• Andreas Gerstmayr's avatar
    fix iteration over CPUs · 7e0784d3
    Andreas Gerstmayr authored
    Since kernel version 4.9.0 BPF stopped working in a KVM guest.
    The problem are calls to perf_event_open with CPU identifiers which do
    not exist (ENODEV). The root cause for this is that the current code
    assumes ascending numbered CPUs. However, this is not always the case
    (e.g. CPU hotplugging).
    
    This patch introduces the get_online_cpus() and get_possible_cpus()
    helper functions and uses the appropriate function for iterations over
    CPUs. The BPF_MAP_TYPE_PERF_EVENT_ARRAY map contains now an entry for
    each possible CPU instead of for each online CPU.
    
    Fixes: #893
    Signed-off-by: default avatarAndreas Gerstmayr <andreas.gerstmayr@catalysts.cc>
    7e0784d3
BPFTable.cc 4.13 KB
/*
 * Copyright (c) 2016 Facebook, Inc.
 *
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 * http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

#include <errno.h>
#include <unistd.h>
#include <cstring>
#include <iostream>

#include "BPFTable.h"

#include "bcc_exception.h"
#include "bcc_syms.h"
#include "libbpf.h"
#include "perf_reader.h"
#include "common.h"

namespace ebpf {

BPFStackTable::~BPFStackTable() {
  for (auto it : pid_sym_)
    bcc_free_symcache(it.second, it.first);
}

std::vector<intptr_t> BPFStackTable::get_stack_addr(int stack_id) {
  std::vector<intptr_t> res;
  stacktrace_t stack;
  if (!lookup(&stack_id, &stack))
    return res;
  for (int i = 0; (i < BPF_MAX_STACK_DEPTH) && (stack.ip[i] != 0); i++)
    res.push_back(stack.ip[i]);
  return res;
}

std::vector<std::string> BPFStackTable::get_stack_symbol(int stack_id,
                                                         int pid) {
  auto addresses = get_stack_addr(stack_id);
  std::vector<std::string> res;
  res.reserve(addresses.size());

  if (pid < 0)
    pid = -1;
  if (pid_sym_.find(pid) == pid_sym_.end())
    pid_sym_[pid] = bcc_symcache_new(pid);
  void* cache = pid_sym_[pid];

  bcc_symbol symbol;
  for (auto addr : addresses)
    if (bcc_symcache_resolve(cache, addr, &symbol) != 0)
      res.push_back("[UNKNOWN]");
    else
      res.push_back(symbol.demangle_name);

  return res;
}

StatusTuple BPFPerfBuffer::open_on_cpu(perf_reader_raw_cb cb, int cpu,
                                       void* cb_cookie) {
  if (cpu_readers_.find(cpu) != cpu_readers_.end())
    return StatusTuple(-1, "Perf buffer already open on CPU %d", cpu);
  auto reader =
      static_cast<perf_reader*>(bpf_open_perf_buffer(cb, cb_cookie, -1, cpu));
  if (reader == nullptr)
    return StatusTuple(-1, "Unable to construct perf reader");
  int reader_fd = perf_reader_fd(reader);
  if (!update(&cpu, &reader_fd)) {
    perf_reader_free(static_cast<void*>(reader));
    return StatusTuple(-1, "Unable to open perf buffer on CPU %d: %s", cpu,
                       strerror(errno));
  }
  cpu_readers_[cpu] = reader;
  readers_.push_back(reader);
  return StatusTuple(0);
}

StatusTuple BPFPerfBuffer::open_all_cpu(perf_reader_raw_cb cb,
                                        void* cb_cookie) {
  if (cpu_readers_.size() != 0 || readers_.size() != 0)
    return StatusTuple(-1, "Previously opened perf buffer not cleaned");

  for (int i: get_online_cpus()) {
    auto res = open_on_cpu(cb, i, cb_cookie);
    if (res.code() != 0) {
      TRY2(close_all_cpu());
      return res;
    }
  }
  return StatusTuple(0);
}

StatusTuple BPFPerfBuffer::close_on_cpu(int cpu) {
  auto it = cpu_readers_.find(cpu);
  if (it == cpu_readers_.end())
    return StatusTuple(0);
  perf_reader_free(static_cast<void*>(it->second));
  if (!remove(const_cast<int*>(&(it->first))))
    return StatusTuple(-1, "Unable to close perf buffer on CPU %d", it->first);
  cpu_readers_.erase(it);
  return StatusTuple(0);
}

StatusTuple BPFPerfBuffer::close_all_cpu() {
  std::string errors;
  bool has_error = false;
  for (int i: get_online_cpus()) {
    auto res = close_on_cpu(i);
    if (res.code() != 0) {
      errors += "Failed to close CPU" + std::to_string(i) + " perf buffer: ";
      errors += res.msg() + "\n";
      has_error = true;
    }
  }
  readers_.clear();
  if (has_error)
    return StatusTuple(-1, errors);
  return StatusTuple(0);
}

void BPFPerfBuffer::poll(int timeout) {
  if (readers_.empty())
    return;
  perf_reader_poll(readers_.size(), readers_.data(), timeout);
}

BPFPerfBuffer::~BPFPerfBuffer() {
  auto res = close_all_cpu();
  if (res.code() != 0)
    std::cerr << "Failed to close all perf buffer on destruction: "
              << res.msg() << std::endl;
}

}  // namespace ebpf