[4.8, C++ Patch] PR 52422

[Paolo Carlini]

Hi,
> On 02/29/2012 06:53 AM, Paolo Carlini wrote:
>> Daniel noticed a couple other SFINAE issues, which I'm fixing in the
>> usual straightforward way. Tested x86_64-linux.
>
> When fixing these things, please add checking for error_mark_node 
> return as well.  For instance, in this patch the calls to 
> decay_conversion from build_conditional_expr_1 could return 
> error_mark_node, in which case we want to return out of 
> build_conditional_expr_1 as well.  Similarly with the call in 
> convert_like_real, and so on.
I see. Thus I added many of those checks, essentially all I could see 
besides some special cases like we call decay_conversion with an 
explicit tf_warning_or_error (thus we have the status quo behavior 
anyway) or when there are already checks, just a few lines below. In the 
feature I'll keep an eye to that.

Does the below look better? (booted and tested x86_64-linux)
>> PS: while working on the patch, I noticed that a few functions, which in
>> the meanwhile acquired a tsubst_flags_t parameter, aren't fully
>> propagating it, are still passing tf_warning_or_error. Apparently a
>> clean-up is in order.
> Please.  :)
Ok, will do (and in case more error_mark_node checks too, of course)

Thanks,
Paolo.

//////////////////////
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: patch_52422_2
URL: <http://gcc.gnu.org/pipermail/gcc-patches/attachments/20120301/fe9bfdff/attachment.ksh>