Update %jinx PR#894
Closed
dozreg-toplud wants to merge 1690 commits into
Closed
Conversation
Added Check of +.feed to make sure its a cell, if its a cell and a moon we crash before running `+veri:dawn` Resolves #818
Addresses https://gist.github.com/Quodss/022a71ebcad581109ecfb9a8359d786f. Adds stricter assertions in arithmetic jets to get rid of jet mismatches from Nock behavior. Rewrites comparison jets to avoid allocations by mpz. Removes unused `drop.c` which had a bugged jet. Adds bloq size assertion in `+swp` jet. Adds loobean check in `+sort` jet. Removes erroneus shortcuts from `+cut`/`+met` jets. Adds atom check in `+cue` jet.
Resolves #
Addresses: > /Users/matt/src/urbit/vere/pkg/vere/db/lmdb.c:308:14: runtime error: load of misaligned address 0x007000017fbe for type 'c3_d' (aka 'unsigned long long'), which requires 8 byte alignment 0x007000017fbe: note: pointer points here 00 00 08 00 01 00 00 00 00 00 00 00 00 00 00 00 41 b0 26 0b 59 a2 13 59 62 26 6e 93 c5 61 17 76 ^ SUMMARY: UndefinedBehaviorSanitizer: undefined-behavior /Users/matt/src/urbit/vere/pkg/vere/db/lmdb.c:308:14
Removes glossary links from runtime error messages and removes/replaces broken docs.urbit.org, developers.urbit.org, and operators.urbit.org links from README files. Removes broken glossary links from double-boot warnings and replaces those with links to the breach instructions.
…f snoc, wasm3 update
…en done" This reverts commit 58f72bd.
All ships on `edge` have been experiencing mysterious crashes in the urth process ever since we merged `next/kelvin/409` to `develop`. It took us two weeks to run this down, but the issue seems to be twofold: 1. We were leaking all effects in urth, causing a surprisingly slow leak on all but the most busy ships. This PR fixes the leak, 2. Unifying equality on the home road was enabled in `vere-v3.5`. `~dozreg-toplud` noticed that unifying equality is unsafe in the presence of uncounted references, and it seems likely that some change in `vere-v4.0` caused this pattern to occur in urth. This problem was mostly showing up as `palloc: page out of heap` or other allocator errors caused by the use-after-free. We will most likely disable the unifying equality on the home road later (at least for urth).
All ships on `edge` have been experiencing mysterious crashes in the urth process ever since we merged `next/kelvin/409` to `develop`. It took us two weeks to run this down, but the issue seems to be twofold: 1. We were leaking all effects in urth, causing a surprisingly slow leak on all but the most busy ships. This PR fixes the leak. 2. Unifying equality on the home road was enabled in `vere-v3.5`. `~dozreg-toplud` noticed that unifying equality is unsafe in the presence of uncounted references, and it seems likely that some change in `vere-v4.0` caused this pattern to occur in urth. This problem was mostly showing up as `palloc: page out of heap` or other allocator errors caused by the use-after-free. We will most likely disable the unifying equality on the home road later (at least for urth).
Is stack alignment done in this way desirable?
This reverts commit b999150.
Contributor
Author
|
current diff with |
Contributor
Author
|
windows build fails |
Merged
pkova
added a commit
that referenced
this pull request
Nov 24, 2025
Implements [UIP-0124](https://github.com/urbit/UIPs/blob/main/UIPS/UIP-0124.md), replaces #648, supersedes #894. Moves Urbit time functions to `manage.c` and defines a per-road stack of absolute deadlines. Virtual timer is set or disarmed any time a new deadline is pushed onto the stack, popped off the stack or if a road is left with some deadlines on its stack. Defines a dynamic hint `%jinx`. If the product of the hint-formula is 0 or a cell, nothing happens, otherwise new deadline is pushed before the hinted formula is executed and popped once it is executed. #### Questions I had to mess with `noun/build.zig` and `jets/e/json_en.c` to make Windows build work, otherwise I was getting build errors. Why was that happening? Was there another, saner way to fix that? #### Warnings When playing around in a fakeship's dojo I saw a leak report once when I ran `|mass`. I then tried to reproduce it, including with `-Dmem-dbg` option and `-g` flag, to no avail.
Collaborator
|
Superseded by #910 |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
6 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Merges
developintosigilante/timeout, resolves name conflincts with%spinstack PR.This still does not work, nested %jinx hints cause crashes. But it at least builds.