mirror of
https://github.com/torvalds/linux.git
synced 2024-11-23 12:42:02 +00:00
063d99b4fa
Commit6afdb859b7
("mm: do not ignore mapping_gfp_mask in page cache allocation paths") has caught some users of hardcoded GFP_KERNEL used in the page cache allocation paths. This, however, wasn't complete and there were others which went unnoticed. Dave Chinner has reported the following deadlock for xfs on loop device: : With the recent merge of the loop device changes, I'm now seeing : XFS deadlock on my single CPU, 1GB RAM VM running xfs/073. : : The deadlocked is as follows: : : kloopd1: loop_queue_read_work : xfs_file_iter_read : lock XFS inode XFS_IOLOCK_SHARED (on image file) : page cache read (GFP_KERNEL) : radix tree alloc : memory reclaim : reclaim XFS inodes : log force to unpin inodes : <wait for log IO completion> : : xfs-cil/loop1: <does log force IO work> : xlog_cil_push : xlog_write : <loop issuing log writes> : xlog_state_get_iclog_space() : <blocks due to all log buffers under write io> : <waits for IO completion> : : kloopd1: loop_queue_write_work : xfs_file_write_iter : lock XFS inode XFS_IOLOCK_EXCL (on image file) : <wait for inode to be unlocked> : : i.e. the kloopd, with it's split read and write work queues, has : introduced a dependency through memory reclaim. i.e. that writes : need to be able to progress for reads make progress. : : The problem, fundamentally, is that mpage_readpages() does a : GFP_KERNEL allocation, rather than paying attention to the inode's : mapping gfp mask, which is set to GFP_NOFS. : : The didn't used to happen, because the loop device used to issue : reads through the splice path and that does: : : error = add_to_page_cache_lru(page, mapping, index, : GFP_KERNEL & mapping_gfp_mask(mapping)); This has changed by commitaa4d86163e
("block: loop: switch to VFS ITER_BVEC"). This patch changes mpage_readpage{s} to follow gfp mask set for the mapping. There are, however, other places which are doing basically the same. lustre:ll_dir_filler is doing GFP_KERNEL from the function which apparently uses GFP_NOFS for other allocations so let's make this consistent. cifs:readpages_get_pages is called from cifs_readpages and __cifs_readpages_from_fscache called from the same path obeys mapping gfp. ramfs_nommu_expand_for_mapping is hardcoding GFP_KERNEL as well regardless it uses mapping_gfp_mask for the page allocation. ext4_mpage_readpages is the called from the page cache allocation path same as read_pages and read_cache_pages As I've noticed in my previous post I cannot say I would be happy about sprinkling mapping_gfp_mask all over the place and it sounds like we should drop gfp_mask argument altogether and use it internally in __add_to_page_cache_locked that would require all the filesystems to use mapping gfp consistently which I am not sure is the case here. From a quick glance it seems that some file system use it all the time while others are selective. Signed-off-by: Michal Hocko <mhocko@suse.com> Reported-by: Dave Chinner <david@fromorbit.com> Cc: "Theodore Ts'o" <tytso@mit.edu> Cc: Ming Lei <ming.lei@canonical.com> Cc: Andreas Dilger <andreas.dilger@intel.com> Cc: Oleg Drokin <oleg.drokin@intel.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Christoph Hellwig <hch@lst.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
273 lines
6.9 KiB
C
273 lines
6.9 KiB
C
/* file-nommu.c: no-MMU version of ramfs
|
|
*
|
|
* Copyright (C) 2005 Red Hat, Inc. All Rights Reserved.
|
|
* Written by David Howells (dhowells@redhat.com)
|
|
*
|
|
* This program is free software; you can redistribute it and/or
|
|
* modify it under the terms of the GNU General Public License
|
|
* as published by the Free Software Foundation; either version
|
|
* 2 of the License, or (at your option) any later version.
|
|
*/
|
|
|
|
#include <linux/module.h>
|
|
#include <linux/fs.h>
|
|
#include <linux/mm.h>
|
|
#include <linux/pagemap.h>
|
|
#include <linux/highmem.h>
|
|
#include <linux/init.h>
|
|
#include <linux/string.h>
|
|
#include <linux/backing-dev.h>
|
|
#include <linux/ramfs.h>
|
|
#include <linux/pagevec.h>
|
|
#include <linux/mman.h>
|
|
#include <linux/sched.h>
|
|
#include <linux/slab.h>
|
|
|
|
#include <asm/uaccess.h>
|
|
#include "internal.h"
|
|
|
|
static int ramfs_nommu_setattr(struct dentry *, struct iattr *);
|
|
static unsigned long ramfs_nommu_get_unmapped_area(struct file *file,
|
|
unsigned long addr,
|
|
unsigned long len,
|
|
unsigned long pgoff,
|
|
unsigned long flags);
|
|
static int ramfs_nommu_mmap(struct file *file, struct vm_area_struct *vma);
|
|
|
|
static unsigned ramfs_mmap_capabilities(struct file *file)
|
|
{
|
|
return NOMMU_MAP_DIRECT | NOMMU_MAP_COPY | NOMMU_MAP_READ |
|
|
NOMMU_MAP_WRITE | NOMMU_MAP_EXEC;
|
|
}
|
|
|
|
const struct file_operations ramfs_file_operations = {
|
|
.mmap_capabilities = ramfs_mmap_capabilities,
|
|
.mmap = ramfs_nommu_mmap,
|
|
.get_unmapped_area = ramfs_nommu_get_unmapped_area,
|
|
.read_iter = generic_file_read_iter,
|
|
.write_iter = generic_file_write_iter,
|
|
.fsync = noop_fsync,
|
|
.splice_read = generic_file_splice_read,
|
|
.splice_write = iter_file_splice_write,
|
|
.llseek = generic_file_llseek,
|
|
};
|
|
|
|
const struct inode_operations ramfs_file_inode_operations = {
|
|
.setattr = ramfs_nommu_setattr,
|
|
.getattr = simple_getattr,
|
|
};
|
|
|
|
/*****************************************************************************/
|
|
/*
|
|
* add a contiguous set of pages into a ramfs inode when it's truncated from
|
|
* size 0 on the assumption that it's going to be used for an mmap of shared
|
|
* memory
|
|
*/
|
|
int ramfs_nommu_expand_for_mapping(struct inode *inode, size_t newsize)
|
|
{
|
|
unsigned long npages, xpages, loop;
|
|
struct page *pages;
|
|
unsigned order;
|
|
void *data;
|
|
int ret;
|
|
gfp_t gfp = mapping_gfp_mask(inode->i_mapping);
|
|
|
|
/* make various checks */
|
|
order = get_order(newsize);
|
|
if (unlikely(order >= MAX_ORDER))
|
|
return -EFBIG;
|
|
|
|
ret = inode_newsize_ok(inode, newsize);
|
|
if (ret)
|
|
return ret;
|
|
|
|
i_size_write(inode, newsize);
|
|
|
|
/* allocate enough contiguous pages to be able to satisfy the
|
|
* request */
|
|
pages = alloc_pages(gfp, order);
|
|
if (!pages)
|
|
return -ENOMEM;
|
|
|
|
/* split the high-order page into an array of single pages */
|
|
xpages = 1UL << order;
|
|
npages = (newsize + PAGE_SIZE - 1) >> PAGE_SHIFT;
|
|
|
|
split_page(pages, order);
|
|
|
|
/* trim off any pages we don't actually require */
|
|
for (loop = npages; loop < xpages; loop++)
|
|
__free_page(pages + loop);
|
|
|
|
/* clear the memory we allocated */
|
|
newsize = PAGE_SIZE * npages;
|
|
data = page_address(pages);
|
|
memset(data, 0, newsize);
|
|
|
|
/* attach all the pages to the inode's address space */
|
|
for (loop = 0; loop < npages; loop++) {
|
|
struct page *page = pages + loop;
|
|
|
|
ret = add_to_page_cache_lru(page, inode->i_mapping, loop,
|
|
gfp);
|
|
if (ret < 0)
|
|
goto add_error;
|
|
|
|
/* prevent the page from being discarded on memory pressure */
|
|
SetPageDirty(page);
|
|
SetPageUptodate(page);
|
|
|
|
unlock_page(page);
|
|
put_page(page);
|
|
}
|
|
|
|
return 0;
|
|
|
|
add_error:
|
|
while (loop < npages)
|
|
__free_page(pages + loop++);
|
|
return ret;
|
|
}
|
|
|
|
/*****************************************************************************/
|
|
/*
|
|
*
|
|
*/
|
|
static int ramfs_nommu_resize(struct inode *inode, loff_t newsize, loff_t size)
|
|
{
|
|
int ret;
|
|
|
|
/* assume a truncate from zero size is going to be for the purposes of
|
|
* shared mmap */
|
|
if (size == 0) {
|
|
if (unlikely(newsize >> 32))
|
|
return -EFBIG;
|
|
|
|
return ramfs_nommu_expand_for_mapping(inode, newsize);
|
|
}
|
|
|
|
/* check that a decrease in size doesn't cut off any shared mappings */
|
|
if (newsize < size) {
|
|
ret = nommu_shrink_inode_mappings(inode, size, newsize);
|
|
if (ret < 0)
|
|
return ret;
|
|
}
|
|
|
|
truncate_setsize(inode, newsize);
|
|
return 0;
|
|
}
|
|
|
|
/*****************************************************************************/
|
|
/*
|
|
* handle a change of attributes
|
|
* - we're specifically interested in a change of size
|
|
*/
|
|
static int ramfs_nommu_setattr(struct dentry *dentry, struct iattr *ia)
|
|
{
|
|
struct inode *inode = d_inode(dentry);
|
|
unsigned int old_ia_valid = ia->ia_valid;
|
|
int ret = 0;
|
|
|
|
/* POSIX UID/GID verification for setting inode attributes */
|
|
ret = inode_change_ok(inode, ia);
|
|
if (ret)
|
|
return ret;
|
|
|
|
/* pick out size-changing events */
|
|
if (ia->ia_valid & ATTR_SIZE) {
|
|
loff_t size = inode->i_size;
|
|
|
|
if (ia->ia_size != size) {
|
|
ret = ramfs_nommu_resize(inode, ia->ia_size, size);
|
|
if (ret < 0 || ia->ia_valid == ATTR_SIZE)
|
|
goto out;
|
|
} else {
|
|
/* we skipped the truncate but must still update
|
|
* timestamps
|
|
*/
|
|
ia->ia_valid |= ATTR_MTIME|ATTR_CTIME;
|
|
}
|
|
}
|
|
|
|
setattr_copy(inode, ia);
|
|
out:
|
|
ia->ia_valid = old_ia_valid;
|
|
return ret;
|
|
}
|
|
|
|
/*****************************************************************************/
|
|
/*
|
|
* try to determine where a shared mapping can be made
|
|
* - we require that:
|
|
* - the pages to be mapped must exist
|
|
* - the pages be physically contiguous in sequence
|
|
*/
|
|
static unsigned long ramfs_nommu_get_unmapped_area(struct file *file,
|
|
unsigned long addr, unsigned long len,
|
|
unsigned long pgoff, unsigned long flags)
|
|
{
|
|
unsigned long maxpages, lpages, nr, loop, ret;
|
|
struct inode *inode = file_inode(file);
|
|
struct page **pages = NULL, **ptr, *page;
|
|
loff_t isize;
|
|
|
|
if (!(flags & MAP_SHARED))
|
|
return addr;
|
|
|
|
/* the mapping mustn't extend beyond the EOF */
|
|
lpages = (len + PAGE_SIZE - 1) >> PAGE_SHIFT;
|
|
isize = i_size_read(inode);
|
|
|
|
ret = -EINVAL;
|
|
maxpages = (isize + PAGE_SIZE - 1) >> PAGE_SHIFT;
|
|
if (pgoff >= maxpages)
|
|
goto out;
|
|
|
|
if (maxpages - pgoff < lpages)
|
|
goto out;
|
|
|
|
/* gang-find the pages */
|
|
ret = -ENOMEM;
|
|
pages = kcalloc(lpages, sizeof(struct page *), GFP_KERNEL);
|
|
if (!pages)
|
|
goto out_free;
|
|
|
|
nr = find_get_pages(inode->i_mapping, pgoff, lpages, pages);
|
|
if (nr != lpages)
|
|
goto out_free_pages; /* leave if some pages were missing */
|
|
|
|
/* check the pages for physical adjacency */
|
|
ptr = pages;
|
|
page = *ptr++;
|
|
page++;
|
|
for (loop = lpages; loop > 1; loop--)
|
|
if (*ptr++ != page++)
|
|
goto out_free_pages;
|
|
|
|
/* okay - all conditions fulfilled */
|
|
ret = (unsigned long) page_address(pages[0]);
|
|
|
|
out_free_pages:
|
|
ptr = pages;
|
|
for (loop = nr; loop > 0; loop--)
|
|
put_page(*ptr++);
|
|
out_free:
|
|
kfree(pages);
|
|
out:
|
|
return ret;
|
|
}
|
|
|
|
/*****************************************************************************/
|
|
/*
|
|
* set up a mapping for shared memory segments
|
|
*/
|
|
static int ramfs_nommu_mmap(struct file *file, struct vm_area_struct *vma)
|
|
{
|
|
if (!(vma->vm_flags & VM_SHARED))
|
|
return -ENOSYS;
|
|
|
|
file_accessed(file);
|
|
vma->vm_ops = &generic_file_vm_ops;
|
|
return 0;
|
|
}
|