This is the mail archive of the gcc-bugs@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]

[Bug target/24193] [4.1 Regression] ICE in extract_insn while compiling libgfortran



------- Comment #6 from wilson at gcc dot gnu dot org  2005-10-06 17:41 -------
Loads can have post_increment addresses with an immediate or register
increment.  Stores can have post_increment address with only an immediate
increment.  We have a special predicate, destination_operand, that checks for
this.  Unfortunately, we haven't been using it consistently.  And people who
have noticed this problem have fixed only the one pattern that they noticed was
broken, instead of fixing all of them.

Scanning the md file, I see that movbi and movti_internal at least are broken. 
There may be more.  I think gr_spill_internal is also broken.  I'm off to a
meeting and will have to finish this later.


-- 

wilson at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |wilson at gcc dot gnu dot
                   |                            |org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=24193


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