diff options
Diffstat (limited to 'drivers/staging/lustre/lustre/llite/vvp_io.c')
-rw-r--r-- | drivers/staging/lustre/lustre/llite/vvp_io.c | 1209 |
1 files changed, 1209 insertions, 0 deletions
diff --git a/drivers/staging/lustre/lustre/llite/vvp_io.c b/drivers/staging/lustre/lustre/llite/vvp_io.c new file mode 100644 index 000000000..91bba7967 --- /dev/null +++ b/drivers/staging/lustre/lustre/llite/vvp_io.c @@ -0,0 +1,1209 @@ +/* + * GPL HEADER START + * + * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 only, + * as published by the Free Software Foundation. + * + * This program 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 version 2 for more details (a copy is included + * in the LICENSE file that accompanied this code). + * + * You should have received a copy of the GNU General Public License + * version 2 along with this program; If not, see + * http://www.sun.com/software/products/lustre/docs/GPLv2.pdf + * + * Please contact Sun Microsystems, Inc., 4150 Network Circle, Santa Clara, + * CA 95054 USA or visit www.sun.com if you need additional information or + * have any questions. + * + * GPL HEADER END + */ +/* + * Copyright (c) 2008, 2010, Oracle and/or its affiliates. All rights reserved. + * Use is subject to license terms. + * + * Copyright (c) 2011, 2012, Intel Corporation. + */ +/* + * This file is part of Lustre, http://www.lustre.org/ + * Lustre is a trademark of Sun Microsystems, Inc. + * + * Implementation of cl_io for VVP layer. + * + * Author: Nikita Danilov <nikita.danilov@sun.com> + * Author: Jinshan Xiong <jinshan.xiong@whamcloud.com> + */ + +#define DEBUG_SUBSYSTEM S_LLITE + + +#include "../include/obd.h" +#include "../include/lustre_lite.h" + +#include "vvp_internal.h" + +static struct vvp_io *cl2vvp_io(const struct lu_env *env, + const struct cl_io_slice *slice); + +/** + * True, if \a io is a normal io, False for splice_{read,write} + */ +int cl_is_normalio(const struct lu_env *env, const struct cl_io *io) +{ + struct vvp_io *vio = vvp_env_io(env); + + LASSERT(io->ci_type == CIT_READ || io->ci_type == CIT_WRITE); + + return vio->cui_io_subtype == IO_NORMAL; +} + +/** + * For swapping layout. The file's layout may have changed. + * To avoid populating pages to a wrong stripe, we have to verify the + * correctness of layout. It works because swapping layout processes + * have to acquire group lock. + */ +static bool can_populate_pages(const struct lu_env *env, struct cl_io *io, + struct inode *inode) +{ + struct ll_inode_info *lli = ll_i2info(inode); + struct ccc_io *cio = ccc_env_io(env); + bool rc = true; + + switch (io->ci_type) { + case CIT_READ: + case CIT_WRITE: + /* don't need lock here to check lli_layout_gen as we have held + * extent lock and GROUP lock has to hold to swap layout */ + if (ll_layout_version_get(lli) != cio->cui_layout_gen) { + io->ci_need_restart = 1; + /* this will return application a short read/write */ + io->ci_continue = 0; + rc = false; + } + case CIT_FAULT: + /* fault is okay because we've already had a page. */ + default: + break; + } + + return rc; +} + +/***************************************************************************** + * + * io operations. + * + */ + +static int vvp_io_fault_iter_init(const struct lu_env *env, + const struct cl_io_slice *ios) +{ + struct vvp_io *vio = cl2vvp_io(env, ios); + struct inode *inode = ccc_object_inode(ios->cis_obj); + + LASSERT(inode == + file_inode(cl2ccc_io(env, ios)->cui_fd->fd_file)); + vio->u.fault.ft_mtime = LTIME_S(inode->i_mtime); + return 0; +} + +static void vvp_io_fini(const struct lu_env *env, const struct cl_io_slice *ios) +{ + struct cl_io *io = ios->cis_io; + struct cl_object *obj = io->ci_obj; + struct ccc_io *cio = cl2ccc_io(env, ios); + + CLOBINVRNT(env, obj, ccc_object_invariant(obj)); + + CDEBUG(D_VFSTRACE, DFID + " ignore/verify layout %d/%d, layout version %d restore needed %d\n", + PFID(lu_object_fid(&obj->co_lu)), + io->ci_ignore_layout, io->ci_verify_layout, + cio->cui_layout_gen, io->ci_restore_needed); + + if (io->ci_restore_needed == 1) { + int rc; + + /* file was detected release, we need to restore it + * before finishing the io + */ + rc = ll_layout_restore(ccc_object_inode(obj)); + /* if restore registration failed, no restart, + * we will return -ENODATA */ + /* The layout will change after restore, so we need to + * block on layout lock hold by the MDT + * as MDT will not send new layout in lvb (see LU-3124) + * we have to explicitly fetch it, all this will be done + * by ll_layout_refresh() + */ + if (rc == 0) { + io->ci_restore_needed = 0; + io->ci_need_restart = 1; + io->ci_verify_layout = 1; + } else { + io->ci_restore_needed = 1; + io->ci_need_restart = 0; + io->ci_verify_layout = 0; + io->ci_result = rc; + } + } + + if (!io->ci_ignore_layout && io->ci_verify_layout) { + __u32 gen = 0; + + /* check layout version */ + ll_layout_refresh(ccc_object_inode(obj), &gen); + io->ci_need_restart = cio->cui_layout_gen != gen; + if (io->ci_need_restart) { + CDEBUG(D_VFSTRACE, + DFID" layout changed from %d to %d.\n", + PFID(lu_object_fid(&obj->co_lu)), + cio->cui_layout_gen, gen); + /* today successful restore is the only possible + * case */ + /* restore was done, clear restoring state */ + ll_i2info(ccc_object_inode(obj))->lli_flags &= + ~LLIF_FILE_RESTORING; + } + } +} + +static void vvp_io_fault_fini(const struct lu_env *env, + const struct cl_io_slice *ios) +{ + struct cl_io *io = ios->cis_io; + struct cl_page *page = io->u.ci_fault.ft_page; + + CLOBINVRNT(env, io->ci_obj, ccc_object_invariant(io->ci_obj)); + + if (page != NULL) { + lu_ref_del(&page->cp_reference, "fault", io); + cl_page_put(env, page); + io->u.ci_fault.ft_page = NULL; + } + vvp_io_fini(env, ios); +} + +static enum cl_lock_mode vvp_mode_from_vma(struct vm_area_struct *vma) +{ + /* + * we only want to hold PW locks if the mmap() can generate + * writes back to the file and that only happens in shared + * writable vmas + */ + if ((vma->vm_flags & VM_SHARED) && (vma->vm_flags & VM_WRITE)) + return CLM_WRITE; + return CLM_READ; +} + +static int vvp_mmap_locks(const struct lu_env *env, + struct ccc_io *vio, struct cl_io *io) +{ + struct ccc_thread_info *cti = ccc_env_info(env); + struct mm_struct *mm = current->mm; + struct vm_area_struct *vma; + struct cl_lock_descr *descr = &cti->cti_descr; + ldlm_policy_data_t policy; + unsigned long addr; + ssize_t count; + int result; + struct iov_iter i; + struct iovec iov; + + LASSERT(io->ci_type == CIT_READ || io->ci_type == CIT_WRITE); + + if (!cl_is_normalio(env, io)) + return 0; + + if (vio->cui_iter == NULL) /* nfs or loop back device write */ + return 0; + + /* No MM (e.g. NFS)? No vmas too. */ + if (mm == NULL) + return 0; + + iov_for_each(iov, i, *(vio->cui_iter)) { + addr = (unsigned long)iov.iov_base; + count = iov.iov_len; + if (count == 0) + continue; + + count += addr & (~CFS_PAGE_MASK); + addr &= CFS_PAGE_MASK; + + down_read(&mm->mmap_sem); + while ((vma = our_vma(mm, addr, count)) != NULL) { + struct inode *inode = file_inode(vma->vm_file); + int flags = CEF_MUST; + + if (ll_file_nolock(vma->vm_file)) { + /* + * For no lock case, a lockless lock will be + * generated. + */ + flags = CEF_NEVER; + } + + /* + * XXX: Required lock mode can be weakened: CIT_WRITE + * io only ever reads user level buffer, and CIT_READ + * only writes on it. + */ + policy_from_vma(&policy, vma, addr, count); + descr->cld_mode = vvp_mode_from_vma(vma); + descr->cld_obj = ll_i2info(inode)->lli_clob; + descr->cld_start = cl_index(descr->cld_obj, + policy.l_extent.start); + descr->cld_end = cl_index(descr->cld_obj, + policy.l_extent.end); + descr->cld_enq_flags = flags; + result = cl_io_lock_alloc_add(env, io, descr); + + CDEBUG(D_VFSTRACE, "lock: %d: [%lu, %lu]\n", + descr->cld_mode, descr->cld_start, + descr->cld_end); + + if (result < 0) { + up_read(&mm->mmap_sem); + return result; + } + + if (vma->vm_end - addr >= count) + break; + + count -= vma->vm_end - addr; + addr = vma->vm_end; + } + up_read(&mm->mmap_sem); + } + return 0; +} + +static int vvp_io_rw_lock(const struct lu_env *env, struct cl_io *io, + enum cl_lock_mode mode, loff_t start, loff_t end) +{ + struct ccc_io *cio = ccc_env_io(env); + int result; + int ast_flags = 0; + + LASSERT(io->ci_type == CIT_READ || io->ci_type == CIT_WRITE); + + ccc_io_update_iov(env, cio, io); + + if (io->u.ci_rw.crw_nonblock) + ast_flags |= CEF_NONBLOCK; + result = vvp_mmap_locks(env, cio, io); + if (result == 0) + result = ccc_io_one_lock(env, io, ast_flags, mode, start, end); + return result; +} + +static int vvp_io_read_lock(const struct lu_env *env, + const struct cl_io_slice *ios) +{ + struct cl_io *io = ios->cis_io; + struct cl_io_rw_common *rd = &io->u.ci_rd.rd; + int result; + + result = vvp_io_rw_lock(env, io, CLM_READ, rd->crw_pos, + rd->crw_pos + rd->crw_count - 1); + + return result; +} + +static int vvp_io_fault_lock(const struct lu_env *env, + const struct cl_io_slice *ios) +{ + struct cl_io *io = ios->cis_io; + struct vvp_io *vio = cl2vvp_io(env, ios); + /* + * XXX LDLM_FL_CBPENDING + */ + return ccc_io_one_lock_index + (env, io, 0, vvp_mode_from_vma(vio->u.fault.ft_vma), + io->u.ci_fault.ft_index, io->u.ci_fault.ft_index); +} + +static int vvp_io_write_lock(const struct lu_env *env, + const struct cl_io_slice *ios) +{ + struct cl_io *io = ios->cis_io; + loff_t start; + loff_t end; + + if (io->u.ci_wr.wr_append) { + start = 0; + end = OBD_OBJECT_EOF; + } else { + start = io->u.ci_wr.wr.crw_pos; + end = start + io->u.ci_wr.wr.crw_count - 1; + } + return vvp_io_rw_lock(env, io, CLM_WRITE, start, end); +} + +static int vvp_io_setattr_iter_init(const struct lu_env *env, + const struct cl_io_slice *ios) +{ + return 0; +} + +/** + * Implementation of cl_io_operations::cio_lock() method for CIT_SETATTR io. + * + * Handles "lockless io" mode when extent locking is done by server. + */ +static int vvp_io_setattr_lock(const struct lu_env *env, + const struct cl_io_slice *ios) +{ + struct ccc_io *cio = ccc_env_io(env); + struct cl_io *io = ios->cis_io; + __u64 new_size; + __u32 enqflags = 0; + + if (cl_io_is_trunc(io)) { + new_size = io->u.ci_setattr.sa_attr.lvb_size; + if (new_size == 0) + enqflags = CEF_DISCARD_DATA; + } else { + if ((io->u.ci_setattr.sa_attr.lvb_mtime >= + io->u.ci_setattr.sa_attr.lvb_ctime) || + (io->u.ci_setattr.sa_attr.lvb_atime >= + io->u.ci_setattr.sa_attr.lvb_ctime)) + return 0; + new_size = 0; + } + cio->u.setattr.cui_local_lock = SETATTR_EXTENT_LOCK; + return ccc_io_one_lock(env, io, enqflags, CLM_WRITE, + new_size, OBD_OBJECT_EOF); +} + +static int vvp_do_vmtruncate(struct inode *inode, size_t size) +{ + int result; + /* + * Only ll_inode_size_lock is taken at this level. + */ + ll_inode_size_lock(inode); + result = inode_newsize_ok(inode, size); + if (result < 0) { + ll_inode_size_unlock(inode); + return result; + } + truncate_setsize(inode, size); + ll_inode_size_unlock(inode); + return result; +} + +static int vvp_io_setattr_trunc(const struct lu_env *env, + const struct cl_io_slice *ios, + struct inode *inode, loff_t size) +{ + inode_dio_wait(inode); + return 0; +} + +static int vvp_io_setattr_time(const struct lu_env *env, + const struct cl_io_slice *ios) +{ + struct cl_io *io = ios->cis_io; + struct cl_object *obj = io->ci_obj; + struct cl_attr *attr = ccc_env_thread_attr(env); + int result; + unsigned valid = CAT_CTIME; + + cl_object_attr_lock(obj); + attr->cat_ctime = io->u.ci_setattr.sa_attr.lvb_ctime; + if (io->u.ci_setattr.sa_valid & ATTR_ATIME_SET) { + attr->cat_atime = io->u.ci_setattr.sa_attr.lvb_atime; + valid |= CAT_ATIME; + } + if (io->u.ci_setattr.sa_valid & ATTR_MTIME_SET) { + attr->cat_mtime = io->u.ci_setattr.sa_attr.lvb_mtime; + valid |= CAT_MTIME; + } + result = cl_object_attr_set(env, obj, attr, valid); + cl_object_attr_unlock(obj); + + return result; +} + +static int vvp_io_setattr_start(const struct lu_env *env, + const struct cl_io_slice *ios) +{ + struct cl_io *io = ios->cis_io; + struct inode *inode = ccc_object_inode(io->ci_obj); + int result = 0; + + mutex_lock(&inode->i_mutex); + if (cl_io_is_trunc(io)) + result = vvp_io_setattr_trunc(env, ios, inode, + io->u.ci_setattr.sa_attr.lvb_size); + if (result == 0) + result = vvp_io_setattr_time(env, ios); + return result; +} + +static void vvp_io_setattr_end(const struct lu_env *env, + const struct cl_io_slice *ios) +{ + struct cl_io *io = ios->cis_io; + struct inode *inode = ccc_object_inode(io->ci_obj); + + if (cl_io_is_trunc(io)) { + /* Truncate in memory pages - they must be clean pages + * because osc has already notified to destroy osc_extents. */ + vvp_do_vmtruncate(inode, io->u.ci_setattr.sa_attr.lvb_size); + inode_dio_write_done(inode); + } + mutex_unlock(&inode->i_mutex); +} + +static void vvp_io_setattr_fini(const struct lu_env *env, + const struct cl_io_slice *ios) +{ + vvp_io_fini(env, ios); +} + +static int vvp_io_read_start(const struct lu_env *env, + const struct cl_io_slice *ios) +{ + struct vvp_io *vio = cl2vvp_io(env, ios); + struct ccc_io *cio = cl2ccc_io(env, ios); + struct cl_io *io = ios->cis_io; + struct cl_object *obj = io->ci_obj; + struct inode *inode = ccc_object_inode(obj); + struct ll_ra_read *bead = &vio->cui_bead; + struct file *file = cio->cui_fd->fd_file; + + int result; + loff_t pos = io->u.ci_rd.rd.crw_pos; + long cnt = io->u.ci_rd.rd.crw_count; + long tot = cio->cui_tot_count; + int exceed = 0; + + CLOBINVRNT(env, obj, ccc_object_invariant(obj)); + + CDEBUG(D_VFSTRACE, "read: -> [%lli, %lli)\n", pos, pos + cnt); + + if (!can_populate_pages(env, io, inode)) + return 0; + + result = ccc_prep_size(env, obj, io, pos, tot, &exceed); + if (result != 0) + return result; + else if (exceed != 0) + goto out; + + LU_OBJECT_HEADER(D_INODE, env, &obj->co_lu, + "Read ino %lu, %lu bytes, offset %lld, size %llu\n", + inode->i_ino, cnt, pos, i_size_read(inode)); + + /* turn off the kernel's read-ahead */ + cio->cui_fd->fd_file->f_ra.ra_pages = 0; + + /* initialize read-ahead window once per syscall */ + if (!vio->cui_ra_window_set) { + vio->cui_ra_window_set = 1; + bead->lrr_start = cl_index(obj, pos); + /* + * XXX: explicit PAGE_CACHE_SIZE + */ + bead->lrr_count = cl_index(obj, tot + PAGE_CACHE_SIZE - 1); + ll_ra_read_in(file, bead); + } + + /* BUG: 5972 */ + file_accessed(file); + switch (vio->cui_io_subtype) { + case IO_NORMAL: + LASSERT(cio->cui_iocb->ki_pos == pos); + result = generic_file_read_iter(cio->cui_iocb, cio->cui_iter); + break; + case IO_SPLICE: + result = generic_file_splice_read(file, &pos, + vio->u.splice.cui_pipe, cnt, + vio->u.splice.cui_flags); + /* LU-1109: do splice read stripe by stripe otherwise if it + * may make nfsd stuck if this read occupied all internal pipe + * buffers. */ + io->ci_continue = 0; + break; + default: + CERROR("Wrong IO type %u\n", vio->cui_io_subtype); + LBUG(); + } + +out: + if (result >= 0) { + if (result < cnt) + io->ci_continue = 0; + io->ci_nob += result; + ll_rw_stats_tally(ll_i2sbi(inode), current->pid, + cio->cui_fd, pos, result, READ); + result = 0; + } + return result; +} + +static void vvp_io_read_fini(const struct lu_env *env, const struct cl_io_slice *ios) +{ + struct vvp_io *vio = cl2vvp_io(env, ios); + struct ccc_io *cio = cl2ccc_io(env, ios); + + if (vio->cui_ra_window_set) + ll_ra_read_ex(cio->cui_fd->fd_file, &vio->cui_bead); + + vvp_io_fini(env, ios); +} + +static int vvp_io_write_start(const struct lu_env *env, + const struct cl_io_slice *ios) +{ + struct ccc_io *cio = cl2ccc_io(env, ios); + struct cl_io *io = ios->cis_io; + struct cl_object *obj = io->ci_obj; + struct inode *inode = ccc_object_inode(obj); + ssize_t result = 0; + loff_t pos = io->u.ci_wr.wr.crw_pos; + size_t cnt = io->u.ci_wr.wr.crw_count; + + if (!can_populate_pages(env, io, inode)) + return 0; + + if (cl_io_is_append(io)) { + /* + * PARALLEL IO This has to be changed for parallel IO doing + * out-of-order writes. + */ + pos = io->u.ci_wr.wr.crw_pos = i_size_read(inode); + cio->cui_iocb->ki_pos = pos; + } else { + LASSERT(cio->cui_iocb->ki_pos == pos); + } + + CDEBUG(D_VFSTRACE, "write: [%lli, %lli)\n", pos, pos + (long long)cnt); + + if (cio->cui_iter == NULL) /* from a temp io in ll_cl_init(). */ + result = 0; + else + result = generic_file_write_iter(cio->cui_iocb, cio->cui_iter); + + if (result > 0) { + if (result < cnt) + io->ci_continue = 0; + io->ci_nob += result; + ll_rw_stats_tally(ll_i2sbi(inode), current->pid, + cio->cui_fd, pos, result, WRITE); + result = 0; + } + return result; +} + +static int vvp_io_kernel_fault(struct vvp_fault_io *cfio) +{ + struct vm_fault *vmf = cfio->fault.ft_vmf; + + cfio->fault.ft_flags = filemap_fault(cfio->ft_vma, vmf); + cfio->fault.ft_flags_valid = 1; + + if (vmf->page) { + CDEBUG(D_PAGE, + "page %p map %p index %lu flags %lx count %u priv %0lx: got addr %p type NOPAGE\n", + vmf->page, vmf->page->mapping, vmf->page->index, + (long)vmf->page->flags, page_count(vmf->page), + page_private(vmf->page), vmf->virtual_address); + if (unlikely(!(cfio->fault.ft_flags & VM_FAULT_LOCKED))) { + lock_page(vmf->page); + cfio->fault.ft_flags |= VM_FAULT_LOCKED; + } + + cfio->ft_vmpage = vmf->page; + return 0; + } + + if (cfio->fault.ft_flags & (VM_FAULT_SIGBUS | VM_FAULT_SIGSEGV)) { + CDEBUG(D_PAGE, "got addr %p - SIGBUS\n", vmf->virtual_address); + return -EFAULT; + } + + if (cfio->fault.ft_flags & VM_FAULT_OOM) { + CDEBUG(D_PAGE, "got addr %p - OOM\n", vmf->virtual_address); + return -ENOMEM; + } + + if (cfio->fault.ft_flags & VM_FAULT_RETRY) + return -EAGAIN; + + CERROR("Unknown error in page fault %d!\n", cfio->fault.ft_flags); + return -EINVAL; +} + + +static int vvp_io_fault_start(const struct lu_env *env, + const struct cl_io_slice *ios) +{ + struct vvp_io *vio = cl2vvp_io(env, ios); + struct cl_io *io = ios->cis_io; + struct cl_object *obj = io->ci_obj; + struct inode *inode = ccc_object_inode(obj); + struct cl_fault_io *fio = &io->u.ci_fault; + struct vvp_fault_io *cfio = &vio->u.fault; + loff_t offset; + int result = 0; + struct page *vmpage = NULL; + struct cl_page *page; + loff_t size; + pgoff_t last; /* last page in a file data region */ + + if (fio->ft_executable && + LTIME_S(inode->i_mtime) != vio->u.fault.ft_mtime) + CWARN("binary "DFID + " changed while waiting for the page fault lock\n", + PFID(lu_object_fid(&obj->co_lu))); + + /* offset of the last byte on the page */ + offset = cl_offset(obj, fio->ft_index + 1) - 1; + LASSERT(cl_index(obj, offset) == fio->ft_index); + result = ccc_prep_size(env, obj, io, 0, offset + 1, NULL); + if (result != 0) + return result; + + /* must return locked page */ + if (fio->ft_mkwrite) { + LASSERT(cfio->ft_vmpage != NULL); + lock_page(cfio->ft_vmpage); + } else { + result = vvp_io_kernel_fault(cfio); + if (result != 0) + return result; + } + + vmpage = cfio->ft_vmpage; + LASSERT(PageLocked(vmpage)); + + if (OBD_FAIL_CHECK(OBD_FAIL_LLITE_FAULT_TRUNC_RACE)) + ll_invalidate_page(vmpage); + + size = i_size_read(inode); + /* Though we have already held a cl_lock upon this page, but + * it still can be truncated locally. */ + if (unlikely((vmpage->mapping != inode->i_mapping) || + (page_offset(vmpage) > size))) { + CDEBUG(D_PAGE, "llite: fault and truncate race happened!\n"); + + /* return +1 to stop cl_io_loop() and ll_fault() will catch + * and retry. */ + result = +1; + goto out; + } + + + if (fio->ft_mkwrite) { + pgoff_t last_index; + /* + * Capture the size while holding the lli_trunc_sem from above + * we want to make sure that we complete the mkwrite action + * while holding this lock. We need to make sure that we are + * not past the end of the file. + */ + last_index = cl_index(obj, size - 1); + if (last_index < fio->ft_index) { + CDEBUG(D_PAGE, + "llite: mkwrite and truncate race happened: %p: 0x%lx 0x%lx\n", + vmpage->mapping, fio->ft_index, last_index); + /* + * We need to return if we are + * passed the end of the file. This will propagate + * up the call stack to ll_page_mkwrite where + * we will return VM_FAULT_NOPAGE. Any non-negative + * value returned here will be silently + * converted to 0. If the vmpage->mapping is null + * the error code would be converted back to ENODATA + * in ll_page_mkwrite0. Thus we return -ENODATA + * to handle both cases + */ + result = -ENODATA; + goto out; + } + } + + page = cl_page_find(env, obj, fio->ft_index, vmpage, CPT_CACHEABLE); + if (IS_ERR(page)) { + result = PTR_ERR(page); + goto out; + } + + /* if page is going to be written, we should add this page into cache + * earlier. */ + if (fio->ft_mkwrite) { + wait_on_page_writeback(vmpage); + if (set_page_dirty(vmpage)) { + struct ccc_page *cp; + + /* vvp_page_assume() calls wait_on_page_writeback(). */ + cl_page_assume(env, io, page); + + cp = cl2ccc_page(cl_page_at(page, &vvp_device_type)); + vvp_write_pending(cl2ccc(obj), cp); + + /* Do not set Dirty bit here so that in case IO is + * started before the page is really made dirty, we + * still have chance to detect it. */ + result = cl_page_cache_add(env, io, page, CRT_WRITE); + LASSERT(cl_page_is_owned(page, io)); + + vmpage = NULL; + if (result < 0) { + cl_page_unmap(env, io, page); + cl_page_discard(env, io, page); + cl_page_disown(env, io, page); + + cl_page_put(env, page); + + /* we're in big trouble, what can we do now? */ + if (result == -EDQUOT) + result = -ENOSPC; + goto out; + } else + cl_page_disown(env, io, page); + } + } + + last = cl_index(obj, size - 1); + /* + * The ft_index is only used in the case of + * a mkwrite action. We need to check + * our assertions are correct, since + * we should have caught this above + */ + LASSERT(!fio->ft_mkwrite || fio->ft_index <= last); + if (fio->ft_index == last) + /* + * Last page is mapped partially. + */ + fio->ft_nob = size - cl_offset(obj, fio->ft_index); + else + fio->ft_nob = cl_page_size(obj); + + lu_ref_add(&page->cp_reference, "fault", io); + fio->ft_page = page; + +out: + /* return unlocked vmpage to avoid deadlocking */ + if (vmpage != NULL) + unlock_page(vmpage); + cfio->fault.ft_flags &= ~VM_FAULT_LOCKED; + return result; +} + +static int vvp_io_fsync_start(const struct lu_env *env, + const struct cl_io_slice *ios) +{ + /* we should mark TOWRITE bit to each dirty page in radix tree to + * verify pages have been written, but this is difficult because of + * race. */ + return 0; +} + +static int vvp_io_read_page(const struct lu_env *env, + const struct cl_io_slice *ios, + const struct cl_page_slice *slice) +{ + struct cl_io *io = ios->cis_io; + struct cl_object *obj = slice->cpl_obj; + struct ccc_page *cp = cl2ccc_page(slice); + struct cl_page *page = slice->cpl_page; + struct inode *inode = ccc_object_inode(obj); + struct ll_sb_info *sbi = ll_i2sbi(inode); + struct ll_file_data *fd = cl2ccc_io(env, ios)->cui_fd; + struct ll_readahead_state *ras = &fd->fd_ras; + struct page *vmpage = cp->cpg_page; + struct cl_2queue *queue = &io->ci_queue; + int rc; + + CLOBINVRNT(env, obj, ccc_object_invariant(obj)); + LASSERT(slice->cpl_obj == obj); + + if (sbi->ll_ra_info.ra_max_pages_per_file && + sbi->ll_ra_info.ra_max_pages) + ras_update(sbi, inode, ras, page->cp_index, + cp->cpg_defer_uptodate); + + /* Sanity check whether the page is protected by a lock. */ + rc = cl_page_is_under_lock(env, io, page); + if (rc != -EBUSY) { + CL_PAGE_HEADER(D_WARNING, env, page, "%s: %d\n", + rc == -ENODATA ? "without a lock" : + "match failed", rc); + if (rc != -ENODATA) + return rc; + } + + if (cp->cpg_defer_uptodate) { + cp->cpg_ra_used = 1; + cl_page_export(env, page, 1); + } + /* + * Add page into the queue even when it is marked uptodate above. + * this will unlock it automatically as part of cl_page_list_disown(). + */ + cl_2queue_add(queue, page); + if (sbi->ll_ra_info.ra_max_pages_per_file && + sbi->ll_ra_info.ra_max_pages) + ll_readahead(env, io, ras, + vmpage->mapping, &queue->c2_qin, fd->fd_flags); + + return 0; +} + +static int vvp_page_sync_io(const struct lu_env *env, struct cl_io *io, + struct cl_page *page, struct ccc_page *cp, + enum cl_req_type crt) +{ + struct cl_2queue *queue; + int result; + + LASSERT(io->ci_type == CIT_READ || io->ci_type == CIT_WRITE); + + queue = &io->ci_queue; + cl_2queue_init_page(queue, page); + + result = cl_io_submit_sync(env, io, crt, queue, 0); + LASSERT(cl_page_is_owned(page, io)); + + if (crt == CRT_READ) + /* + * in CRT_WRITE case page is left locked even in case of + * error. + */ + cl_page_list_disown(env, io, &queue->c2_qin); + cl_2queue_fini(env, queue); + + return result; +} + +/** + * Prepare partially written-to page for a write. + */ +static int vvp_io_prepare_partial(const struct lu_env *env, struct cl_io *io, + struct cl_object *obj, struct cl_page *pg, + struct ccc_page *cp, + unsigned from, unsigned to) +{ + struct cl_attr *attr = ccc_env_thread_attr(env); + loff_t offset = cl_offset(obj, pg->cp_index); + int result; + + cl_object_attr_lock(obj); + result = cl_object_attr_get(env, obj, attr); + cl_object_attr_unlock(obj); + if (result == 0) { + /* + * If are writing to a new page, no need to read old data. + * The extent locking will have updated the KMS, and for our + * purposes here we can treat it like i_size. + */ + if (attr->cat_kms <= offset) { + char *kaddr = kmap_atomic(cp->cpg_page); + + memset(kaddr, 0, cl_page_size(obj)); + kunmap_atomic(kaddr); + } else if (cp->cpg_defer_uptodate) + cp->cpg_ra_used = 1; + else + result = vvp_page_sync_io(env, io, pg, cp, CRT_READ); + /* + * In older implementations, obdo_refresh_inode is called here + * to update the inode because the write might modify the + * object info at OST. However, this has been proven useless, + * since LVB functions will be called when user space program + * tries to retrieve inode attribute. Also, see bug 15909 for + * details. -jay + */ + if (result == 0) + cl_page_export(env, pg, 1); + } + return result; +} + +static int vvp_io_prepare_write(const struct lu_env *env, + const struct cl_io_slice *ios, + const struct cl_page_slice *slice, + unsigned from, unsigned to) +{ + struct cl_object *obj = slice->cpl_obj; + struct ccc_page *cp = cl2ccc_page(slice); + struct cl_page *pg = slice->cpl_page; + struct page *vmpage = cp->cpg_page; + + int result; + + LINVRNT(cl_page_is_vmlocked(env, pg)); + LASSERT(vmpage->mapping->host == ccc_object_inode(obj)); + + result = 0; + + CL_PAGE_HEADER(D_PAGE, env, pg, "preparing: [%d, %d]\n", from, to); + if (!PageUptodate(vmpage)) { + /* + * We're completely overwriting an existing page, so _don't_ + * set it up to date until commit_write + */ + if (from == 0 && to == PAGE_CACHE_SIZE) { + CL_PAGE_HEADER(D_PAGE, env, pg, "full page write\n"); + POISON_PAGE(page, 0x11); + } else + result = vvp_io_prepare_partial(env, ios->cis_io, obj, + pg, cp, from, to); + } else + CL_PAGE_HEADER(D_PAGE, env, pg, "uptodate\n"); + return result; +} + +static int vvp_io_commit_write(const struct lu_env *env, + const struct cl_io_slice *ios, + const struct cl_page_slice *slice, + unsigned from, unsigned to) +{ + struct cl_object *obj = slice->cpl_obj; + struct cl_io *io = ios->cis_io; + struct ccc_page *cp = cl2ccc_page(slice); + struct cl_page *pg = slice->cpl_page; + struct inode *inode = ccc_object_inode(obj); + struct ll_sb_info *sbi = ll_i2sbi(inode); + struct ll_inode_info *lli = ll_i2info(inode); + struct page *vmpage = cp->cpg_page; + + int result; + int tallyop; + loff_t size; + + LINVRNT(cl_page_is_vmlocked(env, pg)); + LASSERT(vmpage->mapping->host == inode); + + LU_OBJECT_HEADER(D_INODE, env, &obj->co_lu, "committing page write\n"); + CL_PAGE_HEADER(D_PAGE, env, pg, "committing: [%d, %d]\n", from, to); + + /* + * queue a write for some time in the future the first time we + * dirty the page. + * + * This is different from what other file systems do: they usually + * just mark page (and some of its buffers) dirty and rely on + * balance_dirty_pages() to start a write-back. Lustre wants write-back + * to be started earlier for the following reasons: + * + * (1) with a large number of clients we need to limit the amount + * of cached data on the clients a lot; + * + * (2) large compute jobs generally want compute-only then io-only + * and the IO should complete as quickly as possible; + * + * (3) IO is batched up to the RPC size and is async until the + * client max cache is hit + * (/proc/fs/lustre/osc/OSC.../max_dirty_mb) + * + */ + if (!PageDirty(vmpage)) { + tallyop = LPROC_LL_DIRTY_MISSES; + result = cl_page_cache_add(env, io, pg, CRT_WRITE); + if (result == 0) { + /* page was added into cache successfully. */ + set_page_dirty(vmpage); + vvp_write_pending(cl2ccc(obj), cp); + } else if (result == -EDQUOT) { + pgoff_t last_index = i_size_read(inode) >> PAGE_CACHE_SHIFT; + bool need_clip = true; + + /* + * Client ran out of disk space grant. Possible + * strategies are: + * + * (a) do a sync write, renewing grant; + * + * (b) stop writing on this stripe, switch to the + * next one. + * + * (b) is a part of "parallel io" design that is the + * ultimate goal. (a) is what "old" client did, and + * what the new code continues to do for the time + * being. + */ + if (last_index > pg->cp_index) { + to = PAGE_CACHE_SIZE; + need_clip = false; + } else if (last_index == pg->cp_index) { + int size_to = i_size_read(inode) & ~CFS_PAGE_MASK; + if (to < size_to) + to = size_to; + } + if (need_clip) + cl_page_clip(env, pg, 0, to); + result = vvp_page_sync_io(env, io, pg, cp, CRT_WRITE); + if (result) + CERROR("Write page %lu of inode %p failed %d\n", + pg->cp_index, inode, result); + } + } else { + tallyop = LPROC_LL_DIRTY_HITS; + result = 0; + } + ll_stats_ops_tally(sbi, tallyop, 1); + + /* Inode should be marked DIRTY even if no new page was marked DIRTY + * because page could have been not flushed between 2 modifications. + * It is important the file is marked DIRTY as soon as the I/O is done + * Indeed, when cache is flushed, file could be already closed and it + * is too late to warn the MDT. + * It is acceptable that file is marked DIRTY even if I/O is dropped + * for some reasons before being flushed to OST. + */ + if (result == 0) { + spin_lock(&lli->lli_lock); + lli->lli_flags |= LLIF_DATA_MODIFIED; + spin_unlock(&lli->lli_lock); + } + + size = cl_offset(obj, pg->cp_index) + to; + + ll_inode_size_lock(inode); + if (result == 0) { + if (size > i_size_read(inode)) { + cl_isize_write_nolock(inode, size); + CDEBUG(D_VFSTRACE, DFID" updating i_size %lu\n", + PFID(lu_object_fid(&obj->co_lu)), + (unsigned long)size); + } + cl_page_export(env, pg, 1); + } else { + if (size > i_size_read(inode)) + cl_page_discard(env, io, pg); + } + ll_inode_size_unlock(inode); + return result; +} + +static const struct cl_io_operations vvp_io_ops = { + .op = { + [CIT_READ] = { + .cio_fini = vvp_io_read_fini, + .cio_lock = vvp_io_read_lock, + .cio_start = vvp_io_read_start, + .cio_advance = ccc_io_advance + }, + [CIT_WRITE] = { + .cio_fini = vvp_io_fini, + .cio_lock = vvp_io_write_lock, + .cio_start = vvp_io_write_start, + .cio_advance = ccc_io_advance + }, + [CIT_SETATTR] = { + .cio_fini = vvp_io_setattr_fini, + .cio_iter_init = vvp_io_setattr_iter_init, + .cio_lock = vvp_io_setattr_lock, + .cio_start = vvp_io_setattr_start, + .cio_end = vvp_io_setattr_end + }, + [CIT_FAULT] = { + .cio_fini = vvp_io_fault_fini, + .cio_iter_init = vvp_io_fault_iter_init, + .cio_lock = vvp_io_fault_lock, + .cio_start = vvp_io_fault_start, + .cio_end = ccc_io_end + }, + [CIT_FSYNC] = { + .cio_start = vvp_io_fsync_start, + .cio_fini = vvp_io_fini + }, + [CIT_MISC] = { + .cio_fini = vvp_io_fini + } + }, + .cio_read_page = vvp_io_read_page, + .cio_prepare_write = vvp_io_prepare_write, + .cio_commit_write = vvp_io_commit_write +}; + +int vvp_io_init(const struct lu_env *env, struct cl_object *obj, + struct cl_io *io) +{ + struct vvp_io *vio = vvp_env_io(env); + struct ccc_io *cio = ccc_env_io(env); + struct inode *inode = ccc_object_inode(obj); + int result; + + CLOBINVRNT(env, obj, ccc_object_invariant(obj)); + + CDEBUG(D_VFSTRACE, DFID + " ignore/verify layout %d/%d, layout version %d restore needed %d\n", + PFID(lu_object_fid(&obj->co_lu)), + io->ci_ignore_layout, io->ci_verify_layout, + cio->cui_layout_gen, io->ci_restore_needed); + + CL_IO_SLICE_CLEAN(cio, cui_cl); + cl_io_slice_add(io, &cio->cui_cl, obj, &vvp_io_ops); + vio->cui_ra_window_set = 0; + result = 0; + if (io->ci_type == CIT_READ || io->ci_type == CIT_WRITE) { + size_t count; + struct ll_inode_info *lli = ll_i2info(inode); + + count = io->u.ci_rw.crw_count; + /* "If nbyte is 0, read() will return 0 and have no other + * results." -- Single Unix Spec */ + if (count == 0) + result = 1; + else + cio->cui_tot_count = count; + + /* for read/write, we store the jobid in the inode, and + * it'll be fetched by osc when building RPC. + * + * it's not accurate if the file is shared by different + * jobs. + */ + lustre_get_jobid(lli->lli_jobid); + } else if (io->ci_type == CIT_SETATTR) { + if (!cl_io_is_trunc(io)) + io->ci_lockreq = CILR_MANDATORY; + } + + /* ignore layout change for generic CIT_MISC but not for glimpse. + * io context for glimpse must set ci_verify_layout to true, + * see cl_glimpse_size0() for details. */ + if (io->ci_type == CIT_MISC && !io->ci_verify_layout) + io->ci_ignore_layout = 1; + + /* Enqueue layout lock and get layout version. We need to do this + * even for operations requiring to open file, such as read and write, + * because it might not grant layout lock in IT_OPEN. */ + if (result == 0 && !io->ci_ignore_layout) { + result = ll_layout_refresh(inode, &cio->cui_layout_gen); + if (result == -ENOENT) + /* If the inode on MDS has been removed, but the objects + * on OSTs haven't been destroyed (async unlink), layout + * fetch will return -ENOENT, we'd ignore this error + * and continue with dirty flush. LU-3230. */ + result = 0; + if (result < 0) + CERROR("%s: refresh file layout " DFID " error %d.\n", + ll_get_fsname(inode->i_sb, NULL, 0), + PFID(lu_object_fid(&obj->co_lu)), result); + } + + return result; +} + +static struct vvp_io *cl2vvp_io(const struct lu_env *env, + const struct cl_io_slice *slice) +{ + /* Calling just for assertion */ + cl2ccc_io(env, slice); + return vvp_env_io(env); +} |