Re: [RFC] Consolidate vmlinux.lds.S files

[Sam Ravnborg <sam _at_ xxxxxxxxxxxx>]

On Mon, Jan 13, 2003 at 12:10:11PM -0700, Eric W. Biederman wrote:
> 
> I consider it a bigger bug that CodingStyle is not followed.  If there
> was an appropriate amount of whitespace linker scripts would be easier
> to read.
I noticed that as well, arm being one of the good guys here.
At least I tried to be consistent with coding style.

> Beyond that linker scripts native support include directives.
But they do not know of standard include paths, and you will see
linker scripts pulling in .h files as well. See ia64 as of today
as a good example.

> And included
> files can have additional SECTIONS attributes.  In particular:
> 
> SECTIONS {
>         .section1 {
> 		*(.section1)                       
>         }
> }
> SECTIONS {
> 	.section2 {
> 		*(.section2)
>         }
> }
> 
> is legal.
> 
> So I would recommend having an arch specific linker script that included
> the architecture independent parts.  And have each part have it's own
> SECTIONS attribute tag.
> 
> Something like:
> 
> ---- arch/i386/vmlinux.lds.S ----
> OUTPUT_FORMAT("elf32-i386", "elf32-i386", "elf32-i386")
> OUTPUT_ARCH(i386)
> ENTRY(_start)
> vmlinux.lds.S
> jiffies = jiffies_64;
> SECTIONS
> {
> 	. = 0xC0000000 + 0x100000;
> 	/* read-only */
> 	_text = .;                    /* Text and read-only data */
> 	.text : {
>         	*(.text)
>         	*(.fixup)
>         	*(.gnu.warning)
>         } = 0x9090
> 
> 	_etext = .;                   /* End of text section */
> 	.rodata : { *(.rodata) *(.rodata.*) }
> 	.kstrtab : { *(.kstrtab) }
> }
> INCLUDE ../../kernel/exceptions.lds
> INCLUDE ../../init/initramfs.lds
> 
> ---- kernel/exceptions.lds ----
> SECTIONS {
> 	. = ALIGN(16);                /* Exception table */
> 	__start___ex_table = .;
> 	__ex_table : { *(__ex_table) }
> 	__stop___ex_table = .;
> }
> 
> ---- init/initramfs.lds ----
> SECTIONS {
> 	. = ALIGN(4096);
> 	__initramfs_start = .;
> 	.init.ramfs : { *(.init.ramfs) }
> 	__initramfs_end = .;
> }
Different architectures have different alingments for ramfs for example.
I do not know why, but wanted to support that.

Good point with more sections btw.

	Sam
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo _at_ vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/