From 3dbb60739acab60b8d2a7c79b0154d11a84341d5 Mon Sep 17 00:00:00 2001 From: Boaz Harrosh Date: Thu, 2 Jul 2009 21:36:58 +0300 Subject: libosd: osd_dev_info: Unique Identification of an OSD device Define an osd_dev_info structure that Uniquely identifies an OSD device lun on the network. The identification is built from unique target attributes and is the same for all network/SAN machines. osduld_info_lookup() - NEW New API that will lookup an osd_dev by its osd_dev_info. This is used by pNFS-objects for cross network global device identification. osduld_device_info() - NEW Given an osd_dev handle returns its associated osd_dev_info. This is used by exofs to encode the device information for network clients. (Get-device-info). The ULD fetches this information at startup and hangs it on each OSD device. (This is a fast operation that can be called at any condition) osd_auto_detect_ver() - REVISED Now returns an osd_dev_info structure. Is only called once by ULD as before. See added comments for how to use. Signed-off-by: Boaz Harrosh --- drivers/scsi/osd/osd_initiator.c | 22 ++++++++++++----- drivers/scsi/osd/osd_uld.c | 53 +++++++++++++++++++++++++++++++++++++++- include/scsi/osd_initiator.h | 37 +++++++++++++++++++++++++--- 3 files changed, 101 insertions(+), 11 deletions(-) diff --git a/drivers/scsi/osd/osd_initiator.c b/drivers/scsi/osd/osd_initiator.c index 7a117c18114c..26e1b414a118 100644 --- a/drivers/scsi/osd/osd_initiator.c +++ b/drivers/scsi/osd/osd_initiator.c @@ -73,7 +73,8 @@ static const char *_osd_ver_desc(struct osd_request *or) #define ATTR_DEF_RI(id, len) ATTR_DEF(OSD_APAGE_ROOT_INFORMATION, id, len) -static int _osd_print_system_info(struct osd_dev *od, void *caps) +static int _osd_get_print_system_info(struct osd_dev *od, + void *caps, struct osd_dev_info *odi) { struct osd_request *or; struct osd_attr get_attrs[] = { @@ -137,8 +138,13 @@ static int _osd_print_system_info(struct osd_dev *od, void *caps) OSD_INFO("PRODUCT_SERIAL_NUMBER [%s]\n", (char *)pFirst); - pFirst = get_attrs[a].val_ptr; - OSD_INFO("OSD_NAME [%s]\n", (char *)pFirst); + odi->osdname_len = get_attrs[a].len; + if (odi->osdname_len) { + odi->osdname = kzalloc(odi->osdname_len + 1, GFP_KERNEL); + memcpy(odi->osdname, get_attrs[a].val_ptr, odi->osdname_len); + } else + odi->osdname = NULL; + OSD_INFO("OSD_NAME [%s]\n", odi->osdname); a++; pFirst = get_attrs[a++].val_ptr; @@ -171,6 +177,9 @@ static int _osd_print_system_info(struct osd_dev *od, void *caps) sid_dump, sizeof(sid_dump), true); OSD_INFO("OSD_SYSTEM_ID(%d)\n" " [%s]\n", len, sid_dump); + + odi->systemid_len = len; + memcpy(odi->systemid, get_attrs[a].val_ptr, len); a++; } out: @@ -178,16 +187,17 @@ out: return ret; } -int osd_auto_detect_ver(struct osd_dev *od, void *caps) +int osd_auto_detect_ver(struct osd_dev *od, + void *caps, struct osd_dev_info *odi) { int ret; /* Auto-detect the osd version */ - ret = _osd_print_system_info(od, caps); + ret = _osd_get_print_system_info(od, caps, odi); if (ret) { osd_dev_set_ver(od, OSD_VER1); OSD_DEBUG("converting to OSD1\n"); - ret = _osd_print_system_info(od, caps); + ret = _osd_get_print_system_info(od, caps, odi); } return ret; diff --git a/drivers/scsi/osd/osd_uld.c b/drivers/scsi/osd/osd_uld.c index 0bdef3390902..8c069f961fb3 100644 --- a/drivers/scsi/osd/osd_uld.c +++ b/drivers/scsi/osd/osd_uld.c @@ -87,6 +87,7 @@ struct osd_uld_device { struct osd_dev od; struct gendisk *disk; struct device *class_member; + struct osd_dev_info odi; }; static void __uld_get(struct osd_uld_device *oud); @@ -216,6 +217,48 @@ free_od: } EXPORT_SYMBOL(osduld_path_lookup); +static inline bool _the_same_or_null(const u8 *a1, unsigned a1_len, + const u8 *a2, unsigned a2_len) +{ + if (!a2_len) /* User string is Empty means don't care */ + return true; + + if (a1_len != a2_len) + return false; + + return 0 == memcmp(a1, a2, a1_len); +} + +struct osd_dev *osduld_info_lookup(const struct osd_dev_info *odi) +{ + unsigned i; + + for (i = 0; i < SCSI_OSD_MAX_MINOR; i++) { + char dev_str[16]; + struct osd_uld_device *oud; + struct osd_dev *od; + + sprintf(dev_str, "/dev/osd%d", i); + od = osduld_path_lookup(dev_str); + if (IS_ERR(od)) + continue; + + oud = od->file->private_data; + if (_the_same_or_null(oud->odi.systemid, oud->odi.systemid_len, + odi->systemid, odi->systemid_len) && + _the_same_or_null(oud->odi.osdname, oud->odi.osdname_len, + odi->osdname, odi->osdname_len)) { + OSD_DEBUG("found device sysid_len=%d osdname=%d\n", + odi->systemid_len, odi->osdname_len); + return od; + } + osduld_put_device(od); + } + + return ERR_PTR(-ENODEV); +} +EXPORT_SYMBOL(osduld_info_lookup); + void osduld_put_device(struct osd_dev *od) { @@ -230,6 +273,13 @@ void osduld_put_device(struct osd_dev *od) } EXPORT_SYMBOL(osduld_put_device); +const struct osd_dev_info *osduld_device_info(struct osd_dev *od) +{ + struct osd_uld_device *oud = od->file->private_data; + return &oud->odi; +} +EXPORT_SYMBOL(osduld_device_info); + /* * Scsi Device operations */ @@ -250,7 +300,7 @@ static int __detect_osd(struct osd_uld_device *oud) OSD_ERR("warning: scsi_test_unit_ready failed\n"); osd_sec_init_nosec_doall_caps(caps, &osd_root_object, false, true); - if (osd_auto_detect_ver(&oud->od, caps)) + if (osd_auto_detect_ver(&oud->od, caps, &oud->odi)) return -ENODEV; return 0; @@ -402,6 +452,7 @@ static void __remove(struct kref *kref) put_disk(oud->disk); ida_remove(&osd_minor_ida, oud->minor); + kfree(oud->odi.osdname); kfree(oud); } diff --git a/include/scsi/osd_initiator.h b/include/scsi/osd_initiator.h index 02bd9f716357..0d29cc33028e 100644 --- a/include/scsi/osd_initiator.h +++ b/include/scsi/osd_initiator.h @@ -56,10 +56,23 @@ struct osd_dev { #endif }; -/* Retrieve/return osd_dev(s) for use by Kernel clients */ -struct osd_dev *osduld_path_lookup(const char *dev_name); /*Use IS_ERR/ERR_PTR*/ +/* Unique Identification of an OSD device */ +struct osd_dev_info { + unsigned systemid_len; + u8 systemid[OSD_SYSTEMID_LEN]; + unsigned osdname_len; + u8 *osdname; +}; + +/* Retrieve/return osd_dev(s) for use by Kernel clients + * Use IS_ERR/ERR_PTR on returned "osd_dev *". + */ +struct osd_dev *osduld_path_lookup(const char *dev_name); +struct osd_dev *osduld_info_lookup(const struct osd_dev_info *odi); void osduld_put_device(struct osd_dev *od); +const struct osd_dev_info *osduld_device_info(struct osd_dev *od); + /* Add/remove test ioctls from external modules */ typedef int (do_test_fn)(struct osd_dev *od, unsigned cmd, unsigned long arg); int osduld_register_test(unsigned ioctl, do_test_fn *do_test); @@ -69,8 +82,24 @@ void osduld_unregister_test(unsigned ioctl); void osd_dev_init(struct osd_dev *od, struct scsi_device *scsi_device); void osd_dev_fini(struct osd_dev *od); -/* some hi level device operations */ -int osd_auto_detect_ver(struct osd_dev *od, void *caps); /* GFP_KERNEL */ +/** + * osd_auto_detect_ver - Detect the OSD version, return Unique Identification + * + * @od: OSD target lun handle + * @caps: Capabilities authorizing OSD root read attributes access + * @odi: Retrieved information uniquely identifying the osd target lun + * Note: odi->osdname must be kfreed by caller. + * + * Auto detects the OSD version of the OSD target and sets the @od + * accordingly. Meanwhile also returns the "system id" and "osd name" root + * attributes which uniquely identify the OSD target. This member is usually + * called by the ULD. ULD users should call osduld_device_info(). + * This rutine allocates osd requests and memory at GFP_KERNEL level and might + * sleep. + */ +int osd_auto_detect_ver(struct osd_dev *od, + void *caps, struct osd_dev_info *odi); + static inline struct request_queue *osd_request_queue(struct osd_dev *od) { return od->scsi_device->request_queue; -- cgit v1.2.3 From f6c3a386118b576c5f092a1e67bc685687406fba Mon Sep 17 00:00:00 2001 From: Boaz Harrosh Date: Thu, 2 Jul 2009 21:37:33 +0300 Subject: libosd: osd_dev_is_ver1 - Minor API cleanup define a new osd_dev_is_ver1 that operates on devices and the old osd_req_is_ver1 uses that new API. Signed-off-by: Boaz Harrosh --- include/scsi/osd_initiator.h | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-) diff --git a/include/scsi/osd_initiator.h b/include/scsi/osd_initiator.h index 0d29cc33028e..7bda1091fb6e 100644 --- a/include/scsi/osd_initiator.h +++ b/include/scsi/osd_initiator.h @@ -113,6 +113,15 @@ static inline void osd_dev_set_ver(struct osd_dev *od, enum osd_std_version v) #endif } +static inline bool osd_dev_is_ver1(struct osd_dev *od) +{ +#ifdef OSD_VER1_SUPPORT + return od->version == OSD_VER1; +#else + return false; +#endif +} + struct osd_request; typedef void (osd_req_done_fn)(struct osd_request *or, void *private); @@ -149,14 +158,9 @@ struct osd_request { int async_error; }; -/* OSD Version control */ static inline bool osd_req_is_ver1(struct osd_request *or) { -#ifdef OSD_VER1_SUPPORT - return or->osd_dev->version == OSD_VER1; -#else - return false; -#endif + return osd_dev_is_ver1(or->osd_dev); } /* -- cgit v1.2.3 From bfe4fdcc729d3138e4883fe9767f30efa380b023 Mon Sep 17 00:00:00 2001 From: Boaz Harrosh Date: Mon, 24 Aug 2009 20:27:45 +0300 Subject: libosd: osd_sense: OSD_CFO_PERMISSIONS Add one more important cdb_field_offset that can be returned with scsi_invalid_field_in_cdb. It is the offset of the permissions_bit_mask field in the capabilities structure. Interestingly, the offset is the same for V1/V2 Signed-off-by: Boaz Harrosh --- include/scsi/osd_sense.h | 3 +++ 1 file changed, 3 insertions(+) diff --git a/include/scsi/osd_sense.h b/include/scsi/osd_sense.h index ff9b33c773c7..91db543a5502 100644 --- a/include/scsi/osd_sense.h +++ b/include/scsi/osd_sense.h @@ -255,6 +255,9 @@ enum osdv2_cdb_field_offset { OSD_CFO_STARTING_BYTE = OSD_CDB_OFFSET(v2.start_address), OSD_CFO_PARTITION_ID = OSD_CDB_OFFSET(partition), OSD_CFO_OBJECT_ID = OSD_CDB_OFFSET(object), + OSD_CFO_PERMISSIONS = sizeof(struct osd_cdb_head) + + offsetof(struct osd_capability_head, + permissions_bit_mask), }; #endif /* ndef __OSD_SENSE_H__ */ -- cgit v1.2.3 From ff5ca6efc35e050aee67e4e506b7ff0e23971565 Mon Sep 17 00:00:00 2001 From: Thadeu Lima de Souza Cascardo Date: Mon, 27 Jul 2009 13:26:32 -0300 Subject: trivial: some small fixes in exofs documentation Add exofs.txt to filesystems Documentation index and fix some typos, identation and grammar. Signed-off-by: Thadeu Lima de Souza Cascardo Signed-off-by: Boaz Harrosh --- Documentation/filesystems/00-INDEX | 2 ++ Documentation/filesystems/exofs.txt | 23 ++++++++++++----------- 2 files changed, 14 insertions(+), 11 deletions(-) diff --git a/Documentation/filesystems/00-INDEX b/Documentation/filesystems/00-INDEX index f15621ee5599..7001782ab932 100644 --- a/Documentation/filesystems/00-INDEX +++ b/Documentation/filesystems/00-INDEX @@ -36,6 +36,8 @@ dnotify.txt - info about directory notification in Linux. ecryptfs.txt - docs on eCryptfs: stacked cryptographic filesystem for Linux. +exofs.txt + - info, usage, mount options, design about EXOFS. ext2.txt - info, mount options and specifications for the Ext2 filesystem. ext3.txt diff --git a/Documentation/filesystems/exofs.txt b/Documentation/filesystems/exofs.txt index 0ced74c2f73c..abd2a9b5b787 100644 --- a/Documentation/filesystems/exofs.txt +++ b/Documentation/filesystems/exofs.txt @@ -60,13 +60,13 @@ USAGE mkfs.exofs --pid=65536 --format /dev/osd0 - The --format is optional if not specified no OSD_FORMAT will be - preformed and a clean file system will be created in the specified pid, + The --format is optional. If not specified, no OSD_FORMAT will be + performed and a clean file system will be created in the specified pid, in the available space of the target. (Use --format=size_in_meg to limit the total LUN space available) - If pid already exist it will be deleted and a new one will be created in it's - place. Be careful. + If pid already exists, it will be deleted and a new one will be created in + its place. Be careful. An exofs lives inside a single OSD partition. You can create multiple exofs filesystems on the same device using multiple pids. @@ -81,7 +81,7 @@ USAGE 7. For reference (See do-exofs example script): do-exofs start - an example of how to perform the above steps. - do-exofs stop - an example of how to unmount the file system. + do-exofs stop - an example of how to unmount the file system. do-exofs format - an example of how to format and mkfs a new exofs. 8. Extra compilation flags (uncomment in fs/exofs/Kbuild): @@ -104,8 +104,8 @@ Where: exofs specific options: Options are separated by commas (,) pid= - The partition number to mount/create as container of the filesystem. - This option is mandatory - to= - Timeout in ticks for a single command + This option is mandatory. + to= - Timeout in ticks for a single command. default is (60 * HZ) [for debugging only] =============================================================================== @@ -116,7 +116,7 @@ DESIGN with a special ID (defined in common.h). Information included in the file system control block is used to fill the in-memory superblock structure at mount time. This object is created before - the file system is used by mkexofs.c It contains information such as: + the file system is used by mkexofs.c. It contains information such as: - The file system's magic number - The next inode number to be allocated @@ -134,8 +134,8 @@ DESIGN attributes. This applies to both regular files and other types (directories, device files, symlinks, etc.). -* Credentials are generated per object (inode and superblock) when they is - created in memory (read off disk or created). The credential works for all +* Credentials are generated per object (inode and superblock) when they are + created in memory (read from disk or created). The credential works for all operations and is used as long as the object remains in memory. * Async OSD operations are used whenever possible, but the target may execute @@ -145,7 +145,8 @@ DESIGN from executing in reverse order: - The following are handled with the OBJ_CREATED and OBJ_2BCREATED flags. OBJ_CREATED is set when we know the object exists on the OSD - - in create's callback function, and when we successfully do a read_inode. + in create's callback function, and when we successfully do a + read_inode. OBJ_2BCREATED is set in the beginning of the create function, so we know that we should wait. - create/delete: delete should wait until the object is created -- cgit v1.2.3 From fd5cf19dbd1a7a19b1fc779d51d2ae3e364d34db Mon Sep 17 00:00:00 2001 From: Boaz Harrosh Date: Tue, 14 Jul 2009 11:06:08 +0300 Subject: exofs: More sane debug print debug prints should be somewhat useful without actually reading the source code Signed-off-by: Boaz Harrosh --- fs/exofs/inode.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/fs/exofs/inode.c b/fs/exofs/inode.c index 6c10f7476699..44748613be07 100644 --- a/fs/exofs/inode.c +++ b/fs/exofs/inode.c @@ -950,8 +950,7 @@ struct inode *exofs_iget(struct super_block *sb, unsigned long ino) #ifdef EXOFS_DEBUG_OBJ_ISIZE if ((inode->i_size != sanity) && (!exofs_inode_is_fast_symlink(inode))) { - EXOFS_ERR("WARNING: Size of object from inode and " - "attributes differ (%lld != %llu)\n", + EXOFS_ERR("WARNING: Size of inode=%llu != object=%llu\n", inode->i_size, _LLU(sanity)); } #endif -- cgit v1.2.3 From 21b4164655f3acbe4588fadd878470fc37e6cd50 Mon Sep 17 00:00:00 2001 From: Boaz Harrosh Date: Tue, 4 Aug 2009 20:40:29 +0300 Subject: exofs: refactor exofs_i_info initialization into common helper There are two places that initialize inodes: exofs_iget() and exofs_new_inode() As more members of exofs_i_info that need initialization are added this code will grow. (soon) Signed-off-by: Boaz Harrosh Signed-off-by: Benny Halevy --- fs/exofs/inode.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/fs/exofs/inode.c b/fs/exofs/inode.c index 44748613be07..cd034e62a54c 100644 --- a/fs/exofs/inode.c +++ b/fs/exofs/inode.c @@ -906,6 +906,12 @@ out: return ret; } + +static void __oi_init(struct exofs_i_info *oi) +{ + init_waitqueue_head(&oi->i_wq); + oi->i_flags = 0; +} /* * Fill in an inode read from the OSD and set it up for use */ @@ -923,13 +929,13 @@ struct inode *exofs_iget(struct super_block *sb, unsigned long ino) if (!(inode->i_state & I_NEW)) return inode; oi = exofs_i(inode); + __oi_init(oi); /* read the inode from the osd */ ret = exofs_get_inode(sb, oi, &fcb, &sanity); if (ret) goto bad_inode; - init_waitqueue_head(&oi->i_wq); set_obj_created(oi); /* copy stuff from on-disk struct to in-memory struct */ @@ -1060,8 +1066,8 @@ struct inode *exofs_new_inode(struct inode *dir, int mode) return ERR_PTR(-ENOMEM); oi = exofs_i(inode); + __oi_init(oi); - init_waitqueue_head(&oi->i_wq); set_obj_2bcreated(oi); sbi = sb->s_fs_info; -- cgit v1.2.3 From ba3690fca034aedc73ed4ed63bd93f0769e683ef Mon Sep 17 00:00:00 2001 From: Boaz Harrosh Date: Wed, 19 Aug 2009 17:31:25 +0300 Subject: exofs: remove BKL from super operations the two places inside exofs that where taking the BKL were: exofs_put_super() - .put_super and exofs_sync_fs() - which is .sync_fs and is also called from .write_super. Now exofs_sync_fs() is protected from itself by also taking the sb_lock. exofs_put_super() directly calls exofs_sync_fs() so there is no danger between these two either. In anyway there is absolutely nothing dangerous been done inside exofs_sync_fs(). Unless there is some subtle race with the actual lifetime of the super_block in regard to .put_super and some other parts of the VFS. Which is highly unlikely. Signed-off-by: Boaz Harrosh --- fs/exofs/super.c | 6 ------ 1 file changed, 6 deletions(-) diff --git a/fs/exofs/super.c b/fs/exofs/super.c index 5ab10c3bbebe..9f500dec3b59 100644 --- a/fs/exofs/super.c +++ b/fs/exofs/super.c @@ -214,7 +214,6 @@ int exofs_sync_fs(struct super_block *sb, int wait) } lock_super(sb); - lock_kernel(); sbi = sb->s_fs_info; fscb->s_nextid = cpu_to_le64(sbi->s_nextid); fscb->s_numfiles = cpu_to_le32(sbi->s_numfiles); @@ -245,7 +244,6 @@ int exofs_sync_fs(struct super_block *sb, int wait) out: if (or) osd_end_request(or); - unlock_kernel(); unlock_super(sb); kfree(fscb); return ret; @@ -268,8 +266,6 @@ static void exofs_put_super(struct super_block *sb) int num_pend; struct exofs_sb_info *sbi = sb->s_fs_info; - lock_kernel(); - if (sb->s_dirt) exofs_write_super(sb); @@ -286,8 +282,6 @@ static void exofs_put_super(struct super_block *sb) osduld_put_device(sbi->s_dev); kfree(sb->s_fs_info); sb->s_fs_info = NULL; - - unlock_kernel(); } /* -- cgit v1.2.3