Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <bcef12c5-35b5-f6a8-24c8-d190f87c3cff@kernel.dk>
Date:   Fri, 20 May 2022 10:06:06 -0600
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux aarch64; rv:91.0) Gecko/20100101
 Thunderbird/91.8.1
Subject: Re: [PATCH v4 0/3] random: convert to using iters, for Al Viro
Content-Language: en-US
To:     "Jason A. Donenfeld" <Jason@zx2c4.com>
Cc:     Theodore Ts'o <tytso@mit.edu>, Christoph Hellwig <hch@lst.de>,
        LKML <linux-kernel@vger.kernel.org>,
        Al Viro <viro@zeniv.linux.org.uk>
References: <20220520094459.116240-1-Jason@zx2c4.com>
 <Yoey+FOYO69lS5qP@zx2c4.com> <bbdeeca7-9c99-e661-6596-f04ea783a30b@kernel.dk>
 <Yoe2PPA/pzqhf+GZ@zx2c4.com> <0a6ed6b9-0917-0d83-5c45-70ff58fad429@kernel.dk>
 <Yoe52RaMayvgodVW@zx2c4.com> <79024675-9a4a-cb2b-a0ed-91067ef05783@kernel.dk>
 <Yoe7ykNXjUerhywY@zx2c4.com>
From:   Jens Axboe <axboe@kernel.dk>
In-Reply-To: <Yoe7ykNXjUerhywY@zx2c4.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 5/20/22 10:03 AM, Jason A. Donenfeld wrote:
>> Then I'd say there are only two options:
>>
>> - Add a helper that provides splice for something that only has
>>   read/write set.
> 
> That'd be fine with me, but wouldn't it involve bringing back set_fs(),
> because of the copy_to_user() in there?

I haven't even looked into whether it's currently feasible or not, just
mentioning it as a potential option. But the better one is definetely
the next one...

>> - Just accept that we're 3% slower reading from /dev/urandom for now,
>>   and maybe 1-2% for small reads. Can't really imagine this being a huge
>>   issue, how many high throughput /dev/urandom read situations exist in
>>   the real world?
> 
> An option three might be that eventually the VFS overhead is worked out
> and read_iter() reaches parity. One can hope, I guess.

And that will certainly happen, especially as we have other paths that
don't really have the choice, they have to use the iterator versions.
And if we can get a bit closer, then that also opens the door more
generic conversions so we can kill ->read/->write for almost all cases
(except those weirdo ones that Al pointed out).

-- 
Jens Axboe