xfs
[Top] [All Lists]

Re: [PATCH 06/11] reflink: more tests

To: "Darrick J. Wong" <darrick.wong@xxxxxxxxxx>
Subject: Re: [PATCH 06/11] reflink: more tests
From: Dave Chinner <david@xxxxxxxxxxxxx>
Date: Mon, 21 Dec 2015 17:31:11 +1100
Cc: linux-ext4@xxxxxxxxxxxxxxx, linux-btrfs@xxxxxxxxxxxxxxx, fstests@xxxxxxxxxxxxxxx, xfs@xxxxxxxxxxx
Delivered-to: xfs@xxxxxxxxxxx
In-reply-to: <20151219091131.15320.12391.stgit@xxxxxxxxxxxxxxxx>
References: <20151219091052.15320.94973.stgit@xxxxxxxxxxxxxxxx> <20151219091131.15320.12391.stgit@xxxxxxxxxxxxxxxx>
User-agent: Mutt/1.5.21 (2010-09-15)
On Sat, Dec 19, 2015 at 01:11:31AM -0800, Darrick J. Wong wrote:
> Add more tests for unaligned copy-on-write things, and explicitly
> test the ability to pass "len == 0" to mean reflink/dedupe all
> the way to the end of the file".
> 
> Signed-off-by: Darrick J. Wong <darrick.wong@xxxxxxxxxx>
.....
> +_cleanup()
> +{
> +    cd /
> +    rm -rf "$tmp".* "$TESTDIR"
> +}
.....
> +
> +TESTDIR="$TEST_DIR/test-$seq"
> +rm -rf "$TESTDIR"
> +mkdir "$TESTDIR"

This use of TESTDIR is highly confusing. When I see TESTDIR I think
that it's the mount point of the test device, not potentially some
directory on the scratch device.

I see that this occurs through lots of tests, not just these new
ones. Can you do a pass across the tests with, say, sed and rename
all these to something less confusing? e.g. "testdir", in lower
case, makes it clear that it's a local variable, not the global test
device mount point.... (i.e. test local variables are lower case,
exported test harnes variables are upper case ;)

Separate patch/pull req is fine.

Cheers,

dave.
-- 
Dave Chinner
david@xxxxxxxxxxxxx

<Prev in Thread] Current Thread [Next in Thread>