From bgustavsson@REDACTED Tue Feb 2 11:38:16 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Tue, 2 Feb 2010 11:38:16 +0100 Subject: [erlang-patches] Improvements in the Report Browser In-Reply-To: <6672d0161001041018n75c904ccwf298a1640765533d@mail.gmail.com> References: <110EAD81-677A-479D-B797-72F9A55BDB70@gmail.com> <6672d0160912210230w5443d4a1wb3f3c249d792daf5@mail.gmail.com> <6B2422AA-528F-497B-B71C-DAC2699C75EB@gmail.com> <6672d0160912220902h491baaedi1c472e8945b69245@mail.gmail.com> <87CCF4EE-A7AB-402F-9D05-929739052C85@gmail.com> <6672d0161001041018n75c904ccwf298a1640765533d@mail.gmail.com> Message-ID: <6672d0161002020238l1d9dc100g6d9f632c1dea22a2@mail.gmail.com> 2010/1/4 Bj?rn Gustavsson : > On Fri, Dec 25, 2009 at 10:03 AM, Videla Alvaro wrote: > >> http://github.com/videlalvaro/otp/tree/va/rb-improvements > > Thanks! Included in 'pu'. We have now reviewed this patch. In order to include it in the next release, the following changes must be made: * There is no need for a new function to replace rb:grep/1. Since rb:grep/1 is (usually) used manually to browse reports, the minor potential incompatibility introduced by changing the regular expression engine is acceptable even in a minor release. Therefore, rb:grep/1 should be changed to use the re module. * It must not be a requirement to pass a compiled regular expression. rb:grep/1 should still accept a string, and automatically compile it. It could also optionally allow a compiled regular expression. (The new rb:filter/1 function should work in the same way.) -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From bgustavsson@REDACTED Tue Feb 2 21:59:12 2010 From: bgustavsson@REDACTED (=?ISO-8859-1?Q?Bj=F6rn_Gustavsson?=) Date: Tue, 02 Feb 2010 21:59:12 +0100 Subject: What's cooking in erlang/otp (2010-02-02) Message-ID: <4B689220.1080708@gmail.com> Provided you have the latest development branch (ccase/r13b04) it should now be possible to build the documentation. (Topic branches based on earlier revisions of the development branch need to be rebased.) The description for building the documentation can be found here: http://wiki.github.com/erlang/otp/documentation R13B04 will be released at the end of February. There will be a feature freeze for the erts, kernel, stdlib, and compiler applications on February 10. After that, only bug fixes will be included in those applications. ------------------------------------------------------------ [Graduated] * bg/cleanup-tests (eunit, kernel, stdlib) (2010-01-24) 5 commits + file_SUITE: eliminate a warning for an unused variable (a656a27) + kernel tests: modernize guard tests (fc1ddbb) + unicode_SUITE: replace deprecated concat_binary/1 with list_to_binary/1 (0cd1623) + stdlib tests: modernize guard tests (ac3d53f) + Test suites: fix creation of Emakefiles (b8b3a8c) * bg/hipe-fixes (compiler, stdlib) (2010-01-25) 3 commits + compiler: make ignore_native_errors also handle internal hipe errors (9e2ac0d) + re tests: disable native-code compilation for huge modules (24e2773) + Teach the compiler the no_native option (fe14832) * cf/epp-macro-overloading (otp, parsetools, stdlib) (2010-02-01) 8 commits + yecc_SUITE: Adjustment for modified error tuple (1ab56cb) + epp_SUITE: Increase code coverage (77a1272) + Minor corrections and clean-ups (ec618ee) + documentation: Macros overloading partly rewritten (c416896) + update the documentation on preprocessor in the reference manual (4c18274) + epp: change rules to choose the right version of a macro (27d7ca0) + epp: Add support of macros overloading (5023556) + epp: fix bug in the function scan_undef (0d15f8a) * sf/erl_match-longlong (erl_interface) (2010-01-20) 1 commit + erl_interface: Allow erl_match() to match ERL_LONGLONG and ERL_U_LONGLONG terms (938ccfe) ------------------------------------------------------------ [New topics] * bg/erl_bif_types (hipe) (2010-01-31) 1 commit - erl_bif_types: add 'http_bin' as a valid packet option (7c72a77) * bg/temp-hipe-fixes (compiler, hipe, kernel) (2010-01-31) 4 commits - Crudely fix return type for the lists:key{search,find,member}/3 (5c9a017) - seq_trace_SUITE: Don't native-compile (c569f85) - Disable native code if on_load is used (101dcc8) - andor_SUITE: Don't native-compile (cb44959) This branch contains temporary workarounds to avoid failing test cases and is never intended to graduate. * dgud/wx_list_ctrl (wx) (2010-02-02) 5 commits - Generate without trailing spaces (cfd2ba4) - Fix correct starting copyright year (5b585a4) - Do not emit trailing whitespace (ce3cd31) - Added tests for wxListCtrl:getItem/2. (aa17e2b) - Added wxListCtrl:getEditCtrl (not available on Mac) (e6433fc) * egil/lcnt (erts, tools) (2010-02-01) 6 commits - Add test suite for lcnt in tools (e26bedd) - Add lcnt:rt_opt/1 bindings to erts_debug (c24d153) - Add runtime option to enable/disable lcnt stats (dd665ed) - Add auto width on string output (12990af) - Add lcnt documentation (41d4913) - Add lock profiling tool (c41fa05) * rani/inet_gethostbyname_fixes (erts, kernel) (2010-01-28) 1 commit - inet:gethostbyname improved to parse IP strings and look up own hostname (3fbd6e6) ------------------------------------------------------------ [Stalled] * sc/sctp-connect-nowait (kernel) (2009-12-21) 1 commit - Implement a non-blocking SCTP connect (ae3d819) Approved in principle, but the implementation needs some finish touches as pointed out by Raimo, and also documentation and test cases. * jn/supervisor_child_count_only (stdlib) (2010-01-25) 1 commit - Add count_children/1 to supervisor.erl to determine the number of children being managed without the memory impact of which_children/1 (10ef53f) Test cases fail when the emulator has been started with the +Meamin option. * va/rb-improvements (sasl) (2009-12-25) 4 commits - New function to filter reports by date (5391383) - New rb:filter/1 function to ease report filtering (ed28348) - Add rb:re/1 to grep reports using the re module (e5bd091) - Fix minor typo in read_report/1 (07c6ad4) There is no need to introduce a new function rb:re/1, but rb:grep/1 can be modified to use the 're' module. Also, it should not be necessary to compile the regular expression beforehand. ------------------------------------------------------------ [Cooking] * at/odbc_osx_fix (odbc) (2009-11-30) 1 commit - Workaround broken ODBC implementation on OSX 10.5 (88c71c9) * db/tv_nthtail_fix (tv) (2009-12-01) 1 commit - Fix for tv which restarts while trying to open a table (abe0759) The author of this topic branch has found the real reason for the problem and has made some progress in resolving it. * dgud/emacs-mode (tools) (2010-01-26) 2 commits - Added the old style skeletons, and a variable to change. (2cbe853) - Move code skeletons to a separate file. Use the updated skeletons from the Erlware Mode. (b4fd161) * gc/hipe_darwin_amd64 (erts) (2010-01-12) 4 commits - Fix hipe memory allocation problems on darwin/amd64. (32ca14f) - Porting x86 darwin fixes to amd64 darwin hipe asm/m4 code. (427f1db) - Automatically enable hipe for darwin/amd64 builds. (43ae4b7) - Allow configure to enable_hipe for darwin/amd64 builds. (733ce7a) * is/mnesia-send-compressed (mnesia) (2010-01-10) 1 commit - Add option to compress data when copying tables between Mnesia nodes (3757d8f) Should graduate soon. * jb/atom-table-size (erts) (2010-01-29) 1 commit - Add the +t emulator option to change the maximum number of atoms (4895776) Should graduate soon. * jn/sasl-format_report (sasl) (2010-01-27) 1 commit - Added sasl_report:format_report/3 to return a string rather than send an io (ee8f29f) * jp/dependencies_makefile (compiler, erts) (2010-01-12) 2 commits - squash! fix snprintf for Windows (bf2708d) - Add dependencies Makefile generation to erlc(1) and compile(3) (7cbc437) Approved in principle by the OTP Technical Board; now awaits a closer review of the implementation. * ks/bsl_system_limit (hipe) (2010-01-26) 1 commit - Fix a HiPE compiler bug evaluating an expression that throws system_limit. (dcada69) * pan/enable_m32_build (erts, otp) (2009-12-08) 1 commit - Teach configure --enable-m32-build for forcing a 32-bit build (15c7817) * po/odbc-update-delete-no-rows (odbc) (2009-11-26) 1 commit - SQL_NO_DATA and SQLSTATE 00000 indicate success for updates/deletes that affect no rows (1a564f8) From videlalvaro@REDACTED Wed Feb 3 13:46:03 2010 From: videlalvaro@REDACTED (Alvaro Videla) Date: Wed, 3 Feb 2010 20:46:03 +0800 Subject: [erlang-patches] Improvements in the Report Browser In-Reply-To: <6672d0161002020238l1d9dc100g6d9f632c1dea22a2@mail.gmail.com> References: <110EAD81-677A-479D-B797-72F9A55BDB70@gmail.com> <6672d0160912210230w5443d4a1wb3f3c249d792daf5@mail.gmail.com> <6B2422AA-528F-497B-B71C-DAC2699C75EB@gmail.com> <6672d0160912220902h491baaedi1c472e8945b69245@mail.gmail.com> <87CCF4EE-A7AB-402F-9D05-929739052C85@gmail.com> <6672d0161001041018n75c904ccwf298a1640765533d@mail.gmail.com> <6672d0161002020238l1d9dc100g6d9f632c1dea22a2@mail.gmail.com> Message-ID: <1E301D2D-899A-41BE-807A-21D8174ACB7D@gmail.com> I will take a look into this an let you know when I'm finished On Feb 2, 2010, at 6:38 PM, Bj?rn Gustavsson wrote: > 2010/1/4 Bj?rn Gustavsson : >> On Fri, Dec 25, 2009 at 10:03 AM, Videla Alvaro wrote: >> >>> http://github.com/videlalvaro/otp/tree/va/rb-improvements >> >> Thanks! Included in 'pu'. > > We have now reviewed this patch. In order to include it in the next release, > the following changes must be made: > > * There is no need for a new function to replace rb:grep/1. Since rb:grep/1 is > (usually) used manually to browse reports, the minor potential incompatibility > introduced by changing the regular expression engine is acceptable even in > a minor release. Therefore, rb:grep/1 should be changed to use the re module. > > * It must not be a requirement to pass a compiled regular expression. rb:grep/1 > should still accept a string, and automatically compile it. It could > also optionally > allow a compiled regular expression. (The new rb:filter/1 function should work > in the same way.) > > -- > Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From tuncer.ayaz@REDACTED Thu Feb 4 02:39:04 2010 From: tuncer.ayaz@REDACTED (Tuncer Ayaz) Date: Thu, 4 Feb 2010 02:39:04 +0100 Subject: Fix typo in kernel/test/code_SUITE Message-ID: <4ac8254d1002031739m6749936do4c0a95fa24ec0561@mail.gmail.com> There was a small typo in the doc function clause of ext_mod_dep. $ git fetch git://github.com/tuncer/otp code_suite_typo From tuncer.ayaz@REDACTED Thu Feb 4 03:13:05 2010 From: tuncer.ayaz@REDACTED (Tuncer Ayaz) Date: Thu, 4 Feb 2010 03:13:05 +0100 Subject: Fix handling of archive (.ez) files in code:clash/0 Message-ID: <4ac8254d1002031813i6203e4aeu5995fe9fc51d006a@mail.gmail.com> code:clash/0 did not take into account .ez files when listing contents of code path entries. Using erl_prim_loader:list_dir/1 instead of file:list_dir/1 fixes the problem. $ git fetch git://github.com/tuncer/otp.git clash_ez From bgustavsson@REDACTED Thu Feb 4 09:39:16 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Thu, 4 Feb 2010 09:39:16 +0100 Subject: [erlang-patches] Fix handling of archive (.ez) files in code:clash/0 In-Reply-To: <4ac8254d1002031813i6203e4aeu5995fe9fc51d006a@mail.gmail.com> References: <4ac8254d1002031813i6203e4aeu5995fe9fc51d006a@mail.gmail.com> Message-ID: <6672d0161002040039h5d16f2d9v2a9aa01b78be7f32@mail.gmail.com> On Thu, Feb 4, 2010 at 3:13 AM, Tuncer Ayaz wrote: > code:clash/0 did not take into account .ez files > when listing contents of code path entries. > Using erl_prim_loader:list_dir/1 instead of > file:list_dir/1 fixes the problem. > > $ git fetch git://github.com/tuncer/otp.git clash_ez Thanks! Included in pu. -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From bgustavsson@REDACTED Thu Feb 4 09:40:11 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Thu, 4 Feb 2010 09:40:11 +0100 Subject: [erlang-patches] Fix typo in kernel/test/code_SUITE In-Reply-To: <4ac8254d1002031739m6749936do4c0a95fa24ec0561@mail.gmail.com> References: <4ac8254d1002031739m6749936do4c0a95fa24ec0561@mail.gmail.com> Message-ID: <6672d0161002040040yb5f2f96t5bec00a9e844bc90@mail.gmail.com> On Thu, Feb 4, 2010 at 2:39 AM, Tuncer Ayaz wrote: > There was a small typo in the doc function clause of ext_mod_dep. > > $ git fetch git://github.com/tuncer/otp code_suite_typo Thanks! I will apply the correction directly on the development branch. -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From tomas@REDACTED Thu Feb 4 11:18:15 2010 From: tomas@REDACTED (Tomas Johansson) Date: Thu, 4 Feb 2010 11:18:15 +0100 Subject: New asn1 compiler option Message-ID: <9e69592f1002040218u603552c1o7f525f094c0b302b@mail.gmail.com> I've added a new compiler option to the asn1 compiler called macro_name_prefix. It works similar to the existing option record_name_prefix, but for macros instead. It is needed when multiple protocols that contains macros with identical names are included in a single module. I've also added the existing record_name_prefix option to the documentation since it was missing. Couldn't find any tests in the asn1 application, but I used the asn1 files in the examples/ subdir to verify that my changes have the desired effect. git fetch git://github.com/jotsen/otp.git asn1_macro_prefix From tuncer.ayaz@REDACTED Thu Feb 4 20:06:46 2010 From: tuncer.ayaz@REDACTED (Tuncer Ayaz) Date: Thu, 4 Feb 2010 20:06:46 +0100 Subject: [erlang-patches] Fix handling of archive (.ez) files in code:clash/0 In-Reply-To: <6672d0161002040039h5d16f2d9v2a9aa01b78be7f32@mail.gmail.com> References: <4ac8254d1002031813i6203e4aeu5995fe9fc51d006a@mail.gmail.com> <6672d0161002040039h5d16f2d9v2a9aa01b78be7f32@mail.gmail.com> Message-ID: <4ac8254d1002041106u78599907yae41ab49bcf1be9f@mail.gmail.com> 2010/2/4 Bj?rn Gustavsson : > On Thu, Feb 4, 2010 at 3:13 AM, Tuncer Ayaz wrote: >> code:clash/0 did not take into account .ez files >> when listing contents of code path entries. >> Using erl_prim_loader:list_dir/1 instead of >> file:list_dir/1 fixes the problem. >> >> $ git fetch git://github.com/tuncer/otp.git clash_ez > > Thanks! Included in pu. Added tests for code:clash/0 to the same topic branch. From zl9d97p02@REDACTED Fri Feb 5 00:45:07 2010 From: zl9d97p02@REDACTED (Simon Cornish) Date: Thu, 4 Feb 2010 15:45:07 -0800 Subject: [erlang-patches] Non-blocking SCTP connect [Was: [erlang-patches] SCTP improvements] In-Reply-To: References: <17551-1260988040-19738@sneakemail.com> <32461-1261465043-568221@sneakemail.com> <20100125151205.GA24872@erix.ericsson.se> <3867-1264434760-792744@sneakemail.com> <20100126091434.GA30687@erix.ericsson.se> Message-ID: <25128-1265327134-719583@sneakemail.com> Hello, API refinements as suggested by Raimo, test cases and documentation. git fetch git://github.com/dotsimon/otp.git sctp_connect_nowait Regards, Simon From bgustavsson@REDACTED Fri Feb 5 07:00:20 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Fri, 5 Feb 2010 07:00:20 +0100 Subject: [erlang-patches] Non-blocking SCTP connect [Was: [erlang-patches] SCTP improvements] In-Reply-To: <25128-1265327134-719583@sneakemail.com> References: <17551-1260988040-19738@sneakemail.com> <32461-1261465043-568221@sneakemail.com> <20100125151205.GA24872@erix.ericsson.se> <3867-1264434760-792744@sneakemail.com> <20100126091434.GA30687@erix.ericsson.se> <25128-1265327134-719583@sneakemail.com> Message-ID: <6672d0161002042200l39cac1dbh9f2e913684ca187e@mail.gmail.com> On Fri, Feb 5, 2010 at 12:45 AM, Simon Cornish wrote: > Hello, > API refinements as suggested by Raimo, test cases and documentation. > > git fetch git://github.com/dotsimon/otp.git sctp_connect_nowait Thanks! I cherry-picked your commit and put into the previous branch. (Since your branch contained merges.) There are some compilation warnings: ./gen_sctp.erl:129: Warning: variable 'Addr' is unused ./gen_sctp.erl:129: Warning: variable 'Opts' is unused ./gen_sctp.erl:129: Warning: variable 'Port' is unused ./gen_sctp.erl:129: Warning: variable 'S' is unused ./gen_sctp.erl:129: Warning: variable 'Timeout' is unused -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From bgustavsson@REDACTED Fri Feb 5 07:04:11 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Fri, 5 Feb 2010 07:04:11 +0100 Subject: [erlang-patches] New asn1 compiler option In-Reply-To: <9e69592f1002040218u603552c1o7f525f094c0b302b@mail.gmail.com> References: <9e69592f1002040218u603552c1o7f525f094c0b302b@mail.gmail.com> Message-ID: <6672d0161002042204l508267dcjedfb4db63d9b26a6@mail.gmail.com> On Thu, Feb 4, 2010 at 11:18 AM, Tomas Johansson wrote: > Couldn't find any tests in the asn1 application [...] We have not published them yet. > git fetch git://github.com/jotsen/otp.git asn1_macro_prefix Thanks! Will be included in pu next time I push. -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From raimo+erlang-patches@REDACTED Fri Feb 5 09:57:08 2010 From: raimo+erlang-patches@REDACTED (Raimo Niskanen) Date: Fri, 5 Feb 2010 09:57:08 +0100 Subject: [erlang-patches] Non-blocking SCTP connect [Was: [erlang-patches] SCTP improvements] In-Reply-To: <6672d0161002042200l39cac1dbh9f2e913684ca187e@mail.gmail.com> References: <17551-1260988040-19738@sneakemail.com> <32461-1261465043-568221@sneakemail.com> <20100125151205.GA24872@erix.ericsson.se> <3867-1264434760-792744@sneakemail.com> <20100126091434.GA30687@erix.ericsson.se> <25128-1265327134-719583@sneakemail.com> <6672d0161002042200l39cac1dbh9f2e913684ca187e@mail.gmail.com> Message-ID: <20100205085708.GB29445@erix.ericsson.se> On Fri, Feb 05, 2010 at 07:00:20AM +0100, Bj?rn Gustavsson wrote: > On Fri, Feb 5, 2010 at 12:45 AM, Simon Cornish wrote: > > Hello, > > API refinements as suggested by Raimo, test cases and documentation. > > > > git fetch git://github.com/dotsimon/otp.git sctp_connect_nowait > > Thanks! I cherry-picked your commit and put into the previous branch. > (Since your branch contained merges.) > > There are some compilation warnings: > > ./gen_sctp.erl:129: Warning: variable 'Addr' is unused > ./gen_sctp.erl:129: Warning: variable 'Opts' is unused > ./gen_sctp.erl:129: Warning: variable 'Port' is unused > ./gen_sctp.erl:129: Warning: variable 'S' is unused > ./gen_sctp.erl:129: Warning: variable 'Timeout' is unused > > -- > Bj?rn Gustavsson, Erlang/OTP, Ericsson AB Who fixes them? > > ________________________________________________________________ > erlang-patches (at) erlang.org mailing list. > See http://www.erlang.org/faq.html > To unsubscribe; mailto:erlang-patches-unsubscribe@REDACTED > -- / Raimo Niskanen, Erlang/OTP, Ericsson AB From raimo+erlang-patches@REDACTED Fri Feb 5 09:58:08 2010 From: raimo+erlang-patches@REDACTED (Raimo Niskanen) Date: Fri, 5 Feb 2010 09:58:08 +0100 Subject: [erlang-patches] Non-blocking SCTP connect [Was: [erlang-patches] SCTP improvements] In-Reply-To: <25128-1265327134-719583@sneakemail.com> References: <17551-1260988040-19738@sneakemail.com> <32461-1261465043-568221@sneakemail.com> <20100125151205.GA24872@erix.ericsson.se> <3867-1264434760-792744@sneakemail.com> <20100126091434.GA30687@erix.ericsson.se> <25128-1265327134-719583@sneakemail.com> Message-ID: <20100205085808.GC29445@erix.ericsson.se> On Thu, Feb 04, 2010 at 03:45:07PM -0800, Simon Cornish wrote: > Hello, > API refinements as suggested by Raimo, test cases and documentation. > Awsome! > git fetch git://github.com/dotsimon/otp.git sctp_connect_nowait > > Regards, > Simon > > ________________________________________________________________ > erlang-patches (at) erlang.org mailing list. > See http://www.erlang.org/faq.html > To unsubscribe; mailto:erlang-patches-unsubscribe@REDACTED -- / Raimo Niskanen, Erlang/OTP, Ericsson AB From bgustavsson@REDACTED Fri Feb 5 10:18:47 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Fri, 5 Feb 2010 10:18:47 +0100 Subject: [erlang-patches] Fix handling of archive (.ez) files in code:clash/0 In-Reply-To: <4ac8254d1002041106u78599907yae41ab49bcf1be9f@mail.gmail.com> References: <4ac8254d1002031813i6203e4aeu5995fe9fc51d006a@mail.gmail.com> <6672d0161002040039h5d16f2d9v2a9aa01b78be7f32@mail.gmail.com> <4ac8254d1002041106u78599907yae41ab49bcf1be9f@mail.gmail.com> Message-ID: <6672d0161002050118y13b3f991j137e89c64acb1901@mail.gmail.com> 2010/2/4 Tuncer Ayaz : > 2010/2/4 Bj?rn Gustavsson : >> Thanks! Included in pu. > > Added tests for code:clash/0 to the same topic branch. > Looks good, with a few minor issues. The branch will probably graduate early next week. The following patch will be included in your branch when I'll push the pu branch later today. I intend to combine it with your commit before the branch is graduated. (In separate commit I have also added a missing is_list(Config) guard test for another function in the same file.) --- Subject: [PATCH 1/2] fixup! fix minor issues To be combined with the previous commit. * Include the test in the all/1 function. * Add "when is_list(Config)" to make sure that the test case properly ignores unknown atomic value. * Don't bother providing the 'doc' clause without any useful information. Also omit the 'suite' clause. --- lib/kernel/test/code_SUITE.erl | 7 ++----- 1 files changed, 2 insertions(+), 5 deletions(-) diff --git a/lib/kernel/test/code_SUITE.erl b/lib/kernel/test/code_SUITE.erl index c8063ca..7712973 100644 --- a/lib/kernel/test/code_SUITE.erl +++ b/lib/kernel/test/code_SUITE.erl @@ -48,7 +48,7 @@ all(suite) -> delete, purge, soft_purge, is_loaded, all_loaded, load_binary, dir_req, object_code, set_path_file, pa_pz_option, add_del_path, - dir_disappeared, ext_mod_dep, + dir_disappeared, ext_mod_dep, clash, load_cached, start_node_with_cache, add_and_rehash, where_is_file_no_cache, where_is_file_cached, purge_stacktrace, mult_lib_roots, bad_erl_libs, @@ -545,10 +545,7 @@ add_del_path(Config) -> ok. - -clash(suite) -> []; -clash(doc) -> ["clash"]; -clash(Config) -> +clash(Config) when is_list(Config) -> DDir = ?config(data_dir,Config)++"clash/", P = code:get_path(), -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From tuncer.ayaz@REDACTED Fri Feb 5 13:08:07 2010 From: tuncer.ayaz@REDACTED (Tuncer Ayaz) Date: Fri, 5 Feb 2010 13:08:07 +0100 Subject: [erlang-patches] Fix handling of archive (.ez) files in code:clash/0 In-Reply-To: <6672d0161002050118y13b3f991j137e89c64acb1901@mail.gmail.com> References: <4ac8254d1002031813i6203e4aeu5995fe9fc51d006a@mail.gmail.com> <6672d0161002040039h5d16f2d9v2a9aa01b78be7f32@mail.gmail.com> <4ac8254d1002041106u78599907yae41ab49bcf1be9f@mail.gmail.com> <6672d0161002050118y13b3f991j137e89c64acb1901@mail.gmail.com> Message-ID: <4ac8254d1002050408h17cd90b1y3e584a656952bb7e@mail.gmail.com> 2010/2/5 Bj?rn Gustavsson : > 2010/2/4 Tuncer Ayaz : >> 2010/2/4 Bj?rn Gustavsson : >>> Thanks! Included in pu. >> >> Added tests for code:clash/0 to the same topic branch. >> > > Looks good, with a few minor issues. The branch will > probably graduate early next week. Great, thanks for the quick review. Many reltool, rebar and riak users will be thankful if this gets included in R13B04 :). > The following patch will be included in your branch > when I'll push the pu branch later today. I intend to combine > it with your commit before the branch is graduated. Please see http://github.com/tuncer/otp/commit/02d75ce for a rebased tree with all the changes. > (In separate commit I have also added a missing > is_list(Config) guard test for another function in the > same file.) Thanks! > --- > Subject: [PATCH 1/2] fixup! fix minor issues > > To be combined with the previous commit. > > * Include the test in the all/1 function. > > * Add "when is_list(Config)" to make sure that the test > case properly ignores unknown atomic value. > > * Don't bother providing the 'doc' clause without any useful > information. Also omit the 'suite' clause. > --- > lib/kernel/test/code_SUITE.erl | 7 ++----- > 1 files changed, 2 insertions(+), 5 deletions(-) > > diff --git a/lib/kernel/test/code_SUITE.erl b/lib/kernel/test/code_SUITE.erl > index c8063ca..7712973 100644 > --- a/lib/kernel/test/code_SUITE.erl > +++ b/lib/kernel/test/code_SUITE.erl > @@ -48,7 +48,7 @@ all(suite) -> > delete, purge, soft_purge, is_loaded, all_loaded, > load_binary, dir_req, object_code, set_path_file, > pa_pz_option, add_del_path, > - dir_disappeared, ext_mod_dep, > + dir_disappeared, ext_mod_dep, clash, Interesting as this was part of the rebased tree to be found as http://github.com/tuncer/otp/commit/585f77 and also in a previous state http://github.com/tuncer/otp/commit/5fcd29 Maybe you took an earlier tree. > load_cached, start_node_with_cache, add_and_rehash, > where_is_file_no_cache, where_is_file_cached, > purge_stacktrace, mult_lib_roots, bad_erl_libs, > @@ -545,10 +545,7 @@ add_del_path(Config) -> > ok. > > > - > -clash(suite) -> []; > -clash(doc) -> ["clash"]; > -clash(Config) -> > +clash(Config) when is_list(Config) -> > DDir = ?config(data_dir,Config)++"clash/", > P = code:get_path(), I've pushed a rebased atop of ccase/r13b04_dev tree with these changes but have to ask why we remove the two clauses. Every other test seems to have the suite and doc clause even if they only return [] in both clauses. From bgustavsson@REDACTED Fri Feb 5 14:45:14 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Fri, 5 Feb 2010 14:45:14 +0100 Subject: [erlang-patches] Fix handling of archive (.ez) files in code:clash/0 In-Reply-To: <4ac8254d1002050408h17cd90b1y3e584a656952bb7e@mail.gmail.com> References: <4ac8254d1002031813i6203e4aeu5995fe9fc51d006a@mail.gmail.com> <6672d0161002040039h5d16f2d9v2a9aa01b78be7f32@mail.gmail.com> <4ac8254d1002041106u78599907yae41ab49bcf1be9f@mail.gmail.com> <6672d0161002050118y13b3f991j137e89c64acb1901@mail.gmail.com> <4ac8254d1002050408h17cd90b1y3e584a656952bb7e@mail.gmail.com> Message-ID: <6672d0161002050545y33f7a64bib5db901c16f88e90@mail.gmail.com> 2010/2/5 Tuncer Ayaz : >> diff --git a/lib/kernel/test/code_SUITE.erl b/lib/kernel/test/code_SUITE.erl >> index c8063ca..7712973 100644 >> --- a/lib/kernel/test/code_SUITE.erl >> +++ b/lib/kernel/test/code_SUITE.erl >> @@ -48,7 +48,7 @@ all(suite) -> >> delete, purge, soft_purge, is_loaded, all_loaded, >> load_binary, dir_req, object_code, set_path_file, >> pa_pz_option, add_del_path, >> - dir_disappeared, ext_mod_dep, >> + dir_disappeared, ext_mod_dep, clash, > > Interesting as this was part of the rebased tree > to be found as > http://github.com/tuncer/otp/commit/585f77 > and also in a previous state > http://github.com/tuncer/otp/commit/5fcd29 > > Maybe you took an earlier tree. Yes, this is strange. I fetched your branch early this morning. > > I've pushed a rebased atop of ccase/r13b04_dev tree > with these changes but have to ask why we remove > the two clauses. Every other test seems to have > the suite and doc clause even if they only return [] > in both clauses. > Matter of taste. "suite" is optional, so I see no reason to put it in in new test cases. On the other hand, I don't find it important enough to change old test cases. "doc" is OK if it it actually gives more information than the name of the test case. Since it didn't, I ripped it out. We currently do not have any tool that will collect only document strings and produce a nice report. I am not sure that anyone is interested in such a report. The documentation strings are usually not detailed enough to be of any real use out-of-context. -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From bgustavsson@REDACTED Fri Feb 5 14:57:08 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Fri, 5 Feb 2010 14:57:08 +0100 Subject: [erlang-patches] Fix handling of archive (.ez) files in code:clash/0 In-Reply-To: <4ac8254d1002050408h17cd90b1y3e584a656952bb7e@mail.gmail.com> References: <4ac8254d1002031813i6203e4aeu5995fe9fc51d006a@mail.gmail.com> <6672d0161002040039h5d16f2d9v2a9aa01b78be7f32@mail.gmail.com> <4ac8254d1002041106u78599907yae41ab49bcf1be9f@mail.gmail.com> <6672d0161002050118y13b3f991j137e89c64acb1901@mail.gmail.com> <4ac8254d1002050408h17cd90b1y3e584a656952bb7e@mail.gmail.com> Message-ID: <6672d0161002050557t28df6e7aka8cc00a9aa1aea86@mail.gmail.com> 2010/2/5 Tuncer Ayaz : >> diff --git a/lib/kernel/test/code_SUITE.erl b/lib/kernel/test/code_SUITE.erl >> index c8063ca..7712973 100644 >> --- a/lib/kernel/test/code_SUITE.erl >> +++ b/lib/kernel/test/code_SUITE.erl >> @@ -48,7 +48,7 @@ all(suite) -> >> delete, purge, soft_purge, is_loaded, all_loaded, >> load_binary, dir_req, object_code, set_path_file, >> pa_pz_option, add_del_path, >> - dir_disappeared, ext_mod_dep, >> + dir_disappeared, ext_mod_dep, clash, > > Interesting as this was part of the rebased tree > to be found as > http://github.com/tuncer/otp/commit/585f77 > and also in a previous state > http://github.com/tuncer/otp/commit/5fcd29 Disregard my previous answer. You exported clash/1, but didn't add it to the list of test cases to run. That's what my patch corrects. -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From raimo+erlang-patches@REDACTED Fri Feb 5 15:24:57 2010 From: raimo+erlang-patches@REDACTED (Raimo Niskanen) Date: Fri, 5 Feb 2010 15:24:57 +0100 Subject: [erlang-patches] Non-blocking SCTP connect [Was: [erlang-patches] SCTP improvements] In-Reply-To: <20100205085808.GC29445@erix.ericsson.se> References: <17551-1260988040-19738@sneakemail.com> <32461-1261465043-568221@sneakemail.com> <20100125151205.GA24872@erix.ericsson.se> <3867-1264434760-792744@sneakemail.com> <20100126091434.GA30687@erix.ericsson.se> <25128-1265327134-719583@sneakemail.com> <20100205085808.GC29445@erix.ericsson.se> Message-ID: <20100205142456.GA21696@erix.ericsson.se> On Fri, Feb 05, 2010 at 09:58:08AM +0100, Raimo Niskanen wrote: > On Thu, Feb 04, 2010 at 03:45:07PM -0800, Simon Cornish wrote: > > Hello, > > API refinements as suggested by Raimo, test cases and documentation. > > > > Awsome! > Bjorn fixed the warnings (as you already know), and I fixed the testcase to be a bit more flexible (one machine SLES 10 SP1 x86_64 produced different timing and return values in our test runs). It should be pushed anytime now. > > git fetch git://github.com/dotsimon/otp.git sctp_connect_nowait > > > > Regards, > > Simon > > > > ________________________________________________________________ > > erlang-patches (at) erlang.org mailing list. > > See http://www.erlang.org/faq.html > > To unsubscribe; mailto:erlang-patches-unsubscribe@REDACTED > > -- > > / Raimo Niskanen, Erlang/OTP, Ericsson AB > > ________________________________________________________________ > erlang-patches (at) erlang.org mailing list. > See http://www.erlang.org/faq.html > To unsubscribe; mailto:erlang-patches-unsubscribe@REDACTED -- / Raimo Niskanen, Erlang/OTP, Ericsson AB From tuncer.ayaz@REDACTED Fri Feb 5 18:19:35 2010 From: tuncer.ayaz@REDACTED (Tuncer Ayaz) Date: Fri, 5 Feb 2010 18:19:35 +0100 Subject: [erlang-patches] Fix handling of archive (.ez) files in code:clash/0 In-Reply-To: <6672d0161002050557t28df6e7aka8cc00a9aa1aea86@mail.gmail.com> References: <4ac8254d1002031813i6203e4aeu5995fe9fc51d006a@mail.gmail.com> <6672d0161002040039h5d16f2d9v2a9aa01b78be7f32@mail.gmail.com> <4ac8254d1002041106u78599907yae41ab49bcf1be9f@mail.gmail.com> <6672d0161002050118y13b3f991j137e89c64acb1901@mail.gmail.com> <4ac8254d1002050408h17cd90b1y3e584a656952bb7e@mail.gmail.com> <6672d0161002050557t28df6e7aka8cc00a9aa1aea86@mail.gmail.com> Message-ID: <4ac8254d1002050919w14c5cec8u72133e0ec4911b38@mail.gmail.com> 2010/2/5 Bj?rn Gustavsson : > 2010/2/5 Tuncer Ayaz : >>> diff --git a/lib/kernel/test/code_SUITE.erl b/lib/kernel/test/code_SUITE.erl >>> index c8063ca..7712973 100644 >>> --- a/lib/kernel/test/code_SUITE.erl >>> +++ b/lib/kernel/test/code_SUITE.erl >>> @@ -48,7 +48,7 @@ all(suite) -> >>> ? ? ?delete, purge, soft_purge, is_loaded, all_loaded, >>> ? ? ?load_binary, dir_req, object_code, set_path_file, >>> ? ? ?pa_pz_option, add_del_path, >>> - ? ? dir_disappeared, ext_mod_dep, >>> + ? ? dir_disappeared, ext_mod_dep, clash, >> >> Interesting as this was part of the rebased tree >> to be found as >> http://github.com/tuncer/otp/commit/585f77 >> and also in a previous state >> http://github.com/tuncer/otp/commit/5fcd29 > > Disregard my previous answer. > > You exported clash/1, but didn't add it > to the list of test cases to run. That's what > my patch corrects. Oh, right, forgot that as I only ran the new test and not the suite as a whole. Thanks. For completeness I've pushed a corrected branch http://github.com/tuncer/otp/commit/a9e3d2 From tuncer.ayaz@REDACTED Fri Feb 5 18:26:35 2010 From: tuncer.ayaz@REDACTED (Tuncer Ayaz) Date: Fri, 5 Feb 2010 18:26:35 +0100 Subject: [erlang-patches] Fix handling of archive (.ez) files in code:clash/0 In-Reply-To: <4ac8254d1002050919w14c5cec8u72133e0ec4911b38@mail.gmail.com> References: <4ac8254d1002031813i6203e4aeu5995fe9fc51d006a@mail.gmail.com> <6672d0161002040039h5d16f2d9v2a9aa01b78be7f32@mail.gmail.com> <4ac8254d1002041106u78599907yae41ab49bcf1be9f@mail.gmail.com> <6672d0161002050118y13b3f991j137e89c64acb1901@mail.gmail.com> <4ac8254d1002050408h17cd90b1y3e584a656952bb7e@mail.gmail.com> <6672d0161002050557t28df6e7aka8cc00a9aa1aea86@mail.gmail.com> <4ac8254d1002050919w14c5cec8u72133e0ec4911b38@mail.gmail.com> Message-ID: <4ac8254d1002050926me236cc7w408d64c8158dbfc4@mail.gmail.com> 2010/2/5 Tuncer Ayaz : > 2010/2/5 Bj?rn Gustavsson : >> 2010/2/5 Tuncer Ayaz : >>>> diff --git a/lib/kernel/test/code_SUITE.erl b/lib/kernel/test/code_SUITE.erl >>>> index c8063ca..7712973 100644 >>>> --- a/lib/kernel/test/code_SUITE.erl >>>> +++ b/lib/kernel/test/code_SUITE.erl >>>> @@ -48,7 +48,7 @@ all(suite) -> >>>> ? ? ?delete, purge, soft_purge, is_loaded, all_loaded, >>>> ? ? ?load_binary, dir_req, object_code, set_path_file, >>>> ? ? ?pa_pz_option, add_del_path, >>>> - ? ? dir_disappeared, ext_mod_dep, >>>> + ? ? dir_disappeared, ext_mod_dep, clash, >>> >>> Interesting as this was part of the rebased tree >>> to be found as >>> http://github.com/tuncer/otp/commit/585f77 >>> and also in a previous state >>> http://github.com/tuncer/otp/commit/5fcd29 >> >> Disregard my previous answer. >> >> You exported clash/1, but didn't add it >> to the list of test cases to run. That's what >> my patch corrects. > > Oh, right, forgot that as I only ran the new > test and not the suite as a whole. > > Thanks. > > For completeness I've pushed a corrected branch > http://github.com/tuncer/otp/commit/a9e3d2 And rebased on top of latest ccase/r13b04_dev http://github.com/tuncer/otp/commit/7ae7fc5 From bgustavsson@REDACTED Fri Feb 5 20:16:43 2010 From: bgustavsson@REDACTED (=?ISO-8859-1?Q?Bj=F6rn_Gustavsson?=) Date: Fri, 05 Feb 2010 20:16:43 +0100 Subject: What's cooking in erlang/otp (2010-02-05) Message-ID: <4B6C6E9B.9020405@gmail.com> Here is the complete release schedule for R13B04: * February 10 (next Wednesday) there will be a feature freeze the erts, kernel, stdlib, and compiler applications. We hope to graduate several branches early next week before the freeze. * February 17 is the feature freeze for rest of Erlang/OTP. * February 24 is the day of the release. ------------------------------------------------------------ [Graduated] * dgud/wx_list_ctrl (wx) (2010-02-02) 5 commits + Generate without trailing spaces (cfd2ba4) + Fix correct starting copyright year (5b585a4) + Do not emit trailing whitespace (ce3cd31) + Added tests for wxListCtrl:getItem/2. (aa17e2b) + Added wxListCtrl:getEditCtrl (not available on Mac) (e6433fc) * egil/gitignore (.gitignore, otp) (2010-02-03) 2 commits + Ignore files generated when running/building tests (c53fe4b) + Update .gitignore files (2f1da83) * is/mnesia-send-compressed (mnesia) (2010-01-10) 1 commit + Add option to compress data when copying tables between Mnesia nodes (3757d8f) * jb/atom-table-size (erts) (2010-01-29) 1 commit + Add the +t emulator option to change the maximum number of atoms (4895776) ------------------------------------------------------------ [New topics] * egil/egd_render (percept) (2010-02-04) 1 commit - Fix egd_render transparent to use float constants (e7842b4) * ks/cleanups (compiler, gs, wx) (2010-02-05) 7 commits - wx: eliminate warnings and tidy as suggested by tidier (282b169) - gs: eliminate warnings and tidy as suggested by tidier (cdfa268) - compiler: support the configure option --enable-native-libs (0d75684) - compiler Makefile: alphabetize module names (589cddd) - sys_pre_expand: tidy as suggested by tidier (53832b3) - v3_core: remember line numbers for attributes (c7998f9) - compile.erl: eliminate compiler warning (4a76b7e) * ks/hipe-fixes (erts, hipe) (2010-02-05) 3 commits - Remove $Id$ (bec836e) - Adds types and specs to files in erts/preloaded. (952cd10) - Fix a HiPE compiler bug evaluating an expression that throws system_limit. (449ac18) * ta/code_clash_ez (kernel) (2010-02-05) 4 commits - code_SUITE: add missing is_list(Config) guard (a1263f0) - fixup! fix minor issues (92ecaf5) - Add tests for code:clash/0 (58477c2) - Fix handling of archive (.ez) files in code:clash/0 (49da83d) Should graduate early next week. * tj/asn1_macro_prefix (asn1) (2010-02-04) 1 commit - Add support for prefixing macro names generated by the compiler (e88fa38) ------------------------------------------------------------ [Stalled] * jn/supervisor_child_count_only (stdlib) (2010-01-25) 1 commit - Add count_children/1 to supervisor.erl to determine the number of children being managed without the memory impact of which_children/1 (10ef53f) Test cases fail when the emulator has been started with the +Meamin option. * va/rb-improvements (sasl) (2009-12-25) 4 commits - New function to filter reports by date (5391383) - New rb:filter/1 function to ease report filtering (ed28348) - Add rb:re/1 to grep reports using the re module (e5bd091) - Fix minor typo in read_report/1 (07c6ad4) There is no need to introduce a new function rb:re/1, but rb:grep/1 can be modified to use the 're' module. Also, it should not be necessary to compile the regular expression beforehand. ------------------------------------------------------------ [Cooking] * at/odbc_osx_fix (odbc) (2009-11-30) 1 commit - Workaround broken ODBC implementation on OSX 10.5 (88c71c9) Being reviewed. * bg/erl_bif_types (hipe) (2010-02-03) 2 commits - Add types for erts_debug:dist_ext_to_term/2 (249f73a) - erl_bif_types: add 'http_bin' as a valid packet option (7c72a77) * bg/temp-hipe-fixes (compiler, hipe, kernel) (2010-01-31) 4 commits - Crudely fix return type for the lists:key{search,find,member}/3 (5c9a017) - seq_trace_SUITE: Don't native-compile (c569f85) - Disable native code if on_load is used (101dcc8) - andor_SUITE: Don't native-compile (cb44959) This branch contains temporary workarounds to avoid failing test cases and is never intended to graduate. * db/tv_nthtail_fix (tv) (2009-12-01) 1 commit - Fix for tv which restarts while trying to open a table (abe0759) The author of this topic branch has found the real reason for the problem and has made some progress in resolving it. * dgud/emacs-mode (tools) (2010-01-26) 2 commits - Added the old style skeletons, and a variable to change. (2cbe853) - Move code skeletons to a separate file. Use the updated skeletons from the Erlware Mode. (b4fd161) * egil/lcnt (erts, tools) (2010-02-01) 6 commits - Add test suite for lcnt in tools (e26bedd) - Add lcnt:rt_opt/1 bindings to erts_debug (c24d153) - Add runtime option to enable/disable lcnt stats (dd665ed) - Add auto width on string output (12990af) - Add lcnt documentation (41d4913) - Add lock profiling tool (c41fa05) * gc/hipe_darwin_amd64 (erts) (2010-01-12) 4 commits - Fix hipe memory allocation problems on darwin/amd64. (32ca14f) - Porting x86 darwin fixes to amd64 darwin hipe asm/m4 code. (427f1db) - Automatically enable hipe for darwin/amd64 builds. (43ae4b7) - Allow configure to enable_hipe for darwin/amd64 builds. (733ce7a) Will probably graduate next week. * jn/sasl-format_report (sasl) (2010-01-27) 1 commit - Added sasl_report:format_report/3 to return a string rather than send an io (ee8f29f) * jp/dependencies_makefile (compiler, erts) (2010-01-12) 2 commits - squash! fix snprintf for Windows (bf2708d) - Add dependencies Makefile generation to erlc(1) and compile(3) (7cbc437) Approved in principle by the OTP Technical Board; now awaits a closer review of the implementation. * pan/enable_m32_build (erts, otp) (2009-12-08) 1 commit - Teach configure --enable-m32-build for forcing a 32-bit build (15c7817) * po/odbc-update-delete-no-rows (odbc) (2009-11-26) 1 commit - SQL_NO_DATA and SQLSTATE 00000 indicate success for updates/deletes that affect no rows (1a564f8) Being reviewed. * rani/inet_gethostbyname_fixes (erts, kernel) (2010-02-04) 6 commits - inet_res_SUITE: testcase fixes for legacy DNS resolver (Solaris 8) (30c73ff) - inet_res: /etc/resolv.conf: use domain as default search list (3311a87) - inet: delayed/avoided read of /etc/resolv.conf and /etc/hosts (6212bcc) - inet_gethost_native: workaround for empty result hostname on MacOS X (b0a6854) - inet_res_SUITE: testcase fix for empty domain name (9bf452c) - inet:gethostbyname improved to parse IP strings and look up own hostname (3fbd6e6) * sc/sctp-connect-nowait (kernel) (2010-02-05) 3 commits - fixup: gen_sctp_SUITE a little bit more flexible (a359f88) - Refined implementation of non-blocking SCTP association establishment with the new API gen_sctp:connect_init/4,5. Included are documentation and test cases. Tests passed ok under Solaris 10/x86. (6f2423c) - Implement a non-blocking SCTP connect (ae3d819) Should graduate early next week. From videlalvaro@REDACTED Sat Feb 6 13:06:15 2010 From: videlalvaro@REDACTED (Alvaro Videla) Date: Sat, 6 Feb 2010 20:06:15 +0800 Subject: [erlang-patches] Improvements in the Report Browser In-Reply-To: <6672d0161002020238l1d9dc100g6d9f632c1dea22a2@mail.gmail.com> References: <110EAD81-677A-479D-B797-72F9A55BDB70@gmail.com> <6672d0160912210230w5443d4a1wb3f3c249d792daf5@mail.gmail.com> <6B2422AA-528F-497B-B71C-DAC2699C75EB@gmail.com> <6672d0160912220902h491baaedi1c472e8945b69245@mail.gmail.com> <87CCF4EE-A7AB-402F-9D05-929739052C85@gmail.com> <6672d0161001041018n75c904ccwf298a1640765533d@mail.gmail.com> <6672d0161002020238l1d9dc100g6d9f632c1dea22a2@mail.gmail.com> Message-ID: I just pushed the changes here: http://github.com/videlalvaro/otp/tree/va/rb-improvements Thanks for the feedback, Alvaro On Feb 2, 2010, at 6:38 PM, Bj?rn Gustavsson wrote: > 2010/1/4 Bj?rn Gustavsson : >> On Fri, Dec 25, 2009 at 10:03 AM, Videla Alvaro wrote: >> >>> http://github.com/videlalvaro/otp/tree/va/rb-improvements >> >> Thanks! Included in 'pu'. > > We have now reviewed this patch. In order to include it in the next release, > the following changes must be made: > > * There is no need for a new function to replace rb:grep/1. Since rb:grep/1 is > (usually) used manually to browse reports, the minor potential incompatibility > introduced by changing the regular expression engine is acceptable even in > a minor release. Therefore, rb:grep/1 should be changed to use the re module. > > * It must not be a requirement to pass a compiled regular expression. rb:grep/1 > should still accept a string, and automatically compile it. It could > also optionally > allow a compiled regular expression. (The new rb:filter/1 function should work > in the same way.) > > -- > Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From bgustavsson@REDACTED Sat Feb 6 13:24:23 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Sat, 6 Feb 2010 13:24:23 +0100 Subject: [erlang-patches] Improvements in the Report Browser In-Reply-To: References: <110EAD81-677A-479D-B797-72F9A55BDB70@gmail.com> <6672d0160912210230w5443d4a1wb3f3c249d792daf5@mail.gmail.com> <6B2422AA-528F-497B-B71C-DAC2699C75EB@gmail.com> <6672d0160912220902h491baaedi1c472e8945b69245@mail.gmail.com> <87CCF4EE-A7AB-402F-9D05-929739052C85@gmail.com> <6672d0161001041018n75c904ccwf298a1640765533d@mail.gmail.com> <6672d0161002020238l1d9dc100g6d9f632c1dea22a2@mail.gmail.com> Message-ID: <6672d0161002060424k6c2e2c4n87d34bfd0a8107d3@mail.gmail.com> 2010/2/6 Alvaro Videla : > I just pushed the changes here: > > http://github.com/videlalvaro/otp/tree/va/rb-improvements > Thanks! It will be included in pu the next I'll push it (probably tomorrow). -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From tuncer.ayaz@REDACTED Tue Feb 9 00:51:58 2010 From: tuncer.ayaz@REDACTED (Tuncer Ayaz) Date: Tue, 9 Feb 2010 00:51:58 +0100 Subject: Make newline in RHS of lib/configure sed substition more portable Message-ID: <4ac8254d1002081551n7a5ba8cek553f6ac52a5690f3@mail.gmail.com> Commit http://github.com/erlang/otp/commit/1b78e8 seems to have introduced a non-portable RHS parameter to a sed substitution string for expanding lib/configure's template file. The escaped newline character in the substitution string does not work as expected. On a Linux machine it inserts an additional blank line and on OSX it only inserts a leading 'n' which gives you a broken lib/configure script. I've successfully tested the fix on Linux and OSX. git fetch git://github.com/tuncer/otp.git otp_build_sed_newline From jay@REDACTED Tue Feb 9 02:53:57 2010 From: jay@REDACTED (Jay Nelson) Date: Mon, 8 Feb 2010 17:53:57 -0800 Subject: make release_docs failure Message-ID: When I tried to run 'make release_docs' I get an error that 'fo' is not found. It is when the first pdf (stdlib) is being converted. How do I get the 'fo' executable? I am running on Mac OS X. jay From hio@REDACTED Tue Feb 9 07:22:03 2010 From: hio@REDACTED (YAMASHINA Hio) Date: Tue, 9 Feb 2010 15:22:03 +0900 Subject: patch for packet size at ssl:send of new_ssl implementation Message-ID: <20100209152203.78c57dc6.hio@hio.jp> Hi all. Even if there is {packet,PacketType} option enabled, no size bytes are prepened by ssl:send with {ssl_imp,new}. If using with {ssl_imp,old}, it is prepended correctly. git fetch git://github.com/hio/otp.git packet_option_for_new_ssl_send Reproduction code and samlpe ssl certification are attached. Ceritication was created with:: > openssl genrsa -out server_key.pem > seq 1 7 | sed -e s/.// | openssl req -new -x509 \ -key server_key.pem -out server_cert.pem && echo Cient process sends some words with some {packet,N} option. Server process receives {packet,raw}, thus it can see prepended size bytes. Current implementation outputs:: client: send: (packet=0) "test" server: recv: (packet=?) "test" client: send: (packet=2) "AAAA" server: recv: (packet=?) "AAAA" client: send: (packet=1) "BBBB" server: recv: (packet=?) "BBBB" client: send: (packet=4) "CCCC" server: recv: (packet=?) "CCCC" client: send: (packet=0) "DDDD" server: recv: (packet=?) "DDDD" Server process receives just client send ones. After patched:: client: send: (packet=0) "test" server: recv: (packet=?) "test" client: send: (packet=2) "AAAA" server: recv: (packet=2) "\x00\x04AAAA" client: send: (packet=1) "BBBB" server: recv: (packet=1) "\x04BBBB" client: send: (packet=4) "CCCC" server: recv: (packet=4) "\x00\x00\x00\x04CCCC" client: send: (packet=0) "DDDD" server: recv: (packet=?) "DDDD" Size bytes are prepended currectly. Regards. -- YAMASHINA Hio , -------------- next part -------------- A non-text attachment was scrubbed... Name: server_key.pem Type: application/octet-stream Size: 497 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: test_ssl.erl Type: application/octet-stream Size: 3700 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: server_cert.pem Type: application/octet-stream Size: 814 bytes Desc: not available URL: From bgustavsson@REDACTED Tue Feb 9 07:39:33 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Tue, 9 Feb 2010 07:39:33 +0100 Subject: [erlang-patches] Make newline in RHS of lib/configure sed substition more portable In-Reply-To: <4ac8254d1002081551n7a5ba8cek553f6ac52a5690f3@mail.gmail.com> References: <4ac8254d1002081551n7a5ba8cek553f6ac52a5690f3@mail.gmail.com> Message-ID: <6672d0161002082239x68cc2194x1c09415d4e0fa0b9@mail.gmail.com> On Tue, Feb 9, 2010 at 12:51 AM, Tuncer Ayaz wrote: > The escaped newline character in the substitution string > does not work as expected. > On a Linux machine it inserts an additional blank line > and on OSX it only inserts a leading 'n' which gives you > a broken lib/configure script. > > I've successfully tested the fix on Linux and OSX. > > git fetch git://github.com/tuncer/otp.git otp_build_sed_newline Thanks for noticing this. I have amended the commit message to include the commit ID of the commit that introduced the non-portable NL. I am not an expert on shell script portability, but I noted that you use a backslash in your definition of the NL variable: NL="\ " while all shell scripts in the Git project (source code and test suites), don't use a backslash and single quotes: LF=' ' Is is just matter of taste or is the variant used in Git (even) more portable? -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From bgustavsson@REDACTED Tue Feb 9 07:59:10 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Tue, 9 Feb 2010 07:59:10 +0100 Subject: [erlang-patches] patch for packet size at ssl:send of new_ssl implementation In-Reply-To: <20100209152203.78c57dc6.hio@hio.jp> References: <20100209152203.78c57dc6.hio@hio.jp> Message-ID: <6672d0161002082259j5c100ed4o90abfc693930088@mail.gmail.com> On Tue, Feb 9, 2010 at 7:22 AM, YAMASHINA Hio wrote: > > Even if there is {packet,PacketType} option enabled, > no size bytes are prepened by ssl:send with {ssl_imp,new}. > If using with {ssl_imp,old}, it is prepended correctly. > > git fetch git://github.com/hio/otp.git packet_option_for_new_ssl_send Thanks! Will include in pu. I have added a commit where I fixed up the indentation and coding style. -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From tuncer.ayaz@REDACTED Tue Feb 9 14:04:29 2010 From: tuncer.ayaz@REDACTED (Tuncer Ayaz) Date: Tue, 9 Feb 2010 14:04:29 +0100 Subject: [erlang-patches] Make newline in RHS of lib/configure sed substition more portable In-Reply-To: <6672d0161002082239x68cc2194x1c09415d4e0fa0b9@mail.gmail.com> References: <4ac8254d1002081551n7a5ba8cek553f6ac52a5690f3@mail.gmail.com> <6672d0161002082239x68cc2194x1c09415d4e0fa0b9@mail.gmail.com> Message-ID: <4ac8254d1002090504g3ad815efv2a1249155747e533@mail.gmail.com> 2010/2/9 Bj?rn Gustavsson : > On Tue, Feb 9, 2010 at 12:51 AM, Tuncer Ayaz wrote: >> The escaped newline character in the substitution string >> does not work as expected. >> On a Linux machine it inserts an additional blank line >> and on OSX it only inserts a leading 'n' which gives you >> a broken lib/configure script. >> >> I've successfully tested the fix on Linux and OSX. >> >> git fetch git://github.com/tuncer/otp.git otp_build_sed_newline > > Thanks for noticing this. I have amended the commit message to include > the commit ID of the commit that introduced the non-portable NL. Thanks, commit message adjusted in tuncer/erlang@REDACTED > I am not an expert on shell script portability, but I noted that you use > a backslash in your definition of the NL variable: > > NL="\ > " > > while all shell scripts in the Git project (source code and test suites), > don't use a backslash and single quotes: > > LF=' > ' > > Is is just matter of taste or is the variant used in Git (even) more portable? Without the BSL sed complains about the RHS: "unescaped newline inside substitute pattern" When I use single quotes on both Linux and OSX it works but inserts an additional blank line after the break. From smith.winston.101@REDACTED Tue Feb 9 16:26:48 2010 From: smith.winston.101@REDACTED (Winston Smith) Date: Tue, 9 Feb 2010 10:26:48 -0500 Subject: Submitting patches via git format-patch/git am Message-ID: I know that you suggest the best way to submit patches is to send a link to a git repository that contains the patch(es), but do/will you guys also accept patches via email in git "format-patch" form? I've used this with other open source projects and it works very well. For those unfamiliar with this, see: man git-format-patch man git-am Thanks! From bgustavsson@REDACTED Tue Feb 9 17:07:05 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Tue, 9 Feb 2010 17:07:05 +0100 Subject: [erlang-patches] Make newline in RHS of lib/configure sed substition more portable In-Reply-To: <4ac8254d1002090504g3ad815efv2a1249155747e533@mail.gmail.com> References: <4ac8254d1002081551n7a5ba8cek553f6ac52a5690f3@mail.gmail.com> <6672d0161002082239x68cc2194x1c09415d4e0fa0b9@mail.gmail.com> <4ac8254d1002090504g3ad815efv2a1249155747e533@mail.gmail.com> Message-ID: <6672d0161002090807h343af0b2p5a4c6c604372651e@mail.gmail.com> 2010/2/9 Tuncer Ayaz : > > Without the BSL sed complains about the RHS: > "unescaped newline inside substitute pattern" > > When I use single quotes on both Linux and OSX it works but > inserts an additional blank line after the break. Thanks! We will use your patch without changes, then. I have applied it directly to the dev branch. -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From bgustavsson@REDACTED Tue Feb 9 17:21:13 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Tue, 9 Feb 2010 17:21:13 +0100 Subject: [erlang-patches] Submitting patches via git format-patch/git am In-Reply-To: References: Message-ID: <6672d0161002090821l3c372386s5d4cb77f2bb36bd9@mail.gmail.com> On Tue, Feb 9, 2010 at 4:26 PM, Winston Smith wrote: > I know that you suggest the best way to submit patches is to send a > link to a git repository that contains the patch(es), but do/will you > guys also accept patches via email in git "format-patch" form? We do. > I've used this with other open source projects and it works very well. Yes. It makes it much easier to review a patch and comment it line-by-line. The drawback is that it can be tricky to configure some email clients so that they don't line wrap or remove spaces in patches (making them impossible to apply), so therefore we didn't want to require patches to be sent that way. -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From jonas.falkevik@REDACTED Tue Feb 9 17:33:28 2010 From: jonas.falkevik@REDACTED (Jonas Falkevik) Date: Tue, 9 Feb 2010 17:33:28 +0100 Subject: erl_bif_timer.c - setup_bif_timer() - returning ref upon erts_pid2proc failure? Message-ID: Hi, It looks like the function setup_bif_timer could return a ref without setting up the timer. Which is used by the erlang:send_after/3 and could in theory return a ref where it should not. Am I missing something? http://github.com/falkevik/otp/commit/61f57e2e25a Regards, Jonas Falkevik From smith.winston.101@REDACTED Tue Feb 9 17:52:40 2010 From: smith.winston.101@REDACTED (Winston Smith) Date: Tue, 9 Feb 2010 11:52:40 -0500 Subject: [PATCH] Add cross-compile configuration for Atmel AVR32-Linux (NGW100) Message-ID: Implemented erl-xcomp-avr32-atmel-linux-gnu.conf to support cross compilation of Erlang on Atmel's AVR32-Linux: http://www.atmel.no/buildroot/ Built using a Ubuntu 9.10 host, with buildroot-2.3.0 and tested on on Atmel's NGW100 development board. Please let me know if there are any issues. Many thanks to Rickard Green for enabling this! W. -------------- next part -------------- A non-text attachment was scrubbed... Name: 0001-Add-cross-compile-configuration-for-Atmel-AVR32-Linu.patch Type: text/x-patch Size: 11728 bytes Desc: not available URL: From jonas.falkevik@REDACTED Tue Feb 9 19:29:15 2010 From: jonas.falkevik@REDACTED (Jonas Falkevik) Date: Tue, 9 Feb 2010 19:29:15 +0100 Subject: erl_bif_timer.c - setup_bif_timer() - returning ref upon erts_pid2proc failure? In-Reply-To: References: Message-ID: This breaks the spec where the pid could be dead. Sorry. On Feb 9, 2010, at 5:33 PM, Jonas Falkevik wrote: > Hi, > It looks like the function setup_bif_timer could return a ref > without setting up the timer. > > Which is used by the erlang:send_after/3 and could in theory return > a ref where it should not. > > Am I missing something? > > http://github.com/falkevik/otp/commit/61f57e2e25a > > > Regards, > Jonas Falkevik From bgustavsson@REDACTED Tue Feb 9 19:53:07 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Tue, 9 Feb 2010 19:53:07 +0100 Subject: [erlang-patches] [PATCH] Add cross-compile configuration for Atmel AVR32-Linux (NGW100) In-Reply-To: References: Message-ID: <6672d0161002091053j30b9618cu2c8c454d30d45e41@mail.gmail.com> On Tue, Feb 9, 2010 at 5:52 PM, Winston Smith wrote: > Implemented erl-xcomp-avr32-atmel-linux-gnu.conf to support cross > compilation of Erlang on Atmel's AVR32-Linux: > ? ?http://www.atmel.no/buildroot/ > > Built using a Ubuntu 9.10 host, with buildroot-2.3.0 and tested on on > Atmel's NGW100 development board. Thanks! Included in 'pu'. -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From bgustavsson@REDACTED Tue Feb 9 20:07:59 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Tue, 9 Feb 2010 20:07:59 +0100 Subject: [erlang-patches] Submitting patches via git format-patch/git am In-Reply-To: <6672d0161002090821l3c372386s5d4cb77f2bb36bd9@mail.gmail.com> References: <6672d0161002090821l3c372386s5d4cb77f2bb36bd9@mail.gmail.com> Message-ID: <6672d0161002091107x47360e3co69db7e115dfba437@mail.gmail.com> 2010/2/9 Bj?rn Gustavsson : > On Tue, Feb 9, 2010 at 4:26 PM, Winston Smith > wrote: >> I know that you suggest the best way to submit patches is to send a >> link to a git repository that contains the patch(es), but do/will you >> guys also accept patches via email in git "format-patch" form? > > We do. > When I said that, I assumed that you meant *inlined* patches. Inlined patches in an email message can be reviewed and commented on easily and a several patches in a series can be saved to a mail folder and applied in one go. I see no advantages with *attached* patches. Was I too vague in my second paragraph? What I meant was that we prefer to be able to do "git fetch", because inlined patches may be corrupted by the email client and we would constantly need to ask people to re-send them. So if you know how to send patches inline without your email client garbling them (or using "git send-email"), feel free to send them that way. But please avoid attached patches. -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From smith.winston.101@REDACTED Tue Feb 9 22:44:40 2010 From: smith.winston.101@REDACTED (Winston Smith) Date: Tue, 9 Feb 2010 16:44:40 -0500 Subject: [erlang-patches] Submitting patches via git format-patch/git am In-Reply-To: <6672d0161002091107x47360e3co69db7e115dfba437@mail.gmail.com> References: <6672d0161002090821l3c372386s5d4cb77f2bb36bd9@mail.gmail.com> <6672d0161002091107x47360e3co69db7e115dfba437@mail.gmail.com> Message-ID: 2010/2/9 Bj?rn Gustavsson : > When I said that, I assumed that you meant *inlined* patches. > Inlined patches in an email message can be reviewed and > commented on easily and a several patches in a series > can be saved to a mail folder and applied in one go. Yes, that makes sense! > I see no advantages with *attached* patches. > > Was I too vague in my second paragraph? What I meant > was that we prefer to be able to do "git fetch", because > inlined patches may be corrupted by the email client and > we would constantly need to ask people to re-send them. I did understand what you meant, however I assumed that by attaching, I thought I'd avoid any possible corruption! > So if you know how to send patches inline without your > email client garbling them (or using "git send-email"), > feel free to send them that way. But please avoid > attached patches. Absolutely, I've installed git send-email and I'll use it next time. Thanks for the guidance. From bernie@REDACTED Wed Feb 10 01:04:46 2010 From: bernie@REDACTED (Bernard Duggan) Date: Wed, 10 Feb 2010 11:04:46 +1100 Subject: Minor documentation fixes Message-ID: <4B71F81E.4060202@m5net.com> This patch just fixes a few minor documentation typos that I've stumbled across in the current release. git fetch git://github.com/bernardd/otp.git doc_fixes Cheers, Bernard From zl9d97p02@REDACTED Wed Feb 10 01:11:31 2010 From: zl9d97p02@REDACTED (Simon Cornish) Date: Tue, 9 Feb 2010 16:11:31 -0800 Subject: ASN.1 BER decode of NULL type broken Message-ID: <22249-1265760692-327986@sneakemail.com> Hi, The asn1rt_bin_ber:decode_null function is broken for many cases. Example: when a NULL type element is present and not the last element in a sequence. Attached a sample ASN.1 module, demonstration of the fault and patch against ccase/r13b04_dev Sorry I don't have time today to wrestle with git to get this into a branch. Hope it's simple enough to be included directly. Regards, Simon -------------- next part -------------- liveoak:asn1dn simon$ which erl /Users/simon/src/otp/git/r13b04_dev/bin/erl liveoak:asn1dn simon$ erlc +bber_bin -pa . DecodeNull.asn Erlang ASN.1 version "1.6.12" compiling "/private/tmp/simon/asn1dn/DecodeNull.asn" Compiler Options: [ber,report_errors, {cwd,"/private/tmp/simon/asn1dn"}, {outdir,"/private/tmp/simon/asn1dn"}, bber_bin, {i,"."}, {i,"/private/tmp/simon/asn1dn"}] --{generated,"/private/tmp/simon/asn1dn/DecodeNull.asn1db"}-- --{generated,"/private/tmp/simon/asn1dn/DecodeNull.hrl"}-- --{generated,"/private/tmp/simon/asn1dn/DecodeNull.erl"}-- ok liveoak:asn1dn simon$ erl -pa . Erlang R13B04 (erts-5.7.5) [source] [smp:2:2] [rq:2] [async-threads:0] [hipe] [kernel-poll:false] Eshell V5.7.5 (abort with ^G) 1> rr('DecodeNull'). ['Seq1'] 2> S = #'Seq1'{opt11 = 'NULL', opt12 = 'NULL'}. #'Seq1'{opt11 = 'NULL',opt12 = 'NULL'} 3> {ok, Enc} = asn1rt:encode('DecodeNull', 'Seq1', S). {ok,[48,8,171,2,5,0,172,2,5,0]} 4> asn1rt:decode('DecodeNull', 'Seq1', Enc). {error,{asn1,{unexpected,<<172,2,5,0>>}}} 5> -------------- next part -------------- A non-text attachment was scrubbed... Name: DecodeNull.asn Type: application/octet-stream Size: 176 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: asn1-decode_null.patch Type: application/octet-stream Size: 671 bytes Desc: not available URL: From kenji.rikitake@REDACTED Wed Feb 10 06:13:04 2010 From: kenji.rikitake@REDACTED (Kenji Rikitake) Date: Wed, 10 Feb 2010 14:13:04 +0900 Subject: Fix calendar:local_time_to_universal_time_dst/2 bug of return value -1 from mktime Message-ID: <20100210051304.GA48531@k2r.org> I am struggling to merge FreeBSD port patches into my GitHub repository. This one is a long-standing fix which is proposed and fixed by Paul Guyot. I do not claim any credit for this patch. This patch has been tested for at least a year in FreeBSD port of Erlang (lang/erlang). git fetch git://github.com/jj1bdx/otp.git local_to_univ http://github.com/jj1bdx/otp/commit/1461281981f9e9f5542138b6807cbed603a38d99 Patch originally created by Paul Guyot Submitted at the following erlang-bugs article and the attachment containing this patch: http://www.erlang.org/pipermail/erlang-bugs/2008-November/001077.html http://www.erlang.org/pipermail/erlang-bugs/attachments/20081101/c1bab4d9/attachment.obj Regards, Kenji Rikitake From bgustavsson@REDACTED Wed Feb 10 07:49:34 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Wed, 10 Feb 2010 07:49:34 +0100 Subject: [erlang-patches] Minor documentation fixes In-Reply-To: <4B71F81E.4060202@m5net.com> References: <4B71F81E.4060202@m5net.com> Message-ID: <6672d0161002092249p49f5eac2o732dae4a2886d3f0@mail.gmail.com> On Wed, Feb 10, 2010 at 1:04 AM, Bernard Duggan wrote: > This patch just fixes a few minor documentation typos that I've stumbled across in the current release. > > git fetch git://github.com/bernardd/otp.git doc_fixes Thanks! Will be included in pu. -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From bgustavsson@REDACTED Wed Feb 10 07:57:52 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Wed, 10 Feb 2010 07:57:52 +0100 Subject: [erlang-patches] Fix calendar:local_time_to_universal_time_dst/2 bug of return value -1 from mktime In-Reply-To: <20100210051304.GA48531@k2r.org> References: <20100210051304.GA48531@k2r.org> Message-ID: <6672d0161002092257s10b3dbd1u1fd225c93ef17048@mail.gmail.com> On Wed, Feb 10, 2010 at 6:13 AM, Kenji Rikitake wrote: > I am struggling to merge FreeBSD port patches into my GitHub repository. > This one is a long-standing fix which is proposed and fixed by > Paul Guyot. I do not claim any credit for this patch. > This patch has been tested for at least a year in FreeBSD port of Erlang (lang/erlang). > > git fetch git://github.com/jj1bdx/otp.git local_to_univ Thanks! Will include in pu. -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From kenji.rikitake@REDACTED Wed Feb 10 08:47:39 2010 From: kenji.rikitake@REDACTED (Kenji Rikitake) Date: Wed, 10 Feb 2010 16:47:39 +0900 Subject: Change lib/wx/configure.in DEBUG_CFLAGS and CFLAGS to remove C compiler "-g" from non-debug build Message-ID: <20100210074739.GA69344@k2r.org> git fetch git://github.com/jj1bdx/otp.git wx_debugflag_fix Change lib/wx/configure.in DEBUG_CFLAGS and CFLAGS to remove C compiler "-g" from non-debug build This patch removes "-g" from the CFLAGS for building wx. The UNIX portion of this patch has been tested and working on the FreeBSD port. For other OSes this is still untested. This patch is originally taken from FreeBSD port patch files. Regards, Kenji Rikitake From kenji.rikitake@REDACTED Wed Feb 10 08:58:32 2010 From: kenji.rikitake@REDACTED (Kenji Rikitake) Date: Wed, 10 Feb 2010 16:58:32 +0900 Subject: Fix typos of lib/sasl/src/rb.erl and lib/snmp/src/misc/snmp_config.erl Message-ID: <20100210075832.GB69344@k2r.org> git fetch git://github.com/jj1bdx/otp.git sasl_snmp_typofix This patch fixes typos of the following two files: lib/sasl/src/rb.erl lib/snmp/src/misc/snmp_config.erl This patch is originally taken from FreeBSD port patch files. Regards, Kenji Rikitake From kenji.rikitake@REDACTED Wed Feb 10 09:11:35 2010 From: kenji.rikitake@REDACTED (Kenji Rikitake) Date: Wed, 10 Feb 2010 17:11:35 +0900 Subject: Fix ODBC compilation error on FreeBSD by adding netinet/in.h to #include macros Message-ID: <20100210081135.GC69344@k2r.org> git fetch git://github.com/jj1bdx/otp.git odbc_netinet Fix ODBC compilation error on FreeBSD by adding netinet/in.h to #include macros This fix modifies lib/odbc/c_src_odbcserver.c This has been tested on FreeBSD port, and will not affect Linux. This patch is originally taken from FreeBSD port patch files, by Giacomo Olgeni. See http://www.erlang.org/cgi-bin/ezmlm-cgi?3:mss:335:200811:hijljjocpmpkhgkhfkdl for the details too Regards, Kenji Rikitake From kenji.rikitake@REDACTED Wed Feb 10 09:25:28 2010 From: kenji.rikitake@REDACTED (Kenji Rikitake) Date: Wed, 10 Feb 2010 17:25:28 +0900 Subject: Fix semantic error of http:request/4 when stream_to() in Options is set to {self, once} Message-ID: <20100210082528.GD69344@k2r.org> git fetch git://github.com/jj1bdx/otp.git r13b03_http_client Fix semantic error of http:request/4 when stream_to() in Options is set to {self, once} Fix the bug in lib/inets/src/http_client/http.erl of the following behavior: error handling when stream_to() = {self} in Options was not executed whenb the stream_to() option is set to {self, once}. Originally from FreeBSD port patch files. See also the "Stream" section of http://www.erlang.org/doc/man/http.html#request-4 Regards, Kenji Rikitake From dgud@REDACTED Wed Feb 10 09:32:41 2010 From: dgud@REDACTED (Dan Gudmundsson) Date: Wed, 10 Feb 2010 09:32:41 +0100 Subject: [erlang-patches] Change lib/wx/configure.in DEBUG_CFLAGS and CFLAGS to remove C compiler "-g" from non-debug build In-Reply-To: <20100210074739.GA69344@k2r.org> References: <20100210074739.GA69344@k2r.org> Message-ID: <93df43b61002100032h37cf468el5ce30906970ca450@mail.gmail.com> We don't like the idea to by default remove the debug info, it's the only chance we have to find bugs and specially in wx where users errors in the code can crash the emulator. /Dan On Wed, Feb 10, 2010 at 8:47 AM, Kenji Rikitake wrote: > git fetch git://github.com/jj1bdx/otp.git wx_debugflag_fix > > Change lib/wx/configure.in DEBUG_CFLAGS and CFLAGS to remove C compiler > "-g" from non-debug build > > This patch removes "-g" from the CFLAGS for building wx. > The UNIX portion of this patch has been tested and working > on the FreeBSD port. ?For other OSes this is still untested. > > This patch is originally taken from FreeBSD port patch files. > > Regards, > Kenji Rikitake > > > ________________________________________________________________ > erlang-patches (at) erlang.org mailing list. > See http://www.erlang.org/faq.html > To unsubscribe; mailto:erlang-patches-unsubscribe@REDACTED > > From kenji.rikitake@REDACTED Wed Feb 10 09:38:28 2010 From: kenji.rikitake@REDACTED (Kenji Rikitake) Date: Wed, 10 Feb 2010 17:38:28 +0900 Subject: Change lib/gs/src/tool_utils.erl to select multiple browsers other than Netscape Message-ID: <20100210083828.GE69344@k2r.org> git fetch git://github.com/jj1bdx/otp.git r13b03_gs_browser Change lib/gs/src/tool_utils.erl to select multiple browsers other than Netscape This patch changes lib/gs/src/tools_utils.erl to choose multiple browsers as in newly-added BROWSERS macro and internal functions find_browser/1 and unix_url_command/1 so that the users can choose the browsers other than currenly hard-coded "Netscape". Originally from FreeBSD port patch files. Regards, Kenji Rikitake From jay@REDACTED Wed Feb 10 08:45:45 2010 From: jay@REDACTED (Jay Nelson) Date: Tue, 9 Feb 2010 23:45:45 -0800 Subject: Fixed test suite for supervisor_child_count_only Message-ID: <690BED90-6D48-44D6-AD81-D7272B3D5ADD@duomark.com> supervisor_SUITE will skip the count_children_memory test if +Meamin is used. git fetch git://github.com/jaynel/otp.git supervisor_child_count_only jay From kenji.rikitake@REDACTED Wed Feb 10 09:51:20 2010 From: kenji.rikitake@REDACTED (Kenji Rikitake) Date: Wed, 10 Feb 2010 17:51:20 +0900 Subject: Fix Makefile permission problem of crypto library on FreeBSD Erlang/OTP building Message-ID: <20100210085120.GG69344@k2r.org> git fetch git://github.com/jj1bdx/otp.git r13b03_crypto_makefile This patch changes INSTALL_PROGRAM to INSTALL_DATA for the Makefile at lib/crypto/priv/obj, which discontinues the building procedure of Erlang/OTP on FreeBSD. Originally posted by Giacomo Olgeni as a part of FreeBSD port patch files. See http://www.erlang.org/cgi-bin/ezmlm-cgi?3:mss:335:200811:hijljjocpmpkhgkhfkdl for the detailed analysis. Regards, Kenji Rikitake From kenji.rikitake@REDACTED Wed Feb 10 10:04:12 2010 From: kenji.rikitake@REDACTED (Kenji Rikitake) Date: Wed, 10 Feb 2010 18:04:12 +0900 Subject: Fix the shell operator bug of "==" to "=" in erts/configure.in Message-ID: <20100210090412.GA71672@k2r.org> git fetch git://github.com/jj1bdx/otp.git r13b03_erts_configure In UNIX test command, "==" is incorrect for string comparison; "=" is the right operator. Originally pointed out by Giacomo Olgeni in the following article: http://www.erlang.org/pipermail/erlang-patches/2008-November/000334.html and available as a FreeBSD port patch. Accordingly fixed in erts/configure.in Regards, Kenji Rikitake From kenji.rikitake@REDACTED Wed Feb 10 09:48:24 2010 From: kenji.rikitake@REDACTED (Kenji Rikitake) Date: Wed, 10 Feb 2010 17:48:24 +0900 Subject: [erlang-patches] Change lib/wx/configure.in DEBUG_CFLAGS and CFLAGS to remove C compiler "-g" from non-debug build In-Reply-To: <93df43b61002100032h37cf468el5ce30906970ca450@mail.gmail.com> References: <20100210074739.GA69344@k2r.org> <93df43b61002100032h37cf468el5ce30906970ca450@mail.gmail.com> Message-ID: <20100210084824.GF69344@k2r.org> Thanks for the explanation, Dan, and I understand it. You can decline my patch about this. Regards, Kenji Rikitake In the message <93df43b61002100032h37cf468el5ce30906970ca450@REDACTED> dated Wed, Feb 10, 2010 at 09:32:17AM +0100, Dan Gudmundsson writes: > We don't like the idea to by default remove the debug info, it's the > only chance we have to > find bugs and specially in wx where users errors in the code can crash > the emulator. > > /Dan From kenji.rikitake@REDACTED Wed Feb 10 10:18:28 2010 From: kenji.rikitake@REDACTED (Kenji Rikitake) Date: Wed, 10 Feb 2010 18:18:28 +0900 Subject: Change erts/etc/unix/Install.src to explicitly set exit code to 0 at the end of the script Message-ID: <20100210091828.GA71996@k2r.org> git fetch git://github.com/jj1bdx/otp.git r13b03_installsrc_exitcode Change erts/etc/unix/Install.src to explicitly set exit code to 0 at the end of the script This patch will ensure the Install.src script to exit with the exit code 0 when the execution finished at the end of the script. Originally from FreeBSD port patch. Regards, Kenji Rikitake From lars@REDACTED Wed Feb 10 11:55:53 2010 From: lars@REDACTED (Lars Thorsen) Date: Wed, 10 Feb 2010 11:55:53 +0100 Subject: [erlang-patches] make release_docs failure In-Reply-To: References: Message-ID: <4B7290B9.5010109@erix.ericsson.se> > From: Jay Nelson > Date: Tue, Feb 9, 2010 at 2:53 AM > Subject: [erlang-patches] make release_docs failure > To: erlang-patches@REDACTED > > > When I tried to run 'make release_docs' I get an error that 'fo' is not found. > It is when the first pdf (stdlib) is being converted. > > How do I get the 'fo' executable? I am running on Mac OS X. > > jay > > > ________________________________________________________________ > erlang-patches (at) erlang.org mailing list. > See http://www.erlang.org/faq.html > To unsubscribe; mailto:erlang-patches-unsubscribe@REDACTED > Hi Jay, Apaches FOP can be fetched from "http://xmlgraphics.apache.org/fop/". Regards Lars Erlang/OTP, Ericsson AB From bgustavsson@REDACTED Wed Feb 10 13:08:52 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Wed, 10 Feb 2010 13:08:52 +0100 Subject: [erlang-patches] Change lib/gs/src/tool_utils.erl to select multiple browsers other than Netscape In-Reply-To: <20100210083828.GE69344@k2r.org> References: <20100210083828.GE69344@k2r.org> Message-ID: <6672d0161002100408o4996810qad7f25b0f763da30@mail.gmail.com> On Wed, Feb 10, 2010 at 9:38 AM, Kenji Rikitake wrote: > git fetch git://github.com/jj1bdx/otp.git r13b03_gs_browser Thanks! I will include it in pu, but it will not be considered for graduation until the function that uses the deprecated regexp module has been updated to use the re module instead. -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From bgustavsson@REDACTED Wed Feb 10 13:09:38 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Wed, 10 Feb 2010 13:09:38 +0100 Subject: [erlang-patches] Fix the shell operator bug of "==" to "=" in erts/configure.in In-Reply-To: <20100210090412.GA71672@k2r.org> References: <20100210090412.GA71672@k2r.org> Message-ID: <6672d0161002100409j57ffbf4ehc165edb3bfa6b976@mail.gmail.com> On Wed, Feb 10, 2010 at 10:04 AM, Kenji Rikitake wrote: > git fetch git://github.com/jj1bdx/otp.git r13b03_erts_configure Thanks! We will probably apply the patch directly on the dev branch. -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From bgustavsson@REDACTED Wed Feb 10 13:17:39 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Wed, 10 Feb 2010 13:17:39 +0100 Subject: [erlang-patches] Change erts/etc/unix/Install.src to explicitly set exit code to 0 at the end of the script In-Reply-To: <20100210091828.GA71996@k2r.org> References: <20100210091828.GA71996@k2r.org> Message-ID: <6672d0161002100417w3daf570eue5512a5091bbcb83@mail.gmail.com> On Wed, Feb 10, 2010 at 10:18 AM, Kenji Rikitake wrote: > git fetch git://github.com/jj1bdx/otp.git r13b03_installsrc_exitcode Thanks! Seems to be a harmless improvement, so we'll probably apply it directly to the dev branch. -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From kenji.rikitake@REDACTED Wed Feb 10 14:42:20 2010 From: kenji.rikitake@REDACTED (Kenji Rikitake) Date: Wed, 10 Feb 2010 22:42:20 +0900 Subject: Fix beam/erl_bif_re.c building bug due to wrong version of PCRE is used Message-ID: <20100210134220.GA76956@k2r.org> git fetch git://github.com/jj1bdx/otp.git r13b03_compile_pcre Reported by Giacomo Olgeni in the follwing message http://www.erlang.org/cgi-bin/ezmlm-cgi?3:mss:335:200811:hijljjocpmpkhgkhfkdl Giacomo also reported this patch is still needed in the latest build (R13B03) of FreeBSD Erlang port. Patch summary: Sorting the $(CC) command line flag so that the local $(INCLUDES) show up as the first option is required to link with correct PCRE version. This patch is mandatory for FreeBSD port build of lang/erlang (and will not harm the building procedure of other OS) Regards, Kenji Rikitake From kenji.rikitake@REDACTED Wed Feb 10 14:52:07 2010 From: kenji.rikitake@REDACTED (Kenji Rikitake) Date: Wed, 10 Feb 2010 22:52:07 +0900 Subject: Fix building error of SCTP in FreeBSD 7.1 and later Message-ID: <20100210135207.GB76956@k2r.org> git fetch git://github.com/jj1bdx/otp.git r13b03_sctp_freebsd Fix building error of SCTP in FreeBSD 7.1 and later This patch fixes building error of SCTP in FreeBSD 7.1 and later. See http://www.erlang.org/cgi-bin/ezmlm-cgi?2:mss:1257:200904:hppnfialjjdlgecdiehe for the details. Patch summary: R13A patch for solving a compilation error when building erts/emulator/drivers/common/inet_drv.c by Kenji Rikitake 12-APR-2009 Symptom solved by this patch: When building R13A in FreeBSD 7.1-RELEASE, the compiler flag HAVE_STRUCT_SCTP_PADDRPARAMS_SPP_SACKDELAY is NOT enabled. Some code in erts/emulator/drivers/common/inet_drv.c incorrectly assumes HAVE_STRUCT_SCTP_PADDRPARAMS_SPP_SACKDELAY is always true when HAVE_STRUCT_SCTP_PADDRPARAMS_SPP_FLAGS is true in config.h. This assumption causes a compilation error. Regards, Kenji Rikitake From kenji.rikitake@REDACTED Wed Feb 10 15:08:25 2010 From: kenji.rikitake@REDACTED (Kenji Rikitake) Date: Wed, 10 Feb 2010 23:08:25 +0900 Subject: My comment and acknowledgments of the FreeBSD patches to R13B03 In-Reply-To: <4B6C6E9B.9020405@gmail.com> References: <4B6C6E9B.9020405@gmail.com> Message-ID: <20100210140825.GC76956@k2r.org> The reason I decided to take a risk of submitting many patches today from FreeBSD Erlang port was due to the deadline announced as follows: In the message <4B6C6E9B.9020405@REDACTED> dated Fri, Feb 05, 2010 at 08:16:19PM +0100, Bjorn Gustavsson writes: > * February 10 (next Wednesday) there will be a feature > freeze the erts, kernel, stdlib, and compiler applications. > We hope to graduate several branches early next week > before the freeze. The OTP team suggested creating individual branch for each problem, so I followed their suggestion and posted a number of messages today. I would like to thank Giacomo Olgeni and Paul Guyot for their kind approval and permission for submitting them to my Github repository. Regards, Kenji Rikitake (jj1bdx at Github) From bgustavsson@REDACTED Wed Feb 10 15:26:31 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Wed, 10 Feb 2010 15:26:31 +0100 Subject: [erlang-patches] Fix calendar:local_time_to_universal_time_dst/2 bug of return value -1 from mktime In-Reply-To: <6672d0161002092257s10b3dbd1u1fd225c93ef17048@mail.gmail.com> References: <20100210051304.GA48531@k2r.org> <6672d0161002092257s10b3dbd1u1fd225c93ef17048@mail.gmail.com> Message-ID: <6672d0161002100626y36ab72f4oc451c9478865be75@mail.gmail.com> 2010/2/10 Bj?rn Gustavsson : > On Wed, Feb 10, 2010 at 6:13 AM, Kenji Rikitake wrote: >> I am struggling to merge FreeBSD port patches into my GitHub repository. >> This one is a long-standing fix which is proposed and fixed by >> Paul Guyot. I do not claim any credit for this patch. >> This patch has been tested for at least a year in FreeBSD port of Erlang (lang/erlang). >> >> git fetch git://github.com/jj1bdx/otp.git local_to_univ > > Thanks! Will include in pu. Result from our review: We don't want the change in calendar.erl. The documentation says that local_time_to_universal_time_dst/1 returns an empty list for a time that does not exist. We don't want the same return value for a valid time that cannot be handled because of an implementation detail. We may consider using the patch for erl_time_sup.c, if you can provide some information to show that it is really needed. Note that the BIF does a range check of its arguments. Are there some values of the arguments that slip through the range check and cause mktime() to return -1? Since such a change is considered a bug fix, we can accept it after the feature freeze today. -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From jean-sebastien.pedron@REDACTED Wed Feb 10 15:39:20 2010 From: jean-sebastien.pedron@REDACTED (=?UTF-8?B?SmVhbi1Tw6liYXN0aWVuIFDDqWRyb24=?=) Date: Wed, 10 Feb 2010 15:39:20 +0100 Subject: Fix error handling of httpc_manager and its starter process Message-ID: <4B72C518.8040406@dumbbell.fr> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi, Here is a patch that fixes error handling of httpc_manager when there's a problem with the starter process or httpc_handler. The symptoms were either no answer sent to the caller or multiple answers sent for the same request. The branch is named "inets_httpc_manager_error_handling": git fetch git://github.com/dumbbell/otp.git inets_httpc_manager_error_handling Regards, - -- Jean-S?bastien P?dron http://www.dumbbell.fr/ PGP Key: http://www.dumbbell.fr/pgp/pubkey.asc -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.10 (FreeBSD) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAktyxRgACgkQa+xGJsFYOlNL7gCfTwNcanE0al3WHx9WDKSacoUq Jf8An383GgDxWXbG4ffeGzv0X9b84K0r =LeGn -----END PGP SIGNATURE----- From bgustavsson@REDACTED Wed Feb 10 15:55:34 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Wed, 10 Feb 2010 15:55:34 +0100 Subject: [erlang-patches] Fixed test suite for supervisor_child_count_only In-Reply-To: <690BED90-6D48-44D6-AD81-D7272B3D5ADD@duomark.com> References: <690BED90-6D48-44D6-AD81-D7272B3D5ADD@duomark.com> Message-ID: <6672d0161002100655t6b167a25wa3cd9c344ba5741@mail.gmail.com> On Wed, Feb 10, 2010 at 8:45 AM, Jay Nelson wrote: > supervisor_SUITE will skip the count_children_memory test if +Meamin is > used. > > git fetch git://github.com/jaynel/otp.git supervisor_child_count_only Thanks! Will be included in pu. We found a few stylistic problems that we want fixed before we can graduate the branch. I have done the fixes, but want your approval before I include them. (I intend to combine them with your commit.) If you approve to the changes below, just reply to this email and tell me that you approve, and we should be able to graduate this branch before the feature freeze. Date: Wed, 10 Feb 2010 15:42:38 +0100 Subject: [PATCH 1/2] fixup! Add count_children/1 to supervisor.erl to determine the number of children being managed without the memory impact of which_children/1 Use tuples, not lists, for collection of terms of a fixed size. --- lib/stdlib/src/supervisor.erl | 16 ++++++++-------- 1 files changed, 8 insertions(+), 8 deletions(-) diff --git a/lib/stdlib/src/supervisor.erl b/lib/stdlib/src/supervisor.erl index d5c95b4..342b71f 100644 --- a/lib/stdlib/src/supervisor.erl +++ b/lib/stdlib/src/supervisor.erl @@ -320,10 +320,10 @@ handle_call(count_children, _From, State) when ?is_simple(State) -> handle_call(count_children, _From, State) -> %% Specs and children are together on the children list... - [Specs, Active, Supers, Workers] = + {Specs, Active, Supers, Workers} = lists:foldl(fun(Child, Counts) -> count_child(Child, Counts) - end, [0,0,0,0], State#state.children), + end, {0,0,0,0}, State#state.children), %% Reformat counts to a property list. Reply = [{specs, Specs}, {active, Active}, @@ -332,16 +332,16 @@ handle_call(count_children, _From, State) -> count_child(#child{pid = Pid, child_type = worker}, - [Specs, Active, Supers, Workers]) -> + {Specs, Active, Supers, Workers}) -> case is_pid(Pid) andalso is_process_alive(Pid) of - true -> [Specs+1, Active+1, Supers, Workers+1]; - false -> [Specs+1, Active, Supers, Workers+1] + true -> {Specs+1, Active+1, Supers, Workers+1}; + false -> {Specs+1, Active, Supers, Workers+1} end; count_child(#child{pid = Pid, child_type = supervisor}, - [Specs, Active, Supers, Workers]) -> + {Specs, Active, Supers, Workers}) -> case is_pid(Pid) andalso is_process_alive(Pid) of - true -> [Specs+1, Active+1, Supers+1, Workers]; - false -> [Specs+1, Active, Supers+1, Workers] + true -> {Specs+1, Active+1, Supers+1, Workers}; + false -> {Specs+1, Active, Supers+1, Workers} end. Date: Wed, 10 Feb 2010 15:48:18 +0100 Subject: [PATCH 2/2] fixup! Add count_children/1 to supervisor.erl to determine the number of children being managed without the memory impact of which_children/1 Remove redundant parenthesis. --- lib/stdlib/test/supervisor_SUITE.erl | 6 +++--- 1 files changed, 3 insertions(+), 3 deletions(-) diff --git a/lib/stdlib/test/supervisor_SUITE.erl b/lib/stdlib/test/supervisor_SUITE.erl index aa4abdb..039ea29 100644 --- a/lib/stdlib/test/supervisor_SUITE.erl +++ b/lib/stdlib/test/supervisor_SUITE.erl @@ -1330,18 +1330,18 @@ count_children_memory(Config) when is_list(Config) -> false -> ?line test_server:fail({count_children, used_more_memory}) end, - case (Size7 =< Size6) of + case Size7 =< Size6 of true -> ok; false -> ?line test_server:fail({count_children, used_more_memory}) end, - case (Size4 > Size3) of + case Size4 > Size3 of true -> ok; false -> ?line test_server:fail({which_children, used_no_memory}) end, - case (Size6 > Size5) of + case Size6 > Size5 of true -> ok; false -> ?line test_server:fail({which_children, used_no_memory}) -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From bgustavsson@REDACTED Wed Feb 10 17:12:32 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Wed, 10 Feb 2010 17:12:32 +0100 Subject: [erlang-patches] Fix typos of lib/sasl/src/rb.erl and lib/snmp/src/misc/snmp_config.erl In-Reply-To: <20100210075832.GB69344@k2r.org> References: <20100210075832.GB69344@k2r.org> Message-ID: <6672d0161002100812k11981e8ek7d905b798f2d14b2@mail.gmail.com> On Wed, Feb 10, 2010 at 8:58 AM, Kenji Rikitake wrote: > git fetch git://github.com/jj1bdx/otp.git sasl_snmp_typofix > > This patch fixes typos of the following two files: > lib/sasl/src/rb.erl Will be included in pu > lib/snmp/src/misc/snmp_config.erl Micael Karlberg will handle this change. In general, unrelated changes to different applications are better put into separate commits. -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From bgustavsson@REDACTED Wed Feb 10 17:26:44 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Wed, 10 Feb 2010 17:26:44 +0100 Subject: [erlang-bugs] Fix beam/erl_bif_re.c building bug due to wrong version of PCRE is used In-Reply-To: <20100210134220.GA76956@k2r.org> References: <20100210134220.GA76956@k2r.org> Message-ID: <6672d0161002100826v20e97c5cne5bff086f531db9c@mail.gmail.com> On Wed, Feb 10, 2010 at 2:42 PM, Kenji Rikitake wrote: > git fetch git://github.com/jj1bdx/otp.git r13b03_compile_pcre Thanks! Will include in pu. -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From bgustavsson@REDACTED Wed Feb 10 17:29:13 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Wed, 10 Feb 2010 17:29:13 +0100 Subject: [erlang-bugs] Fix Makefile permission problem of crypto library on FreeBSD Erlang/OTP building In-Reply-To: <20100210085120.GG69344@k2r.org> References: <20100210085120.GG69344@k2r.org> Message-ID: <6672d0161002100829v9a732d7q4fa0feb2399f296@mail.gmail.com> On Wed, Feb 10, 2010 at 9:51 AM, Kenji Rikitake wrote: > git fetch git://github.com/jj1bdx/otp.git r13b03_crypto_makefile > Thanks! Will include in pu. -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From bgustavsson@REDACTED Wed Feb 10 18:03:38 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Wed, 10 Feb 2010 18:03:38 +0100 Subject: Thanks! and Import Message to all contributors Message-ID: <6672d0161002100903l46189c2cxee28d17331506294@mail.gmail.com> Thanks to everyone that has sent us patches. Submission of patches for erts/kernel/stdlib/compiler is now closed for this release (except for corrections of serious bugs). I now realize that it was a mistake to have the same dead-line for our internal feature freeze as for submission of patches. Because of that mistake, we *may* graduate some branches after the freeze, if we'll find them OTP-worthy. To make sure that I have not forgotten any of the patches, please check out http://erlang.github.com/otp/ to make sure that your topic branch is listed if I have said that we should include it there. If you have not received any answer at all to your email, please send a reminder. There is a new deadline for applications except erts/kernel/stdlib/compiler: Monday (Feb 15) 10.00 CET Emails with new patches received after that time will be put into a special folder and taken care of after the release, unless we happen to have a lot of free time on our hands. :) -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From michael.santos@REDACTED Wed Feb 10 23:34:16 2010 From: michael.santos@REDACTED (Michael Santos) Date: Wed, 10 Feb 2010 17:34:16 -0500 Subject: patch for CVE-2008-2371 Message-ID: <20100210223416.GA3056@ecn.lan> Running the following code will crash the emulator: re:compile(<<"(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]">>, [unicode]). Erlang uses PCRE 7.6. This issue was identified as CVE-2008-2371 and was fixed in PCRE 7.8. A patch can be found here: git fetch git://github.com/msantos/otp.git pcre-CVE-2008-2371 It is taken directly from: http://vcs.pcre.org/viewvc?revision=360&view=revision Does the Erlang/OTP team have a policy on security advisories, so users and package maintainers can evaluate their risk? From kenji.rikitake@REDACTED Thu Feb 11 00:20:20 2010 From: kenji.rikitake@REDACTED (Kenji Rikitake) Date: Thu, 11 Feb 2010 08:20:20 +0900 Subject: [erlang-patches] Fix calendar:local_time_to_universal_time_dst/2 bug of return value -1 from mktime In-Reply-To: <6672d0161002100626y36ab72f4oc451c9478865be75@mail.gmail.com> References: <20100210051304.GA48531@k2r.org> <6672d0161002092257s10b3dbd1u1fd225c93ef17048@mail.gmail.com> <6672d0161002100626y36ab72f4oc451c9478865be75@mail.gmail.com> Message-ID: <20100210232020.GA83470@k2r.org> Bjorn and all: I would like the mktime fix of erl_time_sup.c to be included ASAP. (check out git fetch git://github.com/jj1bdx/otp.git local_to_univ for the patch details) In the message <6672d0161002100626y36ab72f4oc451c9478865be75@REDACTED> dated Wed, Feb 10, 2010 at 03:26:07PM +0100, Bjorn Gustavsson writes: > We don't want the change in calendar.erl. > > The documentation says that local_time_to_universal_time_dst/1 > returns an empty list for a time that does not exist. We don't want > the same return value for a valid time that cannot be handled because > of an implementation detail. I will not argue against this decision for the time being. > We may consider using the patch for erl_time_sup.c, if you can > provide some information to show that it is really needed. Note > that the BIF does a range check of its arguments. Are there some > values of the arguments that slip through the range check and > cause mktime() to return -1? > > Since such a change is considered a bug fix, we can accept > it after the feature freeze today. Paul Guyot summarized the case in http://www.erlang.org/pipermail/erlang-bugs/2008-November/001077.html as: (quote) "I found a bug in Erlang/OTP R12B-4 that occurs on FreeBSD 7.0 with TZ set to UTC. On such systems, mktime(3) returns -1 when is_dst is set to true. According to SUSv3, the OS can return -1 when the time since Epoch cannot be represented, and therefore FreeBSD is perfectly allowed to do so. http://www.opengroup.org/onlinepubs/009695399/functions/mktime.html " (unquote) The POSIX doc for mktime in the above opengroup.org link shows the following definition for the return value: (quote) RETURN VALUE The mktime() function shall return the specified time since the Epoch encoded as a value of type time_t. If the time since the Epoch cannot be represented, the function shall return the value (time_t)-1 and may set errno to indicate the error. (unquote) And a source code of error checking example as: (quote) /* What day of the week is July 4, 2001? */ #include #include struct tm time_str; char daybuf[20]; int main(void) { time_str.tm_year = 2001 - 1900; time_str.tm_mon = 7 - 1; time_str.tm_mday = 4; time_str.tm_hour = 0; time_str.tm_min = 0; time_str.tm_sec = 1; time_str.tm_isdst = -1; if (mktime(&time_str) == -1) /* <-- HERE */ (void)puts("-unknown-"); else { (void)strftime(daybuf, sizeof(daybuf), "%A", &time_str); (void)puts(daybuf); } return 0; } (unquote) So I think the -1 return value of mktime should be properly handled. Regards, Kenji Rikitake > > -- > Bj?rn Gustavsson, Erlang/OTP, Ericsson AB > > ________________________________________________________________ > erlang-patches (at) erlang.org mailing list. > See http://www.erlang.org/faq.html > To unsubscribe; mailto:erlang-patches-unsubscribe@REDACTED From kenji.rikitake@REDACTED Thu Feb 11 00:46:53 2010 From: kenji.rikitake@REDACTED (Kenji Rikitake) Date: Thu, 11 Feb 2010 08:46:53 +0900 Subject: [erlang-bugs] Fix building error of SCTP in FreeBSD 7.1 and later In-Reply-To: <20100210153945.GA22452@erix.ericsson.se> References: <20100210135207.GB76956@k2r.org> <20100210153945.GA22452@erix.ericsson.se> Message-ID: <20100210234653.GA83952@k2r.org> Raimo: Thanks for comments. I will rework this patch after R13B04 comes out. (I have other higher-priority projects than this one) Kenji Rikitake In the message <20100210153945.GA22452@REDACTED> dated Wed, Feb 10, 2010 at 04:39:21PM +0100, Raimo Niskanen writes: > On Wed, Feb 10, 2010 at 10:52:07PM +0900, Kenji Rikitake wrote: > > git fetch git://github.com/jj1bdx/otp.git r13b03_sctp_freebsd > > > > Fix building error of SCTP in FreeBSD 7.1 and later > > > > This patch fixes building error of SCTP in FreeBSD 7.1 and later. > > > > See > > http://www.erlang.org/cgi-bin/ezmlm-cgi?2:mss:1257:200904:hppnfialjjdlgecdiehe > > for the details. > > It seems Giacomo Olgeni's patch that this patch depends on > is missing. erts/configure will not define > HAVE_STRUCT_SCTP_PADDRPARAMS_SPP_SACKDELAY so it will > be disabled on all platforms even the ones that have > that flag... And that would be bad. > > See also: > http://www.erlang.org/cgi-bin/ezmlm-cgi/2/1264 > sorry about me not managing to rework that issue, > other emergencies came in the way... > > > > > Patch summary: > > > > R13A patch for solving a compilation error when building > > erts/emulator/drivers/common/inet_drv.c > > by Kenji Rikitake 12-APR-2009 > > > > Symptom solved by this patch: > > > > When building R13A in FreeBSD 7.1-RELEASE, > > the compiler flag > > HAVE_STRUCT_SCTP_PADDRPARAMS_SPP_SACKDELAY > > is NOT enabled. > > Some code in > > erts/emulator/drivers/common/inet_drv.c > > incorrectly assumes > > HAVE_STRUCT_SCTP_PADDRPARAMS_SPP_SACKDELAY > > is always true when > > HAVE_STRUCT_SCTP_PADDRPARAMS_SPP_FLAGS > > is true in config.h. > > This assumption causes a compilation error. > > > > Regards, > > Kenji Rikitake From bgustavsson@REDACTED Thu Feb 11 07:38:19 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Thu, 11 Feb 2010 07:38:19 +0100 Subject: [erlang-patches] Fix calendar:local_time_to_universal_time_dst/2 bug of return value -1 from mktime In-Reply-To: <20100210232020.GA83470@k2r.org> References: <20100210051304.GA48531@k2r.org> <6672d0161002092257s10b3dbd1u1fd225c93ef17048@mail.gmail.com> <6672d0161002100626y36ab72f4oc451c9478865be75@mail.gmail.com> <20100210232020.GA83470@k2r.org> Message-ID: <6672d0161002102238w7a62fa55gbf1d62758674b371@mail.gmail.com> 2010/2/11 Kenji Rikitake : > I would like the mktime fix of erl_time_sup.c to be included ASAP. I will include a fix in a new branch in pu and unless it causes any problems in our daily builds (unlikely), it will be included in R13B04. See the end of the email for the fix and my description of it. >> The documentation says that local_time_to_universal_time_dst/1 >> returns an empty list for a time that does not exist. We don't want >> the same return value for a valid time that cannot be handled because >> of an implementation detail. > > I will not argue against this decision for the time being. We may change our minds if the patch is to be included in a major release (such as R14, where some incompatibilities are allowed) and the patch is ready in good time before the release so we don't have to rush the review. I agree that calendar will need some sort of update to properly handle badargs from local_time_to_universal_time(). Here is my suggested patch (which will soon be included in pu): Subject: [PATCH] erl_time_sup.c: test for error return from mktime() When converting from local time to universal time in erlang:localtime_to_universal_time/2, the return value from mktime() is not checked, so if it returns -1 (an error return), the BIF will return a time just before the start of the epoch (namely {{1969,12,31},{23,59,59}}). There is a range check in erlang:localtime_to_universal_time/2 that causes an badarg exception if the arguments are out their valid ranges (for example, year less than 1970 or month greater than 12). At least on FreeBSD systems, however, mktime() returns -1 when the TZ is set to UTC and is_dst is set to true. Therefore it is necessary to check the return value of mktime() and cause a badarg exception if it is -1. Reported by Kenji Rikitake and Paul Guyot. See: http://www.erlang.org/pipermail/erlang-bugs/2008-November/001077.html --- erts/emulator/beam/erl_time_sup.c | 3 +++ 1 files changed, 3 insertions(+), 0 deletions(-) diff --git a/erts/emulator/beam/erl_time_sup.c b/erts/emulator/beam/erl_time_sup.c index c15f85f..51202fb 100644 --- a/erts/emulator/beam/erl_time_sup.c +++ b/erts/emulator/beam/erl_time_sup.c @@ -650,6 +650,9 @@ local_to_univ(Sint *year, Sint *month, Sint *day, t.tm_sec = *second; t.tm_isdst = isdst; the_clock = mktime(&t); + if (the_clock == -1) { + return 0; + } #ifdef HAVE_GMTIME_R gmtime_r(&the_clock, (tm = &tmbuf)); #else -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From bgustavsson@REDACTED Thu Feb 11 08:14:48 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Thu, 11 Feb 2010 08:14:48 +0100 Subject: [erlang-patches] patch for CVE-2008-2371 In-Reply-To: <20100210223416.GA3056@ecn.lan> References: <20100210223416.GA3056@ecn.lan> Message-ID: <6672d0161002102314s64a5671foa0cbe384ed9ab0ad@mail.gmail.com> On Wed, Feb 10, 2010 at 11:34 PM, Michael Santos wrote: > > Running the following code will crash the emulator: > > ?re:compile(<<"(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]">>, [unicode]). > > Erlang uses PCRE 7.6. This issue was identified as CVE-2008-2371 and > was fixed in PCRE 7.8. > > A patch can be found here: > > ?git fetch git://github.com/msantos/otp.git pcre-CVE-2008-2371 > Thanks! Will include in pu. I amended your commit to include the following test case: diff --git a/lib/stdlib/test/re_SUITE.erl b/lib/stdlib/test/re_SUITE.erl index fa50ba3..02683f9 100644 --- a/lib/stdlib/test/re_SUITE.erl +++ b/lib/stdlib/test/re_SUITE.erl @@ -18,12 +18,12 @@ %% -module(re_SUITE). --export([all/1, pcre/1,compile_options/1,run_options/1,combined_options/1,replace_autogen/1,global_capture/1,replace_input_types +-export([all/1, pcre/1,compile_options/1,run_options/1,combined_options/1,replace_autogen/1,global_capture/1,replace_input_types -include("test_server.hrl"). -include_lib("kernel/include/file.hrl"). -all(suite) -> [pcre,compile_options,run_options,combined_options,replace_autogen,global_capture,replace_input_types,replace_retu +all(suite) -> [pcre,compile_options,run_options,combined_options,replace_autogen,global_capture,replace_input_types,replace_retu pcre(doc) -> ["Run all applicable tests from the PCRE testsuites."]; @@ -538,3 +538,9 @@ error_handling(Config) when is_list(Config) -> ?t:timetrap_cancel(Dog), ok. +pcre_cve_2008_2371(doc) -> + "Fix as in http://vcs.pcre.org/viewvc?revision=360&view=revision"; +pcre_cve_2008_2371(Config) when is_list(Config) -> + %% Make sure it doesn't crash the emulator. + re:compile(<<"(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]">>, [unicode]), + ok. > Does the Erlang/OTP team have a policy on security advisories, so users > and package maintainers can evaluate their risk? I don't really understand that question, so I assume that the answer is no. -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From kenji.rikitake@REDACTED Thu Feb 11 09:25:18 2010 From: kenji.rikitake@REDACTED (Kenji Rikitake) Date: Thu, 11 Feb 2010 17:25:18 +0900 Subject: [erlang-patches] Fix calendar:local_time_to_universal_time_dst/2 bug of return value -1 from mktime In-Reply-To: <6672d0161002102238w7a62fa55gbf1d62758674b371@mail.gmail.com> References: <20100210051304.GA48531@k2r.org> <6672d0161002092257s10b3dbd1u1fd225c93ef17048@mail.gmail.com> <6672d0161002100626y36ab72f4oc451c9478865be75@mail.gmail.com> <20100210232020.GA83470@k2r.org> <6672d0161002102238w7a62fa55gbf1d62758674b371@mail.gmail.com> Message-ID: <20100211082518.GA84396@k2r.org> I'll keep this terse: In the message <6672d0161002102238w7a62fa55gbf1d62758674b371@REDACTED> dated Thu, Feb 11, 2010 at 07:37:55AM +0100, Bj?rn Gustavsson writes: > 2010/2/11 Kenji Rikitake : > > I would like the mktime fix of erl_time_sup.c to be included ASAP. > > I will include a fix in a new branch in pu and unless it causes any > problems in our daily builds (unlikely), it will be included in R13B04. > See the end of the email for the fix and my description of it. The patch is confirmed and I have no objection. > >> The documentation says that local_time_to_universal_time_dst/1 > >> returns an empty list for a time that does not exist. We don't want > >> the same return value for a valid time that cannot be handled because > >> of an implementation detail. > > > > I will not argue against this decision for the time being. > > We may change our minds if the patch is to be included in > a major release (such as R14, where some incompatibilities > are allowed) and the patch is ready in good time before the > release so we don't have to rush the review. I agree that calendar > will need some sort of update to properly handle badargs from > local_time_to_universal_time(). I'd be glad if Erlang/OTP team will review the semantics of local_time_to_universal_time() again before R14. > Here is my suggested patch (which will soon be included in pu): > > Subject: [PATCH] erl_time_sup.c: test for error return from mktime() Patch confirmed. (rest deleted) Regards, Kenji Rikitake From bgustavsson@REDACTED Thu Feb 11 09:53:56 2010 From: bgustavsson@REDACTED (=?ISO-8859-1?Q?Bj=F6rn_Gustavsson?=) Date: Thu, 11 Feb 2010 09:53:56 +0100 Subject: What's cooking in erlang/otp (2010-02-11) Message-ID: <4B73C5A4.7080407@gmail.com> The dead-line for erts, kernel, stdlib, and compiler has passed. Any new patches for any of those applications will not be considered for inclusion in R13B04 (unless they are corrections for serious bugs). Because of the unexpected amount of patches turning up on the day of the feature freeze there is now a revised schedule: * February 15, 10.00 CET. Emails with patches received after that time will basically be ignored until the release is over, unless they are corrections for serious bugs. * February 17 is the feature freeze Erlang/OTP. * February 24 is the day of the release. ------------------------------------------------------------ [Graduated] * egil/egd_render (percept) (2010-02-04) 1 commit + Fix egd_render transparent to use float constants (e7842b4) * egil/lcnt (erts, tools) (2010-02-01) 6 commits + Add test suite for lcnt in tools (9a22cca) + Add lcnt:rt_opt/1 bindings to erts_debug (98b2678) + Add runtime option to enable/disable lcnt stats (628aa0e) + Add auto width on string output (cfff4e9) + Add lcnt documentation (eec56b2) + Add lock profiling tool (e3c82f0) * gc/hipe_darwin_amd64 (erts) (2010-01-12) 4 commits + Fix hipe memory allocation problems on darwin/amd64 (56c0170) + Porting x86 darwin fixes to amd64 darwin hipe asm/m4 code (d17dd27) + Automatically enable hipe for darwin/amd64 builds (022fa2d) + Allow configure to enable_hipe for darwin/amd64 builds (8421cad) * jn/supervisor_child_count_only (stdlib) (2010-01-25) 1 commit - Add count_children/1 to supervisor.erl to determine the number of children being managed without the memory impact of which_children/1 (1686757) * ks/compiler (compiler) (2010-02-05) 3 commits + compiler: keep line numbers for attributes (a612e99) + compiler Makefile: alphabetize module names (054b4fd) + compile.erl: eliminate compiler warning (4e34336) * ks/erl_lint (stdlib) (2010-02-09) 2 commits + erl_lint_SUITE: adjust failing test case (a037310) + Allow recursive types and check for undefined types (b07101c) * mp/hipe_unified_loader (kernel) (2010-02-09) 1 commit + hipe_unified_loader: only block SMP scheduling when necessary (4a158f3) * pan/enable_m32_build (erts, otp) (2009-12-08) 1 commit + Teach configure --enable-m32-build for forcing a 32-bit build (cad901a) * po/odbc-update-delete-no-rows (odbc) (2009-11-26) 1 commit + SQL_NO_DATA and SQLSTATE 00000 indicate success for updates/deletes that affect no rows (1a564f8) * rani/inet_gethostbyname_fixes (erts, kernel) (2010-02-04) 6 commits + inet_res_SUITE: testcase fixes for legacy DNS resolver (Solaris 8) (7dbd225) + inet_res: /etc/resolv.conf: use domain as default search list (8de5ac7) + inet: delayed/avoided read of /etc/resolv.conf and /etc/hosts (0377ecd) + inet_gethost_native: workaround for empty result hostname on MacOS X (a53dfdc) + inet_res_SUITE: testcase fix for empty domain name (502afaf) + inet:gethostbyname improved to parse IP strings and look up own hostname (cc90d8c) * sc/sctp-connect-nowait (kernel) (2009-12-21) 1 commit + Implement a non-blocking SCTP connect (827bb85) * ta/code_clash_ez (kernel) (2010-02-05) 3 commits + code_SUITE: add missing is_list(Config) guard (c390221) + Add tests for code:clash/0 (79194d5) + Fix handling of archive (.ez) files in code:clash/0 (49da83d) * ta/otp_build_sed_newline (otp) (2010-02-09) 1 commit . Make newline in RHS of lib/configure sed substition more portable (b5a5513) (Included directly in ccase/r13b04_dev.) * ws/cross-compile-configuration (otp) (2010-02-09) 1 commit + Add cross-compile configuration for Atmel AVR32-Linux (NGW100) (2274aa6) ------------------------------------------------------------ [New topics] * bd/doc-fixes (erts, mnesia, stdlib) (2010-02-10) 1 commit - Fix minor documentation errors (3cb8a87) * bg/safer-local_to_univ (erts) (2010-02-11) 1 commit - erl_time_sup.c: test for error return from mktime() (0f62f0b) * kl/extensibility-implied (asn1) (2010-02-05) 3 commits - Updated ticket reference (a1a09bc) - Add support for EXTENSIBILITY IMPLIED, SET/SEQ OF NamedType (debc35b) - Updated version number for the next version of ASN1 (48d0efe) * kr/crypto_makefile (crypto) (2010-02-10) 1 commit - Fix Makefile permission problem of crypto library on FreeBSD Erlang/OTP building (a7491f2) * kr/gs-browser (gs) (2010-02-10) 1 commit - Change lib/gs/src/tool_utils.erl to select multiple browsers other than Netscape (802d159) * kr/odbc-netinet (odbc) (2010-02-10) 1 commit - Fix ODBC compilation error on FreeBSD by adding netinet/in.h to #include macros (5c6be09) * kr/sasl-typo-fix (sasl) (2010-02-10) 1 commit - sasl: Fix typo (b798e9b) * ks/kernel-cleanups (kernel) (2010-02-09) 2 commits - Add types and specs + cleanup. (e86ced3) - Clean up as suggested by tidier. (b730728) * ks/stdlib-cleanups (stdlib) (2010-02-08) 1 commit - stdlib: clean up as suggested by tidier (48654b0) Reviewed and approved. Graduation pending. * ms/pcre-security (erts, stdlib) (2010-02-10) 1 commit - Fix CVE-2008-2371 (outer level option with alternatives caused crash). (bb6370a) * yh/packet_option_for_new_ssl_send (ssl) (2010-02-09) 2 commits - fixup! prepend packet size bytes in ssl:send() in new_ssl implementation (5fff0e3) - prepend packet size bytes in ssl:send() in new_ssl implementation (df6b01c) ------------------------------------------------------------ [Cooking] * at/odbc_osx_fix (odbc) (2009-11-30) 1 commit - Workaround broken ODBC implementation on OSX 10.5 (88c71c9) Being reviewed. * bg/erl_bif_types (hipe) (2010-02-03) 2 commits - Add types for erts_debug:dist_ext_to_term/2 (249f73a) - erl_bif_types: add 'http_bin' as a valid packet option (7c72a77) * bg/temp-hipe-fixes (compiler, hipe, kernel) (2010-01-31) 4 commits - Crudely fix return type for the lists:key{search,find,member}/3 (0e3ab3e) - seq_trace_SUITE: Don't native-compile (c9fc04b) - Disable native code if on_load is used (bd4fe28) - andor_SUITE: Don't native-compile (8aadc4c) This branch contains temporary workarounds to avoid failing test cases and is never intended to graduate. * db/tv_nthtail_fix (tv) (2009-12-01) 1 commit - Fix for tv which restarts while trying to open a table (abe0759) The author of this topic branch has found the real reason for the problem and has made some progress in resolving it. * dgud/emacs-mode (tools) (2010-01-26) 2 commits - Added the old style skeletons, and a variable to change. (2cbe853) - Move code skeletons to a separate file. Use the updated skeletons from the Erlware Mode. (b4fd161) * jn/sasl-format_report (sasl) (2010-01-27) 1 commit - Added sasl_report:format_report/3 to return a string rather than send an io (ee8f29f) * jp/dependencies_makefile (compiler, erts) (2010-01-12) 2 commits - squash! fix snprintf for Windows (bf2708d) - Add dependencies Makefile generation to erlc(1) and compile(3) (7cbc437) Approved in principle by the OTP Technical Board; now awaits a closer review of the implementation. Unfortunately, will not make it for R13B04. * ks/cleanups (asn1, debugger, docbuilder, gs, os_mon, parsetools, percept, runtime_tools, sasl, wx) (2010-02-07) 11 commits - percept: Modernize types and specs and clean up as suggested by tidier (10422fb) - debugger: Clean up as suggested by tidier (cf178ed) - docbuilder: Clean up as suggested by tidier (222ae3f) - parsetools: Modernize types and specs + cleanup by tidier (5659cd0) - runtime_tools: Modernize specs (774e7fd) - sasl: Eliminate tuple used as fun (f48ed8b) - sasl: Add missing modules to app file (45a39a2) - asn1: Clean up as suggested by tidier (405578b) - os_mon: Modernize types and specs (7b7f8b7) - wx: Clean up as suggested by tidier (9b79d9b) - gs: Clean up as suggested by tidier (afd194b) * ks/hipe-fixes (erts, hipe) (2010-02-05) 3 commits - Remove $Id$ (44e6b63) - Adds types and specs to files in erts/preloaded. (d0db7f4) - Fix a HiPE compiler bug evaluating an expression that throws system_limit. (9df43dc) * tj/asn1_macro_prefix (asn1) (2010-02-04) 1 commit - Add support for prefixing macro names generated by the compiler (e88fa38) * va/rb-improvements (sasl) (2010-02-06) 5 commits - uses module re for regular expressions (f15b05a) - New function to filter reports by date (58f4cd1) - New rb:filter/1 function to ease report filtering (55eccfa) - Add rb:re/1 to grep reports using the re module (202d6ac) - Fix minor typo in read_report/1 (7d645b7) From mikpe@REDACTED Thu Feb 11 21:18:53 2010 From: mikpe@REDACTED (Mikael Pettersson) Date: Thu, 11 Feb 2010 21:18:53 +0100 Subject: unbreak event tracer's app.src Message-ID: <19316.26157.33142.865819@pilspetsen.it.uu.se> lib/et/src/et.app.src lists a module (et_contents_viewer) for which there is no source module, so it doesn't get built or installed. This breaks tools like HiPE's big test suite that parse installed .app files and expect to find a .beam file for every listed module. Fixed by deleting the offending line. I also noticed a duplicated line for the et_viewer module, and deleted that too. --- otp-0211/lib/et/src/et.app.src.~1~ 2010-02-11 15:14:03.000000000 +0100 +++ otp-0211/lib/et/src/et.app.src 2010-02-11 20:43:33.000000000 +0100 @@ -24,12 +24,10 @@ [ et, et_collector, - et_contents_viewer, et_gs_contents_viewer, et_gs_viewer, et_selector, et_viewer, - et_viewer, et_wx_contents_viewer, et_wx_viewer ]}, From per@REDACTED Thu Feb 11 22:05:47 2010 From: per@REDACTED (Per Hedeland) Date: Thu, 11 Feb 2010 22:05:47 +0100 (CET) Subject: Bug with agents.conf for snmpm config in R13B03 Message-ID: <201002112105.o1BL5l66062698@pluto.hedeland.org> Hi, Unless I'm missing something, using agents.conf to configure agents for the SNMP manager in R13B03 produces agent registrations that are invalid at least for trap reception, since there is no 'reg_type' config item - the snmpm_config:get_agent_user_info(Addr, Port) call in snmpm_server:do_handle_snmp_trap/4 fails with {error,{reg_type_not_found,Target}}. The patch below (sorry, no git at this time) fixes the problem, but is not obviously correct - it "forces" "new style" snmpm_user callbacks. Then again, if the "old style" isn't documented anymore, and this has been broken for a while (since R13A/R13B?), maybe this is better than changing the file format to allow a choice of old/new (where at least "new" would presumably be different from the current, i.e. sort-of-non-backwards-compat at this point). About "missing something", it is possible to provide the 'reg_type' via the DefaultAgentConfig in users.conf - but I assume that this is not the intent, since I can't find 'reg_type' documented anywhere. (That solution is fine by me though, I just don't want to use it without "official blessing".) Thanks --Per Hedeland --- lib/snmp/src/manager/snmpm_config.erl.orig 2009-11-20 09:48:36.000000000 +0100 +++ lib/snmp/src/manager/snmpm_config.erl 2010-02-11 21:01:15.000000000 +0100 @@ -1618,7 +1618,8 @@ {ok, Addr} -> snmp_conf:check_integer(Port, {gt, 0}), Conf = - [{address, Addr}, + [{reg_type, target_name}, + {address, Addr}, {port, Port}, {community, Comm}, {engine_id, EngineId}, From bernie@REDACTED Fri Feb 12 01:38:29 2010 From: bernie@REDACTED (Bernard Duggan) Date: Fri, 12 Feb 2010 11:38:29 +1100 Subject: Add mnesia activity event subscription Message-ID: <4B74A305.3010200@m5net.com> Hi all, Following a discussion over on erlang-questions, I've implemented an additional mnesia subscription type called 'activity'. This feature allows a process that is watching changes through the mnesia:subscribe({table, _, _}) system to know when a particular activity has completed and therefore to group changes received through the subscription system into groups corresponding to a single transaction. While this grouping by ActivityID was previously possible, it was not possible to know when the /last/ change for a given transaction had been received and therefore be able to forward on a complete transaction change set to some other process. What I'm looking for here is feedback rather than immediate inclusion - This patch is currently not OTP-worthy, if only because it needs documentation - I didn't want to spend too much time on that aspect, though, if the judgement is going to be that this is a totally inappropriate feature or I've gone about it in entirely the wrong way :) On a related note, is there a test suite for mnesia? I couldn't find one that I might be able to add test cases to. git fetch git://github.com/bernardd/otp.git activity_subscribe Cheers, Bernard From bgustavsson@REDACTED Fri Feb 12 06:55:55 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Fri, 12 Feb 2010 06:55:55 +0100 Subject: [erlang-patches] unbreak event tracer's app.src In-Reply-To: <19316.26157.33142.865819@pilspetsen.it.uu.se> References: <19316.26157.33142.865819@pilspetsen.it.uu.se> Message-ID: <6672d0161002112155m36801579x1fb649b69ba4f94d@mail.gmail.com> On Thu, Feb 11, 2010 at 9:18 PM, Mikael Pettersson wrote: > lib/et/src/et.app.src lists a module (et_contents_viewer) for which > there is no source module, so it doesn't get built or installed. > This breaks tools like HiPE's big test suite that parse installed .app > files and expect to find a .beam file for every listed module. > > Fixed by deleting the offending line. I also noticed a duplicated > line for the et_viewer module, and deleted that too. > Thanks! Will include in pu. -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From bgustavsson@REDACTED Fri Feb 12 07:09:45 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Fri, 12 Feb 2010 07:09:45 +0100 Subject: [erlang-patches] Add mnesia activity event subscription In-Reply-To: <4B74A305.3010200@m5net.com> References: <4B74A305.3010200@m5net.com> Message-ID: <6672d0161002112209y5a5f400btefdb2fe84d5d157b@mail.gmail.com> On Fri, Feb 12, 2010 at 1:38 AM, Bernard Duggan wrote: > What I'm looking for here is feedback rather than immediate inclusion - > This patch is currently not OTP-worthy, if only because it needs > documentation - I didn't want to spend too much time on that aspect, > though, if the judgement is going to be that this is a totally > inappropriate feature or I've gone about it in entirely the wrong way :) That is OK. I will include it in pu to keep track of it. > On a related note, is there a test suite for mnesia? ?I couldn't find > one that I might be able to add test cases to. Yes, there is, but we haven't included it in the git repository yet. For each test suite, we'll need to make sure that it have copyright headers and that the contents is otherwise appropriate for public release. > git fetch git://github.com/bernardd/otp.git activity_subscribe Thanks! -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From kenneth.lundin@REDACTED Fri Feb 12 08:37:14 2010 From: kenneth.lundin@REDACTED (Kenneth Lundin) Date: Fri, 12 Feb 2010 08:37:14 +0100 Subject: [erlang-patches] patch for CVE-2008-2371 In-Reply-To: <20100210223416.GA3056@ecn.lan> References: <20100210223416.GA3056@ecn.lan> Message-ID: > > Does the Erlang/OTP team have a policy on security advisories, so users > and package maintainers can evaluate their risk? > Thanks for the patch. I don't really understand what you mean with "policy on security advisories". Please explain more I really want to learn if we have missed something important. This was a bug in the pcre package that we use for the re module in Erlang. As Erlang/OTP is a development system the users can build any type of system with it. And it is the users built system that might be exposed for security issues. We cannot make Erlang/OTP in a way that guarantee that the system build by the user with erlang7OTP does not have security issues or can crash on certain input etc. Of course it is a very bad idea to let the external user to a system freely create regular expressions which are used directly as parameters to the re:compile function. And that is true even if there is no bug in pcre. A complex regexp can for sure break the system anyway by consuming memory or by taking very long time to compile or use in searches. /Kenneth Erlang/OTP Ericsson From bgustavsson@REDACTED Fri Feb 12 16:06:55 2010 From: bgustavsson@REDACTED (=?ISO-8859-1?Q?Bj=F6rn_Gustavsson?=) Date: Fri, 12 Feb 2010 16:06:55 +0100 Subject: What's cooking in erlang/otp (2010-02-12) Message-ID: <4B756E8F.5020200@gmail.com> ------------------------------------------------------------ [Graduated] * bd/doc-fixes (erts, mnesia, stdlib) (2010-02-10) 1 commit + Fix minor documentation errors (3cb8a87) * kr/crypto_makefile (crypto) (2010-02-10) 1 commit + Fix Makefile permission problem of crypto library on FreeBSD Erlang/OTP building (a7491f2) * ks/erts (erts) (2010-02-05) 2 commits + hipe_bif0.c: Remove $Id$ (5cd9121) + erts/preloaded: Add types and specs (f28ac67) * ks/stdlib-cleanups (stdlib) (2010-02-08) 1 commit + stdlib: clean up as suggested by tidier (48654b0) * mp/et-app-file (et) (2010-02-11) 1 commit + unbreak event tracer's app.src (c3faaf4) * ms/pcre-security (erts, stdlib) (2010-02-10) 1 commit + Fix CVE-2008-2371 (outer level option with alternatives caused crash). (bb6370a) * yh/packet_option_for_new_ssl_send (ssl) (2010-02-11) 4 commits + Fixed ssl:setopts(Socket, binary) which was didn't work for 'new' ssl. (464f2ba) + Fixed bug file cache bug and improved the error messages. (a39cf4a) + Allow ssl:listen/2 to be called with option {ssl_imp, old}. (590a061) + prepend packet size bytes in ssl:send() in new_ssl implementation (f435df0) Dan Gudmundsson corrected error handling and added a test case to the original commit, and also added three independent bug fixes to the same branch (for convenience). ------------------------------------------------------------ [New topics] * bd/mnesia-activity-subscription (mnesia) (2010-02-12) 1 commit - Add mnesia activity subscription message (69816f8) * ks/cleanups-after-r13b04 (kernel, stdlib) (2010-02-10) 4 commits - stdlib: Use fun object instead of {M,F} tuple (9825e53) - ets: Cleanup as suggested by tidier (4d656fd) - kernel: Add types and specs + cleanup (f1c6d01) - kernel: Clean up as suggested by tidier (c695f3f) * ks/hipe (dialyzer, hipe, typer) (2010-02-10) 4 commits - dialyzer: New version for the R13B04 release (49d7f36) - hipe: Miscellaneous additions (6049203) - typer: New version for the R13B04 release (6d1f40f) - Fix a HiPE compiler bug evaluating an expression that throws system_limit (665bcd7) To be included in R13B04. ------------------------------------------------------------ [Stalled] * kr/gs-browser (gs) (2010-02-10) 1 commit - Change lib/gs/src/tool_utils.erl to select multiple browsers other than Netscape (802d159) This topic branch introduces a new call to the deprecated regexp module. ------------------------------------------------------------ [Cooking] * at/odbc_osx_fix (odbc) (2009-11-30) 1 commit - Workaround broken ODBC implementation on OSX 10.5 (88c71c9) Being reviewed. * bg/erl_bif_types (hipe) (2010-02-03) 2 commits - Add types for erts_debug:dist_ext_to_term/2 (249f73a) - erl_bif_types: add 'http_bin' as a valid packet option (7c72a77) * bg/safer-local_to_univ (erts) (2010-02-11) 1 commit - erl_time_sup.c: test for error return from mktime() (0f62f0b) * bg/temp-hipe-fixes (compiler, hipe, kernel) (2010-01-31) 4 commits - Crudely fix return type for the lists:key{search,find,member}/3 (0e3ab3e) - seq_trace_SUITE: Don't native-compile (c9fc04b) - Disable native code if on_load is used (bd4fe28) - andor_SUITE: Don't native-compile (8aadc4c) This branch contains temporary workarounds to avoid failing test cases and is never intended to graduate. * db/tv_nthtail_fix (tv) (2009-12-01) 1 commit - Fix for tv which restarts while trying to open a table (abe0759) The author of this topic branch has found the real reason for the problem and has made some progress in resolving it. * dgud/emacs-mode (tools) (2010-01-26) 2 commits - Added the old style skeletons, and a variable to change. (2cbe853) - Move code skeletons to a separate file. Use the updated skeletons from the Erlware Mode. (b4fd161) * jn/sasl-format_report (sasl) (2010-01-27) 1 commit - Added sasl_report:format_report/3 to return a string rather than send an io (ee8f29f) * jp/dependencies_makefile (compiler, erts) (2010-01-12) 2 commits - squash! fix snprintf for Windows (bf2708d) - Add dependencies Makefile generation to erlc(1) and compile(3) (7cbc437) Approved in principle by the OTP Technical Board; now awaits a closer review of the implementation. Unfortunately, will not make it for R13B04. * kl/extensibility-implied (asn1) (2010-02-05) 3 commits - Updated ticket reference (a1a09bc) - Add support for EXTENSIBILITY IMPLIED, SET/SEQ OF NamedType (debc35b) - Updated version number for the next version of ASN1 (48d0efe) * kr/odbc-netinet (odbc) (2010-02-10) 1 commit - Fix ODBC compilation error on FreeBSD by adding netinet/in.h to #include macros (5c6be09) * kr/sasl-typo-fix (sasl) (2010-02-10) 1 commit - sasl: Fix typo (b798e9b) * ks/cleanups (asn1, debugger, docbuilder, gs, os_mon, parsetools, percept, runtime_tools, sasl, wx) (2010-02-07) 11 commits - percept: Modernize types and specs and clean up as suggested by tidier (10422fb) - debugger: Clean up as suggested by tidier (cf178ed) - docbuilder: Clean up as suggested by tidier (222ae3f) - parsetools: Modernize types and specs + cleanup by tidier (5659cd0) - runtime_tools: Modernize specs (774e7fd) - sasl: Eliminate tuple used as fun (f48ed8b) - sasl: Add missing modules to app file (45a39a2) - asn1: Clean up as suggested by tidier (405578b) - os_mon: Modernize types and specs (7b7f8b7) - wx: Clean up as suggested by tidier (9b79d9b) - gs: Clean up as suggested by tidier (afd194b) * tj/asn1_macro_prefix (asn1) (2010-02-04) 1 commit - Add support for prefixing macro names generated by the compiler (e88fa38) * va/rb-improvements (sasl) (2010-02-06) 5 commits - uses module re for regular expressions (f15b05a) - New function to filter reports by date (58f4cd1) - New rb:filter/1 function to ease report filtering (55eccfa) - Add rb:re/1 to grep reports using the re module (202d6ac) - Fix minor typo in read_report/1 (7d645b7) From kenji.rikitake@REDACTED Fri Feb 12 16:22:31 2010 From: kenji.rikitake@REDACTED (Kenji Rikitake) Date: Sat, 13 Feb 2010 00:22:31 +0900 Subject: [erlang-patches] patch for CVE-2008-2371 In-Reply-To: References: <20100210223416.GA3056@ecn.lan> Message-ID: <20100212152231.GA9024@k2r.org> For "security advisories": I suggest the Erlang/OTP team to review the operating system advisories available on the net. It's not about guaranteeing something, but the developers of software generally wants an abridged or summarized statements for bugs like: The bug itself Symptoms Workarounds How it might affect to related software How it will be fixed (or not) FreeBSD Security Advisory might be a good starting point. http://www.freebsd.org/security/advisories.html (Not all people have enough time to read through the source code and experiment. Unfortunately I'm one of those lazy ones too.) Regards, Kenji Rikitake In the message dated Fri, Feb 12, 2010 at 08:36:50AM +0100, Kenneth Lundin writes: > I don't really understand what you mean with "policy on security > advisories". Please explain more > I really want to learn if we have missed something important. From michael.santos@REDACTED Mon Feb 15 16:01:39 2010 From: michael.santos@REDACTED (Michael Santos) Date: Mon, 15 Feb 2010 10:01:39 -0500 Subject: [erlang-patches] patch for CVE-2008-2371 In-Reply-To: References: <20100210223416.GA3056@ecn.lan> Message-ID: <20100215150139.GA7400@ecn.lan> On Fri, Feb 12, 2010 at 08:37:14AM +0100, Kenneth Lundin wrote: > > > > Does the Erlang/OTP team have a policy on security advisories, so users > > and package maintainers can evaluate their risk? > > > Thanks for the patch. > > I don't really understand what you mean with "policy on security > advisories". Please explain more > I really want to learn if we have missed something important. > > This was a bug in the pcre package that we use for the re module in Erlang. > > As Erlang/OTP is a development system the users can build any type of > system with it. > And it is the users built system that might be exposed for security > issues. We cannot make Erlang/OTP > in a way that guarantee that the system build by the user with > erlang7OTP does not have security issues > or can crash on certain input etc. Thanks for the feedback! Kenji Rikitake summarized the issue pretty well. It's mainly a convenience for users and projects making packages for end users. Seems to be a common practice in opensource languages: Java: http://blogs.sun.com/security/tags/java Ruby: http://www.ruby-lang.org/en/security/ Python: http://www.python.org/news/security/ Python's page is somewhat anemic and not up to date, so it doesn't look like this approach is working out for them. Maybe a note or a tag that could be grep'ed for in the ChangeLog would help. > Of course it is a very bad idea to let the external user to a system > freely create regular expressions > which are used directly as parameters to the re:compile function. And > that is true even if there is no bug > in pcre. A complex regexp can for sure break the system anyway by > consuming memory or by taking > very long time to compile or use in searches. I think people who need to take external input for creating regular expressions understand that these may result in excessive resource usage that could hang or crash the node. Erlang has excellent recovery support for that situation. I don't think people expect the node will crash, run shellcode and join a botnet :) From bernie@REDACTED Wed Feb 17 01:55:04 2010 From: bernie@REDACTED (Bernard Duggan) Date: Wed, 17 Feb 2010 11:55:04 +1100 Subject: [erlang-patches] Add mnesia activity event subscription In-Reply-To: <6672d0161002112209y5a5f400btefdb2fe84d5d157b@mail.gmail.com> References: <4B74A305.3010200@m5net.com> <6672d0161002112209y5a5f400btefdb2fe84d5d157b@mail.gmail.com> Message-ID: <4B7B3E68.4030708@m5net.com> Hi Bj?rn et al, I've modified the activity event message to be more consistent with the other mnesia event message types and also added documentation describing the event type to the mnesia user's guide. Changes are on the same branch: git fetch git://github.com/bernardd/otp.git activity_subscribe Cheers, Bernard From bgustavsson@REDACTED Thu Feb 18 12:24:47 2010 From: bgustavsson@REDACTED (=?ISO-8859-1?Q?Bj=F6rn_Gustavsson?=) Date: Thu, 18 Feb 2010 12:24:47 +0100 Subject: What's cooking in erlang/otp (2010-02-18) Message-ID: <4B7D237F.1020803@gmail.com> This is the final What's cooking before the R13B04 release. I have updated the Stalled section with some more branches, and added an "Action expected from" line to each, to make it clear who is responsible for making the next step to advance the topic. I will take this opportunity to thank everyone that have sent us patches. There are currently in the dev branch one or more commits from 32 persons from outside the Erlang/OTP team (a few more patches have been included in some way or another into the release without being visible in the history in the git repository). I have not counted, but I think we have never included so many pathes from open-source contributors in any previous release. ------------------------------------------------------------ [Graduated] * bg/erl_bif_types (hipe) (2010-02-13) 4 commits + Add types for erts_debug:lock_counters/1 (6d2d8b9) + Add types for os:timestamp/0 (3f2a3a5) + Add types for erts_debug:dist_ext_to_term/2 (11d4aab) + erl_bif_types: add 'http_bin' as a valid packet option (4df5137) * dgud/emacs-mode (tools) (2010-01-26) 2 commits + Added the old style skeletons, and a variable to change. (2cbe853) + Move code skeletons to a separate file. Use the updated skeletons from the Erlware Mode. (b4fd161) * jn/sasl-format_report (sasl) (2010-01-27) 1 commit + Added sasl_report:format_report/3 to return a string rather than send an io (109a257) * kr/odbc-netinet (odbc) (2010-02-10) 1 commit + Fix ODBC compilation error on FreeBSD by adding netinet/in.h to #include macros (5c6be09) * ks/cleanups (asn1, os_mon, parsetools, percept, runtime_tools, sasl, wx) (2010-02-13) 13 commits + percept: Clean up as suggested by tidier (a8bf1fe) + percept: Modernize types and specs (f2592e6) + parsetools: Don't use 'try...of' when 'try' will do (c12397f) + parsetools: Use %% for comments at the beginning of a line (152088b) + parsetools: Replace lists:keysearch/3 with lists:keyfind/3 (b573c21) + parsetools: Modernize types and specs (3117c77) + parsetools: Replace TABs with spaces (08e2d74) + runtime_tools: Modernize specs (2b451b0) + sasl: Eliminate tuple used as fun (c394654) + sasl: Add missing modules to app file (b8527e1) + asn1: Clean up as suggested by tidier (37b274d) + os_mon: Modernize types and specs (c40e8d0) . . . * ks/hipe (dialyzer, hipe, syntax_tools, typer) (2010-02-16) 8 commits + dialyzer: Fix system_limit exception in race analysis (abe48c2) + syntax_tools: Add types and specs for most exported functions (8b7dd06) + syntax_tools: Support the --enable-native-libs configure option (649d313) + syntax_tools: Remove $Id$ annotations (b366554) + dialyzer: New version for the R13B04 release (a9c8a24) + hipe: Miscellaneous additions (cdd3a82) + typer: New version for the R13B04 release (eab3576) + Fix a HiPE compiler bug evaluating an expression that throws system_limit (63c1378) * tj/asn1_macro_prefix (asn1) (2010-02-04) 1 commit + Add support for prefixing macro names generated by the compiler (9d247b1) * va/rb-improvements (sasl) (2009-12-25) 4 commits + New rb:filter/2 to filter reports by date (242c70e) + New rb:filter/1 function to ease report filtering (e99ec8e) + Modify rb:grep/1 to grep reports using the re module (501b923) + Fix minor typo in read_report/1 (2356f4e) ------------------------------------------------------------ [New topics] * bg/compiler-bugs (otp) (2010-02-16) 1 commit - bug: inlining a module causes an internal compiler error (f77e8aa) Reminder for a (low priority) bug to fix after the R13B04 release. * gl/jinterface-pom (jinterface, otp) (2010-02-15) 2 commits - Exclude jinterface build artifact from git (37e9289) - Add pom.xml for jinterface build (5405d48) * mh/ssh_channel-init-no-catch (ssh) (2010-01-25) 1 commit - Remove pointless catch from ssh_channel:init/1. (bb7eaa5) * ra/include-public_key-in-ssh (ssh) (2010-01-20) 1 commit - Added public_key among ssh applications. (3a0c2c5) ------------------------------------------------------------ [Stalled] * at/odbc_osx_fix (odbc) (2009-11-30) 1 commit - Workaround broken ODBC implementation on OSX 10.5 (88c71c9) The patch in itself is good. It solves the problem on Mac OS X in the only way that seems possible. Unfortunately, our test cases fail on Linux, and fixing it seems to require a backward-incompatible change to the API of the odbc module or ugly workarounds. The reviewer (Jakob Cederlund) will email the details and suggest and/or implement a solution. Action expected from: OTP/Jakob Cederlund * bg/safer-local_to_univ (erts) (2010-02-11) 1 commit - erl_time_sup.c: test for error return from mktime() (0f62f0b) We will not include this branch in R13B04 for two reasons: o We are not 100% sure that this change is harmless on all platforms in all timezones. o We have an idea how the underlying problem can be fixed in the calendar module (only) and I plan to try it out after the release. We will keep this branch in pu as a reminder. Action expected from: Bj??rn Gustavsson * db/tv_nthtail_fix (tv) (2009-12-01) 1 commit - Fix for tv which restarts while trying to open a table (abe0759) I should have put this into the stalled state earlier. Since our long-term goal is to migrate from gs towards wx, maintaining the old gs applications has a low priority for the Erlang/OTP team. Therefore, we will accept patches that seem sound, but we are unlikely to spend much of our own time fixing bugs. The author of this topic branch has found the real reason for the problem, but uncovered additional issues. Action expected from: The topic author - to decide whether to include the topic branch (which is a workaround, but does eliminate the crash) or to wait for a better fix from himself or someone else with an interest to advance the topic. * jp/dependencies_makefile (compiler, erts) (2010-02-14) 3 commits - squash! eliminate dialyzer warning (434e2ca) - squash! fix snprintf for Windows (bf2708d) - Add dependencies Makefile generation to erlc(1) and compile(3) (7cbc437) Approved in principle by the OTP Technical Board; now awaits a closer review of the implementation. Action expected from: Bj??rn Gustavsson (the reviewer) * kr/gs-browser (gs) (2010-02-10) 1 commit - Change lib/gs/src/tool_utils.erl to select multiple browsers other than Netscape (802d159) This topic branch introduces a new call to the deprecated regexp module. Action expected from: Topic author or anyone else with an interest to advance the topic ------------------------------------------------------------ [Cooking] * bd/mnesia-activity-subscription (mnesia) (2010-02-12) 1 commit - Add mnesia activity subscription message (a564dd3) * bg/temp-hipe-fixes (compiler, hipe, kernel) (2010-01-31) 4 commits - Crudely fix return type for the lists:key{search,find,member}/3 (0e3ab3e) - seq_trace_SUITE: Don't native-compile (c9fc04b) - Disable native code if on_load is used (bd4fe28) - andor_SUITE: Don't native-compile (8aadc4c) This branch contains temporary workarounds to avoid failing test cases and is never intended to graduate. * kl/extensibility-implied (asn1) (2010-02-05) 3 commits - Updated ticket reference (46d01f9) - Add support for EXTENSIBILITY IMPLIED, SET/SEQ OF NamedType (185187c) - Updated version number for the next version of ASN1 (6f855f3) * ks/cleanups-after-r13b04 (debugger, docbuilder, gs, kernel, stdlib) (2010-02-07) 8 commits - debugger: Clean up as suggested by tidier (3ee6718) - docbuilder: Clean up as suggested by tidier (e9ddf18) - gs: Clean up as suggested by tidier (b72b682) - Take out type declarations from file.hrl. (bbaee4c) - stdlib: Use fun object instead of {M,F} tuple (4e68a89) - ets: Cleanup as suggested by tidier (83b3cad) - kernel: Add types and specs + cleanup (649066e) - kernel: Clean up as suggested by tidier (70439b9) ------------------------------------------------------------ [Dropped] * kr/sasl-typo-fix (sasl) (2010-02-10) 1 commit . sasl: Fix typo (b798e9b) Duplicate of commit 2356f4e in va/rb-improvements. From andrew@REDACTED Thu Feb 18 15:54:50 2010 From: andrew@REDACTED (Andrew Thompson) Date: Thu, 18 Feb 2010 09:54:50 -0500 Subject: [erlang-patches] What's cooking in erlang/otp (2010-02-18) In-Reply-To: <4B7D237F.1020803@gmail.com> References: <4B7D237F.1020803@gmail.com> Message-ID: <20100218145450.GB8518@hijacked.us> The ODBC OSX fix doesn't affect 10.6 - which doesn't have the bug. I did test the fix on freebsd, windows and (I thought) linux. Either way, I don't really mind if you don't keep the patch - I don't do any development with macs and my mac using coworker updated to 10.6 (which broke ODBC in a whole other way - 32 bit vs 64 bit nonsense). Andrew From bgustavsson@REDACTED Sat Feb 20 11:39:27 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Sat, 20 Feb 2010 11:39:27 +0100 Subject: Patch approval statistics for R13B04 Message-ID: <6672d0161002200239n40266c5ayd3b942e7c747a5d@mail.gmail.com> Now that R13B04 is feature-frozen, I can present some statistics about the patches we have received. In the git repository, there are commits from 32 persons from outside the Erlang/OTP group in the ccase/r13b04_dev branch (i.e. to be included in R13B04). Manually counting through the "What's cooking" emails sent to the erlang-patches mailing list, I got the following statistics: Number of patches included in R13B04: 51 Total number of dropped patches: 7 Stalled patches: 5 Of the dropped patches, 4 pointed out bugs and misfeatures that we chose to address in a different way. One dropped patch updated a deprecated module. Two patches were dropped because we found serious issues and the authors didn't submit revised patches. Note that of the 51 approved patches, several went through one more revisions by their authors to make them OTP worthy before we approved them. Statistics can never predict the future, but it seems that patches of good quality (OTP worthy) will have a good chance to be included in future OTP releases. Note that we never approve or reject unseen patches. An idea that seems good when explained in an email could turn out to have issues when run through our test suites, and conversely, an idea that seemed bad maybe seemed bad only because we didn't fully understand it. So instead of writing an email beforehand explaining what your patch will do, I suggest that you put the effort into a good commit message that contains a good description of the problem and/or use case, and a rationale to justify the way you chose to solve it. That information will speed up the reviewing process, and more importantly, it will help anyone 5 years from now to understand why the **** the change was made. Thanks to everyone that have sent us patches so far. I am looking forward to receiving many more good patches. -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From bile@REDACTED Sun Feb 21 11:13:03 2010 From: bile@REDACTED (bile@REDACTED) Date: Sun, 21 Feb 2010 05:13:03 -0500 Subject: spawn to rpc, filter, etc. Message-ID: <20100221051303.66bb8b72@landofbile.com> git fetch git://github.com/bile/otp.git spawn_rpc git fetch git://github.com/bile/otp.git net_kernel_catchall git fetch git://github.com/bile/otp.git kernel_stops - spawn_rpc It moves net_kernel:spawn into the rpc module to normalize remote 'apply'ing somewhat. erlang:spawn/4 sends messages to rex. net_kernel's spawn now calls rex's allowing older code to call newer but not reverse. Additionally a function which can be used to filter incoming rpc/spawn calls. fun(M,F,A,From) -> bool(). Set by calling rpc:filter(fun()). On returning true it allows the call otherwise it error(undef)'s as would an actually undefined MFA. It'd be useful in partially sandboxing connected nodes. I'm thinking it'd be nice to offer a default filter which white listed all expected calls made by appmon and other such apps so nonstandard calls would fail but the system would function otherwise. - net_kernel_catchall Adds handle_call(_,_,State) to net_kernel so it doesn't exit when receiving a random gen_server message. Just doesn't reply like rex. - kernel_stops Since rex and other non-'minimal' apps are configured to be permanent under kernel_sup it causes it to exit given it's 0 restart attempts should rpc:stop() be called. Changed rex, global_group, global and inet_db to be transient and return the same value when Mod:stop()ed. From tuncer.ayaz@REDACTED Sun Feb 21 13:08:37 2010 From: tuncer.ayaz@REDACTED (Tuncer Ayaz) Date: Sun, 21 Feb 2010 13:08:37 +0100 Subject: Fix escript:foldl/3 for beam and archive files Message-ID: <4ac8254d1002210408i6ddfb3cdn1629d53b7e9c5172@mail.gmail.com> This adapts the other two case clauses in escript:foldl/3 to the changes introduced by Hakan Mattson for OTP-8434 Improved handling of typed records in escripts in 6af2ac91005276add18b1c9bbf4c8fe4f8c6e040. A partial fix has already been applied by Hakan in 4ec4a06edb5baaa0af2840943230c4a0be3a93cf. git fetch git://github.com/tuncer/otp.git escript_foldl From fritchie@REDACTED Mon Feb 22 22:52:39 2010 From: fritchie@REDACTED (Scott Lystig Fritchie) Date: Mon, 22 Feb 2010 15:52:39 -0600 Subject: Patch for configurable ERTS_DE_BUSY_LIMIT Message-ID: <27950.1266875559@snookles.snookles.com> Hi, all. I sent a longer post to the erlang-questions about this. In a busy system sending large messages in quick succession, I see 'busy_dist_port' messages coming to my monitoring proc, sometimes 5 or even 15 times per second for the same process. And then there are the times where multiple processes are blocked due all sending to procs on the same destination node. Don't get me wrong, this kind of scheduler-controlled congestion mechanism is a good thing. But the 128KB limit feels cramped. When using 4MB instead, in a system with modern 64-bit multi-core CPUs and gigabit Ethernet, the latency jitter appears to drop significantly for us. Patch relative to R13B03. -Scott --- snip --- snip --- snip --- snip --- snip --- snip --- diff -ur otp_src_R13B03.orig/erts/emulator/beam/dist.c otp_src_R13B03/erts/emulator/beam/dist.c --- otp_src_R13B03.orig/erts/emulator/beam/dist.c 2009-11-20 05:29:24.000000000 -0800 +++ otp_src_R13B03/erts/emulator/beam/dist.c 2010-02-14 15:06:32.000000000 -0800 @@ -97,6 +97,9 @@ #define PASS_THROUGH 'p' /* This code should go */ int erts_is_alive; /* System must be blocked on change */ +#define ERTS_DE_BUSY_LIMIT (128*1024) +int erts_de_busy_limit = ERTS_DE_BUSY_LIMIT; + /* distribution trap functions */ Export* dsend2_trap = NULL; @@ -1407,8 +1410,6 @@ return -1; } -#define ERTS_DE_BUSY_LIMIT (128*1024) - static int dsig_send(ErtsDSigData *dsdp, Eterm ctl, Eterm msg, int force_busy) { diff -ur otp_src_R13B03.orig/erts/emulator/beam/erl_init.c otp_src_R13B03/erts/emulator/beam/erl_init.c --- otp_src_R13B03.orig/erts/emulator/beam/erl_init.c 2009-09-18 07:07:46.000000000 -0700 +++ otp_src_R13B03/erts/emulator/beam/erl_init.c 2010-02-14 16:08:33.000000000 -0800 @@ -788,6 +788,7 @@ char envbuf[21]; /* enough for any 64-bit integer */ size_t envbufsz; int async_max_threads = erts_async_max_threads; + extern int erts_de_busy_limit; early_init(&argc, argv); @@ -807,6 +808,9 @@ if (erts_sys_getenv("ERL_THREAD_POOL_SIZE", envbuf, &envbufsz) == 0) { async_max_threads = atoi(envbuf); } + if (erts_sys_getenv("ERTS_DE_BUSY_LIMIT", envbuf, &envbufsz) == 0) { + erts_de_busy_limit = atoi(envbuf); + } #ifdef DEBUG From tuncer.ayaz@REDACTED Tue Feb 23 23:42:55 2010 From: tuncer.ayaz@REDACTED (Tuncer Ayaz) Date: Tue, 23 Feb 2010 23:42:55 +0100 Subject: Fix escript:foldl/3 for beam and archive files In-Reply-To: <4ac8254d1002210408i6ddfb3cdn1629d53b7e9c5172@mail.gmail.com> References: <4ac8254d1002210408i6ddfb3cdn1629d53b7e9c5172@mail.gmail.com> Message-ID: <4ac8254d1002231442l5a37126bn262daed9064d231f@mail.gmail.com> On Sun, Feb 21, 2010 at 1:08 PM, Tuncer Ayaz wrote: > This adapts the other two case clauses in escript:foldl/3 to > the changes introduced by Hakan Mattson for > OTP-8434 ?Improved handling of typed records in escripts > in 6af2ac91005276add18b1c9bbf4c8fe4f8c6e040. > A partial fix has already been applied by Hakan in > 4ec4a06edb5baaa0af2840943230c4a0be3a93cf. > > git fetch git://github.com/tuncer/otp.git escript_foldl Fix has been applied by Bjorn and Hakan to the R13B04 branch. Thanks! From kagato@REDACTED Wed Feb 24 11:50:22 2010 From: kagato@REDACTED (Jayson Vantuyl) Date: Wed, 24 Feb 2010 02:50:22 -0800 Subject: PATCH: fixed rpc handling of uncaught throws Message-ID: <854292B9-A613-4FF9-8035-0A6E81CF1365@souja.net> rpc:call/4 and friends would silently transform a thrown value into a return value. This could silently consume exceptions that would be errors in other cases, which was undesireable. In the case of exits and errors, these functions already synthesize an EXIT tuple which is quite similar to the messages received when trapping exits of other processes. To stay consistent with this behavior, rpc:call now generates a {nocatch,Value} error (with stacktrace, and wrapped in a badrpc tuple). As such, when you execute a function via rpc, you can now expect for it to either return the returned value, or generate an EXIT tuple that is identical to the one that would be generated by a trapped exit but with the From field removed. git fetch git://github.com/jvantuyl/otp.git rpccall_update_to_use_try -- Jayson Vantuyl kagato@REDACTED From bgustavsson@REDACTED Wed Feb 24 16:30:02 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Wed, 24 Feb 2010 16:30:02 +0100 Subject: [erlang-patches] PATCH: fixed rpc handling of uncaught throws In-Reply-To: <854292B9-A613-4FF9-8035-0A6E81CF1365@souja.net> References: <854292B9-A613-4FF9-8035-0A6E81CF1365@souja.net> Message-ID: <6672d0161002240730y6d2faaccnd806f70e6cd42d0f@mail.gmail.com> On Wed, Feb 24, 2010 at 11:50 AM, Jayson Vantuyl wrote: > git fetch git://github.com/jvantuyl/otp.git rpccall_update_to_use_try Thanks! Will include in pu. -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From Antonio.Musumeci@REDACTED Wed Feb 24 16:34:16 2010 From: Antonio.Musumeci@REDACTED (Musumeci, Antonio S) Date: Wed, 24 Feb 2010 10:34:16 -0500 Subject: [erlang-patches] PATCH: fixed rpc handling of uncaught throws In-Reply-To: <6672d0161002240730y6d2faaccnd806f70e6cd42d0f@mail.gmail.com> References: <854292B9-A613-4FF9-8035-0A6E81CF1365@souja.net> <6672d0161002240730y6d2faaccnd806f70e6cd42d0f@mail.gmail.com> Message-ID: <01489E237C2C8F45AF7898E135F48E8001FA72D9D1@NYWEXMBX2129.msad.ms.com> Speaking of rpc has anyone taken a look at the patches I submitted a couple days ago? http://www.erlang.org/cgi-bin/ezmlm-cgi?3:mss:826:201002:ediinmnoamiklchlkamk -----Original Message----- From: erlang-patches@REDACTED [mailto:erlang-patches@REDACTED] On Behalf Of Bj?rn Gustavsson Sent: Wednesday, February 24, 2010 10:30 AM To: Jayson Vantuyl Cc: erlang-patches@REDACTED Subject: Re: [erlang-patches] PATCH: fixed rpc handling of uncaught throws On Wed, Feb 24, 2010 at 11:50 AM, Jayson Vantuyl wrote: > git fetch git://github.com/jvantuyl/otp.git rpccall_update_to_use_try Thanks! Will include in pu. -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB ________________________________________________________________ erlang-patches (at) erlang.org mailing list. See http://www.erlang.org/faq.html To unsubscribe; mailto:erlang-patches-unsubscribe@REDACTED -------------------------------------------------------------------------- NOTICE: If received in error, please destroy, and notify sender. Sender does not intend to waive confidentiality or privilege. Use of this email is prohibited when received in error. We may monitor and store emails to the extent permitted by applicable law. From bgustavsson@REDACTED Wed Feb 24 17:30:30 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Wed, 24 Feb 2010 17:30:30 +0100 Subject: [erlang-patches] spawn to rpc, filter, etc. In-Reply-To: <20100221051303.66bb8b72@landofbile.com> References: <20100221051303.66bb8b72@landofbile.com> Message-ID: <6672d0161002240830r2412c8b7t45b1cbc463a98e08@mail.gmail.com> On Sun, Feb 21, 2010 at 11:13 AM, wrote: > git fetch git://github.com/bile/otp.git spawn_rpc > git fetch git://github.com/bile/otp.git net_kernel_catchall > git fetch git://github.com/bile/otp.git kernel_stops Please revise your commit messages to follow the guidelines in: http://wiki.github.com/erlang/otp/writing-good-commit-messages -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From andrew@REDACTED Wed Feb 24 17:53:58 2010 From: andrew@REDACTED (Andrew Thompson) Date: Wed, 24 Feb 2010 11:53:58 -0500 Subject: format_status analog for error_info Message-ID: <20100224165358.GF1751@hijacked.us> So, with the release of R13B4 I was excited about the new format_status callback for gen_server and gen_fsm. Unfortunately I guess I misunderstood the intent of the patch (and foolishly didn't look at the implementation) and so the functionality is not what I'd hoped. What I was hoping for was to be able to use that callback to also modify the State information fed into error_info so that the problem of a process with massive internal state wasting vast amounts of memory sending its internal state to the SASL error_logger could be avoided if you used that callback to trim out the large and useless data (I don't care to see the giant binary my code was working on 90% of the time). Any thoughts on the best way to do this. Right now I see 3 options: * Piggyback on the terminate() callback. If the return value is of a special form (maybe just {ok, State}) then don't ignore the result and use the new State as the argument passed to error_info() * Add another new callback called something like format_error that gets called (if exported) before error_info and can modify the state data. * Use format_status (maybe passing 'terminate' as the Opt, instead of * 'normal') and look for the [{data, [{"State", State}]}] part of the output (or something else?). I know that lots of people struggle with error_logger getting swamped by large crash messages when procesesses with large state die (facebook even put it in one of their slide decks), so I do intend to fix this - I'm just not sure of the better way (and the way that maximises the chances of a patch being accepted). Thanks and congratulations on the new release, Andrew From bgustavsson@REDACTED Wed Feb 24 18:42:03 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Wed, 24 Feb 2010 18:42:03 +0100 Subject: Branch names after the R13B04 release Message-ID: <6672d0161002240942r422b919bwdd145732b92f827f@mail.gmail.com> The ccase/r13b04_dev branch has been merged to the master branch. The new development branch to base further patches on is called "dev". The description of branches at: http://wiki.github.com/erlang/otp/branches has been updated. -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From Antonio.Musumeci@REDACTED Wed Feb 24 18:47:44 2010 From: Antonio.Musumeci@REDACTED (Musumeci, Antonio S) Date: Wed, 24 Feb 2010 12:47:44 -0500 Subject: [erlang-patches] spawn to rpc, filter, etc. In-Reply-To: <6672d0161002240830r2412c8b7t45b1cbc463a98e08@mail.gmail.com> References: <20100221051303.66bb8b72@landofbile.com> <6672d0161002240830r2412c8b7t45b1cbc463a98e08@mail.gmail.com> Message-ID: <01489E237C2C8F45AF7898E135F48E8001FA72DA4F@NYWEXMBX2129.msad.ms.com> Fixed. -----Original Message----- From: erlang-patches@REDACTED [mailto:erlang-patches@REDACTED] On Behalf Of Bj?rn Gustavsson Sent: Wednesday, February 24, 2010 11:30 AM To: bile@REDACTED Cc: erlang-patches@REDACTED Subject: Re: [erlang-patches] spawn to rpc, filter, etc. On Sun, Feb 21, 2010 at 11:13 AM, wrote: > git fetch git://github.com/bile/otp.git spawn_rpc git fetch > git://github.com/bile/otp.git net_kernel_catchall git fetch > git://github.com/bile/otp.git kernel_stops Please revise your commit messages to follow the guidelines in: http://wiki.github.com/erlang/otp/writing-good-commit-messages -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB ________________________________________________________________ erlang-patches (at) erlang.org mailing list. See http://www.erlang.org/faq.html To unsubscribe; mailto:erlang-patches-unsubscribe@REDACTED -------------------------------------------------------------------------- NOTICE: If received in error, please destroy, and notify sender. Sender does not intend to waive confidentiality or privilege. Use of this email is prohibited when received in error. We may monitor and store emails to the extent permitted by applicable law. From bgustavsson@REDACTED Wed Feb 24 18:56:26 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Wed, 24 Feb 2010 18:56:26 +0100 Subject: Writing good commit messages Message-ID: <6672d0161002240956x67b861bau451b649997e578cc@mail.gmail.com> There is a new wiki page about writing good commit messages: http://wiki.github.com/erlang/otp/writing-good-commit-messages -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From andrew@REDACTED Wed Feb 24 22:08:53 2010 From: andrew@REDACTED (Andrew Thompson) Date: Wed, 24 Feb 2010 16:08:53 -0500 Subject: [erlang-patches] format_status analog for error_info In-Reply-To: <20100224165358.GF1751@hijacked.us> References: <20100224165358.GF1751@hijacked.us> Message-ID: <20100224210853.GI1751@hijacked.us> Okay, I took a stab at implementing it: git fetch git://github.com/Vagabond/otp.git error_info_modify_state_data Attached is a very simple gen_server that illustrates the new behaviour. Comments welcome. Andrew -------------- next part -------------- % module to demonstrate proposed change to terminate() behaviour -module(crash_server). -behaviour(gen_server). %% gen_server callbacks -export([init/1, handle_call/3, handle_cast/2, handle_info/2, terminate/2, code_change/3]). -export([start/0]). start() -> gen_server:start(?MODULE, [], []). -record(state, { foo = "Foo", bar = "bar", baz = <<"baz">> }). init([]) -> erlang:send_after(500, self(), crash), {ok, #state{}}. handle_call(_Event, _From, State) -> {reply, ok, State}. handle_cast(_Event, State) -> {noreply, State}. handle_info(crash, State) -> {stop, crash, State}; handle_info(_Event, State) -> {noreply, State}. code_change(_OldVsn, State, _Extra) -> {ok, State}. terminate(Reason, State) -> {error_info, State#state{baz = redacted}}. From vinoski@REDACTED Wed Feb 24 23:22:48 2010 From: vinoski@REDACTED (Steve Vinoski) Date: Wed, 24 Feb 2010 17:22:48 -0500 Subject: [erlang-patches] format_status analog for error_info In-Reply-To: <20100224165358.GF1751@hijacked.us> References: <20100224165358.GF1751@hijacked.us> Message-ID: <65b2728e1002241422l258b5d21l992be2601d6553a3@mail.gmail.com> On Wed, Feb 24, 2010 at 11:53 AM, Andrew Thompson wrote: > So, with the release of R13B4 I was excited about the new format_status > callback for gen_server and gen_fsm. Unfortunately I guess I > misunderstood the intent of the patch (and foolishly didn't look at the > implementation) and so the functionality is not what I'd hoped. > That patch works solely with sys:get_status, and puts back together two things that at some point in the past were taken apart. I have no idea when the division between get_status and format_status took place or why, so in creating the patch I merely put them back together as best as I could. What I was hoping for was to be able to use that callback to also modify > the State information fed into error_info so that the problem of a > process with massive internal state wasting vast amounts of memory > sending its internal state to the SASL error_logger could be avoided if > you used that callback to trim out the large and useless data (I don't > care to see the giant binary my code was working on 90% of the time). > I don't believe format_status is called any other time except as a result of a sys:get_status call. Any thoughts on the best way to do this. Right now I see 3 options: > > * Piggyback on the terminate() callback. If the return value is of a > special form (maybe just {ok, State}) then don't ignore the result and > use the new State as the argument passed to error_info() > * Add another new callback called something like format_error that gets > called (if exported) before error_info and can modify the state data. > * Use format_status (maybe passing 'terminate' as the Opt, instead of > * 'normal') and look for the [{data, [{"State", State}]}] part > of the output (or something else?). > > I know that lots of people struggle with error_logger getting swamped by > large crash messages when procesesses with large state die (facebook > even put it in one of their slide decks), so I do intend to fix this - > I'm just not sure of the better way (and the way that maximises the > chances of a patch being accepted). I think since format_status is already an optional export, reusing it for the purposes you mention would make sense. Probably not much point in inventing another mechanism to do something similar, and overloading existing functions (like terminate) has a chance of breaking something. Since format_status is new (again! :-) ) the chance of breaking anything by reusing it for this is pretty much nil. --steve From andrew@REDACTED Wed Feb 24 23:59:59 2010 From: andrew@REDACTED (Andrew Thompson) Date: Wed, 24 Feb 2010 17:59:59 -0500 Subject: [erlang-patches] format_status analog for error_info In-Reply-To: <65b2728e1002241422l258b5d21l992be2601d6553a3@mail.gmail.com> References: <20100224165358.GF1751@hijacked.us> <65b2728e1002241422l258b5d21l992be2601d6553a3@mail.gmail.com> Message-ID: <20100224225959.GJ1751@hijacked.us> On Wed, Feb 24, 2010 at 05:22:48PM -0500, Steve Vinoski wrote: > > I think since format_status is already an optional export, reusing it for > the purposes you mention would make sense. Probably not much point in > inventing another mechanism to do something similar, and overloading > existing functions (like terminate) has a chance of breaking something. > Since format_status is new (again! :-) ) the chance of breaking anything by > reusing it for this is pretty much nil. > The reason I was a little leery of format_status was that its return value is a little odd to my eyes, is that format standardized at all or can it be more or less freeform (even missing the "State" key)? Using format_status for this has the happy side effect of ensuring that neither sys:get_status or an error log message will choke and die on trying to prettyprint a gigantic term. Maybe we could call format_status with a different first argument to indicate that the status is for an error, not for normal operation? Andrew From kenji.rikitake@REDACTED Thu Feb 25 00:54:04 2010 From: kenji.rikitake@REDACTED (Kenji Rikitake) Date: Thu, 25 Feb 2010 08:54:04 +0900 Subject: Erlang R13B04 *EXPERIMENTAL* FreeBSD port Message-ID: <20100224235404.GA74854@k2r.org> FreeBSD *EXPERIMENTAL* (unofficial) port for R13B04 available at http://www.ne.jp/asahi/bdx/info/software/erlang-r13b04,1-beta-20100225.tar.gz KNOWN issues: * NO JAVA test yet (WITHOUT_JAVA=yes set in the Makefile) * some patches are not applied (et and wx; both libs build OK) Giacomo: please check this one for building the official port. Regards, Kenji Rikitake From bgustavsson@REDACTED Thu Feb 25 07:49:28 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Thu, 25 Feb 2010 07:49:28 +0100 Subject: [erlang-patches] format_status analog for error_info In-Reply-To: <20100224210853.GI1751@hijacked.us> References: <20100224165358.GF1751@hijacked.us> <20100224210853.GI1751@hijacked.us> Message-ID: <6672d0161002242249g36bf72cemf8eb73c36b81c849@mail.gmail.com> On Wed, Feb 24, 2010 at 10:08 PM, Andrew Thompson wrote: > Okay, I took a stab at implementing it: > > git fetch git://github.com/Vagabond/otp.git error_info_modify_state_data Thanks! Will include in 'pu'. One question: Would the same kind of change also make sense for the gen_event module? -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From vinoski@REDACTED Thu Feb 25 08:25:17 2010 From: vinoski@REDACTED (Steve Vinoski) Date: Thu, 25 Feb 2010 02:25:17 -0500 Subject: [erlang-patches] format_status analog for error_info In-Reply-To: <20100224225959.GJ1751@hijacked.us> References: <20100224165358.GF1751@hijacked.us> <65b2728e1002241422l258b5d21l992be2601d6553a3@mail.gmail.com> <20100224225959.GJ1751@hijacked.us> Message-ID: <65b2728e1002242325j18704f6bs2030d498bdb8d949@mail.gmail.com> On Wed, Feb 24, 2010 at 5:59 PM, Andrew Thompson wrote: > On Wed, Feb 24, 2010 at 05:22:48PM -0500, Steve Vinoski wrote: > > > > I think since format_status is already an optional export, reusing it for > > the purposes you mention would make sense. Probably not much point in > > inventing another mechanism to do something similar, and overloading > > existing functions (like terminate) has a chance of breaking something. > > Since format_status is new (again! :-) ) the chance of breaking anything > by > > reusing it for this is pretty much nil. > > > > The reason I was a little leery of format_status was that its return > value is a little odd to my eyes, is that format standardized at all or > can it be more or less freeform (even missing the "State" key)? > The default for gen_server (for example) is [{data, [{"State", State}]}], but it really can be anything you want. The unit test, for example, just returns [format_status_called]. Note that this freeform nature fits really well with how error_info works too. > Using format_status for this has the happy side effect of ensuring that > neither sys:get_status or an error log message will choke and die on > trying to prettyprint a gigantic term. Maybe we could call format_status > with a different first argument to indicate that the status is for an > error, not for normal operation? > Sure, I think that would work. Only sys.erl currently initiates a call to format_status, so only the atom 'normal' is currently ever passed. A possible patch for gen_server.erl, for example, to reuse format_status for error printing might look like this (this is just an example, not intended an official patch submission): diff --git a/lib/stdlib/src/gen_server.erl b/lib/stdlib/src/gen_server.erl index f1a9a31..f4ab898 100644 --- a/lib/stdlib/src/gen_server.erl +++ b/lib/stdlib/src/gen_server.erl @@ -705,7 +705,18 @@ terminate(Reason, Name, Msg, Mod, State, Debug) -> {shutdown,_}=Shutdown -> exit(Shutdown); _ -> - error_info(Reason, Name, Msg, State, Debug), + NState = + case erlang:function_exported(Mod, format_status, 2) of + true -> + FmtArgs = [get(), State], + case catch Mod:format_status(error, FmtArgs) of + {'EXIT', _} -> State; + Else -> Else + end; + _ -> + State + end, + error_info(Reason, Name, Msg, NState, Debug), exit(Reason) end end. I think this could work pretty well. Shall I work up an official patch? --steve From bgustavsson@REDACTED Thu Feb 25 09:02:07 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Thu, 25 Feb 2010 09:02:07 +0100 Subject: [erlang-patches] spawn to rpc, filter, etc. In-Reply-To: <20100221051303.66bb8b72@landofbile.com> References: <20100221051303.66bb8b72@landofbile.com> Message-ID: <6672d0161002250002x292abe4doe202fba914f412ce@mail.gmail.com> Thinks! On Sun, Feb 21, 2010 at 11:13 AM, wrote: > git fetch git://github.com/bile/otp.git spawn_rpc This patch touches very sensitive code paths that should not be changed lightly. Therefore, we'll need very good commit messages and some clarifications before I will consider including it even in 'pu'. Have you run the kernel test suite? Your first commit message starts with: "Move spawn/4 to rpc, add filter" which is a clear sign that the commit should be split into two commits using "git add -p". The first commit should do the moving of the spawn/4 functions and the commit message should give a very good reason why that is useful and/or which problem it solves. Also, what will happen if an Erlang node with the new code communicates with an older Erlang node running R13B04? The next commit says: "Add forward compatibilities to net_kernel with regard to spawn/4" I don't understand what you mean with forward compatibility. What kind of change do you plan in the future? Or do you mean backward compatibility, and should this change be squeezed into the commit that moves the spawn/4 functions? Another minor issue: this commit introduce compilation warnings for unused variables. > git fetch git://github.com/bile/otp.git net_kernel_catchall > git fetch git://github.com/bile/otp.git kernel_stops Will include in 'pu'. -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From rickard@REDACTED Thu Feb 25 15:44:26 2010 From: rickard@REDACTED (Rickard Green) Date: Thu, 25 Feb 2010 15:44:26 +0100 Subject: Patch for configurable ERTS_DE_BUSY_LIMIT In-Reply-To: <27950.1266875559@snookles.snookles.com> References: <27950.1266875559@snookles.snookles.com> Message-ID: <4B868CCA.9010804@erlang.org> It seems reasonable to add a configuration parameter for this. However: We don't add new configuration parameters using environment variables. All new configuration parameters should be possible to pass as command line arguments. Documentation of the parameter and a test case are needed. You also need to submit the patch via git . Regards, Rickard Green, Erlang/OTP, Ericsson AB. Scott Lystig Fritchie wrote: > Hi, all. I sent a longer post to the erlang-questions about this. > > In a busy system sending large messages in quick succession, I see > 'busy_dist_port' messages coming to my monitoring proc, sometimes 5 or > even 15 times per second for the same process. And then there are the > times where multiple processes are blocked due all sending to procs on > the same destination node. > > Don't get me wrong, this kind of scheduler-controlled congestion > mechanism is a good thing. But the 128KB limit feels cramped. When > using 4MB instead, in a system with modern 64-bit multi-core CPUs and > gigabit Ethernet, the latency jitter appears to drop significantly for > us. > > Patch relative to R13B03. > > -Scott > > --- snip --- snip --- snip --- snip --- snip --- snip --- > > diff -ur otp_src_R13B03.orig/erts/emulator/beam/dist.c otp_src_R13B03/erts/emulator/beam/dist.c > --- otp_src_R13B03.orig/erts/emulator/beam/dist.c 2009-11-20 05:29:24.000000000 -0800 > +++ otp_src_R13B03/erts/emulator/beam/dist.c 2010-02-14 15:06:32.000000000 -0800 > @@ -97,6 +97,9 @@ > #define PASS_THROUGH 'p' /* This code should go */ > > int erts_is_alive; /* System must be blocked on change */ > +#define ERTS_DE_BUSY_LIMIT (128*1024) > +int erts_de_busy_limit =RTS_DE_BUSY_LIMIT; > + > > /* distribution trap functions */ > Export* dsend2_trap =ULL; > @@ -1407,8 +1410,6 @@ > return -1; > } > > -#define ERTS_DE_BUSY_LIMIT (128*1024) > - > static int > dsig_send(ErtsDSigData *dsdp, Eterm ctl, Eterm msg, int force_busy) > { > diff -ur otp_src_R13B03.orig/erts/emulator/beam/erl_init.c otp_src_R13B03/erts/emulator/beam/erl_init.c > --- otp_src_R13B03.orig/erts/emulator/beam/erl_init.c 2009-09-18 07:07:46.000000000 -0700 > +++ otp_src_R13B03/erts/emulator/beam/erl_init.c 2010-02-14 16:08:33.000000000 -0800 > @@ -788,6 +788,7 @@ > char envbuf[21]; /* enough for any 64-bit integer */ > size_t envbufsz; > int async_max_threads =rts_async_max_threads; > + extern int erts_de_busy_limit; > > early_init(&argc, argv); > > @@ -807,6 +808,9 @@ > if (erts_sys_getenv("ERL_THREAD_POOL_SIZE", envbuf, &envbufsz) =0) { > async_max_threads =toi(envbuf); > } > + if (erts_sys_getenv("ERTS_DE_BUSY_LIMIT", envbuf, &envbufsz) =0) { > + erts_de_busy_limit =toi(envbuf); > + } > > > #ifdef DEBUG > From andrew@REDACTED Thu Feb 25 15:53:30 2010 From: andrew@REDACTED (Andrew Thompson) Date: Thu, 25 Feb 2010 09:53:30 -0500 Subject: [erlang-patches] format_status analog for error_info In-Reply-To: <65b2728e1002242325j18704f6bs2030d498bdb8d949@mail.gmail.com> References: <20100224165358.GF1751@hijacked.us> <65b2728e1002241422l258b5d21l992be2601d6553a3@mail.gmail.com> <20100224225959.GJ1751@hijacked.us> <65b2728e1002242325j18704f6bs2030d498bdb8d949@mail.gmail.com> Message-ID: <20100225145330.GK1751@hijacked.us> On Thu, Feb 25, 2010 at 02:25:17AM -0500, Steve Vinoski wrote: > Sure, I think that would work. Only sys.erl currently initiates a call to > format_status, so only the atom 'normal' is currently ever passed. A > possible patch for gen_server.erl, for example, to reuse format_status for > error printing might look like this (this is just an example, not intended > an official patch submission): > > diff --git a/lib/stdlib/src/gen_server.erl b/lib/stdlib/src/gen_server.erl > index f1a9a31..f4ab898 100644 > --- a/lib/stdlib/src/gen_server.erl > +++ b/lib/stdlib/src/gen_server.erl > @@ -705,7 +705,18 @@ terminate(Reason, Name, Msg, Mod, State, Debug) -> > {shutdown,_}=Shutdown -> > exit(Shutdown); > _ -> > - error_info(Reason, Name, Msg, State, Debug), > + NState = > + case erlang:function_exported(Mod, format_status, > 2) of > + true -> > + FmtArgs = [get(), State], > + case catch Mod:format_status(error, > FmtArgs) of > + {'EXIT', _} -> State; > + Else -> Else > + end; > + _ -> > + State > + end, > + error_info(Reason, Name, Msg, NState, Debug), > exit(Reason) > end > end. > So in this case, format_status when the first argument is 'error' is expected to return something more like the passed in state data rather than the {data, [{"State", State}]} structure? (although I guess it doesn't really matter since its printed by ~p anyway). > I think this could work pretty well. Shall I work up an official patch? Yeah, go for it and maybe we can get some feedback on which way the OTP team prefers. Andrew From andrew@REDACTED Thu Feb 25 16:34:03 2010 From: andrew@REDACTED (Andrew Thompson) Date: Thu, 25 Feb 2010 10:34:03 -0500 Subject: [erlang-patches] format_status analog for error_info In-Reply-To: <6672d0161002242249g36bf72cemf8eb73c36b81c849@mail.gmail.com> References: <20100224165358.GF1751@hijacked.us> <20100224210853.GI1751@hijacked.us> <6672d0161002242249g36bf72cemf8eb73c36b81c849@mail.gmail.com> Message-ID: <20100225153403.GL1751@hijacked.us> On Thu, Feb 25, 2010 at 07:49:28AM +0100, Bj??rn Gustavsson wrote: > On Wed, Feb 24, 2010 at 10:08 PM, Andrew Thompson wrote: > > Okay, I took a stab at implementing it: > > > > git fetch git://github.com/Vagabond/otp.git error_info_modify_state_data > > Thanks! Will include in 'pu'. > > One question: Would the same kind of change also make sense > for the gen_event module? > Yeah, I forgot that gen_event had a terminate(). I can certainly add support for that too (it looks a tiny bit more complicated). However, would you prefer Steve's format_status alternative or is the OTP team fine with my approach? I don't really mind which way we do it, so long as it becomes possible. Thanks, Andrew From andrew@REDACTED Thu Feb 25 16:56:42 2010 From: andrew@REDACTED (Andrew Thompson) Date: Thu, 25 Feb 2010 10:56:42 -0500 Subject: [erlang-patches] format_status analog for error_info In-Reply-To: <20100225153403.GL1751@hijacked.us> References: <20100224165358.GF1751@hijacked.us> <20100224210853.GI1751@hijacked.us> <6672d0161002242249g36bf72cemf8eb73c36b81c849@mail.gmail.com> <20100225153403.GL1751@hijacked.us> Message-ID: <20100225155642.GM1751@hijacked.us> On Thu, Feb 25, 2010 at 10:34:03AM -0500, Andrew Thompson wrote: > On Thu, Feb 25, 2010 at 07:49:28AM +0100, Bj??rn Gustavsson wrote: > > On Wed, Feb 24, 2010 at 10:08 PM, Andrew Thompson wrote: > > > Okay, I took a stab at implementing it: > > > > > > git fetch git://github.com/Vagabond/otp.git error_info_modify_state_data > > > > Thanks! Will include in 'pu'. > > > > One question: Would the same kind of change also make sense > > for the gen_event module? > > > Yeah, I forgot that gen_event had a terminate(). I can certainly add > support for that too (it looks a tiny bit more complicated). > > However, would you prefer Steve's format_status alternative or is the > OTP team fine with my approach? I don't really mind which way we do it, > so long as it becomes possible. > Hah! gen_event actually uses the result of terminate() to start the new handler - I'd forgotten that. That makes things a lot more complicated. Andrew From bgustavsson@REDACTED Thu Feb 25 17:16:38 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Thu, 25 Feb 2010 17:16:38 +0100 Subject: [erlang-patches] format_status analog for error_info In-Reply-To: <20100225155642.GM1751@hijacked.us> References: <20100224165358.GF1751@hijacked.us> <20100224210853.GI1751@hijacked.us> <6672d0161002242249g36bf72cemf8eb73c36b81c849@mail.gmail.com> <20100225153403.GL1751@hijacked.us> <20100225155642.GM1751@hijacked.us> Message-ID: <6672d0161002250816m616b5761k3fa67759df07d95f@mail.gmail.com> On Thu, Feb 25, 2010 at 4:56 PM, Andrew Thompson wrote: > > Hah! gen_event actually uses the result of terminate() to start the new > handler - I'd forgotten that. That makes things a lot more complicated. So perhaps it would be better to use Steve's approach? Other than that, I personally don't have any opinion yet on which approach is the better one. -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From andrew@REDACTED Thu Feb 25 17:31:55 2010 From: andrew@REDACTED (Andrew Thompson) Date: Thu, 25 Feb 2010 11:31:55 -0500 Subject: [erlang-patches] format_status analog for error_info In-Reply-To: <6672d0161002250816m616b5761k3fa67759df07d95f@mail.gmail.com> References: <20100224165358.GF1751@hijacked.us> <20100224210853.GI1751@hijacked.us> <6672d0161002242249g36bf72cemf8eb73c36b81c849@mail.gmail.com> <20100225153403.GL1751@hijacked.us> <20100225155642.GM1751@hijacked.us> <6672d0161002250816m616b5761k3fa67759df07d95f@mail.gmail.com> Message-ID: <20100225163155.GN1751@hijacked.us> On Thu, Feb 25, 2010 at 05:16:38PM +0100, Bj??rn Gustavsson wrote: > On Thu, Feb 25, 2010 at 4:56 PM, Andrew Thompson wrote: > > > > Hah! gen_event actually uses the result of terminate() to start the new > > handler - I'd forgotten that. That makes things a lot more complicated. > > So perhaps it would be better to use Steve's approach? > > Other than that, I personally don't have any opinion yet on which > approach is the better one. > I'm inclined to agree - except that gen_event doesn't support the callback module having a format_status callback, at least not yet. In theory sys:get_status on a gen_event could call Mod:format_status for all its handlers with their respective states on their respective modules. Andrew From vinoski@REDACTED Fri Feb 26 08:30:07 2010 From: vinoski@REDACTED (Steve Vinoski) Date: Fri, 26 Feb 2010 02:30:07 -0500 Subject: [erlang-patches] format_status analog for error_info In-Reply-To: <20100225163155.GN1751@hijacked.us> References: <20100224165358.GF1751@hijacked.us> <20100224210853.GI1751@hijacked.us> <6672d0161002242249g36bf72cemf8eb73c36b81c849@mail.gmail.com> <20100225153403.GL1751@hijacked.us> <20100225155642.GM1751@hijacked.us> <6672d0161002250816m616b5761k3fa67759df07d95f@mail.gmail.com> <20100225163155.GN1751@hijacked.us> Message-ID: <65b2728e1002252330p552c28nb59ef09e8cf5abe5@mail.gmail.com> On Thu, Feb 25, 2010 at 11:31 AM, Andrew Thompson wrote: > On Thu, Feb 25, 2010 at 05:16:38PM +0100, Bj??rn Gustavsson wrote: > > On Thu, Feb 25, 2010 at 4:56 PM, Andrew Thompson > wrote: > > > > > > Hah! gen_event actually uses the result of terminate() to start the new > > > handler - I'd forgotten that. That makes things a lot more complicated. > > > > So perhaps it would be better to use Steve's approach? > > > > Other than that, I personally don't have any opinion yet on which > > approach is the better one. > > > > I'm inclined to agree - except that gen_event doesn't support the > callback module having a format_status callback, at least not yet. > > In theory sys:get_status on a gen_event could call Mod:format_status for > all its handlers with their respective states on their respective > modules. > Nice idea -- I've implemented this in the patch I'm working on, and it works just fine. There's a backwards compatibility issue here, though: any event handler that just happens to already export a format_status/2 fun for some other purpose will find that sys:get_status calls on its gen_event module will, with this change, result in unexpected calls to its format_status function. However, this is no different than my original format_status patch for gen_server and gen_fsm that's now part of R13B04 that started this thread, since any gen_server or gen_fsm callback module that similarly just happened to export a format_status/2 fun now finds with R13B04 that sys:get_status calls will invoke that function. I think this is OK, though, since the likelihood of this being pervasive in practice is almost certainly zero. --steve From martti.kuparinen@REDACTED Fri Feb 26 13:57:16 2010 From: martti.kuparinen@REDACTED (Martti Kuparinen) Date: Fri, 26 Feb 2010 14:57:16 +0200 Subject: Two printf format fixes Message-ID: <4B87C52C.6030105@iki.fi> Hi, The attached patch fixes these two warnings in R13B04 on my 64-bit Ubuntu 9.10 with GCC 4.4.1 esock_utils.c:38: warning: format ?%d? expects type ?int?, but argument 3 has type ?size_t? esock_utils.c:50: warning: format ?%d? expects type ?int?, but argument 3 has type ?size_t? http://en.wikipedia.org/wiki/Printf#printf_format_placeholders Martti -------------- next part -------------- A non-text attachment was scrubbed... Name: esock_utils.c.diff Type: text/x-patch Size: 625 bytes Desc: not available URL: From tuncer.ayaz@REDACTED Fri Feb 26 14:31:54 2010 From: tuncer.ayaz@REDACTED (Tuncer Ayaz) Date: Fri, 26 Feb 2010 14:31:54 +0100 Subject: [erlang-patches] Two printf format fixes In-Reply-To: <4B87C52C.6030105@iki.fi> References: <4B87C52C.6030105@iki.fi> Message-ID: <4ac8254d1002260531v21aeab56ocf562dbb5414b928@mail.gmail.com> On Fri, Feb 26, 2010 at 1:57 PM, Martti Kuparinen wrote: > Hi, > > The attached patch fixes these two warnings in R13B04 on my 64-bit > Ubuntu 9.10 with GCC 4.4.1 > > esock_utils.c:38: warning: format ?%d? expects type ?int?, but > argument 3 has type ?size_t? > esock_utils.c:50: warning: format ?%d? expects type ?int?, but > argument 3 has type ?size_t? > > > http://en.wikipedia.org/wiki/Printf#printf_format_placeholders Hi Martti, please have a look at the following patch submission guides: http://wiki.github.com/erlang/otp/submitting-patches http://wiki.github.com/erlang/otp/writing-good-commit-messages It would be great if you can resubmit your patch either as an inline patch via 'git send-email' or as a topic branch on a public git repo. That makes life a lot easier for the OTP team. Thanks. From jan@REDACTED Fri Feb 26 19:54:35 2010 From: jan@REDACTED (Jan Lehnardt) Date: Fri, 26 Feb 2010 10:54:35 -0800 Subject: Enable FILE_SHARE_DELETE flag on Windows Message-ID: <120BD7A9-6A5C-4E1B-A3D0-9007B2A2D56B@apache.org> Dear OTP Team, the CouchDB team would like to propose the following patch to the windows file driver: http://github.com/janl/otp/commit/01e91f903650454fecebf4ec8c8b5ee1f5b35253 It adds an additional flag to the CreateFile() syscall to allow multiple descriptors to the same file to all request a deletion or rename of a file. For more details, see "dwShareMode" on http://msdn.microsoft.com/en-us/library/aa363858(VS.85).aspx Cheers Jan -- From vinoski@REDACTED Sun Feb 28 07:13:37 2010 From: vinoski@REDACTED (Steve Vinoski) Date: Sun, 28 Feb 2010 01:13:37 -0500 Subject: [erlang-patches] format_status analog for error_info In-Reply-To: <65b2728e1002252330p552c28nb59ef09e8cf5abe5@mail.gmail.com> References: <20100224165358.GF1751@hijacked.us> <20100224210853.GI1751@hijacked.us> <6672d0161002242249g36bf72cemf8eb73c36b81c849@mail.gmail.com> <20100225153403.GL1751@hijacked.us> <20100225155642.GM1751@hijacked.us> <6672d0161002250816m616b5761k3fa67759df07d95f@mail.gmail.com> <20100225163155.GN1751@hijacked.us> <65b2728e1002252330p552c28nb59ef09e8cf5abe5@mail.gmail.com> Message-ID: <65b2728e1002272213hf4e6072n7672710f2b0aae29@mail.gmail.com> On Fri, Feb 26, 2010 at 2:30 AM, Steve Vinoski wrote: > > > On Thu, Feb 25, 2010 at 11:31 AM, Andrew Thompson wrote: > >> On Thu, Feb 25, 2010 at 05:16:38PM +0100, Bj??rn Gustavsson wrote: >> > On Thu, Feb 25, 2010 at 4:56 PM, Andrew Thompson >> wrote: >> > > >> > > Hah! gen_event actually uses the result of terminate() to start the >> new >> > > handler - I'd forgotten that. That makes things a lot more >> complicated. >> > >> > So perhaps it would be better to use Steve's approach? >> > >> > Other than that, I personally don't have any opinion yet on which >> > approach is the better one. >> > >> >> I'm inclined to agree - except that gen_event doesn't support the >> callback module having a format_status callback, at least not yet. >> >> In theory sys:get_status on a gen_event could call Mod:format_status for >> all its handlers with their respective states on their respective >> modules. >> > > Nice idea -- I've implemented this in the patch I'm working on, and it > works just fine. > This patch is ready: git fetch git://github.com/vinoski/otp.git format_status_error_info Since gen_event did not previously have format_status callback support, I made these changes in two commits: the first to add termination support to existing format_status support in gen_fsm and gen_server along with doc and unit tests, and the second commit to add the format_status callback support to gen_event along with related doc and unit tests. --steve From bile@REDACTED Sun Feb 28 07:16:26 2010 From: bile@REDACTED (bile@REDACTED) Date: Sun, 28 Feb 2010 01:16:26 -0500 Subject: [erlang-patches] spawn to rpc, filter, etc. In-Reply-To: <6672d0161002250002x292abe4doe202fba914f412ce@mail.gmail.com> References: <20100221051303.66bb8b72@landofbile.com> <6672d0161002250002x292abe4doe202fba914f412ce@mail.gmail.com> Message-ID: <20100228011626.347d0a85@otis> On Thu, 25 Feb 2010 09:02:07 +0100 Bj?rn Gustavsson wrote: > Thinks! > > On Sun, Feb 21, 2010 at 11:13 AM, wrote: > > git fetch git://github.com/bile/otp.git spawn_rpc > > This patch touches very sensitive code paths that should > not be changed lightly. Therefore, we'll need very good > commit messages and some clarifications before I will > consider including it even in 'pu'. > > Have you run the kernel test suite? Yes. Everything passes as expected... particularly bif_SUITE. I've reworked the patch and hopefully clarified the reasons for the changes. The idea is to provide basic filtering / sandboxing abilities to remote spawning and rpc. The merging of net_kernel:spawn with rpc was to centralize both remote 'apply'ing logic as well as the filtering. Due to the change older versions will be able to communicate with newer ones but the other way as spawn requests would be sent to rex instead of net_kernel. I wasn't sure if it'd appropriate to have the filters effect local calls given you ultimately have full control over the local node so I've left them unfiltered at this time. -Antonio From bgustavsson@REDACTED Sun Feb 28 15:27:03 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Sun, 28 Feb 2010 15:27:03 +0100 Subject: [erlang-patches] Enable FILE_SHARE_DELETE flag on Windows In-Reply-To: <120BD7A9-6A5C-4E1B-A3D0-9007B2A2D56B@apache.org> References: <120BD7A9-6A5C-4E1B-A3D0-9007B2A2D56B@apache.org> Message-ID: <6672d0161002280627k46010181la036e686028d8df0@mail.gmail.com> On Fri, Feb 26, 2010 at 7:54 PM, Jan Lehnardt wrote: > Dear OTP Team, > > the CouchDB team would like to propose the following patch to the > windows file driver: > > ?http://github.com/janl/otp/commit/01e91f903650454fecebf4ec8c8b5ee1f5b35253 > Thanks! I have combined and slightly shortened the first two lines in the commit messages (to make it easier to see what the commit is about when browsing in tools such as gitk). If you plan to submit more patches, please have a look at: http://wiki.github.com/erlang/otp/writing-good-commit-messages How close will semantics be to Unix? For instance, if open a file for reading, and another Windows process deletes the file, will I be able to continue reading the file? -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From bgustavsson@REDACTED Sun Feb 28 15:28:32 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Sun, 28 Feb 2010 15:28:32 +0100 Subject: [erlang-patches] format_status analog for error_info In-Reply-To: <65b2728e1002272213hf4e6072n7672710f2b0aae29@mail.gmail.com> References: <20100224165358.GF1751@hijacked.us> <20100224210853.GI1751@hijacked.us> <6672d0161002242249g36bf72cemf8eb73c36b81c849@mail.gmail.com> <20100225153403.GL1751@hijacked.us> <20100225155642.GM1751@hijacked.us> <6672d0161002250816m616b5761k3fa67759df07d95f@mail.gmail.com> <20100225163155.GN1751@hijacked.us> <65b2728e1002252330p552c28nb59ef09e8cf5abe5@mail.gmail.com> <65b2728e1002272213hf4e6072n7672710f2b0aae29@mail.gmail.com> Message-ID: <6672d0161002280628u14fa786cwa818ad7df1db02b7@mail.gmail.com> On Sun, Feb 28, 2010 at 7:13 AM, Steve Vinoski wrote: > This patch is ready: > > git fetch git://github.com/vinoski/otp.git format_status_error_info > Thanks! Included in 'pu'. -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From bgustavsson@REDACTED Sun Feb 28 15:44:12 2010 From: bgustavsson@REDACTED (=?UTF-8?Q?Bj=C3=B6rn_Gustavsson?=) Date: Sun, 28 Feb 2010 15:44:12 +0100 Subject: [erlang-patches] spawn to rpc, filter, etc. In-Reply-To: <20100228011626.347d0a85@otis> References: <20100221051303.66bb8b72@landofbile.com> <6672d0161002250002x292abe4doe202fba914f412ce@mail.gmail.com> <20100228011626.347d0a85@otis> Message-ID: <6672d0161002280644m19387427ud50db6820fc036f1@mail.gmail.com> On Sun, Feb 28, 2010 at 7:16 AM, wrote: > I've reworked the patch and hopefully clarified the reasons for the > changes. The idea is to provide basic filtering / sandboxing abilities > to remote spawning and rpc. The merging of net_kernel:spawn with rpc > was to centralize both remote 'apply'ing logic as well as the > filtering. Due to the change older versions will be able to communicate > with newer ones but the other way as spawn requests would be sent to > rex instead of net_kernel. Thanks! Now it's much clearer. I have included it in 'pu' for now so that it will be easier to review, but I think that it is unlikely that it will be approved in its current form (because spawning from a new node to an old node no longer works). -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From tuncer.ayaz@REDACTED Sun Feb 28 18:07:23 2010 From: tuncer.ayaz@REDACTED (Tuncer Ayaz) Date: Sun, 28 Feb 2010 18:07:23 +0100 Subject: [PATCH] Fix typo in efficiency guide section 2.7 Message-ID: <1267376843-368-1-git-send-email-tuncer.ayaz@gmail.com> Noticed and corrected typo in efficiency guide section 2.7. Signed-off-by: Tuncer Ayaz --- system/doc/efficiency_guide/myths.xml | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/system/doc/efficiency_guide/myths.xml b/system/doc/efficiency_guide/myths.xml index 76a7236..6383657 100644 --- a/system/doc/efficiency_guide/myths.xml +++ b/system/doc/efficiency_guide/myths.xml @@ -169,7 +169,7 @@ vanilla_reverse([], Acc) -> In Erlang, you'll have to think a little more about how the strings are used and choose an appropriate representation and use the re instead of the obsolete - regexp module if you are going to use regualr expressions.

+ regexp module if you are going to use regular expressions.

-- 1.7.0 From bile@REDACTED Sun Feb 28 19:35:47 2010 From: bile@REDACTED (bile@REDACTED) Date: Sun, 28 Feb 2010 13:35:47 -0500 Subject: [erlang-patches] spawn to rpc, filter, etc. In-Reply-To: <6672d0161002280644m19387427ud50db6820fc036f1@mail.gmail.com> References: <20100221051303.66bb8b72@landofbile.com> <6672d0161002250002x292abe4doe202fba914f412ce@mail.gmail.com> <20100228011626.347d0a85@otis> <6672d0161002280644m19387427ud50db6820fc036f1@mail.gmail.com> Message-ID: <20100228133547.1d2e68e4@otis> > Thanks! Now it's much clearer. > > I have included it in 'pu' for now so that it will be easier to > review, but I think that it is unlikely that it will be approved in > its current form (because spawning from a new node to an old node no > longer works). > (Apologies if this email has made it multiple times to the list. I seem to be having some delivery issues.) I'm open to suggestions. It could be changed to not send messages to rex but net_kernel like originally, net_kernel could forward them to rex like now for compatibility but instead change the From field to the originator and allow rex to reply back. I didn't do that originally because 1) I didn't want the extra hop and 2) proxying through net_kernel allows the filter to know it's from an older node. Or revert the merging of spawn with rpc and duplicate the filter logic. I'm not really a fan of either but would prefer the former if the new -> old node spawning won't be acceptable. From bile@REDACTED Sun Feb 28 05:47:52 2010 From: bile@REDACTED (bile@REDACTED) Date: Sat, 27 Feb 2010 23:47:52 -0500 Subject: [erlang-patches] spawn to rpc, filter, etc. In-Reply-To: <6672d0161002250002x292abe4doe202fba914f412ce@mail.gmail.com> References: <20100221051303.66bb8b72@landofbile.com> <6672d0161002250002x292abe4doe202fba914f412ce@mail.gmail.com> Message-ID: <20100227234752.688406f4@landofbile.com> On Thu, 25 Feb 2010 09:02:07 +0100 Bj?rn Gustavsson wrote: > Thinks! > > On Sun, Feb 21, 2010 at 11:13 AM, wrote: > > git fetch git://github.com/bile/otp.git spawn_rpc > > This patch touches very sensitive code paths that should > not be changed lightly. Therefore, we'll need very good > commit messages and some clarifications before I will > consider including it even in 'pu'. > > Have you run the kernel test suite? Yes. Everything passes as expected... particularly bif_SUITE. I've reworked the patch and hopefully clarified the reasons for the changes. The idea is to provide basic filtering / sandboxing abilities to remote spawning and rpc. The merging of net_kernel:spawn with rpc was to centralize both remote 'apply'ing logic as well as the filtering. Due to the change older versions will be able to communicate with newer ones but the other way as spawn requests would be sent to rex instead of net_kernel. I wasn't sure if it'd appropriate to have the filters effect local calls given you ultimately have full control over the local node so I've left them unfiltered at this time. -Antonio From bile@REDACTED Sun Feb 28 16:22:47 2010 From: bile@REDACTED (bile@REDACTED) Date: Sun, 28 Feb 2010 10:22:47 -0500 Subject: [erlang-patches] spawn to rpc, filter, etc. In-Reply-To: <6672d0161002280644m19387427ud50db6820fc036f1@mail.gmail.com> References: <20100221051303.66bb8b72@landofbile.com> <6672d0161002250002x292abe4doe202fba914f412ce@mail.gmail.com> <20100228011626.347d0a85@otis> <6672d0161002280644m19387427ud50db6820fc036f1@mail.gmail.com> Message-ID: <20100228102247.06741060@otis> > Thanks! Now it's much clearer. > > I have included it in 'pu' for now so that it will be easier to > review, but I think that it is unlikely that it will be approved in > its current form (because spawning from a new node to an old node no > longer works). > I'm open to suggestions. It could be changed to not send messages to rex but net_kernel like originally, net_kernel could forward them to rex like now for compatibility but instead change the From field to the originator and allow rex to reply back. I didn't do that originally because 1) I didn't want the extra hop and 2) proxying through net_kernel allows the filter to know it's from an older node. Or revert the merging of spawn with rpc and duplicate the filter logic. I'm not really a fan of either but would prefer the former if the new -> old node spawning won't be acceptable.