| Message ID | 20240125072214.318382-4-aharivel@redhat.com |
|---|---|
| State | New |
| Headers | show |
| Series | Add support for the RAPL MSRs series | expand |
On Thu, Jan 25, 2024 at 08:22:14AM +0100, Anthony Harivel wrote: > diff --git a/docs/specs/rapl-msr.rst b/docs/specs/rapl-msr.rst > new file mode 100644 > index 000000000000..04d27c198fc0 > --- /dev/null > +++ b/docs/specs/rapl-msr.rst > @@ -0,0 +1,133 @@ > +================ > +RAPL MSR support > +================ > + > +Current Limitations > +------------------- > + > +- Works only on Intel host CPUs because AMD CPUs are using different MSR > + addresses. The privileged helper program is validating an allow list of MSRs. If those MSRs are only correct on Intel hosts, then the validation is incomplete, and it could be allowing unprivileged processes on AMD hosts to access forbidden MSRS whose address happen to clash with the Intel RAPL MSRs. IOW, the privileged helper needs to call cpuid() and validate that the current host vendor is Intel. I suspect we also need a feature check of some kind to validate that the intel processor supports this features, since old ones definitely didn't, and we shouldn't assume all future ones will either. > diff --git a/include/sysemu/kvm_int.h b/include/sysemu/kvm_int.h > index 882e37e12c5b..ee2fe8817833 100644 > --- a/include/sysemu/kvm_int.h > +++ b/include/sysemu/kvm_int.h > diff --git a/target/i386/kvm/kvm.c b/target/i386/kvm/kvm.c > index 76a66246eb72..e6cb315c0a90 100644 > --- a/target/i386/kvm/kvm.c > +++ b/target/i386/kvm/kvm.c > @@ -16,16 +16,22 @@ > #include "qapi/qapi-events-run-state.h" > #include "qapi/error.h" > #include "qapi/visitor.h" > +#include <math.h> > +#include <stdint.h> stdint.h always gets included via osdep.h > #include <sys/ioctl.h> > #include <sys/utsname.h> > #include <sys/syscall.h> > +#include <sys/resource.h> > +#include <sys/time.h> > > #include <linux/kvm.h> > +#include <unistd.h> unistd.h also always gets included via osdep.h > #include "standard-headers/asm-x86/kvm_para.h" > #include "hw/xen/interface/arch-x86/cpuid.h" > > #include "cpu.h" > #include "host-cpu.h" > +#include "vmsr_energy.h" > #include "sysemu/sysemu.h" > #include "sysemu/hw_accel.h" > #include "sysemu/kvm_int.h" > @@ -2509,6 +2558,265 @@ static void register_smram_listener(Notifier *n, void *unused) > &smram_address_space, 1, "kvm-smram"); > } > > +static void *kvm_msr_energy_thread(void *data) > +{ > + KVMState *s = data; > + struct KVMMsrEnergy *vmsr = &s->msr_energy; > + > + g_autofree package_energy_stat *pkg_stat = NULL; > + g_autofree thread_stat *thd_stat = NULL; > + g_autofree pid_t *thread_ids = NULL; > + g_autofree CPUState *cpu = NULL; > + unsigned int maxpkgs, maxcpus, maxticks; > + g_autofree unsigned int *vpkgs_energy_stat = NULL; > + unsigned int num_threads = 0; > + unsigned int tmp_num_threads = 0; > + pid_t pid; > + > + X86CPUTopoIDs topo_ids; > + > + > + rcu_register_thread(); > + > + /* Get QEMU PID*/ > + pid = getpid(); > + > + /* Nb of CPUS per packages */ > + maxcpus = vmsr_get_maxcpus(0); > + > + /* Nb of Physical Packages on the system */ > + maxpkgs = vmsr_get_max_physical_package(maxcpus); > + > + /* Those MSR values should not change as well */ > + vmsr->msr_unit = vmsr_read_msr(MSR_RAPL_POWER_UNIT, 0, pid, > + s->msr_energy.socket_path); > + vmsr->msr_limit = vmsr_read_msr(MSR_PKG_POWER_LIMIT, 0, pid, > + s->msr_energy.socket_path); > + vmsr->msr_info = vmsr_read_msr(MSR_PKG_POWER_INFO, 0, pid, > + s->msr_energy.socket_path); > + > + /* Allocate memory for each package energy status */ > + pkg_stat = (package_energy_stat *) > + g_new0(package_energy_stat, maxpkgs); Return value of g_new0 shouldn't need casting. > + > + /* Pre-allocate memory for thread stats */ > + thd_stat = g_new0(thread_stat, 1); > + > + /* Pre-allocate memory for holding Virtual Package Energy counter */ > + vpkgs_energy_stat = g_new0(unsigned int, vmsr->sockets); > + > + /* > + * Max numbers of ticks per package > + * time in second * number of ticks/second * Number of cores / package > + * ex: for 100 ticks/second/CPU, 12 CPUs per Package gives 1200 ticks max > + */ > + maxticks = (MSR_ENERGY_THREAD_SLEEP_US / 1000000) > + * sysconf(_SC_CLK_TCK) * maxcpus; > + > + while (true) { > + /* Get all qemu threads id */ > + thread_ids = vmsr_get_thread_ids(pid, &num_threads); > + > + if (thread_ids == NULL) { > + goto clean; > + } > + > + if (tmp_num_threads < num_threads) { > + thd_stat = g_renew(thread_stat, thd_stat, num_threads); > + } > + > + tmp_num_threads = num_threads; > + > + /* Populate all the thread stats */ > + for (int i = 0; i < num_threads; i++) { > + thd_stat[i].utime = g_new0(unsigned long long, 2); > + thd_stat[i].stime = g_new0(unsigned long long, 2); > + thd_stat[i].thread_id = thread_ids[i]; > + vmsr_read_thread_stat(&thd_stat[i], pid, 0); > + thd_stat[i].numa_node_id = numa_node_of_cpu(thd_stat[i].cpu_id); > + } > + > + /* Retrieve all packages power plane energy counter */ > + for (int i = 0; i <= maxpkgs; i++) { > + for (int j = 0; j < num_threads; j++) { > + /* > + * Use the first thread we found that ran on the CPU > + * of the package to read the packages energy counter > + */ This says we're using a thread ID > + if (thd_stat[j].numa_node_id == i) { > + pkg_stat[i].e_start = > + vmsr_read_msr(MSR_PKG_ENERGY_STATUS, i, pid, but here we're using a pid ID, which is the thread ID of the initial thread. > + s->msr_energy.socket_path); > + break; > + } > + } > + } This API design for vmsr_read_msr() is incredibly inefficient. We're making (maxpkgs * num_threads) calls to vmsr_read_msr(), and every one of those is opening and closing the socket. Why isn't QEMU opening the socket once and then sending all the requests over the same socket ? > + > + /* Sleep a short period while the other threads are working */ > + usleep(MSR_ENERGY_THREAD_SLEEP_US); > + > + /* > + * Retrieve all packages power plane energy counter > + * Calculate the delta of all packages > + */ > + for (int i = 0; i <= maxpkgs; i++) { > + for (int j = 0; j < num_threads; j++) { > + /* > + * Use the first thread we found that ran on the CPU > + * of the package to read the packages energy counter > + */ > + if (thd_stat[j].numa_node_id == i) { > + pkg_stat[i].e_end = > + vmsr_read_msr(MSR_PKG_ENERGY_STATUS, > + thd_stat[j].cpu_id, This is passing the 'cpu_id' field, but in the previous call to vmsr_read_msr 20 lines earlier, we seemd to pass a value that was the NUMA node id. > + thd_stat[j].thread_id, This is now passing a thread_id, while previously we passed the PID id. > + s->msr_energy.socket_path); > + /* > + * Prevent the case we have migrate the VM > + * during the sleep period or any other cases > + * were energy counter might be lower after > + * the sleep. > + */ > + if (pkg_stat[i].e_end > pkg_stat[i].e_start) { > + pkg_stat[i].e_delta = > + pkg_stat[i].e_end - pkg_stat[i].e_start; > + } else { > + pkg_stat[i].e_delta = 0; > + } > + break; > + } > + } > + } > + > + /* Delta of ticks spend by each thread between the sample */ > + for (int i = 0; i < num_threads; i++) { > + if (vmsr_read_thread_stat(&thd_stat[i], pid, 1) != 0) { > + /* > + * We don't count the dead thread > + * i.e threads that existed before the sleep > + * and not anymore > + */ > + thd_stat[i].delta_ticks = 0; > + } else { > + vmsr_delta_ticks(thd_stat, i); > + } > + } > + > + /* > + * Identify the vCPU threads > + * Calculate the Number of vCPU per package > + */ > + CPU_FOREACH(cpu) { > + for (int i = 0; i < num_threads; i++) { > + if (cpu->thread_id == thd_stat[i].thread_id) { > + thd_stat[i].is_vcpu = true; > + thd_stat[i].vcpu_id = cpu->cpu_index; > + pkg_stat[thd_stat[i].numa_node_id].nb_vcpu++; > + thd_stat[i].acpi_id = kvm_arch_vcpu_id(cpu); > + break; > + } > + } > + } > + > + /* Retrieve the virtual package number of each vCPU */ > + for (int i = 0; i < vmsr->x86_cpu_list->len; i++) { > + for (int j = 0; j < num_threads; j++) { > + if ((thd_stat[j].acpi_id == vmsr->x86_cpu_list->cpus[i].arch_id) > + && (thd_stat[j].is_vcpu == true)) { > + x86_topo_ids_from_apicid(thd_stat[j].acpi_id, > + &vmsr->topo_info, &topo_ids); > + thd_stat[j].vpkg = topo_ids.pkg_id; > + } > + } > + } > + > + /* Calculate the total energy of all non-vCPU thread */ > + for (int i = 0; i < num_threads; i++) { > + double temp; > + if ((thd_stat[i].is_vcpu != true) && > + (thd_stat[i].delta_ticks > 0)) { > + temp = vmsr_get_ratio(pkg_stat, thd_stat, maxticks, i); > + pkg_stat[thd_stat[i].numa_node_id].e_ratio > + += (uint64_t)lround(temp); > + } > + } > + > + /* Calculate the ratio per non-vCPU thread of each package */ > + for (int i = 0; i <= maxpkgs; i++) { > + if (pkg_stat[i].nb_vcpu > 0) { > + pkg_stat[i].e_ratio = pkg_stat[i].e_ratio / pkg_stat[i].nb_vcpu; > + } > + } > + > + /* > + * Calculate the energy for each Package: > + * Energy Package = sum of each vCPU energy that belongs to the package > + */ > + for (int i = 0; i < num_threads; i++) { > + double temp; > + > + if ((thd_stat[i].is_vcpu == true) && \ > + (thd_stat[i].delta_ticks > 0)) { > + temp = vmsr_get_ratio(pkg_stat, thd_stat, maxticks, i); > + > + vpkgs_energy_stat[thd_stat[i].vpkg] += (uint64_t)lround(temp); > + vpkgs_energy_stat[thd_stat[i].vpkg] += > + pkg_stat[thd_stat[i].numa_node_id].e_ratio; > + } > + } > + > + /* > + * Finally populate the vmsr register of each vCPU with the total > + * package value to emulate the real hardware where each CPU return the > + * value of the package it belongs. > + */ > + for (int i = 0; i < num_threads; i++) { > + if ((thd_stat[i].is_vcpu == true) && \ > + (thd_stat[i].delta_ticks > 0)) { > + vmsr->msr_value[thd_stat[i].vcpu_id] = \ > + vpkgs_energy_stat[thd_stat[i].vpkg]; > + } > + } > + > + /* Zero out the memory */ > + for (int i = 0; i < num_threads; i++) { > + memset(thd_stat[i].utime, 0, 2 * sizeof(unsigned long long)); > + memset(thd_stat[i].stime, 0, 2 * sizeof(unsigned long long)); > + } > + memset(thd_stat, 0, num_threads * sizeof(thread_stat)); > + memset(thread_ids, 0, sizeof(pid_t)); > + } > + > +clean: > + rcu_unregister_thread(); > + return NULL; > +} > + > +static int kvm_msr_energy_thread_init(KVMState *s, MachineState *ms) > +{ > + struct KVMMsrEnergy *r = &s->msr_energy; > + > + vmsr_init_topo_info(&r->topo_info, ms); > + > + /* Retrieve the number of vCPU */ > + r->cpus = ms->smp.cpus; > + > + /* Retrieve the number of sockets */ > + r->sockets = ms->smp.sockets; > + > + /* Allocate register memory (MSR_PKG_STATUS) for each vCPU */ > + r->msr_value = g_new0(uint64_t, r->cpus); > + > + /* Retrieve the CPUArchIDlist */ > + r->x86_cpu_list = x86_possible_cpu_arch_ids(ms); > + > + qemu_thread_create(&r->msr_thr, "kvm-msr", > + kvm_msr_energy_thread, > + s, QEMU_THREAD_JOINABLE); > + > + return 0; > +} > + > int kvm_arch_get_default_type(MachineState *ms) > { > return 0; > @@ -2711,6 +3019,46 @@ int kvm_arch_init(MachineState *ms, KVMState *s) > strerror(-ret)); > exit(1); > } > + > + if (s->msr_energy.enable == true) { > + > + r = kvm_filter_msr(s, MSR_RAPL_POWER_UNIT, > + kvm_rdmsr_rapl_power_unit, NULL); > + if (!r) { > + error_report("Could not install MSR_RAPL_POWER_UNIT \ > + handler: %s", > + strerror(-ret)); > + exit(1); > + } > + > + r = kvm_filter_msr(s, MSR_PKG_POWER_LIMIT, > + kvm_rdmsr_pkg_power_limit, NULL); > + if (!r) { > + error_report("Could not install MSR_PKG_POWER_LIMIT \ > + handler: %s", > + strerror(-ret)); > + exit(1); > + } > + > + r = kvm_filter_msr(s, MSR_PKG_POWER_INFO, > + kvm_rdmsr_pkg_power_info, NULL); > + if (!r) { > + error_report("Could not install MSR_PKG_POWER_INFO \ > + handler: %s", > + strerror(-ret)); > + exit(1); > + } > + r = kvm_filter_msr(s, MSR_PKG_ENERGY_STATUS, > + kvm_rdmsr_pkg_energy_status, NULL); > + if (!r) { > + error_report("Could not install MSR_PKG_ENERGY_STATUS \ > + handler: %s", > + strerror(-ret)); > + exit(1); > + } else { > + kvm_msr_energy_thread_init(s, ms); > + } > + } > } > > return 0; > diff --git a/target/i386/kvm/meson.build b/target/i386/kvm/meson.build > index 84d9143e6029..16010638df69 100644 > --- a/target/i386/kvm/meson.build > +++ b/target/i386/kvm/meson.build > @@ -3,6 +3,7 @@ i386_kvm_ss = ss.source_set() > i386_kvm_ss.add(files( > 'kvm.c', > 'kvm-cpu.c', > + 'vmsr_energy.c', > )) > > i386_kvm_ss.add(when: 'CONFIG_XEN_EMU', if_true: files('xen-emu.c')) > diff --git a/target/i386/kvm/vmsr_energy.c b/target/i386/kvm/vmsr_energy.c > new file mode 100644 > index 000000000000..27588630efa4 > --- /dev/null > +++ b/target/i386/kvm/vmsr_energy.c > @@ -0,0 +1,295 @@ > +/* > + * QEMU KVM support -- x86 virtual energy-related MSR. > + * > + * Copyright 2023 Red Hat, Inc. 2023 > + * > + * Author: > + * Anthony Harivel <aharivel@redhat.com> > + * > + * This work is licensed under the terms of the GNU GPL, version 2 or later. > + * See the COPYING file in the top-level directory. > + * > + */ > + > +#include "vmsr_energy.h" > +#include "qapi/error.h" > +#include "io/channel.h" > +#include "io/channel-socket.h" > +#include "hw/boards.h" > + > +#define MAX_PATH_LEN 256 > +#define MAX_LINE_LEN 500 > + > +static char *compute_default_paths(void) > +{ > + g_autofree char *state = qemu_get_local_state_dir(); > + > + return g_build_filename(state, "run", "qemu-vmsr-helper.sock", NULL); > +} > + > +static int vmsr_helper_socket_read(QIOChannel *ioc, > + void *buf, int sz, Error **errp) > +{ > + ssize_t r = qio_channel_read_all(ioc, buf, sz, errp); > + > + if (r < 0) { > + object_unref(OBJECT(ioc)); > + ioc = NULL; > + return -EINVAL; > + } > + > + return 0; > +} > + > +static int vmsr_helper_socket_write(QIOChannel *ioc, > + int fd, > + const void *buf, int sz, Error **errp) > +{ > + size_t nfds = (fd != -1); > + while (sz > 0) { > + struct iovec iov; > + ssize_t n_written; > + > + iov.iov_base = (void *)buf; > + iov.iov_len = sz; > + n_written = qio_channel_writev_full(QIO_CHANNEL(ioc), &iov, 1, > + nfds ? &fd : NULL, nfds, 0, errp); > + > + if (n_written <= 0) { > + assert(n_written != QIO_CHANNEL_ERR_BLOCK); > + object_unref(OBJECT(ioc)); > + ioc = NULL; > + return n_written < 0 ? -EINVAL : 0; > + } > + > + nfds = 0; > + buf += n_written; > + sz -= n_written; > + } > + > + return 0; > +} THis method seems pointless. It is being called with fd == -1. The caller should just directly call qio_channel_write_all which will write all data. > + > +uint64_t vmsr_read_msr(uint32_t reg, unsigned int cpu_id, uint32_t tid, > + const char *path) > +{ > + uint64_t data = 0; > + char *socket_path = NULL; > + unsigned int buffer[3]; > + > + if (path == NULL) { > + socket_path = compute_default_paths(); > + } else { > + socket_path = g_strdup(path); > + } > + > + SocketAddress saddr = { > + .type = SOCKET_ADDRESS_TYPE_UNIX, > + .u.q_unix.path = socket_path > + }; > + QIOChannelSocket *sioc = qio_channel_socket_new(); > + Error *local_err = NULL; > + > + int r; > + > + qio_channel_set_name(QIO_CHANNEL(sioc), "vmsr-helper"); > + qio_channel_socket_connect_sync(sioc, > + &saddr, > + &local_err); > + g_free(socket_path); > + if (local_err) { > + goto out_close; > + } > + > + /* > + * Send the required arguments: > + * 1. RAPL MSR register to read > + * 2. On which CPU ID > + * 3. From which vCPU (Thread ID) > + */ > + buffer[0] = reg; > + buffer[1] = cpu_id; > + buffer[2] = tid; IMHO it is nicer to declare a packed struct for the fields being sent, and share its decl between the client and server code. > + > + r = vmsr_helper_socket_write(QIO_CHANNEL(sioc), > + -1, > + &buffer, sizeof(buffer), > + &local_err); > + if (r < 0) { > + goto out_close; > + } > + > + r = vmsr_helper_socket_read(QIO_CHANNEL(sioc), > + &data, sizeof(data), > + &local_err); > + if (r < 0) { > + data = 0; > + goto out_close; > + } > + > +out_close: > + /* Close socket. */ > + qio_channel_close(QIO_CHANNEL(sioc), NULL); > + object_unref(OBJECT(sioc)); > + return data; > +} > + > +/* Retrieve the max number of physical CPU on the package */ > +unsigned int vmsr_get_maxcpus(unsigned int package_num) > +{ > + int k, ncpus; > + unsigned int maxcpus; > + struct bitmask *cpus; > + > + cpus = numa_allocate_cpumask(); > + ncpus = cpus->size; > + > + if (numa_node_to_cpus(package_num, cpus) < 0) { > + return 0; > + } > + > + maxcpus = 0; > + for (k = 0; k < ncpus; k++) { > + if (numa_bitmask_isbitset(cpus, k)) { > + maxcpus++; > + } > + } > + > + return maxcpus; > +} > + > +/* Retrieve the maximum number of physical packages */ > +unsigned int vmsr_get_max_physical_package(unsigned int max_cpus) > +{ > + unsigned int packageCount = 0; > + const char *dir = "/sys/devices/system/cpu/"; > + int *uniquePackages; Mark this as g_autofree() to eliminuate the multiple calls go g_free()in each exit path > + > + char *filePath; > + FILE *file; > + > + uniquePackages = g_new0(int, max_cpus); > + > + for (int i = 0; i < max_cpus; i++) { > + filePath = g_build_filename(dir, g_strdup_printf("cpu%d", i), > + "topology/physical_package_id", NULL); Leaking the g_strdup_printf result. Also filePath is allocated on every iteration, but only freed outside the loop, so another leak. Declare filePath inside the loop and mark it g_autofree > + > + file = fopen(filePath, "r"); > + > + if (file == NULL) { > + perror("Error opening file"); > + g_free(filePath); > + g_free(uniquePackages); > + return 0; > + } > + > + char packageId[10]; > + if (fgets(packageId, sizeof(packageId), file) == NULL) { > + packageCount = 0; > + } > + > + fclose(file); > + > + int currentPackageId = atoi(packageId); > + > + bool isUnique = true; > + for (int j = 0; j < packageCount; j++) { > + if (uniquePackages[j] == currentPackageId) { > + isUnique = false; > + break; > + } > + } > + > + if (isUnique) { > + uniquePackages[packageCount] = currentPackageId; > + packageCount++; > + > + if (packageCount >= max_cpus) { > + break; > + } > + } > + } > + > + g_free(filePath); > + g_free(uniquePackages); > + return (packageCount == 0) ? 1 : packageCount; > +} > +int vmsr_read_thread_stat(struct thread_stat *thread, int pid, int index) > +{ > + char *path; > + path = g_new0(char, MAX_PATH_LEN); This pointer is overwritten on the very next line, so is leaked > + > + path = g_build_filename(g_strdup_printf("/proc/%u/task/%d/stat", pid, \ > + thread->thread_id), NULL); This result is never freed, and the intermediate g_strdup_printf() is also not freed. Using g_build_filename is pointless if you're going to use g_strdup_printf() to buld the entire filename in one go. > + > + FILE *file = fopen(path, "r"); > + if (file == NULL) { > + return -1; > + } > + > + if (fscanf(file, "%*d (%*[^)]) %*c %*d %*d %*d %*d %*d %*u %*u %*u %*u %*u" > + " %llu %llu %*d %*d %*d %*d %*d %*d %*u %*u %*d %*u %*u" > + " %*u %*u %*u %*u %*u %*u %*u %*u %*u %*d %*u %*u %u", > + &thread->utime[index], &thread->stime[index], &thread->cpu_id) != 3) > + return -1; > + > + fclose(file); > + return 0; > +} > + > +/* Read QEMU stat task folder to retrieve all QEMU threads ID */ > +pid_t *vmsr_get_thread_ids(pid_t pid, unsigned int *num_threads) > +{ > + char *path = g_build_filename("/proc", g_strdup_printf("%d/task", pid), NULL); Leaking the g_strdup_printf result > + > + DIR *dir = opendir(path); > + if (dir == NULL) { > + perror("opendir"); > + g_free(path); Mark 'path' as g_autofree so g_free() isn't needed in every exit-path > + return NULL; > + } > + > + pid_t *thread_ids = NULL; > + unsigned int thread_count = 0; > + > + struct dirent *ent; > + while ((ent = readdir(dir)) != NULL) { > + if (ent->d_name[0] == '.') { > + continue; > + } > + pid_t tid = atoi(ent->d_name); > + if (pid != tid) { > + thread_ids = g_renew(pid_t, thread_ids, (thread_count + 1)); > + thread_ids[thread_count] = tid; > + thread_count++; > + } > + } > + > + closedir(dir); > + > + *num_threads = thread_count; > + g_free(path); > + return thread_ids; > +} With regards, Daniel
On Thu, Jan 25, 2024 at 08:22:14AM +0100, Anthony Harivel wrote: > Starting with the "Sandy Bridge" generation, Intel CPUs provide a RAPL > interface (Running Average Power Limit) for advertising the accumulated > energy consumption of various power domains (e.g. CPU packages, DRAM, > etc.). > > The consumption is reported via MSRs (model specific registers) like > MSR_PKG_ENERGY_STATUS for the CPU package power domain. These MSRs are > 64 bits registers that represent the accumulated energy consumption in > micro Joules. They are updated by microcode every ~1ms. > > For now, KVM always returns 0 when the guest requests the value of > these MSRs. Use the KVM MSR filtering mechanism to allow QEMU handle > these MSRs dynamically in userspace. > > To limit the amount of system calls for every MSR call, create a new > thread in QEMU that updates the "virtual" MSR values asynchronously. > > Each vCPU has its own vMSR to reflect the independence of vCPUs. The > thread updates the vMSR values with the ratio of energy consumed of > the whole physical CPU package the vCPU thread runs on and the > thread's utime and stime values. > > All other non-vCPU threads are also taken into account. Their energy > consumption is evenly distributed among all vCPUs threads running on > the same physical CPU package. > > To overcome the problem that reading the RAPL MSR requires priviliged > access, a socket communication between QEMU and the qemu-vmsr-helper is > mandatory. You can specified the socket path in the parameter. > > This feature is activated with -accel kvm,rapl=true,path=/path/sock.sock > > Actual limitation: > - Works only on Intel host CPU because AMD CPUs are using different MSR > adresses. > > - Only the Package Power-Plane (MSR_PKG_ENERGY_STATUS) is reported at > the moment. > > Signed-off-by: Anthony Harivel <aharivel@redhat.com> > --- > accel/kvm/kvm-all.c | 27 +++ > docs/specs/index.rst | 1 + > docs/specs/rapl-msr.rst | 133 +++++++++++++ > include/sysemu/kvm_int.h | 17 ++ > target/i386/cpu.h | 8 + > target/i386/kvm/kvm.c | 348 ++++++++++++++++++++++++++++++++++ > target/i386/kvm/meson.build | 1 + > target/i386/kvm/vmsr_energy.c | 295 ++++++++++++++++++++++++++++ > target/i386/kvm/vmsr_energy.h | 87 +++++++++ > 9 files changed, 917 insertions(+) > create mode 100644 docs/specs/rapl-msr.rst > create mode 100644 target/i386/kvm/vmsr_energy.c > create mode 100644 target/i386/kvm/vmsr_energy.h > > diff --git a/include/sysemu/kvm_int.h b/include/sysemu/kvm_int.h > index 882e37e12c5b..ee2fe8817833 100644 > --- a/include/sysemu/kvm_int.h > +++ b/include/sysemu/kvm_int.h > @@ -14,6 +14,8 @@ > #include "qemu/accel.h" > #include "qemu/queue.h" > #include "sysemu/kvm.h" > +#include "hw/boards.h" > +#include "hw/i386/topology.h" > > typedef struct KVMSlot > { > @@ -48,6 +50,20 @@ typedef struct KVMMemoryListener { > > #define KVM_MSI_HASHTAB_SIZE 256 > > +struct KVMMsrEnergy { > + bool enable; > + char *socket_path; > + QemuThread msr_thr; > + unsigned int cpus; > + unsigned int sockets; > + X86CPUTopoInfo topo_info; > + const CPUArchIdList *cpu_list; > + uint64_t *msr_value; > + uint64_t msr_unit; > + uint64_t msr_limit; > + uint64_t msr_info; > +}; > + > enum KVMDirtyRingReaperState { > KVM_DIRTY_RING_REAPER_NONE = 0, > /* The reaper is sleeping */ > @@ -114,6 +130,7 @@ struct KVMState > bool kvm_dirty_ring_with_bitmap; > uint64_t kvm_eager_split_size; /* Eager Page Splitting chunk size */ > struct KVMDirtyRingReaper reaper; > + struct KVMMsrEnergy msr_energy; > NotifyVmexitOption notify_vmexit; > uint32_t notify_window; > uint32_t xen_version; > @@ -2509,6 +2558,265 @@ static void register_smram_listener(Notifier *n, void *unused) > &smram_address_space, 1, "kvm-smram"); > } > > +static void *kvm_msr_energy_thread(void *data) > +{ > + KVMState *s = data; > + struct KVMMsrEnergy *vmsr = &s->msr_energy; > + > + g_autofree package_energy_stat *pkg_stat = NULL; > + g_autofree thread_stat *thd_stat = NULL; > + g_autofree pid_t *thread_ids = NULL; > + g_autofree CPUState *cpu = NULL; > + unsigned int maxpkgs, maxcpus, maxticks; > + g_autofree unsigned int *vpkgs_energy_stat = NULL; > + unsigned int num_threads = 0; > + unsigned int tmp_num_threads = 0; > + pid_t pid; > + > + X86CPUTopoIDs topo_ids; > + > + > + rcu_register_thread(); > + > + /* Get QEMU PID*/ > + pid = getpid(); > + > + /* Nb of CPUS per packages */ > + maxcpus = vmsr_get_maxcpus(0); > + > + /* Nb of Physical Packages on the system */ > + maxpkgs = vmsr_get_max_physical_package(maxcpus); > + > + /* Those MSR values should not change as well */ > + vmsr->msr_unit = vmsr_read_msr(MSR_RAPL_POWER_UNIT, 0, pid, > + s->msr_energy.socket_path); > + vmsr->msr_limit = vmsr_read_msr(MSR_PKG_POWER_LIMIT, 0, pid, > + s->msr_energy.socket_path); > + vmsr->msr_info = vmsr_read_msr(MSR_PKG_POWER_INFO, 0, pid, > + s->msr_energy.socket_path); > + > + /* Allocate memory for each package energy status */ > + pkg_stat = (package_energy_stat *) > + g_new0(package_energy_stat, maxpkgs); > + > + /* Pre-allocate memory for thread stats */ > + thd_stat = g_new0(thread_stat, 1); > + > + /* Pre-allocate memory for holding Virtual Package Energy counter */ > + vpkgs_energy_stat = g_new0(unsigned int, vmsr->sockets); > + > + /* > + * Max numbers of ticks per package > + * time in second * number of ticks/second * Number of cores / package > + * ex: for 100 ticks/second/CPU, 12 CPUs per Package gives 1200 ticks max > + */ > + maxticks = (MSR_ENERGY_THREAD_SLEEP_US / 1000000) > + * sysconf(_SC_CLK_TCK) * maxcpus; > + > + while (true) { > + /* Get all qemu threads id */ > + thread_ids = vmsr_get_thread_ids(pid, &num_threads); This is allolcating thread_ids > + > + if (thread_ids == NULL) { > + goto clean; > + } > + > + if (tmp_num_threads < num_threads) { > + thd_stat = g_renew(thread_stat, thd_stat, num_threads); > + } > + > + tmp_num_threads = num_threads; > + > + /* Populate all the thread stats */ > + for (int i = 0; i < num_threads; i++) { > + thd_stat[i].utime = g_new0(unsigned long long, 2); > + thd_stat[i].stime = g_new0(unsigned long long, 2); > + thd_stat[i].thread_id = thread_ids[i]; > + vmsr_read_thread_stat(&thd_stat[i], pid, 0); > + thd_stat[i].numa_node_id = numa_node_of_cpu(thd_stat[i].cpu_id); > + } This loop is allocating memory for utime & stime.. > + > + /* Zero out the memory */ > + for (int i = 0; i < num_threads; i++) { > + memset(thd_stat[i].utime, 0, 2 * sizeof(unsigned long long)); > + memset(thd_stat[i].stime, 0, 2 * sizeof(unsigned long long)); > + } > + memset(thd_stat, 0, num_threads * sizeof(thread_stat)); ..and here we're just wiping the thd_stat pointer, so this is surey a memory leak of the stime & utime fields. > + memset(thread_ids, 0, sizeof(pid_t)); Nothing is free'ing thread_ids which was allocated earlier in the loop. > + } > + > +clean: > + rcu_unregister_thread(); > + return NULL; > +} > + > +static int kvm_msr_energy_thread_init(KVMState *s, MachineState *ms) > +{ > + struct KVMMsrEnergy *r = &s->msr_energy; > + > + vmsr_init_topo_info(&r->topo_info, ms); > + > + /* Retrieve the number of vCPU */ > + r->cpus = ms->smp.cpus; > + > + /* Retrieve the number of sockets */ > + r->sockets = ms->smp.sockets; > + > + /* Allocate register memory (MSR_PKG_STATUS) for each vCPU */ > + r->msr_value = g_new0(uint64_t, r->cpus); > + > + /* Retrieve the CPUArchIDlist */ > + r->x86_cpu_list = x86_possible_cpu_arch_ids(ms); > + > + qemu_thread_create(&r->msr_thr, "kvm-msr", > + kvm_msr_energy_thread, > + s, QEMU_THREAD_JOINABLE); > + > + return 0; > +} > + > int kvm_arch_get_default_type(MachineState *ms) > { > return 0; > @@ -2711,6 +3019,46 @@ int kvm_arch_init(MachineState *ms, KVMState *s) > strerror(-ret)); > exit(1); > } > + > + if (s->msr_energy.enable == true) { This looks to be where we need to check that both the host CPU vendor is intel, and the guest CPU vendor is intel, and that the host CPU has the RAPL feature we're using. > + > + r = kvm_filter_msr(s, MSR_RAPL_POWER_UNIT, > + kvm_rdmsr_rapl_power_unit, NULL); > + if (!r) { > + error_report("Could not install MSR_RAPL_POWER_UNIT \ > + handler: %s", > + strerror(-ret)); > + exit(1); > + } > + > + r = kvm_filter_msr(s, MSR_PKG_POWER_LIMIT, > + kvm_rdmsr_pkg_power_limit, NULL); > + if (!r) { > + error_report("Could not install MSR_PKG_POWER_LIMIT \ > + handler: %s", > + strerror(-ret)); > + exit(1); > + } > + > + r = kvm_filter_msr(s, MSR_PKG_POWER_INFO, > + kvm_rdmsr_pkg_power_info, NULL); > + if (!r) { > + error_report("Could not install MSR_PKG_POWER_INFO \ > + handler: %s", > + strerror(-ret)); > + exit(1); > + } > + r = kvm_filter_msr(s, MSR_PKG_ENERGY_STATUS, > + kvm_rdmsr_pkg_energy_status, NULL); > + if (!r) { > + error_report("Could not install MSR_PKG_ENERGY_STATUS \ > + handler: %s", > + strerror(-ret)); > + exit(1); > + } else { > + kvm_msr_energy_thread_init(s, ms); > + } > + } > } > > return 0; With regards, Daniel
On Thu, Jan 25, 2024 at 08:22:14AM +0100, Anthony Harivel wrote: > Starting with the "Sandy Bridge" generation, Intel CPUs provide a RAPL > interface (Running Average Power Limit) for advertising the accumulated > energy consumption of various power domains (e.g. CPU packages, DRAM, > etc.). > > The consumption is reported via MSRs (model specific registers) like > MSR_PKG_ENERGY_STATUS for the CPU package power domain. These MSRs are > 64 bits registers that represent the accumulated energy consumption in > micro Joules. They are updated by microcode every ~1ms. > > For now, KVM always returns 0 when the guest requests the value of > these MSRs. Use the KVM MSR filtering mechanism to allow QEMU handle > these MSRs dynamically in userspace. > > To limit the amount of system calls for every MSR call, create a new > thread in QEMU that updates the "virtual" MSR values asynchronously. > > Each vCPU has its own vMSR to reflect the independence of vCPUs. The > thread updates the vMSR values with the ratio of energy consumed of > the whole physical CPU package the vCPU thread runs on and the > thread's utime and stime values. > > All other non-vCPU threads are also taken into account. Their energy > consumption is evenly distributed among all vCPUs threads running on > the same physical CPU package. > > To overcome the problem that reading the RAPL MSR requires priviliged > access, a socket communication between QEMU and the qemu-vmsr-helper is > mandatory. You can specified the socket path in the parameter. > > This feature is activated with -accel kvm,rapl=true,path=/path/sock.sock > > Actual limitation: > - Works only on Intel host CPU because AMD CPUs are using different MSR > adresses. > > - Only the Package Power-Plane (MSR_PKG_ENERGY_STATUS) is reported at > the moment. > > Signed-off-by: Anthony Harivel <aharivel@redhat.com> > --- > accel/kvm/kvm-all.c | 27 +++ > docs/specs/index.rst | 1 + > docs/specs/rapl-msr.rst | 133 +++++++++++++ > include/sysemu/kvm_int.h | 17 ++ > target/i386/cpu.h | 8 + > target/i386/kvm/kvm.c | 348 ++++++++++++++++++++++++++++++++++ > target/i386/kvm/meson.build | 1 + > target/i386/kvm/vmsr_energy.c | 295 ++++++++++++++++++++++++++++ > target/i386/kvm/vmsr_energy.h | 87 +++++++++ > 9 files changed, 917 insertions(+) > create mode 100644 docs/specs/rapl-msr.rst > create mode 100644 target/i386/kvm/vmsr_energy.c > create mode 100644 target/i386/kvm/vmsr_energy.h > > @@ -2509,6 +2558,265 @@ static void register_smram_listener(Notifier *n, void *unused) > &smram_address_space, 1, "kvm-smram"); > } > > +static void *kvm_msr_energy_thread(void *data) > +{ > + KVMState *s = data; > + struct KVMMsrEnergy *vmsr = &s->msr_energy; > + > + g_autofree package_energy_stat *pkg_stat = NULL; > + g_autofree thread_stat *thd_stat = NULL; > + g_autofree pid_t *thread_ids = NULL; > + g_autofree CPUState *cpu = NULL; > + unsigned int maxpkgs, maxcpus, maxticks; > + g_autofree unsigned int *vpkgs_energy_stat = NULL; > + unsigned int num_threads = 0; > + unsigned int tmp_num_threads = 0; > + pid_t pid; > + > + X86CPUTopoIDs topo_ids; > + > + > + rcu_register_thread(); > + > + /* Get QEMU PID*/ > + pid = getpid(); > + > + /* Nb of CPUS per packages */ > + maxcpus = vmsr_get_maxcpus(0); > + > + /* Nb of Physical Packages on the system */ > + maxpkgs = vmsr_get_max_physical_package(maxcpus); This function can fail so this needs to be checked & reported. > + > + /* Those MSR values should not change as well */ > + vmsr->msr_unit = vmsr_read_msr(MSR_RAPL_POWER_UNIT, 0, pid, > + s->msr_energy.socket_path); > + vmsr->msr_limit = vmsr_read_msr(MSR_PKG_POWER_LIMIT, 0, pid, > + s->msr_energy.socket_path); > + vmsr->msr_info = vmsr_read_msr(MSR_PKG_POWER_INFO, 0, pid, > + s->msr_energy.socket_path); This function can fail for a variety of reasons, most especially if someone gave an incorrect socket path, or if the daemon is not running. This is not getting diagnosed, and even if we try to report it here, we're in a background thread at this point. I think we need to connect and report errors before even starting this thread, so that QEMU startup gets aborted upon configuration error. > + > + /* Allocate memory for each package energy status */ > + pkg_stat = (package_energy_stat *) > + g_new0(package_energy_stat, maxpkgs); > + > + /* Pre-allocate memory for thread stats */ > + thd_stat = g_new0(thread_stat, 1); > + > + /* Pre-allocate memory for holding Virtual Package Energy counter */ > + vpkgs_energy_stat = g_new0(unsigned int, vmsr->sockets); > + > + /* > + * Max numbers of ticks per package > + * time in second * number of ticks/second * Number of cores / package > + * ex: for 100 ticks/second/CPU, 12 CPUs per Package gives 1200 ticks max > + */ > + maxticks = (MSR_ENERGY_THREAD_SLEEP_US / 1000000) > + * sysconf(_SC_CLK_TCK) * maxcpus; > + > + while (true) { > + /* Get all qemu threads id */ > + thread_ids = vmsr_get_thread_ids(pid, &num_threads); > + > + if (thread_ids == NULL) { > + goto clean; > + } > + > + if (tmp_num_threads < num_threads) { > + thd_stat = g_renew(thread_stat, thd_stat, num_threads); > + } > + > + tmp_num_threads = num_threads; > + > + /* Populate all the thread stats */ > + for (int i = 0; i < num_threads; i++) { > + thd_stat[i].utime = g_new0(unsigned long long, 2); > + thd_stat[i].stime = g_new0(unsigned long long, 2); > + thd_stat[i].thread_id = thread_ids[i]; > + vmsr_read_thread_stat(&thd_stat[i], pid, 0); It is non-obvious that the 3rd parameter here is an index into the utime & stime array. This function would be saner to review if called as: vmsr_read_thread_stat(pid, thd_stat[i].thread_id, &thd_stat[i].utime[0], &thd_stat[i].stime[0], &thd_stat[i].cpu_id); so we see what are input parameters and what are output parameters. Also this method can fail, eg if the thread has exited already, so we need to take that into account and stop trying to get info for that thread in later code. eg by setting 'thread_id' to 0 and then skipping any thread_id == 0 later. > + thd_stat[i].numa_node_id = numa_node_of_cpu(thd_stat[i].cpu_id); > + } > + > + /* Retrieve all packages power plane energy counter */ > + for (int i = 0; i <= maxpkgs; i++) { > + for (int j = 0; j < num_threads; j++) { > + /* > + * Use the first thread we found that ran on the CPU > + * of the package to read the packages energy counter > + */ > + if (thd_stat[j].numa_node_id == i) { 'i' is a CPU ID value, while 'numa_node_id' is a NUMA node ID value. I don't think it is semantically valid to compare them for equality. I'm not sure the NUMA node is even relevant, since IIUC from the docs earlier, the power values are scoped per package, which would mean per CPU socket. > + pkg_stat[i].e_start = > + vmsr_read_msr(MSR_PKG_ENERGY_STATUS, i, pid, > + s->msr_energy.socket_path); > + break; > + } > + } > + } > + > + /* Sleep a short period while the other threads are working */ > + usleep(MSR_ENERGY_THREAD_SLEEP_US); > + > + /* > + * Retrieve all packages power plane energy counter > + * Calculate the delta of all packages > + */ > + for (int i = 0; i <= maxpkgs; i++) { > + for (int j = 0; j < num_threads; j++) { > + /* > + * Use the first thread we found that ran on the CPU > + * of the package to read the packages energy counter > + */ > + if (thd_stat[j].numa_node_id == i) { > + pkg_stat[i].e_end = > + vmsr_read_msr(MSR_PKG_ENERGY_STATUS, > + thd_stat[j].cpu_id, > + thd_stat[j].thread_id, > + s->msr_energy.socket_path); > + /* > + * Prevent the case we have migrate the VM > + * during the sleep period or any other cases > + * were energy counter might be lower after > + * the sleep. > + */ > + if (pkg_stat[i].e_end > pkg_stat[i].e_start) { > + pkg_stat[i].e_delta = > + pkg_stat[i].e_end - pkg_stat[i].e_start; > + } else { > + pkg_stat[i].e_delta = 0; > + } > + break; > + } > + } > + } > + > + /* Delta of ticks spend by each thread between the sample */ > + for (int i = 0; i < num_threads; i++) { > + if (vmsr_read_thread_stat(&thd_stat[i], pid, 1) != 0) { Here we fetch the thread 'cpu_id' again. What is expected behaviour if 'cpu_id' has changed since we read it earlier ? The delta ticks calculation will be OK is cpu_is is still within the same package as earlier, but if it moved to another package it is not valid. > + /* > + * We don't count the dead thread > + * i.e threads that existed before the sleep > + * and not anymore > + */ > + thd_stat[i].delta_ticks = 0; > + } else { > + vmsr_delta_ticks(thd_stat, i); > + } > + } > + > + /* > + * Identify the vCPU threads > + * Calculate the Number of vCPU per package > + */ > + CPU_FOREACH(cpu) { > + for (int i = 0; i < num_threads; i++) { > + if (cpu->thread_id == thd_stat[i].thread_id) { > + thd_stat[i].is_vcpu = true; > + thd_stat[i].vcpu_id = cpu->cpu_index; > + pkg_stat[thd_stat[i].numa_node_id].nb_vcpu++; A numa node isn't a package AFAICT. > + thd_stat[i].acpi_id = kvm_arch_vcpu_id(cpu); > + break; > + } > + } > + } > + > + /* Retrieve the virtual package number of each vCPU */ > + for (int i = 0; i < vmsr->x86_cpu_list->len; i++) { > + for (int j = 0; j < num_threads; j++) { > + if ((thd_stat[j].acpi_id == vmsr->x86_cpu_list->cpus[i].arch_id) > + && (thd_stat[j].is_vcpu == true)) { > + x86_topo_ids_from_apicid(thd_stat[j].acpi_id, > + &vmsr->topo_info, &topo_ids); > + thd_stat[j].vpkg = topo_ids.pkg_id; > + } > + } > + } > + > + /* Calculate the total energy of all non-vCPU thread */ > + for (int i = 0; i < num_threads; i++) { > + double temp; > + if ((thd_stat[i].is_vcpu != true) && > + (thd_stat[i].delta_ticks > 0)) { > + temp = vmsr_get_ratio(pkg_stat, thd_stat, maxticks, i); > + pkg_stat[thd_stat[i].numa_node_id].e_ratio > + += (uint64_t)lround(temp); > + } > + } > + > + /* Calculate the ratio per non-vCPU thread of each package */ > + for (int i = 0; i <= maxpkgs; i++) { > + if (pkg_stat[i].nb_vcpu > 0) { > + pkg_stat[i].e_ratio = pkg_stat[i].e_ratio / pkg_stat[i].nb_vcpu; > + } > + } > + > + /* > + * Calculate the energy for each Package: > + * Energy Package = sum of each vCPU energy that belongs to the package > + */ > + for (int i = 0; i < num_threads; i++) { > + double temp; > + > + if ((thd_stat[i].is_vcpu == true) && \ > + (thd_stat[i].delta_ticks > 0)) { > + temp = vmsr_get_ratio(pkg_stat, thd_stat, maxticks, i); > + > + vpkgs_energy_stat[thd_stat[i].vpkg] += (uint64_t)lround(temp); > + vpkgs_energy_stat[thd_stat[i].vpkg] += > + pkg_stat[thd_stat[i].numa_node_id].e_ratio; > + } > + } > + > + /* > + * Finally populate the vmsr register of each vCPU with the total > + * package value to emulate the real hardware where each CPU return the > + * value of the package it belongs. > + */ > + for (int i = 0; i < num_threads; i++) { > + if ((thd_stat[i].is_vcpu == true) && \ > + (thd_stat[i].delta_ticks > 0)) { > + vmsr->msr_value[thd_stat[i].vcpu_id] = \ > + vpkgs_energy_stat[thd_stat[i].vpkg]; > + } > + } > + > + /* Zero out the memory */ > + for (int i = 0; i < num_threads; i++) { > + memset(thd_stat[i].utime, 0, 2 * sizeof(unsigned long long)); > + memset(thd_stat[i].stime, 0, 2 * sizeof(unsigned long long)); > + } > + memset(thd_stat, 0, num_threads * sizeof(thread_stat)); > + memset(thread_ids, 0, sizeof(pid_t)); > + } > + > +clean: > + rcu_unregister_thread(); > + return NULL; > +} With regards, Daniel
Daniel P. Berrangé, Jan 29, 2024 at 20:29: > On Thu, Jan 25, 2024 at 08:22:14AM +0100, Anthony Harivel wrote: > > diff --git a/docs/specs/rapl-msr.rst b/docs/specs/rapl-msr.rst > > new file mode 100644 > > index 000000000000..04d27c198fc0 > > --- /dev/null > > +++ b/docs/specs/rapl-msr.rst > > @@ -0,0 +1,133 @@ > > +================ > > +RAPL MSR support > > +================ > > > + > > +Current Limitations > > +------------------- > > + > > +- Works only on Intel host CPUs because AMD CPUs are using different MSR > > + addresses. > > The privileged helper program is validating an allow list of MSRs. > > If those MSRs are only correct on Intel hosts, then the validation > is incomplete, and it could be allowing unprivileged processes on > AMD hosts to access forbidden MSRS whose address happen to clash > with the Intel RAPL MSRs. > > IOW, the privileged helper needs to call cpuid() and validate that > the current host vendor is Intel. > > I suspect we also need a feature check of some kind to validate > that the intel processor supports this features, since old ones > definitely didn't, and we shouldn't assume all future ones will > either. > To validate that the processor supports the RAPL feature I propose to check this on the Host: $ cat /sys/class/powercap/intel-rapl/enabled 1 The only down side is that INTEL RAPL drivers needs to be mounted then. We don't need it because we directly read the MSRs. Regards, Anthony
On Tue, Feb 20, 2024 at 03:00:56PM +0100, Anthony Harivel wrote: > Daniel P. Berrangé, Jan 29, 2024 at 20:29: > > On Thu, Jan 25, 2024 at 08:22:14AM +0100, Anthony Harivel wrote: > > > diff --git a/docs/specs/rapl-msr.rst b/docs/specs/rapl-msr.rst > > > new file mode 100644 > > > index 000000000000..04d27c198fc0 > > > --- /dev/null > > > +++ b/docs/specs/rapl-msr.rst > > > @@ -0,0 +1,133 @@ > > > +================ > > > +RAPL MSR support > > > +================ > > > > > + > > > +Current Limitations > > > +------------------- > > > + > > > +- Works only on Intel host CPUs because AMD CPUs are using different MSR > > > + addresses. > > > > The privileged helper program is validating an allow list of MSRs. > > > > If those MSRs are only correct on Intel hosts, then the validation > > is incomplete, and it could be allowing unprivileged processes on > > AMD hosts to access forbidden MSRS whose address happen to clash > > with the Intel RAPL MSRs. > > > > IOW, the privileged helper needs to call cpuid() and validate that > > the current host vendor is Intel. > > > > I suspect we also need a feature check of some kind to validate > > that the intel processor supports this features, since old ones > > definitely didn't, and we shouldn't assume all future ones will > > either. > > > > To validate that the processor supports the RAPL feature I propose > to check this on the Host: > > $ cat /sys/class/powercap/intel-rapl/enabled > 1 > > > The only down side is that INTEL RAPL drivers needs to be > mounted then. We don't need it because we directly read the MSRs. Awkward. I've looked around in the kernel drivers and didn't uncover any better option than this. With regards, Daniel
Hi Daniel, > > + if (s->msr_energy.enable == true) { > > This looks to be where we need to check that both the host CPU > vendor is intel, and the guest CPU vendor is intel, and that > the host CPU has the RAPL feature we're using. > > With regards, > Daniel > -- > |: https://berrange.com -o- https://www.flickr.com/photos/dberrange :| > |: https://libvirt.org -o- https://fstop138.berrange.com :| > |: https://entangle-photo.org -o- https://www.instagram.com/dberrange :| Checking for the host cpu and RAPL enable is fine and done. But checking for guest CPU is confusing me. The RAPL feature is enable only with KVM enable. This means "-cpu" can only be "host" or its derivative that essentially copy the host CPU definition, no? That means if we are already checking the host cpu we don't need to do anything for the guest, do we ? Regards, Anthony
On Mon, Mar 04, 2024 at 03:41:02PM +0100, Anthony Harivel wrote: > > Hi Daniel, > > > > + if (s->msr_energy.enable == true) { > > > > This looks to be where we need to check that both the host CPU > > vendor is intel, and the guest CPU vendor is intel, and that > > the host CPU has the RAPL feature we're using. > > Checking for the host cpu and RAPL enable is fine and done. > > But checking for guest CPU is confusing me. > The RAPL feature is enable only with KVM enable. > This means "-cpu" can only be "host" or its derivative that essentially > copy the host CPU definition, no? KVM can use any named CPU. > That means if we are already checking the host cpu we don't need to do > anything for the guest, do we ? When I first wrote this I though it would be as simple as checknig a CPUID feature flag. That appears to not be the case, however, as Linux is just checking for various CPU models directly. With that in mind perhaps we should just check of the guest CPU model vendor == CPUID_VENDOR_INTEL and leave it at that. eg, create an error if running an AMD CPU such as $QEMU -cpu EPYC With regards, Daniel
Daniel P. Berrangé, Mar 04, 2024 at 15:48: > On Mon, Mar 04, 2024 at 03:41:02PM +0100, Anthony Harivel wrote: > > > > Hi Daniel, > > > > > > + if (s->msr_energy.enable == true) { > > > > > > This looks to be where we need to check that both the host CPU > > > vendor is intel, and the guest CPU vendor is intel, and that > > > the host CPU has the RAPL feature we're using. > > > > Checking for the host cpu and RAPL enable is fine and done. > > > > But checking for guest CPU is confusing me. > > The RAPL feature is enable only with KVM enable. > > This means "-cpu" can only be "host" or its derivative that essentially > > copy the host CPU definition, no? > > KVM can use any named CPU. > > > That means if we are already checking the host cpu we don't need to do > > anything for the guest, do we ? > > When I first wrote this I though it would be as simple as checknig a > CPUID feature flag. That appears to not be the case, however, as Linux > is just checking for various CPU models directly. With that in mind > perhaps we should just check of the guest CPU model vendor > == CPUID_VENDOR_INTEL and leave it at that. > > eg, create an error if running an AMD CPU such as $QEMU -cpu EPYC The idea looks good to me. Now the hiccups of this solution is that I cannot find a way to reach CPUArchState at this level of code (i.e kvm_arch_init() ) with only the MachineState or the KVMState. I can only reach the topology with x86_possible_cpu_arch_ids(). CPUArchState struct is holding the cpuid_vendor variables where we can use IS_INTEL_CPU() for checking. Maybe you know the trick that I miss ? Regards, Anthony > > With regards, > Daniel > -- > |: https://berrange.com -o- https://www.flickr.com/photos/dberrange :| > |: https://libvirt.org -o- https://fstop138.berrange.com :| > |: https://entangle-photo.org -o- https://www.instagram.com/dberrange :|
On Tue, Mar 05, 2024 at 02:25:09PM +0100, Anthony Harivel wrote: > Daniel P. Berrangé, Mar 04, 2024 at 15:48: > > On Mon, Mar 04, 2024 at 03:41:02PM +0100, Anthony Harivel wrote: > > > > > > Hi Daniel, > > > > > > > > + if (s->msr_energy.enable == true) { > > > > > > > > This looks to be where we need to check that both the host CPU > > > > vendor is intel, and the guest CPU vendor is intel, and that > > > > the host CPU has the RAPL feature we're using. > > > > > > Checking for the host cpu and RAPL enable is fine and done. > > > > > > But checking for guest CPU is confusing me. > > > The RAPL feature is enable only with KVM enable. > > > This means "-cpu" can only be "host" or its derivative that essentially > > > copy the host CPU definition, no? > > > > KVM can use any named CPU. > > > > > That means if we are already checking the host cpu we don't need to do > > > anything for the guest, do we ? > > > > When I first wrote this I though it would be as simple as checknig a > > CPUID feature flag. That appears to not be the case, however, as Linux > > is just checking for various CPU models directly. With that in mind > > perhaps we should just check of the guest CPU model vendor > > == CPUID_VENDOR_INTEL and leave it at that. > > > > eg, create an error if running an AMD CPU such as $QEMU -cpu EPYC > > The idea looks good to me. Now the hiccups of this solution is that > I cannot find a way to reach CPUArchState at this level of code (i.e > kvm_arch_init() ) with only the MachineState or the KVMState. > I can only reach the topology with x86_possible_cpu_arch_ids(). > > CPUArchState struct is holding the cpuid_vendor variables where we can > use IS_INTEL_CPU() for checking. > > Maybe you know the trick that I miss ? I think perhaps you can do a check in kvm_cpu_realizefn() from target/i386/kvm/kvm-cpu.c, as you have CPUX86State state which is what IS_INTEL_CPU wants. With regards, Daniel
Hi Daniel, > > + > > + /* Retrieve all packages power plane energy counter */ > > + for (int i = 0; i <= maxpkgs; i++) { > > + for (int j = 0; j < num_threads; j++) { > > + /* > > + * Use the first thread we found that ran on the CPU > > + * of the package to read the packages energy counter > > + */ > > This says we're using a thread ID > > > + if (thd_stat[j].numa_node_id == i) { > > + pkg_stat[i].e_start = > > + vmsr_read_msr(MSR_PKG_ENERGY_STATUS, i, pid, > > but here we're using a pid ID, which is the thread ID of the initial > thread. > > > + s->msr_energy.socket_path); > > + break; > > + } > > + } > > + } > > This API design for vmsr_read_msr() is incredibly inefficient. > We're making (maxpkgs * num_threads) calls to vmsr_read_msr(), > and every one of those is opening and closing the socket. > > Why isn't QEMU opening the socket once and then sending all > the requests over the same socket ? > The usage of pid here is a mistake, thanks for pointing this out. However, I'm more sceptical about the fact that the loop is inefficient. The confusion could definitely be because of the poor variable naming, and I apologize about that. Let me try to explain what it's supposed to do: Imagine we are running on machine that has i packages. QEMU has j threads running on whichever packages. We need to get the current packages energy of each packages that are used by the QEMU threads. (could be all i packages, only 1, 2.. we don't know what we need yet) So it loops first on the packages "0", and look if any thread has run on this packages. If no, test the next thread. if yes, we need the value, we call the vmsr_read_msr() then break and now loop for the next package, i.e package "1". And this until all packages has been tested. So in the end, we 'only' have maximum "maxpkgs" calls of vmsr_read_msr(). Hope that's ok and that clear up the confusion! Regards, Anthony
Hi Daniel, Paolo, Here my last questions before wrapping up and send v4, or maybe call off my attempt to add RAPL interface in QEMU. Daniel P. Berrangé, Jan 30, 2024 at 10:39: > > + rcu_register_thread(); > > + > > + /* Get QEMU PID*/ > > + pid = getpid(); > > + > > + /* Nb of CPUS per packages */ > > + maxcpus = vmsr_get_maxcpus(0); > > + > > + /* Nb of Physical Packages on the system */ > > + maxpkgs = vmsr_get_max_physical_package(maxcpus); > > This function can fail so this needs to be checked & reported. > > > + > > + /* Those MSR values should not change as well */ > > + vmsr->msr_unit = vmsr_read_msr(MSR_RAPL_POWER_UNIT, 0, pid, > > + s->msr_energy.socket_path); > > + vmsr->msr_limit = vmsr_read_msr(MSR_PKG_POWER_LIMIT, 0, pid, > > + s->msr_energy.socket_path); > > + vmsr->msr_info = vmsr_read_msr(MSR_PKG_POWER_INFO, 0, pid, > > + s->msr_energy.socket_path); > > This function can fail for a variety of reasons, most especially if someone > gave an incorrect socket path, or if the daemon is not running. This is not > getting diagnosed, and even if we try to report it here, we're in a background > thread at this point. > > I think we need to connect and report errors before even starting this > thread, so that QEMU startup gets aborted upon configuration error. > Fair enough. Would it be ok to do the sanity check before rcu_register_thread() and "return NULL;" in case of error or would you prefer me to check all of this before even calling the qemu_thread_create() ? > > + /* Populate all the thread stats */ > > + for (int i = 0; i < num_threads; i++) { > > + thd_stat[i].utime = g_new0(unsigned long long, 2); > > + thd_stat[i].stime = g_new0(unsigned long long, 2); > > + thd_stat[i].thread_id = thread_ids[i]; > > + vmsr_read_thread_stat(&thd_stat[i], pid, 0); > > It is non-obvious that the 3rd parameter here is an index into > the utime & stime array. This function would be saner to review > if called as: > > vmsr_read_thread_stat(pid, > thd_stat[i].thread_id, > &thd_stat[i].utime[0], > &thd_stat[i].stime[0], > &thd_stat[i].cpu_id); > > so we see what are input parameters and what are output parameters. > > Also this method can fail, eg if the thread has exited already, > so we need to take that into account and stop trying to get info > for that thread in later code. eg by setting 'thread_id' to 0 > and then skipping any thread_id == 0 later. > > Good point. I'll rework the function and return "thread_id" to 0 in case of failure in order to test it later on. > > + thd_stat[i].numa_node_id = numa_node_of_cpu(thd_stat[i].cpu_id); > > + } > > + > > + /* Retrieve all packages power plane energy counter */ > > + for (int i = 0; i <= maxpkgs; i++) { > > + for (int j = 0; j < num_threads; j++) { > > + /* > > + * Use the first thread we found that ran on the CPU > > + * of the package to read the packages energy counter > > + */ > > + if (thd_stat[j].numa_node_id == i) { > > 'i' is a CPU ID value, while 'numa_node_id' is a NUMA node ID value. > I don't think it is semantically valid to compare them for equality. > > I'm not sure the NUMA node is even relevant, since IIUC from the docs > earlier, the power values are scoped per package, which would mean per > CPU socket. > 'i' here is the package number on the host. I'm using functions of libnuma to populate the maxpkgs of the host. I tested this on different Intel CPU with multiple packages and this has always returned the good number of packages. A false positive ? So here I'm checking if the thread has run on the package number 'i'. I populate 'numa_node_id' with numa_node_of_cpu(). I did not wanted to reinvent the wheel and the only lib that was talking about "node" was libnuma. Maybe I'm wrong assuming that a "node" (defined as an area where all memory has the same speed as seen from a particular CPU) could lead me to the packages number ? And this is what I see you wrote below: "A numa node isn't a package AFAICT." Regards, Anthony
On Tue, Mar 12, 2024 at 12:21:14PM +0100, Anthony Harivel wrote: > Daniel P. Berrangé, Jan 30, 2024 at 10:39: > > > + rcu_register_thread(); > > > + > > > + /* Get QEMU PID*/ > > > + pid = getpid(); > > > + > > > + /* Nb of CPUS per packages */ > > > + maxcpus = vmsr_get_maxcpus(0); > > > + > > > + /* Nb of Physical Packages on the system */ > > > + maxpkgs = vmsr_get_max_physical_package(maxcpus); > > > > This function can fail so this needs to be checked & reported. > > > > > + > > > + /* Those MSR values should not change as well */ > > > + vmsr->msr_unit = vmsr_read_msr(MSR_RAPL_POWER_UNIT, 0, pid, > > > + s->msr_energy.socket_path); > > > + vmsr->msr_limit = vmsr_read_msr(MSR_PKG_POWER_LIMIT, 0, pid, > > > + s->msr_energy.socket_path); > > > + vmsr->msr_info = vmsr_read_msr(MSR_PKG_POWER_INFO, 0, pid, > > > + s->msr_energy.socket_path); > > > > This function can fail for a variety of reasons, most especially if someone > > gave an incorrect socket path, or if the daemon is not running. This is not > > getting diagnosed, and even if we try to report it here, we're in a background > > thread at this point. > > > > I think we need to connect and report errors before even starting this > > thread, so that QEMU startup gets aborted upon configuration error. > > > > Fair enough. Would it be ok to do the sanity check before > rcu_register_thread() and "return NULL;" in case of error or would you > prefer me to check all of this before even calling the > qemu_thread_create() ? I think it is best to initialize in kvm_msr_energy_thread_init(), prior to thread creation, as that avoids some complexity cleaning up and reporting errors. > > > + thd_stat[i].numa_node_id = numa_node_of_cpu(thd_stat[i].cpu_id); > > > + } > > > + > > > + /* Retrieve all packages power plane energy counter */ > > > + for (int i = 0; i <= maxpkgs; i++) { > > > + for (int j = 0; j < num_threads; j++) { > > > + /* > > > + * Use the first thread we found that ran on the CPU > > > + * of the package to read the packages energy counter > > > + */ > > > + if (thd_stat[j].numa_node_id == i) { > > > > 'i' is a CPU ID value, while 'numa_node_id' is a NUMA node ID value. > > I don't think it is semantically valid to compare them for equality. > > > > I'm not sure the NUMA node is even relevant, since IIUC from the docs > > earlier, the power values are scoped per package, which would mean per > > CPU socket. > > > > 'i' here is the package number on the host. Opps, yes, 'maxpkgs' is iterating over CPU /socket/ ID numbers The point still stands though. NUMA node ID numbers are not guaranteed to be the same as socket ID numbers. Very often then will be the same (which makes it annoying to test as it is easy to not realize the difference), but we can't rely on that. > I'm using functions of libnuma to populate the maxpkgs of the host. > I tested this on different Intel CPU with multiple packages and this > has always returned the good number of packages. A false positive ? maxpkgs comes from vmsr_get_max_physical_package() which you're reading from sysfs, rather than libnuma. > So here I'm checking if the thread has run on the package number 'i'. > I populate 'numa_node_id' with numa_node_of_cpu(). > > I did not wanted to reinvent the wheel and the only lib that was talking > about "node" was libnuma. I'm not actually convinced we need to use libnuma at all. IIUC, you're just trying to track all CPUs within the same physical socket (package). I don't think we need to care about NUMA nodes to do that tracking. > Maybe I'm wrong assuming that a "node" (defined as an area where all > memory has the same speed as seen from a particular CPU) could lead me > to the packages number ? Historically you could have multiple sockets in the same NUMA node ie a m:1 mapping. These days with AMD sockets, you can have 1 socket compromising many NUMA nodes, as individual dies within a socket are each their own NUMA node. So a 1:m mapping On Intel I think it is still typical to have 1 socket per numa node, but again I don't think we can rely on that 1:1 mapping. Fortunately I don't think it matters, since it looks like you don't really need to track NUMA nodes, only sockets (phnysical package IDs) With regards, Daniel
Hi Daniel, Daniel P. Berrangé, Mar 12, 2024 at 16:49: > The point still stands though. NUMA node ID numbers are not > guaranteed to be the same as socket ID numbers. Very often > then will be the same (which makes it annoying to test as it > is easy to not realize the difference), but we can't rely on > that. > > > I'm using functions of libnuma to populate the maxpkgs of the host. > > I tested this on different Intel CPU with multiple packages and this > > has always returned the good number of packages. A false positive ? > > maxpkgs comes from vmsr_get_max_physical_package() which you're > reading from sysfs, rather than libnuma. > > > So here I'm checking if the thread has run on the package number 'i'. > > I populate 'numa_node_id' with numa_node_of_cpu(). > > > > I did not wanted to reinvent the wheel and the only lib that was talking > > about "node" was libnuma. > > I'm not actually convinced we need to use libnuma at all. IIUC, you're > just trying to track all CPUs within the same physical socket (package). > I don't think we need to care about NUMA nodes to do that tracking. > Alright, having a deeper look I'm actually using NUMA for 2 info: - How many cpu per Package: this helps me calculate the ratio. - To whom package the cpu belongs: to calculate the ratio with the right package energy counter. Without libnuma, I'm bit confused on how to handle this. Should I parse /sys/bus/node/devices/node* to know how many packages ? Should I parse /sys/bus/node/devices/node0/cpu0/topology/core_cpus_list to handle which cpu belongs to which package ? Would that be too cumbusome for the user to enter the detail about how many packages and how many cpu per pakages ? i.e: -kvm,rapl=true,maxpkgs=2,cpupkgs=8,rapl-helper-socket=/path/sock.sock > > Maybe I'm wrong assuming that a "node" (defined as an area where all > > memory has the same speed as seen from a particular CPU) could lead me > > to the packages number ? > > Historically you could have multiple sockets in the same NUMA node > ie a m:1 mapping. > > These days with AMD sockets, you can have 1 socket compromising > many NUMA nodes, as individual dies within a socket are each their > own NUMA node. So a 1:m mapping > > On Intel I think it is still typical to have 1 socket per numa > node, but again I don't think we can rely on that 1:1 mapping. > > Fortunately I don't think it matters, since it looks like you > don't really need to track NUMA nodes, only sockets (phnysical > package IDs) > Very informative, thanks ! > With regards, > Daniel > -- > |: https://berrange.com -o- https://www.flickr.com/photos/dberrange :| > |: https://libvirt.org -o- https://fstop138.berrange.com :| > |: https://entangle-photo.org -o- https://www.instagram.com/dberrange :| Regards, Anthony
On Wed, Mar 13, 2024 at 11:48:19AM +0100, Anthony Harivel wrote: > Hi Daniel, > > Daniel P. Berrangé, Mar 12, 2024 at 16:49: > > > The point still stands though. NUMA node ID numbers are not > > guaranteed to be the same as socket ID numbers. Very often > > then will be the same (which makes it annoying to test as it > > is easy to not realize the difference), but we can't rely on > > that. > > > > > I'm using functions of libnuma to populate the maxpkgs of the host. > > > I tested this on different Intel CPU with multiple packages and this > > > has always returned the good number of packages. A false positive ? > > > > maxpkgs comes from vmsr_get_max_physical_package() which you're > > reading from sysfs, rather than libnuma. > > > > > So here I'm checking if the thread has run on the package number 'i'. > > > I populate 'numa_node_id' with numa_node_of_cpu(). > > > > > > I did not wanted to reinvent the wheel and the only lib that was talking > > > about "node" was libnuma. > > > > I'm not actually convinced we need to use libnuma at all. IIUC, you're > > just trying to track all CPUs within the same physical socket (package). > > I don't think we need to care about NUMA nodes to do that tracking. > > > > Alright, having a deeper look I'm actually using NUMA for 2 info: > > - How many cpu per Package: this helps me calculate the ratio. > > - To whom package the cpu belongs: to calculate the ratio with the right > package energy counter. > > Without libnuma, I'm bit confused on how to handle this. > > Should I parse /sys/bus/node/devices/node* to know how many packages ? > Should I parse /sys/bus/node/devices/node0/cpu0/topology/core_cpus_list > to handle which cpu belongs to which package ? You don't need to access it via the /node/ hierarchy The canonical path for CPUs would be /sys/devices/system/cpu/cpuNNN/topology The core_cpus_list file is giving you hyper-thread siblings within a core, which I don't think is what you want. If you're after discrete physical packages, then 'package_cpus_list' gives you all CPUs within a physical socket (package) I believe. > Would that be too cumbusome for the user to enter the detail about how > many packages and how many cpu per pakages ? > > i.e: > -kvm,rapl=true,maxpkgs=2,cpupkgs=8,rapl-helper-socket=/path/sock.sock That won't cope with asymmetrical CPU configurations, so I think it is preferrable to read the info from sysfs. With regards, Daniel
Hi Daniel, > You don't need to access it via the /node/ hierarchy > > The canonical path for CPUs would be > > /sys/devices/system/cpu/cpuNNN/topology > > The core_cpus_list file is giving you hyper-thread siblings within > a core, which I don't think is what you want. > > If you're after discrete physical packages, then 'package_cpus_list' > gives you all CPUs within a physical socket (package) I believe. > Yes, this could work. However, on laptop, I've got: cat package_cpus_list 0-11 Where on server: package_cpus_list 0,2,4,6,8,10,12,14,16,18,20,22,24,26,28,30,32,34,36,38,40,42,44,46 I asked my teammate: always the same results. This is I guess due to either a difference in the kernel version or how the kernel is handling the case where there is only one package, versus the case with multiple packages. Anyway, writing a C function to handle both cases might not be easy. The other solution would be to loop through /sys/devices/system/cpu/cpuNNN/ and update a table of integer of a fixed size (*) where I increment table[0] for package_0 when I encounter a CPU that belongs to package_0 and so on. Reading cpuNNN/topology/physical_package_id is giving to whom package the cpu belongs to. This is a bit tedious but a safer solution, I think. (*): Maybe dynamic allocation is better ? Thanks again for your guidance. Regards, Anthony
On Thu, Mar 14, 2024 at 09:26:53AM +0100, Anthony Harivel wrote: > > Hi Daniel, > > > > You don't need to access it via the /node/ hierarchy > > > > The canonical path for CPUs would be > > > > /sys/devices/system/cpu/cpuNNN/topology > > > > The core_cpus_list file is giving you hyper-thread siblings within > > a core, which I don't think is what you want. > > > > If you're after discrete physical packages, then 'package_cpus_list' > > gives you all CPUs within a physical socket (package) I believe. > > > > Yes, this could work. > However, on laptop, I've got: > cat package_cpus_list > 0-11 > > Where on server: > package_cpus_list > 0,2,4,6,8,10,12,14,16,18,20,22,24,26,28,30,32,34,36,38,40,42,44,46 > > I asked my teammate: always the same results. This is I guess due to > either a difference in the kernel version or how the kernel is handling > the case where there is only one package, versus the case with multiple > packages. Both are the same data format - it is a list of ranges. In the laptop example, there's only a single range to parse. In the server example there are many ranges but since each range is only 1 cpu, it has collapsed the ranges to the single CPU id. > Anyway, writing a C function to handle both cases might not be easy. Approximately * Read the whole file with g_get_file_contents * Use g_strsplit(data, ",", 0) to get a list of ranges * Iterate over the return list of ranges and g_strsplit(range, "-", 2); - The returned list should be either a single element (if there was no '-'), or a pair of elements (if there was a N-M) With regards, Daniel
diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c index 49e755ec4ad2..d63a6af91291 100644 --- a/accel/kvm/kvm-all.c +++ b/accel/kvm/kvm-all.c @@ -3603,6 +3603,21 @@ static void kvm_set_device(Object *obj, s->device = g_strdup(value); } +static void kvm_set_kvm_rapl(Object *obj, bool value, Error **errp) +{ + KVMState *s = KVM_STATE(obj); + s->msr_energy.enable = value; +} + +static void kvm_set_kvm_rapl_socket_path(Object *obj, + const char *str, + Error **errp) +{ + KVMState *s = KVM_STATE(obj); + g_free(s->msr_energy.socket_path); + s->msr_energy.socket_path = g_strdup(str); +} + static void kvm_accel_instance_init(Object *obj) { KVMState *s = KVM_STATE(obj); @@ -3622,6 +3637,7 @@ static void kvm_accel_instance_init(Object *obj) s->xen_gnttab_max_frames = 64; s->xen_evtchn_max_pirq = 256; s->device = NULL; + s->msr_energy.enable = false; } /** @@ -3666,6 +3682,17 @@ static void kvm_accel_class_init(ObjectClass *oc, void *data) object_class_property_set_description(oc, "device", "Path to the device node to use (default: /dev/kvm)"); + object_class_property_add_bool(oc, "rapl", + NULL, + kvm_set_kvm_rapl); + object_class_property_set_description(oc, "rapl", + "Allow energy related MSRs for RAPL interface in Guest"); + + object_class_property_add_str(oc, "rapl-helper-socket", NULL, + kvm_set_kvm_rapl_socket_path); + object_class_property_set_description(oc, "rapl-helper-socket", + "Socket Path for comminucating with the Virtual MSR helper daemon"); + kvm_arch_accel_class_init(oc); } diff --git a/docs/specs/index.rst b/docs/specs/index.rst index b3f482b0aa58..b426ebb7713c 100644 --- a/docs/specs/index.rst +++ b/docs/specs/index.rst @@ -32,3 +32,4 @@ guest hardware that is specific to QEMU. virt-ctlr vmcoreinfo vmgenid + rapl-msr diff --git a/docs/specs/rapl-msr.rst b/docs/specs/rapl-msr.rst new file mode 100644 index 000000000000..04d27c198fc0 --- /dev/null +++ b/docs/specs/rapl-msr.rst @@ -0,0 +1,133 @@ +================ +RAPL MSR support +================ + +The RAPL interface (Running Average Power Limit) is advertising the accumulated +energy consumption of various power domains (e.g. CPU packages, DRAM, etc.). + +The consumption is reported via MSRs (model specific registers) like +MSR_PKG_ENERGY_STATUS for the CPU package power domain. These MSRs are 64 bits +registers that represent the accumulated energy consumption in micro Joules. + +Thanks to the MSR Filtering patch [#a]_ not all MSRs are handled by KVM. Some +of them can now be handled by the userspace (QEMU). It uses a mechanism called +"MSR filtering" where a list of MSRs is given at init time of a VM to KVM so +that a callback is put in place. The design of this patch uses only this +mechanism for handling the MSRs between guest/host. + +At the moment the following MSRs are involved: + +.. code:: C + + #define MSR_RAPL_POWER_UNIT 0x00000606 + #define MSR_PKG_POWER_LIMIT 0x00000610 + #define MSR_PKG_ENERGY_STATUS 0x00000611 + #define MSR_PKG_POWER_INFO 0x00000614 + +The ``*_POWER_UNIT``, ``*_POWER_LIMIT``, ``*_POWER INFO`` are part of the RAPL +spec and specify the power limit of the package, provide range of parameter(min +power, max power,..) and also the information of the multiplier for the energy +counter to calculate the power. Those MSRs are populated once at the beginning +by reading the host CPU MSRs and are given back to the guest 1:1 when +requested. + +The MSR_PKG_ENERGY_STATUS is a counter; it represents the total amount of +energy consumed since the last time the register was cleared. If you multiply +it with the UNIT provided above you'll get the power in micro-joules. This +counter is always increasing and it increases more or less faster depending on +the consumption of the package. This counter is supposed to overflow at some +point. + +Each core belonging to the same Package reading the MSR_PKG_ENERGY_STATUS (i.e +"rdmsr 0x611") will retrieve the same value. The value represents the energy +for the whole package. Whatever Core reading it will get the same value and a +core that belongs to PKG-0 will not be able to get the value of PKG-1 and +vice-versa. + +High level implementation +------------------------- + +In order to update the value of the virtual MSR, a QEMU thread is created. +The thread is basically just an infinity loop that does: + +1. Snapshot of the time metrics of all QEMU threads (Time spent scheduled in + Userspace and System) + +2. Snapshot of the actual MSR_PKG_ENERGY_STATUS counter of all packages where + the QEMU threads are running on. + +3. Sleep for 1 second - During this pause the vcpu and other non-vcpu threads + will do what they have to do and so the energy counter will increase. + +4. Repeat 2. and 3. and calculate the delta of every metrics representing the + time spent scheduled for each QEMU thread *and* the energy spent by the + packages during the pause. + +5. Filter the vcpu threads and the non-vcpu threads. + +6. Retrieve the topology of the Virtual Machine. This helps identify which + vCPU is running on which virtual package. + +7. The total energy spent by the non-vcpu threads is divided by the number + of vcpu threads so that each vcpu thread will get an equal part of the + energy spent by the QEMU workers. + +8. Calculate the ratio of energy spent per vcpu threads. + +9. Calculate the energy for each virtual package. + +10. The virtual MSRs are updated for each virtual package. Each vCPU that + belongs to the same package will return the same value when accessing the + the MSR. + +11. Loop back to 1. + +Ratio calculation +----------------- + +In Linux, a process has an execution time associated with it. The scheduler is +dividing the time in clock ticks. The number of clock ticks per second can be +found by the sysconf system call. A typical value of clock ticks per second is +100. So a core can run a process at the maximum of 100 ticks per second. If a +package has 4 cores, 400 ticks maximum can be scheduled on all the cores +of the package for a period of 1 second. + +The /proc/[pid]/stat [#b]_ is a sysfs file that can give the executed time of a +process with the [pid] as the process ID. It gives the amount of ticks the +process has been scheduled in userspace (utime) and kernel space (stime). + +By reading those metrics for a thread, one can calculate the ratio of time the +package has spent executing the thread. + +Example: + +A 4 cores package can schedule a maximum of 400 ticks per second with 100 ticks +per second per core. If a thread was scheduled for 100 ticks between a second +on this package, that means my thread has been scheduled for 1/4 of the whole +package. With that, the calculation of the energy spent by the thread on this +package during this whole second is 1/4 of the total energy spent by the +package. + +Usage +----- + +This feature is activated with -accel +kvm,rapl=true,rapl-helper-socket=/path/sock.sock + +It is important that the socket path is the same as the one +:program:`qemu-vmsr-helper` is listneing to. + +Current Limitations +------------------- + +- Works only on Intel host CPUs because AMD CPUs are using different MSR + addresses. + +- Only the Package Power-Plane (MSR_PKG_ENERGY_STATUS) is reported at the + moment. + +References +---------- + +.. [#a] https://patchwork.kernel.org/project/kvm/patch/20200916202951.23760-7-graf@amazon.com/ +.. [#b] https://man7.org/linux/man-pages/man5/proc.5.html diff --git a/include/sysemu/kvm_int.h b/include/sysemu/kvm_int.h index 882e37e12c5b..ee2fe8817833 100644 --- a/include/sysemu/kvm_int.h +++ b/include/sysemu/kvm_int.h @@ -14,6 +14,8 @@ #include "qemu/accel.h" #include "qemu/queue.h" #include "sysemu/kvm.h" +#include "hw/boards.h" +#include "hw/i386/topology.h" typedef struct KVMSlot { @@ -48,6 +50,20 @@ typedef struct KVMMemoryListener { #define KVM_MSI_HASHTAB_SIZE 256 +struct KVMMsrEnergy { + bool enable; + char *socket_path; + QemuThread msr_thr; + unsigned int cpus; + unsigned int sockets; + X86CPUTopoInfo topo_info; + const CPUArchIdList *cpu_list; + uint64_t *msr_value; + uint64_t msr_unit; + uint64_t msr_limit; + uint64_t msr_info; +}; + enum KVMDirtyRingReaperState { KVM_DIRTY_RING_REAPER_NONE = 0, /* The reaper is sleeping */ @@ -114,6 +130,7 @@ struct KVMState bool kvm_dirty_ring_with_bitmap; uint64_t kvm_eager_split_size; /* Eager Page Splitting chunk size */ struct KVMDirtyRingReaper reaper; + struct KVMMsrEnergy msr_energy; NotifyVmexitOption notify_vmexit; uint32_t notify_window; uint32_t xen_version; diff --git a/target/i386/cpu.h b/target/i386/cpu.h index 7f0786e8b98f..8d861296b1af 100644 --- a/target/i386/cpu.h +++ b/target/i386/cpu.h @@ -396,6 +396,10 @@ typedef enum X86Seg { #define MSR_IA32_TSX_CTRL 0x122 #define MSR_IA32_TSCDEADLINE 0x6e0 #define MSR_IA32_PKRS 0x6e1 +#define MSR_RAPL_POWER_UNIT 0x00000606 +#define MSR_PKG_POWER_LIMIT 0x00000610 +#define MSR_PKG_ENERGY_STATUS 0x00000611 +#define MSR_PKG_POWER_INFO 0x00000614 #define MSR_ARCH_LBR_CTL 0x000014ce #define MSR_ARCH_LBR_DEPTH 0x000014cf #define MSR_ARCH_LBR_FROM_0 0x00001500 @@ -1788,6 +1792,10 @@ typedef struct CPUArchState { uintptr_t retaddr; + /* RAPL MSR */ + uint64_t msr_rapl_power_unit; + uint64_t msr_pkg_energy_status; + /* Fields up to this point are cleared by a CPU reset */ struct {} end_reset_fields; diff --git a/target/i386/kvm/kvm.c b/target/i386/kvm/kvm.c index 76a66246eb72..e6cb315c0a90 100644 --- a/target/i386/kvm/kvm.c +++ b/target/i386/kvm/kvm.c @@ -16,16 +16,22 @@ #include "qapi/qapi-events-run-state.h" #include "qapi/error.h" #include "qapi/visitor.h" +#include <math.h> +#include <stdint.h> #include <sys/ioctl.h> #include <sys/utsname.h> #include <sys/syscall.h> +#include <sys/resource.h> +#include <sys/time.h> #include <linux/kvm.h> +#include <unistd.h> #include "standard-headers/asm-x86/kvm_para.h" #include "hw/xen/interface/arch-x86/cpuid.h" #include "cpu.h" #include "host-cpu.h" +#include "vmsr_energy.h" #include "sysemu/sysemu.h" #include "sysemu/hw_accel.h" #include "sysemu/kvm_int.h" @@ -2475,6 +2481,49 @@ static bool kvm_rdmsr_core_thread_count(X86CPU *cpu, uint32_t msr, return true; } +static bool kvm_rdmsr_rapl_power_unit(X86CPU *cpu, uint32_t msr, + uint64_t *val) +{ + + CPUState *cs = CPU(cpu); + + *val = cs->kvm_state->msr_energy.msr_unit; + + return true; +} + +static bool kvm_rdmsr_pkg_power_limit(X86CPU *cpu, uint32_t msr, + uint64_t *val) +{ + + CPUState *cs = CPU(cpu); + + *val = cs->kvm_state->msr_energy.msr_limit; + + return true; +} + +static bool kvm_rdmsr_pkg_power_info(X86CPU *cpu, uint32_t msr, + uint64_t *val) +{ + + CPUState *cs = CPU(cpu); + + *val = cs->kvm_state->msr_energy.msr_info; + + return true; +} + +static bool kvm_rdmsr_pkg_energy_status(X86CPU *cpu, uint32_t msr, + uint64_t *val) +{ + + CPUState *cs = CPU(cpu); + *val = cs->kvm_state->msr_energy.msr_value[cs->cpu_index]; + + return true; +} + static Notifier smram_machine_done; static KVMMemoryListener smram_listener; static AddressSpace smram_address_space; @@ -2509,6 +2558,265 @@ static void register_smram_listener(Notifier *n, void *unused) &smram_address_space, 1, "kvm-smram"); } +static void *kvm_msr_energy_thread(void *data) +{ + KVMState *s = data; + struct KVMMsrEnergy *vmsr = &s->msr_energy; + + g_autofree package_energy_stat *pkg_stat = NULL; + g_autofree thread_stat *thd_stat = NULL; + g_autofree pid_t *thread_ids = NULL; + g_autofree CPUState *cpu = NULL; + unsigned int maxpkgs, maxcpus, maxticks; + g_autofree unsigned int *vpkgs_energy_stat = NULL; + unsigned int num_threads = 0; + unsigned int tmp_num_threads = 0; + pid_t pid; + + X86CPUTopoIDs topo_ids; + + + rcu_register_thread(); + + /* Get QEMU PID*/ + pid = getpid(); + + /* Nb of CPUS per packages */ + maxcpus = vmsr_get_maxcpus(0); + + /* Nb of Physical Packages on the system */ + maxpkgs = vmsr_get_max_physical_package(maxcpus); + + /* Those MSR values should not change as well */ + vmsr->msr_unit = vmsr_read_msr(MSR_RAPL_POWER_UNIT, 0, pid, + s->msr_energy.socket_path); + vmsr->msr_limit = vmsr_read_msr(MSR_PKG_POWER_LIMIT, 0, pid, + s->msr_energy.socket_path); + vmsr->msr_info = vmsr_read_msr(MSR_PKG_POWER_INFO, 0, pid, + s->msr_energy.socket_path); + + /* Allocate memory for each package energy status */ + pkg_stat = (package_energy_stat *) + g_new0(package_energy_stat, maxpkgs); + + /* Pre-allocate memory for thread stats */ + thd_stat = g_new0(thread_stat, 1); + + /* Pre-allocate memory for holding Virtual Package Energy counter */ + vpkgs_energy_stat = g_new0(unsigned int, vmsr->sockets); + + /* + * Max numbers of ticks per package + * time in second * number of ticks/second * Number of cores / package + * ex: for 100 ticks/second/CPU, 12 CPUs per Package gives 1200 ticks max + */ + maxticks = (MSR_ENERGY_THREAD_SLEEP_US / 1000000) + * sysconf(_SC_CLK_TCK) * maxcpus; + + while (true) { + /* Get all qemu threads id */ + thread_ids = vmsr_get_thread_ids(pid, &num_threads); + + if (thread_ids == NULL) { + goto clean; + } + + if (tmp_num_threads < num_threads) { + thd_stat = g_renew(thread_stat, thd_stat, num_threads); + } + + tmp_num_threads = num_threads; + + /* Populate all the thread stats */ + for (int i = 0; i < num_threads; i++) { + thd_stat[i].utime = g_new0(unsigned long long, 2); + thd_stat[i].stime = g_new0(unsigned long long, 2); + thd_stat[i].thread_id = thread_ids[i]; + vmsr_read_thread_stat(&thd_stat[i], pid, 0); + thd_stat[i].numa_node_id = numa_node_of_cpu(thd_stat[i].cpu_id); + } + + /* Retrieve all packages power plane energy counter */ + for (int i = 0; i <= maxpkgs; i++) { + for (int j = 0; j < num_threads; j++) { + /* + * Use the first thread we found that ran on the CPU + * of the package to read the packages energy counter + */ + if (thd_stat[j].numa_node_id == i) { + pkg_stat[i].e_start = + vmsr_read_msr(MSR_PKG_ENERGY_STATUS, i, pid, + s->msr_energy.socket_path); + break; + } + } + } + + /* Sleep a short period while the other threads are working */ + usleep(MSR_ENERGY_THREAD_SLEEP_US); + + /* + * Retrieve all packages power plane energy counter + * Calculate the delta of all packages + */ + for (int i = 0; i <= maxpkgs; i++) { + for (int j = 0; j < num_threads; j++) { + /* + * Use the first thread we found that ran on the CPU + * of the package to read the packages energy counter + */ + if (thd_stat[j].numa_node_id == i) { + pkg_stat[i].e_end = + vmsr_read_msr(MSR_PKG_ENERGY_STATUS, + thd_stat[j].cpu_id, + thd_stat[j].thread_id, + s->msr_energy.socket_path); + /* + * Prevent the case we have migrate the VM + * during the sleep period or any other cases + * were energy counter might be lower after + * the sleep. + */ + if (pkg_stat[i].e_end > pkg_stat[i].e_start) { + pkg_stat[i].e_delta = + pkg_stat[i].e_end - pkg_stat[i].e_start; + } else { + pkg_stat[i].e_delta = 0; + } + break; + } + } + } + + /* Delta of ticks spend by each thread between the sample */ + for (int i = 0; i < num_threads; i++) { + if (vmsr_read_thread_stat(&thd_stat[i], pid, 1) != 0) { + /* + * We don't count the dead thread + * i.e threads that existed before the sleep + * and not anymore + */ + thd_stat[i].delta_ticks = 0; + } else { + vmsr_delta_ticks(thd_stat, i); + } + } + + /* + * Identify the vCPU threads + * Calculate the Number of vCPU per package + */ + CPU_FOREACH(cpu) { + for (int i = 0; i < num_threads; i++) { + if (cpu->thread_id == thd_stat[i].thread_id) { + thd_stat[i].is_vcpu = true; + thd_stat[i].vcpu_id = cpu->cpu_index; + pkg_stat[thd_stat[i].numa_node_id].nb_vcpu++; + thd_stat[i].acpi_id = kvm_arch_vcpu_id(cpu); + break; + } + } + } + + /* Retrieve the virtual package number of each vCPU */ + for (int i = 0; i < vmsr->x86_cpu_list->len; i++) { + for (int j = 0; j < num_threads; j++) { + if ((thd_stat[j].acpi_id == vmsr->x86_cpu_list->cpus[i].arch_id) + && (thd_stat[j].is_vcpu == true)) { + x86_topo_ids_from_apicid(thd_stat[j].acpi_id, + &vmsr->topo_info, &topo_ids); + thd_stat[j].vpkg = topo_ids.pkg_id; + } + } + } + + /* Calculate the total energy of all non-vCPU thread */ + for (int i = 0; i < num_threads; i++) { + double temp; + if ((thd_stat[i].is_vcpu != true) && + (thd_stat[i].delta_ticks > 0)) { + temp = vmsr_get_ratio(pkg_stat, thd_stat, maxticks, i); + pkg_stat[thd_stat[i].numa_node_id].e_ratio + += (uint64_t)lround(temp); + } + } + + /* Calculate the ratio per non-vCPU thread of each package */ + for (int i = 0; i <= maxpkgs; i++) { + if (pkg_stat[i].nb_vcpu > 0) { + pkg_stat[i].e_ratio = pkg_stat[i].e_ratio / pkg_stat[i].nb_vcpu; + } + } + + /* + * Calculate the energy for each Package: + * Energy Package = sum of each vCPU energy that belongs to the package + */ + for (int i = 0; i < num_threads; i++) { + double temp; + + if ((thd_stat[i].is_vcpu == true) && \ + (thd_stat[i].delta_ticks > 0)) { + temp = vmsr_get_ratio(pkg_stat, thd_stat, maxticks, i); + + vpkgs_energy_stat[thd_stat[i].vpkg] += (uint64_t)lround(temp); + vpkgs_energy_stat[thd_stat[i].vpkg] += + pkg_stat[thd_stat[i].numa_node_id].e_ratio; + } + } + + /* + * Finally populate the vmsr register of each vCPU with the total + * package value to emulate the real hardware where each CPU return the + * value of the package it belongs. + */ + for (int i = 0; i < num_threads; i++) { + if ((thd_stat[i].is_vcpu == true) && \ + (thd_stat[i].delta_ticks > 0)) { + vmsr->msr_value[thd_stat[i].vcpu_id] = \ + vpkgs_energy_stat[thd_stat[i].vpkg]; + } + } + + /* Zero out the memory */ + for (int i = 0; i < num_threads; i++) { + memset(thd_stat[i].utime, 0, 2 * sizeof(unsigned long long)); + memset(thd_stat[i].stime, 0, 2 * sizeof(unsigned long long)); + } + memset(thd_stat, 0, num_threads * sizeof(thread_stat)); + memset(thread_ids, 0, sizeof(pid_t)); + } + +clean: + rcu_unregister_thread(); + return NULL; +} + +static int kvm_msr_energy_thread_init(KVMState *s, MachineState *ms) +{ + struct KVMMsrEnergy *r = &s->msr_energy; + + vmsr_init_topo_info(&r->topo_info, ms); + + /* Retrieve the number of vCPU */ + r->cpus = ms->smp.cpus; + + /* Retrieve the number of sockets */ + r->sockets = ms->smp.sockets; + + /* Allocate register memory (MSR_PKG_STATUS) for each vCPU */ + r->msr_value = g_new0(uint64_t, r->cpus); + + /* Retrieve the CPUArchIDlist */ + r->x86_cpu_list = x86_possible_cpu_arch_ids(ms); + + qemu_thread_create(&r->msr_thr, "kvm-msr", + kvm_msr_energy_thread, + s, QEMU_THREAD_JOINABLE); + + return 0; +} + int kvm_arch_get_default_type(MachineState *ms) { return 0; @@ -2711,6 +3019,46 @@ int kvm_arch_init(MachineState *ms, KVMState *s) strerror(-ret)); exit(1); } + + if (s->msr_energy.enable == true) { + + r = kvm_filter_msr(s, MSR_RAPL_POWER_UNIT, + kvm_rdmsr_rapl_power_unit, NULL); + if (!r) { + error_report("Could not install MSR_RAPL_POWER_UNIT \ + handler: %s", + strerror(-ret)); + exit(1); + } + + r = kvm_filter_msr(s, MSR_PKG_POWER_LIMIT, + kvm_rdmsr_pkg_power_limit, NULL); + if (!r) { + error_report("Could not install MSR_PKG_POWER_LIMIT \ + handler: %s", + strerror(-ret)); + exit(1); + } + + r = kvm_filter_msr(s, MSR_PKG_POWER_INFO, + kvm_rdmsr_pkg_power_info, NULL); + if (!r) { + error_report("Could not install MSR_PKG_POWER_INFO \ + handler: %s", + strerror(-ret)); + exit(1); + } + r = kvm_filter_msr(s, MSR_PKG_ENERGY_STATUS, + kvm_rdmsr_pkg_energy_status, NULL); + if (!r) { + error_report("Could not install MSR_PKG_ENERGY_STATUS \ + handler: %s", + strerror(-ret)); + exit(1); + } else { + kvm_msr_energy_thread_init(s, ms); + } + } } return 0; diff --git a/target/i386/kvm/meson.build b/target/i386/kvm/meson.build index 84d9143e6029..16010638df69 100644 --- a/target/i386/kvm/meson.build +++ b/target/i386/kvm/meson.build @@ -3,6 +3,7 @@ i386_kvm_ss = ss.source_set() i386_kvm_ss.add(files( 'kvm.c', 'kvm-cpu.c', + 'vmsr_energy.c', )) i386_kvm_ss.add(when: 'CONFIG_XEN_EMU', if_true: files('xen-emu.c')) diff --git a/target/i386/kvm/vmsr_energy.c b/target/i386/kvm/vmsr_energy.c new file mode 100644 index 000000000000..27588630efa4 --- /dev/null +++ b/target/i386/kvm/vmsr_energy.c @@ -0,0 +1,295 @@ +/* + * QEMU KVM support -- x86 virtual energy-related MSR. + * + * Copyright 2023 Red Hat, Inc. 2023 + * + * Author: + * Anthony Harivel <aharivel@redhat.com> + * + * This work is licensed under the terms of the GNU GPL, version 2 or later. + * See the COPYING file in the top-level directory. + * + */ + +#include "vmsr_energy.h" +#include "qapi/error.h" +#include "io/channel.h" +#include "io/channel-socket.h" +#include "hw/boards.h" + +#define MAX_PATH_LEN 256 +#define MAX_LINE_LEN 500 + +static char *compute_default_paths(void) +{ + g_autofree char *state = qemu_get_local_state_dir(); + + return g_build_filename(state, "run", "qemu-vmsr-helper.sock", NULL); +} + +static int vmsr_helper_socket_read(QIOChannel *ioc, + void *buf, int sz, Error **errp) +{ + ssize_t r = qio_channel_read_all(ioc, buf, sz, errp); + + if (r < 0) { + object_unref(OBJECT(ioc)); + ioc = NULL; + return -EINVAL; + } + + return 0; +} + +static int vmsr_helper_socket_write(QIOChannel *ioc, + int fd, + const void *buf, int sz, Error **errp) +{ + size_t nfds = (fd != -1); + while (sz > 0) { + struct iovec iov; + ssize_t n_written; + + iov.iov_base = (void *)buf; + iov.iov_len = sz; + n_written = qio_channel_writev_full(QIO_CHANNEL(ioc), &iov, 1, + nfds ? &fd : NULL, nfds, 0, errp); + + if (n_written <= 0) { + assert(n_written != QIO_CHANNEL_ERR_BLOCK); + object_unref(OBJECT(ioc)); + ioc = NULL; + return n_written < 0 ? -EINVAL : 0; + } + + nfds = 0; + buf += n_written; + sz -= n_written; + } + + return 0; +} + +uint64_t vmsr_read_msr(uint32_t reg, unsigned int cpu_id, uint32_t tid, + const char *path) +{ + uint64_t data = 0; + char *socket_path = NULL; + unsigned int buffer[3]; + + if (path == NULL) { + socket_path = compute_default_paths(); + } else { + socket_path = g_strdup(path); + } + + SocketAddress saddr = { + .type = SOCKET_ADDRESS_TYPE_UNIX, + .u.q_unix.path = socket_path + }; + QIOChannelSocket *sioc = qio_channel_socket_new(); + Error *local_err = NULL; + + int r; + + qio_channel_set_name(QIO_CHANNEL(sioc), "vmsr-helper"); + qio_channel_socket_connect_sync(sioc, + &saddr, + &local_err); + g_free(socket_path); + if (local_err) { + goto out_close; + } + + /* + * Send the required arguments: + * 1. RAPL MSR register to read + * 2. On which CPU ID + * 3. From which vCPU (Thread ID) + */ + buffer[0] = reg; + buffer[1] = cpu_id; + buffer[2] = tid; + + r = vmsr_helper_socket_write(QIO_CHANNEL(sioc), + -1, + &buffer, sizeof(buffer), + &local_err); + if (r < 0) { + goto out_close; + } + + r = vmsr_helper_socket_read(QIO_CHANNEL(sioc), + &data, sizeof(data), + &local_err); + if (r < 0) { + data = 0; + goto out_close; + } + +out_close: + /* Close socket. */ + qio_channel_close(QIO_CHANNEL(sioc), NULL); + object_unref(OBJECT(sioc)); + return data; +} + +/* Retrieve the max number of physical CPU on the package */ +unsigned int vmsr_get_maxcpus(unsigned int package_num) +{ + int k, ncpus; + unsigned int maxcpus; + struct bitmask *cpus; + + cpus = numa_allocate_cpumask(); + ncpus = cpus->size; + + if (numa_node_to_cpus(package_num, cpus) < 0) { + return 0; + } + + maxcpus = 0; + for (k = 0; k < ncpus; k++) { + if (numa_bitmask_isbitset(cpus, k)) { + maxcpus++; + } + } + + return maxcpus; +} + +/* Retrieve the maximum number of physical packages */ +unsigned int vmsr_get_max_physical_package(unsigned int max_cpus) +{ + unsigned int packageCount = 0; + const char *dir = "/sys/devices/system/cpu/"; + int *uniquePackages; + + char *filePath; + FILE *file; + + uniquePackages = g_new0(int, max_cpus); + + for (int i = 0; i < max_cpus; i++) { + filePath = g_build_filename(dir, g_strdup_printf("cpu%d", i), + "topology/physical_package_id", NULL); + + file = fopen(filePath, "r"); + + if (file == NULL) { + perror("Error opening file"); + g_free(filePath); + g_free(uniquePackages); + return 0; + } + + char packageId[10]; + if (fgets(packageId, sizeof(packageId), file) == NULL) { + packageCount = 0; + } + + fclose(file); + + int currentPackageId = atoi(packageId); + + bool isUnique = true; + for (int j = 0; j < packageCount; j++) { + if (uniquePackages[j] == currentPackageId) { + isUnique = false; + break; + } + } + + if (isUnique) { + uniquePackages[packageCount] = currentPackageId; + packageCount++; + + if (packageCount >= max_cpus) { + break; + } + } + } + + g_free(filePath); + g_free(uniquePackages); + return (packageCount == 0) ? 1 : packageCount; +} +int vmsr_read_thread_stat(struct thread_stat *thread, int pid, int index) +{ + char *path; + path = g_new0(char, MAX_PATH_LEN); + + path = g_build_filename(g_strdup_printf("/proc/%u/task/%d/stat", pid, \ + thread->thread_id), NULL); + + FILE *file = fopen(path, "r"); + if (file == NULL) { + return -1; + } + + if (fscanf(file, "%*d (%*[^)]) %*c %*d %*d %*d %*d %*d %*u %*u %*u %*u %*u" + " %llu %llu %*d %*d %*d %*d %*d %*d %*u %*u %*d %*u %*u" + " %*u %*u %*u %*u %*u %*u %*u %*u %*u %*d %*u %*u %u", + &thread->utime[index], &thread->stime[index], &thread->cpu_id) != 3) + return -1; + + fclose(file); + return 0; +} + +/* Read QEMU stat task folder to retrieve all QEMU threads ID */ +pid_t *vmsr_get_thread_ids(pid_t pid, unsigned int *num_threads) +{ + char *path = g_build_filename("/proc", g_strdup_printf("%d/task", pid), NULL); + + DIR *dir = opendir(path); + if (dir == NULL) { + perror("opendir"); + g_free(path); + return NULL; + } + + pid_t *thread_ids = NULL; + unsigned int thread_count = 0; + + struct dirent *ent; + while ((ent = readdir(dir)) != NULL) { + if (ent->d_name[0] == '.') { + continue; + } + pid_t tid = atoi(ent->d_name); + if (pid != tid) { + thread_ids = g_renew(pid_t, thread_ids, (thread_count + 1)); + thread_ids[thread_count] = tid; + thread_count++; + } + } + + closedir(dir); + + *num_threads = thread_count; + g_free(path); + return thread_ids; +} + +void vmsr_delta_ticks(thread_stat *thd_stat, int i) +{ + thd_stat[i].delta_ticks = (thd_stat[i].utime[1] + thd_stat[i].stime[1]) + - (thd_stat[i].utime[0] + thd_stat[i].stime[0]); +} + +double vmsr_get_ratio(package_energy_stat *pkg_stat, + thread_stat *thd_stat, + int maxticks, int i) { + + return (pkg_stat[thd_stat[i].numa_node_id].e_delta / 100.0) + * ((100.0 / maxticks) * thd_stat[i].delta_ticks); +} + +void vmsr_init_topo_info(X86CPUTopoInfo *topo_info, + const MachineState *ms) +{ + topo_info->dies_per_pkg = ms->smp.dies; + topo_info->cores_per_die = ms->smp.cores; + topo_info->threads_per_core = ms->smp.threads; +} diff --git a/target/i386/kvm/vmsr_energy.h b/target/i386/kvm/vmsr_energy.h new file mode 100644 index 000000000000..385d53b9c16c --- /dev/null +++ b/target/i386/kvm/vmsr_energy.h @@ -0,0 +1,87 @@ +/* + * QEMU KVM support -- x86 virtual energy-related MSR. + * + * Copyright 2023 Red Hat, Inc. 2023 + * + * Author: + * Anthony Harivel <aharivel@redhat.com> + * + * This work is licensed under the terms of the GNU GPL, version 2 or later. + * See the COPYING file in the top-level directory. + * + */ + +#ifndef VMSR_ENERGY_H +#define VMSR_ENERGY_H + +#include "qemu/osdep.h" +#include <numa.h> +#include <stdint.h> +#include "hw/i386/topology.h" + +/* + * Define the interval time in micro seconds between 2 samples of + * energy related MSRs + */ +#define MSR_ENERGY_THREAD_SLEEP_US 1000000.0 + +/* + * Thread statistic + * @ thread_id: TID (thread ID) + * @ is_vcpu: true is thread is vCPU thread + * @ cpu_id: CPU number last executed on + * @ vcpu_id: vCPU ID + * @ numa_node_id:node number of the CPU + * @ vpkg: virtual package number + * @ acpi_id: APIC id of the vCPU + * @ utime: amount of clock ticks the thread + * has been scheduled in User mode + * @ stime: amount of clock ticks the thread + * has been scheduled in System mode + * @ delta_ticks: delta of utime+stime between + * the two samples (before/after sleep) + */ +struct thread_stat { + unsigned int thread_id; + bool is_vcpu; + unsigned int cpu_id; + unsigned int vcpu_id; + unsigned int numa_node_id; + unsigned int vpkg; + unsigned long acpi_id; + unsigned long long *utime; + unsigned long long *stime; + unsigned long long delta_ticks; +}; + +/* + * Package statistic + * @ e_start: package energy counter before the sleep + * @ e_end: package energy counter after the sleep + * @ e_delta: delta of package energy counter + * @ e_ratio: store the energy ratio of non-vCPU thread + * @ nb_vcpu: number of vCPU running on this package + */ +struct package_energy_stat { + uint64_t e_start; + uint64_t e_end; + uint64_t e_delta; + uint64_t e_ratio; + unsigned int nb_vcpu; +}; + +typedef struct thread_stat thread_stat; +typedef struct package_energy_stat package_energy_stat; + +uint64_t vmsr_read_msr(uint32_t reg, unsigned int cpu_id, + unsigned int tid, const char *path); +void vmsr_delta_ticks(thread_stat *thd_stat, int i); +unsigned int vmsr_get_maxcpus(unsigned int package_num); +unsigned int vmsr_get_max_physical_package(unsigned int max_cpus); +int vmsr_read_thread_stat(struct thread_stat *thread, int pid, int index); +pid_t *vmsr_get_thread_ids(pid_t pid, unsigned int *num_threads); +double vmsr_get_ratio(package_energy_stat *pkg_stat, + thread_stat *thd_stat, + int maxticks, int i); +void vmsr_init_topo_info(X86CPUTopoInfo *topo_info, const MachineState *ms); +#endif /* VMSR_ENERGY_H */
Starting with the "Sandy Bridge" generation, Intel CPUs provide a RAPL interface (Running Average Power Limit) for advertising the accumulated energy consumption of various power domains (e.g. CPU packages, DRAM, etc.). The consumption is reported via MSRs (model specific registers) like MSR_PKG_ENERGY_STATUS for the CPU package power domain. These MSRs are 64 bits registers that represent the accumulated energy consumption in micro Joules. They are updated by microcode every ~1ms. For now, KVM always returns 0 when the guest requests the value of these MSRs. Use the KVM MSR filtering mechanism to allow QEMU handle these MSRs dynamically in userspace. To limit the amount of system calls for every MSR call, create a new thread in QEMU that updates the "virtual" MSR values asynchronously. Each vCPU has its own vMSR to reflect the independence of vCPUs. The thread updates the vMSR values with the ratio of energy consumed of the whole physical CPU package the vCPU thread runs on and the thread's utime and stime values. All other non-vCPU threads are also taken into account. Their energy consumption is evenly distributed among all vCPUs threads running on the same physical CPU package. To overcome the problem that reading the RAPL MSR requires priviliged access, a socket communication between QEMU and the qemu-vmsr-helper is mandatory. You can specified the socket path in the parameter. This feature is activated with -accel kvm,rapl=true,path=/path/sock.sock Actual limitation: - Works only on Intel host CPU because AMD CPUs are using different MSR adresses. - Only the Package Power-Plane (MSR_PKG_ENERGY_STATUS) is reported at the moment. Signed-off-by: Anthony Harivel <aharivel@redhat.com> --- accel/kvm/kvm-all.c | 27 +++ docs/specs/index.rst | 1 + docs/specs/rapl-msr.rst | 133 +++++++++++++ include/sysemu/kvm_int.h | 17 ++ target/i386/cpu.h | 8 + target/i386/kvm/kvm.c | 348 ++++++++++++++++++++++++++++++++++ target/i386/kvm/meson.build | 1 + target/i386/kvm/vmsr_energy.c | 295 ++++++++++++++++++++++++++++ target/i386/kvm/vmsr_energy.h | 87 +++++++++ 9 files changed, 917 insertions(+) create mode 100644 docs/specs/rapl-msr.rst create mode 100644 target/i386/kvm/vmsr_energy.c create mode 100644 target/i386/kvm/vmsr_energy.h