This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: reload1.c and i386.md patches for reload bug
>Shouldn't this be added to subdi3_1 and others (maybe ashldi3_{1,2},
>ashrdi3_{1,2}, lshrdi3_{1,2}) too?
Yes, it should be in subdi3_1. I don't think it does anything useful for the
shift patterns, since they have predicates that don't allow MEMs, or constants
in operand 1, and hence ix86_binary_operator_ok will always return one for
them.
I'll write a subdi3_1 patch and do a bootstrap tomorrow to test it.
Jim