diff options
Diffstat (limited to 'fs/gfs2/rgrp.c')
| -rw-r--r-- | fs/gfs2/rgrp.c | 1383 |
1 files changed, 861 insertions, 522 deletions
diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c index 69317435faa7..b14e54b38ee8 100644 --- a/fs/gfs2/rgrp.c +++ b/fs/gfs2/rgrp.c @@ -1,12 +1,11 @@ +// SPDX-License-Identifier: GPL-2.0-only /* * Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved. * Copyright (C) 2004-2008 Red Hat, Inc. All rights reserved. - * - * This copyrighted material is made available to anyone wishing to use, - * modify, copy, or redistribute it subject to the terms and conditions - * 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> @@ -32,19 +31,28 @@ #include "log.h" #include "inode.h" #include "trace_gfs2.h" +#include "dir.h" #define BFITNOENT ((u32)~0) #define NO_BLOCK ((u64)~0) -#if BITS_PER_LONG == 32 -#define LBITMASK (0x55555555UL) -#define LBITSKIP55 (0x55555555UL) -#define LBITSKIP00 (0x00000000UL) -#else -#define LBITMASK (0x5555555555555555UL) -#define LBITSKIP55 (0x5555555555555555UL) -#define LBITSKIP00 (0x0000000000000000UL) -#endif +struct gfs2_rbm { + struct gfs2_rgrpd *rgd; + u32 offset; /* The offset is bitmap relative */ + int bii; /* Bitmap index */ +}; + +static inline struct gfs2_bitmap *rbm_bi(const struct gfs2_rbm *rbm) +{ + return rbm->rgd->rd_bits + rbm->bii; +} + +static inline u64 gfs2_rbm_to_block(const struct gfs2_rbm *rbm) +{ + BUG_ON(rbm->offset >= rbm->rgd->rd_data); + return rbm->rgd->rd_data0 + (rbm_bi(rbm)->bi_start * GFS2_NBBY) + + rbm->offset; +} /* * These routines are used by the resource group routines (rgrp.c) @@ -57,6 +65,11 @@ * 3 = Used (metadata) */ +struct gfs2_extent { + struct gfs2_rbm rbm; + u32 len; +}; + static const char valid_change[16] = { /* current */ /* n */ 0, 1, 1, 1, @@ -65,8 +78,8 @@ static const char valid_change[16] = { 1, 0, 0, 0 }; -static int gfs2_rbm_find(struct gfs2_rbm *rbm, u8 state, u32 minext, - const struct gfs2_inode *ip, bool nowrap); +static int gfs2_rbm_find(struct gfs2_rbm *rbm, u8 state, u32 *minext, + struct gfs2_blkreserv *rs, bool nowrap); /** @@ -81,32 +94,36 @@ static inline void gfs2_setbit(const struct gfs2_rbm *rbm, bool do_clone, unsigned char new_state) { unsigned char *byte1, *byte2, *end, cur_state; - unsigned int buflen = rbm->bi->bi_len; + struct gfs2_bitmap *bi = rbm_bi(rbm); + unsigned int buflen = bi->bi_bytes; const unsigned int bit = (rbm->offset % GFS2_NBBY) * GFS2_BIT_SIZE; - byte1 = rbm->bi->bi_bh->b_data + rbm->bi->bi_offset + (rbm->offset / GFS2_NBBY); - end = rbm->bi->bi_bh->b_data + rbm->bi->bi_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%x old_state=%d, " - "new_state=%d\n", rbm->offset, cur_state, new_state); - printk(KERN_WARNING "GFS2: rgrp=0x%llx bi_start=0x%x\n", - (unsigned long long)rbm->rgd->rd_addr, - rbm->bi->bi_start); - printk(KERN_WARNING "GFS2: bi_offset=0x%x bi_len=0x%x\n", - rbm->bi->bi_offset, rbm->bi->bi_len); + struct gfs2_sbd *sdp = rbm->rgd->rd_sbd; + + fs_warn(sdp, "buf_blk = 0x%x old_state=%d, new_state=%d\n", + rbm->offset, cur_state, new_state); + fs_warn(sdp, "rgrp=0x%llx bi_start=0x%x biblk: 0x%llx\n", + (unsigned long long)rbm->rgd->rd_addr, bi->bi_start, + (unsigned long long)bi->bi_bh->b_blocknr); + fs_warn(sdp, "bi_offset=0x%x bi_bytes=0x%x block=0x%llx\n", + bi->bi_offset, bi->bi_bytes, + (unsigned long long)gfs2_rbm_to_block(rbm)); dump_stack(); gfs2_consist_rgrpd(rbm->rgd); return; } *byte1 ^= (cur_state ^ new_state) << bit; - if (do_clone && rbm->bi->bi_clone) { - byte2 = rbm->bi->bi_clone + rbm->bi->bi_offset + (rbm->offset / 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; } @@ -115,16 +132,26 @@ static inline void gfs2_setbit(const struct gfs2_rbm *rbm, bool do_clone, /** * gfs2_testbit - test a bit in the bitmaps * @rbm: The bit to test + * @use_clone: If true, test the clone bitmap, not the official bitmap. + * + * Some callers like gfs2_unaligned_extlen need to test the clone bitmaps, + * not the "real" bitmaps, to avoid allocating recently freed blocks. * * Returns: The two bit block state of the requested bit */ -static inline u8 gfs2_testbit(const struct gfs2_rbm *rbm) +static inline u8 gfs2_testbit(const struct gfs2_rbm *rbm, bool use_clone) { - const u8 *buffer = rbm->bi->bi_bh->b_data + rbm->bi->bi_offset; + struct gfs2_bitmap *bi = rbm_bi(rbm); + const u8 *buffer; const u8 *byte; unsigned int bit; + if (use_clone && bi->bi_clone) + buffer = bi->bi_clone; + else + buffer = bi->bi_bh->b_data; + buffer += bi->bi_offset; byte = buffer + (rbm->offset / GFS2_NBBY); bit = (rbm->offset % GFS2_NBBY) * GFS2_BIT_SIZE; @@ -132,13 +159,13 @@ static inline u8 gfs2_testbit(const struct gfs2_rbm *rbm) } /** - * gfs2_bit_search + * gfs2_bit_search - search bitmap for a state * @ptr: Pointer to bitmap data * @mask: Mask to use (normally 0x55555.... but adjusted for search start) * @state: The state we are searching for * - * We xor the bitmap data with a patter which is the bitwise opposite - * of what we are looking for, this gives rise to a pattern of ones + * We xor the bitmap data with a pattern which is the bitwise opposite + * of what we are looking for. This gives rise to a pattern of ones * wherever there is a match. Since we have two bits per entry, we * take this pattern, shift it down by one place and then and it with * the original. All the even bit positions (0,2,4, etc) then represent @@ -166,7 +193,7 @@ 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 + * @start: start of the new reservation * @len: number of blocks in the new reservation * @rs: existing reservation to compare against * @@ -174,13 +201,11 @@ static inline u64 gfs2_bit_search(const __le64 *ptr, u64 mask, u8 state) * -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) +static inline int rs_cmp(u64 start, u32 len, struct gfs2_blkreserv *rs) { - u64 startblk = gfs2_rbm_to_block(&rs->rs_rbm); - - if (blk >= startblk + rs->rs_free) + if (start >= rs->rs_start + rs->rs_requested) return 1; - if (blk + len - 1 < startblk) + if (rs->rs_start >= start + len) return -1; return 0; } @@ -251,53 +276,80 @@ static u32 gfs2_bitfit(const u8 *buf, const unsigned int len, static int gfs2_rbm_from_block(struct gfs2_rbm *rbm, u64 block) { - u64 rblock = block - rbm->rgd->rd_data0; - u32 x; - - if (WARN_ON_ONCE(rblock > UINT_MAX)) - return -EINVAL; - if (block >= rbm->rgd->rd_data0 + rbm->rgd->rd_data) + if (!rgrp_contains_block(rbm->rgd, block)) return -E2BIG; - - rbm->bi = rbm->rgd->rd_bits; - rbm->offset = (u32)(rblock); + rbm->bii = 0; + rbm->offset = block - rbm->rgd->rd_data0; /* Check if the block is within the first block */ - if (rbm->offset < (rbm->bi->bi_start + rbm->bi->bi_len) * GFS2_NBBY) + 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; - x = rbm->offset / rbm->rgd->rd_sbd->sd_blocks_per_bitmap; - rbm->offset -= x * rbm->rgd->rd_sbd->sd_blocks_per_bitmap; - rbm->bi += x; + 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_add - add a number of blocks to an rbm + * @rbm: The rbm with rgd already set correctly + * @blocks: The number of blocks to add to rpm + * + * This function takes an existing rbm structure and adds a number of blocks to + * it. + * + * Returns: True if the new rbm would point past the end of the rgrp. + */ + +static bool gfs2_rbm_add(struct gfs2_rbm *rbm, u32 blocks) +{ + struct gfs2_rgrpd *rgd = rbm->rgd; + struct gfs2_bitmap *bi = rgd->rd_bits + rbm->bii; + + if (rbm->offset + blocks < bi->bi_blocks) { + rbm->offset += blocks; + return false; + } + blocks -= bi->bi_blocks - rbm->offset; + + for(;;) { + bi++; + if (bi == rgd->rd_bits + rgd->rd_length) + return true; + if (blocks < bi->bi_blocks) { + rbm->offset = blocks; + rbm->bii = bi - rgd->rd_bits; + return false; + } + blocks -= bi->bi_blocks; + } +} + +/** * 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 + * Returns: true if a non-free block is encountered or the end of the resource + * group is reached. */ static bool gfs2_unaligned_extlen(struct gfs2_rbm *rbm, u32 n_unaligned, u32 *len) { - u64 block; u32 n; u8 res; for (n = 0; n < n_unaligned; n++) { - res = gfs2_testbit(rbm); + res = gfs2_testbit(rbm, true); if (res != GFS2_BLKST_FREE) return true; (*len)--; if (*len == 0) return true; - block = gfs2_rbm_to_block(rbm); - if (gfs2_rbm_from_block(rbm, block + 1)) + if (gfs2_rbm_add(rbm, 1)) return true; } @@ -306,7 +358,7 @@ static bool gfs2_unaligned_extlen(struct gfs2_rbm *rbm, u32 n_unaligned, u32 *le /** * gfs2_free_extlen - Return extent length of free blocks - * @rbm: Starting position + * @rrbm: Starting position * @len: Max length to check * * Starting at the block specified by the rbm, see how many free blocks @@ -328,6 +380,7 @@ static u32 gfs2_free_extlen(const struct gfs2_rbm *rrbm, u32 len) u32 chunk_size; u8 *ptr, *start, *end; u64 block; + struct gfs2_bitmap *bi; if (n_unaligned && gfs2_unaligned_extlen(&rbm, 4 - n_unaligned, &len)) @@ -336,11 +389,12 @@ static u32 gfs2_free_extlen(const struct gfs2_rbm *rrbm, u32 len) n_unaligned = len & 3; /* Start is now byte aligned */ while (len > 3) { - start = rbm.bi->bi_bh->b_data; - if (rbm.bi->bi_clone) - start = rbm.bi->bi_clone; - end = start + rbm.bi->bi_bh->b_size; - start += rbm.bi->bi_offset; + bi = rbm_bi(&rbm); + start = bi->bi_bh->b_data; + if (bi->bi_clone) + start = bi->bi_clone; + start += bi->bi_offset; + end = start + bi->bi_bytes; BUG_ON(rbm.offset & 3); start += (rbm.offset / GFS2_NBBY); bytes = min_t(u32, len / GFS2_NBBY, (end - start)); @@ -425,45 +479,45 @@ void gfs2_rgrp_verify(struct gfs2_rgrpd *rgd) count[x] += gfs2_bitcount(rgd, bi->bi_bh->b_data + bi->bi_offset, - bi->bi_len, x); + bi->bi_bytes, x); } if (count[0] != rgd->rd_free) { - if (gfs2_consist_rgrpd(rgd)) - fs_err(sdp, "free data mismatch: %u != %u\n", - count[0], rgd->rd_free); + gfs2_lm(sdp, "free data mismatch: %u != %u\n", + count[0], rgd->rd_free); + gfs2_consist_rgrpd(rgd); return; } tmp = rgd->rd_data - rgd->rd_free - rgd->rd_dinodes; if (count[1] != tmp) { - if (gfs2_consist_rgrpd(rgd)) - fs_err(sdp, "used data mismatch: %u != %u\n", - count[1], tmp); + gfs2_lm(sdp, "used data mismatch: %u != %u\n", + count[1], tmp); + gfs2_consist_rgrpd(rgd); return; } if (count[2] + count[3] != rgd->rd_dinodes) { - if (gfs2_consist_rgrpd(rgd)) - fs_err(sdp, "used metadata mismatch: %u != %u\n", - count[2] + count[3], rgd->rd_dinodes); + gfs2_lm(sdp, "used metadata mismatch: %u != %u\n", + count[2] + count[3], rgd->rd_dinodes); + gfs2_consist_rgrpd(rgd); return; } } -static inline int rgrp_contains_block(struct gfs2_rgrpd *rgd, u64 block) -{ - u64 first = rgd->rd_data0; - u64 last = first + rgd->rd_data; - return first <= block && block < last; -} - /** * gfs2_blk2rgrpd - Find resource group for a given data/meta block number * @sdp: The GFS2 superblock * @blk: The data block number * @exact: True if this needs to be an exact match * + * The @exact argument should be set to true by most callers. The exception + * is when we need to match blocks which are not represented by the rgrp + * bitmap, but which are part of the rgrp (i.e. padding blocks) which are + * there for alignment purposes. Another way of looking at it is that @exact + * matches only valid data/metadata blocks, but with @exact false, it will + * match any block within the extent of the rgrp. + * * Returns: The resource group, or NULL if not found */ @@ -544,6 +598,13 @@ struct gfs2_rgrpd *gfs2_rgrpd_get_next(struct gfs2_rgrpd *rgd) return rgd; } +void check_and_update_goal(struct gfs2_inode *ip) +{ + struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); + if (!ip->i_goal || gfs2_blk2rgrpd(sdp, ip->i_goal, 1) == NULL) + ip->i_goal = ip->i_no_addr; +} + void gfs2_free_clones(struct gfs2_rgrpd *rgd) { int x; @@ -555,36 +616,16 @@ 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) +static void dump_rs(struct seq_file *seq, const struct gfs2_blkreserv *rs, + const char *fs_id_buf) { - int error = 0; + struct gfs2_inode *ip = container_of(rs, struct gfs2_inode, i_res); - 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); + gfs2_print_dbg(seq, "%s B: n:%llu s:%llu f:%u\n", + fs_id_buf, + (unsigned long long)ip->i_no_addr, + (unsigned long long)rs->rs_start, + rs->rs_requested); } /** @@ -599,18 +640,22 @@ static void __rs_deltree(struct gfs2_blkreserv *rs) if (!gfs2_rs_active(rs)) return; - rgd = rs->rs_rbm.rgd; + rgd = rs->rs_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) { - /* 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; - rs->rs_free = 0; - clear_bit(GBF_FULL, &rs->rs_rbm.bi->bi_flags); - smp_mb__after_clear_bit(); + if (rs->rs_requested) { + /* return requested blocks to the rgrp */ + BUG_ON(rs->rs_rgd->rd_requested < rs->rs_requested); + rs->rs_rgd->rd_requested -= rs->rs_requested; + + /* 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_requested; + rs->rs_requested = 0; } } @@ -623,10 +668,11 @@ void gfs2_rs_deltree(struct gfs2_blkreserv *rs) { struct gfs2_rgrpd *rgd; - rgd = rs->rs_rbm.rgd; + rgd = rs->rs_rgd; if (rgd) { spin_lock(&rgd->rd_rsspin); __rs_deltree(rs); + BUG_ON(rs->rs_requested); spin_unlock(&rgd->rd_rsspin); } } @@ -641,12 +687,8 @@ void gfs2_rs_delete(struct gfs2_inode *ip) struct inode *inode = &ip->i_inode; down_write(&ip->i_rw_mutex); - if (ip->i_res && atomic_read(&inode->i_writecount) <= 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; - } + if (atomic_read(&inode->i_writecount) <= 1) + gfs2_rs_deltree(&ip->i_res); up_write(&ip->i_rw_mutex); } @@ -684,31 +726,25 @@ void gfs2_clear_rgrpd(struct gfs2_sbd *sdp) rb_erase(n, &sdp->sd_rindex_tree); if (gl) { - spin_lock(&gl->gl_spin); - gl->gl_object = NULL; - spin_unlock(&gl->gl_spin); - gfs2_glock_add_to_lru(gl); + if (gl->gl_state != LM_ST_UNLOCKED) { + gfs2_glock_cb(gl, LM_ST_UNLOCKED); + flush_delayed_work(&gl->gl_work); + } + gfs2_rgrp_brelse(rgd); + glock_clear_object(gl, rgd); gfs2_glock_put(gl); } gfs2_free_clones(rgd); - kfree(rgd->rd_bits); return_all_reservations(rgd); + kfree(rgd->rd_bits); + rgd->rd_bits = NULL; 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); -} - /** - * gfs2_compute_bitstructs - Compute the bitmap sizes + * compute_bitstructs - Compute the bitmap sizes * @rgd: The resource group descriptor * * Calculates bitmap descriptors, one for each block that contains bitmap data @@ -742,26 +778,30 @@ static int compute_bitstructs(struct gfs2_rgrpd *rgd) bytes = bytes_left; bi->bi_offset = sizeof(struct gfs2_rgrp); bi->bi_start = 0; - bi->bi_len = bytes; + bi->bi_bytes = 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_bytes = 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_bytes = bytes; + bi->bi_blocks = bytes * GFS2_NBBY; /* other blocks */ } else { bytes = sdp->sd_sb.sb_bsize - sizeof(struct gfs2_meta_header); bi->bi_offset = sizeof(struct gfs2_meta_header); bi->bi_start = rgd->rd_bitbytes - bytes_left; - bi->bi_len = bytes; + bi->bi_bytes = bytes; + bi->bi_blocks = bytes * GFS2_NBBY; } bytes_left -= bytes; @@ -772,12 +812,21 @@ static int compute_bitstructs(struct gfs2_rgrpd *rgd) return -EIO; } bi = rgd->rd_bits + (length - 1); - if ((bi->bi_start + bi->bi_len) * GFS2_NBBY != rgd->rd_data) { - if (gfs2_consist_rgrpd(rgd)) { - gfs2_rindex_print(rgd); - fs_err(sdp, "start=%u len=%u offset=%u\n", - bi->bi_start, bi->bi_len, bi->bi_offset); - } + if ((bi->bi_start + bi->bi_bytes) * GFS2_NBBY != rgd->rd_data) { + gfs2_lm(sdp, + "ri_addr=%llu " + "ri_length=%u " + "ri_data0=%llu " + "ri_data=%u " + "ri_bitbytes=%u " + "start=%u len=%u offset=%u\n", + (unsigned long long)rgd->rd_addr, + rgd->rd_length, + (unsigned long long)rgd->rd_data0, + rgd->rd_data, + rgd->rd_bitbytes, + bi->bi_start, bi->bi_bytes, bi->bi_offset); + gfs2_consist_rgrpd(rgd); return -EIO; } @@ -872,37 +921,71 @@ static int read_rindex_entry(struct gfs2_inode *ip) 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) - goto fail; + mutex_init(&rgd->rd_mutex); error = gfs2_glock_get(sdp, rgd->rd_addr, &gfs2_rgrp_glops, CREATE, &rgd->rd_gl); if (error) goto fail; - rgd->rd_gl->gl_object = rgd; + error = compute_bitstructs(rgd); + if (error) + goto fail_glock; + rgd->rd_rgl = (struct gfs2_rgrp_lvb *)rgd->rd_gl->gl_lksb.sb_lvbptr; - rgd->rd_flags &= ~GFS2_RDF_UPTODATE; + rgd->rd_flags &= ~GFS2_RDF_PREFERRED; if (rgd->rd_data > sdp->sd_max_rg_data) sdp->sd_max_rg_data = rgd->rd_data; spin_lock(&sdp->sd_rindex_spin); error = rgd_insert(rgd); spin_unlock(&sdp->sd_rindex_spin); - if (!error) + if (!error) { + glock_set_object(rgd->rd_gl, rgd); return 0; + } error = 0; /* someone else read in the rgrp; free it and ignore it */ +fail_glock: gfs2_glock_put(rgd->rd_gl); fail: kfree(rgd->rd_bits); + rgd->rd_bits = NULL; kmem_cache_free(gfs2_rgrpd_cachep, rgd); return error; } /** + * set_rgrp_preferences - Run all the rgrps, selecting some we prefer to use + * @sdp: the GFS2 superblock + * + * The purpose of this function is to select a subset of the resource groups + * and mark them as PREFERRED. We do it in such a way that each node prefers + * to use a unique set of rgrps to minimize glock contention. + */ +static void set_rgrp_preferences(struct gfs2_sbd *sdp) +{ + struct gfs2_rgrpd *rgd, *first; + int i; + + /* Skip an initial number of rgrps, based on this node's journal ID. + That should start each node out on its own set. */ + rgd = gfs2_rgrpd_get_first(sdp); + for (i = 0; i < sdp->sd_lockstruct.ls_jid; i++) + rgd = gfs2_rgrpd_get_next(rgd); + first = rgd; + + do { + rgd->rd_flags |= GFS2_RDF_PREFERRED; + for (i = 0; i < sdp->sd_journals; i++) { + rgd = gfs2_rgrpd_get_next(rgd); + if (!rgd || rgd == first) + break; + } + } while (rgd && rgd != first); +} + +/** * gfs2_ri_update - Pull in a new resource index from the disk * @ip: pointer to the rindex inode * @@ -921,6 +1004,12 @@ static int gfs2_ri_update(struct gfs2_inode *ip) if (error < 0) return error; + if (RB_EMPTY_ROOT(&sdp->sd_rindex_tree)) { + fs_err(sdp, "no resource groups found in the file system.\n"); + return -ENOENT; + } + set_rgrp_preferences(sdp); + sdp->sd_rindex_uptodate = 1; return 0; } @@ -979,49 +1068,80 @@ static void gfs2_rgrp_in(struct gfs2_rgrpd *rgd, const void *buf) rgd->rd_free = be32_to_cpu(str->rg_free); rgd->rd_dinodes = be32_to_cpu(str->rg_dinodes); rgd->rd_igeneration = be64_to_cpu(str->rg_igeneration); + /* rd_data0, rd_data and rd_bitbytes already set from rindex */ +} + +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 gfs2_rgrp_out(struct gfs2_rgrpd *rgd, void *buf) { + struct gfs2_rgrpd *next = gfs2_rgrpd_get_next(rgd); struct gfs2_rgrp *str = buf; + u32 crc; str->rg_flags = cpu_to_be32(rgd->rd_flags & ~GFS2_RDF_MASK); str->rg_free = cpu_to_be32(rgd->rd_free); str->rg_dinodes = cpu_to_be32(rgd->rd_dinodes); - str->__pad = cpu_to_be32(0); + if (next == NULL) + str->rg_skip = 0; + else if (next->rd_addr > rgd->rd_addr) + str->rg_skip = cpu_to_be32(next->rd_addr - rgd->rd_addr); str->rg_igeneration = cpu_to_be64(rgd->rd_igeneration); + str->rg_data0 = cpu_to_be64(rgd->rd_data0); + str->rg_data = cpu_to_be32(rgd->rd_data); + str->rg_bitbytes = cpu_to_be32(rgd->rd_bitbytes); + str->rg_crc = 0; + crc = gfs2_disk_hash(buf, sizeof(struct gfs2_rgrp)); + str->rg_crc = cpu_to_be32(crc); + memset(&str->rg_reserved, 0, sizeof(str->rg_reserved)); + gfs2_rgrp_ondisk2lvb(rgd->rd_rgl, buf); } 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; + struct gfs2_sbd *sdp = rgd->rd_sbd; + int valid = 1; - 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); + if (rgl->rl_flags != str->rg_flags) { + fs_warn(sdp, "GFS2: rgd: %llu lvb flag mismatch %u/%u", + (unsigned long long)rgd->rd_addr, + be32_to_cpu(rgl->rl_flags), be32_to_cpu(str->rg_flags)); + valid = 0; + } + if (rgl->rl_free != str->rg_free) { + fs_warn(sdp, "GFS2: rgd: %llu lvb free mismatch %u/%u", + (unsigned long long)rgd->rd_addr, + be32_to_cpu(rgl->rl_free), be32_to_cpu(str->rg_free)); + valid = 0; + } + if (rgl->rl_dinodes != str->rg_dinodes) { + fs_warn(sdp, "GFS2: rgd: %llu lvb dinode mismatch %u/%u", + (unsigned long long)rgd->rd_addr, + be32_to_cpu(rgl->rl_dinodes), + be32_to_cpu(str->rg_dinodes)); + valid = 0; + } + if (rgl->rl_igeneration != str->rg_igeneration) { + fs_warn(sdp, "GFS2: rgd: %llu lvb igen mismatch %llu/%llu", + (unsigned long long)rgd->rd_addr, + (unsigned long long)be64_to_cpu(rgl->rl_igeneration), + (unsigned long long)be64_to_cpu(str->rg_igeneration)); + valid = 0; + } + return valid; } static u32 count_unlinked(struct gfs2_rgrpd *rgd) @@ -1035,8 +1155,8 @@ static u32 count_unlinked(struct gfs2_rgrpd *rgd) 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, + while (goal < bi->bi_blocks) { + goal = gfs2_bitfit(buffer, bi->bi_bytes, goal, GFS2_BLKST_UNLINKED); if (goal == BFITNOENT) break; @@ -1048,21 +1168,38 @@ static u32 count_unlinked(struct gfs2_rgrpd *rgd) return count; } +static void rgrp_set_bitmap_flags(struct gfs2_rgrpd *rgd) +{ + struct gfs2_bitmap *bi; + int x; + + if (rgd->rd_free) { + for (x = 0; x < rgd->rd_length; x++) { + bi = rgd->rd_bits + x; + clear_bit(GBF_FULL, &bi->bi_flags); + } + } else { + for (x = 0; x < rgd->rd_length; x++) { + bi = rgd->rd_bits + x; + set_bit(GBF_FULL, &bi->bi_flags); + } + } +} /** - * gfs2_rgrp_bh_get - Read in a RG's header and bitmaps - * @rgd: the struct gfs2_rgrpd describing the RG to read in + * gfs2_rgrp_go_instantiate - Read in a RG's header and bitmaps + * @gl: the glock representing the rgrpd to read in * * Read in all of a Resource Group's header and bitmap blocks. - * Caller must eventually call gfs2_rgrp_relse() to free the bitmaps. + * Caller must eventually call gfs2_rgrp_brelse() to free the bitmaps. * * Returns: errno */ -int gfs2_rgrp_bh_get(struct gfs2_rgrpd *rgd) +int gfs2_rgrp_go_instantiate(struct gfs2_glock *gl) { + struct gfs2_rgrpd *rgd = gl->gl_object; struct gfs2_sbd *sdp = rgd->rd_sbd; - struct gfs2_glock *gl = rgd->rd_gl; unsigned int length = rgd->rd_length; struct gfs2_bitmap *bi; unsigned int x, y; @@ -1073,7 +1210,7 @@ int gfs2_rgrp_bh_get(struct gfs2_rgrpd *rgd) for (x = 0; x < length; x++) { bi = rgd->rd_bits + x; - error = gfs2_meta_read(gl, rgd->rd_addr + x, 0, &bi->bi_bh); + error = gfs2_meta_read(gl, rgd->rd_addr + x, 0, 0, &bi->bi_bh); if (error) goto fail; } @@ -1090,19 +1227,18 @@ int gfs2_rgrp_bh_get(struct gfs2_rgrpd *rgd) } } - if (!(rgd->rd_flags & GFS2_RDF_UPTODATE)) { - for (x = 0; x < length; x++) - clear_bit(GBF_FULL, &rgd->rd_bits[x].bi_flags); - 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; - } - if (be32_to_cpu(GFS2_MAGIC) != rgd->rd_rgl->rl_magic) { + gfs2_rgrp_in(rgd, (rgd->rd_bits[0].bi_bh)->b_data); + rgrp_set_bitmap_flags(rgd); + rgd->rd_flags |= GFS2_RDF_CHECK; + rgd->rd_free_clone = rgd->rd_free; + GLOCK_BUG_ON(rgd->rd_gl, rgd->rd_reserved); + /* 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) { + } else if (sdp->sd_args.ar_rgrplvb) { if (!gfs2_rgrp_lvb_valid(rgd)){ gfs2_consist_rgrpd(rgd); error = -EIO; @@ -1120,52 +1256,44 @@ fail: bi->bi_bh = NULL; gfs2_assert_warn(sdp, !bi->bi_clone); } - return error; } -int update_rgrp_lvb(struct gfs2_rgrpd *rgd) +static int update_rgrp_lvb(struct gfs2_rgrpd *rgd, struct gfs2_holder *gh) { u32 rl_flags; - if (rgd->rd_flags & GFS2_RDF_UPTODATE) + if (!test_bit(GLF_INSTANTIATE_NEEDED, &gh->gh_gl->gl_flags)) return 0; - if (be32_to_cpu(GFS2_MAGIC) != rgd->rd_rgl->rl_magic) - return gfs2_rgrp_bh_get(rgd); + if (cpu_to_be32(GFS2_MAGIC) != rgd->rd_rgl->rl_magic) + return gfs2_instantiate(gh); 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); + rgd->rd_flags |= (rl_flags | 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); + rgrp_set_bitmap_flags(rgd); rgd->rd_free_clone = rgd->rd_free; + GLOCK_BUG_ON(rgd->rd_gl, rgd->rd_reserved); + /* max out the rgrp allocation failure point */ + rgd->rd_extfail_pt = 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((struct gfs2_rgrpd *)gh->gh_gl->gl_object); -} - /** - * gfs2_rgrp_go_unlock - Release RG bitmaps read in with gfs2_rgrp_bh_get() - * @gh: The glock holder for the resource group + * gfs2_rgrp_brelse - Release RG bitmaps read in with gfs2_rgrp_bh_get() + * @rgd: The resource group * */ -void gfs2_rgrp_go_unlock(struct gfs2_holder *gh) +void gfs2_rgrp_brelse(struct gfs2_rgrpd *rgd) { - struct gfs2_rgrpd *rgd = gh->gh_gl->gl_object; int x, length = rgd->rd_length; for (x = 0; x < length; x++) { @@ -1175,7 +1303,7 @@ void gfs2_rgrp_go_unlock(struct gfs2_holder *gh) bi->bi_bh = NULL; } } - + set_bit(GLF_INSTANTIATE_NEEDED, &rgd->rd_gl->gl_flags); } int gfs2_rgrp_send_discards(struct gfs2_sbd *sdp, u64 offset, @@ -1186,12 +1314,12 @@ int gfs2_rgrp_send_discards(struct gfs2_sbd *sdp, u64 offset, u64 blk; sector_t start = 0; sector_t nr_blks = 0; - int rv; + int rv = -EIO; unsigned int x; u32 trimmed = 0; u8 diff; - for (x = 0; x < bi->bi_len; x++) { + for (x = 0; x < bi->bi_bytes; x++) { const u8 *clone = bi->bi_clone ? bi->bi_clone : bi->bi_bh->b_data; clone += bi->bi_offset; clone += x; @@ -1240,9 +1368,9 @@ start_new_extent: fail: if (sdp->sd_args.ar_discard) - fs_warn(sdp, "error %d on discard request, turning discards off for this filesystem", rv); + fs_warn(sdp, "error %d on discard request, turning discards off for this filesystem\n", rv); sdp->sd_args.ar_discard = 0; - return -EIO; + return rv; } /** @@ -1257,7 +1385,7 @@ int gfs2_fitrim(struct file *filp, void __user *argp) { 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 block_device *bdev = sdp->sd_vfs->s_bdev; struct buffer_head *bh; struct gfs2_rgrpd *rgd; struct gfs2_rgrpd *rgd_end; @@ -1273,7 +1401,10 @@ int gfs2_fitrim(struct file *filp, void __user *argp) if (!capable(CAP_SYS_ADMIN)) return -EPERM; - if (!blk_queue_discard(q)) + if (!test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags)) + return -EROFS; + + if (!bdev_max_discard_sectors(bdev)) return -EOPNOTSUPP; if (copy_from_user(&r, argp, sizeof(r))) @@ -1285,8 +1416,8 @@ int gfs2_fitrim(struct file *filp, void __user *argp) start = r.start >> bs_shift; end = start + (r.len >> bs_shift); - minlen = max_t(u64, r.minlen, - q->limits.discard_granularity) >> bs_shift; + minlen = max_t(u64, r.minlen, sdp->sd_sb.sb_bsize); + minlen = max_t(u64, minlen, bdev_discard_granularity(bdev)) >> bs_shift; if (end <= start || minlen > sdp->sd_max_rg_data) return -EINVAL; @@ -1300,7 +1431,8 @@ int gfs2_fitrim(struct file *filp, void __user *argp) while (1) { - ret = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, &gh); + ret = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, + LM_FLAG_NODE_SCOPE, &gh); if (ret) goto out; @@ -1308,9 +1440,11 @@ int gfs2_fitrim(struct file *filp, void __user *argp) /* Trim each bitmap in the rgrp */ for (x = 0; x < rgd->rd_length; x++) { struct gfs2_bitmap *bi = rgd->rd_bits + x; + rgrp_lock_local(rgd); ret = gfs2_rgrp_send_discards(sdp, rgd->rd_data0, NULL, bi, minlen, &amt); + rgrp_unlock_local(rgd); if (ret) { gfs2_glock_dq_uninit(&gh); goto out; @@ -1322,10 +1456,11 @@ int gfs2_fitrim(struct file *filp, void __user *argp) ret = gfs2_trans_begin(sdp, RES_RG_HDR, 0); if (ret == 0) { bh = rgd->rd_bits[0].bi_bh; + rgrp_lock_local(rgd); 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); + rgrp_unlock_local(rgd); gfs2_trans_end(sdp); } } @@ -1354,9 +1489,8 @@ 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); + struct gfs2_blkreserv *rs = &ip->i_res; + struct gfs2_rgrpd *rgd = rs->rs_rgd; BUG_ON(gfs2_rs_active(rs)); @@ -1367,7 +1501,7 @@ static void rs_insert(struct gfs2_inode *ip) rb_entry(*newn, struct gfs2_blkreserv, rs_node); parent = *newn; - rc = rs_cmp(fsblock, rs->rs_free, cur); + rc = rs_cmp(rs->rs_start, rs->rs_requested, cur); if (rc > 0) newn = &((*newn)->rb_right); else if (rc < 0) @@ -1383,37 +1517,75 @@ static void rs_insert(struct gfs2_inode *ip) rb_insert_color(&rs->rs_node, &rgd->rd_rstree); /* Do our rgrp accounting for the reservation */ - rgd->rd_reserved += rs->rs_free; /* blocks reserved */ + rgd->rd_requested += rs->rs_requested; /* blocks requested */ spin_unlock(&rgd->rd_rsspin); trace_gfs2_rs(rs, TRACE_RS_INSERT); } /** + * rgd_free - return the number of free blocks we can allocate + * @rgd: the resource group + * @rs: The reservation to free + * + * This function returns the number of free blocks for an rgrp. + * That's the clone-free blocks (blocks that are free, not including those + * still being used for unlinked files that haven't been deleted.) + * + * It also subtracts any blocks reserved by someone else, but does not + * include free blocks that are still part of our current reservation, + * because obviously we can (and will) allocate them. + */ +static inline u32 rgd_free(struct gfs2_rgrpd *rgd, struct gfs2_blkreserv *rs) +{ + u32 tot_reserved, tot_free; + + if (WARN_ON_ONCE(rgd->rd_requested < rs->rs_requested)) + return 0; + tot_reserved = rgd->rd_requested - rs->rs_requested; + + if (rgd->rd_free_clone < tot_reserved) + tot_reserved = 0; + + tot_free = rgd->rd_free_clone - tot_reserved; + + return tot_free; +} + +/** * 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 - * @requested: number of blocks required for this allocation + * @ap: the allocation parameters * */ static void rg_mblk_search(struct gfs2_rgrpd *rgd, struct gfs2_inode *ip, - unsigned requested) + const struct gfs2_alloc_parms *ap) { struct gfs2_rbm rbm = { .rgd = rgd, }; u64 goal; - struct gfs2_blkreserv *rs = ip->i_res; + struct gfs2_blkreserv *rs = &ip->i_res; u32 extlen; - u32 free_blocks = rgd->rd_free_clone - rgd->rd_reserved; + u32 free_blocks, blocks_available; int ret; struct inode *inode = &ip->i_inode; + spin_lock(&rgd->rd_rsspin); + free_blocks = rgd_free(rgd, rs); + if (rgd->rd_free_clone < rgd->rd_requested) + free_blocks = 0; + blocks_available = rgd->rd_free_clone - rgd->rd_reserved; + if (rgd == rs->rs_rgd) + blocks_available += rs->rs_reserved; + spin_unlock(&rgd->rd_rsspin); + if (S_ISDIR(inode->i_mode)) extlen = 1; else { - extlen = max_t(u32, atomic_read(&rs->rs_sizehint), requested); - extlen = clamp(extlen, RGRP_RSRV_MINBLKS, free_blocks); + extlen = max_t(u32, atomic_read(&ip->i_sizehint), ap->target); + extlen = clamp(extlen, (u32)RGRP_RSRV_MINBLKS, free_blocks); } - if ((rgd->rd_free_clone < rgd->rd_reserved) || (free_blocks < extlen)) + if (free_blocks < extlen || blocks_available < extlen) return; /* Find bitmap block that contains bits for goal block */ @@ -1425,11 +1597,10 @@ static void rg_mblk_search(struct gfs2_rgrpd *rgd, struct gfs2_inode *ip, if (WARN_ON(gfs2_rbm_from_block(&rbm, goal))) return; - ret = gfs2_rbm_find(&rbm, GFS2_BLKST_FREE, extlen, ip, true); + ret = gfs2_rbm_find(&rbm, GFS2_BLKST_FREE, &extlen, &ip->i_res, true); if (ret == 0) { - rs->rs_rbm = rbm; - rs->rs_free = extlen; - rs->rs_inum = ip->i_no_addr; + rs->rs_start = gfs2_rbm_to_block(&rbm); + rs->rs_requested = extlen; rs_insert(ip); } else { if (goal == rgd->rd_last_alloc + rgd->rd_data0) @@ -1442,7 +1613,7 @@ static void rg_mblk_search(struct gfs2_rgrpd *rgd, struct gfs2_inode *ip, * @rgd: The resource group * @block: The starting block * @length: The required length - * @ip: Ignore any reservations for this inode + * @ignore_rs: Reservation to ignore * * If the block does not appear in any reservation, then return the * block number unchanged. If it does appear in the reservation, then @@ -1452,7 +1623,7 @@ static void rg_mblk_search(struct gfs2_rgrpd *rgd, struct gfs2_inode *ip, static u64 gfs2_next_unreserved_block(struct gfs2_rgrpd *rgd, u64 block, u32 length, - const struct gfs2_inode *ip) + struct gfs2_blkreserv *ignore_rs) { struct gfs2_blkreserv *rs; struct rb_node *n; @@ -1472,8 +1643,8 @@ static u64 gfs2_next_unreserved_block(struct gfs2_rgrpd *rgd, u64 block, } if (n) { - while ((rs_cmp(block, length, rs) == 0) && (ip->i_res != rs)) { - block = gfs2_rbm_to_block(&rs->rs_rbm) + rs->rs_free; + while (rs_cmp(block, length, rs) == 0 && rs != ignore_rs) { + block = rs->rs_start + rs->rs_requested; n = n->rb_right; if (n == NULL) break; @@ -1488,8 +1659,9 @@ static u64 gfs2_next_unreserved_block(struct gfs2_rgrpd *rgd, u64 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 + * @rs: Our own reservation * @minext: The minimum extent length + * @maxext: A pointer to the maximum extent structure * * 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 @@ -1501,22 +1673,21 @@ static u64 gfs2_next_unreserved_block(struct gfs2_rgrpd *rgd, u64 block, */ static int gfs2_reservation_check_and_update(struct gfs2_rbm *rbm, - const struct gfs2_inode *ip, - u32 minext) + struct gfs2_blkreserv *rs, + u32 minext, + struct gfs2_extent *maxext) { u64 block = gfs2_rbm_to_block(rbm); u32 extlen = 1; u64 nblock; - int ret; /* * If we have a minimum extent length, then skip over any extent * which is less than the min extent length in size. */ - if (minext) { + if (minext > 1) { extlen = gfs2_free_extlen(rbm, minext); - nblock = block + extlen; - if (extlen < minext) + if (extlen <= maxext->len) goto fail; } @@ -1524,13 +1695,24 @@ static int gfs2_reservation_check_and_update(struct gfs2_rbm *rbm, * 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) - return 0; + nblock = gfs2_next_unreserved_block(rbm->rgd, block, extlen, rs); + if (nblock == block) { + if (!minext || extlen >= minext) + return 0; + + if (extlen > maxext->len) { + maxext->len = extlen; + maxext->rbm = *rbm; + } + } else { + u64 len = nblock - block; + if (len >= (u64)1 << 32) + return -E2BIG; + extlen = len; + } fail: - ret = gfs2_rbm_from_block(rbm, nblock); - if (ret < 0) - return ret; + if (gfs2_rbm_add(rbm, extlen)) + return -E2BIG; return 1; } @@ -1538,92 +1720,112 @@ fail: * 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: The requested extent length (0 for a single block) - * @ip: If set, check for reservations + * @minext: Pointer to the requested extent length + * This is updated to be the actual reservation size. + * @rs: Our own reservation (NULL to skip checking for reservations) * @nowrap: Stop looking at the end of the rgrp, rather than wrapping * around until we've reached the starting point. * * 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) +static int gfs2_rbm_find(struct gfs2_rbm *rbm, u8 state, u32 *minext, + struct gfs2_blkreserv *rs, bool nowrap) { + bool scan_from_start = rbm->bii == 0 && rbm->offset == 0; struct buffer_head *bh; - struct gfs2_bitmap *initial_bi; - u32 initial_offset; + int last_bii; u32 offset; u8 *buffer; - int index; - int n = 0; - int iters = rbm->rgd->rd_length; + bool wrapped = false; 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. + /* + * Determine the last bitmap to search. If we're not starting at the + * beginning of a bitmap, we need to search that bitmap twice to scan + * the entire resource group. */ - if (rbm->offset != 0) - iters++; + last_bii = rbm->bii - (rbm->offset == 0); while(1) { - if (test_bit(GBF_FULL, &rbm->bi->bi_flags) && + bi = rbm_bi(rbm); + if (test_bit(GBF_FULL, &bi->bi_flags) && (state == GFS2_BLKST_FREE)) goto next_bitmap; - bh = rbm->bi->bi_bh; - buffer = bh->b_data + rbm->bi->bi_offset; + bh = bi->bi_bh; + buffer = bh->b_data + bi->bi_offset; WARN_ON(!buffer_uptodate(bh)); - if (state != GFS2_BLKST_UNLINKED && rbm->bi->bi_clone) - buffer = rbm->bi->bi_clone + rbm->bi->bi_offset; - initial_offset = rbm->offset; - offset = gfs2_bitfit(buffer, rbm->bi->bi_len, rbm->offset, state); - if (offset == BFITNOENT) - goto bitmap_full; + if (state != GFS2_BLKST_UNLINKED && bi->bi_clone) + buffer = bi->bi_clone + bi->bi_offset; + offset = gfs2_bitfit(buffer, bi->bi_bytes, rbm->offset, state); + if (offset == BFITNOENT) { + if (state == GFS2_BLKST_FREE && rbm->offset == 0) + set_bit(GBF_FULL, &bi->bi_flags); + goto next_bitmap; + } rbm->offset = offset; - if (ip == NULL) + if (!rs || !minext) return 0; - initial_bi = rbm->bi; - ret = gfs2_reservation_check_and_update(rbm, ip, minext); + ret = gfs2_reservation_check_and_update(rbm, rs, *minext, + &maxext); if (ret == 0) return 0; - if (ret > 0) { - n += (rbm->bi - initial_bi); + if (ret > 0) goto next_iter; - } if (ret == -E2BIG) { - index = 0; + rbm->bii = 0; rbm->offset = 0; - n += (rbm->bi - initial_bi); 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) - set_bit(GBF_FULL, &rbm->bi->bi_flags); - next_bitmap: /* Find next bitmap in the rgrp */ rbm->offset = 0; - index = rbm->bi - rbm->rgd->rd_bits; - index++; - if (index == rbm->rgd->rd_length) - index = 0; + rbm->bii++; + if (rbm->bii == rbm->rgd->rd_length) + rbm->bii = 0; res_covered_end_of_rgrp: - rbm->bi = &rbm->rgd->rd_bits[index]; - if ((index == 0) && nowrap) - break; - n++; + if (rbm->bii == 0) { + if (wrapped) + break; + wrapped = true; + if (nowrap) + break; + } next_iter: - if (n >= iters) + /* Have we scanned the entire resource group? */ + if (wrapped && rbm->bii > last_bii) break; } + if (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 (wrapped && (scan_from_start || rbm->bii > last_bii) && + *minext < rbm->rgd->rd_extfail_pt) + rbm->rgd->rd_extfail_pt = *minext - 1; + + /* 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; } @@ -1645,12 +1847,11 @@ static void try_rgrp_unlink(struct gfs2_rgrpd *rgd, u64 *last_unlinked, u64 skip struct gfs2_inode *ip; int error; int found = 0; - struct gfs2_rbm rbm = { .rgd = rgd, .bi = rgd->rd_bits, .offset = 0 }; + struct gfs2_rbm rbm = { .rgd = rgd, .bii = 0, .offset = 0 }; while (1) { - down_write(&sdp->sd_log_flush_lock); - error = gfs2_rbm_find(&rbm, GFS2_BLKST_UNLINKED, 0, NULL, true); - up_write(&sdp->sd_log_flush_lock); + error = gfs2_rbm_find(&rbm, GFS2_BLKST_UNLINKED, NULL, NULL, + true); if (error == -ENOSPC) break; if (WARN_ON_ONCE(error)) @@ -1665,7 +1866,7 @@ static void try_rgrp_unlink(struct gfs2_rgrpd *rgd, u64 *last_unlinked, u64 skip continue; *last_unlinked = block; - error = gfs2_glock_get(sdp, block, &gfs2_inode_glops, CREATE, &gl); + error = gfs2_glock_get(sdp, block, &gfs2_iopen_glops, CREATE, &gl); if (error) continue; @@ -1678,7 +1879,7 @@ static void try_rgrp_unlink(struct gfs2_rgrpd *rgd, u64 *last_unlinked, u64 skip */ ip = gl->gl_object; - if (ip || queue_work(gfs2_delete_workqueue, &gl->gl_delete) == 0) + if (ip || !gfs2_queue_verify_delete(gl, false)) gfs2_glock_put(gl); else found++; @@ -1722,17 +1923,26 @@ static void try_rgrp_unlink(struct gfs2_rgrpd *rgd, u64 *last_unlinked, u64 skip static bool gfs2_rgrp_congested(const struct gfs2_rgrpd *rgd, int loops) { const struct gfs2_glock *gl = rgd->rd_gl; - const struct gfs2_sbd *sdp = gl->gl_sbd; + const struct gfs2_sbd *sdp = gl->gl_name.ln_sbd; struct gfs2_lkstats *st; - s64 r_dcount, l_dcount; - s64 r_srttb, l_srttb; + u64 r_dcount, l_dcount; + u64 l_srttb, a_srttb = 0; s64 srttb_diff; - s64 sqr_diff; - s64 var; + u64 sqr_diff; + u64 var; + int cpu, nonzero = 0; preempt_disable(); + for_each_present_cpu(cpu) { + st = &per_cpu_ptr(sdp->sd_lkstats, cpu)->lkstats[LM_TYPE_RGRP]; + if (st->stats[GFS2_LKS_SRTTB]) { + a_srttb += st->stats[GFS2_LKS_SRTTB]; + nonzero++; + } + } st = &this_cpu_ptr(sdp->sd_lkstats)->lkstats[LM_TYPE_RGRP]; - r_srttb = st->stats[GFS2_LKS_SRTTB]; + if (nonzero) + do_div(a_srttb, nonzero); r_dcount = st->stats[GFS2_LKS_DCOUNT]; var = st->stats[GFS2_LKS_SRTTVARB] + gl->gl_stats.stats[GFS2_LKS_SRTTVARB]; @@ -1741,10 +1951,10 @@ static bool gfs2_rgrp_congested(const struct gfs2_rgrpd *rgd, int loops) 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)) + if ((l_dcount < 1) || (r_dcount < 1) || (a_srttb == 0)) return false; - srttb_diff = r_srttb - l_srttb; + srttb_diff = a_srttb - l_srttb; sqr_diff = srttb_diff * srttb_diff; var *= 2; @@ -1757,7 +1967,7 @@ static bool gfs2_rgrp_congested(const struct gfs2_rgrpd *rgd, int loops) } /** - * gfs2_rgrp_used_recently + * gfs2_rgrp_used_recently - test if an rgrp has been used recently * @rs: The block reservation with the rgrp to test * @msecs: The time limit in milliseconds * @@ -1769,7 +1979,7 @@ static bool gfs2_rgrp_used_recently(const struct gfs2_blkreserv *rs, u64 tdiff; tdiff = ktime_to_ns(ktime_sub(ktime_get_real(), - rs->rs_rbm.rgd->rd_gl->gl_dstamp)); + rs->rs_rgd->rd_gl->gl_dstamp)); return tdiff > (msecs * 1000 * 1000); } @@ -1777,10 +1987,8 @@ static bool gfs2_rgrp_used_recently(const struct gfs2_blkreserv *rs, static u32 gfs2_orlov_skip(const struct gfs2_inode *ip) { const struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); - u32 skip; - get_random_bytes(&skip, sizeof(skip)); - return skip % sdp->sd_rgrps; + return get_random_u32() % sdp->sd_rgrps; } static bool gfs2_select_rgrp(struct gfs2_rgrpd **pos, const struct gfs2_rgrpd *begin) @@ -1798,104 +2006,164 @@ static bool gfs2_select_rgrp(struct gfs2_rgrpd **pos, const struct gfs2_rgrpd *b } /** + * fast_to_acquire - determine if a resource group will be fast to acquire + * @rgd: The rgrp + * + * If this is one of our preferred rgrps, it should be quicker to acquire, + * because we tried to set ourselves up as dlm lock master. + */ +static inline int fast_to_acquire(struct gfs2_rgrpd *rgd) +{ + struct gfs2_glock *gl = rgd->rd_gl; + + if (gl->gl_state != LM_ST_UNLOCKED && list_empty(&gl->gl_holders) && + !test_bit(GLF_DEMOTE_IN_PROGRESS, &gl->gl_flags) && + !test_bit(GLF_DEMOTE, &gl->gl_flags)) + return 1; + if (rgd->rd_flags & GFS2_RDF_PREFERRED) + return 1; + return 0; +} + +/** * gfs2_inplace_reserve - Reserve space in the filesystem * @ip: the inode to reserve space for - * @requested: the number of blocks to be reserved + * @ap: the allocation parameters * - * Returns: errno + * We try our best to find an rgrp that has at least ap->target blocks + * available. After a couple of passes (loops == 2), the prospects of finding + * such an rgrp diminish. At this stage, we return the first rgrp that has + * at least ap->min_target blocks available. + * + * Returns: 0 on success, + * -ENOMEM if a suitable rgrp can't be found + * errno otherwise */ -int gfs2_inplace_reserve(struct gfs2_inode *ip, u32 requested, u32 aflags) +int gfs2_inplace_reserve(struct gfs2_inode *ip, struct gfs2_alloc_parms *ap) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); struct gfs2_rgrpd *begin = NULL; - struct gfs2_blkreserv *rs = ip->i_res; - int error = 0, rg_locked, flags = 0; + struct gfs2_blkreserv *rs = &ip->i_res; + int error = 0, flags = LM_FLAG_NODE_SCOPE; + bool rg_locked; u64 last_unlinked = NO_BLOCK; + u32 target = ap->target; int loops = 0; - u32 skip = 0; + u32 free_blocks, blocks_available, skip = 0; + + BUG_ON(rs->rs_reserved); if (sdp->sd_args.ar_rgrplvb) flags |= GL_SKIP; - if (gfs2_assert_warn(sdp, requested)) + if (gfs2_assert_warn(sdp, target)) return -EINVAL; if (gfs2_rs_active(rs)) { - begin = rs->rs_rbm.rgd; - flags = 0; /* Yoda: Do or do not. There is no try */ - } else if (ip->i_rgd && rgrp_contains_block(ip->i_rgd, ip->i_goal)) { - rs->rs_rbm.rgd = begin = ip->i_rgd; + begin = rs->rs_rgd; + } else if (rs->rs_rgd && + rgrp_contains_block(rs->rs_rgd, ip->i_goal)) { + begin = rs->rs_rgd; } else { - rs->rs_rbm.rgd = begin = gfs2_blk2rgrpd(sdp, ip->i_goal, 1); + check_and_update_goal(ip); + rs->rs_rgd = begin = gfs2_blk2rgrpd(sdp, ip->i_goal, 1); } - if (S_ISDIR(ip->i_inode.i_mode) && (aflags & GFS2_AF_ORLOV)) + if (S_ISDIR(ip->i_inode.i_mode) && (ap->aflags & GFS2_AF_ORLOV)) skip = gfs2_orlov_skip(ip); - if (rs->rs_rbm.rgd == NULL) + if (rs->rs_rgd == NULL) return -EBADSLT; while (loops < 3) { - rg_locked = 1; + struct gfs2_rgrpd *rgd; - if (!gfs2_glock_is_locked_by_me(rs->rs_rbm.rgd->rd_gl)) { - rg_locked = 0; + rg_locked = gfs2_glock_is_locked_by_me(rs->rs_rgd->rd_gl); + if (rg_locked) { + rgrp_lock_local(rs->rs_rgd); + } else { 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, + if (!gfs2_rs_active(rs)) { + if (loops == 0 && + !fast_to_acquire(rs->rs_rgd)) + goto next_rgrp; + if ((loops < 2) && + gfs2_rgrp_used_recently(rs, 1000) && + gfs2_rgrp_congested(rs->rs_rgd, loops)) + goto next_rgrp; + } + error = gfs2_glock_nq_init(rs->rs_rgd->rd_gl, LM_ST_EXCLUSIVE, flags, - &rs->rs_rgd_gh); + &ip->i_rgd_gh); if (unlikely(error)) return error; + rgrp_lock_local(rs->rs_rgd); if (!gfs2_rs_active(rs) && (loops < 2) && - gfs2_rgrp_congested(rs->rs_rbm.rgd, loops)) + gfs2_rgrp_congested(rs->rs_rgd, loops)) goto skip_rgrp; if (sdp->sd_args.ar_rgrplvb) { - error = update_rgrp_lvb(rs->rs_rbm.rgd); + error = update_rgrp_lvb(rs->rs_rgd, + &ip->i_rgd_gh); if (unlikely(error)) { - gfs2_glock_dq_uninit(&rs->rs_rgd_gh); + rgrp_unlock_local(rs->rs_rgd); + gfs2_glock_dq_uninit(&ip->i_rgd_gh); return error; } } } - /* Skip unuseable resource groups */ - if (rs->rs_rbm.rgd->rd_flags & (GFS2_RGF_NOALLOC | GFS2_RDF_ERROR)) + /* Skip unusable resource groups */ + if ((rs->rs_rgd->rd_flags & (GFS2_RGF_NOALLOC | + GFS2_RDF_ERROR)) || + (loops == 0 && target > rs->rs_rgd->rd_extfail_pt)) goto skip_rgrp; - if (sdp->sd_args.ar_rgrplvb) - gfs2_rgrp_bh_get(rs->rs_rbm.rgd); + if (sdp->sd_args.ar_rgrplvb) { + error = gfs2_instantiate(&ip->i_rgd_gh); + if (error) + goto skip_rgrp; + } /* Get a reservation if we don't already have one */ if (!gfs2_rs_active(rs)) - rg_mblk_search(rs->rs_rbm.rgd, ip, requested); + rg_mblk_search(rs->rs_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 >= requested) { - ip->i_rgd = rs->rs_rbm.rgd; - return 0; + rgd = rs->rs_rgd; + spin_lock(&rgd->rd_rsspin); + free_blocks = rgd_free(rgd, rs); + blocks_available = rgd->rd_free_clone - rgd->rd_reserved; + if (free_blocks < target || blocks_available < target) { + spin_unlock(&rgd->rd_rsspin); + goto check_rgrp; } - - /* Drop reservation, if we couldn't use reserved rgrp */ - if (gfs2_rs_active(rs)) - gfs2_rs_deltree(rs); + rs->rs_reserved = ap->target; + if (rs->rs_reserved > blocks_available) + rs->rs_reserved = blocks_available; + rgd->rd_reserved += rs->rs_reserved; + spin_unlock(&rgd->rd_rsspin); + rgrp_unlock_local(rs->rs_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, + if (rs->rs_rgd->rd_flags & GFS2_RDF_CHECK) + try_rgrp_unlink(rs->rs_rgd, &last_unlinked, ip->i_no_addr); skip_rgrp: + rgrp_unlock_local(rs->rs_rgd); + + /* 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); + gfs2_glock_dq_uninit(&ip->i_rgd_gh); next_rgrp: /* Find the next rgrp, and continue looking */ - if (gfs2_select_rgrp(&rs->rs_rbm.rgd, begin)) + if (gfs2_select_rgrp(&rs->rs_rgd, begin)) continue; if (skip) continue; @@ -1912,8 +2180,12 @@ next_rgrp: return error; } /* Flushing the log may release space */ - if (loops == 2) - gfs2_log_flush(sdp, NULL); + if (loops == 2) { + if (ap->min_target) + target = ap->min_target; + gfs2_log_flush(sdp, NULL, GFS2_LOG_HEAD_FLUSH_NORMAL | + GFS2_LFC_INPLACE_RESERVE); + } } return -ENOSPC; @@ -1928,32 +2200,21 @@ next_rgrp: void gfs2_inplace_release(struct gfs2_inode *ip) { - struct gfs2_blkreserv *rs = ip->i_res; - - if (rs->rs_rgd_gh.gh_gl) - gfs2_glock_dq_uninit(&rs->rs_rgd_gh); -} - -/** - * gfs2_get_block_type - Check a block in a RG is of given type - * @rgd: the resource group holding the block - * @block: the block number - * - * Returns: The block type (GFS2_BLKST_*) - */ - -static unsigned char gfs2_get_block_type(struct gfs2_rgrpd *rgd, u64 block) -{ - struct gfs2_rbm rbm = { .rgd = rgd, }; - int ret; + struct gfs2_blkreserv *rs = &ip->i_res; - ret = gfs2_rbm_from_block(&rbm, block); - WARN_ON_ONCE(ret != 0); + if (rs->rs_reserved) { + struct gfs2_rgrpd *rgd = rs->rs_rgd; - return gfs2_testbit(&rbm); + spin_lock(&rgd->rd_rsspin); + GLOCK_BUG_ON(rgd->rd_gl, rgd->rd_reserved < rs->rs_reserved); + rgd->rd_reserved -= rs->rs_reserved; + spin_unlock(&rgd->rd_rsspin); + rs->rs_reserved = 0; + } + if (gfs2_holder_initialized(&ip->i_rgd_gh)) + gfs2_glock_dq_uninit(&ip->i_rgd_gh); } - /** * gfs2_alloc_extent - allocate an extent from a given bitmap * @rbm: the resource group information @@ -1973,14 +2234,14 @@ static void gfs2_alloc_extent(const struct gfs2_rbm *rbm, bool dinode, *n = 1; block = gfs2_rbm_to_block(rbm); - gfs2_trans_add_meta(rbm->rgd->rd_gl, rbm->bi->bi_bh); + 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) { ret = gfs2_rbm_from_block(&pos, block); - if (ret || gfs2_testbit(&pos) != GFS2_BLKST_FREE) + if (ret || gfs2_testbit(&pos, true) != GFS2_BLKST_FREE) break; - gfs2_trans_add_meta(pos.rgd->rd_gl, pos.bi->bi_bh); + gfs2_trans_add_meta(pos.rgd->rd_gl, rbm_bi(&pos)->bi_bh); gfs2_setbit(&pos, true, GFS2_BLKST_USED); (*n)++; block++; @@ -1990,77 +2251,84 @@ static void gfs2_alloc_extent(const struct gfs2_rbm *rbm, bool dinode, /** * rgblk_free - Change alloc state of given block(s) * @sdp: the filesystem + * @rgd: the resource group the blocks are in * @bstart: the start of a run of blocks to free * @blen: the length of the block run (all must lie within ONE RG!) * @new_state: GFS2_BLKST_XXX the after-allocation block state - * - * Returns: Resource group containing the block(s) */ -static struct gfs2_rgrpd *rgblk_free(struct gfs2_sbd *sdp, u64 bstart, - u32 blen, unsigned char new_state) +static void rgblk_free(struct gfs2_sbd *sdp, struct gfs2_rgrpd *rgd, + u64 bstart, u32 blen, unsigned char new_state) { struct gfs2_rbm rbm; + struct gfs2_bitmap *bi, *bi_prev = NULL; - 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; - } - + rbm.rgd = rgd; + if (WARN_ON_ONCE(gfs2_rbm_from_block(&rbm, bstart))) + return; while (blen--) { - gfs2_rbm_from_block(&rbm, bstart); - bstart++; - if (!rbm.bi->bi_clone) { - rbm.bi->bi_clone = kmalloc(rbm.bi->bi_bh->b_size, - GFP_NOFS | __GFP_NOFAIL); - memcpy(rbm.bi->bi_clone + rbm.bi->bi_offset, - rbm.bi->bi_bh->b_data + rbm.bi->bi_offset, - rbm.bi->bi_len); + bi = rbm_bi(&rbm); + if (bi != bi_prev) { + 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_bytes); + } + gfs2_trans_add_meta(rbm.rgd->rd_gl, bi->bi_bh); + bi_prev = bi; } - gfs2_trans_add_meta(rbm.rgd->rd_gl, rbm.bi->bi_bh); gfs2_setbit(&rbm, false, new_state); + gfs2_rbm_add(&rbm, 1); } - - return rbm.rgd; } /** * gfs2_rgrp_dump - print out an rgrp * @seq: The iterator - * @gl: The glock in question + * @rgd: The rgrp in question + * @fs_id_buf: pointer to file system id (if requested) * */ -int gfs2_rgrp_dump(struct seq_file *seq, const struct gfs2_glock *gl) +void gfs2_rgrp_dump(struct seq_file *seq, struct gfs2_rgrpd *rgd, + const char *fs_id_buf) { - 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 r:%u\n", + spin_lock(&rgd->rd_rsspin); + gfs2_print_dbg(seq, "%s R: n:%llu f:%02x b:%u/%u i:%u q:%u r:%u e:%u\n", + fs_id_buf, (unsigned long long)rgd->rd_addr, rgd->rd_flags, rgd->rd_free, rgd->rd_free_clone, rgd->rd_dinodes, - rgd->rd_reserved); - spin_lock(&rgd->rd_rsspin); + rgd->rd_requested, rgd->rd_reserved, rgd->rd_extfail_pt); + if (rgd->rd_sbd->sd_args.ar_rgrplvb && rgd->rd_rgl) { + struct gfs2_rgrp_lvb *rgl = rgd->rd_rgl; + + gfs2_print_dbg(seq, "%s L: f:%02x b:%u i:%u\n", fs_id_buf, + be32_to_cpu(rgl->rl_flags), + be32_to_cpu(rgl->rl_free), + be32_to_cpu(rgl->rl_dinodes)); + } 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); + dump_rs(seq, trs, fs_id_buf); } spin_unlock(&rgd->rd_rsspin); - return 0; } static void gfs2_rgrp_error(struct gfs2_rgrpd *rgd) { struct gfs2_sbd *sdp = rgd->rd_sbd; + char fs_id_buf[sizeof(sdp->sd_fsname) + 7]; + fs_warn(sdp, "rgrp %llu has an error, marking it readonly until umount\n", (unsigned long long)rgd->rd_addr); fs_warn(sdp, "umount on all nodes and run fsck.gfs2 to fix the error\n"); - gfs2_rgrp_dump(NULL, rgd->rd_gl); + sprintf(fs_id_buf, "fsid=%s: ", sdp->sd_fsname); + gfs2_rgrp_dump(NULL, rgd, fs_id_buf); rgd->rd_flags |= GFS2_RDF_ERROR; } @@ -2078,28 +2346,61 @@ static void gfs2_rgrp_error(struct gfs2_rgrpd *rgd) 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_blkreserv *rs = &ip->i_res; struct gfs2_rgrpd *rgd = rbm->rgd; - unsigned rlen; - u64 block; - int ret; - spin_lock(&rgd->rd_rsspin); + BUG_ON(rs->rs_reserved < len); + rs->rs_reserved -= len; 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; + u64 start = gfs2_rbm_to_block(rbm); + + if (rs->rs_start == start) { + unsigned int rlen; + + rs->rs_start += len; + rlen = min(rs->rs_requested, len); + rs->rs_requested -= rlen; + rgd->rd_requested -= rlen; trace_gfs2_rs(rs, TRACE_RS_CLAIM); - if (rs->rs_free && !ret) - goto out; + if (rs->rs_start < rgd->rd_data0 + rgd->rd_data && + rs->rs_requested) + return; + /* We used up our block reservation, so we should + reserve more blocks next time. */ + atomic_add(RGRP_RSRV_ADDBLKS, &ip->i_sizehint); } __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)) { + goal = ip->i_res.rs_start; + } else { + 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; + } + if (WARN_ON_ONCE(gfs2_rbm_from_block(rbm, goal))) { + rbm->bii = 0; + rbm->offset = 0; + } } /** @@ -2108,56 +2409,46 @@ out: * @bn: Used to return the starting block number * @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 * * Returns: 0 or error */ int gfs2_alloc_blocks(struct gfs2_inode *ip, u64 *bn, unsigned int *nblocks, - bool dinode, u64 *generation) + bool dinode) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); struct buffer_head *dibh; - struct gfs2_rbm rbm = { .rgd = ip->i_rgd, }; - unsigned int ndata; - u64 goal; + struct gfs2_rbm rbm = { .rgd = ip->i_res.rs_rgd, }; u64 block; /* block, within the file system scope */ - int error; - - if (gfs2_rs_active(ip->i_res)) - goal = gfs2_rbm_to_block(&ip->i_res->rs_rbm); - else 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; + u32 minext = 1; + int error = -ENOSPC; - gfs2_rbm_from_block(&rbm, goal); - error = gfs2_rbm_find(&rbm, GFS2_BLKST_FREE, 0, ip, false); + BUG_ON(ip->i_res.rs_reserved < *nblocks); + rgrp_lock_local(rbm.rgd); + if (gfs2_rs_active(&ip->i_res)) { + gfs2_set_alloc_start(&rbm, ip, dinode); + error = gfs2_rbm_find(&rbm, GFS2_BLKST_FREE, &minext, &ip->i_res, false); + } if (error == -ENOSPC) { - gfs2_rbm_from_block(&rbm, goal); - error = gfs2_rbm_find(&rbm, GFS2_BLKST_FREE, 0, NULL, false); + gfs2_set_alloc_start(&rbm, ip, dinode); + error = gfs2_rbm_find(&rbm, GFS2_BLKST_FREE, &minext, NULL, false); } /* Since all blocks are reserved in advance, this shouldn't happen */ if (error) { - fs_warn(sdp, "inum=%llu error=%d, nblocks=%u, full=%d\n", + 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)); + test_bit(GBF_FULL, &rbm.rgd->rd_bits->bi_flags), + rbm.rgd->rd_extfail_pt); goto rgrp_error; } 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--; - if (!dinode) { - ip->i_goal = block + ndata - 1; + ip->i_goal = block + *nblocks - 1; error = gfs2_meta_inode_buffer(ip, &dibh); if (error == 0) { struct gfs2_dinode *di = @@ -2168,36 +2459,47 @@ int gfs2_alloc_blocks(struct gfs2_inode *ip, u64 *bn, unsigned int *nblocks, brelse(dibh); } } - if (rbm.rgd->rd_free < *nblocks) { - printk(KERN_WARNING "nblocks=%u\n", *nblocks); + spin_lock(&rbm.rgd->rd_rsspin); + gfs2_adjust_reservation(ip, &rbm, *nblocks); + if (rbm.rgd->rd_free < *nblocks || rbm.rgd->rd_reserved < *nblocks) { + fs_warn(sdp, "nblocks=%u\n", *nblocks); + spin_unlock(&rbm.rgd->rd_rsspin); goto rgrp_error; } - + GLOCK_BUG_ON(rbm.rgd->rd_gl, rbm.rgd->rd_reserved < *nblocks); + GLOCK_BUG_ON(rbm.rgd->rd_gl, rbm.rgd->rd_free_clone < *nblocks); + GLOCK_BUG_ON(rbm.rgd->rd_gl, rbm.rgd->rd_free < *nblocks); + rbm.rgd->rd_reserved -= *nblocks; + rbm.rgd->rd_free_clone -= *nblocks; rbm.rgd->rd_free -= *nblocks; + spin_unlock(&rbm.rgd->rd_rsspin); if (dinode) { + u64 generation; + rbm.rgd->rd_dinodes++; - *generation = rbm.rgd->rd_igeneration++; - if (*generation == 0) - *generation = rbm.rgd->rd_igeneration++; + generation = rbm.rgd->rd_igeneration++; + if (generation == 0) + generation = rbm.rgd->rd_igeneration++; + ip->i_generation = generation; } 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); + rgrp_unlock_local(rbm.rgd); gfs2_statfs_change(sdp, 0, -(s64)*nblocks, dinode ? 1 : 0); if (dinode) - gfs2_trans_add_unrevoke(sdp, block, 1); + gfs2_trans_remove_revoke(sdp, block, *nblocks); gfs2_quota_change(ip, *nblocks, ip->i_inode.i_uid, ip->i_inode.i_gid); - 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: + rgrp_unlock_local(rbm.rgd); gfs2_rgrp_error(rbm.rgd); return -EIO; } @@ -2205,45 +2507,47 @@ rgrp_error: /** * __gfs2_free_blocks - free a contiguous run of block(s) * @ip: the inode these blocks are being freed from + * @rgd: the resource group the blocks are in * @bstart: first block of a run of contiguous blocks * @blen: the length of the block run * @meta: 1 if the blocks represent metadata * */ -void __gfs2_free_blocks(struct gfs2_inode *ip, u64 bstart, u32 blen, int meta) +void __gfs2_free_blocks(struct gfs2_inode *ip, struct gfs2_rgrpd *rgd, + u64 bstart, u32 blen, int meta) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); - struct gfs2_rgrpd *rgd; - rgd = rgblk_free(sdp, bstart, blen, GFS2_BLKST_FREE); - if (!rgd) - return; + rgrp_lock_local(rgd); + rgblk_free(sdp, rgd, bstart, blen, GFS2_BLKST_FREE); trace_gfs2_block_alloc(ip, rgd, bstart, blen, GFS2_BLKST_FREE); rgd->rd_free += blen; 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); + rgrp_unlock_local(rgd); /* Directories keep their data in the metadata address space */ - if (meta || ip->i_depth) - gfs2_meta_wipe(ip, bstart, blen); + if (meta || ip->i_depth || gfs2_is_jdata(ip)) + gfs2_journal_wipe(ip, bstart, blen); } /** * gfs2_free_meta - free a contiguous run of data block(s) * @ip: the inode these blocks are being freed from + * @rgd: the resource group the blocks are in * @bstart: first block of a run of contiguous blocks * @blen: the length of the block run * */ -void gfs2_free_meta(struct gfs2_inode *ip, u64 bstart, u32 blen) +void gfs2_free_meta(struct gfs2_inode *ip, struct gfs2_rgrpd *rgd, + u64 bstart, u32 blen) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); - __gfs2_free_blocks(ip, bstart, blen, 1); + __gfs2_free_blocks(ip, rgd, bstart, blen, 1); gfs2_statfs_change(sdp, 0, +blen, 0); gfs2_quota_change(ip, -(s64)blen, ip->i_inode.i_uid, ip->i_inode.i_gid); } @@ -2255,26 +2559,24 @@ void gfs2_unlink_di(struct inode *inode) struct gfs2_rgrpd *rgd; u64 blkno = ip->i_no_addr; - rgd = rgblk_free(sdp, blkno, 1, GFS2_BLKST_UNLINKED); + rgd = gfs2_blk2rgrpd(sdp, blkno, true); if (!rgd) return; + rgrp_lock_local(rgd); + rgblk_free(sdp, rgd, blkno, 1, GFS2_BLKST_UNLINKED); 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); + be32_add_cpu(&rgd->rd_rgl->rl_unlinked, 1); + rgrp_unlock_local(rgd); } -static void gfs2_free_uninit_di(struct gfs2_rgrpd *rgd, u64 blkno) +void gfs2_free_di(struct gfs2_rgrpd *rgd, struct gfs2_inode *ip) { struct gfs2_sbd *sdp = rgd->rd_sbd; - struct gfs2_rgrpd *tmp_rgd; - - tmp_rgd = rgblk_free(sdp, blkno, 1, GFS2_BLKST_FREE); - if (!tmp_rgd) - return; - gfs2_assert_withdraw(sdp, rgd == tmp_rgd); + rgrp_lock_local(rgd); + rgblk_free(sdp, rgd, ip->i_no_addr, 1, GFS2_BLKST_FREE); if (!rgd->rd_dinodes) gfs2_consist_rgrpd(rgd); rgd->rd_dinodes--; @@ -2282,19 +2584,13 @@ static void gfs2_free_uninit_di(struct gfs2_rgrpd *rgd, u64 blkno) 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); + be32_add_cpu(&rgd->rd_rgl->rl_unlinked, -1); + rgrp_unlock_local(rgd); gfs2_statfs_change(sdp, 0, +1, -1); -} - - -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, 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); + gfs2_journal_wipe(ip, ip->i_no_addr, 1); } /** @@ -2303,6 +2599,10 @@ void gfs2_free_di(struct gfs2_rgrpd *rgd, struct gfs2_inode *ip) * @no_addr: The block number to check * @type: The block type we are looking for * + * The inode glock of @no_addr must be held. The @type to check for is either + * GFS2_BLKST_DINODE or GFS2_BLKST_UNLINKED; checking for type GFS2_BLKST_FREE + * or GFS2_BLKST_USED would make no sense. + * * Returns: 0 if the block type matches the expected type * -ESTALE if it doesn't match * or -ve errno if something went wrong while checking @@ -2312,6 +2612,7 @@ int gfs2_check_blk_type(struct gfs2_sbd *sdp, u64 no_addr, unsigned int type) { struct gfs2_rgrpd *rgd; struct gfs2_holder rgd_gh; + struct gfs2_rbm rbm; int error = -EINVAL; rgd = gfs2_blk2rgrpd(sdp, no_addr, 1); @@ -2322,10 +2623,22 @@ int gfs2_check_blk_type(struct gfs2_sbd *sdp, u64 no_addr, unsigned int type) if (error) goto fail; - if (gfs2_get_block_type(rgd, no_addr) != type) - error = -ESTALE; + rbm.rgd = rgd; + error = gfs2_rbm_from_block(&rbm, no_addr); + if (!WARN_ON_ONCE(error)) { + /* + * No need to take the local resource group lock here; the + * inode glock of @no_addr provides the necessary + * synchronization in case the block is an inode. (In case + * the block is not an inode, the block type will not match + * the @type we are looking for.) + */ + if (gfs2_testbit(&rbm, false) != type) + error = -ESTALE; + } gfs2_glock_dq_uninit(&rgd_gh); + fail: return error; } @@ -2354,19 +2667,34 @@ void gfs2_rlist_add(struct gfs2_inode *ip, struct gfs2_rgrp_list *rlist, if (gfs2_assert_warn(sdp, !rlist->rl_ghs)) return; - if (ip->i_rgd && rgrp_contains_block(ip->i_rgd, block)) - rgd = ip->i_rgd; - else + /* + * The resource group last accessed is kept in the last position. + */ + + if (rlist->rl_rgrps) { + rgd = rlist->rl_rgd[rlist->rl_rgrps - 1]; + if (rgrp_contains_block(rgd, block)) + return; rgd = gfs2_blk2rgrpd(sdp, block, 1); + } else { + rgd = ip->i_res.rs_rgd; + if (!rgd || !rgrp_contains_block(rgd, block)) + rgd = gfs2_blk2rgrpd(sdp, block, 1); + } + if (!rgd) { - fs_err(sdp, "rlist_add: no rgrp for block %llu\n", (unsigned long long)block); + fs_err(sdp, "rlist_add: no rgrp for block %llu\n", + (unsigned long long)block); return; } - ip->i_rgd = rgd; - for (x = 0; x < rlist->rl_rgrps; x++) - if (rlist->rl_rgd[x] == rgd) + for (x = 0; x < rlist->rl_rgrps; x++) { + if (rlist->rl_rgd[x] == rgd) { + swap(rlist->rl_rgd[x], + rlist->rl_rgd[rlist->rl_rgrps - 1]); return; + } + } if (rlist->rl_rgrps == rlist->rl_space) { new_space = rlist->rl_space + 10; @@ -2391,27 +2719,29 @@ void gfs2_rlist_add(struct gfs2_inode *ip, struct gfs2_rgrp_list *rlist, * gfs2_rlist_alloc - all RGs have been added to the rlist, now allocate * 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 + * @state: the state we're requesting + * @flags: the modifier flags * * FIXME: Don't use NOFAIL * */ -void gfs2_rlist_alloc(struct gfs2_rgrp_list *rlist, unsigned int state) +void gfs2_rlist_alloc(struct gfs2_rgrp_list *rlist, + unsigned int state, u16 flags) { unsigned int x; - rlist->rl_ghs = kcalloc(rlist->rl_rgrps, sizeof(struct gfs2_holder), - GFP_NOFS | __GFP_NOFAIL); + rlist->rl_ghs = kmalloc_array(rlist->rl_rgrps, + sizeof(struct gfs2_holder), + GFP_NOFS | __GFP_NOFAIL); for (x = 0; x < rlist->rl_rgrps; x++) - gfs2_holder_init(rlist->rl_rgd[x]->rd_gl, - state, 0, - &rlist->rl_ghs[x]); + gfs2_holder_init(rlist->rl_rgd[x]->rd_gl, state, flags, + &rlist->rl_ghs[x]); } /** * gfs2_rlist_free - free a resource group list - * @list: the list of resource groups + * @rlist: the list of resource groups * */ @@ -2429,3 +2759,12 @@ void gfs2_rlist_free(struct gfs2_rgrp_list *rlist) } } +void rgrp_lock_local(struct gfs2_rgrpd *rgd) +{ + mutex_lock(&rgd->rd_mutex); +} + +void rgrp_unlock_local(struct gfs2_rgrpd *rgd) +{ + mutex_unlock(&rgd->rd_mutex); +} |