Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Thu, 25 Jun 2020 11:48:32 -0700
From:   "Darrick J. Wong" <darrick.wong@oracle.com>
To:     "Matthew Wilcox (Oracle)" <willy@infradead.org>
Cc:     linux-kernel@vger.kernel.org, linux-mm@kvack.org,
        linux-xfs@vger.kernel.org, dm-devel@redhat.com,
        Mikulas Patocka <mpatocka@redhat.com>,
        Jens Axboe <axboe@kernel.dk>, NeilBrown <neilb@suse.de>,
        Yafang Shao <laoar.shao@gmail.com>
Subject: Re: [PATCH 0/6] Overhaul memalloc_no*
Message-ID: <20200625184832.GP7606@magnolia>
References: <20200625113122.7540-1-willy@infradead.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20200625113122.7540-1-willy@infradead.org>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Thu, Jun 25, 2020 at 12:31:16PM +0100, Matthew Wilcox (Oracle) wrote:
> I want a memalloc_nowait like we have memalloc_noio and memalloc_nofs
> for an upcoming patch series, and Jens also wants it for non-blocking
> io_uring.  It turns out we already have dm-bufio which could benefit
> from memalloc_nowait, so it may as well go into the tree now.
> 
> The biggest problem is that we're basically out of PF_ flags, so we need
> to find somewhere else to store the PF_MEMALLOC_NOWAIT flag.  It turns
> out the PF_ flags are really supposed to be used for flags which are
> accessed from other tasks, and the MEMALLOC flags are only going to
> be used by this task.  So shuffling everything around frees up some PF
> flags and generally makes the world a better place.

So, uh, how does this intersect with the patch "xfs: reintroduce
PF_FSTRANS for transaction reservation recursion protection" that
re-adds PF_TRANS because uh I guess we lost some subtlety or another at
some point?

I don't have any strong opinion on this series one way or another, but
seeing as that patch was generating discussion about how PF_MEMALLOC_NO*
is not quite the same as PF_TRANS, I kinda want all these competing PF
tweaks and reworks to come together into a single series to review,
rather than the four(?) different people submitting conflicting changes.

[adding Yafang Shao to cc]

--D

> Patch series also available from
> http://git.infradead.org/users/willy/linux.git/shortlog/refs/heads/memalloc
> 
> Matthew Wilcox (Oracle) (6):
>   mm: Replace PF_MEMALLOC_NOIO with memalloc_noio
>   mm: Add become_kswapd and restore_kswapd
>   xfs: Convert to memalloc_nofs_save
>   mm: Replace PF_MEMALLOC_NOFS with memalloc_nofs
>   mm: Replace PF_MEMALLOC_NOIO with memalloc_nocma
>   mm: Add memalloc_nowait
> 
>  drivers/block/loop.c           |  3 +-
>  drivers/md/dm-bufio.c          | 30 ++++--------
>  drivers/md/dm-zoned-metadata.c |  5 +-
>  fs/iomap/buffered-io.c         |  2 +-
>  fs/xfs/kmem.c                  |  2 +-
>  fs/xfs/libxfs/xfs_btree.c      | 14 +++---
>  fs/xfs/xfs_aops.c              |  4 +-
>  fs/xfs/xfs_buf.c               |  2 +-
>  fs/xfs/xfs_linux.h             |  6 ---
>  fs/xfs/xfs_trans.c             | 14 +++---
>  fs/xfs/xfs_trans.h             |  2 +-
>  include/linux/sched.h          |  7 +--
>  include/linux/sched/mm.h       | 84 ++++++++++++++++++++++++++--------
>  kernel/sys.c                   |  8 ++--
>  mm/vmscan.c                    | 16 +------
>  15 files changed, 105 insertions(+), 94 deletions(-)
> 
> -- 
> 2.27.0
>