LoongArch: Add vDSO syscall __vdso_getcpu()

We test 20 million times of getcpu(), the real syscall version take 25
seconds, while the vsyscall version take only 2.4 seconds.

Signed-off-by: Rui Wang <wangrui@loongson.cn>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>

1 files changed, 43 insertions, 0 deletions

diff --git a/arch/loongarch/vdso/vgetcpu.c b/arch/loongarch/vdso/vgetcpu.c
new file mode 100644
index 000000000000..43a0078e4418
--- /dev/null
+++ b/arch/loongarch/vdso/vgetcpu.c
@@ -0,0 +1,43 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Fast user context implementation of getcpu()
+ */
+
+#include <asm/vdso.h>
+#include <linux/getcpu.h>
+
+static __always_inline int read_cpu_id(void)
+{
+	int cpu_id;
+
+	__asm__ __volatile__(
+	"	rdtime.d $zero, %0\n"
+	: "=r" (cpu_id)
+	:
+	: "memory");
+
+	return cpu_id;
+}
+
+static __always_inline const struct vdso_pcpu_data *get_pcpu_data(void)
+{
+	return (struct vdso_pcpu_data *)(get_vdso_base() - VDSO_DATA_SIZE);
+}
+
+int __vdso_getcpu(unsigned int *cpu, unsigned int *node, struct getcpu_cache *unused)
+{
+	int cpu_id;
+	const struct vdso_pcpu_data *data;
+
+	cpu_id = read_cpu_id();
+
+	if (cpu)
+		*cpu = cpu_id;
+
+	if (node) {
+		data = get_pcpu_data();
+		*node = data[cpu_id].node;
+	}
+
+	return 0;
+}