> > Le 16/12/2016 à 14:47, Krzeminski, Marcin (Nokia - PL/Wroclaw) a écrit : > >> -----Original Message----- > >> From: Krzeminski, Marcin (Nokia - PL/Wroclaw) > >> Sent: Tuesday, December 13, 2016 10:46 AM > >> To: Cyrille Pitchen <[email protected]>; > >> [email protected]; [email protected]; > >> [email protected]; [email protected]; linux- > >> [email protected] > >> Cc: [email protected]; [email protected] > >> Subject: RE: [PATCH v4 4/8] mtd: spi-nor: add support of SPI protocols > >> like SPI > >> 1-2-2 and SPI 1-4-4 > >> > >> Cyrille, > >> > >>> -----Original Message----- > >>> From: linux-mtd [mailto:[email protected]] On > >>> Behalf Of Cyrille Pitchen > >>> Sent: Monday, November 21, 2016 3:16 PM > >>> To: [email protected]; [email protected]; > >>> [email protected]; [email protected]; linux- > >>> [email protected] > >>> Cc: Cyrille Pitchen <[email protected]>; > >>> [email protected]; [email protected] > >>> Subject: [PATCH v4 4/8] mtd: spi-nor: add support of SPI protocols > >>> like SPI 1- > >>> 2-2 and SPI 1-4-4 > >>> > >>> This patch changes the prototype of spi_nor_scan(): its 3rd parameter > >>> is replaced by a const struct spi_nor_modes pointer, which tells the > >>> spi-nor framework about which SPI protocols are supported by the SPI > >> controller. > >>> > >>> Besides, this patch also introduces a new > >>> spi_nor_basic_flash_parameter structure telling the spi-nor framework > >>> about the SPI protocols supported by the SPI memory and the needed op > >> codes to use these SPI protocols. > >>> > >>> Currently the struct spi_nor_basic_flash_parameter is filled with > >>> legacy values but a later patch will allow to fill it dynamically by > >>> reading the > >>> JESD216 Serial Flash Discoverable Parameter (SFDP) tables from the SPI > >>> memory. > >>> > >>> With both structures, the spi-nor framework can now compute the best > >>> match between SPI protocols supported by both the (Q)SPI memory and > >>> controller hence selecting the relevant op codes for (Fast) Read, Page > >>> Program and Sector Erase operations. > >>> > >>> The spi_nor_basic_flash_parameter structure also provides the spi-nor > >>> framework with the number of dummy cycles to be used with each Fast > >>> Read commands and the erase block size associated to the erase block > >>> op codes. > >>> > >>> Finally the spi_nor_basic_flash_parameter structure, through the > >>> optional > >>> .enable_quad_io() hook, tells the spi-nor framework how to set the > >>> Quad Enable (QE) bit of the QSPI memory to enable its Quad SPI features. > >>> > >>> Signed-off-by: Cyrille Pitchen <[email protected]> > >>> --- > >>> drivers/mtd/devices/m25p80.c | 17 ++- > >>> drivers/mtd/spi-nor/atmel-quadspi.c | 83 ++++++---- > >>> drivers/mtd/spi-nor/cadence-quadspi.c | 18 ++- > >>> drivers/mtd/spi-nor/fsl-quadspi.c | 8 +- > >>> drivers/mtd/spi-nor/hisi-sfc.c | 32 +++- > >>> drivers/mtd/spi-nor/mtk-quadspi.c | 16 +- > >>> drivers/mtd/spi-nor/nxp-spifi.c | 21 +-- > >>> drivers/mtd/spi-nor/spi-nor.c | 280 > >> +++++++++++++++++++++++++++- > >>> ------ > [...] > >>> +static int spi_nor_setup(struct spi_nor *nor, const struct flash_info > >>> *info, > >>> + const struct spi_nor_basic_flash_parameter > >>> *params, > >>> + const struct spi_nor_modes *modes) > >>> { > >>> + bool enable_quad_io; > >>> + u32 rd_modes, wr_modes, mask; > >>> + const struct spi_nor_erase_type *erase_type = NULL; > >>> + const struct spi_nor_read *read; > >>> + int rd_pindex, wr_pindex, i, err = 0; > >>> + u8 erase_size = SNOR_ERASE_64K; > >> > >> Erase size could be configurable, then user can chose best sector size that > >> match his use case on multi-sized flash. > >> > >>> + > >>> + /* 2-2-2 or 4-4-4 modes are not supported yet. */ > >>> + mask = (SNOR_MODE_2_2_2 | SNOR_MODE_4_4_4); > > IMHO could be nice to put a warning here :) > > > > Then maybe only a dev_dbg() because many developers complain that there are > already too many debug messages in the boot logs.
Sure, also if-ed printout when user request 222 o 444 mode will be fine. Just imagine reaction of an impulsive programmer who requested Full Dual/Quad mode and framework silently ignore it :) Thanks, Marcin > > > Thanks, > > Marcin > > >
