Message ID | 20200911144703.25409-2-m.felsch@pengutronix.de |
---|---|
State | Changes Requested |
Delegated to: | Vignesh R |
Headers | show |
Series | MTD: SST SPI-NOR fixes | expand |
On 9/11/20 8:17 PM, Marco Felsch wrote: > The sst write support for devices using the special SST_WRITE routine > is broken since commit commit df5c21002cf4 ("mtd: spi-nor: use spi-mem > dirmap API") because the spi_nor_create_write_dirmap() function checks > SPINOR_OP_AAI_WP and sst_write_second. These checks are not valid during > probe. The check seems also to be broken since the "op->addr.nbytes = 0" > causes the devm_spi_mem_dirmap_create() function to return > PTR_ERR(-EINVAL) and the probe() function will fail. > > It seems that the commit only copy'n'pasted the existing logic. Use the > correct SST_WRITE flag and return 0 to fix both issues. > > Fixes: df5c21002cf4 ("mtd: spi-nor: use spi-mem dirmap API") > Signed-off-by: Marco Felsch <m.felsch@pengutronix.de> > --- > drivers/mtd/spi-nor/core.c | 12 +++++++++--- > 1 file changed, 9 insertions(+), 3 deletions(-) > > diff --git a/drivers/mtd/spi-nor/core.c b/drivers/mtd/spi-nor/core.c > index 65eff4ce6ab1..31869ac245a8 100644 > --- a/drivers/mtd/spi-nor/core.c > +++ b/drivers/mtd/spi-nor/core.c > @@ -3289,15 +3289,21 @@ static int spi_nor_create_write_dirmap(struct spi_nor *nor) > }; > struct spi_mem_op *op = &info.op_tmpl; > > + /* > + * Most SST SPI-NOR's have a special write routine.which should so s/SPI-NOR/SPI NOR. > + * dirmap.wdesc is not supported for these. Or How about more readable version: "Most SST flashes have special sequence for writing data to the flash and therefore cannot support writes through direct mapping APIs." > + */ > + if (nor->info->flags & SST_WRITE) { > + nor->dirmap.wdesc = NULL; nor->dirmap.wdesc is known to be NULL at this point. So no need to set to NULL again. > + return 0; > + } > + > /* get transfer protocols. */ > op->cmd.buswidth = spi_nor_get_protocol_inst_nbits(nor->write_proto); > op->addr.buswidth = spi_nor_get_protocol_addr_nbits(nor->write_proto); > op->dummy.buswidth = op->addr.buswidth; > op->data.buswidth = spi_nor_get_protocol_data_nbits(nor->write_proto); > > - if (nor->program_opcode == SPINOR_OP_AAI_WP && nor->sst_write_second) > - op->addr.nbytes = 0; > - > nor->dirmap.wdesc = devm_spi_mem_dirmap_create(nor->dev, nor->spimem, > &info); > return PTR_ERR_OR_ZERO(nor->dirmap.wdesc); >
On 9/14/20 3:00 PM, Vignesh Raghavendra wrote: > EXTERNAL EMAIL: Do not click links or open attachments unless you know the content is safe > > On 9/11/20 8:17 PM, Marco Felsch wrote: >> The sst write support for devices using the special SST_WRITE routine >> is broken since commit commit df5c21002cf4 ("mtd: spi-nor: use spi-mem >> dirmap API") because the spi_nor_create_write_dirmap() function checks >> SPINOR_OP_AAI_WP and sst_write_second. These checks are not valid during >> probe. The check seems also to be broken since the "op->addr.nbytes = 0" >> causes the devm_spi_mem_dirmap_create() function to return >> PTR_ERR(-EINVAL) and the probe() function will fail. >> >> It seems that the commit only copy'n'pasted the existing logic. Use the >> correct SST_WRITE flag and return 0 to fix both issues. >> >> Fixes: df5c21002cf4 ("mtd: spi-nor: use spi-mem dirmap API") >> Signed-off-by: Marco Felsch <m.felsch@pengutronix.de> >> --- >> drivers/mtd/spi-nor/core.c | 12 +++++++++--- >> 1 file changed, 9 insertions(+), 3 deletions(-) >> >> diff --git a/drivers/mtd/spi-nor/core.c b/drivers/mtd/spi-nor/core.c >> index 65eff4ce6ab1..31869ac245a8 100644 >> --- a/drivers/mtd/spi-nor/core.c >> +++ b/drivers/mtd/spi-nor/core.c >> @@ -3289,15 +3289,21 @@ static int spi_nor_create_write_dirmap(struct spi_nor *nor) >> }; >> struct spi_mem_op *op = &info.op_tmpl; >> >> + /* >> + * Most SST SPI-NOR's have a special write routine.which should so > > s/SPI-NOR/SPI NOR. > >> + * dirmap.wdesc is not supported for these. > > Or How about more readable version: > > "Most SST flashes have special sequence for writing data to the flash > and therefore cannot support writes through direct mapping APIs." > >> + */ >> + if (nor->info->flags & SST_WRITE) { >> + nor->dirmap.wdesc = NULL; > > nor->dirmap.wdesc is known to be NULL at this point. So no need to set > to NULL again. > >> + return 0; >> + } >> + >> /* get transfer protocols. */ >> op->cmd.buswidth = spi_nor_get_protocol_inst_nbits(nor->write_proto); >> op->addr.buswidth = spi_nor_get_protocol_addr_nbits(nor->write_proto); >> op->dummy.buswidth = op->addr.buswidth; >> op->data.buswidth = spi_nor_get_protocol_data_nbits(nor->write_proto); >> >> - if (nor->program_opcode == SPINOR_OP_AAI_WP && nor->sst_write_second) >> - op->addr.nbytes = 0; >> - >> nor->dirmap.wdesc = devm_spi_mem_dirmap_create(nor->dev, nor->spimem, >> &info); >> return PTR_ERR_OR_ZERO(nor->dirmap.wdesc); >> With Vignesh's comments addressed, one can add: Reviewed-by: Tudor Ambarus <tudor.ambarus@microchip.com>
diff --git a/drivers/mtd/spi-nor/core.c b/drivers/mtd/spi-nor/core.c index 65eff4ce6ab1..31869ac245a8 100644 --- a/drivers/mtd/spi-nor/core.c +++ b/drivers/mtd/spi-nor/core.c @@ -3289,15 +3289,21 @@ static int spi_nor_create_write_dirmap(struct spi_nor *nor) }; struct spi_mem_op *op = &info.op_tmpl; + /* + * Most SST SPI-NOR's have a special write routine.which should so + * dirmap.wdesc is not supported for these. + */ + if (nor->info->flags & SST_WRITE) { + nor->dirmap.wdesc = NULL; + return 0; + } + /* get transfer protocols. */ op->cmd.buswidth = spi_nor_get_protocol_inst_nbits(nor->write_proto); op->addr.buswidth = spi_nor_get_protocol_addr_nbits(nor->write_proto); op->dummy.buswidth = op->addr.buswidth; op->data.buswidth = spi_nor_get_protocol_data_nbits(nor->write_proto); - if (nor->program_opcode == SPINOR_OP_AAI_WP && nor->sst_write_second) - op->addr.nbytes = 0; - nor->dirmap.wdesc = devm_spi_mem_dirmap_create(nor->dev, nor->spimem, &info); return PTR_ERR_OR_ZERO(nor->dirmap.wdesc);
The sst write support for devices using the special SST_WRITE routine is broken since commit commit df5c21002cf4 ("mtd: spi-nor: use spi-mem dirmap API") because the spi_nor_create_write_dirmap() function checks SPINOR_OP_AAI_WP and sst_write_second. These checks are not valid during probe. The check seems also to be broken since the "op->addr.nbytes = 0" causes the devm_spi_mem_dirmap_create() function to return PTR_ERR(-EINVAL) and the probe() function will fail. It seems that the commit only copy'n'pasted the existing logic. Use the correct SST_WRITE flag and return 0 to fix both issues. Fixes: df5c21002cf4 ("mtd: spi-nor: use spi-mem dirmap API") Signed-off-by: Marco Felsch <m.felsch@pengutronix.de> --- drivers/mtd/spi-nor/core.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-)