| Age | Commit message (Collapse) | Author |
|---|
|
|
|
* dgud/erts/fix-driver-thread-name:
Fix thread name from driver api
|
|
* egil/fix-make_internal_hash:
erts: Fix missing case in make_internal_hash
|
|
* hb/doc/refman_type_fix:
doc: Use module() in the definition of the built-in type mfa()
|
|
As for the port number in the parent commit, a FQDN can't be arbitrarily
long, at most 255 octets. Make decode fail if it's more.
|
|
A port number is a 16-bit integer, but the regexp used to parse it in
commit 1590920 slavishly followed the RFC 6733 grammar in matching an
arbitrary number of digits. Make decode fail if it's anything more than
5, to avoid doing erlang:list_to_integer/1 on arbitrarily large lists.
Also make it fail if the resulting integer is outside of the expected
range.
|
|
To bound the length of incoming messages that will be decoded. A message
longer than the specified number of bytes is discarded. An
incoming_maxlen_exceeded counter is incremented to make note of the
occurrence.
The motivation is to prevent a sufficiently malicious peer from
generating significant load by sending long messages with many AVPs for
diameter to decode. The 24-bit message length header accomodates
(16#FFFFFF - 20) div 12 = 1398099
Unsigned32 AVPs for example, which the current record-valued decode is
too slow with in practice. A bound of 16#FFFF bytes allows for 5461
small AVPs, which is probably more than enough for the majority of
applications, but the default is the full 16#FFFFFF.
|
|
* sverk/hamt-encode-size-bug/OTP-12585:
erts: Fix bug in term_to_binary size estimation for hamt
erts: Optimize term_to_binary size estimation
|
|
|
|
The most expensive part of random_ref_comp/1 is
do_matches_loop_comp/4. The running time was reduced from roughly
80 seconds to roughly 40 seconds on my computer when I modified
the caller to pre-compile the pattern.
|
|
The time to run io_proto_SUITE:unicode_options_gen/1 on my computer
was about 110 seconds. After modifying the test case so that it
does several file operations in parallel the time was reduced to
about 60 seconds.
|
|
|
|
|
|
|
|
|
|
|
|
* rickard/time_api/OTP-11997:
Misc fixes
Conflicts:
erts/preloaded/ebin/erlang.beam
erts/preloaded/ebin/erts_internal.beam
|
|
|
|
|
|
* egil/maint/process_dictionary-initial-size/OTP-12535:
erts: Document option 'hpds'
erts: Enable command line argument for initial pd size
|
|
* egil/maps/refactor-tagscheme/OTP-12585:
erts: Refactor Map - use multiple values ESTACK_PUSHN
erts: GC needs the size even if the frag is not referenced
Revert "hipe: Handle separate hashmap tag correctly"
erts: Combine flat and hash maps under one unifying tag
|
|
It was possible to configure the option, but doing so caused the service
to fail when starting a watchdog process:
{function_clause,
[{diameter_service,'-spawn_opts/1-lc$^0/1-0-',
[false],
[{file,"base/diameter_service.erl"},{line,846}]},
{diameter_service,start,5,
[{file,"base/diameter_service.erl"},{line,820}]},
{diameter_service,start,3,
[{file,"base/diameter_service.erl"},{line,782}]},
{diameter_service,handle_call,3,
[{file,"base/diameter_service.erl"},{line,385}]},
{gen_server,try_handle_call,4,[{file,"gen_server.erl"},{line,607}]},
{gen_server,handle_msg,5,[{file,"gen_server.erl"},{line,639}]},
{proc_lib,init_p_do_apply,3,[{file,"proc_lib.erl"},{line,237}]}]}
Tests for the option in the config suite were also missing.
Bungled in commit 78b3dc6.
|
|
|
|
* peppe/common_test/failing_telnet_tests:
Add documentation and make get_data behaviour backwards compatible
Introduce polling feature in ct_telnet
Fix problem with invalid timeouts because of truncated floats
OTP-12627
|
|
|
|
|
|
for tuples and maps containing ascii strings (lists).
|
|
|
|
|
|
* rickard/time_api/OTP-11997:
Fix erts_sys_hrtime() fallback
|
|
|
|
* bjorn/doc:
cerl_trees: Fix incorrect EDoc reference to the cerl module
cerl: Correct incorrect EDoc references
|
|
* rickard/time_api/OTP-11997:
Skip not updated test-cases
Fixes and cleanup
|
|
|
|
|
|
* erland/OTP18/inets/time_api/OTP-12441:
inets: Add new module inets_time_compat with new time API The new module is backwards compatible.
|
|
* erland/ssh/OTP18/time_api/OTP-12444:
ssh: Use new time API
|
|
This reverts commit 4dc5f194b0a5a6f052763bb7494b0458f98d96ca.
|
|
|
|
* sverk/hamt-term2bin-bug/OTP-12585:
erts: Optimize hashmap_get
erts: Remove HAMT_SUBTAG_NODE_ARRAY
erts: Fix bug in binary_to_term for hamt when yielding
erts: Rename to flatmap_from_validated_list
|
|
list_to_integer(lists:duplicate(10000000,$0)).
crashed due to overflow when calculating nr heap words.
|
|
Conflicts:
erts/emulator/beam/erl_map.c
erts/emulator/test/map_SUITE.erl
|
|
* egil/fix-term-cmp/OTP-12623:
erts: Fix comparison of exact terms
|
|
|
|
Use '+hpds size' to set initial process dictionary size for spawned processes.
|
|
|
|
* erland/diameter/time_api_change/OTP-12439:
Update with changes in new time api otp_SUITE: change filter för diameter
|
|
|
|
This will also fix a bug in term_to_binary
treating full nodes as tuples and emiting LIST_EXT for leafs.
|
|
Must save hamt_list in context.
|
