linux/fs/gfs2/aops.c

820 lines
20 KiB
C
Raw Normal View History

// 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.
*/
#include <linux/sched.h>
#include <linux/slab.h>
#include <linux/spinlock.h>
#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/writeback.h>
#include <linux/swap.h>
#include <linux/gfs2_ondisk.h>
#include <linux/backing-dev.h>
#include <linux/uio.h>
#include <trace/events/writeback.h>
#include <linux/sched/signal.h>
#include "gfs2.h"
#include "incore.h"
#include "bmap.h"
#include "glock.h"
#include "inode.h"
#include "log.h"
#include "meta_io.h"
#include "quota.h"
#include "trans.h"
[GFS2] Make journaled data files identical to normal files on disk This is a very large patch, with a few still to be resolved issues so you might want to check out the previous head of the tree since this is known to be unstable. Fixes for the various bugs will be forthcoming shortly. This patch removes the special data format which has been used up till now for journaled data files. Directories still retain the old format so that they will remain on disk compatible with earlier releases. As a result you can now do the following with journaled data files: 1) mmap them 2) export them over NFS 3) convert to/from normal files whenever you want to (the zero length restriction is gone) In addition the level at which GFS' locking is done has changed for all files (since they all now use the page cache) such that the locking is done at the page cache level rather than the level of the fs operations. This should mean that things like loopback mounts and other things which touch the page cache directly should now work. Current known issues: 1. There is a lock mode inversion problem related to the resource group hold function which needs to be resolved. 2. Any significant amount of I/O causes an oops with an offset of hex 320 (NULL pointer dereference) which appears to be related to a journaled data buffer appearing on a list where it shouldn't be. 3. Direct I/O writes are disabled for the time being (will reappear later) 4. There is probably a deadlock between the page lock and GFS' locks under certain combinations of mmap and fs operation I/O. 5. Issue relating to ref counting on internally used inodes causes a hang on umount (discovered before this patch, and not fixed by it) 6. One part of the directory metadata is different from GFS1 and will need to be resolved before next release. Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
2006-02-08 11:50:51 +00:00
#include "rgrp.h"
#include "super.h"
#include "util.h"
#include "glops.h"
#include "aops.h"
void gfs2_page_add_databufs(struct gfs2_inode *ip, struct page *page,
unsigned int from, unsigned int len)
{
struct buffer_head *head = page_buffers(page);
unsigned int bsize = head->b_size;
struct buffer_head *bh;
unsigned int to = from + len;
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)
continue;
if (start >= to)
break;
set_buffer_uptodate(bh);
gfs2_trans_add_data(ip->i_gl, bh);
}
}
/**
* gfs2_get_block_noalloc - Fills in a buffer head with details about a block
* @inode: The inode
* @lblock: The block number to look up
* @bh_result: The buffer head to return the result in
* @create: Non-zero if we may add block to the file
*
* Returns: errno
*/
static int gfs2_get_block_noalloc(struct inode *inode, sector_t lblock,
struct buffer_head *bh_result, int create)
{
int error;
error = gfs2_block_map(inode, lblock, bh_result, 0);
if (error)
return error;
if (!buffer_mapped(bh_result))
return -ENODATA;
return 0;
}
/**
* gfs2_writepage - Write page for writeback mappings
* @page: The page
* @wbc: The writeback control
*/
static int gfs2_writepage(struct page *page, struct writeback_control *wbc)
{
[GFS2] Make journaled data files identical to normal files on disk This is a very large patch, with a few still to be resolved issues so you might want to check out the previous head of the tree since this is known to be unstable. Fixes for the various bugs will be forthcoming shortly. This patch removes the special data format which has been used up till now for journaled data files. Directories still retain the old format so that they will remain on disk compatible with earlier releases. As a result you can now do the following with journaled data files: 1) mmap them 2) export them over NFS 3) convert to/from normal files whenever you want to (the zero length restriction is gone) In addition the level at which GFS' locking is done has changed for all files (since they all now use the page cache) such that the locking is done at the page cache level rather than the level of the fs operations. This should mean that things like loopback mounts and other things which touch the page cache directly should now work. Current known issues: 1. There is a lock mode inversion problem related to the resource group hold function which needs to be resolved. 2. Any significant amount of I/O causes an oops with an offset of hex 320 (NULL pointer dereference) which appears to be related to a journaled data buffer appearing on a list where it shouldn't be. 3. Direct I/O writes are disabled for the time being (will reappear later) 4. There is probably a deadlock between the page lock and GFS' locks under certain combinations of mmap and fs operation I/O. 5. Issue relating to ref counting on internally used inodes causes a hang on umount (discovered before this patch, and not fixed by it) 6. One part of the directory metadata is different from GFS1 and will need to be resolved before next release. Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
2006-02-08 11:50:51 +00:00
struct inode *inode = page->mapping->host;
struct gfs2_inode *ip = GFS2_I(inode);
struct gfs2_sbd *sdp = GFS2_SB(inode);
struct iomap_writepage_ctx wpc = { };
if (gfs2_assert_withdraw(sdp, gfs2_glock_is_held_excl(ip->i_gl)))
goto out;
if (current->journal_info)
goto redirty;
return iomap_writepage(page, wbc, &wpc, &gfs2_writeback_ops);
redirty:
redirty_page_for_writepage(wbc, page);
out:
unlock_page(page);
return 0;
}
/**
* gfs2_write_jdata_page - gfs2 jdata-specific version of block_write_full_page
* @page: The page to write
* @wbc: The writeback control
*
* This is the same as calling block_write_full_page, but it also
* writes pages outside of i_size
*/
static int gfs2_write_jdata_page(struct page *page,
struct writeback_control *wbc)
{
struct inode * const inode = page->mapping->host;
loff_t i_size = i_size_read(inode);
const pgoff_t end_index = i_size >> PAGE_SHIFT;
unsigned offset;
/*
* The page straddles i_size. It must be zeroed out on each and every
* writepage invocation because it may be mmapped. "A file is mapped
* in multiples of the page size. For a file that is not a multiple of
* the page size, the remaining memory is zeroed when mapped, and
* writes to that region are not written out to the file."
*/
offset = i_size & (PAGE_SIZE - 1);
if (page->index == end_index && offset)
zero_user_segment(page, offset, PAGE_SIZE);
return __block_write_full_page(inode, page, gfs2_get_block_noalloc, wbc,
end_buffer_async_write);
}
/**
* __gfs2_jdata_writepage - The core of jdata writepage
* @page: The page to write
* @wbc: The writeback control
*
* This is shared between writepage and writepages and implements the
* core of the writepage operation. If a transaction is required then
* PageChecked will have been set and the transaction will have
* already been started before this is called.
*/
static int __gfs2_jdata_writepage(struct page *page, struct writeback_control *wbc)
{
struct inode *inode = page->mapping->host;
struct gfs2_inode *ip = GFS2_I(inode);
struct gfs2_sbd *sdp = GFS2_SB(inode);
if (PageChecked(page)) {
ClearPageChecked(page);
if (!page_has_buffers(page)) {
create_empty_buffers(page, inode->i_sb->s_blocksize,
BIT(BH_Dirty)|BIT(BH_Uptodate));
}
gfs2_page_add_databufs(ip, page, 0, sdp->sd_vfs->s_blocksize);
}
return gfs2_write_jdata_page(page, wbc);
}
/**
* gfs2_jdata_writepage - Write complete page
* @page: Page to write
* @wbc: The writeback control
*
* Returns: errno
*
*/
static int gfs2_jdata_writepage(struct page *page, struct writeback_control *wbc)
{
struct inode *inode = page->mapping->host;
struct gfs2_inode *ip = GFS2_I(inode);
struct gfs2_sbd *sdp = GFS2_SB(inode);
if (gfs2_assert_withdraw(sdp, gfs2_glock_is_held_excl(ip->i_gl)))
goto out;
if (PageChecked(page) || current->journal_info)
goto out_ignore;
return __gfs2_jdata_writepage(page, wbc);
[GFS2] Make journaled data files identical to normal files on disk This is a very large patch, with a few still to be resolved issues so you might want to check out the previous head of the tree since this is known to be unstable. Fixes for the various bugs will be forthcoming shortly. This patch removes the special data format which has been used up till now for journaled data files. Directories still retain the old format so that they will remain on disk compatible with earlier releases. As a result you can now do the following with journaled data files: 1) mmap them 2) export them over NFS 3) convert to/from normal files whenever you want to (the zero length restriction is gone) In addition the level at which GFS' locking is done has changed for all files (since they all now use the page cache) such that the locking is done at the page cache level rather than the level of the fs operations. This should mean that things like loopback mounts and other things which touch the page cache directly should now work. Current known issues: 1. There is a lock mode inversion problem related to the resource group hold function which needs to be resolved. 2. Any significant amount of I/O causes an oops with an offset of hex 320 (NULL pointer dereference) which appears to be related to a journaled data buffer appearing on a list where it shouldn't be. 3. Direct I/O writes are disabled for the time being (will reappear later) 4. There is probably a deadlock between the page lock and GFS' locks under certain combinations of mmap and fs operation I/O. 5. Issue relating to ref counting on internally used inodes causes a hang on umount (discovered before this patch, and not fixed by it) 6. One part of the directory metadata is different from GFS1 and will need to be resolved before next release. Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
2006-02-08 11:50:51 +00:00
out_ignore:
redirty_page_for_writepage(wbc, page);
out:
[GFS2] Make journaled data files identical to normal files on disk This is a very large patch, with a few still to be resolved issues so you might want to check out the previous head of the tree since this is known to be unstable. Fixes for the various bugs will be forthcoming shortly. This patch removes the special data format which has been used up till now for journaled data files. Directories still retain the old format so that they will remain on disk compatible with earlier releases. As a result you can now do the following with journaled data files: 1) mmap them 2) export them over NFS 3) convert to/from normal files whenever you want to (the zero length restriction is gone) In addition the level at which GFS' locking is done has changed for all files (since they all now use the page cache) such that the locking is done at the page cache level rather than the level of the fs operations. This should mean that things like loopback mounts and other things which touch the page cache directly should now work. Current known issues: 1. There is a lock mode inversion problem related to the resource group hold function which needs to be resolved. 2. Any significant amount of I/O causes an oops with an offset of hex 320 (NULL pointer dereference) which appears to be related to a journaled data buffer appearing on a list where it shouldn't be. 3. Direct I/O writes are disabled for the time being (will reappear later) 4. There is probably a deadlock between the page lock and GFS' locks under certain combinations of mmap and fs operation I/O. 5. Issue relating to ref counting on internally used inodes causes a hang on umount (discovered before this patch, and not fixed by it) 6. One part of the directory metadata is different from GFS1 and will need to be resolved before next release. Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
2006-02-08 11:50:51 +00:00
unlock_page(page);
return 0;
}
/**
* gfs2_writepages - Write a bunch of dirty pages back to disk
* @mapping: The mapping to write
* @wbc: Write-back control
*
* Used for both ordered and writeback modes.
*/
static int gfs2_writepages(struct address_space *mapping,
struct writeback_control *wbc)
{
struct gfs2_sbd *sdp = gfs2_mapping2sbd(mapping);
struct iomap_writepage_ctx wpc = { };
int ret;
/*
* Even if we didn't write any pages here, we might still be holding
* dirty pages in the ail. We forcibly flush the ail because we don't
* want balance_dirty_pages() to loop indefinitely trying to write out
* pages held in the ail that it can't find.
*/
ret = iomap_writepages(mapping, wbc, &wpc, &gfs2_writeback_ops);
if (ret == 0)
set_bit(SDF_FORCE_AIL_FLUSH, &sdp->sd_flags);
return ret;
}
/**
* gfs2_write_jdata_pagevec - Write back a pagevec's worth of pages
* @mapping: The mapping
* @wbc: The writeback control
* @pvec: The vector of pages
* @nr_pages: The number of pages to write
* @done_index: Page index
*
* Returns: non-zero if loop should terminate, zero otherwise
*/
static int gfs2_write_jdata_pagevec(struct address_space *mapping,
struct writeback_control *wbc,
struct pagevec *pvec,
int nr_pages,
pgoff_t *done_index)
{
struct inode *inode = mapping->host;
struct gfs2_sbd *sdp = GFS2_SB(inode);
unsigned nrblocks = nr_pages * (PAGE_SIZE >> inode->i_blkbits);
int i;
int ret;
ret = gfs2_trans_begin(sdp, nrblocks, nrblocks);
if (ret < 0)
return ret;
for(i = 0; i < nr_pages; i++) {
struct page *page = pvec->pages[i];
*done_index = page->index;
lock_page(page);
if (unlikely(page->mapping != mapping)) {
continue_unlock:
unlock_page(page);
continue;
}
if (!PageDirty(page)) {
/* someone wrote it for us */
goto continue_unlock;
}
if (PageWriteback(page)) {
if (wbc->sync_mode != WB_SYNC_NONE)
wait_on_page_writeback(page);
else
goto continue_unlock;
}
BUG_ON(PageWriteback(page));
if (!clear_page_dirty_for_io(page))
goto continue_unlock;
trace_wbc_writepage(wbc, inode_to_bdi(inode));
ret = __gfs2_jdata_writepage(page, wbc);
if (unlikely(ret)) {
if (ret == AOP_WRITEPAGE_ACTIVATE) {
unlock_page(page);
ret = 0;
} else {
/*
* done_index is set past this page,
* so media errors will not choke
* background writeout for the entire
* file. This has consequences for
* range_cyclic semantics (ie. it may
* not be suitable for data integrity
* writeout).
*/
*done_index = page->index + 1;
ret = 1;
break;
}
}
/*
* We stop writing back only if we are not doing
* integrity sync. In case of integrity sync we have to
* keep going until we have written all the pages
* we tagged for writeback prior to entering this loop.
*/
if (--wbc->nr_to_write <= 0 && wbc->sync_mode == WB_SYNC_NONE) {
ret = 1;
break;
}
}
gfs2_trans_end(sdp);
return ret;
}
/**
* gfs2_write_cache_jdata - Like write_cache_pages but different
* @mapping: The mapping to write
* @wbc: The writeback control
*
* The reason that we use our own function here is that we need to
* start transactions before we grab page locks. This allows us
* to get the ordering right.
*/
static int gfs2_write_cache_jdata(struct address_space *mapping,
struct writeback_control *wbc)
{
int ret = 0;
int done = 0;
struct pagevec pvec;
int nr_pages;
treewide: Remove uninitialized_var() usage Using uninitialized_var() is dangerous as it papers over real bugs[1] (or can in the future), and suppresses unrelated compiler warnings (e.g. "unused variable"). If the compiler thinks it is uninitialized, either simply initialize the variable or make compiler changes. In preparation for removing[2] the[3] macro[4], remove all remaining needless uses with the following script: git grep '\buninitialized_var\b' | cut -d: -f1 | sort -u | \ xargs perl -pi -e \ 's/\buninitialized_var\(([^\)]+)\)/\1/g; s:\s*/\* (GCC be quiet|to make compiler happy) \*/$::g;' drivers/video/fbdev/riva/riva_hw.c was manually tweaked to avoid pathological white-space. No outstanding warnings were found building allmodconfig with GCC 9.3.0 for x86_64, i386, arm64, arm, powerpc, powerpc64le, s390x, mips, sparc64, alpha, and m68k. [1] https://lore.kernel.org/lkml/20200603174714.192027-1-glider@google.com/ [2] https://lore.kernel.org/lkml/CA+55aFw+Vbj0i=1TGqCR5vQkCzWJ0QxK6CernOU6eedsudAixw@mail.gmail.com/ [3] https://lore.kernel.org/lkml/CA+55aFwgbgqhbp1fkxvRKEpzyR5J8n1vKT1VZdz9knmPuXhOeg@mail.gmail.com/ [4] https://lore.kernel.org/lkml/CA+55aFz2500WfbKXAx8s67wrm9=yVJu65TpLgN_ybYNv0VEOKA@mail.gmail.com/ Reviewed-by: Leon Romanovsky <leonro@mellanox.com> # drivers/infiniband and mlx4/mlx5 Acked-by: Jason Gunthorpe <jgg@mellanox.com> # IB Acked-by: Kalle Valo <kvalo@codeaurora.org> # wireless drivers Reviewed-by: Chao Yu <yuchao0@huawei.com> # erofs Signed-off-by: Kees Cook <keescook@chromium.org>
2020-06-03 20:09:38 +00:00
pgoff_t writeback_index;
pgoff_t index;
pgoff_t end;
pgoff_t done_index;
int cycled;
int range_whole = 0;
xa_mark_t tag;
pagevec_init(&pvec);
if (wbc->range_cyclic) {
writeback_index = mapping->writeback_index; /* prev offset */
index = writeback_index;
if (index == 0)
cycled = 1;
else
cycled = 0;
end = -1;
} else {
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 12:29:47 +00:00
index = wbc->range_start >> PAGE_SHIFT;
end = wbc->range_end >> PAGE_SHIFT;
if (wbc->range_start == 0 && wbc->range_end == LLONG_MAX)
range_whole = 1;
cycled = 1; /* ignore range_cyclic tests */
}
if (wbc->sync_mode == WB_SYNC_ALL || wbc->tagged_writepages)
tag = PAGECACHE_TAG_TOWRITE;
else
tag = PAGECACHE_TAG_DIRTY;
retry:
if (wbc->sync_mode == WB_SYNC_ALL || wbc->tagged_writepages)
tag_pages_for_writeback(mapping, index, end);
done_index = index;
while (!done && (index <= end)) {
nr_pages = pagevec_lookup_range_tag(&pvec, mapping, &index, end,
tag);
if (nr_pages == 0)
break;
ret = gfs2_write_jdata_pagevec(mapping, wbc, &pvec, nr_pages, &done_index);
if (ret)
done = 1;
if (ret > 0)
ret = 0;
pagevec_release(&pvec);
cond_resched();
}
if (!cycled && !done) {
/*
* range_cyclic:
* We hit the last page and there is more work to be done: wrap
* back to the start of the file
*/
cycled = 1;
index = 0;
end = writeback_index - 1;
goto retry;
}
if (wbc->range_cyclic || (range_whole && wbc->nr_to_write > 0))
mapping->writeback_index = done_index;
return ret;
}
/**
* gfs2_jdata_writepages - Write a bunch of dirty pages back to disk
* @mapping: The mapping to write
* @wbc: The writeback control
*
*/
static int gfs2_jdata_writepages(struct address_space *mapping,
struct writeback_control *wbc)
{
struct gfs2_inode *ip = GFS2_I(mapping->host);
struct gfs2_sbd *sdp = GFS2_SB(mapping->host);
int ret;
ret = gfs2_write_cache_jdata(mapping, wbc);
if (ret == 0 && wbc->sync_mode == WB_SYNC_ALL) {
gfs2_log_flush(sdp, ip->i_gl, GFS2_LOG_HEAD_FLUSH_NORMAL |
GFS2_LFC_JDATA_WPAGES);
ret = gfs2_write_cache_jdata(mapping, wbc);
}
return ret;
}
/**
* stuffed_readpage - Fill in a Linux page with stuffed file data
* @ip: the inode
* @page: the page
*
* Returns: errno
*/
static int stuffed_readpage(struct gfs2_inode *ip, struct page *page)
{
struct buffer_head *dibh;
u64 dsize = i_size_read(&ip->i_inode);
void *kaddr;
int error;
/*
* Due to the order of unstuffing files and ->fault(), we can be
* asked for a zero page in the case of a stuffed file being extended,
* so we need to supply one here. It doesn't happen often.
*/
if (unlikely(page->index)) {
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 12:29:47 +00:00
zero_user(page, 0, PAGE_SIZE);
SetPageUptodate(page);
return 0;
}
error = gfs2_meta_inode_buffer(ip, &dibh);
if (error)
return error;
kaddr = kmap_atomic(page);
if (dsize > gfs2_max_stuffed_size(ip))
dsize = gfs2_max_stuffed_size(ip);
memcpy(kaddr, dibh->b_data + sizeof(struct gfs2_dinode), dsize);
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 12:29:47 +00:00
memset(kaddr + dsize, 0, PAGE_SIZE - dsize);
kunmap_atomic(kaddr);
flush_dcache_page(page);
brelse(dibh);
SetPageUptodate(page);
return 0;
}
static int __gfs2_readpage(void *file, struct page *page)
{
struct inode *inode = page->mapping->host;
struct gfs2_inode *ip = GFS2_I(inode);
struct gfs2_sbd *sdp = GFS2_SB(inode);
int error;
if (!gfs2_is_jdata(ip) ||
(i_blocksize(inode) == PAGE_SIZE && !page_has_buffers(page))) {
error = iomap_readpage(page, &gfs2_iomap_ops);
} else if (gfs2_is_stuffed(ip)) {
error = stuffed_readpage(ip, page);
unlock_page(page);
} else {
error = mpage_readpage(page, gfs2_block_map);
}
if (unlikely(gfs2_withdrawn(sdp)))
return -EIO;
return error;
}
/**
* gfs2_readpage - read a page of a file
* @file: The file to read
* @page: The page of the file
*/
static int gfs2_readpage(struct file *file, struct page *page)
{
return __gfs2_readpage(file, page);
}
/**
* gfs2_internal_read - read an internal file
* @ip: The gfs2 inode
* @buf: The buffer to fill
* @pos: The file position
* @size: The amount to read
*
*/
int gfs2_internal_read(struct gfs2_inode *ip, char *buf, loff_t *pos,
unsigned size)
{
struct address_space *mapping = ip->i_inode.i_mapping;
unsigned long index = *pos >> PAGE_SHIFT;
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 12:29:47 +00:00
unsigned offset = *pos & (PAGE_SIZE - 1);
unsigned copied = 0;
unsigned amt;
struct page *page;
void *p;
do {
amt = size - copied;
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 12:29:47 +00:00
if (offset + size > PAGE_SIZE)
amt = PAGE_SIZE - offset;
page = read_cache_page(mapping, index, __gfs2_readpage, NULL);
if (IS_ERR(page))
return PTR_ERR(page);
p = kmap_atomic(page);
memcpy(buf + copied, p + offset, amt);
kunmap_atomic(p);
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 12:29:47 +00:00
put_page(page);
copied += amt;
index++;
offset = 0;
} while(copied < size);
(*pos) += size;
return size;
}
/**
* gfs2_readahead - Read a bunch of pages at once
* @file: The file to read from
* @mapping: Address space info
* @pages: List of pages to read
* @nr_pages: Number of pages to read
*
* 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 don't handle stuffed files here we let readpage do the honours.
* 3. mpage_readahead() does most of the heavy lifting in the common case.
* 4. gfs2_block_map() is relied upon to set BH_Boundary in the right places.
*/
static void gfs2_readahead(struct readahead_control *rac)
{
struct inode *inode = rac->mapping->host;
struct gfs2_inode *ip = GFS2_I(inode);
if (gfs2_is_stuffed(ip))
;
else if (gfs2_is_jdata(ip))
mpage_readahead(rac, gfs2_block_map);
else
iomap_readahead(rac, &gfs2_iomap_ops);
}
[GFS2] kernel changes to support new gfs2_grow command This is another revision of my gfs2 kernel patch that allows gfs2_grow to function properly. Steve Whitehouse expressed some concerns about the previous patch and I restructured it based on his comments. The previous patch was doing the statfs_change at file close time, under its own transaction. The current patch does the statfs_change inside the gfs2_commit_write function, which keeps it under the umbrella of the inode transaction. I can't call ri_update to re-read the rindex file during the transaction because the transaction may have outstanding unwritten buffers attached to the rgrps that would be otherwise blown away. So instead, I created a new function, gfs2_ri_total, that will re-read the rindex file just to total the file system space for the sake of the statfs_change. The ri_update will happen later, when gfs2 realizes the version number has changed, as it happened before my patch. Since the statfs_change is happening at write_commit time and there may be multiple writes to the rindex file for one grow operation. So one consequence of this restructuring is that instead of getting one kernel message to indicate the change, you may see several. For example, before when you did a gfs2_grow, you'd get a single message like: GFS2: File system extended by 247876 blocks (968MB) Now you get something like: GFS2: File system extended by 207896 blocks (812MB) GFS2: File system extended by 39980 blocks (156MB) This version has also been successfully run against the hours-long "gfs2_fsck_hellfire" test that does several gfs2_grow and gfs2_fsck while interjecting file system damage. It does this repeatedly under a variety Resource Group conditions. Signed-off-By: Bob Peterson <rpeterso@redhat.com> Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
2007-05-09 14:37:57 +00:00
/**
* adjust_fs_space - Adjusts the free space available due to gfs2_grow
* @inode: the rindex inode
*/
void adjust_fs_space(struct inode *inode)
[GFS2] kernel changes to support new gfs2_grow command This is another revision of my gfs2 kernel patch that allows gfs2_grow to function properly. Steve Whitehouse expressed some concerns about the previous patch and I restructured it based on his comments. The previous patch was doing the statfs_change at file close time, under its own transaction. The current patch does the statfs_change inside the gfs2_commit_write function, which keeps it under the umbrella of the inode transaction. I can't call ri_update to re-read the rindex file during the transaction because the transaction may have outstanding unwritten buffers attached to the rgrps that would be otherwise blown away. So instead, I created a new function, gfs2_ri_total, that will re-read the rindex file just to total the file system space for the sake of the statfs_change. The ri_update will happen later, when gfs2 realizes the version number has changed, as it happened before my patch. Since the statfs_change is happening at write_commit time and there may be multiple writes to the rindex file for one grow operation. So one consequence of this restructuring is that instead of getting one kernel message to indicate the change, you may see several. For example, before when you did a gfs2_grow, you'd get a single message like: GFS2: File system extended by 247876 blocks (968MB) Now you get something like: GFS2: File system extended by 207896 blocks (812MB) GFS2: File system extended by 39980 blocks (156MB) This version has also been successfully run against the hours-long "gfs2_fsck_hellfire" test that does several gfs2_grow and gfs2_fsck while interjecting file system damage. It does this repeatedly under a variety Resource Group conditions. Signed-off-By: Bob Peterson <rpeterso@redhat.com> Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
2007-05-09 14:37:57 +00:00
{
struct gfs2_sbd *sdp = GFS2_SB(inode);
struct gfs2_inode *m_ip = GFS2_I(sdp->sd_statfs_inode);
struct gfs2_inode *l_ip = GFS2_I(sdp->sd_sc_inode);
[GFS2] kernel changes to support new gfs2_grow command This is another revision of my gfs2 kernel patch that allows gfs2_grow to function properly. Steve Whitehouse expressed some concerns about the previous patch and I restructured it based on his comments. The previous patch was doing the statfs_change at file close time, under its own transaction. The current patch does the statfs_change inside the gfs2_commit_write function, which keeps it under the umbrella of the inode transaction. I can't call ri_update to re-read the rindex file during the transaction because the transaction may have outstanding unwritten buffers attached to the rgrps that would be otherwise blown away. So instead, I created a new function, gfs2_ri_total, that will re-read the rindex file just to total the file system space for the sake of the statfs_change. The ri_update will happen later, when gfs2 realizes the version number has changed, as it happened before my patch. Since the statfs_change is happening at write_commit time and there may be multiple writes to the rindex file for one grow operation. So one consequence of this restructuring is that instead of getting one kernel message to indicate the change, you may see several. For example, before when you did a gfs2_grow, you'd get a single message like: GFS2: File system extended by 247876 blocks (968MB) Now you get something like: GFS2: File system extended by 207896 blocks (812MB) GFS2: File system extended by 39980 blocks (156MB) This version has also been successfully run against the hours-long "gfs2_fsck_hellfire" test that does several gfs2_grow and gfs2_fsck while interjecting file system damage. It does this repeatedly under a variety Resource Group conditions. Signed-off-By: Bob Peterson <rpeterso@redhat.com> Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
2007-05-09 14:37:57 +00:00
struct gfs2_statfs_change_host *m_sc = &sdp->sd_statfs_master;
struct gfs2_statfs_change_host *l_sc = &sdp->sd_statfs_local;
struct buffer_head *m_bh, *l_bh;
[GFS2] kernel changes to support new gfs2_grow command This is another revision of my gfs2 kernel patch that allows gfs2_grow to function properly. Steve Whitehouse expressed some concerns about the previous patch and I restructured it based on his comments. The previous patch was doing the statfs_change at file close time, under its own transaction. The current patch does the statfs_change inside the gfs2_commit_write function, which keeps it under the umbrella of the inode transaction. I can't call ri_update to re-read the rindex file during the transaction because the transaction may have outstanding unwritten buffers attached to the rgrps that would be otherwise blown away. So instead, I created a new function, gfs2_ri_total, that will re-read the rindex file just to total the file system space for the sake of the statfs_change. The ri_update will happen later, when gfs2 realizes the version number has changed, as it happened before my patch. Since the statfs_change is happening at write_commit time and there may be multiple writes to the rindex file for one grow operation. So one consequence of this restructuring is that instead of getting one kernel message to indicate the change, you may see several. For example, before when you did a gfs2_grow, you'd get a single message like: GFS2: File system extended by 247876 blocks (968MB) Now you get something like: GFS2: File system extended by 207896 blocks (812MB) GFS2: File system extended by 39980 blocks (156MB) This version has also been successfully run against the hours-long "gfs2_fsck_hellfire" test that does several gfs2_grow and gfs2_fsck while interjecting file system damage. It does this repeatedly under a variety Resource Group conditions. Signed-off-By: Bob Peterson <rpeterso@redhat.com> Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
2007-05-09 14:37:57 +00:00
u64 fs_total, new_free;
if (gfs2_trans_begin(sdp, 2 * RES_STATFS, 0) != 0)
return;
[GFS2] kernel changes to support new gfs2_grow command This is another revision of my gfs2 kernel patch that allows gfs2_grow to function properly. Steve Whitehouse expressed some concerns about the previous patch and I restructured it based on his comments. The previous patch was doing the statfs_change at file close time, under its own transaction. The current patch does the statfs_change inside the gfs2_commit_write function, which keeps it under the umbrella of the inode transaction. I can't call ri_update to re-read the rindex file during the transaction because the transaction may have outstanding unwritten buffers attached to the rgrps that would be otherwise blown away. So instead, I created a new function, gfs2_ri_total, that will re-read the rindex file just to total the file system space for the sake of the statfs_change. The ri_update will happen later, when gfs2 realizes the version number has changed, as it happened before my patch. Since the statfs_change is happening at write_commit time and there may be multiple writes to the rindex file for one grow operation. So one consequence of this restructuring is that instead of getting one kernel message to indicate the change, you may see several. For example, before when you did a gfs2_grow, you'd get a single message like: GFS2: File system extended by 247876 blocks (968MB) Now you get something like: GFS2: File system extended by 207896 blocks (812MB) GFS2: File system extended by 39980 blocks (156MB) This version has also been successfully run against the hours-long "gfs2_fsck_hellfire" test that does several gfs2_grow and gfs2_fsck while interjecting file system damage. It does this repeatedly under a variety Resource Group conditions. Signed-off-By: Bob Peterson <rpeterso@redhat.com> Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
2007-05-09 14:37:57 +00:00
/* Total up the file system space, according to the latest rindex. */
fs_total = gfs2_ri_total(sdp);
if (gfs2_meta_inode_buffer(m_ip, &m_bh) != 0)
goto out;
[GFS2] kernel changes to support new gfs2_grow command This is another revision of my gfs2 kernel patch that allows gfs2_grow to function properly. Steve Whitehouse expressed some concerns about the previous patch and I restructured it based on his comments. The previous patch was doing the statfs_change at file close time, under its own transaction. The current patch does the statfs_change inside the gfs2_commit_write function, which keeps it under the umbrella of the inode transaction. I can't call ri_update to re-read the rindex file during the transaction because the transaction may have outstanding unwritten buffers attached to the rgrps that would be otherwise blown away. So instead, I created a new function, gfs2_ri_total, that will re-read the rindex file just to total the file system space for the sake of the statfs_change. The ri_update will happen later, when gfs2 realizes the version number has changed, as it happened before my patch. Since the statfs_change is happening at write_commit time and there may be multiple writes to the rindex file for one grow operation. So one consequence of this restructuring is that instead of getting one kernel message to indicate the change, you may see several. For example, before when you did a gfs2_grow, you'd get a single message like: GFS2: File system extended by 247876 blocks (968MB) Now you get something like: GFS2: File system extended by 207896 blocks (812MB) GFS2: File system extended by 39980 blocks (156MB) This version has also been successfully run against the hours-long "gfs2_fsck_hellfire" test that does several gfs2_grow and gfs2_fsck while interjecting file system damage. It does this repeatedly under a variety Resource Group conditions. Signed-off-By: Bob Peterson <rpeterso@redhat.com> Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
2007-05-09 14:37:57 +00:00
spin_lock(&sdp->sd_statfs_spin);
gfs2_statfs_change_in(m_sc, m_bh->b_data +
sizeof(struct gfs2_dinode));
[GFS2] kernel changes to support new gfs2_grow command This is another revision of my gfs2 kernel patch that allows gfs2_grow to function properly. Steve Whitehouse expressed some concerns about the previous patch and I restructured it based on his comments. The previous patch was doing the statfs_change at file close time, under its own transaction. The current patch does the statfs_change inside the gfs2_commit_write function, which keeps it under the umbrella of the inode transaction. I can't call ri_update to re-read the rindex file during the transaction because the transaction may have outstanding unwritten buffers attached to the rgrps that would be otherwise blown away. So instead, I created a new function, gfs2_ri_total, that will re-read the rindex file just to total the file system space for the sake of the statfs_change. The ri_update will happen later, when gfs2 realizes the version number has changed, as it happened before my patch. Since the statfs_change is happening at write_commit time and there may be multiple writes to the rindex file for one grow operation. So one consequence of this restructuring is that instead of getting one kernel message to indicate the change, you may see several. For example, before when you did a gfs2_grow, you'd get a single message like: GFS2: File system extended by 247876 blocks (968MB) Now you get something like: GFS2: File system extended by 207896 blocks (812MB) GFS2: File system extended by 39980 blocks (156MB) This version has also been successfully run against the hours-long "gfs2_fsck_hellfire" test that does several gfs2_grow and gfs2_fsck while interjecting file system damage. It does this repeatedly under a variety Resource Group conditions. Signed-off-By: Bob Peterson <rpeterso@redhat.com> Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
2007-05-09 14:37:57 +00:00
if (fs_total > (m_sc->sc_total + l_sc->sc_total))
new_free = fs_total - (m_sc->sc_total + l_sc->sc_total);
else
new_free = 0;
spin_unlock(&sdp->sd_statfs_spin);
fs_warn(sdp, "File system extended by %llu blocks.\n",
(unsigned long long)new_free);
[GFS2] kernel changes to support new gfs2_grow command This is another revision of my gfs2 kernel patch that allows gfs2_grow to function properly. Steve Whitehouse expressed some concerns about the previous patch and I restructured it based on his comments. The previous patch was doing the statfs_change at file close time, under its own transaction. The current patch does the statfs_change inside the gfs2_commit_write function, which keeps it under the umbrella of the inode transaction. I can't call ri_update to re-read the rindex file during the transaction because the transaction may have outstanding unwritten buffers attached to the rgrps that would be otherwise blown away. So instead, I created a new function, gfs2_ri_total, that will re-read the rindex file just to total the file system space for the sake of the statfs_change. The ri_update will happen later, when gfs2 realizes the version number has changed, as it happened before my patch. Since the statfs_change is happening at write_commit time and there may be multiple writes to the rindex file for one grow operation. So one consequence of this restructuring is that instead of getting one kernel message to indicate the change, you may see several. For example, before when you did a gfs2_grow, you'd get a single message like: GFS2: File system extended by 247876 blocks (968MB) Now you get something like: GFS2: File system extended by 207896 blocks (812MB) GFS2: File system extended by 39980 blocks (156MB) This version has also been successfully run against the hours-long "gfs2_fsck_hellfire" test that does several gfs2_grow and gfs2_fsck while interjecting file system damage. It does this repeatedly under a variety Resource Group conditions. Signed-off-By: Bob Peterson <rpeterso@redhat.com> Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
2007-05-09 14:37:57 +00:00
gfs2_statfs_change(sdp, new_free, new_free, 0);
if (gfs2_meta_inode_buffer(l_ip, &l_bh) != 0)
goto out2;
update_statfs(sdp, m_bh, l_bh);
brelse(l_bh);
out2:
brelse(m_bh);
out:
sdp->sd_rindex_uptodate = 0;
gfs2_trans_end(sdp);
[GFS2] kernel changes to support new gfs2_grow command This is another revision of my gfs2 kernel patch that allows gfs2_grow to function properly. Steve Whitehouse expressed some concerns about the previous patch and I restructured it based on his comments. The previous patch was doing the statfs_change at file close time, under its own transaction. The current patch does the statfs_change inside the gfs2_commit_write function, which keeps it under the umbrella of the inode transaction. I can't call ri_update to re-read the rindex file during the transaction because the transaction may have outstanding unwritten buffers attached to the rgrps that would be otherwise blown away. So instead, I created a new function, gfs2_ri_total, that will re-read the rindex file just to total the file system space for the sake of the statfs_change. The ri_update will happen later, when gfs2 realizes the version number has changed, as it happened before my patch. Since the statfs_change is happening at write_commit time and there may be multiple writes to the rindex file for one grow operation. So one consequence of this restructuring is that instead of getting one kernel message to indicate the change, you may see several. For example, before when you did a gfs2_grow, you'd get a single message like: GFS2: File system extended by 247876 blocks (968MB) Now you get something like: GFS2: File system extended by 207896 blocks (812MB) GFS2: File system extended by 39980 blocks (156MB) This version has also been successfully run against the hours-long "gfs2_fsck_hellfire" test that does several gfs2_grow and gfs2_fsck while interjecting file system damage. It does this repeatedly under a variety Resource Group conditions. Signed-off-By: Bob Peterson <rpeterso@redhat.com> Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
2007-05-09 14:37:57 +00:00
}
/**
* jdata_set_page_dirty - Page dirtying function
* @page: The page to dirty
*
* Returns: 1 if it dirtyed the page, or 0 otherwise
*/
static int jdata_set_page_dirty(struct page *page)
{
gfs2: Only set PageChecked if we have a transaction With jdata writes, we frequently got into situations where gfs2 deadlocked because of this calling sequence: gfs2_ail1_start gfs2_ail1_flush - for every tr on the sd_ail1_list: gfs2_ail1_start_one - for every bd on the tr's tr_ail1_list: generic_writepages write_cache_pages passing __writepage() calls clear_page_dirty_for_io which calls set_page_dirty: which calls jdata_set_page_dirty which sets PageChecked. __writepage() calls mapping->a_ops->writepage AKA gfs2_jdata_writepage However, gfs2_jdata_writepage checks if PageChecked is set, and if so, it ignores the write and redirties the page. The problem is that write_cache_pages calls clear_page_dirty_for_io, which often calls set_page_dirty(). See comments in page-writeback.c starting with "Yes, Virginia". If it's jdata, set_page_dirty will call jdata_set_page_dirty which will set PageChecked. That causes a conflict because it makes it look like the page has been redirtied by another writer, in which case we need to skip writing it and redirty the page. That ends up in a deadlock because it isn't a "real" writer and nothing will ever clear PageChecked. If we do have a real writer, it will have started a transaction. So this patch checks if a transaction is in use, and if not, it skips setting PageChecked. That way, the page will be dirtied, cleaned, and written appropriately. Signed-off-by: Bob Peterson <rpeterso@redhat.com> Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
2020-08-19 13:55:18 +00:00
if (current->journal_info)
SetPageChecked(page);
return __set_page_dirty_buffers(page);
}
/**
* gfs2_bmap - Block map function
* @mapping: Address space info
* @lblock: The block to map
*
* Returns: The disk address for the block or 0 on hole or error
*/
static sector_t gfs2_bmap(struct address_space *mapping, sector_t lblock)
{
struct gfs2_inode *ip = GFS2_I(mapping->host);
struct gfs2_holder i_gh;
sector_t dblock = 0;
int error;
error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &i_gh);
if (error)
return 0;
if (!gfs2_is_stuffed(ip))
dblock = iomap_bmap(mapping, lblock, &gfs2_iomap_ops);
gfs2_glock_dq_uninit(&i_gh);
return dblock;
}
static void gfs2_discard(struct gfs2_sbd *sdp, struct buffer_head *bh)
{
struct gfs2_bufdata *bd;
lock_buffer(bh);
gfs2_log_lock(sdp);
clear_buffer_dirty(bh);
bd = bh->b_private;
if (bd) {
if (!list_empty(&bd->bd_list) && !buffer_pinned(bh))
list_del_init(&bd->bd_list);
else {
spin_lock(&sdp->sd_ail_lock);
gfs2_remove_from_journal(bh, REMOVE_JDATA);
spin_unlock(&sdp->sd_ail_lock);
}
}
bh->b_bdev = NULL;
clear_buffer_mapped(bh);
clear_buffer_req(bh);
clear_buffer_new(bh);
gfs2_log_unlock(sdp);
unlock_buffer(bh);
}
static void gfs2_invalidatepage(struct page *page, unsigned int offset,
unsigned int length)
{
struct gfs2_sbd *sdp = GFS2_SB(page->mapping->host);
unsigned int stop = offset + length;
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 12:29:47 +00:00
int partial_page = (offset || length < PAGE_SIZE);
struct buffer_head *bh, *head;
unsigned long pos = 0;
BUG_ON(!PageLocked(page));
if (!partial_page)
ClearPageChecked(page);
if (!page_has_buffers(page))
goto out;
bh = head = page_buffers(page);
do {
if (pos + bh->b_size > stop)
return;
if (offset <= pos)
gfs2_discard(sdp, bh);
pos += bh->b_size;
bh = bh->b_this_page;
} while (bh != head);
out:
if (!partial_page)
try_to_release_page(page, 0);
}
/**
* gfs2_releasepage - free the metadata associated with a page
* @page: the page that's being released
* @gfp_mask: passed from Linux VFS, ignored by us
*
* Calls try_to_free_buffers() to free the buffers and put the page if the
* buffers can be released.
*
* Returns: 1 if the page was put or else 0
*/
int gfs2_releasepage(struct page *page, gfp_t gfp_mask)
{
struct address_space *mapping = page->mapping;
struct gfs2_sbd *sdp = gfs2_mapping2sbd(mapping);
struct buffer_head *bh, *head;
struct gfs2_bufdata *bd;
if (!page_has_buffers(page))
return 0;
/*
* From xfs_vm_releasepage: mm accommodates an old ext3 case where
* clean pages might not have had the dirty bit cleared. Thus, it can
* send actual dirty pages to ->releasepage() via shrink_active_list().
*
* As a workaround, we skip pages that contain dirty buffers below.
* Once ->releasepage isn't called on dirty pages anymore, we can warn
* on dirty buffers like we used to here again.
*/
gfs2_log_lock(sdp);
head = bh = page_buffers(page);
do {
if (atomic_read(&bh->b_count))
goto cannot_release;
bd = bh->b_private;
if (bd && bd->bd_tr)
goto cannot_release;
if (buffer_dirty(bh) || WARN_ON(buffer_pinned(bh)))
goto cannot_release;
bh = bh->b_this_page;
} while(bh != head);
head = bh = page_buffers(page);
do {
bd = bh->b_private;
if (bd) {
gfs2_assert_warn(sdp, bd->bd_bh == bh);
bd->bd_bh = NULL;
bh->b_private = NULL;
/*
* The bd may still be queued as a revoke, in which
* case we must not dequeue nor free it.
*/
if (!bd->bd_blkno && !list_empty(&bd->bd_list))
list_del_init(&bd->bd_list);
if (list_empty(&bd->bd_list))
kmem_cache_free(gfs2_bufdata_cachep, bd);
}
bh = bh->b_this_page;
} while (bh != head);
gfs2_log_unlock(sdp);
return try_to_free_buffers(page);
cannot_release:
gfs2_log_unlock(sdp);
return 0;
}
static const struct address_space_operations gfs2_aops = {
.writepage = gfs2_writepage,
.writepages = gfs2_writepages,
.readpage = gfs2_readpage,
.readahead = gfs2_readahead,
.set_page_dirty = iomap_set_page_dirty,
.releasepage = iomap_releasepage,
.invalidatepage = iomap_invalidatepage,
.bmap = gfs2_bmap,
.direct_IO = noop_direct_IO,
.migratepage = iomap_migrate_page,
.is_partially_uptodate = iomap_is_partially_uptodate,
.error_remove_page = generic_error_remove_page,
};
static const struct address_space_operations gfs2_jdata_aops = {
.writepage = gfs2_jdata_writepage,
.writepages = gfs2_jdata_writepages,
.readpage = gfs2_readpage,
.readahead = gfs2_readahead,
.set_page_dirty = jdata_set_page_dirty,
.bmap = gfs2_bmap,
.invalidatepage = gfs2_invalidatepage,
.releasepage = gfs2_releasepage,
GFS2: Pagecache usage optimization on GFS2 I introduced "is_partially_uptodate" aops for GFS2. A page can have multiple buffers and even if a page is not uptodate, some buffers can be uptodate on pagesize != blocksize environment. This aops checks that all buffers which correspond to a part of a file that we want to read are uptodate. If so, we do not have to issue actual read IO to HDD even if a page is not uptodate because the portion we want to read are uptodate. "block_is_partially_uptodate" function is already used by ext2/3/4. With the following patch random read/write mixed workloads or random read after random write workloads can be optimized and we can get performance improvement. I did a performance test using the sysbench. #sysbench --num-threads=16 --max-requests=200000 --test=fileio --file-num=1 --file-block-size=8K --file-total-size=2G --file-test-mode=rndrw --file-fsync-freq=0 --file-rw-ratio=1 run -2.6.29-rc6 Test execution summary: total time: 202.6389s total number of events: 200000 total time taken by event execution: 2580.0480 per-request statistics: min: 0.0000s avg: 0.0129s max: 49.5852s approx. 95 percentile: 0.0462s -2.6.29-rc6-patched Test execution summary: total time: 177.8639s total number of events: 200000 total time taken by event execution: 2419.0199 per-request statistics: min: 0.0000s avg: 0.0121s max: 52.4306s approx. 95 percentile: 0.0444s arch: ia64 pagesize: 16k blocksize: 4k Signed-off-by: Hisashi Hifumi <hifumi.hisashi@oss.ntt.co.jp> Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
2009-03-03 02:45:20 +00:00
.is_partially_uptodate = block_is_partially_uptodate,
.error_remove_page = generic_error_remove_page,
};
void gfs2_set_aops(struct inode *inode)
{
if (gfs2_is_jdata(GFS2_I(inode)))
inode->i_mapping->a_ops = &gfs2_jdata_aops;
else
inode->i_mapping->a_ops = &gfs2_aops;
}