zircon/kernel/lib/syscalls/syscalls.cc

// Copyright 2016 The Fuchsia Authors
//
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file or at
// https://opensource.org/licenses/MIT

#include <err.h>
#include <inttypes.h>
#include <lib/ktrace.h>
#include <lib/syscalls/zx-syscall-numbers.h>
#include <lib/userabi/vdso.h>
#include <platform.h>
#include <stdint.h>
#include <trace.h>

#include <kernel/stats.h>
#include <kernel/thread.h>
#include <object/process_dispatcher.h>
#include <syscalls/syscalls.h>

#include "priv.h"
#include "vdso-valid-sysret.h"

#define LOCAL_TRACE 0

// Main syscall dispatch routine. For every syscall in the system stamp out a separate
// wrapper_<name of syscall> routine using the do_syscall inline function instantiated
// from an a header generated from an external tool.
//
// The end result is a wrapper_<syscall> that does per syscall argument validation and
// argument marshalling to an inner routine called sys_<syscall>.

namespace {

__NO_INLINE int sys_invalid_syscall(uint64_t num, uint64_t pc, uintptr_t vdso_code_address) {
  LTRACEF("invalid syscall %lu from PC %#lx vDSO code %#lx\n", num, pc, vdso_code_address);
  Thread::Current::SignalPolicyException();
  return ZX_ERR_BAD_SYSCALL;
}

struct syscall_pre_out {
  uintptr_t vdso_code_address;
  ProcessDispatcher* current_process;
};

// N.B. Interrupts must be disabled on entry and they will be disabled on exit.
// The reason is the two calls two arch_curr_cpu_num in the ktrace calls: we
// don't want the cpu changing during the call.

// Try to do as much as possible in the shared preamble code to maximize code reuse
// between syscalls.
__NO_INLINE syscall_pre_out do_syscall_pre(uint64_t syscall_num, uint64_t pc) {
  ktrace_tiny(TAG_SYSCALL_ENTER, (static_cast<uint32_t>(syscall_num) << 8) | arch_curr_cpu_num());

  CPU_STATS_INC(syscalls);

  /* re-enable interrupts to maintain kernel preemptiveness
     This must be done after the above ktrace_tiny call, and after the
     above CPU_STATS_INC call as it also calls arch_curr_cpu_num. */
  arch_enable_ints();

  LTRACEF_LEVEL(2, "t %p syscall num %" PRIu64 " ip/pc %#" PRIx64 "\n", Thread::Current::Get(),
                syscall_num, pc);

  ProcessDispatcher* current_process = ProcessDispatcher::GetCurrent();
  uintptr_t vdso_code_address = current_process->vdso_code_address();

  return {vdso_code_address, current_process};
}

__NO_INLINE syscall_result do_syscall_post(uint64_t ret, uint64_t syscall_num) {
  LTRACEF_LEVEL(2, "t %p ret %#" PRIx64 "\n", Thread::Current::Get(), ret);

  /* re-disable interrupts on the way out
     This must be done before the below ktrace_tiny call. */
  arch_disable_ints();

  ktrace_tiny(TAG_SYSCALL_EXIT, (static_cast<uint32_t>(syscall_num << 8)) | arch_curr_cpu_num());

  // The assembler caller will re-disable interrupts at the appropriate time.
  return {ret, Thread::Current::Get()->IsSignaled()};
}

}  // namespace

// Stamped out syscall veneer routine for every syscall. Try to maximize shared code by forcing
// most of the setup and teardown code into non-inlined preamble and postamble code.
template <typename T>
inline syscall_result do_syscall(uint64_t syscall_num, uint64_t pc, bool (*valid_pc)(uintptr_t),
                                 T make_call) {
  // Call the shared preamble code
  auto pre_ret = do_syscall_pre(syscall_num, pc);
  const uintptr_t vdso_code_address = pre_ret.vdso_code_address;
  ProcessDispatcher* current_process = pre_ret.current_process;

  // Validate the user space program counter originated from the vdso at the proper location,
  // otherwise call through to the invalid syscall handler
  uint64_t ret;
  if (unlikely(!valid_pc(pc - vdso_code_address))) {
    ret = sys_invalid_syscall(syscall_num, pc, vdso_code_address);
  } else {
    // Per syscall inlined routine to marshall args appropriately
    ret = make_call(current_process);
  }

  // Call through to the shared postamble code
  return do_syscall_post(ret, syscall_num);
}

// Called when an out of bounds syscall number is passed from user space
syscall_result unknown_syscall(uint64_t syscall_num, uint64_t pc) {
  return do_syscall(
      syscall_num, pc, [](uintptr_t) { return false; },
      [&](ProcessDispatcher*) {
        __builtin_unreachable();
        return ZX_ERR_INTERNAL;
      });
}

// Autogenerated per-syscall wrapper functions.
#include <lib/syscalls/kernel-wrappers.inc>