Subject: + mm-zswap-support-multiple-swap-devices.patch added to -mm tree
To: minchan@xxxxxxxxxx,bob.liu@xxxxxxxxxx,cai.liu@xxxxxxxxxxx,sjennings@xxxxxxxxxxxxxx,weijie.yang.kh@xxxxxxxxx
From: akpm@xxxxxxxxxxxxxxxxxxxx
Date: Tue, 25 Feb 2014 12:54:45 -0800
The patch titled
Subject: mm/zswap: support multiple swap devices
has been added to the -mm tree. Its filename is
mm-zswap-support-multiple-swap-devices.patch
This patch should soon appear at
http://ozlabs.org/~akpm/mmots/broken-out/mm-zswap-support-multiple-swap-devices.patch
and later at
http://ozlabs.org/~akpm/mmotm/broken-out/mm-zswap-support-multiple-swap-devices.patch
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/SubmitChecklist when testing your code ***
The -mm tree is included into linux-next and is updated
there every 3-4 working days
------------------------------------------------------
From: Minchan Kim <minchan@xxxxxxxxxx>
Subject: mm/zswap: support multiple swap devices
Cai Liu reporeted that now zbud pool pages counting has a problem when
multiple swap is used because it just counts only one swap intead of all
of swap so zswap cannot control writeback properly. The result is
unnecessary writeback or no writeback when we should really writeback.
IOW, it made zswap crazy.
Another problem in zswap is:
For example, let's assume we use two swap A and B with different priority
and A already has charged 19% long time ago and let's assume that A swap
is full now so VM start to use B so that B has charged 1% recently. It
menas zswap charged (19% + 1%) is full by default. Then, if VM want to
swap out more pages into B, zbud_reclaim_page would be evict one of pages
in B's pool and it would be repeated continuously. It's totally LRU
reverse problem and swap thrashing in B would happen.
This patch makes zswap consider mutliple swap by creating *a* zbud pool
which will be shared by multiple swap so all of zswap pages in multiple
swap keep order by LRU so it can prevent above two problems.
Signed-off-by: Minchan Kim <minchan@xxxxxxxxxx>
Reported-by: Cai Liu <cai.liu@xxxxxxxxxxx>
Suggested-by: Weijie Yang <weijie.yang.kh@xxxxxxxxx>
Cc: Seth Jennings <sjennings@xxxxxxxxxxxxxx>
Reviewed-by: Bob Liu <bob.liu@xxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
mm/zswap.c | 64 ++++++++++++++++++++++++++-------------------------
1 file changed, 33 insertions(+), 31 deletions(-)
diff -puN mm/zswap.c~mm-zswap-support-multiple-swap-devices mm/zswap.c
--- a/mm/zswap.c~mm-zswap-support-multiple-swap-devices
+++ a/mm/zswap.c
@@ -89,6 +89,9 @@ static unsigned int zswap_max_pool_perce
module_param_named(max_pool_percent,
zswap_max_pool_percent, uint, 0644);
+/* zbud_pool is shared by all of zswap backend */
+static struct zbud_pool *zswap_pool;
+
/*********************************
* compression functions
**********************************/
@@ -189,7 +192,6 @@ struct zswap_header {
struct zswap_tree {
struct rb_root rbroot;
spinlock_t lock;
- struct zbud_pool *pool;
};
static struct zswap_tree *zswap_trees[MAX_SWAPFILES];
@@ -285,13 +287,12 @@ static void zswap_rb_erase(struct rb_roo
* Carries out the common pattern of freeing and entry's zbud allocation,
* freeing the entry itself, and decrementing the number of stored pages.
*/
-static void zswap_free_entry(struct zswap_tree *tree,
- struct zswap_entry *entry)
+static void zswap_free_entry(struct zswap_entry *entry)
{
- zbud_free(tree->pool, entry->handle);
+ zbud_free(zswap_pool, entry->handle);
zswap_entry_cache_free(entry);
atomic_dec(&zswap_stored_pages);
- zswap_pool_pages = zbud_get_pool_size(tree->pool);
+ zswap_pool_pages = zbud_get_pool_size(zswap_pool);
}
/* caller must hold the tree lock */
@@ -311,7 +312,7 @@ static void zswap_entry_put(struct zswap
BUG_ON(refcount < 0);
if (refcount == 0) {
zswap_rb_erase(&tree->rbroot, entry);
- zswap_free_entry(tree, entry);
+ zswap_free_entry(entry);
}
}
@@ -545,7 +546,6 @@ static int zswap_writeback_entry(struct
zbud_unmap(pool, handle);
tree = zswap_trees[swp_type(swpentry)];
offset = swp_offset(swpentry);
- BUG_ON(pool != tree->pool);
/* find and ref zswap entry */
spin_lock(&tree->lock);
@@ -573,13 +573,13 @@ static int zswap_writeback_entry(struct
case ZSWAP_SWAPCACHE_NEW: /* page is locked */
/* decompress */
dlen = PAGE_SIZE;
- src = (u8 *)zbud_map(tree->pool, entry->handle) +
+ src = (u8 *)zbud_map(zswap_pool, entry->handle) +
sizeof(struct zswap_header);
dst = kmap_atomic(page);
ret = zswap_comp_op(ZSWAP_COMPOP_DECOMPRESS, src,
entry->length, dst, &dlen);
kunmap_atomic(dst);
- zbud_unmap(tree->pool, entry->handle);
+ zbud_unmap(zswap_pool, entry->handle);
BUG_ON(ret);
BUG_ON(dlen != PAGE_SIZE);
@@ -652,7 +652,7 @@ static int zswap_frontswap_store(unsigne
/* reclaim space if needed */
if (zswap_is_full()) {
zswap_pool_limit_hit++;
- if (zbud_reclaim_page(tree->pool, 8)) {
+ if (zbud_reclaim_page(zswap_pool, 8)) {
zswap_reject_reclaim_fail++;
ret = -ENOMEM;
goto reject;
@@ -679,7 +679,7 @@ static int zswap_frontswap_store(unsigne
/* store */
len = dlen + sizeof(struct zswap_header);
- ret = zbud_alloc(tree->pool, len, __GFP_NORETRY | __GFP_NOWARN,
+ ret = zbud_alloc(zswap_pool, len, __GFP_NORETRY | __GFP_NOWARN,
&handle);
if (ret == -ENOSPC) {
zswap_reject_compress_poor++;
@@ -689,11 +689,11 @@ static int zswap_frontswap_store(unsigne
zswap_reject_alloc_fail++;
goto freepage;
}
- zhdr = zbud_map(tree->pool, handle);
+ zhdr = zbud_map(zswap_pool, handle);
zhdr->swpentry = swp_entry(type, offset);
buf = (u8 *)(zhdr + 1);
memcpy(buf, dst, dlen);
- zbud_unmap(tree->pool, handle);
+ zbud_unmap(zswap_pool, handle);
put_cpu_var(zswap_dstmem);
/* populate entry */
@@ -716,7 +716,7 @@ static int zswap_frontswap_store(unsigne
/* update stats */
atomic_inc(&zswap_stored_pages);
- zswap_pool_pages = zbud_get_pool_size(tree->pool);
+ zswap_pool_pages = zbud_get_pool_size(zswap_pool);
return 0;
@@ -752,13 +752,13 @@ static int zswap_frontswap_load(unsigned
/* decompress */
dlen = PAGE_SIZE;
- src = (u8 *)zbud_map(tree->pool, entry->handle) +
+ src = (u8 *)zbud_map(zswap_pool, entry->handle) +
sizeof(struct zswap_header);
dst = kmap_atomic(page);
ret = zswap_comp_op(ZSWAP_COMPOP_DECOMPRESS, src, entry->length,
dst, &dlen);
kunmap_atomic(dst);
- zbud_unmap(tree->pool, entry->handle);
+ zbud_unmap(zswap_pool, entry->handle);
BUG_ON(ret);
spin_lock(&tree->lock);
@@ -804,11 +804,9 @@ static void zswap_frontswap_invalidate_a
/* walk the tree and free everything */
spin_lock(&tree->lock);
rbtree_postorder_for_each_entry_safe(entry, n, &tree->rbroot, rbnode)
- zswap_free_entry(tree, entry);
+ zswap_free_entry(entry);
tree->rbroot = RB_ROOT;
spin_unlock(&tree->lock);
-
- zbud_destroy_pool(tree->pool);
kfree(tree);
zswap_trees[type] = NULL;
}
@@ -822,20 +820,14 @@ static void zswap_frontswap_init(unsigne
struct zswap_tree *tree;
tree = kzalloc(sizeof(struct zswap_tree), GFP_KERNEL);
- if (!tree)
- goto err;
- tree->pool = zbud_create_pool(GFP_KERNEL, &zswap_zbud_ops);
- if (!tree->pool)
- goto freetree;
+ if (!tree) {
+ pr_err("alloc failed, zswap disabled for swap type %d\n", type);
+ return;
+ }
+
tree->rbroot = RB_ROOT;
spin_lock_init(&tree->lock);
zswap_trees[type] = tree;
- return;
-
-freetree:
- kfree(tree);
-err:
- pr_err("alloc failed, zswap disabled for swap type %d\n", type);
}
static struct frontswap_ops zswap_frontswap_ops = {
@@ -907,9 +899,16 @@ static int __init init_zswap(void)
return 0;
pr_info("loading zswap\n");
+
+ zswap_pool = zbud_create_pool(GFP_KERNEL, &zswap_zbud_ops);
+ if (!zswap_pool) {
+ pr_err("zbud pool creation failed\n");
+ goto error;
+ }
+
if (zswap_entry_cache_create()) {
pr_err("entry cache creation failed\n");
- goto error;
+ goto cachefail;
}
if (zswap_comp_init()) {
pr_err("compressor initialization failed\n");
@@ -919,6 +918,7 @@ static int __init init_zswap(void)
pr_err("per-cpu initialization failed\n");
goto pcpufail;
}
+
frontswap_register_ops(&zswap_frontswap_ops);
if (zswap_debugfs_init())
pr_warn("debugfs initialization failed\n");
@@ -927,6 +927,8 @@ pcpufail:
zswap_comp_exit();
compfail:
zswap_entry_cache_destory();
+cachefail:
+ zbud_destroy_pool(zswap_pool);
error:
return -ENOMEM;
}
_
Patches currently in -mm which might be from minchan@xxxxxxxxxx are
zram-avoid-null-access-when-fail-to-alloc-meta.patch
fs-cachefiles-use-add_to_page_cache_lru.patch
lib-radix-tree-radix_tree_delete_item.patch
mm-shmem-save-one-radix-tree-lookup-when-truncating-swapped-pages.patch
mm-filemap-move-radix-tree-hole-searching-here.patch
mm-fs-prepare-for-non-page-entries-in-page-cache-radix-trees.patch
mm-fs-store-shadow-entries-in-page-cache.patch
mm-thrash-detection-based-file-cache-sizing.patch
mm-keep-page-cache-radix-tree-nodes-in-check.patch
mm-keep-page-cache-radix-tree-nodes-in-check-fix.patch
mm-keep-page-cache-radix-tree-nodes-in-check-fix-fix.patch
zram-drop-init_done-struct-zram-member.patch
zram-do-not-pass-rw-argument-to-__zram_make_request.patch
zram-remove-good-and-bad-compress-stats.patch
zram-use-atomic64_t-for-all-zram-stats.patch
zram-remove-zram-stats-code-duplication.patch
zram-report-failed-read-and-write-stats.patch
zram-drop-not-used-table-count-member.patch
zram-move-zram-size-warning-to-documentation.patch
zram-document-failed_reads-failed_writes-stats.patch
zram-delete-zram_init_device.patch
mm-zswap-fix-trivial-typo-and-arrange-indentation.patch
mm-zswap-update-zsmalloc-in-comment-to-zbud.patch
mm-zswap-support-multiple-swap-devices.patch
debugging-keep-track-of-page-owners.patch
--
To unsubscribe from this list: send the line "unsubscribe mm-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
