Re: [v6,2/8] dmaengine: Add DMA_MEMCPY_SG transaction op

2017-11-15 Thread Dave Jiang


On 11/15/2017 09:13 AM, Stefan Roese wrote:
> Hi Vinod,
> 
> On 15.11.2017 16:52, Vinod Koul wrote:
>> On Mon, Nov 13, 2017 at 09:28:46AM +0100, Stefan Roese wrote:
>>> Hi Vinod,
>>>
>>> On 31.08.2017 12:57, Robin Murphy wrote:
 On 30/08/17 19:25, Dave Jiang wrote:
> On 08/30/2017 11:18 AM, Robin Murphy wrote:
>> On 25/08/17 21:59, Dave Jiang wrote:
>>> Adding a dmaengine transaction operation that allows copy to/from a
>>> scatterlist and a flat buffer.
>>
>> Apologies if I'm late to the party, but doesn't DMA_SG already cover
>> this use-case? As far as I can see, all this does is save the caller
> >from setting up a single-entry scatterlist to describe the buffer - even
>> if such a simplified interface is justified it seems like something that
>> could be implemented as a wrapper around dmaengine_prep_dma_sg() rather
>> than the providers having to implement a whole extra callback.
>>
>
> DMA_SG is queued to be removed in 4.14. There is no in kernel consumer
> for the code.

 Ah, I see, that's what I was missing. So we're effectively just
 replacing that interface with a more pragmatic alternative - that makes
 sense.
>>>
>>> What are the plans with this new DMA_MEMCPY_SG interface? When will it
>>> hit mainline or is something missing?
>>
>> The old one was removed in 4.14 so if you have a usage feel free to send a
>> patch to add this with usage.
> 
> No, its not the "old one" (DMA_SG) but the "new one" (DMA_MEMCPY_SG)
> I'm referring to (this email thread). My impression was, that this
> new interface has (or will get) in-kernel users and will be pulled at
> some time.
> 

Decided to hold off on the submission for now. If you need it and have
an upstream consumer for it, feel free to push the change.
___
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm


Re: [v6,2/8] dmaengine: Add DMA_MEMCPY_SG transaction op

2017-11-15 Thread Stefan Roese

Hi Vinod,

On 15.11.2017 16:52, Vinod Koul wrote:

On Mon, Nov 13, 2017 at 09:28:46AM +0100, Stefan Roese wrote:

Hi Vinod,

On 31.08.2017 12:57, Robin Murphy wrote:

On 30/08/17 19:25, Dave Jiang wrote:

On 08/30/2017 11:18 AM, Robin Murphy wrote:

On 25/08/17 21:59, Dave Jiang wrote:

Adding a dmaengine transaction operation that allows copy to/from a
scatterlist and a flat buffer.


Apologies if I'm late to the party, but doesn't DMA_SG already cover
this use-case? As far as I can see, all this does is save the caller

>from setting up a single-entry scatterlist to describe the buffer - even

if such a simplified interface is justified it seems like something that
could be implemented as a wrapper around dmaengine_prep_dma_sg() rather
than the providers having to implement a whole extra callback.



DMA_SG is queued to be removed in 4.14. There is no in kernel consumer
for the code.


Ah, I see, that's what I was missing. So we're effectively just
replacing that interface with a more pragmatic alternative - that makes
sense.


What are the plans with this new DMA_MEMCPY_SG interface? When will it
hit mainline or is something missing?


The old one was removed in 4.14 so if you have a usage feel free to send a
patch to add this with usage.


No, its not the "old one" (DMA_SG) but the "new one" (DMA_MEMCPY_SG)
I'm referring to (this email thread). My impression was, that this
new interface has (or will get) in-kernel users and will be pulled at
some time.

Thanks,
Stefan
___
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm


Re: [v6,2/8] dmaengine: Add DMA_MEMCPY_SG transaction op

2017-11-15 Thread Vinod Koul
On Mon, Nov 13, 2017 at 09:28:46AM +0100, Stefan Roese wrote:
> Hi Vinod,
> 
> On 31.08.2017 12:57, Robin Murphy wrote:
> >On 30/08/17 19:25, Dave Jiang wrote:
> >>On 08/30/2017 11:18 AM, Robin Murphy wrote:
> >>>On 25/08/17 21:59, Dave Jiang wrote:
> Adding a dmaengine transaction operation that allows copy to/from a
> scatterlist and a flat buffer.
> >>>
> >>>Apologies if I'm late to the party, but doesn't DMA_SG already cover
> >>>this use-case? As far as I can see, all this does is save the caller
> >>>from setting up a single-entry scatterlist to describe the buffer - even
> >>>if such a simplified interface is justified it seems like something that
> >>>could be implemented as a wrapper around dmaengine_prep_dma_sg() rather
> >>>than the providers having to implement a whole extra callback.
> >>>
> >>
> >>DMA_SG is queued to be removed in 4.14. There is no in kernel consumer
> >>for the code.
> >
> >Ah, I see, that's what I was missing. So we're effectively just
> >replacing that interface with a more pragmatic alternative - that makes
> >sense.
> 
> What are the plans with this new DMA_MEMCPY_SG interface? When will it
> hit mainline or is something missing?

The old one was removed in 4.14 so if you have a usage feel free to send a
patch to add this with usage.

Thanks
-- 
~Vinod
___
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm


Re: [v6,2/8] dmaengine: Add DMA_MEMCPY_SG transaction op

2017-11-13 Thread Stefan Roese

Hi Vinod,

On 31.08.2017 12:57, Robin Murphy wrote:

On 30/08/17 19:25, Dave Jiang wrote:

On 08/30/2017 11:18 AM, Robin Murphy wrote:

On 25/08/17 21:59, Dave Jiang wrote:

Adding a dmaengine transaction operation that allows copy to/from a
scatterlist and a flat buffer.


Apologies if I'm late to the party, but doesn't DMA_SG already cover
this use-case? As far as I can see, all this does is save the caller
from setting up a single-entry scatterlist to describe the buffer - even
if such a simplified interface is justified it seems like something that
could be implemented as a wrapper around dmaengine_prep_dma_sg() rather
than the providers having to implement a whole extra callback.



DMA_SG is queued to be removed in 4.14. There is no in kernel consumer
for the code.


Ah, I see, that's what I was missing. So we're effectively just
replacing that interface with a more pragmatic alternative - that makes
sense.


What are the plans with this new DMA_MEMCPY_SG interface? When will it
hit mainline or is something missing?

Thanks,
Stefan
___
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm


Re: [v6,2/8] dmaengine: Add DMA_MEMCPY_SG transaction op

2017-08-31 Thread Robin Murphy
On 30/08/17 19:25, Dave Jiang wrote:
> On 08/30/2017 11:18 AM, Robin Murphy wrote:
>> On 25/08/17 21:59, Dave Jiang wrote:
>>> Adding a dmaengine transaction operation that allows copy to/from a
>>> scatterlist and a flat buffer.
>>
>> Apologies if I'm late to the party, but doesn't DMA_SG already cover
>> this use-case? As far as I can see, all this does is save the caller
>> from setting up a single-entry scatterlist to describe the buffer - even
>> if such a simplified interface is justified it seems like something that
>> could be implemented as a wrapper around dmaengine_prep_dma_sg() rather
>> than the providers having to implement a whole extra callback.
>>
> 
> DMA_SG is queued to be removed in 4.14. There is no in kernel consumer
> for the code.

Ah, I see, that's what I was missing. So we're effectively just
replacing that interface with a more pragmatic alternative - that makes
sense.

Thanks,
Robin.

>>>
>>> Signed-off-by: Dave Jiang 
>>> ---
>>>  Documentation/dmaengine/provider.txt |3 +++
>>>  drivers/dma/dmaengine.c  |2 ++
>>>  include/linux/dmaengine.h|   19 +++
>>>  3 files changed, 24 insertions(+)
>>>
>>> diff --git a/Documentation/dmaengine/provider.txt 
>>> b/Documentation/dmaengine/provider.txt
>>> index a75f52f..6241e36 100644
>>> --- a/Documentation/dmaengine/provider.txt
>>> +++ b/Documentation/dmaengine/provider.txt
>>> @@ -181,6 +181,9 @@ Currently, the types available are:
>>>  - Used by the client drivers to register a callback that will be
>>>called on a regular basis through the DMA controller interrupt
>>>  
>>> +  * DMA_MEMCPY_SG
>>> +- The device supports scatterlist to/from memory.
>>> +
>>>* DMA_PRIVATE
>>>  - The devices only supports slave transfers, and as such isn't
>>>available for async transfers.
>>> diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
>>> index 428b141..4d2c4e1 100644
>>> --- a/drivers/dma/dmaengine.c
>>> +++ b/drivers/dma/dmaengine.c
>>> @@ -937,6 +937,8 @@ int dma_async_device_register(struct dma_device *device)
>>> !device->device_prep_dma_memset);
>>> BUG_ON(dma_has_cap(DMA_INTERRUPT, device->cap_mask) &&
>>> !device->device_prep_dma_interrupt);
>>> +   BUG_ON(dma_has_cap(DMA_MEMCPY_SG, device->cap_mask) &&
>>> +   !device->device_prep_dma_memcpy_sg);
>>> BUG_ON(dma_has_cap(DMA_CYCLIC, device->cap_mask) &&
>>> !device->device_prep_dma_cyclic);
>>> BUG_ON(dma_has_cap(DMA_INTERLEAVE, device->cap_mask) &&
>>> diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
>>> index 64fbd38..0c91411 100644
>>> --- a/include/linux/dmaengine.h
>>> +++ b/include/linux/dmaengine.h
>>> @@ -67,6 +67,7 @@ enum dma_transaction_type {
>>> DMA_PQ_VAL,
>>> DMA_MEMSET,
>>> DMA_MEMSET_SG,
>>> +   DMA_MEMCPY_SG,
>>> DMA_INTERRUPT,
>>> DMA_PRIVATE,
>>> DMA_ASYNC_TX,
>>> @@ -692,6 +693,7 @@ struct dma_filter {
>>>   * @device_prep_dma_pq_val: prepares a pqzero_sum operation
>>>   * @device_prep_dma_memset: prepares a memset operation
>>>   * @device_prep_dma_memset_sg: prepares a memset operation over a scatter 
>>> list
>>> + * @device_prep_dma_memcpy_sg: prepares memcpy between scatterlist and 
>>> buffer
>>>   * @device_prep_dma_interrupt: prepares an end of chain interrupt operation
>>>   * @device_prep_slave_sg: prepares a slave dma operation
>>>   * @device_prep_dma_cyclic: prepare a cyclic dma operation suitable for 
>>> audio.
>>> @@ -768,6 +770,10 @@ struct dma_device {
>>> struct dma_async_tx_descriptor *(*device_prep_dma_memset_sg)(
>>> struct dma_chan *chan, struct scatterlist *sg,
>>> unsigned int nents, int value, unsigned long flags);
>>> +   struct dma_async_tx_descriptor *(*device_prep_dma_memcpy_sg)(
>>> +   struct dma_chan *chan,
>>> +   struct scatterlist *sg, unsigned int sg_nents,
>>> +   dma_addr_t buf, bool to_sg, unsigned long flags);
>>> struct dma_async_tx_descriptor *(*device_prep_dma_interrupt)(
>>> struct dma_chan *chan, unsigned long flags);
>>>  
>>> @@ -899,6 +905,19 @@ static inline struct dma_async_tx_descriptor 
>>> *dmaengine_prep_dma_memcpy(
>>> len, flags);
>>>  }
>>>  
>>> +static inline struct dma_async_tx_descriptor *dmaengine_prep_dma_memcpy_sg(
>>> +   struct dma_chan *chan, struct scatterlist *sg,
>>> +   unsigned int sg_nents, dma_addr_t buf, bool to_sg,
>>> +   unsigned long flags)
>>> +{
>>> +   if (!chan || !chan->device ||
>>> +   !chan->device->device_prep_dma_memcpy_sg)
>>> +   return NULL;
>>> +
>>> +   return chan->device->device_prep_dma_memcpy_sg(chan, sg, sg_nents,
>>> +  buf, to_sg, flags);
>>> +}
>>> +
>>>  /**
>>>   * dmaengine_terminate_all() - Terminate all active DMA transfers
>>>   * @chan: The 

Re: [v6,2/8] dmaengine: Add DMA_MEMCPY_SG transaction op

2017-08-30 Thread Dave Jiang
On 08/30/2017 11:18 AM, Robin Murphy wrote:
> On 25/08/17 21:59, Dave Jiang wrote:
>> Adding a dmaengine transaction operation that allows copy to/from a
>> scatterlist and a flat buffer.
> 
> Apologies if I'm late to the party, but doesn't DMA_SG already cover
> this use-case? As far as I can see, all this does is save the caller
> from setting up a single-entry scatterlist to describe the buffer - even
> if such a simplified interface is justified it seems like something that
> could be implemented as a wrapper around dmaengine_prep_dma_sg() rather
> than the providers having to implement a whole extra callback.
> 

DMA_SG is queued to be removed in 4.14. There is no in kernel consumer
for the code.

> Robin.
> 
>>
>> Signed-off-by: Dave Jiang 
>> ---
>>  Documentation/dmaengine/provider.txt |3 +++
>>  drivers/dma/dmaengine.c  |2 ++
>>  include/linux/dmaengine.h|   19 +++
>>  3 files changed, 24 insertions(+)
>>
>> diff --git a/Documentation/dmaengine/provider.txt 
>> b/Documentation/dmaengine/provider.txt
>> index a75f52f..6241e36 100644
>> --- a/Documentation/dmaengine/provider.txt
>> +++ b/Documentation/dmaengine/provider.txt
>> @@ -181,6 +181,9 @@ Currently, the types available are:
>>  - Used by the client drivers to register a callback that will be
>>called on a regular basis through the DMA controller interrupt
>>  
>> +  * DMA_MEMCPY_SG
>> +- The device supports scatterlist to/from memory.
>> +
>>* DMA_PRIVATE
>>  - The devices only supports slave transfers, and as such isn't
>>available for async transfers.
>> diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
>> index 428b141..4d2c4e1 100644
>> --- a/drivers/dma/dmaengine.c
>> +++ b/drivers/dma/dmaengine.c
>> @@ -937,6 +937,8 @@ int dma_async_device_register(struct dma_device *device)
>>  !device->device_prep_dma_memset);
>>  BUG_ON(dma_has_cap(DMA_INTERRUPT, device->cap_mask) &&
>>  !device->device_prep_dma_interrupt);
>> +BUG_ON(dma_has_cap(DMA_MEMCPY_SG, device->cap_mask) &&
>> +!device->device_prep_dma_memcpy_sg);
>>  BUG_ON(dma_has_cap(DMA_CYCLIC, device->cap_mask) &&
>>  !device->device_prep_dma_cyclic);
>>  BUG_ON(dma_has_cap(DMA_INTERLEAVE, device->cap_mask) &&
>> diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
>> index 64fbd38..0c91411 100644
>> --- a/include/linux/dmaengine.h
>> +++ b/include/linux/dmaengine.h
>> @@ -67,6 +67,7 @@ enum dma_transaction_type {
>>  DMA_PQ_VAL,
>>  DMA_MEMSET,
>>  DMA_MEMSET_SG,
>> +DMA_MEMCPY_SG,
>>  DMA_INTERRUPT,
>>  DMA_PRIVATE,
>>  DMA_ASYNC_TX,
>> @@ -692,6 +693,7 @@ struct dma_filter {
>>   * @device_prep_dma_pq_val: prepares a pqzero_sum operation
>>   * @device_prep_dma_memset: prepares a memset operation
>>   * @device_prep_dma_memset_sg: prepares a memset operation over a scatter 
>> list
>> + * @device_prep_dma_memcpy_sg: prepares memcpy between scatterlist and 
>> buffer
>>   * @device_prep_dma_interrupt: prepares an end of chain interrupt operation
>>   * @device_prep_slave_sg: prepares a slave dma operation
>>   * @device_prep_dma_cyclic: prepare a cyclic dma operation suitable for 
>> audio.
>> @@ -768,6 +770,10 @@ struct dma_device {
>>  struct dma_async_tx_descriptor *(*device_prep_dma_memset_sg)(
>>  struct dma_chan *chan, struct scatterlist *sg,
>>  unsigned int nents, int value, unsigned long flags);
>> +struct dma_async_tx_descriptor *(*device_prep_dma_memcpy_sg)(
>> +struct dma_chan *chan,
>> +struct scatterlist *sg, unsigned int sg_nents,
>> +dma_addr_t buf, bool to_sg, unsigned long flags);
>>  struct dma_async_tx_descriptor *(*device_prep_dma_interrupt)(
>>  struct dma_chan *chan, unsigned long flags);
>>  
>> @@ -899,6 +905,19 @@ static inline struct dma_async_tx_descriptor 
>> *dmaengine_prep_dma_memcpy(
>>  len, flags);
>>  }
>>  
>> +static inline struct dma_async_tx_descriptor *dmaengine_prep_dma_memcpy_sg(
>> +struct dma_chan *chan, struct scatterlist *sg,
>> +unsigned int sg_nents, dma_addr_t buf, bool to_sg,
>> +unsigned long flags)
>> +{
>> +if (!chan || !chan->device ||
>> +!chan->device->device_prep_dma_memcpy_sg)
>> +return NULL;
>> +
>> +return chan->device->device_prep_dma_memcpy_sg(chan, sg, sg_nents,
>> +   buf, to_sg, flags);
>> +}
>> +
>>  /**
>>   * dmaengine_terminate_all() - Terminate all active DMA transfers
>>   * @chan: The channel for which to terminate the transfers
>>
___
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm


Re: [v6,2/8] dmaengine: Add DMA_MEMCPY_SG transaction op

2017-08-30 Thread Robin Murphy
On 25/08/17 21:59, Dave Jiang wrote:
> Adding a dmaengine transaction operation that allows copy to/from a
> scatterlist and a flat buffer.

Apologies if I'm late to the party, but doesn't DMA_SG already cover
this use-case? As far as I can see, all this does is save the caller
from setting up a single-entry scatterlist to describe the buffer - even
if such a simplified interface is justified it seems like something that
could be implemented as a wrapper around dmaengine_prep_dma_sg() rather
than the providers having to implement a whole extra callback.

Robin.

> 
> Signed-off-by: Dave Jiang 
> ---
>  Documentation/dmaengine/provider.txt |3 +++
>  drivers/dma/dmaengine.c  |2 ++
>  include/linux/dmaengine.h|   19 +++
>  3 files changed, 24 insertions(+)
> 
> diff --git a/Documentation/dmaengine/provider.txt 
> b/Documentation/dmaengine/provider.txt
> index a75f52f..6241e36 100644
> --- a/Documentation/dmaengine/provider.txt
> +++ b/Documentation/dmaengine/provider.txt
> @@ -181,6 +181,9 @@ Currently, the types available are:
>  - Used by the client drivers to register a callback that will be
>called on a regular basis through the DMA controller interrupt
>  
> +  * DMA_MEMCPY_SG
> +- The device supports scatterlist to/from memory.
> +
>* DMA_PRIVATE
>  - The devices only supports slave transfers, and as such isn't
>available for async transfers.
> diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
> index 428b141..4d2c4e1 100644
> --- a/drivers/dma/dmaengine.c
> +++ b/drivers/dma/dmaengine.c
> @@ -937,6 +937,8 @@ int dma_async_device_register(struct dma_device *device)
>   !device->device_prep_dma_memset);
>   BUG_ON(dma_has_cap(DMA_INTERRUPT, device->cap_mask) &&
>   !device->device_prep_dma_interrupt);
> + BUG_ON(dma_has_cap(DMA_MEMCPY_SG, device->cap_mask) &&
> + !device->device_prep_dma_memcpy_sg);
>   BUG_ON(dma_has_cap(DMA_CYCLIC, device->cap_mask) &&
>   !device->device_prep_dma_cyclic);
>   BUG_ON(dma_has_cap(DMA_INTERLEAVE, device->cap_mask) &&
> diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
> index 64fbd38..0c91411 100644
> --- a/include/linux/dmaengine.h
> +++ b/include/linux/dmaengine.h
> @@ -67,6 +67,7 @@ enum dma_transaction_type {
>   DMA_PQ_VAL,
>   DMA_MEMSET,
>   DMA_MEMSET_SG,
> + DMA_MEMCPY_SG,
>   DMA_INTERRUPT,
>   DMA_PRIVATE,
>   DMA_ASYNC_TX,
> @@ -692,6 +693,7 @@ struct dma_filter {
>   * @device_prep_dma_pq_val: prepares a pqzero_sum operation
>   * @device_prep_dma_memset: prepares a memset operation
>   * @device_prep_dma_memset_sg: prepares a memset operation over a scatter 
> list
> + * @device_prep_dma_memcpy_sg: prepares memcpy between scatterlist and buffer
>   * @device_prep_dma_interrupt: prepares an end of chain interrupt operation
>   * @device_prep_slave_sg: prepares a slave dma operation
>   * @device_prep_dma_cyclic: prepare a cyclic dma operation suitable for 
> audio.
> @@ -768,6 +770,10 @@ struct dma_device {
>   struct dma_async_tx_descriptor *(*device_prep_dma_memset_sg)(
>   struct dma_chan *chan, struct scatterlist *sg,
>   unsigned int nents, int value, unsigned long flags);
> + struct dma_async_tx_descriptor *(*device_prep_dma_memcpy_sg)(
> + struct dma_chan *chan,
> + struct scatterlist *sg, unsigned int sg_nents,
> + dma_addr_t buf, bool to_sg, unsigned long flags);
>   struct dma_async_tx_descriptor *(*device_prep_dma_interrupt)(
>   struct dma_chan *chan, unsigned long flags);
>  
> @@ -899,6 +905,19 @@ static inline struct dma_async_tx_descriptor 
> *dmaengine_prep_dma_memcpy(
>   len, flags);
>  }
>  
> +static inline struct dma_async_tx_descriptor *dmaengine_prep_dma_memcpy_sg(
> + struct dma_chan *chan, struct scatterlist *sg,
> + unsigned int sg_nents, dma_addr_t buf, bool to_sg,
> + unsigned long flags)
> +{
> + if (!chan || !chan->device ||
> + !chan->device->device_prep_dma_memcpy_sg)
> + return NULL;
> +
> + return chan->device->device_prep_dma_memcpy_sg(chan, sg, sg_nents,
> +buf, to_sg, flags);
> +}
> +
>  /**
>   * dmaengine_terminate_all() - Terminate all active DMA transfers
>   * @chan: The channel for which to terminate the transfers
> 
___
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm