* lisp/emacs-lisp/comp.el (comp-mvar-symbol-p): As all slots into
a `comp-cstr' are in or fix this logic.
(comp-mvar-cons-p): Likewise.
* test/src/comp-tests.el (bug-44968): New testcase.
* test/src/comp-test-funcs.el (comp-test-44968-f): New test
function.
* lisp/term/xterm.el (xterm--read-event-for-query): Prevent
recording the characters read as the xterm response to a query,
so as not to show them in 'view-lossage'. (Bug#44908)
* test/lisp/net/tramp-tests.el (tramp-test-vec): Check for remote
`tramp-test-temporary-file-directory'.
(tramp-test11-copy-file, tramp-test12-rename-file):
Do not skip for tramp-gvfs.el.
(tramp--test-sh-p): Use `tramp-test-vec'.
* lisp/windmove.el (windmove-display-in-direction)
(windmove-delete-in-direction, windmove-swap-states-in-direction):
Add 'nomini' as MINIBUF arg of window-in-direction.
* lisp/window.el (window-in-direction): Rename arg MINI to MINIBUF.
Update docstring from walk-window-tree.
Send MINIBUF arg to walk-window-tree unchanged.
* lisp/vc/log-view.el (log-view-commit-body): Inherit from
font-lock-comment-face. This makes expanded commit messages in
log-view look the same as they did prior to commit 1f0b929430 (consult
bug#44424) (bug#44937).
Go back to using `write-file' to save bookmarks, instead of using
`write-region'. This means numbered backups of the bookmark file may
get made again, depending on the value of `bookmark-version-control'.
Thanks especially to Drew Adams and Eli Zaretskii for their
persistence in tracking down information relevant to this change.
* lisp/mail/rmailout.el (rmail-output): Fix off-by-one error in
deciding when to advance to the next message under non-nil
'rmail-output-reset-deleted-flag'. (Bug#44839)
* test/lisp/net/tramp-tests.el (tramp-test11-copy-file)
(tramp-test12-rename-file): Do not skip for tramp-gvfs.el.
* lisp/net/tramp-gvfs.el (tramp-gvfs-do-copy-or-rename-file):
Add sanity checks.
We used to store module runtime and environment pointers in the static
lists Vmodule_runtimes and Vmodule_environments. However, this is
incorrect because these objects have to be kept per-thread. With this
naive approach, interleaving module function calls in separate threads
leads to environments being removed in the wrong order, which in turn
can cause local module values to be incorrectly garbage-collected.
Instead, turn Vmodule_runtimes and Vmodule_environments into
hashtables keyed by the thread objects. The fix is relatively
localized and should therefore be safe enough for the release branch.
Module assertions now have to walk the pointer list for the current
thread, which is more correct since they now only find environments
for the current thread.
Also add a unit test that exemplifies the problem. It interleaves two
module calls in two threads so that the first call ends while the
second one is still active. Without this change, this test triggers
an assertion failure.
* src/emacs-module.c (Fmodule_load, initialize_environment)
(finalize_environment, finalize_runtime_unwind): Store runtime and
environment pointers in per-thread lists.
(syms_of_module): Initialize runtimes and environments hashtables.
(module_assert_runtime, module_assert_env, value_to_lisp): Consider
only objects for the current thread.
(module_gc_hash_table_size, module_hash_push, module_hash_pop): New
generic hashtable helper functions.
(module_objects, module_push_pointer, module_pop_pointer): New helper
functions to main thread-specific lists of runtime and environment
pointers.
(mark_modules): Mark all environments in all threads.
* test/data/emacs-module/mod-test.c (Fmod_test_funcall): New test
function.
(emacs_module_init): Bind it.
* test/src/emacs-module-tests.el (emacs-module-tests--variable): New
helper type to guard access to state in a thread-safe way.
(emacs-module-tests--wait-for-variable)
(emacs-module-tests--change-variable): New helper functions.
(emacs-module-tests/interleaved-threads): New unit test.
We used to store module runtime and environment pointers in the static
lists Vmodule_runtimes and Vmodule_environments. However, this is
incorrect because these objects have to be kept per-thread. With this
naive approach, interleaving module function calls in separate threads
leads to environments being removed in the wrong order, which in turn
can cause local module values to be incorrectly garbage-collected.
The fix isn't completely trivial: specbinding the lists wouldn't work
either, because then the garbage collector wouldn't find the
environments in other threads than the current ones, again leading to
objects being garbage-collected incorrectly. While introducing custom
pseudovector types would fix this, it's simpler to put the runtime and
environment pointers into the specbinding list as new specbinding
kinds. This works since we need to unwind them anyway, and we only
ever treat the lists as a stack. The thread switching machinery
ensures that the specbinding lists are thread-local, and that all
elements of the specbinding lists in all threads are marked during
garbage collection.
Module assertions now have to walk the specbinding list for the
current thread, which is more correct since they now only find
environments for the current thread. As a result, we can now remove
the faulty Vmodule_runtimes and Vmodule_environments variables
entirely.
Also add a unit test that exemplifies the problem. It interleaves two
module calls in two threads so that the first call ends while the
second one is still active. Without this change, this test triggers
an assertion failure.
* src/lisp.h (enum specbind_tag): Add new tags for module runtimes and
environments.
* src/eval.c (record_unwind_protect_module): New function to record a
module object in the specpdl list.
(do_one_unbind): Unwind module objects.
(backtrace_eval_unrewind, default_toplevel_binding, lexbound_p)
(Fbacktrace__locals): Deal with new specbinding types.
(mark_specpdl): Mark module environments as needed.
* src/alloc.c (garbage_collect): Remove call to 'mark-modules'.
Garbage collection of module values is now handled as part of marking
the specpdl of each thread.
* src/emacs-module.c (Fmodule_load, funcall_module): Use specpdl to
record module runtimes and environments.
(module_assert_runtime, module_assert_env, value_to_lisp): Walk
through specpdl list instead of list variables.
(mark_module_environment): Rename from 'mark_modules'. Don't attempt
to walk though current thread's environments only, since that would
miss other threads.
(initialize_environment, finalize_environment): Don't change
Vmodule_environments variable; environments are now in the specpdl
list.
(finalize_environment_unwind, finalize_runtime_unwind): Make 'extern'
since do_one_unbind now calls them.
(finalize_runtime_unwind): Don't change Vmodule_runtimes variable;
runtimes are now in the specpdl list.
(syms_of_module): Remove Vmodule_runtimes and Vmodule_environments.
* test/data/emacs-module/mod-test.c (Fmod_test_funcall): New test
function.
(emacs_module_init): Bind it.
* test/src/emacs-module-tests.el (emacs-module-tests--variable): New
helper type to guard access to state in a thread-safe way.
(emacs-module-tests--wait-for-variable)
(emacs-module-tests--change-variable): New helper functions.
(emacs-module-tests/interleaved-threads): New unit test.
When 'omake' is included in compilation-error-regexp-alist, which it
still is by default, then all other rules are modified to match with
an extra leading 6 spaces as well. The 'cucumber' pattern relied on
this in order to work as intended.
* lisp/progmodes/compile.el (compilation-error-regexp-alist-alist):
Extend the 'cucumber' pattern so that it works even when 'omake'
is not included. Move it below the 'gnu' rule so that it doesn't
match anything else.
* lisp/international/mule-cmds.el (mule--ucs-names-affixation):
Replace mule--ucs-names-annotation to display chars in prefixes
that implements two FIXME items.
(read-char-by-name): Let-bind completion-tab-width to 4.
Use affixation-function instead of annotation-function.
* lisp/minibuffer.el (completion-tab-width): New variable.
(completion--insert-strings): Align colwidth to tab positions
when completion-tab-width is non-nil.
* lisp/simple.el (completion-setup-function): Set tab-width to
completion-tab-width when completion-tab-width is non-nil.
https://lists.gnu.org/archive/html/emacs-devel/2020-11/msg01263.html
