Ping ~Aaron
On Wed, Jul 17, 2013 at 8:57 PM, Aaron Ballman <[email protected]> wrote: > There are type attributes and target-specific attributes which get > somewhat special semantic handling. By codifying this as part of the > tablegen file, we can generate more specific code in the future as > well as more cleanly express intent in the tablegen. > > This patch is laying some groundwork for eventually removing much of > the redundancy from SemaDeclAttr.cpp. More patches on that are > coming. In and of itself, this patch removes some code from Attr.td > (namely ASTNode = 0 declarations on various type attributes), and > fixes a mistake with Ptr32, Ptr64, UPtr and SPtr generating AST nodes > that are never actually used anywhere. > > Note that this patch does not add a TypeAttr base class like it does > for InheritableAttr (et al) -- there are no type attributes which > generate AST nodes currently (this is the same reason there is no > TYPE_ATTR macro). > > ~Aaron _______________________________________________ cfe-commits mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
