[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20201123144914.GA667@lst.de>
Date: Mon, 23 Nov 2020 15:49:14 +0100
From: Christoph Hellwig <hch@....de>
To: 廖威雄 <gmpy.liaowx@...il.com>
Cc: Christoph Hellwig <hch@....de>, Kees Cook <keescook@...omium.org>,
Anton Vorontsov <anton@...msg.org>,
Colin Cross <ccross@...roid.com>,
Tony Luck <tony.luck@...el.com>,
WeiXiong Liao <liaoweixiong@...winnertech.com>,
linux-mtd@...ts.infradead.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 8/9] pstore/blk: use the normal block device I/O path
On Sun, Nov 08, 2020 at 10:43:29PM +0800, 廖威雄 wrote:
> > + const char devname[] = "/dev/pstore-blk";
> > + dev_t dev = name_to_dev_t(blkdev);
> > +
> > + if (!dev)
> > + return;
> > + init_unlink(devname);
> > + init_mknod(devname, S_IFBLK | 0600, new_encode_dev(dev));
> > + __pstore_blk_init(devname);
>
> How about pstore/blk making /dev/pstore-blk node no matter build-in
> or module. Then pstore/blk always opens /dev/pstore-blk. By this way,
> blkdev can also be a device number if pstore/blk built as a module.
Please read the comment above blkdev_get_by_dev on why no one should
add new interfaces based on the device number.
Powered by blists - more mailing lists