Thread-safety improvements. Add tests with multiple threads. CI improvements.

.github/workflows/ci.yml

@@ -17,21 +17,25 @@ concurrency:
 
 jobs:
   test:
-    name: Julia ${{ matrix.version }} - ${{ matrix.os }} - ${{ matrix.arch }} - ${{ github.event_name }}
+    name: julia -t${{ matrix.threads}} - ${{ matrix.version }} - ${{ matrix.os }} - ${{ matrix.arch }} - ${{ github.event_name }}
     runs-on: ${{ matrix.os }}
+    timeout-minutes: 30
     strategy:
       fail-fast: false
       matrix:
-        version:
-          - 'nightly'
-        os:
-          - ubuntu-latest
-          - macOS-latest
-          - windows-latest
-        arch:
-          - x64
-          - x86
+        threads:
+          # - '1'
+          - '4,4'
+        version: [nightly]
+        os: [ubuntu-latest, windows-latest, macOS-latest]
+        arch: [x64, x86, aarch64]
         exclude:
+          - os: ubuntu-latest
+            arch: aarch64
+          - os: windows-latest
+            arch: aarch64
+          - os: macOS-latest
+            arch: x64
           - os: macOS-latest
             arch: x86
     steps:
@@ -44,6 +48,7 @@ jobs:
       - uses: julia-actions/julia-runtest@v1
         env:
           JULIA_DISTRIBUTED_TESTING_STANDALONE: 1
+          JULIA_NUM_THREADS: '${{ matrix.threads}}'
       - uses: julia-actions/julia-processcoverage@v1
       - uses: codecov/codecov-action@v5
         with:

src/cluster.jl

@@ -151,7 +151,7 @@
 end
 
 function check_worker_state(w::Worker)
-    if w.state === W_CREATED
+    if (@atomic w.state) === W_CREATED
         if !isclusterlazy()
             if PGRP.topology === :all_to_all
                 # Since higher pids connect with lower pids, the remote worker
@@ -163,10 +163,10 @@
         else
             w.ct_time = time()
             if myid() > w.id
-                t = Threads.@spawn Threads.threadpool() exec_conn_func(w)
+                t = @async exec_conn_func(w)
             else
                 # route request via node 1
-                t = Threads.@spawn Threads.threadpool() remotecall_fetch((p,to_id) -> remotecall_fetch(exec_conn_func, p, to_id), 1, w.id, myid())
+                t = @async remotecall_fetch((p,to_id) -> remotecall_fetch(exec_conn_func, p, to_id), 1, w.id, myid())
             end
             errormonitor(t)
             wait_for_conn(w)
@@ -190,20 +190,14 @@
 end
 
 function wait_for_conn(w)
-    if w.state === W_CREATED
+    if (@atomic w.state) === W_CREATED
         timeout =  worker_timeout() - (time() - w.ct_time)
         timeout <= 0 && error("peer $(w.id) has not connected to $(myid())")
 
-        T = Threads.@spawn Threads.threadpool() begin
-            sleep($timeout)
-            lock(w.c_state) do
-                notify(w.c_state; all=true)
-            end
-        end
-        errormonitor(T)
-        lock(w.c_state) do
-            wait(w.c_state)
-            w.state === W_CREATED && error("peer $(w.id) didn't connect to $(myid()) within $timeout seconds")
+        if timedwait(() -> (@atomic w.state) === W_CONNECTED, timeout) === :timed_out
+            # Notify any waiters on the state and throw
+            @lock w.c_state notify(w.c_state)
+            error("peer $(w.id) didn't connect to $(myid()) within $timeout seconds")
         end
     end
     nothing
@@ -258,7 +252,7 @@
     else
         sock = listen(interface, LPROC.bind_port)
     end
-    errormonitor(Threads.@spawn while isopen(sock)
+    errormonitor(@async while isopen(sock)
         client = accept(sock)
         process_messages(client, client, true)
     end)
@@ -290,7 +284,7 @@
 
 
 function redirect_worker_output(ident, stream)
-    t = Threads.@spawn while !eof(stream)
+    t = @async while !eof(stream)
         line = readline(stream)
         if startswith(line, "      From worker ")
             # stdout's of "additional" workers started from an initial worker on a host are not available
@@ -329,7 +323,7 @@
     leader = String[]
     try
         while ntries > 0
-            readtask = Threads.@spawn Threads.threadpool() readline(io)
+            readtask = @async readline(io)
             yield()
             while !istaskdone(readtask) && ((time_ns() - t0) < timeout)
                 sleep(0.05)
@@ -430,7 +424,7 @@
 
 ```julia
 # On busy clusters, call `addprocs` asynchronously
-t = Threads.@spawn addprocs(...)
+t = @async addprocs(...)
 ```
 
 ```julia
@@ -496,13 +490,14 @@
     # call manager's `launch` is a separate task. This allows the master
     # process initiate the connection setup process as and when workers come
     # online
-    t_launch = Threads.@spawn Threads.threadpool() launch(manager, params, launched, launch_ntfy)
+    # NOTE: Must be `@async`. See FIXME above
+    t_launch = @async launch(manager, params, launched, launch_ntfy)
 
     @sync begin
         while true
             if isempty(launched)
                 istaskdone(t_launch) && break
-                Threads.@spawn Threads.threadpool() begin
+                @async begin # NOTE: Must be `@async`. See FIXME above
                     sleep(1)
                     notify(launch_ntfy)
                 end
@@ -512,7 +507,8 @@
             if !isempty(launched)
                 wconfig = popfirst!(launched)
                 let wconfig=wconfig
-                    Threads.@spawn Threads.threadpool() setup_launched_worker(manager, wconfig, launched_q)
+                    # NOTE: Must be `@async`. See FIXME above
+                    @async setup_launched_worker(manager, wconfig, launched_q)
                 end
             end
         end
@@ -592,7 +588,7 @@
             wconfig.port = port
 
             let wconfig=wconfig
-                Threads.@spawn Threads.threadpool() begin
+                @async begin
                     pid = create_worker(manager, wconfig)
                     remote_do(redirect_output_from_additional_worker, frompid, pid, port)
                     push!(launched_q, pid)
@@ -660,7 +656,7 @@
         for jw in PGRP.workers
             if (jw.id != 1) && (jw.id < w.id)
                 # wait for wl to join
-                if jw.state === W_CREATED
+                if (@atomic jw.state) === W_CREATED
                     lock(jw.c_state) do
                         wait(jw.c_state)
                     end
@@ -688,7 +684,7 @@
 
         for wl in wlist
             lock(wl.c_state) do
-                if wl.state === W_CREATED
+                if (@atomic wl.state) === W_CREATED
                     # wait for wl to join
                     wait(wl.c_state)
                 end
@@ -758,7 +754,7 @@
     end
 
     errormonitor(
-        Threads.@spawn begin
+        @async begin
             timeout = worker_timeout()
             if timedwait(() -> !haskey(map_pid_wrkr, 1), timeout) === :timed_out
                 print(stderr, "Master process (id 1) could not connect within $(timeout) seconds.\nexiting.\n")
@@ -890,7 +886,7 @@
         n = length(PGRP.workers)
         # filter out workers in the process of being setup/shutdown.
         for jw in PGRP.workers
-            if !isa(jw, LocalProcess) && (jw.state !== W_CONNECTED)
+            if !isa(jw, LocalProcess) && ((@atomic jw.state) !== W_CONNECTED)
                 n = n - 1
             end
         end
@@ -941,7 +937,7 @@
 function procs()
     if myid() == 1 || (PGRP.topology === :all_to_all  && !isclusterlazy())
         # filter out workers in the process of being setup/shutdown.
-        return Int[x.id for x in PGRP.workers if isa(x, LocalProcess) || (x.state === W_CONNECTED)]
+        return Int[x.id for x in PGRP.workers if isa(x, LocalProcess) || ((@atomic x.state) === W_CONNECTED)]
     else
         return Int[x.id for x in PGRP.workers]
     end
@@ -950,7 +946,7 @@
 function id_in_procs(id)  # faster version of `id in procs()`
     if myid() == 1 || (PGRP.topology === :all_to_all  && !isclusterlazy())
         for x in PGRP.workers
-            if (x.id::Int) == id && (isa(x, LocalProcess) || (x::Worker).state === W_CONNECTED)
+            if (x.id::Int) == id && (isa(x, LocalProcess) || (@atomic (x::Worker).state) === W_CONNECTED)
                 return true
             end
         end
@@ -972,7 +968,7 @@
 """
 function procs(pid::Integer)
     if myid() == 1
-        all_workers = [x for x in PGRP.workers if isa(x, LocalProcess) || (x.state === W_CONNECTED)]
+        all_workers = [x for x in PGRP.workers if isa(x, LocalProcess) || ((@atomic x.state) === W_CONNECTED)]
         if (pid == 1) || (isa(map_pid_wrkr[pid].manager, LocalManager))
             Int[x.id for x in filter(w -> (w.id==1) || (isa(w.manager, LocalManager)), all_workers)]
         else
@@ -1050,13 +1046,13 @@
 
     pids = vcat(pids...)
     if waitfor == 0
-        t = Threads.@spawn Threads.threadpool() _rmprocs(pids, typemax(Int))
+        t = @async _rmprocs(pids, typemax(Int))
         yield()
         return t
     else
         _rmprocs(pids, waitfor)
         # return a dummy task object that user code can wait on.
-        return Threads.@spawn Threads.threadpool() nothing
+        return @async nothing
     end
 end
 
@@ -1079,11 +1075,11 @@
 
         start = time_ns()
         while (time_ns() - start) < waitfor*1e9
-            all(w -> w.state === W_TERMINATED, rmprocset) && break
+            all(w -> (@atomic w.state) === W_TERMINATED, rmprocset) && break
             sleep(min(0.1, waitfor - (time_ns() - start)/1e9))
         end
 
-        unremoved = [wrkr.id for wrkr in filter(w -> w.state !== W_TERMINATED, rmprocset)]
+        unremoved = [wrkr.id for wrkr in filter(w -> (@atomic w.state) !== W_TERMINATED, rmprocset)]
         if length(unremoved) > 0
             estr = string("rmprocs: pids ", unremoved, " not terminated after ", waitfor, " seconds.")
             throw(ErrorException(estr))
@@ -1239,7 +1235,7 @@
     @assert myid() == 1
     @sync begin
         for pid in pids
-            Threads.@spawn Threads.threadpool() interrupt(pid)
+            @async interrupt(pid)
         end
     end
 end

src/macros.jl

@@ -230,7 +230,7 @@ function remotecall_eval(m::Module, procs, ex)
         # execute locally last as we do not want local execution to block serialization
         # of the request to remote nodes.
         for _ in 1:run_locally
-            Threads.@spawn Threads.threadpool() Core.eval(m, ex)
+            @async Core.eval(m, ex)
         end
     end
     nothing
@@ -275,7 +275,7 @@ function preduce(reducer, f, R)
 end
 
 function pfor(f, R)
-    t = Threads.@spawn Threads.threadpool() @sync for c in splitrange(Int(firstindex(R)), Int(lastindex(R)), nworkers())
+    t = @async @sync for c in splitrange(Int(firstindex(R)), Int(lastindex(R)), nworkers())
         @spawnat :any f(R, first(c), last(c))
     end
     errormonitor(t)

src/managers.jl

@@ -113,7 +113,7 @@
 
 * `exeflags`: additional flags passed to the worker processes. It can either be a `Cmd`, a `String`
   holding one flag, or a collection of strings, with one element per flag.
-  E.g. `\`--threads=auto project=.\``, `"--compile-trace=stderr"` or `["--threads=auto", "--compile=all"]`. 
+  E.g. `\`--threads=auto project=.\``, `"--compile-trace=stderr"` or `["--threads=auto", "--compile=all"]`.
 
 * `topology`: Specifies how the workers connect to each other. Sending a message between
   unconnected workers results in an error.
@@ -178,7 +178,7 @@
     # Wait for all launches to complete.
     @sync for (i, (machine, cnt)) in enumerate(manager.machines)
         let machine=machine, cnt=cnt
-            Threads.@spawn Threads.threadpool() try
+            @async try
                 launch_on_machine(manager, $machine, $cnt, params, launched, launch_ntfy)
             catch e
                 print(stderr, "exception launching on machine $(machine) : $(e)\n")
@@ -740,16 +740,24 @@
     nothing
 end
 
-function kill(manager::LocalManager, pid::Int, config::WorkerConfig; exit_timeout = 15, term_timeout = 15)
+function kill(manager::LocalManager, pid::Int, config::WorkerConfig; profile_wait = 6, exit_timeout = 15, term_timeout = 15)
+    # profile_wait = 6 is 1s for profile, 5s for the report to show
     # First, try sending `exit()` to the remote over the usual control channels
     remote_do(exit, pid)
 
-    timer_task = Threads.@spawn Threads.threadpool() begin
+    timer_task = @async begin
         sleep(exit_timeout)
 
         # Check to see if our child exited, and if not, send an actual kill signal
         if !process_exited(config.process)
-            @warn("Failed to gracefully kill worker $(pid), sending SIGQUIT")
+            @warn "Failed to gracefully kill worker $(pid)"
+            profile_sig = Sys.iswindows() ? nothing : Sys.isbsd() ? ("SIGINFO", 29) : ("SIGUSR1" , 10)
+            if profile_sig !== nothing
+                @warn("Sending profile $(profile_sig[1]) to worker $(pid)")
+                kill(config.process, profile_sig[2])
+                sleep(profile_wait)
+            end
+            @warn("Sending SIGQUIT to worker $(pid)")
             kill(config.process, Base.SIGQUIT)
 
             sleep(term_timeout)

src/messages.jl

@@ -194,13 +194,13 @@
 function flush_gc_msgs()
     try
         for w in (PGRP::ProcessGroup).workers
-            if isa(w,Worker) && (w.state == W_CONNECTED) && w.gcflag
+            if isa(w,Worker) && ((@atomic w.state) == W_CONNECTED) && w.gcflag
                 flush_gc_msgs(w)
             end
         end
     catch e
         bt = catch_backtrace()
-        Threads.@spawn showerror(stderr, e, bt)
+        @async showerror(stderr, e, bt)
     end
 end