Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752771Ab3JWJqN (ORCPT ); Wed, 23 Oct 2013 05:46:13 -0400 Received: from cantor2.suse.de ([195.135.220.15]:58117 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751568Ab3JWJqL (ORCPT ); Wed, 23 Oct 2013 05:46:11 -0400 Message-ID: <52679ADD.3080202@suse.cz> Date: Wed, 23 Oct 2013 11:46:05 +0200 From: Vlastimil Babka User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Thunderbird/24.0 MIME-Version: 1.0 To: Davidlohr Bueso , Andrew Morton , Linus Torvalds Cc: Ingo Molnar , Michel Lespinasse , Peter Zijlstra , Rik van Riel , Tim Chen , aswin@hp.com, linux-mm , linux-kernel@vger.kernel.org Subject: Re: [PATCH 2/3] mm/mlock: prepare params outside critical region References: <1382057438-3306-1-git-send-email-davidlohr@hp.com> <1382057438-3306-3-git-send-email-davidlohr@hp.com> In-Reply-To: <1382057438-3306-3-git-send-email-davidlohr@hp.com> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2540 Lines: 84 On 10/18/2013 02:50 AM, Davidlohr Bueso wrote: > All mlock related syscalls prepare lock limits, lengths and > start parameters with the mmap_sem held. Move this logic > outside of the critical region. For the case of mlock, continue > incrementing the amount already locked by mm->locked_vm with > the rwsem taken. > > Signed-off-by: Davidlohr Bueso > Cc: Michel Lespinasse > Cc: Vlastimil Babka Acked-by: Vlastimil Babka > --- > mm/mlock.c | 18 +++++++++++------- > 1 file changed, 11 insertions(+), 7 deletions(-) > > diff --git a/mm/mlock.c b/mm/mlock.c > index d480cd6..aa7de13 100644 > --- a/mm/mlock.c > +++ b/mm/mlock.c > @@ -689,19 +689,21 @@ SYSCALL_DEFINE2(mlock, unsigned long, start, size_t, len) > > lru_add_drain_all(); /* flush pagevec */ > > - down_write(¤t->mm->mmap_sem); > len = PAGE_ALIGN(len + (start & ~PAGE_MASK)); > start &= PAGE_MASK; > > - locked = len >> PAGE_SHIFT; > - locked += current->mm->locked_vm; > - > lock_limit = rlimit(RLIMIT_MEMLOCK); > lock_limit >>= PAGE_SHIFT; > + locked = len >> PAGE_SHIFT; > + > + down_write(¤t->mm->mmap_sem); > + > + locked += current->mm->locked_vm; > > /* check against resource limits */ > if ((locked <= lock_limit) || capable(CAP_IPC_LOCK)) > error = do_mlock(start, len, 1); > + > up_write(¤t->mm->mmap_sem); > if (!error) > error = __mm_populate(start, len, 0); > @@ -712,11 +714,13 @@ SYSCALL_DEFINE2(munlock, unsigned long, start, size_t, len) > { > int ret; > > - down_write(¤t->mm->mmap_sem); > len = PAGE_ALIGN(len + (start & ~PAGE_MASK)); > start &= PAGE_MASK; > + > + down_write(¤t->mm->mmap_sem); > ret = do_mlock(start, len, 0); > up_write(¤t->mm->mmap_sem); > + > return ret; > } > > @@ -761,12 +765,12 @@ SYSCALL_DEFINE1(mlockall, int, flags) > if (flags & MCL_CURRENT) > lru_add_drain_all(); /* flush pagevec */ > > - down_write(¤t->mm->mmap_sem); > - > lock_limit = rlimit(RLIMIT_MEMLOCK); > lock_limit >>= PAGE_SHIFT; > > ret = -ENOMEM; > + down_write(¤t->mm->mmap_sem); > + > if (!(flags & MCL_CURRENT) || (current->mm->total_vm <= lock_limit) || > capable(CAP_IPC_LOCK)) > ret = do_mlockall(flags); > -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/