Merge tag 'arc-5.6-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/vgupta/arc

Pull ARC updates from Vineet Gupta:

 - Wire up clone3 syscall

 - ARCv2 FPU state save/restore across context switch

 - AXS10x platform and misc fixes

* tag 'arc-5.6-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/vgupta/arc:
  ARCv2: fpu: preserve userspace fpu state
  ARC: fpu: declutter code, move bits out into fpu.h
  ARC: wireup clone3 syscall
  ARC: [plat-axs10x]: Add missing multicast filter number to GMAC node
  ARC: update feature support for jump-labels

13 files changed, 120 insertions, 40 deletions

diff --git a/arch/arc/Kconfig b/arch/arc/Kconfig
index 5f448201955b..ff2a393b635c 100644
--- a/arch/arc/Kconfig
+++ b/arch/arc/Kconfig
@@ -28,6 +28,7 @@ config ARC
 	select GENERIC_SMP_IDLE_THREAD
 	select HAVE_ARCH_KGDB
 	select HAVE_ARCH_TRACEHOOK
+	select HAVE_COPY_THREAD_TLS
 	select HAVE_DEBUG_STACKOVERFLOW
 	select HAVE_DEBUG_KMEMLEAK
 	select HAVE_FUTEX_CMPXCHG if FUTEX
@@ -350,9 +351,8 @@ config NODES_SHIFT
 	  Accessing memory beyond 1GB (with or w/o PAE) requires 2 memory
 	  zones.
 
-if ISA_ARCOMPACT
-
 config ARC_COMPACT_IRQ_LEVELS
+	depends on ISA_ARCOMPACT
 	bool "Setup Timer IRQ as high Priority"
 	# if SMP, LV2 enabled ONLY if ARC implementation has LV2 re-entrancy
 	depends on !SMP
@@ -360,14 +360,10 @@ config ARC_COMPACT_IRQ_LEVELS
 config ARC_FPU_SAVE_RESTORE
 	bool "Enable FPU state persistence across context switch"
 	help
-	  Double Precision Floating Point unit had dedicated regs which
-	  need to be saved/restored across context-switch.
-	  Note that ARC FPU is overly simplistic, unlike say x86, which has
-	  hardware pieces to allow software to conditionally save/restore,
-	  based on actual usage of FPU by a task. Thus our implemn does
-	  this for all tasks in system.
-
-endif #ISA_ARCOMPACT
+	  ARCompact FPU has internal registers to assist with Double precision
+	  Floating Point operations. There are control and stauts registers
+	  for floating point exceptions and rounding modes. These are
+	  preserved across task context switch when enabled.
 
 config ARC_CANT_LLSC
 	def_bool n
diff --git a/arch/arc/boot/dts/axs10x_mb.dtsi b/arch/arc/boot/dts/axs10x_mb.dtsi
index f9a5c9ddcae7..1d109b06e7d8 100644
--- a/arch/arc/boot/dts/axs10x_mb.dtsi
+++ b/arch/arc/boot/dts/axs10x_mb.dtsi
@@ -78,6 +78,7 @@
 			interrupt-names = "macirq";
 			phy-mode = "rgmii";
 			snps,pbl = < 32 >;
+			snps,multicast-filter-bins = <256>;
 			clocks = <&apbclk>;
 			clock-names = "stmmaceth";
 			max-speed = <100>;
diff --git a/arch/arc/include/asm/arcregs.h b/arch/arc/include/asm/arcregs.h
index 5134f0baf33c..f7e432448e4b 100644
--- a/arch/arc/include/asm/arcregs.h
+++ b/arch/arc/include/asm/arcregs.h
@@ -39,6 +39,8 @@
 #define ARC_REG_CLUSTER_BCR	0xcf
 #define ARC_REG_AUX_ICCM	0x208	/* ICCM Base Addr (ARCv2) */
 #define ARC_REG_LPB_CTRL	0x488	/* ARCv2 Loop Buffer control */
+#define ARC_REG_FPU_CTRL	0x300
+#define ARC_REG_FPU_STATUS	0x301
 
 /* Common for ARCompact and ARCv2 status register */
 #define ARC_REG_STATUS32	0x0A
diff --git a/arch/arc/include/asm/fpu.h b/arch/arc/include/asm/fpu.h
new file mode 100644
index 000000000000..64347250fdf5
--- /dev/null
+++ b/arch/arc/include/asm/fpu.h
@@ -0,0 +1,55 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (C) 2020 Synopsys, Inc. (www.synopsys.com)
+ *
+ */
+
+#ifndef _ASM_ARC_FPU_H
+#define _ASM_ARC_FPU_H
+
+#ifdef CONFIG_ARC_FPU_SAVE_RESTORE
+
+#include <asm/ptrace.h>
+
+#ifdef CONFIG_ISA_ARCOMPACT
+
+/* These DPFP regs need to be saved/restored across ctx-sw */
+struct arc_fpu {
+	struct {
+		unsigned int l, h;
+	} aux_dpfp[2];
+};
+
+#define fpu_init_task(regs)
+
+#else
+
+/*
+ * ARCv2 FPU Control aux register
+ *   - bits to enable Traps on Exceptions
+ *   - Rounding mode
+ *
+ * ARCv2 FPU Status aux register
+ *   - FPU exceptions flags (Inv, Div-by-Zero, overflow, underflow, inexact)
+ *   - Flag Write Enable to clear flags explicitly (vs. by fpu instructions
+ *     only
+ */
+
+struct arc_fpu {
+	unsigned int ctrl, status;
+};
+
+extern void fpu_init_task(struct pt_regs *regs);
+
+#endif	/* !CONFIG_ISA_ARCOMPACT */
+
+extern void fpu_save_restore(struct task_struct *p, struct task_struct *n);
+
+#else	/* !CONFIG_ARC_FPU_SAVE_RESTORE */
+
+#define fpu_save_restore(p, n)
+#define fpu_init_task(regs)
+
+#endif	/* CONFIG_ARC_FPU_SAVE_RESTORE */
+
+#endif	/* _ASM_ARC_FPU_H */
diff --git a/arch/arc/include/asm/processor.h b/arch/arc/include/asm/processor.h
index 706edeaa5583..ec532d1e0725 100644
--- a/arch/arc/include/asm/processor.h
+++ b/arch/arc/include/asm/processor.h
@@ -14,15 +14,7 @@
 #ifndef __ASSEMBLY__
 
 #include <asm/ptrace.h>
-
-#ifdef CONFIG_ARC_FPU_SAVE_RESTORE
-/* These DPFP regs need to be saved/restored across ctx-sw */
-struct arc_fpu {
-	struct {
-		unsigned int l, h;
-	} aux_dpfp[2];
-};
-#endif
+#include <asm/fpu.h>
 
 #ifdef CONFIG_ARC_PLAT_EZNPS
 struct eznps_dp {
diff --git a/arch/arc/include/asm/switch_to.h b/arch/arc/include/asm/switch_to.h
index 77f123385e96..aadf65b2b56c 100644
--- a/arch/arc/include/asm/switch_to.h
+++ b/arch/arc/include/asm/switch_to.h
@@ -9,19 +9,7 @@
 #ifndef __ASSEMBLY__
 
 #include <linux/sched.h>
-
-#ifdef CONFIG_ARC_FPU_SAVE_RESTORE
-
-extern void fpu_save_restore(struct task_struct *p, struct task_struct *n);
-#define ARC_FPU_PREV(p, n)	fpu_save_restore(p, n)
-#define ARC_FPU_NEXT(t)
-
-#else
-
-#define ARC_FPU_PREV(p, n)
-#define ARC_FPU_NEXT(n)
-
-#endif /* !CONFIG_ARC_FPU_SAVE_RESTORE */
+#include <asm/fpu.h>
 
 #ifdef CONFIG_ARC_PLAT_EZNPS
 extern void dp_save_restore(struct task_struct *p, struct task_struct *n);
@@ -36,9 +24,8 @@ struct task_struct *__switch_to(struct task_struct *p, struct task_struct *n);
 #define switch_to(prev, next, last)	\
 do {					\
 	ARC_EZNPS_DP_PREV(prev, next);	\
-	ARC_FPU_PREV(prev, next);	\
+	fpu_save_restore(prev, next);	\
 	last = __switch_to(prev, next);\
-	ARC_FPU_NEXT(next);		\
 	mb();				\
 } while (0)
 
diff --git a/arch/arc/include/asm/syscalls.h b/arch/arc/include/asm/syscalls.h
index 7ddba13e9b59..c3f4714a4f5c 100644
--- a/arch/arc/include/asm/syscalls.h
+++ b/arch/arc/include/asm/syscalls.h
@@ -11,6 +11,7 @@
 #include <linux/types.h>
 
 int sys_clone_wrapper(int, int, int, int, int);
+int sys_clone3_wrapper(void *, size_t);
 int sys_cacheflush(uint32_t, uint32_t uint32_t);
 int sys_arc_settls(void *);
 int sys_arc_gettls(void);
diff --git a/arch/arc/include/uapi/asm/unistd.h b/arch/arc/include/uapi/asm/unistd.h
index 5eafa1115162..fa2713ae6bea 100644
--- a/arch/arc/include/uapi/asm/unistd.h
+++ b/arch/arc/include/uapi/asm/unistd.h
@@ -21,6 +21,7 @@
 #define __ARCH_WANT_SET_GET_RLIMIT
 #define __ARCH_WANT_SYS_EXECVE
 #define __ARCH_WANT_SYS_CLONE
+#define __ARCH_WANT_SYS_CLONE3
 #define __ARCH_WANT_SYS_VFORK
 #define __ARCH_WANT_SYS_FORK
 #define __ARCH_WANT_TIME32_SYSCALLS
diff --git a/arch/arc/kernel/Makefile b/arch/arc/kernel/Makefile
index e784f5396dda..75539670431a 100644
--- a/arch/arc/kernel/Makefile
+++ b/arch/arc/kernel/Makefile
@@ -23,7 +23,9 @@ obj-$(CONFIG_PERF_EVENTS)		+= perf_event.o
 obj-$(CONFIG_JUMP_LABEL)		+= jump_label.o
 
 obj-$(CONFIG_ARC_FPU_SAVE_RESTORE)	+= fpu.o
+ifdef CONFIG_ISA_ARCOMPACT
 CFLAGS_fpu.o   += -mdpfp
+endif
 
 ifdef CONFIG_ARC_DW2_UNWIND
 CFLAGS_ctx_sw.o += -fno-omit-frame-pointer
diff --git a/arch/arc/kernel/entry.S b/arch/arc/kernel/entry.S
index 1f6bb184a44d..60406ec62eb8 100644
--- a/arch/arc/kernel/entry.S
+++ b/arch/arc/kernel/entry.S
@@ -35,6 +35,18 @@ ENTRY(sys_clone_wrapper)
 	b .Lret_from_system_call
 END(sys_clone_wrapper)
 
+ENTRY(sys_clone3_wrapper)
+	SAVE_CALLEE_SAVED_USER
+	bl  @sys_clone3
+	DISCARD_CALLEE_SAVED_USER
+
+	GET_CURR_THR_INFO_FLAGS   r10
+	btst r10, TIF_SYSCALL_TRACE
+	bnz  tracesys_exit
+
+	b .Lret_from_system_call
+END(sys_clone3_wrapper)
+
 ENTRY(ret_from_fork)
 	; when the forked child comes here from the __switch_to function
 	; r0 has the last task pointer.
diff --git a/arch/arc/kernel/fpu.c b/arch/arc/kernel/fpu.c
index 07e22b563fbb..c67c0f0f5f77 100644
--- a/arch/arc/kernel/fpu.c
+++ b/arch/arc/kernel/fpu.c
@@ -6,7 +6,9 @@
  */
 
 #include <linux/sched.h>
-#include <asm/switch_to.h>
+#include <asm/fpu.h>
+
+#ifdef CONFIG_ISA_ARCOMPACT
 
 /*
  * To save/restore FPU regs, simplest scheme would use LR/SR insns.
@@ -50,3 +52,28 @@ void fpu_save_restore(struct task_struct *prev, struct task_struct *next)
 		: "r" (zero), "r" (*(readfrom + 3)), "r" (*(readfrom + 2))
 	);
 }
+
+#else
+
+void fpu_init_task(struct pt_regs *regs)
+{
+	/* default rounding mode */
+	write_aux_reg(ARC_REG_FPU_CTRL, 0x100);
+
+	/* set "Write enable" to allow explicit write to exception flags */
+	write_aux_reg(ARC_REG_FPU_STATUS, 0x80000000);
+}
+
+void fpu_save_restore(struct task_struct *prev, struct task_struct *next)
+{
+	struct arc_fpu *save = &prev->thread.fpu;
+	struct arc_fpu *restore = &next->thread.fpu;
+
+	save->ctrl = read_aux_reg(ARC_REG_FPU_CTRL);
+	save->status = read_aux_reg(ARC_REG_FPU_STATUS);
+
+	write_aux_reg(ARC_REG_FPU_CTRL, restore->ctrl);
+	write_aux_reg(ARC_REG_FPU_STATUS, restore->status);
+}
+
+#endif
diff --git a/arch/arc/kernel/process.c b/arch/arc/kernel/process.c
index e1889ce3faf9..315528f04bc1 100644
--- a/arch/arc/kernel/process.c
+++ b/arch/arc/kernel/process.c
@@ -20,6 +20,8 @@
 #include <linux/elf.h>
 #include <linux/tick.h>
 
+#include <asm/fpu.h>
+
 SYSCALL_DEFINE1(arc_settls, void *, user_tls_data_ptr)
 {
 	task_thread_info(current)->thr_ptr = (unsigned int)user_tls_data_ptr;
@@ -171,9 +173,8 @@ asmlinkage void ret_from_fork(void);
  * |    user_r25    |
  * ------------------  <===== END of PAGE
  */
-int copy_thread(unsigned long clone_flags,
-		unsigned long usp, unsigned long kthread_arg,
-		struct task_struct *p)
+int copy_thread_tls(unsigned long clone_flags, unsigned long usp,
+	unsigned long kthread_arg, struct task_struct *p, unsigned long tls)
 {
 	struct pt_regs *c_regs;        /* child's pt_regs */
 	unsigned long *childksp;       /* to unwind out of __switch_to() */
@@ -231,7 +232,7 @@ int copy_thread(unsigned long clone_flags,
 		 * set task's userland tls data ptr from 4th arg
 		 * clone C-lib call is difft from clone sys-call
 		 */
-		task_thread_info(p)->thr_ptr = regs->r3;
+		task_thread_info(p)->thr_ptr = tls;
 	} else {
 		/* Normal fork case: set parent's TLS ptr in child */
 		task_thread_info(p)->thr_ptr =
@@ -264,7 +265,7 @@ int copy_thread(unsigned long clone_flags,
 /*
  * Do necessary setup to start up a new user task
  */
-void start_thread(struct pt_regs * regs, unsigned long pc, unsigned long usp)
+void start_thread(struct pt_regs *regs, unsigned long pc, unsigned long usp)
 {
 	regs->sp = usp;
 	regs->ret = pc;
@@ -280,6 +281,8 @@ void start_thread(struct pt_regs * regs, unsigned long pc, unsigned long usp)
 	regs->eflags = 0;
 #endif
 
+	fpu_init_task(regs);
+
 	/* bogus seed values for debugging */
 	regs->lp_start = 0x10;
 	regs->lp_end = 0x80;
diff --git a/arch/arc/kernel/sys.c b/arch/arc/kernel/sys.c
index fddecc76efb7..1069446bdc58 100644
--- a/arch/arc/kernel/sys.c
+++ b/arch/arc/kernel/sys.c
@@ -7,6 +7,7 @@
 #include <asm/syscalls.h>
 
 #define sys_clone	sys_clone_wrapper
+#define sys_clone3	sys_clone3_wrapper
 
 #undef __SYSCALL
 #define __SYSCALL(nr, call) [nr] = (call),