aboutsummaryrefslogtreecommitdiffstats
path: root/lib
AgeCommit message (Collapse)Author
2017-11-30Add a mutable binary buffer type (prim_buffer)John Högberg
2017-11-30Update syntax_tools to support the stacktrace variableBjörn Gustavsson
2017-11-30Use the new syntax in more test suitesBjörn Gustavsson
2017-11-30Add syntax in try/catch to retrieve the stacktrace directlyBjörn Gustavsson
This commit adds a new syntax for retrieving the stacktrace without calling erlang:get_stacktrace/0. That allow us to deprecate erlang:get_stacktrace/0 and ultimately remove it. The problem with erlang:get_stacktrace/0 is that it can keep huge terms in a process for an indefinite time after an exception. The stacktrace can be huge after a 'function_clause' exception or a failed call to a BIF or operator, because the arguments for the call will be included in the stacktrace. For example: 1> catch abs(lists:seq(1, 1000)). {'EXIT',{badarg,[{erlang,abs, [[1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20|...]], []}, {erl_eval,do_apply,6,[{file,"erl_eval.erl"},{line,674}]}, {erl_eval,expr,5,[{file,"erl_eval.erl"},{line,431}]}, {shell,exprs,7,[{file,"shell.erl"},{line,687}]}, {shell,eval_exprs,7,[{file,"shell.erl"},{line,642}]}, {shell,eval_loop,3,[{file,"shell.erl"},{line,627}]}]}} 2> erlang:get_stacktrace(). [{erlang,abs, [[1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22, 23,24|...]], []}, {erl_eval,do_apply,6,[{file,"erl_eval.erl"},{line,674}]}, {erl_eval,expr,5,[{file,"erl_eval.erl"},{line,431}]}, {shell,exprs,7,[{file,"shell.erl"},{line,687}]}, {shell,eval_exprs,7,[{file,"shell.erl"},{line,642}]}, {shell,eval_loop,3,[{file,"shell.erl"},{line,627}]}] 3> We can extend the syntax for clauses in try/catch to optionally bind the stacktrace to a variable. Here is an example using the current syntax: try Expr catch C:E -> Stk = erlang:get_stacktrace(), . . . In the new syntax, it would look like: try Expr catch C:E:Stk -> . . . Only a variable (not a pattern) is allowed in the stacktrace position, to discourage matching of the stacktrace. (Matching would also be expensive, because the raw format of the stacktrace would have to be converted to the cooked form before matching.) Note that: try Expr catch E -> . . . is a shorthand for: try Expr catch throw:E -> . . . If the stacktrace is to be retrieved for a throw, the 'throw:' prefix must be explicitly included: try Expr catch throw:E:Stk -> . . .
2017-11-30erl_parse: Use a new pat_expr rule for pattern expressionsBjörn Gustavsson
In the grammar file for the Erlang language, patterns are parsed as expressions. erl_lint will then weed out expressions that are not legal patterns. The rule sharing causes problems if we were to introduce new syntax, for example a pattern followed by a ':'. There would be a shift/reduce conflict, and a pattern followed by a ':' would be parsed as a remote call. Introduce a new pat_expr rule to express exactly the subset of expressions that is allowed in pattern. Note: For the moment, we must allow full expressions in case clauses to keep 'merl' working.
2017-11-30Stop trying to maximize the use of x(0)Björn Gustavsson
X register 0 used to be mapped to a hardware register, and therefore faster than the other registers. Because of that, the compiler tried to use x(0) as much as possible as a temporary register. That was changed a few releases ago. X register 0 is now placed in the array of all X registers and has no special speed advantage compared to the other registers. Remove the code in the compiler that attempts to use x(0) as much as possible. As a result, the following type of instruction will be much less frequent: {put_list,Src,{x,0},{x,0}} Instead, the following type of instruction will be more frequent: {put_list,Src,{x,X},{x,X}} (Where X is an arbitrary X register.) Update the runtime system to specialize that kind of put_list instruction.
2017-11-30Merge branch 'maint'Dan Gudmundsson
* maint: Fix type create_option() in mnesia
2017-11-30Merge pull request #1647 from aboroska/fix-mnesia-type-create_optionDan Gudmundsson
Fix type create_option() in mnesia
2017-11-30Clean up collection of basic blocksBjörn Gustavsson
2017-11-30Merge branch 'bjorn/compiler/coverage'Björn Gustavsson
* bjorn/compiler/coverage: v3_codegen: Remove check of operand for bs_context_to_binary beam_jump: Eliminate a repeated clause beam_asm: No longer allow iolists as contents in chunk/2 beam_utils_SUITE: Cover more lines in beam_utils guard_SUITE: Add a test case to cover beam_dead:turn_op/1 Recognize 'nil' as a literal in beam_utils:bif_to_test/3 Cover more code in beam_bsm:btb_opt_1/3 Add test to cover a line in v3_kernel:opt_single_valued/3
2017-11-30Merge branch 'maint'Dan Gudmundsson
* maint: Avoid falling measurements testcases on slow machines stdlib: string optimize special case for ASCII stdlib: Minor unicode_util opts
2017-11-30Merge branch 'dgud/stdlib/optimize-string-lists' into maintDan Gudmundsson
* dgud/stdlib/optimize-string-lists: Avoid falling measurements testcases on slow machines stdlib: string optimize special case for ASCII stdlib: Minor unicode_util opts OTP-14670
2017-11-29Minor grammar tweak in ETS documentationIsaac Whitfield
2017-11-29Avoid falling measurements testcases on slow machinesDan Gudmundsson
2017-11-29stdlib: string optimize special case for ASCIIDan Gudmundsson
Avoid unicode_util module call for ASCII strings
2017-11-29stdlib: Minor unicode_util optsDan Gudmundsson
Exit early for Latin-1
2017-11-29Merge branch 'maint'Ingela Anderton Andin
2017-11-29Merge branch 'ingela/ssl/ERL-521/OTP-14794' into maintIngela Anderton Andin
* ingela/ssl/ERL-521/OTP-14794: ssl: Make sure all possible data is delivered
2017-11-29Stop checking DNS name for SNIRaimo Niskanen
2017-11-29Merge branch 'maint'Ingela Anderton Andin
Conflicts: lib/ssl/src/ssl_connection.erl
2017-11-29Merge branch 'ingela/ssl/cleaner-code' into maintIngela Anderton Andin
* ingela/ssl/cleaner-code: ssl: Add gracefullness to dtls code ssl: Remove old softupgrade code ssl: Use genstamtem properly
2017-11-29v3_codegen: Remove check of operand for bs_context_to_binaryBjörn Gustavsson
The bs_context_to_binary instruction only allows a register operand. v3_codegen has a test to ensure that the operand is a register. That test is no longer necessary. There used to be a possibility that optimizations in sys_core_fold and the inliner could change the operand for bs_context_to_binary to a binary literal. Since 09112806c15a81b that can no longer happen, because no more optimizations are run after the introduction of the bs_context_to_binary instruction.
2017-11-28ssl: Align timing just in caseIngela Anderton Andin
2017-11-28ssl: Make sure all possible data is deliveredIngela Anderton Andin
2017-11-28add hibernate_after to the gen options type specTristan Sloughter
2017-11-28stdlib: Do not check base64 input more than neededHans Bolinder
Often the decode functions return a function_clause error, but not always, and the errors have not been consistent between modifications of the base64 module. Now the errors are returned as they happen--no attempt to make them look nice is done. The alternative, to ensure that, for example, {badarg, Culprit} is always returned upon bad input, was deemed pointless.
2017-11-28stdlib: Minor optimization of base64Hans Bolinder
Using /bits instead of /binary is faster when constructing binaries.
2017-11-28stdlib: Use binary_to_list in base64 when it is fasterHans Bolinder
It is sometimes faster to use binary_to_list on input, at least when not considering time for garbage collections. The three functions all return a list, and the temporary list created by binary_to_list should be reclaimed very fast. The implementation from before is thus kept.
2017-11-28stdlib: Optimize base64 functionsHans Bolinder
A few test cases with zeroes are added. They were not handled correctly before. The access of DECODE_MAP is moved into the inlined function b64d, for symmetry. The function b64e is also inlined. The speed-up is small, but measurable. Note: encode(List), decode(List) and mime_decode(List) no longer call list_to_binary. This can break code that calls the functions with I/O-lists as input.
2017-11-28Merge branch 'maint'Hans Bolinder
* maint: dialyzer: Extend the map implementation's handling of ?unit dialyzer: Use string:find() instead of string:str()
2017-11-28Merge branch 'hasse/dialyzer/maps_anon_fun/OTP-14743' into maintHans Bolinder
* hasse/dialyzer/maps_anon_fun/OTP-14743: dialyzer: Extend the map implementation's handling of ?unit dialyzer: Use string:find() instead of string:str()
2017-11-28Merge branch 'maint'Hans Nilsson
* maint: public_key: Fix failing pkix_verify_hostname_subjAltName_IP TC
2017-11-28Merge branch 'hans/ssh/public_key/fix_bad_cert_merge' into maintHans Nilsson
* hans/ssh/public_key/fix_bad_cert_merge: public_key: Fix failing pkix_verify_hostname_subjAltName_IP TC
2017-11-28inets: Fix httpc path handlingPéter Dimitrov
Fixed handling of URL paths that contain space (%20) characters. Change-Id: Ic09691e35c70ee005af7637a6ec3ae588097bec1
2017-11-27Fix type create_option() in mnesiaAndras Boroska
The create_option() type in mnesia lacks the user_properties field which is used by mnesia in various places.
2017-11-27Merge branch 'maint'Dan Gudmundsson
* maint: Update HOWTO with wxWidgets build examples https should be mapped to url and not seealso Update kernel appup for OTP-20.2
2017-11-27Merge branch 'dgud/erl_docgen/https2url' into maintDan Gudmundsson
* dgud/erl_docgen/https2url: Update HOWTO with wxWidgets build examples https should be mapped to url and not seealso
2017-11-27Merge branch 'siri/appups-20.2' into maintSiri Hansen
* siri/appups-20.2: Update kernel appup for OTP-20.2
2017-11-27public_key: Fix failing pkix_verify_hostname_subjAltName_IP TCHans Nilsson
Was due to an unresolved merge conflict in a453532fe31eecd8c39b5035702ee20b402fd687
2017-11-27beam_jump: Eliminate a repeated clauseBjörn Gustavsson
This clause seems to have been introduced in cac51274eb9a.
2017-11-27beam_asm: No longer allow iolists as contents in chunk/2Björn Gustavsson
The clause that converted an iolist to a binary was never executed. Note that chunk/2 is called for all chunks in the {extra_chunks,Chunks} option. This change will enforce that the contents of each chunk must be a binary (as documented).
2017-11-27beam_utils_SUITE: Cover more lines in beam_utilsBjörn Gustavsson
2017-11-27guard_SUITE: Add a test case to cover beam_dead:turn_op/1Björn Gustavsson
f9a323d10a9f5d added consistent operand order for equality comparisons. As a result, beam_dead:turn_op/1 is no longer covered. We must keep the uncovered lines in beam_dead to ensure that beam_dead can handle BEAM assembly code from another source than v3_codegen that might not follow the operand order convention. The only way to cover the lines is to use BEAM assembly in the test case.
2017-11-27Recognize 'nil' as a literal in beam_utils:bif_to_test/3Björn Gustavsson
beam_utils:bif_to_test/3 is supposed to never put a literal operand as the first operand in is_eq_exact or is_ne_exact, but 'nil' was not recognized as a literal.
2017-11-27Cover more code in beam_bsm:btb_opt_1/3Björn Gustavsson
2017-11-27Add test to cover a line in v3_kernel:opt_single_valued/3Björn Gustavsson
The uncovered line was added in 6753bbcc3fdb0.
2017-11-27Merge pull request #1545 from michalmuskala/move-optBjörn Gustavsson
Place move S x0 instructions at the end of blocks
2017-11-24Merge branch 'hans/public_key/base64_fix/OTP-14788'Hans Nilsson
* hans/public_key/base64_fix/OTP-14788: public_key: Fix bad calls to base64:*
2017-11-24ssl: Add gracefullness to dtls codeIngela Anderton Andin
Also make tls code a little more direct for easier uderstanding
2017-11-24ssl: Remove old softupgrade codeIngela Anderton Andin
This code is only relevant in version is was written in.