| Message ID | 20220809211246.251006-3-eajames@linux.ibm.com (mailing list archive) |
|---|---|
| State | Changes Requested |
| Headers | show |
| Series | iio: pressure: dps310: Reset chip if MEAS_CFG is corrupt | expand |
On Wed, Aug 10, 2022 at 12:12 AM Eddie James <eajames@linux.ibm.com> wrote: > > Corruption of the MEAS_CFG register has been observed soon after > system boot. In order to recover this scenario, check MEAS_CFG if > measurement isn't ready, and if it's incorrect, reset the DPS310 > and execute the startup procedure. Looks like both patches miss the Fixes tag. Can you add them? ... > +/* > + * Called with lock held. Returns a negative value on error, a positive value > + * when the device is not ready, and zero when the device is ready. > + */ > +static int dps310_check_reset_meas_cfg(struct dps310_data *data, int ready_bit) > +{ > + int meas_cfg; > + int rc = regmap_read(data->regmap, DPS310_MEAS_CFG, &meas_cfg); > + > + if (rc < 0) > + return rc; Please, split definition and assignment. > + /* Device is ready, proceed to measurement */ > + if (meas_cfg & ready_bit) > + return 0; > + > + /* Device is OK, just not ready */ > + if (meas_cfg & (DPS310_PRS_EN | DPS310_TEMP_EN | DPS310_BACKGROUND)) > + return 1; > + > + /* DPS310 register state corrupt, better start from scratch */ > + rc = regmap_write(data->regmap, DPS310_RESET, DPS310_RESET_MAGIC); > + if (rc < 0) > + return rc; > + > + /* Wait for device chip access: 2.5ms in specification */ > + usleep_range(2500, 12000); > + > + /* Reinitialize the chip */ > + rc = dps310_startup(data); > + if (rc) > + return rc; > + > + dev_info(&data->client->dev, > + "recovered from corrupted MEAS_CFG=%02x\n", meas_cfg); > + return 1; > +} > + > static int dps310_read_pres_raw(struct dps310_data *data) > { > int rc; > @@ -405,16 +443,26 @@ static int dps310_read_pres_raw(struct dps310_data *data) > if (mutex_lock_interruptible(&data->lock)) > return -EINTR; > > - rate = dps310_get_pres_samp_freq(data); > - timeout = DPS310_POLL_TIMEOUT_US(rate); > - > - /* Poll for sensor readiness; base the timeout upon the sample rate. */ > - rc = regmap_read_poll_timeout(data->regmap, DPS310_MEAS_CFG, ready, > - ready & DPS310_PRS_RDY, > - DPS310_POLL_SLEEP_US(timeout), timeout); > - if (rc) > + rc = dps310_check_reset_meas_cfg(data, DPS310_PRS_RDY); > + if (rc < 0) > goto done; > > + if (rc > 0) { > + rate = dps310_get_pres_samp_freq(data); > + timeout = DPS310_POLL_TIMEOUT_US(rate); > + > + /* > + * Poll for sensor readiness; base the timeout upon the sample > + * rate. > + */ > + rc = regmap_read_poll_timeout(data->regmap, DPS310_MEAS_CFG, > + ready, ready & DPS310_PRS_RDY, > + DPS310_POLL_SLEEP_US(timeout), > + timeout); > + if (rc) > + goto done; > + } If you split the condition body to a helper, it can be rewritten like (also note special definition for positive returned numbers): rc = ..._reset_meas_cfg(...); if (rc == DPS310_MEAS_NOT_READY) rc = ..._new_helper_func(...); if (rc) goto done; and looking at this it might be worth considering calling that conditional in the middle in the _reset_meas_cfg(), so the latter will return either 0 or negative error code. > + rc = dps310_check_reset_meas_cfg(data, DPS310_TMP_RDY); > if (rc < 0) > goto done; > > + if (rc > 0) { > + rate = dps310_get_temp_samp_freq(data); Okay, I see this function is different, but still you may realize a helper from below and something like above suggestion can still be achieved. > + timeout = DPS310_POLL_TIMEOUT_US(rate); > + > + /* > + * Poll for sensor readiness; base the timeout upon the sample > + * rate. > + */ > + rc = regmap_read_poll_timeout(data->regmap, DPS310_MEAS_CFG, > + ready, ready & DPS310_TMP_RDY, > + DPS310_POLL_SLEEP_US(timeout), > + timeout); > + if (rc < 0) Why out of a sudden ' < 0'? > + goto done; > + } As per above. > rc = dps310_read_temp_ready(data);
On 8/12/22 17:13, Andy Shevchenko wrote: > On Wed, Aug 10, 2022 at 12:12 AM Eddie James <eajames@linux.ibm.com> wrote: >> Corruption of the MEAS_CFG register has been observed soon after >> system boot. In order to recover this scenario, check MEAS_CFG if >> measurement isn't ready, and if it's incorrect, reset the DPS310 >> and execute the startup procedure. > Looks like both patches miss the Fixes tag. Can you add them? Well this isn't really a software fix - there's no identifiable bug in the driver. Just trying to recover the chip in this observed mystery scenario. > > ... > >> +/* >> + * Called with lock held. Returns a negative value on error, a positive value >> + * when the device is not ready, and zero when the device is ready. >> + */ >> +static int dps310_check_reset_meas_cfg(struct dps310_data *data, int ready_bit) >> +{ >> + int meas_cfg; >> + int rc = regmap_read(data->regmap, DPS310_MEAS_CFG, &meas_cfg); >> + >> + if (rc < 0) >> + return rc; > Please, split definition and assignment. Ack. > >> + /* Device is ready, proceed to measurement */ >> + if (meas_cfg & ready_bit) >> + return 0; >> + >> + /* Device is OK, just not ready */ >> + if (meas_cfg & (DPS310_PRS_EN | DPS310_TEMP_EN | DPS310_BACKGROUND)) >> + return 1; >> + >> + /* DPS310 register state corrupt, better start from scratch */ >> + rc = regmap_write(data->regmap, DPS310_RESET, DPS310_RESET_MAGIC); >> + if (rc < 0) >> + return rc; >> + >> + /* Wait for device chip access: 2.5ms in specification */ >> + usleep_range(2500, 12000); >> + >> + /* Reinitialize the chip */ >> + rc = dps310_startup(data); >> + if (rc) >> + return rc; >> + >> + dev_info(&data->client->dev, >> + "recovered from corrupted MEAS_CFG=%02x\n", meas_cfg); >> + return 1; >> +} >> + >> static int dps310_read_pres_raw(struct dps310_data *data) >> { >> int rc; >> @@ -405,16 +443,26 @@ static int dps310_read_pres_raw(struct dps310_data *data) >> if (mutex_lock_interruptible(&data->lock)) >> return -EINTR; >> >> - rate = dps310_get_pres_samp_freq(data); >> - timeout = DPS310_POLL_TIMEOUT_US(rate); >> - >> - /* Poll for sensor readiness; base the timeout upon the sample rate. */ >> - rc = regmap_read_poll_timeout(data->regmap, DPS310_MEAS_CFG, ready, >> - ready & DPS310_PRS_RDY, >> - DPS310_POLL_SLEEP_US(timeout), timeout); >> - if (rc) >> + rc = dps310_check_reset_meas_cfg(data, DPS310_PRS_RDY); >> + if (rc < 0) >> goto done; >> >> + if (rc > 0) { >> + rate = dps310_get_pres_samp_freq(data); >> + timeout = DPS310_POLL_TIMEOUT_US(rate); >> + >> + /* >> + * Poll for sensor readiness; base the timeout upon the sample >> + * rate. >> + */ >> + rc = regmap_read_poll_timeout(data->regmap, DPS310_MEAS_CFG, >> + ready, ready & DPS310_PRS_RDY, >> + DPS310_POLL_SLEEP_US(timeout), >> + timeout); >> + if (rc) >> + goto done; >> + } > If you split the condition body to a helper, it can be rewritten like > (also note special definition for positive returned numbers): > > rc = ..._reset_meas_cfg(...); > if (rc == DPS310_MEAS_NOT_READY) > rc = ..._new_helper_func(...); > if (rc) > goto done; > > and looking at this it might be worth considering calling that > conditional in the middle in the _reset_meas_cfg(), so the latter will > return either 0 or negative error code. To be honest that looks more complicated than the way it is now? And I don't think I can make it common between the temp and pressure without some complicated macro business. > >> + rc = dps310_check_reset_meas_cfg(data, DPS310_TMP_RDY); >> if (rc < 0) >> goto done; >> >> + if (rc > 0) { >> + rate = dps310_get_temp_samp_freq(data); > Okay, I see this function is different, but still you may realize a > helper from below and something like above suggestion can still be > achieved. > >> + timeout = DPS310_POLL_TIMEOUT_US(rate); >> + >> + /* >> + * Poll for sensor readiness; base the timeout upon the sample >> + * rate. >> + */ >> + rc = regmap_read_poll_timeout(data->regmap, DPS310_MEAS_CFG, >> + ready, ready & DPS310_TMP_RDY, >> + DPS310_POLL_SLEEP_US(timeout), >> + timeout); >> + if (rc < 0) > Why out of a sudden ' < 0'? Good point, I'll fix that. > >> + goto done; >> + } > As per above. > >> rc = dps310_read_temp_ready(data); >
On Mon, 15 Aug 2022 08:59:03 -0500 Eddie James <eajames@linux.ibm.com> wrote: > On 8/12/22 17:13, Andy Shevchenko wrote: > > On Wed, Aug 10, 2022 at 12:12 AM Eddie James <eajames@linux.ibm.com> wrote: > >> Corruption of the MEAS_CFG register has been observed soon after > >> system boot. In order to recover this scenario, check MEAS_CFG if > >> measurement isn't ready, and if it's incorrect, reset the DPS310 > >> and execute the startup procedure. > > Looks like both patches miss the Fixes tag. Can you add them? > > > Well this isn't really a software fix - there's no identifiable bug in > the driver. Just trying to recover the chip in this observed mystery > scenario. The tag is useful as well for where to backport this to. Probably just tag the driver introduction. Your description makes it clear we aren't finding faults in the driver - just that it didn't cover this undocumented case! Jonathan
diff --git a/drivers/iio/pressure/dps310.c b/drivers/iio/pressure/dps310.c index c706a8b423b5..3a1aeeea3cdd 100644 --- a/drivers/iio/pressure/dps310.c +++ b/drivers/iio/pressure/dps310.c @@ -393,6 +393,44 @@ static int dps310_get_temp_k(struct dps310_data *data) return scale_factors[ilog2(rc)]; } +/* + * Called with lock held. Returns a negative value on error, a positive value + * when the device is not ready, and zero when the device is ready. + */ +static int dps310_check_reset_meas_cfg(struct dps310_data *data, int ready_bit) +{ + int meas_cfg; + int rc = regmap_read(data->regmap, DPS310_MEAS_CFG, &meas_cfg); + + if (rc < 0) + return rc; + + /* Device is ready, proceed to measurement */ + if (meas_cfg & ready_bit) + return 0; + + /* Device is OK, just not ready */ + if (meas_cfg & (DPS310_PRS_EN | DPS310_TEMP_EN | DPS310_BACKGROUND)) + return 1; + + /* DPS310 register state corrupt, better start from scratch */ + rc = regmap_write(data->regmap, DPS310_RESET, DPS310_RESET_MAGIC); + if (rc < 0) + return rc; + + /* Wait for device chip access: 2.5ms in specification */ + usleep_range(2500, 12000); + + /* Reinitialize the chip */ + rc = dps310_startup(data); + if (rc) + return rc; + + dev_info(&data->client->dev, + "recovered from corrupted MEAS_CFG=%02x\n", meas_cfg); + return 1; +} + static int dps310_read_pres_raw(struct dps310_data *data) { int rc; @@ -405,16 +443,26 @@ static int dps310_read_pres_raw(struct dps310_data *data) if (mutex_lock_interruptible(&data->lock)) return -EINTR; - rate = dps310_get_pres_samp_freq(data); - timeout = DPS310_POLL_TIMEOUT_US(rate); - - /* Poll for sensor readiness; base the timeout upon the sample rate. */ - rc = regmap_read_poll_timeout(data->regmap, DPS310_MEAS_CFG, ready, - ready & DPS310_PRS_RDY, - DPS310_POLL_SLEEP_US(timeout), timeout); - if (rc) + rc = dps310_check_reset_meas_cfg(data, DPS310_PRS_RDY); + if (rc < 0) goto done; + if (rc > 0) { + rate = dps310_get_pres_samp_freq(data); + timeout = DPS310_POLL_TIMEOUT_US(rate); + + /* + * Poll for sensor readiness; base the timeout upon the sample + * rate. + */ + rc = regmap_read_poll_timeout(data->regmap, DPS310_MEAS_CFG, + ready, ready & DPS310_PRS_RDY, + DPS310_POLL_SLEEP_US(timeout), + timeout); + if (rc) + goto done; + } + rc = regmap_bulk_read(data->regmap, DPS310_PRS_BASE, val, sizeof(val)); if (rc < 0) goto done; @@ -454,16 +502,26 @@ static int dps310_read_temp_raw(struct dps310_data *data) if (mutex_lock_interruptible(&data->lock)) return -EINTR; - rate = dps310_get_temp_samp_freq(data); - timeout = DPS310_POLL_TIMEOUT_US(rate); - - /* Poll for sensor readiness; base the timeout upon the sample rate. */ - rc = regmap_read_poll_timeout(data->regmap, DPS310_MEAS_CFG, ready, - ready & DPS310_TMP_RDY, - DPS310_POLL_SLEEP_US(timeout), timeout); + rc = dps310_check_reset_meas_cfg(data, DPS310_TMP_RDY); if (rc < 0) goto done; + if (rc > 0) { + rate = dps310_get_temp_samp_freq(data); + timeout = DPS310_POLL_TIMEOUT_US(rate); + + /* + * Poll for sensor readiness; base the timeout upon the sample + * rate. + */ + rc = regmap_read_poll_timeout(data->regmap, DPS310_MEAS_CFG, + ready, ready & DPS310_TMP_RDY, + DPS310_POLL_SLEEP_US(timeout), + timeout); + if (rc < 0) + goto done; + } + rc = dps310_read_temp_ready(data); done: