ceph: sync read inline data

we can't use getattr to fetch inline data while holding Fr cap,
because it can cause deadlock. If we need to sync read inline data,
drop cap refs first, then use getattr to fetch inline data.

Signed-off-by: Yan, Zheng <zyan@redhat.com>

1 files changed, 61 insertions, 5 deletions

diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c
index 5d2b88e3ff0b..13413d7440d6 100644
--- a/fs/ceph/addr.c
+++ b/fs/ceph/addr.c
@@ -192,17 +192,30 @@ static int readpage_nounlock(struct file *filp, struct page *page)
         struct ceph_osd_client *osdc =
                 &ceph_inode_to_client(inode)->client->osdc;
         int err = 0;
+        u64 off = page_offset(page);
         u64 len = PAGE_CACHE_SIZE;
 
-        err = ceph_readpage_from_fscache(inode, page);
+        if (off >= i_size_read(inode)) {
+                zero_user_segment(page, err, PAGE_CACHE_SIZE);
+                SetPageUptodate(page);
+                return 0;
+        }
 
+        /*
+         * Uptodate inline data should have been added into page cache
+         * while getting Fcr caps.
+         */
+        if (ci->i_inline_version != CEPH_INLINE_NONE)
+                return -EINVAL;
+
+        err = ceph_readpage_from_fscache(inode, page);
         if (err == 0)
                 goto out;
 
         dout("readpage inode %p file %p page %p index %lu\n",
              inode, filp, page, page->index);
         err = ceph_osdc_readpages(osdc, ceph_vino(inode), &ci->i_layout,
-                                  (u64) page_offset(page), &len,
+                                  off, &len,
                                   ci->i_truncate_seq, ci->i_truncate_size,
                                   &page, 1, 0);
         if (err == -ENOENT)
@@ -384,6 +397,9 @@ static int ceph_readpages(struct file *file, struct address_space *mapping,
         int rc = 0;
         int max = 0;
 
+        if (ceph_inode(inode)->i_inline_version != CEPH_INLINE_NONE)
+                return -EINVAL;
+
         rc = ceph_readpages_from_fscache(mapping->host, mapping, page_list,
                                          &nr_pages);
 
@@ -1219,8 +1235,8 @@ static int ceph_filemap_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
                 want = CEPH_CAP_FILE_CACHE;
         while (1) {
                 got = 0;
-                ret = ceph_get_caps(ci, CEPH_CAP_FILE_RD, want, -1,
-                                    &got, &pinned_page);
+                ret = ceph_get_caps(ci, CEPH_CAP_FILE_RD, want,
+                                    -1, &got, &pinned_page);
                 if (ret == 0)
                         break;
                 if (ret != -ERESTARTSYS) {
@@ -1231,7 +1247,11 @@ static int ceph_filemap_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
         dout("filemap_fault %p %llu~%zd got cap refs on %s\n",
              inode, off, (size_t)PAGE_CACHE_SIZE, ceph_cap_string(got));
 
-        ret = filemap_fault(vma, vmf);
+        if ((got & (CEPH_CAP_FILE_CACHE | CEPH_CAP_FILE_LAZYIO)) ||
+            ci->i_inline_version == CEPH_INLINE_NONE)
+                ret = filemap_fault(vma, vmf);
+        else
+                ret = -EAGAIN;
 
         dout("filemap_fault %p %llu~%zd dropping cap refs on %s ret %d\n",
              inode, off, (size_t)PAGE_CACHE_SIZE, ceph_cap_string(got), ret);
@@ -1239,6 +1259,42 @@ static int ceph_filemap_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
                 page_cache_release(pinned_page);
         ceph_put_cap_refs(ci, got);
 
+        if (ret != -EAGAIN)
+                return ret;
+
+        /* read inline data */
+        if (off >= PAGE_CACHE_SIZE) {
+                /* does not support inline data > PAGE_SIZE */
+                ret = VM_FAULT_SIGBUS;
+        } else {
+                int ret1;
+                struct address_space *mapping = inode->i_mapping;
+                struct page *page = find_or_create_page(mapping, 0,
+                                                mapping_gfp_mask(mapping) &
+                                                ~__GFP_FS);
+                if (!page) {
+                        ret = VM_FAULT_OOM;
+                        goto out;
+                }
+                ret1 = __ceph_do_getattr(inode, page,
+                                         CEPH_STAT_CAP_INLINE_DATA, true);
+                if (ret1 < 0 || off >= i_size_read(inode)) {
+                        unlock_page(page);
+                        page_cache_release(page);
+                        ret = VM_FAULT_SIGBUS;
+                        goto out;
+                }
+                if (ret1 < PAGE_CACHE_SIZE)
+                        zero_user_segment(page, ret1, PAGE_CACHE_SIZE);
+                else
+                        flush_dcache_page(page);
+                SetPageUptodate(page);
+                vmf->page = page;
+                ret = VM_FAULT_MAJOR | VM_FAULT_LOCKED;
+        }
+out:
+        dout("filemap_fault %p %llu~%zd read inline data ret %d\n",
+             inode, off, (size_t)PAGE_CACHE_SIZE, ret);
         return ret;
 }