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

Pull MIPS fixes from Ralf Baechle:
 "More 3.18 fixes for MIPS:

   - backtraces were not quite working on on 64-bit kernels
   - loongson needs a different cache coherency setting
   - Loongson 3 is a MIPS64 R2 version but due to erratum we treat is an
     older architecture revision.
   - fix build errors due to undefined references to __node_distances
     for certain configurations.
   - fix instruction decodig in the jump label code.
   - for certain configurations copy_{from,to}_user destroy the content
     of $3 so that register needs to be marked as clobbed by the calling
     code.
   - Hardware Table Walker fixes.
   - fill the delay slot of the last instruction of memcpy otherwise
     whatever ends up there randomly might have undesirable effects.
   - ensure get_user/__get_user always zero the variable to be read even
     in case of an error"

* 'upstream' of git://git.linux-mips.org/pub/scm/ralf/upstream-linus:
  MIPS: jump_label.c: Handle the microMIPS J instruction encoding
  MIPS: jump_label.c: Correct the span of the J instruction
  MIPS: Zero variable read by get_user / __get_user in case of an error.
  MIPS: lib: memcpy: Restore NOP on delay slot before returning to caller
  MIPS: tlb-r4k: Add missing HTW stop/start sequences
  MIPS: asm: uaccess: Add v1 register to clobber list on EVA
  MIPS: oprofile: Fix backtrace on 64-bit kernel
  MIPS: Loongson: Set Loongson-3's ISA level to MIPS64R1
  MIPS: Loongson: Fix the write-combine CCA value setting
  MIPS: IP27: Fix __node_distances undefined error
  MIPS: Loongson3: Fix __node_distances undefined error
diff --git a/arch/mips/include/asm/jump_label.h b/arch/mips/include/asm/jump_label.h
index e194f95..fdbff44 100644
--- a/arch/mips/include/asm/jump_label.h
+++ b/arch/mips/include/asm/jump_label.h
@@ -20,9 +20,15 @@
 #define WORD_INSN ".word"
 #endif
 
+#ifdef CONFIG_CPU_MICROMIPS
+#define NOP_INSN "nop32"
+#else
+#define NOP_INSN "nop"
+#endif
+
 static __always_inline bool arch_static_branch(struct static_key *key)
 {
-	asm_volatile_goto("1:\tnop\n\t"
+	asm_volatile_goto("1:\t" NOP_INSN "\n\t"
 		"nop\n\t"
 		".pushsection __jump_table,  \"aw\"\n\t"
 		WORD_INSN " 1b, %l[l_yes], %0\n\t"
diff --git a/arch/mips/include/asm/mach-loongson/cpu-feature-overrides.h b/arch/mips/include/asm/mach-loongson/cpu-feature-overrides.h
index 7d28f95..6d69332 100644
--- a/arch/mips/include/asm/mach-loongson/cpu-feature-overrides.h
+++ b/arch/mips/include/asm/mach-loongson/cpu-feature-overrides.h
@@ -41,10 +41,8 @@
 #define cpu_has_mcheck		0
 #define cpu_has_mdmx		0
 #define cpu_has_mips16		0
-#define cpu_has_mips32r1	0
 #define cpu_has_mips32r2	0
 #define cpu_has_mips3d		0
-#define cpu_has_mips64r1	0
 #define cpu_has_mips64r2	0
 #define cpu_has_mipsmt		0
 #define cpu_has_prefetch	0
diff --git a/arch/mips/include/asm/uaccess.h b/arch/mips/include/asm/uaccess.h
index a109510..133678a 100644
--- a/arch/mips/include/asm/uaccess.h
+++ b/arch/mips/include/asm/uaccess.h
@@ -301,7 +301,8 @@
 			__get_kernel_common((x), size, __gu_ptr);	\
 		else							\
 			__get_user_common((x), size, __gu_ptr);		\
-	}								\
+	} else								\
+		(x) = 0;						\
 									\
 	__gu_err;							\
 })
@@ -316,6 +317,7 @@
 	"	.insn						\n"	\
 	"	.section .fixup,\"ax\"				\n"	\
 	"3:	li	%0, %4					\n"	\
+	"	move	%1, $0					\n"	\
 	"	j	2b					\n"	\
 	"	.previous					\n"	\
 	"	.section __ex_table,\"a\"			\n"	\
@@ -630,6 +632,7 @@
 	"	.insn						\n"	\
 	"	.section .fixup,\"ax\"				\n"	\
 	"3:	li	%0, %4					\n"	\
+	"	move	%1, $0					\n"	\
 	"	j	2b					\n"	\
 	"	.previous					\n"	\
 	"	.section __ex_table,\"a\"			\n"	\
@@ -773,10 +776,11 @@
 	"jal\t" #destination "\n\t"
 #endif
 
-#ifndef CONFIG_CPU_DADDI_WORKAROUNDS
-#define DADDI_SCRATCH "$0"
-#else
+#if defined(CONFIG_CPU_DADDI_WORKAROUNDS) || (defined(CONFIG_EVA) &&	\
+					      defined(CONFIG_CPU_HAS_PREFETCH))
 #define DADDI_SCRATCH "$3"
+#else
+#define DADDI_SCRATCH "$0"
 #endif
 
 extern size_t __copy_user(void *__to, const void *__from, size_t __n);
diff --git a/arch/mips/kernel/cpu-probe.c b/arch/mips/kernel/cpu-probe.c
index 94c4a0c..d5a4f38 100644
--- a/arch/mips/kernel/cpu-probe.c
+++ b/arch/mips/kernel/cpu-probe.c
@@ -757,31 +757,34 @@
 			c->cputype = CPU_LOONGSON2;
 			__cpu_name[cpu] = "ICT Loongson-2";
 			set_elf_platform(cpu, "loongson2e");
+			set_isa(c, MIPS_CPU_ISA_III);
 			break;
 		case PRID_REV_LOONGSON2F:
 			c->cputype = CPU_LOONGSON2;
 			__cpu_name[cpu] = "ICT Loongson-2";
 			set_elf_platform(cpu, "loongson2f");
+			set_isa(c, MIPS_CPU_ISA_III);
 			break;
 		case PRID_REV_LOONGSON3A:
 			c->cputype = CPU_LOONGSON3;
-			c->writecombine = _CACHE_UNCACHED_ACCELERATED;
 			__cpu_name[cpu] = "ICT Loongson-3";
 			set_elf_platform(cpu, "loongson3a");
+			set_isa(c, MIPS_CPU_ISA_M64R1);
 			break;
 		case PRID_REV_LOONGSON3B_R1:
 		case PRID_REV_LOONGSON3B_R2:
 			c->cputype = CPU_LOONGSON3;
 			__cpu_name[cpu] = "ICT Loongson-3";
 			set_elf_platform(cpu, "loongson3b");
+			set_isa(c, MIPS_CPU_ISA_M64R1);
 			break;
 		}
 
-		set_isa(c, MIPS_CPU_ISA_III);
 		c->options = R4K_OPTS |
 			     MIPS_CPU_FPU | MIPS_CPU_LLSC |
 			     MIPS_CPU_32FPR;
 		c->tlbsize = 64;
+		c->writecombine = _CACHE_UNCACHED_ACCELERATED;
 		break;
 	case PRID_IMP_LOONGSON_32:  /* Loongson-1 */
 		decode_configs(c);
diff --git a/arch/mips/kernel/jump_label.c b/arch/mips/kernel/jump_label.c
index 6001610..dda800e 100644
--- a/arch/mips/kernel/jump_label.c
+++ b/arch/mips/kernel/jump_label.c
@@ -18,31 +18,53 @@
 
 #ifdef HAVE_JUMP_LABEL
 
-#define J_RANGE_MASK ((1ul << 28) - 1)
+/*
+ * Define parameters for the standard MIPS and the microMIPS jump
+ * instruction encoding respectively:
+ *
+ * - the ISA bit of the target, either 0 or 1 respectively,
+ *
+ * - the amount the jump target address is shifted right to fit in the
+ *   immediate field of the machine instruction, either 2 or 1,
+ *
+ * - the mask determining the size of the jump region relative to the
+ *   delay-slot instruction, either 256MB or 128MB,
+ *
+ * - the jump target alignment, either 4 or 2 bytes.
+ */
+#define J_ISA_BIT	IS_ENABLED(CONFIG_CPU_MICROMIPS)
+#define J_RANGE_SHIFT	(2 - J_ISA_BIT)
+#define J_RANGE_MASK	((1ul << (26 + J_RANGE_SHIFT)) - 1)
+#define J_ALIGN_MASK	((1ul << J_RANGE_SHIFT) - 1)
 
 void arch_jump_label_transform(struct jump_entry *e,
 			       enum jump_label_type type)
 {
+	union mips_instruction *insn_p;
 	union mips_instruction insn;
-	union mips_instruction *insn_p =
-		(union mips_instruction *)(unsigned long)e->code;
 
-	/* Jump only works within a 256MB aligned region. */
-	BUG_ON((e->target & ~J_RANGE_MASK) != (e->code & ~J_RANGE_MASK));
+	insn_p = (union mips_instruction *)msk_isa16_mode(e->code);
 
-	/* Target must have 4 byte alignment. */
-	BUG_ON((e->target & 3) != 0);
+	/* Jump only works within an aligned region its delay slot is in. */
+	BUG_ON((e->target & ~J_RANGE_MASK) != ((e->code + 4) & ~J_RANGE_MASK));
+
+	/* Target must have the right alignment and ISA must be preserved. */
+	BUG_ON((e->target & J_ALIGN_MASK) != J_ISA_BIT);
 
 	if (type == JUMP_LABEL_ENABLE) {
-		insn.j_format.opcode = j_op;
-		insn.j_format.target = (e->target & J_RANGE_MASK) >> 2;
+		insn.j_format.opcode = J_ISA_BIT ? mm_j32_op : j_op;
+		insn.j_format.target = e->target >> J_RANGE_SHIFT;
 	} else {
 		insn.word = 0; /* nop */
 	}
 
 	get_online_cpus();
 	mutex_lock(&text_mutex);
-	*insn_p = insn;
+	if (IS_ENABLED(CONFIG_CPU_MICROMIPS)) {
+		insn_p->halfword[0] = insn.word >> 16;
+		insn_p->halfword[1] = insn.word;
+	} else
+		*insn_p = insn;
 
 	flush_icache_range((unsigned long)insn_p,
 			   (unsigned long)insn_p + sizeof(*insn_p));
diff --git a/arch/mips/lib/memcpy.S b/arch/mips/lib/memcpy.S
index c17ef80..5d3238a 100644
--- a/arch/mips/lib/memcpy.S
+++ b/arch/mips/lib/memcpy.S
@@ -503,6 +503,7 @@
 	STOREB(t0, NBYTES-2(dst), .Ls_exc_p1\@)
 .Ldone\@:
 	jr	ra
+	 nop
 	.if __memcpy == 1
 	END(memcpy)
 	.set __memcpy, 0
diff --git a/arch/mips/loongson/loongson-3/numa.c b/arch/mips/loongson/loongson-3/numa.c
index 37ed184..42323bc 100644
--- a/arch/mips/loongson/loongson-3/numa.c
+++ b/arch/mips/loongson/loongson-3/numa.c
@@ -33,6 +33,7 @@
 
 static struct node_data prealloc__node_data[MAX_NUMNODES];
 unsigned char __node_distances[MAX_NUMNODES][MAX_NUMNODES];
+EXPORT_SYMBOL(__node_distances);
 struct node_data *__node_data[MAX_NUMNODES];
 EXPORT_SYMBOL(__node_data);
 
diff --git a/arch/mips/mm/tlb-r4k.c b/arch/mips/mm/tlb-r4k.c
index fa6ebd4..c3917e2 100644
--- a/arch/mips/mm/tlb-r4k.c
+++ b/arch/mips/mm/tlb-r4k.c
@@ -299,6 +299,7 @@
 
 	local_irq_save(flags);
 
+	htw_stop();
 	pid = read_c0_entryhi() & ASID_MASK;
 	address &= (PAGE_MASK << 1);
 	write_c0_entryhi(address | pid);
@@ -346,6 +347,7 @@
 			tlb_write_indexed();
 	}
 	tlbw_use_hazard();
+	htw_start();
 	flush_itlb_vm(vma);
 	local_irq_restore(flags);
 }
@@ -422,6 +424,7 @@
 
 	local_irq_save(flags);
 	/* Save old context and create impossible VPN2 value */
+	htw_stop();
 	old_ctx = read_c0_entryhi();
 	old_pagemask = read_c0_pagemask();
 	wired = read_c0_wired();
@@ -443,6 +446,7 @@
 
 	write_c0_entryhi(old_ctx);
 	write_c0_pagemask(old_pagemask);
+	htw_start();
 out:
 	local_irq_restore(flags);
 	return ret;
diff --git a/arch/mips/oprofile/backtrace.c b/arch/mips/oprofile/backtrace.c
index 6854ed5..83a1dfd 100644
--- a/arch/mips/oprofile/backtrace.c
+++ b/arch/mips/oprofile/backtrace.c
@@ -92,7 +92,7 @@
 				/* This marks the end of the previous function,
 				   which means we overran. */
 				break;
-			stack_size = (unsigned) stack_adjustment;
+			stack_size = (unsigned long) stack_adjustment;
 		} else if (is_ra_save_ins(&ip)) {
 			int ra_slot = ip.i_format.simmediate;
 			if (ra_slot < 0)
diff --git a/arch/mips/sgi-ip27/ip27-memory.c b/arch/mips/sgi-ip27/ip27-memory.c
index a95c00f..a304bcc 100644
--- a/arch/mips/sgi-ip27/ip27-memory.c
+++ b/arch/mips/sgi-ip27/ip27-memory.c
@@ -107,6 +107,7 @@
 }
 
 unsigned char __node_distances[MAX_COMPACT_NODES][MAX_COMPACT_NODES];
+EXPORT_SYMBOL(__node_distances);
 
 static int __init compute_node_distance(nasid_t nasid_a, nasid_t nasid_b)
 {