Merge branch 'upstream' of git://git.linux-mips.org/pub/scm/ralf/upstream-linus

Pull MIPS fixes from Ralf Baechle:
 "A fair number of 4.2 fixes also because Markos opened the flood gates.

   - Patch up the math used calculate the location for the page bitmap.

   - The FDC (Not what you think, FDC stands for Fast Debug Channel) IRQ
     around was causing issues on non-Malta platforms, so move the code
     to a Malta specific location.

   - A spelling fix replicated through several files.

   - Fix to the emulation of an R2 instruction for R6 cores.

   - Fix the JR emulation for R6.

   - Further patching of mindless 64 bit issues.

   - Ensure the kernel won't crash on CPUs with L2 caches with >= 8
     ways.

   - Use compat_sys_getsockopt for O32 ABI on 64 bit kernels.

   - Fix cache flushing for multithreaded cores.

   - A build fix"

* 'upstream' of git://git.linux-mips.org/pub/scm/ralf/upstream-linus:
  MIPS: O32: Use compat_sys_getsockopt.
  MIPS: c-r4k: Extend way_string array
  MIPS: Pistachio: Support CDMM & Fast Debug Channel
  MIPS: Malta: Make GIC FDC IRQ workaround Malta specific
  MIPS: c-r4k: Fix cache flushing for MT cores
  Revert "MIPS: Kconfig: Disable SMP/CPS for 64-bit"
  MIPS: cps-vec: Use macros for various arithmetics and memory operations
  MIPS: kernel: cps-vec: Replace KSEG0 with CKSEG0
  MIPS: kernel: cps-vec: Use ta0-ta3 pseudo-registers for 64-bit
  MIPS: kernel: cps-vec: Replace mips32r2 ISA level with mips64r2
  MIPS: kernel: cps-vec: Replace 'la' macro with PTR_LA
  MIPS: kernel: smp-cps: Fix 64-bit compatibility errors due to pointer casting
  MIPS: Fix erroneous JR emulation for MIPS R6
  MIPS: Fix branch emulation for BLTC and BGEC instructions
  MIPS: kernel: traps: Fix broken indentation
  MIPS: bootmem: Don't use memory holes for page bitmap
  MIPS: O32: Do not handle require 32 bytes from the stack to be readable.
  MIPS, CPUFREQ: Fix spelling of Institute.
  MIPS: Lemote 2F: Fix build caused by recent mass rename.

27 files changed, 214 insertions, 123 deletions

diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index 2a14585c90d2..aab7e46cadd5 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -2231,7 +2231,7 @@ config MIPS_CMP
 
 config MIPS_CPS
 	bool "MIPS Coherent Processing System support"
-	depends on SYS_SUPPORTS_MIPS_CPS && !64BIT
+	depends on SYS_SUPPORTS_MIPS_CPS
 	select MIPS_CM
 	select MIPS_CPC
 	select MIPS_CPS_PM if HOTPLUG_CPU
diff --git a/arch/mips/include/asm/mach-loongson64/mmzone.h b/arch/mips/include/asm/mach-loongson64/mmzone.h
index 37c08a27b4f0..c9f7e231e66b 100644
--- a/arch/mips/include/asm/mach-loongson64/mmzone.h
+++ b/arch/mips/include/asm/mach-loongson64/mmzone.h
@@ -1,6 +1,6 @@
 /*
  * Copyright (C) 2010 Loongson Inc. & Lemote Inc. &
- *                    Insititute of Computing Technology
+ *                    Institute of Computing Technology
  * Author:  Xiang Gao, gaoxiang@ict.ac.cn
  *          Huacai Chen, chenhc@lemote.com
  *          Xiaofu Meng, Shuangshuang Zhang
diff --git a/arch/mips/include/asm/smp.h b/arch/mips/include/asm/smp.h
index 2b25d1ba1ea0..16f1ea9ab191 100644
--- a/arch/mips/include/asm/smp.h
+++ b/arch/mips/include/asm/smp.h
@@ -23,6 +23,7 @@
 extern int smp_num_siblings;
 extern cpumask_t cpu_sibling_map[];
 extern cpumask_t cpu_core_map[];
+extern cpumask_t cpu_foreign_map;
 
 #define raw_smp_processor_id() (current_thread_info()->cpu)
 
diff --git a/arch/mips/kernel/branch.c b/arch/mips/kernel/branch.c
index c0c5e5972256..d8f9b357b222 100644
--- a/arch/mips/kernel/branch.c
+++ b/arch/mips/kernel/branch.c
@@ -600,7 +600,7 @@ int __compute_return_epc_for_insn(struct pt_regs *regs,
 		break;
 
 	case blezl_op: /* not really i_format */
-		if (NO_R6EMU)
+		if (!insn.i_format.rt && NO_R6EMU)
 			goto sigill_r6;
 	case blez_op:
 		/*
@@ -635,7 +635,7 @@ int __compute_return_epc_for_insn(struct pt_regs *regs,
 		break;
 
 	case bgtzl_op:
-		if (NO_R6EMU)
+		if (!insn.i_format.rt && NO_R6EMU)
 			goto sigill_r6;
 	case bgtz_op:
 		/*
diff --git a/arch/mips/kernel/cps-vec.S b/arch/mips/kernel/cps-vec.S
index 55b759a0019e..1b6ca634e646 100644
--- a/arch/mips/kernel/cps-vec.S
+++ b/arch/mips/kernel/cps-vec.S
@@ -60,7 +60,7 @@ LEAF(mips_cps_core_entry)
 	 nop
 
 	/* This is an NMI */
-	la	k0, nmi_handler
+	PTR_LA	k0, nmi_handler
 	jr	k0
 	 nop
 
@@ -107,10 +107,10 @@ not_nmi:
 	mul	t1, t1, t0
 	mul	t1, t1, t2
 
-	li	a0, KSEG0
-	add	a1, a0, t1
+	li	a0, CKSEG0
+	PTR_ADD	a1, a0, t1
 1:	cache	Index_Store_Tag_I, 0(a0)
-	add	a0, a0, t0
+	PTR_ADD	a0, a0, t0
 	bne	a0, a1, 1b
 	 nop
 icache_done:
@@ -134,12 +134,12 @@ icache_done:
 	mul	t1, t1, t0
 	mul	t1, t1, t2
 
-	li	a0, KSEG0
-	addu	a1, a0, t1
-	subu	a1, a1, t0
+	li	a0, CKSEG0
+	PTR_ADDU a1, a0, t1
+	PTR_SUBU a1, a1, t0
 1:	cache	Index_Store_Tag_D, 0(a0)
 	bne	a0, a1, 1b
-	 add	a0, a0, t0
+	 PTR_ADD a0, a0, t0
 dcache_done:
 
 	/* Set Kseg0 CCA to that in s0 */
@@ -152,11 +152,11 @@ dcache_done:
 
 	/* Enter the coherent domain */
 	li	t0, 0xff
-	sw	t0, GCR_CL_COHERENCE_OFS(v1)
+	PTR_S	t0, GCR_CL_COHERENCE_OFS(v1)
 	ehb
 
 	/* Jump to kseg0 */
-	la	t0, 1f
+	PTR_LA	t0, 1f
 	jr	t0
 	 nop
 
@@ -178,9 +178,9 @@ dcache_done:
 	 nop
 
 	/* Off we go! */
-	lw	t1, VPEBOOTCFG_PC(v0)
-	lw	gp, VPEBOOTCFG_GP(v0)
-	lw	sp, VPEBOOTCFG_SP(v0)
+	PTR_L	t1, VPEBOOTCFG_PC(v0)
+	PTR_L	gp, VPEBOOTCFG_GP(v0)
+	PTR_L	sp, VPEBOOTCFG_SP(v0)
 	jr	t1
 	 nop
 	END(mips_cps_core_entry)
@@ -217,7 +217,7 @@ LEAF(excep_intex)
 
 .org 0x480
 LEAF(excep_ejtag)
-	la	k0, ejtag_debug_handler
+	PTR_LA	k0, ejtag_debug_handler
 	jr	k0
 	 nop
 	END(excep_ejtag)
@@ -229,7 +229,7 @@ LEAF(mips_cps_core_init)
 	 nop
 
 	.set	push
-	.set	mips32r2
+	.set	mips64r2
 	.set	mt
 
 	/* Only allow 1 TC per VPE to execute... */
@@ -237,7 +237,7 @@ LEAF(mips_cps_core_init)
 
 	/* ...and for the moment only 1 VPE */
 	dvpe
-	la	t1, 1f
+	PTR_LA	t1, 1f
 	jr.hb	t1
 	 nop
 
@@ -250,25 +250,25 @@ LEAF(mips_cps_core_init)
 	mfc0	t0, CP0_MVPCONF0
 	srl	t0, t0, MVPCONF0_PVPE_SHIFT
 	andi	t0, t0, (MVPCONF0_PVPE >> MVPCONF0_PVPE_SHIFT)
-	addiu	t7, t0, 1
+	addiu	ta3, t0, 1
 
 	/* If there's only 1, we're done */
 	beqz	t0, 2f
 	 nop
 
 	/* Loop through each VPE within this core */
-	li	t5, 1
+	li	ta1, 1
 
 1:	/* Operate on the appropriate TC */
-	mtc0	t5, CP0_VPECONTROL
+	mtc0	ta1, CP0_VPECONTROL
 	ehb
 
 	/* Bind TC to VPE (1:1 TC:VPE mapping) */
-	mttc0	t5, CP0_TCBIND
+	mttc0	ta1, CP0_TCBIND
 
 	/* Set exclusive TC, non-active, master */
 	li	t0, VPECONF0_MVP
-	sll	t1, t5, VPECONF0_XTC_SHIFT
+	sll	t1, ta1, VPECONF0_XTC_SHIFT
 	or	t0, t0, t1
 	mttc0	t0, CP0_VPECONF0
 
@@ -280,8 +280,8 @@ LEAF(mips_cps_core_init)
 	mttc0	t0, CP0_TCHALT
 
 	/* Next VPE */
-	addiu	t5, t5, 1
-	slt	t0, t5, t7
+	addiu	ta1, ta1, 1
+	slt	t0, ta1, ta3
 	bnez	t0, 1b
 	 nop
 
@@ -298,19 +298,19 @@ LEAF(mips_cps_core_init)
 
 LEAF(mips_cps_boot_vpes)
 	/* Retrieve CM base address */
-	la	t0, mips_cm_base
-	lw	t0, 0(t0)
+	PTR_LA	t0, mips_cm_base
+	PTR_L	t0, 0(t0)
 
 	/* Calculate a pointer to this cores struct core_boot_config */
-	lw	t0, GCR_CL_ID_OFS(t0)
+	PTR_L	t0, GCR_CL_ID_OFS(t0)
 	li	t1, COREBOOTCFG_SIZE
 	mul	t0, t0, t1
-	la	t1, mips_cps_core_bootcfg
-	lw	t1, 0(t1)
-	addu	t0, t0, t1
+	PTR_LA	t1, mips_cps_core_bootcfg
+	PTR_L	t1, 0(t1)
+	PTR_ADDU t0, t0, t1
 
 	/* Calculate this VPEs ID. If the core doesn't support MT use 0 */
-	has_mt	t6, 1f
+	has_mt	ta2, 1f
 	 li	t9, 0
 
 	/* Find the number of VPEs present in the core */
@@ -334,24 +334,24 @@ LEAF(mips_cps_boot_vpes)
 1:	/* Calculate a pointer to this VPEs struct vpe_boot_config */
 	li	t1, VPEBOOTCFG_SIZE
 	mul	v0, t9, t1
-	lw	t7, COREBOOTCFG_VPECONFIG(t0)
-	addu	v0, v0, t7
+	PTR_L	ta3, COREBOOTCFG_VPECONFIG(t0)
+	PTR_ADDU v0, v0, ta3
 
 #ifdef CONFIG_MIPS_MT
 
 	/* If the core doesn't support MT then return */
-	bnez	t6, 1f
+	bnez	ta2, 1f
 	 nop
 	jr	ra
 	 nop
 
 	.set	push
-	.set	mips32r2
+	.set	mips64r2
 	.set	mt
 
 1:	/* Enter VPE configuration state */
 	dvpe
-	la	t1, 1f
+	PTR_LA	t1, 1f
 	jr.hb	t1
 	 nop
 1:	mfc0	t1, CP0_MVPCONTROL
@@ -360,12 +360,12 @@ LEAF(mips_cps_boot_vpes)
 	ehb
 
 	/* Loop through each VPE */
-	lw	t6, COREBOOTCFG_VPEMASK(t0)
-	move	t8, t6
-	li	t5, 0
+	PTR_L	ta2, COREBOOTCFG_VPEMASK(t0)
+	move	t8, ta2
+	li	ta1, 0
 
 	/* Check whether the VPE should be running. If not, skip it */
-1:	andi	t0, t6, 1
+1:	andi	t0, ta2, 1
 	beqz	t0, 2f
 	 nop
 
@@ -373,7 +373,7 @@ LEAF(mips_cps_boot_vpes)
 	mfc0	t0, CP0_VPECONTROL
 	ori	t0, t0, VPECONTROL_TARGTC
 	xori	t0, t0, VPECONTROL_TARGTC
-	or	t0, t0, t5
+	or	t0, t0, ta1
 	mtc0	t0, CP0_VPECONTROL
 	ehb
 
@@ -384,8 +384,8 @@ LEAF(mips_cps_boot_vpes)
 
 	/* Calculate a pointer to the VPEs struct vpe_boot_config */
 	li	t0, VPEBOOTCFG_SIZE
-	mul	t0, t0, t5
-	addu	t0, t0, t7
+	mul	t0, t0, ta1
+	addu	t0, t0, ta3
 
 	/* Set the TC restart PC */
 	lw	t1, VPEBOOTCFG_PC(t0)
@@ -423,9 +423,9 @@ LEAF(mips_cps_boot_vpes)
 	mttc0	t0, CP0_VPECONF0
 
 	/* Next VPE */
-2:	srl	t6, t6, 1
-	addiu	t5, t5, 1
-	bnez	t6, 1b
+2:	srl	ta2, ta2, 1
+	addiu	ta1, ta1, 1
+	bnez	ta2, 1b
 	 nop
 
 	/* Leave VPE configuration state */
@@ -445,7 +445,7 @@ LEAF(mips_cps_boot_vpes)
 	/* This VPE should be offline, halt the TC */
 	li	t0, TCHALT_H
 	mtc0	t0, CP0_TCHALT
-	la	t0, 1f
+	PTR_LA	t0, 1f
 1:	jr.hb	t0
 	 nop
 
@@ -466,10 +466,10 @@ LEAF(mips_cps_boot_vpes)
 	.set	noat
 	lw	$1, TI_CPU(gp)
 	sll	$1, $1, LONGLOG
-	la	\dest, __per_cpu_offset
+	PTR_LA	\dest, __per_cpu_offset
 	addu	$1, $1, \dest
 	lw	$1, 0($1)
-	la	\dest, cps_cpu_state
+	PTR_LA	\dest, cps_cpu_state
 	addu	\dest, \dest, $1
 	.set	pop
 	.endm
diff --git a/arch/mips/kernel/scall32-o32.S b/arch/mips/kernel/scall32-o32.S
index 6e8de80bb446..4cc13508d967 100644
--- a/arch/mips/kernel/scall32-o32.S
+++ b/arch/mips/kernel/scall32-o32.S
@@ -73,10 +73,11 @@ NESTED(handle_sys, PT_SIZE, sp)
 	.set    noreorder
 	.set	nomacro
 
-1:	user_lw(t5, 16(t0))		# argument #5 from usp
-4:	user_lw(t6, 20(t0))		# argument #6 from usp
-3:	user_lw(t7, 24(t0))		# argument #7 from usp
-2:	user_lw(t8, 28(t0))		# argument #8 from usp
+load_a4: user_lw(t5, 16(t0))		# argument #5 from usp
+load_a5: user_lw(t6, 20(t0))		# argument #6 from usp
+load_a6: user_lw(t7, 24(t0))		# argument #7 from usp
+load_a7: user_lw(t8, 28(t0))		# argument #8 from usp
+loads_done:
 
 	sw	t5, 16(sp)		# argument #5 to ksp
 	sw	t6, 20(sp)		# argument #6 to ksp
@@ -85,10 +86,10 @@ NESTED(handle_sys, PT_SIZE, sp)
 	.set	pop
 
 	.section __ex_table,"a"
-	PTR	1b,bad_stack
-	PTR	2b,bad_stack
-	PTR	3b,bad_stack
-	PTR	4b,bad_stack
+	PTR	load_a4, bad_stack_a4
+	PTR	load_a5, bad_stack_a5
+	PTR	load_a6, bad_stack_a6
+	PTR	load_a7, bad_stack_a7
 	.previous
 
 	lw	t0, TI_FLAGS($28)	# syscall tracing enabled?
@@ -153,8 +154,8 @@ syscall_trace_entry:
 /* ------------------------------------------------------------------------ */
 
 	/*
-	 * The stackpointer for a call with more than 4 arguments is bad.
-	 * We probably should handle this case a bit more drastic.
+	 * Our open-coded access area sanity test for the stack pointer
+	 * failed. We probably should handle this case a bit more drastic.
 	 */
 bad_stack:
 	li	v0, EFAULT
@@ -163,6 +164,22 @@ bad_stack:
 	sw	t0, PT_R7(sp)
 	j	o32_syscall_exit
 
+bad_stack_a4:
+	li	t5, 0
+	b	load_a5
+
+bad_stack_a5:
+	li	t6, 0
+	b	load_a6
+
+bad_stack_a6:
+	li	t7, 0
+	b	load_a7
+
+bad_stack_a7:
+	li	t8, 0
+	b	loads_done
+
 	/*
 	 * The system call does not exist in this kernel
 	 */
diff --git a/arch/mips/kernel/scall64-o32.S b/arch/mips/kernel/scall64-o32.S
index d07b210fbeff..f543ff4feef9 100644
--- a/arch/mips/kernel/scall64-o32.S
+++ b/arch/mips/kernel/scall64-o32.S
@@ -69,16 +69,17 @@ NESTED(handle_sys, PT_SIZE, sp)
 	daddu	t1, t0, 32
 	bltz	t1, bad_stack
 
-1:	lw	a4, 16(t0)		# argument #5 from usp
-2:	lw	a5, 20(t0)		# argument #6 from usp
-3:	lw	a6, 24(t0)		# argument #7 from usp
-4:	lw	a7, 28(t0)		# argument #8 from usp (for indirect syscalls)
+load_a4: lw	a4, 16(t0)		# argument #5 from usp
+load_a5: lw	a5, 20(t0)		# argument #6 from usp
+load_a6: lw	a6, 24(t0)		# argument #7 from usp
+load_a7: lw	a7, 28(t0)		# argument #8 from usp
+loads_done:
 
 	.section __ex_table,"a"
-	PTR	1b, bad_stack
-	PTR	2b, bad_stack
-	PTR	3b, bad_stack
-	PTR	4b, bad_stack
+	PTR	load_a4, bad_stack_a4
+	PTR	load_a5, bad_stack_a5
+	PTR	load_a6, bad_stack_a6
+	PTR	load_a7, bad_stack_a7
 	.previous
 
 	li	t1, _TIF_WORK_SYSCALL_ENTRY
@@ -167,6 +168,22 @@ bad_stack:
 	sd	t0, PT_R7(sp)
 	j	o32_syscall_exit
 
+bad_stack_a4:
+	li	a4, 0
+	b	load_a5
+
+bad_stack_a5:
+	li	a5, 0
+	b	load_a6
+
+bad_stack_a6:
+	li	a6, 0
+	b	load_a7
+
+bad_stack_a7:
+	li	a7, 0
+	b	loads_done
+
 not_o32_scall:
 	/*
 	 * This is not an o32 compatibility syscall, pass it on
@@ -383,7 +400,7 @@ EXPORT(sys32_call_table)
 	PTR	sys_connect			/* 4170 */
 	PTR	sys_getpeername
 	PTR	sys_getsockname
-	PTR	sys_getsockopt
+	PTR	compat_sys_getsockopt
 	PTR	sys_listen
 	PTR	compat_sys_recv			/* 4175 */
 	PTR	compat_sys_recvfrom
diff --git a/arch/mips/kernel/setup.c b/arch/mips/kernel/setup.c
index be73c491182b..008b3378653a 100644
--- a/arch/mips/kernel/setup.c
+++ b/arch/mips/kernel/setup.c
@@ -337,6 +337,11 @@ static void __init bootmem_init(void)
 			min_low_pfn = start;
 		if (end <= reserved_end)
 			continue;
+#ifdef CONFIG_BLK_DEV_INITRD
+		/* mapstart should be after initrd_end */
+		if (initrd_end && end <= (unsigned long)PFN_UP(__pa(initrd_end)))
+			continue;
+#endif
 		if (start >= mapstart)
 			continue;
 		mapstart = max(reserved_end, start);
@@ -366,14 +371,6 @@ static void __init bootmem_init(void)
 		max_low_pfn = PFN_DOWN(HIGHMEM_START);
 	}
 
-#ifdef CONFIG_BLK_DEV_INITRD
-	/*
-	 * mapstart should be after initrd_end
-	 */
-	if (initrd_end)
-		mapstart = max(mapstart, (unsigned long)PFN_UP(__pa(initrd_end)));
-#endif
-
 	/*
 	 * Initialize the boot-time allocator with low memory only.
 	 */
diff --git a/arch/mips/kernel/smp-cps.c b/arch/mips/kernel/smp-cps.c
index 4251d390b5b6..c88937745b4e 100644
--- a/arch/mips/kernel/smp-cps.c
+++ b/arch/mips/kernel/smp-cps.c
@@ -133,7 +133,7 @@ static void __init cps_prepare_cpus(unsigned int max_cpus)
 	/*
 	 * Patch the start of mips_cps_core_entry to provide:
 	 *
-	 * v0 = CM base address
+	 * v1 = CM base address
 	 * s0 = kseg0 CCA
 	 */
 	entry_code = (u32 *)&mips_cps_core_entry;
@@ -369,7 +369,7 @@ void play_dead(void)
 
 static void wait_for_sibling_halt(void *ptr_cpu)
 {
-	unsigned cpu = (unsigned)ptr_cpu;
+	unsigned cpu = (unsigned long)ptr_cpu;
 	unsigned vpe_id = cpu_vpe_id(&cpu_data[cpu]);
 	unsigned halted;
 	unsigned long flags;
@@ -430,7 +430,7 @@ static void cps_cpu_die(unsigned int cpu)
 		 */
 		err = smp_call_function_single(cpu_death_sibling,
 					       wait_for_sibling_halt,
-					       (void *)cpu, 1);
+					       (void *)(unsigned long)cpu, 1);
 		if (err)
 			panic("Failed to call remote sibling CPU\n");
 	}
diff --git a/arch/mips/kernel/smp.c b/arch/mips/kernel/smp.c
index faa46ebd9dda..d0744cc77ea7 100644
--- a/arch/mips/kernel/smp.c
+++ b/arch/mips/kernel/smp.c
@@ -63,6 +63,13 @@ EXPORT_SYMBOL(cpu_sibling_map);
 cpumask_t cpu_core_map[NR_CPUS] __read_mostly;
 EXPORT_SYMBOL(cpu_core_map);
 
+/*
+ * A logcal cpu mask containing only one VPE per core to
+ * reduce the number of IPIs on large MT systems.
+ */
+cpumask_t cpu_foreign_map __read_mostly;
+EXPORT_SYMBOL(cpu_foreign_map);
+
 /* representing cpus for which sibling maps can be computed */
 static cpumask_t cpu_sibling_setup_map;
 
@@ -103,6 +110,29 @@ static inline void set_cpu_core_map(int cpu)
 	}
 }
 
+/*
+ * Calculate a new cpu_foreign_map mask whenever a
+ * new cpu appears or disappears.
+ */
+static inline void calculate_cpu_foreign_map(void)
+{
+	int i, k, core_present;
+	cpumask_t temp_foreign_map;
+
+	/* Re-calculate the mask */
+	for_each_online_cpu(i) {
+		core_present = 0;
+		for_each_cpu(k, &temp_foreign_map)
+			if (cpu_data[i].package == cpu_data[k].package &&
+			    cpu_data[i].core == cpu_data[k].core)
+				core_present = 1;
+		if (!core_present)
+			cpumask_set_cpu(i, &temp_foreign_map);
+	}
+
+	cpumask_copy(&cpu_foreign_map, &temp_foreign_map);
+}
+
 struct plat_smp_ops *mp_ops;
 EXPORT_SYMBOL(mp_ops);
 
@@ -146,6 +176,8 @@ asmlinkage void start_secondary(void)
 	set_cpu_sibling_map(cpu);
 	set_cpu_core_map(cpu);
 
+	calculate_cpu_foreign_map();
+
 	cpumask_set_cpu(cpu, &cpu_callin_map);
 
 	synchronise_count_slave(cpu);
@@ -173,9 +205,18 @@ void __irq_entry smp_call_function_interrupt(void)
 static void stop_this_cpu(void *dummy)
 {
 	/*
-	 * Remove this CPU:
+	 * Remove this CPU. Be a bit slow here and
+	 * set the bits for every online CPU so we don't miss
+	 * any IPI whilst taking this VPE down.
 	 */
+
+	cpumask_copy(&cpu_foreign_map, cpu_online_mask);
+
+	/* Make it visible to every other CPU */
+	smp_mb();
+
 	set_cpu_online(smp_processor_id(), false);
+	calculate_cpu_foreign_map();
 	local_irq_disable();
 	while (1);
 }
@@ -197,6 +238,7 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
 	mp_ops->prepare_cpus(max_cpus);
 	set_cpu_sibling_map(0);
 	set_cpu_core_map(0);
+	calculate_cpu_foreign_map();
 #ifndef CONFIG_HOTPLUG_CPU
 	init_cpu_present(cpu_possible_mask);
 #endif
diff --git a/arch/mips/kernel/traps.c b/arch/mips/kernel/traps.c
index 2a7b38ed23f0..e207a43b5f8f 100644
--- a/arch/mips/kernel/traps.c
+++ b/arch/mips/kernel/traps.c
@@ -2130,10 +2130,10 @@ void per_cpu_trap_init(bool is_boot_cpu)
 	BUG_ON(current->mm);
 	enter_lazy_tlb(&init_mm, current);
 
-		/* Boot CPU's cache setup in setup_arch(). */
-		if (!is_boot_cpu)
-			cpu_cache_init();
-		tlb_init();
+	/* Boot CPU's cache setup in setup_arch(). */
+	if (!is_boot_cpu)
+		cpu_cache_init();
+	tlb_init();
 	TLBMISS_HANDLER_SETUP();
 }
 
diff --git a/arch/mips/loongson64/common/bonito-irq.c b/arch/mips/loongson64/common/bonito-irq.c
index cc0e4fd548e6..4e116d23bab3 100644
--- a/arch/mips/loongson64/common/bonito-irq.c
+++ b/arch/mips/loongson64/common/bonito-irq.c
@@ -3,7 +3,7 @@
  * Author: Jun Sun, jsun@mvista.com or jsun@junsun.net
  * Copyright (C) 2000, 2001 Ralf Baechle (ralf@gnu.org)
  *
- * Copyright (C) 2007 Lemote Inc. & Insititute of Computing Technology
+ * Copyright (C) 2007 Lemote Inc. & Institute of Computing Technology
  * Author: Fuxin Zhang, zhangfx@lemote.com
  *
  *  This program is free software; you can redistribute	 it and/or modify it
diff --git a/arch/mips/loongson64/common/cmdline.c b/arch/mips/loongson64/common/cmdline.c
index 72fed003a536..01fbed137028 100644
--- a/arch/mips/loongson64/common/cmdline.c
+++ b/arch/mips/loongson64/common/cmdline.c
@@ -6,7 +6,7 @@
  * Copyright 2003 ICT CAS
  * Author: Michael Guo <guoyi@ict.ac.cn>
  *
- * Copyright (C) 2007 Lemote Inc. & Insititute of Computing Technology
+ * Copyright (C) 2007 Lemote Inc. & Institute of Computing Technology
  * Author: Fuxin Zhang, zhangfx@lemote.com
  *
  * Copyright (C) 2009 Lemote Inc.
diff --git a/arch/mips/loongson64/common/cs5536/cs5536_mfgpt.c b/arch/mips/loongson64/common/cs5536/cs5536_mfgpt.c
index 12c75db23420..875037063a80 100644
--- a/arch/mips/loongson64/common/cs5536/cs5536_mfgpt.c
+++ b/arch/mips/loongson64/common/cs5536/cs5536_mfgpt.c
@@ -1,7 +1,7 @@
 /*
  * CS5536 General timer functions
  *
- * Copyright (C) 2007 Lemote Inc. & Insititute of Computing Technology
+ * Copyright (C) 2007 Lemote Inc. & Institute of Computing Technology
  * Author: Yanhua, yanh@lemote.com
  *
  * Copyright (C) 2009 Lemote Inc.
diff --git a/arch/mips/loongson64/common/env.c b/arch/mips/loongson64/common/env.c
index 22f04ca2ff3e..f6c44dd332e2 100644
--- a/arch/mips/loongson64/common/env.c
+++ b/arch/mips/loongson64/common/env.c
@@ -6,7 +6,7 @@
  * Copyright 2003 ICT CAS
  * Author: Michael Guo <guoyi@ict.ac.cn>
  *
- * Copyright (C) 2007 Lemote Inc. & Insititute of Computing Technology
+ * Copyright (C) 2007 Lemote Inc. & Institute of Computing Technology
  * Author: Fuxin Zhang, zhangfx@lemote.com
  *
  * Copyright (C) 2009 Lemote Inc.
diff --git a/arch/mips/loongson64/common/irq.c b/arch/mips/loongson64/common/irq.c
index 687003b19b45..d36d969a4a87 100644
--- a/arch/mips/loongson64/common/irq.c
+++ b/arch/mips/loongson64/common/irq.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 2007 Lemote Inc. & Insititute of Computing Technology
+ * Copyright (C) 2007 Lemote Inc. & Institute of Computing Technology
  * Author: Fuxin Zhang, zhangfx@lemote.com
  *
  *  This program is free software; you can redistribute	 it and/or modify it
diff --git a/arch/mips/loongson64/common/setup.c b/arch/mips/loongson64/common/setup.c
index d477dd6bb326..2dc5122f0e09 100644
--- a/arch/mips/loongson64/common/setup.c
+++ b/arch/mips/loongson64/common/setup.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 2007 Lemote Inc. & Insititute of Computing Technology
+ * Copyright (C) 2007 Lemote Inc. & Institute of Computing Technology
  * Author: Fuxin Zhang, zhangfx@lemote.com
  *
  *  This program is free software; you can redistribute	 it and/or modify it
diff --git a/arch/mips/loongson64/fuloong-2e/irq.c b/arch/mips/loongson64/fuloong-2e/irq.c
index ef5ec8f3de5f..892963f860b7 100644
--- a/arch/mips/loongson64/fuloong-2e/irq.c
+++ b/arch/mips/loongson64/fuloong-2e/irq.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 2007 Lemote Inc. & Insititute of Computing Technology
+ * Copyright (C) 2007 Lemote Inc. & Institute of Computing Technology
  * Author: Fuxin Zhang, zhangfx@lemote.com
  *
  *  This program is free software; you can redistribute	 it and/or modify it
diff --git a/arch/mips/loongson64/lemote-2f/clock.c b/arch/mips/loongson64/lemote-2f/clock.c
index 462e34d46b4a..a78fb657068c 100644
--- a/arch/mips/loongson64/lemote-2f/clock.c
+++ b/arch/mips/loongson64/lemote-2f/clock.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 2006 - 2008 Lemote Inc. & Insititute of Computing Technology
+ * Copyright (C) 2006 - 2008 Lemote Inc. & Institute of Computing Technology
  * Author: Yanhua, yanh@lemote.com
  *
  * This file is subject to the terms and conditions of the GNU General Public
@@ -15,7 +15,7 @@
 #include <linux/spinlock.h>
 
 #include <asm/clock.h>
-#include <asm/mach-loongson/loongson.h>
+#include <asm/mach-loongson64/loongson.h>
 
 static LIST_HEAD(clock_list);
 static DEFINE_SPINLOCK(clock_lock);
diff --git a/arch/mips/loongson64/loongson-3/numa.c b/arch/mips/loongson64/loongson-3/numa.c
index 12d14ed48778..6f9e010cec4d 100644
--- a/arch/mips/loongson64/loongson-3/numa.c
+++ b/arch/mips/loongson64/loongson-3/numa.c
@@ -1,6 +1,6 @@
 /*
  * Copyright (C) 2010 Loongson Inc. & Lemote Inc. &
- *                    Insititute of Computing Technology
+ *                    Institute of Computing Technology
  * Author:  Xiang Gao, gaoxiang@ict.ac.cn
  *          Huacai Chen, chenhc@lemote.com
  *          Xiaofu Meng, Shuangshuang Zhang
diff --git a/arch/mips/math-emu/cp1emu.c b/arch/mips/math-emu/cp1emu.c
index 22b9b2cb9219..712f17a2ecf2 100644
--- a/arch/mips/math-emu/cp1emu.c
+++ b/arch/mips/math-emu/cp1emu.c
@@ -451,7 +451,7 @@ static int isBranchInstr(struct pt_regs *regs, struct mm_decoded_insn dec_insn,
 			/* Fall through */
 		case jr_op:
 			/* For R6, JR already emulated in jalr_op */
-			if (NO_R6EMU && insn.r_format.opcode == jr_op)
+			if (NO_R6EMU && insn.r_format.func == jr_op)
 				break;
 			*contpc = regs->regs[insn.r_format.rs];
 			return 1;
@@ -551,7 +551,7 @@ static int isBranchInstr(struct pt_regs *regs, struct mm_decoded_insn dec_insn,
 				dec_insn.next_pc_inc;
 		return 1;
 	case blezl_op:
-		if (NO_R6EMU)
+		if (!insn.i_format.rt && NO_R6EMU)
 			break;
 	case blez_op:
 
@@ -588,7 +588,7 @@ static int isBranchInstr(struct pt_regs *regs, struct mm_decoded_insn dec_insn,
 				dec_insn.next_pc_inc;
 		return 1;
 	case bgtzl_op:
-		if (NO_R6EMU)
+		if (!insn.i_format.rt && NO_R6EMU)
 			break;
 	case bgtz_op:
 		/*
diff --git a/arch/mips/mm/c-r4k.c b/arch/mips/mm/c-r4k.c
index 7f660dc67596..fbea4432f3f2 100644
--- a/arch/mips/mm/c-r4k.c
+++ b/arch/mips/mm/c-r4k.c
@@ -37,6 +37,7 @@
 #include <asm/cacheflush.h> /* for run_uncached() */
 #include <asm/traps.h>
 #include <asm/dma-coherence.h>
+#include <asm/mips-cm.h>
 
 /*
  * Special Variant of smp_call_function for use by cache functions:
@@ -51,9 +52,16 @@ static inline void r4k_on_each_cpu(void (*func) (void *info), void *info)
 {
 	preempt_disable();
 
-#ifndef CONFIG_MIPS_MT_SMP
-	smp_call_function(func, info, 1);
-#endif
+	/*
+	 * The Coherent Manager propagates address-based cache ops to other
+	 * cores but not index-based ops. However, r4k_on_each_cpu is used
+	 * in both cases so there is no easy way to tell what kind of op is
+	 * executed to the other cores. The best we can probably do is
+	 * to restrict that call when a CM is not present because both
+	 * CM-based SMP protocols (CMP & CPS) restrict index-based cache ops.
+	 */
+	if (!mips_cm_present())
+		smp_call_function_many(&cpu_foreign_map, func, info, 1);
 	func(info);
 	preempt_enable();
 }
@@ -937,7 +945,9 @@ static void b5k_instruction_hazard(void)
 }
 
 static char *way_string[] = { NULL, "direct mapped", "2-way",
-	"3-way", "4-way", "5-way", "6-way", "7-way", "8-way"
+	"3-way", "4-way", "5-way", "6-way", "7-way", "8-way",
+	"9-way", "10-way", "11-way", "12-way",
+	"13-way", "14-way", "15-way", "16-way",
 };
 
 static void probe_pcache(void)
diff --git a/arch/mips/mti-malta/malta-time.c b/arch/mips/mti-malta/malta-time.c
index 185e68261f45..5625b190edc0 100644
--- a/arch/mips/mti-malta/malta-time.c
+++ b/arch/mips/mti-malta/malta-time.c
@@ -119,18 +119,24 @@ void read_persistent_clock(struct timespec *ts)
 
 int get_c0_fdc_int(void)
 {
-	int mips_cpu_fdc_irq;
+	/*
+	 * Some cores claim the FDC is routable through the GIC, but it doesn't
+	 * actually seem to be connected for those Malta bitstreams.
+	 */
+	switch (current_cpu_type()) {
+	case CPU_INTERAPTIV:
+	case CPU_PROAPTIV:
+		return -1;
+	};
 
 	if (cpu_has_veic)
-		mips_cpu_fdc_irq = -1;
+		return -1;
 	else if (gic_present)
-		mips_cpu_fdc_irq = gic_get_c0_fdc_int();
+		return gic_get_c0_fdc_int();
 	else if (cp0_fdc_irq >= 0)
-		mips_cpu_fdc_irq = MIPS_CPU_IRQ_BASE + cp0_fdc_irq;
+		return MIPS_CPU_IRQ_BASE + cp0_fdc_irq;
 	else
-		mips_cpu_fdc_irq = -1;
-
-	return mips_cpu_fdc_irq;
+		return -1;
 }
 
 int get_c0_perfcount_int(void)
diff --git a/arch/mips/pistachio/init.c b/arch/mips/pistachio/init.c
index d2dc836523a3..8bd8ebb20a72 100644
--- a/arch/mips/pistachio/init.c
+++ b/arch/mips/pistachio/init.c
@@ -63,13 +63,19 @@ void __init plat_mem_setup(void)
 	plat_setup_iocoherency();
 }
 
-#define DEFAULT_CPC_BASE_ADDR 0x1bde0000
+#define DEFAULT_CPC_BASE_ADDR	0x1bde0000
+#define DEFAULT_CDMM_BASE_ADDR	0x1bdd0000
 
 phys_addr_t mips_cpc_default_phys_base(void)
 {
 	return DEFAULT_CPC_BASE_ADDR;
 }
 
+phys_addr_t mips_cdmm_phys_base(void)
+{
+	return DEFAULT_CDMM_BASE_ADDR;
+}
+
 static void __init mips_nmi_setup(void)
 {
 	void *base;
diff --git a/arch/mips/pistachio/time.c b/arch/mips/pistachio/time.c
index 67889fcea8aa..7c73fcb92a10 100644
--- a/arch/mips/pistachio/time.c
+++ b/arch/mips/pistachio/time.c
@@ -27,6 +27,11 @@ int get_c0_perfcount_int(void)
 	return gic_get_c0_perfcount_int();
 }
 
+int get_c0_fdc_int(void)
+{
+	return gic_get_c0_fdc_int();
+}
+
 void __init plat_time_init(void)
 {
 	struct device_node *np;
diff --git a/drivers/cpufreq/loongson2_cpufreq.c b/drivers/cpufreq/loongson2_cpufreq.c
index fc897babab55..e362860c2b50 100644
--- a/drivers/cpufreq/loongson2_cpufreq.c
+++ b/drivers/cpufreq/loongson2_cpufreq.c
@@ -3,7 +3,7 @@
  *
  * The 2E revision of loongson processor not support this feature.
  *
- * Copyright (C) 2006 - 2008 Lemote Inc. & Insititute of Computing Technology
+ * Copyright (C) 2006 - 2008 Lemote Inc. & Institute of Computing Technology
  * Author: Yanhua, yanh@lemote.com
  *
  * This file is subject to the terms and conditions of the GNU General Public
diff --git a/drivers/irqchip/irq-mips-gic.c b/drivers/irqchip/irq-mips-gic.c
index 4400edd1a6c7..b7d54d428b5e 100644
--- a/drivers/irqchip/irq-mips-gic.c
+++ b/drivers/irqchip/irq-mips-gic.c
@@ -257,16 +257,6 @@ int gic_get_c0_fdc_int(void)
 		return MIPS_CPU_IRQ_BASE + cp0_fdc_irq;
 	}
 
-	/*
-	 * Some cores claim the FDC is routable but it doesn't actually seem to
-	 * be connected.
-	 */
-	switch (current_cpu_type()) {
-	case CPU_INTERAPTIV:
-	case CPU_PROAPTIV:
-		return -1;
-	}
-
 	return irq_create_mapping(gic_irq_domain,
 				  GIC_LOCAL_TO_HWIRQ(GIC_LOCAL_INT_FDC));
 }