk->sk_callback_lock); - queue->sock->sk->sk_user_data = queue; - queue->state_change = queue->sock->sk->sk_state_change; - queue->data_ready = queue->sock->sk->sk_data_ready; - queue->write_space = queue->sock->sk->sk_write_space; - queue->sock->sk->sk_data_ready = nvme_tcp_data_ready; - queue->sock->sk->sk_state_change = nvme_tcp_state_change; - queue->sock->sk->sk_write_space = nvme_tcp_write_space; -#ifdef CONFIG_NET_RX_BUSY_POLL - queue->sock->sk->sk_ll_usec = 1; -#endif - write_unlock_bh(&queue->sock->sk->sk_callback_lock); return 0; @@ -1628,11 +1899,9 @@ err_init_connect: kernel_sock_shutdown(queue->sock, SHUT_RDWR); err_rcv_pdu: kfree(queue->pdu); -err_crypto: - if (queue->hdr_digest || queue->data_digest) - nvme_tcp_free_crypto(queue); err_sock: - sock_release(queue->sock); + /* ->sock will be released by fput() */ + fput(queue->sock->file); queue->sock = NULL; err_destroy_mutex: mutex_destroy(&queue->send_mutex); @@ -1640,7 +1909,7 @@ err_destroy_mutex: return ret; } -static void nvme_tcp_restore_sock_calls(struct nvme_tcp_queue *queue) +static void nvme_tcp_restore_sock_ops(struct nvme_tcp_queue *queue) { struct socket *sock = queue->sock; @@ -1655,11 +1924,11 @@ static void nvme_tcp_restore_sock_calls(struct nvme_tcp_queue *queue) static void __nvme_tcp_stop_queue(struct nvme_tcp_queue *queue) { kernel_sock_shutdown(queue->sock, SHUT_RDWR); - nvme_tcp_restore_sock_calls(queue); + nvme_tcp_restore_sock_ops(queue); cancel_work_sync(&queue->io_work); } -static void nvme_tcp_stop_queue(struct nvme_ctrl *nctrl, int qid) +static void nvme_tcp_stop_queue_nowait(struct nvme_ctrl *nctrl, int qid) { struct nvme_tcp_ctrl *ctrl = to_tcp_ctrl(nctrl); struct nvme_tcp_queue *queue = &ctrl->queues[qid]; @@ -1667,27 +1936,79 @@ static void nvme_tcp_stop_queue(struct nvme_ctrl *nctrl, int qid) if (!test_bit(NVME_TCP_Q_ALLOCATED, &queue->flags)) return; + if (test_and_clear_bit(NVME_TCP_Q_IO_CPU_SET, &queue->flags)) + atomic_dec(&nvme_tcp_cpu_queues[queue->io_cpu]); + mutex_lock(&queue->queue_lock); if (test_and_clear_bit(NVME_TCP_Q_LIVE, &queue->flags)) __nvme_tcp_stop_queue(queue); + /* Stopping the queue will disable TLS */ + queue->tls_enabled = false; mutex_unlock(&queue->queue_lock); } +static void nvme_tcp_wait_queue(struct nvme_ctrl *nctrl, int qid) +{ + struct nvme_tcp_ctrl *ctrl = to_tcp_ctrl(nctrl); + struct nvme_tcp_queue *queue = &ctrl->queues[qid]; + int timeout = 100; + + while (timeout > 0) { + if (!test_bit(NVME_TCP_Q_ALLOCATED, &queue->flags) || + !sk_wmem_alloc_get(queue->sock->sk)) + return; + msleep(2); + timeout -= 2; + } + dev_warn(nctrl->device, + "qid %d: timeout draining sock wmem allocation expired\n", + qid); +} + +static void nvme_tcp_stop_queue(struct nvme_ctrl *nctrl, int qid) +{ + nvme_tcp_stop_queue_nowait(nctrl, qid); + nvme_tcp_wait_queue(nctrl, qid); +} + + +static void nvme_tcp_setup_sock_ops(struct nvme_tcp_queue *queue) +{ + write_lock_bh(&queue->sock->sk->sk_callback_lock); + queue->sock->sk->sk_user_data = queue; + queue->state_change = queue->sock->sk->sk_state_change; + queue->data_ready = queue->sock->sk->sk_data_ready; + queue->write_space = queue->sock->sk->sk_write_space; + queue->sock->sk->sk_data_ready = nvme_tcp_data_ready; + queue->sock->sk->sk_state_change = nvme_tcp_state_change; + queue->sock->sk->sk_write_space = nvme_tcp_write_space; +#ifdef CONFIG_NET_RX_BUSY_POLL + queue->sock->sk->sk_ll_usec = 1; +#endif + write_unlock_bh(&queue->sock->sk->sk_callback_lock); +} + static int nvme_tcp_start_queue(struct nvme_ctrl *nctrl, int idx) { struct nvme_tcp_ctrl *ctrl = to_tcp_ctrl(nctrl); + struct nvme_tcp_queue *queue = &ctrl->queues[idx]; int ret; - if (idx) + queue->rd_enabled = true; + nvme_tcp_init_recv_ctx(queue); + nvme_tcp_setup_sock_ops(queue); + + if (idx) { + nvme_tcp_set_queue_io_cpu(queue); ret = nvmf_connect_io_queue(nctrl, idx); - else + } else ret = nvmf_connect_admin_queue(nctrl); if (!ret) { - set_bit(NVME_TCP_Q_LIVE, &ctrl->queues[idx].flags); + set_bit(NVME_TCP_Q_LIVE, &queue->flags); } else { - if (test_bit(NVME_TCP_Q_ALLOCATED, &ctrl->queues[idx].flags)) - __nvme_tcp_stop_queue(&ctrl->queues[idx]); + if (test_bit(NVME_TCP_Q_ALLOCATED, &queue->flags)) + __nvme_tcp_stop_queue(queue); dev_err(nctrl->device, "failed to connect queue: %d ret=%d\n", idx, ret); } @@ -1718,7 +2039,9 @@ static void nvme_tcp_stop_io_queues(struct nvme_ctrl *ctrl) int i; for (i = 1; i < ctrl->queue_count; i++) - nvme_tcp_stop_queue(ctrl, i); + nvme_tcp_stop_queue_nowait(ctrl, i); + for (i = 1; i < ctrl->queue_count; i++) + nvme_tcp_wait_queue(ctrl, i); } static int nvme_tcp_start_io_queues(struct nvme_ctrl *ctrl, @@ -1743,8 +2066,23 @@ out_stop_queues: static int nvme_tcp_alloc_admin_queue(struct nvme_ctrl *ctrl) { int ret; + key_serial_t pskid = 0; + + if (nvme_tcp_tls_configured(ctrl)) { + if (ctrl->opts->tls_key) + pskid = key_serial(ctrl->opts->tls_key); + else if (ctrl->opts->tls) { + pskid = nvme_tls_psk_default(ctrl->opts->keyring, + ctrl->opts->host->nqn, + ctrl->opts->subsysnqn); + if (!pskid) { + dev_err(ctrl->device, "no valid PSK found\n"); + return -ENOKEY; + } + } + } - ret = nvme_tcp_alloc_queue(ctrl, 0); + ret = nvme_tcp_alloc_queue(ctrl, 0, pskid); if (ret) return ret; @@ -1763,8 +2101,30 @@ static int __nvme_tcp_alloc_io_queues(struct nvme_ctrl *ctrl) { int i, ret; + if (nvme_tcp_tls_configured(ctrl)) { + if (ctrl->opts->concat) { + /* + * The generated PSK is stored in the + * fabric options + */ + if (!ctrl->opts->tls_key) { + dev_err(ctrl->device, "no PSK generated\n"); + return -ENOKEY; + } + if (ctrl->tls_pskid && + ctrl->tls_pskid != key_serial(ctrl->opts->tls_key)) { + dev_err(ctrl->device, "Stale PSK id %08x\n", ctrl->tls_pskid); + ctrl->tls_pskid = 0; + } + } else if (!ctrl->tls_pskid) { + dev_err(ctrl->device, "no PSK negotiated\n"); + return -ENOKEY; + } + } + for (i = 1; i < ctrl->queue_count; i++) { - ret = nvme_tcp_alloc_queue(ctrl, i); + ret = nvme_tcp_alloc_queue(ctrl, i, + ctrl->tls_pskid); if (ret) goto out_free_queues; } @@ -1778,58 +2138,12 @@ out_free_queues: return ret; } -static unsigned int nvme_tcp_nr_io_queues(struct nvme_ctrl *ctrl) -{ - unsigned int nr_io_queues; - - nr_io_queues = min(ctrl->opts->nr_io_queues, num_online_cpus()); - nr_io_queues += min(ctrl->opts->nr_write_queues, num_online_cpus()); - nr_io_queues += min(ctrl->opts->nr_poll_queues, num_online_cpus()); - - return nr_io_queues; -} - -static void nvme_tcp_set_io_queues(struct nvme_ctrl *nctrl, - unsigned int nr_io_queues) -{ - struct nvme_tcp_ctrl *ctrl = to_tcp_ctrl(nctrl); - struct nvmf_ctrl_options *opts = nctrl->opts; - - if (opts->nr_write_queues && opts->nr_io_queues < nr_io_queues) { - /* - * separate read/write queues - * hand out dedicated default queues only after we have - * sufficient read queues. - */ - ctrl->io_queues[HCTX_TYPE_READ] = opts->nr_io_queues; - nr_io_queues -= ctrl->io_queues[HCTX_TYPE_READ]; - ctrl->io_queues[HCTX_TYPE_DEFAULT] = - min(opts->nr_write_queues, nr_io_queues); - nr_io_queues -= ctrl->io_queues[HCTX_TYPE_DEFAULT]; - } else { - /* - * shared read/write queues - * either no write queues were requested, or we don't have - * sufficient queue count to have dedicated default queues. - */ - ctrl->io_queues[HCTX_TYPE_DEFAULT] = - min(opts->nr_io_queues, nr_io_queues); - nr_io_queues -= ctrl->io_queues[HCTX_TYPE_DEFAULT]; - } - - if (opts->nr_poll_queues && nr_io_queues) { - /* map dedicated poll queues only if we have queues left */ - ctrl->io_queues[HCTX_TYPE_POLL] = - min(opts->nr_poll_queues, nr_io_queues); - } -} - static int nvme_tcp_alloc_io_queues(struct nvme_ctrl *ctrl) { unsigned int nr_io_queues; int ret; - nr_io_queues = nvme_tcp_nr_io_queues(ctrl); + nr_io_queues = nvmf_nr_io_queues(ctrl->opts); ret = nvme_set_queue_count(ctrl, &nr_io_queues); if (ret) return ret; @@ -1844,19 +2158,11 @@ static int nvme_tcp_alloc_io_queues(struct nvme_ctrl *ctrl) dev_info(ctrl->device, "creating %d I/O queues.\n", nr_io_queues); - nvme_tcp_set_io_queues(ctrl, nr_io_queues); - + nvmf_set_io_queues(ctrl->opts, nr_io_queues, + to_tcp_ctrl(ctrl)->io_queues); return __nvme_tcp_alloc_io_queues(ctrl); } -static void nvme_tcp_destroy_io_queues(struct nvme_ctrl *ctrl, bool remove) -{ - nvme_tcp_stop_io_queues(ctrl); - if (remove) - nvme_remove_io_tag_set(ctrl); - nvme_tcp_free_io_queues(ctrl); -} - static int nvme_tcp_configure_io_queues(struct nvme_ctrl *ctrl, bool new) { int ret, nr_queues; @@ -1876,7 +2182,7 @@ static int nvme_tcp_configure_io_queues(struct nvme_ctrl *ctrl, bool new) /* * Only start IO queues for which we have allocated the tagset - * and limitted it to the available queues. On reconnects, the + * and limited it to the available queues. On reconnects, the * queue number might have changed. */ nr_queues = min(ctrl->tagset->nr_hw_queues + 1, ctrl->queue_count); @@ -1885,6 +2191,7 @@ static int nvme_tcp_configure_io_queues(struct nvme_ctrl *ctrl, bool new) goto out_cleanup_connect_q; if (!new) { + nvme_start_freeze(ctrl); nvme_unquiesce_io_queues(ctrl); if (!nvme_wait_freeze_timeout(ctrl, NVME_IO_TIMEOUT)) { /* @@ -1893,6 +2200,7 @@ static int nvme_tcp_configure_io_queues(struct nvme_ctrl *ctrl, bool new) * to be safe. */ ret = -ENODEV; + nvme_unfreeze(ctrl); goto out_wait_freeze_timed_out; } blk_mq_update_nr_hw_queues(ctrl->tagset, @@ -1924,14 +2232,6 @@ out_free_io_queues: return ret; } -static void nvme_tcp_destroy_admin_queue(struct nvme_ctrl *ctrl, bool remove) -{ - nvme_tcp_stop_queue(ctrl, 0); - if (remove) - nvme_remove_admin_tag_set(ctrl); - nvme_tcp_free_admin_queue(ctrl); -} - static int nvme_tcp_configure_admin_queue(struct nvme_ctrl *ctrl, bool new) { int error; @@ -1953,6 +2253,9 @@ static int nvme_tcp_configure_admin_queue(struct nvme_ctrl *ctrl, bool new) if (error) goto out_cleanup_tagset; + if (ctrl->opts->concat && !ctrl->tls_pskid) + return 0; + error = nvme_enable_ctrl(ctrl); if (error) goto out_stop_queue; @@ -1986,9 +2289,16 @@ static void nvme_tcp_teardown_admin_queue(struct nvme_ctrl *ctrl, blk_sync_queue(ctrl->admin_q); nvme_tcp_stop_queue(ctrl, 0); nvme_cancel_admin_tagset(ctrl); - if (remove) + if (remove) { nvme_unquiesce_admin_queue(ctrl); - nvme_tcp_destroy_admin_queue(ctrl, remove); + nvme_remove_admin_tag_set(ctrl); + } + nvme_tcp_free_admin_queue(ctrl); + if (ctrl->tls_pskid) { + dev_dbg(ctrl->device, "Wipe negotiated TLS_PSK %08x\n", + ctrl->tls_pskid); + ctrl->tls_pskid = 0; + } } static void nvme_tcp_teardown_io_queues(struct nvme_ctrl *ctrl, @@ -1996,37 +2306,61 @@ static void nvme_tcp_teardown_io_queues(struct nvme_ctrl *ctrl, { if (ctrl->queue_count <= 1) return; - nvme_quiesce_admin_queue(ctrl); - nvme_start_freeze(ctrl); nvme_quiesce_io_queues(ctrl); nvme_sync_io_queues(ctrl); nvme_tcp_stop_io_queues(ctrl); nvme_cancel_tagset(ctrl); - if (remove) + if (remove) { nvme_unquiesce_io_queues(ctrl); - nvme_tcp_destroy_io_queues(ctrl, remove); + nvme_remove_io_tag_set(ctrl); + } + nvme_tcp_free_io_queues(ctrl); } -static void nvme_tcp_reconnect_or_remove(struct nvme_ctrl *ctrl) +static void nvme_tcp_reconnect_or_remove(struct nvme_ctrl *ctrl, + int status) { + enum nvme_ctrl_state state = nvme_ctrl_state(ctrl); + /* If we are resetting/deleting then do nothing */ - if (ctrl->state != NVME_CTRL_CONNECTING) { - WARN_ON_ONCE(ctrl->state == NVME_CTRL_NEW || - ctrl->state == NVME_CTRL_LIVE); + if (state != NVME_CTRL_CONNECTING) { + WARN_ON_ONCE(state == NVME_CTRL_NEW || state == NVME_CTRL_LIVE); return; } - if (nvmf_should_reconnect(ctrl)) { + if (nvmf_should_reconnect(ctrl, status)) { dev_info(ctrl->device, "Reconnecting in %d seconds...\n", ctrl->opts->reconnect_delay); queue_delayed_work(nvme_wq, &to_tcp_ctrl(ctrl)->connect_work, ctrl->opts->reconnect_delay * HZ); } else { - dev_info(ctrl->device, "Removing controller...\n"); + dev_info(ctrl->device, "Removing controller (%d)...\n", + status); nvme_delete_ctrl(ctrl); } } +/* + * The TLS key is set by secure concatenation after negotiation has been + * completed on the admin queue. We need to revoke the key when: + * - concatenation is enabled (otherwise it's a static key set by the user) + * and + * - the generated key is present in ctrl->tls_key (otherwise there's nothing + * to revoke) + * and + * - a valid PSK key ID has been set in ctrl->tls_pskid (otherwise TLS + * negotiation has not run). + * + * We cannot always revoke the key as nvme_tcp_alloc_admin_queue() is called + * twice during secure concatenation, once on a 'normal' connection to run the + * DH-HMAC-CHAP negotiation (which generates the key, so it _must not_ be set), + * and once after the negotiation (which uses the key, so it _must_ be set). + */ +static bool nvme_tcp_key_revoke_needed(struct nvme_ctrl *ctrl) +{ + return ctrl->opts->concat && ctrl->opts->tls_key && ctrl->tls_pskid; +} + static int nvme_tcp_setup_ctrl(struct nvme_ctrl *ctrl, bool new) { struct nvmf_ctrl_options *opts = ctrl->opts; @@ -2036,6 +2370,16 @@ static int nvme_tcp_setup_ctrl(struct nvme_ctrl *ctrl, bool new) if (ret) return ret; + if (ctrl->opts->concat && !ctrl->tls_pskid) { + /* See comments for nvme_tcp_key_revoke_needed() */ + dev_dbg(ctrl->device, "restart admin queue for secure concatenation\n"); + nvme_stop_keep_alive(ctrl); + nvme_tcp_teardown_admin_queue(ctrl, false); + ret = nvme_tcp_configure_admin_queue(ctrl, false); + if (ret) + goto destroy_admin; + } + if (ctrl->icdoff) { ret = -EOPNOTSUPP; dev_err(ctrl->device, "icdoff is not supported!\n"); @@ -2072,8 +2416,10 @@ static int nvme_tcp_setup_ctrl(struct nvme_ctrl *ctrl, bool new) * unless we're during creation of a new controller to * avoid races with teardown flow. */ - WARN_ON_ONCE(ctrl->state != NVME_CTRL_DELETING && - ctrl->state != NVME_CTRL_DELETING_NOIO); + enum nvme_ctrl_state state = nvme_ctrl_state(ctrl); + + WARN_ON_ONCE(state != NVME_CTRL_DELETING && + state != NVME_CTRL_DELETING_NOIO); WARN_ON_ONCE(new); ret = -EINVAL; goto destroy_io; @@ -2088,14 +2434,13 @@ destroy_io: nvme_sync_io_queues(ctrl); nvme_tcp_stop_io_queues(ctrl); nvme_cancel_tagset(ctrl); - nvme_tcp_destroy_io_queues(ctrl, new); + if (new) + nvme_remove_io_tag_set(ctrl); + nvme_tcp_free_io_queues(ctrl); } destroy_admin: - nvme_quiesce_admin_queue(ctrl); - blk_sync_queue(ctrl->admin_q); - nvme_tcp_stop_queue(ctrl, 0); - nvme_cancel_admin_tagset(ctrl); - nvme_tcp_destroy_admin_queue(ctrl, new); + nvme_stop_keep_alive(ctrl); + nvme_tcp_teardown_admin_queue(ctrl, new); return ret; } @@ -2104,23 +2449,25 @@ static void nvme_tcp_reconnect_ctrl_work(struct work_struct *work) struct nvme_tcp_ctrl *tcp_ctrl = container_of(to_delayed_work(work), struct nvme_tcp_ctrl, connect_work); struct nvme_ctrl *ctrl = &tcp_ctrl->ctrl; + int ret; ++ctrl->nr_reconnects; - if (nvme_tcp_setup_ctrl(ctrl, false)) + ret = nvme_tcp_setup_ctrl(ctrl, false); + if (ret) goto requeue; - dev_info(ctrl->device, "Successfully reconnected (%d attempt)\n", - ctrl->nr_reconnects); + dev_info(ctrl->device, "Successfully reconnected (attempt %d/%d)\n", + ctrl->nr_reconnects, ctrl->opts->max_reconnects); ctrl->nr_reconnects = 0; return; requeue: - dev_info(ctrl->device, "Failed reconnect attempt %d\n", - ctrl->nr_reconnects); - nvme_tcp_reconnect_or_remove(ctrl); + dev_info(ctrl->device, "Failed reconnect attempt %d/%d\n", + ctrl->nr_reconnects, ctrl->opts->max_reconnects); + nvme_tcp_reconnect_or_remove(ctrl, ret); } static void nvme_tcp_error_recovery_work(struct work_struct *work) @@ -2129,6 +2476,8 @@ static void nvme_tcp_error_recovery_work(struct work_struct *work) struct nvme_tcp_ctrl, err_work); struct nvme_ctrl *ctrl = &tcp_ctrl->ctrl; + if (nvme_tcp_key_revoke_needed(ctrl)) + nvme_auth_revoke_tls_key(ctrl); nvme_stop_keep_alive(ctrl); flush_work(&ctrl->async_event_work); nvme_tcp_teardown_io_queues(ctrl, false); @@ -2140,12 +2489,14 @@ static void nvme_tcp_error_recovery_work(struct work_struct *work) if (!nvme_change_ctrl_state(ctrl, NVME_CTRL_CONNECTING)) { /* state change failure is ok if we started ctrl delete */ - WARN_ON_ONCE(ctrl->state != NVME_CTRL_DELETING && - ctrl->state != NVME_CTRL_DELETING_NOIO); + enum nvme_ctrl_state state = nvme_ctrl_state(ctrl); + + WARN_ON_ONCE(state != NVME_CTRL_DELETING && + state != NVME_CTRL_DELETING_NOIO); return; } - nvme_tcp_reconnect_or_remove(ctrl); + nvme_tcp_reconnect_or_remove(ctrl, 0); } static void nvme_tcp_teardown_ctrl(struct nvme_ctrl *ctrl, bool shutdown) @@ -2165,25 +2516,31 @@ static void nvme_reset_ctrl_work(struct work_struct *work) { struct nvme_ctrl *ctrl = container_of(work, struct nvme_ctrl, reset_work); + int ret; + if (nvme_tcp_key_revoke_needed(ctrl)) + nvme_auth_revoke_tls_key(ctrl); nvme_stop_ctrl(ctrl); nvme_tcp_teardown_ctrl(ctrl, false); if (!nvme_change_ctrl_state(ctrl, NVME_CTRL_CONNECTING)) { /* state change failure is ok if we started ctrl delete */ - WARN_ON_ONCE(ctrl->state != NVME_CTRL_DELETING && - ctrl->state != NVME_CTRL_DELETING_NOIO); + enum nvme_ctrl_state state = nvme_ctrl_state(ctrl); + + WARN_ON_ONCE(state != NVME_CTRL_DELETING && + state != NVME_CTRL_DELETING_NOIO); return; } - if (nvme_tcp_setup_ctrl(ctrl, false)) + ret = nvme_tcp_setup_ctrl(ctrl, false); + if (ret) goto out_fail; return; out_fail: ++ctrl->nr_reconnects; - nvme_tcp_reconnect_or_remove(ctrl); + nvme_tcp_reconnect_or_remove(ctrl, ret); } static void nvme_tcp_stop_ctrl(struct nvme_ctrl *ctrl) @@ -2264,8 +2621,10 @@ static void nvme_tcp_submit_async_event(struct nvme_ctrl *arg) ctrl->async_req.offset = 0; ctrl->async_req.curr_bio = NULL; ctrl->async_req.data_len = 0; + init_llist_node(&ctrl->async_req.lentry); + INIT_LIST_HEAD(&ctrl->async_req.entry); - nvme_tcp_queue_request(&ctrl->async_req, true, true); + nvme_tcp_queue_request(&ctrl->async_req, true); } static void nvme_tcp_complete_timed_out(struct request *rq) @@ -2281,13 +2640,16 @@ static enum blk_eh_timer_return nvme_tcp_timeout(struct request *rq) { struct nvme_tcp_request *req = blk_mq_rq_to_pdu(rq); struct nvme_ctrl *ctrl = &req->queue->ctrl->ctrl; - struct nvme_tcp_cmd_pdu *pdu = req->pdu; + struct nvme_tcp_cmd_pdu *pdu = nvme_tcp_req_cmd_pdu(req); + struct nvme_command *cmd = &pdu->cmd; + int qid = nvme_tcp_queue_id(req->queue); dev_warn(ctrl->device, - "queue %d: timeout request %#x type %d\n", - nvme_tcp_queue_id(req->queue), rq->tag, pdu->hdr.type); + "I/O tag %d (%04x) type %d opcode %#x (%s) QID %d timeout\n", + rq->tag, nvme_cid(rq), pdu->hdr.type, cmd->common.opcode, + nvme_fabrics_opcode_str(qid, cmd), qid); - if (ctrl->state != NVME_CTRL_LIVE) { + if (nvme_ctrl_state(ctrl) != NVME_CTRL_LIVE) { /* * If we are resetting, connecting or deleting we should * complete immediately because we may block controller @@ -2317,7 +2679,7 @@ static blk_status_t nvme_tcp_map_data(struct nvme_tcp_queue *queue, struct request *rq) { struct nvme_tcp_request *req = blk_mq_rq_to_pdu(rq); - struct nvme_tcp_cmd_pdu *pdu = req->pdu; + struct nvme_tcp_cmd_pdu *pdu = nvme_tcp_req_cmd_pdu(req); struct nvme_command *c = &pdu->cmd; c->common.flags |= NVME_CMD_SGL_METABUF; @@ -2337,7 +2699,7 @@ static blk_status_t nvme_tcp_setup_cmd_pdu(struct nvme_ns *ns, struct request *rq) { struct nvme_tcp_request *req = blk_mq_rq_to_pdu(rq); - struct nvme_tcp_cmd_pdu *pdu = req->pdu; + struct nvme_tcp_cmd_pdu *pdu = nvme_tcp_req_cmd_pdu(req); struct nvme_tcp_queue *queue = req->queue; u8 hdgst = nvme_tcp_hdgst_len(queue), ddgst = 0; blk_status_t ret; @@ -2414,7 +2776,7 @@ static blk_status_t nvme_tcp_queue_rq(struct blk_mq_hw_ctx *hctx, nvme_start_request(rq); - nvme_tcp_queue_request(req, true, bd->last); + nvme_tcp_queue_request(req, bd->last); return BLK_STS_OK; } @@ -2422,50 +2784,15 @@ static blk_status_t nvme_tcp_queue_rq(struct blk_mq_hw_ctx *hctx, static void nvme_tcp_map_queues(struct blk_mq_tag_set *set) { struct nvme_tcp_ctrl *ctrl = to_tcp_ctrl(set->driver_data); - struct nvmf_ctrl_options *opts = ctrl->ctrl.opts; - - if (opts->nr_write_queues && ctrl->io_queues[HCTX_TYPE_READ]) { - /* separate read/write queues */ - set->map[HCTX_TYPE_DEFAULT].nr_queues = - ctrl->io_queues[HCTX_TYPE_DEFAULT]; - set->map[HCTX_TYPE_DEFAULT].queue_offset = 0; - set->map[HCTX_TYPE_READ].nr_queues = - ctrl->io_queues[HCTX_TYPE_READ]; - set->map[HCTX_TYPE_READ].queue_offset = - ctrl->io_queues[HCTX_TYPE_DEFAULT]; - } else { - /* shared read/write queues */ - set->map[HCTX_TYPE_DEFAULT].nr_queues = - ctrl->io_queues[HCTX_TYPE_DEFAULT]; - set->map[HCTX_TYPE_DEFAULT].queue_offset = 0; - set->map[HCTX_TYPE_READ].nr_queues = - ctrl->io_queues[HCTX_TYPE_DEFAULT]; - set->map[HCTX_TYPE_READ].queue_offset = 0; - } - blk_mq_map_queues(&set->map[HCTX_TYPE_DEFAULT]); - blk_mq_map_queues(&set->map[HCTX_TYPE_READ]); - - if (opts->nr_poll_queues && ctrl->io_queues[HCTX_TYPE_POLL]) { - /* map dedicated poll queues only if we have queues left */ - set->map[HCTX_TYPE_POLL].nr_queues = - ctrl->io_queues[HCTX_TYPE_POLL]; - set->map[HCTX_TYPE_POLL].queue_offset = - ctrl->io_queues[HCTX_TYPE_DEFAULT] + - ctrl->io_queues[HCTX_TYPE_READ]; - blk_mq_map_queues(&set->map[HCTX_TYPE_POLL]); - } - - dev_info(ctrl->ctrl.device, - "mapped %d/%d/%d default/read/poll queues.\n", - ctrl->io_queues[HCTX_TYPE_DEFAULT], - ctrl->io_queues[HCTX_TYPE_READ], - ctrl->io_queues[HCTX_TYPE_POLL]); + + nvmf_map_queues(set, &ctrl->ctrl, ctrl->io_queues); } static int nvme_tcp_poll(struct blk_mq_hw_ctx *hctx, struct io_comp_batch *iob) { struct nvme_tcp_queue *queue = hctx->driver_data; struct sock *sk = queue->sock->sk; + int ret; if (!test_bit(NVME_TCP_Q_LIVE, &queue->flags)) return 0; @@ -2473,9 +2800,9 @@ static int nvme_tcp_poll(struct blk_mq_hw_ctx *hctx, struct io_comp_batch *iob) set_bit(NVME_TCP_Q_POLLING, &queue->flags); if (sk_can_busy_loop(sk) && skb_queue_empty_lockless(&sk->sk_receive_queue)) sk_busy_loop(sk, true); - nvme_tcp_try_recv(queue); + ret = nvme_tcp_try_recv(queue); clear_bit(NVME_TCP_Q_POLLING, &queue->flags); - return queue->nr_cqe; + return ret < 0 ? ret : queue->nr_cqe; } static int nvme_tcp_get_address(struct nvme_ctrl *ctrl, char *buf, int size) @@ -2486,6 +2813,11 @@ static int nvme_tcp_get_address(struct nvme_ctrl *ctrl, char *buf, int size) len = nvmf_get_address(ctrl, buf, size); + if (!test_bit(NVME_TCP_Q_LIVE, &queue->flags)) + return len; + + mutex_lock(&queue->queue_lock); + ret = kernel_getsockname(queue->sock, (struct sockaddr *)&src_addr); if (ret > 0) { if (len > 0) @@ -2494,6 +2826,8 @@ static int nvme_tcp_get_address(struct nvme_ctrl *ctrl, char *buf, int size) (len) ? "," : "", &src_addr); } + mutex_unlock(&queue->queue_lock); + return len; } @@ -2525,11 +2859,13 @@ static const struct nvme_ctrl_ops nvme_tcp_ctrl_ops = { .reg_read32 = nvmf_reg_read32, .reg_read64 = nvmf_reg_read64, .reg_write32 = nvmf_reg_write32, + .subsystem_reset = nvmf_subsystem_reset, .free_ctrl = nvme_tcp_free_ctrl, .submit_async_event = nvme_tcp_submit_async_event, .delete_ctrl = nvme_tcp_delete_ctrl, .get_address = nvme_tcp_get_address, .stop_ctrl = nvme_tcp_stop_ctrl, + .get_virt_boundary = nvmf_get_virt_boundary, }; static bool @@ -2549,7 +2885,7 @@ nvme_tcp_existing_controller(struct nvmf_ctrl_options *opts) return found; } -static struct nvme_ctrl *nvme_tcp_create_ctrl(struct device *dev, +static struct nvme_tcp_ctrl *nvme_tcp_alloc_ctrl(struct device *dev, struct nvmf_ctrl_options *opts) { struct nvme_tcp_ctrl *ctrl; @@ -2624,6 +2960,28 @@ static struct nvme_ctrl *nvme_tcp_create_ctrl(struct device *dev, if (ret) goto out_kfree_queues; + return ctrl; +out_kfree_queues: + kfree(ctrl->queues); +out_free_ctrl: + kfree(ctrl); + return ERR_PTR(ret); +} + +static struct nvme_ctrl *nvme_tcp_create_ctrl(struct device *dev, + struct nvmf_ctrl_options *opts) +{ + struct nvme_tcp_ctrl *ctrl; + int ret; + + ctrl = nvme_tcp_alloc_ctrl(dev, opts); + if (IS_ERR(ctrl)) + return ERR_CAST(ctrl); + + ret = nvme_add_ctrl(&ctrl->ctrl); + if (ret) + goto out_put_ctrl; + if (!nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_CONNECTING)) { WARN_ON_ONCE(1); ret = -EINTR; @@ -2634,8 +2992,8 @@ static struct nvme_ctrl *nvme_tcp_create_ctrl(struct device *dev, if (ret) goto out_uninit_ctrl; - dev_info(ctrl->ctrl.device, "new ctrl: NQN \"%s\", addr %pISp\n", - nvmf_ctrl_subsysnqn(&ctrl->ctrl), &ctrl->addr); + dev_info(ctrl->ctrl.device, "new ctrl: NQN \"%s\", addr %pISp, hostnqn: %s\n", + nvmf_ctrl_subsysnqn(&ctrl->ctrl), &ctrl->addr, opts->host->nqn); mutex_lock(&nvme_tcp_ctrl_mutex); list_add_tail(&ctrl->list, &nvme_tcp_ctrl_list); @@ -2645,15 +3003,11 @@ static struct nvme_ctrl *nvme_tcp_create_ctrl(struct device *dev, out_uninit_ctrl: nvme_uninit_ctrl(&ctrl->ctrl); +out_put_ctrl: nvme_put_ctrl(&ctrl->ctrl); if (ret > 0) ret = -EIO; return ERR_PTR(ret); -out_kfree_queues: - kfree(ctrl->queues); -out_free_ctrl: - kfree(ctrl); - return ERR_PTR(ret); } static struct nvmf_transport_ops nvme_tcp_transport = { @@ -2664,17 +3018,35 @@ static struct nvmf_transport_ops nvme_tcp_transport = { NVMF_OPT_HOST_TRADDR | NVMF_OPT_CTRL_LOSS_TMO | NVMF_OPT_HDR_DIGEST | NVMF_OPT_DATA_DIGEST | NVMF_OPT_NR_WRITE_QUEUES | NVMF_OPT_NR_POLL_QUEUES | - NVMF_OPT_TOS | NVMF_OPT_HOST_IFACE, + NVMF_OPT_TOS | NVMF_OPT_HOST_IFACE | NVMF_OPT_TLS | + NVMF_OPT_KEYRING | NVMF_OPT_TLS_KEY | NVMF_OPT_CONCAT, .create_ctrl = nvme_tcp_create_ctrl, }; static int __init nvme_tcp_init_module(void) { - nvme_tcp_wq = alloc_workqueue("nvme_tcp_wq", - WQ_MEM_RECLAIM | WQ_HIGHPRI, 0); + unsigned int wq_flags = WQ_MEM_RECLAIM | WQ_HIGHPRI | WQ_SYSFS; + int cpu; + + BUILD_BUG_ON(sizeof(struct nvme_tcp_hdr) != 8); + BUILD_BUG_ON(sizeof(struct nvme_tcp_cmd_pdu) != 72); + BUILD_BUG_ON(sizeof(struct nvme_tcp_data_pdu) != 24); + BUILD_BUG_ON(sizeof(struct nvme_tcp_rsp_pdu) != 24); + BUILD_BUG_ON(sizeof(struct nvme_tcp_r2t_pdu) != 24); + BUILD_BUG_ON(sizeof(struct nvme_tcp_icreq_pdu) != 128); + BUILD_BUG_ON(sizeof(struct nvme_tcp_icresp_pdu) != 128); + BUILD_BUG_ON(sizeof(struct nvme_tcp_term_pdu) != 24); + + if (wq_unbound) + wq_flags |= WQ_UNBOUND; + + nvme_tcp_wq = alloc_workqueue("nvme_tcp_wq", wq_flags, 0); if (!nvme_tcp_wq) return -ENOMEM; + for_each_possible_cpu(cpu) + atomic_set(&nvme_tcp_cpu_queues[cpu], 0); + nvmf_register_transport(&nvme_tcp_transport); return 0; } @@ -2697,4 +3069,5 @@ static void __exit nvme_tcp_cleanup_module(void) module_init(nvme_tcp_init_module); module_exit(nvme_tcp_cleanup_module); +MODULE_DESCRIPTION("NVMe host TCP transport driver"); MODULE_LICENSE("GPL v2"); diff --git a/drivers/nvme/host/trace.c b/drivers/nvme/host/trace.c index 1c36fcedea20..ad25ad1e4041 100644 --- a/drivers/nvme/host/trace.c +++ b/drivers/nvme/host/trace.c @@ -4,7 +4,7 @@ * Copyright (c) 2018 Johannes Thumshirn, SUSE Linux GmbH */ -#include <asm/unaligned.h> +#include <linux/unaligned.h> #include "trace.h" static const char *nvme_trace_delete_sq(struct trace_seq *p, u8 *cdw10) @@ -119,7 +119,10 @@ static const char *nvme_trace_get_lba_status(struct trace_seq *p, static const char *nvme_trace_admin_format_nvm(struct trace_seq *p, u8 *cdw10) { const char *ret = trace_seq_buffer_ptr(p); - u8 lbaf = cdw10[0] & 0xF; + /* + * lbafu(bit 13:12) is already in the upper 4 bits, lbafl: bit 03:00. + */ + u8 lbaf = (cdw10[1] & 0x30) | (cdw10[0] & 0xF); u8 mset = (cdw10[0] >> 4) & 0x1; u8 pi = (cdw10[0] >> 5) & 0x7; u8 pil = cdw10[1] & 0x1; @@ -164,12 +167,27 @@ static const char *nvme_trace_dsm(struct trace_seq *p, u8 *cdw10) static const char *nvme_trace_zone_mgmt_send(struct trace_seq *p, u8 *cdw10) { + static const char * const zsa_strs[] = { + [0x01] = "close zone", + [0x02] = "finish zone", + [0x03] = "open zone", + [0x04] = "reset zone", + [0x05] = "offline zone", + [0x10] = "set zone descriptor extension" + }; const char *ret = trace_seq_buffer_ptr(p); u64 slba = get_unaligned_le64(cdw10); + const char *zsa_str; u8 zsa = cdw10[12]; u8 all = cdw10[13]; - trace_seq_printf(p, "slba=%llu, zsa=%u, all=%u", slba, zsa, all); + if (zsa < ARRAY_SIZE(zsa_strs) && zsa_strs[zsa]) + zsa_str = zsa_strs[zsa]; + else + zsa_str = "reserved"; + + trace_seq_printf(p, "slba=%llu, zsa=%u:%s, all=%u", + slba, zsa, zsa_str, all); trace_seq_putc(p, 0); return ret; @@ -177,15 +195,132 @@ static const char *nvme_trace_zone_mgmt_send(struct trace_seq *p, u8 *cdw10) static const char *nvme_trace_zone_mgmt_recv(struct trace_seq *p, u8 *cdw10) { + static const char * const zrasf_strs[] = { + [0x00] = "list all zones", + [0x01] = "list the zones in the ZSE: Empty state", + [0x02] = "list the zones in the ZSIO: Implicitly Opened state", + [0x03] = "list the zones in the ZSEO: Explicitly Opened state", + [0x04] = "list the zones in the ZSC: Closed state", + [0x05] = "list the zones in the ZSF: Full state", + [0x06] = "list the zones in the ZSRO: Read Only state", + [0x07] = "list the zones in the ZSO: Offline state", + [0x09] = "list the zones that have the zone attribute" + }; const char *ret = trace_seq_buffer_ptr(p); u64 slba = get_unaligned_le64(cdw10); u32 numd = get_unaligned_le32(cdw10 + 8); u8 zra = cdw10[12]; u8 zrasf = cdw10[13]; + const char *zrasf_str; u8 pr = cdw10[14]; - trace_seq_printf(p, "slba=%llu, numd=%u, zra=%u, zrasf=%u, pr=%u", - slba, numd, zra, zrasf, pr); + if (zrasf < ARRAY_SIZE(zrasf_strs) && zrasf_strs[zrasf]) + zrasf_str = zrasf_strs[zrasf]; + else + zrasf_str = "reserved"; + + trace_seq_printf(p, "slba=%llu, numd=%u, zra=%u, zrasf=%u:%s, pr=%u", + slba, numd, zra, zrasf, zrasf_str, pr); + trace_seq_putc(p, 0); + + return ret; +} + +static const char *nvme_trace_resv_reg(struct trace_seq *p, u8 *cdw10) +{ + static const char * const rrega_strs[] = { + [0x00] = "register", + [0x01] = "unregister", + [0x02] = "replace", + }; + const char *ret = trace_seq_buffer_ptr(p); + u8 rrega = cdw10[0] & 0x7; + u8 iekey = (cdw10[0] >> 3) & 0x1; + u8 ptpl = (cdw10[3] >> 6) & 0x3; + const char *rrega_str; + + if (rrega < ARRAY_SIZE(rrega_strs) && rrega_strs[rrega]) + rrega_str = rrega_strs[rrega]; + else + rrega_str = "reserved"; + + trace_seq_printf(p, "rrega=%u:%s, iekey=%u, ptpl=%u", + rrega, rrega_str, iekey, ptpl); + trace_seq_putc(p, 0); + + return ret; +} + +static const char * const rtype_strs[] = { + [0x00] = "reserved", + [0x01] = "write exclusive", + [0x02] = "exclusive access", + [0x03] = "write exclusive registrants only", + [0x04] = "exclusive access registrants only", + [0x05] = "write exclusive all registrants", + [0x06] = "exclusive access all registrants", +}; + +static const char *nvme_trace_resv_acq(struct trace_seq *p, u8 *cdw10) +{ + static const char * const racqa_strs[] = { + [0x00] = "acquire", + [0x01] = "preempt", + [0x02] = "preempt and abort", + }; + const char *ret = trace_seq_buffer_ptr(p); + u8 racqa = cdw10[0] & 0x7; + u8 iekey = (cdw10[0] >> 3) & 0x1; + u8 rtype = cdw10[1]; + const char *racqa_str = "reserved"; + const char *rtype_str = "reserved"; + + if (racqa < ARRAY_SIZE(racqa_strs) && racqa_strs[racqa]) + racqa_str = racqa_strs[racqa]; + + if (rtype < ARRAY_SIZE(rtype_strs) && rtype_strs[rtype]) + rtype_str = rtype_strs[rtype]; + + trace_seq_printf(p, "racqa=%u:%s, iekey=%u, rtype=%u:%s", + racqa, racqa_str, iekey, rtype, rtype_str); + trace_seq_putc(p, 0); + + return ret; +} + +static const char *nvme_trace_resv_rel(struct trace_seq *p, u8 *cdw10) +{ + static const char * const rrela_strs[] = { + [0x00] = "release", + [0x01] = "clear", + }; + const char *ret = trace_seq_buffer_ptr(p); + u8 rrela = cdw10[0] & 0x7; + u8 iekey = (cdw10[0] >> 3) & 0x1; + u8 rtype = cdw10[1]; + const char *rrela_str = "reserved"; + const char *rtype_str = "reserved"; + + if (rrela < ARRAY_SIZE(rrela_strs) && rrela_strs[rrela]) + rrela_str = rrela_strs[rrela]; + + if (rtype < ARRAY_SIZE(rtype_strs) && rtype_strs[rtype]) + rtype_str = rtype_strs[rtype]; + + trace_seq_printf(p, "rrela=%u:%s, iekey=%u, rtype=%u:%s", + rrela, rrela_str, iekey, rtype, rtype_str); + trace_seq_putc(p, 0); + + return ret; +} + +static const char *nvme_trace_resv_report(struct trace_seq *p, u8 *cdw10) +{ + const char *ret = trace_seq_buffer_ptr(p); + u32 numd = get_unaligned_le32(cdw10); + u8 eds = cdw10[4] & 0x1; + + trace_seq_printf(p, "numd=%u, eds=%u", numd, eds); trace_seq_putc(p, 0); return ret; @@ -243,6 +378,14 @@ const char *nvme_trace_parse_nvm_cmd(struct trace_seq *p, return nvme_trace_zone_mgmt_send(p, cdw10); case nvme_cmd_zone_mgmt_recv: return nvme_trace_zone_mgmt_recv(p, cdw10); + case nvme_cmd_resv_register: + return nvme_trace_resv_reg(p, cdw10); + case nvme_cmd_resv_acquire: + return nvme_trace_resv_acq(p, cdw10); + case nvme_cmd_resv_release: + return nvme_trace_resv_rel(p, cdw10); + case nvme_cmd_resv_report: + return nvme_trace_resv_report(p, cdw10); default: return nvme_trace_common(p, cdw10); } diff --git a/drivers/nvme/host/trace.h b/drivers/nvme/host/trace.h index 6f0eaf6a1528..4fb5922ffdac 100644 --- a/drivers/nvme/host/trace.h +++ b/drivers/nvme/host/trace.h @@ -127,15 +127,12 @@ TRACE_EVENT(nvme_async_event, ), TP_printk("nvme%d: NVME_AEN=%#08x [%s]", __entry->ctrl_id, __entry->result, - __print_symbolic(__entry->result, - aer_name(NVME_AER_NOTICE_NS_CHANGED), - aer_name(NVME_AER_NOTICE_ANA), - aer_name(NVME_AER_NOTICE_FW_ACT_STARTING), - aer_name(NVME_AER_NOTICE_DISC_CHANGED), - aer_name(NVME_AER_ERROR), - aer_name(NVME_AER_SMART), - aer_name(NVME_AER_CSS), - aer_name(NVME_AER_VS)) + __print_symbolic(__entry->result & 0x7, + aer_name(NVME_AER_ERROR), + aer_name(NVME_AER_SMART), + aer_name(NVME_AER_NOTICE), + aer_name(NVME_AER_CSS), + aer_name(NVME_AER_VS)) ) ); diff --git a/drivers/nvme/host/zns.c b/drivers/nvme/host/zns.c index 12316ab51bda..deea2dbef5b8 100644 --- a/drivers/nvme/host/zns.c +++ b/drivers/nvme/host/zns.c @@ -7,17 +7,6 @@ #include <linux/vmalloc.h> #include "nvme.h" -int nvme_revalidate_zones(struct nvme_ns *ns) -{ - struct request_queue *q = ns->queue; - int ret; - - ret = blk_revalidate_disk_zones(ns->disk, NULL); - if (!ret) - blk_queue_max_zone_append_sectors(q, ns->ctrl->max_zone_append); - return ret; -} - static int nvme_set_max_append(struct nvme_ctrl *ctrl) { struct nvme_command c = { }; @@ -46,10 +35,10 @@ static int nvme_set_max_append(struct nvme_ctrl *ctrl) return 0; } -int nvme_update_zone_info(struct nvme_ns *ns, unsigned lbaf) +int nvme_query_zone_info(struct nvme_ns *ns, unsigned lbaf, + struct nvme_zone_info *zi) { struct nvme_effects_log *log = ns->head->effects; - struct request_queue *q = ns->queue; struct nvme_command c = { }; struct nvme_id_ns_zns *id; int status; @@ -100,24 +89,33 @@ int nvme_update_zone_info(struct nvme_ns *ns, unsigned lbaf) goto free_data; } - ns->zsze = nvme_lba_to_sect(ns, le64_to_cpu(id->lbafe[lbaf].zsze)); - if (!is_power_of_2(ns->zsze)) { + zi->zone_size = le64_to_cpu(id->lbafe[lbaf].zsze); + if (!is_power_of_2(zi->zone_size)) { dev_warn(ns->ctrl->device, - "invalid zone size:%llu for namespace:%u\n", - ns->zsze, ns->head->ns_id); + "invalid zone size: %llu for namespace: %u\n", + zi->zone_size, ns->head->ns_id); status = -ENODEV; goto free_data; } + zi->max_open_zones = le32_to_cpu(id->mor) + 1; + zi->max_active_zones = le32_to_cpu(id->mar) + 1; - disk_set_zoned(ns->disk, BLK_ZONED_HM); - blk_queue_flag_set(QUEUE_FLAG_ZONE_RESETALL, q); - disk_set_max_open_zones(ns->disk, le32_to_cpu(id->mor) + 1); - disk_set_max_active_zones(ns->disk, le32_to_cpu(id->mar) + 1); free_data: kfree(id); return status; } +void nvme_update_zone_info(struct nvme_ns *ns, struct queue_limits *lim, + struct nvme_zone_info *zi) +{ + lim->features |= BLK_FEAT_ZONED; + lim->max_open_zones = zi->max_open_zones; + lim->max_active_zones = zi->max_active_zones; + lim->max_hw_zone_append_sectors = ns->ctrl->max_zone_append; + lim->chunk_sectors = ns->head->zsze = + nvme_lba_to_sect(ns->head, zi->zone_size); +} + static void *nvme_zns_alloc_report_buffer(struct nvme_ns *ns, unsigned int nr_zones, size_t *buflen) { @@ -129,7 +127,7 @@ static void *nvme_zns_alloc_report_buffer(struct nvme_ns *ns, sizeof(struct nvme_zone_descriptor); nr_zones = min_t(unsigned int, nr_zones, - get_capacity(ns->disk) >> ilog2(ns->zsze)); + get_capacity(ns->disk) >> ilog2(ns->head->zsze)); bufsize = sizeof(struct nvme_zone_report) + nr_zones * sizeof(struct nvme_zone_descriptor); @@ -150,32 +148,32 @@ static void *nvme_zns_alloc_report_buffer(struct nvme_ns *ns, static int nvme_zone_parse_entry(struct nvme_ns *ns, struct nvme_zone_descriptor *entry, - unsigned int idx, report_zones_cb cb, - void *data) + unsigned int idx, + struct blk_report_zones_args *args) { + struct nvme_ns_head *head = ns->head; struct blk_zone zone = { }; if ((entry->zt & 0xf) != NVME_ZONE_TYPE_SEQWRITE_REQ) { - dev_err(ns->ctrl->device, "invalid zone type %#x\n", - entry->zt); + dev_err(ns->ctrl->device, "invalid zone type %#x\n", entry->zt); return -EINVAL; } zone.type = BLK_ZONE_TYPE_SEQWRITE_REQ; zone.cond = entry->zs >> 4; - zone.len = ns->zsze; - zone.capacity = nvme_lba_to_sect(ns, le64_to_cpu(entry->zcap)); - zone.start = nvme_lba_to_sect(ns, le64_to_cpu(entry->zslba)); + zone.len = head->zsze; + zone.capacity = nvme_lba_to_sect(head, le64_to_cpu(entry->zcap)); + zone.start = nvme_lba_to_sect(head, le64_to_cpu(entry->zslba)); if (zone.cond == BLK_ZONE_COND_FULL) zone.wp = zone.start + zone.len; else - zone.wp = nvme_lba_to_sect(ns, le64_to_cpu(entry->wp)); + zone.wp = nvme_lba_to_sect(head, le64_to_cpu(entry->wp)); - return cb(&zone, idx, data); + return disk_report_zone(ns->disk, &zone, idx, args); } int nvme_ns_report_zones(struct nvme_ns *ns, sector_t sector, - unsigned int nr_zones, report_zones_cb cb, void *data) + unsigned int nr_zones, struct blk_report_zones_args *args) { struct nvme_zone_report *report; struct nvme_command c = { }; @@ -197,11 +195,11 @@ int nvme_ns_report_zones(struct nvme_ns *ns, sector_t sector, c.zmr.zrasf = NVME_ZRASF_ZONE_REPORT_ALL; c.zmr.pr = NVME_REPORT_ZONE_PARTIAL; - sector &= ~(ns->zsze - 1); + sector &= ~(ns->head->zsze - 1); while (zone_idx < nr_zones && sector < get_capacity(ns->disk)) { memset(report, 0, buflen); - c.zmr.slba = cpu_to_le64(nvme_sect_to_lba(ns, sector)); + c.zmr.slba = cpu_to_le64(nvme_sect_to_lba(ns->head, sector)); ret = nvme_submit_sync_cmd(ns->queue, &c, report, buflen); if (ret) { if (ret > 0) @@ -215,13 +213,13 @@ int nvme_ns_report_zones(struct nvme_ns *ns, sector_t sector, for (i = 0; i < nz && zone_idx < nr_zones; i++) { ret = nvme_zone_parse_entry(ns, &report->entries[i], - zone_idx, cb, data); + zone_idx, args); if (ret) goto out_free; zone_idx++; } - sector += ns->zsze * nz; + sector += ns->head->zsze * nz; } if (zone_idx > 0) @@ -240,7 +238,7 @@ blk_status_t nvme_setup_zone_mgmt_send(struct nvme_ns *ns, struct request *req, c->zms.opcode = nvme_cmd_zone_mgmt_send; c->zms.nsid = cpu_to_le32(ns->head->ns_id); - c->zms.slba = cpu_to_le64(nvme_sect_to_lba(ns, blk_rq_pos(req))); + c->zms.slba = cpu_to_le64(nvme_sect_to_lba(ns->head, blk_rq_pos(req))); c->zms.zsa = action; if (req_op(req) == REQ_OP_ZONE_RESET_ALL) diff --git a/drivers/nvme/target/Kconfig b/drivers/nvme/target/Kconfig index 79fc64035ee3..4904097dfd49 100644 --- a/drivers/nvme/target/Kconfig +++ b/drivers/nvme/target/Kconfig @@ -3,8 +3,9 @@ config NVME_TARGET tristate "NVMe Target support" depends on BLOCK - depends on CONFIGFS_FS - select BLK_DEV_INTEGRITY_T10 if BLK_DEV_INTEGRITY + select CONFIGFS_FS + select NVME_KEYRING if NVME_TARGET_TCP_TLS + select KEYS if NVME_TARGET_TCP_TLS select SGL_ALLOC help This enabled target side support for the NVMe protocol, that is @@ -16,6 +17,15 @@ config NVME_TARGET To configure the NVMe target you probably want to use the nvmetcli tool from http://git.infradead.org/users/hch/nvmetcli.git. +config NVME_TARGET_DEBUGFS + bool "NVMe Target debugfs support" + depends on NVME_TARGET + help + This enables debugfs support to display the connected controllers + to each subsystem + + If unsure, say N. + config NVME_TARGET_PASSTHRU bool "NVMe Target Passthrough support" depends on NVME_TARGET @@ -84,17 +94,36 @@ config NVME_TARGET_TCP If unsure, say N. +config NVME_TARGET_TCP_TLS + bool "NVMe over Fabrics TCP target TLS encryption support" + depends on NVME_TARGET_TCP + select NET_HANDSHAKE + select TLS + help + Enables TLS encryption for the NVMe TCP target using the netlink handshake API. + + The TLS handshake daemon is available at + https://github.com/oracle/ktls-utils. + + If unsure, say N. + config NVME_TARGET_AUTH - bool "NVMe over Fabrics In-band Authentication support" + bool "NVMe over Fabrics In-band Authentication in target side" depends on NVME_TARGET - select NVME_COMMON - select CRYPTO - select CRYPTO_HMAC - select CRYPTO_SHA256 - select CRYPTO_SHA512 - select CRYPTO_DH - select CRYPTO_DH_RFC7919_GROUPS + select NVME_AUTH + help + This enables support for NVMe over Fabrics In-band Authentication in + target side. + + If unsure, say N. + +config NVME_TARGET_PCI_EPF + tristate "NVMe PCI Endpoint Function target support" + depends on NVME_TARGET && PCI_ENDPOINT + depends on NVME_CORE=y || NVME_CORE=NVME_TARGET help - This enables support for NVMe over Fabrics In-band Authentication + This enables the NVMe PCI Endpoint Function target driver support, + which allows creating a NVMe PCI controller using an endpoint mode + capable PCI controller. If unsure, say N. diff --git a/drivers/nvme/target/Makefile b/drivers/nvme/target/Makefile index c66820102493..ed8522911d1f 100644 --- a/drivers/nvme/target/Makefile +++ b/drivers/nvme/target/Makefile @@ -8,9 +8,11 @@ obj-$(CONFIG_NVME_TARGET_RDMA) += nvmet-rdma.o obj-$(CONFIG_NVME_TARGET_FC) += nvmet-fc.o obj-$(CONFIG_NVME_TARGET_FCLOOP) += nvme-fcloop.o obj-$(CONFIG_NVME_TARGET_TCP) += nvmet-tcp.o +obj-$(CONFIG_NVME_TARGET_PCI_EPF) += nvmet-pci-epf.o nvmet-y += core.o configfs.o admin-cmd.o fabrics-cmd.o \ - discovery.o io-cmd-file.o io-cmd-bdev.o + discovery.o io-cmd-file.o io-cmd-bdev.o pr.o +nvmet-$(CONFIG_NVME_TARGET_DEBUGFS) += debugfs.o nvmet-$(CONFIG_NVME_TARGET_PASSTHRU) += passthru.o nvmet-$(CONFIG_BLK_DEV_ZONED) += zns.o nvmet-$(CONFIG_NVME_TARGET_AUTH) += fabrics-cmd-auth.o auth.o @@ -19,4 +21,5 @@ nvmet-rdma-y += rdma.o nvmet-fc-y += fc.o nvme-fcloop-y += fcloop.o nvmet-tcp-y += tcp.o +nvmet-pci-epf-y += pci-epf.o nvmet-$(CONFIG_TRACING) += trace.o diff --git a/drivers/nvme/target/admin-cmd.c b/drivers/nvme/target/admin-cmd.c index 6a54ed6fb121..3da31bb1183e 100644 --- a/drivers/nvme/target/admin-cmd.c +++ b/drivers/nvme/target/admin-cmd.c @@ -9,9 +9,136 @@ #include <linux/part_stat.h> #include <generated/utsrelease.h> -#include <asm/unaligned.h> +#include <linux/unaligned.h> #include "nvmet.h" +static void nvmet_execute_delete_sq(struct nvmet_req *req) +{ + struct nvmet_ctrl *ctrl = req->sq->ctrl; + u16 sqid = le16_to_cpu(req->cmd->delete_queue.qid); + u16 status; + + if (!nvmet_is_pci_ctrl(ctrl)) { + status = nvmet_report_invalid_opcode(req); + goto complete; + } + + if (!sqid) { + status = NVME_SC_QID_INVALID | NVME_STATUS_DNR; + goto complete; + } + + status = nvmet_check_sqid(ctrl, sqid, false); + if (status != NVME_SC_SUCCESS) + goto complete; + + status = ctrl->ops->delete_sq(ctrl, sqid); + +complete: + nvmet_req_complete(req, status); +} + +static void nvmet_execute_create_sq(struct nvmet_req *req) +{ + struct nvmet_ctrl *ctrl = req->sq->ctrl; + struct nvme_command *cmd = req->cmd; + u16 sqid = le16_to_cpu(cmd->create_sq.sqid); + u16 cqid = le16_to_cpu(cmd->create_sq.cqid); + u16 sq_flags = le16_to_cpu(cmd->create_sq.sq_flags); + u16 qsize = le16_to_cpu(cmd->create_sq.qsize); + u64 prp1 = le64_to_cpu(cmd->create_sq.prp1); + u16 status; + + if (!nvmet_is_pci_ctrl(ctrl)) { + status = nvmet_report_invalid_opcode(req); + goto complete; + } + + if (!sqid) { + status = NVME_SC_QID_INVALID | NVME_STATUS_DNR; + goto complete; + } + + status = nvmet_check_sqid(ctrl, sqid, true); + if (status != NVME_SC_SUCCESS) + goto complete; + + status = nvmet_check_io_cqid(ctrl, cqid, false); + if (status != NVME_SC_SUCCESS) { + pr_err("SQ %u: Invalid CQID %u\n", sqid, cqid); + goto complete; + } + + if (!qsize || qsize > NVME_CAP_MQES(ctrl->cap)) { + status = NVME_SC_QUEUE_SIZE | NVME_STATUS_DNR; + goto complete; + } + + status = ctrl->ops->create_sq(ctrl, sqid, cqid, sq_flags, qsize, prp1); + +complete: + nvmet_req_complete(req, status); +} + +static void nvmet_execute_delete_cq(struct nvmet_req *req) +{ + struct nvmet_ctrl *ctrl = req->sq->ctrl; + u16 cqid = le16_to_cpu(req->cmd->delete_queue.qid); + u16 status; + + if (!nvmet_is_pci_ctrl(ctrl)) { + status = nvmet_report_invalid_opcode(req); + goto complete; + } + + status = nvmet_check_io_cqid(ctrl, cqid, false); + if (status != NVME_SC_SUCCESS) + goto complete; + + if (!ctrl->cqs[cqid] || nvmet_cq_in_use(ctrl->cqs[cqid])) { + /* Some SQs are still using this CQ */ + status = NVME_SC_QID_INVALID | NVME_STATUS_DNR; + goto complete; + } + + status = ctrl->ops->delete_cq(ctrl, cqid); + +complete: + nvmet_req_complete(req, status); +} + +static void nvmet_execute_create_cq(struct nvmet_req *req) +{ + struct nvmet_ctrl *ctrl = req->sq->ctrl; + struct nvme_command *cmd = req->cmd; + u16 cqid = le16_to_cpu(cmd->create_cq.cqid); + u16 cq_flags = le16_to_cpu(cmd->create_cq.cq_flags); + u16 qsize = le16_to_cpu(cmd->create_cq.qsize); + u16 irq_vector = le16_to_cpu(cmd->create_cq.irq_vector); + u64 prp1 = le64_to_cpu(cmd->create_cq.prp1); + u16 status; + + if (!nvmet_is_pci_ctrl(ctrl)) { + status = nvmet_report_invalid_opcode(req); + goto complete; + } + + status = nvmet_check_io_cqid(ctrl, cqid, true); + if (status != NVME_SC_SUCCESS) + goto complete; + + if (!qsize || qsize > NVME_CAP_MQES(ctrl->cap)) { + status = NVME_SC_QUEUE_SIZE | NVME_STATUS_DNR; + goto complete; + } + + status = ctrl->ops->create_cq(ctrl, cqid, cq_flags, qsize, + prp1, irq_vector); + +complete: + nvmet_req_complete(req, status); +} + u32 nvmet_get_log_page_len(struct nvme_command *cmd) { u32 len = le16_to_cpu(cmd->get_log_page.numdu); @@ -71,6 +198,35 @@ static void nvmet_execute_get_log_page_error(struct nvmet_req *req) nvmet_req_complete(req, 0); } +static void nvmet_execute_get_supported_log_pages(struct nvmet_req *req) +{ + struct nvme_supported_log *logs; + u16 status; + + logs = kzalloc(sizeof(*logs), GFP_KERNEL); + if (!logs) { + status = NVME_SC_INTERNAL; + goto out; + } + + logs->lids[NVME_LOG_SUPPORTED] = cpu_to_le32(NVME_LIDS_LSUPP); + logs->lids[NVME_LOG_ERROR] = cpu_to_le32(NVME_LIDS_LSUPP); + logs->lids[NVME_LOG_SMART] = cpu_to_le32(NVME_LIDS_LSUPP); + logs->lids[NVME_LOG_FW_SLOT] = cpu_to_le32(NVME_LIDS_LSUPP); + logs->lids[NVME_LOG_CHANGED_NS] = cpu_to_le32(NVME_LIDS_LSUPP); + logs->lids[NVME_LOG_CMD_EFFECTS] = cpu_to_le32(NVME_LIDS_LSUPP); + logs->lids[NVME_LOG_ENDURANCE_GROUP] = cpu_to_le32(NVME_LIDS_LSUPP); + logs->lids[NVME_LOG_ANA] = cpu_to_le32(NVME_LIDS_LSUPP); + logs->lids[NVME_LOG_FEATURES] = cpu_to_le32(NVME_LIDS_LSUPP); + logs->lids[NVME_LOG_RMI] = cpu_to_le32(NVME_LIDS_LSUPP); + logs->lids[NVME_LOG_RESERVATION] = cpu_to_le32(NVME_LIDS_LSUPP); + + status = nvmet_copy_to_sgl(req, 0, logs, sizeof(*logs)); + kfree(logs); +out: + nvmet_req_complete(req, status); +} + static u16 nvmet_get_smart_log_nsid(struct nvmet_req *req, struct nvme_smart_log *slog) { @@ -110,7 +266,7 @@ static u16 nvmet_get_smart_log_all(struct nvmet_req *req, unsigned long idx; ctrl = req->sq->ctrl; - xa_for_each(&ctrl->subsys->namespaces, idx, ns) { + nvmet_for_each_enabled_ns(&ctrl->subsys->namespaces, idx, ns) { /* we don't have the right data for file backed ns */ if (!ns->bdev) continue; @@ -130,6 +286,45 @@ static u16 nvmet_get_smart_log_all(struct nvmet_req *req, return NVME_SC_SUCCESS; } +static void nvmet_execute_get_log_page_rmi(struct nvmet_req *req) +{ + struct nvme_rotational_media_log *log; + struct gendisk *disk; + u16 status; + + req->cmd->common.nsid = cpu_to_le32(le16_to_cpu( + req->cmd->get_log_page.lsi)); + status = nvmet_req_find_ns(req); + if (status) + goto out; + + if (!req->ns->bdev || bdev_nonrot(req->ns->bdev)) { + status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; + goto out; + } + + if (req->transfer_len != sizeof(*log)) { + status = NVME_SC_SGL_INVALID_DATA | NVME_STATUS_DNR; + goto out; + } + + log = kzalloc(sizeof(*log), GFP_KERNEL); + if (!log) + goto out; + + log->endgid = req->cmd->get_log_page.lsi; + disk = req->ns->bdev->bd_disk; + if (disk && disk->ia_ranges) + log->numa = cpu_to_le16(disk->ia_ranges->nr_ia_ranges); + else + log->numa = cpu_to_le16(1); + + status = nvmet_copy_to_sgl(req, 0, log, sizeof(*log)); + kfree(log); +out: + nvmet_req_complete(req, status); +} + static void nvmet_execute_get_log_page_smart(struct nvmet_req *req) { struct nvme_smart_log *log; @@ -162,8 +357,18 @@ out: nvmet_req_complete(req, status); } -static void nvmet_get_cmd_effects_nvm(struct nvme_effects_log *log) +static void nvmet_get_cmd_effects_admin(struct nvmet_ctrl *ctrl, + struct nvme_effects_log *log) { + /* For a PCI target controller, advertize support for the . */ + if (nvmet_is_pci_ctrl(ctrl)) { + log->acs[nvme_admin_delete_sq] = + log->acs[nvme_admin_create_sq] = + log->acs[nvme_admin_delete_cq] = + log->acs[nvme_admin_create_cq] = + cpu_to_le32(NVME_CMD_EFFECTS_CSUPP); + } + log->acs[nvme_admin_get_log_page] = log->acs[nvme_admin_identify] = log->acs[nvme_admin_abort_cmd] = @@ -172,10 +377,17 @@ static void nvmet_get_cmd_effects_nvm(struct nvme_effects_log *log) log->acs[nvme_admin_async_event] = log->acs[nvme_admin_keep_alive] = cpu_to_le32(NVME_CMD_EFFECTS_CSUPP); +} +static void nvmet_get_cmd_effects_nvm(struct nvme_effects_log *log) +{ log->iocs[nvme_cmd_read] = log->iocs[nvme_cmd_flush] = log->iocs[nvme_cmd_dsm] = + log->iocs[nvme_cmd_resv_acquire] = + log->iocs[nvme_cmd_resv_register] = + log->iocs[nvme_cmd_resv_release] = + log->iocs[nvme_cmd_resv_report] = cpu_to_le32(NVME_CMD_EFFECTS_CSUPP); log->iocs[nvme_cmd_write] = log->iocs[nvme_cmd_write_zeroes] = @@ -193,6 +405,7 @@ static void nvmet_get_cmd_effects_zns(struct nvme_effects_log *log) static void nvmet_execute_get_log_cmd_effects_ns(struct nvmet_req *req) { + struct nvmet_ctrl *ctrl = req->sq->ctrl; struct nvme_effects_log *log; u16 status = NVME_SC_SUCCESS; @@ -204,6 +417,7 @@ static void nvmet_execute_get_log_cmd_effects_ns(struct nvmet_req *req) switch (req->cmd->get_log_page.csi) { case NVME_CSI_NVM: + nvmet_get_cmd_effects_admin(ctrl, log); nvmet_get_cmd_effects_nvm(log); break; case NVME_CSI_ZNS: @@ -211,6 +425,7 @@ static void nvmet_execute_get_log_cmd_effects_ns(struct nvmet_req *req) status = NVME_SC_INVALID_IO_CMD_SET; goto free; } + nvmet_get_cmd_effects_admin(ctrl, log); nvmet_get_cmd_effects_nvm(log); nvmet_get_cmd_effects_zns(log); break; @@ -259,9 +474,10 @@ static u32 nvmet_format_ana_group(struct nvmet_req *req, u32 grpid, u32 count = 0; if (!(req->cmd->get_log_page.lsp & NVME_ANA_LOG_RGO)) { - xa_for_each(&ctrl->subsys->namespaces, idx, ns) + nvmet_for_each_enabled_ns(&ctrl->subsys->namespaces, idx, ns) { if (ns->anagrpid == grpid) desc->nsids[count++] = cpu_to_le32(ns->nsid); + } } desc->grpid = cpu_to_le32(grpid); @@ -272,6 +488,49 @@ static u32 nvmet_format_ana_group(struct nvmet_req *req, u32 grpid, return struct_size(desc, nsids, count); } +static void nvmet_execute_get_log_page_endgrp(struct nvmet_req *req) +{ + u64 host_reads, host_writes, data_units_read, data_units_written; + struct nvme_endurance_group_log *log; + u16 status; + + /* + * The target driver emulates each endurance group as its own + * namespace, reusing the nsid as the endurance group identifier. + */ + req->cmd->common.nsid = cpu_to_le32(le16_to_cpu( + req->cmd->get_log_page.lsi)); + status = nvmet_req_find_ns(req); + if (status) + goto out; + + log = kzalloc(sizeof(*log), GFP_KERNEL); + if (!log) { + status = NVME_SC_INTERNAL; + goto out; + } + + if (!req->ns->bdev) + goto copy; + + host_reads = part_stat_read(req->ns->bdev, ios[READ]); + data_units_read = + DIV_ROUND_UP(part_stat_read(req->ns->bdev, sectors[READ]), 1000); + host_writes = part_stat_read(req->ns->bdev, ios[WRITE]); + data_units_written = + DIV_ROUND_UP(part_stat_read(req->ns->bdev, sectors[WRITE]), 1000); + + put_unaligned_le64(host_reads, &log->hrc[0]); + put_unaligned_le64(data_units_read, &log->dur[0]); + put_unaligned_le64(host_writes, &log->hwc[0]); + put_unaligned_le64(data_units_written, &log->duw[0]); +copy: + status = nvmet_copy_to_sgl(req, 0, log, sizeof(*log)); + kfree(log); +out: + nvmet_req_complete(req, status); +} + static void nvmet_execute_get_log_page_ana(struct nvmet_req *req) { struct nvme_ana_rsp_hdr hdr = { 0, }; @@ -317,12 +576,44 @@ out: nvmet_req_complete(req, status); } +static void nvmet_execute_get_log_page_features(struct nvmet_req *req) +{ + struct nvme_supported_features_log *features; + u16 status; + + features = kzalloc(sizeof(*features), GFP_KERNEL); + if (!features) { + status = NVME_SC_INTERNAL; + goto out; + } + + features->fis[NVME_FEAT_NUM_QUEUES] = + cpu_to_le32(NVME_FIS_FSUPP | NVME_FIS_CSCPE); + features->fis[NVME_FEAT_KATO] = + cpu_to_le32(NVME_FIS_FSUPP | NVME_FIS_CSCPE); + features->fis[NVME_FEAT_ASYNC_EVENT] = + cpu_to_le32(NVME_FIS_FSUPP | NVME_FIS_CSCPE); + features->fis[NVME_FEAT_HOST_ID] = + cpu_to_le32(NVME_FIS_FSUPP | NVME_FIS_CSCPE); + features->fis[NVME_FEAT_WRITE_PROTECT] = + cpu_to_le32(NVME_FIS_FSUPP | NVME_FIS_NSCPE); + features->fis[NVME_FEAT_RESV_MASK] = + cpu_to_le32(NVME_FIS_FSUPP | NVME_FIS_NSCPE); + + status = nvmet_copy_to_sgl(req, 0, features, sizeof(*features)); + kfree(features); +out: + nvmet_req_complete(req, status); +} + static void nvmet_execute_get_log_page(struct nvmet_req *req) { if (!nvmet_check_transfer_len(req, nvmet_get_log_page_len(req->cmd))) return; switch (req->cmd->get_log_page.lid) { + case NVME_LOG_SUPPORTED: + return nvmet_execute_get_supported_log_pages(req); case NVME_LOG_ERROR: return nvmet_execute_get_log_page_error(req); case NVME_LOG_SMART: @@ -338,13 +629,21 @@ static void nvmet_execute_get_log_page(struct nvmet_req *req) return nvmet_execute_get_log_changed_ns(req); case NVME_LOG_CMD_EFFECTS: return nvmet_execute_get_log_cmd_effects_ns(req); + case NVME_LOG_ENDURANCE_GROUP: + return nvmet_execute_get_log_page_endgrp(req); case NVME_LOG_ANA: return nvmet_execute_get_log_page_ana(req); + case NVME_LOG_FEATURES: + return nvmet_execute_get_log_page_features(req); + case NVME_LOG_RMI: + return nvmet_execute_get_log_page_rmi(req); + case NVME_LOG_RESERVATION: + return nvmet_execute_get_log_page_resv(req); } pr_debug("unhandled lid %d on qid %d\n", req->cmd->get_log_page.lid, req->sq->qid); req->error_loc = offsetof(struct nvme_get_log_page_command, lid); - nvmet_req_complete(req, NVME_SC_INVALID_FIELD | NVME_SC_DNR); + nvmet_req_complete(req, NVME_SC_INVALID_FIELD | NVME_STATUS_DNR); } static void nvmet_execute_identify_ctrl(struct nvmet_req *req) @@ -352,7 +651,7 @@ static void nvmet_execute_identify_ctrl(struct nvmet_req *req) struct nvmet_ctrl *ctrl = req->sq->ctrl; struct nvmet_subsys *subsys = ctrl->subsys; struct nvme_id_ctrl *id; - u32 cmd_capsule_size; + u32 cmd_capsule_size, ctratt; u16 status = 0; if (!subsys->subsys_discovered) { @@ -367,9 +666,8 @@ static void nvmet_execute_identify_ctrl(struct nvmet_req *req) goto out; } - /* XXX: figure out how to assign real vendors IDs. */ - id->vid = 0; - id->ssvid = 0; + id->vid = cpu_to_le16(subsys->vendor_id); + id->ssvid = cpu_to_le16(subsys->subsys_vendor_id); memcpy(id->sn, ctrl->subsys->serial, NVMET_SN_MAX_SIZE); memcpy_and_pad(id->mn, sizeof(id->mn), subsys->model_number, @@ -401,14 +699,16 @@ static void nvmet_execute_identify_ctrl(struct nvmet_req *req) /* XXX: figure out what to do about RTD3R/RTD3 */ id->oaes = cpu_to_le32(NVMET_AEN_CFG_OPTIONAL); - id->ctratt = cpu_to_le32(NVME_CTRL_ATTR_HID_128_BIT | - NVME_CTRL_ATTR_TBKAS); + ctratt = NVME_CTRL_ATTR_HID_128_BIT | NVME_CTRL_ATTR_TBKAS; + if (nvmet_is_pci_ctrl(ctrl)) + ctratt |= NVME_CTRL_ATTR_RHII; + id->ctratt = cpu_to_le32(ctratt); id->oacs = 0; /* * We don't really have a practical limit on the number of abort - * comands. But we don't do anything useful for abort either, so + * commands. But we don't do anything useful for abort either, so * no point in allowing more abort commands than the spec requires. */ id->acl = 3; @@ -428,12 +728,13 @@ static void nvmet_execute_identify_ctrl(struct nvmet_req *req) id->cqes = (0x4 << 4) | 0x4; /* no enforcement soft-limit for maxcmd - pick arbitrary high value */ - id->maxcmd = cpu_to_le16(NVMET_MAX_CMD); + id->maxcmd = cpu_to_le16(NVMET_MAX_CMD(ctrl)); id->nn = cpu_to_le32(NVMET_MAX_NAMESPACES); id->mnan = cpu_to_le32(NVMET_MAX_NAMESPACES); id->oncs = cpu_to_le16(NVME_CTRL_ONCS_DSM | - NVME_CTRL_ONCS_WRITE_ZEROES); + NVME_CTRL_ONCS_WRITE_ZEROES | + NVME_CTRL_ONCS_RESERVATIONS); /* XXX: don't report vwc if the underlying device is write through */ id->vwc = NVME_CTRL_VWC_PRESENT; @@ -445,11 +746,12 @@ static void nvmet_execute_identify_ctrl(struct nvmet_req *req) id->awun = 0; id->awupf = 0; - id->sgls = cpu_to_le32(1 << 0); /* we always support SGLs */ + /* we always support SGLs */ + id->sgls = cpu_to_le32(NVME_CTRL_SGLS_BYTE_ALIGNED); if (ctrl->ops->flags & NVMF_KEYED_SGLS) - id->sgls |= cpu_to_le32(1 << 2); + id->sgls |= cpu_to_le32(NVME_CTRL_SGLS_KSDBDS); if (req->port->inline_data_size) - id->sgls |= cpu_to_le32(1 << 20); + id->sgls |= cpu_to_le32(NVME_CTRL_SGLS_SAOS); strscpy(id->subnqn, ctrl->subsys->subsysnqn, sizeof(id->subnqn)); @@ -467,6 +769,13 @@ static void nvmet_execute_identify_ctrl(struct nvmet_req *req) id->msdbd = ctrl->ops->msdbd; + /* + * Endurance group identifier is 16 bits, so we can't let namespaces + * overflow that since we reuse the nsid + */ + BUILD_BUG_ON(NVMET_MAX_NAMESPACES > USHRT_MAX); + id->endgidmax = cpu_to_le16(NVMET_MAX_NAMESPACES); + id->anacap = (1 << 0) | (1 << 1) | (1 << 2) | (1 << 3) | (1 << 4); id->anatt = 10; /* random value */ id->anagrpmax = cpu_to_le32(NVMET_MAX_ANAGRPS); @@ -496,7 +805,7 @@ static void nvmet_execute_identify_ns(struct nvmet_req *req) if (le32_to_cpu(req->cmd->identify.nsid) == NVME_NSID_ALL) { req->error_loc = offsetof(struct nvme_identify, nsid); - status = NVME_SC_INVALID_NS | NVME_SC_DNR; + status = NVME_SC_INVALID_NS | NVME_STATUS_DNR; goto out; } @@ -551,6 +860,21 @@ static void nvmet_execute_identify_ns(struct nvmet_req *req) id->nmic = NVME_NS_NMIC_SHARED; id->anagrpid = cpu_to_le32(req->ns->anagrpid); + if (req->ns->pr.enable) + id->rescap = NVME_PR_SUPPORT_WRITE_EXCLUSIVE | + NVME_PR_SUPPORT_EXCLUSIVE_ACCESS | + NVME_PR_SUPPORT_WRITE_EXCLUSIVE_REG_ONLY | + NVME_PR_SUPPORT_EXCLUSIVE_ACCESS_REG_ONLY | + NVME_PR_SUPPORT_WRITE_EXCLUSIVE_ALL_REGS | + NVME_PR_SUPPORT_EXCLUSIVE_ACCESS_ALL_REGS | + NVME_PR_SUPPORT_IEKEY_VER_1_3_DEF; + + /* + * Since we don't know any better, every namespace is its own endurance + * group. + */ + id->endgid = cpu_to_le16(req->ns->nsid); + memcpy(&id->nguid, &req->ns->nguid, sizeof(id->nguid)); id->lbaf[0].ds = req->ns->blksize_shift; @@ -576,7 +900,40 @@ out: nvmet_req_complete(req, status); } -static void nvmet_execute_identify_nslist(struct nvmet_req *req) +static void nvmet_execute_identify_endgrp_list(struct nvmet_req *req) +{ + u16 min_endgid = le16_to_cpu(req->cmd->identify.cnssid); + static const int buf_size = NVME_IDENTIFY_DATA_SIZE; + struct nvmet_ctrl *ctrl = req->sq->ctrl; + struct nvmet_ns *ns; + unsigned long idx; + __le16 *list; + u16 status; + int i = 1; + + list = kzalloc(buf_size, GFP_KERNEL); + if (!list) { + status = NVME_SC_INTERNAL; + goto out; + } + + nvmet_for_each_enabled_ns(&ctrl->subsys->namespaces, idx, ns) { + if (ns->nsid <= min_endgid) + continue; + + list[i++] = cpu_to_le16(ns->nsid); + if (i == buf_size / sizeof(__le16)) + break; + } + + list[0] = cpu_to_le16(i - 1); + status = nvmet_copy_to_sgl(req, 0, list, buf_size); + kfree(list); +out: + nvmet_req_complete(req, status); +} + +static void nvmet_execute_identify_nslist(struct nvmet_req *req, bool match_css) { static const int buf_size = NVME_IDENTIFY_DATA_SIZE; struct nvmet_ctrl *ctrl = req->sq->ctrl; @@ -587,15 +944,27 @@ static void nvmet_execute_identify_nslist(struct nvmet_req *req) u16 status = 0; int i = 0; + /* + * NSID values 0xFFFFFFFE and NVME_NSID_ALL are invalid + * See NVMe Base Specification, Active Namespace ID list (CNS 02h). + */ + if (min_nsid == 0xFFFFFFFE || min_nsid == NVME_NSID_ALL) { + req->error_loc = offsetof(struct nvme_identify, nsid); + status = NVME_SC_INVALID_NS | NVME_STATUS_DNR; + goto out; + } + list = kzalloc(buf_size, GFP_KERNEL); if (!list) { status = NVME_SC_INTERNAL; goto out; } - xa_for_each(&ctrl->subsys->namespaces, idx, ns) { + nvmet_for_each_enabled_ns(&ctrl->subsys->namespaces, idx, ns) { if (ns->nsid <= min_nsid) continue; + if (match_css && req->ns->csi != req->cmd->identify.csi) + continue; list[i++] = cpu_to_le32(ns->nsid); if (i == buf_size / sizeof(__le32)) break; @@ -662,27 +1031,73 @@ static void nvmet_execute_identify_desclist(struct nvmet_req *req) if (sg_zero_buffer(req->sg, req->sg_cnt, NVME_IDENTIFY_DATA_SIZE - off, off) != NVME_IDENTIFY_DATA_SIZE - off) - status = NVME_SC_INTERNAL | NVME_SC_DNR; + status = NVME_SC_INTERNAL | NVME_STATUS_DNR; out: nvmet_req_complete(req, status); } -static bool nvmet_handle_identify_desclist(struct nvmet_req *req) +static void nvmet_execute_identify_ctrl_nvm(struct nvmet_req *req) { - switch (req->cmd->identify.csi) { - case NVME_CSI_NVM: - nvmet_execute_identify_desclist(req); - return true; - case NVME_CSI_ZNS: - if (IS_ENABLED(CONFIG_BLK_DEV_ZONED)) { - nvmet_execute_identify_desclist(req); - return true; - } - return false; - default: - return false; + /* Not supported: return zeroes */ + nvmet_req_complete(req, + nvmet_zero_sgl(req, 0, sizeof(struct nvme_id_ctrl_nvm))); +} + +static void nvme_execute_identify_ns_nvm(struct nvmet_req *req) +{ + u16 status; + struct nvme_id_ns_nvm *id; + + status = nvmet_req_find_ns(req); + if (status) + goto out; + + id = kzalloc(sizeof(*id), GFP_KERNEL); + if (!id) { + status = NVME_SC_INTERNAL; + goto out; } + status = nvmet_copy_to_sgl(req, 0, id, sizeof(*id)); + kfree(id); +out: + nvmet_req_complete(req, status); +} + +static void nvmet_execute_id_cs_indep(struct nvmet_req *req) +{ + struct nvme_id_ns_cs_indep *id; + u16 status; + + status = nvmet_req_find_ns(req); + if (status) + goto out; + + id = kzalloc(sizeof(*id), GFP_KERNEL); + if (!id) { + status = NVME_SC_INTERNAL; + goto out; + } + + id->nstat = NVME_NSTAT_NRDY; + id->anagrpid = cpu_to_le32(req->ns->anagrpid); + id->nmic = NVME_NS_NMIC_SHARED; + if (req->ns->readonly) + id->nsattr |= NVME_NS_ATTR_RO; + if (req->ns->bdev && !bdev_nonrot(req->ns->bdev)) + id->nsfeat |= NVME_NS_ROTATIONAL; + /* + * We need flush command to flush the file's metadata, + * so report supporting vwc if backend is file, even + * though buffered_io is disable. + */ + if (req->ns->bdev && !bdev_write_cache(req->ns->bdev)) + id->nsfeat |= NVME_NS_VWC_NOT_PRESENT; + + status = nvmet_copy_to_sgl(req, 0, id, sizeof(*id)); + kfree(id); +out: + nvmet_req_complete(req, status); } static void nvmet_execute_identify(struct nvmet_req *req) @@ -692,61 +1107,65 @@ static void nvmet_execute_identify(struct nvmet_req *req) switch (req->cmd->identify.cns) { case NVME_ID_CNS_NS: - switch (req->cmd->identify.csi) { - case NVME_CSI_NVM: - return nvmet_execute_identify_ns(req); - default: - break; - } - break; - case NVME_ID_CNS_CS_NS: - if (IS_ENABLED(CONFIG_BLK_DEV_ZONED)) { - switch (req->cmd->identify.csi) { - case NVME_CSI_ZNS: - return nvmet_execute_identify_cns_cs_ns(req); - default: - break; - } - } - break; + nvmet_execute_identify_ns(req); + return; case NVME_ID_CNS_CTRL: + nvmet_execute_identify_ctrl(req); + return; + case NVME_ID_CNS_NS_ACTIVE_LIST: + nvmet_execute_identify_nslist(req, false); + return; + case NVME_ID_CNS_NS_DESC_LIST: + nvmet_execute_identify_desclist(req); + return; + case NVME_ID_CNS_CS_NS: switch (req->cmd->identify.csi) { case NVME_CSI_NVM: - return nvmet_execute_identify_ctrl(req); - } - break; - case NVME_ID_CNS_CS_CTRL: - if (IS_ENABLED(CONFIG_BLK_DEV_ZONED)) { - switch (req->cmd->identify.csi) { - case NVME_CSI_ZNS: - return nvmet_execute_identify_cns_cs_ctrl(req); - default: - break; + nvme_execute_identify_ns_nvm(req); + return; + case NVME_CSI_ZNS: + if (IS_ENABLED(CONFIG_BLK_DEV_ZONED)) { + nvmet_execute_identify_ns_zns(req); + return; } + break; } break; - case NVME_ID_CNS_NS_ACTIVE_LIST: + case NVME_ID_CNS_CS_CTRL: switch (req->cmd->identify.csi) { case NVME_CSI_NVM: - return nvmet_execute_identify_nslist(req); - default: + nvmet_execute_identify_ctrl_nvm(req); + return; + case NVME_CSI_ZNS: + if (IS_ENABLED(CONFIG_BLK_DEV_ZONED)) { + nvmet_execute_identify_ctrl_zns(req); + return; + } break; } break; - case NVME_ID_CNS_NS_DESC_LIST: - if (nvmet_handle_identify_desclist(req) == true) - return; - break; + case NVME_ID_CNS_NS_ACTIVE_LIST_CS: + nvmet_execute_identify_nslist(req, true); + return; + case NVME_ID_CNS_NS_CS_INDEP: + nvmet_execute_id_cs_indep(req); + return; + case NVME_ID_CNS_ENDGRP_LIST: + nvmet_execute_identify_endgrp_list(req); + return; } - nvmet_req_cns_error_complete(req); + pr_debug("unhandled identify cns %d on qid %d\n", + req->cmd->identify.cns, req->sq->qid); + req->error_loc = offsetof(struct nvme_identify, cns); + nvmet_req_complete(req, NVME_SC_INVALID_FIELD | NVME_STATUS_DNR); } /* * A "minimum viable" abort implementation: the command is mandatory in the * spec, but we are not required to do any useful work. We couldn't really * do a useful abort, so don't bother even with waiting for the command - * to be exectuted and return immediately telling the command to abort + * to be executed and return immediately telling the command to abort * wasn't found. */ static void nvmet_execute_abort(struct nvmet_req *req) @@ -822,7 +1241,7 @@ u16 nvmet_set_feat_async_event(struct nvmet_req *req, u32 mask) if (val32 & ~mask) { req->error_loc = offsetof(struct nvme_common_command, cdw11); - return NVME_SC_INVALID_FIELD | NVME_SC_DNR; + return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; } WRITE_ONCE(req->sq->ctrl->aen_enabled, val32); @@ -831,6 +1250,92 @@ u16 nvmet_set_feat_async_event(struct nvmet_req *req, u32 mask) return 0; } +static u16 nvmet_set_feat_host_id(struct nvmet_req *req) +{ + struct nvmet_ctrl *ctrl = req->sq->ctrl; + + if (!nvmet_is_pci_ctrl(ctrl)) + return NVME_SC_CMD_SEQ_ERROR | NVME_STATUS_DNR; + + /* + * The NVMe base specifications v2.1 recommends supporting 128-bits host + * IDs (section 5.1.25.1.28.1). However, that same section also says + * that "The controller may support a 64-bit Host Identifier and/or an + * extended 128-bit Host Identifier". So simplify this support and do + * not support 64-bits host IDs to avoid needing to check that all + * controllers associated with the same subsystem all use the same host + * ID size. + */ + if (!(req->cmd->common.cdw11 & cpu_to_le32(1 << 0))) { + req->error_loc = offsetof(struct nvme_common_command, cdw11); + return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; + } + + return nvmet_copy_from_sgl(req, 0, &req->sq->ctrl->hostid, + sizeof(req->sq->ctrl->hostid)); +} + +static u16 nvmet_set_feat_irq_coalesce(struct nvmet_req *req) +{ + struct nvmet_ctrl *ctrl = req->sq->ctrl; + u32 cdw11 = le32_to_cpu(req->cmd->common.cdw11); + struct nvmet_feat_irq_coalesce irqc = { + .time = (cdw11 >> 8) & 0xff, + .thr = cdw11 & 0xff, + }; + + /* + * This feature is not supported for fabrics controllers and mandatory + * for PCI controllers. + */ + if (!nvmet_is_pci_ctrl(ctrl)) { + req->error_loc = offsetof(struct nvme_common_command, cdw10); + return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; + } + + return ctrl->ops->set_feature(ctrl, NVME_FEAT_IRQ_COALESCE, &irqc); +} + +static u16 nvmet_set_feat_irq_config(struct nvmet_req *req) +{ + struct nvmet_ctrl *ctrl = req->sq->ctrl; + u32 cdw11 = le32_to_cpu(req->cmd->common.cdw11); + struct nvmet_feat_irq_config irqcfg = { + .iv = cdw11 & 0xffff, + .cd = (cdw11 >> 16) & 0x1, + }; + + /* + * This feature is not supported for fabrics controllers and mandatory + * for PCI controllers. + */ + if (!nvmet_is_pci_ctrl(ctrl)) { + req->error_loc = offsetof(struct nvme_common_command, cdw10); + return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; + } + + return ctrl->ops->set_feature(ctrl, NVME_FEAT_IRQ_CONFIG, &irqcfg); +} + +static u16 nvmet_set_feat_arbitration(struct nvmet_req *req) +{ + struct nvmet_ctrl *ctrl = req->sq->ctrl; + u32 cdw11 = le32_to_cpu(req->cmd->common.cdw11); + struct nvmet_feat_arbitration arb = { + .hpw = (cdw11 >> 24) & 0xff, + .mpw = (cdw11 >> 16) & 0xff, + .lpw = (cdw11 >> 8) & 0xff, + .ab = cdw11 & 0x3, + }; + + if (!ctrl->ops->set_feature) { + req->error_loc = offsetof(struct nvme_common_command, cdw10); + return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; + } + + return ctrl->ops->set_feature(ctrl, NVME_FEAT_ARBITRATION, &arb); +} + void nvmet_execute_set_features(struct nvmet_req *req) { struct nvmet_subsys *subsys = nvmet_req_subsys(req); @@ -840,20 +1345,29 @@ void nvmet_execute_set_features(struct nvmet_req *req) u16 nsqr; u16 ncqr; - if (!nvmet_check_transfer_len(req, 0)) + if (!nvmet_check_data_len_lte(req, 0)) return; switch (cdw10 & 0xff) { + case NVME_FEAT_ARBITRATION: + status = nvmet_set_feat_arbitration(req); + break; case NVME_FEAT_NUM_QUEUES: ncqr = (cdw11 >> 16) & 0xffff; nsqr = cdw11 & 0xffff; if (ncqr == 0xffff || nsqr == 0xffff) { - status = NVME_SC_INVALID_FIELD | NVME_SC_DNR; + status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; break; } nvmet_set_result(req, (subsys->max_qid - 1) | ((subsys->max_qid - 1) << 16)); break; + case NVME_FEAT_IRQ_COALESCE: + status = nvmet_set_feat_irq_coalesce(req); + break; + case NVME_FEAT_IRQ_CONFIG: + status = nvmet_set_feat_irq_config(req); + break; case NVME_FEAT_KATO: status = nvmet_set_feat_kato(req); break; @@ -861,14 +1375,17 @@ void nvmet_execute_set_features(struct nvmet_req *req) status = nvmet_set_feat_async_event(req, NVMET_AEN_CFG_ALL); break; case NVME_FEAT_HOST_ID: - status = NVME_SC_CMD_SEQ_ERROR | NVME_SC_DNR; + status = nvmet_set_feat_host_id(req); break; case NVME_FEAT_WRITE_PROTECT: status = nvmet_set_feat_write_protect(req); break; + case NVME_FEAT_RESV_MASK: + status = nvmet_set_feat_resv_notif_mask(req, cdw11); + break; default: req->error_loc = offsetof(struct nvme_common_command, cdw10); - status = NVME_SC_INVALID_FIELD | NVME_SC_DNR; + status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; break; } @@ -895,6 +1412,79 @@ static u16 nvmet_get_feat_write_protect(struct nvmet_req *req) return 0; } +static u16 nvmet_get_feat_irq_coalesce(struct nvmet_req *req) +{ + struct nvmet_ctrl *ctrl = req->sq->ctrl; + struct nvmet_feat_irq_coalesce irqc = { }; + u16 status; + + /* + * This feature is not supported for fabrics controllers and mandatory + * for PCI controllers. + */ + if (!nvmet_is_pci_ctrl(ctrl)) { + req->error_loc = offsetof(struct nvme_common_command, cdw10); + return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; + } + + status = ctrl->ops->get_feature(ctrl, NVME_FEAT_IRQ_COALESCE, &irqc); + if (status != NVME_SC_SUCCESS) + return status; + + nvmet_set_result(req, ((u32)irqc.time << 8) | (u32)irqc.thr); + + return NVME_SC_SUCCESS; +} + +static u16 nvmet_get_feat_irq_config(struct nvmet_req *req) +{ + struct nvmet_ctrl *ctrl = req->sq->ctrl; + u32 iv = le32_to_cpu(req->cmd->common.cdw11) & 0xffff; + struct nvmet_feat_irq_config irqcfg = { .iv = iv }; + u16 status; + + /* + * This feature is not supported for fabrics controllers and mandatory + * for PCI controllers. + */ + if (!nvmet_is_pci_ctrl(ctrl)) { + req->error_loc = offsetof(struct nvme_common_command, cdw10); + return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; + } + + status = ctrl->ops->get_feature(ctrl, NVME_FEAT_IRQ_CONFIG, &irqcfg); + if (status != NVME_SC_SUCCESS) + return status; + + nvmet_set_result(req, ((u32)irqcfg.cd << 16) | iv); + + return NVME_SC_SUCCESS; +} + +static u16 nvmet_get_feat_arbitration(struct nvmet_req *req) +{ + struct nvmet_ctrl *ctrl = req->sq->ctrl; + struct nvmet_feat_arbitration arb = { }; + u16 status; + + if (!ctrl->ops->get_feature) { + req->error_loc = offsetof(struct nvme_common_command, cdw10); + return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; + } + + status = ctrl->ops->get_feature(ctrl, NVME_FEAT_ARBITRATION, &arb); + if (status != NVME_SC_SUCCESS) + return status; + + nvmet_set_result(req, + ((u32)arb.hpw << 24) | + ((u32)arb.mpw << 16) | + ((u32)arb.lpw << 8) | + (arb.ab & 0x3)); + + return NVME_SC_SUCCESS; +} + void nvmet_get_feat_kato(struct nvmet_req *req) { nvmet_set_result(req, req->sq->ctrl->kato * 1000); @@ -921,21 +1511,24 @@ void nvmet_execute_get_features(struct nvmet_req *req) * need to come up with some fake values for these. */ #if 0 - case NVME_FEAT_ARBITRATION: - break; case NVME_FEAT_POWER_MGMT: break; case NVME_FEAT_TEMP_THRESH: break; case NVME_FEAT_ERR_RECOVERY: break; + case NVME_FEAT_WRITE_ATOMIC: + break; +#endif + case NVME_FEAT_ARBITRATION: + status = nvmet_get_feat_arbitration(req); + break; case NVME_FEAT_IRQ_COALESCE: + status = nvmet_get_feat_irq_coalesce(req); break; case NVME_FEAT_IRQ_CONFIG: + status = nvmet_get_feat_irq_config(req); break; - case NVME_FEAT_WRITE_ATOMIC: - break; -#endif case NVME_FEAT_ASYNC_EVENT: nvmet_get_feat_async_event(req); break; @@ -954,7 +1547,7 @@ void nvmet_execute_get_features(struct nvmet_req *req) if (!(req->cmd->common.cdw11 & cpu_to_le32(1 << 0))) { req->error_loc = offsetof(struct nvme_common_command, cdw11); - status = NVME_SC_INVALID_FIELD | NVME_SC_DNR; + status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; break; } @@ -964,10 +1557,13 @@ void nvmet_execute_get_features(struct nvmet_req *req) case NVME_FEAT_WRITE_PROTECT: status = nvmet_get_feat_write_protect(req); break; + case NVME_FEAT_RESV_MASK: + status = nvmet_get_feat_resv_notif_mask(req); + break; default: req->error_loc = offsetof(struct nvme_common_command, cdw10); - status = NVME_SC_INVALID_FIELD | NVME_SC_DNR; + status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; break; } @@ -984,7 +1580,7 @@ void nvmet_execute_async_event(struct nvmet_req *req) mutex_lock(&ctrl->lock); if (ctrl->nr_async_event_cmds >= NVMET_ASYNC_EVENTS) { mutex_unlock(&ctrl->lock); - nvmet_req_complete(req, NVME_SC_ASYNC_LIMIT | NVME_SC_DNR); + nvmet_req_complete(req, NVME_SC_ASYNC_LIMIT | NVME_STATUS_DNR); return; } ctrl->async_event_cmds[ctrl->nr_async_event_cmds++] = req; @@ -1013,6 +1609,27 @@ out: nvmet_req_complete(req, status); } +u32 nvmet_admin_cmd_data_len(struct nvmet_req *req) +{ + struct nvme_command *cmd = req->cmd; + + if (nvme_is_fabrics(cmd)) + return nvmet_fabrics_admin_cmd_data_len(req); + if (nvmet_is_disc_subsys(nvmet_req_subsys(req))) + return nvmet_discovery_cmd_data_len(req); + + switch (cmd->common.opcode) { + case nvme_admin_get_log_page: + return nvmet_get_log_page_len(cmd); + case nvme_admin_identify: + return NVME_IDENTIFY_DATA_SIZE; + case nvme_admin_get_features: + return nvmet_feat_data_len(req, le32_to_cpu(cmd->common.cdw10)); + default: + return 0; + } +} + u16 nvmet_parse_admin_cmd(struct nvmet_req *req) { struct nvme_command *cmd = req->cmd; @@ -1020,8 +1637,6 @@ u16 nvmet_parse_admin_cmd(struct nvmet_req *req) if (nvme_is_fabrics(cmd)) return nvmet_parse_fabrics_admin_cmd(req); - if (unlikely(!nvmet_check_auth_status(req))) - return NVME_SC_AUTH_REQUIRED | NVME_SC_DNR; if (nvmet_is_disc_subsys(nvmet_req_subsys(req))) return nvmet_parse_discovery_cmd(req); @@ -1029,13 +1644,30 @@ u16 nvmet_parse_admin_cmd(struct nvmet_req *req) if (unlikely(ret)) return ret; + /* For PCI controllers, admin commands shall not use SGL. */ + if (nvmet_is_pci_ctrl(req->sq->ctrl) && !req->sq->qid && + cmd->common.flags & NVME_CMD_SGL_ALL) + return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; + if (nvmet_is_passthru_req(req)) return nvmet_parse_passthru_admin_cmd(req); switch (cmd->common.opcode) { + case nvme_admin_delete_sq: + req->execute = nvmet_execute_delete_sq; + return 0; + case nvme_admin_create_sq: + req->execute = nvmet_execute_create_sq; + return 0; case nvme_admin_get_log_page: req->execute = nvmet_execute_get_log_page; return 0; + case nvme_admin_delete_cq: + req->execute = nvmet_execute_delete_cq; + return 0; + case nvme_admin_create_cq: + req->execute = nvmet_execute_create_cq; + return 0; case nvme_admin_identify: req->execute = nvmet_execute_identify; return 0; diff --git a/drivers/nvme/target/auth.c b/drivers/nvme/target/auth.c index 4dcddcf95279..2eadeb7e06f2 100644 --- a/drivers/nvme/target/auth.c +++ b/drivers/nvme/target/auth.c @@ -15,7 +15,8 @@ #include <linux/ctype.h> #include <linux/random.h> #include <linux/nvme-auth.h> -#include <asm/unaligned.h> +#include <linux/nvme-keyring.h> +#include <linux/unaligned.h> #include "nvmet.h" @@ -25,6 +26,18 @@ int nvmet_auth_set_key(struct nvmet_host *host, const char *secret, unsigned char key_hash; char *dhchap_secret; + if (!strlen(secret)) { + if (set_ctrl) { + kfree(host->dhchap_ctrl_secret); + host->dhchap_ctrl_secret = NULL; + host->dhchap_ctrl_key_hash = 0; + } else { + kfree(host->dhchap_secret); + host->dhchap_secret = NULL; + host->dhchap_key_hash = 0; + } + return 0; + } if (sscanf(secret, "DHHC-1:%hhd:%*s", &key_hash) != 1) return -EINVAL; if (key_hash > 3) { @@ -44,6 +57,7 @@ int nvmet_auth_set_key(struct nvmet_host *host, const char *secret, dhchap_secret = kstrdup(secret, GFP_KERNEL); if (!dhchap_secret) return -ENOMEM; + down_write(&nvmet_config_sem); if (set_ctrl) { kfree(host->dhchap_ctrl_secret); host->dhchap_ctrl_secret = strim(dhchap_secret); @@ -53,6 +67,7 @@ int nvmet_auth_set_key(struct nvmet_host *host, const char *secret, host->dhchap_secret = strim(dhchap_secret); host->dhchap_key_hash = key_hash; } + up_write(&nvmet_config_sem); return 0; } @@ -101,6 +116,7 @@ int nvmet_setup_dhgroup(struct nvmet_ctrl *ctrl, u8 dhgroup_id) pr_debug("%s: ctrl %d failed to generate private key, err %d\n", __func__, ctrl->cntlid, ret); kfree_sensitive(ctrl->dh_key); + ctrl->dh_key = NULL; return ret; } ctrl->dh_keysize = crypto_kpp_maxsize(ctrl->dh_tfm); @@ -124,12 +140,11 @@ int nvmet_setup_dhgroup(struct nvmet_ctrl *ctrl, u8 dhgroup_id) return ret; } -int nvmet_setup_auth(struct nvmet_ctrl *ctrl) +u8 nvmet_setup_auth(struct nvmet_ctrl *ctrl, struct nvmet_sq *sq) { int ret = 0; struct nvmet_host_link *p; struct nvmet_host *host = NULL; - const char *hash_name; down_read(&nvmet_config_sem); if (nvmet_is_disc_subsys(ctrl->subsys)) @@ -147,13 +162,21 @@ int nvmet_setup_auth(struct nvmet_ctrl *ctrl) } if (!host) { pr_debug("host %s not found\n", ctrl->hostnqn); - ret = -EPERM; + ret = NVME_AUTH_DHCHAP_FAILURE_FAILED; + goto out_unlock; + } + + if (nvmet_queue_tls_keyid(sq)) { + pr_debug("host %s tls enabled\n", ctrl->hostnqn); goto out_unlock; } ret = nvmet_setup_dhgroup(ctrl, host->dhchap_dhgroup_id); - if (ret < 0) + if (ret < 0) { pr_warn("Failed to setup DH group"); + ret = NVME_AUTH_DHCHAP_FAILURE_DHGROUP_UNUSABLE; + goto out_unlock; + } if (!host->dhchap_secret) { pr_debug("No authentication provided\n"); @@ -164,12 +187,6 @@ int nvmet_setup_auth(struct nvmet_ctrl *ctrl) pr_debug("Re-use existing hash ID %d\n", ctrl->shash_id); } else { - hash_name = nvme_auth_hmac_name(host->dhchap_hash_id); - if (!hash_name) { - pr_warn("Hash ID %d invalid\n", host->dhchap_hash_id); - ret = -EINVAL; - goto out_unlock; - } ctrl->shash_id = host->dhchap_hash_id; } @@ -178,7 +195,7 @@ int nvmet_setup_auth(struct nvmet_ctrl *ctrl) ctrl->host_key = nvme_auth_extract_key(host->dhchap_secret + 10, host->dhchap_key_hash); if (IS_ERR(ctrl->host_key)) { - ret = PTR_ERR(ctrl->host_key); + ret = NVME_AUTH_DHCHAP_FAILURE_NOT_USABLE; ctrl->host_key = NULL; goto out_free_hash; } @@ -196,7 +213,7 @@ int nvmet_setup_auth(struct nvmet_ctrl *ctrl) ctrl->ctrl_key = nvme_auth_extract_key(host->dhchap_ctrl_secret + 10, host->dhchap_ctrl_key_hash); if (IS_ERR(ctrl->ctrl_key)) { - ret = PTR_ERR(ctrl->ctrl_key); + ret = NVME_AUTH_DHCHAP_FAILURE_NOT_USABLE; ctrl->ctrl_key = NULL; goto out_free_hash; } @@ -222,6 +239,9 @@ out_unlock: void nvmet_auth_sq_free(struct nvmet_sq *sq) { cancel_delayed_work(&sq->auth_expired_work); +#ifdef CONFIG_NVME_TARGET_TCP_TLS + sq->tls_key = NULL; +#endif kfree(sq->dhchap_c1); sq->dhchap_c1 = NULL; kfree(sq->dhchap_c2); @@ -250,13 +270,22 @@ void nvmet_destroy_auth(struct nvmet_ctrl *ctrl) nvme_auth_free_key(ctrl->ctrl_key); ctrl->ctrl_key = NULL; } +#ifdef CONFIG_NVME_TARGET_TCP_TLS + if (ctrl->tls_key) { + key_put(ctrl->tls_key); + ctrl->tls_key = NULL; + } +#endif } bool nvmet_check_auth_status(struct nvmet_req *req) { - if (req->sq->ctrl->host_key && - !req->sq->authenticated) - return false; + if (req->sq->ctrl->host_key) { + if (req->sq->qid > 0) + return true; + if (!req->sq->authenticated) + return false; + } return true; } @@ -264,10 +293,11 @@ int nvmet_auth_host_hash(struct nvmet_req *req, u8 *response, unsigned int shash_len) { struct crypto_shash *shash_tfm; - struct shash_desc *shash; + SHASH_DESC_ON_STACK(shash, shash_tfm); struct nvmet_ctrl *ctrl = req->sq->ctrl; const char *hash_name; - u8 *challenge = req->sq->dhchap_c1, *host_response; + u8 *challenge = req->sq->dhchap_c1; + struct nvme_dhchap_key *transformed_key; u8 buf[4]; int ret; @@ -284,21 +314,22 @@ int nvmet_auth_host_hash(struct nvmet_req *req, u8 *response, } if (shash_len != crypto_shash_digestsize(shash_tfm)) { - pr_debug("%s: hash len mismatch (len %d digest %d)\n", - __func__, shash_len, - crypto_shash_digestsize(shash_tfm)); + pr_err("%s: hash len mismatch (len %d digest %d)\n", + __func__, shash_len, + crypto_shash_digestsize(shash_tfm)); ret = -EINVAL; goto out_free_tfm; } - host_response = nvme_auth_transform_key(ctrl->host_key, ctrl->hostnqn); - if (IS_ERR(host_response)) { - ret = PTR_ERR(host_response); + transformed_key = nvme_auth_transform_key(ctrl->host_key, + ctrl->hostnqn); + if (IS_ERR(transformed_key)) { + ret = PTR_ERR(transformed_key); goto out_free_tfm; } - ret = crypto_shash_setkey(shash_tfm, host_response, - ctrl->host_key->len); + ret = crypto_shash_setkey(shash_tfm, transformed_key->key, + transformed_key->len); if (ret) goto out_free_response; @@ -314,19 +345,13 @@ int nvmet_auth_host_hash(struct nvmet_req *req, u8 *response, req->sq->dhchap_c1, challenge, shash_len); if (ret) - goto out_free_response; + goto out; } pr_debug("ctrl %d qid %d host response seq %u transaction %d\n", ctrl->cntlid, req->sq->qid, req->sq->dhchap_s1, req->sq->dhchap_tid); - shash = kzalloc(sizeof(*shash) + crypto_shash_descsize(shash_tfm), - GFP_KERNEL); - if (!shash) { - ret = -ENOMEM; - goto out_free_response; - } shash->tfm = shash_tfm; ret = crypto_shash_init(shash); if (ret) @@ -342,33 +367,33 @@ int nvmet_auth_host_hash(struct nvmet_req *req, u8 *response, ret = crypto_shash_update(shash, buf, 2); if (ret) goto out; - memset(buf, 0, 4); + *buf = req->sq->sc_c; ret = crypto_shash_update(shash, buf, 1); if (ret) goto out; ret = crypto_shash_update(shash, "HostHost", 8); if (ret) goto out; + memset(buf, 0, 4); ret = crypto_shash_update(shash, ctrl->hostnqn, strlen(ctrl->hostnqn)); if (ret) goto out; ret = crypto_shash_update(shash, buf, 1); if (ret) goto out; - ret = crypto_shash_update(shash, ctrl->subsysnqn, - strlen(ctrl->subsysnqn)); + ret = crypto_shash_update(shash, ctrl->subsys->subsysnqn, + strlen(ctrl->subsys->subsysnqn)); if (ret) goto out; ret = crypto_shash_final(shash, response); out: if (challenge != req->sq->dhchap_c1) kfree(challenge); - kfree(shash); out_free_response: - kfree_sensitive(host_response); + nvme_auth_free_key(transformed_key); out_free_tfm: crypto_free_shash(shash_tfm); - return 0; + return ret; } int nvmet_auth_ctrl_hash(struct nvmet_req *req, u8 *response, @@ -378,7 +403,8 @@ int nvmet_auth_ctrl_hash(struct nvmet_req *req, u8 *response, struct shash_desc *shash; struct nvmet_ctrl *ctrl = req->sq->ctrl; const char *hash_name; - u8 *challenge = req->sq->dhchap_c2, *ctrl_response; + u8 *challenge = req->sq->dhchap_c2; + struct nvme_dhchap_key *transformed_key; u8 buf[4]; int ret; @@ -402,15 +428,15 @@ int nvmet_auth_ctrl_hash(struct nvmet_req *req, u8 *response, goto out_free_tfm; } - ctrl_response = nvme_auth_transform_key(ctrl->ctrl_key, - ctrl->subsysnqn); - if (IS_ERR(ctrl_response)) { - ret = PTR_ERR(ctrl_response); + transformed_key = nvme_auth_transform_key(ctrl->ctrl_key, + ctrl->subsys->subsysnqn); + if (IS_ERR(transformed_key)) { + ret = PTR_ERR(transformed_key); goto out_free_tfm; } - ret = crypto_shash_setkey(shash_tfm, ctrl_response, - ctrl->ctrl_key->len); + ret = crypto_shash_setkey(shash_tfm, transformed_key->key, + transformed_key->len); if (ret) goto out_free_response; @@ -426,14 +452,14 @@ int nvmet_auth_ctrl_hash(struct nvmet_req *req, u8 *response, req->sq->dhchap_c2, challenge, shash_len); if (ret) - goto out_free_response; + goto out_free_challenge; } shash = kzalloc(sizeof(*shash) + crypto_shash_descsize(shash_tfm), GFP_KERNEL); if (!shash) { ret = -ENOMEM; - goto out_free_response; + goto out_free_challenge; } shash->tfm = shash_tfm; @@ -458,8 +484,8 @@ int nvmet_auth_ctrl_hash(struct nvmet_req *req, u8 *response, ret = crypto_shash_update(shash, "Controller", 10); if (ret) goto out; - ret = crypto_shash_update(shash, ctrl->subsysnqn, - strlen(ctrl->subsysnqn)); + ret = crypto_shash_update(shash, ctrl->subsys->subsysnqn, + strlen(ctrl->subsys->subsysnqn)); if (ret) goto out; ret = crypto_shash_update(shash, buf, 1); @@ -470,14 +496,15 @@ int nvmet_auth_ctrl_hash(struct nvmet_req *req, u8 *response, goto out; ret = crypto_shash_final(shash, response); out: + kfree(shash); +out_free_challenge: if (challenge != req->sq->dhchap_c2) kfree(challenge); - kfree(shash); out_free_response: - kfree_sensitive(ctrl_response); + nvme_auth_free_key(transformed_key); out_free_tfm: crypto_free_shash(shash_tfm); - return 0; + return ret; } int nvmet_auth_ctrl_exponential(struct nvmet_req *req, @@ -526,3 +553,59 @@ int nvmet_auth_ctrl_sesskey(struct nvmet_req *req, return ret; } + +void nvmet_auth_insert_psk(struct nvmet_sq *sq) +{ + int hash_len = nvme_auth_hmac_hash_len(sq->ctrl->shash_id); + u8 *psk, *digest, *tls_psk; + size_t psk_len; + int ret; +#ifdef CONFIG_NVME_TARGET_TCP_TLS + struct key *tls_key = NULL; +#endif + + ret = nvme_auth_generate_psk(sq->ctrl->shash_id, + sq->dhchap_skey, + sq->dhchap_skey_len, + sq->dhchap_c1, sq->dhchap_c2, + hash_len, &psk, &psk_len); + if (ret) { + pr_warn("%s: ctrl %d qid %d failed to generate PSK, error %d\n", + __func__, sq->ctrl->cntlid, sq->qid, ret); + return; + } + ret = nvme_auth_generate_digest(sq->ctrl->shash_id, psk, psk_len, + sq->ctrl->subsys->subsysnqn, + sq->ctrl->hostnqn, &digest); + if (ret) { + pr_warn("%s: ctrl %d qid %d failed to generate digest, error %d\n", + __func__, sq->ctrl->cntlid, sq->qid, ret); + goto out_free_psk; + } + ret = nvme_auth_derive_tls_psk(sq->ctrl->shash_id, psk, psk_len, + digest, &tls_psk); + if (ret) { + pr_warn("%s: ctrl %d qid %d failed to derive TLS PSK, error %d\n", + __func__, sq->ctrl->cntlid, sq->qid, ret); + goto out_free_digest; + } +#ifdef CONFIG_NVME_TARGET_TCP_TLS + tls_key = nvme_tls_psk_refresh(NULL, sq->ctrl->hostnqn, + sq->ctrl->subsys->subsysnqn, + sq->ctrl->shash_id, tls_psk, psk_len, + digest); + if (IS_ERR(tls_key)) { + pr_warn("%s: ctrl %d qid %d failed to refresh key, error %ld\n", + __func__, sq->ctrl->cntlid, sq->qid, PTR_ERR(tls_key)); + tls_key = NULL; + } + if (sq->ctrl->tls_key) + key_put(sq->ctrl->tls_key); + sq->ctrl->tls_key = tls_key; +#endif + kfree_sensitive(tls_psk); +out_free_digest: + kfree_sensitive(digest); +out_free_psk: + kfree_sensitive(psk); +} diff --git a/drivers/nvme/target/configfs.c b/drivers/nvme/target/configfs.c index 907143870da5..e44ef69dffc2 100644 --- a/drivers/nvme/target/configfs.c +++ b/drivers/nvme/target/configfs.c @@ -15,8 +15,10 @@ #ifdef CONFIG_NVME_TARGET_AUTH #include <linux/nvme-auth.h> #endif +#include <linux/nvme-keyring.h> #include <crypto/hash.h> #include <crypto/kpp.h> +#include <linux/nospec.h> #include "nvmet.h" @@ -35,6 +37,7 @@ static struct nvmet_type_name_map nvmet_transport[] = { { NVMF_TRTYPE_RDMA, "rdma" }, { NVMF_TRTYPE_FC, "fc" }, { NVMF_TRTYPE_TCP, "tcp" }, + { NVMF_TRTYPE_PCI, "pci" }, { NVMF_TRTYPE_LOOP, "loop" }, }; @@ -44,6 +47,7 @@ static const struct nvmet_type_name_map nvmet_addr_family[] = { { NVMF_ADDR_FAMILY_IP6, "ipv6" }, { NVMF_ADDR_FAMILY_IB, "ib" }, { NVMF_ADDR_FAMILY_FC, "fc" }, + { NVMF_ADDR_FAMILY_PCI, "pci" }, { NVMF_ADDR_FAMILY_LOOP, "loop" }, }; @@ -159,10 +163,14 @@ static const struct nvmet_type_name_map nvmet_addr_treq[] = { { NVMF_TREQ_NOT_REQUIRED, "not required" }, }; +static inline u8 nvmet_port_disc_addr_treq_mask(struct nvmet_port *port) +{ + return (port->disc_addr.treq & ~NVME_TREQ_SECURE_CHANNEL_MASK); +} + static ssize_t nvmet_addr_treq_show(struct config_item *item, char *page) { - u8 treq = to_nvmet_port(item)->disc_addr.treq & - NVME_TREQ_SECURE_CHANNEL_MASK; + u8 treq = nvmet_port_disc_addr_treq_secure_channel(to_nvmet_port(item)); int i; for (i = 0; i < ARRAY_SIZE(nvmet_addr_treq); i++) { @@ -178,7 +186,7 @@ static ssize_t nvmet_addr_treq_store(struct config_item *item, const char *page, size_t count) { struct nvmet_port *port = to_nvmet_port(item); - u8 treq = port->disc_addr.treq & ~NVME_TREQ_SECURE_CHANNEL_MASK; + u8 treq = nvmet_port_disc_addr_treq_mask(port); int i; if (nvmet_is_port_enabled(port, __func__)) @@ -193,6 +201,20 @@ static ssize_t nvmet_addr_treq_store(struct config_item *item, return -EINVAL; found: + if (port->disc_addr.trtype == NVMF_TRTYPE_TCP && + port->disc_addr.tsas.tcp.sectype == NVMF_TCP_SECTYPE_TLS13) { + switch (nvmet_addr_treq[i].type) { + case NVMF_TREQ_NOT_SPECIFIED: + pr_debug("treq '%s' not allowed for TLS1.3\n", + nvmet_addr_treq[i].name); + return -EINVAL; + case NVMF_TREQ_NOT_REQUIRED: + pr_warn("Allow non-TLS connections while TLS1.3 is enabled\n"); + break; + default: + break; + } + } treq |= nvmet_addr_treq[i].type; port->disc_addr.treq = treq; return count; @@ -253,6 +275,32 @@ static ssize_t nvmet_param_inline_data_size_store(struct config_item *item, CONFIGFS_ATTR(nvmet_, param_inline_data_size); +static ssize_t nvmet_param_max_queue_size_show(struct config_item *item, + char *page) +{ + struct nvmet_port *port = to_nvmet_port(item); + + return snprintf(page, PAGE_SIZE, "%d\n", port->max_queue_size); +} + +static ssize_t nvmet_param_max_queue_size_store(struct config_item *item, + const char *page, size_t count) +{ + struct nvmet_port *port = to_nvmet_port(item); + int ret; + + if (nvmet_is_port_enabled(port, __func__)) + return -EACCES; + ret = kstrtoint(page, 0, &port->max_queue_size); + if (ret) { + pr_err("Invalid value '%s' for max_queue_size\n", page); + return -EINVAL; + } + return count; +} + +CONFIGFS_ATTR(nvmet_, param_max_queue_size); + #ifdef CONFIG_BLK_DEV_INTEGRITY static ssize_t nvmet_param_pi_enable_show(struct config_item *item, char *page) @@ -303,6 +351,11 @@ static void nvmet_port_init_tsas_rdma(struct nvmet_port *port) port->disc_addr.tsas.rdma.cms = NVMF_RDMA_CMS_RDMA_CM; } +static void nvmet_port_init_tsas_tcp(struct nvmet_port *port, int sectype) +{ + port->disc_addr.tsas.tcp.sectype = sectype; +} + static ssize_t nvmet_addr_trtype_store(struct config_item *item, const char *page, size_t count) { @@ -325,11 +378,120 @@ found: port->disc_addr.trtype = nvmet_transport[i].type; if (port->disc_addr.trtype == NVMF_TRTYPE_RDMA) nvmet_port_init_tsas_rdma(port); + else if (port->disc_addr.trtype == NVMF_TRTYPE_TCP) + nvmet_port_init_tsas_tcp(port, NVMF_TCP_SECTYPE_NONE); return count; } CONFIGFS_ATTR(nvmet_, addr_trtype); +static const struct nvmet_type_name_map nvmet_addr_tsas_tcp[] = { + { NVMF_TCP_SECTYPE_NONE, "none" }, + { NVMF_TCP_SECTYPE_TLS13, "tls1.3" }, +}; + +static const struct nvmet_type_name_map nvmet_addr_tsas_rdma[] = { + { NVMF_RDMA_QPTYPE_CONNECTED, "connected" }, + { NVMF_RDMA_QPTYPE_DATAGRAM, "datagram" }, +}; + +static ssize_t nvmet_addr_tsas_show(struct config_item *item, + char *page) +{ + struct nvmet_port *port = to_nvmet_port(item); + int i; + + if (port->disc_addr.trtype == NVMF_TRTYPE_TCP) { + for (i = 0; i < ARRAY_SIZE(nvmet_addr_tsas_tcp); i++) { + if (port->disc_addr.tsas.tcp.sectype == nvmet_addr_tsas_tcp[i].type) + return sprintf(page, "%s\n", nvmet_addr_tsas_tcp[i].name); + } + } else if (port->disc_addr.trtype == NVMF_TRTYPE_RDMA) { + for (i = 0; i < ARRAY_SIZE(nvmet_addr_tsas_rdma); i++) { + if (port->disc_addr.tsas.rdma.qptype == nvmet_addr_tsas_rdma[i].type) + return sprintf(page, "%s\n", nvmet_addr_tsas_rdma[i].name); + } + } + return sprintf(page, "\n"); +} + +static u8 nvmet_addr_tsas_rdma_store(const char *page) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(nvmet_addr_tsas_rdma); i++) { + if (sysfs_streq(page, nvmet_addr_tsas_rdma[i].name)) + return nvmet_addr_tsas_rdma[i].type; + } + return NVMF_RDMA_QPTYPE_INVALID; +} + +static u8 nvmet_addr_tsas_tcp_store(const char *page) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(nvmet_addr_tsas_tcp); i++) { + if (sysfs_streq(page, nvmet_addr_tsas_tcp[i].name)) + return nvmet_addr_tsas_tcp[i].type; + } + return NVMF_TCP_SECTYPE_INVALID; +} + +static ssize_t nvmet_addr_tsas_store(struct config_item *item, + const char *page, size_t count) +{ + struct nvmet_port *port = to_nvmet_port(item); + u8 treq = nvmet_port_disc_addr_treq_mask(port); + u8 sectype, qptype; + + if (nvmet_is_port_enabled(port, __func__)) + return -EACCES; + + if (port->disc_addr.trtype == NVMF_TRTYPE_RDMA) { + qptype = nvmet_addr_tsas_rdma_store(page); + if (qptype == port->disc_addr.tsas.rdma.qptype) + return count; + } else if (port->disc_addr.trtype == NVMF_TRTYPE_TCP) { + sectype = nvmet_addr_tsas_tcp_store(page); + if (sectype != NVMF_TCP_SECTYPE_INVALID) + goto found; + } + + pr_err("Invalid value '%s' for tsas\n", page); + return -EINVAL; + +found: + if (sectype == NVMF_TCP_SECTYPE_TLS13) { + if (!IS_ENABLED(CONFIG_NVME_TARGET_TCP_TLS)) { + pr_err("TLS is not supported\n"); + return -EINVAL; + } + if (!port->keyring) { + pr_err("TLS keyring not configured\n"); + return -EINVAL; + } + } + + nvmet_port_init_tsas_tcp(port, sectype); + /* + * If TLS is enabled TREQ should be set to 'required' per default + */ + if (sectype == NVMF_TCP_SECTYPE_TLS13) { + u8 sc = nvmet_port_disc_addr_treq_secure_channel(port); + + if (sc == NVMF_TREQ_NOT_SPECIFIED) + treq |= NVMF_TREQ_REQUIRED; + else + treq |= sc; + } else { + treq |= NVMF_TREQ_NOT_SPECIFIED; + } + port->disc_addr.treq = treq; + return count; +} + +CONFIGFS_ATTR(nvmet_, addr_tsas); + /* * Namespace structures & file operation functions below */ @@ -509,6 +671,7 @@ static ssize_t nvmet_ns_ana_grpid_store(struct config_item *item, down_write(&nvmet_ana_sem); oldgrpid = ns->anagrpid; + newgrpid = array_index_nospec(newgrpid, NVMET_MAX_ANAGRPS); nvmet_ana_group_enabled[newgrpid]++; ns->anagrpid = newgrpid; nvmet_ana_group_enabled[oldgrpid]--; @@ -536,10 +699,18 @@ static ssize_t nvmet_ns_enable_store(struct config_item *item, if (kstrtobool(page, &enable)) return -EINVAL; + /* + * take a global nvmet_config_sem because the disable routine has a + * window where it releases the subsys-lock, giving a chance to + * a parallel enable to concurrently execute causing the disable to + * have a misaccounting of the ns percpu_ref. + */ + down_write(&nvmet_config_sem); if (enable) ret = nvmet_ns_enable(ns); else nvmet_ns_disable(ns); + up_write(&nvmet_config_sem); return ret ? ret : count; } @@ -600,6 +771,32 @@ static ssize_t nvmet_ns_revalidate_size_store(struct config_item *item, CONFIGFS_ATTR_WO(nvmet_ns_, revalidate_size); +static ssize_t nvmet_ns_resv_enable_show(struct config_item *item, char *page) +{ + return sysfs_emit(page, "%d\n", to_nvmet_ns(item)->pr.enable); +} + +static ssize_t nvmet_ns_resv_enable_store(struct config_item *item, + const char *page, size_t count) +{ + struct nvmet_ns *ns = to_nvmet_ns(item); + bool val; + + if (kstrtobool(page, &val)) + return -EINVAL; + + mutex_lock(&ns->subsys->lock); + if (ns->enabled) { + pr_err("the ns:%d is already enabled.\n", ns->nsid); + mutex_unlock(&ns->subsys->lock); + return -EINVAL; + } + ns->pr.enable = val; + mutex_unlock(&ns->subsys->lock); + return count; +} +CONFIGFS_ATTR(nvmet_ns_, resv_enable); + static struct configfs_attribute *nvmet_ns_attrs[] = { &nvmet_ns_attr_device_path, &nvmet_ns_attr_device_nguid, @@ -608,6 +805,7 @@ static struct configfs_attribute *nvmet_ns_attrs[] = { &nvmet_ns_attr_enable, &nvmet_ns_attr_buffered_io, &nvmet_ns_attr_revalidate_size, + &nvmet_ns_attr_resv_enable, #ifdef CONFIG_PCI_P2PDMA &nvmet_ns_attr_p2pmem, #endif @@ -1162,7 +1360,7 @@ static ssize_t nvmet_subsys_attr_cntlid_min_store(struct config_item *item, return -EINVAL; down_write(&nvmet_config_sem); - if (cntlid_min >= to_subsys(item)->cntlid_max) + if (cntlid_min > to_subsys(item)->cntlid_max) goto out_unlock; to_subsys(item)->cntlid_min = cntlid_min; up_write(&nvmet_config_sem); @@ -1192,7 +1390,7 @@ static ssize_t nvmet_subsys_attr_cntlid_max_store(struct config_item *item, return -EINVAL; down_write(&nvmet_config_sem); - if (cntlid_max <= to_subsys(item)->cntlid_min) + if (cntlid_max < to_subsys(item)->cntlid_min) goto out_unlock; to_subsys(item)->cntlid_max = cntlid_max; up_write(&nvmet_config_sem); @@ -1204,6 +1402,49 @@ out_unlock: } CONFIGFS_ATTR(nvmet_subsys_, attr_cntlid_max); +static ssize_t nvmet_subsys_attr_vendor_id_show(struct config_item *item, + char *page) +{ + return snprintf(page, PAGE_SIZE, "0x%x\n", to_subsys(item)->vendor_id); +} + +static ssize_t nvmet_subsys_attr_vendor_id_store(struct config_item *item, + const char *page, size_t count) +{ + u16 vid; + + if (kstrtou16(page, 0, &vid)) + return -EINVAL; + + down_write(&nvmet_config_sem); + to_subsys(item)->vendor_id = vid; + up_write(&nvmet_config_sem); + return count; +} +CONFIGFS_ATTR(nvmet_subsys_, attr_vendor_id); + +static ssize_t nvmet_subsys_attr_subsys_vendor_id_show(struct config_item *item, + char *page) +{ + return snprintf(page, PAGE_SIZE, "0x%x\n", + to_subsys(item)->subsys_vendor_id); +} + +static ssize_t nvmet_subsys_attr_subsys_vendor_id_store(struct config_item *item, + const char *page, size_t count) +{ + u16 ssvid; + + if (kstrtou16(page, 0, &ssvid)) + return -EINVAL; + + down_write(&nvmet_config_sem); + to_subsys(item)->subsys_vendor_id = ssvid; + up_write(&nvmet_config_sem); + return count; +} +CONFIGFS_ATTR(nvmet_subsys_, attr_subsys_vendor_id); + static ssize_t nvmet_subsys_attr_model_show(struct config_item *item, char *page) { @@ -1432,6 +1673,8 @@ static struct configfs_attribute *nvmet_subsys_attrs[] = { &nvmet_subsys_attr_attr_serial, &nvmet_subsys_attr_attr_cntlid_min, &nvmet_subsys_attr_attr_cntlid_max, + &nvmet_subsys_attr_attr_vendor_id, + &nvmet_subsys_attr_attr_subsys_vendor_id, &nvmet_subsys_attr_attr_model, &nvmet_subsys_attr_attr_qid_max, &nvmet_subsys_attr_attr_ieee_oui, @@ -1473,6 +1716,11 @@ static struct config_group *nvmet_subsys_make(struct config_group *group, return ERR_PTR(-EINVAL); } + if (sysfs_streq(name, nvmet_disc_subsys->subsysnqn)) { + pr_err("can't create subsystem using unique discovery NQN\n"); + return ERR_PTR(-EINVAL); + } + subsys = nvmet_subsys_alloc(name, NVME_NQN_NVME); if (IS_ERR(subsys)) return ERR_CAST(subsys); @@ -1581,6 +1829,7 @@ static struct config_group *nvmet_referral_make( return ERR_PTR(-ENOMEM); INIT_LIST_HEAD(&port->entry); + port->disc_addr.trtype = NVMF_TRTYPE_MAX; config_group_init_type_name(&port->group, name, &nvmet_referral_type); return &port->group; @@ -1700,6 +1949,7 @@ static struct config_group *nvmet_ana_groups_make_group( grp->grpid = grpid; down_write(&nvmet_ana_sem); + grpid = array_index_nospec(grpid, NVMET_MAX_ANAGRPS); nvmet_ana_group_enabled[grpid]++; up_write(&nvmet_ana_sem); @@ -1731,6 +1981,7 @@ static void nvmet_port_release(struct config_item *item) flush_workqueue(nvmet_wq); list_del(&port->global_entry); + key_put(port->keyring); kfree(port->ana_state); kfree(port); } @@ -1741,7 +1992,9 @@ static struct configfs_attribute *nvmet_port_attrs[] = { &nvmet_attr_addr_traddr, &nvmet_attr_addr_trsvcid, &nvmet_attr_addr_trtype, + &nvmet_attr_addr_tsas, &nvmet_attr_param_inline_data_size, + &nvmet_attr_param_max_queue_size, #ifdef CONFIG_BLK_DEV_INTEGRITY &nvmet_attr_param_pi_enable, #endif @@ -1779,6 +2032,14 @@ static struct config_group *nvmet_ports_make(struct config_group *group, return ERR_PTR(-ENOMEM); } + if (IS_ENABLED(CONFIG_NVME_TARGET_TCP_TLS) && nvme_keyring_id()) { + port->keyring = key_lookup(nvme_keyring_id()); + if (IS_ERR(port->keyring)) { + pr_warn("NVMe keyring not available, disabling TLS\n"); + port->keyring = NULL; + } + } + for (i = 1; i <= NVMET_MAX_ANAGRPS; i++) { if (i == NVMET_DEFAULT_ANA_GRPID) port->ana_state[1] = NVME_ANA_OPTIMIZED; @@ -1792,7 +2053,9 @@ static struct config_group *nvmet_ports_make(struct config_group *group, INIT_LIST_HEAD(&port->subsystems); INIT_LIST_HEAD(&port->referrals); port->inline_data_size = -1; /* < 0 == let the transport choose */ + port->max_queue_size = -1; /* < 0 == let the transport choose */ + port->disc_addr.trtype = NVMF_TRTYPE_MAX; port->disc_addr.portid = cpu_to_le16(portid); port->disc_addr.adrfam = NVMF_ADDR_FAMILY_MAX; port->disc_addr.treq = NVMF_TREQ_DISABLE_SQFLOW; @@ -1837,11 +2100,17 @@ static struct config_group nvmet_ports_group; static ssize_t nvmet_host_dhchap_key_show(struct config_item *item, char *page) { - u8 *dhchap_secret = to_host(item)->dhchap_secret; + u8 *dhchap_secret; + ssize_t ret; + down_read(&nvmet_config_sem); + dhchap_secret = to_host(item)->dhchap_secret; if (!dhchap_secret) - return sprintf(page, "\n"); - return sprintf(page, "%s\n", dhchap_secret); + ret = sprintf(page, "\n"); + else + ret = sprintf(page, "%s\n", dhchap_secret); + up_read(&nvmet_config_sem); + return ret; } static ssize_t nvmet_host_dhchap_key_store(struct config_item *item, @@ -1865,10 +2134,16 @@ static ssize_t nvmet_host_dhchap_ctrl_key_show(struct config_item *item, char *page) { u8 *dhchap_secret = to_host(item)->dhchap_ctrl_secret; + ssize_t ret; + down_read(&nvmet_config_sem); + dhchap_secret = to_host(item)->dhchap_ctrl_secret; if (!dhchap_secret) - return sprintf(page, "\n"); - return sprintf(page, "%s\n", dhchap_secret); + ret = sprintf(page, "\n"); + else + ret = sprintf(page, "%s\n", dhchap_secret); + up_read(&nvmet_config_sem); + return ret; } static ssize_t nvmet_host_dhchap_ctrl_key_store(struct config_item *item, @@ -2006,7 +2281,56 @@ static const struct config_item_type nvmet_hosts_type = { static struct config_group nvmet_hosts_group; +static ssize_t nvmet_root_discovery_nqn_show(struct config_item *item, + char *page) +{ + return snprintf(page, PAGE_SIZE, "%s\n", nvmet_disc_subsys->subsysnqn); +} + +static ssize_t nvmet_root_discovery_nqn_store(struct config_item *item, + const char *page, size_t count) +{ + struct list_head *entry; + char *old_nqn, *new_nqn; + size_t len; + + len = strcspn(page, "\n"); + if (!len || len > NVMF_NQN_FIELD_LEN - 1) + return -EINVAL; + + new_nqn = kstrndup(page, len, GFP_KERNEL); + if (!new_nqn) + return -ENOMEM; + + down_write(&nvmet_config_sem); + list_for_each(entry, &nvmet_subsystems_group.cg_children) { + struct config_item *item = + container_of(entry, struct config_item, ci_entry); + + if (!strncmp(config_item_name(item), page, len)) { + pr_err("duplicate NQN %s\n", config_item_name(item)); + up_write(&nvmet_config_sem); + kfree(new_nqn); + return -EINVAL; + } + } + old_nqn = nvmet_disc_subsys->subsysnqn; + nvmet_disc_subsys->subsysnqn = new_nqn; + up_write(&nvmet_config_sem); + + kfree(old_nqn); + return len; +} + +CONFIGFS_ATTR(nvmet_root_, discovery_nqn); + +static struct configfs_attribute *nvmet_root_attrs[] = { + &nvmet_root_attr_discovery_nqn, + NULL, +}; + static const struct config_item_type nvmet_root_type = { + .ct_attrs = nvmet_root_attrs, .ct_owner = THIS_MODULE, }; diff --git a/drivers/nvme/target/core.c b/drivers/nvme/target/core.c index f66ed13d7c11..cc88e5a28c8a 100644 --- a/drivers/nvme/target/core.c +++ b/drivers/nvme/target/core.c @@ -16,6 +16,7 @@ #include "trace.h" #include "nvmet.h" +#include "debugfs.h" struct kmem_cache *nvmet_bvec_cache; struct workqueue_struct *buffered_io_wq; @@ -39,7 +40,7 @@ EXPORT_SYMBOL_GPL(nvmet_wq); * - the nvmet_transports array * * When updating any of those lists/structures write lock should be obtained, - * while when reading (popolating discovery log page or checking host-subsystem + * while when reading (populating discovery log page or checking host-subsystem * link) read lock is obtained to allow concurrent reads. */ DECLARE_RWSEM(nvmet_config_sem); @@ -55,20 +56,13 @@ inline u16 errno_to_nvme_status(struct nvmet_req *req, int errno) return NVME_SC_SUCCESS; case -ENOSPC: req->error_loc = offsetof(struct nvme_rw_command, length); - return NVME_SC_CAP_EXCEEDED | NVME_SC_DNR; + return NVME_SC_CAP_EXCEEDED | NVME_STATUS_DNR; case -EREMOTEIO: req->error_loc = offsetof(struct nvme_rw_command, slba); - return NVME_SC_LBA_RANGE | NVME_SC_DNR; + return NVME_SC_LBA_RANGE | NVME_STATUS_DNR; case -EOPNOTSUPP: req->error_loc = offsetof(struct nvme_common_command, opcode); - switch (req->cmd->common.opcode) { - case nvme_cmd_dsm: - case nvme_cmd_write_zeroes: - return NVME_SC_ONCS_NOT_SUPPORTED | NVME_SC_DNR; - default: - return NVME_SC_INVALID_OPCODE | NVME_SC_DNR; - } - break; + return NVME_SC_INVALID_OPCODE | NVME_STATUS_DNR; case -ENODATA: req->error_loc = offsetof(struct nvme_rw_command, nsid); return NVME_SC_ACCESS_DENIED; @@ -76,7 +70,7 @@ inline u16 errno_to_nvme_status(struct nvmet_req *req, int errno) fallthrough; default: req->error_loc = offsetof(struct nvme_common_command, opcode); - return NVME_SC_INTERNAL | NVME_SC_DNR; + return NVME_SC_INTERNAL | NVME_STATUS_DNR; } } @@ -86,7 +80,7 @@ u16 nvmet_report_invalid_opcode(struct nvmet_req *req) req->sq->qid); req->error_loc = offsetof(struct nvme_common_command, opcode); - return NVME_SC_INVALID_OPCODE | NVME_SC_DNR; + return NVME_SC_INVALID_OPCODE | NVME_STATUS_DNR; } static struct nvmet_subsys *nvmet_find_get_subsys(struct nvmet_port *port, @@ -97,7 +91,7 @@ u16 nvmet_copy_to_sgl(struct nvmet_req *req, off_t off, const void *buf, { if (sg_pcopy_from_buffer(req->sg, req->sg_cnt, buf, len, off) != len) { req->error_loc = offsetof(struct nvme_common_command, dptr); - return NVME_SC_SGL_INVALID_DATA | NVME_SC_DNR; + return NVME_SC_SGL_INVALID_DATA | NVME_STATUS_DNR; } return 0; } @@ -106,7 +100,7 @@ u16 nvmet_copy_from_sgl(struct nvmet_req *req, off_t off, void *buf, size_t len) { if (sg_pcopy_to_buffer(req->sg, req->sg_cnt, buf, len, off) != len) { req->error_loc = offsetof(struct nvme_common_command, dptr); - return NVME_SC_SGL_INVALID_DATA | NVME_SC_DNR; + return NVME_SC_SGL_INVALID_DATA | NVME_STATUS_DNR; } return 0; } @@ -115,7 +109,7 @@ u16 nvmet_zero_sgl(struct nvmet_req *req, off_t off, size_t len) { if (sg_zero_buffer(req->sg, req->sg_cnt, len, off) != len) { req->error_loc = offsetof(struct nvme_common_command, dptr); - return NVME_SC_SGL_INVALID_DATA | NVME_SC_DNR; + return NVME_SC_SGL_INVALID_DATA | NVME_STATUS_DNR; } return 0; } @@ -126,7 +120,7 @@ static u32 nvmet_max_nsid(struct nvmet_subsys *subsys) unsigned long idx; u32 nsid = 0; - xa_for_each(&subsys->namespaces, idx, cur) + nvmet_for_each_enabled_ns(&subsys->namespaces, idx, cur) nsid = cur->nsid; return nsid; @@ -145,7 +139,7 @@ static void nvmet_async_events_failall(struct nvmet_ctrl *ctrl) while (ctrl->nr_async_event_cmds) { req = ctrl->async_event_cmds[--ctrl->nr_async_event_cmds]; mutex_unlock(&ctrl->lock); - nvmet_req_complete(req, NVME_SC_INTERNAL | NVME_SC_DNR); + nvmet_req_complete(req, NVME_SC_INTERNAL | NVME_STATUS_DNR); mutex_lock(&ctrl->lock); } mutex_unlock(&ctrl->lock); @@ -248,7 +242,7 @@ void nvmet_ns_changed(struct nvmet_subsys *subsys, u32 nsid) nvmet_add_to_changed_ns_log(ctrl, cpu_to_le32(nsid)); if (nvmet_aen_bit_disabled(ctrl, NVME_AEN_BIT_NS_ATTR)) continue; - nvmet_add_async_event(ctrl, NVME_AER_TYPE_NOTICE, + nvmet_add_async_event(ctrl, NVME_AER_NOTICE, NVME_AER_NOTICE_NS_CHANGED, NVME_LOG_CHANGED_NS); } @@ -265,7 +259,7 @@ void nvmet_send_ana_event(struct nvmet_subsys *subsys, continue; if (nvmet_aen_bit_disabled(ctrl, NVME_AEN_BIT_ANA_CHANGE)) continue; - nvmet_add_async_event(ctrl, NVME_AER_TYPE_NOTICE, + nvmet_add_async_event(ctrl, NVME_AER_NOTICE, NVME_AER_NOTICE_ANA, NVME_LOG_ANA); } mutex_unlock(&subsys->lock); @@ -323,6 +317,9 @@ int nvmet_enable_port(struct nvmet_port *port) lockdep_assert_held(&nvmet_config_sem); + if (port->disc_addr.trtype == NVMF_TRTYPE_MAX) + return -EINVAL; + ops = nvmet_transports[port->disc_addr.trtype]; if (!ops) { up_write(&nvmet_config_sem); @@ -358,6 +355,18 @@ int nvmet_enable_port(struct nvmet_port *port) if (port->inline_data_size < 0) port->inline_data_size = 0; + /* + * If the transport didn't set the max_queue_size properly, then clamp + * it to the target limits. Also set default values in case the + * transport didn't set it at all. + */ + if (port->max_queue_size < 0) + port->max_queue_size = NVMET_MAX_QUEUE_SIZE; + else + port->max_queue_size = clamp_t(int, port->max_queue_size, + NVMET_MIN_QUEUE_SIZE, + NVMET_MAX_QUEUE_SIZE); + port->enabled = true; port->tr_ops = ops; return 0; @@ -425,11 +434,17 @@ void nvmet_stop_keep_alive_timer(struct nvmet_ctrl *ctrl) u16 nvmet_req_find_ns(struct nvmet_req *req) { u32 nsid = le32_to_cpu(req->cmd->common.nsid); + struct nvmet_subsys *subsys = nvmet_req_subsys(req); - req->ns = xa_load(&nvmet_req_subsys(req)->namespaces, nsid); - if (unlikely(!req->ns)) { + req->ns = xa_load(&subsys->namespaces, nsid); + if (unlikely(!req->ns || !req->ns->enabled)) { req->error_loc = offsetof(struct nvme_common_command, nsid); - return NVME_SC_INVALID_NS | NVME_SC_DNR; + if (!req->ns) /* ns doesn't exist! */ + return NVME_SC_INVALID_NS | NVME_STATUS_DNR; + + /* ns exists but it's disabled */ + req->ns = NULL; + return NVME_SC_INTERNAL_PATH_ERROR; } percpu_ref_get(&req->ns->ref); @@ -498,9 +513,6 @@ static int nvmet_p2pmem_ns_enable(struct nvmet_ns *ns) return 0; } -/* - * Note: ctrl->subsys->lock should be held when calling this function - */ static void nvmet_p2pmem_ns_add_p2p(struct nvmet_ctrl *ctrl, struct nvmet_ns *ns) { @@ -508,6 +520,8 @@ static void nvmet_p2pmem_ns_add_p2p(struct nvmet_ctrl *ctrl, struct pci_dev *p2p_dev; int ret; + lockdep_assert_held(&ctrl->subsys->lock); + if (!ctrl->p2p_client || !ns->use_p2pmem) return; @@ -566,10 +580,6 @@ int nvmet_ns_enable(struct nvmet_ns *ns) if (ns->enabled) goto out_unlock; - ret = -EMFILE; - if (subsys->nr_namespaces == NVMET_MAX_NAMESPACES) - goto out_unlock; - ret = nvmet_bdev_ns_enable(ns); if (ret == -ENOTBLK) ret = nvmet_file_ns_enable(ns); @@ -583,30 +593,25 @@ int nvmet_ns_enable(struct nvmet_ns *ns) list_for_each_entry(ctrl, &subsys->ctrls, subsys_entry) nvmet_p2pmem_ns_add_p2p(ctrl, ns); - ret = percpu_ref_init(&ns->ref, nvmet_destroy_namespace, - 0, GFP_KERNEL); - if (ret) - goto out_dev_put; - - if (ns->nsid > subsys->max_nsid) - subsys->max_nsid = ns->nsid; - - ret = xa_insert(&subsys->namespaces, ns->nsid, ns, GFP_KERNEL); - if (ret) - goto out_restore_subsys_maxnsid; + if (ns->pr.enable) { + ret = nvmet_pr_init_ns(ns); + if (ret) + goto out_dev_put; + } - subsys->nr_namespaces++; + if (percpu_ref_init(&ns->ref, nvmet_destroy_namespace, 0, GFP_KERNEL)) + goto out_pr_exit; nvmet_ns_changed(subsys, ns->nsid); ns->enabled = true; + xa_set_mark(&subsys->namespaces, ns->nsid, NVMET_NS_ENABLED); ret = 0; out_unlock: mutex_unlock(&subsys->lock); return ret; - -out_restore_subsys_maxnsid: - subsys->max_nsid = nvmet_max_nsid(subsys); - percpu_ref_exit(&ns->ref); +out_pr_exit: + if (ns->pr.enable) + nvmet_pr_exit_ns(ns); out_dev_put: list_for_each_entry(ctrl, &subsys->ctrls, subsys_entry) pci_dev_put(radix_tree_delete(&ctrl->p2p_ns_map, ns->nsid)); @@ -625,9 +630,7 @@ void nvmet_ns_disable(struct nvmet_ns *ns) goto out_unlock; ns->enabled = false; - xa_erase(&ns->subsys->namespaces, ns->nsid); - if (ns->nsid == subsys->max_nsid) - subsys->max_nsid = nvmet_max_nsid(subsys); + xa_clear_mark(&subsys->namespaces, ns->nsid, NVMET_NS_ENABLED); list_for_each_entry(ctrl, &subsys->ctrls, subsys_entry) pci_dev_put(radix_tree_delete(&ctrl->p2p_ns_map, ns->nsid)); @@ -638,7 +641,7 @@ void nvmet_ns_disable(struct nvmet_ns *ns) * Now that we removed the namespaces from the lookup list, we * can kill the per_cpu ref and wait for any remaining references * to be dropped, as well as a RCU grace period for anyone only - * using the namepace under rcu_read_lock(). Note that we can't + * using the namespace under rcu_read_lock(). Note that we can't * use call_rcu here as we need to ensure the namespaces have * been fully destroyed before unloading the module. */ @@ -647,9 +650,10 @@ void nvmet_ns_disable(struct nvmet_ns *ns) wait_for_completion(&ns->disable_done); percpu_ref_exit(&ns->ref); - mutex_lock(&subsys->lock); + if (ns->pr.enable) + nvmet_pr_exit_ns(ns); - subsys->nr_namespaces--; + mutex_lock(&subsys->lock); nvmet_ns_changed(subsys, ns->nsid); nvmet_ns_dev_disable(ns); out_unlock: @@ -658,8 +662,19 @@ out_unlock: void nvmet_ns_free(struct nvmet_ns *ns) { + struct nvmet_subsys *subsys = ns->subsys; + nvmet_ns_disable(ns); + mutex_lock(&subsys->lock); + + xa_erase(&subsys->namespaces, ns->nsid); + if (ns->nsid == subsys->max_nsid) + subsys->max_nsid = nvmet_max_nsid(subsys); + + subsys->nr_namespaces--; + mutex_unlock(&subsys->lock); + down_write(&nvmet_ana_sem); nvmet_ana_group_enabled[ns->anagrpid]--; up_write(&nvmet_ana_sem); @@ -672,15 +687,30 @@ struct nvmet_ns *nvmet_ns_alloc(struct nvmet_subsys *subsys, u32 nsid) { struct nvmet_ns *ns; + mutex_lock(&subsys->lock); + + if (subsys->nr_namespaces == NVMET_MAX_NAMESPACES) + goto out_unlock; + ns = kzalloc(sizeof(*ns), GFP_KERNEL); if (!ns) - return NULL; + goto out_unlock; init_completion(&ns->disable_done); ns->nsid = nsid; ns->subsys = subsys; + if (ns->nsid > subsys->max_nsid) + subsys->max_nsid = nsid; + + if (xa_insert(&subsys->namespaces, ns->nsid, ns, GFP_KERNEL)) + goto out_exit; + + subsys->nr_namespaces++; + + mutex_unlock(&subsys->lock); + down_write(&nvmet_ana_sem); ns->anagrpid = NVMET_DEFAULT_ANA_GRPID; nvmet_ana_group_enabled[ns->anagrpid]++; @@ -691,6 +721,12 @@ struct nvmet_ns *nvmet_ns_alloc(struct nvmet_subsys *subsys, u32 nsid) ns->csi = NVME_CSI_NVM; return ns; +out_exit: + subsys->max_nsid = nvmet_max_nsid(subsys); + kfree(ns); +out_unlock: + mutex_unlock(&subsys->lock); + return NULL; } static void nvmet_update_sq_head(struct nvmet_req *req) @@ -738,6 +774,7 @@ static void nvmet_set_error(struct nvmet_req *req, u16 status) static void __nvmet_req_complete(struct nvmet_req *req, u16 status) { struct nvmet_ns *ns = req->ns; + struct nvmet_pr_per_ctrl_ref *pc_ref = req->pc_ref; if (!req->sq->sqhd_disabled) nvmet_update_sq_head(req); @@ -750,22 +787,59 @@ static void __nvmet_req_complete(struct nvmet_req *req, u16 status) trace_nvmet_req_complete(req); req->ops->queue_response(req); + + if (pc_ref) + nvmet_pr_put_ns_pc_ref(pc_ref); if (ns) nvmet_put_namespace(ns); } void nvmet_req_complete(struct nvmet_req *req, u16 status) { + struct nvmet_sq *sq = req->sq; + __nvmet_req_complete(req, status); - percpu_ref_put(&req->sq->ref); + percpu_ref_put(&sq->ref); } EXPORT_SYMBOL_GPL(nvmet_req_complete); +void nvmet_cq_init(struct nvmet_cq *cq) +{ + refcount_set(&cq->ref, 1); +} +EXPORT_SYMBOL_GPL(nvmet_cq_init); + +bool nvmet_cq_get(struct nvmet_cq *cq) +{ + return refcount_inc_not_zero(&cq->ref); +} +EXPORT_SYMBOL_GPL(nvmet_cq_get); + +void nvmet_cq_put(struct nvmet_cq *cq) +{ + if (refcount_dec_and_test(&cq->ref)) + nvmet_cq_destroy(cq); +} +EXPORT_SYMBOL_GPL(nvmet_cq_put); + void nvmet_cq_setup(struct nvmet_ctrl *ctrl, struct nvmet_cq *cq, u16 qid, u16 size) { cq->qid = qid; cq->size = size; + + ctrl->cqs[qid] = cq; +} + +void nvmet_cq_destroy(struct nvmet_cq *cq) +{ + struct nvmet_ctrl *ctrl = cq->ctrl; + + if (ctrl) { + ctrl->cqs[cq->qid] = NULL; + nvmet_ctrl_put(cq->ctrl); + cq->ctrl = NULL; + } } void nvmet_sq_setup(struct nvmet_ctrl *ctrl, struct nvmet_sq *sq, @@ -785,6 +859,99 @@ static void nvmet_confirm_sq(struct percpu_ref *ref) complete(&sq->confirm_done); } +u16 nvmet_check_cqid(struct nvmet_ctrl *ctrl, u16 cqid, bool create) +{ + if (!ctrl->cqs) + return NVME_SC_INTERNAL | NVME_STATUS_DNR; + + if (cqid > ctrl->subsys->max_qid) + return NVME_SC_QID_INVALID | NVME_STATUS_DNR; + + if ((create && ctrl->cqs[cqid]) || (!create && !ctrl->cqs[cqid])) + return NVME_SC_QID_INVALID | NVME_STATUS_DNR; + + return NVME_SC_SUCCESS; +} + +u16 nvmet_check_io_cqid(struct nvmet_ctrl *ctrl, u16 cqid, bool create) +{ + if (!cqid) + return NVME_SC_QID_INVALID | NVME_STATUS_DNR; + return nvmet_check_cqid(ctrl, cqid, create); +} + +bool nvmet_cq_in_use(struct nvmet_cq *cq) +{ + return refcount_read(&cq->ref) > 1; +} +EXPORT_SYMBOL_GPL(nvmet_cq_in_use); + +u16 nvmet_cq_create(struct nvmet_ctrl *ctrl, struct nvmet_cq *cq, + u16 qid, u16 size) +{ + u16 status; + + status = nvmet_check_cqid(ctrl, qid, true); + if (status != NVME_SC_SUCCESS) + return status; + + if (!kref_get_unless_zero(&ctrl->ref)) + return NVME_SC_INTERNAL | NVME_STATUS_DNR; + cq->ctrl = ctrl; + + nvmet_cq_init(cq); + nvmet_cq_setup(ctrl, cq, qid, size); + + return NVME_SC_SUCCESS; +} +EXPORT_SYMBOL_GPL(nvmet_cq_create); + +u16 nvmet_check_sqid(struct nvmet_ctrl *ctrl, u16 sqid, + bool create) +{ + if (!ctrl->sqs) + return NVME_SC_INTERNAL | NVME_STATUS_DNR; + + if (sqid > ctrl->subsys->max_qid) + return NVME_SC_QID_INVALID | NVME_STATUS_DNR; + + if ((create && ctrl->sqs[sqid]) || + (!create && !ctrl->sqs[sqid])) + return NVME_SC_QID_INVALID | NVME_STATUS_DNR; + + return NVME_SC_SUCCESS; +} + +u16 nvmet_sq_create(struct nvmet_ctrl *ctrl, struct nvmet_sq *sq, + struct nvmet_cq *cq, u16 sqid, u16 size) +{ + u16 status; + int ret; + + if (!kref_get_unless_zero(&ctrl->ref)) + return NVME_SC_INTERNAL | NVME_STATUS_DNR; + + status = nvmet_check_sqid(ctrl, sqid, true); + if (status != NVME_SC_SUCCESS) + return status; + + ret = nvmet_sq_init(sq, cq); + if (ret) { + status = NVME_SC_INTERNAL | NVME_STATUS_DNR; + goto ctrl_put; + } + + nvmet_sq_setup(ctrl, sq, sqid, size); + sq->ctrl = ctrl; + + return NVME_SC_SUCCESS; + +ctrl_put: + nvmet_ctrl_put(ctrl); + return status; +} +EXPORT_SYMBOL_GPL(nvmet_sq_create); + void nvmet_sq_destroy(struct nvmet_sq *sq) { struct nvmet_ctrl *ctrl = sq->ctrl; @@ -800,6 +967,16 @@ void nvmet_sq_destroy(struct nvmet_sq *sq) wait_for_completion(&sq->free_done); percpu_ref_exit(&sq->ref); nvmet_auth_sq_free(sq); + nvmet_cq_put(sq->cq); + + /* + * we must reference the ctrl again after waiting for inflight IO + * to complete. Because admin connect may have sneaked in after we + * store sq->ctrl locally, but before we killed the percpu_ref. the + * admin connect allocates and assigns sq->ctrl, which now needs a + * final ref put, as this ctrl is going away. + */ + ctrl = sq->ctrl; if (ctrl) { /* @@ -823,18 +1000,23 @@ static void nvmet_sq_free(struct percpu_ref *ref) complete(&sq->free_done); } -int nvmet_sq_init(struct nvmet_sq *sq) +int nvmet_sq_init(struct nvmet_sq *sq, struct nvmet_cq *cq) { int ret; + if (!nvmet_cq_get(cq)) + return -EINVAL; + ret = percpu_ref_init(&sq->ref, nvmet_sq_free, 0, GFP_KERNEL); if (ret) { pr_err("percpu_ref init failed!\n"); + nvmet_cq_put(cq); return ret; } init_completion(&sq->free_done); init_completion(&sq->confirm_done); nvmet_auth_sq_init(sq); + sq->cq = cq; return 0; } @@ -869,6 +1051,33 @@ static inline u16 nvmet_io_cmd_check_access(struct nvmet_req *req) return 0; } +static u32 nvmet_io_cmd_transfer_len(struct nvmet_req *req) +{ + struct nvme_command *cmd = req->cmd; + u32 metadata_len = 0; + + if (nvme_is_fabrics(cmd)) + return nvmet_fabrics_io_cmd_data_len(req); + + if (!req->ns) + return 0; + + switch (req->cmd->common.opcode) { + case nvme_cmd_read: + case nvme_cmd_write: + case nvme_cmd_zone_append: + if (req->sq->ctrl->pi_support && nvmet_ns_has_pi(req->ns)) + metadata_len = nvmet_rw_metadata_len(req); + return nvmet_rw_data_len(req) + metadata_len; + case nvme_cmd_dsm: + return nvmet_dsm_len(req); + case nvme_cmd_zone_mgmt_recv: + return (le32_to_cpu(req->cmd->zmr.numd) + 1) << 2; + default: + return 0; + } +} + static u16 nvmet_parse_io_cmd(struct nvmet_req *req) { struct nvme_command *cmd = req->cmd; @@ -878,7 +1087,7 @@ static u16 nvmet_parse_io_cmd(struct nvmet_req *req) return nvmet_parse_fabrics_io_cmd(req); if (unlikely(!nvmet_check_auth_status(req))) - return NVME_SC_AUTH_REQUIRED | NVME_SC_DNR; + return NVME_SC_AUTH_REQUIRED | NVME_STATUS_DNR; ret = nvmet_check_ctrl_status(req); if (unlikely(ret)) @@ -902,27 +1111,48 @@ static u16 nvmet_parse_io_cmd(struct nvmet_req *req) return ret; } + if (req->ns->pr.enable) { + ret = nvmet_parse_pr_cmd(req); + if (!ret) + return ret; + } + switch (req->ns->csi) { case NVME_CSI_NVM: if (req->ns->file) - return nvmet_file_parse_io_cmd(req); - return nvmet_bdev_parse_io_cmd(req); + ret = nvmet_file_parse_io_cmd(req); + else + ret = nvmet_bdev_parse_io_cmd(req); + break; case NVME_CSI_ZNS: if (IS_ENABLED(CONFIG_BLK_DEV_ZONED)) - return nvmet_bdev_zns_parse_io_cmd(req); - return NVME_SC_INVALID_IO_CMD_SET; + ret = nvmet_bdev_zns_parse_io_cmd(req); + else + ret = NVME_SC_INVALID_IO_CMD_SET; + break; default: - return NVME_SC_INVALID_IO_CMD_SET; + ret = NVME_SC_INVALID_IO_CMD_SET; + } + if (ret) + return ret; + + if (req->ns->pr.enable) { + ret = nvmet_pr_check_cmd_access(req); + if (ret) + return ret; + + ret = nvmet_pr_get_ns_pc_ref(req); } + return ret; } -bool nvmet_req_init(struct nvmet_req *req, struct nvmet_cq *cq, - struct nvmet_sq *sq, const struct nvmet_fabrics_ops *ops) +bool nvmet_req_init(struct nvmet_req *req, struct nvmet_sq *sq, + const struct nvmet_fabrics_ops *ops) { u8 flags = req->cmd->common.flags; u16 status; - req->cq = cq; + req->cq = sq->cq; req->sq = sq; req->ops = ops; req->sg = NULL; @@ -931,28 +1161,33 @@ bool nvmet_req_init(struct nvmet_req *req, struct nvmet_cq *cq, req->metadata_sg_cnt = 0; req->transfer_len = 0; req->metadata_len = 0; + req->cqe->result.u64 = 0; req->cqe->status = 0; req->cqe->sq_head = 0; req->ns = NULL; req->error_loc = NVMET_NO_ERROR_LOC; req->error_slba = 0; + req->pc_ref = NULL; /* no support for fused commands yet */ if (unlikely(flags & (NVME_CMD_FUSE_FIRST | NVME_CMD_FUSE_SECOND))) { req->error_loc = offsetof(struct nvme_common_command, flags); - status = NVME_SC_INVALID_FIELD | NVME_SC_DNR; + status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; goto fail; } /* * For fabrics, PSDT field shall describe metadata pointer (MPTR) that * contains an address of a single contiguous physical buffer that is - * byte aligned. + * byte aligned. For PCI controllers, this is optional so not enforced. */ if (unlikely((flags & NVME_CMD_SGL_ALL) != NVME_CMD_SGL_METABUF)) { - req->error_loc = offsetof(struct nvme_common_command, flags); - status = NVME_SC_INVALID_FIELD | NVME_SC_DNR; - goto fail; + if (!req->sq->ctrl || !nvmet_is_pci_ctrl(req->sq->ctrl)) { + req->error_loc = + offsetof(struct nvme_common_command, flags); + status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; + goto fail; + } } if (unlikely(!req->sq->ctrl)) @@ -969,7 +1204,7 @@ bool nvmet_req_init(struct nvmet_req *req, struct nvmet_cq *cq, trace_nvmet_req_init(req, req->cmd); if (unlikely(!percpu_ref_tryget_live(&sq->ref))) { - status = NVME_SC_INVALID_FIELD | NVME_SC_DNR; + status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; goto fail; } @@ -987,16 +1222,34 @@ EXPORT_SYMBOL_GPL(nvmet_req_init); void nvmet_req_uninit(struct nvmet_req *req) { percpu_ref_put(&req->sq->ref); + if (req->pc_ref) + nvmet_pr_put_ns_pc_ref(req->pc_ref); if (req->ns) nvmet_put_namespace(req->ns); } EXPORT_SYMBOL_GPL(nvmet_req_uninit); +size_t nvmet_req_transfer_len(struct nvmet_req *req) +{ + if (likely(req->sq->qid != 0)) + return nvmet_io_cmd_transfer_len(req); + if (unlikely(!req->sq->ctrl)) + return nvmet_connect_cmd_data_len(req); + return nvmet_admin_cmd_data_len(req); +} +EXPORT_SYMBOL_GPL(nvmet_req_transfer_len); + bool nvmet_check_transfer_len(struct nvmet_req *req, size_t len) { if (unlikely(len != req->transfer_len)) { + u16 status; + req->error_loc = offsetof(struct nvme_common_command, dptr); - nvmet_req_complete(req, NVME_SC_SGL_INVALID_DATA | NVME_SC_DNR); + if (req->cmd->common.flags & NVME_CMD_SGL_ALL) + status = NVME_SC_SGL_INVALID_DATA; + else + status = NVME_SC_INVALID_FIELD; + nvmet_req_complete(req, status | NVME_STATUS_DNR); return false; } @@ -1007,8 +1260,14 @@ EXPORT_SYMBOL_GPL(nvmet_check_transfer_len); bool nvmet_check_data_len_lte(struct nvmet_req *req, size_t data_len) { if (unlikely(data_len > req->transfer_len)) { + u16 status; + req->error_loc = offsetof(struct nvme_common_command, dptr); - nvmet_req_complete(req, NVME_SC_SGL_INVALID_DATA | NVME_SC_DNR); + if (req->cmd->common.flags & NVME_CMD_SGL_ALL) + status = NVME_SC_SGL_INVALID_DATA; + else + status = NVME_SC_INVALID_FIELD; + nvmet_req_complete(req, status | NVME_STATUS_DNR); return false; } @@ -1099,41 +1358,6 @@ void nvmet_req_free_sgls(struct nvmet_req *req) } EXPORT_SYMBOL_GPL(nvmet_req_free_sgls); -static inline bool nvmet_cc_en(u32 cc) -{ - return (cc >> NVME_CC_EN_SHIFT) & 0x1; -} - -static inline u8 nvmet_cc_css(u32 cc) -{ - return (cc >> NVME_CC_CSS_SHIFT) & 0x7; -} - -static inline u8 nvmet_cc_mps(u32 cc) -{ - return (cc >> NVME_CC_MPS_SHIFT) & 0xf; -} - -static inline u8 nvmet_cc_ams(u32 cc) -{ - return (cc >> NVME_CC_AMS_SHIFT) & 0x7; -} - -static inline u8 nvmet_cc_shn(u32 cc) -{ - return (cc >> NVME_CC_SHN_SHIFT) & 0x3; -} - -static inline u8 nvmet_cc_iosqes(u32 cc) -{ - return (cc >> NVME_CC_IOSQES_SHIFT) & 0xf; -} - -static inline u8 nvmet_cc_iocqes(u32 cc) -{ - return (cc >> NVME_CC_IOCQES_SHIFT) & 0xf; -} - static inline bool nvmet_css_supported(u8 cc_css) { switch (cc_css << NVME_CC_CSS_SHIFT) { @@ -1210,6 +1434,7 @@ void nvmet_update_cc(struct nvmet_ctrl *ctrl, u32 new) ctrl->csts &= ~NVME_CSTS_SHST_CMPLT; mutex_unlock(&ctrl->lock); } +EXPORT_SYMBOL_GPL(nvmet_update_cc); static void nvmet_init_cap(struct nvmet_ctrl *ctrl) { @@ -1221,9 +1446,10 @@ static void nvmet_init_cap(struct nvmet_ctrl *ctrl) ctrl->cap |= (15ULL << 24); /* maximum queue entries supported: */ if (ctrl->ops->get_max_queue_size) - ctrl->cap |= ctrl->ops->get_max_queue_size(ctrl) - 1; + ctrl->cap |= min_t(u16, ctrl->ops->get_max_queue_size(ctrl), + ctrl->port->max_queue_size) - 1; else - ctrl->cap |= NVMET_QUEUE_SIZE - 1; + ctrl->cap |= ctrl->port->max_queue_size - 1; if (nvmet_is_passthru_subsys(ctrl->subsys)) nvmet_passthrough_override_cap(ctrl); @@ -1276,18 +1502,18 @@ u16 nvmet_check_ctrl_status(struct nvmet_req *req) if (unlikely(!(req->sq->ctrl->cc & NVME_CC_ENABLE))) { pr_err("got cmd %d while CC.EN == 0 on qid = %d\n", req->cmd->common.opcode, req->sq->qid); - return NVME_SC_CMD_SEQ_ERROR | NVME_SC_DNR; + return NVME_SC_CMD_SEQ_ERROR | NVME_STATUS_DNR; } if (unlikely(!(req->sq->ctrl->csts & NVME_CSTS_RDY))) { pr_err("got cmd %d while CSTS.RDY == 0 on qid = %d\n", req->cmd->common.opcode, req->sq->qid); - return NVME_SC_CMD_SEQ_ERROR | NVME_SC_DNR; + return NVME_SC_CMD_SEQ_ERROR | NVME_STATUS_DNR; } if (unlikely(!nvmet_check_auth_status(req))) { pr_warn("qid %d not authenticated\n", req->sq->qid); - return NVME_SC_AUTH_REQUIRED | NVME_SC_DNR; + return NVME_SC_AUTH_REQUIRED | NVME_STATUS_DNR; } return 0; } @@ -1312,32 +1538,30 @@ bool nvmet_host_allowed(struct nvmet_subsys *subsys, const char *hostnqn) return false; } -/* - * Note: ctrl->subsys->lock should be held when calling this function - */ static void nvmet_setup_p2p_ns_map(struct nvmet_ctrl *ctrl, - struct nvmet_req *req) + struct device *p2p_client) { struct nvmet_ns *ns; unsigned long idx; - if (!req->p2p_client) + lockdep_assert_held(&ctrl->subsys->lock); + + if (!p2p_client) return; - ctrl->p2p_client = get_device(req->p2p_client); + ctrl->p2p_client = get_device(p2p_client); - xa_for_each(&ctrl->subsys->namespaces, idx, ns) + nvmet_for_each_enabled_ns(&ctrl->subsys->namespaces, idx, ns) nvmet_p2pmem_ns_add_p2p(ctrl, ns); } -/* - * Note: ctrl->subsys->lock should be held when calling this function - */ static void nvmet_release_p2p_ns_map(struct nvmet_ctrl *ctrl) { struct radix_tree_iter iter; void __rcu **slot; + lockdep_assert_held(&ctrl->subsys->lock); + radix_tree_for_each_slot(slot, &ctrl->p2p_ns_map, &iter, 0) pci_dev_put(radix_tree_deref_slot(slot)); @@ -1353,44 +1577,44 @@ static void nvmet_fatal_error_handler(struct work_struct *work) ctrl->ops->delete_ctrl(ctrl); } -u16 nvmet_alloc_ctrl(const char *subsysnqn, const char *hostnqn, - struct nvmet_req *req, u32 kato, struct nvmet_ctrl **ctrlp) +struct nvmet_ctrl *nvmet_alloc_ctrl(struct nvmet_alloc_ctrl_args *args) { struct nvmet_subsys *subsys; struct nvmet_ctrl *ctrl; + u32 kato = args->kato; + u8 dhchap_status; int ret; - u16 status; - status = NVME_SC_CONNECT_INVALID_PARAM | NVME_SC_DNR; - subsys = nvmet_find_get_subsys(req->port, subsysnqn); + args->status = NVME_SC_CONNECT_INVALID_PARAM | NVME_STATUS_DNR; + subsys = nvmet_find_get_subsys(args->port, args->subsysnqn); if (!subsys) { pr_warn("connect request for invalid subsystem %s!\n", - subsysnqn); - req->cqe->result.u32 = IPO_IATTR_CONNECT_DATA(subsysnqn); - req->error_loc = offsetof(struct nvme_common_command, dptr); - goto out; + args->subsysnqn); + args->result = IPO_IATTR_CONNECT_DATA(subsysnqn); + args->error_loc = offsetof(struct nvme_common_command, dptr); + return NULL; } down_read(&nvmet_config_sem); - if (!nvmet_host_allowed(subsys, hostnqn)) { + if (!nvmet_host_allowed(subsys, args->hostnqn)) { pr_info("connect by host %s for subsystem %s not allowed\n", - hostnqn, subsysnqn); - req->cqe->result.u32 = IPO_IATTR_CONNECT_DATA(hostnqn); + args->hostnqn, args->subsysnqn); + args->result = IPO_IATTR_CONNECT_DATA(hostnqn); up_read(&nvmet_config_sem); - status = NVME_SC_CONNECT_INVALID_HOST | NVME_SC_DNR; - req->error_loc = offsetof(struct nvme_common_command, dptr); + args->status = NVME_SC_CONNECT_INVALID_HOST | NVME_STATUS_DNR; + args->error_loc = offsetof(struct nvme_common_command, dptr); goto out_put_subsystem; } up_read(&nvmet_config_sem); - status = NVME_SC_INTERNAL; + args->status = NVME_SC_INTERNAL; ctrl = kzalloc(sizeof(*ctrl), GFP_KERNEL); if (!ctrl) goto out_put_subsystem; mutex_init(&ctrl->lock); - ctrl->port = req->port; - ctrl->ops = req->ops; + ctrl->port = args->port; + ctrl->ops = args->ops; #ifdef CONFIG_NVME_TARGET_PASSTHRU /* By default, set loop targets to clear IDS by default */ @@ -1404,11 +1628,11 @@ u16 nvmet_alloc_ctrl(const char *subsysnqn, const char *hostnqn, INIT_WORK(&ctrl->fatal_err_work, nvmet_fatal_error_handler); INIT_DELAYED_WORK(&ctrl->ka_work, nvmet_keep_alive_timer); - memcpy(ctrl->subsysnqn, subsysnqn, NVMF_NQN_SIZE); - memcpy(ctrl->hostnqn, hostnqn, NVMF_NQN_SIZE); + memcpy(ctrl->hostnqn, args->hostnqn, NVMF_NQN_SIZE); kref_init(&ctrl->ref); ctrl->subsys = subsys; + ctrl->pi_support = ctrl->port->pi_enable && ctrl->subsys->pi_support; nvmet_init_cap(ctrl); WRITE_ONCE(ctrl->aen_enabled, NVMET_AEN_CFG_OPTIONAL); @@ -1423,15 +1647,17 @@ u16 nvmet_alloc_ctrl(const char *subsysnqn, const char *hostnqn, if (!ctrl->sqs) goto out_free_changed_ns_list; - if (subsys->cntlid_min > subsys->cntlid_max) + ctrl->cqs = kcalloc(subsys->max_qid + 1, sizeof(struct nvmet_cq *), + GFP_KERNEL); + if (!ctrl->cqs) goto out_free_sqs; ret = ida_alloc_range(&cntlid_ida, subsys->cntlid_min, subsys->cntlid_max, GFP_KERNEL); if (ret < 0) { - status = NVME_SC_CONNECT_CTRL_BUSY | NVME_SC_DNR; - goto out_free_sqs; + args->status = NVME_SC_CONNECT_CTRL_BUSY | NVME_STATUS_DNR; + goto out_free_cqs; } ctrl->cntlid = ret; @@ -1451,13 +1677,47 @@ u16 nvmet_alloc_ctrl(const char *subsysnqn, const char *hostnqn, nvmet_start_keep_alive_timer(ctrl); mutex_lock(&subsys->lock); + ret = nvmet_ctrl_init_pr(ctrl); + if (ret) + goto init_pr_fail; list_add_tail(&ctrl->subsys_entry, &subsys->ctrls); - nvmet_setup_p2p_ns_map(ctrl, req); + nvmet_setup_p2p_ns_map(ctrl, args->p2p_client); + nvmet_debugfs_ctrl_setup(ctrl); mutex_unlock(&subsys->lock); - *ctrlp = ctrl; - return 0; + if (args->hostid) + uuid_copy(&ctrl->hostid, args->hostid); + dhchap_status = nvmet_setup_auth(ctrl, args->sq); + if (dhchap_status) { + pr_err("Failed to setup authentication, dhchap status %u\n", + dhchap_status); + nvmet_ctrl_put(ctrl); + if (dhchap_status == NVME_AUTH_DHCHAP_FAILURE_FAILED) + args->status = + NVME_SC_CONNECT_INVALID_HOST | NVME_STATUS_DNR; + else + args->status = NVME_SC_INTERNAL; + return NULL; + } + + args->status = NVME_SC_SUCCESS; + + pr_info("Created %s controller %d for subsystem %s for NQN %s%s%s%s.\n", + nvmet_is_disc_subsys(ctrl->subsys) ? "discovery" : "nvm", + ctrl->cntlid, ctrl->subsys->subsysnqn, ctrl->hostnqn, + ctrl->pi_support ? " T10-PI is enabled" : "", + nvmet_has_auth(ctrl, args->sq) ? " with DH-HMAC-CHAP" : "", + nvmet_queue_tls_keyid(args->sq) ? ", TLS" : ""); + + return ctrl; + +init_pr_fail: + mutex_unlock(&subsys->lock); + nvmet_stop_keep_alive_timer(ctrl); + ida_free(&cntlid_ida, ctrl->cntlid); +out_free_cqs: + kfree(ctrl->cqs); out_free_sqs: kfree(ctrl->sqs); out_free_changed_ns_list: @@ -1466,9 +1726,9 @@ out_free_ctrl: kfree(ctrl); out_put_subsystem: nvmet_subsys_put(subsys); -out: - return status; + return NULL; } +EXPORT_SYMBOL_GPL(nvmet_alloc_ctrl); static void nvmet_ctrl_free(struct kref *ref) { @@ -1476,6 +1736,7 @@ static void nvmet_ctrl_free(struct kref *ref) struct nvmet_subsys *subsys = ctrl->subsys; mutex_lock(&subsys->lock); + nvmet_ctrl_destroy_pr(ctrl); nvmet_release_p2p_ns_map(ctrl); list_del(&ctrl->subsys_entry); mutex_unlock(&subsys->lock); @@ -1487,10 +1748,13 @@ static void nvmet_ctrl_free(struct kref *ref) nvmet_destroy_auth(ctrl); + nvmet_debugfs_ctrl_free(ctrl); + ida_free(&cntlid_ida, ctrl->cntlid); nvmet_async_events_free(ctrl); kfree(ctrl->sqs); + kfree(ctrl->cqs); kfree(ctrl->changed_ns_list); kfree(ctrl); @@ -1501,6 +1765,7 @@ void nvmet_ctrl_put(struct nvmet_ctrl *ctrl) { kref_put(&ctrl->ref, nvmet_ctrl_free); } +EXPORT_SYMBOL_GPL(nvmet_ctrl_put); void nvmet_ctrl_fatal_error(struct nvmet_ctrl *ctrl) { @@ -1513,6 +1778,14 @@ void nvmet_ctrl_fatal_error(struct nvmet_ctrl *ctrl) } EXPORT_SYMBOL_GPL(nvmet_ctrl_fatal_error); +ssize_t nvmet_ctrl_host_traddr(struct nvmet_ctrl *ctrl, + char *traddr, size_t traddr_len) +{ + if (!ctrl->ops->host_traddr) + return -EOPNOTSUPP; + return ctrl->ops->host_traddr(ctrl, traddr, traddr_len); +} + static struct nvmet_subsys *nvmet_find_get_subsys(struct nvmet_port *port, const char *subsysnqn) { @@ -1528,6 +1801,13 @@ static struct nvmet_subsys *nvmet_find_get_subsys(struct nvmet_port *port, } down_read(&nvmet_config_sem); + if (!strncmp(nvmet_disc_subsys->subsysnqn, subsysnqn, + NVMF_NQN_SIZE)) { + if (kref_get_unless_zero(&nvmet_disc_subsys->ref)) { + up_read(&nvmet_config_sem); + return nvmet_disc_subsys; + } + } list_for_each_entry(p, &port->subsystems, entry) { if (!strncmp(p->subsys->subsysnqn, subsysnqn, NVMF_NQN_SIZE)) { @@ -1600,8 +1880,14 @@ struct nvmet_subsys *nvmet_subsys_alloc(const char *subsysnqn, INIT_LIST_HEAD(&subsys->ctrls); INIT_LIST_HEAD(&subsys->hosts); + ret = nvmet_debugfs_subsys_setup(subsys); + if (ret) + goto free_subsysnqn; + return subsys; +free_subsysnqn: + kfree(subsys->subsysnqn); free_fr: kfree(subsys->firmware_rev); free_mn: @@ -1616,8 +1902,12 @@ static void nvmet_subsys_free(struct kref *ref) struct nvmet_subsys *subsys = container_of(ref, struct nvmet_subsys, ref); + WARN_ON_ONCE(!list_empty(&subsys->ctrls)); + WARN_ON_ONCE(!list_empty(&subsys->hosts)); WARN_ON_ONCE(!xa_empty(&subsys->namespaces)); + nvmet_debugfs_subsys_free(subsys); + xa_destroy(&subsys->namespaces); nvmet_passthru_subsys_free(subsys); @@ -1663,21 +1953,29 @@ static int __init nvmet_init(void) if (!buffered_io_wq) goto out_free_zbd_work_queue; - nvmet_wq = alloc_workqueue("nvmet-wq", WQ_MEM_RECLAIM, 0); + nvmet_wq = alloc_workqueue("nvmet-wq", + WQ_MEM_RECLAIM | WQ_UNBOUND | WQ_SYSFS, 0); if (!nvmet_wq) goto out_free_buffered_work_queue; - error = nvmet_init_discovery(); + error = nvmet_init_debugfs(); if (error) goto out_free_nvmet_work_queue; + error = nvmet_init_discovery(); + if (error) + goto out_exit_debugfs; + error = nvmet_init_configfs(); if (error) goto out_exit_discovery; + return 0; out_exit_discovery: nvmet_exit_discovery(); +out_exit_debugfs: + nvmet_exit_debugfs(); out_free_nvmet_work_queue: destroy_workqueue(nvmet_wq); out_free_buffered_work_queue: @@ -1693,6 +1991,7 @@ static void __exit nvmet_exit(void) { nvmet_exit_configfs(); nvmet_exit_discovery(); + nvmet_exit_debugfs(); ida_destroy(&cntlid_ida); destroy_workqueue(nvmet_wq); destroy_workqueue(buffered_io_wq); @@ -1706,4 +2005,5 @@ static void __exit nvmet_exit(void) module_init(nvmet_init); module_exit(nvmet_exit); +MODULE_DESCRIPTION("NVMe target core framework"); MODULE_LICENSE("GPL v2"); diff --git a/drivers/nvme/target/debugfs.c b/drivers/nvme/target/debugfs.c new file mode 100644 index 000000000000..5dcbd5aa86e1 --- /dev/null +++ b/drivers/nvme/target/debugfs.c @@ -0,0 +1,229 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * DebugFS interface for the NVMe target. + * Copyright (c) 2022-2024 Shadow + * Copyright (c) 2024 SUSE LLC + */ + +#include <linux/debugfs.h> +#include <linux/fs.h> +#include <linux/init.h> +#include <linux/kernel.h> + +#include "nvmet.h" +#include "debugfs.h" + +static struct dentry *nvmet_debugfs; + +#define NVMET_DEBUGFS_ATTR(field) \ + static int field##_open(struct inode *inode, struct file *file) \ + { return single_open(file, field##_show, inode->i_private); } \ + \ + static const struct file_operations field##_fops = { \ + .open = field##_open, \ + .read = seq_read, \ + .release = single_release, \ + } + +#define NVMET_DEBUGFS_RW_ATTR(field) \ + static int field##_open(struct inode *inode, struct file *file) \ + { return single_open(file, field##_show, inode->i_private); } \ + \ + static const struct file_operations field##_fops = { \ + .open = field##_open, \ + .read = seq_read, \ + .write = field##_write, \ + .release = single_release, \ + } + +static int nvmet_ctrl_hostnqn_show(struct seq_file *m, void *p) +{ + struct nvmet_ctrl *ctrl = m->private; + + seq_puts(m, ctrl->hostnqn); + return 0; +} +NVMET_DEBUGFS_ATTR(nvmet_ctrl_hostnqn); + +static int nvmet_ctrl_kato_show(struct seq_file *m, void *p) +{ + struct nvmet_ctrl *ctrl = m->private; + + seq_printf(m, "%d\n", ctrl->kato); + return 0; +} +NVMET_DEBUGFS_ATTR(nvmet_ctrl_kato); + +static int nvmet_ctrl_port_show(struct seq_file *m, void *p) +{ + struct nvmet_ctrl *ctrl = m->private; + + seq_printf(m, "%d\n", le16_to_cpu(ctrl->port->disc_addr.portid)); + return 0; +} +NVMET_DEBUGFS_ATTR(nvmet_ctrl_port); + +static const char *const csts_state_names[] = { + [NVME_CSTS_RDY] = "ready", + [NVME_CSTS_CFS] = "fatal", + [NVME_CSTS_NSSRO] = "reset", + [NVME_CSTS_SHST_OCCUR] = "shutdown", + [NVME_CSTS_SHST_CMPLT] = "completed", + [NVME_CSTS_PP] = "paused", +}; + +static int nvmet_ctrl_state_show(struct seq_file *m, void *p) +{ + struct nvmet_ctrl *ctrl = m->private; + bool sep = false; + int i; + + for (i = 0; i < ARRAY_SIZE(csts_state_names); i++) { + int state = BIT(i); + + if (!(ctrl->csts & state)) + continue; + if (sep) + seq_puts(m, "|"); + sep = true; + if (csts_state_names[state]) + seq_puts(m, csts_state_names[state]); + else + seq_printf(m, "%d", state); + } + if (sep) + seq_printf(m, "\n"); + return 0; +} + +static ssize_t nvmet_ctrl_state_write(struct file *file, const char __user *buf, + size_t count, loff_t *ppos) +{ + struct seq_file *m = file->private_data; + struct nvmet_ctrl *ctrl = m->private; + char reset[16]; + + if (count >= sizeof(reset)) + return -EINVAL; + if (copy_from_user(reset, buf, count)) + return -EFAULT; + if (!memcmp(reset, "fatal", 5)) + nvmet_ctrl_fatal_error(ctrl); + else + return -EINVAL; + return count; +} +NVMET_DEBUGFS_RW_ATTR(nvmet_ctrl_state); + +static int nvmet_ctrl_host_traddr_show(struct seq_file *m, void *p) +{ + struct nvmet_ctrl *ctrl = m->private; + ssize_t size; + char buf[NVMF_TRADDR_SIZE + 1]; + + size = nvmet_ctrl_host_traddr(ctrl, buf, NVMF_TRADDR_SIZE); + if (size < 0) { + buf[0] = '\0'; + size = 0; + } + buf[size] = '\0'; + seq_printf(m, "%s\n", buf); + return 0; +} +NVMET_DEBUGFS_ATTR(nvmet_ctrl_host_traddr); + +#ifdef CONFIG_NVME_TARGET_TCP_TLS +static int nvmet_ctrl_tls_key_show(struct seq_file *m, void *p) +{ + struct nvmet_ctrl *ctrl = m->private; + key_serial_t keyid = nvmet_queue_tls_keyid(ctrl->sqs[0]); + + seq_printf(m, "%08x\n", keyid); + return 0; +} +NVMET_DEBUGFS_ATTR(nvmet_ctrl_tls_key); + +static int nvmet_ctrl_tls_concat_show(struct seq_file *m, void *p) +{ + struct nvmet_ctrl *ctrl = m->private; + + seq_printf(m, "%d\n", ctrl->concat); + return 0; +} +NVMET_DEBUGFS_ATTR(nvmet_ctrl_tls_concat); +#endif + +int nvmet_debugfs_ctrl_setup(struct nvmet_ctrl *ctrl) +{ + char name[32]; + struct dentry *parent = ctrl->subsys->debugfs_dir; + int ret; + + if (!parent) + return -ENODEV; + snprintf(name, sizeof(name), "ctrl%d", ctrl->cntlid); + ctrl->debugfs_dir = debugfs_create_dir(name, parent); + if (IS_ERR(ctrl->debugfs_dir)) { + ret = PTR_ERR(ctrl->debugfs_dir); + ctrl->debugfs_dir = NULL; + return ret; + } + debugfs_create_file("port", S_IRUSR, ctrl->debugfs_dir, ctrl, + &nvmet_ctrl_port_fops); + debugfs_create_file("hostnqn", S_IRUSR, ctrl->debugfs_dir, ctrl, + &nvmet_ctrl_hostnqn_fops); + debugfs_create_file("kato", S_IRUSR, ctrl->debugfs_dir, ctrl, + &nvmet_ctrl_kato_fops); + debugfs_create_file("state", S_IRUSR | S_IWUSR, ctrl->debugfs_dir, ctrl, + &nvmet_ctrl_state_fops); + debugfs_create_file("host_traddr", S_IRUSR, ctrl->debugfs_dir, ctrl, + &nvmet_ctrl_host_traddr_fops); +#ifdef CONFIG_NVME_TARGET_TCP_TLS + debugfs_create_file("tls_concat", S_IRUSR, ctrl->debugfs_dir, ctrl, + &nvmet_ctrl_tls_concat_fops); + debugfs_create_file("tls_key", S_IRUSR, ctrl->debugfs_dir, ctrl, + &nvmet_ctrl_tls_key_fops); +#endif + return 0; +} + +void nvmet_debugfs_ctrl_free(struct nvmet_ctrl *ctrl) +{ + debugfs_remove_recursive(ctrl->debugfs_dir); +} + +int nvmet_debugfs_subsys_setup(struct nvmet_subsys *subsys) +{ + int ret = 0; + + subsys->debugfs_dir = debugfs_create_dir(subsys->subsysnqn, + nvmet_debugfs); + if (IS_ERR(subsys->debugfs_dir)) { + ret = PTR_ERR(subsys->debugfs_dir); + subsys->debugfs_dir = NULL; + } + return ret; +} + +void nvmet_debugfs_subsys_free(struct nvmet_subsys *subsys) +{ + debugfs_remove_recursive(subsys->debugfs_dir); +} + +int __init nvmet_init_debugfs(void) +{ + struct dentry *parent; + + parent = debugfs_create_dir("nvmet", NULL); + if (IS_ERR(parent)) { + pr_warn("%s: failed to create debugfs directory\n", "nvmet"); + return PTR_ERR(parent); + } + nvmet_debugfs = parent; + return 0; +} + +void nvmet_exit_debugfs(void) +{ + debugfs_remove_recursive(nvmet_debugfs); +} diff --git a/drivers/nvme/target/debugfs.h b/drivers/nvme/target/debugfs.h new file mode 100644 index 000000000000..cfb8bbf6a297 --- /dev/null +++ b/drivers/nvme/target/debugfs.h @@ -0,0 +1,42 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * DebugFS interface for the NVMe target. + * Copyright (c) 2022-2024 Shadow + * Copyright (c) 2024 SUSE LLC + */ +#ifndef NVMET_DEBUGFS_H +#define NVMET_DEBUGFS_H + +#include <linux/types.h> + +#ifdef CONFIG_NVME_TARGET_DEBUGFS +int nvmet_debugfs_subsys_setup(struct nvmet_subsys *subsys); +void nvmet_debugfs_subsys_free(struct nvmet_subsys *subsys); +int nvmet_debugfs_ctrl_setup(struct nvmet_ctrl *ctrl); +void nvmet_debugfs_ctrl_free(struct nvmet_ctrl *ctrl); + +int __init nvmet_init_debugfs(void); +void nvmet_exit_debugfs(void); +#else +static inline int nvmet_debugfs_subsys_setup(struct nvmet_subsys *subsys) +{ + return 0; +} +static inline void nvmet_debugfs_subsys_free(struct nvmet_subsys *subsys){} + +static inline int nvmet_debugfs_ctrl_setup(struct nvmet_ctrl *ctrl) +{ + return 0; +} +static inline void nvmet_debugfs_ctrl_free(struct nvmet_ctrl *ctrl) {} + +static inline int __init nvmet_init_debugfs(void) +{ + return 0; +} + +static inline void nvmet_exit_debugfs(void) {} + +#endif + +#endif /* NVMET_DEBUGFS_H */ diff --git a/drivers/nvme/target/discovery.c b/drivers/nvme/target/discovery.c index 668d257fa986..c06f3e04296c 100644 --- a/drivers/nvme/target/discovery.c +++ b/drivers/nvme/target/discovery.c @@ -21,7 +21,7 @@ static void __nvmet_disc_changed(struct nvmet_port *port, if (nvmet_aen_bit_disabled(ctrl, NVME_AEN_BIT_DISC_CHANGE)) return; - nvmet_add_async_event(ctrl, NVME_AER_TYPE_NOTICE, + nvmet_add_async_event(ctrl, NVME_AER_NOTICE, NVME_AER_NOTICE_DISC_CHANGED, NVME_LOG_DISC); } @@ -119,7 +119,7 @@ static void nvmet_format_discovery_entry(struct nvmf_disc_rsp_page_hdr *hdr, memcpy(e->trsvcid, port->disc_addr.trsvcid, NVMF_TRSVCID_SIZE); memcpy(e->traddr, traddr, NVMF_TRADDR_SIZE); memcpy(e->tsas.common, port->disc_addr.tsas.common, NVMF_TSAS_SIZE); - strncpy(e->subnqn, subsys_nqn, NVMF_NQN_SIZE); + strscpy(e->subnqn, subsys_nqn, NVMF_NQN_SIZE); } /* @@ -179,7 +179,7 @@ static void nvmet_execute_disc_get_log_page(struct nvmet_req *req) if (req->cmd->get_log_page.lid != NVME_LOG_DISC) { req->error_loc = offsetof(struct nvme_get_log_page_command, lid); - status = NVME_SC_INVALID_FIELD | NVME_SC_DNR; + status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; goto out; } @@ -187,7 +187,7 @@ static void nvmet_execute_disc_get_log_page(struct nvmet_req *req) if (offset & 0x3) { req->error_loc = offsetof(struct nvme_get_log_page_command, lpo); - status = NVME_SC_INVALID_FIELD | NVME_SC_DNR; + status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; goto out; } @@ -224,6 +224,9 @@ static void nvmet_execute_disc_get_log_page(struct nvmet_req *req) } list_for_each_entry(r, &req->port->referrals, entry) { + if (r->disc_addr.trtype == NVMF_TRTYPE_PCI) + continue; + nvmet_format_discovery_entry(hdr, r, NVME_DISC_SUBSYS_NAME, r->disc_addr.traddr, @@ -256,7 +259,7 @@ static void nvmet_execute_disc_identify(struct nvmet_req *req) if (req->cmd->identify.cns != NVME_ID_CNS_CTRL) { req->error_loc = offsetof(struct nvme_identify, cns); - status = NVME_SC_INVALID_FIELD | NVME_SC_DNR; + status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; goto out; } @@ -282,7 +285,7 @@ static void nvmet_execute_disc_identify(struct nvmet_req *req) id->lpa = (1 << 2); /* no enforcement soft-limit for maxcmd - pick arbitrary high value */ - id->maxcmd = cpu_to_le16(NVMET_MAX_CMD); + id->maxcmd = cpu_to_le16(NVMET_MAX_CMD(ctrl)); id->sgls = cpu_to_le32(1 << 0); /* we always support SGLs */ if (ctrl->ops->flags & NVMF_KEYED_SGLS) @@ -320,7 +323,7 @@ static void nvmet_execute_disc_set_features(struct nvmet_req *req) default: req->error_loc = offsetof(struct nvme_common_command, cdw10); - stat = NVME_SC_INVALID_FIELD | NVME_SC_DNR; + stat = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; break; } @@ -345,13 +348,27 @@ static void nvmet_execute_disc_get_features(struct nvmet_req *req) default: req->error_loc = offsetof(struct nvme_common_command, cdw10); - stat = NVME_SC_INVALID_FIELD | NVME_SC_DNR; + stat = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; break; } nvmet_req_complete(req, stat); } +u32 nvmet_discovery_cmd_data_len(struct nvmet_req *req) +{ + struct nvme_command *cmd = req->cmd; + + switch (cmd->common.opcode) { + case nvme_admin_get_log_page: + return nvmet_get_log_page_len(req->cmd); + case nvme_admin_identify: + return NVME_IDENTIFY_DATA_SIZE; + default: + return 0; + } +} + u16 nvmet_parse_discovery_cmd(struct nvmet_req *req) { struct nvme_command *cmd = req->cmd; @@ -361,7 +378,7 @@ u16 nvmet_parse_discovery_cmd(struct nvmet_req *req) cmd->common.opcode); req->error_loc = offsetof(struct nvme_common_command, opcode); - return NVME_SC_INVALID_OPCODE | NVME_SC_DNR; + return NVME_SC_INVALID_OPCODE | NVME_STATUS_DNR; } switch (cmd->common.opcode) { @@ -386,7 +403,7 @@ u16 nvmet_parse_discovery_cmd(struct nvmet_req *req) default: pr_debug("unhandled cmd %d\n", cmd->common.opcode); req->error_loc = offsetof(struct nvme_common_command, opcode); - return NVME_SC_INVALID_OPCODE | NVME_SC_DNR; + return NVME_SC_INVALID_OPCODE | NVME_STATUS_DNR; } } diff --git a/drivers/nvme/target/fabrics-cmd-auth.c b/drivers/nvme/target/fabrics-cmd-auth.c index 7970a7640e58..5946681cb0e3 100644 --- a/drivers/nvme/target/fabrics-cmd-auth.c +++ b/drivers/nvme/target/fabrics-cmd-auth.c @@ -31,7 +31,7 @@ void nvmet_auth_sq_init(struct nvmet_sq *sq) sq->dhchap_step = NVME_AUTH_DHCHAP_MESSAGE_NEGOTIATE; } -static u16 nvmet_auth_negotiate(struct nvmet_req *req, void *d) +static u8 nvmet_auth_negotiate(struct nvmet_req *req, void *d) { struct nvmet_ctrl *ctrl = req->sq->ctrl; struct nvmf_auth_dhchap_negotiate_data *data = d; @@ -43,8 +43,27 @@ static u16 nvmet_auth_negotiate(struct nvmet_req *req, void *d) data->auth_protocol[0].dhchap.halen, data->auth_protocol[0].dhchap.dhlen); req->sq->dhchap_tid = le16_to_cpu(data->t_id); - if (data->sc_c) - return NVME_AUTH_DHCHAP_FAILURE_CONCAT_MISMATCH; + req->sq->sc_c = data->sc_c; + if (data->sc_c != NVME_AUTH_SECP_NOSC) { + if (!IS_ENABLED(CONFIG_NVME_TARGET_TCP_TLS)) + return NVME_AUTH_DHCHAP_FAILURE_CONCAT_MISMATCH; + /* Secure concatenation can only be enabled on the admin queue */ + if (req->sq->qid) + return NVME_AUTH_DHCHAP_FAILURE_CONCAT_MISMATCH; + switch (data->sc_c) { + case NVME_AUTH_SECP_NEWTLSPSK: + if (nvmet_queue_tls_keyid(req->sq)) + return NVME_AUTH_DHCHAP_FAILURE_CONCAT_MISMATCH; + break; + case NVME_AUTH_SECP_REPLACETLSPSK: + if (!nvmet_queue_tls_keyid(req->sq)) + return NVME_AUTH_DHCHAP_FAILURE_CONCAT_MISMATCH; + break; + default: + return NVME_AUTH_DHCHAP_FAILURE_CONCAT_MISMATCH; + } + ctrl->concat = true; + } if (data->napd != 1) return NVME_AUTH_DHCHAP_FAILURE_HASH_UNUSABLE; @@ -103,13 +122,19 @@ static u16 nvmet_auth_negotiate(struct nvmet_req *req, void *d) nvme_auth_dhgroup_name(fallback_dhgid)); ctrl->dh_gid = fallback_dhgid; } + if (ctrl->dh_gid == NVME_AUTH_DHGROUP_NULL && ctrl->concat) { + pr_debug("%s: ctrl %d qid %d: NULL DH group invalid " + "for secure channel concatenation\n", __func__, + ctrl->cntlid, req->sq->qid); + return NVME_AUTH_DHCHAP_FAILURE_CONCAT_MISMATCH; + } pr_debug("%s: ctrl %d qid %d: selected DH group %s (%d)\n", __func__, ctrl->cntlid, req->sq->qid, nvme_auth_dhgroup_name(ctrl->dh_gid), ctrl->dh_gid); return 0; } -static u16 nvmet_auth_reply(struct nvmet_req *req, void *d) +static u8 nvmet_auth_reply(struct nvmet_req *req, void *d) { struct nvmet_ctrl *ctrl = req->sq->ctrl; struct nvmf_auth_dhchap_reply_data *data = d; @@ -148,12 +173,22 @@ static u16 nvmet_auth_reply(struct nvmet_req *req, void *d) if (memcmp(data->rval, response, data->hl)) { pr_info("ctrl %d qid %d host response mismatch\n", ctrl->cntlid, req->sq->qid); + pr_debug("ctrl %d qid %d rval %*ph\n", + ctrl->cntlid, req->sq->qid, data->hl, data->rval); + pr_debug("ctrl %d qid %d response %*ph\n", + ctrl->cntlid, req->sq->qid, data->hl, response); kfree(response); return NVME_AUTH_DHCHAP_FAILURE_FAILED; } kfree(response); pr_debug("%s: ctrl %d qid %d host authenticated\n", __func__, ctrl->cntlid, req->sq->qid); + if (!data->cvalid && ctrl->concat) { + pr_debug("%s: ctrl %d qid %d invalid challenge\n", + __func__, ctrl->cntlid, req->sq->qid); + return NVME_AUTH_DHCHAP_FAILURE_FAILED; + } + req->sq->dhchap_s2 = le32_to_cpu(data->seqnum); if (data->cvalid) { req->sq->dhchap_c2 = kmemdup(data->rval + data->hl, data->hl, GFP_KERNEL); @@ -163,22 +198,39 @@ static u16 nvmet_auth_reply(struct nvmet_req *req, void *d) pr_debug("%s: ctrl %d qid %d challenge %*ph\n", __func__, ctrl->cntlid, req->sq->qid, data->hl, req->sq->dhchap_c2); - req->sq->dhchap_s2 = le32_to_cpu(data->seqnum); - } else { + } + /* + * NVMe Base Spec 2.2 section 8.3.4.5.4: DH-HMAC-CHAP_Reply message + * Sequence Number (SEQNUM): [ .. ] + * The value 0h is used to indicate that bidirectional authentication + * is not performed, but a challenge value C2 is carried in order to + * generate a pre-shared key (PSK) for subsequent establishment of a + * secure channel. + */ + if (req->sq->dhchap_s2 == 0) { + if (ctrl->concat) + nvmet_auth_insert_psk(req->sq); req->sq->authenticated = true; + kfree(req->sq->dhchap_c2); req->sq->dhchap_c2 = NULL; - } + } else if (!data->cvalid) + req->sq->authenticated = true; return 0; } -static u16 nvmet_auth_failure2(void *d) +static u8 nvmet_auth_failure2(void *d) { struct nvmf_auth_dhchap_failure_data *data = d; return data->rescode_exp; } +u32 nvmet_auth_send_data_len(struct nvmet_req *req) +{ + return le32_to_cpu(req->cmd->auth_send.tl); +} + void nvmet_execute_auth_send(struct nvmet_req *req) { struct nvmet_ctrl *ctrl = req->sq->ctrl; @@ -186,28 +238,29 @@ void nvmet_execute_auth_send(struct nvmet_req *req) void *d; u32 tl; u16 status = 0; + u8 dhchap_status; if (req->cmd->auth_send.secp != NVME_AUTH_DHCHAP_PROTOCOL_IDENTIFIER) { - status = NVME_SC_INVALID_FIELD | NVME_SC_DNR; + status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; req->error_loc = offsetof(struct nvmf_auth_send_command, secp); goto done; } if (req->cmd->auth_send.spsp0 != 0x01) { - status = NVME_SC_INVALID_FIELD | NVME_SC_DNR; + status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; req->error_loc = offsetof(struct nvmf_auth_send_command, spsp0); goto done; } if (req->cmd->auth_send.spsp1 != 0x01) { - status = NVME_SC_INVALID_FIELD | NVME_SC_DNR; + status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; req->error_loc = offsetof(struct nvmf_auth_send_command, spsp1); goto done; } - tl = le32_to_cpu(req->cmd->auth_send.tl); + tl = nvmet_auth_send_data_len(req); if (!tl) { - status = NVME_SC_INVALID_FIELD | NVME_SC_DNR; + status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; req->error_loc = offsetof(struct nvmf_auth_send_command, tl); goto done; @@ -237,30 +290,32 @@ void nvmet_execute_auth_send(struct nvmet_req *req) if (data->auth_type == NVME_AUTH_COMMON_MESSAGES) { if (data->auth_id == NVME_AUTH_DHCHAP_MESSAGE_NEGOTIATE) { /* Restart negotiation */ - pr_debug("%s: ctrl %d qid %d reset negotiation\n", __func__, - ctrl->cntlid, req->sq->qid); + pr_debug("%s: ctrl %d qid %d reset negotiation\n", + __func__, ctrl->cntlid, req->sq->qid); if (!req->sq->qid) { - if (nvmet_setup_auth(ctrl) < 0) { - status = NVME_SC_INTERNAL; - pr_err("ctrl %d qid 0 failed to setup" - "re-authentication", + dhchap_status = nvmet_setup_auth(ctrl, req->sq); + if (dhchap_status) { + pr_err("ctrl %d qid 0 failed to setup re-authentication\n", ctrl->cntlid); - goto done_failure1; + req->sq->dhchap_status = dhchap_status; + req->sq->dhchap_step = + NVME_AUTH_DHCHAP_MESSAGE_FAILURE1; + goto done_kfree; } } - req->sq->dhchap_step = NVME_AUTH_DHCHAP_MESSAGE_NEGOTIATE; + req->sq->dhchap_step = + NVME_AUTH_DHCHAP_MESSAGE_NEGOTIATE; } else if (data->auth_id != req->sq->dhchap_step) goto done_failure1; /* Validate negotiation parameters */ - status = nvmet_auth_negotiate(req, d); - if (status == 0) + dhchap_status = nvmet_auth_negotiate(req, d); + if (dhchap_status == 0) req->sq->dhchap_step = NVME_AUTH_DHCHAP_MESSAGE_CHALLENGE; else { req->sq->dhchap_step = NVME_AUTH_DHCHAP_MESSAGE_FAILURE1; - req->sq->dhchap_status = status; - status = 0; + req->sq->dhchap_status = dhchap_status; } goto done_kfree; } @@ -284,35 +339,32 @@ void nvmet_execute_auth_send(struct nvmet_req *req) switch (data->auth_id) { case NVME_AUTH_DHCHAP_MESSAGE_REPLY: - status = nvmet_auth_reply(req, d); - if (status == 0) + dhchap_status = nvmet_auth_reply(req, d); + if (dhchap_status == 0) req->sq->dhchap_step = NVME_AUTH_DHCHAP_MESSAGE_SUCCESS1; else { req->sq->dhchap_step = NVME_AUTH_DHCHAP_MESSAGE_FAILURE1; - req->sq->dhchap_status = status; - status = 0; + req->sq->dhchap_status = dhchap_status; } goto done_kfree; - break; case NVME_AUTH_DHCHAP_MESSAGE_SUCCESS2: + if (ctrl->concat) + nvmet_auth_insert_psk(req->sq); req->sq->authenticated = true; pr_debug("%s: ctrl %d qid %d ctrl authenticated\n", __func__, ctrl->cntlid, req->sq->qid); goto done_kfree; - break; case NVME_AUTH_DHCHAP_MESSAGE_FAILURE2: - status = nvmet_auth_failure2(d); - if (status) { + dhchap_status = nvmet_auth_failure2(d); + if (dhchap_status) { pr_warn("ctrl %d qid %d: authentication failed (%d)\n", - ctrl->cntlid, req->sq->qid, status); - req->sq->dhchap_status = status; + ctrl->cntlid, req->sq->qid, dhchap_status); + req->sq->dhchap_status = dhchap_status; req->sq->authenticated = false; - status = 0; } goto done_kfree; - break; default: req->sq->dhchap_status = NVME_AUTH_DHCHAP_FAILURE_INCORRECT_MESSAGE; @@ -320,7 +372,6 @@ void nvmet_execute_auth_send(struct nvmet_req *req) NVME_AUTH_DHCHAP_MESSAGE_FAILURE2; req->sq->authenticated = false; goto done_kfree; - break; } done_failure1: req->sq->dhchap_status = NVME_AUTH_DHCHAP_FAILURE_INCORRECT_MESSAGE; @@ -336,20 +387,21 @@ done: pr_debug("%s: ctrl %d qid %d nvme status %x error loc %d\n", __func__, ctrl->cntlid, req->sq->qid, status, req->error_loc); - req->cqe->result.u64 = 0; - nvmet_req_complete(req, status); if (req->sq->dhchap_step != NVME_AUTH_DHCHAP_MESSAGE_SUCCESS2 && req->sq->dhchap_step != NVME_AUTH_DHCHAP_MESSAGE_FAILURE2) { unsigned long auth_expire_secs = ctrl->kato ? ctrl->kato : 120; mod_delayed_work(system_wq, &req->sq->auth_expired_work, auth_expire_secs * HZ); - return; + goto complete; } /* Final states, clear up variables */ nvmet_auth_sq_free(req->sq); if (req->sq->dhchap_step == NVME_AUTH_DHCHAP_MESSAGE_FAILURE2) nvmet_ctrl_fatal_error(ctrl); + +complete: + nvmet_req_complete(req, status); } static int nvmet_auth_challenge(struct nvmet_req *req, void *d, int al) @@ -431,6 +483,11 @@ static void nvmet_auth_failure1(struct nvmet_req *req, void *d, int al) data->rescode_exp = req->sq->dhchap_status; } +u32 nvmet_auth_receive_data_len(struct nvmet_req *req) +{ + return le32_to_cpu(req->cmd->auth_receive.al); +} + void nvmet_execute_auth_receive(struct nvmet_req *req) { struct nvmet_ctrl *ctrl = req->sq->ctrl; @@ -439,26 +496,26 @@ void nvmet_execute_auth_receive(struct nvmet_req *req) u16 status = 0; if (req->cmd->auth_receive.secp != NVME_AUTH_DHCHAP_PROTOCOL_IDENTIFIER) { - status = NVME_SC_INVALID_FIELD | NVME_SC_DNR; + status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; req->error_loc = offsetof(struct nvmf_auth_receive_command, secp); goto done; } if (req->cmd->auth_receive.spsp0 != 0x01) { - status = NVME_SC_INVALID_FIELD | NVME_SC_DNR; + status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; req->error_loc = offsetof(struct nvmf_auth_receive_command, spsp0); goto done; } if (req->cmd->auth_receive.spsp1 != 0x01) { - status = NVME_SC_INVALID_FIELD | NVME_SC_DNR; + status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; req->error_loc = offsetof(struct nvmf_auth_receive_command, spsp1); goto done; } - al = le32_to_cpu(req->cmd->auth_receive.al); + al = nvmet_auth_receive_data_len(req); if (!al) { - status = NVME_SC_INVALID_FIELD | NVME_SC_DNR; + status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; req->error_loc = offsetof(struct nvmf_auth_receive_command, al); goto done; @@ -483,15 +540,6 @@ void nvmet_execute_auth_receive(struct nvmet_req *req) status = NVME_SC_INTERNAL; break; } - if (status) { - req->sq->dhchap_status = status; - nvmet_auth_failure1(req, d, al); - pr_warn("ctrl %d qid %d: challenge status (%x)\n", - ctrl->cntlid, req->sq->qid, - req->sq->dhchap_status); - status = 0; - break; - } req->sq->dhchap_step = NVME_AUTH_DHCHAP_MESSAGE_REPLY; break; case NVME_AUTH_DHCHAP_MESSAGE_SUCCESS1: @@ -526,12 +574,11 @@ void nvmet_execute_auth_receive(struct nvmet_req *req) status = nvmet_copy_to_sgl(req, 0, d, al); kfree(d); done: - req->cqe->result.u64 = 0; - nvmet_req_complete(req, status); if (req->sq->dhchap_step == NVME_AUTH_DHCHAP_MESSAGE_SUCCESS2) nvmet_auth_sq_free(req->sq); else if (req->sq->dhchap_step == NVME_AUTH_DHCHAP_MESSAGE_FAILURE1) { nvmet_auth_sq_free(req->sq); nvmet_ctrl_fatal_error(ctrl); } + nvmet_req_complete(req, status); } diff --git a/drivers/nvme/target/fabrics-cmd.c b/drivers/nvme/target/fabrics-cmd.c index 43b5bd8bb6a5..7b8d8b397802 100644 --- a/drivers/nvme/target/fabrics-cmd.c +++ b/drivers/nvme/target/fabrics-cmd.c @@ -18,7 +18,7 @@ static void nvmet_execute_prop_set(struct nvmet_req *req) if (req->cmd->prop_set.attrib & 1) { req->error_loc = offsetof(struct nvmf_property_set_command, attrib); - status = NVME_SC_INVALID_FIELD | NVME_SC_DNR; + status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; goto out; } @@ -29,7 +29,7 @@ static void nvmet_execute_prop_set(struct nvmet_req *req) default: req->error_loc = offsetof(struct nvmf_property_set_command, offset); - status = NVME_SC_INVALID_FIELD | NVME_SC_DNR; + status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; } out: nvmet_req_complete(req, status); @@ -50,7 +50,7 @@ static void nvmet_execute_prop_get(struct nvmet_req *req) val = ctrl->cap; break; default: - status = NVME_SC_INVALID_FIELD | NVME_SC_DNR; + status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; break; } } else { @@ -64,8 +64,11 @@ static void nvmet_execute_prop_get(struct nvmet_req *req) case NVME_REG_CSTS: val = ctrl->csts; break; + case NVME_REG_CRTO: + val = NVME_CAP_TIMEOUT(ctrl->csts); + break; default: - status = NVME_SC_INVALID_FIELD | NVME_SC_DNR; + status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; break; } } @@ -82,6 +85,22 @@ static void nvmet_execute_prop_get(struct nvmet_req *req) nvmet_req_complete(req, status); } +u32 nvmet_fabrics_admin_cmd_data_len(struct nvmet_req *req) +{ + struct nvme_command *cmd = req->cmd; + + switch (cmd->fabrics.fctype) { +#ifdef CONFIG_NVME_TARGET_AUTH + case nvme_fabrics_type_auth_send: + return nvmet_auth_send_data_len(req); + case nvme_fabrics_type_auth_receive: + return nvmet_auth_receive_data_len(req); +#endif + default: + return 0; + } +} + u16 nvmet_parse_fabrics_admin_cmd(struct nvmet_req *req) { struct nvme_command *cmd = req->cmd; @@ -105,12 +124,28 @@ u16 nvmet_parse_fabrics_admin_cmd(struct nvmet_req *req) pr_debug("received unknown capsule type 0x%x\n", cmd->fabrics.fctype); req->error_loc = offsetof(struct nvmf_common_command, fctype); - return NVME_SC_INVALID_OPCODE | NVME_SC_DNR; + return NVME_SC_INVALID_OPCODE | NVME_STATUS_DNR; } return 0; } +u32 nvmet_fabrics_io_cmd_data_len(struct nvmet_req *req) +{ + struct nvme_command *cmd = req->cmd; + + switch (cmd->fabrics.fctype) { +#ifdef CONFIG_NVME_TARGET_AUTH + case nvme_fabrics_type_auth_send: + return nvmet_auth_send_data_len(req); + case nvme_fabrics_type_auth_receive: + return nvmet_auth_receive_data_len(req); +#endif + default: + return 0; + } +} + u16 nvmet_parse_fabrics_io_cmd(struct nvmet_req *req) { struct nvme_command *cmd = req->cmd; @@ -128,7 +163,7 @@ u16 nvmet_parse_fabrics_io_cmd(struct nvmet_req *req) pr_debug("received unknown capsule type 0x%x\n", cmd->fabrics.fctype); req->error_loc = offsetof(struct nvmf_common_command, fctype); - return NVME_SC_INVALID_OPCODE | NVME_SC_DNR; + return NVME_SC_INVALID_OPCODE | NVME_STATUS_DNR; } return 0; @@ -147,29 +182,38 @@ static u16 nvmet_install_queue(struct nvmet_ctrl *ctrl, struct nvmet_req *req) pr_warn("queue size zero!\n"); req->error_loc = offsetof(struct nvmf_connect_command, sqsize); req->cqe->result.u32 = IPO_IATTR_CONNECT_SQE(sqsize); - ret = NVME_SC_CONNECT_INVALID_PARAM | NVME_SC_DNR; + ret = NVME_SC_CONNECT_INVALID_PARAM | NVME_STATUS_DNR; goto err; } if (ctrl->sqs[qid] != NULL) { pr_warn("qid %u has already been created\n", qid); req->error_loc = offsetof(struct nvmf_connect_command, qid); - return NVME_SC_CMD_SEQ_ERROR | NVME_SC_DNR; + return NVME_SC_CMD_SEQ_ERROR | NVME_STATUS_DNR; } - if (sqsize > mqes) { + /* for fabrics, this value applies to only the I/O Submission Queues */ + if (qid && sqsize > mqes) { pr_warn("sqsize %u is larger than MQES supported %u cntlid %d\n", sqsize, mqes, ctrl->cntlid); req->error_loc = offsetof(struct nvmf_connect_command, sqsize); req->cqe->result.u32 = IPO_IATTR_CONNECT_SQE(sqsize); - return NVME_SC_CONNECT_INVALID_PARAM | NVME_SC_DNR; + return NVME_SC_CONNECT_INVALID_PARAM | NVME_STATUS_DNR; } old = cmpxchg(&req->sq->ctrl, NULL, ctrl); if (old) { pr_warn("queue already connected!\n"); req->error_loc = offsetof(struct nvmf_connect_command, opcode); - return NVME_SC_CONNECT_CTRL_BUSY | NVME_SC_DNR; + return NVME_SC_CONNECT_CTRL_BUSY | NVME_STATUS_DNR; + } + + kref_get(&ctrl->ref); + old = cmpxchg(&req->cq->ctrl, NULL, ctrl); + if (old) { + pr_warn("queue already connected!\n"); + req->error_loc = offsetof(struct nvmf_connect_command, opcode); + return NVME_SC_CONNECT_CTRL_BUSY | NVME_STATUS_DNR; } /* note: convert queue size from 0's-based value to 1's-based value */ @@ -198,10 +242,26 @@ err: return ret; } -static u32 nvmet_connect_result(struct nvmet_ctrl *ctrl) +static u32 nvmet_connect_result(struct nvmet_ctrl *ctrl, struct nvmet_sq *sq) { + bool needs_auth = nvmet_has_auth(ctrl, sq); + key_serial_t keyid = nvmet_queue_tls_keyid(sq); + + /* Do not authenticate I/O queues */ + if (sq->qid) + needs_auth = false; + + if (keyid) + pr_debug("%s: ctrl %d qid %d should %sauthenticate, tls psk %08x\n", + __func__, ctrl->cntlid, sq->qid, + needs_auth ? "" : "not ", keyid); + else + pr_debug("%s: ctrl %d qid %d should %sauthenticate%s\n", + __func__, ctrl->cntlid, sq->qid, + needs_auth ? "" : "not ", + ctrl->concat ? ", secure concatenation" : ""); return (u32)ctrl->cntlid | - (nvmet_has_auth(ctrl) ? NVME_CONNECT_AUTHREQ_ATR : 0); + (needs_auth ? NVME_CONNECT_AUTHREQ_ATR : 0); } static void nvmet_execute_admin_connect(struct nvmet_req *req) @@ -209,77 +269,68 @@ static void nvmet_execute_admin_connect(struct nvmet_req *req) struct nvmf_connect_command *c = &req->cmd->connect; struct nvmf_connect_data *d; struct nvmet_ctrl *ctrl = NULL; - u16 status = 0; - int ret; + struct nvmet_alloc_ctrl_args args = { + .port = req->port, + .sq = req->sq, + .ops = req->ops, + .p2p_client = req->p2p_client, + .kato = le32_to_cpu(c->kato), + }; if (!nvmet_check_transfer_len(req, sizeof(struct nvmf_connect_data))) return; d = kmalloc(sizeof(*d), GFP_KERNEL); if (!d) { - status = NVME_SC_INTERNAL; + args.status = NVME_SC_INTERNAL; goto complete; } - status = nvmet_copy_from_sgl(req, 0, d, sizeof(*d)); - if (status) + args.status = nvmet_copy_from_sgl(req, 0, d, sizeof(*d)); + if (args.status) goto out; - /* zero out initial completion result, assign values as needed */ - req->cqe->result.u32 = 0; - if (c->recfmt != 0) { pr_warn("invalid connect version (%d).\n", le16_to_cpu(c->recfmt)); - req->error_loc = offsetof(struct nvmf_connect_command, recfmt); - status = NVME_SC_CONNECT_FORMAT | NVME_SC_DNR; + args.error_loc = offsetof(struct nvmf_connect_command, recfmt); + args.status = NVME_SC_CONNECT_FORMAT | NVME_STATUS_DNR; goto out; } if (unlikely(d->cntlid != cpu_to_le16(0xffff))) { pr_warn("connect attempt for invalid controller ID %#x\n", d->cntlid); - status = NVME_SC_CONNECT_INVALID_PARAM | NVME_SC_DNR; - req->cqe->result.u32 = IPO_IATTR_CONNECT_DATA(cntlid); + args.status = NVME_SC_CONNECT_INVALID_PARAM | NVME_STATUS_DNR; + args.result = IPO_IATTR_CONNECT_DATA(cntlid); goto out; } - status = nvmet_alloc_ctrl(d->subsysnqn, d->hostnqn, req, - le32_to_cpu(c->kato), &ctrl); - if (status) - goto out; - - ctrl->pi_support = ctrl->port->pi_enable && ctrl->subsys->pi_support; + d->subsysnqn[NVMF_NQN_FIELD_LEN - 1] = '\0'; + d->hostnqn[NVMF_NQN_FIELD_LEN - 1] = '\0'; - uuid_copy(&ctrl->hostid, &d->hostid); + args.subsysnqn = d->subsysnqn; + args.hostnqn = d->hostnqn; + args.hostid = &d->hostid; + args.kato = le32_to_cpu(c->kato); - ret = nvmet_setup_auth(ctrl); - if (ret < 0) { - pr_err("Failed to setup authentication, error %d\n", ret); - nvmet_ctrl_put(ctrl); - if (ret == -EPERM) - status = (NVME_SC_CONNECT_INVALID_HOST | NVME_SC_DNR); - else - status = NVME_SC_INTERNAL; + ctrl = nvmet_alloc_ctrl(&args); + if (!ctrl) goto out; - } - status = nvmet_install_queue(ctrl, req); - if (status) { + args.status = nvmet_install_queue(ctrl, req); + if (args.status) { nvmet_ctrl_put(ctrl); goto out; } - pr_info("creating %s controller %d for subsystem %s for NQN %s%s%s.\n", - nvmet_is_disc_subsys(ctrl->subsys) ? "discovery" : "nvm", - ctrl->cntlid, ctrl->subsys->subsysnqn, ctrl->hostnqn, - ctrl->pi_support ? " T10-PI is enabled" : "", - nvmet_has_auth(ctrl) ? " with DH-HMAC-CHAP" : ""); - req->cqe->result.u32 = cpu_to_le32(nvmet_connect_result(ctrl)); + args.result = cpu_to_le32(nvmet_connect_result(ctrl, req->sq)); out: kfree(d); complete: - nvmet_req_complete(req, status); + req->error_loc = args.error_loc; + req->cqe->result.u32 = args.result; + nvmet_req_complete(req, args.status); } static void nvmet_execute_io_connect(struct nvmet_req *req) @@ -288,7 +339,7 @@ static void nvmet_execute_io_connect(struct nvmet_req *req) struct nvmf_connect_data *d; struct nvmet_ctrl *ctrl; u16 qid = le16_to_cpu(c->qid); - u16 status = 0; + u16 status; if (!nvmet_check_transfer_len(req, sizeof(struct nvmf_connect_data))) return; @@ -303,26 +354,25 @@ static void nvmet_execute_io_connect(struct nvmet_req *req) if (status) goto out; - /* zero out initial completion result, assign values as needed */ - req->cqe->result.u32 = 0; - if (c->recfmt != 0) { pr_warn("invalid connect version (%d).\n", le16_to_cpu(c->recfmt)); - status = NVME_SC_CONNECT_FORMAT | NVME_SC_DNR; + status = NVME_SC_CONNECT_FORMAT | NVME_STATUS_DNR; goto out; } + d->subsysnqn[NVMF_NQN_FIELD_LEN - 1] = '\0'; + d->hostnqn[NVMF_NQN_FIELD_LEN - 1] = '\0'; ctrl = nvmet_ctrl_find_get(d->subsysnqn, d->hostnqn, le16_to_cpu(d->cntlid), req); if (!ctrl) { - status = NVME_SC_CONNECT_INVALID_PARAM | NVME_SC_DNR; + status = NVME_SC_CONNECT_INVALID_PARAM | NVME_STATUS_DNR; goto out; } if (unlikely(qid > ctrl->subsys->max_qid)) { pr_warn("invalid queue id (%d)\n", qid); - status = NVME_SC_CONNECT_INVALID_PARAM | NVME_SC_DNR; + status = NVME_SC_CONNECT_INVALID_PARAM | NVME_STATUS_DNR; req->cqe->result.u32 = IPO_IATTR_CONNECT_SQE(qid); goto out_ctrl_put; } @@ -332,7 +382,7 @@ static void nvmet_execute_io_connect(struct nvmet_req *req) goto out_ctrl_put; pr_debug("adding queue %d to ctrl %d.\n", qid, ctrl->cntlid); - req->cqe->result.u32 = cpu_to_le32(nvmet_connect_result(ctrl)); + req->cqe->result.u32 = cpu_to_le32(nvmet_connect_result(ctrl, req->sq)); out: kfree(d); complete: @@ -344,6 +394,17 @@ out_ctrl_put: goto out; } +u32 nvmet_connect_cmd_data_len(struct nvmet_req *req) +{ + struct nvme_command *cmd = req->cmd; + + if (!nvme_is_fabrics(cmd) || + cmd->fabrics.fctype != nvme_fabrics_type_connect) + return 0; + + return sizeof(struct nvmf_connect_data); +} + u16 nvmet_parse_connect_cmd(struct nvmet_req *req) { struct nvme_command *cmd = req->cmd; @@ -352,13 +413,13 @@ u16 nvmet_parse_connect_cmd(struct nvmet_req *req) pr_debug("invalid command 0x%x on unconnected queue.\n", cmd->fabrics.opcode); req->error_loc = offsetof(struct nvme_common_command, opcode); - return NVME_SC_INVALID_OPCODE | NVME_SC_DNR; + return NVME_SC_INVALID_OPCODE | NVME_STATUS_DNR; } if (cmd->fabrics.fctype != nvme_fabrics_type_connect) { pr_debug("invalid capsule type 0x%x on unconnected queue.\n", cmd->fabrics.fctype); req->error_loc = offsetof(struct nvmf_common_command, fctype); - return NVME_SC_INVALID_OPCODE | NVME_SC_DNR; + return NVME_SC_INVALID_OPCODE | NVME_STATUS_DNR; } if (cmd->connect.qid == 0) diff --git a/drivers/nvme/target/fc.c b/drivers/nvme/target/fc.c index ab2627e17bb9..0d9784004c9b 100644 --- a/drivers/nvme/target/fc.c +++ b/drivers/nvme/target/fc.c @@ -54,6 +54,8 @@ struct nvmet_fc_ls_req_op { /* for an LS RQST XMT */ int ls_error; struct list_head lsreq_list; /* tgtport->ls_req_list */ bool req_queued; + + struct work_struct put_work; }; @@ -145,8 +147,8 @@ struct nvmet_fc_tgt_queue { struct list_head avail_defer_list; struct workqueue_struct *work_q; struct kref ref; - struct rcu_head rcu; - struct nvmet_fc_fcp_iod fod[]; /* array of fcp_iods */ + /* array of fcp_iods */ + struct nvmet_fc_fcp_iod fod[] /* __counted_by(sqsize) */; } __aligned(sizeof(unsigned long long)); struct nvmet_fc_hostport { @@ -165,26 +167,11 @@ struct nvmet_fc_tgt_assoc { struct nvmet_fc_hostport *hostport; struct nvmet_fc_ls_iod *rcv_disconn; struct list_head a_list; - struct nvmet_fc_tgt_queue __rcu *queues[NVMET_NR_QUEUES + 1]; + struct nvmet_fc_tgt_queue *queues[NVMET_NR_QUEUES + 1]; struct kref ref; struct work_struct del_work; - struct rcu_head rcu; }; - -static inline int -nvmet_fc_iodnum(struct nvmet_fc_ls_iod *iodptr) -{ - return (iodptr - iodptr->tgtport->iod); -} - -static inline int -nvmet_fc_fodnum(struct nvmet_fc_fcp_iod *fodptr) -{ - return (fodptr - fodptr->queue->fod); -} - - /* * Association and Connection IDs: * @@ -248,6 +235,14 @@ static int nvmet_fc_tgt_a_get(struct nvmet_fc_tgt_assoc *assoc); static void nvmet_fc_tgt_q_put(struct nvmet_fc_tgt_queue *queue); static int nvmet_fc_tgt_q_get(struct nvmet_fc_tgt_queue *queue); static void nvmet_fc_tgtport_put(struct nvmet_fc_tgtport *tgtport); +static void nvmet_fc_put_lsop_work(struct work_struct *work) +{ + struct nvmet_fc_ls_req_op *lsop = + container_of(work, struct nvmet_fc_ls_req_op, put_work); + + nvmet_fc_tgtport_put(lsop->tgtport); + kfree(lsop); +} static int nvmet_fc_tgtport_get(struct nvmet_fc_tgtport *tgtport); static void nvmet_fc_handle_fcp_rqst(struct nvmet_fc_tgtport *tgtport, struct nvmet_fc_fcp_iod *fod); @@ -359,7 +354,7 @@ __nvmet_fc_finish_ls_req(struct nvmet_fc_ls_req_op *lsop) if (!lsop->req_queued) { spin_unlock_irqrestore(&tgtport->lock, flags); - return; + goto out_putwork; } list_del(&lsop->lsreq_list); @@ -372,7 +367,8 @@ __nvmet_fc_finish_ls_req(struct nvmet_fc_ls_req_op *lsop) (lsreq->rqstlen + lsreq->rsplen), DMA_BIDIRECTIONAL); - nvmet_fc_tgtport_put(tgtport); +out_putwork: + queue_work(nvmet_wq, &lsop->put_work); } static int @@ -393,6 +389,7 @@ __nvmet_fc_send_ls_req(struct nvmet_fc_tgtport *tgtport, lsreq->done = done; lsop->req_queued = false; INIT_LIST_HEAD(&lsop->lsreq_list); + INIT_WORK(&lsop->put_work, nvmet_fc_put_lsop_work); lsreq->rqstdma = fc_dma_map_single(tgtport->dev, lsreq->rqstaddr, lsreq->rqstlen + lsreq->rsplen, @@ -452,8 +449,6 @@ nvmet_fc_disconnect_assoc_done(struct nvmefc_ls_req *lsreq, int status) __nvmet_fc_finish_ls_req(lsop); /* fc-nvme target doesn't care about success or failure of cmd */ - - kfree(lsop); } /* @@ -464,7 +459,7 @@ nvmet_fc_disconnect_assoc_done(struct nvmefc_ls_req *lsreq, int status) * down, and the related FC-NVME Association ID and Connection IDs * become invalid. * - * The behavior of the fc-nvme target is such that it's + * The behavior of the fc-nvme target is such that its * understanding of the association and connections will implicitly * be torn down. The action is implicit as it may be due to a loss of * connectivity with the fc-nvme host, so the target may never get a @@ -488,16 +483,14 @@ nvmet_fc_xmt_disconnect_assoc(struct nvmet_fc_tgt_assoc *assoc) * message is normal. Otherwise, send unless the hostport has * already been invalidated by the lldd. */ - if (!tgtport->ops->ls_req || !assoc->hostport || - assoc->hostport->invalid) + if (!tgtport->ops->ls_req || assoc->hostport->invalid) return; lsop = kzalloc((sizeof(*lsop) + sizeof(*discon_rqst) + sizeof(*discon_acc) + tgtport->ops->lsrqst_priv_sz), GFP_KERNEL); if (!lsop) { - dev_info(tgtport->dev, - "{%d:%d} send Disconnect Association failed: ENOMEM\n", + pr_info("{%d:%d}: send Disconnect Association failed: ENOMEM\n", tgtport->fc_target_port.port_num, assoc->a_id); return; } @@ -519,8 +512,7 @@ nvmet_fc_xmt_disconnect_assoc(struct nvmet_fc_tgt_assoc *assoc) ret = nvmet_fc_send_ls_req_async(tgtport, lsop, nvmet_fc_disconnect_assoc_done); if (ret) { - dev_info(tgtport->dev, - "{%d:%d} XMT Disconnect Association failed: %d\n", + pr_info("{%d:%d}: XMT Disconnect Association failed: %d\n", tgtport->fc_target_port.port_num, assoc->a_id, ret); kfree(lsop); } @@ -801,14 +793,11 @@ nvmet_fc_alloc_target_queue(struct nvmet_fc_tgt_assoc *assoc, if (!queue) return NULL; - if (!nvmet_fc_tgt_a_get(assoc)) - goto out_free_queue; - queue->work_q = alloc_workqueue("ntfc%d.%d.%d", 0, 0, assoc->tgtport->fc_target_port.port_num, assoc->a_id, qid); if (!queue->work_q) - goto out_a_put; + goto out_free_queue; queue->qid = qid; queue->sqsize = sqsize; @@ -825,20 +814,20 @@ nvmet_fc_alloc_target_queue(struct nvmet_fc_tgt_assoc *assoc, nvmet_fc_prep_fcp_iodlist(assoc->tgtport, queue); - ret = nvmet_sq_init(&queue->nvme_sq); + nvmet_cq_init(&queue->nvme_cq); + ret = nvmet_sq_init(&queue->nvme_sq, &queue->nvme_cq); if (ret) goto out_fail_iodlist; WARN_ON(assoc->queues[qid]); - rcu_assign_pointer(assoc->queues[qid], queue); + assoc->queues[qid] = queue; return queue; out_fail_iodlist: + nvmet_cq_put(&queue->nvme_cq); nvmet_fc_destroy_fcp_iodlist(assoc->tgtport, queue); destroy_workqueue(queue->work_q); -out_a_put: - nvmet_fc_tgt_a_put(assoc); out_free_queue: kfree(queue); return NULL; @@ -851,15 +840,11 @@ nvmet_fc_tgt_queue_free(struct kref *ref) struct nvmet_fc_tgt_queue *queue = container_of(ref, struct nvmet_fc_tgt_queue, ref); - rcu_assign_pointer(queue->assoc->queues[queue->qid], NULL); - nvmet_fc_destroy_fcp_iodlist(queue->assoc->tgtport, queue); - nvmet_fc_tgt_a_put(queue->assoc); - destroy_workqueue(queue->work_q); - kfree_rcu(queue, rcu); + kfree(queue); } static void @@ -949,6 +934,7 @@ nvmet_fc_delete_target_queue(struct nvmet_fc_tgt_queue *queue) flush_workqueue(queue->work_q); nvmet_sq_destroy(&queue->nvme_sq); + nvmet_cq_put(&queue->nvme_cq); nvmet_fc_tgt_q_put(queue); } @@ -968,7 +954,7 @@ nvmet_fc_find_target_queue(struct nvmet_fc_tgtport *tgtport, rcu_read_lock(); list_for_each_entry_rcu(assoc, &tgtport->assoc_list, a_list) { if (association_id == assoc->association_id) { - queue = rcu_dereference(assoc->queues[qid]); + queue = assoc->queues[qid]; if (queue && (!atomic_read(&queue->connected) || !nvmet_fc_tgt_q_get(queue))) @@ -1010,16 +996,6 @@ nvmet_fc_hostport_get(struct nvmet_fc_hostport *hostport) return kref_get_unless_zero(&hostport->ref); } -static void -nvmet_fc_free_hostport(struct nvmet_fc_hostport *hostport) -{ - /* if LLDD not implemented, leave as NULL */ - if (!hostport || !hostport->hosthandle) - return; - - nvmet_fc_hostport_put(hostport); -} - static struct nvmet_fc_hostport * nvmet_fc_match_hostport(struct nvmet_fc_tgtport *tgtport, void *hosthandle) { @@ -1030,7 +1006,7 @@ nvmet_fc_match_hostport(struct nvmet_fc_tgtport *tgtport, void *hosthandle) list_for_each_entry(host, &tgtport->host_list, host_list) { if (host->hosthandle == hosthandle && !host->invalid) { if (nvmet_fc_hostport_get(host)) - return (host); + return host; } } @@ -1043,33 +1019,24 @@ nvmet_fc_alloc_hostport(struct nvmet_fc_tgtport *tgtport, void *hosthandle) struct nvmet_fc_hostport *newhost, *match = NULL; unsigned long flags; + /* + * Caller holds a reference on tgtport. + */ + /* if LLDD not implemented, leave as NULL */ if (!hosthandle) return NULL; - /* - * take reference for what will be the newly allocated hostport if - * we end up using a new allocation - */ - if (!nvmet_fc_tgtport_get(tgtport)) - return ERR_PTR(-EINVAL); - spin_lock_irqsave(&tgtport->lock, flags); match = nvmet_fc_match_hostport(tgtport, hosthandle); spin_unlock_irqrestore(&tgtport->lock, flags); - if (match) { - /* no new allocation - release reference */ - nvmet_fc_tgtport_put(tgtport); + if (match) return match; - } newhost = kzalloc(sizeof(*newhost), GFP_KERNEL); - if (!newhost) { - /* no new allocation - release reference */ - nvmet_fc_tgtport_put(tgtport); + if (!newhost) return ERR_PTR(-ENOMEM); - } spin_lock_irqsave(&tgtport->lock, flags); match = nvmet_fc_match_hostport(tgtport, hosthandle); @@ -1077,9 +1044,8 @@ nvmet_fc_alloc_hostport(struct nvmet_fc_tgtport *tgtport, void *hosthandle) /* new allocation not needed */ kfree(newhost); newhost = match; - /* no new allocation - release reference */ - nvmet_fc_tgtport_put(tgtport); } else { + nvmet_fc_tgtport_get(tgtport); newhost->tgtport = tgtport; newhost->hosthandle = hosthandle; INIT_LIST_HEAD(&newhost->host_list); @@ -1093,23 +1059,62 @@ nvmet_fc_alloc_hostport(struct nvmet_fc_tgtport *tgtport, void *hosthandle) } static void -nvmet_fc_delete_assoc(struct work_struct *work) +nvmet_fc_delete_assoc_work(struct work_struct *work) { struct nvmet_fc_tgt_assoc *assoc = container_of(work, struct nvmet_fc_tgt_assoc, del_work); + struct nvmet_fc_tgtport *tgtport = assoc->tgtport; nvmet_fc_delete_target_assoc(assoc); nvmet_fc_tgt_a_put(assoc); + nvmet_fc_tgtport_put(tgtport); +} + +static void +nvmet_fc_schedule_delete_assoc(struct nvmet_fc_tgt_assoc *assoc) +{ + int terminating; + + terminating = atomic_xchg(&assoc->terminating, 1); + + /* if already terminating, do nothing */ + if (terminating) + return; + + nvmet_fc_tgtport_get(assoc->tgtport); + if (!queue_work(nvmet_wq, &assoc->del_work)) + nvmet_fc_tgtport_put(assoc->tgtport); +} + +static bool +nvmet_fc_assoc_exists(struct nvmet_fc_tgtport *tgtport, u64 association_id) +{ + struct nvmet_fc_tgt_assoc *a; + bool found = false; + + rcu_read_lock(); + list_for_each_entry_rcu(a, &tgtport->assoc_list, a_list) { + if (association_id == a->association_id) { + found = true; + break; + } + } + rcu_read_unlock(); + + return found; } static struct nvmet_fc_tgt_assoc * nvmet_fc_alloc_target_assoc(struct nvmet_fc_tgtport *tgtport, void *hosthandle) { - struct nvmet_fc_tgt_assoc *assoc, *tmpassoc; + struct nvmet_fc_tgt_assoc *assoc; unsigned long flags; + bool done; u64 ran; int idx; - bool needrandom = true; + + if (!tgtport->pe) + return NULL; assoc = kzalloc(sizeof(*assoc), GFP_KERNEL); if (!assoc) @@ -1119,43 +1124,34 @@ nvmet_fc_alloc_target_assoc(struct nvmet_fc_tgtport *tgtport, void *hosthandle) if (idx < 0) goto out_free_assoc; - if (!nvmet_fc_tgtport_get(tgtport)) - goto out_ida; - assoc->hostport = nvmet_fc_alloc_hostport(tgtport, hosthandle); if (IS_ERR(assoc->hostport)) - goto out_put; + goto out_ida; assoc->tgtport = tgtport; + nvmet_fc_tgtport_get(tgtport); assoc->a_id = idx; INIT_LIST_HEAD(&assoc->a_list); kref_init(&assoc->ref); - INIT_WORK(&assoc->del_work, nvmet_fc_delete_assoc); + INIT_WORK(&assoc->del_work, nvmet_fc_delete_assoc_work); atomic_set(&assoc->terminating, 0); - while (needrandom) { + done = false; + do { get_random_bytes(&ran, sizeof(ran) - BYTES_FOR_QID); ran = ran << BYTES_FOR_QID_SHIFT; spin_lock_irqsave(&tgtport->lock, flags); - needrandom = false; - list_for_each_entry(tmpassoc, &tgtport->assoc_list, a_list) { - if (ran == tmpassoc->association_id) { - needrandom = true; - break; - } - } - if (!needrandom) { + if (!nvmet_fc_assoc_exists(tgtport, ran)) { assoc->association_id = ran; list_add_tail_rcu(&assoc->a_list, &tgtport->assoc_list); + done = true; } spin_unlock_irqrestore(&tgtport->lock, flags); - } + } while (!done); return assoc; -out_put: - nvmet_fc_tgtport_put(tgtport); out_ida: ida_free(&tgtport->assoc_cnt, idx); out_free_assoc: @@ -1171,24 +1167,27 @@ nvmet_fc_target_assoc_free(struct kref *ref) struct nvmet_fc_tgtport *tgtport = assoc->tgtport; struct nvmet_fc_ls_iod *oldls; unsigned long flags; + int i; + + for (i = NVMET_NR_QUEUES; i >= 0; i--) { + if (assoc->queues[i]) + nvmet_fc_delete_target_queue(assoc->queues[i]); + } /* Send Disconnect now that all i/o has completed */ nvmet_fc_xmt_disconnect_assoc(assoc); - nvmet_fc_free_hostport(assoc->hostport); + nvmet_fc_hostport_put(assoc->hostport); spin_lock_irqsave(&tgtport->lock, flags); - list_del_rcu(&assoc->a_list); oldls = assoc->rcv_disconn; spin_unlock_irqrestore(&tgtport->lock, flags); /* if pending Rcv Disconnect Association LS, send rsp now */ if (oldls) nvmet_fc_xmt_ls_rsp(tgtport, oldls); ida_free(&tgtport->assoc_cnt, assoc->a_id); - dev_info(tgtport->dev, - "{%d:%d} Association freed\n", + pr_info("{%d:%d}: Association freed\n", tgtport->fc_target_port.port_num, assoc->a_id); - kfree_rcu(assoc, rcu); - nvmet_fc_tgtport_put(tgtport); + kfree(assoc); } static void @@ -1207,38 +1206,25 @@ static void nvmet_fc_delete_target_assoc(struct nvmet_fc_tgt_assoc *assoc) { struct nvmet_fc_tgtport *tgtport = assoc->tgtport; - struct nvmet_fc_tgt_queue *queue; - int i, terminating; - - terminating = atomic_xchg(&assoc->terminating, 1); + unsigned long flags; + int i; - /* if already terminating, do nothing */ - if (terminating) - return; + spin_lock_irqsave(&tgtport->lock, flags); + list_del_rcu(&assoc->a_list); + spin_unlock_irqrestore(&tgtport->lock, flags); + synchronize_rcu(); + /* ensure all in-flight I/Os have been processed */ for (i = NVMET_NR_QUEUES; i >= 0; i--) { - rcu_read_lock(); - queue = rcu_dereference(assoc->queues[i]); - if (!queue) { - rcu_read_unlock(); - continue; - } - - if (!nvmet_fc_tgt_q_get(queue)) { - rcu_read_unlock(); - continue; - } - rcu_read_unlock(); - nvmet_fc_delete_target_queue(queue); - nvmet_fc_tgt_q_put(queue); + if (assoc->queues[i]) + flush_workqueue(assoc->queues[i]->work_q); } - dev_info(tgtport->dev, - "{%d:%d} Association deleted\n", + pr_info("{%d:%d}: Association deleted\n", tgtport->fc_target_port.port_num, assoc->a_id); - nvmet_fc_tgt_a_put(assoc); + nvmet_fc_tgtport_put(tgtport); } static struct nvmet_fc_tgt_assoc * @@ -1269,6 +1255,7 @@ nvmet_fc_portentry_bind(struct nvmet_fc_tgtport *tgtport, { lockdep_assert_held(&nvmet_fc_tgtlock); + nvmet_fc_tgtport_get(tgtport); pe->tgtport = tgtport; tgtport->pe = pe; @@ -1288,8 +1275,10 @@ nvmet_fc_portentry_unbind(struct nvmet_fc_port_entry *pe) unsigned long flags; spin_lock_irqsave(&nvmet_fc_tgtlock, flags); - if (pe->tgtport) + if (pe->tgtport) { + nvmet_fc_tgtport_put(pe->tgtport); pe->tgtport->pe = NULL; + } list_del(&pe->pe_list); spin_unlock_irqrestore(&nvmet_fc_tgtlock, flags); } @@ -1307,8 +1296,10 @@ nvmet_fc_portentry_unbind_tgt(struct nvmet_fc_tgtport *tgtport) spin_lock_irqsave(&nvmet_fc_tgtlock, flags); pe = tgtport->pe; - if (pe) + if (pe) { + nvmet_fc_tgtport_put(pe->tgtport); pe->tgtport = NULL; + } tgtport->pe = NULL; spin_unlock_irqrestore(&nvmet_fc_tgtlock, flags); } @@ -1331,6 +1322,9 @@ nvmet_fc_portentry_rebind_tgt(struct nvmet_fc_tgtport *tgtport) list_for_each_entry(pe, &nvmet_fc_portentry_list, pe_list) { if (tgtport->fc_target_port.node_name == pe->node_name && tgtport->fc_target_port.port_name == pe->port_name) { + if (!nvmet_fc_tgtport_get(tgtport)) + continue; + WARN_ON(pe->tgtport); tgtport->pe = pe; pe->tgtport = tgtport; @@ -1343,7 +1337,7 @@ nvmet_fc_portentry_rebind_tgt(struct nvmet_fc_tgtport *tgtport) /** * nvmet_fc_register_targetport - transport entry point called by an * LLDD to register the existence of a local - * NVME subystem FC port. + * NVME subsystem FC port. * @pinfo: pointer to information about the port to be registered * @template: LLDD entrypoints and operational parameters for the port * @dev: physical hardware device node port corresponds to. Will be @@ -1449,11 +1443,6 @@ nvmet_fc_free_tgtport(struct kref *ref) struct nvmet_fc_tgtport *tgtport = container_of(ref, struct nvmet_fc_tgtport, ref); struct device *dev = tgtport->dev; - unsigned long flags; - - spin_lock_irqsave(&nvmet_fc_tgtlock, flags); - list_del(&tgtport->tgt_list); - spin_unlock_irqrestore(&nvmet_fc_tgtlock, flags); nvmet_fc_free_ls_iodlist(tgtport); @@ -1491,9 +1480,8 @@ __nvmet_fc_free_assocs(struct nvmet_fc_tgtport *tgtport) list_for_each_entry_rcu(assoc, &tgtport->assoc_list, a_list) { if (!nvmet_fc_tgt_a_get(assoc)) continue; - if (!queue_work(nvmet_wq, &assoc->del_work)) - /* already deleting - release local reference */ - nvmet_fc_tgt_a_put(assoc); + nvmet_fc_schedule_delete_assoc(assoc); + nvmet_fc_tgt_a_put(assoc); } rcu_read_unlock(); } @@ -1539,16 +1527,14 @@ nvmet_fc_invalidate_host(struct nvmet_fc_target_port *target_port, spin_lock_irqsave(&tgtport->lock, flags); list_for_each_entry_safe(assoc, next, &tgtport->assoc_list, a_list) { - if (!assoc->hostport || - assoc->hostport->hosthandle != hosthandle) + if (assoc->hostport->hosthandle != hosthandle) continue; if (!nvmet_fc_tgt_a_get(assoc)) continue; assoc->hostport->invalid = 1; noassoc = false; - if (!queue_work(nvmet_wq, &assoc->del_work)) - /* already deleting - release local reference */ - nvmet_fc_tgt_a_put(assoc); + nvmet_fc_schedule_delete_assoc(assoc); + nvmet_fc_tgt_a_put(assoc); } spin_unlock_irqrestore(&tgtport->lock, flags); @@ -1580,7 +1566,7 @@ nvmet_fc_delete_ctrl(struct nvmet_ctrl *ctrl) rcu_read_lock(); list_for_each_entry_rcu(assoc, &tgtport->assoc_list, a_list) { - queue = rcu_dereference(assoc->queues[0]); + queue = assoc->queues[0]; if (queue && queue->nvme_sq.ctrl == ctrl) { if (nvmet_fc_tgt_a_get(assoc)) found_ctrl = true; @@ -1592,9 +1578,8 @@ nvmet_fc_delete_ctrl(struct nvmet_ctrl *ctrl) nvmet_fc_tgtport_put(tgtport); if (found_ctrl) { - if (!queue_work(nvmet_wq, &assoc->del_work)) - /* already deleting - release local reference */ - nvmet_fc_tgt_a_put(assoc); + nvmet_fc_schedule_delete_assoc(assoc); + nvmet_fc_tgt_a_put(assoc); return; } @@ -1603,6 +1588,39 @@ nvmet_fc_delete_ctrl(struct nvmet_ctrl *ctrl) spin_unlock_irqrestore(&nvmet_fc_tgtlock, flags); } +static void +nvmet_fc_free_pending_reqs(struct nvmet_fc_tgtport *tgtport) +{ + struct nvmet_fc_ls_req_op *lsop; + struct nvmefc_ls_req *lsreq; + struct nvmet_fc_ls_iod *iod; + int i; + + iod = tgtport->iod; + for (i = 0; i < NVMET_LS_CTX_COUNT; iod++, i++) + cancel_work(&iod->work); + + /* + * After this point the connection is lost and thus any pending + * request can't be processed by the normal completion path. This + * is likely a request from nvmet_fc_send_ls_req_async. + */ + while ((lsop = list_first_entry_or_null(&tgtport->ls_req_list, + struct nvmet_fc_ls_req_op, lsreq_list))) { + list_del(&lsop->lsreq_list); + + if (!lsop->req_queued) + continue; + + lsreq = &lsop->ls_req; + fc_dma_unmap_single(tgtport->dev, lsreq->rqstdma, + (lsreq->rqstlen + lsreq->rsplen), + DMA_BIDIRECTIONAL); + nvmet_fc_tgtport_put(tgtport); + kfree(lsop); + } +} + /** * nvmet_fc_unregister_targetport - transport entry point called by an * LLDD to deregister/remove a previously @@ -1618,19 +1636,20 @@ int nvmet_fc_unregister_targetport(struct nvmet_fc_target_port *target_port) { struct nvmet_fc_tgtport *tgtport = targetport_to_tgtport(target_port); + unsigned long flags; + + spin_lock_irqsave(&nvmet_fc_tgtlock, flags); + list_del(&tgtport->tgt_list); + spin_unlock_irqrestore(&nvmet_fc_tgtlock, flags); nvmet_fc_portentry_unbind_tgt(tgtport); /* terminate any outstanding associations */ __nvmet_fc_free_assocs(tgtport); - /* - * should terminate LS's as well. However, LS's will be generated - * at the tail end of association termination, so they likely don't - * exist yet. And even if they did, it's worthwhile to just let - * them finish and targetport ref counting will clean things up. - */ + flush_workqueue(nvmet_wq); + nvmet_fc_free_pending_reqs(tgtport); nvmet_fc_tgtport_put(tgtport); return 0; @@ -1685,15 +1704,17 @@ nvmet_fc_ls_create_association(struct nvmet_fc_tgtport *tgtport, else { queue = nvmet_fc_alloc_target_queue(iod->assoc, 0, be16_to_cpu(rqst->assoc_cmd.sqsize)); - if (!queue) + if (!queue) { ret = VERR_QUEUE_ALLOC_FAIL; + nvmet_fc_tgt_a_put(iod->assoc); + } } } if (ret) { - dev_err(tgtport->dev, - "Create Association LS failed: %s\n", - validation_errors[ret]); + pr_err("{%d}: Create Association LS failed: %s\n", + tgtport->fc_target_port.port_num, + validation_errors[ret]); iod->lsrsp->rsplen = nvme_fc_format_rjt(acc, sizeof(*acc), rqst->w0.ls_cmd, FCNVME_RJT_RC_LOGIC, @@ -1705,8 +1726,7 @@ nvmet_fc_ls_create_association(struct nvmet_fc_tgtport *tgtport, atomic_set(&queue->connected, 1); queue->sqhd = 0; /* best place to init value */ - dev_info(tgtport->dev, - "{%d:%d} Association created\n", + pr_info("{%d:%d}: Association created\n", tgtport->fc_target_port.port_num, iod->assoc->a_id); /* format a response */ @@ -1784,9 +1804,9 @@ nvmet_fc_ls_create_connection(struct nvmet_fc_tgtport *tgtport, } if (ret) { - dev_err(tgtport->dev, - "Create Connection LS failed: %s\n", - validation_errors[ret]); + pr_err("{%d}: Create Connection LS failed: %s\n", + tgtport->fc_target_port.port_num, + validation_errors[ret]); iod->lsrsp->rsplen = nvme_fc_format_rjt(acc, sizeof(*acc), rqst->w0.ls_cmd, (ret == VERR_NO_ASSOC) ? @@ -1846,9 +1866,9 @@ nvmet_fc_ls_disconnect(struct nvmet_fc_tgtport *tgtport, } if (ret || !assoc) { - dev_err(tgtport->dev, - "Disconnect LS failed: %s\n", - validation_errors[ret]); + pr_err("{%d}: Disconnect LS failed: %s\n", + tgtport->fc_target_port.port_num, + validation_errors[ret]); iod->lsrsp->rsplen = nvme_fc_format_rjt(acc, sizeof(*acc), rqst->w0.ls_cmd, (ret == VERR_NO_ASSOC) ? @@ -1867,9 +1887,6 @@ nvmet_fc_ls_disconnect(struct nvmet_fc_tgtport *tgtport, sizeof(struct fcnvme_ls_disconnect_assoc_acc)), FCNVME_LS_DISCONNECT_ASSOC); - /* release get taken in nvmet_fc_find_target_assoc */ - nvmet_fc_tgt_a_put(assoc); - /* * The rules for LS response says the response cannot * go back until ABTS's have been sent for all outstanding @@ -1884,11 +1901,8 @@ nvmet_fc_ls_disconnect(struct nvmet_fc_tgtport *tgtport, assoc->rcv_disconn = iod; spin_unlock_irqrestore(&tgtport->lock, flags); - nvmet_fc_delete_target_assoc(assoc); - if (oldls) { - dev_info(tgtport->dev, - "{%d:%d} Multiple Disconnect Association LS's " + pr_info("{%d:%d}: Multiple Disconnect Association LS's " "received\n", tgtport->fc_target_port.port_num, assoc->a_id); /* overwrite good response with bogus failure */ @@ -1901,6 +1915,9 @@ nvmet_fc_ls_disconnect(struct nvmet_fc_tgtport *tgtport, nvmet_fc_xmt_ls_rsp(tgtport, oldls); } + nvmet_fc_schedule_delete_assoc(assoc); + nvmet_fc_tgt_a_put(assoc); + return false; } @@ -2028,8 +2045,8 @@ nvmet_fc_rcv_ls_req(struct nvmet_fc_target_port *target_port, struct fcnvme_ls_rqst_w0 *w0 = (struct fcnvme_ls_rqst_w0 *)lsreqbuf; if (lsreqbuf_len > sizeof(union nvmefc_ls_requests)) { - dev_info(tgtport->dev, - "RCV %s LS failed: payload too large (%d)\n", + pr_info("{%d}: RCV %s LS failed: payload too large (%d)\n", + tgtport->fc_target_port.port_num, (w0->ls_cmd <= NVME_FC_LAST_LS_CMD_VALUE) ? nvmefc_ls_names[w0->ls_cmd] : "", lsreqbuf_len); @@ -2037,8 +2054,8 @@ nvmet_fc_rcv_ls_req(struct nvmet_fc_target_port *target_port, } if (!nvmet_fc_tgtport_get(tgtport)) { - dev_info(tgtport->dev, - "RCV %s LS failed: target deleting\n", + pr_info("{%d}: RCV %s LS failed: target deleting\n", + tgtport->fc_target_port.port_num, (w0->ls_cmd <= NVME_FC_LAST_LS_CMD_VALUE) ? nvmefc_ls_names[w0->ls_cmd] : ""); return -ESHUTDOWN; @@ -2046,8 +2063,8 @@ nvmet_fc_rcv_ls_req(struct nvmet_fc_target_port *target_port, iod = nvmet_fc_alloc_ls_iod(tgtport); if (!iod) { - dev_info(tgtport->dev, - "RCV %s LS failed: context allocation failed\n", + pr_info("{%d}: RCV %s LS failed: context allocation failed\n", + tgtport->fc_target_port.port_num, (w0->ls_cmd <= NVME_FC_LAST_LS_CMD_VALUE) ? nvmefc_ls_names[w0->ls_cmd] : ""); nvmet_fc_tgtport_put(tgtport); @@ -2291,7 +2308,7 @@ nvmet_fc_transfer_fcp_data(struct nvmet_fc_tgtport *tgtport, ret = tgtport->ops->fcp_op(&tgtport->fc_target_port, fod->fcpreq); if (ret) { /* - * should be ok to set w/o lock as its in the thread of + * should be ok to set w/o lock as it's in the thread of * execution (not an async timer routine) and doesn't * contend with any clearing action */ @@ -2537,8 +2554,9 @@ nvmet_fc_handle_fcp_rqst(struct nvmet_fc_tgtport *tgtport, fod->req.cmd = &fod->cmdiubuf.sqe; fod->req.cqe = &fod->rspiubuf.cqe; - if (tgtport->pe) - fod->req.port = tgtport->pe->port; + if (!tgtport->pe) + goto transport_error; + fod->req.port = tgtport->pe->port; /* clear any response payload */ memset(&fod->rspiubuf, 0, sizeof(fod->rspiubuf)); @@ -2546,10 +2564,8 @@ nvmet_fc_handle_fcp_rqst(struct nvmet_fc_tgtport *tgtport, fod->data_sg = NULL; fod->data_sg_cnt = 0; - ret = nvmet_req_init(&fod->req, - &fod->queue->nvme_cq, - &fod->queue->nvme_sq, - &nvmet_fc_tgt_fcp_ops); + ret = nvmet_req_init(&fod->req, &fod->queue->nvme_sq, + &nvmet_fc_tgt_fcp_ops); if (!ret) { /* bad SQE content or invalid ctrl state */ /* nvmet layer has already called op done to send rsp. */ @@ -2608,7 +2624,7 @@ transport_error: * and the api of the FC LLDD which may issue a hw command to send the * response, but the LLDD may not get the hw completion for that command * and upcall the nvmet_fc layer before a new command may be - * asynchronously received - its possible for a command to be received + * asynchronously received - it's possible for a command to be received * before the LLDD and nvmet_fc have recycled the job structure. It gives * the appearance of more commands received than fits in the sq. * To alleviate this scenario, a temporary queue is maintained in the @@ -2875,12 +2891,17 @@ nvmet_fc_add_port(struct nvmet_port *port) list_for_each_entry(tgtport, &nvmet_fc_target_list, tgt_list) { if ((tgtport->fc_target_port.node_name == traddr.nn) && (tgtport->fc_target_port.port_name == traddr.pn)) { + if (!nvmet_fc_tgtport_get(tgtport)) + continue; + /* a FC port can only be 1 nvmet port id */ if (!tgtport->pe) { nvmet_fc_portentry_bind(tgtport, pe, port); ret = 0; } else ret = -EALREADY; + + nvmet_fc_tgtport_put(tgtport); break; } } @@ -2896,9 +2917,22 @@ static void nvmet_fc_remove_port(struct nvmet_port *port) { struct nvmet_fc_port_entry *pe = port->priv; + struct nvmet_fc_tgtport *tgtport = NULL; + unsigned long flags; + + spin_lock_irqsave(&nvmet_fc_tgtlock, flags); + if (pe->tgtport && nvmet_fc_tgtport_get(pe->tgtport)) + tgtport = pe->tgtport; + spin_unlock_irqrestore(&nvmet_fc_tgtlock, flags); nvmet_fc_portentry_unbind(pe); + if (tgtport) { + /* terminate any outstanding associations */ + __nvmet_fc_free_assocs(tgtport); + nvmet_fc_tgtport_put(tgtport); + } + kfree(pe); } @@ -2906,10 +2940,53 @@ static void nvmet_fc_discovery_chg(struct nvmet_port *port) { struct nvmet_fc_port_entry *pe = port->priv; - struct nvmet_fc_tgtport *tgtport = pe->tgtport; + struct nvmet_fc_tgtport *tgtport = NULL; + unsigned long flags; + + spin_lock_irqsave(&nvmet_fc_tgtlock, flags); + if (pe->tgtport && nvmet_fc_tgtport_get(pe->tgtport)) + tgtport = pe->tgtport; + spin_unlock_irqrestore(&nvmet_fc_tgtlock, flags); + + if (!tgtport) + return; if (tgtport && tgtport->ops->discovery_event) tgtport->ops->discovery_event(&tgtport->fc_target_port); + + nvmet_fc_tgtport_put(tgtport); +} + +static ssize_t +nvmet_fc_host_traddr(struct nvmet_ctrl *ctrl, + char *traddr, size_t traddr_size) +{ + struct nvmet_sq *sq = ctrl->sqs[0]; + struct nvmet_fc_tgt_queue *queue = + container_of(sq, struct nvmet_fc_tgt_queue, nvme_sq); + struct nvmet_fc_tgtport *tgtport = queue->assoc ? queue->assoc->tgtport : NULL; + struct nvmet_fc_hostport *hostport = queue->assoc ? queue->assoc->hostport : NULL; + u64 wwnn, wwpn; + ssize_t ret = 0; + + if (!tgtport || !nvmet_fc_tgtport_get(tgtport)) + return -ENODEV; + if (!hostport || !nvmet_fc_hostport_get(hostport)) { + ret = -ENODEV; + goto out_put; + } + + if (tgtport->ops->host_traddr) { + ret = tgtport->ops->host_traddr(hostport->hosthandle, &wwnn, &wwpn); + if (ret) + goto out_put_host; + ret = snprintf(traddr, traddr_size, "nn-0x%llx:pn-0x%llx", wwnn, wwpn); + } +out_put_host: + nvmet_fc_hostport_put(hostport); +out_put: + nvmet_fc_tgtport_put(tgtport); + return ret; } static const struct nvmet_fabrics_ops nvmet_fc_tgt_fcp_ops = { @@ -2921,6 +2998,7 @@ static const struct nvmet_fabrics_ops nvmet_fc_tgt_fcp_ops = { .queue_response = nvmet_fc_fcp_nvme_cmd_done, .delete_ctrl = nvmet_fc_delete_ctrl, .discovery_chg = nvmet_fc_discovery_chg, + .host_traddr = nvmet_fc_host_traddr, }; static int __init nvmet_fc_init_module(void) @@ -2930,6 +3008,9 @@ static int __init nvmet_fc_init_module(void) static void __exit nvmet_fc_exit_module(void) { + /* ensure any shutdown operation, e.g. delete ctrls have finished */ + flush_workqueue(nvmet_wq); + /* sanity check - all lports should be removed */ if (!list_empty(&nvmet_fc_target_list)) pr_warn("%s: targetport list not empty\n", __func__); @@ -2942,4 +3023,5 @@ static void __exit nvmet_fc_exit_module(void) module_init(nvmet_fc_init_module); module_exit(nvmet_fc_exit_module); +MODULE_DESCRIPTION("NVMe target FC transport driver"); MODULE_LICENSE("GPL v2"); diff --git a/drivers/nvme/target/fcloop.c b/drivers/nvme/target/fcloop.c index 5c16372f3b53..c30e9a3e014f 100644 --- a/drivers/nvme/target/fcloop.c +++ b/drivers/nvme/target/fcloop.c @@ -207,13 +207,16 @@ static LIST_HEAD(fcloop_nports); struct fcloop_lport { struct nvme_fc_local_port *localport; struct list_head lport_list; - struct completion unreg_done; + refcount_t ref; }; struct fcloop_lport_priv { struct fcloop_lport *lport; }; +/* The port is already being removed, avoid double free */ +#define PORT_DELETED 0 + struct fcloop_rport { struct nvme_fc_remote_port *remoteport; struct nvmet_fc_target_port *targetport; @@ -222,6 +225,7 @@ struct fcloop_rport { spinlock_t lock; struct list_head ls_list; struct work_struct ls_work; + unsigned long flags; }; struct fcloop_tport { @@ -232,6 +236,7 @@ struct fcloop_tport { spinlock_t lock; struct list_head ls_list; struct work_struct ls_work; + unsigned long flags; }; struct fcloop_nport { @@ -239,7 +244,7 @@ struct fcloop_nport { struct fcloop_tport *tport; struct fcloop_lport *lport; struct list_head nport_list; - struct kref ref; + refcount_t ref; u64 node_name; u64 port_name; u32 port_role; @@ -249,7 +254,6 @@ struct fcloop_nport { struct fcloop_lsreq { struct nvmefc_ls_req *lsreq; struct nvmefc_ls_rsp ls_rsp; - int lsdir; /* H2T or T2H */ int status; struct list_head ls_list; /* fcloop_rport->ls_list */ }; @@ -274,7 +278,7 @@ struct fcloop_fcpreq { u32 inistate; bool active; bool aborted; - struct kref ref; + refcount_t ref; struct work_struct fcp_rcv_work; struct work_struct abort_rcv_work; struct work_struct tio_done_work; @@ -287,6 +291,9 @@ struct fcloop_ini_fcpreq { spinlock_t inilock; }; +/* SLAB cache for fcloop_lsreq structures */ +static struct kmem_cache *lsreq_cache; + static inline struct fcloop_lsreq * ls_rsp_to_lsreq(struct nvmefc_ls_rsp *lsrsp) { @@ -337,6 +344,7 @@ fcloop_rport_lsrqst_work(struct work_struct *work) * callee may free memory containing tls_req. * do not reference lsreq after this. */ + kmem_cache_free(lsreq_cache, tls_req); spin_lock(&rport->lock); } @@ -348,17 +356,20 @@ fcloop_h2t_ls_req(struct nvme_fc_local_port *localport, struct nvme_fc_remote_port *remoteport, struct nvmefc_ls_req *lsreq) { - struct fcloop_lsreq *tls_req = lsreq->private; struct fcloop_rport *rport = remoteport->private; + struct fcloop_lsreq *tls_req; int ret = 0; + tls_req = kmem_cache_alloc(lsreq_cache, GFP_KERNEL); + if (!tls_req) + return -ENOMEM; tls_req->lsreq = lsreq; INIT_LIST_HEAD(&tls_req->ls_list); if (!rport->targetport) { tls_req->status = -ECONNREFUSED; spin_lock(&rport->lock); - list_add_tail(&rport->ls_list, &tls_req->ls_list); + list_add_tail(&tls_req->ls_list, &rport->ls_list); spin_unlock(&rport->lock); queue_work(nvmet_wq, &rport->ls_work); return ret; @@ -388,14 +399,17 @@ fcloop_h2t_xmt_ls_rsp(struct nvmet_fc_target_port *targetport, lsrsp->done(lsrsp); - if (remoteport) { - rport = remoteport->private; - spin_lock(&rport->lock); - list_add_tail(&rport->ls_list, &tls_req->ls_list); - spin_unlock(&rport->lock); - queue_work(nvmet_wq, &rport->ls_work); + if (!remoteport) { + kmem_cache_free(lsreq_cache, tls_req); + return 0; } + rport = remoteport->private; + spin_lock(&rport->lock); + list_add_tail(&tls_req->ls_list, &rport->ls_list); + spin_unlock(&rport->lock); + queue_work(nvmet_wq, &rport->ls_work); + return 0; } @@ -421,6 +435,7 @@ fcloop_tport_lsrqst_work(struct work_struct *work) * callee may free memory containing tls_req. * do not reference lsreq after this. */ + kmem_cache_free(lsreq_cache, tls_req); spin_lock(&tport->lock); } @@ -431,8 +446,8 @@ static int fcloop_t2h_ls_req(struct nvmet_fc_target_port *targetport, void *hosthandle, struct nvmefc_ls_req *lsreq) { - struct fcloop_lsreq *tls_req = lsreq->private; struct fcloop_tport *tport = targetport->private; + struct fcloop_lsreq *tls_req; int ret = 0; /* @@ -440,13 +455,17 @@ fcloop_t2h_ls_req(struct nvmet_fc_target_port *targetport, void *hosthandle, * hosthandle ignored as fcloop currently is * 1:1 tgtport vs remoteport */ + + tls_req = kmem_cache_alloc(lsreq_cache, GFP_KERNEL); + if (!tls_req) + return -ENOMEM; tls_req->lsreq = lsreq; INIT_LIST_HEAD(&tls_req->ls_list); if (!tport->remoteport) { tls_req->status = -ECONNREFUSED; spin_lock(&tport->lock); - list_add_tail(&tport->ls_list, &tls_req->ls_list); + list_add_tail(&tls_req->ls_list, &tport->ls_list); spin_unlock(&tport->lock); queue_work(nvmet_wq, &tport->ls_work); return ret; @@ -456,6 +475,9 @@ fcloop_t2h_ls_req(struct nvmet_fc_target_port *targetport, void *hosthandle, ret = nvme_fc_rcv_ls_req(tport->remoteport, &tls_req->ls_rsp, lsreq->rqstaddr, lsreq->rqstlen); + if (ret) + kmem_cache_free(lsreq_cache, tls_req); + return ret; } @@ -470,18 +492,32 @@ fcloop_t2h_xmt_ls_rsp(struct nvme_fc_local_port *localport, struct nvmet_fc_target_port *targetport = rport->targetport; struct fcloop_tport *tport; + if (!targetport) { + /* + * The target port is gone. The target doesn't expect any + * response anymore and thus lsreq can't be accessed anymore. + * + * We end up here from delete association exchange: + * nvmet_fc_xmt_disconnect_assoc sends an async request. + * + * Return success because this is what LLDDs do; silently + * drop the response. + */ + lsrsp->done(lsrsp); + kmem_cache_free(lsreq_cache, tls_req); + return 0; + } + memcpy(lsreq->rspaddr, lsrsp->rspbuf, ((lsreq->rsplen < lsrsp->rsplen) ? lsreq->rsplen : lsrsp->rsplen)); lsrsp->done(lsrsp); - if (targetport) { - tport = targetport->private; - spin_lock(&tport->lock); - list_add_tail(&tport->ls_list, &tls_req->ls_list); - spin_unlock(&tport->lock); - queue_work(nvmet_wq, &tport->ls_work); - } + tport = targetport->private; + spin_lock(&tport->lock); + list_add_tail(&tls_req->ls_list, &tport->ls_list); + spin_unlock(&tport->lock); + queue_work(nvmet_wq, &tport->ls_work); return 0; } @@ -492,6 +528,16 @@ fcloop_t2h_host_release(void *hosthandle) /* host handle ignored for now */ } +static int +fcloop_t2h_host_traddr(void *hosthandle, u64 *wwnn, u64 *wwpn) +{ + struct fcloop_rport *rport = hosthandle; + + *wwnn = rport->lport->localport->node_name; + *wwpn = rport->lport->localport->port_name; + return 0; +} + /* * Simulate reception of RSCN and converting it to a initiator transport * call to rescan a remote port. @@ -524,24 +570,18 @@ fcloop_tgt_discovery_evt(struct nvmet_fc_target_port *tgtport) } static void -fcloop_tfcp_req_free(struct kref *ref) +fcloop_tfcp_req_put(struct fcloop_fcpreq *tfcp_req) { - struct fcloop_fcpreq *tfcp_req = - container_of(ref, struct fcloop_fcpreq, ref); + if (!refcount_dec_and_test(&tfcp_req->ref)) + return; kfree(tfcp_req); } -static void -fcloop_tfcp_req_put(struct fcloop_fcpreq *tfcp_req) -{ - kref_put(&tfcp_req->ref, fcloop_tfcp_req_free); -} - static int fcloop_tfcp_req_get(struct fcloop_fcpreq *tfcp_req) { - return kref_get_unless_zero(&tfcp_req->ref); + return refcount_inc_not_zero(&tfcp_req->ref); } static void @@ -561,7 +601,8 @@ fcloop_call_host_done(struct nvmefc_fcp_req *fcpreq, } /* release original io reference on tgt struct */ - fcloop_tfcp_req_put(tfcp_req); + if (tfcp_req) + fcloop_tfcp_req_put(tfcp_req); } static bool drop_fabric_opcode; @@ -613,11 +654,13 @@ fcloop_fcp_recv_work(struct work_struct *work) { struct fcloop_fcpreq *tfcp_req = container_of(work, struct fcloop_fcpreq, fcp_rcv_work); - struct nvmefc_fcp_req *fcpreq = tfcp_req->fcpreq; + struct nvmefc_fcp_req *fcpreq; + unsigned long flags; int ret = 0; bool aborted = false; - spin_lock_irq(&tfcp_req->reqlock); + spin_lock_irqsave(&tfcp_req->reqlock, flags); + fcpreq = tfcp_req->fcpreq; switch (tfcp_req->inistate) { case INI_IO_START: tfcp_req->inistate = INI_IO_ACTIVE; @@ -626,26 +669,27 @@ fcloop_fcp_recv_work(struct work_struct *work) aborted = true; break; default: - spin_unlock_irq(&tfcp_req->reqlock); + spin_unlock_irqrestore(&tfcp_req->reqlock, flags); WARN_ON(1); return; } - spin_unlock_irq(&tfcp_req->reqlock); - - if (unlikely(aborted)) - ret = -ECANCELED; - else { - if (likely(!check_for_drop(tfcp_req))) - ret = nvmet_fc_rcv_fcp_req(tfcp_req->tport->targetport, - &tfcp_req->tgt_fcp_req, - fcpreq->cmdaddr, fcpreq->cmdlen); - else - pr_info("%s: dropped command ********\n", __func__); + spin_unlock_irqrestore(&tfcp_req->reqlock, flags); + + if (unlikely(aborted)) { + /* the abort handler will call fcloop_call_host_done */ + return; } + + if (unlikely(check_for_drop(tfcp_req))) { + pr_info("%s: dropped command ********\n", __func__); + return; + } + + ret = nvmet_fc_rcv_fcp_req(tfcp_req->tport->targetport, + &tfcp_req->tgt_fcp_req, + fcpreq->cmdaddr, fcpreq->cmdlen); if (ret) fcloop_call_host_done(fcpreq, tfcp_req, ret); - - return; } static void @@ -655,21 +699,24 @@ fcloop_fcp_abort_recv_work(struct work_struct *work) container_of(work, struct fcloop_fcpreq, abort_rcv_work); struct nvmefc_fcp_req *fcpreq; bool completed = false; + unsigned long flags; - spin_lock_irq(&tfcp_req->reqlock); - fcpreq = tfcp_req->fcpreq; + spin_lock_irqsave(&tfcp_req->reqlock, flags); switch (tfcp_req->inistate) { case INI_IO_ABORTED: + fcpreq = tfcp_req->fcpreq; + tfcp_req->fcpreq = NULL; break; case INI_IO_COMPLETED: completed = true; break; default: - spin_unlock_irq(&tfcp_req->reqlock); + spin_unlock_irqrestore(&tfcp_req->reqlock, flags); + fcloop_tfcp_req_put(tfcp_req); WARN_ON(1); return; } - spin_unlock_irq(&tfcp_req->reqlock); + spin_unlock_irqrestore(&tfcp_req->reqlock, flags); if (unlikely(completed)) { /* remove reference taken in original abort downcall */ @@ -681,10 +728,6 @@ fcloop_fcp_abort_recv_work(struct work_struct *work) nvmet_fc_rcv_fcp_abort(tfcp_req->tport->targetport, &tfcp_req->tgt_fcp_req); - spin_lock_irq(&tfcp_req->reqlock); - tfcp_req->fcpreq = NULL; - spin_unlock_irq(&tfcp_req->reqlock); - fcloop_call_host_done(fcpreq, tfcp_req, -ECANCELED); /* call_host_done releases reference for abort downcall */ } @@ -699,11 +742,12 @@ fcloop_tgt_fcprqst_done_work(struct work_struct *work) struct fcloop_fcpreq *tfcp_req = container_of(work, struct fcloop_fcpreq, tio_done_work); struct nvmefc_fcp_req *fcpreq; + unsigned long flags; - spin_lock_irq(&tfcp_req->reqlock); + spin_lock_irqsave(&tfcp_req->reqlock, flags); fcpreq = tfcp_req->fcpreq; tfcp_req->inistate = INI_IO_COMPLETED; - spin_unlock_irq(&tfcp_req->reqlock); + spin_unlock_irqrestore(&tfcp_req->reqlock, flags); fcloop_call_host_done(fcpreq, tfcp_req, tfcp_req->status); } @@ -737,7 +781,7 @@ fcloop_fcp_req(struct nvme_fc_local_port *localport, INIT_WORK(&tfcp_req->fcp_rcv_work, fcloop_fcp_recv_work); INIT_WORK(&tfcp_req->abort_rcv_work, fcloop_fcp_abort_recv_work); INIT_WORK(&tfcp_req->tio_done_work, fcloop_tgt_fcprqst_done_work); - kref_init(&tfcp_req->ref); + refcount_set(&tfcp_req->ref, 1); queue_work(nvmet_wq, &tfcp_req->fcp_rcv_work); @@ -807,13 +851,14 @@ fcloop_fcp_op(struct nvmet_fc_target_port *tgtport, u32 rsplen = 0, xfrlen = 0; int fcp_err = 0, active, aborted; u8 op = tgt_fcpreq->op; + unsigned long flags; - spin_lock_irq(&tfcp_req->reqlock); + spin_lock_irqsave(&tfcp_req->reqlock, flags); fcpreq = tfcp_req->fcpreq; active = tfcp_req->active; aborted = tfcp_req->aborted; tfcp_req->active = true; - spin_unlock_irq(&tfcp_req->reqlock); + spin_unlock_irqrestore(&tfcp_req->reqlock, flags); if (unlikely(active)) /* illegal - call while i/o active */ @@ -821,9 +866,9 @@ fcloop_fcp_op(struct nvmet_fc_target_port *tgtport, if (unlikely(aborted)) { /* target transport has aborted i/o prior */ - spin_lock_irq(&tfcp_req->reqlock); + spin_lock_irqsave(&tfcp_req->reqlock, flags); tfcp_req->active = false; - spin_unlock_irq(&tfcp_req->reqlock); + spin_unlock_irqrestore(&tfcp_req->reqlock, flags); tgt_fcpreq->transferred_length = 0; tgt_fcpreq->fcp_error = -ECANCELED; tgt_fcpreq->done(tgt_fcpreq); @@ -880,9 +925,9 @@ fcloop_fcp_op(struct nvmet_fc_target_port *tgtport, break; } - spin_lock_irq(&tfcp_req->reqlock); + spin_lock_irqsave(&tfcp_req->reqlock, flags); tfcp_req->active = false; - spin_unlock_irq(&tfcp_req->reqlock); + spin_unlock_irqrestore(&tfcp_req->reqlock, flags); tgt_fcpreq->transferred_length = xfrlen; tgt_fcpreq->fcp_error = fcp_err; @@ -896,15 +941,16 @@ fcloop_tgt_fcp_abort(struct nvmet_fc_target_port *tgtport, struct nvmefc_tgt_fcp_req *tgt_fcpreq) { struct fcloop_fcpreq *tfcp_req = tgt_fcp_req_to_fcpreq(tgt_fcpreq); + unsigned long flags; /* * mark aborted only in case there were 2 threads in transport * (one doing io, other doing abort) and only kills ops posted * after the abort request */ - spin_lock_irq(&tfcp_req->reqlock); + spin_lock_irqsave(&tfcp_req->reqlock, flags); tfcp_req->aborted = true; - spin_unlock_irq(&tfcp_req->reqlock); + spin_unlock_irqrestore(&tfcp_req->reqlock, flags); tfcp_req->status = NVME_SC_INTERNAL; @@ -946,19 +992,23 @@ fcloop_fcp_abort(struct nvme_fc_local_port *localport, struct fcloop_ini_fcpreq *inireq = fcpreq->private; struct fcloop_fcpreq *tfcp_req; bool abortio = true; + unsigned long flags; spin_lock(&inireq->inilock); tfcp_req = inireq->tfcp_req; - if (tfcp_req) - fcloop_tfcp_req_get(tfcp_req); + if (tfcp_req) { + if (!fcloop_tfcp_req_get(tfcp_req)) + tfcp_req = NULL; + } spin_unlock(&inireq->inilock); - if (!tfcp_req) + if (!tfcp_req) { /* abort has already been called */ - return; + goto out_host_done; + } /* break initiator/target relationship for io */ - spin_lock_irq(&tfcp_req->reqlock); + spin_lock_irqsave(&tfcp_req->reqlock, flags); switch (tfcp_req->inistate) { case INI_IO_START: case INI_IO_ACTIVE: @@ -968,11 +1018,11 @@ fcloop_fcp_abort(struct nvme_fc_local_port *localport, abortio = false; break; default: - spin_unlock_irq(&tfcp_req->reqlock); + spin_unlock_irqrestore(&tfcp_req->reqlock, flags); WARN_ON(1); - return; + goto out_host_done; } - spin_unlock_irq(&tfcp_req->reqlock); + spin_unlock_irqrestore(&tfcp_req->reqlock, flags); if (abortio) /* leave the reference while the work item is scheduled */ @@ -984,32 +1034,56 @@ fcloop_fcp_abort(struct nvme_fc_local_port *localport, */ fcloop_tfcp_req_put(tfcp_req); } + + return; + +out_host_done: + fcloop_call_host_done(fcpreq, tfcp_req, -ECANCELED); } static void -fcloop_nport_free(struct kref *ref) +fcloop_lport_put(struct fcloop_lport *lport) { - struct fcloop_nport *nport = - container_of(ref, struct fcloop_nport, ref); unsigned long flags; + if (!refcount_dec_and_test(&lport->ref)) + return; + spin_lock_irqsave(&fcloop_lock, flags); - list_del(&nport->nport_list); + list_del(&lport->lport_list); spin_unlock_irqrestore(&fcloop_lock, flags); - kfree(nport); + kfree(lport); +} + +static int +fcloop_lport_get(struct fcloop_lport *lport) +{ + return refcount_inc_not_zero(&lport->ref); } static void fcloop_nport_put(struct fcloop_nport *nport) { - kref_put(&nport->ref, fcloop_nport_free); + unsigned long flags; + + if (!refcount_dec_and_test(&nport->ref)) + return; + + spin_lock_irqsave(&fcloop_lock, flags); + list_del(&nport->nport_list); + spin_unlock_irqrestore(&fcloop_lock, flags); + + if (nport->lport) + fcloop_lport_put(nport->lport); + + kfree(nport); } static int fcloop_nport_get(struct fcloop_nport *nport) { - return kref_get_unless_zero(&nport->ref); + return refcount_inc_not_zero(&nport->ref); } static void @@ -1018,26 +1092,49 @@ fcloop_localport_delete(struct nvme_fc_local_port *localport) struct fcloop_lport_priv *lport_priv = localport->private; struct fcloop_lport *lport = lport_priv->lport; - /* release any threads waiting for the unreg to complete */ - complete(&lport->unreg_done); + fcloop_lport_put(lport); } static void fcloop_remoteport_delete(struct nvme_fc_remote_port *remoteport) { struct fcloop_rport *rport = remoteport->private; + bool put_port = false; + unsigned long flags; flush_work(&rport->ls_work); - fcloop_nport_put(rport->nport); + + spin_lock_irqsave(&fcloop_lock, flags); + if (!test_and_set_bit(PORT_DELETED, &rport->flags)) + put_port = true; + rport->nport->rport = NULL; + spin_unlock_irqrestore(&fcloop_lock, flags); + + if (put_port) { + WARN_ON(!list_empty(&rport->ls_list)); + fcloop_nport_put(rport->nport); + } } static void fcloop_targetport_delete(struct nvmet_fc_target_port *targetport) { struct fcloop_tport *tport = targetport->private; + bool put_port = false; + unsigned long flags; flush_work(&tport->ls_work); - fcloop_nport_put(tport->nport); + + spin_lock_irqsave(&fcloop_lock, flags); + if (!test_and_set_bit(PORT_DELETED, &tport->flags)) + put_port = true; + tport->nport->tport = NULL; + spin_unlock_irqrestore(&fcloop_lock, flags); + + if (put_port) { + WARN_ON(!list_empty(&tport->ls_list)); + fcloop_nport_put(tport->nport); + } } #define FCLOOP_HW_QUEUES 4 @@ -1061,7 +1158,6 @@ static struct nvme_fc_port_template fctemplate = { /* sizes of additional private data for data structures */ .local_priv_sz = sizeof(struct fcloop_lport_priv), .remote_priv_sz = sizeof(struct fcloop_rport), - .lsrqst_priv_sz = sizeof(struct fcloop_lsreq), .fcprqst_priv_sz = sizeof(struct fcloop_ini_fcpreq), }; @@ -1075,6 +1171,7 @@ static struct nvmet_fc_target_template tgttemplate = { .ls_req = fcloop_t2h_ls_req, .ls_abort = fcloop_t2h_ls_abort, .host_release = fcloop_t2h_host_release, + .host_traddr = fcloop_t2h_host_traddr, .max_hw_queues = FCLOOP_HW_QUEUES, .max_sgl_segments = FCLOOP_SGL_SEGS, .max_dif_sgl_segments = FCLOOP_SGL_SEGS, @@ -1083,7 +1180,6 @@ static struct nvmet_fc_target_template tgttemplate = { .target_features = 0, /* sizes of additional private data for data structures */ .target_priv_sz = sizeof(struct fcloop_tport), - .lsrqst_priv_sz = sizeof(struct fcloop_lsreq), }; static ssize_t @@ -1130,6 +1226,7 @@ fcloop_create_local_port(struct device *dev, struct device_attribute *attr, lport->localport = localport; INIT_LIST_HEAD(&lport->lport_list); + refcount_set(&lport->ref, 1); spin_lock_irqsave(&fcloop_lock, flags); list_add_tail(&lport->lport_list, &fcloop_lports); @@ -1146,59 +1243,94 @@ out_free_lport: return ret ? ret : count; } +static int +__localport_unreg(struct fcloop_lport *lport) +{ + return nvme_fc_unregister_localport(lport->localport); +} -static void -__unlink_local_port(struct fcloop_lport *lport) +static struct fcloop_nport * +__fcloop_nport_lookup(u64 node_name, u64 port_name) { - list_del(&lport->lport_list); + struct fcloop_nport *nport; + + list_for_each_entry(nport, &fcloop_nports, nport_list) { + if (nport->node_name != node_name || + nport->port_name != port_name) + continue; + + if (fcloop_nport_get(nport)) + return nport; + + break; + } + + return NULL; } -static int -__wait_localport_unreg(struct fcloop_lport *lport) +static struct fcloop_nport * +fcloop_nport_lookup(u64 node_name, u64 port_name) { - int ret; + struct fcloop_nport *nport; + unsigned long flags; - init_completion(&lport->unreg_done); + spin_lock_irqsave(&fcloop_lock, flags); + nport = __fcloop_nport_lookup(node_name, port_name); + spin_unlock_irqrestore(&fcloop_lock, flags); - ret = nvme_fc_unregister_localport(lport->localport); + return nport; +} - wait_for_completion(&lport->unreg_done); +static struct fcloop_lport * +__fcloop_lport_lookup(u64 node_name, u64 port_name) +{ + struct fcloop_lport *lport; - kfree(lport); + list_for_each_entry(lport, &fcloop_lports, lport_list) { + if (lport->localport->node_name != node_name || + lport->localport->port_name != port_name) + continue; - return ret; + if (fcloop_lport_get(lport)) + return lport; + + break; + } + + return NULL; } +static struct fcloop_lport * +fcloop_lport_lookup(u64 node_name, u64 port_name) +{ + struct fcloop_lport *lport; + unsigned long flags; + + spin_lock_irqsave(&fcloop_lock, flags); + lport = __fcloop_lport_lookup(node_name, port_name); + spin_unlock_irqrestore(&fcloop_lock, flags); + + return lport; +} static ssize_t fcloop_delete_local_port(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { - struct fcloop_lport *tlport, *lport = NULL; + struct fcloop_lport *lport; u64 nodename, portname; - unsigned long flags; int ret; ret = fcloop_parse_nm_options(dev, &nodename, &portname, buf); if (ret) return ret; - spin_lock_irqsave(&fcloop_lock, flags); - - list_for_each_entry(tlport, &fcloop_lports, lport_list) { - if (tlport->localport->node_name == nodename && - tlport->localport->port_name == portname) { - lport = tlport; - __unlink_local_port(lport); - break; - } - } - spin_unlock_irqrestore(&fcloop_lock, flags); - + lport = fcloop_lport_lookup(nodename, portname); if (!lport) return -ENOENT; - ret = __wait_localport_unreg(lport); + ret = __localport_unreg(lport); + fcloop_lport_put(lport); return ret ? ret : count; } @@ -1206,8 +1338,8 @@ fcloop_delete_local_port(struct device *dev, struct device_attribute *attr, static struct fcloop_nport * fcloop_alloc_nport(const char *buf, size_t count, bool remoteport) { - struct fcloop_nport *newnport, *nport = NULL; - struct fcloop_lport *tmplport, *lport = NULL; + struct fcloop_nport *newnport, *nport; + struct fcloop_lport *lport; struct fcloop_ctrl_options *opts; unsigned long flags; u32 opts_mask = (remoteport) ? RPORT_OPTS : TGTPORT_OPTS; @@ -1222,10 +1354,8 @@ fcloop_alloc_nport(const char *buf, size_t count, bool remoteport) goto out_free_opts; /* everything there ? */ - if ((opts->mask & opts_mask) != opts_mask) { - ret = -EINVAL; + if ((opts->mask & opts_mask) != opts_mask) goto out_free_opts; - } newnport = kzalloc(sizeof(*newnport), GFP_KERNEL); if (!newnport) @@ -1238,63 +1368,64 @@ fcloop_alloc_nport(const char *buf, size_t count, bool remoteport) newnport->port_role = opts->roles; if (opts->mask & NVMF_OPT_FCADDR) newnport->port_id = opts->fcaddr; - kref_init(&newnport->ref); + refcount_set(&newnport->ref, 1); spin_lock_irqsave(&fcloop_lock, flags); - - list_for_each_entry(tmplport, &fcloop_lports, lport_list) { - if (tmplport->localport->node_name == opts->wwnn && - tmplport->localport->port_name == opts->wwpn) - goto out_invalid_opts; - - if (tmplport->localport->node_name == opts->lpwwnn && - tmplport->localport->port_name == opts->lpwwpn) - lport = tmplport; + lport = __fcloop_lport_lookup(opts->wwnn, opts->wwpn); + if (lport) { + /* invalid configuration */ + fcloop_lport_put(lport); + goto out_free_newnport; } if (remoteport) { - if (!lport) - goto out_invalid_opts; - newnport->lport = lport; - } - - list_for_each_entry(nport, &fcloop_nports, nport_list) { - if (nport->node_name == opts->wwnn && - nport->port_name == opts->wwpn) { - if ((remoteport && nport->rport) || - (!remoteport && nport->tport)) { - nport = NULL; - goto out_invalid_opts; - } - - fcloop_nport_get(nport); - - spin_unlock_irqrestore(&fcloop_lock, flags); - - if (remoteport) - nport->lport = lport; - if (opts->mask & NVMF_OPT_ROLES) - nport->port_role = opts->roles; - if (opts->mask & NVMF_OPT_FCADDR) - nport->port_id = opts->fcaddr; + lport = __fcloop_lport_lookup(opts->lpwwnn, opts->lpwwpn); + if (!lport) { + /* invalid configuration */ goto out_free_newnport; } } - list_add_tail(&newnport->nport_list, &fcloop_nports); + nport = __fcloop_nport_lookup(opts->wwnn, opts->wwpn); + if (nport) { + if ((remoteport && nport->rport) || + (!remoteport && nport->tport)) { + /* invalid configuration */ + goto out_put_nport; + } + + /* found existing nport, discard the new nport */ + kfree(newnport); + } else { + list_add_tail(&newnport->nport_list, &fcloop_nports); + nport = newnport; + } + if (opts->mask & NVMF_OPT_ROLES) + nport->port_role = opts->roles; + if (opts->mask & NVMF_OPT_FCADDR) + nport->port_id = opts->fcaddr; + if (lport) { + if (!nport->lport) + nport->lport = lport; + else + fcloop_lport_put(lport); + } spin_unlock_irqrestore(&fcloop_lock, flags); kfree(opts); - return newnport; + return nport; -out_invalid_opts: - spin_unlock_irqrestore(&fcloop_lock, flags); +out_put_nport: + if (lport) + fcloop_lport_put(lport); + fcloop_nport_put(nport); out_free_newnport: + spin_unlock_irqrestore(&fcloop_lock, flags); kfree(newnport); out_free_opts: kfree(opts); - return nport; + return NULL; } static ssize_t @@ -1335,6 +1466,7 @@ fcloop_create_remote_port(struct device *dev, struct device_attribute *attr, rport->nport = nport; rport->lport = nport->lport; nport->rport = rport; + rport->flags = 0; spin_lock_init(&rport->lock); INIT_WORK(&rport->ls_work, fcloop_rport_lsrqst_work); INIT_LIST_HEAD(&rport->ls_list); @@ -1348,6 +1480,8 @@ __unlink_remote_port(struct fcloop_nport *nport) { struct fcloop_rport *rport = nport->rport; + lockdep_assert_held(&fcloop_lock); + if (rport && nport->tport) nport->tport->remoteport = NULL; nport->rport = NULL; @@ -1358,9 +1492,6 @@ __unlink_remote_port(struct fcloop_nport *nport) static int __remoteport_unreg(struct fcloop_nport *nport, struct fcloop_rport *rport) { - if (!rport) - return -EALREADY; - return nvme_fc_unregister_remoteport(rport->remoteport); } @@ -1368,8 +1499,8 @@ static ssize_t fcloop_delete_remote_port(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { - struct fcloop_nport *nport = NULL, *tmpport; - static struct fcloop_rport *rport; + struct fcloop_nport *nport; + struct fcloop_rport *rport; u64 nodename, portname; unsigned long flags; int ret; @@ -1378,24 +1509,24 @@ fcloop_delete_remote_port(struct device *dev, struct device_attribute *attr, if (ret) return ret; - spin_lock_irqsave(&fcloop_lock, flags); - - list_for_each_entry(tmpport, &fcloop_nports, nport_list) { - if (tmpport->node_name == nodename && - tmpport->port_name == portname && tmpport->rport) { - nport = tmpport; - rport = __unlink_remote_port(nport); - break; - } - } + nport = fcloop_nport_lookup(nodename, portname); + if (!nport) + return -ENOENT; + spin_lock_irqsave(&fcloop_lock, flags); + rport = __unlink_remote_port(nport); spin_unlock_irqrestore(&fcloop_lock, flags); - if (!nport) - return -ENOENT; + if (!rport) { + ret = -ENOENT; + goto out_nport_put; + } ret = __remoteport_unreg(nport, rport); +out_nport_put: + fcloop_nport_put(nport); + return ret ? ret : count; } @@ -1433,6 +1564,7 @@ fcloop_create_target_port(struct device *dev, struct device_attribute *attr, tport->nport = nport; tport->lport = nport->lport; nport->tport = tport; + tport->flags = 0; spin_lock_init(&tport->lock); INIT_WORK(&tport->ls_work, fcloop_tport_lsrqst_work); INIT_LIST_HEAD(&tport->ls_list); @@ -1446,6 +1578,8 @@ __unlink_target_port(struct fcloop_nport *nport) { struct fcloop_tport *tport = nport->tport; + lockdep_assert_held(&fcloop_lock); + if (tport && nport->rport) nport->rport->targetport = NULL; nport->tport = NULL; @@ -1456,9 +1590,6 @@ __unlink_target_port(struct fcloop_nport *nport) static int __targetport_unreg(struct fcloop_nport *nport, struct fcloop_tport *tport) { - if (!tport) - return -EALREADY; - return nvmet_fc_unregister_targetport(tport->targetport); } @@ -1466,8 +1597,8 @@ static ssize_t fcloop_delete_target_port(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { - struct fcloop_nport *nport = NULL, *tmpport; - struct fcloop_tport *tport = NULL; + struct fcloop_nport *nport; + struct fcloop_tport *tport; u64 nodename, portname; unsigned long flags; int ret; @@ -1476,24 +1607,24 @@ fcloop_delete_target_port(struct device *dev, struct device_attribute *attr, if (ret) return ret; - spin_lock_irqsave(&fcloop_lock, flags); - - list_for_each_entry(tmpport, &fcloop_nports, nport_list) { - if (tmpport->node_name == nodename && - tmpport->port_name == portname && tmpport->tport) { - nport = tmpport; - tport = __unlink_target_port(nport); - break; - } - } + nport = fcloop_nport_lookup(nodename, portname); + if (!nport) + return -ENOENT; + spin_lock_irqsave(&fcloop_lock, flags); + tport = __unlink_target_port(nport); spin_unlock_irqrestore(&fcloop_lock, flags); - if (!nport) - return -ENOENT; + if (!tport) { + ret = -ENOENT; + goto out_nport_put; + } ret = __targetport_unreg(nport, tport); +out_nport_put: + fcloop_nport_put(nport); + return ret ? ret : count; } @@ -1554,23 +1685,29 @@ static const struct attribute_group *fcloop_dev_attr_groups[] = { NULL, }; -static struct class *fcloop_class; +static const struct class fcloop_class = { + .name = "fcloop", +}; static struct device *fcloop_device; - static int __init fcloop_init(void) { int ret; - fcloop_class = class_create(THIS_MODULE, "fcloop"); - if (IS_ERR(fcloop_class)) { + lsreq_cache = kmem_cache_create("lsreq_cache", + sizeof(struct fcloop_lsreq), 0, + 0, NULL); + if (!lsreq_cache) + return -ENOMEM; + + ret = class_register(&fcloop_class); + if (ret) { pr_err("couldn't register class fcloop\n"); - ret = PTR_ERR(fcloop_class); - return ret; + goto out_destroy_cache; } fcloop_device = device_create_with_groups( - fcloop_class, NULL, MKDEV(0, 0), NULL, + &fcloop_class, NULL, MKDEV(0, 0), NULL, fcloop_dev_attr_groups, "ctl"); if (IS_ERR(fcloop_device)) { pr_err("couldn't create ctl device!\n"); @@ -1583,14 +1720,16 @@ static int __init fcloop_init(void) return 0; out_destroy_class: - class_destroy(fcloop_class); + class_unregister(&fcloop_class); +out_destroy_cache: + kmem_cache_destroy(lsreq_cache); return ret; } static void __exit fcloop_exit(void) { - struct fcloop_lport *lport = NULL; - struct fcloop_nport *nport = NULL; + struct fcloop_lport *lport; + struct fcloop_nport *nport; struct fcloop_tport *tport; struct fcloop_rport *rport; unsigned long flags; @@ -1601,7 +1740,7 @@ static void __exit fcloop_exit(void) for (;;) { nport = list_first_entry_or_null(&fcloop_nports, typeof(*nport), nport_list); - if (!nport) + if (!nport || !fcloop_nport_get(nport)) break; tport = __unlink_target_port(nport); @@ -1609,13 +1748,21 @@ static void __exit fcloop_exit(void) spin_unlock_irqrestore(&fcloop_lock, flags); - ret = __targetport_unreg(nport, tport); - if (ret) - pr_warn("%s: Failed deleting target port\n", __func__); + if (tport) { + ret = __targetport_unreg(nport, tport); + if (ret) + pr_warn("%s: Failed deleting target port\n", + __func__); + } - ret = __remoteport_unreg(nport, rport); - if (ret) - pr_warn("%s: Failed deleting remote port\n", __func__); + if (rport) { + ret = __remoteport_unreg(nport, rport); + if (ret) + pr_warn("%s: Failed deleting remote port\n", + __func__); + } + + fcloop_nport_put(nport); spin_lock_irqsave(&fcloop_lock, flags); } @@ -1623,17 +1770,17 @@ static void __exit fcloop_exit(void) for (;;) { lport = list_first_entry_or_null(&fcloop_lports, typeof(*lport), lport_list); - if (!lport) + if (!lport || !fcloop_lport_get(lport)) break; - __unlink_local_port(lport); - spin_unlock_irqrestore(&fcloop_lock, flags); - ret = __wait_localport_unreg(lport); + ret = __localport_unreg(lport); if (ret) pr_warn("%s: Failed deleting local port\n", __func__); + fcloop_lport_put(lport); + spin_lock_irqsave(&fcloop_lock, flags); } @@ -1641,11 +1788,13 @@ static void __exit fcloop_exit(void) put_device(fcloop_device); - device_destroy(fcloop_class, MKDEV(0, 0)); - class_destroy(fcloop_class); + device_destroy(&fcloop_class, MKDEV(0, 0)); + class_unregister(&fcloop_class); + kmem_cache_destroy(lsreq_cache); } module_init(fcloop_init); module_exit(fcloop_exit); +MODULE_DESCRIPTION("NVMe target FC loop transport driver"); MODULE_LICENSE("GPL v2"); diff --git a/drivers/nvme/target/io-cmd-bdev.c b/drivers/nvme/target/io-cmd-bdev.c index c2d6cea0236b..8d246b8ca604 100644 --- a/drivers/nvme/target/io-cmd-bdev.c +++ b/drivers/nvme/target/io-cmd-bdev.c @@ -36,7 +36,7 @@ void nvmet_bdev_set_limits(struct block_device *bdev, struct nvme_id_ns *id) */ id->nsfeat |= 1 << 4; /* NPWG = Namespace Preferred Write Granularity. 0's based */ - id->npwg = lpp0b; + id->npwg = to0based(bdev_io_min(bdev) / bdev_logical_block_size(bdev)); /* NPWA = Namespace Preferred Write Alignment. 0's based */ id->npwa = id->npwg; /* NPDG = Namespace Preferred Deallocate Granularity. 0's based */ @@ -46,13 +46,18 @@ void nvmet_bdev_set_limits(struct block_device *bdev, struct nvme_id_ns *id) id->npda = id->npdg; /* NOWS = Namespace Optimal Write Size */ id->nows = to0based(bdev_io_opt(bdev) / bdev_logical_block_size(bdev)); + + /* Set WZDS and DRB if device supports unmapped write zeroes */ + if (bdev_write_zeroes_unmap_sectors(bdev)) + id->dlfeat = (1 << 3) | 0x1; } void nvmet_bdev_ns_disable(struct nvmet_ns *ns) { - if (ns->bdev) { - blkdev_put(ns->bdev, FMODE_WRITE | FMODE_READ); + if (ns->bdev_file) { + fput(ns->bdev_file); ns->bdev = NULL; + ns->bdev_file = NULL; } } @@ -60,15 +65,17 @@ static void nvmet_bdev_ns_enable_integrity(struct nvmet_ns *ns) { struct blk_integrity *bi = bdev_get_integrity(ns->bdev); - if (bi) { - ns->metadata_size = bi->tuple_size; - if (bi->profile == &t10_pi_type1_crc) + if (!bi) + return; + + if (bi->csum_type == BLK_INTEGRITY_CSUM_CRC) { + ns->metadata_size = bi->metadata_size; + if (bi->flags & BLK_INTEGRITY_REF_TAG) ns->pi_type = NVME_NS_DPS_PI_TYPE1; - else if (bi->profile == &t10_pi_type3_crc) - ns->pi_type = NVME_NS_DPS_PI_TYPE3; else - /* Unsupported metadata type */ - ns->metadata_size = 0; + ns->pi_type = NVME_NS_DPS_PI_TYPE3; + } else { + ns->metadata_size = 0; } } @@ -84,23 +91,24 @@ int nvmet_bdev_ns_enable(struct nvmet_ns *ns) if (ns->buffered_io) return -ENOTBLK; - ns->bdev = blkdev_get_by_path(ns->device_path, - FMODE_READ | FMODE_WRITE, NULL); - if (IS_ERR(ns->bdev)) { - ret = PTR_ERR(ns->bdev); + ns->bdev_file = bdev_file_open_by_path(ns->device_path, + BLK_OPEN_READ | BLK_OPEN_WRITE, NULL, NULL); + if (IS_ERR(ns->bdev_file)) { + ret = PTR_ERR(ns->bdev_file); if (ret != -ENOTBLK) { - pr_err("failed to open block device %s: (%ld)\n", - ns->device_path, PTR_ERR(ns->bdev)); + pr_err("failed to open block device %s: (%d)\n", + ns->device_path, ret); } - ns->bdev = NULL; + ns->bdev_file = NULL; return ret; } + ns->bdev = file_bdev(ns->bdev_file); ns->size = bdev_nr_bytes(ns->bdev); ns->blksize_shift = blksize_bits(bdev_logical_block_size(ns->bdev)); ns->pi_type = 0; ns->metadata_size = 0; - if (IS_ENABLED(CONFIG_BLK_DEV_INTEGRITY_T10)) + if (IS_ENABLED(CONFIG_BLK_DEV_INTEGRITY)) nvmet_bdev_ns_enable_integrity(ns); if (bdev_is_zoned(ns->bdev)) { @@ -129,27 +137,20 @@ u16 blk_to_nvme_status(struct nvmet_req *req, blk_status_t blk_sts) * Right now there exists M : 1 mapping between block layer error * to the NVMe status code (see nvme_error_status()). For consistency, * when we reverse map we use most appropriate NVMe Status code from - * the group of the NVMe staus codes used in the nvme_error_status(). + * the group of the NVMe status codes used in the nvme_error_status(). */ switch (blk_sts) { case BLK_STS_NOSPC: - status = NVME_SC_CAP_EXCEEDED | NVME_SC_DNR; + status = NVME_SC_CAP_EXCEEDED | NVME_STATUS_DNR; req->error_loc = offsetof(struct nvme_rw_command, length); break; case BLK_STS_TARGET: - status = NVME_SC_LBA_RANGE | NVME_SC_DNR; + status = NVME_SC_LBA_RANGE | NVME_STATUS_DNR; req->error_loc = offsetof(struct nvme_rw_command, slba); break; case BLK_STS_NOTSUPP: + status = NVME_SC_INVALID_OPCODE | NVME_STATUS_DNR; req->error_loc = offsetof(struct nvme_common_command, opcode); - switch (req->cmd->common.opcode) { - case nvme_cmd_dsm: - case nvme_cmd_write_zeroes: - status = NVME_SC_ONCS_NOT_SUPPORTED | NVME_SC_DNR; - break; - default: - status = NVME_SC_INVALID_OPCODE | NVME_SC_DNR; - } break; case BLK_STS_MEDIUM: status = NVME_SC_ACCESS_DENIED; @@ -157,7 +158,7 @@ u16 blk_to_nvme_status(struct nvmet_req *req, blk_status_t blk_sts) break; case BLK_STS_IOERR: default: - status = NVME_SC_INTERNAL | NVME_SC_DNR; + status = NVME_SC_INTERNAL | NVME_STATUS_DNR; req->error_loc = offsetof(struct nvme_common_command, opcode); } @@ -206,12 +207,11 @@ static int nvmet_bdev_alloc_bip(struct nvmet_req *req, struct bio *bio, return PTR_ERR(bip); } - bip->bip_iter.bi_size = bio_integrity_bytes(bi, bio_sectors(bio)); /* virtual start sector must be in integrity interval units */ bip_set_seed(bip, bio->bi_iter.bi_sector >> (bi->interval_exp - SECTOR_SHIFT)); - resid = bip->bip_iter.bi_size; + resid = bio_integrity_bytes(bi, bio_sectors(bio)); while (resid > 0 && sg_miter_next(miter)) { len = min_t(size_t, miter->length, resid); rc = bio_integrity_add_page(bio, miter->page, len, @@ -269,6 +269,9 @@ static void nvmet_bdev_execute_rw(struct nvmet_req *req) iter_flags = SG_MITER_FROM_SG; } + if (req->cmd->rw.control & cpu_to_le16(NVME_RW_LR)) + opf |= REQ_FAILFAST_DEV; + if (is_pci_p2pdma_page(sg_page(req->sg))) opf |= REQ_NOMERGE; @@ -355,7 +358,7 @@ u16 nvmet_bdev_flush(struct nvmet_req *req) return 0; if (blkdev_issue_flush(req->ns->bdev)) - return NVME_SC_INTERNAL | NVME_SC_DNR; + return NVME_SC_INTERNAL | NVME_STATUS_DNR; return 0; } diff --git a/drivers/nvme/target/io-cmd-file.c b/drivers/nvme/target/io-cmd-file.c index 871c4f32f443..2d068439b129 100644 --- a/drivers/nvme/target/io-cmd-file.c +++ b/drivers/nvme/target/io-cmd-file.c @@ -73,13 +73,6 @@ err: return ret; } -static void nvmet_file_init_bvec(struct bio_vec *bv, struct scatterlist *sg) -{ - bv->bv_page = sg_page(sg); - bv->bv_offset = sg->offset; - bv->bv_len = sg->length; -} - static ssize_t nvmet_file_submit_bvec(struct nvmet_req *req, loff_t pos, unsigned long nr_segs, size_t count, int ki_flags) { @@ -146,7 +139,8 @@ static bool nvmet_file_execute_io(struct nvmet_req *req, int ki_flags) memset(&req->f.iocb, 0, sizeof(struct kiocb)); for_each_sg(req->sg, sg, req->sg_cnt, i) { - nvmet_file_init_bvec(&req->f.bvec[bv_cnt], sg); + bvec_set_page(&req->f.bvec[bv_cnt], sg_page(sg), sg->length, + sg->offset); len += req->f.bvec[bv_cnt].bv_len; total_len += req->f.bvec[bv_cnt].bv_len; bv_cnt++; diff --git a/drivers/nvme/target/loop.c b/drivers/nvme/target/loop.c index f2d24b2d992f..fc8e7c9ad858 100644 --- a/drivers/nvme/target/loop.c +++ b/drivers/nvme/target/loop.c @@ -33,10 +33,12 @@ struct nvme_loop_ctrl { struct list_head list; struct blk_mq_tag_set tag_set; - struct nvme_loop_iod async_event_iod; struct nvme_ctrl ctrl; struct nvmet_port *port; + + /* Must be last --ends in a flexible-array member. */ + struct nvme_loop_iod async_event_iod; }; static inline struct nvme_loop_ctrl *to_loop_ctrl(struct nvme_ctrl *ctrl) @@ -148,8 +150,7 @@ static blk_status_t nvme_loop_queue_rq(struct blk_mq_hw_ctx *hctx, nvme_start_request(req); iod->cmd.common.flags |= NVME_CMD_SGL_METABUF; iod->req.port = queue->ctrl->port; - if (!nvmet_req_init(&iod->req, &queue->nvme_cq, - &queue->nvme_sq, &nvme_loop_ops)) + if (!nvmet_req_init(&iod->req, &queue->nvme_sq, &nvme_loop_ops)) return BLK_STS_OK; if (blk_rq_nr_phys_segments(req)) { @@ -162,7 +163,7 @@ static blk_status_t nvme_loop_queue_rq(struct blk_mq_hw_ctx *hctx, } iod->req.sg = iod->sg_table.sgl; - iod->req.sg_cnt = blk_rq_map_sg(req->q, req, iod->sg_table.sgl); + iod->req.sg_cnt = blk_rq_map_sg(req, iod->sg_table.sgl); iod->req.transfer_len = blk_rq_payload_bytes(req); } @@ -181,8 +182,7 @@ static void nvme_loop_submit_async_event(struct nvme_ctrl *arg) iod->cmd.common.command_id = NVME_AQ_BLK_MQ_DEPTH; iod->cmd.common.flags |= NVME_CMD_SGL_METABUF; - if (!nvmet_req_init(&iod->req, &queue->nvme_cq, &queue->nvme_sq, - &nvme_loop_ops)) { + if (!nvmet_req_init(&iod->req, &queue->nvme_sq, &nvme_loop_ops)) { dev_err(ctrl->ctrl.device, "failed async event work\n"); return; } @@ -265,7 +265,15 @@ static void nvme_loop_destroy_admin_queue(struct nvme_loop_ctrl *ctrl) { if (!test_and_clear_bit(NVME_LOOP_Q_LIVE, &ctrl->queues[0].flags)) return; + /* + * It's possible that some requests might have been added + * after admin queue is stopped/quiesced. So now start the + * queue to flush these requests to the completion. + */ + nvme_unquiesce_admin_queue(&ctrl->ctrl); + nvmet_sq_destroy(&ctrl->queues[0].nvme_sq); + nvmet_cq_put(&ctrl->queues[0].nvme_cq); nvme_remove_admin_tag_set(&ctrl->ctrl); } @@ -295,8 +303,15 @@ static void nvme_loop_destroy_io_queues(struct nvme_loop_ctrl *ctrl) for (i = 1; i < ctrl->ctrl.queue_count; i++) { clear_bit(NVME_LOOP_Q_LIVE, &ctrl->queues[i].flags); nvmet_sq_destroy(&ctrl->queues[i].nvme_sq); + nvmet_cq_put(&ctrl->queues[i].nvme_cq); } ctrl->ctrl.queue_count = 1; + /* + * It's possible that some requests might have been added + * after io queue is stopped/quiesced. So now start the + * queue to flush these requests to the completion. + */ + nvme_unquiesce_io_queues(&ctrl->ctrl); } static int nvme_loop_init_io_queues(struct nvme_loop_ctrl *ctrl) @@ -314,9 +329,13 @@ static int nvme_loop_init_io_queues(struct nvme_loop_ctrl *ctrl) for (i = 1; i <= nr_io_queues; i++) { ctrl->queues[i].ctrl = ctrl; - ret = nvmet_sq_init(&ctrl->queues[i].nvme_sq); - if (ret) + nvmet_cq_init(&ctrl->queues[i].nvme_cq); + ret = nvmet_sq_init(&ctrl->queues[i].nvme_sq, + &ctrl->queues[i].nvme_cq); + if (ret) { + nvmet_cq_put(&ctrl->queues[i].nvme_cq); goto out_destroy_queues; + } ctrl->ctrl.queue_count++; } @@ -347,9 +366,13 @@ static int nvme_loop_configure_admin_queue(struct nvme_loop_ctrl *ctrl) int error; ctrl->queues[0].ctrl = ctrl; - error = nvmet_sq_init(&ctrl->queues[0].nvme_sq); - if (error) + nvmet_cq_init(&ctrl->queues[0].nvme_cq); + error = nvmet_sq_init(&ctrl->queues[0].nvme_sq, + &ctrl->queues[0].nvme_cq); + if (error) { + nvmet_cq_put(&ctrl->queues[0].nvme_cq); return error; + } ctrl->ctrl.queue_count = 1; error = nvme_alloc_admin_tag_set(&ctrl->ctrl, &ctrl->admin_tag_set, @@ -373,7 +396,7 @@ static int nvme_loop_configure_admin_queue(struct nvme_loop_ctrl *ctrl) goto out_cleanup_tagset; ctrl->ctrl.max_hw_sectors = - (NVME_LOOP_MAX_SEGMENTS - 1) << (PAGE_SHIFT - 9); + (NVME_LOOP_MAX_SEGMENTS - 1) << PAGE_SECTORS_SHIFT; nvme_unquiesce_admin_queue(&ctrl->ctrl); @@ -388,6 +411,7 @@ out_cleanup_tagset: nvme_remove_admin_tag_set(&ctrl->ctrl); out_free_sq: nvmet_sq_destroy(&ctrl->queues[0].nvme_sq); + nvmet_cq_put(&ctrl->queues[0].nvme_cq); return error; } @@ -400,7 +424,7 @@ static void nvme_loop_shutdown_ctrl(struct nvme_loop_ctrl *ctrl) } nvme_quiesce_admin_queue(&ctrl->ctrl); - if (ctrl->ctrl.state == NVME_CTRL_LIVE) + if (nvme_ctrl_state(&ctrl->ctrl) == NVME_CTRL_LIVE) nvme_disable_ctrl(&ctrl->ctrl, true); nvme_cancel_admin_tagset(&ctrl->ctrl); @@ -434,8 +458,10 @@ static void nvme_loop_reset_ctrl_work(struct work_struct *work) nvme_loop_shutdown_ctrl(ctrl); if (!nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_CONNECTING)) { - if (ctrl->ctrl.state != NVME_CTRL_DELETING && - ctrl->ctrl.state != NVME_CTRL_DELETING_NOIO) + enum nvme_ctrl_state state = nvme_ctrl_state(&ctrl->ctrl); + + if (state != NVME_CTRL_DELETING && + state != NVME_CTRL_DELETING_NOIO) /* state change failure for non-deleted ctrl? */ WARN_ON_ONCE(1); return; @@ -466,6 +492,8 @@ static void nvme_loop_reset_ctrl_work(struct work_struct *work) out_destroy_io: nvme_loop_destroy_io_queues(ctrl); out_destroy_admin: + nvme_quiesce_admin_queue(&ctrl->ctrl); + nvme_cancel_admin_tagset(&ctrl->ctrl); nvme_loop_destroy_admin_queue(ctrl); out_disable: dev_warn(ctrl->ctrl.device, "Removing after reset failure\n"); @@ -483,6 +511,7 @@ static const struct nvme_ctrl_ops nvme_loop_ctrl_ops = { .submit_async_event = nvme_loop_submit_async_event, .delete_ctrl = nvme_loop_delete_ctrl_host, .get_address = nvmf_get_address, + .get_virt_boundary = nvme_get_virt_boundary, }; static int nvme_loop_create_io_queues(struct nvme_loop_ctrl *ctrl) @@ -551,6 +580,10 @@ static struct nvme_ctrl *nvme_loop_create_ctrl(struct device *dev, goto out; } + ret = nvme_add_ctrl(&ctrl->ctrl); + if (ret) + goto out_put_ctrl; + if (!nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_CONNECTING)) WARN_ON_ONCE(1); @@ -600,11 +633,14 @@ static struct nvme_ctrl *nvme_loop_create_ctrl(struct device *dev, return &ctrl->ctrl; out_remove_admin_queue: + nvme_quiesce_admin_queue(&ctrl->ctrl); + nvme_cancel_admin_tagset(&ctrl->ctrl); nvme_loop_destroy_admin_queue(ctrl); out_free_queues: kfree(ctrl->queues); out_uninit_ctrl: nvme_uninit_ctrl(&ctrl->ctrl); +out_put_ctrl: nvme_put_ctrl(&ctrl->ctrl); out: if (ret > 0) @@ -684,5 +720,6 @@ static void __exit nvme_loop_cleanup_module(void) module_init(nvme_loop_init_module); module_exit(nvme_loop_cleanup_module); +MODULE_DESCRIPTION("NVMe target loop transport driver"); MODULE_LICENSE("GPL v2"); MODULE_ALIAS("nvmet-transport-254"); /* 254 == NVMF_TRTYPE_LOOP */ diff --git a/drivers/nvme/target/nvmet.h b/drivers/nvme/target/nvmet.h index 89bedfcd974c..b664b584fdc8 100644 --- a/drivers/nvme/target/nvmet.h +++ b/drivers/nvme/target/nvmet.h @@ -20,9 +20,11 @@ #include <linux/blkdev.h> #include <linux/radix-tree.h> #include <linux/t10-pi.h> +#include <linux/kfifo.h> -#define NVMET_DEFAULT_VS NVME_VS(1, 3, 0) +#define NVMET_DEFAULT_VS NVME_VS(2, 1, 0) +#define NVMET_NS_ENABLED XA_MARK_1 #define NVMET_ASYNC_EVENTS 4 #define NVMET_ERROR_LOG_SLOTS 128 #define NVMET_NO_ERROR_LOC ((u16)-1) @@ -30,6 +32,13 @@ #define NVMET_MN_MAX_SIZE 40 #define NVMET_SN_MAX_SIZE 20 #define NVMET_FR_MAX_SIZE 8 +#define NVMET_PR_LOG_QUEUE_SIZE 64 + +#define nvmet_for_each_ns(xa, index, entry) \ + xa_for_each(xa, index, entry) + +#define nvmet_for_each_enabled_ns(xa, index, entry) \ + xa_for_each_marked(xa, index, entry, NVMET_NS_ENABLED) /* * Supported optional AENs: @@ -56,8 +65,41 @@ #define IPO_IATTR_CONNECT_SQE(x) \ (cpu_to_le32(offsetof(struct nvmf_connect_command, x))) +struct nvmet_pr_registrant { + u64 rkey; + uuid_t hostid; + enum nvme_pr_type rtype; + struct list_head entry; + struct rcu_head rcu; +}; + +struct nvmet_pr { + bool enable; + unsigned long notify_mask; + atomic_t generation; + struct nvmet_pr_registrant __rcu *holder; + /* + * During the execution of the reservation command, mutual + * exclusion is required throughout the process. However, + * while waiting asynchronously for the 'per controller + * percpu_ref' to complete before the 'preempt and abort' + * command finishes, a semaphore is needed to ensure mutual + * exclusion instead of a mutex. + */ + struct semaphore pr_sem; + struct list_head registrant_list; +}; + +struct nvmet_pr_per_ctrl_ref { + struct percpu_ref ref; + struct completion free_done; + struct completion confirm_done; + uuid_t hostid; +}; + struct nvmet_ns { struct percpu_ref ref; + struct file *bdev_file; struct block_device *bdev; struct file *file; bool readonly; @@ -79,11 +121,13 @@ struct nvmet_ns { struct completion disable_done; mempool_t *bvec_pool; - int use_p2pmem; struct pci_dev *p2p_dev; + int use_p2pmem; int pi_type; int metadata_size; u8 csi; + struct nvmet_pr pr; + struct xarray pr_per_ctrl_refs; }; static inline struct nvmet_ns *to_nvmet_ns(struct config_item *item) @@ -97,23 +141,27 @@ static inline struct device *nvmet_ns_dev(struct nvmet_ns *ns) } struct nvmet_cq { + struct nvmet_ctrl *ctrl; u16 qid; u16 size; + refcount_t ref; }; struct nvmet_sq { struct nvmet_ctrl *ctrl; struct percpu_ref ref; + struct nvmet_cq *cq; u16 qid; u16 size; u32 sqhd; bool sqhd_disabled; #ifdef CONFIG_NVME_TARGET_AUTH - struct delayed_work auth_expired_work; bool authenticated; + struct delayed_work auth_expired_work; u16 dhchap_tid; - u16 dhchap_status; - int dhchap_step; + u8 sc_c; + u8 dhchap_status; + u8 dhchap_step; u8 *dhchap_c1; u8 *dhchap_c2; u32 dhchap_s1; @@ -121,6 +169,9 @@ struct nvmet_sq { u8 *dhchap_skey; int dhchap_skey_len; #endif +#ifdef CONFIG_NVME_TARGET_TCP_TLS + struct key *tls_key; +#endif struct completion free_done; struct completion confirm_done; }; @@ -158,9 +209,11 @@ struct nvmet_port { struct config_group ana_groups_group; struct nvmet_ana_group ana_default_group; enum nvme_ana_state *ana_state; + struct key *keyring; void *priv; bool enabled; int inline_data_size; + int max_queue_size; const struct nvmet_fabrics_ops *tr_ops; bool pi_enable; }; @@ -178,9 +231,29 @@ static inline struct nvmet_port *ana_groups_to_port( ana_groups_group); } +static inline u8 nvmet_port_disc_addr_treq_secure_channel(struct nvmet_port *port) +{ + return (port->disc_addr.treq & NVME_TREQ_SECURE_CHANNEL_MASK); +} + +static inline bool nvmet_port_secure_channel_required(struct nvmet_port *port) +{ + return nvmet_port_disc_addr_treq_secure_channel(port) == NVMF_TREQ_REQUIRED; +} + +struct nvmet_pr_log_mgr { + struct mutex lock; + u64 lost_count; + u64 counter; + DECLARE_KFIFO(log_queue, struct nvme_pr_log, NVMET_PR_LOG_QUEUE_SIZE); +}; + struct nvmet_ctrl { struct nvmet_subsys *subsys; struct nvmet_sq **sqs; + struct nvmet_cq **cqs; + + void *drvdata; bool reset_tbkas; @@ -212,16 +285,18 @@ struct nvmet_ctrl { __le32 *changed_ns_list; u32 nr_changed_ns; - char subsysnqn[NVMF_NQN_FIELD_LEN]; char hostnqn[NVMF_NQN_FIELD_LEN]; struct device *p2p_client; struct radix_tree_root p2p_ns_map; - +#ifdef CONFIG_NVME_TARGET_DEBUGFS + struct dentry *debugfs_dir; +#endif spinlock_t error_lock; u64 err_counter; struct nvme_error_slot slots[NVMET_ERROR_LOG_SLOTS]; bool pi_support; + bool concat; #ifdef CONFIG_NVME_TARGET_AUTH struct nvme_dhchap_key *host_key; struct nvme_dhchap_key *ctrl_key; @@ -231,6 +306,10 @@ struct nvmet_ctrl { u8 *dh_key; size_t dh_keysize; #endif +#ifdef CONFIG_NVME_TARGET_TCP_TLS + struct key *tls_key; +#endif + struct nvmet_pr_log_mgr pr_log_mgr; }; struct nvmet_subsys { @@ -249,7 +328,9 @@ struct nvmet_subsys { struct list_head hosts; bool allow_any_host; - +#ifdef CONFIG_NVME_TARGET_DEBUGFS + struct dentry *debugfs_dir; +#endif u16 max_qid; u64 ver; @@ -263,6 +344,8 @@ struct nvmet_subsys { struct config_group namespaces_group; struct config_group allowed_hosts_group; + u16 vendor_id; + u16 subsys_vendor_id; char *model_number; u32 ieee_oui; char *firmware_rev; @@ -337,10 +420,24 @@ struct nvmet_fabrics_ops { void (*delete_ctrl)(struct nvmet_ctrl *ctrl); void (*disc_traddr)(struct nvmet_req *req, struct nvmet_port *port, char *traddr); + ssize_t (*host_traddr)(struct nvmet_ctrl *ctrl, + char *traddr, size_t traddr_len); u16 (*install_queue)(struct nvmet_sq *nvme_sq); void (*discovery_chg)(struct nvmet_port *port); u8 (*get_mdts)(const struct nvmet_ctrl *ctrl); u16 (*get_max_queue_size)(const struct nvmet_ctrl *ctrl); + + /* Operations mandatory for PCI target controllers */ + u16 (*create_sq)(struct nvmet_ctrl *ctrl, u16 sqid, u16 cqid, u16 flags, + u16 qsize, u64 prp1); + u16 (*delete_sq)(struct nvmet_ctrl *ctrl, u16 sqid); + u16 (*create_cq)(struct nvmet_ctrl *ctrl, u16 cqid, u16 flags, + u16 qsize, u64 prp1, u16 irq_vector); + u16 (*delete_cq)(struct nvmet_ctrl *ctrl, u16 cqid); + u16 (*set_feature)(const struct nvmet_ctrl *ctrl, u8 feat, + void *feat_data); + u16 (*get_feature)(const struct nvmet_ctrl *ctrl, u8 feat, + void *feat_data); }; #define NVMET_MAX_INLINE_BIOVEC 8 @@ -377,6 +474,9 @@ struct nvmet_req { struct work_struct zmgmt_work; } z; #endif /* CONFIG_BLK_DEV_ZONED */ + struct { + struct work_struct abort_work; + } r; }; int sg_cnt; int metadata_sg_cnt; @@ -393,6 +493,7 @@ struct nvmet_req { struct device *p2p_client; u16 error_loc; u64 error_slba; + struct nvmet_pr_per_ctrl_ref *pc_ref; }; #define NVMET_MAX_MPOOL_BVEC 16 @@ -446,18 +547,24 @@ void nvmet_start_keep_alive_timer(struct nvmet_ctrl *ctrl); void nvmet_stop_keep_alive_timer(struct nvmet_ctrl *ctrl); u16 nvmet_parse_connect_cmd(struct nvmet_req *req); +u32 nvmet_connect_cmd_data_len(struct nvmet_req *req); void nvmet_bdev_set_limits(struct block_device *bdev, struct nvme_id_ns *id); u16 nvmet_bdev_parse_io_cmd(struct nvmet_req *req); u16 nvmet_file_parse_io_cmd(struct nvmet_req *req); u16 nvmet_bdev_zns_parse_io_cmd(struct nvmet_req *req); +u32 nvmet_admin_cmd_data_len(struct nvmet_req *req); u16 nvmet_parse_admin_cmd(struct nvmet_req *req); +u32 nvmet_discovery_cmd_data_len(struct nvmet_req *req); u16 nvmet_parse_discovery_cmd(struct nvmet_req *req); u16 nvmet_parse_fabrics_admin_cmd(struct nvmet_req *req); +u32 nvmet_fabrics_admin_cmd_data_len(struct nvmet_req *req); u16 nvmet_parse_fabrics_io_cmd(struct nvmet_req *req); +u32 nvmet_fabrics_io_cmd_data_len(struct nvmet_req *req); -bool nvmet_req_init(struct nvmet_req *req, struct nvmet_cq *cq, - struct nvmet_sq *sq, const struct nvmet_fabrics_ops *ops); +bool nvmet_req_init(struct nvmet_req *req, struct nvmet_sq *sq, + const struct nvmet_fabrics_ops *ops); void nvmet_req_uninit(struct nvmet_req *req); +size_t nvmet_req_transfer_len(struct nvmet_req *req); bool nvmet_check_transfer_len(struct nvmet_req *req, size_t len); bool nvmet_check_data_len_lte(struct nvmet_req *req, size_t data_len); void nvmet_req_complete(struct nvmet_req *req, u16 status); @@ -468,23 +575,51 @@ void nvmet_execute_set_features(struct nvmet_req *req); void nvmet_execute_get_features(struct nvmet_req *req); void nvmet_execute_keep_alive(struct nvmet_req *req); +u16 nvmet_check_cqid(struct nvmet_ctrl *ctrl, u16 cqid, bool create); +u16 nvmet_check_io_cqid(struct nvmet_ctrl *ctrl, u16 cqid, bool create); +void nvmet_cq_init(struct nvmet_cq *cq); void nvmet_cq_setup(struct nvmet_ctrl *ctrl, struct nvmet_cq *cq, u16 qid, u16 size); +u16 nvmet_cq_create(struct nvmet_ctrl *ctrl, struct nvmet_cq *cq, u16 qid, + u16 size); +void nvmet_cq_destroy(struct nvmet_cq *cq); +bool nvmet_cq_get(struct nvmet_cq *cq); +void nvmet_cq_put(struct nvmet_cq *cq); +bool nvmet_cq_in_use(struct nvmet_cq *cq); +u16 nvmet_check_sqid(struct nvmet_ctrl *ctrl, u16 sqid, bool create); void nvmet_sq_setup(struct nvmet_ctrl *ctrl, struct nvmet_sq *sq, u16 qid, u16 size); +u16 nvmet_sq_create(struct nvmet_ctrl *ctrl, struct nvmet_sq *sq, + struct nvmet_cq *cq, u16 qid, u16 size); void nvmet_sq_destroy(struct nvmet_sq *sq); -int nvmet_sq_init(struct nvmet_sq *sq); +int nvmet_sq_init(struct nvmet_sq *sq, struct nvmet_cq *cq); void nvmet_ctrl_fatal_error(struct nvmet_ctrl *ctrl); void nvmet_update_cc(struct nvmet_ctrl *ctrl, u32 new); -u16 nvmet_alloc_ctrl(const char *subsysnqn, const char *hostnqn, - struct nvmet_req *req, u32 kato, struct nvmet_ctrl **ctrlp); + +struct nvmet_alloc_ctrl_args { + struct nvmet_port *port; + struct nvmet_sq *sq; + char *subsysnqn; + char *hostnqn; + uuid_t *hostid; + const struct nvmet_fabrics_ops *ops; + struct device *p2p_client; + u32 kato; + __le32 result; + u16 error_loc; + u16 status; +}; + +struct nvmet_ctrl *nvmet_alloc_ctrl(struct nvmet_alloc_ctrl_args *args); struct nvmet_ctrl *nvmet_ctrl_find_get(const char *subsysnqn, const char *hostnqn, u16 cntlid, struct nvmet_req *req); void nvmet_ctrl_put(struct nvmet_ctrl *ctrl); u16 nvmet_check_ctrl_status(struct nvmet_req *req); +ssize_t nvmet_ctrl_host_traddr(struct nvmet_ctrl *ctrl, + char *traddr, size_t traddr_len); struct nvmet_subsys *nvmet_subsys_alloc(const char *subsysnqn, enum nvme_subsys_type type); @@ -531,9 +666,10 @@ void nvmet_subsys_disc_changed(struct nvmet_subsys *subsys, void nvmet_add_async_event(struct nvmet_ctrl *ctrl, u8 event_type, u8 event_info, u8 log_page); -#define NVMET_QUEUE_SIZE 1024 +#define NVMET_MIN_QUEUE_SIZE 16 +#define NVMET_MAX_QUEUE_SIZE 1024 #define NVMET_NR_QUEUES 128 -#define NVMET_MAX_CMD NVMET_QUEUE_SIZE +#define NVMET_MAX_CMD(ctrl) (NVME_CAP_MQES(ctrl->cap) + 1) /* * Nice round number that makes a list of nsids fit into a page. @@ -581,8 +717,8 @@ bool nvmet_ns_revalidate(struct nvmet_ns *ns); u16 blk_to_nvme_status(struct nvmet_req *req, blk_status_t blk_sts); bool nvmet_bdev_zns_enable(struct nvmet_ns *ns); -void nvmet_execute_identify_cns_cs_ctrl(struct nvmet_req *req); -void nvmet_execute_identify_cns_cs_ns(struct nvmet_req *req); +void nvmet_execute_identify_ctrl_zns(struct nvmet_req *req); +void nvmet_execute_identify_ns_zns(struct nvmet_req *req); void nvmet_bdev_execute_zone_mgmt_recv(struct nvmet_req *req); void nvmet_bdev_execute_zone_mgmt_send(struct nvmet_req *req); void nvmet_bdev_execute_zone_append(struct nvmet_req *req); @@ -617,6 +753,11 @@ static inline bool nvmet_is_disc_subsys(struct nvmet_subsys *subsys) return subsys->type != NVME_NQN_NVME; } +static inline bool nvmet_is_pci_ctrl(struct nvmet_ctrl *ctrl) +{ + return ctrl->port->disc_addr.trtype == NVMF_TRTYPE_PCI; +} + #ifdef CONFIG_NVME_TARGET_PASSTHRU void nvmet_passthru_subsys_free(struct nvmet_subsys *subsys); int nvmet_passthru_ctrl_enable(struct nvmet_subsys *subsys); @@ -658,10 +799,45 @@ void nvmet_passthrough_override_cap(struct nvmet_ctrl *ctrl); u16 errno_to_nvme_status(struct nvmet_req *req, int errno); u16 nvmet_report_invalid_opcode(struct nvmet_req *req); +static inline bool nvmet_cc_en(u32 cc) +{ + return (cc & NVME_CC_ENABLE) >> NVME_CC_EN_SHIFT; +} + +static inline u8 nvmet_cc_css(u32 cc) +{ + return (cc & NVME_CC_CSS_MASK) >> NVME_CC_CSS_SHIFT; +} + +static inline u8 nvmet_cc_mps(u32 cc) +{ + return (cc & NVME_CC_MPS_MASK) >> NVME_CC_MPS_SHIFT; +} + +static inline u8 nvmet_cc_ams(u32 cc) +{ + return (cc & NVME_CC_AMS_MASK) >> NVME_CC_AMS_SHIFT; +} + +static inline u8 nvmet_cc_shn(u32 cc) +{ + return (cc & NVME_CC_SHN_MASK) >> NVME_CC_SHN_SHIFT; +} + +static inline u8 nvmet_cc_iosqes(u32 cc) +{ + return (cc & NVME_CC_IOSQES_MASK) >> NVME_CC_IOSQES_SHIFT; +} + +static inline u8 nvmet_cc_iocqes(u32 cc) +{ + return (cc & NVME_CC_IOCQES_MASK) >> NVME_CC_IOCQES_SHIFT; +} + /* Convert a 32-bit number to a 16-bit 0's based number */ static inline __le16 to0based(u32 a) { - return cpu_to_le16(max(1U, min(1U << 16, a)) - 1); + return cpu_to_le16(clamp(a, 1U, 1U << 16) - 1); } static inline bool nvmet_ns_has_pi(struct nvmet_ns *ns) @@ -687,27 +863,39 @@ static inline bool nvmet_use_inline_bvec(struct nvmet_req *req) req->sg_cnt <= NVMET_MAX_INLINE_BIOVEC; } -static inline void nvmet_req_cns_error_complete(struct nvmet_req *req) -{ - pr_debug("unhandled identify cns %d on qid %d\n", - req->cmd->identify.cns, req->sq->qid); - req->error_loc = offsetof(struct nvme_identify, cns); - nvmet_req_complete(req, NVME_SC_INVALID_FIELD | NVME_SC_DNR); -} - static inline void nvmet_req_bio_put(struct nvmet_req *req, struct bio *bio) { if (bio != &req->b.inline_bio) bio_put(bio); + else + bio_uninit(bio); } +#ifdef CONFIG_NVME_TARGET_TCP_TLS +static inline key_serial_t nvmet_queue_tls_keyid(struct nvmet_sq *sq) +{ + return sq->tls_key ? key_serial(sq->tls_key) : 0; +} +static inline void nvmet_sq_put_tls_key(struct nvmet_sq *sq) +{ + if (sq->tls_key) { + key_put(sq->tls_key); + sq->tls_key = NULL; + } +} +#else +static inline key_serial_t nvmet_queue_tls_keyid(struct nvmet_sq *sq) { return 0; } +static inline void nvmet_sq_put_tls_key(struct nvmet_sq *sq) {} +#endif #ifdef CONFIG_NVME_TARGET_AUTH +u32 nvmet_auth_send_data_len(struct nvmet_req *req); void nvmet_execute_auth_send(struct nvmet_req *req); +u32 nvmet_auth_receive_data_len(struct nvmet_req *req); void nvmet_execute_auth_receive(struct nvmet_req *req); int nvmet_auth_set_key(struct nvmet_host *host, const char *secret, bool set_ctrl); int nvmet_auth_set_host_hash(struct nvmet_host *host, const char *hash); -int nvmet_setup_auth(struct nvmet_ctrl *ctrl); +u8 nvmet_setup_auth(struct nvmet_ctrl *ctrl, struct nvmet_sq *sq); void nvmet_auth_sq_init(struct nvmet_sq *sq); void nvmet_destroy_auth(struct nvmet_ctrl *ctrl); void nvmet_auth_sq_free(struct nvmet_sq *sq); @@ -717,16 +905,18 @@ int nvmet_auth_host_hash(struct nvmet_req *req, u8 *response, unsigned int hash_len); int nvmet_auth_ctrl_hash(struct nvmet_req *req, u8 *response, unsigned int hash_len); -static inline bool nvmet_has_auth(struct nvmet_ctrl *ctrl) +static inline bool nvmet_has_auth(struct nvmet_ctrl *ctrl, struct nvmet_sq *sq) { - return ctrl->host_key != NULL; + return ctrl->host_key != NULL && !nvmet_queue_tls_keyid(sq); } int nvmet_auth_ctrl_exponential(struct nvmet_req *req, u8 *buf, int buf_size); int nvmet_auth_ctrl_sesskey(struct nvmet_req *req, u8 *buf, int buf_size); +void nvmet_auth_insert_psk(struct nvmet_sq *sq); #else -static inline int nvmet_setup_auth(struct nvmet_ctrl *ctrl) +static inline u8 nvmet_setup_auth(struct nvmet_ctrl *ctrl, + struct nvmet_sq *sq) { return 0; } @@ -739,11 +929,49 @@ static inline bool nvmet_check_auth_status(struct nvmet_req *req) { return true; } -static inline bool nvmet_has_auth(struct nvmet_ctrl *ctrl) +static inline bool nvmet_has_auth(struct nvmet_ctrl *ctrl, + struct nvmet_sq *sq) { return false; } static inline const char *nvmet_dhchap_dhgroup_name(u8 dhgid) { return NULL; } +static inline void nvmet_auth_insert_psk(struct nvmet_sq *sq) {}; #endif +int nvmet_pr_init_ns(struct nvmet_ns *ns); +u16 nvmet_parse_pr_cmd(struct nvmet_req *req); +u16 nvmet_pr_check_cmd_access(struct nvmet_req *req); +int nvmet_ctrl_init_pr(struct nvmet_ctrl *ctrl); +void nvmet_ctrl_destroy_pr(struct nvmet_ctrl *ctrl); +void nvmet_pr_exit_ns(struct nvmet_ns *ns); +void nvmet_execute_get_log_page_resv(struct nvmet_req *req); +u16 nvmet_set_feat_resv_notif_mask(struct nvmet_req *req, u32 mask); +u16 nvmet_get_feat_resv_notif_mask(struct nvmet_req *req); +u16 nvmet_pr_get_ns_pc_ref(struct nvmet_req *req); +static inline void nvmet_pr_put_ns_pc_ref(struct nvmet_pr_per_ctrl_ref *pc_ref) +{ + percpu_ref_put(&pc_ref->ref); +} + +/* + * Data for the get_feature() and set_feature() operations of PCI target + * controllers. + */ +struct nvmet_feat_irq_coalesce { + u8 thr; + u8 time; +}; + +struct nvmet_feat_irq_config { + u16 iv; + bool cd; +}; + +struct nvmet_feat_arbitration { + u8 hpw; + u8 mpw; + u8 lpw; + u8 ab; +}; + #endif /* _NVMET_H */ diff --git a/drivers/nvme/target/passthru.c b/drivers/nvme/target/passthru.c index adc0958755d6..96648ec2fadb 100644 --- a/drivers/nvme/target/passthru.c +++ b/drivers/nvme/target/passthru.c @@ -13,7 +13,7 @@ #include "../host/nvme.h" #include "nvmet.h" -MODULE_IMPORT_NS(NVME_TARGET_PASSTHRU); +MODULE_IMPORT_NS("NVME_TARGET_PASSTHRU"); /* * xarray to maintain one passthru subsystem per nvme controller. @@ -99,17 +99,17 @@ static u16 nvmet_passthru_override_id_ctrl(struct nvmet_req *req) /* * The passthru NVMe driver may have a limit on the number of segments - * which depends on the host's memory fragementation. To solve this, + * which depends on the host's memory fragmentation. To solve this, * ensure mdts is limited to the pages equal to the number of segments. */ - max_hw_sectors = min_not_zero(pctrl->max_segments << (PAGE_SHIFT - 9), + max_hw_sectors = min_not_zero(pctrl->max_segments << PAGE_SECTORS_SHIFT, pctrl->max_hw_sectors); /* - * nvmet_passthru_map_sg is limitted to using a single bio so limit + * nvmet_passthru_map_sg is limited to using a single bio so limit * the mdts based on BIO_MAX_VECS as well */ - max_hw_sectors = min_not_zero(BIO_MAX_VECS << (PAGE_SHIFT - 9), + max_hw_sectors = min_not_zero(BIO_MAX_VECS << PAGE_SECTORS_SHIFT, max_hw_sectors); page_shift = NVME_CAP_MPSMIN(ctrl->cap) + 12; @@ -132,7 +132,7 @@ static u16 nvmet_passthru_override_id_ctrl(struct nvmet_req *req) id->sqes = min_t(__u8, ((0x6 << 4) | 0x6), id->sqes); id->cqes = min_t(__u8, ((0x4 << 4) | 0x4), id->cqes); - id->maxcmd = cpu_to_le16(NVMET_MAX_CMD); + id->maxcmd = cpu_to_le16(NVMET_MAX_CMD(ctrl)); /* don't support fuse commands */ id->fuses = 0; @@ -147,10 +147,10 @@ static u16 nvmet_passthru_override_id_ctrl(struct nvmet_req *req) * When passthru controller is setup using nvme-loop transport it will * export the passthru ctrl subsysnqn (PCIe NVMe ctrl) and will fail in * the nvme/host/core.c in the nvme_init_subsystem()->nvme_active_ctrl() - * code path with duplicate ctr subsynqn. In order to prevent that we + * code path with duplicate ctrl subsysnqn. In order to prevent that we * mask the passthru-ctrl subsysnqn with the target ctrl subsysnqn. */ - memcpy(id->subnqn, ctrl->subsysnqn, sizeof(id->subnqn)); + memcpy(id->subnqn, ctrl->subsys->subsysnqn, sizeof(id->subnqn)); /* use fabric id-ctrl values */ id->ioccsz = cpu_to_le32((sizeof(struct nvme_command) + @@ -216,22 +216,23 @@ static void nvmet_passthru_execute_cmd_work(struct work_struct *w) struct nvmet_req *req = container_of(w, struct nvmet_req, p.work); struct request *rq = req->p.rq; struct nvme_ctrl *ctrl = nvme_req(rq)->ctrl; + struct nvme_ns *ns = rq->q->queuedata; u32 effects; int status; - status = nvme_execute_passthru_rq(rq, &effects); - + effects = nvme_passthru_start(ctrl, ns, req->cmd->common.opcode); + status = nvme_execute_rq(rq, false); if (status == NVME_SC_SUCCESS && req->cmd->common.opcode == nvme_admin_identify) { switch (req->cmd->identify.cns) { case NVME_ID_CNS_CTRL: - nvmet_passthru_override_id_ctrl(req); + status = nvmet_passthru_override_id_ctrl(req); break; case NVME_ID_CNS_NS: - nvmet_passthru_override_id_ns(req); + status = nvmet_passthru_override_id_ns(req); break; case NVME_ID_CNS_NS_DESC_LIST: - nvmet_passthru_override_id_descs(req); + status = nvmet_passthru_override_id_descs(req); break; } } else if (status < 0) @@ -242,7 +243,7 @@ static void nvmet_passthru_execute_cmd_work(struct work_struct *w) blk_mq_free_request(rq); if (effects) - nvme_passthru_end(ctrl, effects, req->cmd, status); + nvme_passthru_end(ctrl, ns, effects, req->cmd, status); } static enum rq_end_io_ret nvmet_passthru_req_done(struct request *rq, @@ -260,6 +261,7 @@ static int nvmet_passthru_map_sg(struct nvmet_req *req, struct request *rq) { struct scatterlist *sg; struct bio *bio; + int ret = -EINVAL; int i; if (req->sg_cnt > BIO_MAX_VECS) @@ -276,16 +278,19 @@ static int nvmet_passthru_map_sg(struct nvmet_req *req, struct request *rq) } for_each_sg(req->sg, sg, req->sg_cnt, i) { - if (bio_add_pc_page(rq->q, bio, sg_page(sg), sg->length, - sg->offset) < sg->length) { - nvmet_req_bio_put(req, bio); - return -EINVAL; - } + if (bio_add_page(bio, sg_page(sg), sg->length, sg->offset) < + sg->length) + goto out_bio_put; } - blk_rq_bio_prep(rq, bio, req->sg_cnt); - + ret = blk_rq_append_bio(rq, bio); + if (ret) + goto out_bio_put; return 0; + +out_bio_put: + nvmet_req_bio_put(req, bio); + return ret; } static void nvmet_passthru_execute_cmd(struct nvmet_req *req) @@ -305,7 +310,7 @@ static void nvmet_passthru_execute_cmd(struct nvmet_req *req) ns = nvme_find_get_ns(ctrl, nsid); if (unlikely(!ns)) { pr_err("failed to get passthru ns nsid:%u\n", nsid); - status = NVME_SC_INVALID_NS | NVME_SC_DNR; + status = NVME_SC_INVALID_NS | NVME_STATUS_DNR; goto out; } @@ -425,7 +430,7 @@ u16 nvmet_parse_passthru_io_cmd(struct nvmet_req *req) * emulated in the future if regular targets grow support for * this feature. */ - return NVME_SC_INVALID_OPCODE | NVME_SC_DNR; + return NVME_SC_INVALID_OPCODE | NVME_STATUS_DNR; } return nvmet_setup_passthru_command(req); @@ -477,7 +482,7 @@ static u16 nvmet_passthru_get_set_features(struct nvmet_req *req) case NVME_FEAT_RESV_PERSIST: /* No reservations, see nvmet_parse_passthru_io_cmd() */ default: - return NVME_SC_INVALID_OPCODE | NVME_SC_DNR; + return NVME_SC_INVALID_OPCODE | NVME_STATUS_DNR; } } @@ -528,16 +533,14 @@ u16 nvmet_parse_passthru_admin_cmd(struct nvmet_req *req) case NVME_FEAT_HOST_ID: req->execute = nvmet_execute_get_features; return NVME_SC_SUCCESS; + case NVME_FEAT_FDP: + return nvmet_setup_passthru_command(req); default: return nvmet_passthru_get_set_features(req); } break; case nvme_admin_identify: switch (req->cmd->identify.cns) { - case NVME_ID_CNS_CTRL: - req->execute = nvmet_passthru_execute_cmd; - req->p.use_workqueue = true; - return NVME_SC_SUCCESS; case NVME_ID_CNS_CS_CTRL: switch (req->cmd->identify.csi) { case NVME_CSI_ZNS: @@ -545,8 +548,10 @@ u16 nvmet_parse_passthru_admin_cmd(struct nvmet_req *req) req->p.use_workqueue = true; return NVME_SC_SUCCESS; } - return NVME_SC_INVALID_OPCODE | NVME_SC_DNR; + return NVME_SC_INVALID_OPCODE | NVME_STATUS_DNR; + case NVME_ID_CNS_CTRL: case NVME_ID_CNS_NS: + case NVME_ID_CNS_NS_DESC_LIST: req->execute = nvmet_passthru_execute_cmd; req->p.use_workqueue = true; return NVME_SC_SUCCESS; @@ -557,7 +562,7 @@ u16 nvmet_parse_passthru_admin_cmd(struct nvmet_req *req) req->p.use_workqueue = true; return NVME_SC_SUCCESS; } - return NVME_SC_INVALID_OPCODE | NVME_SC_DNR; + return NVME_SC_INVALID_OPCODE | NVME_STATUS_DNR; default: return nvmet_setup_passthru_command(req); } @@ -601,7 +606,7 @@ int nvmet_passthru_ctrl_enable(struct nvmet_subsys *subsys) goto out_put_file; } - old = xa_cmpxchg(&passthru_subsystems, ctrl->cntlid, NULL, + old = xa_cmpxchg(&passthru_subsystems, ctrl->instance, NULL, subsys, GFP_KERNEL); if (xa_is_err(old)) { ret = xa_err(old); @@ -634,7 +639,7 @@ out_unlock: static void __nvmet_passthru_ctrl_disable(struct nvmet_subsys *subsys) { if (subsys->passthru_ctrl) { - xa_erase(&passthru_subsystems, subsys->passthru_ctrl->cntlid); + xa_erase(&passthru_subsystems, subsys->passthru_ctrl->instance); module_put(subsys->passthru_ctrl->ops->module); nvme_put_ctrl(subsys->passthru_ctrl); } diff --git a/drivers/nvme/target/pci-epf.c b/drivers/nvme/target/pci-epf.c new file mode 100644 index 000000000000..f858a6c9d7cb --- /dev/null +++ b/drivers/nvme/target/pci-epf.c @@ -0,0 +1,2651 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * NVMe PCI Endpoint Function target driver. + * + * Copyright (c) 2024, Western Digital Corporation or its affiliates. + * Copyright (c) 2024, Rick Wertenbroek <rick.wertenbroek@gmail.com> + * REDS Institute, HEIG-VD, HES-SO, Switzerland + */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include <linux/delay.h> +#include <linux/dmaengine.h> +#include <linux/io.h> +#include <linux/mempool.h> +#include <linux/module.h> +#include <linux/mutex.h> +#include <linux/nvme.h> +#include <linux/pci_ids.h> +#include <linux/pci-epc.h> +#include <linux/pci-epf.h> +#include <linux/pci_regs.h> +#include <linux/slab.h> + +#include "nvmet.h" + +static LIST_HEAD(nvmet_pci_epf_ports); +static DEFINE_MUTEX(nvmet_pci_epf_ports_mutex); + +/* + * Default and maximum allowed data transfer size. For the default, + * allow up to 128 page-sized segments. For the maximum allowed, + * use 4 times the default (which is completely arbitrary). + */ +#define NVMET_PCI_EPF_MAX_SEGS 128 +#define NVMET_PCI_EPF_MDTS_KB \ + (NVMET_PCI_EPF_MAX_SEGS << (PAGE_SHIFT - 10)) +#define NVMET_PCI_EPF_MAX_MDTS_KB (NVMET_PCI_EPF_MDTS_KB * 4) + +/* + * IRQ vector coalescing threshold: by default, post 8 CQEs before raising an + * interrupt vector to the host. This default 8 is completely arbitrary and can + * be changed by the host with a nvme_set_features command. + */ +#define NVMET_PCI_EPF_IV_THRESHOLD 8 + +/* + * BAR CC register and SQ polling intervals. + */ +#define NVMET_PCI_EPF_CC_POLL_INTERVAL msecs_to_jiffies(10) +#define NVMET_PCI_EPF_SQ_POLL_INTERVAL msecs_to_jiffies(5) +#define NVMET_PCI_EPF_SQ_POLL_IDLE msecs_to_jiffies(5000) + +/* + * SQ arbitration burst default: fetch at most 8 commands at a time from an SQ. + */ +#define NVMET_PCI_EPF_SQ_AB 8 + +/* + * Handling of CQs is normally immediate, unless we fail to map a CQ or the CQ + * is full, in which case we retry the CQ processing after this interval. + */ +#define NVMET_PCI_EPF_CQ_RETRY_INTERVAL msecs_to_jiffies(1) + +enum nvmet_pci_epf_queue_flags { + NVMET_PCI_EPF_Q_LIVE = 0, /* The queue is live */ + NVMET_PCI_EPF_Q_IRQ_ENABLED, /* IRQ is enabled for this queue */ +}; + +/* + * IRQ vector descriptor. + */ +struct nvmet_pci_epf_irq_vector { + unsigned int vector; + unsigned int ref; + bool cd; + int nr_irqs; +}; + +struct nvmet_pci_epf_queue { + union { + struct nvmet_sq nvme_sq; + struct nvmet_cq nvme_cq; + }; + struct nvmet_pci_epf_ctrl *ctrl; + unsigned long flags; + + u64 pci_addr; + size_t pci_size; + struct pci_epc_map pci_map; + + u16 qid; + u16 depth; + u16 vector; + u16 head; + u16 tail; + u16 phase; + u32 db; + + size_t qes; + + struct nvmet_pci_epf_irq_vector *iv; + struct workqueue_struct *iod_wq; + struct delayed_work work; + spinlock_t lock; + struct list_head list; +}; + +/* + * PCI Root Complex (RC) address data segment for mapping an admin or + * I/O command buffer @buf of @length bytes to the PCI address @pci_addr. + */ +struct nvmet_pci_epf_segment { + void *buf; + u64 pci_addr; + u32 length; +}; + +/* + * Command descriptors. + */ +struct nvmet_pci_epf_iod { + struct list_head link; + + struct nvmet_req req; + struct nvme_command cmd; + struct nvme_completion cqe; + unsigned int status; + + struct nvmet_pci_epf_ctrl *ctrl; + + struct nvmet_pci_epf_queue *sq; + struct nvmet_pci_epf_queue *cq; + + /* Data transfer size and direction for the command. */ + size_t data_len; + enum dma_data_direction dma_dir; + + /* + * PCI Root Complex (RC) address data segments: if nr_data_segs is 1, we + * use only @data_seg. Otherwise, the array of segments @data_segs is + * allocated to manage multiple PCI address data segments. @data_sgl and + * @data_sgt are used to setup the command request for execution by the + * target core. + */ + unsigned int nr_data_segs; + struct nvmet_pci_epf_segment data_seg; + struct nvmet_pci_epf_segment *data_segs; + struct scatterlist data_sgl; + struct sg_table data_sgt; + + struct work_struct work; + struct completion done; +}; + +/* + * PCI target controller private data. + */ +struct nvmet_pci_epf_ctrl { + struct nvmet_pci_epf *nvme_epf; + struct nvmet_port *port; + struct nvmet_ctrl *tctrl; + struct device *dev; + + unsigned int nr_queues; + struct nvmet_pci_epf_queue *sq; + struct nvmet_pci_epf_queue *cq; + unsigned int sq_ab; + + mempool_t iod_pool; + void *bar; + u64 cap; + u32 cc; + u32 csts; + + size_t io_sqes; + size_t io_cqes; + + size_t mps_shift; + size_t mps; + size_t mps_mask; + + unsigned int mdts; + + struct delayed_work poll_cc; + struct delayed_work poll_sqs; + + struct mutex irq_lock; + struct nvmet_pci_epf_irq_vector *irq_vectors; + unsigned int irq_vector_threshold; + + bool link_up; + bool enabled; +}; + +/* + * PCI EPF driver private data. + */ +struct nvmet_pci_epf { + struct pci_epf *epf; + + const struct pci_epc_features *epc_features; + + void *reg_bar; + size_t msix_table_offset; + + unsigned int irq_type; + unsigned int nr_vectors; + + struct nvmet_pci_epf_ctrl ctrl; + + bool dma_enabled; + struct dma_chan *dma_tx_chan; + struct mutex dma_tx_lock; + struct dma_chan *dma_rx_chan; + struct mutex dma_rx_lock; + + struct mutex mmio_lock; + + /* PCI endpoint function configfs attributes. */ + struct config_group group; + __le16 portid; + char subsysnqn[NVMF_NQN_SIZE]; + unsigned int mdts_kb; +}; + +static inline u32 nvmet_pci_epf_bar_read32(struct nvmet_pci_epf_ctrl *ctrl, + u32 off) +{ + __le32 *bar_reg = ctrl->bar + off; + + return le32_to_cpu(READ_ONCE(*bar_reg)); +} + +static inline void nvmet_pci_epf_bar_write32(struct nvmet_pci_epf_ctrl *ctrl, + u32 off, u32 val) +{ + __le32 *bar_reg = ctrl->bar + off; + + WRITE_ONCE(*bar_reg, cpu_to_le32(val)); +} + +static inline u64 nvmet_pci_epf_bar_read64(struct nvmet_pci_epf_ctrl *ctrl, + u32 off) +{ + return (u64)nvmet_pci_epf_bar_read32(ctrl, off) | + ((u64)nvmet_pci_epf_bar_read32(ctrl, off + 4) << 32); +} + +static inline void nvmet_pci_epf_bar_write64(struct nvmet_pci_epf_ctrl *ctrl, + u32 off, u64 val) +{ + nvmet_pci_epf_bar_write32(ctrl, off, val & 0xFFFFFFFF); + nvmet_pci_epf_bar_write32(ctrl, off + 4, (val >> 32) & 0xFFFFFFFF); +} + +static inline int nvmet_pci_epf_mem_map(struct nvmet_pci_epf *nvme_epf, + u64 pci_addr, size_t size, struct pci_epc_map *map) +{ + struct pci_epf *epf = nvme_epf->epf; + + return pci_epc_mem_map(epf->epc, epf->func_no, epf->vfunc_no, + pci_addr, size, map); +} + +static inline void nvmet_pci_epf_mem_unmap(struct nvmet_pci_epf *nvme_epf, + struct pci_epc_map *map) +{ + struct pci_epf *epf = nvme_epf->epf; + + pci_epc_mem_unmap(epf->epc, epf->func_no, epf->vfunc_no, map); +} + +struct nvmet_pci_epf_dma_filter { + struct device *dev; + u32 dma_mask; +}; + +static bool nvmet_pci_epf_dma_filter(struct dma_chan *chan, void *arg) +{ + struct nvmet_pci_epf_dma_filter *filter = arg; + struct dma_slave_caps caps; + + memset(&caps, 0, sizeof(caps)); + dma_get_slave_caps(chan, &caps); + + return chan->device->dev == filter->dev && + (filter->dma_mask & caps.directions); +} + +static void nvmet_pci_epf_init_dma(struct nvmet_pci_epf *nvme_epf) +{ + struct pci_epf *epf = nvme_epf->epf; + struct device *dev = &epf->dev; + struct nvmet_pci_epf_dma_filter filter; + struct dma_chan *chan; + dma_cap_mask_t mask; + + mutex_init(&nvme_epf->dma_rx_lock); + mutex_init(&nvme_epf->dma_tx_lock); + + dma_cap_zero(mask); + dma_cap_set(DMA_SLAVE, mask); + + filter.dev = epf->epc->dev.parent; + filter.dma_mask = BIT(DMA_DEV_TO_MEM); + + chan = dma_request_channel(mask, nvmet_pci_epf_dma_filter, &filter); + if (!chan) + goto out_dma_no_rx; + + nvme_epf->dma_rx_chan = chan; + + filter.dma_mask = BIT(DMA_MEM_TO_DEV); + chan = dma_request_channel(mask, nvmet_pci_epf_dma_filter, &filter); + if (!chan) + goto out_dma_no_tx; + + nvme_epf->dma_tx_chan = chan; + + nvme_epf->dma_enabled = true; + + dev_dbg(dev, "Using DMA RX channel %s, maximum segment size %u B\n", + dma_chan_name(nvme_epf->dma_rx_chan), + dma_get_max_seg_size(dmaengine_get_dma_device(nvme_epf-> + dma_rx_chan))); + + dev_dbg(dev, "Using DMA TX channel %s, maximum segment size %u B\n", + dma_chan_name(nvme_epf->dma_tx_chan), + dma_get_max_seg_size(dmaengine_get_dma_device(nvme_epf-> + dma_tx_chan))); + + return; + +out_dma_no_tx: + dma_release_channel(nvme_epf->dma_rx_chan); + nvme_epf->dma_rx_chan = NULL; + +out_dma_no_rx: + mutex_destroy(&nvme_epf->dma_rx_lock); + mutex_destroy(&nvme_epf->dma_tx_lock); + nvme_epf->dma_enabled = false; + + dev_info(&epf->dev, "DMA not supported, falling back to MMIO\n"); +} + +static void nvmet_pci_epf_deinit_dma(struct nvmet_pci_epf *nvme_epf) +{ + if (!nvme_epf->dma_enabled) + return; + + dma_release_channel(nvme_epf->dma_tx_chan); + nvme_epf->dma_tx_chan = NULL; + dma_release_channel(nvme_epf->dma_rx_chan); + nvme_epf->dma_rx_chan = NULL; + mutex_destroy(&nvme_epf->dma_rx_lock); + mutex_destroy(&nvme_epf->dma_tx_lock); + nvme_epf->dma_enabled = false; +} + +static int nvmet_pci_epf_dma_transfer(struct nvmet_pci_epf *nvme_epf, + struct nvmet_pci_epf_segment *seg, enum dma_data_direction dir) +{ + struct pci_epf *epf = nvme_epf->epf; + struct dma_async_tx_descriptor *desc; + struct dma_slave_config sconf = {}; + struct device *dev = &epf->dev; + struct device *dma_dev; + struct dma_chan *chan; + dma_cookie_t cookie; + dma_addr_t dma_addr; + struct mutex *lock; + int ret; + + switch (dir) { + case DMA_FROM_DEVICE: + lock = &nvme_epf->dma_rx_lock; + chan = nvme_epf->dma_rx_chan; + sconf.direction = DMA_DEV_TO_MEM; + sconf.src_addr = seg->pci_addr; + break; + case DMA_TO_DEVICE: + lock = &nvme_epf->dma_tx_lock; + chan = nvme_epf->dma_tx_chan; + sconf.direction = DMA_MEM_TO_DEV; + sconf.dst_addr = seg->pci_addr; + break; + default: + return -EINVAL; + } + + mutex_lock(lock); + + dma_dev = dmaengine_get_dma_device(chan); + dma_addr = dma_map_single(dma_dev, seg->buf, seg->length, dir); + ret = dma_mapping_error(dma_dev, dma_addr); + if (ret) + goto unlock; + + ret = dmaengine_slave_config(chan, &sconf); + if (ret) { + dev_err(dev, "Failed to configure DMA channel\n"); + goto unmap; + } + + desc = dmaengine_prep_slave_single(chan, dma_addr, seg->length, + sconf.direction, DMA_CTRL_ACK); + if (!desc) { + dev_err(dev, "Failed to prepare DMA\n"); + ret = -EIO; + goto unmap; + } + + cookie = dmaengine_submit(desc); + ret = dma_submit_error(cookie); + if (ret) { + dev_err(dev, "Failed to do DMA submit (err=%d)\n", ret); + goto unmap; + } + + if (dma_sync_wait(chan, cookie) != DMA_COMPLETE) { + dev_err(dev, "DMA transfer failed\n"); + ret = -EIO; + } + + dmaengine_terminate_sync(chan); + +unmap: + dma_unmap_single(dma_dev, dma_addr, seg->length, dir); + +unlock: + mutex_unlock(lock); + + return ret; +} + +static int nvmet_pci_epf_mmio_transfer(struct nvmet_pci_epf *nvme_epf, + struct nvmet_pci_epf_segment *seg, enum dma_data_direction dir) +{ + u64 pci_addr = seg->pci_addr; + u32 length = seg->length; + void *buf = seg->buf; + struct pci_epc_map map; + int ret = -EINVAL; + + /* + * Note: MMIO transfers do not need serialization but this is a + * simple way to avoid using too many mapping windows. + */ + mutex_lock(&nvme_epf->mmio_lock); + + while (length) { + ret = nvmet_pci_epf_mem_map(nvme_epf, pci_addr, length, &map); + if (ret) + break; + + switch (dir) { + case DMA_FROM_DEVICE: + memcpy_fromio(buf, map.virt_addr, map.pci_size); + break; + case DMA_TO_DEVICE: + memcpy_toio(map.virt_addr, buf, map.pci_size); + break; + default: + ret = -EINVAL; + goto unlock; + } + + pci_addr += map.pci_size; + buf += map.pci_size; + length -= map.pci_size; + + nvmet_pci_epf_mem_unmap(nvme_epf, &map); + } + +unlock: + mutex_unlock(&nvme_epf->mmio_lock); + + return ret; +} + +static inline int nvmet_pci_epf_transfer_seg(struct nvmet_pci_epf *nvme_epf, + struct nvmet_pci_epf_segment *seg, enum dma_data_direction dir) +{ + if (nvme_epf->dma_enabled) + return nvmet_pci_epf_dma_transfer(nvme_epf, seg, dir); + + return nvmet_pci_epf_mmio_transfer(nvme_epf, seg, dir); +} + +static inline int nvmet_pci_epf_transfer(struct nvmet_pci_epf_ctrl *ctrl, + void *buf, u64 pci_addr, u32 length, + enum dma_data_direction dir) +{ + struct nvmet_pci_epf_segment seg = { + .buf = buf, + .pci_addr = pci_addr, + .length = length, + }; + + return nvmet_pci_epf_transfer_seg(ctrl->nvme_epf, &seg, dir); +} + +static int nvmet_pci_epf_alloc_irq_vectors(struct nvmet_pci_epf_ctrl *ctrl) +{ + ctrl->irq_vectors = kcalloc(ctrl->nr_queues, + sizeof(struct nvmet_pci_epf_irq_vector), + GFP_KERNEL); + if (!ctrl->irq_vectors) + return -ENOMEM; + + mutex_init(&ctrl->irq_lock); + + return 0; +} + +static void nvmet_pci_epf_free_irq_vectors(struct nvmet_pci_epf_ctrl *ctrl) +{ + if (ctrl->irq_vectors) { + mutex_destroy(&ctrl->irq_lock); + kfree(ctrl->irq_vectors); + ctrl->irq_vectors = NULL; + } +} + +static struct nvmet_pci_epf_irq_vector * +nvmet_pci_epf_find_irq_vector(struct nvmet_pci_epf_ctrl *ctrl, u16 vector) +{ + struct nvmet_pci_epf_irq_vector *iv; + int i; + + lockdep_assert_held(&ctrl->irq_lock); + + for (i = 0; i < ctrl->nr_queues; i++) { + iv = &ctrl->irq_vectors[i]; + if (iv->ref && iv->vector == vector) + return iv; + } + + return NULL; +} + +static struct nvmet_pci_epf_irq_vector * +nvmet_pci_epf_add_irq_vector(struct nvmet_pci_epf_ctrl *ctrl, u16 vector) +{ + struct nvmet_pci_epf_irq_vector *iv; + int i; + + mutex_lock(&ctrl->irq_lock); + + iv = nvmet_pci_epf_find_irq_vector(ctrl, vector); + if (iv) { + iv->ref++; + goto unlock; + } + + for (i = 0; i < ctrl->nr_queues; i++) { + iv = &ctrl->irq_vectors[i]; + if (!iv->ref) + break; + } + + if (WARN_ON_ONCE(!iv)) + goto unlock; + + iv->ref = 1; + iv->vector = vector; + iv->nr_irqs = 0; + +unlock: + mutex_unlock(&ctrl->irq_lock); + + return iv; +} + +static void nvmet_pci_epf_remove_irq_vector(struct nvmet_pci_epf_ctrl *ctrl, + u16 vector) +{ + struct nvmet_pci_epf_irq_vector *iv; + + mutex_lock(&ctrl->irq_lock); + + iv = nvmet_pci_epf_find_irq_vector(ctrl, vector); + if (iv) { + iv->ref--; + if (!iv->ref) { + iv->vector = 0; + iv->nr_irqs = 0; + } + } + + mutex_unlock(&ctrl->irq_lock); +} + +static bool nvmet_pci_epf_should_raise_irq(struct nvmet_pci_epf_ctrl *ctrl, + struct nvmet_pci_epf_queue *cq, bool force) +{ + struct nvmet_pci_epf_irq_vector *iv = cq->iv; + bool ret; + + /* IRQ coalescing for the admin queue is not allowed. */ + if (!cq->qid) + return true; + + if (iv->cd) + return true; + + if (force) { + ret = iv->nr_irqs > 0; + } else { + iv->nr_irqs++; + ret = iv->nr_irqs >= ctrl->irq_vector_threshold; + } + if (ret) + iv->nr_irqs = 0; + + return ret; +} + +static void nvmet_pci_epf_raise_irq(struct nvmet_pci_epf_ctrl *ctrl, + struct nvmet_pci_epf_queue *cq, bool force) +{ + struct nvmet_pci_epf *nvme_epf = ctrl->nvme_epf; + struct pci_epf *epf = nvme_epf->epf; + int ret = 0; + + if (!test_bit(NVMET_PCI_EPF_Q_LIVE, &cq->flags) || + !test_bit(NVMET_PCI_EPF_Q_IRQ_ENABLED, &cq->flags)) + return; + + mutex_lock(&ctrl->irq_lock); + + if (!nvmet_pci_epf_should_raise_irq(ctrl, cq, force)) + goto unlock; + + switch (nvme_epf->irq_type) { + case PCI_IRQ_MSIX: + case PCI_IRQ_MSI: + /* + * If we fail to raise an MSI or MSI-X interrupt, it is likely + * because the host is using legacy INTX IRQs (e.g. BIOS, + * grub), but we can fallback to the INTX type only if the + * endpoint controller supports this type. + */ + ret = pci_epc_raise_irq(epf->epc, epf->func_no, epf->vfunc_no, + nvme_epf->irq_type, cq->vector + 1); + if (!ret || !nvme_epf->epc_features->intx_capable) + break; + fallthrough; + case PCI_IRQ_INTX: + ret = pci_epc_raise_irq(epf->epc, epf->func_no, epf->vfunc_no, + PCI_IRQ_INTX, 0); + break; + default: + WARN_ON_ONCE(1); + ret = -EINVAL; + break; + } + + if (ret) + dev_err_ratelimited(ctrl->dev, + "CQ[%u]: Failed to raise IRQ (err=%d)\n", + cq->qid, ret); + +unlock: + mutex_unlock(&ctrl->irq_lock); +} + +static inline const char *nvmet_pci_epf_iod_name(struct nvmet_pci_epf_iod *iod) +{ + return nvme_opcode_str(iod->sq->qid, iod->cmd.common.opcode); +} + +static void nvmet_pci_epf_exec_iod_work(struct work_struct *work); + +static struct nvmet_pci_epf_iod * +nvmet_pci_epf_alloc_iod(struct nvmet_pci_epf_queue *sq) +{ + struct nvmet_pci_epf_ctrl *ctrl = sq->ctrl; + struct nvmet_pci_epf_iod *iod; + + iod = mempool_alloc(&ctrl->iod_pool, GFP_KERNEL); + if (unlikely(!iod)) + return NULL; + + memset(iod, 0, sizeof(*iod)); + iod->req.cmd = &iod->cmd; + iod->req.cqe = &iod->cqe; + iod->req.port = ctrl->port; + iod->ctrl = ctrl; + iod->sq = sq; + iod->cq = &ctrl->cq[sq->qid]; + INIT_LIST_HEAD(&iod->link); + iod->dma_dir = DMA_NONE; + INIT_WORK(&iod->work, nvmet_pci_epf_exec_iod_work); + init_completion(&iod->done); + + return iod; +} + +/* + * Allocate or grow a command table of PCI segments. + */ +static int nvmet_pci_epf_alloc_iod_data_segs(struct nvmet_pci_epf_iod *iod, + int nsegs) +{ + struct nvmet_pci_epf_segment *segs; + int nr_segs = iod->nr_data_segs + nsegs; + + segs = krealloc(iod->data_segs, + nr_segs * sizeof(struct nvmet_pci_epf_segment), + GFP_KERNEL | __GFP_ZERO); + if (!segs) + return -ENOMEM; + + iod->nr_data_segs = nr_segs; + iod->data_segs = segs; + + return 0; +} + +static void nvmet_pci_epf_free_iod(struct nvmet_pci_epf_iod *iod) +{ + int i; + + if (iod->data_segs) { + for (i = 0; i < iod->nr_data_segs; i++) + kfree(iod->data_segs[i].buf); + if (iod->data_segs != &iod->data_seg) + kfree(iod->data_segs); + } + if (iod->data_sgt.nents > 1) + sg_free_table(&iod->data_sgt); + mempool_free(iod, &iod->ctrl->iod_pool); +} + +static int nvmet_pci_epf_transfer_iod_data(struct nvmet_pci_epf_iod *iod) +{ + struct nvmet_pci_epf *nvme_epf = iod->ctrl->nvme_epf; + struct nvmet_pci_epf_segment *seg = &iod->data_segs[0]; + int i, ret; + + /* Split the data transfer according to the PCI segments. */ + for (i = 0; i < iod->nr_data_segs; i++, seg++) { + ret = nvmet_pci_epf_transfer_seg(nvme_epf, seg, iod->dma_dir); + if (ret) { + iod->status = NVME_SC_DATA_XFER_ERROR | NVME_STATUS_DNR; + return ret; + } + } + + return 0; +} + +static inline u32 nvmet_pci_epf_prp_ofst(struct nvmet_pci_epf_ctrl *ctrl, + u64 prp) +{ + return prp & ctrl->mps_mask; +} + +static inline size_t nvmet_pci_epf_prp_size(struct nvmet_pci_epf_ctrl *ctrl, + u64 prp) +{ + return ctrl->mps - nvmet_pci_epf_prp_ofst(ctrl, prp); +} + +/* + * Transfer a PRP list from the host and return the number of prps. + */ +static int nvmet_pci_epf_get_prp_list(struct nvmet_pci_epf_ctrl *ctrl, u64 prp, + size_t xfer_len, __le64 *prps) +{ + size_t nr_prps = (xfer_len + ctrl->mps_mask) >> ctrl->mps_shift; + u32 length; + int ret; + + /* + * Compute the number of PRPs required for the number of bytes to + * transfer (xfer_len). If this number overflows the memory page size + * with the PRP list pointer specified, only return the space available + * in the memory page, the last PRP in there will be a PRP list pointer + * to the remaining PRPs. + */ + length = min(nvmet_pci_epf_prp_size(ctrl, prp), nr_prps << 3); + ret = nvmet_pci_epf_transfer(ctrl, prps, prp, length, DMA_FROM_DEVICE); + if (ret) + return ret; + + return length >> 3; +} + +static int nvmet_pci_epf_iod_parse_prp_list(struct nvmet_pci_epf_ctrl *ctrl, + struct nvmet_pci_epf_iod *iod) +{ + struct nvme_command *cmd = &iod->cmd; + struct nvmet_pci_epf_segment *seg; + size_t size = 0, ofst, prp_size, xfer_len; + size_t transfer_len = iod->data_len; + int nr_segs, nr_prps = 0; + u64 pci_addr, prp; + int i = 0, ret; + __le64 *prps; + + prps = kzalloc(ctrl->mps, GFP_KERNEL); + if (!prps) + goto err_internal; + + /* + * Allocate PCI segments for the command: this considers the worst case + * scenario where all prps are discontiguous, so get as many segments + * as we can have prps. In practice, most of the time, we will have + * far less PCI segments than prps. + */ + prp = le64_to_cpu(cmd->common.dptr.prp1); + if (!prp) + goto err_invalid_field; + + ofst = nvmet_pci_epf_prp_ofst(ctrl, prp); + nr_segs = (transfer_len + ofst + ctrl->mps - 1) >> ctrl->mps_shift; + + ret = nvmet_pci_epf_alloc_iod_data_segs(iod, nr_segs); + if (ret) + goto err_internal; + + /* Set the first segment using prp1. */ + seg = &iod->data_segs[0]; + seg->pci_addr = prp; + seg->length = nvmet_pci_epf_prp_size(ctrl, prp); + + size = seg->length; + pci_addr = prp + size; + nr_segs = 1; + + /* + * Now build the PCI address segments using the PRP lists, starting + * from prp2. + */ + prp = le64_to_cpu(cmd->common.dptr.prp2); + if (!prp) + goto err_invalid_field; + + while (size < transfer_len) { + xfer_len = transfer_len - size; + + if (!nr_prps) { + nr_prps = nvmet_pci_epf_get_prp_list(ctrl, prp, + xfer_len, prps); + if (nr_prps < 0) + goto err_internal; + + i = 0; + ofst = 0; + } + + /* Current entry */ + prp = le64_to_cpu(prps[i]); + if (!prp) + goto err_invalid_field; + + /* Did we reach the last PRP entry of the list? */ + if (xfer_len > ctrl->mps && i == nr_prps - 1) { + /* We need more PRPs: PRP is a list pointer. */ + nr_prps = 0; + continue; + } + + /* Only the first PRP is allowed to have an offset. */ + if (nvmet_pci_epf_prp_ofst(ctrl, prp)) + goto err_invalid_offset; + + if (prp != pci_addr) { + /* Discontiguous prp: new segment. */ + nr_segs++; + if (WARN_ON_ONCE(nr_segs > iod->nr_data_segs)) + goto err_internal; + + seg++; + seg->pci_addr = prp; + seg->length = 0; + pci_addr = prp; + } + + prp_size = min_t(size_t, ctrl->mps, xfer_len); + seg->length += prp_size; + pci_addr += prp_size; + size += prp_size; + + i++; + } + + iod->nr_data_segs = nr_segs; + ret = 0; + + if (size != transfer_len) { + dev_err(ctrl->dev, + "PRPs transfer length mismatch: got %zu B, need %zu B\n", + size, transfer_len); + goto err_internal; + } + + kfree(prps); + + return 0; + +err_invalid_offset: + dev_err(ctrl->dev, "PRPs list invalid offset\n"); + iod->status = NVME_SC_PRP_INVALID_OFFSET | NVME_STATUS_DNR; + goto err; + +err_invalid_field: + dev_err(ctrl->dev, "PRPs list invalid field\n"); + iod->status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; + goto err; + +err_internal: + dev_err(ctrl->dev, "PRPs list internal error\n"); + iod->status = NVME_SC_INTERNAL | NVME_STATUS_DNR; + +err: + kfree(prps); + return -EINVAL; +} + +static int nvmet_pci_epf_iod_parse_prp_simple(struct nvmet_pci_epf_ctrl *ctrl, + struct nvmet_pci_epf_iod *iod) +{ + struct nvme_command *cmd = &iod->cmd; + size_t transfer_len = iod->data_len; + int ret, nr_segs = 1; + u64 prp1, prp2 = 0; + size_t prp1_size; + + prp1 = le64_to_cpu(cmd->common.dptr.prp1); + prp1_size = nvmet_pci_epf_prp_size(ctrl, prp1); + + /* For commands crossing a page boundary, we should have prp2. */ + if (transfer_len > prp1_size) { + prp2 = le64_to_cpu(cmd->common.dptr.prp2); + if (!prp2) { + iod->status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; + return -EINVAL; + } + if (nvmet_pci_epf_prp_ofst(ctrl, prp2)) { + iod->status = + NVME_SC_PRP_INVALID_OFFSET | NVME_STATUS_DNR; + return -EINVAL; + } + if (prp2 != prp1 + prp1_size) + nr_segs = 2; + } + + if (nr_segs == 1) { + iod->nr_data_segs = 1; + iod->data_segs = &iod->data_seg; + iod->data_segs[0].pci_addr = prp1; + iod->data_segs[0].length = transfer_len; + return 0; + } + + ret = nvmet_pci_epf_alloc_iod_data_segs(iod, nr_segs); + if (ret) { + iod->status = NVME_SC_INTERNAL | NVME_STATUS_DNR; + return ret; + } + + iod->data_segs[0].pci_addr = prp1; + iod->data_segs[0].length = prp1_size; + iod->data_segs[1].pci_addr = prp2; + iod->data_segs[1].length = transfer_len - prp1_size; + + return 0; +} + +static int nvmet_pci_epf_iod_parse_prps(struct nvmet_pci_epf_iod *iod) +{ + struct nvmet_pci_epf_ctrl *ctrl = iod->ctrl; + u64 prp1 = le64_to_cpu(iod->cmd.common.dptr.prp1); + size_t ofst; + + /* Get the PCI address segments for the command using its PRPs. */ + ofst = nvmet_pci_epf_prp_ofst(ctrl, prp1); + if (ofst & 0x3) { + iod->status = NVME_SC_PRP_INVALID_OFFSET | NVME_STATUS_DNR; + return -EINVAL; + } + + if (iod->data_len + ofst <= ctrl->mps * 2) + return nvmet_pci_epf_iod_parse_prp_simple(ctrl, iod); + + return nvmet_pci_epf_iod_parse_prp_list(ctrl, iod); +} + +/* + * Transfer an SGL segment from the host and return the number of data + * descriptors and the next segment descriptor, if any. + */ +static struct nvme_sgl_desc * +nvmet_pci_epf_get_sgl_segment(struct nvmet_pci_epf_ctrl *ctrl, + struct nvme_sgl_desc *desc, unsigned int *nr_sgls) +{ + struct nvme_sgl_desc *sgls; + u32 length = le32_to_cpu(desc->length); + int nr_descs, ret; + void *buf; + + buf = kmalloc(length, GFP_KERNEL); + if (!buf) + return NULL; + + ret = nvmet_pci_epf_transfer(ctrl, buf, le64_to_cpu(desc->addr), length, + DMA_FROM_DEVICE); + if (ret) { + kfree(buf); + return NULL; + } + + sgls = buf; + nr_descs = length / sizeof(struct nvme_sgl_desc); + if (sgls[nr_descs - 1].type == (NVME_SGL_FMT_SEG_DESC << 4) || + sgls[nr_descs - 1].type == (NVME_SGL_FMT_LAST_SEG_DESC << 4)) { + /* + * We have another SGL segment following this one: do not count + * it as a regular data SGL descriptor and return it to the + * caller. + */ + *desc = sgls[nr_descs - 1]; + nr_descs--; + } else { + /* We do not have another SGL segment after this one. */ + desc->length = 0; + } + + *nr_sgls = nr_descs; + + return sgls; +} + +static int nvmet_pci_epf_iod_parse_sgl_segments(struct nvmet_pci_epf_ctrl *ctrl, + struct nvmet_pci_epf_iod *iod) +{ + struct nvme_command *cmd = &iod->cmd; + struct nvme_sgl_desc seg = cmd->common.dptr.sgl; + struct nvme_sgl_desc *sgls = NULL; + int n = 0, i, nr_sgls; + int ret; + + /* + * We do not support inline data nor keyed SGLs, so we should be seeing + * only segment descriptors. + */ + if (seg.type != (NVME_SGL_FMT_SEG_DESC << 4) && + seg.type != (NVME_SGL_FMT_LAST_SEG_DESC << 4)) { + iod->status = NVME_SC_SGL_INVALID_TYPE | NVME_STATUS_DNR; + return -EIO; + } + + while (seg.length) { + sgls = nvmet_pci_epf_get_sgl_segment(ctrl, &seg, &nr_sgls); + if (!sgls) { + iod->status = NVME_SC_INTERNAL | NVME_STATUS_DNR; + return -EIO; + } + + /* Grow the PCI segment table as needed. */ + ret = nvmet_pci_epf_alloc_iod_data_segs(iod, nr_sgls); + if (ret) { + iod->status = NVME_SC_INTERNAL | NVME_STATUS_DNR; + goto out; + } + + /* + * Parse the SGL descriptors to build the PCI segment table, + * checking the descriptor type as we go. + */ + for (i = 0; i < nr_sgls; i++) { + if (sgls[i].type != (NVME_SGL_FMT_DATA_DESC << 4)) { + iod->status = NVME_SC_SGL_INVALID_TYPE | + NVME_STATUS_DNR; + goto out; + } + iod->data_segs[n].pci_addr = le64_to_cpu(sgls[i].addr); + iod->data_segs[n].length = le32_to_cpu(sgls[i].length); + n++; + } + + kfree(sgls); + } + + out: + if (iod->status != NVME_SC_SUCCESS) { + kfree(sgls); + return -EIO; + } + + return 0; +} + +static int nvmet_pci_epf_iod_parse_sgls(struct nvmet_pci_epf_iod *iod) +{ + struct nvmet_pci_epf_ctrl *ctrl = iod->ctrl; + struct nvme_sgl_desc *sgl = &iod->cmd.common.dptr.sgl; + + if (sgl->type == (NVME_SGL_FMT_DATA_DESC << 4)) { + /* Single data descriptor case. */ + iod->nr_data_segs = 1; + iod->data_segs = &iod->data_seg; + iod->data_seg.pci_addr = le64_to_cpu(sgl->addr); + iod->data_seg.length = le32_to_cpu(sgl->length); + return 0; + } + + return nvmet_pci_epf_iod_parse_sgl_segments(ctrl, iod); +} + +static int nvmet_pci_epf_alloc_iod_data_buf(struct nvmet_pci_epf_iod *iod) +{ + struct nvmet_pci_epf_ctrl *ctrl = iod->ctrl; + struct nvmet_req *req = &iod->req; + struct nvmet_pci_epf_segment *seg; + struct scatterlist *sg; + int ret, i; + + if (iod->data_len > ctrl->mdts) { + iod->status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; + return -EINVAL; + } + + /* + * Get the PCI address segments for the command data buffer using either + * its SGLs or PRPs. + */ + if (iod->cmd.common.flags & NVME_CMD_SGL_ALL) + ret = nvmet_pci_epf_iod_parse_sgls(iod); + else + ret = nvmet_pci_epf_iod_parse_prps(iod); + if (ret) + return ret; + + /* Get a command buffer using SGLs matching the PCI segments. */ + if (iod->nr_data_segs == 1) { + sg_init_table(&iod->data_sgl, 1); + iod->data_sgt.sgl = &iod->data_sgl; + iod->data_sgt.nents = 1; + iod->data_sgt.orig_nents = 1; + } else { + ret = sg_alloc_table(&iod->data_sgt, iod->nr_data_segs, + GFP_KERNEL); + if (ret) + goto err_nomem; + } + + for_each_sgtable_sg(&iod->data_sgt, sg, i) { + seg = &iod->data_segs[i]; + seg->buf = kmalloc(seg->length, GFP_KERNEL); + if (!seg->buf) + goto err_nomem; + sg_set_buf(sg, seg->buf, seg->length); + } + + req->transfer_len = iod->data_len; + req->sg = iod->data_sgt.sgl; + req->sg_cnt = iod->data_sgt.nents; + + return 0; + +err_nomem: + iod->status = NVME_SC_INTERNAL | NVME_STATUS_DNR; + return -ENOMEM; +} + +static void nvmet_pci_epf_complete_iod(struct nvmet_pci_epf_iod *iod) +{ + struct nvmet_pci_epf_queue *cq = iod->cq; + unsigned long flags; + + /* Print an error message for failed commands, except AENs. */ + iod->status = le16_to_cpu(iod->cqe.status) >> 1; + if (iod->status && iod->cmd.common.opcode != nvme_admin_async_event) + dev_err(iod->ctrl->dev, + "CQ[%d]: Command %s (0x%x) status 0x%0x\n", + iod->sq->qid, nvmet_pci_epf_iod_name(iod), + iod->cmd.common.opcode, iod->status); + + /* + * Add the command to the list of completed commands and schedule the + * CQ work. + */ + spin_lock_irqsave(&cq->lock, flags); + list_add_tail(&iod->link, &cq->list); + queue_delayed_work(system_highpri_wq, &cq->work, 0); + spin_unlock_irqrestore(&cq->lock, flags); +} + +static void nvmet_pci_epf_drain_queue(struct nvmet_pci_epf_queue *queue) +{ + struct nvmet_pci_epf_iod *iod; + unsigned long flags; + + spin_lock_irqsave(&queue->lock, flags); + while (!list_empty(&queue->list)) { + iod = list_first_entry(&queue->list, struct nvmet_pci_epf_iod, + link); + list_del_init(&iod->link); + nvmet_pci_epf_free_iod(iod); + } + spin_unlock_irqrestore(&queue->lock, flags); +} + +static int nvmet_pci_epf_add_port(struct nvmet_port *port) +{ + mutex_lock(&nvmet_pci_epf_ports_mutex); + list_add_tail(&port->entry, &nvmet_pci_epf_ports); + mutex_unlock(&nvmet_pci_epf_ports_mutex); + return 0; +} + +static void nvmet_pci_epf_remove_port(struct nvmet_port *port) +{ + mutex_lock(&nvmet_pci_epf_ports_mutex); + list_del_init(&port->entry); + mutex_unlock(&nvmet_pci_epf_ports_mutex); +} + +static struct nvmet_port * +nvmet_pci_epf_find_port(struct nvmet_pci_epf_ctrl *ctrl, __le16 portid) +{ + struct nvmet_port *p, *port = NULL; + + mutex_lock(&nvmet_pci_epf_ports_mutex); + list_for_each_entry(p, &nvmet_pci_epf_ports, entry) { + if (p->disc_addr.portid == portid) { + port = p; + break; + } + } + mutex_unlock(&nvmet_pci_epf_ports_mutex); + + return port; +} + +static void nvmet_pci_epf_queue_response(struct nvmet_req *req) +{ + struct nvmet_pci_epf_iod *iod = + container_of(req, struct nvmet_pci_epf_iod, req); + + iod->status = le16_to_cpu(req->cqe->status) >> 1; + + /* + * If the command failed or we have no data to transfer, complete the + * command immediately. + */ + if (iod->status || !iod->data_len || iod->dma_dir != DMA_TO_DEVICE) { + nvmet_pci_epf_complete_iod(iod); + return; + } + + complete(&iod->done); +} + +static u8 nvmet_pci_epf_get_mdts(const struct nvmet_ctrl *tctrl) +{ + struct nvmet_pci_epf_ctrl *ctrl = tctrl->drvdata; + int page_shift = NVME_CAP_MPSMIN(tctrl->cap) + 12; + + return ilog2(ctrl->mdts) - page_shift; +} + +static u16 nvmet_pci_epf_create_cq(struct nvmet_ctrl *tctrl, + u16 cqid, u16 flags, u16 qsize, u64 pci_addr, u16 vector) +{ + struct nvmet_pci_epf_ctrl *ctrl = tctrl->drvdata; + struct nvmet_pci_epf_queue *cq = &ctrl->cq[cqid]; + u16 status; + int ret; + + if (test_bit(NVMET_PCI_EPF_Q_LIVE, &cq->flags)) + return NVME_SC_QID_INVALID | NVME_STATUS_DNR; + + if (!(flags & NVME_QUEUE_PHYS_CONTIG)) + return NVME_SC_INVALID_QUEUE | NVME_STATUS_DNR; + + cq->pci_addr = pci_addr; + cq->qid = cqid; + cq->depth = qsize + 1; + cq->vector = vector; + cq->head = 0; + cq->tail = 0; + cq->phase = 1; + cq->db = NVME_REG_DBS + (((cqid * 2) + 1) * sizeof(u32)); + nvmet_pci_epf_bar_write32(ctrl, cq->db, 0); + + if (!cqid) + cq->qes = sizeof(struct nvme_completion); + else + cq->qes = ctrl->io_cqes; + cq->pci_size = cq->qes * cq->depth; + + if (flags & NVME_CQ_IRQ_ENABLED) { + cq->iv = nvmet_pci_epf_add_irq_vector(ctrl, vector); + if (!cq->iv) + return NVME_SC_INTERNAL | NVME_STATUS_DNR; + set_bit(NVMET_PCI_EPF_Q_IRQ_ENABLED, &cq->flags); + } + + status = nvmet_cq_create(tctrl, &cq->nvme_cq, cqid, cq->depth); + if (status != NVME_SC_SUCCESS) + goto err; + + /* + * Map the CQ PCI address space and since PCI endpoint controllers may + * return a partial mapping, check that the mapping is large enough. + */ + ret = nvmet_pci_epf_mem_map(ctrl->nvme_epf, cq->pci_addr, cq->pci_size, + &cq->pci_map); + if (ret) { + dev_err(ctrl->dev, "Failed to map CQ %u (err=%d)\n", + cq->qid, ret); + goto err_internal; + } + + if (cq->pci_map.pci_size < cq->pci_size) { + dev_err(ctrl->dev, "Invalid partial mapping of queue %u\n", + cq->qid); + goto err_unmap_queue; + } + + set_bit(NVMET_PCI_EPF_Q_LIVE, &cq->flags); + + if (test_bit(NVMET_PCI_EPF_Q_IRQ_ENABLED, &cq->flags)) + dev_dbg(ctrl->dev, + "CQ[%u]: %u entries of %zu B, IRQ vector %u\n", + cqid, qsize, cq->qes, cq->vector); + else + dev_dbg(ctrl->dev, + "CQ[%u]: %u entries of %zu B, IRQ disabled\n", + cqid, qsize, cq->qes); + + return NVME_SC_SUCCESS; + +err_unmap_queue: + nvmet_pci_epf_mem_unmap(ctrl->nvme_epf, &cq->pci_map); +err_internal: + status = NVME_SC_INTERNAL | NVME_STATUS_DNR; +err: + if (test_and_clear_bit(NVMET_PCI_EPF_Q_IRQ_ENABLED, &cq->flags)) + nvmet_pci_epf_remove_irq_vector(ctrl, cq->vector); + return status; +} + +static u16 nvmet_pci_epf_delete_cq(struct nvmet_ctrl *tctrl, u16 cqid) +{ + struct nvmet_pci_epf_ctrl *ctrl = tctrl->drvdata; + struct nvmet_pci_epf_queue *cq = &ctrl->cq[cqid]; + + if (!test_and_clear_bit(NVMET_PCI_EPF_Q_LIVE, &cq->flags)) + return NVME_SC_QID_INVALID | NVME_STATUS_DNR; + + cancel_delayed_work_sync(&cq->work); + nvmet_pci_epf_drain_queue(cq); + if (test_and_clear_bit(NVMET_PCI_EPF_Q_IRQ_ENABLED, &cq->flags)) + nvmet_pci_epf_remove_irq_vector(ctrl, cq->vector); + nvmet_pci_epf_mem_unmap(ctrl->nvme_epf, &cq->pci_map); + nvmet_cq_put(&cq->nvme_cq); + + return NVME_SC_SUCCESS; +} + +static u16 nvmet_pci_epf_create_sq(struct nvmet_ctrl *tctrl, + u16 sqid, u16 cqid, u16 flags, u16 qsize, u64 pci_addr) +{ + struct nvmet_pci_epf_ctrl *ctrl = tctrl->drvdata; + struct nvmet_pci_epf_queue *sq = &ctrl->sq[sqid]; + struct nvmet_pci_epf_queue *cq = &ctrl->cq[cqid]; + u16 status; + + if (test_bit(NVMET_PCI_EPF_Q_LIVE, &sq->flags)) + return NVME_SC_QID_INVALID | NVME_STATUS_DNR; + + if (!(flags & NVME_QUEUE_PHYS_CONTIG)) + return NVME_SC_INVALID_QUEUE | NVME_STATUS_DNR; + + sq->pci_addr = pci_addr; + sq->qid = sqid; + sq->depth = qsize + 1; + sq->head = 0; + sq->tail = 0; + sq->phase = 0; + sq->db = NVME_REG_DBS + (sqid * 2 * sizeof(u32)); + nvmet_pci_epf_bar_write32(ctrl, sq->db, 0); + if (!sqid) + sq->qes = 1UL << NVME_ADM_SQES; + else + sq->qes = ctrl->io_sqes; + sq->pci_size = sq->qes * sq->depth; + + status = nvmet_sq_create(tctrl, &sq->nvme_sq, &cq->nvme_cq, sqid, + sq->depth); + if (status != NVME_SC_SUCCESS) + return status; + + sq->iod_wq = alloc_workqueue("sq%d_wq", WQ_UNBOUND, + min_t(int, sq->depth, WQ_MAX_ACTIVE), sqid); + if (!sq->iod_wq) { + dev_err(ctrl->dev, "Failed to create SQ %d work queue\n", sqid); + status = NVME_SC_INTERNAL | NVME_STATUS_DNR; + goto out_destroy_sq; + } + + set_bit(NVMET_PCI_EPF_Q_LIVE, &sq->flags); + + dev_dbg(ctrl->dev, "SQ[%u]: %u entries of %zu B\n", + sqid, qsize, sq->qes); + + return NVME_SC_SUCCESS; + +out_destroy_sq: + nvmet_sq_destroy(&sq->nvme_sq); + return status; +} + +static u16 nvmet_pci_epf_delete_sq(struct nvmet_ctrl *tctrl, u16 sqid) +{ + struct nvmet_pci_epf_ctrl *ctrl = tctrl->drvdata; + struct nvmet_pci_epf_queue *sq = &ctrl->sq[sqid]; + + if (!test_and_clear_bit(NVMET_PCI_EPF_Q_LIVE, &sq->flags)) + return NVME_SC_QID_INVALID | NVME_STATUS_DNR; + + destroy_workqueue(sq->iod_wq); + sq->iod_wq = NULL; + + nvmet_pci_epf_drain_queue(sq); + + if (sq->nvme_sq.ctrl) + nvmet_sq_destroy(&sq->nvme_sq); + + return NVME_SC_SUCCESS; +} + +static u16 nvmet_pci_epf_get_feat(const struct nvmet_ctrl *tctrl, + u8 feat, void *data) +{ + struct nvmet_pci_epf_ctrl *ctrl = tctrl->drvdata; + struct nvmet_feat_arbitration *arb; + struct nvmet_feat_irq_coalesce *irqc; + struct nvmet_feat_irq_config *irqcfg; + struct nvmet_pci_epf_irq_vector *iv; + u16 status; + + switch (feat) { + case NVME_FEAT_ARBITRATION: + arb = data; + if (!ctrl->sq_ab) + arb->ab = 0x7; + else + arb->ab = ilog2(ctrl->sq_ab); + return NVME_SC_SUCCESS; + + case NVME_FEAT_IRQ_COALESCE: + irqc = data; + irqc->thr = ctrl->irq_vector_threshold; + irqc->time = 0; + return NVME_SC_SUCCESS; + + case NVME_FEAT_IRQ_CONFIG: + irqcfg = data; + mutex_lock(&ctrl->irq_lock); + iv = nvmet_pci_epf_find_irq_vector(ctrl, irqcfg->iv); + if (iv) { + irqcfg->cd = iv->cd; + status = NVME_SC_SUCCESS; + } else { + status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; + } + mutex_unlock(&ctrl->irq_lock); + return status; + + default: + return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; + } +} + +static u16 nvmet_pci_epf_set_feat(const struct nvmet_ctrl *tctrl, + u8 feat, void *data) +{ + struct nvmet_pci_epf_ctrl *ctrl = tctrl->drvdata; + struct nvmet_feat_arbitration *arb; + struct nvmet_feat_irq_coalesce *irqc; + struct nvmet_feat_irq_config *irqcfg; + struct nvmet_pci_epf_irq_vector *iv; + u16 status; + + switch (feat) { + case NVME_FEAT_ARBITRATION: + arb = data; + if (arb->ab == 0x7) + ctrl->sq_ab = 0; + else + ctrl->sq_ab = 1 << arb->ab; + return NVME_SC_SUCCESS; + + case NVME_FEAT_IRQ_COALESCE: + /* + * Since we do not implement precise IRQ coalescing timing, + * ignore the time field. + */ + irqc = data; + ctrl->irq_vector_threshold = irqc->thr + 1; + return NVME_SC_SUCCESS; + + case NVME_FEAT_IRQ_CONFIG: + irqcfg = data; + mutex_lock(&ctrl->irq_lock); + iv = nvmet_pci_epf_find_irq_vector(ctrl, irqcfg->iv); + if (iv) { + iv->cd = irqcfg->cd; + status = NVME_SC_SUCCESS; + } else { + status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; + } + mutex_unlock(&ctrl->irq_lock); + return status; + + default: + return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; + } +} + +static const struct nvmet_fabrics_ops nvmet_pci_epf_fabrics_ops = { + .owner = THIS_MODULE, + .type = NVMF_TRTYPE_PCI, + .add_port = nvmet_pci_epf_add_port, + .remove_port = nvmet_pci_epf_remove_port, + .queue_response = nvmet_pci_epf_queue_response, + .get_mdts = nvmet_pci_epf_get_mdts, + .create_cq = nvmet_pci_epf_create_cq, + .delete_cq = nvmet_pci_epf_delete_cq, + .create_sq = nvmet_pci_epf_create_sq, + .delete_sq = nvmet_pci_epf_delete_sq, + .get_feature = nvmet_pci_epf_get_feat, + .set_feature = nvmet_pci_epf_set_feat, +}; + +static void nvmet_pci_epf_cq_work(struct work_struct *work); + +static void nvmet_pci_epf_init_queue(struct nvmet_pci_epf_ctrl *ctrl, + unsigned int qid, bool sq) +{ + struct nvmet_pci_epf_queue *queue; + + if (sq) { + queue = &ctrl->sq[qid]; + } else { + queue = &ctrl->cq[qid]; + INIT_DELAYED_WORK(&queue->work, nvmet_pci_epf_cq_work); + } + queue->ctrl = ctrl; + queue->qid = qid; + spin_lock_init(&queue->lock); + INIT_LIST_HEAD(&queue->list); +} + +static int nvmet_pci_epf_alloc_queues(struct nvmet_pci_epf_ctrl *ctrl) +{ + unsigned int qid; + + ctrl->sq = kcalloc(ctrl->nr_queues, + sizeof(struct nvmet_pci_epf_queue), GFP_KERNEL); + if (!ctrl->sq) + return -ENOMEM; + + ctrl->cq = kcalloc(ctrl->nr_queues, + sizeof(struct nvmet_pci_epf_queue), GFP_KERNEL); + if (!ctrl->cq) { + kfree(ctrl->sq); + ctrl->sq = NULL; + return -ENOMEM; + } + + for (qid = 0; qid < ctrl->nr_queues; qid++) { + nvmet_pci_epf_init_queue(ctrl, qid, true); + nvmet_pci_epf_init_queue(ctrl, qid, false); + } + + return 0; +} + +static void nvmet_pci_epf_free_queues(struct nvmet_pci_epf_ctrl *ctrl) +{ + kfree(ctrl->sq); + ctrl->sq = NULL; + kfree(ctrl->cq); + ctrl->cq = NULL; +} + +static void nvmet_pci_epf_exec_iod_work(struct work_struct *work) +{ + struct nvmet_pci_epf_iod *iod = + container_of(work, struct nvmet_pci_epf_iod, work); + struct nvmet_req *req = &iod->req; + int ret; + + if (!iod->ctrl->link_up) { + nvmet_pci_epf_free_iod(iod); + return; + } + + if (!test_bit(NVMET_PCI_EPF_Q_LIVE, &iod->sq->flags)) { + iod->status = NVME_SC_QID_INVALID | NVME_STATUS_DNR; + goto complete; + } + + /* + * If nvmet_req_init() fails (e.g., unsupported opcode) it will call + * __nvmet_req_complete() internally which will call + * nvmet_pci_epf_queue_response() and will complete the command directly. + */ + if (!nvmet_req_init(req, &iod->sq->nvme_sq, &nvmet_pci_epf_fabrics_ops)) + return; + + iod->data_len = nvmet_req_transfer_len(req); + if (iod->data_len) { + /* + * Get the data DMA transfer direction. Here "device" means the + * PCI root-complex host. + */ + if (nvme_is_write(&iod->cmd)) + iod->dma_dir = DMA_FROM_DEVICE; + else + iod->dma_dir = DMA_TO_DEVICE; + + /* + * Setup the command data buffer and get the command data from + * the host if needed. + */ + ret = nvmet_pci_epf_alloc_iod_data_buf(iod); + if (!ret && iod->dma_dir == DMA_FROM_DEVICE) + ret = nvmet_pci_epf_transfer_iod_data(iod); + if (ret) { + nvmet_req_uninit(req); + goto complete; + } + } + + req->execute(req); + + /* + * If we do not have data to transfer after the command execution + * finishes, nvmet_pci_epf_queue_response() will complete the command + * directly. No need to wait for the completion in this case. + */ + if (!iod->data_len || iod->dma_dir != DMA_TO_DEVICE) + return; + + wait_for_completion(&iod->done); + + if (iod->status != NVME_SC_SUCCESS) + return; + + WARN_ON_ONCE(!iod->data_len || iod->dma_dir != DMA_TO_DEVICE); + nvmet_pci_epf_transfer_iod_data(iod); + +complete: + nvmet_pci_epf_complete_iod(iod); +} + +static int nvmet_pci_epf_process_sq(struct nvmet_pci_epf_ctrl *ctrl, + struct nvmet_pci_epf_queue *sq) +{ + struct nvmet_pci_epf_iod *iod; + int ret, n = 0; + u16 head = sq->head; + + sq->tail = nvmet_pci_epf_bar_read32(ctrl, sq->db); + while (head != sq->tail && (!ctrl->sq_ab || n < ctrl->sq_ab)) { + iod = nvmet_pci_epf_alloc_iod(sq); + if (!iod) + break; + + /* Get the NVMe command submitted by the host. */ + ret = nvmet_pci_epf_transfer(ctrl, &iod->cmd, + sq->pci_addr + head * sq->qes, + sq->qes, DMA_FROM_DEVICE); + if (ret) { + /* Not much we can do... */ + nvmet_pci_epf_free_iod(iod); + break; + } + + dev_dbg(ctrl->dev, "SQ[%u]: head %u, tail %u, command %s\n", + sq->qid, head, sq->tail, + nvmet_pci_epf_iod_name(iod)); + + head++; + if (head == sq->depth) + head = 0; + WRITE_ONCE(sq->head, head); + n++; + + queue_work_on(WORK_CPU_UNBOUND, sq->iod_wq, &iod->work); + + sq->tail = nvmet_pci_epf_bar_read32(ctrl, sq->db); + } + + return n; +} + +static void nvmet_pci_epf_poll_sqs_work(struct work_struct *work) +{ + struct nvmet_pci_epf_ctrl *ctrl = + container_of(work, struct nvmet_pci_epf_ctrl, poll_sqs.work); + struct nvmet_pci_epf_queue *sq; + unsigned long limit = jiffies; + unsigned long last = 0; + int i, nr_sqs; + + while (ctrl->link_up && ctrl->enabled) { + nr_sqs = 0; + /* Do round-robin arbitration. */ + for (i = 0; i < ctrl->nr_queues; i++) { + sq = &ctrl->sq[i]; + if (!test_bit(NVMET_PCI_EPF_Q_LIVE, &sq->flags)) + continue; + if (nvmet_pci_epf_process_sq(ctrl, sq)) + nr_sqs++; + } + + /* + * If we have been running for a while, reschedule to let other + * tasks run and to avoid RCU stalls. + */ + if (time_is_before_jiffies(limit + secs_to_jiffies(1))) { + cond_resched(); + limit = jiffies; + continue; + } + + if (nr_sqs) { + last = jiffies; + continue; + } + + /* + * If we have not received any command on any queue for more + * than NVMET_PCI_EPF_SQ_POLL_IDLE, assume we are idle and + * reschedule. This avoids "burning" a CPU when the controller + * is idle for a long time. + */ + if (time_is_before_jiffies(last + NVMET_PCI_EPF_SQ_POLL_IDLE)) + break; + + cpu_relax(); + } + + schedule_delayed_work(&ctrl->poll_sqs, NVMET_PCI_EPF_SQ_POLL_INTERVAL); +} + +static void nvmet_pci_epf_cq_work(struct work_struct *work) +{ + struct nvmet_pci_epf_queue *cq = + container_of(work, struct nvmet_pci_epf_queue, work.work); + struct nvmet_pci_epf_ctrl *ctrl = cq->ctrl; + struct nvme_completion *cqe; + struct nvmet_pci_epf_iod *iod; + unsigned long flags; + int ret = 0, n = 0; + + while (test_bit(NVMET_PCI_EPF_Q_LIVE, &cq->flags) && ctrl->link_up) { + + /* Check that the CQ is not full. */ + cq->head = nvmet_pci_epf_bar_read32(ctrl, cq->db); + if (cq->head == cq->tail + 1) { + ret = -EAGAIN; + break; + } + + spin_lock_irqsave(&cq->lock, flags); + iod = list_first_entry_or_null(&cq->list, + struct nvmet_pci_epf_iod, link); + if (iod) + list_del_init(&iod->link); + spin_unlock_irqrestore(&cq->lock, flags); + + if (!iod) + break; + + /* + * Post the IOD completion entry. If the IOD request was + * executed (req->execute() called), the CQE is already + * initialized. However, the IOD may have been failed before + * that, leaving the CQE not properly initialized. So always + * initialize it here. + */ + cqe = &iod->cqe; + cqe->sq_head = cpu_to_le16(READ_ONCE(iod->sq->head)); + cqe->sq_id = cpu_to_le16(iod->sq->qid); + cqe->command_id = iod->cmd.common.command_id; + cqe->status = cpu_to_le16((iod->status << 1) | cq->phase); + + dev_dbg(ctrl->dev, + "CQ[%u]: %s status 0x%x, result 0x%llx, head %u, tail %u, phase %u\n", + cq->qid, nvmet_pci_epf_iod_name(iod), iod->status, + le64_to_cpu(cqe->result.u64), cq->head, cq->tail, + cq->phase); + + memcpy_toio(cq->pci_map.virt_addr + cq->tail * cq->qes, + cqe, cq->qes); + + cq->tail++; + if (cq->tail >= cq->depth) { + cq->tail = 0; + cq->phase ^= 1; + } + + nvmet_pci_epf_free_iod(iod); + + /* Signal the host. */ + nvmet_pci_epf_raise_irq(ctrl, cq, false); + n++; + } + + /* + * We do not support precise IRQ coalescing time (100ns units as per + * NVMe specifications). So if we have posted completion entries without + * reaching the interrupt coalescing threshold, raise an interrupt. + */ + if (n) + nvmet_pci_epf_raise_irq(ctrl, cq, true); + + if (ret < 0) + queue_delayed_work(system_highpri_wq, &cq->work, + NVMET_PCI_EPF_CQ_RETRY_INTERVAL); +} + +static void nvmet_pci_epf_clear_ctrl_config(struct nvmet_pci_epf_ctrl *ctrl) +{ + struct nvmet_ctrl *tctrl = ctrl->tctrl; + + /* Initialize controller status. */ + tctrl->csts = 0; + ctrl->csts = 0; + nvmet_pci_epf_bar_write32(ctrl, NVME_REG_CSTS, ctrl->csts); + + /* Initialize controller configuration and start polling. */ + tctrl->cc = 0; + ctrl->cc = 0; + nvmet_pci_epf_bar_write32(ctrl, NVME_REG_CC, ctrl->cc); +} + +static int nvmet_pci_epf_enable_ctrl(struct nvmet_pci_epf_ctrl *ctrl) +{ + u64 pci_addr, asq, acq; + u32 aqa; + u16 status, qsize; + + if (ctrl->enabled) + return 0; + + dev_info(ctrl->dev, "Enabling controller\n"); + + ctrl->mps_shift = nvmet_cc_mps(ctrl->cc) + 12; + ctrl->mps = 1UL << ctrl->mps_shift; + ctrl->mps_mask = ctrl->mps - 1; + + ctrl->io_sqes = 1UL << nvmet_cc_iosqes(ctrl->cc); + if (ctrl->io_sqes < sizeof(struct nvme_command)) { + dev_err(ctrl->dev, "Unsupported I/O SQES %zu (need %zu)\n", + ctrl->io_sqes, sizeof(struct nvme_command)); + goto err; + } + + ctrl->io_cqes = 1UL << nvmet_cc_iocqes(ctrl->cc); + if (ctrl->io_cqes < sizeof(struct nvme_completion)) { + dev_err(ctrl->dev, "Unsupported I/O CQES %zu (need %zu)\n", + ctrl->io_cqes, sizeof(struct nvme_completion)); + goto err; + } + + /* Create the admin queue. */ + aqa = nvmet_pci_epf_bar_read32(ctrl, NVME_REG_AQA); + asq = nvmet_pci_epf_bar_read64(ctrl, NVME_REG_ASQ); + acq = nvmet_pci_epf_bar_read64(ctrl, NVME_REG_ACQ); + + qsize = (aqa & 0x0fff0000) >> 16; + pci_addr = acq & GENMASK_ULL(63, 12); + status = nvmet_pci_epf_create_cq(ctrl->tctrl, 0, + NVME_CQ_IRQ_ENABLED | NVME_QUEUE_PHYS_CONTIG, + qsize, pci_addr, 0); + if (status != NVME_SC_SUCCESS) { + dev_err(ctrl->dev, "Failed to create admin completion queue\n"); + goto err; + } + + qsize = aqa & 0x00000fff; + pci_addr = asq & GENMASK_ULL(63, 12); + status = nvmet_pci_epf_create_sq(ctrl->tctrl, 0, 0, + NVME_QUEUE_PHYS_CONTIG, qsize, pci_addr); + if (status != NVME_SC_SUCCESS) { + dev_err(ctrl->dev, "Failed to create admin submission queue\n"); + nvmet_pci_epf_delete_cq(ctrl->tctrl, 0); + goto err; + } + + ctrl->sq_ab = NVMET_PCI_EPF_SQ_AB; + ctrl->irq_vector_threshold = NVMET_PCI_EPF_IV_THRESHOLD; + ctrl->enabled = true; + ctrl->csts = NVME_CSTS_RDY; + + /* Start polling the controller SQs. */ + schedule_delayed_work(&ctrl->poll_sqs, 0); + + return 0; + +err: + nvmet_pci_epf_clear_ctrl_config(ctrl); + return -EINVAL; +} + +static void nvmet_pci_epf_disable_ctrl(struct nvmet_pci_epf_ctrl *ctrl, + bool shutdown) +{ + int qid; + + if (!ctrl->enabled) + return; + + dev_info(ctrl->dev, "%s controller\n", + shutdown ? "Shutting down" : "Disabling"); + + ctrl->enabled = false; + cancel_delayed_work_sync(&ctrl->poll_sqs); + + /* Delete all I/O queues first. */ + for (qid = 1; qid < ctrl->nr_queues; qid++) + nvmet_pci_epf_delete_sq(ctrl->tctrl, qid); + + for (qid = 1; qid < ctrl->nr_queues; qid++) + nvmet_pci_epf_delete_cq(ctrl->tctrl, qid); + + /* Delete the admin queue last. */ + nvmet_pci_epf_delete_sq(ctrl->tctrl, 0); + nvmet_pci_epf_delete_cq(ctrl->tctrl, 0); + + ctrl->csts &= ~NVME_CSTS_RDY; + if (shutdown) { + ctrl->csts |= NVME_CSTS_SHST_CMPLT; + ctrl->cc &= ~NVME_CC_ENABLE; + nvmet_pci_epf_bar_write32(ctrl, NVME_REG_CC, ctrl->cc); + } +} + +static void nvmet_pci_epf_poll_cc_work(struct work_struct *work) +{ + struct nvmet_pci_epf_ctrl *ctrl = + container_of(work, struct nvmet_pci_epf_ctrl, poll_cc.work); + u32 old_cc, new_cc; + int ret; + + if (!ctrl->tctrl) + return; + + old_cc = ctrl->cc; + new_cc = nvmet_pci_epf_bar_read32(ctrl, NVME_REG_CC); + if (new_cc == old_cc) + goto reschedule_work; + + ctrl->cc = new_cc; + + if (nvmet_cc_en(new_cc) && !nvmet_cc_en(old_cc)) { + ret = nvmet_pci_epf_enable_ctrl(ctrl); + if (ret) + goto reschedule_work; + } + + if (!nvmet_cc_en(new_cc) && nvmet_cc_en(old_cc)) + nvmet_pci_epf_disable_ctrl(ctrl, false); + + if (nvmet_cc_shn(new_cc) && !nvmet_cc_shn(old_cc)) + nvmet_pci_epf_disable_ctrl(ctrl, true); + + if (!nvmet_cc_shn(new_cc) && nvmet_cc_shn(old_cc)) + ctrl->csts &= ~NVME_CSTS_SHST_CMPLT; + + nvmet_update_cc(ctrl->tctrl, ctrl->cc); + nvmet_pci_epf_bar_write32(ctrl, NVME_REG_CSTS, ctrl->csts); + +reschedule_work: + schedule_delayed_work(&ctrl->poll_cc, NVMET_PCI_EPF_CC_POLL_INTERVAL); +} + +static void nvmet_pci_epf_init_bar(struct nvmet_pci_epf_ctrl *ctrl) +{ + struct nvmet_ctrl *tctrl = ctrl->tctrl; + + ctrl->bar = ctrl->nvme_epf->reg_bar; + + /* Copy the target controller capabilities as a base. */ + ctrl->cap = tctrl->cap; + + /* Contiguous Queues Required (CQR). */ + ctrl->cap |= 0x1ULL << 16; + + /* Set Doorbell stride to 4B (DSTRB). */ + ctrl->cap &= ~GENMASK_ULL(35, 32); + + /* Clear NVM Subsystem Reset Supported (NSSRS). */ + ctrl->cap &= ~(0x1ULL << 36); + + /* Clear Boot Partition Support (BPS). */ + ctrl->cap &= ~(0x1ULL << 45); + + /* Clear Persistent Memory Region Supported (PMRS). */ + ctrl->cap &= ~(0x1ULL << 56); + + /* Clear Controller Memory Buffer Supported (CMBS). */ + ctrl->cap &= ~(0x1ULL << 57); + + nvmet_pci_epf_bar_write64(ctrl, NVME_REG_CAP, ctrl->cap); + nvmet_pci_epf_bar_write32(ctrl, NVME_REG_VS, tctrl->subsys->ver); + + nvmet_pci_epf_clear_ctrl_config(ctrl); +} + +static int nvmet_pci_epf_create_ctrl(struct nvmet_pci_epf *nvme_epf, + unsigned int max_nr_queues) +{ + struct nvmet_pci_epf_ctrl *ctrl = &nvme_epf->ctrl; + struct nvmet_alloc_ctrl_args args = {}; + char hostnqn[NVMF_NQN_SIZE]; + uuid_t id; + int ret; + + memset(ctrl, 0, sizeof(*ctrl)); + ctrl->dev = &nvme_epf->epf->dev; + mutex_init(&ctrl->irq_lock); + ctrl->nvme_epf = nvme_epf; + ctrl->mdts = nvme_epf->mdts_kb * SZ_1K; + INIT_DELAYED_WORK(&ctrl->poll_cc, nvmet_pci_epf_poll_cc_work); + INIT_DELAYED_WORK(&ctrl->poll_sqs, nvmet_pci_epf_poll_sqs_work); + + ret = mempool_init_kmalloc_pool(&ctrl->iod_pool, + max_nr_queues * NVMET_MAX_QUEUE_SIZE, + sizeof(struct nvmet_pci_epf_iod)); + if (ret) { + dev_err(ctrl->dev, "Failed to initialize IOD mempool\n"); + return ret; + } + + ctrl->port = nvmet_pci_epf_find_port(ctrl, nvme_epf->portid); + if (!ctrl->port) { + dev_err(ctrl->dev, "Port not found\n"); + ret = -EINVAL; + goto out_mempool_exit; + } + + /* Create the target controller. */ + uuid_gen(&id); + snprintf(hostnqn, NVMF_NQN_SIZE, + "nqn.2014-08.org.nvmexpress:uuid:%pUb", &id); + args.port = ctrl->port; + args.subsysnqn = nvme_epf->subsysnqn; + memset(&id, 0, sizeof(uuid_t)); + args.hostid = &id; + args.hostnqn = hostnqn; + args.ops = &nvmet_pci_epf_fabrics_ops; + + ctrl->tctrl = nvmet_alloc_ctrl(&args); + if (!ctrl->tctrl) { + dev_err(ctrl->dev, "Failed to create target controller\n"); + ret = -ENOMEM; + goto out_mempool_exit; + } + ctrl->tctrl->drvdata = ctrl; + + /* We do not support protection information for now. */ + if (ctrl->tctrl->pi_support) { + dev_err(ctrl->dev, + "Protection information (PI) is not supported\n"); + ret = -ENOTSUPP; + goto out_put_ctrl; + } + + /* Allocate our queues, up to the maximum number. */ + ctrl->nr_queues = min(ctrl->tctrl->subsys->max_qid + 1, max_nr_queues); + ret = nvmet_pci_epf_alloc_queues(ctrl); + if (ret) + goto out_put_ctrl; + + /* + * Allocate the IRQ vectors descriptors. We cannot have more than the + * maximum number of queues. + */ + ret = nvmet_pci_epf_alloc_irq_vectors(ctrl); + if (ret) + goto out_free_queues; + + dev_info(ctrl->dev, + "New PCI ctrl \"%s\", %u I/O queues, mdts %u B\n", + ctrl->tctrl->subsys->subsysnqn, ctrl->nr_queues - 1, + ctrl->mdts); + + /* Initialize BAR 0 using the target controller CAP. */ + nvmet_pci_epf_init_bar(ctrl); + + return 0; + +out_free_queues: + nvmet_pci_epf_free_queues(ctrl); +out_put_ctrl: + nvmet_ctrl_put(ctrl->tctrl); + ctrl->tctrl = NULL; +out_mempool_exit: + mempool_exit(&ctrl->iod_pool); + return ret; +} + +static void nvmet_pci_epf_start_ctrl(struct nvmet_pci_epf_ctrl *ctrl) +{ + + dev_info(ctrl->dev, "PCI link up\n"); + ctrl->link_up = true; + + schedule_delayed_work(&ctrl->poll_cc, NVMET_PCI_EPF_CC_POLL_INTERVAL); +} + +static void nvmet_pci_epf_stop_ctrl(struct nvmet_pci_epf_ctrl *ctrl) +{ + dev_info(ctrl->dev, "PCI link down\n"); + ctrl->link_up = false; + + cancel_delayed_work_sync(&ctrl->poll_cc); + + nvmet_pci_epf_disable_ctrl(ctrl, false); + nvmet_pci_epf_clear_ctrl_config(ctrl); +} + +static void nvmet_pci_epf_destroy_ctrl(struct nvmet_pci_epf_ctrl *ctrl) +{ + if (!ctrl->tctrl) + return; + + dev_info(ctrl->dev, "Destroying PCI ctrl \"%s\"\n", + ctrl->tctrl->subsys->subsysnqn); + + nvmet_pci_epf_stop_ctrl(ctrl); + + nvmet_pci_epf_free_queues(ctrl); + nvmet_pci_epf_free_irq_vectors(ctrl); + + nvmet_ctrl_put(ctrl->tctrl); + ctrl->tctrl = NULL; + + mempool_exit(&ctrl->iod_pool); +} + +static int nvmet_pci_epf_configure_bar(struct nvmet_pci_epf *nvme_epf) +{ + struct pci_epf *epf = nvme_epf->epf; + const struct pci_epc_features *epc_features = nvme_epf->epc_features; + size_t reg_size, reg_bar_size; + size_t msix_table_size = 0; + + /* + * The first free BAR will be our register BAR and per NVMe + * specifications, it must be BAR 0. + */ + if (pci_epc_get_first_free_bar(epc_features) != BAR_0) { + dev_err(&epf->dev, "BAR 0 is not free\n"); + return -ENODEV; + } + + /* + * While NVMe PCIe Transport Specification 1.1, section 2.1.10, claims + * that the BAR0 type is Implementation Specific, in NVMe 1.1, the type + * is required to be 64-bit. Thus, for interoperability, always set the + * type to 64-bit. In the rare case that the PCI EPC does not support + * configuring BAR0 as 64-bit, the call to pci_epc_set_bar() will fail, + * and we will return failure back to the user. + */ + epf->bar[BAR_0].flags |= PCI_BASE_ADDRESS_MEM_TYPE_64; + + /* + * Calculate the size of the register bar: NVMe registers first with + * enough space for the doorbells, followed by the MSI-X table + * if supported. + */ + reg_size = NVME_REG_DBS + (NVMET_NR_QUEUES * 2 * sizeof(u32)); + reg_size = ALIGN(reg_size, 8); + + if (epc_features->msix_capable) { + size_t pba_size; + + msix_table_size = PCI_MSIX_ENTRY_SIZE * epf->msix_interrupts; + nvme_epf->msix_table_offset = reg_size; + pba_size = ALIGN(DIV_ROUND_UP(epf->msix_interrupts, 8), 8); + + reg_size += msix_table_size + pba_size; + } + + if (epc_features->bar[BAR_0].type == BAR_FIXED) { + if (reg_size > epc_features->bar[BAR_0].fixed_size) { + dev_err(&epf->dev, + "BAR 0 size %llu B too small, need %zu B\n", + epc_features->bar[BAR_0].fixed_size, + reg_size); + return -ENOMEM; + } + reg_bar_size = epc_features->bar[BAR_0].fixed_size; + } else { + reg_bar_size = ALIGN(reg_size, max(epc_features->align, 4096)); + } + + nvme_epf->reg_bar = pci_epf_alloc_space(epf, reg_bar_size, BAR_0, + epc_features, PRIMARY_INTERFACE); + if (!nvme_epf->reg_bar) { + dev_err(&epf->dev, "Failed to allocate BAR 0\n"); + return -ENOMEM; + } + memset(nvme_epf->reg_bar, 0, reg_bar_size); + + return 0; +} + +static void nvmet_pci_epf_free_bar(struct nvmet_pci_epf *nvme_epf) +{ + struct pci_epf *epf = nvme_epf->epf; + + if (!nvme_epf->reg_bar) + return; + + pci_epf_free_space(epf, nvme_epf->reg_bar, BAR_0, PRIMARY_INTERFACE); + nvme_epf->reg_bar = NULL; +} + +static void nvmet_pci_epf_clear_bar(struct nvmet_pci_epf *nvme_epf) +{ + struct pci_epf *epf = nvme_epf->epf; + + pci_epc_clear_bar(epf->epc, epf->func_no, epf->vfunc_no, + &epf->bar[BAR_0]); +} + +static int nvmet_pci_epf_init_irq(struct nvmet_pci_epf *nvme_epf) +{ + const struct pci_epc_features *epc_features = nvme_epf->epc_features; + struct pci_epf *epf = nvme_epf->epf; + int ret; + + /* Enable MSI-X if supported, otherwise, use MSI. */ + if (epc_features->msix_capable && epf->msix_interrupts) { + ret = pci_epc_set_msix(epf->epc, epf->func_no, epf->vfunc_no, + epf->msix_interrupts, BAR_0, + nvme_epf->msix_table_offset); + if (ret) { + dev_err(&epf->dev, "Failed to configure MSI-X\n"); + return ret; + } + + nvme_epf->nr_vectors = epf->msix_interrupts; + nvme_epf->irq_type = PCI_IRQ_MSIX; + + return 0; + } + + if (epc_features->msi_capable && epf->msi_interrupts) { + ret = pci_epc_set_msi(epf->epc, epf->func_no, epf->vfunc_no, + epf->msi_interrupts); + if (ret) { + dev_err(&epf->dev, "Failed to configure MSI\n"); + return ret; + } + + nvme_epf->nr_vectors = epf->msi_interrupts; + nvme_epf->irq_type = PCI_IRQ_MSI; + + return 0; + } + + /* MSI and MSI-X are not supported: fall back to INTx. */ + nvme_epf->nr_vectors = 1; + nvme_epf->irq_type = PCI_IRQ_INTX; + + return 0; +} + +static int nvmet_pci_epf_epc_init(struct pci_epf *epf) +{ + struct nvmet_pci_epf *nvme_epf = epf_get_drvdata(epf); + const struct pci_epc_features *epc_features = nvme_epf->epc_features; + struct nvmet_pci_epf_ctrl *ctrl = &nvme_epf->ctrl; + unsigned int max_nr_queues = NVMET_NR_QUEUES; + int ret; + + /* For now, do not support virtual functions. */ + if (epf->vfunc_no > 0) { + dev_err(&epf->dev, "Virtual functions are not supported\n"); + return -EINVAL; + } + + /* + * Cap the maximum number of queues we can support on the controller + * with the number of IRQs we can use. + */ + if (epc_features->msix_capable && epf->msix_interrupts) { + dev_info(&epf->dev, + "PCI endpoint controller supports MSI-X, %u vectors\n", + epf->msix_interrupts); + max_nr_queues = min(max_nr_queues, epf->msix_interrupts); + } else if (epc_features->msi_capable && epf->msi_interrupts) { + dev_info(&epf->dev, + "PCI endpoint controller supports MSI, %u vectors\n", + epf->msi_interrupts); + max_nr_queues = min(max_nr_queues, epf->msi_interrupts); + } + + if (max_nr_queues < 2) { + dev_err(&epf->dev, "Invalid maximum number of queues %u\n", + max_nr_queues); + return -EINVAL; + } + + /* Create the target controller. */ + ret = nvmet_pci_epf_create_ctrl(nvme_epf, max_nr_queues); + if (ret) { + dev_err(&epf->dev, + "Failed to create NVMe PCI target controller (err=%d)\n", + ret); + return ret; + } + + nvmet_pci_epf_init_dma(nvme_epf); + + /* Set device ID, class, etc. */ + epf->header->vendorid = ctrl->tctrl->subsys->vendor_id; + epf->header->subsys_vendor_id = ctrl->tctrl->subsys->subsys_vendor_id; + ret = pci_epc_write_header(epf->epc, epf->func_no, epf->vfunc_no, + epf->header); + if (ret) { + dev_err(&epf->dev, + "Failed to write configuration header (err=%d)\n", ret); + goto out_destroy_ctrl; + } + + ret = pci_epc_set_bar(epf->epc, epf->func_no, epf->vfunc_no, + &epf->bar[BAR_0]); + if (ret) { + dev_err(&epf->dev, "Failed to set BAR 0 (err=%d)\n", ret); + goto out_destroy_ctrl; + } + + /* + * Enable interrupts and start polling the controller BAR if we do not + * have a link up notifier. + */ + ret = nvmet_pci_epf_init_irq(nvme_epf); + if (ret) + goto out_clear_bar; + + if (!epc_features->linkup_notifier) + nvmet_pci_epf_start_ctrl(&nvme_epf->ctrl); + + return 0; + +out_clear_bar: + nvmet_pci_epf_clear_bar(nvme_epf); +out_destroy_ctrl: + nvmet_pci_epf_destroy_ctrl(&nvme_epf->ctrl); + return ret; +} + +static void nvmet_pci_epf_epc_deinit(struct pci_epf *epf) +{ + struct nvmet_pci_epf *nvme_epf = epf_get_drvdata(epf); + struct nvmet_pci_epf_ctrl *ctrl = &nvme_epf->ctrl; + + nvmet_pci_epf_destroy_ctrl(ctrl); + + nvmet_pci_epf_deinit_dma(nvme_epf); + nvmet_pci_epf_clear_bar(nvme_epf); +} + +static int nvmet_pci_epf_link_up(struct pci_epf *epf) +{ + struct nvmet_pci_epf *nvme_epf = epf_get_drvdata(epf); + struct nvmet_pci_epf_ctrl *ctrl = &nvme_epf->ctrl; + + nvmet_pci_epf_start_ctrl(ctrl); + + return 0; +} + +static int nvmet_pci_epf_link_down(struct pci_epf *epf) +{ + struct nvmet_pci_epf *nvme_epf = epf_get_drvdata(epf); + struct nvmet_pci_epf_ctrl *ctrl = &nvme_epf->ctrl; + + nvmet_pci_epf_stop_ctrl(ctrl); + + return 0; +} + +static const struct pci_epc_event_ops nvmet_pci_epf_event_ops = { + .epc_init = nvmet_pci_epf_epc_init, + .epc_deinit = nvmet_pci_epf_epc_deinit, + .link_up = nvmet_pci_epf_link_up, + .link_down = nvmet_pci_epf_link_down, +}; + +static int nvmet_pci_epf_bind(struct pci_epf *epf) +{ + struct nvmet_pci_epf *nvme_epf = epf_get_drvdata(epf); + const struct pci_epc_features *epc_features; + struct pci_epc *epc = epf->epc; + int ret; + + if (WARN_ON_ONCE(!epc)) + return -EINVAL; + + epc_features = pci_epc_get_features(epc, epf->func_no, epf->vfunc_no); + if (!epc_features) { + dev_err(&epf->dev, "epc_features not implemented\n"); + return -EOPNOTSUPP; + } + nvme_epf->epc_features = epc_features; + + ret = nvmet_pci_epf_configure_bar(nvme_epf); + if (ret) + return ret; + + return 0; +} + +static void nvmet_pci_epf_unbind(struct pci_epf *epf) +{ + struct nvmet_pci_epf *nvme_epf = epf_get_drvdata(epf); + struct pci_epc *epc = epf->epc; + + nvmet_pci_epf_destroy_ctrl(&nvme_epf->ctrl); + + if (epc->init_complete) { + nvmet_pci_epf_deinit_dma(nvme_epf); + nvmet_pci_epf_clear_bar(nvme_epf); + } + + nvmet_pci_epf_free_bar(nvme_epf); +} + +static struct pci_epf_header nvme_epf_pci_header = { + .vendorid = PCI_ANY_ID, + .deviceid = PCI_ANY_ID, + .progif_code = 0x02, /* NVM Express */ + .baseclass_code = PCI_BASE_CLASS_STORAGE, + .subclass_code = 0x08, /* Non-Volatile Memory controller */ + .interrupt_pin = PCI_INTERRUPT_INTA, +}; + +static int nvmet_pci_epf_probe(struct pci_epf *epf, + const struct pci_epf_device_id *id) +{ + struct nvmet_pci_epf *nvme_epf; + int ret; + + nvme_epf = devm_kzalloc(&epf->dev, sizeof(*nvme_epf), GFP_KERNEL); + if (!nvme_epf) + return -ENOMEM; + + ret = devm_mutex_init(&epf->dev, &nvme_epf->mmio_lock); + if (ret) + return ret; + + nvme_epf->epf = epf; + nvme_epf->mdts_kb = NVMET_PCI_EPF_MDTS_KB; + + epf->event_ops = &nvmet_pci_epf_event_ops; + epf->header = &nvme_epf_pci_header; + epf_set_drvdata(epf, nvme_epf); + + return 0; +} + +#define to_nvme_epf(epf_group) \ + container_of(epf_group, struct nvmet_pci_epf, group) + +static ssize_t nvmet_pci_epf_portid_show(struct config_item *item, char *page) +{ + struct config_group *group = to_config_group(item); + struct nvmet_pci_epf *nvme_epf = to_nvme_epf(group); + + return sysfs_emit(page, "%u\n", le16_to_cpu(nvme_epf->portid)); +} + +static ssize_t nvmet_pci_epf_portid_store(struct config_item *item, + const char *page, size_t len) +{ + struct config_group *group = to_config_group(item); + struct nvmet_pci_epf *nvme_epf = to_nvme_epf(group); + u16 portid; + + /* Do not allow setting this when the function is already started. */ + if (nvme_epf->ctrl.tctrl) + return -EBUSY; + + if (!len) + return -EINVAL; + + if (kstrtou16(page, 0, &portid)) + return -EINVAL; + + nvme_epf->portid = cpu_to_le16(portid); + + return len; +} + +CONFIGFS_ATTR(nvmet_pci_epf_, portid); + +static ssize_t nvmet_pci_epf_subsysnqn_show(struct config_item *item, + char *page) +{ + struct config_group *group = to_config_group(item); + struct nvmet_pci_epf *nvme_epf = to_nvme_epf(group); + + return sysfs_emit(page, "%s\n", nvme_epf->subsysnqn); +} + +static ssize_t nvmet_pci_epf_subsysnqn_store(struct config_item *item, + const char *page, size_t len) +{ + struct config_group *group = to_config_group(item); + struct nvmet_pci_epf *nvme_epf = to_nvme_epf(group); + + /* Do not allow setting this when the function is already started. */ + if (nvme_epf->ctrl.tctrl) + return -EBUSY; + + if (!len) + return -EINVAL; + + strscpy(nvme_epf->subsysnqn, page, len); + + return len; +} + +CONFIGFS_ATTR(nvmet_pci_epf_, subsysnqn); + +static ssize_t nvmet_pci_epf_mdts_kb_show(struct config_item *item, char *page) +{ + struct config_group *group = to_config_group(item); + struct nvmet_pci_epf *nvme_epf = to_nvme_epf(group); + + return sysfs_emit(page, "%u\n", nvme_epf->mdts_kb); +} + +static ssize_t nvmet_pci_epf_mdts_kb_store(struct config_item *item, + const char *page, size_t len) +{ + struct config_group *group = to_config_group(item); + struct nvmet_pci_epf *nvme_epf = to_nvme_epf(group); + unsigned long mdts_kb; + int ret; + + if (nvme_epf->ctrl.tctrl) + return -EBUSY; + + ret = kstrtoul(page, 0, &mdts_kb); + if (ret) + return ret; + if (!mdts_kb) + mdts_kb = NVMET_PCI_EPF_MDTS_KB; + else if (mdts_kb > NVMET_PCI_EPF_MAX_MDTS_KB) + mdts_kb = NVMET_PCI_EPF_MAX_MDTS_KB; + + if (!is_power_of_2(mdts_kb)) + return -EINVAL; + + nvme_epf->mdts_kb = mdts_kb; + + return len; +} + +CONFIGFS_ATTR(nvmet_pci_epf_, mdts_kb); + +static struct configfs_attribute *nvmet_pci_epf_attrs[] = { + &nvmet_pci_epf_attr_portid, + &nvmet_pci_epf_attr_subsysnqn, + &nvmet_pci_epf_attr_mdts_kb, + NULL, +}; + +static const struct config_item_type nvmet_pci_epf_group_type = { + .ct_attrs = nvmet_pci_epf_attrs, + .ct_owner = THIS_MODULE, +}; + +static struct config_group *nvmet_pci_epf_add_cfs(struct pci_epf *epf, + struct config_group *group) +{ + struct nvmet_pci_epf *nvme_epf = epf_get_drvdata(epf); + + config_group_init_type_name(&nvme_epf->group, "nvme", + &nvmet_pci_epf_group_type); + + return &nvme_epf->group; +} + +static const struct pci_epf_device_id nvmet_pci_epf_ids[] = { + { .name = "nvmet_pci_epf" }, + {}, +}; + +static struct pci_epf_ops nvmet_pci_epf_ops = { + .bind = nvmet_pci_epf_bind, + .unbind = nvmet_pci_epf_unbind, + .add_cfs = nvmet_pci_epf_add_cfs, +}; + +static struct pci_epf_driver nvmet_pci_epf_driver = { + .driver.name = "nvmet_pci_epf", + .probe = nvmet_pci_epf_probe, + .id_table = nvmet_pci_epf_ids, + .ops = &nvmet_pci_epf_ops, + .owner = THIS_MODULE, +}; + +static int __init nvmet_pci_epf_init_module(void) +{ + int ret; + + ret = pci_epf_register_driver(&nvmet_pci_epf_driver); + if (ret) + return ret; + + ret = nvmet_register_transport(&nvmet_pci_epf_fabrics_ops); + if (ret) { + pci_epf_unregister_driver(&nvmet_pci_epf_driver); + return ret; + } + + return 0; +} + +static void __exit nvmet_pci_epf_cleanup_module(void) +{ + nvmet_unregister_transport(&nvmet_pci_epf_fabrics_ops); + pci_epf_unregister_driver(&nvmet_pci_epf_driver); +} + +module_init(nvmet_pci_epf_init_module); +module_exit(nvmet_pci_epf_cleanup_module); + +MODULE_DESCRIPTION("NVMe PCI Endpoint Function target driver"); +MODULE_AUTHOR("Damien Le Moal <dlemoal@kernel.org>"); +MODULE_LICENSE("GPL"); diff --git a/drivers/nvme/target/pr.c b/drivers/nvme/target/pr.c new file mode 100644 index 000000000000..cd22d8333314 --- /dev/null +++ b/drivers/nvme/target/pr.c @@ -0,0 +1,1155 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * NVMe over Fabrics Persist Reservation. + * Copyright (c) 2024 Guixin Liu, Alibaba Group. + * All rights reserved. + */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt +#include <linux/unaligned.h> +#include "nvmet.h" + +#define NVMET_PR_NOTIFI_MASK_ALL \ + (1 << NVME_PR_NOTIFY_BIT_REG_PREEMPTED | \ + 1 << NVME_PR_NOTIFY_BIT_RESV_RELEASED | \ + 1 << NVME_PR_NOTIFY_BIT_RESV_PREEMPTED) + +static inline bool nvmet_pr_parse_ignore_key(u32 cdw10) +{ + /* Ignore existing key, bit 03. */ + return (cdw10 >> 3) & 1; +} + +static inline struct nvmet_ns *nvmet_pr_to_ns(struct nvmet_pr *pr) +{ + return container_of(pr, struct nvmet_ns, pr); +} + +static struct nvmet_pr_registrant * +nvmet_pr_find_registrant(struct nvmet_pr *pr, uuid_t *hostid) +{ + struct nvmet_pr_registrant *reg; + + list_for_each_entry_rcu(reg, &pr->registrant_list, entry) { + if (uuid_equal(®->hostid, hostid)) + return reg; + } + return NULL; +} + +u16 nvmet_set_feat_resv_notif_mask(struct nvmet_req *req, u32 mask) +{ + u32 nsid = le32_to_cpu(req->cmd->common.nsid); + struct nvmet_ctrl *ctrl = req->sq->ctrl; + struct nvmet_ns *ns; + unsigned long idx; + u16 status; + + if (mask & ~(NVMET_PR_NOTIFI_MASK_ALL)) { + req->error_loc = offsetof(struct nvme_common_command, cdw11); + return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; + } + + if (nsid != U32_MAX) { + status = nvmet_req_find_ns(req); + if (status) + return status; + if (!req->ns->pr.enable) + return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; + + WRITE_ONCE(req->ns->pr.notify_mask, mask); + goto success; + } + + nvmet_for_each_enabled_ns(&ctrl->subsys->namespaces, idx, ns) { + if (ns->pr.enable) + WRITE_ONCE(ns->pr.notify_mask, mask); + } + +success: + nvmet_set_result(req, mask); + return NVME_SC_SUCCESS; +} + +u16 nvmet_get_feat_resv_notif_mask(struct nvmet_req *req) +{ + u16 status; + + status = nvmet_req_find_ns(req); + if (status) + return status; + + if (!req->ns->pr.enable) + return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; + + nvmet_set_result(req, READ_ONCE(req->ns->pr.notify_mask)); + return status; +} + +void nvmet_execute_get_log_page_resv(struct nvmet_req *req) +{ + struct nvmet_pr_log_mgr *log_mgr = &req->sq->ctrl->pr_log_mgr; + struct nvme_pr_log next_log = {0}; + struct nvme_pr_log log = {0}; + u16 status = NVME_SC_SUCCESS; + u64 lost_count; + u64 cur_count; + u64 next_count; + + mutex_lock(&log_mgr->lock); + if (!kfifo_get(&log_mgr->log_queue, &log)) + goto out; + + /* + * We can't get the last in kfifo. + * Utilize the current count and the count from the next log to + * calculate the number of lost logs, while also addressing cases + * of overflow. If there is no subsequent log, the number of lost + * logs is equal to the lost_count within the nvmet_pr_log_mgr. + */ + cur_count = le64_to_cpu(log.count); + if (kfifo_peek(&log_mgr->log_queue, &next_log)) { + next_count = le64_to_cpu(next_log.count); + if (next_count > cur_count) + lost_count = next_count - cur_count - 1; + else + lost_count = U64_MAX - cur_count + next_count - 1; + } else { + lost_count = log_mgr->lost_count; + } + + log.count = cpu_to_le64((cur_count + lost_count) == 0 ? + 1 : (cur_count + lost_count)); + log_mgr->lost_count -= lost_count; + + log.nr_pages = kfifo_len(&log_mgr->log_queue); + +out: + status = nvmet_copy_to_sgl(req, 0, &log, sizeof(log)); + mutex_unlock(&log_mgr->lock); + nvmet_req_complete(req, status); +} + +static void nvmet_pr_add_resv_log(struct nvmet_ctrl *ctrl, u8 log_type, + u32 nsid) +{ + struct nvmet_pr_log_mgr *log_mgr = &ctrl->pr_log_mgr; + struct nvme_pr_log log = {0}; + + mutex_lock(&log_mgr->lock); + log_mgr->counter++; + if (log_mgr->counter == 0) + log_mgr->counter = 1; + + log.count = cpu_to_le64(log_mgr->counter); + log.type = log_type; + log.nsid = cpu_to_le32(nsid); + + if (!kfifo_put(&log_mgr->log_queue, log)) { + pr_info("a reservation log lost, cntlid:%d, log_type:%d, nsid:%d\n", + ctrl->cntlid, log_type, nsid); + log_mgr->lost_count++; + } + + mutex_unlock(&log_mgr->lock); +} + +static void nvmet_pr_resv_released(struct nvmet_pr *pr, uuid_t *hostid) +{ + struct nvmet_ns *ns = nvmet_pr_to_ns(pr); + struct nvmet_subsys *subsys = ns->subsys; + struct nvmet_ctrl *ctrl; + + if (test_bit(NVME_PR_NOTIFY_BIT_RESV_RELEASED, &pr->notify_mask)) + return; + + mutex_lock(&subsys->lock); + list_for_each_entry(ctrl, &subsys->ctrls, subsys_entry) { + if (!uuid_equal(&ctrl->hostid, hostid) && + nvmet_pr_find_registrant(pr, &ctrl->hostid)) { + nvmet_pr_add_resv_log(ctrl, + NVME_PR_LOG_RESERVATION_RELEASED, ns->nsid); + nvmet_add_async_event(ctrl, NVME_AER_CSS, + NVME_AEN_RESV_LOG_PAGE_AVALIABLE, + NVME_LOG_RESERVATION); + } + } + mutex_unlock(&subsys->lock); +} + +static void nvmet_pr_send_event_to_host(struct nvmet_pr *pr, uuid_t *hostid, + u8 log_type) +{ + struct nvmet_ns *ns = nvmet_pr_to_ns(pr); + struct nvmet_subsys *subsys = ns->subsys; + struct nvmet_ctrl *ctrl; + + mutex_lock(&subsys->lock); + list_for_each_entry(ctrl, &subsys->ctrls, subsys_entry) { + if (uuid_equal(hostid, &ctrl->hostid)) { + nvmet_pr_add_resv_log(ctrl, log_type, ns->nsid); + nvmet_add_async_event(ctrl, NVME_AER_CSS, + NVME_AEN_RESV_LOG_PAGE_AVALIABLE, + NVME_LOG_RESERVATION); + } + } + mutex_unlock(&subsys->lock); +} + +static void nvmet_pr_resv_preempted(struct nvmet_pr *pr, uuid_t *hostid) +{ + if (test_bit(NVME_PR_NOTIFY_BIT_RESV_PREEMPTED, &pr->notify_mask)) + return; + + nvmet_pr_send_event_to_host(pr, hostid, + NVME_PR_LOG_RESERVATOIN_PREEMPTED); +} + +static void nvmet_pr_registration_preempted(struct nvmet_pr *pr, + uuid_t *hostid) +{ + if (test_bit(NVME_PR_NOTIFY_BIT_REG_PREEMPTED, &pr->notify_mask)) + return; + + nvmet_pr_send_event_to_host(pr, hostid, + NVME_PR_LOG_REGISTRATION_PREEMPTED); +} + +static inline void nvmet_pr_set_new_holder(struct nvmet_pr *pr, u8 new_rtype, + struct nvmet_pr_registrant *reg) +{ + reg->rtype = new_rtype; + rcu_assign_pointer(pr->holder, reg); +} + +static u16 nvmet_pr_register(struct nvmet_req *req, + struct nvmet_pr_register_data *d) +{ + struct nvmet_ctrl *ctrl = req->sq->ctrl; + struct nvmet_pr_registrant *new, *reg; + struct nvmet_pr *pr = &req->ns->pr; + u16 status = NVME_SC_SUCCESS; + u64 nrkey = le64_to_cpu(d->nrkey); + + new = kmalloc(sizeof(*new), GFP_KERNEL); + if (!new) + return NVME_SC_INTERNAL; + + down(&pr->pr_sem); + reg = nvmet_pr_find_registrant(pr, &ctrl->hostid); + if (reg) { + if (reg->rkey != nrkey) + status = NVME_SC_RESERVATION_CONFLICT | NVME_STATUS_DNR; + kfree(new); + goto out; + } + + memset(new, 0, sizeof(*new)); + INIT_LIST_HEAD(&new->entry); + new->rkey = nrkey; + uuid_copy(&new->hostid, &ctrl->hostid); + list_add_tail_rcu(&new->entry, &pr->registrant_list); + +out: + up(&pr->pr_sem); + return status; +} + +static void nvmet_pr_unregister_one(struct nvmet_pr *pr, + struct nvmet_pr_registrant *reg) +{ + struct nvmet_pr_registrant *first_reg; + struct nvmet_pr_registrant *holder; + u8 original_rtype; + + list_del_rcu(®->entry); + + holder = rcu_dereference_protected(pr->holder, 1); + if (reg != holder) + goto out; + + original_rtype = holder->rtype; + if (original_rtype == NVME_PR_WRITE_EXCLUSIVE_ALL_REGS || + original_rtype == NVME_PR_EXCLUSIVE_ACCESS_ALL_REGS) { + first_reg = list_first_or_null_rcu(&pr->registrant_list, + struct nvmet_pr_registrant, entry); + if (first_reg) + first_reg->rtype = original_rtype; + rcu_assign_pointer(pr->holder, first_reg); + } else { + rcu_assign_pointer(pr->holder, NULL); + + if (original_rtype == NVME_PR_WRITE_EXCLUSIVE_REG_ONLY || + original_rtype == NVME_PR_EXCLUSIVE_ACCESS_REG_ONLY) + nvmet_pr_resv_released(pr, ®->hostid); + } +out: + kfree_rcu(reg, rcu); +} + +static u16 nvmet_pr_unregister(struct nvmet_req *req, + struct nvmet_pr_register_data *d, + bool ignore_key) +{ + u16 status = NVME_SC_RESERVATION_CONFLICT | NVME_STATUS_DNR; + struct nvmet_ctrl *ctrl = req->sq->ctrl; + struct nvmet_pr *pr = &req->ns->pr; + struct nvmet_pr_registrant *reg; + + down(&pr->pr_sem); + list_for_each_entry_rcu(reg, &pr->registrant_list, entry) { + if (uuid_equal(®->hostid, &ctrl->hostid)) { + if (ignore_key || reg->rkey == le64_to_cpu(d->crkey)) { + status = NVME_SC_SUCCESS; + nvmet_pr_unregister_one(pr, reg); + } + break; + } + } + up(&pr->pr_sem); + + return status; +} + +static void nvmet_pr_update_reg_rkey(struct nvmet_pr_registrant *reg, + void *attr) +{ + reg->rkey = *(u64 *)attr; +} + +static u16 nvmet_pr_update_reg_attr(struct nvmet_pr *pr, + struct nvmet_pr_registrant *reg, + void (*change_attr)(struct nvmet_pr_registrant *reg, + void *attr), + void *attr) +{ + struct nvmet_pr_registrant *holder; + struct nvmet_pr_registrant *new; + + holder = rcu_dereference_protected(pr->holder, 1); + if (reg != holder) { + change_attr(reg, attr); + return NVME_SC_SUCCESS; + } + + new = kmalloc(sizeof(*new), GFP_ATOMIC); + if (!new) + return NVME_SC_INTERNAL; + + new->rkey = holder->rkey; + new->rtype = holder->rtype; + uuid_copy(&new->hostid, &holder->hostid); + INIT_LIST_HEAD(&new->entry); + + change_attr(new, attr); + list_replace_rcu(&holder->entry, &new->entry); + rcu_assign_pointer(pr->holder, new); + kfree_rcu(holder, rcu); + + return NVME_SC_SUCCESS; +} + +static u16 nvmet_pr_replace(struct nvmet_req *req, + struct nvmet_pr_register_data *d, + bool ignore_key) +{ + u16 status = NVME_SC_RESERVATION_CONFLICT | NVME_STATUS_DNR; + struct nvmet_ctrl *ctrl = req->sq->ctrl; + struct nvmet_pr *pr = &req->ns->pr; + struct nvmet_pr_registrant *reg; + u64 nrkey = le64_to_cpu(d->nrkey); + + down(&pr->pr_sem); + list_for_each_entry_rcu(reg, &pr->registrant_list, entry) { + if (uuid_equal(®->hostid, &ctrl->hostid)) { + if (ignore_key || reg->rkey == le64_to_cpu(d->crkey)) + status = nvmet_pr_update_reg_attr(pr, reg, + nvmet_pr_update_reg_rkey, + &nrkey); + break; + } + } + up(&pr->pr_sem); + return status; +} + +static void nvmet_execute_pr_register(struct nvmet_req *req) +{ + u32 cdw10 = le32_to_cpu(req->cmd->common.cdw10); + bool ignore_key = nvmet_pr_parse_ignore_key(cdw10); + struct nvmet_pr_register_data *d; + u8 reg_act = cdw10 & 0x07; /* Reservation Register Action, bit 02:00 */ + u16 status; + + d = kmalloc(sizeof(*d), GFP_KERNEL); + if (!d) { + status = NVME_SC_INTERNAL; + goto out; + } + + status = nvmet_copy_from_sgl(req, 0, d, sizeof(*d)); + if (status) + goto free_data; + + switch (reg_act) { + case NVME_PR_REGISTER_ACT_REG: + status = nvmet_pr_register(req, d); + break; + case NVME_PR_REGISTER_ACT_UNREG: + status = nvmet_pr_unregister(req, d, ignore_key); + break; + case NVME_PR_REGISTER_ACT_REPLACE: + status = nvmet_pr_replace(req, d, ignore_key); + break; + default: + req->error_loc = offsetof(struct nvme_common_command, cdw10); + status = NVME_SC_INVALID_OPCODE | NVME_STATUS_DNR; + break; + } +free_data: + kfree(d); +out: + if (!status) + atomic_inc(&req->ns->pr.generation); + nvmet_req_complete(req, status); +} + +static u16 nvmet_pr_acquire(struct nvmet_req *req, + struct nvmet_pr_registrant *reg, + u8 rtype) +{ + struct nvmet_pr *pr = &req->ns->pr; + struct nvmet_pr_registrant *holder; + + holder = rcu_dereference_protected(pr->holder, 1); + if (holder && reg != holder) + return NVME_SC_RESERVATION_CONFLICT | NVME_STATUS_DNR; + if (holder && reg == holder) { + if (holder->rtype == rtype) + return NVME_SC_SUCCESS; + return NVME_SC_RESERVATION_CONFLICT | NVME_STATUS_DNR; + } + + nvmet_pr_set_new_holder(pr, rtype, reg); + return NVME_SC_SUCCESS; +} + +static void nvmet_pr_confirm_ns_pc_ref(struct percpu_ref *ref) +{ + struct nvmet_pr_per_ctrl_ref *pc_ref = + container_of(ref, struct nvmet_pr_per_ctrl_ref, ref); + + complete(&pc_ref->confirm_done); +} + +static void nvmet_pr_set_ctrl_to_abort(struct nvmet_req *req, uuid_t *hostid) +{ + struct nvmet_pr_per_ctrl_ref *pc_ref; + struct nvmet_ns *ns = req->ns; + unsigned long idx; + + xa_for_each(&ns->pr_per_ctrl_refs, idx, pc_ref) { + if (uuid_equal(&pc_ref->hostid, hostid)) { + percpu_ref_kill_and_confirm(&pc_ref->ref, + nvmet_pr_confirm_ns_pc_ref); + wait_for_completion(&pc_ref->confirm_done); + } + } +} + +static u16 nvmet_pr_unreg_all_host_by_prkey(struct nvmet_req *req, u64 prkey, + uuid_t *send_hostid, + bool abort) +{ + u16 status = NVME_SC_RESERVATION_CONFLICT | NVME_STATUS_DNR; + struct nvmet_pr_registrant *reg, *tmp; + struct nvmet_pr *pr = &req->ns->pr; + uuid_t hostid; + + list_for_each_entry_safe(reg, tmp, &pr->registrant_list, entry) { + if (reg->rkey == prkey) { + status = NVME_SC_SUCCESS; + uuid_copy(&hostid, ®->hostid); + if (abort) + nvmet_pr_set_ctrl_to_abort(req, &hostid); + nvmet_pr_unregister_one(pr, reg); + if (!uuid_equal(&hostid, send_hostid)) + nvmet_pr_registration_preempted(pr, &hostid); + } + } + return status; +} + +static void nvmet_pr_unreg_all_others_by_prkey(struct nvmet_req *req, + u64 prkey, + uuid_t *send_hostid, + bool abort) +{ + struct nvmet_pr_registrant *reg, *tmp; + struct nvmet_pr *pr = &req->ns->pr; + uuid_t hostid; + + list_for_each_entry_safe(reg, tmp, &pr->registrant_list, entry) { + if (reg->rkey == prkey && + !uuid_equal(®->hostid, send_hostid)) { + uuid_copy(&hostid, ®->hostid); + if (abort) + nvmet_pr_set_ctrl_to_abort(req, &hostid); + nvmet_pr_unregister_one(pr, reg); + nvmet_pr_registration_preempted(pr, &hostid); + } + } +} + +static void nvmet_pr_unreg_all_others(struct nvmet_req *req, + uuid_t *send_hostid, + bool abort) +{ + struct nvmet_pr_registrant *reg, *tmp; + struct nvmet_pr *pr = &req->ns->pr; + uuid_t hostid; + + list_for_each_entry_safe(reg, tmp, &pr->registrant_list, entry) { + if (!uuid_equal(®->hostid, send_hostid)) { + uuid_copy(&hostid, ®->hostid); + if (abort) + nvmet_pr_set_ctrl_to_abort(req, &hostid); + nvmet_pr_unregister_one(pr, reg); + nvmet_pr_registration_preempted(pr, &hostid); + } + } +} + +static void nvmet_pr_update_holder_rtype(struct nvmet_pr_registrant *reg, + void *attr) +{ + u8 new_rtype = *(u8 *)attr; + + reg->rtype = new_rtype; +} + +static u16 nvmet_pr_preempt(struct nvmet_req *req, + struct nvmet_pr_registrant *reg, + u8 rtype, + struct nvmet_pr_acquire_data *d, + bool abort) +{ + struct nvmet_ctrl *ctrl = req->sq->ctrl; + struct nvmet_pr *pr = &req->ns->pr; + struct nvmet_pr_registrant *holder; + enum nvme_pr_type original_rtype; + u64 prkey = le64_to_cpu(d->prkey); + u16 status; + + holder = rcu_dereference_protected(pr->holder, 1); + if (!holder) + return nvmet_pr_unreg_all_host_by_prkey(req, prkey, + &ctrl->hostid, abort); + + original_rtype = holder->rtype; + if (original_rtype == NVME_PR_WRITE_EXCLUSIVE_ALL_REGS || + original_rtype == NVME_PR_EXCLUSIVE_ACCESS_ALL_REGS) { + if (!prkey) { + /* + * To prevent possible access from other hosts, and + * avoid terminate the holder, set the new holder + * first before unregistering. + */ + nvmet_pr_set_new_holder(pr, rtype, reg); + nvmet_pr_unreg_all_others(req, &ctrl->hostid, abort); + return NVME_SC_SUCCESS; + } + return nvmet_pr_unreg_all_host_by_prkey(req, prkey, + &ctrl->hostid, abort); + } + + if (holder == reg) { + status = nvmet_pr_update_reg_attr(pr, holder, + nvmet_pr_update_holder_rtype, &rtype); + if (!status && original_rtype != rtype) + nvmet_pr_resv_released(pr, ®->hostid); + return status; + } + + if (prkey == holder->rkey) { + /* + * Same as before, set the new holder first. + */ + nvmet_pr_set_new_holder(pr, rtype, reg); + nvmet_pr_unreg_all_others_by_prkey(req, prkey, &ctrl->hostid, + abort); + if (original_rtype != rtype) + nvmet_pr_resv_released(pr, ®->hostid); + return NVME_SC_SUCCESS; + } + + if (prkey) + return nvmet_pr_unreg_all_host_by_prkey(req, prkey, + &ctrl->hostid, abort); + return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; +} + +static void nvmet_pr_do_abort(struct work_struct *w) +{ + struct nvmet_req *req = container_of(w, struct nvmet_req, r.abort_work); + struct nvmet_pr_per_ctrl_ref *pc_ref; + struct nvmet_ns *ns = req->ns; + unsigned long idx; + + /* + * The target does not support abort, just wait per-controller ref to 0. + */ + xa_for_each(&ns->pr_per_ctrl_refs, idx, pc_ref) { + if (percpu_ref_is_dying(&pc_ref->ref)) { + wait_for_completion(&pc_ref->free_done); + reinit_completion(&pc_ref->confirm_done); + reinit_completion(&pc_ref->free_done); + percpu_ref_resurrect(&pc_ref->ref); + } + } + + up(&ns->pr.pr_sem); + nvmet_req_complete(req, NVME_SC_SUCCESS); +} + +static u16 __nvmet_execute_pr_acquire(struct nvmet_req *req, + struct nvmet_pr_registrant *reg, + u8 acquire_act, + u8 rtype, + struct nvmet_pr_acquire_data *d) +{ + u16 status; + + switch (acquire_act) { + case NVME_PR_ACQUIRE_ACT_ACQUIRE: + status = nvmet_pr_acquire(req, reg, rtype); + goto out; + case NVME_PR_ACQUIRE_ACT_PREEMPT: + status = nvmet_pr_preempt(req, reg, rtype, d, false); + goto inc_gen; + case NVME_PR_ACQUIRE_ACT_PREEMPT_AND_ABORT: + status = nvmet_pr_preempt(req, reg, rtype, d, true); + goto inc_gen; + default: + req->error_loc = offsetof(struct nvme_common_command, cdw10); + status = NVME_SC_INVALID_OPCODE | NVME_STATUS_DNR; + goto out; + } +inc_gen: + if (!status) + atomic_inc(&req->ns->pr.generation); +out: + return status; +} + +static void nvmet_execute_pr_acquire(struct nvmet_req *req) +{ + u32 cdw10 = le32_to_cpu(req->cmd->common.cdw10); + bool ignore_key = nvmet_pr_parse_ignore_key(cdw10); + /* Reservation type, bit 15:08 */ + u8 rtype = (u8)((cdw10 >> 8) & 0xff); + /* Reservation acquire action, bit 02:00 */ + u8 acquire_act = cdw10 & 0x07; + struct nvmet_ctrl *ctrl = req->sq->ctrl; + struct nvmet_pr_acquire_data *d = NULL; + struct nvmet_pr *pr = &req->ns->pr; + struct nvmet_pr_registrant *reg; + u16 status = NVME_SC_SUCCESS; + + if (ignore_key || + rtype < NVME_PR_WRITE_EXCLUSIVE || + rtype > NVME_PR_EXCLUSIVE_ACCESS_ALL_REGS) { + status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; + goto out; + } + + d = kmalloc(sizeof(*d), GFP_KERNEL); + if (!d) { + status = NVME_SC_INTERNAL; + goto out; + } + + status = nvmet_copy_from_sgl(req, 0, d, sizeof(*d)); + if (status) + goto free_data; + + status = NVME_SC_RESERVATION_CONFLICT | NVME_STATUS_DNR; + down(&pr->pr_sem); + list_for_each_entry_rcu(reg, &pr->registrant_list, entry) { + if (uuid_equal(®->hostid, &ctrl->hostid) && + reg->rkey == le64_to_cpu(d->crkey)) { + status = __nvmet_execute_pr_acquire(req, reg, + acquire_act, rtype, d); + break; + } + } + + if (!status && acquire_act == NVME_PR_ACQUIRE_ACT_PREEMPT_AND_ABORT) { + kfree(d); + INIT_WORK(&req->r.abort_work, nvmet_pr_do_abort); + queue_work(nvmet_wq, &req->r.abort_work); + return; + } + + up(&pr->pr_sem); + +free_data: + kfree(d); +out: + nvmet_req_complete(req, status); +} + +static u16 nvmet_pr_release(struct nvmet_req *req, + struct nvmet_pr_registrant *reg, + u8 rtype) +{ + struct nvmet_pr *pr = &req->ns->pr; + struct nvmet_pr_registrant *holder; + u8 original_rtype; + + holder = rcu_dereference_protected(pr->holder, 1); + if (!holder || reg != holder) + return NVME_SC_SUCCESS; + + original_rtype = holder->rtype; + if (original_rtype != rtype) + return NVME_SC_RESERVATION_CONFLICT | NVME_STATUS_DNR; + + rcu_assign_pointer(pr->holder, NULL); + + if (original_rtype != NVME_PR_WRITE_EXCLUSIVE && + original_rtype != NVME_PR_EXCLUSIVE_ACCESS) + nvmet_pr_resv_released(pr, ®->hostid); + + return NVME_SC_SUCCESS; +} + +static void nvmet_pr_clear(struct nvmet_req *req) +{ + struct nvmet_pr_registrant *reg, *tmp; + struct nvmet_pr *pr = &req->ns->pr; + + rcu_assign_pointer(pr->holder, NULL); + + list_for_each_entry_safe(reg, tmp, &pr->registrant_list, entry) { + list_del_rcu(®->entry); + if (!uuid_equal(&req->sq->ctrl->hostid, ®->hostid)) + nvmet_pr_resv_preempted(pr, ®->hostid); + kfree_rcu(reg, rcu); + } + + atomic_inc(&pr->generation); +} + +static u16 __nvmet_execute_pr_release(struct nvmet_req *req, + struct nvmet_pr_registrant *reg, + u8 release_act, u8 rtype) +{ + switch (release_act) { + case NVME_PR_RELEASE_ACT_RELEASE: + return nvmet_pr_release(req, reg, rtype); + case NVME_PR_RELEASE_ACT_CLEAR: + nvmet_pr_clear(req); + return NVME_SC_SUCCESS; + default: + req->error_loc = offsetof(struct nvme_common_command, cdw10); + return NVME_SC_INVALID_OPCODE | NVME_STATUS_DNR; + } +} + +static void nvmet_execute_pr_release(struct nvmet_req *req) +{ + u32 cdw10 = le32_to_cpu(req->cmd->common.cdw10); + bool ignore_key = nvmet_pr_parse_ignore_key(cdw10); + u8 rtype = (u8)((cdw10 >> 8) & 0xff); /* Reservation type, bit 15:08 */ + u8 release_act = cdw10 & 0x07; /* Reservation release action, bit 02:00 */ + struct nvmet_ctrl *ctrl = req->sq->ctrl; + struct nvmet_pr *pr = &req->ns->pr; + struct nvmet_pr_release_data *d; + struct nvmet_pr_registrant *reg; + u16 status; + + if (ignore_key) { + status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; + goto out; + } + + d = kmalloc(sizeof(*d), GFP_KERNEL); + if (!d) { + status = NVME_SC_INTERNAL; + goto out; + } + + status = nvmet_copy_from_sgl(req, 0, d, sizeof(*d)); + if (status) + goto free_data; + + status = NVME_SC_RESERVATION_CONFLICT | NVME_STATUS_DNR; + down(&pr->pr_sem); + list_for_each_entry_rcu(reg, &pr->registrant_list, entry) { + if (uuid_equal(®->hostid, &ctrl->hostid) && + reg->rkey == le64_to_cpu(d->crkey)) { + status = __nvmet_execute_pr_release(req, reg, + release_act, rtype); + break; + } + } + up(&pr->pr_sem); +free_data: + kfree(d); +out: + nvmet_req_complete(req, status); +} + +static void nvmet_execute_pr_report(struct nvmet_req *req) +{ + u32 cdw11 = le32_to_cpu(req->cmd->common.cdw11); + u32 cdw10 = le32_to_cpu(req->cmd->common.cdw10); + u32 num_bytes = 4 * (cdw10 + 1); /* cdw10 is number of dwords */ + u8 eds = cdw11 & 1; /* Extended data structure, bit 00 */ + struct nvme_registered_ctrl_ext *ctrl_eds; + struct nvme_reservation_status_ext *data; + struct nvmet_pr *pr = &req->ns->pr; + struct nvmet_pr_registrant *holder; + struct nvmet_pr_registrant *reg; + u16 num_ctrls = 0; + u16 status; + u8 rtype; + + /* nvmet hostid(uuid_t) is 128 bit. */ + if (!eds) { + req->error_loc = offsetof(struct nvme_common_command, cdw11); + status = NVME_SC_HOST_ID_INCONSIST | NVME_STATUS_DNR; + goto out; + } + + if (num_bytes < sizeof(struct nvme_reservation_status_ext)) { + req->error_loc = offsetof(struct nvme_common_command, cdw10); + status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; + goto out; + } + + data = kzalloc(num_bytes, GFP_KERNEL); + if (!data) { + status = NVME_SC_INTERNAL; + goto out; + } + data->gen = cpu_to_le32(atomic_read(&pr->generation)); + data->ptpls = 0; + ctrl_eds = data->regctl_eds; + + rcu_read_lock(); + holder = rcu_dereference(pr->holder); + rtype = holder ? holder->rtype : 0; + data->rtype = rtype; + + list_for_each_entry_rcu(reg, &pr->registrant_list, entry) { + num_ctrls++; + /* + * continue to get the number of all registrans. + */ + if (((void *)ctrl_eds + sizeof(*ctrl_eds)) > + ((void *)data + num_bytes)) + continue; + /* + * Dynamic controller, set cntlid to 0xffff. + */ + ctrl_eds->cntlid = cpu_to_le16(NVME_CNTLID_DYNAMIC); + if (rtype == NVME_PR_WRITE_EXCLUSIVE_ALL_REGS || + rtype == NVME_PR_EXCLUSIVE_ACCESS_ALL_REGS) + ctrl_eds->rcsts = 1; + if (reg == holder) + ctrl_eds->rcsts = 1; + uuid_copy((uuid_t *)&ctrl_eds->hostid, ®->hostid); + ctrl_eds->rkey = cpu_to_le64(reg->rkey); + ctrl_eds++; + } + rcu_read_unlock(); + + put_unaligned_le16(num_ctrls, data->regctl); + status = nvmet_copy_to_sgl(req, 0, data, num_bytes); + kfree(data); +out: + nvmet_req_complete(req, status); +} + +u16 nvmet_parse_pr_cmd(struct nvmet_req *req) +{ + struct nvme_command *cmd = req->cmd; + + switch (cmd->common.opcode) { + case nvme_cmd_resv_register: + req->execute = nvmet_execute_pr_register; + break; + case nvme_cmd_resv_acquire: + req->execute = nvmet_execute_pr_acquire; + break; + case nvme_cmd_resv_release: + req->execute = nvmet_execute_pr_release; + break; + case nvme_cmd_resv_report: + req->execute = nvmet_execute_pr_report; + break; + default: + return 1; + } + return NVME_SC_SUCCESS; +} + +static bool nvmet_is_req_write_cmd_group(struct nvmet_req *req) +{ + u8 opcode = req->cmd->common.opcode; + + if (req->sq->qid) { + switch (opcode) { + case nvme_cmd_flush: + case nvme_cmd_write: + case nvme_cmd_write_zeroes: + case nvme_cmd_dsm: + case nvme_cmd_zone_append: + case nvme_cmd_zone_mgmt_send: + return true; + default: + return false; + } + } + return false; +} + +static bool nvmet_is_req_read_cmd_group(struct nvmet_req *req) +{ + u8 opcode = req->cmd->common.opcode; + + if (req->sq->qid) { + switch (opcode) { + case nvme_cmd_read: + case nvme_cmd_zone_mgmt_recv: + return true; + default: + return false; + } + } + return false; +} + +u16 nvmet_pr_check_cmd_access(struct nvmet_req *req) +{ + struct nvmet_ctrl *ctrl = req->sq->ctrl; + struct nvmet_pr_registrant *holder; + struct nvmet_ns *ns = req->ns; + struct nvmet_pr *pr = &ns->pr; + u16 status = NVME_SC_SUCCESS; + + rcu_read_lock(); + holder = rcu_dereference(pr->holder); + if (!holder) + goto unlock; + if (uuid_equal(&ctrl->hostid, &holder->hostid)) + goto unlock; + + /* + * The Reservation command group is checked in executing, + * allow it here. + */ + switch (holder->rtype) { + case NVME_PR_WRITE_EXCLUSIVE: + if (nvmet_is_req_write_cmd_group(req)) + status = NVME_SC_RESERVATION_CONFLICT | NVME_STATUS_DNR; + break; + case NVME_PR_EXCLUSIVE_ACCESS: + if (nvmet_is_req_read_cmd_group(req) || + nvmet_is_req_write_cmd_group(req)) + status = NVME_SC_RESERVATION_CONFLICT | NVME_STATUS_DNR; + break; + case NVME_PR_WRITE_EXCLUSIVE_REG_ONLY: + case NVME_PR_WRITE_EXCLUSIVE_ALL_REGS: + if ((nvmet_is_req_write_cmd_group(req)) && + !nvmet_pr_find_registrant(pr, &ctrl->hostid)) + status = NVME_SC_RESERVATION_CONFLICT | NVME_STATUS_DNR; + break; + case NVME_PR_EXCLUSIVE_ACCESS_REG_ONLY: + case NVME_PR_EXCLUSIVE_ACCESS_ALL_REGS: + if ((nvmet_is_req_read_cmd_group(req) || + nvmet_is_req_write_cmd_group(req)) && + !nvmet_pr_find_registrant(pr, &ctrl->hostid)) + status = NVME_SC_RESERVATION_CONFLICT | NVME_STATUS_DNR; + break; + default: + pr_warn("the reservation type is set wrong, type:%d\n", + holder->rtype); + break; + } + +unlock: + rcu_read_unlock(); + if (status) + req->error_loc = offsetof(struct nvme_common_command, opcode); + return status; +} + +u16 nvmet_pr_get_ns_pc_ref(struct nvmet_req *req) +{ + struct nvmet_pr_per_ctrl_ref *pc_ref; + + pc_ref = xa_load(&req->ns->pr_per_ctrl_refs, + req->sq->ctrl->cntlid); + if (unlikely(!percpu_ref_tryget_live(&pc_ref->ref))) + return NVME_SC_INTERNAL; + req->pc_ref = pc_ref; + return NVME_SC_SUCCESS; +} + +static void nvmet_pr_ctrl_ns_all_cmds_done(struct percpu_ref *ref) +{ + struct nvmet_pr_per_ctrl_ref *pc_ref = + container_of(ref, struct nvmet_pr_per_ctrl_ref, ref); + + complete(&pc_ref->free_done); +} + +static int nvmet_pr_alloc_and_insert_pc_ref(struct nvmet_ns *ns, + unsigned long idx, + uuid_t *hostid) +{ + struct nvmet_pr_per_ctrl_ref *pc_ref; + int ret; + + pc_ref = kmalloc(sizeof(*pc_ref), GFP_ATOMIC); + if (!pc_ref) + return -ENOMEM; + + ret = percpu_ref_init(&pc_ref->ref, nvmet_pr_ctrl_ns_all_cmds_done, + PERCPU_REF_ALLOW_REINIT, GFP_KERNEL); + if (ret) + goto free; + + init_completion(&pc_ref->free_done); + init_completion(&pc_ref->confirm_done); + uuid_copy(&pc_ref->hostid, hostid); + + ret = xa_insert(&ns->pr_per_ctrl_refs, idx, pc_ref, GFP_KERNEL); + if (ret) + goto exit; + return ret; +exit: + percpu_ref_exit(&pc_ref->ref); +free: + kfree(pc_ref); + return ret; +} + +int nvmet_ctrl_init_pr(struct nvmet_ctrl *ctrl) +{ + struct nvmet_subsys *subsys = ctrl->subsys; + struct nvmet_pr_per_ctrl_ref *pc_ref; + struct nvmet_ns *ns = NULL; + unsigned long idx; + int ret; + + ctrl->pr_log_mgr.counter = 0; + ctrl->pr_log_mgr.lost_count = 0; + mutex_init(&ctrl->pr_log_mgr.lock); + INIT_KFIFO(ctrl->pr_log_mgr.log_queue); + + /* + * Here we are under subsys lock, if an ns not in subsys->namespaces, + * we can make sure that ns is not enabled, and not call + * nvmet_pr_init_ns(), see more details in nvmet_ns_enable(). + * So just check ns->pr.enable. + */ + nvmet_for_each_enabled_ns(&subsys->namespaces, idx, ns) { + if (ns->pr.enable) { + ret = nvmet_pr_alloc_and_insert_pc_ref(ns, ctrl->cntlid, + &ctrl->hostid); + if (ret) + goto free_per_ctrl_refs; + } + } + return 0; + +free_per_ctrl_refs: + nvmet_for_each_enabled_ns(&subsys->namespaces, idx, ns) { + if (ns->pr.enable) { + pc_ref = xa_erase(&ns->pr_per_ctrl_refs, ctrl->cntlid); + if (pc_ref) + percpu_ref_exit(&pc_ref->ref); + kfree(pc_ref); + } + } + return ret; +} + +void nvmet_ctrl_destroy_pr(struct nvmet_ctrl *ctrl) +{ + struct nvmet_pr_per_ctrl_ref *pc_ref; + struct nvmet_ns *ns; + unsigned long idx; + + kfifo_free(&ctrl->pr_log_mgr.log_queue); + mutex_destroy(&ctrl->pr_log_mgr.lock); + + nvmet_for_each_enabled_ns(&ctrl->subsys->namespaces, idx, ns) { + if (ns->pr.enable) { + pc_ref = xa_erase(&ns->pr_per_ctrl_refs, ctrl->cntlid); + if (pc_ref) + percpu_ref_exit(&pc_ref->ref); + kfree(pc_ref); + } + } +} + +int nvmet_pr_init_ns(struct nvmet_ns *ns) +{ + struct nvmet_subsys *subsys = ns->subsys; + struct nvmet_pr_per_ctrl_ref *pc_ref; + struct nvmet_ctrl *ctrl = NULL; + unsigned long idx; + int ret; + + ns->pr.holder = NULL; + atomic_set(&ns->pr.generation, 0); + sema_init(&ns->pr.pr_sem, 1); + INIT_LIST_HEAD(&ns->pr.registrant_list); + ns->pr.notify_mask = 0; + + xa_init(&ns->pr_per_ctrl_refs); + + list_for_each_entry(ctrl, &subsys->ctrls, subsys_entry) { + ret = nvmet_pr_alloc_and_insert_pc_ref(ns, ctrl->cntlid, + &ctrl->hostid); + if (ret) + goto free_per_ctrl_refs; + } + return 0; + +free_per_ctrl_refs: + xa_for_each(&ns->pr_per_ctrl_refs, idx, pc_ref) { + xa_erase(&ns->pr_per_ctrl_refs, idx); + percpu_ref_exit(&pc_ref->ref); + kfree(pc_ref); + } + return ret; +} + +void nvmet_pr_exit_ns(struct nvmet_ns *ns) +{ + struct nvmet_pr_registrant *reg, *tmp; + struct nvmet_pr_per_ctrl_ref *pc_ref; + struct nvmet_pr *pr = &ns->pr; + unsigned long idx; + + list_for_each_entry_safe(reg, tmp, &pr->registrant_list, entry) { + list_del(®->entry); + kfree(reg); + } + + xa_for_each(&ns->pr_per_ctrl_refs, idx, pc_ref) { + /* + * No command on ns here, we can safely free pc_ref. + */ + pc_ref = xa_erase(&ns->pr_per_ctrl_refs, idx); + percpu_ref_exit(&pc_ref->ref); + kfree(pc_ref); + } + + xa_destroy(&ns->pr_per_ctrl_refs); +} diff --git a/drivers/nvme/target/rdma.c b/drivers/nvme/target/rdma.c index 4597bca43a6d..9c12b2361a6d 100644 --- a/drivers/nvme/target/rdma.c +++ b/drivers/nvme/target/rdma.c @@ -16,7 +16,7 @@ #include <linux/string.h> #include <linux/wait.h> #include <linux/inet.h> -#include <asm/unaligned.h> +#include <linux/unaligned.h> #include <rdma/ib_verbs.h> #include <rdma/rdma_cm.h> @@ -37,6 +37,10 @@ #define NVMET_RDMA_MAX_MDTS 8 #define NVMET_RDMA_MAX_METADATA_MDTS 5 +#define NVMET_RDMA_BACKLOG 128 + +#define NVMET_RDMA_DISCRETE_RSP_TAG -1 + struct nvmet_rdma_srq; struct nvmet_rdma_cmd { @@ -51,7 +55,6 @@ struct nvmet_rdma_cmd { enum { NVMET_RDMA_REQ_INLINE_DATA = (1 << 0), - NVMET_RDMA_REQ_INVALIDATE_RKEY = (1 << 1), }; struct nvmet_rdma_rsp { @@ -74,7 +77,7 @@ struct nvmet_rdma_rsp { u32 invalidate_rkey; struct list_head wait_list; - struct list_head free_list; + int tag; }; enum nvmet_rdma_queue_state { @@ -97,8 +100,7 @@ struct nvmet_rdma_queue { struct nvmet_sq nvme_sq; struct nvmet_rdma_rsp *rsps; - struct list_head free_rsps; - spinlock_t rsps_lock; + struct sbitmap rsp_tags; struct nvmet_rdma_cmd *cmds; struct work_struct release_work; @@ -171,7 +173,8 @@ static void nvmet_rdma_queue_disconnect(struct nvmet_rdma_queue *queue); static void nvmet_rdma_free_rsp(struct nvmet_rdma_device *ndev, struct nvmet_rdma_rsp *r); static int nvmet_rdma_alloc_rsp(struct nvmet_rdma_device *ndev, - struct nvmet_rdma_rsp *r); + struct nvmet_rdma_rsp *r, + int tag); static const struct nvmet_fabrics_ops nvmet_rdma_ops; @@ -209,15 +212,12 @@ static inline bool nvmet_rdma_need_data_out(struct nvmet_rdma_rsp *rsp) static inline struct nvmet_rdma_rsp * nvmet_rdma_get_rsp(struct nvmet_rdma_queue *queue) { - struct nvmet_rdma_rsp *rsp; - unsigned long flags; + struct nvmet_rdma_rsp *rsp = NULL; + int tag; - spin_lock_irqsave(&queue->rsps_lock, flags); - rsp = list_first_entry_or_null(&queue->free_rsps, - struct nvmet_rdma_rsp, free_list); - if (likely(rsp)) - list_del(&rsp->free_list); - spin_unlock_irqrestore(&queue->rsps_lock, flags); + tag = sbitmap_get(&queue->rsp_tags); + if (tag >= 0) + rsp = &queue->rsps[tag]; if (unlikely(!rsp)) { int ret; @@ -225,13 +225,12 @@ nvmet_rdma_get_rsp(struct nvmet_rdma_queue *queue) rsp = kzalloc(sizeof(*rsp), GFP_KERNEL); if (unlikely(!rsp)) return NULL; - ret = nvmet_rdma_alloc_rsp(queue->dev, rsp); + ret = nvmet_rdma_alloc_rsp(queue->dev, rsp, + NVMET_RDMA_DISCRETE_RSP_TAG); if (unlikely(ret)) { kfree(rsp); return NULL; } - - rsp->allocated = true; } return rsp; @@ -240,17 +239,13 @@ nvmet_rdma_get_rsp(struct nvmet_rdma_queue *queue) static inline void nvmet_rdma_put_rsp(struct nvmet_rdma_rsp *rsp) { - unsigned long flags; - - if (unlikely(rsp->allocated)) { + if (unlikely(rsp->tag == NVMET_RDMA_DISCRETE_RSP_TAG)) { nvmet_rdma_free_rsp(rsp->queue->dev, rsp); kfree(rsp); return; } - spin_lock_irqsave(&rsp->queue->rsps_lock, flags); - list_add_tail(&rsp->free_list, &rsp->queue->free_rsps); - spin_unlock_irqrestore(&rsp->queue->rsps_lock, flags); + sbitmap_clear_bit(&rsp->queue->rsp_tags, rsp->tag); } static void nvmet_rdma_free_inline_pages(struct nvmet_rdma_device *ndev, @@ -372,7 +367,7 @@ nvmet_rdma_alloc_cmds(struct nvmet_rdma_device *ndev, struct nvmet_rdma_cmd *cmds; int ret = -EINVAL, i; - cmds = kcalloc(nr_cmds, sizeof(struct nvmet_rdma_cmd), GFP_KERNEL); + cmds = kvcalloc(nr_cmds, sizeof(struct nvmet_rdma_cmd), GFP_KERNEL); if (!cmds) goto out; @@ -387,7 +382,7 @@ nvmet_rdma_alloc_cmds(struct nvmet_rdma_device *ndev, out_free: while (--i >= 0) nvmet_rdma_free_cmd(ndev, cmds + i, admin); - kfree(cmds); + kvfree(cmds); out: return ERR_PTR(ret); } @@ -399,11 +394,11 @@ static void nvmet_rdma_free_cmds(struct nvmet_rdma_device *ndev, for (i = 0; i < nr_cmds; i++) nvmet_rdma_free_cmd(ndev, cmds + i, admin); - kfree(cmds); + kvfree(cmds); } static int nvmet_rdma_alloc_rsp(struct nvmet_rdma_device *ndev, - struct nvmet_rdma_rsp *r) + struct nvmet_rdma_rsp *r, int tag) { /* NVMe CQE / RDMA SEND */ r->req.cqe = kmalloc(sizeof(*r->req.cqe), GFP_KERNEL); @@ -431,6 +426,7 @@ static int nvmet_rdma_alloc_rsp(struct nvmet_rdma_device *ndev, r->read_cqe.done = nvmet_rdma_read_data_done; /* Data Out / RDMA WRITE */ r->write_cqe.done = nvmet_rdma_write_data_done; + r->tag = tag; return 0; @@ -453,33 +449,33 @@ nvmet_rdma_alloc_rsps(struct nvmet_rdma_queue *queue) { struct nvmet_rdma_device *ndev = queue->dev; int nr_rsps = queue->recv_queue_size * 2; - int ret = -EINVAL, i; + int ret = -ENOMEM, i; - queue->rsps = kcalloc(nr_rsps, sizeof(struct nvmet_rdma_rsp), + if (sbitmap_init_node(&queue->rsp_tags, nr_rsps, -1, GFP_KERNEL, + NUMA_NO_NODE, false, true)) + goto out; + + queue->rsps = kvcalloc(nr_rsps, sizeof(struct nvmet_rdma_rsp), GFP_KERNEL); if (!queue->rsps) - goto out; + goto out_free_sbitmap; for (i = 0; i < nr_rsps; i++) { struct nvmet_rdma_rsp *rsp = &queue->rsps[i]; - ret = nvmet_rdma_alloc_rsp(ndev, rsp); + ret = nvmet_rdma_alloc_rsp(ndev, rsp, i); if (ret) goto out_free; - - list_add_tail(&rsp->free_list, &queue->free_rsps); } return 0; out_free: - while (--i >= 0) { - struct nvmet_rdma_rsp *rsp = &queue->rsps[i]; - - list_del(&rsp->free_list); - nvmet_rdma_free_rsp(ndev, rsp); - } - kfree(queue->rsps); + while (--i >= 0) + nvmet_rdma_free_rsp(ndev, &queue->rsps[i]); + kvfree(queue->rsps); +out_free_sbitmap: + sbitmap_free(&queue->rsp_tags); out: return ret; } @@ -489,13 +485,10 @@ static void nvmet_rdma_free_rsps(struct nvmet_rdma_queue *queue) struct nvmet_rdma_device *ndev = queue->dev; int i, nr_rsps = queue->recv_queue_size * 2; - for (i = 0; i < nr_rsps; i++) { - struct nvmet_rdma_rsp *rsp = &queue->rsps[i]; - - list_del(&rsp->free_list); - nvmet_rdma_free_rsp(ndev, rsp); - } - kfree(queue->rsps); + for (i = 0; i < nr_rsps; i++) + nvmet_rdma_free_rsp(ndev, &queue->rsps[i]); + kvfree(queue->rsps); + sbitmap_free(&queue->rsp_tags); } static int nvmet_rdma_post_recv(struct nvmet_rdma_device *ndev, @@ -585,8 +578,8 @@ static void nvmet_rdma_set_sig_domain(struct blk_integrity *bi, if (control & NVME_RW_PRINFO_PRCHK_REF) domain->sig.dif.ref_remap = true; - domain->sig.dif.app_tag = le16_to_cpu(cmd->rw.apptag); - domain->sig.dif.apptag_check_mask = le16_to_cpu(cmd->rw.appmask); + domain->sig.dif.app_tag = le16_to_cpu(cmd->rw.lbat); + domain->sig.dif.apptag_check_mask = le16_to_cpu(cmd->rw.lbatm); domain->sig.dif.app_escape = true; if (pi_type == NVME_NS_DPS_PI_TYPE3) domain->sig.dif.ref_escape = true; @@ -720,7 +713,7 @@ static void nvmet_rdma_queue_response(struct nvmet_req *req) struct rdma_cm_id *cm_id = rsp->queue->cm_id; struct ib_send_wr *first_wr; - if (rsp->flags & NVMET_RDMA_REQ_INVALIDATE_RKEY) { + if (rsp->invalidate_rkey) { rsp->send_wr.opcode = IB_WR_SEND_WITH_INV; rsp->send_wr.ex.invalidate_rkey = rsp->invalidate_rkey; } else { @@ -859,12 +852,12 @@ static u16 nvmet_rdma_map_sgl_inline(struct nvmet_rdma_rsp *rsp) if (!nvme_is_write(rsp->req.cmd)) { rsp->req.error_loc = offsetof(struct nvme_common_command, opcode); - return NVME_SC_INVALID_FIELD | NVME_SC_DNR; + return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; } if (off + len > rsp->queue->dev->inline_data_size) { pr_err("invalid inline data offset!\n"); - return NVME_SC_SGL_INVALID_OFFSET | NVME_SC_DNR; + return NVME_SC_SGL_INVALID_OFFSET | NVME_STATUS_DNR; } /* no data command? */ @@ -903,10 +896,8 @@ static u16 nvmet_rdma_map_sgl_keyed(struct nvmet_rdma_rsp *rsp, goto error_out; rsp->n_rdma += ret; - if (invalidate) { + if (invalidate) rsp->invalidate_rkey = key; - rsp->flags |= NVMET_RDMA_REQ_INVALIDATE_RKEY; - } return 0; @@ -928,7 +919,7 @@ static u16 nvmet_rdma_map_sgl(struct nvmet_rdma_rsp *rsp) pr_err("invalid SGL subtype: %#x\n", sgl->type); rsp->req.error_loc = offsetof(struct nvme_common_command, dptr); - return NVME_SC_INVALID_FIELD | NVME_SC_DNR; + return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; } case NVME_KEY_SGL_FMT_DATA_DESC: switch (sgl->type & 0xf) { @@ -940,12 +931,12 @@ static u16 nvmet_rdma_map_sgl(struct nvmet_rdma_rsp *rsp) pr_err("invalid SGL subtype: %#x\n", sgl->type); rsp->req.error_loc = offsetof(struct nvme_common_command, dptr); - return NVME_SC_INVALID_FIELD | NVME_SC_DNR; + return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; } default: pr_err("invalid SGL type: %#x\n", sgl->type); rsp->req.error_loc = offsetof(struct nvme_common_command, dptr); - return NVME_SC_SGL_INVALID_TYPE | NVME_SC_DNR; + return NVME_SC_SGL_INVALID_TYPE | NVME_STATUS_DNR; } } @@ -985,8 +976,7 @@ static void nvmet_rdma_handle_command(struct nvmet_rdma_queue *queue, cmd->send_sge.addr, cmd->send_sge.length, DMA_TO_DEVICE); - if (!nvmet_req_init(&cmd->req, &queue->nvme_cq, - &queue->nvme_sq, &nvmet_rdma_ops)) + if (!nvmet_req_init(&cmd->req, &queue->nvme_sq, &nvmet_rdma_ops)) return; status = nvmet_rdma_map_sgl(cmd); @@ -1005,6 +995,27 @@ out_err: nvmet_req_complete(&cmd->req, status); } +static bool nvmet_rdma_recv_not_live(struct nvmet_rdma_queue *queue, + struct nvmet_rdma_rsp *rsp) +{ + unsigned long flags; + bool ret = true; + + spin_lock_irqsave(&queue->state_lock, flags); + /* + * recheck queue state is not live to prevent a race condition + * with RDMA_CM_EVENT_ESTABLISHED handler. + */ + if (queue->state == NVMET_RDMA_Q_LIVE) + ret = false; + else if (queue->state == NVMET_RDMA_Q_CONNECTING) + list_add_tail(&rsp->wait_list, &queue->rsp_wait_list); + else + nvmet_rdma_put_rsp(rsp); + spin_unlock_irqrestore(&queue->state_lock, flags); + return ret; +} + static void nvmet_rdma_recv_done(struct ib_cq *cq, struct ib_wc *wc) { struct nvmet_rdma_cmd *cmd = @@ -1045,18 +1056,11 @@ static void nvmet_rdma_recv_done(struct ib_cq *cq, struct ib_wc *wc) rsp->req.cmd = cmd->nvme_cmd; rsp->req.port = queue->port; rsp->n_rdma = 0; + rsp->invalidate_rkey = 0; - if (unlikely(queue->state != NVMET_RDMA_Q_LIVE)) { - unsigned long flags; - - spin_lock_irqsave(&queue->state_lock, flags); - if (queue->state == NVMET_RDMA_Q_CONNECTING) - list_add_tail(&rsp->wait_list, &queue->rsp_wait_list); - else - nvmet_rdma_put_rsp(rsp); - spin_unlock_irqrestore(&queue->state_lock, flags); + if (unlikely(queue->state != NVMET_RDMA_Q_LIVE) && + nvmet_rdma_recv_not_live(queue, rsp)) return; - } nvmet_rdma_handle_command(queue, rsp); } @@ -1348,6 +1352,7 @@ static void nvmet_rdma_free_queue(struct nvmet_rdma_queue *queue) pr_debug("freeing queue %d\n", queue->idx); nvmet_sq_destroy(&queue->nvme_sq); + nvmet_cq_put(&queue->nvme_cq); nvmet_rdma_destroy_queue_ib(queue); if (!queue->nsrq) { @@ -1431,7 +1436,8 @@ nvmet_rdma_alloc_queue(struct nvmet_rdma_device *ndev, goto out_reject; } - ret = nvmet_sq_init(&queue->nvme_sq); + nvmet_cq_init(&queue->nvme_cq); + ret = nvmet_sq_init(&queue->nvme_sq, &queue->nvme_cq); if (ret) { ret = NVME_RDMA_CM_NO_RSC; goto out_free_queue; @@ -1455,8 +1461,6 @@ nvmet_rdma_alloc_queue(struct nvmet_rdma_device *ndev, INIT_LIST_HEAD(&queue->rsp_wait_list); INIT_LIST_HEAD(&queue->rsp_wr_wait_list); spin_lock_init(&queue->rsp_wr_wait_lock); - INIT_LIST_HEAD(&queue->free_rsps); - spin_lock_init(&queue->rsps_lock); INIT_LIST_HEAD(&queue->queue_list); queue->idx = ida_alloc(&nvmet_rdma_queue_ida, GFP_KERNEL); @@ -1514,6 +1518,7 @@ out_ida_remove: out_destroy_sq: nvmet_sq_destroy(&queue->nvme_sq); out_free_queue: + nvmet_cq_put(&queue->nvme_cq); kfree(queue); out_reject: nvmet_rdma_cm_reject(cm_id, ret); @@ -1583,8 +1588,19 @@ static int nvmet_rdma_queue_connect(struct rdma_cm_id *cm_id, } if (queue->host_qid == 0) { - /* Let inflight controller teardown complete */ - flush_workqueue(nvmet_wq); + struct nvmet_rdma_queue *q; + int pending = 0; + + /* Check for pending controller teardown */ + mutex_lock(&nvmet_rdma_queue_mutex); + list_for_each_entry(q, &nvmet_rdma_queue_list, queue_list) { + if (q->nvme_sq.ctrl == queue->nvme_sq.ctrl && + q->state == NVMET_RDMA_Q_DISCONNECTING) + pending++; + } + mutex_unlock(&nvmet_rdma_queue_mutex); + if (pending > NVMET_RDMA_BACKLOG) + return NVME_SC_CONNECT_CTRL_BUSY; } ret = nvmet_rdma_cm_accept(cm_id, queue, &event->param.conn); @@ -1715,7 +1731,7 @@ static void nvmet_rdma_queue_connect_fail(struct rdma_cm_id *cm_id, * We registered an ib_client to handle device removal for queues, * so we only need to handle the listening port cm_ids. In this case * we nullify the priv to prevent double cm_id destruction and destroying - * the cm_id implicitely by returning a non-zero rc to the callout. + * the cm_id implicitly by returning a non-zero rc to the callout. */ static int nvmet_rdma_device_removal(struct rdma_cm_id *cm_id, struct nvmet_rdma_queue *queue) @@ -1726,7 +1742,7 @@ static int nvmet_rdma_device_removal(struct rdma_cm_id *cm_id, /* * This is a queue cm_id. we have registered * an ib_client to handle queues removal - * so don't interfear and just return. + * so don't interfere and just return. */ return 0; } @@ -1744,7 +1760,7 @@ static int nvmet_rdma_device_removal(struct rdma_cm_id *cm_id, /* * We need to return 1 so that the core will destroy - * it's own ID. What a great API design.. + * its own ID. What a great API design.. */ return 1; } @@ -1803,18 +1819,14 @@ static int nvmet_rdma_cm_handler(struct rdma_cm_id *cm_id, static void nvmet_rdma_delete_ctrl(struct nvmet_ctrl *ctrl) { - struct nvmet_rdma_queue *queue; + struct nvmet_rdma_queue *queue, *n; -restart: mutex_lock(&nvmet_rdma_queue_mutex); - list_for_each_entry(queue, &nvmet_rdma_queue_list, queue_list) { - if (queue->nvme_sq.ctrl == ctrl) { - list_del_init(&queue->queue_list); - mutex_unlock(&nvmet_rdma_queue_mutex); - - __nvmet_rdma_queue_disconnect(queue); - goto restart; - } + list_for_each_entry_safe(queue, n, &nvmet_rdma_queue_list, queue_list) { + if (queue->nvme_sq.ctrl != ctrl) + continue; + list_del_init(&queue->queue_list); + __nvmet_rdma_queue_disconnect(queue); } mutex_unlock(&nvmet_rdma_queue_mutex); } @@ -1880,7 +1892,7 @@ static int nvmet_rdma_enable_port(struct nvmet_rdma_port *port) goto out_destroy_id; } - ret = rdma_listen(cm_id, 128); + ret = rdma_listen(cm_id, NVMET_RDMA_BACKLOG); if (ret) { pr_err("listening to %pISpcs failed (%d)\n", addr, ret); goto out_destroy_id; @@ -1943,6 +1955,14 @@ static int nvmet_rdma_add_port(struct nvmet_port *nport) nport->inline_data_size = NVMET_RDMA_MAX_INLINE_DATA_SIZE; } + if (nport->max_queue_size < 0) { + nport->max_queue_size = NVME_RDMA_DEFAULT_QUEUE_SIZE; + } else if (nport->max_queue_size > NVME_RDMA_MAX_QUEUE_SIZE) { + pr_warn("max_queue_size %u is too large, reducing to %u\n", + nport->max_queue_size, NVME_RDMA_MAX_QUEUE_SIZE); + nport->max_queue_size = NVME_RDMA_MAX_QUEUE_SIZE; + } + ret = inet_pton_with_scope(&init_net, af, nport->disc_addr.traddr, nport->disc_addr.trsvcid, &port->addr); if (ret) { @@ -1981,7 +2001,7 @@ static void nvmet_rdma_disc_port_addr(struct nvmet_req *req, struct nvmet_rdma_port *port = nport->priv; struct rdma_cm_id *cm_id = port->cm_id; - if (inet_addr_is_any((struct sockaddr *)&cm_id->route.addr.src_addr)) { + if (inet_addr_is_any(&cm_id->route.addr.src_addr)) { struct nvmet_rdma_rsp *rsp = container_of(req, struct nvmet_rdma_rsp, req); struct rdma_cm_id *req_cm_id = rsp->queue->cm_id; @@ -1993,6 +2013,17 @@ static void nvmet_rdma_disc_port_addr(struct nvmet_req *req, } } +static ssize_t nvmet_rdma_host_port_addr(struct nvmet_ctrl *ctrl, + char *traddr, size_t traddr_len) +{ + struct nvmet_sq *nvme_sq = ctrl->sqs[0]; + struct nvmet_rdma_queue *queue = + container_of(nvme_sq, struct nvmet_rdma_queue, nvme_sq); + + return snprintf(traddr, traddr_len, "%pISc", + (struct sockaddr *)&queue->cm_id->route.addr.dst_addr); +} + static u8 nvmet_rdma_get_mdts(const struct nvmet_ctrl *ctrl) { if (ctrl->pi_support) @@ -2002,6 +2033,8 @@ static u8 nvmet_rdma_get_mdts(const struct nvmet_ctrl *ctrl) static u16 nvmet_rdma_get_max_queue_size(const struct nvmet_ctrl *ctrl) { + if (ctrl->pi_support) + return NVME_RDMA_MAX_METADATA_QUEUE_SIZE; return NVME_RDMA_MAX_QUEUE_SIZE; } @@ -2015,6 +2048,7 @@ static const struct nvmet_fabrics_ops nvmet_rdma_ops = { .queue_response = nvmet_rdma_queue_response, .delete_ctrl = nvmet_rdma_delete_ctrl, .disc_traddr = nvmet_rdma_disc_port_addr, + .host_traddr = nvmet_rdma_host_port_addr, .get_mdts = nvmet_rdma_get_mdts, .get_max_queue_size = nvmet_rdma_get_max_queue_size, }; @@ -2091,5 +2125,6 @@ static void __exit nvmet_rdma_exit(void) module_init(nvmet_rdma_init); module_exit(nvmet_rdma_exit); +MODULE_DESCRIPTION("NVMe target RDMA transport driver"); MODULE_LICENSE("GPL v2"); MODULE_ALIAS("nvmet-transport-1"); /* 1 == NVMF_TRTYPE_RDMA */ diff --git a/drivers/nvme/target/tcp.c b/drivers/nvme/target/tcp.c index cc05c094de22..15416ff0eac4 100644 --- a/drivers/nvme/target/tcp.c +++ b/drivers/nvme/target/tcp.c @@ -7,17 +7,49 @@ #include <linux/module.h> #include <linux/init.h> #include <linux/slab.h> +#include <linux/crc32c.h> #include <linux/err.h> #include <linux/nvme-tcp.h> +#include <linux/nvme-keyring.h> #include <net/sock.h> #include <net/tcp.h> +#include <net/tls.h> +#include <net/tls_prot.h> +#include <net/handshake.h> #include <linux/inet.h> #include <linux/llist.h> -#include <crypto/hash.h> +#include <trace/events/sock.h> #include "nvmet.h" #define NVMET_TCP_DEF_INLINE_DATA_SIZE (4 * PAGE_SIZE) +#define NVMET_TCP_MAXH2CDATA 0x400000 /* 16M arbitrary limit */ +#define NVMET_TCP_BACKLOG 128 + +static int param_store_val(const char *str, int *val, int min, int max) +{ + int ret, new_val; + + ret = kstrtoint(str, 10, &new_val); + if (ret) + return -EINVAL; + + if (new_val < min || new_val > max) + return -EINVAL; + + *val = new_val; + return 0; +} + +static int set_params(const char *str, const struct kernel_param *kp) +{ + return param_store_val(str, kp->arg, 0, INT_MAX); +} + +static const struct kernel_param_ops set_param_ops = { + .set = set_params, + .get = param_get_int, +}; /* Define the socket priority to use for connections were it is desirable * that the NIC consider performing optimized packet processing or filtering. @@ -26,8 +58,8 @@ * values that may be unique for some NIC implementations. */ static int so_priority; -module_param(so_priority, int, 0644); -MODULE_PARM_DESC(so_priority, "nvmet tcp socket optimize priority"); +device_param_cb(so_priority, &set_param_ops, &so_priority, 0644); +MODULE_PARM_DESC(so_priority, "nvmet tcp socket optimize priority: Default 0"); /* Define a time period (in usecs) that io_work() shall sample an activated * queue before determining it to be idle. This optional module behavior @@ -35,9 +67,20 @@ MODULE_PARM_DESC(so_priority, "nvmet tcp socket optimize priority"); * using advanced interrupt moderation techniques. */ static int idle_poll_period_usecs; -module_param(idle_poll_period_usecs, int, 0644); +device_param_cb(idle_poll_period_usecs, &set_param_ops, + &idle_poll_period_usecs, 0644); MODULE_PARM_DESC(idle_poll_period_usecs, - "nvmet tcp io_work poll till idle time period in usecs"); + "nvmet tcp io_work poll till idle time period in usecs: Default 0"); + +#ifdef CONFIG_NVME_TARGET_TCP_TLS +/* + * TLS handshake timeout + */ +static int tls_handshake_timeout = 10; +module_param(tls_handshake_timeout, int, 0644); +MODULE_PARM_DESC(tls_handshake_timeout, + "nvme TLS handshake timeout in seconds (default 10)"); +#endif #define NVMET_TCP_RECV_BUDGET 8 #define NVMET_TCP_SEND_BUDGET 8 @@ -77,6 +120,7 @@ struct nvmet_tcp_cmd { u32 pdu_len; u32 pdu_recv; int sg_idx; + char recv_cbuf[CMSG_LEN(sizeof(char))]; struct msghdr recv_msg; struct bio_vec *iov; u32 flags; @@ -95,8 +139,10 @@ struct nvmet_tcp_cmd { enum nvmet_tcp_queue_state { NVMET_TCP_Q_CONNECTING, + NVMET_TCP_Q_TLS_HANDSHAKE, NVMET_TCP_Q_LIVE, NVMET_TCP_Q_DISCONNECTING, + NVMET_TCP_Q_FAILED, }; struct nvmet_tcp_queue { @@ -105,6 +151,7 @@ struct nvmet_tcp_queue { struct work_struct io_work; struct nvmet_cq nvme_cq; struct nvmet_sq nvme_sq; + struct kref kref; /* send state */ struct nvmet_tcp_cmd *cmds; @@ -125,8 +172,10 @@ struct nvmet_tcp_queue { /* digest state */ bool hdr_digest; bool data_digest; - struct ahash_request *snd_hash; - struct ahash_request *rcv_hash; + + /* TLS state */ + key_serial_t tls_pskid; + struct delayed_work tls_handshake_tmo_work; unsigned long poll_end; @@ -243,14 +292,9 @@ static inline u8 nvmet_tcp_ddgst_len(struct nvmet_tcp_queue *queue) return queue->data_digest ? NVME_TCP_DIGEST_LENGTH : 0; } -static inline void nvmet_tcp_hdgst(struct ahash_request *hash, - void *pdu, size_t len) +static inline void nvmet_tcp_hdgst(void *pdu, size_t len) { - struct scatterlist sg; - - sg_init_one(&sg, pdu, len); - ahash_request_set_crypt(hash, &sg, pdu + len, len); - crypto_ahash_digest(hash); + put_unaligned_le32(~crc32c(~0, pdu, len), pdu + len); } static int nvmet_tcp_verify_hdgst(struct nvmet_tcp_queue *queue, @@ -267,7 +311,7 @@ static int nvmet_tcp_verify_hdgst(struct nvmet_tcp_queue *queue, } recv_digest = *(__le32 *)(pdu + hdr->hlen); - nvmet_tcp_hdgst(queue->rcv_hash, pdu, len); + nvmet_tcp_hdgst(pdu, len); exp_digest = *(__le32 *)(pdu + hdr->hlen); if (recv_digest != exp_digest) { pr_err("queue %d: header digest error: recv %#x expected %#x\n", @@ -296,6 +340,7 @@ static int nvmet_tcp_check_ddgst(struct nvmet_tcp_queue *queue, void *pdu) return 0; } +/* If cmd buffers are NULL, no operation is performed */ static void nvmet_tcp_free_cmd_buffers(struct nvmet_tcp_cmd *cmd) { kfree(cmd->iov); @@ -321,9 +366,8 @@ static void nvmet_tcp_build_pdu_iovec(struct nvmet_tcp_cmd *cmd) while (length) { u32 iov_len = min_t(u32, length, sg->length - sg_offset); - iov->bv_page = sg_page(sg); - iov->bv_len = sg->length; - iov->bv_offset = sg->offset + sg_offset; + bvec_set_page(iov, sg_page(sg), iov_len, + sg->offset + sg_offset); length -= iov_len; sg = sg_next(sg); @@ -346,6 +390,7 @@ static void nvmet_tcp_fatal_error(struct nvmet_tcp_queue *queue) static void nvmet_tcp_socket_error(struct nvmet_tcp_queue *queue, int status) { + queue->rcv_state = NVMET_TCP_RECV_ERR; if (status == -EPIPE || status == -ECONNRESET) kernel_sock_shutdown(queue->sock, SHUT_RDWR); else @@ -363,10 +408,10 @@ static int nvmet_tcp_map_data(struct nvmet_tcp_cmd *cmd) if (sgl->type == ((NVME_SGL_FMT_DATA_DESC << 4) | NVME_SGL_FMT_OFFSET)) { if (!nvme_is_write(cmd->req.cmd)) - return NVME_SC_INVALID_FIELD | NVME_SC_DNR; + return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; if (len > cmd->req.port->inline_data_size) - return NVME_SC_SGL_INVALID_OFFSET | NVME_SC_DNR; + return NVME_SC_SGL_INVALID_OFFSET | NVME_STATUS_DNR; cmd->pdu_len = len; } cmd->req.transfer_len += len; @@ -389,12 +434,24 @@ err: return NVME_SC_INTERNAL; } -static void nvmet_tcp_calc_ddgst(struct ahash_request *hash, - struct nvmet_tcp_cmd *cmd) +static void nvmet_tcp_calc_ddgst(struct nvmet_tcp_cmd *cmd) { - ahash_request_set_crypt(hash, cmd->req.sg, - (void *)&cmd->exp_ddgst, cmd->req.transfer_len); - crypto_ahash_digest(hash); + size_t total_len = cmd->req.transfer_len; + struct scatterlist *sg = cmd->req.sg; + u32 crc = ~0; + + while (total_len) { + size_t len = min_t(size_t, total_len, sg->length); + + /* + * Note that the scatterlist does not contain any highmem pages, + * as it was allocated by sgl_alloc() with GFP_KERNEL. + */ + crc = crc32c(crc, sg_virt(sg), len); + total_len -= len; + sg = sg_next(sg); + } + cmd->exp_ddgst = cpu_to_le32(~crc); } static void nvmet_setup_c2h_data_pdu(struct nvmet_tcp_cmd *cmd) @@ -421,19 +478,18 @@ static void nvmet_setup_c2h_data_pdu(struct nvmet_tcp_cmd *cmd) if (queue->data_digest) { pdu->hdr.flags |= NVME_TCP_F_DDGST; - nvmet_tcp_calc_ddgst(queue->snd_hash, cmd); + nvmet_tcp_calc_ddgst(cmd); } if (cmd->queue->hdr_digest) { pdu->hdr.flags |= NVME_TCP_F_HDGST; - nvmet_tcp_hdgst(queue->snd_hash, pdu, sizeof(*pdu)); + nvmet_tcp_hdgst(pdu, sizeof(*pdu)); } } static void nvmet_setup_r2t_pdu(struct nvmet_tcp_cmd *cmd) { struct nvme_tcp_r2t_pdu *pdu = cmd->r2t_pdu; - struct nvmet_tcp_queue *queue = cmd->queue; u8 hdgst = nvmet_tcp_hdgst_len(cmd->queue); cmd->offset = 0; @@ -451,14 +507,13 @@ static void nvmet_setup_r2t_pdu(struct nvmet_tcp_cmd *cmd) pdu->r2t_offset = cpu_to_le32(cmd->rbytes_done); if (cmd->queue->hdr_digest) { pdu->hdr.flags |= NVME_TCP_F_HDGST; - nvmet_tcp_hdgst(queue->snd_hash, pdu, sizeof(*pdu)); + nvmet_tcp_hdgst(pdu, sizeof(*pdu)); } } static void nvmet_setup_response_pdu(struct nvmet_tcp_cmd *cmd) { struct nvme_tcp_rsp_pdu *pdu = cmd->rsp_pdu; - struct nvmet_tcp_queue *queue = cmd->queue; u8 hdgst = nvmet_tcp_hdgst_len(cmd->queue); cmd->offset = 0; @@ -471,7 +526,7 @@ static void nvmet_setup_response_pdu(struct nvmet_tcp_cmd *cmd) pdu->hdr.plen = cpu_to_le32(pdu->hdr.hlen + hdgst); if (cmd->queue->hdr_digest) { pdu->hdr.flags |= NVME_TCP_F_HDGST; - nvmet_tcp_hdgst(queue->snd_hash, pdu, sizeof(*pdu)); + nvmet_tcp_hdgst(pdu, sizeof(*pdu)); } } @@ -518,10 +573,16 @@ static void nvmet_tcp_queue_response(struct nvmet_req *req) struct nvmet_tcp_cmd *cmd = container_of(req, struct nvmet_tcp_cmd, req); struct nvmet_tcp_queue *queue = cmd->queue; + enum nvmet_tcp_recv_state queue_state; + struct nvmet_tcp_cmd *queue_cmd; struct nvme_sgl_desc *sgl; u32 len; - if (unlikely(cmd == queue->cmd)) { + /* Pairs with store_release in nvmet_prepare_receive_pdu() */ + queue_state = smp_load_acquire(&queue->rcv_state); + queue_cmd = READ_ONCE(queue->cmd); + + if (unlikely(cmd == queue_cmd)) { sgl = &cmd->req.cmd->common.dptr.sgl; len = le32_to_cpu(sgl->length); @@ -530,7 +591,7 @@ static void nvmet_tcp_queue_response(struct nvmet_req *req) * Avoid using helpers, this might happen before * nvmet_req_init is completed. */ - if (queue->rcv_state == NVMET_TCP_RECV_PDU && + if (queue_state == NVMET_TCP_RECV_PDU && len && len <= cmd->req.port->inline_data_size && nvme_is_write(cmd->req.cmd)) return; @@ -550,13 +611,17 @@ static void nvmet_tcp_execute_request(struct nvmet_tcp_cmd *cmd) static int nvmet_try_send_data_pdu(struct nvmet_tcp_cmd *cmd) { + struct msghdr msg = { + .msg_flags = MSG_DONTWAIT | MSG_MORE | MSG_SPLICE_PAGES, + }; + struct bio_vec bvec; u8 hdgst = nvmet_tcp_hdgst_len(cmd->queue); int left = sizeof(*cmd->data_pdu) - cmd->offset + hdgst; int ret; - ret = kernel_sendpage(cmd->queue->sock, virt_to_page(cmd->data_pdu), - offset_in_page(cmd->data_pdu) + cmd->offset, - left, MSG_DONTWAIT | MSG_MORE | MSG_SENDPAGE_NOTLAST); + bvec_set_virt(&bvec, (void *)cmd->data_pdu + cmd->offset, left); + iov_iter_bvec(&msg.msg_iter, ITER_SOURCE, &bvec, 1, left); + ret = sock_sendmsg(cmd->queue->sock, &msg); if (ret <= 0) return ret; @@ -577,17 +642,21 @@ static int nvmet_try_send_data(struct nvmet_tcp_cmd *cmd, bool last_in_batch) int ret; while (cmd->cur_sg) { + struct msghdr msg = { + .msg_flags = MSG_DONTWAIT | MSG_SPLICE_PAGES, + }; struct page *page = sg_page(cmd->cur_sg); + struct bio_vec bvec; u32 left = cmd->cur_sg->length - cmd->offset; - int flags = MSG_DONTWAIT; if ((!last_in_batch && cmd->queue->send_list_len) || cmd->wbytes_done + left < cmd->req.transfer_len || queue->data_digest || !queue->nvme_sq.sqhd_disabled) - flags |= MSG_MORE | MSG_SENDPAGE_NOTLAST; + msg.msg_flags |= MSG_MORE; - ret = kernel_sendpage(cmd->queue->sock, page, cmd->offset, - left, flags); + bvec_set_page(&bvec, page, left, cmd->offset); + iov_iter_bvec(&msg.msg_iter, ITER_SOURCE, &bvec, 1, left); + ret = sock_sendmsg(cmd->queue->sock, &msg); if (ret <= 0) return ret; @@ -623,18 +692,20 @@ static int nvmet_try_send_data(struct nvmet_tcp_cmd *cmd, bool last_in_batch) static int nvmet_try_send_response(struct nvmet_tcp_cmd *cmd, bool last_in_batch) { + struct msghdr msg = { .msg_flags = MSG_DONTWAIT | MSG_SPLICE_PAGES, }; + struct bio_vec bvec; u8 hdgst = nvmet_tcp_hdgst_len(cmd->queue); int left = sizeof(*cmd->rsp_pdu) - cmd->offset + hdgst; - int flags = MSG_DONTWAIT; int ret; if (!last_in_batch && cmd->queue->send_list_len) - flags |= MSG_MORE | MSG_SENDPAGE_NOTLAST; + msg.msg_flags |= MSG_MORE; else - flags |= MSG_EOR; + msg.msg_flags |= MSG_EOR; - ret = kernel_sendpage(cmd->queue->sock, virt_to_page(cmd->rsp_pdu), - offset_in_page(cmd->rsp_pdu) + cmd->offset, left, flags); + bvec_set_virt(&bvec, (void *)cmd->rsp_pdu + cmd->offset, left); + iov_iter_bvec(&msg.msg_iter, ITER_SOURCE, &bvec, 1, left); + ret = sock_sendmsg(cmd->queue->sock, &msg); if (ret <= 0) return ret; cmd->offset += ret; @@ -651,18 +722,20 @@ static int nvmet_try_send_response(struct nvmet_tcp_cmd *cmd, static int nvmet_try_send_r2t(struct nvmet_tcp_cmd *cmd, bool last_in_batch) { + struct msghdr msg = { .msg_flags = MSG_DONTWAIT | MSG_SPLICE_PAGES, }; + struct bio_vec bvec; u8 hdgst = nvmet_tcp_hdgst_len(cmd->queue); int left = sizeof(*cmd->r2t_pdu) - cmd->offset + hdgst; - int flags = MSG_DONTWAIT; int ret; if (!last_in_batch && cmd->queue->send_list_len) - flags |= MSG_MORE | MSG_SENDPAGE_NOTLAST; + msg.msg_flags |= MSG_MORE; else - flags |= MSG_EOR; + msg.msg_flags |= MSG_EOR; - ret = kernel_sendpage(cmd->queue->sock, virt_to_page(cmd->r2t_pdu), - offset_in_page(cmd->r2t_pdu) + cmd->offset, left, flags); + bvec_set_virt(&bvec, (void *)cmd->r2t_pdu + cmd->offset, left); + iov_iter_bvec(&msg.msg_iter, ITER_SOURCE, &bvec, 1, left); + ret = sock_sendmsg(cmd->queue->sock, &msg); if (ret <= 0) return ret; cmd->offset += ret; @@ -782,46 +855,11 @@ static void nvmet_prepare_receive_pdu(struct nvmet_tcp_queue *queue) { queue->offset = 0; queue->left = sizeof(struct nvme_tcp_hdr); - queue->cmd = NULL; - queue->rcv_state = NVMET_TCP_RECV_PDU; -} - -static void nvmet_tcp_free_crypto(struct nvmet_tcp_queue *queue) -{ - struct crypto_ahash *tfm = crypto_ahash_reqtfm(queue->rcv_hash); - - ahash_request_free(queue->rcv_hash); - ahash_request_free(queue->snd_hash); - crypto_free_ahash(tfm); -} - -static int nvmet_tcp_alloc_crypto(struct nvmet_tcp_queue *queue) -{ - struct crypto_ahash *tfm; - - tfm = crypto_alloc_ahash("crc32c", 0, CRYPTO_ALG_ASYNC); - if (IS_ERR(tfm)) - return PTR_ERR(tfm); - - queue->snd_hash = ahash_request_alloc(tfm, GFP_KERNEL); - if (!queue->snd_hash) - goto free_tfm; - ahash_request_set_callback(queue->snd_hash, 0, NULL, NULL); - - queue->rcv_hash = ahash_request_alloc(tfm, GFP_KERNEL); - if (!queue->rcv_hash) - goto free_snd_hash; - ahash_request_set_callback(queue->rcv_hash, 0, NULL, NULL); - - return 0; -free_snd_hash: - ahash_request_free(queue->snd_hash); -free_tfm: - crypto_free_ahash(tfm); - return -ENOMEM; + WRITE_ONCE(queue->cmd, NULL); + /* Ensure rcv_state is visible only after queue->cmd is set */ + smp_store_release(&queue->rcv_state, NVMET_TCP_RECV_PDU); } - static int nvmet_tcp_handle_icreq(struct nvmet_tcp_queue *queue) { struct nvme_tcp_icreq_pdu *icreq = &queue->pdu.icreq; @@ -834,6 +872,7 @@ static int nvmet_tcp_handle_icreq(struct nvmet_tcp_queue *queue) pr_err("bad nvme-tcp pdu length (%d)\n", le32_to_cpu(icreq->hdr.plen)); nvmet_tcp_fatal_error(queue); + return -EPROTO; } if (icreq->pfv != NVME_TCP_PFV_1_0) { @@ -849,11 +888,6 @@ static int nvmet_tcp_handle_icreq(struct nvmet_tcp_queue *queue) queue->hdr_digest = !!(icreq->digest & NVME_TCP_HDR_DIGEST_ENABLE); queue->data_digest = !!(icreq->digest & NVME_TCP_DATA_DIGEST_ENABLE); - if (queue->hdr_digest || queue->data_digest) { - ret = nvmet_tcp_alloc_crypto(queue); - if (ret) - return ret; - } memset(icresp, 0, sizeof(*icresp)); icresp->hdr.type = nvme_tcp_icresp; @@ -861,7 +895,7 @@ static int nvmet_tcp_handle_icreq(struct nvmet_tcp_queue *queue) icresp->hdr.pdo = 0; icresp->hdr.plen = cpu_to_le32(icresp->hdr.hlen); icresp->pfv = cpu_to_le16(NVME_TCP_PFV_1_0); - icresp->maxdata = cpu_to_le32(0x400000); /* 16M arbitrary limit */ + icresp->maxdata = cpu_to_le32(NVMET_TCP_MAXH2CDATA); icresp->cpda = 0; if (queue->hdr_digest) icresp->digest |= NVME_TCP_HDR_DIGEST_ENABLE; @@ -871,16 +905,14 @@ static int nvmet_tcp_handle_icreq(struct nvmet_tcp_queue *queue) iov.iov_base = icresp; iov.iov_len = sizeof(*icresp); ret = kernel_sendmsg(queue->sock, &msg, &iov, 1, iov.iov_len); - if (ret < 0) - goto free_crypto; + if (ret < 0) { + queue->state = NVMET_TCP_Q_FAILED; + return ret; /* queue removal will cleanup */ + } queue->state = NVMET_TCP_Q_LIVE; nvmet_prepare_receive_pdu(queue); return 0; -free_crypto: - if (queue->hdr_digest || queue->data_digest) - nvmet_tcp_free_crypto(queue); - return ret; } static void nvmet_tcp_handle_req_failure(struct nvmet_tcp_queue *queue, @@ -918,13 +950,13 @@ static int nvmet_tcp_handle_h2c_data_pdu(struct nvmet_tcp_queue *queue) { struct nvme_tcp_data_pdu *data = &queue->pdu.data; struct nvmet_tcp_cmd *cmd; + unsigned int exp_data_len; if (likely(queue->nr_cmds)) { if (unlikely(data->ttag >= queue->nr_cmds)) { pr_err("queue %d: received out of bound ttag %u, nr_cmds %u\n", queue->idx, data->ttag, queue->nr_cmds); - nvmet_tcp_fatal_error(queue); - return -EPROTO; + goto err_proto; } cmd = &queue->cmds[data->ttag]; } else { @@ -935,19 +967,32 @@ static int nvmet_tcp_handle_h2c_data_pdu(struct nvmet_tcp_queue *queue) pr_err("ttag %u unexpected data offset %u (expected %u)\n", data->ttag, le32_to_cpu(data->data_offset), cmd->rbytes_done); - /* FIXME: use path and transport errors */ - nvmet_req_complete(&cmd->req, - NVME_SC_INVALID_FIELD | NVME_SC_DNR); - return -EPROTO; + goto err_proto; } + exp_data_len = le32_to_cpu(data->hdr.plen) - + nvmet_tcp_hdgst_len(queue) - + nvmet_tcp_ddgst_len(queue) - + sizeof(*data); + cmd->pdu_len = le32_to_cpu(data->data_length); + if (unlikely(cmd->pdu_len != exp_data_len || + cmd->pdu_len == 0 || + cmd->pdu_len > NVMET_TCP_MAXH2CDATA)) { + pr_err("H2CData PDU len %u is invalid\n", cmd->pdu_len); + goto err_proto; + } cmd->pdu_recv = 0; nvmet_tcp_build_pdu_iovec(cmd); queue->cmd = cmd; queue->rcv_state = NVMET_TCP_RECV_DATA; return 0; + +err_proto: + /* FIXME: use proper transport errors */ + nvmet_tcp_fatal_error(queue); + return -EPROTO; } static int nvmet_tcp_done_recv_pdu(struct nvmet_tcp_queue *queue) @@ -994,12 +1039,12 @@ static int nvmet_tcp_done_recv_pdu(struct nvmet_tcp_queue *queue) req = &queue->cmd->req; memcpy(req->cmd, nvme_cmd, sizeof(*nvme_cmd)); - if (unlikely(!nvmet_req_init(req, &queue->nvme_cq, - &queue->nvme_sq, &nvmet_tcp_ops))) { - pr_err("failed cmd %p id %d opcode %d, data_len: %d\n", + if (unlikely(!nvmet_req_init(req, &queue->nvme_sq, &nvmet_tcp_ops))) { + pr_err("failed cmd %p id %d opcode %d, data_len: %d, status: %04x\n", req->cmd, req->cmd->common.command_id, req->cmd->common.opcode, - le32_to_cpu(req->cmd->common.dptr.sgl.length)); + le32_to_cpu(req->cmd->common.dptr.sgl.length), + le16_to_cpu(req->cqe->status)); nvmet_tcp_handle_req_failure(queue, queue->cmd, req); return 0; @@ -1061,20 +1106,65 @@ static inline bool nvmet_tcp_pdu_valid(u8 type) return false; } +static int nvmet_tcp_tls_record_ok(struct nvmet_tcp_queue *queue, + struct msghdr *msg, char *cbuf) +{ + struct cmsghdr *cmsg = (struct cmsghdr *)cbuf; + u8 ctype, level, description; + int ret = 0; + + ctype = tls_get_record_type(queue->sock->sk, cmsg); + switch (ctype) { + case 0: + break; + case TLS_RECORD_TYPE_DATA: + break; + case TLS_RECORD_TYPE_ALERT: + tls_alert_recv(queue->sock->sk, msg, &level, &description); + if (level == TLS_ALERT_LEVEL_FATAL) { + pr_err("queue %d: TLS Alert desc %u\n", + queue->idx, description); + ret = -ENOTCONN; + } else { + pr_warn("queue %d: TLS Alert desc %u\n", + queue->idx, description); + ret = -EAGAIN; + } + break; + default: + /* discard this record type */ + pr_err("queue %d: TLS record %d unhandled\n", + queue->idx, ctype); + ret = -EAGAIN; + break; + } + return ret; +} + static int nvmet_tcp_try_recv_pdu(struct nvmet_tcp_queue *queue) { struct nvme_tcp_hdr *hdr = &queue->pdu.cmd.hdr; - int len; + int len, ret; struct kvec iov; + char cbuf[CMSG_LEN(sizeof(char))] = {}; struct msghdr msg = { .msg_flags = MSG_DONTWAIT }; recv: iov.iov_base = (void *)&queue->pdu + queue->offset; iov.iov_len = queue->left; + if (queue->tls_pskid) { + msg.msg_control = cbuf; + msg.msg_controllen = sizeof(cbuf); + } len = kernel_recvmsg(queue->sock, &msg, &iov, 1, iov.iov_len, msg.msg_flags); if (unlikely(len < 0)) return len; + if (queue->tls_pskid) { + ret = nvmet_tcp_tls_record_ok(queue, &msg, cbuf); + if (ret < 0) + return ret; + } queue->offset += len; queue->left -= len; @@ -1118,7 +1208,7 @@ static void nvmet_tcp_prep_recv_ddgst(struct nvmet_tcp_cmd *cmd) { struct nvmet_tcp_queue *queue = cmd->queue; - nvmet_tcp_calc_ddgst(queue->rcv_hash, cmd); + nvmet_tcp_calc_ddgst(cmd); queue->offset = 0; queue->left = NVME_TCP_DIGEST_LENGTH; queue->rcv_state = NVMET_TCP_RECV_DDGST; @@ -1127,16 +1217,22 @@ static void nvmet_tcp_prep_recv_ddgst(struct nvmet_tcp_cmd *cmd) static int nvmet_tcp_try_recv_data(struct nvmet_tcp_queue *queue) { struct nvmet_tcp_cmd *cmd = queue->cmd; - int ret; + int len, ret; while (msg_data_left(&cmd->recv_msg)) { - ret = sock_recvmsg(cmd->queue->sock, &cmd->recv_msg, + len = sock_recvmsg(cmd->queue->sock, &cmd->recv_msg, cmd->recv_msg.msg_flags); - if (ret <= 0) - return ret; + if (len <= 0) + return len; + if (queue->tls_pskid) { + ret = nvmet_tcp_tls_record_ok(cmd->queue, + &cmd->recv_msg, cmd->recv_cbuf); + if (ret < 0) + return ret; + } - cmd->pdu_recv += ret; - cmd->rbytes_done += ret; + cmd->pdu_recv += len; + cmd->rbytes_done += len; } if (queue->data_digest) { @@ -1154,20 +1250,30 @@ static int nvmet_tcp_try_recv_data(struct nvmet_tcp_queue *queue) static int nvmet_tcp_try_recv_ddgst(struct nvmet_tcp_queue *queue) { struct nvmet_tcp_cmd *cmd = queue->cmd; - int ret; + int ret, len; + char cbuf[CMSG_LEN(sizeof(char))] = {}; struct msghdr msg = { .msg_flags = MSG_DONTWAIT }; struct kvec iov = { .iov_base = (void *)&cmd->recv_ddgst + queue->offset, .iov_len = queue->left }; - ret = kernel_recvmsg(queue->sock, &msg, &iov, 1, + if (queue->tls_pskid) { + msg.msg_control = cbuf; + msg.msg_controllen = sizeof(cbuf); + } + len = kernel_recvmsg(queue->sock, &msg, &iov, 1, iov.iov_len, msg.msg_flags); - if (unlikely(ret < 0)) - return ret; + if (unlikely(len < 0)) + return len; + if (queue->tls_pskid) { + ret = nvmet_tcp_tls_record_ok(queue, &msg, cbuf); + if (ret < 0) + return ret; + } - queue->offset += ret; - queue->left -= ret; + queue->offset += len; + queue->left -= len; if (queue->left) return -EAGAIN; @@ -1245,14 +1351,27 @@ done: return ret; } +static void nvmet_tcp_release_queue(struct kref *kref) +{ + struct nvmet_tcp_queue *queue = + container_of(kref, struct nvmet_tcp_queue, kref); + + WARN_ON(queue->state != NVMET_TCP_Q_DISCONNECTING); + queue_work(nvmet_wq, &queue->release_work); +} + static void nvmet_tcp_schedule_release_queue(struct nvmet_tcp_queue *queue) { - spin_lock(&queue->state_lock); + spin_lock_bh(&queue->state_lock); + if (queue->state == NVMET_TCP_Q_TLS_HANDSHAKE) { + /* Socket closed during handshake */ + tls_handshake_cancel(queue->sock->sk); + } if (queue->state != NVMET_TCP_Q_DISCONNECTING) { queue->state = NVMET_TCP_Q_DISCONNECTING; - queue_work(nvmet_wq, &queue->release_work); + kref_put(&queue->kref, nvmet_tcp_release_queue); } - spin_unlock(&queue->state_lock); + spin_unlock_bh(&queue->state_lock); } static inline void nvmet_tcp_arm_queue_deadline(struct nvmet_tcp_queue *queue) @@ -1334,6 +1453,10 @@ static int nvmet_tcp_alloc_cmd(struct nvmet_tcp_queue *queue, if (!c->r2t_pdu) goto out_free_data; + if (queue->state == NVMET_TCP_Q_TLS_HANDSHAKE) { + c->recv_msg.msg_control = c->recv_cbuf; + c->recv_msg.msg_controllen = sizeof(c->recv_cbuf); + } c->recv_msg.msg_flags = MSG_DONTWAIT | MSG_NOSIGNAL; list_add_tail(&c->entry, &queue->free_list); @@ -1361,7 +1484,7 @@ static int nvmet_tcp_alloc_cmds(struct nvmet_tcp_queue *queue) struct nvmet_tcp_cmd *cmds; int i, ret = -EINVAL, nr_cmds = queue->nr_cmds; - cmds = kcalloc(nr_cmds, sizeof(struct nvmet_tcp_cmd), GFP_KERNEL); + cmds = kvcalloc(nr_cmds, sizeof(struct nvmet_tcp_cmd), GFP_KERNEL); if (!cmds) goto out; @@ -1377,7 +1500,7 @@ static int nvmet_tcp_alloc_cmds(struct nvmet_tcp_queue *queue) out_free: while (--i >= 0) nvmet_tcp_free_cmd(cmds + i); - kfree(cmds); + kvfree(cmds); out: return ret; } @@ -1391,13 +1514,16 @@ static void nvmet_tcp_free_cmds(struct nvmet_tcp_queue *queue) nvmet_tcp_free_cmd(cmds + i); nvmet_tcp_free_cmd(&queue->connect); - kfree(cmds); + kvfree(cmds); } static void nvmet_tcp_restore_socket_callbacks(struct nvmet_tcp_queue *queue) { struct socket *sock = queue->sock; + if (!queue->state_change) + return; + write_lock_bh(&sock->sk->sk_callback_lock); sock->sk->sk_data_ready = queue->data_ready; sock->sk->sk_state_change = queue->state_change; @@ -1427,18 +1553,13 @@ static void nvmet_tcp_free_cmd_data_in_buffers(struct nvmet_tcp_queue *queue) struct nvmet_tcp_cmd *cmd = queue->cmds; int i; - for (i = 0; i < queue->nr_cmds; i++, cmd++) { - if (nvmet_tcp_need_data_in(cmd)) - nvmet_tcp_free_cmd_buffers(cmd); - } - - if (!queue->nr_cmds && nvmet_tcp_need_data_in(&queue->connect)) - nvmet_tcp_free_cmd_buffers(&queue->connect); + for (i = 0; i < queue->nr_cmds; i++, cmd++) + nvmet_tcp_free_cmd_buffers(cmd); + nvmet_tcp_free_cmd_buffers(&queue->connect); } static void nvmet_tcp_release_queue_work(struct work_struct *w) { - struct page *page; struct nvmet_tcp_queue *queue = container_of(w, struct nvmet_tcp_queue, release_work); @@ -1447,22 +1568,22 @@ static void nvmet_tcp_release_queue_work(struct work_struct *w) mutex_unlock(&nvmet_tcp_queue_mutex); nvmet_tcp_restore_socket_callbacks(queue); + cancel_delayed_work_sync(&queue->tls_handshake_tmo_work); cancel_work_sync(&queue->io_work); /* stop accepting incoming data */ queue->rcv_state = NVMET_TCP_RECV_ERR; + nvmet_sq_put_tls_key(&queue->nvme_sq); nvmet_tcp_uninit_data_in_cmds(queue); nvmet_sq_destroy(&queue->nvme_sq); + nvmet_cq_put(&queue->nvme_cq); cancel_work_sync(&queue->io_work); nvmet_tcp_free_cmd_data_in_buffers(queue); - sock_release(queue->sock); + /* ->sock will be released by fput() */ + fput(queue->sock->file); nvmet_tcp_free_cmds(queue); - if (queue->hdr_digest || queue->data_digest) - nvmet_tcp_free_crypto(queue); ida_free(&nvmet_tcp_queue_ida, queue->idx); - - page = virt_to_head_page(queue->pf_cache.va); - __page_frag_cache_drain(page, queue->pf_cache.pagecnt_bias); + page_frag_cache_drain(&queue->pf_cache); kfree(queue); } @@ -1470,10 +1591,17 @@ static void nvmet_tcp_data_ready(struct sock *sk) { struct nvmet_tcp_queue *queue; + trace_sk_data_ready(sk); + read_lock_bh(&sk->sk_callback_lock); queue = sk->sk_user_data; - if (likely(queue)) - queue_work_on(queue_cpu(queue), nvmet_tcp_wq, &queue->io_work); + if (likely(queue)) { + if (queue->data_ready) + queue->data_ready(sk); + if (queue->state != NVMET_TCP_Q_TLS_HANDSHAKE) + queue_work_on(queue_cpu(queue), nvmet_tcp_wq, + &queue->io_work); + } read_unlock_bh(&sk->sk_callback_lock); } @@ -1581,38 +1709,209 @@ static int nvmet_tcp_set_queue_sock(struct nvmet_tcp_queue *queue) return ret; } -static int nvmet_tcp_alloc_queue(struct nvmet_tcp_port *port, +#ifdef CONFIG_NVME_TARGET_TCP_TLS +static int nvmet_tcp_try_peek_pdu(struct nvmet_tcp_queue *queue) +{ + struct nvme_tcp_hdr *hdr = &queue->pdu.cmd.hdr; + int len, ret; + struct kvec iov = { + .iov_base = (u8 *)&queue->pdu + queue->offset, + .iov_len = sizeof(struct nvme_tcp_hdr), + }; + char cbuf[CMSG_LEN(sizeof(char))] = {}; + struct msghdr msg = { + .msg_control = cbuf, + .msg_controllen = sizeof(cbuf), + .msg_flags = MSG_PEEK, + }; + + if (nvmet_port_secure_channel_required(queue->port->nport)) + return 0; + + len = kernel_recvmsg(queue->sock, &msg, &iov, 1, + iov.iov_len, msg.msg_flags); + if (unlikely(len < 0)) { + pr_debug("queue %d: peek error %d\n", + queue->idx, len); + return len; + } + + ret = nvmet_tcp_tls_record_ok(queue, &msg, cbuf); + if (ret < 0) + return ret; + + if (len < sizeof(struct nvme_tcp_hdr)) { + pr_debug("queue %d: short read, %d bytes missing\n", + queue->idx, (int)iov.iov_len - len); + return -EAGAIN; + } + pr_debug("queue %d: hdr type %d hlen %d plen %d size %d\n", + queue->idx, hdr->type, hdr->hlen, hdr->plen, + (int)sizeof(struct nvme_tcp_icreq_pdu)); + if (hdr->type == nvme_tcp_icreq && + hdr->hlen == sizeof(struct nvme_tcp_icreq_pdu) && + hdr->plen == cpu_to_le32(sizeof(struct nvme_tcp_icreq_pdu))) { + pr_debug("queue %d: icreq detected\n", + queue->idx); + return len; + } + return 0; +} + +static int nvmet_tcp_tls_key_lookup(struct nvmet_tcp_queue *queue, + key_serial_t peerid) +{ + struct key *tls_key = nvme_tls_key_lookup(peerid); + int status = 0; + + if (IS_ERR(tls_key)) { + pr_warn("%s: queue %d failed to lookup key %x\n", + __func__, queue->idx, peerid); + spin_lock_bh(&queue->state_lock); + queue->state = NVMET_TCP_Q_FAILED; + spin_unlock_bh(&queue->state_lock); + status = PTR_ERR(tls_key); + } else { + pr_debug("%s: queue %d using TLS PSK %x\n", + __func__, queue->idx, peerid); + queue->nvme_sq.tls_key = tls_key; + } + return status; +} + +static void nvmet_tcp_tls_handshake_done(void *data, int status, + key_serial_t peerid) +{ + struct nvmet_tcp_queue *queue = data; + + pr_debug("queue %d: TLS handshake done, key %x, status %d\n", + queue->idx, peerid, status); + spin_lock_bh(&queue->state_lock); + if (WARN_ON(queue->state != NVMET_TCP_Q_TLS_HANDSHAKE)) { + spin_unlock_bh(&queue->state_lock); + return; + } + if (!status) { + queue->tls_pskid = peerid; + queue->state = NVMET_TCP_Q_CONNECTING; + } else + queue->state = NVMET_TCP_Q_FAILED; + spin_unlock_bh(&queue->state_lock); + + cancel_delayed_work_sync(&queue->tls_handshake_tmo_work); + + if (!status) + status = nvmet_tcp_tls_key_lookup(queue, peerid); + + if (status) + nvmet_tcp_schedule_release_queue(queue); + else + nvmet_tcp_set_queue_sock(queue); + kref_put(&queue->kref, nvmet_tcp_release_queue); +} + +static void nvmet_tcp_tls_handshake_timeout(struct work_struct *w) +{ + struct nvmet_tcp_queue *queue = container_of(to_delayed_work(w), + struct nvmet_tcp_queue, tls_handshake_tmo_work); + + pr_warn("queue %d: TLS handshake timeout\n", queue->idx); + /* + * If tls_handshake_cancel() fails we've lost the race with + * nvmet_tcp_tls_handshake_done() */ + if (!tls_handshake_cancel(queue->sock->sk)) + return; + spin_lock_bh(&queue->state_lock); + if (WARN_ON(queue->state != NVMET_TCP_Q_TLS_HANDSHAKE)) { + spin_unlock_bh(&queue->state_lock); + return; + } + queue->state = NVMET_TCP_Q_FAILED; + spin_unlock_bh(&queue->state_lock); + nvmet_tcp_schedule_release_queue(queue); + kref_put(&queue->kref, nvmet_tcp_release_queue); +} + +static int nvmet_tcp_tls_handshake(struct nvmet_tcp_queue *queue) +{ + int ret = -EOPNOTSUPP; + struct tls_handshake_args args; + + if (queue->state != NVMET_TCP_Q_TLS_HANDSHAKE) { + pr_warn("cannot start TLS in state %d\n", queue->state); + return -EINVAL; + } + + kref_get(&queue->kref); + pr_debug("queue %d: TLS ServerHello\n", queue->idx); + memset(&args, 0, sizeof(args)); + args.ta_sock = queue->sock; + args.ta_done = nvmet_tcp_tls_handshake_done; + args.ta_data = queue; + args.ta_keyring = key_serial(queue->port->nport->keyring); + args.ta_timeout_ms = tls_handshake_timeout * 1000; + + ret = tls_server_hello_psk(&args, GFP_KERNEL); + if (ret) { + kref_put(&queue->kref, nvmet_tcp_release_queue); + pr_err("failed to start TLS, err=%d\n", ret); + } else { + queue_delayed_work(nvmet_wq, &queue->tls_handshake_tmo_work, + tls_handshake_timeout * HZ); + } + return ret; +} +#else +static void nvmet_tcp_tls_handshake_timeout(struct work_struct *w) {} +#endif + +static void nvmet_tcp_alloc_queue(struct nvmet_tcp_port *port, struct socket *newsock) { struct nvmet_tcp_queue *queue; + struct file *sock_file = NULL; int ret; queue = kzalloc(sizeof(*queue), GFP_KERNEL); - if (!queue) - return -ENOMEM; + if (!queue) { + ret = -ENOMEM; + goto out_release; + } INIT_WORK(&queue->release_work, nvmet_tcp_release_queue_work); INIT_WORK(&queue->io_work, nvmet_tcp_io_work); + kref_init(&queue->kref); queue->sock = newsock; queue->port = port; queue->nr_cmds = 0; spin_lock_init(&queue->state_lock); - queue->state = NVMET_TCP_Q_CONNECTING; + if (queue->port->nport->disc_addr.tsas.tcp.sectype == + NVMF_TCP_SECTYPE_TLS13) + queue->state = NVMET_TCP_Q_TLS_HANDSHAKE; + else + queue->state = NVMET_TCP_Q_CONNECTING; INIT_LIST_HEAD(&queue->free_list); init_llist_head(&queue->resp_list); INIT_LIST_HEAD(&queue->resp_send_list); + sock_file = sock_alloc_file(queue->sock, O_CLOEXEC, NULL); + if (IS_ERR(sock_file)) { + ret = PTR_ERR(sock_file); + goto out_free_queue; + } + queue->idx = ida_alloc(&nvmet_tcp_queue_ida, GFP_KERNEL); if (queue->idx < 0) { ret = queue->idx; - goto out_free_queue; + goto out_sock; } ret = nvmet_tcp_alloc_cmd(queue, &queue->connect); if (ret) goto out_ida_remove; - ret = nvmet_sq_init(&queue->nvme_sq); + nvmet_cq_init(&queue->nvme_cq); + ret = nvmet_sq_init(&queue->nvme_sq, &queue->nvme_cq); if (ret) goto out_free_connect; @@ -1622,23 +1921,51 @@ static int nvmet_tcp_alloc_queue(struct nvmet_tcp_port *port, list_add_tail(&queue->queue_list, &nvmet_tcp_queue_list); mutex_unlock(&nvmet_tcp_queue_mutex); + INIT_DELAYED_WORK(&queue->tls_handshake_tmo_work, + nvmet_tcp_tls_handshake_timeout); +#ifdef CONFIG_NVME_TARGET_TCP_TLS + if (queue->state == NVMET_TCP_Q_TLS_HANDSHAKE) { + struct sock *sk = queue->sock->sk; + + /* Restore the default callbacks before starting upcall */ + write_lock_bh(&sk->sk_callback_lock); + sk->sk_user_data = NULL; + sk->sk_data_ready = port->data_ready; + write_unlock_bh(&sk->sk_callback_lock); + if (!nvmet_tcp_try_peek_pdu(queue)) { + if (!nvmet_tcp_tls_handshake(queue)) + return; + /* TLS handshake failed, terminate the connection */ + goto out_destroy_sq; + } + /* Not a TLS connection, continue with normal processing */ + queue->state = NVMET_TCP_Q_CONNECTING; + } +#endif + ret = nvmet_tcp_set_queue_sock(queue); if (ret) goto out_destroy_sq; - return 0; + return; out_destroy_sq: mutex_lock(&nvmet_tcp_queue_mutex); list_del_init(&queue->queue_list); mutex_unlock(&nvmet_tcp_queue_mutex); nvmet_sq_destroy(&queue->nvme_sq); out_free_connect: + nvmet_cq_put(&queue->nvme_cq); nvmet_tcp_free_cmd(&queue->connect); out_ida_remove: ida_free(&nvmet_tcp_queue_ida, queue->idx); +out_sock: + fput(queue->sock->file); out_free_queue: kfree(queue); - return ret; +out_release: + pr_err("failed to allocate queue, error %d\n", ret); + if (!sock_file) + sock_release(newsock); } static void nvmet_tcp_accept_work(struct work_struct *w) @@ -1655,11 +1982,7 @@ static void nvmet_tcp_accept_work(struct work_struct *w) pr_warn("failed to accept err=%d\n", ret); return; } - ret = nvmet_tcp_alloc_queue(port, newsock); - if (ret) { - pr_err("failed to allocate queue\n"); - sock_release(newsock); - } + nvmet_tcp_alloc_queue(port, newsock); } } @@ -1667,6 +1990,8 @@ static void nvmet_tcp_listen_data_ready(struct sock *sk) { struct nvmet_tcp_port *port; + trace_sk_data_ready(sk); + read_lock_bh(&sk->sk_callback_lock); port = sk->sk_user_data; if (!port) @@ -1730,14 +2055,14 @@ static int nvmet_tcp_add_port(struct nvmet_port *nport) if (so_priority > 0) sock_set_priority(port->sock->sk, so_priority); - ret = kernel_bind(port->sock, (struct sockaddr *)&port->addr, + ret = kernel_bind(port->sock, (struct sockaddr_unsized *)&port->addr, sizeof(port->addr)); if (ret) { pr_err("failed to bind port socket %d\n", ret); goto err_sock; } - ret = kernel_listen(port->sock, 128); + ret = kernel_listen(port->sock, NVMET_TCP_BACKLOG); if (ret) { pr_err("failed to listen %d on port sock\n", ret); goto err_sock; @@ -1803,13 +2128,26 @@ static u16 nvmet_tcp_install_queue(struct nvmet_sq *sq) container_of(sq, struct nvmet_tcp_queue, nvme_sq); if (sq->qid == 0) { - /* Let inflight controller teardown complete */ - flush_workqueue(nvmet_wq); + struct nvmet_tcp_queue *q; + int pending = 0; + + /* Check for pending controller teardown */ + mutex_lock(&nvmet_tcp_queue_mutex); + list_for_each_entry(q, &nvmet_tcp_queue_list, queue_list) { + if (q->nvme_sq.ctrl == sq->ctrl && + q->state == NVMET_TCP_Q_DISCONNECTING) + pending++; + } + mutex_unlock(&nvmet_tcp_queue_mutex); + if (pending > NVMET_TCP_BACKLOG) + return NVME_SC_CONNECT_CTRL_BUSY; } queue->nr_cmds = sq->size * 2; - if (nvmet_tcp_alloc_cmds(queue)) + if (nvmet_tcp_alloc_cmds(queue)) { + queue->nr_cmds = 0; return NVME_SC_INTERNAL; + } return 0; } @@ -1818,7 +2156,7 @@ static void nvmet_tcp_disc_port_addr(struct nvmet_req *req, { struct nvmet_tcp_port *port = nport->priv; - if (inet_addr_is_any((struct sockaddr *)&port->addr)) { + if (inet_addr_is_any(&port->addr)) { struct nvmet_tcp_cmd *cmd = container_of(req, struct nvmet_tcp_cmd, req); struct nvmet_tcp_queue *queue = cmd->queue; @@ -1829,6 +2167,19 @@ static void nvmet_tcp_disc_port_addr(struct nvmet_req *req, } } +static ssize_t nvmet_tcp_host_port_addr(struct nvmet_ctrl *ctrl, + char *traddr, size_t traddr_len) +{ + struct nvmet_sq *sq = ctrl->sqs[0]; + struct nvmet_tcp_queue *queue = + container_of(sq, struct nvmet_tcp_queue, nvme_sq); + + if (queue->sockaddr_peer.ss_family == AF_UNSPEC) + return -EINVAL; + return snprintf(traddr, traddr_len, "%pISc", + (struct sockaddr *)&queue->sockaddr_peer); +} + static const struct nvmet_fabrics_ops nvmet_tcp_ops = { .owner = THIS_MODULE, .type = NVMF_TRTYPE_TCP, @@ -1839,6 +2190,7 @@ static const struct nvmet_fabrics_ops nvmet_tcp_ops = { .delete_ctrl = nvmet_tcp_delete_ctrl, .install_queue = nvmet_tcp_install_queue, .disc_traddr = nvmet_tcp_disc_port_addr, + .host_traddr = nvmet_tcp_host_port_addr, }; static int __init nvmet_tcp_init(void) @@ -1874,10 +2226,12 @@ static void __exit nvmet_tcp_exit(void) flush_workqueue(nvmet_wq); destroy_workqueue(nvmet_tcp_wq); + ida_destroy(&nvmet_tcp_queue_ida); } module_init(nvmet_tcp_init); module_exit(nvmet_tcp_exit); +MODULE_DESCRIPTION("NVMe target TCP transport driver"); MODULE_LICENSE("GPL v2"); MODULE_ALIAS("nvmet-transport-3"); /* 3 == NVMF_TRTYPE_TCP */ diff --git a/drivers/nvme/target/trace.c b/drivers/nvme/target/trace.c index bff454d46255..6dbc7036f2e4 100644 --- a/drivers/nvme/target/trace.c +++ b/drivers/nvme/target/trace.c @@ -4,7 +4,7 @@ * Copyright (c) 2018 Johannes Thumshirn, SUSE Linux GmbH */ -#include <asm/unaligned.h> +#include <linux/unaligned.h> #include "trace.h" static const char *nvmet_trace_admin_identify(struct trace_seq *p, u8 *cdw10) @@ -119,6 +119,167 @@ const char *nvmet_trace_parse_admin_cmd(struct trace_seq *p, } } +static const char *nvmet_trace_zone_mgmt_send(struct trace_seq *p, u8 *cdw10) +{ + static const char * const zsa_strs[] = { + [0x01] = "close zone", + [0x02] = "finish zone", + [0x03] = "open zone", + [0x04] = "reset zone", + [0x05] = "offline zone", + [0x10] = "set zone descriptor extension" + }; + const char *ret = trace_seq_buffer_ptr(p); + u64 slba = get_unaligned_le64(cdw10); + const char *zsa_str; + u8 zsa = cdw10[12]; + u8 all = cdw10[13]; + + if (zsa < ARRAY_SIZE(zsa_strs) && zsa_strs[zsa]) + zsa_str = zsa_strs[zsa]; + else + zsa_str = "reserved"; + + trace_seq_printf(p, "slba=%llu, zsa=%u:%s, all=%u", + slba, zsa, zsa_str, all); + trace_seq_putc(p, 0); + + return ret; +} + +static const char *nvmet_trace_zone_mgmt_recv(struct trace_seq *p, u8 *cdw10) +{ + static const char * const zrasf_strs[] = { + [0x00] = "list all zones", + [0x01] = "list the zones in the ZSE: Empty state", + [0x02] = "list the zones in the ZSIO: Implicitly Opened state", + [0x03] = "list the zones in the ZSEO: Explicitly Opened state", + [0x04] = "list the zones in the ZSC: Closed state", + [0x05] = "list the zones in the ZSF: Full state", + [0x06] = "list the zones in the ZSRO: Read Only state", + [0x07] = "list the zones in the ZSO: Offline state", + [0x09] = "list the zones that have the zone attribute" + }; + const char *ret = trace_seq_buffer_ptr(p); + u64 slba = get_unaligned_le64(cdw10); + u32 numd = get_unaligned_le32(&cdw10[8]); + u8 zra = cdw10[12]; + u8 zrasf = cdw10[13]; + const char *zrasf_str; + u8 pr = cdw10[14]; + + if (zrasf < ARRAY_SIZE(zrasf_strs) && zrasf_strs[zrasf]) + zrasf_str = zrasf_strs[zrasf]; + else + zrasf_str = "reserved"; + + trace_seq_printf(p, "slba=%llu, numd=%u, zra=%u, zrasf=%u:%s, pr=%u", + slba, numd, zra, zrasf, zrasf_str, pr); + trace_seq_putc(p, 0); + + return ret; +} + +static const char *nvmet_trace_resv_reg(struct trace_seq *p, u8 *cdw10) +{ + static const char * const rrega_strs[] = { + [0x00] = "register", + [0x01] = "unregister", + [0x02] = "replace", + }; + const char *ret = trace_seq_buffer_ptr(p); + u8 rrega = cdw10[0] & 0x7; + u8 iekey = (cdw10[0] >> 3) & 0x1; + u8 ptpl = (cdw10[3] >> 6) & 0x3; + const char *rrega_str; + + if (rrega < ARRAY_SIZE(rrega_strs) && rrega_strs[rrega]) + rrega_str = rrega_strs[rrega]; + else + rrega_str = "reserved"; + + trace_seq_printf(p, "rrega=%u:%s, iekey=%u, ptpl=%u", + rrega, rrega_str, iekey, ptpl); + trace_seq_putc(p, 0); + + return ret; +} + +static const char * const rtype_strs[] = { + [0x00] = "reserved", + [0x01] = "write exclusive", + [0x02] = "exclusive access", + [0x03] = "write exclusive registrants only", + [0x04] = "exclusive access registrants only", + [0x05] = "write exclusive all registrants", + [0x06] = "exclusive access all registrants", +}; + +static const char *nvmet_trace_resv_acq(struct trace_seq *p, u8 *cdw10) +{ + static const char * const racqa_strs[] = { + [0x00] = "acquire", + [0x01] = "preempt", + [0x02] = "preempt and abort", + }; + const char *ret = trace_seq_buffer_ptr(p); + u8 racqa = cdw10[0] & 0x7; + u8 iekey = (cdw10[0] >> 3) & 0x1; + u8 rtype = cdw10[1]; + const char *racqa_str = "reserved"; + const char *rtype_str = "reserved"; + + if (racqa < ARRAY_SIZE(racqa_strs) && racqa_strs[racqa]) + racqa_str = racqa_strs[racqa]; + + if (rtype < ARRAY_SIZE(rtype_strs) && rtype_strs[rtype]) + rtype_str = rtype_strs[rtype]; + + trace_seq_printf(p, "racqa=%u:%s, iekey=%u, rtype=%u:%s", + racqa, racqa_str, iekey, rtype, rtype_str); + trace_seq_putc(p, 0); + + return ret; +} + +static const char *nvmet_trace_resv_rel(struct trace_seq *p, u8 *cdw10) +{ + static const char * const rrela_strs[] = { + [0x00] = "release", + [0x01] = "clear", + }; + const char *ret = trace_seq_buffer_ptr(p); + u8 rrela = cdw10[0] & 0x7; + u8 iekey = (cdw10[0] >> 3) & 0x1; + u8 rtype = cdw10[1]; + const char *rrela_str = "reserved"; + const char *rtype_str = "reserved"; + + if (rrela < ARRAY_SIZE(rrela_strs) && rrela_strs[rrela]) + rrela_str = rrela_strs[rrela]; + + if (rtype < ARRAY_SIZE(rtype_strs) && rtype_strs[rtype]) + rtype_str = rtype_strs[rtype]; + + trace_seq_printf(p, "rrela=%u:%s, iekey=%u, rtype=%u:%s", + rrela, rrela_str, iekey, rtype, rtype_str); + trace_seq_putc(p, 0); + + return ret; +} + +static const char *nvmet_trace_resv_report(struct trace_seq *p, u8 *cdw10) +{ + const char *ret = trace_seq_buffer_ptr(p); + u32 numd = get_unaligned_le32(cdw10); + u8 eds = cdw10[4] & 0x1; + + trace_seq_printf(p, "numd=%u, eds=%u", numd, eds); + trace_seq_putc(p, 0); + + return ret; +} + const char *nvmet_trace_parse_nvm_cmd(struct trace_seq *p, u8 opcode, u8 *cdw10) { @@ -126,9 +287,22 @@ const char *nvmet_trace_parse_nvm_cmd(struct trace_seq *p, case nvme_cmd_read: case nvme_cmd_write: case nvme_cmd_write_zeroes: + case nvme_cmd_zone_append: return nvmet_trace_read_write(p, cdw10); case nvme_cmd_dsm: return nvmet_trace_dsm(p, cdw10); + case nvme_cmd_zone_mgmt_send: + return nvmet_trace_zone_mgmt_send(p, cdw10); + case nvme_cmd_zone_mgmt_recv: + return nvmet_trace_zone_mgmt_recv(p, cdw10); + case nvme_cmd_resv_register: + return nvmet_trace_resv_reg(p, cdw10); + case nvme_cmd_resv_acquire: + return nvmet_trace_resv_acq(p, cdw10); + case nvme_cmd_resv_release: + return nvmet_trace_resv_rel(p, cdw10); + case nvme_cmd_resv_report: + return nvmet_trace_resv_report(p, cdw10); default: return nvmet_trace_common(p, cdw10); } @@ -176,6 +350,34 @@ static const char *nvmet_trace_fabrics_property_get(struct trace_seq *p, return ret; } +static const char *nvmet_trace_fabrics_auth_send(struct trace_seq *p, u8 *spc) +{ + const char *ret = trace_seq_buffer_ptr(p); + u8 spsp0 = spc[1]; + u8 spsp1 = spc[2]; + u8 secp = spc[3]; + u32 tl = get_unaligned_le32(spc + 4); + + trace_seq_printf(p, "spsp0=%02x, spsp1=%02x, secp=%02x, tl=%u", + spsp0, spsp1, secp, tl); + trace_seq_putc(p, 0); + return ret; +} + +static const char *nvmet_trace_fabrics_auth_receive(struct trace_seq *p, u8 *spc) +{ + const char *ret = trace_seq_buffer_ptr(p); + u8 spsp0 = spc[1]; + u8 spsp1 = spc[2]; + u8 secp = spc[3]; + u32 al = get_unaligned_le32(spc + 4); + + trace_seq_printf(p, "spsp0=%02x, spsp1=%02x, secp=%02x, al=%u", + spsp0, spsp1, secp, al); + trace_seq_putc(p, 0); + return ret; +} + static const char *nvmet_trace_fabrics_common(struct trace_seq *p, u8 *spc) { const char *ret = trace_seq_buffer_ptr(p); @@ -195,6 +397,10 @@ const char *nvmet_trace_parse_fabrics_cmd(struct trace_seq *p, return nvmet_trace_fabrics_connect(p, spc); case nvme_fabrics_type_property_get: return nvmet_trace_fabrics_property_get(p, spc); + case nvme_fabrics_type_auth_send: + return nvmet_trace_fabrics_auth_send(p, spc); + case nvme_fabrics_type_auth_receive: + return nvmet_trace_fabrics_auth_receive(p, spc); default: return nvmet_trace_fabrics_common(p, spc); } @@ -211,7 +417,7 @@ const char *nvmet_trace_disk_name(struct trace_seq *p, char *name) return ret; } -const char *nvmet_trace_ctrl_name(struct trace_seq *p, struct nvmet_ctrl *ctrl) +const char *nvmet_trace_ctrl_id(struct trace_seq *p, u16 ctrl_id) { const char *ret = trace_seq_buffer_ptr(p); @@ -224,8 +430,8 @@ const char *nvmet_trace_ctrl_name(struct trace_seq *p, struct nvmet_ctrl *ctrl) * If we can know the extra data of the connect command in this stage, * we can update this print statement later. */ - if (ctrl) - trace_seq_printf(p, "%d", ctrl->cntlid); + if (ctrl_id) + trace_seq_printf(p, "%d", ctrl_id); else trace_seq_printf(p, "_"); trace_seq_putc(p, 0); diff --git a/drivers/nvme/target/trace.h b/drivers/nvme/target/trace.h index 6109b3806b12..7f7ebf9558e5 100644 --- a/drivers/nvme/target/trace.h +++ b/drivers/nvme/target/trace.h @@ -32,18 +32,24 @@ const char *nvmet_trace_parse_fabrics_cmd(struct trace_seq *p, u8 fctype, nvmet_trace_parse_nvm_cmd(p, opcode, cdw10) : \ nvmet_trace_parse_admin_cmd(p, opcode, cdw10))) -const char *nvmet_trace_ctrl_name(struct trace_seq *p, struct nvmet_ctrl *ctrl); -#define __print_ctrl_name(ctrl) \ - nvmet_trace_ctrl_name(p, ctrl) +const char *nvmet_trace_ctrl_id(struct trace_seq *p, u16 ctrl_id); +#define __print_ctrl_id(ctrl_id) \ + nvmet_trace_ctrl_id(p, ctrl_id) const char *nvmet_trace_disk_name(struct trace_seq *p, char *name); #define __print_disk_name(name) \ nvmet_trace_disk_name(p, name) #ifndef TRACE_HEADER_MULTI_READ -static inline struct nvmet_ctrl *nvmet_req_to_ctrl(struct nvmet_req *req) +static inline u16 nvmet_req_to_ctrl_id(struct nvmet_req *req) { - return req->sq->ctrl; + /* + * The queue and controller pointers are not valid until an association + * has been established. + */ + if (!req->sq || !req->sq->ctrl) + return 0; + return req->sq->ctrl->cntlid; } static inline void __assign_req_name(char *name, struct nvmet_req *req) @@ -53,8 +59,7 @@ static inline void __assign_req_name(char *name, struct nvmet_req *req) return; } - strncpy(name, req->ns->device_path, - min_t(size_t, DISK_NAME_LEN, strlen(req->ns->device_path))); + strscpy_pad(name, req->ns->device_path, DISK_NAME_LEN); } #endif @@ -63,7 +68,7 @@ TRACE_EVENT(nvmet_req_init, TP_ARGS(req, cmd), TP_STRUCT__entry( __field(struct nvme_command *, cmd) - __field(struct nvmet_ctrl *, ctrl) + __field(u16, ctrl_id) __array(char, disk, DISK_NAME_LEN) __field(int, qid) __field(u16, cid) @@ -76,7 +81,7 @@ TRACE_EVENT(nvmet_req_init, ), TP_fast_assign( __entry->cmd = cmd; - __entry->ctrl = nvmet_req_to_ctrl(req); + __entry->ctrl_id = nvmet_req_to_ctrl_id(req); __assign_req_name(__entry->disk, req); __entry->qid = req->sq->qid; __entry->cid = cmd->common.command_id; @@ -85,12 +90,12 @@ TRACE_EVENT(nvmet_req_init, __entry->flags = cmd->common.flags; __entry->nsid = le32_to_cpu(cmd->common.nsid); __entry->metadata = le64_to_cpu(cmd->common.metadata); - memcpy(__entry->cdw10, &cmd->common.cdw10, + memcpy(__entry->cdw10, &cmd->common.cdws, sizeof(__entry->cdw10)); ), TP_printk("nvmet%s: %sqid=%d, cmdid=%u, nsid=%u, flags=%#x, " "meta=%#llx, cmd=(%s, %s)", - __print_ctrl_name(__entry->ctrl), + __print_ctrl_id(__entry->ctrl_id), __print_disk_name(__entry->disk), __entry->qid, __entry->cid, __entry->nsid, __entry->flags, __entry->metadata, @@ -104,7 +109,7 @@ TRACE_EVENT(nvmet_req_complete, TP_PROTO(struct nvmet_req *req), TP_ARGS(req), TP_STRUCT__entry( - __field(struct nvmet_ctrl *, ctrl) + __field(u16, ctrl_id) __array(char, disk, DISK_NAME_LEN) __field(int, qid) __field(int, cid) @@ -112,7 +117,7 @@ TRACE_EVENT(nvmet_req_complete, __field(u16, status) ), TP_fast_assign( - __entry->ctrl = nvmet_req_to_ctrl(req); + __entry->ctrl_id = nvmet_req_to_ctrl_id(req); __entry->qid = req->cq->qid; __entry->cid = req->cqe->command_id; __entry->result = le64_to_cpu(req->cqe->result.u64); @@ -120,7 +125,7 @@ TRACE_EVENT(nvmet_req_complete, __assign_req_name(__entry->disk, req); ), TP_printk("nvmet%s: %sqid=%d, cmdid=%u, res=%#llx, status=%#x", - __print_ctrl_name(__entry->ctrl), + __print_ctrl_id(__entry->ctrl_id), __print_disk_name(__entry->disk), __entry->qid, __entry->cid, __entry->result, __entry->status) diff --git a/drivers/nvme/target/zns.c b/drivers/nvme/target/zns.c index 1254cf57e008..15a579cf528c 100644 --- a/drivers/nvme/target/zns.c +++ b/drivers/nvme/target/zns.c @@ -52,14 +52,10 @@ bool nvmet_bdev_zns_enable(struct nvmet_ns *ns) if (get_capacity(bd_disk) & (bdev_zone_sectors(ns->bdev) - 1)) return false; /* - * ZNS does not define a conventional zone type. If the underlying - * device has a bitmap set indicating the existence of conventional - * zones, reject the device. Otherwise, use report zones to detect if - * the device has conventional zones. + * ZNS does not define a conventional zone type. Use report zones + * to detect if the device has conventional zones and reject it if + * it does. */ - if (ns->bdev->bd_disk->conv_zones_bitmap) - return false; - ret = blkdev_report_zones(ns->bdev, 0, bdev_nr_zones(ns->bdev), validate_conv_zones_cb, NULL); if (ret < 0) @@ -70,7 +66,7 @@ bool nvmet_bdev_zns_enable(struct nvmet_ns *ns) return true; } -void nvmet_execute_identify_cns_cs_ctrl(struct nvmet_req *req) +void nvmet_execute_identify_ctrl_zns(struct nvmet_req *req) { u8 zasl = req->sq->ctrl->subsys->zasl; struct nvmet_ctrl *ctrl = req->sq->ctrl; @@ -95,16 +91,16 @@ out: nvmet_req_complete(req, status); } -void nvmet_execute_identify_cns_cs_ns(struct nvmet_req *req) +void nvmet_execute_identify_ns_zns(struct nvmet_req *req) { - struct nvme_id_ns_zns *id_zns; + struct nvme_id_ns_zns *id_zns = NULL; u64 zsze; u16 status; u32 mar, mor; if (le32_to_cpu(req->cmd->identify.nsid) == NVME_NSID_ALL) { req->error_loc = offsetof(struct nvme_identify, nsid); - status = NVME_SC_INVALID_NS | NVME_SC_DNR; + status = NVME_SC_INVALID_NS | NVME_STATUS_DNR; goto out; } @@ -118,16 +114,18 @@ void nvmet_execute_identify_cns_cs_ns(struct nvmet_req *req) if (status) goto done; - if (!bdev_is_zoned(req->ns->bdev)) { - req->error_loc = offsetof(struct nvme_identify, nsid); - goto done; - } - if (nvmet_ns_revalidate(req->ns)) { mutex_lock(&req->ns->subsys->lock); nvmet_ns_changed(req->ns->subsys, req->ns->nsid); mutex_unlock(&req->ns->subsys->lock); } + + if (!bdev_is_zoned(req->ns->bdev)) { + status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; + req->error_loc = offsetof(struct nvme_identify, nsid); + goto out; + } + zsze = (bdev_zone_sectors(req->ns->bdev) << 9) >> req->ns->blksize_shift; id_zns->lbafe[0].zsze = cpu_to_le64(zsze); @@ -148,8 +146,8 @@ void nvmet_execute_identify_cns_cs_ns(struct nvmet_req *req) done: status = nvmet_copy_to_sgl(req, 0, id_zns, sizeof(*id_zns)); - kfree(id_zns); out: + kfree(id_zns); nvmet_req_complete(req, status); } @@ -160,17 +158,17 @@ static u16 nvmet_bdev_validate_zone_mgmt_recv(struct nvmet_req *req) if (sect >= get_capacity(req->ns->bdev->bd_disk)) { req->error_loc = offsetof(struct nvme_zone_mgmt_recv_cmd, slba); - return NVME_SC_LBA_RANGE | NVME_SC_DNR; + return NVME_SC_LBA_RANGE | NVME_STATUS_DNR; } if (out_bufsize < sizeof(struct nvme_zone_report)) { req->error_loc = offsetof(struct nvme_zone_mgmt_recv_cmd, numd); - return NVME_SC_INVALID_FIELD | NVME_SC_DNR; + return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; } if (req->cmd->zmr.zra != NVME_ZRA_ZONE_REPORT) { req->error_loc = offsetof(struct nvme_zone_mgmt_recv_cmd, zra); - return NVME_SC_INVALID_FIELD | NVME_SC_DNR; + return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; } switch (req->cmd->zmr.pr) { @@ -179,7 +177,7 @@ static u16 nvmet_bdev_validate_zone_mgmt_recv(struct nvmet_req *req) break; default: req->error_loc = offsetof(struct nvme_zone_mgmt_recv_cmd, pr); - return NVME_SC_INVALID_FIELD | NVME_SC_DNR; + return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; } switch (req->cmd->zmr.zrasf) { @@ -195,7 +193,7 @@ static u16 nvmet_bdev_validate_zone_mgmt_recv(struct nvmet_req *req) default: req->error_loc = offsetof(struct nvme_zone_mgmt_recv_cmd, zrasf); - return NVME_SC_INVALID_FIELD | NVME_SC_DNR; + return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; } return NVME_SC_SUCCESS; @@ -254,8 +252,7 @@ static unsigned long nvmet_req_nr_zones_from_slba(struct nvmet_req *req) { unsigned int sect = nvmet_lba_to_sect(req->ns, req->cmd->zmr.slba); - return bdev_nr_zones(req->ns->bdev) - - (sect >> ilog2(bdev_zone_sectors(req->ns->bdev))); + return bdev_nr_zones(req->ns->bdev) - bdev_zone_no(req->ns->bdev, sect); } static unsigned long get_nr_zones_from_buf(struct nvmet_req *req, u32 bufsize) @@ -344,7 +341,7 @@ static u16 blkdev_zone_mgmt_errno_to_nvme_status(int ret) return NVME_SC_SUCCESS; case -EINVAL: case -EIO: - return NVME_SC_ZONE_INVALID_TRANSITION | NVME_SC_DNR; + return NVME_SC_ZONE_INVALID_TRANSITION | NVME_STATUS_DNR; default: return NVME_SC_INTERNAL; } @@ -455,8 +452,7 @@ static u16 nvmet_bdev_execute_zmgmt_send_all(struct nvmet_req *req) switch (zsa_req_op(req->cmd->zms.zsa)) { case REQ_OP_ZONE_RESET: ret = blkdev_zone_mgmt(req->ns->bdev, REQ_OP_ZONE_RESET, 0, - get_capacity(req->ns->bdev->bd_disk), - GFP_KERNEL); + get_capacity(req->ns->bdev->bd_disk)); if (ret < 0) return blkdev_zone_mgmt_errno_to_nvme_status(ret); break; @@ -467,7 +463,7 @@ static u16 nvmet_bdev_execute_zmgmt_send_all(struct nvmet_req *req) default: /* this is needed to quiet compiler warning */ req->error_loc = offsetof(struct nvme_zone_mgmt_send_cmd, zsa); - return NVME_SC_INVALID_FIELD | NVME_SC_DNR; + return NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; } return NVME_SC_SUCCESS; @@ -485,7 +481,7 @@ static void nvmet_bdev_zmgmt_send_work(struct work_struct *w) if (op == REQ_OP_LAST) { req->error_loc = offsetof(struct nvme_zone_mgmt_send_cmd, zsa); - status = NVME_SC_ZONE_INVALID_TRANSITION | NVME_SC_DNR; + status = NVME_SC_ZONE_INVALID_TRANSITION | NVME_STATUS_DNR; goto out; } @@ -497,17 +493,17 @@ static void nvmet_bdev_zmgmt_send_work(struct work_struct *w) if (sect >= get_capacity(bdev->bd_disk)) { req->error_loc = offsetof(struct nvme_zone_mgmt_send_cmd, slba); - status = NVME_SC_LBA_RANGE | NVME_SC_DNR; + status = NVME_SC_LBA_RANGE | NVME_STATUS_DNR; goto out; } if (sect & (zone_sectors - 1)) { req->error_loc = offsetof(struct nvme_zone_mgmt_send_cmd, slba); - status = NVME_SC_INVALID_FIELD | NVME_SC_DNR; + status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; goto out; } - ret = blkdev_zone_mgmt(bdev, op, sect, zone_sectors, GFP_KERNEL); + ret = blkdev_zone_mgmt(bdev, op, sect, zone_sectors); if (ret < 0) status = blkdev_zone_mgmt_errno_to_nvme_status(ret); @@ -541,13 +537,21 @@ void nvmet_bdev_execute_zone_append(struct nvmet_req *req) u16 status = NVME_SC_SUCCESS; unsigned int total_len = 0; struct scatterlist *sg; + u32 data_len = nvmet_rw_data_len(req); struct bio *bio; int sg_cnt; - /* Request is completed on len mismatch in nvmet_check_transter_len() */ + /* Request is completed on len mismatch in nvmet_check_transfer_len() */ if (!nvmet_check_transfer_len(req, nvmet_rw_data_len(req))) return; + if (data_len > + bdev_max_zone_append_sectors(req->ns->bdev) << SECTOR_SHIFT) { + req->error_loc = offsetof(struct nvme_rw_command, length); + status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; + goto out; + } + if (!req->sg_cnt) { nvmet_req_complete(req, 0); return; @@ -555,13 +559,13 @@ void nvmet_bdev_execute_zone_append(struct nvmet_req *req) if (sect >= get_capacity(req->ns->bdev->bd_disk)) { req->error_loc = offsetof(struct nvme_rw_command, slba); - status = NVME_SC_LBA_RANGE | NVME_SC_DNR; + status = NVME_SC_LBA_RANGE | NVME_STATUS_DNR; goto out; } if (sect & (bdev_zone_sectors(req->ns->bdev) - 1)) { req->error_loc = offsetof(struct nvme_rw_command, slba); - status = NVME_SC_INVALID_FIELD | NVME_SC_DNR; + status = NVME_SC_INVALID_FIELD | NVME_STATUS_DNR; goto out; } @@ -580,21 +584,17 @@ void nvmet_bdev_execute_zone_append(struct nvmet_req *req) bio->bi_opf |= REQ_FUA; for_each_sg(req->sg, sg, req->sg_cnt, sg_cnt) { - struct page *p = sg_page(sg); - unsigned int l = sg->length; - unsigned int o = sg->offset; - unsigned int ret; + unsigned int len = sg->length; - ret = bio_add_zone_append_page(bio, p, l, o); - if (ret != sg->length) { + if (bio_add_page(bio, sg_page(sg), len, sg->offset) != len) { status = NVME_SC_INTERNAL; goto out_put_bio; } - total_len += sg->length; + total_len += len; } - if (total_len != nvmet_rw_data_len(req)) { - status = NVME_SC_INTERNAL | NVME_SC_DNR; + if (total_len != data_len) { + status = NVME_SC_INTERNAL | NVME_STATUS_DNR; goto out_put_bio; } |