Re: [PATCH 3/5] sandbox: parse libftdi options

[Sascha Hauer <s.hauer@pengutronix.de>]

On Thu, Oct 19, 2017 at 02:55:40PM +0300, Antony Pavlov wrote:
> On Mon, 16 Oct 2017 16:06:18 +0200
> Sascha Hauer <s.hauer@pengutronix.de> wrote:
> 
> > On Mon, Oct 16, 2017 at 05:10:11PM +0300, Antony Pavlov wrote:
> > > On Mon, 16 Oct 2017 10:21:58 +0200
> > > Sascha Hauer <s.hauer@pengutronix.de> wrote:
> > > 
> > > > On Tue, Oct 10, 2017 at 03:26:29PM +0300, Antony Pavlov wrote:
> > > > > Signed-off-by: Antony Pavlov <antonynpavlov@gmail.com>
> > > > > ---
> > > > >  arch/sandbox/Makefile    |  2 +-
> > > > >  arch/sandbox/os/common.c | 12 ++++++--
> > > > >  arch/sandbox/os/ftdi.c   | 79 +++++++++++++++++++++++++++++++++++++++++++++++-
> > > > >  3 files changed, 89 insertions(+), 4 deletions(-)
> > > > > 
> > > > > diff --git a/arch/sandbox/os/ftdi.c b/arch/sandbox/os/ftdi.c
> > > > > index 34e9165787..e3e46ed12d 100644
> > > > > --- a/arch/sandbox/os/ftdi.c
> > > > > +++ b/arch/sandbox/os/ftdi.c
> > > > > @@ -20,6 +20,7 @@
> > > > >  #include <unistd.h>
> > > > >  #include <ftdi.h>
> > > > >  #include <errno.h>
> > > > > +#include <string.h>
> > > > >  #include <mach/linux.h>
> > > > >  
> > > > >  #define FTDI_VID		0x0403	/* Vendor Id */
> > > > > @@ -38,6 +39,8 @@ struct ft2232_bitbang {
> > > > >  
> > > > >  static struct ft2232_bitbang ftbb;
> > > > >  
> > > > > +extern const char *libftdi_options;
> > > > > +
> > > > >  static inline int ftdi_flush(struct ftdi_context *ftdi)
> > > > >  {
> > > > >  	uint8_t buf[1];
> > > > > @@ -116,6 +119,67 @@ void barebox_libftdi1_gpio_set_value(struct ft2232_bitbang *ftbb,
> > > > >  		ftbb->odata &= ~BIT(off);
> > > > >  }
> > > > >  
> > > > > +/* This is a somewhat hacked function similar in some ways to strtok().
> > > > > + * It will look for needle with a subsequent '=' in haystack, return a copy of
> > > > > + * needle and remove everything from the first occurrence of needle to the next
> > > > > + * delimiter from haystack.
> > > > > + */
> > > > > +static char *extract_param(const char *const *haystack, const char *needle,
> > > > > +			const char *delim)
> > > > > +{
> > > > 
> > > > Parsing comma separated option lists is something we do more than once.
> > > > Right now we already have parseopt_b and parseopt_hu. Would be nice to
> > > > have this function alongside with the existing functions. Also
> > > > parseopt_* look simpler to follow, it may be worth adopting the code for
> > > > this function.
> > > 
> > > At the moment the parseopt_* functions are in the fs/parseopt.c file.
> > > Adding the parseopt_ul() for parsing u32 option value in arch/sandbox/os/ftdi.c
> > > will lead to moving fs/parseopt.c file to common code, e.g. to lib/.
> > > 
> > > Is it ok to add 'obj-y += parseopt.o' to the lib/Makefile file?
> > 
> > obj-y should be fine. ATM fs/parseopt.c is always compiled aswell.
> 
> Currently parseopt_hu() returns void and has no explicit flag for parsing error
> situation. I think that it is reasonable to use explicit error flag in new
> parseopt_ul() for parsing u32 option value, e.g. use function return value
> as an error flag.
> 
> Any suggestion?

Yes, returning an error is a good idea and given that the return value
is currently unused this is a good place.

Sascha

-- 
Pengutronix e.K.                           |                             |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |
Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |

_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox