From 45bce8f3e3436bbe2e03dd2b076abdce79ffabb7 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Thu, 29 Nov 2012 10:21:43 -0800 Subject: fs/buffer.c: make block-size be per-page and protected by the page lock This makes the buffer size handling be a per-page thing, which allows us to not have to worry about locking too much when changing the buffer size. If a page doesn't have buffers, we still need to read the block size from the inode, but we can do that with ACCESS_ONCE(), so that even if the size is changing, we get a consistent value. This doesn't convert all functions - many of the buffer functions are used purely by filesystems, which in turn results in the buffer size being fixed at mount-time. So they don't have the same consistency issues that the raw device access can have. Signed-off-by: Linus Torvalds --- fs/buffer.c | 79 +++++++++++++++++++++++++++++++++++++------------------------ 1 file changed, 48 insertions(+), 31 deletions(-) (limited to 'fs/buffer.c') diff --git a/fs/buffer.c b/fs/buffer.c index b5f044283edb..28a74ff5324b 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -1552,6 +1552,28 @@ void unmap_underlying_metadata(struct block_device *bdev, sector_t block) } EXPORT_SYMBOL(unmap_underlying_metadata); +/* + * Size is a power-of-two in the range 512..PAGE_SIZE, + * and the case we care about most is PAGE_SIZE. + * + * So this *could* possibly be written with those + * constraints in mind (relevant mostly if some + * architecture has a slow bit-scan instruction) + */ +static inline int block_size_bits(unsigned int blocksize) +{ + return ilog2(blocksize); +} + +static struct buffer_head *create_page_buffers(struct page *page, struct inode *inode, unsigned int b_state) +{ + BUG_ON(!PageLocked(page)); + + if (!page_has_buffers(page)) + create_empty_buffers(page, 1 << ACCESS_ONCE(inode->i_blkbits), b_state); + return page_buffers(page); +} + /* * NOTE! All mapped/uptodate combinations are valid: * @@ -1589,19 +1611,13 @@ static int __block_write_full_page(struct inode *inode, struct page *page, sector_t block; sector_t last_block; struct buffer_head *bh, *head; - const unsigned blocksize = 1 << inode->i_blkbits; + unsigned int blocksize, bbits; int nr_underway = 0; int write_op = (wbc->sync_mode == WB_SYNC_ALL ? WRITE_SYNC : WRITE); - BUG_ON(!PageLocked(page)); - - last_block = (i_size_read(inode) - 1) >> inode->i_blkbits; - - if (!page_has_buffers(page)) { - create_empty_buffers(page, blocksize, + head = create_page_buffers(page, inode, (1 << BH_Dirty)|(1 << BH_Uptodate)); - } /* * Be very careful. We have no exclusion from __set_page_dirty_buffers @@ -1613,9 +1629,12 @@ static int __block_write_full_page(struct inode *inode, struct page *page, * handle that here by just cleaning them. */ - block = (sector_t)page->index << (PAGE_CACHE_SHIFT - inode->i_blkbits); - head = page_buffers(page); bh = head; + blocksize = bh->b_size; + bbits = block_size_bits(blocksize); + + block = (sector_t)page->index << (PAGE_CACHE_SHIFT - bbits); + last_block = (i_size_read(inode) - 1) >> bbits; /* * Get all the dirty buffers mapped to disk addresses and @@ -1806,12 +1825,10 @@ int __block_write_begin(struct page *page, loff_t pos, unsigned len, BUG_ON(to > PAGE_CACHE_SIZE); BUG_ON(from > to); - blocksize = 1 << inode->i_blkbits; - if (!page_has_buffers(page)) - create_empty_buffers(page, blocksize, 0); - head = page_buffers(page); + head = create_page_buffers(page, inode, 0); + blocksize = head->b_size; + bbits = block_size_bits(blocksize); - bbits = inode->i_blkbits; block = (sector_t)page->index << (PAGE_CACHE_SHIFT - bbits); for(bh = head, block_start = 0; bh != head || !block_start; @@ -1881,11 +1898,11 @@ static int __block_commit_write(struct inode *inode, struct page *page, unsigned blocksize; struct buffer_head *bh, *head; - blocksize = 1 << inode->i_blkbits; + bh = head = page_buffers(page); + blocksize = bh->b_size; - for(bh = head = page_buffers(page), block_start = 0; - bh != head || !block_start; - block_start=block_end, bh = bh->b_this_page) { + block_start = 0; + do { block_end = block_start + blocksize; if (block_end <= from || block_start >= to) { if (!buffer_uptodate(bh)) @@ -1895,7 +1912,10 @@ static int __block_commit_write(struct inode *inode, struct page *page, mark_buffer_dirty(bh); } clear_buffer_new(bh); - } + + block_start = block_end; + bh = bh->b_this_page; + } while (bh != head); /* * If this is a partial write which happened to make all buffers @@ -2020,7 +2040,6 @@ EXPORT_SYMBOL(generic_write_end); int block_is_partially_uptodate(struct page *page, read_descriptor_t *desc, unsigned long from) { - struct inode *inode = page->mapping->host; unsigned block_start, block_end, blocksize; unsigned to; struct buffer_head *bh, *head; @@ -2029,13 +2048,13 @@ int block_is_partially_uptodate(struct page *page, read_descriptor_t *desc, if (!page_has_buffers(page)) return 0; - blocksize = 1 << inode->i_blkbits; + head = page_buffers(page); + blocksize = head->b_size; to = min_t(unsigned, PAGE_CACHE_SIZE - from, desc->count); to = from + to; if (from < blocksize && to > PAGE_CACHE_SIZE - blocksize) return 0; - head = page_buffers(page); bh = head; block_start = 0; do { @@ -2068,18 +2087,16 @@ int block_read_full_page(struct page *page, get_block_t *get_block) struct inode *inode = page->mapping->host; sector_t iblock, lblock; struct buffer_head *bh, *head, *arr[MAX_BUF_PER_PAGE]; - unsigned int blocksize; + unsigned int blocksize, bbits; int nr, i; int fully_mapped = 1; - BUG_ON(!PageLocked(page)); - blocksize = 1 << inode->i_blkbits; - if (!page_has_buffers(page)) - create_empty_buffers(page, blocksize, 0); - head = page_buffers(page); + head = create_page_buffers(page, inode, 0); + blocksize = head->b_size; + bbits = block_size_bits(blocksize); - iblock = (sector_t)page->index << (PAGE_CACHE_SHIFT - inode->i_blkbits); - lblock = (i_size_read(inode)+blocksize-1) >> inode->i_blkbits; + iblock = (sector_t)page->index << (PAGE_CACHE_SHIFT - bbits); + lblock = (i_size_read(inode)+blocksize-1) >> bbits; bh = head; nr = 0; i = 0; -- cgit v1.2.3-70-g09d2 From bbec0270bdd887f96377065ee38b8848b5afa395 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Thu, 29 Nov 2012 12:31:52 -0800 Subject: blkdev_max_block: make private to fs/buffer.c We really don't want to look at the block size for the raw block device accesses in fs/block-dev.c, because it may be changing from under us. So get rid of the max_block logic entirely, since the caller should already have done it anyway. That leaves the only user of this function in fs/buffer.c, so move the whole function there and make it static. Signed-off-by: Linus Torvalds --- fs/block_dev.c | 55 +----------------------------------------------------- fs/buffer.c | 14 +++++++++++++- include/linux/fs.h | 1 - 3 files changed, 14 insertions(+), 56 deletions(-) (limited to 'fs/buffer.c') diff --git a/fs/block_dev.c b/fs/block_dev.c index 47a949d8a07e..a1e09b4fe1ba 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -70,19 +70,6 @@ static void bdev_inode_switch_bdi(struct inode *inode, spin_unlock(&dst->wb.list_lock); } -sector_t blkdev_max_block(struct block_device *bdev) -{ - sector_t retval = ~((sector_t)0); - loff_t sz = i_size_read(bdev->bd_inode); - - if (sz) { - unsigned int size = block_size(bdev); - unsigned int sizebits = blksize_bits(size); - retval = (sz >> sizebits); - } - return retval; -} - /* Kill _all_ buffers and pagecache , dirty or not.. */ void kill_bdev(struct block_device *bdev) { @@ -163,52 +150,12 @@ static int blkdev_get_block(struct inode *inode, sector_t iblock, struct buffer_head *bh, int create) { - if (iblock >= blkdev_max_block(I_BDEV(inode))) { - if (create) - return -EIO; - - /* - * for reads, we're just trying to fill a partial page. - * return a hole, they will have to call get_block again - * before they can fill it, and they will get -EIO at that - * time - */ - return 0; - } bh->b_bdev = I_BDEV(inode); bh->b_blocknr = iblock; set_buffer_mapped(bh); return 0; } -static int -blkdev_get_blocks(struct inode *inode, sector_t iblock, - struct buffer_head *bh, int create) -{ - sector_t end_block = blkdev_max_block(I_BDEV(inode)); - unsigned long max_blocks = bh->b_size >> inode->i_blkbits; - - if ((iblock + max_blocks) > end_block) { - max_blocks = end_block - iblock; - if ((long)max_blocks <= 0) { - if (create) - return -EIO; /* write fully beyond EOF */ - /* - * It is a read which is fully beyond EOF. We return - * a !buffer_mapped buffer - */ - max_blocks = 0; - } - } - - bh->b_bdev = I_BDEV(inode); - bh->b_blocknr = iblock; - bh->b_size = max_blocks << inode->i_blkbits; - if (max_blocks) - set_buffer_mapped(bh); - return 0; -} - static ssize_t blkdev_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov, loff_t offset, unsigned long nr_segs) @@ -217,7 +164,7 @@ blkdev_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov, struct inode *inode = file->f_mapping->host; return __blockdev_direct_IO(rw, iocb, inode, I_BDEV(inode), iov, offset, - nr_segs, blkdev_get_blocks, NULL, NULL, 0); + nr_segs, blkdev_get_block, NULL, NULL, 0); } int __sync_blockdev(struct block_device *bdev, int wait) diff --git a/fs/buffer.c b/fs/buffer.c index 28a74ff5324b..3586fb05c8ce 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -911,6 +911,18 @@ link_dev_buffers(struct page *page, struct buffer_head *head) attach_page_buffers(page, head); } +static sector_t blkdev_max_block(struct block_device *bdev, unsigned int size) +{ + sector_t retval = ~((sector_t)0); + loff_t sz = i_size_read(bdev->bd_inode); + + if (sz) { + unsigned int sizebits = blksize_bits(size); + retval = (sz >> sizebits); + } + return retval; +} + /* * Initialise the state of a blockdev page's buffers. */ @@ -921,7 +933,7 @@ init_page_buffers(struct page *page, struct block_device *bdev, struct buffer_head *head = page_buffers(page); struct buffer_head *bh = head; int uptodate = PageUptodate(page); - sector_t end_block = blkdev_max_block(I_BDEV(bdev->bd_inode)); + sector_t end_block = blkdev_max_block(I_BDEV(bdev->bd_inode), size); do { if (!buffer_mapped(bh)) { diff --git a/include/linux/fs.h b/include/linux/fs.h index 44f288e9726d..75fe9a134803 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2047,7 +2047,6 @@ extern void unregister_blkdev(unsigned int, const char *); extern struct block_device *bdget(dev_t); extern struct block_device *bdgrab(struct block_device *bdev); extern void bd_set_size(struct block_device *, loff_t size); -extern sector_t blkdev_max_block(struct block_device *bdev); extern void bd_forget(struct inode *inode); extern void bdput(struct block_device *); extern void invalidate_bdev(struct block_device *); -- cgit v1.2.3-70-g09d2 From 57302e0ddf8a210a66fd8a1a2fa50844863b5ded Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Tue, 4 Dec 2012 08:25:11 -0800 Subject: vfs: avoid "attempt to access beyond end of device" warnings The block device access simplification that avoided accessing the (racy) block size information (commit bbec0270bdd8: "blkdev_max_block: make private to fs/buffer.c") no longer checks the maximum block size in the block mapping path. That was _almost_ as simple as just removing the code entirely, because the readers and writers all check the size of the device anyway, so under normal circumstances it "just worked". However, the block size may be such that the end of the device may straddle one single buffer_head. At which point we may still want to access the end of the device, but the buffer we use to access it partially extends past the end. The 'bd_set_size()' function intentionally sets the block size to avoid this, but mounting the device - or setting the block size by hand to some other value - can modify that block size. So instead, teach 'submit_bh()' about the special case of the buffer head straddling the end of the device, and turning such an access into a smaller IO access, avoiding the problem. This, btw, also means that unlike before, we can now access the whole device regardless of device block size setting. So now, even if the device size is only 512-byte aligned, we can read and write even the last sector even when having a much bigger block size for accessing the rest of the device. So with this, we could now get rid of the 'bd_set_size()' block size code entirely - resulting in faster IO for the common case - but that would be a separate patch. Reported-and-tested-by: Romain Francoise Reporeted-and-tested-by: Meelis Roos Reported-by: Tony Luck Signed-off-by: Linus Torvalds --- fs/buffer.c | 52 ++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 52 insertions(+) (limited to 'fs/buffer.c') diff --git a/fs/buffer.c b/fs/buffer.c index 3586fb05c8ce..c4e11390a44c 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -2893,6 +2893,55 @@ static void end_bio_bh_io_sync(struct bio *bio, int err) bio_put(bio); } +/* + * This allows us to do IO even on the odd last sectors + * of a device, even if the bh block size is some multiple + * of the physical sector size. + * + * We'll just truncate the bio to the size of the device, + * and clear the end of the buffer head manually. + * + * Truly out-of-range accesses will turn into actual IO + * errors, this only handles the "we need to be able to + * do IO at the final sector" case. + */ +static void guard_bh_eod(int rw, struct bio *bio, struct buffer_head *bh) +{ + sector_t maxsector; + unsigned bytes; + + maxsector = i_size_read(bio->bi_bdev->bd_inode) >> 9; + if (!maxsector) + return; + + /* + * If the *whole* IO is past the end of the device, + * let it through, and the IO layer will turn it into + * an EIO. + */ + if (unlikely(bio->bi_sector >= maxsector)) + return; + + maxsector -= bio->bi_sector; + bytes = bio->bi_size; + if (likely((bytes >> 9) <= maxsector)) + return; + + /* Uhhuh. We've got a bh that straddles the device size! */ + bytes = maxsector << 9; + + /* Truncate the bio.. */ + bio->bi_size = bytes; + bio->bi_io_vec[0].bv_len = bytes; + + /* ..and clear the end of the buffer for reads */ + if (rw & READ) { + void *kaddr = kmap_atomic(bh->b_page); + memset(kaddr + bh_offset(bh) + bytes, 0, bh->b_size - bytes); + kunmap_atomic(kaddr); + } +} + int submit_bh(int rw, struct buffer_head * bh) { struct bio *bio; @@ -2929,6 +2978,9 @@ int submit_bh(int rw, struct buffer_head * bh) bio->bi_end_io = end_bio_bh_io_sync; bio->bi_private = bh; + /* Take care of bh's that straddle the end of the device */ + guard_bh_eod(rw, bio, bh); + bio_get(bio); submit_bio(rw, bio); -- cgit v1.2.3-70-g09d2 From 27d7c2a006a81c04fab00b8cd81b99af3b32738d Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 5 Dec 2012 20:01:24 +0300 Subject: vfs: clear to the end of the buffer on partial buffer reads READ is zero so the "rw & READ" test is always false. The intended test was "((rw & RW_MASK) == READ)". Signed-off-by: Dan Carpenter Signed-off-by: Linus Torvalds --- fs/buffer.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs/buffer.c') diff --git a/fs/buffer.c b/fs/buffer.c index c4e11390a44c..ec0aca8ba6bf 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -2935,7 +2935,7 @@ static void guard_bh_eod(int rw, struct bio *bio, struct buffer_head *bh) bio->bi_io_vec[0].bv_len = bytes; /* ..and clear the end of the buffer for reads */ - if (rw & READ) { + if ((rw & RW_MASK) == READ) { void *kaddr = kmap_atomic(bh->b_page); memset(kaddr + bh_offset(bh) + bytes, 0, bh->b_size - bytes); kunmap_atomic(kaddr); -- cgit v1.2.3-70-g09d2