public inbox for u-boot@lists.denx.de
 help / color / mirror / Atom feed
From: Albert ARIBAUD <albert.u.boot@aribaud.net>
To: u-boot@lists.denx.de
Subject: [U-Boot] [PATCH 1/3] ARM: lib: Remove CONFIG_ARCH_CPU_INIT dependency
Date: Thu, 01 Mar 2012 22:44:09 +0100	[thread overview]
Message-ID: <4F4FEDA9.3080307@aribaud.net> (raw)
In-Reply-To: <CAOMZO5BpvjbY6g3m3CCO_JWEz=3dv3diApJyS8QkMRh8gX=uUw@mail.gmail.com>

Le 01/03/2012 21:57, Fabio Estevam a ?crit :
> On Thu, Mar 1, 2012 at 5:51 PM, Albert ARIBAUD
> <albert.u.boot@aribaud.net>  wrote:
>
>> This adds (some) (functionally dead) code to all boards that did not require
>> arch_cpu_init().
>
> Yes, it adds a "return 0" only. Is this terribly bad?

Well...

> Isn't it better than having to select CONFIG_ARCH_CPU_INIT for every
> board to select arch_cpu_init(), which is supposed to be CPU
> dependent?

It adds a "#define CONFIG_ARCH_CPU_INTI" only. Is this terribly bad ?

If what bothers you is that some board all repeat something that should 
be done only once, then it should bother you just as much that many 
boards repeat something that should not be done at all most of the time.

I understand that you see a repetition of CONFIG_ARCH_CPU_INIT that you 
suspect is highly redundant. But then, you should cure the repetition, 
not the existence of CONFIG_ARCH_CPU_INIT.

How about this approach: since the problem is the presence of CPU- (or 
Soc-)level elements in board-level config files, and since there are 
probably many other such items in board config files (CPU type in a SoC 
and HW IP configurations, for instance) then why not introduce some 
cpu-config and soc-config file which would group all those cpu- (resp 
soc-)level items, including any "#define CONFIG_ARCH_CPU_INIT", and 
replace these items in board config files with a single "#include 
"<cpu>-config.h" (or "<soc>-config.h")? This would significantly reduce 
board config file size.

One could even choose either of two courses depending on how many or how 
few boards require CONFIG_ARCH_CPU_INIT: if many boards require it, let 
the cpu- or soc-common config contain the define, and the few boards 
which don't will have to explicitly undefine it; if only a few boards 
need the define, then it won't be done in the cpu- or soc-common config 
but in each board that requires it.

And CPUs, SoCs and boards that don't need ARCH_CPU_INIT at all won't 
grow a single useless byte.

Amicalement,
-- 
Albert.

  reply	other threads:[~2012-03-01 21:44 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-03-01 14:02 [U-Boot] [PATCH 1/3] ARM: lib: Remove CONFIG_ARCH_CPU_INIT dependency Fabio Estevam
2012-03-01 14:02 ` [U-Boot] [PATCH 2/3] ARM: mx6: Remove CONFIG_ARCH_CPU_INIT Fabio Estevam
2012-03-01 18:31   ` Marek Vasut
2012-07-07 11:25     ` Albert ARIBAUD
2012-03-01 14:02 ` [U-Boot] [PATCH 3/3] ARM: mx28: " Fabio Estevam
2012-03-01 18:31   ` Marek Vasut
2012-07-07 11:31     ` Albert ARIBAUD
2012-07-19 17:28       ` Marek Vasut
2012-03-01 18:31 ` [U-Boot] [PATCH 1/3] ARM: lib: Remove CONFIG_ARCH_CPU_INIT dependency Marek Vasut
2012-03-01 20:51 ` Albert ARIBAUD
2012-03-01 20:57   ` Fabio Estevam
2012-03-01 21:44     ` Albert ARIBAUD [this message]
2012-03-05 17:56       ` Tom Rini
2012-03-06  7:26         ` Albert ARIBAUD
2012-03-03 10:01 ` Stefano Babic
2012-03-05 17:52 ` Tom Rini
2012-03-05 17:58   ` Fabio Estevam
2012-03-05 18:12     ` Fabio Estevam

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=4F4FEDA9.3080307@aribaud.net \
    --to=albert.u.boot@aribaud.net \
    --cc=u-boot@lists.denx.de \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox