* Some musb Blackfin porting issues
@ 2007-09-06 6:52 Bryan Wu
2007-09-06 14:43 ` Tony Lindgren
0 siblings, 1 reply; 3+ messages in thread
From: Bryan Wu @ 2007-09-06 6:52 UTC (permalink / raw)
To: David Brownell, Tony Lindgren, linux-omap-open-source
Dear David and Tony:
I almost imported the musb stack to our Blackfin tree. But there are
some code highly coupled with ARM Linux architecture. Please give me
some idea about following issues:
1. Clock framework is not available in Blackfin Linux arch currently. Is
it possible to skip it?
2. Registers layout in musb-regs.h is different with BF54x processor. I
plan to write a musb-bf54x.h to replace the whole definition in
musb-regs.h. Is that possible or do it conflict with current musb
design?
in musb-regs.h:
---
#ifdef CONFIG_BF54x
#include "musb-bf54x.h"
#else
..... /* original code from musb-regs.h */
#endif
--
Thanks a lot
- Bryan Wu
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: Some musb Blackfin porting issues
2007-09-06 6:52 Some musb Blackfin porting issues Bryan Wu
@ 2007-09-06 14:43 ` Tony Lindgren
2007-09-06 18:44 ` David Brownell
0 siblings, 1 reply; 3+ messages in thread
From: Tony Lindgren @ 2007-09-06 14:43 UTC (permalink / raw)
To: Bryan Wu; +Cc: David Brownell, linux-omap-open-source
* Bryan Wu <bryan.wu@analog.com> [070905 23:55]:
> Dear David and Tony:
>
> I almost imported the musb stack to our Blackfin tree. But there are
> some code highly coupled with ARM Linux architecture. Please give me
> some idea about following issues:
>
> 1. Clock framework is not available in Blackfin Linux arch currently. Is
> it possible to skip it?
Yes, you can have custom clock enable/disable functions, see tusb_set_clock()
in arch/arm/mach-omap2/board-n800-usb.c.
> 2. Registers layout in musb-regs.h is different with BF54x processor. I
> plan to write a musb-bf54x.h to replace the whole definition in
> musb-regs.h. Is that possible or do it conflict with current musb
> design?
> in musb-regs.h:
> ---
> #ifdef CONFIG_BF54x
> #include "musb-bf54x.h"
> #else
> ..... /* original code from musb-regs.h */
> #endif
> --
I guess that's the way to go for now to get things working. Dave?
Tony
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: Some musb Blackfin porting issues
2007-09-06 14:43 ` Tony Lindgren
@ 2007-09-06 18:44 ` David Brownell
0 siblings, 0 replies; 3+ messages in thread
From: David Brownell @ 2007-09-06 18:44 UTC (permalink / raw)
To: tony, bryan.wu; +Cc: linux-omap-open-source
> > I almost imported the musb stack to our Blackfin tree. But there are
> > some code highly coupled with ARM Linux architecture. Please give me
> > some idea about following issues:
> >
> > 1. Clock framework is not available in Blackfin Linux arch currently. Is
> > it possible to skip it?
>
> Yes, you can have custom clock enable/disable functions, see tusb_set_clock()
> in arch/arm/mach-omap2/board-n800-usb.c.
One minor issue would then be needing stub implementations of the clk_*()
functions so the driver can link. The same issue comes up on x86 builds,
which would be good to allow as part of routine build tests.
Maybe you can just provide a set of inlined NOP clock stubs in the musb code
that would be included as needed. I'd suggest you avoid trying to solve
that general cross-arch problem, at least initially.
> > 2. Registers layout in musb-regs.h is different with BF54x processor. I
> > plan to write a musb-bf54x.h to replace the whole definition in
> > musb-regs.h. Is that possible or do it conflict with current musb
> > design?
> > in musb-regs.h:
> > ---
> > #ifdef CONFIG_BF54x
> > #include "musb-bf54x.h"
> > #else
> > ..... /* original code from musb-regs.h */
> > #endif
> > --
>
> I guess that's the way to go for now to get things working. Dave?
Something like that, sure. But I'd like it better if it were more
like two different #include files, one for Blackfin and one for the
standard register layouts (matching Mentor specs).
Are those registers the same widths? If the differences go beyond
just addressing, I'd forsee trouble ahead.
- Dave
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2007-09-06 18:44 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2007-09-06 6:52 Some musb Blackfin porting issues Bryan Wu
2007-09-06 14:43 ` Tony Lindgren
2007-09-06 18:44 ` David Brownell
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox