why is sim_addr_range_hit_p tagged inline?

why is sim_addr_range_hit_p tagged inline?

[DJ Delorie]

In sim/sim-arange.h:

  /* Return non-zero if ADDR is in range AR, traversing the entire tree.
     If no range is specified, that is defined to mean "everything".  */
  extern INLINE int
  sim_addr_range_hit_p (ADDR_RANGE * /*ar*/, address_word /*addr*/);
  #define ADDR_RANGE_HIT_P(ar, addr) \
    ((ar)->range_tree == NULL || sim_addr_range_hit_p ((ar), (addr)))

Why is the INLINE there?  sim_addr_range_hit_p() is only defined in
sim-arange.c

Re: why is sim_addr_range_hit_p tagged inline?

[Andreas Schwab]

DJ Delorie <dj@redhat.com> writes:

> In sim/sim-arange.h:
>
>   /* Return non-zero if ADDR is in range AR, traversing the entire tree.
>      If no range is specified, that is defined to mean "everything".  */
>   extern INLINE int
>   sim_addr_range_hit_p (ADDR_RANGE * /*ar*/, address_word /*addr*/);
>   #define ADDR_RANGE_HIT_P(ar, addr) \
>     ((ar)->range_tree == NULL || sim_addr_range_hit_p ((ar), (addr)))
>
> Why is the INLINE there?  sim_addr_range_hit_p() is only defined in
> sim-arange.c

Unless HAVE_INLINE is defined.

Andreas.

-- 
Andreas Schwab, schwab@linux-m68k.org
GPG Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: why is sim_addr_range_hit_p tagged inline?

[DJ Delorie]

> Unless HAVE_INLINE is defined.

Ah, ick.  I noticed this because I had a report from a Mac user that
they were getting "multiply defined symbols" for that function.

Re: why is sim_addr_range_hit_p tagged inline?

[Mike Frysinger]

[-- Attachment #1: Type: text/plain, Size: 438 bytes --]

On 14 Nov 2014 15:16, DJ Delorie wrote:
> > Unless HAVE_INLINE is defined.
> 
> Ah, ick.  I noticed this because I had a report from a Mac user that
> they were getting "multiply defined symbols" for that function.

the sim code base heavily inlines things for the sake of speed.  but i'm not 
sure what kind of perf difference it makes nowadays on modern systems.  would be 
worth seeing if we could just drop it all ...
-mike

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]