Perl 6 - the future is here, just unevenly distributed

IRC log for #parrot, 2014-12-02

Parrot | source cross referenced

| Channels | #parrot index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
01:03 particle1 joined #parrot
02:43 kid51 joined #parrot
03:14 kid51 joined #parrot
03:17 kid51_ joined #parrot
03:24 kid51_ joined #parrot
03:32 kid51_ joined #parrot
04:25 sorear joined #parrot
07:21 FROGGS joined #parrot
11:27 dalek parrot/smoke-me/rpa-offset-gh1152: 63ed0d4 | rurban++ | src/pmc/ (2 files):
11:27 dalek parrot/smoke-me/rpa-offset-gh1152: [pmc] rpa: fix visit
11:27 dalek parrot/smoke-me/rpa-offset-gh1152:
11:27 dalek parrot/smoke-me/rpa-offset-gh1152: we need an offset for visit, we cannot use the fixedpmcarray
11:27 dalek parrot/smoke-me/rpa-offset-gh1152: method.
11:27 dalek parrot/smoke-me/rpa-offset-gh1152: review: https://github.com/parrot/parrot/commit/63ed0d4246
11:27 dalek parrot/smoke-me/rpa-offset-gh1152: 6b4db01 | rurban++ | / (2 files):
11:27 dalek parrot/smoke-me/rpa-offset-gh1152: [pmc] rpa: fix resize with offset optimization
11:27 dalek parrot/smoke-me/rpa-offset-gh1152:
11:27 dalek parrot/smoke-me/rpa-offset-gh1152: we moved the wrong items, and had wrong slack overwrites.
11:27 dalek parrot/smoke-me/rpa-offset-gh1152: Add testcases.
11:28 dalek parrot/smoke-me/rpa-offset-gh1152: review: https://github.com/parrot/parrot/commit/6b4db013c8
11:28 dalek parrot/smoke-me/rpa-offset-gh1152: c8a7211 | rurban++ | / (4 files):
11:28 dalek parrot/smoke-me/rpa-offset-gh1152: [pmc]: various minor refactors
11:28 dalek parrot/smoke-me/rpa-offset-gh1152:
11:28 dalek parrot/smoke-me/rpa-offset-gh1152: * Access attr struct more directly.
11:28 dalek parrot/smoke-me/rpa-offset-gh1152: * Fix one syntax problem with UNLIKELY with non-std compilers.
11:28 dalek parrot/smoke-me/rpa-offset-gh1152: * Optimize *1.5 by using integer math.
11:28 dalek parrot/smoke-me/rpa-offset-gh1152: * Change rpa overallocation to use 1.5, not 2. This is better to handle
11:28 dalek parrot/smoke-me/rpa-offset-gh1152:   free blocks.
11:28 dalek parrot/smoke-me/rpa-offset-gh1152: * Always fill slack space, as we extend rpa via set. In a better version we
11:28 dalek parrot/smoke-me/rpa-offset-gh1152:   should fill the slack space only when needed: only on set.
11:28 dalek parrot/smoke-me/rpa-offset-gh1152: * Change push_{non-pmc} to use push_pmc and not set_keyed_int.
11:28 dalek parrot/smoke-me/rpa-offset-gh1152:   push is offset optimized, set not.
11:28 dalek parrot/smoke-me/rpa-offset-gh1152:
11:28 dalek parrot/smoke-me/rpa-offset-gh1152: Still WIP. object was probably modified after being freed
11:28 dalek parrot/smoke-me/rpa-offset-gh1152: review: https://github.com/parrot/parrot/commit/c8a7211520
11:38 travis-ci joined #parrot
11:38 travis-ci parrot/parrot#1361 (smoke-me/rpa-offset-gh1152 - c8a7211 : Reini Urban): The build is still failing.
11:38 travis-ci Change view : https://github.com/parrot/parrot/compare/5ff0fc926be9...c8a7211520f9
11:38 travis-ci Build details : http://travis-ci.org/parrot/parrot/builds/42726894
11:38 travis-ci left #parrot
13:09 kid51 joined #parrot
14:23 dalek parrot/smoke-me/rpa-offset-gh1152: b4e7b00 | rurban++ | / (4 files):
14:23 dalek parrot/smoke-me/rpa-offset-gh1152: [pmc]: rpa: optimize slack filling with PMCNULL
14:23 dalek parrot/smoke-me/rpa-offset-gh1152:
14:23 dalek parrot/smoke-me/rpa-offset-gh1152: Only fill the rest with PMCNULL when necessary, i.e. when
14:23 dalek parrot/smoke-me/rpa-offset-gh1152: we set an index > size, the intermediate indices between size and index-1.
14:23 dalek parrot/smoke-me/rpa-offset-gh1152: This saves a lot of slack filling code cycles.
14:23 dalek parrot/smoke-me/rpa-offset-gh1152:
14:23 dalek parrot/smoke-me/rpa-offset-gh1152: Error on delete oob.
14:23 dalek parrot/smoke-me/rpa-offset-gh1152: review: https://github.com/parrot/parrot/commit/b4e7b00f7e
14:23 dalek parrot/smoke-me/rpa-offset-gh1152: 6a1917f | rurban++ | src/pmc/resizablepmcarray.pmc:
14:23 dalek parrot/smoke-me/rpa-offset-gh1152: [pmc]: rpa: optimize delete_keyed_int
14:23 dalek parrot/smoke-me/rpa-offset-gh1152:
14:23 dalek parrot/smoke-me/rpa-offset-gh1152: on index=0 just bump the offset as with shift,
14:23 dalek parrot/smoke-me/rpa-offset-gh1152: else memmove en-block, not item by item in a loop.
14:23 dalek parrot/smoke-me/rpa-offset-gh1152:
14:24 dalek parrot/smoke-me/rpa-offset-gh1152: This passes now most rpa related tests, just reverse has a conflict
14:24 dalek parrot/smoke-me/rpa-offset-gh1152: with fpa. with reverse renamed to nreverse make test and benchmarking
14:24 dalek parrot/smoke-me/rpa-offset-gh1152: is possible.
14:24 dalek parrot/smoke-me/rpa-offset-gh1152: review: https://github.com/parrot/parrot/commit/6a1917f001
14:41 travis-ci joined #parrot
14:41 travis-ci parrot/parrot#1362 (smoke-me/rpa-offset-gh1152 - 6a1917f : Reini Urban): The build is still failing.
14:41 travis-ci Change view : https://github.com/parrot/parrot/compare/c8a7211520f9...6a1917f00185
14:41 travis-ci Build details : http://travis-ci.org/parrot/parrot/builds/42744107
14:41 travis-ci left #parrot
14:44 dalek parrot/smoke-me/rpa-offset-gh1152: c3742a5 | rurban++ | / (4 files):
14:44 dalek parrot/smoke-me/rpa-offset-gh1152: [ext] fix the remaining rpa reverse problem
14:44 dalek parrot/smoke-me/rpa-offset-gh1152:
14:44 dalek parrot/smoke-me/rpa-offset-gh1152: nqp ResizablePMCArray provided its own reverse method, which is now
14:44 dalek parrot/smoke-me/rpa-offset-gh1152: not needed anymore. We use the better pir method instead.
14:44 dalek parrot/smoke-me/rpa-offset-gh1152:
14:44 dalek parrot/smoke-me/rpa-offset-gh1152: Also found a syntax problem in parrot_old
14:44 dalek parrot/smoke-me/rpa-offset-gh1152:
14:44 dalek parrot/smoke-me/rpa-offset-gh1152: Remaining todos:
14:44 dalek parrot/smoke-me/rpa-offset-gh1152: t/op/lexicals.t                           (Wstat: 256 Tests: 54 Failed: 1)
14:44 dalek parrot/smoke-me/rpa-offset-gh1152:   Failed test:  27
14:44 dalek parrot/smoke-me/rpa-offset-gh1152: t/compilers/pge/p5regex/p5rx.t            (Wstat: 0 Tests: 960 Failed: 7)
14:44 dalek parrot/smoke-me/rpa-offset-gh1152:   Failed tests:  172, 420-421, 442-443, 869-870
14:44 dalek parrot/smoke-me/rpa-offset-gh1152: t/src/extend_vtable.t                     (Wstat: 512 Tests: 134 Failed: 2)
14:44 dalek parrot/smoke-me/rpa-offset-gh1152:   Failed tests:  34, 37
14:44 dalek parrot/smoke-me/rpa-offset-gh1152: review: https://github.com/parrot/parrot/commit/c3742a5c08
15:05 travis-ci joined #parrot
15:05 travis-ci parrot/parrot#1363 (smoke-me/rpa-offset-gh1152 - c3742a5 : Reini Urban): The build has errored.
15:05 travis-ci Change view : https://github.com/parrot/parrot/compare/6a1917f00185...c3742a5c0879
15:05 travis-ci Build details : http://travis-ci.org/parrot/parrot/builds/42746413
15:05 travis-ci left #parrot
15:55 Reini joined #parrot
15:56 Reini rpa's almost ready. let's now see how much faster they are. should be faster than qrpa's.
15:59 dalek parrot/smoke-me/rpa-offset-gh1152: ec0b57a | rurban++ | / (4 files):
15:59 dalek parrot/smoke-me/rpa-offset-gh1152: [core] apifify -t20 as PARROT_TRACE_ARRAY_STATE_FLAG
15:59 dalek parrot/smoke-me/rpa-offset-gh1152:
15:59 dalek parrot/smoke-me/rpa-offset-gh1152: Only available without --optimize, i.e. #ifndef NDEBUG
15:59 dalek parrot/smoke-me/rpa-offset-gh1152: review: https://github.com/parrot/parrot/commit/ec0b57a3b1
15:59 dalek parrot/smoke-me/rpa-offset-gh1152: 8005345 | rurban++ | docs/ (2 files):
15:59 dalek parrot/smoke-me/rpa-offset-gh1152: [docs] Update stability and memory_internals
15:59 dalek parrot/smoke-me/rpa-offset-gh1152:
15:59 dalek parrot/smoke-me/rpa-offset-gh1152: stability was written in 2007. Just the deprecation cycle
15:59 dalek parrot/smoke-me/rpa-offset-gh1152: needed to be updated.
15:59 dalek parrot/smoke-me/rpa-offset-gh1152:
15:59 dalek parrot/smoke-me/rpa-offset-gh1152: memory_internals talks as if mark&sweep is the only GC.
15:59 dalek parrot/smoke-me/rpa-offset-gh1152: Improve that.
15:59 dalek parrot/smoke-me/rpa-offset-gh1152: review: https://github.com/parrot/parrot/commit/8005345686
15:59 dalek parrot: f413a4b | rurban++ | t/compilers/pct/complete_workflow.t:
15:59 dalek parrot: [test] fix GLOB* temp file creation
15:59 dalek parrot:
15:59 dalek parrot: regression from commit caba76f589a9aa.
15:59 dalek parrot: mixed up handle and filename, as run_command actually requires
15:59 dalek parrot: a name and not a handle for STDERR.
15:59 dalek parrot: review: https://github.com/parrot/parrot/commit/f413a4bc28
15:59 dalek parrot: 6e5ea61 | rurban++ | docs/ (2 files):
15:59 dalek parrot: [docs] Update stability and memory_internals
15:59 dalek parrot:
15:59 dalek parrot: stability was written in 2007. Just the deprecation cycle
15:59 dalek parrot: needed to be updated.
15:59 dalek parrot:
15:59 dalek parrot: memory_internals talks as if mark&sweep is the only GC.
15:59 dalek parrot: Improve that.
15:59 dalek parrot: review: https://github.com/parrot/parrot/commit/6e5ea6188a
16:06 dalek Heuristic branch merge: pushed 20 commits to parrot/smoke-me/rpa-offset-gh1152 by rurban
16:08 Reini 10% slower, huh?
16:16 Reini nope, too high load. on my dedicated bench machine it's -22.96%
16:16 Reini 20% was expected. good
16:19 Reini maybe another −20% with ms2 and I call it done
16:28 Reini yep, and perl6 is also faster. good
16:29 Reini lot of shift + push in the parser engines
16:31 travis-ci joined #parrot
16:31 travis-ci parrot/parrot#1364 (master - 6e5ea61 : Reini Urban): The build passed.
16:31 travis-ci Change view : https://github.com/parrot/parrot/compare/ac68d44a0582...6e5ea6188a89
16:31 travis-ci Build details : http://travis-ci.org/parrot/parrot/builds/42755786
16:31 travis-ci left #parrot
16:42 travis-ci joined #parrot
16:42 travis-ci parrot/parrot#1364 (smoke-me/rpa-offset-gh1152 - 8005345 : Reini Urban): The build has errored.
16:42 travis-ci Change view : https://github.com/parrot/parrot/compare/c3742a5c0879...8005345686fa
16:42 travis-ci Build details : http://travis-ci.org/parrot/parrot/builds/42755797
16:42 travis-ci left #parrot
16:54 Reini And now I can repro the performance win on my mac also: 13.59% faster
16:54 Reini this is big
16:55 Reini but there's still a double free or corruption
16:56 Reini in splice
17:06 travis-ci joined #parrot
17:06 travis-ci parrot/parrot#1365 (smoke-me/rpa-offset-gh1152 - 3e1e5df : Reini Urban): The build has errored.
17:06 travis-ci Change view : https://github.com/parrot/parrot/compare/8005345686fa...3e1e5df0c9c7
17:06 travis-ci Build details : http://travis-ci.org/parrot/parrot/builds/42756804
17:06 travis-ci left #parrot
18:28 dalek parrot/smoke-me/rpa-offset-gh1152: bd63ea7 | rurban++ | src/pmc/resizablepmcarray.pmc:
18:28 dalek parrot/smoke-me/rpa-offset-gh1152: [pmc] rpa: fix a resize edgecase with size=0
18:28 dalek parrot/smoke-me/rpa-offset-gh1152:
18:28 dalek parrot/smoke-me/rpa-offset-gh1152: avoid negative offsets.
18:28 dalek parrot/smoke-me/rpa-offset-gh1152: repro:
18:28 dalek parrot/smoke-me/rpa-offset-gh1152:   valgrind ./parrot_old -t20 parrot-nqp.pbc --target=pir compilers/data_json/JSON.nqp
18:28 dalek parrot/smoke-me/rpa-offset-gh1152: review: https://github.com/parrot/parrot/commit/bd63ea782e
18:58 travis-ci joined #parrot
18:58 travis-ci parrot/parrot#1366 (smoke-me/rpa-offset-gh1152 - bd63ea7 : Reini Urban): The build failed.
18:58 travis-ci Change view : https://github.com/parrot/parrot/compare/3e1e5df0c9c7...bd63ea782e0d
18:58 travis-ci Build details : http://travis-ci.org/parrot/parrot/builds/42772372
18:58 travis-ci left #parrot
19:07 FROGGS joined #parrot
21:55 Reini another 2.5% faster, better shift
22:08 lauleon joined #parrot
22:08 dalek parrot/smoke-me/rpa-offset-gh1152: 2c49485 | rurban++ | src/pmc/resizablepmcarray.pmc:
22:08 dalek parrot/smoke-me/rpa-offset-gh1152: [pmc] rpa: fix invalid read in delete_keyed_int, off by one
22:08 dalek parrot/smoke-me/rpa-offset-gh1152:
22:08 dalek parrot/smoke-me/rpa-offset-gh1152: with (1,9,10) and key=1 we read one too far.
22:08 dalek parrot/smoke-me/rpa-offset-gh1152: repro:
22:08 dalek parrot/smoke-me/rpa-offset-gh1152:   valgrind ./parrot_old -t20 t/compilers/pge/p5regex/p5rx.t
22:08 dalek parrot/smoke-me/rpa-offset-gh1152: review: https://github.com/parrot/parrot/commit/2c4948572c
22:08 dalek parrot/smoke-me/rpa-offset-gh1152: e4fe79c | rurban++ | src/pmc/resizablepmcarray.pmc:
22:08 dalek parrot/smoke-me/rpa-offset-gh1152: [pmc] rpa: shift optim, use data struct directly
22:08 dalek parrot/smoke-me/rpa-offset-gh1152:
22:08 dalek parrot/smoke-me/rpa-offset-gh1152: do not use seperate SELF and temp values for the attrs.
22:08 dalek parrot/smoke-me/rpa-offset-gh1152: 2.40% faster on my mac, %ebp addressing for data.
22:08 dalek parrot/smoke-me/rpa-offset-gh1152: review: https://github.com/parrot/parrot/commit/e4fe79c647
22:08 dalek parrot/smoke-me/rpa-offset-gh1152: 73aacbb | rurban++ | src/pmc/resizablepmcarray.pmc:
22:08 dalek parrot/smoke-me/rpa-offset-gh1152: [pmc] rpa: internal rename trace macros
22:08 dalek parrot/smoke-me/rpa-offset-gh1152:
22:09 dalek parrot/smoke-me/rpa-offset-gh1152: add TRACE_RPAdata
22:09 dalek parrot/smoke-me/rpa-offset-gh1152: review: https://github.com/parrot/parrot/commit/73aacbb414
22:24 travis-ci joined #parrot
22:24 travis-ci parrot/parrot#1367 (smoke-me/rpa-offset-gh1152 - 73aacbb : Reini Urban): The build is still failing.
22:24 travis-ci Change view : https://github.com/parrot/parrot/compare/bd63ea782e0d...73aacbb414a9
22:24 travis-ci Build details : http://travis-ci.org/parrot/parrot/builds/42796246
22:24 travis-ci left #parrot
22:52 Reini left #parrot
22:53 Reini joined #parrot
22:54 Reini joined #parrot
23:06 freudsfeud joined #parrot
23:30 kid51 joined #parrot

| Channels | #parrot index | Today | | Search | Google Search | Plain-Text | summary

Parrot | source cross referenced