|
Parrot 4.10.0 "[Red-eared Parakeet]" parrot.org/ | Log: irclog.perlgeek.de/parrot Set by moderator on 21 November 2012. |
|||
|
00:02
kid51 joined
00:05
Psyche^ joined
|
|||
| whiteknight | rurban: the idea is a good one. Last time I proposed it some other people weren't happy with it | 00:06 | |
| it doesn't improve performance by itself, but it's an API we can start using now in preparation for 6model | 00:07 | ||
|
01:14
Mike-PerlRecruiter_ joined
|
|||
| dalek | rrot/report_gmp_version: 582558b | dukeleto++ | config/auto/gmp (2 files): [config] Report GMP version from Configure.pl This information was not stored or shown anywhere else, which makes it inconvenient to debug issues relating to GMP. |
01:45 | |
| rrot: 243c292 | dukeleto++ | / (2 files): [t] Fix some coding standard tests to make Travis happy |
01:54 | ||
| rrot/check_for_git_binary_gh477: b704856 | dukeleto++ | lib/Parrot/ (4 files): Detect run-time support of git instead of assuming .git implies a git binary, #477 |
02:33 | ||
| rrot/check_for_git_binary_gh477: c83ac34 | dukeleto++ | lib/Parrot/Git.pm: Allow Parrot::Git to properly export functions |
|||
|
02:38
kid51 joined
|
|||
| kid51 | Is there a definitive way I can determine whether, when I know I'm "on Intel", I'm on "32bit" or "64bit"? | 02:39 | |
| sorear | what kind of answer do you want? | 02:46 | |
| do you want to know if the processor supports 64-bit? | 02:47 | ||
| do you want to know if the C compiler generates 64-bit executables by default? | |||
| do you want to know if some specific program is compiled as a 64-bit program? | |||
| do you want to know if the C compiler can generate 64-bit executables? | |||
| kid51 | Either of these things: | 02:48 | |
| 1. When I build Parrot on i386 vs x86_64, where is that reflected in myconfig, config_lib.pir or lib/Parrot/Config/Generated.pm? | |||
| 2. Some command line thing that makes that distinction | 02:49 | ||
| Really so that I can comment intelligently on Parrot github issues or Perl 5 rt.perl.org tickets. | 02:50 | ||
| sorear | kid51: is there a config item for sizeof(void*) or sizeof(INTVAL)? | ||
| kid51 | Probably, but I'd have to grep. | ||
| Can you look at config/auto/sizes.pm? | 02:51 | ||
| sorear | it'd be 4 or 8 | 02:53 | |
| I don't have a parrot handy | |||
| kid51 | Alright, on my Darwin/PPC (Mac OS X 10.4.11), I have intvalsize => 4. | ||
| On this Darwin/Intel, I have intvalsize > 8 | 02:54 | ||
| (both from lib/Parrot/Config/Generated.pm) | |||
| I *believe* the Intel is x86_64, but I want to be able to state that accurately. | |||
| On this Darwin/Intel box, when I say: ./parrot --version | 02:56 | ||
| I get "Parrot version 4.10.0-devel built for i386-darwin" | |||
| However, when I say 'uname -a', I get (in part): | 02:57 | ||
| root:xnu-1699.32.7~1/RELEASE_X86_64 x86_64 | 02:58 | ||
| ... which is why I am confused | |||
| dalek | rrot/native_pbc2: b40373b | rurban++ | src/packfile/pf_items.c: fix 8_10_le=>8_8_le converter Do not memcpy 16 to 10. |
04:02 | |
| Heuristic branch merge: pushed 70 commits to parrot/native_pbc2 by rurban | 04:03 | ||
| benabik | msg kid51 I suggest using `file parrot` to discover the type of the executable. My --version says i386-darwin, but file reports 'Mach-O 64-bit executable x86_64' | 04:19 | |
| aloha | OK. I'll deliver the message. | ||
| benabik | OTOH, that suggests our system detection code could use some work. :-/ | ||
| On the gripping hand, even `gcc -v` reports 'i686-apple-darwin11', so... | 04:20 | ||
| dalek | kudo/nom: 798dbdd | (Solomon Foster)++ | src/ (3 files): Add Texas versions of the Set and Bag operators. |
04:48 | |
|
04:51
Reini joined
|
|||
| Reini | I found a rpath trick on darwin! blogs.oracle.com/dipol/entry/dynam...th_and_mac | 04:51 | |
| it works | |||
| install_name_tool -id "@loader_path/blib/lib/libparrot.dylib" blib/lib/libparrot.dylib | 04:52 | ||
| make parrot | |||
|
05:08
PerlJam joined
05:21
Util joined
05:42
dalek joined
05:56
Util joined
05:58
dalek joined
|
|||
| dalek | rrot/getprotobyname_gh606: b21673b | rurban++ | / (7 files): [GH #606] Refactor Socket.getprotobyname, add test Put it into platform/generic/socket.c |
06:06 | |
|
06:14
tadzik joined
06:19
PerlJam joined,
Util joined
|
|||
| benabik | msg Reini I suggest looking at the first comment. There's apparently even more flexible options like @rpath. developer.apple.com/mac/library/rel...ols/RN-Id/ | 06:21 | |
| aloha | OK. I'll deliver the message. | ||
|
06:28
Util joined,
tadzik joined
06:30
dalek joined
06:39
Util joined
07:38
tadzik joined
07:55
p6eval joined
08:24
benabik joined
10:25
migimunz joined
10:35
woosley left
10:42
migimunz2 joined
10:47
drift joined
11:05
migimunz joined
11:26
migimunz joined
11:58
migimunz joined
12:01
migimunz2 joined
12:28
drift joined
12:59
Coke joined
13:14
Mike-PerlRecruiter_ joined
13:21
Util joined
13:44
benabik joined
14:11
bluescreen joined
14:16
PacoAir joined
14:24
Reini joined
14:29
PacoAir joined
14:36
Reini joined
|
|||
| rurban | benabik: Yes, @rpath (since 10.5) or @loader_path (since 10.4, similar to $ORIGIN) or @executable_path (ever). | 14:40 | |
|
15:11
benabik joined
15:38
dmalcolm joined
|
|||
| rurban | dukeleto: check_for_git_binary_gh477 tested ok on linux and win32 (without git) | 15:43 | |
|
16:07
PerlJam joined
|
|||
| rurban | I'll merge it after some more tests. I found more win32 troubles | 16:12 | |
|
16:29
benabik joined
|
|||
| ttbot | Parrot b7048560 i386-linux-thread-multi make error tt.taptinder.org/cmdinfo/103265 | 17:04 | |
|
17:44
benabik joined
|
|||
| dukeleto | i think that ttbot error is because it tested an older commit where the git stuff didn't work correctly yet | 18:30 | |
|
18:53
tuxit joined
|
|||
| rurban | yes, that was b7048560 | 18:53 | |
| I could not find the darwin threads problem yet, with chameneos | 18:54 | ||
|
19:51
bouncy joined
19:59
Reini joined
21:00
Reini joined
21:03
benabik joined
21:06
schmooster joined
21:46
benabik_ joined
21:56
Reini joined
21:58
benabik joined
22:32
whiteknight joined
|
|||
| whiteknight | good evening, #parrot | 22:39 | |
|
23:59
sivoais joined
|
|||