RE: [microblaze-uclinux] xenet_FifoSend(struct sk_buff *orig_skb,structnet_device*dev)

["Brettschneider Falk" <fbrettschneider@xxxxxxxxxxxxxxx>]

Hi Jim,
thanks very much for your help.

Today I found out my patch skbuff.c.diff (see http://www.itee.uq.edu.au/~listarch/microblaze-uclinux/archive/2008/04/msg00083.html) even makes it unnecessary to memmove() in drivers/net/xilinx_emac/adapter.c:FifoRecvHandler()! :-)
Reason: Since I move the start address of skb->data 2 bytes already in __alloc_skb I don't need to realign the data by memmove anymore.

And if I want to do data moving with your move_halfword_higher() regardless of my patch, it violates your DWORD-alignment requirement of the src pointer. This was the reason why move_halfword_higher() doesn't work for me. But anyway, as found out today, with the patch we don't need it anymore.

I still have to measure and compare the times "with skbuff.c.diff but without move_halfword_higher()" against "without skbuff.c.diff but with move_halfword_higer()".

Cheers, Falk

___________________________
microblaze-uclinux mailing list
microblaze-uclinux@xxxxxxxxxxxxxx
Project Home Page : http://www.itee.uq.edu.au/~jwilliams/mblaze-uclinux
Mailing List Archive : http://www.itee.uq.edu.au/~listarch/microblaze-uclinux/