All of lore.kernel.org
 help / color / mirror / Atom feed
From: Gao Xiang <gaoxiang25@huawei.com>
To: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	<devel@driverdev.osuosl.org>
Cc: <linux-fsdevel@vger.kernel.org>, <linux-kernel@vger.kernel.org>,
	<linux-erofs@lists.ozlabs.org>, <yuchao0@huawei.com>,
	<miaoxie@huawei.com>, <weidu.du@huawei.com>, <hsiangkao@aol.com>,
	<chao@kernel.org>, Gao Xiang <gaoxiang25@huawei.com>
Subject: [PATCH 24/25] staging: erofs: introduce cached decompression
Date: Thu, 26 Jul 2018 20:22:07 +0800	[thread overview]
Message-ID: <1532607728-103372-25-git-send-email-gaoxiang25@huawei.com> (raw)
In-Reply-To: <1532607728-103372-1-git-send-email-gaoxiang25@huawei.com>

This patch adds an optional choice which can be
enabled by users in order to cache both incomplete
ends of compressed clusters as a complement to
the in-place decompression in order to boost random
read, but it costs more memory than the in-place
decompression only.

Signed-off-by: Gao Xiang <gaoxiang25@huawei.com>
---
 drivers/staging/erofs/Kconfig     |  38 ++++++
 drivers/staging/erofs/internal.h  |  26 ++++
 drivers/staging/erofs/super.c     |  73 ++++++++++
 drivers/staging/erofs/unzip_vle.c | 274 ++++++++++++++++++++++++++++++++++++++
 drivers/staging/erofs/utils.c     |  17 ++-
 5 files changed, 427 insertions(+), 1 deletion(-)

diff --git a/drivers/staging/erofs/Kconfig b/drivers/staging/erofs/Kconfig
index b55ce1c..663b755 100644
--- a/drivers/staging/erofs/Kconfig
+++ b/drivers/staging/erofs/Kconfig
@@ -101,3 +101,41 @@ config EROFS_FS_CLUSTER_PAGE_LIMIT
 	  than 2. Otherwise, the image cannot be mounted
 	  correctly on this kernel.
 
+choice
+	prompt "EROFS VLE Data Decompression mode"
+	depends on EROFS_FS_ZIP
+	default EROFS_FS_ZIP_CACHE_BIPOLAR
+	help
+	  EROFS supports three options for VLE decompression.
+	  "In-place Decompression Only" consumes the minimum memory
+	  with lowest random read.
+
+	  "Bipolar Cached Decompression" consumes the maximum memory
+	  with highest random read.
+
+	  If unsure, select "Bipolar Cached Decompression"
+
+config EROFS_FS_ZIP_NO_CACHE
+	bool "In-place Decompression Only"
+	help
+	  Read compressed data into page cache and do in-place
+	  decompression directly.
+
+config EROFS_FS_ZIP_CACHE_UNIPOLAR
+	bool "Unipolar Cached Decompression"
+	help
+	  For each request, it caches the last compressed page
+	  for further reading.
+	  It still decompresses in place for the rest compressed pages.
+
+config EROFS_FS_ZIP_CACHE_BIPOLAR
+	bool "Bipolar Cached Decompression"
+	help
+	  For each request, it caches the both end compressed pages
+	  for further reading.
+	  It still decompresses in place for the rest compressed pages.
+
+	  Recommended for performance priority.
+
+endchoice
+
diff --git a/drivers/staging/erofs/internal.h b/drivers/staging/erofs/internal.h
index 3adec7d..669f93a 100644
--- a/drivers/staging/erofs/internal.h
+++ b/drivers/staging/erofs/internal.h
@@ -58,6 +58,18 @@ struct erofs_fault_info {
 };
 #endif
 
+#ifdef CONFIG_EROFS_FS_ZIP_CACHE_BIPOLAR
+#define EROFS_FS_ZIP_CACHE_LVL	(2)
+#elif defined(EROFS_FS_ZIP_CACHE_UNIPOLAR)
+#define EROFS_FS_ZIP_CACHE_LVL	(1)
+#else
+#define EROFS_FS_ZIP_CACHE_LVL	(0)
+#endif
+
+#if (!defined(EROFS_FS_HAS_MANAGED_CACHE) && (EROFS_FS_ZIP_CACHE_LVL > 0))
+#define EROFS_FS_HAS_MANAGED_CACHE
+#endif
+
 /* EROFS_SUPER_MAGIC_V1 to represent the whole file system */
 #define EROFS_SUPER_MAGIC   EROFS_SUPER_MAGIC_V1
 
@@ -82,6 +94,11 @@ struct erofs_sb_info {
 
 	/* the dedicated workstation for compression */
 	struct radix_tree_root workstn_tree;
+
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+	struct inode *managed_cache;
+#endif
+
 #endif
 
 	u32 build_time_nsec;
@@ -240,6 +257,15 @@ static inline void erofs_workstation_cleanup_all(struct super_block *sb)
 	erofs_shrink_workstation(EROFS_SB(sb), ~0UL, true);
 }
 
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+#define EROFS_UNALLOCATED_CACHED_PAGE	((void *)0x5F0EF00D)
+
+extern int try_to_free_all_cached_pages(struct erofs_sb_info *sbi,
+	struct erofs_workgroup *egrp);
+extern int try_to_free_cached_page(struct address_space *mapping,
+	struct page *page);
+#endif
+
 #endif
 
 /* we strictly follow PAGE_SIZE and no buffer head yet */
diff --git a/drivers/staging/erofs/super.c b/drivers/staging/erofs/super.c
index 2bd433a..97da5c8 100644
--- a/drivers/staging/erofs/super.c
+++ b/drivers/staging/erofs/super.c
@@ -256,6 +256,63 @@ static int parse_options(struct super_block *sb, char *options)
 	return 0;
 }
 
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+
+static const struct address_space_operations managed_cache_aops;
+
+static int managed_cache_releasepage(struct page *page, gfp_t gfp_mask)
+{
+	int ret = 1;	/* 0 - busy */
+	struct address_space *const mapping = page->mapping;
+
+	BUG_ON(!PageLocked(page));
+	BUG_ON(mapping->a_ops != &managed_cache_aops);
+
+	if (PagePrivate(page))
+		ret = try_to_free_cached_page(mapping, page);
+
+	return ret;
+}
+
+static void managed_cache_invalidatepage(struct page *page,
+	unsigned int offset, unsigned int length)
+{
+	const unsigned int stop = length + offset;
+
+	BUG_ON(!PageLocked(page));
+
+	/* Check for overflow */
+	BUG_ON(stop > PAGE_SIZE || stop < length);
+
+	if (offset == 0 && stop == PAGE_SIZE)
+		while (!managed_cache_releasepage(page, GFP_NOFS))
+			cond_resched();
+}
+
+static const struct address_space_operations managed_cache_aops = {
+	.releasepage = managed_cache_releasepage,
+	.invalidatepage = managed_cache_invalidatepage,
+};
+
+static struct inode *erofs_init_managed_cache(struct super_block *sb)
+{
+	struct inode *inode = new_inode(sb);
+
+	if (unlikely(inode == NULL))
+		return ERR_PTR(-ENOMEM);
+
+	set_nlink(inode, 1);
+	inode->i_size = OFFSET_MAX;
+
+	inode->i_mapping->a_ops = &managed_cache_aops;
+	mapping_set_gfp_mask(inode->i_mapping,
+			     GFP_NOFS | __GFP_HIGHMEM |
+			     __GFP_MOVABLE |  __GFP_NOFAIL);
+	return inode;
+}
+
+#endif
+
 static int erofs_read_super(struct super_block *sb,
 	const char *dev_name, void *data, int silent)
 {
@@ -307,6 +364,14 @@ static int erofs_read_super(struct super_block *sb,
 	INIT_RADIX_TREE(&sbi->workstn_tree, GFP_ATOMIC);
 #endif
 
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+	sbi->managed_cache = erofs_init_managed_cache(sb);
+	if (IS_ERR(sbi->managed_cache)) {
+		err = PTR_ERR(sbi->managed_cache);
+		goto err_init_managed_cache;
+	}
+#endif
+
 	/* get the root inode */
 	inode = erofs_iget(sb, ROOT_NID(sbi), true);
 	if (IS_ERR(inode)) {
@@ -361,6 +426,10 @@ static int erofs_read_super(struct super_block *sb,
 	if (sb->s_root == NULL)
 		iput(inode);
 err_iget:
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+	iput(sbi->managed_cache);
+err_init_managed_cache:
+#endif
 err_parseopt:
 err_sbread:
 	sb->s_fs_info = NULL;
@@ -386,6 +455,10 @@ static void erofs_put_super(struct super_block *sb)
 	infoln("unmounted for %s", sbi->dev_name);
 	__putname(sbi->dev_name);
 
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+	iput(sbi->managed_cache);
+#endif
+
 	mutex_lock(&sbi->umount_mutex);
 
 #ifdef CONFIG_EROFS_FS_ZIP
diff --git a/drivers/staging/erofs/unzip_vle.c b/drivers/staging/erofs/unzip_vle.c
index f0ead60..7671fe8 100644
--- a/drivers/staging/erofs/unzip_vle.c
+++ b/drivers/staging/erofs/unzip_vle.c
@@ -95,6 +95,111 @@ struct z_erofs_vle_work_builder {
 #define VLE_WORK_BUILDER_INIT()	\
 	{ .work = NULL, .role = Z_EROFS_VLE_WORK_PRIMARY_FOLLOWED }
 
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+
+static bool grab_managed_cache_pages(struct address_space *mapping,
+				     erofs_blk_t start,
+				     struct page **compressed_pages,
+				     int clusterblks,
+				     bool reserve_allocation)
+{
+	bool noio = true;
+	unsigned int i;
+
+	/* TODO: optimize by introducing find_get_pages_range */
+	for (i = 0; i < clusterblks; ++i) {
+		struct page *page, *found;
+
+		if (READ_ONCE(compressed_pages[i]) != NULL)
+			continue;
+
+		page = found = find_get_page(mapping, start + i);
+		if (found == NULL) {
+			noio = false;
+			if (!reserve_allocation)
+				continue;
+			page = EROFS_UNALLOCATED_CACHED_PAGE;
+		}
+
+		if (NULL == cmpxchg(compressed_pages + i, NULL, page))
+			continue;
+
+		if (found != NULL)
+			put_page(found);
+	}
+	return noio;
+}
+
+/* called by erofs_shrinker to get rid of all compressed_pages */
+int try_to_free_all_cached_pages(struct erofs_sb_info *sbi,
+				 struct erofs_workgroup *egrp)
+{
+	struct z_erofs_vle_workgroup *const grp =
+		container_of(egrp, struct z_erofs_vle_workgroup, obj);
+	struct address_space *const mapping = sbi->managed_cache->i_mapping;
+	const int clusterpages = erofs_clusterpages(sbi);
+	int i;
+
+	/*
+	 * refcount of workgroup is now freezed as 1,
+	 * therefore no need to worry about available decompression users.
+	 */
+	for (i = 0; i < clusterpages; ++i) {
+		struct page *page = grp->compressed_pages[i];
+
+		if (page == NULL || page->mapping != mapping)
+			continue;
+
+		/* block other users from reclaiming or migrating the page */
+		if (!trylock_page(page))
+			return -EBUSY;
+
+		/* barrier is implied in the following 'unlock_page' */
+		WRITE_ONCE(grp->compressed_pages[i], NULL);
+
+		set_page_private(page, 0);
+		ClearPagePrivate(page);
+
+		unlock_page(page);
+		put_page(page);
+	}
+	return 0;
+}
+
+int try_to_free_cached_page(struct address_space *mapping, struct page *page)
+{
+	struct erofs_sb_info *const sbi = EROFS_SB(mapping->host->i_sb);
+	const unsigned int clusterpages = erofs_clusterpages(sbi);
+
+	struct z_erofs_vle_workgroup *grp;
+	int ret = 0;	/* 0 - busy */
+
+	/* prevent the workgroup from being freed */
+	rcu_read_lock();
+	grp = (void *)page_private(page);
+
+	if (erofs_workgroup_try_to_freeze(&grp->obj, 1)) {
+		unsigned int i;
+
+		for (i = 0; i < clusterpages; ++i) {
+			if (grp->compressed_pages[i] == page) {
+				WRITE_ONCE(grp->compressed_pages[i], NULL);
+				ret = 1;
+				break;
+			}
+		}
+		erofs_workgroup_unfreeze(&grp->obj, 1);
+	}
+	rcu_read_unlock();
+
+	if (ret) {
+		ClearPagePrivate(page);
+		put_page(page);
+	}
+	return ret;
+}
+#endif
+
 /* page_type must be Z_EROFS_PAGE_TYPE_EXCLUSIVE */
 static inline bool try_to_reuse_as_compressed_page(
 	struct z_erofs_vle_work_builder *b,
@@ -463,6 +568,9 @@ struct z_erofs_vle_frontend {
 	z_erofs_vle_owned_workgrp_t owned_head;
 
 	bool initial;
+#if (EROFS_FS_ZIP_CACHE_LVL >= 2)
+	erofs_off_t cachedzone_la;
+#endif
 };
 
 #define VLE_FRONTEND_INIT(__i) { \
@@ -489,6 +597,12 @@ static int z_erofs_do_read_page(struct z_erofs_vle_frontend *fe,
 	bool tight = builder_is_followed(builder);
 	struct z_erofs_vle_work *work = builder->work;
 
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+	struct address_space *const mngda = sbi->managed_cache->i_mapping;
+	struct z_erofs_vle_workgroup *grp;
+	bool noio_outoforder;
+#endif
+
 	enum z_erofs_page_type page_type;
 	unsigned cur, end, spiltted, index;
 	int err;
@@ -529,6 +643,21 @@ static int z_erofs_do_read_page(struct z_erofs_vle_frontend *fe,
 	if (unlikely(err))
 		goto err_out;
 
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+	grp = fe->builder.grp;
+
+	/* let's do out-of-order decompression for noio */
+	noio_outoforder = grab_managed_cache_pages(mngda,
+		erofs_blknr(map->m_pa),
+		grp->compressed_pages, erofs_blknr(map->m_plen),
+		/* compressed page caching selection strategy */
+		fe->initial | (EROFS_FS_ZIP_CACHE_LVL >= 2 ?
+			map->m_la < fe->cachedzone_la : 0));
+
+	if (noio_outoforder && builder_is_followed(builder))
+		builder->role = Z_EROFS_VLE_WORK_PRIMARY;
+#endif
+
 	tight &= builder_is_followed(builder);
 	work = builder->work;
 hitted:
@@ -607,15 +736,39 @@ static inline void z_erofs_vle_read_endio(struct bio *bio)
 	const blk_status_t err = bio->bi_status;
 	unsigned i;
 	struct bio_vec *bvec;
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+	struct address_space *mngda = NULL;
+#endif
 
 	bio_for_each_segment_all(bvec, bio, i) {
 		struct page *page = bvec->bv_page;
+		bool cachemngd = false;
 
 		DBG_BUGON(PageUptodate(page));
 		BUG_ON(page->mapping == NULL);
 
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+		if (unlikely(mngda == NULL && !z_erofs_is_stagingpage(page))) {
+			struct inode *const inode = page->mapping->host;
+			struct super_block *const sb = inode->i_sb;
+
+			mngda = EROFS_SB(sb)->managed_cache->i_mapping;
+		}
+
+		/*
+		 * If mngda has not gotten, it equals NULL,
+		 * however, page->mapping never be NULL if working properly.
+		 */
+		cachemngd = (page->mapping == mngda);
+#endif
+
 		if (unlikely(err))
 			SetPageError(page);
+		else if (cachemngd)
+			SetPageUptodate(page);
+
+		if (cachemngd)
+			unlock_page(page);
 	}
 
 	z_erofs_vle_unzip_kickoff(bio->bi_private, -1);
@@ -630,6 +783,9 @@ static int z_erofs_vle_unzip(struct super_block *sb,
 	struct list_head *page_pool)
 {
 	struct erofs_sb_info *const sbi = EROFS_SB(sb);
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+	struct address_space *const mngda = sbi->managed_cache->i_mapping;
+#endif
 	const unsigned clusterpages = erofs_clusterpages(sbi);
 
 	struct z_erofs_pagevec_ctor ctor;
@@ -727,6 +883,13 @@ static int z_erofs_vle_unzip(struct super_block *sb,
 
 		if (z_erofs_is_stagingpage(page))
 			continue;
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+		else if (page->mapping == mngda) {
+			BUG_ON(PageLocked(page));
+			BUG_ON(!PageUptodate(page));
+			continue;
+		}
+#endif
 
 		/* only non-head page could be reused as a compressed page */
 		pagenr = z_erofs_onlinepage_index(page);
@@ -804,6 +967,10 @@ static int z_erofs_vle_unzip(struct super_block *sb,
 	for (i = 0; i < clusterpages; ++i) {
 		page = compressed_pages[i];
 
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+		if (page->mapping == mngda)
+			continue;
+#endif
 		/* recycle all individual staging pages */
 		(void)z_erofs_gather_if_stagingpage(page_pool, page);
 
@@ -898,7 +1065,31 @@ static void z_erofs_vle_unzip_wq(struct work_struct *work)
 	return io;
 }
 
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+/* true - unlocked (noio), false - locked (need submit io) */
+static inline bool recover_managed_page(struct z_erofs_vle_workgroup *grp,
+					struct page *page)
+{
+	wait_on_page_locked(page);
+	if (PagePrivate(page) && PageUptodate(page))
+		return true;
+
+	lock_page(page);
+	if (unlikely(!PagePrivate(page))) {
+		set_page_private(page, (unsigned long)grp);
+		SetPagePrivate(page);
+	}
+	if (unlikely(PageUptodate(page))) {
+		unlock_page(page);
+		return true;
+	}
+	return false;
+}
+
+#define __FSIO_1 1
+#else
 #define __FSIO_1 0
+#endif
 
 static bool z_erofs_vle_submit_all(struct super_block *sb,
 				   z_erofs_vle_owned_workgrp_t owned_head,
@@ -909,6 +1100,10 @@ static bool z_erofs_vle_submit_all(struct super_block *sb,
 	struct erofs_sb_info *const sbi = EROFS_SB(sb);
 	const unsigned clusterpages = erofs_clusterpages(sbi);
 	const gfp_t gfp = GFP_NOFS;
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+	struct address_space *const mngda = sbi->managed_cache->i_mapping;
+	struct z_erofs_vle_workgroup *lstgrp_noio = NULL, *lstgrp_io = NULL;
+#endif
 	struct z_erofs_vle_unzip_io *ios[1 + __FSIO_1];
 	struct bio *bio;
 	tagptr1_t bi_private;
@@ -924,6 +1119,10 @@ static bool z_erofs_vle_submit_all(struct super_block *sb,
 	 * force_fg == 1, (io, fg_io[0]) no io, (io, fg_io[1]) need submit io
 	 * force_fg == 0, (io, fg_io[0]) no io; (io[1], bg_io) need submit io
 	 */
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+	ios[0] = prepare_io_handler(sb, fg_io + 0, false);
+#endif
+
 	if (force_fg) {
 		ios[__FSIO_1] = prepare_io_handler(sb, fg_io + __FSIO_1, false);
 		bi_private = tagptr_fold(tagptr1_t, ios[__FSIO_1], 0);
@@ -944,6 +1143,10 @@ static bool z_erofs_vle_submit_all(struct super_block *sb,
 		struct page **compressed_pages, *oldpage, *page;
 		pgoff_t first_index;
 		unsigned i = 0;
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+		unsigned int noio = 0;
+		bool cachemngd;
+#endif
 		int err;
 
 		/* no possible 'owned_head' equals the following */
@@ -964,15 +1167,40 @@ static bool z_erofs_vle_submit_all(struct super_block *sb,
 		/* fulfill all compressed pages */
 		oldpage = page = READ_ONCE(compressed_pages[i]);
 
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+		cachemngd = false;
+
+		if (page == EROFS_UNALLOCATED_CACHED_PAGE) {
+			cachemngd = true;
+			goto do_allocpage;
+		} else if (page != NULL) {
+			if (page->mapping != mngda)
+				BUG_ON(PageUptodate(page));
+			else if (recover_managed_page(grp, page)) {
+				/* page is uptodate, skip io submission */
+				force_submit = true;
+				++noio;
+				goto skippage;
+			}
+		} else {
+do_allocpage:
+#else
 		if (page != NULL)
 			BUG_ON(PageUptodate(page));
 		else {
+#endif
 			page = __stagingpage_alloc(pagepool, gfp);
 
 			if (oldpage != cmpxchg(compressed_pages + i,
 				oldpage, page)) {
 				list_add(&page->lru, pagepool);
 				goto repeat;
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+			} else if (cachemngd && !add_to_page_cache_lru(page,
+				mngda, first_index + i, gfp)) {
+				set_page_private(page, (unsigned long)grp);
+				SetPagePrivate(page);
+#endif
 			}
 		}
 
@@ -996,14 +1224,51 @@ static bool z_erofs_vle_submit_all(struct super_block *sb,
 
 		force_submit = false;
 		last_index = first_index + i;
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+skippage:
+#endif
 		if (++i < clusterpages)
 			goto repeat;
+
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+		if (noio < clusterpages) {
+			lstgrp_io = grp;
+		} else {
+			z_erofs_vle_owned_workgrp_t iogrp_next =
+				owned_head == Z_EROFS_VLE_WORKGRP_TAIL ?
+				Z_EROFS_VLE_WORKGRP_TAIL_CLOSED :
+				owned_head;
+
+			if (lstgrp_io == NULL)
+				ios[1]->head = iogrp_next;
+			else
+				WRITE_ONCE(lstgrp_io->next, iogrp_next);
+
+			if (lstgrp_noio == NULL)
+				ios[0]->head = grp;
+			else
+				WRITE_ONCE(lstgrp_noio->next, grp);
+
+			lstgrp_noio = grp;
+		}
+#endif
 	} while (owned_head != Z_EROFS_VLE_WORKGRP_TAIL);
 
 	if (bio != NULL)
 		__submit_bio(bio, REQ_OP_READ, 0);
 
+#ifndef EROFS_FS_HAS_MANAGED_CACHE
 	BUG_ON(!nr_bios);
+#else
+	if (lstgrp_noio != NULL)
+		WRITE_ONCE(lstgrp_noio->next, Z_EROFS_VLE_WORKGRP_TAIL_CLOSED);
+
+	if (!force_fg && !nr_bios) {
+		kvfree(container_of(ios[1],
+			struct z_erofs_vle_unzip_io_sb, io));
+		return true;
+	}
+#endif
 
 	z_erofs_vle_unzip_kickoff(tagptr_cast_ptr(bi_private), nr_bios);
 	return true;
@@ -1019,6 +1284,9 @@ static void z_erofs_submit_and_unzip(struct z_erofs_vle_frontend *f,
 	if (!z_erofs_vle_submit_all(sb, f->owned_head, pagepool, io, force_fg))
 		return;
 
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+	z_erofs_vle_unzip_all(sb, &io[0], pagepool);
+#endif
 	if (!force_fg)
 		return;
 
@@ -1038,6 +1306,9 @@ static int z_erofs_vle_normalaccess_readpage(struct file *file,
 	int err;
 	LIST_HEAD(pagepool);
 
+#if (EROFS_FS_ZIP_CACHE_LVL >= 2)
+	f.cachedzone_la = page->index << PAGE_SHIFT;
+#endif
 	err = z_erofs_do_read_page(&f, page, &pagepool);
 	(void)z_erofs_vle_work_iter_end(&f.builder);
 
@@ -1068,6 +1339,9 @@ static inline int __z_erofs_vle_normalaccess_readpages(
 	struct page *head = NULL;
 	LIST_HEAD(pagepool);
 
+#if (EROFS_FS_ZIP_CACHE_LVL >= 2)
+	f.cachedzone_la = lru_to_page(pages)->index << PAGE_SHIFT;
+#endif
 	for (; nr_pages; --nr_pages) {
 		struct page *page = lru_to_page(pages);
 
diff --git a/drivers/staging/erofs/utils.c b/drivers/staging/erofs/utils.c
index 6530035..ee70bb9 100644
--- a/drivers/staging/erofs/utils.c
+++ b/drivers/staging/erofs/utils.c
@@ -143,13 +143,28 @@ unsigned long erofs_shrink_workstation(struct erofs_sb_info *sbi,
 		if (cleanup)
 			BUG_ON(cnt != 1);
 
+#ifndef EROFS_FS_HAS_MANAGED_CACHE
 		else if (cnt > 1)
+#else
+		if (!erofs_workgroup_try_to_freeze(grp, 1))
+#endif
 			continue;
 
 		if (radix_tree_delete(&sbi->workstn_tree,
-			grp->index) != grp)
+			grp->index) != grp) {
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+skip:
+			erofs_workgroup_unfreeze(grp, 1);
+#endif
 			continue;
+		}
 
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+		if (try_to_free_all_cached_pages(sbi, grp))
+			goto skip;
+
+		erofs_workgroup_unfreeze(grp, 1);
+#endif
 		/* (rarely) grabbed again when freeing */
 		erofs_workgroup_put(grp);
 
-- 
1.9.1


WARNING: multiple messages have this Message-ID (diff)
From: gaoxiang25@huawei.com (Gao Xiang)
Subject: [PATCH 24/25] staging: erofs: introduce cached decompression
Date: Thu, 26 Jul 2018 20:22:07 +0800	[thread overview]
Message-ID: <1532607728-103372-25-git-send-email-gaoxiang25@huawei.com> (raw)
In-Reply-To: <1532607728-103372-1-git-send-email-gaoxiang25@huawei.com>

This patch adds an optional choice which can be
enabled by users in order to cache both incomplete
ends of compressed clusters as a complement to
the in-place decompression in order to boost random
read, but it costs more memory than the in-place
decompression only.

Signed-off-by: Gao Xiang <gaoxiang25 at huawei.com>
---
 drivers/staging/erofs/Kconfig     |  38 ++++++
 drivers/staging/erofs/internal.h  |  26 ++++
 drivers/staging/erofs/super.c     |  73 ++++++++++
 drivers/staging/erofs/unzip_vle.c | 274 ++++++++++++++++++++++++++++++++++++++
 drivers/staging/erofs/utils.c     |  17 ++-
 5 files changed, 427 insertions(+), 1 deletion(-)

diff --git a/drivers/staging/erofs/Kconfig b/drivers/staging/erofs/Kconfig
index b55ce1c..663b755 100644
--- a/drivers/staging/erofs/Kconfig
+++ b/drivers/staging/erofs/Kconfig
@@ -101,3 +101,41 @@ config EROFS_FS_CLUSTER_PAGE_LIMIT
 	  than 2. Otherwise, the image cannot be mounted
 	  correctly on this kernel.
 
+choice
+	prompt "EROFS VLE Data Decompression mode"
+	depends on EROFS_FS_ZIP
+	default EROFS_FS_ZIP_CACHE_BIPOLAR
+	help
+	  EROFS supports three options for VLE decompression.
+	  "In-place Decompression Only" consumes the minimum memory
+	  with lowest random read.
+
+	  "Bipolar Cached Decompression" consumes the maximum memory
+	  with highest random read.
+
+	  If unsure, select "Bipolar Cached Decompression"
+
+config EROFS_FS_ZIP_NO_CACHE
+	bool "In-place Decompression Only"
+	help
+	  Read compressed data into page cache and do in-place
+	  decompression directly.
+
+config EROFS_FS_ZIP_CACHE_UNIPOLAR
+	bool "Unipolar Cached Decompression"
+	help
+	  For each request, it caches the last compressed page
+	  for further reading.
+	  It still decompresses in place for the rest compressed pages.
+
+config EROFS_FS_ZIP_CACHE_BIPOLAR
+	bool "Bipolar Cached Decompression"
+	help
+	  For each request, it caches the both end compressed pages
+	  for further reading.
+	  It still decompresses in place for the rest compressed pages.
+
+	  Recommended for performance priority.
+
+endchoice
+
diff --git a/drivers/staging/erofs/internal.h b/drivers/staging/erofs/internal.h
index 3adec7d..669f93a 100644
--- a/drivers/staging/erofs/internal.h
+++ b/drivers/staging/erofs/internal.h
@@ -58,6 +58,18 @@ struct erofs_fault_info {
 };
 #endif
 
+#ifdef CONFIG_EROFS_FS_ZIP_CACHE_BIPOLAR
+#define EROFS_FS_ZIP_CACHE_LVL	(2)
+#elif defined(EROFS_FS_ZIP_CACHE_UNIPOLAR)
+#define EROFS_FS_ZIP_CACHE_LVL	(1)
+#else
+#define EROFS_FS_ZIP_CACHE_LVL	(0)
+#endif
+
+#if (!defined(EROFS_FS_HAS_MANAGED_CACHE) && (EROFS_FS_ZIP_CACHE_LVL > 0))
+#define EROFS_FS_HAS_MANAGED_CACHE
+#endif
+
 /* EROFS_SUPER_MAGIC_V1 to represent the whole file system */
 #define EROFS_SUPER_MAGIC   EROFS_SUPER_MAGIC_V1
 
@@ -82,6 +94,11 @@ struct erofs_sb_info {
 
 	/* the dedicated workstation for compression */
 	struct radix_tree_root workstn_tree;
+
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+	struct inode *managed_cache;
+#endif
+
 #endif
 
 	u32 build_time_nsec;
@@ -240,6 +257,15 @@ static inline void erofs_workstation_cleanup_all(struct super_block *sb)
 	erofs_shrink_workstation(EROFS_SB(sb), ~0UL, true);
 }
 
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+#define EROFS_UNALLOCATED_CACHED_PAGE	((void *)0x5F0EF00D)
+
+extern int try_to_free_all_cached_pages(struct erofs_sb_info *sbi,
+	struct erofs_workgroup *egrp);
+extern int try_to_free_cached_page(struct address_space *mapping,
+	struct page *page);
+#endif
+
 #endif
 
 /* we strictly follow PAGE_SIZE and no buffer head yet */
diff --git a/drivers/staging/erofs/super.c b/drivers/staging/erofs/super.c
index 2bd433a..97da5c8 100644
--- a/drivers/staging/erofs/super.c
+++ b/drivers/staging/erofs/super.c
@@ -256,6 +256,63 @@ static int parse_options(struct super_block *sb, char *options)
 	return 0;
 }
 
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+
+static const struct address_space_operations managed_cache_aops;
+
+static int managed_cache_releasepage(struct page *page, gfp_t gfp_mask)
+{
+	int ret = 1;	/* 0 - busy */
+	struct address_space *const mapping = page->mapping;
+
+	BUG_ON(!PageLocked(page));
+	BUG_ON(mapping->a_ops != &managed_cache_aops);
+
+	if (PagePrivate(page))
+		ret = try_to_free_cached_page(mapping, page);
+
+	return ret;
+}
+
+static void managed_cache_invalidatepage(struct page *page,
+	unsigned int offset, unsigned int length)
+{
+	const unsigned int stop = length + offset;
+
+	BUG_ON(!PageLocked(page));
+
+	/* Check for overflow */
+	BUG_ON(stop > PAGE_SIZE || stop < length);
+
+	if (offset == 0 && stop == PAGE_SIZE)
+		while (!managed_cache_releasepage(page, GFP_NOFS))
+			cond_resched();
+}
+
+static const struct address_space_operations managed_cache_aops = {
+	.releasepage = managed_cache_releasepage,
+	.invalidatepage = managed_cache_invalidatepage,
+};
+
+static struct inode *erofs_init_managed_cache(struct super_block *sb)
+{
+	struct inode *inode = new_inode(sb);
+
+	if (unlikely(inode == NULL))
+		return ERR_PTR(-ENOMEM);
+
+	set_nlink(inode, 1);
+	inode->i_size = OFFSET_MAX;
+
+	inode->i_mapping->a_ops = &managed_cache_aops;
+	mapping_set_gfp_mask(inode->i_mapping,
+			     GFP_NOFS | __GFP_HIGHMEM |
+			     __GFP_MOVABLE |  __GFP_NOFAIL);
+	return inode;
+}
+
+#endif
+
 static int erofs_read_super(struct super_block *sb,
 	const char *dev_name, void *data, int silent)
 {
@@ -307,6 +364,14 @@ static int erofs_read_super(struct super_block *sb,
 	INIT_RADIX_TREE(&sbi->workstn_tree, GFP_ATOMIC);
 #endif
 
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+	sbi->managed_cache = erofs_init_managed_cache(sb);
+	if (IS_ERR(sbi->managed_cache)) {
+		err = PTR_ERR(sbi->managed_cache);
+		goto err_init_managed_cache;
+	}
+#endif
+
 	/* get the root inode */
 	inode = erofs_iget(sb, ROOT_NID(sbi), true);
 	if (IS_ERR(inode)) {
@@ -361,6 +426,10 @@ static int erofs_read_super(struct super_block *sb,
 	if (sb->s_root == NULL)
 		iput(inode);
 err_iget:
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+	iput(sbi->managed_cache);
+err_init_managed_cache:
+#endif
 err_parseopt:
 err_sbread:
 	sb->s_fs_info = NULL;
@@ -386,6 +455,10 @@ static void erofs_put_super(struct super_block *sb)
 	infoln("unmounted for %s", sbi->dev_name);
 	__putname(sbi->dev_name);
 
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+	iput(sbi->managed_cache);
+#endif
+
 	mutex_lock(&sbi->umount_mutex);
 
 #ifdef CONFIG_EROFS_FS_ZIP
diff --git a/drivers/staging/erofs/unzip_vle.c b/drivers/staging/erofs/unzip_vle.c
index f0ead60..7671fe8 100644
--- a/drivers/staging/erofs/unzip_vle.c
+++ b/drivers/staging/erofs/unzip_vle.c
@@ -95,6 +95,111 @@ struct z_erofs_vle_work_builder {
 #define VLE_WORK_BUILDER_INIT()	\
 	{ .work = NULL, .role = Z_EROFS_VLE_WORK_PRIMARY_FOLLOWED }
 
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+
+static bool grab_managed_cache_pages(struct address_space *mapping,
+				     erofs_blk_t start,
+				     struct page **compressed_pages,
+				     int clusterblks,
+				     bool reserve_allocation)
+{
+	bool noio = true;
+	unsigned int i;
+
+	/* TODO: optimize by introducing find_get_pages_range */
+	for (i = 0; i < clusterblks; ++i) {
+		struct page *page, *found;
+
+		if (READ_ONCE(compressed_pages[i]) != NULL)
+			continue;
+
+		page = found = find_get_page(mapping, start + i);
+		if (found == NULL) {
+			noio = false;
+			if (!reserve_allocation)
+				continue;
+			page = EROFS_UNALLOCATED_CACHED_PAGE;
+		}
+
+		if (NULL == cmpxchg(compressed_pages + i, NULL, page))
+			continue;
+
+		if (found != NULL)
+			put_page(found);
+	}
+	return noio;
+}
+
+/* called by erofs_shrinker to get rid of all compressed_pages */
+int try_to_free_all_cached_pages(struct erofs_sb_info *sbi,
+				 struct erofs_workgroup *egrp)
+{
+	struct z_erofs_vle_workgroup *const grp =
+		container_of(egrp, struct z_erofs_vle_workgroup, obj);
+	struct address_space *const mapping = sbi->managed_cache->i_mapping;
+	const int clusterpages = erofs_clusterpages(sbi);
+	int i;
+
+	/*
+	 * refcount of workgroup is now freezed as 1,
+	 * therefore no need to worry about available decompression users.
+	 */
+	for (i = 0; i < clusterpages; ++i) {
+		struct page *page = grp->compressed_pages[i];
+
+		if (page == NULL || page->mapping != mapping)
+			continue;
+
+		/* block other users from reclaiming or migrating the page */
+		if (!trylock_page(page))
+			return -EBUSY;
+
+		/* barrier is implied in the following 'unlock_page' */
+		WRITE_ONCE(grp->compressed_pages[i], NULL);
+
+		set_page_private(page, 0);
+		ClearPagePrivate(page);
+
+		unlock_page(page);
+		put_page(page);
+	}
+	return 0;
+}
+
+int try_to_free_cached_page(struct address_space *mapping, struct page *page)
+{
+	struct erofs_sb_info *const sbi = EROFS_SB(mapping->host->i_sb);
+	const unsigned int clusterpages = erofs_clusterpages(sbi);
+
+	struct z_erofs_vle_workgroup *grp;
+	int ret = 0;	/* 0 - busy */
+
+	/* prevent the workgroup from being freed */
+	rcu_read_lock();
+	grp = (void *)page_private(page);
+
+	if (erofs_workgroup_try_to_freeze(&grp->obj, 1)) {
+		unsigned int i;
+
+		for (i = 0; i < clusterpages; ++i) {
+			if (grp->compressed_pages[i] == page) {
+				WRITE_ONCE(grp->compressed_pages[i], NULL);
+				ret = 1;
+				break;
+			}
+		}
+		erofs_workgroup_unfreeze(&grp->obj, 1);
+	}
+	rcu_read_unlock();
+
+	if (ret) {
+		ClearPagePrivate(page);
+		put_page(page);
+	}
+	return ret;
+}
+#endif
+
 /* page_type must be Z_EROFS_PAGE_TYPE_EXCLUSIVE */
 static inline bool try_to_reuse_as_compressed_page(
 	struct z_erofs_vle_work_builder *b,
@@ -463,6 +568,9 @@ struct z_erofs_vle_frontend {
 	z_erofs_vle_owned_workgrp_t owned_head;
 
 	bool initial;
+#if (EROFS_FS_ZIP_CACHE_LVL >= 2)
+	erofs_off_t cachedzone_la;
+#endif
 };
 
 #define VLE_FRONTEND_INIT(__i) { \
@@ -489,6 +597,12 @@ static int z_erofs_do_read_page(struct z_erofs_vle_frontend *fe,
 	bool tight = builder_is_followed(builder);
 	struct z_erofs_vle_work *work = builder->work;
 
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+	struct address_space *const mngda = sbi->managed_cache->i_mapping;
+	struct z_erofs_vle_workgroup *grp;
+	bool noio_outoforder;
+#endif
+
 	enum z_erofs_page_type page_type;
 	unsigned cur, end, spiltted, index;
 	int err;
@@ -529,6 +643,21 @@ static int z_erofs_do_read_page(struct z_erofs_vle_frontend *fe,
 	if (unlikely(err))
 		goto err_out;
 
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+	grp = fe->builder.grp;
+
+	/* let's do out-of-order decompression for noio */
+	noio_outoforder = grab_managed_cache_pages(mngda,
+		erofs_blknr(map->m_pa),
+		grp->compressed_pages, erofs_blknr(map->m_plen),
+		/* compressed page caching selection strategy */
+		fe->initial | (EROFS_FS_ZIP_CACHE_LVL >= 2 ?
+			map->m_la < fe->cachedzone_la : 0));
+
+	if (noio_outoforder && builder_is_followed(builder))
+		builder->role = Z_EROFS_VLE_WORK_PRIMARY;
+#endif
+
 	tight &= builder_is_followed(builder);
 	work = builder->work;
 hitted:
@@ -607,15 +736,39 @@ static inline void z_erofs_vle_read_endio(struct bio *bio)
 	const blk_status_t err = bio->bi_status;
 	unsigned i;
 	struct bio_vec *bvec;
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+	struct address_space *mngda = NULL;
+#endif
 
 	bio_for_each_segment_all(bvec, bio, i) {
 		struct page *page = bvec->bv_page;
+		bool cachemngd = false;
 
 		DBG_BUGON(PageUptodate(page));
 		BUG_ON(page->mapping == NULL);
 
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+		if (unlikely(mngda == NULL && !z_erofs_is_stagingpage(page))) {
+			struct inode *const inode = page->mapping->host;
+			struct super_block *const sb = inode->i_sb;
+
+			mngda = EROFS_SB(sb)->managed_cache->i_mapping;
+		}
+
+		/*
+		 * If mngda has not gotten, it equals NULL,
+		 * however, page->mapping never be NULL if working properly.
+		 */
+		cachemngd = (page->mapping == mngda);
+#endif
+
 		if (unlikely(err))
 			SetPageError(page);
+		else if (cachemngd)
+			SetPageUptodate(page);
+
+		if (cachemngd)
+			unlock_page(page);
 	}
 
 	z_erofs_vle_unzip_kickoff(bio->bi_private, -1);
@@ -630,6 +783,9 @@ static int z_erofs_vle_unzip(struct super_block *sb,
 	struct list_head *page_pool)
 {
 	struct erofs_sb_info *const sbi = EROFS_SB(sb);
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+	struct address_space *const mngda = sbi->managed_cache->i_mapping;
+#endif
 	const unsigned clusterpages = erofs_clusterpages(sbi);
 
 	struct z_erofs_pagevec_ctor ctor;
@@ -727,6 +883,13 @@ static int z_erofs_vle_unzip(struct super_block *sb,
 
 		if (z_erofs_is_stagingpage(page))
 			continue;
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+		else if (page->mapping == mngda) {
+			BUG_ON(PageLocked(page));
+			BUG_ON(!PageUptodate(page));
+			continue;
+		}
+#endif
 
 		/* only non-head page could be reused as a compressed page */
 		pagenr = z_erofs_onlinepage_index(page);
@@ -804,6 +967,10 @@ static int z_erofs_vle_unzip(struct super_block *sb,
 	for (i = 0; i < clusterpages; ++i) {
 		page = compressed_pages[i];
 
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+		if (page->mapping == mngda)
+			continue;
+#endif
 		/* recycle all individual staging pages */
 		(void)z_erofs_gather_if_stagingpage(page_pool, page);
 
@@ -898,7 +1065,31 @@ static void z_erofs_vle_unzip_wq(struct work_struct *work)
 	return io;
 }
 
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+/* true - unlocked (noio), false - locked (need submit io) */
+static inline bool recover_managed_page(struct z_erofs_vle_workgroup *grp,
+					struct page *page)
+{
+	wait_on_page_locked(page);
+	if (PagePrivate(page) && PageUptodate(page))
+		return true;
+
+	lock_page(page);
+	if (unlikely(!PagePrivate(page))) {
+		set_page_private(page, (unsigned long)grp);
+		SetPagePrivate(page);
+	}
+	if (unlikely(PageUptodate(page))) {
+		unlock_page(page);
+		return true;
+	}
+	return false;
+}
+
+#define __FSIO_1 1
+#else
 #define __FSIO_1 0
+#endif
 
 static bool z_erofs_vle_submit_all(struct super_block *sb,
 				   z_erofs_vle_owned_workgrp_t owned_head,
@@ -909,6 +1100,10 @@ static bool z_erofs_vle_submit_all(struct super_block *sb,
 	struct erofs_sb_info *const sbi = EROFS_SB(sb);
 	const unsigned clusterpages = erofs_clusterpages(sbi);
 	const gfp_t gfp = GFP_NOFS;
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+	struct address_space *const mngda = sbi->managed_cache->i_mapping;
+	struct z_erofs_vle_workgroup *lstgrp_noio = NULL, *lstgrp_io = NULL;
+#endif
 	struct z_erofs_vle_unzip_io *ios[1 + __FSIO_1];
 	struct bio *bio;
 	tagptr1_t bi_private;
@@ -924,6 +1119,10 @@ static bool z_erofs_vle_submit_all(struct super_block *sb,
 	 * force_fg == 1, (io, fg_io[0]) no io, (io, fg_io[1]) need submit io
 	 * force_fg == 0, (io, fg_io[0]) no io; (io[1], bg_io) need submit io
 	 */
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+	ios[0] = prepare_io_handler(sb, fg_io + 0, false);
+#endif
+
 	if (force_fg) {
 		ios[__FSIO_1] = prepare_io_handler(sb, fg_io + __FSIO_1, false);
 		bi_private = tagptr_fold(tagptr1_t, ios[__FSIO_1], 0);
@@ -944,6 +1143,10 @@ static bool z_erofs_vle_submit_all(struct super_block *sb,
 		struct page **compressed_pages, *oldpage, *page;
 		pgoff_t first_index;
 		unsigned i = 0;
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+		unsigned int noio = 0;
+		bool cachemngd;
+#endif
 		int err;
 
 		/* no possible 'owned_head' equals the following */
@@ -964,15 +1167,40 @@ static bool z_erofs_vle_submit_all(struct super_block *sb,
 		/* fulfill all compressed pages */
 		oldpage = page = READ_ONCE(compressed_pages[i]);
 
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+		cachemngd = false;
+
+		if (page == EROFS_UNALLOCATED_CACHED_PAGE) {
+			cachemngd = true;
+			goto do_allocpage;
+		} else if (page != NULL) {
+			if (page->mapping != mngda)
+				BUG_ON(PageUptodate(page));
+			else if (recover_managed_page(grp, page)) {
+				/* page is uptodate, skip io submission */
+				force_submit = true;
+				++noio;
+				goto skippage;
+			}
+		} else {
+do_allocpage:
+#else
 		if (page != NULL)
 			BUG_ON(PageUptodate(page));
 		else {
+#endif
 			page = __stagingpage_alloc(pagepool, gfp);
 
 			if (oldpage != cmpxchg(compressed_pages + i,
 				oldpage, page)) {
 				list_add(&page->lru, pagepool);
 				goto repeat;
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+			} else if (cachemngd && !add_to_page_cache_lru(page,
+				mngda, first_index + i, gfp)) {
+				set_page_private(page, (unsigned long)grp);
+				SetPagePrivate(page);
+#endif
 			}
 		}
 
@@ -996,14 +1224,51 @@ static bool z_erofs_vle_submit_all(struct super_block *sb,
 
 		force_submit = false;
 		last_index = first_index + i;
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+skippage:
+#endif
 		if (++i < clusterpages)
 			goto repeat;
+
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+		if (noio < clusterpages) {
+			lstgrp_io = grp;
+		} else {
+			z_erofs_vle_owned_workgrp_t iogrp_next =
+				owned_head == Z_EROFS_VLE_WORKGRP_TAIL ?
+				Z_EROFS_VLE_WORKGRP_TAIL_CLOSED :
+				owned_head;
+
+			if (lstgrp_io == NULL)
+				ios[1]->head = iogrp_next;
+			else
+				WRITE_ONCE(lstgrp_io->next, iogrp_next);
+
+			if (lstgrp_noio == NULL)
+				ios[0]->head = grp;
+			else
+				WRITE_ONCE(lstgrp_noio->next, grp);
+
+			lstgrp_noio = grp;
+		}
+#endif
 	} while (owned_head != Z_EROFS_VLE_WORKGRP_TAIL);
 
 	if (bio != NULL)
 		__submit_bio(bio, REQ_OP_READ, 0);
 
+#ifndef EROFS_FS_HAS_MANAGED_CACHE
 	BUG_ON(!nr_bios);
+#else
+	if (lstgrp_noio != NULL)
+		WRITE_ONCE(lstgrp_noio->next, Z_EROFS_VLE_WORKGRP_TAIL_CLOSED);
+
+	if (!force_fg && !nr_bios) {
+		kvfree(container_of(ios[1],
+			struct z_erofs_vle_unzip_io_sb, io));
+		return true;
+	}
+#endif
 
 	z_erofs_vle_unzip_kickoff(tagptr_cast_ptr(bi_private), nr_bios);
 	return true;
@@ -1019,6 +1284,9 @@ static void z_erofs_submit_and_unzip(struct z_erofs_vle_frontend *f,
 	if (!z_erofs_vle_submit_all(sb, f->owned_head, pagepool, io, force_fg))
 		return;
 
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+	z_erofs_vle_unzip_all(sb, &io[0], pagepool);
+#endif
 	if (!force_fg)
 		return;
 
@@ -1038,6 +1306,9 @@ static int z_erofs_vle_normalaccess_readpage(struct file *file,
 	int err;
 	LIST_HEAD(pagepool);
 
+#if (EROFS_FS_ZIP_CACHE_LVL >= 2)
+	f.cachedzone_la = page->index << PAGE_SHIFT;
+#endif
 	err = z_erofs_do_read_page(&f, page, &pagepool);
 	(void)z_erofs_vle_work_iter_end(&f.builder);
 
@@ -1068,6 +1339,9 @@ static inline int __z_erofs_vle_normalaccess_readpages(
 	struct page *head = NULL;
 	LIST_HEAD(pagepool);
 
+#if (EROFS_FS_ZIP_CACHE_LVL >= 2)
+	f.cachedzone_la = lru_to_page(pages)->index << PAGE_SHIFT;
+#endif
 	for (; nr_pages; --nr_pages) {
 		struct page *page = lru_to_page(pages);
 
diff --git a/drivers/staging/erofs/utils.c b/drivers/staging/erofs/utils.c
index 6530035..ee70bb9 100644
--- a/drivers/staging/erofs/utils.c
+++ b/drivers/staging/erofs/utils.c
@@ -143,13 +143,28 @@ unsigned long erofs_shrink_workstation(struct erofs_sb_info *sbi,
 		if (cleanup)
 			BUG_ON(cnt != 1);
 
+#ifndef EROFS_FS_HAS_MANAGED_CACHE
 		else if (cnt > 1)
+#else
+		if (!erofs_workgroup_try_to_freeze(grp, 1))
+#endif
 			continue;
 
 		if (radix_tree_delete(&sbi->workstn_tree,
-			grp->index) != grp)
+			grp->index) != grp) {
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+skip:
+			erofs_workgroup_unfreeze(grp, 1);
+#endif
 			continue;
+		}
 
+#ifdef EROFS_FS_HAS_MANAGED_CACHE
+		if (try_to_free_all_cached_pages(sbi, grp))
+			goto skip;
+
+		erofs_workgroup_unfreeze(grp, 1);
+#endif
 		/* (rarely) grabbed again when freeing */
 		erofs_workgroup_put(grp);
 
-- 
1.9.1

  parent reply	other threads:[~2018-07-26 12:29 UTC|newest]

Thread overview: 112+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-05-31 11:06 [NOMERGE] [RFC PATCH 00/12] erofs: introduce erofs file system Gao Xiang
2018-06-01  7:48 ` Richard Weinberger
2018-06-01  9:11   ` Gao Xiang
2018-06-01  9:28     ` Richard Weinberger
2018-06-01 11:16       ` Gao Xiang
2018-06-07 10:26         ` Pavel Machek
2018-07-27  0:55       ` Joey Pabalinas
2018-07-27  0:57         ` Joey Pabalinas
2018-07-26 12:21 ` [PATCH 00/25] staging: " Gao Xiang
2018-07-26 12:21   ` Gao Xiang
2018-07-26 12:21   ` [PATCH 01/25] staging: erofs: add on-disk layout Gao Xiang
2018-07-26 12:21     ` Gao Xiang
2018-07-26 12:21   ` [PATCH 02/25] staging: erofs: add erofs in-memory stuffs Gao Xiang
2018-07-26 12:21     ` Gao Xiang
2018-07-26 12:21   ` [PATCH 03/25] staging: erofs: add super block operations Gao Xiang
2018-07-26 12:21     ` Gao Xiang
2018-07-26 12:21   ` [PATCH 04/25] staging: erofs: add raw address_space operations Gao Xiang
2018-07-26 12:21     ` Gao Xiang
2018-07-26 12:21   ` [PATCH 05/25] staging: erofs: add inode operations Gao Xiang
2018-07-26 12:21     ` Gao Xiang
2018-07-26 12:21   ` [PATCH 06/25] staging: erofs: add directory operations Gao Xiang
2018-07-26 12:21     ` Gao Xiang
2018-07-26 12:21   ` [PATCH 07/25] staging: erofs: add namei functions Gao Xiang
2018-07-26 12:21     ` Gao Xiang
2018-07-26 12:21   ` [PATCH 08/25] staging: erofs: update Kconfig and Makefile Gao Xiang
2018-07-26 12:21     ` Gao Xiang
2018-07-26 12:21   ` [PATCH 09/25] staging: erofs: introduce xattr & acl support Gao Xiang
2018-07-26 12:21     ` Gao Xiang
2018-07-26 12:21   ` [PATCH 10/25] staging: erofs: support special inode Gao Xiang
2018-07-26 12:21     ` Gao Xiang
2018-07-26 12:21   ` [PATCH 11/25] staging: erofs: introduce error injection infrastructure Gao Xiang
2018-07-26 12:21     ` Gao Xiang
2018-07-26 12:21   ` [PATCH 12/25] staging: erofs: support tracepoint Gao Xiang
2018-07-26 12:21     ` Gao Xiang
2018-07-26 12:21   ` [PATCH 13/25] staging: erofs: <linux/tagptr.h>: introduce tagged pointer Gao Xiang
2018-07-26 12:21     ` Gao Xiang
2018-07-26 12:21   ` [PATCH 14/25] staging: erofs: introduce pagevec for unzip subsystem Gao Xiang
2018-07-26 12:21     ` Gao Xiang
2018-07-26 12:21   ` [PATCH 15/25] staging: erofs: add erofs_map_blocks_iter Gao Xiang
2018-07-26 12:21     ` Gao Xiang
2018-07-26 12:21   ` [PATCH 16/25] staging: erofs: add erofs_allocpage Gao Xiang
2018-07-26 12:21     ` Gao Xiang
2018-07-26 12:22   ` [PATCH 17/25] staging: erofs: globalize prepare_bio and __submit_bio Gao Xiang
2018-07-26 12:22     ` Gao Xiang
2018-07-26 12:22   ` [PATCH 18/25] staging: erofs: introduce a customized LZ4 decompression Gao Xiang
2018-07-26 12:22     ` Gao Xiang
2018-07-26 12:22   ` [PATCH 19/25] staging: erofs: add a generic z_erofs VLE decompressor Gao Xiang
2018-07-26 12:22     ` Gao Xiang
2018-07-26 12:22   ` [PATCH 20/25] staging: erofs: introduce superblock registration Gao Xiang
2018-07-26 12:22     ` Gao Xiang
2018-07-26 12:22   ` [PATCH 21/25] staging: erofs: introduce erofs shrinker Gao Xiang
2018-07-26 12:22     ` Gao Xiang
2018-07-26 12:22   ` [PATCH 22/25] staging: erofs: introduce workstation for decompression Gao Xiang
2018-07-26 12:22     ` Gao Xiang
2018-07-26 12:22   ` [PATCH 23/25] staging: erofs: introduce VLE decompression support Gao Xiang
2018-07-26 12:22     ` Gao Xiang
2018-07-26 12:22   ` Gao Xiang [this message]
2018-07-26 12:22     ` [PATCH 24/25] staging: erofs: introduce cached decompression Gao Xiang
2018-07-26 12:22   ` [PATCH 25/25] staging: erofs: add a TODO and update MAINTAINERS for staging Gao Xiang
2018-07-26 12:22     ` Gao Xiang
2018-07-28  7:10     ` [PATCH] staging: erofs: fix a compile warning of Z_EROFS_VLE_VMAP_ONSTACK_PAGES Gao Xiang
2018-07-28  7:10       ` Gao Xiang
2018-07-28 10:43       ` Chao Yu
2018-07-28 10:43         ` Chao Yu
2018-07-29  5:34       ` [PATCH 1/2] staging: erofs: fix compile error without built-in decompression support Gao Xiang
2018-07-29  5:34         ` Gao Xiang
2018-07-29  5:37         ` [PATCH 2/2] staging: erofs: fix conditional uninitialized `pcn' in z_erofs_map_blocks_iter Gao Xiang
2018-07-29  5:37           ` Gao Xiang
2018-07-30  1:51           ` [PATCH] staging: erofs: use the wrapped PTR_ERR_OR_ZERO instead of open code Gao Xiang
2018-07-30  1:51             ` Gao Xiang
2018-07-30  6:58             ` Chao Yu
2018-07-30  6:58               ` Chao Yu
2018-08-01  6:38             ` [PATCH 1/2] staging: erofs: add the missing break in z_erofs_map_blocks_iter Gao Xiang
2018-08-01  6:38               ` Gao Xiang
2018-08-01  6:38               ` [PATCH 2/2] staging: erofs: remove a redundant marco in xattr Gao Xiang
2018-08-01  6:38                 ` Gao Xiang
2018-08-01  9:02               ` [PATCH 1/2] staging: erofs: add the missing break in z_erofs_map_blocks_iter Dan Carpenter
2018-08-01  9:02                 ` Dan Carpenter
2018-08-01  9:19                 ` Gao Xiang
2018-08-01  9:19                   ` Gao Xiang
2018-08-01  9:36                   ` [PATCH RESEND " Gao Xiang
2018-08-01  9:36                     ` Gao Xiang
2018-08-01 11:36                     ` Dan Carpenter
2018-08-01 11:36                       ` Dan Carpenter
2018-08-01 12:08                       ` Gao Xiang
2018-08-01 12:08                         ` Gao Xiang
2018-07-30  2:07           ` [PATCH 2/2] staging: erofs: fix conditional uninitialized `pcn' " Chao Yu
2018-07-30  2:07             ` Chao Yu
2018-07-30  2:07         ` [PATCH 1/2] staging: erofs: fix compile error without built-in decompression support Chao Yu
2018-07-30  2:07           ` Chao Yu
2018-07-30  2:32           ` Gao Xiang
2018-07-30  2:32             ` Gao Xiang
2018-07-30  3:07             ` Chao Yu
2018-07-30  3:07               ` Chao Yu
2018-07-30  3:55               ` Gao Xiang
2018-07-30  3:55                 ` Gao Xiang
2018-07-30  3:34           ` [FOR INTERNAL REVIEW] [PATCH RESEND 1/3] staging: erofs: fix incorrect code in erofs_shrink_scan Gao Xiang
2018-07-30  3:34             ` [FOR INTERNAL REVIEW] [PATCH RESEND 2/3] staging: erofs: add 'erofs_' prefixes for try_to_free_(all_)cached_page(s) Gao Xiang
2018-07-30  6:57               ` Chao Yu
2018-07-30  3:34             ` [FOR INTERNAL REVIEW] [PATCH RESEND 3/3] staging: erofs: fix conditional uninitialized `pcn' in z_erofs_map_blocks_iter Gao Xiang
2018-07-30  6:56             ` [FOR INTERNAL REVIEW] [PATCH RESEND 1/3] staging: erofs: fix incorrect code in erofs_shrink_scan Chao Yu
2018-07-27  0:25   ` [PATCH 00/25] staging: erofs: introduce erofs file system Christian Kujau
2018-07-27  1:39     ` Gao Xiang
2018-07-27  1:39       ` Gao Xiang
2018-07-27  1:56       ` Gao Xiang
2018-07-27  1:56         ` Gao Xiang
2018-07-28  7:25   ` Greg Kroah-Hartman
2018-07-28  7:25     ` Greg Kroah-Hartman
2018-07-28  9:33     ` Gao Xiang
2018-07-28  9:33       ` Gao Xiang
2018-07-28 10:34     ` Chao Yu
2018-07-28 10:34       ` Chao Yu

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1532607728-103372-25-git-send-email-gaoxiang25@huawei.com \
    --to=gaoxiang25@huawei.com \
    --cc=chao@kernel.org \
    --cc=devel@driverdev.osuosl.org \
    --cc=gregkh@linuxfoundation.org \
    --cc=hsiangkao@aol.com \
    --cc=linux-erofs@lists.ozlabs.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=miaoxie@huawei.com \
    --cc=weidu.du@huawei.com \
    --cc=yuchao0@huawei.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.