Merge git://git.kernel.org/pub/scm/linux/kernel/git/cmetcalf/linux-tile

Pull arch/tile updates from Chris Metcalf:
 "These fix a few stray build issues seen in linux-next, and also add
  the minimal required support for perf to tilegx"

* git://git.kernel.org/pub/scm/linux/kernel/git/cmetcalf/linux-tile:
  arch/tile: remove unused variable 'devcap'
  tile: Fix vDSO compilation issue with allyesconfig
  perf tools: Allow building for tile
  tile/perf: Support perf_events on tilegx and tilepro
  tile: Enable NMIs on return from handle_nmi() without errors
  tile: Add support for handling PMC hardware
  tile: don't use __get_cpu_var() with structure-typed arguments
  tile: avoid overflow in ns2cycles

arch/tile/Kconfig

@@ -3,6 +3,8 @@
 
 config TILE
 	def_bool y
+	select HAVE_PERF_EVENTS
+	select USE_PMC if PERF_EVENTS
 	select HAVE_DMA_ATTRS
 	select HAVE_DMA_API_DEBUG
 	select HAVE_KVM if !TILEGX
@@ -66,6 +68,10 @@ config HUGETLB_SUPER_PAGES
 config GENERIC_TIME_VSYSCALL
 	def_bool y
 
+# Enable PMC if PERF_EVENTS, OPROFILE, or WATCHPOINTS are enabled.
+config USE_PMC
+	bool
+
 # FIXME: tilegx can implement a more efficient rwsem.
 config RWSEM_GENERIC_SPINLOCK
 	def_bool y

arch/tile/include/asm/perf_event.h

@@ -0,0 +1,22 @@
+/*
+ * Copyright 2014 Tilera Corporation. All Rights Reserved.
+ *
+ *   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, GOOD TITLE or
+ *   NON INFRINGEMENT.  See the GNU General Public License for
+ *   more details.
+ */
+
+#ifndef _ASM_TILE_PERF_EVENT_H
+#define _ASM_TILE_PERF_EVENT_H
+
+#include <linux/percpu.h>
+DECLARE_PER_CPU(u64, perf_irqs);
+
+unsigned long handle_syscall_link_address(void);
+#endif /* _ASM_TILE_PERF_EVENT_H */

arch/tile/include/asm/pmc.h

@@ -0,0 +1,64 @@
+/*
+ * Copyright 2014 Tilera Corporation. All Rights Reserved.
+ *
+ *   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, GOOD TITLE or
+ *   NON INFRINGEMENT.  See the GNU General Public License for
+ *   more details.
+ */
+
+#ifndef _ASM_TILE_PMC_H
+#define _ASM_TILE_PMC_H
+
+#include <linux/ptrace.h>
+
+#define TILE_BASE_COUNTERS	2
+
+/* Bitfields below are derived from SPR PERF_COUNT_CTL*/
+#ifndef __tilegx__
+/* PERF_COUNT_CTL on TILEPro */
+#define TILE_CTL_EXCL_USER	(1 << 7) /* exclude user level */
+#define TILE_CTL_EXCL_KERNEL	(1 << 8) /* exclude kernel level */
+#define TILE_CTL_EXCL_HV	(1 << 9) /* exclude hypervisor level */
+
+#define TILE_SEL_MASK		0x7f	/* 7 bits for event SEL,
+					COUNT_0_SEL */
+#define TILE_PLM_MASK		0x780	/* 4 bits priv level msks,
+					COUNT_0_MASK*/
+#define TILE_EVENT_MASK	(TILE_SEL_MASK | TILE_PLM_MASK)
+
+#else /* __tilegx__*/
+/* PERF_COUNT_CTL on TILEGx*/
+#define TILE_CTL_EXCL_USER	(1 << 10) /* exclude user level */
+#define TILE_CTL_EXCL_KERNEL	(1 << 11) /* exclude kernel level */
+#define TILE_CTL_EXCL_HV	(1 << 12) /* exclude hypervisor level */
+
+#define TILE_SEL_MASK		0x3f	/* 6 bits for event SEL,
+					COUNT_0_SEL*/
+#define TILE_BOX_MASK		0x1c0	/* 3 bits box msks,
+					COUNT_0_BOX */
+#define TILE_PLM_MASK		0x3c00	/* 4 bits priv level msks,
+					COUNT_0_MASK */
+#define TILE_EVENT_MASK	(TILE_SEL_MASK | TILE_BOX_MASK | TILE_PLM_MASK)
+#endif /* __tilegx__*/
+
+/* Takes register and fault number.  Returns error to disable the interrupt. */
+typedef int (*perf_irq_t)(struct pt_regs *, int);
+
+int userspace_perf_handler(struct pt_regs *regs, int fault);
+
+perf_irq_t reserve_pmc_hardware(perf_irq_t new_perf_irq);
+void release_pmc_hardware(void);
+
+unsigned long pmc_get_overflow(void);
+void pmc_ack_overflow(unsigned long status);
+
+void unmask_pmc_interrupts(void);
+void mask_pmc_interrupts(void);
+
+#endif /* _ASM_TILE_PMC_H */

arch/tile/kernel/Makefile

@@ -25,6 +25,8 @@ obj-$(CONFIG_PCI)		+= pci_gx.o
 else
 obj-$(CONFIG_PCI)		+= pci.o
 endif
+obj-$(CONFIG_PERF_EVENTS)	+= perf_event.o
+obj-$(CONFIG_USE_PMC)		+= pmc.o
 obj-$(CONFIG_TILE_USB)		+= usb.o
 obj-$(CONFIG_TILE_HVGLUE_TRACE)	+= hvglue_trace.o
 obj-$(CONFIG_FUNCTION_TRACER)	+= ftrace.o mcount_64.o

arch/tile/kernel/intvec_32.S

@@ -313,13 +313,13 @@ intvec_\vecname:
 	 movei  r3, 0
 	}
 	.else
-	.ifc \c_routine, op_handle_perf_interrupt
+	.ifc \c_routine, handle_perf_interrupt
 	{
 	 mfspr  r2, PERF_COUNT_STS
 	 movei  r3, -1   /* not used, but set for consistency */
 	}
 	.else
-	.ifc \c_routine, op_handle_aux_perf_interrupt
+	.ifc \c_routine, handle_perf_interrupt
 	{
 	 mfspr  r2, AUX_PERF_COUNT_STS
 	 movei  r3, -1   /* not used, but set for consistency */
@@ -946,6 +946,13 @@ STD_ENTRY(interrupt_return)
 	bzt     r30, .Lrestore_regs
 3:
 
+	/* We are relying on INT_PERF_COUNT at 33, and AUX_PERF_COUNT at 48 */
+	{
+	 moveli r0, lo16(1 << (INT_PERF_COUNT - 32))
+	 bz     r31, .Lrestore_regs
+	}
+	auli    r0, r0, ha16(1 << (INT_AUX_PERF_COUNT - 32))
+	mtspr   SPR_INTERRUPT_MASK_RESET_K_1, r0
 
 	/*
 	 * We now commit to returning from this interrupt, since we will be
@@ -1171,6 +1178,10 @@ handle_nmi:
 	 PTREGS_PTR(r0, PTREGS_OFFSET_BASE)
 	}
 	FEEDBACK_REENTER(handle_nmi)
+	{
+	 movei  r30, 1
+	 seq    r31, r0, zero
+	}
 	j       interrupt_return
 	STD_ENDPROC(handle_nmi)
 
@@ -1835,8 +1846,9 @@ int_unalign:
 /* Include .intrpt array of interrupt vectors */
 	.section ".intrpt", "ax"
 
-#define op_handle_perf_interrupt bad_intr
-#define op_handle_aux_perf_interrupt bad_intr
+#ifndef CONFIG_USE_PMC
+#define handle_perf_interrupt bad_intr
+#endif
 
 #ifndef CONFIG_HARDWALL
 #define do_hardwall_trap bad_intr
@@ -1877,7 +1889,7 @@ int_unalign:
 	int_hand     INT_IDN_AVAIL, IDN_AVAIL, bad_intr
 	int_hand     INT_UDN_AVAIL, UDN_AVAIL, bad_intr
 	int_hand     INT_PERF_COUNT, PERF_COUNT, \
-		     op_handle_perf_interrupt, handle_nmi
+		     handle_perf_interrupt, handle_nmi
 	int_hand     INT_INTCTRL_3, INTCTRL_3, bad_intr
 #if CONFIG_KERNEL_PL == 2
 	dc_dispatch  INT_INTCTRL_2, INTCTRL_2
@@ -1902,7 +1914,7 @@ int_unalign:
 	int_hand     INT_SN_CPL, SN_CPL, bad_intr
 	int_hand     INT_DOUBLE_FAULT, DOUBLE_FAULT, do_trap
 	int_hand     INT_AUX_PERF_COUNT, AUX_PERF_COUNT, \
-		     op_handle_aux_perf_interrupt, handle_nmi
+		     handle_perf_interrupt, handle_nmi
 
 	/* Synthetic interrupt delivered only by the simulator */
 	int_hand     INT_BREAKPOINT, BREAKPOINT, do_breakpoint

arch/tile/kernel/intvec_64.S

@@ -509,10 +509,10 @@ intvec_\vecname:
 	.ifc \c_routine, do_trap
 	mfspr   r2, GPV_REASON
 	.else
-	.ifc \c_routine, op_handle_perf_interrupt
+	.ifc \c_routine, handle_perf_interrupt
 	mfspr   r2, PERF_COUNT_STS
 	.else
-	.ifc \c_routine, op_handle_aux_perf_interrupt
+	.ifc \c_routine, handle_perf_interrupt
 	mfspr   r2, AUX_PERF_COUNT_STS
 	.endif
 	.endif
@@ -971,6 +971,15 @@ STD_ENTRY(interrupt_return)
 	beqzt   r30, .Lrestore_regs
 3:
 
+#if INT_PERF_COUNT + 1 != INT_AUX_PERF_COUNT
+# error Bad interrupt assumption
+#endif
+	{
+	 movei  r0, 3   /* two adjacent bits for the PERF_COUNT mask */
+	 beqz   r31, .Lrestore_regs
+	}
+	shli    r0, r0, INT_PERF_COUNT
+	mtspr   SPR_INTERRUPT_MASK_RESET_K, r0
 
 	/*
 	 * We now commit to returning from this interrupt, since we will be
@@ -1187,7 +1196,7 @@ handle_nmi:
 	FEEDBACK_REENTER(handle_nmi)
 	{
 	 movei  r30, 1
-	 move   r31, r0
+	 cmpeq  r31, r0, zero
 	}
 	j       interrupt_return
 	STD_ENDPROC(handle_nmi)
@@ -1491,8 +1500,9 @@ STD_ENTRY(fill_ra_stack)
 	.global intrpt_start
 intrpt_start:
 
-#define op_handle_perf_interrupt bad_intr
-#define op_handle_aux_perf_interrupt bad_intr
+#ifndef CONFIG_USE_PMC
+#define handle_perf_interrupt bad_intr
+#endif
 
 #ifndef CONFIG_HARDWALL
 #define do_hardwall_trap bad_intr
@@ -1540,9 +1550,9 @@ intrpt_start:
 #endif
 	int_hand     INT_IPI_0, IPI_0, bad_intr
 	int_hand     INT_PERF_COUNT, PERF_COUNT, \
-		     op_handle_perf_interrupt, handle_nmi
+		     handle_perf_interrupt, handle_nmi
 	int_hand     INT_AUX_PERF_COUNT, AUX_PERF_COUNT, \
-		     op_handle_perf_interrupt, handle_nmi
+		     handle_perf_interrupt, handle_nmi
 	int_hand     INT_INTCTRL_3, INTCTRL_3, bad_intr
 #if CONFIG_KERNEL_PL == 2
 	dc_dispatch  INT_INTCTRL_2, INTCTRL_2

arch/tile/kernel/irq.c

@@ -21,6 +21,7 @@
 #include <hv/drv_pcie_rc_intf.h>
 #include <arch/spr_def.h>
 #include <asm/traps.h>
+#include <linux/perf_event.h>
 
 /* Bit-flag stored in irq_desc->chip_data to indicate HW-cleared irqs. */
 #define IS_HW_CLEARED 1
@@ -260,6 +261,23 @@ void ack_bad_irq(unsigned int irq)
 	pr_err("unexpected IRQ trap at vector %02x\n", irq);
 }
 
+/*
+ * /proc/interrupts printing:
+ */
+int arch_show_interrupts(struct seq_file *p, int prec)
+{
+#ifdef CONFIG_PERF_EVENTS
+	int i;
+
+	seq_printf(p, "%*s: ", prec, "PMI");
+
+	for_each_online_cpu(i)
+		seq_printf(p, "%10llu ", per_cpu(perf_irqs, i));
+	seq_puts(p, "  perf_events\n");
+#endif
+	return 0;
+}
+
 /*
  * Generic, controller-independent functions:
  */

arch/tile/kernel/messaging.c

@@ -68,8 +68,8 @@ void hv_message_intr(struct pt_regs *regs, int intnum)
 #endif
 
 	while (1) {
-		rmi = hv_receive_message(__get_cpu_var(msg_state),
-					 (HV_VirtAddr) message,
+		HV_MsgState *state = this_cpu_ptr(&msg_state);
+		rmi = hv_receive_message(*state, (HV_VirtAddr) message,
 					 sizeof(message));
 		if (rmi.msglen == 0)
 			break;

arch/tile/kernel/pci.c

@@ -250,8 +250,6 @@ static void fixup_read_and_payload_sizes(void)
 
 	/* Scan for the smallest maximum payload size. */
 	for_each_pci_dev(dev) {
-		u32 devcap;
-
 		if (!pci_is_pcie(dev))
 			continue;
 

arch/tile/kernel/pmc.c

@@ -0,0 +1,121 @@
+/*
+ * Copyright 2014 Tilera Corporation. All Rights Reserved.
+ *
+ *   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, GOOD TITLE or
+ *   NON INFRINGEMENT.  See the GNU General Public License for
+ *   more details.
+ */
+
+#include <linux/errno.h>
+#include <linux/spinlock.h>
+#include <linux/module.h>
+#include <linux/atomic.h>
+#include <linux/interrupt.h>
+
+#include <asm/processor.h>
+#include <asm/pmc.h>
+
+perf_irq_t perf_irq = NULL;
+int handle_perf_interrupt(struct pt_regs *regs, int fault)
+{
+	int retval;
+
+	if (!perf_irq)
+		panic("Unexpected PERF_COUNT interrupt %d\n", fault);
+
+	nmi_enter();
+	retval = perf_irq(regs, fault);
+	nmi_exit();
+	return retval;
+}
+
+/* Reserve PMC hardware if it is available. */
+perf_irq_t reserve_pmc_hardware(perf_irq_t new_perf_irq)
+{
+	return cmpxchg(&perf_irq, NULL, new_perf_irq);
+}
+EXPORT_SYMBOL(reserve_pmc_hardware);
+
+/* Release PMC hardware. */
+void release_pmc_hardware(void)
+{
+	perf_irq = NULL;
+}
+EXPORT_SYMBOL(release_pmc_hardware);
+
+
+/*
+ * Get current overflow status of each performance counter,
+ * and auxiliary performance counter.
+ */
+unsigned long
+pmc_get_overflow(void)
+{
+	unsigned long status;
+
+	/*
+	 * merge base+aux into a single vector
+	 */
+	status = __insn_mfspr(SPR_PERF_COUNT_STS);
+	status |= __insn_mfspr(SPR_AUX_PERF_COUNT_STS) << TILE_BASE_COUNTERS;
+	return status;
+}
+
+/*
+ * Clear the status bit for the corresponding counter, if written
+ * with a one.
+ */
+void
+pmc_ack_overflow(unsigned long status)
+{
+	/*
+	 * clear overflow status by writing ones
+	 */
+	__insn_mtspr(SPR_PERF_COUNT_STS, status);
+	__insn_mtspr(SPR_AUX_PERF_COUNT_STS, status >> TILE_BASE_COUNTERS);
+}
+
+/*
+ * The perf count interrupts are masked and unmasked explicitly,
+ * and only here.  The normal irq_enable() does not enable them,
+ * and irq_disable() does not disable them.  That lets these
+ * routines drive the perf count interrupts orthogonally.
+ *
+ * We also mask the perf count interrupts on entry to the perf count
+ * interrupt handler in assembly code, and by default unmask them
+ * again (with interrupt critical section protection) just before
+ * returning from the interrupt.  If the perf count handler returns
+ * a non-zero error code, then we don't re-enable them before returning.
+ *
+ * For Pro, we rely on both interrupts being in the same word to update
+ * them atomically so we never have one enabled and one disabled.
+ */
+
+#if CHIP_HAS_SPLIT_INTR_MASK()
+# if INT_PERF_COUNT < 32 || INT_AUX_PERF_COUNT < 32
+#  error Fix assumptions about which word PERF_COUNT interrupts are in
+# endif
+#endif
+
+static inline unsigned long long pmc_mask(void)
+{
+	unsigned long long mask = 1ULL << INT_PERF_COUNT;
+	mask |= 1ULL << INT_AUX_PERF_COUNT;
+	return mask;
+}
+
+void unmask_pmc_interrupts(void)
+{
+	interrupt_mask_reset_mask(pmc_mask());
+}
+
+void mask_pmc_interrupts(void)
+{
+	interrupt_mask_set_mask(pmc_mask());
+}

arch/tile/kernel/time.c

@@ -236,7 +236,15 @@ cycles_t ns2cycles(unsigned long nsecs)
 	 * clock frequency.
 	 */
 	struct clock_event_device *dev = &__raw_get_cpu_var(tile_timer);
-	return ((u64)nsecs * dev->mult) >> dev->shift;
+
+	/*
+	 * as in clocksource.h and x86's timer.h, we split the calculation
+	 * into 2 parts to avoid unecessary overflow of the intermediate
+	 * value. This will not lead to any loss of precision.
+	 */
+	u64 quot = (u64)nsecs >> dev->shift;
+	u64 rem  = (u64)nsecs & ((1ULL << dev->shift) - 1);
+	return quot * dev->mult + ((rem * dev->mult) >> dev->shift);
 }
 
 void update_vsyscall_tz(void)

arch/tile/kernel/vdso/Makefile

@@ -104,7 +104,7 @@ $(obj-vdso32:%=%): KBUILD_AFLAGS = $(KBUILD_AFLAGS_32)
 $(obj-vdso32:%=%): KBUILD_CFLAGS = $(KBUILD_CFLAGS_32)
 
 $(obj)/vgettimeofday32.o: $(obj)/vgettimeofday.c
-	$(call if_changed,cc_o_c)
+	$(call if_changed_rule,cc_o_c)
 
 $(obj)/vrt_sigreturn32.o: $(obj)/vrt_sigreturn.S
 	$(call if_changed,as_o_S)

tools/perf/config/Makefile.arch

@@ -5,7 +5,8 @@ ARCH ?= $(shell echo $(uname_M) | sed -e s/i.86/i386/ -e s/sun4u/sparc64/ \
                                   -e s/arm.*/arm/ -e s/sa110/arm/ \
                                   -e s/s390x/s390/ -e s/parisc64/parisc/ \
                                   -e s/ppc.*/powerpc/ -e s/mips.*/mips/ \
-                                  -e s/sh[234].*/sh/ -e s/aarch64.*/arm64/ )
+                                  -e s/sh[234].*/sh/ -e s/aarch64.*/arm64/ \
+                                  -e s/tile.*/tile/ )
 
 # Additional ARCH settings for x86
 ifeq ($(ARCH),i386)

tools/perf/perf.h

@@ -145,6 +145,14 @@
 #define CPUINFO_PROC	"core ID"
 #endif
 
+#ifdef __tile__
+#define mb()		asm volatile ("mf" ::: "memory")
+#define wmb()		asm volatile ("mf" ::: "memory")
+#define rmb()		asm volatile ("mf" ::: "memory")
+#define cpu_relax()	asm volatile ("mfspr zero, PASS" ::: "memory")
+#define CPUINFO_PROC    "model name"
+#endif
+
 #define barrier() asm volatile ("" ::: "memory")
 
 #ifndef cpu_relax