linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Andy Lutomirski <luto@kernel.org>
To: Jens Axboe <axboe@kernel.dk>
Cc: Andy Lutomirski <luto@kernel.org>, Christoph Hellwig <hch@lst.de>,
	Linux FS Devel <linux-fsdevel@vger.kernel.org>,
	linux-aio@kvack.org, linux-block@vger.kernel.org,
	Jeff Moyer <jmoyer@redhat.com>, Avi Kivity <avi@scylladb.com>,
	Linux API <linux-api@vger.kernel.org>,
	linux-man <linux-man@vger.kernel.org>
Subject: Re: [PATCH 05/18] Add io_uring IO interface
Date: Wed, 30 Jan 2019 21:11:46 -0800	[thread overview]
Message-ID: <CALCETrVwHcWcJB4r=14=uB2NHGasW+rRRWVoun6-UY_t3+AU2A@mail.gmail.com> (raw)
In-Reply-To: <dab3fa32-3887-3264-9fe1-bc3b7abc58b3@kernel.dk>

>>> On Jan 28, 2019, at 5:20 PM, Jens Axboe <axboe@kernel.dk> wrote:
>>>
>>> On 1/28/19 5:47 PM, Andy Lutomirski wrote:
>>> On Mon, Jan 28, 2019 at 6:57 AM Christoph Hellwig <hch@lst.de> wrote:
>>>
>>> [please make sure linux-api and linux-man are CCed on new syscalls
>>> so that we get API experts to review them]
>>
>>>> +static int io_import_iovec(struct io_ring_ctx *ctx, int rw,
>>>> +                        const struct io_uring_sqe *sqe,
>>>> +                        struct iovec **iovec, struct iov_iter *iter)
>>>> +{
>>>> +     void __user *buf = u64_to_user_ptr(sqe->addr);
>>>> +
>>>> +#ifdef CONFIG_COMPAT
>>>> +     if (ctx->compat)
>>>> +             return compat_import_iovec(rw, buf, sqe->len, UIO_FASTIOV,
>>>> +                                             iovec, iter);
>>>> +#endif
>>>
>>> I think we can just check in_compat_syscall() here, which means we
>>> can kill the ->compat member, and the separate compat version of the
>>> setup syscall.
>>
>> Since this whole API is new, I don't suppose you could introduce a
>> struct iovec64 or similar and just make the ABI be identical for
>> 64-bit and 32-bit code?
>
> Sure, that would be straight forward. Is there a strong reason to do
> so outside of "that would be nice"? It's not like it's a huge amount
> of code.

Here are some minor-ish benefits:

 - It avoids having a code path that is only used with 32 bit code on
64 bit kernels and is therefore rarely tested.  (In this particular
case, the code path doesn't diverge much, but for most compat
syscalls, it's almost an entirely separate implementation of the main
syscall code.)

 - It makes life easier for tools like strace.

 - It minimizes the chance of making a giant mess on x32, which isn't
really native or compat.

--Andy

  parent reply	other threads:[~2019-01-31  5:12 UTC|newest]

Thread overview: 64+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-01-23 15:35 [PATCHSET v7] io_uring IO interface Jens Axboe
2019-01-23 15:35 ` [PATCH 01/18] fs: add an iopoll method to struct file_operations Jens Axboe
2019-01-28 14:25   ` Christoph Hellwig
2019-01-28 16:13     ` Jens Axboe
2019-01-23 15:35 ` [PATCH 02/18] block: wire up block device iopoll method Jens Axboe
2019-01-23 15:35 ` [PATCH 03/18] block: add bio_set_polled() helper Jens Axboe
2019-01-28 14:26   ` Christoph Hellwig
2019-01-23 15:35 ` [PATCH 04/18] iomap: wire up the iopoll method Jens Axboe
2019-01-23 15:35 ` [PATCH 05/18] Add io_uring IO interface Jens Axboe
2019-01-28 14:57   ` Christoph Hellwig
2019-01-28 16:26     ` Jens Axboe
2019-01-28 16:34       ` Christoph Hellwig
2019-01-28 19:32         ` Jens Axboe
2019-01-28 18:25     ` Jens Axboe
2019-01-29  6:30       ` Christoph Hellwig
2019-01-29 11:58         ` Arnd Bergmann
2019-01-29 15:20           ` Jens Axboe
2019-01-29 16:18             ` Arnd Bergmann
2019-01-29 16:19               ` Jens Axboe
2019-01-29 16:26                 ` Arnd Bergmann
2019-01-29 16:28                   ` Jens Axboe
2019-01-29 16:46                     ` Arnd Bergmann
2019-01-29  0:47     ` Andy Lutomirski
2019-01-29  1:20       ` Jens Axboe
2019-01-29  6:45         ` Christoph Hellwig
2019-01-29 12:05           ` Arnd Bergmann
2019-01-31  5:11         ` Andy Lutomirski [this message]
2019-01-31 16:37           ` Jens Axboe
2019-01-23 15:35 ` [PATCH 06/18] io_uring: add fsync support Jens Axboe
2019-01-23 15:35 ` [PATCH 07/13] io_uring: add support for pre-mapped user IO buffers Jens Axboe
2019-01-23 15:35 ` [PATCH 07/18] io_uring: support for IO polling Jens Axboe
2019-01-28 15:02   ` Christoph Hellwig
2019-01-28 16:46     ` Jens Axboe
2019-01-29  6:27       ` Christoph Hellwig
2019-01-29 13:20         ` Jens Axboe
2019-01-23 15:35 ` [PATCH 08/18] fs: add fget_many() and fput_many() Jens Axboe
2019-01-28 14:29   ` Christoph Hellwig
2019-01-28 16:48     ` Jens Axboe
2019-01-23 15:35 ` [PATCH 08/13] io_uring: add file set registration Jens Axboe
2019-01-23 15:35 ` [PATCH 09/13] io_uring: add submission polling Jens Axboe
2019-01-28 15:09   ` Christoph Hellwig
2019-01-28 17:05     ` Jens Axboe
2019-01-29  6:29       ` Christoph Hellwig
2019-01-29 13:21         ` Jens Axboe
2019-01-28 21:13   ` Jeff Moyer
2019-01-28 21:28     ` Jens Axboe
2019-01-23 15:35 ` [PATCH 09/18] io_uring: use fget/fput_many() for file references Jens Axboe
2019-01-23 15:35 ` [PATCH 10/13] io_uring: add io_kiocb ref count Jens Axboe
2019-01-23 15:35 ` [PATCH 10/18] io_uring: batch io_kiocb allocation Jens Axboe
2019-01-23 15:35 ` [PATCH 11/18] block: implement bio helper to add iter bvec pages to bio Jens Axboe
2019-01-28 14:31   ` Christoph Hellwig
2019-01-28 16:54     ` Jens Axboe
2019-01-23 15:35 ` [PATCH 11/13] io_uring: add support for IORING_OP_POLL Jens Axboe
2019-01-23 15:35 ` [PATCH 12/18] io_uring: add support for pre-mapped user IO buffers Jens Axboe
2019-01-23 15:35 ` [PATCH 12/13] io_uring: allow workqueue item to handle multiple buffered requests Jens Axboe
2019-01-23 15:35 ` [PATCH 13/18] io_uring: add file set registration Jens Axboe
2019-01-23 15:35 ` [PATCH 13/13] io_uring: add io_uring_event cache hit information Jens Axboe
2019-01-23 15:35 ` [PATCH 14/18] io_uring: add submission polling Jens Axboe
2019-01-23 15:35 ` [PATCH 15/18] io_uring: add io_kiocb ref count Jens Axboe
2019-01-23 15:35 ` [PATCH 16/18] io_uring: add support for IORING_OP_POLL Jens Axboe
2019-01-23 15:35 ` [PATCH 17/18] io_uring: allow workqueue item to handle multiple buffered requests Jens Axboe
2019-01-23 15:35 ` [PATCH 18/18] io_uring: add io_uring_event cache hit information Jens Axboe
     [not found] <20190128213538.13486-1-axboe@kernel.dk>
     [not found] ` <20190128213538.13486-6-axboe@kernel.dk>
     [not found]   ` <CAG48ez0vDqDH4ks7q4L3F+xt-4kVQrN1yw34QwFAmwQyy27FTw@mail.gmail.com>
     [not found]     ` <e9326a77-54c5-e2b8-d9e5-663261462597@kernel.dk>
     [not found]       ` <CAG48ez17NW0GJVRC6dFcHZTgQifFz5og1XCUbXkHKhr6f=j74Q@mail.gmail.com>
     [not found]         ` <05cb18f7a97a6151c305cdb7240c4abc995aed59.camel@fb.com>
     [not found]           ` <CAG48ez09iOOnPz83b6HxktYHTfouS2GD6i3PfQEjp8WCp+3-VA@mail.gmail.com>
2019-02-01 17:23             ` [PATCH 05/18] Add io_uring IO interface Jann Horn
2019-02-01 18:05               ` Al Viro

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='CALCETrVwHcWcJB4r=14=uB2NHGasW+rRRWVoun6-UY_t3+AU2A@mail.gmail.com' \
    --to=luto@kernel.org \
    --cc=avi@scylladb.com \
    --cc=axboe@kernel.dk \
    --cc=hch@lst.de \
    --cc=jmoyer@redhat.com \
    --cc=linux-aio@kvack.org \
    --cc=linux-api@vger.kernel.org \
    --cc=linux-block@vger.kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-man@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body. 
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).