fix dynamic stack allocs for amd64

The arm64 might have the same problem but it
is currently unable to handle them even in
instruction selection.

Thanks to Jean Dao for reporting the bug.

1 files changed, 12 insertions, 4 deletions

diff --git a/amd64/emit.c b/amd64/emit.c
index 51833b4..b8fa655 100644
--- a/amd64/emit.c
+++ b/amd64/emit.c
@@ -488,10 +488,10 @@ emitins(Ins i, Fn *fn, FILE *f)
 	}
 }
 
-static int
+static uint64_t
 framesz(Fn *fn)
 {
-	int i, o, f;
+	uint64_t i, o, f;
 
 	/* specific to NAlign == 3 */
 	for (i=0, o=0; i<NCLR; i++)
@@ -512,7 +512,8 @@ amd64_emitfn(Fn *fn, FILE *f)
 	static int id0;
 	Blk *b, *s;
 	Ins *i, itmp;
-	int *r, c, fs, o, n, lbl;
+	int *r, c, o, n, lbl;
+	uint64_t fs;
 
 	fprintf(f, ".text\n");
 	if (fn->export)
@@ -525,7 +526,7 @@ amd64_emitfn(Fn *fn, FILE *f)
 	);
 	fs = framesz(fn);
 	if (fs)
-		fprintf(f, "\tsub $%d, %%rsp\n", fs);
+		fprintf(f, "\tsub $%"PRIu64", %%rsp\n", fs);
 	if (fn->vararg) {
 		o = -176;
 		for (r=amd64_sysv_rsave; r<&amd64_sysv_rsave[6]; r++, o+=8)
@@ -537,6 +538,7 @@ amd64_emitfn(Fn *fn, FILE *f)
 		if (fn->reg & BIT(*r)) {
 			itmp.arg[0] = TMP(*r);
 			emitf("pushq %L0", &itmp, fn, f);
+			fs += 8;
 		}
 
 	for (lbl=0, b=fn->start; b; b=b->link) {
@@ -547,6 +549,12 @@ amd64_emitfn(Fn *fn, FILE *f)
 		lbl = 1;
 		switch (b->jmp.type) {
 		case Jret0:
+			if (fn->dynalloc)
+				fprintf(f,
+					"\tmovq %%rbp, %%rsp\n"
+					"\tsubq $%"PRIu64", %%rsp\n",
+					fs
+				);
 			for (r=&amd64_sysv_rclob[NCLR]; r>amd64_sysv_rclob;)
 				if (fn->reg & BIT(*--r)) {
 					itmp.arg[0] = TMP(*r);