|
00:51
MasterDuke joined
|
|||
| Geth | MoarVM/main: 2a0dc7f171 | (Timo Paulssen)++ | 2 files latin1.c and ascii.c can return str_consts.empty for 0-length string except the very first time when creating the 0-length string constant |
00:58 | |
| MoarVM/main: f2c57c0e35 | MasterDuke17++ (committed using GitHub Web editor) | 2 files Merge pull request #1827 from MoarVM/better_empty_strings |
|||
| MoarVM/main: abab265d06 | (Timo Paulssen)++ | src/spesh/optimize.c When guarding conc or type and type is known, guard just conc/type The difference in moar bytecode is very small, but in JITted code it's a bit bigger since the non-"just" variant also jits a spesh slot access first. |
|||
| MoarVM/main: 2b5dcaee98 | MasterDuke17++ (committed using GitHub Web editor) | src/spesh/optimize.c Merge pull request #1826 from MoarVM/turn_guard_into_just_if_possible |
|||
|
02:49
MasterDuke left
06:49
sena_kun joined
07:24
sena_kun left
|
|||
| timo | what's not awesome is that when we merge coolroot into main, everybody will have to re-run Configure so that config.h is regenerated from config.h.in | 16:58 | |
| i wonder if there's a way to get a little output to people when that goes wrong so they know what's going on immediately | 17:00 | ||
| lizmat | Configure on NQP ? | 17:04 | |
| timo | no Configure on moar | 17:05 | |
| i guess you need to do that anyway whenever we bump versions since otherwise every step in the make process will yell at you? | 17:06 | ||
| ab5tract | Yeah, I think that one kicks in automatically on version bumps | 17:10 | |
| I’d have to check to be sure tho | 17:11 | ||
| Then again, I make clean every time.. | 17:12 | ||
| timo | make clean in moar isn't enough i think? that leaves config.h and config.c undeleted, so that you can "make clean; make all" without trouble | 17:16 | |
| "make distclean" will remove them though, and force you to reconfigure | |||
| Geth | MoarVM/coolroot: f2dcdcae00 | (Timo Paulssen)++ | 98 files A re-imagining of the MVMROOT macros where we don't have to put the code block inside the macro's argument list, which means the whole block is no longer considered a single statement by tools like gdb, profilers, and so on. |
17:17 | |
| ab5tract | timo: I meant realclean indeed, but I trigger that in the rakudo directory | 17:25 | |
| timo | ok, you rely on --build-moar of Configure.pl then? | 17:27 | |
| ab5tract | Indeed | 17:32 | |
| timo | i'll quickly look into the possibility of making MVMROOT a var arg macro so we don't need the numbers any more either | 17:37 | |
| never mind, that's a terrible idea | 17:40 | ||
| Geth | MoarVM/coolroot: e5e9553e87 | (Timo Paulssen)++ | 98 files A re-imagining of the MVMROOT macros where we don't have to put the code block inside the macro's argument list, which means the whole block is no longer considered a single statement by tools like gdb, profilers, and so on. |
17:41 | |
| timo | i forgot a change in one file somehow | 17:43 | |
| i can't find a "re-run failed" button for azure pipelines, and i clicked "re-run" on a few checks on the github side of things and it doesn't seem to be doing anything even though it popped up a "successfully requested X to be re run" or something | 17:58 | ||
| it's impressive how slow msvc is at compiling moar | 18:04 | ||
| oh i think i found the button | 18:09 | ||
|
18:49
sena_kun joined
|
|||
| Geth | MoarVM/main: ac82e446f1 | (Timo Paulssen)++ (committed by timo) | 98 files A re-imagining of the MVMROOT macros where we don't have to put the code block inside the macro's argument list, which means the whole block is no longer considered a single statement by tools like gdb, profilers, and so on. |
19:13 | |
| timo | very glad to have this out of the door | 19:14 | |
| lizmat | ok, so now time to bump Rakudo, right? | 19:15 | |
| timo | not strictly necessary, but of course it'd help to get testing in it | 19:17 | |
| lizmat | right, on it | ||
| timo | i did some research to make sure the compiler spits out the same thing for old mvmroot and new mvmroot | 21:50 | |
|
22:17
sena_kun left
|
|||