RE: [PATCH 5/5] I/OAT: Add completion callback for async_tx interface use
>From: David Miller [mailto:[EMAIL PROTECTED] > >From: Andrew Morton <[EMAIL PROTECTED]> >Date: Wed, 17 Oct 2007 17:44:38 -0700 > >> On Wed, 17 Oct 2007 17:14:39 -0700 >> Shannon Nelson <[EMAIL PROTECTED]> wrote: >> >> > + tx->callback = (void *)ioat_dma_test_callback; >> >> and when I remove this cast I get >> >> drivers/dma/ioat_dma.c: In function 'ioat_dma_self_test': >> drivers/dma/ioat_dma.c:718: warning: assignment from >incompatible pointer type >> >> because ioat_dma_test_callback isn't void-returning. >Something is wrong >> here. I assume that ioat_dma_test_callback() should just be >> void-returning? > >In fact this might crash on systems that conditionally pop >return values based upon whether the function is void or not. > Yep - thanks. That's my misuse of Dan's dma_async_tx_callback typedef. sln - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: [PATCH 5/5] I/OAT: Add completion callback for async_tx interface use
From: Andrew Morton <[EMAIL PROTECTED]> Date: Wed, 17 Oct 2007 17:44:38 -0700 > On Wed, 17 Oct 2007 17:14:39 -0700 > Shannon Nelson <[EMAIL PROTECTED]> wrote: > > > + tx->callback = (void *)ioat_dma_test_callback; > > and when I remove this cast I get > > drivers/dma/ioat_dma.c: In function 'ioat_dma_self_test': > drivers/dma/ioat_dma.c:718: warning: assignment from incompatible pointer type > > because ioat_dma_test_callback isn't void-returning. Something is wrong > here. I assume that ioat_dma_test_callback() should just be > void-returning? In fact this might crash on systems that conditionally pop return values based upon whether the function is void or not. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
RE: [PATCH 5/5] I/OAT: Add completion callback for async_tx interface use
>From: Andrew Morton [mailto:[EMAIL PROTECTED]
>
>On Wed, 17 Oct 2007 17:14:39 -0700
>Shannon Nelson <[EMAIL PROTECTED]> wrote:
[...]
>> +static dma_async_tx_callback ioat_dma_test_callback(void
>*dma_async_param)
>> +{
>> +printk(KERN_ERR "ioatdma: ioat_dma_test_callback(0x%04llx)\n",
>> +(u64)dma_async_param);
>> +return 0;
>> +}
>
>This wanted to be `return NULL'. I'll fix.
Thanks.
>
>> /**
>> * ioat_dma_self_test - Perform a IOAT transaction to
>verify the HW works.
>> * @device: device to be tested
>> @@ -691,6 +715,8 @@ static int ioat_dma_self_test(struct
>ioatdma_device *device)
>> addr = dma_map_single(dma_chan->device->dev, dest,
>IOAT_TEST_SIZE,
>> DMA_FROM_DEVICE);
>> ioat_set_dest(addr, tx, 0);
>> +tx->callback = (void *)ioat_dma_test_callback;
>
>This cast is unneeded, surely? It had better be..
>
>> +tx->callback_param = (void *)0x8086;
>
>eh?
I suppose I could have used "42"...
sln
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Re: [PATCH 5/5] I/OAT: Add completion callback for async_tx interface use
On Wed, 17 Oct 2007 17:14:39 -0700
Shannon Nelson <[EMAIL PROTECTED]> wrote:
> + printk(KERN_ERR "ioatdma: ioat_dma_test_callback(0x%04llx)\n",
> + (u64)dma_async_param);
Generates a warning on 32-bit:
drivers/dma/ioat_dma.c: In function 'ioat_dma_test_callback':
drivers/dma/ioat_dma.c:660: warning: cast from pointer to integer of different
size
and we can never ever print u64's anwyay: we dont' know what type the arch
uses to implement them. The usual fix is to cast to unsigned long long,
but in this case %p will work nicely, no?
> + tx->callback = (void *)ioat_dma_test_callback;
and when I remove this cast I get
drivers/dma/ioat_dma.c: In function 'ioat_dma_self_test':
drivers/dma/ioat_dma.c:718: warning: assignment from incompatible pointer type
because ioat_dma_test_callback isn't void-returning. Something is wrong
here. I assume that ioat_dma_test_callback() should just be
void-returning?
diff -puN
drivers/dma/ioat_dma.c~i-oat-add-completion-callback-for-async_tx-interface-use-fix
drivers/dma/ioat_dma.c
---
a/drivers/dma/ioat_dma.c~i-oat-add-completion-callback-for-async_tx-interface-use-fix
+++ a/drivers/dma/ioat_dma.c
@@ -654,11 +654,10 @@ static void ioat_dma_start_null_desc(str
*/
#define IOAT_TEST_SIZE 2000
-static dma_async_tx_callback ioat_dma_test_callback(void *dma_async_param)
+static void ioat_dma_test_callback(void *dma_async_param)
{
- printk(KERN_ERR "ioatdma: ioat_dma_test_callback(0x%04llx)\n",
- (u64)dma_async_param);
- return 0;
+ printk(KERN_ERR "ioatdma: ioat_dma_test_callback(0x%p)\n",
+ dma_async_param);
}
/**
@@ -715,7 +714,7 @@ static int ioat_dma_self_test(struct ioa
addr = dma_map_single(dma_chan->device->dev, dest, IOAT_TEST_SIZE,
DMA_FROM_DEVICE);
ioat_set_dest(addr, tx, 0);
- tx->callback = (void *)ioat_dma_test_callback;
+ tx->callback = ioat_dma_test_callback;
tx->callback_param = (void *)0x8086;
cookie = ioat_tx_submit(tx);
if (cookie < 0) {
_
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Re: [PATCH 5/5] I/OAT: Add completion callback for async_tx interface use
On Wed, 17 Oct 2007 17:14:39 -0700
Shannon Nelson <[EMAIL PROTECTED]> wrote:
> The async_tx interface includes a completion callback. This adds support
> for using that callback, including using interrupts on completion.
>
> Signed-off-by: Shannon Nelson <[EMAIL PROTECTED]>
> ---
>
> drivers/dma/ioat_dma.c | 26 ++
> 1 files changed, 26 insertions(+), 0 deletions(-)
>
> diff --git a/drivers/dma/ioat_dma.c b/drivers/dma/ioat_dma.c
> index 117ac38..f560527 100644
> --- a/drivers/dma/ioat_dma.c
> +++ b/drivers/dma/ioat_dma.c
> @@ -212,6 +212,18 @@ static dma_cookie_t ioat_tx_submit(struct
> dma_async_tx_descriptor *tx)
> } while (len && (new = ioat_dma_get_next_descriptor(ioat_chan)));
>
> hw->ctl = IOAT_DMA_DESCRIPTOR_CTL_CP_STS;
> + if (new->async_tx.callback) {
> + hw->ctl |= IOAT_DMA_DESCRIPTOR_CTL_INT_GN;
> + if (first != new) {
> + /* move callback into to last desc */
> + new->async_tx.callback = first->async_tx.callback;
> + new->async_tx.callback_param
> + = first->async_tx.callback_param;
> + first->async_tx.callback = NULL;
> + first->async_tx.callback_param = NULL;
> + }
> + }
> +
> new->tx_cnt = desc_count;
> new->async_tx.ack = orig_ack; /* client is in control of this ack */
>
> @@ -516,6 +528,11 @@ static void ioat_dma_memcpy_cleanup(struct ioat_dma_chan
> *ioat_chan)
> pci_unmap_addr(desc, src),
> pci_unmap_len(desc, len),
> PCI_DMA_TODEVICE);
> + if (desc->async_tx.callback) {
> + desc->async_tx.callback(
> + desc->async_tx.callback_param);
> + desc->async_tx.callback = NULL;
> + }
> }
>
> if (desc->async_tx.phys != phys_complete) {
> @@ -637,6 +654,13 @@ static void ioat_dma_start_null_desc(struct
> ioat_dma_chan *ioat_chan)
> */
> #define IOAT_TEST_SIZE 2000
>
> +static dma_async_tx_callback ioat_dma_test_callback(void *dma_async_param)
> +{
> + printk(KERN_ERR "ioatdma: ioat_dma_test_callback(0x%04llx)\n",
> + (u64)dma_async_param);
> + return 0;
> +}
This wanted to be `return NULL'. I'll fix.
> /**
> * ioat_dma_self_test - Perform a IOAT transaction to verify the HW works.
> * @device: device to be tested
> @@ -691,6 +715,8 @@ static int ioat_dma_self_test(struct ioatdma_device
> *device)
> addr = dma_map_single(dma_chan->device->dev, dest, IOAT_TEST_SIZE,
> DMA_FROM_DEVICE);
> ioat_set_dest(addr, tx, 0);
> + tx->callback = (void *)ioat_dma_test_callback;
This cast is unneeded, surely? It had better be..
> + tx->callback_param = (void *)0x8086;
eh?
> cookie = ioat_tx_submit(tx);
> if (cookie < 0) {
> dev_err(&device->pdev->dev,
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
