On Wed, 28 Nov 2018 14:17:12 +0000 <[email protected]> wrote:
> On 11/28/2018 09:57 AM, Boris Brezillon wrote: > > On Tue, 20 Nov 2018 11:55:21 +0000 > > <[email protected]> wrote: > > > >> + > >> + /* > >> + * We set nor->addr_width here to skip spi_nor_set_4byte_opcodes() > >> + * later because this latest function implements a legacy quirk for > >> + * the erase size of Spansion memory. However this quirk is no longer > >> + * needed with new SFDP compliant memories. > >> + */ > >> + nor->addr_width = 4; > >> + nor->flags |= SPI_NOR_4B_OPCODES; > > > > You mean SNOR_F_4B_OPCODES (the one introduced here [1]), because > > SPI_NOR_4B_OPCODES should only be used for flash_info->flags and might > > soon conflict with another SNOR_F_ flag? > > > > yes, you're right. > > > [1]http://patchwork.ozlabs.org/patch/991476/ > > > > Can you apply your patch? Will submit a new version afterwards. Actually, I realized setting SNOR_F_4B_OPCODES when the BFPT advertises 4_BYTES_ONLY is incorrect as 4bytes only can mean "use the 3B opcodes but pass address on 4 bytes". Here is a new version of this patch [1]. Feel free to pick it up and send it along with your "SFDP 4-byte Address Instruction Table" patch (I have not reason to send it alone since the problem I was trying to solve is no longer fixed by [1]). [1]https://github.com/bbrezillon/linux-0day/commit/a953b6b435ec67bca00d6666f472db5f6dca4f63
