From a76cccbeef9e91b5f799e8853acac1ed1fc833cb Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Tue, 31 Jul 2012 14:55:51 +1000 Subject: xfs: fix uninitialised variable in xfs_rtbuf_get() Results in this assert failure in generic/090: XFS: Assertion failed: *nmap >= 1, file: fs/xfs/xfs_bmap.c, line: 4363 ..... Call Trace: [] xfs_bmapi_read+0x6b/0x370 [] xfs_rtbuf_get+0x42/0x130 [] xfs_rtget_summary+0x89/0x120 [] xfs_rtallocate_extent_size+0xce/0x340 [] xfs_rtallocate_extent+0x240/0x290 [] xfs_bmap_rtalloc+0x1ba/0x340 [] xfs_bmap_alloc+0x35/0x40 [] xfs_bmapi_allocate+0xf1/0x350 [] xfs_bmapi_write+0x66e/0xa60 [] xfs_iomap_write_direct+0x22a/0x3f0 [] __xfs_get_blocks+0x38b/0x5d0 [] xfs_get_blocks_direct+0x14/0x20 [] do_blockdev_direct_IO+0xf71/0x1eb0 [] __blockdev_direct_IO+0x55/0x60 [] xfs_vm_direct_IO+0x11a/0x1e0 [] generic_file_direct_write+0xd7/0x1b0 [] xfs_file_dio_aio_write+0x13c/0x320 [] xfs_file_aio_write+0x1c2/0x1d0 [] do_sync_write+0xa7/0xe0 [] vfs_write+0xa8/0x160 [] sys_pwrite64+0x92/0xb0 [] system_call_fastpath+0x16/0x1b Signed-off-by: Dave Chinner Signed-off-by: Ben Myers --- fs/xfs/xfs_rtalloc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/xfs/xfs_rtalloc.c b/fs/xfs/xfs_rtalloc.c index 92d4331cd4f1..ca28a4ba4b54 100644 --- a/fs/xfs/xfs_rtalloc.c +++ b/fs/xfs/xfs_rtalloc.c @@ -857,7 +857,7 @@ xfs_rtbuf_get( xfs_buf_t *bp; /* block buffer, result */ xfs_inode_t *ip; /* bitmap or summary inode */ xfs_bmbt_irec_t map; - int nmap; + int nmap = 1; int error; /* error value */ ip = issum ? mp->m_rsumip : mp->m_rbmip; -- cgit v1.2.3-58-ga151 From 1ed845df60f3f02d4b7cd9fcad79ccb69c289f5c Mon Sep 17 00:00:00 2001 From: Alex Elder Date: Wed, 1 Aug 2012 09:56:49 -0500 Subject: xfs: kill struct declarations in xfs_mount.h I noticed that "struct xfs_mount_args" was still declared in "fs/xfs/xfs_mount.h". That struct doesn't even exist any more (and is obviously not referenced elsewhere in that header file). While in there, delete four other unneeded struct declarations in that file. Doing so highlights that "fs/xfs/xfs_trace.h" was relying indirectly on "xfs_mount.h" to be #included in order to declare "struct xfs_bmbt_irec", so add that declaration to resolve that issue. Signed-off-by: Alex Elder Signed-off-by: Ben Myers --- fs/xfs/xfs_mount.h | 5 ----- fs/xfs/xfs_trace.h | 1 + 2 files changed, 1 insertion(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h index 05a05a7b6119..deee09e534dc 100644 --- a/fs/xfs/xfs_mount.h +++ b/fs/xfs/xfs_mount.h @@ -54,12 +54,7 @@ typedef struct xfs_trans_reservations { #include "xfs_sync.h" struct xlog; -struct xfs_mount_args; struct xfs_inode; -struct xfs_bmbt_irec; -struct xfs_bmap_free; -struct xfs_extdelta; -struct xfs_swapext; struct xfs_mru_cache; struct xfs_nameops; struct xfs_ail; diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h index e5795dd6013a..7d36ccf57f93 100644 --- a/fs/xfs/xfs_trace.h +++ b/fs/xfs/xfs_trace.h @@ -37,6 +37,7 @@ struct xlog_recover; struct xlog_recover_item; struct xfs_buf_log_format; struct xfs_inode_log_format; +struct xfs_bmbt_irec; DECLARE_EVENT_CLASS(xfs_attr_list_class, TP_PROTO(struct xfs_attr_list_context *ctx), -- cgit v1.2.3-58-ga151 From c4982110ae93d7575503feb81d15e93c0c5f393c Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 7 Aug 2012 02:02:02 -0400 Subject: xfs: unlock the AGI buffer when looping in xfs_dialloc Also update some commens in the area to make the code easier to read. Signed-off-by: Christoph Hellwig Reviewed-by: Mark Tinguely Reviewed-by: Dave Chinner Signed-off-by: Ben Myers --- fs/xfs/xfs_ialloc.c | 17 +++++++++-------- 1 file changed, 9 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_ialloc.c b/fs/xfs/xfs_ialloc.c index 21e37b55f7e5..5aceb3f8ecd6 100644 --- a/fs/xfs/xfs_ialloc.c +++ b/fs/xfs/xfs_ialloc.c @@ -962,23 +962,22 @@ xfs_dialloc( if (!pag->pagi_freecount && !okalloc) goto nextag; + /* + * Then read in the AGI buffer and recheck with the AGI buffer + * lock held. + */ error = xfs_ialloc_read_agi(mp, tp, agno, &agbp); if (error) goto out_error; - /* - * Once the AGI has been read in we have to recheck - * pagi_freecount with the AGI buffer lock held. - */ if (pag->pagi_freecount) { xfs_perag_put(pag); goto out_alloc; } - if (!okalloc) { - xfs_trans_brelse(tp, agbp); - goto nextag; - } + if (!okalloc) + goto nextag_relse_buffer; + error = xfs_ialloc_ag_alloc(tp, agbp, &ialloced); if (error) { @@ -1007,6 +1006,8 @@ xfs_dialloc( return 0; } +nextag_relse_buffer: + xfs_trans_brelse(tp, agbp); nextag: xfs_perag_put(pag); if (++agno == mp->m_sb.sb_agcount) -- cgit v1.2.3-58-ga151 From 643bfc061c47e9c7661324a09fb0a0bc6601e5d6 Mon Sep 17 00:00:00 2001 From: Tomas Racek Date: Tue, 14 Aug 2012 10:35:04 +0200 Subject: xfs: check for possible overflow in xfs_ioc_trim If range.start or range.minlen is bigger than filesystem size, return invalid value error. This fixes possible overflow in BTOBB macro when passed value was nearly ULLONG_MAX. Signed-off-by: Tomas Racek Reviewed-by: Dave Chinner Signed-off-by: Ben Myers --- fs/xfs/xfs_discard.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_discard.c b/fs/xfs/xfs_discard.c index f9c3fe304a17..69cf4fcde03e 100644 --- a/fs/xfs/xfs_discard.c +++ b/fs/xfs/xfs_discard.c @@ -179,12 +179,14 @@ xfs_ioc_trim( * used by the fstrim application. In the end it really doesn't * matter as trimming blocks is an advisory interface. */ + if (range.start >= XFS_FSB_TO_B(mp, mp->m_sb.sb_dblocks) || + range.minlen > XFS_FSB_TO_B(mp, XFS_ALLOC_AG_MAX_USABLE(mp))) + return -XFS_ERROR(EINVAL); + start = BTOBB(range.start); end = start + BTOBBT(range.len) - 1; minlen = BTOBB(max_t(u64, granularity, range.minlen)); - if (XFS_BB_TO_FSB(mp, start) >= mp->m_sb.sb_dblocks) - return -XFS_ERROR(EINVAL); if (end > XFS_FSB_TO_BB(mp, mp->m_sb.sb_dblocks) - 1) end = XFS_FSB_TO_BB(mp, mp->m_sb.sb_dblocks)- 1; -- cgit v1.2.3-58-ga151 From e599b3253c5e49f7a2a579eef2fc2aa066989ef5 Mon Sep 17 00:00:00 2001 From: Carlos Maiolino Date: Fri, 10 Aug 2012 15:01:51 -0300 Subject: xfs: fix race while discarding buffers [V4] While xfs_buftarg_shrink() is freeing buffers from the dispose list (filled with buffers from lru list), there is a possibility to have xfs_buf_stale() racing with it, and removing buffers from dispose list before xfs_buftarg_shrink() does it. This happens because xfs_buftarg_shrink() handle the dispose list without locking and the test condition in xfs_buf_stale() checks for the buffer being in *any* list: if (!list_empty(&bp->b_lru)) If the buffer happens to be on dispose list, this causes the buffer counter of lru list (btp->bt_lru_nr) to be decremented twice (once in xfs_buftarg_shrink() and another in xfs_buf_stale()) causing a wrong account usage of the lru list. This may cause xfs_buftarg_shrink() to return a wrong value to the memory shrinker shrink_slab(), and such account error may also cause an underflowed value to be returned; since the counter is lower than the current number of items in the lru list, a decrement may happen when the counter is 0, causing an underflow on the counter. The fix uses a new flag field (and a new buffer flag) to serialize buffer handling during the shrink process. The new flag field has been designed to use btp->bt_lru_lock/unlock instead of xfs_buf_lock/unlock mechanism. dchinner, sandeen, aquini and aris also deserve credits for this. Signed-off-by: Carlos Maiolino Reviewed-by: Ben Myers Reviewed-by: Dave Chinner Signed-off-by: Ben Myers --- fs/xfs/xfs_buf.c | 5 ++++- fs/xfs/xfs_buf.h | 41 ++++++++++++++++++++++++----------------- 2 files changed, 28 insertions(+), 18 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c index d7a9dd735e1e..933b7930b863 100644 --- a/fs/xfs/xfs_buf.c +++ b/fs/xfs/xfs_buf.c @@ -96,6 +96,7 @@ xfs_buf_lru_add( atomic_inc(&bp->b_hold); list_add_tail(&bp->b_lru, &btp->bt_lru); btp->bt_lru_nr++; + bp->b_lru_flags &= ~_XBF_LRU_DISPOSE; } spin_unlock(&btp->bt_lru_lock); } @@ -154,7 +155,8 @@ xfs_buf_stale( struct xfs_buftarg *btp = bp->b_target; spin_lock(&btp->bt_lru_lock); - if (!list_empty(&bp->b_lru)) { + if (!list_empty(&bp->b_lru) && + !(bp->b_lru_flags & _XBF_LRU_DISPOSE)) { list_del_init(&bp->b_lru); btp->bt_lru_nr--; atomic_dec(&bp->b_hold); @@ -1501,6 +1503,7 @@ xfs_buftarg_shrink( */ list_move(&bp->b_lru, &dispose); btp->bt_lru_nr--; + bp->b_lru_flags |= _XBF_LRU_DISPOSE; } spin_unlock(&btp->bt_lru_lock); diff --git a/fs/xfs/xfs_buf.h b/fs/xfs/xfs_buf.h index d03b73b9604e..7c0b6a0a1557 100644 --- a/fs/xfs/xfs_buf.h +++ b/fs/xfs/xfs_buf.h @@ -38,27 +38,28 @@ typedef enum { XBRW_ZERO = 3, /* Zero target memory */ } xfs_buf_rw_t; -#define XBF_READ (1 << 0) /* buffer intended for reading from device */ -#define XBF_WRITE (1 << 1) /* buffer intended for writing to device */ -#define XBF_READ_AHEAD (1 << 2) /* asynchronous read-ahead */ -#define XBF_ASYNC (1 << 4) /* initiator will not wait for completion */ -#define XBF_DONE (1 << 5) /* all pages in the buffer uptodate */ -#define XBF_STALE (1 << 6) /* buffer has been staled, do not find it */ +#define XBF_READ (1 << 0) /* buffer intended for reading from device */ +#define XBF_WRITE (1 << 1) /* buffer intended for writing to device */ +#define XBF_READ_AHEAD (1 << 2) /* asynchronous read-ahead */ +#define XBF_ASYNC (1 << 4) /* initiator will not wait for completion */ +#define XBF_DONE (1 << 5) /* all pages in the buffer uptodate */ +#define XBF_STALE (1 << 6) /* buffer has been staled, do not find it */ /* I/O hints for the BIO layer */ -#define XBF_SYNCIO (1 << 10)/* treat this buffer as synchronous I/O */ -#define XBF_FUA (1 << 11)/* force cache write through mode */ -#define XBF_FLUSH (1 << 12)/* flush the disk cache before a write */ +#define XBF_SYNCIO (1 << 10)/* treat this buffer as synchronous I/O */ +#define XBF_FUA (1 << 11)/* force cache write through mode */ +#define XBF_FLUSH (1 << 12)/* flush the disk cache before a write */ /* flags used only as arguments to access routines */ -#define XBF_TRYLOCK (1 << 16)/* lock requested, but do not wait */ -#define XBF_UNMAPPED (1 << 17)/* do not map the buffer */ +#define XBF_TRYLOCK (1 << 16)/* lock requested, but do not wait */ +#define XBF_UNMAPPED (1 << 17)/* do not map the buffer */ /* flags used only internally */ -#define _XBF_PAGES (1 << 20)/* backed by refcounted pages */ -#define _XBF_KMEM (1 << 21)/* backed by heap memory */ -#define _XBF_DELWRI_Q (1 << 22)/* buffer on a delwri queue */ -#define _XBF_COMPOUND (1 << 23)/* compound buffer */ +#define _XBF_PAGES (1 << 20)/* backed by refcounted pages */ +#define _XBF_KMEM (1 << 21)/* backed by heap memory */ +#define _XBF_DELWRI_Q (1 << 22)/* buffer on a delwri queue */ +#define _XBF_COMPOUND (1 << 23)/* compound buffer */ +#define _XBF_LRU_DISPOSE (1 << 24)/* buffer being discarded */ typedef unsigned int xfs_buf_flags_t; @@ -72,12 +73,13 @@ typedef unsigned int xfs_buf_flags_t; { XBF_SYNCIO, "SYNCIO" }, \ { XBF_FUA, "FUA" }, \ { XBF_FLUSH, "FLUSH" }, \ - { XBF_TRYLOCK, "TRYLOCK" }, /* should never be set */\ + { XBF_TRYLOCK, "TRYLOCK" }, /* should never be set */\ { XBF_UNMAPPED, "UNMAPPED" }, /* ditto */\ { _XBF_PAGES, "PAGES" }, \ { _XBF_KMEM, "KMEM" }, \ { _XBF_DELWRI_Q, "DELWRI_Q" }, \ - { _XBF_COMPOUND, "COMPOUND" } + { _XBF_COMPOUND, "COMPOUND" }, \ + { _XBF_LRU_DISPOSE, "LRU_DISPOSE" } typedef struct xfs_buftarg { dev_t bt_dev; @@ -124,7 +126,12 @@ typedef struct xfs_buf { xfs_buf_flags_t b_flags; /* status flags */ struct semaphore b_sema; /* semaphore for lockables */ + /* + * concurrent access to b_lru and b_lru_flags are protected by + * bt_lru_lock and not by b_sema + */ struct list_head b_lru; /* lru list */ + xfs_buf_flags_t b_lru_flags; /* internal lru status flags */ wait_queue_head_t b_waiters; /* unpin waiters */ struct list_head b_list; struct xfs_perag *b_pag; /* contains rbtree root */ -- cgit v1.2.3-58-ga151 From 834ab12228fad777a11007a24cb6286b02c9a41c Mon Sep 17 00:00:00 2001 From: Jeff Liu Date: Tue, 21 Aug 2012 17:11:45 +0800 Subject: xfs: Remove type argument from xfs_seek_data()/xfs_seek_hole() The type is already indicated by the function naming explicitly, so this argument can be omitted from those calls. Signed-off-by: Jie Liu Reviewed-by: Mark Tinguely Reviewed-by: Dave Chinner Signed-off-by: Ben Myers --- fs/xfs/xfs_file.c | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index 56afcdb2377d..92ba18f841f1 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -962,8 +962,7 @@ xfs_vm_page_mkwrite( STATIC loff_t xfs_seek_data( struct file *file, - loff_t start, - u32 type) + loff_t start) { struct inode *inode = file->f_mapping->host; struct xfs_inode *ip = XFS_I(inode); @@ -1029,8 +1028,7 @@ out_unlock: STATIC loff_t xfs_seek_hole( struct file *file, - loff_t start, - u32 type) + loff_t start) { struct inode *inode = file->f_mapping->host; struct xfs_inode *ip = XFS_I(inode); @@ -1092,9 +1090,9 @@ xfs_file_llseek( case SEEK_SET: return generic_file_llseek(file, offset, origin); case SEEK_DATA: - return xfs_seek_data(file, offset, origin); + return xfs_seek_data(file, offset); case SEEK_HOLE: - return xfs_seek_hole(file, offset, origin); + return xfs_seek_hole(file, offset); default: return -EINVAL; } -- cgit v1.2.3-58-ga151 From d126d43f631f996daeee5006714fed914be32368 Mon Sep 17 00:00:00 2001 From: Jeff Liu Date: Tue, 21 Aug 2012 17:11:57 +0800 Subject: xfs: Introduce a helper routine to probe data or hole offset from page cache Introduce helpers to probe data or hole offset from page cache. Signed-off-by: Jie Liu Reviewed-by: Mark Tinguely Reviewed-by: Dave Chinner Signed-off-by: Ben Myers --- fs/xfs/xfs_file.c | 219 ++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 219 insertions(+) (limited to 'fs') diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index 92ba18f841f1..d78a746b6c7c 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -36,6 +36,7 @@ #include #include +#include static const struct vm_operations_struct xfs_file_vm_ops; @@ -959,6 +960,224 @@ xfs_vm_page_mkwrite( return block_page_mkwrite(vma, vmf, xfs_get_blocks); } +/* + * This type is designed to indicate the type of offset we would like + * to search from page cache for either xfs_seek_data() or xfs_seek_hole(). + */ +enum { + HOLE_OFF = 0, + DATA_OFF, +}; + +/* + * Lookup the desired type of offset from the given page. + * + * On success, return true and the offset argument will point to the + * start of the region that was found. Otherwise this function will + * return false and keep the offset argument unchanged. + */ +STATIC bool +xfs_lookup_buffer_offset( + struct page *page, + loff_t *offset, + unsigned int type) +{ + loff_t lastoff = page_offset(page); + bool found = false; + struct buffer_head *bh, *head; + + bh = head = page_buffers(page); + do { + /* + * Unwritten extents that have data in the page + * cache covering them can be identified by the + * BH_Unwritten state flag. Pages with multiple + * buffers might have a mix of holes, data and + * unwritten extents - any buffer with valid + * data in it should have BH_Uptodate flag set + * on it. + */ + if (buffer_unwritten(bh) || + buffer_uptodate(bh)) { + if (type == DATA_OFF) + found = true; + } else { + if (type == HOLE_OFF) + found = true; + } + + if (found) { + *offset = lastoff; + break; + } + lastoff += bh->b_size; + } while ((bh = bh->b_this_page) != head); + + return found; +} + +/* + * This routine is called to find out and return a data or hole offset + * from the page cache for unwritten extents according to the desired + * type for xfs_seek_data() or xfs_seek_hole(). + * + * The argument offset is used to tell where we start to search from the + * page cache. Map is used to figure out the end points of the range to + * lookup pages. + * + * Return true if the desired type of offset was found, and the argument + * offset is filled with that address. Otherwise, return false and keep + * offset unchanged. + */ +STATIC bool +xfs_find_get_desired_pgoff( + struct inode *inode, + struct xfs_bmbt_irec *map, + unsigned int type, + loff_t *offset) +{ + struct xfs_inode *ip = XFS_I(inode); + struct xfs_mount *mp = ip->i_mount; + struct pagevec pvec; + pgoff_t index; + pgoff_t end; + loff_t endoff; + loff_t startoff = *offset; + loff_t lastoff = startoff; + bool found = false; + + pagevec_init(&pvec, 0); + + index = startoff >> PAGE_CACHE_SHIFT; + endoff = XFS_FSB_TO_B(mp, map->br_startoff + map->br_blockcount); + end = endoff >> PAGE_CACHE_SHIFT; + do { + int want; + unsigned nr_pages; + unsigned int i; + + want = min_t(pgoff_t, end - index, PAGEVEC_SIZE); + nr_pages = pagevec_lookup(&pvec, inode->i_mapping, index, + want); + /* + * No page mapped into given range. If we are searching holes + * and if this is the first time we got into the loop, it means + * that the given offset is landed in a hole, return it. + * + * If we have already stepped through some block buffers to find + * holes but they all contains data. In this case, the last + * offset is already updated and pointed to the end of the last + * mapped page, if it does not reach the endpoint to search, + * that means there should be a hole between them. + */ + if (nr_pages == 0) { + /* Data search found nothing */ + if (type == DATA_OFF) + break; + + ASSERT(type == HOLE_OFF); + if (lastoff == startoff || lastoff < endoff) { + found = true; + *offset = lastoff; + } + break; + } + + /* + * At lease we found one page. If this is the first time we + * step into the loop, and if the first page index offset is + * greater than the given search offset, a hole was found. + */ + if (type == HOLE_OFF && lastoff == startoff && + lastoff < page_offset(pvec.pages[0])) { + found = true; + break; + } + + for (i = 0; i < nr_pages; i++) { + struct page *page = pvec.pages[i]; + loff_t b_offset; + + /* + * At this point, the page may be truncated or + * invalidated (changing page->mapping to NULL), + * or even swizzled back from swapper_space to tmpfs + * file mapping. However, page->index will not change + * because we have a reference on the page. + * + * Searching done if the page index is out of range. + * If the current offset is not reaches the end of + * the specified search range, there should be a hole + * between them. + */ + if (page->index > end) { + if (type == HOLE_OFF && lastoff < endoff) { + *offset = lastoff; + found = true; + } + goto out; + } + + lock_page(page); + /* + * Page truncated or invalidated(page->mapping == NULL). + * We can freely skip it and proceed to check the next + * page. + */ + if (unlikely(page->mapping != inode->i_mapping)) { + unlock_page(page); + continue; + } + + if (!page_has_buffers(page)) { + unlock_page(page); + continue; + } + + found = xfs_lookup_buffer_offset(page, &b_offset, type); + if (found) { + /* + * The found offset may be less than the start + * point to search if this is the first time to + * come here. + */ + *offset = max_t(loff_t, startoff, b_offset); + unlock_page(page); + goto out; + } + + /* + * We either searching data but nothing was found, or + * searching hole but found a data buffer. In either + * case, probably the next page contains the desired + * things, update the last offset to it so. + */ + lastoff = page_offset(page) + PAGE_SIZE; + unlock_page(page); + } + + /* + * The number of returned pages less than our desired, search + * done. In this case, nothing was found for searching data, + * but we found a hole behind the last offset. + */ + if (nr_pages < want) { + if (type == HOLE_OFF) { + *offset = lastoff; + found = true; + } + break; + } + + index = pvec.pages[i - 1]->index + 1; + pagevec_release(&pvec); + } while (index <= end); + +out: + pagevec_release(&pvec); + return found; +} + STATIC loff_t xfs_seek_data( struct file *file, -- cgit v1.2.3-58-ga151 From 52f1acc8b56a333fbc7218711c3fa2fb3bf78b92 Mon Sep 17 00:00:00 2001 From: Jeff Liu Date: Tue, 21 Aug 2012 17:12:07 +0800 Subject: xfs: xfs_seek_data() refinement with unwritten extents check up from page cache xfs_seek_data() refinement with unwritten extents check up from page cache. Signed-off-by: Jie Liu Reviewed-by: Mark Tinguely Reviewed-by: Dave Chinner Signed-off-by: Ben Myers --- fs/xfs/xfs_file.c | 72 +++++++++++++++++++++++++++++++++++++++++-------------- 1 file changed, 54 insertions(+), 18 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index d78a746b6c7c..3f9107431dfb 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -1186,8 +1186,6 @@ xfs_seek_data( struct inode *inode = file->f_mapping->host; struct xfs_inode *ip = XFS_I(inode); struct xfs_mount *mp = ip->i_mount; - struct xfs_bmbt_irec map[2]; - int nmap = 2; loff_t uninitialized_var(offset); xfs_fsize_t isize; xfs_fileoff_t fsbno; @@ -1203,36 +1201,74 @@ xfs_seek_data( goto out_unlock; } - fsbno = XFS_B_TO_FSBT(mp, start); - /* * Try to read extents from the first block indicated * by fsbno to the end block of the file. */ + fsbno = XFS_B_TO_FSBT(mp, start); end = XFS_B_TO_FSB(mp, isize); + for (;;) { + struct xfs_bmbt_irec map[2]; + int nmap = 2; + unsigned int i; - error = xfs_bmapi_read(ip, fsbno, end - fsbno, map, &nmap, - XFS_BMAPI_ENTIRE); - if (error) - goto out_unlock; + error = xfs_bmapi_read(ip, fsbno, end - fsbno, map, &nmap, + XFS_BMAPI_ENTIRE); + if (error) + goto out_unlock; - /* - * Treat unwritten extent as data extent since it might - * contains dirty data in page cache. - */ - if (map[0].br_startblock != HOLESTARTBLOCK) { - offset = max_t(loff_t, start, - XFS_FSB_TO_B(mp, map[0].br_startoff)); - } else { + /* No extents at given offset, must be beyond EOF */ + if (nmap == 0) { + error = ENXIO; + goto out_unlock; + } + + for (i = 0; i < nmap; i++) { + offset = max_t(loff_t, start, + XFS_FSB_TO_B(mp, map[i].br_startoff)); + + /* Landed in a data extent */ + if (map[i].br_startblock == DELAYSTARTBLOCK || + (map[i].br_state == XFS_EXT_NORM && + !isnullstartblock(map[i].br_startblock))) + goto out; + + /* + * Landed in an unwritten extent, try to search data + * from page cache. + */ + if (map[i].br_state == XFS_EXT_UNWRITTEN) { + if (xfs_find_get_desired_pgoff(inode, &map[i], + DATA_OFF, &offset)) + goto out; + } + } + + /* + * map[0] is hole or its an unwritten extent but + * without data in page cache. Probably means that + * we are reading after EOF if nothing in map[1]. + */ if (nmap == 1) { error = ENXIO; goto out_unlock; } - offset = max_t(loff_t, start, - XFS_FSB_TO_B(mp, map[1].br_startoff)); + ASSERT(i > 1); + + /* + * Nothing was found, proceed to the next round of search + * if reading offset not beyond or hit EOF. + */ + fsbno = map[i - 1].br_startoff + map[i - 1].br_blockcount; + start = XFS_FSB_TO_B(mp, fsbno); + if (start >= isize) { + error = ENXIO; + goto out_unlock; + } } +out: if (offset != file->f_pos) file->f_pos = offset; -- cgit v1.2.3-58-ga151 From b686d1f79acb65c6a34473c15fcfa2ee54aed8e2 Mon Sep 17 00:00:00 2001 From: Jeff Liu Date: Tue, 21 Aug 2012 17:12:18 +0800 Subject: xfs: xfs_seek_hole() refinement with hole searching from page cache for unwritten extents xfs_seek_hole() refinement with hole searching from page cache for unwritten extent. Signed-off-by: Jie Liu Reviewed-by: Mark Tinguely Reviewed-by: Dave Chinner Signed-off-by: Ben Myers --- fs/xfs/xfs_file.c | 78 +++++++++++++++++++++++++++++++++++++++++++++++-------- 1 file changed, 67 insertions(+), 11 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index 3f9107431dfb..1eaeb8be3aae 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -1289,9 +1289,9 @@ xfs_seek_hole( struct xfs_inode *ip = XFS_I(inode); struct xfs_mount *mp = ip->i_mount; loff_t uninitialized_var(offset); - loff_t holeoff; xfs_fsize_t isize; xfs_fileoff_t fsbno; + xfs_filblks_t end; uint lock; int error; @@ -1307,21 +1307,77 @@ xfs_seek_hole( } fsbno = XFS_B_TO_FSBT(mp, start); - error = xfs_bmap_first_unused(NULL, ip, 1, &fsbno, XFS_DATA_FORK); - if (error) - goto out_unlock; + end = XFS_B_TO_FSB(mp, isize); + + for (;;) { + struct xfs_bmbt_irec map[2]; + int nmap = 2; + unsigned int i; + + error = xfs_bmapi_read(ip, fsbno, end - fsbno, map, &nmap, + XFS_BMAPI_ENTIRE); + if (error) + goto out_unlock; + + /* No extents at given offset, must be beyond EOF */ + if (nmap == 0) { + error = ENXIO; + goto out_unlock; + } + + for (i = 0; i < nmap; i++) { + offset = max_t(loff_t, start, + XFS_FSB_TO_B(mp, map[i].br_startoff)); + + /* Landed in a hole */ + if (map[i].br_startblock == HOLESTARTBLOCK) + goto out; + + /* + * Landed in an unwritten extent, try to search hole + * from page cache. + */ + if (map[i].br_state == XFS_EXT_UNWRITTEN) { + if (xfs_find_get_desired_pgoff(inode, &map[i], + HOLE_OFF, &offset)) + goto out; + } + } - holeoff = XFS_FSB_TO_B(mp, fsbno); - if (holeoff <= start) - offset = start; - else { /* - * xfs_bmap_first_unused() could return a value bigger than - * isize if there are no more holes past the supplied offset. + * map[0] contains data or its unwritten but contains + * data in page cache, probably means that we are + * reading after EOF. We should fix offset to point + * to the end of the file(i.e., there is an implicit + * hole at the end of any file). */ - offset = min_t(loff_t, holeoff, isize); + if (nmap == 1) { + offset = isize; + break; + } + + ASSERT(i > 1); + + /* + * Both mappings contains data, proceed to the next round of + * search if the current reading offset not beyond or hit EOF. + */ + fsbno = map[i - 1].br_startoff + map[i - 1].br_blockcount; + start = XFS_FSB_TO_B(mp, fsbno); + if (start >= isize) { + offset = isize; + break; + } } +out: + /* + * At this point, we must have found a hole. However, the returned + * offset may be bigger than the file size as it may be aligned to + * page boundary for unwritten extents, we need to deal with this + * situation in particular. + */ + offset = min_t(loff_t, offset, isize); if (offset != file->f_pos) file->f_pos = offset; -- cgit v1.2.3-58-ga151 From 4026c9fde9c67266932afd209e25bfef4474a1be Mon Sep 17 00:00:00 2001 From: Ben Myers Date: Thu, 13 Sep 2012 16:18:47 -0500 Subject: xfs: stop the sync worker before xfs_unmountfs Cancel work of the xfs_sync_worker before teardown of the log in xfs_unmountfs. This prevents occasional crashes on unmount like so: PID: 21602 TASK: ee9df060 CPU: 0 COMMAND: "kworker/0:3" #0 [c5377d28] crash_kexec at c0292c94 #1 [c5377d80] oops_end at c07090c2 #2 [c5377d98] no_context at c06f614e #3 [c5377dbc] __bad_area_nosemaphore at c06f6281 #4 [c5377df4] bad_area_nosemaphore at c06f629b #5 [c5377e00] do_page_fault at c070b0cb #6 [c5377e7c] error_code (via page_fault) at c070892c EAX: f300c6a8 EBX: f300c6a8 ECX: 000000c0 EDX: 000000c0 EBP: c5377ed0 DS: 007b ESI: 00000000 ES: 007b EDI: 00000001 GS: ffffad20 CS: 0060 EIP: c0481ad0 ERR: ffffffff EFLAGS: 00010246 #7 [c5377eb0] atomic64_read_cx8 at c0481ad0 #8 [c5377ebc] xlog_assign_tail_lsn_locked at f7cc7c6e [xfs] #9 [c5377ed4] xfs_trans_ail_delete_bulk at f7ccd520 [xfs] #10 [c5377f0c] xfs_buf_iodone at f7ccb602 [xfs] #11 [c5377f24] xfs_buf_do_callbacks at f7cca524 [xfs] #12 [c5377f30] xfs_buf_iodone_callbacks at f7cca5da [xfs] #13 [c5377f4c] xfs_buf_iodone_work at f7c718d0 [xfs] #14 [c5377f58] process_one_work at c024ee4c #15 [c5377f98] worker_thread at c024f43d #16 [c5377fbc] kthread at c025326b #17 [c5377fe8] kernel_thread_helper at c070e834 PID: 26653 TASK: e79143b0 CPU: 3 COMMAND: "umount" #0 [cde0fda0] __schedule at c0706595 #1 [cde0fe28] schedule at c0706b89 #2 [cde0fe30] schedule_timeout at c0705600 #3 [cde0fe94] __down_common at c0706098 #4 [cde0fec8] __down at c0706122 #5 [cde0fed0] down at c025936f #6 [cde0fee0] xfs_buf_lock at f7c7131d [xfs] #7 [cde0ff00] xfs_freesb at f7cc2236 [xfs] #8 [cde0ff10] xfs_fs_put_super at f7c80f21 [xfs] #9 [cde0ff1c] generic_shutdown_super at c0333d7a #10 [cde0ff38] kill_block_super at c0333e0f #11 [cde0ff48] deactivate_locked_super at c0334218 #12 [cde0ff58] deactivate_super at c033495d #13 [cde0ff68] mntput_no_expire at c034bc13 #14 [cde0ff7c] sys_umount at c034cc69 #15 [cde0ffa0] sys_oldumount at c034ccd4 #16 [cde0ffb0] system_call at c0707e66 commit 11159a05 added this to xfs_log_unmount and needs to be cleaned up at a later date. Signed-off-by: Ben Myers Reviewed-by: Dave Chinner Reviewed-by: Mark Tinguely --- fs/xfs/xfs_super.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index bdaf4cb9f4a2..19e2380fb867 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -919,6 +919,7 @@ xfs_fs_put_super( struct xfs_mount *mp = XFS_M(sb); xfs_filestream_unmount(mp); + cancel_delayed_work_sync(&mp->m_sync_work); xfs_unmountfs(mp); xfs_syncd_stop(mp); xfs_freesb(mp); -- cgit v1.2.3-58-ga151 From c3a58fecdd1934a8538ada9073107625f5151687 Mon Sep 17 00:00:00 2001 From: Carlos Maiolino Date: Fri, 17 Aug 2012 18:19:38 -0300 Subject: Make inode64 a remountable option Actually, there is no reason about why a user must umount and mount a XFS filesystem to enable 'inode64' option. So, this patch makes this a remountable option. Signed-off-by: Carlos Maiolino Reviewed-by: Brian Foster Reviewed-by: Dave Chinner Signed-off-by: Ben Myers --- fs/xfs/xfs_super.c | 30 +++++++++++++++++++++++++++++- 1 file changed, 29 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index 19e2380fb867..c416a01fcb14 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -120,12 +120,13 @@ mempool_t *xfs_ioend_pool; * in the future, too. */ enum { - Opt_barrier, Opt_nobarrier, Opt_err + Opt_barrier, Opt_nobarrier, Opt_inode64, Opt_err }; static const match_table_t tokens = { {Opt_barrier, "barrier"}, {Opt_nobarrier, "nobarrier"}, + {Opt_inode64, "inode64"}, {Opt_err, NULL} }; @@ -1031,6 +1032,30 @@ xfs_restore_resvblks(struct xfs_mount *mp) xfs_reserve_blocks(mp, &resblks, NULL); } +STATIC void +xfs_set_inode64(struct xfs_mount *mp) +{ + int i = 0; + + for (i = 0; i < mp->m_sb.sb_agcount; i++) { + struct xfs_perag *pag; + + pag = xfs_perag_get(mp, i); + pag->pagi_inodeok = 1; + pag->pagf_metadata = 0; + xfs_perag_put(pag); + } + + /* There is no need for lock protection on m_flags, + * the rw_semaphore of the VFS superblock is locked + * during mount/umount/remount operations, so this is + * enough to avoid concurency on the m_flags field + */ + mp->m_flags &= ~(XFS_MOUNT_32BITINODES | + XFS_MOUNT_SMALL_INUMS); + mp->m_maxagi = i; +} + STATIC int xfs_fs_remount( struct super_block *sb, @@ -1056,6 +1081,9 @@ xfs_fs_remount( case Opt_nobarrier: mp->m_flags &= ~XFS_MOUNT_BARRIER; break; + case Opt_inode64: + xfs_set_inode64(mp); + break; default: /* * Logically we would return an error here to prevent -- cgit v1.2.3-58-ga151 From 8aea3ff411b2ce8fe7b46644298ed243a920eb24 Mon Sep 17 00:00:00 2001 From: Carlos Maiolino Date: Thu, 20 Sep 2012 10:32:36 -0300 Subject: xfs: Fix m_agirotor reset during AG selection xfs_ialloc_next_ag() currently resets m_agirotor when it is equal to m_maxagi: if (++mp->m_agirotor == mp->m_maxagi) mp->m_agirotor = 0; But, if for some reason mp->m_maxagi changes to a lower value than current m_agirotor, this condition will never be true, causing m_agirotor to exceed the maximum allowed value (m_maxagi). This implies mainly during lookups for xfs_perag structs in its radix tree, since the agno value used for the lookup is based on m_agirotor. An out-of-range m_agirotor may cause a lookup failure which in case will return NULL. As an example, the value of m_maxagi is decreased during inode64->inode32 remount process, case where I've found this problem. Signed-off-by: Carlos Maiolino Reviewed-by: Christoph Hellwig Reviewed-by: Mark Tinguely Signed-off-by: Ben Myers --- fs/xfs/xfs_ialloc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/xfs/xfs_ialloc.c b/fs/xfs/xfs_ialloc.c index 5aceb3f8ecd6..445bf1aef31c 100644 --- a/fs/xfs/xfs_ialloc.c +++ b/fs/xfs/xfs_ialloc.c @@ -431,7 +431,7 @@ xfs_ialloc_next_ag( spin_lock(&mp->m_agirotor_lock); agno = mp->m_agirotor; - if (++mp->m_agirotor == mp->m_maxagi) + if (++mp->m_agirotor >= mp->m_maxagi) mp->m_agirotor = 0; spin_unlock(&mp->m_agirotor_lock); -- cgit v1.2.3-58-ga151 From 08bf540412ed82a15cb9068249ad49b410a7b082 Mon Sep 17 00:00:00 2001 From: Carlos Maiolino Date: Thu, 20 Sep 2012 10:32:37 -0300 Subject: xfs: make inode64 as the default allocation mode since 64-bit inodes can be accessed while using inode32, and these can also be used on 32-bit kernels, there is no reason to still keep inode32 as the default mount option. If the filesystem cannot handle 64bit inode numbers (i.e CONFIG_LBDAF is not enabled and BITS_PER_LONG == 32), XFS_MOUNT_SMALL_INUMS will still be set by default, so inode64 is not an unconditional default value. Signed-off-by: Carlos Maiolino Reviewed-by: Christoph Hellwig Reviewed-by: Mark Tinguely Signed-off-by: Ben Myers --- fs/xfs/xfs_super.c | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'fs') diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index c416a01fcb14..996257d36fd1 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -88,6 +88,8 @@ mempool_t *xfs_ioend_pool; * unwritten extent conversion */ #define MNTOPT_NOBARRIER "nobarrier" /* .. disable */ #define MNTOPT_64BITINODE "inode64" /* inodes can be allocated anywhere */ +#define MNTOPT_32BITINODE "inode32" /* inode allocation limited to + * XFS_MAXINUMBER_32 */ #define MNTOPT_IKEEP "ikeep" /* do not free empty inode clusters */ #define MNTOPT_NOIKEEP "noikeep" /* free empty inode clusters */ #define MNTOPT_LARGEIO "largeio" /* report large I/O sizes in stat() */ @@ -198,7 +200,9 @@ xfs_parseargs( */ mp->m_flags |= XFS_MOUNT_BARRIER; mp->m_flags |= XFS_MOUNT_COMPAT_IOSIZE; +#if !XFS_BIG_INUMS mp->m_flags |= XFS_MOUNT_SMALL_INUMS; +#endif /* * These can be overridden by the mount option parsing. @@ -295,6 +299,8 @@ xfs_parseargs( return EINVAL; } dswidth = simple_strtoul(value, &eov, 10); + } else if (!strcmp(this_char, MNTOPT_32BITINODE)) { + mp->m_flags |= XFS_MOUNT_SMALL_INUMS; } else if (!strcmp(this_char, MNTOPT_64BITINODE)) { mp->m_flags &= ~XFS_MOUNT_SMALL_INUMS; #if !XFS_BIG_INUMS @@ -493,6 +499,7 @@ xfs_showargs( { XFS_MOUNT_FILESTREAMS, "," MNTOPT_FILESTREAM }, { XFS_MOUNT_GRPID, "," MNTOPT_GRPID }, { XFS_MOUNT_DISCARD, "," MNTOPT_DISCARD }, + { XFS_MOUNT_SMALL_INUMS, "," MNTOPT_32BITINODE }, { 0, NULL } }; static struct proc_xfs_info xfs_info_unset[] = { -- cgit v1.2.3-58-ga151 From 2d2194f61fddab3a9731b6e7a7ae3a4a19dd810c Mon Sep 17 00:00:00 2001 From: Carlos Maiolino Date: Thu, 20 Sep 2012 10:32:38 -0300 Subject: xfs: reduce code duplication handling inode32/64 options Add xfs_set_inode32() to be used to enable inode32 allocation mode. this will reduce the amount of duplicated code needed to mount/remount a filesystem with inode32 option. This patch also changes xfs_set_inode64() to return the maximum AG number that inodes can be allocated instead of set mp->m_maxagi by itself, so that the behaviour is the same as xfs_set_inode32(). This simplifies code that calls these functions and needs to know the maximum AG that inodes can be allocated in. Signed-off-by: Carlos Maiolino Reviewed-by: Christoph Hellwig Reviewed-by: Mark Tinguely Signed-off-by: Ben Myers --- fs/xfs/xfs_mount.c | 43 +++----------------------- fs/xfs/xfs_super.c | 89 +++++++++++++++++++++++++++++++++++++++--------------- fs/xfs/xfs_super.h | 2 ++ 3 files changed, 72 insertions(+), 62 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index 29c2f83d4147..b2bd3a0e6376 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -440,7 +440,7 @@ xfs_initialize_perag( xfs_agnumber_t agcount, xfs_agnumber_t *maxagi) { - xfs_agnumber_t index, max_metadata; + xfs_agnumber_t index; xfs_agnumber_t first_initialised = 0; xfs_perag_t *pag; xfs_agino_t agino; @@ -500,43 +500,10 @@ xfs_initialize_perag( else mp->m_flags &= ~XFS_MOUNT_32BITINODES; - if (mp->m_flags & XFS_MOUNT_32BITINODES) { - /* - * Calculate how much should be reserved for inodes to meet - * the max inode percentage. - */ - if (mp->m_maxicount) { - __uint64_t icount; - - icount = sbp->sb_dblocks * sbp->sb_imax_pct; - do_div(icount, 100); - icount += sbp->sb_agblocks - 1; - do_div(icount, sbp->sb_agblocks); - max_metadata = icount; - } else { - max_metadata = agcount; - } - - for (index = 0; index < agcount; index++) { - ino = XFS_AGINO_TO_INO(mp, index, agino); - if (ino > XFS_MAXINUMBER_32) { - index++; - break; - } - - pag = xfs_perag_get(mp, index); - pag->pagi_inodeok = 1; - if (index < max_metadata) - pag->pagf_metadata = 1; - xfs_perag_put(pag); - } - } else { - for (index = 0; index < agcount; index++) { - pag = xfs_perag_get(mp, index); - pag->pagi_inodeok = 1; - xfs_perag_put(pag); - } - } + if (mp->m_flags & XFS_MOUNT_32BITINODES) + index = xfs_set_inode32(mp); + else + index = xfs_set_inode64(mp); if (maxagi) *maxagi = index; diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index 996257d36fd1..d6619d685531 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -599,6 +599,71 @@ xfs_max_file_offset( return (((__uint64_t)pagefactor) << bitshift) - 1; } +xfs_agnumber_t +xfs_set_inode32(struct xfs_mount *mp) +{ + xfs_agnumber_t index = 0; + xfs_sb_t *sbp = &mp->m_sb; + xfs_agnumber_t max_metadata; + xfs_agino_t agino = XFS_OFFBNO_TO_AGINO(mp, sbp->sb_agblocks -1, 0); + xfs_ino_t ino = XFS_AGINO_TO_INO(mp, sbp->sb_agcount -1, agino); + xfs_perag_t *pag; + + /* Calculate how much should be reserved for inodes to meet + * the max inode percentage. + */ + if (mp->m_maxicount) { + __uint64_t icount; + + icount = sbp->sb_dblocks * sbp->sb_imax_pct; + do_div(icount, 100); + icount += sbp->sb_agblocks - 1; + do_div(icount, sbp->sb_agblocks); + max_metadata = icount; + } else { + max_metadata = sbp->sb_agcount; + } + + for (index = 0; index < sbp->sb_agcount; index++) { + ino = XFS_AGINO_TO_INO(mp, index, agino); + if (ino > XFS_MAXINUMBER_32) { + index++; + break; + } + + pag = xfs_perag_get(mp, index); + pag->pagi_inodeok = 1; + if (index < max_metadata) + pag->pagf_metadata = 1; + xfs_perag_put(pag); + } + return index; +} + +xfs_agnumber_t +xfs_set_inode64(struct xfs_mount *mp) +{ + xfs_agnumber_t index = 0; + + for (index = 0; index < mp->m_sb.sb_agcount; index++) { + struct xfs_perag *pag; + + pag = xfs_perag_get(mp, index); + pag->pagi_inodeok = 1; + pag->pagf_metadata = 0; + xfs_perag_put(pag); + } + + /* There is no need for lock protection on m_flags, + * the rw_semaphore of the VFS superblock is locked + * during mount/umount/remount operations, so this is + * enough to avoid concurency on the m_flags field + */ + mp->m_flags &= ~(XFS_MOUNT_32BITINODES | + XFS_MOUNT_SMALL_INUMS); + return index; +} + STATIC int xfs_blkdev_get( xfs_mount_t *mp, @@ -1039,30 +1104,6 @@ xfs_restore_resvblks(struct xfs_mount *mp) xfs_reserve_blocks(mp, &resblks, NULL); } -STATIC void -xfs_set_inode64(struct xfs_mount *mp) -{ - int i = 0; - - for (i = 0; i < mp->m_sb.sb_agcount; i++) { - struct xfs_perag *pag; - - pag = xfs_perag_get(mp, i); - pag->pagi_inodeok = 1; - pag->pagf_metadata = 0; - xfs_perag_put(pag); - } - - /* There is no need for lock protection on m_flags, - * the rw_semaphore of the VFS superblock is locked - * during mount/umount/remount operations, so this is - * enough to avoid concurency on the m_flags field - */ - mp->m_flags &= ~(XFS_MOUNT_32BITINODES | - XFS_MOUNT_SMALL_INUMS); - mp->m_maxagi = i; -} - STATIC int xfs_fs_remount( struct super_block *sb, diff --git a/fs/xfs/xfs_super.h b/fs/xfs/xfs_super.h index 09b0c26b2245..9de4a920ba05 100644 --- a/fs/xfs/xfs_super.h +++ b/fs/xfs/xfs_super.h @@ -75,6 +75,8 @@ struct block_device; extern __uint64_t xfs_max_file_offset(unsigned int); extern void xfs_blkdev_issue_flush(struct xfs_buftarg *); +extern xfs_agnumber_t xfs_set_inode32(struct xfs_mount *); +extern xfs_agnumber_t xfs_set_inode64(struct xfs_mount *); extern const struct export_operations xfs_export_operations; extern const struct xattr_handler *xfs_xattr_handlers[]; -- cgit v1.2.3-58-ga151 From 4c0837224c677db35cd85b04a77504c496cadb66 Mon Sep 17 00:00:00 2001 From: Carlos Maiolino Date: Thu, 20 Sep 2012 10:32:39 -0300 Subject: xfs: Fix mp->m_maxagi update during inode64 remount With the changes made on xfs_set_inode64(), to make it behave as xfs_set_inode32() (now leaving to the caller the responsibility to update mp->m_maxagi), we use the return value of xfs_set_inode64() to update mp->m_maxagi during remount. Signed-off-by: Carlos Maiolino Reviewed-by: Mark Tinguely Signed-off-by: Ben Myers --- fs/xfs/xfs_super.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index d6619d685531..aeb03f9a8967 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -1130,7 +1130,7 @@ xfs_fs_remount( mp->m_flags &= ~XFS_MOUNT_BARRIER; break; case Opt_inode64: - xfs_set_inode64(mp); + mp->m_maxagi = xfs_set_inode64(mp); break; default: /* -- cgit v1.2.3-58-ga151 From 4056c1d08d2a7c50ae7414db7c1783ba45b4835d Mon Sep 17 00:00:00 2001 From: Carlos Maiolino Date: Thu, 20 Sep 2012 10:32:40 -0300 Subject: xfs: add inode64->inode32 transition into xfs_set_inode32() To make inode32 a remountable option, xfs_set_inode32() should be able to make a transition from inode64 option, disabling inode allocation on higher AGs. Signed-off-by: Carlos Maiolino Reviewed-by: Christoph Hellwig Reviewed-by: Mark Tinguely Signed-off-by: Ben Myers --- fs/xfs/xfs_super.c | 15 ++++++++++++--- 1 file changed, 12 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index aeb03f9a8967..168d4984ce89 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -603,6 +603,7 @@ xfs_agnumber_t xfs_set_inode32(struct xfs_mount *mp) { xfs_agnumber_t index = 0; + xfs_agnumber_t maxagi = 0; xfs_sb_t *sbp = &mp->m_sb; xfs_agnumber_t max_metadata; xfs_agino_t agino = XFS_OFFBNO_TO_AGINO(mp, sbp->sb_agblocks -1, 0); @@ -626,18 +627,26 @@ xfs_set_inode32(struct xfs_mount *mp) for (index = 0; index < sbp->sb_agcount; index++) { ino = XFS_AGINO_TO_INO(mp, index, agino); + if (ino > XFS_MAXINUMBER_32) { - index++; - break; + pag = xfs_perag_get(mp, index); + pag->pagi_inodeok = 0; + pag->pagf_metadata = 0; + xfs_perag_put(pag); + continue; } pag = xfs_perag_get(mp, index); pag->pagi_inodeok = 1; + maxagi++; if (index < max_metadata) pag->pagf_metadata = 1; xfs_perag_put(pag); } - return index; + mp->m_flags |= (XFS_MOUNT_32BITINODES | + XFS_MOUNT_SMALL_INUMS); + + return maxagi; } xfs_agnumber_t -- cgit v1.2.3-58-ga151 From 2ea0392983a82f7dc3055568ae0f2558724d119b Mon Sep 17 00:00:00 2001 From: Carlos Maiolino Date: Thu, 20 Sep 2012 10:32:41 -0300 Subject: xfs: Make inode32 a remountable option As inode64 is the default option now, and was also made remountable previously, inode32 can also be remounted on-the-fly when it is needed. Signed-off-by: Carlos Maiolino Reviewed-by: Christoph Hellwig Reviewed-by: Mark Tinguely Signed-off-by: Ben Myers --- fs/xfs/xfs_super.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index 168d4984ce89..d93f2c7364cc 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -122,13 +122,18 @@ mempool_t *xfs_ioend_pool; * in the future, too. */ enum { - Opt_barrier, Opt_nobarrier, Opt_inode64, Opt_err + Opt_barrier, + Opt_nobarrier, + Opt_inode64, + Opt_inode32, + Opt_err }; static const match_table_t tokens = { {Opt_barrier, "barrier"}, {Opt_nobarrier, "nobarrier"}, {Opt_inode64, "inode64"}, + {Opt_inode32, "inode32"}, {Opt_err, NULL} }; @@ -1141,6 +1146,9 @@ xfs_fs_remount( case Opt_inode64: mp->m_maxagi = xfs_set_inode64(mp); break; + case Opt_inode32: + mp->m_maxagi = xfs_set_inode32(mp); + break; default: /* * Logically we would return an error here to prevent -- cgit v1.2.3-58-ga151