kclient: use capsnap to get snapped file size

diff --git a/src/kernel/addr.c b/src/kernel/addr.c
index 07265f36cd7b7e549d7d1ec24b68af3b1bc455c8..44e24d72556ca2ca036210f4c0f32c02b2901fc7 100644 (file)
--- a/src/kernel/addr.c
+++ b/src/kernel/addr.c
@@ -339,7 +339,8 @@ out:
  *
  * Caller holds i_lock.
  */
-static struct ceph_snap_context *__get_oldest_context(struct inode *inode)
+static struct ceph_snap_context *__get_oldest_context(struct inode *inode,
+                                                     u64 *snap_size)
 {
        struct ceph_inode_info *ci = ceph_inode(inode);
        struct ceph_snap_context *snapc = NULL;
@@ -350,12 +351,14 @@ static struct ceph_snap_context *__get_oldest_context(struct inode *inode)
                capsnap = list_entry(p, struct ceph_cap_snap, ci_item);
                dout(20, " cap_snap %p snapc %p has %d dirty pages\n", capsnap,
                     capsnap->context, capsnap->dirty_pages);
-               if (capsnap->dirty_pages)
+               if (capsnap->dirty_pages) {
+                       snapc = ceph_get_snap_context(capsnap->context);
+                       if (snap_size)
+                               *snap_size = capsnap->size;
                        break;
+               }
        }
-       if (capsnap && capsnap->dirty_pages) {
-               snapc = ceph_get_snap_context(capsnap->context);
-       } else if (ci->i_snap_realm) {
+       if (!snapc && ci->i_snap_realm) {
                snapc = ceph_get_snap_context(ci->i_snap_realm->cached_context);
                dout(20, " head snapc %p has %d dirty pages\n",
                     snapc, ci->i_wrbuffer_ref_head);
@@ -363,12 +366,13 @@ static struct ceph_snap_context *__get_oldest_context(struct inode *inode)
        return snapc;
 }
 
-static struct ceph_snap_context *get_oldest_context(struct inode *inode)
+static struct ceph_snap_context *get_oldest_context(struct inode *inode,
+                                                   u64 *snap_size)
 {
        struct ceph_snap_context *snapc = NULL;
 
        spin_lock(&inode->i_lock);
-       snapc = __get_oldest_context(inode);
+       snapc = __get_oldest_context(inode, snap_size);
        spin_unlock(&inode->i_lock);
        return snapc;
 }
@@ -391,6 +395,7 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc)
        loff_t i_size;
        int err = 0;
        struct ceph_snap_context *snapc;
+       u64 snap_size = 0;
 
        dout(10, "writepage %p idx %lu\n", page, page->index);
 
@@ -402,22 +407,13 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc)
        ci = ceph_inode(inode);
        osdc = &ceph_inode_to_client(inode)->osdc;
 
-       /* is this a partial page at end of file? */
-       spin_lock(&inode->i_lock);
-       i_size = ci->i_as_size;
-       if (i_size < page_off + len)
-               len = i_size - page_off;
-       spin_unlock(&inode->i_lock);
-       dout(10, "writepage %p page %p index %lu on %llu~%u\n",
-            inode, page, page->index, page_off, len);
-
        /* verify this is a writeable snap context */
        snapc = (void *)page->private;
        if (snapc == NULL) {
                dout(20, "writepage %p page %p not dirty?\n", inode, page);
                goto out;
        }
-       if (snapc != get_oldest_context(inode)) {
+       if (snapc != get_oldest_context(inode, &snap_size)) {
                dout(10, "writepage %p page %p snapc %p not writeable - noop\n",
                     inode, page, (void *)page->private);
                /* we should only noop if called by kswapd */
@@ -425,6 +421,17 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc)
                goto out;
        }
 
+       /* is this a partial page at end of file? */
+       if (snap_size)
+               i_size = snap_size;
+       else
+               i_size = i_size_read(inode);
+       if (i_size < page_off + len)
+               len = i_size - page_off;
+
+       dout(10, "writepage %p page %p index %lu on %llu~%u\n",
+            inode, page, page->index, page_off, len);
+
        set_page_writeback(page);
        err = ceph_osdc_writepages(osdc, ceph_vino(inode),
                                   &ci->i_layout, snapc,
@@ -565,6 +572,7 @@ static int ceph_writepages_start(struct address_space *mapping,
        unsigned wsize = 1 << inode->i_blkbits;
        struct ceph_osd_request *req = NULL;
        int do_sync;
+       u64 snap_size = 0;
 
        /*
         * Include a 'sync' in the OSD request if this is a data
@@ -618,7 +626,7 @@ static int ceph_writepages_start(struct address_space *mapping,
 retry:
        /* find oldest snap context with dirty data */
        ceph_put_snap_context(snapc);
-       snapc = get_oldest_context(inode);
+       snapc = get_oldest_context(inode, &snap_size);
        if (!snapc) {
                /* hmm, why does writepages get called when there
                   is no dirty data? */
@@ -643,7 +651,7 @@ retry:
                int pvec_pages, locked_pages;
                struct page *page;
                int want;
-               u64 offset, len, as_size;
+               u64 offset, len;
                struct ceph_osd_request_head *reqhead;
                struct ceph_osd_op *op;         
 
@@ -697,11 +705,11 @@ get_more_pages:
                                dout(20, "waiting on writeback %p\n", page);
                                wait_on_page_writeback(page);
                        }
-                       spin_lock(&inode->i_lock);
-                       as_size = ci->i_as_size;
-                       spin_unlock(&inode->i_lock);
-                       if (page_offset(page) >= as_size) {
-                               dout(20, "%p > as_size %llu\n", page, as_size);
+                       if ((snap_size && page_offset(page) > snap_size) ||
+                           (!snap_size &&
+                            page_offset(page) > i_size_read(inode))) {
+                               dout(20, "%p page eof %llu\n", page, snap_size ?
+                                    snap_size : i_size_read(inode));
                                done = 1;
                                unlock_page(page);
                                break;
@@ -793,10 +801,8 @@ get_more_pages:
 
                /* submit the write */
                offset = req->r_pages[0]->index << PAGE_CACHE_SHIFT;
-               spin_lock(&inode->i_lock);
-               len = min(ci->i_as_size - offset,
+               len = min((snap_size ? snap_size : i_size_read(inode)) - offset,
                          (u64)locked_pages << PAGE_CACHE_SHIFT);
-               spin_unlock(&inode->i_lock);
                dout(10, "writepages got %d pages at %llu~%llu\n",
                     locked_pages, offset, len);
 
@@ -860,7 +866,7 @@ out_free:
 static int context_is_writeable_or_written(struct inode *inode,
                                           struct ceph_snap_context *snapc)
 {
-       struct ceph_snap_context *oldest = get_oldest_context(inode);
+       struct ceph_snap_context *oldest = get_oldest_context(inode, NULL);
        return !oldest || snapc->seq <= oldest->seq;
 }
 
@@ -912,7 +918,7 @@ retry_locked:
                 * this page is already dirty in another (older) snap
                 * context!  is it writeable now?
                 */
-               snapc = get_oldest_context(inode);
+               snapc = get_oldest_context(inode, NULL);
                up_read(&mdsc->snap_rwsem);
 
                if (snapc != (void *)page->private) {

diff --git a/src/kernel/inode.c b/src/kernel/inode.c
index 8c2c43d6f42083d7e2f443302f96220ecdd37e97..9fe9ae1523c7d287bec726e1c63d17c87c2fc4bb 100644 (file)
--- a/src/kernel/inode.c
+++ b/src/kernel/inode.c
@@ -288,7 +288,6 @@ struct inode *ceph_alloc_inode(struct super_block *sb)
        for (i = 0; i < CEPH_FILE_MODE_NUM; i++)
                ci->i_nr_by_mode[i] = 0;
 
-       ci->i_as_size = 0;
        ci->i_truncate_seq = 0;
        ci->i_truncate_size = 0;
        ci->i_truncate_pending = 0;
@@ -365,8 +364,6 @@ int ceph_fill_file_size(struct inode *inode, int issued,
                inode->i_size = size;
                inode->i_blocks = (size + (1<<9) - 1) >> 9;
                ci->i_reported_size = size;
-               if (ci->i_as_size < size)
-                       ci->i_as_size = size;
                if (truncate_seq != ci->i_truncate_seq) {
                        dout(10, "truncate_seq %u -> %u\n",
                             ci->i_truncate_seq, truncate_seq);
@@ -1156,8 +1153,6 @@ int ceph_inode_set_size(struct inode *inode, loff_t size)
        dout(30, "set_size %p %llu -> %llu\n", inode, inode->i_size, size);
        inode->i_size = size;
        inode->i_blocks = (size + (1 << 9) - 1) >> 9;
-       if (ci->i_as_size < size)
-               ci->i_as_size = size;
 
        /* tell the MDS if we are approaching max_size */
        if ((size << 1) >= ci->i_max_size &&
@@ -1291,7 +1286,6 @@ retry:
        truncate_inode_pages(inode->i_mapping, to);
 
        spin_lock(&inode->i_lock);
-       ci->i_as_size = to;
        ci->i_truncate_pending--;
        if (ci->i_truncate_pending == 0)
                wake = 1;
@@ -1448,9 +1442,7 @@ int ceph_setattr(struct dentry *dentry, struct iattr *attr)
                if ((issued & CEPH_CAP_FILE_EXCL) &&
                    attr->ia_size > inode->i_size) {
                        inode->i_size = attr->ia_size;
-                       if (ci->i_as_size < attr->ia_size) {
-                               ci->i_as_size = attr->ia_size;
-                       } else {
+                       if (attr->ia_size < inode->i_size) {
                                ci->i_truncate_size = attr->ia_size;
                                ci->i_truncate_pending++;
                                queue_trunc = 1;

diff --git a/src/kernel/super.h b/src/kernel/super.h
index 3921a6e6931a8ab47ef4d76a9c7c2add55ec2b63..4651284acef196042806539bbbac7c89ffb0a781 100644 (file)
--- a/src/kernel/super.h
+++ b/src/kernel/super.h
@@ -320,7 +320,6 @@ struct ceph_inode_info {
 
        int i_nr_by_mode[CEPH_FILE_MODE_NUM];  /* open file counts */
 
-       loff_t i_as_size;          /* address space size (pre-truncation) */
        u32 i_truncate_seq;        /* last truncate to smaller size */
        u64 i_truncate_size;       /*  and the size we last truncated down to */
        int i_truncate_pending;    /*  still need to call vmtruncate */
@@ -804,6 +803,7 @@ extern int ceph_inode_set_size(struct inode *inode, loff_t size);
 extern void ceph_inode_writeback(struct work_struct *work);
 extern void ceph_vmtruncate_work(struct work_struct *work);
 extern void __ceph_do_pending_vmtruncate(struct inode *inode);
+extern void __ceph_queue_vmtruncate(struct inode *inode);
 
 extern int ceph_do_getattr(struct inode *inode, int mask);
 extern int ceph_permission(struct inode *inode, int mask);