[ Upstream commit a7ab400d6fe73d0119fdc234e9982a6f80faea9f ]
During our recent testing with fadvise(FADV_DONTNEED), we find that if
given offset/length is not page-aligned, the last page will not be
discarded. The tool we use is vmtouch (https://hoytech.com/vmtouch/),
we map a 10KB-sized file into memory and then try to run this tool to
evict the whole file mapping, but the last single page always remains
staying in the memory:
$./vmtouch -e test_10K
Files: 1
Directories: 0
Evicted Pages: 3 (12K)
Elapsed: 2.1e-05 seconds
$./vmtouch test_10K
Files: 1
Directories: 0
Resident Pages: 1/3 4K/12K 33.3%
Elapsed: 5.5e-05 seconds
However when we test with an older kernel, say 3.10, this problem is
gone. So we wonder if this is a regression:
$./vmtouch -e test_10K
Files: 1
Directories: 0
Evicted Pages: 3 (12K)
Elapsed: 8.2e-05 seconds
$./vmtouch test_10K
Files: 1
Directories: 0
Resident Pages: 0/3 0/12K 0% <-- partial page also discarded
Elapsed: 5e-05 seconds
After digging a little bit into this problem, we find it seems not a
regression. Not discarding partial page is likely to be on purpose
according to commit 441c228f817f ("mm: fadvise: document the
fadvise(FADV_DONTNEED) behaviour for partial pages") written by Mel
Gorman. He explained why partial pages should be preserved instead of
being discarded when using fadvise(FADV_DONTNEED).
However, the interesting part is that the actual code did NOT work as
the same as it was described, the partial page was still discarded
anyway, due to a calculation mistake of `end_index' passed to
invalidate_mapping_pages(). This mistake has not been fixed until
recently, that's why we fail to reproduce our problem in old kernels.
The fix is done in commit 18aba41cbf ("mm/fadvise.c: do not discard
partial pages with POSIX_FADV_DONTNEED") by Oleg Drokin.
Back to the original testing, our problem becomes that there is a
special case that, if the page-unaligned `endbyte' is also the end of
file, it is not necessary at all to preserve the last partial page, as
we all know no one else will use the rest of it. It should be safe
enough if we just discard the whole page. So we add an EOF check in
this patch.
We also find a poosbile real world issue in mainline kernel. Assume
such scenario: A userspace backup application want to backup a huge
amount of small files (<4k) at once, the developer might (I guess) want
to use fadvise(FADV_DONTNEED) to save memory. However, FADV_DONTNEED
won't really happen since the only page mapped is a partial page, and
kernel will preserve it. Our patch also fixes this problem, since we
know the endbyte is EOF, so we discard it.
Here is a simple reproducer to reproduce and verify each scenario we
described above:
test_fadvise.c
==============================
#include <sys/mman.h>
#include <sys/stat.h>
#include <fcntl.h>
#include <stdlib.h>
#include <string.h>
#include <stdio.h>
#include <unistd.h>
int main(int argc, char **argv)
{
int i, fd, ret, len;
struct stat buf;
void *addr;
unsigned char *vec;
char *strbuf;
ssize_t pagesize = getpagesize();
ssize_t filesize;
fd = open(argv[1], O_RDWR|O_CREAT, S_IRUSR|S_IWUSR);
if (fd < 0)
return -1;
filesize = strtoul(argv[2], NULL, 10);
strbuf = malloc(filesize);
memset(strbuf, 42, filesize);
write(fd, strbuf, filesize);
free(strbuf);
fsync(fd);
len = (filesize + pagesize - 1) / pagesize;
printf("length of pages: %d\n", len);
addr = mmap(NULL, filesize, PROT_READ, MAP_SHARED, fd, 0);
if (addr == MAP_FAILED)
return -1;
ret = posix_fadvise(fd, 0, filesize, POSIX_FADV_DONTNEED);
if (ret < 0)
return -1;
vec = malloc(len);
ret = mincore(addr, filesize, (void *)vec);
if (ret < 0)
return -1;
for (i = 0; i < len; i++)
printf("pages[%d]: %x\n", i, vec[i] & 0x1);
free(vec);
close(fd);
return 0;
}
==============================
Test 1: running on kernel with commit 18aba41cbf reverted:
[root@caspar ~]# uname -r
4.15.0-rc6.revert+
[root@caspar ~]# ./test_fadvise file1 1024
length of pages: 1
pages[0]: 0 # <-- partial page discarded
[root@caspar ~]# ./test_fadvise file2 8192
length of pages: 2
pages[0]: 0
pages[1]: 0
[root@caspar ~]# ./test_fadvise file3 10240
length of pages: 3
pages[0]: 0
pages[1]: 0
pages[2]: 0 # <-- partial page discarded
Test 2: running on mainline kernel:
[root@caspar ~]# uname -r
4.15.0-rc6+
[root@caspar ~]# ./test_fadvise test1 1024
length of pages: 1
pages[0]: 1 # <-- partial and the only page not discarded
[root@caspar ~]# ./test_fadvise test2 8192
length of pages: 2
pages[0]: 0
pages[1]: 0
[root@caspar ~]# ./test_fadvise test3 10240
length of pages: 3
pages[0]: 0
pages[1]: 0
pages[2]: 1 # <-- partial page not discarded
Test 3: running on kernel with this patch:
[root@caspar ~]# uname -r
4.15.0-rc6.patched+
[root@caspar ~]# ./test_fadvise test1 1024
length of pages: 1
pages[0]: 0 # <-- partial page and EOF, discarded
[root@caspar ~]# ./test_fadvise test2 8192
length of pages: 2
pages[0]: 0
pages[1]: 0
[root@caspar ~]# ./test_fadvise test3 10240
length of pages: 3
pages[0]: 0
pages[1]: 0
pages[2]: 0 # <-- partial page and EOF, discarded
[akpm@linux-foundation.org: tweak code comment]
Link: http://lkml.kernel.org/r/5222da9ee20e1695eaabb69f631f200d6e6b8876.1515132470.git.jinli.zjl@alibaba-inc.com
Signed-off-by: shidao.ytt <shidao.ytt@alibaba-inc.com>
Signed-off-by: Caspar Zhang <jinli.zjl@alibaba-inc.com>
Reviewed-by: Oliver Yang <zhiche.yy@alibaba-inc.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Sasha Levin <alexander.levin@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
I noticed that the logic in the fadvise64_64 syscall is incorrect for partial pages. While first page of the region is correctly skipped if it is partial, the last page of the region is mistakenly discarded. This leads to problems for applications that read data in non-page-aligned chunks discarding already processed data between the reads. A somewhat misguided application that does something like write(XX bytes (non-page-alligned)); drop the data it just wrote; repeat gets a significant penalty in performance as a result. Link: http://lkml.kernel.org/r/1464917140-1506698-1-git-send-email-green@linuxhacker.ru Signed-off-by: Oleg Drokin <green@linuxhacker.ru> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
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>
In several places, bdi_congested() and its wrappers are used to
determine whether more IOs should be issued. With cgroup writeback
support, this question can't be answered solely based on the bdi
(backing_dev_info). It's dependent on whether the filesystem and bdi
support cgroup writeback and the blkcg the inode is associated with.
This patch implements inode_congested() and its wrappers which take
@inode and determines the congestion state considering cgroup
writeback. The new functions replace bdi_*congested() calls in places
where the query is about specific inode and task.
There are several filesystem users which also fit this criteria but
they should be updated when each filesystem implements cgroup
writeback support.
v2: Now that a given inode is associated with only one wb, congestion
state can be determined independent from the asking task. Drop
@task. Spotted by Vivek. Also, converted to take @inode instead
of @mapping and renamed to inode_congested().
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Jan Kara <jack@suse.cz>
Cc: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
All callers of get_xip_mem() are now gone. Remove checks for it, initialisers of it, documentation of it and the only implementation of it. Also remove mm/filemap_xip.c as it is now empty. Also remove documentation of the long-gone get_xip_page(). Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com> Cc: Andreas Dilger <andreas.dilger@intel.com> Cc: Boaz Harrosh <boaz@plexistor.com> Cc: Christoph Hellwig <hch@lst.de> Cc: Dave Chinner <david@fromorbit.com> Cc: Jan Kara <jack@suse.cz> Cc: Jens Axboe <axboe@kernel.dk> Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com> Cc: Randy Dunlap <rdunlap@infradead.org> Cc: Ross Zwisler <ross.zwisler@linux.intel.com> Cc: Theodore Ts'o <tytso@mit.edu> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Now that we got rid of the bdi abuse on character devices we can always use sb->s_bdi to get at the backing_dev_info for a file, except for the block device special case. Export inode_to_bdi and replace uses of mapping->backing_dev_info with it to prepare for the removal of mapping->backing_dev_info. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Tejun Heo <tj@kernel.org> Reviewed-by: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
A random seek IO benchmark appeared to regress because of a change to readahead but the real problem was the benchmark. To ensure the IO request accesssed disk, it used fadvise(FADV_DONTNEED) on a block boundary (512K) but the hint is ignored by the kernel. This is correct but not necessarily obvious behaviour. As much as I dislike comment patches, the explanation for this behaviour predates current git history. Clarify why it behaves like this in case someone "fixes" fadvise or readahead for the wrong reasons. Signed-off-by: Mel Gorman <mgorman@suse.de> Cc: Michael Kerrisk <mtk.manpages@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
... and convert a bunch of SYSCALL_DEFINE ones to SYSCALL_DEFINE<n>, killing the boilerplate crap around them. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Pull vfs pile (part one) from Al Viro: "Assorted stuff - cleaning namei.c up a bit, fixing ->d_name/->d_parent locking violations, etc. The most visible changes here are death of FS_REVAL_DOT (replaced with "has ->d_weak_revalidate()") and a new helper getting from struct file to inode. Some bits of preparation to xattr method interface changes. Misc patches by various people sent this cycle *and* ocfs2 fixes from several cycles ago that should've been upstream right then. PS: the next vfs pile will be xattr stuff." * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (46 commits) saner proc_get_inode() calling conventions proc: avoid extra pde_put() in proc_fill_super() fs: change return values from -EACCES to -EPERM fs/exec.c: make bprm_mm_init() static ocfs2/dlm: use GFP_ATOMIC inside a spin_lock ocfs2: fix possible use-after-free with AIO ocfs2: Fix oops in ocfs2_fast_symlink_readpage() code path get_empty_filp()/alloc_file() leave both ->f_pos and ->f_version zero target: writev() on single-element vector is pointless export kernel_write(), convert open-coded instances fs: encode_fh: return FILEID_INVALID if invalid fid_type kill f_vfsmnt vfs: kill FS_REVAL_DOT by adding a d_weak_revalidate dentry op nfsd: handle vfs_getattr errors in acl protocol switch vfs_getattr() to struct path default SET_PERSONALITY() in linux/elf.h ceph: prepopulate inodes only when request is aborted d_hash_and_lookup(): export, switch open-coded instances 9p: switch v9fs_set_create_acl() to inode+fid, do it before d_instantiate() 9p: split dropping the acls from v9fs_set_create_acl() ...
Rob van der Heij reported the following (paraphrased) on private mail.
The scenario is that I want to avoid backups to fill up the page
cache and purge stuff that is more likely to be used again (this is
with s390x Linux on z/VM, so I don't give it as much memory that
we don't care anymore). So I have something with LD_PRELOAD that
intercepts the close() call (from tar, in this case) and issues
a posix_fadvise() just before closing the file.
This mostly works, except for small files (less than 14 pages)
that remains in page cache after the face.
Unfortunately Rob has not had a chance to test this exact patch but the
test program below should be reproducing the problem he described.
The issue is the per-cpu pagevecs for LRU additions. If the pages are
added by one CPU but fadvise() is called on another then the pages
remain resident as the invalidate_mapping_pages() only drains the local
pagevecs via its call to pagevec_release(). The user-visible effect is
that a program that uses fadvise() properly is not obeyed.
A possible fix for this is to put the necessary smarts into
invalidate_mapping_pages() to globally drain the LRU pagevecs if a
pagevec page could not be discarded. The downside with this is that an
inode cache shrink would send a global IPI and memory pressure
potentially causing global IPI storms is very undesirable.
Instead, this patch adds a check during fadvise(POSIX_FADV_DONTNEED) to
check if invalidate_mapping_pages() discarded all the requested pages.
If a subset of pages are discarded it drains the LRU pagevecs and tries
again. If the second attempt fails, it assumes it is due to the pages
being mapped, locked or dirty and does not care. With this patch, an
application using fadvise() correctly will be obeyed but there is a
downside that a malicious application can force the kernel to send
global IPIs and increase overhead.
If accepted, I would like this to be considered as a -stable candidate.
It's not an urgent issue but it's a system call that is not working as
advertised which is weak.
The following test program demonstrates the problem. It should never
report that pages are still resident but will without this patch. It
assumes that CPU 0 and 1 exist.
int main() {
int fd;
int pagesize = getpagesize();
ssize_t written = 0, expected;
char *buf;
unsigned char *vec;
int resident, i;
cpu_set_t set;
/* Prepare a buffer for writing */
expected = FILESIZE_PAGES * pagesize;
buf = malloc(expected + 1);
if (buf == NULL) {
printf("ENOMEM\n");
exit(EXIT_FAILURE);
}
buf[expected] = 0;
memset(buf, 'a', expected);
/* Prepare the mincore vec */
vec = malloc(FILESIZE_PAGES);
if (vec == NULL) {
printf("ENOMEM\n");
exit(EXIT_FAILURE);
}
/* Bind ourselves to CPU 0 */
CPU_ZERO(&set);
CPU_SET(0, &set);
if (sched_setaffinity(getpid(), sizeof(set), &set) == -1) {
perror("sched_setaffinity");
exit(EXIT_FAILURE);
}
/* open file, unlink and write buffer */
fd = open("fadvise-test-file", O_CREAT|O_EXCL|O_RDWR);
if (fd == -1) {
perror("open");
exit(EXIT_FAILURE);
}
unlink("fadvise-test-file");
while (written < expected) {
ssize_t this_write;
this_write = write(fd, buf + written, expected - written);
if (this_write == -1) {
perror("write");
exit(EXIT_FAILURE);
}
written += this_write;
}
free(buf);
/*
* Force ourselves to another CPU. If fadvise only flushes the local
* CPUs pagevecs then the fadvise will fail to discard all file pages
*/
CPU_ZERO(&set);
CPU_SET(1, &set);
if (sched_setaffinity(getpid(), sizeof(set), &set) == -1) {
perror("sched_setaffinity");
exit(EXIT_FAILURE);
}
/* sync and fadvise to discard the page cache */
fsync(fd);
if (posix_fadvise(fd, 0, expected, POSIX_FADV_DONTNEED) == -1) {
perror("posix_fadvise");
exit(EXIT_FAILURE);
}
/* map the file and use mincore to see which parts of it are resident */
buf = mmap(NULL, expected, PROT_READ, MAP_SHARED, fd, 0);
if (buf == NULL) {
perror("mmap");
exit(EXIT_FAILURE);
}
if (mincore(buf, expected, vec) == -1) {
perror("mincore");
exit(EXIT_FAILURE);
}
/* Check residency */
for (i = 0, resident = 0; i < FILESIZE_PAGES; i++) {
if (vec[i])
resident++;
}
if (resident != 0) {
printf("Nr unexpected pages resident: %d\n", resident);
exit(EXIT_FAILURE);
}
munmap(buf, expected);
close(fd);
free(vec);
exit(EXIT_SUCCESS);
}
Signed-off-by: Mel Gorman <mgorman@suse.de>
Reported-by: Rob van der Heij <rvdheij@gmail.com>
Tested-by: Rob van der Heij <rvdheij@gmail.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>