* RE: pcmcia
@ 2001-11-30 8:39 Guillermo A. Loyola
2001-11-30 17:35 ` pcmcia Pete Popov
0 siblings, 1 reply; 6+ messages in thread
From: Guillermo A. Loyola @ 2001-11-30 8:39 UTC (permalink / raw)
To: 'Pete Popov', linux-mips, sforge
> The pcmcia variable ioaddr_t should be a 32 bit type for my socket
> driver. Is there any harm to other mips pcmcia socket drivers if we
> apply the patch below?
We need the same here, how about doing this instead:
#ifdef __i386__
typedef u_short ioaddr_t;
#else
typedef u_int ioaddr_t;
#endif
Gmo.
^ permalink raw reply [flat|nested] 6+ messages in thread
* RE: pcmcia
2001-11-30 8:39 pcmcia Guillermo A. Loyola
@ 2001-11-30 17:35 ` Pete Popov
2001-11-30 17:54 ` pcmcia Alan Cox
0 siblings, 1 reply; 6+ messages in thread
From: Pete Popov @ 2001-11-30 17:35 UTC (permalink / raw)
To: Guillermo A. Loyola; +Cc: linux-mips, sforge
On Fri, 2001-11-30 at 00:39, Guillermo A. Loyola wrote:
> > The pcmcia variable ioaddr_t should be a 32 bit type for my socket
> > driver. Is there any harm to other mips pcmcia socket drivers if we
> > apply the patch below?
>
> We need the same here, how about doing this instead:
>
> #ifdef __i386__
> typedef u_short ioaddr_t;
> #else
> typedef u_int ioaddr_t;
> #endif
That probably makes more sense. I wasn't sure if it's only x86 that
needs? ioaddr_t to be a 16 bit type.
Pete
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: pcmcia
2001-11-30 17:35 ` pcmcia Pete Popov
@ 2001-11-30 17:54 ` Alan Cox
2001-11-30 17:50 ` pcmcia Pete Popov
2001-11-30 17:54 ` pcmcia Alan Cox
0 siblings, 2 replies; 6+ messages in thread
From: Alan Cox @ 2001-11-30 17:54 UTC (permalink / raw)
To: Pete Popov; +Cc: Guillermo A. Loyola, linux-mips, sforge
> > We need the same here, how about doing this instead:
> >
> > #ifdef __i386__
> > typedef u_short ioaddr_t;
> > #else
> > typedef u_int ioaddr_t;
> > #endif
>
> That probably makes more sense. I wasn't sure if it's only x86 that
> needs? ioaddr_t to be a 16 bit type.
Is there any platform where making it int actually -breaks-. At least for
2.5 it would seem a lot saner to just make it bigger and see
Alan
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: pcmcia
2001-11-30 17:54 ` pcmcia Alan Cox
@ 2001-11-30 17:50 ` Pete Popov
2001-11-30 17:54 ` pcmcia Alan Cox
1 sibling, 0 replies; 6+ messages in thread
From: Pete Popov @ 2001-11-30 17:50 UTC (permalink / raw)
To: Alan Cox; +Cc: Guillermo A. Loyola, linux-mips, sforge
On Fri, 2001-11-30 at 09:54, Alan Cox wrote:
> > > We need the same here, how about doing this instead:
> > >
> > > #ifdef __i386__
> > > typedef u_short ioaddr_t;
> > > #else
> > > typedef u_int ioaddr_t;
> > > #endif
> >
> > That probably makes more sense. I wasn't sure if it's only x86 that
> > needs? ioaddr_t to be a 16 bit type.
>
> Is there any platform where making it int actually -breaks-.
I can't see how it would break anything ... but I've said that before.
It's not a variable which maps a hardware register, a protocol field,
etc, so it should be safe to just make it an int.
> At least for 2.5 it would seem a lot saner to just make it bigger and see
Pete
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: pcmcia
2001-11-30 17:54 ` pcmcia Alan Cox
2001-11-30 17:50 ` pcmcia Pete Popov
@ 2001-11-30 17:54 ` Alan Cox
1 sibling, 0 replies; 6+ messages in thread
From: Alan Cox @ 2001-11-30 17:54 UTC (permalink / raw)
To: Pete Popov; +Cc: Guillermo A. Loyola, linux-mips, sforge
> > We need the same here, how about doing this instead:
> >
> > #ifdef __i386__
> > typedef u_short ioaddr_t;
> > #else
> > typedef u_int ioaddr_t;
> > #endif
>
> That probably makes more sense. I wasn't sure if it's only x86 that
> needs? ioaddr_t to be a 16 bit type.
Is there any platform where making it int actually -breaks-. At least for
2.5 it would seem a lot saner to just make it bigger and see
Alan
^ permalink raw reply [flat|nested] 6+ messages in thread
* pcmcia
@ 2001-11-29 20:32 Pete Popov
0 siblings, 0 replies; 6+ messages in thread
From: Pete Popov @ 2001-11-29 20:32 UTC (permalink / raw)
To: linux-mips, sforge
The pcmcia variable ioaddr_t should be a 32 bit type for my socket
driver. Is there any harm to other mips pcmcia socket drivers if we
apply the patch below? If not, it would make it so much easier if I
don't have to debug this problem with each new kernel (having forgotten
about the need for this patch)...
--- linux-orig/include/pcmcia/cs_types.h Mon Nov 5 16:55:31 2001
+++ linux/include/pcmcia/cs_types.h Thu Nov 29 12:27:42 2001
@@ -36,7 +36,7 @@
#include <sys/types.h>
#endif
-#ifdef __arm__
+#if defined(__arm__) || defined(__mips__)
typedef u_int ioaddr_t;
#else
typedef u_short ioaddr_t;
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2001-11-30 18:51 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2001-11-30 8:39 pcmcia Guillermo A. Loyola
2001-11-30 17:35 ` pcmcia Pete Popov
2001-11-30 17:54 ` pcmcia Alan Cox
2001-11-30 17:50 ` pcmcia Pete Popov
2001-11-30 17:54 ` pcmcia Alan Cox
-- strict thread matches above, loose matches on Subject: below --
2001-11-29 20:32 pcmcia Pete Popov
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox