Merge branch '4.14-features' of git://git.linux-mips.org/pub/scm/ralf/upstream-linus

Pull MIPS updates from Ralf Baechle:
 "This is the main pull request for 4.14 for MIPS; below a summary of
  the non-merge commits:

  CM:
   - Rename mips_cm_base to mips_gcr_base
   - Specify register size when generating accessors
   - Use BIT/GENMASK for register fields, order & drop shifts
   - Add cluster & block args to mips_cm_lock_other()

  CPC:
   - Use common CPS accessor generation macros
   - Use BIT/GENMASK for register fields, order & drop shifts
   - Introduce register modify (set/clear/change) accessors
   - Use change_*, set_* & clear_* where appropriate
   - Add CM/CPC 3.5 register definitions
   - Use GlobalNumber macros rather than magic numbers
   - Have asm/mips-cps.h include CM & CPC headers
   - Cluster support for topology functions
   - Detect CPUs in secondary clusters

  CPS:
   - Read GIC_VL_IDENT directly, not via irqchip driver

  DMA:
   - Consolidate coherent and non-coherent dma_alloc code
   - Don't use dma_cache_sync to implement fd_cacheflush

  FPU emulation / FP assist code:
   - Another series of 14 commits fixing corner cases such as NaN
     propgagation and other special input values.
   - Zero bits 32-63 of the result for a CLASS.D instruction.
   - Enhanced statics via debugfs
   - Do not use bools for arithmetic. GCC 7.1 moans about this.
   - Correct user fault_addr type

  Generic MIPS:
   - Enhancement of stack backtraces
   - Cleanup from non-existing options
   - Handle non word sized instructions when examining frame
   - Fix detection and decoding of ADDIUSP instruction
   - Fix decoding of SWSP16 instruction
   - Refactor handling of stack pointer in get_frame_info
   - Remove unreachable code from force_fcr31_sig()
   - Convert to using %pOF instead of full_name
   - Remove the R6000 support.
   - Move FP code from *_switch.S to *_fpu.S
   - Remove unused ST_OFF from r2300_switch.S
   - Allow platform to specify multiple its.S files
   - Add #includes to various files to ensure code builds reliable and
     without warning..
   - Remove __invalidate_kernel_vmap_range
   - Remove plat_timer_setup
   - Declare various variables & functions static
   - Abstract CPU core & VP(E) ID access through accessor functions
   - Store core & VP IDs in GlobalNumber-style variable
   - Unify checks for sibling CPUs
   - Add CPU cluster number accessors
   - Prevent direct use of generic_defconfig
   - Make CONFIG_MIPS_MT_SMP default y
   - Add __ioread64_copy
   - Remove unnecessary inclusions of linux/irqchip/mips-gic.h

  GIC:
   - Introduce asm/mips-gic.h with accessor functions
   - Use new GIC accessor functions in mips-gic-timer
   - Remove counter access functions from irq-mips-gic.c
   - Remove gic_read_local_vp_id() from irq-mips-gic.c
   - Simplify shared interrupt pending/mask reads in irq-mips-gic.c
   - Simplify gic_local_irq_domain_map() in irq-mips-gic.c
   - Drop gic_(re)set_mask() functions in irq-mips-gic.c
   - Remove gic_set_polarity(), gic_set_trigger(), gic_set_dual_edge(),
     gic_map_to_pin() and gic_map_to_vpe() from irq-mips-gic.c.
   - Convert remaining shared reg access, local int mask access and
     remaining local reg access to new accessors
   - Move GIC_LOCAL_INT_* to asm/mips-gic.h
   - Remove GIC_CPU_INT* macros from irq-mips-gic.c
   - Move various definitions to the driver
   - Remove gic_get_usm_range()
   - Remove __gic_irq_dispatch() forward declaration
   - Remove gic_init()
   - Use mips_gic_present() in place of gic_present and remove
     gic_present
   - Move gic_get_c0_*_int() to asm/mips-gic.h
   - Remove linux/irqchip/mips-gic.h
   - Inline __gic_init()
   - Inline gic_basic_init()
   - Make pcpu_masks a per-cpu variable
   - Use pcpu_masks to avoid reading GIC_SH_MASK*
   - Clean up mti, reserved-cpu-vectors handling
   - Use cpumask_first_and() in gic_set_affinity()
   - Let the core set struct irq_common_data affinity

  microMIPS:
   - Fix microMIPS stack unwinding on big endian systems

  MIPS-GIC:
   - SYNC after enabling GIC region

  NUMA:
   - Remove the unused parent_node() macro

  R6:
   - Constify r2_decoder_tables
   - Add accessor & bit definitions for GlobalNumber

  SMP:
   - Constify smp ops
   - Allow boot_secondary SMP op to return errors

  VDSO:
   - Drop gic_get_usm_range() usage
   - Avoid use of linux/irqchip/mips-gic.h

  Platform changes:

  Alchemy:
   - Add devboard machine type to cpuinfo
   - update cpu feature overrides
   - Threaded carddetect irqs for devboards

  AR7:
   - allow NULL clock for clk_get_rate

  BCM63xx:
   - Fix ENETDMA_6345_MAXBURST_REG offset
   - Allow NULL clock for clk_get_rate

  CI20:
   - Enable GPIO and RTC drivers in defconfig
   - Add ethernet and fixed-regulator nodes to DTS

  Generic platform:
   - Move Boston and NI 169445 FIT image source to their own files
   - Include asm/bootinfo.h for plat_fdt_relocated()
   - Include asm/time.h for get_c0_*_int()
   - Include asm/bootinfo.h for plat_fdt_relocated()
   - Include asm/time.h for get_c0_*_int()
   - Allow filtering enabled boards by requirements
   - Don't explicitly disable CONFIG_USB_SUPPORT
   - Bump default NR_CPUS to 16

  JZ4700:
   - Probe the jz4740-rtc driver from devicetree

  Lantiq:
   - Drop check of boot select from the spi-falcon driver.
   - Drop check of boot select from the lantiq-flash MTD driver.
   - Access boot cause register in the watchdog driver through regmap
   - Add device tree binding documentation for the watchdog driver
   - Add docs for the RCU DT bindings.
   - Convert the fpi bus driver to a platform_driver
   - Remove ltq_reset_cause() and ltq_boot_select(
   - Switch to a proper reset driver
   - Switch to a new drivers/soc GPHY driver
   - Add an USB PHY driver for the Lantiq SoCs using the RCU module
   - Use of_platform_default_populate instead of __dt_register_buses
   - Enable MFD_SYSCON to be able to use it for the RCU MFD
   - Replace ltq_boot_select() with dummy implementation.

  Loongson 2F:
   - Allow NULL clock for clk_get_rate

  Malta:
   - Use new GIC accessor functions

  NI 169445:
   - Add support for NI 169445 board.
   - Only include in 32r2el kernels

  Octeon:
   - Add support for watchdog of 78XX SOCs.
   - Add support for watchdog of CN68XX SOCs.
   - Expose support for mips32r1, mips32r2 and mips64r1
   - Enable more drivers in config file
   - Add support for accessing the boot vector.
   - Remove old boot vector code from watchdog driver
   - Define watchdog registers for 70xx, 73xx, 78xx, F75xx.
   - Make CSR functions node aware.
   - Allow access to CIU3 IRQ domains.
   - Misc cleanups in the watchdog driver

  Omega2+:
   - New board, add support and defconfig

  Pistachio:
   - Enable Root FS on NFS in defconfig

  Ralink:
   - Add Mediatek MT7628A SoC
   - Allow NULL clock for clk_get_rate
   - Explicitly request exclusive reset control in the pci-mt7620 PCI driver.

  SEAD3:
   - Only include in 32 bit kernels by default

  VoCore:
   - Add VoCore as a vendor t0 dt-bindings
   - Add defconfig file"

* '4.14-features' of git://git.linux-mips.org/pub/scm/ralf/upstream-linus: (167 commits)
  MIPS: Refactor handling of stack pointer in get_frame_info
  MIPS: Stacktrace: Fix microMIPS stack unwinding on big endian systems
  MIPS: microMIPS: Fix decoding of swsp16 instruction
  MIPS: microMIPS: Fix decoding of addiusp instruction
  MIPS: microMIPS: Fix detection of addiusp instruction
  MIPS: Handle non word sized instructions when examining frame
  MIPS: ralink: allow NULL clock for clk_get_rate
  MIPS: Loongson 2F: allow NULL clock for clk_get_rate
  MIPS: BCM63XX: allow NULL clock for clk_get_rate
  MIPS: AR7: allow NULL clock for clk_get_rate
  MIPS: BCM63XX: fix ENETDMA_6345_MAXBURST_REG offset
  mips: Save all registers when saving the frame
  MIPS: Add DWARF unwinding to assembly
  MIPS: Make SAVE_SOME more standard
  MIPS: Fix issues in backtraces
  MIPS: jz4780: DTS: Probe the jz4740-rtc driver from devicetree
  MIPS: Ci20: Enable RTC driver
  watchdog: octeon-wdt: Add support for 78XX SOCs.
  watchdog: octeon-wdt: Add support for cn68XX SOCs.
  watchdog: octeon-wdt: File cleaning.
  ...

1 files changed, 272 insertions, 12 deletions

diff --git a/arch/mips/math-emu/cp1emu.c b/arch/mips/math-emu/cp1emu.c
index f08a7b4facb9..192542dbd972 100644
--- a/arch/mips/math-emu/cp1emu.c
+++ b/arch/mips/math-emu/cp1emu.c
@@ -58,7 +58,7 @@ static int fpu_emu(struct pt_regs *, struct mips_fpu_struct *,
 	mips_instruction);
 
 static int fpux_emu(struct pt_regs *,
-	struct mips_fpu_struct *, mips_instruction, void *__user *);
+	struct mips_fpu_struct *, mips_instruction, void __user **);
 
 /* Control registers */
 
@@ -830,12 +830,12 @@ do {									\
 } while (0)
 
 #define DIFROMREG(di, x)						\
-	((di) = get_fpr64(&ctx->fpr[(x) & ~(cop1_64bit(xcp) == 0)], 0))
+	((di) = get_fpr64(&ctx->fpr[(x) & ~(cop1_64bit(xcp) ^ 1)], 0))
 
 #define DITOREG(di, x)							\
 do {									\
 	unsigned fpr, i;						\
-	fpr = (x) & ~(cop1_64bit(xcp) == 0);				\
+	fpr = (x) & ~(cop1_64bit(xcp) ^ 1);				\
 	set_fpr64(&ctx->fpr[fpr], 0, di);				\
 	for (i = 1; i < ARRAY_SIZE(ctx->fpr[x].val64); i++)		\
 		set_fpr64(&ctx->fpr[fpr], i, 0);			\
@@ -973,7 +973,7 @@ static inline void cop1_ctc(struct pt_regs *xcp, struct mips_fpu_struct *ctx,
  */
 
 static int cop1Emulate(struct pt_regs *xcp, struct mips_fpu_struct *ctx,
-		struct mm_decoded_insn dec_insn, void *__user *fault_addr)
+		struct mm_decoded_insn dec_insn, void __user **fault_addr)
 {
 	unsigned long contpc = xcp->cp0_epc + dec_insn.pc_inc;
 	unsigned int cond, cbit, bit0;
@@ -1195,9 +1195,11 @@ emul:
 			bit0 = get_fpr32(fpr, 0) & 0x1;
 			switch (MIPSInst_RS(ir)) {
 			case bc1eqz_op:
+				MIPS_FPU_EMU_INC_STATS(bc1eqz);
 				cond = bit0 == 0;
 				break;
 			case bc1nez_op:
+				MIPS_FPU_EMU_INC_STATS(bc1nez);
 				cond = bit0 != 0;
 				break;
 			}
@@ -1230,6 +1232,7 @@ emul:
 				break;
 			}
 branch_common:
+			MIPS_FPU_EMU_INC_STATS(branches);
 			set_delay_slot(xcp);
 			if (cond) {
 				/*
@@ -1460,7 +1463,7 @@ DEF3OP(nmadd, dp, ieee754dp_mul, ieee754dp_add, ieee754dp_neg);
 DEF3OP(nmsub, dp, ieee754dp_mul, ieee754dp_sub, ieee754dp_neg);
 
 static int fpux_emu(struct pt_regs *xcp, struct mips_fpu_struct *ctx,
-	mips_instruction ir, void *__user *fault_addr)
+	mips_instruction ir, void __user **fault_addr)
 {
 	unsigned rcsr = 0;	/* resulting csr */
 
@@ -1682,15 +1685,19 @@ static int fpu_emu(struct pt_regs *xcp, struct mips_fpu_struct *ctx,
 		switch (MIPSInst_FUNC(ir)) {
 			/* binary ops */
 		case fadd_op:
+			MIPS_FPU_EMU_INC_STATS(add_s);
 			handler.b = ieee754sp_add;
 			goto scopbop;
 		case fsub_op:
+			MIPS_FPU_EMU_INC_STATS(sub_s);
 			handler.b = ieee754sp_sub;
 			goto scopbop;
 		case fmul_op:
+			MIPS_FPU_EMU_INC_STATS(mul_s);
 			handler.b = ieee754sp_mul;
 			goto scopbop;
 		case fdiv_op:
+			MIPS_FPU_EMU_INC_STATS(div_s);
 			handler.b = ieee754sp_div;
 			goto scopbop;
 
@@ -1699,6 +1706,7 @@ static int fpu_emu(struct pt_regs *xcp, struct mips_fpu_struct *ctx,
 			if (!cpu_has_mips_2_3_4_5_r)
 				return SIGILL;
 
+			MIPS_FPU_EMU_INC_STATS(sqrt_s);
 			handler.u = ieee754sp_sqrt;
 			goto scopuop;
 
@@ -1711,6 +1719,7 @@ static int fpu_emu(struct pt_regs *xcp, struct mips_fpu_struct *ctx,
 			if (!cpu_has_mips_4_5_64_r2_r6)
 				return SIGILL;
 
+			MIPS_FPU_EMU_INC_STATS(rsqrt_s);
 			handler.u = fpemu_sp_rsqrt;
 			goto scopuop;
 
@@ -1718,6 +1727,7 @@ static int fpu_emu(struct pt_regs *xcp, struct mips_fpu_struct *ctx,
 			if (!cpu_has_mips_4_5_64_r2_r6)
 				return SIGILL;
 
+			MIPS_FPU_EMU_INC_STATS(recip_s);
 			handler.u = fpemu_sp_recip;
 			goto scopuop;
 
@@ -1754,6 +1764,7 @@ static int fpu_emu(struct pt_regs *xcp, struct mips_fpu_struct *ctx,
 			if (!cpu_has_mips_r6)
 				return SIGILL;
 
+			MIPS_FPU_EMU_INC_STATS(seleqz_s);
 			SPFROMREG(rv.s, MIPSInst_FT(ir));
 			if (rv.w & 0x1)
 				rv.w = 0;
@@ -1765,6 +1776,7 @@ static int fpu_emu(struct pt_regs *xcp, struct mips_fpu_struct *ctx,
 			if (!cpu_has_mips_r6)
 				return SIGILL;
 
+			MIPS_FPU_EMU_INC_STATS(selnez_s);
 			SPFROMREG(rv.s, MIPSInst_FT(ir));
 			if (rv.w & 0x1)
 				SPFROMREG(rv.s, MIPSInst_FS(ir));
@@ -1778,6 +1790,7 @@ static int fpu_emu(struct pt_regs *xcp, struct mips_fpu_struct *ctx,
 			if (!cpu_has_mips_r6)
 				return SIGILL;
 
+			MIPS_FPU_EMU_INC_STATS(maddf_s);
 			SPFROMREG(ft, MIPSInst_FT(ir));
 			SPFROMREG(fs, MIPSInst_FS(ir));
 			SPFROMREG(fd, MIPSInst_FD(ir));
@@ -1791,6 +1804,7 @@ static int fpu_emu(struct pt_regs *xcp, struct mips_fpu_struct *ctx,
 			if (!cpu_has_mips_r6)
 				return SIGILL;
 
+			MIPS_FPU_EMU_INC_STATS(msubf_s);
 			SPFROMREG(ft, MIPSInst_FT(ir));
 			SPFROMREG(fs, MIPSInst_FS(ir));
 			SPFROMREG(fd, MIPSInst_FD(ir));
@@ -1804,9 +1818,9 @@ static int fpu_emu(struct pt_regs *xcp, struct mips_fpu_struct *ctx,
 			if (!cpu_has_mips_r6)
 				return SIGILL;
 
+			MIPS_FPU_EMU_INC_STATS(rint_s);
 			SPFROMREG(fs, MIPSInst_FS(ir));
-			rv.l = ieee754sp_tlong(fs);
-			rv.s = ieee754sp_flong(rv.l);
+			rv.s = ieee754sp_rint(fs);
 			goto copcsr;
 		}
 
@@ -1816,6 +1830,7 @@ static int fpu_emu(struct pt_regs *xcp, struct mips_fpu_struct *ctx,
 			if (!cpu_has_mips_r6)
 				return SIGILL;
 
+			MIPS_FPU_EMU_INC_STATS(class_s);
 			SPFROMREG(fs, MIPSInst_FS(ir));
 			rv.w = ieee754sp_2008class(fs);
 			rfmt = w_fmt;
@@ -1828,6 +1843,7 @@ static int fpu_emu(struct pt_regs *xcp, struct mips_fpu_struct *ctx,
 			if (!cpu_has_mips_r6)
 				return SIGILL;
 
+			MIPS_FPU_EMU_INC_STATS(min_s);
 			SPFROMREG(ft, MIPSInst_FT(ir));
 			SPFROMREG(fs, MIPSInst_FS(ir));
 			rv.s = ieee754sp_fmin(fs, ft);
@@ -1840,6 +1856,7 @@ static int fpu_emu(struct pt_regs *xcp, struct mips_fpu_struct *ctx,
 			if (!cpu_has_mips_r6)
 				return SIGILL;
 
+			MIPS_FPU_EMU_INC_STATS(mina_s);
 			SPFROMREG(ft, MIPSInst_FT(ir));
 			SPFROMREG(fs, MIPSInst_FS(ir));
 			rv.s = ieee754sp_fmina(fs, ft);
@@ -1852,6 +1869,7 @@ static int fpu_emu(struct pt_regs *xcp, struct mips_fpu_struct *ctx,
 			if (!cpu_has_mips_r6)
 				return SIGILL;
 
+			MIPS_FPU_EMU_INC_STATS(max_s);
 			SPFROMREG(ft, MIPSInst_FT(ir));
 			SPFROMREG(fs, MIPSInst_FS(ir));
 			rv.s = ieee754sp_fmax(fs, ft);
@@ -1864,6 +1882,7 @@ static int fpu_emu(struct pt_regs *xcp, struct mips_fpu_struct *ctx,
 			if (!cpu_has_mips_r6)
 				return SIGILL;
 
+			MIPS_FPU_EMU_INC_STATS(maxa_s);
 			SPFROMREG(ft, MIPSInst_FT(ir));
 			SPFROMREG(fs, MIPSInst_FS(ir));
 			rv.s = ieee754sp_fmaxa(fs, ft);
@@ -1871,15 +1890,18 @@ static int fpu_emu(struct pt_regs *xcp, struct mips_fpu_struct *ctx,
 		}
 
 		case fabs_op:
+			MIPS_FPU_EMU_INC_STATS(abs_s);
 			handler.u = ieee754sp_abs;
 			goto scopuop;
 
 		case fneg_op:
+			MIPS_FPU_EMU_INC_STATS(neg_s);
 			handler.u = ieee754sp_neg;
 			goto scopuop;
 
 		case fmov_op:
 			/* an easy one */
+			MIPS_FPU_EMU_INC_STATS(mov_s);
 			SPFROMREG(rv.s, MIPSInst_FS(ir));
 			goto copcsr;
 
@@ -1922,12 +1944,14 @@ copcsr:
 			return SIGILL;	/* not defined */
 
 		case fcvtd_op:
+			MIPS_FPU_EMU_INC_STATS(cvt_d_s);
 			SPFROMREG(fs, MIPSInst_FS(ir));
 			rv.d = ieee754dp_fsp(fs);
 			rfmt = d_fmt;
 			goto copcsr;
 
 		case fcvtw_op:
+			MIPS_FPU_EMU_INC_STATS(cvt_w_s);
 			SPFROMREG(fs, MIPSInst_FS(ir));
 			rv.w = ieee754sp_tint(fs);
 			rfmt = w_fmt;
@@ -1940,6 +1964,15 @@ copcsr:
 			if (!cpu_has_mips_2_3_4_5_r)
 				return SIGILL;
 
+			if (MIPSInst_FUNC(ir) == fceil_op)
+				MIPS_FPU_EMU_INC_STATS(ceil_w_s);
+			if (MIPSInst_FUNC(ir) == ffloor_op)
+				MIPS_FPU_EMU_INC_STATS(floor_w_s);
+			if (MIPSInst_FUNC(ir) == fround_op)
+				MIPS_FPU_EMU_INC_STATS(round_w_s);
+			if (MIPSInst_FUNC(ir) == ftrunc_op)
+				MIPS_FPU_EMU_INC_STATS(trunc_w_s);
+
 			oldrm = ieee754_csr.rm;
 			SPFROMREG(fs, MIPSInst_FS(ir));
 			ieee754_csr.rm = MIPSInst_FUNC(ir);
@@ -1952,6 +1985,7 @@ copcsr:
 			if (!cpu_has_mips_r6)
 				return SIGILL;
 
+			MIPS_FPU_EMU_INC_STATS(sel_s);
 			SPFROMREG(fd, MIPSInst_FD(ir));
 			if (fd.bits & 0x1)
 				SPFROMREG(rv.s, MIPSInst_FT(ir));
@@ -1963,6 +1997,7 @@ copcsr:
 			if (!cpu_has_mips_3_4_5_64_r2_r6)
 				return SIGILL;
 
+			MIPS_FPU_EMU_INC_STATS(cvt_l_s);
 			SPFROMREG(fs, MIPSInst_FS(ir));
 			rv.l = ieee754sp_tlong(fs);
 			rfmt = l_fmt;
@@ -1975,6 +2010,15 @@ copcsr:
 			if (!cpu_has_mips_3_4_5_64_r2_r6)
 				return SIGILL;
 
+			if (MIPSInst_FUNC(ir) == fceill_op)
+				MIPS_FPU_EMU_INC_STATS(ceil_l_s);
+			if (MIPSInst_FUNC(ir) == ffloorl_op)
+				MIPS_FPU_EMU_INC_STATS(floor_l_s);
+			if (MIPSInst_FUNC(ir) == froundl_op)
+				MIPS_FPU_EMU_INC_STATS(round_l_s);
+			if (MIPSInst_FUNC(ir) == ftruncl_op)
+				MIPS_FPU_EMU_INC_STATS(trunc_l_s);
+
 			oldrm = ieee754_csr.rm;
 			SPFROMREG(fs, MIPSInst_FS(ir));
 			ieee754_csr.rm = MIPSInst_FUNC(ir);
@@ -2016,15 +2060,19 @@ copcsr:
 		switch (MIPSInst_FUNC(ir)) {
 			/* binary ops */
 		case fadd_op:
+			MIPS_FPU_EMU_INC_STATS(add_d);
 			handler.b = ieee754dp_add;
 			goto dcopbop;
 		case fsub_op:
+			MIPS_FPU_EMU_INC_STATS(sub_d);
 			handler.b = ieee754dp_sub;
 			goto dcopbop;
 		case fmul_op:
+			MIPS_FPU_EMU_INC_STATS(mul_d);
 			handler.b = ieee754dp_mul;
 			goto dcopbop;
 		case fdiv_op:
+			MIPS_FPU_EMU_INC_STATS(div_d);
 			handler.b = ieee754dp_div;
 			goto dcopbop;
 
@@ -2033,6 +2081,7 @@ copcsr:
 			if (!cpu_has_mips_2_3_4_5_r)
 				return SIGILL;
 
+			MIPS_FPU_EMU_INC_STATS(sqrt_d);
 			handler.u = ieee754dp_sqrt;
 			goto dcopuop;
 		/*
@@ -2044,12 +2093,14 @@ copcsr:
 			if (!cpu_has_mips_4_5_64_r2_r6)
 				return SIGILL;
 
+			MIPS_FPU_EMU_INC_STATS(rsqrt_d);
 			handler.u = fpemu_dp_rsqrt;
 			goto dcopuop;
 		case frecip_op:
 			if (!cpu_has_mips_4_5_64_r2_r6)
 				return SIGILL;
 
+			MIPS_FPU_EMU_INC_STATS(recip_d);
 			handler.u = fpemu_dp_recip;
 			goto dcopuop;
 		case fmovc_op:
@@ -2083,6 +2134,7 @@ copcsr:
 			if (!cpu_has_mips_r6)
 				return SIGILL;
 
+			MIPS_FPU_EMU_INC_STATS(seleqz_d);
 			DPFROMREG(rv.d, MIPSInst_FT(ir));
 			if (rv.l & 0x1)
 				rv.l = 0;
@@ -2094,6 +2146,7 @@ copcsr:
 			if (!cpu_has_mips_r6)
 				return SIGILL;
 
+			MIPS_FPU_EMU_INC_STATS(selnez_d);
 			DPFROMREG(rv.d, MIPSInst_FT(ir));
 			if (rv.l & 0x1)
 				DPFROMREG(rv.d, MIPSInst_FS(ir));
@@ -2107,6 +2160,7 @@ copcsr:
 			if (!cpu_has_mips_r6)
 				return SIGILL;
 
+			MIPS_FPU_EMU_INC_STATS(maddf_d);
 			DPFROMREG(ft, MIPSInst_FT(ir));
 			DPFROMREG(fs, MIPSInst_FS(ir));
 			DPFROMREG(fd, MIPSInst_FD(ir));
@@ -2120,6 +2174,7 @@ copcsr:
 			if (!cpu_has_mips_r6)
 				return SIGILL;
 
+			MIPS_FPU_EMU_INC_STATS(msubf_d);
 			DPFROMREG(ft, MIPSInst_FT(ir));
 			DPFROMREG(fs, MIPSInst_FS(ir));
 			DPFROMREG(fd, MIPSInst_FD(ir));
@@ -2133,9 +2188,9 @@ copcsr:
 			if (!cpu_has_mips_r6)
 				return SIGILL;
 
+			MIPS_FPU_EMU_INC_STATS(rint_d);
 			DPFROMREG(fs, MIPSInst_FS(ir));
-			rv.l = ieee754dp_tlong(fs);
-			rv.d = ieee754dp_flong(rv.l);
+			rv.d = ieee754dp_rint(fs);
 			goto copcsr;
 		}
 
@@ -2145,9 +2200,10 @@ copcsr:
 			if (!cpu_has_mips_r6)
 				return SIGILL;
 
+			MIPS_FPU_EMU_INC_STATS(class_d);
 			DPFROMREG(fs, MIPSInst_FS(ir));
-			rv.w = ieee754dp_2008class(fs);
-			rfmt = w_fmt;
+			rv.l = ieee754dp_2008class(fs);
+			rfmt = l_fmt;
 			break;
 		}
 
@@ -2157,6 +2213,7 @@ copcsr:
 			if (!cpu_has_mips_r6)
 				return SIGILL;
 
+			MIPS_FPU_EMU_INC_STATS(min_d);
 			DPFROMREG(ft, MIPSInst_FT(ir));
 			DPFROMREG(fs, MIPSInst_FS(ir));
 			rv.d = ieee754dp_fmin(fs, ft);
@@ -2169,6 +2226,7 @@ copcsr:
 			if (!cpu_has_mips_r6)
 				return SIGILL;
 
+			MIPS_FPU_EMU_INC_STATS(mina_d);
 			DPFROMREG(ft, MIPSInst_FT(ir));
 			DPFROMREG(fs, MIPSInst_FS(ir));
 			rv.d = ieee754dp_fmina(fs, ft);
@@ -2181,6 +2239,7 @@ copcsr:
 			if (!cpu_has_mips_r6)
 				return SIGILL;
 
+			MIPS_FPU_EMU_INC_STATS(max_d);
 			DPFROMREG(ft, MIPSInst_FT(ir));
 			DPFROMREG(fs, MIPSInst_FS(ir));
 			rv.d = ieee754dp_fmax(fs, ft);
@@ -2193,6 +2252,7 @@ copcsr:
 			if (!cpu_has_mips_r6)
 				return SIGILL;
 
+			MIPS_FPU_EMU_INC_STATS(maxa_d);
 			DPFROMREG(ft, MIPSInst_FT(ir));
 			DPFROMREG(fs, MIPSInst_FS(ir));
 			rv.d = ieee754dp_fmaxa(fs, ft);
@@ -2200,15 +2260,18 @@ copcsr:
 		}
 
 		case fabs_op:
+			MIPS_FPU_EMU_INC_STATS(abs_d);
 			handler.u = ieee754dp_abs;
 			goto dcopuop;
 
 		case fneg_op:
+			MIPS_FPU_EMU_INC_STATS(neg_d);
 			handler.u = ieee754dp_neg;
 			goto dcopuop;
 
 		case fmov_op:
 			/* an easy one */
+			MIPS_FPU_EMU_INC_STATS(mov_d);
 			DPFROMREG(rv.d, MIPSInst_FS(ir));
 			goto copcsr;
 
@@ -2228,6 +2291,7 @@ dcopuop:
 		 * unary conv ops
 		 */
 		case fcvts_op:
+			MIPS_FPU_EMU_INC_STATS(cvt_s_d);
 			DPFROMREG(fs, MIPSInst_FS(ir));
 			rv.s = ieee754sp_fdp(fs);
 			rfmt = s_fmt;
@@ -2237,6 +2301,7 @@ dcopuop:
 			return SIGILL;	/* not defined */
 
 		case fcvtw_op:
+			MIPS_FPU_EMU_INC_STATS(cvt_w_d);
 			DPFROMREG(fs, MIPSInst_FS(ir));
 			rv.w = ieee754dp_tint(fs);	/* wrong */
 			rfmt = w_fmt;
@@ -2249,6 +2314,15 @@ dcopuop:
 			if (!cpu_has_mips_2_3_4_5_r)
 				return SIGILL;
 
+			if (MIPSInst_FUNC(ir) == fceil_op)
+				MIPS_FPU_EMU_INC_STATS(ceil_w_d);
+			if (MIPSInst_FUNC(ir) == ffloor_op)
+				MIPS_FPU_EMU_INC_STATS(floor_w_d);
+			if (MIPSInst_FUNC(ir) == fround_op)
+				MIPS_FPU_EMU_INC_STATS(round_w_d);
+			if (MIPSInst_FUNC(ir) == ftrunc_op)
+				MIPS_FPU_EMU_INC_STATS(trunc_w_d);
+
 			oldrm = ieee754_csr.rm;
 			DPFROMREG(fs, MIPSInst_FS(ir));
 			ieee754_csr.rm = MIPSInst_FUNC(ir);
@@ -2261,6 +2335,7 @@ dcopuop:
 			if (!cpu_has_mips_r6)
 				return SIGILL;
 
+			MIPS_FPU_EMU_INC_STATS(sel_d);
 			DPFROMREG(fd, MIPSInst_FD(ir));
 			if (fd.bits & 0x1)
 				DPFROMREG(rv.d, MIPSInst_FT(ir));
@@ -2272,6 +2347,7 @@ dcopuop:
 			if (!cpu_has_mips_3_4_5_64_r2_r6)
 				return SIGILL;
 
+			MIPS_FPU_EMU_INC_STATS(cvt_l_d);
 			DPFROMREG(fs, MIPSInst_FS(ir));
 			rv.l = ieee754dp_tlong(fs);
 			rfmt = l_fmt;
@@ -2284,6 +2360,15 @@ dcopuop:
 			if (!cpu_has_mips_3_4_5_64_r2_r6)
 				return SIGILL;
 
+			if (MIPSInst_FUNC(ir) == fceill_op)
+				MIPS_FPU_EMU_INC_STATS(ceil_l_d);
+			if (MIPSInst_FUNC(ir) == ffloorl_op)
+				MIPS_FPU_EMU_INC_STATS(floor_l_d);
+			if (MIPSInst_FUNC(ir) == froundl_op)
+				MIPS_FPU_EMU_INC_STATS(round_l_d);
+			if (MIPSInst_FUNC(ir) == ftruncl_op)
+				MIPS_FPU_EMU_INC_STATS(trunc_l_d);
+
 			oldrm = ieee754_csr.rm;
 			DPFROMREG(fs, MIPSInst_FS(ir));
 			ieee754_csr.rm = MIPSInst_FUNC(ir);
@@ -2325,12 +2410,14 @@ dcopuop:
 		switch (MIPSInst_FUNC(ir)) {
 		case fcvts_op:
 			/* convert word to single precision real */
+			MIPS_FPU_EMU_INC_STATS(cvt_s_w);
 			SPFROMREG(fs, MIPSInst_FS(ir));
 			rv.s = ieee754sp_fint(fs.bits);
 			rfmt = s_fmt;
 			goto copcsr;
 		case fcvtd_op:
 			/* convert word to double precision real */
+			MIPS_FPU_EMU_INC_STATS(cvt_d_w);
 			SPFROMREG(fs, MIPSInst_FS(ir));
 			rv.d = ieee754dp_fint(fs.bits);
 			rfmt = d_fmt;
@@ -2350,6 +2437,90 @@ dcopuop:
 			    (MIPSInst_FUNC(ir) & 0x20))
 				return SIGILL;
 
+			if (!sig) {
+				if (!(MIPSInst_FUNC(ir) & PREDICATE_BIT)) {
+					switch (cmpop) {
+					case 0:
+					MIPS_FPU_EMU_INC_STATS(cmp_af_s);
+					break;
+					case 1:
+					MIPS_FPU_EMU_INC_STATS(cmp_un_s);
+					break;
+					case 2:
+					MIPS_FPU_EMU_INC_STATS(cmp_eq_s);
+					break;
+					case 3:
+					MIPS_FPU_EMU_INC_STATS(cmp_ueq_s);
+					break;
+					case 4:
+					MIPS_FPU_EMU_INC_STATS(cmp_lt_s);
+					break;
+					case 5:
+					MIPS_FPU_EMU_INC_STATS(cmp_ult_s);
+					break;
+					case 6:
+					MIPS_FPU_EMU_INC_STATS(cmp_le_s);
+					break;
+					case 7:
+					MIPS_FPU_EMU_INC_STATS(cmp_ule_s);
+					break;
+					}
+				} else {
+					switch (cmpop) {
+					case 1:
+					MIPS_FPU_EMU_INC_STATS(cmp_or_s);
+					break;
+					case 2:
+					MIPS_FPU_EMU_INC_STATS(cmp_une_s);
+					break;
+					case 3:
+					MIPS_FPU_EMU_INC_STATS(cmp_ne_s);
+					break;
+					}
+				}
+			} else {
+				if (!(MIPSInst_FUNC(ir) & PREDICATE_BIT)) {
+					switch (cmpop) {
+					case 0:
+					MIPS_FPU_EMU_INC_STATS(cmp_saf_s);
+					break;
+					case 1:
+					MIPS_FPU_EMU_INC_STATS(cmp_sun_s);
+					break;
+					case 2:
+					MIPS_FPU_EMU_INC_STATS(cmp_seq_s);
+					break;
+					case 3:
+					MIPS_FPU_EMU_INC_STATS(cmp_sueq_s);
+					break;
+					case 4:
+					MIPS_FPU_EMU_INC_STATS(cmp_slt_s);
+					break;
+					case 5:
+					MIPS_FPU_EMU_INC_STATS(cmp_sult_s);
+					break;
+					case 6:
+					MIPS_FPU_EMU_INC_STATS(cmp_sle_s);
+					break;
+					case 7:
+					MIPS_FPU_EMU_INC_STATS(cmp_sule_s);
+					break;
+					}
+				} else {
+					switch (cmpop) {
+					case 1:
+					MIPS_FPU_EMU_INC_STATS(cmp_sor_s);
+					break;
+					case 2:
+					MIPS_FPU_EMU_INC_STATS(cmp_sune_s);
+					break;
+					case 3:
+					MIPS_FPU_EMU_INC_STATS(cmp_sne_s);
+					break;
+					}
+				}
+			}
+
 			/* fmt is w_fmt for single precision so fix it */
 			rfmt = s_fmt;
 			/* default to false */
@@ -2394,6 +2565,7 @@ dcopuop:
 			break;
 			}
 		}
+		break;
 	}
 
 	case l_fmt:
@@ -2406,11 +2578,13 @@ dcopuop:
 		switch (MIPSInst_FUNC(ir)) {
 		case fcvts_op:
 			/* convert long to single precision real */
+			MIPS_FPU_EMU_INC_STATS(cvt_s_l);
 			rv.s = ieee754sp_flong(bits);
 			rfmt = s_fmt;
 			goto copcsr;
 		case fcvtd_op:
 			/* convert long to double precision real */
+			MIPS_FPU_EMU_INC_STATS(cvt_d_l);
 			rv.d = ieee754dp_flong(bits);
 			rfmt = d_fmt;
 			goto copcsr;
@@ -2424,6 +2598,90 @@ dcopuop:
 			    (MIPSInst_FUNC(ir) & 0x20))
 				return SIGILL;
 
+			if (!sig) {
+				if (!(MIPSInst_FUNC(ir) & PREDICATE_BIT)) {
+					switch (cmpop) {
+					case 0:
+					MIPS_FPU_EMU_INC_STATS(cmp_af_d);
+					break;
+					case 1:
+					MIPS_FPU_EMU_INC_STATS(cmp_un_d);
+					break;
+					case 2:
+					MIPS_FPU_EMU_INC_STATS(cmp_eq_d);
+					break;
+					case 3:
+					MIPS_FPU_EMU_INC_STATS(cmp_ueq_d);
+					break;
+					case 4:
+					MIPS_FPU_EMU_INC_STATS(cmp_lt_d);
+					break;
+					case 5:
+					MIPS_FPU_EMU_INC_STATS(cmp_ult_d);
+					break;
+					case 6:
+					MIPS_FPU_EMU_INC_STATS(cmp_le_d);
+					break;
+					case 7:
+					MIPS_FPU_EMU_INC_STATS(cmp_ule_d);
+					break;
+					}
+				} else {
+					switch (cmpop) {
+					case 1:
+					MIPS_FPU_EMU_INC_STATS(cmp_or_d);
+					break;
+					case 2:
+					MIPS_FPU_EMU_INC_STATS(cmp_une_d);
+					break;
+					case 3:
+					MIPS_FPU_EMU_INC_STATS(cmp_ne_d);
+					break;
+					}
+				}
+			} else {
+				if (!(MIPSInst_FUNC(ir) & PREDICATE_BIT)) {
+					switch (cmpop) {
+					case 0:
+					MIPS_FPU_EMU_INC_STATS(cmp_saf_d);
+					break;
+					case 1:
+					MIPS_FPU_EMU_INC_STATS(cmp_sun_d);
+					break;
+					case 2:
+					MIPS_FPU_EMU_INC_STATS(cmp_seq_d);
+					break;
+					case 3:
+					MIPS_FPU_EMU_INC_STATS(cmp_sueq_d);
+					break;
+					case 4:
+					MIPS_FPU_EMU_INC_STATS(cmp_slt_d);
+					break;
+					case 5:
+					MIPS_FPU_EMU_INC_STATS(cmp_sult_d);
+					break;
+					case 6:
+					MIPS_FPU_EMU_INC_STATS(cmp_sle_d);
+					break;
+					case 7:
+					MIPS_FPU_EMU_INC_STATS(cmp_sule_d);
+					break;
+					}
+				} else {
+					switch (cmpop) {
+					case 1:
+					MIPS_FPU_EMU_INC_STATS(cmp_sor_d);
+					break;
+					case 2:
+					MIPS_FPU_EMU_INC_STATS(cmp_sune_d);
+					break;
+					case 3:
+					MIPS_FPU_EMU_INC_STATS(cmp_sne_d);
+					break;
+					}
+				}
+			}
+
 			/* fmt is l_fmt for double precision so fix it */
 			rfmt = d_fmt;
 			/* default to false */
@@ -2468,6 +2726,8 @@ dcopuop:
 			break;
 			}
 		}
+		break;
+
 	default:
 		return SIGILL;
 	}
@@ -2553,7 +2813,7 @@ dcopuop:
  * For simplicity we always terminate upon an ISA mode switch.
  */
 int fpu_emulator_cop1Handler(struct pt_regs *xcp, struct mips_fpu_struct *ctx,
-	int has_fpu, void *__user *fault_addr)
+	int has_fpu, void __user **fault_addr)
 {
 	unsigned long oldepc, prevepc;
 	struct mm_decoded_insn dec_insn;