Re: [PATCH] SPI: DUAL and QUAD support

[Matthieu CASTET <matthieu.castet@parrot.com>]

Le Mon, 29 Jul 2013 15:25:51 +0100,
yuhang wang <wangyuhang2014@gmail.com> a écrit :

> Hi,
> 
> 2013/7/29 Matthieu CASTET <matthieu.castet@parrot.com>:
> > Le Mon, 29 Jul 2013 11:53:14 +0100,
> > wangyuhang <wangyuhang2014@gmail.com> a écrit :
> >
> >> Hi,
> >>
> >> modify two things.
> >> 1:
> >> >> @@ -1316,6 +1324,12 @@ int spi_setup(struct spi_device *spi)
> >> >>         /* help drivers fail *cleanly* when they need options
> >> >>          * that aren't supported with their current master
> >> >>          */
> >> >> +       if (((spi->mode >> 8) & 0x03) == 0x03 ||
> >> >> +               ((spi->mode >> 10) & 0x03) == 0x03) {
> >> >> +               dev_err(&spi->dev,
> >> >> +               "setup: can not select dual and quad at the same
> >> >> time\n")
> >> >> +               return -EINVAL;
> >> >> +       }
> >> >> +               return -EINVAL;
> >> >> +       }
> >>
> >> >This code won't work if the constants you added for mode flags
> >> >change value.  More importantly, anyone searching the code for
> >> >SPI_TX_DUAL, etc. won't find this code.
> >>
> >> so use the certain macro to replace.
> >>
> >> 2:
> >> >You need to make sure that existing userspace binaries can run
> >> >unmodified, changing the types will change the layout of the ioctl
> >> >arguments.  This may mean that you need to add new ioctls if you
> >> >need to change types.
> >>
> >> I recovered the SPI_IOC_RD_LSB_FIRST ioctl.
> >> Add new ioctl to fit u16 mode.
> >> In addition,may SPI_IOC_RD_LSB_FIRST should be del in a better way.
> >>
> >> Signed-off-by: wangyuhang <wangyuhang2014@gmail.com>
> >> ---
> >
> >>  /* IOCTL commands */
> >> @@ -54,6 +61,8 @@
> >>   * @tx_buf: Holds pointer to userspace buffer with transmit data,
> >> or null.
> >>   *     If no data is provided, zeroes are shifted out.
> >>   * @rx_buf: Holds pointer to userspace buffer for receive data, or
> >> null.
> >> + * @tx_nbits: number of bits used for writting.
> >> + * @rx_nbits: number of bits used for reading.
> >>   * @len: Length of tx and rx buffers, in bytes.
> >>   * @speed_hz: Temporary override of the device's bitrate.
> >>   * @bits_per_word: Temporary override of the device's wordsize.
> >> @@ -85,6 +94,8 @@
> >>  struct spi_ioc_transfer {
> >>         __u64           tx_buf;
> >>         __u64           rx_buf;
> >> +       __u8            tx_nbits;
> >> +       __u8            rx_nbits;
> >>
> >>         __u32           len;
> >>         __u32           speed_hz;
> > You still break old userspace ABI.
> >
> > You need to create a new structure, if you need to add fields.
> >
> I dont think adding the members will cause the program in userspace
> any errors. User do not have to set these members.
Did you try to run on your patch kernel a program build without the
patch ?


if I build my program with :

> >>  struct spi_ioc_transfer {
> >>         __u64           tx_buf;
> >>         __u64           rx_buf;
> >>
> >>         __u32           len;
> >>         __u32           speed_hz;

Now I run it on a new kernel.
The kernel won't parse correctly my structure : len will go in
tx_nbits, rx_nbits and len.

Keep in mind that it is binary interface, not source interface. You
should not have to rebuild your binary.

Matthieu