2010-06-16 18:30:31

by John Sheehan

Subject: [PATCH 1/2] Staging: comedi: fix style errors in adl_pci9111.c

From: John Sheehan <[email protected]>

a patch for adl_pci9111.c, to fix the following
errors reported by the checkpatch.pl tool,

-Macros with complex values should be enclosed in parenthesis
-need consistent spacing around '&' (ctx:WxV)
-spaces required around that '==' (ctx:VxV)
-Macros with complex values should be enclosed in parenthesis
-trailing statements should be on next line

Signed-off-by: John Sheehan <[email protected]>
---
drivers/staging/comedi/drivers/adl_pci9111.c | 25 +++++++++++++------------
1 files changed, 13 insertions(+), 12 deletions(-)

diff --git a/drivers/staging/comedi/drivers/adl_pci9111.c b/drivers/staging/comedi/drivers/adl_pci9111.c
index 36a254c..b9501ce 100644
--- a/drivers/staging/comedi/drivers/adl_pci9111.c
+++ b/drivers/staging/comedi/drivers/adl_pci9111.c
@@ -177,7 +177,7 @@ TODO:
#define PCI9111_FIFO_FULL_MASK 0x40
#define PCI9111_AD_BUSY_MASK 0x80

-#define PCI9111_IO_BASE dev->iobase
+#define PCI9111_IO_BASE (dev->iobase)

/*
* Define inlined function
@@ -190,7 +190,7 @@ TODO:
outb(flags, PCI9111_IO_BASE+PCI9111_REGISTER_TRIGGER_MODE_CONTROL)

#define pci9111_interrupt_and_fifo_get() \
- ((inb(PCI9111_IO_BASE+PCI9111_REGISTER_AD_MODE_INTERRUPT_READBACK) >> 4) &0x03)
+ ((inb(PCI9111_IO_BASE+PCI9111_REGISTER_AD_MODE_INTERRUPT_READBACK) >> 4)&0x03)

#define pci9111_interrupt_and_fifo_set(flags) \
outb(flags, PCI9111_IO_BASE+PCI9111_REGISTER_INTERRUPT_CONTROL)
@@ -208,35 +208,35 @@ TODO:

#define pci9111_is_fifo_full() \
((inb(PCI9111_IO_BASE+PCI9111_REGISTER_RANGE_STATUS_READBACK)& \
- PCI9111_FIFO_FULL_MASK)==0)
+ PCI9111_FIFO_FULL_MASK) == 0)

#define pci9111_is_fifo_half_full() \
((inb(PCI9111_IO_BASE+PCI9111_REGISTER_RANGE_STATUS_READBACK)& \
- PCI9111_FIFO_HALF_FULL_MASK)==0)
+ PCI9111_FIFO_HALF_FULL_MASK) == 0)

#define pci9111_is_fifo_empty() \
((inb(PCI9111_IO_BASE+PCI9111_REGISTER_RANGE_STATUS_READBACK)& \
- PCI9111_FIFO_EMPTY_MASK)==0)
+ PCI9111_FIFO_EMPTY_MASK) == 0)

#define pci9111_ai_channel_set(channel) \
outb((channel)&PCI9111_CHANNEL_MASK, PCI9111_IO_BASE+PCI9111_REGISTER_AD_CHANNEL_CONTROL)

#define pci9111_ai_channel_get() \
- inb(PCI9111_IO_BASE+PCI9111_REGISTER_AD_CHANNEL_READBACK)&PCI9111_CHANNEL_MASK
+ (inb(PCI9111_IO_BASE+PCI9111_REGISTER_AD_CHANNEL_READBACK)&(PCI9111_CHANNEL_MASK))

#define pci9111_ai_range_set(range) \
outb((range)&PCI9111_RANGE_MASK, PCI9111_IO_BASE+PCI9111_REGISTER_INPUT_SIGNAL_RANGE)

#define pci9111_ai_range_get() \
- inb(PCI9111_IO_BASE+PCI9111_REGISTER_RANGE_STATUS_READBACK)&PCI9111_RANGE_MASK
+ (inb(PCI9111_IO_BASE+PCI9111_REGISTER_RANGE_STATUS_READBACK)&(PCI9111_RANGE_MASK))

#define pci9111_ai_get_data() \
- ((inw(PCI9111_IO_BASE+PCI9111_REGISTER_AD_FIFO_VALUE)>>4)&PCI9111_AI_RESOLUTION_MASK) \
- ^ PCI9111_AI_RESOLUTION_2_CMP_BIT
+ (((inw(PCI9111_IO_BASE+PCI9111_REGISTER_AD_FIFO_VALUE)>>4)&PCI9111_AI_RESOLUTION_MASK) \
+ ^ (PCI9111_AI_RESOLUTION_2_CMP_BIT))

#define pci9111_hr_ai_get_data() \
- (inw(PCI9111_IO_BASE+PCI9111_REGISTER_AD_FIFO_VALUE) & PCI9111_HR_AI_RESOLUTION_MASK) \
- ^ PCI9111_HR_AI_RESOLUTION_2_CMP_BIT
+ ((inw(PCI9111_IO_BASE+PCI9111_REGISTER_AD_FIFO_VALUE) & PCI9111_HR_AI_RESOLUTION_MASK) \
+ ^ (PCI9111_HR_AI_RESOLUTION_2_CMP_BIT))

#define pci9111_ao_set_data(data) \
outw(data&PCI9111_AO_RESOLUTION_MASK, PCI9111_IO_BASE+PCI9111_REGISTER_DA_OUTPUT)
@@ -551,7 +551,8 @@ static int pci9111_ai_cancel(struct comedi_device *dev,
#define pci9111_check_trigger_src(src, flags) \
tmp = src; \
src &= flags; \
- if (!src || tmp != src) error++
+ if (!src || tmp != src) \
+ error++

static int
pci9111_ai_do_cmd_test(struct comedi_device *dev,
--
1.7.0.4

2010-06-16 20:22:38

by Charles Clément

Subject: Re: [PATCH 1/2] Staging: comedi: fix style errors in adl_pci9111.c

Hi,

On Wed, Jun 16, 2010 at 07:30:09PM +0100, John Sheehan wrote:
> From: John Sheehan <[email protected]>

> @@ -190,7 +190,7 @@ TODO:
> outb(flags, PCI9111_IO_BASE+PCI9111_REGISTER_TRIGGER_MODE_CONTROL)
>
> #define pci9111_interrupt_and_fifo_get() \
> - ((inb(PCI9111_IO_BASE+PCI9111_REGISTER_AD_MODE_INTERRUPT_READBACK) >> 4) &0x03)
> + ((inb(PCI9111_IO_BASE+PCI9111_REGISTER_AD_MODE_INTERRUPT_READBACK) >> 4)&0x03)

Doesn't checkpatch.pl suggest to add a whitespace after the & in that case?
It would seem more readable.

--
Charles Cl?ment

2010-06-20 20:05:28

by Charles Clément

Subject: Re: [PATCH 1/2] Staging: comedi: fix style errors in adl_pci9111.c

On Thu, Jun 17, 2010 at 10:33:36AM +0100, John Sheehan wrote:
> Hi Charles,
>
> 2010/6/16 Charles Cl?ment <[email protected]>:
> > Hi,
> >
> > On Wed, Jun 16, 2010 at 07:30:09PM +0100, John Sheehan wrote:
> >> From: John Sheehan <[email protected]>
> >
> >> @@ -190,7 +190,7 @@ TODO:
> >> ? ?outb(flags, PCI9111_IO_BASE+PCI9111_REGISTER_TRIGGER_MODE_CONTROL)
> >>
> >> ?#define pci9111_interrupt_and_fifo_get() \
> >> - ?((inb(PCI9111_IO_BASE+PCI9111_REGISTER_AD_MODE_INTERRUPT_READBACK) >> 4) &0x03)
> >> + ?((inb(PCI9111_IO_BASE+PCI9111_REGISTER_AD_MODE_INTERRUPT_READBACK) >> 4)&0x03)
> >
> > Doesn't checkpatch.pl suggest to add a whitespace after the & in that case?
>
> no, surprisingly enough, it just requires the amount of whitespace at
> either side of the & to be consistent

Any specific reason to take this off the list?

>
> John
> > It would seem more readable.
> >
> > --
> > Charles Cl?ment
> >
> >

--
Charles Cl?ment