Trans-TU influence on phases 1 through 7: prior art?
137 views
Skip to first unread message
James Widman
Mar 5, 2014, 1:40:07 AM3/5/14
to mod...@isocpp.org
I’m going to find compiler documentation that mentions this (with any C++ compiler), but in case anyone out there knows already:
In any C90, C99, C11, C++98, C++2003, C++2011, or C++2014 implementation, is there a pragma that has the following effect?
Suppose we have two translation units, X and Y.
During phase 4 for X, we encounter a pragma, and the execution of that pragma in the context of X is documented as altering one or more phases between 1 through 7 inclusive during translation of Y.
That is, at least one of the phases 1-7 for Y behave differently depending on whether X is translated.
If anyone can name that pragma, could you please post a link to further info?
—James
In any C90, C99, C11, C++98, C++2003, C++2011, or C++2014 implementation, is there a pragma that has the following effect?
Suppose we have two translation units, X and Y.
During phase 4 for X, we encounter a pragma, and the execution of that pragma in the context of X is documented as altering one or more phases between 1 through 7 inclusive during translation of Y.
That is, at least one of the phases 1-7 for Y behave differently depending on whether X is translated.
If anyone can name that pragma, could you please post a link to further info?
—James
James Widman
Mar 5, 2014, 1:43:27 AM3/5/14
to mod...@isocpp.org
On Mar 5, 2014, at 1:40 AM, James Widman <james....@gmail.com> wrote:
> I’m going to find compiler documentation that mentions this (with any C++ compiler),
—James
Reply all
Reply to author
Forward
0 new messages