[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20150723095329.GA26420@infradead.org>
Date: Thu, 23 Jul 2015 02:53:29 -0700
From: Christoph Hellwig <hch@...radead.org>
To: Matias Bj??rling <mb@...htnvm.io>
Cc: hch@...radead.org, axboe@...com, linux-fsdevel@...r.kernel.org,
linux-kernel@...r.kernel.org, linux-nvme@...ts.infradead.org,
Stephen.Bates@...s.com, keith.busch@...el.com
Subject: Re: [PATCH v5 1/5] lightnvm: Support for Open-Channel SSDs
Hi Matias,
I like this new architecture.
Minor nitpicks below:
> @@ -0,0 +1,598 @@
> +/*
> + * core.c - Open-channel SSD integration core
No need to state the file name in the top of the file comments, it
doesn't add value and get stale easily.
> +static LIST_HEAD(_targets);
> +static LIST_HEAD(_bms);
> +static LIST_HEAD(_devices);
> +static DECLARE_RWSEM(_lock);
Please add a nvm_ prefix to these.
> +static int nvm_ioctl(struct block_device *bdev, fmode_t mode, unsigned int cmd,
> + unsigned long arg)
> +{
> + return 0;
> +}
> +
> +static int nvm_open(struct block_device *bdev, fmode_t mode)
> +{
> + return 0;
> +}
> +
> +static void nvm_release(struct gendisk *disk, fmode_t mode)
> +{
> +}
No need to implement these empty methods.
> +/* _lock must be taken */
Turn this into an lockdep_assert_held in the code, please.
> +int nvm_register(struct request_queue *q, struct gendisk *disk,
> + struct nvm_dev_ops *ops)
> +{
No need to pass a gendisk here, in fact LighNVM low level driver
shouldn't even need to allocate one. The only thing you seem to be
using here is the name anyway.
> +void nvm_unregister(struct gendisk *disk)
Same here.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists