Automatic merge of 'next' into merge (2025-02-14 17:57)

Author: maddy-kerneldev

Documentation/ABI/testing/sysfs-kernel-fadump

@@ -55,4 +55,5 @@ Date:		May 2024
 Contact:	[email protected]
 Description:	read/write
 		This is a special sysfs file available to setup additional
-		parameters to be passed to capture kernel.
+		parameters to be passed to capture kernel. For HASH MMU it
+		is exported only if RMA size higher than 768MB.

Documentation/arch/powerpc/firmware-assisted-dump.rst

@@ -120,6 +120,28 @@ to ensure that crash data is preserved to process later.
    e.g.
      # echo 1 > /sys/firmware/opal/mpipl/release_core
 
+-- Support for Additional Kernel Arguments in Fadump
+   Fadump has a feature that allows passing additional kernel arguments
+   to the fadump kernel. This feature was primarily designed to disable
+   kernel functionalities that are not required for the fadump kernel
+   and to reduce its memory footprint while collecting the dump.
+
+  Command to Add Additional Kernel Parameters to Fadump:
+  e.g.
+  # echo "nr_cpus=16" > /sys/kernel/fadump/bootargs_append
+
+  The above command is sufficient to add additional arguments to fadump.
+  An explicit service restart is not required.
+
+  Command to Retrieve the Additional Fadump Arguments:
+  e.g.
+  # cat /sys/kernel/fadump/bootargs_append
+
+Note: Additional kernel arguments for fadump with HASH MMU is only
+      supported if the RMA size is greater than 768 MB. If the RMA
+      size is less than 768 MB, the kernel does not export the
+      /sys/kernel/fadump/bootargs_append sysfs node.
+
 Implementation details:
 -----------------------
 

Documentation/arch/powerpc/papr_hcalls.rst

@@ -289,6 +289,17 @@ to be issued multiple times in order to be completely serviced. The
 subsequent hcalls to the hypervisor until the hcall is completely serviced
 at which point H_SUCCESS or other error is returned by the hypervisor.
 
+**H_HTM**
+
+| Input: flags, target, operation (op), op-param1, op-param2, op-param3
+| Out: *dumphtmbufferdata*
+| Return Value: *H_Success,H_Busy,H_LongBusyOrder,H_Partial,H_Parameter,
+		 H_P2,H_P3,H_P4,H_P5,H_P6,H_State,H_Not_Available,H_Authority*
+
+H_HTM supports setup, configuration, control and dumping of Hardware Trace
+Macro (HTM) function and its data. HTM buffer stores tracing data for functions
+like core instruction, core LLAT and nest.
+
 References
 ==========
 .. [1] "Power Architecture Platform Reference"

arch/powerpc/include/asm/hvcall.h

@@ -348,6 +348,7 @@
 #define H_SCM_FLUSH		0x44C
 #define H_GET_ENERGY_SCALE_INFO	0x450
 #define H_PKS_SIGNED_UPDATE	0x454
+#define H_HTM                   0x458
 #define H_WATCHDOG		0x45C
 #define H_GUEST_GET_CAPABILITIES 0x460
 #define H_GUEST_SET_CAPABILITIES 0x464
@@ -498,6 +499,39 @@
 #define H_GUEST_CAP_POWER11	(1UL<<(63-3))
 #define H_GUEST_CAP_BITMAP2	(1UL<<(63-63))
 
+/*
+ * Defines for H_HTM - Macros for hardware trace macro (HTM) function.
+ */
+#define H_HTM_FLAGS_HARDWARE_TARGET    (1ul << 63)
+#define H_HTM_FLAGS_LOGICAL_TARGET     (1ul << 62)
+#define H_HTM_FLAGS_PROCID_TARGET      (1ul << 61)
+#define H_HTM_FLAGS_NOWRAP             (1ul << 60)
+
+#define H_HTM_OP_SHIFT                 (63-15)
+#define H_HTM_OP(x)                    ((unsigned long)(x)<<H_HTM_OP_SHIFT)
+#define H_HTM_OP_CAPABILITIES          0x01
+#define H_HTM_OP_STATUS                        0x02
+#define H_HTM_OP_SETUP                 0x03
+#define H_HTM_OP_CONFIGURE             0x04
+#define H_HTM_OP_START                 0x05
+#define H_HTM_OP_STOP                  0x06
+#define H_HTM_OP_DECONFIGURE           0x07
+#define H_HTM_OP_DUMP_DETAILS          0x08
+#define H_HTM_OP_DUMP_DATA             0x09
+#define H_HTM_OP_DUMP_SYSMEM_CONF      0x0a
+#define H_HTM_OP_DUMP_SYSPROC_CONF     0x0b
+
+#define H_HTM_TYPE_SHIFT               (63-31)
+#define H_HTM_TYPE(x)                  ((unsigned long)(x)<<H_HTM_TYPE_SHIFT)
+#define H_HTM_TYPE_NEST                        0x01
+#define H_HTM_TYPE_CORE                        0x02
+#define H_HTM_TYPE_LLAT                        0x03
+#define H_HTM_TYPE_GLOBAL              0xff
+
+#define H_HTM_TARGET_NODE_INDEX(x)             ((unsigned long)(x)<<(63-15))
+#define H_HTM_TARGET_NODAL_CHIP_INDEX(x)       ((unsigned long)(x)<<(63-31))
+#define H_HTM_TARGET_CORE_INDEX_ON_CHIP(x)     ((unsigned long)(x)<<(63-47))
+
 #ifndef __ASSEMBLY__
 #include <linux/types.h>
 

arch/powerpc/include/asm/mmzone.h

@@ -29,6 +29,7 @@ extern cpumask_var_t node_to_cpumask_map[];
 #ifdef CONFIG_MEMORY_HOTPLUG
 extern unsigned long max_pfn;
 u64 memory_hotplug_max(void);
+u64 hot_add_drconf_memory_max(void);
 #else
 #define memory_hotplug_max() memblock_end_of_DRAM()
 #endif

arch/powerpc/include/asm/plpar_wrappers.h

@@ -65,6 +65,27 @@ static inline long register_dtl(unsigned long cpu, unsigned long vpa)
 	return vpa_call(H_VPA_REG_DTL, cpu, vpa);
 }
 
+static inline long htm_call(unsigned long flags, unsigned long target,
+               unsigned long operation, unsigned long param1,
+               unsigned long param2, unsigned long param3)
+{
+       return plpar_hcall_norets(H_HTM, flags, target, operation,
+                                 param1, param2, param3);
+}
+
+static inline long htm_get_dump_hardware(unsigned long nodeindex,
+               unsigned long nodalchipindex, unsigned long coreindexonchip,
+               unsigned long type, unsigned long addr, unsigned long size,
+               unsigned long offset)
+{
+       return htm_call(H_HTM_FLAGS_HARDWARE_TARGET,
+                       H_HTM_TARGET_NODE_INDEX(nodeindex) |
+                       H_HTM_TARGET_NODAL_CHIP_INDEX(nodalchipindex) |
+                       H_HTM_TARGET_CORE_INDEX_ON_CHIP(coreindexonchip),
+                       H_HTM_OP(H_HTM_OP_DUMP_DATA) | H_HTM_TYPE(type),
+                       addr, size, offset);
+}
+
 extern void vpa_init(int cpu);
 
 static inline long plpar_pte_enter(unsigned long flags,

arch/powerpc/include/asm/prom.h

@@ -17,6 +17,8 @@
 struct device_node;
 struct property;
 
+#define MIN_RMA			768		/* Minimum RMA (in MB) for CAS negotiation */
+
 #define OF_DT_BEGIN_NODE	0x1		/* Start of node, full name */
 #define OF_DT_END_NODE		0x2		/* End node */
 #define OF_DT_PROP		0x3		/* Property: name off, size,

arch/powerpc/kernel/fadump.c

@@ -33,6 +33,7 @@
 #include <asm/fadump-internal.h>
 #include <asm/setup.h>
 #include <asm/interrupt.h>
+#include <asm/prom.h>
 
 /*
  * The CPU who acquired the lock to trigger the fadump crash should
@@ -1764,19 +1765,19 @@ void __init fadump_setup_param_area(void)
 		range_end = memblock_end_of_DRAM();
 	} else {
 		/*
-		 * Passing additional parameters is supported for hash MMU only
-		 * if the first memory block size is 768MB or higher.
+		 * Memory range for passing additional parameters for HASH MMU
+		 * must meet the following conditions:
+		 * 1. The first memory block size must be higher than the
+		 *    minimum RMA (MIN_RMA) size. Bootloader can use memory
+		 *    upto RMA size. So it should be avoided.
+		 * 2. The range should be between MIN_RMA and RMA size (ppc64_rma_size)
+		 * 3. It must not overlap with the fadump reserved area.
 		 */
-		if (ppc64_rma_size < 0x30000000)
+		if (ppc64_rma_size < MIN_RMA*1024*1024)
 			return;
 
-		/*
-		 * 640 MB to 768 MB is not used by PFW/bootloader. So, try reserving
-		 * memory for passing additional parameters in this range to avoid
-		 * being stomped on by PFW/bootloader.
-		 */
-		range_start = 0x2A000000;
-		range_end = range_start + 0x4000000;
+		range_start = MIN_RMA * 1024 * 1024;
+		range_end = min(ppc64_rma_size, fw_dump.boot_mem_top);
 	}
 
 	fw_dump.param_area = memblock_phys_alloc_range(COMMAND_LINE_SIZE,

arch/powerpc/kernel/prom_init.c

@@ -1061,7 +1061,7 @@ static const struct ibm_arch_vec ibm_architecture_vec_template __initconst = {
 		.virt_base = cpu_to_be32(0xffffffff),
 		.virt_size = cpu_to_be32(0xffffffff),
 		.load_base = cpu_to_be32(0xffffffff),
-		.min_rma = cpu_to_be32(512),		/* 512MB min RMA */
+		.min_rma = cpu_to_be32(MIN_RMA),
 		.min_load = cpu_to_be32(0xffffffff),	/* full client load */
 		.min_rma_percent = 0,	/* min RMA percentage of total RAM */
 		.max_pft_size = 48,	/* max log_2(hash table size) */

arch/powerpc/mm/numa.c

@@ -1336,7 +1336,7 @@ int hot_add_scn_to_nid(unsigned long scn_addr)
 	return nid;
 }
 
-static u64 hot_add_drconf_memory_max(void)
+u64 hot_add_drconf_memory_max(void)
 {
 	struct device_node *memory = NULL;
 	struct device_node *dn = NULL;

arch/powerpc/perf/core-book3s.c

@@ -2222,6 +2222,10 @@ static struct pmu power_pmu = {
 #define PERF_SAMPLE_ADDR_TYPE  (PERF_SAMPLE_ADDR |		\
 				PERF_SAMPLE_PHYS_ADDR |		\
 				PERF_SAMPLE_DATA_PAGE_SIZE)
+
+#define SIER_TYPE_SHIFT	15
+#define SIER_TYPE_MASK	(0x7ull << SIER_TYPE_SHIFT)
+
 /*
  * A counter has overflowed; update its count and record
  * things if requested.  Note that interrupts are hard-disabled
@@ -2290,6 +2294,22 @@ static void record_and_restart(struct perf_event *event, unsigned long val,
 	    is_kernel_addr(mfspr(SPRN_SIAR)))
 		record = 0;
 
+	/*
+	 * SIER[46-48] presents instruction type of the sampled instruction.
+	 * In ISA v3.0 and before values "0" and "7" are considered reserved.
+	 * In ISA v3.1, value "7" has been used to indicate "larx/stcx".
+	 * Drop the sample if "type" has reserved values for this field with a
+	 * ISA version check.
+	 */
+	if (event->attr.sample_type & PERF_SAMPLE_DATA_SRC &&
+			ppmu->get_mem_data_src) {
+		val = (regs->dar & SIER_TYPE_MASK) >> SIER_TYPE_SHIFT;
+		if (val == 0 || (val == 7 && !cpu_has_feature(CPU_FTR_ARCH_31))) {
+			record = 0;
+			atomic64_inc(&event->lost_samples);
+		}
+	}
+
 	/*
 	 * Finally record data if requested.
 	 */

arch/powerpc/perf/isa207-common.c

@@ -319,10 +319,18 @@ void isa207_get_mem_data_src(union perf_mem_data_src *dsrc, u32 flags,
 		return;
 	}
 
-	sier = mfspr(SPRN_SIER);
+	/*
+	 * Use regs-dar for SPRN_SIER which is saved
+	 * during perf_read_regs at the beginning
+	 * of the PMU interrupt handler to avoid multiple
+	 * reads of SPRN_SIER
+	 */
+	sier = regs->dar;
 	val = (sier & ISA207_SIER_TYPE_MASK) >> ISA207_SIER_TYPE_SHIFT;
-	if (val != 1 && val != 2 && !(val == 7 && cpu_has_feature(CPU_FTR_ARCH_31)))
+	if (val != 1 && val != 2 && !(val == 7 && cpu_has_feature(CPU_FTR_ARCH_31))) {
+		dsrc->val = 0;
 		return;
+	}
 
 	idx = (sier & ISA207_SIER_LDST_MASK) >> ISA207_SIER_LDST_SHIFT;
 	sub_idx = (sier & ISA207_SIER_DATA_SRC_MASK) >> ISA207_SIER_DATA_SRC_SHIFT;
@@ -338,8 +346,12 @@ void isa207_get_mem_data_src(union perf_mem_data_src *dsrc, u32 flags,
 		 * to determine the exact instruction type. If the sampling
 		 * criteria is neither load or store, set the type as default
 		 * to NA.
+		 *
+		 * Use regs->dsisr for MMCRA which is saved during perf_read_regs
+		 * at the beginning of the PMU interrupt handler to avoid
+		 * multiple reads of SPRN_MMCRA
 		 */
-		mmcra = mfspr(SPRN_MMCRA);
+		mmcra = regs->dsisr;
 
 		op_type = (mmcra >> MMCRA_SAMP_ELIG_SHIFT) & MMCRA_SAMP_ELIG_MASK;
 		switch (op_type) {

arch/powerpc/platforms/pseries/Kconfig

@@ -128,6 +128,15 @@ config CMM
 	  will be reused for other LPARs. The interface allows firmware to
 	  balance memory across many LPARs.
 
+config HTMDUMP
+	tristate "PowerVM data dumper"
+	depends on PPC_PSERIES && DEBUG_FS
+	default m
+	help
+	  Select this option, if you want to enable the kernel debugfs
+	  interface to dump the Hardware Trace Macro (HTM) function data
+	  in the LPAR.
+
 config HV_PERF_CTRS
 	bool "Hypervisor supplied PMU events (24x7 & GPCI)"
 	default y

arch/powerpc/platforms/pseries/Makefile

@@ -19,6 +19,7 @@ obj-$(CONFIG_HVC_CONSOLE)	+= hvconsole.o
 obj-$(CONFIG_HVCS)		+= hvcserver.o
 obj-$(CONFIG_HCALL_STATS)	+= hvCall_inst.o
 obj-$(CONFIG_CMM)		+= cmm.o
+obj-$(CONFIG_HTMDUMP)		+= htmdump.o
 obj-$(CONFIG_IO_EVENT_IRQ)	+= io_event_irq.o
 obj-$(CONFIG_LPARCFG)		+= lparcfg.o
 obj-$(CONFIG_IBMVIO)		+= vio.o