Re: [PATCH 15/19] mkfs: don't treat files as though they are block devic

To:	xfs-oss <xfs@xxxxxxxxxxx>
Subject:	Re: [PATCH 15/19] mkfs: don't treat files as though they are block devices
From:	Jan Tulak <jtulak@xxxxxxxxxx>
Date:	Thu, 21 Apr 2016 14:43:21 +0200
Delivered-to:	xfs@xxxxxxxxxxx
In-reply-to:	<1461231593-31294-16-git-send-email-jtulak@xxxxxxxxxx>
References:	<1461231593-31294-1-git-send-email-jtulak@xxxxxxxxxx> <1461231593-31294-16-git-send-email-jtulak@xxxxxxxxxx>

On Thu, Apr 21, 2016 at 11:39 AM, Jan Tulak <jtulak@xxxxxxxxxx> wrote:

From: Dave Chinner <dchinner@xxxxxxxxxx>

THIS PATCH HAS KNOWN ISSUES - it fails xfs/206 and xfs/216 tests, as it
shrinks a file instead just not using it entirely, when -d size is used.

âSo the shrinking is happening here:

3127 Â Â Â Â/*

3128 Â Â Â Â * If the data area is a file, then grow it out to its final size

3129 Â Â Â Â * so that the reads for the end of the device in the mount code

3130 Â Â Â Â * will succeed.

3131 Â Â Â Â */

3132 Â Â Â Âif (xi.disfile && ftruncate64(xi.dfd, dblocks * blocksize) < 0) {â

Before the patch, xi.disfile was 0 and so it didn't shrink the file to the size of the new FS.

Now, what is the correct solve to this? Tests are written for the old behaviour, but this shrinking seems to be an intentional thing.

It seems that the FS works ok even when this truncating is not applied, so I think that I should remove this chunk (or change it to xi.dcreate=1 only), and keep the old behaviour.

What do you think about it, guys?

Cheers,

Jan

Jan Tulak

jtulak@xxxxxxxxxxÂ/ jan@xxxxxxxx

<Prev in Thread]	Current Thread	[Next in Thread>
[PATCH 04/19] mkfs: validate all input values, (continued) [PATCH 04/19] mkfs: validate all input values, Jan Tulak [PATCH 08/19] mkfs: getbool is redundant, Jan Tulak [PATCH 09/19] mkfs: use getnum_checked for all ranged parameters, Jan Tulak [PATCH 10/19] mkfs: add respecification detection to generic parsing, Jan Tulak [PATCH 07/19] mkfs: structify input parameter passing, Jan Tulak [PATCH 11/19] mkfs: table based parsing for converted parameters, Jan Tulak [PATCH 12/19] mkfs: merge getnum, Jan Tulak [PATCH 14/19] mkfs: add string options to generic parsing, Jan Tulak [PATCH 17/19] mkfs: unit conversions are case insensitive, Jan Tulak [PATCH 15/19] mkfs: don't treat files as though they are block devices, Jan Tulak Re: [PATCH 15/19] mkfs: don't treat files as though they are block devices, Jan Tulak <= Re: [PATCH 15/19] mkfs: don't treat files as though they are block devices, Eric Sandeen Re: [PATCH 15/19] mkfs: don't treat files as though they are block devices, Jan Tulak [PATCH 15/19 v2] mkfs: don't treat files as though they are block devices, Jan Tulak [PATCH 15/19 v3] mkfs: don't treat files as though they are block devices, Jan Tulak Re: [PATCH 15/19 v3] mkfs: don't treat files as though they are block devices, Eric Sandeen [PATCH 16/19] mkfs: move spinodes crc check, Jan Tulak [PATCH 13/19] mkfs: encode conflicts into parsing table, Jan Tulak [PATCH 18/19] mkfs: add optional 'reason' for illegal_option, Jan Tulak [PATCH 19/19] mkfs: conflicting values with disabled crc should fail, Jan Tulak [RFC PATCH] xfstests: Add mkfs input validation tests, Jan Tulak

<Prev in Thread]

Current Thread

[Next in Thread>

Previous by Date:	Re: [XFS] Any process to a particular XFS device hung in D state forever., Brian Foster
Next by Date:	[PATCH] xfsdump: fix race condition between lseek() and read()/write(), Eryu Guan
Previous by Thread:	[PATCH 15/19] mkfs: don't treat files as though they are block devices, Jan Tulak
Next by Thread:	Re: [PATCH 15/19] mkfs: don't treat files as though they are block devices, Eric Sandeen
Indexes:	[Date] [Thread] [Top] [All Lists]

Previous by Date:

Re: [XFS] Any process to a particular XFS device hung in D state forever., Brian Foster

Next by Date:

[PATCH] xfsdump: fix race condition between lseek() and read()/write(), Eryu Guan

Previous by Thread:

[PATCH 15/19] mkfs: don't treat files as though they are block devices, Jan Tulak

Next by Thread:

Re: [PATCH 15/19] mkfs: don't treat files as though they are block devices, Eric Sandeen

Indexes:

[Date] [Thread] [Top] [All Lists]