Use separate lines instead of semicolons.

macOS aarch64 assembly dialect treats ; as comment instead of a newline

ok tb@, jsing@

1 files changed, 10 insertions, 4 deletions

diff --git a/src/lib/libcrypto/bn/arch/aarch64/bn_arch.h b/src/lib/libcrypto/bn/arch/aarch64/bn_arch.h
index 9d61bc9114..b382d1a860 100644
--- a/src/lib/libcrypto/bn/arch/aarch64/bn_arch.h
+++ b/src/lib/libcrypto/bn/arch/aarch64/bn_arch.h
@@ -1,4 +1,4 @@
-/*      $OpenBSD: bn_arch.h,v 1.5 2023/02/17 05:46:57 jsing Exp $ */
+/*      $OpenBSD: bn_arch.h,v 1.6 2023/02/25 15:39:40 bcook Exp $ */
 /*
  * Copyright (c) 2023 Joel Sing <jsing@openbsd.org>
  *
@@ -31,7 +31,9 @@ bn_addw(BN_ULONG a, BN_ULONG b, BN_ULONG *out_r1, BN_ULONG *out_r0)
 {
         BN_ULONG carry, r0;
 
-        __asm__ ("adds %1, %2, %3; cset %0, cs"
+        __asm__ (
+                "adds %1, %2, %3 \n"
+                "cset %0, cs"
             : "=r"(carry), "=r"(r0)
             : "r"(a), "r"(b)
             : "cc");
@@ -48,7 +50,9 @@ bn_mulw(BN_ULONG a, BN_ULONG b, BN_ULONG *out_r1, BN_ULONG *out_r0)
         BN_ULONG r1, r0;
 
         /* Unsigned multiplication using a umulh/mul pair. */
-        __asm__ ("umulh %0, %2, %3; mul %1, %2, %3"
+        __asm__ (
+                "umulh %0, %2, %3 \n"
+                "mul %1, %2, %3"
             : "=&r"(r1), "=r"(r0)
             : "r"(a), "r"(b));
 
@@ -63,7 +67,9 @@ bn_subw(BN_ULONG a, BN_ULONG b, BN_ULONG *out_borrow, BN_ULONG *out_r0)
 {
         BN_ULONG borrow, r0;
 
-        __asm__ ("subs %1, %2, %3; cset %0, cc"
+        __asm__ (
+                "subs %1, %2, %3 \n"
+                "cset %0, cc"
             : "=r"(borrow), "=r"(r0)
             : "r"(a), "r"(b)
             : "cc");