On Mon, 29 Oct 2018 14:39:24 +0100 Miquel Raynal <miquel.raynal@xxxxxxxxxxx> wrote: > Miquel Raynal <miquel.raynal@xxxxxxxxxxx> wrote on Mon, 29 Oct 2018 > 14:36:47 +0100: > > > Hi Boris, > > > > Boris Brezillon <boris.brezillon@xxxxxxxxxxx> wrote on Tue, 23 Oct 2018 > > 20:50:02 +0200: > > > > > Add a wrapper to prevent drivers and core code from directly calling > > > the ->select_chip hook which we are about to deprecate. > > > > > > Signed-off-by: Boris Brezillon <boris.brezillon@xxxxxxxxxxx> > > > --- > > > drivers/mtd/nand/raw/gpmi-nand/gpmi-nand.c | 23 +++-- > > > drivers/mtd/nand/raw/jz4740_nand.c | 4 +- > > > drivers/mtd/nand/raw/nand_base.c | 114 ++++++++++++++------- > > > drivers/mtd/nand/raw/r852.c | 4 +- > > > include/linux/mtd/rawnand.h | 4 + > > > 5 files changed, 98 insertions(+), 51 deletions(-) > > > > > > > So far I am glad to see all these changes. > > > > About the ->select_chip() removal, I wonder if it would not be better > > to also change the local variables "chipnr" or "chip_number" (or > > even "i") that suggest that this ID selects a chip, while it > > actually selects a die in a chip (and it is possible to have multiple > > die on a chip, so multiple CS for one single NAND chip). I agree. > > > > Do you think it is worth the change ? If yes, would it fit in this patch > > or is it better to do this change elsewhere? > > > > This request actually applies to the following patches as well. Maybe we > could even find a uniform way to name it, "die_nr" or something like > this? Target is the name used in the ONFI spec, hence the function names. Note that die is not accurate since you might have several dies exposed through a single CS line (that's called LUNs). ______________________________________________________ Linux MTD discussion mailing list http://lists.infradead.org/mailman/listinfo/linux-mtd/