| Commit message (Collapse) | Author | Age | Files | Lines |
| ... | |
| |
|
|
|
|
|
|
| |
When we forward an email that matched an inbox regexp, make sure the
author and the original report email are Cc'd. Otherwise the automation
cannot determine the original bug report.
Improve tests.
|
| |
|
|
|
| |
The impact score is deducted from the title.
Impact is max(known_titles).
|
| | |
|
| |
|
|
|
|
| |
For some specified inboxes, forward the emails that contain syz
commands.
Add tests to verify the behavior.
|
| |
|
|
|
| |
The API method can be used to send a raw email on behalf of the GAE
instance.
|
| | |
|
| |
|
|
|
| |
GKE streams all stdout as severity=ERROR messages, which makes it too
noisy to be useful on the Admin page of syzbot.
|
| |
|
|
| |
It is a bit shorter.
|
| |
|
|
| |
Get parameters are the internal interface.
|
| |
|
|
|
| |
1. Refactor handleHeatmap.
2. Introduce function options. Build them from http.Request.
|
| |
|
|
|
| |
We're currently using alternative solution:
https://cloud.google.com/datastore/docs/backups
|
| |
|
|
| |
The lines about folders don't look actionable.
|
| |
|
|
| |
getSpannerClient returns prod client as a default.
|
| |
|
|
|
| |
1. Set coverage db.
2. Use appUrl instead of r[.URL].Host.
|
| |
|
|
| |
We periodically send coverage reports for the regressions detection.
|
| |
|
|
|
|
|
|
|
| |
We are using incorrect *Sort functions when using some sortTable functions.
Also, this is causing the current syz repro sort to not work.
Thereforce we need to modify it to use the correct *Sort functions
and remove unnecessary tab spaces to pass the intended text to
the *Sort functions.
|
| |
|
|
|
|
|
| |
To make them accessible for non-admins, we need explicit checks in
access.go.
Extend TestAccess to cover fsck logs.
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
cover.Format controls the resulting view.
It allows to:
1. Remove records with 0 covered blocks.
2. Remove lines with low (user defined) coverage.
3. Order records by the covered lines drop value.
The corresponding GET parameters are:
1. Implicitly enabled for onlyUnique records.
2. min-cover-lines-drop=%d
3. order-by-cover-lines-drop=1
|
| |
|
|
|
| |
It allows to see records older than 12 periods.
And it allows to specify the target for coverage regression analysis.
|
| |
|
|
|
|
|
|
| |
Currently it's only possible to understand total number of uncovered
blocks in a function (implicitly defined by Instrumented field).
This does not allow to render coverage data, nor do detailed analysis
on line level. Export detailed info about both covered and uncovered blocks.
This allows to e.g. calculate coverage percent for kernel interfaces.
|
| |
|
|
|
|
|
| |
The existing pkg/html package transitively takes too many dependencies,
which complicates the reuse of some of its methods.
Split off a pkg/html/urlutil package.
|
| |
|
|
|
|
| |
It was previously disabled because line coverage required data from BigQuery.
It was too expensive.
Coverage numbers are available in the Spanner now.
|
| |
|
|
|
|
|
| |
8 cores, 12 gig machines can't process quarter long aggregation for some reason.
The process requires only 1 gig and do the stream processing.
It is something else I don't fully understand.
Local experiments show high memory consumptions by the git processes.
|
| |
|
|
|
| |
Quarter long aggregations are killed.
OpsAgent may help to understand why.
|
| | |
|
| |
|
|
|
| |
We changes the machine type to c3-highmem-8.
But forgot to double the CPU resources.
|
| |
|
|
|
|
| |
Some time ago we added the per-manager statistics.
Histogram was broken since that moment.
Closes #5867.
|
| |
|
|
| |
Coverage merging often take more than 1 hour.
|
| |
|
|
| |
New code will be limited to max 7 function params.
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
Context: #5829.
Let's not pretend that the revoked reproducer never existed and still
report it. It will avoid unexpected side-effects for the higher-level
logic.
There may be better ways to resolve the bug, but let's first just get it
fixed to prevent syzbot from spamming the mailing lists.
Add a test to verify the new behavior.
|
| |
|
|
|
| |
The export is quite big but is generated fast.
Every line is a valid json object representing the single program coverage.
|
| | |
|
| |
|
|
|
| |
It currently generates &subsystem=null&manager=null&unique-only=null.
I'd like to see &subsystem=&manager=&unique-only=
|
| |
|
|
|
| |
Gerrit servers return base64 encoded file content.
To get the raw data, &format=TEXT request param is needed.
|
| | |
|
| | |
|
| |
|
|
| |
Wrong closing tags etc.
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
| |
It looks cheaper than appengine.NewContext().
|
| |
|
|
| |
It is available in preview since Nov. 19 2024.
|
| |
|
|
| |
It will simplify the json API links.
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
| |
1. Init coveragedb client once and propagate it through context to enable mocking.
2. Always init coverage handlers. It simplifies testing.
3. Read webGit and coveragedb client from ctx to make it mockable.
4. Use int for file line number and int64 for merged coverage.
5. Add tests.
|
| |
|
|
|
|
|
|
|
|
|
|
| |
1. Make heatmap testable, move out the spanner client instantiation.
2. Generate spannerdb.ReadOnlyTransaction mocks.
3. Generate spannerdb.RowIterator mocks.
4. Generate spannerdb.Row mocks.
5. Prepare spannerdb fixture.
6. Fixed html control name + value.
7. Added multiple tests.
8. Show line coverage from selected manager.
9. Propagate coverage url params to file coverage url.
|
| |
|
|
| |
They are shorter, more readable, and don't require temp vars.
|
| | |
|