On 06/17/14 14:35, Kai Tietz wrote:
I just did retest my testcase with recent source. I can't reproduce
this missed optimization before sched2 pass anymore. I moved second
peephole2 pass just before split_before_sched2 and everything got
caught.
Let's go with this if your idea of using a define_split doesn't work out.
To remove first peephole2 pass seems to cause weaker code for
impossible pushes, etc
OK.
Nevertheless it might be a point to make this new peephole instead a
define_split? I admit that this operation isn't a split, nevertheless
we would avoid a second peephole pass.
Doesn't hurt to try and as you say, if we can avoid a 2nd peep2 pass,
that's good.
jeff
