Automatic merge of 'next-test' into merge-test (2024-06-19 21:38)

arch/powerpc/configs/ppc6xx_defconfig

@@ -12,7 +12,6 @@ CONFIG_TASK_XACCT=y
 CONFIG_TASK_IO_ACCOUNTING=y
 CONFIG_CGROUPS=y
 CONFIG_CGROUP_SCHED=y
-CONFIG_RT_GROUP_SCHED=y
 CONFIG_CGROUP_DEVICE=y
 CONFIG_CGROUP_CPUACCT=y
 CONFIG_USER_NS=y

arch/powerpc/include/asm/kexec.h

@@ -103,10 +103,8 @@ int load_crashdump_segments_ppc64(struct kimage *image,
 int setup_purgatory_ppc64(struct kimage *image, const void *slave_code,
 			  const void *fdt, unsigned long kernel_load_addr,
 			  unsigned long fdt_load_addr);
-unsigned int kexec_extra_fdt_size_ppc64(struct kimage *image);
-int setup_new_fdt_ppc64(const struct kimage *image, void *fdt,
-			unsigned long initrd_load_addr,
-			unsigned long initrd_len, const char *cmdline);
+unsigned int kexec_extra_fdt_size_ppc64(struct kimage *image, struct crash_mem *rmem);
+int setup_new_fdt_ppc64(const struct kimage *image, void *fdt, struct crash_mem *rmem);
 #endif /* CONFIG_PPC64 */
 
 #endif /* CONFIG_KEXEC_FILE */

arch/powerpc/include/asm/perf_event_server.h

@@ -89,7 +89,8 @@ struct power_pmu {
 #define PPMU_NO_SIAR		0x00000100 /* Do not use SIAR */
 #define PPMU_ARCH_31		0x00000200 /* Has MMCR3, SIER2 and SIER3 */
 #define PPMU_P10_DD1		0x00000400 /* Is power10 DD1 processor version */
-#define PPMU_HAS_ATTR_CONFIG1	0x00000800 /* Using config1 attribute */
+#define PPMU_P10		0x00000800 /* For power10 pmu */
+#define PPMU_HAS_ATTR_CONFIG1	0x00001000 /* Using config1 attribute */
 
 /*
  * Values for flags to get_alternatives()

arch/powerpc/kexec/core_64.c

@@ -456,9 +456,15 @@ static int add_node_props(void *fdt, int node_offset, const struct device_node *
  * @fdt:              Flattened device tree of the kernel.
  *
  * Returns 0 on success, negative errno on error.
+ *
+ * Note: expecting no subnodes under /cpus/<node> with device_type == "cpu".
+ * If this changes, update this function to include them.
  */
 int update_cpus_node(void *fdt)
 {
+	int prev_node_offset;
+	const char *device_type;
+	const struct fdt_property *prop;
 	struct device_node *cpus_node, *dn;
 	int cpus_offset, cpus_subnode_offset, ret = 0;
 
@@ -469,30 +475,44 @@ int update_cpus_node(void *fdt)
 		return cpus_offset;
 	}
 
-	if (cpus_offset > 0) {
-		ret = fdt_del_node(fdt, cpus_offset);
+	prev_node_offset = cpus_offset;
+	/* Delete sub-nodes of /cpus node with device_type == "cpu" */
+	for (cpus_subnode_offset = fdt_first_subnode(fdt, cpus_offset); cpus_subnode_offset >= 0;) {
+		/* Ignore nodes that do not have a device_type property or device_type != "cpu" */
+		prop = fdt_get_property(fdt, cpus_subnode_offset, "device_type", NULL);
+		if (!prop || strcmp(prop->data, "cpu")) {
+			prev_node_offset = cpus_subnode_offset;
+			goto next_node;
+		}
+
+		ret = fdt_del_node(fdt, cpus_subnode_offset);
 		if (ret < 0) {
-			pr_err("Error deleting /cpus node: %s\n", fdt_strerror(ret));
-			return -EINVAL;
+			pr_err("Failed to delete a cpus sub-node: %s\n", fdt_strerror(ret));
+			return ret;
 		}
+next_node:
+		if (prev_node_offset == cpus_offset)
+			cpus_subnode_offset = fdt_first_subnode(fdt, cpus_offset);
+		else
+			cpus_subnode_offset = fdt_next_subnode(fdt, prev_node_offset);
 	}
 
-	/* Add cpus node to fdt */
-	cpus_offset = fdt_add_subnode(fdt, fdt_path_offset(fdt, "/"), "cpus");
-	if (cpus_offset < 0) {
-		pr_err("Error creating /cpus node: %s\n", fdt_strerror(cpus_offset));
+	cpus_node = of_find_node_by_path("/cpus");
+	/* Fail here to avoid kexec/kdump kernel boot hung */
+	if (!cpus_node) {
+		pr_err("No /cpus node found\n");
 		return -EINVAL;
 	}
 
-	/* Add cpus node properties */
-	cpus_node = of_find_node_by_path("/cpus");
-	ret = add_node_props(fdt, cpus_offset, cpus_node);
-	of_node_put(cpus_node);
-	if (ret < 0)
-		return ret;
+	/* Add all /cpus sub-nodes of device_type == "cpu" to FDT */
+	for_each_child_of_node(cpus_node, dn) {
+		/* Ignore device nodes that do not have a device_type property
+		 * or device_type != "cpu".
+		 */
+		device_type = of_get_property(dn, "device_type", NULL);
+		if (!device_type || strcmp(device_type, "cpu"))
+			continue;
 
-	/* Loop through all subnodes of cpus and add them to fdt */
-	for_each_node_by_type(dn, "cpu") {
 		cpus_subnode_offset = fdt_add_subnode(fdt, cpus_offset, dn->full_name);
 		if (cpus_subnode_offset < 0) {
 			pr_err("Unable to add %s subnode: %s\n", dn->full_name,
@@ -506,6 +526,7 @@ int update_cpus_node(void *fdt)
 			goto out;
 	}
 out:
+	of_node_put(cpus_node);
 	of_node_put(dn);
 	return ret;
 }

arch/powerpc/kexec/elf_64.c

@@ -23,6 +23,7 @@
 #include <linux/of_fdt.h>
 #include <linux/slab.h>
 #include <linux/types.h>
+#include <asm/kexec_ranges.h>
 
 static void *elf64_load(struct kimage *image, char *kernel_buf,
 			unsigned long kernel_len, char *initrd,
@@ -36,6 +37,7 @@ static void *elf64_load(struct kimage *image, char *kernel_buf,
 	const void *slave_code;
 	struct elfhdr ehdr;
 	char *modified_cmdline = NULL;
+	struct crash_mem *rmem = NULL;
 	struct kexec_elf_info elf_info;
 	struct kexec_buf kbuf = { .image = image, .buf_min = 0,
 				  .buf_max = ppc64_rma_size };
@@ -102,17 +104,20 @@ static void *elf64_load(struct kimage *image, char *kernel_buf,
 		kexec_dprintk("Loaded initrd at 0x%lx\n", initrd_load_addr);
 	}
 
+	ret = get_reserved_memory_ranges(&rmem);
+	if (ret)
+		goto out;
+
 	fdt = of_kexec_alloc_and_setup_fdt(image, initrd_load_addr,
 					   initrd_len, cmdline,
-					   kexec_extra_fdt_size_ppc64(image));
+					   kexec_extra_fdt_size_ppc64(image, rmem));
 	if (!fdt) {
 		pr_err("Error setting up the new device tree.\n");
 		ret = -EINVAL;
 		goto out;
 	}
 
-	ret = setup_new_fdt_ppc64(image, fdt, initrd_load_addr,
-				  initrd_len, cmdline);
+	ret = setup_new_fdt_ppc64(image, fdt, rmem);
 	if (ret)
 		goto out_free_fdt;
 
@@ -146,6 +151,7 @@ static void *elf64_load(struct kimage *image, char *kernel_buf,
 out_free_fdt:
 	kvfree(fdt);
 out:
+	kfree(rmem);
 	kfree(modified_cmdline);
 	kexec_free_elf_info(&elf_info);
 

arch/powerpc/kexec/file_load_64.c

@@ -803,10 +803,9 @@ static unsigned int cpu_node_size(void)
 	return size;
 }
 
-static unsigned int kdump_extra_fdt_size_ppc64(struct kimage *image)
+static unsigned int kdump_extra_fdt_size_ppc64(struct kimage *image, unsigned int cpu_nodes)
 {
-	unsigned int cpu_nodes, extra_size = 0;
-	struct device_node *dn;
+	unsigned int extra_size = 0;
 	u64 usm_entries;
 #ifdef CONFIG_CRASH_HOTPLUG
 	unsigned int possible_cpu_nodes;
@@ -826,18 +825,6 @@ static unsigned int kdump_extra_fdt_size_ppc64(struct kimage *image)
 		extra_size += (unsigned int)(usm_entries * sizeof(u64));
 	}
 
-	/*
-	 * Get the number of CPU nodes in the current DT. This allows to
-	 * reserve places for CPU nodes added since the boot time.
-	 */
-	cpu_nodes = 0;
-	for_each_node_by_type(dn, "cpu") {
-		cpu_nodes++;
-	}
-
-	if (cpu_nodes > boot_cpu_node_count)
-		extra_size += (cpu_nodes - boot_cpu_node_count) * cpu_node_size();
-
 #ifdef CONFIG_CRASH_HOTPLUG
 	/*
 	 * Make sure enough space is reserved to accommodate possible CPU nodes
@@ -861,16 +848,30 @@ static unsigned int kdump_extra_fdt_size_ppc64(struct kimage *image)
  *
  * Returns the estimated extra size needed for kexec/kdump kernel FDT.
  */
-unsigned int kexec_extra_fdt_size_ppc64(struct kimage *image)
+unsigned int kexec_extra_fdt_size_ppc64(struct kimage *image, struct crash_mem *rmem)
 {
-	unsigned int extra_size = 0;
+	struct device_node *dn;
+	unsigned int cpu_nodes = 0, extra_size = 0;
 
 	// Budget some space for the password blob. There's already extra space
 	// for the key name
 	if (plpks_is_available())
 		extra_size += (unsigned int)plpks_get_passwordlen();
 
-	return extra_size + kdump_extra_fdt_size_ppc64(image);
+	/* Get the number of CPU nodes in the current device tree */
+	for_each_node_by_type(dn, "cpu") {
+		cpu_nodes++;
+	}
+
+	/* Consider extra space for CPU nodes added since the boot time */
+	if (cpu_nodes > boot_cpu_node_count)
+		extra_size += (cpu_nodes - boot_cpu_node_count) * cpu_node_size();
+
+	/* Consider extra space for reserved memory ranges if any */
+	if (rmem->nr_ranges > 0)
+		extra_size += sizeof(struct fdt_reserve_entry) * rmem->nr_ranges;
+
+	return extra_size + kdump_extra_fdt_size_ppc64(image, cpu_nodes);
 }
 
 static int copy_property(void *fdt, int node_offset, const struct device_node *dn,
@@ -924,18 +925,13 @@ static int update_pci_dma_nodes(void *fdt, const char *dmapropname)
  *                       being loaded.
  * @image:               kexec image being loaded.
  * @fdt:                 Flattened device tree for the next kernel.
- * @initrd_load_addr:    Address where the next initrd will be loaded.
- * @initrd_len:          Size of the next initrd, or 0 if there will be none.
- * @cmdline:             Command line for the next kernel, or NULL if there will
- *                       be none.
+ * @rmem:                Reserved memory ranges.
  *
  * Returns 0 on success, negative errno on error.
  */
-int setup_new_fdt_ppc64(const struct kimage *image, void *fdt,
-			unsigned long initrd_load_addr,
-			unsigned long initrd_len, const char *cmdline)
+int setup_new_fdt_ppc64(const struct kimage *image, void *fdt, struct crash_mem *rmem)
 {
-	struct crash_mem *umem = NULL, *rmem = NULL;
+	struct crash_mem *umem = NULL;
 	int i, nr_ranges, ret;
 
 #ifdef CONFIG_CRASH_DUMP
@@ -991,10 +987,6 @@ int setup_new_fdt_ppc64(const struct kimage *image, void *fdt,
 		goto out;
 
 	/* Update memory reserve map */
-	ret = get_reserved_memory_ranges(&rmem);
-	if (ret)
-		goto out;
-
 	nr_ranges = rmem ? rmem->nr_ranges : 0;
 	for (i = 0; i < nr_ranges; i++) {
 		u64 base, size;
@@ -1014,7 +1006,6 @@ int setup_new_fdt_ppc64(const struct kimage *image, void *fdt,
 		ret = plpks_populate_fdt(fdt);
 
 out:
-	kfree(rmem);
 	kfree(umem);
 	return ret;
 }

arch/powerpc/perf/core-book3s.c

@@ -266,51 +266,44 @@ static inline u32 perf_flags_from_msr(struct pt_regs *regs)
 static inline u32 perf_get_misc_flags(struct pt_regs *regs)
 {
 	bool use_siar = regs_use_siar(regs);
-	unsigned long mmcra = regs->dsisr;
-	int marked = mmcra & MMCRA_SAMPLE_ENABLE;
+	unsigned long siar;
+	unsigned long addr;
 
 	if (!use_siar)
 		return perf_flags_from_msr(regs);
 
-	/*
-	 * Check the address in SIAR to identify the
-	 * privilege levels since the SIER[MSR_HV, MSR_PR]
-	 * bits are not set for marked events in power10
-	 * DD1.
-	 */
-	if (marked && (ppmu->flags & PPMU_P10_DD1)) {
-		unsigned long siar = mfspr(SPRN_SIAR);
-		if (siar) {
-			if (is_kernel_addr(siar))
-				return PERF_RECORD_MISC_KERNEL;
-			return PERF_RECORD_MISC_USER;
-		} else {
-			if (is_kernel_addr(regs->nip))
-				return PERF_RECORD_MISC_KERNEL;
-			return PERF_RECORD_MISC_USER;
-		}
-	}
-
 	/*
 	 * If we don't have flags in MMCRA, rather than using
 	 * the MSR, we intuit the flags from the address in
 	 * SIAR which should give slightly more reliable
 	 * results
 	 */
 	if (ppmu->flags & PPMU_NO_SIPR) {
-		unsigned long siar = mfspr(SPRN_SIAR);
+		siar = mfspr(SPRN_SIAR);
 		if (is_kernel_addr(siar))
 			return PERF_RECORD_MISC_KERNEL;
 		return PERF_RECORD_MISC_USER;
 	}
 
 	/* PR has priority over HV, so order below is important */
-	if (regs_sipr(regs))
-		return PERF_RECORD_MISC_USER;
-
-	if (regs_sihv(regs) && (freeze_events_kernel != MMCR0_FCHV))
+	if (regs_sipr(regs)) {
+		if (!(ppmu->flags & PPMU_P10))
+			return PERF_RECORD_MISC_USER;
+	} else if (regs_sihv(regs) && (freeze_events_kernel != MMCR0_FCHV))
 		return PERF_RECORD_MISC_HYPERVISOR;
 
+	/*
+	 * Check the address in SIAR to identify the
+	 * privilege levels since the SIER[MSR_HV, MSR_PR]
+	 * bits are not set correctly in power10 sometimes
+	 */
+	if (ppmu->flags & PPMU_P10) {
+		siar = mfspr(SPRN_SIAR);
+		addr = siar ? siar : regs->nip;
+		if (!is_kernel_addr(addr))
+			return PERF_RECORD_MISC_USER;
+	}
+
 	return PERF_RECORD_MISC_KERNEL;
 }
 

arch/powerpc/perf/power10-pmu.c

@@ -593,7 +593,8 @@ static struct power_pmu power10_pmu = {
 	.get_mem_weight		= isa207_get_mem_weight,
 	.disable_pmc		= isa207_disable_pmc,
 	.flags			= PPMU_HAS_SIER | PPMU_ARCH_207S |
-				  PPMU_ARCH_31 | PPMU_HAS_ATTR_CONFIG1,
+				  PPMU_ARCH_31 | PPMU_HAS_ATTR_CONFIG1 |
+				  PPMU_P10,
 	.n_generic		= ARRAY_SIZE(power10_generic_events),
 	.generic_events		= power10_generic_events,
 	.cache_events		= &power10_cache_events,