Merge tag 'riscv-for-linus-4.19-mw0' of...

Merge tag 'riscv-for-linus-4.19-mw0' of git://git.kernel.org/pub/scm/linux/kernel/git/palmer/riscv-linux

Pull RISC-V updates from Palmer Dabbelt:
 "This contains some major improvements to the RISC-V port, including
  the necessary interrupt controller and timer support to actually make
  it to userspace. Support for three devices has been added:

   - the ISA-mandated timers on RISC-V systems.

   - the ISA-mandated first-level interrupt controller on RISC-V
     systems, which is handled as part of our core arch code because
     it's very small and tightly tied to the ISA.

   - SiFive's platform-level interrupt controller, which talks to the
     actual devices.

  In addition to these new devices, there are a handful of cleanups all
  over the RISC-V tree:

   - build fixes for various configurations:
      * A fix to the vDSO build's makefile so it respects CFLAGS.
      * The addition of __lshrti3, a libgcc derived function necessary
        for some 32-bit configurations.
      * !SMP && PERF_EVENTS

   - Cleanups to the arch code to remove the remnants of old versions of
     the drivers that were just properly submitted.
      * Some dead code from the timer driver, most of which wasn't ever
        even compiled.
      * Cleanups of some interrupt #defines, which are now local to the
        interrupt handling code.

   - Fixes to ptrace(), which while not being sufficient to fully make
     GDB work are at least sufficient to get simple GDB tasks to work.

   - Early printk support via RISC-V's architecturally mandated SBI
     console device.

   - A fix to our early debug trap handler to ensure it's always
     aligned.

  These patches have all been through a fairly extensive review process,
  but as this enables a whole pile of functionality (ie, userspace) I'm
  confident we'll need to submit a few more patches. The only concrete
  issues I know about are the sys_riscv_flush_icache patches, but as I
  managed to screw those up on Friday I figured it'd be best to let them
  bake another week.

  This tag boots a Fedora root filesystem on QEMU's master branch for
  me, and before this morning's rebase (from 4.18-rc8 to 4.18) it booted
  on the HiFive Unleashed.

  Thanks to Christoph Hellwig and the other guys at WD for getting the
  new drivers in shape!"

* tag 'riscv-for-linus-4.19-mw0' of git://git.kernel.org/pub/scm/linux/kernel/git/palmer/riscv-linux:
  dt-bindings: interrupt-controller: SiFive Plaform Level Interrupt Controller
  dt-bindings: interrupt-controller: RISC-V local interrupt controller
  RISC-V: Fix !CONFIG_SMP compilation error
  irqchip: add a SiFive PLIC driver
  RISC-V: Add the directive for alignment of stvec's value
  clocksource: new RISC-V SBI timer driver
  RISC-V: implement low-level interrupt handling
  RISC-V: add a definition for the SIE SEIE bit
  RISC-V: remove INTERRUPT_CAUSE_* defines from asm/irq.h
  RISC-V: simplify software interrupt / IPI code
  RISC-V: remove timer leftovers
  RISC-V: Add early printk support via the SBI console
  RISC-V: Don't increment sepc after breakpoint.
  RISC-V: implement __lshrti3.
  RISC-V: Use KBUILD_CFLAGS instead of KCFLAGS when building the vDSO

Documentation/devicetree/bindings/interrupt-controller/riscv,cpu-intc.txt

+RISC-V Hart-Level Interrupt Controller (HLIC)
+---------------------------------------------
+
+RISC-V cores include Control Status Registers (CSRs) which are local to each
+CPU core (HART in RISC-V terminology) and can be read or written by software.
+Some of these CSRs are used to control local interrupts connected to the core.
+Every interrupt is ultimately routed through a hart's HLIC before it
+interrupts that hart.
+
+The RISC-V supervisor ISA manual specifies three interrupt sources that are
+attached to every HLIC: software interrupts, the timer interrupt, and external
+interrupts.  Software interrupts are used to send IPIs between cores.  The
+timer interrupt comes from an architecturally mandated real-time timer that is
+controller via Supervisor Binary Interface (SBI) calls and CSR reads.  External
+interrupts connect all other device interrupts to the HLIC, which are routed
+via the platform-level interrupt controller (PLIC).
+
+All RISC-V systems that conform to the supervisor ISA specification are
+required to have a HLIC with these three interrupt sources present.  Since the
+interrupt map is defined by the ISA it's not listed in the HLIC's device tree
+entry, though external interrupt controllers (like the PLIC, for example) will
+need to define how their interrupts map to the relevant HLICs.  This means
+a PLIC interrupt property will typically list the HLICs for all present HARTs
+in the system.
+
+Required properties:
+- compatible : "riscv,cpu-intc"
+- #interrupt-cells : should be <1>
+- interrupt-controller : Identifies the node as an interrupt controller
+
+Furthermore, this interrupt-controller MUST be embedded inside the cpu
+definition of the hart whose CSRs control these local interrupts.
+
+An example device tree entry for a HLIC is show below.
+
+	cpu1: cpu@1 {
+		compatible = "riscv";
+		...
+		cpu1-intc: interrupt-controller {
+			#interrupt-cells = <1>;
+			compatible = "riscv,cpu-intc", "sifive,fu540-c000-cpu-intc";
+			interrupt-controller;
+		};
+	};

Documentation/devicetree/bindings/interrupt-controller/sifive,plic-1.0.0.txt

+SiFive Platform-Level Interrupt Controller (PLIC)
+-------------------------------------------------
+
+SiFive SOCs include an implementation of the Platform-Level Interrupt Controller
+(PLIC) high-level specification in the RISC-V Privileged Architecture
+specification.  The PLIC connects all external interrupts in the system to all
+hart contexts in the system, via the external interrupt source in each hart.
+
+A hart context is a privilege mode in a hardware execution thread.  For example,
+in an 4 core system with 2-way SMT, you have 8 harts and probably at least two
+privilege modes per hart; machine mode and supervisor mode.
+
+Each interrupt can be enabled on per-context basis.  Any context can claim
+a pending enabled interrupt and then release it once it has been handled.
+
+Each interrupt has a configurable priority.  Higher priority interrupts are
+serviced first.  Each context can specify a priority threshold. Interrupts
+with priority below this threshold will not cause the PLIC to raise its
+interrupt line leading to the context.
+
+While the PLIC supports both edge-triggered and level-triggered interrupts,
+interrupt handlers are oblivious to this distinction and therefore it is not
+specified in the PLIC device-tree binding.
+
+While the RISC-V ISA doesn't specify a memory layout for the PLIC, the
+"sifive,plic-1.0.0" device is a concrete implementation of the PLIC that
+contains a specific memory layout, which is documented in chapter 8 of the
+SiFive U5 Coreplex Series Manual <https://static.dev.sifive.com/U54-MC-RVCoreIP.pdf>.
+
+Required properties:
+- compatible : "sifive,plic-1.0.0" and a string identifying the actual
+  detailed implementation in case that specific bugs need to be worked around.
+- #address-cells : should be <0> or more.
+- #interrupt-cells : should be <1> or more.
+- interrupt-controller : Identifies the node as an interrupt controller.
+- reg : Should contain 1 register range (address and length).
+- interrupts-extended : Specifies which contexts are connected to the PLIC,
+  with "-1" specifying that a context is not present.  Each node pointed
+  to should be a riscv,cpu-intc node, which has a riscv node as parent.
+- riscv,ndev: Specifies how many external interrupts are supported by
+  this controller.
+
+Example:
+
+	plic: interrupt-controller@c000000 {
+		#address-cells = <0>;
+		#interrupt-cells = <1>;
+		compatible = "sifive,plic-1.0.0", "sifive,fu540-c000-plic";
+		interrupt-controller;
+		interrupts-extended = <
+			&cpu0-intc 11
+			&cpu1-intc 11 &cpu1-intc 9
+			&cpu2-intc 11 &cpu2-intc 9
+			&cpu3-intc 11 &cpu3-intc 9
+			&cpu4-intc 11 &cpu4-intc 9>;
+		reg = <0xc000000 0x4000000>;
+		riscv,ndev = <10>;
+	};

arch/riscv/Makefile

@@ -25,6 +25,9 @@ ifeq ($(CONFIG_ARCH_RV64I),y)
 
 	KBUILD_CFLAGS += -mabi=lp64
 	KBUILD_AFLAGS += -mabi=lp64
+	
+	KBUILD_CFLAGS	+= $(call cc-ifversion, -ge, 0500, -DCONFIG_ARCH_SUPPORTS_INT128)
+
 	KBUILD_MARCH = rv64im
 	LDFLAGS += -melf64lriscv
 else

arch/riscv/configs/defconfig

@@ -76,3 +76,4 @@ CONFIG_ROOT_NFS=y
 CONFIG_CRYPTO_USER_API_HASH=y
 CONFIG_MODULES=y
 CONFIG_MODULE_UNLOAD=y
+CONFIG_SIFIVE_PLIC=y

arch/riscv/include/asm/csr.h

@@ -54,6 +54,7 @@
 /* Interrupt Enable and Interrupt Pending flags */
 #define SIE_SSIE _AC(0x00000002, UL) /* Software Interrupt Enable */
 #define SIE_STIE _AC(0x00000020, UL) /* Timer Interrupt Enable */
+#define SIE_SEIE _AC(0x00000200, UL) /* External Interrupt Enable */
 
 #define EXC_INST_MISALIGNED     0
 #define EXC_INST_ACCESS         1

arch/riscv/include/asm/irq.h

@@ -17,11 +17,8 @@
 
 #define NR_IRQS         0
 
-#define INTERRUPT_CAUSE_SOFTWARE    1
-#define INTERRUPT_CAUSE_TIMER       5
-#define INTERRUPT_CAUSE_EXTERNAL    9
-
 void riscv_timer_interrupt(void);
+void riscv_software_interrupt(void);
 
 #include <asm-generic/irq.h>
 

arch/riscv/include/asm/perf_event.h

@@ -10,6 +10,7 @@
 
 #include <linux/perf_event.h>
 #include <linux/ptrace.h>
+#include <linux/interrupt.h>
 
 #define RISCV_BASE_COUNTERS	2
 

arch/riscv/include/asm/smp.h

@@ -24,9 +24,6 @@
 
 #ifdef CONFIG_SMP
 
-/* SMP initialization hook for setup_arch */
-void __init init_clockevent(void);
-
 /* SMP initialization hook for setup_arch */
 void __init setup_smp(void);
 
@@ -44,9 +41,6 @@ void arch_send_call_function_single_ipi(int cpu);
  */
 #define raw_smp_processor_id() (*((int*)((char*)get_current() + TASK_TI_CPU)))
 
-/* Interprocessor interrupt handler */
-irqreturn_t handle_ipi(void);
-
 #endif /* CONFIG_SMP */
 
 #endif /* _ASM_RISCV_SMP_H */

arch/riscv/kernel/entry.S

@@ -168,8 +168,8 @@ ENTRY(handle_exception)
 
 	/* Handle interrupts */
 	move a0, sp /* pt_regs */
-	REG_L a1, handle_arch_irq
-	jr a1
+	move a1, s4 /* scause */
+	tail do_IRQ
 1:
 	/* Exceptions run with interrupts enabled */
 	csrs sstatus, SR_SIE

arch/riscv/kernel/head.S

@@ -94,6 +94,7 @@ relocate:
 	or a0, a0, a1
 	sfence.vma
 	csrw sptbr, a0
+.align 2
 1:
 	/* Set trap vector to spin forever to help debug */
 	la a0, .Lsecondary_park
@@ -143,6 +144,7 @@ relocate:
 	tail smp_callin
 #endif
 
+.align 2
 .Lsecondary_park:
 	/* We lack SMP support or have too many harts, so park this hart */
 	wfi

arch/riscv/kernel/irq.c

+// SPDX-License-Identifier: GPL-2.0
 /*
  * Copyright (C) 2012 Regents of the University of California
  * Copyright (C) 2017 SiFive
- *
- *   This program is free software; you can redistribute it and/or
- *   modify it under the terms of the GNU General Public License
- *   as published by the Free Software Foundation, version 2.
- *
- *   This program is distributed in the hope that it will be useful,
- *   but WITHOUT ANY WARRANTY; without even the implied warranty of
- *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- *   GNU General Public License for more details.
+ * Copyright (C) 2018 Christoph Hellwig
  */
 
 #include <linux/interrupt.h>
 #include <linux/irqchip.h>
 #include <linux/irqdomain.h>
 
+/*
+ * Possible interrupt causes:
+ */
+#define INTERRUPT_CAUSE_SOFTWARE    1
+#define INTERRUPT_CAUSE_TIMER       5
+#define INTERRUPT_CAUSE_EXTERNAL    9
+
+/*
+ * The high order bit of the trap cause register is always set for
+ * interrupts, which allows us to differentiate them from exceptions
+ * quickly.  The INTERRUPT_CAUSE_* macros don't contain that bit, so we
+ * need to mask it off.
+ */
+#define INTERRUPT_CAUSE_FLAG	(1UL << (__riscv_xlen - 1))
+
+asmlinkage void __irq_entry do_IRQ(struct pt_regs *regs, unsigned long cause)
+{
+	struct pt_regs *old_regs = set_irq_regs(regs);
+
+	irq_enter();
+	switch (cause & ~INTERRUPT_CAUSE_FLAG) {
+	case INTERRUPT_CAUSE_TIMER:
+		riscv_timer_interrupt();
+		break;
+#ifdef CONFIG_SMP
+	case INTERRUPT_CAUSE_SOFTWARE:
+		/*
+		 * We only use software interrupts to pass IPIs, so if a non-SMP
+		 * system gets one, then we don't know what to do.
+		 */
+		riscv_software_interrupt();
+		break;
+#endif
+	case INTERRUPT_CAUSE_EXTERNAL:
+		handle_arch_irq(regs);
+		break;
+	default:
+		panic("unexpected interrupt cause");
+	}
+	irq_exit();
+
+	set_irq_regs(old_regs);
+}
+
 void __init init_IRQ(void)
 {
 	irqchip_init();

arch/riscv/kernel/perf_event.c

@@ -27,7 +27,6 @@
 #include <linux/mutex.h>
 #include <linux/bitmap.h>
 #include <linux/irq.h>
-#include <linux/interrupt.h>
 #include <linux/perf_event.h>
 #include <linux/atomic.h>
 #include <linux/of.h>

arch/riscv/kernel/setup.c

@@ -39,6 +39,27 @@
 #include <asm/tlbflush.h>
 #include <asm/thread_info.h>
 
+#ifdef CONFIG_EARLY_PRINTK
+static void sbi_console_write(struct console *co, const char *buf,
+			      unsigned int n)
+{
+	int i;
+
+	for (i = 0; i < n; ++i) {
+		if (buf[i] == '\n')
+			sbi_console_putchar('\r');
+		sbi_console_putchar(buf[i]);
+	}
+}
+
+struct console riscv_sbi_early_console_dev __initdata = {
+	.name	= "early",
+	.write	= sbi_console_write,
+	.flags	= CON_PRINTBUFFER | CON_BOOT | CON_ANYTIME,
+	.index	= -1
+};
+#endif
+
 #ifdef CONFIG_DUMMY_CONSOLE
 struct screen_info screen_info = {
 	.orig_video_lines	= 30,
@@ -195,6 +216,12 @@ static void __init setup_bootmem(void)
 
 void __init setup_arch(char **cmdline_p)
 {
+#if defined(CONFIG_EARLY_PRINTK)
+       if (likely(early_console == NULL)) {
+               early_console = &riscv_sbi_early_console_dev;
+               register_console(early_console);
+       }
+#endif
 	*cmdline_p = boot_command_line;
 
 	parse_early_param();

arch/riscv/kernel/smp.c

@@ -45,7 +45,7 @@ int setup_profiling_timer(unsigned int multiplier)
 	return -EINVAL;
 }
 
-irqreturn_t handle_ipi(void)
+void riscv_software_interrupt(void)
 {
 	unsigned long *pending_ipis = &ipi_data[smp_processor_id()].bits;
 
@@ -60,7 +60,7 @@ irqreturn_t handle_ipi(void)
 
 		ops = xchg(pending_ipis, 0);
 		if (ops == 0)
-			return IRQ_HANDLED;
+			return;
 
 		if (ops & (1 << IPI_RESCHEDULE))
 			scheduler_ipi();
@@ -73,8 +73,6 @@ irqreturn_t handle_ipi(void)
 		/* Order data access and bit testing. */
 		mb();
 	}
-
-	return IRQ_HANDLED;
 }
 
 static void

arch/riscv/kernel/smpboot.c

@@ -104,7 +104,6 @@ asmlinkage void __init smp_callin(void)
 	current->active_mm = mm;
 
 	trap_init();
-	init_clockevent();
 	notify_cpu_starting(smp_processor_id());
 	set_cpu_online(smp_processor_id(), 1);
 	local_flush_tlb_all();

arch/riscv/kernel/time.c

@@ -13,38 +13,11 @@
  */
 
 #include <linux/clocksource.h>
-#include <linux/clockchips.h>
 #include <linux/delay.h>
-
-#ifdef CONFIG_RISCV_TIMER
-#include <linux/timer_riscv.h>
-#endif
-
 #include <asm/sbi.h>
 
 unsigned long riscv_timebase;
 
-DECLARE_PER_CPU(struct clock_event_device, riscv_clock_event);
-
-void riscv_timer_interrupt(void)
-{
-#ifdef CONFIG_RISCV_TIMER
-	/*
-	 * FIXME: This needs to be cleaned up along with the rest of the IRQ
-	 * handling cleanup.  See irq.c for more details.
-	 */
-	struct clock_event_device *evdev = this_cpu_ptr(&riscv_clock_event);
-
-	evdev->event_handler(evdev);
-#endif
-}
-
-void __init init_clockevent(void)
-{
-	timer_probe();
-	csr_set(sie, SIE_STIE);
-}
-
 void __init time_init(void)
 {
 	struct device_node *cpu;
@@ -56,6 +29,5 @@ void __init time_init(void)
 	riscv_timebase = prop;
 
 	lpj_fine = riscv_timebase / HZ;
-
-	init_clockevent();
+	timer_probe();
 }

arch/riscv/kernel/traps.c

@@ -138,7 +138,6 @@ asmlinkage void do_trap_break(struct pt_regs *regs)
 #endif /* CONFIG_GENERIC_BUG */
 
 	force_sig_fault(SIGTRAP, TRAP_BRKPT, (void __user *)(regs->sepc), current);
-	regs->sepc += 0x4;
 }
 
 #ifdef CONFIG_GENERIC_BUG

arch/riscv/kernel/vdso/Makefile

@@ -52,8 +52,8 @@ $(obj)/%.so: $(obj)/%.so.dbg FORCE
 # Add -lgcc so rv32 gets static muldi3 and lshrdi3 definitions.
 # Make sure only to export the intended __vdso_xxx symbol offsets.
 quiet_cmd_vdsold = VDSOLD  $@
-      cmd_vdsold = $(CC) $(KCFLAGS) $(call cc-option, -no-pie) -nostdlib $(SYSCFLAGS_$(@F)) \
-                           -Wl,-T,$(filter-out FORCE,$^) -o $@.tmp -lgcc && \
+      cmd_vdsold = $(CC) $(KBUILD_CFLAGS) $(call cc-option, -no-pie) -nostdlib -nostartfiles $(SYSCFLAGS_$(@F)) \
+                           -Wl,-T,$(filter-out FORCE,$^) -o $@.tmp && \
                    $(CROSS_COMPILE)objcopy \
                            $(patsubst %, -G __vdso_%, $(vdso-syms)) $@.tmp $@
 

arch/riscv/lib/Makefile

@@ -2,5 +2,6 @@ lib-y	+= delay.o
 lib-y	+= memcpy.o
 lib-y	+= memset.o
 lib-y	+= uaccess.o
+lib-y	+= tishift.o
 
 lib-$(CONFIG_32BIT) += udivdi3.o

arch/riscv/lib/tishift.S

+/*
+ * Copyright (C) 2018 Free Software Foundation, Inc.
+ *
+ *   This program is free software; you can redistribute it and/or
+ *   modify it under the terms of the GNU General Public License
+ *   as published by the Free Software Foundation, version 2.
+ *
+ *   This program is distributed in the hope that it will be useful,
+ *   but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *   GNU General Public License for more details.
+ */
+  .globl __lshrti3
+__lshrti3:
+  beqz  a2, .L1
+  li    a5,64
+  sub   a5,a5,a2
+  addi  sp,sp,-16
+  sext.w a4,a5
+  blez  a5, .L2
+  sext.w a2,a2
+  sll   a4,a1,a4
+  srl   a0,a0,a2
+  srl   a1,a1,a2
+  or    a0,a0,a4
+  sd    a1,8(sp)
+  sd    a0,0(sp)
+  ld    a0,0(sp)
+  ld    a1,8(sp)
+  addi  sp,sp,16
+  ret
+.L1:
+  ret
+.L2:
+  negw  a4,a4
+  srl   a1,a1,a4
+  sd    a1,0(sp)
+  sd    zero,8(sp)
+  ld    a0,0(sp)
+  ld    a1,8(sp)
+  addi  sp,sp,16
+  ret

drivers/clocksource/Kconfig

@@ -609,4 +609,15 @@ config ATCPIT100_TIMER
 	help
 	  This option enables support for the Andestech ATCPIT100 timers.
 
+config RISCV_TIMER
+	bool "Timer for the RISC-V platform"
+	depends on RISCV
+	default y
+	select TIMER_PROBE
+	select TIMER_OF
+	help
+	  This enables the per-hart timer built into all RISC-V systems, which
+	  is accessed via both the SBI and the rdcycle instruction.  This is
+	  required for all RISC-V systems.
+
 endmenu

drivers/clocksource/Makefile

@@ -78,3 +78,4 @@ obj-$(CONFIG_H8300_TPU)			+= h8300_tpu.o
 obj-$(CONFIG_CLKSRC_ST_LPC)		+= clksrc_st_lpc.o
 obj-$(CONFIG_X86_NUMACHIP)		+= numachip.o
 obj-$(CONFIG_ATCPIT100_TIMER)		+= timer-atcpit100.o
+obj-$(CONFIG_RISCV_TIMER)		+= riscv_timer.o

drivers/clocksource/riscv_timer.c

+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2012 Regents of the University of California
+ * Copyright (C) 2017 SiFive
+ */
+#include <linux/clocksource.h>
+#include <linux/clockchips.h>
+#include <linux/cpu.h>
+#include <linux/delay.h>
+#include <linux/irq.h>
+#include <asm/sbi.h>
+
+/*
+ * All RISC-V systems have a timer attached to every hart.  These timers can be
+ * read by the 'rdcycle' pseudo instruction, and can use the SBI to setup
+ * events.  In order to abstract the architecture-specific timer reading and
+ * setting functions away from the clock event insertion code, we provide
+ * function pointers to the clockevent subsystem that perform two basic
+ * operations: rdtime() reads the timer on the current CPU, and
+ * next_event(delta) sets the next timer event to 'delta' cycles in the future.
+ * As the timers are inherently a per-cpu resource, these callbacks perform
+ * operations on the current hart.  There is guaranteed to be exactly one timer
+ * per hart on all RISC-V systems.
+ */
+
+static int riscv_clock_next_event(unsigned long delta,
+		struct clock_event_device *ce)
+{
+	csr_set(sie, SIE_STIE);
+	sbi_set_timer(get_cycles64() + delta);
+	return 0;
+}
+
+static DEFINE_PER_CPU(struct clock_event_device, riscv_clock_event) = {
+	.name			= "riscv_timer_clockevent",
+	.features		= CLOCK_EVT_FEAT_ONESHOT,
+	.rating			= 100,
+	.set_next_event		= riscv_clock_next_event,
+};
+
+/*
+ * It is guaranteed that all the timers across all the harts are synchronized
+ * within one tick of each other, so while this could technically go
+ * backwards when hopping between CPUs, practically it won't happen.
+ */
+static unsigned long long riscv_clocksource_rdtime(struct clocksource *cs)
+{
+	return get_cycles64();
+}
+
+static DEFINE_PER_CPU(struct clocksource, riscv_clocksource) = {
+	.name		= "riscv_clocksource",
+	.rating		= 300,
+	.mask		= CLOCKSOURCE_MASK(BITS_PER_LONG),
+	.flags		= CLOCK_SOURCE_IS_CONTINUOUS,
+	.read		= riscv_clocksource_rdtime,
+};
+
+static int riscv_timer_starting_cpu(unsigned int cpu)
+{
+	struct clock_event_device *ce = per_cpu_ptr(&riscv_clock_event, cpu);
+
+	ce->cpumask = cpumask_of(cpu);
+	clockevents_config_and_register(ce, riscv_timebase, 100, 0x7fffffff);
+
+	csr_set(sie, SIE_STIE);
+	return 0;
+}
+
+static int riscv_timer_dying_cpu(unsigned int cpu)
+{
+	csr_clear(sie, SIE_STIE);
+	return 0;
+}
+
+/* called directly from the low-level interrupt handler */
+void riscv_timer_interrupt(void)
+{
+	struct clock_event_device *evdev = this_cpu_ptr(&riscv_clock_event);
+
+	csr_clear(sie, SIE_STIE);
+	evdev->event_handler(evdev);
+}
+
+static int __init riscv_timer_init_dt(struct device_node *n)
+{
+	int cpu_id = riscv_of_processor_hart(n), error;
+	struct clocksource *cs;
+
+	if (cpu_id != smp_processor_id())
+		return 0;
+
+	cs = per_cpu_ptr(&riscv_clocksource, cpu_id);
+	clocksource_register_hz(cs, riscv_timebase);
+
+	error = cpuhp_setup_state(CPUHP_AP_RISCV_TIMER_STARTING,
+			 "clockevents/riscv/timer:starting",
+			 riscv_timer_starting_cpu, riscv_timer_dying_cpu);
+	if (error)
+		pr_err("RISCV timer register failed [%d] for cpu = [%d]\n",
+		       error, cpu_id);
+	return error;
+}
+
+TIMER_OF_DECLARE(riscv_timer, "riscv", riscv_timer_init_dt);

drivers/irqchip/Kconfig

@@ -372,3 +372,15 @@ config QCOM_PDC
 	  IRQs for Qualcomm Technologies Inc (QTI) mobile chips.
 
 endmenu
+
+config SIFIVE_PLIC
+	bool "SiFive Platform-Level Interrupt Controller"
+	depends on RISCV
+	help
+	   This enables support for the PLIC chip found in SiFive (and
+	   potentially other) RISC-V systems.  The PLIC controls devices
+	   interrupts and connects them to each core's local interrupt
+	   controller.  Aside from timer and software interrupts, all other
+	   interrupt sources are subordinate to the PLIC.
+
+	   If you don't know what to do here, say Y.

drivers/irqchip/Makefile

@@ -87,3 +87,4 @@ obj-$(CONFIG_MESON_IRQ_GPIO)		+= irq-meson-gpio.o
 obj-$(CONFIG_GOLDFISH_PIC) 		+= irq-goldfish-pic.o
 obj-$(CONFIG_NDS32)			+= irq-ativic32.o
 obj-$(CONFIG_QCOM_PDC)			+= qcom-pdc.o
+obj-$(CONFIG_SIFIVE_PLIC)		+= irq-sifive-plic.o

drivers/irqchip/irq-sifive-plic.c

+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2017 SiFive
+ * Copyright (C) 2018 Christoph Hellwig
+ */
+#define pr_fmt(fmt) "plic: " fmt
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/irq.h>
+#include <linux/irqchip.h>
+#include <linux/irqdomain.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <linux/of_irq.h>
+#include <linux/platform_device.h>
+#include <linux/spinlock.h>
+
+/*
+ * This driver implements a version of the RISC-V PLIC with the actual layout
+ * specified in chapter 8 of the SiFive U5 Coreplex Series Manual:
+ *
+ *     https://static.dev.sifive.com/U54-MC-RVCoreIP.pdf
+ *
+ * The largest number supported by devices marked as 'sifive,plic-1.0.0', is
+ * 1024, of which device 0 is defined as non-existent by the RISC-V Privileged
+ * Spec.
+ */
+
+#define MAX_DEVICES			1024
+#define MAX_CONTEXTS			15872
+
+/*
+ * Each interrupt source has a priority register associated with it.
+ * We always hardwire it to one in Linux.
+ */
+#define PRIORITY_BASE			0
+#define     PRIORITY_PER_ID		4
+
+/*
+ * Each hart context has a vector of interrupt enable bits associated with it.
+ * There's one bit for each interrupt source.
+ */
+#define ENABLE_BASE			0x2000
+#define     ENABLE_PER_HART		0x80
+
+/*
+ * Each hart context has a set of control registers associated with it.  Right
+ * now there's only two: a source priority threshold over which the hart will
+ * take an interrupt, and a register to claim interrupts.
+ */
+#define CONTEXT_BASE			0x200000
+#define     CONTEXT_PER_HART		0x1000
+#define     CONTEXT_THRESHOLD		0x00
+#define     CONTEXT_CLAIM		0x04
+
+static void __iomem *plic_regs;
+
+struct plic_handler {
+	bool			present;
+	int			ctxid;
+};
+static DEFINE_PER_CPU(struct plic_handler, plic_handlers);
+
+static inline void __iomem *plic_hart_offset(int ctxid)
+{
+	return plic_regs + CONTEXT_BASE + ctxid * CONTEXT_PER_HART;
+}
+
+static inline u32 __iomem *plic_enable_base(int ctxid)
+{
+	return plic_regs + ENABLE_BASE + ctxid * ENABLE_PER_HART;
+}
+
+/*
+ * Protect mask operations on the registers given that we can't assume that
+ * atomic memory operations work on them.
+ */
+static DEFINE_RAW_SPINLOCK(plic_toggle_lock);
+
+static inline void plic_toggle(int ctxid, int hwirq, int enable)
+{
+	u32 __iomem *reg = plic_enable_base(ctxid) + (hwirq / 32);
+	u32 hwirq_mask = 1 << (hwirq % 32);
+
+	raw_spin_lock(&plic_toggle_lock);
+	if (enable)
+		writel(readl(reg) | hwirq_mask, reg);
+	else
+		writel(readl(reg) & ~hwirq_mask, reg);
+	raw_spin_unlock(&plic_toggle_lock);
+}
+
+static inline void plic_irq_toggle(struct irq_data *d, int enable)
+{
+	int cpu;
+
+	writel(enable, plic_regs + PRIORITY_BASE + d->hwirq * PRIORITY_PER_ID);
+	for_each_cpu(cpu, irq_data_get_affinity_mask(d)) {
+		struct plic_handler *handler = per_cpu_ptr(&plic_handlers, cpu);
+
+		if (handler->present)
+			plic_toggle(handler->ctxid, d->hwirq, enable);
+	}
+}
+
+static void plic_irq_enable(struct irq_data *d)
+{
+	plic_irq_toggle(d, 1);
+}
+
+static void plic_irq_disable(struct irq_data *d)
+{
+	plic_irq_toggle(d, 0);
+}
+
+static struct irq_chip plic_chip = {
+	.name		= "SiFive PLIC",
+	/*
+	 * There is no need to mask/unmask PLIC interrupts.  They are "masked"
+	 * by reading claim and "unmasked" when writing it back.
+	 */
+	.irq_enable	= plic_irq_enable,
+	.irq_disable	= plic_irq_disable,
+};
+
+static int plic_irqdomain_map(struct irq_domain *d, unsigned int irq,
+			      irq_hw_number_t hwirq)
+{
+	irq_set_chip_and_handler(irq, &plic_chip, handle_simple_irq);
+	irq_set_chip_data(irq, NULL);
+	irq_set_noprobe(irq);
+	return 0;
+}
+
+static const struct irq_domain_ops plic_irqdomain_ops = {
+	.map		= plic_irqdomain_map,
+	.xlate		= irq_domain_xlate_onecell,
+};
+
+static struct irq_domain *plic_irqdomain;
+
+/*
+ * Handling an interrupt is a two-step process: first you claim the interrupt
+ * by reading the claim register, then you complete the interrupt by writing
+ * that source ID back to the same claim register.  This automatically enables
+ * and disables the interrupt, so there's nothing else to do.
+ */
+static void plic_handle_irq(struct pt_regs *regs)
+{
+	struct plic_handler *handler = this_cpu_ptr(&plic_handlers);
+	void __iomem *claim = plic_hart_offset(handler->ctxid) + CONTEXT_CLAIM;
+	irq_hw_number_t hwirq;
+
+	WARN_ON_ONCE(!handler->present);
+
+	csr_clear(sie, SIE_SEIE);
+	while ((hwirq = readl(claim))) {
+		int irq = irq_find_mapping(plic_irqdomain, hwirq);
+
+		if (unlikely(irq <= 0))
+			pr_warn_ratelimited("can't find mapping for hwirq %lu\n",
+					hwirq);
+		else
+			generic_handle_irq(irq);
+		writel(hwirq, claim);
+	}
+	csr_set(sie, SIE_SEIE);
+}
+
+/*
+ * Walk up the DT tree until we find an active RISC-V core (HART) node and
+ * extract the cpuid from it.
+ */
+static int plic_find_hart_id(struct device_node *node)
+{
+	for (; node; node = node->parent) {
+		if (of_device_is_compatible(node, "riscv"))
+			return riscv_of_processor_hart(node);
+	}
+
+	return -1;
+}
+
+static int __init plic_init(struct device_node *node,
+		struct device_node *parent)
+{
+	int error = 0, nr_handlers, nr_mapped = 0, i;
+	u32 nr_irqs;
+
+	if (plic_regs) {
+		pr_warn("PLIC already present.\n");
+		return -ENXIO;
+	}
+
+	plic_regs = of_iomap(node, 0);
+	if (WARN_ON(!plic_regs))
+		return -EIO;
+
+	error = -EINVAL;
+	of_property_read_u32(node, "riscv,ndev", &nr_irqs);
+	if (WARN_ON(!nr_irqs))
+		goto out_iounmap;
+
+	nr_handlers = of_irq_count(node);
+	if (WARN_ON(!nr_handlers))
+		goto out_iounmap;
+	if (WARN_ON(nr_handlers < num_possible_cpus()))
+		goto out_iounmap;
+
+	error = -ENOMEM;
+	plic_irqdomain = irq_domain_add_linear(node, nr_irqs + 1,
+			&plic_irqdomain_ops, NULL);
+	if (WARN_ON(!plic_irqdomain))
+		goto out_iounmap;
+
+	for (i = 0; i < nr_handlers; i++) {
+		struct of_phandle_args parent;
+		struct plic_handler *handler;
+		irq_hw_number_t hwirq;
+		int cpu;
+
+		if (of_irq_parse_one(node, i, &parent)) {
+			pr_err("failed to parse parent for context %d.\n", i);
+			continue;
+		}
+
+		/* skip context holes */
+		if (parent.args[0] == -1)
+			continue;
+
+		cpu = plic_find_hart_id(parent.np);
+		if (cpu < 0) {
+			pr_warn("failed to parse hart ID for context %d.\n", i);
+			continue;
+		}
+
+		handler = per_cpu_ptr(&plic_handlers, cpu);
+		handler->present = true;
+		handler->ctxid = i;
+
+		/* priority must be > threshold to trigger an interrupt */
+		writel(0, plic_hart_offset(i) + CONTEXT_THRESHOLD);
+		for (hwirq = 1; hwirq <= nr_irqs; hwirq++)
+			plic_toggle(i, hwirq, 0);
+		nr_mapped++;
+	}
+
+	pr_info("mapped %d interrupts to %d (out of %d) handlers.\n",
+		nr_irqs, nr_mapped, nr_handlers);
+	set_handle_irq(plic_handle_irq);
+	return 0;
+
+out_iounmap:
+	iounmap(plic_regs);
+	return error;
+}
+
+IRQCHIP_DECLARE(sifive_plic, "sifive,plic-1.0.0", plic_init);
+IRQCHIP_DECLARE(riscv_plic0, "riscv,plic0", plic_init); /* for legacy systems */

include/linux/cpuhotplug.h

@@ -125,6 +125,7 @@ enum cpuhp_state {
 	CPUHP_AP_MARCO_TIMER_STARTING,
 	CPUHP_AP_MIPS_GIC_TIMER_STARTING,
 	CPUHP_AP_ARC_TIMER_STARTING,
+	CPUHP_AP_RISCV_TIMER_STARTING,
 	CPUHP_AP_KVM_STARTING,
 	CPUHP_AP_KVM_ARM_VGIC_INIT_STARTING,
 	CPUHP_AP_KVM_ARM_VGIC_STARTING,