diff options
Diffstat (limited to 'tools/power')
48 files changed, 3662 insertions, 1377 deletions
diff --git a/tools/power/acpi/.gitignore b/tools/power/acpi/.gitignore index 0b319fc8bb17..eada0297ef88 100644 --- a/tools/power/acpi/.gitignore +++ b/tools/power/acpi/.gitignore @@ -2,4 +2,5 @@ /acpidbg /acpidump /ec +/pfrut /include/ diff --git a/tools/power/acpi/Makefile b/tools/power/acpi/Makefile index a249c50ebf55..5ff1d9c864d0 100644 --- a/tools/power/acpi/Makefile +++ b/tools/power/acpi/Makefile @@ -9,18 +9,18 @@ include ../../scripts/Makefile.include .NOTPARALLEL: -all: acpidbg acpidump ec -clean: acpidbg_clean acpidump_clean ec_clean -install: acpidbg_install acpidump_install ec_install -uninstall: acpidbg_uninstall acpidump_uninstall ec_uninstall +all: acpidbg acpidump ec pfrut +clean: acpidbg_clean acpidump_clean ec_clean pfrut_clean +install: acpidbg_install acpidump_install ec_install pfrut_install +uninstall: acpidbg_uninstall acpidump_uninstall ec_uninstall pfrut_uninstall -acpidbg acpidump ec: FORCE +acpidbg acpidump ec pfrut: FORCE $(call descend,tools/$@,all) -acpidbg_clean acpidump_clean ec_clean: +acpidbg_clean acpidump_clean ec_clean pfrut_clean: $(call descend,tools/$(@:_clean=),clean) -acpidbg_install acpidump_install ec_install: +acpidbg_install acpidump_install ec_install pfrut_install: $(call descend,tools/$(@:_install=),install) -acpidbg_uninstall acpidump_uninstall ec_uninstall: +acpidbg_uninstall acpidump_uninstall ec_uninstall pfrut_uninstall: $(call descend,tools/$(@:_uninstall=),uninstall) .PHONY: FORCE diff --git a/tools/power/acpi/Makefile.rules b/tools/power/acpi/Makefile.rules index 1d7616f5d0ae..b71aada77688 100644 --- a/tools/power/acpi/Makefile.rules +++ b/tools/power/acpi/Makefile.rules @@ -9,7 +9,7 @@ objdir := $(OUTPUT)tools/$(TOOL)/ toolobjs := $(addprefix $(objdir),$(TOOL_OBJS)) $(OUTPUT)$(TOOL): $(toolobjs) FORCE $(ECHO) " LD " $(subst $(OUTPUT),,$@) - $(QUIET) $(LD) $(CFLAGS) $(LDFLAGS) $(toolobjs) -L$(OUTPUT) -o $@ + $(QUIET) $(LD) $(CFLAGS) $(toolobjs) $(LDFLAGS) -L$(OUTPUT) -o $@ $(ECHO) " STRIP " $(subst $(OUTPUT),,$@) $(QUIET) $(STRIPCMD) $@ diff --git a/tools/power/acpi/common/cmfsize.c b/tools/power/acpi/common/cmfsize.c index 185b8c588e1d..38f9b9da8170 100644 --- a/tools/power/acpi/common/cmfsize.c +++ b/tools/power/acpi/common/cmfsize.c @@ -3,7 +3,7 @@ * * Module Name: cmfsize - Common get file size function * - * Copyright (C) 2000 - 2021, Intel Corp. + * Copyright (C) 2000 - 2022, Intel Corp. * *****************************************************************************/ diff --git a/tools/power/acpi/common/getopt.c b/tools/power/acpi/common/getopt.c index 3c265bc917a1..96fd6cec78e2 100644 --- a/tools/power/acpi/common/getopt.c +++ b/tools/power/acpi/common/getopt.c @@ -3,7 +3,7 @@ * * Module Name: getopt * - * Copyright (C) 2000 - 2021, Intel Corp. + * Copyright (C) 2000 - 2022, Intel Corp. * *****************************************************************************/ diff --git a/tools/power/acpi/man/pfrut.8 b/tools/power/acpi/man/pfrut.8 new file mode 100644 index 000000000000..3db574770e8d --- /dev/null +++ b/tools/power/acpi/man/pfrut.8 @@ -0,0 +1,137 @@ +.TH "PFRUT" "8" "October 2021" "pfrut 1.0" "" +.hy +.SH Name +.PP +pfrut \- Platform Firmware Runtime Update and Telemetry tool +.SH SYNOPSIS +.PP +\f[B]pfrut\f[R] [\f[I]Options\f[R]] +.SH DESCRIPTION +.PP +The PFRUT(Platform Firmware Runtime Update and Telemetry) kernel interface is designed +to +.PD 0 +.P +.PD +interact with the platform firmware interface defined in the +.PD 0 +.P +.PD +Management Mode Firmware Runtime +Update (https://uefi.org/sites/default/files/resources/Intel_MM_OS_Interface_Spec_Rev100.pdf) +.PD 0 +.P +.PD +\f[B]pfrut\f[R] is the tool to interact with the kernel interface. +.PD 0 +.P +.PD +.SH OPTIONS +.TP +.B \f[B]\-h\f[R], \f[B]\-\-help\f[R] +Display helper information. +.TP +.B \f[B]\-l\f[R], \f[B]\-\-load\f[R] +Load the capsule file into the system. +To be more specific, the capsule file will be copied to the +communication buffer. +.TP +.B \f[B]\-s\f[R], \f[B]\-\-stage\f[R] +Stage the capsule image from communication buffer into Management Mode +and perform authentication. +.TP +.B \f[B]\-a\f[R], \f[B]\-\-activate\f[R] +Activate a previous staged capsule image. +.TP +.B \f[B]\-u\f[R], \f[B]\-\-update\f[R] +Perform both stage and activation actions. +.TP +.B \f[B]\-q\f[R], \f[B]\-\-query\f[R] +Query the update capability. +.TP +.B \f[B]\-d\f[R], \f[B]\-\-setrev\f[R] +Set the revision ID of code injection/driver update. +.TP +.B \f[B]\-D\f[R], \f[B]\-\-setrevlog\f[R] +Set the revision ID of telemetry. +.TP +.B \f[B]\-G\f[R], \f[B]\-\-getloginfo\f[R] +Get telemetry log information and print it out. +.TP +.B \f[B]\-T\f[R], \f[B]\-\-type\f[R] +Set the telemetry log data type. +.TP +.B \f[B]\-L\f[R], \f[B]\-\-level\f[R] +Set the telemetry log level. +.TP +.B \f[B]\-R\f[R], \f[B]\-\-read\f[R] +Read all the telemetry data and print it out. +.SH EXAMPLES +.PP +\f[B]pfrut \-G\f[R] +.PP +log_level:4 +.PD 0 +.P +.PD +log_type:0 +.PD 0 +.P +.PD +log_revid:2 +.PD 0 +.P +.PD +max_data_size:65536 +.PD 0 +.P +.PD +chunk1_size:0 +.PD 0 +.P +.PD +chunk2_size:1401 +.PD 0 +.P +.PD +rollover_cnt:0 +.PD 0 +.P +.PD +reset_cnt:4 +.PP +\f[B]pfru \-q\f[R] +.PP +code injection image type:794bf8b2\-6e7b\-454e\-885f\-3fb9bb185402 +.PD 0 +.P +.PD +fw_version:0 +.PD 0 +.P +.PD +code_rt_version:1 +.PD 0 +.P +.PD +driver update image type:0e5f0b14\-f849\-7945\-ad81\-bc7b6d2bb245 +.PD 0 +.P +.PD +drv_rt_version:0 +.PD 0 +.P +.PD +drv_svn:0 +.PD 0 +.P +.PD +platform id:39214663\-b1a8\-4eaa\-9024\-f2bb53ea4723 +.PD 0 +.P +.PD +oem id:a36db54f\-ea2a\-e14e\-b7c4\-b5780e51ba3d +.PP +\f[B]pfrut \-l yours.cap \-u \-T 1 \-L 4\f[R] +.SH AUTHORS +Chen Yu. diff --git a/tools/power/acpi/os_specific/service_layers/oslinuxtbl.c b/tools/power/acpi/os_specific/service_layers/oslinuxtbl.c index ccabdbaae6a4..bd08f36df4a7 100644 --- a/tools/power/acpi/os_specific/service_layers/oslinuxtbl.c +++ b/tools/power/acpi/os_specific/service_layers/oslinuxtbl.c @@ -3,7 +3,7 @@ * * Module Name: oslinuxtbl - Linux OSL for obtaining ACPI tables * - * Copyright (C) 2000 - 2021, Intel Corp. + * Copyright (C) 2000 - 2022, Intel Corp. * *****************************************************************************/ diff --git a/tools/power/acpi/os_specific/service_layers/osunixdir.c b/tools/power/acpi/os_specific/service_layers/osunixdir.c index edd99274cd12..5107892d054b 100644 --- a/tools/power/acpi/os_specific/service_layers/osunixdir.c +++ b/tools/power/acpi/os_specific/service_layers/osunixdir.c @@ -3,7 +3,7 @@ * * Module Name: osunixdir - Unix directory access interfaces * - * Copyright (C) 2000 - 2021, Intel Corp. + * Copyright (C) 2000 - 2022, Intel Corp. * *****************************************************************************/ diff --git a/tools/power/acpi/os_specific/service_layers/osunixmap.c b/tools/power/acpi/os_specific/service_layers/osunixmap.c index fee0022560d5..6ff4edd8dc3b 100644 --- a/tools/power/acpi/os_specific/service_layers/osunixmap.c +++ b/tools/power/acpi/os_specific/service_layers/osunixmap.c @@ -3,7 +3,7 @@ * * Module Name: osunixmap - Unix OSL for file mappings * - * Copyright (C) 2000 - 2021, Intel Corp. + * Copyright (C) 2000 - 2022, Intel Corp. * *****************************************************************************/ diff --git a/tools/power/acpi/os_specific/service_layers/osunixxf.c b/tools/power/acpi/os_specific/service_layers/osunixxf.c index 0861728da562..b3651a04d68c 100644 --- a/tools/power/acpi/os_specific/service_layers/osunixxf.c +++ b/tools/power/acpi/os_specific/service_layers/osunixxf.c @@ -3,7 +3,7 @@ * * Module Name: osunixxf - UNIX OSL interfaces * - * Copyright (C) 2000 - 2021, Intel Corp. + * Copyright (C) 2000 - 2022, Intel Corp. * *****************************************************************************/ diff --git a/tools/power/acpi/tools/acpidump/acpidump.h b/tools/power/acpi/tools/acpidump/acpidump.h index e0ebc1dab1cc..153249c87fd7 100644 --- a/tools/power/acpi/tools/acpidump/acpidump.h +++ b/tools/power/acpi/tools/acpidump/acpidump.h @@ -3,7 +3,7 @@ * * Module Name: acpidump.h - Include file for acpi_dump utility * - * Copyright (C) 2000 - 2021, Intel Corp. + * Copyright (C) 2000 - 2022, Intel Corp. * *****************************************************************************/ diff --git a/tools/power/acpi/tools/acpidump/apdump.c b/tools/power/acpi/tools/acpidump/apdump.c index 444e3d78bd89..d54dde02b87d 100644 --- a/tools/power/acpi/tools/acpidump/apdump.c +++ b/tools/power/acpi/tools/acpidump/apdump.c @@ -3,7 +3,7 @@ * * Module Name: apdump - Dump routines for ACPI tables (acpidump) * - * Copyright (C) 2000 - 2021, Intel Corp. + * Copyright (C) 2000 - 2022, Intel Corp. * *****************************************************************************/ diff --git a/tools/power/acpi/tools/acpidump/apfiles.c b/tools/power/acpi/tools/acpidump/apfiles.c index da0c6e13042b..2d9b45a9b526 100644 --- a/tools/power/acpi/tools/acpidump/apfiles.c +++ b/tools/power/acpi/tools/acpidump/apfiles.c @@ -3,7 +3,7 @@ * * Module Name: apfiles - File-related functions for acpidump utility * - * Copyright (C) 2000 - 2021, Intel Corp. + * Copyright (C) 2000 - 2022, Intel Corp. * *****************************************************************************/ diff --git a/tools/power/acpi/tools/acpidump/apmain.c b/tools/power/acpi/tools/acpidump/apmain.c index a4cf6042fcfd..44b23fc53dd9 100644 --- a/tools/power/acpi/tools/acpidump/apmain.c +++ b/tools/power/acpi/tools/acpidump/apmain.c @@ -3,7 +3,7 @@ * * Module Name: apmain - Main module for the acpidump utility * - * Copyright (C) 2000 - 2021, Intel Corp. + * Copyright (C) 2000 - 2022, Intel Corp. * *****************************************************************************/ diff --git a/tools/power/acpi/tools/pfrut/Makefile b/tools/power/acpi/tools/pfrut/Makefile new file mode 100644 index 000000000000..61c1a96fd433 --- /dev/null +++ b/tools/power/acpi/tools/pfrut/Makefile @@ -0,0 +1,23 @@ +# SPDX-License-Identifier: GPL-2.0+ + +include ../../Makefile.config + +TOOL = pfrut +EXTRA_INSTALL = install-man +EXTRA_UNINSTALL = uninstall-man + +CFLAGS += -Wall -O2 +CFLAGS += -DPFRUT_HEADER='"../../../../../include/uapi/linux/pfrut.h"' +LDFLAGS += -luuid + +TOOL_OBJS = \ + pfrut.o + +include ../../Makefile.rules + +install-man: $(srctree)/man/pfrut.8 + $(ECHO) " INST " pfrut.8 + $(QUIET) $(INSTALL_DATA) -D $< $(DESTDIR)$(mandir)/man8/pfrut.8 +uninstall-man: + $(ECHO) " UNINST " pfrut.8 + $(QUIET) rm -f $(DESTDIR)$(mandir)/man8/pfrut.8 diff --git a/tools/power/acpi/tools/pfrut/pfrut.c b/tools/power/acpi/tools/pfrut/pfrut.c new file mode 100644 index 000000000000..52aa0351533c --- /dev/null +++ b/tools/power/acpi/tools/pfrut/pfrut.c @@ -0,0 +1,424 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Platform Firmware Runtime Update tool to do Management + * Mode code injection/driver update and telemetry retrieval. + * + * This tool uses the interfaces provided by pfr_update and + * pfr_telemetry drivers. These interfaces are exposed via + * /dev/pfr_update and /dev/pfr_telemetry. Write operation + * on the /dev/pfr_update is to load the EFI capsule into + * kernel space. Mmap/read operations on /dev/pfr_telemetry + * could be used to read the telemetry data to user space. + */ +#define _GNU_SOURCE +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +#include <sys/types.h> +#include <sys/stat.h> +#include <fcntl.h> +#include <unistd.h> +#include <getopt.h> +#include <sys/ioctl.h> +#include <sys/mman.h> +#include <uuid/uuid.h> +#include PFRUT_HEADER + +char *capsule_name; +int action, query_cap, log_type, log_level, log_read, log_getinfo, + revid, log_revid; +int set_log_level, set_log_type, + set_revid, set_log_revid; + +char *progname; + +#define LOG_ERR 0 +#define LOG_WARN 1 +#define LOG_INFO 2 +#define LOG_VERB 4 +#define LOG_EXEC_IDX 0 +#define LOG_HISTORY_IDX 1 +#define REVID_1 1 +#define REVID_2 2 + +static int valid_log_level(int level) +{ + return level == LOG_ERR || level == LOG_WARN || + level == LOG_INFO || level == LOG_VERB; +} + +static int valid_log_type(int type) +{ + return type == LOG_EXEC_IDX || type == LOG_HISTORY_IDX; +} + +static inline int valid_log_revid(int id) +{ + return id == REVID_1 || id == REVID_2; +} + +static void help(void) +{ + fprintf(stderr, + "usage: %s [OPTIONS]\n" + " code injection:\n" + " -l, --load\n" + " -s, --stage\n" + " -a, --activate\n" + " -u, --update [stage and activate]\n" + " -q, --query\n" + " -d, --revid update\n" + " telemetry:\n" + " -G, --getloginfo\n" + " -T, --type(0:execution, 1:history)\n" + " -L, --level(0, 1, 2, 4)\n" + " -R, --read\n" + " -D, --revid log\n", + progname); +} + +char *option_string = "l:sauqd:GT:L:RD:h"; +static struct option long_options[] = { + {"load", required_argument, 0, 'l'}, + {"stage", no_argument, 0, 's'}, + {"activate", no_argument, 0, 'a'}, + {"update", no_argument, 0, 'u'}, + {"query", no_argument, 0, 'q'}, + {"getloginfo", no_argument, 0, 'G'}, + {"type", required_argument, 0, 'T'}, + {"level", required_argument, 0, 'L'}, + {"read", no_argument, 0, 'R'}, + {"setrev", required_argument, 0, 'd'}, + {"setrevlog", required_argument, 0, 'D'}, + {"help", no_argument, 0, 'h'}, + {} +}; + +static void parse_options(int argc, char **argv) +{ + int option_index = 0; + char *pathname; + int opt; + + pathname = strdup(argv[0]); + progname = basename(pathname); + + while ((opt = getopt_long_only(argc, argv, option_string, + long_options, &option_index)) != -1) { + switch (opt) { + case 'l': + capsule_name = optarg; + break; + case 's': + action = 1; + break; + case 'a': + action = 2; + break; + case 'u': + action = 3; + break; + case 'q': + query_cap = 1; + break; + case 'G': + log_getinfo = 1; + break; + case 'T': + log_type = atoi(optarg); + set_log_type = 1; + break; + case 'L': + log_level = atoi(optarg); + set_log_level = 1; + break; + case 'R': + log_read = 1; + break; + case 'd': + revid = atoi(optarg); + set_revid = 1; + break; + case 'D': + log_revid = atoi(optarg); + set_log_revid = 1; + break; + case 'h': + help(); + exit(0); + default: + break; + } + } +} + +void print_cap(struct pfru_update_cap_info *cap) +{ + char *uuid; + + uuid = malloc(37); + if (!uuid) { + perror("Can not allocate uuid buffer\n"); + exit(1); + } + + uuid_unparse(cap->code_type, uuid); + printf("code injection image type:%s\n", uuid); + printf("fw_version:%d\n", cap->fw_version); + printf("code_rt_version:%d\n", cap->code_rt_version); + + uuid_unparse(cap->drv_type, uuid); + printf("driver update image type:%s\n", uuid); + printf("drv_rt_version:%d\n", cap->drv_rt_version); + printf("drv_svn:%d\n", cap->drv_svn); + + uuid_unparse(cap->platform_id, uuid); + printf("platform id:%s\n", uuid); + uuid_unparse(cap->oem_id, uuid); + printf("oem id:%s\n", uuid); + printf("oem information length:%d\n", cap->oem_info_len); + + free(uuid); +} + +int main(int argc, char *argv[]) +{ + int fd_update, fd_update_log, fd_capsule; + struct pfrt_log_data_info data_info; + struct pfrt_log_info info; + struct pfru_update_cap_info cap; + void *addr_map_capsule; + struct stat st; + char *log_buf; + int ret; + + if (getuid() != 0) { + printf("Please run the tool as root - Exiting.\n"); + return 1; + } + + parse_options(argc, argv); + + fd_update = open("/dev/acpi_pfr_update0", O_RDWR); + if (fd_update < 0) { + printf("PFRU device not supported - Quit...\n"); + return 1; + } + + fd_update_log = open("/dev/acpi_pfr_telemetry0", O_RDWR); + if (fd_update_log < 0) { + printf("PFRT device not supported - Quit...\n"); + return 1; + } + + if (query_cap) { + ret = ioctl(fd_update, PFRU_IOC_QUERY_CAP, &cap); + if (ret) + perror("Query Update Capability info failed."); + else + print_cap(&cap); + + close(fd_update); + close(fd_update_log); + + return ret; + } + + if (log_getinfo) { + ret = ioctl(fd_update_log, PFRT_LOG_IOC_GET_DATA_INFO, &data_info); + if (ret) { + perror("Get telemetry data info failed."); + close(fd_update); + close(fd_update_log); + + return 1; + } + + ret = ioctl(fd_update_log, PFRT_LOG_IOC_GET_INFO, &info); + if (ret) { + perror("Get telemetry info failed."); + close(fd_update); + close(fd_update_log); + + return 1; + } + + printf("log_level:%d\n", info.log_level); + printf("log_type:%d\n", info.log_type); + printf("log_revid:%d\n", info.log_revid); + printf("max_data_size:%d\n", data_info.max_data_size); + printf("chunk1_size:%d\n", data_info.chunk1_size); + printf("chunk2_size:%d\n", data_info.chunk2_size); + printf("rollover_cnt:%d\n", data_info.rollover_cnt); + printf("reset_cnt:%d\n", data_info.reset_cnt); + + return 0; + } + + info.log_level = -1; + info.log_type = -1; + info.log_revid = -1; + + if (set_log_level) { + if (!valid_log_level(log_level)) { + printf("Invalid log level %d\n", + log_level); + } else { + info.log_level = log_level; + } + } + + if (set_log_type) { + if (!valid_log_type(log_type)) { + printf("Invalid log type %d\n", + log_type); + } else { + info.log_type = log_type; + } + } + + if (set_log_revid) { + if (!valid_log_revid(log_revid)) { + printf("Invalid log revid %d, unchanged.\n", + log_revid); + } else { + info.log_revid = log_revid; + } + } + + ret = ioctl(fd_update_log, PFRT_LOG_IOC_SET_INFO, &info); + if (ret) { + perror("Log information set failed.(log_level, log_type, log_revid)"); + close(fd_update); + close(fd_update_log); + + return 1; + } + + if (set_revid) { + ret = ioctl(fd_update, PFRU_IOC_SET_REV, &revid); + if (ret) { + perror("pfru update revid set failed"); + close(fd_update); + close(fd_update_log); + + return 1; + } + + printf("pfru update revid set to %d\n", revid); + } + + if (capsule_name) { + fd_capsule = open(capsule_name, O_RDONLY); + if (fd_capsule < 0) { + perror("Can not open capsule file..."); + close(fd_update); + close(fd_update_log); + + return 1; + } + + if (fstat(fd_capsule, &st) < 0) { + perror("Can not fstat capsule file..."); + close(fd_capsule); + close(fd_update); + close(fd_update_log); + + return 1; + } + + addr_map_capsule = mmap(NULL, st.st_size, PROT_READ, MAP_SHARED, + fd_capsule, 0); + if (addr_map_capsule == MAP_FAILED) { + perror("Failed to mmap capsule file."); + close(fd_capsule); + close(fd_update); + close(fd_update_log); + + return 1; + } + + ret = write(fd_update, (char *)addr_map_capsule, st.st_size); + printf("Load %d bytes of capsule file into the system\n", + ret); + + if (ret == -1) { + perror("Failed to load capsule file"); + close(fd_capsule); + close(fd_update); + close(fd_update_log); + + return 1; + } + + munmap(addr_map_capsule, st.st_size); + close(fd_capsule); + printf("Load done.\n"); + } + + if (action) { + if (action == 1) { + ret = ioctl(fd_update, PFRU_IOC_STAGE, NULL); + } else if (action == 2) { + ret = ioctl(fd_update, PFRU_IOC_ACTIVATE, NULL); + } else if (action == 3) { + ret = ioctl(fd_update, PFRU_IOC_STAGE_ACTIVATE, NULL); + } else { + close(fd_update); + close(fd_update_log); + + return 1; + } + printf("Update finished, return %d\n", ret); + } + + close(fd_update); + + if (log_read) { + void *p_mmap; + int max_data_sz; + + ret = ioctl(fd_update_log, PFRT_LOG_IOC_GET_DATA_INFO, &data_info); + if (ret) { + perror("Get telemetry data info failed."); + close(fd_update_log); + + return 1; + } + + max_data_sz = data_info.max_data_size; + if (!max_data_sz) { + printf("No telemetry data available.\n"); + close(fd_update_log); + + return 1; + } + + log_buf = malloc(max_data_sz + 1); + if (!log_buf) { + perror("log_buf allocate failed."); + close(fd_update_log); + + return 1; + } + + p_mmap = mmap(NULL, max_data_sz, PROT_READ, MAP_SHARED, fd_update_log, 0); + if (p_mmap == MAP_FAILED) { + perror("mmap error."); + close(fd_update_log); + + return 1; + } + + memcpy(log_buf, p_mmap, max_data_sz); + log_buf[max_data_sz] = '\0'; + printf("%s\n", log_buf); + free(log_buf); + + munmap(p_mmap, max_data_sz); + } + + close(fd_update_log); + + return 0; +} diff --git a/tools/power/cpupower/Makefile b/tools/power/cpupower/Makefile index 3b1594447f29..e9b6de314654 100644 --- a/tools/power/cpupower/Makefile +++ b/tools/power/cpupower/Makefile @@ -143,9 +143,9 @@ UTIL_HEADERS = utils/helpers/helpers.h utils/idle_monitor/cpupower-monitor.h \ utils/helpers/bitmask.h \ utils/idle_monitor/idle_monitors.h utils/idle_monitor/idle_monitors.def -LIB_HEADERS = lib/cpufreq.h lib/cpupower.h lib/cpuidle.h -LIB_SRC = lib/cpufreq.c lib/cpupower.c lib/cpuidle.c -LIB_OBJS = lib/cpufreq.o lib/cpupower.o lib/cpuidle.o +LIB_HEADERS = lib/cpufreq.h lib/cpupower.h lib/cpuidle.h lib/acpi_cppc.h +LIB_SRC = lib/cpufreq.c lib/cpupower.c lib/cpuidle.c lib/acpi_cppc.c +LIB_OBJS = lib/cpufreq.o lib/cpupower.o lib/cpuidle.o lib/acpi_cppc.o LIB_OBJS := $(addprefix $(OUTPUT),$(LIB_OBJS)) override CFLAGS += -pipe diff --git a/tools/power/cpupower/ToDo b/tools/power/cpupower/TODO index b196a139a3e4..b196a139a3e4 100644 --- a/tools/power/cpupower/ToDo +++ b/tools/power/cpupower/TODO diff --git a/tools/power/cpupower/debug/i386/dump_psb.c b/tools/power/cpupower/debug/i386/dump_psb.c index 2c768cf70128..6fb81b42ea61 100644 --- a/tools/power/cpupower/debug/i386/dump_psb.c +++ b/tools/power/cpupower/debug/i386/dump_psb.c @@ -1,7 +1,5 @@ -/* - * dump_psb. (c) 2004, Dave Jones, Red Hat Inc. - * Licensed under the GPL v2. - */ +// SPDX-License-Identifier: GPL-2.0-only +// dump_psb. (c) 2004, Dave Jones, Red Hat Inc. #include <fcntl.h> #include <stdio.h> diff --git a/tools/power/cpupower/lib/acpi_cppc.c b/tools/power/cpupower/lib/acpi_cppc.c new file mode 100644 index 000000000000..c401ac331e9f --- /dev/null +++ b/tools/power/cpupower/lib/acpi_cppc.c @@ -0,0 +1,59 @@ +// SPDX-License-Identifier: GPL-2.0-only + +#include <stdio.h> +#include <errno.h> +#include <stdlib.h> +#include <string.h> +#include <sys/types.h> +#include <sys/stat.h> +#include <fcntl.h> +#include <unistd.h> + +#include "cpupower_intern.h" +#include "acpi_cppc.h" + +/* ACPI CPPC sysfs access ***********************************************/ + +static int acpi_cppc_read_file(unsigned int cpu, const char *fname, + char *buf, size_t buflen) +{ + char path[SYSFS_PATH_MAX]; + + snprintf(path, sizeof(path), PATH_TO_CPU "cpu%u/acpi_cppc/%s", + cpu, fname); + return cpupower_read_sysfs(path, buf, buflen); +} + +static const char * const acpi_cppc_value_files[] = { + [HIGHEST_PERF] = "highest_perf", + [LOWEST_PERF] = "lowest_perf", + [NOMINAL_PERF] = "nominal_perf", + [LOWEST_NONLINEAR_PERF] = "lowest_nonlinear_perf", + [LOWEST_FREQ] = "lowest_freq", + [NOMINAL_FREQ] = "nominal_freq", + [REFERENCE_PERF] = "reference_perf", + [WRAPAROUND_TIME] = "wraparound_time" +}; + +unsigned long acpi_cppc_get_data(unsigned int cpu, enum acpi_cppc_value which) +{ + unsigned long long value; + unsigned int len; + char linebuf[MAX_LINE_LEN]; + char *endp; + + if (which >= MAX_CPPC_VALUE_FILES) + return 0; + + len = acpi_cppc_read_file(cpu, acpi_cppc_value_files[which], + linebuf, sizeof(linebuf)); + if (len == 0) + return 0; + + value = strtoull(linebuf, &endp, 0); + + if (endp == linebuf || errno == ERANGE) + return 0; + + return value; +} diff --git a/tools/power/cpupower/lib/acpi_cppc.h b/tools/power/cpupower/lib/acpi_cppc.h new file mode 100644 index 000000000000..85ca83080316 --- /dev/null +++ b/tools/power/cpupower/lib/acpi_cppc.h @@ -0,0 +1,21 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ + +#ifndef __ACPI_CPPC_H__ +#define __ACPI_CPPC_H__ + +enum acpi_cppc_value { + HIGHEST_PERF, + LOWEST_PERF, + NOMINAL_PERF, + LOWEST_NONLINEAR_PERF, + LOWEST_FREQ, + NOMINAL_FREQ, + REFERENCE_PERF, + WRAPAROUND_TIME, + MAX_CPPC_VALUE_FILES +}; + +unsigned long acpi_cppc_get_data(unsigned int cpu, + enum acpi_cppc_value which); + +#endif /* _ACPI_CPPC_H */ diff --git a/tools/power/cpupower/lib/cpufreq.c b/tools/power/cpupower/lib/cpufreq.c index c3b56db8b921..1516d23c17c9 100644 --- a/tools/power/cpupower/lib/cpufreq.c +++ b/tools/power/cpupower/lib/cpufreq.c @@ -83,20 +83,21 @@ static const char *cpufreq_value_files[MAX_CPUFREQ_VALUE_READ_FILES] = { [STATS_NUM_TRANSITIONS] = "stats/total_trans" }; - -static unsigned long sysfs_cpufreq_get_one_value(unsigned int cpu, - enum cpufreq_value which) +unsigned long cpufreq_get_sysfs_value_from_table(unsigned int cpu, + const char **table, + unsigned int index, + unsigned int size) { unsigned long value; unsigned int len; char linebuf[MAX_LINE_LEN]; char *endp; - if (which >= MAX_CPUFREQ_VALUE_READ_FILES) + if (!table || index >= size || !table[index]) return 0; - len = sysfs_cpufreq_read_file(cpu, cpufreq_value_files[which], - linebuf, sizeof(linebuf)); + len = sysfs_cpufreq_read_file(cpu, table[index], linebuf, + sizeof(linebuf)); if (len == 0) return 0; @@ -109,6 +110,14 @@ static unsigned long sysfs_cpufreq_get_one_value(unsigned int cpu, return value; } +static unsigned long sysfs_cpufreq_get_one_value(unsigned int cpu, + enum cpufreq_value which) +{ + return cpufreq_get_sysfs_value_from_table(cpu, cpufreq_value_files, + which, + MAX_CPUFREQ_VALUE_READ_FILES); +} + /* read access to files which contain one string */ enum cpufreq_string { @@ -124,7 +133,7 @@ static const char *cpufreq_string_files[MAX_CPUFREQ_STRING_FILES] = { static char *sysfs_cpufreq_get_one_string(unsigned int cpu, - enum cpufreq_string which) + enum cpufreq_string which) { char linebuf[MAX_LINE_LEN]; char *result; diff --git a/tools/power/cpupower/lib/cpufreq.h b/tools/power/cpupower/lib/cpufreq.h index 95f4fd9e2656..2f3c84035806 100644 --- a/tools/power/cpupower/lib/cpufreq.h +++ b/tools/power/cpupower/lib/cpufreq.h @@ -203,6 +203,18 @@ int cpufreq_modify_policy_governor(unsigned int cpu, char *governor); int cpufreq_set_frequency(unsigned int cpu, unsigned long target_frequency); +/* + * get the sysfs value from specific table + * + * Read the value with the sysfs file name from specific table. Does + * only work if the cpufreq driver has the specific sysfs interfaces. + */ + +unsigned long cpufreq_get_sysfs_value_from_table(unsigned int cpu, + const char **table, + unsigned int index, + unsigned int size); + #ifdef __cplusplus } #endif diff --git a/tools/power/cpupower/man/cpupower-frequency-info.1 b/tools/power/cpupower/man/cpupower-frequency-info.1 index 6aa8d239dff9..dd545b499480 100644 --- a/tools/power/cpupower/man/cpupower-frequency-info.1 +++ b/tools/power/cpupower/man/cpupower-frequency-info.1 @@ -53,6 +53,9 @@ human\-readable output for the \-f, \-w, \-s and \-y parameters. \fB\-n\fR \fB\-\-no-rounding\fR Output frequencies and latencies without rounding off values. .TP +\fB\-c\fR \fB\-\-perf\fR +Get performances and frequencies capabilities of CPPC, by reading it from hardware (only available on the hardware with CPPC). +.TP .SH "REMARKS" .LP By default only values of core zero are displayed. How to display settings of diff --git a/tools/power/cpupower/man/cpupower-idle-set.1 b/tools/power/cpupower/man/cpupower-idle-set.1 index 21916cff7516..8cef3c71e19e 100644 --- a/tools/power/cpupower/man/cpupower-idle-set.1 +++ b/tools/power/cpupower/man/cpupower-idle-set.1 @@ -4,7 +4,7 @@ cpupower\-idle\-set \- Utility to set cpu idle state specific kernel options .SH "SYNTAX" .LP -cpupower [ \-c cpulist ] idle\-info [\fIoptions\fP] +cpupower [ \-c cpulist ] idle\-set [\fIoptions\fP] .SH "DESCRIPTION" .LP The cpupower idle\-set subcommand allows to set cpu idle, also called cpu diff --git a/tools/power/cpupower/utils/cpufreq-info.c b/tools/power/cpupower/utils/cpufreq-info.c index f9895e31ff5a..0646f615fe2d 100644 --- a/tools/power/cpupower/utils/cpufreq-info.c +++ b/tools/power/cpupower/utils/cpufreq-info.c @@ -84,43 +84,6 @@ static void proc_cpufreq_output(void) } static int no_rounding; -static void print_speed(unsigned long speed) -{ - unsigned long tmp; - - if (no_rounding) { - if (speed > 1000000) - printf("%u.%06u GHz", ((unsigned int) speed/1000000), - ((unsigned int) speed%1000000)); - else if (speed > 1000) - printf("%u.%03u MHz", ((unsigned int) speed/1000), - (unsigned int) (speed%1000)); - else - printf("%lu kHz", speed); - } else { - if (speed > 1000000) { - tmp = speed%10000; - if (tmp >= 5000) - speed += 10000; - printf("%u.%02u GHz", ((unsigned int) speed/1000000), - ((unsigned int) (speed%1000000)/10000)); - } else if (speed > 100000) { - tmp = speed%1000; - if (tmp >= 500) - speed += 1000; - printf("%u MHz", ((unsigned int) speed/1000)); - } else if (speed > 1000) { - tmp = speed%100; - if (tmp >= 50) - speed += 100; - printf("%u.%01u MHz", ((unsigned int) speed/1000), - ((unsigned int) (speed%1000)/100)); - } - } - - return; -} - static void print_duration(unsigned long duration) { unsigned long tmp; @@ -183,9 +146,12 @@ static int get_boost_mode_x86(unsigned int cpu) printf(_(" Supported: %s\n"), support ? _("yes") : _("no")); printf(_(" Active: %s\n"), active ? _("yes") : _("no")); - if ((cpupower_cpu_info.vendor == X86_VENDOR_AMD && - cpupower_cpu_info.family >= 0x10) || - cpupower_cpu_info.vendor == X86_VENDOR_HYGON) { + if (cpupower_cpu_info.vendor == X86_VENDOR_AMD && + cpupower_cpu_info.caps & CPUPOWER_CAP_AMD_PSTATE) { + return 0; + } else if ((cpupower_cpu_info.vendor == X86_VENDOR_AMD && + cpupower_cpu_info.family >= 0x10) || + cpupower_cpu_info.vendor == X86_VENDOR_HYGON) { ret = decode_pstates(cpu, b_states, pstates, &pstate_no); if (ret) return ret; @@ -254,11 +220,11 @@ static int get_boost_mode(unsigned int cpu) if (freqs) { printf(_(" boost frequency steps: ")); while (freqs->next) { - print_speed(freqs->frequency); + print_speed(freqs->frequency, no_rounding); printf(", "); freqs = freqs->next; } - print_speed(freqs->frequency); + print_speed(freqs->frequency, no_rounding); printf("\n"); cpufreq_put_available_frequencies(freqs); } @@ -277,7 +243,7 @@ static int get_freq_kernel(unsigned int cpu, unsigned int human) return -EINVAL; } if (human) { - print_speed(freq); + print_speed(freq, no_rounding); } else printf("%lu", freq); printf(_(" (asserted by call to kernel)\n")); @@ -296,7 +262,7 @@ static int get_freq_hardware(unsigned int cpu, unsigned int human) return -EINVAL; } if (human) { - print_speed(freq); + print_speed(freq, no_rounding); } else printf("%lu", freq); printf(_(" (asserted by call to hardware)\n")); @@ -316,9 +282,9 @@ static int get_hardware_limits(unsigned int cpu, unsigned int human) if (human) { printf(_(" hardware limits: ")); - print_speed(min); + print_speed(min, no_rounding); printf(" - "); - print_speed(max); + print_speed(max, no_rounding); printf("\n"); } else { printf("%lu %lu\n", min, max); @@ -350,9 +316,9 @@ static int get_policy(unsigned int cpu) return -EINVAL; } printf(_(" current policy: frequency should be within ")); - print_speed(policy->min); + print_speed(policy->min, no_rounding); printf(_(" and ")); - print_speed(policy->max); + print_speed(policy->max, no_rounding); printf(".\n "); printf(_("The governor \"%s\" may decide which speed to use\n" @@ -436,7 +402,7 @@ static int get_freq_stats(unsigned int cpu, unsigned int human) struct cpufreq_stats *stats = cpufreq_get_stats(cpu, &total_time); while (stats) { if (human) { - print_speed(stats->frequency); + print_speed(stats->frequency, no_rounding); printf(":%.2f%%", (100.0 * stats->time_in_state) / total_time); } else @@ -472,6 +438,17 @@ static int get_latency(unsigned int cpu, unsigned int human) return 0; } +/* --performance / -c */ + +static int get_perf_cap(unsigned int cpu) +{ + if (cpupower_cpu_info.vendor == X86_VENDOR_AMD && + cpupower_cpu_info.caps & CPUPOWER_CAP_AMD_PSTATE) + amd_pstate_show_perf_and_freq(cpu, no_rounding); + + return 0; +} + static void debug_output_one(unsigned int cpu) { struct cpufreq_available_frequencies *freqs; @@ -486,11 +463,11 @@ static void debug_output_one(unsigned int cpu) if (freqs) { printf(_(" available frequency steps: ")); while (freqs->next) { - print_speed(freqs->frequency); + print_speed(freqs->frequency, no_rounding); printf(", "); freqs = freqs->next; } - print_speed(freqs->frequency); + print_speed(freqs->frequency, no_rounding); printf("\n"); cpufreq_put_available_frequencies(freqs); } @@ -500,6 +477,7 @@ static void debug_output_one(unsigned int cpu) if (get_freq_hardware(cpu, 1) < 0) get_freq_kernel(cpu, 1); get_boost_mode(cpu); + get_perf_cap(cpu); } static struct option info_opts[] = { @@ -518,6 +496,7 @@ static struct option info_opts[] = { {"proc", no_argument, NULL, 'o'}, {"human", no_argument, NULL, 'm'}, {"no-rounding", no_argument, NULL, 'n'}, + {"performance", no_argument, NULL, 'c'}, { }, }; @@ -531,7 +510,7 @@ int cmd_freq_info(int argc, char **argv) int output_param = 0; do { - ret = getopt_long(argc, argv, "oefwldpgrasmybn", info_opts, + ret = getopt_long(argc, argv, "oefwldpgrasmybnc", info_opts, NULL); switch (ret) { case '?': @@ -554,6 +533,7 @@ int cmd_freq_info(int argc, char **argv) case 'e': case 's': case 'y': + case 'c': if (output_param) { output_param = -1; cont = 0; @@ -660,6 +640,9 @@ int cmd_freq_info(int argc, char **argv) case 'y': ret = get_latency(cpu, human); break; + case 'c': + ret = get_perf_cap(cpu); + break; } if (ret) return ret; diff --git a/tools/power/cpupower/utils/helpers/amd.c b/tools/power/cpupower/utils/helpers/amd.c index 97f2c857048e..c519cc89c97f 100644 --- a/tools/power/cpupower/utils/helpers/amd.c +++ b/tools/power/cpupower/utils/helpers/amd.c @@ -8,7 +8,10 @@ #include <pci/pci.h> #include "helpers/helpers.h" +#include "cpufreq.h" +#include "acpi_cppc.h" +/* ACPI P-States Helper Functions for AMD Processors ***************/ #define MSR_AMD_PSTATE_STATUS 0xc0010063 #define MSR_AMD_PSTATE 0xc0010064 #define MSR_AMD_PSTATE_LIMIT 0xc0010061 @@ -146,4 +149,78 @@ int amd_pci_get_num_boost_states(int *active, int *states) pci_cleanup(pci_acc); return 0; } + +/* ACPI P-States Helper Functions for AMD Processors ***************/ + +/* AMD P-State Helper Functions ************************************/ +enum amd_pstate_value { + AMD_PSTATE_HIGHEST_PERF, + AMD_PSTATE_MAX_FREQ, + AMD_PSTATE_LOWEST_NONLINEAR_FREQ, + MAX_AMD_PSTATE_VALUE_READ_FILES, +}; + +static const char *amd_pstate_value_files[MAX_AMD_PSTATE_VALUE_READ_FILES] = { + [AMD_PSTATE_HIGHEST_PERF] = "amd_pstate_highest_perf", + [AMD_PSTATE_MAX_FREQ] = "amd_pstate_max_freq", + [AMD_PSTATE_LOWEST_NONLINEAR_FREQ] = "amd_pstate_lowest_nonlinear_freq", +}; + +static unsigned long amd_pstate_get_data(unsigned int cpu, + enum amd_pstate_value value) +{ + return cpufreq_get_sysfs_value_from_table(cpu, + amd_pstate_value_files, + value, + MAX_AMD_PSTATE_VALUE_READ_FILES); +} + +void amd_pstate_boost_init(unsigned int cpu, int *support, int *active) +{ + unsigned long highest_perf, nominal_perf, cpuinfo_min, + cpuinfo_max, amd_pstate_max; + + highest_perf = amd_pstate_get_data(cpu, AMD_PSTATE_HIGHEST_PERF); + nominal_perf = acpi_cppc_get_data(cpu, NOMINAL_PERF); + + *support = highest_perf > nominal_perf ? 1 : 0; + if (!(*support)) + return; + + cpufreq_get_hardware_limits(cpu, &cpuinfo_min, &cpuinfo_max); + amd_pstate_max = amd_pstate_get_data(cpu, AMD_PSTATE_MAX_FREQ); + + *active = cpuinfo_max == amd_pstate_max ? 1 : 0; +} + +void amd_pstate_show_perf_and_freq(unsigned int cpu, int no_rounding) +{ + printf(_(" AMD PSTATE Highest Performance: %lu. Maximum Frequency: "), + amd_pstate_get_data(cpu, AMD_PSTATE_HIGHEST_PERF)); + /* + * If boost isn't active, the cpuinfo_max doesn't indicate real max + * frequency. So we read it back from amd-pstate sysfs entry. + */ + print_speed(amd_pstate_get_data(cpu, AMD_PSTATE_MAX_FREQ), no_rounding); + printf(".\n"); + + printf(_(" AMD PSTATE Nominal Performance: %lu. Nominal Frequency: "), + acpi_cppc_get_data(cpu, NOMINAL_PERF)); + print_speed(acpi_cppc_get_data(cpu, NOMINAL_FREQ) * 1000, + no_rounding); + printf(".\n"); + + printf(_(" AMD PSTATE Lowest Non-linear Performance: %lu. Lowest Non-linear Frequency: "), + acpi_cppc_get_data(cpu, LOWEST_NONLINEAR_PERF)); + print_speed(amd_pstate_get_data(cpu, AMD_PSTATE_LOWEST_NONLINEAR_FREQ), + no_rounding); + printf(".\n"); + + printf(_(" AMD PSTATE Lowest Performance: %lu. Lowest Frequency: "), + acpi_cppc_get_data(cpu, LOWEST_PERF)); + print_speed(acpi_cppc_get_data(cpu, LOWEST_FREQ) * 1000, no_rounding); + printf(".\n"); +} + +/* AMD P-State Helper Functions ************************************/ #endif /* defined(__i386__) || defined(__x86_64__) */ diff --git a/tools/power/cpupower/utils/helpers/cpuid.c b/tools/power/cpupower/utils/helpers/cpuid.c index 72eb43593180..eae91f11d187 100644 --- a/tools/power/cpupower/utils/helpers/cpuid.c +++ b/tools/power/cpupower/utils/helpers/cpuid.c @@ -149,6 +149,19 @@ out: if (ext_cpuid_level >= 0x80000008 && cpuid_ebx(0x80000008) & (1 << 4)) cpu_info->caps |= CPUPOWER_CAP_AMD_RDPRU; + + if (cpupower_amd_pstate_enabled()) { + cpu_info->caps |= CPUPOWER_CAP_AMD_PSTATE; + + /* + * If AMD P-State is enabled, the firmware will treat + * AMD P-State function as high priority. + */ + cpu_info->caps &= ~CPUPOWER_CAP_AMD_CPB; + cpu_info->caps &= ~CPUPOWER_CAP_AMD_CPB_MSR; + cpu_info->caps &= ~CPUPOWER_CAP_AMD_HW_PSTATE; + cpu_info->caps &= ~CPUPOWER_CAP_AMD_PSTATEDEF; + } } if (cpu_info->vendor == X86_VENDOR_INTEL) { diff --git a/tools/power/cpupower/utils/helpers/helpers.h b/tools/power/cpupower/utils/helpers/helpers.h index 33ffacee7fcb..96e4bede078b 100644 --- a/tools/power/cpupower/utils/helpers/helpers.h +++ b/tools/power/cpupower/utils/helpers/helpers.h @@ -11,6 +11,7 @@ #include <libintl.h> #include <locale.h> +#include <stdbool.h> #include "helpers/bitmask.h" #include <cpupower.h> @@ -73,6 +74,7 @@ enum cpupower_cpu_vendor {X86_VENDOR_UNKNOWN = 0, X86_VENDOR_INTEL, #define CPUPOWER_CAP_AMD_HW_PSTATE 0x00000100 #define CPUPOWER_CAP_AMD_PSTATEDEF 0x00000200 #define CPUPOWER_CAP_AMD_CPB_MSR 0x00000400 +#define CPUPOWER_CAP_AMD_PSTATE 0x00000800 #define CPUPOWER_AMD_CPBDIS 0x02000000 @@ -135,6 +137,16 @@ extern int decode_pstates(unsigned int cpu, int boost_states, extern int cpufreq_has_boost_support(unsigned int cpu, int *support, int *active, int * states); + +/* AMD P-State stuff **************************/ +bool cpupower_amd_pstate_enabled(void); +void amd_pstate_boost_init(unsigned int cpu, + int *support, int *active); +void amd_pstate_show_perf_and_freq(unsigned int cpu, + int no_rounding); + +/* AMD P-State stuff **************************/ + /* * CPUID functions returning a single datum */ @@ -167,6 +179,15 @@ static inline int cpufreq_has_boost_support(unsigned int cpu, int *support, int *active, int * states) { return -1; } +static inline bool cpupower_amd_pstate_enabled(void) +{ return false; } +static inline void amd_pstate_boost_init(unsigned int cpu, int *support, + int *active) +{} +static inline void amd_pstate_show_perf_and_freq(unsigned int cpu, + int no_rounding) +{} + /* cpuid and cpuinfo helpers **************************/ static inline unsigned int cpuid_eax(unsigned int op) { return 0; }; @@ -184,5 +205,6 @@ extern struct bitmask *offline_cpus; void get_cpustate(void); void print_online_cpus(void); void print_offline_cpus(void); +void print_speed(unsigned long speed, int no_rounding); #endif /* __CPUPOWERUTILS_HELPERS__ */ diff --git a/tools/power/cpupower/utils/helpers/misc.c b/tools/power/cpupower/utils/helpers/misc.c index fc6e34511721..9547b29254a7 100644 --- a/tools/power/cpupower/utils/helpers/misc.c +++ b/tools/power/cpupower/utils/helpers/misc.c @@ -3,9 +3,11 @@ #include <stdio.h> #include <errno.h> #include <stdlib.h> +#include <string.h> #include "helpers/helpers.h" #include "helpers/sysfs.h" +#include "cpufreq.h" #if defined(__i386__) || defined(__x86_64__) @@ -39,6 +41,8 @@ int cpufreq_has_boost_support(unsigned int cpu, int *support, int *active, if (ret) return ret; } + } else if (cpupower_cpu_info.caps & CPUPOWER_CAP_AMD_PSTATE) { + amd_pstate_boost_init(cpu, support, active); } else if (cpupower_cpu_info.caps & CPUPOWER_CAP_INTEL_IDA) *support = *active = 1; return 0; @@ -83,6 +87,22 @@ int cpupower_intel_set_perf_bias(unsigned int cpu, unsigned int val) return 0; } +bool cpupower_amd_pstate_enabled(void) +{ + char *driver = cpufreq_get_driver(0); + bool ret = false; + + if (!driver) + return ret; + + if (!strcmp(driver, "amd-pstate")) + ret = true; + + cpufreq_put_driver(driver); + + return ret; +} + #endif /* #if defined(__i386__) || defined(__x86_64__) */ /* get_cpustate @@ -144,3 +164,43 @@ void print_offline_cpus(void) printf(_("cpupower set operation was not performed on them\n")); } } + +/* + * print_speed + * + * Print the exact CPU frequency with appropriate unit + */ +void print_speed(unsigned long speed, int no_rounding) +{ + unsigned long tmp; + + if (no_rounding) { + if (speed > 1000000) + printf("%u.%06u GHz", ((unsigned int)speed / 1000000), + ((unsigned int)speed % 1000000)); + else if (speed > 1000) + printf("%u.%03u MHz", ((unsigned int)speed / 1000), + (unsigned int)(speed % 1000)); + else + printf("%lu kHz", speed); + } else { + if (speed > 1000000) { + tmp = speed % 10000; + if (tmp >= 5000) + speed += 10000; + printf("%u.%02u GHz", ((unsigned int)speed / 1000000), + ((unsigned int)(speed % 1000000) / 10000)); + } else if (speed > 100000) { + tmp = speed % 1000; + if (tmp >= 500) + speed += 1000; + printf("%u MHz", ((unsigned int)speed / 1000)); + } else if (speed > 1000) { + tmp = speed % 100; + if (tmp >= 50) + speed += 100; + printf("%u.%01u MHz", ((unsigned int)speed / 1000), + ((unsigned int)(speed % 1000) / 100)); + } + } +} diff --git a/tools/power/pm-graph/README b/tools/power/pm-graph/README index da468bd510ca..3213dbe63b74 100644 --- a/tools/power/pm-graph/README +++ b/tools/power/pm-graph/README @@ -6,22 +6,22 @@ |_| |___/ |_| pm-graph: suspend/resume/boot timing analysis tools - Version: 5.8 + Version: 5.10 Author: Todd Brandt <todd.e.brandt@intel.com> - Home Page: https://01.org/pm-graph + Home Page: https://www.intel.com/content/www/us/en/developer/topic-technology/open/pm-graph/overview.html Report bugs/issues at bugzilla.kernel.org Tools/pm-graph - https://bugzilla.kernel.org/buglist.cgi?component=pm-graph&product=Tools Full documentation available online & in man pages - Getting Started: - https://01.org/pm-graph/documentation/getting-started + https://www.intel.com/content/www/us/en/developer/articles/technical/usage.html - - Config File Format: - https://01.org/pm-graph/documentation/3-config-file-format + - Feature Summary: + https://www.intel.com/content/www/us/en/developer/topic-technology/open/pm-graph/features.html - upstream version in git: - https://github.com/intel/pm-graph/ + git clone https://github.com/intel/pm-graph/ Table of Contents - Overview @@ -97,8 +97,8 @@ (kernel/pre-3.15/enable_trace_events_suspend_resume.patch) (kernel/pre-3.15/enable_trace_events_device_pm_callback.patch) - If you're using a kernel older than 3.15.0, the following - additional kernel parameters are required: + If you're using bootgraph, or sleepgraph with a kernel older than 3.15.0, + the following additional kernel parameters are required: (e.g. in file /etc/default/grub) GRUB_CMDLINE_LINUX_DEFAULT="... initcall_debug log_buf_len=32M ..." diff --git a/tools/power/pm-graph/bootgraph.py b/tools/power/pm-graph/bootgraph.py index 2823cd3122f7..f96f50e0c336 100755 --- a/tools/power/pm-graph/bootgraph.py +++ b/tools/power/pm-graph/bootgraph.py @@ -69,22 +69,24 @@ class SystemValues(aslib.SystemValues): bootloader = 'grub' blexec = [] def __init__(self): - self.hostname = platform.node() + self.kernel, self.hostname = 'unknown', platform.node() self.testtime = datetime.now().strftime('%Y-%m-%d_%H:%M:%S') if os.path.exists('/proc/version'): fp = open('/proc/version', 'r') - val = fp.read().strip() + self.kernel = self.kernelVersion(fp.read().strip()) fp.close() - self.kernel = self.kernelVersion(val) - else: - self.kernel = 'unknown' self.testdir = datetime.now().strftime('boot-%y%m%d-%H%M%S') def kernelVersion(self, msg): - return msg.split()[2] + m = re.match('^[Ll]inux *[Vv]ersion *(?P<v>\S*) .*', msg) + if m: + return m.group('v') + return 'unknown' def checkFtraceKernelVersion(self): - val = tuple(map(int, self.kernel.split('-')[0].split('.'))) - if val >= (4, 10, 0): - return True + m = re.match('^(?P<x>[0-9]*)\.(?P<y>[0-9]*)\.(?P<z>[0-9]*).*', self.kernel) + if m: + val = tuple(map(int, m.groups())) + if val >= (4, 10, 0): + return True return False def kernelParams(self): cmdline = 'initcall_debug log_buf_len=32M' diff --git a/tools/power/pm-graph/config/custom-timeline-functions.cfg b/tools/power/pm-graph/config/custom-timeline-functions.cfg index 962e5768681c..4f80ad7d7275 100644 --- a/tools/power/pm-graph/config/custom-timeline-functions.cfg +++ b/tools/power/pm-graph/config/custom-timeline-functions.cfg @@ -125,7 +125,7 @@ acpi_suspend_begin: suspend_console: acpi_pm_prepare: syscore_suspend: -arch_thaw_secondary_cpus_end: +arch_enable_nonboot_cpus_end: syscore_resume: acpi_pm_finish: resume_console: diff --git a/tools/power/pm-graph/sleepgraph.8 b/tools/power/pm-graph/sleepgraph.8 index 5126271de98a..643271b6fc6f 100644 --- a/tools/power/pm-graph/sleepgraph.8 +++ b/tools/power/pm-graph/sleepgraph.8 @@ -78,6 +78,9 @@ This helps maintain the consistency of test data for better comparison. If a wifi connection is available, check that it reconnects after resume. Include the reconnect time in the total resume time calculation and treat wifi timeouts as resume failures. +.TP +\fB-wifitrace\fR +Trace through the wifi reconnect time and include it in the timeline. .SS "advanced" .TP diff --git a/tools/power/pm-graph/sleepgraph.py b/tools/power/pm-graph/sleepgraph.py index ffd50953a024..cfe343306e08 100755 --- a/tools/power/pm-graph/sleepgraph.py +++ b/tools/power/pm-graph/sleepgraph.py @@ -66,8 +66,13 @@ from threading import Thread from subprocess import call, Popen, PIPE import base64 +debugtiming = False +mystarttime = time.time() def pprint(msg): - print(msg) + if debugtiming: + print('[%09.3f] %s' % (time.time()-mystarttime, msg)) + else: + print(msg) sys.stdout.flush() def ascii(text): @@ -81,19 +86,21 @@ def ascii(text): # store system values and test parameters class SystemValues: title = 'SleepGraph' - version = '5.8' + version = '5.10' ansi = False rs = 0 display = '' gzip = False sync = False wifi = False + netfix = False verbose = False testlog = True dmesglog = True ftracelog = False acpidebug = True tstat = True + wifitrace = False mindevlen = 0.0001 mincglen = 0.0 cgphase = '' @@ -108,6 +115,7 @@ class SystemValues: cpucount = 0 memtotal = 204800 memfree = 204800 + osversion = '' srgap = 0 cgexp = False testdir = '' @@ -116,6 +124,8 @@ class SystemValues: fpdtpath = '/sys/firmware/acpi/tables/FPDT' epath = '/sys/kernel/debug/tracing/events/power/' pmdpath = '/sys/power/pm_debug_messages' + s0ixpath = '/sys/module/intel_pmc_core/parameters/warn_on_s0ix_failures' + s0ixres = '/sys/devices/system/cpu/cpuidle/low_power_idle_system_residency_us' acpipath='/sys/module/acpi/parameters/debug_level' traceevents = [ 'suspend_resume', @@ -156,6 +166,7 @@ class SystemValues: ftop = False usetraceevents = False usetracemarkers = True + useftrace = True usekprobes = True usedevsrc = False useprocmon = False @@ -171,6 +182,7 @@ class SystemValues: tmstart = 'SUSPEND START %Y%m%d-%H:%M:%S.%f' tmend = 'RESUME COMPLETE %Y%m%d-%H:%M:%S.%f' tracefuncs = { + 'async_synchronize_full': {}, 'sys_sync': {}, 'ksys_sync': {}, '__pm_notifier_call_chain': {}, @@ -279,16 +291,23 @@ class SystemValues: 'intel_fbdev_set_suspend': {}, } infocmds = [ + [0, 'sysinfo', 'uname', '-a'], + [0, 'cpuinfo', 'head', '-7', '/proc/cpuinfo'], [0, 'kparams', 'cat', '/proc/cmdline'], [0, 'mcelog', 'mcelog'], [0, 'pcidevices', 'lspci', '-tv'], - [0, 'usbdevices', 'lsusb', '-t'], + [0, 'usbdevices', 'lsusb', '-tv'], + [0, 'acpidevices', 'sh', '-c', 'ls -l /sys/bus/acpi/devices/*/physical_node'], + [0, 's0ix_require', 'cat', '/sys/kernel/debug/pmc_core/substate_requirements'], + [0, 's0ix_debug', 'cat', '/sys/kernel/debug/pmc_core/slp_s0_debug_status'], + [1, 's0ix_residency', 'cat', '/sys/kernel/debug/pmc_core/slp_s0_residency_usec'], [1, 'interrupts', 'cat', '/proc/interrupts'], [1, 'wakeups', 'cat', '/sys/kernel/debug/wakeup_sources'], [2, 'gpecounts', 'sh', '-c', 'grep -v invalid /sys/firmware/acpi/interrupts/*'], [2, 'suspendstats', 'sh', '-c', 'grep -v invalid /sys/power/suspend_stats/*'], [2, 'cpuidle', 'sh', '-c', 'grep -v invalid /sys/devices/system/cpu/cpu*/cpuidle/state*/s2idle/*'], [2, 'battery', 'sh', '-c', 'grep -v invalid /sys/class/power_supply/*/*'], + [2, 'thermal', 'sh', '-c', 'grep . /sys/class/thermal/thermal_zone*/temp'], ] cgblacklist = [] kprobes = dict() @@ -358,8 +377,19 @@ class SystemValues: self.outputResult({'error':msg}) sys.exit(1) return False - def usable(self, file): - return (os.path.exists(file) and os.path.getsize(file) > 0) + def usable(self, file, ishtml=False): + if not os.path.exists(file) or os.path.getsize(file) < 1: + return False + if ishtml: + try: + fp = open(file, 'r') + res = fp.read(1000) + fp.close() + except: + return False + if '<html>' not in res: + return False + return True def getExec(self, cmd): try: fp = Popen(['which', cmd], stdout=PIPE, stderr=PIPE).stdout @@ -413,12 +443,16 @@ class SystemValues: r = info['bios-release-date'] if 'bios-release-date' in info else '' self.sysstamp = '# sysinfo | man:%s | plat:%s | cpu:%s | bios:%s | biosdate:%s | numcpu:%d | memsz:%d | memfr:%d' % \ (m, p, c, b, r, self.cpucount, self.memtotal, self.memfree) + if self.osversion: + self.sysstamp += ' | os:%s' % self.osversion def printSystemInfo(self, fatal=False): self.rootCheck(True) out = dmidecode(self.mempath, fatal) if len(out) < 1: return fmt = '%-24s: %s' + if self.osversion: + print(fmt % ('os-version', self.osversion)) for name in sorted(out): print(fmt % (name, out[name])) print(fmt % ('cpucount', ('%d' % self.cpucount))) @@ -426,20 +460,25 @@ class SystemValues: print(fmt % ('memfree', ('%d kB' % self.memfree))) def cpuInfo(self): self.cpucount = 0 - fp = open('/proc/cpuinfo', 'r') - for line in fp: - if re.match('^processor[ \t]*:[ \t]*[0-9]*', line): - self.cpucount += 1 - fp.close() - fp = open('/proc/meminfo', 'r') - for line in fp: - m = re.match('^MemTotal:[ \t]*(?P<sz>[0-9]*) *kB', line) - if m: - self.memtotal = int(m.group('sz')) - m = re.match('^MemFree:[ \t]*(?P<sz>[0-9]*) *kB', line) - if m: - self.memfree = int(m.group('sz')) - fp.close() + if os.path.exists('/proc/cpuinfo'): + with open('/proc/cpuinfo', 'r') as fp: + for line in fp: + if re.match('^processor[ \t]*:[ \t]*[0-9]*', line): + self.cpucount += 1 + if os.path.exists('/proc/meminfo'): + with open('/proc/meminfo', 'r') as fp: + for line in fp: + m = re.match('^MemTotal:[ \t]*(?P<sz>[0-9]*) *kB', line) + if m: + self.memtotal = int(m.group('sz')) + m = re.match('^MemFree:[ \t]*(?P<sz>[0-9]*) *kB', line) + if m: + self.memfree = int(m.group('sz')) + if os.path.exists('/etc/os-release'): + with open('/etc/os-release', 'r') as fp: + for line in fp: + if line.startswith('PRETTY_NAME='): + self.osversion = line[12:].strip().replace('"', '') def initTestOutput(self, name): self.prefix = self.hostname v = open('/proc/version', 'r').read().strip() @@ -698,6 +737,8 @@ class SystemValues: return False return True def fsetVal(self, val, path): + if not self.useftrace: + return False return self.setVal(val, self.tpath+path) def getVal(self, file): res = '' @@ -711,9 +752,11 @@ class SystemValues: pass return res def fgetVal(self, path): + if not self.useftrace: + return '' return self.getVal(self.tpath+path) def cleanupFtrace(self): - if(self.usecallgraph or self.usetraceevents or self.usedevsrc): + if self.useftrace: self.fsetVal('0', 'events/kprobes/enable') self.fsetVal('', 'kprobe_events') self.fsetVal('1024', 'buffer_size_kb') @@ -734,13 +777,14 @@ class SystemValues: return True return False def initFtrace(self, quiet=False): + if not self.useftrace: + return if not quiet: sysvals.printSystemInfo(False) - pprint('INITIALIZING FTRACE...') + pprint('INITIALIZING FTRACE') # turn trace off self.fsetVal('0', 'tracing_on') self.cleanupFtrace() - self.testVal(self.pmdpath, 'basic', '1') # set the trace clock to global self.fsetVal('global', 'trace_clock') self.fsetVal('nop', 'current_tracer') @@ -766,6 +810,10 @@ class SystemValues: # set trace type self.fsetVal('function_graph', 'current_tracer') self.fsetVal('', 'set_ftrace_filter') + # temporary hack to fix https://bugzilla.kernel.org/show_bug.cgi?id=212761 + fp = open(self.tpath+'set_ftrace_notrace', 'w') + fp.write('native_queued_spin_lock_slowpath\ndev_driver_string') + fp.close() # set trace format options self.fsetVal('print-parent', 'trace_options') self.fsetVal('funcgraph-abstime', 'trace_options') @@ -797,7 +845,7 @@ class SystemValues: for name in self.dev_tracefuncs: self.defaultKprobe(name, self.dev_tracefuncs[name]) if not quiet: - pprint('INITIALIZING KPROBES...') + pprint('INITIALIZING KPROBES') self.addKprobes(self.verbose) if(self.usetraceevents): # turn trace events on @@ -846,6 +894,8 @@ class SystemValues: fp.write('# turbostat %s\n' % test['turbo']) if 'wifi' in test: fp.write('# wifi %s\n' % test['wifi']) + if 'netfix' in test: + fp.write('# netfix %s\n' % test['netfix']) if test['error'] or len(testdata) > 1: fp.write('# enter_sleep_error %s\n' % test['error']) return fp @@ -865,6 +915,8 @@ class SystemValues: fp.write('error%s: %s\n' % (n, testdata['error'])) else: fp.write('result%s: pass\n' % n) + if 'mode' in testdata: + fp.write('mode%s: %s\n' % (n, testdata['mode'])) for v in ['suspend', 'resume', 'boot', 'lastinit']: if v in testdata: fp.write('%s%s: %.3f\n' % (v, n, testdata[v])) @@ -901,6 +953,8 @@ class SystemValues: fp.write(text) fp.close() def dlog(self, text): + if not self.dmesgfile: + return self.putlog(self.dmesgfile, '# %s\n' % text) def flog(self, text): self.putlog(self.ftracefile, text) @@ -954,34 +1008,31 @@ class SystemValues: dirname = props[dev].syspath if not dirname or not os.path.exists(dirname): continue - with open(dirname+'/power/async') as fp: - text = fp.read() - props[dev].isasync = False - if 'enabled' in text: + props[dev].isasync = False + if os.path.exists(dirname+'/power/async'): + fp = open(dirname+'/power/async') + if 'enabled' in fp.read(): props[dev].isasync = True + fp.close() fields = os.listdir(dirname) - if 'product' in fields: - with open(dirname+'/product', 'rb') as fp: - props[dev].altname = ascii(fp.read()) - elif 'name' in fields: - with open(dirname+'/name', 'rb') as fp: - props[dev].altname = ascii(fp.read()) - elif 'model' in fields: - with open(dirname+'/model', 'rb') as fp: - props[dev].altname = ascii(fp.read()) - elif 'description' in fields: - with open(dirname+'/description', 'rb') as fp: - props[dev].altname = ascii(fp.read()) - elif 'id' in fields: - with open(dirname+'/id', 'rb') as fp: - props[dev].altname = ascii(fp.read()) - elif 'idVendor' in fields and 'idProduct' in fields: - idv, idp = '', '' - with open(dirname+'/idVendor', 'rb') as fp: - idv = ascii(fp.read()).strip() - with open(dirname+'/idProduct', 'rb') as fp: - idp = ascii(fp.read()).strip() - props[dev].altname = '%s:%s' % (idv, idp) + for file in ['product', 'name', 'model', 'description', 'id', 'idVendor']: + if file not in fields: + continue + try: + with open(os.path.join(dirname, file), 'rb') as fp: + props[dev].altname = ascii(fp.read()) + except: + continue + if file == 'idVendor': + idv, idp = props[dev].altname.strip(), '' + try: + with open(os.path.join(dirname, 'idProduct'), 'rb') as fp: + idp = ascii(fp.read()).strip() + except: + props[dev].altname = '' + break + props[dev].altname = '%s:%s' % (idv, idp) + break if props[dev].altname: out = props[dev].altname.strip().replace('\n', ' ')\ .replace(',', ' ').replace(';', ' ') @@ -1047,7 +1098,7 @@ class SystemValues: self.cmd1[name] = self.dictify(info, delta) elif not debug and delta and name in self.cmd1: before, after = self.cmd1[name], self.dictify(info, delta) - dinfo = ('\t%s\n' % before['@']) if '@' in before else '' + dinfo = ('\t%s\n' % before['@']) if '@' in before and len(before) > 1 else '' prefix = self.commonPrefix(list(before.keys())) for key in sorted(before): if key in after and before[key] != after[key]: @@ -1086,6 +1137,15 @@ class SystemValues: self.cfgdef[file] = fp.read().strip() fp.write(value) fp.close() + def s0ixSupport(self): + if not os.path.exists(self.s0ixres) or not os.path.exists(self.mempowerfile): + return False + fp = open(sysvals.mempowerfile, 'r') + data = fp.read().strip() + fp.close() + if '[s2idle]' in data: + return True + return False def haveTurbostat(self): if not self.tstat: return False @@ -1099,7 +1159,7 @@ class SystemValues: self.vprint(out) return True return False - def turbostat(self): + def turbostat(self, s0ixready): cmd = self.getExec('turbostat') rawout = keyline = valline = '' fullcmd = '%s -q -S echo freeze > %s' % (cmd, self.powerfile) @@ -1126,8 +1186,18 @@ class SystemValues: for key in keyline: idx = keyline.index(key) val = valline[idx] + if key == 'SYS%LPI' and not s0ixready and re.match('^[0\.]*$', val): + continue out.append('%s=%s' % (key, val)) return '|'.join(out) + def netfixon(self, net='both'): + cmd = self.getExec('netfix') + if not cmd: + return '' + fp = Popen([cmd, '-s', net, 'on'], stdout=PIPE, stderr=PIPE).stdout + out = ascii(fp.read()).strip() + fp.close() + return out def wifiDetails(self, dev): try: info = open('/sys/class/net/%s/device/uevent' % dev, 'r').read().strip() @@ -1144,12 +1214,12 @@ class SystemValues: except: return '' for line in reversed(w.split('\n')): - m = re.match(' *(?P<dev>.*): (?P<stat>[0-9a-f]*) .*', w.split('\n')[-1]) + m = re.match(' *(?P<dev>.*): (?P<stat>[0-9a-f]*) .*', line) if not m or (dev and dev != m.group('dev')): continue return m.group('dev') return '' - def pollWifi(self, dev, timeout=60): + def pollWifi(self, dev, timeout=10): start = time.time() while (time.time() - start) < timeout: w = self.checkWifi(dev) @@ -1278,15 +1348,29 @@ class SystemValues: for i in self.rslist: self.setVal(self.rstgt, i) pprint('runtime suspend settings restored on %d devices' % len(self.rslist)) + def start(self, pm): + if self.useftrace: + self.dlog('start ftrace tracing') + self.fsetVal('1', 'tracing_on') + if self.useprocmon: + self.dlog('start the process monitor') + pm.start() + def stop(self, pm): + if self.useftrace: + if self.useprocmon: + self.dlog('stop the process monitor') + pm.stop() + self.dlog('stop ftrace tracing') + self.fsetVal('0', 'tracing_on') sysvals = SystemValues() switchvalues = ['enable', 'disable', 'on', 'off', 'true', 'false', '1', '0'] switchoff = ['disable', 'off', 'false', '0'] suspendmodename = { - 'freeze': 'Freeze (S0)', - 'standby': 'Standby (S1)', - 'mem': 'Suspend (S3)', - 'disk': 'Hibernate (S4)' + 'standby': 'standby (S1)', + 'freeze': 'freeze (S2idle)', + 'mem': 'suspend (S3)', + 'disk': 'hibernate (S4)' } # Class: DevProps @@ -1376,6 +1460,7 @@ class Data: 'INVALID' : r'(?i).*\bINVALID\b.*', 'CRASH' : r'(?i).*\bCRASHED\b.*', 'TIMEOUT' : r'(?i).*\bTIMEOUT\b.*', + 'ABORT' : r'(?i).*\bABORT\b.*', 'IRQ' : r'.*\bgenirq: .*', 'TASKFAIL': r'.*Freezing of tasks *.*', 'ACPI' : r'.*\bACPI *(?P<b>[A-Za-z]*) *Error[: ].*', @@ -1574,19 +1659,20 @@ class Data: ubiquitous = False if kprobename in dtf and 'ub' in dtf[kprobename]: ubiquitous = True - title = cdata+' '+rdata - mstr = '\(.*\) *(?P<args>.*) *\((?P<caller>.*)\+.* arg1=(?P<ret>.*)' - m = re.match(mstr, title) - if m: - c = m.group('caller') - a = m.group('args').strip() - r = m.group('ret') + mc = re.match('\(.*\) *(?P<args>.*)', cdata) + mr = re.match('\((?P<caller>\S*).* arg1=(?P<ret>.*)', rdata) + if mc and mr: + c = mr.group('caller').split('+')[0] + a = mc.group('args').strip() + r = mr.group('ret') if len(r) > 6: r = '' else: r = 'ret=%s ' % r if ubiquitous and c in dtf and 'ub' in dtf[c]: return False + else: + return False color = sysvals.kprobeColor(kprobename) e = DevFunction(displayname, a, c, r, start, end, ubiquitous, proc, pid, color) tgtdev['src'].append(e) @@ -1703,6 +1789,14 @@ class Data: e.time = self.trimTimeVal(e.time, t0, dT, left) e.end = self.trimTimeVal(e.end, t0, dT, left) e.length = e.end - e.time + if('cpuexec' in d): + cpuexec = dict() + for e in d['cpuexec']: + c0, cN = e + c0 = self.trimTimeVal(c0, t0, dT, left) + cN = self.trimTimeVal(cN, t0, dT, left) + cpuexec[(c0, cN)] = d['cpuexec'][e] + d['cpuexec'] = cpuexec for dir in ['suspend', 'resume']: list = [] for e in self.errorinfo[dir]: @@ -1724,9 +1818,9 @@ class Data: if 'waking' in self.dmesg[lp]: tCnt = self.dmesg[lp]['waking'][0] if self.dmesg[lp]['waking'][1] >= 0.001: - tTry = '-%.0f' % (round(self.dmesg[lp]['waking'][1] * 1000)) + tTry = '%.0f' % (round(self.dmesg[lp]['waking'][1] * 1000)) else: - tTry = '-%.3f' % (self.dmesg[lp]['waking'][1] * 1000) + tTry = '%.3f' % (self.dmesg[lp]['waking'][1] * 1000) text = '%.0f (%s ms waking %d times)' % (tL * 1000, tTry, tCnt) else: text = '%.0f' % (tL * 1000) @@ -2017,75 +2111,43 @@ class Data: return d def addProcessUsageEvent(self, name, times): # get the start and end times for this process - maxC = 0 - tlast = 0 - start = -1 - end = -1 + cpuexec = dict() + tlast = start = end = -1 for t in sorted(times): - if tlast == 0: + if tlast < 0: tlast = t continue - if name in self.pstl[t]: - if start == -1 or tlast < start: + if name in self.pstl[t] and self.pstl[t][name] > 0: + if start < 0: start = tlast - if end == -1 or t > end: - end = t + end, key = t, (tlast, t) + maxj = (t - tlast) * 1024.0 + cpuexec[key] = min(1.0, float(self.pstl[t][name]) / maxj) tlast = t - if start == -1 or end == -1: - return 0 + if start < 0 or end < 0: + return # add a new action for this process and get the object out = self.newActionGlobal(name, start, end, -3) - if not out: - return 0 - phase, devname = out - dev = self.dmesg[phase]['list'][devname] - # get the cpu exec data - tlast = 0 - clast = 0 - cpuexec = dict() - for t in sorted(times): - if tlast == 0 or t <= start or t > end: - tlast = t - continue - list = self.pstl[t] - c = 0 - if name in list: - c = list[name] - if c > maxC: - maxC = c - if c != clast: - key = (tlast, t) - cpuexec[key] = c - tlast = t - clast = c - dev['cpuexec'] = cpuexec - return maxC + if out: + phase, devname = out + dev = self.dmesg[phase]['list'][devname] + dev['cpuexec'] = cpuexec def createProcessUsageEvents(self): - # get an array of process names - proclist = [] + # get an array of process names and times + proclist = {'sus': dict(), 'res': dict()} + tdata = {'sus': [], 'res': []} for t in sorted(self.pstl): - pslist = self.pstl[t] - for ps in sorted(pslist): - if ps not in proclist: - proclist.append(ps) - # get a list of data points for suspend and resume - tsus = [] - tres = [] - for t in sorted(self.pstl): - if t < self.tSuspended: - tsus.append(t) - else: - tres.append(t) + dir = 'sus' if t < self.tSuspended else 'res' + for ps in sorted(self.pstl[t]): + if ps not in proclist[dir]: + proclist[dir][ps] = 0 + tdata[dir].append(t) # process the events for suspend and resume - if len(proclist) > 0: + if len(proclist['sus']) > 0 or len(proclist['res']) > 0: sysvals.vprint('Process Execution:') - for ps in proclist: - c = self.addProcessUsageEvent(ps, tsus) - if c > 0: - sysvals.vprint('%25s (sus): %d' % (ps, c)) - c = self.addProcessUsageEvent(ps, tres) - if c > 0: - sysvals.vprint('%25s (res): %d' % (ps, c)) + for dir in ['sus', 'res']: + for ps in sorted(proclist[dir]): + self.addProcessUsageEvent(ps, tdata[dir]) def handleEndMarker(self, time, msg=''): dm = self.dmesg self.setEnd(time, msg) @@ -2107,6 +2169,30 @@ class Data: # set resume complete to end at end marker if 'resume_complete' in dm: dm['resume_complete']['end'] = time + def initcall_debug_call(self, line, quick=False): + m = re.match('.*(\[ *)(?P<t>[0-9\.]*)(\]) .* (?P<f>.*)\: '+\ + 'PM: *calling .* @ (?P<n>.*), parent: (?P<p>.*)', line) + if not m: + m = re.match('.*(\[ *)(?P<t>[0-9\.]*)(\]) .* (?P<f>.*)\: '+\ + 'calling .* @ (?P<n>.*), parent: (?P<p>.*)', line) + if not m: + m = re.match('.*(\[ *)(?P<t>[0-9\.]*)(\]) calling '+\ + '(?P<f>.*)\+ @ (?P<n>.*), parent: (?P<p>.*)', line) + if m: + return True if quick else m.group('t', 'f', 'n', 'p') + return False if quick else ('', '', '', '') + def initcall_debug_return(self, line, quick=False): + m = re.match('.*(\[ *)(?P<t>[0-9\.]*)(\]) .* (?P<f>.*)\: PM: '+\ + '.* returned (?P<r>[0-9]*) after (?P<dt>[0-9]*) usecs', line) + if not m: + m = re.match('.*(\[ *)(?P<t>[0-9\.]*)(\]) .* (?P<f>.*)\: '+\ + '.* returned (?P<r>[0-9]*) after (?P<dt>[0-9]*) usecs', line) + if not m: + m = re.match('.*(\[ *)(?P<t>[0-9\.]*)(\]) call '+\ + '(?P<f>.*)\+ returned .* after (?P<dt>.*) usecs', line) + if m: + return True if quick else m.group('t', 'f', 'dt') + return False if quick else ('', '', '') def debugPrint(self): for p in self.sortedPhases(): list = self.dmesg[p]['list'] @@ -2880,10 +2966,11 @@ class TestProps: cmdlinefmt = '^# command \| (?P<cmd>.*)' kparamsfmt = '^# kparams \| (?P<kp>.*)' devpropfmt = '# Device Properties: .*' - pinfofmt = '# platform-(?P<val>[a-z,A-Z,0-9]*): (?P<info>.*)' + pinfofmt = '# platform-(?P<val>[a-z,A-Z,0-9,_]*): (?P<info>.*)' tracertypefmt = '# tracer: (?P<t>.*)' firmwarefmt = '# fwsuspend (?P<s>[0-9]*) fwresume (?P<r>[0-9]*)$' procexecfmt = 'ps - (?P<ps>.*)$' + procmultifmt = '@(?P<n>[0-9]*)\|(?P<ps>.*)$' ftrace_line_fmt_fg = \ '^ *(?P<time>[0-9\.]*) *\| *(?P<cpu>[0-9]*)\)'+\ ' *(?P<proc>.*)-(?P<pid>[0-9]*) *\|'+\ @@ -2893,6 +2980,9 @@ class TestProps: '(?P<flags>\S*) *(?P<time>[0-9\.]*): *'+\ '(?P<msg>.*)' machinesuspend = 'machine_suspend\[.*' + multiproclist = dict() + multiproctime = 0.0 + multiproccnt = 0 def __init__(self): self.stamp = '' self.sysinfo = '' @@ -3063,6 +3153,7 @@ class TestRun: self.ttemp = dict() class ProcessMonitor: + maxchars = 512 def __init__(self): self.proclist = dict() self.running = False @@ -3088,19 +3179,23 @@ class ProcessMonitor: if ujiff > 0 or kjiff > 0: running[pid] = ujiff + kjiff process.wait() - out = '' + out = [''] for pid in running: jiffies = running[pid] val = self.proclist[pid] - if out: - out += ',' - out += '%s-%s %d' % (val['name'], pid, jiffies) - return 'ps - '+out + if len(out[-1]) > self.maxchars: + out.append('') + elif len(out[-1]) > 0: + out[-1] += ',' + out[-1] += '%s-%s %d' % (val['name'], pid, jiffies) + if len(out) > 1: + for line in out: + sysvals.fsetVal('ps - @%d|%s' % (len(out), line), 'trace_marker') + else: + sysvals.fsetVal('ps - %s' % out[0], 'trace_marker') def processMonitor(self, tid): while self.running: - out = self.procstat() - if out: - sysvals.fsetVal(out, 'trace_marker') + self.procstat() def start(self): self.thread = Thread(target=self.processMonitor, args=(0,)) self.running = True @@ -3116,7 +3211,7 @@ class ProcessMonitor: # markers, and/or kprobes required for primary parsing. def doesTraceLogHaveTraceEvents(): kpcheck = ['_cal: (', '_ret: ('] - techeck = ['suspend_resume', 'device_pm_callback'] + techeck = ['suspend_resume', 'device_pm_callback', 'tracing_mark_write'] tmcheck = ['SUSPEND START', 'RESUME COMPLETE'] sysvals.usekprobes = False fp = sysvals.openlog(sysvals.ftracefile, 'r') @@ -3139,12 +3234,11 @@ def doesTraceLogHaveTraceEvents(): check.remove(i) tmcheck = check fp.close() - sysvals.usetraceevents = True if len(techeck) < 2 else False + sysvals.usetraceevents = True if len(techeck) < 3 else False sysvals.usetracemarkers = True if len(tmcheck) == 0 else False # Function: appendIncompleteTraceLog # Description: -# [deprecated for kernel 3.15 or newer] # Adds callgraph data which lacks trace event data. This is only # for timelines generated from 3.15 or older # Arguments: @@ -3246,6 +3340,61 @@ def appendIncompleteTraceLog(testruns): dev['ftrace'] = cg break +# Function: loadTraceLog +# Description: +# load the ftrace file into memory and fix up any ordering issues +# Output: +# TestProps instance and an array of lines in proper order +def loadTraceLog(): + tp, data, lines, trace = TestProps(), dict(), [], [] + tf = sysvals.openlog(sysvals.ftracefile, 'r') + for line in tf: + # remove any latent carriage returns + line = line.replace('\r\n', '') + if tp.stampInfo(line, sysvals): + continue + # ignore all other commented lines + if line[0] == '#': + continue + # ftrace line: parse only valid lines + m = re.match(tp.ftrace_line_fmt, line) + if(not m): + continue + dur = m.group('dur') if tp.cgformat else 'traceevent' + info = (m.group('time'), m.group('proc'), m.group('pid'), + m.group('msg'), dur) + # group the data by timestamp + t = float(info[0]) + if t in data: + data[t].append(info) + else: + data[t] = [info] + # we only care about trace event ordering + if (info[3].startswith('suspend_resume:') or \ + info[3].startswith('tracing_mark_write:')) and t not in trace: + trace.append(t) + tf.close() + for t in sorted(data): + first, last, blk = [], [], data[t] + if len(blk) > 1 and t in trace: + # move certain lines to the start or end of a timestamp block + for i in range(len(blk)): + if 'SUSPEND START' in blk[i][3]: + first.append(i) + elif re.match('.* timekeeping_freeze.*begin', blk[i][3]): + last.append(i) + elif re.match('.* timekeeping_freeze.*end', blk[i][3]): + first.append(i) + elif 'RESUME COMPLETE' in blk[i][3]: + last.append(i) + if len(first) == 1 and len(last) == 0: + blk.insert(0, blk.pop(first[0])) + elif len(last) == 1 and len(first) == 0: + blk.append(blk.pop(last[0])) + for info in blk: + lines.append(info) + return (tp, lines) + # Function: parseTraceLog # Description: # Analyze an ftrace log output file generated from this app during @@ -3271,32 +3420,12 @@ def parseTraceLog(live=False): # extract the callgraph and traceevent data s2idle_enter = hwsus = False - tp = TestProps() testruns, testdata = [], [] testrun, data, limbo = 0, 0, True - tf = sysvals.openlog(sysvals.ftracefile, 'r') phase = 'suspend_prepare' - for line in tf: - # remove any latent carriage returns - line = line.replace('\r\n', '') - if tp.stampInfo(line, sysvals): - continue - # ignore all other commented lines - if line[0] == '#': - continue - # ftrace line: parse only valid lines - m = re.match(tp.ftrace_line_fmt, line) - if(not m): - continue + tp, tf = loadTraceLog() + for m_time, m_proc, m_pid, m_msg, m_param3 in tf: # gather the basic message data from the line - m_time = m.group('time') - m_proc = m.group('proc') - m_pid = m.group('pid') - m_msg = m.group('msg') - if(tp.cgformat): - m_param3 = m.group('dur') - else: - m_param3 = 'traceevent' if(m_time and m_pid and m_msg): t = FTraceLine(m_time, m_msg, m_param3) pid = int(m_pid) @@ -3320,16 +3449,33 @@ def parseTraceLog(live=False): continue # process cpu exec line if t.type == 'tracing_mark_write': + if t.name == 'CMD COMPLETE' and data.tKernRes == 0: + data.tKernRes = t.time m = re.match(tp.procexecfmt, t.name) if(m): - proclist = dict() - for ps in m.group('ps').split(','): + parts, msg = 1, m.group('ps') + m = re.match(tp.procmultifmt, msg) + if(m): + parts, msg = int(m.group('n')), m.group('ps') + if tp.multiproccnt == 0: + tp.multiproctime = t.time + tp.multiproclist = dict() + proclist = tp.multiproclist + tp.multiproccnt += 1 + else: + proclist = dict() + tp.multiproccnt = 0 + for ps in msg.split(','): val = ps.split() - if not val: + if not val or len(val) != 2: continue name = val[0].replace('--', '-') proclist[name] = int(val[1]) - data.pstl[t.time] = proclist + if parts == 1: + data.pstl[t.time] = proclist + elif parts == tp.multiproccnt: + data.pstl[tp.multiproctime] = proclist + tp.multiproccnt = 0 continue # find the end of resume if(t.endMarker()): @@ -3523,6 +3669,9 @@ def parseTraceLog(live=False): e = next((x for x in reversed(tp.ktemp[key]) if x['end'] < 0), 0) if not e: continue + if (t.time - e['begin']) * 1000 < sysvals.mindevlen: + tp.ktemp[key].pop() + continue e['end'] = t.time e['rdata'] = kprobedata # end of kernel resume @@ -3545,7 +3694,6 @@ def parseTraceLog(live=False): testrun.ftemp[key].append(FTraceCallGraph(pid, sysvals)) if(res == -1): testrun.ftemp[key][-1].addLine(t) - tf.close() if len(testdata) < 1: sysvals.vprint('WARNING: ftrace start marker is missing') if data and not data.devicegroups: @@ -3667,7 +3815,13 @@ def parseTraceLog(live=False): if p not in data.dmesg: if not terr: ph = p if 'machine' in p else lp - terr = '%s%s failed in %s phase' % (sysvals.suspendmode, tn, ph) + if p == 'suspend_machine': + sm = sysvals.suspendmode + if sm in suspendmodename: + sm = suspendmodename[sm] + terr = 'test%s did not enter %s power mode' % (tn, sm) + else: + terr = '%s%s failed in %s phase' % (sysvals.suspendmode, tn, ph) pprint('TEST%s FAILED: %s' % (tn, terr)) error.append(terr) if data.tSuspended == 0: @@ -3708,9 +3862,7 @@ def parseTraceLog(live=False): # Function: loadKernelLog # Description: -# [deprecated for kernel 3.15.0 or newer] # load the dmesg file into memory and fix up any ordering issues -# The dmesg filename is taken from sysvals # Output: # An array of empty Data objects with only their dmesgtext attributes set def loadKernelLog(): @@ -3736,7 +3888,8 @@ def loadKernelLog(): if(not m): continue msg = m.group("msg") - if(re.match('PM: Syncing filesystems.*', msg)): + if re.match('PM: Syncing filesystems.*', msg) or \ + re.match('PM: suspend entry.*', msg): if(data): testruns.append(data) data = Data(len(testruns)) @@ -3747,11 +3900,17 @@ def loadKernelLog(): if(m): sysvals.stamp['kernel'] = m.group('k') m = re.match('PM: Preparing system for (?P<m>.*) sleep', msg) - if(m): + if not m: + m = re.match('PM: Preparing system for sleep \((?P<m>.*)\)', msg) + if m: sysvals.stamp['mode'] = sysvals.suspendmode = m.group('m') data.dmesgtext.append(line) lf.close() + if sysvals.suspendmode == 's2idle': + sysvals.suspendmode = 'freeze' + elif sysvals.suspendmode == 'deep': + sysvals.suspendmode = 'mem' if data: testruns.append(data) if len(testruns) < 1: @@ -3762,12 +3921,9 @@ def loadKernelLog(): for data in testruns: last = '' for line in data.dmesgtext: - mc = re.match('.*(\[ *)(?P<t>[0-9\.]*)(\]) calling '+\ - '(?P<f>.*)\+ @ .*, parent: .*', line) - mr = re.match('.*(\[ *)(?P<t>[0-9\.]*)(\]) call '+\ - '(?P<f>.*)\+ returned .* after (?P<dt>.*) usecs', last) - if(mc and mr and (mc.group('t') == mr.group('t')) and - (mc.group('f') == mr.group('f'))): + ct, cf, n, p = data.initcall_debug_call(line) + rt, rf, l = data.initcall_debug_return(last) + if ct and rt and ct == rt and cf == rf: i = data.dmesgtext.index(last) j = data.dmesgtext.index(line) data.dmesgtext[i] = line @@ -3777,7 +3933,6 @@ def loadKernelLog(): # Function: parseKernelLog # Description: -# [deprecated for kernel 3.15.0 or newer] # Analyse a dmesg log output file generated from this app during # the execution phase. Create a set of device structures in memory # for subsequent formatting in the html output file @@ -3796,30 +3951,30 @@ def parseKernelLog(data): # dmesg phase match table dm = { - 'suspend_prepare': ['PM: Syncing filesystems.*'], - 'suspend': ['PM: Entering [a-z]* sleep.*', 'Suspending console.*'], - 'suspend_late': ['PM: suspend of devices complete after.*'], - 'suspend_noirq': ['PM: late suspend of devices complete after.*'], - 'suspend_machine': ['PM: noirq suspend of devices complete after.*'], - 'resume_machine': ['ACPI: Low-level resume complete.*'], - 'resume_noirq': ['ACPI: Waking up from system sleep state.*'], - 'resume_early': ['PM: noirq resume of devices complete after.*'], - 'resume': ['PM: early resume of devices complete after.*'], - 'resume_complete': ['PM: resume of devices complete after.*'], + 'suspend_prepare': ['PM: Syncing filesystems.*', 'PM: suspend entry.*'], + 'suspend': ['PM: Entering [a-z]* sleep.*', 'Suspending console.*', + 'PM: Suspending system .*'], + 'suspend_late': ['PM: suspend of devices complete after.*', + 'PM: freeze of devices complete after.*'], + 'suspend_noirq': ['PM: late suspend of devices complete after.*', + 'PM: late freeze of devices complete after.*'], + 'suspend_machine': ['PM: suspend-to-idle', + 'PM: noirq suspend of devices complete after.*', + 'PM: noirq freeze of devices complete after.*'], + 'resume_machine': ['PM: Timekeeping suspended for.*', + 'ACPI: Low-level resume complete.*', + 'ACPI: resume from mwait', + 'Suspended for [0-9\.]* seconds'], + 'resume_noirq': ['PM: resume from suspend-to-idle', + 'ACPI: Waking up from system sleep state.*'], + 'resume_early': ['PM: noirq resume of devices complete after.*', + 'PM: noirq restore of devices complete after.*'], + 'resume': ['PM: early resume of devices complete after.*', + 'PM: early restore of devices complete after.*'], + 'resume_complete': ['PM: resume of devices complete after.*', + 'PM: restore of devices complete after.*'], 'post_resume': ['.*Restarting tasks \.\.\..*'], } - if(sysvals.suspendmode == 'standby'): - dm['resume_machine'] = ['PM: Restoring platform NVS memory'] - elif(sysvals.suspendmode == 'disk'): - dm['suspend_late'] = ['PM: freeze of devices complete after.*'] - dm['suspend_noirq'] = ['PM: late freeze of devices complete after.*'] - dm['suspend_machine'] = ['PM: noirq freeze of devices complete after.*'] - dm['resume_machine'] = ['PM: Restoring platform NVS memory'] - dm['resume_early'] = ['PM: noirq restore of devices complete after.*'] - dm['resume'] = ['PM: early restore of devices complete after.*'] - dm['resume_complete'] = ['PM: restore of devices complete after.*'] - elif(sysvals.suspendmode == 'freeze'): - dm['resume_machine'] = ['ACPI: resume from mwait'] # action table (expected events that occur and show up in dmesg) at = { @@ -3867,12 +4022,13 @@ def parseKernelLog(data): for s in dm[p]: if(re.match(s, msg)): phasechange, phase = True, p + dm[p] = [s] break # hack for determining resume_machine end for freeze if(not sysvals.usetraceevents and sysvals.suspendmode == 'freeze' \ and phase == 'resume_machine' and \ - re.match('calling (?P<f>.*)\+ @ .*, parent: .*', msg)): + data.initcall_debug_call(line, True)): data.setPhase(phase, ktime, False) phase = 'resume_noirq' data.setPhase(phase, ktime, True) @@ -3945,26 +4101,18 @@ def parseKernelLog(data): # -- device callbacks -- if(phase in data.sortedPhases()): # device init call - if(re.match('calling (?P<f>.*)\+ @ .*, parent: .*', msg)): - sm = re.match('calling (?P<f>.*)\+ @ '+\ - '(?P<n>.*), parent: (?P<p>.*)', msg); - f = sm.group('f') - n = sm.group('n') - p = sm.group('p') - if(f and n and p): - data.newAction(phase, f, int(n), p, ktime, -1, '') - # device init return - elif(re.match('call (?P<f>.*)\+ returned .* after '+\ - '(?P<t>.*) usecs', msg)): - sm = re.match('call (?P<f>.*)\+ returned .* after '+\ - '(?P<t>.*) usecs(?P<a>.*)', msg); - f = sm.group('f') - t = sm.group('t') - list = data.dmesg[phase]['list'] - if(f in list): - dev = list[f] - dev['length'] = int(t) - dev['end'] = ktime + t, f, n, p = data.initcall_debug_call(line) + if t and f and n and p: + data.newAction(phase, f, int(n), p, ktime, -1, '') + else: + # device init return + t, f, l = data.initcall_debug_return(line) + if t and f and l: + list = data.dmesg[phase]['list'] + if(f in list): + dev = list[f] + dev['length'] = int(l) + dev['end'] = ktime # if trace events are not available, these are better than nothing if(not sysvals.usetraceevents): @@ -4006,6 +4154,8 @@ def parseKernelLog(data): # fill in any missing phases phasedef = data.phasedef terr, lp = '', 'suspend_prepare' + if lp not in data.dmesg: + doError('dmesg log format has changed, could not find start of suspend') for p in sorted(phasedef, key=lambda k:phasedef[k]['order']): if p not in data.dmesg: if not terr: @@ -4061,6 +4211,8 @@ def callgraphHTML(sv, hf, num, cg, title, color, devid): fmt = '<n>(%.3f ms @ '+sv.timeformat+')</n>' flen = fmt % (line.length*1000, line.time) if line.isLeaf(): + if line.length * 1000 < sv.mincglen: + continue hf.write(html_func_leaf.format(line.name, flen)) elif line.freturn: hf.write(html_func_end) @@ -4675,14 +4827,11 @@ def createHTML(testruns, testfail): if('cpuexec' in dev): for t in sorted(dev['cpuexec']): start, end = t - j = float(dev['cpuexec'][t]) / 5 - if j > 1.0: - j = 1.0 height = '%.3f' % (rowheight/3) top = '%.3f' % (rowtop + devtl.scaleH + 2*rowheight/3) left = '%f' % (((start-m0)*100)/mTotal) width = '%f' % ((end-start)*100/mTotal) - color = 'rgba(255, 0, 0, %f)' % j + color = 'rgba(255, 0, 0, %f)' % dev['cpuexec'][t] devtl.html += \ html_cpuexec.format(left, top, height, width, color) if('src' not in dev): @@ -5301,17 +5450,9 @@ def executeSuspend(quiet=False): call('sync', shell=True) sv.dlog('read dmesg') sv.initdmesg() - # start ftrace - if(sv.usecallgraph or sv.usetraceevents): - if not quiet: - pprint('START TRACING') - sv.dlog('start ftrace tracing') - sv.fsetVal('1', 'tracing_on') - if sv.useprocmon: - sv.dlog('start the process monitor') - pm.start() - sv.dlog('run the cmdinfo list before') + sv.dlog('cmdinfo before') sv.cmdinfo(True) + sv.start(pm) # execute however many s/r runs requested for count in range(1,sv.execcount+1): # x2delay in between test runs @@ -5334,8 +5475,7 @@ def executeSuspend(quiet=False): sv.dlog('enable RTC wake alarm') sv.rtcWakeAlarmOn() # start of suspend trace marker - if(sv.usecallgraph or sv.usetraceevents): - sv.fsetVal(datetime.now().strftime(sv.tmstart), 'trace_marker') + sv.fsetVal(datetime.now().strftime(sv.tmstart), 'trace_marker') # predelay delay if(count == 1 and sv.predelay > 0): sv.fsetVal('WAIT %d' % sv.predelay, 'trace_marker') @@ -5349,6 +5489,7 @@ def executeSuspend(quiet=False): if res != 0: tdata['error'] = 'cmd returned %d' % res else: + s0ixready = sv.s0ixSupport() mode = sv.suspendmode if sv.memmode and os.path.exists(sv.mempowerfile): mode = 'mem' @@ -5358,9 +5499,10 @@ def executeSuspend(quiet=False): sv.testVal(sv.diskpowerfile, 'radio', sv.diskmode) if sv.acpidebug: sv.testVal(sv.acpipath, 'acpi', '0xe') - if mode == 'freeze' and sv.haveTurbostat(): + if ((mode == 'freeze') or (sv.memmode == 's2idle')) \ + and sv.haveTurbostat(): # execution will pause here - turbo = sv.turbostat() + turbo = sv.turbostat(s0ixready) if turbo: tdata['turbo'] = turbo else: @@ -5371,7 +5513,8 @@ def executeSuspend(quiet=False): pf.close() except Exception as e: tdata['error'] = str(e) - sv.dlog('system returned from resume') + sv.fsetVal('CMD COMPLETE', 'trace_marker') + sv.dlog('system returned') # reset everything sv.testVal('restoreall') if(sv.rtcwake): @@ -5384,30 +5527,32 @@ def executeSuspend(quiet=False): sv.fsetVal('WAIT END', 'trace_marker') # return from suspend pprint('RESUME COMPLETE') - if(sv.usecallgraph or sv.usetraceevents): + if(count < sv.execcount): sv.fsetVal(datetime.now().strftime(sv.tmend), 'trace_marker') + elif(not sv.wifitrace): + sv.fsetVal(datetime.now().strftime(sv.tmend), 'trace_marker') + sv.stop(pm) if sv.wifi and wifi: tdata['wifi'] = sv.pollWifi(wifi) sv.dlog('wifi check, %s' % tdata['wifi']) + if(count == sv.execcount and sv.wifitrace): + sv.fsetVal(datetime.now().strftime(sv.tmend), 'trace_marker') + sv.stop(pm) + if sv.netfix: + tdata['netfix'] = sv.netfixon() + sv.dlog('netfix, %s' % tdata['netfix']) if(sv.suspendmode == 'mem' or sv.suspendmode == 'command'): sv.dlog('read the ACPI FPDT') tdata['fw'] = getFPDT(False) testdata.append(tdata) - sv.dlog('run the cmdinfo list after') + sv.dlog('cmdinfo after') cmdafter = sv.cmdinfo(False) - # stop ftrace - if(sv.usecallgraph or sv.usetraceevents): - if sv.useprocmon: - sv.dlog('stop the process monitor') - pm.stop() - sv.fsetVal('0', 'tracing_on') # grab a copy of the dmesg output if not quiet: pprint('CAPTURING DMESG') - sysvals.dlog('EXECUTION TRACE END') sv.getdmesg(testdata) # grab a copy of the ftrace output - if(sv.usecallgraph or sv.usetraceevents): + if sv.useftrace: if not quiet: pprint('CAPTURING TRACE') op = sv.writeDatafileHeader(sv.ftracefile, testdata) @@ -5838,13 +5983,19 @@ def statusCheck(probecheck=False): pprint(' please choose one with -m') # check if ftrace is available - res = sysvals.colorText('NO') - ftgood = sysvals.verifyFtrace() - if(ftgood): - res = 'YES' - elif(sysvals.usecallgraph): - status = 'ftrace is not properly supported' - pprint(' is ftrace supported: %s' % res) + if sysvals.useftrace: + res = sysvals.colorText('NO') + sysvals.useftrace = sysvals.verifyFtrace() + efmt = '"{0}" uses ftrace, and it is not properly supported' + if sysvals.useftrace: + res = 'YES' + elif sysvals.usecallgraph: + status = efmt.format('-f') + elif sysvals.usedevsrc: + status = efmt.format('-dev') + elif sysvals.useprocmon: + status = efmt.format('-proc') + pprint(' is ftrace supported: %s' % res) # check if kprobes are available if sysvals.usekprobes: @@ -5857,8 +6008,8 @@ def statusCheck(probecheck=False): pprint(' are kprobes supported: %s' % res) # what data source are we using - res = 'DMESG' - if(ftgood): + res = 'DMESG (very limited, ftrace is preferred)' + if sysvals.useftrace: sysvals.usetraceevents = True for e in sysvals.traceevents: if not os.path.exists(sysvals.epath+e): @@ -5879,7 +6030,7 @@ def statusCheck(probecheck=False): pprint(' optional commands this tool may use for info:') no = sysvals.colorText('MISSING') yes = sysvals.colorText('FOUND', 32) - for c in ['turbostat', 'mcelog', 'lspci', 'lsusb']: + for c in ['turbostat', 'mcelog', 'lspci', 'lsusb', 'netfix']: if c == 'turbostat': res = yes if sysvals.haveTurbostat() else no else: @@ -5971,7 +6122,7 @@ def processData(live=False, quiet=False): if not sysvals.stamp: pprint('ERROR: data does not include the expected stamp') return (testruns, {'error': 'timeline generation failed'}) - shown = ['bios', 'biosdate', 'cpu', 'host', 'kernel', 'man', 'memfr', + shown = ['os', 'bios', 'biosdate', 'cpu', 'host', 'kernel', 'man', 'memfr', 'memsz', 'mode', 'numcpu', 'plat', 'time', 'wifi'] sysvals.vprint('System Info:') for key in sorted(sysvals.stamp): @@ -6052,6 +6203,8 @@ def runTest(n=0, quiet=False): if sysvals.display: ret = sysvals.displayControl('init') sysvals.dlog('xset display init, ret = %d' % ret) + sysvals.testVal(sysvals.pmdpath, 'basic', '1') + sysvals.testVal(sysvals.s0ixpath, 'basic', 'Y') sysvals.dlog('initialize ftrace') sysvals.initFtrace(quiet) @@ -6145,9 +6298,12 @@ def data_from_html(file, outpath, issues, fulldetail=False): elist[err[0]] += 1 for i in elist: ilist.append('%sx%d' % (i, elist[i]) if elist[i] > 1 else i) - wifi = find_in_html(html, 'Wifi Resume: ', '</td>') - if wifi: - extra['wifi'] = wifi + line = find_in_html(log, '# wifi ', '\n') + if line: + extra['wifi'] = line + line = find_in_html(log, '# netfix ', '\n') + if line: + extra['netfix'] = line low = find_in_html(html, 'freeze time: <b>', ' ms</b>') for lowstr in ['waking', '+']: if not low: @@ -6182,6 +6338,8 @@ def data_from_html(file, outpath, issues, fulldetail=False): if not m: continue name, time, phase = m.group('n'), m.group('t'), m.group('p') + if name == 'async_synchronize_full': + continue if ' async' in name or ' sync' in name: name = ' '.join(name.split(' ')[:-1]) if phase.startswith('suspend'): @@ -6243,7 +6401,7 @@ def genHtml(subdir, force=False): sysvals.ftracefile = file sysvals.setOutputFile() if (sysvals.dmesgfile or sysvals.ftracefile) and sysvals.htmlfile and \ - (force or not sysvals.usable(sysvals.htmlfile)): + (force or not sysvals.usable(sysvals.htmlfile, True)): pprint('FTRACE: %s' % sysvals.ftracefile) if sysvals.dmesgfile: pprint('DMESG : %s' % sysvals.dmesgfile) @@ -6533,6 +6691,8 @@ def printHelp(): ' -skiphtml Run the test and capture the trace logs, but skip the timeline (default: disabled)\n'\ ' -result fn Export a results table to a text file for parsing.\n'\ ' -wifi If a wifi connection is available, check that it reconnects after resume.\n'\ + ' -wifitrace Trace kernel execution through wifi reconnect.\n'\ + ' -netfix Use netfix to reset the network in the event it fails to resume.\n'\ ' [testprep]\n'\ ' -sync Sync the filesystems before starting the test\n'\ ' -rs on/off Enable/disable runtime suspend for all devices, restore all after test\n'\ @@ -6615,6 +6775,8 @@ if __name__ == '__main__': elif(arg == '-v'): pprint("Version %s" % sysvals.version) sys.exit(0) + elif(arg == '-debugtiming'): + debugtiming = True elif(arg == '-x2'): sysvals.execcount = 2 elif(arg == '-x2delay'): @@ -6657,6 +6819,10 @@ if __name__ == '__main__': sysvals.sync = True elif(arg == '-wifi'): sysvals.wifi = True + elif(arg == '-wifitrace'): + sysvals.wifitrace = True + elif(arg == '-netfix'): + sysvals.netfix = True elif(arg == '-gzip'): sysvals.gzip = True elif(arg == '-info'): @@ -6819,7 +6985,7 @@ if __name__ == '__main__': sysvals.outdir = val sysvals.notestrun = True if(os.path.isdir(val) == False): - doError('%s is not accessible' % val) + doError('%s is not accesible' % val) elif(arg == '-filter'): try: val = next(args) @@ -6942,12 +7108,11 @@ if __name__ == '__main__': time.sleep(sysvals.multitest['delay']) fmt = 'suspend-%y%m%d-%H%M%S' sysvals.testdir = os.path.join(sysvals.outdir, datetime.now().strftime(fmt)) - ret = runTest(i+1, True) + ret = runTest(i+1, not sysvals.verbose) failcnt = 0 if not ret else failcnt + 1 if sysvals.maxfail > 0 and failcnt >= sysvals.maxfail: pprint('Maximum fail count of %d reached, aborting multitest' % (sysvals.maxfail)) break - time.sleep(5) sysvals.resetlog() sysvals.multistat(False, i, finish) if 'time' in sysvals.multitest and datetime.now() >= finish: diff --git a/tools/power/x86/amd_pstate_tracer/amd_pstate_trace.py b/tools/power/x86/amd_pstate_tracer/amd_pstate_trace.py new file mode 100755 index 000000000000..2dea4032ac56 --- /dev/null +++ b/tools/power/x86/amd_pstate_tracer/amd_pstate_trace.py @@ -0,0 +1,354 @@ +#!/usr/bin/env python3 +# SPDX-License-Identifier: GPL-2.0-only +# -*- coding: utf-8 -*- +# +""" This utility can be used to debug and tune the performance of the +AMD P-State driver. It imports intel_pstate_tracer to analyze AMD P-State +trace event. + +Prerequisites: + Python version 2.7.x or higher + gnuplot 5.0 or higher + gnuplot-py 1.8 or higher + (Most of the distributions have these required packages. They may be called + gnuplot-py, phython-gnuplot or phython3-gnuplot, gnuplot-nox, ... ) + + Kernel config for Linux trace is enabled + + see print_help(): for Usage and Output details + +""" +from __future__ import print_function +from datetime import datetime +import subprocess +import os +import time +import re +import signal +import sys +import getopt +import Gnuplot +from numpy import * +from decimal import * +sys.path.append('../intel_pstate_tracer') +#import intel_pstate_tracer +import intel_pstate_tracer as ipt + +__license__ = "GPL version 2" + +MAX_CPUS = 256 +# Define the csv file columns +C_COMM = 15 +C_ELAPSED = 14 +C_SAMPLE = 13 +C_DURATION = 12 +C_LOAD = 11 +C_TSC = 10 +C_APERF = 9 +C_MPERF = 8 +C_FREQ = 7 +C_MAX_PERF = 6 +C_DES_PERF = 5 +C_MIN_PERF = 4 +C_USEC = 3 +C_SEC = 2 +C_CPU = 1 + +global sample_num, last_sec_cpu, last_usec_cpu, start_time, test_name, trace_file + +getcontext().prec = 11 + +sample_num =0 +last_sec_cpu = [0] * MAX_CPUS +last_usec_cpu = [0] * MAX_CPUS + +def plot_per_cpu_freq(cpu_index): + """ Plot per cpu frequency """ + + file_name = 'cpu{:0>3}.csv'.format(cpu_index) + if os.path.exists(file_name): + output_png = "cpu%03d_frequency.png" % cpu_index + g_plot = ipt.common_gnuplot_settings() + g_plot('set output "' + output_png + '"') + g_plot('set yrange [0:7]') + g_plot('set ytics 0, 1') + g_plot('set ylabel "CPU Frequency (GHz)"') + g_plot('set title "{} : frequency : CPU {:0>3} : {:%F %H:%M}"'.format(test_name, cpu_index, datetime.now())) + g_plot('set ylabel "CPU frequency"') + g_plot('set key off') + ipt.set_4_plot_linestyles(g_plot) + g_plot('plot "' + file_name + '" using {:d}:{:d} with linespoints linestyle 1 axis x1y1'.format(C_ELAPSED, C_FREQ)) + +def plot_per_cpu_des_perf(cpu_index): + """ Plot per cpu desired perf """ + + file_name = 'cpu{:0>3}.csv'.format(cpu_index) + if os.path.exists(file_name): + output_png = "cpu%03d_des_perf.png" % cpu_index + g_plot = ipt.common_gnuplot_settings() + g_plot('set output "' + output_png + '"') + g_plot('set yrange [0:255]') + g_plot('set ylabel "des perf"') + g_plot('set title "{} : cpu des perf : CPU {:0>3} : {:%F %H:%M}"'.format(test_name, cpu_index, datetime.now())) + g_plot('set key off') + ipt.set_4_plot_linestyles(g_plot) + g_plot('plot "' + file_name + '" using {:d}:{:d} with linespoints linestyle 1 axis x1y1'.format(C_ELAPSED, C_DES_PERF)) + +def plot_per_cpu_load(cpu_index): + """ Plot per cpu load """ + + file_name = 'cpu{:0>3}.csv'.format(cpu_index) + if os.path.exists(file_name): + output_png = "cpu%03d_load.png" % cpu_index + g_plot = ipt.common_gnuplot_settings() + g_plot('set output "' + output_png + '"') + g_plot('set yrange [0:100]') + g_plot('set ytics 0, 10') + g_plot('set ylabel "CPU load (percent)"') + g_plot('set title "{} : cpu load : CPU {:0>3} : {:%F %H:%M}"'.format(test_name, cpu_index, datetime.now())) + g_plot('set key off') + ipt.set_4_plot_linestyles(g_plot) + g_plot('plot "' + file_name + '" using {:d}:{:d} with linespoints linestyle 1 axis x1y1'.format(C_ELAPSED, C_LOAD)) + +def plot_all_cpu_frequency(): + """ Plot all cpu frequencies """ + + output_png = 'all_cpu_frequencies.png' + g_plot = ipt.common_gnuplot_settings() + g_plot('set output "' + output_png + '"') + g_plot('set ylabel "CPU Frequency (GHz)"') + g_plot('set title "{} : cpu frequencies : {:%F %H:%M}"'.format(test_name, datetime.now())) + + title_list = subprocess.check_output('ls cpu???.csv | sed -e \'s/.csv//\'',shell=True).decode('utf-8').replace('\n', ' ') + plot_str = "plot for [i in title_list] i.'.csv' using {:d}:{:d} pt 7 ps 1 title i".format(C_ELAPSED, C_FREQ) + g_plot('title_list = "{}"'.format(title_list)) + g_plot(plot_str) + +def plot_all_cpu_des_perf(): + """ Plot all cpu desired perf """ + + output_png = 'all_cpu_des_perf.png' + g_plot = ipt.common_gnuplot_settings() + g_plot('set output "' + output_png + '"') + g_plot('set ylabel "des perf"') + g_plot('set title "{} : cpu des perf : {:%F %H:%M}"'.format(test_name, datetime.now())) + + title_list = subprocess.check_output('ls cpu???.csv | sed -e \'s/.csv//\'',shell=True).decode('utf-8').replace('\n', ' ') + plot_str = "plot for [i in title_list] i.'.csv' using {:d}:{:d} pt 255 ps 1 title i".format(C_ELAPSED, C_DES_PERF) + g_plot('title_list = "{}"'.format(title_list)) + g_plot(plot_str) + +def plot_all_cpu_load(): + """ Plot all cpu load """ + + output_png = 'all_cpu_load.png' + g_plot = ipt.common_gnuplot_settings() + g_plot('set output "' + output_png + '"') + g_plot('set yrange [0:100]') + g_plot('set ylabel "CPU load (percent)"') + g_plot('set title "{} : cpu load : {:%F %H:%M}"'.format(test_name, datetime.now())) + + title_list = subprocess.check_output('ls cpu???.csv | sed -e \'s/.csv//\'',shell=True).decode('utf-8').replace('\n', ' ') + plot_str = "plot for [i in title_list] i.'.csv' using {:d}:{:d} pt 255 ps 1 title i".format(C_ELAPSED, C_LOAD) + g_plot('title_list = "{}"'.format(title_list)) + g_plot(plot_str) + +def store_csv(cpu_int, time_pre_dec, time_post_dec, min_perf, des_perf, max_perf, freq_ghz, mperf, aperf, tsc, common_comm, load, duration_ms, sample_num, elapsed_time, cpu_mask): + """ Store master csv file information """ + + global graph_data_present + + if cpu_mask[cpu_int] == 0: + return + + try: + f_handle = open('cpu.csv', 'a') + string_buffer = "CPU_%03u, %05u, %06u, %u, %u, %u, %.4f, %u, %u, %u, %.2f, %.3f, %u, %.3f, %s\n" % (cpu_int, int(time_pre_dec), int(time_post_dec), int(min_perf), int(des_perf), int(max_perf), freq_ghz, int(mperf), int(aperf), int(tsc), load, duration_ms, sample_num, elapsed_time, common_comm) + f_handle.write(string_buffer) + f_handle.close() + except: + print('IO error cpu.csv') + return + + graph_data_present = True; + + +def cleanup_data_files(): + """ clean up existing data files """ + + if os.path.exists('cpu.csv'): + os.remove('cpu.csv') + f_handle = open('cpu.csv', 'a') + f_handle.write('common_cpu, common_secs, common_usecs, min_perf, des_perf, max_perf, freq, mperf, aperf, tsc, load, duration_ms, sample_num, elapsed_time, common_comm') + f_handle.write('\n') + f_handle.close() + +def read_trace_data(file_name, cpu_mask): + """ Read and parse trace data """ + + global current_max_cpu + global sample_num, last_sec_cpu, last_usec_cpu, start_time + + try: + data = open(file_name, 'r').read() + except: + print('Error opening ', file_name) + sys.exit(2) + + for line in data.splitlines(): + search_obj = \ + re.search(r'(^(.*?)\[)((\d+)[^\]])(.*?)(\d+)([.])(\d+)(.*?amd_min_perf=)(\d+)(.*?amd_des_perf=)(\d+)(.*?amd_max_perf=)(\d+)(.*?freq=)(\d+)(.*?mperf=)(\d+)(.*?aperf=)(\d+)(.*?tsc=)(\d+)' + , line) + + if search_obj: + cpu = search_obj.group(3) + cpu_int = int(cpu) + cpu = str(cpu_int) + + time_pre_dec = search_obj.group(6) + time_post_dec = search_obj.group(8) + min_perf = search_obj.group(10) + des_perf = search_obj.group(12) + max_perf = search_obj.group(14) + freq = search_obj.group(16) + mperf = search_obj.group(18) + aperf = search_obj.group(20) + tsc = search_obj.group(22) + + common_comm = search_obj.group(2).replace(' ', '') + + if sample_num == 0 : + start_time = Decimal(time_pre_dec) + Decimal(time_post_dec) / Decimal(1000000) + sample_num += 1 + + if last_sec_cpu[cpu_int] == 0 : + last_sec_cpu[cpu_int] = time_pre_dec + last_usec_cpu[cpu_int] = time_post_dec + else : + duration_us = (int(time_pre_dec) - int(last_sec_cpu[cpu_int])) * 1000000 + (int(time_post_dec) - int(last_usec_cpu[cpu_int])) + duration_ms = Decimal(duration_us) / Decimal(1000) + last_sec_cpu[cpu_int] = time_pre_dec + last_usec_cpu[cpu_int] = time_post_dec + elapsed_time = Decimal(time_pre_dec) + Decimal(time_post_dec) / Decimal(1000000) - start_time + load = Decimal(int(mperf)*100)/ Decimal(tsc) + freq_ghz = Decimal(freq)/Decimal(1000000) + store_csv(cpu_int, time_pre_dec, time_post_dec, min_perf, des_perf, max_perf, freq_ghz, mperf, aperf, tsc, common_comm, load, duration_ms, sample_num, elapsed_time, cpu_mask) + + if cpu_int > current_max_cpu: + current_max_cpu = cpu_int +# Now separate the main overall csv file into per CPU csv files. + ipt.split_csv(current_max_cpu, cpu_mask) + + +def signal_handler(signal, frame): + print(' SIGINT: Forcing cleanup before exit.') + if interval: + ipt.disable_trace(trace_file) + ipt.clear_trace_file() + ipt.free_trace_buffer() + sys.exit(0) + +trace_file = "/sys/kernel/debug/tracing/events/amd_cpu/enable" +signal.signal(signal.SIGINT, signal_handler) + +interval = "" +file_name = "" +cpu_list = "" +test_name = "" +memory = "10240" +graph_data_present = False; + +valid1 = False +valid2 = False + +cpu_mask = zeros((MAX_CPUS,), dtype=int) + + +try: + opts, args = getopt.getopt(sys.argv[1:],"ht:i:c:n:m:",["help","trace_file=","interval=","cpu=","name=","memory="]) +except getopt.GetoptError: + ipt.print_help('amd_pstate') + sys.exit(2) +for opt, arg in opts: + if opt == '-h': + print() + sys.exit() + elif opt in ("-t", "--trace_file"): + valid1 = True + location = os.path.realpath(os.path.join(os.getcwd(), os.path.dirname(__file__))) + file_name = os.path.join(location, arg) + elif opt in ("-i", "--interval"): + valid1 = True + interval = arg + elif opt in ("-c", "--cpu"): + cpu_list = arg + elif opt in ("-n", "--name"): + valid2 = True + test_name = arg + elif opt in ("-m", "--memory"): + memory = arg + +if not (valid1 and valid2): + ipt.print_help('amd_pstate') + sys.exit() + +if cpu_list: + for p in re.split("[,]", cpu_list): + if int(p) < MAX_CPUS : + cpu_mask[int(p)] = 1 +else: + for i in range (0, MAX_CPUS): + cpu_mask[i] = 1 + +if not os.path.exists('results'): + os.mkdir('results') + ipt.fix_ownership('results') + +os.chdir('results') +if os.path.exists(test_name): + print('The test name directory already exists. Please provide a unique test name. Test re-run not supported, yet.') + sys.exit() +os.mkdir(test_name) +ipt.fix_ownership(test_name) +os.chdir(test_name) + +cur_version = sys.version_info +print('python version (should be >= 2.7):') +print(cur_version) + +cleanup_data_files() + +if interval: + file_name = "/sys/kernel/debug/tracing/trace" + ipt.clear_trace_file() + ipt.set_trace_buffer_size(memory) + ipt.enable_trace(trace_file) + time.sleep(int(interval)) + ipt.disable_trace(trace_file) + +current_max_cpu = 0 + +read_trace_data(file_name, cpu_mask) + +if interval: + ipt.clear_trace_file() + ipt.free_trace_buffer() + +if graph_data_present == False: + print('No valid data to plot') + sys.exit(2) + +for cpu_no in range(0, current_max_cpu + 1): + plot_per_cpu_freq(cpu_no) + plot_per_cpu_des_perf(cpu_no) + plot_per_cpu_load(cpu_no) + +plot_all_cpu_des_perf() +plot_all_cpu_frequency() +plot_all_cpu_load() + +for root, dirs, files in os.walk('.'): + for f in files: + ipt.fix_ownership(f) + +os.chdir('../../') diff --git a/tools/power/x86/intel-speed-select/Build b/tools/power/x86/intel-speed-select/Build index b61456d75190..81e36bd578b1 100644 --- a/tools/power/x86/intel-speed-select/Build +++ b/tools/power/x86/intel-speed-select/Build @@ -1 +1 @@ -intel-speed-select-y += isst-config.o isst-core.o isst-display.o +intel-speed-select-y += isst-config.o isst-core.o isst-display.o isst-daemon.o hfi-events.o diff --git a/tools/power/x86/intel-speed-select/Makefile b/tools/power/x86/intel-speed-select/Makefile index 12c6939dca2a..7221f2f55e8b 100644 --- a/tools/power/x86/intel-speed-select/Makefile +++ b/tools/power/x86/intel-speed-select/Makefile @@ -13,8 +13,8 @@ endif # Do not use make's built-in rules # (this improves performance and avoids hard-to-debug behaviour); MAKEFLAGS += -r - -override CFLAGS += -O2 -Wall -g -D_GNU_SOURCE -I$(OUTPUT)include +override CFLAGS += -O2 -Wall -g -D_GNU_SOURCE -I$(OUTPUT)include -I/usr/include/libnl3 +override LDFLAGS += -lnl-genl-3 -lnl-3 ALL_TARGETS := intel-speed-select ALL_PROGRAMS := $(patsubst %,$(OUTPUT)%,$(ALL_TARGETS)) @@ -31,19 +31,23 @@ $(OUTPUT)include/linux/isst_if.h: ../../../../include/uapi/linux/isst_if.h mkdir -p $(OUTPUT)include/linux 2>&1 || true ln -sf $(CURDIR)/../../../../include/uapi/linux/isst_if.h $@ -prepare: $(OUTPUT)include/linux/isst_if.h +$(OUTPUT)include/linux/thermal.h: ../../../../include/uapi/linux/thermal.h + mkdir -p $(OUTPUT)include/linux 2>&1 || true + ln -sf $(CURDIR)/../../../../include/uapi/linux/thermal.h $@ + +prepare: $(OUTPUT)include/linux/isst_if.h $(OUTPUT)include/linux/thermal.h ISST_IN := $(OUTPUT)intel-speed-select-in.o $(ISST_IN): prepare FORCE $(Q)$(MAKE) $(build)=intel-speed-select $(OUTPUT)intel-speed-select: $(ISST_IN) - $(QUIET_LINK)$(CC) $(CFLAGS) $(LDFLAGS) $< -o $@ + $(QUIET_LINK)$(CC) $(CFLAGS) $< $(LDFLAGS) -o $@ clean: rm -f $(ALL_PROGRAMS) rm -rf $(OUTPUT)include/linux/isst_if.h - find $(if $(OUTPUT),$(OUTPUT),.) -name '*.o' -delete -o -name '\.*.d' -delete + find $(or $(OUTPUT),.) -name '*.o' -delete -o -name '\.*.d' -delete install: $(ALL_PROGRAMS) install -d -m 755 $(DESTDIR)$(bindir); \ diff --git a/tools/power/x86/intel-speed-select/hfi-events.c b/tools/power/x86/intel-speed-select/hfi-events.c new file mode 100644 index 000000000000..be96e90cc2a1 --- /dev/null +++ b/tools/power/x86/intel-speed-select/hfi-events.c @@ -0,0 +1,312 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Intel Speed Select -- Read HFI events for OOB + * Copyright (c) 2022 Intel Corporation. + */ + +/* + * This file incorporates work covered by the following copyright and + * permission notice: + + * WPA Supplicant - driver interaction with Linux nl80211/cfg80211 + * Copyright (c) 2003-2008, Jouni Malinen <j@w1.fi> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * Alternatively, this software may be distributed under the terms of + * BSD license. + * + * Requires + * libnl-genl-3-dev + * + * For Fedora/CenOS + * dnf install libnl3-devel + * For Ubuntu + * apt install libnl-3-dev libnl-genl-3-dev + */ + +#include <stdio.h> +#include <stdlib.h> +#include <stdarg.h> +#include <string.h> +#include <unistd.h> +#include <fcntl.h> +#include <sys/file.h> +#include <sys/types.h> +#include <sys/stat.h> +#include <errno.h> +#include <getopt.h> +#include <signal.h> +#include <netlink/genl/genl.h> +#include <netlink/genl/family.h> +#include <netlink/genl/ctrl.h> + +#include <linux/thermal.h> +#include "isst.h" + +struct hfi_event_data { + struct nl_sock *nl_handle; + struct nl_cb *nl_cb; +}; + +struct hfi_event_data drv; + +static int ack_handler(struct nl_msg *msg, void *arg) +{ + int *err = arg; + *err = 0; + return NL_STOP; +} + +static int finish_handler(struct nl_msg *msg, void *arg) +{ + int *ret = arg; + *ret = 0; + return NL_SKIP; +} + +static int error_handler(struct sockaddr_nl *nla, struct nlmsgerr *err, + void *arg) +{ + int *ret = arg; + *ret = err->error; + return NL_SKIP; +} + +static int seq_check_handler(struct nl_msg *msg, void *arg) +{ + return NL_OK; +} + +static int send_and_recv_msgs(struct hfi_event_data *drv, + struct nl_msg *msg, + int (*valid_handler)(struct nl_msg *, void *), + void *valid_data) +{ + struct nl_cb *cb; + int err = -ENOMEM; + + cb = nl_cb_clone(drv->nl_cb); + if (!cb) + goto out; + + err = nl_send_auto_complete(drv->nl_handle, msg); + if (err < 0) + goto out; + + err = 1; + + nl_cb_err(cb, NL_CB_CUSTOM, error_handler, &err); + nl_cb_set(cb, NL_CB_FINISH, NL_CB_CUSTOM, finish_handler, &err); + nl_cb_set(cb, NL_CB_ACK, NL_CB_CUSTOM, ack_handler, &err); + + if (valid_handler) + nl_cb_set(cb, NL_CB_VALID, NL_CB_CUSTOM, + valid_handler, valid_data); + + while (err > 0) + nl_recvmsgs(drv->nl_handle, cb); + out: + nl_cb_put(cb); + nlmsg_free(msg); + return err; +} + +struct family_data { + const char *group; + int id; +}; + +static int family_handler(struct nl_msg *msg, void *arg) +{ + struct family_data *res = arg; + struct nlattr *tb[CTRL_ATTR_MAX + 1]; + struct genlmsghdr *gnlh = nlmsg_data(nlmsg_hdr(msg)); + struct nlattr *mcgrp; + int i; + + nla_parse(tb, CTRL_ATTR_MAX, genlmsg_attrdata(gnlh, 0), + genlmsg_attrlen(gnlh, 0), NULL); + if (!tb[CTRL_ATTR_MCAST_GROUPS]) + return NL_SKIP; + + nla_for_each_nested(mcgrp, tb[CTRL_ATTR_MCAST_GROUPS], i) { + struct nlattr *tb2[CTRL_ATTR_MCAST_GRP_MAX + 1]; + nla_parse(tb2, CTRL_ATTR_MCAST_GRP_MAX, nla_data(mcgrp), + nla_len(mcgrp), NULL); + if (!tb2[CTRL_ATTR_MCAST_GRP_NAME] || + !tb2[CTRL_ATTR_MCAST_GRP_ID] || + strncmp(nla_data(tb2[CTRL_ATTR_MCAST_GRP_NAME]), + res->group, + nla_len(tb2[CTRL_ATTR_MCAST_GRP_NAME])) != 0) + continue; + res->id = nla_get_u32(tb2[CTRL_ATTR_MCAST_GRP_ID]); + break; + } + + return 0; +} + +static int nl_get_multicast_id(struct hfi_event_data *drv, + const char *family, const char *group) +{ + struct nl_msg *msg; + int ret = -1; + struct family_data res = { group, -ENOENT }; + + msg = nlmsg_alloc(); + if (!msg) + return -ENOMEM; + genlmsg_put(msg, 0, 0, genl_ctrl_resolve(drv->nl_handle, "nlctrl"), + 0, 0, CTRL_CMD_GETFAMILY, 0); + NLA_PUT_STRING(msg, CTRL_ATTR_FAMILY_NAME, family); + + ret = send_and_recv_msgs(drv, msg, family_handler, &res); + msg = NULL; + if (ret == 0) + ret = res.id; + +nla_put_failure: + nlmsg_free(msg); + return ret; +} + +struct perf_cap { + int cpu; + int perf; + int eff; +}; + +static void process_hfi_event(struct perf_cap *perf_cap) +{ + struct isst_id id; + + set_isst_id(&id, perf_cap->cpu); + process_level_change(&id); +} + +static int handle_event(struct nl_msg *n, void *arg) +{ + struct nlmsghdr *nlh = nlmsg_hdr(n); + struct genlmsghdr *genlhdr = genlmsg_hdr(nlh); + struct nlattr *attrs[THERMAL_GENL_ATTR_MAX + 1]; + int ret; + struct perf_cap perf_cap = {0}; + + ret = genlmsg_parse(nlh, 0, attrs, THERMAL_GENL_ATTR_MAX, NULL); + + debug_printf("Received event %d parse_rer:%d\n", genlhdr->cmd, ret); + if (genlhdr->cmd == THERMAL_GENL_EVENT_CPU_CAPABILITY_CHANGE) { + struct nlattr *cap; + int j, index = 0; + + debug_printf("THERMAL_GENL_EVENT_CPU_CAPABILITY_CHANGE\n"); + nla_for_each_nested(cap, attrs[THERMAL_GENL_ATTR_CPU_CAPABILITY], j) { + switch (index) { + case 0: + perf_cap.cpu = nla_get_u32(cap); + break; + case 1: + perf_cap.perf = nla_get_u32(cap); + break; + case 2: + perf_cap.eff = nla_get_u32(cap); + break; + default: + break; + } + ++index; + if (index == 3) { + index = 0; + process_hfi_event(&perf_cap); + } + } + } + + return 0; +} + +static int _hfi_exit; + +static int check_hf_suport(void) +{ + unsigned int eax = 0, ebx = 0, ecx = 0, edx = 0; + + __cpuid(6, eax, ebx, ecx, edx); + if (eax & BIT(19)) + return 1; + + return 0; +} + +int hfi_main(void) +{ + struct nl_sock *sock; + struct nl_cb *cb; + int err = 0; + int mcast_id; + int no_block = 0; + + if (!check_hf_suport()) { + fprintf(stderr, "CPU Doesn't support HFI\n"); + return -1; + } + + sock = nl_socket_alloc(); + if (!sock) { + fprintf(stderr, "nl_socket_alloc failed\n"); + return -1; + } + + if (genl_connect(sock)) { + fprintf(stderr, "genl_connect(sk_event) failed\n"); + goto free_sock; + } + + drv.nl_handle = sock; + drv.nl_cb = cb = nl_cb_alloc(NL_CB_DEFAULT); + if (drv.nl_cb == NULL) { + printf("Failed to allocate netlink callbacks"); + goto free_sock; + } + + mcast_id = nl_get_multicast_id(&drv, THERMAL_GENL_FAMILY_NAME, + THERMAL_GENL_EVENT_GROUP_NAME); + if (mcast_id < 0) { + fprintf(stderr, "nl_get_multicast_id failed\n"); + goto free_sock; + } + + if (nl_socket_add_membership(sock, mcast_id)) { + fprintf(stderr, "nl_socket_add_membership failed"); + goto free_sock; + } + + nl_cb_set(cb, NL_CB_SEQ_CHECK, NL_CB_CUSTOM, seq_check_handler, 0); + nl_cb_set(cb, NL_CB_VALID, NL_CB_CUSTOM, handle_event, NULL); + + if (no_block) + nl_socket_set_nonblocking(sock); + + debug_printf("hfi is initialized\n"); + + while (!_hfi_exit && !err) { + err = nl_recvmsgs(sock, cb); + debug_printf("nl_recv_message err:%d\n", err); + } + + return 0; + + /* Netlink library doesn't have calls to dealloc cb or disconnect */ +free_sock: + nl_socket_free(sock); + + return -1; +} + +void hfi_exit(void) +{ + _hfi_exit = 1; +} diff --git a/tools/power/x86/intel-speed-select/isst-config.c b/tools/power/x86/intel-speed-select/isst-config.c index bf9fd3549a1d..a160bad291eb 100644 --- a/tools/power/x86/intel-speed-select/isst-config.c +++ b/tools/power/x86/intel-speed-select/isst-config.c @@ -15,7 +15,8 @@ struct process_cmd_struct { int arg; }; -static const char *version_str = "v1.10"; +static const char *version_str = "v1.13"; + static const int supported_api_ver = 1; static struct isst_if_platform_info isst_platform_info; static char *progname; @@ -62,6 +63,7 @@ struct _cpu_map { unsigned short die_id; unsigned short punit_cpu; unsigned short punit_cpu_core; + unsigned short initialized; }; struct _cpu_map *cpu_map; @@ -297,10 +299,16 @@ static void store_cpu_topology(void) fclose(fp); } -int get_physical_package_id(int cpu) +static int get_physical_package_id(int cpu) { int ret; + if (cpu < 0) + return -1; + + if (cpu_map && cpu_map[cpu].initialized) + return cpu_map[cpu].pkg_id; + ret = parse_int_file(0, "/sys/devices/system/cpu/cpu%d/topology/physical_package_id", cpu); @@ -315,10 +323,16 @@ int get_physical_package_id(int cpu) return ret; } -int get_physical_core_id(int cpu) +static int get_physical_core_id(int cpu) { int ret; + if (cpu < 0) + return -1; + + if (cpu_map && cpu_map[cpu].initialized) + return cpu_map[cpu].core_id; + ret = parse_int_file(0, "/sys/devices/system/cpu/cpu%d/topology/core_id", cpu); @@ -333,10 +347,16 @@ int get_physical_core_id(int cpu) return ret; } -int get_physical_die_id(int cpu) +static int get_physical_die_id(int cpu) { int ret; + if (cpu < 0) + return -1; + + if (cpu_map && cpu_map[cpu].initialized) + return cpu_map[cpu].die_id; + ret = parse_int_file(0, "/sys/devices/system/cpu/cpu%d/topology/die_id", cpu); @@ -358,6 +378,31 @@ int get_physical_die_id(int cpu) return ret; } +void set_isst_id(struct isst_id *id, int cpu) +{ + id->cpu = cpu; + + id->pkg = get_physical_package_id(cpu); + if (id < 0 || id->pkg >= MAX_PACKAGE_COUNT) + id->pkg = -1; + + id->die = get_physical_die_id(cpu); + if (id < 0 || id->die >= MAX_DIE_PER_PACKAGE) + id->die = -1; +} + +int is_cpu_in_power_domain(int cpu, struct isst_id *id) +{ + struct isst_id tid; + + set_isst_id(&tid, cpu); + + if (id->pkg == tid.pkg && id->die == tid.die) + return 1; + + return 0; +} + int get_cpufreq_base_freq(int cpu) { return parse_int_file(0, "/sys/devices/system/cpu/cpu%d/cpufreq/base_frequency", cpu); @@ -368,7 +413,7 @@ int get_topo_max_cpus(void) return topo_max_cpus; } -static void set_cpu_online_offline(int cpu, int state) +void set_cpu_online_offline(int cpu, int state) { char buffer[128]; int fd, ret; @@ -409,15 +454,14 @@ static void force_all_cpus_online(void) unlink("/var/run/isst_cpu_topology.dat"); } -#define MAX_PACKAGE_COUNT 8 -#define MAX_DIE_PER_PACKAGE 2 -static void for_each_online_package_in_set(void (*callback)(int, void *, void *, - void *, void *), - void *arg1, void *arg2, void *arg3, - void *arg4) +void for_each_online_package_in_set(void (*callback)(struct isst_id *, void *, void *, + void *, void *), + void *arg1, void *arg2, void *arg3, + void *arg4) { int max_packages[MAX_PACKAGE_COUNT * MAX_PACKAGE_COUNT]; int pkg_index = 0, i; + struct isst_id id; memset(max_packages, 0xff, sizeof(max_packages)); for (i = 0; i < topo_max_cpus; ++i) { @@ -451,18 +495,20 @@ static void for_each_online_package_in_set(void (*callback)(int, void *, void *, } } + set_isst_id(&id, i); if (!skip && online && callback) { - callback(i, arg1, arg2, arg3, arg4); + callback(&id, arg1, arg2, arg3, arg4); max_packages[pkg_index++] = pkg_id; } } } static void for_each_online_target_cpu_in_set( - void (*callback)(int, void *, void *, void *, void *), void *arg1, + void (*callback)(struct isst_id *, void *, void *, void *, void *), void *arg1, void *arg2, void *arg3, void *arg4) { int i, found = 0; + struct isst_id id; for (i = 0; i < topo_max_cpus; ++i) { int online; @@ -476,8 +522,9 @@ static void for_each_online_target_cpu_in_set( online = 1; /* online entry for CPU 0 needs some special configs */ + set_isst_id(&id, i); if (online && callback) { - callback(i, arg1, arg2, arg3, arg4); + callback(&id, arg1, arg2, arg3, arg4); found = 1; } } @@ -537,47 +584,8 @@ void free_cpu_set(cpu_set_t *cpu_set) } static int cpu_cnt[MAX_PACKAGE_COUNT][MAX_DIE_PER_PACKAGE]; -static long long core_mask[MAX_PACKAGE_COUNT][MAX_DIE_PER_PACKAGE]; -static void set_cpu_present_cpu_mask(void) -{ - size_t size; - DIR *dir; - int i; - - size = alloc_cpu_set(&present_cpumask); - present_cpumask_size = size; - for (i = 0; i < topo_max_cpus; ++i) { - char buffer[256]; - snprintf(buffer, sizeof(buffer), - "/sys/devices/system/cpu/cpu%d", i); - dir = opendir(buffer); - if (dir) { - int pkg_id, die_id; - - CPU_SET_S(i, size, present_cpumask); - die_id = get_physical_die_id(i); - if (die_id < 0) - die_id = 0; - - pkg_id = get_physical_package_id(i); - if (pkg_id < 0) { - fprintf(stderr, "Failed to get package id, CPU %d may be offline\n", i); - continue; - } - if (pkg_id < MAX_PACKAGE_COUNT && - die_id < MAX_DIE_PER_PACKAGE) { - int core_id = get_physical_core_id(i); - - cpu_cnt[pkg_id][die_id]++; - core_mask[pkg_id][die_id] |= (1ULL << core_id); - } - } - closedir(dir); - } -} - -int get_max_punit_core_id(int pkg_id, int die_id) +int get_max_punit_core_id(struct isst_id *id) { int max_id = 0; int i; @@ -587,60 +595,74 @@ int get_max_punit_core_id(int pkg_id, int die_id) if (!CPU_ISSET_S(i, present_cpumask_size, present_cpumask)) continue; - if (cpu_map[i].pkg_id == pkg_id && - cpu_map[i].die_id == die_id && - cpu_map[i].punit_cpu_core > max_id) + if (is_cpu_in_power_domain(i, id) && + cpu_map[i].punit_cpu_core > max_id) max_id = cpu_map[i].punit_cpu_core; } return max_id; } -int get_cpu_count(int pkg_id, int die_id) +int get_cpu_count(struct isst_id *id) { - if (pkg_id < MAX_PACKAGE_COUNT && die_id < MAX_DIE_PER_PACKAGE) - return cpu_cnt[pkg_id][die_id]; + if (id->pkg < 0 || id->die < 0) + return 0; - return 0; -} - -static void set_cpu_target_cpu_mask(void) -{ - size_t size; - int i; - - size = alloc_cpu_set(&target_cpumask); - target_cpumask_size = size; - for (i = 0; i < max_target_cpus; ++i) { - if (!CPU_ISSET_S(target_cpus[i], present_cpumask_size, - present_cpumask)) - continue; - - CPU_SET_S(target_cpus[i], size, target_cpumask); - } + return cpu_cnt[id->pkg][id->die]; } static void create_cpu_map(void) { const char *pathname = "/dev/isst_interface"; + size_t size; + DIR *dir; int i, fd = 0; struct isst_if_cpu_maps map; - cpu_map = malloc(sizeof(*cpu_map) * topo_max_cpus); + /* Use calloc to make sure the memory is initialized to Zero */ + cpu_map = calloc(topo_max_cpus, sizeof(*cpu_map)); if (!cpu_map) err(3, "cpumap"); fd = open(pathname, O_RDWR); - if (fd < 0) + if (fd < 0 && !is_clx_n_platform()) err(-1, "%s open failed", pathname); + size = alloc_cpu_set(&present_cpumask); + present_cpumask_size = size; + for (i = 0; i < topo_max_cpus; ++i) { - if (!CPU_ISSET_S(i, present_cpumask_size, present_cpumask)) + char buffer[256]; + int pkg_id, die_id, core_id; + + /* check if CPU is online */ + snprintf(buffer, sizeof(buffer), + "/sys/devices/system/cpu/cpu%d", i); + dir = opendir(buffer); + if (!dir) continue; + closedir(dir); + CPU_SET_S(i, size, present_cpumask); + + pkg_id = get_physical_package_id(i); + die_id = get_physical_die_id(i); + core_id = get_physical_core_id(i); + + if (pkg_id < 0 || die_id < 0 || core_id < 0) + continue; + + cpu_map[i].pkg_id = pkg_id; + cpu_map[i].die_id = die_id; + cpu_map[i].core_id = core_id; + cpu_map[i].initialized = 1; + + cpu_cnt[pkg_id][die_id]++; + + if (fd < 0) + continue; map.cmd_count = 1; map.cpu_map[0].logical_cpu = i; - debug_printf(" map logical_cpu:%d\n", map.cpu_map[0].logical_cpu); if (ioctl(fd, ISST_IF_GET_PHY_ID, &map) == -1) { @@ -649,9 +671,6 @@ static void create_cpu_map(void) map.cpu_map[0].logical_cpu); continue; } - cpu_map[i].core_id = get_physical_core_id(i); - cpu_map[i].pkg_id = get_physical_package_id(i); - cpu_map[i].die_id = get_physical_die_id(i); cpu_map[i].punit_cpu = map.cpu_map[0].physical_cpu; cpu_map[i].punit_cpu_core = (map.cpu_map[0].physical_cpu >> 1); // shift to get core id @@ -662,35 +681,27 @@ static void create_cpu_map(void) cpu_map[i].pkg_id, cpu_map[i].punit_cpu, cpu_map[i].punit_cpu_core); } - - if (fd) + if (fd >= 0) close(fd); -} -int find_logical_cpu(int pkg_id, int die_id, int punit_core_id) -{ - int i; + size = alloc_cpu_set(&target_cpumask); + target_cpumask_size = size; + for (i = 0; i < max_target_cpus; ++i) { + if (!CPU_ISSET_S(target_cpus[i], present_cpumask_size, + present_cpumask)) + continue; - for (i = 0; i < topo_max_cpus; ++i) { - if (cpu_map[i].pkg_id == pkg_id && - cpu_map[i].die_id == die_id && - cpu_map[i].punit_cpu_core == punit_core_id) - return i; + CPU_SET_S(target_cpus[i], size, target_cpumask); } - - return -EINVAL; } -void set_cpu_mask_from_punit_coremask(int cpu, unsigned long long core_mask, +void set_cpu_mask_from_punit_coremask(struct isst_id *id, unsigned long long core_mask, size_t core_cpumask_size, cpu_set_t *core_cpumask, int *cpu_cnt) { int i, cnt = 0; - int die_id, pkg_id; *cpu_cnt = 0; - die_id = get_physical_die_id(cpu); - pkg_id = get_physical_package_id(cpu); for (i = 0; i < 64; ++i) { if (core_mask & BIT_ULL(i)) { @@ -700,8 +711,7 @@ void set_cpu_mask_from_punit_coremask(int cpu, unsigned long long core_mask, if (!CPU_ISSET_S(j, present_cpumask_size, present_cpumask)) continue; - if (cpu_map[j].pkg_id == pkg_id && - cpu_map[j].die_id == die_id && + if (is_cpu_in_power_domain(j, id) && cpu_map[j].punit_cpu_core == i) { CPU_SET_S(j, core_cpumask_size, core_cpumask); @@ -932,6 +942,7 @@ static void isst_print_extended_platform_info(void) struct isst_pkg_ctdp pkg_dev; int ret, i, j; FILE *filep; + struct isst_id id; for (i = 0; i < 256; ++i) { char path[256]; @@ -948,7 +959,8 @@ static void isst_print_extended_platform_info(void) fclose(filep); - ret = isst_get_ctdp_levels(i, &pkg_dev); + set_isst_id(&id, i); + ret = isst_get_ctdp_levels(&id, &pkg_dev); if (ret) return; @@ -965,7 +977,7 @@ static void isst_print_extended_platform_info(void) fprintf(outf, "TDP level change control is unlocked, max level: %d \n", pkg_dev.levels); for (j = 0; j <= pkg_dev.levels; ++j) { - ret = isst_get_ctdp_control(i, j, &ctdp_level); + ret = isst_get_ctdp_control(&id, j, &ctdp_level); if (ret) continue; @@ -986,7 +998,7 @@ static void isst_print_extended_platform_info(void) else fprintf(outf, "Intel(R) SST-BF (feature base-freq) is not supported\n"); - ret = isst_read_pm_config(i, &cp_state, &cp_cap); + ret = isst_read_pm_config(&id, &cp_state, &cp_cap); if (ret) { fprintf(outf, "Intel(R) SST-CP (feature core-power) status is unknown\n"); return; @@ -1008,6 +1020,10 @@ static void isst_print_platform_information(void) exit(0); } + /* Early initialization to create working cpu_map */ + set_max_cpu_num(); + create_cpu_map(); + fd = open(pathname, O_RDWR); if (fd < 0) err(-1, "%s open failed", pathname); @@ -1032,18 +1048,18 @@ static void isst_print_platform_information(void) } static char *local_str0, *local_str1; -static void exec_on_get_ctdp_cpu(int cpu, void *arg1, void *arg2, void *arg3, +static void exec_on_get_ctdp_cpu(struct isst_id *id, void *arg1, void *arg2, void *arg3, void *arg4) { - int (*fn_ptr)(int cpu, void *arg); + int (*fn_ptr)(struct isst_id *id, void *arg); int ret; fn_ptr = arg1; - ret = fn_ptr(cpu, arg2); + ret = fn_ptr(id, arg2); if (ret) isst_display_error_info_message(1, "get_tdp_* failed", 0, 0); else - isst_ctdp_display_core_info(cpu, outf, arg3, + isst_ctdp_display_core_info(id, outf, arg3, *(unsigned int *)arg4, local_str0, local_str1); } @@ -1111,9 +1127,9 @@ static int clx_n_get_base_ratio(void) return (int)(value); } -static int clx_n_config(int cpu) +static int clx_n_config(struct isst_id *id) { - int i, ret, pkg_id, die_id; + int i, ret; unsigned long cpu_bf; struct isst_pkg_ctdp_level_info *ctdp_level; struct isst_pbf_info *pbf_info; @@ -1135,15 +1151,11 @@ static int clx_n_config(int cpu) pbf_info->p1_high = 0; pbf_info->p1_low = ~0; - pkg_id = get_physical_package_id(cpu); - die_id = get_physical_die_id(cpu); - for (i = 0; i < topo_max_cpus; i++) { if (!CPU_ISSET_S(i, present_cpumask_size, present_cpumask)) continue; - if (pkg_id != get_physical_package_id(i) || - die_id != get_physical_die_id(i)) + if (!is_cpu_in_power_domain(i, id)) continue; CPU_SET_S(i, ctdp_level->core_cpumask_size, @@ -1180,8 +1192,7 @@ static int clx_n_config(int cpu) if (!CPU_ISSET_S(i, present_cpumask_size, present_cpumask)) continue; - if (pkg_id != get_physical_package_id(i) || - die_id != get_physical_die_id(i)) + if (!is_cpu_in_power_domain(i, id)) continue; cpu_bf = parse_int_file(1, @@ -1207,7 +1218,7 @@ error_ret: return ret; } -static void dump_clx_n_config_for_cpu(int cpu, void *arg1, void *arg2, +static void dump_clx_n_config_for_cpu(struct isst_id *id, void *arg1, void *arg2, void *arg3, void *arg4) { int ret; @@ -1217,7 +1228,7 @@ static void dump_clx_n_config_for_cpu(int cpu, void *arg1, void *arg2, exit(0); } - ret = clx_n_config(cpu); + ret = clx_n_config(id); if (ret) { debug_printf("clx_n_config failed"); } else { @@ -1227,27 +1238,27 @@ static void dump_clx_n_config_for_cpu(int cpu, void *arg1, void *arg2, ctdp_level = &clx_n_pkg_dev.ctdp_level[0]; pbf_info = &ctdp_level->pbf_info; clx_n_pkg_dev.processed = 1; - isst_ctdp_display_information(cpu, outf, tdp_level, &clx_n_pkg_dev); + isst_ctdp_display_information(id, outf, tdp_level, &clx_n_pkg_dev); free_cpu_set(ctdp_level->core_cpumask); free_cpu_set(pbf_info->core_cpumask); } } -static void dump_isst_config_for_cpu(int cpu, void *arg1, void *arg2, +static void dump_isst_config_for_cpu(struct isst_id *id, void *arg1, void *arg2, void *arg3, void *arg4) { struct isst_pkg_ctdp pkg_dev; int ret; memset(&pkg_dev, 0, sizeof(pkg_dev)); - ret = isst_get_process_ctdp(cpu, tdp_level, &pkg_dev); + ret = isst_get_process_ctdp(id, tdp_level, &pkg_dev); if (ret) { - isst_display_error_info_message(1, "Failed to get perf-profile info on cpu", 1, cpu); + isst_display_error_info_message(1, "Failed to get perf-profile info on cpu", 1, id->cpu); isst_ctdp_display_information_end(outf); exit(1); } else { - isst_ctdp_display_information(cpu, outf, tdp_level, &pkg_dev); - isst_get_process_ctdp_complete(cpu, &pkg_dev); + isst_ctdp_display_information(id, outf, tdp_level, &pkg_dev); + isst_get_process_ctdp_complete(id, &pkg_dev); } } @@ -1283,23 +1294,21 @@ static void dump_isst_config(int arg) static void adjust_scaling_max_from_base_freq(int cpu); -static void set_tdp_level_for_cpu(int cpu, void *arg1, void *arg2, void *arg3, +static void set_tdp_level_for_cpu(struct isst_id *id, void *arg1, void *arg2, void *arg3, void *arg4) { int ret; - ret = isst_set_tdp_level(cpu, tdp_level); + ret = isst_set_tdp_level(id, tdp_level); if (ret) { isst_display_error_info_message(1, "Set TDP level failed", 0, 0); isst_ctdp_display_information_end(outf); exit(1); } else { - isst_display_result(cpu, outf, "perf-profile", "set_tdp_level", + isst_display_result(id, outf, "perf-profile", "set_tdp_level", ret); if (force_online_offline) { struct isst_pkg_ctdp_level_info ctdp_level; - int pkg_id = get_physical_package_id(cpu); - int die_id = get_physical_die_id(cpu); /* Wait for updated base frequencies */ usleep(2000); @@ -1307,7 +1316,7 @@ static void set_tdp_level_for_cpu(int cpu, void *arg1, void *arg2, void *arg3, fprintf(stderr, "Option is set to online/offline\n"); ctdp_level.core_cpumask_size = alloc_cpu_set(&ctdp_level.core_cpumask); - ret = isst_get_coremask_info(cpu, tdp_level, &ctdp_level); + ret = isst_get_coremask_info(id, tdp_level, &ctdp_level); if (ret) { isst_display_error_info_message(1, "Can't get coremask, online/offline option is ignored", 0, 0); return; @@ -1315,7 +1324,7 @@ static void set_tdp_level_for_cpu(int cpu, void *arg1, void *arg2, void *arg3, if (ctdp_level.cpu_count) { int i, max_cpus = get_topo_max_cpus(); for (i = 0; i < max_cpus; ++i) { - if (pkg_id != get_physical_package_id(i) || die_id != get_physical_die_id(i)) + if (!is_cpu_in_power_domain(i, id)) continue; if (CPU_ISSET_S(i, ctdp_level.core_cpumask_size, ctdp_level.core_cpumask)) { fprintf(stderr, "online cpu %d\n", i); @@ -1358,12 +1367,12 @@ static void set_tdp_level(int arg) isst_ctdp_display_information_end(outf); } -static void clx_n_dump_pbf_config_for_cpu(int cpu, void *arg1, void *arg2, +static void clx_n_dump_pbf_config_for_cpu(struct isst_id *id, void *arg1, void *arg2, void *arg3, void *arg4) { int ret; - ret = clx_n_config(cpu); + ret = clx_n_config(id); if (ret) { isst_display_error_info_message(1, "clx_n_config failed", 0, 0); } else { @@ -1372,25 +1381,25 @@ static void clx_n_dump_pbf_config_for_cpu(int cpu, void *arg1, void *arg2, ctdp_level = &clx_n_pkg_dev.ctdp_level[0]; pbf_info = &ctdp_level->pbf_info; - isst_pbf_display_information(cpu, outf, tdp_level, pbf_info); + isst_pbf_display_information(id, outf, tdp_level, pbf_info); free_cpu_set(ctdp_level->core_cpumask); free_cpu_set(pbf_info->core_cpumask); } } -static void dump_pbf_config_for_cpu(int cpu, void *arg1, void *arg2, void *arg3, +static void dump_pbf_config_for_cpu(struct isst_id *id, void *arg1, void *arg2, void *arg3, void *arg4) { struct isst_pbf_info pbf_info; int ret; - ret = isst_get_pbf_info(cpu, tdp_level, &pbf_info); + ret = isst_get_pbf_info(id, tdp_level, &pbf_info); if (ret) { isst_display_error_info_message(1, "Failed to get base-freq info at this level", 1, tdp_level); isst_ctdp_display_information_end(outf); exit(1); } else { - isst_pbf_display_information(cpu, outf, tdp_level, &pbf_info); + isst_pbf_display_information(id, outf, tdp_level, &pbf_info); isst_get_pbf_info_complete(&pbf_info); } } @@ -1427,12 +1436,12 @@ static void dump_pbf_config(int arg) isst_ctdp_display_information_end(outf); } -static int set_clos_param(int cpu, int clos, int epp, int wt, int min, int max) +static int set_clos_param(struct isst_id *id, int clos, int epp, int wt, int min, int max) { struct isst_clos_config clos_config; int ret; - ret = isst_pm_get_clos(cpu, clos, &clos_config); + ret = isst_pm_get_clos(id, clos, &clos_config); if (ret) { isst_display_error_info_message(1, "isst_pm_get_clos failed", 0, 0); return ret; @@ -1441,7 +1450,7 @@ static int set_clos_param(int cpu, int clos, int epp, int wt, int min, int max) clos_config.clos_max = max; clos_config.epp = epp; clos_config.clos_prop_prio = wt; - ret = isst_set_clos(cpu, clos, &clos_config); + ret = isst_set_clos(id, clos, &clos_config); if (ret) { isst_display_error_info_message(1, "isst_set_clos failed", 0, 0); return ret; @@ -1503,14 +1512,14 @@ static void adjust_scaling_min_from_base_freq(int cpu) set_cpufreq_scaling_min_max(cpu, 0, base_freq); } -static int set_clx_pbf_cpufreq_scaling_min_max(int cpu) +static int set_clx_pbf_cpufreq_scaling_min_max(struct isst_id *id) { struct isst_pkg_ctdp_level_info *ctdp_level; struct isst_pbf_info *pbf_info; - int i, pkg_id, die_id, freq, freq_high, freq_low; + int i, freq, freq_high, freq_low; int ret; - ret = clx_n_config(cpu); + ret = clx_n_config(id); if (ret) { debug_printf("cpufreq_scaling_min_max failed for CLX"); return ret; @@ -1521,11 +1530,8 @@ static int set_clx_pbf_cpufreq_scaling_min_max(int cpu) freq_high = pbf_info->p1_high * 100000; freq_low = pbf_info->p1_low * 100000; - pkg_id = get_physical_package_id(cpu); - die_id = get_physical_die_id(cpu); for (i = 0; i < get_topo_max_cpus(); ++i) { - if (pkg_id != get_physical_package_id(i) || - die_id != get_physical_die_id(i)) + if (!is_cpu_in_power_domain(i, id)) continue; if (CPU_ISSET_S(i, pbf_info->core_cpumask_size, @@ -1588,84 +1594,75 @@ static int set_cpufreq_scaling_min_max_from_cpuinfo(int cpu, int cpuinfo_max, in return 0; } -static void set_scaling_min_to_cpuinfo_max(int cpu) +static void set_scaling_min_to_cpuinfo_max(struct isst_id *id) { - int i, pkg_id, die_id; + int i; - pkg_id = get_physical_package_id(cpu); - die_id = get_physical_die_id(cpu); for (i = 0; i < get_topo_max_cpus(); ++i) { - if (pkg_id != get_physical_package_id(i) || - die_id != get_physical_die_id(i)) + if (!is_cpu_in_power_domain(i, id)) continue; + adjust_scaling_max_from_base_freq(i); set_cpufreq_scaling_min_max_from_cpuinfo(i, 1, 0); adjust_scaling_min_from_base_freq(i); } } -static void set_scaling_min_to_cpuinfo_min(int cpu) +static void set_scaling_min_to_cpuinfo_min(struct isst_id *id) { - int i, pkg_id, die_id; + int i; - pkg_id = get_physical_package_id(cpu); - die_id = get_physical_die_id(cpu); for (i = 0; i < get_topo_max_cpus(); ++i) { - if (pkg_id != get_physical_package_id(i) || - die_id != get_physical_die_id(i)) + if (!is_cpu_in_power_domain(i, id)) continue; + adjust_scaling_max_from_base_freq(i); set_cpufreq_scaling_min_max_from_cpuinfo(i, 0, 0); } } -static void set_scaling_max_to_cpuinfo_max(int cpu) +static void set_scaling_max_to_cpuinfo_max(struct isst_id *id) { - int i, pkg_id, die_id; + int i; - pkg_id = get_physical_package_id(cpu); - die_id = get_physical_die_id(cpu); for (i = 0; i < get_topo_max_cpus(); ++i) { - if (pkg_id != get_physical_package_id(i) || - die_id != get_physical_die_id(i)) + if (!is_cpu_in_power_domain(i, id)) continue; set_cpufreq_scaling_min_max_from_cpuinfo(i, 1, 1); } } -static int set_core_priority_and_min(int cpu, int mask_size, +static int set_core_priority_and_min(struct isst_id *id, int mask_size, cpu_set_t *cpu_mask, int min_high, int min_low) { - int pkg_id, die_id, ret, i; + int ret, i; if (!CPU_COUNT_S(mask_size, cpu_mask)) return -1; - ret = set_clos_param(cpu, 0, 0, 0, min_high, 0xff); + ret = set_clos_param(id, 0, 0, 0, min_high, 0xff); if (ret) return ret; - ret = set_clos_param(cpu, 1, 15, 15, min_low, 0xff); + ret = set_clos_param(id, 1, 15, 15, min_low, 0xff); if (ret) return ret; - ret = set_clos_param(cpu, 2, 15, 15, min_low, 0xff); + ret = set_clos_param(id, 2, 15, 15, min_low, 0xff); if (ret) return ret; - ret = set_clos_param(cpu, 3, 15, 15, min_low, 0xff); + ret = set_clos_param(id, 3, 15, 15, min_low, 0xff); if (ret) return ret; - pkg_id = get_physical_package_id(cpu); - die_id = get_physical_die_id(cpu); for (i = 0; i < get_topo_max_cpus(); ++i) { int clos; + struct isst_id tid; - if (pkg_id != get_physical_package_id(i) || - die_id != get_physical_die_id(i)) + if (!is_cpu_in_power_domain(i, id)) continue; if (CPU_ISSET_S(i, mask_size, cpu_mask)) @@ -1674,7 +1671,8 @@ static int set_core_priority_and_min(int cpu, int mask_size, clos = 3; debug_printf("Associate cpu: %d clos: %d\n", i, clos); - ret = isst_clos_associate(i, clos); + set_isst_id(&tid, i); + ret = isst_clos_associate(&tid, clos); if (ret) { isst_display_error_info_message(1, "isst_clos_associate failed", 0, 0); return ret; @@ -1684,20 +1682,20 @@ static int set_core_priority_and_min(int cpu, int mask_size, return 0; } -static int set_pbf_core_power(int cpu) +static int set_pbf_core_power(struct isst_id *id) { struct isst_pbf_info pbf_info; struct isst_pkg_ctdp pkg_dev; int ret; - ret = isst_get_ctdp_levels(cpu, &pkg_dev); + ret = isst_get_ctdp_levels(id, &pkg_dev); if (ret) { debug_printf("isst_get_ctdp_levels failed"); return ret; } debug_printf("Current_level: %d\n", pkg_dev.current_level); - ret = isst_get_pbf_info(cpu, pkg_dev.current_level, &pbf_info); + ret = isst_get_pbf_info(id, pkg_dev.current_level, &pbf_info); if (ret) { debug_printf("isst_get_pbf_info failed"); return ret; @@ -1705,7 +1703,7 @@ static int set_pbf_core_power(int cpu) debug_printf("p1_high: %d p1_low: %d\n", pbf_info.p1_high, pbf_info.p1_low); - ret = set_core_priority_and_min(cpu, pbf_info.core_cpumask_size, + ret = set_core_priority_and_min(id, pbf_info.core_cpumask_size, pbf_info.core_cpumask, pbf_info.p1_high, pbf_info.p1_low); if (ret) { @@ -1713,7 +1711,7 @@ static int set_pbf_core_power(int cpu) return ret; } - ret = isst_pm_qos_config(cpu, 1, 1); + ret = isst_pm_qos_config(id, 1, 1); if (ret) { debug_printf("isst_pm_qos_config failed"); return ret; @@ -1722,7 +1720,7 @@ static int set_pbf_core_power(int cpu) return 0; } -static void set_pbf_for_cpu(int cpu, void *arg1, void *arg2, void *arg3, +static void set_pbf_for_cpu(struct isst_id *id, void *arg1, void *arg2, void *arg3, void *arg4) { struct isst_pkg_ctdp_level_info ctdp_level; @@ -1733,22 +1731,22 @@ static void set_pbf_for_cpu(int cpu, void *arg1, void *arg2, void *arg3, if (is_clx_n_platform()) { ret = 0; if (status) { - set_clx_pbf_cpufreq_scaling_min_max(cpu); + set_clx_pbf_cpufreq_scaling_min_max(id); } else { - set_scaling_max_to_cpuinfo_max(cpu); - set_scaling_min_to_cpuinfo_min(cpu); + set_scaling_max_to_cpuinfo_max(id); + set_scaling_min_to_cpuinfo_min(id); } goto disp_result; } - ret = isst_get_ctdp_levels(cpu, &pkg_dev); + ret = isst_get_ctdp_levels(id, &pkg_dev); if (ret) { isst_display_error_info_message(1, "Failed to get number of levels", 0, 0); goto disp_result; } - ret = isst_get_ctdp_control(cpu, pkg_dev.current_level, &ctdp_level); + ret = isst_get_ctdp_control(id, pkg_dev.current_level, &ctdp_level); if (ret) { isst_display_error_info_message(1, "Failed to get current level", 0, 0); goto disp_result; @@ -1761,34 +1759,34 @@ static void set_pbf_for_cpu(int cpu, void *arg1, void *arg2, void *arg3, } if (auto_mode && status) { - ret = set_pbf_core_power(cpu); + ret = set_pbf_core_power(id); if (ret) goto disp_result; } - ret = isst_set_pbf_fact_status(cpu, 1, status); + ret = isst_set_pbf_fact_status(id, 1, status); if (ret) { debug_printf("isst_set_pbf_fact_status failed"); if (auto_mode) - isst_pm_qos_config(cpu, 0, 0); + isst_pm_qos_config(id, 0, 0); } else { if (auto_mode) { if (status) - set_scaling_min_to_cpuinfo_max(cpu); + set_scaling_min_to_cpuinfo_max(id); else - set_scaling_min_to_cpuinfo_min(cpu); + set_scaling_min_to_cpuinfo_min(id); } } if (auto_mode && !status) - isst_pm_qos_config(cpu, 0, 1); + isst_pm_qos_config(id, 0, 1); disp_result: if (status) - isst_display_result(cpu, outf, "base-freq", "enable", + isst_display_result(id, outf, "base-freq", "enable", ret); else - isst_display_result(cpu, outf, "base-freq", "disable", + isst_display_result(id, outf, "base-freq", "disable", ret); } @@ -1837,19 +1835,19 @@ static void set_pbf_enable(int arg) isst_ctdp_display_information_end(outf); } -static void dump_fact_config_for_cpu(int cpu, void *arg1, void *arg2, +static void dump_fact_config_for_cpu(struct isst_id *id, void *arg1, void *arg2, void *arg3, void *arg4) { struct isst_fact_info fact_info; int ret; - ret = isst_get_fact_info(cpu, tdp_level, fact_bucket, &fact_info); + ret = isst_get_fact_info(id, tdp_level, fact_bucket, &fact_info); if (ret) { isst_display_error_info_message(1, "Failed to get turbo-freq info at this level", 1, tdp_level); isst_ctdp_display_information_end(outf); exit(1); } else { - isst_fact_display_information(cpu, outf, tdp_level, fact_bucket, + isst_fact_display_information(id, outf, tdp_level, fact_bucket, fact_avx, &fact_info); } } @@ -1883,7 +1881,7 @@ static void dump_fact_config(int arg) isst_ctdp_display_information_end(outf); } -static void set_fact_for_cpu(int cpu, void *arg1, void *arg2, void *arg3, +static void set_fact_for_cpu(struct isst_id *id, void *arg1, void *arg2, void *arg3, void *arg4) { struct isst_pkg_ctdp_level_info ctdp_level; @@ -1891,13 +1889,19 @@ static void set_fact_for_cpu(int cpu, void *arg1, void *arg2, void *arg3, int ret; int status = *(int *)arg4; - ret = isst_get_ctdp_levels(cpu, &pkg_dev); + if (status && no_turbo()) { + isst_display_error_info_message(1, "Turbo mode is disabled", 0, 0); + ret = -1; + goto disp_results; + } + + ret = isst_get_ctdp_levels(id, &pkg_dev); if (ret) { isst_display_error_info_message(1, "Failed to get number of levels", 0, 0); goto disp_results; } - ret = isst_get_ctdp_control(cpu, pkg_dev.current_level, &ctdp_level); + ret = isst_get_ctdp_control(id, pkg_dev.current_level, &ctdp_level); if (ret) { isst_display_error_info_message(1, "Failed to get current level", 0, 0); goto disp_results; @@ -1910,16 +1914,16 @@ static void set_fact_for_cpu(int cpu, void *arg1, void *arg2, void *arg3, } if (status) { - ret = isst_pm_qos_config(cpu, 1, 1); + ret = isst_pm_qos_config(id, 1, 1); if (ret) goto disp_results; } - ret = isst_set_pbf_fact_status(cpu, 0, status); + ret = isst_set_pbf_fact_status(id, 0, status); if (ret) { debug_printf("isst_set_pbf_fact_status failed"); if (auto_mode) - isst_pm_qos_config(cpu, 0, 0); + isst_pm_qos_config(id, 0, 0); goto disp_results; } @@ -1928,31 +1932,32 @@ static void set_fact_for_cpu(int cpu, void *arg1, void *arg2, void *arg3, if (status) { struct isst_pkg_ctdp pkg_dev; - ret = isst_get_ctdp_levels(cpu, &pkg_dev); + ret = isst_get_ctdp_levels(id, &pkg_dev); if (!ret) - ret = isst_set_trl(cpu, fact_trl); + ret = isst_set_trl(id, fact_trl); if (ret && auto_mode) - isst_pm_qos_config(cpu, 0, 0); + isst_pm_qos_config(id, 0, 0); } else { if (auto_mode) - isst_pm_qos_config(cpu, 0, 0); + isst_pm_qos_config(id, 0, 0); } disp_results: if (status) { - isst_display_result(cpu, outf, "turbo-freq", "enable", ret); + isst_display_result(id, outf, "turbo-freq", "enable", ret); if (ret) fact_enable_fail = ret; } else { /* Since we modified TRL during Fact enable, restore it */ - isst_set_trl_from_current_tdp(cpu, fact_trl); - isst_display_result(cpu, outf, "turbo-freq", "disable", ret); + isst_set_trl_from_current_tdp(id, fact_trl); + isst_display_result(id, outf, "turbo-freq", "disable", ret); } } static void set_fact_enable(int arg) { int i, ret, enable = arg; + struct isst_id id; if (cmd_help) { if (enable) { @@ -2026,19 +2031,20 @@ static void set_fact_enable(int arg) if (!CPU_ISSET_S(i, present_cpumask_size, present_cpumask)) continue; - ret = set_clos_param(i, 0, 0, 0, 0, 0xff); + set_isst_id(&id, i); + ret = set_clos_param(&id, 0, 0, 0, 0, 0xff); if (ret) goto error_disp; - ret = set_clos_param(i, 1, 15, 15, 0, 0xff); + ret = set_clos_param(&id, 1, 15, 15, 0, 0xff); if (ret) goto error_disp; - ret = set_clos_param(i, 2, 15, 15, 0, 0xff); + ret = set_clos_param(&id, 2, 15, 15, 0, 0xff); if (ret) goto error_disp; - ret = set_clos_param(i, 3, 15, 15, 0, 0xff); + ret = set_clos_param(&id, 3, 15, 15, 0, 0xff); if (ret) goto error_disp; @@ -2048,21 +2054,22 @@ static void set_fact_enable(int arg) clos = 3; debug_printf("Associate cpu: %d clos: %d\n", i, clos); - ret = isst_clos_associate(i, clos); + ret = isst_clos_associate(&id, clos); if (ret) goto error_disp; } - isst_display_result(-1, outf, "turbo-freq --auto", "enable", 0); + set_isst_id(&id, -1); + isst_display_result(&id, outf, "turbo-freq --auto", "enable", 0); } return; error_disp: - isst_display_result(i, outf, "turbo-freq --auto", "enable", ret); + isst_display_result(&id, outf, "turbo-freq --auto", "enable", ret); } -static void enable_clos_qos_config(int cpu, void *arg1, void *arg2, void *arg3, +static void enable_clos_qos_config(struct isst_id *id, void *arg1, void *arg2, void *arg3, void *arg4) { int ret; @@ -2071,15 +2078,15 @@ static void enable_clos_qos_config(int cpu, void *arg1, void *arg2, void *arg3, if (is_skx_based_platform()) clos_priority_type = 1; - ret = isst_pm_qos_config(cpu, status, clos_priority_type); + ret = isst_pm_qos_config(id, status, clos_priority_type); if (ret) isst_display_error_info_message(1, "isst_pm_qos_config failed", 0, 0); if (status) - isst_display_result(cpu, outf, "core-power", "enable", + isst_display_result(id, outf, "core-power", "enable", ret); else - isst_display_result(cpu, outf, "core-power", "disable", + isst_display_result(id, outf, "core-power", "disable", ret); } @@ -2118,17 +2125,17 @@ static void set_clos_enable(int arg) isst_ctdp_display_information_end(outf); } -static void dump_clos_config_for_cpu(int cpu, void *arg1, void *arg2, +static void dump_clos_config_for_cpu(struct isst_id *id, void *arg1, void *arg2, void *arg3, void *arg4) { struct isst_clos_config clos_config; int ret; - ret = isst_pm_get_clos(cpu, current_clos, &clos_config); + ret = isst_pm_get_clos(id, current_clos, &clos_config); if (ret) isst_display_error_info_message(1, "isst_pm_get_clos failed", 0, 0); else - isst_clos_display_information(cpu, outf, current_clos, + isst_clos_display_information(id, outf, current_clos, &clos_config); } @@ -2157,19 +2164,19 @@ static void dump_clos_config(int arg) isst_ctdp_display_information_end(outf); } -static void get_clos_info_for_cpu(int cpu, void *arg1, void *arg2, void *arg3, +static void get_clos_info_for_cpu(struct isst_id *id, void *arg1, void *arg2, void *arg3, void *arg4) { int enable, ret, prio_type; - ret = isst_clos_get_clos_information(cpu, &enable, &prio_type); + ret = isst_clos_get_clos_information(id, &enable, &prio_type); if (ret) isst_display_error_info_message(1, "isst_clos_get_info failed", 0, 0); else { int cp_state, cp_cap; - isst_read_pm_config(cpu, &cp_state, &cp_cap); - isst_clos_display_clos_information(cpu, outf, enable, prio_type, + isst_read_pm_config(id, &cp_state, &cp_cap); + isst_clos_display_clos_information(id, outf, enable, prio_type, cp_state, cp_cap); } } @@ -2194,25 +2201,22 @@ static void dump_clos_info(int arg) } -static void set_clos_config_for_cpu(int cpu, void *arg1, void *arg2, void *arg3, +static void set_clos_config_for_cpu(struct isst_id *id, void *arg1, void *arg2, void *arg3, void *arg4) { struct isst_clos_config clos_config; int ret; - clos_config.pkg_id = get_physical_package_id(cpu); - clos_config.die_id = get_physical_die_id(cpu); - clos_config.epp = clos_epp; clos_config.clos_prop_prio = clos_prop_prio; clos_config.clos_min = clos_min; clos_config.clos_max = clos_max; clos_config.clos_desired = clos_desired; - ret = isst_set_clos(cpu, current_clos, &clos_config); + ret = isst_set_clos(id, current_clos, &clos_config); if (ret) isst_display_error_info_message(1, "isst_set_clos failed", 0, 0); else - isst_display_result(cpu, outf, "core-power", "config", ret); + isst_display_result(id, outf, "core-power", "config", ret); } static void set_clos_config(int arg) @@ -2268,16 +2272,16 @@ static void set_clos_config(int arg) isst_ctdp_display_information_end(outf); } -static void set_clos_assoc_for_cpu(int cpu, void *arg1, void *arg2, void *arg3, +static void set_clos_assoc_for_cpu(struct isst_id *id, void *arg1, void *arg2, void *arg3, void *arg4) { int ret; - ret = isst_clos_associate(cpu, current_clos); + ret = isst_clos_associate(id, current_clos); if (ret) debug_printf("isst_clos_associate failed"); else - isst_display_result(cpu, outf, "core-power", "assoc", ret); + isst_display_result(id, outf, "core-power", "assoc", ret); } static void set_clos_assoc(int arg) @@ -2305,16 +2309,16 @@ static void set_clos_assoc(int arg) } } -static void get_clos_assoc_for_cpu(int cpu, void *arg1, void *arg2, void *arg3, +static void get_clos_assoc_for_cpu(struct isst_id *id, void *arg1, void *arg2, void *arg3, void *arg4) { int clos, ret; - ret = isst_clos_get_assoc_status(cpu, &clos); + ret = isst_clos_get_assoc_status(id, &clos); if (ret) isst_display_error_info_message(1, "isst_clos_get_assoc_status failed", 0, 0); else - isst_clos_display_assoc_information(cpu, outf, clos); + isst_clos_display_assoc_information(id, outf, clos); } static void get_clos_assoc(int arg) @@ -2336,27 +2340,28 @@ static void get_clos_assoc(int arg) isst_ctdp_display_information_end(outf); } -static void set_turbo_mode_for_cpu(int cpu, int status) +static void set_turbo_mode_for_cpu(struct isst_id *id, int status) { int base_freq; if (status) { - base_freq = get_cpufreq_base_freq(cpu); - set_cpufreq_scaling_min_max(cpu, 1, base_freq); + base_freq = get_cpufreq_base_freq(id->cpu); + set_cpufreq_scaling_min_max(id->cpu, 1, base_freq); } else { - set_scaling_max_to_cpuinfo_max(cpu); + set_scaling_max_to_cpuinfo_max(id); } if (status) { - isst_display_result(cpu, outf, "turbo-mode", "enable", 0); + isst_display_result(id, outf, "turbo-mode", "enable", 0); } else { - isst_display_result(cpu, outf, "turbo-mode", "disable", 0); + isst_display_result(id, outf, "turbo-mode", "disable", 0); } } static void set_turbo_mode(int arg) { int i, enable = arg; + struct isst_id id; if (cmd_help) { if (enable) @@ -2378,14 +2383,16 @@ static void set_turbo_mode(int arg) online = 1; /* online entry for CPU 0 needs some special configs */ - if (online) - set_turbo_mode_for_cpu(i, enable); + if (online) { + set_isst_id(&id, i); + set_turbo_mode_for_cpu(&id, enable); + } } isst_ctdp_display_information_end(outf); } -static void get_set_trl(int cpu, void *arg1, void *arg2, void *arg3, +static void get_set_trl(struct isst_id *id, void *arg1, void *arg2, void *arg3, void *arg4) { unsigned long long trl; @@ -2398,16 +2405,16 @@ static void get_set_trl(int cpu, void *arg1, void *arg2, void *arg3, } if (set) { - ret = isst_set_trl(cpu, fact_trl); - isst_display_result(cpu, outf, "turbo-mode", "set-trl", ret); + ret = isst_set_trl(id, fact_trl); + isst_display_result(id, outf, "turbo-mode", "set-trl", ret); return; } - ret = isst_get_trl(cpu, &trl); + ret = isst_get_trl(id, &trl); if (ret) - isst_display_result(cpu, outf, "turbo-mode", "get-trl", ret); + isst_display_result(id, outf, "turbo-mode", "get-trl", ret); else - isst_trl_display_information(cpu, outf, trl); + isst_trl_display_information(id, outf, trl); } static void process_trl(int arg) @@ -2747,9 +2754,6 @@ void process_command(int argc, char **argv, } } - if (!is_clx_n_platform()) - create_cpu_map(); - i = 0; while (cmds[i].feature) { if (!strcmp(cmds[i].feature, feature) && @@ -2801,7 +2805,9 @@ static void usage(void) printf("\t[-p|--pause] : Delay between two mail box commands in milliseconds\n"); printf("\t[-r|--retry] : Retry count for mail box commands on failure, default 3\n"); printf("\t[-v|--version] : Print version\n"); - + printf("\t[-b|--oob : Start a daemon to process HFI events for perf profile change from Out of Band agent.\n"); + printf("\t[-n|--no-daemon : Don't run as daemon. By default --oob will turn on daemon mode\n"); + printf("\t[-w|--delay : Delay for reading config level state change in OOB poll mode.\n"); printf("\nResult format\n"); printf("\tResult display uses a common format for each command:\n"); printf("\tResults are formatted in text/JSON with\n"); @@ -2835,6 +2841,9 @@ static void cmdline(int argc, char **argv) int opt, force_cpus_online = 0; int option_index = 0; int ret; + int oob_mode = 0; + int poll_interval = -1; + int no_daemon = 0; static struct option long_options[] = { { "all-cpus-online", no_argument, 0, 'a' }, @@ -2847,6 +2856,9 @@ static void cmdline(int argc, char **argv) { "out", required_argument, 0, 'o' }, { "retry", required_argument, 0, 'r' }, { "version", no_argument, 0, 'v' }, + { "oob", no_argument, 0, 'b' }, + { "no-daemon", no_argument, 0, 'n' }, + { "poll-interval", required_argument, 0, 'w' }, { 0, 0, 0, 0 } }; @@ -2873,7 +2885,7 @@ static void cmdline(int argc, char **argv) } progname = argv[0]; - while ((opt = getopt_long_only(argc, argv, "+c:df:hio:va", long_options, + while ((opt = getopt_long_only(argc, argv, "+c:df:hio:vabw:n", long_options, &option_index)) != -1) { switch (opt) { case 'a': @@ -2918,12 +2930,26 @@ static void cmdline(int argc, char **argv) case 'v': print_version(); break; + case 'b': + oob_mode = 1; + break; + case 'n': + no_daemon = 1; + break; + case 'w': + ret = strtol(optarg, &ptr, 10); + if (!ret) { + fprintf(stderr, "Invalid poll interval count\n"); + exit(0); + } + poll_interval = ret; + break; default: usage(); } } - if (optind > (argc - 2)) { + if (optind > (argc - 2) && !oob_mode) { usage(); exit(0); } @@ -2931,8 +2957,17 @@ static void cmdline(int argc, char **argv) if (force_cpus_online) force_all_cpus_online(); store_cpu_topology(); - set_cpu_present_cpu_mask(); - set_cpu_target_cpu_mask(); + create_cpu_map(); + + if (oob_mode) { + if (debug_flag) + fprintf(stderr, "OOB mode is enabled in debug mode\n"); + + ret = isst_daemon(debug_flag, poll_interval, no_daemon); + if (ret) + fprintf(stderr, "OOB mode enable failed\n"); + goto out; + } if (!is_clx_n_platform()) { ret = isst_fill_platform_info(); diff --git a/tools/power/x86/intel-speed-select/isst-core.c b/tools/power/x86/intel-speed-select/isst-core.c index 4431c8a0d40a..f701b45c832c 100644 --- a/tools/power/x86/intel-speed-select/isst-core.c +++ b/tools/power/x86/intel-speed-select/isst-core.c @@ -6,7 +6,7 @@ #include "isst.h" -int isst_write_pm_config(int cpu, int cp_state) +int isst_write_pm_config(struct isst_id *id, int cp_state) { unsigned int req, resp; int ret; @@ -16,27 +16,27 @@ int isst_write_pm_config(int cpu, int cp_state) else req = 0; - ret = isst_send_mbox_command(cpu, WRITE_PM_CONFIG, PM_FEATURE, 0, req, + ret = isst_send_mbox_command(id->cpu, WRITE_PM_CONFIG, PM_FEATURE, 0, req, &resp); if (ret) return ret; - debug_printf("cpu:%d WRITE_PM_CONFIG resp:%x\n", cpu, resp); + debug_printf("cpu:%d WRITE_PM_CONFIG resp:%x\n", id->cpu, resp); return 0; } -int isst_read_pm_config(int cpu, int *cp_state, int *cp_cap) +int isst_read_pm_config(struct isst_id *id, int *cp_state, int *cp_cap) { unsigned int resp; int ret; - ret = isst_send_mbox_command(cpu, READ_PM_CONFIG, PM_FEATURE, 0, 0, + ret = isst_send_mbox_command(id->cpu, READ_PM_CONFIG, PM_FEATURE, 0, 0, &resp); if (ret) return ret; - debug_printf("cpu:%d READ_PM_CONFIG resp:%x\n", cpu, resp); + debug_printf("cpu:%d READ_PM_CONFIG resp:%x\n", id->cpu, resp); *cp_state = resp & BIT(16); *cp_cap = resp & BIT(0) ? 1 : 0; @@ -44,12 +44,12 @@ int isst_read_pm_config(int cpu, int *cp_state, int *cp_cap) return 0; } -int isst_get_ctdp_levels(int cpu, struct isst_pkg_ctdp *pkg_dev) +int isst_get_ctdp_levels(struct isst_id *id, struct isst_pkg_ctdp *pkg_dev) { unsigned int resp; int ret; - ret = isst_send_mbox_command(cpu, CONFIG_TDP, + ret = isst_send_mbox_command(id->cpu, CONFIG_TDP, CONFIG_TDP_GET_LEVELS_INFO, 0, 0, &resp); if (ret) { pkg_dev->levels = 0; @@ -60,7 +60,7 @@ int isst_get_ctdp_levels(int cpu, struct isst_pkg_ctdp *pkg_dev) return 0; } - debug_printf("cpu:%d CONFIG_TDP_GET_LEVELS_INFO resp:%x\n", cpu, resp); + debug_printf("cpu:%d CONFIG_TDP_GET_LEVELS_INFO resp:%x\n", id->cpu, resp); pkg_dev->version = resp & 0xff; pkg_dev->levels = (resp >> 8) & 0xff; @@ -71,14 +71,14 @@ int isst_get_ctdp_levels(int cpu, struct isst_pkg_ctdp *pkg_dev) return 0; } -int isst_get_ctdp_control(int cpu, int config_index, +int isst_get_ctdp_control(struct isst_id *id, int config_index, struct isst_pkg_ctdp_level_info *ctdp_level) { int cp_state, cp_cap; unsigned int resp; int ret; - ret = isst_send_mbox_command(cpu, CONFIG_TDP, + ret = isst_send_mbox_command(id->cpu, CONFIG_TDP, CONFIG_TDP_GET_TDP_CONTROL, 0, config_index, &resp); if (ret) @@ -89,30 +89,30 @@ int isst_get_ctdp_control(int cpu, int config_index, ctdp_level->fact_enabled = !!(resp & BIT(16)); ctdp_level->pbf_enabled = !!(resp & BIT(17)); - ret = isst_read_pm_config(cpu, &cp_state, &cp_cap); + ret = isst_read_pm_config(id, &cp_state, &cp_cap); if (ret) { - debug_printf("cpu:%d pm_config is not supported \n", cpu); + debug_printf("cpu:%d pm_config is not supported\n", id->cpu); } else { - debug_printf("cpu:%d pm_config SST-CP state:%d cap:%d \n", cpu, cp_state, cp_cap); + debug_printf("cpu:%d pm_config SST-CP state:%d cap:%d\n", id->cpu, cp_state, cp_cap); ctdp_level->sst_cp_support = cp_cap; ctdp_level->sst_cp_enabled = cp_state; } debug_printf( "cpu:%d CONFIG_TDP_GET_TDP_CONTROL resp:%x fact_support:%d pbf_support: %d fact_enabled:%d pbf_enabled:%d\n", - cpu, resp, ctdp_level->fact_support, ctdp_level->pbf_support, + id->cpu, resp, ctdp_level->fact_support, ctdp_level->pbf_support, ctdp_level->fact_enabled, ctdp_level->pbf_enabled); return 0; } -int isst_get_tdp_info(int cpu, int config_index, +int isst_get_tdp_info(struct isst_id *id, int config_index, struct isst_pkg_ctdp_level_info *ctdp_level) { unsigned int resp; int ret; - ret = isst_send_mbox_command(cpu, CONFIG_TDP, CONFIG_TDP_GET_TDP_INFO, + ret = isst_send_mbox_command(id->cpu, CONFIG_TDP, CONFIG_TDP_GET_TDP_INFO, 0, config_index, &resp); if (ret) { isst_display_error_info_message(1, "Invalid level, Can't get TDP information at level", 1, config_index); @@ -124,18 +124,18 @@ int isst_get_tdp_info(int cpu, int config_index, debug_printf( "cpu:%d ctdp:%d CONFIG_TDP_GET_TDP_INFO resp:%x tdp_ratio:%d pkg_tdp:%d\n", - cpu, config_index, resp, ctdp_level->tdp_ratio, + id->cpu, config_index, resp, ctdp_level->tdp_ratio, ctdp_level->pkg_tdp); return 0; } -int isst_get_pwr_info(int cpu, int config_index, +int isst_get_pwr_info(struct isst_id *id, int config_index, struct isst_pkg_ctdp_level_info *ctdp_level) { unsigned int resp; int ret; - ret = isst_send_mbox_command(cpu, CONFIG_TDP, CONFIG_TDP_GET_PWR_INFO, + ret = isst_send_mbox_command(id->cpu, CONFIG_TDP, CONFIG_TDP_GET_PWR_INFO, 0, config_index, &resp); if (ret) return ret; @@ -145,18 +145,18 @@ int isst_get_pwr_info(int cpu, int config_index, debug_printf( "cpu:%d ctdp:%d CONFIG_TDP_GET_PWR_INFO resp:%x pkg_max_power:%d pkg_min_power:%d\n", - cpu, config_index, resp, ctdp_level->pkg_max_power, + id->cpu, config_index, resp, ctdp_level->pkg_max_power, ctdp_level->pkg_min_power); return 0; } -void isst_get_uncore_p0_p1_info(int cpu, int config_index, +void isst_get_uncore_p0_p1_info(struct isst_id *id, int config_index, struct isst_pkg_ctdp_level_info *ctdp_level) { unsigned int resp; int ret; - ret = isst_send_mbox_command(cpu, CONFIG_TDP, + ret = isst_send_mbox_command(id->cpu, CONFIG_TDP, CONFIG_TDP_GET_UNCORE_P0_P1_INFO, 0, config_index, &resp); if (ret) { @@ -169,16 +169,16 @@ void isst_get_uncore_p0_p1_info(int cpu, int config_index, ctdp_level->uncore_p1 = (resp & GENMASK(15, 8)) >> 8; debug_printf( "cpu:%d ctdp:%d CONFIG_TDP_GET_UNCORE_P0_P1_INFO resp:%x uncore p0:%d uncore p1:%d\n", - cpu, config_index, resp, ctdp_level->uncore_p0, + id->cpu, config_index, resp, ctdp_level->uncore_p0, ctdp_level->uncore_p1); } -void isst_get_p1_info(int cpu, int config_index, +void isst_get_p1_info(struct isst_id *id, int config_index, struct isst_pkg_ctdp_level_info *ctdp_level) { unsigned int resp; int ret; - ret = isst_send_mbox_command(cpu, CONFIG_TDP, CONFIG_TDP_GET_P1_INFO, 0, + ret = isst_send_mbox_command(id->cpu, CONFIG_TDP, CONFIG_TDP_GET_P1_INFO, 0, config_index, &resp); if (ret) { ctdp_level->sse_p1 = 0; @@ -192,17 +192,17 @@ void isst_get_p1_info(int cpu, int config_index, ctdp_level->avx512_p1 = (resp & GENMASK(23, 16)) >> 16; debug_printf( "cpu:%d ctdp:%d CONFIG_TDP_GET_P1_INFO resp:%x sse_p1:%d avx2_p1:%d avx512_p1:%d\n", - cpu, config_index, resp, ctdp_level->sse_p1, + id->cpu, config_index, resp, ctdp_level->sse_p1, ctdp_level->avx2_p1, ctdp_level->avx512_p1); } -void isst_get_uncore_mem_freq(int cpu, int config_index, +void isst_get_uncore_mem_freq(struct isst_id *id, int config_index, struct isst_pkg_ctdp_level_info *ctdp_level) { unsigned int resp; int ret; - ret = isst_send_mbox_command(cpu, CONFIG_TDP, CONFIG_TDP_GET_MEM_FREQ, + ret = isst_send_mbox_command(id->cpu, CONFIG_TDP, CONFIG_TDP_GET_MEM_FREQ, 0, config_index, &resp); if (ret) { ctdp_level->mem_freq = 0; @@ -226,16 +226,16 @@ void isst_get_uncore_mem_freq(int cpu, int config_index, } debug_printf( "cpu:%d ctdp:%d CONFIG_TDP_GET_MEM_FREQ resp:%x uncore mem_freq:%d\n", - cpu, config_index, resp, ctdp_level->mem_freq); + id->cpu, config_index, resp, ctdp_level->mem_freq); } -int isst_get_tjmax_info(int cpu, int config_index, +int isst_get_tjmax_info(struct isst_id *id, int config_index, struct isst_pkg_ctdp_level_info *ctdp_level) { unsigned int resp; int ret; - ret = isst_send_mbox_command(cpu, CONFIG_TDP, CONFIG_TDP_GET_TJMAX_INFO, + ret = isst_send_mbox_command(id->cpu, CONFIG_TDP, CONFIG_TDP_GET_TJMAX_INFO, 0, config_index, &resp); if (ret) return ret; @@ -244,12 +244,12 @@ int isst_get_tjmax_info(int cpu, int config_index, debug_printf( "cpu:%d ctdp:%d CONFIG_TDP_GET_TJMAX_INFO resp:%x t_proc_hot:%d\n", - cpu, config_index, resp, ctdp_level->t_proc_hot); + id->cpu, config_index, resp, ctdp_level->t_proc_hot); return 0; } -int isst_get_coremask_info(int cpu, int config_index, +int isst_get_coremask_info(struct isst_id *id, int config_index, struct isst_pkg_ctdp_level_info *ctdp_level) { unsigned int resp; @@ -260,7 +260,7 @@ int isst_get_coremask_info(int cpu, int config_index, unsigned long long mask; int cpu_count = 0; - ret = isst_send_mbox_command(cpu, CONFIG_TDP, + ret = isst_send_mbox_command(id->cpu, CONFIG_TDP, CONFIG_TDP_GET_CORE_MASK, 0, (i << 8) | config_index, &resp); if (ret) @@ -268,27 +268,27 @@ int isst_get_coremask_info(int cpu, int config_index, debug_printf( "cpu:%d ctdp:%d mask:%d CONFIG_TDP_GET_CORE_MASK resp:%x\n", - cpu, config_index, i, resp); + id->cpu, config_index, i, resp); mask = (unsigned long long)resp << (32 * i); - set_cpu_mask_from_punit_coremask(cpu, mask, + set_cpu_mask_from_punit_coremask(id, mask, ctdp_level->core_cpumask_size, ctdp_level->core_cpumask, &cpu_count); ctdp_level->cpu_count += cpu_count; - debug_printf("cpu:%d ctdp:%d mask:%d cpu count:%d\n", cpu, + debug_printf("cpu:%d ctdp:%d mask:%d cpu count:%d\n", id->cpu, config_index, i, ctdp_level->cpu_count); } return 0; } -int isst_get_get_trl_from_msr(int cpu, int *trl) +int isst_get_get_trl_from_msr(struct isst_id *id, int *trl) { unsigned long long msr_trl; int ret; - ret = isst_send_msr_command(cpu, 0x1AD, 0, &msr_trl); + ret = isst_send_msr_command(id->cpu, 0x1AD, 0, &msr_trl); if (ret) return ret; @@ -304,13 +304,13 @@ int isst_get_get_trl_from_msr(int cpu, int *trl) return 0; } -int isst_get_get_trl(int cpu, int level, int avx_level, int *trl) +int isst_get_get_trl(struct isst_id *id, int level, int avx_level, int *trl) { unsigned int req, resp; int ret; req = level | (avx_level << 16); - ret = isst_send_mbox_command(cpu, CONFIG_TDP, + ret = isst_send_mbox_command(id->cpu, CONFIG_TDP, CONFIG_TDP_GET_TURBO_LIMIT_RATIOS, 0, req, &resp); if (ret) @@ -318,7 +318,7 @@ int isst_get_get_trl(int cpu, int level, int avx_level, int *trl) debug_printf( "cpu:%d CONFIG_TDP_GET_TURBO_LIMIT_RATIOS req:%x resp:%x\n", - cpu, req, resp); + id->cpu, req, resp); trl[0] = resp & GENMASK(7, 0); trl[1] = (resp & GENMASK(15, 8)) >> 8; @@ -326,13 +326,13 @@ int isst_get_get_trl(int cpu, int level, int avx_level, int *trl) trl[3] = (resp & GENMASK(31, 24)) >> 24; req = level | BIT(8) | (avx_level << 16); - ret = isst_send_mbox_command(cpu, CONFIG_TDP, + ret = isst_send_mbox_command(id->cpu, CONFIG_TDP, CONFIG_TDP_GET_TURBO_LIMIT_RATIOS, 0, req, &resp); if (ret) return ret; - debug_printf("cpu:%d CONFIG_TDP_GET_TURBO_LIMIT req:%x resp:%x\n", cpu, + debug_printf("cpu:%d CONFIG_TDP_GET_TURBO_LIMIT req:%x resp:%x\n", id->cpu, req, resp); trl[4] = resp & GENMASK(7, 0); @@ -343,61 +343,37 @@ int isst_get_get_trl(int cpu, int level, int avx_level, int *trl) return 0; } -int isst_get_trl_bucket_info(int cpu, unsigned long long *buckets_info) +int isst_get_trl_bucket_info(struct isst_id *id, unsigned long long *buckets_info) { int ret; - debug_printf("cpu:%d bucket info via MSR\n", cpu); + debug_printf("cpu:%d bucket info via MSR\n", id->cpu); *buckets_info = 0; - ret = isst_send_msr_command(cpu, 0x1ae, 0, buckets_info); + ret = isst_send_msr_command(id->cpu, 0x1ae, 0, buckets_info); if (ret) return ret; - debug_printf("cpu:%d bucket info via MSR successful 0x%llx\n", cpu, + debug_printf("cpu:%d bucket info via MSR successful 0x%llx\n", id->cpu, *buckets_info); return 0; } -int isst_set_tdp_level_msr(int cpu, int tdp_level) -{ - unsigned long long level = tdp_level; - int ret; - - debug_printf("cpu: tdp_level via MSR %d\n", cpu, tdp_level); - - if (isst_get_config_tdp_lock_status(cpu)) { - isst_display_error_info_message(1, "tdp_locked", 0, 0); - return -1; - } - - if (tdp_level > 2) - return -1; /* invalid value */ - - ret = isst_send_msr_command(cpu, 0x64b, 1, &level); - if (ret) - return ret; - - debug_printf("cpu: tdp_level via MSR successful %d\n", cpu, tdp_level); - - return 0; -} - -int isst_set_tdp_level(int cpu, int tdp_level) +int isst_set_tdp_level(struct isst_id *id, int tdp_level) { unsigned int resp; int ret; - if (isst_get_config_tdp_lock_status(cpu)) { + if (isst_get_config_tdp_lock_status(id)) { isst_display_error_info_message(1, "TDP is locked", 0, 0); return -1; } - ret = isst_send_mbox_command(cpu, CONFIG_TDP, CONFIG_TDP_SET_LEVEL, 0, + ret = isst_send_mbox_command(id->cpu, CONFIG_TDP, CONFIG_TDP_SET_LEVEL, 0, tdp_level, &resp); if (ret) { isst_display_error_info_message(1, "Set TDP level failed for level", 1, tdp_level); @@ -407,14 +383,14 @@ int isst_set_tdp_level(int cpu, int tdp_level) return 0; } -int isst_get_pbf_info(int cpu, int level, struct isst_pbf_info *pbf_info) +int isst_get_pbf_info(struct isst_id *id, int level, struct isst_pbf_info *pbf_info) { struct isst_pkg_ctdp_level_info ctdp_level; struct isst_pkg_ctdp pkg_dev; int i, ret, max_punit_core, max_mask_index; unsigned int req, resp; - ret = isst_get_ctdp_levels(cpu, &pkg_dev); + ret = isst_get_ctdp_levels(id, &pkg_dev); if (ret) { isst_display_error_info_message(1, "Failed to get number of levels", 0, 0); return ret; @@ -425,7 +401,7 @@ int isst_get_pbf_info(int cpu, int level, struct isst_pbf_info *pbf_info) return -1; } - ret = isst_get_ctdp_control(cpu, level, &ctdp_level); + ret = isst_get_ctdp_control(id, level, &ctdp_level); if (ret) return ret; @@ -436,14 +412,14 @@ int isst_get_pbf_info(int cpu, int level, struct isst_pbf_info *pbf_info) pbf_info->core_cpumask_size = alloc_cpu_set(&pbf_info->core_cpumask); - max_punit_core = get_max_punit_core_id(get_physical_package_id(cpu), get_physical_die_id(cpu)); + max_punit_core = get_max_punit_core_id(id); max_mask_index = max_punit_core > 32 ? 2 : 1; for (i = 0; i < max_mask_index; ++i) { unsigned long long mask; int count; - ret = isst_send_mbox_command(cpu, CONFIG_TDP, + ret = isst_send_mbox_command(id->cpu, CONFIG_TDP, CONFIG_TDP_PBF_GET_CORE_MASK_INFO, 0, (i << 8) | level, &resp); if (ret) @@ -451,23 +427,23 @@ int isst_get_pbf_info(int cpu, int level, struct isst_pbf_info *pbf_info) debug_printf( "cpu:%d CONFIG_TDP_PBF_GET_CORE_MASK_INFO resp:%x\n", - cpu, resp); + id->cpu, resp); mask = (unsigned long long)resp << (32 * i); - set_cpu_mask_from_punit_coremask(cpu, mask, + set_cpu_mask_from_punit_coremask(id, mask, pbf_info->core_cpumask_size, pbf_info->core_cpumask, &count); } req = level; - ret = isst_send_mbox_command(cpu, CONFIG_TDP, + ret = isst_send_mbox_command(id->cpu, CONFIG_TDP, CONFIG_TDP_PBF_GET_P1HI_P1LO_INFO, 0, req, &resp); if (ret) return ret; - debug_printf("cpu:%d CONFIG_TDP_PBF_GET_P1HI_P1LO_INFO resp:%x\n", cpu, + debug_printf("cpu:%d CONFIG_TDP_PBF_GET_P1HI_P1LO_INFO resp:%x\n", id->cpu, resp); pbf_info->p1_low = resp & 0xff; @@ -475,21 +451,21 @@ int isst_get_pbf_info(int cpu, int level, struct isst_pbf_info *pbf_info) req = level; ret = isst_send_mbox_command( - cpu, CONFIG_TDP, CONFIG_TDP_PBF_GET_TDP_INFO, 0, req, &resp); + id->cpu, CONFIG_TDP, CONFIG_TDP_PBF_GET_TDP_INFO, 0, req, &resp); if (ret) return ret; - debug_printf("cpu:%d CONFIG_TDP_PBF_GET_TDP_INFO resp:%x\n", cpu, resp); + debug_printf("cpu:%d CONFIG_TDP_PBF_GET_TDP_INFO resp:%x\n", id->cpu, resp); pbf_info->tdp = resp & 0xffff; req = level; ret = isst_send_mbox_command( - cpu, CONFIG_TDP, CONFIG_TDP_PBF_GET_TJ_MAX_INFO, 0, req, &resp); + id->cpu, CONFIG_TDP, CONFIG_TDP_PBF_GET_TJ_MAX_INFO, 0, req, &resp); if (ret) return ret; - debug_printf("cpu:%d CONFIG_TDP_PBF_GET_TJ_MAX_INFO resp:%x\n", cpu, + debug_printf("cpu:%d CONFIG_TDP_PBF_GET_TJ_MAX_INFO resp:%x\n", id->cpu, resp); pbf_info->t_control = (resp >> 8) & 0xff; pbf_info->t_prochot = resp & 0xff; @@ -502,7 +478,7 @@ void isst_get_pbf_info_complete(struct isst_pbf_info *pbf_info) free_cpu_set(pbf_info->core_cpumask); } -int isst_set_pbf_fact_status(int cpu, int pbf, int enable) +int isst_set_pbf_fact_status(struct isst_id *id, int pbf, int enable) { struct isst_pkg_ctdp pkg_dev; struct isst_pkg_ctdp_level_info ctdp_level; @@ -510,13 +486,13 @@ int isst_set_pbf_fact_status(int cpu, int pbf, int enable) unsigned int req = 0, resp; int ret; - ret = isst_get_ctdp_levels(cpu, &pkg_dev); + ret = isst_get_ctdp_levels(id, &pkg_dev); if (ret) - debug_printf("cpu:%d No support for dynamic ISST\n", cpu); + debug_printf("cpu:%d No support for dynamic ISST\n", id->cpu); current_level = pkg_dev.current_level; - ret = isst_get_ctdp_control(cpu, current_level, &ctdp_level); + ret = isst_get_ctdp_control(id, current_level, &ctdp_level); if (ret) return ret; @@ -542,18 +518,18 @@ int isst_set_pbf_fact_status(int cpu, int pbf, int enable) req &= ~BIT(16); } - ret = isst_send_mbox_command(cpu, CONFIG_TDP, + ret = isst_send_mbox_command(id->cpu, CONFIG_TDP, CONFIG_TDP_SET_TDP_CONTROL, 0, req, &resp); if (ret) return ret; debug_printf("cpu:%d CONFIG_TDP_SET_TDP_CONTROL pbf/fact:%d req:%x\n", - cpu, pbf, req); + id->cpu, pbf, req); return 0; } -int isst_get_fact_bucket_info(int cpu, int level, +int isst_get_fact_bucket_info(struct isst_id *id, int level, struct isst_fact_bucket_info *bucket_info) { unsigned int resp; @@ -563,7 +539,7 @@ int isst_get_fact_bucket_info(int cpu, int level, int j; ret = isst_send_mbox_command( - cpu, CONFIG_TDP, + id->cpu, CONFIG_TDP, CONFIG_TDP_GET_FACT_HP_TURBO_LIMIT_NUMCORES, 0, (i << 8) | level, &resp); if (ret) @@ -571,7 +547,7 @@ int isst_get_fact_bucket_info(int cpu, int level, debug_printf( "cpu:%d CONFIG_TDP_GET_FACT_HP_TURBO_LIMIT_NUMCORES index:%d level:%d resp:%x\n", - cpu, i, level, resp); + id->cpu, i, level, resp); for (j = 0; j < 4; ++j) { bucket_info[j + (i * 4)].high_priority_cores_count = @@ -584,7 +560,7 @@ int isst_get_fact_bucket_info(int cpu, int level, int j; ret = isst_send_mbox_command( - cpu, CONFIG_TDP, + id->cpu, CONFIG_TDP, CONFIG_TDP_GET_FACT_HP_TURBO_LIMIT_RATIOS, 0, (k << 16) | (i << 8) | level, &resp); if (ret) @@ -592,7 +568,7 @@ int isst_get_fact_bucket_info(int cpu, int level, debug_printf( "cpu:%d CONFIG_TDP_GET_FACT_HP_TURBO_LIMIT_RATIOS index:%d level:%d avx:%d resp:%x\n", - cpu, i, level, k, resp); + id->cpu, i, level, k, resp); for (j = 0; j < 4; ++j) { switch (k) { @@ -618,14 +594,14 @@ int isst_get_fact_bucket_info(int cpu, int level, return 0; } -int isst_get_fact_info(int cpu, int level, int fact_bucket, struct isst_fact_info *fact_info) +int isst_get_fact_info(struct isst_id *id, int level, int fact_bucket, struct isst_fact_info *fact_info) { struct isst_pkg_ctdp_level_info ctdp_level; struct isst_pkg_ctdp pkg_dev; unsigned int resp; int j, ret, print; - ret = isst_get_ctdp_levels(cpu, &pkg_dev); + ret = isst_get_ctdp_levels(id, &pkg_dev); if (ret) { isst_display_error_info_message(1, "Failed to get number of levels", 0, 0); return ret; @@ -636,7 +612,7 @@ int isst_get_fact_info(int cpu, int level, int fact_bucket, struct isst_fact_inf return -1; } - ret = isst_get_ctdp_control(cpu, level, &ctdp_level); + ret = isst_get_ctdp_control(id, level, &ctdp_level); if (ret) return ret; @@ -645,20 +621,20 @@ int isst_get_fact_info(int cpu, int level, int fact_bucket, struct isst_fact_inf return -1; } - ret = isst_send_mbox_command(cpu, CONFIG_TDP, + ret = isst_send_mbox_command(id->cpu, CONFIG_TDP, CONFIG_TDP_GET_FACT_LP_CLIPPING_RATIO, 0, level, &resp); if (ret) return ret; debug_printf("cpu:%d CONFIG_TDP_GET_FACT_LP_CLIPPING_RATIO resp:%x\n", - cpu, resp); + id->cpu, resp); fact_info->lp_clipping_ratio_license_sse = resp & 0xff; fact_info->lp_clipping_ratio_license_avx2 = (resp >> 8) & 0xff; fact_info->lp_clipping_ratio_license_avx512 = (resp >> 16) & 0xff; - ret = isst_get_fact_bucket_info(cpu, level, fact_info->bucket_info); + ret = isst_get_fact_bucket_info(id, level, fact_info->bucket_info); if (ret) return ret; @@ -680,32 +656,32 @@ int isst_get_fact_info(int cpu, int level, int fact_bucket, struct isst_fact_inf return 0; } -int isst_get_trl(int cpu, unsigned long long *trl) +int isst_get_trl(struct isst_id *id, unsigned long long *trl) { int ret; - ret = isst_send_msr_command(cpu, 0x1AD, 0, trl); + ret = isst_send_msr_command(id->cpu, 0x1AD, 0, trl); if (ret) return ret; return 0; } -int isst_set_trl(int cpu, unsigned long long trl) +int isst_set_trl(struct isst_id *id, unsigned long long trl) { int ret; if (!trl) trl = 0xFFFFFFFFFFFFFFFFULL; - ret = isst_send_msr_command(cpu, 0x1AD, 1, &trl); + ret = isst_send_msr_command(id->cpu, 0x1AD, 1, &trl); if (ret) return ret; return 0; } -int isst_set_trl_from_current_tdp(int cpu, unsigned long long trl) +int isst_set_trl_from_current_tdp(struct isst_id *id, unsigned long long trl) { unsigned long long msr_trl; int ret; @@ -717,11 +693,11 @@ int isst_set_trl_from_current_tdp(int cpu, unsigned long long trl) int trl[8]; int i; - ret = isst_get_ctdp_levels(cpu, &pkg_dev); + ret = isst_get_ctdp_levels(id, &pkg_dev); if (ret) return ret; - ret = isst_get_get_trl(cpu, pkg_dev.current_level, 0, trl); + ret = isst_get_get_trl(id, pkg_dev.current_level, 0, trl); if (ret) return ret; @@ -732,7 +708,7 @@ int isst_set_trl_from_current_tdp(int cpu, unsigned long long trl) msr_trl |= (_trl << (i * 8)); } } - ret = isst_send_msr_command(cpu, 0x1AD, 1, &msr_trl); + ret = isst_send_msr_command(id->cpu, 0x1AD, 1, &msr_trl); if (ret) return ret; @@ -740,12 +716,12 @@ int isst_set_trl_from_current_tdp(int cpu, unsigned long long trl) } /* Return 1 if locked */ -int isst_get_config_tdp_lock_status(int cpu) +int isst_get_config_tdp_lock_status(struct isst_id *id) { unsigned long long tdp_control = 0; int ret; - ret = isst_send_msr_command(cpu, 0x64b, 0, &tdp_control); + ret = isst_send_msr_command(id->cpu, 0x64b, 0, &tdp_control); if (ret) return ret; @@ -754,7 +730,7 @@ int isst_get_config_tdp_lock_status(int cpu) return ret; } -void isst_get_process_ctdp_complete(int cpu, struct isst_pkg_ctdp *pkg_dev) +void isst_get_process_ctdp_complete(struct isst_id *id, struct isst_pkg_ctdp *pkg_dev) { int i; @@ -771,19 +747,19 @@ void isst_get_process_ctdp_complete(int cpu, struct isst_pkg_ctdp *pkg_dev) } } -int isst_get_process_ctdp(int cpu, int tdp_level, struct isst_pkg_ctdp *pkg_dev) +int isst_get_process_ctdp(struct isst_id *id, int tdp_level, struct isst_pkg_ctdp *pkg_dev) { int i, ret, valid = 0; if (pkg_dev->processed) return 0; - ret = isst_get_ctdp_levels(cpu, pkg_dev); + ret = isst_get_ctdp_levels(id, pkg_dev); if (ret) return ret; debug_printf("cpu: %d ctdp enable:%d current level: %d levels:%d\n", - cpu, pkg_dev->enabled, pkg_dev->current_level, + id->cpu, pkg_dev->enabled, pkg_dev->current_level, pkg_dev->levels); if (tdp_level != 0xff && tdp_level > pkg_dev->levels) { @@ -800,16 +776,16 @@ int isst_get_process_ctdp(int cpu, int tdp_level, struct isst_pkg_ctdp *pkg_dev) if (tdp_level != 0xff && i != tdp_level) continue; - debug_printf("cpu:%d Get Information for TDP level:%d\n", cpu, + debug_printf("cpu:%d Get Information for TDP level:%d\n", id->cpu, i); ctdp_level = &pkg_dev->ctdp_level[i]; ctdp_level->level = i; - ctdp_level->control_cpu = cpu; - ctdp_level->pkg_id = get_physical_package_id(cpu); - ctdp_level->die_id = get_physical_die_id(cpu); + ctdp_level->control_cpu = id->cpu; + ctdp_level->pkg_id = id->pkg; + ctdp_level->die_id = id->die; - ret = isst_get_ctdp_control(cpu, i, ctdp_level); + ret = isst_get_ctdp_control(id, i, ctdp_level); if (ret) continue; @@ -818,13 +794,13 @@ int isst_get_process_ctdp(int cpu, int tdp_level, struct isst_pkg_ctdp *pkg_dev) ctdp_level->processed = 1; if (ctdp_level->pbf_support) { - ret = isst_get_pbf_info(cpu, i, &ctdp_level->pbf_info); + ret = isst_get_pbf_info(id, i, &ctdp_level->pbf_info); if (!ret) ctdp_level->pbf_found = 1; } if (ctdp_level->fact_support) { - ret = isst_get_fact_info(cpu, i, 0xff, + ret = isst_get_fact_info(id, i, 0xff, &ctdp_level->fact_info); if (ret) return ret; @@ -833,76 +809,76 @@ int isst_get_process_ctdp(int cpu, int tdp_level, struct isst_pkg_ctdp *pkg_dev) if (!pkg_dev->enabled && is_skx_based_platform()) { int freq; - freq = get_cpufreq_base_freq(cpu); + freq = get_cpufreq_base_freq(id->cpu); if (freq > 0) { ctdp_level->sse_p1 = freq / 100000; ctdp_level->tdp_ratio = ctdp_level->sse_p1; } - isst_get_get_trl_from_msr(cpu, ctdp_level->trl_sse_active_cores); - isst_get_trl_bucket_info(cpu, &ctdp_level->buckets_info); + isst_get_get_trl_from_msr(id, ctdp_level->trl_sse_active_cores); + isst_get_trl_bucket_info(id, &ctdp_level->buckets_info); continue; } - ret = isst_get_tdp_info(cpu, i, ctdp_level); + ret = isst_get_tdp_info(id, i, ctdp_level); if (ret) return ret; - ret = isst_get_pwr_info(cpu, i, ctdp_level); + ret = isst_get_pwr_info(id, i, ctdp_level); if (ret) return ret; - ret = isst_get_tjmax_info(cpu, i, ctdp_level); + ret = isst_get_tjmax_info(id, i, ctdp_level); if (ret) return ret; ctdp_level->core_cpumask_size = alloc_cpu_set(&ctdp_level->core_cpumask); - ret = isst_get_coremask_info(cpu, i, ctdp_level); + ret = isst_get_coremask_info(id, i, ctdp_level); if (ret) return ret; - ret = isst_get_trl_bucket_info(cpu, &ctdp_level->buckets_info); + ret = isst_get_trl_bucket_info(id, &ctdp_level->buckets_info); if (ret) return ret; - ret = isst_get_get_trl(cpu, i, 0, + ret = isst_get_get_trl(id, i, 0, ctdp_level->trl_sse_active_cores); if (ret) return ret; - ret = isst_get_get_trl(cpu, i, 1, + ret = isst_get_get_trl(id, i, 1, ctdp_level->trl_avx_active_cores); if (ret) return ret; - ret = isst_get_get_trl(cpu, i, 2, + ret = isst_get_get_trl(id, i, 2, ctdp_level->trl_avx_512_active_cores); if (ret) return ret; - isst_get_uncore_p0_p1_info(cpu, i, ctdp_level); - isst_get_p1_info(cpu, i, ctdp_level); - isst_get_uncore_mem_freq(cpu, i, ctdp_level); + isst_get_uncore_p0_p1_info(id, i, ctdp_level); + isst_get_p1_info(id, i, ctdp_level); + isst_get_uncore_mem_freq(id, i, ctdp_level); } if (!valid) - isst_display_error_info_message(0, "Invalid level, Can't get TDP control information at specified levels on cpu", 1, cpu); + isst_display_error_info_message(0, "Invalid level, Can't get TDP control information at specified levels on cpu", 1, id->cpu); return 0; } -int isst_clos_get_clos_information(int cpu, int *enable, int *type) +int isst_clos_get_clos_information(struct isst_id *id, int *enable, int *type) { unsigned int resp; int ret; - ret = isst_send_mbox_command(cpu, CONFIG_CLOS, CLOS_PM_QOS_CONFIG, 0, 0, + ret = isst_send_mbox_command(id->cpu, CONFIG_CLOS, CLOS_PM_QOS_CONFIG, 0, 0, &resp); if (ret) return ret; - debug_printf("cpu:%d CLOS_PM_QOS_CONFIG resp:%x\n", cpu, resp); + debug_printf("cpu:%d CLOS_PM_QOS_CONFIG resp:%x\n", id->cpu, resp); if (resp & BIT(1)) *enable = 1; @@ -917,7 +893,7 @@ int isst_clos_get_clos_information(int cpu, int *enable, int *type) return 0; } -int isst_pm_qos_config(int cpu, int enable_clos, int priority_type) +int isst_pm_qos_config(struct isst_id *id, int enable_clos, int priority_type) { unsigned int req, resp; int ret; @@ -926,13 +902,13 @@ int isst_pm_qos_config(int cpu, int enable_clos, int priority_type) struct isst_pkg_ctdp pkg_dev; struct isst_pkg_ctdp_level_info ctdp_level; - ret = isst_get_ctdp_levels(cpu, &pkg_dev); + ret = isst_get_ctdp_levels(id, &pkg_dev); if (ret) { debug_printf("isst_get_ctdp_levels\n"); return ret; } - ret = isst_get_ctdp_control(cpu, pkg_dev.current_level, + ret = isst_get_ctdp_control(id, pkg_dev.current_level, &ctdp_level); if (ret) return ret; @@ -941,23 +917,23 @@ int isst_pm_qos_config(int cpu, int enable_clos, int priority_type) isst_display_error_info_message(1, "Ignoring request, turbo-freq feature is still enabled", 0, 0); return -EINVAL; } - ret = isst_write_pm_config(cpu, 0); + ret = isst_write_pm_config(id, 0); if (ret) isst_display_error_info_message(0, "WRITE_PM_CONFIG command failed, ignoring error", 0, 0); } else { - ret = isst_write_pm_config(cpu, 1); + ret = isst_write_pm_config(id, 1); if (ret) isst_display_error_info_message(0, "WRITE_PM_CONFIG command failed, ignoring error", 0, 0); } - ret = isst_send_mbox_command(cpu, CONFIG_CLOS, CLOS_PM_QOS_CONFIG, 0, 0, + ret = isst_send_mbox_command(id->cpu, CONFIG_CLOS, CLOS_PM_QOS_CONFIG, 0, 0, &resp); if (ret) { isst_display_error_info_message(1, "CLOS_PM_QOS_CONFIG command failed", 0, 0); return ret; } - debug_printf("cpu:%d CLOS_PM_QOS_CONFIG resp:%x\n", cpu, resp); + debug_printf("cpu:%d CLOS_PM_QOS_CONFIG resp:%x\n", id->cpu, resp); req = resp; @@ -974,30 +950,27 @@ int isst_pm_qos_config(int cpu, int enable_clos, int priority_type) else req = req & ~BIT(2); - ret = isst_send_mbox_command(cpu, CONFIG_CLOS, CLOS_PM_QOS_CONFIG, + ret = isst_send_mbox_command(id->cpu, CONFIG_CLOS, CLOS_PM_QOS_CONFIG, BIT(MBOX_CMD_WRITE_BIT), req, &resp); if (ret) return ret; - debug_printf("cpu:%d CLOS_PM_QOS_CONFIG priority type:%d req:%x\n", cpu, + debug_printf("cpu:%d CLOS_PM_QOS_CONFIG priority type:%d req:%x\n", id->cpu, priority_type, req); return 0; } -int isst_pm_get_clos(int cpu, int clos, struct isst_clos_config *clos_config) +int isst_pm_get_clos(struct isst_id *id, int clos, struct isst_clos_config *clos_config) { unsigned int resp; int ret; - ret = isst_send_mbox_command(cpu, CONFIG_CLOS, CLOS_PM_CLOS, clos, 0, + ret = isst_send_mbox_command(id->cpu, CONFIG_CLOS, CLOS_PM_CLOS, clos, 0, &resp); if (ret) return ret; - clos_config->pkg_id = get_physical_package_id(cpu); - clos_config->die_id = get_physical_die_id(cpu); - clos_config->epp = resp & 0x0f; clos_config->clos_prop_prio = (resp >> 4) & 0x0f; clos_config->clos_min = (resp >> 8) & 0xff; @@ -1007,7 +980,7 @@ int isst_pm_get_clos(int cpu, int clos, struct isst_clos_config *clos_config) return 0; } -int isst_set_clos(int cpu, int clos, struct isst_clos_config *clos_config) +int isst_set_clos(struct isst_id *id, int clos, struct isst_clos_config *clos_config) { unsigned int req, resp; unsigned int param; @@ -1021,53 +994,53 @@ int isst_set_clos(int cpu, int clos, struct isst_clos_config *clos_config) param = BIT(MBOX_CMD_WRITE_BIT) | clos; - ret = isst_send_mbox_command(cpu, CONFIG_CLOS, CLOS_PM_CLOS, param, req, + ret = isst_send_mbox_command(id->cpu, CONFIG_CLOS, CLOS_PM_CLOS, param, req, &resp); if (ret) return ret; - debug_printf("cpu:%d CLOS_PM_CLOS param:%x req:%x\n", cpu, param, req); + debug_printf("cpu:%d CLOS_PM_CLOS param:%x req:%x\n", id->cpu, param, req); return 0; } -int isst_clos_get_assoc_status(int cpu, int *clos_id) +int isst_clos_get_assoc_status(struct isst_id *id, int *clos_id) { unsigned int resp; unsigned int param; int core_id, ret; - core_id = find_phy_core_num(cpu); + core_id = find_phy_core_num(id->cpu); param = core_id; - ret = isst_send_mbox_command(cpu, CONFIG_CLOS, CLOS_PQR_ASSOC, param, 0, + ret = isst_send_mbox_command(id->cpu, CONFIG_CLOS, CLOS_PQR_ASSOC, param, 0, &resp); if (ret) return ret; - debug_printf("cpu:%d CLOS_PQR_ASSOC param:%x resp:%x\n", cpu, param, + debug_printf("cpu:%d CLOS_PQR_ASSOC param:%x resp:%x\n", id->cpu, param, resp); *clos_id = (resp >> 16) & 0x03; return 0; } -int isst_clos_associate(int cpu, int clos_id) +int isst_clos_associate(struct isst_id *id, int clos_id) { unsigned int req, resp; unsigned int param; int core_id, ret; req = (clos_id & 0x03) << 16; - core_id = find_phy_core_num(cpu); + core_id = find_phy_core_num(id->cpu); param = BIT(MBOX_CMD_WRITE_BIT) | core_id; - ret = isst_send_mbox_command(cpu, CONFIG_CLOS, CLOS_PQR_ASSOC, param, + ret = isst_send_mbox_command(id->cpu, CONFIG_CLOS, CLOS_PQR_ASSOC, param, req, &resp); if (ret) return ret; - debug_printf("cpu:%d CLOS_PQR_ASSOC param:%x req:%x\n", cpu, param, + debug_printf("cpu:%d CLOS_PQR_ASSOC param:%x req:%x\n", id->cpu, param, req); return 0; diff --git a/tools/power/x86/intel-speed-select/isst-daemon.c b/tools/power/x86/intel-speed-select/isst-daemon.c new file mode 100644 index 000000000000..0699137c0901 --- /dev/null +++ b/tools/power/x86/intel-speed-select/isst-daemon.c @@ -0,0 +1,242 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Intel Speed Select -- Allow speed select to daemonize + * Copyright (c) 2022 Intel Corporation. + */ + +#include <stdio.h> +#include <stdlib.h> +#include <stdarg.h> +#include <string.h> +#include <unistd.h> +#include <fcntl.h> +#include <sys/file.h> +#include <sys/types.h> +#include <sys/stat.h> +#include <errno.h> +#include <getopt.h> +#include <signal.h> +#include <time.h> + +#include "isst.h" + +static int per_package_levels_info[MAX_PACKAGE_COUNT][MAX_DIE_PER_PACKAGE]; +static time_t per_package_levels_tm[MAX_PACKAGE_COUNT][MAX_DIE_PER_PACKAGE]; + +static void init_levels(void) +{ + int i, j; + + for (i = 0; i < MAX_PACKAGE_COUNT; ++i) + for (j = 0; j < MAX_DIE_PER_PACKAGE; ++j) + per_package_levels_info[i][j] = -1; +} + +void process_level_change(struct isst_id *id) +{ + struct isst_pkg_ctdp_level_info ctdp_level; + struct isst_pkg_ctdp pkg_dev; + time_t tm; + int ret; + + if (id->pkg < 0 || id->die < 0) { + debug_printf("Invalid package/die info for cpu:%d\n", id->cpu); + return; + } + + tm = time(NULL); + if (tm - per_package_levels_tm[id->pkg][id->die] < 2) + return; + + per_package_levels_tm[id->pkg][id->die] = tm; + + ret = isst_get_ctdp_levels(id, &pkg_dev); + if (ret) { + debug_printf("Can't get tdp levels for cpu:%d\n", id->cpu); + return; + } + + debug_printf("Get Config level %d pkg:%d die:%d current_level:%d\n", id->cpu, + id->pkg, id->die, pkg_dev.current_level); + + if (pkg_dev.locked) { + debug_printf("config TDP s locked \n"); + return; + } + + if (per_package_levels_info[id->pkg][id->die] == pkg_dev.current_level) + return; + + debug_printf("**Config level change for cpu:%d pkg:%d die:%d from %d to %d\n", + id->cpu, id->pkg, id->die, per_package_levels_info[id->pkg][id->die], + pkg_dev.current_level); + + per_package_levels_info[id->pkg][id->die] = pkg_dev.current_level; + + ctdp_level.core_cpumask_size = + alloc_cpu_set(&ctdp_level.core_cpumask); + ret = isst_get_coremask_info(id, pkg_dev.current_level, &ctdp_level); + if (ret) { + free_cpu_set(ctdp_level.core_cpumask); + debug_printf("Can't get core_mask:%d\n", id->cpu); + return; + } + + if (ctdp_level.cpu_count) { + int i, max_cpus = get_topo_max_cpus(); + for (i = 0; i < max_cpus; ++i) { + if (!is_cpu_in_power_domain(i, id)) + continue; + if (CPU_ISSET_S(i, ctdp_level.core_cpumask_size, ctdp_level.core_cpumask)) { + fprintf(stderr, "online cpu %d\n", i); + set_cpu_online_offline(i, 1); + } else { + fprintf(stderr, "offline cpu %d\n", i); + set_cpu_online_offline(i, 0); + } + } + } + + free_cpu_set(ctdp_level.core_cpumask); +} + +static void _poll_for_config_change(struct isst_id *id, void *arg1, void *arg2, + void *arg3, void *arg4) +{ + process_level_change(id); +} + +static void poll_for_config_change(void) +{ + for_each_online_package_in_set(_poll_for_config_change, NULL, NULL, + NULL, NULL); +} + +static int done = 0; +static int pid_file_handle; + +static void signal_handler(int sig) +{ + switch (sig) { + case SIGINT: + case SIGTERM: + done = 1; + hfi_exit(); + exit(0); + break; + default: + break; + } +} + +static void daemonize(char *rundir, char *pidfile) +{ + int pid, sid, i; + char str[10]; + struct sigaction sig_actions; + sigset_t sig_set; + int ret; + + if (getppid() == 1) + return; + + sigemptyset(&sig_set); + sigaddset(&sig_set, SIGCHLD); + sigaddset(&sig_set, SIGTSTP); + sigaddset(&sig_set, SIGTTOU); + sigaddset(&sig_set, SIGTTIN); + sigprocmask(SIG_BLOCK, &sig_set, NULL); + + sig_actions.sa_handler = signal_handler; + sigemptyset(&sig_actions.sa_mask); + sig_actions.sa_flags = 0; + + sigaction(SIGHUP, &sig_actions, NULL); + sigaction(SIGTERM, &sig_actions, NULL); + sigaction(SIGINT, &sig_actions, NULL); + + pid = fork(); + if (pid < 0) { + /* Could not fork */ + exit(EXIT_FAILURE); + } + if (pid > 0) + exit(EXIT_SUCCESS); + + umask(027); + + sid = setsid(); + if (sid < 0) + exit(EXIT_FAILURE); + + /* close all descriptors */ + for (i = getdtablesize(); i >= 0; --i) + close(i); + + i = open("/dev/null", O_RDWR); + ret = dup(i); + if (ret == -1) + exit(EXIT_FAILURE); + + ret = dup(i); + if (ret == -1) + exit(EXIT_FAILURE); + + ret = chdir(rundir); + if (ret == -1) + exit(EXIT_FAILURE); + + pid_file_handle = open(pidfile, O_RDWR | O_CREAT, 0600); + if (pid_file_handle == -1) { + /* Couldn't open lock file */ + exit(1); + } + /* Try to lock file */ +#ifdef LOCKF_SUPPORT + if (lockf(pid_file_handle, F_TLOCK, 0) == -1) { +#else + if (flock(pid_file_handle, LOCK_EX|LOCK_NB) < 0) { +#endif + /* Couldn't get lock on lock file */ + fprintf(stderr, "Couldn't get lock file %d\n", getpid()); + exit(1); + } + snprintf(str, sizeof(str), "%d\n", getpid()); + ret = write(pid_file_handle, str, strlen(str)); + if (ret == -1) + exit(EXIT_FAILURE); + + close(i); +} + +int isst_daemon(int debug_mode, int poll_interval, int no_daemon) +{ + int ret; + + if (!no_daemon && poll_interval < 0 && !debug_mode) { + fprintf(stderr, "OOB mode is enabled and will run as daemon\n"); + daemonize((char *) "/tmp/", + (char *)"/tmp/hfi-events.pid"); + } else { + signal(SIGINT, signal_handler); + } + + init_levels(); + + if (poll_interval < 0) { + ret = hfi_main(); + if (ret) { + fprintf(stderr, "HFI initialization failed\n"); + } + fprintf(stderr, "Must specify poll-interval\n"); + return ret; + } + + debug_printf("Starting loop\n"); + while (!done) { + sleep(poll_interval); + poll_for_config_change(); + } + + return 0; +} diff --git a/tools/power/x86/intel-speed-select/isst-display.c b/tools/power/x86/intel-speed-select/isst-display.c index f97d8859ada7..b19f57d30f55 100644 --- a/tools/power/x86/intel-speed-select/isst-display.c +++ b/tools/power/x86/intel-speed-select/isst-display.c @@ -166,29 +166,27 @@ static void format_and_print(FILE *outf, int level, char *header, char *value) last_level = level; } -static int print_package_info(int cpu, FILE *outf) +static int print_package_info(struct isst_id *id, FILE *outf) { char header[256]; if (out_format_is_json()) { snprintf(header, sizeof(header), "package-%d:die-%d:cpu-%d", - get_physical_package_id(cpu), get_physical_die_id(cpu), - cpu); + id->pkg, id->die, id->cpu); format_and_print(outf, 1, header, NULL); return 1; } - snprintf(header, sizeof(header), "package-%d", - get_physical_package_id(cpu)); + snprintf(header, sizeof(header), "package-%d", id->pkg); format_and_print(outf, 1, header, NULL); - snprintf(header, sizeof(header), "die-%d", get_physical_die_id(cpu)); + snprintf(header, sizeof(header), "die-%d", id->die); format_and_print(outf, 2, header, NULL); - snprintf(header, sizeof(header), "cpu-%d", cpu); + snprintf(header, sizeof(header), "cpu-%d", id->cpu); format_and_print(outf, 3, header, NULL); return 3; } -static void _isst_pbf_display_information(int cpu, FILE *outf, int level, +static void _isst_pbf_display_information(struct isst_id *id, FILE *outf, int level, struct isst_pbf_info *pbf_info, int disp_level) { @@ -231,7 +229,7 @@ static void _isst_pbf_display_information(int cpu, FILE *outf, int level, format_and_print(outf, disp_level + 1, header, value); } -static void _isst_fact_display_information(int cpu, FILE *outf, int level, +static void _isst_fact_display_information(struct isst_id *id, FILE *outf, int level, int fact_bucket, int fact_avx, struct isst_fact_info *fact_info, int base_level) @@ -319,7 +317,7 @@ static void _isst_fact_display_information(int cpu, FILE *outf, int level, format_and_print(outf, base_level + 2, header, value); } -void isst_ctdp_display_core_info(int cpu, FILE *outf, char *prefix, +void isst_ctdp_display_core_info(struct isst_id *id, FILE *outf, char *prefix, unsigned int val, char *str0, char *str1) { char header[256]; @@ -328,17 +326,14 @@ void isst_ctdp_display_core_info(int cpu, FILE *outf, char *prefix, if (out_format_is_json()) { snprintf(header, sizeof(header), "package-%d:die-%d:cpu-%d", - get_physical_package_id(cpu), get_physical_die_id(cpu), - cpu); + id->pkg, id->die, id->cpu); format_and_print(outf, level++, header, NULL); } else { - snprintf(header, sizeof(header), "package-%d", - get_physical_package_id(cpu)); + snprintf(header, sizeof(header), "package-%d", id->pkg); format_and_print(outf, level++, header, NULL); - snprintf(header, sizeof(header), "die-%d", - get_physical_die_id(cpu)); + snprintf(header, sizeof(header), "die-%d", id->die); format_and_print(outf, level++, header, NULL); - snprintf(header, sizeof(header), "cpu-%d", cpu); + snprintf(header, sizeof(header), "cpu-%d", id->cpu); format_and_print(outf, level++, header, NULL); } @@ -353,7 +348,7 @@ void isst_ctdp_display_core_info(int cpu, FILE *outf, char *prefix, format_and_print(outf, 1, NULL, NULL); } -void isst_ctdp_display_information(int cpu, FILE *outf, int tdp_level, +void isst_ctdp_display_information(struct isst_id *id, FILE *outf, int tdp_level, struct isst_pkg_ctdp *pkg_dev) { char header[256]; @@ -362,7 +357,7 @@ void isst_ctdp_display_information(int cpu, FILE *outf, int tdp_level, int i; if (pkg_dev->processed) - level = print_package_info(cpu, outf); + level = print_package_info(id, outf); for (i = 0; i <= pkg_dev->levels; ++i) { struct isst_pkg_ctdp_level_info *ctdp_level; @@ -377,8 +372,7 @@ void isst_ctdp_display_information(int cpu, FILE *outf, int tdp_level, format_and_print(outf, level + 1, header, NULL); snprintf(header, sizeof(header), "cpu-count"); - j = get_cpu_count(get_physical_die_id(cpu), - get_physical_die_id(cpu)); + j = get_cpu_count(id); snprintf(value, sizeof(value), "%d", j); format_and_print(outf, level + 2, header, value); @@ -485,7 +479,7 @@ void isst_ctdp_display_information(int cpu, FILE *outf, int tdp_level, if (is_clx_n_platform()) { if (ctdp_level->pbf_support) - _isst_pbf_display_information(cpu, outf, + _isst_pbf_display_information(id, outf, tdp_level, &ctdp_level->pbf_info, level + 2); @@ -557,11 +551,11 @@ void isst_ctdp_display_information(int cpu, FILE *outf, int tdp_level, } if (ctdp_level->pbf_support) - _isst_pbf_display_information(cpu, outf, i, + _isst_pbf_display_information(id, outf, i, &ctdp_level->pbf_info, level + 2); if (ctdp_level->fact_support) - _isst_fact_display_information(cpu, outf, i, 0xff, 0xff, + _isst_fact_display_information(id, outf, i, 0xff, 0xff, &ctdp_level->fact_info, level + 2); } @@ -583,36 +577,36 @@ void isst_ctdp_display_information_end(FILE *outf) start = 0; } -void isst_pbf_display_information(int cpu, FILE *outf, int level, +void isst_pbf_display_information(struct isst_id *id, FILE *outf, int level, struct isst_pbf_info *pbf_info) { int _level; - _level = print_package_info(cpu, outf); - _isst_pbf_display_information(cpu, outf, level, pbf_info, _level + 1); + _level = print_package_info(id, outf); + _isst_pbf_display_information(id, outf, level, pbf_info, _level + 1); format_and_print(outf, 1, NULL, NULL); } -void isst_fact_display_information(int cpu, FILE *outf, int level, +void isst_fact_display_information(struct isst_id *id, FILE *outf, int level, int fact_bucket, int fact_avx, struct isst_fact_info *fact_info) { int _level; - _level = print_package_info(cpu, outf); - _isst_fact_display_information(cpu, outf, level, fact_bucket, fact_avx, + _level = print_package_info(id, outf); + _isst_fact_display_information(id, outf, level, fact_bucket, fact_avx, fact_info, _level + 1); format_and_print(outf, 1, NULL, NULL); } -void isst_clos_display_information(int cpu, FILE *outf, int clos, +void isst_clos_display_information(struct isst_id *id, FILE *outf, int clos, struct isst_clos_config *clos_config) { char header[256]; char value[256]; int level; - level = print_package_info(cpu, outf); + level = print_package_info(id, outf); snprintf(header, sizeof(header), "core-power"); format_and_print(outf, level + 1, header, NULL); @@ -647,7 +641,7 @@ void isst_clos_display_information(int cpu, FILE *outf, int clos, format_and_print(outf, level, NULL, NULL); } -void isst_clos_display_clos_information(int cpu, FILE *outf, +void isst_clos_display_clos_information(struct isst_id *id, FILE *outf, int clos_enable, int type, int state, int cap) { @@ -655,7 +649,7 @@ void isst_clos_display_clos_information(int cpu, FILE *outf, char value[256]; int level; - level = print_package_info(cpu, outf); + level = print_package_info(id, outf); snprintf(header, sizeof(header), "core-power"); format_and_print(outf, level + 1, header, NULL); @@ -691,13 +685,13 @@ void isst_clos_display_clos_information(int cpu, FILE *outf, format_and_print(outf, level, NULL, NULL); } -void isst_clos_display_assoc_information(int cpu, FILE *outf, int clos) +void isst_clos_display_assoc_information(struct isst_id *id, FILE *outf, int clos) { char header[256]; char value[256]; int level; - level = print_package_info(cpu, outf); + level = print_package_info(id, outf); snprintf(header, sizeof(header), "get-assoc"); format_and_print(outf, level + 1, header, NULL); @@ -709,15 +703,15 @@ void isst_clos_display_assoc_information(int cpu, FILE *outf, int clos) format_and_print(outf, level, NULL, NULL); } -void isst_display_result(int cpu, FILE *outf, char *feature, char *cmd, +void isst_display_result(struct isst_id *id, FILE *outf, char *feature, char *cmd, int result) { char header[256]; char value[256]; int level = 3; - if (cpu >= 0) - level = print_package_info(cpu, outf); + if (id->cpu >= 0) + level = print_package_info(id, outf); snprintf(header, sizeof(header), "%s", feature); format_and_print(outf, level + 1, header, NULL); @@ -772,13 +766,13 @@ void isst_display_error_info_message(int error, char *msg, int arg_valid, int ar format_and_print(outf, 0, NULL, NULL); } -void isst_trl_display_information(int cpu, FILE *outf, unsigned long long trl) +void isst_trl_display_information(struct isst_id *id, FILE *outf, unsigned long long trl) { char header[256]; char value[256]; int level; - level = print_package_info(cpu, outf); + level = print_package_info(id, outf); snprintf(header, sizeof(header), "get-trl"); format_and_print(outf, level + 1, header, NULL); diff --git a/tools/power/x86/intel-speed-select/isst.h b/tools/power/x86/intel-speed-select/isst.h index 1aa15d5ea57c..409fcc9c8033 100644 --- a/tools/power/x86/intel-speed-select/isst.h +++ b/tools/power/x86/intel-speed-select/isst.h @@ -76,9 +76,17 @@ #define DISP_FREQ_MULTIPLIER 100 +#define MAX_PACKAGE_COUNT 8 +#define MAX_DIE_PER_PACKAGE 2 + +/* Unified structure to specific a CPU or a Power Domain */ +struct isst_id { + int cpu; + int pkg; + int die; +}; + struct isst_clos_config { - int pkg_id; - int die_id; unsigned char epp; unsigned char clos_prop_prio; unsigned char clos_min; @@ -168,22 +176,20 @@ struct isst_pkg_ctdp { struct isst_pkg_ctdp_level_info ctdp_level[ISST_MAX_TDP_LEVELS]; }; +extern int is_cpu_in_power_domain(int cpu, struct isst_id *id); extern int get_topo_max_cpus(void); -extern int get_cpu_count(int pkg_id, int die_id); -extern int get_max_punit_core_id(int pkg_id, int die_id); +extern int get_cpu_count(struct isst_id *id); +extern int get_max_punit_core_id(struct isst_id *id); /* Common interfaces */ FILE *get_output_file(void); extern void debug_printf(const char *format, ...); extern int out_format_is_json(void); -extern int get_physical_package_id(int cpu); -extern int get_physical_die_id(int cpu); +extern void set_isst_id(struct isst_id *id, int cpu); extern size_t alloc_cpu_set(cpu_set_t **cpu_set); extern void free_cpu_set(cpu_set_t *cpu_set); -extern int find_logical_cpu(int pkg_id, int die_id, int phy_cpu); -extern int find_phy_cpu_num(int logical_cpu); extern int find_phy_core_num(int logical_cpu); -extern void set_cpu_mask_from_punit_coremask(int cpu, +extern void set_cpu_mask_from_punit_coremask(struct isst_id *id, unsigned long long core_mask, size_t core_cpumask_size, cpu_set_t *core_cpumask, @@ -197,67 +203,74 @@ extern int isst_send_mbox_command(unsigned int cpu, unsigned char command, extern int isst_send_msr_command(unsigned int cpu, unsigned int command, int write, unsigned long long *req_resp); -extern int isst_get_ctdp_levels(int cpu, struct isst_pkg_ctdp *pkg_dev); -extern int isst_get_ctdp_control(int cpu, int config_index, +extern int isst_get_ctdp_levels(struct isst_id *id, struct isst_pkg_ctdp *pkg_dev); +extern int isst_get_ctdp_control(struct isst_id *id, int config_index, struct isst_pkg_ctdp_level_info *ctdp_level); -extern int isst_get_coremask_info(int cpu, int config_index, +extern int isst_get_coremask_info(struct isst_id *id, int config_index, struct isst_pkg_ctdp_level_info *ctdp_level); -extern int isst_get_process_ctdp(int cpu, int tdp_level, +extern int isst_get_process_ctdp(struct isst_id *id, int tdp_level, struct isst_pkg_ctdp *pkg_dev); -extern void isst_get_process_ctdp_complete(int cpu, +extern void isst_get_process_ctdp_complete(struct isst_id *id, struct isst_pkg_ctdp *pkg_dev); -extern void isst_ctdp_display_information(int cpu, FILE *outf, int tdp_level, +extern void isst_ctdp_display_information(struct isst_id *id, FILE *outf, int tdp_level, struct isst_pkg_ctdp *pkg_dev); -extern void isst_ctdp_display_core_info(int cpu, FILE *outf, char *prefix, +extern void isst_ctdp_display_core_info(struct isst_id *id, FILE *outf, char *prefix, unsigned int val, char *str0, char *str1); extern void isst_ctdp_display_information_start(FILE *outf); extern void isst_ctdp_display_information_end(FILE *outf); -extern void isst_pbf_display_information(int cpu, FILE *outf, int level, +extern void isst_pbf_display_information(struct isst_id *id, FILE *outf, int level, struct isst_pbf_info *info); -extern int isst_set_tdp_level(int cpu, int tdp_level); -extern int isst_set_tdp_level_msr(int cpu, int tdp_level); -extern int isst_set_pbf_fact_status(int cpu, int pbf, int enable); -extern int isst_get_pbf_info(int cpu, int level, +extern int isst_set_tdp_level(struct isst_id *id, int tdp_level); +extern int isst_set_pbf_fact_status(struct isst_id *id, int pbf, int enable); +extern int isst_get_pbf_info(struct isst_id *id, int level, struct isst_pbf_info *pbf_info); extern void isst_get_pbf_info_complete(struct isst_pbf_info *pbf_info); -extern int isst_get_fact_info(int cpu, int level, int fact_bucket, +extern int isst_get_fact_info(struct isst_id *id, int level, int fact_bucket, struct isst_fact_info *fact_info); -extern int isst_get_fact_bucket_info(int cpu, int level, +extern int isst_get_fact_bucket_info(struct isst_id *id, int level, struct isst_fact_bucket_info *bucket_info); -extern void isst_fact_display_information(int cpu, FILE *outf, int level, +extern void isst_fact_display_information(struct isst_id *id, FILE *outf, int level, int fact_bucket, int fact_avx, struct isst_fact_info *fact_info); -extern int isst_set_trl(int cpu, unsigned long long trl); -extern int isst_get_trl(int cpu, unsigned long long *trl); -extern int isst_set_trl_from_current_tdp(int cpu, unsigned long long trl); -extern int isst_get_config_tdp_lock_status(int cpu); +extern int isst_set_trl(struct isst_id *id, unsigned long long trl); +extern int isst_get_trl(struct isst_id *id, unsigned long long *trl); +extern int isst_set_trl_from_current_tdp(struct isst_id *id, unsigned long long trl); +extern int isst_get_config_tdp_lock_status(struct isst_id *id); -extern int isst_pm_qos_config(int cpu, int enable_clos, int priority_type); -extern int isst_pm_get_clos(int cpu, int clos, +extern int isst_pm_qos_config(struct isst_id *id, int enable_clos, int priority_type); +extern int isst_pm_get_clos(struct isst_id *id, int clos, struct isst_clos_config *clos_config); -extern int isst_set_clos(int cpu, int clos, +extern int isst_set_clos(struct isst_id *id, int clos, struct isst_clos_config *clos_config); -extern int isst_clos_associate(int cpu, int clos); -extern int isst_clos_get_assoc_status(int cpu, int *clos_id); -extern void isst_clos_display_information(int cpu, FILE *outf, int clos, +extern int isst_clos_associate(struct isst_id *id, int clos); +extern int isst_clos_get_assoc_status(struct isst_id *id, int *clos_id); +extern void isst_clos_display_information(struct isst_id *id, FILE *outf, int clos, struct isst_clos_config *clos_config); -extern void isst_clos_display_assoc_information(int cpu, FILE *outf, int clos); -extern int isst_read_reg(unsigned short reg, unsigned int *val); -extern int isst_write_reg(int reg, unsigned int val); +extern void isst_clos_display_assoc_information(struct isst_id *id, FILE *outf, int clos); -extern void isst_display_result(int cpu, FILE *outf, char *feature, char *cmd, +extern void isst_display_result(struct isst_id *id, FILE *outf, char *feature, char *cmd, int result); -extern int isst_clos_get_clos_information(int cpu, int *enable, int *type); -extern void isst_clos_display_clos_information(int cpu, FILE *outf, +extern int isst_clos_get_clos_information(struct isst_id *id, int *enable, int *type); +extern void isst_clos_display_clos_information(struct isst_id *id, FILE *outf, int clos_enable, int type, int state, int cap); extern int is_clx_n_platform(void); extern int get_cpufreq_base_freq(int cpu); -extern int isst_read_pm_config(int cpu, int *cp_state, int *cp_cap); +extern int isst_read_pm_config(struct isst_id *id, int *cp_state, int *cp_cap); extern void isst_display_error_info_message(int error, char *msg, int arg_valid, int arg); extern int is_skx_based_platform(void); extern int is_spr_platform(void); extern int is_icx_platform(void); -extern void isst_trl_display_information(int cpu, FILE *outf, unsigned long long trl); +extern void isst_trl_display_information(struct isst_id *id, FILE *outf, unsigned long long trl); + +extern void set_cpu_online_offline(int cpu, int state); +extern void for_each_online_package_in_set(void (*callback)(struct isst_id *, void *, void *, + void *, void *), + void *arg1, void *arg2, void *arg3, + void *arg4); +extern int isst_daemon(int debug_mode, int poll_interval, int no_daemon); +extern void process_level_change(struct isst_id *id); +extern int hfi_main(void); +extern void hfi_exit(void); #endif diff --git a/tools/power/x86/intel_pstate_tracer/intel_pstate_tracer.py b/tools/power/x86/intel_pstate_tracer/intel_pstate_tracer.py index e15e20696d17..b46e9eb8f5aa 100755 --- a/tools/power/x86/intel_pstate_tracer/intel_pstate_tracer.py +++ b/tools/power/x86/intel_pstate_tracer/intel_pstate_tracer.py @@ -63,7 +63,7 @@ C_USEC = 3 C_SEC = 2 C_CPU = 1 -global sample_num, last_sec_cpu, last_usec_cpu, start_time, testname +global sample_num, last_sec_cpu, last_usec_cpu, start_time, testname, trace_file # 11 digits covers uptime to 115 days getcontext().prec = 11 @@ -72,17 +72,17 @@ sample_num =0 last_sec_cpu = [0] * MAX_CPUS last_usec_cpu = [0] * MAX_CPUS -def print_help(): - print('intel_pstate_tracer.py:') +def print_help(driver_name): + print('%s_tracer.py:'%driver_name) print(' Usage:') print(' If the trace file is available, then to simply parse and plot, use (sudo not required):') - print(' ./intel_pstate_tracer.py [-c cpus] -t <trace_file> -n <test_name>') + print(' ./%s_tracer.py [-c cpus] -t <trace_file> -n <test_name>'%driver_name) print(' Or') - print(' ./intel_pstate_tracer.py [--cpu cpus] ---trace_file <trace_file> --name <test_name>') + print(' ./%s_tracer.py [--cpu cpus] ---trace_file <trace_file> --name <test_name>'%driver_name) print(' To generate trace file, parse and plot, use (sudo required):') - print(' sudo ./intel_pstate_tracer.py [-c cpus] -i <interval> -n <test_name> -m <kbytes>') + print(' sudo ./%s_tracer.py [-c cpus] -i <interval> -n <test_name> -m <kbytes>'%driver_name) print(' Or') - print(' sudo ./intel_pstate_tracer.py [--cpu cpus] --interval <interval> --name <test_name> --memory <kbytes>') + print(' sudo ./%s_tracer.py [--cpu cpus] --interval <interval> --name <test_name> --memory <kbytes>'%driver_name) print(' Optional argument:') print(' cpus: comma separated list of CPUs') print(' kbytes: Kilo bytes of memory per CPU to allocate to the trace buffer. Default: 10240') @@ -323,7 +323,7 @@ def set_4_plot_linestyles(g_plot): g_plot('set style line 3 linetype 1 linecolor rgb "purple" pointtype -1') g_plot('set style line 4 linetype 1 linecolor rgb "blue" pointtype -1') -def store_csv(cpu_int, time_pre_dec, time_post_dec, core_busy, scaled, _from, _to, mperf, aperf, tsc, freq_ghz, io_boost, common_comm, load, duration_ms, sample_num, elapsed_time, tsc_ghz): +def store_csv(cpu_int, time_pre_dec, time_post_dec, core_busy, scaled, _from, _to, mperf, aperf, tsc, freq_ghz, io_boost, common_comm, load, duration_ms, sample_num, elapsed_time, tsc_ghz, cpu_mask): """ Store master csv file information """ global graph_data_present @@ -342,11 +342,9 @@ def store_csv(cpu_int, time_pre_dec, time_post_dec, core_busy, scaled, _from, _t graph_data_present = True; -def split_csv(): +def split_csv(current_max_cpu, cpu_mask): """ seperate the all csv file into per CPU csv files. """ - global current_max_cpu - if os.path.exists('cpu.csv'): for index in range(0, current_max_cpu + 1): if cpu_mask[int(index)] != 0: @@ -381,27 +379,25 @@ def clear_trace_file(): print('IO error clearing trace file ') sys.exit(2) -def enable_trace(): +def enable_trace(trace_file): """ Enable trace """ try: - open('/sys/kernel/debug/tracing/events/power/pstate_sample/enable' - , 'w').write("1") + open(trace_file,'w').write("1") except: print('IO error enabling trace ') sys.exit(2) -def disable_trace(): +def disable_trace(trace_file): """ Disable trace """ try: - open('/sys/kernel/debug/tracing/events/power/pstate_sample/enable' - , 'w').write("0") + open(trace_file, 'w').write("0") except: print('IO error disabling trace ') sys.exit(2) -def set_trace_buffer_size(): +def set_trace_buffer_size(memory): """ Set trace buffer size """ try: @@ -421,7 +417,7 @@ def free_trace_buffer(): print('IO error freeing trace buffer ') sys.exit(2) -def read_trace_data(filename): +def read_trace_data(filename, cpu_mask): """ Read and parse trace data """ global current_max_cpu @@ -481,135 +477,137 @@ def read_trace_data(filename): tsc_ghz = Decimal(0) if duration_ms != Decimal(0) : tsc_ghz = Decimal(tsc)/duration_ms/Decimal(1000000) - store_csv(cpu_int, time_pre_dec, time_post_dec, core_busy, scaled, _from, _to, mperf, aperf, tsc, freq_ghz, io_boost, common_comm, load, duration_ms, sample_num, elapsed_time, tsc_ghz) + store_csv(cpu_int, time_pre_dec, time_post_dec, core_busy, scaled, _from, _to, mperf, aperf, tsc, freq_ghz, io_boost, common_comm, load, duration_ms, sample_num, elapsed_time, tsc_ghz, cpu_mask) if cpu_int > current_max_cpu: current_max_cpu = cpu_int # End of for each trace line loop # Now seperate the main overall csv file into per CPU csv files. - split_csv() + split_csv(current_max_cpu, cpu_mask) def signal_handler(signal, frame): print(' SIGINT: Forcing cleanup before exit.') if interval: - disable_trace() + disable_trace(trace_file) clear_trace_file() # Free the memory free_trace_buffer() sys.exit(0) -signal.signal(signal.SIGINT, signal_handler) +if __name__ == "__main__": + trace_file = "/sys/kernel/debug/tracing/events/power/pstate_sample/enable" + signal.signal(signal.SIGINT, signal_handler) -interval = "" -filename = "" -cpu_list = "" -testname = "" -memory = "10240" -graph_data_present = False; + interval = "" + filename = "" + cpu_list = "" + testname = "" + memory = "10240" + graph_data_present = False; -valid1 = False -valid2 = False + valid1 = False + valid2 = False -cpu_mask = zeros((MAX_CPUS,), dtype=int) + cpu_mask = zeros((MAX_CPUS,), dtype=int) -try: - opts, args = getopt.getopt(sys.argv[1:],"ht:i:c:n:m:",["help","trace_file=","interval=","cpu=","name=","memory="]) -except getopt.GetoptError: - print_help() - sys.exit(2) -for opt, arg in opts: - if opt == '-h': - print() + try: + opts, args = getopt.getopt(sys.argv[1:],"ht:i:c:n:m:",["help","trace_file=","interval=","cpu=","name=","memory="]) + except getopt.GetoptError: + print_help('intel_pstate') + sys.exit(2) + for opt, arg in opts: + if opt == '-h': + print_help('intel_pstate') + sys.exit() + elif opt in ("-t", "--trace_file"): + valid1 = True + location = os.path.realpath(os.path.join(os.getcwd(), os.path.dirname(__file__))) + filename = os.path.join(location, arg) + elif opt in ("-i", "--interval"): + valid1 = True + interval = arg + elif opt in ("-c", "--cpu"): + cpu_list = arg + elif opt in ("-n", "--name"): + valid2 = True + testname = arg + elif opt in ("-m", "--memory"): + memory = arg + + if not (valid1 and valid2): + print_help('intel_pstate') sys.exit() - elif opt in ("-t", "--trace_file"): - valid1 = True - location = os.path.realpath(os.path.join(os.getcwd(), os.path.dirname(__file__))) - filename = os.path.join(location, arg) - elif opt in ("-i", "--interval"): - valid1 = True - interval = arg - elif opt in ("-c", "--cpu"): - cpu_list = arg - elif opt in ("-n", "--name"): - valid2 = True - testname = arg - elif opt in ("-m", "--memory"): - memory = arg - -if not (valid1 and valid2): - print_help() - sys.exit() - -if cpu_list: - for p in re.split("[,]", cpu_list): - if int(p) < MAX_CPUS : - cpu_mask[int(p)] = 1 -else: - for i in range (0, MAX_CPUS): - cpu_mask[i] = 1 - -if not os.path.exists('results'): - os.mkdir('results') + + if cpu_list: + for p in re.split("[,]", cpu_list): + if int(p) < MAX_CPUS : + cpu_mask[int(p)] = 1 + else: + for i in range (0, MAX_CPUS): + cpu_mask[i] = 1 + + if not os.path.exists('results'): + os.mkdir('results') + # The regular user needs to own the directory, not root. + fix_ownership('results') + + os.chdir('results') + if os.path.exists(testname): + print('The test name directory already exists. Please provide a unique test name. Test re-run not supported, yet.') + sys.exit() + os.mkdir(testname) # The regular user needs to own the directory, not root. - fix_ownership('results') - -os.chdir('results') -if os.path.exists(testname): - print('The test name directory already exists. Please provide a unique test name. Test re-run not supported, yet.') - sys.exit() -os.mkdir(testname) -# The regular user needs to own the directory, not root. -fix_ownership(testname) -os.chdir(testname) - -# Temporary (or perhaps not) -cur_version = sys.version_info -print('python version (should be >= 2.7):') -print(cur_version) - -# Left as "cleanup" for potential future re-run ability. -cleanup_data_files() - -if interval: - filename = "/sys/kernel/debug/tracing/trace" - clear_trace_file() - set_trace_buffer_size() - enable_trace() - print('Sleeping for ', interval, 'seconds') - time.sleep(int(interval)) - disable_trace() - -current_max_cpu = 0 - -read_trace_data(filename) - -if interval: - clear_trace_file() - # Free the memory - free_trace_buffer() - -if graph_data_present == False: - print('No valid data to plot') - sys.exit(2) - -for cpu_no in range(0, current_max_cpu + 1): - plot_perf_busy_with_sample(cpu_no) - plot_perf_busy(cpu_no) - plot_durations(cpu_no) - plot_loads(cpu_no) - -plot_pstate_cpu_with_sample() -plot_pstate_cpu() -plot_load_cpu() -plot_frequency_cpu() -plot_duration_cpu() -plot_scaled_cpu() -plot_boost_cpu() -plot_ghz_cpu() - -# It is preferrable, but not necessary, that the regular user owns the files, not root. -for root, dirs, files in os.walk('.'): - for f in files: - fix_ownership(f) - -os.chdir('../../') + fix_ownership(testname) + os.chdir(testname) + + # Temporary (or perhaps not) + cur_version = sys.version_info + print('python version (should be >= 2.7):') + print(cur_version) + + # Left as "cleanup" for potential future re-run ability. + cleanup_data_files() + + if interval: + filename = "/sys/kernel/debug/tracing/trace" + clear_trace_file() + set_trace_buffer_size(memory) + enable_trace(trace_file) + print('Sleeping for ', interval, 'seconds') + time.sleep(int(interval)) + disable_trace(trace_file) + + current_max_cpu = 0 + + read_trace_data(filename, cpu_mask) + + if interval: + clear_trace_file() + # Free the memory + free_trace_buffer() + + if graph_data_present == False: + print('No valid data to plot') + sys.exit(2) + + for cpu_no in range(0, current_max_cpu + 1): + plot_perf_busy_with_sample(cpu_no) + plot_perf_busy(cpu_no) + plot_durations(cpu_no) + plot_loads(cpu_no) + + plot_pstate_cpu_with_sample() + plot_pstate_cpu() + plot_load_cpu() + plot_frequency_cpu() + plot_duration_cpu() + plot_scaled_cpu() + plot_boost_cpu() + plot_ghz_cpu() + + # It is preferrable, but not necessary, that the regular user owns the files, not root. + for root, dirs, files in os.walk('.'): + for f in files: + fix_ownership(f) + + os.chdir('../../') diff --git a/tools/power/x86/turbostat/Makefile b/tools/power/x86/turbostat/Makefile index f3e3c94ab9bd..92e139b9c792 100644 --- a/tools/power/x86/turbostat/Makefile +++ b/tools/power/x86/turbostat/Makefile @@ -9,7 +9,7 @@ ifeq ("$(origin O)", "command line") endif turbostat : turbostat.c -override CFLAGS += -O2 -Wall -I../../../include +override CFLAGS += -O2 -Wall -Wextra -I../../../include override CFLAGS += -DMSRHEADER='"../../../../arch/x86/include/asm/msr-index.h"' override CFLAGS += -DINTEL_FAMILY_HEADER='"../../../../arch/x86/include/asm/intel-family.h"' override CFLAGS += -D_FILE_OFFSET_BITS=64 diff --git a/tools/power/x86/turbostat/turbostat.8 b/tools/power/x86/turbostat/turbostat.8 index 9b17097bc3d7..c7b26a3603af 100644 --- a/tools/power/x86/turbostat/turbostat.8 +++ b/tools/power/x86/turbostat/turbostat.8 @@ -92,40 +92,66 @@ displays the statistics gathered since it was forked. .SH ROW DESCRIPTIONS The system configuration dump (if --quiet is not used) is followed by statistics. The first row of the statistics labels the content of each column (below). The second row of statistics is the system summary line. The system summary line has a '-' in the columns for the Package, Core, and CPU. The contents of the system summary line depends on the type of column. Columns that count items (eg. IRQ) show the sum across all CPUs in the system. Columns that show a percentage show the average across all CPUs in the system. Columns that dump raw MSR values simply show 0 in the summary. After the system summary row, each row describes a specific Package/Core/CPU. Note that if the --cpu parameter is used to limit which specific CPUs are displayed, turbostat will still collect statistics for all CPUs in the system and will still show the system summary for all CPUs in the system. .SH COLUMN DESCRIPTIONS -.nf +.PP \fBusec\fP For each CPU, the number of microseconds elapsed during counter collection, including thread migration -- if any. This counter is disabled by default, and is enabled with "--enable usec", or --debug. On the summary row, usec refers to the total elapsed time to collect the counters on all cpus. +.PP \fBTime_Of_Day_Seconds\fP For each CPU, the gettimeofday(2) value (seconds.subsec since Epoch) when the counters ending the measurement interval were collected. This column is disabled by default, and can be enabled with "--enable Time_Of_Day_Seconds" or "--debug". On the summary row, Time_Of_Day_Seconds refers to the timestamp following collection of counters on the last CPU. +.PP \fBCore\fP processor core number. Note that multiple CPUs per core indicate support for Intel(R) Hyper-Threading Technology (HT). +.PP \fBCPU\fP Linux CPU (logical processor) number. Yes, it is okay that on many systems the CPUs are not listed in numerical order -- for efficiency reasons, turbostat runs in topology order, so HT siblings appear together. +.PP \fBPackage\fP processor package number -- not present on systems with a single processor package. +.PP \fBAvg_MHz\fP number of cycles executed divided by time elapsed. Note that this includes idle-time when 0 instructions are executed. +.PP \fBBusy%\fP percent of the measurement interval that the CPU executes instructions, aka. % of time in "C0" state. +.PP \fBBzy_MHz\fP average clock rate while the CPU was not idle (ie. in "c0" state). +.PP \fBTSC_MHz\fP average MHz that the TSC ran during the entire interval. +.PP \fBIRQ\fP The number of interrupts serviced by that CPU during the measurement interval. The system total line is the sum of interrupts serviced across all CPUs. turbostat parses /proc/interrupts to generate this summary. +.PP \fBSMI\fP The number of System Management Interrupts serviced CPU during the measurement interval. While this counter is actually per-CPU, SMI are triggered on all processors, so the number should be the same for all CPUs. +.PP \fBC1, C2, C3...\fP The number times Linux requested the C1, C2, C3 idle state during the measurement interval. The system summary line shows the sum for all CPUs. These are C-state names as exported in /sys/devices/system/cpu/cpu*/cpuidle/state*/name. While their names are generic, their attributes are processor specific. They the system description section of output shows what MWAIT sub-states they are mapped to on each system. +.PP \fBC1%, C2%, C3%\fP The residency percentage that Linux requested C1, C2, C3.... The system summary is the average of all CPUs in the system. Note that these are software, reflecting what was requested. The hardware counters reflect what was actually achieved. +.PP \fBCPU%c1, CPU%c3, CPU%c6, CPU%c7\fP show the percentage residency in hardware core idle states. These numbers are from hardware residency counters. +.PP \fBCoreTmp\fP Degrees Celsius reported by the per-core Digital Thermal Sensor. +.PP \fBPkgTmp\fP Degrees Celsius reported by the per-package Package Thermal Monitor. +.PP \fBGFX%rc6\fP The percentage of time the GPU is in the "render C6" state, rc6, during the measurement interval. From /sys/class/drm/card0/power/rc6_residency_ms. +.PP \fBGFXMHz\fP Instantaneous snapshot of what sysfs presents at the end of the measurement interval. From /sys/class/graphics/fb0/device/drm/card0/gt_cur_freq_mhz. +.PP \fBPkg%pc2, Pkg%pc3, Pkg%pc6, Pkg%pc7\fP percentage residency in hardware package idle states. These numbers are from hardware residency counters. +.PP \fBPkgWatt\fP Watts consumed by the whole package. +.PP \fBCorWatt\fP Watts consumed by the core part of the package. +.PP \fBGFXWatt\fP Watts consumed by the Graphics part of the package -- available only on client processors. +.PP \fBRAMWatt\fP Watts consumed by the DRAM DIMMS -- available only on server processors. +.PP \fBPKG_%\fP percent of the interval that RAPL throttling was active on the Package. Note that the system summary is the sum of the package throttling time, and thus may be higher than 100% on a multi-package system. Note that the meaning of this field is model specific. For example, some hardware increments this counter when RAPL responds to thermal limits, but does not increment this counter when RAPL responds to power limits. Comparing PkgWatt and PkgTmp to system limits is necessary. +.PP \fBRAM_%\fP percent of the interval that RAPL throttling was active on DRAM. -.fi +.PP +\fBUncMHz\fP uncore MHz, instantaneous sample. .SH TOO MUCH INFORMATION EXAMPLE By default, turbostat dumps all possible information -- a system configuration header, followed by columns for all counters. This is ideal for remote debugging, use the "--out" option to save everything to a text file, and get that file to the expert helping you debug. .PP When you are not interested in all that information, and there are several ways to see only what you want. First the "--quiet" option will skip the configuration information, and turbostat will show only the counter columns. Second, you can reduce the columns with the "--hide" and "--show" options. If you use the "--show" option, then turbostat will show only the columns you list. If you use the "--hide" option, turbostat will show all columns, except the ones you list. .PP -To find out what columns are available for --show and --hide, the "--list" option is available. For convenience, the special strings "sysfs" can be used to refer to all of the sysfs C-state counters at once: +To find out what columns are available for --show and --hide, the "--list" option is available. Usually, the CATEGORY names above are used to refer to groups of counters. Also, for convenience, the special string "sysfs" can be used to refer to all of the sysfs C-state counters at once: +.PP .nf sudo ./turbostat --show sysfs --quiet sleep 10 10.003837 sec @@ -158,32 +184,29 @@ Without a command to fork, turbostat displays statistics ever 5 seconds. Periodic output goes to stdout, by default, unless --out is used to specify an output file. The 5-second interval can be changed with the "-i sec" option. .nf -sudo ./turbostat --quiet --hide sysfs,IRQ,SMI,CoreTmp,PkgTmp,GFX%rc6,GFXMHz,PkgWatt,CorWatt,GFXWatt - Core CPU Avg_MHz Busy% Bzy_MHz TSC_MHz CPU%c1 CPU%c3 CPU%c6 CPU%c7 - - - 488 12.52 3900 3498 12.50 0.00 0.00 74.98 - 0 0 5 0.13 3900 3498 99.87 0.00 0.00 0.00 - 0 4 3897 99.99 3900 3498 0.01 - 1 1 0 0.00 3856 3498 0.01 0.00 0.00 99.98 - 1 5 0 0.00 3861 3498 0.01 - 2 2 1 0.02 3889 3498 0.03 0.00 0.00 99.95 - 2 6 0 0.00 3863 3498 0.05 - 3 3 0 0.01 3869 3498 0.02 0.00 0.00 99.97 - 3 7 0 0.00 3878 3498 0.03 - Core CPU Avg_MHz Busy% Bzy_MHz TSC_MHz CPU%c1 CPU%c3 CPU%c6 CPU%c7 - - - 491 12.59 3900 3498 12.42 0.00 0.00 74.99 - 0 0 27 0.69 3900 3498 99.31 0.00 0.00 0.00 - 0 4 3898 99.99 3900 3498 0.01 - 1 1 0 0.00 3883 3498 0.01 0.00 0.00 99.99 - 1 5 0 0.00 3898 3498 0.01 - 2 2 0 0.01 3889 3498 0.02 0.00 0.00 99.98 - 2 6 0 0.00 3889 3498 0.02 - 3 3 0 0.00 3856 3498 0.01 0.00 0.00 99.99 - 3 7 0 0.00 3897 3498 0.01 +sudo turbostat --quiet --show CPU,frequency + Core CPU Avg_MHz Busy% Bzy_MHz TSC_MHz CPU%c7 UncMhz + - - 524 12.48 4198 3096 74.53 3800 + 0 0 4 0.09 4081 3096 98.88 3800 + 0 4 1 0.02 4063 3096 + 1 1 2 0.06 4063 3096 99.60 + 1 5 2 0.05 4070 3096 + 2 2 4178 99.52 4199 3096 0.00 + 2 6 3 0.08 4159 3096 + 3 3 1 0.04 4046 3096 99.66 + 3 7 0 0.01 3989 3096 + Core CPU Avg_MHz Busy% Bzy_MHz TSC_MHz CPU%c7 UncMhz + - - 525 12.52 4198 3096 74.54 3800 + 0 0 4 0.10 4051 3096 99.49 3800 + 0 4 2 0.04 3993 3096 + 1 1 3 0.07 4054 3096 99.56 + 1 5 4 0.10 4018 3096 + 2 2 4178 99.51 4199 3096 0.00 + 2 6 4 0.09 4143 3096 + 3 3 2 0.06 4026 3096 99.10 + 3 7 7 0.17 4074 3096 .fi -This example also shows the use of the --hide option to skip columns that are not wanted. -Note that cpu4 in this example is 99.99% busy, while the other CPUs are all under 1% busy. -Notice that cpu4's HT sibling is cpu0, which is under 1% busy, but can get into CPU%c1 only, -because its cpu4's activity on shared hardware keeps it from entering a deeper C-state. +This example also shows the use of the --show option to show only the desired columns. .SH SYSTEM CONFIGURATION INFORMATION EXAMPLE @@ -191,61 +214,86 @@ By default, turbostat always dumps system configuration information before taking measurements. In the example above, "--quiet" is used to suppress that output. Here is an example of the configuration information: .nf -turbostat version 2017.02.15 - Len Brown <lenb@kernel.org> -CPUID(0): GenuineIntel 13 CPUID levels; family:model:stepping 0x6:3c:3 (6:60:3) -CPUID(1): SSE3 MONITOR - EIST TM2 TSC MSR ACPI-TM TM -CPUID(6): APERF, TURBO, DTS, PTM, No-HWP, No-HWPnotify, No-HWPwindow, No-HWPepp, No-HWPpkg, EPB -cpu4: MSR_IA32_MISC_ENABLE: 0x00850089 (TCC EIST No-MWAIT PREFETCH TURBO) -CPUID(7): No-SGX -cpu4: MSR_MISC_PWR_MGMT: 0x00400000 (ENable-EIST_Coordination DISable-EPB DISable-OOB) -RAPL: 3121 sec. Joule Counter Range, at 84 Watts -cpu4: MSR_PLATFORM_INFO: 0x80838f3012300 +turbostat version 2022.04.16 - Len Brown <lenb@kernel.org> +Kernel command line: BOOT_IMAGE=/boot/vmlinuz-5.18.0-rc6-00001-ge6891250e3b5 ... +CPUID(0): GenuineIntel 0x16 CPUID levels +CPUID(1): family:model:stepping 0x6:9e:9 (6:158:9) microcode 0xea +CPUID(0x80000000): max_extended_levels: 0x80000008 +CPUID(1): SSE3 MONITOR - EIST TM2 TSC MSR ACPI-TM HT TM +CPUID(6): APERF, TURBO, DTS, PTM, HWP, HWPnotify, HWPwindow, HWPepp, No-HWPpkg, EPB +cpu7: MSR_IA32_MISC_ENABLE: 0x00850089 (TCC EIST MWAIT PREFETCH TURBO) +CPUID(7): SGX +cpu7: MSR_IA32_FEATURE_CONTROL: 0x00000005 (Locked ) +CPUID(0x15): eax_crystal: 2 ebx_tsc: 258 ecx_crystal_hz: 0 +TSC: 3096 MHz (24000000 Hz * 258 / 2 / 1000000) +CPUID(0x16): base_mhz: 3100 max_mhz: 4200 bus_mhz: 100 +cpu7: MSR_MISC_PWR_MGMT: 0x00401cc0 (ENable-EIST_Coordination DISable-EPB DISable-OOB) +RAPL: 5825 sec. Joule Counter Range, at 45 Watts +cpu7: MSR_PLATFORM_INFO: 0x80839f1011f00 8 * 100.0 = 800.0 MHz max efficiency frequency -35 * 100.0 = 3500.0 MHz base frequency -cpu4: MSR_IA32_POWER_CTL: 0x0004005d (C1E auto-promotion: DISabled) -cpu4: MSR_TURBO_RATIO_LIMIT: 0x25262727 -37 * 100.0 = 3700.0 MHz max turbo 4 active cores -38 * 100.0 = 3800.0 MHz max turbo 3 active cores -39 * 100.0 = 3900.0 MHz max turbo 2 active cores -39 * 100.0 = 3900.0 MHz max turbo 1 active cores -cpu4: MSR_CONFIG_TDP_NOMINAL: 0x00000023 (base_ratio=35) -cpu4: MSR_CONFIG_TDP_LEVEL_1: 0x00000000 () -cpu4: MSR_CONFIG_TDP_LEVEL_2: 0x00000000 () -cpu4: MSR_CONFIG_TDP_CONTROL: 0x80000000 ( lock=1) -cpu4: MSR_TURBO_ACTIVATION_RATIO: 0x00000000 (MAX_NON_TURBO_RATIO=0 lock=0) -cpu4: MSR_PKG_CST_CONFIG_CONTROL: 0x1e000400 (UNdemote-C3, UNdemote-C1, demote-C3, demote-C1, UNlocked: pkg-cstate-limit=0: pc0) -cpu4: POLL: CPUIDLE CORE POLL IDLE -cpu4: C1: MWAIT 0x00 -cpu4: C1E: MWAIT 0x01 -cpu4: C3: MWAIT 0x10 -cpu4: C6: MWAIT 0x20 -cpu4: C7s: MWAIT 0x32 -cpu4: MSR_MISC_FEATURE_CONTROL: 0x00000000 (L2-Prefetch L2-Prefetch-pair L1-Prefetch L1-IP-Prefetch) -cpu0: MSR_IA32_ENERGY_PERF_BIAS: 0x00000006 (balanced) -cpu0: MSR_CORE_PERF_LIMIT_REASONS, 0x31200000 (Active: ) (Logged: Transitions, MultiCoreTurbo, Amps, Auto-HWP, ) -cpu0: MSR_GFX_PERF_LIMIT_REASONS, 0x00000000 (Active: ) (Logged: ) -cpu0: MSR_RING_PERF_LIMIT_REASONS, 0x0d000000 (Active: ) (Logged: Amps, PkgPwrL1, PkgPwrL2, ) +31 * 100.0 = 3100.0 MHz base frequency +cpu7: MSR_IA32_POWER_CTL: 0x002c005d (C1E auto-promotion: DISabled) +cpu7: MSR_TURBO_RATIO_LIMIT: 0x2728292a +39 * 100.0 = 3900.0 MHz max turbo 4 active cores +40 * 100.0 = 4000.0 MHz max turbo 3 active cores +41 * 100.0 = 4100.0 MHz max turbo 2 active cores +42 * 100.0 = 4200.0 MHz max turbo 1 active cores +cpu7: MSR_CONFIG_TDP_NOMINAL: 0x0000001f (base_ratio=31) +cpu7: MSR_CONFIG_TDP_LEVEL_1: 0x00000000 () +cpu7: MSR_CONFIG_TDP_LEVEL_2: 0x00000000 () +cpu7: MSR_CONFIG_TDP_CONTROL: 0x80000000 ( lock=1) +cpu7: MSR_TURBO_ACTIVATION_RATIO: 0x00000000 (MAX_NON_TURBO_RATIO=0 lock=0) +cpu7: MSR_PKG_CST_CONFIG_CONTROL: 0x1e008008 (UNdemote-C3, UNdemote-C1, demote-C3, demote-C1, locked, pkg-cstate-limit=8 (unlimited)) +Uncore Frequency pkg0 die0: 800 - 3900 MHz (800 - 3900 MHz) +/dev/cpu_dma_latency: 2000000000 usec (default) +current_driver: intel_idle +current_governor: menu +current_governor_ro: menu +cpu7: POLL: CPUIDLE CORE POLL IDLE +cpu7: C1: MWAIT 0x00 +cpu7: C1E: MWAIT 0x01 +cpu7: C3: MWAIT 0x10 +cpu7: C6: MWAIT 0x20 +cpu7: C7s: MWAIT 0x33 +cpu7: C8: MWAIT 0x40 +cpu7: C9: MWAIT 0x50 +cpu7: C10: MWAIT 0x60 +cpu7: cpufreq driver: intel_pstate +cpu7: cpufreq governor: performance +cpufreq intel_pstate no_turbo: 0 +cpu7: MSR_MISC_FEATURE_CONTROL: 0x00000000 (L2-Prefetch L2-Prefetch-pair L1-Prefetch L1-IP-Prefetch) +cpu0: MSR_PM_ENABLE: 0x00000001 (HWP) +cpu0: MSR_HWP_CAPABILITIES: 0x01101f53 (high 83 guar 31 eff 16 low 1) +cpu0: MSR_HWP_REQUEST: 0x00005353 (min 83 max 83 des 0 epp 0x0 window 0x0 pkg 0x0) +cpu0: MSR_HWP_INTERRUPT: 0x00000001 (EN_Guaranteed_Perf_Change, Dis_Excursion_Min) +cpu0: MSR_HWP_STATUS: 0x00000004 (No-Guaranteed_Perf_Change, No-Excursion_Min) +cpu0: EPB: 6 (balanced) cpu0: MSR_RAPL_POWER_UNIT: 0x000a0e03 (0.125000 Watts, 0.000061 Joules, 0.000977 sec.) -cpu0: MSR_PKG_POWER_INFO: 0x000002a0 (84 W TDP, RAPL 0 - 0 W, 0.000000 sec.) -cpu0: MSR_PKG_POWER_LIMIT: 0x428348001a82a0 (UNlocked) -cpu0: PKG Limit #1: ENabled (84.000000 Watts, 8.000000 sec, clamp DISabled) -cpu0: PKG Limit #2: ENabled (105.000000 Watts, 0.002441* sec, clamp DISabled) +cpu0: MSR_PKG_POWER_INFO: 0x00000168 (45 W TDP, RAPL 0 - 0 W, 0.000000 sec.) +cpu0: MSR_PKG_POWER_LIMIT: 0x42820800218208 (UNlocked) +cpu0: PKG Limit #1: ENabled (65.000 Watts, 64.000000 sec, clamp ENabled) +cpu0: PKG Limit #2: ENabled (65.000 Watts, 0.002441* sec, clamp DISabled) +cpu0: MSR_VR_CURRENT_CONFIG: 0x00000000 +cpu0: PKG Limit #4: 0.000000 Watts (UNlocked) +cpu0: MSR_DRAM_POWER_LIMIT: 0x5400de00000000 (UNlocked) +cpu0: DRAM Limit: DISabled (0.000 Watts, 0.000977 sec, clamp DISabled) cpu0: MSR_PP0_POLICY: 0 cpu0: MSR_PP0_POWER_LIMIT: 0x00000000 (UNlocked) -cpu0: Cores Limit: DISabled (0.000000 Watts, 0.000977 sec, clamp DISabled) +cpu0: Cores Limit: DISabled (0.000 Watts, 0.000977 sec, clamp DISabled) cpu0: MSR_PP1_POLICY: 0 cpu0: MSR_PP1_POWER_LIMIT: 0x00000000 (UNlocked) -cpu0: GFX Limit: DISabled (0.000000 Watts, 0.000977 sec, clamp DISabled) -cpu0: MSR_IA32_TEMPERATURE_TARGET: 0x00641400 (100 C) -cpu0: MSR_IA32_PACKAGE_THERM_STATUS: 0x884c0800 (24 C) -cpu0: MSR_IA32_THERM_STATUS: 0x884c0000 (24 C +/- 1) -cpu1: MSR_IA32_THERM_STATUS: 0x88510000 (19 C +/- 1) -cpu2: MSR_IA32_THERM_STATUS: 0x884e0000 (22 C +/- 1) -cpu3: MSR_IA32_THERM_STATUS: 0x88510000 (19 C +/- 1) -cpu4: MSR_PKGC3_IRTL: 0x00008842 (valid, 67584 ns) -cpu4: MSR_PKGC6_IRTL: 0x00008873 (valid, 117760 ns) -cpu4: MSR_PKGC7_IRTL: 0x00008891 (valid, 148480 ns) +cpu0: GFX Limit: DISabled (0.000 Watts, 0.000977 sec, clamp DISabled) +cpu0: MSR_IA32_TEMPERATURE_TARGET: 0x00640000 (100 C) (100 default - 0 offset) +cpu0: MSR_IA32_PACKAGE_THERM_STATUS: 0x88200800 (68 C) +cpu0: MSR_IA32_PACKAGE_THERM_INTERRUPT: 0x00000003 (100 C, 100 C) +cpu7: MSR_PKGC3_IRTL: 0x0000884e (valid, 79872 ns) +cpu7: MSR_PKGC6_IRTL: 0x00008876 (valid, 120832 ns) +cpu7: MSR_PKGC7_IRTL: 0x00008894 (valid, 151552 ns) +cpu7: MSR_PKGC8_IRTL: 0x000088fa (valid, 256000 ns) +cpu7: MSR_PKGC9_IRTL: 0x0000894c (valid, 339968 ns) +cpu7: MSR_PKGC10_IRTL: 0x00008bf2 (valid, 1034240 ns) .fi +.PP The \fBmax efficiency\fP frequency, a.k.a. Low Frequency Mode, is the frequency available at the minimum package voltage. The \fBTSC frequency\fP is the base frequency of the processor -- this should match the brand string @@ -292,7 +340,7 @@ starts a new interval. must be run as root. Alternatively, non-root users can be enabled to run turbostat this way: -# setcap cap_sys_rawio=ep ./turbostat +# setcap cap_sys_admin,cap_sys_rawio,cap_sys_nice=+ep ./turbostat # chmod +r /dev/cpu/*/msr diff --git a/tools/power/x86/turbostat/turbostat.c b/tools/power/x86/turbostat/turbostat.c index 47d3ba895d6d..aba460410dbd 100644 --- a/tools/power/x86/turbostat/turbostat.c +++ b/tools/power/x86/turbostat/turbostat.c @@ -3,7 +3,7 @@ * turbostat -- show CPU frequency and C-state residency * on modern Intel and AMD processors. * - * Copyright (c) 2021 Intel Corporation. + * Copyright (c) 2022 Intel Corporation. * Len Brown <len.brown@intel.com> */ @@ -37,6 +37,173 @@ #include <asm/unistd.h> #include <stdbool.h> +#define UNUSED(x) (void)(x) + +/* + * This list matches the column headers, except + * 1. built-in only, the sysfs counters are not here -- we learn of those at run-time + * 2. Core and CPU are moved to the end, we can't have strings that contain them + * matching on them for --show and --hide. + */ + +/* + * buffer size used by sscanf() for added column names + * Usually truncated to 7 characters, but also handles 18 columns for raw 64-bit counters + */ +#define NAME_BYTES 20 +#define PATH_BYTES 128 + +enum counter_scope { SCOPE_CPU, SCOPE_CORE, SCOPE_PACKAGE }; +enum counter_type { COUNTER_ITEMS, COUNTER_CYCLES, COUNTER_SECONDS, COUNTER_USEC }; +enum counter_format { FORMAT_RAW, FORMAT_DELTA, FORMAT_PERCENT }; + +struct msr_counter { + unsigned int msr_num; + char name[NAME_BYTES]; + char path[PATH_BYTES]; + unsigned int width; + enum counter_type type; + enum counter_format format; + struct msr_counter *next; + unsigned int flags; +#define FLAGS_HIDE (1 << 0) +#define FLAGS_SHOW (1 << 1) +#define SYSFS_PERCPU (1 << 1) +}; + +struct msr_counter bic[] = { + { 0x0, "usec", "", 0, 0, 0, NULL, 0 }, + { 0x0, "Time_Of_Day_Seconds", "", 0, 0, 0, NULL, 0 }, + { 0x0, "Package", "", 0, 0, 0, NULL, 0 }, + { 0x0, "Node", "", 0, 0, 0, NULL, 0 }, + { 0x0, "Avg_MHz", "", 0, 0, 0, NULL, 0 }, + { 0x0, "Busy%", "", 0, 0, 0, NULL, 0 }, + { 0x0, "Bzy_MHz", "", 0, 0, 0, NULL, 0 }, + { 0x0, "TSC_MHz", "", 0, 0, 0, NULL, 0 }, + { 0x0, "IRQ", "", 0, 0, 0, NULL, 0 }, + { 0x0, "SMI", "", 32, 0, FORMAT_DELTA, NULL, 0 }, + { 0x0, "sysfs", "", 0, 0, 0, NULL, 0 }, + { 0x0, "CPU%c1", "", 0, 0, 0, NULL, 0 }, + { 0x0, "CPU%c3", "", 0, 0, 0, NULL, 0 }, + { 0x0, "CPU%c6", "", 0, 0, 0, NULL, 0 }, + { 0x0, "CPU%c7", "", 0, 0, 0, NULL, 0 }, + { 0x0, "ThreadC", "", 0, 0, 0, NULL, 0 }, + { 0x0, "CoreTmp", "", 0, 0, 0, NULL, 0 }, + { 0x0, "CoreCnt", "", 0, 0, 0, NULL, 0 }, + { 0x0, "PkgTmp", "", 0, 0, 0, NULL, 0 }, + { 0x0, "GFX%rc6", "", 0, 0, 0, NULL, 0 }, + { 0x0, "GFXMHz", "", 0, 0, 0, NULL, 0 }, + { 0x0, "Pkg%pc2", "", 0, 0, 0, NULL, 0 }, + { 0x0, "Pkg%pc3", "", 0, 0, 0, NULL, 0 }, + { 0x0, "Pkg%pc6", "", 0, 0, 0, NULL, 0 }, + { 0x0, "Pkg%pc7", "", 0, 0, 0, NULL, 0 }, + { 0x0, "Pkg%pc8", "", 0, 0, 0, NULL, 0 }, + { 0x0, "Pkg%pc9", "", 0, 0, 0, NULL, 0 }, + { 0x0, "Pk%pc10", "", 0, 0, 0, NULL, 0 }, + { 0x0, "CPU%LPI", "", 0, 0, 0, NULL, 0 }, + { 0x0, "SYS%LPI", "", 0, 0, 0, NULL, 0 }, + { 0x0, "PkgWatt", "", 0, 0, 0, NULL, 0 }, + { 0x0, "CorWatt", "", 0, 0, 0, NULL, 0 }, + { 0x0, "GFXWatt", "", 0, 0, 0, NULL, 0 }, + { 0x0, "PkgCnt", "", 0, 0, 0, NULL, 0 }, + { 0x0, "RAMWatt", "", 0, 0, 0, NULL, 0 }, + { 0x0, "PKG_%", "", 0, 0, 0, NULL, 0 }, + { 0x0, "RAM_%", "", 0, 0, 0, NULL, 0 }, + { 0x0, "Pkg_J", "", 0, 0, 0, NULL, 0 }, + { 0x0, "Cor_J", "", 0, 0, 0, NULL, 0 }, + { 0x0, "GFX_J", "", 0, 0, 0, NULL, 0 }, + { 0x0, "RAM_J", "", 0, 0, 0, NULL, 0 }, + { 0x0, "Mod%c6", "", 0, 0, 0, NULL, 0 }, + { 0x0, "Totl%C0", "", 0, 0, 0, NULL, 0 }, + { 0x0, "Any%C0", "", 0, 0, 0, NULL, 0 }, + { 0x0, "GFX%C0", "", 0, 0, 0, NULL, 0 }, + { 0x0, "CPUGFX%", "", 0, 0, 0, NULL, 0 }, + { 0x0, "Core", "", 0, 0, 0, NULL, 0 }, + { 0x0, "CPU", "", 0, 0, 0, NULL, 0 }, + { 0x0, "APIC", "", 0, 0, 0, NULL, 0 }, + { 0x0, "X2APIC", "", 0, 0, 0, NULL, 0 }, + { 0x0, "Die", "", 0, 0, 0, NULL, 0 }, + { 0x0, "GFXAMHz", "", 0, 0, 0, NULL, 0 }, + { 0x0, "IPC", "", 0, 0, 0, NULL, 0 }, + { 0x0, "CoreThr", "", 0, 0, 0, NULL, 0 }, + { 0x0, "UncMHz", "", 0, 0, 0, NULL, 0 }, +}; + +#define MAX_BIC (sizeof(bic) / sizeof(struct msr_counter)) +#define BIC_USEC (1ULL << 0) +#define BIC_TOD (1ULL << 1) +#define BIC_Package (1ULL << 2) +#define BIC_Node (1ULL << 3) +#define BIC_Avg_MHz (1ULL << 4) +#define BIC_Busy (1ULL << 5) +#define BIC_Bzy_MHz (1ULL << 6) +#define BIC_TSC_MHz (1ULL << 7) +#define BIC_IRQ (1ULL << 8) +#define BIC_SMI (1ULL << 9) +#define BIC_sysfs (1ULL << 10) +#define BIC_CPU_c1 (1ULL << 11) +#define BIC_CPU_c3 (1ULL << 12) +#define BIC_CPU_c6 (1ULL << 13) +#define BIC_CPU_c7 (1ULL << 14) +#define BIC_ThreadC (1ULL << 15) +#define BIC_CoreTmp (1ULL << 16) +#define BIC_CoreCnt (1ULL << 17) +#define BIC_PkgTmp (1ULL << 18) +#define BIC_GFX_rc6 (1ULL << 19) +#define BIC_GFXMHz (1ULL << 20) +#define BIC_Pkgpc2 (1ULL << 21) +#define BIC_Pkgpc3 (1ULL << 22) +#define BIC_Pkgpc6 (1ULL << 23) +#define BIC_Pkgpc7 (1ULL << 24) +#define BIC_Pkgpc8 (1ULL << 25) +#define BIC_Pkgpc9 (1ULL << 26) +#define BIC_Pkgpc10 (1ULL << 27) +#define BIC_CPU_LPI (1ULL << 28) +#define BIC_SYS_LPI (1ULL << 29) +#define BIC_PkgWatt (1ULL << 30) +#define BIC_CorWatt (1ULL << 31) +#define BIC_GFXWatt (1ULL << 32) +#define BIC_PkgCnt (1ULL << 33) +#define BIC_RAMWatt (1ULL << 34) +#define BIC_PKG__ (1ULL << 35) +#define BIC_RAM__ (1ULL << 36) +#define BIC_Pkg_J (1ULL << 37) +#define BIC_Cor_J (1ULL << 38) +#define BIC_GFX_J (1ULL << 39) +#define BIC_RAM_J (1ULL << 40) +#define BIC_Mod_c6 (1ULL << 41) +#define BIC_Totl_c0 (1ULL << 42) +#define BIC_Any_c0 (1ULL << 43) +#define BIC_GFX_c0 (1ULL << 44) +#define BIC_CPUGFX (1ULL << 45) +#define BIC_Core (1ULL << 46) +#define BIC_CPU (1ULL << 47) +#define BIC_APIC (1ULL << 48) +#define BIC_X2APIC (1ULL << 49) +#define BIC_Die (1ULL << 50) +#define BIC_GFXACTMHz (1ULL << 51) +#define BIC_IPC (1ULL << 52) +#define BIC_CORE_THROT_CNT (1ULL << 53) +#define BIC_UNCORE_MHZ (1ULL << 54) + +#define BIC_TOPOLOGY (BIC_Package | BIC_Node | BIC_CoreCnt | BIC_PkgCnt | BIC_Core | BIC_CPU | BIC_Die ) +#define BIC_THERMAL_PWR ( BIC_CoreTmp | BIC_PkgTmp | BIC_PkgWatt | BIC_CorWatt | BIC_GFXWatt | BIC_RAMWatt | BIC_PKG__ | BIC_RAM__) +#define BIC_FREQUENCY ( BIC_Avg_MHz | BIC_Busy | BIC_Bzy_MHz | BIC_TSC_MHz | BIC_GFXMHz | BIC_GFXACTMHz | BIC_UNCORE_MHZ) +#define BIC_IDLE ( BIC_sysfs | BIC_CPU_c1 | BIC_CPU_c3 | BIC_CPU_c6 | BIC_CPU_c7 | BIC_GFX_rc6 | BIC_Pkgpc2 | BIC_Pkgpc3 | BIC_Pkgpc6 | BIC_Pkgpc7 | BIC_Pkgpc8 | BIC_Pkgpc9 | BIC_Pkgpc10 | BIC_CPU_LPI | BIC_SYS_LPI | BIC_Mod_c6 | BIC_Totl_c0 | BIC_Any_c0 | BIC_GFX_c0 | BIC_CPUGFX) +#define BIC_OTHER ( BIC_IRQ | BIC_SMI | BIC_ThreadC | BIC_CoreTmp | BIC_IPC) + +#define BIC_DISABLED_BY_DEFAULT (BIC_USEC | BIC_TOD | BIC_APIC | BIC_X2APIC) + +unsigned long long bic_enabled = (0xFFFFFFFFFFFFFFFFULL & ~BIC_DISABLED_BY_DEFAULT); +unsigned long long bic_present = BIC_USEC | BIC_TOD | BIC_sysfs | BIC_APIC | BIC_X2APIC; + +#define DO_BIC(COUNTER_NAME) (bic_enabled & bic_present & COUNTER_NAME) +#define DO_BIC_READ(COUNTER_NAME) (bic_present & COUNTER_NAME) +#define ENABLE_BIC(COUNTER_NAME) (bic_enabled |= COUNTER_NAME) +#define BIC_PRESENT(COUNTER_BIT) (bic_present |= COUNTER_BIT) +#define BIC_NOT_PRESENT(COUNTER_BIT) (bic_present &= ~COUNTER_BIT) +#define BIC_IS_ENABLED(COUNTER_BIT) (bic_enabled & COUNTER_BIT) + char *proc_stat = "/proc/stat"; FILE *outf; int *fd_percpu; @@ -48,6 +215,7 @@ struct timespec interval_ts = { 5, 0 }; unsigned int model_orig; unsigned int num_iterations; +unsigned int header_iterations; unsigned int debug; unsigned int quiet; unsigned int shown; @@ -62,6 +230,8 @@ unsigned int do_slm_cstates; unsigned int use_c1_residency_msr; unsigned int has_aperf; unsigned int has_epb; +unsigned int has_turbo; +unsigned int is_hybrid; unsigned int do_irtl_snb; unsigned int do_irtl_hsw; unsigned int units = 1000000; /* MHz etc */ @@ -159,13 +329,6 @@ int ignore_stdin; #define MAX(a, b) ((a) > (b) ? (a) : (b)) -/* - * buffer size used by sscanf() for added column names - * Usually truncated to 7 characters, but also handles 18 columns for raw 64-bit counters - */ -#define NAME_BYTES 20 -#define PATH_BYTES 128 - int backwards_count; char *progname; @@ -205,6 +368,7 @@ struct core_data { unsigned int core_temp_c; unsigned int core_energy; /* MSR_CORE_ENERGY_STAT */ unsigned int core_id; + unsigned long long core_throt_cnt; unsigned long long counter[MAX_ADDED_COUNTERS]; } *core_even, *core_odd; @@ -233,6 +397,7 @@ struct pkg_data { unsigned long long rapl_pkg_perf_status; /* MSR_PKG_PERF_STATUS */ unsigned long long rapl_dram_perf_status; /* MSR_DRAM_PERF_STATUS */ unsigned int pkg_temp_c; + unsigned int uncore_mhz; unsigned long long counter[MAX_ADDED_COUNTERS]; } *package_even, *package_odd; @@ -255,24 +420,6 @@ struct pkg_data { #define GET_PKG(pkg_base, pkg_no) (pkg_base + pkg_no) -enum counter_scope { SCOPE_CPU, SCOPE_CORE, SCOPE_PACKAGE }; -enum counter_type { COUNTER_ITEMS, COUNTER_CYCLES, COUNTER_SECONDS, COUNTER_USEC }; -enum counter_format { FORMAT_RAW, FORMAT_DELTA, FORMAT_PERCENT }; - -struct msr_counter { - unsigned int msr_num; - char name[NAME_BYTES]; - char path[PATH_BYTES]; - unsigned int width; - enum counter_type type; - enum counter_format format; - struct msr_counter *next; - unsigned int flags; -#define FLAGS_HIDE (1 << 0) -#define FLAGS_SHOW (1 << 1) -#define SYSFS_PERCPU (1 << 1) -}; - /* * The accumulated sum of MSR is defined as a monotonic * increasing MSR, it will be accumulated periodically, @@ -522,8 +669,10 @@ static int perf_instr_count_open(int cpu_num) /* counter for cpu_num, including user + kernel and all processes */ fd = perf_event_open(&pea, -1, cpu_num, -1, 0); - if (fd == -1) - err(-1, "cpu%d: perf instruction counter\n", cpu_num); + if (fd == -1) { + warn("cpu%d: perf instruction counter", cpu_num); + BIC_NOT_PRESENT(BIC_IPC); + } return fd; } @@ -550,143 +699,10 @@ int get_msr(int cpu, off_t offset, unsigned long long *msr) return 0; } -/* - * This list matches the column headers, except - * 1. built-in only, the sysfs counters are not here -- we learn of those at run-time - * 2. Core and CPU are moved to the end, we can't have strings that contain them - * matching on them for --show and --hide. - */ -struct msr_counter bic[] = { - { 0x0, "usec" }, - { 0x0, "Time_Of_Day_Seconds" }, - { 0x0, "Package" }, - { 0x0, "Node" }, - { 0x0, "Avg_MHz" }, - { 0x0, "Busy%" }, - { 0x0, "Bzy_MHz" }, - { 0x0, "TSC_MHz" }, - { 0x0, "IRQ" }, - { 0x0, "SMI", "", 32, 0, FORMAT_DELTA, NULL }, - { 0x0, "sysfs" }, - { 0x0, "CPU%c1" }, - { 0x0, "CPU%c3" }, - { 0x0, "CPU%c6" }, - { 0x0, "CPU%c7" }, - { 0x0, "ThreadC" }, - { 0x0, "CoreTmp" }, - { 0x0, "CoreCnt" }, - { 0x0, "PkgTmp" }, - { 0x0, "GFX%rc6" }, - { 0x0, "GFXMHz" }, - { 0x0, "Pkg%pc2" }, - { 0x0, "Pkg%pc3" }, - { 0x0, "Pkg%pc6" }, - { 0x0, "Pkg%pc7" }, - { 0x0, "Pkg%pc8" }, - { 0x0, "Pkg%pc9" }, - { 0x0, "Pk%pc10" }, - { 0x0, "CPU%LPI" }, - { 0x0, "SYS%LPI" }, - { 0x0, "PkgWatt" }, - { 0x0, "CorWatt" }, - { 0x0, "GFXWatt" }, - { 0x0, "PkgCnt" }, - { 0x0, "RAMWatt" }, - { 0x0, "PKG_%" }, - { 0x0, "RAM_%" }, - { 0x0, "Pkg_J" }, - { 0x0, "Cor_J" }, - { 0x0, "GFX_J" }, - { 0x0, "RAM_J" }, - { 0x0, "Mod%c6" }, - { 0x0, "Totl%C0" }, - { 0x0, "Any%C0" }, - { 0x0, "GFX%C0" }, - { 0x0, "CPUGFX%" }, - { 0x0, "Core" }, - { 0x0, "CPU" }, - { 0x0, "APIC" }, - { 0x0, "X2APIC" }, - { 0x0, "Die" }, - { 0x0, "GFXAMHz" }, - { 0x0, "IPC" }, -}; - -#define MAX_BIC (sizeof(bic) / sizeof(struct msr_counter)) -#define BIC_USEC (1ULL << 0) -#define BIC_TOD (1ULL << 1) -#define BIC_Package (1ULL << 2) -#define BIC_Node (1ULL << 3) -#define BIC_Avg_MHz (1ULL << 4) -#define BIC_Busy (1ULL << 5) -#define BIC_Bzy_MHz (1ULL << 6) -#define BIC_TSC_MHz (1ULL << 7) -#define BIC_IRQ (1ULL << 8) -#define BIC_SMI (1ULL << 9) -#define BIC_sysfs (1ULL << 10) -#define BIC_CPU_c1 (1ULL << 11) -#define BIC_CPU_c3 (1ULL << 12) -#define BIC_CPU_c6 (1ULL << 13) -#define BIC_CPU_c7 (1ULL << 14) -#define BIC_ThreadC (1ULL << 15) -#define BIC_CoreTmp (1ULL << 16) -#define BIC_CoreCnt (1ULL << 17) -#define BIC_PkgTmp (1ULL << 18) -#define BIC_GFX_rc6 (1ULL << 19) -#define BIC_GFXMHz (1ULL << 20) -#define BIC_Pkgpc2 (1ULL << 21) -#define BIC_Pkgpc3 (1ULL << 22) -#define BIC_Pkgpc6 (1ULL << 23) -#define BIC_Pkgpc7 (1ULL << 24) -#define BIC_Pkgpc8 (1ULL << 25) -#define BIC_Pkgpc9 (1ULL << 26) -#define BIC_Pkgpc10 (1ULL << 27) -#define BIC_CPU_LPI (1ULL << 28) -#define BIC_SYS_LPI (1ULL << 29) -#define BIC_PkgWatt (1ULL << 30) -#define BIC_CorWatt (1ULL << 31) -#define BIC_GFXWatt (1ULL << 32) -#define BIC_PkgCnt (1ULL << 33) -#define BIC_RAMWatt (1ULL << 34) -#define BIC_PKG__ (1ULL << 35) -#define BIC_RAM__ (1ULL << 36) -#define BIC_Pkg_J (1ULL << 37) -#define BIC_Cor_J (1ULL << 38) -#define BIC_GFX_J (1ULL << 39) -#define BIC_RAM_J (1ULL << 40) -#define BIC_Mod_c6 (1ULL << 41) -#define BIC_Totl_c0 (1ULL << 42) -#define BIC_Any_c0 (1ULL << 43) -#define BIC_GFX_c0 (1ULL << 44) -#define BIC_CPUGFX (1ULL << 45) -#define BIC_Core (1ULL << 46) -#define BIC_CPU (1ULL << 47) -#define BIC_APIC (1ULL << 48) -#define BIC_X2APIC (1ULL << 49) -#define BIC_Die (1ULL << 50) -#define BIC_GFXACTMHz (1ULL << 51) -#define BIC_IPC (1ULL << 52) - -#define BIC_TOPOLOGY (BIC_Package | BIC_Node | BIC_CoreCnt | BIC_PkgCnt | BIC_Core | BIC_CPU | BIC_Die ) -#define BIC_THERMAL_PWR ( BIC_CoreTmp | BIC_PkgTmp | BIC_PkgWatt | BIC_CorWatt | BIC_GFXWatt | BIC_RAMWatt | BIC_PKG__ | BIC_RAM__) -#define BIC_FREQUENCY ( BIC_Avg_MHz | BIC_Busy | BIC_Bzy_MHz | BIC_TSC_MHz | BIC_GFXMHz | BIC_GFXACTMHz ) -#define BIC_IDLE ( BIC_sysfs | BIC_CPU_c1 | BIC_CPU_c3 | BIC_CPU_c6 | BIC_CPU_c7 | BIC_GFX_rc6 | BIC_Pkgpc2 | BIC_Pkgpc3 | BIC_Pkgpc6 | BIC_Pkgpc7 | BIC_Pkgpc8 | BIC_Pkgpc9 | BIC_Pkgpc10 | BIC_CPU_LPI | BIC_SYS_LPI | BIC_Mod_c6 | BIC_Totl_c0 | BIC_Any_c0 | BIC_GFX_c0 | BIC_CPUGFX) -#define BIC_OTHER ( BIC_IRQ | BIC_SMI | BIC_ThreadC | BIC_CoreTmp | BIC_IPC) - -#define BIC_DISABLED_BY_DEFAULT (BIC_USEC | BIC_TOD | BIC_APIC | BIC_X2APIC) - -unsigned long long bic_enabled = (0xFFFFFFFFFFFFFFFFULL & ~BIC_DISABLED_BY_DEFAULT); -unsigned long long bic_present = BIC_USEC | BIC_TOD | BIC_sysfs | BIC_APIC | BIC_X2APIC; - -#define DO_BIC(COUNTER_NAME) (bic_enabled & bic_present & COUNTER_NAME) -#define DO_BIC_READ(COUNTER_NAME) (bic_present & COUNTER_NAME) -#define ENABLE_BIC(COUNTER_NAME) (bic_enabled |= COUNTER_NAME) -#define BIC_PRESENT(COUNTER_BIT) (bic_present |= COUNTER_BIT) -#define BIC_NOT_PRESENT(COUNTER_BIT) (bic_present &= ~COUNTER_BIT) -#define BIC_IS_ENABLED(COUNTER_BIT) (bic_enabled & COUNTER_BIT) - #define MAX_DEFERRED 16 +char *deferred_add_names[MAX_DEFERRED]; char *deferred_skip_names[MAX_DEFERRED]; +int deferred_add_index; int deferred_skip_index; /* @@ -720,6 +736,8 @@ void help(void) " -l, --list list column headers only\n" " -n, --num_iterations num\n" " number of the measurement iterations\n" + " -N, --header_iterations num\n" + " print header every num iterations\n" " -o, --out file\n" " create or truncate \"file\" for all output\n" " -q, --quiet skip decoding system configuration header\n" @@ -741,7 +759,7 @@ void help(void) */ unsigned long long bic_lookup(char *name_list, enum show_hide_mode mode) { - int i; + unsigned int i; unsigned long long retval = 0; while (name_list) { @@ -752,40 +770,51 @@ unsigned long long bic_lookup(char *name_list, enum show_hide_mode mode) if (comma) *comma = '\0'; - if (!strcmp(name_list, "all")) - return ~0; - if (!strcmp(name_list, "topology")) - return BIC_TOPOLOGY; - if (!strcmp(name_list, "power")) - return BIC_THERMAL_PWR; - if (!strcmp(name_list, "idle")) - return BIC_IDLE; - if (!strcmp(name_list, "frequency")) - return BIC_FREQUENCY; - if (!strcmp(name_list, "other")) - return BIC_OTHER; - if (!strcmp(name_list, "all")) - return 0; - for (i = 0; i < MAX_BIC; ++i) { if (!strcmp(name_list, bic[i].name)) { retval |= (1ULL << i); break; } + if (!strcmp(name_list, "all")) { + retval |= ~0; + break; + } else if (!strcmp(name_list, "topology")) { + retval |= BIC_TOPOLOGY; + break; + } else if (!strcmp(name_list, "power")) { + retval |= BIC_THERMAL_PWR; + break; + } else if (!strcmp(name_list, "idle")) { + retval |= BIC_IDLE; + break; + } else if (!strcmp(name_list, "frequency")) { + retval |= BIC_FREQUENCY; + break; + } else if (!strcmp(name_list, "other")) { + retval |= BIC_OTHER; + break; + } + } if (i == MAX_BIC) { if (mode == SHOW_LIST) { - fprintf(stderr, "Invalid counter name: %s\n", name_list); - exit(-1); - } - deferred_skip_names[deferred_skip_index++] = name_list; - if (debug) - fprintf(stderr, "deferred \"%s\"\n", name_list); - if (deferred_skip_index >= MAX_DEFERRED) { - fprintf(stderr, "More than max %d un-recognized --skip options '%s'\n", - MAX_DEFERRED, name_list); - help(); - exit(1); + deferred_add_names[deferred_add_index++] = name_list; + if (deferred_add_index >= MAX_DEFERRED) { + fprintf(stderr, "More than max %d un-recognized --add options '%s'\n", + MAX_DEFERRED, name_list); + help(); + exit(1); + } + } else { + deferred_skip_names[deferred_skip_index++] = name_list; + if (debug) + fprintf(stderr, "deferred \"%s\"\n", name_list); + if (deferred_skip_index >= MAX_DEFERRED) { + fprintf(stderr, "More than max %d un-recognized --skip options '%s'\n", + MAX_DEFERRED, name_list); + help(); + exit(1); + } } } @@ -872,6 +901,9 @@ void print_header(char *delim) if (DO_BIC(BIC_CoreTmp)) outp += sprintf(outp, "%sCoreTmp", (printed++ ? delim : "")); + if (DO_BIC(BIC_CORE_THROT_CNT)) + outp += sprintf(outp, "%sCoreThr", (printed++ ? delim : "")); + if (do_rapl && !rapl_joules) { if (DO_BIC(BIC_CorWatt) && (do_rapl & RAPL_PER_CORE_ENERGY)) outp += sprintf(outp, "%sCorWatt", (printed++ ? delim : "")); @@ -961,6 +993,9 @@ void print_header(char *delim) if (DO_BIC(BIC_RAM__)) outp += sprintf(outp, "%sRAM_%%", (printed++ ? delim : "")); } + if (DO_BIC(BIC_UNCORE_MHZ)) + outp += sprintf(outp, "%sUncMHz", (printed++ ? delim : "")); + for (mp = sys.pp; mp; mp = mp->next) { if (mp->format == FORMAT_RAW) { if (mp->width == 64) @@ -1011,6 +1046,7 @@ int dump_counters(struct thread_data *t, struct core_data *c, struct pkg_data *p outp += sprintf(outp, "c6: %016llX\n", c->c6); outp += sprintf(outp, "c7: %016llX\n", c->c7); outp += sprintf(outp, "DTS: %dC\n", c->core_temp_c); + outp += sprintf(outp, "cpu_throt_count: %016llX\n", c->core_throt_cnt); outp += sprintf(outp, "Joules: %0X\n", c->core_energy); for (i = 0, mp = sys.cp; mp; i++, mp = mp->next) { @@ -1225,6 +1261,10 @@ int format_counters(struct thread_data *t, struct core_data *c, struct pkg_data if (DO_BIC(BIC_CoreTmp)) outp += sprintf(outp, "%s%d", (printed++ ? delim : ""), c->core_temp_c); + /* Core throttle count */ + if (DO_BIC(BIC_CORE_THROT_CNT)) + outp += sprintf(outp, "%s%lld", (printed++ ? delim : ""), c->core_throt_cnt); + for (i = 0, mp = sys.cp; mp; i++, mp = mp->next) { if (mp->format == FORMAT_RAW) { if (mp->width == 32) @@ -1311,6 +1351,7 @@ int format_counters(struct thread_data *t, struct core_data *c, struct pkg_data if (DO_BIC(BIC_PkgWatt)) outp += sprintf(outp, fmt8, (printed++ ? delim : ""), p->energy_pkg * rapl_energy_units / interval_float); + if (DO_BIC(BIC_CorWatt) && !(do_rapl & RAPL_PER_CORE_ENERGY)) outp += sprintf(outp, fmt8, (printed++ ? delim : ""), p->energy_cores * rapl_energy_units / interval_float); @@ -1337,6 +1378,9 @@ int format_counters(struct thread_data *t, struct core_data *c, struct pkg_data outp += sprintf(outp, fmt8, (printed++ ? delim : ""), 100.0 * p->rapl_dram_perf_status * rapl_time_units / interval_float); + /* UncMHz */ + if (DO_BIC(BIC_UNCORE_MHZ)) + outp += sprintf(outp, "%s%d", (printed++ ? delim : ""), p->uncore_mhz); for (i = 0, mp = sys.pp; mp; i++, mp = mp->next) { if (mp->format == FORMAT_RAW) { @@ -1386,14 +1430,14 @@ void flush_output_stderr(void) void format_all_counters(struct thread_data *t, struct core_data *c, struct pkg_data *p) { - static int printed; + static int count; - if (!printed || !summary_only) + if ((!count || (header_iterations && !(count % header_iterations))) || !summary_only) print_header("\t"); format_counters(&average.threads, &average.cores, &average.packages); - printed = 1; + count++; if (summary_only) return; @@ -1438,6 +1482,7 @@ int delta_package(struct pkg_data *new, struct pkg_data *old) else old->gfx_rc6_ms = new->gfx_rc6_ms - old->gfx_rc6_ms; + old->uncore_mhz = new->uncore_mhz; old->gfx_mhz = new->gfx_mhz; old->gfx_act_mhz = new->gfx_act_mhz; @@ -1467,6 +1512,7 @@ void delta_core(struct core_data *new, struct core_data *old) old->c6 = new->c6 - old->c6; old->c7 = new->c7 - old->c7; old->core_temp_c = new->core_temp_c; + old->core_throt_cnt = new->core_throt_cnt; old->mc6_us = new->mc6_us - old->mc6_us; DELTA_WRAP32(new->core_energy, old->core_energy); @@ -1626,6 +1672,7 @@ void clear_counters(struct thread_data *t, struct core_data *c, struct pkg_data c->mc6_us = 0; c->core_temp_c = 0; c->core_energy = 0; + c->core_throt_cnt = 0; p->pkg_wtd_core_c0 = 0; p->pkg_any_core_c0 = 0; @@ -1654,6 +1701,7 @@ void clear_counters(struct thread_data *t, struct core_data *c, struct pkg_data p->pkg_temp_c = 0; p->gfx_rc6_ms = 0; + p->uncore_mhz = 0; p->gfx_mhz = 0; p->gfx_act_mhz = 0; for (i = 0, mp = sys.tp; mp; i++, mp = mp->next) @@ -1710,6 +1758,7 @@ int sum_counters(struct thread_data *t, struct core_data *c, struct pkg_data *p) average.cores.mc6_us += c->mc6_us; average.cores.core_temp_c = MAX(average.cores.core_temp_c, c->core_temp_c); + average.cores.core_throt_cnt = MAX(average.cores.core_throt_cnt, c->core_throt_cnt); average.cores.core_energy += c->core_energy; @@ -1752,6 +1801,7 @@ int sum_counters(struct thread_data *t, struct core_data *c, struct pkg_data *p) average.packages.energy_gfx += p->energy_gfx; average.packages.gfx_rc6_ms = p->gfx_rc6_ms; + average.packages.uncore_mhz = p->uncore_mhz; average.packages.gfx_mhz = p->gfx_mhz; average.packages.gfx_act_mhz = p->gfx_act_mhz; @@ -1912,6 +1962,16 @@ int get_mp(int cpu, struct msr_counter *mp, unsigned long long *counterp) return 0; } +unsigned long long get_uncore_mhz(int package, int die) +{ + char path[128]; + + sprintf(path, "/sys/devices/system/cpu/intel_uncore_frequency/package_0%d_die_0%d/current_freq_khz", package, + die); + + return (snapshot_sysfs_counter(path) / 1000); +} + int get_epb(int cpu) { char path[128 + PATH_BYTES]; @@ -1987,6 +2047,26 @@ void get_apic_id(struct thread_data *t) fprintf(outf, "cpu%d: BIOS BUG: apic 0x%x x2apic 0x%x\n", t->cpu_id, t->apic_id, t->x2apic_id); } +int get_core_throt_cnt(int cpu, unsigned long long *cnt) +{ + char path[128 + PATH_BYTES]; + unsigned long long tmp; + FILE *fp; + int ret; + + sprintf(path, "/sys/devices/system/cpu/cpu%d/thermal_throttle/core_throttle_count", cpu); + fp = fopen(path, "r"); + if (!fp) + return -1; + ret = fscanf(fp, "%lld", &tmp); + fclose(fp); + if (ret != 1) + return -1; + *cnt = tmp; + + return 0; +} + /* * get_counters(...) * migrate to cpu @@ -2129,6 +2209,9 @@ retry: c->core_temp_c = tj_max - ((msr >> 16) & 0x7F); } + if (DO_BIC(BIC_CORE_THROT_CNT)) + get_core_throt_cnt(cpu, &c->core_throt_cnt); + if (do_rapl & RAPL_AMD_F17H) { if (get_msr(cpu, MSR_CORE_ENERGY_STAT, &msr)) return -14; @@ -2238,6 +2321,10 @@ retry: if (DO_BIC(BIC_GFX_rc6)) p->gfx_rc6_ms = gfx_cur_rc6_ms; + /* n.b. assume die0 uncore frequency applies to whole package */ + if (DO_BIC(BIC_UNCORE_MHZ)) + p->uncore_mhz = get_uncore_mhz(p->package_id, 0); + if (DO_BIC(BIC_GFXMHz)) p->gfx_mhz = gfx_cur_mhz; @@ -2323,7 +2410,7 @@ int skx_pkg_cstate_limits[16] = }; int icx_pkg_cstate_limits[16] = - { PCL__0, PCL__2, PCL__6, PCLRSV, PCLRSV, PCLRSV, PCLRSV, PCLUNL, PCLRSV, PCLRSV, PCLRSV, PCLRSV, PCLRSV, PCLRSV, + { PCL__0, PCL__2, PCL__6, PCL__6, PCLRSV, PCLRSV, PCLRSV, PCLUNL, PCLRSV, PCLRSV, PCLRSV, PCLRSV, PCLRSV, PCLRSV, PCLRSV, PCLRSV }; @@ -2428,24 +2515,30 @@ int has_turbo_ratio_group_limits(int family, int model) if (!genuine_intel) return 0; + if (family != 6) + return 0; + switch (model) { case INTEL_FAM6_ATOM_GOLDMONT: case INTEL_FAM6_SKYLAKE_X: case INTEL_FAM6_ICELAKE_X: + case INTEL_FAM6_SAPPHIRERAPIDS_X: case INTEL_FAM6_ATOM_GOLDMONT_D: case INTEL_FAM6_ATOM_TREMONT_D: return 1; + default: + return 0; } - return 0; } -static void dump_turbo_ratio_limits(int family, int model) +static void dump_turbo_ratio_limits(int trl_msr_offset, int family, int model) { unsigned long long msr, core_counts; - unsigned int ratio, group_size; + int shift; - get_msr(base_cpu, MSR_TURBO_RATIO_LIMIT, &msr); - fprintf(outf, "cpu%d: MSR_TURBO_RATIO_LIMIT: 0x%08llx\n", base_cpu, msr); + get_msr(base_cpu, trl_msr_offset, &msr); + fprintf(outf, "cpu%d: MSR_%sTURBO_RATIO_LIMIT: 0x%08llx\n", + base_cpu, trl_msr_offset == MSR_SECONDARY_TURBO_RATIO_LIMIT ? "SECONDARY" : "", msr); if (has_turbo_ratio_group_limits(family, model)) { get_msr(base_cpu, MSR_TURBO_RATIO_LIMIT1, &core_counts); @@ -2454,53 +2547,16 @@ static void dump_turbo_ratio_limits(int family, int model) core_counts = 0x0807060504030201; } - ratio = (msr >> 56) & 0xFF; - group_size = (core_counts >> 56) & 0xFF; - if (ratio) - fprintf(outf, "%d * %.1f = %.1f MHz max turbo %d active cores\n", - ratio, bclk, ratio * bclk, group_size); - - ratio = (msr >> 48) & 0xFF; - group_size = (core_counts >> 48) & 0xFF; - if (ratio) - fprintf(outf, "%d * %.1f = %.1f MHz max turbo %d active cores\n", - ratio, bclk, ratio * bclk, group_size); - - ratio = (msr >> 40) & 0xFF; - group_size = (core_counts >> 40) & 0xFF; - if (ratio) - fprintf(outf, "%d * %.1f = %.1f MHz max turbo %d active cores\n", - ratio, bclk, ratio * bclk, group_size); - - ratio = (msr >> 32) & 0xFF; - group_size = (core_counts >> 32) & 0xFF; - if (ratio) - fprintf(outf, "%d * %.1f = %.1f MHz max turbo %d active cores\n", - ratio, bclk, ratio * bclk, group_size); - - ratio = (msr >> 24) & 0xFF; - group_size = (core_counts >> 24) & 0xFF; - if (ratio) - fprintf(outf, "%d * %.1f = %.1f MHz max turbo %d active cores\n", - ratio, bclk, ratio * bclk, group_size); - - ratio = (msr >> 16) & 0xFF; - group_size = (core_counts >> 16) & 0xFF; - if (ratio) - fprintf(outf, "%d * %.1f = %.1f MHz max turbo %d active cores\n", - ratio, bclk, ratio * bclk, group_size); + for (shift = 56; shift >= 0; shift -= 8) { + unsigned int ratio, group_size; - ratio = (msr >> 8) & 0xFF; - group_size = (core_counts >> 8) & 0xFF; - if (ratio) - fprintf(outf, "%d * %.1f = %.1f MHz max turbo %d active cores\n", - ratio, bclk, ratio * bclk, group_size); + ratio = (msr >> shift) & 0xFF; + group_size = (core_counts >> shift) & 0xFF; + if (ratio) + fprintf(outf, "%d * %.1f = %.1f MHz max turbo %d active cores\n", + ratio, bclk, ratio * bclk, group_size); + } - ratio = (msr >> 0) & 0xFF; - group_size = (core_counts >> 0) & 0xFF; - if (ratio) - fprintf(outf, "%d * %.1f = %.1f MHz max turbo %d active cores\n", - ratio, bclk, ratio * bclk, group_size); return; } @@ -2913,7 +2969,7 @@ int get_thread_siblings(struct cpu_topology *thiscpu) } } } - } while (!strncmp(&character, ",", 1)); + } while (character == ','); fclose(filep); return CPU_COUNT_S(size, thiscpu->put_ids); @@ -3027,6 +3083,8 @@ void set_max_cpu_num(void) */ int count_cpus(int cpu) { + UNUSED(cpu); + topo.num_cpus++; return 0; } @@ -3361,6 +3419,9 @@ static int update_msr_sum(struct thread_data *t, struct core_data *c, struct pkg int i, ret; int cpu = t->cpu_id; + UNUSED(c); + UNUSED(p); + for (i = IDX_PKG_ENERGY; i < IDX_COUNT; i++) { unsigned long long msr_cur, msr_last; off_t offset; @@ -3387,6 +3448,8 @@ static int update_msr_sum(struct thread_data *t, struct core_data *c, struct pkg static void msr_record_handler(union sigval v) { + UNUSED(v); + for_all_cpus(update_msr_sum, EVEN_COUNTERS); } @@ -3439,6 +3502,9 @@ release_msr: /* * set_my_sched_priority(pri) * return previous + * + * if non-root, do this: + * # /sbin/setcap cap_sys_rawio,cap_sys_nice=+ep /usr/bin/turbostat */ int set_my_sched_priority(int priority) { @@ -3457,7 +3523,7 @@ int set_my_sched_priority(int priority) errno = 0; retval = getpriority(PRIO_PROCESS, 0); if (retval != priority) - err(-1, "getpriority(%d) != setpriority(%d)", retval, priority); + err(retval, "getpriority(%d) != setpriority(%d)", retval, priority); return original_priority; } @@ -3466,7 +3532,7 @@ void turbostat_loop() { int retval; int restarted = 0; - int done_iters = 0; + unsigned int done_iters = 0; setup_signal_handler(); @@ -3669,6 +3735,7 @@ int probe_nhm_msrs(unsigned int family, unsigned int model) has_misc_feature_control = 1; break; case INTEL_FAM6_SKYLAKE_X: /* SKX */ + case INTEL_FAM6_SAPPHIRERAPIDS_X: /* SPR */ pkg_cstate_limits = skx_pkg_cstate_limits; has_misc_feature_control = 1; break; @@ -3678,6 +3745,7 @@ int probe_nhm_msrs(unsigned int family, unsigned int model) break; case INTEL_FAM6_ATOM_SILVERMONT: /* BYT */ no_MSR_MISC_PWR_MGMT = 1; + /* FALLTHRU */ case INTEL_FAM6_ATOM_SILVERMONT_D: /* AVN */ pkg_cstate_limits = slv_pkg_cstate_limits; break; @@ -3721,6 +3789,9 @@ int has_slv_msrs(unsigned int family, unsigned int model) if (!genuine_intel) return 0; + if (family != 6) + return 0; + switch (model) { case INTEL_FAM6_ATOM_SILVERMONT: case INTEL_FAM6_ATOM_SILVERMONT_MID: @@ -3736,6 +3807,9 @@ int is_dnv(unsigned int family, unsigned int model) if (!genuine_intel) return 0; + if (family != 6) + return 0; + switch (model) { case INTEL_FAM6_ATOM_GOLDMONT_D: return 1; @@ -3749,6 +3823,9 @@ int is_bdx(unsigned int family, unsigned int model) if (!genuine_intel) return 0; + if (family != 6) + return 0; + switch (model) { case INTEL_FAM6_BROADWELL_X: return 1; @@ -3762,6 +3839,9 @@ int is_skx(unsigned int family, unsigned int model) if (!genuine_intel) return 0; + if (family != 6) + return 0; + switch (model) { case INTEL_FAM6_SKYLAKE_X: return 1; @@ -3775,6 +3855,9 @@ int is_icx(unsigned int family, unsigned int model) if (!genuine_intel) return 0; + if (family != 6) + return 0; + switch (model) { case INTEL_FAM6_ICELAKE_X: return 1; @@ -3782,11 +3865,30 @@ int is_icx(unsigned int family, unsigned int model) return 0; } +int is_spr(unsigned int family, unsigned int model) +{ + + if (!genuine_intel) + return 0; + + if (family != 6) + return 0; + + switch (model) { + case INTEL_FAM6_SAPPHIRERAPIDS_X: + return 1; + } + return 0; +} + int is_ehl(unsigned int family, unsigned int model) { if (!genuine_intel) return 0; + if (family != 6) + return 0; + switch (model) { case INTEL_FAM6_ATOM_TREMONT: return 1; @@ -3799,6 +3901,9 @@ int is_jvl(unsigned int family, unsigned int model) if (!genuine_intel) return 0; + if (family != 6) + return 0; + switch (model) { case INTEL_FAM6_ATOM_TREMONT_D: return 1; @@ -3811,6 +3916,9 @@ int has_turbo_ratio_limit(unsigned int family, unsigned int model) if (has_slv_msrs(family, model)) return 0; + if (family != 6) + return 0; + switch (model) { /* Nehalem compatible, but do not include turbo-ratio limit support */ case INTEL_FAM6_NEHALEM_EX: /* Nehalem-EX Xeon - Beckton */ @@ -3890,6 +3998,7 @@ int has_glm_turbo_ratio_limit(unsigned int family, unsigned int model) case INTEL_FAM6_ATOM_GOLDMONT: case INTEL_FAM6_SKYLAKE_X: case INTEL_FAM6_ICELAKE_X: + case INTEL_FAM6_SAPPHIRERAPIDS_X: return 1; default: return 0; @@ -3917,7 +4026,7 @@ int has_config_tdp(unsigned int family, unsigned int model) case INTEL_FAM6_CANNONLAKE_L: /* CNL */ case INTEL_FAM6_SKYLAKE_X: /* SKX */ case INTEL_FAM6_ICELAKE_X: /* ICX */ - + case INTEL_FAM6_SAPPHIRERAPIDS_X: /* SPR */ case INTEL_FAM6_XEON_PHI_KNL: /* Knights Landing */ return 1; default: @@ -3972,21 +4081,23 @@ static void remove_underbar(char *s) *to = 0; } -static void dump_cstate_pstate_config_info(unsigned int family, unsigned int model) +static void dump_turbo_ratio_info(unsigned int family, unsigned int model) { - if (!do_nhm_platform_info) + if (!has_turbo) return; - dump_nhm_platform_info(); - if (has_hsw_turbo_ratio_limit(family, model)) dump_hsw_turbo_ratio_limits(); if (has_ivt_turbo_ratio_limit(family, model)) dump_ivt_turbo_ratio_limits(); - if (has_turbo_ratio_limit(family, model)) - dump_turbo_ratio_limits(family, model); + if (has_turbo_ratio_limit(family, model)) { + dump_turbo_ratio_limits(MSR_TURBO_RATIO_LIMIT, family, model); + + if (is_hybrid) + dump_turbo_ratio_limits(MSR_SECONDARY_TURBO_RATIO_LIMIT, family, model); + } if (has_atom_turbo_ratio_limit(family, model)) dump_atom_turbo_ratio_limits(); @@ -3996,10 +4107,36 @@ static void dump_cstate_pstate_config_info(unsigned int family, unsigned int mod if (has_config_tdp(family, model)) dump_config_tdp(); +} + +static void dump_cstate_pstate_config_info(unsigned int family, unsigned int model) +{ + if (!do_nhm_platform_info) + return; + dump_nhm_platform_info(); + dump_turbo_ratio_info(family, model); dump_nhm_cst_cfg(); } +static int read_sysfs_int(char *path) +{ + FILE *input; + int retval = -1; + + input = fopen(path, "r"); + if (input == NULL) { + if (debug) + fprintf(outf, "NSFOD %s\n", path); + return (-1); + } + if (fscanf(input, "%d", &retval) != 1) + err(1, "%s: failed to read int from file", path); + fclose(input); + + return (retval); +} + static void dump_sysfs_file(char *path) { FILE *input; @@ -4018,6 +4155,48 @@ static void dump_sysfs_file(char *path) fprintf(outf, "%s: %s", strrchr(path, '/') + 1, cpuidle_buf); } +static void intel_uncore_frequency_probe(void) +{ + int i, j; + char path[128]; + + if (!genuine_intel) + return; + + if (access("/sys/devices/system/cpu/intel_uncore_frequency/package_00_die_00", R_OK)) + return; + + if (!access("/sys/devices/system/cpu/intel_uncore_frequency/package_00_die_00/current_freq_khz", R_OK)) + BIC_PRESENT(BIC_UNCORE_MHZ); + + if (quiet) + return; + + for (i = 0; i < topo.num_packages; ++i) { + for (j = 0; j < topo.num_die; ++j) { + int k, l; + + sprintf(path, "/sys/devices/system/cpu/intel_uncore_frequency/package_0%d_die_0%d/min_freq_khz", + i, j); + k = read_sysfs_int(path); + sprintf(path, "/sys/devices/system/cpu/intel_uncore_frequency/package_0%d_die_0%d/max_freq_khz", + i, j); + l = read_sysfs_int(path); + fprintf(outf, "Uncore Frequency pkg%d die%d: %d - %d MHz ", i, j, k / 1000, l / 1000); + + sprintf(path, + "/sys/devices/system/cpu/intel_uncore_frequency/package_0%d_die_0%d/initial_min_freq_khz", + i, j); + k = read_sysfs_int(path); + sprintf(path, + "/sys/devices/system/cpu/intel_uncore_frequency/package_0%d_die_0%d/initial_max_freq_khz", + i, j); + l = read_sysfs_int(path); + fprintf(outf, "(%d - %d MHz)\n", k / 1000, l / 1000); + } + } +} + static void dump_sysfs_cstate_config(void) { char path[64]; @@ -4125,6 +4304,9 @@ int print_epb(struct thread_data *t, struct core_data *c, struct pkg_data *p) char *epb_string; int cpu, epb; + UNUSED(c); + UNUSED(p); + if (!has_epb) return 0; @@ -4171,6 +4353,9 @@ int print_hwp(struct thread_data *t, struct core_data *c, struct pkg_data *p) unsigned long long msr; int cpu; + UNUSED(c); + UNUSED(p); + if (!has_hwp) return 0; @@ -4254,6 +4439,9 @@ int print_perf_limit(struct thread_data *t, struct core_data *c, struct pkg_data unsigned long long msr; int cpu; + UNUSED(c); + UNUSED(p); + cpu = t->cpu_id; /* per-package */ @@ -4359,6 +4547,8 @@ double get_tdp_intel(unsigned int model) double get_tdp_amd(unsigned int family) { + UNUSED(family); + /* This is the max stock TDP of HEDT/Server Fam17h+ chips */ return 280.0; } @@ -4376,6 +4566,7 @@ static double rapl_dram_energy_units_probe(int model, double rapl_energy_units) case INTEL_FAM6_BROADWELL_X: /* BDX */ case INTEL_FAM6_SKYLAKE_X: /* SKX */ case INTEL_FAM6_XEON_PHI_KNL: /* KNL */ + case INTEL_FAM6_ICELAKE_X: /* ICX */ return (rapl_dram_energy_units = 15.3 / 1000000); default: return (rapl_energy_units); @@ -4465,6 +4656,7 @@ void rapl_probe_intel(unsigned int family, unsigned int model) case INTEL_FAM6_BROADWELL_X: /* BDX */ case INTEL_FAM6_SKYLAKE_X: /* SKX */ case INTEL_FAM6_ICELAKE_X: /* ICX */ + case INTEL_FAM6_SAPPHIRERAPIDS_X: /* SPR */ case INTEL_FAM6_XEON_PHI_KNL: /* KNL */ do_rapl = RAPL_PKG | RAPL_DRAM | RAPL_DRAM_POWER_INFO | RAPL_DRAM_PERF_STATUS | RAPL_PKG_PERF_STATUS | @@ -4559,6 +4751,8 @@ void rapl_probe_amd(unsigned int family, unsigned int model) unsigned int has_rapl = 0; double tdp; + UNUSED(model); + if (max_extended_level >= 0x80000007) { __cpuid(0x80000007, eax, ebx, ecx, edx); /* RAPL (Fam 17h+) */ @@ -4617,6 +4811,7 @@ void perf_limit_reasons_probe(unsigned int family, unsigned int model) case INTEL_FAM6_HASWELL_L: /* HSW */ case INTEL_FAM6_HASWELL_G: /* HSW */ do_gfx_perf_limit_reasons = 1; + /* FALLTHRU */ case INTEL_FAM6_HASWELL_X: /* HSX */ do_core_perf_limit_reasons = 1; do_ring_perf_limit_reasons = 1; @@ -4627,13 +4822,19 @@ void perf_limit_reasons_probe(unsigned int family, unsigned int model) void automatic_cstate_conversion_probe(unsigned int family, unsigned int model) { - if (is_skx(family, model) || is_bdx(family, model) || is_icx(family, model)) + if (family != 6) + return; + + switch (model) { + case INTEL_FAM6_BROADWELL_X: + case INTEL_FAM6_SKYLAKE_X: has_automatic_cstate_conversion = 1; + } } void prewake_cstate_probe(unsigned int family, unsigned int model) { - if (is_icx(family, model)) + if (is_icx(family, model) || is_spr(family, model)) dis_cstate_prewake = 1; } @@ -4643,6 +4844,9 @@ int print_thermal(struct thread_data *t, struct core_data *c, struct pkg_data *p unsigned int dts, dts2; int cpu; + UNUSED(c); + UNUSED(p); + if (!(do_dts || do_ptm)) return 0; @@ -4698,7 +4902,7 @@ int print_thermal(struct thread_data *t, struct core_data *c, struct pkg_data *p void print_power_limit_msr(int cpu, unsigned long long msr, char *label) { - fprintf(outf, "cpu%d: %s: %sabled (%f Watts, %f sec, clamp %sabled)\n", + fprintf(outf, "cpu%d: %s: %sabled (%0.3f Watts, %f sec, clamp %sabled)\n", cpu, label, ((msr >> 15) & 1) ? "EN" : "DIS", ((msr >> 0) & 0x7FFF) * rapl_power_units, @@ -4714,6 +4918,9 @@ int print_rapl(struct thread_data *t, struct core_data *c, struct pkg_data *p) const char *msr_name; int cpu; + UNUSED(c); + UNUSED(p); + if (!do_rapl) return 0; @@ -4762,12 +4969,19 @@ int print_rapl(struct thread_data *t, struct core_data *c, struct pkg_data *p) cpu, msr, (msr >> 63) & 1 ? "" : "UN"); print_power_limit_msr(cpu, msr, "PKG Limit #1"); - fprintf(outf, "cpu%d: PKG Limit #2: %sabled (%f Watts, %f* sec, clamp %sabled)\n", + fprintf(outf, "cpu%d: PKG Limit #2: %sabled (%0.3f Watts, %f* sec, clamp %sabled)\n", cpu, ((msr >> 47) & 1) ? "EN" : "DIS", ((msr >> 32) & 0x7FFF) * rapl_power_units, (1.0 + (((msr >> 54) & 0x3) / 4.0)) * (1 << ((msr >> 49) & 0x1F)) * rapl_time_units, ((msr >> 48) & 1) ? "EN" : "DIS"); + + if (get_msr(cpu, MSR_VR_CURRENT_CONFIG, &msr)) + return -9; + + fprintf(outf, "cpu%d: MSR_VR_CURRENT_CONFIG: 0x%08llx\n", cpu, msr); + fprintf(outf, "cpu%d: PKG Limit #4: %f Watts (%slocked)\n", + cpu, ((msr >> 0) & 0x1FFF) * rapl_power_units, (msr >> 31) & 1 ? "" : "UN"); } if (do_rapl & RAPL_DRAM_POWER_INFO) { @@ -4830,6 +5044,9 @@ int has_snb_msrs(unsigned int family, unsigned int model) if (!genuine_intel) return 0; + if (family != 6) + return 0; + switch (model) { case INTEL_FAM6_SANDYBRIDGE: case INTEL_FAM6_SANDYBRIDGE_X: @@ -4846,6 +5063,7 @@ int has_snb_msrs(unsigned int family, unsigned int model) case INTEL_FAM6_CANNONLAKE_L: /* CNL */ case INTEL_FAM6_SKYLAKE_X: /* SKX */ case INTEL_FAM6_ICELAKE_X: /* ICX */ + case INTEL_FAM6_SAPPHIRERAPIDS_X: /* SPR */ case INTEL_FAM6_ATOM_GOLDMONT: /* BXT */ case INTEL_FAM6_ATOM_GOLDMONT_PLUS: case INTEL_FAM6_ATOM_GOLDMONT_D: /* DNV */ @@ -4873,6 +5091,9 @@ int has_c8910_msrs(unsigned int family, unsigned int model) if (!genuine_intel) return 0; + if (family != 6) + return 0; + switch (model) { case INTEL_FAM6_HASWELL_L: /* HSW */ case INTEL_FAM6_BROADWELL: /* BDW */ @@ -4899,6 +5120,9 @@ int has_skl_msrs(unsigned int family, unsigned int model) if (!genuine_intel) return 0; + if (family != 6) + return 0; + switch (model) { case INTEL_FAM6_SKYLAKE_L: /* SKL */ case INTEL_FAM6_CANNONLAKE_L: /* CNL */ @@ -4911,6 +5135,10 @@ int is_slm(unsigned int family, unsigned int model) { if (!genuine_intel) return 0; + + if (family != 6) + return 0; + switch (model) { case INTEL_FAM6_ATOM_SILVERMONT: /* BYT */ case INTEL_FAM6_ATOM_SILVERMONT_D: /* AVN */ @@ -4923,6 +5151,10 @@ int is_knl(unsigned int family, unsigned int model) { if (!genuine_intel) return 0; + + if (family != 6) + return 0; + switch (model) { case INTEL_FAM6_XEON_PHI_KNL: /* KNL */ return 1; @@ -4935,6 +5167,9 @@ int is_cnl(unsigned int family, unsigned int model) if (!genuine_intel) return 0; + if (family != 6) + return 0; + switch (model) { case INTEL_FAM6_CANNONLAKE_L: /* CNL */ return 1; @@ -4989,6 +5224,9 @@ int get_cpu_type(struct thread_data *t, struct core_data *c, struct pkg_data *p) { unsigned int eax, ebx, ecx, edx; + UNUSED(c); + UNUSED(p); + if (!genuine_intel) return 0; @@ -5025,6 +5263,9 @@ int set_temperature_target(struct thread_data *t, struct core_data *c, struct pk unsigned int tcc_default, tcc_offset; int cpu; + UNUSED(c); + UNUSED(p); + /* tj_max is used only for dts or ptm */ if (!(do_dts || do_ptm)) return 0; @@ -5209,13 +5450,18 @@ unsigned int intel_model_duplicates(unsigned int model) case INTEL_FAM6_LAKEFIELD: case INTEL_FAM6_ALDERLAKE: case INTEL_FAM6_ALDERLAKE_L: + case INTEL_FAM6_ALDERLAKE_N: + case INTEL_FAM6_RAPTORLAKE: + case INTEL_FAM6_RAPTORLAKE_P: + case INTEL_FAM6_RAPTORLAKE_S: + case INTEL_FAM6_METEORLAKE: + case INTEL_FAM6_METEORLAKE_L: return INTEL_FAM6_CANNONLAKE_L; case INTEL_FAM6_ATOM_TREMONT_L: return INTEL_FAM6_ATOM_TREMONT; case INTEL_FAM6_ICELAKE_D: - case INTEL_FAM6_SAPPHIRERAPIDS_X: return INTEL_FAM6_ICELAKE_X; } return model; @@ -5246,7 +5492,7 @@ void print_dev_latency(void) } /* - * Linux-perf manages the the HW instructions-retired counter + * Linux-perf manages the HW instructions-retired counter * by enabling when requested, and hiding rollover */ void linux_perf_init(void) @@ -5268,7 +5514,6 @@ void process_cpuid() { unsigned int eax, ebx, ecx, edx; unsigned int fms, family, model, stepping, ecx_flags, edx_flags; - unsigned int has_turbo; unsigned long long ucode_patch = 0; eax = ebx = ecx = edx = 0; @@ -5391,7 +5636,10 @@ void process_cpuid() __cpuid_count(0x7, 0, eax, ebx, ecx, edx); has_sgx = ebx & (1 << 2); - fprintf(outf, "CPUID(7): %sSGX\n", has_sgx ? "" : "No-"); + + is_hybrid = edx & (1 << 15); + + fprintf(outf, "CPUID(7): %sSGX %sHybrid\n", has_sgx ? "" : "No-", is_hybrid ? "" : "No-"); if (has_sgx) decode_feature_control_msr(); @@ -5502,7 +5750,7 @@ void process_cpuid() BIC_NOT_PRESENT(BIC_Pkgpc7); use_c1_residency_msr = 1; } - if (is_skx(family, model) || is_icx(family, model)) { + if (is_skx(family, model) || is_icx(family, model) || is_spr(family, model)) { BIC_NOT_PRESENT(BIC_CPU_c3); BIC_NOT_PRESENT(BIC_Pkgpc3); BIC_NOT_PRESENT(BIC_CPU_c7); @@ -5547,6 +5795,7 @@ void process_cpuid() if (!quiet) dump_cstate_pstate_config_info(family, model); + intel_uncore_frequency_probe(); if (!quiet) print_dev_latency(); @@ -5572,6 +5821,11 @@ void process_cpuid() else BIC_NOT_PRESENT(BIC_CPU_LPI); + if (!access("/sys/devices/system/cpu/cpu0/thermal_throttle/core_throttle_count", R_OK)) + BIC_PRESENT(BIC_CORE_THROT_CNT); + else + BIC_NOT_PRESENT(BIC_CORE_THROT_CNT); + if (!access(sys_lpi_file_sysfs, R_OK)) { sys_lpi_file = sys_lpi_file_sysfs; BIC_PRESENT(BIC_SYS_LPI); @@ -5601,11 +5855,6 @@ int dir_filter(const struct dirent *dirp) return 0; } -int open_dev_cpu_msr(int dummy1) -{ - return 0; -} - void topology_probe() { int i; @@ -5896,6 +6145,9 @@ void turbostat_init() if (!quiet && do_irtl_snb) print_irtl(); + + if (DO_BIC(BIC_IPC)) + (void)get_instr_count_fd(base_cpu); } int fork_it(char **argv) @@ -5973,7 +6225,30 @@ int get_and_dump_counters(void) void print_version() { - fprintf(outf, "turbostat version 21.05.04" " - Len Brown <lenb@kernel.org>\n"); + fprintf(outf, "turbostat version 2022.10.04 - Len Brown <lenb@kernel.org>\n"); +} + +#define COMMAND_LINE_SIZE 2048 + +void print_bootcmd(void) +{ + char bootcmd[COMMAND_LINE_SIZE]; + FILE *fp; + int ret; + + memset(bootcmd, 0, COMMAND_LINE_SIZE); + fp = fopen("/proc/cmdline", "r"); + if (!fp) + return; + + ret = fread(bootcmd, sizeof(char), COMMAND_LINE_SIZE - 1, fp); + if (ret) { + bootcmd[ret] = '\0'; + /* the last character is already '\n' */ + fprintf(outf, "Kernel command line: %s", bootcmd); + } + + fclose(fp); } int add_counter(unsigned int msr_num, char *path, char *name, @@ -6138,6 +6413,16 @@ next: } } +int is_deferred_add(char *name) +{ + int i; + + for (i = 0; i < deferred_add_index; ++i) + if (!strcmp(name, deferred_add_names[i])) + return 1; + return 0; +} + int is_deferred_skip(char *name) { int i; @@ -6156,9 +6441,6 @@ void probe_sysfs(void) int state; char *sp; - if (!DO_BIC(BIC_sysfs)) - return; - for (state = 10; state >= 0; --state) { sprintf(path, "/sys/devices/system/cpu/cpu%d/cpuidle/state%d/name", base_cpu, state); @@ -6181,6 +6463,9 @@ void probe_sysfs(void) sprintf(path, "cpuidle/state%d/time", state); + if (!DO_BIC(BIC_sysfs) && !is_deferred_add(name_buf)) + continue; + if (is_deferred_skip(name_buf)) continue; @@ -6206,6 +6491,9 @@ void probe_sysfs(void) sprintf(path, "cpuidle/state%d/usage", state); + if (!DO_BIC(BIC_sysfs) && !is_deferred_add(name_buf)) + continue; + if (is_deferred_skip(name_buf)) continue; @@ -6313,6 +6601,7 @@ void cmdline(int argc, char **argv) { "interval", required_argument, 0, 'i' }, { "IPC", no_argument, 0, 'I' }, { "num_iterations", required_argument, 0, 'n' }, + { "header_iterations", required_argument, 0, 'N' }, { "help", no_argument, 0, 'h' }, { "hide", required_argument, 0, 'H' }, // meh, -h taken by --help { "Joules", no_argument, 0, 'J' }, @@ -6394,6 +6683,14 @@ void cmdline(int argc, char **argv) exit(2); } break; + case 'N': + header_iterations = strtod(optarg, NULL); + + if (header_iterations <= 0) { + fprintf(outf, "iterations %d should be positive number\n", header_iterations); + exit(2); + } + break; case 's': /* * --show: show only those specified @@ -6425,13 +6722,17 @@ int main(int argc, char **argv) outf = stderr; cmdline(argc, argv); - if (!quiet) + if (!quiet) { print_version(); + print_bootcmd(); + } probe_sysfs(); turbostat_init(); + msr_sum_record(); + /* dump counters and exit */ if (dump_only) return get_and_dump_counters(); @@ -6443,7 +6744,6 @@ int main(int argc, char **argv) return 0; } - msr_sum_record(); /* * if any params left, it must be a command to fork */ |