Can I/OAT DMA engineer access PCI MMIO space

Can I/OAT DMA engineer access PCI MMIO space

[康剑斌]

I try to use ioatdam to copy data from system memory to pci MMIO space:

=======================
void mmio_copy(void *dst, void *src, size_t len)
struct async_submit_ctl submit;
struct memcpy_context ctx;

atomic_set(&ctx.cnt, 1);
init_completion(&ctx.cmp);
init_async_submit(&submit, ASYNC_TX_ACK, NULL,
done_memcpy, &ctx, NULL);

while (len) {
size_t clen = (len > PAGE_SIZE) ? PAGE_SIZE : len;
struct page *p_dst, *p_src;

p_dst = virt_to_page(dst);
p_src = virt_to_page(src);

atomic_inc(&ctx.cnt);

ntb_async_memcpy(p_dst, p_src, 0, 0, clen, &submit);

dst += clen;
src += clen;
len -= clen;
}

if (atomic_dec_and_test(&ctx.cnt))
return;

async_tx_issue_pending_all();
wait_for_completion(&ctx.cmp);
=================

If dst points to a memory space, the operation would pass.
But if dst points to a pci MMIO space, it failed with kernel oops.
It seems the code:
BUG_ON(is_ioat_bug(chanerr));
in drivers/dma/ioat/dma_v3.c, line 365 cause the oops.
Is there anyway to access pci MMIO space using ioat?
The datasheet says that ioat supports MMIO access.

Re: Can I/OAT DMA engineer access PCI MMIO space

[Koul, Vinod]

On Fri, 2011-04-29 at 18:13 +0800, 康剑斌 wrote:
> I try to use ioatdam to copy data from system memory to pci MMIO space:

> If dst points to a memory space, the operation would pass.
> But if dst points to a pci MMIO space, it failed with kernel oops.
> It seems the code:
> BUG_ON(is_ioat_bug(chanerr));
> in drivers/dma/ioat/dma_v3.c, line 365 cause the oops.
> Is there anyway to access pci MMIO space using ioat?
> The datasheet says that ioat supports MMIO access.
Did you map the IO memory in kernel using ioremap and friends first?

-- 
~Vinod

Re: Can I/OAT DMA engineer access PCI MMIO space

[康剑斌]

>> I try to use ioatdam to copy data from system memory to pci MMIO space:
>> If dst points to a memory space, the operation would pass.
>> But if dst points to a pci MMIO space, it failed with kernel oops.
>> It seems the code:
>> BUG_ON(is_ioat_bug(chanerr));
>> in drivers/dma/ioat/dma_v3.c, line 365 cause the oops.
>> Is there anyway to access pci MMIO space using ioat?
>> The datasheet says that ioat supports MMIO access.
> Did you map the IO memory in kernel using ioremap and friends first?
>
yes, I had used 'ioremap_nocache' to map the IO memory and I can use
memcpy to copy data to this region. The async_tx should have been 
correctly configured as
I can use aync_memcpy to copy data between different system memory address.

Re: Can I/OAT DMA engineer access PCI MMIO space

[Koul, Vinod]

On Tue, 2011-05-03 at 10:21 +0800, 康剑斌 wrote:
> >> I try to use ioatdam to copy data from system memory to pci MMIO space:
> >> If dst points to a memory space, the operation would pass.
> >> But if dst points to a pci MMIO space, it failed with kernel oops.
> >> It seems the code:
> >> BUG_ON(is_ioat_bug(chanerr));
> >> in drivers/dma/ioat/dma_v3.c, line 365 cause the oops.
> >> Is there anyway to access pci MMIO space using ioat?
> >> The datasheet says that ioat supports MMIO access.
> > Did you map the IO memory in kernel using ioremap and friends first?
> >
> yes, I had used 'ioremap_nocache' to map the IO memory and I can use
> memcpy to copy data to this region. The async_tx should have been 
> correctly configured as
> I can use aync_memcpy to copy data between different system memory address.
Then you should be using memcpy_toio() and friends

~Vinod

-- 
~Vinod

Re: Can I/OAT DMA engineer access PCI MMIO space

[康剑斌]

>> yes, I had used 'ioremap_nocache' to map the IO memory and I can use
>> memcpy to copy data to this region. The async_tx should have been
>> correctly configured as
>> I can use aync_memcpy to copy data between different system memory address.
> Then you should be using memcpy_toio() and friends
>
Do you mean that if I have mapped the mmio, I can' use I/OAT dma 
transfer to this region any more?
I can use memcpy to copy data, but it consumes lots of cpu as PCI access 
is too slow.
If I can use i/oat dma and asyc_tx api to do the job, the performance 
should be imporved.
Thanks

Re: Can I/OAT DMA engineer access PCI MMIO space

[Dan Williams]

On 5/2/2011 11:31 PM, 康剑斌 wrote:
>
>>> yes, I had used 'ioremap_nocache' to map the IO memory and I can use
>>> memcpy to copy data to this region. The async_tx should have been
>>> correctly configured as
>>> I can use aync_memcpy to copy data between different system memory address.
>> Then you should be using memcpy_toio() and friends
>>
> Do you mean that if I have mapped the mmio, I can' use I/OAT dma
> transfer to this region any more?
> I can use memcpy to copy data, but it consumes lots of cpu as PCI access
> is too slow.
> If I can use i/oat dma and asyc_tx api to do the job, the performance
> should be imporved.
> Thanks

The async_tx api only supports memory-to-memory transfers.  To write to 
mmio space with ioatdma you would need a custom method, like the 
dma-slave support in other drivers, to program the descriptors with the 
physical mmio bus address.

--
Dan

Re: Can I/OAT DMA engineer access PCI MMIO space

[康剑斌]

于 2011年05月03日 23:58, Dan Williams 写道:
>
>> Do you mean that if I have mapped the mmio, I can' use I/OAT dma
>> transfer to this region any more?
>> I can use memcpy to copy data, but it consumes lots of cpu as PCI access
>> is too slow.
>> If I can use i/oat dma and asyc_tx api to do the job, the performance
>> should be imporved.
>> Thanks
>
>
> The async_tx api only supports memory-to-memory transfers.  To write 
> to mmio space with ioatdma you would need a custom method, like the 
> dma-slave support in other drivers, to program the descriptors with 
> the physical mmio bus address.
>
> -- 
> Dan
Thanks.
I directly read pci bar address and program it into descriptors, ioatdma 
works.
Some problem is, when PCI transfer failed (Using a NTB connect to 
another system, and the system power down),
ioatdma will cause kernel oops.

BUG_ON(is_ioat_bug(chanerr));
in drivers/dma/ioat/dma_v3.c, line 365

It seems that HW reports a 'IOAT_CHANERR_DEST_ADDR_ERR', and drivers 
can't recover from this situation.
What does dma-slave mean? Just like DMA_SLAVE flag existing in other DMA 
drivers?

Re: Can I/OAT DMA engineer access PCI MMIO space

[Dan Williams]

[ adding Dave ]

On 5/5/2011 1:45 AM, 康剑斌 wrote:
> Thanks.
> I directly read pci bar address and program it into descriptors, ioatdma
> works.
> Some problem is, when PCI transfer failed (Using a NTB connect to
> another system, and the system power down),
> ioatdma will cause kernel oops.
>
> BUG_ON(is_ioat_bug(chanerr));
> in drivers/dma/ioat/dma_v3.c, line 365
>
> It seems that HW reports a 'IOAT_CHANERR_DEST_ADDR_ERR', and drivers
> can't recover from this situation.

Ah ok, this is expected with the current upstream ioatdma driver.  The 
driver assumes that all transfers are mem-to-mem (ASYNC_TX_DMA or 
NET_DMA) and that a destination address error is a fatal error (similar 
to a kernel page fault).

With NTB, where failures are expected, the driver would need to be 
modified to expect the error, recover from it, and report it to the 
application.

> What does dma-slave mean? Just like DMA_SLAVE flag existing in other DMA
> drivers?

Yes, DMA_SLAVE is the generic framework to associate a dma offload 
device with an mmio peripheral.

--
Dan

Re: Can I/OAT DMA engineer access PCI MMIO space

[Dan Williams]

On 5/2/2011 7:21 PM, 康剑斌 wrote:
>
>>> I try to use ioatdam to copy data from system memory to pci MMIO space:
>>> If dst points to a memory space, the operation would pass.
>>> But if dst points to a pci MMIO space, it failed with kernel oops.
>>> It seems the code:
>>> BUG_ON(is_ioat_bug(chanerr));
>>> in drivers/dma/ioat/dma_v3.c, line 365 cause the oops.
>>> Is there anyway to access pci MMIO space using ioat?
>>> The datasheet says that ioat supports MMIO access.
>> Did you map the IO memory in kernel using ioremap and friends first?
>>
> yes, I had used 'ioremap_nocache' to map the IO memory and I can use
> memcpy to copy data to this region. The async_tx should have been
> correctly configured as
> I can use aync_memcpy to copy data between different system memory address.
>

ioremap maps mmio space for the cpu not bus mastering devices.  You need 
to program the physical address into the descriptor.

--
Dan