[PATCH v4 8/8] vga: Optimise console scrolling

[Matthew Wilcox]

From: Matthew Wilcox <mawilcox@xxxxxxxxxxxxx>

Where possible, call memset16(), memmove() or memcpy() instead of using
open-coded loops.  I don't like the calling convention that uses a byte
count instead of a count of u16s, but it's a little late to change that.
Reduces code size of fbcon.o by almost 400 bytes on my laptop build.

Signed-off-by: Matthew Wilcox <mawilcox@xxxxxxxxxxxxx>
---
 arch/mips/include/asm/vga.h    |  7 +++++++
 arch/powerpc/include/asm/vga.h |  8 ++++++++
 arch/sparc/include/asm/vga.h   | 25 +++++++++++++++++++++++++
 include/linux/vt_buffer.h      | 12 ++++++++++++
 4 files changed, 52 insertions(+)

diff --git a/arch/mips/include/asm/vga.h b/arch/mips/include/asm/vga.h
index f82c83749a08..975ff51f80c4 100644
--- a/arch/mips/include/asm/vga.h
+++ b/arch/mips/include/asm/vga.h
@@ -6,6 +6,7 @@
 #ifndef _ASM_VGA_H
 #define _ASM_VGA_H
 
+#include <linux/string.h>
 #include <asm/addrspace.h>
 #include <asm/byteorder.h>
 
@@ -40,9 +41,15 @@ static inline u16 scr_readw(volatile const u16 *addr)
 	return le16_to_cpu(*addr);
 }
 
+static inline void scr_memsetw(u16 *s, u16 v, unsigned int count)
+{
+	memset16(s, cpu_to_le16(v), count / 2);
+}
+
 #define scr_memcpyw(d, s, c) memcpy(d, s, c)
 #define scr_memmovew(d, s, c) memmove(d, s, c)
 #define VT_BUF_HAVE_MEMCPYW
 #define VT_BUF_HAVE_MEMMOVEW
+#define VT_BUF_HAVE_MEMSETW
 
 #endif /* _ASM_VGA_H */
diff --git a/arch/powerpc/include/asm/vga.h b/arch/powerpc/include/asm/vga.h
index ab3acd2f2786..7a7b541b7493 100644
--- a/arch/powerpc/include/asm/vga.h
+++ b/arch/powerpc/include/asm/vga.h
@@ -33,8 +33,16 @@ static inline u16 scr_readw(volatile const u16 *addr)
 	return le16_to_cpu(*addr);
 }
 
+#define VT_BUF_HAVE_MEMSETW
+static inline void scr_memsetw(u16 *s, u16 v, unsigned int n)
+{
+	memset16(s, cpu_to_le16(v), n / 2);
+}
+
 #define VT_BUF_HAVE_MEMCPYW
+#define VT_BUF_HAVE_MEMMOVEW
 #define scr_memcpyw	memcpy
+#define scr_memmovew	memmove
 
 #endif /* !CONFIG_VGA_CONSOLE && !CONFIG_MDA_CONSOLE */
 
diff --git a/arch/sparc/include/asm/vga.h b/arch/sparc/include/asm/vga.h
index ec0e9967d93d..f54e8b6fb197 100644
--- a/arch/sparc/include/asm/vga.h
+++ b/arch/sparc/include/asm/vga.h
@@ -8,9 +8,13 @@
 #define _LINUX_ASM_VGA_H_
 
 #include <linux/bug.h>
+#include <linux/string.h>
 #include <asm/types.h>
 
 #define VT_BUF_HAVE_RW
+#define VT_BUF_HAVE_MEMSETW
+#define VT_BUF_HAVE_MEMCPYW
+#define VT_BUF_HAVE_MEMMOVEW
 
 #undef scr_writew
 #undef scr_readw
@@ -29,6 +33,27 @@ static inline u16 scr_readw(const u16 *addr)
 	return *addr;
 }
 
+static inline void scr_memsetw(u16 *p, u16 v, unsigned int n)
+{
+	BUG_ON((long) p >= 0);
+
+	memset16(p, cpu_to_le16(v), n / 2);
+}
+
+static inline void scr_memcpyw(u16 *d, u16 *s, unsigned int n)
+{
+	BUG_ON((long) d >= 0);
+
+	memcpy(d, s, n);
+}
+
+static inline void scr_memmovew(u16 *d, u16 *s, unsigned int n)
+{
+	BUG_ON((long) d >= 0);
+
+	memmove(d, s, n);
+}
+
 #define VGA_MAP_MEM(x,s) (x)
 
 #endif
diff --git a/include/linux/vt_buffer.h b/include/linux/vt_buffer.h
index f38c10ba3ff5..31b92fcd8f03 100644
--- a/include/linux/vt_buffer.h
+++ b/include/linux/vt_buffer.h
@@ -26,24 +26,33 @@
 #ifndef VT_BUF_HAVE_MEMSETW
 static inline void scr_memsetw(u16 *s, u16 c, unsigned int count)
 {
+#ifdef VT_BUF_HAVE_RW
 	count /= 2;
 	while (count--)
 		scr_writew(c, s++);
+#else
+	memset16(s, c, count / 2);
+#endif
 }
 #endif
 
 #ifndef VT_BUF_HAVE_MEMCPYW
 static inline void scr_memcpyw(u16 *d, const u16 *s, unsigned int count)
 {
+#ifdef VT_BUF_HAVE_RW
 	count /= 2;
 	while (count--)
 		scr_writew(scr_readw(s++), d++);
+#else
+	memcpy(d, s, count);
+#endif
 }
 #endif
 
 #ifndef VT_BUF_HAVE_MEMMOVEW
 static inline void scr_memmovew(u16 *d, const u16 *s, unsigned int count)
 {
+#ifdef VT_BUF_HAVE_RW
 	if (d < s)
 		scr_memcpyw(d, s, count);
 	else {
@@ -53,6 +62,9 @@ static inline void scr_memmovew(u16 *d, const u16 *s, unsigned int count)
 		while (count--)
 			scr_writew(scr_readw(--s), --d);
 	}
+#else
+	memmove(d, s, count);
+#endif
 }
 #endif
 
-- 
2.13.2