[PATCH] fix sierra_net endianess bug

[PATCH] fix sierra_net endianess bug

[Maris Paupe]

Hello,
I have found that sierra_net does not work properly on big-endian
systems, it works if bytes are flipped in this particular location.

Signed-off-by: Maris Paupe <marisp@mt.lv>
---
diff --git a/drivers/net/usb/sierra_net.c b/drivers/net/usb/sierra_net.c
index b59cf20..cf11924 100644
--- a/drivers/net/usb/sierra_net.c
+++ b/drivers/net/usb/sierra_net.c
@@ -779,7 +779,7 @@ static int sierra_net_bind(struct usbnet *dev, struct usb_interface *intf)
  	dev_dbg(&dev->udev->dev, "Fw attr: %x\n", fwattr);
  
  	/* test whether firmware supports DHCP */
-	if (!(status == sizeof(fwattr) && (fwattr & SWI_GET_FW_ATTR_MASK))) {
+	if (!(status == sizeof(fwattr) && (le16_to_cpu(fwattr) & SWI_GET_FW_ATTR_MASK))) {
  		/* found incompatible firmware version */
  		dev_err(&dev->udev->dev, "Incompatible driver and firmware"
  			" versions\n");
---

Re: [PATCH] fix sierra_net endianess bug

[Andreas Schwab]

Maris Paupe <marisp@mt.lv> writes:

> I have found that sierra_net does not work properly on big-endian
> systems, it works if bytes are flipped in this particular location.

That should probably be fixed in sierra_net_get_fw_attr (and the author
of sierra_net_get_fw_attr should be hit with a cluestick to not use
kmalloc for allocating a single u16).

Andreas.

-- 
Andreas Schwab, schwab@linux-m68k.org
GPG Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: [PATCH] fix sierra_net endianess bug

[Ben Hutchings]

On Wed, 2012-03-14 at 16:26 +0100, Andreas Schwab wrote:
> Maris Paupe <marisp@mt.lv> writes:
> 
> > I have found that sierra_net does not work properly on big-endian
> > systems, it works if bytes are flipped in this particular location.
> 
> That should probably be fixed in sierra_net_get_fw_attr (and the author
> of sierra_net_get_fw_attr should be hit with a cluestick to not use
> kmalloc for allocating a single u16).

You cannot use stack buffers for DMA, which includes all USB I/O.

Ben.

-- 
Ben Hutchings, Staff Engineer, Solarflare
Not speaking for my employer; that's the marketing department's job.
They asked us to note that Solarflare product names are trademarked.

Re: [PATCH] fix sierra_net endianess bug

[Andreas Schwab]

Ben Hutchings <bhutchings@solarflare.com> writes:

> On Wed, 2012-03-14 at 16:26 +0100, Andreas Schwab wrote:
>> Maris Paupe <marisp@mt.lv> writes:
>> 
>> > I have found that sierra_net does not work properly on big-endian
>> > systems, it works if bytes are flipped in this particular location.
>> 
>> That should probably be fixed in sierra_net_get_fw_attr (and the author
>> of sierra_net_get_fw_attr should be hit with a cluestick to not use
>> kmalloc for allocating a single u16).
>
> You cannot use stack buffers for DMA, which includes all USB I/O.

Ok, sorry, I didn't know that.

Andreas.

-- 
Andreas Schwab, schwab@linux-m68k.org
GPG Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."