This is the mail archive of the fortran@gcc.gnu.org mailing list for the GNU Fortran 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]

[FORTRAN PATCH] Avoid another useless NON_LVALUE_EXPR


Another trivial tweak to avoid fold-const.c constructing a NON_LVALUE_EXPR
caused by the gfortran front-end folding a multiplication by one.  I'm
actually fairly close to deprecating GCC's NON_LVALUE_EXPR tree code by
increasing the C front-end's use of "struct c_expr", but in the meantime
this makes gfortran's -fdump-tree-original output slightly more readable.

Tested on x86_64-unknown-linux-gnu with a full "make bootstrap", including
gfortran, and regression tested with a top-level "make -k check" with no
new failures.

Ok for mainline?

2007-01-04  Roger Sayle  <roger@eyesopen.com>

        * trans-array.c (gfc_conv_loop_setup):  Test whether the loop
        stride is one, to avoid fold_build2 introducing a useless
        NON_LVALUE_EXPR node.

Roger
--

Attachment: patchf1.txt
Description: Text document


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