linux-kernel - Re: [PATCH] selftests: futex: fix run

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: Windows password security audit tool. GUI, reports in PDF.

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

Message-ID: <95d21b7c-be98-a8f5-fe67-ec4877d4a946@osg.samsung.com>
Date:   Wed, 2 Aug 2017 19:06:39 -0600
From:   Shuah Khan <shuahkh@....samsung.com>
To:     Darren Hart <dvhart@...radead.org>
Cc:     shuah@...nel.org, tglx@...utronix.de, mingo@...hat.com,
        peterz@...radead.org, linux-kernel@...r.kernel.org,
        linux-kselftest@...r.kernel.org, Shuah Khan <shuah@...nel.org>,
        Shuah Khan <shuahkh@....samsung.com>
Subject: Re: [PATCH] selftests: futex: fix run_tests target

On 08/02/2017 06:52 PM, Darren Hart wrote:
> On Wed, Aug 02, 2017 at 04:31:42PM -0600, Shuah Khan wrote:
>> make -C tools/testing/selftests/futex/ run_tests doesn't run the tests.
>> Running run_tests target only when $(OUTPUT) is the $(PWD) is incorrect.
>> When $(OUTPUT) is empty, run_tests will not run. Fix it.
>>
>> Fixes: a8ba798bc8ec ("selftests: enable O and KBUILD_OUTPUT")
>> Signed-off-by: Shuah Khan <shuahkh@....samsung.com>
> 
> So this indeed addresses the problem you reported.
> 
> I'm curious about why this was changed like it was. I presume to avoid
> some issues with relative paths, like the ./run.sh and the "cd
> functional" within run.sh.... but I haven't been able to trigger that.> 
> When you say $(OUTPUT) is empty - when does that occur? Because if it
> was in fact "" then the change below would attempt to execute /run.sh
> ... which is surely not what we want.

This will not happen. At this stage, OUTPUT will not be empty.

I didn't explain the condition correctly in my change log. When OUTPUT
isn't set (which is the usual case when test is run using:

make -C tools/testing/selftests/futex run_tests

lib.mk sets it to OUTPUT := $(shell pwd)

So OUTPUT will be the current directory when run_tests are run.

> 
>> ---
>>  tools/testing/selftests/futex/Makefile | 2 +-
>>  1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/tools/testing/selftests/futex/Makefile b/tools/testing/selftests/futex/Makefile
>> index e2fbb890aef9..7c647f619d63 100644
>> --- a/tools/testing/selftests/futex/Makefile
>> +++ b/tools/testing/selftests/futex/Makefile
>> @@ -14,7 +14,7 @@ all:
>>  	done
>>  
>>  override define RUN_TESTS
>> -	@if [ `dirname $(OUTPUT)` = $(PWD) ]; then ./run.sh; fi

The above change was made in conjunction with the
a8ba798bc8ec ("selftests: enable O and KBUILD_OUTPUT")

The above change didn't anticipate failures when tests are run outside
the selftests/Makefile umbrella via make -C tools/testing/selftests/testdir

>> +	$(OUTPUT)/run.sh
>>  endef
>>  
>>  override define INSTALL_RULE
>> -- 
>> 2.11.0
>>
>>
> 

thanks,
-- Shuah