Add Neon support to ARMv7

This patch enables the use of the Neon extension on ARMv7 (Cortex-A8).
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>

arch/arm/Kconfig

@@ -806,6 +806,13 @@ config VFPv3
 	depends on VFP
 	default y if CPU_V7
 
+config NEON
+	bool "NEON Advanced SIMD Extension support"
+	depends on VFPv3 && CPU_V7
+	help
+	  Say Y to include support code for NEON, the ARMv7 Advanced SIMD
+	  Extensions.
+
 endmenu
 
 menu "Userspace binary formats"

arch/arm/kernel/entry-armv.S

@@ -473,12 +473,32 @@ __und_usr:
  * co-processor instructions.  However, we have to watch out
  * for the ARM6/ARM7 SWI bug.
  *
+ * Neon is a special case that has to be handled here. Not all
+ * Neon instructions are co-processor instructions, so we have
+ * to make a special case of checking for them. Plus, there's
+ * five groups of them, so we have a table of mask/opcode pairs
+ * to check against, and if any match then we branch off into the
+ * Neon handler code.
+ *
  * Emulators may wish to make use of the following registers:
  *  r0  = instruction opcode.
  *  r2  = PC+4
  *  r10 = this threads thread_info structure.
  */
 call_fpe:
+#ifdef CONFIG_NEON
+	adr	r6, .LCneon_opcodes
+2:
+	ldr	r7, [r6],#4			@ mask value
+	cmp	r7, #0
+	beq	1f				@ if mask is 0 then we've done
+	and	r8, r0, r7
+	ldr	r7, [r6],#4			@ opcode bits matching in mask
+	cmp	r8, r7
+	beq	undef_neon			@ if == then it's a neon opcode
+	b	2b
+1:
+#endif
 	tst	r0, #0x08000000			@ only CDP/CPRT/LDC/STC have bit 27
 #if defined(CONFIG_CPU_ARM610) || defined(CONFIG_CPU_ARM710)
 	and	r8, r0, #0x0f000000		@ mask out op-code bits
@@ -528,6 +548,29 @@ call_fpe:
 	mov	pc, lr				@ CP#14 (Debug)
 	mov	pc, lr				@ CP#15 (Control)
 
+#ifdef CONFIG_NEON
+	.align	6
+
+.LCneon_opcodes:
+	.word	0xfe000000			@ mask
+	.word	0xf2000000			@ opcode
+
+	.word	0x0e000f00			@ mask
+	.word	0x0c000b00			@ opcode
+
+	.word	0xff100000			@ mask
+	.word	0xf4000000			@ opcode
+
+	.word	0x0f000f10			@ mask
+	.word	0x0e000b10			@ opcode
+
+	.word	0x0fe00fd0			@ mask
+	.word	0x0c400b10			@ opcode
+
+	.word	0x00000000			@ mask
+	.word	0x00000000			@ opcode
+#endif
+
 do_fpe:
 	enable_irq
 	ldr	r4, .LCfp
@@ -548,6 +591,13 @@ ENTRY(fp_enter)
 	.word	fpundefinstr
 	.text
 
+#ifdef CONFIG_NEON
+undef_neon:
+	get_thread_info r10
+	enable_irq
+	b	do_vfp				@ Let VFP handler handle this
+#endif
+
 fpundefinstr:
 	mov	r0, sp
 	adr	lr, ret_from_exception