------- Comment #9 from pault at gcc dot gnu dot org 2007-05-15 10:56 ------- (In reply to comment #8) > The patch below works and regtests OK. I told a lie - the version that regtests OK has the line
+ if (!integer_onep (info->stride[dim])) + stride_ne_one = 1; a bit further down, within the subsequent 'else' block. In fact, I have analysed this problem fully and am now regtesting a 'proper' patch, which I hope to submit tonight. Paul -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31867
