Steven Rostedt wrote:

On Fri, 16 Nov 2007, Nelson, Shannon wrote:

quoted

first->async_tx.phys;

quoted

-       __list_splice(&new_chain, ioat_chan->used_desc.prev);
+       list_splice_tail(&new_chain, ioat_chan->used_desc.prev);

NAK.

These functions do insertions differently.  The 'prev' is pointing to
the last valid descriptor in the queue and you really want to get the
new_chain stuck on after this.  Your list_splice_tail() will insert the
new_chain just before it which will muck up the order of the DMA
requests.

You might have more success with
	list_splice_tail(&new_chain, ioat_chan->used_desc);
where used_desc points to the whole list, rather than using the .prev
pointer to a specific node.

Please copy me on future ioatdma related comments.

And people wonder why we post RT related patches to LKML. This is exactly
why!

Thanks for the response Shannon!

-- Steve

Hi

I've tried this change, but it still don't compile. So I propose this patch.

regards,
trem

------------------------

Index: linux-2.6.23/drivers/dma/ioat_dma.c
===================================================================

--- linux-2.6.23.orig/drivers/dma/ioat_dma.c
+++ linux-2.6.23/drivers/dma/ioat_dma.c

@@ -244,7 +244,7 @@ static dma_cookie_t ioat_tx_submit(struc
        /* write address into NextDescriptor field of last desc in chain */
        to_ioat_desc(ioat_chan->used_desc.prev)->hw->next =
                                                        first->async_tx.phys;
-       __list_splice(&new_chain, ioat_chan->used_desc.prev);
+       list_splice_tail(&new_chain, &ioat_chan->used_desc);
 
        ioat_chan->pending += desc_count;
        if (ioat_chan->pending >= 4) {