X-Git-Url: https://git.whamcloud.com/?p=fs%2Flustre-release.git;a=blobdiff_plain;f=lustre%2Fllite%2Ffile.c;h=675a5a1967f4b46e5d64a1911f31a857b70e7e94;hp=33da78289d36e65ccb195b3639d2a879a69456e6;hb=cda353e6efae5013a26aedbe49d8aa6fb8fe456e;hpb=ae723cf8161f7af15fb7c2541fa7d4131a816af9 diff --git a/lustre/llite/file.c b/lustre/llite/file.c index 33da782..675a5a1 100644 --- a/lustre/llite/file.c +++ b/lustre/llite/file.c @@ -42,11 +42,11 @@ #include #include #include -#ifdef HAVE_UIDGID_HEADER -# include -#endif +#include +#include #include +#include #include #include "cl_object.h" @@ -58,6 +58,12 @@ struct split_param { __u16 sp_mirror_id; }; +struct pcc_param { + __u64 pa_data_version; + __u32 pa_archive_id; + __u32 pa_layout_gen; +}; + static int ll_put_grouplock(struct inode *inode, struct file *file, unsigned long arg); @@ -73,6 +79,7 @@ static struct ll_file_data *ll_file_data_get(void) return NULL; fd->fd_write_failed = false; + pcc_file_init(&fd->fd_pcc_file); return fd; } @@ -139,8 +146,7 @@ static int ll_close_inode_openhandle(struct inode *inode, if (class_exp2obd(md_exp) == NULL) { CERROR("%s: invalid MDC connection handle closing "DFID"\n", - ll_get_fsname(inode->i_sb, NULL, 0), - PFID(&lli->lli_fid)); + ll_i2sbi(inode)->ll_fsname, PFID(&lli->lli_fid)); GOTO(out, rc = 0); } @@ -157,6 +163,7 @@ static int ll_close_inode_openhandle(struct inode *inode, op_data->op_attr_blocks += ((struct inode *)data)->i_blocks; op_data->op_attr.ia_valid |= ATTR_SIZE; op_data->op_xvalid |= OP_XVALID_BLOCKS; + /* fallthrough */ case MDS_CLOSE_LAYOUT_SPLIT: case MDS_CLOSE_LAYOUT_SWAP: { struct split_param *sp = data; @@ -191,6 +198,17 @@ static int ll_close_inode_openhandle(struct inode *inode, break; } + case MDS_PCC_ATTACH: { + struct pcc_param *param = data; + + LASSERT(data != NULL); + op_data->op_bias |= MDS_HSM_RELEASE | MDS_PCC_ATTACH; + op_data->op_archive_id = param->pa_archive_id; + op_data->op_data_version = param->pa_data_version; + op_data->op_lease_handle = och->och_lease_handle; + break; + } + case MDS_HSM_RELEASE: LASSERT(data != NULL); op_data->op_bias |= MDS_HSM_RELEASE; @@ -221,6 +239,12 @@ static int ll_close_inode_openhandle(struct inode *inode, body = req_capsule_server_get(&req->rq_pill, &RMF_MDT_BODY); if (!(body->mbo_valid & OBD_MD_CLOSE_INTENT_EXECED)) rc = -EBUSY; + + if (bias & MDS_PCC_ATTACH) { + struct pcc_param *param = data; + + param->pa_layout_gen = body->mbo_layout_gen; + } } ll_finish_md_op_data(op_data); @@ -283,7 +307,7 @@ static int ll_md_close(struct inode *inode, struct file *file) .l_inodebits = { MDS_INODELOCK_OPEN }, }; __u64 flags = LDLM_FL_BLOCK_GRANTED | LDLM_FL_TEST_LOCK; - struct ll_file_data *fd = LUSTRE_FPRIVATE(file); + struct ll_file_data *fd = file->private_data; struct ll_inode_info *lli = ll_i2info(inode); struct lustre_handle lockh; enum ldlm_mode lockmode; @@ -330,12 +354,14 @@ static int ll_md_close(struct inode *inode, struct file *file) } mutex_unlock(&lli->lli_och_mutex); - if (!md_lock_match(ll_i2mdexp(inode), flags, ll_inode2fid(inode), + /* LU-4398: do not cache write open lock if the file has exec bit */ + if ((lockmode == LCK_CW && inode->i_mode & S_IXUGO) || + !md_lock_match(ll_i2mdexp(inode), flags, ll_inode2fid(inode), LDLM_IBITS, &policy, lockmode, &lockh)) rc = ll_md_real_close(inode, fd->fd_omode); out: - LUSTRE_FPRIVATE(file) = NULL; + file->private_data = NULL; ll_file_data_put(fd); RETURN(rc); @@ -348,19 +374,19 @@ out: */ int ll_file_release(struct inode *inode, struct file *file) { - struct ll_file_data *fd; - struct ll_sb_info *sbi = ll_i2sbi(inode); - struct ll_inode_info *lli = ll_i2info(inode); - int rc; - ENTRY; + struct ll_file_data *fd; + struct ll_sb_info *sbi = ll_i2sbi(inode); + struct ll_inode_info *lli = ll_i2info(inode); + ktime_t kstart = ktime_get(); + int rc; + + ENTRY; CDEBUG(D_VFSTRACE, "VFS Op:inode="DFID"(%p)\n", PFID(ll_inode2fid(inode)), inode); - if (inode->i_sb->s_root != file_dentry(file)) - ll_stats_ops_tally(sbi, LPROC_LL_RELEASE, 1); - fd = LUSTRE_FPRIVATE(file); - LASSERT(fd != NULL); + fd = file->private_data; + LASSERT(fd != NULL); /* The last ref on @file, maybe not the the owner pid of statahead, * because parent and child process can share the same file handle. */ @@ -368,11 +394,13 @@ int ll_file_release(struct inode *inode, struct file *file) ll_deauthorize_statahead(inode, fd); if (inode->i_sb->s_root == file_dentry(file)) { - LUSTRE_FPRIVATE(file) = NULL; + file->private_data = NULL; ll_file_data_put(fd); - RETURN(0); + GOTO(out, rc = 0); } + pcc_file_release(inode, file); + if (!S_ISDIR(inode->i_mode)) { if (lli->lli_clob != NULL) lov_read_and_clear_async_rc(lli->lli_clob); @@ -384,6 +412,10 @@ int ll_file_release(struct inode *inode, struct file *file) if (CFS_FAIL_TIMEOUT_MS(OBD_FAIL_PTLRPC_DUMP_LOG, cfs_fail_val)) libcfs_debug_dumplog(); +out: + if (!rc && inode->i_sb->s_root != file_dentry(file)) + ll_stats_ops_tally(sbi, LPROC_LL_RELEASE, + ktime_us_delta(ktime_get(), kstart)); RETURN(rc); } @@ -391,52 +423,71 @@ static inline int ll_dom_readpage(void *data, struct page *page) { struct niobuf_local *lnb = data; void *kaddr; + int rc = 0; - kaddr = ll_kmap_atomic(page, KM_USER0); + struct inode *inode = page2inode(page); + + kaddr = kmap_atomic(page); memcpy(kaddr, lnb->lnb_data, lnb->lnb_len); if (lnb->lnb_len < PAGE_SIZE) memset(kaddr + lnb->lnb_len, 0, PAGE_SIZE - lnb->lnb_len); flush_dcache_page(page); SetPageUptodate(page); - ll_kunmap_atomic(kaddr, KM_USER0); + kunmap_atomic(kaddr); + + if (inode && IS_ENCRYPTED(inode) && S_ISREG(inode->i_mode)) { + if (!llcrypt_has_encryption_key(inode)) + CDEBUG(D_SEC, "no enc key for "DFID"\n", + PFID(ll_inode2fid(inode))); + else { + unsigned int offs = 0; + + while (offs < PAGE_SIZE) { + /* decrypt only if page is not empty */ + if (memcmp(page_address(page) + offs, + page_address(ZERO_PAGE(0)), + LUSTRE_ENCRYPTION_UNIT_SIZE) == 0) + break; + + rc = llcrypt_decrypt_pagecache_blocks(page, + LUSTRE_ENCRYPTION_UNIT_SIZE, + offs); + if (rc) + break; + + offs += LUSTRE_ENCRYPTION_UNIT_SIZE; + } + } + } unlock_page(page); - return 0; + return rc; } -void ll_dom_finish_open(struct inode *inode, struct ptlrpc_request *req, - struct lookup_intent *it) +void ll_dom_finish_open(struct inode *inode, struct ptlrpc_request *req) { + struct lu_env *env; + struct cl_io *io; struct ll_inode_info *lli = ll_i2info(inode); struct cl_object *obj = lli->lli_clob; struct address_space *mapping = inode->i_mapping; struct page *vmpage; struct niobuf_remote *rnb; + struct mdt_body *body; char *data; - struct lustre_handle lockh; - struct ldlm_lock *lock; unsigned long index, start; struct niobuf_local lnb; - bool dom_lock = false; + __u16 refcheck; + int rc; ENTRY; if (obj == NULL) RETURN_EXIT; - if (it->it_lock_mode != 0) { - lockh.cookie = it->it_lock_handle; - lock = ldlm_handle2lock(&lockh); - if (lock != NULL) - dom_lock = ldlm_has_dom(lock); - LDLM_LOCK_PUT(lock); - } - if (!dom_lock) - RETURN_EXIT; - - if (!req_capsule_has_field(&req->rq_pill, &RMF_NIOBUF_INLINE, - RCL_SERVER)) + if (!req_capsule_field_present(&req->rq_pill, &RMF_NIOBUF_INLINE, + RCL_SERVER)) RETURN_EXIT; rnb = req_capsule_server_get(&req->rq_pill, &RMF_NIOBUF_INLINE); @@ -448,30 +499,44 @@ void ll_dom_finish_open(struct inode *inode, struct ptlrpc_request *req, * client PAGE_SIZE to be used on that client, if server's PAGE_SIZE is * smaller then offset may be not aligned and that data is just ignored. */ - if (rnb->rnb_offset % PAGE_SIZE) + if (rnb->rnb_offset & ~PAGE_MASK) RETURN_EXIT; - /* Server returns whole file or just file tail if it fills in - * reply buffer, in both cases total size should be inode size. + /* Server returns whole file or just file tail if it fills in reply + * buffer, in both cases total size should be equal to the file size. */ - if (rnb->rnb_offset + rnb->rnb_len < i_size_read(inode)) { - CERROR("%s: server returns off/len %llu/%u < i_size %llu\n", - ll_get_fsname(inode->i_sb, NULL, 0), rnb->rnb_offset, - rnb->rnb_len, i_size_read(inode)); + body = req_capsule_server_get(&req->rq_pill, &RMF_MDT_BODY); + if (rnb->rnb_offset + rnb->rnb_len != body->mbo_dom_size && + !(inode && IS_ENCRYPTED(inode))) { + CERROR("%s: server returns off/len %llu/%u but size %llu\n", + ll_i2sbi(inode)->ll_fsname, rnb->rnb_offset, + rnb->rnb_len, body->mbo_dom_size); RETURN_EXIT; } - CDEBUG(D_INFO, "Get data along with open at %llu len %i, i_size %llu\n", - rnb->rnb_offset, rnb->rnb_len, i_size_read(inode)); + env = cl_env_get(&refcheck); + if (IS_ERR(env)) + RETURN_EXIT; + io = vvp_env_thread_io(env); + io->ci_obj = obj; + io->ci_ignore_layout = 1; + rc = cl_io_init(env, io, CIT_MISC, obj); + if (rc) + GOTO(out_io, rc); + + CDEBUG(D_INFO, "Get data along with open at %llu len %i, size %llu\n", + rnb->rnb_offset, rnb->rnb_len, body->mbo_dom_size); data = (char *)rnb + sizeof(*rnb); lnb.lnb_file_offset = rnb->rnb_offset; - start = lnb.lnb_file_offset / PAGE_SIZE; + start = lnb.lnb_file_offset >> PAGE_SHIFT; index = 0; - LASSERT(lnb.lnb_file_offset % PAGE_SIZE == 0); + LASSERT((lnb.lnb_file_offset & ~PAGE_MASK) == 0); lnb.lnb_page_offset = 0; do { + struct cl_page *page; + lnb.lnb_data = data + (index << PAGE_SHIFT); lnb.lnb_len = rnb->rnb_len - (index << PAGE_SHIFT); if (lnb.lnb_len > PAGE_SIZE) @@ -482,14 +547,38 @@ void ll_dom_finish_open(struct inode *inode, struct ptlrpc_request *req, if (IS_ERR(vmpage)) { CWARN("%s: cannot fill page %lu for "DFID " with data: rc = %li\n", - ll_get_fsname(inode->i_sb, NULL, 0), - index + start, PFID(lu_object_fid(&obj->co_lu)), + ll_i2sbi(inode)->ll_fsname, index + start, + PFID(lu_object_fid(&obj->co_lu)), PTR_ERR(vmpage)); break; } + lock_page(vmpage); + if (vmpage->mapping == NULL) { + unlock_page(vmpage); + put_page(vmpage); + /* page was truncated */ + break; + } + /* attach VM page to CL page cache */ + page = cl_page_find(env, obj, vmpage->index, vmpage, + CPT_CACHEABLE); + if (IS_ERR(page)) { + ClearPageUptodate(vmpage); + unlock_page(vmpage); + put_page(vmpage); + break; + } + cl_page_export(env, page, 1); + cl_page_put(env, page); + unlock_page(vmpage); put_page(vmpage); index++; } while (rnb->rnb_len > (index << PAGE_SHIFT)); + +out_io: + cl_io_fini(env, io); + cl_env_put(env, &refcheck); + EXIT; } @@ -510,12 +599,14 @@ static int ll_intent_file_open(struct dentry *de, void *lmm, int lmmsize, /* if server supports open-by-fid, or file name is invalid, don't pack * name in open request */ - if (!(exp_connect_flags(sbi->ll_md_exp) & OBD_CONNECT_OPEN_BY_FID)) { + if (OBD_FAIL_CHECK(OBD_FAIL_LLITE_OPEN_BY_NAME) || + !(exp_connect_flags(sbi->ll_md_exp) & OBD_CONNECT_OPEN_BY_FID)) { retry: len = de->d_name.len; - name = kmalloc(len, GFP_NOFS); + name = kmalloc(len + 1, GFP_NOFS); if (!name) RETURN(-ENOMEM); + /* race here */ spin_lock(&de->d_lock); if (len != de->d_name.len) { @@ -524,12 +615,12 @@ retry: goto retry; } memcpy(name, de->d_name.name, len); + name[len] = '\0'; spin_unlock(&de->d_lock); if (!lu_name_is_valid_2(name, len)) { kfree(name); - name = NULL; - len = 0; + RETURN(-ESTALE); } } @@ -569,8 +660,21 @@ retry: rc = ll_prep_inode(&de->d_inode, req, NULL, itp); if (!rc && itp->it_lock_mode) { - ll_dom_finish_open(de->d_inode, req, itp); - ll_set_lock_data(sbi->ll_md_exp, de->d_inode, itp, NULL); + __u64 bits = 0; + + /* If we got a lock back and it has a LOOKUP bit set, + * make sure the dentry is marked as valid so we can find it. + * We don't need to care about actual hashing since other bits + * of kernel will deal with that later. + */ + ll_set_lock_data(sbi->ll_md_exp, de->d_inode, itp, &bits); + if (bits & MDS_INODELOCK_LOOKUP) + d_lustre_revalidate(de); + /* if DoM bit returned along with LAYOUT bit then there + * can be read-on-open data returned. + */ + if (bits & MDS_INODELOCK_DOM && bits & MDS_INODELOCK_LAYOUT) + ll_dom_finish_open(de->d_inode, req); } out: @@ -611,7 +715,7 @@ static int ll_local_open(struct file *file, struct lookup_intent *it, struct inode *inode = file_inode(file); ENTRY; - LASSERT(!LUSTRE_FPRIVATE(file)); + LASSERT(!file->private_data); LASSERT(fd != NULL); @@ -623,7 +727,7 @@ static int ll_local_open(struct file *file, struct lookup_intent *it, RETURN(rc); } - LUSTRE_FPRIVATE(file) = fd; + file->private_data = fd; ll_readahead_init(inode, &fd->fd_ras); fd->fd_omode = it->it_flags & (FMODE_READ | FMODE_WRITE | FMODE_EXEC); @@ -655,6 +759,7 @@ int ll_file_open(struct inode *inode, struct file *file) struct obd_client_handle **och_p = NULL; __u64 *och_usecount = NULL; struct ll_file_data *fd; + ktime_t kstart = ktime_get(); int rc = 0; ENTRY; @@ -664,6 +769,12 @@ int ll_file_open(struct inode *inode, struct file *file) it = file->private_data; /* XXX: compat macro */ file->private_data = NULL; /* prevent ll_local_open assertion */ + if (S_ISREG(inode->i_mode)) { + rc = llcrypt_file_open(inode, file); + if (rc) + GOTO(out_nofiledata, rc); + } + fd = ll_file_data_get(); if (fd == NULL) GOTO(out_nofiledata, rc = -ENOMEM); @@ -673,18 +784,18 @@ int ll_file_open(struct inode *inode, struct file *file) ll_authorize_statahead(inode, fd); if (inode->i_sb->s_root == file_dentry(file)) { - LUSTRE_FPRIVATE(file) = fd; - RETURN(0); - } + file->private_data = fd; + RETURN(0); + } if (!it || !it->it_disposition) { - /* Convert f_flags into access mode. We cannot use file->f_mode, - * because everything but O_ACCMODE mask was stripped from - * there */ - if ((oit.it_flags + 1) & O_ACCMODE) - oit.it_flags++; - if (file->f_flags & O_TRUNC) - oit.it_flags |= FMODE_WRITE; + /* Convert f_flags into access mode. We cannot use file->f_mode, + * because everything but O_ACCMODE mask was stripped from + * there */ + if ((oit.it_flags + 1) & O_ACCMODE) + oit.it_flags++; + if (file->f_flags & O_TRUNC) + oit.it_flags |= FMODE_WRITE; /* kernel only call f_op->open in dentry_open. filp_open calls * dentry_open after call to open_namei that checks permissions. @@ -694,62 +805,65 @@ int ll_file_open(struct inode *inode, struct file *file) if (oit.it_flags & (FMODE_WRITE | FMODE_READ)) oit.it_flags |= MDS_OPEN_OWNEROVERRIDE; - /* We do not want O_EXCL here, presumably we opened the file - * already? XXX - NFS implications? */ - oit.it_flags &= ~O_EXCL; + /* We do not want O_EXCL here, presumably we opened the file + * already? XXX - NFS implications? */ + oit.it_flags &= ~O_EXCL; - /* bug20584, if "it_flags" contains O_CREAT, the file will be - * created if necessary, then "IT_CREAT" should be set to keep - * consistent with it */ - if (oit.it_flags & O_CREAT) - oit.it_op |= IT_CREAT; + /* bug20584, if "it_flags" contains O_CREAT, the file will be + * created if necessary, then "IT_CREAT" should be set to keep + * consistent with it */ + if (oit.it_flags & O_CREAT) + oit.it_op |= IT_CREAT; - it = &oit; - } + it = &oit; + } restart: - /* Let's see if we have file open on MDS already. */ - if (it->it_flags & FMODE_WRITE) { - och_p = &lli->lli_mds_write_och; - och_usecount = &lli->lli_open_fd_write_count; - } else if (it->it_flags & FMODE_EXEC) { - och_p = &lli->lli_mds_exec_och; - och_usecount = &lli->lli_open_fd_exec_count; - } else { - och_p = &lli->lli_mds_read_och; - och_usecount = &lli->lli_open_fd_read_count; - } + /* Let's see if we have file open on MDS already. */ + if (it->it_flags & FMODE_WRITE) { + och_p = &lli->lli_mds_write_och; + och_usecount = &lli->lli_open_fd_write_count; + } else if (it->it_flags & FMODE_EXEC) { + och_p = &lli->lli_mds_exec_och; + och_usecount = &lli->lli_open_fd_exec_count; + } else { + och_p = &lli->lli_mds_read_och; + och_usecount = &lli->lli_open_fd_read_count; + } mutex_lock(&lli->lli_och_mutex); - if (*och_p) { /* Open handle is present */ - if (it_disposition(it, DISP_OPEN_OPEN)) { - /* Well, there's extra open request that we do not need, - let's close it somehow. This will decref request. */ - rc = it_open_error(DISP_OPEN_OPEN, it); - if (rc) { + if (*och_p) { /* Open handle is present */ + if (it_disposition(it, DISP_OPEN_OPEN)) { + /* Well, there's extra open request that we do not need, + * let's close it somehow. This will decref request. */ + rc = it_open_error(DISP_OPEN_OPEN, it); + if (rc) { mutex_unlock(&lli->lli_och_mutex); - GOTO(out_openerr, rc); - } + GOTO(out_openerr, rc); + } ll_release_openhandle(file_dentry(file), it); - } - (*och_usecount)++; + } + (*och_usecount)++; - rc = ll_local_open(file, it, fd, NULL); - if (rc) { - (*och_usecount)--; + rc = ll_local_open(file, it, fd, NULL); + if (rc) { + (*och_usecount)--; mutex_unlock(&lli->lli_och_mutex); - GOTO(out_openerr, rc); - } - } else { - LASSERT(*och_usecount == 0); + GOTO(out_openerr, rc); + } + } else { + LASSERT(*och_usecount == 0); if (!it->it_disposition) { - struct ll_dentry_data *ldd = ll_d2d(file->f_path.dentry); - /* We cannot just request lock handle now, new ELC code - means that one of other OPEN locks for this file - could be cancelled, and since blocking ast handler - would attempt to grab och_mutex as well, that would - result in a deadlock */ + struct dentry *dentry = file_dentry(file); + struct ll_dentry_data *ldd; + + /* We cannot just request lock handle now, new ELC code + * means that one of other OPEN locks for this file + * could be cancelled, and since blocking ast handler + * would attempt to grab och_mutex as well, that would + * result in a deadlock + */ mutex_unlock(&lli->lli_och_mutex); /* * Normally called under two situations: @@ -766,34 +880,37 @@ restart: * lookup path only, since ll_iget_for_nfs always calls * ll_d_init(). */ + ldd = ll_d2d(dentry); if (ldd && ldd->lld_nfs_dentry) { ldd->lld_nfs_dentry = 0; - it->it_flags |= MDS_OPEN_LOCK; + if (!filename_is_volatile(dentry->d_name.name, + dentry->d_name.len, + NULL)) + it->it_flags |= MDS_OPEN_LOCK; } - /* + /* * Always specify MDS_OPEN_BY_FID because we don't want * to get file with different fid. */ it->it_flags |= MDS_OPEN_BY_FID; - rc = ll_intent_file_open(file_dentry(file), NULL, 0, - it); - if (rc) - GOTO(out_openerr, rc); + rc = ll_intent_file_open(dentry, NULL, 0, it); + if (rc) + GOTO(out_openerr, rc); - goto restart; - } - OBD_ALLOC(*och_p, sizeof (struct obd_client_handle)); - if (!*och_p) - GOTO(out_och_free, rc = -ENOMEM); + goto restart; + } + OBD_ALLOC(*och_p, sizeof(struct obd_client_handle)); + if (!*och_p) + GOTO(out_och_free, rc = -ENOMEM); - (*och_usecount)++; + (*och_usecount)++; - /* md_intent_lock() didn't get a request ref if there was an - * open error, so don't do cleanup on the request here - * (bug 3430) */ - /* XXX (green): Should not we bail out on any error here, not - * just open error? */ + /* md_intent_lock() didn't get a request ref if there was an + * open error, so don't do cleanup on the request here + * (bug 3430) */ + /* XXX (green): Should not we bail out on any error here, not + * just open error? */ rc = it_open_error(DISP_OPEN_OPEN, it); if (rc != 0) GOTO(out_och_free, rc); @@ -806,35 +923,45 @@ restart: if (rc) GOTO(out_och_free, rc); } + + rc = pcc_file_open(inode, file); + if (rc) + GOTO(out_och_free, rc); + mutex_unlock(&lli->lli_och_mutex); - fd = NULL; - /* Must do this outside lli_och_mutex lock to prevent deadlock where - different kind of OPEN lock for this same inode gets cancelled - by ldlm_cancel_lru */ - if (!S_ISREG(inode->i_mode)) - GOTO(out_och_free, rc); + /* lockless for direct IO so that it can do IO in parallel */ + if (file->f_flags & O_DIRECT) + fd->fd_flags |= LL_FILE_LOCKLESS_IO; + fd = NULL; + /* Must do this outside lli_och_mutex lock to prevent deadlock where + different kind of OPEN lock for this same inode gets cancelled + by ldlm_cancel_lru */ + if (!S_ISREG(inode->i_mode)) + GOTO(out_och_free, rc); cl_lov_delay_create_clear(&file->f_flags); GOTO(out_och_free, rc); out_och_free: - if (rc) { - if (och_p && *och_p) { - OBD_FREE(*och_p, sizeof (struct obd_client_handle)); - *och_p = NULL; /* OBD_FREE writes some magic there */ - (*och_usecount)--; - } + if (rc) { + if (och_p && *och_p) { + OBD_FREE(*och_p, sizeof(struct obd_client_handle)); + *och_p = NULL; /* OBD_FREE writes some magic there */ + (*och_usecount)--; + } mutex_unlock(&lli->lli_och_mutex); out_openerr: if (lli->lli_opendir_key == fd) ll_deauthorize_statahead(inode, fd); + if (fd != NULL) ll_file_data_put(fd); - } else { - ll_stats_ops_tally(ll_i2sbi(inode), LPROC_LL_OPEN, 1); - } + } else { + ll_stats_ops_tally(ll_i2sbi(inode), LPROC_LL_OPEN, + ktime_us_delta(ktime_get(), kstart)); + } out_nofiledata: if (it && it_disposition(it, DISP_ENQ_OPEN_REF)) { @@ -842,7 +969,7 @@ out_nofiledata: it_clear_disposition(it, DISP_ENQ_OPEN_REF); } - return rc; + return rc; } static int ll_md_blocking_lease_ast(struct ldlm_lock *lock, @@ -877,7 +1004,7 @@ static int ll_lease_och_acquire(struct inode *inode, struct file *file, struct lustre_handle *old_open_handle) { struct ll_inode_info *lli = ll_i2info(inode); - struct ll_file_data *fd = LUSTRE_FPRIVATE(file); + struct ll_file_data *fd = file->private_data; struct obd_client_handle **och_p; __u64 *och_usecount; int rc = 0; @@ -921,7 +1048,7 @@ out_unlock: static int ll_lease_och_release(struct inode *inode, struct file *file) { struct ll_inode_info *lli = ll_i2info(inode); - struct ll_file_data *fd = LUSTRE_FPRIVATE(file); + struct ll_file_data *fd = file->private_data; struct obd_client_handle **och_p; struct obd_client_handle *old_och = NULL; __u64 *och_usecount; @@ -1022,15 +1149,17 @@ ll_lease_open(struct inode *inode, struct file *file, fmode_t fmode, GOTO(out_release_it, rc); LASSERT(it_disposition(&it, DISP_ENQ_OPEN_REF)); - ll_och_fill(sbi->ll_md_exp, &it, och); + rc = ll_och_fill(sbi->ll_md_exp, &it, och); + if (rc) + GOTO(out_release_it, rc); if (!it_disposition(&it, DISP_OPEN_LEASE)) /* old server? */ GOTO(out_close, rc = -EOPNOTSUPP); /* already get lease, handle lease lock */ ll_set_lock_data(sbi->ll_md_exp, inode, &it, NULL); - if (it.it_lock_mode == 0 || - it.it_lock_bits != MDS_INODELOCK_OPEN) { + if (!it.it_lock_mode || + !(it.it_lock_bits & MDS_INODELOCK_OPEN)) { /* open lock must return for lease */ CERROR(DFID "lease granted but no open lock, %d/%llu.\n", PFID(ll_inode2fid(inode)), it.it_lock_mode, @@ -1052,8 +1181,7 @@ out_close: rc2 = ll_close_inode_openhandle(inode, och, 0, NULL); if (rc2 < 0) CERROR("%s: error closing file "DFID": %d\n", - ll_get_fsname(inode->i_sb, NULL, 0), - PFID(&ll_i2info(inode)->lli_fid), rc2); + sbi->ll_fsname, PFID(&ll_i2info(inode)->lli_fid), rc2); och = NULL; /* och has been freed in ll_close_inode_openhandle() */ out_release_it: ll_intent_release(&it); @@ -1097,7 +1225,7 @@ static int ll_swap_layouts_close(struct obd_client_handle *och, ENTRY; CDEBUG(D_INODE, "%s: biased close of file "DFID"\n", - ll_get_fsname(inode->i_sb, NULL, 0), PFID(fid1)); + ll_i2sbi(inode)->ll_fsname, PFID(fid1)); rc = ll_check_swap_layouts_validity(inode, inode2); if (rc < 0) @@ -1238,11 +1366,10 @@ int ll_merge_attr(const struct lu_env *env, struct inode *inode) * POSIX. Solving this problem needs to send an RPC to MDT for each * read, this will hurt performance. */ - if (inode->i_atime.tv_sec < lli->lli_atime || - lli->lli_update_atime) { + if (ll_file_test_and_clear_flag(lli, LLIF_UPDATE_ATIME) || + inode->i_atime.tv_sec < lli->lli_atime) inode->i_atime.tv_sec = lli->lli_atime; - lli->lli_update_atime = 0; - } + inode->i_mtime.tv_sec = lli->lli_mtime; inode->i_ctime.tv_sec = lli->lli_ctime; @@ -1293,7 +1420,7 @@ out_size_unlock: */ void ll_io_set_mirror(struct cl_io *io, const struct file *file) { - struct ll_file_data *fd = LUSTRE_FPRIVATE(file); + struct ll_file_data *fd = file->private_data; /* clear layout version for generic(non-resync) I/O in case it carries * stale layout version due to I/O restart */ @@ -1332,16 +1459,17 @@ static bool file_is_noatime(const struct file *file) if ((mnt->mnt_flags & MNT_NODIRATIME) && S_ISDIR(inode->i_mode)) return true; - if ((inode->i_sb->s_flags & MS_NODIRATIME) && S_ISDIR(inode->i_mode)) + if ((inode->i_sb->s_flags & SB_NODIRATIME) && S_ISDIR(inode->i_mode)) return true; return false; } -static void ll_io_init(struct cl_io *io, struct file *file, enum cl_io_type iot) +void ll_io_init(struct cl_io *io, struct file *file, enum cl_io_type iot, + struct vvp_io_args *args) { struct inode *inode = file_inode(file); - struct ll_file_data *fd = LUSTRE_FPRIVATE(file); + struct ll_file_data *fd = file->private_data; io->u.ci_rw.crw_nonblock = file->f_flags & O_NONBLOCK; io->ci_lock_no_expand = fd->ll_lock_no_expand; @@ -1351,7 +1479,13 @@ static void ll_io_init(struct cl_io *io, struct file *file, enum cl_io_type iot) io->u.ci_wr.wr_sync = !!(file->f_flags & O_SYNC || file->f_flags & O_DIRECT || IS_SYNC(inode)); +#ifdef HAVE_GENERIC_WRITE_SYNC_2ARGS + io->u.ci_wr.wr_sync |= !!(args && + (args->u.normal.via_iocb->ki_flags & + IOCB_DSYNC)); +#endif } + io->ci_obj = ll_i2info(inode)->lli_clob; io->ci_lockreq = CILR_MAYBE; if (ll_file_nolock(file)) { @@ -1361,6 +1495,7 @@ static void ll_io_init(struct cl_io *io, struct file *file, enum cl_io_type iot) io->ci_lockreq = CILR_MANDATORY; } io->ci_noatime = file_is_noatime(file); + io->ci_async_readahead = false; /* FLR: only use non-delay I/O for read as there is only one * avaliable mirror for write. */ @@ -1405,16 +1540,17 @@ ll_file_io_generic(const struct lu_env *env, struct vvp_io_args *args, struct file *file, enum cl_io_type iot, loff_t *ppos, size_t count) { - struct vvp_io *vio = vvp_env_io(env); - struct inode *inode = file_inode(file); - struct ll_inode_info *lli = ll_i2info(inode); - struct ll_file_data *fd = LUSTRE_FPRIVATE(file); - struct range_lock range; - struct cl_io *io; - ssize_t result = 0; - int rc = 0; - unsigned retried = 0; - bool restarted = false; + struct vvp_io *vio = vvp_env_io(env); + struct inode *inode = file_inode(file); + struct ll_inode_info *lli = ll_i2info(inode); + struct ll_file_data *fd = file->private_data; + struct range_lock range; + struct cl_io *io; + ssize_t result = 0; + int rc = 0; + unsigned int retried = 0, ignore_lockless = 0; + bool is_aio = false; + struct cl_dio_aio *ci_aio = NULL; ENTRY; @@ -1422,9 +1558,20 @@ ll_file_io_generic(const struct lu_env *env, struct vvp_io_args *args, file_dentry(file)->d_name.name, iot == CIT_READ ? "read" : "write", *ppos, count); + io = vvp_env_thread_io(env); + if (file->f_flags & O_DIRECT) { + if (!is_sync_kiocb(args->u.normal.via_iocb)) + is_aio = true; + ci_aio = cl_aio_alloc(args->u.normal.via_iocb); + if (!ci_aio) + GOTO(out, rc = -ENOMEM); + } + restart: io = vvp_env_thread_io(env); - ll_io_init(io, file, iot); + ll_io_init(io, file, iot, args); + io->ci_aio = ci_aio; + io->ci_ignore_lockless = ignore_lockless; io->ci_ndelay_tried = retried; if (cl_io_rw_init(env, io, iot, *ppos, count) == 0) { @@ -1435,35 +1582,23 @@ restart: else range_lock_init(&range, *ppos, *ppos + count - 1); - vio->vui_fd = LUSTRE_FPRIVATE(file); - vio->vui_io_subtype = args->via_io_subtype; - - switch (vio->vui_io_subtype) { - case IO_NORMAL: - vio->vui_iter = args->u.normal.via_iter; - vio->vui_iocb = args->u.normal.via_iocb; - /* Direct IO reads must also take range lock, - * or multiple reads will try to work on the same pages - * See LU-6227 for details. */ - if (((iot == CIT_WRITE) || - (iot == CIT_READ && (file->f_flags & O_DIRECT))) && - !(vio->vui_fd->fd_flags & LL_FILE_GROUP_LOCKED)) { - CDEBUG(D_VFSTRACE, "Range lock "RL_FMT"\n", - RL_PARA(&range)); - rc = range_lock(&lli->lli_write_tree, &range); - if (rc < 0) - GOTO(out, rc); + vio->vui_fd = file->private_data; + vio->vui_iter = args->u.normal.via_iter; + vio->vui_iocb = args->u.normal.via_iocb; + /* Direct IO reads must also take range lock, + * or multiple reads will try to work on the same pages + * See LU-6227 for details. + */ + if (((iot == CIT_WRITE) || + (iot == CIT_READ && (file->f_flags & O_DIRECT))) && + !(vio->vui_fd->fd_flags & LL_FILE_GROUP_LOCKED)) { + CDEBUG(D_VFSTRACE, "Range lock "RL_FMT"\n", + RL_PARA(&range)); + rc = range_lock(&lli->lli_write_tree, &range); + if (rc < 0) + GOTO(out, rc); - range_locked = true; - } - break; - case IO_SPLICE: - vio->u.splice.vui_pipe = args->u.splice.via_pipe; - vio->u.splice.vui_flags = args->u.splice.via_flags; - break; - default: - CERROR("unknown IO subtype %u\n", vio->vui_io_subtype); - LBUG(); + range_locked = true; } ll_cl_add(file, env, io, LCC_RW); @@ -1480,13 +1615,22 @@ restart: rc = io->ci_result; } + /* + * In order to move forward AIO, ci_nob was increased, + * but that doesn't mean io have been finished, it just + * means io have been submited, we will always return + * EIOCBQUEUED to the caller, So we could only return + * number of bytes in non-AIO case. + */ if (io->ci_nob > 0) { - result += io->ci_nob; - count -= io->ci_nob; - *ppos = io->u.ci_wr.wr.crw_pos; /* for splice */ + if (!is_aio) { + result += io->ci_nob; + *ppos = io->u.ci_wr.wr.crw_pos; /* for splice */ + } + count -= io->ci_nob; /* prepare IO restart */ - if (count > 0 && args->via_io_subtype == IO_NORMAL) + if (count > 0) args->u.normal.via_iter = vio->vui_iter; } out: @@ -1497,7 +1641,8 @@ out: file->f_path.dentry->d_name.name, iot, rc, result, io->ci_need_restart); - if ((rc == 0 || rc == -ENODATA) && count > 0 && io->ci_need_restart) { + if ((rc == 0 || rc == -ENODATA || rc == -ENOLCK) && + count > 0 && io->ci_need_restart) { CDEBUG(D_VFSTRACE, "%s: restart %s from %lld, count: %zu, ret: %zd, rc: %d\n", file_dentry(file)->d_name.name, @@ -1505,10 +1650,31 @@ out: *ppos, count, result, rc); /* preserve the tried count for FLR */ retried = io->ci_ndelay_tried; - restarted = true; + ignore_lockless = io->ci_ignore_lockless; goto restart; } + if (io->ci_aio) { + /* + * VFS will call aio_complete() if no -EIOCBQUEUED + * is returned for AIO, so we can not call aio_complete() + * in our end_io(). + */ + if (rc != -EIOCBQUEUED) + io->ci_aio->cda_no_aio_complete = 1; + /** + * Drop one extra reference so that end_io() could be + * called for this IO context, we could call it after + * we make sure all AIO requests have been proceed. + */ + cl_sync_io_note(env, &io->ci_aio->cda_sync, + rc == -EIOCBQUEUED ? 0 : rc); + if (!is_aio) { + cl_aio_free(io->ci_aio); + io->ci_aio = NULL; + } + } + if (iot == CIT_READ) { if (result > 0) ll_stats_ops_tally(ll_i2sbi(inode), @@ -1595,7 +1761,7 @@ ll_do_fast_read(struct kiocb *iocb, struct iov_iter *iter) if (result > 0) { ll_heat_add(file_inode(iocb->ki_filp), CIT_READ, result); ll_stats_ops_tally(ll_i2sbi(file_inode(iocb->ki_filp)), - LPROC_LL_READ_BYTES, result); + LPROC_LL_READ_BYTES, result); } return result; @@ -1608,9 +1774,32 @@ static ssize_t ll_file_read_iter(struct kiocb *iocb, struct iov_iter *to) { struct lu_env *env; struct vvp_io_args *args; + struct file *file = iocb->ki_filp; ssize_t result; ssize_t rc2; __u16 refcheck; + ktime_t kstart = ktime_get(); + bool cached; + + if (!iov_iter_count(to)) + return 0; + + /** + * Currently when PCC read failed, we do not fall back to the + * normal read path, just return the error. + * The resaon is that: for RW-PCC, the file data may be modified + * in the PCC and inconsistent with the data on OSTs (or file + * data has been removed from the Lustre file system), at this + * time, fallback to the normal read path may read the wrong + * data. + * TODO: for RO-PCC (readonly PCC), fall back to normal read + * path: read data from data copy on OSTs. + */ + result = pcc_file_read_iter(iocb, to, &cached); + if (cached) + GOTO(out, result); + + ll_ras_enter(file, iocb->ki_pos, iov_iter_count(to)); result = ll_do_fast_read(iocb, to); if (result < 0 || iov_iter_count(to) == 0) @@ -1620,11 +1809,11 @@ static ssize_t ll_file_read_iter(struct kiocb *iocb, struct iov_iter *to) if (IS_ERR(env)) return PTR_ERR(env); - args = ll_env_args(env, IO_NORMAL); + args = ll_env_args(env); args->u.normal.via_iter = to; args->u.normal.via_iocb = iocb; - rc2 = ll_file_io_generic(env, args, iocb->ki_filp, CIT_READ, + rc2 = ll_file_io_generic(env, args, file, CIT_READ, &iocb->ki_pos, iov_iter_count(to)); if (rc2 > 0) result += rc2; @@ -1633,6 +1822,14 @@ static ssize_t ll_file_read_iter(struct kiocb *iocb, struct iov_iter *to) cl_env_put(env, &refcheck); out: + if (result > 0) { + ll_rw_stats_tally(ll_i2sbi(file_inode(file)), current->pid, + file->private_data, iocb->ki_pos, result, + READ); + ll_stats_ops_tally(ll_i2sbi(file_inode(file)), LPROC_LL_READ, + ktime_us_delta(ktime_get(), kstart)); + } + return result; } @@ -1703,17 +1900,38 @@ static ssize_t ll_file_write_iter(struct kiocb *iocb, struct iov_iter *from) struct vvp_io_args *args; struct lu_env *env; ssize_t rc_tiny = 0, rc_normal; + struct file *file = iocb->ki_filp; __u16 refcheck; + bool cached; + ktime_t kstart = ktime_get(); + int result; ENTRY; + if (!iov_iter_count(from)) + GOTO(out, rc_normal = 0); + + /** + * When PCC write failed, we usually do not fall back to the normal + * write path, just return the error. But there is a special case when + * returned error code is -ENOSPC due to running out of space on PCC HSM + * bakcend. At this time, it will fall back to normal I/O path and + * retry the I/O. As the file is in HSM released state, it will restore + * the file data to OSTs first and redo the write again. And the + * restore process will revoke the layout lock and detach the file + * from PCC cache automatically. + */ + result = pcc_file_write_iter(iocb, from, &cached); + if (cached && result != -ENOSPC && result != -EDQUOT) + GOTO(out, rc_normal = result); + /* NB: we can't do direct IO for tiny writes because they use the page * cache, we can't do sync writes because tiny writes can't flush * pages, and we can't do append writes because we can't guarantee the * required DLM locks are held to protect file size. */ - if (ll_sbi_has_tiny_write(ll_i2sbi(file_inode(iocb->ki_filp))) && - !(iocb->ki_filp->f_flags & (O_DIRECT | O_SYNC | O_APPEND))) + if (ll_sbi_has_tiny_write(ll_i2sbi(file_inode(file))) && + !(file->f_flags & (O_DIRECT | O_SYNC | O_APPEND))) rc_tiny = ll_do_tiny_write(iocb, from); /* In case of error, go on and try normal write - Only stop if tiny @@ -1726,12 +1944,12 @@ static ssize_t ll_file_write_iter(struct kiocb *iocb, struct iov_iter *from) if (IS_ERR(env)) return PTR_ERR(env); - args = ll_env_args(env, IO_NORMAL); + args = ll_env_args(env); args->u.normal.via_iter = from; args->u.normal.via_iocb = iocb; - rc_normal = ll_file_io_generic(env, args, iocb->ki_filp, CIT_WRITE, - &iocb->ki_pos, iov_iter_count(from)); + rc_normal = ll_file_io_generic(env, args, file, CIT_WRITE, + &iocb->ki_pos, iov_iter_count(from)); /* On success, combine bytes written. */ if (rc_tiny >= 0 && rc_normal > 0) @@ -1744,6 +1962,14 @@ static ssize_t ll_file_write_iter(struct kiocb *iocb, struct iov_iter *from) cl_env_put(env, &refcheck); out: + if (rc_normal > 0) { + ll_rw_stats_tally(ll_i2sbi(file_inode(file)), current->pid, + file->private_data, iocb->ki_pos, + rc_normal, WRITE); + ll_stats_ops_tally(ll_i2sbi(file_inode(file)), LPROC_LL_WRITE, + ktime_us_delta(ktime_get(), kstart)); + } + RETURN(rc_normal); } @@ -1752,7 +1978,8 @@ out: * XXX: exact copy from kernel code (__generic_file_aio_write_nolock) */ static int ll_file_get_iov_count(const struct iovec *iov, - unsigned long *nr_segs, size_t *count) + unsigned long *nr_segs, size_t *count, + int access_flags) { size_t cnt = 0; unsigned long seg; @@ -1767,7 +1994,7 @@ static int ll_file_get_iov_count(const struct iovec *iov, cnt += iv->iov_len; if (unlikely((ssize_t)(cnt|iv->iov_len) < 0)) return -EINVAL; - if (access_ok(VERIFY_READ, iv->iov_base, iv->iov_len)) + if (access_ok(access_flags, iv->iov_base, iv->iov_len)) continue; if (seg == 0) return -EFAULT; @@ -1787,10 +2014,13 @@ static ssize_t ll_file_aio_read(struct kiocb *iocb, const struct iovec *iov, ssize_t result; ENTRY; - result = ll_file_get_iov_count(iov, &nr_segs, &iov_count); + result = ll_file_get_iov_count(iov, &nr_segs, &iov_count, VERIFY_READ); if (result) RETURN(result); + if (!iov_count) + RETURN(0); + # ifdef HAVE_IOV_ITER_INIT_DIRECTION iov_iter_init(&to, READ, iov, nr_segs, iov_count); # else /* !HAVE_IOV_ITER_INIT_DIRECTION */ @@ -1808,8 +2038,12 @@ static ssize_t ll_file_read(struct file *file, char __user *buf, size_t count, struct iovec iov = { .iov_base = buf, .iov_len = count }; struct kiocb kiocb; ssize_t result; + ENTRY; + if (!count) + RETURN(0); + init_sync_kiocb(&kiocb, file); kiocb.ki_pos = *ppos; #ifdef HAVE_KIOCB_KI_LEFT @@ -1836,10 +2070,13 @@ static ssize_t ll_file_aio_write(struct kiocb *iocb, const struct iovec *iov, ssize_t result; ENTRY; - result = ll_file_get_iov_count(iov, &nr_segs, &iov_count); + result = ll_file_get_iov_count(iov, &nr_segs, &iov_count, VERIFY_WRITE); if (result) RETURN(result); + if (!iov_count) + RETURN(0); + # ifdef HAVE_IOV_ITER_INIT_DIRECTION iov_iter_init(&from, WRITE, iov, nr_segs, iov_count); # else /* !HAVE_IOV_ITER_INIT_DIRECTION */ @@ -1861,6 +2098,9 @@ static ssize_t ll_file_write(struct file *file, const char __user *buf, ENTRY; + if (!count) + RETURN(0); + init_sync_kiocb(&kiocb, file); kiocb.ki_pos = *ppos; #ifdef HAVE_KIOCB_KI_LEFT @@ -1876,32 +2116,6 @@ static ssize_t ll_file_write(struct file *file, const char __user *buf, } #endif /* !HAVE_FILE_OPERATIONS_READ_WRITE_ITER */ -/* - * Send file content (through pagecache) somewhere with helper - */ -static ssize_t ll_file_splice_read(struct file *in_file, loff_t *ppos, - struct pipe_inode_info *pipe, size_t count, - unsigned int flags) -{ - struct lu_env *env; - struct vvp_io_args *args; - ssize_t result; - __u16 refcheck; - ENTRY; - - env = cl_env_get(&refcheck); - if (IS_ERR(env)) - RETURN(PTR_ERR(env)); - - args = ll_env_args(env, IO_SPLICE); - args->u.splice.via_pipe = pipe; - args->u.splice.via_flags = flags; - - result = ll_file_io_generic(env, args, in_file, CIT_READ, ppos, count); - cl_env_put(env, &refcheck); - RETURN(result); -} - int ll_lov_setstripe_ea_info(struct inode *inode, struct dentry *dentry, __u64 flags, struct lov_user_md *lum, int lum_size) { @@ -1912,6 +2126,12 @@ int ll_lov_setstripe_ea_info(struct inode *inode, struct dentry *dentry, int rc; ENTRY; + if ((__swab32(lum->lmm_magic) & le32_to_cpu(LOV_MAGIC_MASK)) == + le32_to_cpu(LOV_MAGIC_MAGIC)) { + /* this code will only exist for big-endian systems */ + lustre_swab_lov_user_md(lum, 0); + } + ll_inode_size_lock(inode); rc = ll_intent_file_open(dentry, lum, lum_size, &oit); if (rc < 0) @@ -1971,16 +2191,18 @@ int ll_lov_getstripe_ea_info(struct inode *inode, const char *filename, if (lmm->lmm_magic != cpu_to_le32(LOV_MAGIC_V1) && lmm->lmm_magic != cpu_to_le32(LOV_MAGIC_V3) && - lmm->lmm_magic != cpu_to_le32(LOV_MAGIC_COMP_V1)) + lmm->lmm_magic != cpu_to_le32(LOV_MAGIC_COMP_V1) && + lmm->lmm_magic != cpu_to_le32(LOV_MAGIC_FOREIGN)) GOTO(out, rc = -EPROTO); - /* - * This is coming from the MDS, so is probably in - * little endian. We convert it to host endian before - * passing it to userspace. - */ - if (LOV_MAGIC != cpu_to_le32(LOV_MAGIC)) { - int stripe_count; + /* + * This is coming from the MDS, so is probably in + * little endian. We convert it to host endian before + * passing it to userspace. + */ + if ((lmm->lmm_magic & __swab32(LOV_MAGIC_MAGIC)) == + __swab32(LOV_MAGIC_MAGIC)) { + int stripe_count = 0; if (lmm->lmm_magic == cpu_to_le32(LOV_MAGIC_V1) || lmm->lmm_magic == cpu_to_le32(LOV_MAGIC_V3)) { @@ -1990,27 +2212,19 @@ int ll_lov_getstripe_ea_info(struct inode *inode, const char *filename, stripe_count = 0; } - /* if function called for directory - we should - * avoid swab not existent lsm objects */ - if (lmm->lmm_magic == cpu_to_le32(LOV_MAGIC_V1)) { - lustre_swab_lov_user_md_v1( - (struct lov_user_md_v1 *)lmm); - if (S_ISREG(body->mbo_mode)) - lustre_swab_lov_user_md_objects( - ((struct lov_user_md_v1 *)lmm)->lmm_objects, - stripe_count); - } else if (lmm->lmm_magic == cpu_to_le32(LOV_MAGIC_V3)) { - lustre_swab_lov_user_md_v3( - (struct lov_user_md_v3 *)lmm); - if (S_ISREG(body->mbo_mode)) - lustre_swab_lov_user_md_objects( - ((struct lov_user_md_v3 *)lmm)->lmm_objects, - stripe_count); - } else if (lmm->lmm_magic == - cpu_to_le32(LOV_MAGIC_COMP_V1)) { - lustre_swab_lov_comp_md_v1( - (struct lov_comp_md_v1 *)lmm); - } + lustre_swab_lov_user_md((struct lov_user_md *)lmm, 0); + + /* if function called for directory - we should + * avoid swab not existent lsm objects */ + if (lmm->lmm_magic == LOV_MAGIC_V1 && S_ISREG(body->mbo_mode)) + lustre_swab_lov_user_md_objects( + ((struct lov_user_md_v1 *)lmm)->lmm_objects, + stripe_count); + else if (lmm->lmm_magic == LOV_MAGIC_V3 && + S_ISREG(body->mbo_mode)) + lustre_swab_lov_user_md_objects( + ((struct lov_user_md_v3 *)lmm)->lmm_objects, + stripe_count); } out: @@ -2093,20 +2307,28 @@ static int ll_lov_setstripe(struct inode *inode, struct file *file, GOTO(out, rc); rc = ll_file_getstripe(inode, arg, lum_size); + if (S_ISREG(inode->i_mode) && IS_ENCRYPTED(inode) && + ll_i2info(inode)->lli_clob) { + struct iattr attr = { 0 }; + + rc = cl_setattr_ost(ll_i2info(inode)->lli_clob, &attr, + OP_XVALID_FLAGS, LUSTRE_ENCRYPT_FL); + } } cl_lov_delay_create_clear(&file->f_flags); out: - OBD_FREE(klum, lum_size); + OBD_FREE_LARGE(klum, lum_size); RETURN(rc); } + static int ll_get_grouplock(struct inode *inode, struct file *file, unsigned long arg) { struct ll_inode_info *lli = ll_i2info(inode); struct cl_object *obj = lli->lli_clob; - struct ll_file_data *fd = LUSTRE_FPRIVATE(file); + struct ll_file_data *fd = file->private_data; struct ll_grouplock grouplock; int rc; ENTRY; @@ -2116,18 +2338,28 @@ ll_get_grouplock(struct inode *inode, struct file *file, unsigned long arg) RETURN(-EINVAL); } - if (ll_file_nolock(file)) - RETURN(-EOPNOTSUPP); + if (ll_file_nolock(file)) + RETURN(-EOPNOTSUPP); +retry: + if (file->f_flags & O_NONBLOCK) { + if (!mutex_trylock(&lli->lli_group_mutex)) + RETURN(-EAGAIN); + } else + mutex_lock(&lli->lli_group_mutex); - spin_lock(&lli->lli_lock); if (fd->fd_flags & LL_FILE_GROUP_LOCKED) { CWARN("group lock already existed with gid %lu\n", fd->fd_grouplock.lg_gid); - spin_unlock(&lli->lli_lock); - RETURN(-EINVAL); + GOTO(out, rc = -EINVAL); + } + if (arg != lli->lli_group_gid && lli->lli_group_users != 0) { + if (file->f_flags & O_NONBLOCK) + GOTO(out, rc = -EAGAIN); + mutex_unlock(&lli->lli_group_mutex); + wait_var_event(&lli->lli_group_users, !lli->lli_group_users); + GOTO(retry, rc = 0); } LASSERT(fd->fd_grouplock.lg_lock == NULL); - spin_unlock(&lli->lli_lock); /** * XXX: group lock needs to protect all OST objects while PFL @@ -2147,7 +2379,7 @@ ll_get_grouplock(struct inode *inode, struct file *file, unsigned long arg) env = cl_env_get(&refcheck); if (IS_ERR(env)) - RETURN(PTR_ERR(env)); + GOTO(out, rc = PTR_ERR(env)); rc = cl_object_layout_get(env, obj, &cl); if (!rc && cl.cl_is_composite) @@ -2156,62 +2388,68 @@ ll_get_grouplock(struct inode *inode, struct file *file, unsigned long arg) cl_env_put(env, &refcheck); if (rc) - RETURN(rc); + GOTO(out, rc); } rc = cl_get_grouplock(ll_i2info(inode)->lli_clob, arg, (file->f_flags & O_NONBLOCK), &grouplock); - if (rc) - RETURN(rc); - spin_lock(&lli->lli_lock); - if (fd->fd_flags & LL_FILE_GROUP_LOCKED) { - spin_unlock(&lli->lli_lock); - CERROR("another thread just won the race\n"); - cl_put_grouplock(&grouplock); - RETURN(-EINVAL); - } + if (rc) + GOTO(out, rc); fd->fd_flags |= LL_FILE_GROUP_LOCKED; fd->fd_grouplock = grouplock; - spin_unlock(&lli->lli_lock); + if (lli->lli_group_users == 0) + lli->lli_group_gid = grouplock.lg_gid; + lli->lli_group_users++; CDEBUG(D_INFO, "group lock %lu obtained\n", arg); - RETURN(0); +out: + mutex_unlock(&lli->lli_group_mutex); + + RETURN(rc); } static int ll_put_grouplock(struct inode *inode, struct file *file, unsigned long arg) { struct ll_inode_info *lli = ll_i2info(inode); - struct ll_file_data *fd = LUSTRE_FPRIVATE(file); + struct ll_file_data *fd = file->private_data; struct ll_grouplock grouplock; + int rc; ENTRY; - spin_lock(&lli->lli_lock); + mutex_lock(&lli->lli_group_mutex); if (!(fd->fd_flags & LL_FILE_GROUP_LOCKED)) { - spin_unlock(&lli->lli_lock); - CWARN("no group lock held\n"); - RETURN(-EINVAL); - } + CWARN("no group lock held\n"); + GOTO(out, rc = -EINVAL); + } LASSERT(fd->fd_grouplock.lg_lock != NULL); if (fd->fd_grouplock.lg_gid != arg) { CWARN("group lock %lu doesn't match current id %lu\n", arg, fd->fd_grouplock.lg_gid); - spin_unlock(&lli->lli_lock); - RETURN(-EINVAL); + GOTO(out, rc = -EINVAL); } grouplock = fd->fd_grouplock; memset(&fd->fd_grouplock, 0, sizeof(fd->fd_grouplock)); fd->fd_flags &= ~LL_FILE_GROUP_LOCKED; - spin_unlock(&lli->lli_lock); cl_put_grouplock(&grouplock); + + lli->lli_group_users--; + if (lli->lli_group_users == 0) { + lli->lli_group_gid = 0; + wake_up_var(&lli->lli_group_users); + } CDEBUG(D_INFO, "group lock %lu released\n", arg); - RETURN(0); + GOTO(out, rc = 0); +out: + mutex_unlock(&lli->lli_group_mutex); + + RETURN(rc); } /** @@ -2246,7 +2484,9 @@ int ll_release_openhandle(struct dentry *dentry, struct lookup_intent *it) if (!och) GOTO(out, rc = -ENOMEM); - ll_och_fill(ll_i2sbi(inode)->ll_md_exp, it, och); + rc = ll_och_fill(ll_i2sbi(inode)->ll_md_exp, it, och); + if (rc) + GOTO(out, rc); rc = ll_close_inode_openhandle(inode, och, 0, NULL); out: @@ -2445,7 +2685,7 @@ int ll_hsm_release(struct inode *inode) ENTRY; CDEBUG(D_INODE, "%s: Releasing file "DFID".\n", - ll_get_fsname(inode->i_sb, NULL, 0), + ll_i2sbi(inode)->ll_fsname, PFID(&ll_i2info(inode)->lli_fid)); och = ll_lease_open(inode, NULL, FMODE_WRITE, MDS_OPEN_RELEASE); @@ -2785,9 +3025,9 @@ int ll_file_lock_ahead(struct file *file, struct llapi_lu_ladvise *ladvise) ENTRY; - CDEBUG(D_VFSTRACE, "Lock request: file=%.*s, inode=%p, mode=%s " - "start=%llu, end=%llu\n", dentry->d_name.len, - dentry->d_name.name, dentry->d_inode, + CDEBUG(D_VFSTRACE, + "Lock request: file=%pd, inode=%p, mode=%s start=%llu, end=%llu\n", + dentry, dentry->d_inode, user_lockname[ladvise->lla_lockahead_mode], (__u64) start, (__u64) end); @@ -2853,6 +3093,7 @@ static const char *const ladvise_names[] = LU_LADVISE_NAMES; static int ll_ladvise_sanity(struct inode *inode, struct llapi_lu_ladvise *ladvise) { + struct ll_sb_info *sbi = ll_i2sbi(inode); enum lu_ladvise_type advice = ladvise->lla_advice; /* Note the peradvice flags is a 32 bit field, so per advice flags must * be in the first 32 bits of enum ladvise_flags */ @@ -2862,9 +3103,9 @@ static int ll_ladvise_sanity(struct inode *inode, if (advice > LU_LADVISE_MAX || advice == LU_LADVISE_INVALID) { rc = -EINVAL; - CDEBUG(D_VFSTRACE, "%s: advice with value '%d' not recognized," - "last supported advice is %s (value '%d'): rc = %d\n", - ll_get_fsname(inode->i_sb, NULL, 0), advice, + CDEBUG(D_VFSTRACE, + "%s: advice with value '%d' not recognized, last supported advice is %s (value '%d'): rc = %d\n", + sbi->ll_fsname, advice, ladvise_names[LU_LADVISE_MAX-1], LU_LADVISE_MAX-1, rc); GOTO(out, rc); } @@ -2875,8 +3116,7 @@ static int ll_ladvise_sanity(struct inode *inode, if (flags & ~LF_LOCKNOEXPAND_MASK) { rc = -EINVAL; CDEBUG(D_VFSTRACE, "%s: Invalid flags (%x) for %s: " - "rc = %d\n", - ll_get_fsname(inode->i_sb, NULL, 0), flags, + "rc = %d\n", sbi->ll_fsname, flags, ladvise_names[advice], rc); GOTO(out, rc); } @@ -2887,12 +3127,12 @@ static int ll_ladvise_sanity(struct inode *inode, ladvise->lla_lockahead_mode == 0) { rc = -EINVAL; CDEBUG(D_VFSTRACE, "%s: Invalid mode (%d) for %s: " - "rc = %d\n", - ll_get_fsname(inode->i_sb, NULL, 0), + "rc = %d\n", sbi->ll_fsname, ladvise->lla_lockahead_mode, ladvise_names[advice], rc); GOTO(out, rc); } + /* fallthrough */ case LU_LADVISE_WILLREAD: case LU_LADVISE_DONTNEED: default: @@ -2901,16 +3141,14 @@ static int ll_ladvise_sanity(struct inode *inode, if (flags & ~LF_DEFAULT_MASK) { rc = -EINVAL; CDEBUG(D_VFSTRACE, "%s: Invalid flags (%x) for %s: " - "rc = %d\n", - ll_get_fsname(inode->i_sb, NULL, 0), flags, + "rc = %d\n", sbi->ll_fsname, flags, ladvise_names[advice], rc); GOTO(out, rc); } if (ladvise->lla_start >= ladvise->lla_end) { rc = -EINVAL; CDEBUG(D_VFSTRACE, "%s: Invalid range (%llu to %llu) " - "for %s: rc = %d\n", - ll_get_fsname(inode->i_sb, NULL, 0), + "for %s: rc = %d\n", sbi->ll_fsname, ladvise->lla_start, ladvise->lla_end, ladvise_names[advice], rc); GOTO(out, rc); @@ -2974,7 +3212,7 @@ static int ll_ladvise(struct inode *inode, struct file *file, __u64 flags, static int ll_lock_noexpand(struct file *file, int flags) { - struct ll_file_data *fd = LUSTRE_FPRIVATE(file); + struct ll_file_data *fd = file->private_data; fd->ll_lock_no_expand = !(flags & LF_UNSET); @@ -3068,6 +3306,11 @@ int ll_ioctl_fssetxattr(struct inode *inode, unsigned int cmd, if (obj == NULL) GOTO(out_fsxattr, rc); + /* Avoiding OST RPC if this is only project ioctl */ + if (fsxattr.fsx_xflags == 0 || + fsxattr.fsx_xflags == FS_XFLAG_PROJINHERIT) + GOTO(out_fsxattr, rc); + OBD_ALLOC_PTR(attr); if (attr == NULL) GOTO(out_fsxattr, rc = -ENOMEM); @@ -3084,17 +3327,20 @@ static long ll_file_unlock_lease(struct file *file, struct ll_ioc_lease *ioc, unsigned long arg) { struct inode *inode = file_inode(file); - struct ll_file_data *fd = LUSTRE_FPRIVATE(file); + struct ll_file_data *fd = file->private_data; struct ll_inode_info *lli = ll_i2info(inode); struct obd_client_handle *och = NULL; struct split_param sp; - bool lease_broken; + struct pcc_param param; + bool lease_broken = false; fmode_t fmode = 0; enum mds_op_bias bias = 0; struct file *layout_file = NULL; void *data = NULL; size_t data_size = 0; - long rc; + bool attached = false; + long rc, rc2 = 0; + ENTRY; mutex_lock(&lli->lli_och_mutex); @@ -3105,22 +3351,22 @@ static long ll_file_unlock_lease(struct file *file, struct ll_ioc_lease *ioc, mutex_unlock(&lli->lli_och_mutex); if (och == NULL) - GOTO(out, rc = -ENOLCK); + RETURN(-ENOLCK); fmode = och->och_flags; switch (ioc->lil_flags) { case LL_LEASE_RESYNC_DONE: if (ioc->lil_count > IOC_IDS_MAX) - GOTO(out, rc = -EINVAL); + GOTO(out_lease_close, rc = -EINVAL); data_size = offsetof(typeof(*ioc), lil_ids[ioc->lil_count]); OBD_ALLOC(data, data_size); if (!data) - GOTO(out, rc = -ENOMEM); + GOTO(out_lease_close, rc = -ENOMEM); if (copy_from_user(data, (void __user *)arg, data_size)) - GOTO(out, rc = -EFAULT); + GOTO(out_lease_close, rc = -EFAULT); bias = MDS_CLOSE_RESYNC_DONE; break; @@ -3128,19 +3374,19 @@ static long ll_file_unlock_lease(struct file *file, struct ll_ioc_lease *ioc, int fd; if (ioc->lil_count != 1) - GOTO(out, rc = -EINVAL); + GOTO(out_lease_close, rc = -EINVAL); arg += sizeof(*ioc); if (copy_from_user(&fd, (void __user *)arg, sizeof(__u32))) - GOTO(out, rc = -EFAULT); + GOTO(out_lease_close, rc = -EFAULT); layout_file = fget(fd); if (!layout_file) - GOTO(out, rc = -EBADF); + GOTO(out_lease_close, rc = -EBADF); if ((file->f_flags & O_ACCMODE) == O_RDONLY || (layout_file->f_flags & O_ACCMODE) == O_RDONLY) - GOTO(out, rc = -EPERM); + GOTO(out_lease_close, rc = -EPERM); data = file_inode(layout_file); bias = MDS_CLOSE_LAYOUT_MERGE; @@ -3151,20 +3397,20 @@ static long ll_file_unlock_lease(struct file *file, struct ll_ioc_lease *ioc, int mirror_id; if (ioc->lil_count != 2) - GOTO(out, rc = -EINVAL); + GOTO(out_lease_close, rc = -EINVAL); arg += sizeof(*ioc); if (copy_from_user(&fdv, (void __user *)arg, sizeof(__u32))) - GOTO(out, rc = -EFAULT); + GOTO(out_lease_close, rc = -EFAULT); arg += sizeof(__u32); if (copy_from_user(&mirror_id, (void __user *)arg, sizeof(__u32))) - GOTO(out, rc = -EFAULT); + GOTO(out_lease_close, rc = -EFAULT); layout_file = fget(fdv); if (!layout_file) - GOTO(out, rc = -EBADF); + GOTO(out_lease_close, rc = -EBADF); sp.sp_inode = file_inode(layout_file); sp.sp_mirror_id = (__u16)mirror_id; @@ -3172,11 +3418,35 @@ static long ll_file_unlock_lease(struct file *file, struct ll_ioc_lease *ioc, bias = MDS_CLOSE_LAYOUT_SPLIT; break; } + case LL_LEASE_PCC_ATTACH: + if (ioc->lil_count != 1) + RETURN(-EINVAL); + + arg += sizeof(*ioc); + if (copy_from_user(¶m.pa_archive_id, (void __user *)arg, + sizeof(__u32))) + GOTO(out_lease_close, rc2 = -EFAULT); + + rc2 = pcc_readwrite_attach(file, inode, param.pa_archive_id); + if (rc2) + GOTO(out_lease_close, rc2); + + attached = true; + /* Grab latest data version */ + rc2 = ll_data_version(inode, ¶m.pa_data_version, + LL_DV_WR_FLUSH); + if (rc2) + GOTO(out_lease_close, rc2); + + data = ¶m; + bias = MDS_PCC_ATTACH; + break; default: /* without close intent */ break; } +out_lease_close: rc = ll_lease_close_intent(och, inode, &lease_broken, bias, data); if (rc < 0) GOTO(out, rc); @@ -3200,6 +3470,14 @@ out: if (layout_file) fput(layout_file); break; + case LL_LEASE_PCC_ATTACH: + if (!rc) + rc = rc2; + rc = pcc_readwrite_attach_fini(file, inode, + param.pa_layout_gen, + lease_broken, rc, + attached); + break; } if (!rc) @@ -3212,7 +3490,7 @@ static long ll_file_set_lease(struct file *file, struct ll_ioc_lease *ioc, { struct inode *inode = file_inode(file); struct ll_inode_info *lli = ll_i2info(inode); - struct ll_file_data *fd = LUSTRE_FPRIVATE(file); + struct ll_file_data *fd = file->private_data; struct obd_client_handle *och = NULL; __u64 open_flags = 0; bool lease_broken; @@ -3290,7 +3568,7 @@ static void ll_heat_get(struct inode *inode, struct lu_heat *heat) spin_unlock(&lli->lli_heat_lock); } -static int ll_heat_set(struct inode *inode, __u64 flags) +static int ll_heat_set(struct inode *inode, enum lu_heat_flag flags) { struct ll_inode_info *lli = ll_i2info(inode); int rc = 0; @@ -3313,7 +3591,7 @@ static long ll_file_ioctl(struct file *file, unsigned int cmd, unsigned long arg) { struct inode *inode = file_inode(file); - struct ll_file_data *fd = LUSTRE_FPRIVATE(file); + struct ll_file_data *fd = file->private_data; int flags, rc; ENTRY; @@ -3686,9 +3964,9 @@ out_ladvise: fd->fd_designated_mirror = (__u32)arg; RETURN(0); } - case LL_IOC_FSGETXATTR: + case FS_IOC_FSGETXATTR: RETURN(ll_ioctl_fsgetxattr(inode, cmd, arg)); - case LL_IOC_FSSETXATTR: + case FS_IOC_FSSETXATTR: RETURN(ll_ioctl_fssetxattr(inode, cmd, arg)); case BLKSSZGET: RETURN(put_user(PAGE_SIZE, (int __user *)arg)); @@ -3723,101 +4001,161 @@ out_ladvise: rc = ll_heat_set(inode, flags); RETURN(rc); } + case LL_IOC_PCC_DETACH: { + struct lu_pcc_detach *detach; + + OBD_ALLOC_PTR(detach); + if (detach == NULL) + RETURN(-ENOMEM); + + if (copy_from_user(detach, + (const struct lu_pcc_detach __user *)arg, + sizeof(*detach))) + GOTO(out_detach_free, rc = -EFAULT); + + if (!S_ISREG(inode->i_mode)) + GOTO(out_detach_free, rc = -EINVAL); + + if (!inode_owner_or_capable(inode)) + GOTO(out_detach_free, rc = -EPERM); + + rc = pcc_ioctl_detach(inode, detach->pccd_opt); +out_detach_free: + OBD_FREE_PTR(detach); + RETURN(rc); + } + case LL_IOC_PCC_STATE: { + struct lu_pcc_state __user *ustate = + (struct lu_pcc_state __user *)arg; + struct lu_pcc_state *state; + + OBD_ALLOC_PTR(state); + if (state == NULL) + RETURN(-ENOMEM); + + if (copy_from_user(state, ustate, sizeof(*state))) + GOTO(out_state, rc = -EFAULT); + + rc = pcc_ioctl_state(file, inode, state); + if (rc) + GOTO(out_state, rc); + + if (copy_to_user(ustate, state, sizeof(*state))) + GOTO(out_state, rc = -EFAULT); + +out_state: + OBD_FREE_PTR(state); + RETURN(rc); + } +#ifdef HAVE_LUSTRE_CRYPTO + case LL_IOC_SET_ENCRYPTION_POLICY: + if (!ll_sbi_has_encrypt(ll_i2sbi(inode))) + return -EOPNOTSUPP; + return llcrypt_ioctl_set_policy(file, (const void __user *)arg); + case LL_IOC_GET_ENCRYPTION_POLICY_EX: + if (!ll_sbi_has_encrypt(ll_i2sbi(inode))) + return -EOPNOTSUPP; + return llcrypt_ioctl_get_policy_ex(file, (void __user *)arg); + case LL_IOC_ADD_ENCRYPTION_KEY: + if (!ll_sbi_has_encrypt(ll_i2sbi(inode))) + return -EOPNOTSUPP; + return llcrypt_ioctl_add_key(file, (void __user *)arg); + case LL_IOC_REMOVE_ENCRYPTION_KEY: + if (!ll_sbi_has_encrypt(ll_i2sbi(inode))) + return -EOPNOTSUPP; + return llcrypt_ioctl_remove_key(file, (void __user *)arg); + case LL_IOC_REMOVE_ENCRYPTION_KEY_ALL_USERS: + if (!ll_sbi_has_encrypt(ll_i2sbi(inode))) + return -EOPNOTSUPP; + return llcrypt_ioctl_remove_key_all_users(file, + (void __user *)arg); + case LL_IOC_GET_ENCRYPTION_KEY_STATUS: + if (!ll_sbi_has_encrypt(ll_i2sbi(inode))) + return -EOPNOTSUPP; + return llcrypt_ioctl_get_key_status(file, (void __user *)arg); +#endif default: RETURN(obd_iocontrol(cmd, ll_i2dtexp(inode), 0, NULL, (void __user *)arg)); } } -#ifndef HAVE_FILE_LLSEEK_SIZE -static inline loff_t -llseek_execute(struct file *file, loff_t offset, loff_t maxsize) +loff_t ll_lseek(struct inode *inode, loff_t offset, int whence) { - if (offset < 0 && !(file->f_mode & FMODE_UNSIGNED_OFFSET)) - return -EINVAL; - if (offset > maxsize) - return -EINVAL; + struct lu_env *env; + struct cl_io *io; + struct cl_lseek_io *lsio; + __u16 refcheck; + int rc; + loff_t retval; - if (offset != file->f_pos) { - file->f_pos = offset; - file->f_version = 0; - } - return offset; -} + ENTRY; -static loff_t -generic_file_llseek_size(struct file *file, loff_t offset, int origin, - loff_t maxsize, loff_t eof) -{ - struct inode *inode = file_inode(file); + env = cl_env_get(&refcheck); + if (IS_ERR(env)) + RETURN(PTR_ERR(env)); - switch (origin) { - case SEEK_END: - offset += eof; - break; - case SEEK_CUR: - /* - * Here we special-case the lseek(fd, 0, SEEK_CUR) - * position-querying operation. Avoid rewriting the "same" - * f_pos value back to the file because a concurrent read(), - * write() or lseek() might have altered it - */ - if (offset == 0) - return file->f_pos; - /* - * f_lock protects against read/modify/write race with other - * SEEK_CURs. Note that parallel writes and reads behave - * like SEEK_SET. - */ - inode_lock(inode); - offset = llseek_execute(file, file->f_pos + offset, maxsize); - inode_unlock(inode); - return offset; - case SEEK_DATA: - /* - * In the generic case the entire file is data, so as long as - * offset isn't at the end of the file then the offset is data. - */ - if (offset >= eof) - return -ENXIO; - break; - case SEEK_HOLE: - /* - * There is a virtual hole at the end of the file, so as long as - * offset isn't i_size or larger, return i_size. - */ - if (offset >= eof) - return -ENXIO; - offset = eof; - break; - } + io = vvp_env_thread_io(env); + io->ci_obj = ll_i2info(inode)->lli_clob; + + lsio = &io->u.ci_lseek; + lsio->ls_start = offset; + lsio->ls_whence = whence; + lsio->ls_result = -ENXIO; + + do { + rc = cl_io_init(env, io, CIT_LSEEK, io->ci_obj); + if (!rc) + rc = cl_io_loop(env, io); + else + rc = io->ci_result; + retval = rc ? : lsio->ls_result; + cl_io_fini(env, io); + } while (unlikely(io->ci_need_restart)); + + cl_env_put(env, &refcheck); - return llseek_execute(file, offset, maxsize); + RETURN(retval); } -#endif static loff_t ll_file_seek(struct file *file, loff_t offset, int origin) { struct inode *inode = file_inode(file); - loff_t retval, eof = 0; + loff_t retval = offset, eof = 0; + ktime_t kstart = ktime_get(); ENTRY; - retval = offset + ((origin == SEEK_END) ? i_size_read(inode) : - (origin == SEEK_CUR) ? file->f_pos : 0); + CDEBUG(D_VFSTRACE, "VFS Op:inode="DFID"(%p), to=%llu=%#llx(%d)\n", PFID(ll_inode2fid(inode)), inode, retval, retval, origin); - ll_stats_ops_tally(ll_i2sbi(inode), LPROC_LL_LLSEEK, 1); - if (origin == SEEK_END || origin == SEEK_HOLE || origin == SEEK_DATA) { + if (origin == SEEK_END) { retval = ll_glimpse_size(inode); if (retval != 0) RETURN(retval); eof = i_size_read(inode); } - retval = ll_generic_file_llseek_size(file, offset, origin, - ll_file_maxbytes(inode), eof); + if (origin == SEEK_HOLE || origin == SEEK_DATA) { + if (offset < 0) + return -ENXIO; + + /* flush local cache first if any */ + cl_sync_file_range(inode, offset, OBD_OBJECT_EOF, + CL_FSYNC_LOCAL, 0); + + retval = ll_lseek(inode, offset, origin); + if (retval < 0) + return retval; + retval = vfs_setpos(file, retval, ll_file_maxbytes(inode)); + } else { + retval = generic_file_llseek_size(file, offset, origin, + ll_file_maxbytes(inode), eof); + } + if (retval >= 0) + ll_stats_ops_tally(ll_i2sbi(inode), LPROC_LL_LLSEEK, + ktime_us_delta(ktime_get(), kstart)); RETURN(retval); } @@ -3825,7 +4163,7 @@ static int ll_flush(struct file *file, fl_owner_t id) { struct inode *inode = file_inode(file); struct ll_inode_info *lli = ll_i2info(inode); - struct ll_file_data *fd = LUSTRE_FPRIVATE(file); + struct ll_file_data *fd = file->private_data; int rc, err; LASSERT(!S_ISDIR(inode->i_mode)); @@ -3901,40 +4239,25 @@ int cl_sync_file_range(struct inode *inode, loff_t start, loff_t end, * file_dentry() as is done otherwise. */ -#ifdef HAVE_FILE_FSYNC_4ARGS int ll_fsync(struct file *file, loff_t start, loff_t end, int datasync) { struct dentry *dentry = file_dentry(file); -#elif defined(HAVE_FILE_FSYNC_2ARGS) -int ll_fsync(struct file *file, int datasync) -{ - struct dentry *dentry = file_dentry(file); - loff_t start = 0; - loff_t end = LLONG_MAX; -#else -int ll_fsync(struct file *file, struct dentry *dentry, int datasync) -{ - loff_t start = 0; - loff_t end = LLONG_MAX; -#endif struct inode *inode = dentry->d_inode; struct ll_inode_info *lli = ll_i2info(inode); struct ptlrpc_request *req; + ktime_t kstart = ktime_get(); int rc, err; + ENTRY; - CDEBUG(D_VFSTRACE, "VFS Op:inode="DFID"(%p)\n", - PFID(ll_inode2fid(inode)), inode); - ll_stats_ops_tally(ll_i2sbi(inode), LPROC_LL_FSYNC, 1); + CDEBUG(D_VFSTRACE, + "VFS Op:inode="DFID"(%p), start %lld, end %lld, datasync %d\n", + PFID(ll_inode2fid(inode)), inode, start, end, datasync); -#ifdef HAVE_FILE_FSYNC_4ARGS - rc = filemap_write_and_wait_range(inode->i_mapping, start, end); - inode_lock(inode); -#else /* fsync's caller has already called _fdata{sync,write}, we want * that IO to finish before calling the osc and mdc sync methods */ - rc = filemap_fdatawait(inode->i_mapping); -#endif + rc = filemap_write_and_wait_range(inode->i_mapping, start, end); + inode_lock(inode); /* catch async errors that were recorded back when async writeback * failed for pages in this mapping. */ @@ -3957,9 +4280,16 @@ int ll_fsync(struct file *file, struct dentry *dentry, int datasync) ptlrpc_req_finished(req); if (S_ISREG(inode->i_mode)) { - struct ll_file_data *fd = LUSTRE_FPRIVATE(file); + struct ll_file_data *fd = file->private_data; + bool cached; - err = cl_sync_file_range(inode, start, end, CL_FSYNC_ALL, 0); + /* Sync metadata on MDT first, and then sync the cached data + * on PCC. + */ + err = pcc_fsync(file, start, end, datasync, &cached); + if (!cached) + err = cl_sync_file_range(inode, start, end, + CL_FSYNC_ALL, 0); if (rc == 0 && err < 0) rc = err; if (rc < 0) @@ -3968,9 +4298,11 @@ int ll_fsync(struct file *file, struct dentry *dentry, int datasync) fd->fd_write_failed = false; } -#ifdef HAVE_FILE_FSYNC_4ARGS inode_unlock(inode); -#endif + + if (!rc) + ll_stats_ops_tally(ll_i2sbi(inode), LPROC_LL_FSYNC, + ktime_us_delta(ktime_get(), kstart)); RETURN(rc); } @@ -3988,6 +4320,7 @@ ll_file_flock(struct file *file, int cmd, struct file_lock *file_lock) struct lustre_handle lockh = { 0 }; union ldlm_policy_data flock = { { 0 } }; int fl_type = file_lock->fl_type; + ktime_t kstart = ktime_get(); __u64 flags = 0; int rc; int rc2 = 0; @@ -3996,23 +4329,22 @@ ll_file_flock(struct file *file, int cmd, struct file_lock *file_lock) CDEBUG(D_VFSTRACE, "VFS Op:inode="DFID" file_lock=%p\n", PFID(ll_inode2fid(inode)), file_lock); - ll_stats_ops_tally(ll_i2sbi(inode), LPROC_LL_FLOCK, 1); - - if (file_lock->fl_flags & FL_FLOCK) { - LASSERT((cmd == F_SETLKW) || (cmd == F_SETLK)); - /* flocks are whole-file locks */ - flock.l_flock.end = OFFSET_MAX; - /* For flocks owner is determined by the local file desctiptor*/ - flock.l_flock.owner = (unsigned long)file_lock->fl_file; - } else if (file_lock->fl_flags & FL_POSIX) { - flock.l_flock.owner = (unsigned long)file_lock->fl_owner; - flock.l_flock.start = file_lock->fl_start; - flock.l_flock.end = file_lock->fl_end; - } else { - RETURN(-EINVAL); - } - flock.l_flock.pid = file_lock->fl_pid; + if (file_lock->fl_flags & FL_FLOCK) { + LASSERT((cmd == F_SETLKW) || (cmd == F_SETLK)); + /* flocks are whole-file locks */ + flock.l_flock.end = OFFSET_MAX; + /* For flocks owner is determined by the local file desctiptor*/ + flock.l_flock.owner = (unsigned long)file_lock->fl_file; + } else if (file_lock->fl_flags & FL_POSIX) { + flock.l_flock.owner = (unsigned long)file_lock->fl_owner; + flock.l_flock.start = file_lock->fl_start; + flock.l_flock.end = file_lock->fl_end; + } else { + RETURN(-EINVAL); + } + flock.l_flock.pid = file_lock->fl_pid; +#if defined(HAVE_LM_COMPARE_OWNER) || defined(lm_compare_owner) /* Somewhat ugly workaround for svc lockd. * lockd installs custom fl_lmops->lm_compare_owner that checks * for the fl_owner to be the same (which it always is on local node @@ -4022,6 +4354,7 @@ ll_file_flock(struct file *file, int cmd, struct file_lock *file_lock) * pointer space for current->files are not intersecting */ if (file_lock->fl_lmops && file_lock->fl_lmops->lm_compare_owner) flock.l_flock.owner = (unsigned long)file_lock->fl_pid; +#endif switch (fl_type) { case F_RDLCK: @@ -4114,7 +4447,10 @@ ll_file_flock(struct file *file, int cmd, struct file_lock *file_lock) ll_finish_md_op_data(op_data); - RETURN(rc); + if (!rc) + ll_stats_ops_tally(ll_i2sbi(inode), LPROC_LL_FLOCK, + ktime_us_delta(ktime_get(), kstart)); + RETURN(rc); } int ll_get_fid_by_name(struct inode *parent, const char *name, @@ -4199,10 +4535,9 @@ int ll_migrate(struct inode *parent, struct file *file, struct lmv_user_md *lum, if (!(exp_connect_flags2(ll_i2sbi(parent)->ll_md_exp) & OBD_CONNECT2_DIR_MIGRATE)) { if (le32_to_cpu(lum->lum_stripe_count) > 1 || - ll_i2info(child_inode)->lli_lsm_md) { + ll_dir_striped(child_inode)) { CERROR("%s: MDT doesn't support stripe directory " - "migration!\n", - ll_get_fsname(parent->i_sb, NULL, 0)); + "migration!\n", ll_i2sbi(parent)->ll_fsname); GOTO(out_iput, rc = -EOPNOTSUPP); } } @@ -4224,7 +4559,7 @@ int ll_migrate(struct inode *parent, struct file *file, struct lmv_user_md *lum, op_data->op_fid3 = *ll_inode2fid(child_inode); if (!fid_is_sane(&op_data->op_fid3)) { CERROR("%s: migrate %s, but FID "DFID" is insane\n", - ll_get_fsname(parent->i_sb, NULL, 0), name, + ll_i2sbi(parent)->ll_fsname, name, PFID(&op_data->op_fid3)); GOTO(out_unlock, rc = -EINVAL); } @@ -4305,9 +4640,20 @@ out_iput: static int ll_file_noflock(struct file *file, int cmd, struct file_lock *file_lock) { - ENTRY; + struct ll_file_data *fd = file->private_data; + ENTRY; - RETURN(-ENOSYS); + /* + * In order to avoid flood of warning messages, only print one message + * for one file. And the entire message rate on the client is limited + * by CDEBUG_LIMIT too. + */ + if (!(fd->fd_flags & LL_FILE_FLOCK_WARNING)) { + fd->fd_flags |= LL_FILE_FLOCK_WARNING; + CDEBUG_LIMIT(D_TTY | D_CONSOLE, + "flock disabled, mount with '-o [local]flock' to enable\r\n"); + } + RETURN(-ENOSYS); } /** @@ -4339,25 +4685,25 @@ int ll_have_md_lock(struct inode *inode, __u64 *bits, enum ldlm_mode l_req_mode) ldlm_lockname[mode]); flags = LDLM_FL_BLOCK_GRANTED | LDLM_FL_CBPENDING | LDLM_FL_TEST_LOCK; - for (i = 0; i <= MDS_INODELOCK_MAXSHIFT && *bits != 0; i++) { - policy.l_inodebits.bits = *bits & (1 << i); + for (i = 0; i < MDS_INODELOCK_NUMBITS && *bits != 0; i++) { + policy.l_inodebits.bits = *bits & BIT(i); if (policy.l_inodebits.bits == 0) continue; - if (md_lock_match(ll_i2mdexp(inode), flags, fid, LDLM_IBITS, - &policy, mode, &lockh)) { - struct ldlm_lock *lock; - - lock = ldlm_handle2lock(&lockh); - if (lock) { - *bits &= - ~(lock->l_policy_data.l_inodebits.bits); - LDLM_LOCK_PUT(lock); - } else { - *bits &= ~policy.l_inodebits.bits; - } - } - } + if (md_lock_match(ll_i2mdexp(inode), flags, fid, LDLM_IBITS, + &policy, mode, &lockh)) { + struct ldlm_lock *lock; + + lock = ldlm_handle2lock(&lockh); + if (lock) { + *bits &= + ~(lock->l_policy_data.l_inodebits.bits); + LDLM_LOCK_PUT(lock); + } else { + *bits &= ~policy.l_inodebits.bits; + } + } + } RETURN(*bits == 0); } @@ -4387,8 +4733,7 @@ static int ll_inode_revalidate_fini(struct inode *inode, int rc) /* If it is striped directory, and there is bad stripe * Let's revalidate the dentry again, instead of returning * error */ - if (S_ISDIR(inode->i_mode) && - ll_i2info(inode)->lli_lsm_md != NULL) + if (ll_dir_striped(inode)) return 0; /* This path cannot be hit for regular files unless in @@ -4399,7 +4744,7 @@ static int ll_inode_revalidate_fini(struct inode *inode, int rc) } else if (rc != 0) { CDEBUG_LIMIT((rc == -EACCES || rc == -EIDRM) ? D_INFO : D_ERROR, "%s: revalidate FID "DFID" error: rc = %d\n", - ll_get_fsname(inode->i_sb, NULL, 0), + ll_i2sbi(inode)->ll_fsname, PFID(ll_inode2fid(inode)), rc); } @@ -4408,6 +4753,7 @@ static int ll_inode_revalidate_fini(struct inode *inode, int rc) static int ll_inode_revalidate(struct dentry *dentry, enum ldlm_intent_flags op) { + struct inode *parent; struct inode *inode = dentry->d_inode; struct obd_export *exp = ll_i2mdexp(inode); struct lookup_intent oit = { @@ -4415,18 +4761,30 @@ static int ll_inode_revalidate(struct dentry *dentry, enum ldlm_intent_flags op) }; struct ptlrpc_request *req = NULL; struct md_op_data *op_data; + const char *name = NULL; + size_t namelen = 0; int rc = 0; ENTRY; CDEBUG(D_VFSTRACE, "VFS Op:inode="DFID"(%p),name=%s\n", PFID(ll_inode2fid(inode)), inode, dentry->d_name.name); - /* Call getattr by fid, so do not provide name at all. */ - op_data = ll_prep_md_op_data(NULL, inode, inode, NULL, 0, 0, + if (exp_connect_flags2(exp) & OBD_CONNECT2_GETATTR_PFID) { + parent = dentry->d_parent->d_inode; + name = dentry->d_name.name; + namelen = dentry->d_name.len; + } else { + parent = inode; + } + + op_data = ll_prep_md_op_data(NULL, parent, inode, name, namelen, 0, LUSTRE_OPC_ANY, NULL); if (IS_ERR(op_data)) RETURN(PTR_ERR(op_data)); + /* Call getattr by fid */ + if (exp_connect_flags2(exp) & OBD_CONNECT2_GETATTR_PFID) + op_data->op_flags = MF_GETATTR_BY_FID; rc = md_intent_lock(exp, op_data, &oit, &req, &ll_md_blocking_ast, 0); ll_finish_md_op_data(op_data); if (rc < 0) { @@ -4444,11 +4802,8 @@ static int ll_inode_revalidate(struct dentry *dentry, enum ldlm_intent_flags op) * do_lookup() -> ll_revalidate_it(). We cannot use d_drop * here to preserve get_cwd functionality on 2.6. * Bug 10503 */ - if (!dentry->d_inode->i_nlink) { - ll_lock_dcache(inode); - d_lustre_invalidate(dentry, 0); - ll_unlock_dcache(inode); - } + if (!dentry->d_inode->i_nlink) + d_lustre_invalidate(dentry); ll_lookup_finish_locks(&oit, dentry); out: @@ -4464,8 +4819,12 @@ static int ll_merge_md_attr(struct inode *inode) int rc; LASSERT(lli->lli_lsm_md != NULL); + + if (!lmv_dir_striped(lli->lli_lsm_md)) + RETURN(0); + down_read(&lli->lli_lsm_sem); - rc = md_merge_attr(ll_i2mdexp(inode), ll_i2info(inode)->lli_lsm_md, + rc = md_merge_attr(ll_i2mdexp(inode), &lli->lli_fid, lli->lli_lsm_md, &attr, ll_md_blocking_ast); up_read(&lli->lli_lsm_sem); if (rc != 0) @@ -4482,39 +4841,66 @@ static int ll_merge_md_attr(struct inode *inode) RETURN(0); } -static inline dev_t ll_compat_encode_dev(dev_t dev) -{ - /* The compat_sys_*stat*() syscalls will fail unless the - * device majors and minors are both less than 256. Note that - * the value returned here will be passed through - * old_encode_dev() in cp_compat_stat(). And so we are not - * trying to return a valid compat (u16) device number, just - * one that will pass the old_valid_dev() check. */ - - return MKDEV(MAJOR(dev) & 0xff, MINOR(dev) & 0xff); -} - -#ifdef HAVE_INODEOPS_ENHANCED_GETATTR -int ll_getattr(const struct path *path, struct kstat *stat, - u32 request_mask, unsigned int flags) -{ - struct dentry *de = path->dentry; -#else -int ll_getattr(struct vfsmount *mnt, struct dentry *de, struct kstat *stat) +int ll_getattr_dentry(struct dentry *de, struct kstat *stat, u32 request_mask, + unsigned int flags) { -#endif struct inode *inode = de->d_inode; struct ll_sb_info *sbi = ll_i2sbi(inode); struct ll_inode_info *lli = ll_i2info(inode); + struct inode *dir = de->d_parent->d_inode; + bool need_glimpse = true; + ktime_t kstart = ktime_get(); int rc; - ll_stats_ops_tally(sbi, LPROC_LL_GETATTR, 1); + /* The OST object(s) determine the file size, blocks and mtime. */ + if (!(request_mask & STATX_SIZE || request_mask & STATX_BLOCKS || + request_mask & STATX_MTIME)) + need_glimpse = false; + + if (dentry_may_statahead(dir, de)) + ll_start_statahead(dir, de, need_glimpse && + !(flags & AT_STATX_DONT_SYNC)); + + if (flags & AT_STATX_DONT_SYNC) + GOTO(fill_attr, rc = 0); rc = ll_inode_revalidate(de, IT_GETATTR); if (rc < 0) RETURN(rc); if (S_ISREG(inode->i_mode)) { + bool cached; + + if (!need_glimpse) + GOTO(fill_attr, rc); + + rc = pcc_inode_getattr(inode, request_mask, flags, &cached); + if (cached && rc < 0) + RETURN(rc); + + if (cached) + GOTO(fill_attr, rc); + + /* + * If the returned attr is masked with OBD_MD_FLSIZE & + * OBD_MD_FLBLOCKS & OBD_MD_FLMTIME, it means that the file size + * or blocks obtained from MDT is strictly correct, and the file + * is usually not being modified by clients, and the [a|m|c]time + * got from MDT is also strictly correct. + * Under this circumstance, it does not need to send glimpse + * RPCs to OSTs for file attributes such as the size and blocks. + */ + if (lli->lli_attr_valid & OBD_MD_FLSIZE && + lli->lli_attr_valid & OBD_MD_FLBLOCKS && + lli->lli_attr_valid & OBD_MD_FLMTIME) { + inode->i_mtime.tv_sec = lli->lli_mtime; + if (lli->lli_attr_valid & OBD_MD_FLATIME) + inode->i_atime.tv_sec = lli->lli_atime; + if (lli->lli_attr_valid & OBD_MD_FLCTIME) + inode->i_ctime.tv_sec = lli->lli_ctime; + GOTO(fill_attr, rc); + } + /* In case of restore, the MDT has the right size and has * already send it back without granting the layout lock, * inode is up-to-date so glimpse is useless. @@ -4529,18 +4915,21 @@ int ll_getattr(struct vfsmount *mnt, struct dentry *de, struct kstat *stat) } } else { /* If object isn't regular a file then don't validate size. */ - if (S_ISDIR(inode->i_mode) && - lli->lli_lsm_md != NULL) { + if (ll_dir_striped(inode)) { rc = ll_merge_md_attr(inode); if (rc < 0) RETURN(rc); } - inode->i_atime.tv_sec = lli->lli_atime; - inode->i_mtime.tv_sec = lli->lli_mtime; - inode->i_ctime.tv_sec = lli->lli_ctime; + if (lli->lli_attr_valid & OBD_MD_FLATIME) + inode->i_atime.tv_sec = lli->lli_atime; + if (lli->lli_attr_valid & OBD_MD_FLMTIME) + inode->i_mtime.tv_sec = lli->lli_mtime; + if (lli->lli_attr_valid & OBD_MD_FLCTIME) + inode->i_ctime.tv_sec = lli->lli_ctime; } +fill_attr: OBD_FAIL_TIMEOUT(OBD_FAIL_GETATTR_DELAY, 30); if (ll_need_32bit_api(sbi)) { @@ -4559,13 +4948,146 @@ int ll_getattr(struct vfsmount *mnt, struct dentry *de, struct kstat *stat) stat->atime = inode->i_atime; stat->mtime = inode->i_mtime; stat->ctime = inode->i_ctime; - stat->blksize = sbi->ll_stat_blksize ?: 1 << inode->i_blkbits; + /* stat->blksize is used to tell about preferred IO size */ + if (sbi->ll_stat_blksize) + stat->blksize = sbi->ll_stat_blksize; + else if (S_ISREG(inode->i_mode)) + stat->blksize = 1 << min(PTLRPC_MAX_BRW_BITS + 1, + LL_MAX_BLKSIZE_BITS); + else + stat->blksize = 1 << inode->i_sb->s_blocksize_bits; stat->nlink = inode->i_nlink; stat->size = i_size_read(inode); stat->blocks = inode->i_blocks; - return 0; +#ifdef HAVE_INODEOPS_ENHANCED_GETATTR + if (flags & AT_STATX_DONT_SYNC) { + if (stat->size == 0 && + lli->lli_attr_valid & OBD_MD_FLLAZYSIZE) + stat->size = lli->lli_lazysize; + if (stat->blocks == 0 && + lli->lli_attr_valid & OBD_MD_FLLAZYBLOCKS) + stat->blocks = lli->lli_lazyblocks; + } + + if (lli->lli_attr_valid & OBD_MD_FLBTIME) { + stat->result_mask |= STATX_BTIME; + stat->btime.tv_sec = lli->lli_btime; + } + + stat->attributes_mask = STATX_ATTR_IMMUTABLE | STATX_ATTR_APPEND; + stat->attributes |= ll_inode_to_ext_flags(inode->i_flags); + stat->result_mask &= request_mask; +#endif + + ll_stats_ops_tally(sbi, LPROC_LL_GETATTR, + ktime_us_delta(ktime_get(), kstart)); + + return 0; +} + +#ifdef HAVE_INODEOPS_ENHANCED_GETATTR +int ll_getattr(const struct path *path, struct kstat *stat, + u32 request_mask, unsigned int flags) +{ + return ll_getattr_dentry(path->dentry, stat, request_mask, flags); +} +#else +int ll_getattr(struct vfsmount *mnt, struct dentry *de, struct kstat *stat) +{ + return ll_getattr_dentry(de, stat, STATX_BASIC_STATS, + AT_STATX_SYNC_AS_STAT); +} +#endif + +int cl_falloc(struct inode *inode, int mode, loff_t offset, loff_t len) +{ + struct lu_env *env; + struct cl_io *io; + __u16 refcheck; + int rc; loff_t sa_falloc_end; + loff_t size = i_size_read(inode); + + ENTRY; + + env = cl_env_get(&refcheck); + if (IS_ERR(env)) + RETURN(PTR_ERR(env)); + + io = vvp_env_thread_io(env); + io->ci_obj = ll_i2info(inode)->lli_clob; + io->ci_verify_layout = 1; + io->u.ci_setattr.sa_parent_fid = lu_object_fid(&io->ci_obj->co_lu); + io->u.ci_setattr.sa_falloc_mode = mode; + io->u.ci_setattr.sa_falloc_offset = offset; + io->u.ci_setattr.sa_falloc_len = len; + io->u.ci_setattr.sa_falloc_end = io->u.ci_setattr.sa_falloc_offset + + io->u.ci_setattr.sa_falloc_len; + io->u.ci_setattr.sa_subtype = CL_SETATTR_FALLOCATE; + sa_falloc_end = io->u.ci_setattr.sa_falloc_end; + if (sa_falloc_end > size) { + /* Check new size against VFS/VM file size limit and rlimit */ + rc = inode_newsize_ok(inode, sa_falloc_end); + if (rc) + goto out; + if (sa_falloc_end > ll_file_maxbytes(inode)) { + CDEBUG(D_INODE, "file size too large %llu > %llu\n", + (unsigned long long)(sa_falloc_end), + ll_file_maxbytes(inode)); + rc = -EFBIG; + goto out; + } + io->u.ci_setattr.sa_attr.lvb_size = sa_falloc_end; + if (!(mode & FALLOC_FL_KEEP_SIZE)) + io->u.ci_setattr.sa_avalid |= ATTR_SIZE; + } else { + io->u.ci_setattr.sa_attr.lvb_size = size; + } + +again: + if (cl_io_init(env, io, CIT_SETATTR, io->ci_obj) == 0) + rc = cl_io_loop(env, io); + else + rc = io->ci_result; + + cl_io_fini(env, io); + if (unlikely(io->ci_need_restart)) + goto again; + +out: + cl_env_put(env, &refcheck); + RETURN(rc); +} + +long ll_fallocate(struct file *filp, int mode, loff_t offset, loff_t len) +{ + struct inode *inode = filp->f_path.dentry->d_inode; + int rc; + + /* + * Encrypted inodes can't handle collapse range or zero range or insert + * range since we would need to re-encrypt blocks with a different IV or + * XTS tweak (which are based on the logical block number). + * Similar to what ext4 does. + */ + if (IS_ENCRYPTED(inode) && + (mode & (FALLOC_FL_COLLAPSE_RANGE | FALLOC_FL_INSERT_RANGE | + FALLOC_FL_ZERO_RANGE))) + RETURN(-EOPNOTSUPP); + + /* + * Only mode == 0 (which is standard prealloc) is supported now. + * Punch is not supported yet. + */ + if (mode & ~FALLOC_FL_KEEP_SIZE) + RETURN(-EOPNOTSUPP); + + ll_stats_ops_tally(ll_i2sbi(inode), LPROC_LL_FALLOCATE, 1); + + rc = cl_falloc(inode, mode, offset, len); + + RETURN(rc); } static int ll_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, @@ -4621,7 +5143,7 @@ struct posix_acl *ll_get_acl(struct inode *inode, int type) } #ifdef HAVE_IOP_SET_ACL -#ifdef CONFIG_FS_POSIX_ACL +#ifdef CONFIG_LUSTRE_FS_POSIX_ACL int ll_set_acl(struct inode *inode, struct posix_acl *acl, int type) { struct ll_sb_info *sbi = ll_i2sbi(inode); @@ -4677,50 +5199,10 @@ out: set_cached_acl(inode, type, acl); RETURN(rc); } -#endif /* CONFIG_FS_POSIX_ACL */ +#endif /* CONFIG_LUSTRE_FS_POSIX_ACL */ #endif /* HAVE_IOP_SET_ACL */ -#ifndef HAVE_GENERIC_PERMISSION_2ARGS -static int -# ifdef HAVE_GENERIC_PERMISSION_4ARGS -ll_check_acl(struct inode *inode, int mask, unsigned int flags) -# else -ll_check_acl(struct inode *inode, int mask) -# endif -{ -# ifdef CONFIG_FS_POSIX_ACL - struct posix_acl *acl; - int rc; - ENTRY; - -# ifdef HAVE_GENERIC_PERMISSION_4ARGS - if (flags & IPERM_FLAG_RCU) - return -ECHILD; -# endif - acl = ll_get_acl(inode, ACL_TYPE_ACCESS); - - if (!acl) - RETURN(-EAGAIN); - - rc = posix_acl_permission(inode, acl, mask); - posix_acl_release(acl); - - RETURN(rc); -# else /* !CONFIG_FS_POSIX_ACL */ - return -EAGAIN; -# endif /* CONFIG_FS_POSIX_ACL */ -} -#endif /* HAVE_GENERIC_PERMISSION_2ARGS */ - -#ifdef HAVE_GENERIC_PERMISSION_4ARGS -int ll_inode_permission(struct inode *inode, int mask, unsigned int flags) -#else -# ifdef HAVE_INODE_PERMISION_2ARGS int ll_inode_permission(struct inode *inode, int mask) -# else -int ll_inode_permission(struct inode *inode, int mask, struct nameidata *nd) -# endif -#endif { int rc = 0; struct ll_sb_info *sbi; @@ -4729,24 +5211,23 @@ int ll_inode_permission(struct inode *inode, int mask, struct nameidata *nd) const struct cred *old_cred = NULL; cfs_cap_t cap; bool squash_id = false; + ktime_t kstart = ktime_get(); + ENTRY; -#ifdef MAY_NOT_BLOCK if (mask & MAY_NOT_BLOCK) return -ECHILD; -#elif defined(HAVE_GENERIC_PERMISSION_4ARGS) - if (flags & IPERM_FLAG_RCU) - return -ECHILD; -#endif - /* as root inode are NOT getting validated in lookup operation, - * need to do it before permission check. */ + /* + * as root inode are NOT getting validated in lookup operation, + * need to do it before permission check. + */ - if (inode == inode->i_sb->s_root->d_inode) { + if (inode == inode->i_sb->s_root->d_inode) { rc = ll_inode_revalidate(inode->i_sb->s_root, IT_LOOKUP); - if (rc) - RETURN(rc); - } + if (rc) + RETURN(rc); + } CDEBUG(D_VFSTRACE, "VFS Op:inode="DFID"(%p), inode mode %x mask %o\n", PFID(ll_inode2fid(inode)), inode, inode->i_mode, mask); @@ -4773,20 +5254,23 @@ int ll_inode_permission(struct inode *inode, int mask, struct nameidata *nd) cred->fsuid = make_kuid(&init_user_ns, squash->rsi_uid); cred->fsgid = make_kgid(&init_user_ns, squash->rsi_gid); for (cap = 0; cap < sizeof(cfs_cap_t) * 8; cap++) { - if ((1 << cap) & CFS_CAP_FS_MASK) + if (BIT(cap) & CFS_CAP_FS_MASK) cap_lower(cred->cap_effective, cap); } old_cred = override_creds(cred); } - ll_stats_ops_tally(sbi, LPROC_LL_INODE_PERM, 1); - rc = ll_generic_permission(inode, mask, flags, ll_check_acl); + rc = generic_permission(inode, mask); /* restore current process's credentials and FS capability */ if (squash_id) { revert_creds(old_cred); put_cred(cred); } + if (!rc) + ll_stats_ops_tally(sbi, LPROC_LL_INODE_PERM, + ktime_us_delta(ktime_get(), kstart)); + RETURN(rc); } @@ -4810,9 +5294,14 @@ struct file_operations ll_file_operations = { .release = ll_file_release, .mmap = ll_file_mmap, .llseek = ll_file_seek, - .splice_read = ll_file_splice_read, +#ifndef HAVE_DEFAULT_FILE_SPLICE_READ_EXPORT + .splice_read = generic_file_splice_read, +#else + .splice_read = pcc_file_splice_read, +#endif .fsync = ll_fsync, - .flush = ll_flush + .flush = ll_flush, + .fallocate = ll_fallocate, }; struct file_operations ll_file_operations_flock = { @@ -4834,11 +5323,16 @@ struct file_operations ll_file_operations_flock = { .release = ll_file_release, .mmap = ll_file_mmap, .llseek = ll_file_seek, - .splice_read = ll_file_splice_read, +#ifndef HAVE_DEFAULT_FILE_SPLICE_READ_EXPORT + .splice_read = generic_file_splice_read, +#else + .splice_read = pcc_file_splice_read, +#endif .fsync = ll_fsync, .flush = ll_flush, .flock = ll_file_flock, - .lock = ll_file_flock + .lock = ll_file_flock, + .fallocate = ll_fallocate, }; /* These are for -o noflock - to return ENOSYS on flock calls */ @@ -4861,11 +5355,16 @@ struct file_operations ll_file_operations_noflock = { .release = ll_file_release, .mmap = ll_file_mmap, .llseek = ll_file_seek, - .splice_read = ll_file_splice_read, +#ifndef HAVE_DEFAULT_FILE_SPLICE_READ_EXPORT + .splice_read = generic_file_splice_read, +#else + .splice_read = pcc_file_splice_read, +#endif .fsync = ll_fsync, .flush = ll_flush, .flock = ll_file_noflock, - .lock = ll_file_noflock + .lock = ll_file_noflock, + .fallocate = ll_fallocate, }; struct inode_operations ll_file_inode_operations = { @@ -4879,9 +5378,7 @@ struct inode_operations ll_file_inode_operations = { #endif .listxattr = ll_listxattr, .fiemap = ll_fiemap, -#ifdef HAVE_IOP_GET_ACL .get_acl = ll_get_acl, -#endif #ifdef HAVE_IOP_SET_ACL .set_acl = ll_set_acl, #endif @@ -5074,8 +5571,7 @@ out: /* wait for IO to complete if it's still being used. */ if (wait_layout) { CDEBUG(D_INODE, "%s: "DFID"(%p) wait for layout reconf\n", - ll_get_fsname(inode->i_sb, NULL, 0), - PFID(&lli->lli_fid), inode); + sbi->ll_fsname, PFID(&lli->lli_fid), inode); memset(&conf, 0, sizeof conf); conf.coc_opc = OBJECT_CONF_WAIT; @@ -5085,8 +5581,7 @@ out: rc = -EAGAIN; CDEBUG(D_INODE, "%s file="DFID" waiting layout return: %d\n", - ll_get_fsname(inode->i_sb, NULL, 0), - PFID(&lli->lli_fid), rc); + sbi->ll_fsname, PFID(&lli->lli_fid), rc); } RETURN(rc); } @@ -5124,8 +5619,7 @@ static int ll_layout_intent(struct inode *inode, struct layout_intent *intent) it.it_flags = FMODE_WRITE; LDLM_DEBUG_NOLOCK("%s: requeue layout lock for file "DFID"(%p)", - ll_get_fsname(inode->i_sb, NULL, 0), - PFID(&lli->lli_fid), inode); + sbi->ll_fsname, PFID(&lli->lli_fid), inode); rc = md_intent_lock(sbi->ll_md_exp, op_data, &it, &req, &ll_md_blocking_ast, 0); @@ -5184,7 +5678,8 @@ int ll_layout_refresh(struct inode *inode, __u32 *gen) /* mostly layout lock is caching on the local side, so try to * match it before grabbing layout lock mutex. */ mode = ll_take_md_lock(inode, MDS_INODELOCK_LAYOUT, &lockh, 0, - LCK_CR | LCK_CW | LCK_PR | LCK_PW); + LCK_CR | LCK_CW | LCK_PR | + LCK_PW | LCK_EX); if (mode != 0) { /* hit cached lock */ rc = ll_layout_lock_set(&lockh, mode, inode); if (rc == -EAGAIN)