From: Sascha Hauer <s.hauer@pengutronix.de>
To: Antony Pavlov <antonynpavlov@gmail.com>
Cc: barebox@lists.infradead.org
Subject: Re: [RFC 2/3] MIPS: boot: realize _start in C
Date: Tue, 20 Nov 2012 08:58:18 +0100 [thread overview]
Message-ID: <20121120075818.GB10369@pengutronix.de> (raw)
In-Reply-To: <1353154239-31574-3-git-send-email-antonynpavlov@gmail.com>
On Sat, Nov 17, 2012 at 04:10:38PM +0400, Antony Pavlov wrote:
> The pbl will be incharge of the lowlevel init and
> barebox relocation. So we can skip it in barebox itself
> and use simple _start.
>
> Signed-off-by: Antony Pavlov <antonynpavlov@gmail.com>
> ---
> arch/mips/boot/Makefile | 1 -
> arch/mips/boot/main_entry.c | 18 ++++++
> arch/mips/boot/start.S | 142 -------------------------------------------
> 3 files changed, 18 insertions(+), 143 deletions(-)
> delete mode 100644 arch/mips/boot/start.S
>
> diff --git a/arch/mips/boot/Makefile b/arch/mips/boot/Makefile
> index 6b093f1..4f11a48 100644
> --- a/arch/mips/boot/Makefile
> +++ b/arch/mips/boot/Makefile
> @@ -1,4 +1,3 @@
> -obj-y += start.o
> obj-y += main_entry.o
>
> pbl-y += start-pbl.o main_entry-pbl.o
> diff --git a/arch/mips/boot/main_entry.c b/arch/mips/boot/main_entry.c
> index 0a33c45..422d40b 100644
> --- a/arch/mips/boot/main_entry.c
> +++ b/arch/mips/boot/main_entry.c
> @@ -22,6 +22,8 @@
> #include <asm/cpu-features.h>
> #include <asm/mipsregs.h>
> #include <asm/addrspace.h>
> +#include <asm/regdef.h>
> +#include <asm-generic/memory_layout.h>
>
> extern void start_barebox(void);
> extern void handle_reserved(void);
> @@ -92,3 +94,19 @@ void main_entry(void)
>
> start_barebox();
> }
> +
> +void __naked __section(.text_entry) _start(void)
> +{
> + u32 r;
> +
> + /* setup the stack */
> + r = STACK_BASE + STACK_SIZE - 16;
> +
> + __asm__ __volatile__(
> + "move\t$sp, %0\n\t"
> + : /* no outputs */
> + : "r" (r)
> + );
> +
> + main_entry();
I wonder that the C function skips several things the Assembler version
does. Is this done in the pbl now? Does this patch make the pbl
mandatory? We haven't done this on Arm, mostly because we were not
confident enough that it works on every board. That may be different on
mips, but should be mentioned in the commit log. But maybe I am
misreading the patches...
Sascha
> +}
> diff --git a/arch/mips/boot/start.S b/arch/mips/boot/start.S
> deleted file mode 100644
> index 7e2ae5e..0000000
> --- a/arch/mips/boot/start.S
> +++ /dev/null
> @@ -1,142 +0,0 @@
> -/*
> - * Startup Code for MIPS CPU
> - *
> - * Copyright (C) 2011 Antony Pavlov <antonynpavlov@gmail.com>
> - * Used code copyrighted (C) 2009 by Shinya Kuribayashi <skuribay@pobox.com>
> - *
> - * This file is part of barebox.
> - * 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 version 2
> - * as published by the Free Software Foundation.
> - *
> - * 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.
> - *
> - */
> -
> -#include <asm/regdef.h>
> -#include <asm/mipsregs.h>
> -#include <asm/asm.h>
> -#include <asm-generic/memory_layout.h>
> -#include <generated/compile.h>
> -#include <generated/utsrelease.h>
> -
> - /*
> - * ADR macro instruction (inspired by ARM)
> - *
> - * ARM architecture doesn't have PC-relative jump instruction
> - * like MIPS' B/BAL insns. When ARM makes PC-relative jumps,
> - * it uses ADR insn. ADR is used to get a destination address
> - * of 'label' against current PC. With this, ARM can safely
> - * make PC-relative jumps.
> - */
> - .macro ADR rd label temp
> - .set push
> - .set noreorder
> - move \temp, ra # preserve ra beforehand
> - bal _pc
> - nop
> -_pc: addiu \rd, ra, \label - _pc # label is assumed to be
> - move ra, \temp # within pc +/- 32KB
> - .set pop
> - .endm
> -
> - .set noreorder
> - .text
> - .section ".text_bare_init"
> - .align 4
> -
> -EXPORT(_start)
> -
> - b __start
> - nop
> -
> - .org 0x10
> - .ascii "barebox " UTS_RELEASE " " UTS_VERSION
> - .byte 0
> -
> - .align 4
> -__start:
> - /* disable watchpoints */
> - mtc0 zero, CP0_WATCHLO
> - mtc0 zero, CP0_WATCHHI
> -
> - /* disable interrupts */
> - mfc0 k0, CP0_STATUS
> - li k1, ~ST0_IE
> - and k0, k1
> - mtc0 k0, CP0_STATUS
> -
> - /* copy barebox to link location */
> - ADR a0, _start, t1 /* a0 <- pc-relative position of _start */
> -
> - la a1, _start /* link (RAM) _start address */
> -
> - beq a0, a1, stack_setup
> - nop
> -
> - la t0, _start
> - la t1, __bss_start
> - subu t2, t1, t0 /* t2 <- size of barebox */
> - addu a2, a0, t2 /* a2 <- source end address */
> -
> -#define LONGSIZE 4
> -
> -copy_loop:
> - /* copy from source address [a0] */
> - lw t4, LONGSIZE * 0(a0)
> - lw t5, LONGSIZE * 1(a0)
> - lw t6, LONGSIZE * 2(a0)
> - lw t7, LONGSIZE * 3(a0)
> - /* copy fo target address [a1] */
> - sw t4, LONGSIZE * 0(a1)
> - sw t5, LONGSIZE * 1(a1)
> - sw t6, LONGSIZE * 2(a1)
> - sw t7, LONGSIZE * 3(a1)
> - addi a0, LONGSIZE * 4
> - subu t3, a0, a2
> - blez t3, copy_loop
> - addi a1, LONGSIZE * 4
> -
> - /*
> - * Dominic Sweetman, See MIPS Run, Morgan Kaufmann, 2nd edition, 2006
> - *
> - * 11.2.2 Stack Argument Structure in o32
> - * ...
> - * At the point where a function is called, sp must be
> - * eight-byte-aligned, matching the alignment of the largest
> - * basic types -- a long long integer or a floating-point double.
> - * The eight-byte alignment is not required by 32-bit MIPS integer
> - * hardware, but it's essential for compatibility with CPUs with
> - * 64-bit registers, and thus part of the rules. Subroutines fit
> - * in with this by always adjusting the stack pointer by a multiple
> - * of eight.
> - * ...
> - * SGI's n32 and n64 standards call for the stack to be maintained
> - * with 16-byte alignment.
> - *
> - */
> -
> -#if (STACK_BASE + STACK_SIZE) % 16 != 0
> -#error stack pointer must be 16-byte-aligned
> -#endif
> -
> -stack_setup:
> - la sp, STACK_BASE + STACK_SIZE
> -
> - /* reserve four 32-bit argument slots */
> - addiu sp, -16
> -
> - la v0, main_entry
> - jal v0
> - nop
> -
> - /* No return */
> -
> -__error:
> - b __error
> - nop
> --
> 1.7.10.4
>
>
> _______________________________________________
> barebox mailing list
> barebox@lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/barebox
>
--
Pengutronix e.K. | |
Industrial Linux Solutions | http://www.pengutronix.de/ |
Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0 |
Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 |
_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox
next prev parent reply other threads:[~2012-11-20 7:58 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-11-17 12:10 [RFC 0/3] MIPS: add pre-bootloader (pbl) image support Antony Pavlov
2012-11-17 12:10 ` [RFC 1/3] " Antony Pavlov
2012-11-20 7:59 ` Sascha Hauer
2012-11-24 15:45 ` Antony Pavlov
2012-11-26 8:05 ` Sascha Hauer
2012-11-17 12:10 ` [RFC 2/3] MIPS: boot: realize _start in C Antony Pavlov
2012-11-20 7:58 ` Sascha Hauer [this message]
2012-11-24 16:18 ` Antony Pavlov
2012-11-17 12:10 ` [RFC 3/3] MIPS: qemu-malta_defconfig: use pbl Antony Pavlov
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=20121120075818.GB10369@pengutronix.de \
--to=s.hauer@pengutronix.de \
--cc=antonynpavlov@gmail.com \
--cc=barebox@lists.infradead.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