Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1487832
| From | Joe Perches <joe@perches.com> |
|---|---|
| Newsgroups | linux.kernel |
| Subject | Re: iomux-mx3.h: possible macro precedence issue |
| Date | 2016-09-21 08:20 +0200 |
| Message-ID | <sjJ2V-5pe-1@gated-at.bofh.it> (permalink) |
| References | <sjzwC-7F4-33@gated-at.bofh.it> <sjITg-5i1-9@gated-at.bofh.it> |
| Organization | linux.* mail to news gateway |
On Wed, 2016-09-21 at 08:06 +0200, Uwe Kleine-König wrote:\
> On Tue, Sep 20, 2016 at 01:02:24PM -0700, Joe Perches wrote:
> > Julia Lawall wrote a script
> >
> > Link: http://lkml.kernel.org/r/alpine.DEB.2.10.1609201503260.2914@hadrien
> >
> > that found a possible issue with macro argument precedence.
> >
> >
> > diff -u -p a/arch/arm/mach-imx/iomux-mx3.h b/arch/arm/mach-imx/iomux-mx3.h
> > --- a/arch/arm/mach-imx/iomux-mx3.h
> > +++ b/arch/arm/mach-imx/iomux-mx3.h
> > @@ -529,7 +529,7 @@ enum iomux_pins {
> > #define MX31_PIN_DCD_DTE1__DCD_DTE2 IOMUX_MODE(MX31_PIN_DCD_DTE1, IOMUX_CONFIG_ALT1)
> > #define MX31_PIN_RI_DTE1__RI_DTE2 IOMUX_MODE(MX31_PIN_RI_DTE1, IOMUX_CONFIG_ALT1)
> > #define MX31_PIN_DSR_DTE1__DSR_DTE2 IOMUX_MODE(MX31_PIN_DSR_DTE1, IOMUX_CONFIG_ALT1)
> > -#define MX31_PIN_DTR_DTE1__DTR_DTE2 IOMUX_MODE(MX31_PIN_DTR_DTE1, IOMUX_OCONFIG_ALT3 | IOMUX_ICONFIG_NONE)
> > +#define MX31_PIN_DTR_DTE1__DTR_DTE2 IOMUX_MODE(MX31_PIN_DTR_DTE1, (IOMUX_OCONFIG_ALT3 | IOMUX_ICONFIG_NONE))
> > #define MX31_PIN_PC_RST__CTS5 IOMUX_MODE(MX31_PIN_PC_RST, IOMUX_CONFIG_ALT2)
> > #define MX31_PIN_PC_VS2__RTS5 IOMUX_MODE(MX31_PIN_PC_VS2, IOMUX_CONFIG_ALT2)
> > #define MX31_PIN_PC_BVD2__TXD5 IOMUX_MODE(MX31_PIN_PC_BVD2, IOMUX_CONFIG_ALT2)
>
>
> I assume this is the only problematic definition?
yes..
> If so, there is a
> single affected platform: arch/arm/mach-imx/mach-kzm_arm11_01.c
>
> > this may be intentional, but perhaps a solution
> > would be to use parentheses in the #define IOMUX_MODE
> >
> > ---
> >
> > diff --git a/arch/arm/mach-imx/iomux-mx3.h b/arch/arm/mach-imx/iomux-mx3.h
> > index 368667b..3f9ede2 100644
> > --- a/arch/arm/mach-imx/iomux-mx3.h
> > +++ b/arch/arm/mach-imx/iomux-mx3.h
> > @@ -156,7 +156,7 @@ void mxc_iomux_mode(unsigned int pin_mode);
> > > > (((gpionum << IOMUX_GPIONUM_SHIFT) & IOMUX_GPIONUM_MASK) | \
> > > > (padnum & IOMUX_PADNUM_MASK))
> >
> > -#define IOMUX_MODE(pin, mode) (pin | mode << IOMUX_MODE_SHIFT)
> > +#define IOMUX_MODE(pin, mode) ((pin) | ((mode) << IOMUX_MODE_SHIFT))
> >
> > #define IOMUX_TO_GPIO(iomux_pin) \
> > ((iomux_pin & IOMUX_GPIONUM_MASK) >> IOMUX_GPIONUM_SHIFT)
>
>
> My preference is to fix IOMUX_MODE instead of adding the parents to
> MX31_PIN_DTR_DTE1__DTR_DTE2.
Mine too.
> Joe, do you create a proper patch?
I just don't have the hardware and don't know what was intended
by the existing code. It does appear to be a defect though.
Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread
iomux-mx3.h: possible macro precedence issue Joe Perches <joe@perches.com> - 2016-09-20 22:10 +0200
Re: iomux-mx3.h: possible macro precedence issue Uwe Kleine-König <u.kleine-koenig@pengutronix.de> - 2016-09-21 08:10 +0200
Re: iomux-mx3.h: possible macro precedence issue Joe Perches <joe@perches.com> - 2016-09-21 08:20 +0200
Re: iomux-mx3.h: possible macro precedence issue Uwe Kleine-König <u.kleine-koenig@pengutronix.de> - 2016-09-21 08:50 +0200
csiph-web