[PATCH] rpmsg: fix build warning when compiling 64-bit

[PATCH] rpmsg: fix build warning when compiling 64-bit

[Mark Asselstine]

dev_dbg() in rpmsg_probe() made use of the %x formatting
which expects an 'unsigned int' which dma_addr_t is not
when compiling for 64-bit. Casting to a 'void *' and
using %p will avoid this and not force an upcast to a
64-bit value for 32-bit builds.

Signed-off-by: Mark Asselstine <mark.asselstine@windriver.com>
CC: Ohad Ben-Cohen <ohad@wizery.com>
---
 drivers/rpmsg/virtio_rpmsg_bus.c |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/rpmsg/virtio_rpmsg_bus.c b/drivers/rpmsg/virtio_rpmsg_bus.c
index 8980ac2..28558f6 100644
--- a/drivers/rpmsg/virtio_rpmsg_bus.c
+++ b/drivers/rpmsg/virtio_rpmsg_bus.c
@@ -891,8 +891,8 @@ static int rpmsg_probe(struct virtio_device *vdev)
 	if (!bufs_va)
 		goto vqs_del;
 
-	dev_dbg(&vdev->dev, "buffers: va %p, dma 0x%x\n", bufs_va,
-						vrp->bufs_dma);
+	dev_dbg(&vdev->dev, "buffers: va %p, dma %p\n", bufs_va,
+						(void *)vrp->bufs_dma);
 
 	/* half of the buffers is dedicated for RX */
 	vrp->rbufs = bufs_va;
-- 
1.7.5.4

[PATCH] rpmsg: fix build warning when compiling 64-bit

[Arnd Bergmann]

On Wednesday 29 February 2012, Mark Asselstine wrote:
> dev_dbg() in rpmsg_probe() made use of the %x formatting
> which expects an 'unsigned int' which dma_addr_t is not
> when compiling for 64-bit. Casting to a 'void *' and
> using %p will avoid this and not force an upcast to a
> 64-bit value for 32-bit builds.

This does not work on 32-bit platforms with a 64-bit dma_addr_t
though. Better cast to unsigned long long and print using 0x%llx.

	Arnd

[PATCH V2] rpmsg: fix build warning when dma_addr_t is 64-bit

[Mark Asselstine]

dev_dbg() in rpmsg_probe() made use of the %x formatting that
expects an 'unsigned int' which dma_addr_t is not in cases where
dma_addr_t is 64-bit (CONFIG_ARCH_DMA_ADDR_T_64BIT). Casting to
a 'unsigned long long' and using %llx will avoid this.

Signed-off-by: Mark Asselstine <mark.asselstine@windriver.com>
CC: Ohad Ben-Cohen <ohad@wizery.com>
CC: Arnd Bergmann <arnd@arndb.de>
---
 drivers/rpmsg/virtio_rpmsg_bus.c |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/rpmsg/virtio_rpmsg_bus.c b/drivers/rpmsg/virtio_rpmsg_bus.c
index 8980ac2..c142e7b 100644
--- a/drivers/rpmsg/virtio_rpmsg_bus.c
+++ b/drivers/rpmsg/virtio_rpmsg_bus.c
@@ -891,8 +891,8 @@ static int rpmsg_probe(struct virtio_device *vdev)
 	if (!bufs_va)
 		goto vqs_del;
 
-	dev_dbg(&vdev->dev, "buffers: va %p, dma 0x%x\n", bufs_va,
-						vrp->bufs_dma);
+	dev_dbg(&vdev->dev, "buffers: va %p, dma 0x%llx\n", bufs_va,
+					(unsigned long long)vrp->bufs_dma);
 
 	/* half of the buffers is dedicated for RX */
 	vrp->rbufs = bufs_va;
-- 
1.7.5.4

[PATCH V2] rpmsg: fix build warning when dma_addr_t is 64-bit

[Ohad Ben-Cohen]

On Wed, Feb 29, 2012 at 5:41 PM, Mark Asselstine
<mark.asselstine@windriver.com> wrote:
> dev_dbg() in rpmsg_probe() made use of the %x formatting that
> expects an 'unsigned int' which dma_addr_t is not in cases where
> dma_addr_t is 64-bit (CONFIG_ARCH_DMA_ADDR_T_64BIT). Casting to
> a 'unsigned long long' and using %llx will avoid this.
>
> Signed-off-by: Mark Asselstine <mark.asselstine@windriver.com>
> CC: Ohad Ben-Cohen <ohad@wizery.com>
> CC: Arnd Bergmann <arnd@arndb.de>

Thanks, Mark !

I'll send this to Arnd later this week together with the other
remoteproc/rpmsg patches.

Ohad.