Re: Errors in pairs

[Nathan Sidwell]

On 03/24/2018 04:06 PM, Volker Reichelt wrote:

Hi everybody,

while bug-hunting I noticed that we emit lots of erros in pairs in
check_final_overrider (cp/search.c), e.g.:

   error ("invalid covariant return type for %q+#D", overrider);
   error ("  overriding %q+#D", basefn);

I would expect the second line to be emitted as a note (using inform).
Is there a reason for this or should that be changed?


Probably a sign the code predates 'inform' and/or the error-limit 
option.  As Paolo says, this should be changed.


nathan

--
Nathan Sidwell

Re: Errors in pairs

[Volker Reichelt]

On 03/24/2018 10:35 PM, Paolo Carlini wrote:

Hi,

On 24/03/2018 21:06, Volker Reichelt wrote:

Hi everybody,

while bug-hunting I noticed that we emit lots of erros in pairs in
check_final_overrider (cp/search.c), e.g.:

  error ("invalid covariant return type for %q+#D", overrider);
  error ("  overriding %q+#D", basefn);

I would expect the second line to be emitted as a note (using inform).
Is there a reason for this or should that be changed?
IMHO that should be changed. As you may or may not have noticed, 
personally I took care of many of such issues, but obviously not of 
this one (I'm not surprised, I remember thinking a few times "I should 
come back to this one too..."). I encourage you to send patches!


Paolo.



OK, here we go:

https://gcc.gnu.org/ml/gcc-patches/2018-03/msg01364.html

Regards,
Volker

Re: Errors in pairs

[Paolo Carlini]

Hi,

On 24/03/2018 21:06, Volker Reichelt wrote:

Hi everybody,

while bug-hunting I noticed that we emit lots of erros in pairs in
check_final_overrider (cp/search.c), e.g.:

  error ("invalid covariant return type for %q+#D", overrider);
  error ("  overriding %q+#D", basefn);

I would expect the second line to be emitted as a note (using inform).
Is there a reason for this or should that be changed?
IMHO that should be changed. As you may or may not have noticed, 
personally I took care of many of such issues, but obviously not of this 
one (I'm not surprised, I remember thinking a few times "I should come 
back to this one too..."). I encourage you to send patches!


Paolo.

Errors in pairs

[Volker Reichelt]

Hi everybody,

while bug-hunting I noticed that we emit lots of erros in pairs in
check_final_overrider (cp/search.c), e.g.:

  error ("invalid covariant return type for %q+#D", overrider);
  error ("  overriding %q+#D", basefn);

I would expect the second line to be emitted as a note (using inform).
Is there a reason for this or should that be changed?

Regards,
Volker