Merge tag 'arm64-upstream' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux

Pull arm64 updates from Will Deacon:
 "CPU features:

   - Remove ARM64_HAS_NO_HW_PREFETCH copy_page() optimisation for ye
     olde Thunder-X machines

   - Avoid mapping KPTI trampoline when it is not required

   - Make CPU capability API more robust during early initialisation

  Early idreg overrides:

   - Remove dependencies on core kernel helpers from the early
     command-line parsing logic in preparation for moving this code
     before the kernel is mapped

  FPsimd:

   - Restore kernel-mode fpsimd context lazily, allowing us to run
     fpsimd code sequences in the kernel with pre-emption enabled

  KBuild:

   - Install 'vmlinuz.efi' when CONFIG_EFI_ZBOOT=y

   - Makefile cleanups

  LPA2 prep:

   - Preparatory work for enabling the 'LPA2' extension, which will
     introduce 52-bit virtual and physical addressing even with 4KiB
     pages (including for KVM guests).

  Misc:

   - Remove dead code and fix a typo

  MM:

   - Pass NUMA node information for IRQ stack allocations

  Perf:

   - Add perf support for the Synopsys DesignWare PCIe PMU

   - Add support for event counting thresholds (FEAT_PMUv3_TH)
     introduced in Armv8.8

   - Add support for i.MX8DXL SoCs to the IMX DDR PMU driver.

   - Minor PMU driver fixes and optimisations

  RIP VPIPT:

   - Remove what support we had for the obsolete VPIPT I-cache policy

  Selftests:

   - Improvements to the SVE and SME selftests

  Stacktrace:

   - Refactor kernel unwind logic so that it can used by BPF unwinding
     and, eventually, reliable backtracing

  Sysregs:

   - Update a bunch of register definitions based on the latest XML drop
     from Arm"

* tag 'arm64-upstream' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux: (87 commits)
  kselftest/arm64: Don't probe the current VL for unsupported vector types
  efi/libstub: zboot: do not use $(shell ...) in cmd_copy_and_pad
  arm64: properly install vmlinuz.efi
  arm64/sysreg: Add missing system instruction definitions for FGT
  arm64/sysreg: Add missing system register definitions for FGT
  arm64/sysreg: Add missing ExtTrcBuff field definition to ID_AA64DFR0_EL1
  arm64/sysreg: Add missing Pauth_LR field definitions to ID_AA64ISAR1_EL1
  arm64: memory: remove duplicated include
  arm: perf: Fix ARCH=arm build with GCC
  arm64: Align boot cpucap handling with system cpucap handling
  arm64: Cleanup system cpucap handling
  MAINTAINERS: add maintainers for DesignWare PCIe PMU driver
  drivers/perf: add DesignWare PCIe PMU driver
  PCI: Move pci_clear_and_set_dword() helper to PCI header
  PCI: Add Alibaba Vendor ID to linux/pci_ids.h
  docs: perf: Add description for Synopsys DesignWare PCIe PMU driver
  arm64: irq: set the correct node for shadow call stack
  Revert "perf/arm_dmc620: Remove duplicate format attribute #defines"
  arm64: fpsimd: Implement lazy restore for kernel mode FPSIMD
  arm64: fpsimd: Preserve/restore kernel mode NEON at context switch
  ...

3 files changed, 17 insertions, 105 deletions

diff --git a/arch/arm/kernel/perf_event_v6.c b/arch/arm/kernel/perf_event_v6.c
index 1ae99deeec54..8fc080c9e4fb 100644
--- a/arch/arm/kernel/perf_event_v6.c
+++ b/arch/arm/kernel/perf_event_v6.c
@@ -268,10 +268,8 @@ static inline void armv6pmu_write_counter(struct perf_event *event, u64 value)
 
 static void armv6pmu_enable_event(struct perf_event *event)
 {
-	unsigned long val, mask, evt, flags;
-	struct arm_pmu *cpu_pmu = to_arm_pmu(event->pmu);
+	unsigned long val, mask, evt;
 	struct hw_perf_event *hwc = &event->hw;
-	struct pmu_hw_events *events = this_cpu_ptr(cpu_pmu->hw_events);
 	int idx = hwc->idx;
 
 	if (ARMV6_CYCLE_COUNTER == idx) {
@@ -294,12 +292,10 @@ static void armv6pmu_enable_event(struct perf_event *event)
 	 * Mask out the current event and set the counter to count the event
 	 * that we're interested in.
 	 */
-	raw_spin_lock_irqsave(&events->pmu_lock, flags);
 	val = armv6_pmcr_read();
 	val &= ~mask;
 	val |= evt;
 	armv6_pmcr_write(val);
-	raw_spin_unlock_irqrestore(&events->pmu_lock, flags);
 }
 
 static irqreturn_t
@@ -362,26 +358,20 @@ armv6pmu_handle_irq(struct arm_pmu *cpu_pmu)
 
 static void armv6pmu_start(struct arm_pmu *cpu_pmu)
 {
-	unsigned long flags, val;
-	struct pmu_hw_events *events = this_cpu_ptr(cpu_pmu->hw_events);
+	unsigned long val;
 
-	raw_spin_lock_irqsave(&events->pmu_lock, flags);
 	val = armv6_pmcr_read();
 	val |= ARMV6_PMCR_ENABLE;
 	armv6_pmcr_write(val);
-	raw_spin_unlock_irqrestore(&events->pmu_lock, flags);
 }
 
 static void armv6pmu_stop(struct arm_pmu *cpu_pmu)
 {
-	unsigned long flags, val;
-	struct pmu_hw_events *events = this_cpu_ptr(cpu_pmu->hw_events);
+	unsigned long val;
 
-	raw_spin_lock_irqsave(&events->pmu_lock, flags);
 	val = armv6_pmcr_read();
 	val &= ~ARMV6_PMCR_ENABLE;
 	armv6_pmcr_write(val);
-	raw_spin_unlock_irqrestore(&events->pmu_lock, flags);
 }
 
 static int
@@ -419,10 +409,8 @@ static void armv6pmu_clear_event_idx(struct pmu_hw_events *cpuc,
 
 static void armv6pmu_disable_event(struct perf_event *event)
 {
-	unsigned long val, mask, evt, flags;
-	struct arm_pmu *cpu_pmu = to_arm_pmu(event->pmu);
+	unsigned long val, mask, evt;
 	struct hw_perf_event *hwc = &event->hw;
-	struct pmu_hw_events *events = this_cpu_ptr(cpu_pmu->hw_events);
 	int idx = hwc->idx;
 
 	if (ARMV6_CYCLE_COUNTER == idx) {
@@ -444,20 +432,16 @@ static void armv6pmu_disable_event(struct perf_event *event)
 	 * of ETM bus signal assertion cycles. The external reporting should
 	 * be disabled and so this should never increment.
 	 */
-	raw_spin_lock_irqsave(&events->pmu_lock, flags);
 	val = armv6_pmcr_read();
 	val &= ~mask;
 	val |= evt;
 	armv6_pmcr_write(val);
-	raw_spin_unlock_irqrestore(&events->pmu_lock, flags);
 }
 
 static void armv6mpcore_pmu_disable_event(struct perf_event *event)
 {
-	unsigned long val, mask, flags, evt = 0;
-	struct arm_pmu *cpu_pmu = to_arm_pmu(event->pmu);
+	unsigned long val, mask, evt = 0;
 	struct hw_perf_event *hwc = &event->hw;
-	struct pmu_hw_events *events = this_cpu_ptr(cpu_pmu->hw_events);
 	int idx = hwc->idx;
 
 	if (ARMV6_CYCLE_COUNTER == idx) {
@@ -475,12 +459,10 @@ static void armv6mpcore_pmu_disable_event(struct perf_event *event)
 	 * Unlike UP ARMv6, we don't have a way of stopping the counters. We
 	 * simply disable the interrupt reporting.
 	 */
-	raw_spin_lock_irqsave(&events->pmu_lock, flags);
 	val = armv6_pmcr_read();
 	val &= ~mask;
 	val |= evt;
 	armv6_pmcr_write(val);
-	raw_spin_unlock_irqrestore(&events->pmu_lock, flags);
 }
 
 static int armv6_map_event(struct perf_event *event)
diff --git a/arch/arm/kernel/perf_event_v7.c b/arch/arm/kernel/perf_event_v7.c
index eb2190477da1..a3322e2b3ea4 100644
--- a/arch/arm/kernel/perf_event_v7.c
+++ b/arch/arm/kernel/perf_event_v7.c
@@ -870,10 +870,8 @@ static void armv7_pmnc_dump_regs(struct arm_pmu *cpu_pmu)
 
 static void armv7pmu_enable_event(struct perf_event *event)
 {
-	unsigned long flags;
 	struct hw_perf_event *hwc = &event->hw;
 	struct arm_pmu *cpu_pmu = to_arm_pmu(event->pmu);
-	struct pmu_hw_events *events = this_cpu_ptr(cpu_pmu->hw_events);
 	int idx = hwc->idx;
 
 	if (!armv7_pmnc_counter_valid(cpu_pmu, idx)) {
@@ -886,7 +884,6 @@ static void armv7pmu_enable_event(struct perf_event *event)
 	 * Enable counter and interrupt, and set the counter to count
 	 * the event that we're interested in.
 	 */
-	raw_spin_lock_irqsave(&events->pmu_lock, flags);
 
 	/*
 	 * Disable counter
@@ -910,16 +907,12 @@ static void armv7pmu_enable_event(struct perf_event *event)
 	 * Enable counter
 	 */
 	armv7_pmnc_enable_counter(idx);
-
-	raw_spin_unlock_irqrestore(&events->pmu_lock, flags);
 }
 
 static void armv7pmu_disable_event(struct perf_event *event)
 {
-	unsigned long flags;
 	struct hw_perf_event *hwc = &event->hw;
 	struct arm_pmu *cpu_pmu = to_arm_pmu(event->pmu);
-	struct pmu_hw_events *events = this_cpu_ptr(cpu_pmu->hw_events);
 	int idx = hwc->idx;
 
 	if (!armv7_pmnc_counter_valid(cpu_pmu, idx)) {
@@ -931,7 +924,6 @@ static void armv7pmu_disable_event(struct perf_event *event)
 	/*
 	 * Disable counter and interrupt
 	 */
-	raw_spin_lock_irqsave(&events->pmu_lock, flags);
 
 	/*
 	 * Disable counter
@@ -942,8 +934,6 @@ static void armv7pmu_disable_event(struct perf_event *event)
 	 * Disable interrupt for this counter
 	 */
 	armv7_pmnc_disable_intens(idx);
-
-	raw_spin_unlock_irqrestore(&events->pmu_lock, flags);
 }
 
 static irqreturn_t armv7pmu_handle_irq(struct arm_pmu *cpu_pmu)
@@ -1009,24 +999,14 @@ static irqreturn_t armv7pmu_handle_irq(struct arm_pmu *cpu_pmu)
 
 static void armv7pmu_start(struct arm_pmu *cpu_pmu)
 {
-	unsigned long flags;
-	struct pmu_hw_events *events = this_cpu_ptr(cpu_pmu->hw_events);
-
-	raw_spin_lock_irqsave(&events->pmu_lock, flags);
 	/* Enable all counters */
 	armv7_pmnc_write(armv7_pmnc_read() | ARMV7_PMNC_E);
-	raw_spin_unlock_irqrestore(&events->pmu_lock, flags);
 }
 
 static void armv7pmu_stop(struct arm_pmu *cpu_pmu)
 {
-	unsigned long flags;
-	struct pmu_hw_events *events = this_cpu_ptr(cpu_pmu->hw_events);
-
-	raw_spin_lock_irqsave(&events->pmu_lock, flags);
 	/* Disable all counters */
 	armv7_pmnc_write(armv7_pmnc_read() & ~ARMV7_PMNC_E);
-	raw_spin_unlock_irqrestore(&events->pmu_lock, flags);
 }
 
 static int armv7pmu_get_event_idx(struct pmu_hw_events *cpuc,
@@ -1072,8 +1052,10 @@ static int armv7pmu_set_event_filter(struct hw_perf_event *event,
 {
 	unsigned long config_base = 0;
 
-	if (attr->exclude_idle)
-		return -EPERM;
+	if (attr->exclude_idle) {
+		pr_debug("ARM performance counters do not support mode exclusion\n");
+		return -EOPNOTSUPP;
+	}
 	if (attr->exclude_user)
 		config_base |= ARMV7_EXCLUDE_USER;
 	if (attr->exclude_kernel)
@@ -1492,14 +1474,10 @@ static void krait_clearpmu(u32 config_base)
 
 static void krait_pmu_disable_event(struct perf_event *event)
 {
-	unsigned long flags;
 	struct hw_perf_event *hwc = &event->hw;
 	int idx = hwc->idx;
-	struct arm_pmu *cpu_pmu = to_arm_pmu(event->pmu);
-	struct pmu_hw_events *events = this_cpu_ptr(cpu_pmu->hw_events);
 
 	/* Disable counter and interrupt */
-	raw_spin_lock_irqsave(&events->pmu_lock, flags);
 
 	/* Disable counter */
 	armv7_pmnc_disable_counter(idx);
@@ -1512,23 +1490,17 @@ static void krait_pmu_disable_event(struct perf_event *event)
 
 	/* Disable interrupt for this counter */
 	armv7_pmnc_disable_intens(idx);
-
-	raw_spin_unlock_irqrestore(&events->pmu_lock, flags);
 }
 
 static void krait_pmu_enable_event(struct perf_event *event)
 {
-	unsigned long flags;
 	struct hw_perf_event *hwc = &event->hw;
 	int idx = hwc->idx;
-	struct arm_pmu *cpu_pmu = to_arm_pmu(event->pmu);
-	struct pmu_hw_events *events = this_cpu_ptr(cpu_pmu->hw_events);
 
 	/*
 	 * Enable counter and interrupt, and set the counter to count
 	 * the event that we're interested in.
 	 */
-	raw_spin_lock_irqsave(&events->pmu_lock, flags);
 
 	/* Disable counter */
 	armv7_pmnc_disable_counter(idx);
@@ -1548,8 +1520,6 @@ static void krait_pmu_enable_event(struct perf_event *event)
 
 	/* Enable counter */
 	armv7_pmnc_enable_counter(idx);
-
-	raw_spin_unlock_irqrestore(&events->pmu_lock, flags);
 }
 
 static void krait_pmu_reset(void *info)
@@ -1825,14 +1795,10 @@ static void scorpion_clearpmu(u32 config_base)
 
 static void scorpion_pmu_disable_event(struct perf_event *event)
 {
-	unsigned long flags;
 	struct hw_perf_event *hwc = &event->hw;
 	int idx = hwc->idx;
-	struct arm_pmu *cpu_pmu = to_arm_pmu(event->pmu);
-	struct pmu_hw_events *events = this_cpu_ptr(cpu_pmu->hw_events);
 
 	/* Disable counter and interrupt */
-	raw_spin_lock_irqsave(&events->pmu_lock, flags);
 
 	/* Disable counter */
 	armv7_pmnc_disable_counter(idx);
@@ -1845,23 +1811,17 @@ static void scorpion_pmu_disable_event(struct perf_event *event)
 
 	/* Disable interrupt for this counter */
 	armv7_pmnc_disable_intens(idx);
-
-	raw_spin_unlock_irqrestore(&events->pmu_lock, flags);
 }
 
 static void scorpion_pmu_enable_event(struct perf_event *event)
 {
-	unsigned long flags;
 	struct hw_perf_event *hwc = &event->hw;
 	int idx = hwc->idx;
-	struct arm_pmu *cpu_pmu = to_arm_pmu(event->pmu);
-	struct pmu_hw_events *events = this_cpu_ptr(cpu_pmu->hw_events);
 
 	/*
 	 * Enable counter and interrupt, and set the counter to count
 	 * the event that we're interested in.
 	 */
-	raw_spin_lock_irqsave(&events->pmu_lock, flags);
 
 	/* Disable counter */
 	armv7_pmnc_disable_counter(idx);
@@ -1881,8 +1841,6 @@ static void scorpion_pmu_enable_event(struct perf_event *event)
 
 	/* Enable counter */
 	armv7_pmnc_enable_counter(idx);
-
-	raw_spin_unlock_irqrestore(&events->pmu_lock, flags);
 }
 
 static void scorpion_pmu_reset(void *info)
diff --git a/arch/arm/kernel/perf_event_xscale.c b/arch/arm/kernel/perf_event_xscale.c
index f6cdcacfb96d..7a2ba1c689a7 100644
--- a/arch/arm/kernel/perf_event_xscale.c
+++ b/arch/arm/kernel/perf_event_xscale.c
@@ -203,10 +203,8 @@ xscale1pmu_handle_irq(struct arm_pmu *cpu_pmu)
 
 static void xscale1pmu_enable_event(struct perf_event *event)
 {
-	unsigned long val, mask, evt, flags;
-	struct arm_pmu *cpu_pmu = to_arm_pmu(event->pmu);
+	unsigned long val, mask, evt;
 	struct hw_perf_event *hwc = &event->hw;
-	struct pmu_hw_events *events = this_cpu_ptr(cpu_pmu->hw_events);
 	int idx = hwc->idx;
 
 	switch (idx) {
@@ -229,20 +227,16 @@ static void xscale1pmu_enable_event(struct perf_event *event)
 		return;
 	}
 
-	raw_spin_lock_irqsave(&events->pmu_lock, flags);
 	val = xscale1pmu_read_pmnc();
 	val &= ~mask;
 	val |= evt;
 	xscale1pmu_write_pmnc(val);
-	raw_spin_unlock_irqrestore(&events->pmu_lock, flags);
 }
 
 static void xscale1pmu_disable_event(struct perf_event *event)
 {
-	unsigned long val, mask, evt, flags;
-	struct arm_pmu *cpu_pmu = to_arm_pmu(event->pmu);
+	unsigned long val, mask, evt;
 	struct hw_perf_event *hwc = &event->hw;
-	struct pmu_hw_events *events = this_cpu_ptr(cpu_pmu->hw_events);
 	int idx = hwc->idx;
 
 	switch (idx) {
@@ -263,12 +257,10 @@ static void xscale1pmu_disable_event(struct perf_event *event)
 		return;
 	}
 
-	raw_spin_lock_irqsave(&events->pmu_lock, flags);
 	val = xscale1pmu_read_pmnc();
 	val &= ~mask;
 	val |= evt;
 	xscale1pmu_write_pmnc(val);
-	raw_spin_unlock_irqrestore(&events->pmu_lock, flags);
 }
 
 static int
@@ -300,26 +292,20 @@ static void xscalepmu_clear_event_idx(struct pmu_hw_events *cpuc,
 
 static void xscale1pmu_start(struct arm_pmu *cpu_pmu)
 {
-	unsigned long flags, val;
-	struct pmu_hw_events *events = this_cpu_ptr(cpu_pmu->hw_events);
+	unsigned long val;
 
-	raw_spin_lock_irqsave(&events->pmu_lock, flags);
 	val = xscale1pmu_read_pmnc();
 	val |= XSCALE_PMU_ENABLE;
 	xscale1pmu_write_pmnc(val);
-	raw_spin_unlock_irqrestore(&events->pmu_lock, flags);
 }
 
 static void xscale1pmu_stop(struct arm_pmu *cpu_pmu)
 {
-	unsigned long flags, val;
-	struct pmu_hw_events *events = this_cpu_ptr(cpu_pmu->hw_events);
+	unsigned long val;
 
-	raw_spin_lock_irqsave(&events->pmu_lock, flags);
 	val = xscale1pmu_read_pmnc();
 	val &= ~XSCALE_PMU_ENABLE;
 	xscale1pmu_write_pmnc(val);
-	raw_spin_unlock_irqrestore(&events->pmu_lock, flags);
 }
 
 static inline u64 xscale1pmu_read_counter(struct perf_event *event)
@@ -549,10 +535,8 @@ xscale2pmu_handle_irq(struct arm_pmu *cpu_pmu)
 
 static void xscale2pmu_enable_event(struct perf_event *event)
 {
-	unsigned long flags, ien, evtsel;
-	struct arm_pmu *cpu_pmu = to_arm_pmu(event->pmu);
+	unsigned long ien, evtsel;
 	struct hw_perf_event *hwc = &event->hw;
-	struct pmu_hw_events *events = this_cpu_ptr(cpu_pmu->hw_events);
 	int idx = hwc->idx;
 
 	ien = xscale2pmu_read_int_enable();
@@ -587,18 +571,14 @@ static void xscale2pmu_enable_event(struct perf_event *event)
 		return;
 	}
 
-	raw_spin_lock_irqsave(&events->pmu_lock, flags);
 	xscale2pmu_write_event_select(evtsel);
 	xscale2pmu_write_int_enable(ien);
-	raw_spin_unlock_irqrestore(&events->pmu_lock, flags);
 }
 
 static void xscale2pmu_disable_event(struct perf_event *event)
 {
-	unsigned long flags, ien, evtsel, of_flags;
-	struct arm_pmu *cpu_pmu = to_arm_pmu(event->pmu);
+	unsigned long ien, evtsel, of_flags;
 	struct hw_perf_event *hwc = &event->hw;
-	struct pmu_hw_events *events = this_cpu_ptr(cpu_pmu->hw_events);
 	int idx = hwc->idx;
 
 	ien = xscale2pmu_read_int_enable();
@@ -638,11 +618,9 @@ static void xscale2pmu_disable_event(struct perf_event *event)
 		return;
 	}
 
-	raw_spin_lock_irqsave(&events->pmu_lock, flags);
 	xscale2pmu_write_event_select(evtsel);
 	xscale2pmu_write_int_enable(ien);
 	xscale2pmu_write_overflow_flags(of_flags);
-	raw_spin_unlock_irqrestore(&events->pmu_lock, flags);
 }
 
 static int
@@ -663,26 +641,20 @@ out:
 
 static void xscale2pmu_start(struct arm_pmu *cpu_pmu)
 {
-	unsigned long flags, val;
-	struct pmu_hw_events *events = this_cpu_ptr(cpu_pmu->hw_events);
+	unsigned long val;
 
-	raw_spin_lock_irqsave(&events->pmu_lock, flags);
 	val = xscale2pmu_read_pmnc() & ~XSCALE_PMU_CNT64;
 	val |= XSCALE_PMU_ENABLE;
 	xscale2pmu_write_pmnc(val);
-	raw_spin_unlock_irqrestore(&events->pmu_lock, flags);
 }
 
 static void xscale2pmu_stop(struct arm_pmu *cpu_pmu)
 {
-	unsigned long flags, val;
-	struct pmu_hw_events *events = this_cpu_ptr(cpu_pmu->hw_events);
+	unsigned long val;
 
-	raw_spin_lock_irqsave(&events->pmu_lock, flags);
 	val = xscale2pmu_read_pmnc();
 	val &= ~XSCALE_PMU_ENABLE;
 	xscale2pmu_write_pmnc(val);
-	raw_spin_unlock_irqrestore(&events->pmu_lock, flags);
 }
 
 static inline u64 xscale2pmu_read_counter(struct perf_event *event)