Dev ad9740 wip #3035

sarpadi · 2025-12-04T18:06:38Z

PR Description

Please replace this comment with a summary of your changes, and add any context
necessary to understand them. List any dependencies required for this change.
To check the checkboxes below, insert a 'x' between square brackets (without
any space), or simply check them after publishing the PR.
If you changes include a breaking change, please specify dependent PRs in the
description and try to push all related PRs simultaneously.

PR Type

Bug fix (a change that fixes an issue)
New feature (a change that adds new functionality)
Breaking change (a change that affects other repos or cause CIs to fail)

PR Checklist

I have conducted a self-review of my own code changes
I have compiled my changes, including the documentation
I have tested the changes on the relevant hardware
I have updated the documentation outside this repo accordingly
I have provided links for the relevant upstream lore

DDS is generating an apparently correct sine wave; in ramp mode a periodic signal is generated more similar to an asymetric PWM than to a ramp

machschmitt

Hi @sarpadi ,

Having a first look at this mess.
I'm not really getting what this wants to do with the iio-backend. I guess, it might make more sense to Nuno.

machschmitt · 2025-12-05T18:52:22Z

drivers/iio/dac/ad9740.c

+	/* Data format: true = 2's complement, false = offset binary */
+	bool twos_complement;


comment doesn't make much sense and/or is not meaningful. What is "offset binary"?

machschmitt · 2025-12-05T18:54:55Z

drivers/iio/dac/ad9740.c

+static int ad9740_read_raw(struct iio_dev *indio_dev,
+			   struct iio_chan_spec const *chan,
+			   int *val, int *val2, long mask)
+{
+	return -EINVAL;
+}


IIRC, this is not needed due to nothing can be read from the device

machschmitt · 2025-12-05T18:55:25Z

drivers/iio/dac/ad9740.c

+			    struct iio_chan_spec const *chan,
+			    int val, int val2, long mask)
+{
+	return -EINVAL;


this should probably have a proper implementation

machschmitt · 2025-12-05T18:57:46Z

drivers/iio/dac/ad9740.c

+	dev_info(st->dev, "Buffer enable requested - starting DMA streaming\n");
+	dev_info(st->dev, "========================================\n");
+
+	mutex_lock(&st->lock);


can use guard() instead

machschmitt · 2025-12-05T18:58:24Z

drivers/iio/dac/ad9740.c

+	dev_info(st->dev, "Buffer disable requested - stopping DMA streaming\n");
+	dev_info(st->dev, "========================================\n");
+
+	mutex_lock(&st->lock);


machschmitt · 2025-12-05T19:17:25Z

drivers/iio/dac/ad9740.c

+
+	dev_info(&pdev->dev, "Configuring IIO device structure\n");
+	indio_dev->name = st->chip_info->name;
+	indio_dev->modes = INDIO_DIRECT_MODE | INDIO_BUFFER_HARDWARE;


INDIO_BUFFER_HARDWARE should be added within the proper buffer_setup() call

machschmitt · 2025-12-05T19:24:11Z

drivers/iio/dac/ad9740.c

+	/* Make a modifiable copy of the channel spec for backend extension */
+	struct iio_chan_spec *channels;
+	const struct iio_chan_spec_ext_info *backend_ext_info;
+	struct iio_chan_spec_ext_info *merged_ext_info;
+	int num_our_ext_info, num_backend_ext_info, i, j;
+
+	dev_info(&pdev->dev, "Creating modifiable channel spec\n");
+	channels = devm_kmemdup(&pdev->dev, st->chip_info->channels,
+				sizeof(struct iio_chan_spec) * st->chip_info->num_channels,
+				GFP_KERNEL);
+	if (!channels) {
+		dev_err(&pdev->dev, "Failed to allocate channel spec\n");
+		return -ENOMEM;
+	}
+
+	/* Extend channel with DDS controls from backend */
+	dev_info(&pdev->dev, "Extending channel spec with backend DDS controls\n");
+	ret = iio_backend_extend_chan_spec(st->back, &channels[0]);
+	if (ret) {
+		dev_err(&pdev->dev, "Failed to extend channel spec: %d\n", ret);
+		return ret;
+	}
+
+	/* Merge our ext_info (data_source) with backend's ext_info (DDS) */
+	backend_ext_info = channels[0].ext_info;
+
+	/* Count entries in both arrays */
+	num_our_ext_info = ARRAY_SIZE(ad9740_ext_info) - 1; /* exclude terminator */
+	num_backend_ext_info = 0;
+	if (backend_ext_info) {
+		while (backend_ext_info[num_backend_ext_info].name)
+			num_backend_ext_info++;
+	}
+
+	dev_info(&pdev->dev, "Merging ext_info: %d our attrs + %d backend attrs\n",
+		 num_our_ext_info, num_backend_ext_info);
+
+	/* Allocate merged array: our entries + backend entries + terminator */
+	merged_ext_info = devm_kcalloc(&pdev->dev,
+				       num_our_ext_info + num_backend_ext_info + 1,
+				       sizeof(*merged_ext_info), GFP_KERNEL);
+	if (!merged_ext_info) {
+		dev_err(&pdev->dev, "Failed to allocate merged ext_info\n");
+		return -ENOMEM;
+	}
+
+	/* Copy our ext_info first */
+	for (i = 0; i < num_our_ext_info; i++)
+		merged_ext_info[i] = ad9740_ext_info[i];
+
+	/* Then append backend ext_info */
+	for (j = 0; j < num_backend_ext_info; j++)
+		merged_ext_info[i + j] = backend_ext_info[j];
+
+	/* Terminator is already zero from kcalloc */
+	channels[0].ext_info = merged_ext_info;
+	dev_info(&pdev->dev, "Extended attributes merged successfully\n");


i'm not sure about this channel juggling. Nuno might understand it better (or not?).

machschmitt · 2025-12-05T19:25:06Z

drivers/iio/dac/ad9740.c

+	/* Setup DAC and backend */
+	ret = ad9740_setup(st);
+	if (ret) {
+		dev_err(&pdev->dev, "AD9740 setup failed: %d\n", ret);


dev_err_probe

machschmitt · 2025-12-05T19:26:21Z

drivers/iio/dac/ad9740.c

+
+	/* Register IIO device */
+	dev_info(&pdev->dev, "Registering IIO device\n");
+	ret = devm_iio_device_register(&pdev->dev, indio_dev);


i see there's a lot of debugging, but should probably just return devm_iio_... here when it gets well tested

machschmitt · 2025-12-05T19:30:40Z

drivers/iio/dac/adi-axi-dac.c

 static const struct of_device_id axi_dac_of_match[] = {
 	{ .compatible = "adi,axi-dac-9.1.b", .data = &dac_generic },
 	{ .compatible = "adi,axi-ad3552r", .data = &dac_ad3552r },
+	{ .compatible = "adi,axi-ad9740", .data = &dac_ad9740 },


i haven't looked into the dac data sheet nor the HDL in details but, why can't this use the generic dac backend?

sarpadi added 11 commits October 3, 2025 10:28

ad9740: Add dts

54e4e28

ad9740: enable aux output; set f=128MHz

ca74221

ad9740: Fix adf dts

da812e1

ad9740: Cleanup

1fff25a

ad9740: DDS and Ramp mode working

f888616

DDS is generating an apparently correct sine wave; in ramp mode a periodic signal is generated more similar to an asymetric PWM than to a ramp

ad9740: DMA mode working but not in cyclic mode

64739bd

ad9740: DMA mode working at 120MHz

0cc3596

ad9740: DTS: Update ADF clk to 210MHz

73b298b

AD9740: Add support for all 4 devices

b7d39cd

ad9740: Fix MSB alignment

a699234

ad9740: Clean up ad974x references

318468a

machschmitt reviewed Dec 5, 2025

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Dev ad9740 wip #3035

Dev ad9740 wip #3035

Uh oh!

sarpadi commented Dec 4, 2025

Uh oh!

machschmitt left a comment

Uh oh!

machschmitt Dec 5, 2025

Uh oh!

machschmitt Dec 5, 2025

Uh oh!

machschmitt Dec 5, 2025

Uh oh!

machschmitt Dec 5, 2025

Uh oh!

machschmitt Dec 5, 2025

Uh oh!

machschmitt Dec 5, 2025

Uh oh!

machschmitt Dec 5, 2025

Uh oh!

machschmitt Dec 5, 2025

Uh oh!

machschmitt Dec 5, 2025

Uh oh!

machschmitt Dec 5, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

		/* Data format: true = 2's complement, false = offset binary */
		bool twos_complement;

Dev ad9740 wip #3035

Are you sure you want to change the base?

Dev ad9740 wip #3035

Uh oh!

Conversation

sarpadi commented Dec 4, 2025

PR Description

PR Type

PR Checklist

Uh oh!

machschmitt left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants