LU-8509 llite: drop_caches hangs in cl_inode_fini()

This patch releases cl_pages on error in ll_write_begin()
to avoid memory and object reference leaks. Also, it
reuses per-cpu lu_env in ll_invalidatepage() in the same
way as done in ll_releasepage().

Change-Id: I11a658941aff4dacd3cc83685ae3df67f28093a2
Signed-off-by: Andrew Perepechko <andrew.perepechko@seagate.com>
Seagate-bug-id: MRP-3504
Reviewed-on: http://review.whamcloud.com/22745
Reviewed-by: Jinshan Xiong <jinshan.xiong@intel.com>
Reviewed-by: Bobi Jam <bobijam@hotmail.com>
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

diff --git a/lustre/llite/rw26.c b/lustre/llite/rw26.c
index 1369c09..acf5a64 100644 (file)
--- a/lustre/llite/rw26.c
+++ b/lustre/llite/rw26.c
@@ -88,8 +88,6 @@ static void ll_invalidatepage(struct page *vmpage,
         struct cl_page   *page;
         struct cl_object *obj;
 
-       __u16 refcheck;
-
         LASSERT(PageLocked(vmpage));
         LASSERT(!PageWriteback(vmpage));
 
@@ -103,20 +101,22 @@ static void ll_invalidatepage(struct page *vmpage,
 #else
        if (offset == 0) {
 #endif
-                env = cl_env_get(&refcheck);
-                if (!IS_ERR(env)) {
-                        inode = vmpage->mapping->host;
-                        obj = ll_i2info(inode)->lli_clob;
-                        if (obj != NULL) {
-                                page = cl_vmpage_page(vmpage, obj);
-                                if (page != NULL) {
-                                        cl_page_delete(env, page);
-                                        cl_page_put(env, page);
-                                }
-                        } else
-                                LASSERT(vmpage->private == 0);
-                        cl_env_put(env, &refcheck);
-                }
+               /* See the comment in ll_releasepage() */
+               env = cl_env_percpu_get();
+               LASSERT(!IS_ERR(env));
+
+               inode = vmpage->mapping->host;
+               obj = ll_i2info(inode)->lli_clob;
+               if (obj != NULL) {
+                       page = cl_vmpage_page(vmpage, obj);
+                       if (page != NULL) {
+                               cl_page_delete(env, page);
+                               cl_page_put(env, page);
+                       }
+               } else
+                       LASSERT(vmpage->private == 0);
+
+               cl_env_percpu_put(env);
         }
 }
 
@@ -632,9 +632,9 @@ static int ll_write_begin(struct file *file, struct address_space *mapping,
                          struct page **pagep, void **fsdata)
 {
        struct ll_cl_context *lcc;
-       const struct lu_env  *env;
+       const struct lu_env  *env = NULL;
        struct cl_io   *io;
-       struct cl_page *page;
+       struct cl_page *page = NULL;
 
        struct cl_object *clob = ll_i2info(mapping->host)->lli_clob;
        pgoff_t index = pos >> PAGE_SHIFT;
@@ -721,6 +721,10 @@ out:
                        unlock_page(vmpage);
                        put_page(vmpage);
                }
+               if (!IS_ERR_OR_NULL(page)) {
+                       lu_ref_del(&page->cp_reference, "cl_io", io);
+                       cl_page_put(env, page);
+               }
                if (io)
                        io->ci_result = result;
        } else {

diff --git a/lustre/tests/sanity.sh b/lustre/tests/sanity.sh
index 320c39a..810f99d 100755 (executable)
--- a/lustre/tests/sanity.sh
+++ b/lustre/tests/sanity.sh
@@ -15790,6 +15790,26 @@ test_407() {
 }
 run_test 407 "transaction fail should cause operation fail"
 
+test_408() {
+       dd if=/dev/zero of=$DIR/$tfile bs=4096 count=1 oflag=direct
+
+       #define OBD_FAIL_OSC_BRW_PREP_REQ2        0x40a
+       lctl set_param fail_loc=0x8000040a
+       # let ll_prepare_partial_page() fail
+       dd if=/dev/zero of=$DIR/$tfile bs=2048 count=1 conv=notrunc || true
+
+       rm -f $DIR/$tfile
+
+       # create at least 100 unused inodes so that
+       # shrink_icache_memory(0) should not return 0
+       touch $DIR/$tfile-{0..100}
+       rm -f $DIR/$tfile-{0..100}
+       sync
+
+       echo 2 > /proc/sys/vm/drop_caches
+}
+run_test 408 "drop_caches should not hang due to page leaks"
+
 #
 # tests that do cleanup/setup should be run at the end
 #