From mboxrd@z Thu Jan 1 00:00:00 1970 From: Vlastimil Babka Subject: Re: [PATCH 12/18] aio: make aio_setup_ring killable Date: Fri, 11 Mar 2016 12:57:42 +0100 Message-ID: <56E2B2B6.4040307@suse.cz> References: <1456752417-9626-1-git-send-email-mhocko@kernel.org> <1456752417-9626-13-git-send-email-mhocko@kernel.org> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-2; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from mx2.suse.de ([195.135.220.15]:35083 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751731AbcCKL5p (ORCPT ); Fri, 11 Mar 2016 06:57:45 -0500 In-Reply-To: <1456752417-9626-13-git-send-email-mhocko@kernel.org> Sender: linux-arch-owner@vger.kernel.org List-ID: To: Michal Hocko , LKML Cc: Andrew Morton , linux-mm@kvack.org, Alex Deucher , Alex Thorlton , Andrea Arcangeli , Andy Lutomirski , Benjamin LaHaise , =?UTF-8?Q?Christian_K=c3=b6nig?= , Daniel Vetter , Dave Hansen , David Airlie , Davidlohr Bueso , David Rientjes , "H . Peter Anvin" , Hugh Dickins , Ingo Molnar , Johannes Weiner , "Kirill A . Shutemov" , Konstantin Khlebnikov , linux-arch@vger.kernel.org, Mel Gorman , Oleg Nesterov , Peter Zijlstra Pe On 02/29/2016 02:26 PM, Michal Hocko wrote: > From: Michal Hocko > > aio_setup_ring waits for mmap_sem in writable mode. If the waiting > task gets killed by the oom killer it would block oom_reaper from > asynchronous address space reclaim and reduce the chances of timely > OOM resolving. Wait for the lock in the killable mode and return with > EINTR if the task got killed while waiting. This will also expedite > the return to the userspace and do_exit. > > Cc: Benjamin LaHaise > Cc: Alexander Viro > Signed-off-by: Michal Hocko The error handling could be changed to labels, but: Acked-by: Vlastimil Babka > --- > fs/aio.c | 7 ++++++- > 1 file changed, 6 insertions(+), 1 deletion(-) > > diff --git a/fs/aio.c b/fs/aio.c > index 56bcdf4105f4..1c2e7e2c1b2b 100644 > --- a/fs/aio.c > +++ b/fs/aio.c > @@ -520,7 +520,12 @@ static int aio_setup_ring(struct kioctx *ctx) > ctx->mmap_size = nr_pages * PAGE_SIZE; > pr_debug("attempting mmap of %lu bytes\n", ctx->mmap_size); > > - down_write(&mm->mmap_sem); > + if (down_write_killable(&mm->mmap_sem)) { > + ctx->mmap_size = 0; > + aio_free_ring(ctx); > + return -EINTR; > + } > + > ctx->mmap_base = do_mmap_pgoff(ctx->aio_ring_file, 0, ctx->mmap_size, > PROT_READ | PROT_WRITE, > MAP_SHARED, 0, &unused); > From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mx2.suse.de ([195.135.220.15]:35083 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751731AbcCKL5p (ORCPT ); Fri, 11 Mar 2016 06:57:45 -0500 Subject: Re: [PATCH 12/18] aio: make aio_setup_ring killable References: <1456752417-9626-1-git-send-email-mhocko@kernel.org> <1456752417-9626-13-git-send-email-mhocko@kernel.org> From: Vlastimil Babka Message-ID: <56E2B2B6.4040307@suse.cz> Date: Fri, 11 Mar 2016 12:57:42 +0100 MIME-Version: 1.0 In-Reply-To: <1456752417-9626-13-git-send-email-mhocko@kernel.org> Content-Type: text/plain; charset=iso-8859-2; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-arch-owner@vger.kernel.org List-ID: To: Michal Hocko , LKML Cc: Andrew Morton , linux-mm@kvack.org, Alex Deucher , Alex Thorlton , Andrea Arcangeli , Andy Lutomirski , Benjamin LaHaise , =?UTF-8?Q?Christian_K=c3=b6nig?= , Daniel Vetter , Dave Hansen , David Airlie , Davidlohr Bueso , David Rientjes , "H . Peter Anvin" , Hugh Dickins , Ingo Molnar , Johannes Weiner , "Kirill A . Shutemov" , Konstantin Khlebnikov , linux-arch@vger.kernel.org, Mel Gorman , Oleg Nesterov , Peter Zijlstra , Petr Cermak , Thomas Gleixner , Michal Hocko , Alexander Viro Message-ID: <20160311115742.Ic6IUBL7_SHEZ6nS9mtgUE23GM3-PutWkxC9JtkvsRM@z> On 02/29/2016 02:26 PM, Michal Hocko wrote: > From: Michal Hocko > > aio_setup_ring waits for mmap_sem in writable mode. If the waiting > task gets killed by the oom killer it would block oom_reaper from > asynchronous address space reclaim and reduce the chances of timely > OOM resolving. Wait for the lock in the killable mode and return with > EINTR if the task got killed while waiting. This will also expedite > the return to the userspace and do_exit. > > Cc: Benjamin LaHaise > Cc: Alexander Viro > Signed-off-by: Michal Hocko The error handling could be changed to labels, but: Acked-by: Vlastimil Babka > --- > fs/aio.c | 7 ++++++- > 1 file changed, 6 insertions(+), 1 deletion(-) > > diff --git a/fs/aio.c b/fs/aio.c > index 56bcdf4105f4..1c2e7e2c1b2b 100644 > --- a/fs/aio.c > +++ b/fs/aio.c > @@ -520,7 +520,12 @@ static int aio_setup_ring(struct kioctx *ctx) > ctx->mmap_size = nr_pages * PAGE_SIZE; > pr_debug("attempting mmap of %lu bytes\n", ctx->mmap_size); > > - down_write(&mm->mmap_sem); > + if (down_write_killable(&mm->mmap_sem)) { > + ctx->mmap_size = 0; > + aio_free_ring(ctx); > + return -EINTR; > + } > + > ctx->mmap_base = do_mmap_pgoff(ctx->aio_ring_file, 0, ctx->mmap_size, > PROT_READ | PROT_WRITE, > MAP_SHARED, 0, &unused); >