Merge: perf: sync with upstream v6.17

MR: https://gitlab.com/redhat/centos-stream/src/kernel/centos-stream-10/-/merge_requests/1556

JIRA: https://issues.redhat.com/browse/RHEL-78200

Sync perf codebase with upstream v6.17.

Signed-off-by: Anubhav Shelat <ashelat@redhat.com>

Approved-by: Michael Petlan <mpetlan@redhat.com>
Approved-by: Steve Best <sbest@redhat.com>
Approved-by: jbrnak <jbrnak@redhat.com>
Approved-by: CKI KWF Bot <cki-ci-bot+kwf-gitlab-com@redhat.com>

Merged-by: CKI GitLab Kmaint Pipeline Bot <26919896-cki-kmaint-pipeline-bot@users.noreply.gitlab.com>

Author: CKI KWF Bot

Documentation/admin-guide/syscall-user-dispatch.rst

@@ -53,20 +53,25 @@ following prctl:
 
   prctl(PR_SET_SYSCALL_USER_DISPATCH, <op>, <offset>, <length>, [selector])
 
-<op> is either PR_SYS_DISPATCH_ON or PR_SYS_DISPATCH_OFF, to enable and
-disable the mechanism globally for that thread.  When
-PR_SYS_DISPATCH_OFF is used, the other fields must be zero.
-
-[<offset>, <offset>+<length>) delimit a memory region interval
-from which syscalls are always executed directly, regardless of the
-userspace selector.  This provides a fast path for the C library, which
-includes the most common syscall dispatchers in the native code
-applications, and also provides a way for the signal handler to return
+<op> is either PR_SYS_DISPATCH_EXCLUSIVE_ON/PR_SYS_DISPATCH_INCLUSIVE_ON
+or PR_SYS_DISPATCH_OFF, to enable and disable the mechanism globally for
+that thread.  When PR_SYS_DISPATCH_OFF is used, the other fields must be zero.
+
+For PR_SYS_DISPATCH_EXCLUSIVE_ON [<offset>, <offset>+<length>) delimit
+a memory region interval from which syscalls are always executed directly,
+regardless of the userspace selector.  This provides a fast path for the
+C library, which includes the most common syscall dispatchers in the native
+code applications, and also provides a way for the signal handler to return
 without triggering a nested SIGSYS on (rt\_)sigreturn.  Users of this
 interface should make sure that at least the signal trampoline code is
 included in this region. In addition, for syscalls that implement the
 trampoline code on the vDSO, that trampoline is never intercepted.
 
+For PR_SYS_DISPATCH_INCLUSIVE_ON [<offset>, <offset>+<length>) delimit
+a memory region interval from which syscalls are dispatched based on
+the userspace selector. Syscalls from outside of the range are always
+executed directly.
+
 [selector] is a pointer to a char-sized region in the process memory
 region, that provides a quick way to enable disable syscall redirection
 thread-wide, without the need to invoke the kernel directly.  selector

arch/powerpc/kernel/eeh_driver.c

@@ -907,7 +907,7 @@ void eeh_handle_normal_event(struct eeh_pe *pe)
 		/* FIXME: Use the same format as dump_stack() */
 		pr_err("EEH: Call Trace:\n");
 		for (i = 0; i < pe->trace_entries; i++)
-			pr_err("EEH: [%pK] %pS\n", ptrs[i], ptrs[i]);
+			pr_err("EEH: [%p] %pS\n", ptrs[i], ptrs[i]);
 
 		pe->trace_entries = 0;
 	}

arch/powerpc/perf/hv-24x7.c

@@ -713,27 +713,27 @@ static ssize_t catalog_event_len_validate(struct hv_24x7_event_data *event,
 	ev_len = be16_to_cpu(event->length);
 
 	if (ev_len % 16)
-		pr_info("event %zu has length %zu not divisible by 16: event=%pK\n",
+		pr_info("event %zu has length %zu not divisible by 16: event=%p\n",
 				event_idx, ev_len, event);
 
 	ev_end = (__u8 *)event + ev_len;
 	if (ev_end > end) {
-		pr_warn("event %zu has .length=%zu, ends after buffer end: ev_end=%pK > end=%pK, offset=%zu\n",
+		pr_warn("event %zu has .length=%zu, ends after buffer end: ev_end=%p > end=%p, offset=%zu\n",
 				event_idx, ev_len, ev_end, end,
 				offset);
 		return -1;
 	}
 
 	calc_ev_end = event_end(event, end);
 	if (!calc_ev_end) {
-		pr_warn("event %zu has a calculated length which exceeds buffer length %zu: event=%pK end=%pK, offset=%zu\n",
+		pr_warn("event %zu has a calculated length which exceeds buffer length %zu: event=%p end=%p, offset=%zu\n",
 			event_idx, event_data_bytes, event, end,
 			offset);
 		return -1;
 	}
 
 	if (calc_ev_end > ev_end) {
-		pr_warn("event %zu exceeds its own length: event=%pK, end=%pK, offset=%zu, calc_ev_end=%pK\n",
+		pr_warn("event %zu exceeds its own length: event=%p, end=%p, offset=%zu, calc_ev_end=%p\n",
 			event_idx, event, ev_end, offset, calc_ev_end);
 		return -1;
 	}

arch/x86/events/intel/uncore.c

@@ -1814,6 +1814,12 @@ static const struct intel_uncore_init_fun lnl_uncore_init __initconst = {
 	.mmio_init = lnl_uncore_mmio_init,
 };
 
+static const struct intel_uncore_init_fun ptl_uncore_init __initconst = {
+	.cpu_init = ptl_uncore_cpu_init,
+	.mmio_init = ptl_uncore_mmio_init,
+	.use_discovery = true,
+};
+
 static const struct intel_uncore_init_fun icx_uncore_init __initconst = {
 	.cpu_init = icx_uncore_cpu_init,
 	.pci_init = icx_uncore_pci_init,
@@ -1895,6 +1901,7 @@ static const struct x86_cpu_id intel_uncore_match[] __initconst = {
 	X86_MATCH_VFM(INTEL_ARROWLAKE_U,	&mtl_uncore_init),
 	X86_MATCH_VFM(INTEL_ARROWLAKE_H,	&mtl_uncore_init),
 	X86_MATCH_VFM(INTEL_LUNARLAKE_M,	&lnl_uncore_init),
+	X86_MATCH_VFM(INTEL_PANTHERLAKE_L,	&ptl_uncore_init),
 	X86_MATCH_VFM(INTEL_SAPPHIRERAPIDS_X,	&spr_uncore_init),
 	X86_MATCH_VFM(INTEL_EMERALDRAPIDS_X,	&spr_uncore_init),
 	X86_MATCH_VFM(INTEL_GRANITERAPIDS_X,	&gnr_uncore_init),

arch/x86/events/intel/uncore.h

@@ -612,10 +612,12 @@ void tgl_uncore_cpu_init(void);
 void adl_uncore_cpu_init(void);
 void lnl_uncore_cpu_init(void);
 void mtl_uncore_cpu_init(void);
+void ptl_uncore_cpu_init(void);
 void tgl_uncore_mmio_init(void);
 void tgl_l_uncore_mmio_init(void);
 void adl_uncore_mmio_init(void);
 void lnl_uncore_mmio_init(void);
+void ptl_uncore_mmio_init(void);
 int snb_pci2phy_map_init(int devid);
 
 /* uncore_snbep.c */

arch/x86/events/intel/uncore_discovery.c

@@ -273,32 +273,15 @@ uncore_ignore_unit(struct uncore_unit_discovery *unit, int *ignore)
 	return false;
 }
 
-static int parse_discovery_table(struct pci_dev *dev, int die,
-				 u32 bar_offset, bool *parsed,
-				 int *ignore)
+static int __parse_discovery_table(resource_size_t addr, int die,
+				   bool *parsed, int *ignore)
 {
 	struct uncore_global_discovery global;
 	struct uncore_unit_discovery unit;
 	void __iomem *io_addr;
-	resource_size_t addr;
 	unsigned long size;
-	u32 val;
 	int i;
 
-	pci_read_config_dword(dev, bar_offset, &val);
-
-	if (val & ~PCI_BASE_ADDRESS_MEM_MASK & ~PCI_BASE_ADDRESS_MEM_TYPE_64)
-		return -EINVAL;
-
-	addr = (resource_size_t)(val & PCI_BASE_ADDRESS_MEM_MASK);
-#ifdef CONFIG_PHYS_ADDR_T_64BIT
-	if ((val & PCI_BASE_ADDRESS_MEM_TYPE_MASK) == PCI_BASE_ADDRESS_MEM_TYPE_64) {
-		u32 val2;
-
-		pci_read_config_dword(dev, bar_offset + 4, &val2);
-		addr |= ((resource_size_t)val2) << 32;
-	}
-#endif
 	size = UNCORE_DISCOVERY_GLOBAL_MAP_SIZE;
 	io_addr = ioremap(addr, size);
 	if (!io_addr)
@@ -341,7 +324,32 @@ static int parse_discovery_table(struct pci_dev *dev, int die,
 	return 0;
 }
 
-bool intel_uncore_has_discovery_tables(int *ignore)
+static int parse_discovery_table(struct pci_dev *dev, int die,
+				 u32 bar_offset, bool *parsed,
+				 int *ignore)
+{
+	resource_size_t addr;
+	u32 val;
+
+	pci_read_config_dword(dev, bar_offset, &val);
+
+	if (val & ~PCI_BASE_ADDRESS_MEM_MASK & ~PCI_BASE_ADDRESS_MEM_TYPE_64)
+		return -EINVAL;
+
+	addr = (resource_size_t)(val & PCI_BASE_ADDRESS_MEM_MASK);
+#ifdef CONFIG_PHYS_ADDR_T_64BIT
+	if ((val & PCI_BASE_ADDRESS_MEM_TYPE_MASK) == PCI_BASE_ADDRESS_MEM_TYPE_64) {
+		u32 val2;
+
+		pci_read_config_dword(dev, bar_offset + 4, &val2);
+		addr |= ((resource_size_t)val2) << 32;
+	}
+#endif
+
+	return __parse_discovery_table(addr, die, parsed, ignore);
+}
+
+static bool intel_uncore_has_discovery_tables_pci(int *ignore)
 {
 	u32 device, val, entry_id, bar_offset;
 	int die, dvsec = 0, ret = true;
@@ -390,6 +398,45 @@ bool intel_uncore_has_discovery_tables(int *ignore)
 	return ret;
 }
 
+static bool intel_uncore_has_discovery_tables_msr(int *ignore)
+{
+	unsigned long *die_mask;
+	bool parsed = false;
+	int cpu, die;
+	u64 base;
+
+	die_mask = kcalloc(BITS_TO_LONGS(uncore_max_dies()),
+			   sizeof(unsigned long), GFP_KERNEL);
+	if (!die_mask)
+		return false;
+
+	cpus_read_lock();
+	for_each_online_cpu(cpu) {
+		die = topology_logical_die_id(cpu);
+		if (__test_and_set_bit(die, die_mask))
+			continue;
+
+		if (rdmsrl_safe_on_cpu(cpu, UNCORE_DISCOVERY_MSR, &base))
+			continue;
+
+		if (!base)
+			continue;
+
+		__parse_discovery_table(base, die, &parsed, ignore);
+	}
+
+	cpus_read_unlock();
+
+	kfree(die_mask);
+	return parsed;
+}
+
+bool intel_uncore_has_discovery_tables(int *ignore)
+{
+	return intel_uncore_has_discovery_tables_msr(ignore) ||
+	       intel_uncore_has_discovery_tables_pci(ignore);
+}
+
 void intel_uncore_clear_discovery_tables(void)
 {
 	struct intel_uncore_discovery_type *type, *next;
@@ -603,7 +650,7 @@ void intel_generic_uncore_mmio_init_box(struct intel_uncore_box *box)
 	}
 
 	addr = unit->addr;
-	box->io_addr = ioremap(addr, UNCORE_GENERIC_MMIO_SIZE);
+	box->io_addr = ioremap(addr, type->mmio_map_size);
 	if (!box->io_addr) {
 		pr_warn("Uncore type %d box %d: ioremap error for 0x%llx.\n",
 			type->type_id, unit->id, (unsigned long long)addr);

arch/x86/events/intel/uncore_discovery.h

@@ -1,5 +1,8 @@
 /* SPDX-License-Identifier: GPL-2.0-only */
 
+/* Store the full address of the global discovery table */
+#define UNCORE_DISCOVERY_MSR			0x201e
+
 /* Generic device ID of a discovery table device */
 #define UNCORE_DISCOVERY_TABLE_DEVICE		0x09a7
 /* Capability ID for a discovery table device */
@@ -168,3 +171,7 @@ bool intel_generic_uncore_assign_hw_event(struct perf_event *event,
 					  struct intel_uncore_box *box);
 void uncore_find_add_unit(struct intel_uncore_discovery_unit *node,
 			  struct rb_root *root, u16 *num_units);
+struct intel_uncore_type **
+uncore_get_uncores(enum uncore_access_type type_id, int num_extra,
+		   struct intel_uncore_type **extra, int max_num_types,
+		   struct intel_uncore_type **uncores);

arch/x86/events/intel/uncore_snb.c

@@ -1854,3 +1854,82 @@ void lnl_uncore_mmio_init(void)
 }
 
 /* end of Lunar Lake MMIO uncore support */
+
+/* Panther Lake uncore support */
+
+#define UNCORE_PTL_MAX_NUM_UNCORE_TYPES		42
+#define UNCORE_PTL_TYPE_IMC			6
+#define UNCORE_PTL_TYPE_SNCU			34
+#define UNCORE_PTL_TYPE_HBO			41
+
+#define PTL_UNCORE_GLOBAL_CTL_OFFSET		0x380
+
+static struct intel_uncore_type ptl_uncore_imc = {
+	.name			= "imc",
+	.mmio_map_size		= 0xf00,
+};
+
+static void ptl_uncore_sncu_init_box(struct intel_uncore_box *box)
+{
+	intel_generic_uncore_mmio_init_box(box);
+
+	/* Clear the global freeze bit */
+	if (box->io_addr)
+		writel(0, box->io_addr + PTL_UNCORE_GLOBAL_CTL_OFFSET);
+}
+
+static struct intel_uncore_ops ptl_uncore_sncu_ops = {
+	.init_box		= ptl_uncore_sncu_init_box,
+	.exit_box		= uncore_mmio_exit_box,
+	.disable_box		= intel_generic_uncore_mmio_disable_box,
+	.enable_box		= intel_generic_uncore_mmio_enable_box,
+	.disable_event		= intel_generic_uncore_mmio_disable_event,
+	.enable_event		= intel_generic_uncore_mmio_enable_event,
+	.read_counter		= uncore_mmio_read_counter,
+};
+
+static struct intel_uncore_type ptl_uncore_sncu = {
+	.name			= "sncu",
+	.ops			= &ptl_uncore_sncu_ops,
+	.mmio_map_size		= 0xf00,
+};
+
+static struct intel_uncore_type ptl_uncore_hbo = {
+	.name			= "hbo",
+	.mmio_map_size		= 0xf00,
+};
+
+static struct intel_uncore_type *ptl_uncores[UNCORE_PTL_MAX_NUM_UNCORE_TYPES] = {
+	[UNCORE_PTL_TYPE_IMC] = &ptl_uncore_imc,
+	[UNCORE_PTL_TYPE_SNCU] = &ptl_uncore_sncu,
+	[UNCORE_PTL_TYPE_HBO] = &ptl_uncore_hbo,
+};
+
+#define UNCORE_PTL_MMIO_EXTRA_UNCORES		1
+
+static struct intel_uncore_type *ptl_mmio_extra_uncores[UNCORE_PTL_MMIO_EXTRA_UNCORES] = {
+	&adl_uncore_imc_free_running,
+};
+
+void ptl_uncore_mmio_init(void)
+{
+	uncore_mmio_uncores = uncore_get_uncores(UNCORE_ACCESS_MMIO,
+						 UNCORE_PTL_MMIO_EXTRA_UNCORES,
+						 ptl_mmio_extra_uncores,
+						 UNCORE_PTL_MAX_NUM_UNCORE_TYPES,
+						 ptl_uncores);
+}
+
+static struct intel_uncore_type *ptl_msr_uncores[] = {
+	&mtl_uncore_cbox,
+	NULL
+};
+
+void ptl_uncore_cpu_init(void)
+{
+	mtl_uncore_cbox.num_boxes = 6;
+	mtl_uncore_cbox.ops = &lnl_uncore_msr_ops;
+	uncore_msr_uncores = ptl_msr_uncores;
+}
+
+/* end of Panther Lake uncore support */

arch/x86/events/intel/uncore_snbep.c

@@ -6408,9 +6408,11 @@ static void uncore_type_customized_copy(struct intel_uncore_type *to_type,
 		to_type->get_topology = from_type->get_topology;
 	if (from_type->cleanup_mapping)
 		to_type->cleanup_mapping = from_type->cleanup_mapping;
+	if (from_type->mmio_map_size)
+		to_type->mmio_map_size = from_type->mmio_map_size;
 }
 
-static struct intel_uncore_type **
+struct intel_uncore_type **
 uncore_get_uncores(enum uncore_access_type type_id, int num_extra,
 		   struct intel_uncore_type **extra, int max_num_types,
 		   struct intel_uncore_type **uncores)

drivers/perf/arm_pmu.c

@@ -342,12 +342,10 @@ armpmu_add(struct perf_event *event, int flags)
 	if (idx < 0)
 		return idx;
 
-	/*
-	 * If there is an event in the counter we are going to use then make
-	 * sure it is disabled.
-	 */
+	/* The newly-allocated counter should be empty */
+	WARN_ON_ONCE(hw_events->events[idx]);
+
 	event->hw.idx = idx;
-	armpmu->disable(event);
 	hw_events->events[idx] = event;
 
 	hwc->state = PERF_HES_STOPPED | PERF_HES_UPTODATE;