cosmopolitan/libc/calls/poll-nt.c
Justine Tunney ec480f5aa0
Make improvements
- Every unit test now passes on Apple Silicon. The final piece of this
  puzzle was porting our POSIX threads cancelation support, since that
  works differently on ARM64 XNU vs. AMD64. Our semaphore support on
  Apple Silicon is also superior now compared to AMD64, thanks to the
  grand central dispatch library which lets *NSYNC locks go faster.

- The Cosmopolitan runtime is now more stable, particularly on Windows.
  To do this, thread local storage is mandatory at all runtime levels,
  and the innermost packages of the C library is no longer being built
  using ASAN. TLS is being bootstrapped with a 128-byte TIB during the
  process startup phase, and then later on the runtime re-allocates it
  either statically or dynamically to support code using _Thread_local.
  fork() and execve() now do a better job cooperating with threads. We
  can now check how much stack memory is left in the process or thread
  when functions like kprintf() / execve() etc. call alloca(), so that
  ENOMEM can be raised, reduce a buffer size, or just print a warning.

- POSIX signal emulation is now implemented the same way kernels do it
  with pthread_kill() and raise(). Any thread can interrupt any other
  thread, regardless of what it's doing. If it's blocked on read/write
  then the killer thread will cancel its i/o operation so that EINTR can
  be returned in the mark thread immediately. If it's doing a tight CPU
  bound operation, then that's also interrupted by the signal delivery.
  Signal delivery works now by suspending a thread and pushing context
  data structures onto its stack, and redirecting its execution to a
  trampoline function, which calls SetThreadContext(GetCurrentThread())
  when it's done.

- We're now doing a better job managing locks and handles. On NetBSD we
  now close semaphore file descriptors in forked children. Semaphores on
  Windows can now be canceled immediately, which means mutexes/condition
  variables will now go faster. Apple Silicon semaphores can be canceled
  too. We're now using Apple's pthread_yield() funciton. Apple _nocancel
  syscalls are now used on XNU when appropriate to ensure pthread_cancel
  requests aren't lost. The MbedTLS library has been updated to support
  POSIX thread cancelations. See tool/build/runitd.c for an example of
  how it can be used for production multi-threaded tls servers. Handles
  on Windows now leak less often across processes. All i/o operations on
  Windows are now overlapped, which means file pointers can no longer be
  inherited across dup() and fork() for the time being.

- We now spawn a thread on Windows to deliver SIGCHLD and wakeup wait4()
  which means, for example, that posix_spawn() now goes 3x faster. POSIX
  spawn is also now more correct. Like Musl, it's now able to report the
  failure code of execve() via a pipe although our approach favors using
  shared memory to do that on systems that have a true vfork() function.

- We now spawn a thread to deliver SIGALRM to threads when setitimer()
  is used. This enables the most precise wakeups the OS makes possible.

- The Cosmopolitan runtime now uses less memory. On NetBSD for example,
  it turned out the kernel would actually commit the PT_GNU_STACK size
  which caused RSS to be 6mb for every process. Now it's down to ~4kb.
  On Apple Silicon, we reduce the mandatory upstream thread size to the
  smallest possible size to reduce the memory overhead of Cosmo threads.
  The examples directory has a program called greenbean which can spawn
  a web server on Linux with 10,000 worker threads and have the memory
  usage of the process be ~77mb. The 1024 byte overhead of POSIX-style
  thread-local storage is now optional; it won't be allocated until the
  pthread_setspecific/getspecific functions are called. On Windows, the
  threads that get spawned which are internal to the libc implementation
  use reserve rather than commit memory, which shaves a few hundred kb.

- sigaltstack() is now supported on Windows, however it's currently not
  able to be used to handle stack overflows, since crash signals are
  still generated by WIN32. However the crash handler will still switch
  to the alt stack, which is helpful in environments with tiny threads.

- Test binaries are now smaller. Many of the mandatory dependencies of
  the test runner have been removed. This ensures many programs can do a
  better job only linking the the thing they're testing. This caused the
  test binaries for LIBC_FMT for example, to decrease from 200kb to 50kb

- long double is no longer used in the implementation details of libc,
  except in the APIs that define it. The old code that used long double
  for time (instead of struct timespec) has now been thoroughly removed.

- ShowCrashReports() is now much tinier in MODE=tiny. Instead of doing
  backtraces itself, it'll just print a command you can run on the shell
  using our new `cosmoaddr2line` program to view the backtrace.

- Crash report signal handling now works in a much better way. Instead
  of terminating the process, it now relies on SA_RESETHAND so that the
  default SIG_IGN behavior can terminate the process if necessary.

- Our pledge() functionality has now been fully ported to AARCH64 Linux.
2023-09-18 21:04:47 -07:00

230 lines
8.2 KiB
C

/*-*- mode:c;indent-tabs-mode:nil;c-basic-offset:2;tab-width:8;coding:utf-8 -*-│
│vi: set net ft=c ts=2 sts=2 sw=2 fenc=utf-8 :vi│
╞══════════════════════════════════════════════════════════════════════════════╡
│ Copyright 2020 Justine Alexandra Roberts Tunney │
│ │
│ Permission to use, copy, modify, and/or distribute this software for │
│ any purpose with or without fee is hereby granted, provided that the │
│ above copyright notice and this permission notice appear in all copies. │
│ │
│ THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL │
│ WARRANTIES WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED │
│ WARRANTIES OF MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE │
│ AUTHOR BE LIABLE FOR ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL │
│ DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR │
│ PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER │
│ TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR │
│ PERFORMANCE OF THIS SOFTWARE. │
╚─────────────────────────────────────────────────────────────────────────────*/
#include "libc/calls/calls.h"
#include "libc/calls/internal.h"
#include "libc/calls/sig.internal.h"
#include "libc/calls/state.internal.h"
#include "libc/calls/struct/sigaction.h"
#include "libc/errno.h"
#include "libc/intrin/bits.h"
#include "libc/intrin/strace.internal.h"
#include "libc/macros.internal.h"
#include "libc/mem/mem.h"
#include "libc/nt/enum/filetype.h"
#include "libc/nt/errors.h"
#include "libc/nt/files.h"
#include "libc/nt/ipc.h"
#include "libc/nt/runtime.h"
#include "libc/nt/struct/pollfd.h"
#include "libc/nt/synchronization.h"
#include "libc/nt/thread.h"
#include "libc/nt/thunk/msabi.h"
#include "libc/nt/winsock.h"
#include "libc/sock/internal.h"
#include "libc/sock/struct/pollfd.h"
#include "libc/sock/struct/pollfd.internal.h"
#include "libc/sysv/consts/o.h"
#include "libc/sysv/consts/poll.h"
#include "libc/sysv/consts/sig.h"
#include "libc/sysv/errfuns.h"
#include "libc/thread/posixthread.internal.h"
#ifdef __x86_64__
__static_yoink("WinMainStdin");
/*
* Polls on the New Technology.
*
* This function is used to implement poll() and select(). You may poll
* on both sockets and files at the same time. We also poll for signals
* while poll is polling.
*/
textwindows int sys_poll_nt(struct pollfd *fds, uint64_t nfds, uint64_t *ms,
const sigset_t *sigmask) {
bool ok;
uint32_t avail;
sigset_t oldmask;
struct sys_pollfd_nt pipefds[8];
struct sys_pollfd_nt sockfds[64];
int pipeindices[ARRAYLEN(pipefds)];
int sockindices[ARRAYLEN(sockfds)];
int i, rc, sn, pn, gotinvals, gotpipes, gotsocks, waitfor;
// check for interrupts early before doing work
if (sigmask) {
__sig_mask(SIG_SETMASK, sigmask, &oldmask);
}
if ((rc = _check_interrupts(0))) {
goto ReturnPath;
}
// do the planning
// we need to read static variables
// we might need to spawn threads and open pipes
__fds_lock();
for (gotinvals = rc = sn = pn = i = 0; i < nfds; ++i) {
if (fds[i].fd < 0) continue;
if (__isfdopen(fds[i].fd)) {
if (__isfdkind(fds[i].fd, kFdSocket)) {
if (sn < ARRAYLEN(sockfds)) {
// the magnums for POLLIN/OUT/PRI on NT include the other ones too
// we need to clear ones like POLLNVAL or else WSAPoll shall whine
sockindices[sn] = i;
sockfds[sn].handle = g_fds.p[fds[i].fd].handle;
sockfds[sn].events = fds[i].events & (POLLPRI | POLLIN | POLLOUT);
sockfds[sn].revents = 0;
++sn;
} else {
// too many socket fds
rc = enomem();
break;
}
} else if (pn < ARRAYLEN(pipefds)) {
pipeindices[pn] = i;
pipefds[pn].handle = __resolve_stdin_handle(g_fds.p[fds[i].fd].handle);
pipefds[pn].events = 0;
pipefds[pn].revents = 0;
switch (g_fds.p[fds[i].fd].flags & O_ACCMODE) {
case O_RDONLY:
pipefds[pn].events = fds[i].events & POLLIN;
break;
case O_WRONLY:
pipefds[pn].events = fds[i].events & POLLOUT;
break;
case O_RDWR:
pipefds[pn].events = fds[i].events & (POLLIN | POLLOUT);
break;
default:
__builtin_unreachable();
}
++pn;
} else {
// too many non-socket fds
rc = enomem();
break;
}
} else {
++gotinvals;
}
}
__fds_unlock();
if (rc) {
// failed to create a polling solution
goto ReturnPath;
}
// perform the i/o and sleeping and looping
for (;;) {
// see if input is available on non-sockets
for (gotpipes = i = 0; i < pn; ++i) {
if (pipefds[i].events & POLLOUT) {
// we have no way of polling if a non-socket is writeable yet
// therefore we assume that if it can happen, it shall happen
pipefds[i].revents |= POLLOUT;
}
if (pipefds[i].events & POLLIN) {
if (GetFileType(pipefds[i].handle) == kNtFileTypePipe) {
ok = PeekNamedPipe(pipefds[i].handle, 0, 0, 0, &avail, 0);
POLLTRACE("PeekNamedPipe(%ld, 0, 0, 0, [%'u], 0) → %hhhd% m",
pipefds[i].handle, avail, ok);
if (ok) {
if (avail) {
pipefds[i].revents |= POLLIN;
}
} else {
pipefds[i].revents |= POLLERR;
}
} else {
// we have no way of polling if a non-socket is readable yet
// therefore we assume that if it can happen it shall happen
pipefds[i].revents |= POLLIN;
}
}
if (pipefds[i].revents) {
++gotpipes;
}
}
// if we haven't found any good results yet then here we
// compute a small time slice we don't mind sleeping for
waitfor = gotinvals || gotpipes ? 0 : MIN(__SIG_POLL_INTERVAL_MS, *ms);
if (sn) {
#if _NTTRACE
POLLTRACE("WSAPoll(%p, %u, %'d) out of %'lu", sockfds, sn, waitfor, *ms);
#endif
if ((gotsocks = WSAPoll(sockfds, sn, 0)) == -1) {
rc = __winsockerr();
goto ReturnPath;
}
} else {
gotsocks = 0;
}
if (!gotinvals && !gotsocks && !gotpipes && waitfor) {
POLLTRACE("sleeping for %'d out of %'lu ms", waitfor, *ms);
struct PosixThread *pt = _pthread_self();
pt->abort_errno = 0;
pt->pt_flags |= PT_INSEMAPHORE;
WaitForSingleObject(pt->semaphore, waitfor);
pt->pt_flags &= ~PT_INSEMAPHORE;
if (pt->abort_errno) {
errno = pt->abort_errno;
rc = -1;
goto ReturnPath;
}
*ms -= waitfor; // todo: make more resilient
}
// we gave all the sockets and all the named pipes a shot
// if we found anything at all then it's time to end work
if (gotinvals || gotpipes || gotsocks || *ms <= 0) {
break;
}
// otherwise loop limitlessly for timeout to elapse while
// checking for signal delivery interrupts, along the way
if ((rc = _check_interrupts(0))) {
goto ReturnPath;
}
}
// the system call is going to succeed
// it's now ok to start setting the output memory
for (i = 0; i < nfds; ++i) {
if (fds[i].fd < 0 || __isfdopen(fds[i].fd)) {
fds[i].revents = 0;
} else {
fds[i].revents = POLLNVAL;
}
}
for (i = 0; i < pn; ++i) {
fds[pipeindices[i]].revents = pipefds[i].revents;
}
for (i = 0; i < sn; ++i) {
fds[sockindices[i]].revents = sockfds[i].revents;
}
// and finally return
rc = gotinvals + gotpipes + gotsocks;
ReturnPath:
if (sigmask) {
__sig_mask(SIG_SETMASK, &oldmask, 0);
}
return rc;
}
#endif /* __x86_64__ */