Age | Commit message (Collapse) | Author |
|
OTP-15882
|
|
Add a "proper" behaviour module for the megaco_user behaviour.
This behaviour is not a proper behaviour since its possible to
configure megaco to add additional arguments to *all* of the
functions in the behaviour (with the user_args config option).
So, this behaviour only reflects the *default config* (whith
the user_args set to [], that is no extra args).
OTP-15882
|
|
Made the behaviour definitions of megaco_encoder and
megaco_edist_compress follow "modern" standards (which
satisfies dialyzer).
OTP-15882
|
|
Removed a single ']' that did no good from the doc...
OTP-15882
|
|
Mostly added nowarn_function-attributes to silence dialyzer
(for functions that are "future" proofed).
But actually also found some errors (although the issues have
not been found for almost 15 years, so...).
OTP-15882
|
|
The actual "improvement" has to do with io. That is, making
sure we get any. A test case failed (on a OpenIndiana VM),
which basically only pointed to which line in the test code.
But it was impossible to figure out from that what actually
had gone wrong.
Also removed the 'export all' compiler flag, which caused a
whole bunch of unused function to become visible. Commented
out those...
|
|
The simple test case was faulty in that the simple example
(which it is testing) require that the MGC and MG run on different
erlang nodes. But the test case did not do this. Instead, it ran
both the MGC and the MG on the test server node, which caused
the test case to fail 1 time of 10.
|
|
We did not check the actual result of the timer cancel
operation before continuing, which seems to cause
problems on some platforms (where the timer may actually
expire even though we have 100 ms to cancel).
So now we check the cancel result, and if we did not
succeed, we instead skip the test case.
|
|
|
|
The examples test suite was hard to debug just based
on its output (both plain debug printouts and printouts
generated via dbg). Some timeouts has been added between
the various actions so that it will be easier to understand
the output.
|
|
* kostis/doc/map-type-is-an-alias:
The map() type is an alias; not a predefined type
|
|
* sverker/system_info-procs-bug/ERL-979/OTP-15909:
Fix fatal bug in erts_proc_sig_signal_size
|
|
for suppression file.
|
|
* peterdmv/ssl/fix-cert-error-handling/OTP-15900:
ssl: Fix handling of certificate decoding problems
|
|
* peterdmv/ssl/fix-sign-algs-cert/OTP-15913:
ssl: Backport fix for signature_algorithms_cert
|
|
|
|
This pull request corrects an historical accident.
At some point in time, the set of Erlang terms was extended to include
maps, and at that time the `map()' type was also introduced to the
language as an easy, albeit quite coarse, way to declare map terms.
At some later point in time, the type language was extended with the
ability to declare _type associations_ in maps and also with a notation
to declare mandatory (:=) and optional (=>) type associations.
At that point in time, it should have been realized that the `map()'
type is nothing more than an alias (for `#{any() => any()}') instead
of being a _predefined_ type. (The situation is analogous to e.g.
the `binary()' type being just an alias for `<<_:_*8>>'.)
This pull request corrects the documentation (the reference manual).
|
|
|
|
* bjorn/erts/unoptimized-plus:
Handle addition of bignum + variable in unoptimized code
|
|
* lukas/os_mon/disk_sup_ignore_sqashfs/OTP-15778:
os_mon: Make disk_sup ignore squashfs
|
|
|
|
into maint
|
|
|
|
|
|
Seems the right thing to do as the compiled code is not necessarily
executed by a single CPU thread.
Only warning if old gcc and will work as good/bad as it did before.
|
|
|
|
|
|
to generate nice html from logs of gcov compiled emulator.
|
|
called by
erlang:process_info(memory)
erlang:system_info(procs)
crash dump
break menu
Will crash VM if process has a pending monitor DOWN signal in its queue.
|
|
|
|
Ideally we should ignore all read only filesystems,
but this will have to do for now.
We do this as we don't want disk_sup to report that
a filesystem is almost full if it is read-only.
|
|
This commit fixes interoperability problems with openssl when
the TLS 1.3 server is using the option signature_algs_cert.
In such cases the signature_algorithms_cert extension was encoded
as a signature_algorithms extension and openssl s_client returned
an Illegal Parameter Alert due to its filtering of extension
duplicates.
|
|
* hans/crypto/FIPS_macs/OTP-13872:
crypto: Fix FIPS flags for MACs
crypto: Fix FIPS flags for digests in HMAC
|
|
* hans/crypto/FIPS_digests/OTP-15911:
crypto: Fix FIPS flags for digests
crypto: Rename cipher fips macro
|
|
|
|
|
|
|
|
FORBIDDEN_IN_FIPS -> CIPHER_FORBIDDEN_IN_FIPS
|
|
* hans/crypto/cuddle_docs:
crypto: Fix bad link
|
|
|
|
* ingela/ssl/TLS-hibernate-bug/OTP-15910:
ssl: Fix hibernation bug
|
|
|
|
* kjell/stdlib/ets_ordered_set_slow_react/OTP-15906:
ETS ordered_set: Improvements to the CA tree implementation
|
|
This commit only affects the implementation of ETS `ordered_set`
tables with the `write_concurrency` option enabled. Such tables are
implemented with a data structure that is called the contention
adapting search tree (CA tree). This commit introduces the following
changes:
* This commit causes a join to be triggered in one randomly selected
base node in about one of 1000 read unlock calls for base node
locks. No such joins happened before this commit. Before this
commit, operations that only acquired looks in read-mode never
triggered any contention adaptation. Therefore, the CA tree could
get stuck in a sub-optimal state in certain scenarios. This could
happen, for example, when a CA tree is first populated with parallel
inserts (which will cause splits of base nodes) and then only
read-only operations are applied to the data structure. Benchmark
results from the
`ets_SUITE:lookup_catree_par_vs_seq_init_benchmark/0` benchmark
function (which is included in this commit) shows that this change
can improve the throughput of the CA tree in the scenario described
above.
* Read-only operations will now also increase values of statistics
counters when they detect that they need to wait for other
operations. Only write operation changed statistics counters before
this commit. This improves the statistics that the adaptation
heuristics is based on.
* Additionally, this commit adds an upper and lower limit to the
contention statistics variables in the base nodes. Such limits did
not exist before this commit. This should, for example, make the CA
tree more responsive to contention after long periods of low
contention.
|
|
Return real pend value in erlang:dist_get_stat/1
|
|
alt_dist docs: Correct the mf_getstat description
|
|
* maint-22:
Updated OTP version
Prepare release
# Conflicts:
# make/otp_version_tickets
|
|
garazdawi/lukas/erts/fix_active_n_close_win32/ERL-960/OTP-15901
Fix {active,N} close race condition on windows
|
|
* maint-21:
Updated OTP version
Prepare release
# Conflicts:
# OTP_VERSION
# erts/doc/src/notes.xml
# erts/vsn.mk
# lib/common_test/doc/src/notes.xml
# lib/common_test/vsn.mk
# lib/kernel/doc/src/notes.xml
# lib/kernel/vsn.mk
# lib/public_key/doc/src/notes.xml
# lib/public_key/vsn.mk
# lib/ssl/doc/src/notes.xml
# lib/ssl/vsn.mk
# lib/stdlib/doc/src/notes.xml
# lib/stdlib/vsn.mk
# make/otp_version_tickets
# otp_versions.table
|
|
|