This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: Add checks for tree.value, tree.minval, and tree.maxval andothers


On Thu, 2004-03-25 at 23:44, Richard Kenner wrote:
>     The check for TYPE_DOMAIN seems wrong.  It is not allowing
>     INTEGER_TYPEs, which, according to doc/c-tree.texi, it should.  This
>     causes an ICE in tree-ssa inside c-typeck.c:comptypes() where we have
>     
>       if (TREE_CODE (t2) == INTEGER_TYPE && TYPE_IS_SIZETYPE (t2)
>           && TYPE_DOMAIN (t2) != 0)
>         t2 = TYPE_DOMAIN (t2);
> 
>     OK for mainline after bootstrapping?  
> 
> No.  Instead the TYPE_DOMAIN above should be TYPE_ORIG_SIZE_TYPE like
> comptypes in cp/typeck.c
>
How about the doc inconsistency?  Is that wrong too?  When did it
change?  The code I quoted comes from a patch of yours in March 2000.
There's at least another similar hunk in that file.


Diego.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]