From: Jonathan Santos Jonathan.Santos@analog.com
[ Upstream commit 7e54d932873d91a55d1b89b7389876d78aeeab32 ]
The SYNC_IN pulse width must be at least 1.5 x Tmclk, corresponding to ~2.5 µs at the lowest supported MCLK frequency. Add a 3 µs delay to ensure reliable synchronization timing even for the worst-case scenario.
Signed-off-by: Jonathan Santos Jonathan.Santos@analog.com Reviewed-by: David Lechner dlechner@baylibre.com Reviewed-by: Andy Shevchenko andriy.shevchenko@linux.intel.com Link: https://patch.msgid.link/d3ee92a533cd1207cf5c5cc4d7bdbb5c6c267f68.1749063024... Signed-off-by: Jonathan Cameron Jonathan.Cameron@huawei.com Signed-off-by: Sasha Levin sashal@kernel.org ---
LLM Generated explanations, may be completely bogus:
**Backport Status: YES**
This commit should be backported to stable kernel trees. Here's my extensive analysis:
**1. Bug Fix Nature:** The commit fixes a timing violation bug where the SYNC_IN pulse width was not meeting the hardware's minimum timing requirement. According to the datasheet (as noted in the commit), the SYNC_IN pulse must be at least 1.5 × Tmclk (MCLK period). Without the proper delay, the synchronization pulse may be too short, especially at lower MCLK frequencies, causing unreliable synchronization.
**2. Specific Code Analysis:**
The original code had a critical timing issue: ```c gpiod_set_value(st->gpio_sync_in, 1); gpiod_set_value(st->gpio_sync_in, 0); ```
This creates a pulse with essentially zero width (just the GPIO toggle time), which violates the hardware specification. The fix adds: - A new dedicated function `ad7768_send_sync_pulse()` that ensures proper timing - Uses `gpiod_set_value_cansleep()` instead of `gpiod_set_value()` for better reliability - Adds a 3 µs delay using `fsleep(3)` to guarantee the minimum pulse width
**3. Impact Analysis:** - **User Impact**: Without this fix, users may experience unreliable ADC synchronization, particularly when changing filter decimation rates or when operating at lower MCLK frequencies (0.6 MHz) - **Worst Case**: At 0.6 MHz, Tmclk = 1.67 µs, requiring a minimum pulse width of 2.5 µs. The original code provides essentially 0 µs - **Data Integrity**: Improper synchronization can lead to incorrect ADC readings or timing issues in multi-device setups
**4. Risk Assessment:** - **Low Risk**: The change is minimal and contained - it only adds a 3 µs delay and switches to the cansleep GPIO variant - **No API Changes**: The fix is internal to the driver with no external interface changes - **Conservative Fix**: The 3 µs delay provides margin above the 2.5 µs worst-case requirement
**5. Stable Kernel Criteria:** ✓ Fixes a real bug that affects users (timing violation) ✓ Small, contained change (adds one function, modifies one call site) ✓ No new features or architectural changes ✓ Minimal risk of regression ✓ Clear hardware specification violation being fixed
**6. Supporting Evidence:** - The driver has had multiple previous fixes backported (as seen in git log with "Fixes:" tags) - Similar timing-related fixes are commonly backported for hardware drivers - The AD7768-1 is an active production part used in data acquisition systems where reliability is critical
The commit meets all stable kernel criteria for backporting - it's a clear bug fix for a hardware timing violation that could cause unreliable operation, implemented with minimal changes and low regression risk.
drivers/iio/adc/ad7768-1.c | 23 +++++++++++++++++++---- 1 file changed, 19 insertions(+), 4 deletions(-)
diff --git a/drivers/iio/adc/ad7768-1.c b/drivers/iio/adc/ad7768-1.c index 51134023534a..8b414a102864 100644 --- a/drivers/iio/adc/ad7768-1.c +++ b/drivers/iio/adc/ad7768-1.c @@ -252,6 +252,24 @@ static const struct regmap_config ad7768_regmap24_config = { .max_register = AD7768_REG24_COEFF_DATA, };
+static int ad7768_send_sync_pulse(struct ad7768_state *st) +{ + /* + * The datasheet specifies a minimum SYNC_IN pulse width of 1.5 × Tmclk, + * where Tmclk is the MCLK period. The supported MCLK frequencies range + * from 0.6 MHz to 17 MHz, which corresponds to a minimum SYNC_IN pulse + * width of approximately 2.5 µs in the worst-case scenario (0.6 MHz). + * + * Add a delay to ensure the pulse width is always sufficient to + * trigger synchronization. + */ + gpiod_set_value_cansleep(st->gpio_sync_in, 1); + fsleep(3); + gpiod_set_value_cansleep(st->gpio_sync_in, 0); + + return 0; +} + static int ad7768_set_mode(struct ad7768_state *st, enum ad7768_conv_mode mode) { @@ -339,10 +357,7 @@ static int ad7768_set_dig_fil(struct ad7768_state *st, return ret;
/* A sync-in pulse is required every time the filter dec rate changes */ - gpiod_set_value(st->gpio_sync_in, 1); - gpiod_set_value(st->gpio_sync_in, 0); - - return 0; + return ad7768_send_sync_pulse(st); }
static int ad7768_set_freq(struct ad7768_state *st,