------- Comment #1 from ramana at gcc dot gnu dot org 2010-03-15 11:58 -------
though trunk of a recent vintage generates the following bit of code which is
slightly better. There's no reason why we can't expand in a better manner and
generate the vld1.f32. Marking as an enhancement.
mov r3, #0
ldr r2, [r2, #0] @ float
vdup.32 d16, r3
ldr r3, [r1, #0] @ float
vmov d17, d16 @ v2sf
vmov.32 d17[0], r2
vmov.32 d16[0], r3
vadd.f32 d16, d16, d17
vst1.32 {d16[0]}, [r0]
bx lr
--
ramana at gcc dot gnu dot org changed:
What |Removed |Added
----------------------------------------------------------------------------
Severity|normal |enhancement
Status|UNCONFIRMED |NEW
Component|c |target
Ever Confirmed|0 |1
Keywords| |missed-optimization
Known to fail| |4.4.3 4.5.0
Last reconfirmed|0000-00-00 00:00:00 |2010-03-15 11:58:16
date| |
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43364
