[PATCH 0/3] iio: adc: ade9000: series of probe and channel fixes

[PATCH 0/3] iio: adc: ade9000: series of probe and channel fixes

[Antoniu Miclaus]

This series fixes three issues in the ade9000 driver, all present since
the driver was added.

Patch 1 moves the interrupt requests after the vdd regulator is enabled,
so an early interrupt no longer runs a handler doing SPI accesses against
an unpowered chip.

Patch 2 corrects the overlapping scan_index values for the current and
voltage channels, which caused the Phase B and C waveform buffers to be
configured incorrectly.

Patch 3 fixes the STATUS1 event scan bound so that Phase C dip events
(bit 25) are actually reported.

Antoniu Miclaus (3):
  iio: adc: ade9000: request interrupts after powering the device
  iio: adc: ade9000: fix overlapping scan_index for current and voltage
    channels
  iio: adc: ade9000: handle Phase C dip events in IRQ1 handler

 drivers/iio/adc/ade9000.c | 31 ++++++++++++++++---------------
 1 file changed, 16 insertions(+), 15 deletions(-)

-- 
2.43.0

[PATCH 1/3] iio: adc: ade9000: request interrupts after powering the device

[Antoniu Miclaus]

The IRQ handlers do SPI register accesses, but the interrupts were
requested before the vdd regulator was enabled. An interrupt arriving
while the chip is unpowered runs a handler against a dead chip, causing
SPI errors or garbage reads.

Request the interrupts after enabling the regulator. This keeps irq1
registered before ade9000_reset(), which waits on it, and lets devm free
the interrupts before the regulator is disabled.

Fixes: 81de7b4619fc ("iio: adc: add ade9000 support")
Signed-off-by: Antoniu Miclaus <antoniu.miclaus@analog.com>
---
 drivers/iio/adc/ade9000.c | 25 +++++++++++++------------
 1 file changed, 13 insertions(+), 12 deletions(-)

diff --git a/drivers/iio/adc/ade9000.c b/drivers/iio/adc/ade9000.c
index c6c3ea953fea..51cb43755a1f 100644
--- a/drivers/iio/adc/ade9000.c
+++ b/drivers/iio/adc/ade9000.c
@@ -1705,18 +1705,6 @@ static int ade9000_probe(struct spi_device *spi)
 	if (ret)
 		return ret;
 
-	ret = ade9000_request_irq(dev, "irq0", ade9000_irq0_thread, indio_dev);
-	if (ret)
-		return ret;
-
-	ret = ade9000_request_irq(dev, "irq1", ade9000_irq1_thread, indio_dev);
-	if (ret)
-		return ret;
-
-	ret = ade9000_request_irq(dev, "dready", ade9000_dready_thread, indio_dev);
-	if (ret)
-		return ret;
-
 	/* External CMOS clock input (optional - crystal can be used instead) */
 	st->clkin = devm_clk_get_optional_enabled(dev, NULL);
 	if (IS_ERR(st->clkin))
@@ -1736,6 +1724,19 @@ static int ade9000_probe(struct spi_device *spi)
 		return dev_err_probe(&spi->dev, ret,
 				     "Failed to get and enable vdd regulator\n");
 
+	/* Request IRQs only after powering the chip; the handlers do SPI access. */
+	ret = ade9000_request_irq(dev, "irq0", ade9000_irq0_thread, indio_dev);
+	if (ret)
+		return ret;
+
+	ret = ade9000_request_irq(dev, "irq1", ade9000_irq1_thread, indio_dev);
+	if (ret)
+		return ret;
+
+	ret = ade9000_request_irq(dev, "dready", ade9000_dready_thread, indio_dev);
+	if (ret)
+		return ret;
+
 	indio_dev->channels = ade9000_channels;
 	indio_dev->num_channels = ARRAY_SIZE(ade9000_channels);
 
-- 
2.43.0

[PATCH 2/3] iio: adc: ade9000: fix overlapping scan_index for current and voltage channels

[Antoniu Miclaus]

Current channels used scan_index "num" and voltage "num + 1", so with
phases 0, 1 and 2 the indices overlapped and no longer matched the
IA=0, VA=1, IB=2, VB=3, IC=4, VC=5 layout expected by
ade9000_waveform_buffer_config(). Phase B and C buffers were configured
incorrectly.

Use "num * 2" for current and "num * 2 + 1" for voltage to get the
non-overlapping 0/1, 2/3, 4/5 layout.

Fixes: 81de7b4619fc ("iio: adc: add ade9000 support")
Signed-off-by: Antoniu Miclaus <antoniu.miclaus@analog.com>
---
 drivers/iio/adc/ade9000.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/iio/adc/ade9000.c b/drivers/iio/adc/ade9000.c
index 51cb43755a1f..167b1f4d6279 100644
--- a/drivers/iio/adc/ade9000.c
+++ b/drivers/iio/adc/ade9000.c
@@ -444,7 +444,7 @@ static const struct iio_chan_spec_ext_info ade9000_ext_info[] = {
 			      BIT(IIO_CHAN_INFO_CALIBSCALE),		\
 	.event_spec = ade9000_current_events,				\
 	.num_event_specs = ARRAY_SIZE(ade9000_current_events),		\
-	.scan_index = num,						\
+	.scan_index = num * 2,						\
 	.indexed = 1,							\
 	.scan_type = {							\
 		.sign = 's',						\
@@ -464,7 +464,7 @@ static const struct iio_chan_spec_ext_info ade9000_ext_info[] = {
 			      BIT(IIO_CHAN_INFO_FREQUENCY),		\
 	.event_spec = ade9000_voltage_events,				\
 	.num_event_specs = ARRAY_SIZE(ade9000_voltage_events),		\
-	.scan_index = num + 1,	/* interleave with current channels */	\
+	.scan_index = num * 2 + 1, /* interleave with current channels */	\
 	.indexed = 1,							\
 	.scan_type = {							\
 		.sign = 's',						\
-- 
2.43.0

[PATCH 3/3] iio: adc: ade9000: handle Phase C dip events in IRQ1 handler

[Antoniu Miclaus]

ADE9000_ST1_CROSSING_DEPTH is the exclusive upper bound for scanning the
STATUS1 event bits, so a value of 25 left out the highest event bit,
ADE9000_ST1_DIPC_BIT (bit 25). Phase C dip events were never reported.

Bump the value to 26 so bit 25 is scanned.

Fixes: 81de7b4619fc ("iio: adc: add ade9000 support")
Signed-off-by: Antoniu Miclaus <antoniu.miclaus@analog.com>
---
 drivers/iio/adc/ade9000.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/iio/adc/ade9000.c b/drivers/iio/adc/ade9000.c
index 167b1f4d6279..da6caabfe2a4 100644
--- a/drivers/iio/adc/ade9000.c
+++ b/drivers/iio/adc/ade9000.c
@@ -219,7 +219,7 @@
 #define ADE9000_ST1_ERROR2_BIT		BIT(30)
 #define ADE9000_ST1_ERROR3_BIT		BIT(31)
 #define ADE9000_ST1_CROSSING_FIRST	6
-#define ADE9000_ST1_CROSSING_DEPTH	25
+#define ADE9000_ST1_CROSSING_DEPTH	26
 
 #define ADE9000_WFB_TRG_DIP_BIT		BIT(0)
 #define ADE9000_WFB_TRG_SWELL_BIT	BIT(1)
-- 
2.43.0

Re: [PATCH 1/3] iio: adc: ade9000: request interrupts after powering the device

[Joshua Crofts]

On Fri, 7 Aug 2026 11:56:41 +0300
Antoniu Miclaus <antoniu.miclaus@analog.com> wrote:

> The IRQ handlers do SPI register accesses, but the interrupts were
> requested before the vdd regulator was enabled. An interrupt arriving
> while the chip is unpowered runs a handler against a dead chip, causing
> SPI errors or garbage reads.
> 
> Request the interrupts after enabling the regulator. This keeps irq1
> registered before ade9000_reset(), which waits on it, and lets devm free
> the interrupts before the regulator is disabled.
> 
> Fixes: 81de7b4619fc ("iio: adc: add ade9000 support")
> Signed-off-by: Antoniu Miclaus <antoniu.miclaus@analog.com>
> ---

LGTM.

Reviewed-by: Joshua Crofts <joshua.crofts1@gmail.com>

-- 
Kind regards,
Joshua Crofts