csky: optimize memcpy_{from,to}io() and memset_io()

Optimize memcpy_{from,to}io() and memset_io() by transferring in
64 bit as much as possible with minimized barrier usage.  This
simplest optimization brings faster throughput compare to current
byte-by-byte read and write with barrier in the loop. Code's
skeleton is taken from the powerpc & arm64.

Signed-off-by: Guo Ren <guoren@linux.alibaba.com>
Signed-off-by: Guo Ren <guoren@kernel.org>

arch/csky/include/asm/io.h

@@ -31,6 +31,17 @@
 #define writel(v,c)		({ wmb(); writel_relaxed((v),(c)); mb(); })
 #endif
 
+/*
+ * String version of I/O memory access operations.
+ */
+extern void __memcpy_fromio(void *, const volatile void __iomem *, size_t);
+extern void __memcpy_toio(volatile void __iomem *, const void *, size_t);
+extern void __memset_io(volatile void __iomem *, int, size_t);
+
+#define memset_io(c,v,l)        __memset_io((c),(v),(l))
+#define memcpy_fromio(a,c,l)    __memcpy_fromio((a),(c),(l))
+#define memcpy_toio(c,a,l)      __memcpy_toio((c),(a),(l))
+
 /*
  * I/O memory mapping functions.
  */

arch/csky/kernel/Makefile

@@ -2,7 +2,7 @@
 extra-y := head.o vmlinux.lds
 
 obj-y += entry.o atomic.o signal.o traps.o irq.o time.o vdso.o vdso/
-obj-y += power.o syscall.o syscall_table.o setup.o
+obj-y += power.o syscall.o syscall_table.o setup.o io.o
 obj-y += process.o cpu-probe.o ptrace.o stacktrace.o
 obj-y += probes/
 

arch/csky/kernel/io.c

@@ -0,0 +1,91 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <linux/export.h>
+#include <linux/types.h>
+#include <linux/io.h>
+
+/*
+ * Copy data from IO memory space to "real" memory space.
+ */
+void __memcpy_fromio(void *to, const volatile void __iomem *from, size_t count)
+{
+	while (count && !IS_ALIGNED((unsigned long)from, 4)) {
+		*(u8 *)to = __raw_readb(from);
+		from++;
+		to++;
+		count--;
+	}
+
+	while (count >= 4) {
+		*(u32 *)to = __raw_readl(from);
+		from += 4;
+		to += 4;
+		count -= 4;
+	}
+
+	while (count) {
+		*(u8 *)to = __raw_readb(from);
+		from++;
+		to++;
+		count--;
+	}
+}
+EXPORT_SYMBOL(__memcpy_fromio);
+
+/*
+ * Copy data from "real" memory space to IO memory space.
+ */
+void __memcpy_toio(volatile void __iomem *to, const void *from, size_t count)
+{
+	while (count && !IS_ALIGNED((unsigned long)to, 4)) {
+		__raw_writeb(*(u8 *)from, to);
+		from++;
+		to++;
+		count--;
+	}
+
+	while (count >= 4) {
+		__raw_writel(*(u32 *)from, to);
+		from += 4;
+		to += 4;
+		count -= 4;
+	}
+
+	while (count) {
+		__raw_writeb(*(u8 *)from, to);
+		from++;
+		to++;
+		count--;
+	}
+}
+EXPORT_SYMBOL(__memcpy_toio);
+
+/*
+ * "memset" on IO memory space.
+ */
+void __memset_io(volatile void __iomem *dst, int c, size_t count)
+{
+	u32 qc = (u8)c;
+
+	qc |= qc << 8;
+	qc |= qc << 16;
+
+	while (count && !IS_ALIGNED((unsigned long)dst, 4)) {
+		__raw_writeb(c, dst);
+		dst++;
+		count--;
+	}
+
+	while (count >= 4) {
+		__raw_writel(qc, dst);
+		dst += 4;
+		count -= 4;
+	}
+
+	while (count) {
+		__raw_writeb(c, dst);
+		dst++;
+		count--;
+	}
+}
+EXPORT_SYMBOL(__memset_io);