gc: better compilation of floating point +=
R=ken2
CC=golang-dev
https://golang.org/cl/255042
diff --git a/src/cmd/8g/ggen.c b/src/cmd/8g/ggen.c
index 549488d1..f6fa7da 100644
--- a/src/cmd/8g/ggen.c
+++ b/src/cmd/8g/ggen.c
@@ -471,21 +471,34 @@
}
hard:
- tempname(&n2, nr->type);
- cgen(nr, &n2);
-
- igen(nl, &n1, N);
+ n2.op = 0;
+ n1.op = 0;
+ if(nr->ullman >= nl->ullman || nl->addable) {
+ mgen(nr, &n2, N);
+ nr = &n2;
+ nr = &n2;
+ } else {
+ tempname(&n2, nr->type);
+ cgen(nr, &n2);
+ nr = &n2;
+ }
+ if(!nl->addable) {
+ igen(nl, &n1, N);
+ nl = &n1;
+ }
n3 = *n;
- n3.left = &n1;
- n3.right = &n2;
+ n3.left = nl;
+ n3.right = nr;
n3.op = n->etype;
- tempname(&n4, nl->type);
- cgen(&n3, &n4);
- gmove(&n4, &n1);
+ mgen(&n3, &n4, N);
+ gmove(&n4, nl);
- regfree(&n1);
+ if(n1.op)
+ regfree(&n1);
+ mfree(&n2);
+ mfree(&n4);
ret:
;
