netdev - Re: [RFC PATCH net-next 05/10] selftests: Introduce Makefile variable to list shared bash scripts

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <20231228203848.ypx5iomfk4u36wyc@skbuf>
Date: Thu, 28 Dec 2023 22:38:48 +0200
From: Vladimir Oltean <vladimir.oltean@....com>
To: Benjamin Poirier <bpoirier@...dia.com>
Cc: netdev@...r.kernel.org, Shuah Khan <shuah@...nel.org>,
	Petr Machata <petrm@...dia.com>, Hangbin Liu <liuhangbin@...il.com>
Subject: Re: [RFC PATCH net-next 05/10] selftests: Introduce Makefile
 variable to list shared bash scripts

On Thu, Dec 28, 2023 at 02:28:06PM -0500, Benjamin Poirier wrote:
> I agree with your point about consistency. I have changed TEST_INCLUDES
> to take paths relative to PWD. The implementation is more complicated
> since the paths have to be converted to the old values anyways for
> `rsync -R` but it works. I pasted the overall diff below and it'll be
> part of the next version of the series.
> 
> > 
> > To solve the inconsistency, can it be used like this everywhere?
> > 
> > TEST_INCLUDES := \
> > 	$(SRC_PATH)/net/lib.sh
> 
> After the changes, it's possible to list files using SRC_PATH but I
> didn't do it. Since the point is to make TEST_INCLUDES be more like
> TEST_PROGS, TEST_FILES, ..., I used relative paths.
> For example in net/forwarding/Makefile:
> TEST_INCLUDES := \
> 	../lib.sh

I thought you wanted to avoid the cascade of ../../../ which is confusing,
so I recommended $(SRC_PATH) as a way to avoid that by using absolute
paths instead - which makes it easier to track down the include file
being sourced.

My inconsistency issue was with TEST_INCLUDES being relative to a
different directory than the rest. With that being addressed,
I don't think that using absolute paths for some files would be
inconsistent in a similar way. But I don't mind too much either way.
Feel free to keep the ../../../ scheme.