|
08:53
sena_kun joined
|
|||
| Geth | MoarVM/detailed_moarvm_file_dump: 6 commits pushed by (Timo Paulssen)++ | 10:13 | |
| timo | whew, we really do have a boatload of stringified numbers in core setting's string heap ... | 10:21 | |
| lizmat | timo: aren't values 0..255 special cased? or are you talking about bigger numbers ? | 11:24 | |
| they could be nqp::const flag values ? | |||
| timo | nah, i think the numbers i'm seeing are all cuuids | 12:06 | |
| they go from the lowest numbers up to the tens of thousands or maybe hundreds of thousands? not quite sure. | |||
| lizmat | but why as string ? | 12:08 | |
| ah, remember now: they're not always strings: I::Perl5 uses strings, if I remember correctly | 12:09 | ||
| *integers | |||
| timo | could be | 12:13 | |
| for the medium future, we may want to look into storing stringified integers using less space | 12:14 | ||
| but then we'd spend a little time on first access to the string for generating a stringified integer | 12:15 | ||
| i wonder what we store a closure's name for. shouldn't it be the same as the static code object's? | 12:23 | ||
| lizmat | jnhtn might know :-) | 12:25 | |
| timo | > String heap has 21966 only-digits strings for a total of 136132 bytes | 12:28 | |
| m: say 136132 * 100 / 0x86f50, " percent of bytes are for digits"; say 21966 * 100 / 38616, " percent of strings are digit-strings" | 12:29 | ||
| camelia | 24.626617 percent of bytes are for digits 56.883157 percent of strings are digit-strings |
||
| timo | m: say "but that's only ", 136132 * 100 / 0x1bff5c0, " percent of the total file." | 12:30 | |
| camelia | but that's only 0.46370427 percent of the total file. | ||
| lizmat | and yet another Rakudo Weekly News hits the Net: rakudoweekly.blog/2024/10/21/2024-...ent-alert/ | 16:55 | |
|
21:18
coleman left
21:21
coleman joined
22:41
sena_kun left
|
|||