maemachinebroke/hakurei
1
0
Fork 0
forked from rosa/hakurei
Code Pull Requests Activity
983 Commits 7 Branches 12 Tags
8690419c2d577a27efe3cfcafc39c05b7646926b
Commit Graph
100 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
cat 8690419c2d hst: replace internal/app error 
This turns out to still be quite useful across internal/app and its relatives. Perhaps a cleaner replacement for baseError.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-09-15 01:44:43 +09:00
cat 1cdc6b4246 test/sandbox: create marker in /var/tmp 
This prepares the test suite for private TMPDIR.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-09-14 16:45:17 +09:00
cat 56aad8dc11 test/sandbox/tool: marker pathname from flag 
Since this is going to be placed in a shared directory, it needs to be unique to the identity. Instead of trying to figure out identity from mountinfo, just have the test script pass hardcoded values.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-09-14 15:57:41 +09:00
cat 83c4f8b767 test/sandbox: check extra writable paths 
This is not always obvious from mountinfo.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-09-14 15:12:51 +09:00
cat d0ddd71934 test/sandbox: bind /var/tmp writable 
This makes it possible to place markers with private tmpdir.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-09-14 14:59:53 +09:00
cat 70e02090f7 nix: use slightly less ambiguous type 
I had trouble getting Nix to merge json arrays properly, I am not sure that this helps.

At this point I have given up trying to understand Nix type system, and I am just trying to keep the Nix stuff going with extensive tests until it can be replaced by lkl for testing and planterette for general usage.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-09-14 14:45:14 +09:00
cat ca247b8037 internal/app: mount /dev/shm early 
This avoids covering /dev/shm mounts from hst.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-09-14 01:49:42 +09:00
cat 3f25c3f0af container: initialise cmd early 
This allows use of more cmd methods.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-09-13 20:01:33 +09:00
cat e271fa77aa nix: update flake lock 
Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-09-13 12:07:57 +09:00
cat f876043844 internal/hlog: remove error wrapping 
This was a stopgap solution that lasted for way too long. This finally removes it and prepares internal/app for some major changes.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-09-12 06:52:35 +09:00
cat 6265aea73a system: partial I inherit dispatcher 
This enables I struct methods to be checked.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-09-11 02:02:31 +09:00
cat c8a0effe90 system/wayland: use syscall dispatcher 
This enables wayland op methods to be instrumented.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-09-11 01:48:18 +09:00
cat 8df01b71d4 system: remove test package 
This prepares the Commit and Revert methods for testing via stub.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-09-10 23:50:22 +09:00
cat 985c4dd2fc system/xhost: wrap revert error correctly 
This otherwise creates a confusing error message on a revert failure.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-09-08 04:17:39 +09:00
cat da2b9c01ce system/tmpfiles: do not fail for smaller files 
The limit is meant to be an upper bound. Handle EOF and print verbose message for it instead of failing.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-09-08 03:22:10 +09:00
cat 323d132c40 system/mkdir: use syscall dispatcher 
This enables mkdir op methods to be instrumented.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-09-06 22:30:08 +09:00
cat 6cc2b406a4 system/link: use syscall dispatcher 
This enables hardlink op methods to be instrumented.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-09-06 19:47:58 +09:00
cat fcd0f2ede7 system/output: pass through LinkError 
This has similar formatting to PathError.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-09-06 17:41:06 +09:00
cat e68db7fbfc system: unexport Op implementations 
None of these are valid with their zero value, and the implementations assume they are created by the builder methods. They are by all means an implementation detail and exporting them makes no sense.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-09-06 16:16:03 +09:00
cat ac81cfbedc system/dbus: print incomplete string in buffer 
Not sure if this will ever be reached, but nice to have nonetheless.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-09-06 15:50:29 +09:00
cat 05db06c87b system/dbus: use syscall dispatcher 
This allows dbus op methods and builder to be instrumented.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-09-06 14:25:19 +09:00
cat e603b688ca system/dispatcher: expose test reporting to builder 
This is currently unused but useful for builders with errors.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-09-06 12:59:33 +09:00
cat a9def08533 system/dbus: drop proxy output beyond threshold 
This prevents xdg-dbus-proxy from running the priv process out of memory.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-09-06 02:56:21 +09:00
cat ecaf43358d system/dbus: create context in subtest 
This is causing a huge amount of spurious test failures due to the poor performance of the integration vm. This should finally put an end to the annoyance.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-09-05 05:15:40 +09:00
cat 197fa65b8f system/dbus: remove redundant proxy pairs 
This is left over from before dbus.Final. Remove them now as they serve no purpose.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-09-05 02:07:56 +09:00
cat e81a45e849 container/dispatcher: optional stub wait4 signal association 
This synchronises the wait4 return after the toplevel signal call in lowlastcap_signaled_cancel_forward_error.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-09-04 20:28:49 +09:00
cat 3920acf8c2 container/stub: remove function call in handleExit 
This gets inlined and does not cause problems usually but turns out -coverpkg uninlines it and breaks the recovery.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-09-04 19:39:12 +09:00
cat 19630a9593 container/dispatcher: remove wait4 test log 
Turns out the reporting methods are not safe for concurrent use, despite the claim in testing.T doc comment.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-09-04 05:30:57 +09:00
cat 4051577d6b container/stub: override goexit methods 
FailNow, Fatal, Fatalf, SkipNow, Skip and Skipf must be called from the goroutine created by the test.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-09-04 04:51:49 +09:00
cat ddfb865e2d system/dispatcher: wrap syscall helper functions 
This allows tests to stub all kernel behaviour, like in the container package.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-09-04 04:15:25 +09:00
cat 024d2ff782 system: improve tests of the I struct 
This cleans up for the test overhaul of this package.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-09-03 02:16:10 +09:00
cat 6f719bc3c1 system: update doc commands and remove mutex 
The mutex is not really doing anything, none of these methods make sense when called concurrently anyway. The copylocks analysis is still satisfied by the noCopy struct.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-09-02 04:54:34 +09:00
cat 1b5d20a39b container/dispatcher: stub.Call initialisation helper function 
This keeps composites analysis happy without making the test cases (too) bloated.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-09-02 04:44:08 +09:00
cat 49600a6f46 container/stub: export stub helpers 
These are very useful in many packages containing relatively large amount of code making calls to difficult or impossible to stub functions.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-31 23:11:25 +09:00
cat b489a3bba1 system/output: implement MessageError 
This error is also formatted differently based on state.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-31 13:51:21 +09:00
cat 780e3e5465 container/msg: optionally provide error messages 
This makes handling of fatal errors a lot less squirmy.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-31 11:57:59 +09:00
cat 712cfc06d7 container: wrap container init start errors 
This helps indicate the exact origin and nature of the error. This eliminates generic WrapErr from container.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-30 23:44:48 +09:00
cat f5abce9df5 system: wrap op errors 
This passes more information allowing for better error handling. This eliminates generic WrapErr from system.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-30 22:49:12 +09:00
cat ddb003e39b system/internal/xcb: refactor and clean up 
This package still does not deserve to be out of internal, but at least it is less haunting now. I am still not handling the xcb error though, the struct is almost entirely undocumented and the implementation is unreadable. Not even going to try.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-30 20:02:18 +09:00
cat b12c290f12 system/wayland: improve error descriptions 
A lot of these errors have very short and nondescript descriptions. These are only returned on incorrect API usage, but it makes sense to make them more descriptive anyway.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-30 16:51:40 +09:00
cat 0122593312 system/acl: wrap libacl errors in PathError 
This helps determine which libacl function the errno came from.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-30 13:19:15 +09:00
cat 6aa431d57a system/acl: update test log messages 
Most of these were never updated after UpdatePerm was renamed to Update.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-30 12:55:49 +09:00
cat 08eeafe817 container/mount: unwrap vfs decoder errors 
These are now handled by init. This eliminates generic WrapErr from mount and procPaths.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-29 22:15:05 +09:00
cat d7c7c69a13 container/dispatcher: check simple test errors via reflect 
Again, avoids the errors package concealing unexpected behaviours.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-29 22:12:21 +09:00
cat 50972096cd container/vfs: wrap decoder errors 
This passes line information and handles strconv errors so it reads better.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-29 21:51:31 +09:00
cat 905b9f9785 container/initoverlay: invalid argument type 
This eliminates generic WrapErr from overlay.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-29 02:56:56 +09:00
cat 1c7e634f09 container/dispatcher: check test errors via reflect 
Using the errors package might conceal some incorrect behaviour.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-29 02:35:24 +09:00
cat 8d472ebf2b container/inittmpfs: unwrap out of bounds error 
This eliminates generic WrapErr from tmpfs.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-29 02:15:48 +09:00
cat 4da6463135 container/init: unwrap path errors 
These are also now handled by init properly, so wrapping them in self is meaningless and unreachable.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-29 02:04:09 +09:00
cat eb3385d490 container/initsymlink: unwrap mount errors 
The mount function now wraps its own errors in a much more descriptive type with proper message formatting. Wrapping them no longer makes any sense.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-29 01:46:54 +09:00
cat b8669338da container/initsymlink: unwrap absolute error 
This is now handled properly by the init.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-29 01:43:11 +09:00
cat f24dd4ab8c container/init: handle unwrapped errors 
This is much cleaner from both the return statement and the error handling.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-29 01:37:13 +09:00
cat a462341a0a container: repeat and impossible state types 
This moves repeated Op errors and impossible internal state errors off of msg.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-29 01:12:02 +09:00
cat 84ad9791e2 container: wrap mount syscall errno 
This is the first step to deprecating the generalised error wrapping error message pattern.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-29 01:06:12 +09:00
cat b14690aa77 internal/app: remove seal interface 
This further cleans up the package for the restructure.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-28 01:07:51 +09:00
cat d0b6852cd7 internal/app: remove app interface 
It is very clear at this point that there will not be multiple implementations of App, and the internal/app package will never move out of internal due to hsu.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-28 00:54:44 +09:00
cat da0459aca1 internal/app: update doc comments 
A lot of these comments are quite old and have not been updated to reflect changes.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-28 00:45:57 +09:00
cat 1be8de6f5c internal/app: less strict username regex 
Use the default value of NAME_REGEX from adduser. Should not hurt compatibility while being less strict.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-28 00:22:55 +09:00
cat 0f41d96671 internal: move sysconf wrapper to app 
This should not be used and is not useful in other packages.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-28 00:04:58 +09:00
cat 92f510a647 cmd/hakurei/command: pd run dbus-verbose nil check 
This otherwise dereferences a nil pointer when dbus-verbose is set and either session or system bus are nil.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-09-06 00:09:25 +09:00
cat acb6931f3e app/seal: leave $DISPLAY as is on host abstract 
This helps work around faulty software that misinterprets unix: DISPLAY string.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-27 20:42:03 +09:00
cat 9d932d1039 release: 0.2.1 
Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-26 03:33:45 +09:00
cat 9bc8532d56 container/initdev: mount tmpfs on shm for ro dev 
Programs expect /dev/shm to be a writable tmpfs.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-26 03:27:07 +09:00
cat 07194c74cb release: 0.2.0 
Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-26 02:23:59 +09:00
cat 4cf694d2b3 hst: use hsu userid for share path suffix 
The privileged user is identifier to hakurei through its hsu userid. Using the kernel uid here makes little sense and is a leftover design choice from before hsu was implemented.

Closes #7.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-26 02:16:33 +09:00
cat c9facb746b hst/config: remove data field, rename dir to home 
There is no reason to give the home directory special treatment, as this behaviour can be quite confusing. The home directory also does not necessarily require its own mount point, it could be provided by a parent or simply be ephemeral.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-26 00:56:10 +09:00
cat 878b66022e hst/fsbind: optional ensure source 
This exposes the BindEnsure flag of BindMountOp.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-26 00:50:23 +09:00
cat 2e0a4795f6 container/initbind: optional ensure host directory 
This is used for ensuring persistent data directories specific to the container.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-26 00:44:45 +09:00
cat c328b584c0 hst/fslink: improve string representation 
This shortens the representation of most common use cases and generally improves readability.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-25 22:52:48 +09:00
cat 9585b35d5b hst/config: remove symlink field 
Closes #6.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-25 22:23:54 +09:00
cat 26cafe3e80 hst/fs: implement link fstype 
Symlinks do not require special treatment, and doing this allows placing links in order.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-25 21:57:38 +09:00
cat 125f150784 hst/fs: update doc comments 
The Type method no longer exists on the interface. Update doc comments to reflect that.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-25 21:11:39 +09:00
cat 0dcac55a0c hst/config: remove container etc field 
This no longer needs special treatment since it can be specified as a generic filesystem entry.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-25 19:24:33 +09:00
cat 6d202d73b4 hst/fsbind: optional autoetc behaviour 
This generalises the special field allowing any special behaviour to be matched from target.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-25 18:38:19 +09:00
cat 1438096339 hst/config: handle filesystem entry targeting root 
This allows any fstype supported by hst to be directly mounted on sysroot. A special case in internal/app applies the matching entry early and excludes it from path hiding.

Closes #5.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-25 17:52:57 +09:00
cat 059164d4fa hst/fsbind: optional autoroot behaviour 
This allows autoroot to be configured via Filesystem.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-25 17:44:12 +09:00
cat 8db906ee64 container/dispatcher: remove exit stub test log 
Turns out testing.T does not like being called in defer.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-25 17:33:35 +09:00
cat cedfceded5 container/autoroot: remove prefix field 
This field has been a noop for a long time. Remove it to prevent further confusion.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-25 03:39:20 +09:00
cat 33d2dcce1b container/initoverlay: internal bypass sysroot prefix 
This is for supporting overlay mounts for autoroot.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-25 02:42:22 +09:00
cat 2baa2d7063 container/init: measure init behaviour 
This used to be entirely done via integration tests, with almost no hope of error injection and coverage profile. These tests significantly increase confidence of future work in this area.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-24 04:52:32 +09:00
cat 0166833431 container/dispatcher: start goroutine in dispatcher 
This allows instrumentation of calls from goroutine without relying on finalizers.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-23 21:58:40 +09:00
cat b3da3da525 container/init: avoid multiple lastcap calls 
This reduces the size of []kexpect in the test suite.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-23 11:09:11 +09:00
cat 1b3902df78 container/dispatcher: instrument each goroutine individually 
Scheduler nondeterminism cannot be accounted for, so do this instead.

There should not be any performance penalty as these calls are optimised out for direct.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-23 11:07:16 +09:00
cat ea1e3ebae9 container/params: pass fd instead of file 
The file is very difficult to stub. Pass fd instead as it is the value that is actually useful.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-23 00:16:46 +09:00
cat 1c692bfb79 container/init: call lockOSThread through dispatcher 
This degrades test performance if not stubbed out.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-22 22:24:14 +09:00
cat 141a18999f container: move integration test helpers 
With the new instrumentation it is now possible to run init code outside integration tests.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-22 22:07:19 +09:00
cat afe23600d2 container/path: use syscall dispatcher 
This allows path and mount functions to be instrumented.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-22 22:02:21 +09:00
cat 09d2844981 container/init: wrap syscall helper functions 
This allows tests to stub all kernel behaviour, enabling measurement of all function call arguments and error injection.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-22 19:27:31 +09:00
cat d500d6e559 system/dbus: share host net ns for abstract 
Host abstract unix sockets are only accessible when also in the init net ns.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-21 21:55:23 +09:00
cat 5b73316ae0 container/syscall: doc comments from manpages 
These are pulled straight from the manpages.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-21 00:33:46 +09:00
cat 5d8a2199b6 container/init: op interface valid method 
Check ops early and eliminate duplicate checks.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-21 00:18:50 +09:00
cat a1482ecdd0 container/inittmpfs: check path equivalence by value 
Fixes regression introduced while integrating Absolute.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-20 20:17:28 +09:00
cat a07f9ed84c container/initsymlink: check path equivalence by value 
Fixes regression introduced while integrating Absolute.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-20 20:03:02 +09:00
cat 51304b03af container/initremount: check path equivalence by value 
Fixes regression introduced while integrating Absolute.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-20 19:55:51 +09:00
cat c6397b941f container/initproc: check path equivalence by value 
Fixes regression introduced while integrating Absolute.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-20 19:29:45 +09:00
cat d65e5f817a container/initplace: check path equivalence by value 
Fixes regression introduced while integrating Absolute.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-20 19:19:27 +09:00
cat 696e593898 container/initoverlay: check path equivalence by value 
Fixes regression introduced while integrating Absolute.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-20 17:33:15 +09:00
cat 97ab24feef container/init: use absolute compare method 
More checks are also added.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-20 17:14:36 +09:00
cat 31f0dd36df absolute: efficient equivalence check method 
This is more efficient and makes the call site cleaner.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-20 17:06:38 +09:00
cat 9aec2f46fe container/initdev: check path equivalence by value 
Fixes regression introduced while integrating Absolute.

Signed-off-by: Ophestra <cat@gensokyo.uk>
 2025-08-20 02:55:45 +09:00