linux-arm-kernel.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
From: Dave.Martin@arm.com (Dave Martin)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH 03/19] ARM: fixup_pv_table bug when CPU_ENDIAN_BE8
Date: Wed, 28 Aug 2013 17:54:27 +0100	[thread overview]
Message-ID: <20130828165427.GI3842@localhost.localdomain> (raw)
In-Reply-To: <1377639509-22778-4-git-send-email-ben.dooks@codethink.co.uk>

On Tue, Aug 27, 2013 at 10:38:13PM +0100, Ben Dooks wrote:
> The fixup_pv_table assumes that the instructions are in the same
> endian configuration as the data, but when the CPU is running in
> BE8 the instructions stay in little-endian format.
> 
> Make sure if CONFIG_CPU_ENDIAN_BE8 is set that we do all the
> alterations to the instructions taking in to account the LDR/STR
> will be swapping the data endian-ness.
> 
> Since the code is only modifying a byte, we avoid dual-swapping
> the data, and just change the bits we clear and ORR in (in the
> case where the code is not thumb2).
> 
> For thumb2, we add the necessary rev16 instructions to ensure that
> the instructions are processed in the correct format, as it was
> easier than re-writing the code to contain a mask and shift.

Shame about the ifdef, but ARM_LE_OR_BE32(...) or ARM_NOT_BE8(...) seems
too contrived to be practical...

We could probably do bitfield clobber more neatly with BFI, but
unfortunately v6 plain doesn't have that.  Only v6T2 or v7 and later
have it.

Reviewed-by: Dave Martin <Dave.Martin@arm.com>

Cheers
---Dave

> 
> Signed-off-by: Ben Dooks <ben.dooks@codethink.co.uk>
> ---
>  arch/arm/kernel/head.S |    8 ++++++++
>  1 file changed, 8 insertions(+)
> 
> diff --git a/arch/arm/kernel/head.S b/arch/arm/kernel/head.S
> index 2c7cc1e..9e5906c 100644
> --- a/arch/arm/kernel/head.S
> +++ b/arch/arm/kernel/head.S
> @@ -582,8 +582,10 @@ __fixup_a_pv_table:
>  	b	2f
>  1:	add     r7, r3
>  	ldrh	ip, [r7, #2]
> +ARM_BE8(rev16	ip, ip)
>  	and	ip, 0x8f00
>  	orr	ip, r6	@ mask in offset bits 31-24
> +ARM_BE8(rev16	ip, ip)
>  	strh	ip, [r7, #2]
>  2:	cmp	r4, r5
>  	ldrcc	r7, [r4], #4	@ use branch for delay slot
> @@ -592,8 +594,14 @@ __fixup_a_pv_table:
>  #else
>  	b	2f
>  1:	ldr	ip, [r7, r3]
> +#ifdef CONFIG_CPU_ENDIAN_BE8
> +	@ in BE8, we load data in BE, but instructions still in LE
> +	bic	ip, ip, #0xff000000
> +	orr	ip, ip, r6, lsl#24
> +#else
>  	bic	ip, ip, #0x000000ff
>  	orr	ip, ip, r6	@ mask in offset bits 31-24
> +#endif
>  	str	ip, [r7, r3]
>  2:	cmp	r4, r5
>  	ldrcc	r7, [r4], #4	@ use branch for delay slot
> -- 
> 1.7.10.4
> 
> 
> _______________________________________________
> linux-arm-kernel mailing list
> linux-arm-kernel at lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

  reply	other threads:[~2013-08-28 16:54 UTC|newest]

Thread overview: 62+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-08-27 21:38 3.11-rc7 big-endian support Ben Dooks
2013-08-27 21:38 ` [PATCH 01/19] ARM: fix ARCH_IXP4xx usage of ARCH_SUPPORTS_BIG_ENDIAN Ben Dooks
2013-08-27 21:38 ` [PATCH 02/19] ARM: asm: Add ARM_BE8() assembly helper Ben Dooks
2013-08-27 21:38 ` [PATCH 03/19] ARM: fixup_pv_table bug when CPU_ENDIAN_BE8 Ben Dooks
2013-08-28 16:54   ` Dave Martin [this message]
2013-08-27 21:38 ` [PATCH 04/19] ARM: set BE8 if LE in head code Ben Dooks
2013-08-28 10:21   ` Thomas Petazzoni
2013-08-28 16:42   ` Dave Martin
2013-08-27 21:38 ` [PATCH 05/19] ARM: pl01x debug code endian fix Ben Dooks
2013-08-28 17:03   ` Dave Martin
2013-08-27 21:38 ` [PATCH 06/19] ARM: twd: data " Ben Dooks
2013-08-27 21:38 ` [PATCH 07/19] ARM: smp_scu: data endian fixes Ben Dooks
2013-08-27 21:38 ` [PATCH 08/19] highbank: enable big-endian Ben Dooks
2013-08-28 15:21   ` Rob Herring
2013-08-30 18:50     ` Ben Dooks
2013-08-27 21:38 ` [PATCH 09/19] mvebu: support running big-endian Ben Dooks
2013-08-27 21:38 ` [PATCH 10/19] vexpress: add big endian support Ben Dooks
2013-08-27 21:38 ` [PATCH 11/19] ARM: alignment: correctly decode instructions in BE8 mode Ben Dooks
2013-08-28 16:44   ` Dave Martin
2013-08-27 21:38 ` [PATCH 12/19] ARM: traps: use <asm/opcodes.h> to get correct instruction order Ben Dooks
2013-08-28 11:22   ` Dave Martin
2013-08-27 21:38 ` [PATCH 13/19] ARM: module: correctly relocate instructions in BE8 Ben Dooks
2013-08-28 16:46   ` Dave Martin
2013-08-27 21:38 ` [PATCH 14/19] ARM: set --be8 when linking modules Ben Dooks
2013-08-28 16:38   ` Dave Martin
2013-08-30 18:55     ` Ben Dooks
2013-08-27 21:38 ` [PATCH 15/19] ARM: hardware: fix endian-ness in <hardware/coresight.h> Ben Dooks
2013-08-27 21:38 ` [PATCH 16/19] ARM: net: fix arm instruction endian-ness in bpf_jit_32.c Ben Dooks
2013-08-28 16:47   ` Dave Martin
2013-08-27 21:38 ` [PATCH 17/19] ARM: Correct BUG() assembly to ensure it is endian-agnostic Ben Dooks
2013-08-28 16:48   ` Dave Martin
2013-08-27 21:38 ` [PATCH 18/19] ARM: kdgb: use <asm/opcodes.h> for data to be assembled as intruction Ben Dooks
2013-08-28 16:50   ` Dave Martin
2013-08-27 21:38 ` [PATCH 19/19] ARM: atomic64: fix endian-ness in atomic.h Ben Dooks
2013-08-27 22:02 ` 3.11-rc7 big-endian support Ben Dooks
2013-08-29  9:59   ` Thomas Petazzoni
2013-08-29 10:23     ` Ben Dooks
2013-08-29 10:28       ` Thomas Petazzoni
2013-08-29 15:53         ` Victor Kamensky
2013-08-29 17:17           ` Ben Dooks
2013-08-28 11:45 ` re-send patch series due to mta issues Ben Dooks
2013-08-28 11:45   ` [PATCH 01/19] ARM: fix ARCH_IXP4xx usage of ARCH_SUPPORTS_BIG_ENDIAN Ben Dooks
2013-08-28 11:45   ` [PATCH 02/19] ARM: asm: Add ARM_BE8() assembly helper Ben Dooks
2013-08-28 11:45   ` [PATCH 03/19] ARM: fixup_pv_table bug when CPU_ENDIAN_BE8 Ben Dooks
2013-08-28 11:45   ` [PATCH 04/19] ARM: set BE8 if LE in head code Ben Dooks
2013-08-28 11:45   ` [PATCH 05/19] ARM: pl01x debug code endian fix Ben Dooks
2013-08-28 11:45   ` [PATCH 06/19] ARM: twd: data " Ben Dooks
2013-08-28 11:45   ` [PATCH 07/19] ARM: smp_scu: data endian fixes Ben Dooks
2013-08-28 11:45   ` [PATCH 08/19] highbank: enable big-endian Ben Dooks
2013-08-28 11:45   ` [PATCH 09/19] mvebu: support running big-endian Ben Dooks
2013-08-28 11:45   ` [PATCH 10/19] vexpress: add big endian support Ben Dooks
2013-08-28 11:45   ` [PATCH 11/19] ARM: alignment: correctly decode instructions in BE8 mode Ben Dooks
2013-08-28 19:40     ` Sergei Shtylyov
2013-08-30 18:47       ` Ben Dooks
2013-08-28 11:45   ` [PATCH 12/19] ARM: traps: use <asm/opcodes.h> to get correct instruction order Ben Dooks
2013-08-28 11:45   ` [PATCH 13/19] ARM: module: correctly relocate instructions in BE8 Ben Dooks
2013-08-28 11:45   ` [PATCH 14/19] ARM: set --be8 when linking modules Ben Dooks
2013-08-28 11:45   ` [PATCH 15/19] ARM: hardware: fix endian-ness in <hardware/coresight.h> Ben Dooks
2013-08-28 11:45   ` [PATCH 16/19] ARM: net: fix arm instruction endian-ness in bpf_jit_32.c Ben Dooks
2013-08-28 11:45   ` [PATCH 17/19] ARM: Correct BUG() assembly to ensure it is endian-agnostic Ben Dooks
2013-08-28 11:45   ` [PATCH 18/19] ARM: kdgb: use <asm/opcodes.h> for data to be assembled as intruction Ben Dooks
2013-08-28 11:45   ` [PATCH 19/19] ARM: atomic64: fix endian-ness in atomic.h Ben Dooks

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=20130828165427.GI3842@localhost.localdomain \
    --to=dave.martin@arm.com \
    --cc=linux-arm-kernel@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;
as well as URLs for NNTP newsgroup(s).