aboutsummaryrefslogtreecommitdiffstats
path: root/erts
AgeCommit message (Collapse)Author
2015-06-15Update prim_inet.beamRaimo Niskanen
2015-06-09Fix socket option {linger, {true, 0}} to abort TCP connectionsRory Byrne
Up until now, if {linger, {true, 0}} is set on the socket and there is data in the port driver queue, the connection is not aborted until the port queue is empty and close() is called on the underlying file descriptor. This bug allows an idle TCP client to prevent a server from terminating the connection and freeing resources. This patch fixes the problem by discarding the port queue if the socket is closed when {linger, {true, 0}} is set.
2015-06-09Apply 'show_econnreset' socket option to send errors as wellRory Byrne
Up till now all send errors have been translated into a generic {error, closed}. This patch allows {error, econnreset} to be returned on send errors when it is detected that the TCP peer has sent an RST.
2015-06-09Add 'show_econnreset' TCP socket optionRory Byrne
An ECONNRESET is a socket error which tells us that a TCP peer has sent an RST. The RST indicates that they have aborted the connection and that the payload we have received should not be considered complete. Up until now, the implementation of TCP in inet_drv.c has hidden the receipt of the RST from the user, treating it as though it was just a FIN terminating the read side of the socket. There are many cases where user code needs to be able to distinguish between a socket that was closed normally and one that was aborted. Setting the option {show_econnreset, true} enables the user to receive ECONNRESET errors on both active and passive sockets. A connected socket returned from gen_tcp:accept/1 will inherit the show_econnreset setting of the listening socket. By default this option is set to {show_econnreset, false}. Note that this patch only enables the reporting of ECONNRESET when the socket is being read from. It does not report ECONNRESET (or EPIPE) when the user tries to write to a connection when an RST has already been received. Currently the TCP implementation in inet_drv.c hides all such send errors from the user in favour of returning {error, close}. A separate patch will be needed to enable the reporting of such errors.
2015-06-08Merge branch 'egil/fix-internal_hash-list-opt'Björn-Egil Dahlberg
* egil/fix-internal_hash-list-opt: erts: Fix faulty list optimization in make_internal_hash
2015-06-08erts: Fix faulty list optimization in make_internal_hashBjörn-Egil Dahlberg
Reported-by: Rory Byrne
2015-06-05Merge branch 'henrik/typo-fix-branch'Henrik Nord
* henrik/typo-fix-branch: Fixing typo
2015-06-05Reorder scheduler information in crashdumpsDan Gudmundsson
To make it easier to parse stack trace information from tools
2015-06-04Merge branch 'egil/cuddle-tests'Björn-Egil Dahlberg
* egil/cuddle-tests: erts: Remove remaining ?line macros in trace_meta_SUITE erts: Relax trace_meta_SUITE:on_and_off_test/1 erts: Remove ?line macro in trace_meta_SUITE:on_and_off_test/1 kernel: Remove remaining ?line macros in gen_tcp_misc_SUITE kernel: Relax gen_tcp_misc_SUITE don't wait for timeout kernel: Relax test gen_tcp_misc_SUITE:wrapping_oct/1
2015-06-04erts: Remove remaining ?line macros in trace_meta_SUITEBjörn-Egil Dahlberg
Also, fixed some confusing indentations and whitespace errors. This commit only has whitespace changes and removal of ?line macros.
2015-06-04erts: Relax trace_meta_SUITE:on_and_off_test/1Björn-Egil Dahlberg
2015-06-04erts: Remove ?line macro in trace_meta_SUITE:on_and_off_test/1Björn-Egil Dahlberg
2015-06-03Merge commit fixing typoHenrik Nord
Thanks to Nathanael Bartley
2015-06-03Merge branch 'jh/freebsdcpu/OTP-12796'Björn-Egil Dahlberg
* jh/freebsdcpu/OTP-12796: os_mon: Comply with C89 standard os_mon: Update docs for FreeBSD os_mon: Implement cpu_sup:util/0,1 for FreeBSD
2015-06-02Merge branch 'nybek/fix_gen_tcp_shutdown'Zandra Hird
OTP-12797
2015-06-01Merge branch 'vinoski/enif-raise-exception/OTP-12770'Sverker Eriksson
* vinoski/enif-raise-exception/OTP-12770: Add enif_raise_exception Enhance enif_has_pending_exception Fix for enif_schedule_nif and exceptions Conflicts: erts/doc/src/erl_nif.xml
2015-06-01Merge branch 'sverk/nif-map-docs'Sverker Eriksson
* sverk/nif-map-docs: erts: Cleanup fsummary lines in erl_nif docs erts: Fix alphabetic order in erl_nif doc erts: Add docs for map functions in nif API erts: Rename ErlNifMapIteratorEntry enums
2015-05-29Merge branch 'legoscia/dialyzer/cache-hipe-binaries/OTP-12779'Björn Gustavsson
* legoscia/dialyzer/cache-hipe-binaries/OTP-12779: Dialyzer to cache results of HiPE compilation Change hipe_bifs:system_crc/1 to hipe_bifs:system_crc/0
2015-05-28erts: Cleanup fsummary lines in erl_nif docsSverker Eriksson
by removing all full stop.
2015-05-28erts: Fix alphabetic order in erl_nif docSverker Eriksson
enif_make_reverse_list was at the wrong place
2015-05-28erts: Add docs for map functions in nif APISverker Eriksson
2015-05-28Merge branch 'richcarl/warnings-by-default/OTP-12781'Björn-Egil Dahlberg
* richcarl/warnings-by-default/OTP-12781: stdlib: Use warning channel in test qlc_SUITE:otp_6964/1 stdlib: Fix testcase for qlc_SUITE kernel: Fix code_SUITE with respect to new logger default Map error logger warnings to warning messages by default
2015-05-27Change hipe_bifs:system_crc/1 to hipe_bifs:system_crc/0Magnus Henoch
The macro HIPE_SYSTEM_CRC used to contain a hidden cookie from the VM that generated hipe_literals.hrl. This means that BEAM files containing that macro would be tied to a particular version of the VM. Change hipe_bifs:system_crc such that it doesn't require a hidden cookie to return the desired value.
2015-05-27erts: Rename ErlNifMapIteratorEntry enumsSverker Eriksson
To differentiate between first/last map entry and head/tail which is before/after first/last map entry.
2015-05-26Add enif_raise_exceptionSteve Vinoski
Add enif_raise_exception function to allow NIFs to raise error exceptions holding any Erlang terms. This does not replace or deprecate the enif_make_badarg function, though, because raising badarg errors is so idiomatic in NIFs. Reimplement enif_make_badarg on top of enif_raise_exception. Add new tests for enif_raise_exception for both normal and dirty NIFs. Add documentation for enif_raise_exception.
2015-05-26Enhance enif_has_pending_exceptionSteve Vinoski
Sverker Eriksson came up with the following idea: to handle a future ability for NIFs to raise more than just badarg exceptions, modify the recently-added enif_has_pending_exception function to take a second argument: a pointer to ERL_NIF_TERM. If this argument is a null pointer, ignore it. Otherwise, if the first argument, an ErlNifEnv*, has an associated exception, set the pointed-to ERL_NIF_TERM of the second argument to the value of the exception term. Add new tests and documentation for this modification.
2015-05-26Fix for enif_schedule_nif and exceptionsSteve Vinoski
Fix a place where part of the implementation of enif_schedule_nif was not using the ErlNifEnv exception_thrown field when it should have been. Also make the result of enif_schedule_nif return false when passed to enif_is_exception, and add an assertion for this to the nif_SUITE.c tests.
2015-05-26os_mon: Implement cpu_sup:util/0,1 for FreeBSDWasif Malik and Johannes Huning
Authors: Wasif Malik and Johannes Huning
2015-05-26Merge branch 'egil/cuddle-tests'Björn-Egil Dahlberg
* egil/cuddle-tests: erts: Relax monitor_SUITE:mixer/1 erts: Refactor monitor_SUITE:mixer/1 stdlib: Minor change of unicode_SUITE stdlib: Relax iterations in binary_module_SUITE:random_ref_sr_comp/1 Remove ?line macro in binary_module_SUITE:random_ref_sr_comp/1 stdlib: Relax iterations in binary_module_SUITE:random_ref_comp/1 stdlib: Remove ?line macro in binary_module_SUITE:random_ref_comp/1 stdlib: Relax filesize of io_proto_SUITE:unicode_options_gen/1 stdlib: Relax rand_SUITE timeout debugger: Relax test map_SUITE:t_guard_receive_large/1 tools: Relax lcnt test time stdlib: ets_SUITE cuddling
2015-05-26Merge branch 'jlouis/cocci-noderef-fixes'Henrik Nord
* jlouis/cocci-noderef-fixes: Correct usage of sizeof() for pointer types OTP-12771
2015-05-25erts: Relax monitor_SUITE:mixer/1Björn-Egil Dahlberg
2015-05-25erts: Refactor monitor_SUITE:mixer/1Björn-Egil Dahlberg
Remove ?line macro.
2015-05-25Merge branch 'vinoski/ds-timer-wheel-fix'Rickard Green
* vinoski/ds-timer-wheel-fix: Avoid timer wheel work in dirty schedulers
2015-05-25Merge branch 'rickard/timer-improvement/OTP-12650'Rickard Green
* rickard/timer-improvement/OTP-12650: Fix bug causing timeout to overwrite exit instruction Doc fixes
2015-05-25Merge branch 'rickard/fix_alloc_stat_fix/OTP-12766'Rickard Green
* rickard/fix_alloc_stat_fix/OTP-12766: Fix statistics reported about fix alloc types
2015-05-25Fix bug causing timeout to overwrite exit instructionRickard Green
2015-05-25Doc fixesRickard Green
2015-05-25Fix statistics reported about fix alloc typesRickard Green
2015-05-22Map error logger warnings to warning messages by defaultRichard Carlsson
Also fix and document the broken +We option.
2015-05-22Merge branch 'bk/configure-arch-ppc64le/OTP-12761'Björn-Egil Dahlberg
* bk/configure-arch-ppc64le/OTP-12761: Add architecture ppc64le architecture as a ppc64
2015-05-21Add architecture ppc64le architecture as a ppc64Breno Leitao
ppc64le is a new arch/ABI that runs on IBM POWER processor starting at POWER8 processor, and it should be listed as a ppc64, mainly for hipe enablement. For more information about this new arch/ABI, please check the documents at: https://www-03.ibm.com/technologyconnect/tgcm/TGCMServlet.wss?alias=OpenPOWER
2015-05-21Update runtime dependenciesRickard Green
2015-05-20Revert "Map error logger warnings to warning messages by default"Zandra Hird
This reverts commit 4c4d7fa40e5fb59854724ce74b8aa3546525cb90. This pr is causing some test failures that were missed at first.
2015-05-20Merge branch 'richcarl/warnings-by-default'Zandra Hird
* richcarl/warnings-by-default: Map error logger warnings to warning messages by default OTP-12755
2015-05-20Merge branch 'mikpe/erl_child_setup-android-breakage'Henrik Nord
* mikpe/erl_child_setup-android-breakage: erl_child_setup.c: fix Android breakage OTP-12751
2015-05-18Avoid timer wheel work in dirty schedulersSteve Vinoski
Dirty schedulers are limited in the type of work they can perform. Make sure they do not participate in any work or make any function calls related to timer wheels.
2015-05-18Map error logger warnings to warning messages by defaultRichard Carlsson
Also fix and document the broken +We option.
2015-05-13Revert "Prepare release"Henrik Nord
This reverts commit e09dd66dc4d89c62ddfd8c19791f9678d5d787c6.
2015-05-12Correct usage of sizeof() for pointer typesJesper Louis Andersen
Given some pointer *x, calling sizeof(x) will give us the size of the pointer (4/8 bytes) not the size fo the underlying dereferenced structure. Use coccinelle to search for these occurrences in the source code, and correct them one by one. In the case of erl_node_tables.c, the erts_snprintf() calls used a much too small buffer. - run_erl.c: Use the size of the signal type, not its pointer. - erl_node_tables.c: Use the size of the _BUFFER in erts_snprintf() to make sure we can use the full space.
2015-05-12Fix gen_tcp:shutdown/2 by making it asynchronousRory Byrne
If the driver queue is empty, or the user is requesting a 'read' shutdown, then the shutdown() syscall is performed synchronously, as per the old version of shutdown/2. However, if the user is requesting a 'write' or 'read_write' shutdown, and there is data in the driver queue for the socket, then the shutdown() syscall is delayed and handled asynchronously when the driver queue is written out. This version of shutdown solves a number of issues with the old version. The two main solutions it offers are: * It doesn't block when the TCP peer is idle or slow. This is the expected behaviour when shutdown() is called: the caller needs to be able to continue reading from the socket, not be prevented from doing so. * It doesn't truncate the output. The current version of gen_tcp:shutdown/2 will truncate any outbound data in the driver queue after about 10 seconds if the TCP peer is idle of slow. Worse yet, it doesn't even inform anyone that the data has been truncated: 'ok' is returned to the caller; and a FIN rather than an RST is sent to the TCP peer. For a detailed description of all the problems with the old version of shutdown, please see the EEP Light that was written to justify this patch.