SHM_UNLOCK: fix long unpreemptible section

scan_mapping_unevictable_pages() is used to make SysV SHM_LOCKed pages
evictable again once the shared memory is unlocked.  It does this with
pagevec_lookup()s across the whole object (which might occupy most of
memory), and takes 300ms to unlock 7GB here.  A cond_resched() every
PAGEVEC_SIZE pages would be good.

However, KOSAKI-san points out that this is called under shmem.c's
info->lock, and it's also under shm.c's shm_lock(), both spinlocks.
There is no strong reason for that: we need to take these pages off the
unevictable list soonish, but those locks are not required for it.

So move the call to scan_mapping_unevictable_pages() from shmem.c's
unlock handling up to shm.c's unlock handling.  Remove the recently
added barrier, not needed now we have spin_unlock() before the scan.

Use get_file(), with subsequent fput(), to make sure we have a reference
to mapping throughout scan_mapping_unevictable_pages(): that's something
that was previously guaranteed by the shm_lock().

Remove shmctl's lru_add_drain_all(): we don't fault in pages at SHM_LOCK
time, and we lazily discover them to be Unevictable later, so it serves
no purpose for SHM_LOCK; and serves no purpose for SHM_UNLOCK, since
pages still on pagevec are not marked Unevictable.

The original code avoided redundant rescans by checking VM_LOCKED flag
at its level: now avoid them by checking shp's SHM_LOCKED.

The original code called scan_mapping_unevictable_pages() on a locked
area at shm_destroy() time: perhaps we once had accounting cross-checks
which required that, but not now, so skip the overhead and just let
inode eviction deal with them.

Put check_move_unevictable_page() and scan_mapping_unevictable_pages()
under CONFIG_SHMEM (with stub for the TINY case when ramfs is used),
more as comment than to save space; comment them used for SHM_UNLOCK.

Signed-off-by: Hugh Dickins <hughd@google.com>
Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Minchan Kim <minchan.kim@gmail.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Shaohua Li <shaohua.li@intel.com>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michel Lespinasse <walken@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This commit is contained in:
Hugh Dickins 2012-01-20 14:34:19 -08:00 committed by Linus Torvalds
parent cb78edfdce
commit 85046579bd
3 changed files with 33 additions and 23 deletions

View file

@ -870,9 +870,7 @@ SYSCALL_DEFINE3(shmctl, int, shmid, int, cmd, struct shmid_ds __user *, buf)
case SHM_LOCK:
case SHM_UNLOCK:
{
struct file *uninitialized_var(shm_file);
lru_add_drain_all(); /* drain pagevecs to lru lists */
struct file *shm_file;
shp = shm_lock_check(ns, shmid);
if (IS_ERR(shp)) {
@ -895,22 +893,31 @@ SYSCALL_DEFINE3(shmctl, int, shmid, int, cmd, struct shmid_ds __user *, buf)
err = security_shm_shmctl(shp, cmd);
if (err)
goto out_unlock;
if(cmd==SHM_LOCK) {
shm_file = shp->shm_file;
if (is_file_hugepages(shm_file))
goto out_unlock;
if (cmd == SHM_LOCK) {
struct user_struct *user = current_user();
if (!is_file_hugepages(shp->shm_file)) {
err = shmem_lock(shp->shm_file, 1, user);
if (!err && !(shp->shm_perm.mode & SHM_LOCKED)){
shp->shm_perm.mode |= SHM_LOCKED;
shp->mlock_user = user;
}
err = shmem_lock(shm_file, 1, user);
if (!err && !(shp->shm_perm.mode & SHM_LOCKED)) {
shp->shm_perm.mode |= SHM_LOCKED;
shp->mlock_user = user;
}
} else if (!is_file_hugepages(shp->shm_file)) {
shmem_lock(shp->shm_file, 0, shp->mlock_user);
shp->shm_perm.mode &= ~SHM_LOCKED;
shp->mlock_user = NULL;
goto out_unlock;
}
/* SHM_UNLOCK */
if (!(shp->shm_perm.mode & SHM_LOCKED))
goto out_unlock;
shmem_lock(shm_file, 0, shp->mlock_user);
shp->shm_perm.mode &= ~SHM_LOCKED;
shp->mlock_user = NULL;
get_file(shm_file);
shm_unlock(shp);
scan_mapping_unevictable_pages(shm_file->f_mapping);
fput(shm_file);
goto out;
}
case IPC_RMID:

View file

@ -1068,13 +1068,6 @@ int shmem_lock(struct file *file, int lock, struct user_struct *user)
user_shm_unlock(inode->i_size, user);
info->flags &= ~VM_LOCKED;
mapping_clear_unevictable(file->f_mapping);
/*
* Ensure that a racing putback_lru_page() can see
* the pages of this mapping are evictable when we
* skip them due to !PageLRU during the scan.
*/
smp_mb__after_clear_bit();
scan_mapping_unevictable_pages(file->f_mapping);
}
retval = 0;

View file

@ -3499,6 +3499,7 @@ int page_evictable(struct page *page, struct vm_area_struct *vma)
return 1;
}
#ifdef CONFIG_SHMEM
/**
* check_move_unevictable_page - check page for evictability and move to appropriate zone lru list
* @page: page to check evictability and move to appropriate lru list
@ -3509,6 +3510,8 @@ int page_evictable(struct page *page, struct vm_area_struct *vma)
*
* Restrictions: zone->lru_lock must be held, page must be on LRU and must
* have PageUnevictable set.
*
* This function is only used for SysV IPC SHM_UNLOCK.
*/
static void check_move_unevictable_page(struct page *page, struct zone *zone)
{
@ -3545,6 +3548,8 @@ static void check_move_unevictable_page(struct page *page, struct zone *zone)
*
* Scan all pages in mapping. Check unevictable pages for
* evictability and move them to the appropriate zone lru list.
*
* This function is only used for SysV IPC SHM_UNLOCK.
*/
void scan_mapping_unevictable_pages(struct address_space *mapping)
{
@ -3590,9 +3595,14 @@ void scan_mapping_unevictable_pages(struct address_space *mapping)
pagevec_release(&pvec);
count_vm_events(UNEVICTABLE_PGSCANNED, pg_scanned);
cond_resched();
}
}
#else
void scan_mapping_unevictable_pages(struct address_space *mapping)
{
}
#endif /* CONFIG_SHMEM */
static void warn_scan_unevictable_pages(void)
{