Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net

32 files changed, 232 insertions, 166 deletions

diff --git a/arch/arm/boot/dts/am335x-bone-common.dtsi b/arch/arm/boot/dts/am335x-bone-common.dtsi
index c3255e0c90aa..dbb3f4d2bf84 100644
--- a/arch/arm/boot/dts/am335x-bone-common.dtsi
+++ b/arch/arm/boot/dts/am335x-bone-common.dtsi
@@ -223,6 +223,25 @@
 /include/ "tps65217.dtsi"
 
 &tps {
+	/*
+	 * Configure pmic to enter OFF-state instead of SLEEP-state ("RTC-only
+	 * mode") at poweroff.  Most BeagleBone versions do not support RTC-only
+	 * mode and risk hardware damage if this mode is entered.
+	 *
+	 * For details, see linux-omap mailing list May 2015 thread
+	 *	[PATCH] ARM: dts: am335x-bone* enable pmic-shutdown-controller
+	 * In particular, messages:
+	 *	http://www.spinics.net/lists/linux-omap/msg118585.html
+	 *	http://www.spinics.net/lists/linux-omap/msg118615.html
+	 *
+	 * You can override this later with
+	 *	&tps {  /delete-property/ ti,pmic-shutdown-controller;  }
+	 * if you want to use RTC-only mode and made sure you are not affected
+	 * by the hardware problems. (Tip: double-check by performing a current
+	 * measurement after shutdown: it should be less than 1 mA.)
+	 */
+	ti,pmic-shutdown-controller;
+
 	regulators {
 		dcdc1_reg: regulator@0 {
 			regulator-name = "vdds_dpr";
diff --git a/arch/arm/boot/dts/am35xx-clocks.dtsi b/arch/arm/boot/dts/am35xx-clocks.dtsi
index 518b8fde88b0..18cc826e9db5 100644
--- a/arch/arm/boot/dts/am35xx-clocks.dtsi
+++ b/arch/arm/boot/dts/am35xx-clocks.dtsi
@@ -12,7 +12,7 @@
 		#clock-cells = <0>;
 		compatible = "ti,am35xx-gate-clock";
 		clocks = <&ipss_ick>;
-		reg = <0x059c>;
+		reg = <0x032c>;
 		ti,bit-shift = <1>;
 	};
 
@@ -20,7 +20,7 @@
 		#clock-cells = <0>;
 		compatible = "ti,gate-clock";
 		clocks = <&rmii_ck>;
-		reg = <0x059c>;
+		reg = <0x032c>;
 		ti,bit-shift = <9>;
 	};
 
@@ -28,7 +28,7 @@
 		#clock-cells = <0>;
 		compatible = "ti,am35xx-gate-clock";
 		clocks = <&ipss_ick>;
-		reg = <0x059c>;
+		reg = <0x032c>;
 		ti,bit-shift = <2>;
 	};
 
@@ -36,7 +36,7 @@
 		#clock-cells = <0>;
 		compatible = "ti,gate-clock";
 		clocks = <&pclk_ck>;
-		reg = <0x059c>;
+		reg = <0x032c>;
 		ti,bit-shift = <10>;
 	};
 
@@ -44,7 +44,7 @@
 		#clock-cells = <0>;
 		compatible = "ti,am35xx-gate-clock";
 		clocks = <&ipss_ick>;
-		reg = <0x059c>;
+		reg = <0x032c>;
 		ti,bit-shift = <0>;
 	};
 
@@ -52,7 +52,7 @@
 		#clock-cells = <0>;
 		compatible = "ti,gate-clock";
 		clocks = <&sys_ck>;
-		reg = <0x059c>;
+		reg = <0x032c>;
 		ti,bit-shift = <8>;
 	};
 
@@ -60,7 +60,7 @@
 		#clock-cells = <0>;
 		compatible = "ti,am35xx-gate-clock";
 		clocks = <&sys_ck>;
-		reg = <0x059c>;
+		reg = <0x032c>;
 		ti,bit-shift = <3>;
 	};
 };
diff --git a/arch/arm/boot/dts/armada-xp-linksys-mamba.dts b/arch/arm/boot/dts/armada-xp-linksys-mamba.dts
index a2cf2154dcdb..fdd187c55aa5 100644
--- a/arch/arm/boot/dts/armada-xp-linksys-mamba.dts
+++ b/arch/arm/boot/dts/armada-xp-linksys-mamba.dts
@@ -95,6 +95,11 @@
 
 		internal-regs {
 
+			rtc@10300 {
+				/* No crystal connected to the internal RTC */
+				status = "disabled";
+			};
+
 			/* J10: VCC, NC, RX, NC, TX, GND  */
 			serial@12000 {
 				status = "okay";
diff --git a/arch/arm/boot/dts/dm816x.dtsi b/arch/arm/boot/dts/dm816x.dtsi
index de8427be830a..289806adb343 100644
--- a/arch/arm/boot/dts/dm816x.dtsi
+++ b/arch/arm/boot/dts/dm816x.dtsi
@@ -382,7 +382,7 @@
 			ti,hwmods = "usb_otg_hs";
 
 			usb0: usb@47401000 {
-				compatible = "ti,musb-am33xx";
+				compatible = "ti,musb-dm816";
 				reg = <0x47401400 0x400
 				       0x47401000 0x200>;
 				reg-names = "mc", "control";
@@ -422,7 +422,7 @@
 			};
 
 			usb1: usb@47401800 {
-				compatible = "ti,musb-am33xx";
+				compatible = "ti,musb-dm816";
 				reg = <0x47401c00 0x400
 				       0x47401800 0x200>;
 				reg-names = "mc", "control";
diff --git a/arch/arm/boot/dts/omap3-n900.dts b/arch/arm/boot/dts/omap3-n900.dts
index 5c16145920ea..5f5e0f3d5b64 100644
--- a/arch/arm/boot/dts/omap3-n900.dts
+++ b/arch/arm/boot/dts/omap3-n900.dts
@@ -832,8 +832,8 @@
 		touchscreen-fuzz-x = <4>;
 		touchscreen-fuzz-y = <7>;
 		touchscreen-fuzz-pressure = <2>;
-		touchscreen-max-x = <4096>;
-		touchscreen-max-y = <4096>;
+		touchscreen-size-x = <4096>;
+		touchscreen-size-y = <4096>;
 		touchscreen-max-pressure = <2048>;
 
 		ti,x-plate-ohms = <280>;
diff --git a/arch/arm/mach-exynos/suspend.c b/arch/arm/mach-exynos/suspend.c
index c0b6dccbf7bd..7d23ce04cad5 100644
--- a/arch/arm/mach-exynos/suspend.c
+++ b/arch/arm/mach-exynos/suspend.c
@@ -87,8 +87,8 @@ static unsigned int exynos_pmu_spare3;
 static u32 exynos_irqwake_intmask = 0xffffffff;
 
 static const struct exynos_wkup_irq exynos3250_wkup_irq[] = {
-	{ 105, BIT(1) }, /* RTC alarm */
-	{ 106, BIT(2) }, /* RTC tick */
+	{ 73, BIT(1) }, /* RTC alarm */
+	{ 74, BIT(2) }, /* RTC tick */
 	{ /* sentinel */ },
 };
 
diff --git a/arch/arm/mach-omap2/sleep34xx.S b/arch/arm/mach-omap2/sleep34xx.S
index d1dedc8195ed..eafd120b53f1 100644
--- a/arch/arm/mach-omap2/sleep34xx.S
+++ b/arch/arm/mach-omap2/sleep34xx.S
@@ -203,23 +203,8 @@ save_context_wfi:
 	 */
 	ldr	r1, kernel_flush
 	blx	r1
-	/*
-	 * The kernel doesn't interwork: v7_flush_dcache_all in particluar will
-	 * always return in Thumb state when CONFIG_THUMB2_KERNEL is enabled.
-	 * This sequence switches back to ARM.  Note that .align may insert a
-	 * nop: bx pc needs to be word-aligned in order to work.
-	 */
- THUMB(	.thumb		)
- THUMB(	.align		)
- THUMB(	bx	pc	)
- THUMB(	nop		)
-	.arm
-
 	b	omap3_do_wfi
-
-/*
- * Local variables
- */
+ENDPROC(omap34xx_cpu_suspend)
 omap3_do_wfi_sram_addr:
 	.word omap3_do_wfi_sram
 kernel_flush:
@@ -364,10 +349,7 @@ exit_nonoff_modes:
  * ===================================
  */
 	ldmfd	sp!, {r4 - r11, pc}	@ restore regs and return
-
-/*
- * Local variables
- */
+ENDPROC(omap3_do_wfi)
 sdrc_power:
 	.word	SDRC_POWER_V
 cm_idlest1_core:
diff --git a/arch/arm64/boot/dts/mediatek/mt8173-evb.dts b/arch/arm64/boot/dts/mediatek/mt8173-evb.dts
index 43d54017b779..d0ab012fa379 100644
--- a/arch/arm64/boot/dts/mediatek/mt8173-evb.dts
+++ b/arch/arm64/boot/dts/mediatek/mt8173-evb.dts
@@ -16,7 +16,8 @@
 #include "mt8173.dtsi"
 
 / {
-	model = "mediatek,mt8173-evb";
+	model = "MediaTek MT8173 evaluation board";
+	compatible = "mediatek,mt8173-evb", "mediatek,mt8173";
 
 	aliases {
 		serial0 = &uart0;
diff --git a/arch/ia64/kernel/smpboot.c b/arch/ia64/kernel/smpboot.c
index 15051e9c2c6f..b054c5c6e713 100644
--- a/arch/ia64/kernel/smpboot.c
+++ b/arch/ia64/kernel/smpboot.c
@@ -127,7 +127,7 @@ int smp_num_siblings = 1;
 volatile int ia64_cpu_to_sapicid[NR_CPUS];
 EXPORT_SYMBOL(ia64_cpu_to_sapicid);
 
-static volatile cpumask_t cpu_callin_map;
+static cpumask_t cpu_callin_map;
 
 struct smp_boot_data smp_boot_data __initdata;
 
@@ -477,6 +477,7 @@ do_boot_cpu (int sapicid, int cpu, struct task_struct *idle)
 	for (timeout = 0; timeout < 100000; timeout++) {
 		if (cpumask_test_cpu(cpu, &cpu_callin_map))
 			break;  /* It has booted */
+		barrier(); /* Make sure we re-read cpu_callin_map */
 		udelay(100);
 	}
 	Dprintk("\n");
diff --git a/arch/mips/ath79/setup.c b/arch/mips/ath79/setup.c
index a73c93c3d44a..7fc8397d16f2 100644
--- a/arch/mips/ath79/setup.c
+++ b/arch/mips/ath79/setup.c
@@ -225,7 +225,7 @@ void __init plat_time_init(void)
 	ddr_clk_rate = ath79_get_sys_clk_rate("ddr");
 	ref_clk_rate = ath79_get_sys_clk_rate("ref");
 
-	pr_info("Clocks: CPU:%lu.%03luMHz, DDR:%lu.%03luMHz, AHB:%lu.%03luMHz, Ref:%lu.%03luMHz",
+	pr_info("Clocks: CPU:%lu.%03luMHz, DDR:%lu.%03luMHz, AHB:%lu.%03luMHz, Ref:%lu.%03luMHz\n",
 		cpu_clk_rate / 1000000, (cpu_clk_rate / 1000) % 1000,
 		ddr_clk_rate / 1000000, (ddr_clk_rate / 1000) % 1000,
 		ahb_clk_rate / 1000000, (ahb_clk_rate / 1000) % 1000,
diff --git a/arch/mips/kernel/cpu-probe.c b/arch/mips/kernel/cpu-probe.c
index e36515dcd3b2..209e5b76c1bc 100644
--- a/arch/mips/kernel/cpu-probe.c
+++ b/arch/mips/kernel/cpu-probe.c
@@ -74,13 +74,12 @@ static inline void cpu_set_fpu_fcsr_mask(struct cpuinfo_mips *c)
 {
 	unsigned long sr, mask, fcsr, fcsr0, fcsr1;
 
+	fcsr = c->fpu_csr31;
 	mask = FPU_CSR_ALL_X | FPU_CSR_ALL_E | FPU_CSR_ALL_S | FPU_CSR_RM;
 
 	sr = read_c0_status();
 	__enable_fpu(FPU_AS_IS);
 
-	fcsr = read_32bit_cp1_register(CP1_STATUS);
-
 	fcsr0 = fcsr & mask;
 	write_32bit_cp1_register(CP1_STATUS, fcsr0);
 	fcsr0 = read_32bit_cp1_register(CP1_STATUS);
diff --git a/arch/mips/kernel/irq.c b/arch/mips/kernel/irq.c
index 51f57d841662..3c8a18a00a65 100644
--- a/arch/mips/kernel/irq.c
+++ b/arch/mips/kernel/irq.c
@@ -109,7 +109,7 @@ void __init init_IRQ(void)
 #endif
 }
 
-#ifdef DEBUG_STACKOVERFLOW
+#ifdef CONFIG_DEBUG_STACKOVERFLOW
 static inline void check_stack_overflow(void)
 {
 	unsigned long sp;
diff --git a/arch/mips/kvm/emulate.c b/arch/mips/kvm/emulate.c
index 4b50c5787e25..d5fa3eaf39a1 100644
--- a/arch/mips/kvm/emulate.c
+++ b/arch/mips/kvm/emulate.c
@@ -2409,7 +2409,7 @@ enum emulation_result kvm_mips_complete_mmio_load(struct kvm_vcpu *vcpu,
 		if (vcpu->mmio_needed == 2)
 			*gpr = *(int16_t *) run->mmio.data;
 		else
-			*gpr = *(int16_t *) run->mmio.data;
+			*gpr = *(uint16_t *)run->mmio.data;
 
 		break;
 	case 1:
diff --git a/arch/mips/loongson/loongson-3/smp.c b/arch/mips/loongson/loongson-3/smp.c
index e3c68b5da18d..509877c6e9d9 100644
--- a/arch/mips/loongson/loongson-3/smp.c
+++ b/arch/mips/loongson/loongson-3/smp.c
@@ -272,7 +272,7 @@ void loongson3_ipi_interrupt(struct pt_regs *regs)
 	if (action & SMP_ASK_C0COUNT) {
 		BUG_ON(cpu != 0);
 		c0count = read_c0_count();
-		for (i = 1; i < loongson_sysconf.nr_cpus; i++)
+		for (i = 1; i < num_possible_cpus(); i++)
 			per_cpu(core0_c0count, i) = c0count;
 	}
 }
diff --git a/arch/mips/mm/c-r4k.c b/arch/mips/mm/c-r4k.c
index 0dbb65a51ce5..2e03ab173591 100644
--- a/arch/mips/mm/c-r4k.c
+++ b/arch/mips/mm/c-r4k.c
@@ -1372,7 +1372,7 @@ static int probe_scache(void)
 	scache_size = addr;
 	c->scache.linesz = 16 << ((config & R4K_CONF_SB) >> 22);
 	c->scache.ways = 1;
-	c->dcache.waybit = 0;		/* does not matter */
+	c->scache.waybit = 0;		/* does not matter */
 
 	return 1;
 }
diff --git a/arch/mips/net/bpf_jit.c b/arch/mips/net/bpf_jit.c
index 5d6139390bf8..e23fdf2a9c80 100644
--- a/arch/mips/net/bpf_jit.c
+++ b/arch/mips/net/bpf_jit.c
@@ -681,11 +681,7 @@ static unsigned int get_stack_depth(struct jit_ctx *ctx)
 		sp_off += config_enabled(CONFIG_64BIT) ?
 			(ARGS_USED_BY_JIT + 1) * RSIZE : RSIZE;
 
-	/*
-	 * Subtract the bytes for the last registers since we only care about
-	 * the location on the stack pointer.
-	 */
-	return sp_off - RSIZE;
+	return sp_off;
 }
 
 static void build_prologue(struct jit_ctx *ctx)
diff --git a/arch/mips/ralink/ill_acc.c b/arch/mips/ralink/ill_acc.c
index e20b02e3ae28..e10d10b9e82a 100644
--- a/arch/mips/ralink/ill_acc.c
+++ b/arch/mips/ralink/ill_acc.c
@@ -41,7 +41,7 @@ static irqreturn_t ill_acc_irq_handler(int irq, void *_priv)
 		addr, (type >> ILL_ACC_OFF_S) & ILL_ACC_OFF_M,
 		type & ILL_ACC_LEN_M);
 
-	rt_memc_w32(REG_ILL_ACC_TYPE, REG_ILL_ACC_TYPE);
+	rt_memc_w32(ILL_INT_STATUS, REG_ILL_ACC_TYPE);
 
 	return IRQ_HANDLED;
 }
diff --git a/arch/s390/net/bpf_jit.h b/arch/s390/net/bpf_jit.h
index ba8593a515ba..de156ba3bd71 100644
--- a/arch/s390/net/bpf_jit.h
+++ b/arch/s390/net/bpf_jit.h
@@ -48,7 +48,9 @@ extern u8 sk_load_word[], sk_load_half[], sk_load_byte[];
  * We get 160 bytes stack space from calling function, but only use
  * 11 * 8 byte (old backchain + r15 - r6) for storing registers.
  */
-#define STK_OFF (MAX_BPF_STACK + 8 + 4 + 4 + (160 - 11 * 8))
+#define STK_SPACE	(MAX_BPF_STACK + 8 + 4 + 4 + 160)
+#define STK_160_UNUSED	(160 - 11 * 8)
+#define STK_OFF		(STK_SPACE - STK_160_UNUSED)
 #define STK_OFF_TMP	160	/* Offset of tmp buffer on stack */
 #define STK_OFF_HLEN	168	/* Offset of SKB header length on stack */
 
diff --git a/arch/s390/net/bpf_jit_comp.c b/arch/s390/net/bpf_jit_comp.c
index 20c146d1251a..55423d8be580 100644
--- a/arch/s390/net/bpf_jit_comp.c
+++ b/arch/s390/net/bpf_jit_comp.c
@@ -384,13 +384,16 @@ static void bpf_jit_prologue(struct bpf_jit *jit)
 	}
 	/* Setup stack and backchain */
 	if (jit->seen & SEEN_STACK) {
-		/* lgr %bfp,%r15 (BPF frame pointer) */
-		EMIT4(0xb9040000, BPF_REG_FP, REG_15);
+		if (jit->seen & SEEN_FUNC)
+			/* lgr %w1,%r15 (backchain) */
+			EMIT4(0xb9040000, REG_W1, REG_15);
+		/* la %bfp,STK_160_UNUSED(%r15) (BPF frame pointer) */
+		EMIT4_DISP(0x41000000, BPF_REG_FP, REG_15, STK_160_UNUSED);
 		/* aghi %r15,-STK_OFF */
 		EMIT4_IMM(0xa70b0000, REG_15, -STK_OFF);
 		if (jit->seen & SEEN_FUNC)
-			/* stg %bfp,152(%r15) (backchain) */
-			EMIT6_DISP_LH(0xe3000000, 0x0024, BPF_REG_FP, REG_0,
+			/* stg %w1,152(%r15) (backchain) */
+			EMIT6_DISP_LH(0xe3000000, 0x0024, REG_W1, REG_0,
 				      REG_15, 152);
 	}
 	/*
diff --git a/arch/x86/boot/compressed/misc.h b/arch/x86/boot/compressed/misc.h
index 89dd0d78013a..805d25ca5f1d 100644
--- a/arch/x86/boot/compressed/misc.h
+++ b/arch/x86/boot/compressed/misc.h
@@ -2,15 +2,14 @@
 #define BOOT_COMPRESSED_MISC_H
 
 /*
- * we have to be careful, because no indirections are allowed here, and
- * paravirt_ops is a kind of one. As it will only run in baremetal anyway,
- * we just keep it from happening
+ * Special hack: we have to be careful, because no indirections are allowed here,
+ * and paravirt_ops is a kind of one. As it will only run in baremetal anyway,
+ * we just keep it from happening. (This list needs to be extended when new
+ * paravirt and debugging variants are added.)
  */
 #undef CONFIG_PARAVIRT
+#undef CONFIG_PARAVIRT_SPINLOCKS
 #undef CONFIG_KASAN
-#ifdef CONFIG_X86_32
-#define _ASM_X86_DESC_H 1
-#endif
 
 #include <linux/linkage.h>
 #include <linux/screen_info.h>
diff --git a/arch/x86/include/asm/ptrace.h b/arch/x86/include/asm/ptrace.h
index 19507ffa5d28..5fabf1362942 100644
--- a/arch/x86/include/asm/ptrace.h
+++ b/arch/x86/include/asm/ptrace.h
@@ -107,7 +107,7 @@ static inline unsigned long regs_return_value(struct pt_regs *regs)
 static inline int user_mode(struct pt_regs *regs)
 {
 #ifdef CONFIG_X86_32
-	return (regs->cs & SEGMENT_RPL_MASK) == USER_RPL;
+	return ((regs->cs & SEGMENT_RPL_MASK) | (regs->flags & X86_VM_MASK)) >= USER_RPL;
 #else
 	return !!(regs->cs & 3);
 #endif
diff --git a/arch/x86/include/asm/segment.h b/arch/x86/include/asm/segment.h
index 5a9856eb12ba..7d5a1929d76b 100644
--- a/arch/x86/include/asm/segment.h
+++ b/arch/x86/include/asm/segment.h
@@ -231,11 +231,21 @@
 #define TLS_SIZE			(GDT_ENTRY_TLS_ENTRIES* 8)
 
 #ifdef __KERNEL__
+
+/*
+ * early_idt_handler_array is an array of entry points referenced in the
+ * early IDT.  For simplicity, it's a real array with one entry point
+ * every nine bytes.  That leaves room for an optional 'push $0' if the
+ * vector has no error code (two bytes), a 'push $vector_number' (two
+ * bytes), and a jump to the common entry code (up to five bytes).
+ */
+#define EARLY_IDT_HANDLER_SIZE 9
+
 #ifndef __ASSEMBLY__
 
-extern const char early_idt_handlers[NUM_EXCEPTION_VECTORS][2+2+5];
+extern const char early_idt_handler_array[NUM_EXCEPTION_VECTORS][EARLY_IDT_HANDLER_SIZE];
 #ifdef CONFIG_TRACING
-# define trace_early_idt_handlers early_idt_handlers
+# define trace_early_idt_handler_array early_idt_handler_array
 #endif
 
 /*
diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c
index 87848ebe2bb7..4f7001f28936 100644
--- a/arch/x86/kernel/cpu/perf_event.c
+++ b/arch/x86/kernel/cpu/perf_event.c
@@ -190,6 +190,7 @@ static bool check_hw_exists(void)
 	u64 val, val_fail, val_new= ~0;
 	int i, reg, reg_fail, ret = 0;
 	int bios_fail = 0;
+	int reg_safe = -1;
 
 	/*
 	 * Check to see if the BIOS enabled any of the counters, if so
@@ -204,6 +205,8 @@ static bool check_hw_exists(void)
 			bios_fail = 1;
 			val_fail = val;
 			reg_fail = reg;
+		} else {
+			reg_safe = i;
 		}
 	}
 
@@ -222,11 +225,22 @@ static bool check_hw_exists(void)
 	}
 
 	/*
+	 * If all the counters are enabled, the below test will always
+	 * fail.  The tools will also become useless in this scenario.
+	 * Just fail and disable the hardware counters.
+	 */
+
+	if (reg_safe == -1) {
+		reg = reg_safe;
+		goto msr_fail;
+	}
+
+	/*
 	 * Read the current value, change it and read it back to see if it
 	 * matches, this is needed to detect certain hardware emulators
 	 * (qemu/kvm) that don't trap on the MSR access and always return 0s.
 	 */
-	reg = x86_pmu_event_addr(0);
+	reg = x86_pmu_event_addr(reg_safe);
 	if (rdmsrl_safe(reg, &val))
 		goto msr_fail;
 	val ^= 0xffffUL;
@@ -611,6 +625,7 @@ struct sched_state {
 	int	event;		/* event index */
 	int	counter;	/* counter index */
 	int	unassigned;	/* number of events to be assigned left */
+	int	nr_gp;		/* number of GP counters used */
 	unsigned long used[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
 };
 
@@ -620,27 +635,29 @@ struct sched_state {
 struct perf_sched {
 	int			max_weight;
 	int			max_events;
-	struct perf_event	**events;
-	struct sched_state	state;
+	int			max_gp;
 	int			saved_states;
+	struct event_constraint	**constraints;
+	struct sched_state	state;
 	struct sched_state	saved[SCHED_STATES_MAX];
 };
 
 /*
  * Initialize interator that runs through all events and counters.
  */
-static void perf_sched_init(struct perf_sched *sched, struct perf_event **events,
-			    int num, int wmin, int wmax)
+static void perf_sched_init(struct perf_sched *sched, struct event_constraint **constraints,
+			    int num, int wmin, int wmax, int gpmax)
 {
 	int idx;
 
 	memset(sched, 0, sizeof(*sched));
 	sched->max_events	= num;
 	sched->max_weight	= wmax;
-	sched->events		= events;
+	sched->max_gp		= gpmax;
+	sched->constraints	= constraints;
 
 	for (idx = 0; idx < num; idx++) {
-		if (events[idx]->hw.constraint->weight == wmin)
+		if (constraints[idx]->weight == wmin)
 			break;
 	}
 
@@ -687,7 +704,7 @@ static bool __perf_sched_find_counter(struct perf_sched *sched)
 	if (sched->state.event >= sched->max_events)
 		return false;
 
-	c = sched->events[sched->state.event]->hw.constraint;
+	c = sched->constraints[sched->state.event];
 	/* Prefer fixed purpose counters */
 	if (c->idxmsk64 & (~0ULL << INTEL_PMC_IDX_FIXED)) {
 		idx = INTEL_PMC_IDX_FIXED;
@@ -696,11 +713,16 @@ static bool __perf_sched_find_counter(struct perf_sched *sched)
 				goto done;
 		}
 	}
+
 	/* Grab the first unused counter starting with idx */
 	idx = sched->state.counter;
 	for_each_set_bit_from(idx, c->idxmsk, INTEL_PMC_IDX_FIXED) {
-		if (!__test_and_set_bit(idx, sched->state.used))
+		if (!__test_and_set_bit(idx, sched->state.used)) {
+			if (sched->state.nr_gp++ >= sched->max_gp)
+				return false;
+
 			goto done;
+		}
 	}
 
 	return false;
@@ -745,7 +767,7 @@ static bool perf_sched_next_event(struct perf_sched *sched)
 			if (sched->state.weight > sched->max_weight)
 				return false;
 		}
-		c = sched->events[sched->state.event]->hw.constraint;
+		c = sched->constraints[sched->state.event];
 	} while (c->weight != sched->state.weight);
 
 	sched->state.counter = 0;	/* start with first counter */
@@ -756,12 +778,12 @@ static bool perf_sched_next_event(struct perf_sched *sched)
 /*
  * Assign a counter for each event.
  */
-int perf_assign_events(struct perf_event **events, int n,
-			int wmin, int wmax, int *assign)
+int perf_assign_events(struct event_constraint **constraints, int n,
+			int wmin, int wmax, int gpmax, int *assign)
 {
 	struct perf_sched sched;
 
-	perf_sched_init(&sched, events, n, wmin, wmax);
+	perf_sched_init(&sched, constraints, n, wmin, wmax, gpmax);
 
 	do {
 		if (!perf_sched_find_counter(&sched))
@@ -788,9 +810,9 @@ int x86_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign)
 		x86_pmu.start_scheduling(cpuc);
 
 	for (i = 0, wmin = X86_PMC_IDX_MAX, wmax = 0; i < n; i++) {
-		hwc = &cpuc->event_list[i]->hw;
+		cpuc->event_constraint[i] = NULL;
 		c = x86_pmu.get_event_constraints(cpuc, i, cpuc->event_list[i]);
-		hwc->constraint = c;
+		cpuc->event_constraint[i] = c;
 
 		wmin = min(wmin, c->weight);
 		wmax = max(wmax, c->weight);
@@ -801,7 +823,7 @@ int x86_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign)
 	 */
 	for (i = 0; i < n; i++) {
 		hwc = &cpuc->event_list[i]->hw;
-		c = hwc->constraint;
+		c = cpuc->event_constraint[i];
 
 		/* never assigned */
 		if (hwc->idx == -1)
@@ -821,9 +843,26 @@ int x86_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign)
 	}
 
 	/* slow path */
-	if (i != n)
-		unsched = perf_assign_events(cpuc->event_list, n, wmin,
-					     wmax, assign);
+	if (i != n) {
+		int gpmax = x86_pmu.num_counters;
+
+		/*
+		 * Do not allow scheduling of more than half the available
+		 * generic counters.
+		 *
+		 * This helps avoid counter starvation of sibling thread by
+		 * ensuring at most half the counters cannot be in exclusive
+		 * mode. There is no designated counters for the limits. Any
+		 * N/2 counters can be used. This helps with events with
+		 * specific counter constraints.
+		 */
+		if (is_ht_workaround_enabled() && !cpuc->is_fake &&
+		    READ_ONCE(cpuc->excl_cntrs->exclusive_present))
+			gpmax /= 2;
+
+		unsched = perf_assign_events(cpuc->event_constraint, n, wmin,
+					     wmax, gpmax, assign);
+	}
 
 	/*
 	 * In case of success (unsched = 0), mark events as committed,
@@ -840,7 +879,7 @@ int x86_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign)
 			e = cpuc->event_list[i];
 			e->hw.flags |= PERF_X86_EVENT_COMMITTED;
 			if (x86_pmu.commit_scheduling)
-				x86_pmu.commit_scheduling(cpuc, e, assign[i]);
+				x86_pmu.commit_scheduling(cpuc, i, assign[i]);
 		}
 	}
 
@@ -1292,8 +1331,10 @@ static void x86_pmu_del(struct perf_event *event, int flags)
 		x86_pmu.put_event_constraints(cpuc, event);
 
 	/* Delete the array entry. */
-	while (++i < cpuc->n_events)
+	while (++i < cpuc->n_events) {
 		cpuc->event_list[i-1] = cpuc->event_list[i];
+		cpuc->event_constraint[i-1] = cpuc->event_constraint[i];
+	}
 	--cpuc->n_events;
 
 	perf_event_update_userpage(event);
diff --git a/arch/x86/kernel/cpu/perf_event.h b/arch/x86/kernel/cpu/perf_event.h
index 6ac5cb7a9e14..ef78516850fb 100644
--- a/arch/x86/kernel/cpu/perf_event.h
+++ b/arch/x86/kernel/cpu/perf_event.h
@@ -74,6 +74,7 @@ struct event_constraint {
 #define PERF_X86_EVENT_EXCL		0x0040 /* HT exclusivity on counter */
 #define PERF_X86_EVENT_DYNAMIC		0x0080 /* dynamic alloc'd constraint */
 #define PERF_X86_EVENT_RDPMC_ALLOWED	0x0100 /* grant rdpmc permission */
+#define PERF_X86_EVENT_EXCL_ACCT	0x0200 /* accounted EXCL event */
 
 
 struct amd_nb {
@@ -134,8 +135,6 @@ enum intel_excl_state_type {
 struct intel_excl_states {
 	enum intel_excl_state_type init_state[X86_PMC_IDX_MAX];
 	enum intel_excl_state_type state[X86_PMC_IDX_MAX];
-	int  num_alloc_cntrs;/* #counters allocated */
-	int  max_alloc_cntrs;/* max #counters allowed */
 	bool sched_started; /* true if scheduling has started */
 };
 
@@ -144,6 +143,11 @@ struct intel_excl_cntrs {
 
 	struct intel_excl_states states[2];
 
+	union {
+		u16	has_exclusive[2];
+		u32	exclusive_present;
+	};
+
 	int		refcnt;		/* per-core: #HT threads */
 	unsigned	core_id;	/* per-core: core id */
 };
@@ -172,7 +176,11 @@ struct cpu_hw_events {
 					     added in the current transaction */
 	int			assign[X86_PMC_IDX_MAX]; /* event to counter assignment */
 	u64			tags[X86_PMC_IDX_MAX];
+
 	struct perf_event	*event_list[X86_PMC_IDX_MAX]; /* in enabled order */
+	struct event_constraint	*event_constraint[X86_PMC_IDX_MAX];
+
+	int			n_excl; /* the number of exclusive events */
 
 	unsigned int		group_flag;
 	int			is_fake;
@@ -519,9 +527,7 @@ struct x86_pmu {
 	void		(*put_event_constraints)(struct cpu_hw_events *cpuc,
 						 struct perf_event *event);
 
-	void		(*commit_scheduling)(struct cpu_hw_events *cpuc,
-					     struct perf_event *event,
-					     int cntr);
+	void		(*commit_scheduling)(struct cpu_hw_events *cpuc, int idx, int cntr);
 
 	void		(*start_scheduling)(struct cpu_hw_events *cpuc);
 
@@ -717,8 +723,8 @@ static inline void __x86_pmu_enable_event(struct hw_perf_event *hwc,
 
 void x86_pmu_enable_all(int added);
 
-int perf_assign_events(struct perf_event **events, int n,
-			int wmin, int wmax, int *assign);
+int perf_assign_events(struct event_constraint **constraints, int n,
+			int wmin, int wmax, int gpmax, int *assign);
 int x86_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign);
 
 void x86_pmu_stop(struct perf_event *event, int flags);
@@ -929,4 +935,8 @@ static inline struct intel_shared_regs *allocate_shared_regs(int cpu)
 	return NULL;
 }
 
+static inline int is_ht_workaround_enabled(void)
+{
+	return 0;
+}
 #endif /* CONFIG_CPU_SUP_INTEL */
diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c
index 3998131d1a68..a1e35c9f06b9 100644
--- a/arch/x86/kernel/cpu/perf_event_intel.c
+++ b/arch/x86/kernel/cpu/perf_event_intel.c
@@ -1923,7 +1923,6 @@ intel_start_scheduling(struct cpu_hw_events *cpuc)
 	xl = &excl_cntrs->states[tid];
 
 	xl->sched_started = true;
-	xl->num_alloc_cntrs = 0;
 	/*
 	 * lock shared state until we are done scheduling
 	 * in stop_event_scheduling()
@@ -2000,6 +1999,11 @@ intel_get_excl_constraints(struct cpu_hw_events *cpuc, struct perf_event *event,
 	 * across HT threads
 	 */
 	is_excl = c->flags & PERF_X86_EVENT_EXCL;
+	if (is_excl && !(event->hw.flags & PERF_X86_EVENT_EXCL_ACCT)) {
+		event->hw.flags |= PERF_X86_EVENT_EXCL_ACCT;
+		if (!cpuc->n_excl++)
+			WRITE_ONCE(excl_cntrs->has_exclusive[tid], 1);
+	}
 
 	/*
 	 * xl = state of current HT
@@ -2008,18 +2012,6 @@ intel_get_excl_constraints(struct cpu_hw_events *cpuc, struct perf_event *event,
 	xl = &excl_cntrs->states[tid];
 	xlo = &excl_cntrs->states[o_tid];
 
-	/*
-	 * do not allow scheduling of more than max_alloc_cntrs
-	 * which is set to half the available generic counters.
-	 * this helps avoid counter starvation of sibling thread
-	 * by ensuring at most half the counters cannot be in
-	 * exclusive mode. There is not designated counters for the
-	 * limits. Any N/2 counters can be used. This helps with
-	 * events with specifix counter constraints
-	 */
-	if (xl->num_alloc_cntrs++ == xl->max_alloc_cntrs)
-		return &emptyconstraint;
-
 	cx = c;
 
 	/*
@@ -2106,7 +2098,7 @@ static struct event_constraint *
 intel_get_event_constraints(struct cpu_hw_events *cpuc, int idx,
 			    struct perf_event *event)
 {
-	struct event_constraint *c1 = event->hw.constraint;
+	struct event_constraint *c1 = cpuc->event_constraint[idx];
 	struct event_constraint *c2;
 
 	/*
@@ -2150,6 +2142,11 @@ static void intel_put_excl_constraints(struct cpu_hw_events *cpuc,
 
 	xl = &excl_cntrs->states[tid];
 	xlo = &excl_cntrs->states[o_tid];
+	if (hwc->flags & PERF_X86_EVENT_EXCL_ACCT) {
+		hwc->flags &= ~PERF_X86_EVENT_EXCL_ACCT;
+		if (!--cpuc->n_excl)
+			WRITE_ONCE(excl_cntrs->has_exclusive[tid], 0);
+	}
 
 	/*
 	 * put_constraint may be called from x86_schedule_events()
@@ -2188,8 +2185,6 @@ intel_put_shared_regs_event_constraints(struct cpu_hw_events *cpuc,
 static void intel_put_event_constraints(struct cpu_hw_events *cpuc,
 					struct perf_event *event)
 {
-	struct event_constraint *c = event->hw.constraint;
-
 	intel_put_shared_regs_event_constraints(cpuc, event);
 
 	/*
@@ -2197,19 +2192,14 @@ static void intel_put_event_constraints(struct cpu_hw_events *cpuc,
 	 * all events are subject to and must call the
 	 * put_excl_constraints() routine
 	 */
-	if (c && cpuc->excl_cntrs)
+	if (cpuc->excl_cntrs)
 		intel_put_excl_constraints(cpuc, event);
-
-	/* cleanup dynamic constraint */
-	if (c && (c->flags & PERF_X86_EVENT_DYNAMIC))
-		event->hw.constraint = NULL;
 }
 
-static void intel_commit_scheduling(struct cpu_hw_events *cpuc,
-				    struct perf_event *event, int cntr)
+static void intel_commit_scheduling(struct cpu_hw_events *cpuc, int idx, int cntr)
 {
 	struct intel_excl_cntrs *excl_cntrs = cpuc->excl_cntrs;
-	struct event_constraint *c = event->hw.constraint;
+	struct event_constraint *c = cpuc->event_constraint[idx];
 	struct intel_excl_states *xlo, *xl;
 	int tid = cpuc->excl_thread_id;
 	int o_tid = 1 - tid;
@@ -2639,8 +2629,6 @@ static void intel_pmu_cpu_starting(int cpu)
 		cpuc->lbr_sel = &cpuc->shared_regs->regs[EXTRA_REG_LBR];
 
 	if (x86_pmu.flags & PMU_FL_EXCL_CNTRS) {
-		int h = x86_pmu.num_counters >> 1;
-
 		for_each_cpu(i, topology_thread_cpumask(cpu)) {
 			struct intel_excl_cntrs *c;
 
@@ -2654,11 +2642,6 @@ static void intel_pmu_cpu_starting(int cpu)
 		}
 		cpuc->excl_cntrs->core_id = core_id;
 		cpuc->excl_cntrs->refcnt++;
-		/*
-		 * set hard limit to half the number of generic counters
-		 */
-		cpuc->excl_cntrs->states[0].max_alloc_cntrs = h;
-		cpuc->excl_cntrs->states[1].max_alloc_cntrs = h;
 	}
 }
 
diff --git a/arch/x86/kernel/cpu/perf_event_intel_ds.c b/arch/x86/kernel/cpu/perf_event_intel_ds.c
index 813f75d71175..7f73b3553e2e 100644
--- a/arch/x86/kernel/cpu/perf_event_intel_ds.c
+++ b/arch/x86/kernel/cpu/perf_event_intel_ds.c
@@ -706,9 +706,9 @@ void intel_pmu_pebs_disable(struct perf_event *event)
 
 	cpuc->pebs_enabled &= ~(1ULL << hwc->idx);
 
-	if (event->hw.constraint->flags & PERF_X86_EVENT_PEBS_LDLAT)
+	if (event->hw.flags & PERF_X86_EVENT_PEBS_LDLAT)
 		cpuc->pebs_enabled &= ~(1ULL << (hwc->idx + 32));
-	else if (event->hw.constraint->flags & PERF_X86_EVENT_PEBS_ST)
+	else if (event->hw.flags & PERF_X86_EVENT_PEBS_ST)
 		cpuc->pebs_enabled &= ~(1ULL << 63);
 
 	if (cpuc->enabled)
diff --git a/arch/x86/kernel/cpu/perf_event_intel_pt.c b/arch/x86/kernel/cpu/perf_event_intel_pt.c
index ffe666c2c6b5..123ff1bb2f60 100644
--- a/arch/x86/kernel/cpu/perf_event_intel_pt.c
+++ b/arch/x86/kernel/cpu/perf_event_intel_pt.c
@@ -151,7 +151,7 @@ static int __init pt_pmu_hw_init(void)
 
 		de_attr->attr.attr.name = pt_caps[i].name;
 
-		sysfs_attr_init(&de_attrs->attr.attr);
+		sysfs_attr_init(&de_attr->attr.attr);
 
 		de_attr->attr.attr.mode		= S_IRUGO;
 		de_attr->attr.show		= pt_cap_show;
@@ -615,7 +615,8 @@ static int pt_buffer_reset_markers(struct pt_buffer *buf,
 				   struct perf_output_handle *handle)
 
 {
-	unsigned long idx, npages, end;
+	unsigned long head = local64_read(&buf->head);
+	unsigned long idx, npages, wakeup;
 
 	if (buf->snapshot)
 		return 0;
@@ -634,17 +635,26 @@ static int pt_buffer_reset_markers(struct pt_buffer *buf,
 	buf->topa_index[buf->stop_pos]->stop = 0;
 	buf->topa_index[buf->intr_pos]->intr = 0;
 
-	if (pt_cap_get(PT_CAP_topa_multiple_entries)) {
-		npages = (handle->size + 1) >> PAGE_SHIFT;
-		end = (local64_read(&buf->head) >> PAGE_SHIFT) + npages;
-		/*if (end > handle->wakeup >> PAGE_SHIFT)
-		  end = handle->wakeup >> PAGE_SHIFT;*/
-		idx = end & (buf->nr_pages - 1);
-		buf->stop_pos = idx;
-		idx = (local64_read(&buf->head) >> PAGE_SHIFT) + npages - 1;
-		idx &= buf->nr_pages - 1;
-		buf->intr_pos = idx;
-	}
+	/* how many pages till the STOP marker */
+	npages = handle->size >> PAGE_SHIFT;
+
+	/* if it's on a page boundary, fill up one more page */
+	if (!offset_in_page(head + handle->size + 1))
+		npages++;
+
+	idx = (head >> PAGE_SHIFT) + npages;
+	idx &= buf->nr_pages - 1;
+	buf->stop_pos = idx;
+
+	wakeup = handle->wakeup >> PAGE_SHIFT;
+
+	/* in the worst case, wake up the consumer one page before hard stop */
+	idx = (head >> PAGE_SHIFT) + npages - 1;
+	if (idx > wakeup)
+		idx = wakeup;
+
+	idx &= buf->nr_pages - 1;
+	buf->intr_pos = idx;
 
 	buf->topa_index[buf->stop_pos]->stop = 1;
 	buf->topa_index[buf->intr_pos]->intr = 1;
diff --git a/arch/x86/kernel/cpu/perf_event_intel_uncore.c b/arch/x86/kernel/cpu/perf_event_intel_uncore.c
index c635b8b49e93..dd319e59246b 100644
--- a/arch/x86/kernel/cpu/perf_event_intel_uncore.c
+++ b/arch/x86/kernel/cpu/perf_event_intel_uncore.c
@@ -365,9 +365,8 @@ static int uncore_assign_events(struct intel_uncore_box *box, int assign[], int
 	bitmap_zero(used_mask, UNCORE_PMC_IDX_MAX);
 
 	for (i = 0, wmin = UNCORE_PMC_IDX_MAX, wmax = 0; i < n; i++) {
-		hwc = &box->event_list[i]->hw;
 		c = uncore_get_event_constraint(box, box->event_list[i]);
-		hwc->constraint = c;
+		box->event_constraint[i] = c;
 		wmin = min(wmin, c->weight);
 		wmax = max(wmax, c->weight);
 	}
@@ -375,7 +374,7 @@ static int uncore_assign_events(struct intel_uncore_box *box, int assign[], int
 	/* fastpath, try to reuse previous register */
 	for (i = 0; i < n; i++) {
 		hwc = &box->event_list[i]->hw;
-		c = hwc->constraint;
+		c = box->event_constraint[i];
 
 		/* never assigned */
 		if (hwc->idx == -1)
@@ -395,8 +394,8 @@ static int uncore_assign_events(struct intel_uncore_box *box, int assign[], int
 	}
 	/* slow path */
 	if (i != n)
-		ret = perf_assign_events(box->event_list, n,
-					 wmin, wmax, assign);
+		ret = perf_assign_events(box->event_constraint, n,
+					 wmin, wmax, n, assign);
 
 	if (!assign || ret) {
 		for (i = 0; i < n; i++)
diff --git a/arch/x86/kernel/cpu/perf_event_intel_uncore.h b/arch/x86/kernel/cpu/perf_event_intel_uncore.h
index 6c8c1e7e69d8..f789ec9a0133 100644
--- a/arch/x86/kernel/cpu/perf_event_intel_uncore.h
+++ b/arch/x86/kernel/cpu/perf_event_intel_uncore.h
@@ -97,6 +97,7 @@ struct intel_uncore_box {
 	atomic_t refcnt;
 	struct perf_event *events[UNCORE_PMC_IDX_MAX];
 	struct perf_event *event_list[UNCORE_PMC_IDX_MAX];
+	struct event_constraint *event_constraint[UNCORE_PMC_IDX_MAX];
 	unsigned long active_mask[BITS_TO_LONGS(UNCORE_PMC_IDX_MAX)];
 	u64 tags[UNCORE_PMC_IDX_MAX];
 	struct pci_dev *pci_dev;
diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c
index 2b55ee6db053..5a4668136e98 100644
--- a/arch/x86/kernel/head64.c
+++ b/arch/x86/kernel/head64.c
@@ -167,7 +167,7 @@ asmlinkage __visible void __init x86_64_start_kernel(char * real_mode_data)
 	clear_bss();
 
 	for (i = 0; i < NUM_EXCEPTION_VECTORS; i++)
-		set_intr_gate(i, early_idt_handlers[i]);
+		set_intr_gate(i, early_idt_handler_array[i]);
 	load_idt((const struct desc_ptr *)&idt_descr);
 
 	copy_bootdata(__va(real_mode_data));
diff --git a/arch/x86/kernel/head_32.S b/arch/x86/kernel/head_32.S
index d031bad9e07e..53eeb226657c 100644
--- a/arch/x86/kernel/head_32.S
+++ b/arch/x86/kernel/head_32.S
@@ -478,21 +478,22 @@ is486:
 __INIT
 setup_once:
 	/*
-	 * Set up a idt with 256 entries pointing to ignore_int,
-	 * interrupt gates. It doesn't actually load idt - that needs
-	 * to be done on each CPU. Interrupts are enabled elsewhere,
-	 * when we can be relatively sure everything is ok.
+	 * Set up a idt with 256 interrupt gates that push zero if there
+	 * is no error code and then jump to early_idt_handler_common.
+	 * It doesn't actually load the idt - that needs to be done on
+	 * each CPU. Interrupts are enabled elsewhere, when we can be
+	 * relatively sure everything is ok.
 	 */
 
 	movl $idt_table,%edi
-	movl $early_idt_handlers,%eax
+	movl $early_idt_handler_array,%eax
 	movl $NUM_EXCEPTION_VECTORS,%ecx
 1:
 	movl %eax,(%edi)
 	movl %eax,4(%edi)
 	/* interrupt gate, dpl=0, present */
 	movl $(0x8E000000 + __KERNEL_CS),2(%edi)
-	addl $9,%eax
+	addl $EARLY_IDT_HANDLER_SIZE,%eax
 	addl $8,%edi
 	loop 1b
 
@@ -524,26 +525,28 @@ setup_once:
 	andl $0,setup_once_ref	/* Once is enough, thanks */
 	ret
 
-ENTRY(early_idt_handlers)
+ENTRY(early_idt_handler_array)
 	# 36(%esp) %eflags
 	# 32(%esp) %cs
 	# 28(%esp) %eip
 	# 24(%rsp) error code
 	i = 0
 	.rept NUM_EXCEPTION_VECTORS
-	.if (EXCEPTION_ERRCODE_MASK >> i) & 1
-	ASM_NOP2
-	.else
+	.ifeq (EXCEPTION_ERRCODE_MASK >> i) & 1
 	pushl $0		# Dummy error code, to make stack frame uniform
 	.endif
 	pushl $i		# 20(%esp) Vector number
-	jmp early_idt_handler
+	jmp early_idt_handler_common
 	i = i + 1
+	.fill early_idt_handler_array + i*EARLY_IDT_HANDLER_SIZE - ., 1, 0xcc
 	.endr
-ENDPROC(early_idt_handlers)
+ENDPROC(early_idt_handler_array)
 	
-	/* This is global to keep gas from relaxing the jumps */
-ENTRY(early_idt_handler)
+early_idt_handler_common:
+	/*
+	 * The stack is the hardware frame, an error code or zero, and the
+	 * vector number.
+	 */
 	cld
 
 	cmpl $2,(%esp)		# X86_TRAP_NMI
@@ -603,7 +606,7 @@ ex_entry:
 is_nmi:
 	addl $8,%esp		/* drop vector number and error code */
 	iret
-ENDPROC(early_idt_handler)
+ENDPROC(early_idt_handler_common)
 
 /* This is the default interrupt "handler" :-) */
 	ALIGN
diff --git a/arch/x86/kernel/head_64.S b/arch/x86/kernel/head_64.S
index ae6588b301c2..df7e78057ae0 100644
--- a/arch/x86/kernel/head_64.S
+++ b/arch/x86/kernel/head_64.S
@@ -321,26 +321,28 @@ bad_address:
 	jmp bad_address
 
 	__INIT
-	.globl early_idt_handlers
-early_idt_handlers:
+ENTRY(early_idt_handler_array)
 	# 104(%rsp) %rflags
 	#  96(%rsp) %cs
 	#  88(%rsp) %rip
 	#  80(%rsp) error code
 	i = 0
 	.rept NUM_EXCEPTION_VECTORS
-	.if (EXCEPTION_ERRCODE_MASK >> i) & 1
-	ASM_NOP2
-	.else
+	.ifeq (EXCEPTION_ERRCODE_MASK >> i) & 1
 	pushq $0		# Dummy error code, to make stack frame uniform
 	.endif
 	pushq $i		# 72(%rsp) Vector number
-	jmp early_idt_handler
+	jmp early_idt_handler_common
 	i = i + 1
+	.fill early_idt_handler_array + i*EARLY_IDT_HANDLER_SIZE - ., 1, 0xcc
 	.endr
+ENDPROC(early_idt_handler_array)
 
-/* This is global to keep gas from relaxing the jumps */
-ENTRY(early_idt_handler)
+early_idt_handler_common:
+	/*
+	 * The stack is the hardware frame, an error code or zero, and the
+	 * vector number.
+	 */
 	cld
 
 	cmpl $2,(%rsp)		# X86_TRAP_NMI
@@ -412,7 +414,7 @@ ENTRY(early_idt_handler)
 is_nmi:
 	addq $16,%rsp		# drop vector number and error code
 	INTERRUPT_RETURN
-ENDPROC(early_idt_handler)
+ENDPROC(early_idt_handler_common)
 
 	__INITDATA