aboutsummaryrefslogtreecommitdiffstats
AgeCommit message (Collapse)Author
2019-02-11Merge pull request #2134 from bjorng/bjorn/compiler/propagate-noneBjörn Gustavsson
beam_ssa_type: Propagate the 'none' type from calls
2019-02-11beam_ssa_type: Propagate the 'none' type from callsBjörn Gustavsson
Consider this pseudo code: f(...) -> Val = case Expr of ... -> ... ; ... -> ... ; ... -> my_abort(something_went_wrong) end, %% Here follows code that uses Val. . . . my_abort(Reason) -> throw({error,Reason}). The first two clauses in the case will probably provide some information about the type of the variable `Var`, information that would be useful for optimizing the code that follows the case. However, the third clause would ruin everything. The call to `my_abort/1` could return anything, and thus `Val` could also have any type. 294d66a295f6 introduced module-level type analysis, which will in general keep track of the return type of a local function call. However, it does not improve the optimization for this specific function. When a function never returns, that is, when its type is `none`, it does not propagate the `none` type, but instead pretends that the return type is `any`. This commit extends the handling of functions that don't return to properly handle the `none` type. Any instructions that directly follows the function that does not return will be discarded, and the call will be rewritten to a tail-recursive call. For this specific example, it means that the type for `Val` deduced from the first two clauses will be retained and can be used for optimizing the code after the case.
2019-02-07Merge pull request #2135 from bjorng/bjorn/compiler/optimize-dominatorsBjörn Gustavsson
Optimize ssa_opt_sink for huge functions
2019-02-06Merge branch 'maint'Rickard Green
* maint: Introduce ei_init() Fix bug in ei_accept_tmo Fix build of erl_interface on BSD
2019-02-06Merge branch 'rickard/ei-ext-maint/OTP-15442' into maintRickard Green
* rickard/ei-ext-maint/OTP-15442: Introduce ei_init() Fix bug in ei_accept_tmo Fix build of erl_interface on BSD
2019-02-06Merge pull request #2129 from hanssv/crypto/add_blake2_supportHans Nilsson
Add hash function BLAKE2 to crypto:hash/hmac OTP-15564
2019-02-06Merge branch 'maint'Sverker Eriksson
2019-02-06Merge branch 'sverker/fix-compile-warning' into maintSverker Eriksson
* sverker/fix-compile-warning: erts: Fix compiler warning for IS_SSMALL
2019-02-06Merge PR-2127 from sverker/erts/enif_make_monitor_term/OTP-15362Sverker Eriksson
Add enif_make_monitor_term
2019-02-06Optimize ssa_opt_sink for huge functionsBjörn Gustavsson
The ssa_opt_sink optimization of beam_ssa_opt could get very slow for certain huge functions. 9a190cae9bd7 partly addressed this issue by terminating the optimization early if there happened to be no get_tuple_element instructions at all in the function. This commit addresses the issue more directly by making the dominator calculation in beam_ssa:dominators/1 more efficient. The same algorithm as before is used, but it is implemented in a more efficient way based on the ideas in "A Simple, Fast Dominance Algorithm" (http://www.hipersoft.rice.edu/grads/publications/dom14.pdf). As well as being more efficient, the new implementation also gives an explicit representation of the dominator tree, which makes it possible to simplify and optimize the ssa_opt_sink optimization.
2019-02-06Merge pull request #2132 from jhogberg/john/compiler/tuple-element-typesJohn Högberg
Keep track of tuple elements in type optimization pass
2019-02-05Merge branch 'rickard/ei-ext/OTP-15442' into rickard/ei-ext-maint/OTP-15442Rickard Green
* rickard/ei-ext/OTP-15442: Introduce ei_init() Fix bug in ei_accept_tmo Fix build of erl_interface on BSD
2019-02-05Introduce ei_init()Rickard Green
You previously had to initialize ei via erl_init() which implied that you were forced to link against the erl_interface even when not using it (besides initializing ei).
2019-02-05Add blake2 datatype to crypto.xmlHans Svensson
2019-02-05Merge branch 'maint'Ingela Anderton Andin
Conflicts: lib/ssl/doc/src/ssl.xml lib/ssl/src/ssl.erl lib/ssl/src/ssl_cipher_format.erl lib/ssl/src/tls_handshake.erl
2019-02-05Merge branch 'ingela/ssl/alert-return/OTP-15423' into maintIngela Anderton Andin
* ingela/ssl/alert-return/OTP-15423: ssl: Use specs to generate type documentation ssl: Enhance error handling
2019-02-05Merge branch 'maint'Ingela Anderton Andin
2019-02-05Merge branch 'ingela/ssl/dtls-match-error/OTP-15561' into maintIngela Anderton Andin
* ingela/ssl/dtls-match-error/OTP-15561: ssl: Remove unintended match
2019-02-05ssl: Use specs to generate type documentationIngela Anderton Andin
2019-02-05Merge branch 'maint'Lukas Larsson
2019-02-05Merge branch 'lukas/erts/cerl-etp-rr-support' into maintLukas Larsson
* lukas/erts/cerl-etp-rr-support: erts: Expand rr support in cerl and etp
2019-02-05erts: Expand rr support in cerl and etpLukas Larsson
2019-02-05beam_ssa_type: Track the types of tuple elementsJohn Högberg
Prior to 294d66a295f6c2101fe3c2da630979ad4e736c08 there wasn't much point to keeping track of tuple element types; they were only known when we had inserted or extracted values from a tuple, and in neither case was it likely that we'd extract the same values again. It makes a lot more sense to do so now that type optimizations are applied across functions; if we return a tuple it's very likely that its elements will be extracted soon after, and knowing their types lets us eliminate more type checks. Co-authored-by: Björn Gustavsson <[email protected]>
2019-02-04Fix bug in ei_accept_tmoRickard Green
2019-02-04Merge branch 'maint'Péter Dimitrov
* maint: Updated OTP version Prepare release
2019-02-04Merge branch 'maint-21' into maintPéter Dimitrov
* maint-21: Updated OTP version Prepare release
2019-02-04beam_ssa_type: Infer types based on the result of #b_switch{}John Högberg
2019-02-04beam_ssa_opt: Fix function name printing in sub-pass crash dumpJohn Högberg
2019-02-04beam_validator: Remove unreachable case clauseJohn Högberg
2019-02-04Merge pull request #2126 from bjorng/bjorn/compiler/compilation-speedBjörn Gustavsson
Reduce compilation times
2019-02-04Merge branch 'bjorn/compiler/cuddle-with-tests'Björn Gustavsson
* bjorn/compiler/cuddle-with-tests: inline_SUITE: Don't start a slave node Correct test_lib:is_cloned_mod/1 sys_core_fold_lists: Propagate annotations in expansion of lists functions Parallelize compile_SUITE:bc_options/1 Remove the optimized_guard/1 test case
2019-02-04Add hash function BLAKE2 to crypto:hash/hmacHans Svensson
Adds two hash functions blake2b and blake2s (64 bit hash and 32 bit hash respectively). These are modern and standard hash functions used in blockchains and encrypted communication protocols (e.g. Noise - http://www.noiseprotocol.org/). The hash functions are available in OpenSSL since version 1.1.1. Also add test cases and mention in documentation.
2019-02-04Merge branch 'hans/crypto/fixes/OTP-14732'Hans Nilsson
* hans/crypto/fixes/OTP-14732: otp_test_engine.c fixes crypto: Fix compilation < 1.0.0
2019-02-04Merge branch 'maint'Péter Dimitrov
* maint: Fix typo in erlang.xml inets: Ignore bracket option in format_address Change-Id: I8a215d3872ae74e08d7a17b70ba53535947c032f
2019-02-04otp_test_engine.c fixesHans Nilsson
2019-02-04crypto: Fix compilation < 1.0.0Hans Nilsson
2019-02-04Merge pull request #2095 from hogand/crypto/revamp-filesHans Nilsson
crypto: revamp C code [WIP] OTP-14732
2019-02-04Merge branch 'peterdmv/inets/fix-http-client/ERIERL-289/OTP-15554' into maintPéter Dimitrov
* peterdmv/inets/fix-http-client/ERIERL-289/OTP-15554: inets: Ignore bracket option in format_address Change-Id: If3f19325edb8f8cc1ced717aa125658c00438b70
2019-02-04ssl: Enhance error handlingIngela Anderton Andin
2019-02-04Merge pull request #2128 from amatalai/fix-typo-in-erlang-xmlLukas Larsson
Fix typo in erlang.xml
2019-02-04Fix typo in erlang.xmlTobiasz
2019-02-01erts: Add enif_make_monitor_termSverker Eriksson
2019-02-01ssl: Remove unintended matchIngela Anderton Andin
2019-02-01inline_SUITE: Don't start a slave nodeBjörn Gustavsson
A long time ago there was a good idea to run compiled code in a slave node. Nowadays, not so much.
2019-02-01Correct test_lib:is_cloned_mod/1Björn Gustavsson
test_lib:is_cloned_mod(inline_SUITE) would return true.
2019-02-01sys_core_fold_lists: Propagate annotations in expansion of lists functionsBjörn Gustavsson
There could be a warning with a `no_file` atom instead of filename and line number.
2019-02-01Parallelize compile_SUITE:bc_options/1Björn Gustavsson
2019-02-01Remove the optimized_guard/1 test caseBjörn Gustavsson
With the new SSA code passes, the optimized_guard/1 test case has become really bad at finding unnecessary `and` and `or` instructions.
2019-02-01Make helper functions tail-recursiveBjörn Gustavsson
Two helper functions in beam_ssa_opt and beam_ssa_dead are body-recursive for no good reason. While at it, add some clarifying comments to the functions.
2019-02-01Optimize beam_ssa:def_used/2Björn Gustavsson
beam_ssa:def_used/2 is used by beam_ssa_pre_codegen when reserving Y registers. Do the following optimizations: * Use an ordset instead of a gb_set. When the only operation performed on a set is union/2, an ordset will usually be faster, especially when the result is an ordset. * Use a cerl_set instead of a gb_set for the set of all possible predecessors. cerl_sets is usually faster than gb_sets.