5 files changed, 76 insertions, 79 deletions

diff --git a/fs/exofs/inode.c b/fs/exofs/inode.c
index 5badb0c039d..1562c27a2fa 100644
--- a/fs/exofs/inode.c
+++ b/fs/exofs/inode.c
@@ -37,15 +37,12 @@
 
 #define EXOFS_DBGMSG2(M...) do {} while (0)
 
-enum {MAX_PAGES_KMALLOC = PAGE_SIZE / sizeof(struct page *), };
-
 unsigned exofs_max_io_pages(struct ore_layout *layout,
 			    unsigned expected_pages)
 {
-	unsigned pages = min_t(unsigned, expected_pages, MAX_PAGES_KMALLOC);
+	unsigned pages = min_t(unsigned, expected_pages,
+			       layout->max_io_length / PAGE_SIZE);
 
-	/* TODO: easily support bio chaining */
-	pages =  min_t(unsigned, pages, layout->max_io_length / PAGE_SIZE);
 	return pages;
 }
 
@@ -101,7 +98,8 @@ static void _pcol_reset(struct page_collect *pcol)
 	 * it might not end here. don't be left with nothing
 	 */
 	if (!pcol->expected_pages)
-		pcol->expected_pages = MAX_PAGES_KMALLOC;
+		pcol->expected_pages =
+				exofs_max_io_pages(&pcol->sbi->layout, ~0);
 }
 
 static int pcol_try_alloc(struct page_collect *pcol)
@@ -389,6 +387,8 @@ static int readpage_strip(void *data, struct page *page)
 	size_t len;
 	int ret;
 
+	BUG_ON(!PageLocked(page));
+
 	/* FIXME: Just for debugging, will be removed */
 	if (PageUptodate(page))
 		EXOFS_ERR("PageUptodate(0x%lx, 0x%lx)\n", pcol->inode->i_ino,
@@ -572,8 +572,16 @@ static struct page *__r4w_get_page(void *priv, u64 offset, bool *uptodate)
 
 	if (!pcol->that_locked_page ||
 	    (pcol->that_locked_page->index != index)) {
-		struct page *page = find_get_page(pcol->inode->i_mapping, index);
+		struct page *page;
+		loff_t i_size = i_size_read(pcol->inode);
+
+		if (offset >= i_size) {
+			*uptodate = true;
+			EXOFS_DBGMSG("offset >= i_size index=0x%lx\n", index);
+			return ZERO_PAGE(0);
+		}
 
+		page =  find_get_page(pcol->inode->i_mapping, index);
 		if (!page) {
 			page = find_or_create_page(pcol->inode->i_mapping,
 						   index, GFP_NOFS);
@@ -602,12 +610,13 @@ static void __r4w_put_page(void *priv, struct page *page)
 {
 	struct page_collect *pcol = priv;
 
-	if (pcol->that_locked_page != page) {
+	if ((pcol->that_locked_page != page) && (ZERO_PAGE(0) != page)) {
 		EXOFS_DBGMSG("index=0x%lx\n", page->index);
 		page_cache_release(page);
 		return;
 	}
-	EXOFS_DBGMSG("that_locked_page index=0x%lx\n", page->index);
+	EXOFS_DBGMSG("that_locked_page index=0x%lx\n",
+		     ZERO_PAGE(0) == page ? -1 : page->index);
 }
 
 static const struct _ore_r4w_op _r4w_op = {
diff --git a/fs/exofs/namei.c b/fs/exofs/namei.c
index fc7161d6bf6..4731fd991ef 100644
--- a/fs/exofs/namei.c
+++ b/fs/exofs/namei.c
@@ -46,7 +46,7 @@ static inline int exofs_add_nondir(struct dentry *dentry, struct inode *inode)
 }
 
 static struct dentry *exofs_lookup(struct inode *dir, struct dentry *dentry,
-				   struct nameidata *nd)
+				   unsigned int flags)
 {
 	struct inode *inode;
 	ino_t ino;
@@ -60,7 +60,7 @@ static struct dentry *exofs_lookup(struct inode *dir, struct dentry *dentry,
 }
 
 static int exofs_create(struct inode *dir, struct dentry *dentry, umode_t mode,
-			 struct nameidata *nd)
+			 bool excl)
 {
 	struct inode *inode = exofs_new_inode(dir, mode);
 	int err = PTR_ERR(inode);
diff --git a/fs/exofs/ore.c b/fs/exofs/ore.c
index 49cf230554a..1585db1aa36 100644
--- a/fs/exofs/ore.c
+++ b/fs/exofs/ore.c
@@ -735,13 +735,7 @@ static int _prepare_for_striping(struct ore_io_state *ios)
 out:
 	ios->numdevs = devs_in_group;
 	ios->pages_consumed = cur_pg;
-	if (unlikely(ret)) {
-		if (length == ios->length)
-			return ret;
-		else
-			ios->length -= length;
-	}
-	return 0;
+	return ret;
 }
 
 int ore_create(struct ore_io_state *ios)
@@ -843,11 +837,11 @@ static int _write_mirror(struct ore_io_state *ios, int cur_comp)
 				bio->bi_rw |= REQ_WRITE;
 			}
 
-			osd_req_write(or, _ios_obj(ios, dev), per_dev->offset,
-				      bio, per_dev->length);
+			osd_req_write(or, _ios_obj(ios, cur_comp),
+				      per_dev->offset, bio, per_dev->length);
 			ORE_DBGMSG("write(0x%llx) offset=0x%llx "
 				      "length=0x%llx dev=%d\n",
-				     _LLU(_ios_obj(ios, dev)->id),
+				     _LLU(_ios_obj(ios, cur_comp)->id),
 				     _LLU(per_dev->offset),
 				     _LLU(per_dev->length), dev);
 		} else if (ios->kern_buff) {
@@ -859,20 +853,20 @@ static int _write_mirror(struct ore_io_state *ios, int cur_comp)
 			       (ios->si.unit_off + ios->length >
 				ios->layout->stripe_unit));
 
-			ret = osd_req_write_kern(or, _ios_obj(ios, per_dev->dev),
+			ret = osd_req_write_kern(or, _ios_obj(ios, cur_comp),
 						 per_dev->offset,
 						 ios->kern_buff, ios->length);
 			if (unlikely(ret))
 				goto out;
 			ORE_DBGMSG2("write_kern(0x%llx) offset=0x%llx "
 				      "length=0x%llx dev=%d\n",
-				     _LLU(_ios_obj(ios, dev)->id),
+				     _LLU(_ios_obj(ios, cur_comp)->id),
 				     _LLU(per_dev->offset),
 				     _LLU(ios->length), per_dev->dev);
 		} else {
-			osd_req_set_attributes(or, _ios_obj(ios, dev));
+			osd_req_set_attributes(or, _ios_obj(ios, cur_comp));
 			ORE_DBGMSG2("obj(0x%llx) set_attributes=%d dev=%d\n",
-				     _LLU(_ios_obj(ios, dev)->id),
+				     _LLU(_ios_obj(ios, cur_comp)->id),
 				     ios->out_attr_len, dev);
 		}
 
diff --git a/fs/exofs/ore_raid.c b/fs/exofs/ore_raid.c
index d222c77cfa1..5f376d14fdc 100644
--- a/fs/exofs/ore_raid.c
+++ b/fs/exofs/ore_raid.c
@@ -144,26 +144,26 @@ static void _sp2d_reset(struct __stripe_pages_2d *sp2d,
 {
 	unsigned data_devs = sp2d->data_devs;
 	unsigned group_width = data_devs + sp2d->parity;
-	unsigned p;
+	int p, c;
 
 	if (!sp2d->needed)
 		return;
 
-	for (p = 0; p < sp2d->pages_in_unit; p++) {
-		struct __1_page_stripe *_1ps = &sp2d->_1p_stripes[p];
-
-		if (_1ps->write_count < group_width) {
-			unsigned c;
+	for (c = data_devs - 1; c >= 0; --c)
+		for (p = sp2d->pages_in_unit - 1; p >= 0; --p) {
+			struct __1_page_stripe *_1ps = &sp2d->_1p_stripes[p];
 
-			for (c = 0; c < data_devs; c++)
-				if (_1ps->page_is_read[c]) {
-					struct page *page = _1ps->pages[c];
+			if (_1ps->page_is_read[c]) {
+				struct page *page = _1ps->pages[c];
 
-					r4w->put_page(priv, page);
-					_1ps->page_is_read[c] = false;
-				}
+				r4w->put_page(priv, page);
+				_1ps->page_is_read[c] = false;
+			}
 		}
 
+	for (p = 0; p < sp2d->pages_in_unit; p++) {
+		struct __1_page_stripe *_1ps = &sp2d->_1p_stripes[p];
+
 		memset(_1ps->pages, 0, group_width * sizeof(*_1ps->pages));
 		_1ps->write_count = 0;
 		_1ps->tx = NULL;
@@ -461,16 +461,12 @@ static void _mark_read4write_pages_uptodate(struct ore_io_state *ios, int ret)
  * ios->sp2d[p][*], xor is calculated the same way. These pages are
  * allocated/freed and don't go through cache
  */
-static int _read_4_write(struct ore_io_state *ios)
+static int _read_4_write_first_stripe(struct ore_io_state *ios)
 {
-	struct ore_io_state *ios_read;
 	struct ore_striping_info read_si;
 	struct __stripe_pages_2d *sp2d = ios->sp2d;
 	u64 offset = ios->si.first_stripe_start;
-	u64 last_stripe_end;
-	unsigned bytes_in_stripe = ios->si.bytes_in_stripe;
-	unsigned i, c, p, min_p = sp2d->pages_in_unit, max_p = -1;
-	int ret;
+	unsigned c, p, min_p = sp2d->pages_in_unit, max_p = -1;
 
 	if (offset == ios->offset) /* Go to start collect $200 */
 		goto read_last_stripe;
@@ -478,6 +474,9 @@ static int _read_4_write(struct ore_io_state *ios)
 	min_p = _sp2d_min_pg(sp2d);
 	max_p = _sp2d_max_pg(sp2d);
 
+	ORE_DBGMSG("stripe_start=0x%llx ios->offset=0x%llx min_p=%d max_p=%d\n",
+		   offset, ios->offset, min_p, max_p);
+
 	for (c = 0; ; c++) {
 		ore_calc_stripe_info(ios->layout, offset, 0, &read_si);
 		read_si.obj_offset += min_p * PAGE_SIZE;
@@ -512,6 +511,18 @@ static int _read_4_write(struct ore_io_state *ios)
 	}
 
 read_last_stripe:
+	return 0;
+}
+
+static int _read_4_write_last_stripe(struct ore_io_state *ios)
+{
+	struct ore_striping_info read_si;
+	struct __stripe_pages_2d *sp2d = ios->sp2d;
+	u64 offset;
+	u64 last_stripe_end;
+	unsigned bytes_in_stripe = ios->si.bytes_in_stripe;
+	unsigned c, p, min_p = sp2d->pages_in_unit, max_p = -1;
+
 	offset = ios->offset + ios->length;
 	if (offset % PAGE_SIZE)
 		_add_to_r4w_last_page(ios, &offset);
@@ -527,15 +538,15 @@ read_last_stripe:
 	c = _dev_order(ios->layout->group_width * ios->layout->mirrors_p1,
 		       ios->layout->mirrors_p1, read_si.par_dev, read_si.dev);
 
-	BUG_ON(ios->si.first_stripe_start + bytes_in_stripe != last_stripe_end);
-	/* unaligned IO must be within a single stripe */
-
 	if (min_p == sp2d->pages_in_unit) {
 		/* Didn't do it yet */
 		min_p = _sp2d_min_pg(sp2d);
 		max_p = _sp2d_max_pg(sp2d);
 	}
 
+	ORE_DBGMSG("offset=0x%llx stripe_end=0x%llx min_p=%d max_p=%d\n",
+		   offset, last_stripe_end, min_p, max_p);
+
 	while (offset < last_stripe_end) {
 		struct __1_page_stripe *_1ps = &sp2d->_1p_stripes[p];
 
@@ -568,6 +579,15 @@ read_last_stripe:
 	}
 
 read_it:
+	return 0;
+}
+
+static int _read_4_write_execute(struct ore_io_state *ios)
+{
+	struct ore_io_state *ios_read;
+	unsigned i;
+	int ret;
+
 	ios_read = ios->ios_read_4_write;
 	if (!ios_read)
 		return 0;
@@ -591,6 +611,8 @@ read_it:
 	}
 
 	_mark_read4write_pages_uptodate(ios_read, ret);
+	ore_put_io_state(ios_read);
+	ios->ios_read_4_write = NULL; /* Might need a reuse at last stripe */
 	return 0;
 }
 
@@ -626,8 +648,11 @@ int _ore_add_parity_unit(struct ore_io_state *ios,
 			/* If first stripe, Read in all read4write pages
 			 * (if needed) before we calculate the first parity.
 			 */
-			_read_4_write(ios);
+			_read_4_write_first_stripe(ios);
 		}
+		if (!cur_len) /* If last stripe r4w pages of last stripe */
+			_read_4_write_last_stripe(ios);
+		_read_4_write_execute(ios);
 
 		for (i = 0; i < num_pages; i++) {
 			pages[i] = _raid_page_alloc();
@@ -654,34 +679,14 @@ int _ore_add_parity_unit(struct ore_io_state *ios,
 
 int _ore_post_alloc_raid_stuff(struct ore_io_state *ios)
 {
-	struct ore_layout *layout = ios->layout;
-
 	if (ios->parity_pages) {
+		struct ore_layout *layout = ios->layout;
 		unsigned pages_in_unit = layout->stripe_unit / PAGE_SIZE;
-		unsigned stripe_size = ios->si.bytes_in_stripe;
-		u64 last_stripe, first_stripe;
 
 		if (_sp2d_alloc(pages_in_unit, layout->group_width,
 				layout->parity, &ios->sp2d)) {
 			return -ENOMEM;
 		}
-
-		/* Round io down to last full strip */
-		first_stripe = div_u64(ios->offset, stripe_size);
-		last_stripe = div_u64(ios->offset + ios->length, stripe_size);
-
-		/* If an IO spans more then a single stripe it must end at
-		 * a stripe boundary. The reminder at the end is pushed into the
-		 * next IO.
-		 */
-		if (last_stripe != first_stripe) {
-			ios->length = last_stripe * stripe_size - ios->offset;
-
-			BUG_ON(!ios->length);
-			ios->nr_pages = (ios->length + PAGE_SIZE - 1) /
-					PAGE_SIZE;
-			ios->si.length = ios->length; /*make it consistent */
-		}
 	}
 	return 0;
 }
diff --git a/fs/exofs/super.c b/fs/exofs/super.c
index 433783624d1..dde41a75c7c 100644
--- a/fs/exofs/super.c
+++ b/fs/exofs/super.c
@@ -400,8 +400,6 @@ static int exofs_sync_fs(struct super_block *sb, int wait)
 	ret = ore_write(ios);
 	if (unlikely(ret))
 		EXOFS_ERR("%s: ore_write failed.\n", __func__);
-	else
-		sb->s_dirt = 0;
 
 
 	unlock_super(sb);
@@ -412,14 +410,6 @@ out:
 	return ret;
 }
 
-static void exofs_write_super(struct super_block *sb)
-{
-	if (!(sb->s_flags & MS_RDONLY))
-		exofs_sync_fs(sb, 1);
-	else
-		sb->s_dirt = 0;
-}
-
 static void _exofs_print_device(const char *msg, const char *dev_path,
 				struct osd_dev *od, u64 pid)
 {
@@ -952,7 +942,6 @@ static const struct super_operations exofs_sops = {
 	.write_inode    = exofs_write_inode,
 	.evict_inode    = exofs_evict_inode,
 	.put_super      = exofs_put_super,
-	.write_super    = exofs_write_super,
 	.sync_fs	= exofs_sync_fs,
 	.statfs         = exofs_statfs,
 };