/*
* Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved.
- * Copyright (C) 2004-2005 Red Hat, Inc. All rights reserved.
+ * Copyright (C) 2004-2006 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
#include <linux/completion.h>
#include <linux/buffer_head.h>
#include <linux/pagemap.h>
+#include <linux/pagevec.h>
#include <linux/mpage.h>
#include <linux/fs.h>
#include <linux/gfs2_ondisk.h>
-#include <asm/semaphore.h>
#include "gfs2.h"
#include "lm_interface.h"
#include "log.h"
#include "meta_io.h"
#include "ops_address.h"
-#include "page.h"
#include "quota.h"
#include "trans.h"
#include "rgrp.h"
#include "ops_file.h"
#include "util.h"
+#include "glops.h"
+
+
+static void gfs2_page_add_databufs(struct gfs2_inode *ip, struct page *page,
+ unsigned int from, unsigned int to)
+{
+ struct buffer_head *head = page_buffers(page);
+ unsigned int bsize = head->b_size;
+ struct buffer_head *bh;
+ unsigned int start, end;
+
+ for (bh = head, start = 0; bh != head || !start;
+ bh = bh->b_this_page, start = end) {
+ end = start + bsize;
+ if (end <= from || start >= to)
+ continue;
+ gfs2_trans_add_bh(ip->i_gl, bh, 0);
+ }
+}
/**
* gfs2_get_block - Fills in a buffer head with details about a block
int gfs2_get_block(struct inode *inode, sector_t lblock,
struct buffer_head *bh_result, int create)
{
- struct gfs2_inode *ip = inode->u.generic_ip;
int new = create;
uint64_t dblock;
int error;
+ int boundary;
- error = gfs2_block_map(ip, lblock, &new, &dblock, NULL);
+ error = gfs2_block_map(inode, lblock, &new, &dblock, &boundary);
if (error)
return error;
map_bh(bh_result, inode->i_sb, dblock);
if (new)
set_buffer_new(bh_result);
+ if (boundary)
+ set_buffer_boundary(bh_result);
return 0;
}
static int get_block_noalloc(struct inode *inode, sector_t lblock,
struct buffer_head *bh_result, int create)
{
- struct gfs2_inode *ip = inode->u.generic_ip;
int new = 0;
uint64_t dblock;
int error;
+ int boundary;
- error = gfs2_block_map(ip, lblock, &new, &dblock, NULL);
+ error = gfs2_block_map(inode, lblock, &new, &dblock, &boundary);
if (error)
return error;
if (dblock)
map_bh(bh_result, inode->i_sb, dblock);
- else if (gfs2_assert_withdraw(ip->i_sbd, !create))
+ else if (gfs2_assert_withdraw(GFS2_SB(inode), !create))
error = -EIO;
+ if (boundary)
+ set_buffer_boundary(bh_result);
return error;
}
static int gfs2_writepage(struct page *page, struct writeback_control *wbc)
{
struct inode *inode = page->mapping->host;
- struct gfs2_inode *ip = page->mapping->host->u.generic_ip;
- struct gfs2_sbd *sdp = ip->i_sbd;
+ struct gfs2_inode *ip = GFS2_I(inode);
+ struct gfs2_sbd *sdp = GFS2_SB(inode);
loff_t i_size = i_size_read(inode);
pgoff_t end_index = i_size >> PAGE_CACHE_SHIFT;
unsigned offset;
/* Is the page fully outside i_size? (truncate in progress) */
offset = i_size & (PAGE_CACHE_SIZE-1);
- if (page->index >= end_index+1 || !offset) {
+ if (page->index > end_index || (page->index == end_index && !offset)) {
page->mapping->a_ops->invalidatepage(page, 0);
unlock_page(page);
return 0; /* don't care */
error = gfs2_trans_begin(sdp, RES_DINODE + 1, 0);
if (error)
goto out_ignore;
+ if (!page_has_buffers(page)) {
+ create_empty_buffers(page, inode->i_sb->s_blocksize,
+ (1 << BH_Dirty)|(1 << BH_Uptodate));
+ }
gfs2_page_add_databufs(ip, page, 0, sdp->sd_vfs->s_blocksize-1);
done_trans = 1;
}
return 0;
}
+static int zero_readpage(struct page *page)
+{
+ void *kaddr;
+
+ kaddr = kmap_atomic(page, KM_USER0);
+ memset(kaddr, 0, PAGE_CACHE_SIZE);
+ kunmap_atomic(page, KM_USER0);
+
+ SetPageUptodate(page);
+
+ return 0;
+}
+
/**
* stuffed_readpage - Fill in a Linux page with stuffed file data
* @ip: the inode
void *kaddr;
int error;
+ /* Only the first page of a stuffed file might contain data */
+ if (unlikely(page->index))
+ return zero_readpage(page);
+
error = gfs2_meta_inode_buffer(ip, &dibh);
if (error)
return error;
kaddr = kmap_atomic(page, KM_USER0);
- memcpy((char *)kaddr,
- dibh->b_data + sizeof(struct gfs2_dinode),
+ memcpy(kaddr, dibh->b_data + sizeof(struct gfs2_dinode),
ip->i_di.di_size);
- memset((char *)kaddr + ip->i_di.di_size,
- 0,
- PAGE_CACHE_SIZE - ip->i_di.di_size);
+ memset(kaddr + ip->i_di.di_size, 0, PAGE_CACHE_SIZE - ip->i_di.di_size);
kunmap_atomic(page, KM_USER0);
brelse(dibh);
return 0;
}
-static int zero_readpage(struct page *page)
-{
- void *kaddr;
-
- kaddr = kmap_atomic(page, KM_USER0);
- memset(kaddr, 0, PAGE_CACHE_SIZE);
- kunmap_atomic(page, KM_USER0);
-
- SetPageUptodate(page);
- unlock_page(page);
-
- return 0;
-}
/**
* gfs2_readpage - readpage with locking
static int gfs2_readpage(struct file *file, struct page *page)
{
- struct gfs2_inode *ip = page->mapping->host->u.generic_ip;
- struct gfs2_sbd *sdp = ip->i_sbd;
+ struct gfs2_inode *ip = GFS2_I(page->mapping->host);
+ struct gfs2_sbd *sdp = GFS2_SB(page->mapping->host);
struct gfs2_holder gh;
int error;
+ int do_unlock = 0;
- if (file != &gfs2_internal_file_sentinal) {
- gfs2_holder_init(ip->i_gl, LM_ST_SHARED, GL_ATIME, &gh);
+ if (likely(file != &gfs2_internal_file_sentinal)) {
+ if (file) {
+ struct gfs2_file *gf = file->private_data;
+ if (test_bit(GFF_EXLOCK, &gf->f_flags))
+ goto skip_lock;
+ }
+ gfs2_holder_init(ip->i_gl, LM_ST_SHARED, GL_ATIME|GL_AOP, &gh);
+ do_unlock = 1;
error = gfs2_glock_nq_m_atime(1, &gh);
- if (error)
+ if (unlikely(error))
goto out_unlock;
}
+skip_lock:
if (gfs2_is_stuffed(ip)) {
- if (!page->index) {
- error = stuffed_readpage(ip, page);
- unlock_page(page);
- } else
- error = zero_readpage(page);
+ error = stuffed_readpage(ip, page);
+ unlock_page(page);
} else
error = mpage_readpage(page, gfs2_get_block);
return error;
out_unlock:
unlock_page(page);
+ if (do_unlock)
+ gfs2_holder_uninit(&gh);
+ goto out;
+}
+
+/**
+ * gfs2_readpages - Read a bunch of pages at once
+ *
+ * Some notes:
+ * 1. This is only for readahead, so we can simply ignore any things
+ * which are slightly inconvenient (such as locking conflicts between
+ * the page lock and the glock) and return having done no I/O. Its
+ * obviously not something we'd want to do on too regular a basis.
+ * Any I/O we ignore at this time will be done via readpage later.
+ * 2. We have to handle stuffed files here too.
+ * 3. mpage_readpages() does most of the heavy lifting in the common case.
+ * 4. gfs2_get_block() is relied upon to set BH_Boundary in the right places.
+ * 5. We use LM_FLAG_TRY_1CB here, effectively we then have lock-ahead as
+ * well as read-ahead.
+ */
+static int gfs2_readpages(struct file *file, struct address_space *mapping,
+ struct list_head *pages, unsigned nr_pages)
+{
+ struct inode *inode = mapping->host;
+ struct gfs2_inode *ip = GFS2_I(inode);
+ struct gfs2_sbd *sdp = GFS2_SB(inode);
+ struct gfs2_holder gh;
+ unsigned page_idx;
+ int ret;
+ int do_unlock = 0;
+
+ if (likely(file != &gfs2_internal_file_sentinal)) {
+ if (file) {
+ struct gfs2_file *gf = file->private_data;
+ if (test_bit(GFF_EXLOCK, &gf->f_flags))
+ goto skip_lock;
+ }
+ gfs2_holder_init(ip->i_gl, LM_ST_SHARED,
+ LM_FLAG_TRY_1CB|GL_ATIME|GL_AOP, &gh);
+ do_unlock = 1;
+ ret = gfs2_glock_nq_m_atime(1, &gh);
+ if (ret == GLR_TRYFAILED)
+ goto out_noerror;
+ if (unlikely(ret))
+ goto out_unlock;
+ }
+skip_lock:
+ if (gfs2_is_stuffed(ip)) {
+ struct pagevec lru_pvec;
+ pagevec_init(&lru_pvec, 0);
+ for (page_idx = 0; page_idx < nr_pages; page_idx++) {
+ struct page *page = list_entry(pages->prev, struct page, lru);
+ prefetchw(&page->flags);
+ list_del(&page->lru);
+ if (!add_to_page_cache(page, mapping,
+ page->index, GFP_KERNEL)) {
+ ret = stuffed_readpage(ip, page);
+ unlock_page(page);
+ if (!pagevec_add(&lru_pvec, page))
+ __pagevec_lru_add(&lru_pvec);
+ } else {
+ page_cache_release(page);
+ }
+ }
+ pagevec_lru_add(&lru_pvec);
+ ret = 0;
+ } else {
+ /* What we really want to do .... */
+ ret = mpage_readpages(mapping, pages, nr_pages, gfs2_get_block);
+ }
+
+ if (do_unlock) {
+ gfs2_glock_dq_m(1, &gh);
+ gfs2_holder_uninit(&gh);
+ }
+out:
+ if (unlikely(test_bit(SDF_SHUTDOWN, &sdp->sd_flags)))
+ ret = -EIO;
+ return ret;
+out_noerror:
+ ret = 0;
+out_unlock:
+ /* unlock all pages, we can't do any I/O right now */
+ for (page_idx = 0; page_idx < nr_pages; page_idx++) {
+ struct page *page = list_entry(pages->prev, struct page, lru);
+ list_del(&page->lru);
+ unlock_page(page);
+ page_cache_release(page);
+ }
+ if (do_unlock)
+ gfs2_holder_uninit(&gh);
goto out;
}
static int gfs2_prepare_write(struct file *file, struct page *page,
unsigned from, unsigned to)
{
- struct gfs2_inode *ip = page->mapping->host->u.generic_ip;
- struct gfs2_sbd *sdp = ip->i_sbd;
+ struct gfs2_inode *ip = GFS2_I(page->mapping->host);
+ struct gfs2_sbd *sdp = GFS2_SB(page->mapping->host);
unsigned int data_blocks, ind_blocks, rblocks;
int alloc_required;
int error = 0;
loff_t end = ((loff_t)page->index << PAGE_CACHE_SHIFT) + to;
struct gfs2_alloc *al;
- gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, GL_ATIME, &ip->i_gh);
+ gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, GL_ATIME|GL_AOP, &ip->i_gh);
error = gfs2_glock_nq_m_atime(1, &ip->i_gh);
if (error)
goto out_uninit;
if (gfs2_is_stuffed(ip)) {
if (end > sdp->sd_sb.sb_bsize - sizeof(struct gfs2_dinode)) {
- error = gfs2_unstuff_dinode(ip, gfs2_unstuffer_page,
- page);
+ error = gfs2_unstuff_dinode(ip, page);
if (error == 0)
goto prepare_write;
} else if (!PageUptodate(page))
unsigned from, unsigned to)
{
struct inode *inode = page->mapping->host;
- struct gfs2_inode *ip = inode->u.generic_ip;
- struct gfs2_sbd *sdp = ip->i_sbd;
+ struct gfs2_inode *ip = GFS2_I(inode);
+ struct gfs2_sbd *sdp = GFS2_SB(inode);
int error = -EOPNOTSUPP;
struct buffer_head *dibh;
struct gfs2_alloc *al = &ip->i_alloc;;
static sector_t gfs2_bmap(struct address_space *mapping, sector_t lblock)
{
- struct gfs2_inode *ip = mapping->host->u.generic_ip;
+ struct gfs2_inode *ip = GFS2_I(mapping->host);
struct gfs2_holder i_gh;
sector_t dblock = 0;
int error;
if (bd) {
bd->bd_bh = NULL;
bh->b_private = NULL;
- gfs2_log_unlock(sdp);
- brelse(bh);
- } else
- gfs2_log_unlock(sdp);
+ }
+ gfs2_log_unlock(sdp);
lock_buffer(bh);
clear_buffer_dirty(bh);
static void gfs2_invalidatepage(struct page *page, unsigned long offset)
{
- struct gfs2_sbd *sdp = page->mapping->host->i_sb->s_fs_info;
+ struct gfs2_sbd *sdp = GFS2_SB(page->mapping->host);
struct buffer_head *head, *bh, *next;
unsigned int curr_off = 0;
return;
}
-static ssize_t gfs2_direct_IO_write(struct kiocb *iocb, const struct iovec *iov,
- loff_t offset, unsigned long nr_segs)
+static ssize_t gfs2_direct_IO(int rw, struct kiocb *iocb,
+ const struct iovec *iov, loff_t offset,
+ unsigned long nr_segs)
{
struct file *file = iocb->ki_filp;
struct inode *inode = file->f_mapping->host;
- struct gfs2_inode *ip = inode->u.generic_ip;
+ struct gfs2_inode *ip = GFS2_I(inode);
struct gfs2_holder gh;
int rv;
+ if (rw == READ)
+ mutex_lock(&inode->i_mutex);
/*
- * Shared lock, even though its write, since we do no allocation
+ * Shared lock, even if its a write, since we do no allocation
* on this path. All we need change is atime.
*/
gfs2_holder_init(ip->i_gl, LM_ST_SHARED, GL_ATIME, &gh);
if (rv)
goto out;
+ if (offset > i_size_read(inode))
+ goto out;
+
/*
* Should we return an error here? I can't see that O_DIRECT for
* a journaled file makes any sense. For now we'll silently fall
if (gfs2_is_stuffed(ip))
goto out;
- rv = __blockdev_direct_IO(WRITE, iocb, inode, inode->i_sb->s_bdev,
- iov, offset, nr_segs, gfs2_get_block,
- NULL, DIO_OWN_LOCKING);
+ rv = blockdev_direct_IO_own_locking(rw, iocb, inode,
+ inode->i_sb->s_bdev,
+ iov, offset, nr_segs,
+ gfs2_get_block, NULL);
out:
gfs2_glock_dq_m(1, &gh);
gfs2_holder_uninit(&gh);
+ if (rw == READ)
+ mutex_unlock(&inode->i_mutex);
return rv;
}
/**
- * gfs2_direct_IO
+ * stuck_releasepage - We're stuck in gfs2_releasepage(). Print stuff out.
+ * @bh: the buffer we're stuck on
*
- * This is called with a shared lock already held for the read path.
- * Currently, no locks are held when the write path is called.
*/
-static ssize_t gfs2_direct_IO(int rw, struct kiocb *iocb,
- const struct iovec *iov, loff_t offset,
- unsigned long nr_segs)
+
+static void stuck_releasepage(struct buffer_head *bh)
{
- struct file *file = iocb->ki_filp;
- struct inode *inode = file->f_mapping->host;
- struct gfs2_inode *ip = inode->u.generic_ip;
- struct gfs2_sbd *sdp = ip->i_sbd;
+ struct inode *inode = bh->b_page->mapping->host;
+ struct gfs2_sbd *sdp = inode->i_sb->s_fs_info;
+ struct gfs2_bufdata *bd = bh->b_private;
+ struct gfs2_glock *gl;
+static unsigned limit = 0;
+
+ if (limit++ > 3)
+ return;
+
+ fs_warn(sdp, "stuck in gfs2_releasepage() %p\n", inode);
+ fs_warn(sdp, "blkno = %llu, bh->b_count = %d\n",
+ (unsigned long long)bh->b_blocknr, atomic_read(&bh->b_count));
+ fs_warn(sdp, "pinned = %u\n", buffer_pinned(bh));
+ fs_warn(sdp, "bh->b_private = %s\n", (bd) ? "!NULL" : "NULL");
+
+ if (!bd)
+ return;
+
+ gl = bd->bd_gl;
+
+ fs_warn(sdp, "gl = (%u, %llu)\n",
+ gl->gl_name.ln_type, (unsigned long long)gl->gl_name.ln_number);
- if (rw == WRITE)
- return gfs2_direct_IO_write(iocb, iov, offset, nr_segs);
+ fs_warn(sdp, "bd_list_tr = %s, bd_le.le_list = %s\n",
+ (list_empty(&bd->bd_list_tr)) ? "no" : "yes",
+ (list_empty(&bd->bd_le.le_list)) ? "no" : "yes");
- if (gfs2_assert_warn(sdp, gfs2_glock_is_locked_by_me(ip->i_gl)) ||
- gfs2_assert_warn(sdp, !gfs2_is_stuffed(ip)))
- return -EINVAL;
+ if (gl->gl_ops == &gfs2_inode_glops) {
+ struct gfs2_inode *ip = gl->gl_object;
+ unsigned int x;
- return __blockdev_direct_IO(READ, iocb, inode, inode->i_sb->s_bdev, iov,
- offset, nr_segs, gfs2_get_block, NULL,
- DIO_OWN_LOCKING);
+ if (!ip)
+ return;
+
+ fs_warn(sdp, "ip = %llu %llu\n",
+ (unsigned long long)ip->i_num.no_formal_ino,
+ (unsigned long long)ip->i_num.no_addr);
+
+ for (x = 0; x < GFS2_MAX_META_HEIGHT; x++)
+ fs_warn(sdp, "ip->i_cache[%u] = %s\n",
+ x, (ip->i_cache[x]) ? "!NULL" : "NULL");
+ }
+}
+
+/**
+ * gfs2_aspace_releasepage - free the metadata associated with a page
+ * @page: the page that's being released
+ * @gfp_mask: passed from Linux VFS, ignored by us
+ *
+ * Call try_to_free_buffers() if the buffers in this page can be
+ * released.
+ *
+ * Returns: 0
+ */
+
+int gfs2_releasepage(struct page *page, gfp_t gfp_mask)
+{
+ struct inode *aspace = page->mapping->host;
+ struct gfs2_sbd *sdp = aspace->i_sb->s_fs_info;
+ struct buffer_head *bh, *head;
+ struct gfs2_bufdata *bd;
+ unsigned long t = jiffies + gfs2_tune_get(sdp, gt_stall_secs) * HZ;
+
+ if (!page_has_buffers(page))
+ goto out;
+
+ head = bh = page_buffers(page);
+ do {
+ while (atomic_read(&bh->b_count)) {
+ if (!atomic_read(&aspace->i_writecount))
+ return 0;
+
+ if (time_after_eq(jiffies, t)) {
+ stuck_releasepage(bh);
+ /* should we withdraw here? */
+ return 0;
+ }
+
+ yield();
+ }
+
+ gfs2_assert_warn(sdp, !buffer_pinned(bh));
+
+ bd = bh->b_private;
+ if (bd) {
+ gfs2_assert_warn(sdp, bd->bd_bh == bh);
+ gfs2_assert_warn(sdp, list_empty(&bd->bd_list_tr));
+ gfs2_assert_warn(sdp, list_empty(&bd->bd_le.le_list));
+ gfs2_assert_warn(sdp, !bd->bd_ail);
+ kmem_cache_free(gfs2_bufdata_cachep, bd);
+ bh->b_private = NULL;
+ }
+
+ bh = bh->b_this_page;
+ } while (bh != head);
+
+out:
+ return try_to_free_buffers(page);
}
-struct address_space_operations gfs2_file_aops = {
+const struct address_space_operations gfs2_file_aops = {
.writepage = gfs2_writepage,
.readpage = gfs2_readpage,
+ .readpages = gfs2_readpages,
.sync_page = block_sync_page,
.prepare_write = gfs2_prepare_write,
.commit_write = gfs2_commit_write,
.bmap = gfs2_bmap,
.invalidatepage = gfs2_invalidatepage,
+ .releasepage = gfs2_releasepage,
.direct_IO = gfs2_direct_IO,
};
git.openpandora.org / pandora-kernel.git / blobdiff