Re: [C++ Patch] PR 60999

2014-05-06 Thread Jason Merrill
On 05/06/2014 04:23 AM, Paolo Carlini wrote: On further thought (yesterday was a little tired, sorry), I think that nsdmi-template7.C is telling us that, in fact, we *do* want something true when we are handling an NSDMI inside a non-template class of a template class! Right, sorry, I was think

Re: [C++ Patch] PR 60999

2014-05-06 Thread Paolo Carlini
Hi again, On 05/05/2014 10:48 PM, Jason Merrill wrote: On 05/05/2014 12:25 PM, Paolo Carlini wrote: Good, but is it Ok to use uses_template_parms for that? A few days ago I struggled to find something simpler ready to use, to no avail (well, this is probably well known to you, but there are sur

Re: [C++ Patch] PR 60999

2014-05-05 Thread Paolo Carlini
Hi, On 05/05/2014 10:48 PM, Jason Merrill wrote: On 05/05/2014 12:25 PM, Paolo Carlini wrote: Good, but is it Ok to use uses_template_parms for that? A few days ago I struggled to find something simpler ready to use, to no avail (well, this is probably well known to you, but there are surprisin

Re: [C++ Patch] PR 60999

2014-05-05 Thread Jason Merrill
On 05/05/2014 12:25 PM, Paolo Carlini wrote: Good, but is it Ok to use uses_template_parms for that? A few days ago I struggled to find something simpler ready to use, to no avail (well, this is probably well known to you, but there are surprisingly few places in pt.c where we explain either in c

Re: [C++ Patch] PR 60999

2014-05-05 Thread Paolo Carlini
Hi, On 05/05/2014 03:32 PM, Jason Merrill wrote: On 05/03/2014 06:55 PM, Paolo Carlini wrote: On 05/03/2014 11:14 PM, Jason Merrill wrote: Do you want CLASSTYPE_IS_TEMPLATE here? Uhm, you mean something simple like the attached? Because certainly I tried it and didn't fully work (nsdmi-temp

Re: [C++ Patch] PR 60999

2014-05-05 Thread Jason Merrill
On 05/03/2014 06:55 PM, Paolo Carlini wrote: On 05/03/2014 11:14 PM, Jason Merrill wrote: Do you want CLASSTYPE_IS_TEMPLATE here? Uhm, you mean something simple like the attached? Because certainly I tried it and didn't fully work (nsdmi-template7.C?) and for sure we would regress on the belo

Re: [C++ Patch] PR 60999

2014-05-03 Thread Paolo Carlini
Hi, On 05/03/2014 11:14 PM, Jason Merrill wrote: On 04/30/2014 05:15 AM, Paolo Carlini wrote: decl = (CLASSTYPE_TEMPLATE_INFO (DECL_CONTEXT (decl)) +&& uses_template_parms (DECL_CONTEXT (decl)) Do you want CLASSTYPE_IS_TEMPLATE here? Uhm, you mean something simple like the attac

Re: [C++ Patch] PR 60999

2014-05-03 Thread Jason Merrill
On 04/30/2014 05:15 AM, Paolo Carlini wrote: decl = (CLASSTYPE_TEMPLATE_INFO (DECL_CONTEXT (decl)) + && uses_template_parms (DECL_CONTEXT (decl)) Do you want CLASSTYPE_IS_TEMPLATE here? Jason

[C++ Patch] PR 60999

2014-04-30 Thread Paolo Carlini
Hi, this regression unfortunately has to do with my fix for c++/57887, thus the code in maybe_begin_member_template_processing: if (nsdmi) decl = (CLASSTYPE_TEMPLATE_INFO (DECL_CONTEXT (decl)) ? CLASSTYPE_TI_TEMPLATE (DECL_CONTEXT (decl)) : NULL_TREE); The check is t