2009/1/29 Adrian McMenamin <adrian@xxxxxxxxxxxxxxxxxxxxxxxx>:
> On Thu, January 29, 2009 4:43 pm, Adrian McMenamin wrote:
>> On Thu, January 29, 2009 3:13 am, Paul Mundt wrote:
>>>>
>>> Please do not abuse P2SEGADDR in this fashion. No new code should be
>>> using P2SEGADDR anyways. maple is particularly abusive in this case
>>> since
>>> it bounces back and forth between P2SEGADDR and PHYSADDR to try and
>>> ignore the fact cache flushing has to be handled.
>>>
>>> You may also wish to consider ioremap/ioremap_nocache().
>>>
>>
>> Can I just use dma_cache_sync before each dma run and then use the P1
>> address space?
>>
>
> After the dma run I mean - the cache is flushed before the run - but not
> after.
>
So I have done this:
Removing P2SEGADDR call and flushing caches
---
diff --git a/drivers/sh/maple/maple.c b/drivers/sh/maple/maple.c
index 4d9b8b3..5dca308 100644
--- a/drivers/sh/maple/maple.c
+++ b/drivers/sh/maple/maple.c
@@ -186,7 +186,7 @@ static struct mapleq *maple_allocq(struct
maple_device *mdev)
mq->recvbuf = kmem_cache_zalloc(maple_queue_cache, GFP_KERNEL);
if (!mq->recvbuf)
goto failed_p2;
- mq->recvbuf->buf = (void *)P2SEGADDR(&((mq->recvbuf->bufx)[0]));
+ mq->recvbuf->buf = &((mq->recvbuf->bufx)[0]);
return mq;
@@ -639,9 +639,11 @@ static void maple_dma_handler(struct work_struct *work)
ctrl_outl(0, MAPLE_ENABLE);
if (!list_empty(&maple_sentq)) {
list_for_each_entry_safe(mq, nmq, &maple_sentq, list) {
+ mdev = mq->dev;
recvbuf = mq->recvbuf->buf;
+ dma_cache_sync(&mdev->dev, recvbuf, 0x400,
+ DMA_FROM_DEVICE);
code = recvbuf[0];
- mdev = mq->dev;
kfree(mq->sendbuf);
list_del_init(&mq->list);
switch (code) {
Seems to work...
--
To unsubscribe from this list: send the line "unsubscribe linux-input" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
