Make improvements

- We now serialize the file descriptor table when spawning / executing
  processes on Windows. This means you can now inherit more stuff than
  just standard i/o. It's needed by bash, which duplicates the console
  to file descriptor #255. We also now do a better job serializing the
  environment variables, so you're less likely to encounter E2BIG when
  using your bash shell. We also no longer coerce environ to uppercase

- execve() on Windows now remotely controls its parent process to make
  them spawn a replacement for itself. Then it'll be able to terminate
  immediately once the spawn succeeds, without having to linger around
  for the lifetime as a shell process for proxying the exit code. When
  process worker thread running in the parent sees the child die, it's
  given a handle to the new child, to replace it in the process table.

- execve() and posix_spawn() on Windows will now provide CreateProcess
  an explicit handle list. This allows us to remove handle locks which
  enables better fork/spawn concurrency, with seriously correct thread
  safety. Other codebases like Go use the same technique. On the other
  hand fork() still favors the conventional WIN32 inheritence approach
  which can be a little bit messy, but is *controlled* by guaranteeing
  perfectly clean slates at both the spawning and execution boundaries

- sigset_t is now 64 bits. Having it be 128 bits was a mistake because
  there's no reason to use that and it's only supported by FreeBSD. By
  using the system word size, signal mask manipulation on Windows goes
  very fast. Furthermore @asyncsignalsafe funcs have been rewritten on
  Windows to take advantage of signal masking, now that it's much more
  pleasant to use.

- All the overlapped i/o code on Windows has been rewritten for pretty
  good signal and cancelation safety. We're now able to ensure overlap
  data structures are cleaned up so long as you don't longjmp() out of
  out of a signal handler that interrupted an i/o operation. Latencies
  are also improved thanks to the removal of lots of "busy wait" code.
  Waits should be optimal for everything except poll(), which shall be
  the last and final demon we slay in the win32 i/o horror show.

- getrusage() on Windows is now able to report RUSAGE_CHILDREN as well
  as RUSAGE_SELF, thanks to aggregation in the process manager thread.
This commit is contained in:
Justine Tunney 2023-10-08 05:36:18 -07:00
parent af7cb3c82f
commit 791f79fcb3
No known key found for this signature in database
GPG key ID: BE714B4575D6E328
382 changed files with 4008 additions and 4511 deletions

View file

@ -18,9 +18,8 @@
*/
#include "libc/sysv/consts/sig.h"
#include "ape/sections.internal.h"
#include "libc/assert.h"
#include "libc/atomic.h"
#include "libc/calls/blockcancel.internal.h"
#include "libc/calls/blocksigs.internal.h"
#include "libc/calls/calls.h"
#include "libc/calls/sig.internal.h"
#include "libc/calls/state.internal.h"
@ -56,10 +55,14 @@
#include "libc/sysv/consts/sicode.h"
#include "libc/sysv/consts/ss.h"
#include "libc/thread/posixthread.internal.h"
#include "libc/thread/thread.h"
#include "libc/thread/tls.h"
#ifdef __x86_64__
/**
* @fileoverview Cosmopolitan Signals for Windows.
*/
struct SignalFrame {
struct PosixThread *pt;
struct NtContext *nc;
@ -86,6 +89,17 @@ textwindows bool __sig_ignored(int sig) {
__sig_ignored_by_default(sig));
}
textwindows void __sig_delete(int sig) {
struct Dll *e;
__sig.pending &= ~(1ull << (sig - 1));
_pthread_lock();
for (e = dll_last(_pthread_list); e; e = dll_prev(_pthread_list, e)) {
struct PosixThread *pt = POSIXTHREAD_CONTAINER(e);
pt->tib->tib_sigpending &= ~(1ull << (sig - 1));
}
_pthread_unlock();
}
static textwindows bool __sig_should_use_altstack(unsigned flags,
struct CosmoTib *tib) {
if (!(flags & SA_ONSTACK)) {
@ -141,53 +155,72 @@ static textwindows sigaction_f __sig_handler(unsigned rva) {
textwindows int __sig_raise(int sig, int sic) {
unsigned rva, flags;
struct CosmoTib *tib = __get_tls();
struct PosixThread *pt = (struct PosixThread *)tib->tib_pthread;
ucontext_t ctx = {.uc_sigmask.__bits[0] = tib->tib_sigmask};
struct PosixThread *pt = _pthread_self();
ucontext_t ctx = {.uc_sigmask = pt->tib->tib_sigmask};
if (!__sig_start(pt, sig, &rva, &flags)) return 0;
siginfo_t si = {.si_signo = sig, .si_code = sic};
struct NtContext nc;
nc.ContextFlags = kNtContextAll;
GetThreadContext(GetCurrentThread(), &nc);
_ntcontext2linux(&ctx, &nc);
pt->tib->tib_sigmask |= __sighandmask[sig];
if (!(flags & SA_NODEFER)) {
tib->tib_sigmask |= 1ull << (sig - 1);
pt->tib->tib_sigmask |= 1ull << (sig - 1);
}
STRACE("entering raise(%G) signal handler %t with mask %s → %s", sig,
__sig_handler(rva), DescribeSigset(0, &ctx.uc_sigmask),
DescribeSigset(0, &(sigset_t){{pt->tib->tib_sigmask}}));
NTTRACE("entering raise(%G) signal handler %t with mask %s → %s", sig,
__sig_handler(rva), DescribeSigset(0, &ctx.uc_sigmask),
DescribeSigset(0, (sigset_t *)&pt->tib->tib_sigmask));
__sig_handler(rva)(sig, &si, &ctx);
STRACE("leaving raise(%G) signal handler %t with mask %s → %s", sig,
__sig_handler(rva),
DescribeSigset(0, &(sigset_t){{pt->tib->tib_sigmask}}),
DescribeSigset(0, &ctx.uc_sigmask));
tib->tib_sigmask = ctx.uc_sigmask.__bits[0];
NTTRACE("leaving raise(%G) signal handler %t with mask %s → %s", sig,
__sig_handler(rva),
DescribeSigset(0, (sigset_t *)&pt->tib->tib_sigmask),
DescribeSigset(0, &ctx.uc_sigmask));
pt->tib->tib_sigmask = ctx.uc_sigmask;
return (flags & SA_RESTART) ? 2 : 1;
}
textwindows void __sig_cancel(struct PosixThread *pt, unsigned flags) {
atomic_int *futex;
if (_weaken(WakeByAddressSingle) &&
(futex = atomic_load_explicit(&pt->pt_futex, memory_order_acquire))) {
STRACE("canceling futex");
_weaken(WakeByAddressSingle)(futex);
} else if (!(flags & SA_RESTART) && pt->iohandle > 0) {
STRACE("canceling i/o operation");
if (!CancelIoEx(pt->iohandle, pt->ioverlap)) {
int err = GetLastError();
if (err != kNtErrorNotFound) {
STRACE("CancelIoEx() failed w/ %d", err);
}
}
} else if (pt->pt_flags & PT_INSEMAPHORE) {
STRACE("canceling semaphore");
pt->pt_flags &= ~PT_INSEMAPHORE;
if (!ReleaseSemaphore(pt->semaphore, 1, 0)) {
STRACE("ReleaseSemaphore() failed w/ %d", GetLastError());
}
} else {
STRACE("canceling asynchronously");
// cancels blocking operations being performed by signaled thread
textwindows void __sig_cancel(struct PosixThread *pt, int sig, unsigned flags) {
bool should_restart;
atomic_int *blocker;
// cancelation points need to set pt_blocker before entering a wait op
blocker = atomic_load_explicit(&pt->pt_blocker, memory_order_acquire);
// cancelation points should set it back to this after blocking
// however, code that longjmps might mess it up a tolerable bit
if (blocker == PT_BLOCKER_CPU) {
STRACE("%G delivered to %d asynchronously", sig, _pthread_tid(pt));
return;
}
// most cancelation points can be safely restarted w/o raising eintr
should_restart = (flags & SA_RESTART) && (pt->pt_flags & PT_RESTARTABLE);
// we can cancel another thread's overlapped i/o op after the freeze
if (blocker == PT_BLOCKER_IO) {
if (should_restart) {
STRACE("%G restarting %d's i/o op", sig, _pthread_tid(pt));
} else {
STRACE("%G interupting %d's i/o op", sig, _pthread_tid(pt));
CancelIoEx(pt->pt_iohandle, pt->pt_ioverlap);
}
return;
}
// threads can create semaphores on an as-needed basis
if (blocker == PT_BLOCKER_SEM) {
if (should_restart) {
STRACE("%G restarting %d's semaphore", sig, _pthread_tid(pt));
} else {
STRACE("%G releasing %d's semaphore", sig, _pthread_tid(pt));
pt->pt_flags &= ~PT_RESTARTABLE;
ReleaseSemaphore(pt->pt_semaphore, 1, 0);
}
return;
}
// all other blocking ops that aren't overlap should use futexes
// we force restartable futexes to churn by waking w/o releasing
STRACE("%G waking %d's futex", sig, _pthread_tid(pt));
if (!should_restart) {
atomic_store_explicit(blocker, 1, memory_order_release);
}
WakeByAddressSingle(blocker);
}
static textwindows wontreturn void __sig_panic(const char *msg) {
@ -207,23 +240,21 @@ static textwindows wontreturn void __sig_tramp(struct SignalFrame *sf) {
ucontext_t ctx = {0};
int sig = sf->si.si_signo;
_ntcontext2linux(&ctx, sf->nc);
ctx.uc_sigmask.__bits[0] =
atomic_load_explicit(&sf->pt->tib->tib_sigmask, memory_order_acquire);
ctx.uc_sigmask = sf->pt->tib->tib_sigmask;
sf->pt->tib->tib_sigmask |= __sighandmask[sig];
if (!(sf->flags & SA_NODEFER)) {
sf->pt->tib->tib_sigmask |= 1ull << (sig - 1);
}
++__sig.count;
STRACE("entering __sig_tramp(%G, %t) with mask %s → %s", sig,
__sig_handler(sf->rva), DescribeSigset(0, &ctx.uc_sigmask),
DescribeSigset(0, &(sigset_t){{sf->pt->tib->tib_sigmask}}));
NTTRACE("entering __sig_tramp(%G, %t) with mask %s → %s", sig,
__sig_handler(sf->rva), DescribeSigset(0, &ctx.uc_sigmask),
DescribeSigset(0, (sigset_t *)&sf->pt->tib->tib_sigmask));
__sig_handler(sf->rva)(sig, &sf->si, &ctx);
STRACE("leaving __sig_tramp(%G, %t) with mask %s → %s", sig,
__sig_handler(sf->rva),
DescribeSigset(0, &(sigset_t){{sf->pt->tib->tib_sigmask}}),
DescribeSigset(0, &ctx.uc_sigmask));
atomic_store_explicit(&sf->pt->tib->tib_sigmask, ctx.uc_sigmask.__bits[0],
memory_order_release);
// TDOO(jart): Do we need to do a __sig_check() here?
NTTRACE("leaving __sig_tramp(%G, %t) with mask %s → %s", sig,
__sig_handler(sf->rva),
DescribeSigset(0, (sigset_t *)&sf->pt->tib->tib_sigmask),
DescribeSigset(0, &ctx.uc_sigmask));
sf->pt->tib->tib_sigmask = ctx.uc_sigmask;
_ntlinux2context(sf->nc, &ctx);
SetThreadContext(GetCurrentThread(), sf->nc);
__sig_panic("SetThreadContext(GetCurrentThread)");
@ -275,18 +306,15 @@ static textwindows int __sig_killer(struct PosixThread *pt, int sig, int sic) {
STRACE("SetThreadContext failed w/ %d", GetLastError());
return ESRCH;
}
ResumeThread(th); // doesn't actually resume if doing blocking i/o
pt->abort_errno = EINTR;
__sig_cancel(pt, flags); // we can wake it up immediately by canceling it
ResumeThread(th);
__sig_cancel(pt, sig, flags);
return 0;
}
textwindows int __sig_kill(struct PosixThread *pt, int sig, int sic) {
int rc;
BLOCK_SIGNALS;
BLOCK_CANCELLATIONS;
rc = __sig_killer(pt, sig, sic);
ALLOW_CANCELLATIONS;
ALLOW_SIGNALS;
return rc;
}
@ -302,22 +330,25 @@ textwindows void __sig_generate(int sig, int sic) {
STRACE("terminating on %G due to no handler", sig);
__sig_terminate(sig);
}
pthread_spin_lock(&_pthread_lock);
BLOCK_SIGNALS;
_pthread_lock();
for (e = dll_first(_pthread_list); e; e = dll_next(_pthread_list, e)) {
pt = POSIXTHREAD_CONTAINER(e);
if (atomic_load_explicit(&pt->status, memory_order_acquire) <
if (atomic_load_explicit(&pt->pt_status, memory_order_acquire) <
kPosixThreadTerminated &&
!(pt->tib->tib_sigmask & (1ull << (sig - 1)))) {
mark = pt;
break;
}
}
pthread_spin_unlock(&_pthread_lock);
_pthread_unlock();
ALLOW_SIGNALS;
if (mark) {
STRACE("generating %G by killing %d", sig, _pthread_tid(mark));
__sig_kill(mark, sig, sic);
} else {
STRACE("all threads block %G so adding to pending signals of process", sig);
__sig.pending |= sig;
__sig.pending |= 1ull << (sig - 1);
}
}
@ -380,8 +411,8 @@ static int __sig_crash_sig(struct NtExceptionPointers *ep, int *code) {
}
}
static void __sig_crasher(struct NtExceptionPointers *ep, int code, int sig,
struct CosmoTib *tib) {
static void __sig_unmaskable(struct NtExceptionPointers *ep, int code, int sig,
struct CosmoTib *tib) {
// increment the signal count for getrusage()
++__sig.count;
@ -436,9 +467,13 @@ static void __sig_crasher(struct NtExceptionPointers *ep, int code, int sig,
ucontext_t ctx = {0};
siginfo_t si = {.si_signo = sig, .si_code = code, .si_addr = si_addr};
_ntcontext2linux(&ctx, ep->ContextRecord);
ctx.uc_sigmask.__bits[0] = tib->tib_sigmask;
ctx.uc_sigmask = tib->tib_sigmask;
tib->tib_sigmask |= __sighandmask[sig];
if (!(flags & SA_NODEFER)) {
tib->tib_sigmask |= 1ull << (sig - 1);
}
__sig_handler(rva)(sig, &si, &ctx);
tib->tib_sigmask = ctx.uc_sigmask.__bits[0];
tib->tib_sigmask = ctx.uc_sigmask;
_ntlinux2context(ep->ContextRecord, &ctx);
}
@ -470,10 +505,10 @@ __msabi dontinstrument unsigned __sig_crash(struct NtExceptionPointers *ep) {
struct CosmoTib *tib = __get_tls();
unsigned flags = __sighandflags[sig];
if (__sig_should_use_altstack(flags, tib)) {
__stack_call(ep, code, sig, tib, __sig_crasher,
__stack_call(ep, code, sig, tib, __sig_unmaskable,
tib->tib_sigstack_addr + tib->tib_sigstack_size);
} else {
__sig_crasher(ep, code, sig, tib);
__sig_unmaskable(ep, code, sig, tib);
}
// resume running user program
@ -517,6 +552,7 @@ static textwindows int __sig_checkem(atomic_ulong *sigs, struct CosmoTib *tib,
for (int sig = 1; sig <= 64; ++sig) {
if ((deliverable & (1ull << (sig - 1))) &&
atomic_fetch_and(sigs, ~(1ull << (sig - 1))) & (1ull << (sig - 1))) {
STRACE("found pending %G we can raise now", sig);
handler_was_called |= __sig_raise(sig, SI_KERNEL);
}
}