Re: Fw: [Bug 6421] New: kernel 2.6.10-2.6.16 on alpha: arch/alpha/kernel/io.c, iowrite16_rep() BUG_ON((unsigned long)src & 0x1) triggered

[Andrew Morton <akpm@osdl.org>]

Ingo Oeser <netdev@axxeo.de> wrote:
>
> Stephen Hemminger wrote:
> > Looks like PIO at unaligned addresses doesn't work on alpha...
> 
> Maybe this should be fixed similiar to ioread32_rep in  arch/alpha/kernel/io.c?
> 

I think so, but Ivan thinks networking is bust:


Ivan Kokshaysky <ink@jurassic.park.msu.ru> wrote:
>
> On Fri, Apr 21, 2006 at 04:28:30AM -0700, Andrew Morton wrote:
> > Why is it "silently corrupted"?  It's just misaligned, isn't it?  Networking
> > does that sometimes.
> 
> Because networking does read/write "short" fields in various packet
> header structures. Results are illustrated in a following example:
> 
> char foo[] __attribute__((aligned(8))) = "0123456701234567";
> 
> int main()
> {
> 	short *bar = (short *)&foo[7];
> 	printf("%04x\n", *bar); /* 3037 */
> 	*bar = 0x4241; /* "AB" */
> 	printf("%s\n", foo);
> 	return 0;
> }
> --------
> 0037
> ^^
> 0123456A01234567
>         ^
> Misalignment by two bytes for ints and longs is often unavoidable in
> networking and we can cope with it, but there is no excuse of 1-byte
> misalignment.