Revision 1aef882f023eb7c24d6d77f001bd0ba956fdd861 authored by Linus Torvalds on 24 April 2015, 14:08:41 UTC, committed by Linus Torvalds on 24 April 2015, 14:08:41 UTC
Pull xfs update from Dave Chinner: "This update contains: - RENAME_WHITEOUT support - conversion of per-cpu superblock accounting to use generic counters - new inode mmap lock so that we can lock page faults out of truncate, hole punch and other direct extent manipulation functions to avoid racing mmap writes from causing data corruption - rework of direct IO submission and completion to solve data corruption issue when running concurrent extending DIO writes. Also solves problem of running IO completion transactions in interrupt context during size extending AIO writes. - FALLOC_FL_INSERT_RANGE support for inserting holes into a file via direct extent manipulation to avoid needing to copy data within the file - attribute block header field overflow fix for 64k block size filesystems - Lots of changes to log messaging to be more informative and concise when errors occur. Also prevent a lot of unnecessary log spamming due to cascading failures in error conditions. - lots of cleanups and bug fixes One thing of note is the direct IO fixes that we merged last week after the window opened. Even though a little late, they fix a user reported data corruption and have been pretty well tested. I figured there was not much point waiting another 2 weeks for -rc1 to be released just so I could send them to you..." * tag 'xfs-for-linus-4.1-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/dgc/linux-xfs: (49 commits) xfs: using generic_file_direct_write() is unnecessary xfs: direct IO EOF zeroing needs to drain AIO xfs: DIO write completion size updates race xfs: DIO writes within EOF don't need an ioend xfs: handle DIO overwrite EOF update completion correctly xfs: DIO needs an ioend for writes xfs: move DIO mapping size calculation xfs: factor DIO write mapping from get_blocks xfs: unlock i_mutex in xfs_break_layouts xfs: kill unnecessary firstused overflow check on attr3 leaf removal xfs: use larger in-core attr firstused field and detect overflow xfs: pass attr geometry to attr leaf header conversion functions xfs: disallow ro->rw remount on norecovery mount xfs: xfs_shift_file_space can be static xfs: Add support FALLOC_FL_INSERT_RANGE for fallocate fs: Add support FALLOC_FL_INSERT_RANGE for fallocate xfs: Fix incorrect positive ENOMEM return xfs: xfs_mru_cache_insert() should use GFP_NOFS xfs: %pF is only for function pointers xfs: fix shadow warning in xfs_da3_root_split() ...
swap_cgroup.c
#include <linux/swap_cgroup.h>
#include <linux/vmalloc.h>
#include <linux/mm.h>
#include <linux/swapops.h> /* depends on mm.h include */
static DEFINE_MUTEX(swap_cgroup_mutex);
struct swap_cgroup_ctrl {
struct page **map;
unsigned long length;
spinlock_t lock;
};
static struct swap_cgroup_ctrl swap_cgroup_ctrl[MAX_SWAPFILES];
struct swap_cgroup {
unsigned short id;
};
#define SC_PER_PAGE (PAGE_SIZE/sizeof(struct swap_cgroup))
/*
* SwapCgroup implements "lookup" and "exchange" operations.
* In typical usage, this swap_cgroup is accessed via memcg's charge/uncharge
* against SwapCache. At swap_free(), this is accessed directly from swap.
*
* This means,
* - we have no race in "exchange" when we're accessed via SwapCache because
* SwapCache(and its swp_entry) is under lock.
* - When called via swap_free(), there is no user of this entry and no race.
* Then, we don't need lock around "exchange".
*
* TODO: we can push these buffers out to HIGHMEM.
*/
/*
* allocate buffer for swap_cgroup.
*/
static int swap_cgroup_prepare(int type)
{
struct page *page;
struct swap_cgroup_ctrl *ctrl;
unsigned long idx, max;
ctrl = &swap_cgroup_ctrl[type];
for (idx = 0; idx < ctrl->length; idx++) {
page = alloc_page(GFP_KERNEL | __GFP_ZERO);
if (!page)
goto not_enough_page;
ctrl->map[idx] = page;
}
return 0;
not_enough_page:
max = idx;
for (idx = 0; idx < max; idx++)
__free_page(ctrl->map[idx]);
return -ENOMEM;
}
static struct swap_cgroup *lookup_swap_cgroup(swp_entry_t ent,
struct swap_cgroup_ctrl **ctrlp)
{
pgoff_t offset = swp_offset(ent);
struct swap_cgroup_ctrl *ctrl;
struct page *mappage;
struct swap_cgroup *sc;
ctrl = &swap_cgroup_ctrl[swp_type(ent)];
if (ctrlp)
*ctrlp = ctrl;
mappage = ctrl->map[offset / SC_PER_PAGE];
sc = page_address(mappage);
return sc + offset % SC_PER_PAGE;
}
/**
* swap_cgroup_cmpxchg - cmpxchg mem_cgroup's id for this swp_entry.
* @ent: swap entry to be cmpxchged
* @old: old id
* @new: new id
*
* Returns old id at success, 0 at failure.
* (There is no mem_cgroup using 0 as its id)
*/
unsigned short swap_cgroup_cmpxchg(swp_entry_t ent,
unsigned short old, unsigned short new)
{
struct swap_cgroup_ctrl *ctrl;
struct swap_cgroup *sc;
unsigned long flags;
unsigned short retval;
sc = lookup_swap_cgroup(ent, &ctrl);
spin_lock_irqsave(&ctrl->lock, flags);
retval = sc->id;
if (retval == old)
sc->id = new;
else
retval = 0;
spin_unlock_irqrestore(&ctrl->lock, flags);
return retval;
}
/**
* swap_cgroup_record - record mem_cgroup for this swp_entry.
* @ent: swap entry to be recorded into
* @id: mem_cgroup to be recorded
*
* Returns old value at success, 0 at failure.
* (Of course, old value can be 0.)
*/
unsigned short swap_cgroup_record(swp_entry_t ent, unsigned short id)
{
struct swap_cgroup_ctrl *ctrl;
struct swap_cgroup *sc;
unsigned short old;
unsigned long flags;
sc = lookup_swap_cgroup(ent, &ctrl);
spin_lock_irqsave(&ctrl->lock, flags);
old = sc->id;
sc->id = id;
spin_unlock_irqrestore(&ctrl->lock, flags);
return old;
}
/**
* lookup_swap_cgroup_id - lookup mem_cgroup id tied to swap entry
* @ent: swap entry to be looked up.
*
* Returns ID of mem_cgroup at success. 0 at failure. (0 is invalid ID)
*/
unsigned short lookup_swap_cgroup_id(swp_entry_t ent)
{
return lookup_swap_cgroup(ent, NULL)->id;
}
int swap_cgroup_swapon(int type, unsigned long max_pages)
{
void *array;
unsigned long array_size;
unsigned long length;
struct swap_cgroup_ctrl *ctrl;
if (!do_swap_account)
return 0;
length = DIV_ROUND_UP(max_pages, SC_PER_PAGE);
array_size = length * sizeof(void *);
array = vzalloc(array_size);
if (!array)
goto nomem;
ctrl = &swap_cgroup_ctrl[type];
mutex_lock(&swap_cgroup_mutex);
ctrl->length = length;
ctrl->map = array;
spin_lock_init(&ctrl->lock);
if (swap_cgroup_prepare(type)) {
/* memory shortage */
ctrl->map = NULL;
ctrl->length = 0;
mutex_unlock(&swap_cgroup_mutex);
vfree(array);
goto nomem;
}
mutex_unlock(&swap_cgroup_mutex);
return 0;
nomem:
printk(KERN_INFO "couldn't allocate enough memory for swap_cgroup.\n");
printk(KERN_INFO
"swap_cgroup can be disabled by swapaccount=0 boot option\n");
return -ENOMEM;
}
void swap_cgroup_swapoff(int type)
{
struct page **map;
unsigned long i, length;
struct swap_cgroup_ctrl *ctrl;
if (!do_swap_account)
return;
mutex_lock(&swap_cgroup_mutex);
ctrl = &swap_cgroup_ctrl[type];
map = ctrl->map;
length = ctrl->length;
ctrl->map = NULL;
ctrl->length = 0;
mutex_unlock(&swap_cgroup_mutex);
if (map) {
for (i = 0; i < length; i++) {
struct page *page = map[i];
if (page)
__free_page(page);
}
vfree(map);
}
}
Computing file changes ...