linux-kernel - Re: current binutils trunk fails to build bootable kernel image for some configurations

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] [day] [month] [year] [list]

Message-ID: <20090723060122.GI13233@bubble.grove.modra.org>
Date:	Thu, 23 Jul 2009 15:31:22 +0930
From:	Alan Modra <amodra@...pond.net.au>
To:	Sam Ravnborg <sam@...nborg.org>
Cc:	Matthias Klose <doko@...ntu.com>,
	binutils <binutils@...rceware.org>,
	Bastian Blank <waldi@...ian.org>,
	Linus Torvalds <torvalds@...ux-foundation.org>,
	linux-kernel@...r.kernel.org, Kiko Piris <kernel@...ispons.net>,
	Damien Wyart <damien.wyart@...e.fr>, Greg KH <gregkh@...e.de>,
	Wolfgang Walter <wolfgang.walter@...m.de>
Subject: Re: current binutils trunk fails to build bootable kernel image
	for some configurations

On Thu, Jul 23, 2009 at 07:05:22AM +0200, Sam Ravnborg wrote:
> On Thu, Jul 23, 2009 at 10:19:39AM +0930, Alan Modra wrote:
> >   .data_nosave ALIGN(PAGE_SIZE) : AT(ADDR(.data_nosave) - LOAD_OFFSET) {
> 
> Do you recall when this started to be supported?

Forever.

> I think I tried this and it failed - but I may be wrong here.

I can think of at least two things that might cause my suggestion to
fail.
1) .data_nosave input sections may have alignments greater than
PAGE_SIZE, but we're setting the start address explicity.
2) following sections won't be aligned to PAGE_SIZE if .data_nosave is
empty.

Hmm, there is another option that I'd forgotten about.

  .data_nosave : AT(ADDR(.data_nosave) - LOAD_OFFSET) ALIGN(PAGE_SIZE) {

This avoids (1) above since it only increases alignment.  Support for
this form of alignment was added 2005-09.

> Also - do there exist an option to tell what sections has not been
> covered by a linker script?

No.  You could put

  /DISCARD/ : { *(*) }

at the end of your script, then generate a map file.  Discarded
sections are reported.

-- 
Alan Modra
Australia Development Lab, IBM
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/