From mboxrd@z Thu Jan  1 00:00:00 1970
From: Alexander Holler <holler@ahsoftware.de>
Date: Sat, 02 Apr 2011 10:32:38 +0200
Subject: [U-Boot] [PATCH] USB: Remove __attribute__ ((packed)) for
 struct ehci_hccr and ehci_hcor.
In-Reply-To: <1301697344-4033-1-git-send-email-holler@ahsoftware.de>
References: <1301697344-4033-1-git-send-email-holler@ahsoftware.de>
Message-ID: <4D96DF26.6040408@ahsoftware.de>
List-Id: <u-boot.lists.denx.de>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
To: u-boot@lists.denx.de

Hello,

Am 02.04.2011 00:35, schrieb Alexander Holler:
> Remove __attribute__ ((packed)) to prevent byte access to soc
> registers in some gcc versions.
>
> Having patches to enable ehci for the BeagleBoard lying around for
> several month, this one was the show-stopper.
>
> Credits have to go to Laine Walker-Avina<lwalkera@ieee.org>  for
> finding the problem.
>
> Signed-off-by: Alexander Holler<holler@ahsoftware.de>
> ---
>   drivers/usb/host/ehci.h |    4 ++--
>   1 files changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/usb/host/ehci.h b/drivers/usb/host/ehci.h
> index 945ab64..df9f055 100644
> --- a/drivers/usb/host/ehci.h
> +++ b/drivers/usb/host/ehci.h
> @@ -55,7 +55,7 @@ struct ehci_hccr {
>   #define HCS_N_PORTS(p)		(((p)>>  0)&  0xf)
>   	uint32_t cr_hccparams;
>   	uint8_t cr_hcsp_portrt[8];
> -} __attribute__ ((packed));
> +};
>
>   struct ehci_hcor {
>   	uint32_t or_usbcmd;
> @@ -85,7 +85,7 @@ struct ehci_hcor {
>   #define FLAG_CF		(1<<  0)	/* true:  we'll support "high speed" */
>   	uint32_t or_portsc[CONFIG_SYS_USB_EHCI_MAX_ROOT_PORTS];
>   	uint32_t or_systune;
> -} __attribute__ ((packed));
> +};
>
>   #define USBMODE		0x68		/* USB Device mode */
>   #define USBMODE_SDIS	(1<<  3)	/* Stream disable */

Before I'm killing someones 64bit+ machine with that patch:

The hint I've received was to use

+} __attribute__ ((packed, aligned(4)));

That works too. I haven't seen the original commit (I've only got told 
that credits have to go to Laine Walker-Avina), I don't know if the 
standard says something to that, I don't know if some 64bit+ SoC might 
choose to align that stuff otherwise and I will not send a v2 of that patch.

Regards,

Alexander