linux-kernel - Re: [PATCH] mm/gup_benchmark: GUP_BENCHMARK depends on DEBUG

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: <8d2a373c-d88f-9db2-766d-96bd54e69a40@nvidia.com>
Date:   Sat, 7 Nov 2020 23:53:07 -0800
From:   John Hubbard <jhubbard@...dia.com>
To:     "Song Bao Hua (Barry Song)" <song.bao.hua@...ilicon.com>,
        Randy Dunlap <rdunlap@...radead.org>,
        "akpm@...ux-foundation.org" <akpm@...ux-foundation.org>,
        "linux-mm@...ck.org" <linux-mm@...ck.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
CC:     Linuxarm <linuxarm@...wei.com>,
        Ralph Campbell <rcampbell@...dia.com>,
        John Garry <john.garry@...wei.com>
Subject: Re: [PATCH] mm/gup_benchmark: GUP_BENCHMARK depends on DEBUG_FS

On 11/7/20 11:35 PM, Song Bao Hua (Barry Song) wrote:
> Or do you want this ?
> 
> (Code B)
> diff --git a/mm/Kconfig b/mm/Kconfig
> index 01b0ae0cd9d3..a7ff0d31afd5 100644
> --- a/mm/Kconfig
> +++ b/mm/Kconfig
> @@ -836,6 +836,7 @@ config PERCPU_STATS
> 
>   config GUP_TEST
>          bool "Enable infrastructure for get_user_pages()-related unit tests"
> +       depends on DEBUG_FS
>          help
>            Provides /sys/kernel/debug/gup_test, which in turn provides a way
>            to make ioctl calls that can launch kernel-based unit tests for
> @@ -853,6 +854,9 @@ config GUP_TEST
> 
>            See tools/testing/selftests/vm/gup_test.c
> 
> +comment "GUP_TEST needs to have DEBUG_FS enabled"
> +       depends on !GUP_TEST && !DEBUG_FS
> +
>   config GUP_GET_PTE_LOW_HIGH
>          bool


The above, please. It's your original patch, plus a way to display something
if the "depends on" is not met.


> 
> To be honest, I am not a big fan of both of code A and B. I think "depends on" has
> clearly said everything the redundant comment wants to say.

It's not really just a "comment", in the sense of commenting the Kconfig sources.
It actually has an effect, which is that it displays something in "make menuconfig".
So it's not redundant, because it provides that output *instead* of hiding the
option entirely, when !DEBUG_FS.

Try it out, it's nice.


> 
>    │ Symbol: GUP_TEST [=]
>    │ Type  : bool
>    │ Defined at mm/Kconfig:837
>    │   Prompt: Enable infrastructure for get_user_pages()-related unit tests
>    │   Depends on: DEBUG_FS [=n]
>    │   Location:
>    │ (1) -> Memory Management option
> 
> Menuconfig shows GUP_TEST depends on DEBUG_FS and right now DEBUG_FS is
> "n". so it is impossible to enable GUP_TEST.
> 
> "comment" is a good thing, but it is more likely to be used for a menu or a group
> of configurations to extend a bundle of things.
> 
> On the other hand, If this particular case needs this comment, so do countless
> other configurations in hundreds of Kconfig files.

Well, maybe, yes.

I personally find it quite difficult, having options appear and disappear on me,
in this system. If they all had this "comment" behavior by default, to show up
as a placeholder, I think it would be a better user experience.


thanks,
-- 
John Hubbard
NVIDIA