public inbox for gcc-help@gcc.gnu.org
 help / color / mirror / Atom feed
From: Dukextra <mahuhta@gmail.com>
To: gcc-help@gcc.gnu.org
Subject: Re: "causes a section type conflict"
Date: Thu, 25 Mar 2010 11:40:00 -0000	[thread overview]
Message-ID: <28024536.post@talk.nabble.com> (raw)
In-Reply-To: <mcr7hpce388.fsf@dhcp-172-17-9-151.mtv.corp.google.com>



Ian Lance Taylor-3 wrote:
> 
> 
> Thanks.  Now for the rest of it.  Please give us a complete standalone
> test case.  Please show us the exact output of the linker.
> 
> 


Hi

I tried to do the linking so wrong way. I tried to put every C-code
implementations to same section with using __attribute__ command. And that
is wrong.

My orginal problem caused because I was trying to put variables and
functions to same section. I can continue compiling if I put variables to
different section than functions. And later I had to put also consts to
different sections. (My example doesnt show that problem because it doesnt
use attributed variable)

I think that better way to do this is to edit linker script to put defined
files to different memeory.

example:

	.sdata :
	{
		*(EXCLUDE_FILE(*my_object.o) .sdata) /* put every object's .sdata sections
here but not my_object.o*/
	} > ram
	
	.sdata_my_section : AT (LOADADDR(.sdata)+SIZEOF(.sdata))
	{
		*my_object.o(.sdata) /* put only .sdata sections from my_object.o file*/ 					
	} > my_own_ram_section


And do this for every section... data, sdata, bss, etc...

BUT

Now I have new problem. How I can make symbols in linker script?
I want to list many files in one symbol like this:


__MY_OBJS = my_file1.o my_file2.o my_file3.o

SECTION
{
	.sdata :
	{
		*(EXCLUDE_FILE(*__MY_OBJS) .sdata) 
	} > ram
	
	.sdata_hal : AT (LOADADDR(.sdata)+SIZEOF(.sdata))
	{
		*__MY_OBJS(.sdata) 					
	} > my_own_ram_section
}


Is there any way to do this?

Makefile style $(__MY_OBJS) could be better but that doesnt work too.

Dukex
-- 
View this message in context: http://old.nabble.com/%22causes-a-section-type-conflict%22-tp27847975p28024536.html
Sent from the gcc - Help mailing list archive at Nabble.com.

  reply	other threads:[~2010-03-25  7:00 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-03-10 10:36 Dukextra
2010-03-10 16:39 ` Ian Lance Taylor
2010-03-10 16:50   ` Axel Freyn
2010-03-11  5:13     ` Dukextra
2010-03-16 14:48       ` Ian Lance Taylor
2010-03-25 11:40         ` Dukextra [this message]
2010-03-25 15:34           ` Ian Lance Taylor

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=28024536.post@talk.nabble.com \
    --to=mahuhta@gmail.com \
    --cc=gcc-help@gcc.gnu.org \
    /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;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).