Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Mon, 20 May 2019 07:23:31 -0700
From:   Christoph Hellwig <hch@infradead.org>
To:     Alan Stern <stern@rowland.harvard.edu>
Cc:     Christoph Hellwig <hch@infradead.org>,
        Oliver Neukum <oneukum@suse.com>,
        Jaewon Kim <jaewon31.kim@gmail.com>, linux-mm@kvack.org,
        gregkh@linuxfoundation.org, Jaewon Kim <jaewon31.kim@samsung.com>,
        m.szyprowski@samsung.com, ytk.lee@samsung.com,
        linux-kernel@vger.kernel.org, linux-usb@vger.kernel.org
Subject: Re: [RFC PATCH] usb: host: xhci: allow __GFP_FS in dma allocation
Message-ID: <20190520142331.GA12108@infradead.org>
References: <20190520101206.GA9291@infradead.org>
 <Pine.LNX.4.44L0.1905201011490.1498-100000@iolanthe.rowland.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <Pine.LNX.4.44L0.1905201011490.1498-100000@iolanthe.rowland.org>
User-Agent: Mutt/1.9.2 (2017-12-15)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Mon, May 20, 2019 at 10:16:57AM -0400, Alan Stern wrote:
> What if the allocation requires the kernel to swap some old pages out 
> to the backing store, but the backing store is on the device that the 
> driver is managing?  The swap can't take place until the current I/O 
> operation is complete (assuming the driver can handle only one I/O 
> operation at a time), and the current operation can't complete until 
> the old pages are swapped out.  Result: deadlock.
> 
> Isn't that the whole reason for using GFP_NOIO in the first place?

It is, or rather was.  As it has been incredibly painful to wire
up the gfp_t argument through some callstacks, most notably the
vmalloc allocator which is used by a lot of the DMA allocators on
non-coherent platforms, we now have the memalloc_noio_save and
memalloc_nofs_save functions that mark a thread as not beeing to
go into I/O / FS reclaim.  So even if you use GFP_KERNEL you will
not dip into reclaim with those flags set on the thread.