* [IOC3] IP27: Really set PCI64_ATTR_VIRTUAL, not PCI64_ATTR_PREC.
@ 2006-06-17 17:57 Ralf Baechle
2006-06-20 8:15 ` Ingo Oeser
0 siblings, 1 reply; 3+ messages in thread
From: Ralf Baechle @ 2006-06-17 17:57 UTC (permalink / raw)
To: netdev, Jeff Garzik
IOC3's homegrown DMA mapping functions that are used to optimize things
a little on IP27 set the wrong bit.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
diff --git a/drivers/net/ioc3-eth.c b/drivers/net/ioc3-eth.c
index ae71ed5..e76e6e7 100644
--- a/drivers/net/ioc3-eth.c
+++ b/drivers/net/ioc3-eth.c
@@ -145,7 +145,7 @@ static inline struct sk_buff * ioc3_allo
static inline unsigned long ioc3_map(void *ptr, unsigned long vdev)
{
#ifdef CONFIG_SGI_IP27
- vdev <<= 58; /* Shift to PCI64_ATTR_VIRTUAL */
+ vdev <<= 57; /* Shift to PCI64_ATTR_VIRTUAL */
return vdev | (0xaUL << PCI64_ATTR_TARG_SHFT) | PCI64_ATTR_PREF |
((unsigned long)ptr & TO_PHYS_MASK);
^ permalink raw reply related [flat|nested] 3+ messages in thread* Re: [IOC3] IP27: Really set PCI64_ATTR_VIRTUAL, not PCI64_ATTR_PREC.
2006-06-17 17:57 [IOC3] IP27: Really set PCI64_ATTR_VIRTUAL, not PCI64_ATTR_PREC Ralf Baechle
@ 2006-06-20 8:15 ` Ingo Oeser
2006-06-20 10:02 ` Ralf Baechle
0 siblings, 1 reply; 3+ messages in thread
From: Ingo Oeser @ 2006-06-20 8:15 UTC (permalink / raw)
To: Ralf Baechle; +Cc: netdev, Jeff Garzik
Hi Ralf,
Ralf Baechle :
> IOC3's homegrown DMA mapping functions that are used to optimize things
> a little on IP27 set the wrong bit.
What about using a symbol instead of magic numbers?
That way one at least sees the intention of the coder.
> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
>
> diff --git a/drivers/net/ioc3-eth.c b/drivers/net/ioc3-eth.c
> index ae71ed5..e76e6e7 100644
> --- a/drivers/net/ioc3-eth.c
> +++ b/drivers/net/ioc3-eth.c
> @@ -145,7 +145,7 @@ static inline struct sk_buff * ioc3_allo
> static inline unsigned long ioc3_map(void *ptr, unsigned long vdev)
> {
> #ifdef CONFIG_SGI_IP27
> - vdev <<= 58; /* Shift to PCI64_ATTR_VIRTUAL */
> + vdev <<= 57; /* Shift to PCI64_ATTR_VIRTUAL */
So please use a symbolic value here.
Regards
Ingo Oeser
^ permalink raw reply [flat|nested] 3+ messages in thread* Re: [IOC3] IP27: Really set PCI64_ATTR_VIRTUAL, not PCI64_ATTR_PREC.
2006-06-20 8:15 ` Ingo Oeser
@ 2006-06-20 10:02 ` Ralf Baechle
0 siblings, 0 replies; 3+ messages in thread
From: Ralf Baechle @ 2006-06-20 10:02 UTC (permalink / raw)
To: Ingo Oeser; +Cc: netdev, Jeff Garzik
On Tue, Jun 20, 2006 at 10:15:01AM +0200, Ingo Oeser wrote:
> From: Ingo Oeser <netdev@axxeo.de>
> To: Ralf Baechle <ralf@linux-mips.org>
> Subject: Re: [IOC3] IP27: Really set PCI64_ATTR_VIRTUAL, not PCI64_ATTR_PREC.
> Date: Tue, 20 Jun 2006 10:15:01 +0200
> Cc: netdev@vger.kernel.org, Jeff Garzik <jeff@garzik.org>
> Content-Type: text/plain;
> charset="iso-8859-1"
>
> Hi Ralf,
>
> Ralf Baechle :
> > IOC3's homegrown DMA mapping functions that are used to optimize things
> > a little on IP27 set the wrong bit.
>
> What about using a symbol instead of magic numbers?
> That way one at least sees the intention of the coder.
>
> > Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
> >
> > diff --git a/drivers/net/ioc3-eth.c b/drivers/net/ioc3-eth.c
> > index ae71ed5..e76e6e7 100644
> > --- a/drivers/net/ioc3-eth.c
> > +++ b/drivers/net/ioc3-eth.c
> > @@ -145,7 +145,7 @@ static inline struct sk_buff * ioc3_allo
> > static inline unsigned long ioc3_map(void *ptr, unsigned long vdev)
> > {
> > #ifdef CONFIG_SGI_IP27
> > - vdev <<= 58; /* Shift to PCI64_ATTR_VIRTUAL */
> > + vdev <<= 57; /* Shift to PCI64_ATTR_VIRTUAL */
>
> So please use a symbolic value here.
It is a hack and meant to look like one, loudly marked with #ifdef.
Ralf
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2006-06-20 10:04 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2006-06-17 17:57 [IOC3] IP27: Really set PCI64_ATTR_VIRTUAL, not PCI64_ATTR_PREC Ralf Baechle
2006-06-20 8:15 ` Ingo Oeser
2006-06-20 10:02 ` Ralf Baechle
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).