Parrot 7.0.0 "Crimson Shining Parrot" | parrot.org/ | Log: irclog.perlgeek.de/parrot | #parrotsketch meeting Fridays 13:30 UTC
Set by moderator on 21 January 2015.
00:51 kid51 joined 02:17 kid51_at_dinner_ joined 05:34 dalek joined 05:40 d4l3k_ joined 05:42 dalek joined 06:40 rurban joined 06:51 sorear joined
dalek rrot/smoke-me/icu-gh1188: 66c6d4a | rurban++ | config/auto/icu.pm:
[config] remove internal {without} arg when constant

small refactor to reduce unneeded complexity		 07:04
rrot/smoke-me/icu-gh1188: e4e767a | rurban++ | config/auto/icu.pm:
[config] fix ICU tests drop headers #1188

we forgot to pass the -I directives to _probe_icu. Closes #1188
07:18 travis-ci joined
travis-ci parrot/parrot#1511 (smoke-me/icu-gh1188 - e4e767a : Reini Urban): The build passed. 07:18
Change view : github.com/parrot/parrot/compare/6...e767ae82c4
Build details : travis-ci.org/parrot/parrot/builds/48016980
07:18 travis-ci left 07:19 Timbus joined 07:42 FROGGS joined
dalek rrot/smoke-me/icu-gh1188: 125e507 | rurban++ | config/auto/icu.pm:
[config] one more icuheaders probe fix for GH #1188

First try without icuheaders incpath, then with.
Note: This will still fail with shared icu, if the icu libpath is not in the LD LIBRARY search path. but icu-config does not provide a rpath.		 07:53
rrot: 0c49b30 | rurban++ | src/ (3 files):
[docs] alarm and threads

Update docs
08:06 particle joined
dalek rrot/smoke-me/7.0.1: 24b8b51 | rurban++ | / (21 files):
Prepare 7.0.1 hotfix release

Todo the 2nd t/stress/threads.t test, caused by stringbuilder gc, probably #1123.		 08:34
rrot/smoke-me/7.0.1: 391c214 | rurban++ | config/auto/icu.pm:
[config] remove internal {without} arg when constant

small refactor to reduce unneeded complexity
rrot/smoke-me/7.0.1: 0ba29ff | rurban++ | config/auto/icu.pm:
[config] fix ICU tests drop headers #1188

we forgot to pass the -I directives to _probe_icu. Closes #1188
rrot/smoke-me/7.0.1: cdd184a | rurban++ | config/auto/icu.pm:
[config] one more icuheaders probe fix for GH #1188

First try without icuheaders incpath, then with.
Note: This will still fail with shared icu, if the icu libpath is not in the LD LIBRARY search path. but icu-config does not provide a rpath.
rrot/smoke-me/7.0.1: 00cc593 | rurban++ | / (2 files):
[test] fix t/steps/auto/icu-01.t

remove tests for outdated internal functions and fix some old capture tests
rrot/smoke-me/7.0.1: 29491dc | rurban++ | / (2 files):
Release 7.0.1
08:46 rurban1 joined
dalek rrot/smoke-me/7.0.1: c4ec76d | rurban++ | / (5 files):
Release 7.0.1
 11:09
11:10 kid51 joined 11:24 travis-ci joined
travis-ci parrot/parrot#1514 (smoke-me/7.0.1 - c4ec76d : Reini Urban): The build passed. 11:24
Change view : github.com/parrot/parrot/compare/2...ec76d41488
Build details : travis-ci.org/parrot/parrot/builds/48035528
11:24 travis-ci left
rurban did the release 7.0.1 11:26
dalek rrot: 1bfb43c | rurban++ | docs/project/release_manager_guide.pod:
[docs] cover 7.3 in release_manager_guide
 11:34
rrot: ba5fd2f | rurban++ | docs/project/release_manager_guide.pod:
[docs] minor release_manager_guide improvement
rrot: 0006e12 | rurban++ | / (2 files):
[pmc] rpa: fix custom_mark_destroy confusion

custom_mark_destroy is only needed in the init method, which is fpa.set_integer_native. optimize PObj_custom_mark_destroy_SETALL to use only one setter call.
rrot: 56c7036 | rurban++ | / (3 files):
[GC] fix GC segfault with --optimize

A wrong PARROT_CANNOT_RETURN_NULL in Parrot_pf_get_current_code_segment
  (unchecked return of inter->code, which can be null) causes GC segfaults
when inter->code == NULL.
Fixes GH #1186. This wrong PARROT_CANNOT_RETURN_NULL was added with 3.6.0 in 7dc0e2276d3612afd at Wed Jun 29 20:52:24 2011 but started being fatal after the new packfile refactor with 7.0, when the cc optimizer started optimizing away the bc != NULL check in mark_code_segment()
rrot: f14a5a1 | rurban++ | src/gc/mark_sweep.c:
[GC] optimize away a Parrot_pf_get_current_code_segment call

This was the call with the wrong CANNOT_RETURN_NULL attribute, causing GH #1186. We only need inter->code, and it can be const. The const_table also
rrot: 5069da4 | rurban++ | src/gc/gc_gms.c:
[threads] fix deadlock in gc_gms_mark_and_sweep

move the lock downwards to the guard. Fixes GH #1187
rrot: 24b8b51 | rurban++ | / (21 files):
Prepare 7.0.1 hotfix release

Todo the 2nd t/stress/threads.t test, caused by stringbuilder gc, probably #1123.
rrot: 391c214 | rurban++ | config/auto/icu.pm:
[config] remove internal {without} arg when constant

small refactor to reduce unneeded complexity
rrot: 0ba29ff | rurban++ | config/auto/icu.pm:
[config] fix ICU tests drop headers #1188

we forgot to pass the -I directives to _probe_icu. Closes #1188
rrot: cdd184a | rurban++ | config/auto/icu.pm:
[config] one more icuheaders probe fix for GH #1188

First try without icuheaders incpath, then with.
Note: This will still fail with shared icu, if the icu libpath is not in the LD LIBRARY search path. but icu-config does not provide a rpath.
rrot: 00cc593 | rurban++ | / (2 files):
[test] fix t/steps/auto/icu-01.t

remove tests for outdated internal functions and fix some old capture tests
rrot: c4ec76d | rurban++ | / (5 files):
Release 7.0.1
rrot: 4dbace2 | rurban++ | / (26 files):
Merge branch '7.0.1'

Fixed Conflict:
  \tChangeLog
moderator Parrot 7.0.1 "Crimson Shining Parrot" | parrot.org/ | Log: irclog.perlgeek.de/parrot | #parrotsketch�meeting Fridays 13:30 UTC 11:39
11:50 travis-ci joined
travis-ci parrot/parrot#1515 (master - 4dbace2 : Reini Urban): The build has errored. 11:50
Change view : github.com/parrot/parrot/compare/0...bace20fb85
Build details : travis-ci.org/parrot/parrot/builds/48037729
11:50 travis-ci left
rurban This error was a travis timeout with cpan 12:12
12:14 lauleon joined 12:29 rurban1 joined, Chirag joined 12:36 kid51_ joined 14:05 lauleon joined 14:11 lauleon joined
rurban1 hmm, make install problems on darwin 14:23
resp. make installable
but only within rakudo 14:24
14:36 rurban2 joined 14:38 particle joined 15:40 lauleon joined 15:47 lauleon joined 16:23 rurban joined
rurban aah, make clean was needed for rakudo/install/lib 16:37
16:49 rurban_ joined
dalek rrot/smoke-me/stringbuilder-gh1123: c44d180 | rurban++ | docs/project/release_manager_guide.pod:
[docs] update release_manager_guide

remove some outdated items. there is nothing version specific at the gh wiki and c2 wiki anymore.		 16:56
rrot/smoke-me/stringbuilder-gh1123: c868e4e | rurban++ | / (2 files):
[pmc] change StringBuilder overallocation

by 1.5, not 2. And for >8192 onto the next block size. The optimal overallocation is the golden ration 1.618, but we prefer int ops. A difference is not measurable.
rrot/smoke-me/stringbuilder-gh1123: 0250719 | rurban++ | / (2 files):
[config] change perl5 derived -O2 to -O3 for gcc/clang

Faster --optimize. 5% faster in parrot-bench. Closes GH #1185
rrot/smoke-me/stringbuilder-gh1123: 8cb2a30 | rurban++ | src/pmc/stringbuilder.pmc:
[pmc] disable 2x StringBuilder.substr clone, GH #1123

STRING_substr already creates a copy for us, no need to copy it twice. no regression.
Note that t/stress/gc.t started failing earlier, even in 7.0.0.
rrot/smoke-me/stringbuilder-gh1123: 96ab7d8 | rurban++ | src/pmc/stringbuilder.pmc:
[pmc] StringBuilder.substr needs no WB

GH #1123. 1.14% faster
rrot/smoke-me/stringbuilder-gh1123: 7c93c27 | rurban++ | config/init/optimize.pm:
comment: add GH #1184 testcase for src/gc/system.c amd64
rrot/rurban/thread-lock-gh1187: 3b49c11 | rurban++ | src/gc/gc_gms.c:
threads: refactor thread locks in the GC

Simplify GC locks to new macros:
  - THREAD_COND_{UN,}LOCK only if locked/lockable (with the self->locked guard)
  - THREAD_{UN,}LOCK unconditionally (race free).
17:18 travis-ci joined
travis-ci parrot/parrot#1516 (smoke-me/stringbuilder-gh1123 - 7c93c27 : Reini Urban): The build passed. 17:18
Change view : github.com/parrot/parrot/compare/c...93c27262a4
Build details : travis-ci.org/parrot/parrot/builds/48075491
17:18 travis-ci left 17:43 FROGGS joined 18:33 FROGGS_ joined 18:43 Khisanth joined
dalek rrot/smoke-me/stringbuilder-gh1123: d97d379 | rurban++ | docs/project/release_manager_guide.pod:
[docs] update release_manager_guide

remove some outdated items. there is nothing version specific at the gh wiki and c2 wiki anymore.		 21:19
rrot/smoke-me/stringbuilder-gh1123: f9f1f39 | rurban++ | / (2 files):
RESPONSIBLE_PARTIES + CREDITS: small updates

  cotto and whiteknight are non-active since 5.0
rrot/smoke-me/stringbuilder-gh1123: a00ee7f | rurban++ | config/init/optimize.pm:
comment: add GH #1184 testcase for src/gc/system.c amd64
rrot/smoke-me/stringbuilder-gh1123: d6acd78 | rurban++ | / (2 files):
[config] change perl5 derived -O2 to -O3 for gcc/clang

Faster --optimize. 5% faster in parrot-bench. Closes GH #1185
rrot/smoke-me/stringbuilder-gh1123: 49cb26f | rurban++ | / (2 files):
[pmc] change StringBuilder overallocation

by 1.5, not 2. And for >8192 onto the next block size. The optimal overallocation is the golden ration 1.618, but we prefer int ops. A difference is not measurable.
rrot/smoke-me/stringbuilder-gh1123: 04ec499 | rurban++ | src/pmc/stringbuilder.pmc:
[pmc] disable 2x StringBuilder.substr clone, GH #1123

STRING_substr already creates a copy for us, no need to copy it twice. no regression.
Note that t/stress/gc.t started failing earlier, even in 7.0.0.
rrot/smoke-me/stringbuilder-gh1123: f33e88b | rurban++ | src/pmc/stringbuilder.pmc:
[pmc] StringBuilder.substr needs no WB

GH #1123. 1.14% faster
rrot/smoke-me/stringbuilder-gh1123: f8617f8 | rurban++ | src/pmc/stringbuilder.pmc:
[pmc] Avoid StringBuilder copies on set_pmc, ...

and push_pmc, i_concatenate, when the arg is a StringBuilder PMC already. 0.5% faster.
rrot/smoke-me/stringbuilder-gh1123: e3ed19e | rurban++ | src/io/stringhandle.c:
[core] optimize io_stringhandle_write_b

Reuse write buffer if large enough. This happens very often. But surpringly no perf. improvement measurable.
rrot: d97d379 | rurban++ | docs/project/release_manager_guide.pod:
[docs] update release_manager_guide

remove some outdated items. there is nothing version specific at the gh wiki and c2 wiki anymore.
rrot: f9f1f39 | rurban++ | / (2 files):
RESPONSIBLE_PARTIES + CREDITS: small updates

  cotto and whiteknight are non-active since 5.0
rrot: a00ee7f | rurban++ | config/init/optimize.pm:
comment: add GH #1184 testcase for src/gc/system.c amd64
rrot/rurban/gcms-numfree: c868e4e | rurban++ | / (2 files):
[pmc] change StringBuilder overallocation

by 1.5, not 2. And for >8192 onto the next block size. The optimal overallocation is the golden ration 1.618, but we prefer int ops. A difference is not measurable.
rrot/rurban/gcms-numfree: 0250719 | rurban++ | / (2 files):
[config] change perl5 derived -O2 to -O3 for gcc/clang

Faster --optimize. 5% faster in parrot-bench. Closes GH #1185
rrot/rurban/gcms-numfree: fa1bc2c | rurban++ | src/pmc/stringbuilder.pmc:
[pmc] disable 2x StringBuilder.substr clone, GH #1123

STRING_substr already creates a copy for us, no need to copy it twice. no regression.
Note that t/stress/gc.t started failing earlier, even in 7.0.0.
rrot/rurban/gcms-numfree: 1cfe581 | rurban++ | src/pmc/stringbuilder.pmc:
[pmc] StringBuilder.substr needs no WB

GH #1123. 1.14% faster
rrot/rurban/gcms-numfree: 4d66f6c | rurban++ | config/init/optimize.pm:
comment: add GH #1184 testcase for src/gc/system.c amd64
rrot/rurban/gcms-numfree: 9cf2c34 | rurban++ | src/pmc/stringbuilder.pmc:
[pmc] Avoid StringBuilder copies on set_pmc, ...

and push_pmc, i_concatenate, when the arg is a StringBuilder PMC already. 0.5% faster.		 21:21
rrot/rurban/gcms-numfree: 68c7dac | rurban++ | / (2 files):
RESPONSIBLE_PARTIES + CREDITS: small updates

  cotto and whiteknight are non-active since 5.0
rrot/rurban/gcms-numfree: 53139d9 | rurban++ | src/gc/ (3 files):
gc_ms: skip counting num_free_objects unless GC_STATS (debugging only)
rrot/rurban/gcms-numfree: 9ff9bd6 | rurban++ | / (3 files):
gc: define the default GC, for PARROT_GC_DEFAULT_MS

We need a compile-time definition, to skip num_free_objects on gms or inf
rrot/rurban/gcms-numfree: e3980fd | rurban++ | src/io/stringhandle.c:
[core] optimize io_stringhandle_write_b

WIP. reuse write buffer if large enough. This happens very often
rrot/rurban/gcms-numfree: 825c61c | rurban++ | src/pmc/stringbuilder.pmc:
Revert "[pmc] Avoid StringBuilder copies on set_pmc, ..."

test: This reverts commit 9cf2c341d9b256d5acf6ae27d09673bcfd821f62.
21:36 rurban joined 22:22 lauleon joined 22:38 lauleon joined 22:48 travis-ci joined
travis-ci parrot/parrot#1517 (smoke-me/stringbuilder-gh1123 - e3ed19e : Reini Urban): The build passed. 22:48
Change view : github.com/parrot/parrot/compare/7...ed19e4b4c7
Build details : travis-ci.org/parrot/parrot/builds/48105542
22:48 travis-ci left 22:59 travis-ci joined
travis-ci parrot/parrot#1518 (master - a00ee7f : Reini Urban): The build passed. 22:59
Change view : github.com/parrot/parrot/compare/4...0ee7f05342
Build details : travis-ci.org/parrot/parrot/builds/48105583
22:59 travis-ci left 23:07 lauleon joined 23:12 lauleon joined 23:28 lauleon joined 23:55 kid51 joined

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!