[prev in list] [next in list] [prev in thread] [next in thread] 

List:       linux-kernel
Subject:    Re: [RFC] clustered apic irq affinity fix for i386
From:       Andrew Morton <akpm () digeo ! com>
Date:       2003-04-30 23:36:37
[Download RAW message or body]

Keith Mannthey <kmannth@us.ibm.com> wrote:
> 
> Hello all,
> 	Machines with clustered apics are buggy when it comes to setting irq
> affinity.

You stand accused of crimes against whitespace.

- The patch uses eight-spaces everywhere.  Please use eight-stop tabs.

- Funny comment format:

/*we want to be careful what we write we are in clustered mode
 *if the mask came from pending_irq_balance_apicid we are ok because
 *it was generated with cpu_to_logical_apicid*/

 should be

/*
 * We want to be careful what we write we are in clustered mode if the mask
 * came from pending_irq_balance_apicid we are ok because it was generated
 * with cpu_to_logical_apicid.
 */

- 1e16-column xterms.  Please aim for 80-columns:

static inline void io_apic_write_affinity(unsigned int apic, unsigned int reg,  \
unsigned int mask, unsigned int irq)

  should be:

static inline void io_apic_write_affinity(unsigned int apic,
		unsigned int reg,  unsigned int mask, unsigned int irq)

  or

static inline void
io_apic_write_affinity(unsigned int apic, unsigned int reg,
			unsigned int mask, unsigned int irq)


- non-K&R braces:


       if ((pending_irq_balance_apicid[irq] == mask) || (irqbalance_disabled))
       {
               mask = mask << 24;
               io_apic_write(apic,reg,mask);
       }
       else
       {
               printk ("Trying to write abartry affinity value to ioapic! Not \
allowed!");  }
}

  should be:

       if ((pending_irq_balance_apicid[irq] == mask) || irqbalance_disabled) {
               mask = mask << 24;
               io_apic_write(apic,reg,mask);
       } else {
               printk("Trying to write abartry affinity value to ioapic! Not \
allowed!");  }

  (note: no space between "printk" and "(")

  (s/abartry/arbitrary/)

  (best replace that printk with a BUG() or a WARN_ON(1))

  (or split the string up so it fits in 80-cols)

- Funny comments:

	/*only the first 8 bits are valid*/

  should be:

	/* Only the first 8 bits are valid */


Could you please take a look at all that and resend?

Thanks.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/


[prev in list] [next in list] [prev in thread] [next in thread]