| Age | Commit message (Collapse) | Author |
|---|
|
* origin/peppe/common_test/otp-9930:
Make line numbers show in logs for timetraps and abortions
OTP-9930
|
|
* origin/peppe/common_test/otp-9941-9955:
Fix problem with handling error_logger events
OTP-9941
OTP-9955
|
|
* origin/peppe/common_test/otp-9894:
Fix problem with large fun environment being copied unnecessarily
Implement asynchronous call to print to test case log file
OTP-9894
|
|
* ia/ssh/sftpd-file-handles/OTP-9948:
sftp daemon generates file handles correct
|
|
* hb/stdlib/digraph_condensation/OTP-9953:
Remove loops from the graph created by digraph_utils:condensation/1
|
|
When an attempt to restart a child failed, supervisor would earlier
keep the execution flow and try to restart the child over and over
again until it either succeeded or the restart frequency limit was
reached. If none of these happened, supervisor would hang forever in
this loop.
This commit adds a timer of 0 ms where the control is left back to the
gen_server which implements the supervisor. This way any incoming
request to the supervisor will be handled - which could help breaking
the infinite loop - e.g. shutdown request for the supervisor or for
the problematic child.
This introduces some incompatibilities in stdlib due to new return
values from supervisor:
* restart_child/2 can now return {error,restarting}
* delete_child/2 can now return {error,restarting}
* which_children/1 returns a list of {Id,Child,Type,Mods},
where Child, in addition to the old pid() or 'undefined',
now also can be 'restarting'.
|
|
OTP-9930
|
|
OTP-9941
OTP-9955
|
|
Also always run ssh_sftp_SUITE against own daemon as well as openssh
if existing. Corrected links test case and found that openssh has a
known bug that swaps the create link arguments. Created separate
ticket (OTP-9951) to deal with this.
|
|
* mh/create_cookie_error_msg:
Specify file name and error on create_cookie failure
OTP-9954
|
|
The function digraph_utils:condensation/1 used to create a digraph
containing loops contradicting the documentation which states that the
created digraph is free of cycles.
Thanks to Kostis Sagonas for finding the bug.
|
|
* alind/asn1/parallel-tests:
[asn1] Refactor test cases and remove extra suites
[asn1] Replace 'driver' option with 'nif'
[asn1] Correct test suite dependencies
[asn1] Parallelize test suites
[asn1] Make tables and processes unnamed
[asn1] Abstract out table access to new module
[asn1] Run asn1 compilation in a separate process
|
|
|
|
|
|
|
|
|
|
|
|
- Refactor and clean up asn1_db process
- Remove unused stop function in asn1ct.erl
- Remove infinite loop possibilites in asn1ct_check.erl
- test/1,2,3 now run in separate process
- Update documentation for new test options
|
|
All table access is now performed in a separate module. This will allow
changes to how ETS is handled by changing only this module. Note that
the module exports a very ETS-like interface for now which would have to
be maintained even if the data format would change (to a hash map for
example).
|
|
|
|
|
|
* dgud/mnesia/fix-qlc-warning:
[mnesia] Fix warning in example code
|
|
* dgud/wx/misc-improvements/OTP-9947:
[wx] Add simple taskbaricon test
[wx] Add missing stc function
[wx] Add wxTaskBarIcon class
[wx] Doc overloaded functions
[wx] Fix spec errors
[wx] Fix api bugs in wxDC
[wx] Fix wxGraphicContext bugs
[wx] Remove redundant erts version check
[wx] Generated types for all wx classes
[wx] Generate GL api from latest version
[wx] Cleanup Makefiles
|
|
* ia/ssh/currently-supported-doc/OTP-8109:
Document currently supported algorithms
|
|
* ia/ssh/error-handling:
Prevent client hanging. (OTP-8111)
Added checks of API input
Improve check so that we will not try to read ssh packet length indicator if not sure we have enough data.
Improved error handling
|
|
Restored supervisor tree so that error propagation will work as
intended, although connection processes are set to temporary, instead
of permanent with restart times set to 0, and termination of the
connection subtree is initiated by a temporary process spawned by
ssh_connection_managers terminate. This is done to avoid unwanted
supervisor reports. Pherhaps we need some new supervisor
functionality.
|
|
|
|
* uw/extending_gen:
Add plugin support for alternative name lookup
OTP-9945
|
|
wxSTC:SetEdgeMode was missing for some reason.
|
|
|
|
edoc do not handle overloaded type-specs, so doc them manually
|
|
|
|
* sverk/ets-compress-bug:
erts: Fail binary_to_term if bignum arity is too large
erts: Fix bignum-bug in ETS with compressed option
OTP-9932
|
|
* tst/no_hipe_ceach:
Remove hipe_ceach from hipe.app.src to fix reltool-generated release startup
OTP-9939
|
|
|
|
In order to avoid possible deadlock when the cth_log_redirect hook
prints via ct_logs to the test case log file, the print call must
be made by a separate (temporary) process, instead of ct_logs.
|
|
* ks/compile_info-fix:
compiler: Fix discrepancy in compile_info
OTP-9917
|
|
DC's and GC's is not double buffered by default on windows,
and there is a separate erase event which causes awful flickering
when constant updating a window.
Hack around wx (to be able to use wxBufferPaintDC),
to avoid flickering on windows.
This works on windows because there (and only there)
wxGC:create/1 also takes a memoryDC as argument.
|
|
Sigh, seems that Suse-11 does not default deliver wxWidgets with wxGRAPHICS_CONTEXT enabled,
add fallback to use wxDC again.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
A large 64-bit immediate number will be stored as SMALL_BIG_EXT by ETS
compressed format. When uncompressing, the SMALL_BIG_EXT was first
decoded as as bignum (by bytes_to_big) and then turned into a small
(by big_norm). This works for normal "binary_to_term" as
decoded_size() over-estimates the needed heap size. But for ETS no
over-estimation is done as the real term size is known and stored in
DbTerm.
Fixed by preventing bytes_to_big() from writing bignum digit when the
number is seen to fit in an immediate.
|
|
|
|
not sure we have enough data.
OTP-8380
|
|
The BEAM disassembler used the atom 'none' to signify the absence
of a compile_info chunk in a .beam file. This clashes with the type
declaration of the compile_info field of a #beam_file{} record as
containing a list. Use [] to signify the absence of this chunk.
This simplifies the code and avoids a dialyzer warning.
For fixing a similar problem and for consistency, changed also the
return type of the attributes field of the #beam_file{} record.
This required a change in the beam_disasm test suite.
|
