public inbox for u-boot@lists.denx.de
 help / color / mirror / Atom feed
From: Albert ARIBAUD <albert.aribaud@free.fr>
To: u-boot@lists.denx.de
Subject: [U-Boot] ARM: Incorrect ROM protection range?
Date: Thu, 24 Feb 2011 14:13:13 +0100	[thread overview]
Message-ID: <4D665969.1@free.fr> (raw)
In-Reply-To: <AANLkTi=Zr+u_=JNQrBNyk8VW1QstsZ7LLA0ORce_8wGZ@mail.gmail.com>

Le 24/02/2011 13:58, Po-Yu Chuang a ?crit :
> Hi Albert,
>
> On Thu, Feb 24, 2011 at 7:58 PM, Albert ARIBAUD<albert.aribaud@free.fr>  wrote:
>> Le 24/02/2011 10:39, Wolfgang Denk a ?crit :
>>>
>>> Dear Albert ARIBAUD,
>>>
>>> In message<4D6611A7.5050802@free.fr>    you wrote:
>>>>
>>>>>>> In the U-Boot image itself, knowing the image size could be achieved
>>>>>>> in
>>>>>>> ARM by using a general _end symbol that would be set after the last
>>>>>>> image output section, so _end-_start would equal the image size.
>>>>>>
>>>>>> we have such a "_end" in u-boot.lds files.
>>>>
>>>> I *knew* this name did not pop up in my mind without a reason. :)
>>>>
>>>> Apologies for not having checked.
>>>>
>>>>> I guess we need a __dynsym_end in all u-boot.lds files.
>>>>
>>>> I'd rather go for "_end", which does not tie the solution to dynsym
>>>> being the last section in the image -- imagine for some reason we move
>>>> dynsym inside the image rather than at the end, end the image size will
>>>> be wrong again. With _end, and a suitable comment in the LD file... this
>>>> will be averted where-ever dynsym ends up.
>>>>
>>>> After all, the bug you uncovered was due to using the wrong symbol, a
>>>> BSS related one rather than an image-related one, in the first place; so
>>>> let's try and learn from past mistakes.
>>>
>>> I think the location of _end in the linker scripts needs to be changed
>>> - so far it covers the maximum of dynsym and bss, but it should refer
>>> to the end of dynsym only.
>>
>> Po-Yu Chuang, can you verify Wolfgang's suggestion (and make sure no other
>> place depends on _end) and submit a patch? As this is a fix, a quick patch
>> could still be pulled in for the upcoming release.
>
> I think there is a problem in Wolfgang's suggestion.
>
> Those start.S files assume that BSS section is from __bss_start to _end.
> If we change _end to the end of .dynsym, then we will not clear BSS correctly.
>
> I agree with your comment about _end is a better choice than __dynsym_end,
> but if we changed _end, maybe we need __bss_end? Either way, we need to
> modify all the u-boot.lds and/or start.S.
>
> I still don't know what is the best solution.

Thanks for pointing this out: indeed, _end is actually used as BSS's 
end, not as the image's end -- I did not find any other use of it.

However, not only ARM uses _end: how do we want to handle the change 
overall?

> best regards,
> Po-Yu Chuang

Amicalement,
-- 
Albert.

  reply	other threads:[~2011-02-24 13:13 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-02-24  6:06 [U-Boot] ARM: Incorrect ROM protection range? Po-Yu Chuang
2011-02-24  6:52 ` Albert ARIBAUD
2011-02-24  7:07   ` Wolfgang Denk
2011-02-24  7:11   ` Po-Yu Chuang
2011-02-24  7:33   ` Heiko Schocher
2011-02-24  7:40     ` Po-Yu Chuang
2011-02-24  8:07       ` Albert ARIBAUD
2011-02-24  9:39         ` Wolfgang Denk
2011-02-24 11:58           ` Albert ARIBAUD
2011-02-24 12:58             ` Po-Yu Chuang
2011-02-24 13:13               ` Albert ARIBAUD [this message]
2011-02-24 13:41                 ` Wolfgang Denk
2011-02-24 16:01                   ` Albert ARIBAUD
2011-02-24 18:38                     ` Wolfgang Denk
2011-02-24 18:47                       ` Albert ARIBAUD
2011-02-24 20:17                         ` Wolfgang Denk
2011-02-25  2:41                           ` Po-Yu Chuang
2011-02-25  6:34                             ` Albert ARIBAUD
2011-02-25  7:15                               ` Wolfgang Denk
2011-02-25  8:00                                 ` Po-Yu Chuang
2011-03-01  8:31                                   ` Po-Yu Chuang
2011-03-01 12:22                                     ` Albert ARIBAUD
2011-03-02  2:56                                       ` Po-Yu Chuang

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=4D665969.1@free.fr \
    --to=albert.aribaud@free.fr \
    --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