From: Eric Sandeen <sandeen@sandeen.net>
To: xfs@oss.sgi.com
Subject: Re: [PATCH 06/19] mkfs: validate logarithmic parameters sanely
Date: Wed, 6 Apr 2016 21:52:54 -0500 [thread overview]
Message-ID: <5705CB86.8090701@sandeen.net> (raw)
In-Reply-To: <1458818136-56043-7-git-send-email-jtulak@redhat.com>
On 3/24/16 6:15 AM, jtulak@redhat.com wrote:
> From: Dave Chinner <dchinner@redhat.com>
>
> Testing logarithmic paramters like "-n log=<num>" shows that we do a
> terrible job of validating such input. e.g.:
<<<<< >>>>>>>
This lost pat of Dave's original commit log, which said:
# mkfs.xfs -f -n log=456858480 /dev/vda
> .....
> naming =version 2 bsize=65536 ascii-ci=0 ftype=0
> ....
>
> Yeah, I just asked for a block size of 2^456858480, and it didn't
> get rejected. Great, isn't it?
>
> So, factor out the parsing of logarithmic parameters, and pass in
> the maximum valid value that they can take. These maximum values
> might not be completely accurate (e.g. block/sector sizes will
> affect the eventual valid maximum) but we can get rid of all the
> overflows and stupidities before we get to fine-grained validity
> checking later in mkfs once things like block and sector sizes have
> been finalised.
>
> Signed-off-by: Dave Chinner <dchinner@redhat.com>
> Signed-off-by: Jan Tulak <jtulak@redhat.com>
> Reviewed-by: Brian Foster <bfoster@redhat.com>
otherwise,
Reviewed-by: Eric Sandeen <sandeen@redhat.com>
> ---
> mkfs/xfs_mkfs.c | 79 +++++++++++++++++++++++++++++++++++++--------------------
> 1 file changed, 51 insertions(+), 28 deletions(-)
>
> diff --git a/mkfs/xfs_mkfs.c b/mkfs/xfs_mkfs.c
> index 9394bd3..dda076d 100644
> --- a/mkfs/xfs_mkfs.c
> +++ b/mkfs/xfs_mkfs.c
> @@ -1022,6 +1022,27 @@ getbool(
> return c ? true : false;
> }
>
> +static int
> +getnum_checked(
> + const char *str,
> + long long min_val,
> + long long max_val,
> + const char *illegal_str,
> + char reqval_char,
> + char *reqval_opts[],
> + int reqval_optind)
> +{
> + long long c;
> +
> + if (!str || *str == '\0')
> + reqval(reqval_char, reqval_opts, reqval_optind);
> +
> + c = getnum(str, 0, 0, false);
> + if (c < min_val || c > max_val)
> + illegal(str, illegal_str);
> + return c;
> +}
> +
> int
> main(
> int argc,
> @@ -1179,16 +1200,16 @@ main(
>
> switch (getsubopt(&p, (constpp)bopts, &value)) {
> case B_LOG:
> - if (!value || *value == '\0')
> - reqval('b', bopts, B_LOG);
> if (blflag)
> respec('b', bopts, B_LOG);
> if (bsflag)
> conflict('b', bopts, B_SIZE,
> B_LOG);
> - blocklog = getnum(value, 0, 0, false);
> - if (blocklog <= 0)
> - illegal(value, "b log");
> + blocklog = getnum_checked(value,
> + XFS_MIN_BLOCKSIZE_LOG,
> + XFS_MAX_BLOCKSIZE_LOG,
> + "b log", 'b', bopts,
> + B_LOG);
> blocksize = 1 << blocklog;
> blflag = 1;
> break;
> @@ -1325,16 +1346,16 @@ main(
> nodsflag = 1;
> break;
> case D_SECTLOG:
> - if (!value || *value == '\0')
> - reqval('d', dopts, D_SECTLOG);
> if (slflag)
> respec('d', dopts, D_SECTLOG);
> if (ssflag)
> conflict('d', dopts, D_SECTSIZE,
> D_SECTLOG);
> - sectorlog = getnum(value, 0, 0, false);
> - if (sectorlog <= 0)
> - illegal(value, "d sectlog");
> + sectorlog = getnum_checked(value,
> + XFS_MIN_SECTORSIZE_LOG,
> + XFS_MAX_SECTORSIZE_LOG,
> + "d sectlog", 'd', dopts,
> + D_SECTLOG);
> sectorsize = 1 << sectorlog;
> slflag = 1;
> break;
> @@ -1399,9 +1420,11 @@ main(
> if (isflag)
> conflict('i', iopts, I_SIZE,
> I_LOG);
> - inodelog = getnum(value, 0, 0, false);
> - if (inodelog <= 0)
> - illegal(value, "i log");
> + inodelog = getnum_checked(value,
> + XFS_DINODE_MIN_LOG,
> + XFS_DINODE_MAX_LOG,
> + "i log", 'i', iopts,
> + I_LOG);
> isize = 1 << inodelog;
> ilflag = 1;
> break;
> @@ -1573,16 +1596,16 @@ main(
> lsflag = 1;
> break;
> case L_SECTLOG:
> - if (!value || *value == '\0')
> - reqval('l', lopts, L_SECTLOG);
> if (lslflag)
> respec('l', lopts, L_SECTLOG);
> if (lssflag)
> conflict('l', lopts, L_SECTSIZE,
> L_SECTLOG);
> - lsectorlog = getnum(value, 0, 0, false);
> - if (lsectorlog <= 0)
> - illegal(value, "l sectlog");
> + lsectorlog = getnum_checked(value,
> + XFS_MIN_SECTORSIZE_LOG,
> + XFS_MAX_SECTORSIZE_LOG,
> + "l sectlog", 'l', lopts,
> + L_SECTLOG);
> lsectorsize = 1 << lsectorlog;
> lslflag = 1;
> break;
> @@ -1658,16 +1681,16 @@ _("cannot specify both -m crc=1 and -n ftype\n"));
>
> switch (getsubopt(&p, (constpp)nopts, &value)) {
> case N_LOG:
> - if (!value || *value == '\0')
> - reqval('n', nopts, N_LOG);
> if (nlflag)
> respec('n', nopts, N_LOG);
> if (nsflag)
> conflict('n', nopts, N_SIZE,
> N_LOG);
> - dirblocklog = getnum(value, 0, 0, false);
> - if (dirblocklog <= 0)
> - illegal(value, "n log");
> + dirblocklog = getnum_checked(value,
> + XFS_MIN_REC_DIRSIZE,
> + XFS_MAX_BLOCKSIZE_LOG,
> + "n log", 'n', nopts,
> + N_LOG);
> dirblocksize = 1 << dirblocklog;
> nlflag = 1;
> break;
> @@ -1786,16 +1809,16 @@ _("cannot specify both -m crc=1 and -n ftype\n"));
> switch (getsubopt(&p, (constpp)sopts, &value)) {
> case S_LOG:
> case S_SECTLOG:
> - if (!value || *value == '\0')
> - reqval('s', sopts, S_SECTLOG);
> if (slflag || lslflag)
> respec('s', sopts, S_SECTLOG);
> if (ssflag || lssflag)
> conflict('s', sopts, S_SECTSIZE,
> S_SECTLOG);
> - sectorlog = getnum(value, 0, 0, false);
> - if (sectorlog <= 0)
> - illegal(value, "s sectlog");
> + sectorlog = getnum_checked(value,
> + XFS_MIN_SECTORSIZE_LOG,
> + XFS_MAX_SECTORSIZE_LOG,
> + "s sectlog", 's', sopts,
> + S_SECTLOG);
> lsectorlog = sectorlog;
> sectorsize = 1 << sectorlog;
> lsectorsize = sectorsize;
>
_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs
next prev parent reply other threads:[~2016-04-07 2:53 UTC|newest]
Thread overview: 84+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-03-24 11:15 [PATCH 00/19] mkfs cleaning jtulak
2016-03-24 11:15 ` [PATCH 01/19] xfsprogs: use common code for multi-disk detection jtulak
2016-03-31 20:25 ` Eric Sandeen
2016-04-06 9:05 ` Jan Tulak
2016-03-24 11:15 ` [PATCH 02/19] mkfs: sanitise ftype parameter values jtulak
2016-03-24 16:33 ` Eric Sandeen
2016-03-29 16:11 ` Jan Tulak
2016-03-29 16:17 ` Eric Sandeen
2016-03-29 16:20 ` Jan Tulak
2016-03-29 17:14 ` Jan Tulak
2016-03-24 11:15 ` [PATCH 03/19] mkfs: Sanitise the superblock feature macros jtulak
2016-04-01 2:05 ` Eric Sandeen
2016-04-06 9:12 ` Jan Tulak
2016-04-06 21:01 ` Dave Chinner
2016-04-07 11:53 ` Jan Tulak
2016-04-07 0:12 ` Eric Sandeen
2016-04-07 1:43 ` Eric Sandeen
2016-04-07 13:09 ` Jan Tulak
2016-04-07 13:18 ` Eric Sandeen
2016-04-07 13:27 ` Jan Tulak
2016-03-24 11:15 ` [PATCH 04/19] mkfs: validate all input values jtulak
2016-04-06 23:02 ` Eric Sandeen
2016-04-07 11:15 ` Jan Tulak
2016-03-24 11:15 ` [PATCH 05/19] mkfs: factor boolean option parsing jtulak
2016-04-07 2:48 ` Eric Sandeen
2016-03-24 11:15 ` [PATCH 06/19] mkfs: validate logarithmic parameters sanely jtulak
2016-04-07 2:52 ` Eric Sandeen [this message]
2016-03-24 11:15 ` [PATCH 07/19] mkfs: structify input parameter passing jtulak
2016-04-07 3:14 ` Eric Sandeen
2016-04-07 11:43 ` Jan Tulak
2016-03-24 11:15 ` [PATCH 08/19] mkfs: getbool is redundant jtulak
2016-04-07 17:25 ` Eric Sandeen
2016-04-08 10:30 ` Jan Tulak
2016-04-08 17:41 ` Eric Sandeen
2016-03-24 11:15 ` [PATCH 09/19] mkfs: use getnum_checked for all ranged parameters jtulak
2016-04-07 19:02 ` Eric Sandeen
2016-04-08 10:47 ` Jan Tulak
2016-04-08 15:52 ` Eric Sandeen
2016-03-24 11:15 ` [PATCH 10/19] mkfs: add respecification detection to generic parsing jtulak
2016-04-07 19:06 ` Eric Sandeen
2016-03-24 11:15 ` [PATCH 11/19] mkfs: table based parsing for converted parameters jtulak
2016-04-07 19:08 ` Eric Sandeen
2016-03-24 11:15 ` [PATCH 12/19] mkfs: merge getnum jtulak
2016-04-07 19:14 ` Eric Sandeen
2016-03-24 11:15 ` [PATCH 13/19] mkfs: encode conflicts into parsing table jtulak
2016-04-07 22:40 ` Eric Sandeen
2016-03-24 11:15 ` [PATCH 14/19] mkfs: add string options to generic parsing jtulak
2016-04-07 22:49 ` Eric Sandeen
2016-03-24 11:15 ` [PATCH 15/19] mkfs: don't treat files as though they are block devices jtulak
2016-04-08 0:25 ` Eric Sandeen
2016-04-08 0:32 ` Eric Sandeen
2016-04-08 14:58 ` Jan Tulak
2016-04-08 15:50 ` Eric Sandeen
2016-04-08 15:56 ` Jan Tulak
2016-04-09 4:12 ` Eric Sandeen
2016-04-13 15:43 ` Jan Tulak
2016-04-14 9:49 ` Jan Tulak
2016-04-20 9:51 ` Jan Tulak
2016-04-20 13:17 ` Jan Tulak
2016-04-20 16:53 ` Eric Sandeen
2016-04-21 9:22 ` Jan Tulak
2016-03-24 11:15 ` [PATCH 16/19] mkfs: move spinodes crc check jtulak
2016-03-24 11:15 ` [PATCH 17/19] xfsprogs: disable truncating of files jtulak
2016-04-06 21:42 ` Eric Sandeen
2016-04-07 9:41 ` Jan Tulak
2016-04-08 0:09 ` Dave Chinner
2016-04-08 10:06 ` Jan Tulak
2016-04-08 23:08 ` Dave Chinner
2016-04-13 15:08 ` Jan Tulak
2016-04-13 16:17 ` Eric Sandeen
2016-04-13 16:23 ` Jan Tulak
2016-04-13 16:25 ` Eric Sandeen
2016-04-13 21:37 ` Dave Chinner
2016-04-14 12:31 ` Jan Tulak
2016-03-24 11:15 ` [PATCH 18/19] mkfs: unit conversions are case insensitive jtulak
2016-04-06 21:10 ` Eric Sandeen
2016-04-07 10:50 ` Jan Tulak
2016-04-08 0:41 ` Eric Sandeen
2016-04-08 1:03 ` Dave Chinner
2016-04-08 9:08 ` Jan Tulak
2016-04-08 15:51 ` Eric Sandeen
2016-03-24 11:15 ` [PATCH 19/19] mkfs: add optional 'reason' for illegal_option jtulak
2016-04-06 22:23 ` Eric Sandeen
-- strict thread matches above, loose matches on Subject: below --
2016-04-21 9:39 [PATCH 00/19 v2] mkfs cleaning Jan Tulak
2016-04-21 9:39 ` [PATCH 06/19] mkfs: validate logarithmic parameters sanely Jan Tulak
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=5705CB86.8090701@sandeen.net \
--to=sandeen@sandeen.net \
--cc=xfs@oss.sgi.com \
/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