arm: add a common .lds link script

Most ARM CPUs use a very similar link script. This adds a basic script that can be used by most CPUs.

Two new symbols are introduced which are intended to eventually be defined on all architectures to make things easier for generic relocation and reduce special-case code for each architecture:

__text_start is the start of the text area (equivalent to the existing _start on ARM). It marks the start of the region which must be copied to a new location during relocation.

__image_copy_end is the end of the region which must be copied to a new location during relocation. It is normally equal to the start of the BSS region, but this can vary in some cases (SPL?). Making this an explicit symbol on its own removes any ambiguity and permits common code to always do the right thing.

This new script makes use of CPUDIR, now defined by both Makefile and spl/Makefile, to find the directory containing the start.o object file, which is always placed first in the image.

To permit MMU setup prior to relocation (as used by pxa) we add an area to the link script which contains space for this. This is taken from commit 7f4cfcf. CPUs can put the contents in there using their start.S file. BTW, shouldn't that area be 16KB-aligned?

Signed-off-by Simon Glass<[email protected]>

---
Changes in v3:
- Add more comments in the commit message
- Add section for MMU area, as required by pxa

  arch/arm/cpu/u-boot.lds |   91 +++++++++++++++++++++++++++++++++++++++++++++++
  1 files changed, 91 insertions(+), 0 deletions(-)

create mode 100644 arch/arm/cpu/u-boot.lds

arch/arm/cpu/u-boot.lds new file +46 -0

new file mode 100644
index 0000000..7a859fc
@@ -0,0 +1,91 @@
+/*
+ * Copyright (c) 2004-2008 Texas Instruments
+ *
+ * (C) Copyright 2002
+ * Gary Jennejohn, DENX Software Engineering,<[email protected]>
+ *
+ * See file CREDITS for list of people who contributed to this
+ * project.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 of
+ * the License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston,
+ * MA 02111-1307 USA
+ */
+
+OUTPUT_FORMAT("elf32-littlearm", "elf32-littlearm", "elf32-littlearm")
+OUTPUT_ARCH(arm)
+ENTRY(_start)
+SECTIONS
+{
+	. = 0x00000000;
+
+	. = ALIGN(4);
+	.text :
+	{
+		__text_start = .;
+		CPUDIR/start.o (.text)
+		*(.text)
+	}
+
+	. = ALIGN(4);
+	.rodata : { *(SORT_BY_ALIGNMENT(SORT_BY_NAME(.rodata*))) }
+
+	. = ALIGN(4);
+	.data : {
+		*(.data)

arch/arm/cpu/u-boot.lds new file +4 -0

+	__image_copy_end = .;
+
+	.rel.dyn : {
+		__rel_dyn_start = .;

Quoted 7 lines

Two new symbols are introduced which are intended to eventually be defined on all architectures to make things easier for generic relocation and reduce special-case code for each architecture:

__text_start is the start of the text area (equivalent to the existing _start on ARM). It marks the start of the region which must be copied to a new location during relocation.

Please name it _image_copy_start, not __text_start

Hi Simon,

Le 21/02/2012 07:17, Simon Glass a ?crit :

Quoted 7 lines

Most ARM CPUs use a very similar link script. This adds a basic script that can be used by most CPUs.

Two new symbols are introduced which are intended to eventually be defined on all architectures to make things easier for generic relocation and reduce special-case code for each architecture:

Quoted 3 lines

__text_start is the start of the text area (equivalent to the existing _start on ARM). It marks the start of the region which must be copied to a new location during relocation.

Please name it _image_copy_start, not __text_start

Quoted 88 lines

__image_copy_end is the end of the region which must be copied to a new location during relocation. It is normally equal to the start of the BSS region, but this can vary in some cases (SPL?). Making this an explicit symbol on its own removes any ambiguity and permits common code to always do the right thing.

This new script makes use of CPUDIR, now defined by both Makefile and spl/Makefile, to find the directory containing the start.o object file, which is always placed first in the image.

To permit MMU setup prior to relocation (as used by pxa) we add an area to the link script which contains space for this. This is taken from commit 7f4cfcf. CPUs can put the contents in there using their start.S file. BTW, shouldn't that area be 16KB-aligned?

Signed-off-by Simon Glass<[email protected]>

---
Changes in v3:
- Add more comments in the commit message
- Add section for MMU area, as required by pxa

?arch/arm/cpu/u-boot.lds | ? 91
+++++++++++++++++++++++++++++++++++++++++++++++
?1 files changed, 91 insertions(+), 0 deletions(-) ?create mode 100644 arch/arm/cpu/u-boot.lds

arch/arm/cpu/u-boot.lds new file +54 -0

new file mode 100644
index 0000000..7a859fc
@@ -0,0 +1,91 @@
+/*
+ * Copyright (c) 2004-2008 Texas Instruments
+ *
+ * (C) Copyright 2002
+ * Gary Jennejohn, DENX Software Engineering,<[email protected]>
+ *
+ * See file CREDITS for list of people who contributed to this
+ * project.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 of
+ * the License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston,
+ * MA 02111-1307 USA
+ */
+
+OUTPUT_FORMAT("elf32-littlearm", "elf32-littlearm", "elf32-littlearm")
+OUTPUT_ARCH(arm)
+ENTRY(_start)
+SECTIONS
+{
+ ? ? ? . = 0x00000000;
+
+ ? ? ? . = ALIGN(4);
+ ? ? ? .text :
+ ? ? ? {
+ ? ? ? ? ? ? ? __text_start = .;
+ ? ? ? ? ? ? ? CPUDIR/start.o (.text)
+ ? ? ? ? ? ? ? *(.text)
+ ? ? ? }
+
+ ? ? ? . = ALIGN(4);
+ ? ? ? .rodata : { *(SORT_BY_ALIGNMENT(SORT_BY_NAME(.rodata*))) }
+
+ ? ? ? . = ALIGN(4);
+ ? ? ? .data : {
+ ? ? ? ? ? ? ? *(.data)
+ ? ? ? }
+
+ ? ? ? . = ALIGN(4);
+
+ ? ? ? . = .;
+ ? ? ? __u_boot_cmd_start = .;
+ ? ? ? .u_boot_cmd : { *(.u_boot_cmd) }
+ ? ? ? __u_boot_cmd_end = .;

arch/arm/cpu/u-boot.lds line 56

Quoted 2 lines

arch/arm/cpu/u-boot.lds new file +2 -0

+
+ ? ? ? . = ALIGN(4);

This "ALIGN(4)" is what upsets my edminiv2 build and makes it not binary-identical. It is not in the current arm926ejs liner script. I will check if adding it to the current arm926ejs script produces a working and binary-identical u-boot.

Quoted 6 lines

arch/arm/cpu/u-boot.lds new file +4 -0

+ ? ? ? __image_copy_end = .;
+
+ ? ? ? .rel.dyn : {
+ ? ? ? ? ? ? ? __rel_dyn_start = .;

arch/arm/cpu/u-boot.lds line 70

Quoted 10 lines

arch/arm/cpu/u-boot.lds new file +10 -0

+ ? ? ? ? ? ? ? *(.rel*)
+ ? ? ? ? ? ? ? __rel_dyn_end = .;
+ ? ? ? }
+
+ ? ? ? .dynsym : {
+ ? ? ? ? ? ? ? __dynsym_start = .;
+ ? ? ? ? ? ? ? *(.dynsym)
+ ? ? ? }
+
+ ? ? ? _end = .;

Below is what I assume to be the reservation for MMU.

arch/arm/cpu/u-boot.lds line 75

Quoted 5 lines

arch/arm/cpu/u-boot.lds new file +5 -0

+ ? ? ? . = ALIGN(4096);
+
+ ? ? ? .mmutable : {
+ ? ? ? ? ? ? ? *(.mmutable)
+ ? ? ? }

... and I don't like it at all, more so if it is going to be actually 16 KB, because it seems to me we're wasting memory and creating a hole in the middle of our in-RAM binary, and I'm not sure we need to do it here in the first place. Do we need to have that MMU region mapped over DDR? And if so, do we need to have it mapped in the middle of u-boot? Plus you're placing it after .dynsym and .rel.dyn, but possibly overlaid with .bss. Normally this area should have gone by the time we get to use BSS, but assumption is the mother of all screw-ups.

Quoted 12 lines

arch/arm/cpu/u-boot.lds new file +10 -0

+ ? ? ? .bss __rel_dyn_start (OVERLAY) : {
+ ? ? ? ? ? ? ? __bss_start = .;
+ ? ? ? ? ? ? ? *(.bss)
+ ? ? ? ? ? ? ? ?. = ALIGN(4);
+ ? ? ? ? ? ? ? __bss_end__ = .;
+ ? ? ? }
+
+ ? ? ? /DISCARD/ : { *(.dynstr*) }
+ ? ? ? /DISCARD/ : { *(.dynamic*) }
+ ? ? ? /DISCARD/ : { *(.plt*) }

arch/arm/cpu/u-boot.lds line 88

Quoted 3 lines

arch/arm/cpu/u-boot.lds new file +3 -0

+ ? ? ? /DISCARD/ : { *(.interp*) }
+ ? ? ? /DISCARD/ : { *(.gnu*) }
+}

Amicalement,
--
Albert.

Quoted 7 lines

Quoted 1 line

+ . = ALIGN(4);

This "ALIGN(4)" is what upsets my edminiv2 build and makes it not binary-identical. It is not in the current arm926ejs liner script. I will check if adding it to the current arm926ejs script produces a working and binary-identical u-boot.

OK, but realistically we copy a word at a time (at least) and the .rel.dyn region is going to be word-aligned I think. All we are doing here is making sure that the image ends on a word boundary, which seems reasonable to me.

In other words, I would like the __image_copy_start and __image_copy_end symbols to be word-aligned.

Hi Simon,

Le 21/02/2012 18:02, Simon Glass a ?crit :

Quoted 15 lines

Quoted 8 lines

Quoted 1 line

+ ? ? ? . = ALIGN(4);

This "ALIGN(4)" is what upsets my edminiv2 build and makes it not binary-identical. It is not in the current arm926ejs liner script. I will check if adding it to the current arm926ejs script produces a working and binary-identical u-boot.

OK, but realistically we copy a word at a time (at least) and the .rel.dyn region is going to be word-aligned I think. All we are doing here is making sure that the image ends on a word boundary, which seems reasonable to me.

Quoted 2 lines

In other words, I would like the __image_copy_start and __image_copy_end symbols to be word-aligned.

I understand the requirement and agree to it -- only, it seems to me it is already fulfilled in the current lds files, so I still fail to see what breaks 'near binary identity'.

Amicalement,
--
Albert.

Hi Simon,

Le 21/02/2012 21:14, Simon Glass a ?crit :

Quoted 42 lines

Hi Albert,

Quoted 30 lines · On Tue, Feb 21, 2012 at 11:52 AM, Albert ARIBAUD <[email protected]>…

Hi Simon,

Le 21/02/2012 18:02, Simon Glass a ?crit :

Quoted 14 lines

Quoted 11 lines

Quoted 1 line

+ ? ? ? . = ALIGN(4);

This "ALIGN(4)" is what upsets my edminiv2 build and makes it not binary-identical. It is not in the current arm926ejs liner script. I will
check if adding it to the current arm926ejs script produces a working and
binary-identical u-boot.

Quoted 7 lines

OK, but realistically we copy a word at a time (at least) and the .rel.dyn region is going to be word-aligned I think. All we are doing here is making sure that the image ends on a word boundary, which seems reasonable to me.

In other words, I would like the __image_copy_start and __image_copy_end symbols to be word-aligned.

Quoted 4 lines

I understand the requirement and agree to it -- only, it seems to me it is
already fulfilled in the current lds files, so I still fail to see what breaks 'near binary identity'.

Quoted 3 lines

If we add any symbols or change the value of any of them, then the binary file may change. You can see the differences by doing something like:

This statement is true in a very general fashion, but becomes false for many cases, including the one where the symbol is added at link stage and not referenced in the build, which is the case for edminiv2. Then the ELF file's symbol sections may change, but the .text and .data sections have no reason to.

Plus, you did not address my note that the current .lds already aligns the start and end of copy section to a multiple of 4.

Quoted 5 lines

armv7a-cros-linux-gnueabi-objdump -b binary -m arm -D u-boot.bin

on each file (but I guess you know that).

Yes, plus 'od -t x1z -A x u-boot.bin > u-boot.lst' :) and it does not explain why the .text and .data sections should undergo any change when the .lds introduces a symbol that neither section references.

Quoted 4 lines

Regards,
Simon

Amicalement,
--
Albert.

Hi Albert,

On Tue, Feb 21, 2012 at 5:11 AM, Albert ARIBAUD

Quoted 10 lines · <[email protected]> wrote:

Hi Simon,

Le 21/02/2012 07:17, Simon Glass a ?crit :

Quoted 7 lines

Most ARM CPUs use a very similar link script. This adds a basic script that can be used by most CPUs.

Two new symbols are introduced which are intended to eventually be defined on all architectures to make things easier for generic relocation and reduce special-case code for each architecture:

Quoted 5 lines

Quoted 3 lines

__text_start is the start of the text area (equivalent to the existing _start on ARM). It marks the start of the region which must be copied to a new location during relocation.

Please name it _image_copy_start, not __text_start

OK will do

Quoted 85 lines

Quoted 85 lines

__image_copy_end is the end of the region which must be copied to a new location during relocation. It is normally equal to the start of the BSS region, but this can vary in some cases (SPL?). Making this an explicit symbol on its own removes any ambiguity and permits common code to always do the right thing.

This new script makes use of CPUDIR, now defined by both Makefile and spl/Makefile, to find the directory containing the start.o object file, which is always placed first in the image.

To permit MMU setup prior to relocation (as used by pxa) we add an area to the link script which contains space for this. This is taken from commit 7f4cfcf. CPUs can put the contents in there using their start.S file. BTW, shouldn't that area be 16KB-aligned?

Signed-off-by Simon Glass<[email protected]>

---
Changes in v3:
- Add more comments in the commit message
- Add section for MMU area, as required by pxa

 arch/arm/cpu/u-boot.lds |   91
+++++++++++++++++++++++++++++++++++++++++++++++
 1 files changed, 91 insertions(+), 0 deletions(-)
 create mode 100644 arch/arm/cpu/u-boot.lds

arch/arm/cpu/u-boot.lds new file +53 -0

new file mode 100644
index 0000000..7a859fc
@@ -0,0 +1,91 @@
+/*
+ * Copyright (c) 2004-2008 Texas Instruments
+ *
+ * (C) Copyright 2002
+ * Gary Jennejohn, DENX Software Engineering,<[email protected]>
+ *
+ * See file CREDITS for list of people who contributed to this
+ * project.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 of
+ * the License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston,
+ * MA 02111-1307 USA
+ */
+
+OUTPUT_FORMAT("elf32-littlearm", "elf32-littlearm", "elf32-littlearm")
+OUTPUT_ARCH(arm)
+ENTRY(_start)
+SECTIONS
+{
+       . = 0x00000000;
+
+       . = ALIGN(4);
+       .text :
+       {
+               __text_start = .;
+               CPUDIR/start.o (.text)
+               *(.text)
+       }
+
+       . = ALIGN(4);
+       .rodata : { *(SORT_BY_ALIGNMENT(SORT_BY_NAME(.rodata*))) }
+
+       . = ALIGN(4);
+       .data : {
+               *(.data)
+       }
+
+       . = ALIGN(4);
+
+       . = .;
+       __u_boot_cmd_start = .;
+       .u_boot_cmd : { *(.u_boot_cmd) }

Quoted 8 lines

arch/arm/cpu/u-boot.lds line 56

Quoted 3 lines

arch/arm/cpu/u-boot.lds new file +3 -0

+       __u_boot_cmd_end = .;
+
+       . = ALIGN(4);

This "ALIGN(4)" is what upsets my edminiv2 build and makes it not binary-identical. It is not in the current arm926ejs liner script. I will check if adding it to the current arm926ejs script produces a working and binary-identical u-boot.

OK, but realistically we copy a word at a time (at least) and the .rel.dyn region is going to be word-aligned I think. All we are doing here is making sure that the image ends on a word boundary, which seems reasonable to me.

In other words, I would like the __image_copy_start and __image_copy_end symbols to be word-aligned.

Quoted 23 lines

Quoted 4 lines

arch/arm/cpu/u-boot.lds new file +4 -0

+       __image_copy_end = .;
+
+       .rel.dyn : {
+               __rel_dyn_start = .;

arch/arm/cpu/u-boot.lds line 71

Quoted 10 lines

arch/arm/cpu/u-boot.lds new file +10 -0

+               *(.rel*)
+               __rel_dyn_end = .;
+       }
+
+       .dynsym : {
+               __dynsym_start = .;
+               *(.dynsym)
+       }
+
+       _end = .;

Below is what I assume to be the reservation for MMU.

arch/arm/cpu/u-boot.lds line 77

Quoted 5 lines

arch/arm/cpu/u-boot.lds new file +5 -0

+       . = ALIGN(4096);
+
+       .mmutable : {
+               *(.mmutable)
+       }

Quoted 8 lines

... and I don't like it at all, more so if it is going to be actually 16 KB, because it seems to me we're wasting memory and creating a hole in the middle of our in-RAM binary, and I'm not sure we need to do it here in the first place. Do we need to have that MMU region mapped over DDR? And if so, do we need to have it mapped in the middle of u-boot? Plus you're placing it after .dynsym and .rel.dyn, but possibly overlaid with .bss. Normally this area should have gone by the time we get to use BSS, but assumption is the mother of all screw-ups.

Yes it will be 16KB, although of course only if it exists. At present it is only used for pxa so including it here allows pxa boards to build.

The hole is not in the middle of the binary, but at the end. The BSS may overlay it if large enough, but that doesn't take affect until after relocation.

I'm not thrilled with it either, but it does allow us to have an option to set up the MMU table at compile time for those that need it. Having thought about it a bit more I feel that we should be able to set this up at run-time instead, but still, it is needed at present for pxa, so...

I have question also, for Marek. Should this be 16KB aligned, since ARM needs this I think. The alignment is currently provided by pxa's start.S but we might as well be explicit here I think.

Quoted 10 lines

Quoted 10 lines

arch/arm/cpu/u-boot.lds new file +9 -0

+       .bss __rel_dyn_start (OVERLAY) : {
+               __bss_start = .;
+               *(.bss)
+                . = ALIGN(4);
+               __bss_end__ = .;
+       }
+
+       /DISCARD/ : { *(.dynstr*) }
+       /DISCARD/ : { *(.dynamic*) }

Quoted 8 lines

arch/arm/cpu/u-boot.lds line 91

Quoted 4 lines

arch/arm/cpu/u-boot.lds new file +4 -0

+       /DISCARD/ : { *(.plt*) }
+       /DISCARD/ : { *(.interp*) }
+       /DISCARD/ : { *(.gnu*) }
+}

Amicalement,
--
Albert.

Regards,
Simon

Quoted 171 lines

Hi Albert,

On Tue, Feb 21, 2012 at 5:11 AM, Albert ARIBAUD

Quoted 10 lines · <[email protected]> wrote:

Hi Simon,

Le 21/02/2012 07:17, Simon Glass a ?crit :

Quoted 7 lines

Most ARM CPUs use a very similar link script. This adds a basic script that can be used by most CPUs.

Two new symbols are introduced which are intended to eventually be defined on all architectures to make things easier for generic relocation and reduce special-case code for each architecture:

Quoted 5 lines

Quoted 3 lines

__text_start is the start of the text area (equivalent to the existing _start on ARM). It marks the start of the region which must be copied to a new location during relocation.

Please name it _image_copy_start, not __text_start

OK will do

Quoted 85 lines

Quoted 85 lines

__image_copy_end is the end of the region which must be copied to a new location during relocation. It is normally equal to the start of the BSS region, but this can vary in some cases (SPL?). Making this an explicit symbol on its own removes any ambiguity and permits common code to always do the right thing.

This new script makes use of CPUDIR, now defined by both Makefile and spl/Makefile, to find the directory containing the start.o object file, which is always placed first in the image.

To permit MMU setup prior to relocation (as used by pxa) we add an area to the link script which contains space for this. This is taken from commit 7f4cfcf. CPUs can put the contents in there using their start.S file. BTW, shouldn't that area be 16KB-aligned?

Signed-off-by Simon Glass<[email protected]>

---
Changes in v3:
- Add more comments in the commit message
- Add section for MMU area, as required by pxa

?arch/arm/cpu/u-boot.lds | ? 91
+++++++++++++++++++++++++++++++++++++++++++++++
?1 files changed, 91 insertions(+), 0 deletions(-) ?create mode 100644 arch/arm/cpu/u-boot.lds

arch/arm/cpu/u-boot.lds new file +53 -0

new file mode 100644
index 0000000..7a859fc
@@ -0,0 +1,91 @@
+/*
+ * Copyright (c) 2004-2008 Texas Instruments
+ *
+ * (C) Copyright 2002
+ * Gary Jennejohn, DENX Software Engineering,<[email protected]>
+ *
+ * See file CREDITS for list of people who contributed to this
+ * project.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 of
+ * the License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston,
+ * MA 02111-1307 USA
+ */
+
+OUTPUT_FORMAT("elf32-littlearm", "elf32-littlearm", "elf32-littlearm")
+OUTPUT_ARCH(arm)
+ENTRY(_start)
+SECTIONS
+{
+ ? ? ? . = 0x00000000;
+
+ ? ? ? . = ALIGN(4);
+ ? ? ? .text :
+ ? ? ? {
+ ? ? ? ? ? ? ? __text_start = .;
+ ? ? ? ? ? ? ? CPUDIR/start.o (.text)
+ ? ? ? ? ? ? ? *(.text)
+ ? ? ? }
+
+ ? ? ? . = ALIGN(4);
+ ? ? ? .rodata : { *(SORT_BY_ALIGNMENT(SORT_BY_NAME(.rodata*))) }
+
+ ? ? ? . = ALIGN(4);
+ ? ? ? .data : {
+ ? ? ? ? ? ? ? *(.data)
+ ? ? ? }
+
+ ? ? ? . = ALIGN(4);
+
+ ? ? ? . = .;
+ ? ? ? __u_boot_cmd_start = .;
+ ? ? ? .u_boot_cmd : { *(.u_boot_cmd) }

Quoted 8 lines

arch/arm/cpu/u-boot.lds line 56

Quoted 3 lines

arch/arm/cpu/u-boot.lds new file +3 -0

+ ? ? ? __u_boot_cmd_end = .;
+
+ ? ? ? . = ALIGN(4);

This "ALIGN(4)" is what upsets my edminiv2 build and makes it not binary-identical. It is not in the current arm926ejs liner script. I will check if adding it to the current arm926ejs script produces a working and binary-identical u-boot.

OK, but realistically we copy a word at a time (at least) and the .rel.dyn region is going to be word-aligned I think. All we are doing here is making sure that the image ends on a word boundary, which seems reasonable to me.

In other words, I would like the __image_copy_start and __image_copy_end symbols to be word-aligned.

Quoted 23 lines

Quoted 4 lines

arch/arm/cpu/u-boot.lds new file +4 -0

+ ? ? ? __image_copy_end = .;
+
+ ? ? ? .rel.dyn : {
+ ? ? ? ? ? ? ? __rel_dyn_start = .;

arch/arm/cpu/u-boot.lds line 70

Quoted 10 lines

arch/arm/cpu/u-boot.lds new file +10 -0

+ ? ? ? ? ? ? ? *(.rel*)
+ ? ? ? ? ? ? ? __rel_dyn_end = .;
+ ? ? ? }
+
+ ? ? ? .dynsym : {
+ ? ? ? ? ? ? ? __dynsym_start = .;
+ ? ? ? ? ? ? ? *(.dynsym)
+ ? ? ? }
+
+ ? ? ? _end = .;

Below is what I assume to be the reservation for MMU.

arch/arm/cpu/u-boot.lds line 75

Quoted 5 lines

arch/arm/cpu/u-boot.lds new file +5 -0

+ ? ? ? . = ALIGN(4096);
+
+ ? ? ? .mmutable : {
+ ? ? ? ? ? ? ? *(.mmutable)
+ ? ? ? }

Quoted 8 lines

... and I don't like it at all, more so if it is going to be actually 16 KB, because it seems to me we're wasting memory and creating a hole in the middle of our in-RAM binary, and I'm not sure we need to do it here in the first place. Do we need to have that MMU region mapped over DDR? And if so, do we need to have it mapped in the middle of u-boot? Plus you're placing it after .dynsym and .rel.dyn, but possibly overlaid with .bss. Normally this area should have gone by the time we get to use BSS, but assumption is the mother of all screw-ups.

Yes it will be 16KB, although of course only if it exists. At present it is only used for pxa so including it here allows pxa boards to build.

The hole is not in the middle of the binary, but at the end. The BSS may overlay it if large enough, but that doesn't take affect until after relocation.

I'm not thrilled with it either, but it does allow us to have an option to set up the MMU table at compile time for those that need it. Having thought about it a bit more I feel that we should be able to set this up at run-time instead, but still, it is needed at present for pxa, so...

Quoted 3 lines

I have question also, for Marek. Should this be 16KB aligned, since ARM needs this I think. The alignment is currently provided by pxa's start.S but we might as well be explicit here I think.

Yes, but I believe it should be only 4k aligned. This part is special in more ways actually.

This is only used on PXA250, since that has no SRAM and therefore no place for stack at boottime. But to allow locking cachelines (to create space for stack) we need the MMU table ready. Therefore I created the MMU table with 1:1 mapping in the ROM at compile time and in the init code I point the MMU to this table.

That's about it

M

Quoted 22 lines

Quoted 11 lines

Quoted 11 lines

arch/arm/cpu/u-boot.lds new file +9 -0

+ ? ? ? .bss __rel_dyn_start (OVERLAY) : {
+ ? ? ? ? ? ? ? __bss_start = .;
+ ? ? ? ? ? ? ? *(.bss)
+ ? ? ? ? ? ? ? ?. = ALIGN(4);
+ ? ? ? ? ? ? ? __bss_end__ = .;
+ ? ? ? }
+
+ ? ? ? /DISCARD/ : { *(.dynstr*) }
+ ? ? ? /DISCARD/ : { *(.dynamic*) }

Quoted 8 lines

arch/arm/cpu/u-boot.lds line 88

Quoted 4 lines

arch/arm/cpu/u-boot.lds new file +4 -0

+ ? ? ? /DISCARD/ : { *(.plt*) }
+ ? ? ? /DISCARD/ : { *(.interp*) }
+ ? ? ? /DISCARD/ : { *(.gnu*) }
+}

Amicalement,
--
Albert.

Regards,
Simon