From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754106Ab3JBObf (ORCPT ); Wed, 2 Oct 2013 10:31:35 -0400 Received: from cantor2.suse.de ([195.135.220.15]:49027 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753319Ab3JBO3A (ORCPT ); Wed, 2 Oct 2013 10:29:00 -0400 From: Jan Kara To: LKML Cc: linux-mm@kvack.org, Jan Kara , Mike Marciniszyn , Roland Dreier , linux-rdma@vger.kernel.org Subject: [PATCH 24/26] ib: Convert qib_user_sdma_pin_pages() to use get_user_pages_unlocked() Date: Wed, 2 Oct 2013 16:28:05 +0200 Message-Id: <1380724087-13927-25-git-send-email-jack@suse.cz> X-Mailer: git-send-email 1.8.1.4 In-Reply-To: <1380724087-13927-1-git-send-email-jack@suse.cz> References: <1380724087-13927-1-git-send-email-jack@suse.cz> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Function qib_user_sdma_queue_pkts() gets called with mmap_sem held for writing. Except for get_user_pages() deep down in qib_user_sdma_pin_pages() we don't seem to need mmap_sem at all. Even more interestingly the function qib_user_sdma_queue_pkts() (and also qib_user_sdma_coalesce() called somewhat later) call copy_from_user() which can hit a page fault and we deadlock on trying to get mmap_sem when handling that fault. So just make qib_user_sdma_pin_pages() use get_user_pages_unlocked() and leave mmap_sem locking for mm. CC: Mike Marciniszyn CC: Roland Dreier CC: linux-rdma@vger.kernel.org Signed-off-by: Jan Kara --- drivers/infiniband/hw/qib/qib_user_sdma.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) diff --git a/drivers/infiniband/hw/qib/qib_user_sdma.c b/drivers/infiniband/hw/qib/qib_user_sdma.c index d0a0ea0c14d6..c1b6463acd59 100644 --- a/drivers/infiniband/hw/qib/qib_user_sdma.c +++ b/drivers/infiniband/hw/qib/qib_user_sdma.c @@ -594,8 +594,8 @@ static int qib_user_sdma_pin_pages(const struct qib_devdata *dd, else j = npages; - ret = get_user_pages(current, current->mm, addr, - j, 0, 1, pages, NULL); + ret = get_user_pages_unlocked(current, current->mm, addr, + j, 0, 1, pages); if (ret != j) { i = 0; j = ret; @@ -1294,11 +1294,8 @@ int qib_user_sdma_writev(struct qib_ctxtdata *rcd, int mxp = 8; int ndesc = 0; - down_write(¤t->mm->mmap_sem); ret = qib_user_sdma_queue_pkts(dd, ppd, pq, iov, dim, &list, &mxp, &ndesc); - up_write(¤t->mm->mmap_sem); - if (ret < 0) goto done_unlock; else { -- 1.8.1.4