| Age | Commit message (Collapse) | Author |
|---|
|
|
|
When multiple files with the same name are required, the features_index
hash stores the indexes in `$LOADED_FEATURES` array into a darray.
The dest and src arguments for `MEMMOVE` were wrongly reversed when
inserting a new index in the darray.
[Bug #21568]
|
|
The test introduced in 928fea3 is leaking files because the Tempfile is
not cleaned when the process exits. We see this in the output:
Children under /var/folders/51/_yzlsvf96v9729jtj8_mcp7w0000gn/T/rubytest.7lbdlp:
* -rw------- 1 0 2025-09-10 10:09:51 -0400 test20250910-13775-g4op87.rb
* -rw------- 1 0 2025-09-10 10:09:51 -0400 test20250910-13775-pu621v.rb
* -rw------- 1 0 2025-09-10 10:09:51 -0400 test20250910-13775-4xv688.rb
* -rw------- 1 0 2025-09-10 10:09:51 -0400 test20250910-13775-1rvp4b.rb
|
|
[Bug #21567]
When we require an object that is not a string, it will attempt to convert
it to a string by calling to_str on it. If we modify the $LOADED_FEATURES
array while it calls to_str, Ruby can crash because it can end up inserting
the string in the wrong index in the array.
For example, the following script crashes:
require "tempfile"
class MyString
def initialize(path)
@path = path
end
def to_str
$LOADED_FEATURES.clear
@path
end
def to_path = @path
end
def create_ruby_file = Tempfile.create(["test", ".rb"]).path
require MyString.new(create_ruby_file)
$LOADED_FEATURES.unshift(create_ruby_file)
$LOADED_FEATURES << MyString.new(create_ruby_file)
require create_ruby_file
Crash log:
test.rb:21: [BUG] Segmentation fault at 0x0000000000000004
ruby 3.5.0dev (2025-09-09T09:29:35Z master ce94add7fb) +PRISM [arm64-darwin24]
-- Crash Report log information --------------------------------------------
See Crash Report log file in one of the following locations:
* ~/Library/Logs/DiagnosticReports
* /Library/Logs/DiagnosticReports
for more details.
Don't forget to include the above Crash Report log file in bug reports.
-- Control frame information -----------------------------------------------
c:0003 p:---- s:0011 e:000010 CFUNC :require
c:0002 p:0076 s:0006 e:000005 EVAL test.rb:21 [FINISH]
c:0001 p:0000 s:0003 E:0001b0 DUMMY [FINISH]
-- Ruby level backtrace information ----------------------------------------
test.rb:21:in '<main>'
test.rb:21:in 'require'
-- Threading information ---------------------------------------------------
Total ractor count: 1
Ruby thread count for this ractor: 1
-- Machine register context ------------------------------------------------
x0: 0x0000000000000004 x1: 0x000000000000c800 x2: 0x0000000000000000
x3: 0x0000000000000000 x4: 0x0000000000000205 x5: 0x0000000000000000
x6: 0x0000000000000000 x7: 0x0000000000000001 x18: 0x0000000000000000
x19: 0x0000000209dfc0b0 x20: 0x0000000209dfc018 x21: 0x000000016ee8ab58
x22: 0x0fffffff0009d71d x23: 0x0000000209dfc018 x24: 0x0000000209dfc150
x25: 0x000000016ee8acc0 x26: 0x0000000000000000 x27: 0x0000000000000000
x28: 0x0000000000000000 lr: 0x0000000101244140 fp: 0x000000016ee887f0
sp: 0x000000016ee887d0
-- C level backtrace information -------------------------------------------
miniruby(rb_print_backtrace+0x24) [0x101317b08] vm_dump.c:843
miniruby(rb_print_backtrace) (null):0
miniruby(rb_vm_bugreport+0x26c) [0x101317d94] vm_dump.c:1175
miniruby(rb_bug_for_fatal_signal+0xa4) [0x10105ddac] error.c:1130
miniruby(sig_do_nothing+0x0) [0x1012278c0] signal.c:948
miniruby(sigsegv) (null):0
/usr/lib/system/libsystem_platform.dylib(_sigtramp+0x38) [0x19c1216a4]
miniruby(rb_str_new_frozen+0x1c) [0x101244140] string.c:1495
miniruby(rb_check_realpath_internal+0x68) [0x101077804] file.c:4679
miniruby(rb_check_realpath+0x2c) [0x101077aa4] file.c:4765
miniruby(get_loaded_features_index+0x37c) [0x1010f9c94] load.c:467
miniruby(rb_feature_p+0xd0) [0x1010f8174] load.c:582
miniruby(search_required+0xac) [0x1010f6ad4] load.c:1193
miniruby(require_internal+0x274) [0x1010f7518] load.c:1424
miniruby(rb_require_string_internal+0x94) [0x1010f6830] load.c:1571
miniruby(rb_require_string+0x58) [0x1010f66e8] load.c:1557
miniruby(rb_f_require+0x1c) [0x1010f6684] load.c:1150
miniruby(ractor_safe_call_cfunc_1+0x38) [0x101306c28] vm_insnhelper.c:3696
miniruby(vm_call_cfunc_with_frame_+0x250) [0x1012f857c] vm_insnhelper.c:3873
miniruby(vm_call_cfunc_with_frame+0x6c) [0x1012f8834] vm_insnhelper.c:3919
miniruby(vm_sendish+0x1a8) [0x1012c990c] vm_insnhelper.c:6087
miniruby(vm_exec_core+0x4050) [0x1012cfb48] insns.def:900
miniruby(vm_exec_loop+0x80) [0x1012e5448] vm.c:2666
miniruby(rb_vm_exec+0x134) [0x1012c9b40] vm.c:2645
miniruby(rb_iseq_eval_main+0x34) [0x1012e5628] vm.c:2919
miniruby(rb_ec_exec_node+0xe4) [0x10106d094] eval.c:282
miniruby(ruby_run_node+0x94) [0x10106cf64] eval.c:320
miniruby(rb_main+0x40) [0x100f7499c] main.c:42
miniruby(main+0x60) [0x100f74928] main.c:62
|
|
Notes:
Merged: https://github.com/ruby/ruby/pull/12531
|
|
|
|
|
|
|
|
Launchable reported that this was the most "flaky" test. Perhaps the
default timeout (10 seconds) is too tight for a test that uses
`GC.stress = true`. I try to relax the limit.
|
|
`Kernel#require` converts feature name objects that have the `to_path`
method such as `Pathname`, but had used the original object on error
and had resulted in an unexpected `TypeError`.
|
|
assert_in_out_err adds --disable=gems so we don't need to add
--disable-gems in the args list.
Notes:
Merged: https://github.com/ruby/ruby/pull/8303
|
|
[Bug #19793]
Dummy frames are created at the top level when requiring another file.
While requiring a file, it will try to convert using encodings. Some of
these encodings will not respond to to_str. If method_missing is
redefined on Object, then it will call method_missing and attempt raise
an error. However, the iseq is invalid as it's a dummy frame so it will
write an invalid iseq to the created NoMethodError.
The following script crashes:
```
GC.stress = true
class Object
public :method_missing
end
File.write("/tmp/empty.rb", "")
require "/tmp/empty.rb"
```
With the following backtrace:
```
frame #0: 0x00000001000fa8b8 miniruby`RVALUE_MARKED(obj=4308637824) at gc.c:1638:12
frame #1: 0x00000001000fb440 miniruby`RVALUE_BLACK_P(obj=4308637824) at gc.c:1763:12
frame #2: 0x00000001000facdc miniruby`gc_writebarrier_incremental(a=4308637824, b=4308332208, objspace=0x000000010180b000) at gc.c:8822:9
frame #3: 0x00000001000faad8 miniruby`rb_gc_writebarrier(a=4308637824, b=4308332208) at gc.c:8864:17
frame #4: 0x000000010016aff0 miniruby`rb_obj_written(a=4308637824, oldv=36, b=4308332208, filename="../iseq.c", line=1279) at gc.h:804:9
frame #5: 0x0000000100162a60 miniruby`rb_obj_write(a=4308637824, slot=0x0000000100d09888, b=4308332208, filename="../iseq.c", line=1279) at gc.h:837:5
frame #6: 0x0000000100165b0c miniruby`iseqw_new(iseq=0x0000000100d09880) at iseq.c:1279:9
frame #7: 0x0000000100165a64 miniruby`rb_iseqw_new(iseq=0x0000000100d09880) at iseq.c:1289:12
frame #8: 0x00000001000d8324 miniruby`name_err_init_attr(exc=4309777920, recv=4304780496, method=827660) at error.c:1830:35
frame #9: 0x00000001000d1b80 miniruby`name_err_init(exc=4309777920, mesg=4308332496, recv=4304780496, method=827660) at error.c:1869:12
frame #10: 0x00000001000d1bd4 miniruby`rb_nomethod_err_new(mesg=4308332496, recv=4304780496, method=827660, args=4308332448, priv=0) at error.c:1957:5
frame #11: 0x000000010039049c miniruby`rb_make_no_method_exception(exc=4304914512, format=4308332496, obj=4304780496, argc=1, argv=0x000000016fdfab00, priv=0) at vm_eval.c:959:16
frame #12: 0x00000001003b3274 miniruby`raise_method_missing(ec=0x0000000100b06f40, argc=1, argv=0x000000016fdfab00, obj=4304780496, last_call_status=MISSING_NOENTRY) at vm_eval.c:999:15
frame #13: 0x00000001003945d4 miniruby`rb_method_missing(argc=1, argv=0x000000016fdfab00, obj=4304780496) at vm_eval.c:944:5
...
frame #23: 0x000000010038f5e4 miniruby`rb_vm_call_kw(ec=0x0000000100b06f40, recv=4304780496, id=2865, argc=1, argv=0x000000016fdfab00, me=0x0000000100cbfcf0, kw_splat=0) at vm_eval.c:326:12
frame #24: 0x00000001003c18e4 miniruby`call_method_entry(ec=0x0000000100b06f40, defined_class=4304927952, obj=4304780496, id=2865, cme=0x0000000100cbfcf0, argc=1, argv=0x000000016fdfab00, kw_splat=0) at vm_method.c:2720:20
frame #25: 0x00000001003c440c miniruby`check_funcall_exec(v=6171896792) at vm_eval.c:589:12
frame #26: 0x00000001000dec00 miniruby`rb_vrescue2(b_proc=(miniruby`check_funcall_exec at vm_eval.c:587), data1=6171896792, r_proc=(miniruby`check_funcall_failed at vm_eval.c:596), data2=6171896792, args="Pȗ") at eval.c:919:18
frame #27: 0x00000001000deab0 miniruby`rb_rescue2(b_proc=(miniruby`check_funcall_exec at vm_eval.c:587), data1=6171896792, r_proc=(miniruby`check_funcall_failed at vm_eval.c:596), data2=6171896792) at eval.c:900:17
frame #28: 0x000000010039008c miniruby`check_funcall_missing(ec=0x0000000100b06f40, klass=4304923536, recv=4304780496, mid=3233, argc=0, argv=0x0000000000000000, respond=-1, def=36, kw_splat=0) at vm_eval.c:666:15
frame #29: 0x000000010038fa60 miniruby`rb_check_funcall_default_kw(recv=4304780496, mid=3233, argc=0, argv=0x0000000000000000, def=36, kw_splat=0) at vm_eval.c:703:21
frame #30: 0x000000010038fb04 miniruby`rb_check_funcall(recv=4304780496, mid=3233, argc=0, argv=0x0000000000000000) at vm_eval.c:685:12
frame #31: 0x00000001001c469c miniruby`convert_type_with_id(val=4304780496, tname="String", method=3233, raise=0, index=-1) at object.c:3061:15
frame #32: 0x00000001001c4a4c miniruby`rb_check_convert_type_with_id(val=4304780496, type=5, tname="String", method=3233) at object.c:3153:9
frame #33: 0x00000001002d59f8 miniruby`rb_check_string_type(str=4304780496) at string.c:2571:11
frame #34: 0x000000010014b7b0 miniruby`io_encoding_set(fptr=0x0000000100d09ca0, v1=4304780496, v2=4, opt=4) at io.c:11655:19
frame #35: 0x0000000100139a58 miniruby`rb_io_set_encoding(argc=1, argv=0x000000016fdfb450, io=4308334032) at io.c:13497:5
frame #36: 0x00000001003c0004 miniruby`ractor_safe_call_cfunc_m1(recv=4308334032, argc=1, argv=0x000000016fdfb450, func=(miniruby`rb_io_set_encoding at io.c:13487)) at vm_insnhelper.c:3271:12
...
frame #43: 0x0000000100390b08 miniruby`rb_funcall(recv=4308334032, mid=16593, n=1) at vm_eval.c:1137:12
frame #44: 0x00000001002a43d8 miniruby`load_file_internal(argp_v=6171899936) at ruby.c:2500:5
...
```
|
|
At the time this test first started using `assert_raise_with_message`,
it did not touch `Encoding.default_internal`.
|
|
* Fix temporary methods on Object leaking across test cases.
* Remove temporary classes/modules leaking across test cases.
Notes:
Merged: https://github.com/ruby/ruby/pull/7513
|
|
[Bug #19415]
If multiple threads attemps to load the same file concurrently
it's not a circular dependency issue.
So we check that the existing ThreadShield is owner by the current
fiber before warning about circular dependencies.
Notes:
Merged: https://github.com/ruby/ruby/pull/7257
|
|
This reverts commit fa49651e05a06512e18ccb2f54a7198c9ff579de.
Notes:
Merged: https://github.com/ruby/ruby/pull/7256
|
|
[Bug #19415]
If multiple threads attemps to load the same file concurrently
it's not a circular dependency issue.
So we check that the existing ThreadShield is owner by the current
fiber before warning about circular dependencies.
Notes:
Merged: https://github.com/ruby/ruby/pull/7252
|
|
|
|
|
|
Instead of always using a new anonymous module for Kernel#load if
the wrap argument is not false/nil, use the given module if a module
is provided.
Implements [Feature #6210]
Notes:
Merged: https://github.com/ruby/ruby/pull/4986
|
|
This fixes issues with paths being loaded twice in certain cases
when symlinks are used.
It took me multiple attempts to get this working. My original
attempt tried to convert paths to realpaths before adding them
to $LOADED_FEATURES. Unfortunately, this doesn't work well
with the loaded feature index, which is based off load paths
and not realpaths. While I was able to get require working, I'm
fairly sure the loaded feature index was not being used as
expected, which would have significant performance implications.
Additionally, I was never able to get that approach working with
autoload when autoloading a non-realpath file. It also broke
some specs.
This takes a more conservative approach. Directly before loading the
file, if the file with the same realpath has been required, the
loading of the file is skipped. The realpaths are stored as
fstrings in a hidden hash.
When rebuilding the loaded feature index, the hash of realpaths
is also rebuilt. I'm guessing this makes rebuilding process
slower, but I don think that is a hot path. In general, modifying
loaded features is only done when reloading, and that tends to be
in non-production environments.
Change test_require_with_loaded_features_pop test to use 30 threads
and 300 iterations, instead of 4 threads and 1000 iterations.
I saw only sporadic failures with 4/1000, but consistent failures
30/300 threads. These failures were due to the fact that the
concurrent deletions from $LOADED_FEATURES in other threads can
result in rb_ary_entry returning nil when rebuilding the loaded
features index.
To avoid concurrency issues when rebuilding the loaded features
index, the building of the index itself is left alone, and
afterwards, a separate loop is done on a copy of the loaded feature
snapshot in order to rebuild the realpaths hash.
Fixes [Bug #17885]
Notes:
Merged: https://github.com/ruby/ruby/pull/4887
|
|
The feature names loaded from the default load paths should also
be in the file system encoding.
Notes:
Merged: https://github.com/ruby/ruby/pull/4915
|
|
This reverts commit ddb85c5d2bdf75a83eb163856508691a7436b446.
This commit causes unexpected warnings in TestTranscode#test_loading_race
occasionally in CI.
|
|
This fixes issues with paths being loaded twice in certain cases
when symlinks are used.
It took me multiple attempts to get this working. My original
attempt tried to convert paths to realpaths before adding them
to $LOADED_FEATURES. Unfortunately, this doesn't work well
with the loaded feature index, which is based off load paths
and not realpaths. While I was able to get require working, I'm
fairly sure the loaded feature index was not being used as
expected, which would have significant performance implications.
Additionally, I was never able to get that approach working with
autoload when autoloading a non-realpath file. It also broke
some specs.
This takes a more conservative approach. Directly before loading the
file, if the file with the same realpath has been required, the
loading of the file is skipped. The realpaths are stored as
fstrings in a hidden hash.
When rebuilding the loaded feature index, the hash of realpaths
is also rebuilt. I'm guessing this makes rebuilding process
slower, but I don think that is a hot path. In general, modifying
loaded features is only done when reloading, and that tends to be
in non-production environments.
Change test_require_with_loaded_features_pop test to use 30 threads
and 300 iterations, instead of 4 threads and 1000 iterations.
I saw only sporadic failures with 4/1000, but consistent failures
30/300 threads. These failures were due to the fact that the
concurrent deletions from $LOADED_FEATURES in other threads can
result in rb_ary_entry returning nil when rebuilding the loaded
features index.
To avoid concurrency issues when rebuilding the loaded features
index, the building of the index itself is left alone, and
afterwards, a separate loop is done on a copy of the loaded feature
snapshot in order to rebuild the realpaths hash.
Fixes [Bug #17885]
Notes:
Merged: https://github.com/ruby/ruby/pull/4615
|
|
If $LOADED_FEATURES is changed in the just required file, also the
index table needs to be updated before loaded_features_snapshot is
reset. If the snapshot is reset without updating the table, the
name of the added feature will not be found.
Notes:
Merged: https://github.com/ruby/ruby/pull/4849
|
|
Notes:
Merged: https://github.com/ruby/ruby/pull/4354
|
|
I think it's more friendly and easier to work with to return `nil` when
the feature is not found in the $LOAD_PATH.
Notes:
Merged: https://github.com/ruby/ruby/pull/2317
|
|
The test fails due to unknown reason. Need to debug in future, but
tentatively skipped.
|
|
enc/trans/single_byte is needed to run some tests, however
it will fail to require because $: is empty.
|
|
Retrying after rescued `require` should try to load the same
library again. [Bug #16607]
Notes:
Merged: https://github.com/ruby/ruby/pull/2879
|
|
Dir.tmpdir can return same directory because of rand() value,
so we shouldn't rely on different name.
|
|
related https://bugs.ruby-lang.org/issues/16177
Notes:
Merged: https://github.com/ruby/ruby/pull/2736
|
|
This removes the related tests, and puts the related specs behind
version guards. This affects all code in lib, including some
libraries that may want to support older versions of Ruby.
Notes:
Merged: https://github.com/ruby/ruby/pull/2476
|
|
This removes the security features added by $SAFE = 1, and warns for access
or modification of $SAFE from Ruby-level, as well as warning when calling
all public C functions related to $SAFE.
This modifies some internal functions that took a safe level argument
to no longer take the argument.
rb_require_safe now warns, rb_require_string has been added as a
version that takes a VALUE and does not warn.
One public C function that still takes a safe level argument and that
this doesn't warn for is rb_eval_cmd. We may want to consider
adding an alternative method that does not take a safe level argument,
and warn for rb_eval_cmd.
Notes:
Merged: https://github.com/ruby/ruby/pull/2476
|
|
The taint mechanism is decided to be removed at 2.7. [Feature #16131]
So, this change removes the tests that expects a SecurityError when
requiring a file under $SAFE >= 1.
The reason why they should be removed in advance is because the upstream
of rubygems has already removed a call to "untaint" method, which makes
the tests fail.
|
|
The extension for require method with rubygems affects this test
case. We need to keep this test for the original method.
|
|
|
|
|
|
This reverts commit 7a75baa6e294473f02da512c99f7ef1f76b2d83c.
Revert "Defer skip to avoid errors on ensure"
This reverts commit adfc8d6dbadbccef27d6ec78022650840c7604cc.
because 76e2370f132f83c16c9de39a0a9356579f364527 is hoped to fix the
problem.
|
|
|
|
somehow the test fails only on Wercker in CIs triggered by GitHub.
Actually rubyci and ci.rvm.jp also have issues too, though.
|
|
Moved from RubyVM. [Feature #15903]
|
|
|
|
|
|
* load.c (rb_resolve_feature_path): search the path for already
loaded feature.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@66661 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
|
|
* vm_core.h (rb_vm_t): move `rb_execution_context_t::safe_level` to
`rb_vm_t::safe_level_` because `$SAFE` is a process (VM) global state.
* vm_core.h (rb_proc_t): remove `rb_proc_t::safe_level` because `Proc`
objects don't need to keep `$SAFE` at the creation.
Also make `is_from_method` and `is_lambda` as 1 bit fields.
* cont.c (cont_restore_thread): no need to keep `$SAFE` for Continuation.
* eval.c (ruby_cleanup): use `rb_set_safe_level_force()` instead of access
`vm->safe_level_` directly.
* eval_jump.c: End procs `END{}` doesn't keep `$SAFE`.
* proc.c (proc_dup): removed and introduce `rb_proc_dup` in vm.c.
* safe.c (rb_set_safe_level): don't check `$SAFE` 1 -> 0 changes.
* safe.c (safe_setter): use `rb_set_safe_level()`.
* thread.c (rb_thread_safe_level): `Thread#safe_level` returns `$SAFE`.
It should be obsolete.
* transcode.c (load_transcoder_entry): `rb_safe_level()` only returns
0 or 1 so that this check is not needed.
* vm.c (vm_proc_create_from_captured): don't need to keep `$SAFE` for Proc.
* vm.c (rb_proc_create): renamed to `proc_create`.
* vm.c (rb_proc_dup): moved from proc.c.
* vm.c (vm_invoke_proc): do not need to set and restore `$SAFE`
for `Proc#call`.
* vm_eval.c (rb_eval_cmd): rename a local variable to represent clearer
meaning.
* lib/drb/drb.rb: restore `$SAFE`.
* lib/erb.rb: restore `$SAFE`, too.
* test/lib/leakchecker.rb: check `$SAFE == 0` at the end of tests.
* test/rubygems/test_gem.rb: do not set `$SAFE = 1`.
* bootstraptest/test_proc.rb: catch up this change.
* spec/ruby/optional/capi/string_spec.rb: ditto.
* test/bigdecimal/test_bigdecimal.rb: ditto.
* test/fiddle/test_func.rb: ditto.
* test/fiddle/test_handle.rb: ditto.
* test/net/imap/test_imap_response_parser.rb: ditto.
* test/pathname/test_pathname.rb: ditto.
* test/readline/test_readline.rb: ditto.
* test/ruby/test_file.rb: ditto.
* test/ruby/test_optimization.rb: ditto.
* test/ruby/test_proc.rb: ditto.
* test/ruby/test_require.rb: ditto.
* test/ruby/test_thread.rb: ditto.
* test/rubygems/test_gem_specification.rb: ditto.
* test/test_tempfile.rb: ditto.
* test/test_tmpdir.rb: ditto.
* test/win32ole/test_win32ole.rb: ditto.
* test/win32ole/test_win32ole_event.rb: ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61510 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
|
|
* load.c (rb_construct_expanded_load_path): expand load paths to
real paths to get rid of duplicate loading from symbolic-linked
directories. [Feature #10222]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@59984 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
|
|
* test/ruby/test_require.rb (test_require_with_unc): remove extra
slash. the result of File.expand_path has a slash just follows
the drive letter.
https://github.com/ruby/ruby/commit/4ef849bd9014746821402e21053a53ea61b3690d#commitcomment-22391133
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@59012 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
|
|
* test/ruby/test_require.rb (test_require_with_unc): ignore EPERM.
GetFileAttributes and CreateFile with backup semantics fail at a
symlink in a shared folder to another shared folder.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@59009 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
|
|
* file.c (path_check_0): preserve encoding of path name in warning
message.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@59008 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
|
