diff options
Diffstat (limited to 'fs/gfs2/bmap.c')
| -rw-r--r-- | fs/gfs2/bmap.c | 679 |
1 files changed, 440 insertions, 239 deletions
diff --git a/fs/gfs2/bmap.c b/fs/gfs2/bmap.c index 6d47379e794..e6ee5b6e8d9 100644 --- a/fs/gfs2/bmap.c +++ b/fs/gfs2/bmap.c @@ -7,10 +7,10 @@ * of the GNU General Public License version 2. */ -#include <linux/slab.h> #include <linux/spinlock.h> #include <linux/completion.h> #include <linux/buffer_head.h> +#include <linux/blkdev.h> #include <linux/gfs2_ondisk.h> #include <linux/crc32.h> @@ -22,6 +22,8 @@ #include "meta_io.h" #include "quota.h" #include "rgrp.h" +#include "log.h" +#include "super.h" #include "trans.h" #include "dir.h" #include "util.h" @@ -36,11 +38,6 @@ struct metapath { __u16 mp_list[GFS2_MAX_META_HEIGHT]; }; -typedef int (*block_call_t) (struct gfs2_inode *ip, struct buffer_head *dibh, - struct buffer_head *bh, __be64 *top, - __be64 *bottom, unsigned int height, - void *data); - struct strip_mine { int sm_first; unsigned int sm_height; @@ -51,7 +48,7 @@ struct strip_mine { * @ip: the inode * @dibh: the dinode buffer * @block: the block number that was allocated - * @private: any locked page held by the caller process + * @page: The (optional) page. This is looked up if @page is NULL * * Returns: errno */ @@ -64,7 +61,7 @@ static int gfs2_unstuffer_page(struct gfs2_inode *ip, struct buffer_head *dibh, int release = 0; if (!page || page->index) { - page = grab_cache_page(inode->i_mapping, 0); + page = find_or_create_page(inode->i_mapping, 0, GFP_NOFS); if (!page) return -ENOMEM; release = 1; @@ -72,11 +69,13 @@ static int gfs2_unstuffer_page(struct gfs2_inode *ip, struct buffer_head *dibh, if (!PageUptodate(page)) { void *kaddr = kmap(page); + u64 dsize = i_size_read(inode); + + if (dsize > (dibh->b_size - sizeof(struct gfs2_dinode))) + dsize = dibh->b_size - sizeof(struct gfs2_dinode); - memcpy(kaddr, dibh->b_data + sizeof(struct gfs2_dinode), - ip->i_disksize); - memset(kaddr + ip->i_disksize, 0, - PAGE_CACHE_SIZE - ip->i_disksize); + memcpy(kaddr, dibh->b_data + sizeof(struct gfs2_dinode), dsize); + memset(kaddr + dsize, 0, PAGE_CACHE_SIZE - dsize); kunmap(page); SetPageUptodate(page); @@ -95,7 +94,7 @@ static int gfs2_unstuffer_page(struct gfs2_inode *ip, struct buffer_head *dibh, if (!gfs2_is_jdata(ip)) mark_buffer_dirty(bh); if (!gfs2_is_writeback(ip)) - gfs2_trans_add_bh(ip->i_gl, bh, 0); + gfs2_trans_add_data(ip->i_gl, bh); if (release) { unlock_page(page); @@ -108,8 +107,7 @@ static int gfs2_unstuffer_page(struct gfs2_inode *ip, struct buffer_head *dibh, /** * gfs2_unstuff_dinode - Unstuff a dinode when the data has grown too big * @ip: The GFS2 inode to unstuff - * @unstuffer: the routine that handles unstuffing a non-zero length file - * @private: private data for the unstuffer + * @page: The (optional) page. This is looked up if the @page is NULL * * This routine unstuffs a dinode and returns it to a "normal" state such * that the height can be grown in the traditional way. @@ -131,12 +129,12 @@ int gfs2_unstuff_dinode(struct gfs2_inode *ip, struct page *page) if (error) goto out; - if (ip->i_disksize) { + if (i_size_read(&ip->i_inode)) { /* Get a free block, fill it with the stuffed data, and write it out to disk */ unsigned int n = 1; - error = gfs2_alloc_block(ip, &block, &n); + error = gfs2_alloc_blocks(ip, &block, &n, 0, NULL); if (error) goto out_brelse; if (isdir) { @@ -156,11 +154,11 @@ int gfs2_unstuff_dinode(struct gfs2_inode *ip, struct page *page) /* Set up the pointer to the new block */ - gfs2_trans_add_bh(ip->i_gl, dibh, 1); + gfs2_trans_add_meta(ip->i_gl, dibh); di = (struct gfs2_dinode *)dibh->b_data; gfs2_buffer_clear_tail(dibh, sizeof(struct gfs2_dinode)); - if (ip->i_disksize) { + if (i_size_read(&ip->i_inode)) { *(__be64 *)(di + 1) = cpu_to_be64(block); gfs2_add_inode_blocks(&ip->i_inode, 1); di->di_blocks = cpu_to_be64(gfs2_get_inode_blocks(&ip->i_inode)); @@ -272,6 +270,30 @@ static inline __be64 *metapointer(unsigned int height, const struct metapath *mp return ((__be64 *)(bh->b_data + head_size)) + mp->mp_list[height]; } +static void gfs2_metapath_ra(struct gfs2_glock *gl, + const struct buffer_head *bh, const __be64 *pos) +{ + struct buffer_head *rabh; + const __be64 *endp = (const __be64 *)(bh->b_data + bh->b_size); + const __be64 *t; + + for (t = pos; t < endp; t++) { + if (!*t) + continue; + + rabh = gfs2_getbuf(gl, be64_to_cpu(*t), CREATE); + if (trylock_buffer(rabh)) { + if (!buffer_uptodate(rabh)) { + rabh->b_end_io = end_buffer_read_sync; + submit_bh(READA | REQ_META, rabh); + continue; + } + unlock_buffer(rabh); + } + brelse(rabh); + } +} + /** * lookup_metapath - Walk the metadata tree to a specific point * @ip: The inode @@ -303,7 +325,7 @@ static int lookup_metapath(struct gfs2_inode *ip, struct metapath *mp) if (!dblock) return x + 1; - ret = gfs2_meta_indirect_buffer(ip, x+1, dblock, 0, &mp->mp_bh[x+1]); + ret = gfs2_meta_indirect_buffer(ip, x+1, dblock, &mp->mp_bh[x+1]); if (ret) return ret; } @@ -384,7 +406,7 @@ static inline __be64 *gfs2_indirect_init(struct metapath *mp, BUG_ON(i < 1); BUG_ON(mp->mp_bh[i] != NULL); mp->mp_bh[i] = gfs2_meta_new(gl, bn); - gfs2_trans_add_bh(gl, mp->mp_bh[i], 1); + gfs2_trans_add_meta(gl, mp->mp_bh[i]); gfs2_metatype_set(mp->mp_bh[i], GFS2_METATYPE_IN, GFS2_FORMAT_IN); gfs2_buffer_clear_tail(mp->mp_bh[i], sizeof(struct gfs2_meta_header)); ptr += offset; @@ -431,12 +453,14 @@ static int gfs2_bmap_alloc(struct inode *inode, const sector_t lblock, { struct gfs2_inode *ip = GFS2_I(inode); struct gfs2_sbd *sdp = GFS2_SB(inode); + struct super_block *sb = sdp->sd_vfs; struct buffer_head *dibh = mp->mp_bh[0]; u64 bn, dblock = 0; unsigned n, i, blks, alloced = 0, iblks = 0, branch_start = 0; unsigned dblks = 0; unsigned ptrs_per_blk; const unsigned end_of_metadata = height - 1; + int ret; int eob = 0; enum alloc_state state; __be64 *ptr; @@ -445,7 +469,7 @@ static int gfs2_bmap_alloc(struct inode *inode, const sector_t lblock, BUG_ON(sheight < 1); BUG_ON(dibh == NULL); - gfs2_trans_add_bh(ip->i_gl, dibh, 1); + gfs2_trans_add_meta(ip->i_gl, dibh); if (height == sheight) { struct buffer_head *bh; @@ -480,7 +504,7 @@ static int gfs2_bmap_alloc(struct inode *inode, const sector_t lblock, do { int error; n = blks - alloced; - error = gfs2_alloc_block(ip, &bn, &n); + error = gfs2_alloc_blocks(ip, &bn, &n, 0, NULL); if (error) return error; alloced += n; @@ -521,7 +545,7 @@ static int gfs2_bmap_alloc(struct inode *inode, const sector_t lblock, /* Branching from existing tree */ case ALLOC_GROW_DEPTH: if (i > 1 && i < height) - gfs2_trans_add_bh(ip->i_gl, mp->mp_bh[i-1], 1); + gfs2_trans_add_meta(ip->i_gl, mp->mp_bh[i-1]); for (; i < height && n > 0; i++, n--) gfs2_indirect_init(mp, ip->i_gl, i, mp->mp_list[i-1], bn++); @@ -533,15 +557,24 @@ static int gfs2_bmap_alloc(struct inode *inode, const sector_t lblock, case ALLOC_DATA: BUG_ON(n > dblks); BUG_ON(mp->mp_bh[end_of_metadata] == NULL); - gfs2_trans_add_bh(ip->i_gl, mp->mp_bh[end_of_metadata], 1); + gfs2_trans_add_meta(ip->i_gl, mp->mp_bh[end_of_metadata]); dblks = n; ptr = metapointer(end_of_metadata, mp); dblock = bn; while (n-- > 0) *ptr++ = cpu_to_be64(bn++); + if (buffer_zeronew(bh_map)) { + ret = sb_issue_zeroout(sb, dblock, dblks, + GFP_NOFS); + if (ret) { + fs_err(sdp, + "Failed to zero data buffers\n"); + clear_buffer_zeronew(bh_map); + } + } break; } - } while (state != ALLOC_DATA); + } while ((state != ALLOC_DATA) || !dblock); ip->i_height = height; gfs2_add_inode_blocks(&ip->i_inode, alloced); @@ -667,76 +700,6 @@ int gfs2_extent_map(struct inode *inode, u64 lblock, int *new, u64 *dblock, unsi } /** - * recursive_scan - recursively scan through the end of a file - * @ip: the inode - * @dibh: the dinode buffer - * @mp: the path through the metadata to the point to start - * @height: the height the recursion is at - * @block: the indirect block to look at - * @first: 1 if this is the first block - * @bc: the call to make for each piece of metadata - * @data: data opaque to this function to pass to @bc - * - * When this is first called @height and @block should be zero and - * @first should be 1. - * - * Returns: errno - */ - -static int recursive_scan(struct gfs2_inode *ip, struct buffer_head *dibh, - struct metapath *mp, unsigned int height, - u64 block, int first, block_call_t bc, - void *data) -{ - struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); - struct buffer_head *bh = NULL; - __be64 *top, *bottom; - u64 bn; - int error; - int mh_size = sizeof(struct gfs2_meta_header); - - if (!height) { - error = gfs2_meta_inode_buffer(ip, &bh); - if (error) - return error; - dibh = bh; - - top = (__be64 *)(bh->b_data + sizeof(struct gfs2_dinode)) + mp->mp_list[0]; - bottom = (__be64 *)(bh->b_data + sizeof(struct gfs2_dinode)) + sdp->sd_diptrs; - } else { - error = gfs2_meta_indirect_buffer(ip, height, block, 0, &bh); - if (error) - return error; - - top = (__be64 *)(bh->b_data + mh_size) + - (first ? mp->mp_list[height] : 0); - - bottom = (__be64 *)(bh->b_data + mh_size) + sdp->sd_inptrs; - } - - error = bc(ip, dibh, bh, top, bottom, height, data); - if (error) - goto out; - - if (height < ip->i_height - 1) - for (; top < bottom; top++, first = 0) { - if (!*top) - continue; - - bn = be64_to_cpu(*top); - - error = recursive_scan(ip, dibh, mp, height + 1, bn, - first, bc, data); - if (error) - break; - } - -out: - brelse(bh); - return error; -} - -/** * do_strip - Look for a layer a particular layer of the file and strip it off * @ip: the inode * @dibh: the dinode buffer @@ -744,20 +707,19 @@ out: * @top: The first pointer in the buffer * @bottom: One more than the last pointer * @height: the height this buffer is at - * @data: a pointer to a struct strip_mine + * @sm: a pointer to a struct strip_mine * * Returns: errno */ static int do_strip(struct gfs2_inode *ip, struct buffer_head *dibh, struct buffer_head *bh, __be64 *top, __be64 *bottom, - unsigned int height, void *data) + unsigned int height, struct strip_mine *sm) { - struct strip_mine *sm = data; struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); struct gfs2_rgrp_list rlist; u64 bn, bstart; - u32 blen; + u32 blen, btotal; __be64 *p; unsigned int rg_blocks = 0; int metadata; @@ -765,6 +727,10 @@ static int do_strip(struct gfs2_inode *ip, struct buffer_head *dibh, int x; int error; + error = gfs2_rindex_update(sdp); + if (error) + return error; + if (!*top) sm->sm_first = 0; @@ -779,10 +745,8 @@ static int do_strip(struct gfs2_inode *ip, struct buffer_head *dibh, metadata = (height != ip->i_height - 1); if (metadata) revokes = (height) ? sdp->sd_inptrs : sdp->sd_diptrs; - - error = gfs2_rindex_hold(sdp, &ip->i_alloc->al_ri_gh); - if (error) - return error; + else if (ip->i_depth) + revokes = sdp->sd_inptrs; memset(&rlist, 0, sizeof(struct gfs2_rgrp_list)); bstart = 0; @@ -798,7 +762,7 @@ static int do_strip(struct gfs2_inode *ip, struct buffer_head *dibh, blen++; else { if (bstart) - gfs2_rlist_add(sdp, &rlist, bstart); + gfs2_rlist_add(ip, &rlist, bstart); bstart = bn; blen = 1; @@ -806,7 +770,7 @@ static int do_strip(struct gfs2_inode *ip, struct buffer_head *dibh, } if (bstart) - gfs2_rlist_add(sdp, &rlist, bstart); + gfs2_rlist_add(ip, &rlist, bstart); else goto out; /* Nothing to do */ @@ -822,6 +786,9 @@ static int do_strip(struct gfs2_inode *ip, struct buffer_head *dibh, if (error) goto out_rlist; + if (gfs2_rs_active(ip->i_res)) /* needs to be done with the rgrp glock held */ + gfs2_rs_deltree(ip->i_res); + error = gfs2_trans_begin(sdp, rg_blocks + RES_DINODE + RES_INDIRECT + RES_STATFS + RES_QUOTA, revokes); @@ -830,11 +797,12 @@ static int do_strip(struct gfs2_inode *ip, struct buffer_head *dibh, down_write(&ip->i_rw_mutex); - gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_trans_add_bh(ip->i_gl, bh, 1); + gfs2_trans_add_meta(ip->i_gl, dibh); + gfs2_trans_add_meta(ip->i_gl, bh); bstart = 0; blen = 0; + btotal = 0; for (p = top; p < bottom; p++) { if (!*p) @@ -846,10 +814,8 @@ static int do_strip(struct gfs2_inode *ip, struct buffer_head *dibh, blen++; else { if (bstart) { - if (metadata) - gfs2_free_meta(ip, bstart, blen); - else - gfs2_free_data(ip, bstart, blen); + __gfs2_free_blocks(ip, bstart, blen, metadata); + btotal += blen; } bstart = bn; @@ -860,12 +826,14 @@ static int do_strip(struct gfs2_inode *ip, struct buffer_head *dibh, gfs2_add_inode_blocks(&ip->i_inode, -1); } if (bstart) { - if (metadata) - gfs2_free_meta(ip, bstart, blen); - else - gfs2_free_data(ip, bstart, blen); + __gfs2_free_blocks(ip, bstart, blen, metadata); + btotal += blen; } + gfs2_statfs_change(sdp, 0, +btotal, 0); + gfs2_quota_change(ip, -(s64)btotal, ip->i_inode.i_uid, + ip->i_inode.i_gid); + ip->i_inode.i_mtime = ip->i_inode.i_ctime = CURRENT_TIME; gfs2_dinode_out(ip, dibh->b_data); @@ -879,74 +847,77 @@ out_rg_gunlock: out_rlist: gfs2_rlist_free(&rlist); out: - gfs2_glock_dq_uninit(&ip->i_alloc->al_ri_gh); return error; } /** - * do_grow - Make a file look bigger than it is + * recursive_scan - recursively scan through the end of a file * @ip: the inode - * @size: the size to set the file to + * @dibh: the dinode buffer + * @mp: the path through the metadata to the point to start + * @height: the height the recursion is at + * @block: the indirect block to look at + * @first: 1 if this is the first block + * @sm: data opaque to this function to pass to @bc * - * Called with an exclusive lock on @ip. + * When this is first called @height and @block should be zero and + * @first should be 1. * * Returns: errno */ -static int do_grow(struct gfs2_inode *ip, u64 size) +static int recursive_scan(struct gfs2_inode *ip, struct buffer_head *dibh, + struct metapath *mp, unsigned int height, + u64 block, int first, struct strip_mine *sm) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); - struct gfs2_alloc *al; - struct buffer_head *dibh; + struct buffer_head *bh = NULL; + __be64 *top, *bottom; + u64 bn; int error; + int mh_size = sizeof(struct gfs2_meta_header); - al = gfs2_alloc_get(ip); - if (!al) - return -ENOMEM; + if (!height) { + error = gfs2_meta_inode_buffer(ip, &bh); + if (error) + return error; + dibh = bh; - error = gfs2_quota_lock_check(ip); + top = (__be64 *)(bh->b_data + sizeof(struct gfs2_dinode)) + mp->mp_list[0]; + bottom = (__be64 *)(bh->b_data + sizeof(struct gfs2_dinode)) + sdp->sd_diptrs; + } else { + error = gfs2_meta_indirect_buffer(ip, height, block, &bh); + if (error) + return error; + + top = (__be64 *)(bh->b_data + mh_size) + + (first ? mp->mp_list[height] : 0); + + bottom = (__be64 *)(bh->b_data + mh_size) + sdp->sd_inptrs; + } + + error = do_strip(ip, dibh, bh, top, bottom, height, sm); if (error) goto out; - al->al_requested = sdp->sd_max_height + RES_DATA; + if (height < ip->i_height - 1) { - error = gfs2_inplace_reserve(ip); - if (error) - goto out_gunlock_q; + gfs2_metapath_ra(ip->i_gl, bh, top); - error = gfs2_trans_begin(sdp, - sdp->sd_max_height + al->al_rgd->rd_length + - RES_JDATA + RES_DINODE + RES_STATFS + RES_QUOTA, 0); - if (error) - goto out_ipres; + for (; top < bottom; top++, first = 0) { + if (!*top) + continue; - error = gfs2_meta_inode_buffer(ip, &dibh); - if (error) - goto out_end_trans; + bn = be64_to_cpu(*top); - if (size > sdp->sd_sb.sb_bsize - sizeof(struct gfs2_dinode)) { - if (gfs2_is_stuffed(ip)) { - error = gfs2_unstuff_dinode(ip, NULL); + error = recursive_scan(ip, dibh, mp, height + 1, bn, + first, sm); if (error) - goto out_brelse; + break; } } - - ip->i_disksize = size; - ip->i_inode.i_mtime = ip->i_inode.i_ctime = CURRENT_TIME; - gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(ip, dibh->b_data); - -out_brelse: - brelse(dibh); -out_end_trans: - gfs2_trans_end(sdp); -out_ipres: - gfs2_inplace_release(ip); -out_gunlock_q: - gfs2_quota_unlock(ip); out: - gfs2_alloc_put(ip); + brelse(bh); return error; } @@ -956,11 +927,10 @@ out: * * This is partly borrowed from ext3. */ -static int gfs2_block_truncate_page(struct address_space *mapping) +static int gfs2_block_truncate_page(struct address_space *mapping, loff_t from) { struct inode *inode = mapping->host; struct gfs2_inode *ip = GFS2_I(inode); - loff_t from = inode->i_size; unsigned long index = from >> PAGE_CACHE_SHIFT; unsigned offset = from & (PAGE_CACHE_SIZE-1); unsigned blocksize, iblock, length, pos; @@ -968,7 +938,7 @@ static int gfs2_block_truncate_page(struct address_space *mapping) struct page *page; int err; - page = grab_cache_page(mapping, index); + page = find_or_create_page(mapping, index, GFP_NOFS); if (!page) return 0; @@ -1012,7 +982,7 @@ static int gfs2_block_truncate_page(struct address_space *mapping) } if (!gfs2_is_writeback(ip)) - gfs2_trans_add_bh(ip->i_gl, bh, 0); + gfs2_trans_add_data(ip->i_gl, bh); zero_user(page, offset, length); mark_buffer_dirty(bh); @@ -1022,15 +992,54 @@ unlock: return err; } -static int trunc_start(struct gfs2_inode *ip, u64 size) +#define GFS2_JTRUNC_REVOKES 8192 + +/** + * gfs2_journaled_truncate - Wrapper for truncate_pagecache for jdata files + * @inode: The inode being truncated + * @oldsize: The original (larger) size + * @newsize: The new smaller size + * + * With jdata files, we have to journal a revoke for each block which is + * truncated. As a result, we need to split this into separate transactions + * if the number of pages being truncated gets too large. + */ + +static int gfs2_journaled_truncate(struct inode *inode, u64 oldsize, u64 newsize) +{ + struct gfs2_sbd *sdp = GFS2_SB(inode); + u64 max_chunk = GFS2_JTRUNC_REVOKES * sdp->sd_vfs->s_blocksize; + u64 chunk; + int error; + + while (oldsize != newsize) { + chunk = oldsize - newsize; + if (chunk > max_chunk) + chunk = max_chunk; + truncate_pagecache(inode, oldsize - chunk); + oldsize -= chunk; + gfs2_trans_end(sdp); + error = gfs2_trans_begin(sdp, RES_DINODE, GFS2_JTRUNC_REVOKES); + if (error) + return error; + } + + return 0; +} + +static int trunc_start(struct inode *inode, u64 oldsize, u64 newsize) { - struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); + struct gfs2_inode *ip = GFS2_I(inode); + struct gfs2_sbd *sdp = GFS2_SB(inode); + struct address_space *mapping = inode->i_mapping; struct buffer_head *dibh; int journaled = gfs2_is_jdata(ip); int error; - error = gfs2_trans_begin(sdp, - RES_DINODE + (journaled ? RES_JDATA : 0), 0); + if (journaled) + error = gfs2_trans_begin(sdp, RES_DINODE + RES_JDATA, GFS2_JTRUNC_REVOKES); + else + error = gfs2_trans_begin(sdp, RES_DINODE, 0); if (error) return error; @@ -1038,29 +1047,35 @@ static int trunc_start(struct gfs2_inode *ip, u64 size) if (error) goto out; - if (gfs2_is_stuffed(ip)) { - ip->i_disksize = size; - ip->i_inode.i_mtime = ip->i_inode.i_ctime = CURRENT_TIME; - gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(ip, dibh->b_data); - gfs2_buffer_clear_tail(dibh, sizeof(struct gfs2_dinode) + size); - error = 1; + gfs2_trans_add_meta(ip->i_gl, dibh); + if (gfs2_is_stuffed(ip)) { + gfs2_buffer_clear_tail(dibh, sizeof(struct gfs2_dinode) + newsize); } else { - if (size & (u64)(sdp->sd_sb.sb_bsize - 1)) - error = gfs2_block_truncate_page(ip->i_inode.i_mapping); - - if (!error) { - ip->i_disksize = size; - ip->i_inode.i_mtime = ip->i_inode.i_ctime = CURRENT_TIME; - ip->i_diskflags |= GFS2_DIF_TRUNC_IN_PROG; - gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(ip, dibh->b_data); + if (newsize & (u64)(sdp->sd_sb.sb_bsize - 1)) { + error = gfs2_block_truncate_page(mapping, newsize); + if (error) + goto out_brelse; } + ip->i_diskflags |= GFS2_DIF_TRUNC_IN_PROG; } - brelse(dibh); + i_size_write(inode, newsize); + ip->i_inode.i_mtime = ip->i_inode.i_ctime = CURRENT_TIME; + gfs2_dinode_out(ip, dibh->b_data); + if (journaled) + error = gfs2_journaled_truncate(inode, oldsize, newsize); + else + truncate_pagecache(inode, newsize); + + if (error) { + brelse(dibh); + return error; + } + +out_brelse: + brelse(dibh); out: gfs2_trans_end(sdp); return error; @@ -1080,27 +1095,26 @@ static int trunc_dealloc(struct gfs2_inode *ip, u64 size) lblock = (size - 1) >> sdp->sd_sb.sb_bsize_shift; find_metapath(sdp, lblock, &mp, ip->i_height); - if (!gfs2_alloc_get(ip)) - return -ENOMEM; + error = gfs2_rindex_update(sdp); + if (error) + return error; - error = gfs2_quota_hold(ip, NO_QUOTA_CHANGE, NO_QUOTA_CHANGE); + error = gfs2_quota_hold(ip, NO_UID_QUOTA_CHANGE, NO_GID_QUOTA_CHANGE); if (error) - goto out; + return error; while (height--) { struct strip_mine sm; sm.sm_first = !!size; sm.sm_height = height; - error = recursive_scan(ip, NULL, &mp, 0, 0, 1, do_strip, &sm); + error = recursive_scan(ip, NULL, &mp, 0, 0, 1, &sm); if (error) break; } gfs2_quota_unhold(ip); -out: - gfs2_alloc_put(ip); return error; } @@ -1120,15 +1134,16 @@ static int trunc_end(struct gfs2_inode *ip) if (error) goto out; - if (!ip->i_disksize) { + if (!i_size_read(&ip->i_inode)) { ip->i_height = 0; ip->i_goal = ip->i_no_addr; gfs2_buffer_clear_tail(dibh, sizeof(struct gfs2_dinode)); + gfs2_ordered_del_inode(ip); } ip->i_inode.i_mtime = ip->i_inode.i_ctime = CURRENT_TIME; ip->i_diskflags &= ~GFS2_DIF_TRUNC_IN_PROG; - gfs2_trans_add_bh(ip->i_gl, dibh, 1); + gfs2_trans_add_meta(ip->i_gl, dibh); gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); @@ -1140,92 +1155,168 @@ out: /** * do_shrink - make a file smaller - * @ip: the inode - * @size: the size to make the file - * @truncator: function to truncate the last partial block + * @inode: the inode + * @oldsize: the current inode size + * @newsize: the size to make the file * - * Called with an exclusive lock on @ip. + * Called with an exclusive lock on @inode. The @size must + * be equal to or smaller than the current inode size. * * Returns: errno */ -static int do_shrink(struct gfs2_inode *ip, u64 size) +static int do_shrink(struct inode *inode, u64 oldsize, u64 newsize) { + struct gfs2_inode *ip = GFS2_I(inode); int error; - error = trunc_start(ip, size); + error = trunc_start(inode, oldsize, newsize); if (error < 0) return error; - if (error > 0) + if (gfs2_is_stuffed(ip)) return 0; - error = trunc_dealloc(ip, size); - if (!error) + error = trunc_dealloc(ip, newsize); + if (error == 0) error = trunc_end(ip); return error; } -static int do_touch(struct gfs2_inode *ip, u64 size) +void gfs2_trim_blocks(struct inode *inode) { - struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); + u64 size = inode->i_size; + int ret; + + ret = do_shrink(inode, size, size); + WARN_ON(ret != 0); +} + +/** + * do_grow - Touch and update inode size + * @inode: The inode + * @size: The new size + * + * This function updates the timestamps on the inode and + * may also increase the size of the inode. This function + * must not be called with @size any smaller than the current + * inode size. + * + * Although it is not strictly required to unstuff files here, + * earlier versions of GFS2 have a bug in the stuffed file reading + * code which will result in a buffer overrun if the size is larger + * than the max stuffed file size. In order to prevent this from + * occurring, such files are unstuffed, but in other cases we can + * just update the inode size directly. + * + * Returns: 0 on success, or -ve on error + */ + +static int do_grow(struct inode *inode, u64 size) +{ + struct gfs2_inode *ip = GFS2_I(inode); + struct gfs2_sbd *sdp = GFS2_SB(inode); + struct gfs2_alloc_parms ap = { .target = 1, }; struct buffer_head *dibh; int error; + int unstuff = 0; - error = gfs2_trans_begin(sdp, RES_DINODE, 0); + if (gfs2_is_stuffed(ip) && + (size > (sdp->sd_sb.sb_bsize - sizeof(struct gfs2_dinode)))) { + error = gfs2_quota_lock_check(ip); + if (error) + return error; + + error = gfs2_inplace_reserve(ip, &ap); + if (error) + goto do_grow_qunlock; + unstuff = 1; + } + + error = gfs2_trans_begin(sdp, RES_DINODE + RES_STATFS + RES_RG_BIT + + (sdp->sd_args.ar_quota == GFS2_QUOTA_OFF ? + 0 : RES_QUOTA), 0); if (error) - return error; + goto do_grow_release; - down_write(&ip->i_rw_mutex); + if (unstuff) { + error = gfs2_unstuff_dinode(ip, NULL); + if (error) + goto do_end_trans; + } error = gfs2_meta_inode_buffer(ip, &dibh); if (error) - goto do_touch_out; + goto do_end_trans; + i_size_write(inode, size); ip->i_inode.i_mtime = ip->i_inode.i_ctime = CURRENT_TIME; - gfs2_trans_add_bh(ip->i_gl, dibh, 1); + gfs2_trans_add_meta(ip->i_gl, dibh); gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); -do_touch_out: - up_write(&ip->i_rw_mutex); +do_end_trans: gfs2_trans_end(sdp); +do_grow_release: + if (unstuff) { + gfs2_inplace_release(ip); +do_grow_qunlock: + gfs2_quota_unlock(ip); + } return error; } /** - * gfs2_truncatei - make a file a given size - * @ip: the inode - * @size: the size to make the file - * @truncator: function to truncate the last partial block + * gfs2_setattr_size - make a file a given size + * @inode: the inode + * @newsize: the size to make the file * - * The file size can grow, shrink, or stay the same size. + * The file size can grow, shrink, or stay the same size. This + * is called holding i_mutex and an exclusive glock on the inode + * in question. * * Returns: errno */ -int gfs2_truncatei(struct gfs2_inode *ip, u64 size) +int gfs2_setattr_size(struct inode *inode, u64 newsize) { - int error; + struct gfs2_inode *ip = GFS2_I(inode); + int ret; + u64 oldsize; - if (gfs2_assert_warn(GFS2_SB(&ip->i_inode), S_ISREG(ip->i_inode.i_mode))) - return -EINVAL; + BUG_ON(!S_ISREG(inode->i_mode)); - if (size > ip->i_disksize) - error = do_grow(ip, size); - else if (size < ip->i_disksize) - error = do_shrink(ip, size); - else - /* update time stamps */ - error = do_touch(ip, size); + ret = inode_newsize_ok(inode, newsize); + if (ret) + return ret; - return error; + ret = get_write_access(inode); + if (ret) + return ret; + + inode_dio_wait(inode); + + ret = gfs2_rs_alloc(ip); + if (ret) + goto out; + + oldsize = inode->i_size; + if (newsize >= oldsize) { + ret = do_grow(inode, newsize); + goto out; + } + + gfs2_rs_deltree(ip->i_res); + ret = do_shrink(inode, oldsize, newsize); +out: + put_write_access(inode); + return ret; } int gfs2_truncatei_resume(struct gfs2_inode *ip) { int error; - error = trunc_dealloc(ip, ip->i_disksize); + error = trunc_dealloc(ip, i_size_read(&ip->i_inode)); if (!error) error = trunc_end(ip); return error; @@ -1237,17 +1328,131 @@ int gfs2_file_dealloc(struct gfs2_inode *ip) } /** + * gfs2_free_journal_extents - Free cached journal bmap info + * @jd: The journal + * + */ + +void gfs2_free_journal_extents(struct gfs2_jdesc *jd) +{ + struct gfs2_journal_extent *jext; + + while(!list_empty(&jd->extent_list)) { + jext = list_entry(jd->extent_list.next, struct gfs2_journal_extent, list); + list_del(&jext->list); + kfree(jext); + } +} + +/** + * gfs2_add_jextent - Add or merge a new extent to extent cache + * @jd: The journal descriptor + * @lblock: The logical block at start of new extent + * @dblock: The physical block at start of new extent + * @blocks: Size of extent in fs blocks + * + * Returns: 0 on success or -ENOMEM + */ + +static int gfs2_add_jextent(struct gfs2_jdesc *jd, u64 lblock, u64 dblock, u64 blocks) +{ + struct gfs2_journal_extent *jext; + + if (!list_empty(&jd->extent_list)) { + jext = list_entry(jd->extent_list.prev, struct gfs2_journal_extent, list); + if ((jext->dblock + jext->blocks) == dblock) { + jext->blocks += blocks; + return 0; + } + } + + jext = kzalloc(sizeof(struct gfs2_journal_extent), GFP_NOFS); + if (jext == NULL) + return -ENOMEM; + jext->dblock = dblock; + jext->lblock = lblock; + jext->blocks = blocks; + list_add_tail(&jext->list, &jd->extent_list); + jd->nr_extents++; + return 0; +} + +/** + * gfs2_map_journal_extents - Cache journal bmap info + * @sdp: The super block + * @jd: The journal to map + * + * Create a reusable "extent" mapping from all logical + * blocks to all physical blocks for the given journal. This will save + * us time when writing journal blocks. Most journals will have only one + * extent that maps all their logical blocks. That's because gfs2.mkfs + * arranges the journal blocks sequentially to maximize performance. + * So the extent would map the first block for the entire file length. + * However, gfs2_jadd can happen while file activity is happening, so + * those journals may not be sequential. Less likely is the case where + * the users created their own journals by mounting the metafs and + * laying it out. But it's still possible. These journals might have + * several extents. + * + * Returns: 0 on success, or error on failure + */ + +int gfs2_map_journal_extents(struct gfs2_sbd *sdp, struct gfs2_jdesc *jd) +{ + u64 lblock = 0; + u64 lblock_stop; + struct gfs2_inode *ip = GFS2_I(jd->jd_inode); + struct buffer_head bh; + unsigned int shift = sdp->sd_sb.sb_bsize_shift; + u64 size; + int rc; + + lblock_stop = i_size_read(jd->jd_inode) >> shift; + size = (lblock_stop - lblock) << shift; + jd->nr_extents = 0; + WARN_ON(!list_empty(&jd->extent_list)); + + do { + bh.b_state = 0; + bh.b_blocknr = 0; + bh.b_size = size; + rc = gfs2_block_map(jd->jd_inode, lblock, &bh, 0); + if (rc || !buffer_mapped(&bh)) + goto fail; + rc = gfs2_add_jextent(jd, lblock, bh.b_blocknr, bh.b_size >> shift); + if (rc) + goto fail; + size -= bh.b_size; + lblock += (bh.b_size >> ip->i_inode.i_blkbits); + } while(size > 0); + + fs_info(sdp, "journal %d mapped with %u extents\n", jd->jd_jid, + jd->nr_extents); + return 0; + +fail: + fs_warn(sdp, "error %d mapping journal %u at offset %llu (extent %u)\n", + rc, jd->jd_jid, + (unsigned long long)(i_size_read(jd->jd_inode) - size), + jd->nr_extents); + fs_warn(sdp, "bmap=%d lblock=%llu block=%llu, state=0x%08lx, size=%llu\n", + rc, (unsigned long long)lblock, (unsigned long long)bh.b_blocknr, + bh.b_state, (unsigned long long)bh.b_size); + gfs2_free_journal_extents(jd); + return rc; +} + +/** * gfs2_write_alloc_required - figure out if a write will require an allocation * @ip: the file being written to * @offset: the offset to write to * @len: the number of bytes being written - * @alloc_required: set to 1 if an alloc is required, 0 otherwise * - * Returns: errno + * Returns: 1 if an alloc is required, 0 otherwise */ int gfs2_write_alloc_required(struct gfs2_inode *ip, u64 offset, - unsigned int len, int *alloc_required) + unsigned int len) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); struct buffer_head bh; @@ -1255,26 +1460,23 @@ int gfs2_write_alloc_required(struct gfs2_inode *ip, u64 offset, u64 lblock, lblock_stop, size; u64 end_of_file; - *alloc_required = 0; - if (!len) return 0; if (gfs2_is_stuffed(ip)) { if (offset + len > sdp->sd_sb.sb_bsize - sizeof(struct gfs2_dinode)) - *alloc_required = 1; + return 1; return 0; } - *alloc_required = 1; shift = sdp->sd_sb.sb_bsize_shift; BUG_ON(gfs2_is_dir(ip)); - end_of_file = (ip->i_disksize + sdp->sd_sb.sb_bsize - 1) >> shift; + end_of_file = (i_size_read(&ip->i_inode) + sdp->sd_sb.sb_bsize - 1) >> shift; lblock = offset >> shift; lblock_stop = (offset + len + sdp->sd_sb.sb_bsize - 1) >> shift; if (lblock_stop > end_of_file) - return 0; + return 1; size = (lblock_stop - lblock) << shift; do { @@ -1282,12 +1484,11 @@ int gfs2_write_alloc_required(struct gfs2_inode *ip, u64 offset, bh.b_size = size; gfs2_block_map(&ip->i_inode, lblock, &bh, 0); if (!buffer_mapped(&bh)) - return 0; + return 1; size -= bh.b_size; lblock += (bh.b_size >> ip->i_inode.i_blkbits); } while(size > 0); - *alloc_required = 0; return 0; } |