linux-kernel - Re: [PATCH v2 0/8] powerpc/85xx: p2020: Create one unified machine description

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]

Date:   Mon, 13 Feb 2023 21:11:45 +0100
From:   Pali Rohár <pali@...nel.org>
To:     Christophe Leroy <christophe.leroy@...roup.eu>
Cc:     Michael Ellerman <mpe@...erman.id.au>,
        Nicholas Piggin <npiggin@...il.com>,
        Scott Wood <oss@...error.net>, Sinan Akman <sinan@...teme.com>,
        Martin Kennedy <hurricos@...il.com>,
        "linuxppc-dev@...ts.ozlabs.org" <linuxppc-dev@...ts.ozlabs.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v2 0/8] powerpc/85xx: p2020: Create one unified machine
 description

On Monday 13 February 2023 19:58:15 Christophe Leroy wrote:
> Le 09/02/2023 à 01:15, Pali Rohár a écrit :
> >>
> >> This patch moves all p2020 boards from mpc85xx_rdb.c and mpc85xx_ds.c
> >> files into new p2020.c file, and plus it copies all helper functions
> >> which p2020 boards requires. This patch does not introduce any new code
> >> or functional change. It should be really plain copy/move.
> 
> Yes after looking into it in more details, it is exactly that. You 
> copied all helper functions but this is not said in the commit message.
> I think it should be said, and more important it should be explained why.
> Because this is exactly what I was not understanding, why I couldn't see 
> all moved functions: just because many of them were not moved but copied.
> 
> In the two first pages you made some function static, and then you 
> duplicated it. Why ? Why not keep it global and just use it from one 
> place to the other ?
> 
> Because after patch 3 we have:
> 
> arch/powerpc/platforms/85xx/mpc85xx_rdb.c:static void __init 
> mpc85xx_rdb_pic_init(void)
> arch/powerpc/platforms/85xx/p2020.c:static void __init 
> mpc85xx_rdb_pic_init(void)
> 
> arch/powerpc/platforms/85xx/mpc85xx_ds.c:static void __init 
> mpc85xx_ds_pic_init(void)
> arch/powerpc/platforms/85xx/p2020.c:static void __init 
> mpc85xx_ds_pic_init(void)
> 
> Why not just drop patches 1 and 2 and keep those two functions and all 
> the other common functions like mpc85xx_8259_cascade() 
> mpc85xx_ds_uli_init() and a lot more  in a separate common file ?
> 
> Christophe

After applying all patches there is no mpc85xx_rdb_pic_init() /
mpc85xx_ds_pic_init() function in p2020.c file. There is
p2020_pic_init() in p2020.c but it is slightly different than previous
two functions.

Maybe it could be possible to create one function mpc85xx_pic_init() as
unification of previous 3 functions, but such change would be needed to
test on lot of mpc85xx boards, which would be affected by such change.
And I do not have them for testing. I have only P2020.

So I wrote *_pic_init() function which is p2020 specific, like already
existing ds and rdb specific functions in their own source files.