aboutsummaryrefslogtreecommitdiffstats
path: root/erts/emulator
AgeCommit message (Collapse)Author
2018-05-16New process suspend implementation based on async signalingRickard Green
2018-05-16Teach HiPE to yield from receiveRickard Green
2018-05-07Merge pull request #1802 from michalmuskala/map-is-key-bifBjörn Gustavsson
Introduce is_map_key/2 guard BIF OTP-15037
2018-05-04Merge branch 'john/erts/fix-windows-symlinks/OTP-15062/ERL-615'John Högberg
* john/erts/fix-windows-symlinks/OTP-15062/ERL-615: Stop assuming that all NTFS reparse points are links
2018-05-03Stop assuming that all NTFS reparse points are linksJohn Högberg
This fixes a crash that would occur when using file:read_file_info/1 on a file with a non-link reparse point, which are commonly seen in RSS and OneDrive folders.
2018-04-29Introduce is_map_key/2 guard BIFMichał Muskała
This complements the `map_get/2` guard BIF introduced in #1784. Rationale. `map_get/2` allows accessing map fields in guards, but it might be problematic in more complex guard expressions, for example: foo(X) when map_get(a, X) =:= 1 or is_list(X) -> ... The `is_list/1` part of the guard could never succeed since the `map_get/2` guard would fail the whole guard expression. In this situation, this could be solved by using `;` instead of `or` to separate the guards, but it is not possible in every case. To solve this situation, this PR proposes a `is_map_key/2` guard that allows to check if a map has key inside a guard before trying to access that key. When combined with `is_map/1` this allows to construct a purely boolean guard expression testing a value of a key in a map. Implementation. Given the use case motivating the introduction of this function, the PR contains compiler optimisations that produce optimial code for the following guard expression: foo(X) when is_map(X) and is_map_key(a, X) and map_get(a, X) =:= 1 -> ok; foo(_) -> error. Given all three tests share the failure label, the `is_map_key/2` and `is_map/2` tests are optimised away. As with `map_get/2` the `is_map_key/2` BIF is allowed in match specs.
2018-04-27Merge branch 'sverker/lc-thread-exit-free-fix'Sverker Eriksson
* sverker/lc-thread-exit-free-fix: erts: Fix memory leak in lock checker at thread exit
2018-04-27Merge branch 'rickard/process_info/OTP-14966'Rickard Green
* rickard/process_info/OTP-14966: Fix scheduled process_info() 'status' request Fix handling of process-info requests in receive
2018-04-27Fix scheduled process_info() 'status' requestRickard Green
2018-04-27Merge branch 'siri/kernel/logger/OTP-13295'Siri Hansen
* siri/kernel/logger/OTP-13295: Add documentation of the built-in logger handlers Catch badarg in logger:get_format_depth/0 Add chars_limit option to logger_formatter Don't kill logger process until all other processes are dead Set call timeout for logger_server to infinity Update primary bootstrap Test cuddle for logger Update cth_log_redirect to a logger handler Start using logger internally in kernel and stdlib Remove error_logger process and add logger process Add logger
2018-04-26erts: Optimize monitor signal by message piggybackSverker Eriksson
If no message/signal is sent (to same destination) then monitor signal is flushed when process is scheduled out.
2018-04-26erts: Cleanup some codeSverker Eriksson
2018-04-26Fix handling of process-info requests in receiveRickard Green
2018-04-26Merge branch 'rickard/dirty-schedulers-test-fix/OTP-15046'Rickard Green
* rickard/dirty-schedulers-test-fix/OTP-15046: Do not run tests that conflicts with dirty schedulers test
2018-04-26erts: Fix memory leak in lock checker at thread exitSverker Eriksson
Leak introduced in 865ac3b740d9efa1a0583349929591c757300412.
2018-04-26Test cuddle for loggerSiri Hansen
2018-04-26Remove error_logger process and add logger processSiri Hansen
2018-04-26Merge branch 'rickard/process_info/OTP-14966'Rickard Green
* rickard/process_info/OTP-14966: Restore merge of signal queues in queue_messages() if main lock is held Fix message tracing
2018-04-25Do not run tests that conflicts with dirty schedulers testRickard Green
2018-04-25Restore merge of signal queues in queue_messages() if main lock is heldRickard Green
2018-04-25Merge branch 'map-get-bif' of git://github.com/michalmuskala/otpBjörn Gustavsson
* 'map-get-bif' of git://github.com/michalmuskala/otp: Introduce map_get guard-safe function OTP-15037
2018-04-25Merge branch 'lukas/erts/poll-thread/OTP-14346'Lukas Larsson
* lukas/erts/poll-thread/OTP-14346: erts: nif resource stop from poll-thread is a indirect call
2018-04-25Merge branch 'lukas/erts/dump_SUITE_fix'Lukas Larsson
* lukas/erts/dump_SUITE_fix: erts: Increase file read timeout for signal_abort test
2018-04-24Merge branch 'john/erts/fix-lcnt-toggle-test'John Högberg
* john/erts/fix-lcnt-toggle-test: Disregard locks that can't be toggled in lcnt_SUITE
2018-04-24Merge pull request #1790 from jhogberg/john/erts/more-alloc-info/OTP-14961John Högberg
Improve memory instrumentation OTP-15024 OTP-14961
2018-04-24Introduce map_get guard-safe functionMichał Muskała
Rationale Today all compound data types except for maps can be deconstructed in guards. For tuples we have `element/2` and for lists `hd/1` and `tl/1`. Maps are completely opaque to guards. This means matching on maps can't be abstracted into macros, which is often done with repetitive guards. It also means that maps have to be always selected whole from ETS tables, even when only one field would be enough, which creates a potential efficiency issue. This PR introduces an `erlang:map_get/2` guard-safe function that allows extracting a map field in guard. An alternative to this function would be to introduce the syntax for extracting a value from a map that was planned in the original EEP: `Map#{Key}`. Even outside of guards, since this function is a guard-BIF it is more efficient than using `maps:get/2` (since it does not need to set up the stack), and more convenient from pattern matching on the map (compare: `#{key := Value} = Map, Value` to `map_get(key, Map)`). Performance considerations A common concern against adding this function is the notion that "guards have to be fast" and ideally execute in constant time. While there are some counterexamples (`length/1`), what is more important is the fact that adding those functions does not change in any way the time complexity of pattern matching - it's already possible to match on map fields today directly in patterns - adding this ability to guards will niether slow down or speed up the execution, it will only make certain programs more convenient to write. This first version is very naive and does not perform any optimizations.
2018-04-24mikpe/erts-logger-integer-truncation/PR-1795/OTP-15032Lukas Larsson
fix integer truncation bugs in error logger path
2018-04-24erts: nif resource stop from poll-thread is a indirect callLukas Larsson
2018-04-23Merge branch 'sverker/valgrind-uninit-run_queue-ptr'Sverker Eriksson
* sverker/valgrind-uninit-run_queue-ptr: erts: Fix run queue pointer in proxy process
2018-04-23Merge branch 'sverker/ets-cherries/OTP-15031'Sverker Eriksson
* sverker/ets-cherries/OTP-15031: erts: Use table ref for select continuation erts: Refactor erl_db_hash next() into next_live() erts: Change wrong alloc_fnf
2018-04-23erts: Rewrite memory instrumentationJohn Högberg
This commit replaces the old memory instrumentation with a new implementation that scans carriers instead of wrapping erts_alloc/erts_free. The old implementation could not extract information without halting the emulator, had considerable runtime overhead, and the memory maps it produced were noisy and lacked critical information. Since the new implementation walks through existing data structures there's no longer a need to start the emulator with special flags to get information about carrier utilization/fragmentation. Memory fragmentation is also easier to diagnose as it's presented on a per-carrier basis which eliminates the need to account for "holes" between mmap segments. To help track allocations, each allocation can now be tagged with what it is and who allocated it at the cost of one extra word per allocation. This is controlled on a per-allocator basis with the +M<S>atags option, and is enabled by default for binary_alloc and driver_alloc (which is also used by NIFs).
2018-04-23erts: Increase file read timeout for signal_abort testLukas Larsson
2018-04-23Fix message tracingRickard Green
2018-04-23Merge branch 'lukas/erts/seq_tracer_nif/OTP-15029'Lukas Larsson
* lukas/erts/seq_tracer_nif/OTP-15029: Fix seq_trace erl_tracer bug Fix makefile mkdir warning
2018-04-21fix integer truncation bugs in error logger pathMikael Pettersson
Sending a large term to the error logger has two problems related to the size and sign of the variables used to represent lengths: - the API functions (erts_send_error_term_to_logger() et al) perform an unchecked narrowing conversion from size_t to int when passing dsbufp->str_len to the internal functions; this may both truncate the length and make it negative - do_send_term_to_logger() and do_send_to_logger() multiply the int-typed length by 2 before widening it to Uint and adding a few more values; the intermediate product may overflow causing loss of high bits and a change of sign; if the intermediate product is negative the final size will be an extremely large positive value The end result is that the computed buffer size can be arbitrarily wrong, either too small or too large. While reviewing this code I also found and fixed a potential narrowing bug in erts_set_hole_marker().
2018-04-20erts: Use table ref for select continuationSverker Eriksson
and not the name. For more sane named table semantics. Applies to both select/1 continuation and trap context.
2018-04-20erts: Refactor erl_db_hash next() into next_live()Sverker Eriksson
where argument 'list' is the first candidate (not list->next). Also simplified db_first_hash().
2018-04-20erts: Change wrong alloc_fnfSverker Eriksson
There is no NULL check here !?
2018-04-20Merge PR-1792 from saleyn/float_to_listSverker Eriksson
Optimize performance of float_to_list/2
2018-04-20erts: Fix run queue pointer in proxy processSverker Eriksson
proc->run_queue detected as uninitialized by valgrind but seems harmless in practice as it's not used for proxy processes. "Bug" introduced in OTP-17 by ca0425c6ff85262bc15367f5fd9cbc51cde52b20 and made worse (but still harmless) in master for OTP-21 at fbb10ebc4a37555c7ea7f99e14286d862993976a.
2018-04-19Optimize performance of float_to_list/2Serge Aleynikov
2018-04-19Make stacktraces consistent when backtrace_depth is 0Björn Gustavsson
It is allowed to set the backtrace depth to 0, but when an exception is catched the stacktrace will still contain one element: 1> erlang:system_flag(backtrace_depth, 0). 8 2> catch error(badarg). {'EXIT',{badarg,[{shell,apply_fun,3, [{file,"shell.erl"},{line,908}]}]}} However, when an exception is raised using `erlang:raise/3`, there will be no elements in the stacktrace: 3> catch erlang:raise(error, badarg, [{fake,name,[arg],[]}]). {'EXIT',{badarg,[]}} Since the `error_handler` module uses `erlang:raise/3` to raise an exception when an undefined function is called, there will not be any stacktrace when calling an undefined function: 4> catch undef_module:undef_name(some_argument). {'EXIT',{undef,[]}} Fix this inconsistency by changing `erlang:raise/3` so that it always includes one element in the stacktrace: 3> catch erlang:raise(error, badarg, [{fake,name,[arg],[]}]). {'EXIT',{badarg,[{fake,name,[arg],[]}]}} 4> catch undef_module:undef_name(some_argument). {'EXIT',{undef,[{undef_module,undef_name,[some_argument],[]}]}}
2018-04-18Merge branch 'sverker/lock-check-matrix'Sverker Eriksson
2018-04-18erts: Fix rare deadlock in realloc when +ramv is enabledJohn Högberg
OTP-15024
2018-04-16Disregard locks that can't be toggled in lcnt_SUITEJohn Högberg
2018-04-16erts: Keep track of which NIF a scheduler is executingJohn Högberg
This may be of interest in crash dumps and allows the upcoming allocation tagging feature to track allocations on a per-NIF basis. Note that this is only updated when user code calls a NIF; it's not altered when the emulator calls NIFs during code upgrades or tracing.
2018-04-16erts: Always keep a copy of driver names as an atomJohn Högberg
2018-04-16erts: Update esdp->current_port on immediate port callsJohn Högberg
This increases the accuracy of crash dumps and the upcoming allocation tagging feature.
2018-04-16Merge branch 'rickard/process_info/OTP-14966'Rickard Green
* rickard/process_info/OTP-14966: Fix deadlock in HiPE gc after receive
2018-04-16Fix deadlock in HiPE gc after receiveRickard Green