[PATCH v6 2/2] dma: Add Xilinx AXI Video Direct Memory Access Engine driver support

Jassi Brar jassisinghbrar at gmail.com
Mon Mar 24 11:23:44 EDT 2014


On Mon, Mar 24, 2014 at 8:44 PM, Srikanth Thokala <sthokal at xilinx.com> wrote:
> On Mon, Mar 24, 2014 at 4:21 PM, Jassi Brar <jaswinder.singh at linaro.org> wrote:
>> On 24 March 2014 14:30, Srikanth Thokala <sthokal at xilinx.com> wrote:
>>> Hi Jassi,
>>>
>>> Thanks for the Acked-by.
>>>
>>> On Mon, Mar 24, 2014 at 11:50 AM, Jassi Brar <jassisinghbrar at gmail.com> wrote:
>>>> On Tue, Mar 18, 2014 at 12:36 AM, Srikanth Thokala <sthokal at xilinx.com> wrote:
>>>>
>>>>> +
>>>>> +/**
>>>>> + * xilinx_vdma_prep_slave_sg - prepare a descriptor for a DMA_SLAVE transaction
>>>>
>>>> Should be xilinx_vdma_dma_prep_interleaved here.
>>>
>>> Ok, I will correct it.  Thanks.
>>>
>>>>
>>>>> +/**
>>>>> + * xilinx_vdma_channel_set_config - Configure VDMA channel
>>>>> + * Run-time configuration for Axi VDMA, supports:
>>>>> + * . halt the channel
>>>>> + * . configure interrupt coalescing and inter-packet delay threshold
>>>>> + * . start/stop parking
>>>>> + * . enable genlock
>>>>> + *
>>>>> + * @dchan: DMA channel
>>>>> + * @cfg: VDMA device configuration pointer
>>>>> + *
>>>>> + * Return: '0' on success and failure value on error
>>>>> + */
>>>>> +int xilinx_vdma_channel_set_config(struct dma_chan *dchan,
>>>>> +                                       struct xilinx_vdma_config *cfg)
>>>>> +{
>>>>> +       struct xilinx_vdma_chan *chan = to_xilinx_chan(dchan);
>>>>> +       u32 dmacr;
>>>>> +
>>>>> +       if (cfg->reset)
>>>>> +               return xilinx_vdma_chan_reset(chan);
>>>>> +
>>>>> +       dmacr = vdma_ctrl_read(chan, XILINX_VDMA_REG_DMACR);
>>>>> +
>>>>> +       chan->config.frm_dly = cfg->frm_dly;
>>>>> +       chan->config.park = cfg->park;
>>>>> +
>>>>> +       /* genlock settings */
>>>>> +       chan->config.gen_lock = cfg->gen_lock;
>>>>> +       chan->config.master = cfg->master;
>>>>> +
>>>>> +       if (cfg->gen_lock && chan->genlock) {
>>>>> +               dmacr |= XILINX_VDMA_DMACR_GENLOCK_EN;
>>>>> +               dmacr |= cfg->master << XILINX_VDMA_DMACR_MASTER_SHIFT;
>>>>> +       }
>>>>> +
>>>>> +       chan->config.frm_cnt_en = cfg->frm_cnt_en;
>>>>> +       if (cfg->park)
>>>>> +               chan->config.park_frm = cfg->park_frm;
>>>>> +       else
>>>>> +               chan->config.park_frm = -1;
>>>>> +
>>>>> +       chan->config.coalesc = cfg->coalesc;
>>>>> +       chan->config.delay = cfg->delay;
>>>>> +
>>>>> +       if (cfg->coalesc <= XILINX_VDMA_DMACR_FRAME_COUNT_MAX) {
>>>>> +               dmacr |= cfg->coalesc << XILINX_VDMA_DMACR_FRAME_COUNT_SHIFT;
>>>>> +               chan->config.coalesc = cfg->coalesc;
>>>>> +       }
>>>>> +
>>>>> +       if (cfg->delay <= XILINX_VDMA_DMACR_DELAY_MAX) {
>>>>> +               dmacr |= cfg->delay << XILINX_VDMA_DMACR_DELAY_SHIFT;
>>>>> +               chan->config.delay = cfg->delay;
>>>>> +       }
>>>>> +
>>>>> +       /* FSync Source selection */
>>>>> +       dmacr &= ~XILINX_VDMA_DMACR_FSYNCSRC_MASK;
>>>>> +       dmacr |= cfg->ext_fsync << XILINX_VDMA_DMACR_FSYNCSRC_SHIFT;
>>>>> +
>>>>> +       vdma_ctrl_write(chan, XILINX_VDMA_REG_DMACR, dmacr);
>>>>> +
>>>>> +       return 0;
>>>>> +}
>>>>> +EXPORT_SYMBOL(xilinx_vdma_channel_set_config);
>>>>> +
>>>> Bypassing the DMAEngine api doesn't seem very neat.
>>>> Is there currently any client driver that /changes/ this configuration
>>>> between channel requests? If no, then simply get the configuration
>>>> from DT node and avoid this api.
>>>
>>> This change is recently suggested by Lars in the following patch series,
>>> "dma: Remove comment about embedding dma_slave_config into custom structs"
>>>
>> If some user does indeed need to /change/ the configuration across
>> channel requests, we can keep function. Is there currently any such
>> user?
>
> Yes, we are planning to push these changes soon to drivers/media/.
>
This is not what I asked though. I asked if some client driver changes
the configuration options based on some runtime decision? If yes, then
the api is warranted. Otherwise, you might consider getting the static
configuration from DT. That's all I suggest.



More information about the linux-arm-kernel mailing list