From mboxrd@z Thu Jan 1 00:00:00 1970 From: Marek Vasut Date: Fri, 11 May 2018 13:31:49 +0200 Subject: [U-Boot] [PATCH v3 3/3] spi: mxs_spi: DM conversion In-Reply-To: References: <1525961829-17767-1-git-send-email-akash@openedev.com> <10d2a7f5-7908-7858-7cb2-caecd12e07bc@gmail.com> Message-ID: List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 8bit To: u-boot@lists.denx.de On 05/11/2018 01:09 PM, Akash Gajjar wrote: > Hello Marek, > > On Fri, May 11, 2018 at 4:09 PM, Marek Vasut > wrote: > > On 05/11/2018 12:08 PM, Gajjar Akash wrote: > > Hi Marek, > > > > Thanks for the review comments. > > > >     > -struct mxs_spi_slave { > >     > -     struct spi_slave        slave; > >     > -     uint32_t                max_khz; > >     > -     uint32_t                mode; > >     > -     struct mxs_ssp_regs     *regs; > >     > +struct mxs_spi_priv { > >     > +     struct mxs_ssp_regs *regs; > >     > +     u32     max_khz; > >     > +     u32     mode; > >     > +     u32     bus; > >     > +     u32     cs; > > > >     Type cleanup should be a separate patch > > > >   > > Okay, I will prepare seperate patch for type cleanup. > > > > > >     >  }; > >     >               if (mxs_wait_mask_set(&ssp_regs->hw_ssp_ctrl0_reg, > >     >                       SSP_CTRL0_RUN, MXS_SPI_MAX_TIMEOUT)) { > >     > -                     printf("MXS SPI: Timeout waiting for > start\n"); > >     > +                     debug("MXS SPI: Timeout waiting for > start\n"); > > > >     printf , we don't want to hide errors > > > >   > > okay, will revert it back to printf. > > > > > >     >                       return -ETIMEDOUT; > >     >               } > > > >     > + > >     > +#ifndef __SPI_MXS_H > >     > +#define __SPI_MXS_H > >     > + > >     > +struct mxs_spi_platdata { > >     > +     struct mxs_ssp_regs *regs; > >     > +     u32 bus; > >     > +     u32 max_hz; > >     > +     u32 cs; > > > >     Why is this header here at all ? > > > >   > > I didnt get this comment. do I need to place it somewhere else? > > See the beginning of this email, it seems the same structure exists > twice. > > > ​My intention was to have two individual structure for private and > platform data.​ > But now I could use one structre and access its members using two > structure variables(one for private and one for platadata). > > Is That looks okay? I do not quite understand what you mean, but there is a duplication of information here. That's a problem and should be fixed. -- Best regards, Marek Vasut