update auto-sizes to latest master #718

wants to merge 151 commits into


None yet

gerdr commented Feb 25, 2012

should trigger a CI run...

Whiteknight and others added some commits Oct 19, 2011

@Whiteknight Whiteknight First draft prototypes of most of the new ops we are going to need fo…
…r a PCC revamp. For some reason, mk bootstrap-ops fails with a parse error, though I can't figure out why.
@Whiteknight Whiteknight Fix some parse/build problems with the new ops. Make headerizer and b…
@jkeenan jkeenan Require TAP::Harness::Archive and related modules only if we' making …
…an archive or sending to smolder.
@jkeenan jkeenan Remove the BEGIN block, which is no longer needed. ea13f56
@ayardley ayardley Update document to point to the correct pod checker test because ther…
…e has not

been a '/t/doc/pod.t' since v0.9.1
@ayardley ayardley Minor updates and revisions 948fbf1
@ayardley ayardley Note revision of 'docs/project/release_manager_guide.pod' in ChangeLog 96ff976
@ayardley ayardley Make 'make_html_docs.pl' name conform to other 'make' tools. efc7d2e
@ayardley ayardley Enable mk_html_docs.pl to pass html headers and links to Parrot::Docs…
@ayardley ayardley Minor edit 1af5323
@ayardley ayardley Added a simple body subroutine to process html passed to the module. 659ec6f
@ayardley ayardley Renamed make_html_docs.pl to mk_html_docs.pl to conform with other ma…
…ke named tools.
@Benabik Benabik frontend: Be pessimistic about arguments
The code to handle -o added two arguments to the final array.
However, longopt allows "-o file" to also be passed as
"--output=file".  This causes one argument to cause two to be added.
This means that we can't have the destination array be the same size
as the original.

I've decided to use the most pessimistic case for allocation, one
where every argument causes two to be added (parrot --output=file1
--output=file2 etc).  Alternative solutions include passing it as
"-o=file" instead and having prt0 handle breaking it apart, but this
is the fastest way to fix the memory overrun.
@ayardley ayardley Added comments to clarify the reason for the code updates. 5472b91
@ayardley ayardley Fixed typo. d712500
@ayardley ayardley Fixed typo 93f24ff
@Benabik Benabik Auto-replace TT# with GH# as per GH #707
This takes us from 316 to 145 references to trac tickets.  I suspect
they are all closed tickets.
@Benabik Benabik codeingstd: don't use C++ comments, benabik-- 15dd3bf
@leto leto Add a Travis CI config file 2d8947b
@leto leto [ci] Do a configure at 'script' time, not 'before_script' 7e7e8f4
@leto leto [ci] We use Configure.pl, not ./configure bed6acf
@leto leto [ci] See how Travis CI likes our full test suite 81051fb
@leto leto [ci] Fix a dirth of whitespace a6ccd19
@ayardley ayardley Update doc to fix issue #170. 74a17f0
@ayardley ayardley Fixed typo. c2d7820
@jkeenan jkeenan Merge branch 'kid51/fulltest_easy' 844e6f9
@jkeenan jkeenan Update MANIFEST to include .travis.yml. 5c9014e
@leto leto [ci] Make our codingstd test conveniently skip over .travis.yml 42997bc
@leto leto [ci] Test parrot with LLVM/clang as well as with/without optimization 8b1c382
@jkeenan jkeenan Config step auto::sha1 was removed some time ago! No longer need corr…
…esponding 'steps' test in the repo.
@ayardley ayardley Fixed minor error to earlier updates. b40e9d9
@ayardley ayardley Merge branch 'ayardley/mk_html_docs' 4533a7f
@leto leto [ci] Attempt to install perl dependencies so we can submit to Smolder 5b68a03
@leto leto [ci] Try removing local::lib setup since it is already intalled+confi…
…gured by apt-get
@leto leto [ci] Add zlib 8a1265a
@leto leto [ci] Try to fix a permission issue 9285a10
@leto leto [ci] Try fixing the permisions again and dump our environment 676284f
@fperrad fperrad [docs] fix build 032380e
@leto leto [ci] There is no ~/perl5 directory when local::lib is installed via a…
@leto leto [ci] Correct the dependencies for submitting to smolder via Perl and …
…don't run tests for CPAN dependencies
@leto leto [ci] Try to get the correct local::lib env vars set 1f4a9d3
@leto leto [ci] Since --test=build with return an error code if build tests fail…
…, run our make command unconditionally so Smolder reports get sent
@leto leto [ci] Test plain builds instead of ccache
We weren't getting a lot of ROI out of testing ccache vs. non-ccache
builds, i.e. I have never seen a ccache/non-ccache build pair have
different behaviors.

Much better to test a "plain" parrot compile for each compiler, which
means we turn off every possible Configure.pl flag. This tends to catch
test failures where tests have not been properly skipped if Parrot was
not compiled with the feature turned on.

This has the added benefit of using fewer resources on Travis CI and
running faster.
@leto leto [ci] Remove ccache, add llvm-dev and comment out env dump 0554832
@Whiteknight Whiteknight Merge from master. we had some conflicts in the ops because of the ea…
…rlier green_threads merge, so those conflicts have been resolved and the new ops files have been generated
@jkeenan jkeenan [codingstd] Rebreak line to conform to linelength standard. 0f1422c
@jkeenan jkeenan [codingstd] Rebreak line to avoid cuddled else. ad65ada
@jkeenan jkeenan [codingstd] No space permitted before closing paren. 2cabfb6
@jkeenan jkeenan Revert "[codingstd] No space permitted before closing paren."
This reverts commit 2cabfb6.
@jkeenan jkeenan 'ARGMOD' not appropriate with 'const'; must use 'ARGIN'. f4f5025
@jkeenan jkeenan [codingstd] No hard tabs. 3bc818b
@NotFound NotFound make parrot_hash_iterate_... usable like functions to decreeas dangli…
…ng else and semicolon placing problems
@jkeenan jkeenan [codingstd] Rebreak lines to meet c_parens, c_function_docs and linel…
…ength standards.
@jkeenan jkeenan Improve sentence flow in documentation of what test does. 41bcdab
@jkeenan jkeenan [codingstd] Provide at least stub documentation for recently added ex…
…perimental opcodes.
@jkeenan jkeenan Correct misspelled last name. b62f4b8
@Util Util [codingstd] Add missing ASSERT_ARGS. c_arg_assert.t now passes. 9274546
@leto leto [ci] Add gcc-multilib in hopes of making LLVM/clang happy 1b91854
@leto leto [ci] Turn on IRC notifications
Our test suite (smolder_fulltest) is now passing on Travis CI [0] .
Every commit is being tested with clang, gcc and g++ in a disposable
Ubuntu VM container.

We really need "make allhlltest", but that is so long that it would
timeout with their current settings.

Next step is Rakudo on Travis.

[0] http://travis-ci.org/#!/parrot/parrot/builds/659732
@leto leto Update my URL in CREDITS and some details ae4bf3d
@leto leto [ci] Remove #perl6 from irc notifications 0c54198
@Whiteknight Whiteknight Add an is_equal vtable to Class PMC e9e82f3
@Whiteknight Whiteknight Merge branch 'master' of github.com:parrot/parrot 6e53ad0
@Whiteknight Whiteknight If the first parameter is named 'self' and this is a method, do not a…
…utomagically prepend another 'self' parameter to the argument list
@Whiteknight Whiteknight Add in a test for explict self in methods 7691e5c
@Whiteknight Whiteknight Mention pcc ops in changelog 6fc5e92
@leto leto [ci] With the blessing of cotto++, start sending travis ci emails to …

leto commented on a611cbb Feb 11, 2012



leto commented on beffdd5 Feb 22, 2012

It could be preventing various nasty GC bugs from occuring

ayardley and others added some commits Feb 23, 2012

@ayardley ayardley Update to the release manager guide to instruct release managers to p…
…rovide a more informative topic for #parrot.
@moritz moritz [os] expose unlink and rmdir as separate methods
Perl 6 wants them separate, so it is much more convenient this way.
Also the .rm approach is prone to race conditions between the stat and
the unlink/rmdir call
@Benabik Benabik Fix building on Cygwin
The DLL is cygparrot4_1_0.dll, not cygparrot4.1.0.dll.  This seems to
be a bug with update_version.pl, which should be corrected before the
next release.

Reported by gfldex++ on #perl6
@gerdr gerdr Fix issue #720 by quoting version regex and renaming DLL on Cygwin
The old name was non-standard anyway
@gerdr gerdr fix issue #709 on clang
treats clang like g++
icc handling still broken
@leto leto [doc][ci skip] Add skeleton for next release announcement c1109bb
@leto leto [tools] Script to add a new release announcement skeleton to ChangeLog fe16231
@leto leto [doc][ci skip] Update the release manager guide and relieve poor rele…
…ase managers of manually writing a release announcement skeleton
@bacek bacek Remove new_si op. Closes #203. f8e4a7a
@bacek bacek Rebootstrap ops 41cc2cd
@jkeenan jkeenan [codingstd] Modify file to pass 'perlcritic' re: no subroutine
prototypes; no two-argument 'open'; no 'return undef'.
@bacek bacek Copy prop-related functions from "default" pmc into parrot-wide set. 33dd6db
@bacek bacek Replace prop VTABLEs with function calls f88e925
@bacek bacek Rebootstrap ops 2d2251b
@bacek bacek Mass replace VTABLE with function calls 4a9efbc
@bacek bacek Don't generate VTABLE_getprop b7b2470
@bacek bacek Don't use VTABLE_getprop in imcc b857157
@bacek bacek Regenerate imcc grammar b16b688
@bacek bacek Finally kill props VTABLES cdf2bd9
@Whiteknight Whiteknight Fix possible segfault if we try to get the main_sub from a packfile t…
…hat has no constants or no main_sub
@Whiteknight Whiteknight Merge branch 'master' of github.com:parrot/parrot 021101f
@bacek bacek Mark pmc prop functions with PARROT_EXPORT 3378465
@bacek bacek Update test to latest list of vtables. e40e72a
@bacek bacek Update test to remove prop VTABLEs testing. 6f6d7d0
@leto leto [ci] Tell Travis to use a perl worker and test under perl 5.10 and 5.14 c662128
@bacek bacek Merge branch 'kill_props_vtables' 47bfa1f
@leto leto [ci] Perl versions must be quoted because they are strings 5555d82
@leto leto [ci] Specify a dummy install command so cpanminus is not invoked, sin…
…ce it doesn't know what to do with Parrot
@leto leto [ci] Submitting smoke tests requires LWP::UserAgent as a dependency d82182c
@leto leto [ci] Only run Travis on the master branch and only on 5.14 for now c7f99d7
@bacek bacek Add new getprop variant and deprecate old one. Part of #351 4a482c7
@bacek bacek Rebootstrap ops a8dd597
@petdance petdance remove capture from match 675a804
@petdance petdance removed -Wzero-as-null-pointer-constant because of too many false pos…
@Whiteknight Whiteknight [ci skip] Travis CI shouldn't attempt a build with clang. It's always…
… failing config and we don't know why yet.
@petdance petdance Updating function parameters 48317e5
@petdance petdance adding some C++-specific warnings and a cage check for redundant decl…
@petdance petdance removed old function declarations 2c7436f
@petdance petdance remove redundant declaration 608aa0a
@petdance petdance add -Wstrict-overflow to see what that turns up 2c959c2
@petdance petdance adding some more warnings 075a574
@leto leto Merge pull request #715 from gerdr/gerdr/709-clang
fix issue #709 on clang
@bacek bacek Update pir book to new getprop syntax. 10ae7c6
@bacek bacek Update nqp-rx in-place to new getprop syntax cebe8d7
@bacek bacek Update PCT to new getprop version a7790bf
@bacek bacek Update tests to use new getprop version 5c3c2fe
@bacek bacek Update tetris to use new getprop version 29f72df
@bacek bacek Update runtime to use new getprop version 4b23603
@bacek bacek Update test to new getprop version 179831d
@bacek bacek Add mem_internal_realloc_n_typed for consistency 67b0e5e
@bacek bacek Update PointerArray to use mem_internal_* functions. 6444a67
@bacek bacek Don't use mem_sys_* macros. They are deprecated 8213cff
@gerdr gerdr char is allowed to be unsigned, so explictily request signed type a44b098
@gerdr gerdr remove unnecessary type ptrcast_t 7086655
@gerdr gerdr remove _set_ptrcast() from auto::memalign ea21140
@gerdr gerdr fix misuse of %u to printf values of type size_t 2e75f95
@gerdr gerdr remove needless inclusion of stdint.h a071a37
@gerdr gerdr re-introduce ptrcast_t and use it where actually appropriate dcd0e0a
@gerdr gerdr re-factor auto::sizes 86b2e05
@gerdr gerdr bugfix: change opcode type when changing intval type a17e15c
@jkeenan @gerdr jkeenan Test auto::sizes::runstep(). adaef97
@jkeenan @gerdr jkeenan Add tests for new version of auto::sizes. bcb3786
@jkeenan @gerdr jkeenan Correct number of tests in plan. 4e0236b
@jkeenan @gerdr jkeenan Update copyright date. 6a7a693
@jkeenan @gerdr jkeenan Test all execution paths (branches) for auto::sizes::_set_intval_rang…
@gerdr gerdr add FIXME note about #705 1309ad5
@jkeenan @gerdr jkeenan Test all execution paths for _set_floatval_range(). ce54ea6
@gerdr gerdr use _minimal_ type with size greater than sizeof (void *) for ptrcast
no change on non-exotic platforms, but in priciple the right thing to do

gerdr closed this Mar 5, 2012

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment