[Sheepdog] [PATCH 1/2] collie: fix recovery thread race

[MORITA Kazutaka]

Signed-off-by: MORITA Kazutaka <morita.kazut...@lab.ntt.co.jp>
---
 collie/store.c |   34 +++++++++++++++++-----------------
 1 files changed, 17 insertions(+), 17 deletions(-)

diff --git a/collie/store.c b/collie/store.c
index f1dbdc3..1f17df4 100644
--- a/collie/store.c
+++ b/collie/store.c
@@ -1085,15 +1085,13 @@ static void recover_one_done(struct work *work, int idx)
 
        rw->done++;
 
-       if (rw->done < rw->count && rw->rw_siblings.next == 
&recovery_work_list) {
+       if (rw->done < rw->count && list_empty(&recovery_work_list)) {
                queue_work(dobj_queue, &rw->work);
                return;
        }
 
        dprintf("recovery done, %d\n", rw->epoch);
-       recovering--;
-
-       list_del(&rw->rw_siblings);
+       recovering = 0;
 
        free(rw->buf);
        free(rw);
@@ -1102,7 +1100,9 @@ static void recover_one_done(struct work *work, int idx)
                rw = list_first_entry(&recovery_work_list,
                                      struct recovery_work, rw_siblings);
 
-               recovering++;
+               list_del(&rw->rw_siblings);
+
+               recovering = 1;
                queue_work(dobj_queue, &rw->work);
        }
 }
@@ -1129,7 +1129,7 @@ static int __fill_obj_list(struct recovery_work *rw,
        }
 
        wlen = 0;
-       rlen = 1 << 20;
+       rlen = (1 << 20) - (rw->count * sizeof(uint64_t));
 
        memset(&hdr, 0, sizeof(hdr));
        hdr.opcode = SD_OP_GET_OBJ_LIST;
@@ -1150,10 +1150,10 @@ static int __fill_obj_list(struct recovery_work *rw,
 
        rsp = (struct sd_list_rsp *)&hdr;
 
-       if (rsp->result != SD_RES_SUCCESS) {
+       if (ret || rsp->result != SD_RES_SUCCESS) {
                rw->retry = 1;
                *done_hash = end_hash;
-               eprintf("try again, %d\n", rsp->result);
+               eprintf("try again, %d, %d\n", ret, rsp->result);
                return 0;
        }
 
@@ -1281,7 +1281,7 @@ static void __start_recovery_done(struct work *work, int 
idx)
                return;
        }
 
-       if (rw->count && rw->rw_siblings.next == &recovery_work_list) {
+       if (rw->count && list_empty(&recovery_work_list)) {
                rw->work.fn = recover_one;
                rw->work.done = recover_one_done;
 
@@ -1293,9 +1293,7 @@ static void __start_recovery_done(struct work *work, int 
idx)
        }
 
        dprintf("recovery done, %d\n", rw->epoch);
-       recovering--;
-
-       list_del(&rw->rw_siblings);
+       recovering = 0;
 
        free(rw->buf);
        free(rw);
@@ -1304,7 +1302,9 @@ static void __start_recovery_done(struct work *work, int 
idx)
                rw = list_first_entry(&recovery_work_list,
                                      struct recovery_work, rw_siblings);
 
-               recovering++;
+               list_del(&rw->rw_siblings);
+
+               recovering = 1;
                queue_work(dobj_queue, &rw->work);
        }
 }
@@ -1324,10 +1324,10 @@ int start_recovery(uint32_t epoch)
        rw->work.fn = __start_recovery;
        rw->work.done = __start_recovery_done;
 
-       list_add_tail(&rw->rw_siblings, &recovery_work_list);
-
-       if (!recovering) {
-               recovering++;
+       if (recovering)
+               list_add_tail(&rw->rw_siblings, &recovery_work_list);
+       else {
+               recovering = 1;
                queue_work(dobj_queue, &rw->work);
        }
 
-- 
1.5.6.5

-- 
sheepdog mailing list
sheepdog@lists.wpkg.org
http://lists.wpkg.org/mailman/listinfo/sheepdog