| Age | Commit message (Collapse) | Author |
|---|
|
* bjorn/erts/spawn_opt-doc/OTP-11849:
Document that spawn_opt/5 does not support the 'monitor' option
|
|
* sverk/ets-delete-unfix-race/OTP-11892:
Fix race between ETS table deletion and unfixation
erts: Add etp commands for alloc_util block and carrier inspection
|
|
* lukas/erts/suspend_assert/OTP-11906:
erts: Fix faulty process suspend assert
|
|
|
|
* lukas/erts/etp-lc-macros/OTP-11904:
erts: Add etp-lc-dump and etp-ppc-stacktrace macro
|
|
Symptom: VM crash running mnesia_SUITE
Scenario: Process A terminates while still having fixed table T
and process B "at the same time" deletes table T with ets:delete/1 or
by terminating.
Problem: A table scheduled for deallocation do only have a valid
'common' part. The unfix-table-at-process-exit code tried to read the
hash-specific part of such a table.
Solution: Must back off if DB_DELETE flag is set.
Since: R16B
|
|
* sverk/test-cuddle:
Fix timeout for match_spec_SUITE:otp_9422
|
|
Must receive 'abort' even after loop_runner has finished.
|
|
|
|
etp-block
etp-block2mbc
etp-carrier-blocks
|
|
Conflicts:
OTP_VERSION
erts/vsn.mk
|
|
|
|
|
|
* rickard/sys_info_spec/OTP-11859:
Fixed type spec of erlang:system_info/1
|
|
* sverk/monitor-large-heap-nosmp/OTP-11852:
erts: Fix system_monitor(large_heap) for non-smp VM
|
|
|
|
No message for large_heap monitoring was ever sent on non-smp VM.
Bug introduced in R16B.
|
|
|
|
* egil/test-cuddling:
tools: Refactor testcase
emulator: Increase timetrap timeout for binary_SUITE
emulator: Increase timetrap timeout for match_spec_SUITE
|
|
as it times out on some machines.
Not much point in redoing with lower input reductions when the terms
are so big we will trap several times anyway.
|
|
|
|
|
|
* rickard/runtime_dependencies_test:
Support for ignoring apps in runtime_dependencies test
|
|
|
|
|
|
* egil/maps-doc:
doc: Add Maps example for sequential programming
doc: Mention map expressions and map guards
doc: Add maps to reference manual
edoc: Add map/0 as a new predefined type
doc: Descripe Maps type syntax
doc: Clearify language of user-defined attributes
doc: Document Maps datatype in reference manual
erts: Document map guard functions
erts: Fix is_map/1 spec
|
|
* sverk/suspend-running-race:
erts: Fix bug of scheduling a suspended process
|
|
* sverk/test-cuddle:
erts: Fix bug in test case port_SUITE:otp_5119
erts: Reduce runtime of test port_SUITE:close_deaf_port
|
|
* vinoski/dirty-nif-purge:
prevent NIF purge during dirty NIF execution
fix dirty NIF invalid memory read
|
|
* nox/remove-superfluous-echo:
Remove a superfluous echo in the emulator Makefile
|
|
Reference-count the NIF before and after invoking a NIF on dirty schedulers
to prevent having the NIF purged during the call.
|
|
Dirty NIF support used an Export structure to facilitate calls to dirty
NIFs and finalizers, but Export isn't large enough to hold all necessary
data. This was causing an invalid memory read in beam_emu.c past the end of
the Export object. Add a local extended Export struct to erl_nif.c that can
hold all the necessary data.
|
|
* sverk/change-map-subtag:
erts: Adjust is_external_header() for new map tag
Change the subtag used for maps from 0xB to 0xF
|
|
|
|
* lukas/ose/master-17.0/OTP-11334:
erts: Move debug printout to eliminate gcc warning
|
|
These applications contains generated code with the latin1 directivce.
|
|
|
|
|
|
This is a race that can cause RUNNING (instead of RUNNING_SYS)
set on a SUSPENDED process. The effect of this race happening is probably
quite benign.
The bug was discovered by process_SUITE:processes_gc_trap on debug VM
crashing on last assert in schedule():
/* Never run a suspended process */
ASSERT(!(ERTS_PSFLG_SUSPENDED & erts_smp_atomic32_read_nob(&p->state)));
|
|
to not mistake a map for an external term (pid, port or ref).
|
|
* matwey/systemd/OTP-11829:
Add systemd socket activation for epmd.
Conflicts:
erts/epmd/src/epmd_int.h
Extra ticket: OTP-11828
|
|
* pf/document-escript-feature-only-present-in-examples/OTP-11827:
Document an escript:create/2 hidden feature
|
|
escript:create/2 accepts a 3-elements tuple containing files and
zip:create/3 options to build a zip file.
Also had to update zip typespecs to allow referral from escript docs.
|
|
* lukas/ose/master-17.0/OTP-11334:
ose: Fix erts assert failed printouts
ose: fix for packet_bytes in fd/spawn driver.
ose: Prepare slave for running on OSE
ose: Fix bug when hunting for signal proxy
ose: Implement tcp inet driver for OSE
ose: Add ifdefs for HAVE_UDP
ose: Yielding has to be done differently for background processes.
ose: Print faults in aio sys driver calls
ose: Prinout errno when to_erl read fails
ose: erlang display goes to ramlog printf
ose: Initiate stdin/stdout/stderr
ose: Break lmconf into one per load module
ose: Reset busy port when pdq empty
ose: Restore the owner of the signal
|
|
The subtag chosen for maps breaks crucial assumptions in other parts of
the system, in particular the native code generation scheme used in the
HiPE compiler. Therefore it is better to use 'the other' unused subtag.
The main change here is the use of 0xF subtag for maps instead of 0xB.
The rest of the differences are changes to and additions of comments.
One more comment from my part.
I noticed that the file contains the following two lines:
The comment of the second line is most certainly a copy and paste error
and should be appropriately fixed by the OTP developers. More importantly,
it would be great if that subtag (0x7) turned out to be unused and could
be used for maps instead of the 0xF one. This might turn out very handy
in the future. I can elaborate on this, if needed.
|
|
as it fails with timeout sometimes.
|
|
|
|
|
|
Sometimes the wrong signal, i.e. a read reply, would be
received from this non-selective receive.
|
|
The inet driver for OSE has to handle signals instead of selects
and thus the wrappers for ready_input/output are a little bit
different. However the majority of the inet code remains the same.
|
