This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [3.4-BIB] fix profiling


>>>>> "Nathan" == Nathan Sidwell <nathan@codesourcery.com> writes:

 Nathan> Paul Koning wrote:
 >>>>>>> "Nathan" == Nathan Sidwell <nathan@codesourcery.com> writes:

 Nathan> Honza, why is this change necessary? unsigned will be at
 Nathan> least 32bits (gcc doesn't target 16bit int abis).
 >> Huh?  Pdp11 is (unless overridden by -mint32) a target with 16-bit
 >> ints.
 Nathan> oo, didn't know that. The documentation is incorrect.

 Nathan> doc/portability.texi says 'The main goal of GCC was to make a
 Nathan> good, fast compiler for machines in the class that the GNU
 Nathan> system aims to run on: 32-bit machines that address 8-bit
 Nathan> bytes and have several general registers.'

As a statement of general intent I would not have any quarrel with
that.

 Nathan> er, ok that doesn't say the target has 32bit ints. My
 Nathan> mistake. The gnu coding conventions also say (from autoconf)
 Nathan> 'However, don't make any effort to cater to the possibility
 Nathan> that an @code{int} will be less than 32 bits.  We don't
 Nathan> support 16-bit machines in GNU.'  ... so I didn't :)

Sounds like a general GNU comment.  GCC sometimes goes places where
the larger GNU stuff fears to tread... :-)

 Nathan> I know how to dink the gcov declarations to make them work on
 Nathan> 16-bit int ABIs. Should we?

Perhaps not.  If pdp11 is the one exception, a reasonable answer may
be "if/when pdp11 wants gcov support, it will have to address this".

   paul


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]