diff mbox

[v10,2/2] dmaengine: Add Xilinx zynqmp dma engine driver support

Message ID 1464765839-29018-2-git-send-email-appanad@xilinx.com (mailing list archive)
State Changes Requested
Headers show

Commit Message

Appana Durga Kedareswara rao June 1, 2016, 7:23 a.m. UTC
Added the driver for zynqmp dma engine used in Zynq
UltraScale+ MPSoC. This dma controller supports memory to memory
and memory to I/O buffer transfers.

Signed-off-by: Punnaiah Choudary Kalluri <punnaia@xilinx.com>
Signed-off-by: Kedareswara rao Appana <appanad@xilinx.com>
---
Changes for v10:
- Use 64bit write for 64-bit platforms as suggested by Shubhrajyoti.
Changes for v9:
- Derive the include sg software runtime configuration parameter
  through config strucutre as suggested by the Lars.
Changes for v8:
- Derive the software runtime configuration parameters
  through config strucutre and configure them during transfer
  as suggested by the Lars.
Changes for v7:
- Fixed kbuild compilation warnings.
- Fixed {src,dst}_addr_widths are supposed to be a bitmask of
  supported slave device widths as suggested by Rob.
Changes in v6:
- Removed unnecessary axcache properties from the driver
- Fixed compilation issues
Changes in v5:
- Removed in_interrupt check from the tasklet cleanup as
  suggested by the vinod/lars.
Changes in v4:
- Modified the defines to start with ZYNQMP_DMA perfix
- Changed the zynqmp_dma_alloc_chan_resources to return number of
  allocated descriptors
- Changed the zynqmp_dma_device variable name
- Released the locks before calling user callback
- freeup irq in zynqmp_dma_chan_remove function.
Changes in v3:
- Modified the zynqmp_dma_chan_is_idle function return type to bool
Changes in v2:
- Corrected the function header documentation
- Framework expects bus-width value in bytes. so, fixed it
- Removed magic numbers for bus-width

 drivers/dma/Kconfig             |    6 +
 drivers/dma/xilinx/Makefile     |    1 +
 drivers/dma/xilinx/zynqmp_dma.c | 1269 +++++++++++++++++++++++++++++++++++++++
 3 files changed, 1276 insertions(+)
 create mode 100644 drivers/dma/xilinx/zynqmp_dma.c

Comments

Vinod Koul June 7, 2016, 7:08 a.m. UTC | #1
On Wed, Jun 01, 2016 at 12:53:59PM +0530, Kedareswara rao Appana wrote:
> +config XILINX_ZYNQMP_DMA
> +	tristate "Xilinx ZynqMP DMA Engine"
> +	depends on (ARCH_ZYNQ || MICROBLAZE || ARM64)
> +	select DMA_ENGINE
> +	help
> +	  Enable support for Xilinx ZynqMP DMA controller.

Kconfig and makefile is sorted alphabetically, pls update this

> +#include <linux/bitops.h>
> +#include <linux/dma-mapping.h>
> +#include <linux/dma/xilinx_dma.h>
> +#include <linux/dmaengine.h>
> +#include <linux/dmapool.h>
> +#include <linux/init.h>
> +#include <linux/interrupt.h>
> +#include <linux/io.h>
> +#include <linux/module.h>
> +#include <linux/of_address.h>
> +#include <linux/of_dma.h>
> +#include <linux/of_irq.h>
> +#include <linux/of_platform.h>
> +#include <linux/slab.h>
> +#include <linux/clk.h>
> +#include <linux/io-64-nonatomic-lo-hi.h>

do you really need so many headers?

> +static inline void zynqmp_dma_writeq(struct zynqmp_dma_chan *chan, u32 reg,
> +				     u64 value)
> +{
> +#if defined(CONFIG_PHYS_ADDR_T_64BIT)
> +	writeq(value, chan->regs + reg);
> +#else
> +	lo_hi_writeq(value, chan->regs + reg);
> +#endif

why do you need this? can you explain..

> +static inline bool zynqmp_dma_chan_is_idle(struct zynqmp_dma_chan *chan)
> +{
> +	return chan->idle;
> +

empty line not required


> +static void zynqmp_dma_desc_config_eod(struct zynqmp_dma_chan *chan, void *desc)

eod? 80 line?


> +	val = 0;
> +	if (chan->config.has_sg)
> +		val |= ZYNQMP_DMA_POINT_TYPE_SG;

why not val = and get rid of assign to 0 above

> +	writel(val, chan->regs + ZYNQMP_DMA_CTRL0);

okay why write 0 for no sg mode?

> +
> +	val = 0;
> +	if (chan->is_dmacoherent) {
> +		val |= ZYNQMP_DMA_AXCOHRNT;
> +		val = (val & ~ZYNQMP_DMA_AXCACHE) |
> +			(ZYNQMP_DMA_AXCACHE_VAL << ZYNQMP_DMA_AXCACHE_OFST);
> +	}
> +	writel(val, chan->regs + ZYNQMP_DMA_DSCR_ATTR);

same comments here too

> +	spin_lock_bh(&chan->lock);
> +	desc = list_first_entry(&chan->free_list, struct zynqmp_dma_desc_sw,
> +				 node);

how about:

	desc = list_first_entry(&chan->free_list,
			struct zynqmp_dma_desc_sw, node);

> +	chan->desc_free_cnt++;
> +	list_add_tail(&sdesc->node, &chan->free_list);
> +	list_for_each_entry_safe(child, next, &sdesc->tx_list, node) {
> +		chan->desc_free_cnt++;
> +		INIT_LIST_HEAD(&child->tx_list);
> +		list_move_tail(&child->node, &chan->free_list);
> +	}
> +	INIT_LIST_HEAD(&sdesc->tx_list);

why are you initializing list in free?

> +static int zynqmp_dma_alloc_chan_resources(struct dma_chan *dchan)
> +{
> +	struct zynqmp_dma_chan *chan = to_chan(dchan);
> +	struct zynqmp_dma_desc_sw *desc;
> +	int i;
> +
> +	chan->sw_desc_pool = kzalloc(sizeof(*desc) * ZYNQMP_DMA_NUM_DESCS,
> +				     GFP_KERNEL);
> +	if (!chan->sw_desc_pool)
> +		return -ENOMEM;

empty line here pls

> +static enum dma_status zynqmp_dma_tx_status(struct dma_chan *dchan,
> +				      dma_cookie_t cookie,
> +				      struct dma_tx_state *txstate)
> +{
> +	return dma_cookie_status(dchan, cookie, txstate);

why do you need this wrapper, assign dma_cookie_status as your status
callback.

> +int zynqmp_dma_channel_set_config(struct dma_chan *dchan,
> +				  struct zynqmp_dma_config *cfg)
> +{
> +	struct zynqmp_dma_chan *chan = to_chan(dchan);
> +
> +	chan->config.ovrfetch = cfg->ovrfetch;
> +	chan->config.has_sg = cfg->has_sg;

is this HW capability? if so why would anyone not like to use it!

> +	chan->config.ratectrl = cfg->ratectrl;
> +	chan->config.src_issue = cfg->src_issue;
> +	chan->config.src_burst_len = cfg->src_burst_len;
> +	chan->config.dst_burst_len = cfg->dst_burst_len;

can you describe these parameters?

How would a client know how to configure them?
> +
> +	return 0;
> +}
> +EXPORT_SYMBOL(zynqmp_dma_channel_set_config);

Not _GPL?

> +	chan->bus_width = ZYNQMP_DMA_BUS_WIDTH_64;
> +	chan->config.src_issue = ZYNQMP_DMA_SRC_ISSUE_RST_VAL;
> +	chan->config.dst_burst_len = ZYNQMP_DMA_AWLEN_RST_VAL;
> +	chan->config.src_burst_len = ZYNQMP_DMA_ARLEN_RST_VAL;
> +	err = of_property_read_u32(node, "xlnx,bus-width", &chan->bus_width);
> +	if ((err < 0) && ((chan->bus_width != ZYNQMP_DMA_BUS_WIDTH_64) ||
> +			  (chan->bus_width != ZYNQMP_DMA_BUS_WIDTH_128))) {
> +		dev_err(zdev->dev, "invalid bus-width value");
> +		return err;
> +	}
> +
> +	chan->bus_width = chan->bus_width / 8;

?

why not update it once?
Appana Durga Kedareswara rao June 8, 2016, 7:40 a.m. UTC | #2
Hi Vinod,

	Thanks for the review...

> 
> On Wed, Jun 01, 2016 at 12:53:59PM +0530, Kedareswara rao Appana wrote:
> > +config XILINX_ZYNQMP_DMA
> > +	tristate "Xilinx ZynqMP DMA Engine"
> > +	depends on (ARCH_ZYNQ || MICROBLAZE || ARM64)
> > +	select DMA_ENGINE
> > +	help
> > +	  Enable support for Xilinx ZynqMP DMA controller.
> 
> Kconfig and makefile is sorted alphabetically, pls update this

Ok Sure will fix in the next version...

> 
> > +#include <linux/bitops.h>
> > +#include <linux/dma-mapping.h>
> > +#include <linux/dma/xilinx_dma.h>
> > +#include <linux/dmaengine.h>
> > +#include <linux/dmapool.h>
> > +#include <linux/init.h>
> > +#include <linux/interrupt.h>
> > +#include <linux/io.h>
> > +#include <linux/module.h>
> > +#include <linux/of_address.h>
> > +#include <linux/of_dma.h>
> > +#include <linux/of_irq.h>
> > +#include <linux/of_platform.h>
> > +#include <linux/slab.h>
> > +#include <linux/clk.h>
> > +#include <linux/io-64-nonatomic-lo-hi.h>
> 
> do you really need so many headers?

Will remove unwanted header files in the next version...

> 
> > +static inline void zynqmp_dma_writeq(struct zynqmp_dma_chan *chan, u32
> reg,
> > +				     u64 value)
> > +{
> > +#if defined(CONFIG_PHYS_ADDR_T_64BIT)
> > +	writeq(value, chan->regs + reg);
> > +#else
> > +	lo_hi_writeq(value, chan->regs + reg); #endif
> 
> why do you need this? can you explain..

writeq is available only on 64-bit platforms to make it work for every platform I have modified like this.
Will fix in the next version will use only lo_hi_writeq which is available across all the platforms.

> 
> > +static inline bool zynqmp_dma_chan_is_idle(struct zynqmp_dma_chan
> > +*chan) {
> > +	return chan->idle;
> > +
> 
> empty line not required

OK will fix...
> 
> 
> > +static void zynqmp_dma_desc_config_eod(struct zynqmp_dma_chan *chan,
> > +void *desc)
> 
> eod? 80 line?

It is exactly 80lines and check patch.pl didn't complained about it

> 
> 
> > +	val = 0;
> > +	if (chan->config.has_sg)
> > +		val |= ZYNQMP_DMA_POINT_TYPE_SG;
> 
> why not val = and get rid of assign to 0 above

Ok Will fix in the next version.

> 
> > +	writel(val, chan->regs + ZYNQMP_DMA_CTRL0);
> 
> okay why write 0 for no sg mode?

Ok will fix in the next version..

> 
> > +
> > +	val = 0;
> > +	if (chan->is_dmacoherent) {
> > +		val |= ZYNQMP_DMA_AXCOHRNT;
> > +		val = (val & ~ZYNQMP_DMA_AXCACHE) |
> > +			(ZYNQMP_DMA_AXCACHE_VAL <<
> ZYNQMP_DMA_AXCACHE_OFST);
> > +	}
> > +	writel(val, chan->regs + ZYNQMP_DMA_DSCR_ATTR);
> 
> same comments here too

Ok will fix in the next version..

> 
> > +	spin_lock_bh(&chan->lock);
> > +	desc = list_first_entry(&chan->free_list, struct zynqmp_dma_desc_sw,
> > +				 node);
> 
> how about:
> 
> 	desc = list_first_entry(&chan->free_list,
> 			struct zynqmp_dma_desc_sw, node);

Ok will fix...

> 
> > +	chan->desc_free_cnt++;
> > +	list_add_tail(&sdesc->node, &chan->free_list);
> > +	list_for_each_entry_safe(child, next, &sdesc->tx_list, node) {
> > +		chan->desc_free_cnt++;
> > +		INIT_LIST_HEAD(&child->tx_list);
> > +		list_move_tail(&child->node, &chan->free_list);
> > +	}
> > +	INIT_LIST_HEAD(&sdesc->tx_list);
> 
> why are you initializing list in free?

Will fix in the next version of the patch...

> 
> > +static int zynqmp_dma_alloc_chan_resources(struct dma_chan *dchan) {
> > +	struct zynqmp_dma_chan *chan = to_chan(dchan);
> > +	struct zynqmp_dma_desc_sw *desc;
> > +	int i;
> > +
> > +	chan->sw_desc_pool = kzalloc(sizeof(*desc) *
> ZYNQMP_DMA_NUM_DESCS,
> > +				     GFP_KERNEL);
> > +	if (!chan->sw_desc_pool)
> > +		return -ENOMEM;
> 
> empty line here pls

Ok will fix...

> 
> > +static enum dma_status zynqmp_dma_tx_status(struct dma_chan *dchan,
> > +				      dma_cookie_t cookie,
> > +				      struct dma_tx_state *txstate) {
> > +	return dma_cookie_status(dchan, cookie, txstate);
> 
> why do you need this wrapper, assign dma_cookie_status as your status
> callback.

Ok will fix...

> 
> > +int zynqmp_dma_channel_set_config(struct dma_chan *dchan,
> > +				  struct zynqmp_dma_config *cfg)
> > +{
> > +	struct zynqmp_dma_chan *chan = to_chan(dchan);
> > +
> > +	chan->config.ovrfetch = cfg->ovrfetch;
> > +	chan->config.has_sg = cfg->has_sg;
> 
> is this HW capability? if so why would anyone not like to use it!

Yes it is HW capability. It can be either in simple mode or SG mode
Earlier In the driver this configuration is read from the device-tree 
But as per lars and your suggestion moved it as runtime config parameters.

> 
> > +	chan->config.ratectrl = cfg->ratectrl;
> > +	chan->config.src_issue = cfg->src_issue;
> > +	chan->config.src_burst_len = cfg->src_burst_len;
> > +	chan->config.dst_burst_len = cfg->dst_burst_len;
> 
> can you describe these parameters?
ratectl:
Rate control can be independently enabled per channel. When rate control is enabled, the
DMA channel uses the rate control count to schedule successive data read transactions.
src_issue:
Tells outstanding transaction on SRC.
Burst_len: 
Configures the burst length of the src and dst transfers...

> 
> How would a client know how to configure them?

With the default values of the config parameters driver will work.
If user has specific requirement to change these parameters they can pass
It to the driver using set_config API and all these parameters are
Documented in the include/linux/dma/xilinx_dma.h file...

> > +
> > +	return 0;
> > +}
> > +EXPORT_SYMBOL(zynqmp_dma_channel_set_config);
> 
> Not _GPL?

Ok will fix in the next version...

> 
> > +	chan->bus_width = ZYNQMP_DMA_BUS_WIDTH_64;
> > +	chan->config.src_issue = ZYNQMP_DMA_SRC_ISSUE_RST_VAL;
> > +	chan->config.dst_burst_len = ZYNQMP_DMA_AWLEN_RST_VAL;
> > +	chan->config.src_burst_len = ZYNQMP_DMA_ARLEN_RST_VAL;
> > +	err = of_property_read_u32(node, "xlnx,bus-width", &chan-
> >bus_width);
> > +	if ((err < 0) && ((chan->bus_width != ZYNQMP_DMA_BUS_WIDTH_64) ||
> > +			  (chan->bus_width !=
> ZYNQMP_DMA_BUS_WIDTH_128))) {
> > +		dev_err(zdev->dev, "invalid bus-width value");
> > +		return err;
> > +	}
> > +
> > +	chan->bus_width = chan->bus_width / 8;
> 
> ?
> 
> why not update it once?

Ok will fix in the next version...


Regards,
Kedar.
--
To unsubscribe from this list: send the line "unsubscribe dmaengine" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Vinod Koul June 13, 2016, 5:50 a.m. UTC | #3
On Wed, Jun 08, 2016 at 07:40:52AM +0000, Appana Durga Kedareswara Rao wrote:
> > > +static void zynqmp_dma_desc_config_eod(struct zynqmp_dma_chan *chan,
> > > +void *desc)
> > 
> > eod? 80 line?

What's eod?

> > > +int zynqmp_dma_channel_set_config(struct dma_chan *dchan,
> > > +				  struct zynqmp_dma_config *cfg)
> > > +{
> > > +	struct zynqmp_dma_chan *chan = to_chan(dchan);
> > > +
> > > +	chan->config.ovrfetch = cfg->ovrfetch;
> > > +	chan->config.has_sg = cfg->has_sg;
> > 
> > is this HW capability? if so why would anyone not like to use it!
> 
> Yes it is HW capability. It can be either in simple mode or SG mode
> Earlier In the driver this configuration is read from the device-tree 
> But as per lars and your suggestion moved it as runtime config parameters.

If sg mode is available why would anyone _not_ want it?

I do not think there is point to have this

> 
> > 
> > > +	chan->config.ratectrl = cfg->ratectrl;
> > > +	chan->config.src_issue = cfg->src_issue;
> > > +	chan->config.src_burst_len = cfg->src_burst_len;
> > > +	chan->config.dst_burst_len = cfg->dst_burst_len;
> > 
> > can you describe these parameters?
> ratectl:
> Rate control can be independently enabled per channel. When rate control is enabled, the
> DMA channel uses the rate control count to schedule successive data read transactions.

And how is this used by client?

> src_issue:
> Tells outstanding transaction on SRC.

This should be read only then, right?

> Burst_len: 
> Configures the burst length of the src and dst transfers...

Hmmm, but you are on memcpy, so that should be programmed for throughput?

> > 
> > How would a client know how to configure them?
> 
> With the default values of the config parameters driver will work.

But how will client know what is default!

> If user has specific requirement to change these parameters they can pass
> It to the driver using set_config API and all these parameters are
> Documented in the include/linux/dma/xilinx_dma.h file...

Can you give me an example where user would like to do that
Appana Durga Kedareswara rao June 14, 2016, 8:18 a.m. UTC | #4
Hi Vinod,

	Thanks for the review...

> 
> On Wed, Jun 08, 2016 at 07:40:52AM +0000, Appana Durga Kedareswara Rao
> wrote:
> > > > +static void zynqmp_dma_desc_config_eod(struct zynqmp_dma_chan
> > > > +*chan, void *desc)
> > >
> > > eod? 80 line?
> 
> What's eod?

End of descriptor...

> 
> > > > +int zynqmp_dma_channel_set_config(struct dma_chan *dchan,
> > > > +				  struct zynqmp_dma_config *cfg) {
> > > > +	struct zynqmp_dma_chan *chan = to_chan(dchan);
> > > > +
> > > > +	chan->config.ovrfetch = cfg->ovrfetch;
> > > > +	chan->config.has_sg = cfg->has_sg;
> > >
> > > is this HW capability? if so why would anyone not like to use it!
> >
> > Yes it is HW capability. It can be either in simple mode or SG mode
> > Earlier In the driver this configuration is read from the device-tree
> > But as per lars and your suggestion moved it as runtime config parameters.
> 
> If sg mode is available why would anyone _not_ want it?
> 
> I do not think there is point to have this

You mean always keep the device in SG mode and provide an option 
For simple dma mode if user want to use simple DMA mode??

There are few features that are available in the simple DMA mode won't
Available in SG mode like write only DMA , read only DMA mode etc...

> 
> >
> > >
> > > > +	chan->config.ratectrl = cfg->ratectrl;
> > > > +	chan->config.src_issue = cfg->src_issue;
> > > > +	chan->config.src_burst_len = cfg->src_burst_len;
> > > > +	chan->config.dst_burst_len = cfg->dst_burst_len;
> > >
> > > can you describe these parameters?
> > ratectl:
> > Rate control can be independently enabled per channel. When rate
> > control is enabled, the DMA channel uses the rate control count to schedule
> successive data read transactions.
> 
> And how is this used by client?

When rate control is enabled, ZDMA channel uses the rate control count
To schedule successive data read transactions I mean kind of flow control to schedule 
Transactions at fixed intervals instead of pumping the transfers without delay or whenever bus is available

Rate control count register definition (11:0):
Scheduling interval for SRC AXI transaction, only used if rate control is enabled 


> 
> > src_issue:
> > Tells outstanding transaction on SRC.
> 
> This should be read only then, right?

It is a Read/Write register
http://www.xilinx.com/html_docs/registers/ug1087/ug1087-zynq-ultrascale-registers.html 
By default it is configured for Max transactions.
If user want to limit it they can limit it using this config option.

> 
> > Burst_len:
> > Configures the burst length of the src and dst transfers...
> 
> Hmmm, but you are on memcpy, so that should be programmed for throughput?

Yes...

> 
> > >
> > > How would a client know how to configure them?
> >
> > With the default values of the config parameters driver will work.
> 
> But how will client know what is default!

Default values means IP default state after reset.
If user not aware of the above parameters also still the driver will work for basic functionality.
Do you want me to implement one more API get_config so that 
Whenever user will call the get_config he will know the default values
Of the config parameters?

> 
> > If user has specific requirement to change these parameters they can
> > pass It to the driver using set_config API and all these parameters
> > are Documented in the include/linux/dma/xilinx_dma.h file...
> 
> Can you give me an example where user would like to do that


I am using customized dma test client.
There I am calling this set_config API before triggering memcpy/SG operations.

Regards,
Kedar.

> 
> --
> ~Vinod
> --
> To unsubscribe from this list: send the line "unsubscribe dmaengine" in the body
> of a message to majordomo@vger.kernel.org More majordomo info at
> http://vger.kernel.org/majordomo-info.html
--
To unsubscribe from this list: send the line "unsubscribe dmaengine" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Vinod Koul June 15, 2016, 4:50 p.m. UTC | #5
On Tue, Jun 14, 2016 at 08:18:09AM +0000, Appana Durga Kedareswara Rao wrote:
> > > Yes it is HW capability. It can be either in simple mode or SG mode
> > > Earlier In the driver this configuration is read from the device-tree
> > > But as per lars and your suggestion moved it as runtime config parameters.
> > 
> > If sg mode is available why would anyone _not_ want it?
> > 
> > I do not think there is point to have this
> 
> You mean always keep the device in SG mode and provide an option 
> For simple dma mode if user want to use simple DMA mode??

Yes, why would anyone want to use single if sg is available?

> 
> There are few features that are available in the simple DMA mode won't
> Available in SG mode like write only DMA , read only DMA mode etc...

Can you explain what these are, how they are used by clients?

> > > > > +	chan->config.ratectrl = cfg->ratectrl;
> > > > > +	chan->config.src_issue = cfg->src_issue;
> > > > > +	chan->config.src_burst_len = cfg->src_burst_len;
> > > > > +	chan->config.dst_burst_len = cfg->dst_burst_len;
> > > >
> > > > can you describe these parameters?
> > > ratectl:
> > > Rate control can be independently enabled per channel. When rate
> > > control is enabled, the DMA channel uses the rate control count to schedule
> > successive data read transactions.
> > 
> > And how is this used by client?
> 
> When rate control is enabled, ZDMA channel uses the rate control count
> To schedule successive data read transactions I mean kind of flow control to schedule 
> Transactions at fixed intervals instead of pumping the transfers without delay or whenever bus is available
> 
> Rate control count register definition (11:0):
> Scheduling interval for SRC AXI transaction, only used if rate control is enabled 

Okay, why would anyone want transactions at fixed interval. We are talking
about DMA, so please give me transfers without delay or whenever bus is
available!


> > > src_issue:
> > > Tells outstanding transaction on SRC.
> > 
> > This should be read only then, right?
> 
> It is a Read/Write register
> http://www.xilinx.com/html_docs/registers/ug1087/ug1087-zynq-ultrascale-registers.html 
> By default it is configured for Max transactions.
> If user want to limit it they can limit it using this config option.

And why would they want that?

> > > Burst_len:
> > > Configures the burst length of the src and dst transfers...
> > 
> > Hmmm, but you are on memcpy, so that should be programmed for throughput?
> 
> Yes...

So max burst lengths then, why would anyone configure lesser?

> > > > How would a client know how to configure them?
> > >
> > > With the default values of the config parameters driver will work.
> > 
> > But how will client know what is default!
> 
> Default values means IP default state after reset.
> If user not aware of the above parameters also still the driver will work for basic functionality.
> Do you want me to implement one more API get_config so that 
> Whenever user will call the get_config he will know the default values
> Of the config parameters?

Looking at above I think we do not warrant programming them. Assume defaults
in your driver for performance. Like max burst lengths, Max transactions,
without delay scheduling.

If you disagree, which is fine, please provide the cases where a client
would need to program these to different values.

> > > If user has specific requirement to change these parameters they can
> > > pass It to the driver using set_config API and all these parameters
> > > are Documented in the include/linux/dma/xilinx_dma.h file...
> > 
> > Can you give me an example where user would like to do that
> 
> I am using customized dma test client.
> There I am calling this set_config API before triggering memcpy/SG operations.

Well that is precisely a  problem. The generic applications wont know "your"
custom API and will not configure that!
Appana Durga Kedareswara rao June 16, 2016, 7:19 a.m. UTC | #6
Hi Vinod,

	Thanks for the review...

> 
> On Tue, Jun 14, 2016 at 08:18:09AM +0000, Appana Durga Kedareswara Rao
> wrote:
> > > > Yes it is HW capability. It can be either in simple mode or SG
> > > > mode Earlier In the driver this configuration is read from the
> > > > device-tree But as per lars and your suggestion moved it as runtime config
> parameters.
> > >
> > > If sg mode is available why would anyone _not_ want it?
> > >
> > > I do not think there is point to have this
> >
> > You mean always keep the device in SG mode and provide an option For
> > simple dma mode if user want to use simple DMA mode??
> 
> Yes, why would anyone want to use single if sg is available?
> 
> >
> > There are few features that are available in the simple DMA mode won't
> > Available in SG mode like write only DMA , read only DMA mode etc...
> 
> Can you explain what these are, how they are used by clients?

Currently it is not implemented but there are cases like,

We want to initialize the some region with known value, then write only dma helps
We want to read dummy data from the fifo, then read only dma helps.
Best case is SPI fifo.

> 
> > > > > > +	chan->config.ratectrl = cfg->ratectrl;
> > > > > > +	chan->config.src_issue = cfg->src_issue;
> > > > > > +	chan->config.src_burst_len = cfg->src_burst_len;
> > > > > > +	chan->config.dst_burst_len = cfg->dst_burst_len;
> > > > >
> > > > > can you describe these parameters?
> > > > ratectl:
> > > > Rate control can be independently enabled per channel. When rate
> > > > control is enabled, the DMA channel uses the rate control count to
> > > > schedule
> > > successive data read transactions.
> > >
> > > And how is this used by client?
> >
> > When rate control is enabled, ZDMA channel uses the rate control count
> > To schedule successive data read transactions I mean kind of flow
> > control to schedule Transactions at fixed intervals instead of pumping
> > the transfers without delay or whenever bus is available
> >
> > Rate control count register definition (11:0):
> > Scheduling interval for SRC AXI transaction, only used if rate control
> > is enabled
> 
> Okay, why would anyone want transactions at fixed interval. We are talking
> about DMA, so please give me transfers without delay or whenever bus is
> available!

Could be that there are certain IPs that requires delay in b/w transactions.
Since IP is providing this feature we are exploring it to the user.

> 
> 
> > > > src_issue:
> > > > Tells outstanding transaction on SRC.
> > >
> > > This should be read only then, right?
> >
> > It is a Read/Write register
> > http://www.xilinx.com/html_docs/registers/ug1087/ug1087-zynq-ultrascal
> > e-registers.html By default it is configured for Max transactions.
> > If user want to limit it they can limit it using this config option.
> 
> And why would they want that?

Could be that there are certain IPs that my not support burst operations or
May not support all burst lengths it will be helpful.
Since IP is providing the feature, we are exploring it to the user.

> 
> > > > Burst_len:
> > > > Configures the burst length of the src and dst transfers...
> > >
> > > Hmmm, but you are on memcpy, so that should be programmed for
> throughput?
> >
> > Yes...
> 
> So max burst lengths then, why would anyone configure lesser?

Depends on the destination and source IPs.

> 
> > > > > How would a client know how to configure them?
> > > >
> > > > With the default values of the config parameters driver will work.
> > >
> > > But how will client know what is default!
> >
> > Default values means IP default state after reset.
> > If user not aware of the above parameters also still the driver will work for
> basic functionality.
> > Do you want me to implement one more API get_config so that Whenever
> > user will call the get_config he will know the default values Of the
> > config parameters?
> 
> Looking at above I think we do not warrant programming them. Assume defaults
> in your driver for performance. Like max burst lengths, Max transactions,
> without delay scheduling.
> 
> If you disagree, which is fine, please provide the cases where a client would
> need to program these to different values.

As said above there are use cases for SPI and others but currently we didn't tested
It.

> 
> > > > If user has specific requirement to change these parameters they
> > > > can pass It to the driver using set_config API and all these
> > > > parameters are Documented in the include/linux/dma/xilinx_dma.h file...
> > >
> > > Can you give me an example where user would like to do that
> >
> > I am using customized dma test client.
> > There I am calling this set_config API before triggering memcpy/SG
> operations.
> 
> Well that is precisely a  problem. The generic applications wont know "your"
> custom API and will not configure that!

As said above, generic application can work with default values. 
These are not custom parameters and we can say these are the dma parameters.
i.e we can generalize these and provide it to user as generic API.


Regards,
Kedar.

> 
> --
> ~Vinod
--
To unsubscribe from this list: send the line "unsubscribe dmaengine" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Vinod Koul June 21, 2016, 3:41 p.m. UTC | #7
On Thu, Jun 16, 2016 at 07:19:38AM +0000, Appana Durga Kedareswara Rao wrote:
> Hi Vinod,
> 
> 	Thanks for the review...
> 
> > 
> > On Tue, Jun 14, 2016 at 08:18:09AM +0000, Appana Durga Kedareswara Rao
> > wrote:
> > > > > Yes it is HW capability. It can be either in simple mode or SG
> > > > > mode Earlier In the driver this configuration is read from the
> > > > > device-tree But as per lars and your suggestion moved it as runtime config
> > parameters.
> > > >
> > > > If sg mode is available why would anyone _not_ want it?
> > > >
> > > > I do not think there is point to have this
> > >
> > > You mean always keep the device in SG mode and provide an option For
> > > simple dma mode if user want to use simple DMA mode??
> > 
> > Yes, why would anyone want to use single if sg is available?

If you have sg mode always enabled, but sg_list is size 1, that its
essentially a single

Btw SPI is a slave case, not a memcpy!

> > > There are few features that are available in the simple DMA mode won't
> > > Available in SG mode like write only DMA , read only DMA mode etc...
> > 
> > Can you explain what these are, how they are used by clients?
> 
> Currently it is not implemented but there are cases like,
> 
> We want to initialize the some region with known value, then write only dma helps
> We want to read dummy data from the fifo, then read only dma helps.

Read will be another transfer, perhpas sg lnegth = 1

> Best case is SPI fifo.

Nope

> 
> > 
> > > > > > > +	chan->config.ratectrl = cfg->ratectrl;
> > > > > > > +	chan->config.src_issue = cfg->src_issue;
> > > > > > > +	chan->config.src_burst_len = cfg->src_burst_len;
> > > > > > > +	chan->config.dst_burst_len = cfg->dst_burst_len;
> > > > > >
> > > > > > can you describe these parameters?
> > > > > ratectl:
> > > > > Rate control can be independently enabled per channel. When rate
> > > > > control is enabled, the DMA channel uses the rate control count to
> > > > > schedule
> > > > successive data read transactions.
> > > >
> > > > And how is this used by client?
> > >
> > > When rate control is enabled, ZDMA channel uses the rate control count
> > > To schedule successive data read transactions I mean kind of flow
> > > control to schedule Transactions at fixed intervals instead of pumping
> > > the transfers without delay or whenever bus is available
> > >
> > > Rate control count register definition (11:0):
> > > Scheduling interval for SRC AXI transaction, only used if rate control
> > > is enabled
> > 
> > Okay, why would anyone want transactions at fixed interval. We are talking
> > about DMA, so please give me transfers without delay or whenever bus is
> > available!
> 
> Could be that there are certain IPs that requires delay in b/w transactions.
> Since IP is providing this feature we are exploring it to the user.

I kind of don't agree to that!

> 
> > 
> > 
> > > > > src_issue:
> > > > > Tells outstanding transaction on SRC.
> > > >
> > > > This should be read only then, right?
> > >
> > > It is a Read/Write register
> > > http://www.xilinx.com/html_docs/registers/ug1087/ug1087-zynq-ultrascal
> > > e-registers.html By default it is configured for Max transactions.
> > > If user want to limit it they can limit it using this config option.
> > 
> > And why would they want that?
> 
> Could be that there are certain IPs that my not support burst operations or
> May not support all burst lengths it will be helpful.
> Since IP is providing the feature, we are exploring it to the user.
> 
> > 
> > > > > Burst_len:
> > > > > Configures the burst length of the src and dst transfers...
> > > >
> > > > Hmmm, but you are on memcpy, so that should be programmed for
> > throughput?
> > >
> > > Yes...
> > 
> > So max burst lengths then, why would anyone configure lesser?
> 
> Depends on the destination and source IPs.

Not for memory copy


> 
> > 
> > > > > > How would a client know how to configure them?
> > > > >
> > > > > With the default values of the config parameters driver will work.
> > > >
> > > > But how will client know what is default!
> > >
> > > Default values means IP default state after reset.
> > > If user not aware of the above parameters also still the driver will work for
> > basic functionality.
> > > Do you want me to implement one more API get_config so that Whenever
> > > user will call the get_config he will know the default values Of the
> > > config parameters?
> > 
> > Looking at above I think we do not warrant programming them. Assume defaults
> > in your driver for performance. Like max burst lengths, Max transactions,
> > without delay scheduling.
> > 
> > If you disagree, which is fine, please provide the cases where a client would
> > need to program these to different values.
> 
> As said above there are use cases for SPI and others but currently we didn't tested
> It.
> 
> > 
> > > > > If user has specific requirement to change these parameters they
> > > > > can pass It to the driver using set_config API and all these
> > > > > parameters are Documented in the include/linux/dma/xilinx_dma.h file...
> > > >
> > > > Can you give me an example where user would like to do that
> > >
> > > I am using customized dma test client.
> > > There I am calling this set_config API before triggering memcpy/SG
> > operations.
> > 
> > Well that is precisely a  problem. The generic applications wont know "your"
> > custom API and will not configure that!
> 
> As said above, generic application can work with default values. 
> These are not custom parameters and we can say these are the dma parameters.
> i.e we can generalize these and provide it to user as generic API.

No it doesn't make it generic API. With generic API the IP should be usable
with any generic interface without specfic hoops to go thru. People are
going to test upstream dmatest on your IP and find it not working or
degraded perf, so better stick with generic API.
Punnaiah Choudary Kalluri June 21, 2016, 4:19 p.m. UTC | #8
Hi Vinod,

> -----Original Message-----
> From: Vinod Koul [mailto:vinod.koul@intel.com]
> Sent: Tuesday, June 21, 2016 9:11 PM
> To: Appana Durga Kedareswara Rao <appanad@xilinx.com>
> Cc: robh+dt@kernel.org; pawel.moll@arm.com; mark.rutland@arm.com;
> ijc+devicetree@hellion.org.uk; galak@codeaurora.org; Michal Simek
> <michals@xilinx.com>; Soren Brinkmann <sorenb@xilinx.com>;
> dan.j.williams@intel.com; moritz.fischer@ettus.com;
> laurent.pinchart@ideasonboard.com; luis@debethencourt.com; Srikanth
> Vemula <svemula@xilinx.com>; Anirudha Sarangi <anirudh@xilinx.com>;
> Punnaiah Choudary Kalluri <punnaia@xilinx.com>;
> devicetree@vger.kernel.org; linux-arm-kernel@lists.infradead.org; linux-
> kernel@vger.kernel.org; dmaengine@vger.kernel.org
> Subject: Re: [PATCH v10 2/2] dmaengine: Add Xilinx zynqmp dma engine
> driver support
>
> On Thu, Jun 16, 2016 at 07:19:38AM +0000, Appana Durga Kedareswara Rao
> wrote:
> > Hi Vinod,
> >
> >     Thanks for the review...
> >
> > >
> > > On Tue, Jun 14, 2016 at 08:18:09AM +0000, Appana Durga Kedareswara
> Rao
> > > wrote:
> > > > > > Yes it is HW capability. It can be either in simple mode or SG
> > > > > > mode Earlier In the driver this configuration is read from the
> > > > > > device-tree But as per lars and your suggestion moved it as runtime
> config
> > > parameters.
> > > > >
> > > > > If sg mode is available why would anyone _not_ want it?
> > > > >
> > > > > I do not think there is point to have this
> > > >
> > > > You mean always keep the device in SG mode and provide an option
> For
> > > > simple dma mode if user want to use simple DMA mode??
> > >
> > > Yes, why would anyone want to use single if sg is available?
>
> If you have sg mode always enabled, but sg_list is size 1, that its
> essentially a single
>

True. As we said, simple mode ha few additional features like write only
And read only dma. So, if user want then here is the provision.

> Btw SPI is a slave case, not a memcpy!
>

Correct. We are working on slave dma support and will patch to this driver.

> > > > There are few features that are available in the simple DMA mode
> won't
> > > > Available in SG mode like write only DMA , read only DMA mode etc...
> > >
> > > Can you explain what these are, how they are used by clients?
> >
> > Currently it is not implemented but there are cases like,
> >
> > We want to initialize the some region with known value, then write only
> dma helps
> > We want to read dummy data from the fifo, then read only dma helps.
>
> Read will be another transfer, perhpas sg lnegth = 1
>


In sg case, both source and destination locations need to be configured.
In simple dma and read-only mode, only source address is required. Simple
Dma just read the data from source location and discard that data. It will save
Unnecessary write to destination here.

> > Best case is SPI fifo.
>
> Nope
>

Why?


> >
> > >
> > > > > > > > +       chan->config.ratectrl = cfg->ratectrl;
> > > > > > > > +       chan->config.src_issue = cfg->src_issue;
> > > > > > > > +       chan->config.src_burst_len = cfg->src_burst_len;
> > > > > > > > +       chan->config.dst_burst_len = cfg->dst_burst_len;
> > > > > > >
> > > > > > > can you describe these parameters?
> > > > > > ratectl:
> > > > > > Rate control can be independently enabled per channel. When rate
> > > > > > control is enabled, the DMA channel uses the rate control count to
> > > > > > schedule
> > > > > successive data read transactions.
> > > > >
> > > > > And how is this used by client?
> > > >
> > > > When rate control is enabled, ZDMA channel uses the rate control
> count
> > > > To schedule successive data read transactions I mean kind of flow
> > > > control to schedule Transactions at fixed intervals instead of pumping
> > > > the transfers without delay or whenever bus is available
> > > >
> > > > Rate control count register definition (11:0):
> > > > Scheduling interval for SRC AXI transaction, only used if rate control
> > > > is enabled
> > >
> > > Okay, why would anyone want transactions at fixed interval. We are
> talking
> > > about DMA, so please give me transfers without delay or whenever bus
> is
> > > available!
> >
> > Could be that there are certain IPs that requires delay in b/w transactions.
> > Since IP is providing this feature we are exploring it to the user.
>
> I kind of don't agree to that!
>

It's a kind of rate control mechanism. We can aslo use this feature to prioritize other
Channel by deliberately controlling the current channel transfer scheduling.

> >
> > >
> > >
> > > > > > src_issue:
> > > > > > Tells outstanding transaction on SRC.
> > > > >
> > > > > This should be read only then, right?
> > > >
> > > > It is a Read/Write register
> > > > http://www.xilinx.com/html_docs/registers/ug1087/ug1087-zynq-
> ultrascal
> > > > e-registers.html By default it is configured for Max transactions.
> > > > If user want to limit it they can limit it using this config option.
> > >
> > > And why would they want that?
> >
> > Could be that there are certain IPs that my not support burst operations or
> > May not support all burst lengths it will be helpful.
> > Since IP is providing the feature, we are exploring it to the user.
> >
> > >
> > > > > > Burst_len:
> > > > > > Configures the burst length of the src and dst transfers...
> > > > >
> > > > > Hmmm, but you are on memcpy, so that should be programmed for
> > > throughput?
> > > >
> > > > Yes...
> > >
> > > So max burst lengths then, why would anyone configure lesser?
> >
> > Depends on the destination and source IPs.
>
> Not for memory copy
>

Depends on the system and how source and destination IPs are interconnected.
Sometimes there is a limitation from interconnection also.
Also some flash controllers providing linear access to NAND or NOR memories may
Have limitation in burst length but still user can use memory copy with desired burst
Length.

>
> >
> > >
> > > > > > > How would a client know how to configure them?
> > > > > >
> > > > > > With the default values of the config parameters driver will work.
> > > > >
> > > > > But how will client know what is default!
> > > >
> > > > Default values means IP default state after reset.
> > > > If user not aware of the above parameters also still the driver will work
> for
> > > basic functionality.
> > > > Do you want me to implement one more API get_config so that
> Whenever
> > > > user will call the get_config he will know the default values Of the
> > > > config parameters?
> > >
> > > Looking at above I think we do not warrant programming them. Assume
> defaults
> > > in your driver for performance. Like max burst lengths, Max transactions,
> > > without delay scheduling.
> > >
> > > If you disagree, which is fine, please provide the cases where a client
> would
> > > need to program these to different values.
> >
> > As said above there are use cases for SPI and others but currently we didn't
> tested
> > It.
> >
> > >
> > > > > > If user has specific requirement to change these parameters they
> > > > > > can pass It to the driver using set_config API and all these
> > > > > > parameters are Documented in the include/linux/dma/xilinx_dma.h
> file...
> > > > >
> > > > > Can you give me an example where user would like to do that
> > > >
> > > > I am using customized dma test client.
> > > > There I am calling this set_config API before triggering memcpy/SG
> > > operations.
> > >
> > > Well that is precisely a  problem. The generic applications wont know
> "your"
> > > custom API and will not configure that!
> >
> > As said above, generic application can work with default values.
> > These are not custom parameters and we can say these are the dma
> parameters.
> > i.e we can generalize these and provide it to user as generic API.
>
> No it doesn't make it generic API. With generic API the IP should be usable
> with any generic interface without specfic hoops to go thru. People are
> going to test upstream dmatest on your IP and find it not working or
> degraded perf, so better stick with generic API.
>

By default driver configures the optimized settings and it works with upstream
Dmatest. Even with customized settings it works with dmatest only thing it might
Gain/loss performance but that's up to the configuration user may want.

This dma controller was designed to provide more flexibility to the user by providing
the all possible dma parameters configurable.

Regards,
Punnaiah
> --
> ~Vinod


This email and any attachments are intended for the sole use of the named recipient(s) and contain(s) confidential information that may be proprietary, privileged or copyrighted under applicable law. If you are not the intended recipient, do not read, copy, or forward this email message or any attachments. Delete this email message and any attachments immediately.

--
To unsubscribe from this list: send the line "unsubscribe dmaengine" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Vinod Koul June 21, 2016, 4:38 p.m. UTC | #9
On Tue, Jun 21, 2016 at 04:19:38PM +0000, Punnaiah Choudary Kalluri wrote:
> > > > > > > mode Earlier In the driver this configuration is read from the
> > > > > > > device-tree But as per lars and your suggestion moved it as runtime
> > config
> > > > parameters.
> > > > > >
> > > > > > If sg mode is available why would anyone _not_ want it?
> > > > > >
> > > > > > I do not think there is point to have this
> > > > >
> > > > > You mean always keep the device in SG mode and provide an option
> > For
> > > > > simple dma mode if user want to use simple DMA mode??
> > > >
> > > > Yes, why would anyone want to use single if sg is available?
> >
> > If you have sg mode always enabled, but sg_list is size 1, that its
> > essentially a single
> >
> 
> True. As we said, simple mode ha few additional features like write only
> And read only dma. So, if user want then here is the provision.
> 
> > Btw SPI is a slave case, not a memcpy!
> >
> 
> Correct. We are working on slave dma support and will patch to this driver.

And in slave cases, you can use slave config to pass the values which are
required, so we dont need this custome interface!

> 
> > > > > There are few features that are available in the simple DMA mode
> > won't
> > > > > Available in SG mode like write only DMA , read only DMA mode etc...
> > > >
> > > > Can you explain what these are, how they are used by clients?
> > >
> > > Currently it is not implemented but there are cases like,
> > >
> > > We want to initialize the some region with known value, then write only
> > dma helps
> > > We want to read dummy data from the fifo, then read only dma helps.
> >
> > Read will be another transfer, perhpas sg lnegth = 1
> >
> 
> 
> In sg case, both source and destination locations need to be configured.
> In simple dma and read-only mode, only source address is required. Simple
> Dma just read the data from source location and discard that data. It will save
> Unnecessary write to destination here.
> 
> > > Best case is SPI fifo.
> >
> > Nope
> >
> 
> Why?

beacuse you are messing with APIs. SPI will be slave where we know how to
program the values!

> 
> 
> > >
> > > >
> > > > > > > > > +       chan->config.ratectrl = cfg->ratectrl;
> > > > > > > > > +       chan->config.src_issue = cfg->src_issue;
> > > > > > > > > +       chan->config.src_burst_len = cfg->src_burst_len;
> > > > > > > > > +       chan->config.dst_burst_len = cfg->dst_burst_len;
> > > > > > > >
> > > > > > > > can you describe these parameters?
> > > > > > > ratectl:
> > > > > > > Rate control can be independently enabled per channel. When rate
> > > > > > > control is enabled, the DMA channel uses the rate control count to
> > > > > > > schedule
> > > > > > successive data read transactions.
> > > > > >
> > > > > > And how is this used by client?
> > > > >
> > > > > When rate control is enabled, ZDMA channel uses the rate control
> > count
> > > > > To schedule successive data read transactions I mean kind of flow
> > > > > control to schedule Transactions at fixed intervals instead of pumping
> > > > > the transfers without delay or whenever bus is available
> > > > >
> > > > > Rate control count register definition (11:0):
> > > > > Scheduling interval for SRC AXI transaction, only used if rate control
> > > > > is enabled
> > > >
> > > > Okay, why would anyone want transactions at fixed interval. We are
> > talking
> > > > about DMA, so please give me transfers without delay or whenever bus
> > is
> > > > available!
> > >
> > > Could be that there are certain IPs that requires delay in b/w transactions.
> > > Since IP is providing this feature we are exploring it to the user.
> >
> > I kind of don't agree to that!
> >
> 
> It's a kind of rate control mechanism. We can aslo use this feature to prioritize other
> Channel by deliberately controlling the current channel transfer scheduling.
> 
> > >
> > > >
> > > >
> > > > > > > src_issue:
> > > > > > > Tells outstanding transaction on SRC.
> > > > > >
> > > > > > This should be read only then, right?
> > > > >
> > > > > It is a Read/Write register
> > > > > http://www.xilinx.com/html_docs/registers/ug1087/ug1087-zynq-
> > ultrascal
> > > > > e-registers.html By default it is configured for Max transactions.
> > > > > If user want to limit it they can limit it using this config option.
> > > >
> > > > And why would they want that?
> > >
> > > Could be that there are certain IPs that my not support burst operations or
> > > May not support all burst lengths it will be helpful.
> > > Since IP is providing the feature, we are exploring it to the user.
> > >
> > > >
> > > > > > > Burst_len:
> > > > > > > Configures the burst length of the src and dst transfers...
> > > > > >
> > > > > > Hmmm, but you are on memcpy, so that should be programmed for
> > > > throughput?
> > > > >
> > > > > Yes...
> > > >
> > > > So max burst lengths then, why would anyone configure lesser?
> > >
> > > Depends on the destination and source IPs.
> >
> > Not for memory copy
> >
> 
> Depends on the system and how source and destination IPs are interconnected.
> Sometimes there is a limitation from interconnection also.
> Also some flash controllers providing linear access to NAND or NOR memories may
> Have limitation in burst length but still user can use memory copy with desired burst
> Length.

Some of those cases are treated as slave for obvious reasons.

Please check existing solutions before inventing new ones. Everyone thinks
that their hardware and thereby driver is a novel case, but on closer
inspection that is usually not the case, different falvour yes, novel no!

Okay I am convince now this is not right approach. Please remove this
custom interface and then implement slave for required slave scenarios!

> 
> >
> > >
> > > >
> > > > > > > > How would a client know how to configure them?
> > > > > > >
> > > > > > > With the default values of the config parameters driver will work.
> > > > > >
> > > > > > But how will client know what is default!
> > > > >
> > > > > Default values means IP default state after reset.
> > > > > If user not aware of the above parameters also still the driver will work
> > for
> > > > basic functionality.
> > > > > Do you want me to implement one more API get_config so that
> > Whenever
> > > > > user will call the get_config he will know the default values Of the
> > > > > config parameters?
> > > >
> > > > Looking at above I think we do not warrant programming them. Assume
> > defaults
> > > > in your driver for performance. Like max burst lengths, Max transactions,
> > > > without delay scheduling.
> > > >
> > > > If you disagree, which is fine, please provide the cases where a client
> > would
> > > > need to program these to different values.
> > >
> > > As said above there are use cases for SPI and others but currently we didn't
> > tested
> > > It.
> > >
> > > >
> > > > > > > If user has specific requirement to change these parameters they
> > > > > > > can pass It to the driver using set_config API and all these
> > > > > > > parameters are Documented in the include/linux/dma/xilinx_dma.h
> > file...
> > > > > >
> > > > > > Can you give me an example where user would like to do that
> > > > >
> > > > > I am using customized dma test client.
> > > > > There I am calling this set_config API before triggering memcpy/SG
> > > > operations.
> > > >
> > > > Well that is precisely a  problem. The generic applications wont know
> > "your"
> > > > custom API and will not configure that!
> > >
> > > As said above, generic application can work with default values.
> > > These are not custom parameters and we can say these are the dma
> > parameters.
> > > i.e we can generalize these and provide it to user as generic API.
> >
> > No it doesn't make it generic API. With generic API the IP should be usable
> > with any generic interface without specfic hoops to go thru. People are
> > going to test upstream dmatest on your IP and find it not working or
> > degraded perf, so better stick with generic API.
> >
> 
> By default driver configures the optimized settings and it works with upstream
> Dmatest. Even with customized settings it works with dmatest only thing it might
> Gain/loss performance but that's up to the configuration user may want.
> 
> This dma controller was designed to provide more flexibility to the user by providing
> the all possible dma parameters configurable.
> 
> Regards,
> Punnaiah
> > --
> > ~Vinod
> 
> 
> This email and any attachments are intended for the sole use of the named recipient(s) and contain(s) confidential information that may be proprietary, privileged or copyrighted under applicable law. If you are not the intended recipient, do not read, copy, or forward this email message or any attachments. Delete this email message and any attachments immediately.

Really!
Punnaiah Choudary Kalluri June 21, 2016, 5:29 p.m. UTC | #10
> On Tue, Jun 21, 2016 at 04:19:38PM +0000, Punnaiah Choudary Kalluri wrote:
> > > > > > > > mode Earlier In the driver this configuration is read from the
> > > > > > > > device-tree But as per lars and your suggestion moved it as
> runtime
> > > config
> > > > > parameters.
> > > > > > >
> > > > > > > If sg mode is available why would anyone _not_ want it?
> > > > > > >
> > > > > > > I do not think there is point to have this
> > > > > >
> > > > > > You mean always keep the device in SG mode and provide an
> option
> > > For
> > > > > > simple dma mode if user want to use simple DMA mode??
> > > > >
> > > > > Yes, why would anyone want to use single if sg is available?
> > >
> > > If you have sg mode always enabled, but sg_list is size 1, that its
> > > essentially a single
> > >
> >
> > True. As we said, simple mode ha few additional features like write only
> > And read only dma. So, if user want then here is the provision.
> >
> > > Btw SPI is a slave case, not a memcpy!
> > >
> >
> > Correct. We are working on slave dma support and will patch to this driver.
> 
> And in slave cases, you can use slave config to pass the values which are
> required, so we dont need this custome interface!
> 

Ok agree with you for the scenario that I have mentioned above.

Other simple dma mode feature that I missed to explain is configuring the 
Dma descriptors. It provides a register interface for configuring the dma transaction.
So, no need to maintain the descriptors in memory and it will be useful for the system 
that are in crunch of memory.  

How do you want us to handle this case?

> > > >
> > > > >
> > > > >
> > > > > > > > src_issue:
> > > > > > > > Tells outstanding transaction on SRC.
> > > > > > >
> > > > > > > This should be read only then, right?
> > > > > >
> > > > > > It is a Read/Write register
> > > > > > http://www.xilinx.com/html_docs/registers/ug1087/ug1087-zynq-
> > > ultrascal
> > > > > > e-registers.html By default it is configured for Max transactions.
> > > > > > If user want to limit it they can limit it using this config option.
> > > > >
> > > > > And why would they want that?
> > > >
> > > > Could be that there are certain IPs that my not support burst operations
> or
> > > > May not support all burst lengths it will be helpful.
> > > > Since IP is providing the feature, we are exploring it to the user.
> > > >
> > > > >
> > > > > > > > Burst_len:
> > > > > > > > Configures the burst length of the src and dst transfers...
> > > > > > >
> > > > > > > Hmmm, but you are on memcpy, so that should be programmed
> for
> > > > > throughput?
> > > > > >
> > > > > > Yes...
> > > > >
> > > > > So max burst lengths then, why would anyone configure lesser?
> > > >
> > > > Depends on the destination and source IPs.
> > >
> > > Not for memory copy
> > >
> >
> > Depends on the system and how source and destination IPs are
> interconnected.
> > Sometimes there is a limitation from interconnection also.
> > Also some flash controllers providing linear access to NAND or NOR
> memories may
> > Have limitation in burst length but still user can use memory copy with
> desired burst
> > Length.
> 
> Some of those cases are treated as slave for obvious reasons.
> 
> Please check existing solutions before inventing new ones. Everyone thinks
> that their hardware and thereby driver is a novel case, but on closer
> inspection that is usually not the case, different falvour yes, novel no!
> 
> Okay I am convince now this is not right approach. Please remove this
> custom interface and then implement slave for required slave scenarios!
>

Assume controller is having 8 channels and four of them are used for slave
Dma and others are for memcpy.
Controller didn't have the per channel priority control but providing the rate control
Mechanism. 

So, I need some interface for configuring the rate control per channel at run time irrespective
Of whether the channel is allocated for slave dma or memcpy dma.

Is it wrong having the configurable dma parameters for dma memcpy operation? We are exposing the
Hw capabilities to the user for better dma transaction management.

> >
> >
> > This email and any attachments are intended for the sole use of the named
> recipient(s) and contain(s) confidential information that may be proprietary,
> privileged or copyrighted under applicable law. If you are not the intended
> recipient, do not read, copy, or forward this email message or any
> attachments. Delete this email message and any attachments immediately.
> 
> Really!
> 

My apologies :)

Thanks,
Punnaiah
> 
> --
> ~Vinod
--
To unsubscribe from this list: send the line "unsubscribe dmaengine" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Vinod Koul June 28, 2016, 4:14 a.m. UTC | #11
On Tue, Jun 21, 2016 at 05:29:42PM +0000, Punnaiah Choudary Kalluri wrote:
> 
> Ok agree with you for the scenario that I have mentioned above.
> 
> Other simple dma mode feature that I missed to explain is configuring the 
> Dma descriptors. It provides a register interface for configuring the dma transaction.
> So, no need to maintain the descriptors in memory and it will be useful for the system 
> that are in crunch of memory.  

And why are these not coming out in the first place, which makes me think it
is fishy!

Do you mean programming DMA descriptors to hardware and you can use
registers instead of memory?

> 
> How do you want us to handle this case?
> 

> > Okay I am convince now this is not right approach. Please remove this
> > custom interface and then implement slave for required slave scenarios!
> >
> 
> Assume controller is having 8 channels and four of them are used for slave
> Dma and others are for memcpy.
> Controller didn't have the per channel priority control but providing the rate control
> Mechanism. 

How does the use of few for memcpy and few for slave change things? IMO it
doesn't, you program the channel accordingly

> 
> So, I need some interface for configuring the rate control per channel at run time irrespective
> Of whether the channel is allocated for slave dma or memcpy dma.

why?

> Is it wrong having the configurable dma parameters for dma memcpy operation? We are exposing the
> Hw capabilities to the user for better dma transaction management.

For memcpy yes. Memcpy is a generic case where people do not do driver
specific stuff. So I tend ot push back on that..

For slave, existing APIs allow you to program the additional parameters..
FWIW, rate control is a generic parameter which if you justify enough can be
added to dma_slave_config

Thanks
Punnaiah Choudary Kalluri June 29, 2016, 3:59 a.m. UTC | #12
> -----Original Message-----
> From: Vinod Koul [mailto:vinod.koul@intel.com]
> Sent: Tuesday, June 28, 2016 9:45 AM
> To: Punnaiah Choudary Kalluri <punnaia@xilinx.com>
> Cc: Appana Durga Kedareswara Rao <appanad@xilinx.com>;
> robh+dt@kernel.org; pawel.moll@arm.com; mark.rutland@arm.com;
> ijc+devicetree@hellion.org.uk; galak@codeaurora.org; Michal Simek
> <michals@xilinx.com>; Soren Brinkmann <sorenb@xilinx.com>;
> dan.j.williams@intel.com; moritz.fischer@ettus.com;
> laurent.pinchart@ideasonboard.com; luis@debethencourt.com; Srikanth
> Vemula <svemula@xilinx.com>; Anirudha Sarangi <anirudh@xilinx.com>;
> devicetree@vger.kernel.org; linux-arm-kernel@lists.infradead.org; linux-
> kernel@vger.kernel.org; dmaengine@vger.kernel.org
> Subject: Re: [PATCH v10 2/2] dmaengine: Add Xilinx zynqmp dma engine
> driver support
> 
> On Tue, Jun 21, 2016 at 05:29:42PM +0000, Punnaiah Choudary Kalluri wrote:
> >
> > Ok agree with you for the scenario that I have mentioned above.
> >
> > Other simple dma mode feature that I missed to explain is configuring the
> > Dma descriptors. It provides a register interface for configuring the dma
> transaction.
> > So, no need to maintain the descriptors in memory and it will be useful for
> the system
> > that are in crunch of memory.
> 
> And why are these not coming out in the first place, which makes me think it
> is fishy!
> 
> Do you mean programming DMA descriptors to hardware and you can use
> registers instead of memory?
> 

Yes.

> >
> > How do you want us to handle this case?
> >
> 
> > > Okay I am convince now this is not right approach. Please remove this
> > > custom interface and then implement slave for required slave scenarios!
> > >
> >
> > Assume controller is having 8 channels and four of them are used for slave
> > Dma and others are for memcpy.
> > Controller didn't have the per channel priority control but providing the
> rate control
> > Mechanism.
> 
> How does the use of few for memcpy and few for slave change things? IMO
> it
> doesn't, you program the channel accordingly
> 
> >
> > So, I need some interface for configuring the rate control per channel at
> run time irrespective
> > Of whether the channel is allocated for slave dma or memcpy dma.
> 
> why?
> 

This is to prioritize the channel over other channels at runtime.
Also, if the slave device doesn't have a flow control implemented,
Then rate control is one mechanism for controlling the transactions 
Between the source and destination.

> > Is it wrong having the configurable dma parameters for dma memcpy
> operation? We are exposing the
> > Hw capabilities to the user for better dma transaction management.
> 
> For memcpy yes. Memcpy is a generic case where people do not do driver
> specific stuff. So I tend ot push back on that..
> 

Ok. Then we will consider using slave dma if the memcpy requires custom
Settings (the settings might be for debug purpose or there is real hw design
that mandates changes in default optimized settings).

> For slave, existing APIs allow you to program the additional parameters..
> FWIW, rate control is a generic parameter which if you justify enough can be
> added to dma_slave_config
> 

As said above, rate control will be helpful for the controller that doesn't have 
Per channel priority option and also cases where slave device/controller that
Doesn't have Flow control implemented.

Thanks,
Punnaiah

> Thanks
> --
> ~Vinod
--
To unsubscribe from this list: send the line "unsubscribe dmaengine" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
diff mbox

Patch

diff --git a/drivers/dma/Kconfig b/drivers/dma/Kconfig
index 8c98779..629e339 100644
--- a/drivers/dma/Kconfig
+++ b/drivers/dma/Kconfig
@@ -541,6 +541,12 @@  config ZX_DMA
 	help
 	  Support the DMA engine for ZTE ZX296702 platform devices.
 
+config XILINX_ZYNQMP_DMA
+	tristate "Xilinx ZynqMP DMA Engine"
+	depends on (ARCH_ZYNQ || MICROBLAZE || ARM64)
+	select DMA_ENGINE
+	help
+	  Enable support for Xilinx ZynqMP DMA controller.
 
 # driver files
 source "drivers/dma/bestcomm/Kconfig"
diff --git a/drivers/dma/xilinx/Makefile b/drivers/dma/xilinx/Makefile
index 3c4e9f2..95469dc 100644
--- a/drivers/dma/xilinx/Makefile
+++ b/drivers/dma/xilinx/Makefile
@@ -1 +1,2 @@ 
 obj-$(CONFIG_XILINX_VDMA) += xilinx_vdma.o
+obj-$(CONFIG_XILINX_ZYNQMP_DMA) += zynqmp_dma.o
diff --git a/drivers/dma/xilinx/zynqmp_dma.c b/drivers/dma/xilinx/zynqmp_dma.c
new file mode 100644
index 0000000..dd05eff
--- /dev/null
+++ b/drivers/dma/xilinx/zynqmp_dma.c
@@ -0,0 +1,1269 @@ 
+/*
+ * DMA driver for Xilinx ZynqMP DMA Engine
+ *
+ * Copyright (C) 2016 Xilinx, Inc. All rights reserved.
+ *
+ * This program is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/bitops.h>
+#include <linux/dma-mapping.h>
+#include <linux/dma/xilinx_dma.h>
+#include <linux/dmaengine.h>
+#include <linux/dmapool.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/module.h>
+#include <linux/of_address.h>
+#include <linux/of_dma.h>
+#include <linux/of_irq.h>
+#include <linux/of_platform.h>
+#include <linux/slab.h>
+#include <linux/clk.h>
+#include <linux/io-64-nonatomic-lo-hi.h>
+
+#include "../dmaengine.h"
+
+/* Register Offsets */
+#define ZYNQMP_DMA_ISR			0x100
+#define ZYNQMP_DMA_IMR			0x104
+#define ZYNQMP_DMA_IER			0x108
+#define ZYNQMP_DMA_IDS			0x10C
+#define ZYNQMP_DMA_CTRL0		0x110
+#define ZYNQMP_DMA_CTRL1		0x114
+#define ZYNQMP_DMA_DATA_ATTR		0x120
+#define ZYNQMP_DMA_DSCR_ATTR		0x124
+#define ZYNQMP_DMA_SRC_DSCR_WRD0	0x128
+#define ZYNQMP_DMA_SRC_DSCR_WRD1	0x12C
+#define ZYNQMP_DMA_SRC_DSCR_WRD2	0x130
+#define ZYNQMP_DMA_SRC_DSCR_WRD3	0x134
+#define ZYNQMP_DMA_DST_DSCR_WRD0	0x138
+#define ZYNQMP_DMA_DST_DSCR_WRD1	0x13C
+#define ZYNQMP_DMA_DST_DSCR_WRD2	0x140
+#define ZYNQMP_DMA_DST_DSCR_WRD3	0x144
+#define ZYNQMP_DMA_SRC_START_LSB	0x158
+#define ZYNQMP_DMA_SRC_START_MSB	0x15C
+#define ZYNQMP_DMA_DST_START_LSB	0x160
+#define ZYNQMP_DMA_DST_START_MSB	0x164
+#define ZYNQMP_DMA_RATE_CTRL		0x18C
+#define ZYNQMP_DMA_IRQ_SRC_ACCT		0x190
+#define ZYNQMP_DMA_IRQ_DST_ACCT		0x194
+#define ZYNQMP_DMA_CTRL2		0x200
+
+/* Interrupt registers bit field definitions */
+#define ZYNQMP_DMA_DONE			BIT(10)
+#define ZYNQMP_DMA_AXI_WR_DATA		BIT(9)
+#define ZYNQMP_DMA_AXI_RD_DATA		BIT(8)
+#define ZYNQMP_DMA_AXI_RD_DST_DSCR	BIT(7)
+#define ZYNQMP_DMA_AXI_RD_SRC_DSCR	BIT(6)
+#define ZYNQMP_DMA_IRQ_DST_ACCT_ERR	BIT(5)
+#define ZYNQMP_DMA_IRQ_SRC_ACCT_ERR	BIT(4)
+#define ZYNQMP_DMA_BYTE_CNT_OVRFL	BIT(3)
+#define ZYNQMP_DMA_DST_DSCR_DONE	BIT(2)
+#define ZYNQMP_DMA_INV_APB		BIT(0)
+
+/* Control 0 register bit field definitions */
+#define ZYNQMP_DMA_OVR_FETCH		BIT(7)
+#define ZYNQMP_DMA_POINT_TYPE_SG	BIT(6)
+#define ZYNQMP_DMA_RATE_CTRL_EN		BIT(3)
+
+/* Control 1 register bit field definitions */
+#define ZYNQMP_DMA_SRC_ISSUE		GENMASK(4, 0)
+
+/* Data Attribute register bit field definitions */
+#define ZYNQMP_DMA_ARBURST		GENMASK(27, 26)
+#define ZYNQMP_DMA_ARCACHE		GENMASK(25, 22)
+#define ZYNQMP_DMA_ARCACHE_OFST		22
+#define ZYNQMP_DMA_ARQOS		GENMASK(21, 18)
+#define ZYNQMP_DMA_ARQOS_OFST		18
+#define ZYNQMP_DMA_ARLEN		GENMASK(17, 14)
+#define ZYNQMP_DMA_ARLEN_OFST		14
+#define ZYNQMP_DMA_AWBURST		GENMASK(13, 12)
+#define ZYNQMP_DMA_AWCACHE		GENMASK(11, 8)
+#define ZYNQMP_DMA_AWCACHE_OFST		8
+#define ZYNQMP_DMA_AWQOS		GENMASK(7, 4)
+#define ZYNQMP_DMA_AWQOS_OFST		4
+#define ZYNQMP_DMA_AWLEN		GENMASK(3, 0)
+#define ZYNQMP_DMA_AWLEN_OFST		0
+
+/* Descriptor Attribute register bit field definitions */
+#define ZYNQMP_DMA_AXCOHRNT		BIT(8)
+#define ZYNQMP_DMA_AXCACHE		GENMASK(7, 4)
+#define ZYNQMP_DMA_AXCACHE_OFST		4
+#define ZYNQMP_DMA_AXQOS		GENMASK(3, 0)
+#define ZYNQMP_DMA_AXQOS_OFST		0
+
+/* Control register 2 bit field definitions */
+#define ZYNQMP_DMA_ENABLE		BIT(0)
+
+/* Buffer Descriptor definitions */
+#define ZYNQMP_DMA_DESC_CTRL_STOP	0x10
+#define ZYNQMP_DMA_DESC_CTRL_COMP_INT	0x4
+#define ZYNQMP_DMA_DESC_CTRL_SIZE_256	0x2
+#define ZYNQMP_DMA_DESC_CTRL_COHRNT	0x1
+
+/* Interrupt Mask specific definitions */
+#define ZYNQMP_DMA_INT_ERR	(ZYNQMP_DMA_AXI_RD_DATA | \
+				ZYNQMP_DMA_AXI_WR_DATA | \
+				ZYNQMP_DMA_AXI_RD_DST_DSCR | \
+				ZYNQMP_DMA_AXI_RD_SRC_DSCR | \
+				ZYNQMP_DMA_INV_APB)
+#define ZYNQMP_DMA_INT_OVRFL	(ZYNQMP_DMA_BYTE_CNT_OVRFL | \
+				ZYNQMP_DMA_IRQ_SRC_ACCT_ERR | \
+				ZYNQMP_DMA_IRQ_DST_ACCT_ERR)
+#define ZYNQMP_DMA_INT_DONE	(ZYNQMP_DMA_DONE | ZYNQMP_DMA_DST_DSCR_DONE)
+#define ZYNQMP_DMA_INT_EN_DEFAULT_MASK	(ZYNQMP_DMA_INT_DONE | \
+					ZYNQMP_DMA_INT_ERR | \
+					ZYNQMP_DMA_INT_OVRFL | \
+					ZYNQMP_DMA_DST_DSCR_DONE)
+
+/* Max number of descriptors per channel */
+#define ZYNQMP_DMA_NUM_DESCS	32
+
+/* Max transfer size per descriptor */
+#define ZYNQMP_DMA_MAX_TRANS_LEN	0x40000000
+
+/* Reset values for data attributes */
+#define ZYNQMP_DMA_AXCACHE_VAL		0xF
+#define ZYNQMP_DMA_ARLEN_RST_VAL	0xF
+#define ZYNQMP_DMA_AWLEN_RST_VAL	0xF
+
+#define ZYNQMP_DMA_SRC_ISSUE_RST_VAL	0x1F
+
+#define ZYNQMP_DMA_IDS_DEFAULT_MASK	0xFFF
+
+/* Bus width in bits */
+#define ZYNQMP_DMA_BUS_WIDTH_64		64
+#define ZYNQMP_DMA_BUS_WIDTH_128	128
+
+#define ZYNQMP_DMA_DESC_SIZE(chan)	(chan->desc_size)
+
+#define to_chan(chan)		container_of(chan, struct zynqmp_dma_chan, \
+					     common)
+#define tx_to_desc(tx)		container_of(tx, struct zynqmp_dma_desc_sw, \
+					     async_tx)
+
+/**
+ * struct zynqmp_dma_desc_ll - Hw linked list descriptor
+ * @addr: Buffer address
+ * @size: Size of the buffer
+ * @ctrl: Control word
+ * @nxtdscraddr: Next descriptor base address
+ * @rsvd: Reserved field and for Hw internal use.
+ */
+struct zynqmp_dma_desc_ll {
+	u64 addr;
+	u32 size;
+	u32 ctrl;
+	u64 nxtdscraddr;
+	u64 rsvd;
+}; __aligned(64)
+
+/**
+ * struct zynqmp_dma_desc_sw - Per Transaction structure
+ * @src: Source address for simple mode dma
+ * @dst: Destination address for simple mode dma
+ * @len: Transfer length for simple mode dma
+ * @node: Node in the channel descriptor list
+ * @tx_list: List head for the current transfer
+ * @async_tx: Async transaction descriptor
+ * @src_v: Virtual address of the src descriptor
+ * @src_p: Physical address of the src descriptor
+ * @dst_v: Virtual address of the dst descriptor
+ * @dst_p: Physical address of the dst descriptor
+ */
+struct zynqmp_dma_desc_sw {
+	u64 src;
+	u64 dst;
+	u32 len;
+	struct list_head node;
+	struct list_head tx_list;
+	struct dma_async_tx_descriptor async_tx;
+	struct zynqmp_dma_desc_ll *src_v;
+	dma_addr_t src_p;
+	struct zynqmp_dma_desc_ll *dst_v;
+	dma_addr_t dst_p;
+};
+
+/**
+ * struct zynqmp_dma_chan - Driver specific DMA channel structure
+ * @zdev: Driver specific device structure
+ * @regs: Control registers offset
+ * @lock: Descriptor operation lock
+ * @pending_list: Descriptors waiting
+ * @free_list: Descriptors free
+ * @active_list: Descriptors active
+ * @sw_desc_pool: SW descriptor pool
+ * @done_list: Complete descriptors
+ * @common: DMA common channel
+ * @desc_pool_v: Statically allocated descriptor base
+ * @desc_pool_p: Physical allocated descriptor base
+ * @desc_free_cnt: Descriptor available count
+ * @dev: The dma device
+ * @irq: Channel IRQ
+ * @is_dmacoherent: Tells whether dma operations are coherent or not
+ * @tasklet: Cleanup work after irq
+ * @idle : Channel status;
+ * @desc_size: Size of the low level descriptor
+ * @err: Channel has errors
+ * @bus_width: Bus width
+ * @clk_main: Pointer to main clock
+ * @clk_apb: Pointer to apb clock
+ * @config: Device configuration info
+ */
+struct zynqmp_dma_chan {
+	struct zynqmp_dma_device *zdev;
+	void __iomem *regs;
+	spinlock_t lock;
+	struct list_head pending_list;
+	struct list_head free_list;
+	struct list_head active_list;
+	struct zynqmp_dma_desc_sw *sw_desc_pool;
+	struct list_head done_list;
+	struct dma_chan common;
+	void *desc_pool_v;
+	dma_addr_t desc_pool_p;
+	u32 desc_free_cnt;
+	struct device *dev;
+	int irq;
+	bool is_dmacoherent;
+	struct tasklet_struct tasklet;
+	bool idle;
+	u32 desc_size;
+	bool err;
+	u32 bus_width;
+	struct clk *clk_main;
+	struct clk *clk_apb;
+	struct zynqmp_dma_config config;
+};
+
+/**
+ * struct zynqmp_dma_device - DMA device structure
+ * @dev: Device Structure
+ * @common: DMA device structure
+ * @chan: Driver specific DMA channel
+ */
+struct zynqmp_dma_device {
+	struct device *dev;
+	struct dma_device common;
+	struct zynqmp_dma_chan *chan;
+};
+
+static inline void zynqmp_dma_writeq(struct zynqmp_dma_chan *chan, u32 reg,
+				     u64 value)
+{
+#if defined(CONFIG_PHYS_ADDR_T_64BIT)
+	writeq(value, chan->regs + reg);
+#else
+	lo_hi_writeq(value, chan->regs + reg);
+#endif
+}
+
+/**
+ * zynqmp_dma_chan_is_idle - Provides the channel idle status
+ * @chan: ZynqMP DMA DMA channel pointer
+ *
+ * Return: 'true' if the channel is idle otherwise 'false'
+ */
+static inline bool zynqmp_dma_chan_is_idle(struct zynqmp_dma_chan *chan)
+{
+	return chan->idle;
+
+}
+
+/**
+ * zynqmp_dma_update_desc_to_ctrlr - Updates descriptor to the controller
+ * @chan: ZynqMP DMA DMA channel pointer
+ * @desc: Transaction descriptor pointer
+ */
+static void zynqmp_dma_update_desc_to_ctrlr(struct zynqmp_dma_chan *chan,
+				      struct zynqmp_dma_desc_sw *desc)
+{
+	dma_addr_t addr;
+
+	addr = desc->src_p;
+	zynqmp_dma_writeq(chan, ZYNQMP_DMA_SRC_START_LSB, addr);
+	addr = desc->dst_p;
+	zynqmp_dma_writeq(chan, ZYNQMP_DMA_DST_START_LSB, addr);
+}
+
+/**
+ * zynqmp_dma_desc_config_eod - Mark the descriptor as end descriptor
+ * @chan: ZynqMP DMA channel pointer
+ * @desc: Hw descriptor pointer
+ */
+static void zynqmp_dma_desc_config_eod(struct zynqmp_dma_chan *chan, void *desc)
+{
+	struct zynqmp_dma_desc_ll *hw = (struct zynqmp_dma_desc_ll *)desc;
+
+	hw->ctrl |= ZYNQMP_DMA_DESC_CTRL_STOP;
+	hw++;
+	hw->ctrl |= ZYNQMP_DMA_DESC_CTRL_COMP_INT | ZYNQMP_DMA_DESC_CTRL_STOP;
+}
+
+/**
+ * zynqmp_dma_config_simple_desc - Configure the transfer parameters
+ * @chan: ZynqMP DMA channel pointer
+ * @src: Source buffer address
+ * @dst: Destination buffer address
+ * @len: Transfer length
+ */
+static void zynqmp_dma_config_simple_desc(struct zynqmp_dma_chan *chan,
+					  dma_addr_t src, dma_addr_t dst,
+					  size_t len)
+{
+	u32 val;
+
+	zynqmp_dma_writeq(chan, ZYNQMP_DMA_SRC_DSCR_WRD0, src);
+	writel(len, chan->regs + ZYNQMP_DMA_SRC_DSCR_WRD2);
+
+	if (chan->is_dmacoherent)
+		writel(ZYNQMP_DMA_DESC_CTRL_COHRNT,
+			chan->regs + ZYNQMP_DMA_SRC_DSCR_WRD3);
+	else
+		writel(0, chan->regs + ZYNQMP_DMA_SRC_DSCR_WRD3);
+
+	zynqmp_dma_writeq(chan, ZYNQMP_DMA_DST_DSCR_WRD0, dst);
+	writel(len, chan->regs + ZYNQMP_DMA_DST_DSCR_WRD2);
+
+	if (chan->is_dmacoherent)
+		val = ZYNQMP_DMA_DESC_CTRL_COHRNT |
+				ZYNQMP_DMA_DESC_CTRL_COMP_INT;
+	else
+		val = ZYNQMP_DMA_DESC_CTRL_COMP_INT;
+	writel(val, chan->regs + ZYNQMP_DMA_DST_DSCR_WRD3);
+}
+
+/**
+ * zynqmp_dma_config_sg_ll_desc - Configure the linked list descriptor
+ * @chan: ZynqMP DMA channel pointer
+ * @sdesc: Hw descriptor pointer
+ * @src: Source buffer address
+ * @dst: Destination buffer address
+ * @len: Transfer length
+ * @prev: Previous hw descriptor pointer
+ */
+static void zynqmp_dma_config_sg_ll_desc(struct zynqmp_dma_chan *chan,
+				   struct zynqmp_dma_desc_ll *sdesc,
+				   dma_addr_t src, dma_addr_t dst, size_t len,
+				   struct zynqmp_dma_desc_ll *prev)
+{
+	struct zynqmp_dma_desc_ll *ddesc = sdesc + 1;
+
+	sdesc->size = ddesc->size = len;
+	sdesc->addr = src;
+	ddesc->addr = dst;
+
+	sdesc->ctrl = ddesc->ctrl = ZYNQMP_DMA_DESC_CTRL_SIZE_256;
+	if (chan->is_dmacoherent) {
+		sdesc->ctrl |= ZYNQMP_DMA_DESC_CTRL_COHRNT;
+		ddesc->ctrl |= ZYNQMP_DMA_DESC_CTRL_COHRNT;
+	}
+
+	if (prev) {
+		dma_addr_t addr = chan->desc_pool_p +
+			    ((dma_addr_t)sdesc - (dma_addr_t)chan->desc_pool_v);
+		ddesc = prev + 1;
+		prev->nxtdscraddr = addr;
+		ddesc->nxtdscraddr = addr + ZYNQMP_DMA_DESC_SIZE(chan);
+	}
+}
+
+/**
+ * zynqmp_dma_init - Initialize the channel
+ * @chan: ZynqMP DMA channel pointer
+ */
+static void zynqmp_dma_init(struct zynqmp_dma_chan *chan)
+{
+	u32 val;
+
+	writel(ZYNQMP_DMA_IDS_DEFAULT_MASK, chan->regs + ZYNQMP_DMA_IDS);
+	val = readl(chan->regs + ZYNQMP_DMA_ISR);
+	writel(val, chan->regs + ZYNQMP_DMA_ISR);
+
+	val = 0;
+	if (chan->config.has_sg)
+		val |= ZYNQMP_DMA_POINT_TYPE_SG;
+	writel(val, chan->regs + ZYNQMP_DMA_CTRL0);
+
+	val = 0;
+	if (chan->is_dmacoherent) {
+		val |= ZYNQMP_DMA_AXCOHRNT;
+		val = (val & ~ZYNQMP_DMA_AXCACHE) |
+			(ZYNQMP_DMA_AXCACHE_VAL << ZYNQMP_DMA_AXCACHE_OFST);
+	}
+	writel(val, chan->regs + ZYNQMP_DMA_DSCR_ATTR);
+
+	val = readl(chan->regs + ZYNQMP_DMA_DATA_ATTR);
+	if (chan->is_dmacoherent) {
+		val = (val & ~ZYNQMP_DMA_ARCACHE) |
+			(ZYNQMP_DMA_AXCACHE_VAL << ZYNQMP_DMA_ARCACHE_OFST);
+		val = (val & ~ZYNQMP_DMA_AWCACHE) |
+			(ZYNQMP_DMA_AXCACHE_VAL << ZYNQMP_DMA_AWCACHE_OFST);
+	}
+	writel(val, chan->regs + ZYNQMP_DMA_DATA_ATTR);
+
+	/* Clearing the interrupt account rgisters */
+	val = readl(chan->regs + ZYNQMP_DMA_IRQ_SRC_ACCT);
+	val = readl(chan->regs + ZYNQMP_DMA_IRQ_DST_ACCT);
+
+	chan->idle = true;
+}
+
+/**
+ * zynqmp_dma_tx_submit - Submit DMA transaction
+ * @tx: Async transaction descriptor pointer
+ *
+ * Return: cookie value
+ */
+static dma_cookie_t zynqmp_dma_tx_submit(struct dma_async_tx_descriptor *tx)
+{
+	struct zynqmp_dma_chan *chan = to_chan(tx->chan);
+	struct zynqmp_dma_desc_sw *desc, *new;
+	dma_cookie_t cookie;
+
+	new = tx_to_desc(tx);
+	spin_lock_bh(&chan->lock);
+	cookie = dma_cookie_assign(tx);
+
+	if (!list_empty(&chan->pending_list) && chan->config.has_sg) {
+		desc = list_last_entry(&chan->pending_list,
+				     struct zynqmp_dma_desc_sw, node);
+		if (!list_empty(&desc->tx_list))
+			desc = list_last_entry(&desc->tx_list,
+					       struct zynqmp_dma_desc_sw, node);
+		desc->src_v->nxtdscraddr = new->src_p;
+		desc->src_v->ctrl &= ~ZYNQMP_DMA_DESC_CTRL_STOP;
+		desc->dst_v->nxtdscraddr = new->dst_p;
+		desc->dst_v->ctrl &= ~ZYNQMP_DMA_DESC_CTRL_STOP;
+	}
+
+	list_add_tail(&new->node, &chan->pending_list);
+	spin_unlock_bh(&chan->lock);
+
+	return cookie;
+}
+
+/**
+ * zynqmp_dma_get_descriptor - Get the sw descriptor from the pool
+ * @chan: ZynqMP DMA channel pointer
+ *
+ * Return: The sw descriptor
+ */
+static struct zynqmp_dma_desc_sw *
+zynqmp_dma_get_descriptor(struct zynqmp_dma_chan *chan)
+{
+	struct zynqmp_dma_desc_sw *desc;
+
+	spin_lock_bh(&chan->lock);
+	desc = list_first_entry(&chan->free_list, struct zynqmp_dma_desc_sw,
+				 node);
+	list_del(&desc->node);
+	spin_unlock_bh(&chan->lock);
+
+	INIT_LIST_HEAD(&desc->tx_list);
+	/* Clear the src and dst descriptor memory */
+	if (chan->config.has_sg) {
+		memset((void *)desc->src_v, 0, ZYNQMP_DMA_DESC_SIZE(chan));
+		memset((void *)desc->dst_v, 0, ZYNQMP_DMA_DESC_SIZE(chan));
+	}
+
+	return desc;
+}
+
+/**
+ * zynqmp_dma_free_descriptor - Issue pending transactions
+ * @chan: ZynqMP DMA channel pointer
+ * @sdesc: Transaction descriptor pointer
+ */
+static void zynqmp_dma_free_descriptor(struct zynqmp_dma_chan *chan,
+				 struct zynqmp_dma_desc_sw *sdesc)
+{
+	struct zynqmp_dma_desc_sw *child, *next;
+
+	chan->desc_free_cnt++;
+	list_add_tail(&sdesc->node, &chan->free_list);
+	list_for_each_entry_safe(child, next, &sdesc->tx_list, node) {
+		chan->desc_free_cnt++;
+		INIT_LIST_HEAD(&child->tx_list);
+		list_move_tail(&child->node, &chan->free_list);
+	}
+	INIT_LIST_HEAD(&sdesc->tx_list);
+}
+
+/**
+ * zynqmp_dma_free_desc_list - Free descriptors list
+ * @chan: ZynqMP DMA channel pointer
+ * @list: List to parse and delete the descriptor
+ */
+static void zynqmp_dma_free_desc_list(struct zynqmp_dma_chan *chan,
+				      struct list_head *list)
+{
+	struct zynqmp_dma_desc_sw *desc, *next;
+
+	list_for_each_entry_safe(desc, next, list, node)
+		zynqmp_dma_free_descriptor(chan, desc);
+}
+
+/**
+ * zynqmp_dma_alloc_chan_resources - Allocate channel resources
+ * @dchan: DMA channel
+ *
+ * Return: Number of descriptors on success and failure value on error
+ */
+static int zynqmp_dma_alloc_chan_resources(struct dma_chan *dchan)
+{
+	struct zynqmp_dma_chan *chan = to_chan(dchan);
+	struct zynqmp_dma_desc_sw *desc;
+	int i;
+
+	chan->sw_desc_pool = kzalloc(sizeof(*desc) * ZYNQMP_DMA_NUM_DESCS,
+				     GFP_KERNEL);
+	if (!chan->sw_desc_pool)
+		return -ENOMEM;
+	chan->idle = true;
+	chan->desc_free_cnt = ZYNQMP_DMA_NUM_DESCS;
+
+	INIT_LIST_HEAD(&chan->free_list);
+
+	for (i = 0; i < ZYNQMP_DMA_NUM_DESCS; i++) {
+		desc = chan->sw_desc_pool + i;
+		dma_async_tx_descriptor_init(&desc->async_tx, &chan->common);
+		desc->async_tx.tx_submit = zynqmp_dma_tx_submit;
+		list_add_tail(&desc->node, &chan->free_list);
+	}
+
+	chan->desc_pool_v = dma_zalloc_coherent(chan->dev,
+				(2 * chan->desc_size * ZYNQMP_DMA_NUM_DESCS),
+				&chan->desc_pool_p, GFP_KERNEL);
+	if (!chan->desc_pool_v)
+		return -ENOMEM;
+
+	for (i = 0; i < ZYNQMP_DMA_NUM_DESCS; i++) {
+		desc = chan->sw_desc_pool + i;
+		desc->src_v = (struct zynqmp_dma_desc_ll *) (chan->desc_pool_v +
+					(i * ZYNQMP_DMA_DESC_SIZE(chan) * 2));
+		desc->dst_v = (struct zynqmp_dma_desc_ll *) (desc->src_v + 1);
+		desc->src_p = chan->desc_pool_p +
+				(i * ZYNQMP_DMA_DESC_SIZE(chan) * 2);
+		desc->dst_p = desc->src_p + ZYNQMP_DMA_DESC_SIZE(chan);
+	}
+
+	return ZYNQMP_DMA_NUM_DESCS;
+}
+
+/**
+ * zynqmp_dma_start - Start DMA channel
+ * @chan: ZynqMP DMA channel pointer
+ */
+static void zynqmp_dma_start(struct zynqmp_dma_chan *chan)
+{
+	writel(ZYNQMP_DMA_INT_EN_DEFAULT_MASK, chan->regs + ZYNQMP_DMA_IER);
+	chan->idle = false;
+	writel(ZYNQMP_DMA_ENABLE, chan->regs + ZYNQMP_DMA_CTRL2);
+}
+
+/**
+ * zynqmp_dma_handle_ovfl_int - Process the overflow interrupt
+ * @chan: ZynqMP DMA channel pointer
+ * @status: Interrupt status value
+ */
+static void zynqmp_dma_handle_ovfl_int(struct zynqmp_dma_chan *chan, u32 status)
+{
+	u32 val;
+
+	if (status & ZYNQMP_DMA_IRQ_DST_ACCT_ERR)
+		val = readl(chan->regs + ZYNQMP_DMA_IRQ_DST_ACCT);
+	if (status & ZYNQMP_DMA_IRQ_SRC_ACCT_ERR)
+		val = readl(chan->regs + ZYNQMP_DMA_IRQ_SRC_ACCT);
+}
+
+static void zynqmp_dma_config(struct zynqmp_dma_chan *chan)
+{
+	u32 val;
+
+	val = readl(chan->regs + ZYNQMP_DMA_CTRL0);
+	if (chan->config.ovrfetch)
+		val |= ZYNQMP_DMA_OVR_FETCH;
+	if (chan->config.ratectrl) {
+		val |= ZYNQMP_DMA_RATE_CTRL_EN;
+		writel(chan->config.ratectrl, chan->regs +
+		       ZYNQMP_DMA_RATE_CTRL);
+	}
+	writel(val, chan->regs + ZYNQMP_DMA_CTRL0);
+
+	val = readl(chan->regs + ZYNQMP_DMA_CTRL1);
+	if (chan->config.src_issue)
+		val = (val & ~ZYNQMP_DMA_SRC_ISSUE) | chan->config.src_issue;
+	else
+		val |= ZYNQMP_DMA_SRC_ISSUE_RST_VAL;
+	writel(val, chan->regs + ZYNQMP_DMA_CTRL1);
+
+	val = readl(chan->regs + ZYNQMP_DMA_DATA_ATTR);
+	if (chan->config.src_burst_len)
+		val = (val & ~ZYNQMP_DMA_ARLEN) |
+			(chan->config.src_burst_len << ZYNQMP_DMA_ARLEN_OFST);
+	else
+		val |= ZYNQMP_DMA_ARLEN_RST_VAL;
+
+	if (chan->config.dst_burst_len)
+		val = (val & ~ZYNQMP_DMA_AWLEN) |
+			(chan->config.dst_burst_len << ZYNQMP_DMA_AWLEN_OFST);
+	else
+		val |= ZYNQMP_DMA_AWLEN_RST_VAL;
+	writel(val, chan->regs + ZYNQMP_DMA_DATA_ATTR);
+
+}
+
+/**
+ * zynqmp_dma_start_transfer - Initiate the new transfer
+ * @chan: ZynqMP DMA channel pointer
+ */
+static void zynqmp_dma_start_transfer(struct zynqmp_dma_chan *chan)
+{
+	struct zynqmp_dma_desc_sw *desc;
+
+	if (!zynqmp_dma_chan_is_idle(chan))
+		return;
+
+	zynqmp_dma_config(chan);
+
+	desc = list_first_entry_or_null(&chan->pending_list,
+					struct zynqmp_dma_desc_sw, node);
+	if (!desc)
+		return;
+
+	if (chan->config.has_sg)
+		list_splice_tail_init(&chan->pending_list, &chan->active_list);
+	else
+		list_move_tail(&desc->node, &chan->active_list);
+
+	if (chan->config.has_sg)
+		zynqmp_dma_update_desc_to_ctrlr(chan, desc);
+	else
+		zynqmp_dma_config_simple_desc(chan, desc->src, desc->dst,
+					      desc->len);
+
+	zynqmp_dma_start(chan);
+}
+
+
+/**
+ * zynqmp_dma_chan_desc_cleanup - Cleanup the completed descriptors
+ * @chan: ZynqMP DMA channel
+ */
+static void zynqmp_dma_chan_desc_cleanup(struct zynqmp_dma_chan *chan)
+{
+	struct zynqmp_dma_desc_sw *desc, *next;
+
+	list_for_each_entry_safe(desc, next, &chan->done_list, node) {
+		dma_async_tx_callback callback;
+		void *callback_param;
+
+		list_del(&desc->node);
+
+		callback = desc->async_tx.callback;
+		callback_param = desc->async_tx.callback_param;
+		if (callback) {
+			spin_unlock(&chan->lock);
+			callback(callback_param);
+			spin_lock(&chan->lock);
+		}
+
+		/* Run any dependencies, then free the descriptor */
+		zynqmp_dma_free_descriptor(chan, desc);
+	}
+}
+
+/**
+ * zynqmp_dma_complete_descriptor - Mark the active descriptor as complete
+ * @chan: ZynqMP DMA channel pointer
+ */
+static void zynqmp_dma_complete_descriptor(struct zynqmp_dma_chan *chan)
+{
+	struct zynqmp_dma_desc_sw *desc;
+
+	desc = list_first_entry_or_null(&chan->active_list,
+					struct zynqmp_dma_desc_sw, node);
+	if (!desc)
+		return;
+	list_del(&desc->node);
+	dma_cookie_complete(&desc->async_tx);
+	list_add_tail(&desc->node, &chan->done_list);
+}
+
+/**
+ * zynqmp_dma_issue_pending - Issue pending transactions
+ * @dchan: DMA channel pointer
+ */
+static void zynqmp_dma_issue_pending(struct dma_chan *dchan)
+{
+	struct zynqmp_dma_chan *chan = to_chan(dchan);
+
+	spin_lock_bh(&chan->lock);
+	zynqmp_dma_start_transfer(chan);
+	spin_unlock_bh(&chan->lock);
+}
+
+/**
+ * zynqmp_dma_free_descriptors - Free channel descriptors
+ * @dchan: DMA channel pointer
+ */
+static void zynqmp_dma_free_descriptors(struct zynqmp_dma_chan *chan)
+{
+	zynqmp_dma_free_desc_list(chan, &chan->active_list);
+	zynqmp_dma_free_desc_list(chan, &chan->pending_list);
+	zynqmp_dma_free_desc_list(chan, &chan->done_list);
+}
+
+/**
+ * zynqmp_dma_free_chan_resources - Free channel resources
+ * @dchan: DMA channel pointer
+ */
+static void zynqmp_dma_free_chan_resources(struct dma_chan *dchan)
+{
+	struct zynqmp_dma_chan *chan = to_chan(dchan);
+
+	spin_lock_bh(&chan->lock);
+	zynqmp_dma_free_descriptors(chan);
+	spin_unlock_bh(&chan->lock);
+	dma_free_coherent(chan->dev,
+		(2 * ZYNQMP_DMA_DESC_SIZE(chan) * ZYNQMP_DMA_NUM_DESCS),
+		chan->desc_pool_v, chan->desc_pool_p);
+	kfree(chan->sw_desc_pool);
+}
+
+/**
+ * zynqmp_dma_tx_status - Get dma transaction status
+ * @dchan: DMA channel pointer
+ * @cookie: Transaction identifier
+ * @txstate: Transaction state
+ *
+ * Return: DMA transaction status
+ */
+static enum dma_status zynqmp_dma_tx_status(struct dma_chan *dchan,
+				      dma_cookie_t cookie,
+				      struct dma_tx_state *txstate)
+{
+	return dma_cookie_status(dchan, cookie, txstate);
+}
+
+/**
+ * zynqmp_dma_reset - Reset the channel
+ * @chan: ZynqMP DMA channel pointer
+ */
+static void zynqmp_dma_reset(struct zynqmp_dma_chan *chan)
+{
+	writel(ZYNQMP_DMA_IDS_DEFAULT_MASK, chan->regs + ZYNQMP_DMA_IDS);
+
+	zynqmp_dma_complete_descriptor(chan);
+	zynqmp_dma_chan_desc_cleanup(chan);
+	zynqmp_dma_free_descriptors(chan);
+	zynqmp_dma_init(chan);
+}
+
+/**
+ * zynqmp_dma_irq_handler - ZynqMP DMA Interrupt handler
+ * @irq: IRQ number
+ * @data: Pointer to the ZynqMP DMA channel structure
+ *
+ * Return: IRQ_HANDLED/IRQ_NONE
+ */
+static irqreturn_t zynqmp_dma_irq_handler(int irq, void *data)
+{
+	struct zynqmp_dma_chan *chan = (struct zynqmp_dma_chan *)data;
+	u32 isr, imr, status;
+	irqreturn_t ret = IRQ_NONE;
+
+	isr = readl(chan->regs + ZYNQMP_DMA_ISR);
+	imr = readl(chan->regs + ZYNQMP_DMA_IMR);
+	status = isr & ~imr;
+
+	writel(isr, chan->regs + ZYNQMP_DMA_ISR);
+	if (status & ZYNQMP_DMA_INT_DONE) {
+		tasklet_schedule(&chan->tasklet);
+		ret = IRQ_HANDLED;
+	}
+
+	if (status & ZYNQMP_DMA_DONE)
+		chan->idle = true;
+
+	if (status & ZYNQMP_DMA_INT_ERR) {
+		chan->err = true;
+		tasklet_schedule(&chan->tasklet);
+		dev_err(chan->dev, "Channel %p has errors\n", chan);
+		ret = IRQ_HANDLED;
+	}
+
+	if (status & ZYNQMP_DMA_INT_OVRFL) {
+		zynqmp_dma_handle_ovfl_int(chan, status);
+		dev_info(chan->dev, "Channel %p overflow interrupt\n", chan);
+		ret = IRQ_HANDLED;
+	}
+
+	return ret;
+}
+
+/**
+ * zynqmp_dma_do_tasklet - Schedule completion tasklet
+ * @data: Pointer to the ZynqMP DMA channel structure
+ */
+static void zynqmp_dma_do_tasklet(unsigned long data)
+{
+	struct zynqmp_dma_chan *chan = (struct zynqmp_dma_chan *)data;
+	u32 count;
+
+	spin_lock(&chan->lock);
+
+	if (chan->err) {
+		zynqmp_dma_reset(chan);
+		chan->err = false;
+		goto unlock;
+	}
+
+	count = readl(chan->regs + ZYNQMP_DMA_IRQ_DST_ACCT);
+
+	while (count) {
+		zynqmp_dma_complete_descriptor(chan);
+		zynqmp_dma_chan_desc_cleanup(chan);
+		count--;
+	}
+
+	if (chan->idle)
+		zynqmp_dma_start_transfer(chan);
+
+unlock:
+	spin_unlock(&chan->lock);
+}
+
+/**
+ * zynqmp_dma_device_terminate_all - Aborts all transfers on a channel
+ * @dchan: DMA channel pointer
+ *
+ * Return: Always '0'
+ */
+static int zynqmp_dma_device_terminate_all(struct dma_chan *dchan)
+{
+	struct zynqmp_dma_chan *chan = to_chan(dchan);
+
+	spin_lock_bh(&chan->lock);
+	writel(ZYNQMP_DMA_IDS_DEFAULT_MASK, chan->regs + ZYNQMP_DMA_IDS);
+	zynqmp_dma_free_descriptors(chan);
+	spin_unlock_bh(&chan->lock);
+
+	return 0;
+}
+
+/**
+ * zynqmp_dma_prep_memcpy - prepare descriptors for memcpy transaction
+ * @dchan: DMA channel
+ * @dma_dst: Destination buffer address
+ * @dma_src: Source buffer address
+ * @len: Transfer length
+ * @flags: transfer ack flags
+ *
+ * Return: Async transaction descriptor on success and NULL on failure
+ */
+static struct dma_async_tx_descriptor *zynqmp_dma_prep_memcpy(
+				struct dma_chan *dchan, dma_addr_t dma_dst,
+				dma_addr_t dma_src, size_t len, ulong flags)
+{
+	struct zynqmp_dma_chan *chan;
+	struct zynqmp_dma_desc_sw *new, *first = NULL;
+	void *desc = NULL, *prev = NULL;
+	size_t copy;
+	u32 desc_cnt;
+
+	chan = to_chan(dchan);
+
+	if ((len > ZYNQMP_DMA_MAX_TRANS_LEN) && !chan->config.has_sg)
+		return NULL;
+
+	desc_cnt = DIV_ROUND_UP(len, ZYNQMP_DMA_MAX_TRANS_LEN);
+
+	spin_lock_bh(&chan->lock);
+	if ((desc_cnt > chan->desc_free_cnt) && chan->config.has_sg) {
+		spin_unlock_bh(&chan->lock);
+		dev_dbg(chan->dev, "chan %p descs are not available\n", chan);
+		return NULL;
+	}
+	chan->desc_free_cnt = chan->desc_free_cnt - desc_cnt;
+	spin_unlock_bh(&chan->lock);
+
+	do {
+		/* Allocate and populate the descriptor */
+		new = zynqmp_dma_get_descriptor(chan);
+
+		copy = min_t(size_t, len, ZYNQMP_DMA_MAX_TRANS_LEN);
+		if (chan->config.has_sg) {
+			desc = (struct zynqmp_dma_desc_ll *)new->src_v;
+			zynqmp_dma_config_sg_ll_desc(chan, desc, dma_src,
+						     dma_dst, copy, prev);
+		} else {
+			new->src = dma_src;
+			new->dst = dma_dst;
+			new->len = len;
+		}
+
+		prev = desc;
+		len -= copy;
+		dma_src += copy;
+		dma_dst += copy;
+		if (!first)
+			first = new;
+		else
+			list_add_tail(&new->node, &first->tx_list);
+	} while (len);
+
+	if (chan->config.has_sg)
+		zynqmp_dma_desc_config_eod(chan, desc);
+
+	async_tx_ack(&first->async_tx);
+	first->async_tx.flags = flags;
+	return &first->async_tx;
+}
+
+/**
+ * zynqmp_dma_prep_slave_sg - prepare descriptors for a memory sg transaction
+ * @dchan: DMA channel
+ * @dst_sg: Destination scatter list
+ * @dst_sg_len: Number of entries in destination scatter list
+ * @src_sg: Source scatter list
+ * @src_sg_len: Number of entries in source scatter list
+ * @flags: transfer ack flags
+ *
+ * Return: Async transaction descriptor on success and NULL on failure
+ */
+static struct dma_async_tx_descriptor *zynqmp_dma_prep_sg(
+			struct dma_chan *dchan, struct scatterlist *dst_sg,
+			unsigned int dst_sg_len, struct scatterlist *src_sg,
+			unsigned int src_sg_len, unsigned long flags)
+{
+	struct zynqmp_dma_desc_sw *new, *first = NULL;
+	struct zynqmp_dma_chan *chan = to_chan(dchan);
+	void *desc = NULL, *prev = NULL;
+	size_t len, dst_avail, src_avail;
+	dma_addr_t dma_dst, dma_src;
+	u32 desc_cnt = 0, i;
+	struct scatterlist *sg;
+
+	if (!chan->config.has_sg)
+		return NULL;
+
+	for_each_sg(src_sg, sg, src_sg_len, i)
+		desc_cnt += DIV_ROUND_UP(sg_dma_len(sg),
+					 ZYNQMP_DMA_MAX_TRANS_LEN);
+
+	spin_lock_bh(&chan->lock);
+	if (desc_cnt > chan->desc_free_cnt) {
+		spin_unlock_bh(&chan->lock);
+		dev_dbg(chan->dev, "chan %p descs are not available\n", chan);
+		return NULL;
+	}
+	chan->desc_free_cnt = chan->desc_free_cnt - desc_cnt;
+	spin_unlock_bh(&chan->lock);
+
+	dst_avail = sg_dma_len(dst_sg);
+	src_avail = sg_dma_len(src_sg);
+
+	/* Run until we are out of scatterlist entries */
+	while (true) {
+		/* Allocate and populate the descriptor */
+		new = zynqmp_dma_get_descriptor(chan);
+		desc = (struct zynqmp_dma_desc_ll *)new->src_v;
+		len = min_t(size_t, src_avail, dst_avail);
+		len = min_t(size_t, len, ZYNQMP_DMA_MAX_TRANS_LEN);
+		if (len == 0)
+			goto fetch;
+		dma_dst = sg_dma_address(dst_sg) + sg_dma_len(dst_sg) -
+			dst_avail;
+		dma_src = sg_dma_address(src_sg) + sg_dma_len(src_sg) -
+			src_avail;
+
+		zynqmp_dma_config_sg_ll_desc(chan, desc, dma_src, dma_dst,
+					     len, prev);
+		prev = desc;
+		dst_avail -= len;
+		src_avail -= len;
+
+		if (!first)
+			first = new;
+		else
+			list_add_tail(&new->node, &first->tx_list);
+fetch:
+		/* Fetch the next dst scatterlist entry */
+		if (dst_avail == 0) {
+			if (dst_sg_len == 0)
+				break;
+			dst_sg = sg_next(dst_sg);
+			if (dst_sg == NULL)
+				break;
+			dst_sg_len--;
+			dst_avail = sg_dma_len(dst_sg);
+		}
+		/* Fetch the next src scatterlist entry */
+		if (src_avail == 0) {
+			if (src_sg_len == 0)
+				break;
+			src_sg = sg_next(src_sg);
+			if (src_sg == NULL)
+				break;
+			src_sg_len--;
+			src_avail = sg_dma_len(src_sg);
+		}
+	}
+
+	zynqmp_dma_desc_config_eod(chan, desc);
+	first->async_tx.flags = flags;
+	return &first->async_tx;
+}
+
+/**
+ * zynqmp_dma_channel_set_config - Configure ZYNQMP DMA channel
+ * @dchan: DMA channel
+ * @cfg: ZYNQMP DMA device configuration pointer
+ *
+ * Return: '0' always
+ */
+int zynqmp_dma_channel_set_config(struct dma_chan *dchan,
+				  struct zynqmp_dma_config *cfg)
+{
+	struct zynqmp_dma_chan *chan = to_chan(dchan);
+
+	chan->config.ovrfetch = cfg->ovrfetch;
+	chan->config.has_sg = cfg->has_sg;
+	chan->config.ratectrl = cfg->ratectrl;
+	chan->config.src_issue = cfg->src_issue;
+	chan->config.src_burst_len = cfg->src_burst_len;
+	chan->config.dst_burst_len = cfg->dst_burst_len;
+
+	return 0;
+}
+EXPORT_SYMBOL(zynqmp_dma_channel_set_config);
+
+/**
+ * zynqmp_dma_chan_remove - Channel remove function
+ * @chan: ZynqMP DMA channel pointer
+ */
+static void zynqmp_dma_chan_remove(struct zynqmp_dma_chan *chan)
+{
+	if (!chan)
+		return;
+
+	devm_free_irq(chan->zdev->dev, chan->irq, chan);
+	tasklet_kill(&chan->tasklet);
+	list_del(&chan->common.device_node);
+	clk_disable_unprepare(chan->clk_apb);
+	clk_disable_unprepare(chan->clk_main);
+}
+
+/**
+ * zynqmp_dma_chan_probe - Per Channel Probing
+ * @zdev: Driver specific device structure
+ * @pdev: Pointer to the platform_device structure
+ *
+ * Return: '0' on success and failure value on error
+ */
+static int zynqmp_dma_chan_probe(struct zynqmp_dma_device *zdev,
+			   struct platform_device *pdev)
+{
+	struct zynqmp_dma_chan *chan;
+	struct resource *res;
+	struct device_node *node = pdev->dev.of_node;
+	int err;
+
+	chan = devm_kzalloc(zdev->dev, sizeof(*chan), GFP_KERNEL);
+	if (!chan)
+		return -ENOMEM;
+	chan->dev = zdev->dev;
+	chan->zdev = zdev;
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	chan->regs = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(chan->regs))
+		return PTR_ERR(chan->regs);
+
+	chan->bus_width = ZYNQMP_DMA_BUS_WIDTH_64;
+	chan->config.src_issue = ZYNQMP_DMA_SRC_ISSUE_RST_VAL;
+	chan->config.dst_burst_len = ZYNQMP_DMA_AWLEN_RST_VAL;
+	chan->config.src_burst_len = ZYNQMP_DMA_ARLEN_RST_VAL;
+	err = of_property_read_u32(node, "xlnx,bus-width", &chan->bus_width);
+	if ((err < 0) && ((chan->bus_width != ZYNQMP_DMA_BUS_WIDTH_64) ||
+			  (chan->bus_width != ZYNQMP_DMA_BUS_WIDTH_128))) {
+		dev_err(zdev->dev, "invalid bus-width value");
+		return err;
+	}
+
+	chan->bus_width = chan->bus_width / 8;
+	chan->is_dmacoherent =  of_property_read_bool(node, "dma-coherent");
+
+	zdev->chan = chan;
+	tasklet_init(&chan->tasklet, zynqmp_dma_do_tasklet, (ulong)chan);
+	spin_lock_init(&chan->lock);
+	INIT_LIST_HEAD(&chan->active_list);
+	INIT_LIST_HEAD(&chan->pending_list);
+	INIT_LIST_HEAD(&chan->done_list);
+	INIT_LIST_HEAD(&chan->free_list);
+
+	dma_cookie_init(&chan->common);
+	chan->common.device = &zdev->common;
+	list_add_tail(&chan->common.device_node, &zdev->common.channels);
+
+	zynqmp_dma_init(chan);
+	chan->irq = platform_get_irq(pdev, 0);
+	if (chan->irq < 0)
+		return -ENXIO;
+	err = devm_request_irq(&pdev->dev, chan->irq, zynqmp_dma_irq_handler, 0,
+			       "zynqmp-dma", chan);
+	if (err)
+		return err;
+	chan->clk_main = devm_clk_get(&pdev->dev, "clk_main");
+	if (IS_ERR(chan->clk_main)) {
+		dev_err(&pdev->dev, "main clock not found.\n");
+		return PTR_ERR(chan->clk_main);
+	}
+
+	chan->clk_apb = devm_clk_get(&pdev->dev, "clk_apb");
+	if (IS_ERR(chan->clk_apb)) {
+		dev_err(&pdev->dev, "apb clock not found.\n");
+		return PTR_ERR(chan->clk_apb);
+	}
+
+	err = clk_prepare_enable(chan->clk_main);
+	if (err) {
+		dev_err(&pdev->dev, "Unable to enable main clock.\n");
+		return err;
+	}
+
+	err = clk_prepare_enable(chan->clk_apb);
+	if (err) {
+		clk_disable_unprepare(chan->clk_main);
+		dev_err(&pdev->dev, "Unable to enable apb clock.\n");
+		return err;
+	}
+
+	chan->desc_size = sizeof(struct zynqmp_dma_desc_ll);
+	chan->idle = true;
+	return 0;
+}
+
+/**
+ * of_zynqmp_dma_xlate - Translation function
+ * @dma_spec: Pointer to DMA specifier as found in the device tree
+ * @ofdma: Pointer to DMA controller data
+ *
+ * Return: DMA channel pointer on success and NULL on error
+ */
+static struct dma_chan *of_zynqmp_dma_xlate(struct of_phandle_args *dma_spec,
+					    struct of_dma *ofdma)
+{
+	struct zynqmp_dma_device *zdev = ofdma->of_dma_data;
+
+	return dma_get_slave_channel(&zdev->chan->common);
+}
+
+/**
+ * zynqmp_dma_probe - Driver probe function
+ * @pdev: Pointer to the platform_device structure
+ *
+ * Return: '0' on success and failure value on error
+ */
+static int zynqmp_dma_probe(struct platform_device *pdev)
+{
+	struct zynqmp_dma_device *zdev;
+	struct dma_device *p;
+	int ret;
+
+	zdev = devm_kzalloc(&pdev->dev, sizeof(*zdev), GFP_KERNEL);
+	if (!zdev)
+		return -ENOMEM;
+
+	zdev->dev = &pdev->dev;
+	INIT_LIST_HEAD(&zdev->common.channels);
+
+	dma_set_mask(&pdev->dev, DMA_BIT_MASK(44));
+	dma_cap_set(DMA_SG, zdev->common.cap_mask);
+	dma_cap_set(DMA_MEMCPY, zdev->common.cap_mask);
+
+	p = &zdev->common;
+	p->device_prep_dma_sg = zynqmp_dma_prep_sg;
+	p->device_prep_dma_memcpy = zynqmp_dma_prep_memcpy;
+	p->device_terminate_all = zynqmp_dma_device_terminate_all;
+	p->device_issue_pending = zynqmp_dma_issue_pending;
+	p->device_alloc_chan_resources = zynqmp_dma_alloc_chan_resources;
+	p->device_free_chan_resources = zynqmp_dma_free_chan_resources;
+	p->device_tx_status = zynqmp_dma_tx_status;
+	p->dev = &pdev->dev;
+
+	platform_set_drvdata(pdev, zdev);
+
+	ret = zynqmp_dma_chan_probe(zdev, pdev);
+	if (ret) {
+		dev_err(&pdev->dev, "Probing channel failed\n");
+		goto free_chan_resources;
+	}
+
+	p->dst_addr_widths = BIT(zdev->chan->bus_width);
+	p->src_addr_widths = BIT(zdev->chan->bus_width);
+
+	dma_async_device_register(&zdev->common);
+
+	ret = of_dma_controller_register(pdev->dev.of_node,
+					 of_zynqmp_dma_xlate, zdev);
+	if (ret) {
+		dev_err(&pdev->dev, "Unable to register DMA to DT\n");
+		dma_async_device_unregister(&zdev->common);
+		goto free_chan_resources;
+	}
+
+	dev_info(&pdev->dev, "ZynqMP DMA driver Probe success\n");
+
+	return 0;
+
+free_chan_resources:
+	zynqmp_dma_chan_remove(zdev->chan);
+	return ret;
+}
+
+/**
+ * zynqmp_dma_remove - Driver remove function
+ * @pdev: Pointer to the platform_device structure
+ *
+ * Return: Always '0'
+ */
+static int zynqmp_dma_remove(struct platform_device *pdev)
+{
+	struct zynqmp_dma_device *zdev = platform_get_drvdata(pdev);
+
+	of_dma_controller_free(pdev->dev.of_node);
+	dma_async_device_unregister(&zdev->common);
+
+	zynqmp_dma_chan_remove(zdev->chan);
+
+	return 0;
+}
+
+static const struct of_device_id zynqmp_dma_of_match[] = {
+	{ .compatible = "xlnx,zynqmp-dma-1.0", },
+	{}
+};
+MODULE_DEVICE_TABLE(of, zynqmp_dma_of_match);
+
+static struct platform_driver zynqmp_dma_driver = {
+	.driver = {
+		.name = "xilinx-zynqmp-dma",
+		.of_match_table = zynqmp_dma_of_match,
+	},
+	.probe = zynqmp_dma_probe,
+	.remove = zynqmp_dma_remove,
+};
+
+module_platform_driver(zynqmp_dma_driver);
+
+MODULE_AUTHOR("Xilinx, Inc.");
+MODULE_DESCRIPTION("Xilinx ZynqMP DMA driver");
+MODULE_LICENSE("GPL");