/* -*- mode: c; c-basic-offset: 8; indent-tabs-mode: nil; -*- * vim:expandtab:shiftwidth=8:tabstop=8: * * linux/fs/obdfilter/filter.c * * Copyright (c) 2001-2003 Cluster File Systems, Inc. * Author: Peter Braam * Author: Andreas Dilger * * This file is part of Lustre, http://www.lustre.org. * * Lustre is free software; you can redistribute it and/or * modify it under the terms of version 2 of the GNU General Public * License as published by the Free Software Foundation. * * Lustre is distributed in the hope that it will be useful, * but WITHOUT ANY WARRANTY; without even the implied warranty of * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * GNU General Public License for more details. * * You should have received a copy of the GNU General Public License * along with Lustre; if not, write to the Free Software * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ /* * Invariant: Get O/R i_sem for lookup, if needed, before any journal ops * (which need to get journal_lock, may block if journal full). * * Invariant: Call filter_start_transno() before any journal ops to avoid the * same deadlock problem. We can (and want) to get rid of the * transno sem in favour of the dir/inode i_sem to avoid single * threaded operation on the OST. */ #define EXPORT_SYMTAB #define DEBUG_SUBSYSTEM S_FILTER #include #include #include // XXX kill me soon #include #include #include #include #include #include #include #include #include #include #if (LINUX_VERSION_CODE > KERNEL_VERSION(2,5,0)) #include #endif enum { LPROC_FILTER_READ_BYTES = 0, LPROC_FILTER_WRITE_BYTES = 1, LPROC_FILTER_LAST, }; #define S_SHIFT 12 static char *obd_type_by_mode[S_IFMT >> S_SHIFT] = { [0] NULL, [S_IFREG >> S_SHIFT] "R", [S_IFDIR >> S_SHIFT] "D", [S_IFCHR >> S_SHIFT] "C", [S_IFBLK >> S_SHIFT] "B", [S_IFIFO >> S_SHIFT] "F", [S_IFSOCK >> S_SHIFT] "S", [S_IFLNK >> S_SHIFT] "L" }; static inline const char *obd_mode_to_type(int mode) { return obd_type_by_mode[(mode & S_IFMT) >> S_SHIFT]; } static void filter_ffd_addref(void *ffdp) { struct filter_file_data *ffd = ffdp; atomic_inc(&ffd->ffd_refcount); CDEBUG(D_INFO, "GETting ffd %p : new refcount %d\n", ffd, atomic_read(&ffd->ffd_refcount)); } static struct filter_file_data *filter_ffd_new(void) { struct filter_file_data *ffd; OBD_ALLOC(ffd, sizeof *ffd); if (ffd == NULL) { CERROR("out of memory\n"); return NULL; } atomic_set(&ffd->ffd_refcount, 2); INIT_LIST_HEAD(&ffd->ffd_handle.h_link); class_handle_hash(&ffd->ffd_handle, filter_ffd_addref); return ffd; } static struct filter_file_data *filter_handle2ffd(struct lustre_handle *handle) { struct filter_file_data *ffd = NULL; ENTRY; LASSERT(handle != NULL); ffd = class_handle2object(handle->cookie); if (ffd != NULL) LASSERT(ffd->ffd_file->private_data == ffd); RETURN(ffd); } static void filter_ffd_put(struct filter_file_data *ffd) { CDEBUG(D_INFO, "PUTting ffd %p : new refcount %d\n", ffd, atomic_read(&ffd->ffd_refcount) - 1); LASSERT(atomic_read(&ffd->ffd_refcount) > 0 && atomic_read(&ffd->ffd_refcount) < 0x5a5a); if (atomic_dec_and_test(&ffd->ffd_refcount)) { LASSERT(list_empty(&ffd->ffd_handle.h_link)); OBD_FREE(ffd, sizeof *ffd); } } static void filter_ffd_destroy(struct filter_file_data *ffd) { class_handle_unhash(&ffd->ffd_handle); filter_ffd_put(ffd); } static void filter_commit_cb(struct obd_device *obd, __u64 transno, int error) { obd_transno_commit_cb(obd, transno, error); } /* Assumes caller has already pushed us into the kernel context. */ int filter_finish_transno(struct obd_export *export, void *handle, struct obd_trans_info *oti, int rc) { __u64 last_rcvd; struct obd_device *obd = export->exp_obd; struct filter_obd *filter = &obd->u.filter; struct filter_export_data *fed = &export->exp_filter_data; struct filter_client_data *fcd = fed->fed_fcd; loff_t off; ssize_t written; /* Propagate error code. */ if (rc) RETURN(rc); if (!obd->obd_replayable) RETURN(rc); /* we don't allocate new transnos for replayed requests */ if (oti && oti->oti_transno == 0) { spin_lock(&filter->fo_translock); last_rcvd = le64_to_cpu(filter->fo_fsd->fsd_last_rcvd) + 1; filter->fo_fsd->fsd_last_rcvd = cpu_to_le64(last_rcvd); spin_unlock(&filter->fo_translock); oti->oti_transno = last_rcvd; fcd->fcd_last_rcvd = cpu_to_le64(last_rcvd); fcd->fcd_mount_count = filter->fo_fsd->fsd_mount_count; /* could get xid from oti, if it's ever needed */ fcd->fcd_last_xid = 0; off = fed->fed_lr_off; fsfilt_set_last_rcvd(obd, last_rcvd, handle, filter_commit_cb); written = lustre_fwrite(filter->fo_rcvd_filp, (char *)fcd, sizeof(*fcd), &off); CDEBUG(D_HA, "wrote trans #"LPD64" for client %s at #%d: " "written = "LPSZ"\n", last_rcvd, fcd->fcd_uuid, fed->fed_lr_idx, written); if (written == sizeof(*fcd)) RETURN(0); CERROR("error writing to last_rcvd file: rc = %d\n", (int)written); if (written >= 0) RETURN(-EIO); RETURN(written); } RETURN(0); } static inline void f_dput(struct dentry *dentry) { /* Can't go inside filter_ddelete because it can block */ CDEBUG(D_INODE, "putting %s: %p, count = %d\n", dentry->d_name.name, dentry, atomic_read(&dentry->d_count) - 1); LASSERT(atomic_read(&dentry->d_count) > 0); dput(dentry); } /* Not racy w.r.t. others, because we are the only user of this dentry */ static void filter_drelease(struct dentry *dentry) { if (dentry->d_fsdata) OBD_FREE(dentry->d_fsdata, sizeof(struct filter_dentry_data)); } struct dentry_operations filter_dops = { .d_release = filter_drelease, }; #define LAST_RCVD "last_rcvd" #define INIT_OBJID 2 /* This limit is arbitrary, but for now we fit it in 1 page (32k clients) */ #define FILTER_LR_MAX_CLIENTS (PAGE_SIZE * 8) #define FILTER_LR_MAX_CLIENT_WORDS (FILTER_LR_MAX_CLIENTS/sizeof(unsigned long)) /* Add client data to the FILTER. We use a bitmap to locate a free space * in the last_rcvd file if cl_idx is -1 (i.e. a new client). * Otherwise, we have just read the data from the last_rcvd file and * we know its offset. */ int filter_client_add(struct obd_device *obd, struct filter_obd *filter, struct filter_export_data *fed, int cl_idx) { unsigned long *bitmap = filter->fo_last_rcvd_slots; int new_client = (cl_idx == -1); LASSERT(bitmap != NULL); /* XXX if fcd_uuid were a real obd_uuid, I could use obd_uuid_equals */ if (!strcmp(fed->fed_fcd->fcd_uuid, "OBD_CLASS_UUID")) RETURN(0); /* the bitmap operations can handle cl_idx > sizeof(long) * 8, so * there's no need for extra complication here */ if (new_client) { cl_idx = find_first_zero_bit(bitmap, FILTER_LR_MAX_CLIENTS); repeat: if (cl_idx >= FILTER_LR_MAX_CLIENTS) { CERROR("no client slots - fix FILTER_LR_MAX_CLIENTS\n"); return -ENOMEM; } if (test_and_set_bit(cl_idx, bitmap)) { CERROR("FILTER client %d: found bit is set in bitmap\n", cl_idx); cl_idx = find_next_zero_bit(bitmap, FILTER_LR_MAX_CLIENTS, cl_idx); goto repeat; } } else { if (test_and_set_bit(cl_idx, bitmap)) { CERROR("FILTER client %d: bit already set in bitmap!\n", cl_idx); LBUG(); } } fed->fed_lr_idx = cl_idx; fed->fed_lr_off = le32_to_cpu(filter->fo_fsd->fsd_client_start) + cl_idx * le16_to_cpu(filter->fo_fsd->fsd_client_size); CDEBUG(D_INFO, "client at index %d (%llu) with UUID '%s' added\n", fed->fed_lr_idx, fed->fed_lr_off, fed->fed_fcd->fcd_uuid); if (new_client) { struct obd_run_ctxt saved; loff_t off = fed->fed_lr_off; ssize_t written; void *handle; CDEBUG(D_INFO, "writing client fcd at idx %u (%llu) (len %u)\n", fed->fed_lr_idx,off,(unsigned int)sizeof(*fed->fed_fcd)); push_ctxt(&saved, &filter->fo_ctxt, NULL); /* Transaction eeded to fix for bug 1403 */ handle = fsfilt_start(obd, filter->fo_rcvd_filp->f_dentry->d_inode, FSFILT_OP_SETATTR); if (IS_ERR(handle)) { written = PTR_ERR(handle); CERROR("unable to start transaction: rc %d\n", (int)written); } else { written = lustre_fwrite(filter->fo_rcvd_filp, (char *)fed->fed_fcd, sizeof(*fed->fed_fcd), &off); fsfilt_commit(obd, filter->fo_rcvd_filp->f_dentry->d_inode, handle, 0); } pop_ctxt(&saved, &filter->fo_ctxt, NULL); if (written != sizeof(*fed->fed_fcd)) { if (written < 0) RETURN(written); RETURN(-EIO); } } return 0; } int filter_client_free(struct obd_export *exp, int failover) { struct filter_export_data *fed = &exp->exp_filter_data; struct filter_obd *filter = &exp->exp_obd->u.filter; struct filter_client_data zero_fcd; struct obd_run_ctxt saved; int written; loff_t off; ENTRY; if (!fed->fed_fcd) RETURN(0); if (failover != 0) GOTO(free, 0); /* XXX if fcd_uuid were a real obd_uuid, I could use obd_uuid_equals */ if (!strcmp(fed->fed_fcd->fcd_uuid, "OBD_CLASS_UUID")) GOTO(free, 0); LASSERT(filter->fo_last_rcvd_slots != NULL); off = fed->fed_lr_off; CDEBUG(D_INFO, "freeing client at idx %u (%lld) with UUID '%s'\n", fed->fed_lr_idx, fed->fed_lr_off, fed->fed_fcd->fcd_uuid); if (!test_and_clear_bit(fed->fed_lr_idx, filter->fo_last_rcvd_slots)) { CERROR("FILTER client %u: bit already clear in bitmap!!\n", fed->fed_lr_idx); LBUG(); } memset(&zero_fcd, 0, sizeof zero_fcd); push_ctxt(&saved, &filter->fo_ctxt, NULL); written = lustre_fwrite(filter->fo_rcvd_filp, (const char *)&zero_fcd, sizeof(zero_fcd), &off); /* XXX: this write gets lost sometimes, unless this sync is here. */ if (written > 0) file_fsync(filter->fo_rcvd_filp, filter->fo_rcvd_filp->f_dentry, 1); pop_ctxt(&saved, &filter->fo_ctxt, NULL); if (written != sizeof(zero_fcd)) { CERROR("error zeroing out client %s idx %u (%llu) in %s: %d\n", fed->fed_fcd->fcd_uuid, fed->fed_lr_idx, fed->fed_lr_off, LAST_RCVD, written); } else { CDEBUG(D_INFO, "zeroed disconnecting client %s at idx %u (%llu)\n", fed->fed_fcd->fcd_uuid, fed->fed_lr_idx,fed->fed_lr_off); } free: OBD_FREE(fed->fed_fcd, sizeof(*fed->fed_fcd)); RETURN(0); } static int filter_free_server_data(struct filter_obd *filter) { OBD_FREE(filter->fo_fsd, sizeof(*filter->fo_fsd)); filter->fo_fsd = NULL; OBD_FREE(filter->fo_last_rcvd_slots, FILTER_LR_MAX_CLIENT_WORDS * sizeof(unsigned long)); filter->fo_last_rcvd_slots = NULL; return 0; } /* assumes caller is already in kernel ctxt */ static int filter_update_server_data(struct file *filp, struct filter_server_data *fsd) { loff_t off = 0; int rc; CDEBUG(D_INODE, "server uuid : %s\n", fsd->fsd_uuid); CDEBUG(D_INODE, "server last_objid: "LPU64"\n", le64_to_cpu(fsd->fsd_last_objid)); CDEBUG(D_INODE, "server last_rcvd : "LPU64"\n", le64_to_cpu(fsd->fsd_last_rcvd)); CDEBUG(D_INODE, "server last_mount: "LPU64"\n", le64_to_cpu(fsd->fsd_mount_count)); rc = lustre_fwrite(filp, (char *)fsd, sizeof(*fsd), &off); if (rc != sizeof(*fsd)) { CDEBUG(D_INODE, "error writing filter_server_data: rc = %d\n", rc); RETURN(-EIO); } RETURN(0); } /* assumes caller has already in kernel ctxt */ static int filter_init_server_data(struct obd_device *obd, struct file * filp, __u64 init_lastobjid) { struct filter_obd *filter = &obd->u.filter; struct filter_server_data *fsd; struct filter_client_data *fcd = NULL; struct inode *inode = filp->f_dentry->d_inode; unsigned long last_rcvd_size = inode->i_size; __u64 mount_count = 0; int cl_idx; loff_t off = 0; int rc; /* ensure padding in the struct is the correct size */ LASSERT (offsetof(struct filter_server_data, fsd_padding) + sizeof(fsd->fsd_padding) == FILTER_LR_SERVER_SIZE); LASSERT (offsetof(struct filter_client_data, fcd_padding) + sizeof(fcd->fcd_padding) == FILTER_LR_CLIENT_SIZE); OBD_ALLOC(fsd, sizeof(*fsd)); if (!fsd) RETURN(-ENOMEM); filter->fo_fsd = fsd; OBD_ALLOC(filter->fo_last_rcvd_slots, FILTER_LR_MAX_CLIENT_WORDS * sizeof(unsigned long)); if (filter->fo_last_rcvd_slots == NULL) { OBD_FREE(fsd, sizeof(*fsd)); RETURN(-ENOMEM); } if (last_rcvd_size == 0) { CERROR("%s: initializing new last_rcvd\n", obd->obd_name); memcpy(fsd->fsd_uuid, obd->obd_uuid.uuid,sizeof(fsd->fsd_uuid)); fsd->fsd_last_objid = cpu_to_le64(init_lastobjid); fsd->fsd_last_rcvd = 0; mount_count = fsd->fsd_mount_count = 0; fsd->fsd_server_size = cpu_to_le32(FILTER_LR_SERVER_SIZE); fsd->fsd_client_start = cpu_to_le32(FILTER_LR_CLIENT_START); fsd->fsd_client_size = cpu_to_le16(FILTER_LR_CLIENT_SIZE); fsd->fsd_subdir_count = cpu_to_le16(FILTER_SUBDIR_COUNT); filter->fo_subdir_count = FILTER_SUBDIR_COUNT; } else { ssize_t retval = lustre_fread(filp, (char *)fsd, sizeof(*fsd), &off); if (retval != sizeof(*fsd)) { CDEBUG(D_INODE,"OBD filter: error reading %s\n", LAST_RCVD); GOTO(err_fsd, rc = -EIO); } mount_count = le64_to_cpu(fsd->fsd_mount_count); filter->fo_subdir_count = le16_to_cpu(fsd->fsd_subdir_count); } if (fsd->fsd_feature_incompat) { CERROR("unsupported feature %x\n", le32_to_cpu(fsd->fsd_feature_incompat)); GOTO(err_fsd, rc = -EINVAL); } if (fsd->fsd_feature_rocompat) { CERROR("read-only feature %x\n", le32_to_cpu(fsd->fsd_feature_rocompat)); /* Do something like remount filesystem read-only */ GOTO(err_fsd, rc = -EINVAL); } CDEBUG(D_INODE, "%s: server last_objid: "LPU64"\n", obd->obd_name, le64_to_cpu(fsd->fsd_last_objid)); CDEBUG(D_INODE, "%s: server last_rcvd : "LPU64"\n", obd->obd_name, le64_to_cpu(fsd->fsd_last_rcvd)); CDEBUG(D_INODE, "%s: server last_mount: "LPU64"\n", obd->obd_name, mount_count); CDEBUG(D_INODE, "%s: server data size: %u\n", obd->obd_name, le32_to_cpu(fsd->fsd_server_size)); CDEBUG(D_INODE, "%s: per-client data start: %u\n", obd->obd_name, le32_to_cpu(fsd->fsd_client_start)); CDEBUG(D_INODE, "%s: per-client data size: %u\n", obd->obd_name, le32_to_cpu(fsd->fsd_client_size)); CDEBUG(D_INODE, "%s: server subdir_count: %u\n", obd->obd_name, le16_to_cpu(fsd->fsd_subdir_count)); /* * When we do a clean FILTER shutdown, we save the last_rcvd into * the header. If we find clients with higher last_rcvd values * then those clients may need recovery done. */ if (!obd->obd_replayable) { CERROR("%s: recovery support OFF\n", obd->obd_name); GOTO(out, rc = 0); } for (cl_idx = 0; off < last_rcvd_size; cl_idx++) { __u64 last_rcvd; int mount_age; if (!fcd) { OBD_ALLOC(fcd, sizeof(*fcd)); if (!fcd) GOTO(err_fsd, rc = -ENOMEM); } /* Don't assume off is incremented properly, in case * sizeof(fsd) isn't the same as fsd->fsd_client_size. */ off = le32_to_cpu(fsd->fsd_client_start) + cl_idx * le16_to_cpu(fsd->fsd_client_size); rc = lustre_fread(filp, (char *)fcd, sizeof(*fcd), &off); if (rc != sizeof(*fcd)) { CERROR("error reading FILTER %s offset %d: rc = %d\n", LAST_RCVD, cl_idx, rc); if (rc > 0) /* XXX fatal error or just abort reading? */ rc = -EIO; break; } if (fcd->fcd_uuid[0] == '\0') { CDEBUG(D_INFO, "skipping zeroed client at offset %d\n", cl_idx); continue; } last_rcvd = le64_to_cpu(fcd->fcd_last_rcvd); /* These exports are cleaned up by filter_disconnect(), so they * need to be set up like real exports as filter_connect() does. */ mount_age = mount_count - le64_to_cpu(fcd->fcd_mount_count); if (mount_age < FILTER_MOUNT_RECOV) { struct obd_export *exp = class_new_export(obd); struct filter_export_data *fed; CERROR("RCVRNG CLIENT uuid: %s idx: %d lr: "LPU64 " srv lr: "LPU64" mnt: "LPU64" last mount: " LPU64"\n", fcd->fcd_uuid, cl_idx, last_rcvd, le64_to_cpu(fsd->fsd_last_rcvd), le64_to_cpu(fcd->fcd_mount_count), mount_count); if (exp == NULL) { /* XXX this rc is ignored */ rc = -ENOMEM; break; } memcpy(&exp->exp_client_uuid.uuid, fcd->fcd_uuid, sizeof exp->exp_client_uuid.uuid); fed = &exp->exp_filter_data; fed->fed_fcd = fcd; filter_client_add(obd, filter, fed, cl_idx); /* create helper if export init gets more complex */ INIT_LIST_HEAD(&fed->fed_open_head); spin_lock_init(&fed->fed_lock); fcd = NULL; obd->obd_recoverable_clients++; class_export_put(exp); } else { CDEBUG(D_INFO, "discarded client %d UUID '%s' count "LPU64"\n", cl_idx, fcd->fcd_uuid, le64_to_cpu(fcd->fcd_mount_count)); } CDEBUG(D_OTHER, "client at idx %d has last_rcvd = "LPU64"\n", cl_idx, last_rcvd); if (last_rcvd > le64_to_cpu(filter->fo_fsd->fsd_last_rcvd)) filter->fo_fsd->fsd_last_rcvd = cpu_to_le64(last_rcvd); obd->obd_last_committed = le64_to_cpu(filter->fo_fsd->fsd_last_rcvd); if (obd->obd_recoverable_clients) { CERROR("RECOVERY: %d recoverable clients, last_rcvd " LPU64"\n", obd->obd_recoverable_clients, le64_to_cpu(filter->fo_fsd->fsd_last_rcvd)); obd->obd_next_recovery_transno = obd->obd_last_committed + 1; obd->obd_recovering = 1; } } if (fcd) OBD_FREE(fcd, sizeof(*fcd)); out: fsd->fsd_mount_count = cpu_to_le64(mount_count + 1); /* save it,so mount count and last_recvd is current */ rc = filter_update_server_data(filp, filter->fo_fsd); RETURN(rc); err_fsd: filter_free_server_data(filter); RETURN(rc); } /* setup the object store with correct subdirectories */ static int filter_prep(struct obd_device *obd) { struct obd_run_ctxt saved; struct filter_obd *filter = &obd->u.filter; struct dentry *dentry, *O_dentry; struct file *file; struct inode *inode; int i; int rc = 0; int mode = 0; push_ctxt(&saved, &filter->fo_ctxt, NULL); dentry = simple_mkdir(current->fs->pwd, "O", 0700); CDEBUG(D_INODE, "got/created O: %p\n", dentry); if (IS_ERR(dentry)) { rc = PTR_ERR(dentry); CERROR("cannot open/create O: rc = %d\n", rc); GOTO(out, rc); } filter->fo_dentry_O = dentry; /* * Create directories and/or get dentries for each object type. * This saves us from having to do multiple lookups for each one. */ O_dentry = filter->fo_dentry_O; for (mode = 0; mode < (S_IFMT >> S_SHIFT); mode++) { char *name = obd_type_by_mode[mode]; if (!name) { filter->fo_dentry_O_mode[mode] = NULL; continue; } dentry = simple_mkdir(O_dentry, name, 0700); CDEBUG(D_INODE, "got/created O/%s: %p\n", name, dentry); if (IS_ERR(dentry)) { rc = PTR_ERR(dentry); CERROR("cannot create O/%s: rc = %d\n", name, rc); GOTO(err_O_mode, rc); } filter->fo_dentry_O_mode[mode] = dentry; } file = filp_open(LAST_RCVD, O_RDWR | O_CREAT, 0700); if (!file || IS_ERR(file)) { rc = PTR_ERR(file); CERROR("OBD filter: cannot open/create %s: rc = %d\n", LAST_RCVD, rc); GOTO(err_O_mode, rc); } if (!S_ISREG(file->f_dentry->d_inode->i_mode)) { CERROR("%s is not a regular file!: mode = %o\n", LAST_RCVD, file->f_dentry->d_inode->i_mode); GOTO(err_filp, rc = -ENOENT); } rc = fsfilt_journal_data(obd, file); if (rc) { CERROR("cannot journal data on %s: rc = %d\n", LAST_RCVD, rc); GOTO(err_filp, rc); } /* steal operations */ inode = file->f_dentry->d_inode; filter->fo_fop = file->f_op; filter->fo_iop = inode->i_op; filter->fo_aops = inode->i_mapping->a_ops; rc = filter_init_server_data(obd, file, INIT_OBJID); if (rc) { CERROR("cannot read %s: rc = %d\n", LAST_RCVD, rc); GOTO(err_client, rc); } filter->fo_rcvd_filp = file; if (filter->fo_subdir_count) { O_dentry = filter->fo_dentry_O_mode[S_IFREG >> S_SHIFT]; OBD_ALLOC(filter->fo_dentry_O_sub, filter->fo_subdir_count * sizeof(dentry)); if (!filter->fo_dentry_O_sub) GOTO(err_client, rc = -ENOMEM); for (i = 0; i < filter->fo_subdir_count; i++) { char dir[20]; snprintf(dir, sizeof(dir), "d%u", i); dentry = simple_mkdir(O_dentry, dir, 0700); CDEBUG(D_INODE, "got/created O/R/%s: %p\n", dir,dentry); if (IS_ERR(dentry)) { rc = PTR_ERR(dentry); CERROR("can't create O/R/%s: rc = %d\n",dir,rc); GOTO(err_O_sub, rc); } filter->fo_dentry_O_sub[i] = dentry; } } rc = 0; out: pop_ctxt(&saved, &filter->fo_ctxt, NULL); return(rc); err_O_sub: while (i-- > 0) { struct dentry *dentry = filter->fo_dentry_O_sub[i]; if (dentry) { f_dput(dentry); filter->fo_dentry_O_sub[i] = NULL; } } OBD_FREE(filter->fo_dentry_O_sub, filter->fo_subdir_count * sizeof(dentry)); err_client: class_disconnect_exports(obd, 0); err_filp: if (filp_close(file, 0)) CERROR("can't close %s after error\n", LAST_RCVD); filter->fo_rcvd_filp = NULL; err_O_mode: while (mode-- > 0) { struct dentry *dentry = filter->fo_dentry_O_mode[mode]; if (dentry) { f_dput(dentry); filter->fo_dentry_O_mode[mode] = NULL; } } f_dput(filter->fo_dentry_O); filter->fo_dentry_O = NULL; goto out; } /* cleanup the filter: write last used object id to status file */ static void filter_post(struct obd_device *obd) { struct obd_run_ctxt saved; struct filter_obd *filter = &obd->u.filter; long rc; int mode; /* XXX: filter_update_lastobjid used to call fsync_dev. It might be * best to start a transaction with h_sync, because we removed this * from lastobjid */ push_ctxt(&saved, &filter->fo_ctxt, NULL); rc = filter_update_server_data(filter->fo_rcvd_filp, filter->fo_fsd); if (rc) CERROR("OBD filter: error writing lastobjid: rc = %ld\n", rc); if (filter->fo_rcvd_filp) { rc = file_fsync(filter->fo_rcvd_filp, filter->fo_rcvd_filp->f_dentry, 1); filp_close(filter->fo_rcvd_filp, 0); filter->fo_rcvd_filp = NULL; if (rc) CERROR("last_rcvd file won't closed rc = %ld\n", rc); } if (filter->fo_subdir_count) { int i; for (i = 0; i < filter->fo_subdir_count; i++) { struct dentry *dentry = filter->fo_dentry_O_sub[i]; f_dput(dentry); filter->fo_dentry_O_sub[i] = NULL; } OBD_FREE(filter->fo_dentry_O_sub, filter->fo_subdir_count * sizeof(*filter->fo_dentry_O_sub)); } for (mode = 0; mode < (S_IFMT >> S_SHIFT); mode++) { struct dentry *dentry = filter->fo_dentry_O_mode[mode]; if (dentry) { f_dput(dentry); filter->fo_dentry_O_mode[mode] = NULL; } } f_dput(filter->fo_dentry_O); filter_free_server_data(filter); pop_ctxt(&saved, &filter->fo_ctxt, NULL); } static __u64 filter_next_id(struct filter_obd *filter) { obd_id id; LASSERT(filter->fo_fsd != NULL); spin_lock(&filter->fo_objidlock); id = le64_to_cpu(filter->fo_fsd->fsd_last_objid); filter->fo_fsd->fsd_last_objid = cpu_to_le64(id + 1); spin_unlock(&filter->fo_objidlock); return id; } /* direct cut-n-paste of mds_blocking_ast() */ int filter_blocking_ast(struct ldlm_lock *lock, struct ldlm_lock_desc *desc, void *data, int flag) { int do_ast; ENTRY; if (flag == LDLM_CB_CANCELING) { /* Don't need to do anything here. */ RETURN(0); } /* XXX layering violation! -phil */ l_lock(&lock->l_resource->lr_namespace->ns_lock); /* Get this: if filter_blocking_ast is racing with ldlm_intent_policy, * such that mds_blocking_ast is called just before l_i_p takes the * ns_lock, then by the time we get the lock, we might not be the * correct blocking function anymore. So check, and return early, if * so. */ if (lock->l_blocking_ast != filter_blocking_ast) { l_unlock(&lock->l_resource->lr_namespace->ns_lock); RETURN(0); } lock->l_flags |= LDLM_FL_CBPENDING; do_ast = (!lock->l_readers && !lock->l_writers); l_unlock(&lock->l_resource->lr_namespace->ns_lock); if (do_ast) { struct lustre_handle lockh; int rc; LDLM_DEBUG(lock, "already unused, calling ldlm_cli_cancel"); ldlm_lock2handle(lock, &lockh); rc = ldlm_cli_cancel(&lockh); if (rc < 0) CERROR("ldlm_cli_cancel: %d\n", rc); } else { LDLM_DEBUG(lock, "Lock still has references, will be " "cancelled later"); } RETURN(0); } static int filter_lock_dentry(struct obd_device *obd, struct dentry *de, ldlm_mode_t lock_mode,struct lustre_handle *lockh) { struct ldlm_res_id res_id = { .name = {0} }; int flags = 0, rc; ENTRY; res_id.name[0] = de->d_inode->i_ino; res_id.name[1] = de->d_inode->i_generation; rc = ldlm_cli_enqueue(NULL, NULL, obd->obd_namespace, NULL, res_id, LDLM_PLAIN, NULL, 0, lock_mode, &flags, ldlm_completion_ast, filter_blocking_ast, NULL, lockh); RETURN(rc == ELDLM_OK ? 0 : -ENOLCK); /* XXX translate ldlm code */ } static void filter_parent_unlock(struct dentry *dparent, struct lustre_handle *lockh, ldlm_mode_t lock_mode) { ldlm_lock_decref(lockh, lock_mode); } /* We never dget the object parent, so DON'T dput it either */ static inline struct dentry *filter_parent(struct obd_device *obd, obd_mode mode, obd_id objid) { struct filter_obd *filter = &obd->u.filter; LASSERT(S_ISREG(mode)); /* only regular files for now */ if (!S_ISREG(mode) || filter->fo_subdir_count == 0) return filter->fo_dentry_O_mode[(mode & S_IFMT) >> S_SHIFT]; return filter->fo_dentry_O_sub[objid & (filter->fo_subdir_count - 1)]; } /* We never dget the object parent, so DON'T dput it either */ static inline struct dentry *filter_parent_lock(struct obd_device *obd, obd_mode mode, obd_id objid, ldlm_mode_t lock_mode, struct lustre_handle *lockh) { unsigned long now = jiffies; struct dentry *de = filter_parent(obd, mode, objid); int rc; if (IS_ERR(de)) return de; rc = filter_lock_dentry(obd, de, lock_mode, lockh); if (time_after(jiffies, now + 15*HZ)) CERROR("slow parent lock %lus\n", (jiffies - now) / HZ); return rc ? ERR_PTR(rc) : de; } /* How to get files, dentries, inodes from object id's. * * If dir_dentry is passed, the caller has already locked the parent * appropriately for this operation (normally a write lock). If * dir_dentry is NULL, we do a read lock while we do the lookup to * avoid races with create/destroy and such changing the directory * internal to the filesystem code. */ static struct dentry *filter_fid2dentry(struct obd_device *obd, struct dentry *dir_dentry, obd_mode mode, obd_id id) { struct super_block *sb = obd->u.filter.fo_sb; struct lustre_handle lockh; struct dentry *dparent = dir_dentry; struct dentry *dchild; char name[32]; int len; ENTRY; if (!sb || !sb->s_dev) { CERROR("device not initialized.\n"); RETURN(ERR_PTR(-ENXIO)); } if (id == 0) { CERROR("fatal: invalid object id 0\n"); LBUG(); RETURN(ERR_PTR(-ESTALE)); } len = sprintf(name, LPU64, id); if (!dir_dentry) { dparent = filter_parent_lock(obd, mode, id, LCK_PR, &lockh); if (IS_ERR(dparent)) RETURN(dparent); } CDEBUG(D_INODE, "looking up object O/%*s/%s\n", dparent->d_name.len, dparent->d_name.name, name); dchild = ll_lookup_one_len(name, dparent, len); if (!dir_dentry) filter_parent_unlock(dparent, &lockh, LCK_PR); if (IS_ERR(dchild)) { CERROR("child lookup error %ld\n", PTR_ERR(dchild)); RETURN(dchild); } CDEBUG(D_INODE, "got child objid %s: %p, count = %d\n", name, dchild, atomic_read(&dchild->d_count)); LASSERT(atomic_read(&dchild->d_count) > 0); RETURN(dchild); } static struct file *filter_obj_open(struct obd_export *export, __u64 id, __u32 type, ldlm_mode_t parent_mode, struct lustre_handle *parent_lockh) { struct obd_device *obd = export->exp_obd; struct filter_obd *filter = &obd->u.filter; struct super_block *sb = filter->fo_sb; struct dentry *dchild = NULL, *dparent = NULL; struct filter_export_data *fed = &export->exp_filter_data; struct filter_dentry_data *fdd = NULL; struct filter_file_data *ffd = NULL; struct obd_run_ctxt saved; char name[24]; struct file *file; int len, cleanup_phase = 0; ENTRY; push_ctxt(&saved, &filter->fo_ctxt, NULL); if (!sb || !sb->s_dev) { CERROR("fatal: device not initialized.\n"); GOTO(cleanup, file = ERR_PTR(-ENXIO)); } if (!id) { CERROR("fatal: invalid obdo "LPU64"\n", id); GOTO(cleanup, file = ERR_PTR(-ESTALE)); } if (!(type & S_IFMT)) { CERROR("OBD %s, object "LPU64" has bad type: %o\n", __FUNCTION__, id, type); GOTO(cleanup, file = ERR_PTR(-EINVAL)); } ffd = filter_ffd_new(); if (ffd == NULL) { CERROR("obdfilter: out of memory\n"); GOTO(cleanup, file = ERR_PTR(-ENOMEM)); } cleanup_phase = 1; /* We preallocate this to avoid blocking while holding fo_fddlock */ OBD_ALLOC(fdd, sizeof *fdd); if (fdd == NULL) { CERROR("obdfilter: out of memory\n"); GOTO(cleanup, file = ERR_PTR(-ENOMEM)); } cleanup_phase = 2; dparent = filter_parent_lock(obd, type, id, parent_mode, parent_lockh); if (IS_ERR(dparent)) GOTO(cleanup, file = (void *)dparent); cleanup_phase = 3; len = snprintf(name, sizeof(name), LPU64, id); dchild = ll_lookup_one_len(name, dparent, len); if (IS_ERR(dchild)) GOTO(cleanup, file = (void *)dchild); cleanup_phase = 4; if (dchild->d_inode == NULL) { CERROR("opening non-existent object %s - O_CREAT?\n", name); file = ERR_PTR(-ENOENT); GOTO(cleanup, file); } /* dentry_open does a dput(dchild) and mntput(mnt) on error */ mntget(filter->fo_vfsmnt); file = dentry_open(dchild, filter->fo_vfsmnt, O_RDWR | O_LARGEFILE); if (IS_ERR(file)) { dchild = NULL; /* prevent a double dput in step 4 */ CERROR("error opening %s: rc %ld\n", name, PTR_ERR(file)); GOTO(cleanup, file); } spin_lock(&filter->fo_fddlock); if (dchild->d_fsdata) { spin_unlock(&filter->fo_fddlock); OBD_FREE(fdd, sizeof *fdd); fdd = dchild->d_fsdata; LASSERT(fdd->fdd_magic == FILTER_DENTRY_MAGIC); /* should only happen during client recovery */ if (fdd->fdd_flags & FILTER_FLAG_DESTROY) CDEBUG(D_INODE,"opening destroyed object "LPU64"\n",id); atomic_inc(&fdd->fdd_open_count); } else { atomic_set(&fdd->fdd_open_count, 1); fdd->fdd_magic = FILTER_DENTRY_MAGIC; fdd->fdd_flags = 0; fdd->fdd_objid = id; /* If this is racy, then we can use {cmp}xchg and atomic_add */ dchild->d_fsdata = fdd; spin_unlock(&filter->fo_fddlock); } ffd->ffd_file = file; LASSERT(file->private_data == NULL); file->private_data = ffd; if (!dchild->d_op) dchild->d_op = &filter_dops; else LASSERT(dchild->d_op == &filter_dops); spin_lock(&fed->fed_lock); list_add(&ffd->ffd_export_list, &fed->fed_open_head); spin_unlock(&fed->fed_lock); CDEBUG(D_INODE, "opened objid "LPU64": rc = %p\n", id, file); cleanup: switch (cleanup_phase) { case 4: if (IS_ERR(file)) f_dput(dchild); case 3: if (IS_ERR(file)) filter_parent_unlock(dparent, parent_lockh,parent_mode); case 2: if (IS_ERR(file)) OBD_FREE(fdd, sizeof *fdd); case 1: if (IS_ERR(file)) filter_ffd_destroy(ffd); filter_ffd_put(ffd); case 0: pop_ctxt(&saved, &filter->fo_ctxt, NULL); } RETURN(file); } /* Caller must hold LCK_PW on parent and push us into kernel context. * Caller is also required to ensure that dchild->d_inode exists. */ static int filter_destroy_internal(struct obd_device *obd, struct dentry *dparent, struct dentry *dchild) { struct inode *inode = dchild->d_inode; int rc; ENTRY; if (inode->i_nlink != 1 || atomic_read(&inode->i_count) != 1) { CERROR("destroying objid %*s nlink = %d, count = %d\n", dchild->d_name.len, dchild->d_name.name, inode->i_nlink, atomic_read(&inode->i_count)); } rc = vfs_unlink(dparent->d_inode, dchild); if (rc) CERROR("error unlinking objid %*s: rc %d\n", dchild->d_name.len, dchild->d_name.name, rc); RETURN(rc); } /* If closing because we are failing this device, then don't do the unlink on close. */ static int filter_close_internal(struct obd_export *exp, struct filter_file_data *ffd, struct obd_trans_info *oti, int failover) { struct obd_device *obd = exp->exp_obd; struct filter_obd *filter = &obd->u.filter; struct file *filp = ffd->ffd_file; struct dentry *dchild = dget(filp->f_dentry); struct filter_dentry_data *fdd = dchild->d_fsdata; struct lustre_handle parent_lockh; int rc, rc2, cleanup_phase = 0; struct dentry *dparent = NULL; struct obd_run_ctxt saved; ENTRY; LASSERT(filp->private_data == ffd); LASSERT(fdd); LASSERT(fdd->fdd_magic == FILTER_DENTRY_MAGIC); rc = filp_close(filp, 0); if (atomic_dec_and_test(&fdd->fdd_open_count) && fdd->fdd_flags & FILTER_FLAG_DESTROY && !failover) { void *handle; push_ctxt(&saved, &filter->fo_ctxt, NULL); cleanup_phase = 1; LASSERT(fdd->fdd_objid > 0); dparent = filter_parent_lock(obd, S_IFREG, fdd->fdd_objid, LCK_PW, &parent_lockh); if (IS_ERR(dparent)) GOTO(cleanup, rc = PTR_ERR(dparent)); cleanup_phase = 2; handle = fsfilt_start(obd, dparent->d_inode, FSFILT_OP_UNLINK); if (IS_ERR(handle)) GOTO(cleanup, rc = PTR_ERR(handle)); /* XXX unlink from PENDING directory now too */ rc2 = filter_destroy_internal(obd, dparent, dchild); if (rc2 && !rc) rc = rc2; rc = filter_finish_transno(exp, handle, oti, rc); rc2 = fsfilt_commit(obd, dparent->d_inode, handle, 0); if (rc2) { CERROR("error on commit, err = %d\n", rc2); if (!rc) rc = rc2; } } cleanup: switch(cleanup_phase) { case 2: if (rc || oti == NULL) { filter_parent_unlock(dparent, &parent_lockh, LCK_PW); } else { memcpy(&oti->oti_ack_locks[0].lock, &parent_lockh, sizeof(parent_lockh)); oti->oti_ack_locks[0].mode = LCK_PW; } case 1: pop_ctxt(&saved, &filter->fo_ctxt, NULL); case 0: f_dput(dchild); filter_ffd_destroy(ffd); break; default: CERROR("invalid cleanup_phase %d\n", cleanup_phase); LBUG(); } RETURN(rc); } /* obd methods */ /* mount the file system (secretly) */ static int filter_common_setup(struct obd_device *obd, obd_count len, void *buf, char *option) { struct obd_ioctl_data* data = buf; struct filter_obd *filter = &obd->u.filter; struct vfsmount *mnt; int rc = 0; ENTRY; if (!data->ioc_inlbuf1 || !data->ioc_inlbuf2) RETURN(-EINVAL); obd->obd_fsops = fsfilt_get_ops(data->ioc_inlbuf2); if (IS_ERR(obd->obd_fsops)) RETURN(PTR_ERR(obd->obd_fsops)); mnt = do_kern_mount(data->ioc_inlbuf2, 0, data->ioc_inlbuf1, option); rc = PTR_ERR(mnt); if (IS_ERR(mnt)) GOTO(err_ops, rc); if (data->ioc_inllen3 > 0 && data->ioc_inlbuf3) { if (*data->ioc_inlbuf3 == 'f') { obd->obd_replayable = 1; obd_sync_filter = 1; CERROR("%s: configured for recovery and sync write\n", obd->obd_name); } else { if (*data->ioc_inlbuf3 != 'n') { CERROR("unrecognised flag '%c'\n", *data->ioc_inlbuf3); } } } if (data->ioc_inllen4 > 0 && data->ioc_inlbuf4) { if (*data->ioc_inlbuf4 == '/') { CERROR("filter namespace mount: %s\n", data->ioc_inlbuf4); filter->fo_nspath = strdup(data->ioc_inlbuf4); } else { CERROR("namespace mount must be absolute path: '%s'\n", data->ioc_inlbuf4); } } filter->fo_vfsmnt = mnt; filter->fo_sb = mnt->mnt_sb; filter->fo_fstype = mnt->mnt_sb->s_type->name; CDEBUG(D_SUPER, "%s: mnt = %p\n", filter->fo_fstype, mnt); OBD_SET_CTXT_MAGIC(&filter->fo_ctxt); filter->fo_ctxt.pwdmnt = mnt; filter->fo_ctxt.pwd = mnt->mnt_root; filter->fo_ctxt.fs = get_ds(); rc = filter_prep(obd); if (rc) GOTO(err_mntput, rc); spin_lock_init(&filter->fo_translock); spin_lock_init(&filter->fo_fddlock); spin_lock_init(&filter->fo_objidlock); INIT_LIST_HEAD(&filter->fo_export_list); obd->obd_namespace = ldlm_namespace_new("filter-tgt", LDLM_NAMESPACE_SERVER); if (!obd->obd_namespace) GOTO(err_post, rc = -ENOMEM); ptlrpc_init_client(LDLM_CB_REQUEST_PORTAL, LDLM_CB_REPLY_PORTAL, "filter_ldlm_cb_client", &obd->obd_ldlm_client); RETURN(0); err_post: filter_post(obd); err_mntput: unlock_kernel(); mntput(mnt); filter->fo_sb = 0; lock_kernel(); err_ops: fsfilt_put_ops(obd->obd_fsops); return rc; } static int filter_setup(struct obd_device *obd, obd_count len, void *buf) { struct obd_ioctl_data* data = buf; char *option = NULL; if (!strcmp(data->ioc_inlbuf2, "ext3")) option = "asyncdel"; return filter_common_setup(obd, len, buf, option); } /* sanobd setup methods - use a specific mount option */ static int filter_san_setup(struct obd_device *obd, obd_count len, void *buf) { struct obd_ioctl_data* data = buf; char *option = NULL; if (!data->ioc_inlbuf2) RETURN(-EINVAL); /* for extN/ext3 filesystem, we must mount it with 'writeback' mode */ if (!strcmp(data->ioc_inlbuf2, "extN")) option = "data=writeback"; else if (!strcmp(data->ioc_inlbuf2, "ext3")) option = "data=writeback,asyncdel"; else LBUG(); /* just a reminder */ return filter_common_setup(obd, len, buf, option); } static int filter_cleanup(struct obd_device *obd, int force, int failover) { struct super_block *sb; ENTRY; if (failover) CERROR("%s: shutting down for failover; client state will" " be preserved.\n", obd->obd_name); if (!list_empty(&obd->obd_exports)) { CERROR("%s: still has clients!\n", obd->obd_name); class_disconnect_exports(obd, failover); if (!list_empty(&obd->obd_exports)) { CERROR("still has exports after forced cleanup?\n"); RETURN(-EBUSY); } } ldlm_namespace_free(obd->obd_namespace); sb = obd->u.filter.fo_sb; if (!sb) RETURN(0); filter_post(obd); shrink_dcache_parent(sb->s_root); unlock_kernel(); if (atomic_read(&obd->u.filter.fo_vfsmnt->mnt_count) > 1){ CERROR("%s: mount point busy, mnt_count: %d\n", obd->obd_name, atomic_read(&obd->u.filter.fo_vfsmnt->mnt_count)); } mntput(obd->u.filter.fo_vfsmnt); obd->u.filter.fo_sb = 0; /* destroy_buffers(obd->u.filter.fo_sb->s_dev);*/ fsfilt_put_ops(obd->obd_fsops); lock_kernel(); RETURN(0); } int filter_attach(struct obd_device *obd, obd_count len, void *data) { struct lprocfs_static_vars lvars; int rc; lprocfs_init_vars(&lvars); rc = lprocfs_obd_attach(obd, lvars.obd_vars); if (rc != 0) return rc; rc = lprocfs_alloc_obd_stats(obd, LPROC_FILTER_LAST); if (rc != 0) return rc; /* Init obdfilter private stats here */ lprocfs_counter_init(obd->obd_stats, LPROC_FILTER_READ_BYTES, LPROCFS_CNTR_AVGMINMAX, "read_bytes", "bytes"); lprocfs_counter_init(obd->obd_stats, LPROC_FILTER_WRITE_BYTES, LPROCFS_CNTR_AVGMINMAX, "write_bytes", "bytes"); return rc; } int filter_detach(struct obd_device *dev) { lprocfs_free_obd_stats(dev); return lprocfs_obd_detach(dev); } /* nearly identical to mds_connect */ static int filter_connect(struct lustre_handle *conn, struct obd_device *obd, struct obd_uuid *cluuid) { struct obd_export *exp; struct filter_export_data *fed; struct filter_client_data *fcd; struct filter_obd *filter = &obd->u.filter; int rc; ENTRY; if (!conn || !obd || !cluuid) RETURN(-EINVAL); rc = class_connect(conn, obd, cluuid); if (rc) RETURN(rc); exp = class_conn2export(conn); LASSERT(exp); fed = &exp->exp_filter_data; class_export_put(exp); INIT_LIST_HEAD(&fed->fed_open_head); spin_lock_init(&fed->fed_lock); if (!obd->obd_replayable) RETURN(0); OBD_ALLOC(fcd, sizeof(*fcd)); if (!fcd) { CERROR("filter: out of memory for client data\n"); GOTO(out_export, rc = -ENOMEM); } memcpy(fcd->fcd_uuid, cluuid, sizeof(fcd->fcd_uuid)); fed->fed_fcd = fcd; fcd->fcd_mount_count = cpu_to_le64(filter->fo_fsd->fsd_mount_count); rc = filter_client_add(obd, filter, fed, -1); if (rc) GOTO(out_fcd, rc); RETURN(rc); out_fcd: OBD_FREE(fcd, sizeof(*fcd)); out_export: class_disconnect(conn, 0); RETURN(rc); } static void filter_destroy_export(struct obd_export *exp) { struct filter_export_data *fed = &exp->exp_filter_data; ENTRY; spin_lock(&fed->fed_lock); while (!list_empty(&fed->fed_open_head)) { struct filter_file_data *ffd; ffd = list_entry(fed->fed_open_head.next, typeof(*ffd), ffd_export_list); list_del(&ffd->ffd_export_list); spin_unlock(&fed->fed_lock); CERROR("force close file %*s (hdl %p:"LPX64") on disconnect\n", ffd->ffd_file->f_dentry->d_name.len, ffd->ffd_file->f_dentry->d_name.name, ffd, ffd->ffd_handle.h_cookie); filter_close_internal(exp, ffd, NULL, exp->exp_failover); spin_lock(&fed->fed_lock); } spin_unlock(&fed->fed_lock); if (exp->exp_obd->obd_replayable) filter_client_free(exp, exp->exp_failover); EXIT; } /* also incredibly similar to mds_disconnect */ static int filter_disconnect(struct lustre_handle *conn, int failover) { struct obd_export *exp = class_conn2export(conn); int rc; unsigned long flags; ENTRY; LASSERT(exp); ldlm_cancel_locks_for_export(exp); spin_lock_irqsave(&exp->exp_lock, flags); exp->exp_failover = failover; spin_unlock_irqrestore(&exp->exp_lock, flags); rc = class_disconnect(conn, failover); fsfilt_sync(exp->exp_obd, exp->exp_obd->u.filter.fo_sb); class_export_put(exp); /* XXX cleanup preallocated inodes */ RETURN(rc); } static void filter_from_inode(struct obdo *oa, struct inode *inode, int valid) { int type = oa->o_mode & S_IFMT; ENTRY; CDEBUG(D_INFO, "src inode %lu (%p), dst obdo "LPU64" valid 0x%08x\n", inode->i_ino, inode, oa->o_id, valid); /* Don't copy the inode number in place of the object ID */ obdo_from_inode(oa, inode, valid); oa->o_mode &= ~S_IFMT; oa->o_mode |= type; if (S_ISCHR(inode->i_mode) || S_ISBLK(inode->i_mode)) { obd_rdev rdev = kdev_t_to_nr(inode->i_rdev); oa->o_rdev = rdev; oa->o_valid |= OBD_MD_FLRDEV; } EXIT; } static struct dentry *__filter_oa2dentry(struct lustre_handle *conn, struct obdo *oa, char *what) { struct dentry *dchild = NULL; if (oa->o_valid & OBD_MD_FLHANDLE) { struct lustre_handle *ost_handle = obdo_handle(oa); struct filter_file_data *ffd = filter_handle2ffd(ost_handle); if (ffd != NULL) { struct filter_dentry_data *fdd; dchild = dget(ffd->ffd_file->f_dentry); fdd = dchild->d_fsdata; LASSERT(fdd->fdd_magic == FILTER_DENTRY_MAGIC); filter_ffd_put(ffd); CDEBUG(D_INODE, "got child objid %*s: %p, count = %d\n", dchild->d_name.len, dchild->d_name.name, dchild, atomic_read(&dchild->d_count)); } } if (!dchild) { struct obd_device *obd = class_conn2obd(conn); if (!obd) { CERROR("invalid client cookie "LPX64"\n", conn->cookie); RETURN(ERR_PTR(-EINVAL)); } dchild = filter_fid2dentry(obd, NULL, oa->o_mode, oa->o_id); } if (IS_ERR(dchild)) { CERROR("%s error looking up object: "LPU64"\n", what, oa->o_id); RETURN(dchild); } if (!dchild->d_inode) { CERROR("%s on non-existent object: "LPU64"\n", what, oa->o_id); f_dput(dchild); RETURN(ERR_PTR(-ENOENT)); } return dchild; } #define filter_oa2dentry(conn, oa) __filter_oa2dentry(conn, oa, __FUNCTION__) static int filter_getattr(struct lustre_handle *conn, struct obdo *oa, struct lov_stripe_md *md) { struct dentry *dentry = NULL; int rc = 0; ENTRY; dentry = filter_oa2dentry(conn, oa); if (IS_ERR(dentry)) RETURN(PTR_ERR(dentry)); filter_from_inode(oa, dentry->d_inode, oa->o_valid); f_dput(dentry); RETURN(rc); } /* this is called from filter_truncate() until we have filter_punch() */ static int filter_setattr(struct lustre_handle *conn, struct obdo *oa, struct lov_stripe_md *md, struct obd_trans_info *oti) { struct obd_run_ctxt saved; struct obd_export *export = class_conn2export(conn); struct obd_device *obd = class_conn2obd(conn); struct filter_obd *filter = &obd->u.filter; struct dentry *dentry; struct iattr iattr; struct inode *inode; void * handle; int rc, rc2; ENTRY; dentry = filter_oa2dentry(conn, oa); if (IS_ERR(dentry)) GOTO(out_exp, rc = PTR_ERR(dentry)); iattr_from_obdo(&iattr, oa, oa->o_valid); iattr.ia_mode = (iattr.ia_mode & ~S_IFMT) | S_IFREG; inode = dentry->d_inode; push_ctxt(&saved, &filter->fo_ctxt, NULL); lock_kernel(); if (iattr.ia_valid & ATTR_SIZE) down(&inode->i_sem); handle = fsfilt_start(obd, dentry->d_inode, FSFILT_OP_SETATTR); if (IS_ERR(handle)) GOTO(out_unlock, rc = PTR_ERR(handle)); rc = fsfilt_setattr(obd, dentry, handle, &iattr, 1); rc = filter_finish_transno(export, handle, oti, rc); rc2 = fsfilt_commit(obd, dentry->d_inode, handle, 0); if (rc2) { CERROR("error on commit, err = %d\n", rc2); if (!rc) rc = rc2; } if (iattr.ia_valid & ATTR_SIZE) { up(&inode->i_sem); oa->o_valid = OBD_MD_FLBLOCKS | OBD_MD_FLCTIME | OBD_MD_FLMTIME; obdo_from_inode(oa, inode, oa->o_valid); } out_unlock: unlock_kernel(); pop_ctxt(&saved, &filter->fo_ctxt, NULL); f_dput(dentry); out_exp: class_export_put(export); RETURN(rc); } static int filter_open(struct lustre_handle *conn, struct obdo *oa, struct lov_stripe_md *ea, struct obd_trans_info *oti, struct obd_client_handle *och) { struct obd_export *export = NULL; struct lustre_handle *handle; struct filter_file_data *ffd; struct file *filp; struct lustre_handle parent_lockh; int rc = 0; ENTRY; export = class_conn2export(conn); if (!export) { CDEBUG(D_IOCTL, "invalid client cookie "LPX64"\n", conn->cookie); GOTO(out, rc = -EINVAL); } filp = filter_obj_open(export, oa->o_id, oa->o_mode, LCK_PR, &parent_lockh); if (IS_ERR(filp)) GOTO(out, rc = PTR_ERR(filp)); filter_from_inode(oa, filp->f_dentry->d_inode, oa->o_valid); ffd = filp->private_data; handle = obdo_handle(oa); handle->cookie = ffd->ffd_handle.h_cookie; oa->o_valid |= OBD_MD_FLHANDLE; out: class_export_put(export); if (!rc) { memcpy(&oti->oti_ack_locks[0].lock, &parent_lockh, sizeof(parent_lockh)); oti->oti_ack_locks[0].mode = LCK_PR; } RETURN(rc); } static int filter_close(struct lustre_handle *conn, struct obdo *oa, struct lov_stripe_md *ea, struct obd_trans_info *oti) { struct obd_export *exp = class_conn2export(conn); struct filter_file_data *ffd; struct filter_export_data *fed; int rc; ENTRY; if (!exp) { CDEBUG(D_IOCTL, "invalid client cookie"LPX64"\n", conn->cookie); GOTO(out, rc = -EINVAL); } if (!(oa->o_valid & OBD_MD_FLHANDLE)) { CERROR("no handle for close of objid "LPU64"\n", oa->o_id); GOTO(out, rc = -EINVAL); } ffd = filter_handle2ffd(obdo_handle(oa)); if (ffd == NULL) { CERROR("bad handle ("LPX64") for close\n", obdo_handle(oa)->cookie); GOTO(out, rc = -ESTALE); } fed = &exp->exp_filter_data; spin_lock(&fed->fed_lock); list_del(&ffd->ffd_export_list); spin_unlock(&fed->fed_lock); rc = filter_close_internal(exp, ffd, oti, 0); filter_ffd_put(ffd); GOTO(out, rc); out: class_export_put(exp); return rc; } static int filter_create(struct lustre_handle *conn, struct obdo *oa, struct lov_stripe_md **ea, struct obd_trans_info *oti) { struct obd_export *exp; struct obd_device *obd = class_conn2obd(conn); struct filter_obd *filter = &obd->u.filter; struct obd_run_ctxt saved; struct lustre_handle parent_lockh; struct dentry *dparent; struct dentry *dchild = NULL; struct iattr; void *handle; int err, rc, cleanup_phase; ENTRY; if (!obd) { CERROR("invalid client cookie "LPX64"\n", conn->cookie); RETURN(-EINVAL); } exp = class_conn2export(conn); push_ctxt(&saved, &filter->fo_ctxt, NULL); retry: oa->o_id = filter_next_id(filter); cleanup_phase = 0; dparent = filter_parent_lock(obd, S_IFREG, oa->o_id, LCK_PW, &parent_lockh); if (IS_ERR(dparent)) GOTO(cleanup, rc = PTR_ERR(dparent)); cleanup_phase = 1; dchild = filter_fid2dentry(obd, dparent, S_IFREG, oa->o_id); if (IS_ERR(dchild)) GOTO(cleanup, rc = PTR_ERR(dchild)); if (dchild->d_inode) { /* This would only happen if lastobjid was bad on disk */ CERROR("Serious error: objid %*s already exists; is this " "filesystem corrupt? I will try to work around it.\n", dchild->d_name.len, dchild->d_name.name); f_dput(dchild); filter_parent_unlock(dparent, &parent_lockh, LCK_PW); goto retry; } cleanup_phase = 2; handle = fsfilt_start(obd, dparent->d_inode, FSFILT_OP_CREATE); if (IS_ERR(handle)) GOTO(cleanup, rc = PTR_ERR(handle)); rc = vfs_create(dparent->d_inode, dchild, oa->o_mode); if (rc) CERROR("create failed rc = %d\n", rc); rc = filter_finish_transno(exp, handle, oti, rc); err = filter_update_server_data(filter->fo_rcvd_filp, filter->fo_fsd); if (err) { CERROR("unable to write lastobjid but file created\n"); if (!rc) rc = err; } err = fsfilt_commit(obd, dparent->d_inode, handle, 0); if (err) { CERROR("error on commit, err = %d\n", err); if (!rc) rc = err; } if (rc) GOTO(cleanup, rc); /* Set flags for fields we have set in the inode struct */ oa->o_valid = OBD_MD_FLID | OBD_MD_FLBLKSZ | OBD_MD_FLBLOCKS | OBD_MD_FLMTIME | OBD_MD_FLATIME | OBD_MD_FLCTIME; filter_from_inode(oa, dchild->d_inode, oa->o_valid); EXIT; cleanup: switch(cleanup_phase) { case 2: f_dput(dchild); case 1: /* locked parent dentry */ if (rc || oti == NULL) { filter_parent_unlock(dparent, &parent_lockh, LCK_PW); } else { memcpy(&oti->oti_ack_locks[0].lock, &parent_lockh, sizeof(parent_lockh)); oti->oti_ack_locks[0].mode = LCK_PW; } case 0: pop_ctxt(&saved, &filter->fo_ctxt, NULL); class_export_put(exp); break; default: CERROR("invalid cleanup_phase %d\n", cleanup_phase); LBUG(); } RETURN(rc); } static int filter_destroy(struct lustre_handle *conn, struct obdo *oa, struct lov_stripe_md *ea, struct obd_trans_info *oti) { struct obd_export *exp; struct obd_device *obd = class_conn2obd(conn); struct filter_obd *filter = &obd->u.filter; struct dentry *dparent, *dchild = NULL; struct filter_dentry_data *fdd; struct obd_run_ctxt saved; void *handle = NULL; struct lustre_handle parent_lockh; int rc, rc2, cleanup_phase = 0; ENTRY; if (!obd) { CERROR("invalid client cookie "LPX64"\n", conn->cookie); RETURN(-EINVAL); } exp = class_conn2export(conn); CDEBUG(D_INODE, "destroying objid "LPU64"\n", oa->o_id); push_ctxt(&saved, &filter->fo_ctxt, NULL); dparent = filter_parent_lock(obd, oa->o_mode, oa->o_id, LCK_PW, &parent_lockh); if (IS_ERR(dparent)) GOTO(cleanup, rc = PTR_ERR(dparent)); cleanup_phase = 1; dchild = filter_fid2dentry(obd, dparent, S_IFREG, oa->o_id); if (IS_ERR(dchild)) GOTO(cleanup, rc = -ENOENT); cleanup_phase = 2; if (!dchild->d_inode) { CERROR("destroying non-existent object "LPU64"\n", oa->o_id); GOTO(cleanup, rc = -ENOENT); } handle = fsfilt_start(obd, dparent->d_inode, FSFILT_OP_UNLINK); if (IS_ERR(handle)) GOTO(cleanup, rc = PTR_ERR(handle)); cleanup_phase = 3; fdd = dchild->d_fsdata; if (fdd && atomic_read(&fdd->fdd_open_count)) { LASSERT(fdd->fdd_magic = FILTER_DENTRY_MAGIC); if (!(fdd->fdd_flags & FILTER_FLAG_DESTROY)) { fdd->fdd_flags |= FILTER_FLAG_DESTROY; /* XXX put into PENDING directory in case of crash */ CDEBUG(D_INODE, "defer destroy of %dx open objid "LPU64"\n", atomic_read(&fdd->fdd_open_count), oa->o_id); } else CDEBUG(D_INODE, "repeat destroy of %dx open objid "LPU64"\n", atomic_read(&fdd->fdd_open_count), oa->o_id); GOTO(cleanup, rc = 0); } rc = filter_destroy_internal(obd, dparent, dchild); cleanup: switch(cleanup_phase) { case 3: rc = filter_finish_transno(exp, handle, oti, rc); rc2 = fsfilt_commit(obd, dparent->d_inode, handle, 0); if (rc2) { CERROR("error on commit, err = %d\n", rc2); if (!rc) rc = rc2; } case 2: f_dput(dchild); case 1: if (rc || oti == NULL) { filter_parent_unlock(dparent, &parent_lockh, LCK_PW); } else { memcpy(&oti->oti_ack_locks[0].lock, &parent_lockh, sizeof(parent_lockh)); oti->oti_ack_locks[0].mode = LCK_PW; } case 0: pop_ctxt(&saved, &filter->fo_ctxt, NULL); class_export_put(exp); break; default: CERROR("invalid cleanup_phase %d\n", cleanup_phase); LBUG(); } RETURN(rc); } /* NB start and end are used for punch, but not truncate */ static int filter_truncate(struct lustre_handle *conn, struct obdo *oa, struct lov_stripe_md *lsm, obd_off start, obd_off end, struct obd_trans_info *oti) { int error; ENTRY; if (end != OBD_OBJECT_EOF) CERROR("PUNCH not supported, only truncate: end = "LPX64"\n", end); CDEBUG(D_INODE, "calling truncate for object "LPU64", valid = %x, " "o_size = "LPD64"\n", oa->o_id, oa->o_valid, start); oa->o_size = start; error = filter_setattr(conn, oa, NULL, oti); RETURN(error); } static inline void lustre_put_page(struct page *page) { page_cache_release(page); } static int filter_start_page_read(struct inode *inode, struct niobuf_local *lnb) { struct address_space *mapping = inode->i_mapping; struct page *page; unsigned long index = lnb->offset >> PAGE_SHIFT; int rc; page = grab_cache_page(mapping, index); /* locked page */ if (IS_ERR(page)) return lnb->rc = PTR_ERR(page); lnb->page = page; if (inode->i_size < lnb->offset + lnb->len - 1) lnb->rc = inode->i_size - lnb->offset; else lnb->rc = lnb->len; if (PageUptodate(page)) { unlock_page(page); return 0; } rc = mapping->a_ops->readpage(NULL, page); if (rc < 0) { CERROR("page index %lu, rc = %d\n", index, rc); lnb->page = NULL; lustre_put_page(page); return lnb->rc = rc; } return 0; } static int filter_finish_page_read(struct niobuf_local *lnb) { if (lnb->page == NULL) return 0; if (PageUptodate(lnb->page)) return 0; wait_on_page(lnb->page); if (!PageUptodate(lnb->page)) { CERROR("page index %lu/offset "LPX64" not uptodate\n", lnb->page->index, lnb->offset); GOTO(err_page, lnb->rc = -EIO); } if (PageError(lnb->page)) { CERROR("page index %lu/offset "LPX64" has error\n", lnb->page->index, lnb->offset); GOTO(err_page, lnb->rc = -EIO); } return 0; err_page: lustre_put_page(lnb->page); lnb->page = NULL; return lnb->rc; } static struct page *lustre_get_page_write(struct inode *inode, unsigned long index) { struct address_space *mapping = inode->i_mapping; struct page *page; int rc; page = grab_cache_page(mapping, index); /* locked page */ if (!IS_ERR(page)) { /* Note: Called with "O" and "PAGE_SIZE" this is essentially * a no-op for most filesystems, because we write the whole * page. For partial-page I/O this will read in the page. */ rc = mapping->a_ops->prepare_write(NULL, page, 0, PAGE_SIZE); if (rc) { CERROR("page index %lu, rc = %d\n", index, rc); if (rc != -ENOSPC) LBUG(); GOTO(err_unlock, rc); } /* XXX not sure if we need this if we are overwriting page */ if (PageError(page)) { CERROR("error on page index %lu, rc = %d\n", index, rc); LBUG(); GOTO(err_unlock, rc = -EIO); } } return page; err_unlock: unlock_page(page); lustre_put_page(page); return ERR_PTR(rc); } #if (LINUX_VERSION_CODE > KERNEL_VERSION(2,5,0)) int waitfor_one_page(struct page *page) { wait_on_page_locked(page); return 0; } #endif #if (LINUX_VERSION_CODE < KERNEL_VERSION(2,5,0)) /* We should only change the file mtime (and not the ctime, like * update_inode_times() in generic_file_write()) when we only change data. */ static inline void inode_update_time(struct inode *inode, int ctime_too) { time_t now = CURRENT_TIME; if (inode->i_mtime == now && (!ctime_too || inode->i_ctime == now)) return; inode->i_mtime = now; if (ctime_too) inode->i_ctime = now; mark_inode_dirty_sync(inode); } #endif static int lustre_commit_write(struct niobuf_local *lnb) { struct page *page = lnb->page; unsigned from = lnb->offset & ~PAGE_MASK; unsigned to = from + lnb->len; struct inode *inode = page->mapping->host; int err; LASSERT(to <= PAGE_SIZE); err = page->mapping->a_ops->commit_write(NULL, page, from, to); if (!err && IS_SYNC(inode)) err = waitfor_one_page(page); //SetPageUptodate(page); // the client commit_write will do this SetPageReferenced(page); unlock_page(page); lustre_put_page(page); return err; } int filter_get_page_write(struct inode *inode, struct niobuf_local *lnb, int *pglocked) { unsigned long index = lnb->offset >> PAGE_SHIFT; struct address_space *mapping = inode->i_mapping; struct page *page; int rc; //ASSERT_PAGE_INDEX(index, GOTO(err, rc = -EINVAL)); if (*pglocked) page = grab_cache_page_nowait(mapping, index); /* locked page */ else page = grab_cache_page(mapping, index); /* locked page */ /* This page is currently locked, so get a temporary page instead. */ if (!page) { CDEBUG(D_ERROR,"ino %lu page %ld locked\n", inode->i_ino,index); page = alloc_pages(GFP_KERNEL, 0); /* locked page */ if (!page) { CERROR("no memory for a temp page\n"); GOTO(err, rc = -ENOMEM); } page->index = index; lnb->page = page; lnb->flags |= N_LOCAL_TEMP_PAGE; } else if (!IS_ERR(page)) { (*pglocked)++; rc = mapping->a_ops->prepare_write(NULL, page, lnb->offset & ~PAGE_MASK, lnb->len); if (rc) { if (rc != -ENOSPC) CERROR("page index %lu, rc = %d\n", index, rc); GOTO(err_unlock, rc); } /* XXX not sure if we need this if we are overwriting page */ if (PageError(page)) { CERROR("error on page index %lu, rc = %d\n", index, rc); LBUG(); GOTO(err_unlock, rc = -EIO); } lnb->page = page; } return 0; err_unlock: unlock_page(page); lustre_put_page(page); err: return lnb->rc = rc; } /* * We need to balance prepare_write() calls with commit_write() calls. * If the page has been prepared, but we have no data for it, we don't * want to overwrite valid data on disk, but we still need to zero out * data for space which was newly allocated. Like part of what happens * in __block_prepare_write() for newly allocated blocks. * * XXX currently __block_prepare_write() creates buffers for all the * pages, and the filesystems mark these buffers as BH_New if they * were newly allocated from disk. We use the BH_New flag similarly. */ static int filter_commit_write(struct niobuf_local *lnb, int err) { #if (LINUX_VERSION_CODE < KERNEL_VERSION(2,5,0)) if (err) { unsigned block_start, block_end; struct buffer_head *bh, *head = lnb->page->buffers; unsigned blocksize = head->b_size; /* debugging: just seeing if this ever happens */ CDEBUG(err == -ENOSPC ? D_INODE : D_ERROR, "called for ino %lu:%lu on err %d\n", lnb->page->mapping->host->i_ino, lnb->page->index, err); /* Currently one buffer per page, but in the future... */ for (bh = head, block_start = 0; bh != head || !block_start; block_start = block_end, bh = bh->b_this_page) { block_end = block_start + blocksize; if (buffer_new(bh)) { memset(kmap(lnb->page) + block_start, 0, blocksize); kunmap(lnb->page); } } } #endif return lustre_commit_write(lnb); } static int filter_preprw(int cmd, struct obd_export *exp, struct obdo *obdo, int objcount, struct obd_ioobj *obj, int niocount, struct niobuf_remote *nb, struct niobuf_local *res, void **desc_private, struct obd_trans_info *oti) { struct obd_run_ctxt saved; struct obd_device *obd; struct obd_ioobj *o; struct niobuf_remote *rnb; struct niobuf_local *lnb; struct fsfilt_objinfo *fso; struct dentry *dentry; struct inode *inode; int pglocked = 0, rc = 0, i, j, tot_bytes = 0; unsigned long now = jiffies; ENTRY; memset(res, 0, niocount * sizeof(*res)); obd = exp->exp_obd; if (obd == NULL) RETURN(-EINVAL); // theoretically we support multi-obj BRW RPCs, but until then... LASSERT(objcount == 1); OBD_ALLOC(fso, objcount * sizeof(*fso)); if (!fso) RETURN(-ENOMEM); push_ctxt(&saved, &obd->u.filter.fo_ctxt, NULL); for (i = 0, o = obj; i < objcount; i++, o++) { struct filter_dentry_data *fdd; LASSERT(o->ioo_bufcnt); dentry = filter_fid2dentry(obd, NULL, o->ioo_type, o->ioo_id); if (IS_ERR(dentry)) GOTO(out_objinfo, rc = PTR_ERR(dentry)); fso[i].fso_dentry = dentry; fso[i].fso_bufcnt = o->ioo_bufcnt; if (!dentry->d_inode) { CERROR("trying to BRW to non-existent file "LPU64"\n", o->ioo_id); f_dput(dentry); GOTO(out_objinfo, rc = -ENOENT); } /* If we ever start to support mutli-object BRW RPCs, we will * need to get locks on mulitple inodes (in order) or use the * DLM to do the locking for us (and use the same locking in * filter_setattr() for truncate). That isn't all, because * there still exists the possibility of a truncate starting * a new transaction while holding the ext3 rwsem = write * while some writes (which have started their transactions * here) blocking on the ext3 rwsem = read => lock inversion. * * The handling gets very ugly when dealing with locked pages. * It may be easier to just get rid of the locked page code * (which has problems of its own) and either discover we do * not need it anymore (i.e. it was a symptom of another bug) * or ensure we get the page locks in an appropriate order. */ if (cmd & OBD_BRW_WRITE) down(&dentry->d_inode->i_sem); fdd = dentry->d_fsdata; if (!fdd || !atomic_read(&fdd->fdd_open_count)) CDEBUG(D_PAGE, "I/O to unopened object "LPU64"\n", o->ioo_id); } if (time_after(jiffies, now + 15*HZ)) CERROR("slow prep setup %lus\n", (jiffies - now) / HZ); if (cmd & OBD_BRW_WRITE) { *desc_private = fsfilt_brw_start(obd, objcount, fso, niocount, nb); if (IS_ERR(*desc_private)) { rc = PTR_ERR(*desc_private); CDEBUG(rc == -ENOSPC ? D_INODE : D_ERROR, "error starting transaction: rc = %d\n", rc); *desc_private = NULL; GOTO(out_objinfo, rc); } } for (i = 0, o = obj, rnb = nb, lnb = res; i < objcount; i++, o++) { dentry = fso[i].fso_dentry; inode = dentry->d_inode; for (j = 0; j < o->ioo_bufcnt; j++, rnb++, lnb++) { if (j == 0) lnb->dentry = dentry; else lnb->dentry = dget(dentry); lnb->offset = rnb->offset; lnb->len = rnb->len; lnb->flags = rnb->flags; lnb->start = jiffies; if (cmd & OBD_BRW_WRITE) { rc = filter_get_page_write(inode,lnb,&pglocked); if (rc) up(&dentry->d_inode->i_sem); } else if (inode->i_size <= rnb->offset) { /* If there's no more data, abort early. * lnb->page == NULL and lnb->rc == 0, so it's * easy to detect later. */ f_dput(dentry); lnb->dentry = NULL; break; } else { rc = filter_start_page_read(inode, lnb); } if (rc) { CDEBUG(rc == -ENOSPC ? D_INODE : D_ERROR, "page err %u@"LPU64" %u/%u %p: rc %d\n", lnb->len, lnb->offset, j, o->ioo_bufcnt, dentry, rc); f_dput(dentry); GOTO(out_pages, rc); } tot_bytes += lnb->len; if ((cmd & OBD_BRW_READ) && lnb->rc < lnb->len) { /* Likewise with a partial read */ break; } } } if (time_after(jiffies, now + 15*HZ)) CERROR("slow prep get page %lus\n", (jiffies - now) / HZ); if (cmd & OBD_BRW_READ) { lprocfs_counter_add(obd->obd_stats, LPROC_FILTER_READ_BYTES, tot_bytes); while (lnb-- > res) { rc = filter_finish_page_read(lnb); if (rc) { CERROR("error page %u@"LPU64" %u %p: rc %d\n", lnb->len, lnb->offset, lnb - res, lnb->dentry, rc); f_dput(lnb->dentry); GOTO(out_pages, rc); } } } else lprocfs_counter_add(obd->obd_stats, LPROC_FILTER_WRITE_BYTES, tot_bytes); if (time_after(jiffies, now + 15*HZ)) CERROR("slow prep finish page %lus\n", (jiffies - now) / HZ); EXIT; out: OBD_FREE(fso, objcount * sizeof(*fso)); current->journal_info = NULL; pop_ctxt(&saved, &obd->u.filter.fo_ctxt, NULL); return rc; out_pages: while (lnb-- > res) { if (cmd & OBD_BRW_WRITE) { filter_commit_write(lnb, rc); up(&lnb->dentry->d_inode->i_sem); } else { lustre_put_page(lnb->page); } f_dput(lnb->dentry); } if (cmd & OBD_BRW_WRITE) { filter_finish_transno(exp, *desc_private, oti, rc); fsfilt_commit(obd, filter_parent(obd,S_IFREG,obj->ioo_id)->d_inode, *desc_private, 0); } goto out; /* dropped the dentry refs already (one per page) */ out_objinfo: for (i = 0; i < objcount && fso[i].fso_dentry; i++) { if (cmd & OBD_BRW_WRITE) up(&fso[i].fso_dentry->d_inode->i_sem); f_dput(fso[i].fso_dentry); } goto out; } static int filter_write_locked_page(struct niobuf_local *lnb) { struct page *lpage; void *lpage_addr; void *lnb_addr; int rc; ENTRY; lpage = lustre_get_page_write(lnb->dentry->d_inode, lnb->page->index); if (IS_ERR(lpage)) { /* It is highly unlikely that we would ever get an error here. * The page we want to get was previously locked, so it had to * have already allocated the space, and we were just writing * over the same data, so there would be no hole in the file. * * XXX: possibility of a race with truncate could exist, need * to check that. There are no guarantees w.r.t. * write order even on a local filesystem, although the * normal response would be to return the number of bytes * successfully written and leave the rest to the app. */ rc = PTR_ERR(lpage); CERROR("error getting locked page index %ld: rc = %d\n", lnb->page->index, rc); LBUG(); lustre_commit_write(lnb); RETURN(rc); } /* 2 kmaps == vanishingly small deadlock opportunity */ lpage_addr = kmap(lpage); lnb_addr = kmap(lnb->page); memcpy(lpage_addr, lnb_addr, PAGE_SIZE); kunmap(lnb->page); kunmap(lpage); lustre_put_page(lnb->page); lnb->page = lpage; rc = lustre_commit_write(lnb); if (rc) CERROR("error committing locked page %ld: rc = %d\n", lnb->page->index, rc); RETURN(rc); } static int filter_syncfs(struct obd_export *exp) { struct obd_device *obd = exp->exp_obd; ENTRY; RETURN(fsfilt_sync(obd, obd->u.filter.fo_sb)); } static int filter_commitrw(int cmd, struct obd_export *exp, int objcount, struct obd_ioobj *obj, int niocount, struct niobuf_local *res, void *desc_private, struct obd_trans_info *oti) { struct obd_run_ctxt saved; struct obd_ioobj *o; struct niobuf_local *lnb; struct obd_device *obd = exp->exp_obd; int found_locked = 0, rc = 0, i; unsigned long now = jiffies; /* DEBUGGING OST TIMEOUTS */ ENTRY; push_ctxt(&saved, &obd->u.filter.fo_ctxt, NULL); LASSERT(!current->journal_info); current->journal_info = desc_private; for (i = 0, o = obj, lnb = res; i < objcount; i++, o++) { int j; if (cmd & OBD_BRW_WRITE) { inode_update_time(lnb->dentry->d_inode, 1); up(&lnb->dentry->d_inode->i_sem); } for (j = 0 ; j < o->ioo_bufcnt ; j++, lnb++) { if (lnb->page == NULL) { continue; } if (lnb->flags & N_LOCAL_TEMP_PAGE) { found_locked++; continue; } if (time_after(jiffies, lnb->start + 15*HZ)) CERROR("slow commitrw %lus\n", (jiffies - lnb->start) / HZ); if (cmd & OBD_BRW_WRITE) { int err = filter_commit_write(lnb, 0); if (!rc) rc = err; } else { lustre_put_page(lnb->page); } f_dput(lnb->dentry); if (time_after(jiffies, lnb->start + 15*HZ)) CERROR("slow commit_write %lus\n", (jiffies - lnb->start) / HZ); } } for (i = 0, o = obj, lnb = res; found_locked > 0 && i < objcount; i++, o++) { int j; for (j = 0 ; j < o->ioo_bufcnt ; j++, lnb++) { int err; if (!(lnb->flags & N_LOCAL_TEMP_PAGE)) continue; if (time_after(jiffies, lnb->start + 15*HZ)) CERROR("slow commitrw locked %lus\n", (jiffies - lnb->start) / HZ); err = filter_write_locked_page(lnb); if (!rc) rc = err; f_dput(lnb->dentry); found_locked--; if (time_after(jiffies, lnb->start + 15*HZ)) CERROR("slow commit_write locked %lus\n", (jiffies - lnb->start) / HZ); } } if (cmd & OBD_BRW_WRITE) { /* We just want any dentry for the commit, for now */ struct dentry *dparent = filter_parent(obd, S_IFREG, 0); int err; rc = filter_finish_transno(exp, desc_private, oti, rc); err = fsfilt_commit(obd, dparent->d_inode, desc_private, obd_sync_filter); if (err) rc = err; if (obd_sync_filter) LASSERT(oti->oti_transno <= obd->obd_last_committed); if (time_after(jiffies, now + 15*HZ)) CERROR("slow commitrw commit %lus\n", (jiffies-now)/HZ); } LASSERT(!current->journal_info); pop_ctxt(&saved, &obd->u.filter.fo_ctxt, NULL); RETURN(rc); } static int filter_brw(int cmd, struct lustre_handle *conn, struct lov_stripe_md *lsm, obd_count oa_bufs, struct brw_page *pga, struct obd_trans_info *oti) { struct obd_export *export = class_conn2export(conn); struct obd_ioobj ioo; struct niobuf_local *lnb; struct niobuf_remote *rnb; obd_count i; void *desc_private; int ret = 0; ENTRY; if (export == NULL) RETURN(-EINVAL); OBD_ALLOC(lnb, oa_bufs * sizeof(struct niobuf_local)); OBD_ALLOC(rnb, oa_bufs * sizeof(struct niobuf_remote)); if (lnb == NULL || rnb == NULL) GOTO(out, ret = -ENOMEM); for (i = 0; i < oa_bufs; i++) { rnb[i].offset = pga[i].off; rnb[i].len = pga[i].count; } ioo.ioo_id = lsm->lsm_object_id; ioo.ioo_gr = 0; ioo.ioo_type = S_IFREG; ioo.ioo_bufcnt = oa_bufs; ret = filter_preprw(cmd, export, NULL, 1, &ioo, oa_bufs, rnb, lnb, &desc_private, oti); if (ret != 0) GOTO(out, ret); for (i = 0; i < oa_bufs; i++) { void *virt = kmap(pga[i].pg); obd_off off = pga[i].off & ~PAGE_MASK; void *addr = kmap(lnb[i].page); /* 2 kmaps == vanishingly small deadlock opportunity */ if (cmd & OBD_BRW_WRITE) memcpy(addr + off, virt + off, pga[i].count); else memcpy(virt + off, addr + off, pga[i].count); kunmap(addr); kunmap(virt); } ret = filter_commitrw(cmd, export, 1, &ioo, oa_bufs, lnb, desc_private, oti); out: if (lnb) OBD_FREE(lnb, oa_bufs * sizeof(struct niobuf_local)); if (rnb) OBD_FREE(rnb, oa_bufs * sizeof(struct niobuf_remote)); class_export_put(export); RETURN(ret); } static int filter_san_preprw(int cmd, struct lustre_handle *conn, int objcount, struct obd_ioobj *obj, int niocount, struct niobuf_remote *nb) { struct obd_device *obd; struct obd_ioobj *o = obj; struct niobuf_remote *rnb = nb; int rc = 0; int i; ENTRY; obd = class_conn2obd(conn); if (!obd) { CDEBUG(D_IOCTL, "invalid client cookie "LPX64"\n", conn->cookie); RETURN(-EINVAL); } for (i = 0; i < objcount; i++, o++) { struct dentry *dentry; struct inode *inode; int (*fs_bmap)(struct address_space *, long); int j; dentry = filter_fid2dentry(obd, NULL, o->ioo_type, o->ioo_id); if (IS_ERR(dentry)) GOTO(out, rc = PTR_ERR(dentry)); inode = dentry->d_inode; if (!inode) { CERROR("trying to BRW to non-existent file "LPU64"\n", o->ioo_id); f_dput(dentry); GOTO(out, rc = -ENOENT); } fs_bmap = inode->i_mapping->a_ops->bmap; for (j = 0; j < o->ioo_bufcnt; j++, rnb++) { long block; block = rnb->offset >> inode->i_blkbits; if (cmd == OBD_BRW_READ) { block = fs_bmap(inode->i_mapping, block); } else { loff_t newsize = rnb->offset + rnb->len; /* fs_prep_san_write will also update inode * size for us: * (1) new alloced block * (2) existed block but size extented */ /* FIXME We could call fs_prep_san_write() * only once for all the blocks allocation. * Now call it once for each block, for * simplicity. And if error happens, we * probably need to release previous alloced * block */ rc = fs_prep_san_write(obd, inode, &block, 1, newsize); if (rc) break; } rnb->offset = block; } f_dput(dentry); } out: RETURN(rc); } static int filter_statfs(struct obd_export *exp, struct obd_statfs *osfs) { struct obd_device *obd = exp->exp_obd; ENTRY; RETURN(fsfilt_statfs(obd, obd->u.filter.fo_sb, osfs)); } static int filter_get_info(struct lustre_handle *conn, __u32 keylen, void *key, __u32 *vallen, void *val) { struct obd_device *obd; ENTRY; obd = class_conn2obd(conn); if (!obd) { CDEBUG(D_IOCTL, "invalid client cookie "LPX64"\n", conn->cookie); RETURN(-EINVAL); } if (keylen == strlen("blocksize") && memcmp(key, "blocksize", keylen) == 0) { __u32 *blocksize = val; *vallen = sizeof(*blocksize); *blocksize = obd->u.filter.fo_sb->s_blocksize; RETURN(0); } if (keylen == strlen("blocksize_bits") && memcmp(key, "blocksize_bits", keylen) == 0) { __u32 *blocksize_bits = val; *vallen = sizeof(*blocksize_bits); *blocksize_bits = obd->u.filter.fo_sb->s_blocksize_bits; RETURN(0); } CDEBUG(D_IOCTL, "invalid key\n"); RETURN(-EINVAL); } int filter_copy_data(struct lustre_handle *dst_conn, struct obdo *dst, struct lustre_handle *src_conn, struct obdo *src, obd_size count, obd_off offset, struct obd_trans_info *oti) { struct page *page; struct lov_stripe_md srcmd, dstmd; unsigned long index = 0; int err = 0; LBUG(); /* THIS CODE IS NOT CORRECT -phil */ memset(&srcmd, 0, sizeof(srcmd)); memset(&dstmd, 0, sizeof(dstmd)); srcmd.lsm_object_id = src->o_id; dstmd.lsm_object_id = dst->o_id; ENTRY; CDEBUG(D_INFO, "src: ino "LPU64" blocks "LPU64", size "LPU64 ", dst: ino "LPU64"\n", src->o_id, src->o_blocks, src->o_size, dst->o_id); page = alloc_page(GFP_USER); if (page == NULL) RETURN(-ENOMEM); wait_on_page(page); /* XXX with brw vector I/O, we could batch up reads and writes here, * all we need to do is allocate multiple pages to handle the I/Os * and arrays to handle the request parameters. */ while (index < ((src->o_size + PAGE_SIZE - 1) >> PAGE_SHIFT)) { struct brw_page pg; pg.pg = page; pg.count = PAGE_SIZE; pg.off = (page->index) << PAGE_SHIFT; pg.flag = 0; page->index = index; err = obd_brw(OBD_BRW_READ, src_conn, &srcmd, 1, &pg, NULL); if (err) { EXIT; break; } pg.flag = OBD_BRW_CREATE; CDEBUG(D_INFO, "Read page %ld ...\n", page->index); err = obd_brw(OBD_BRW_WRITE, dst_conn, &dstmd, 1, &pg, oti); /* XXX should handle dst->o_size, dst->o_blocks here */ if (err) { EXIT; break; } CDEBUG(D_INFO, "Wrote page %ld ...\n", page->index); index++; } dst->o_size = src->o_size; dst->o_blocks = src->o_blocks; dst->o_valid |= OBD_MD_FLSIZE | OBD_MD_FLBLOCKS; unlock_page(page); __free_page(page); RETURN(err); } int filter_iocontrol(unsigned int cmd, struct lustre_handle *conn, int len, void *karg, void *uarg) { struct obd_device *obd = class_conn2obd(conn); switch (cmd) { case OBD_IOC_ABORT_RECOVERY: CERROR("aborting recovery for device %s\n", obd->obd_name); target_abort_recovery(obd); RETURN(0); default: RETURN(-EINVAL); } RETURN(0); } static struct obd_ops filter_obd_ops = { o_owner: THIS_MODULE, o_attach: filter_attach, o_detach: filter_detach, o_get_info: filter_get_info, o_setup: filter_setup, o_cleanup: filter_cleanup, o_connect: filter_connect, o_disconnect: filter_disconnect, o_statfs: filter_statfs, o_syncfs: filter_syncfs, o_getattr: filter_getattr, o_create: filter_create, o_setattr: filter_setattr, o_destroy: filter_destroy, o_open: filter_open, o_close: filter_close, o_brw: filter_brw, o_punch: filter_truncate, o_preprw: filter_preprw, o_commitrw: filter_commitrw, o_destroy_export: filter_destroy_export, o_iocontrol: filter_iocontrol, #if 0 o_san_preprw: filter_san_preprw, o_preallocate: filter_preallocate_inodes, o_migrate: filter_migrate, o_copy: filter_copy_data, o_iterate: filter_iterate #endif }; static struct obd_ops filter_sanobd_ops = { o_owner: THIS_MODULE, o_attach: filter_attach, o_detach: filter_detach, o_get_info: filter_get_info, o_setup: filter_san_setup, o_cleanup: filter_cleanup, o_connect: filter_connect, o_disconnect: filter_disconnect, o_statfs: filter_statfs, o_getattr: filter_getattr, o_create: filter_create, o_setattr: filter_setattr, o_destroy: filter_destroy, o_open: filter_open, o_close: filter_close, o_brw: filter_brw, o_punch: filter_truncate, o_preprw: filter_preprw, o_commitrw: filter_commitrw, o_san_preprw: filter_san_preprw, o_destroy_export: filter_destroy_export, o_iocontrol: filter_iocontrol, #if 0 o_preallocate: filter_preallocate_inodes, o_migrate: filter_migrate, o_copy: filter_copy_data, o_iterate: filter_iterate #endif }; static int __init obdfilter_init(void) { struct lprocfs_static_vars lvars; int rc; printk(KERN_INFO "Lustre Filtering OBD driver; info@clusterfs.com\n"); lprocfs_init_vars(&lvars); rc = class_register_type(&filter_obd_ops, lvars.module_vars, OBD_FILTER_DEVICENAME); if (rc) return rc; rc = class_register_type(&filter_sanobd_ops, lvars.module_vars, OBD_FILTER_SAN_DEVICENAME); if (rc) class_unregister_type(OBD_FILTER_DEVICENAME); return rc; } static void __exit obdfilter_exit(void) { class_unregister_type(OBD_FILTER_SAN_DEVICENAME); class_unregister_type(OBD_FILTER_DEVICENAME); } MODULE_AUTHOR("Cluster File Systems, Inc. "); MODULE_DESCRIPTION("Lustre Filtering OBD driver"); MODULE_LICENSE("GPL"); module_init(obdfilter_init); module_exit(obdfilter_exit);