Improve threading and i/o routines

- On Windows connect() can now be interrupted by a signal; connect() w/
  O_NONBLOCK will now raise EINPROGRESS; and connect() with SO_SNDTIMEO
  will raise ETIMEDOUT after the interval has elapsed.

- We now get the AcceptEx(), ConnectEx(), and TransmitFile() functions
  from the WIN32 API the officially blessed way, using WSAIoctl().

- Do nothing on Windows when fsync() is called on a directory handle.
  This was raising EACCES earlier becaues GENERIC_WRITE is required on
  the handle. It's possible to FlushFileBuffers() a directory handle if
  it's opened with write access but MSDN doesn't document what it does.
  If you have any idea, please let us know!

- Prefer manual reset event objects for read() and write() on Windows.

- Do some code cleanup on our dlmalloc customizations.

- Fix errno type error in Windows blocking routines.

- Make the futex polyfill simpler and faster.

Makefile

@@ -322,7 +322,6 @@ COSMOPOLITAN_OBJECTS =			\
 	LIBC_SOCK			\
 	LIBC_NT_WS2_32			\
 	LIBC_NT_IPHLPAPI		\
-	LIBC_NT_MSWSOCK			\
 	LIBC_X				\
 	THIRD_PARTY_GETOPT		\
 	LIBC_LOG			\

libc/calls/fdatasync-nt.c

@@ -18,15 +18,24 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/calls/internal.h"
 #include "libc/calls/syscall_support-nt.internal.h"
+#include "libc/nt/enum/fileflagandattributes.h"
 #include "libc/nt/enum/filetype.h"
 #include "libc/nt/files.h"
+#include "libc/nt/struct/byhandlefileinformation.h"
 #include "libc/sysv/errfuns.h"
 #ifdef __x86_64__
 
 textwindows int sys_fdatasync_nt(int fd, bool fake) {
+  struct NtByHandleFileInformation wst;
   if (!__isfdopen(fd)) return ebadf();
   if (!__isfdkind(fd, kFdFile)) return einval();
   if (GetFileType(g_fds.p[fd].handle) != kNtFileTypeDisk) return einval();
+  if (!GetFileInformationByHandle(g_fds.p[fd].handle, &wst)) return __winerr();
+  if (wst.dwFileAttributes & kNtFileAttributeDirectory) {
+    // Flushing a directory handle is possible, but it needs
+    // kNtGenericWrite access, and MSDN doesn't document it.
+    return 0;
+  }
   if (_check_cancel() == -1) return -1;
   if (_check_signal(false) == -1) return -1;
   if (fake) return 0;

libc/calls/readwrite-nt.c

@@ -33,6 +33,7 @@
 #include "libc/nt/synchronization.h"
 #include "libc/nt/thread.h"
 #include "libc/stdio/sysparam.h"
+#include "libc/str/str.h"
 #include "libc/sysv/consts/o.h"
 #include "libc/sysv/errfuns.h"
 #include "libc/thread/posixthread.internal.h"
@@ -67,10 +68,10 @@ sys_readwrite_nt(int fd, void *data, size_t size, ssize_t offset,
   bool32 ok;
   uint64_t m;
   uint32_t exchanged;
+  int olderror = errno;
   bool eagained = false;
   bool eintered = false;
   bool canceled = false;
-  bool olderror = errno;
   struct PosixThread *pt;
   struct Fd *f = g_fds.p + fd;
 
@@ -124,7 +125,7 @@ sys_readwrite_nt(int fd, void *data, size_t size, ssize_t offset,
   // can only be returned by a single system call in a thread's life
   // another thing we do is check if any pending signals exist, then
   // running as many of them as possible before entering a wait call
-  struct NtOverlapped overlap = {.hEvent = CreateEvent(0, 0, 0, 0),
+  struct NtOverlapped overlap = {.hEvent = CreateEvent(0, 1, 0, 0),
                                  .Pointer = offset};
   struct ReadwriteResources rwc = {handle, &overlap};
   pthread_cleanup_push(UnwindReadwrite, &rwc);

libc/calls/releasefd.c

@@ -25,6 +25,7 @@
 void __releasefd(int fd) {
   int f1, f2;
   if (!(0 <= fd && fd < g_fds.n)) return;
+  g_fds.p[fd].kind = kFdEmpty;
   bzero(g_fds.p + fd, sizeof(*g_fds.p));
   f1 = atomic_load_explicit(&g_fds.f, memory_order_relaxed);
   do {

libc/calls/struct/fd.internal.h

@@ -14,15 +14,17 @@ COSMOPOLITAN_C_START_
 
 struct Fd {
   char kind;
+  bool isbound;
   unsigned flags;
   unsigned mode;
-  int64_t handle;
-  int64_t pointer;
+  long handle;
+  long pointer;
   int family;
   int type;
   int protocol;
-  uint32_t rcvtimeo;
-  uint32_t sndtimeo;
+  unsigned rcvtimeo; /* millis; 0 means wait forever */
+  unsigned sndtimeo; /* millis; 0 means wait forever */
+  void *connect_op;
 };
 
 struct Fds {

libc/intrin/pthread_yield.c

@@ -18,6 +18,7 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/calls/calls.h"
 #include "libc/dce.h"
+#include "libc/nexgen32e/yield.h"
 #include "libc/runtime/runtime.h"
 #include "libc/runtime/syslib.internal.h"
 #include "libc/thread/thread.h"
@@ -30,6 +31,8 @@
 int pthread_yield(void) {
   if (IsXnuSilicon()) {
     __syslib->__pthread_yield_np();
+  } else if (IsOpenbsd()) {
+    spin_yield();  // sched_yield() is punishingly slow on OpenBSD
   } else {
     sched_yield();
   }

libc/intrin/ulock.c

@@ -31,7 +31,7 @@
 int sys_ulock_wait(uint32_t operation, void *addr, uint64_t value,
                    uint32_t timeout_micros) asm("sys_futex_cp");
 
-// returns -1 w/ errno
+// returns number of other waiters, or -1 w/ errno
 int ulock_wait(uint32_t operation, void *addr, uint64_t value,
                uint32_t timeout_micros) {
   int rc;

libc/nt/iocp.h

@@ -39,13 +39,13 @@ typedef void (*NtOverlappedCompletionRoutine)(
 
 int64_t CreateIoCompletionPort(int64_t FileHandleOrNeg1,
                                int64_t opt_ExistingCompletionPortOrZero,
-                               void *StatePointer,
+                               uint64_t CompletionKey,
                                uint32_t NumberOfConcurrentThreads);
 
 bool32 GetQueuedCompletionStatus(int64_t CompletionPort,
                                  uint32_t *lpNumberOfBytesTransferred,
-                                 void *StatePointerPointer,
-                                 struct NtOverlapped **lpOverlapped,
+                                 uint64_t *out_lpCompletionKey,
+                                 struct NtOverlapped **out_lpOverlapped,
                                  uint32_t dwMilliseconds);
 
 bool32 GetQueuedCompletionStatusEx(
@@ -56,7 +56,7 @@ bool32 GetQueuedCompletionStatusEx(
 
 bool32 PostQueuedCompletionStatus(int64_t CompletionPort,
                                   uint32_t dwNumberOfBytesTransferred,
-                                  uint32_t *dwCompletionKey,
+                                  uint64_t dwCompletionKey,
                                   struct NtOverlapped *opt_lpOverlapped);
 
 bool32 SetFileCompletionNotificationModes(int64_t FileHandle,

libc/nt/master.sh

@@ -483,15 +483,6 @@ imp	'GetSaveFileName'					GetSaveFileNameW					comdlg32	1
 imp	'PrintDlg'						PrintDlgW						comdlg32	1
 imp	'ReplaceText'						ReplaceTextW						comdlg32	1
 
-# MSWSOCK.DLL
-#
-#	Name							Actual							DLL		Arity
-imp	'AcceptEx'						AcceptEx						MsWSock		8
-imp	'DisconnectEx'						DisconnectEx						MsWSock		4
-imp	'GetAcceptExSockaddrs'					GetAcceptExSockaddrs					MsWSock		8
-imp	'TransmitFile'						TransmitFile						MsWSock		7
-imp	'WSARecvEx'						WSARecvEx						MsWSock		4
-
 # WS2_32.DLL
 #
 #	Name							Actual							DLL		Arity
@@ -516,6 +507,7 @@ imp	''							setsockopt						ws2_32		5
 imp	''							shutdown						ws2_32		2
 imp	''							socket							ws2_32		3
 imp	''							socket							ws2_32		3
+imp	''							socket							ws2_32		3
 imp	'FreeAddrInfo'						FreeAddrInfoW						ws2_32		1
 imp	'FreeAddrInfoEx'					FreeAddrInfoExW						ws2_32		1
 imp	'GetAddrInfo'						GetAddrInfoW						ws2_32		4

libc/nt/nt.mk

@@ -221,27 +221,6 @@ $(LIBC_NT_IPHLPAPI_A).pkg:				\
 
 #───────────────────────────────────────────────────────────────────────────────
 
-LIBC_NT_ARTIFACTS += LIBC_NT_MSWSOCK_A
-LIBC_NT_MSWSOCK = $(LIBC_NT_MSWSOCK_A_DEPS) $(LIBC_NT_MSWSOCK_A)
-LIBC_NT_MSWSOCK_A = o/$(MODE)/libc/nt/MsWSock.a
-LIBC_NT_MSWSOCK_A_SRCS := $(wildcard libc/nt/MsWSock/*.S)
-LIBC_NT_MSWSOCK_A_OBJS = $(LIBC_NT_MSWSOCK_A_SRCS:%.S=o/$(MODE)/%.o)
-LIBC_NT_MSWSOCK_A_CHECKS = $(LIBC_NT_MSWSOCK_A).pkg
-LIBC_NT_MSWSOCK_A_DIRECTDEPS = LIBC_NT_KERNEL32
-LIBC_NT_MSWSOCK_A_DEPS :=				\
-	$(call uniq,$(foreach x,$(LIBC_NT_MSWSOCK_A_DIRECTDEPS),$($(x))))
-
-$(LIBC_NT_MSWSOCK_A):					\
-		libc/nt/MsWSock/			\
-		$(LIBC_NT_MSWSOCK_A).pkg		\
-		$(LIBC_NT_MSWSOCK_A_OBJS)
-
-$(LIBC_NT_MSWSOCK_A).pkg:				\
-		$(LIBC_NT_MSWSOCK_A_OBJS)		\
-		$(foreach x,$(LIBC_NT_MSWSOCK_A_DIRECTDEPS),$($(x)_A).pkg)
-
-#───────────────────────────────────────────────────────────────────────────────
-
 LIBC_NT_ARTIFACTS += LIBC_NT_IPHLPAPI_A
 LIBC_NT_IPHLPAPI = $(LIBC_NT_IPHLPAPI_A_DEPS) $(LIBC_NT_IPHLPAPI_A)
 LIBC_NT_IPHLPAPI_A = o/$(MODE)/libc/nt/iphlpapi.a

libc/nt/struct/overlappedentry.h

@@ -5,7 +5,7 @@
 COSMOPOLITAN_C_START_
 
 struct NtOverlappedEntry {
-  uint32_t *lpCompletionKey;
+  uint64_t lpCompletionKey;
   struct NtOverlapped *lpOverlapped;
   uint32_t *Internal;
   uint32_t dwNumberOfBytesTransferred;

libc/nt/winsock.h

@@ -297,6 +297,7 @@ int WSACleanup(void);
 int WSAGetLastError(void) nosideeffect;
 void WSASetLastError(int);
 
+int64_t __sys_socket_nt(int, int, int);
 int __sys_bind_nt(uint64_t, const void *, int);
 int __sys_closesocket_nt(uint64_t);
 int __sys_getpeername_nt(uint64_t, void *, uint32_t *);
@@ -342,12 +343,6 @@ int64_t WSAAccept(uint64_t s, struct sockaddr *out_addr,
                   const NtConditionProc opt_lpfnCondition,
                   const uint32_t *opt_dwCallbackData) paramsnonnull((2)) __wur;
 
-bool32 AcceptEx(int64_t sListenSocket, int64_t sAcceptSocket,
-                void *out_lpOutputBuffer /*[recvlen+local+remoteaddrlen]*/,
-                uint32_t dwReceiveDataLength, uint32_t dwLocalAddressLength,
-                uint32_t dwRemoteAddressLength, uint32_t *out_lpdwBytesReceived,
-                struct NtOverlapped *inout_lpOverlapped);
-
 int WSASend(uint64_t s, const struct NtIovec *lpBuffers, uint32_t dwBufferCount,
             uint32_t *opt_out_lpNumberOfBytesSent, uint32_t dwFlags,
             struct NtOverlapped *opt_inout_lpOverlapped,
@@ -494,19 +489,6 @@ int /* success==0 */ WSAGetServiceClassNameByClassId(
     const struct NtGuid *lpServiceClassId, char16_t *out_lpszServiceClassName,
     uint32_t *inout_lpdwBufferLength) paramsnonnull();
 
-bool32 TransmitFile(int64_t hSocket, int64_t hFile,
-                    uint32_t opt_nNumberOfBytesToWrite,
-                    uint32_t opt_nNumberOfBytesPerSend,
-                    struct NtOverlapped *opt_inout_lpOverlapped,
-                    const struct NtTransmitFileBuffers *opt_lpTransmitBuffers,
-                    uint32_t dwReserved);
-
-bool32 AcceptEx(int64_t sListenSocket, int64_t sAcceptSocket,
-                void *out_lpOutputBuffer /*[recvlen+local+remoteaddrlen]*/,
-                uint32_t dwReceiveDataLength, uint32_t dwLocalAddressLength,
-                uint32_t dwRemoteAddressLength, uint32_t *out_lpdwBytesReceived,
-                struct NtOverlapped *inout_lpOverlapped);
-
 void GetAcceptExSockaddrs(
     const void *lpOutputBuffer /*[recvsize+addrsize+addrlen]*/,
     uint32_t dwReceiveDataLength, uint32_t dwLocalAddressLength,

libc/runtime/clone.c

@@ -30,7 +30,6 @@
 #include "libc/intrin/asan.internal.h"
 #include "libc/intrin/atomic.h"
 #include "libc/intrin/describeflags.internal.h"
-#include "libc/intrin/strace.internal.h"
 #include "libc/intrin/ulock.h"
 #include "libc/intrin/weaken.h"
 #include "libc/limits.h"
@@ -667,7 +666,6 @@ errno_t clone(void *func, void *stk, size_t stksz, int flags, void *arg,
                         CLONE_CHILD_SETTID | CLONE_CHILD_CLEARTID)) !=
                  (CLONE_THREAD | CLONE_VM | CLONE_FS | CLONE_FILES |
                   CLONE_SIGHAND | CLONE_SYSVSEM)) {
-    STRACE("cosmo clone() is picky about flags, see clone.c");
     rc = EINVAL;
   } else if (IsXnu()) {
 #ifdef __x86_64__
@@ -695,8 +693,5 @@ errno_t clone(void *func, void *stk, size_t stksz, int flags, void *arg,
     rc = EAGAIN;
   }
 
-  STRACE("clone(%t, %p, %'zu, %#x, %p, %p, %p, %p) → %s", func, stk, stksz,
-         flags, arg, ptid, tls, ctid, DescribeErrno(rc));
-
   return rc;
 }