Hi Sebastian, On lun., nov. 30 2015, Sebastian Hesselbarth <sebastian.hesselbarth@xxxxxxxxx> wrote: > On 29.11.2015 15:35, Thomas Petazzoni wrote: >> Adding Ezequiel Garcia in Cc. >> >> On Sat, 28 Nov 2015 12:14:08 +0100, Sebastian Hesselbarth wrote: >>> The NAND device found on Lenovo ix4-300d uses 4-bit BCH ECC protection. >>> Add the corresponding properties to the NAND node. >> >> If the ONFI information from the NAND flash say that it requires 4 bits >> per 512, then there should be no need to add this information to the >> Device Tree as the pxa3xx_nand driver by default uses the ONFI >> information. > > Thomas, > > as said in the cover letter, this is also DT cleanup with barebox > bootloader in mind. I do not accept what Linux' pxa3xx_nand driver > is doing as a reference here ;) > >> Those properties are only needed when for some reason the vendor has >> chosen to use a ECC strength that doesn't match with the one advertised >> by the flash in its ONFI information (either stronger or weaker). But >> in this case, your commit log is confusing, because it says that the >> "NAND device ... uses 4-bit BCH ECC protection". If it really does, >> then the patch is not needed :-) > > I agree that if ONFI is already advertising 4/512 ECC (and it is), we > do not need the properties. Anyway, IIRC barebox does not yet properly > parse ONFI or at least it does not derive minimum ECC settings from it. > > I'll have to have a closer look at barebox' ONFI parsing capabilites > and can live with this patch not applied even though it does no harm. So for now, I don't apply it. Thanks, Gregory -- Gregory Clement, Free Electrons Kernel, drivers, real-time and embedded Linux development, consulting, training and support. http://free-electrons.com -- To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
