LU-3182 lmv: fix duplicate directory entries

In 4e57f6dd3a156e35ccb587fc5c003805dd73ecb7, I accidentally
introduced a new way for duplicate directory entries to be returned
from readdir().  That patch fails to properly decrement the nlupgs
counter when breaking out of the inner-for loop.  This accounting
error causes an extra iteration of the inner-for loop when processing
the next cfs page and a bad ldp_hash_end value is then saved in the
lu_dirpage.  To fix this, always decrement the nlupgs counter on
entry into the inner loop.

Note: this bug only affects architectures with > 4k-sized pages, e.g.
PowerPC.

Signed-off-by: Ned Bass <bass6@llnl.gov>
Change-Id: I9fa0fd8f34081f834f1d8b4df633d6e08926dcc9
Reviewed-on: http://review.whamcloud.com/6405
Reviewed-by: Fan Yong <fan.yong@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Bobi Jam <bobijam@gmail.com>
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>

diff --git a/lustre/lmv/lmv_obd.c b/lustre/lmv/lmv_obd.c
index c3a93bf..5f8a2ee 100644 (file)
--- a/lustre/lmv/lmv_obd.c
+++ b/lustre/lmv/lmv_obd.c
@@ -2186,7 +2186,7 @@ static void lmv_adjust_dirpages(struct page **pages, int ncfspgs, int nlupgs)
                __u64                   hash_end = dp->ldp_hash_end;
                __u32                   flags = dp->ldp_flags;
 
-               for (; nlupgs > 1; nlupgs--) {
+               while (--nlupgs > 0) {
                        ent = lu_dirent_start(dp);
                        for (end_dirent = ent; ent != NULL;
                             end_dirent = ent, ent = lu_dirent_next(ent));
@@ -2220,6 +2220,7 @@ static void lmv_adjust_dirpages(struct page **pages, int ncfspgs, int nlupgs)
 
                kunmap(pages[i]);
        }
+       LASSERTF(nlupgs == 0, "left = %d", nlupgs);
 }
 #else
 #define lmv_adjust_dirpages(pages, ncfspgs, nlupgs) do {} while (0)

diff --git a/lustre/tests/sanity.sh b/lustre/tests/sanity.sh
index d64d161..a489aba 100644 (file)
--- a/lustre/tests/sanity.sh
+++ b/lustre/tests/sanity.sh
@@ -1114,6 +1114,21 @@ test_24z() {
 }
 run_test 24z "rename one remote dir to another remote dir should fail"
 
+test_24A() { # LU-3182
+       local NFILES=5000
+
+       mkdir -p $DIR/$tdir
+       createmany -m $DIR/$tdir/$tfile $NFILES
+       local t=`ls $DIR/$tdir | wc -l`
+       local u=`ls $DIR/$tdir | sort -u | wc -l`
+       if [ $t -ne $NFILES -o $u -ne $NFILES ] ; then
+               error "Expected $NFILES files, got $t ($u unique)"
+       fi
+
+       rm -rf $DIR/$tdir || error "Can not delete directories"
+}
+run_test 24A "readdir() returns correct number of entries."
+
 test_25a() {
        echo '== symlink sanity ============================================='