| Commit message (Collapse) | Author | Age | Files | Lines |
| | |
|
| |
|
|
|
| |
The current time sorting produces some random order (text is not time).
Let's sort as text, it should produce at least some order (e.g. group PATCH together).
|
| | |
|
| | |
|
| |
|
|
|
| |
Print OK only for non-failed and non-crashed jobs.
Print links to independent assets independently.
|
| | |
|
| |
|
|
| |
Useful to estimate overall syzbot performance.
|
| |
|
|
|
| |
This will help us understand how exactly how we have arrived at the
reproducer.
|
| | |
|
| |
|
|
|
|
|
| |
Currently we print somewhat verbose "💬 0 [3d18h]" when there are
no discussions, the date repeats the reporting date.
It' makes it hard to visually see when there are conversations
and when there are none. Print nothing when there are no discussions.
|
| |
|
|
| |
Display last 500 successful builds for every manager.
|
| | |
|
| |
|
|
|
|
|
| |
In some cases restarting the failed bisection does not make sense,
because we know that the result will be the same.
Let admins choose between invalidating the job and restarting it.
|
| |
|
|
|
|
|
| |
Add a per-namespace page that contains:
1) The list of commits to backport from the trees fuzzed in that
namespace.
2) The list of commits to backport into the namespace.
|
| |
|
|
|
|
|
| |
Currently we only display the link for latest fix/cause bisections on
the bug page.
Also display it inside the collapsible job lists.
|
| |
|
|
| |
Distinguish them from normal fix bisections.
|
| | |
|
| | |
|
| |
|
|
|
| |
Job log can sched more light on what was exactly happening. It's not
really correct to omit it.
|
| |
|
|
|
| |
Currently we only show running/pending/recent jobs. Let admins also see
the list of the latest cause/fix bisections.
|
| |
|
|
|
|
|
|
| |
The dropdown is the only UI element that lets users exit the admin
page, so removing it from the admin page is not advisable. However
since the admin page is not a real namespace, the dropdown will display
the name of the first namespace, when visiting the admin page. This
confuses new users.
|
| |
|
|
|
|
|
|
|
|
| |
Changes to our rootfs, compilers or bisection logic regularly cause
regressions in our bisection accuracy. Retrying them currently entails
fiddling with the GCP datastore directly or mass deleting all failed
bisections.
This change will allow us to retry specific bisections with a single
click.
|
| |
|
|
|
| |
Display them as a collapsible block on the bug info page. Use colors to
distinguish results.
|
| |
|
|
|
|
| |
For each label, allow only one value to be specified.
At the same time, allow multiple different labels (subsystem, origin,
prio, etc) be specified together.
|
| |
|
|
|
|
| |
For now let's just keep them in the patch testing requests list and give
them a different name. Later we'll group the information in a much more
explicit way.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Let bug labels come in three flavours:
1) Bug labels with multiple values (e.g. `subsystems`).
2) Bug labels with only one value (e.g. `prio`).
3) Flags.
Let users configure bug labels via email by issuing the following
commands:
#syz set subsystems: abc, def
#syz set no-reminders
#syz unset no-reminders
Also let users set tags for invididual bugs in reported bug lists:
#syz set <1> some-tag
|
| | |
|
| |
|
|
|
| |
Instead of showing tons of extra information to the user, only show the
collapsed titles with counts and let user toggle the sections they need.
|
| |
|
|
|
|
|
|
|
| |
Collect the information about the discussions under bug reports and
patches that target the reported bugs.
For now only Lore discussions are supported.
Display the discussion list on the bug info page.
|
| |
|
|
|
|
|
| |
Manager name does not tell much to end users.
They are also long and all prefixed with ci[2]-upstream-,
this takes lots of space in the table.
Move it to the end of the table.
|
| |
|
|
|
| |
In some cases we don't have the bisection log, but only the error log.
Adjust the template.
|
| |
|
|
|
| |
Count the number of bugs without any subsystem and let users filter bugs
by this criteria.
|
| | |
|
| | |
|
| |
|
|
| |
Take the counts from the cache, include links to the filtered bug views.
|
| |
|
|
| |
For each subsystem, collect the number of open, fixed and invalid bugs.
|
| |
|
|
|
|
|
| |
Currently it's non obvious whether and what filters are applied to a bug
list.
Display the list to the user and enable them to drop individual filters.
|
| |
|
|
|
| |
To generate filtering URLs, use the ability of context.Context to carry
on variable values.
|
| |
|
|
|
| |
Currently we only underline the difference in emails, do it also in the
web version.
|
| | |
|
| |
|
|
|
| |
The `Instances` header currently leads to nowhere. Let's add there a
link to the tested repos.
|
| |
|
|
| |
Currently the User field is empty, which makes things confusing.
|
| |
|
|
|
|
|
| |
This reflects the situation more adequately than just the recent jobs.
There are times when failed jobs are retried or manually restarted and
the current page just does not show such jobs.
|
| |
|
|
|
|
|
|
|
|
|
| |
Commit 2f3b44ea4 ("dashboard: visualize stale repros") inadvertently
added a single space in the "<td class=repro>" tag. This will break
tooling which relies on HTML scraping and pattern matching. This
patch ensures the formatting is kept consistent whilst honouring the
author's true intentions.
Fixes: 2f3b44ea4 ("dashboard: visualize stale repros")
Signed-off-by: Lee Jones <joneslee@google.com>
|
| |
|
|
|
| |
When rendering a crash list, strike stale repros with a line (like the
<s></s> tag does).
|
| |
|
|
| |
Right now it's sent by email, but still not displayed on the web page.
|
| |
|
|
|
| |
They might confuse the users who search the page.
Show them in the "Fixed" tab instead.
|
| |
|
|
| |
Very handy to assess if it's the right/wrong commit.
|
| |
|
|
| |
Show graph pages to users who have access to the bugs themselves.
|
| |
|
|
|
|
|
| |
It's useful to be able to look at the current config for a particular manager.
Currently it's doable but somewhat ugly (find a recently reported crash and
look at its config).
Show kernel config for each manager directly.
|