Linux-Next Archive on lore.kernel.org
 help / color / Atom feed
From: "Darrick J. Wong" <darrick.wong@oracle.com>
To: Jan Stancek <jstancek@redhat.com>
Cc: Naresh Kamboju <naresh.kamboju@linaro.org>,
	LTP List <ltp@lists.linux.it>,
	Linux-Next Mailing List <linux-next@vger.kernel.org>,
	linux-fsdevel@vger.kernel.org, chrubis <chrubis@suse.cz>,
	open list <linux-kernel@vger.kernel.org>,
	Al Viro <viro@zeniv.linux.org.uk>,
	Mark Brown <broonie@kernel.org>, Arnd Bergmann <arnd@arndb.de>,
	lkft-triage@lists.linaro.org,
	Christoph Hellwig <hch@infradead.org>,
	linux-ext4 <linux-ext4@vger.kernel.org>,
	"Theodore Ts'o" <tytso@mit.edu>
Subject: Re: LTP: diotest4.c:476: read to read-only space. returns 0: Success
Date: Sun, 10 Nov 2019 17:26:14 -0800
Message-ID: <20191111012614.GC6235@magnolia> (raw)
In-Reply-To: <852514139.11036267.1573172443439.JavaMail.zimbra@redhat.com>

[add the other iomap maintainer to cc]
[add the ext4 list to cc since they just added iomap directio support]
[add the ext4 maintainer for the same reason]

On Thu, Nov 07, 2019 at 07:20:43PM -0500, Jan Stancek wrote:
> 
> 
> ----- Original Message -----
> > LTP test case dio04 test failed on 32bit kernel running linux next
> > 20191107 kernel.
> > Linux version 5.4.0-rc6-next-20191107.
> > 
> > diotest4    1  TPASS  :  Negative Offset
> > diotest4    2  TPASS  :  removed
> > diotest4    3  TPASS  :  Odd count of read and write
> > diotest4    4  TPASS  :  Read beyond the file size
> > diotest4    5  TPASS  :  Invalid file descriptor
> > diotest4    6  TPASS  :  Out of range file descriptor
> > diotest4    7  TPASS  :  Closed file descriptor
> > diotest4    8  TPASS  :  removed
> > diotest4    9  TCONF  :  diotest4.c:345: Direct I/O on /dev/null is
> > not supported
> > diotest4   10  TPASS  :  read, write to a mmaped file
> > diotest4   11  TPASS  :  read, write to an unmapped file
> > diotest4   12  TPASS  :  read from file not open for reading
> > diotest4   13  TPASS  :  write to file not open for writing
> > diotest4   14  TPASS  :  read, write with non-aligned buffer
> > diotest4   15  TFAIL  :  diotest4.c:476: read to read-only space.
> > returns 0: Success
> > diotest4   16  TFAIL  :  diotest4.c:180: read, write buffer in read-only
> > space
> > diotest4   17  TFAIL  :  diotest4.c:114: read allows  nonexistant
> > space. returns 0: Success
> > diotest4   18  TFAIL  :  diotest4.c:129: write allows  nonexistant
> > space.returns -1: Invalid argument
> > diotest4   19  TFAIL  :  diotest4.c:180: read, write in non-existant space
> > diotest4   20  TPASS  :  read, write for file with O_SYNC
> > diotest4    0  TINFO  :  2/15 test blocks failed
> 
> Smaller reproducer for 32-bit system and ext4 is:
>   openat(AT_FDCWD, "testdata-4.5918", O_RDWR|O_DIRECT) = 4
>   mmap2(NULL, 4096, PROT_READ, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0xb7f7b000
>   read(4, 0xb7f7b000, 4096)              = 0 // expects -EFAULT
> 
> Problem appears to be conversion in ternary operator at
> iomap_dio_bio_actor() return. Test passes for me with
> following tweak:

I can't do a whole lot with a code snippet that lacks a proper SOB
header.

> diff --git a/fs/iomap/direct-io.c b/fs/iomap/direct-io.c
> index 2f88d64c2a4d..8615b1f78389 100644
> --- a/fs/iomap/direct-io.c
> +++ b/fs/iomap/direct-io.c
> @@ -318,7 +318,7 @@ iomap_dio_bio_actor(struct inode *inode, loff_t pos, loff_t length,
>                 if (pad)
>                         iomap_dio_zero(dio, iomap, pos, fs_block_size - pad);
>         }
> -       return copied ? copied : ret;
> +       return copied ? (loff_t) copied : ret;

I'm a little confused on this proposed fix -- why does casting size_t
(aka unsigned long) to loff_t (long long) on a 32-bit system change the
test outcome?  Does this same diotest4 failure happen with XFS?  I ask
because XFS has been using iomap for directio for ages.

AFAICT @copied is a simple byte counter, and the other variable @n that
gets added to @copied is also a simple byte counter -- nobody should
ever be trying to stuff a negative value in there?

(Or is this a bug with the ext4 code that calls iomap_dio_rw?)

--D

>  }
> 
>  static loff_t
> 

  reply index

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-11-07 13:59 Naresh Kamboju
2019-11-08  0:20 ` Jan Stancek
2019-11-11  1:26   ` Darrick J. Wong [this message]
2019-11-11  8:19     ` Jan Stancek
2019-11-11  8:38       ` Christoph Hellwig
2019-11-11 10:28         ` [PATCH] iomap: fix return value of iomap_dio_bio_actor on 32bit systems Jan Stancek
2019-11-11 10:36           ` Christoph Hellwig
2019-11-12  1:24           ` Darrick J. Wong
2019-11-11 10:38         ` LTP: diotest4.c:476: read to read-only space. returns 0: Success Jan Stancek
2019-11-11 18:26           ` Naresh Kamboju

Reply instructions:

You may reply publically 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=20191111012614.GC6235@magnolia \
    --to=darrick.wong@oracle.com \
    --cc=arnd@arndb.de \
    --cc=broonie@kernel.org \
    --cc=chrubis@suse.cz \
    --cc=hch@infradead.org \
    --cc=jstancek@redhat.com \
    --cc=linux-ext4@vger.kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-next@vger.kernel.org \
    --cc=lkft-triage@lists.linaro.org \
    --cc=ltp@lists.linux.it \
    --cc=naresh.kamboju@linaro.org \
    --cc=tytso@mit.edu \
    --cc=viro@zeniv.linux.org.uk \
    /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

Linux-Next Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/linux-next/0 linux-next/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 linux-next linux-next/ https://lore.kernel.org/linux-next \
		linux-next@vger.kernel.org
	public-inbox-index linux-next

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.linux-next


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git