aboutsummaryrefslogtreecommitdiffstats
path: root/erts
AgeCommit message (Collapse)Author
2019-04-10Decentralized counters for ETS ordered_set with write_concurrencyKjell Winblad
Previously, all ETS tables used centralized counter variables to keep track of the number of items stored and the amount of memory consumed. These counters can cause scalability problems (especially on big NUMA systems). This commit adds an implementation of a decentralized counter and modifies the implementation of ETS so that ETS tables of type ordered_set with write_concurrency enabled use the decentralized counter. [Experiments][1] indicate that this change substantially improves the scalability of ETS ordered_set tables with write_concurrency enabled in scenarios with frequent `ets:insert/2` and `ets:delete/2` calls. The new counter is implemented in the module erts_flxctr (`erts_flxctr.h` and `erts_flxctr.c`). The module has the suffix flxctr as it contains the implementation of a flexible counter (i.e., counter instances can be configured to be either centralized or decentralized). Counters that are configured to be centralized are implemented with a single counter variable which is modified with atomic operations. Decentralized counters are spread over several cache lines (how many can be configured with the parameter `+dcg`). The scheduler threads are mapped to cache lines so that there is no single point of contention when decentralized counters are updated. The thread progress functionality of the Erlang VM is utilized to implement support for linearizable snapshots of decentralized counters. The snapshot functionality is used by the `ets:info/1` and `ets:info/2` functions. [1]: http://winsh.me/ets_catree_benchmark/flxctr_res.html
2019-04-10Merge pull request #2204 from jhogberg/john/erts/optimize-arith-ops/OTP-15740John Högberg
erts: Optimize arithmetic ops using overflow intrinsics
2019-04-10Merge branch 'john/merge-re-replace-fix'John Högberg
* john/merge-re-replace-fix: erts: Skip ERTS_NOINLINE on non-GCC-compatible compilers cleanup: beam_emu.c: use ERTS_NOINLINE instead of NOINLINE stdlib: fix re:replace on LTO builds
2019-04-09erts: Fix dist sequence yield state allocator typeLukas Larsson
The wrong allocator type was given to erts_free.
2019-04-09erts: Fix driver_SUITE:use_fallback_pollsetLukas Larsson
We don't check for io errors as the pollset will be unstable.
2019-04-09erts: Yield when exiting/free process is suspended by deLukas Larsson
Complementory fix to 922fd355831575965
2019-04-09erts: Fix alignment of ErtsDistExternal on sparcLukas Larsson
The edep->data values need to be 64-bit aligned as it contains 64-bit values.
2019-04-09erts: Fix incorrect gcc may be used init warningLukas Larsson
2019-04-09erts: Skip large dist testcases on machines with little memoryLukas Larsson
2019-04-09erts: Fix lock order violation in system_info(procs)Lukas Larsson
2019-04-09erts: Fix buffer alignment bug in alloc_dist_obufLukas Larsson
Cannot do unaligned word writes on sparc!
2019-04-09erts: Fix build of dist_ctrl_get_data return termLukas Larsson
2019-04-09Merge branch 'maint'Lukas Larsson
2019-04-09erts: Fix sys_driver to correctly deselect fdsLukas Larsson
Incorrect precidence rules made the driver never deselect the given fds.
2019-04-08Merge branch 'bjorn/erts/cuddle-with-tests'Björn Gustavsson
* bjorn/erts/cuddle-with-tests: statistics_SUITE: Make wall_clock_update/1 more tolerant
2019-04-08erts: Optimize arithmetic ops using overflow intrinsicsJohn Högberg
2019-04-08erts: Skip ERTS_NOINLINE on non-GCC-compatible compilersJohn Högberg
__declspec(noinline) works fine on MSVC but requires us to place the macro before a function rather than after, which in turn causes early versions of GCC to puke since they only accept __attribute__ at the end of a function declaration. Since this is a new macro that previously only saw use in beam_emu, I figured it's easiest to leave it disabled on MSVC.
2019-04-05Optimize the i_minus instructionBjörn Gustavsson
Measurements show that i_minus instructions (subtraction) are frequent enough to warrant creating specialized i_minus instructions. Thanks to José Valim for doing instruction counting on Elixir code.
2019-04-05Merge branch 'lukas/erl_docgen/add-internal-docs'Björn Gustavsson
OTP-15715 * lukas/erl_docgen/add-internal-docs: cerl_clauses: Fix broken link Fix minor typos Add CountingInstructions to the internal documentation GarbageCollection.md: Fix links to references emd2exml: Remove paragraphs around image tags emd2exml: Clean emphasis markers in link texts emd2exml: Allow short names of anchors for internal links Include figures erts: Fix more internal docs xmllint compiler: Add internal docs section erl_docgen: Add new internal docs chapter to docs
2019-04-05Merge branch 'bjorn/doc/fix-broken-links'Björn Gustavsson
* bjorn/doc/fix-broken-links: Fix broken links
2019-04-05Merge remote-tracking branch 'trofi/master' into john/merge-re-replace-fixJohn Högberg
PR 2194, stdlib: fix re:replace on LTO builds * trofi/master: cleanup: beam_emu.c: use ERTS_NOINLINE instead of NOINLINE stdlib: fix re:replace on LTO builds
2019-04-04Fix broken linksBjörn Gustavsson
A multitude of broken links were found when a bug was fixed in the link checker.
2019-04-04Fix minor typosBjörn Gustavsson
2019-04-04Add CountingInstructions to the internal documentationBjörn Gustavsson
2019-04-04GarbageCollection.md: Fix links to referencesBjörn Gustavsson
2019-04-04Include figuresBjörn Gustavsson
2019-04-04erts: Fix more internal docs xmllintLukas Larsson
2019-04-04erl_docgen: Add new internal docs chapter to docsLukas Larsson
2019-04-03lcnt_SUITE: Ignore socket-NIF locks in toggle testsJohn Högberg
2019-04-02Merge branch 'sverker/enable-big-creation/OTP-15603'Sverker Eriksson
AGAIN * sverker/enable-big-creation/OTP-15603: erts: Document new EPMD response ALIVE2_X_RESP erl_interface: Support 32-bit creation local cnode jinterface: Remove old encoding of pid,port,refs
2019-04-02Merge branch 'john/erts/cuddle-efile-suite'John Högberg
* john/erts/cuddle-efile-suite: efile_SUITE: Fix proc_zero_sized_files on systems with empty /proc
2019-04-01Merge branch 'maint'Sverker Eriksson
* maint: Updated OTP version Prepare release # Conflicts: # OTP_VERSION # make/otp_version_tickets_in_merge
2019-04-01efile_SUITE: Fix proc_zero_sized_files on systems with empty /procJohn Högberg
This test would fail on some versions of FreeBSD because /proc was present but empty. This commit makes the test fail if and only if a known zero-sized file is found and turns out to be empty empty.
2019-04-01Merge branch 'john/kernel/cuddle-file-suite'John Högberg
* john/kernel/cuddle-file-suite: file_name_SUITE: Darwin 18 broke support for 'icky' names file_SUITE: Ignore EILSEQ on Darwin in +fnu/latin1 test
2019-03-29cleanup: beam_emu.c: use ERTS_NOINLINE instead of NOINLINESergei Trofimovich
Signed-off-by: Sergei Trofimovich <[email protected]>
2019-03-29stdlib: fix re:replace on LTO buildsSergei Trofimovich
Fabio Coatti reported elixir build failure in https://bugs.gentoo.org/681778. The minimal reproducer looks like that (from otp git tree): $ ./configure CFLAGS='-O2 -flto' LDFLAGS='-O2 -flto=8' $ make $ ERL_TOP=$PWD \ PATH=$ERL_TOP/bin:$PATH \ \ bin/erl \ \ -noshell -eval 're:replace("a","b","c",[{return,list}]).' \ -s erlang halt {"init terminating in do_boot",{badarg,[{re,replace,["a","b","c",[{return,list}]], [{file,"re.erl"},{line,362}]}, {erl_eval,do_apply,6,[{file,"erl_eval.erl"},{line,680}]}, {init,start_it,1,[]}, {init,start_em,1,[]}, {init,do_boot,3,[]}]}} init terminating in do_boot ({badarg,[{re,replace,[[_],[_],[_],[_]],[{_},{_}]}, {erl_eval,do_apply,6,[{_},{_}]},{init,start_it,1,[]},{init,start_em,1,[]},{init,do_boot,3,[]}]}) Crash dump is being written to: erl_crash.dump...done The failure happens in libpcre2 where stack overflow is mis-identified at function entry of erts_pcre_compile2() compile_regex() if (PUBL(stack_guard) != NULL && PUBL(stack_guard)()) { *errorcodeptr= ERR85; return FALSE; } The stack "overflow" detection happens in thr_wrapper() ethr_set_stacklimit__() because the stack usage code relies on the fact that ethr_set_stacklimit__() and similar functions don't get inlined into callers for stack growth measurement. Before the change inlining avoidance was achieved by putting functions into standalone translation units. LTO makes this technique inefficient. The change marks functions explicitly as __attribute__((__noinline__)) on gcc. Reported-by: Fabio Coatti Bug: https://bugs.gentoo.org/681778 Signed-off-by: Sergei Trofimovich <[email protected]>
2019-03-29Prepare releaseErlang/OTP
2019-03-29Merge branch 'sverker/process_info-reductions-fix/OTP-15709' into maint-21Erlang/OTP
* sverker/process_info-reductions-fix/OTP-15709: erts: Fix bug in process_info(reductions) erts: Use ptr_val for pointer in gc msg copy
2019-03-29Merge branch 'maint'Sverker Eriksson
2019-03-29Merge branch 'sverker/process_info-reductions-fix/OTP-15709' into maintSverker Eriksson
* sverker/process_info-reductions-fix/OTP-15709: erts: Fix bug in process_info(reductions) erts: Use ptr_val for pointer in gc msg copy
2019-03-29Fix out of memory bug in the implementation of mapsKjell Winblad
This commit fixes a bug that could cause a crash or memory usage to grow until the machine ran out of memory when adding a key-value pair to a map. This could happen when inserting a new key-value pair with a key K1 containing a binary B1 into a map M having a key K2 with a binary B2 if the following conditions were met: * size(B1) >= 4294967296 * size(B2) >= 4294967296 * size(M) >= 32 * (size(B1) rem 4294967296) == (size(B2) rem 4294967296) * the first (size(B1) rem 4294967296) bytes are the same both in B1 and B2 * substituting B1 in K1 with B2 would result in a term with the same value as K2 The root cause of the bug is that the map implementation only hashed the first (X modulo 4294967296) bytes of binaries so that different binaries could get hashed to the same hash value independently of the hash seed.
2019-03-29Merge pull request #2169 from llelf/shell-ctrl+rbracketBjörn Gustavsson
tty: document ^] keystroke
2019-03-29file_SUITE: Ignore EILSEQ on Darwin in +fnu/latin1 testJohn Högberg
There's no way to create the file in question, so we may as well skip the test.
2019-03-28erts: Fix bug in process_info(reductions)Sverker Eriksson
returning incorrect result as * current process might not be RUNNING in which case REDS_IN is actually used as def_arg_reg[5] * FCALLS might not have been swapped out * the SAVED_CALLS case was wrong and returned number of reds left
2019-03-28erts: Use ptr_val for pointer in gc msg copySverker Eriksson
Cherry-pick: 3d7d66b84b69f3eb42e8bcd8dc510b6676e31cac Without this fix asserts would trigger in debug build but nothing else would break.
2019-03-28Merge branch 'john/erts/remove-destructive-bs_get_binary2/ERL-901'John Högberg
* john/erts/remove-destructive-bs_get_binary2/ERL-901: erts: Remove unsafe bs_get_binary2 optimization from loader
2019-03-27erts: Remove unsafe bs_get_binary2 optimization from loaderJohn Högberg
A load-time optimization assumed that match contexts had no further uses when a bs_get_binary2 overwrote the match context's register, and figured it would be safe to reuse the match context's memory for the resulting binary. This is no longer safe as of OTP 22, as a match context may be reused after being passed to another function.
2019-03-27Merge branch 'john/erts/seq_trace-sigq-clearing/OTP-15704'John Högberg
* john/erts/seq_trace-sigq-clearing/OTP-15704: erts: Handle all signal types when clearing seq_trace token
2019-03-27erts: Handle all signal types when clearing seq_trace tokenJohn Högberg
We walk through all signal queues in the system when clearing seq_trace, so we must be prepared to encounter all kinds of signals.
2019-03-27Fix unsafe optimization made by the loaderBjörn Gustavsson
Fix the unsafe load-time optimization introduced in 07bdbb3a1edc. https://bugs.erlang.org/browse/ERL-899