Revision f3d17d3d8e6df78e7b872c6f43da186148892406 authored by Keno Fischer on 22 April 2019, 00:52:27 UTC, committed by Keno Fischer on 26 April 2019, 22:01:56 UTC
After fixing #30118, the behavior of #31649 changed. Instead of crashing immediately upon attempting to print `spvec`, the first time would fail with the correct ambiguity. However, any subsequent attempts to print the same value would yet again result in a crash. Upon investigation, the series of events leading to this crash are as follows: 1. Showing the first error message attempts to put `convert(::Type{Any}, ::Nothing)` into the cache (matching the previous `convert(::Type{Any}, x::Any)` method). 2. jl_compilation_sig widens this to `convert(::DataType, x::Any)`. No guard signatures are allocated, because `ml_matches` does not return `convert(::Any, ::sparse_t)` because it is uncallable due to the ambiguity. 3. Upon the next call, we find `convert(::DataType, x::Any)` in the cache and use it. The subsequent successful return conflicts with type inference's result that this call will not return (i.e. is inferred to Union{}). To fix this change the `ml_matches` call in step 2 to include ambiguous matches, thus allocating guard entries for these signatures and preventing the cache match. That works for this issue, but I should note that I have seen additional crashes along similar lines while working on this fix, so there may be additional issues in the vicinity. Fixes #31649
1 parent a030b0a
runtests.jl
# This file is a part of Julia. License is MIT: https://julialang.org/license
using Test
using Distributed
import REPL
using Base.Printf: @sprintf
include("choosetests.jl")
include("testenv.jl")
tests, net_on, exit_on_error, seed = choosetests(ARGS)
tests = unique(tests)
const max_worker_rss = if haskey(ENV, "JULIA_TEST_MAXRSS_MB")
parse(Int, ENV["JULIA_TEST_MAXRSS_MB"]) * 2^20
else
typemax(Csize_t)
end
limited_worker_rss = max_worker_rss != typemax(Csize_t)
function test_path(test)
t = split(test, '/')
if t[1] in STDLIBS
if length(t) == 2
return joinpath(STDLIB_DIR, t[1], "test", t[2])
else
return joinpath(STDLIB_DIR, t[1], "test", "runtests")
end
else
return joinpath(@__DIR__, test)
end
end
# Check all test files exist
isfiles = isfile.(test_path.(tests) .* ".jl")
if !all(isfiles)
error("did not find test files for the following tests: ",
join(tests[.!(isfiles)], ", "))
end
const node1_tests = String[]
function move_to_node1(t)
if t in tests
splice!(tests, findfirst(isequal(t), tests))
push!(node1_tests, t)
end
nothing
end
# Base.compilecache only works from node 1, so precompile test is handled specially
move_to_node1("precompile")
move_to_node1("SharedArrays")
# Ensure things like consuming all kernel pipe memory doesn't interfere with other tests
move_to_node1("stress")
# In a constrained memory environment, run the "distributed" test after all other tests
# since it starts a lot of workers and can easily exceed the maximum memory
limited_worker_rss && move_to_node1("Distributed")
import LinearAlgebra
cd(@__DIR__) do
n = 1
if net_on
n = min(Sys.CPU_THREADS, length(tests))
n > 1 && addprocs_with_testenv(n)
LinearAlgebra.BLAS.set_num_threads(1)
end
skipped = 0
@everywhere include("testdefs.jl")
#pretty print the information about gc and mem usage
testgroupheader = "Test"
workerheader = "(Worker)"
name_align = maximum([length(testgroupheader) + length(" ") + length(workerheader); map(x -> length(x) + 3 + ndigits(nworkers()), tests)])
elapsed_align = length("Time (s)")
gc_align = length("GC (s)")
percent_align = length("GC %")
alloc_align = length("Alloc (MB)")
rss_align = length("RSS (MB)")
printstyled(testgroupheader, color=:white)
printstyled(lpad(workerheader, name_align - length(testgroupheader) + 1), " | ", color=:white)
printstyled("Time (s) | GC (s) | GC % | Alloc (MB) | RSS (MB)\n", color=:white)
results=[]
print_lock = ReentrantLock()
function print_testworker_stats(test, wrkr, resp)
lock(print_lock)
try
printstyled(test, color=:white)
printstyled(lpad("($wrkr)", name_align - length(test) + 1, " "), " | ", color=:white)
time_str = @sprintf("%7.2f",resp[2])
printstyled(lpad(time_str, elapsed_align, " "), " | ", color=:white)
gc_str = @sprintf("%5.2f", resp[5].total_time / 10^9)
printstyled(lpad(gc_str, gc_align, " "), " | ", color=:white)
# since there may be quite a few digits in the percentage,
# the left-padding here is less to make sure everything fits
percent_str = @sprintf("%4.1f", 100 * resp[5].total_time / (10^9 * resp[2]))
printstyled(lpad(percent_str, percent_align, " "), " | ", color=:white)
alloc_str = @sprintf("%5.2f", resp[3] / 2^20)
printstyled(lpad(alloc_str, alloc_align, " "), " | ", color=:white)
rss_str = @sprintf("%5.2f", resp[6] / 2^20)
printstyled(lpad(rss_str, rss_align, " "), "\n", color=:white)
finally
unlock(print_lock)
end
end
all_tests = [tests; node1_tests]
local stdin_monitor
all_tasks = Task[]
try
# Monitor stdin and kill this task on ^C
# but don't do this on Windows, because it may deadlock in the kernel
if !Sys.iswindows() && isa(stdin, Base.TTY)
t = current_task()
stdin_monitor = @async begin
term = REPL.Terminals.TTYTerminal("xterm", stdin, stdout, stderr)
try
REPL.Terminals.raw!(term, true)
while true
if read(term, Char) == '\x3'
Base.throwto(t, InterruptException())
break
end
end
catch e
isa(e, InterruptException) || rethrow()
finally
REPL.Terminals.raw!(term, false)
end
end
end
@sync begin
for p in workers()
@async begin
push!(all_tasks, current_task())
while length(tests) > 0
test = popfirst!(tests)
local resp
wrkr = p
try
resp = remotecall_fetch(runtests, wrkr, test, test_path(test); seed=seed)
catch e
isa(e, InterruptException) && return
resp = [e]
end
push!(results, (test, resp))
if resp[1] isa Exception
if exit_on_error
skipped = length(tests)
empty!(tests)
end
elseif resp[end] > max_worker_rss
if n > 1
rmprocs(wrkr, waitfor=30)
p = addprocs_with_testenv(1)[1]
remotecall_fetch(include, p, "testdefs.jl")
else # single process testing
error("Halting tests. Memory limit reached : $resp > $max_worker_rss")
end
end
!isa(resp[1], Exception) && print_testworker_stats(test, wrkr, resp)
end
if p != 1
# Free up memory =)
rmprocs(p, waitfor=30)
end
end
end
end
n > 1 && length(node1_tests) > 1 && print("\nExecuting tests that run on node 1 only:\n")
for t in node1_tests
# As above, try to run each test
# which must run on node 1. If
# the test fails, catch the error,
# and either way, append the results
# to the overall aggregator
isolate = true
t == "SharedArrays" && (isolate = false)
local resp
try
resp = eval(Expr(:call, () -> runtests(t, test_path(t), isolate, seed=seed))) # runtests is defined by the include above
print_testworker_stats(t, 1, resp)
catch e
resp = [e]
end
push!(results, (t, resp))
end
catch e
isa(e, InterruptException) || rethrow()
# If the test suite was merely interrupted, still print the
# summary, which can be useful to diagnose what's going on
foreach(task -> begin
istaskstarted(task) || return
istaskdone(task) && return
try
schedule(task, InterruptException(); error=true)
catch ex
@error "InterruptException" exception=ex,catch_backtrace()
end
end, all_tasks)
foreach(wait, all_tasks)
finally
if @isdefined stdin_monitor
schedule(stdin_monitor, InterruptException(); error=true)
end
end
#=
` Construct a testset on the master node which will hold results from all the
test files run on workers and on node1. The loop goes through the results,
inserting them as children of the overall testset if they are testsets,
handling errors otherwise.
Since the workers don't return information about passing/broken tests, only
errors or failures, those Result types get passed `nothing` for their test
expressions (and expected/received result in the case of Broken).
If a test failed, returning a `RemoteException`, the error is displayed and
the overall testset has a child testset inserted, with the (empty) Passes
and Brokens from the worker and the full information about all errors and
failures encountered running the tests. This information will be displayed
as a summary at the end of the test run.
If a test failed, returning an `Exception` that is not a `RemoteException`,
it is likely the julia process running the test has encountered some kind
of internal error, such as a segfault. The entire testset is marked as
Errored, and execution continues until the summary at the end of the test
run, where the test file is printed out as the "failed expression".
=#
o_ts = Test.DefaultTestSet("Overall")
Test.push_testset(o_ts)
completed_tests = Set{String}()
for res in results
push!(completed_tests, res[1])
if isa(res[2][1], Test.DefaultTestSet)
Test.push_testset(res[2][1])
Test.record(o_ts, res[2][1])
Test.pop_testset()
elseif isa(res[2][1], Tuple{Int,Int})
fake = Test.DefaultTestSet(res[1])
for i in 1:res[2][1][1]
Test.record(fake, Test.Pass(:test, nothing, nothing, nothing))
end
for i in 1:res[2][1][2]
Test.record(fake, Test.Broken(:test, nothing))
end
Test.push_testset(fake)
Test.record(o_ts, fake)
Test.pop_testset()
elseif isa(res[2][1], RemoteException) && isa(res[2][1].captured.ex, Test.TestSetException)
println("Worker $(res[2][1].pid) failed running test $(res[1]):")
Base.showerror(stdout,res[2][1].captured)
fake = Test.DefaultTestSet(res[1])
for i in 1:res[2][1].captured.ex.pass
Test.record(fake, Test.Pass(:test, nothing, nothing, nothing))
end
for i in 1:res[2][1].captured.ex.broken
Test.record(fake, Test.Broken(:test, nothing))
end
for t in res[2][1].captured.ex.errors_and_fails
Test.record(fake, t)
end
Test.push_testset(fake)
Test.record(o_ts, fake)
Test.pop_testset()
elseif isa(res[2][1], Exception)
# If this test raised an exception that is not a remote testset exception,
# i.e. not a RemoteException capturing a TestSetException that means
# the test runner itself had some problem, so we may have hit a segfault,
# deserialization errors or something similar. Record this testset as Errored.
fake = Test.DefaultTestSet(res[1])
Test.record(fake, Test.Error(:test_error, res[1], res[2][1], [], LineNumberNode(1)))
Test.push_testset(fake)
Test.record(o_ts, fake)
Test.pop_testset()
else
error(string("Unknown result type : ", typeof(res)))
end
end
for test in all_tests
(test in completed_tests) && continue
fake = Test.DefaultTestSet(test)
Test.record(fake, Test.Error(:test_interrupted, test, InterruptException(), [], LineNumberNode(1)))
Test.push_testset(fake)
Test.record(o_ts, fake)
Test.pop_testset()
end
println()
Test.print_test_results(o_ts,1)
if !o_ts.anynonpass
println(" \033[32;1mSUCCESS\033[0m")
else
println(" \033[31;1mFAILURE\033[0m\n")
skipped > 0 &&
println("$skipped test", skipped > 1 ? "s were" : " was", " skipped due to failure.")
println("The global RNG seed was 0x$(string(seed, base = 16)).\n")
Test.print_test_errors(o_ts)
throw(Test.FallbackTestSetException("Test run finished with errors"))
end
end
![swh spinner](/static/img/swh-spinner.gif)
Computing file changes ...