linux-kernel - Re: [PATCH] m68knommu: Clean up linker script using new linker script macros.

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]

Date:	Tue, 8 Sep 2009 10:36:33 -0400 (EDT)
From:	Tim Abbott <tabbott@...lice.com>
To:	Greg Ungerer <gerg@...pgear.com>
cc:	Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
	Sam Ravnborg <sam@...nborg.org>,
	Greg Ungerer <gerg@...inux.org>
Subject: Re: [PATCH] m68knommu: Clean up linker script using new linker script
 macros.

On Tue, 8 Sep 2009, Greg Ungerer wrote:

> But now the bss section doesn't looked properly aligned (at
> least it is not on a page boundary any longer after the init.data):
> 
> Sections:
> Idx Name          Size      VMA       LMA       File off  Algn
>   0 .text         00124ab8  40020000  40020000  00002000  2**4
>                   CONTENTS, ALLOC, LOAD, READONLY, CODE
>   1 .data         00011540  40144ac0  40144ac0  00126ac0  2**4
>                   CONTENTS, ALLOC, LOAD, DATA
>   2 .init.text    0000a96c  40156000  40156000  00138000  2**2
>                   CONTENTS, ALLOC, LOAD, READONLY, CODE
>   3 .init.data    00000b7c  4016096c  4016096c  0014296c  2**2
>                   CONTENTS, ALLOC, LOAD, DATA
>   4 .bss          0000a9e8  401614f0  401614f0  001434e8  2**4
>                   ALLOC
>   5 .comment      00001c8c  00000000  00000000  001434e8  2**0
>                   CONTENTS, READONLY
> 
> Seems kind of odd because:
> 
> 40156000 A __init_begin
> 40162000 A __init_end
> 401614f0 B _sbss

Yes, that is quite odd.  

> The resulting kernels don't boot.

Right, presumably a big chunk of the .sbss code is being freed since _sbss 
is before __init_end.

I'm guessing the issue here is that the __init_begin and __init_end 
markers are now not actually in a section (and thus don't have a memory 
region marker attached).  Does the following patch cause _sbss to not 
overlap the init region?

	-Tim Abbott

diff --git a/arch/m68knommu/kernel/vmlinux.lds.S b/arch/m68knommu/kernel/vmlinux.lds.S
index c380040..81b54cb 100644
--- a/arch/m68knommu/kernel/vmlinux.lds.S
+++ b/arch/m68knommu/kernel/vmlinux.lds.S
@@ -154,12 +154,16 @@ SECTIONS {
 		_edata = . ;
 	} > DATA
 
-	. = ALIGN(PAGE_SIZE);
-	__init_begin = .;
+	.init_begin : {
+		. = ALIGN(PAGE_SIZE);
+		__init_begin = .;
+	} > INIT
 	INIT_TEXT_SECTION(PAGE_SIZE) > INIT
 	INIT_DATA_SECTION(16) > INIT
-	. = ALIGN(PAGE_SIZE);
-	__init_end = .;
+	.init_end : {
+		. = ALIGN(PAGE_SIZE);
+		__init_end = .;
+	} > INIT
 
 	/DISCARD/ : {
 		EXIT_TEXT
--
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/