Received-SPF: pass (google.com: domain of linux-kernel+bounces-187957-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:40f1:3f00::1 as permitted sender) client-ip=2604:1380:40f1:3f00::1;
Date: Thu, 23 May 2024 13:08:02 -0700
From: Andrew Morton <akpm@linux-foundation.org>
To: Liu Wei <liuwei09@cestc.cn>
Cc: willy@infradead.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] mm/filemap: invalidating pages is still necessary when
 io with IOCB_NOWAIT
Message-Id: <20240523130802.730d2790b8e5f691871575c0@linux-foundation.org>
In-Reply-To: <20240513132339.26269-1-liuwei09@cestc.cn>
References: <20240513132339.26269-1-liuwei09@cestc.cn>
Precedence: bulk
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit

On Mon, 13 May 2024 21:23:39 +0800 Liu Wei <liuwei09@cestc.cn> wrote:

> After commit (6be96d3ad3 fs: return if direct I/O will trigger writeback),
> when we issuing AIO with direct I/O and IOCB_NOWAIT on a block device, the
> process context will not be blocked.
> 
> However, if the device already has page cache in memory, EAGAIN will be
> returned. And even when trying to reissue the AIO with direct I/O and
> IOCB_NOWAIT again, we consistently receive EAGAIN.
> 
> Maybe a better way to deal with it: filemap_fdatawrite_range dirty pages
> with WB_SYNC_NONE flag, and invalidate_mapping_pages unmapped pages at
> the same time.

Can't userspace do this?  If EAGAIN, sync the fd and retry the IO?