diff options
Diffstat (limited to 'fs/gfs2/rgrp.c')
| -rw-r--r-- | fs/gfs2/rgrp.c | 1777 |
1 files changed, 1320 insertions, 457 deletions
diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c index 981bfa32121..f4cb9c0d6bb 100644 --- a/fs/gfs2/rgrp.c +++ b/fs/gfs2/rgrp.c @@ -7,6 +7,8 @@ * of the GNU General Public License version 2. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <linux/slab.h> #include <linux/spinlock.h> #include <linux/completion.h> @@ -16,6 +18,7 @@ #include <linux/prefetch.h> #include <linux/blkdev.h> #include <linux/rbtree.h> +#include <linux/random.h> #include "gfs2.h" #include "incore.h" @@ -56,6 +59,11 @@ * 3 = Used (metadata) */ +struct gfs2_extent { + struct gfs2_rbm rbm; + u32 len; +}; + static const char valid_change[16] = { /* current */ /* n */ 0, 1, 1, 1, @@ -64,53 +72,49 @@ static const char valid_change[16] = { 1, 0, 0, 0 }; -static u32 rgblk_search(struct gfs2_rgrpd *rgd, u32 goal, - unsigned char old_state, - struct gfs2_bitmap **rbi); +static int gfs2_rbm_find(struct gfs2_rbm *rbm, u8 state, u32 *minext, + const struct gfs2_inode *ip, bool nowrap, + const struct gfs2_alloc_parms *ap); + /** * gfs2_setbit - Set a bit in the bitmaps - * @buffer: the buffer that holds the bitmaps - * @buflen: the length (in bytes) of the buffer - * @block: the block to set + * @rbm: The position of the bit to set + * @do_clone: Also set the clone bitmap, if it exists * @new_state: the new state of the block * */ -static inline void gfs2_setbit(struct gfs2_rgrpd *rgd, unsigned char *buf1, - unsigned char *buf2, unsigned int offset, - struct gfs2_bitmap *bi, u32 block, +static inline void gfs2_setbit(const struct gfs2_rbm *rbm, bool do_clone, unsigned char new_state) { unsigned char *byte1, *byte2, *end, cur_state; + struct gfs2_bitmap *bi = rbm_bi(rbm); unsigned int buflen = bi->bi_len; - const unsigned int bit = (block % GFS2_NBBY) * GFS2_BIT_SIZE; + const unsigned int bit = (rbm->offset % GFS2_NBBY) * GFS2_BIT_SIZE; - byte1 = buf1 + offset + (block / GFS2_NBBY); - end = buf1 + offset + buflen; + byte1 = bi->bi_bh->b_data + bi->bi_offset + (rbm->offset / GFS2_NBBY); + end = bi->bi_bh->b_data + bi->bi_offset + buflen; BUG_ON(byte1 >= end); cur_state = (*byte1 >> bit) & GFS2_BIT_MASK; if (unlikely(!valid_change[new_state * 4 + cur_state])) { - printk(KERN_WARNING "GFS2: buf_blk = 0x%llx old_state=%d, " - "new_state=%d\n", - (unsigned long long)block, cur_state, new_state); - printk(KERN_WARNING "GFS2: rgrp=0x%llx bi_start=0x%lx\n", - (unsigned long long)rgd->rd_addr, - (unsigned long)bi->bi_start); - printk(KERN_WARNING "GFS2: bi_offset=0x%lx bi_len=0x%lx\n", - (unsigned long)bi->bi_offset, - (unsigned long)bi->bi_len); + pr_warn("buf_blk = 0x%x old_state=%d, new_state=%d\n", + rbm->offset, cur_state, new_state); + pr_warn("rgrp=0x%llx bi_start=0x%x\n", + (unsigned long long)rbm->rgd->rd_addr, bi->bi_start); + pr_warn("bi_offset=0x%x bi_len=0x%x\n", + bi->bi_offset, bi->bi_len); dump_stack(); - gfs2_consist_rgrpd(rgd); + gfs2_consist_rgrpd(rbm->rgd); return; } *byte1 ^= (cur_state ^ new_state) << bit; - if (buf2) { - byte2 = buf2 + offset + (block / GFS2_NBBY); + if (do_clone && bi->bi_clone) { + byte2 = bi->bi_clone + bi->bi_offset + (rbm->offset / GFS2_NBBY); cur_state = (*byte2 >> bit) & GFS2_BIT_MASK; *byte2 ^= (cur_state ^ new_state) << bit; } @@ -118,29 +122,22 @@ static inline void gfs2_setbit(struct gfs2_rgrpd *rgd, unsigned char *buf1, /** * gfs2_testbit - test a bit in the bitmaps - * @buffer: the buffer that holds the bitmaps - * @buflen: the length (in bytes) of the buffer - * @block: the block to read + * @rbm: The bit to test * + * Returns: The two bit block state of the requested bit */ -static inline unsigned char gfs2_testbit(struct gfs2_rgrpd *rgd, - const unsigned char *buffer, - unsigned int buflen, u32 block) +static inline u8 gfs2_testbit(const struct gfs2_rbm *rbm) { - const unsigned char *byte, *end; - unsigned char cur_state; + struct gfs2_bitmap *bi = rbm_bi(rbm); + const u8 *buffer = bi->bi_bh->b_data + bi->bi_offset; + const u8 *byte; unsigned int bit; - byte = buffer + (block / GFS2_NBBY); - bit = (block % GFS2_NBBY) * GFS2_BIT_SIZE; - end = buffer + buflen; - - gfs2_assert(rgd->rd_sbd, byte < end); + byte = buffer + (rbm->offset / GFS2_NBBY); + bit = (rbm->offset % GFS2_NBBY) * GFS2_BIT_SIZE; - cur_state = (*byte >> bit) & GFS2_BIT_MASK; - - return cur_state; + return (*byte >> bit) & GFS2_BIT_MASK; } /** @@ -177,9 +174,30 @@ static inline u64 gfs2_bit_search(const __le64 *ptr, u64 mask, u8 state) } /** + * rs_cmp - multi-block reservation range compare + * @blk: absolute file system block number of the new reservation + * @len: number of blocks in the new reservation + * @rs: existing reservation to compare against + * + * returns: 1 if the block range is beyond the reach of the reservation + * -1 if the block range is before the start of the reservation + * 0 if the block range overlaps with the reservation + */ +static inline int rs_cmp(u64 blk, u32 len, struct gfs2_blkreserv *rs) +{ + u64 startblk = gfs2_rbm_to_block(&rs->rs_rbm); + + if (blk >= startblk + rs->rs_free) + return 1; + if (blk + len - 1 < startblk) + return -1; + return 0; +} + +/** * gfs2_bitfit - Search an rgrp's bitmap buffer to find a bit-pair representing * a block in a given allocation state. - * @buffer: the buffer that holds the bitmaps + * @buf: the buffer that holds the bitmaps * @len: the length (in bytes) of the buffer * @goal: start search at this block's bit-pair (within @buffer) * @state: GFS2_BLKST_XXX the state of the block we're looking for. @@ -207,8 +225,6 @@ static u32 gfs2_bitfit(const u8 *buf, const unsigned int len, u64 mask = 0x5555555555555555ULL; u32 bit; - BUG_ON(state > 3); - /* Mask off bits we don't care about at the start of the search */ mask <<= spoint; tmp = gfs2_bit_search(ptr, mask, state); @@ -230,7 +246,164 @@ static u32 gfs2_bitfit(const u8 *buf, const unsigned int len, } /** + * gfs2_rbm_from_block - Set the rbm based upon rgd and block number + * @rbm: The rbm with rgd already set correctly + * @block: The block number (filesystem relative) + * + * This sets the bi and offset members of an rbm based on a + * resource group and a filesystem relative block number. The + * resource group must be set in the rbm on entry, the bi and + * offset members will be set by this function. + * + * Returns: 0 on success, or an error code + */ + +static int gfs2_rbm_from_block(struct gfs2_rbm *rbm, u64 block) +{ + u64 rblock = block - rbm->rgd->rd_data0; + + if (WARN_ON_ONCE(rblock > UINT_MAX)) + return -EINVAL; + if (block >= rbm->rgd->rd_data0 + rbm->rgd->rd_data) + return -E2BIG; + + rbm->bii = 0; + rbm->offset = (u32)(rblock); + /* Check if the block is within the first block */ + if (rbm->offset < rbm_bi(rbm)->bi_blocks) + return 0; + + /* Adjust for the size diff between gfs2_meta_header and gfs2_rgrp */ + rbm->offset += (sizeof(struct gfs2_rgrp) - + sizeof(struct gfs2_meta_header)) * GFS2_NBBY; + rbm->bii = rbm->offset / rbm->rgd->rd_sbd->sd_blocks_per_bitmap; + rbm->offset -= rbm->bii * rbm->rgd->rd_sbd->sd_blocks_per_bitmap; + return 0; +} + +/** + * gfs2_rbm_incr - increment an rbm structure + * @rbm: The rbm with rgd already set correctly + * + * This function takes an existing rbm structure and increments it to the next + * viable block offset. + * + * Returns: If incrementing the offset would cause the rbm to go past the + * end of the rgrp, true is returned, otherwise false. + * + */ + +static bool gfs2_rbm_incr(struct gfs2_rbm *rbm) +{ + if (rbm->offset + 1 < rbm_bi(rbm)->bi_blocks) { /* in the same bitmap */ + rbm->offset++; + return false; + } + if (rbm->bii == rbm->rgd->rd_length - 1) /* at the last bitmap */ + return true; + + rbm->offset = 0; + rbm->bii++; + return false; +} + +/** + * gfs2_unaligned_extlen - Look for free blocks which are not byte aligned + * @rbm: Position to search (value/result) + * @n_unaligned: Number of unaligned blocks to check + * @len: Decremented for each block found (terminate on zero) + * + * Returns: true if a non-free block is encountered + */ + +static bool gfs2_unaligned_extlen(struct gfs2_rbm *rbm, u32 n_unaligned, u32 *len) +{ + u32 n; + u8 res; + + for (n = 0; n < n_unaligned; n++) { + res = gfs2_testbit(rbm); + if (res != GFS2_BLKST_FREE) + return true; + (*len)--; + if (*len == 0) + return true; + if (gfs2_rbm_incr(rbm)) + return true; + } + + return false; +} + +/** + * gfs2_free_extlen - Return extent length of free blocks + * @rrbm: Starting position + * @len: Max length to check + * + * Starting at the block specified by the rbm, see how many free blocks + * there are, not reading more than len blocks ahead. This can be done + * using memchr_inv when the blocks are byte aligned, but has to be done + * on a block by block basis in case of unaligned blocks. Also this + * function can cope with bitmap boundaries (although it must stop on + * a resource group boundary) + * + * Returns: Number of free blocks in the extent + */ + +static u32 gfs2_free_extlen(const struct gfs2_rbm *rrbm, u32 len) +{ + struct gfs2_rbm rbm = *rrbm; + u32 n_unaligned = rbm.offset & 3; + u32 size = len; + u32 bytes; + u32 chunk_size; + u8 *ptr, *start, *end; + u64 block; + struct gfs2_bitmap *bi; + + if (n_unaligned && + gfs2_unaligned_extlen(&rbm, 4 - n_unaligned, &len)) + goto out; + + n_unaligned = len & 3; + /* Start is now byte aligned */ + while (len > 3) { + bi = rbm_bi(&rbm); + start = bi->bi_bh->b_data; + if (bi->bi_clone) + start = bi->bi_clone; + end = start + bi->bi_bh->b_size; + start += bi->bi_offset; + BUG_ON(rbm.offset & 3); + start += (rbm.offset / GFS2_NBBY); + bytes = min_t(u32, len / GFS2_NBBY, (end - start)); + ptr = memchr_inv(start, 0, bytes); + chunk_size = ((ptr == NULL) ? bytes : (ptr - start)); + chunk_size *= GFS2_NBBY; + BUG_ON(len < chunk_size); + len -= chunk_size; + block = gfs2_rbm_to_block(&rbm); + if (gfs2_rbm_from_block(&rbm, block + chunk_size)) { + n_unaligned = 0; + break; + } + if (ptr) { + n_unaligned = 3; + break; + } + n_unaligned = len & 3; + } + + /* Deal with any bits left over at the end */ + if (n_unaligned) + gfs2_unaligned_extlen(&rbm, n_unaligned, &len); +out: + return size - len; +} + +/** * gfs2_bitcount - count the number of bits in a certain state + * @rgd: the resource group descriptor * @buffer: the buffer that holds the bitmaps * @buflen: the length (in bytes) of the buffer * @state: the state of the block we're looking for @@ -264,7 +437,6 @@ static u32 gfs2_bitcount(struct gfs2_rgrpd *rgd, const u8 *buffer, /** * gfs2_rgrp_verify - Verify that a resource group is consistent - * @sdp: the filesystem * @rgd: the rgrp * */ @@ -322,28 +494,37 @@ static inline int rgrp_contains_block(struct gfs2_rgrpd *rgd, u64 block) /** * gfs2_blk2rgrpd - Find resource group for a given data/meta block number * @sdp: The GFS2 superblock - * @n: The data block number + * @blk: The data block number + * @exact: True if this needs to be an exact match * * Returns: The resource group, or NULL if not found */ -struct gfs2_rgrpd *gfs2_blk2rgrpd(struct gfs2_sbd *sdp, u64 blk) +struct gfs2_rgrpd *gfs2_blk2rgrpd(struct gfs2_sbd *sdp, u64 blk, bool exact) { - struct rb_node **newn; + struct rb_node *n, *next; struct gfs2_rgrpd *cur; spin_lock(&sdp->sd_rindex_spin); - newn = &sdp->sd_rindex_tree.rb_node; - while (*newn) { - cur = rb_entry(*newn, struct gfs2_rgrpd, rd_node); + n = sdp->sd_rindex_tree.rb_node; + while (n) { + cur = rb_entry(n, struct gfs2_rgrpd, rd_node); + next = NULL; if (blk < cur->rd_addr) - newn = &((*newn)->rb_left); + next = n->rb_left; else if (blk >= cur->rd_data0 + cur->rd_data) - newn = &((*newn)->rb_right); - else { + next = n->rb_right; + if (next == NULL) { spin_unlock(&sdp->sd_rindex_spin); + if (exact) { + if (blk < cur->rd_addr) + return NULL; + if (blk >= cur->rd_data0 + cur->rd_data) + return NULL; + } return cur; } + n = next; } spin_unlock(&sdp->sd_rindex_spin); @@ -372,7 +553,7 @@ struct gfs2_rgrpd *gfs2_rgrpd_get_first(struct gfs2_sbd *sdp) /** * gfs2_rgrpd_get_next - get the next RG - * @rgd: A RG + * @rgd: the resource group descriptor * * Returns: The next rgrp */ @@ -407,6 +588,127 @@ void gfs2_free_clones(struct gfs2_rgrpd *rgd) } } +/** + * gfs2_rs_alloc - make sure we have a reservation assigned to the inode + * @ip: the inode for this reservation + */ +int gfs2_rs_alloc(struct gfs2_inode *ip) +{ + int error = 0; + + down_write(&ip->i_rw_mutex); + if (ip->i_res) + goto out; + + ip->i_res = kmem_cache_zalloc(gfs2_rsrv_cachep, GFP_NOFS); + if (!ip->i_res) { + error = -ENOMEM; + goto out; + } + + RB_CLEAR_NODE(&ip->i_res->rs_node); +out: + up_write(&ip->i_rw_mutex); + return error; +} + +static void dump_rs(struct seq_file *seq, const struct gfs2_blkreserv *rs) +{ + gfs2_print_dbg(seq, " B: n:%llu s:%llu b:%u f:%u\n", + (unsigned long long)rs->rs_inum, + (unsigned long long)gfs2_rbm_to_block(&rs->rs_rbm), + rs->rs_rbm.offset, rs->rs_free); +} + +/** + * __rs_deltree - remove a multi-block reservation from the rgd tree + * @rs: The reservation to remove + * + */ +static void __rs_deltree(struct gfs2_blkreserv *rs) +{ + struct gfs2_rgrpd *rgd; + + if (!gfs2_rs_active(rs)) + return; + + rgd = rs->rs_rbm.rgd; + trace_gfs2_rs(rs, TRACE_RS_TREEDEL); + rb_erase(&rs->rs_node, &rgd->rd_rstree); + RB_CLEAR_NODE(&rs->rs_node); + + if (rs->rs_free) { + struct gfs2_bitmap *bi = rbm_bi(&rs->rs_rbm); + + /* return reserved blocks to the rgrp */ + BUG_ON(rs->rs_rbm.rgd->rd_reserved < rs->rs_free); + rs->rs_rbm.rgd->rd_reserved -= rs->rs_free; + /* The rgrp extent failure point is likely not to increase; + it will only do so if the freed blocks are somehow + contiguous with a span of free blocks that follows. Still, + it will force the number to be recalculated later. */ + rgd->rd_extfail_pt += rs->rs_free; + rs->rs_free = 0; + clear_bit(GBF_FULL, &bi->bi_flags); + } +} + +/** + * gfs2_rs_deltree - remove a multi-block reservation from the rgd tree + * @rs: The reservation to remove + * + */ +void gfs2_rs_deltree(struct gfs2_blkreserv *rs) +{ + struct gfs2_rgrpd *rgd; + + rgd = rs->rs_rbm.rgd; + if (rgd) { + spin_lock(&rgd->rd_rsspin); + __rs_deltree(rs); + spin_unlock(&rgd->rd_rsspin); + } +} + +/** + * gfs2_rs_delete - delete a multi-block reservation + * @ip: The inode for this reservation + * @wcount: The inode's write count, or NULL + * + */ +void gfs2_rs_delete(struct gfs2_inode *ip, atomic_t *wcount) +{ + down_write(&ip->i_rw_mutex); + if (ip->i_res && ((wcount == NULL) || (atomic_read(wcount) <= 1))) { + gfs2_rs_deltree(ip->i_res); + BUG_ON(ip->i_res->rs_free); + kmem_cache_free(gfs2_rsrv_cachep, ip->i_res); + ip->i_res = NULL; + } + up_write(&ip->i_rw_mutex); +} + +/** + * return_all_reservations - return all reserved blocks back to the rgrp. + * @rgd: the rgrp that needs its space back + * + * We previously reserved a bunch of blocks for allocation. Now we need to + * give them back. This leave the reservation structures in tact, but removes + * all of their corresponding "no-fly zones". + */ +static void return_all_reservations(struct gfs2_rgrpd *rgd) +{ + struct rb_node *n; + struct gfs2_blkreserv *rs; + + spin_lock(&rgd->rd_rsspin); + while ((n = rb_first(&rgd->rd_rstree))) { + rs = rb_entry(n, struct gfs2_blkreserv, rs_node); + __rs_deltree(rs); + } + spin_unlock(&rgd->rd_rsspin); +} + void gfs2_clear_rgrpd(struct gfs2_sbd *sdp) { struct rb_node *n; @@ -429,17 +731,18 @@ void gfs2_clear_rgrpd(struct gfs2_sbd *sdp) gfs2_free_clones(rgd); kfree(rgd->rd_bits); + return_all_reservations(rgd); kmem_cache_free(gfs2_rgrpd_cachep, rgd); } } static void gfs2_rindex_print(const struct gfs2_rgrpd *rgd) { - printk(KERN_INFO " ri_addr = %llu\n", (unsigned long long)rgd->rd_addr); - printk(KERN_INFO " ri_length = %u\n", rgd->rd_length); - printk(KERN_INFO " ri_data0 = %llu\n", (unsigned long long)rgd->rd_data0); - printk(KERN_INFO " ri_data = %u\n", rgd->rd_data); - printk(KERN_INFO " ri_bitbytes = %u\n", rgd->rd_bitbytes); + pr_info("ri_addr = %llu\n", (unsigned long long)rgd->rd_addr); + pr_info("ri_length = %u\n", rgd->rd_length); + pr_info("ri_data0 = %llu\n", (unsigned long long)rgd->rd_data0); + pr_info("ri_data = %u\n", rgd->rd_data); + pr_info("ri_bitbytes = %u\n", rgd->rd_bitbytes); } /** @@ -478,18 +781,21 @@ static int compute_bitstructs(struct gfs2_rgrpd *rgd) bi->bi_offset = sizeof(struct gfs2_rgrp); bi->bi_start = 0; bi->bi_len = bytes; + bi->bi_blocks = bytes * GFS2_NBBY; /* header block */ } else if (x == 0) { bytes = sdp->sd_sb.sb_bsize - sizeof(struct gfs2_rgrp); bi->bi_offset = sizeof(struct gfs2_rgrp); bi->bi_start = 0; bi->bi_len = bytes; + bi->bi_blocks = bytes * GFS2_NBBY; /* last block */ } else if (x + 1 == length) { bytes = bytes_left; bi->bi_offset = sizeof(struct gfs2_meta_header); bi->bi_start = rgd->rd_bitbytes - bytes_left; bi->bi_len = bytes; + bi->bi_blocks = bytes * GFS2_NBBY; /* other blocks */ } else { bytes = sdp->sd_sb.sb_bsize - @@ -497,6 +803,7 @@ static int compute_bitstructs(struct gfs2_rgrpd *rgd) bi->bi_offset = sizeof(struct gfs2_meta_header); bi->bi_start = rgd->rd_bitbytes - bytes_left; bi->bi_len = bytes; + bi->bi_blocks = bytes * GFS2_NBBY; } bytes_left -= bytes; @@ -521,6 +828,7 @@ static int compute_bitstructs(struct gfs2_rgrpd *rgd) /** * gfs2_ri_total - Total up the file system space, according to the rindex. + * @sdp: the filesystem * */ u64 gfs2_ri_total(struct gfs2_sbd *sdp) @@ -529,27 +837,23 @@ u64 gfs2_ri_total(struct gfs2_sbd *sdp) struct inode *inode = sdp->sd_rindex; struct gfs2_inode *ip = GFS2_I(inode); char buf[sizeof(struct gfs2_rindex)]; - struct file_ra_state ra_state; int error, rgrps; - mutex_lock(&sdp->sd_rindex_mutex); - file_ra_state_init(&ra_state, inode->i_mapping); for (rgrps = 0;; rgrps++) { loff_t pos = rgrps * sizeof(struct gfs2_rindex); if (pos + sizeof(struct gfs2_rindex) > i_size_read(inode)) break; - error = gfs2_internal_read(ip, &ra_state, buf, &pos, + error = gfs2_internal_read(ip, buf, &pos, sizeof(struct gfs2_rindex)); if (error != sizeof(struct gfs2_rindex)) break; total_data += be32_to_cpu(((struct gfs2_rindex *)buf)->ri_data); } - mutex_unlock(&sdp->sd_rindex_mutex); return total_data; } -static void rgd_insert(struct gfs2_rgrpd *rgd) +static int rgd_insert(struct gfs2_rgrpd *rgd) { struct gfs2_sbd *sdp = rgd->rd_sbd; struct rb_node **newn = &sdp->sd_rindex_tree.rb_node, *parent = NULL; @@ -565,24 +869,26 @@ static void rgd_insert(struct gfs2_rgrpd *rgd) else if (rgd->rd_addr > cur->rd_addr) newn = &((*newn)->rb_right); else - return; + return -EEXIST; } rb_link_node(&rgd->rd_node, parent, newn); rb_insert_color(&rgd->rd_node, &sdp->sd_rindex_tree); + sdp->sd_rgrps++; + return 0; } /** * read_rindex_entry - Pull in a new resource index entry from the disk - * @gl: The glock covering the rindex inode + * @ip: Pointer to the rindex inode * * Returns: 0 on success, > 0 on EOF, error code otherwise */ -static int read_rindex_entry(struct gfs2_inode *ip, - struct file_ra_state *ra_state) +static int read_rindex_entry(struct gfs2_inode *ip) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); + const unsigned bsize = sdp->sd_sb.sb_bsize; loff_t pos = sdp->sd_rgrps * sizeof(struct gfs2_rindex); struct gfs2_rindex buf; int error; @@ -591,7 +897,7 @@ static int read_rindex_entry(struct gfs2_inode *ip, if (pos >= i_size_read(&ip->i_inode)) return 1; - error = gfs2_internal_read(ip, ra_state, (char *)&buf, &pos, + error = gfs2_internal_read(ip, (char *)&buf, &pos, sizeof(struct gfs2_rindex)); if (error != sizeof(struct gfs2_rindex)) @@ -608,6 +914,7 @@ static int read_rindex_entry(struct gfs2_inode *ip, rgd->rd_data0 = be64_to_cpu(buf.ri_data0); rgd->rd_data = be32_to_cpu(buf.ri_data); rgd->rd_bitbytes = be32_to_cpu(buf.ri_bitbytes); + spin_lock_init(&rgd->rd_rsspin); error = compute_bitstructs(rgd); if (error) @@ -619,14 +926,20 @@ static int read_rindex_entry(struct gfs2_inode *ip, goto fail; rgd->rd_gl->gl_object = rgd; + rgd->rd_gl->gl_vm.start = rgd->rd_addr * bsize; + rgd->rd_gl->gl_vm.end = rgd->rd_gl->gl_vm.start + (rgd->rd_length * bsize) - 1; + rgd->rd_rgl = (struct gfs2_rgrp_lvb *)rgd->rd_gl->gl_lksb.sb_lvbptr; rgd->rd_flags &= ~GFS2_RDF_UPTODATE; if (rgd->rd_data > sdp->sd_max_rg_data) sdp->sd_max_rg_data = rgd->rd_data; spin_lock(&sdp->sd_rindex_spin); - rgd_insert(rgd); - sdp->sd_rgrps++; + error = rgd_insert(rgd); spin_unlock(&sdp->sd_rindex_spin); - return error; + if (!error) + return 0; + + error = 0; /* someone else read in the rgrp; free it and ignore it */ + gfs2_glock_put(rgd->rd_gl); fail: kfree(rgd->rd_bits); @@ -644,13 +957,10 @@ fail: static int gfs2_ri_update(struct gfs2_inode *ip) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); - struct inode *inode = &ip->i_inode; - struct file_ra_state ra_state; int error; - file_ra_state_init(&ra_state, inode->i_mapping); do { - error = read_rindex_entry(ip, &ra_state); + error = read_rindex_entry(ip); } while (error == 0); if (error < 0) @@ -683,20 +993,22 @@ int gfs2_rindex_update(struct gfs2_sbd *sdp) struct gfs2_glock *gl = ip->i_gl; struct gfs2_holder ri_gh; int error = 0; + int unlock_required = 0; /* Read new copy from disk if we don't have the latest */ if (!sdp->sd_rindex_uptodate) { - mutex_lock(&sdp->sd_rindex_mutex); - error = gfs2_glock_nq_init(gl, LM_ST_SHARED, 0, &ri_gh); - if (error) - return error; + if (!gfs2_glock_is_locked_by_me(gl)) { + error = gfs2_glock_nq_init(gl, LM_ST_SHARED, 0, &ri_gh); + if (error) + return error; + unlock_required = 1; + } if (!sdp->sd_rindex_uptodate) error = gfs2_ri_update(ip); - gfs2_glock_dq_uninit(&ri_gh); - mutex_unlock(&sdp->sd_rindex_mutex); + if (unlock_required) + gfs2_glock_dq_uninit(&ri_gh); } - return error; } @@ -726,8 +1038,64 @@ static void gfs2_rgrp_out(struct gfs2_rgrpd *rgd, void *buf) memset(&str->rg_reserved, 0, sizeof(str->rg_reserved)); } +static int gfs2_rgrp_lvb_valid(struct gfs2_rgrpd *rgd) +{ + struct gfs2_rgrp_lvb *rgl = rgd->rd_rgl; + struct gfs2_rgrp *str = (struct gfs2_rgrp *)rgd->rd_bits[0].bi_bh->b_data; + + if (rgl->rl_flags != str->rg_flags || rgl->rl_free != str->rg_free || + rgl->rl_dinodes != str->rg_dinodes || + rgl->rl_igeneration != str->rg_igeneration) + return 0; + return 1; +} + +static void gfs2_rgrp_ondisk2lvb(struct gfs2_rgrp_lvb *rgl, const void *buf) +{ + const struct gfs2_rgrp *str = buf; + + rgl->rl_magic = cpu_to_be32(GFS2_MAGIC); + rgl->rl_flags = str->rg_flags; + rgl->rl_free = str->rg_free; + rgl->rl_dinodes = str->rg_dinodes; + rgl->rl_igeneration = str->rg_igeneration; + rgl->__pad = 0UL; +} + +static void update_rgrp_lvb_unlinked(struct gfs2_rgrpd *rgd, u32 change) +{ + struct gfs2_rgrp_lvb *rgl = rgd->rd_rgl; + u32 unlinked = be32_to_cpu(rgl->rl_unlinked) + change; + rgl->rl_unlinked = cpu_to_be32(unlinked); +} + +static u32 count_unlinked(struct gfs2_rgrpd *rgd) +{ + struct gfs2_bitmap *bi; + const u32 length = rgd->rd_length; + const u8 *buffer = NULL; + u32 i, goal, count = 0; + + for (i = 0, bi = rgd->rd_bits; i < length; i++, bi++) { + goal = 0; + buffer = bi->bi_bh->b_data + bi->bi_offset; + WARN_ON(!buffer_uptodate(bi->bi_bh)); + while (goal < bi->bi_len * GFS2_NBBY) { + goal = gfs2_bitfit(buffer, bi->bi_len, goal, + GFS2_BLKST_UNLINKED); + if (goal == BFITNOENT) + break; + count++; + goal++; + } + } + + return count; +} + + /** - * gfs2_rgrp_go_lock - Read in a RG's header and bitmaps + * gfs2_rgrp_bh_get - Read in a RG's header and bitmaps * @rgd: the struct gfs2_rgrpd describing the RG to read in * * Read in all of a Resource Group's header and bitmap blocks. @@ -736,9 +1104,8 @@ static void gfs2_rgrp_out(struct gfs2_rgrpd *rgd, void *buf) * Returns: errno */ -int gfs2_rgrp_go_lock(struct gfs2_holder *gh) +static int gfs2_rgrp_bh_get(struct gfs2_rgrpd *rgd) { - struct gfs2_rgrpd *rgd = gh->gh_gl->gl_object; struct gfs2_sbd *sdp = rgd->rd_sbd; struct gfs2_glock *gl = rgd->rd_gl; unsigned int length = rgd->rd_length; @@ -746,6 +1113,9 @@ int gfs2_rgrp_go_lock(struct gfs2_holder *gh) unsigned int x, y; int error; + if (rgd->rd_bits[0].bi_bh != NULL) + return 0; + for (x = 0; x < length; x++) { bi = rgd->rd_bits + x; error = gfs2_meta_read(gl, rgd->rd_addr + x, 0, &bi->bi_bh); @@ -771,8 +1141,23 @@ int gfs2_rgrp_go_lock(struct gfs2_holder *gh) gfs2_rgrp_in(rgd, (rgd->rd_bits[0].bi_bh)->b_data); rgd->rd_flags |= (GFS2_RDF_UPTODATE | GFS2_RDF_CHECK); rgd->rd_free_clone = rgd->rd_free; + /* max out the rgrp allocation failure point */ + rgd->rd_extfail_pt = rgd->rd_free; + } + if (cpu_to_be32(GFS2_MAGIC) != rgd->rd_rgl->rl_magic) { + rgd->rd_rgl->rl_unlinked = cpu_to_be32(count_unlinked(rgd)); + gfs2_rgrp_ondisk2lvb(rgd->rd_rgl, + rgd->rd_bits[0].bi_bh->b_data); + } + else if (sdp->sd_args.ar_rgrplvb) { + if (!gfs2_rgrp_lvb_valid(rgd)){ + gfs2_consist_rgrpd(rgd); + error = -EIO; + goto fail; + } + if (rgd->rd_rgl->rl_unlinked == 0) + rgd->rd_flags &= ~GFS2_RDF_CHECK; } - return 0; fail: @@ -786,9 +1171,42 @@ fail: return error; } +static int update_rgrp_lvb(struct gfs2_rgrpd *rgd) +{ + u32 rl_flags; + + if (rgd->rd_flags & GFS2_RDF_UPTODATE) + return 0; + + if (cpu_to_be32(GFS2_MAGIC) != rgd->rd_rgl->rl_magic) + return gfs2_rgrp_bh_get(rgd); + + rl_flags = be32_to_cpu(rgd->rd_rgl->rl_flags); + rl_flags &= ~GFS2_RDF_MASK; + rgd->rd_flags &= GFS2_RDF_MASK; + rgd->rd_flags |= (rl_flags | GFS2_RDF_UPTODATE | GFS2_RDF_CHECK); + if (rgd->rd_rgl->rl_unlinked == 0) + rgd->rd_flags &= ~GFS2_RDF_CHECK; + rgd->rd_free = be32_to_cpu(rgd->rd_rgl->rl_free); + rgd->rd_free_clone = rgd->rd_free; + rgd->rd_dinodes = be32_to_cpu(rgd->rd_rgl->rl_dinodes); + rgd->rd_igeneration = be64_to_cpu(rgd->rd_rgl->rl_igeneration); + return 0; +} + +int gfs2_rgrp_go_lock(struct gfs2_holder *gh) +{ + struct gfs2_rgrpd *rgd = gh->gh_gl->gl_object; + struct gfs2_sbd *sdp = rgd->rd_sbd; + + if (gh->gh_flags & GL_SKIP && sdp->sd_args.ar_rgrplvb) + return 0; + return gfs2_rgrp_bh_get(rgd); +} + /** * gfs2_rgrp_go_unlock - Release RG bitmaps read in with gfs2_rgrp_bh_get() - * @rgd: the struct gfs2_rgrpd describing the RG to read in + * @gh: The glock holder for the resource group * */ @@ -799,128 +1217,506 @@ void gfs2_rgrp_go_unlock(struct gfs2_holder *gh) for (x = 0; x < length; x++) { struct gfs2_bitmap *bi = rgd->rd_bits + x; - brelse(bi->bi_bh); - bi->bi_bh = NULL; + if (bi->bi_bh) { + brelse(bi->bi_bh); + bi->bi_bh = NULL; + } } } -void gfs2_rgrp_send_discards(struct gfs2_sbd *sdp, u64 offset, +int gfs2_rgrp_send_discards(struct gfs2_sbd *sdp, u64 offset, struct buffer_head *bh, - const struct gfs2_bitmap *bi) + const struct gfs2_bitmap *bi, unsigned minlen, u64 *ptrimmed) { struct super_block *sb = sdp->sd_vfs; - struct block_device *bdev = sb->s_bdev; - const unsigned int sects_per_blk = sdp->sd_sb.sb_bsize / - bdev_logical_block_size(sb->s_bdev); u64 blk; sector_t start = 0; - sector_t nr_sects = 0; + sector_t nr_blks = 0; int rv; unsigned int x; + u32 trimmed = 0; + u8 diff; for (x = 0; x < bi->bi_len; x++) { - const u8 *orig = bh->b_data + bi->bi_offset + x; - const u8 *clone = bi->bi_clone + bi->bi_offset + x; - u8 diff = ~(*orig | (*orig >> 1)) & (*clone | (*clone >> 1)); + const u8 *clone = bi->bi_clone ? bi->bi_clone : bi->bi_bh->b_data; + clone += bi->bi_offset; + clone += x; + if (bh) { + const u8 *orig = bh->b_data + bi->bi_offset + x; + diff = ~(*orig | (*orig >> 1)) & (*clone | (*clone >> 1)); + } else { + diff = ~(*clone | (*clone >> 1)); + } diff &= 0x55; if (diff == 0) continue; blk = offset + ((bi->bi_start + x) * GFS2_NBBY); - blk *= sects_per_blk; /* convert to sectors */ while(diff) { if (diff & 1) { - if (nr_sects == 0) + if (nr_blks == 0) goto start_new_extent; - if ((start + nr_sects) != blk) { - rv = blkdev_issue_discard(bdev, start, - nr_sects, GFP_NOFS, - 0); - if (rv) - goto fail; - nr_sects = 0; + if ((start + nr_blks) != blk) { + if (nr_blks >= minlen) { + rv = sb_issue_discard(sb, + start, nr_blks, + GFP_NOFS, 0); + if (rv) + goto fail; + trimmed += nr_blks; + } + nr_blks = 0; start_new_extent: start = blk; } - nr_sects += sects_per_blk; + nr_blks++; } diff >>= 2; - blk += sects_per_blk; + blk++; } } - if (nr_sects) { - rv = blkdev_issue_discard(bdev, start, nr_sects, GFP_NOFS, 0); + if (nr_blks >= minlen) { + rv = sb_issue_discard(sb, start, nr_blks, GFP_NOFS, 0); if (rv) goto fail; + trimmed += nr_blks; } - return; + if (ptrimmed) + *ptrimmed = trimmed; + return 0; + fail: - fs_warn(sdp, "error %d on discard request, turning discards off for this filesystem", rv); + if (sdp->sd_args.ar_discard) + fs_warn(sdp, "error %d on discard request, turning discards off for this filesystem", rv); sdp->sd_args.ar_discard = 0; + return -EIO; } /** - * gfs2_qadata_get - get the struct gfs2_qadata structure for an inode - * @ip: the incore GFS2 inode structure + * gfs2_fitrim - Generate discard requests for unused bits of the filesystem + * @filp: Any file on the filesystem + * @argp: Pointer to the arguments (also used to pass result) * - * Returns: the struct gfs2_qadata + * Returns: 0 on success, otherwise error code */ -struct gfs2_qadata *gfs2_qadata_get(struct gfs2_inode *ip) +int gfs2_fitrim(struct file *filp, void __user *argp) { - struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); - int error; - BUG_ON(ip->i_qadata != NULL); - ip->i_qadata = kzalloc(sizeof(struct gfs2_qadata), GFP_NOFS); - error = gfs2_rindex_update(sdp); - if (error) - fs_warn(sdp, "rindex update returns %d\n", error); - return ip->i_qadata; + struct inode *inode = file_inode(filp); + struct gfs2_sbd *sdp = GFS2_SB(inode); + struct request_queue *q = bdev_get_queue(sdp->sd_vfs->s_bdev); + struct buffer_head *bh; + struct gfs2_rgrpd *rgd; + struct gfs2_rgrpd *rgd_end; + struct gfs2_holder gh; + struct fstrim_range r; + int ret = 0; + u64 amt; + u64 trimmed = 0; + u64 start, end, minlen; + unsigned int x; + unsigned bs_shift = sdp->sd_sb.sb_bsize_shift; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + if (!blk_queue_discard(q)) + return -EOPNOTSUPP; + + if (copy_from_user(&r, argp, sizeof(r))) + return -EFAULT; + + ret = gfs2_rindex_update(sdp); + if (ret) + return ret; + + start = r.start >> bs_shift; + end = start + (r.len >> bs_shift); + minlen = max_t(u64, r.minlen, + q->limits.discard_granularity) >> bs_shift; + + if (end <= start || minlen > sdp->sd_max_rg_data) + return -EINVAL; + + rgd = gfs2_blk2rgrpd(sdp, start, 0); + rgd_end = gfs2_blk2rgrpd(sdp, end, 0); + + if ((gfs2_rgrpd_get_first(sdp) == gfs2_rgrpd_get_next(rgd_end)) + && (start > rgd_end->rd_data0 + rgd_end->rd_data)) + return -EINVAL; /* start is beyond the end of the fs */ + + while (1) { + + ret = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, &gh); + if (ret) + goto out; + + if (!(rgd->rd_flags & GFS2_RGF_TRIMMED)) { + /* Trim each bitmap in the rgrp */ + for (x = 0; x < rgd->rd_length; x++) { + struct gfs2_bitmap *bi = rgd->rd_bits + x; + ret = gfs2_rgrp_send_discards(sdp, + rgd->rd_data0, NULL, bi, minlen, + &amt); + if (ret) { + gfs2_glock_dq_uninit(&gh); + goto out; + } + trimmed += amt; + } + + /* Mark rgrp as having been trimmed */ + ret = gfs2_trans_begin(sdp, RES_RG_HDR, 0); + if (ret == 0) { + bh = rgd->rd_bits[0].bi_bh; + rgd->rd_flags |= GFS2_RGF_TRIMMED; + gfs2_trans_add_meta(rgd->rd_gl, bh); + gfs2_rgrp_out(rgd, bh->b_data); + gfs2_rgrp_ondisk2lvb(rgd->rd_rgl, bh->b_data); + gfs2_trans_end(sdp); + } + } + gfs2_glock_dq_uninit(&gh); + + if (rgd == rgd_end) + break; + + rgd = gfs2_rgrpd_get_next(rgd); + } + +out: + r.len = trimmed << bs_shift; + if (copy_to_user(argp, &r, sizeof(r))) + return -EFAULT; + + return ret; } /** - * gfs2_blkrsv_get - get the struct gfs2_blkreserv structure for an inode - * @ip: the incore GFS2 inode structure + * rs_insert - insert a new multi-block reservation into the rgrp's rb_tree + * @ip: the inode structure * - * Returns: the struct gfs2_qadata */ +static void rs_insert(struct gfs2_inode *ip) +{ + struct rb_node **newn, *parent = NULL; + int rc; + struct gfs2_blkreserv *rs = ip->i_res; + struct gfs2_rgrpd *rgd = rs->rs_rbm.rgd; + u64 fsblock = gfs2_rbm_to_block(&rs->rs_rbm); + + BUG_ON(gfs2_rs_active(rs)); + + spin_lock(&rgd->rd_rsspin); + newn = &rgd->rd_rstree.rb_node; + while (*newn) { + struct gfs2_blkreserv *cur = + rb_entry(*newn, struct gfs2_blkreserv, rs_node); -static struct gfs2_blkreserv *gfs2_blkrsv_get(struct gfs2_inode *ip) + parent = *newn; + rc = rs_cmp(fsblock, rs->rs_free, cur); + if (rc > 0) + newn = &((*newn)->rb_right); + else if (rc < 0) + newn = &((*newn)->rb_left); + else { + spin_unlock(&rgd->rd_rsspin); + WARN_ON(1); + return; + } + } + + rb_link_node(&rs->rs_node, parent, newn); + rb_insert_color(&rs->rs_node, &rgd->rd_rstree); + + /* Do our rgrp accounting for the reservation */ + rgd->rd_reserved += rs->rs_free; /* blocks reserved */ + spin_unlock(&rgd->rd_rsspin); + trace_gfs2_rs(rs, TRACE_RS_INSERT); +} + +/** + * rg_mblk_search - find a group of multiple free blocks to form a reservation + * @rgd: the resource group descriptor + * @ip: pointer to the inode for which we're reserving blocks + * @ap: the allocation parameters + * + */ + +static void rg_mblk_search(struct gfs2_rgrpd *rgd, struct gfs2_inode *ip, + const struct gfs2_alloc_parms *ap) { - BUG_ON(ip->i_res != NULL); - ip->i_res = kzalloc(sizeof(struct gfs2_blkreserv), GFP_NOFS); - return ip->i_res; + struct gfs2_rbm rbm = { .rgd = rgd, }; + u64 goal; + struct gfs2_blkreserv *rs = ip->i_res; + u32 extlen; + u32 free_blocks = rgd->rd_free_clone - rgd->rd_reserved; + int ret; + struct inode *inode = &ip->i_inode; + + if (S_ISDIR(inode->i_mode)) + extlen = 1; + else { + extlen = max_t(u32, atomic_read(&rs->rs_sizehint), ap->target); + extlen = clamp(extlen, RGRP_RSRV_MINBLKS, free_blocks); + } + if ((rgd->rd_free_clone < rgd->rd_reserved) || (free_blocks < extlen)) + return; + + /* Find bitmap block that contains bits for goal block */ + if (rgrp_contains_block(rgd, ip->i_goal)) + goal = ip->i_goal; + else + goal = rgd->rd_last_alloc + rgd->rd_data0; + + if (WARN_ON(gfs2_rbm_from_block(&rbm, goal))) + return; + + ret = gfs2_rbm_find(&rbm, GFS2_BLKST_FREE, &extlen, ip, true, ap); + if (ret == 0) { + rs->rs_rbm = rbm; + rs->rs_free = extlen; + rs->rs_inum = ip->i_no_addr; + rs_insert(ip); + } else { + if (goal == rgd->rd_last_alloc + rgd->rd_data0) + rgd->rd_last_alloc = 0; + } } /** - * try_rgrp_fit - See if a given reservation will fit in a given RG - * @rgd: the RG data - * @ip: the inode + * gfs2_next_unreserved_block - Return next block that is not reserved + * @rgd: The resource group + * @block: The starting block + * @length: The required length + * @ip: Ignore any reservations for this inode + * + * If the block does not appear in any reservation, then return the + * block number unchanged. If it does appear in the reservation, then + * keep looking through the tree of reservations in order to find the + * first block number which is not reserved. + */ + +static u64 gfs2_next_unreserved_block(struct gfs2_rgrpd *rgd, u64 block, + u32 length, + const struct gfs2_inode *ip) +{ + struct gfs2_blkreserv *rs; + struct rb_node *n; + int rc; + + spin_lock(&rgd->rd_rsspin); + n = rgd->rd_rstree.rb_node; + while (n) { + rs = rb_entry(n, struct gfs2_blkreserv, rs_node); + rc = rs_cmp(block, length, rs); + if (rc < 0) + n = n->rb_left; + else if (rc > 0) + n = n->rb_right; + else + break; + } + + if (n) { + while ((rs_cmp(block, length, rs) == 0) && (ip->i_res != rs)) { + block = gfs2_rbm_to_block(&rs->rs_rbm) + rs->rs_free; + n = n->rb_right; + if (n == NULL) + break; + rs = rb_entry(n, struct gfs2_blkreserv, rs_node); + } + } + + spin_unlock(&rgd->rd_rsspin); + return block; +} + +/** + * gfs2_reservation_check_and_update - Check for reservations during block alloc + * @rbm: The current position in the resource group + * @ip: The inode for which we are searching for blocks + * @minext: The minimum extent length + * @maxext: A pointer to the maximum extent structure * - * If there's room for the requested blocks to be allocated from the RG: + * This checks the current position in the rgrp to see whether there is + * a reservation covering this block. If not then this function is a + * no-op. If there is, then the position is moved to the end of the + * contiguous reservation(s) so that we are pointing at the first + * non-reserved block. * - * Returns: 1 on success (it fits), 0 on failure (it doesn't fit) + * Returns: 0 if no reservation, 1 if @rbm has changed, otherwise an error */ -static int try_rgrp_fit(const struct gfs2_rgrpd *rgd, const struct gfs2_inode *ip) +static int gfs2_reservation_check_and_update(struct gfs2_rbm *rbm, + const struct gfs2_inode *ip, + u32 minext, + struct gfs2_extent *maxext) { - const struct gfs2_blkreserv *rs = ip->i_res; + u64 block = gfs2_rbm_to_block(rbm); + u32 extlen = 1; + u64 nblock; + int ret; - if (rgd->rd_flags & (GFS2_RGF_NOALLOC | GFS2_RDF_ERROR)) - return 0; - if (rgd->rd_free_clone >= rs->rs_requested) - return 1; - return 0; + /* + * If we have a minimum extent length, then skip over any extent + * which is less than the min extent length in size. + */ + if (minext) { + extlen = gfs2_free_extlen(rbm, minext); + if (extlen <= maxext->len) + goto fail; + } + + /* + * Check the extent which has been found against the reservations + * and skip if parts of it are already reserved + */ + nblock = gfs2_next_unreserved_block(rbm->rgd, block, extlen, ip); + if (nblock == block) { + if (!minext || extlen >= minext) + return 0; + + if (extlen > maxext->len) { + maxext->len = extlen; + maxext->rbm = *rbm; + } +fail: + nblock = block + extlen; + } + ret = gfs2_rbm_from_block(rbm, nblock); + if (ret < 0) + return ret; + return 1; } -static inline u32 gfs2_bi2rgd_blk(struct gfs2_bitmap *bi, u32 blk) +/** + * gfs2_rbm_find - Look for blocks of a particular state + * @rbm: Value/result starting position and final position + * @state: The state which we want to find + * @minext: Pointer to the requested extent length (NULL for a single block) + * This is updated to be the actual reservation size. + * @ip: If set, check for reservations + * @nowrap: Stop looking at the end of the rgrp, rather than wrapping + * around until we've reached the starting point. + * @ap: the allocation parameters + * + * Side effects: + * - If looking for free blocks, we set GBF_FULL on each bitmap which + * has no free blocks in it. + * - If looking for free blocks, we set rd_extfail_pt on each rgrp which + * has come up short on a free block search. + * + * Returns: 0 on success, -ENOSPC if there is no block of the requested state + */ + +static int gfs2_rbm_find(struct gfs2_rbm *rbm, u8 state, u32 *minext, + const struct gfs2_inode *ip, bool nowrap, + const struct gfs2_alloc_parms *ap) { - return (bi->bi_start * GFS2_NBBY) + blk; + struct buffer_head *bh; + int initial_bii; + u32 initial_offset; + int first_bii = rbm->bii; + u32 first_offset = rbm->offset; + u32 offset; + u8 *buffer; + int n = 0; + int iters = rbm->rgd->rd_length; + int ret; + struct gfs2_bitmap *bi; + struct gfs2_extent maxext = { .rbm.rgd = rbm->rgd, }; + + /* If we are not starting at the beginning of a bitmap, then we + * need to add one to the bitmap count to ensure that we search + * the starting bitmap twice. + */ + if (rbm->offset != 0) + iters++; + + while(1) { + bi = rbm_bi(rbm); + if (test_bit(GBF_FULL, &bi->bi_flags) && + (state == GFS2_BLKST_FREE)) + goto next_bitmap; + + bh = bi->bi_bh; + buffer = bh->b_data + bi->bi_offset; + WARN_ON(!buffer_uptodate(bh)); + if (state != GFS2_BLKST_UNLINKED && bi->bi_clone) + buffer = bi->bi_clone + bi->bi_offset; + initial_offset = rbm->offset; + offset = gfs2_bitfit(buffer, bi->bi_len, rbm->offset, state); + if (offset == BFITNOENT) + goto bitmap_full; + rbm->offset = offset; + if (ip == NULL) + return 0; + + initial_bii = rbm->bii; + ret = gfs2_reservation_check_and_update(rbm, ip, + minext ? *minext : 0, + &maxext); + if (ret == 0) + return 0; + if (ret > 0) { + n += (rbm->bii - initial_bii); + goto next_iter; + } + if (ret == -E2BIG) { + rbm->bii = 0; + rbm->offset = 0; + n += (rbm->bii - initial_bii); + goto res_covered_end_of_rgrp; + } + return ret; + +bitmap_full: /* Mark bitmap as full and fall through */ + if ((state == GFS2_BLKST_FREE) && initial_offset == 0) { + struct gfs2_bitmap *bi = rbm_bi(rbm); + set_bit(GBF_FULL, &bi->bi_flags); + } + +next_bitmap: /* Find next bitmap in the rgrp */ + rbm->offset = 0; + rbm->bii++; + if (rbm->bii == rbm->rgd->rd_length) + rbm->bii = 0; +res_covered_end_of_rgrp: + if ((rbm->bii == 0) && nowrap) + break; + n++; +next_iter: + if (n >= iters) + break; + } + + if (minext == NULL || state != GFS2_BLKST_FREE) + return -ENOSPC; + + /* If the extent was too small, and it's smaller than the smallest + to have failed before, remember for future reference that it's + useless to search this rgrp again for this amount or more. */ + if ((first_offset == 0) && (first_bii == 0) && + (*minext < rbm->rgd->rd_extfail_pt)) + rbm->rgd->rd_extfail_pt = *minext; + + /* If the maximum extent we found is big enough to fulfill the + minimum requirements, use it anyway. */ + if (maxext.len) { + *rbm = maxext.rbm; + *minext = maxext.len; + return 0; + } + + return -ENOSPC; } /** * try_rgrp_unlink - Look for any unlinked, allocated, but unused inodes * @rgd: The rgrp + * @last_unlinked: block address of the last dinode we unlinked + * @skip: block address we should explicitly not unlink * * Returns: 0 if no error * The inode, if one has been found, in inode. @@ -928,34 +1724,34 @@ static inline u32 gfs2_bi2rgd_blk(struct gfs2_bitmap *bi, u32 blk) static void try_rgrp_unlink(struct gfs2_rgrpd *rgd, u64 *last_unlinked, u64 skip) { - u32 goal = 0, block; - u64 no_addr; + u64 block; struct gfs2_sbd *sdp = rgd->rd_sbd; struct gfs2_glock *gl; struct gfs2_inode *ip; int error; int found = 0; - struct gfs2_bitmap *bi; + struct gfs2_rbm rbm = { .rgd = rgd, .bii = 0, .offset = 0 }; - while (goal < rgd->rd_data) { + while (1) { down_write(&sdp->sd_log_flush_lock); - block = rgblk_search(rgd, goal, GFS2_BLKST_UNLINKED, &bi); + error = gfs2_rbm_find(&rbm, GFS2_BLKST_UNLINKED, NULL, NULL, + true, NULL); up_write(&sdp->sd_log_flush_lock); - if (block == BFITNOENT) + if (error == -ENOSPC) + break; + if (WARN_ON_ONCE(error)) break; - block = gfs2_bi2rgd_blk(bi, block); - /* rgblk_search can return a block < goal, so we need to - keep it marching forward. */ - no_addr = block + rgd->rd_data0; - goal = max(block + 1, goal + 1); - if (*last_unlinked != NO_BLOCK && no_addr <= *last_unlinked) + block = gfs2_rbm_to_block(&rbm); + if (gfs2_rbm_from_block(&rbm, block + 1)) + break; + if (*last_unlinked != NO_BLOCK && block <= *last_unlinked) continue; - if (no_addr == skip) + if (block == skip) continue; - *last_unlinked = no_addr; + *last_unlinked = block; - error = gfs2_glock_get(sdp, no_addr, &gfs2_inode_glops, CREATE, &gl); + error = gfs2_glock_get(sdp, block, &gfs2_inode_glops, CREATE, &gl); if (error) continue; @@ -983,118 +1779,232 @@ static void try_rgrp_unlink(struct gfs2_rgrpd *rgd, u64 *last_unlinked, u64 skip } /** - * get_local_rgrp - Choose and lock a rgrp for allocation - * @ip: the inode to reserve space for - * @rgp: the chosen and locked rgrp + * gfs2_rgrp_congested - Use stats to figure out whether an rgrp is congested + * @rgd: The rgrp in question + * @loops: An indication of how picky we can be (0=very, 1=less so) * - * Try to acquire rgrp in way which avoids contending with others. + * This function uses the recently added glock statistics in order to + * figure out whether a parciular resource group is suffering from + * contention from multiple nodes. This is done purely on the basis + * of timings, since this is the only data we have to work with and + * our aim here is to reject a resource group which is highly contended + * but (very important) not to do this too often in order to ensure that + * we do not land up introducing fragmentation by changing resource + * groups when not actually required. * - * Returns: errno + * The calculation is fairly simple, we want to know whether the SRTTB + * (i.e. smoothed round trip time for blocking operations) to acquire + * the lock for this rgrp's glock is significantly greater than the + * time taken for resource groups on average. We introduce a margin in + * the form of the variable @var which is computed as the sum of the two + * respective variences, and multiplied by a factor depending on @loops + * and whether we have a lot of data to base the decision on. This is + * then tested against the square difference of the means in order to + * decide whether the result is statistically significant or not. + * + * Returns: A boolean verdict on the congestion status */ -static int get_local_rgrp(struct gfs2_inode *ip, u64 *last_unlinked) +static bool gfs2_rgrp_congested(const struct gfs2_rgrpd *rgd, int loops) { - struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); - struct gfs2_rgrpd *rgd, *begin = NULL; - struct gfs2_blkreserv *rs = ip->i_res; - int error, rg_locked, flags = LM_FLAG_TRY; - int loops = 0; + const struct gfs2_glock *gl = rgd->rd_gl; + const struct gfs2_sbd *sdp = gl->gl_sbd; + struct gfs2_lkstats *st; + s64 r_dcount, l_dcount; + s64 r_srttb, l_srttb; + s64 srttb_diff; + s64 sqr_diff; + s64 var; + + preempt_disable(); + st = &this_cpu_ptr(sdp->sd_lkstats)->lkstats[LM_TYPE_RGRP]; + r_srttb = st->stats[GFS2_LKS_SRTTB]; + r_dcount = st->stats[GFS2_LKS_DCOUNT]; + var = st->stats[GFS2_LKS_SRTTVARB] + + gl->gl_stats.stats[GFS2_LKS_SRTTVARB]; + preempt_enable(); + + l_srttb = gl->gl_stats.stats[GFS2_LKS_SRTTB]; + l_dcount = gl->gl_stats.stats[GFS2_LKS_DCOUNT]; + + if ((l_dcount < 1) || (r_dcount < 1) || (r_srttb == 0)) + return false; + + srttb_diff = r_srttb - l_srttb; + sqr_diff = srttb_diff * srttb_diff; + + var *= 2; + if (l_dcount < 8 || r_dcount < 8) + var *= 2; + if (loops == 1) + var *= 2; + + return ((srttb_diff < 0) && (sqr_diff > var)); +} - if (ip->i_rgd && rgrp_contains_block(ip->i_rgd, ip->i_goal)) - rgd = begin = ip->i_rgd; - else - rgd = begin = gfs2_blk2rgrpd(sdp, ip->i_goal); +/** + * gfs2_rgrp_used_recently + * @rs: The block reservation with the rgrp to test + * @msecs: The time limit in milliseconds + * + * Returns: True if the rgrp glock has been used within the time limit + */ +static bool gfs2_rgrp_used_recently(const struct gfs2_blkreserv *rs, + u64 msecs) +{ + u64 tdiff; - if (rgd == NULL) - return -EBADSLT; + tdiff = ktime_to_ns(ktime_sub(ktime_get_real(), + rs->rs_rbm.rgd->rd_gl->gl_dstamp)); - while (loops < 3) { - rg_locked = 0; + return tdiff > (msecs * 1000 * 1000); +} - if (gfs2_glock_is_locked_by_me(rgd->rd_gl)) { - rg_locked = 1; - error = 0; - } else { - error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, - flags, &rs->rs_rgd_gh); - } - switch (error) { - case 0: - if (try_rgrp_fit(rgd, ip)) { - ip->i_rgd = rgd; - return 0; - } - if (rgd->rd_flags & GFS2_RDF_CHECK) - try_rgrp_unlink(rgd, last_unlinked, ip->i_no_addr); - if (!rg_locked) - gfs2_glock_dq_uninit(&rs->rs_rgd_gh); - /* fall through */ - case GLR_TRYFAILED: - rgd = gfs2_rgrpd_get_next(rgd); - if (rgd == begin) { - flags = 0; - loops++; - } - break; - default: - return error; - } - } +static u32 gfs2_orlov_skip(const struct gfs2_inode *ip) +{ + const struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); + u32 skip; - return -ENOSPC; + get_random_bytes(&skip, sizeof(skip)); + return skip % sdp->sd_rgrps; } -static void gfs2_blkrsv_put(struct gfs2_inode *ip) +static bool gfs2_select_rgrp(struct gfs2_rgrpd **pos, const struct gfs2_rgrpd *begin) { - BUG_ON(ip->i_res == NULL); - kfree(ip->i_res); - ip->i_res = NULL; + struct gfs2_rgrpd *rgd = *pos; + struct gfs2_sbd *sdp = rgd->rd_sbd; + + rgd = gfs2_rgrpd_get_next(rgd); + if (rgd == NULL) + rgd = gfs2_rgrpd_get_first(sdp); + *pos = rgd; + if (rgd != begin) /* If we didn't wrap */ + return true; + return false; } /** * gfs2_inplace_reserve - Reserve space in the filesystem * @ip: the inode to reserve space for + * @ap: the allocation parameters * * Returns: errno */ -int gfs2_inplace_reserve(struct gfs2_inode *ip, u32 requested) +int gfs2_inplace_reserve(struct gfs2_inode *ip, const struct gfs2_alloc_parms *ap) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); - struct gfs2_blkreserv *rs; - int error = 0; + struct gfs2_rgrpd *begin = NULL; + struct gfs2_blkreserv *rs = ip->i_res; + int error = 0, rg_locked, flags = 0; u64 last_unlinked = NO_BLOCK; - int tries = 0; - - rs = gfs2_blkrsv_get(ip); - if (!rs) - return -ENOMEM; + int loops = 0; + u32 skip = 0; - rs->rs_requested = requested; - if (gfs2_assert_warn(sdp, requested)) { - error = -EINVAL; - goto out; + if (sdp->sd_args.ar_rgrplvb) + flags |= GL_SKIP; + if (gfs2_assert_warn(sdp, ap->target)) + return -EINVAL; + if (gfs2_rs_active(rs)) { + begin = rs->rs_rbm.rgd; + } else if (ip->i_rgd && rgrp_contains_block(ip->i_rgd, ip->i_goal)) { + rs->rs_rbm.rgd = begin = ip->i_rgd; + } else { + rs->rs_rbm.rgd = begin = gfs2_blk2rgrpd(sdp, ip->i_goal, 1); } + if (S_ISDIR(ip->i_inode.i_mode) && (ap->aflags & GFS2_AF_ORLOV)) + skip = gfs2_orlov_skip(ip); + if (rs->rs_rbm.rgd == NULL) + return -EBADSLT; - do { - error = get_local_rgrp(ip, &last_unlinked); - if (error != -ENOSPC) - break; + while (loops < 3) { + rg_locked = 1; + + if (!gfs2_glock_is_locked_by_me(rs->rs_rbm.rgd->rd_gl)) { + rg_locked = 0; + if (skip && skip--) + goto next_rgrp; + if (!gfs2_rs_active(rs) && (loops < 2) && + gfs2_rgrp_used_recently(rs, 1000) && + gfs2_rgrp_congested(rs->rs_rbm.rgd, loops)) + goto next_rgrp; + error = gfs2_glock_nq_init(rs->rs_rbm.rgd->rd_gl, + LM_ST_EXCLUSIVE, flags, + &rs->rs_rgd_gh); + if (unlikely(error)) + return error; + if (!gfs2_rs_active(rs) && (loops < 2) && + gfs2_rgrp_congested(rs->rs_rbm.rgd, loops)) + goto skip_rgrp; + if (sdp->sd_args.ar_rgrplvb) { + error = update_rgrp_lvb(rs->rs_rbm.rgd); + if (unlikely(error)) { + gfs2_glock_dq_uninit(&rs->rs_rgd_gh); + return error; + } + } + } + + /* Skip unuseable resource groups */ + if ((rs->rs_rbm.rgd->rd_flags & (GFS2_RGF_NOALLOC | + GFS2_RDF_ERROR)) || + (ap->target > rs->rs_rbm.rgd->rd_extfail_pt)) + goto skip_rgrp; + + if (sdp->sd_args.ar_rgrplvb) + gfs2_rgrp_bh_get(rs->rs_rbm.rgd); + + /* Get a reservation if we don't already have one */ + if (!gfs2_rs_active(rs)) + rg_mblk_search(rs->rs_rbm.rgd, ip, ap); + + /* Skip rgrps when we can't get a reservation on first pass */ + if (!gfs2_rs_active(rs) && (loops < 1)) + goto check_rgrp; + + /* If rgrp has enough free space, use it */ + if (rs->rs_rbm.rgd->rd_free_clone >= ap->target) { + ip->i_rgd = rs->rs_rbm.rgd; + return 0; + } + +check_rgrp: + /* Check for unlinked inodes which can be reclaimed */ + if (rs->rs_rbm.rgd->rd_flags & GFS2_RDF_CHECK) + try_rgrp_unlink(rs->rs_rbm.rgd, &last_unlinked, + ip->i_no_addr); +skip_rgrp: + /* Drop reservation, if we couldn't use reserved rgrp */ + if (gfs2_rs_active(rs)) + gfs2_rs_deltree(rs); + + /* Unlock rgrp if required */ + if (!rg_locked) + gfs2_glock_dq_uninit(&rs->rs_rgd_gh); +next_rgrp: + /* Find the next rgrp, and continue looking */ + if (gfs2_select_rgrp(&rs->rs_rbm.rgd, begin)) + continue; + if (skip) + continue; + + /* If we've scanned all the rgrps, but found no free blocks + * then this checks for some less likely conditions before + * trying again. + */ + loops++; /* Check that fs hasn't grown if writing to rindex */ if (ip == GFS2_I(sdp->sd_rindex) && !sdp->sd_rindex_uptodate) { error = gfs2_ri_update(ip); if (error) - break; - continue; + return error; } /* Flushing the log may release space */ - gfs2_log_flush(sdp, NULL); - } while (tries++ < 3); + if (loops == 2) + gfs2_log_flush(sdp, NULL, NORMAL_FLUSH); + } -out: - if (error) - gfs2_blkrsv_put(ip); - return error; + return -ENOSPC; } /** @@ -1110,7 +2020,6 @@ void gfs2_inplace_release(struct gfs2_inode *ip) if (rs->rs_rgd_gh.gh_gl) gfs2_glock_dq_uninit(&rs->rs_rgd_gh); - gfs2_blkrsv_put(ip); } /** @@ -1123,151 +2032,47 @@ void gfs2_inplace_release(struct gfs2_inode *ip) static unsigned char gfs2_get_block_type(struct gfs2_rgrpd *rgd, u64 block) { - struct gfs2_bitmap *bi = NULL; - u32 length, rgrp_block, buf_block; - unsigned int buf; - unsigned char type; - - length = rgd->rd_length; - rgrp_block = block - rgd->rd_data0; - - for (buf = 0; buf < length; buf++) { - bi = rgd->rd_bits + buf; - if (rgrp_block < (bi->bi_start + bi->bi_len) * GFS2_NBBY) - break; - } + struct gfs2_rbm rbm = { .rgd = rgd, }; + int ret; - gfs2_assert(rgd->rd_sbd, buf < length); - buf_block = rgrp_block - bi->bi_start * GFS2_NBBY; + ret = gfs2_rbm_from_block(&rbm, block); + WARN_ON_ONCE(ret != 0); - type = gfs2_testbit(rgd, bi->bi_bh->b_data + bi->bi_offset, - bi->bi_len, buf_block); - - return type; + return gfs2_testbit(&rbm); } -/** - * rgblk_search - find a block in @state - * @rgd: the resource group descriptor - * @goal: the goal block within the RG (start here to search for avail block) - * @state: GFS2_BLKST_XXX the before-allocation state to find - * @dinode: TRUE if the first block we allocate is for a dinode - * @rbi: address of the pointer to the bitmap containing the block found - * - * Walk rgrp's bitmap to find bits that represent a block in @state. - * - * This function never fails, because we wouldn't call it unless we - * know (from reservation results, etc.) that a block is available. - * - * Scope of @goal is just within rgrp, not the whole filesystem. - * Scope of @returned block is just within bitmap, not the whole filesystem. - * - * Returns: the block number found relative to the bitmap rbi - */ - -static u32 rgblk_search(struct gfs2_rgrpd *rgd, u32 goal, - unsigned char state, - struct gfs2_bitmap **rbi) -{ - struct gfs2_bitmap *bi = NULL; - const u32 length = rgd->rd_length; - u32 blk = BFITNOENT; - unsigned int buf, x; - const u8 *buffer = NULL; - - *rbi = NULL; - /* Find bitmap block that contains bits for goal block */ - for (buf = 0; buf < length; buf++) { - bi = rgd->rd_bits + buf; - /* Convert scope of "goal" from rgrp-wide to within found bit block */ - if (goal < (bi->bi_start + bi->bi_len) * GFS2_NBBY) { - goal -= bi->bi_start * GFS2_NBBY; - goto do_search; - } - } - buf = 0; - goal = 0; - -do_search: - /* Search (up to entire) bitmap in this rgrp for allocatable block. - "x <= length", instead of "x < length", because we typically start - the search in the middle of a bit block, but if we can't find an - allocatable block anywhere else, we want to be able wrap around and - search in the first part of our first-searched bit block. */ - for (x = 0; x <= length; x++) { - bi = rgd->rd_bits + buf; - - if (test_bit(GBF_FULL, &bi->bi_flags) && - (state == GFS2_BLKST_FREE)) - goto skip; - - /* The GFS2_BLKST_UNLINKED state doesn't apply to the clone - bitmaps, so we must search the originals for that. */ - buffer = bi->bi_bh->b_data + bi->bi_offset; - WARN_ON(!buffer_uptodate(bi->bi_bh)); - if (state != GFS2_BLKST_UNLINKED && bi->bi_clone) - buffer = bi->bi_clone + bi->bi_offset; - - blk = gfs2_bitfit(buffer, bi->bi_len, goal, state); - if (blk != BFITNOENT) - break; - - if ((goal == 0) && (state == GFS2_BLKST_FREE)) - set_bit(GBF_FULL, &bi->bi_flags); - - /* Try next bitmap block (wrap back to rgrp header if at end) */ -skip: - buf++; - buf %= length; - goal = 0; - } - - if (blk != BFITNOENT) - *rbi = bi; - - return blk; -} /** * gfs2_alloc_extent - allocate an extent from a given bitmap - * @rgd: the resource group descriptor - * @bi: the bitmap within the rgrp - * @blk: the block within the bitmap + * @rbm: the resource group information * @dinode: TRUE if the first block we allocate is for a dinode - * @n: The extent length + * @n: The extent length (value/result) * - * Add the found bitmap buffer to the transaction. + * Add the bitmap buffer to the transaction. * Set the found bits to @new_state to change block's allocation state. - * Returns: starting block number of the extent (fs scope) */ -static u64 gfs2_alloc_extent(struct gfs2_rgrpd *rgd, struct gfs2_bitmap *bi, - u32 blk, bool dinode, unsigned int *n) +static void gfs2_alloc_extent(const struct gfs2_rbm *rbm, bool dinode, + unsigned int *n) { + struct gfs2_rbm pos = { .rgd = rbm->rgd, }; const unsigned int elen = *n; - u32 goal; - const u8 *buffer = NULL; - - *n = 0; - buffer = bi->bi_bh->b_data + bi->bi_offset; - gfs2_trans_add_bh(rgd->rd_gl, bi->bi_bh, 1); - gfs2_setbit(rgd, bi->bi_bh->b_data, bi->bi_clone, bi->bi_offset, - bi, blk, dinode ? GFS2_BLKST_DINODE : GFS2_BLKST_USED); - (*n)++; - goal = blk; + u64 block; + int ret; + + *n = 1; + block = gfs2_rbm_to_block(rbm); + gfs2_trans_add_meta(rbm->rgd->rd_gl, rbm_bi(rbm)->bi_bh); + gfs2_setbit(rbm, true, dinode ? GFS2_BLKST_DINODE : GFS2_BLKST_USED); + block++; while (*n < elen) { - goal++; - if (goal >= (bi->bi_len * GFS2_NBBY)) + ret = gfs2_rbm_from_block(&pos, block); + if (ret || gfs2_testbit(&pos) != GFS2_BLKST_FREE) break; - if (gfs2_testbit(rgd, buffer, bi->bi_len, goal) != - GFS2_BLKST_FREE) - break; - gfs2_setbit(rgd, bi->bi_bh->b_data, bi->bi_clone, bi->bi_offset, - bi, goal, GFS2_BLKST_USED); + gfs2_trans_add_meta(pos.rgd->rd_gl, rbm_bi(&pos)->bi_bh); + gfs2_setbit(&pos, true, GFS2_BLKST_USED); (*n)++; + block++; } - blk = gfs2_bi2rgd_blk(bi, blk); - rgd->rd_last_alloc = blk + *n - 1; - return rgd->rd_data0 + blk; } /** @@ -1283,47 +2088,31 @@ static u64 gfs2_alloc_extent(struct gfs2_rgrpd *rgd, struct gfs2_bitmap *bi, static struct gfs2_rgrpd *rgblk_free(struct gfs2_sbd *sdp, u64 bstart, u32 blen, unsigned char new_state) { - struct gfs2_rgrpd *rgd; - struct gfs2_bitmap *bi = NULL; - u32 length, rgrp_blk, buf_blk; - unsigned int buf; + struct gfs2_rbm rbm; + struct gfs2_bitmap *bi; - rgd = gfs2_blk2rgrpd(sdp, bstart); - if (!rgd) { + rbm.rgd = gfs2_blk2rgrpd(sdp, bstart, 1); + if (!rbm.rgd) { if (gfs2_consist(sdp)) fs_err(sdp, "block = %llu\n", (unsigned long long)bstart); return NULL; } - length = rgd->rd_length; - - rgrp_blk = bstart - rgd->rd_data0; - while (blen--) { - for (buf = 0; buf < length; buf++) { - bi = rgd->rd_bits + buf; - if (rgrp_blk < (bi->bi_start + bi->bi_len) * GFS2_NBBY) - break; - } - - gfs2_assert(rgd->rd_sbd, buf < length); - - buf_blk = rgrp_blk - bi->bi_start * GFS2_NBBY; - rgrp_blk++; - + gfs2_rbm_from_block(&rbm, bstart); + bi = rbm_bi(&rbm); + bstart++; if (!bi->bi_clone) { bi->bi_clone = kmalloc(bi->bi_bh->b_size, GFP_NOFS | __GFP_NOFAIL); memcpy(bi->bi_clone + bi->bi_offset, - bi->bi_bh->b_data + bi->bi_offset, - bi->bi_len); + bi->bi_bh->b_data + bi->bi_offset, bi->bi_len); } - gfs2_trans_add_bh(rgd->rd_gl, bi->bi_bh, 1); - gfs2_setbit(rgd, bi->bi_bh->b_data, NULL, bi->bi_offset, - bi, buf_blk, new_state); + gfs2_trans_add_meta(rbm.rgd->rd_gl, bi->bi_bh); + gfs2_setbit(&rbm, false, new_state); } - return rgd; + return rbm.rgd; } /** @@ -1333,15 +2122,24 @@ static struct gfs2_rgrpd *rgblk_free(struct gfs2_sbd *sdp, u64 bstart, * */ -int gfs2_rgrp_dump(struct seq_file *seq, const struct gfs2_glock *gl) +void gfs2_rgrp_dump(struct seq_file *seq, const struct gfs2_glock *gl) { - const struct gfs2_rgrpd *rgd = gl->gl_object; + struct gfs2_rgrpd *rgd = gl->gl_object; + struct gfs2_blkreserv *trs; + const struct rb_node *n; + if (rgd == NULL) - return 0; - gfs2_print_dbg(seq, " R: n:%llu f:%02x b:%u/%u i:%u\n", + return; + gfs2_print_dbg(seq, " R: n:%llu f:%02x b:%u/%u i:%u r:%u e:%u\n", (unsigned long long)rgd->rd_addr, rgd->rd_flags, - rgd->rd_free, rgd->rd_free_clone, rgd->rd_dinodes); - return 0; + rgd->rd_free, rgd->rd_free_clone, rgd->rd_dinodes, + rgd->rd_reserved, rgd->rd_extfail_pt); + spin_lock(&rgd->rd_rsspin); + for (n = rb_first(&rgd->rd_rstree); n; n = rb_next(&trs->rs_node)) { + trs = rb_entry(n, struct gfs2_blkreserv, rs_node); + dump_rs(seq, trs); + } + spin_unlock(&rgd->rd_rsspin); } static void gfs2_rgrp_error(struct gfs2_rgrpd *rgd) @@ -1355,10 +2153,77 @@ static void gfs2_rgrp_error(struct gfs2_rgrpd *rgd) } /** + * gfs2_adjust_reservation - Adjust (or remove) a reservation after allocation + * @ip: The inode we have just allocated blocks for + * @rbm: The start of the allocated blocks + * @len: The extent length + * + * Adjusts a reservation after an allocation has taken place. If the + * reservation does not match the allocation, or if it is now empty + * then it is removed. + */ + +static void gfs2_adjust_reservation(struct gfs2_inode *ip, + const struct gfs2_rbm *rbm, unsigned len) +{ + struct gfs2_blkreserv *rs = ip->i_res; + struct gfs2_rgrpd *rgd = rbm->rgd; + unsigned rlen; + u64 block; + int ret; + + spin_lock(&rgd->rd_rsspin); + if (gfs2_rs_active(rs)) { + if (gfs2_rbm_eq(&rs->rs_rbm, rbm)) { + block = gfs2_rbm_to_block(rbm); + ret = gfs2_rbm_from_block(&rs->rs_rbm, block + len); + rlen = min(rs->rs_free, len); + rs->rs_free -= rlen; + rgd->rd_reserved -= rlen; + trace_gfs2_rs(rs, TRACE_RS_CLAIM); + if (rs->rs_free && !ret) + goto out; + } + __rs_deltree(rs); + } +out: + spin_unlock(&rgd->rd_rsspin); +} + +/** + * gfs2_set_alloc_start - Set starting point for block allocation + * @rbm: The rbm which will be set to the required location + * @ip: The gfs2 inode + * @dinode: Flag to say if allocation includes a new inode + * + * This sets the starting point from the reservation if one is active + * otherwise it falls back to guessing a start point based on the + * inode's goal block or the last allocation point in the rgrp. + */ + +static void gfs2_set_alloc_start(struct gfs2_rbm *rbm, + const struct gfs2_inode *ip, bool dinode) +{ + u64 goal; + + if (gfs2_rs_active(ip->i_res)) { + *rbm = ip->i_res->rs_rbm; + return; + } + + if (!dinode && rgrp_contains_block(rbm->rgd, ip->i_goal)) + goal = ip->i_goal; + else + goal = rbm->rgd->rd_last_alloc + rbm->rgd->rd_data0; + + gfs2_rbm_from_block(rbm, goal); +} + +/** * gfs2_alloc_blocks - Allocate one or more blocks of data and/or a dinode * @ip: the inode to allocate the block for * @bn: Used to return the starting block number - * @ndata: requested number of blocks/extent length (value/result) + * @nblocks: requested number of blocks/extent length (value/result) * @dinode: 1 if we're allocating a dinode block, else 0 * @generation: the generation number of the inode * @@ -1370,33 +2235,34 @@ int gfs2_alloc_blocks(struct gfs2_inode *ip, u64 *bn, unsigned int *nblocks, { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); struct buffer_head *dibh; - struct gfs2_rgrpd *rgd; + struct gfs2_rbm rbm = { .rgd = ip->i_rgd, }; unsigned int ndata; - u32 goal, blk; /* block, within the rgrp scope */ u64 block; /* block, within the file system scope */ int error; - struct gfs2_bitmap *bi; - /* Only happens if there is a bug in gfs2, return something distinctive - * to ensure that it is noticed. - */ - if (ip->i_res == NULL) - return -ECANCELED; + gfs2_set_alloc_start(&rbm, ip, dinode); + error = gfs2_rbm_find(&rbm, GFS2_BLKST_FREE, NULL, ip, false, NULL); - rgd = ip->i_rgd; - - if (!dinode && rgrp_contains_block(rgd, ip->i_goal)) - goal = ip->i_goal - rgd->rd_data0; - else - goal = rgd->rd_last_alloc; - - blk = rgblk_search(rgd, goal, GFS2_BLKST_FREE, &bi); + if (error == -ENOSPC) { + gfs2_set_alloc_start(&rbm, ip, dinode); + error = gfs2_rbm_find(&rbm, GFS2_BLKST_FREE, NULL, NULL, false, + NULL); + } /* Since all blocks are reserved in advance, this shouldn't happen */ - if (blk == BFITNOENT) + if (error) { + fs_warn(sdp, "inum=%llu error=%d, nblocks=%u, full=%d fail_pt=%d\n", + (unsigned long long)ip->i_no_addr, error, *nblocks, + test_bit(GBF_FULL, &rbm.rgd->rd_bits->bi_flags), + rbm.rgd->rd_extfail_pt); goto rgrp_error; + } - block = gfs2_alloc_extent(rgd, bi, blk, dinode, nblocks); + gfs2_alloc_extent(&rbm, dinode, nblocks); + block = gfs2_rbm_to_block(&rbm); + rbm.rgd->rd_last_alloc = block - rbm.rgd->rd_data0; + if (gfs2_rs_active(ip->i_res)) + gfs2_adjust_reservation(ip, &rbm, *nblocks); ndata = *nblocks; if (dinode) ndata--; @@ -1407,46 +2273,43 @@ int gfs2_alloc_blocks(struct gfs2_inode *ip, u64 *bn, unsigned int *nblocks, if (error == 0) { struct gfs2_dinode *di = (struct gfs2_dinode *)dibh->b_data; - gfs2_trans_add_bh(ip->i_gl, dibh, 1); + gfs2_trans_add_meta(ip->i_gl, dibh); di->di_goal_meta = di->di_goal_data = cpu_to_be64(ip->i_goal); brelse(dibh); } } - if (rgd->rd_free < *nblocks) + if (rbm.rgd->rd_free < *nblocks) { + pr_warn("nblocks=%u\n", *nblocks); goto rgrp_error; + } - rgd->rd_free -= *nblocks; + rbm.rgd->rd_free -= *nblocks; if (dinode) { - rgd->rd_dinodes++; - *generation = rgd->rd_igeneration++; + rbm.rgd->rd_dinodes++; + *generation = rbm.rgd->rd_igeneration++; if (*generation == 0) - *generation = rgd->rd_igeneration++; + *generation = rbm.rgd->rd_igeneration++; } - gfs2_trans_add_bh(rgd->rd_gl, rgd->rd_bits[0].bi_bh, 1); - gfs2_rgrp_out(rgd, rgd->rd_bits[0].bi_bh->b_data); + gfs2_trans_add_meta(rbm.rgd->rd_gl, rbm.rgd->rd_bits[0].bi_bh); + gfs2_rgrp_out(rbm.rgd, rbm.rgd->rd_bits[0].bi_bh->b_data); + gfs2_rgrp_ondisk2lvb(rbm.rgd->rd_rgl, rbm.rgd->rd_bits[0].bi_bh->b_data); gfs2_statfs_change(sdp, 0, -(s64)*nblocks, dinode ? 1 : 0); if (dinode) - gfs2_trans_add_unrevoke(sdp, block, 1); + gfs2_trans_add_unrevoke(sdp, block, *nblocks); - /* - * This needs reviewing to see why we cannot do the quota change - * at this point in the dinode case. - */ - if (ndata) - gfs2_quota_change(ip, ndata, ip->i_inode.i_uid, - ip->i_inode.i_gid); + gfs2_quota_change(ip, *nblocks, ip->i_inode.i_uid, ip->i_inode.i_gid); - rgd->rd_free_clone -= *nblocks; - trace_gfs2_block_alloc(ip, block, *nblocks, + rbm.rgd->rd_free_clone -= *nblocks; + trace_gfs2_block_alloc(ip, rbm.rgd, block, *nblocks, dinode ? GFS2_BLKST_DINODE : GFS2_BLKST_USED); *bn = block; return 0; rgrp_error: - gfs2_rgrp_error(rgd); + gfs2_rgrp_error(rbm.rgd); return -EIO; } @@ -1467,11 +2330,12 @@ void __gfs2_free_blocks(struct gfs2_inode *ip, u64 bstart, u32 blen, int meta) rgd = rgblk_free(sdp, bstart, blen, GFS2_BLKST_FREE); if (!rgd) return; - trace_gfs2_block_alloc(ip, bstart, blen, GFS2_BLKST_FREE); + trace_gfs2_block_alloc(ip, rgd, bstart, blen, GFS2_BLKST_FREE); rgd->rd_free += blen; - - gfs2_trans_add_bh(rgd->rd_gl, rgd->rd_bits[0].bi_bh, 1); + rgd->rd_flags &= ~GFS2_RGF_TRIMMED; + gfs2_trans_add_meta(rgd->rd_gl, rgd->rd_bits[0].bi_bh); gfs2_rgrp_out(rgd, rgd->rd_bits[0].bi_bh->b_data); + gfs2_rgrp_ondisk2lvb(rgd->rd_rgl, rgd->rd_bits[0].bi_bh->b_data); /* Directories keep their data in the metadata address space */ if (meta || ip->i_depth) @@ -1505,9 +2369,11 @@ void gfs2_unlink_di(struct inode *inode) rgd = rgblk_free(sdp, blkno, 1, GFS2_BLKST_UNLINKED); if (!rgd) return; - trace_gfs2_block_alloc(ip, blkno, 1, GFS2_BLKST_UNLINKED); - gfs2_trans_add_bh(rgd->rd_gl, rgd->rd_bits[0].bi_bh, 1); + trace_gfs2_block_alloc(ip, rgd, blkno, 1, GFS2_BLKST_UNLINKED); + gfs2_trans_add_meta(rgd->rd_gl, rgd->rd_bits[0].bi_bh); gfs2_rgrp_out(rgd, rgd->rd_bits[0].bi_bh->b_data); + gfs2_rgrp_ondisk2lvb(rgd->rd_rgl, rgd->rd_bits[0].bi_bh->b_data); + update_rgrp_lvb_unlinked(rgd, 1); } static void gfs2_free_uninit_di(struct gfs2_rgrpd *rgd, u64 blkno) @@ -1525,8 +2391,10 @@ static void gfs2_free_uninit_di(struct gfs2_rgrpd *rgd, u64 blkno) rgd->rd_dinodes--; rgd->rd_free++; - gfs2_trans_add_bh(rgd->rd_gl, rgd->rd_bits[0].bi_bh, 1); + gfs2_trans_add_meta(rgd->rd_gl, rgd->rd_bits[0].bi_bh); gfs2_rgrp_out(rgd, rgd->rd_bits[0].bi_bh->b_data); + gfs2_rgrp_ondisk2lvb(rgd->rd_rgl, rgd->rd_bits[0].bi_bh->b_data); + update_rgrp_lvb_unlinked(rgd, -1); gfs2_statfs_change(sdp, 0, +1, -1); } @@ -1535,7 +2403,7 @@ static void gfs2_free_uninit_di(struct gfs2_rgrpd *rgd, u64 blkno) void gfs2_free_di(struct gfs2_rgrpd *rgd, struct gfs2_inode *ip) { gfs2_free_uninit_di(rgd, ip->i_no_addr); - trace_gfs2_block_alloc(ip, ip->i_no_addr, 1, GFS2_BLKST_FREE); + trace_gfs2_block_alloc(ip, rgd, ip->i_no_addr, 1, GFS2_BLKST_FREE); gfs2_quota_change(ip, -1, ip->i_inode.i_uid, ip->i_inode.i_gid); gfs2_meta_wipe(ip, ip->i_no_addr, 1); } @@ -1555,14 +2423,9 @@ int gfs2_check_blk_type(struct gfs2_sbd *sdp, u64 no_addr, unsigned int type) { struct gfs2_rgrpd *rgd; struct gfs2_holder rgd_gh; - int error; + int error = -EINVAL; - error = gfs2_rindex_update(sdp); - if (error) - return error; - - error = -EINVAL; - rgd = gfs2_blk2rgrpd(sdp, no_addr); + rgd = gfs2_blk2rgrpd(sdp, no_addr, 1); if (!rgd) goto fail; @@ -1605,7 +2468,7 @@ void gfs2_rlist_add(struct gfs2_inode *ip, struct gfs2_rgrp_list *rlist, if (ip->i_rgd && rgrp_contains_block(ip->i_rgd, block)) rgd = ip->i_rgd; else - rgd = gfs2_blk2rgrpd(sdp, block); + rgd = gfs2_blk2rgrpd(sdp, block, 1); if (!rgd) { fs_err(sdp, "rlist_add: no rgrp for block %llu\n", (unsigned long long)block); return; @@ -1640,7 +2503,6 @@ void gfs2_rlist_add(struct gfs2_inode *ip, struct gfs2_rgrp_list *rlist, * and initialize an array of glock holders for them * @rlist: the list of resource groups * @state: the lock state to acquire the RG lock in - * @flags: the modifier flags for the holder structures * * FIXME: Don't use NOFAIL * @@ -1660,7 +2522,7 @@ void gfs2_rlist_alloc(struct gfs2_rgrp_list *rlist, unsigned int state) /** * gfs2_rlist_free - free a resource group list - * @list: the list of resource groups + * @rlist: the list of resource groups * */ @@ -1674,6 +2536,7 @@ void gfs2_rlist_free(struct gfs2_rgrp_list *rlist) for (x = 0; x < rlist->rl_rgrps; x++) gfs2_holder_uninit(&rlist->rl_ghs[x]); kfree(rlist->rl_ghs); + rlist->rl_ghs = NULL; } } |