On 21.03.2013 20:10, Arnd Bergmann wrote:
On Thursday 21 March 2013, Jason Cooper wrote:
On Thu, Mar 21, 2013 at 06:04:59PM +0100, Soeren Moch wrote:
Now I found out what is going on here:
In itd_urb_transaction() we allocate 9 iTDs for each URB with
number_of_packets == 64 in my case. The iTDs are added to
sched->td_list. For a frame-aligned scheduling we need 8 iTDs, the
9th one is released back to the front of the streams free_list in
iso_sched_free(). This iTD was cleared after allocation and has a
frame number of 0 now. So for each allocation when now_frame == 0 we
allocate from the dma_pool, not from the free_list. The attached
patch invalidates the frame number in each iTD before it is sent to
the scheduler. This fixes the problem without the need to iterate
over a iTD list.
Signed-off-by: Soeren Moch <smoch@xxxxxx>
Wow! Great work Soeren! Talk about a long road to a small fix. Thanks
for keeping after it.
+1
I hardly understand half of the description above, but that much sounds
plausible. Is this a bug fix that should get backported to stable kernels?
Arnd
I think this patch should go to stable kernels, too. But I'm not an
expert here...
Soeren
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>