| 15 Mar 2026 |
 emily | given the UNSAFE_TODO I'm guessing this is some neglected corner of the browser | 06:31:33 |
| emily | [0] is precisely GNU extension right | 06:31:42 |
 Vladimír Čunát | Not in standard C | 06:31:43 |
 dramforever | dang | 06:31:49 |
| emily | isn't there an __attribute__((align(…))) or something. maybe something like that got standardized | 06:32:10 |
| Vladimír Čunát | Yes. | 06:32:24 |
| dramforever | oh, lol yeah | 06:32:32 |
| dramforever | https://en.cppreference.com/w/c/language/alignas.html | 06:32:43 |
| dramforever | although still | 06:32:55 |
| dramforever | i'm not sure if that's what they even wanted | 06:33:04 |
| dramforever | i'm [13677/56297] into building chromium-unwrapped and my laptop hasn't exploded yet | 06:33:32 |
| dramforever | i think that's a good sign that i will be able to test this after all | 06:35:19 |
| Vladimír Čunát | In standard C this certainly works to ensure what they apparently wanted. | 06:36:16 |
| Vladimír Čunát | (I do use it.) | 06:36:22 |
| Vladimír Čunát | C++ does have it as well, so probable there as well. | 06:36:42 |
| dramforever | this definitely aligns | 06:36:44 |
| dramforever | but the old code obviously also just didn't align | 06:36:52 |
| dramforever | so maybe they don't actually mean like, type alignment | 06:37:04 |
| dramforever | or maybe they just messed it up | 06:37:20 |
| emily | I think they might mean "pad it out" | 06:46:06 |
| emily | but it's definitely baffling with a flexible array | 06:46:13 |
| emily | maybe you can git blame it. how long can that take on the Chromium repo | 06:46:55 |
| dramforever | the answer is "seconds" btw | 06:56:10 |
| dramforever | there's a "Blame" button | 06:56:14 |
| dramforever | which leads to https://codereview.chromium.org/2490303002/patch/260001/270006 | 06:56:17 |
| dramforever | still no idea what "alignment" means, but it looks like
- chromium intends to at least use
[1] for FAM
- they don't enforce it
| 06:57:29 |
| dramforever | ugh i don't want to figure out how to report chromium bug | 06:58:03 |
| emily |
It's definitely a benefit
| 07:03:42 |
| emily |
base/metrics/persistent_histogram_allocator.cc:247: char name[sizeof(uint64_t)];
Please add a comment about this, so it's clear why it's like this.
| 07:04:13 |
| emily |
commit-bot: I haz the power
4 years, 1 month ago (2016-11-22 02:17:58 UTC) #80
CQ is trying da patch.
| 07:05:10 |
