* [PATCH] d10v_make_iaddr: make it idempotent.
@ 2002-04-12 15:51 Michael Snyder
2002-04-12 16:27 ` Andrew Cagney
0 siblings, 1 reply; 3+ messages in thread
From: Michael Snyder @ 2002-04-12 15:51 UTC (permalink / raw)
To: gdb-patches; +Cc: jimb
2002-04-12 Michael Snyder <msnyder@redhat.com>
* d10v-tdep.c (d10v_make_iaddr): Make it idempotent,
in case it gets applied to an address that is already
in the instruction space.
Index: d10v-tdep.c
===================================================================
RCS file: /cvs/cvsfiles/devo/gdb/d10v-tdep.c,v
retrieving revision 2.88
diff -c -3 -p -r2.88 d10v-tdep.c
*** d10v-tdep.c 2002/04/09 03:00:45 2.88
--- d10v-tdep.c 2002/04/12 22:47:30
*************** d10v_register_virtual_type (int reg_nr)
*** 356,373 ****
return builtin_type_int16;
}
- static CORE_ADDR
- d10v_make_daddr (CORE_ADDR x)
- {
- return ((x) | DMEM_START);
- }
-
- static CORE_ADDR
- d10v_make_iaddr (CORE_ADDR x)
- {
- return (((x) << 2) | IMEM_START);
- }
-
static int
d10v_daddr_p (CORE_ADDR x)
{
--- 356,361 ----
*************** d10v_iaddr_p (CORE_ADDR x)
*** 380,385 ****
--- 368,387 ----
return (((x) & 0x3000000) == IMEM_START);
}
+ static CORE_ADDR
+ d10v_make_daddr (CORE_ADDR x)
+ {
+ return ((x) | DMEM_START);
+ }
+
+ static CORE_ADDR
+ d10v_make_iaddr (CORE_ADDR x)
+ {
+ if (d10v_iaddr_p (x))
+ return x; /* Idempotency -- x is already in the IMEM space. */
+ else
+ return (((x) << 2) | IMEM_START);
+ }
static CORE_ADDR
d10v_convert_iaddr_to_raw (CORE_ADDR x)
^ permalink raw reply [flat|nested] 3+ messages in thread* Re: [PATCH] d10v_make_iaddr: make it idempotent.
2002-04-12 15:51 [PATCH] d10v_make_iaddr: make it idempotent Michael Snyder
@ 2002-04-12 16:27 ` Andrew Cagney
2002-04-12 17:34 ` Michael Snyder
0 siblings, 1 reply; 3+ messages in thread
From: Andrew Cagney @ 2002-04-12 16:27 UTC (permalink / raw)
To: Michael Snyder; +Cc: gdb-patches, jimb
> 2002-04-12 Michael Snyder <msnyder@redhat.com>
>
> * d10v-tdep.c (d10v_make_iaddr): Make it idempotent,
> in case it gets applied to an address that is already
> in the instruction space.
Michael, I don't understand this change. Why does it need to be
idempotent - isn't that a bug somewhere else? And why move the code?
Andrew
> Index: d10v-tdep.c
> ===================================================================
> RCS file: /cvs/cvsfiles/devo/gdb/d10v-tdep.c,v
> retrieving revision 2.88
> diff -c -3 -p -r2.88 d10v-tdep.c
> *** d10v-tdep.c 2002/04/09 03:00:45 2.88
> --- d10v-tdep.c 2002/04/12 22:47:30
> *************** d10v_register_virtual_type (int reg_nr)
> *** 356,373 ****
> return builtin_type_int16;
> }
>
> - static CORE_ADDR
> - d10v_make_daddr (CORE_ADDR x)
> - {
> - return ((x) | DMEM_START);
> - }
> -
> - static CORE_ADDR
> - d10v_make_iaddr (CORE_ADDR x)
> - {
> - return (((x) << 2) | IMEM_START);
> - }
> -
> static int
> d10v_daddr_p (CORE_ADDR x)
> {
> --- 356,361 ----
> *************** d10v_iaddr_p (CORE_ADDR x)
> *** 380,385 ****
> --- 368,387 ----
> return (((x) & 0x3000000) == IMEM_START);
> }
>
> + static CORE_ADDR
> + d10v_make_daddr (CORE_ADDR x)
> + {
> + return ((x) | DMEM_START);
> + }
> +
> + static CORE_ADDR
> + d10v_make_iaddr (CORE_ADDR x)
> + {
> + if (d10v_iaddr_p (x))
> + return x; /* Idempotency -- x is already in the IMEM space. */
> + else
> + return (((x) << 2) | IMEM_START);
> + }
>
> static CORE_ADDR
> d10v_convert_iaddr_to_raw (CORE_ADDR x)
>
^ permalink raw reply [flat|nested] 3+ messages in thread* Re: [PATCH] d10v_make_iaddr: make it idempotent.
2002-04-12 16:27 ` Andrew Cagney
@ 2002-04-12 17:34 ` Michael Snyder
0 siblings, 0 replies; 3+ messages in thread
From: Michael Snyder @ 2002-04-12 17:34 UTC (permalink / raw)
To: Andrew Cagney; +Cc: Michael Snyder, gdb-patches, jimb
Andrew Cagney wrote:
>
> > 2002-04-12 Michael Snyder <msnyder@redhat.com>
> >
> > * d10v-tdep.c (d10v_make_iaddr): Make it idempotent,
> > in case it gets applied to an address that is already
> > in the instruction space.
>
> Michael, I don't understand this change. Why does it need to be
> idempotent - isn't that a bug somewhere else?
If you say "x/i 0x20001234", you don't want it to apply
the transform to that address.
> And why move the code?
Just because I called dq0v_iaddr_p. I have a preference
for ordering functions rather than adding prototypes.
I guess that's not the best thing to do...
>
> Andrew
>
> > Index: d10v-tdep.c
> > ===================================================================
> > RCS file: /cvs/cvsfiles/devo/gdb/d10v-tdep.c,v
> > retrieving revision 2.88
> > diff -c -3 -p -r2.88 d10v-tdep.c
> > *** d10v-tdep.c 2002/04/09 03:00:45 2.88
> > --- d10v-tdep.c 2002/04/12 22:47:30
> > *************** d10v_register_virtual_type (int reg_nr)
> > *** 356,373 ****
> > return builtin_type_int16;
> > }
> >
> > - static CORE_ADDR
> > - d10v_make_daddr (CORE_ADDR x)
> > - {
> > - return ((x) | DMEM_START);
> > - }
> > -
> > - static CORE_ADDR
> > - d10v_make_iaddr (CORE_ADDR x)
> > - {
> > - return (((x) << 2) | IMEM_START);
> > - }
> > -
> > static int
> > d10v_daddr_p (CORE_ADDR x)
> > {
> > --- 356,361 ----
> > *************** d10v_iaddr_p (CORE_ADDR x)
> > *** 380,385 ****
> > --- 368,387 ----
> > return (((x) & 0x3000000) == IMEM_START);
> > }
> >
> > + static CORE_ADDR
> > + d10v_make_daddr (CORE_ADDR x)
> > + {
> > + return ((x) | DMEM_START);
> > + }
> > +
> > + static CORE_ADDR
> > + d10v_make_iaddr (CORE_ADDR x)
> > + {
> > + if (d10v_iaddr_p (x))
> > + return x; /* Idempotency -- x is already in the IMEM space. */
> > + else
> > + return (((x) << 2) | IMEM_START);
> > + }
> >
> > static CORE_ADDR
> > d10v_convert_iaddr_to_raw (CORE_ADDR x)
> >
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2002-04-13 0:34 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2002-04-12 15:51 [PATCH] d10v_make_iaddr: make it idempotent Michael Snyder
2002-04-12 16:27 ` Andrew Cagney
2002-04-12 17:34 ` Michael Snyder
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox