[Bug debug/41097] Inlined variable debug location disappears when ptr type

2009-09-22 Thread aoliva at gcc dot gnu dot org


--- Comment #5 from aoliva at gcc dot gnu dot org  2009-09-22 20:56 ---
... which means the bug is fixed.


-- 

aoliva at gcc dot gnu dot org changed:

   What|Removed |Added

 Status|ASSIGNED|RESOLVED
 Resolution||FIXED


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=41097



[Bug debug/41097] Inlined variable debug location disappears when ptr type

2009-09-22 Thread aoliva at gcc dot gnu dot org


--- Comment #4 from aoliva at gcc dot gnu dot org  2009-09-22 20:53 ---
Jan Hubicka fixed this in revision 151917.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=41097



[Bug debug/41097] Inlined variable debug location disappears when ptr type

2009-09-08 Thread aoliva at gcc dot gnu dot org


--- Comment #3 from aoliva at gcc dot gnu dot org  2009-09-08 17:06 ---
The non-volatile variable goes missing in the inlined copy of m because it is
coalesced with the incoming argument.  With VTA, a note is retained that
associates p_name with the string.  However, we still fail to emit debug
information for p_name, for some so-far unknown reason.


-- 

aoliva at gcc dot gnu dot org changed:

   What|Removed |Added

 AssignedTo|unassigned at gcc dot gnu   |aoliva at gcc dot gnu dot
   |dot org |org
 Status|UNCONFIRMED |ASSIGNED
 Ever Confirmed|0   |1
   Last reconfirmed|-00-00 00:00:00 |2009-09-08 17:06:53
   date||


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=41097



[Bug debug/41097] Inlined variable debug location disappears when ptr type

2009-08-18 Thread mark at gcc dot gnu dot org


--- Comment #2 from mark at gcc dot gnu dot org  2009-08-18 08:11 ---
(In reply to comment #1)
> Wait a second are you saying:
> volatile char * p_name 
> 
> and p_name disappears?
> Well that is because p_name is no longer volatile but what it points to is.

Well, obvious p_name still is there, since it will be put in a register to feed
into the asm statement. But the debuginfo location for it is missing.

> I don't think this is a bug in GCC but rather a misunderstanding on how
> qualifiers apply to types.
>
> If you use a typedef for char*, does it disappear?

You are right, I misunderstood how the volatile qualifier "attached" to the
type. Using:

  typedef char * char_ptr;
  volatile char_ptr p_name = (char_ptr) name;

Does what I want. Thanks. That will allow me to write a workaround for what I
am seeing using typedef typeof(arg).

I am not sure this means the bug is really not there, because I am still a
little surprised (confused?) the location expression goes missing in the
original (and I assume the variable isn't actually fully optimized away). But
since I now have a workaround for that I am already happy. Thanks.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=41097



[Bug debug/41097] Inlined variable debug location disappears when ptr type

2009-08-18 Thread pinskia at gcc dot gnu dot org


--- Comment #1 from pinskia at gcc dot gnu dot org  2009-08-18 07:22 ---
Wait a second are you saying:
volatile char * p_name 

and p_name disappears?
Well that is because p_name is no longer volatile but what it points to is.

I don't think this is a bug in GCC but rather a misunderstanding on how
qualifiers apply to types.

If you use a typedef for char*, does it disappear?


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=41097