-
Notifications
You must be signed in to change notification settings - Fork 1.2k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
syz-cluster: initial code #5620
base: master
Are you sure you want to change the base?
Conversation
dbf79ae
to
0a8c2df
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The first patch is not questionable. Let's review/merge it separately.
var patchSubjectRe = regexp.MustCompile(`\[(?:(?:rfc|resend)\s+)*patch`) | ||
type PatchSubject struct { | ||
Title string | ||
Tags []string // Sometimes there's e.g. "net" or "next-next" in the subject. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
These tags are parsed, but lost later. What's the plan here?
Do we want to not test "RFC" patches? Then these should set Corrupted?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
As mentioned in the comment, the developers sometimes explicitly say that the the patch it to be applied to the specific tree. This information is not used in this initial PR, but will be important for the further series triage implementations.
@@ -20,9 +22,29 @@ type Thread struct { | |||
Messages []*email.Email | |||
} | |||
|
|||
type Series struct { | |||
Subject string |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Does it happen that the subject changes between versions?
If so, do we need to also extract in-reply to message-id, which may be the previous version?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
AFAIK the series are usually not sent in-reply-to the previous ones, so I'd expect the title to be fairly stable. Otherwise there's no chance to relate them.
Dir string | ||
Precious bool | ||
Sandbox bool | ||
Env []string |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Can we unexport some of these fields if they are used only within this package?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I've unexported Precious
, the others are used outside (at least so far).
0a8c2df
to
3ef5a48
Compare
Support filtering by the commit date.
3ef5a48
to
2a65760
Compare
Refactor the code to make it more reusable. Add a method to extract specifically the list of new patch series.
2a65760
to
770edae
Compare
It will let other parts of the system to use the git-specific functionality.
ae3f62b
to
0129842
Compare
The basic code of a K8S-based cluster that: * Aggregates new LKML patch series. * Determines the kernel trees to apply them to. * Builds the basic and the patched kernel. * Displays the results on a web dashboard. This is a very rudimentary version with a lot of TODOs that provides a skeleton for further work. The project makes use of Argo workflows and Spanner DB. Bootstrap is used for the web interface. Overall structure: * syz-cluster/dashboard: a web dashboard listing patch series and their test results. * syz-cluster/series-tracker: polls Lore archives and submits the new patch series to the DB. * syz-cluster/controller: schedules workflows and provides API for them. * syz-cluster/kernel-disk: a cron job that keeps a kernel checkout up to date. * syz-cluster/workflow/*: workflow steps. For the DB structure see syz-cluster/pkg/db/migrations/*.
0129842
to
14daace
Compare
FTR:
|
The basic code of a k8s-based cluster that:
This is a very rudimentary version with a lot of TODOs that provides a skeleton for further work.
The project makes use of Argo workflows and Spanner DB.
Bootstrap is used for the web interface.
Overall structure:
syz-cluster/dashboard
: a web dashboard listing patch series and their test results.syz-cluster/series-tracker
: polls Lore archives and submits the new patch series to the DB.syz-cluster/controller
: schedules workflows and provides API for them.syz-cluster/kernel-disk
: a cron job that keeps a kernel checkout up to date.syz-cluster/workflow/*
: workflow steps.For the DB structure see
syz-cluster/pkg/db/migrations/*
.TODO (for this PR):
go.mod
bumped to1.23.1
syz-cluster
tests on GitHub CI (we'll need at least some local Spanner emulator binary).Questions/thoughts out loud:
1. What to do with the vendor folder
Ourvendor/
folder is quite big, and this PR adds even more modules on top of that. Do we still want to keep that code in our repostory? Is it possible to only keep the modules needed by other components but notsyz-cluster
(since I dogo mod download
in the Docker containers anyway).Filed #5645
2. Some pkg/ packages are too eager to depend on prog/ and sys/
It has made some Dockerfiles more complicated that they could have been.
Should we strive to remove these dependencies?