^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1) // SPDX-License-Identifier: GPL-2.0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3) * linux/fs/ext4/page-io.c
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 5) * This contains the new page_io functions for ext4
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 6) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 7) * Written by Theodore Ts'o, 2010.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 8) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 9)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 10) #include <linux/fs.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 11) #include <linux/time.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 12) #include <linux/highuid.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 13) #include <linux/pagemap.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 14) #include <linux/quotaops.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 15) #include <linux/string.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 16) #include <linux/buffer_head.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 17) #include <linux/writeback.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 18) #include <linux/pagevec.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 19) #include <linux/mpage.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 20) #include <linux/namei.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 21) #include <linux/uio.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 22) #include <linux/bio.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 23) #include <linux/workqueue.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 24) #include <linux/kernel.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 25) #include <linux/slab.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 26) #include <linux/mm.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 27) #include <linux/backing-dev.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 28)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 29) #include "ext4_jbd2.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 30) #include "xattr.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 31) #include "acl.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 32)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 33) static struct kmem_cache *io_end_cachep;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 34) static struct kmem_cache *io_end_vec_cachep;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 35)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 36) int __init ext4_init_pageio(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 37) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 38) io_end_cachep = KMEM_CACHE(ext4_io_end, SLAB_RECLAIM_ACCOUNT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 39) if (io_end_cachep == NULL)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 40) return -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 41)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 42) io_end_vec_cachep = KMEM_CACHE(ext4_io_end_vec, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 43) if (io_end_vec_cachep == NULL) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 44) kmem_cache_destroy(io_end_cachep);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 45) return -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 46) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 47) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 48) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 49)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 50) void ext4_exit_pageio(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 51) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 52) kmem_cache_destroy(io_end_cachep);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 53) kmem_cache_destroy(io_end_vec_cachep);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 54) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 55)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 56) struct ext4_io_end_vec *ext4_alloc_io_end_vec(ext4_io_end_t *io_end)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 57) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 58) struct ext4_io_end_vec *io_end_vec;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 59)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 60) io_end_vec = kmem_cache_zalloc(io_end_vec_cachep, GFP_NOFS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 61) if (!io_end_vec)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 62) return ERR_PTR(-ENOMEM);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 63) INIT_LIST_HEAD(&io_end_vec->list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 64) list_add_tail(&io_end_vec->list, &io_end->list_vec);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 65) return io_end_vec;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 66) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 67)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 68) static void ext4_free_io_end_vec(ext4_io_end_t *io_end)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 69) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 70) struct ext4_io_end_vec *io_end_vec, *tmp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 71)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 72) if (list_empty(&io_end->list_vec))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 73) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 74) list_for_each_entry_safe(io_end_vec, tmp, &io_end->list_vec, list) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 75) list_del(&io_end_vec->list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 76) kmem_cache_free(io_end_vec_cachep, io_end_vec);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 77) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 78) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 79)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 80) struct ext4_io_end_vec *ext4_last_io_end_vec(ext4_io_end_t *io_end)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 81) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 82) BUG_ON(list_empty(&io_end->list_vec));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 83) return list_last_entry(&io_end->list_vec, struct ext4_io_end_vec, list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 84) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 85)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 86) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 87) * Print an buffer I/O error compatible with the fs/buffer.c. This
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 88) * provides compatibility with dmesg scrapers that look for a specific
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 89) * buffer I/O error message. We really need a unified error reporting
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 90) * structure to userspace ala Digital Unix's uerf system, but it's
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 91) * probably not going to happen in my lifetime, due to LKML politics...
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 92) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 93) static void buffer_io_error(struct buffer_head *bh)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 94) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 95) printk_ratelimited(KERN_ERR "Buffer I/O error on device %pg, logical block %llu\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 96) bh->b_bdev,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 97) (unsigned long long)bh->b_blocknr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 98) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 99)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 100) static void ext4_finish_bio(struct bio *bio)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 101) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 102) struct bio_vec *bvec;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 103) struct bvec_iter_all iter_all;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 104)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 105) bio_for_each_segment_all(bvec, bio, iter_all) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 106) struct page *page = bvec->bv_page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 107) struct page *bounce_page = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 108) struct buffer_head *bh, *head;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 109) unsigned bio_start = bvec->bv_offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 110) unsigned bio_end = bio_start + bvec->bv_len;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 111) unsigned under_io = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 112) unsigned long flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 113)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 114) if (!page)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 115) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 116)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 117) if (fscrypt_is_bounce_page(page)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 118) bounce_page = page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 119) page = fscrypt_pagecache_page(bounce_page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 120) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 121)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 122) if (bio->bi_status) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 123) SetPageError(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 124) mapping_set_error(page->mapping, -EIO);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 125) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 126) bh = head = page_buffers(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 127) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 128) * We check all buffers in the page under b_uptodate_lock
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 129) * to avoid races with other end io clearing async_write flags
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 130) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 131) spin_lock_irqsave(&head->b_uptodate_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 132) do {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 133) if (bh_offset(bh) < bio_start ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 134) bh_offset(bh) + bh->b_size > bio_end) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 135) if (buffer_async_write(bh))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 136) under_io++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 137) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 138) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 139) clear_buffer_async_write(bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 140) if (bio->bi_status)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 141) buffer_io_error(bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 142) } while ((bh = bh->b_this_page) != head);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 143) spin_unlock_irqrestore(&head->b_uptodate_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 144) if (!under_io) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 145) fscrypt_free_bounce_page(bounce_page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 146) end_page_writeback(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 147) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 148) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 149) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 150)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 151) static void ext4_release_io_end(ext4_io_end_t *io_end)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 152) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 153) struct bio *bio, *next_bio;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 154)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 155) BUG_ON(!list_empty(&io_end->list));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 156) BUG_ON(io_end->flag & EXT4_IO_END_UNWRITTEN);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 157) WARN_ON(io_end->handle);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 158)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 159) for (bio = io_end->bio; bio; bio = next_bio) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 160) next_bio = bio->bi_private;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 161) ext4_finish_bio(bio);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 162) bio_put(bio);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 163) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 164) ext4_free_io_end_vec(io_end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 165) kmem_cache_free(io_end_cachep, io_end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 166) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 167)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 168) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 169) * Check a range of space and convert unwritten extents to written. Note that
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 170) * we are protected from truncate touching same part of extent tree by the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 171) * fact that truncate code waits for all DIO to finish (thus exclusion from
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 172) * direct IO is achieved) and also waits for PageWriteback bits. Thus we
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 173) * cannot get to ext4_ext_truncate() before all IOs overlapping that range are
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 174) * completed (happens from ext4_free_ioend()).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 175) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 176) static int ext4_end_io_end(ext4_io_end_t *io_end)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 177) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 178) struct inode *inode = io_end->inode;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 179) handle_t *handle = io_end->handle;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 180) int ret = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 181)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 182) ext4_debug("ext4_end_io_nolock: io_end 0x%p from inode %lu,list->next 0x%p,"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 183) "list->prev 0x%p\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 184) io_end, inode->i_ino, io_end->list.next, io_end->list.prev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 185)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 186) io_end->handle = NULL; /* Following call will use up the handle */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 187) ret = ext4_convert_unwritten_io_end_vec(handle, io_end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 188) if (ret < 0 && !ext4_forced_shutdown(EXT4_SB(inode->i_sb))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 189) ext4_msg(inode->i_sb, KERN_EMERG,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 190) "failed to convert unwritten extents to written "
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 191) "extents -- potential data loss! "
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 192) "(inode %lu, error %d)", inode->i_ino, ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 193) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 194) ext4_clear_io_unwritten_flag(io_end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 195) ext4_release_io_end(io_end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 196) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 197) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 198)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 199) static void dump_completed_IO(struct inode *inode, struct list_head *head)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 200) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 201) #ifdef EXT4FS_DEBUG
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 202) struct list_head *cur, *before, *after;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 203) ext4_io_end_t *io_end, *io_end0, *io_end1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 204)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 205) if (list_empty(head))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 206) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 207)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 208) ext4_debug("Dump inode %lu completed io list\n", inode->i_ino);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 209) list_for_each_entry(io_end, head, list) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 210) cur = &io_end->list;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 211) before = cur->prev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 212) io_end0 = container_of(before, ext4_io_end_t, list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 213) after = cur->next;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 214) io_end1 = container_of(after, ext4_io_end_t, list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 215)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 216) ext4_debug("io 0x%p from inode %lu,prev 0x%p,next 0x%p\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 217) io_end, inode->i_ino, io_end0, io_end1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 218) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 219) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 220) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 221)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 222) /* Add the io_end to per-inode completed end_io list. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 223) static void ext4_add_complete_io(ext4_io_end_t *io_end)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 224) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 225) struct ext4_inode_info *ei = EXT4_I(io_end->inode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 226) struct ext4_sb_info *sbi = EXT4_SB(io_end->inode->i_sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 227) struct workqueue_struct *wq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 228) unsigned long flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 229)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 230) /* Only reserved conversions from writeback should enter here */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 231) WARN_ON(!(io_end->flag & EXT4_IO_END_UNWRITTEN));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 232) WARN_ON(!io_end->handle && sbi->s_journal);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 233) spin_lock_irqsave(&ei->i_completed_io_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 234) wq = sbi->rsv_conversion_wq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 235) if (list_empty(&ei->i_rsv_conversion_list))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 236) queue_work(wq, &ei->i_rsv_conversion_work);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 237) list_add_tail(&io_end->list, &ei->i_rsv_conversion_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 238) spin_unlock_irqrestore(&ei->i_completed_io_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 239) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 240)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 241) static int ext4_do_flush_completed_IO(struct inode *inode,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 242) struct list_head *head)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 243) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 244) ext4_io_end_t *io_end;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 245) struct list_head unwritten;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 246) unsigned long flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 247) struct ext4_inode_info *ei = EXT4_I(inode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 248) int err, ret = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 249)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 250) spin_lock_irqsave(&ei->i_completed_io_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 251) dump_completed_IO(inode, head);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 252) list_replace_init(head, &unwritten);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 253) spin_unlock_irqrestore(&ei->i_completed_io_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 254)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 255) while (!list_empty(&unwritten)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 256) io_end = list_entry(unwritten.next, ext4_io_end_t, list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 257) BUG_ON(!(io_end->flag & EXT4_IO_END_UNWRITTEN));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 258) list_del_init(&io_end->list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 259)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 260) err = ext4_end_io_end(io_end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 261) if (unlikely(!ret && err))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 262) ret = err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 263) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 264) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 265) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 266)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 267) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 268) * work on completed IO, to convert unwritten extents to extents
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 269) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 270) void ext4_end_io_rsv_work(struct work_struct *work)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 271) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 272) struct ext4_inode_info *ei = container_of(work, struct ext4_inode_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 273) i_rsv_conversion_work);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 274) ext4_do_flush_completed_IO(&ei->vfs_inode, &ei->i_rsv_conversion_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 275) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 276)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 277) ext4_io_end_t *ext4_init_io_end(struct inode *inode, gfp_t flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 278) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 279) ext4_io_end_t *io_end = kmem_cache_zalloc(io_end_cachep, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 280)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 281) if (io_end) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 282) io_end->inode = inode;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 283) INIT_LIST_HEAD(&io_end->list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 284) INIT_LIST_HEAD(&io_end->list_vec);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 285) atomic_set(&io_end->count, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 286) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 287) return io_end;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 288) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 289)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 290) void ext4_put_io_end_defer(ext4_io_end_t *io_end)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 291) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 292) if (atomic_dec_and_test(&io_end->count)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 293) if (!(io_end->flag & EXT4_IO_END_UNWRITTEN) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 294) list_empty(&io_end->list_vec)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 295) ext4_release_io_end(io_end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 296) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 297) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 298) ext4_add_complete_io(io_end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 299) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 300) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 301)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 302) int ext4_put_io_end(ext4_io_end_t *io_end)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 303) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 304) int err = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 305)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 306) if (atomic_dec_and_test(&io_end->count)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 307) if (io_end->flag & EXT4_IO_END_UNWRITTEN) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 308) err = ext4_convert_unwritten_io_end_vec(io_end->handle,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 309) io_end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 310) io_end->handle = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 311) ext4_clear_io_unwritten_flag(io_end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 312) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 313) ext4_release_io_end(io_end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 314) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 315) return err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 316) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 317)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 318) ext4_io_end_t *ext4_get_io_end(ext4_io_end_t *io_end)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 319) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 320) atomic_inc(&io_end->count);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 321) return io_end;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 322) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 323)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 324) /* BIO completion function for page writeback */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 325) static void ext4_end_bio(struct bio *bio)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 326) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 327) ext4_io_end_t *io_end = bio->bi_private;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 328) sector_t bi_sector = bio->bi_iter.bi_sector;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 329) char b[BDEVNAME_SIZE];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 330)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 331) if (WARN_ONCE(!io_end, "io_end is NULL: %s: sector %Lu len %u err %d\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 332) bio_devname(bio, b),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 333) (long long) bio->bi_iter.bi_sector,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 334) (unsigned) bio_sectors(bio),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 335) bio->bi_status)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 336) ext4_finish_bio(bio);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 337) bio_put(bio);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 338) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 339) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 340) bio->bi_end_io = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 341)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 342) if (bio->bi_status) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 343) struct inode *inode = io_end->inode;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 344)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 345) ext4_warning(inode->i_sb, "I/O error %d writing to inode %lu "
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 346) "starting block %llu)",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 347) bio->bi_status, inode->i_ino,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 348) (unsigned long long)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 349) bi_sector >> (inode->i_blkbits - 9));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 350) mapping_set_error(inode->i_mapping,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 351) blk_status_to_errno(bio->bi_status));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 352) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 353)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 354) if (io_end->flag & EXT4_IO_END_UNWRITTEN) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 355) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 356) * Link bio into list hanging from io_end. We have to do it
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 357) * atomically as bio completions can be racing against each
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 358) * other.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 359) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 360) bio->bi_private = xchg(&io_end->bio, bio);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 361) ext4_put_io_end_defer(io_end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 362) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 363) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 364) * Drop io_end reference early. Inode can get freed once
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 365) * we finish the bio.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 366) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 367) ext4_put_io_end_defer(io_end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 368) ext4_finish_bio(bio);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 369) bio_put(bio);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 370) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 371) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 372)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 373) void ext4_io_submit(struct ext4_io_submit *io)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 374) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 375) struct bio *bio = io->io_bio;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 376)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 377) if (bio) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 378) int io_op_flags = io->io_wbc->sync_mode == WB_SYNC_ALL ?
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 379) REQ_SYNC : 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 380) io->io_bio->bi_write_hint = io->io_end->inode->i_write_hint;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 381) bio_set_op_attrs(io->io_bio, REQ_OP_WRITE, io_op_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 382) submit_bio(io->io_bio);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 383) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 384) io->io_bio = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 385) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 386)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 387) void ext4_io_submit_init(struct ext4_io_submit *io,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 388) struct writeback_control *wbc)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 389) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 390) io->io_wbc = wbc;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 391) io->io_bio = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 392) io->io_end = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 393) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 394)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 395) static void io_submit_init_bio(struct ext4_io_submit *io,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 396) struct buffer_head *bh)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 397) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 398) struct bio *bio;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 399)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 400) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 401) * bio_alloc will _always_ be able to allocate a bio if
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 402) * __GFP_DIRECT_RECLAIM is set, see comments for bio_alloc_bioset().
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 403) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 404) bio = bio_alloc(GFP_NOIO, BIO_MAX_PAGES);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 405) fscrypt_set_bio_crypt_ctx_bh(bio, bh, GFP_NOIO);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 406) bio->bi_iter.bi_sector = bh->b_blocknr * (bh->b_size >> 9);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 407) bio_set_dev(bio, bh->b_bdev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 408) bio->bi_end_io = ext4_end_bio;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 409) bio->bi_private = ext4_get_io_end(io->io_end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 410) io->io_bio = bio;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 411) io->io_next_block = bh->b_blocknr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 412) wbc_init_bio(io->io_wbc, bio);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 413) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 414)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 415) static void io_submit_add_bh(struct ext4_io_submit *io,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 416) struct inode *inode,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 417) struct page *page,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 418) struct buffer_head *bh)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 419) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 420) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 421)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 422) if (io->io_bio && (bh->b_blocknr != io->io_next_block ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 423) !fscrypt_mergeable_bio_bh(io->io_bio, bh))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 424) submit_and_retry:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 425) ext4_io_submit(io);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 426) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 427) if (io->io_bio == NULL) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 428) io_submit_init_bio(io, bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 429) io->io_bio->bi_write_hint = inode->i_write_hint;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 430) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 431) ret = bio_add_page(io->io_bio, page, bh->b_size, bh_offset(bh));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 432) if (ret != bh->b_size)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 433) goto submit_and_retry;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 434) wbc_account_cgroup_owner(io->io_wbc, page, bh->b_size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 435) io->io_next_block++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 436) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 437)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 438) int ext4_bio_write_page(struct ext4_io_submit *io,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 439) struct page *page,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 440) int len,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 441) struct writeback_control *wbc,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 442) bool keep_towrite)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 443) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 444) struct page *bounce_page = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 445) struct inode *inode = page->mapping->host;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 446) unsigned block_start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 447) struct buffer_head *bh, *head;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 448) int ret = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 449) int nr_submitted = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 450) int nr_to_submit = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 451)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 452) BUG_ON(!PageLocked(page));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 453) BUG_ON(PageWriteback(page));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 454)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 455) if (keep_towrite)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 456) set_page_writeback_keepwrite(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 457) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 458) set_page_writeback(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 459) ClearPageError(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 460)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 461) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 462) * Comments copied from block_write_full_page:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 463) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 464) * The page straddles i_size. It must be zeroed out on each and every
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 465) * writepage invocation because it may be mmapped. "A file is mapped
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 466) * in multiples of the page size. For a file that is not a multiple of
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 467) * the page size, the remaining memory is zeroed when mapped, and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 468) * writes to that region are not written out to the file."
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 469) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 470) if (len < PAGE_SIZE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 471) zero_user_segment(page, len, PAGE_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 472) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 473) * In the first loop we prepare and mark buffers to submit. We have to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 474) * mark all buffers in the page before submitting so that
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 475) * end_page_writeback() cannot be called from ext4_bio_end_io() when IO
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 476) * on the first buffer finishes and we are still working on submitting
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 477) * the second buffer.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 478) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 479) bh = head = page_buffers(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 480) do {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 481) block_start = bh_offset(bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 482) if (block_start >= len) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 483) clear_buffer_dirty(bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 484) set_buffer_uptodate(bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 485) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 486) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 487) if (!buffer_dirty(bh) || buffer_delay(bh) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 488) !buffer_mapped(bh) || buffer_unwritten(bh)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 489) /* A hole? We can safely clear the dirty bit */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 490) if (!buffer_mapped(bh))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 491) clear_buffer_dirty(bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 492) if (io->io_bio)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 493) ext4_io_submit(io);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 494) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 495) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 496) if (buffer_new(bh))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 497) clear_buffer_new(bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 498) set_buffer_async_write(bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 499) nr_to_submit++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 500) } while ((bh = bh->b_this_page) != head);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 501)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 502) bh = head = page_buffers(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 503)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 504) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 505) * If any blocks are being written to an encrypted file, encrypt them
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 506) * into a bounce page. For simplicity, just encrypt until the last
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 507) * block which might be needed. This may cause some unneeded blocks
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 508) * (e.g. holes) to be unnecessarily encrypted, but this is rare and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 509) * can't happen in the common case of blocksize == PAGE_SIZE.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 510) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 511) if (fscrypt_inode_uses_fs_layer_crypto(inode) && nr_to_submit) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 512) gfp_t gfp_flags = GFP_NOFS;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 513) unsigned int enc_bytes = round_up(len, i_blocksize(inode));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 514)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 515) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 516) * Since bounce page allocation uses a mempool, we can only use
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 517) * a waiting mask (i.e. request guaranteed allocation) on the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 518) * first page of the bio. Otherwise it can deadlock.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 519) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 520) if (io->io_bio)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 521) gfp_flags = GFP_NOWAIT | __GFP_NOWARN;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 522) retry_encrypt:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 523) bounce_page = fscrypt_encrypt_pagecache_blocks(page, enc_bytes,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 524) 0, gfp_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 525) if (IS_ERR(bounce_page)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 526) ret = PTR_ERR(bounce_page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 527) if (ret == -ENOMEM &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 528) (io->io_bio || wbc->sync_mode == WB_SYNC_ALL)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 529) gfp_flags = GFP_NOFS;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 530) if (io->io_bio)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 531) ext4_io_submit(io);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 532) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 533) gfp_flags |= __GFP_NOFAIL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 534) congestion_wait(BLK_RW_ASYNC, HZ/50);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 535) goto retry_encrypt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 536) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 537)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 538) printk_ratelimited(KERN_ERR "%s: ret = %d\n", __func__, ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 539) redirty_page_for_writepage(wbc, page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 540) do {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 541) clear_buffer_async_write(bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 542) bh = bh->b_this_page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 543) } while (bh != head);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 544) goto unlock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 545) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 546) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 547)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 548) /* Now submit buffers to write */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 549) do {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 550) if (!buffer_async_write(bh))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 551) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 552) io_submit_add_bh(io, inode,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 553) bounce_page ? bounce_page : page, bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 554) nr_submitted++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 555) clear_buffer_dirty(bh);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 556) } while ((bh = bh->b_this_page) != head);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 557)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 558) unlock:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 559) unlock_page(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 560) /* Nothing submitted - we have to end page writeback */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 561) if (!nr_submitted)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 562) end_page_writeback(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 563) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 564) }