> Subject: Re: [PATCH v2 lora-next 1/4] net: lora: sx1301: > convert burst spi functions to regmap raw > > Hi Ben and Mark, > > Am 09.10.18 um 14:52 schrieb Ben Whitten: > > As we have caching disabled we can access the regmap > using raw for our > > firmware reading and writing bursts. > > We also remove the now defunct spi element from the > structure as this > > completes the move to regmap. > > > > Signed-off-by: Ben Whitten > <ben.whitten@xxxxxxxxxxxxx> > > --- > > drivers/net/lora/sx1301.c | 26 +++++++++++++++++------ > --- > > drivers/net/lora/sx1301.h | 2 -- > > 2 files changed, 17 insertions(+), 11 deletions(-) > > > > diff --git a/drivers/net/lora/sx1301.c > b/drivers/net/lora/sx1301.c > > index fd29258..5ab0e2d 100644 > > --- a/drivers/net/lora/sx1301.c > > +++ b/drivers/net/lora/sx1301.c > > @@ -76,19 +76,28 @@ static struct regmap_config > sx1301_regmap_config = { > > > > static int sx1301_read_burst(struct sx1301_priv *priv, u8 > reg, u8 *val, size_t len) > > { > > - u8 addr = reg & 0x7f; > > - return spi_write_then_read(priv->spi, &addr, 1, val, > len); > > + size_t max; > > + > > + max = regmap_get_raw_read_max(priv->regmap); > > + if (max && max < len) { > > + dev_err(priv->dev, "Burst greater then max > raw read\n"); > > + return -EINVAL; > > + } > > + > > + return regmap_raw_read(priv->regmap, reg, val, > len); > > I believe the way we are using it with a single register needs > regmap_noinc_read() instead, in case we ever want to > enable caching? ... Certainly, the noinc was introduced in 4.19 lets rebase lora-next and use this API instead. > > > } > > > > static int sx1301_write_burst(struct sx1301_priv *priv, u8 > reg, const u8 *val, size_t len) > > { > > - u8 addr = reg | BIT(7); > > - struct spi_transfer xfr[2] = { > > - { .tx_buf = &addr, .len = 1 }, > > - { .tx_buf = val, .len = len }, > > - }; > > + size_t max; > > + > > + max = regmap_get_raw_write_max(priv->regmap); > > + if (max && max < len) { > > + dev_err(priv->dev, "Burst greater then max > raw write\n"); > > + return -EINVAL; > > + } > > > > - return spi_sync_transfer(priv->spi, xfr, 2); > > + return regmap_raw_write(priv->regmap, reg, val, > len); > > ... Which would mean we are lacking a noinc API for write > here! > > @Mark/Stefan, any reason noinc was not implemented > symmetrically? > > For sx1276 I have local regmap conversion patches, but I > kept getting > -EINVAL for bursts and haven't figured out why yet. Need to > compare to > your code - I assume you successfully tested this, Ben. Yes I was able to load and verify/run the firmware using this patch on my Sentrius RG1xx board. Thanks, Ben