Mailing List Archive: [PATCH] bt8xx: unaligned access

[PATCH] bt8xx: unaligned access

May 20, 2008, 5:32 PM

Post #1 of 4 (188 views)

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
drivers/media/video/bt8xx/bttv-cards.c | 5 +++--
1 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/media/video/bt8xx/bttv-cards.c b/drivers/media/video/bt8xx/bttv-cards.c
index f20a01c..8ef0424 100644
--- a/drivers/media/video/bt8xx/bttv-cards.c
+++ b/drivers/media/video/bt8xx/bttv-cards.c
@@ -34,6 +34,7 @@
#include <linux/firmware.h>
#include <net/checksum.h>

+#include <asm/unaligned.h>
#include <asm/io.h>

#include "bttvp.h"
@@ -3858,7 +3859,7 @@ static void __devinit osprey_eeprom(struct bttv *btv, const u8 ee[256])
ee += i;

/* found a valid descriptor */
- type = be16_to_cpup((u16*)(ee+4));
+ type = get_unaligned_be16((__be16 *)(ee+4));

switch(type) {
/* 848 based */
@@ -3918,7 +3919,7 @@ static void __devinit osprey_eeprom(struct bttv *btv, const u8 ee[256])
btv->c.nr, type);
break;
}
- serial = be32_to_cpup((u32*)(ee+6));
+ serial = get_unaligned_be32((__be32 *)(ee+6));
}

printk(KERN_INFO "bttv%d: osprey eeprom: card=%d '%s' serial=%u\n",
--
1.5.3.GIT

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

Re: [PATCH] bt8xx: unaligned access [ In reply to ]

harvey.harrison at gmail

May 20, 2008, 5:38 PM

Post #2 of 4 (198 views)

Permalink

On Wed, 2008-05-21 at 01:32 +0100, Al Viro wrote:
> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
> ---
> drivers/media/video/bt8xx/bttv-cards.c | 5 +++--
> 1 files changed, 3 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/media/video/bt8xx/bttv-cards.c b/drivers/media/video/bt8xx/bttv-cards.c
> index f20a01c..8ef0424 100644
> --- a/drivers/media/video/bt8xx/bttv-cards.c
> +++ b/drivers/media/video/bt8xx/bttv-cards.c
> @@ -34,6 +34,7 @@
> #include <linux/firmware.h>
> #include <net/checksum.h>
>
> +#include <asm/unaligned.h>
> #include <asm/io.h>
>
> #include "bttvp.h"
> @@ -3858,7 +3859,7 @@ static void __devinit osprey_eeprom(struct bttv *btv, const u8 ee[256])
> ee += i;
>
> /* found a valid descriptor */
> - type = be16_to_cpup((u16*)(ee+4));
> + type = get_unaligned_be16((__be16 *)(ee+4));

The casts aren't strictly necessary here. Any reason to keep it?

>
> switch(type) {
> /* 848 based */
> @@ -3918,7 +3919,7 @@ static void __devinit osprey_eeprom(struct bttv *btv, const u8 ee[256])
> btv->c.nr, type);
> break;
> }
> - serial = be32_to_cpup((u32*)(ee+6));
> + serial = get_unaligned_be32((__be32 *)(ee+6));

Ditto.

Cheers,

Harvey

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

Re: [PATCH] bt8xx: unaligned access [ In reply to ]

viro at ZenIV

May 20, 2008, 5:43 PM

Post #3 of 4 (187 views)

Permalink

On Tue, May 20, 2008 at 05:38:43PM -0700, Harvey Harrison wrote:
> > - type = be16_to_cpup((u16*)(ee+4));
> > + type = get_unaligned_be16((__be16 *)(ee+4));
>
> The casts aren't strictly necessary here. Any reason to keep it?

I'd rather have them explicit *and* make get_unaligned_* take the pointer
of right type, so that it'd eat things like void * and __be16 *, but warn
on int * et.al. One too many brainos seen in that area...
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

Re: [PATCH] bt8xx: unaligned access [ In reply to ]

harvey.harrison at gmail

May 20, 2008, 5:45 PM

Post #4 of 4 (188 views)

Permalink

On Wed, 2008-05-21 at 01:43 +0100, Al Viro wrote:
> On Tue, May 20, 2008 at 05:38:43PM -0700, Harvey Harrison wrote:
> > > - type = be16_to_cpup((u16*)(ee+4));
> > > + type = get_unaligned_be16((__be16 *)(ee+4));
> >
> > The casts aren't strictly necessary here. Any reason to keep it?
>
> I'd rather have them explicit *and* make get_unaligned_* take the pointer
> of right type, so that it'd eat things like void * and __be16 *, but warn
> on int * et.al. One too many brainos seen in that area...

You want me to patch the unaligned guys to take typed pointers and
follow it through? Or do you mind?

Cheers,

Harvey

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