IB/hfi1: Fix bug that blocks process on exit after port bounce

During the processing of a user SDMA request, if there was an
error before the request counter was increased, the state of
the packet queue could be updated incorrectly, causing the
counter to underflow. As the result, the process could get
stuck later since the counter could never get back to 0.

This patch adds a condition to guard the packet queue update
so that the counter is only decreased if it has been increased
before the error happens.

Reviewed-by: Mitko Haralanov <[email protected]>
Signed-off-by: Jianxin Xiong <[email protected]>
Signed-off-by: Doug Ledford <[email protected]>

diff --git a/drivers/staging/rdma/hfi1/user_sdma.c b/drivers/staging/rdma/hfi1/user_sdma.c
index aed2878c97f149155d38025d8534bc2098918676..29f4795f866cd19185c680d28f95302e206f62e4 100644 (file)
--- a/drivers/staging/rdma/hfi1/user_sdma.c
+++ b/drivers/staging/rdma/hfi1/user_sdma.c
@@ -510,6 +510,7 @@ int hfi1_user_sdma_process_request(struct file *fp, struct iovec *iovec,
        struct sdma_req_info info;
        struct user_sdma_request *req;
        u8 opcode, sc, vl;
+       int req_queued = 0;
 
        if (iovec[idx].iov_len < sizeof(info) + sizeof(req->hdr)) {
                hfi1_cdbg(
@@ -706,6 +707,7 @@ int hfi1_user_sdma_process_request(struct file *fp, struct iovec *iovec,
 
        set_comp_state(pq, cq, info.comp_idx, QUEUED, 0);
        atomic_inc(&pq->n_reqs);
+       req_queued = 1;
        /* Send the first N packets in the request to buy us some time */
        ret = user_sdma_send_pkts(req, pcount);
        if (unlikely(ret < 0 && ret != -EBUSY)) {
@@ -750,7 +752,8 @@ int hfi1_user_sdma_process_request(struct file *fp, struct iovec *iovec,
        return 0;
 free_req:
        user_sdma_free_request(req, true);
-       pq_update(pq);
+       if (req_queued)
+               pq_update(pq);
        set_comp_state(pq, cq, info.comp_idx, ERROR, req->status);
        return ret;
 }