From f01d5cb24ea92494f9e093e6ee411364a686be99 Mon Sep 17 00:00:00 2001 From: Ilya Dryomov Date: Thu, 2 Jun 2016 16:45:08 +0200 Subject: libceph: rename ceph_entity_name_encode() -> ceph_auth_entity_name_encode() Clear up EntityName vs entity_name_t confusion. Signed-off-by: Ilya Dryomov Reviewed-by: Mike Christie Reviewed-by: Alex Elder --- include/linux/ceph/auth.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/ceph/auth.h b/include/linux/ceph/auth.h index 1563265d2097..374bb1c4ef52 100644 --- a/include/linux/ceph/auth.h +++ b/include/linux/ceph/auth.h @@ -104,7 +104,7 @@ extern int ceph_auth_build_hello(struct ceph_auth_client *ac, extern int ceph_handle_auth_reply(struct ceph_auth_client *ac, void *buf, size_t len, void *reply_buf, size_t reply_len); -extern int ceph_entity_name_encode(const char *name, void **p, void *end); +int ceph_auth_entity_name_encode(const char *name, void **p, void *end); extern int ceph_build_auth(struct ceph_auth_client *ac, void *msg_buf, size_t msg_len); -- cgit v1.2.3 From a4ed38d7a180f184a6e7aedd09db9ca4b1e6a71c Mon Sep 17 00:00:00 2001 From: Douglas Fuller Date: Fri, 17 Jul 2015 13:18:07 -0700 Subject: libceph: support for CEPH_OSD_OP_LIST_WATCHERS Add support for this Ceph OSD op, needed to support the RBD exclusive lock feature. Signed-off-by: Douglas Fuller [idryomov@gmail.com: refactor, misc fixes throughout] Signed-off-by: Ilya Dryomov Reviewed-by: Mike Christie Reviewed-by: Alex Elder --- include/linux/ceph/osd_client.h | 15 +++++- net/ceph/osd_client.c | 117 ++++++++++++++++++++++++++++++++++++++++ 2 files changed, 131 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/ceph/osd_client.h b/include/linux/ceph/osd_client.h index 858932304260..19821a191732 100644 --- a/include/linux/ceph/osd_client.h +++ b/include/linux/ceph/osd_client.h @@ -120,6 +120,9 @@ struct ceph_osd_req_op { struct ceph_osd_data request_data; struct ceph_osd_data response_data; } notify; + struct { + struct ceph_osd_data response_data; + } list_watchers; struct { u64 expected_object_size; u64 expected_write_size; @@ -249,6 +252,12 @@ struct ceph_osd_linger_request { size_t *preply_len; }; +struct ceph_watch_item { + struct ceph_entity_name name; + u64 cookie; + struct ceph_entity_addr addr; +}; + struct ceph_osd_client { struct ceph_client *client; @@ -346,7 +355,6 @@ extern void osd_req_op_cls_response_data_pages(struct ceph_osd_request *, struct page **pages, u64 length, u32 alignment, bool pages_from_pool, bool own_pages); - extern void osd_req_op_cls_init(struct ceph_osd_request *osd_req, unsigned int which, u16 opcode, const char *class, const char *method); @@ -434,5 +442,10 @@ int ceph_osdc_notify(struct ceph_osd_client *osdc, size_t *preply_len); int ceph_osdc_watch_check(struct ceph_osd_client *osdc, struct ceph_osd_linger_request *lreq); +int ceph_osdc_list_watchers(struct ceph_osd_client *osdc, + struct ceph_object_id *oid, + struct ceph_object_locator *oloc, + struct ceph_watch_item **watchers, + u32 *num_watchers); #endif diff --git a/net/ceph/osd_client.c b/net/ceph/osd_client.c index a97e7b506612..dd51ec8ce97f 100644 --- a/net/ceph/osd_client.c +++ b/net/ceph/osd_client.c @@ -338,6 +338,9 @@ static void osd_req_op_data_release(struct ceph_osd_request *osd_req, ceph_osd_data_release(&op->notify.request_data); ceph_osd_data_release(&op->notify.response_data); break; + case CEPH_OSD_OP_LIST_WATCHERS: + ceph_osd_data_release(&op->list_watchers.response_data); + break; default: break; } @@ -863,6 +866,8 @@ static u32 osd_req_encode_op(struct ceph_osd_op *dst, case CEPH_OSD_OP_NOTIFY: dst->notify.cookie = cpu_to_le64(src->notify.cookie); break; + case CEPH_OSD_OP_LIST_WATCHERS: + break; case CEPH_OSD_OP_SETALLOCHINT: dst->alloc_hint.expected_object_size = cpu_to_le64(src->alloc_hint.expected_object_size); @@ -1445,6 +1450,10 @@ static void setup_request_data(struct ceph_osd_request *req, ceph_osdc_msg_data_add(req->r_reply, &op->extent.osd_data); break; + case CEPH_OSD_OP_LIST_WATCHERS: + ceph_osdc_msg_data_add(req->r_reply, + &op->list_watchers.response_data); + break; /* both */ case CEPH_OSD_OP_CALL: @@ -3891,6 +3900,114 @@ int ceph_osdc_watch_check(struct ceph_osd_client *osdc, return ret; } +static int decode_watcher(void **p, void *end, struct ceph_watch_item *item) +{ + u8 struct_v; + u32 struct_len; + int ret; + + ret = ceph_start_decoding(p, end, 2, "watch_item_t", + &struct_v, &struct_len); + if (ret) + return ret; + + ceph_decode_copy(p, &item->name, sizeof(item->name)); + item->cookie = ceph_decode_64(p); + *p += 4; /* skip timeout_seconds */ + if (struct_v >= 2) { + ceph_decode_copy(p, &item->addr, sizeof(item->addr)); + ceph_decode_addr(&item->addr); + } + + dout("%s %s%llu cookie %llu addr %s\n", __func__, + ENTITY_NAME(item->name), item->cookie, + ceph_pr_addr(&item->addr.in_addr)); + return 0; +} + +static int decode_watchers(void **p, void *end, + struct ceph_watch_item **watchers, + u32 *num_watchers) +{ + u8 struct_v; + u32 struct_len; + int i; + int ret; + + ret = ceph_start_decoding(p, end, 1, "obj_list_watch_response_t", + &struct_v, &struct_len); + if (ret) + return ret; + + *num_watchers = ceph_decode_32(p); + *watchers = kcalloc(*num_watchers, sizeof(**watchers), GFP_NOIO); + if (!*watchers) + return -ENOMEM; + + for (i = 0; i < *num_watchers; i++) { + ret = decode_watcher(p, end, *watchers + i); + if (ret) { + kfree(*watchers); + return ret; + } + } + + return 0; +} + +/* + * On success, the caller is responsible for: + * + * kfree(watchers); + */ +int ceph_osdc_list_watchers(struct ceph_osd_client *osdc, + struct ceph_object_id *oid, + struct ceph_object_locator *oloc, + struct ceph_watch_item **watchers, + u32 *num_watchers) +{ + struct ceph_osd_request *req; + struct page **pages; + int ret; + + req = ceph_osdc_alloc_request(osdc, NULL, 1, false, GFP_NOIO); + if (!req) + return -ENOMEM; + + ceph_oid_copy(&req->r_base_oid, oid); + ceph_oloc_copy(&req->r_base_oloc, oloc); + req->r_flags = CEPH_OSD_FLAG_READ; + + ret = ceph_osdc_alloc_messages(req, GFP_NOIO); + if (ret) + goto out_put_req; + + pages = ceph_alloc_page_vector(1, GFP_NOIO); + if (IS_ERR(pages)) { + ret = PTR_ERR(pages); + goto out_put_req; + } + + osd_req_op_init(req, 0, CEPH_OSD_OP_LIST_WATCHERS, 0); + ceph_osd_data_pages_init(osd_req_op_data(req, 0, list_watchers, + response_data), + pages, PAGE_SIZE, 0, false, true); + + ceph_osdc_start_request(osdc, req, false); + ret = ceph_osdc_wait_request(osdc, req); + if (ret >= 0) { + void *p = page_address(pages[0]); + void *const end = p + req->r_ops[0].outdata_len; + + ret = decode_watchers(&p, end, watchers, num_watchers); + } + +out_put_req: + ceph_osdc_put_request(req); + return ret; +} +EXPORT_SYMBOL(ceph_osdc_list_watchers); + /* * Call all pending notify callbacks - for use after a watch is * unregistered, to make sure no more callbacks for it will be invoked -- cgit v1.2.3 From 428a715811fe74e8a8f09b830c8d3b5245096f8d Mon Sep 17 00:00:00 2001 From: Douglas Fuller Date: Wed, 17 Jun 2015 14:49:45 -0400 Subject: libceph: add ceph_osdc_call() single-page helper Add a convenience function to osd_client to send Ceph OSD 'class' ops. The interface assumes that the request and reply data each consist of single pages. Signed-off-by: Douglas Fuller Signed-off-by: Ilya Dryomov Reviewed-by: Mike Christie Reviewed-by: Alex Elder --- include/linux/ceph/osd_client.h | 8 +++++++ net/ceph/osd_client.c | 51 +++++++++++++++++++++++++++++++++++++++++ 2 files changed, 59 insertions(+) (limited to 'include') diff --git a/include/linux/ceph/osd_client.h b/include/linux/ceph/osd_client.h index 19821a191732..96337b15a60d 100644 --- a/include/linux/ceph/osd_client.h +++ b/include/linux/ceph/osd_client.h @@ -397,6 +397,14 @@ extern void ceph_osdc_sync(struct ceph_osd_client *osdc); extern void ceph_osdc_flush_notifies(struct ceph_osd_client *osdc); void ceph_osdc_maybe_request_map(struct ceph_osd_client *osdc); +int ceph_osdc_call(struct ceph_osd_client *osdc, + struct ceph_object_id *oid, + struct ceph_object_locator *oloc, + const char *class, const char *method, + unsigned int flags, + struct page *req_page, size_t req_len, + struct page *resp_page, size_t *resp_len); + extern int ceph_osdc_readpages(struct ceph_osd_client *osdc, struct ceph_vino vino, struct ceph_file_layout *layout, diff --git a/net/ceph/osd_client.c b/net/ceph/osd_client.c index dd51ec8ce97f..fbc6b7090c65 100644 --- a/net/ceph/osd_client.c +++ b/net/ceph/osd_client.c @@ -4026,6 +4026,57 @@ void ceph_osdc_maybe_request_map(struct ceph_osd_client *osdc) } EXPORT_SYMBOL(ceph_osdc_maybe_request_map); +/* + * Execute an OSD class method on an object. + * + * @flags: CEPH_OSD_FLAG_* + * @resp_len: out param for reply length + */ +int ceph_osdc_call(struct ceph_osd_client *osdc, + struct ceph_object_id *oid, + struct ceph_object_locator *oloc, + const char *class, const char *method, + unsigned int flags, + struct page *req_page, size_t req_len, + struct page *resp_page, size_t *resp_len) +{ + struct ceph_osd_request *req; + int ret; + + req = ceph_osdc_alloc_request(osdc, NULL, 1, false, GFP_NOIO); + if (!req) + return -ENOMEM; + + ceph_oid_copy(&req->r_base_oid, oid); + ceph_oloc_copy(&req->r_base_oloc, oloc); + req->r_flags = flags; + + ret = ceph_osdc_alloc_messages(req, GFP_NOIO); + if (ret) + goto out_put_req; + + osd_req_op_cls_init(req, 0, CEPH_OSD_OP_CALL, class, method); + if (req_page) + osd_req_op_cls_request_data_pages(req, 0, &req_page, req_len, + 0, false, false); + if (resp_page) + osd_req_op_cls_response_data_pages(req, 0, &resp_page, + PAGE_SIZE, 0, false, false); + + ceph_osdc_start_request(osdc, req, false); + ret = ceph_osdc_wait_request(osdc, req); + if (ret >= 0) { + ret = req->r_ops[0].rval; + if (resp_page) + *resp_len = req->r_ops[0].outdata_len; + } + +out_put_req: + ceph_osdc_put_request(req); + return ret; +} +EXPORT_SYMBOL(ceph_osdc_call); + /* * init, shutdown */ -- cgit v1.2.3 From f66241cb99dac861aa2cedb9f05ffa98d70cbc6e Mon Sep 17 00:00:00 2001 From: Douglas Fuller Date: Thu, 18 Jun 2015 13:06:10 -0700 Subject: libceph: support for advisory locking on RADOS objects This patch adds support for rados lock, unlock and break lock. Based heavily on code by Mike Christie . Signed-off-by: Douglas Fuller Signed-off-by: Ilya Dryomov Reviewed-by: Mike Christie Reviewed-by: Alex Elder --- include/linux/ceph/cls_lock_client.h | 27 ++++++ net/ceph/Makefile | 1 + net/ceph/cls_lock_client.c | 180 +++++++++++++++++++++++++++++++++++ 3 files changed, 208 insertions(+) create mode 100644 include/linux/ceph/cls_lock_client.h create mode 100644 net/ceph/cls_lock_client.c (limited to 'include') diff --git a/include/linux/ceph/cls_lock_client.h b/include/linux/ceph/cls_lock_client.h new file mode 100644 index 000000000000..4e4dffef22bb --- /dev/null +++ b/include/linux/ceph/cls_lock_client.h @@ -0,0 +1,27 @@ +#ifndef _LINUX_CEPH_CLS_LOCK_CLIENT_H +#define _LINUX_CEPH_CLS_LOCK_CLIENT_H + +#include + +enum ceph_cls_lock_type { + CEPH_CLS_LOCK_NONE = 0, + CEPH_CLS_LOCK_EXCLUSIVE = 1, + CEPH_CLS_LOCK_SHARED = 2, +}; + +int ceph_cls_lock(struct ceph_osd_client *osdc, + struct ceph_object_id *oid, + struct ceph_object_locator *oloc, + char *lock_name, u8 type, char *cookie, + char *tag, char *desc, u8 flags); +int ceph_cls_unlock(struct ceph_osd_client *osdc, + struct ceph_object_id *oid, + struct ceph_object_locator *oloc, + char *lock_name, char *cookie); +int ceph_cls_break_lock(struct ceph_osd_client *osdc, + struct ceph_object_id *oid, + struct ceph_object_locator *oloc, + char *lock_name, char *cookie, + struct ceph_entity_name *locker); + +#endif diff --git a/net/ceph/Makefile b/net/ceph/Makefile index 84cbed630c4b..6a5180903e7b 100644 --- a/net/ceph/Makefile +++ b/net/ceph/Makefile @@ -5,6 +5,7 @@ obj-$(CONFIG_CEPH_LIB) += libceph.o libceph-y := ceph_common.o messenger.o msgpool.o buffer.o pagelist.o \ mon_client.o \ + cls_lock_client.o \ osd_client.o osdmap.o crush/crush.o crush/mapper.o crush/hash.o \ debugfs.o \ auth.o auth_none.o \ diff --git a/net/ceph/cls_lock_client.c b/net/ceph/cls_lock_client.c new file mode 100644 index 000000000000..2a314537f958 --- /dev/null +++ b/net/ceph/cls_lock_client.c @@ -0,0 +1,180 @@ +#include + +#include +#include + +#include +#include + +/** + * ceph_cls_lock - grab rados lock for object + * @oid, @oloc: object to lock + * @lock_name: the name of the lock + * @type: lock type (CEPH_CLS_LOCK_EXCLUSIVE or CEPH_CLS_LOCK_SHARED) + * @cookie: user-defined identifier for this instance of the lock + * @tag: user-defined tag + * @desc: user-defined lock description + * @flags: lock flags + * + * All operations on the same lock should use the same tag. + */ +int ceph_cls_lock(struct ceph_osd_client *osdc, + struct ceph_object_id *oid, + struct ceph_object_locator *oloc, + char *lock_name, u8 type, char *cookie, + char *tag, char *desc, u8 flags) +{ + int lock_op_buf_size; + int name_len = strlen(lock_name); + int cookie_len = strlen(cookie); + int tag_len = strlen(tag); + int desc_len = strlen(desc); + void *p, *end; + struct page *lock_op_page; + struct timespec mtime; + int ret; + + lock_op_buf_size = name_len + sizeof(__le32) + + cookie_len + sizeof(__le32) + + tag_len + sizeof(__le32) + + desc_len + sizeof(__le32) + + sizeof(struct ceph_timespec) + + /* flag and type */ + sizeof(u8) + sizeof(u8) + + CEPH_ENCODING_START_BLK_LEN; + if (lock_op_buf_size > PAGE_SIZE) + return -E2BIG; + + lock_op_page = alloc_page(GFP_NOIO); + if (!lock_op_page) + return -ENOMEM; + + p = page_address(lock_op_page); + end = p + lock_op_buf_size; + + /* encode cls_lock_lock_op struct */ + ceph_start_encoding(&p, 1, 1, + lock_op_buf_size - CEPH_ENCODING_START_BLK_LEN); + ceph_encode_string(&p, end, lock_name, name_len); + ceph_encode_8(&p, type); + ceph_encode_string(&p, end, cookie, cookie_len); + ceph_encode_string(&p, end, tag, tag_len); + ceph_encode_string(&p, end, desc, desc_len); + /* only support infinite duration */ + memset(&mtime, 0, sizeof(mtime)); + ceph_encode_timespec(p, &mtime); + p += sizeof(struct ceph_timespec); + ceph_encode_8(&p, flags); + + dout("%s lock_name %s type %d cookie %s tag %s desc %s flags 0x%x\n", + __func__, lock_name, type, cookie, tag, desc, flags); + ret = ceph_osdc_call(osdc, oid, oloc, "lock", "lock", + CEPH_OSD_FLAG_WRITE | CEPH_OSD_FLAG_ONDISK, + lock_op_page, lock_op_buf_size, NULL, NULL); + + dout("%s: status %d\n", __func__, ret); + __free_page(lock_op_page); + return ret; +} +EXPORT_SYMBOL(ceph_cls_lock); + +/** + * ceph_cls_unlock - release rados lock for object + * @oid, @oloc: object to lock + * @lock_name: the name of the lock + * @cookie: user-defined identifier for this instance of the lock + */ +int ceph_cls_unlock(struct ceph_osd_client *osdc, + struct ceph_object_id *oid, + struct ceph_object_locator *oloc, + char *lock_name, char *cookie) +{ + int unlock_op_buf_size; + int name_len = strlen(lock_name); + int cookie_len = strlen(cookie); + void *p, *end; + struct page *unlock_op_page; + int ret; + + unlock_op_buf_size = name_len + sizeof(__le32) + + cookie_len + sizeof(__le32) + + CEPH_ENCODING_START_BLK_LEN; + if (unlock_op_buf_size > PAGE_SIZE) + return -E2BIG; + + unlock_op_page = alloc_page(GFP_NOIO); + if (!unlock_op_page) + return -ENOMEM; + + p = page_address(unlock_op_page); + end = p + unlock_op_buf_size; + + /* encode cls_lock_unlock_op struct */ + ceph_start_encoding(&p, 1, 1, + unlock_op_buf_size - CEPH_ENCODING_START_BLK_LEN); + ceph_encode_string(&p, end, lock_name, name_len); + ceph_encode_string(&p, end, cookie, cookie_len); + + dout("%s lock_name %s cookie %s\n", __func__, lock_name, cookie); + ret = ceph_osdc_call(osdc, oid, oloc, "lock", "unlock", + CEPH_OSD_FLAG_WRITE | CEPH_OSD_FLAG_ONDISK, + unlock_op_page, unlock_op_buf_size, NULL, NULL); + + dout("%s: status %d\n", __func__, ret); + __free_page(unlock_op_page); + return ret; +} +EXPORT_SYMBOL(ceph_cls_unlock); + +/** + * ceph_cls_break_lock - release rados lock for object for specified client + * @oid, @oloc: object to lock + * @lock_name: the name of the lock + * @cookie: user-defined identifier for this instance of the lock + * @locker: current lock owner + */ +int ceph_cls_break_lock(struct ceph_osd_client *osdc, + struct ceph_object_id *oid, + struct ceph_object_locator *oloc, + char *lock_name, char *cookie, + struct ceph_entity_name *locker) +{ + int break_op_buf_size; + int name_len = strlen(lock_name); + int cookie_len = strlen(cookie); + struct page *break_op_page; + void *p, *end; + int ret; + + break_op_buf_size = name_len + sizeof(__le32) + + cookie_len + sizeof(__le32) + + sizeof(u8) + sizeof(__le64) + + CEPH_ENCODING_START_BLK_LEN; + if (break_op_buf_size > PAGE_SIZE) + return -E2BIG; + + break_op_page = alloc_page(GFP_NOIO); + if (!break_op_page) + return -ENOMEM; + + p = page_address(break_op_page); + end = p + break_op_buf_size; + + /* encode cls_lock_break_op struct */ + ceph_start_encoding(&p, 1, 1, + break_op_buf_size - CEPH_ENCODING_START_BLK_LEN); + ceph_encode_string(&p, end, lock_name, name_len); + ceph_encode_copy(&p, locker, sizeof(*locker)); + ceph_encode_string(&p, end, cookie, cookie_len); + + dout("%s lock_name %s cookie %s locker %s%llu\n", __func__, lock_name, + cookie, ENTITY_NAME(*locker)); + ret = ceph_osdc_call(osdc, oid, oloc, "lock", "break_lock", + CEPH_OSD_FLAG_WRITE | CEPH_OSD_FLAG_ONDISK, + break_op_page, break_op_buf_size, NULL, NULL); + + dout("%s: status %d\n", __func__, ret); + __free_page(break_op_page); + return ret; +} +EXPORT_SYMBOL(ceph_cls_break_lock); -- cgit v1.2.3 From d4ed4a530562881cc5225050e42d96034f405aae Mon Sep 17 00:00:00 2001 From: Douglas Fuller Date: Mon, 29 Jun 2015 16:55:42 -0700 Subject: libceph: support for lock.lock_info Add an interface for the Ceph OSD lock.lock_info method and associated data structures. Based heavily on code by Mike Christie . Signed-off-by: Douglas Fuller [idryomov@gmail.com: refactor, misc fixes throughout] Signed-off-by: Ilya Dryomov Reviewed-by: Mike Christie Reviewed-by: Alex Elder --- include/linux/ceph/cls_lock_client.h | 22 ++++++ net/ceph/cls_lock_client.c | 145 +++++++++++++++++++++++++++++++++++ 2 files changed, 167 insertions(+) (limited to 'include') diff --git a/include/linux/ceph/cls_lock_client.h b/include/linux/ceph/cls_lock_client.h index 4e4dffef22bb..84884d8d4710 100644 --- a/include/linux/ceph/cls_lock_client.h +++ b/include/linux/ceph/cls_lock_client.h @@ -9,6 +9,20 @@ enum ceph_cls_lock_type { CEPH_CLS_LOCK_SHARED = 2, }; +struct ceph_locker_id { + struct ceph_entity_name name; /* locker's client name */ + char *cookie; /* locker's cookie */ +}; + +struct ceph_locker_info { + struct ceph_entity_addr addr; /* locker's address */ +}; + +struct ceph_locker { + struct ceph_locker_id id; + struct ceph_locker_info info; +}; + int ceph_cls_lock(struct ceph_osd_client *osdc, struct ceph_object_id *oid, struct ceph_object_locator *oloc, @@ -24,4 +38,12 @@ int ceph_cls_break_lock(struct ceph_osd_client *osdc, char *lock_name, char *cookie, struct ceph_entity_name *locker); +void ceph_free_lockers(struct ceph_locker *lockers, u32 num_lockers); + +int ceph_cls_lock_info(struct ceph_osd_client *osdc, + struct ceph_object_id *oid, + struct ceph_object_locator *oloc, + char *lock_name, u8 *type, char **tag, + struct ceph_locker **lockers, u32 *num_lockers); + #endif diff --git a/net/ceph/cls_lock_client.c b/net/ceph/cls_lock_client.c index 2a314537f958..50f040fdb2a9 100644 --- a/net/ceph/cls_lock_client.c +++ b/net/ceph/cls_lock_client.c @@ -178,3 +178,148 @@ int ceph_cls_break_lock(struct ceph_osd_client *osdc, return ret; } EXPORT_SYMBOL(ceph_cls_break_lock); + +void ceph_free_lockers(struct ceph_locker *lockers, u32 num_lockers) +{ + int i; + + for (i = 0; i < num_lockers; i++) + kfree(lockers[i].id.cookie); + kfree(lockers); +} +EXPORT_SYMBOL(ceph_free_lockers); + +static int decode_locker(void **p, void *end, struct ceph_locker *locker) +{ + u8 struct_v; + u32 len; + char *s; + int ret; + + ret = ceph_start_decoding(p, end, 1, "locker_id_t", &struct_v, &len); + if (ret) + return ret; + + ceph_decode_copy(p, &locker->id.name, sizeof(locker->id.name)); + s = ceph_extract_encoded_string(p, end, NULL, GFP_NOIO); + if (IS_ERR(s)) + return PTR_ERR(s); + + locker->id.cookie = s; + + ret = ceph_start_decoding(p, end, 1, "locker_info_t", &struct_v, &len); + if (ret) + return ret; + + *p += sizeof(struct ceph_timespec); /* skip expiration */ + ceph_decode_copy(p, &locker->info.addr, sizeof(locker->info.addr)); + ceph_decode_addr(&locker->info.addr); + len = ceph_decode_32(p); + *p += len; /* skip description */ + + dout("%s %s%llu cookie %s addr %s\n", __func__, + ENTITY_NAME(locker->id.name), locker->id.cookie, + ceph_pr_addr(&locker->info.addr.in_addr)); + return 0; +} + +static int decode_lockers(void **p, void *end, u8 *type, char **tag, + struct ceph_locker **lockers, u32 *num_lockers) +{ + u8 struct_v; + u32 struct_len; + char *s; + int i; + int ret; + + ret = ceph_start_decoding(p, end, 1, "cls_lock_get_info_reply", + &struct_v, &struct_len); + if (ret) + return ret; + + *num_lockers = ceph_decode_32(p); + *lockers = kcalloc(*num_lockers, sizeof(**lockers), GFP_NOIO); + if (!*lockers) + return -ENOMEM; + + for (i = 0; i < *num_lockers; i++) { + ret = decode_locker(p, end, *lockers + i); + if (ret) + goto err_free_lockers; + } + + *type = ceph_decode_8(p); + s = ceph_extract_encoded_string(p, end, NULL, GFP_NOIO); + if (IS_ERR(s)) { + ret = PTR_ERR(s); + goto err_free_lockers; + } + + *tag = s; + return 0; + +err_free_lockers: + ceph_free_lockers(*lockers, *num_lockers); + return ret; +} + +/* + * On success, the caller is responsible for: + * + * kfree(tag); + * ceph_free_lockers(lockers, num_lockers); + */ +int ceph_cls_lock_info(struct ceph_osd_client *osdc, + struct ceph_object_id *oid, + struct ceph_object_locator *oloc, + char *lock_name, u8 *type, char **tag, + struct ceph_locker **lockers, u32 *num_lockers) +{ + int get_info_op_buf_size; + int name_len = strlen(lock_name); + struct page *get_info_op_page, *reply_page; + size_t reply_len; + void *p, *end; + int ret; + + get_info_op_buf_size = name_len + sizeof(__le32) + + CEPH_ENCODING_START_BLK_LEN; + if (get_info_op_buf_size > PAGE_SIZE) + return -E2BIG; + + get_info_op_page = alloc_page(GFP_NOIO); + if (!get_info_op_page) + return -ENOMEM; + + reply_page = alloc_page(GFP_NOIO); + if (!reply_page) { + __free_page(get_info_op_page); + return -ENOMEM; + } + + p = page_address(get_info_op_page); + end = p + get_info_op_buf_size; + + /* encode cls_lock_get_info_op struct */ + ceph_start_encoding(&p, 1, 1, + get_info_op_buf_size - CEPH_ENCODING_START_BLK_LEN); + ceph_encode_string(&p, end, lock_name, name_len); + + dout("%s lock_name %s\n", __func__, lock_name); + ret = ceph_osdc_call(osdc, oid, oloc, "lock", "get_info", + CEPH_OSD_FLAG_READ, get_info_op_page, + get_info_op_buf_size, reply_page, &reply_len); + + dout("%s: status %d\n", __func__, ret); + if (ret >= 0) { + p = page_address(reply_page); + end = p + reply_len; + + ret = decode_lockers(&p, end, type, tag, lockers, num_lockers); + } + + __free_page(get_info_op_page); + __free_page(reply_page); + return ret; +} +EXPORT_SYMBOL(ceph_cls_lock_info); -- cgit v1.2.3 From 6305a3b415157759bfe4b50a643fac22ea229f5c Mon Sep 17 00:00:00 2001 From: Douglas Fuller Date: Wed, 22 Jul 2015 20:59:52 -0400 Subject: libceph: support for blacklisting clients Reuse ceph_mon_generic_request infrastructure for sending monitor commands. In particular, add support for 'blacklist add' to prevent other, non-responsive clients from making further updates. Signed-off-by: Douglas Fuller [idryomov@gmail.com: refactor, misc fixes throughout] Signed-off-by: Ilya Dryomov Reviewed-by: Mike Christie Reviewed-by: Alex Elder --- include/linux/ceph/ceph_fs.h | 11 ++++++ include/linux/ceph/mon_client.h | 3 ++ net/ceph/mon_client.c | 82 +++++++++++++++++++++++++++++++++++++++++ 3 files changed, 96 insertions(+) (limited to 'include') diff --git a/include/linux/ceph/ceph_fs.h b/include/linux/ceph/ceph_fs.h index 7868d602c0a0..c086e63dcee1 100644 --- a/include/linux/ceph/ceph_fs.h +++ b/include/linux/ceph/ceph_fs.h @@ -138,6 +138,9 @@ struct ceph_dir_layout { #define CEPH_MSG_POOLOP_REPLY 48 #define CEPH_MSG_POOLOP 49 +/* mon commands */ +#define CEPH_MSG_MON_COMMAND 50 +#define CEPH_MSG_MON_COMMAND_ACK 51 /* osd */ #define CEPH_MSG_OSD_MAP 41 @@ -176,6 +179,14 @@ struct ceph_mon_statfs_reply { struct ceph_statfs st; } __attribute__ ((packed)); +struct ceph_mon_command { + struct ceph_mon_request_header monhdr; + struct ceph_fsid fsid; + __le32 num_strs; /* always 1 */ + __le32 str_len; + char str[]; +} __attribute__ ((packed)); + struct ceph_osd_getmap { struct ceph_mon_request_header monhdr; struct ceph_fsid fsid; diff --git a/include/linux/ceph/mon_client.h b/include/linux/ceph/mon_client.h index 24d704d1ea5c..d5a3ecea578d 100644 --- a/include/linux/ceph/mon_client.h +++ b/include/linux/ceph/mon_client.h @@ -141,6 +141,9 @@ int ceph_monc_get_version(struct ceph_mon_client *monc, const char *what, int ceph_monc_get_version_async(struct ceph_mon_client *monc, const char *what, ceph_monc_callback_t cb, u64 private_data); +int ceph_monc_blacklist_add(struct ceph_mon_client *monc, + struct ceph_entity_addr *client_addr); + extern int ceph_monc_open_session(struct ceph_mon_client *monc); extern int ceph_monc_validate_auth(struct ceph_mon_client *monc); diff --git a/net/ceph/mon_client.c b/net/ceph/mon_client.c index ef34a02719d7..a8effc8b7280 100644 --- a/net/ceph/mon_client.c +++ b/net/ceph/mon_client.c @@ -835,6 +835,83 @@ int ceph_monc_get_version_async(struct ceph_mon_client *monc, const char *what, } EXPORT_SYMBOL(ceph_monc_get_version_async); +static void handle_command_ack(struct ceph_mon_client *monc, + struct ceph_msg *msg) +{ + struct ceph_mon_generic_request *req; + void *p = msg->front.iov_base; + void *const end = p + msg->front_alloc_len; + u64 tid = le64_to_cpu(msg->hdr.tid); + + dout("%s msg %p tid %llu\n", __func__, msg, tid); + + ceph_decode_need(&p, end, sizeof(struct ceph_mon_request_header) + + sizeof(u32), bad); + p += sizeof(struct ceph_mon_request_header); + + mutex_lock(&monc->mutex); + req = lookup_generic_request(&monc->generic_request_tree, tid); + if (!req) { + mutex_unlock(&monc->mutex); + return; + } + + req->result = ceph_decode_32(&p); + __finish_generic_request(req); + mutex_unlock(&monc->mutex); + + complete_generic_request(req); + return; + +bad: + pr_err("corrupt mon_command ack, tid %llu\n", tid); + ceph_msg_dump(msg); +} + +int ceph_monc_blacklist_add(struct ceph_mon_client *monc, + struct ceph_entity_addr *client_addr) +{ + struct ceph_mon_generic_request *req; + struct ceph_mon_command *h; + int ret = -ENOMEM; + int len; + + req = alloc_generic_request(monc, GFP_NOIO); + if (!req) + goto out; + + req->request = ceph_msg_new(CEPH_MSG_MON_COMMAND, 256, GFP_NOIO, true); + if (!req->request) + goto out; + + req->reply = ceph_msg_new(CEPH_MSG_MON_COMMAND_ACK, 512, GFP_NOIO, + true); + if (!req->reply) + goto out; + + mutex_lock(&monc->mutex); + register_generic_request(req); + h = req->request->front.iov_base; + h->monhdr.have_version = 0; + h->monhdr.session_mon = cpu_to_le16(-1); + h->monhdr.session_mon_tid = 0; + h->fsid = monc->monmap->fsid; + h->num_strs = cpu_to_le32(1); + len = sprintf(h->str, "{ \"prefix\": \"osd blacklist\", \ + \"blacklistop\": \"add\", \ + \"addr\": \"%pISpc/%u\" }", + &client_addr->in_addr, le32_to_cpu(client_addr->nonce)); + h->str_len = cpu_to_le32(len); + send_generic_request(monc, req); + mutex_unlock(&monc->mutex); + + ret = wait_generic_request(req); +out: + put_generic_request(req); + return ret; +} +EXPORT_SYMBOL(ceph_monc_blacklist_add); + /* * Resend pending generic requests. */ @@ -1139,6 +1216,10 @@ static void dispatch(struct ceph_connection *con, struct ceph_msg *msg) handle_get_version_reply(monc, msg); break; + case CEPH_MSG_MON_COMMAND_ACK: + handle_command_ack(monc, msg); + break; + case CEPH_MSG_MON_MAP: ceph_monc_handle_map(monc, msg); break; @@ -1178,6 +1259,7 @@ static struct ceph_msg *mon_alloc_msg(struct ceph_connection *con, m = ceph_msg_get(monc->m_subscribe_ack); break; case CEPH_MSG_STATFS_REPLY: + case CEPH_MSG_MON_COMMAND_ACK: return get_generic_reply(con, hdr, skip); case CEPH_MSG_AUTH_REPLY: m = ceph_msg_get(monc->m_auth_reply); -- cgit v1.2.3 From 033268a5f01270f0ef20d1a9a078b157f4af97f8 Mon Sep 17 00:00:00 2001 From: Ilya Dryomov Date: Fri, 12 Aug 2016 14:59:58 +0200 Subject: libceph: rename ceph_client_id() -> ceph_client_gid() It's gid / global_id in other places. Signed-off-by: Ilya Dryomov Reviewed-by: Mike Christie Reviewed-by: Alex Elder --- drivers/block/rbd.c | 2 +- include/linux/ceph/libceph.h | 2 +- net/ceph/ceph_common.c | 7 ++++--- 3 files changed, 6 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/drivers/block/rbd.c b/drivers/block/rbd.c index 6c6519f6492a..e0585e9040f1 100644 --- a/drivers/block/rbd.c +++ b/drivers/block/rbd.c @@ -3758,7 +3758,7 @@ static ssize_t rbd_client_id_show(struct device *dev, struct rbd_device *rbd_dev = dev_to_rbd_dev(dev); return sprintf(buf, "client%lld\n", - ceph_client_id(rbd_dev->rbd_client->client)); + ceph_client_gid(rbd_dev->rbd_client->client)); } static ssize_t rbd_pool_show(struct device *dev, diff --git a/include/linux/ceph/libceph.h b/include/linux/ceph/libceph.h index 83fc1fff7061..b4cffff70e44 100644 --- a/include/linux/ceph/libceph.h +++ b/include/linux/ceph/libceph.h @@ -264,7 +264,7 @@ extern struct ceph_client *ceph_create_client(struct ceph_options *opt, void *private, u64 supported_features, u64 required_features); -extern u64 ceph_client_id(struct ceph_client *client); +u64 ceph_client_gid(struct ceph_client *client); extern void ceph_destroy_client(struct ceph_client *client); extern int __ceph_open_session(struct ceph_client *client, unsigned long started); diff --git a/net/ceph/ceph_common.c b/net/ceph/ceph_common.c index bddfcf6f09c2..8a7921767308 100644 --- a/net/ceph/ceph_common.c +++ b/net/ceph/ceph_common.c @@ -566,11 +566,11 @@ int ceph_print_client_options(struct seq_file *m, struct ceph_client *client) } EXPORT_SYMBOL(ceph_print_client_options); -u64 ceph_client_id(struct ceph_client *client) +u64 ceph_client_gid(struct ceph_client *client) { return client->monc.auth->global_id; } -EXPORT_SYMBOL(ceph_client_id); +EXPORT_SYMBOL(ceph_client_gid); /* * create a fresh client instance @@ -685,7 +685,8 @@ int __ceph_open_session(struct ceph_client *client, unsigned long started) return client->auth_err; } - pr_info("client%llu fsid %pU\n", ceph_client_id(client), &client->fsid); + pr_info("client%llu fsid %pU\n", ceph_client_gid(client), + &client->fsid); ceph_debugfs_client_init(client); return 0; -- cgit v1.2.3 From 005a07bf0a92e7f0e73fc9a6c9acc992c5dbd00c Mon Sep 17 00:00:00 2001 From: Ilya Dryomov Date: Thu, 18 Aug 2016 18:38:43 +0200 Subject: rbd: add 'client_addr' sysfs rbd device attribute Export client addr/nonce, so userspace can check if a image is being blacklisted. Signed-off-by: Mike Christie [idryomov@gmail.com: ceph_client_addr(), endianess fix] Signed-off-by: Ilya Dryomov --- Documentation/ABI/testing/sysfs-bus-rbd | 6 ++++++ drivers/block/rbd.c | 13 +++++++++++++ include/linux/ceph/libceph.h | 1 + net/ceph/ceph_common.c | 6 ++++++ 4 files changed, 26 insertions(+) (limited to 'include') diff --git a/Documentation/ABI/testing/sysfs-bus-rbd b/Documentation/ABI/testing/sysfs-bus-rbd index 2ddd680929d8..273e27f2491f 100644 --- a/Documentation/ABI/testing/sysfs-bus-rbd +++ b/Documentation/ABI/testing/sysfs-bus-rbd @@ -43,6 +43,12 @@ Description: Available only if rbd module is inserted with single_major Entries under /sys/bus/rbd/devices// -------------------------------------------- +client_addr + + The ceph unique client entity_addr_t (address + nonce). + The format is
:/: '1.2.3.4:1234/5678' or + '[1:2:3:4:5:6:7:8]:1234/5678'. (August 2016, since 4.9.) + client_id The ceph unique client id that was assigned for this specific session. diff --git a/drivers/block/rbd.c b/drivers/block/rbd.c index fd1a9891b348..69d76c3afcdd 100644 --- a/drivers/block/rbd.c +++ b/drivers/block/rbd.c @@ -4592,6 +4592,17 @@ static ssize_t rbd_minor_show(struct device *dev, return sprintf(buf, "%d\n", rbd_dev->minor); } +static ssize_t rbd_client_addr_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct rbd_device *rbd_dev = dev_to_rbd_dev(dev); + struct ceph_entity_addr *client_addr = + ceph_client_addr(rbd_dev->rbd_client->client); + + return sprintf(buf, "%pISpc/%u\n", &client_addr->in_addr, + le32_to_cpu(client_addr->nonce)); +} + static ssize_t rbd_client_id_show(struct device *dev, struct device_attribute *attr, char *buf) { @@ -4702,6 +4713,7 @@ static DEVICE_ATTR(size, S_IRUGO, rbd_size_show, NULL); static DEVICE_ATTR(features, S_IRUGO, rbd_features_show, NULL); static DEVICE_ATTR(major, S_IRUGO, rbd_major_show, NULL); static DEVICE_ATTR(minor, S_IRUGO, rbd_minor_show, NULL); +static DEVICE_ATTR(client_addr, S_IRUGO, rbd_client_addr_show, NULL); static DEVICE_ATTR(client_id, S_IRUGO, rbd_client_id_show, NULL); static DEVICE_ATTR(pool, S_IRUGO, rbd_pool_show, NULL); static DEVICE_ATTR(pool_id, S_IRUGO, rbd_pool_id_show, NULL); @@ -4716,6 +4728,7 @@ static struct attribute *rbd_attrs[] = { &dev_attr_features.attr, &dev_attr_major.attr, &dev_attr_minor.attr, + &dev_attr_client_addr.attr, &dev_attr_client_id.attr, &dev_attr_pool.attr, &dev_attr_pool_id.attr, diff --git a/include/linux/ceph/libceph.h b/include/linux/ceph/libceph.h index b4cffff70e44..1816c5e26581 100644 --- a/include/linux/ceph/libceph.h +++ b/include/linux/ceph/libceph.h @@ -264,6 +264,7 @@ extern struct ceph_client *ceph_create_client(struct ceph_options *opt, void *private, u64 supported_features, u64 required_features); +struct ceph_entity_addr *ceph_client_addr(struct ceph_client *client); u64 ceph_client_gid(struct ceph_client *client); extern void ceph_destroy_client(struct ceph_client *client); extern int __ceph_open_session(struct ceph_client *client, diff --git a/net/ceph/ceph_common.c b/net/ceph/ceph_common.c index 8a7921767308..464e88599b9d 100644 --- a/net/ceph/ceph_common.c +++ b/net/ceph/ceph_common.c @@ -566,6 +566,12 @@ int ceph_print_client_options(struct seq_file *m, struct ceph_client *client) } EXPORT_SYMBOL(ceph_print_client_options); +struct ceph_entity_addr *ceph_client_addr(struct ceph_client *client) +{ + return &client->msgr.inst.addr; +} +EXPORT_SYMBOL(ceph_client_addr); + u64 ceph_client_gid(struct ceph_client *client) { return client->monc.auth->global_id; -- cgit v1.2.3 From fcff415c9421b417ef91d48f546f3c4566ddc358 Mon Sep 17 00:00:00 2001 From: "Yan, Zheng" Date: Wed, 14 Sep 2016 16:39:51 +0800 Subject: ceph: handle CEPH_SESSION_REJECT message Signed-off-by: Yan, Zheng --- fs/ceph/mds_client.c | 27 ++++++++++++++++++++++----- fs/ceph/mds_client.h | 1 + fs/ceph/strings.c | 2 ++ include/linux/ceph/ceph_fs.h | 1 + 4 files changed, 26 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c index f72d4ae303b2..277ac9c5c99c 100644 --- a/fs/ceph/mds_client.c +++ b/fs/ceph/mds_client.c @@ -370,6 +370,7 @@ const char *ceph_session_state_name(int s) case CEPH_MDS_SESSION_CLOSING: return "closing"; case CEPH_MDS_SESSION_RESTARTING: return "restarting"; case CEPH_MDS_SESSION_RECONNECTING: return "reconnecting"; + case CEPH_MDS_SESSION_REJECTED: return "rejected"; default: return "???"; } } @@ -1378,7 +1379,7 @@ static int request_close_session(struct ceph_mds_client *mdsc, if (!msg) return -ENOMEM; ceph_con_send(&session->s_con, msg); - return 0; + return 1; } /* @@ -2131,6 +2132,10 @@ static int __do_request(struct ceph_mds_client *mdsc, ceph_session_state_name(session->s_state)); if (session->s_state != CEPH_MDS_SESSION_OPEN && session->s_state != CEPH_MDS_SESSION_HUNG) { + if (session->s_state == CEPH_MDS_SESSION_REJECTED) { + err = -EACCES; + goto out_session; + } if (session->s_state == CEPH_MDS_SESSION_NEW || session->s_state == CEPH_MDS_SESSION_CLOSING) __open_session(mdsc, session); @@ -2652,6 +2657,15 @@ static void handle_session(struct ceph_mds_session *session, wake_up_session_caps(session, 0); break; + case CEPH_SESSION_REJECT: + WARN_ON(session->s_state != CEPH_MDS_SESSION_OPENING); + pr_info("mds%d rejected session\n", session->s_mds); + session->s_state = CEPH_MDS_SESSION_REJECTED; + cleanup_session_requests(mdsc, session); + remove_session_caps(session); + wake = 2; /* for good measure */ + break; + default: pr_err("mdsc_handle_session bad op %d mds%d\n", op, mds); WARN_ON(1); @@ -3557,11 +3571,11 @@ void ceph_mdsc_sync(struct ceph_mds_client *mdsc) /* * true if all sessions are closed, or we force unmount */ -static bool done_closing_sessions(struct ceph_mds_client *mdsc) +static bool done_closing_sessions(struct ceph_mds_client *mdsc, int skipped) { if (ACCESS_ONCE(mdsc->fsc->mount_state) == CEPH_MOUNT_SHUTDOWN) return true; - return atomic_read(&mdsc->num_sessions) == 0; + return atomic_read(&mdsc->num_sessions) <= skipped; } /* @@ -3572,6 +3586,7 @@ void ceph_mdsc_close_sessions(struct ceph_mds_client *mdsc) struct ceph_options *opts = mdsc->fsc->client->options; struct ceph_mds_session *session; int i; + int skipped = 0; dout("close_sessions\n"); @@ -3583,7 +3598,8 @@ void ceph_mdsc_close_sessions(struct ceph_mds_client *mdsc) continue; mutex_unlock(&mdsc->mutex); mutex_lock(&session->s_mutex); - __close_session(mdsc, session); + if (__close_session(mdsc, session) <= 0) + skipped++; mutex_unlock(&session->s_mutex); ceph_put_mds_session(session); mutex_lock(&mdsc->mutex); @@ -3591,7 +3607,8 @@ void ceph_mdsc_close_sessions(struct ceph_mds_client *mdsc) mutex_unlock(&mdsc->mutex); dout("waiting for sessions to close\n"); - wait_event_timeout(mdsc->session_close_wq, done_closing_sessions(mdsc), + wait_event_timeout(mdsc->session_close_wq, + done_closing_sessions(mdsc, skipped), ceph_timeout_jiffies(opts->mount_timeout)); /* tear down remaining sessions */ diff --git a/fs/ceph/mds_client.h b/fs/ceph/mds_client.h index 6b3679737d4a..3c6f77b7bb02 100644 --- a/fs/ceph/mds_client.h +++ b/fs/ceph/mds_client.h @@ -121,6 +121,7 @@ enum { CEPH_MDS_SESSION_CLOSING = 5, CEPH_MDS_SESSION_RESTARTING = 6, CEPH_MDS_SESSION_RECONNECTING = 7, + CEPH_MDS_SESSION_REJECTED = 8, }; struct ceph_mds_session { diff --git a/fs/ceph/strings.c b/fs/ceph/strings.c index 89e6bc321df3..913dea163d5c 100644 --- a/fs/ceph/strings.c +++ b/fs/ceph/strings.c @@ -43,6 +43,8 @@ const char *ceph_session_op_name(int op) case CEPH_SESSION_RECALL_STATE: return "recall_state"; case CEPH_SESSION_FLUSHMSG: return "flushmsg"; case CEPH_SESSION_FLUSHMSG_ACK: return "flushmsg_ack"; + case CEPH_SESSION_FORCE_RO: return "force_ro"; + case CEPH_SESSION_REJECT: return "reject"; } return "???"; } diff --git a/include/linux/ceph/ceph_fs.h b/include/linux/ceph/ceph_fs.h index c086e63dcee1..f96de8de4fa7 100644 --- a/include/linux/ceph/ceph_fs.h +++ b/include/linux/ceph/ceph_fs.h @@ -281,6 +281,7 @@ enum { CEPH_SESSION_FLUSHMSG, CEPH_SESSION_FLUSHMSG_ACK, CEPH_SESSION_FORCE_RO, + CEPH_SESSION_REJECT, }; extern const char *ceph_session_op_name(int op); -- cgit v1.2.3