From b064904c509decf9e038f29f903a2304851a913b Mon Sep 17 00:00:00 2001
From: Heiko Carstens <hca@linux.ibm.com>
Date: Tue, 11 Aug 2020 11:16:51 +0200
Subject: [PATCH] s390/checksum: provide csum_ipv6_magic()

This implementation needs only ~30% of the time to calculate the
checksum compared to the generic variant. In addition the compiler
also generates only ~30% of the instructions compared to the generic
variant (on z14, compiled with march=z196).

Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
---
 arch/s390/include/asm/checksum.h | 24 ++++++++++++++++++++++--
 1 file changed, 22 insertions(+), 2 deletions(-)

diff --git a/arch/s390/include/asm/checksum.h b/arch/s390/include/asm/checksum.h
index 6d01c96aeb5c4..f4b42db5d007d 100644
--- a/arch/s390/include/asm/checksum.h
+++ b/arch/s390/include/asm/checksum.h
@@ -13,6 +13,7 @@
 #define _S390_CHECKSUM_H
 
 #include <linux/uaccess.h>
+#include <linux/in6.h>
 
 /*
  * computes the checksum of a memory block at buff, length len,
@@ -115,6 +116,25 @@ static inline __sum16 ip_compute_csum(const void *buff, int len)
 	return csum_fold(csum_partial(buff, len, 0));
 }
 
-#endif /* _S390_CHECKSUM_H */
-
+#define _HAVE_ARCH_IPV6_CSUM
+static inline __sum16 csum_ipv6_magic(const struct in6_addr *saddr,
+				      const struct in6_addr *daddr,
+				      __u32 len, __u8 proto, __wsum csum)
+{
+	__u64 sum = (__force __u64)csum;
+
+	sum += (__force __u32)saddr->s6_addr32[0];
+	sum += (__force __u32)saddr->s6_addr32[1];
+	sum += (__force __u32)saddr->s6_addr32[2];
+	sum += (__force __u32)saddr->s6_addr32[3];
+	sum += (__force __u32)daddr->s6_addr32[0];
+	sum += (__force __u32)daddr->s6_addr32[1];
+	sum += (__force __u32)daddr->s6_addr32[2];
+	sum += (__force __u32)daddr->s6_addr32[3];
+	sum += len;
+	sum += proto;
+	sum += (sum >> 32) | (sum << 32);
+	return csum_fold((__force __wsum)(sum >> 32));
+}
 
+#endif /* _S390_CHECKSUM_H */
-- 
GitLab