| Message ID | 20101105211620.d36ff536.basile@starynkevitch.net |
|---|---|
| State | New |
| Headers | show |
On Fri, 5 Nov 2010 21:16:20 +0100 Basile Starynkevitch <basile@starynkevitch.net> wrote: > +#define POS_HERE(Call) do { pos.file = this_file; pos.line = __LINE__; > \ Something get wrong with my emailer. Please consider the backslash to be on the same line as the #define
On Fri, Nov 5, 2010 at 9:16 PM, Basile Starynkevitch <basile@starynkevitch.net> wrote: > The following patch is a cleaner macro to > force the pos to the actual position. (I actually created > the mistake by indenting the file too GNU-sely). > > > Index: gcc/gengtype.c > =================================================================== > --- gcc/gengtype.c (revision 166376) > +++ gcc/gengtype.c (working copy) > @@ -4531,28 +4536,21 @@ main (int argc, char **argv) > /* These types are set up with #define or else outside of where > we can see them. We should initialize them before calling > read_input_list. */ > - pos.file = this_file; > - pos.line = __LINE__ + 1; > - do_scalar_typedef ("CUMULATIVE_ARGS", &pos); > - pos.line++; > - do_scalar_typedef ("REAL_VALUE_TYPE", &pos); > - pos.line++; > - do_scalar_typedef ("FIXED_VALUE_TYPE", &pos); > - pos.line++; > - do_scalar_typedef ("double_int", &pos); > - pos.line++; > - do_scalar_typedef ("uint64_t", &pos); > - pos.line++; > - do_scalar_typedef ("uint8", &pos); > - pos.line++; > - do_scalar_typedef ("jword", &pos); > - pos.line++; > - do_scalar_typedef ("JCF_u2", &pos); > - pos.line++; > - do_scalar_typedef ("void", &pos); > - pos.line++; > - do_typedef ("PTR", create_pointer (resolve_typedef ("void", > &pos)), > - &pos); > +#define POS_HERE(Call) do { pos.file = this_file; pos.line = __LINE__; > \ > + Call;} while(0) > + POS_HERE (do_scalar_typedef ("CUMULATIVE_ARGS", &pos)); > + POS_HERE (do_scalar_typedef ("REAL_VALUE_TYPE", &pos)); > + POS_HERE (do_scalar_typedef ("FIXED_VALUE_TYPE", &pos)); > + POS_HERE (do_scalar_typedef ("double_int", &pos)); > + POS_HERE (do_scalar_typedef ("uint64_t", &pos)); > + POS_HERE (do_scalar_typedef ("uint8", &pos)); > + POS_HERE (do_scalar_typedef ("jword", &pos)); > + POS_HERE (do_scalar_typedef ("JCF_u2", &pos)); > + POS_HERE (do_scalar_typedef ("void", &pos)); > + POS_HERE (do_typedef ("PTR", > + create_pointer (resolve_typedef ("void", &pos)), > + &pos)); This doesn't look like a 1:1 transformation - who increments pos.line? > +#undef POS_HERE > read_input_list (inputlist); > for (i = 0; i < num_gt_files; i++) > { > ################ gcc/ChangeLog entry > 2010-11-05 Basile Starynkevitch <basile@starynkevitch.net> > * gengtype (main): Get here's position using POS_HERE macro for > do*typedef calls. > ############## > > Ok for trunk? > > PS. Since I am changing web provider next week, I might have transitory > email problems. If an email bounce to basile@starynkevitch.net please > resend it to both basile.starynkevitch@gmail.com & > basile.starynkevitch@cea.fr (but don't use these emails when > basile@starynkevitch.net works) > -- > Basile STARYNKEVITCH http://starynkevitch.net/Basile/ > email: basile<at>starynkevitch<dot>net mobile: +33 6 8501 2359 > 8, rue de la Faiencerie, 92340 Bourg La Reine, France > *** opinions {are only mine, sont seulement les miennes} *** >
On Sat, 6 Nov 2010 15:35:17 +0100 Richard Guenther <richard.guenther@gmail.com> wrote: >> +#define POS_HERE(Call) do { pos.file = this_file; pos.line = __LINE__; \ >> + Call;} while(0) >> + POS_HERE (do_scalar_typedef ("CUMULATIVE_ARGS", &pos)); >> + POS_HERE (do_scalar_typedef ("REAL_VALUE_TYPE", &pos)); > This doesn't look like a 1:1 transformation - who increments pos.line? pos.line is not supposed to be incremented, but to give the accurate line number. It was incremented in 4.5 to avoid using __LINE__ repeatedly. (that was a poor way of doing it, IMHO) Then, I indented gengtype and the line increment got on seperate line, which sucks. Cheers
2010/11/5 Basile Starynkevitch <basile@starynkevitch.net>: > 2010-11-05 Basile Starynkevitch <basile@starynkevitch.net> > * gengtype (main): Get here's position using POS_HERE macro for > do*typedef calls. Richard - This patch is a 1:1 transformation from a pre-indent-run gengtype. Also, it is not terribly critical to get the line numbers right here. Basile - This patch is OK, unless Richard objects. Thanks,
On Mon, 8 Nov 2010 04:21:11 +0200 Laurynas Biveinis <laurynas.biveinis@gmail.com> wrote: > 2010/11/5 Basile Starynkevitch <basile@starynkevitch.net>: > > 2010-11-05 Basile Starynkevitch <basile@starynkevitch.net> > > * gengtype (main): Get here's position using POS_HERE macro for > > do*typedef calls. > This patch is OK, unless Richard objects. Thanks, Committed revision 166428. PS: Laurynas, if mails to me bounce, please use basile@ours.starynkevitch.net or basile.starynkevitch@gmail.com or basile.starynkevitch@cea.fr
2010/11/8 Basile Starynkevitch <basile@starynkevitch.net>: >> This patch is OK, unless Richard objects. Thanks, > > Committed revision 166428. You didn't wait to see if Richard objects. Now let's hope he doesn't :)
Index: gcc/gengtype.c =================================================================== --- gcc/gengtype.c (revision 166376) +++ gcc/gengtype.c (working copy) @@ -4531,28 +4536,21 @@ main (int argc, char **argv) /* These types are set up with #define or else outside of where we can see them. We should initialize them before calling read_input_list. */ - pos.file = this_file; - pos.line = __LINE__ + 1; - do_scalar_typedef ("CUMULATIVE_ARGS", &pos); - pos.line++; - do_scalar_typedef ("REAL_VALUE_TYPE", &pos); - pos.line++; - do_scalar_typedef ("FIXED_VALUE_TYPE", &pos); - pos.line++; - do_scalar_typedef ("double_int", &pos); - pos.line++; - do_scalar_typedef ("uint64_t", &pos); - pos.line++; - do_scalar_typedef ("uint8", &pos); - pos.line++; - do_scalar_typedef ("jword", &pos); - pos.line++; - do_scalar_typedef ("JCF_u2", &pos); - pos.line++; - do_scalar_typedef ("void", &pos); - pos.line++; - do_typedef ("PTR", create_pointer (resolve_typedef ("void", &pos)), - &pos); +#define POS_HERE(Call) do { pos.file = this_file; pos.line = __LINE__; \ + Call;} while(0) + POS_HERE (do_scalar_typedef ("CUMULATIVE_ARGS", &pos)); + POS_HERE (do_scalar_typedef ("REAL_VALUE_TYPE", &pos)); + POS_HERE (do_scalar_typedef ("FIXED_VALUE_TYPE", &pos)); + POS_HERE (do_scalar_typedef ("double_int", &pos)); + POS_HERE (do_scalar_typedef ("uint64_t", &pos)); + POS_HERE (do_scalar_typedef ("uint8", &pos)); + POS_HERE (do_scalar_typedef ("jword", &pos)); + POS_HERE (do_scalar_typedef ("JCF_u2", &pos)); + POS_HERE (do_scalar_typedef ("void", &pos)); + POS_HERE (do_typedef ("PTR", + create_pointer (resolve_typedef ("void", &pos)), + &pos)); +#undef POS_HERE read_input_list (inputlist);
The following patch is a cleaner macro to force the pos to the actual position. (I actually created the mistake by indenting the file too GNU-sely). for (i = 0; i < num_gt_files; i++) { ################ gcc/ChangeLog entry 2010-11-05 Basile Starynkevitch <basile@starynkevitch.net> * gengtype (main): Get here's position using POS_HERE macro for do*typedef calls. ############## Ok for trunk? PS. Since I am changing web provider next week, I might have transitory email problems. If an email bounce to basile@starynkevitch.net please resend it to both basile.starynkevitch@gmail.com & basile.starynkevitch@cea.fr (but don't use these emails when basile@starynkevitch.net works)