| Message ID | 20171213113530.6857-1-sean.nyekjaer@prevas.dk |
|---|---|
| State | Changes Requested |
| Delegated to: | Cyrille Pitchen |
| Headers | show |
| Series | [v2,1/2] mtd: spi-nor: add support for is25lq040b | expand |
Hi Cyrille On 2017-12-13 12:35, Sean Nyekjaer wrote: > Signed-off-by: Sean Nyekjaer <sean.nyekjaer@prevas.dk> > --- > v2: > - removed SECT_4K as the chip are supporting 32K and bulk erase > > drivers/mtd/spi-nor/spi-nor.c | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/drivers/mtd/spi-nor/spi-nor.c b/drivers/mtd/spi-nor/spi-nor.c > index 9178139a39d0..cc4328dff89a 100644 > --- a/drivers/mtd/spi-nor/spi-nor.c > +++ b/drivers/mtd/spi-nor/spi-nor.c > @@ -1048,6 +1048,7 @@ static const struct flash_info spi_nor_ids[] = { > { "is25cd512", INFO(0x7f9d20, 0, 32 * 1024, 2, SECT_4K) }, > { "is25lp128", INFO(0x9d6018, 0, 64 * 1024, 256, > SECT_4K | SPI_NOR_DUAL_READ) }, > + { "is25lq040b", INFO(0x9d4013, 0, 64 * 1024, 64, 0) }, Doh, my bad, there is only 8 blocks on the device (4Mbit not 4MB) The device have the possibility to erase 4, 32, 64 Kbyte, or Chip Erase. Is the SECT_4K preferred in the kernel or should i leave it off? > > /* Macronix */ > { "mx25l512e", INFO(0xc22010, 0, 64 * 1024, 1, SECT_4K) }, /Sean
Hi Sean, Le 14/12/2017 à 11:21, Sean Nyekjær a écrit : > Hi Cyrille > > On 2017-12-13 12:35, Sean Nyekjaer wrote: >> Signed-off-by: Sean Nyekjaer <sean.nyekjaer@prevas.dk> >> --- >> v2: >> - removed SECT_4K as the chip are supporting 32K and bulk erase >> >> drivers/mtd/spi-nor/spi-nor.c | 1 + >> 1 file changed, 1 insertion(+) >> >> diff --git a/drivers/mtd/spi-nor/spi-nor.c b/drivers/mtd/spi-nor/spi-nor.c >> index 9178139a39d0..cc4328dff89a 100644 >> --- a/drivers/mtd/spi-nor/spi-nor.c >> +++ b/drivers/mtd/spi-nor/spi-nor.c >> @@ -1048,6 +1048,7 @@ static const struct flash_info spi_nor_ids[] = { >> { "is25cd512", INFO(0x7f9d20, 0, 32 * 1024, 2, SECT_4K) }, >> { "is25lp128", INFO(0x9d6018, 0, 64 * 1024, 256, >> SECT_4K | SPI_NOR_DUAL_READ) }, >> + { "is25lq040b", INFO(0x9d4013, 0, 64 * 1024, 64, 0) }, > Doh, my bad, there is only 8 blocks on the device (4Mbit not 4MB) > > The device have the possibility to erase 4, 32, 64 Kbyte, or Chip Erase. > Is the SECT_4K preferred in the kernel or should i leave it off? If 4K erase is uniformly supported by the memory, then please add the SECT_4K flag. I'm looking at the datasheet [1]: - Uniform 4 Kbyte Sectors or 32/64 Kbyte Blocks: please add the SECT_4K flag - FAST READ QUAD OUTPUT (FRQO, 6Bh): Fast Read 1-1-4 supported, then please add the SPI_NOR_QUAD_READ flag too - FAST READ DUAL OUTPUT (FRDO, 3Bh): Fast Read 1-1-2 supported, also add the SPI_NOR_DUAL_READ flag. Besides, setting at least one the the SPI_NOR_{DUAL,QUAD}_READ flags will enable parsing of the SFDP tables, which are also supported by this memory: "Supports Serial Flash Discoverable Parameters (SFDP)" So you should set the 3 flags: SECT_4K | SPI_NOR_DUAL_READ | SPI_NOR_QUAD_READ Best regards, Cyrille [1] http://www.issi.com/WW/pdf/25LQ025B-512B-010B-020B-040B.pdf >> /* Macronix */ >> { "mx25l512e", INFO(0xc22010, 0, 64 * 1024, 1, SECT_4K) }, > /Sean >
diff --git a/drivers/mtd/spi-nor/spi-nor.c b/drivers/mtd/spi-nor/spi-nor.c index 9178139a39d0..cc4328dff89a 100644 --- a/drivers/mtd/spi-nor/spi-nor.c +++ b/drivers/mtd/spi-nor/spi-nor.c @@ -1048,6 +1048,7 @@ static const struct flash_info spi_nor_ids[] = { { "is25cd512", INFO(0x7f9d20, 0, 32 * 1024, 2, SECT_4K) }, { "is25lp128", INFO(0x9d6018, 0, 64 * 1024, 256, SECT_4K | SPI_NOR_DUAL_READ) }, + { "is25lq040b", INFO(0x9d4013, 0, 64 * 1024, 64, 0) }, /* Macronix */ { "mx25l512e", INFO(0xc22010, 0, 64 * 1024, 1, SECT_4K) },
Signed-off-by: Sean Nyekjaer <sean.nyekjaer@prevas.dk> --- v2: - removed SECT_4K as the chip are supporting 32K and bulk erase drivers/mtd/spi-nor/spi-nor.c | 1 + 1 file changed, 1 insertion(+)