| Message ID | 1456334649-21437-13-git-send-email-sagig@mellanox.com (mailing list archive) |
|---|---|
| State | Not Applicable |
| Headers | show |
> Now the rdma core offers a QP draining service, use it > instead of our own. This assumes that Doug is taking Steve's QP draining series to 4.6 -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Hi Sagi,
[auto build test ERROR on target/master]
[also build test ERROR on v4.5-rc5 next-20160224]
[if your patch is applied to the wrong git tree, please drop us a note to help improving the system]
url: https://github.com/0day-ci/linux/commits/Sagi-Grimberg/iser-target-patches-for-4-6/20160225-012959
base: https://git.kernel.org/pub/scm/linux/kernel/git/nab/target-pending.git master
config: sparc64-allyesconfig (attached as .config)
reproduce:
wget https://git.kernel.org/cgit/linux/kernel/git/wfg/lkp-tests.git/plain/sbin/make.cross -O ~/bin/make.cross
chmod +x ~/bin/make.cross
# save the attached .config to linux build tree
make.cross ARCH=sparc64
All errors (new ones prefixed by >>):
drivers/infiniband/ulp/isert/ib_isert.c: In function 'isert_disconnected_handler':
>> drivers/infiniband/ulp/isert/ib_isert.c:889:3: error: implicit declaration of function 'ib_drain_qp' [-Werror=implicit-function-declaration]
ib_drain_qp(isert_conn->qp);
^
cc1: some warnings being treated as errors
vim +/ib_drain_qp +889 drivers/infiniband/ulp/isert/ib_isert.c
883 mutex_lock(&isert_conn->mutex);
884 switch (isert_conn->state) {
885 case ISER_CONN_TERMINATING:
886 break;
887 case ISER_CONN_UP:
888 isert_conn_terminate(isert_conn);
> 889 ib_drain_qp(isert_conn->qp);
890 isert_handle_unbound_conn(isert_conn);
891 break;
892 case ISER_CONN_BOUND:
---
0-DAY kernel test infrastructure Open Source Technology Center
https://lists.01.org/pipermail/kbuild-all Intel Corporation
> Hi Sagi, > > [auto build test ERROR on target/master] > [also build test ERROR on v4.5-rc5 next-20160224] > [if your patch is applied to the wrong git tree, please drop us a note to help improving the system] OK, this patch depends on a new API that is expected to be included from doug's tree, and that is why compilation fails. Nic, should we just push this in rc1 after doug pull request landed? -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Thu, 2016-02-25 at 10:49 +0200, Sagi Grimberg wrote: > > Hi Sagi, > > > > [auto build test ERROR on target/master] > > [also build test ERROR on v4.5-rc5 next-20160224] > > [if your patch is applied to the wrong git tree, please drop us a note to help improving the system] > > OK, this patch depends on a new API that is expected to be included from > doug's tree, and that is why compilation fails. > Er, missed this earlier message.. > Nic, should we just push this in rc1 after doug pull request landed? Sure, I'll plan to include in the post -rc1 fixes PULL request. -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
> Er, missed this earlier message.. > >> Nic, should we just push this in rc1 after doug pull request landed? > > Sure, I'll plan to include in the post -rc1 fixes PULL request. Thanks Nic! -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Now that ib_drain_qp and the iser patches in mainline it would be good to get this one in as well through an update to either tree. -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Wed, 2016-03-23 at 00:48 -0700, Christoph Hellwig wrote: > Now that ib_drain_qp and the iser patches in mainline it would be good > to get this one in as well through an update to either tree. > -- Applied to target-pending/master for -rc2. Thanks folks. -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
> Applied to target-pending/master for -rc2. > > Thanks folks. Thanks Nic. -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
diff --git a/drivers/infiniband/ulp/isert/ib_isert.c b/drivers/infiniband/ulp/isert/ib_isert.c index 57e4edd661c5..411e4464ca23 100644 --- a/drivers/infiniband/ulp/isert/ib_isert.c +++ b/drivers/infiniband/ulp/isert/ib_isert.c @@ -63,7 +63,6 @@ isert_rdma_accept(struct isert_conn *isert_conn); struct rdma_cm_id *isert_setup_id(struct isert_np *isert_np); static void isert_release_work(struct work_struct *work); -static void isert_wait4flush(struct isert_conn *isert_conn); static void isert_recv_done(struct ib_cq *cq, struct ib_wc *wc); static void isert_send_done(struct ib_cq *cq, struct ib_wc *wc); static void isert_login_recv_done(struct ib_cq *cq, struct ib_wc *wc); @@ -141,7 +140,7 @@ isert_create_qp(struct isert_conn *isert_conn, attr.qp_context = isert_conn; attr.send_cq = comp->cq; attr.recv_cq = comp->cq; - attr.cap.max_send_wr = ISERT_QP_MAX_REQ_DTOS; + attr.cap.max_send_wr = ISERT_QP_MAX_REQ_DTOS + 1; attr.cap.max_recv_wr = ISERT_QP_MAX_RECV_DTOS + 1; attr.cap.max_send_sge = device->ib_device->attrs.max_sge; isert_conn->max_sge = min(device->ib_device->attrs.max_sge, @@ -887,7 +886,7 @@ isert_disconnected_handler(struct rdma_cm_id *cma_id, break; case ISER_CONN_UP: isert_conn_terminate(isert_conn); - isert_wait4flush(isert_conn); + ib_drain_qp(isert_conn->qp); isert_handle_unbound_conn(isert_conn); break; case ISER_CONN_BOUND: @@ -3213,36 +3212,6 @@ isert_wait4cmds(struct iscsi_conn *conn) } } -static void -isert_beacon_done(struct ib_cq *cq, struct ib_wc *wc) -{ - struct isert_conn *isert_conn = wc->qp->qp_context; - - isert_print_wc(wc); - - isert_info("conn %p completing wait_comp_err\n", isert_conn); - complete(&isert_conn->wait_comp_err); -} - -static void -isert_wait4flush(struct isert_conn *isert_conn) -{ - struct ib_recv_wr *bad_wr; - static struct ib_cqe cqe = { .done = isert_beacon_done }; - - isert_info("conn %p\n", isert_conn); - - init_completion(&isert_conn->wait_comp_err); - isert_conn->beacon.wr_cqe = &cqe; - /* post an indication that all flush errors were consumed */ - if (ib_post_recv(isert_conn->qp, &isert_conn->beacon, &bad_wr)) { - isert_err("conn %p failed to post beacon", isert_conn); - return; - } - - wait_for_completion(&isert_conn->wait_comp_err); -} - /** * isert_put_unsol_pending_cmds() - Drop commands waiting for * unsolicitate dataout @@ -3288,7 +3257,7 @@ static void isert_wait_conn(struct iscsi_conn *conn) isert_conn_terminate(isert_conn); mutex_unlock(&isert_conn->mutex); - isert_wait4flush(isert_conn); + ib_drain_qp(isert_conn->qp); isert_put_unsol_pending_cmds(conn); isert_wait4cmds(conn); isert_wait4logout(isert_conn); @@ -3300,7 +3269,7 @@ static void isert_free_conn(struct iscsi_conn *conn) { struct isert_conn *isert_conn = conn->context; - isert_wait4flush(isert_conn); + ib_drain_qp(isert_conn->qp); isert_put_conn(isert_conn); } diff --git a/drivers/infiniband/ulp/isert/ib_isert.h b/drivers/infiniband/ulp/isert/ib_isert.h index 192788a4820c..147900cbb578 100644 --- a/drivers/infiniband/ulp/isert/ib_isert.h +++ b/drivers/infiniband/ulp/isert/ib_isert.h @@ -209,14 +209,12 @@ struct isert_conn { struct ib_qp *qp; struct isert_device *device; struct mutex mutex; - struct completion wait_comp_err; struct kref kref; struct list_head fr_pool; int fr_pool_size; /* lock to protect fastreg pool */ spinlock_t pool_lock; struct work_struct release_work; - struct ib_recv_wr beacon; bool logout_posted; bool snd_w_inv; };
Now the rdma core offers a QP draining service, use it instead of our own. Signed-off-by: Sagi Grimberg <sagig@mellanox.com> --- drivers/infiniband/ulp/isert/ib_isert.c | 39 ++++----------------------------- drivers/infiniband/ulp/isert/ib_isert.h | 2 -- 2 files changed, 4 insertions(+), 37 deletions(-)