libuv 1.45.0, #include cleanup, probably something else.

git-svn-id: https://www.unprompted.com/svn/projects/tildefriends/trunk@4308 ed5197a5-7fde-0310-b194-c3ffbd925b24

deps/libuv/docs/src/design.rst

@@ -60,16 +60,15 @@ stages of a loop iteration:
     :align: center
 
 
-#. The loop concept of 'now' is updated. The event loop caches the current time at the start of
-   the event loop tick in order to reduce the number of time-related system calls.
+#. The loop concept of 'now' is initially set.
+
+#. Due timers are run if the loop was run with ``UV_RUN_DEFAULT``. All active timers scheduled
+   for a time before the loop's concept of *now* get their callbacks called.
 
 #. If the loop is *alive*  an iteration is started, otherwise the loop will exit immediately. So,
    when is a loop considered to be *alive*? If a loop has active and ref'd handles, active
    requests or closing handles it's considered to be *alive*.
 
-#. Due timers are run. All active timers scheduled for a time before the loop's concept of *now*
-   get their callbacks called.
-
 #. Pending callbacks are called. All I/O callbacks are called right after polling for I/O, for the
    most part. There are cases, however, in which calling such a callback is deferred for the next
    loop iteration. If the previous iteration deferred any I/O callback it will be run at this point.
@@ -101,9 +100,11 @@ stages of a loop iteration:
 #. Close callbacks are called. If a handle was closed by calling :c:func:`uv_close` it will
    get the close callback called.
 
-#. Special case in case the loop was run with ``UV_RUN_ONCE``, as it implies forward progress.
-   It's possible that no I/O callbacks were fired after blocking for I/O, but some time has passed
-   so there might be timers which are due, those timers get their callbacks called.
+#. The loop concept of 'now' is updated.
+
+#. Due timers are run. Note that 'now' is not updated again until the next loop iteration.
+   So if a timer became due while other timers were being processed, it won't be run until
+   the following event loop iteration.
 
 #. Iteration ends. If the loop was run with ``UV_RUN_NOWAIT`` or ``UV_RUN_ONCE`` modes the
    iteration ends and :c:func:`uv_run` will return. If the loop was run with ``UV_RUN_DEFAULT``

deps/libuv/docs/src/fs.rst

@@ -12,6 +12,12 @@ otherwise it will be performed asynchronously.
 All file operations are run on the threadpool. See :ref:`threadpool` for information
 on the threadpool size.
 
+Starting with libuv v1.45.0, some file operations on Linux are handed off to
+`io_uring <https://en.wikipedia.org/wiki/Io_uring>` when possible. Apart from
+a (sometimes significant) increase in throughput there should be no change in
+observable behavior. Libuv reverts to using its threadpool when the necessary
+kernel features are unavailable or unsuitable.
+
 .. note::
      On Windows `uv_fs_*` functions use utf-8 encoding.
 
@@ -24,7 +30,8 @@ Data types
 
 .. c:type:: uv_timespec_t
 
-    Portable equivalent of ``struct timespec``.
+    Y2K38-unsafe data type for storing times with nanosecond resolution.
+    Will be replaced with :c:type:`uv_timespec64_t` in libuv v2.0.
 
     ::
 
@@ -160,6 +167,10 @@ Data types
             size_t nentries;
         } uv_dir_t;
 
+.. c:type:: void (*uv_fs_cb)(uv_fs_t* req)
+
+    Callback called when a request is completed asynchronously.
+
 
 Public members
 ^^^^^^^^^^^^^^
@@ -218,7 +229,8 @@ API
 
 .. c:function:: int uv_fs_read(uv_loop_t* loop, uv_fs_t* req, uv_file file, const uv_buf_t bufs[], unsigned int nbufs, int64_t offset, uv_fs_cb cb)
 
-    Equivalent to :man:`preadv(2)`.
+    Equivalent to :man:`preadv(2)`. If the `offset` argument is `-1`, then
+    the current file offset is used and updated.
 
     .. warning::
         On Windows, under non-MSVC environments (e.g. when GCC or Clang is used
@@ -231,7 +243,8 @@ API
 
 .. c:function:: int uv_fs_write(uv_loop_t* loop, uv_fs_t* req, uv_file file, const uv_buf_t bufs[], unsigned int nbufs, int64_t offset, uv_fs_cb cb)
 
-    Equivalent to :man:`pwritev(2)`.
+    Equivalent to :man:`pwritev(2)`. If the `offset` argument is `-1`, then
+    the current file offset is used and updated.
 
     .. warning::
         On Windows, under non-MSVC environments (e.g. when GCC or Clang is used
@@ -463,10 +476,6 @@ API
         The background story and some more details on these issues can be checked
         `here <https://github.com/nodejs/node/issues/7726>`_.
 
-    .. note::
-      This function is not implemented on Windows XP and Windows Server 2003.
-      On these systems, UV_ENOSYS is returned.
-
     .. versionadded:: 1.8.0
 
 .. c:function:: int uv_fs_chown(uv_loop_t* loop, uv_fs_t* req, const char* path, uv_uid_t uid, uv_gid_t gid, uv_fs_cb cb)

deps/libuv/docs/src/guide/networking.rst

@@ -164,7 +164,7 @@ IPv6 stack only
 
 IPv6 sockets can be used for both IPv4 and IPv6 communication. If you want to
 restrict the socket to IPv6 only, pass the ``UV_UDP_IPV6ONLY`` flag to
-``uv_udp_bind`` [#]_.
+``uv_udp_bind``.
 
 Multicast
 ~~~~~~~~~
@@ -250,7 +250,6 @@ times, with each address being reported once.
 ----
 
 .. [#] https://beej.us/guide/bgnet/html/#broadcast-packetshello-world
-.. [#] on Windows only supported on Windows Vista and later.
 .. [#] https://www.tldp.org/HOWTO/Multicast-HOWTO-6.html#ss6.1
 .. [#] libuv use the system ``getaddrinfo`` in the libuv threadpool. libuv
     v0.8.0 and earlier also included c-ares_ as an alternative, but this has been

deps/libuv/docs/src/guide/utilities.rst

@@ -235,7 +235,7 @@ Our downloader is to be invoked as::
 
     $ ./uvwget [url1] [url2] ...
 
-So we add each argument as an URL
+So we add each argument as a URL
 
 .. rubric:: uvwget/main.c - Adding urls
 .. literalinclude:: ../../code/uvwget/main.c

deps/libuv/docs/src/handle.rst

@@ -153,6 +153,9 @@ API
     In-progress requests, like uv_connect_t or uv_write_t, are cancelled and
     have their callbacks called asynchronously with status=UV_ECANCELED.
 
+    `close_cb` can be `NULL` in cases where no cleanup or deallocation is
+    necessary.
+
 .. c:function:: void uv_ref(uv_handle_t* handle)
 
     Reference the given handle. References are idempotent, that is, if a handle

deps/libuv/docs/src/metrics.rst

@@ -4,8 +4,46 @@
 Metrics operations
 ======================
 
-libuv provides a metrics API to track the amount of time the event loop has
-spent idle in the kernel's event provider.
+libuv provides a metrics API to track various internal operations of the event
+loop.
+
+
+Data types
+----------
+
+.. c:type:: uv_metrics_t
+
+    The struct that contains event loop metrics. It is recommended to retrieve
+    these metrics in a :c:type:`uv_prepare_cb` in order to make sure there are
+    no inconsistencies with the metrics counters.
+
+    ::
+
+        typedef struct {
+            uint64_t loop_count;
+            uint64_t events;
+            uint64_t events_waiting;
+            /* private */
+            uint64_t* reserved[13];
+        } uv_metrics_t;
+
+
+Public members
+^^^^^^^^^^^^^^
+
+.. c:member:: uint64_t uv_metrics_t.loop_count
+
+    Number of event loop iterations.
+
+.. c:member:: uint64_t uv_metrics_t.events
+
+    Number of events that have been processed by the event handler.
+
+.. c:member:: uint64_t uv_metrics_t.events_waiting
+
+    Number of events that were waiting to be processed when the event provider
+    was called.
+
 
 API
 ---
@@ -25,3 +63,9 @@ API
         :c:type:`UV_METRICS_IDLE_TIME`.
 
     .. versionadded:: 1.39.0
+
+.. c:function:: int uv_metrics_info(uv_loop_t* loop, uv_metrics_t* metrics)
+
+    Copy the current set of event loop metrics to the ``metrics`` pointer.
+
+    .. versionadded:: 1.45.0

deps/libuv/docs/src/misc.rst

@@ -73,7 +73,8 @@ Data types
 
 .. c:type:: uv_timeval_t
 
-    Data type for storing times.
+    Y2K38-unsafe data type for storing times with microsecond resolution.
+    Will be replaced with :c:type:`uv_timeval64_t` in libuv v2.0.
 
     ::
 
@@ -84,7 +85,7 @@ Data types
 
 .. c:type:: uv_timeval64_t
 
-    Alternative data type for storing times.
+    Y2K38-safe data type for storing times with microsecond resolution.
 
     ::
 
@@ -93,6 +94,28 @@ Data types
             int32_t tv_usec;
         } uv_timeval64_t;
 
+.. c:type:: uv_timespec64_t
+
+    Y2K38-safe data type for storing times with nanosecond resolution.
+
+    ::
+
+        typedef struct {
+            int64_t tv_sec;
+            int32_t tv_nsec;
+        } uv_timespec64_t;
+
+.. c:enum:: uv_clock_id
+
+    Clock source for :c:func:`uv_clock_gettime`.
+
+    ::
+
+        typedef enum {
+          UV_CLOCK_MONOTONIC,
+          UV_CLOCK_REALTIME
+        } uv_clock_id;
+
 .. c:type:: uv_rusage_t
 
     Data type for resource usage results.
@@ -119,7 +142,10 @@ Data types
         } uv_rusage_t;
 
     Members marked with `(X)` are unsupported on Windows.
-    See :man:`getrusage(2)` for supported fields on Unix
+    See :man:`getrusage(2)` for supported fields on UNIX-like platforms.
+
+    The maximum resident set size is reported in kilobytes, the unit most
+    platforms use natively.
 
 .. c:type:: uv_cpu_info_t
 
@@ -211,7 +237,7 @@ API
     type of the stdio streams.
 
     For :man:`isatty(3)` equivalent functionality use this function and test
-    for ``UV_TTY``.
+    for `UV_TTY`.
 
 .. c:function:: int uv_replace_allocator(uv_malloc_func malloc_func, uv_realloc_func realloc_func, uv_calloc_func calloc_func, uv_free_func free_func)
 
@@ -225,8 +251,8 @@ API
     after all resources have been freed and thus libuv doesn't reference
     any allocated memory chunk.
 
-    On success, it returns 0, if any of the function pointers is NULL it
-    returns UV_EINVAL.
+    On success, it returns 0, if any of the function pointers is `NULL` it
+    returns `UV_EINVAL`.
 
     .. warning:: There is no protection against changing the allocator multiple
                  times. If the user changes it they are responsible for making
@@ -362,6 +388,13 @@ API
 
     Frees the `cpu_infos` array previously allocated with :c:func:`uv_cpu_info`.
 
+.. c:function:: int uv_cpumask_size(void)
+
+    Returns the maximum size of the mask used for process/thread affinities,
+    or `UV_ENOTSUP` if affinities are not supported on the current platform.
+
+    .. versionadded:: 1.45.0
+
 .. c:function:: int uv_interface_addresses(uv_interface_address_t** addresses, int* count)
 
     Gets address information about the network interfaces on the system. An
@@ -541,18 +574,21 @@ API
 
 .. c:function:: uint64_t uv_get_free_memory(void)
 
-    Gets the amount of free memory available in the system, as reported by the kernel (in bytes).
+    Gets the amount of free memory available in the system, as reported by
+    the kernel (in bytes). Returns 0 when unknown.
 
 .. c:function:: uint64_t uv_get_total_memory(void)
 
     Gets the total amount of physical memory in the system (in bytes).
+    Returns 0 when unknown.
 
 .. c:function:: uint64_t uv_get_constrained_memory(void)
 
-    Gets the amount of memory available to the process (in bytes) based on
+    Gets the total amount of memory available to the process (in bytes) based on
     limits imposed by the OS. If there is no such constraint, or the constraint
-    is unknown, `0` is returned. Note that it is not unusual for this value to
-    be less than or greater than :c:func:`uv_get_total_memory`.
+    is unknown, `0` is returned. If there is a constraining mechanism, but there
+    is no constraint set, `UINT64_MAX` is returned. Note that it is not unusual
+    for this value to be less than or greater than :c:func:`uv_get_total_memory`.
 
     .. note::
         This function currently only returns a non-zero value on Linux, based
@@ -560,9 +596,23 @@ API
 
     .. versionadded:: 1.29.0
 
+.. c:function:: uint64_t uv_get_available_memory(void)
+
+    Gets the amount of free memory that is still available to the process (in bytes).
+    This differs from :c:func:`uv_get_free_memory` in that it takes into account any
+    limits imposed by the OS. If there is no such constraint, or the constraint
+    is unknown, the amount returned will be identical to :c:func:`uv_get_free_memory`.
+
+    .. note::
+        This function currently only returns a value that is different from
+        what :c:func:`uv_get_free_memory` reports on Linux, based
+        on cgroups if it is present.
+
+    .. versionadded:: 1.45.0
+
 .. c:function:: uint64_t uv_hrtime(void)
 
-    Returns the current high-resolution real time. This is expressed in
+    Returns the current high-resolution timestamp. This is expressed in
     nanoseconds. It is relative to an arbitrary time in the past. It is not
     related to the time of day and therefore not subject to clock drift. The
     primary use is for measuring performance between intervals.
@@ -571,6 +621,19 @@ API
         Not every platform can support nanosecond resolution; however, this value will always
         be in nanoseconds.
 
+.. c:function:: int uv_clock_gettime(uv_clock_id clock_id, uv_timespec64_t* ts)
+
+    Obtain the current system time from a high-resolution real-time or monotonic
+    clock source.
+
+    The real-time clock counts from the UNIX epoch (1970-01-01) and is subject
+    to time adjustments; it can jump back in time.
+
+    The monotonic clock counts from an arbitrary point in the past and never
+    jumps back in time.
+
+    .. versionadded:: 1.45.0
+
 .. c:function:: void uv_print_all_handles(uv_loop_t* loop, FILE* stream)
 
     Prints all handles associated with the given `loop` to the given `stream`.

deps/libuv/docs/src/poll.rst

@@ -101,7 +101,9 @@ API
     with one of the `UV_E*` error codes (see :ref:`errors`). The user should
     not close the socket while the handle is active. If the user does that
     anyway, the callback *may* be called reporting an error status, but this is
-    **not** guaranteed.
+    **not** guaranteed. If `status == UV_EBADF` polling is discontinued for the
+    file handle and no further events will be reported. The user should
+    then call :c:func:`uv_close` on the handle.
 
     .. note::
         Calling :c:func:`uv_poll_start` on a handle that is already active is

deps/libuv/docs/src/threading.rst

@@ -88,6 +88,46 @@ Threads
 
     .. versionadded:: 1.26.0
 
+.. c:function:: int uv_thread_setaffinity(uv_thread_t* tid, char* cpumask, char* oldmask, size_t mask_size)
+
+    Sets the specified thread's affinity to cpumask, which is specified in
+    bytes. Optionally returning the previous affinity setting in oldmask.
+    On Unix, uses :man:`pthread_getaffinity_np(3)` to get the affinity setting
+    and maps the cpu_set_t to bytes in oldmask. Then maps the bytes in cpumask
+    to a cpu_set_t and uses :man:`pthread_setaffinity_np(3)`. On Windows, maps
+    the bytes in cpumask to a bitmask and uses SetThreadAffinityMask() which
+    returns the previous affinity setting.
+
+    The mask_size specifies the number of entries (bytes) in cpumask / oldmask,
+    and must be greater-than-or-equal-to :c:func:`uv_cpumask_size`.
+
+    .. note::
+        Thread affinity setting is not atomic on Windows. Unsupported on macOS.
+
+    .. versionadded:: 1.45.0
+
+.. c:function:: int uv_thread_getaffinity(uv_thread_t* tid, char* cpumask, size_t mask_size)
+
+    Gets the specified thread's affinity setting. On Unix, this maps the
+    cpu_set_t returned by :man:`pthread_getaffinity_np(3)` to bytes in cpumask.
+
+    The mask_size specifies the number of entries (bytes) in cpumask,
+    and must be greater-than-or-equal-to :c:func:`uv_cpumask_size`.
+
+    .. note::
+        Thread affinity getting is not atomic on Windows. Unsupported on macOS.
+
+    .. versionadded:: 1.45.0
+
+.. c:function:: int uv_thread_getcpu(void)
+
+    Gets the CPU number on which the calling thread is running.
+
+    .. note::
+        Currently only implemented on Windows, Linux and FreeBSD.
+
+    .. versionadded:: 1.45.0
+
 .. c:function:: uv_thread_t uv_thread_self(void)
 .. c:function:: int uv_thread_join(uv_thread_t *tid)
 .. c:function:: int uv_thread_equal(const uv_thread_t* t1, const uv_thread_t* t2)

deps/libuv/docs/src/threadpool.rst

@@ -14,6 +14,9 @@ is 1024).
 
 .. versionchanged:: 1.30.0 the maximum UV_THREADPOOL_SIZE allowed was increased from 128 to 1024.
 
+.. versionchanged:: 1.45.0 threads now have an 8 MB stack instead of the
+   (sometimes too low) platform default.
+
 The threadpool is global and shared across all event loops. When a particular
 function makes use of the threadpool (i.e. when using :c:func:`uv_queue_work`)
 libuv preallocates and initializes the maximum number of threads allowed by

deps/libuv/docs/src/udp.rst

@@ -56,7 +56,7 @@ Data types
             /*
              * Indicates if IP_RECVERR/IPV6_RECVERR will be set when binding the handle.
              * This sets IP_RECVERR for IPv4 and IPV6_RECVERR for IPv6 UDP sockets on
-             * Linux. This stops the Linux kernel from supressing some ICMP error messages
+             * Linux. This stops the Linux kernel from suppressing some ICMP error messages
              * and enables full ICMP error reporting for faster failover.
              * This flag is no-op on platforms other than Linux.
              */