1078847 - GenerateAsymmetricKeyTask fails if no private key usages are passed

Status: RESOLVED FIXED

(Core :: DOM: Security, defect)

Description

[Tim Taubert [:ttaubert] (inactive)]

Small typo from bug 1037892. Also I think we should modify the condition to allow generating a key pair where only the public *or* the private key has any usages. This behavior isn't defined in the spec afaict, should it be added?

Blocking DH as this makes a few tests fail there.

Comment 1

[Tim Taubert [:ttaubert] (inactive)]

Attachment: 0007-Bug-1078847-GenerateAsymmetricKeyTask-should-allow-g.patch

Comment 2

[Tim Taubert [:ttaubert] (inactive)]

(In reply to Tim Taubert [:ttaubert] from comment #0)
> This behavior isn't defined in the spec afaict, should it be added?

I was referring to behavior of bailing out and making .generateKey() fail if there are no valid usages for any of the keys in the key pair.

Comment 3

[Tim Taubert [:ttaubert] (inactive)]

Try:

https://treeherder.mozilla.org/ui/#/jobs?repo=try&revision=ee02b0672bc6

Comment 4

[Tim Taubert [:ttaubert] (inactive)]

https://hg.mozilla.org/integration/mozilla-inbound/rev/5c095b4e178c

Comment 5

[Carsten Book [:Tomcat]]

https://hg.mozilla.org/mozilla-central/rev/5c095b4e178c

Comment 6

[Tim Taubert [:ttaubert] (inactive)]

Comment on attachment 8500736 [details] [diff] [review]
0007-Bug-1078847-GenerateAsymmetricKeyTask-should-allow-g.patch

Approval Request Comment
[Feature/regressing bug #]: bug 1037892
[User impact if declined]: Asymmetric key generation fails when the private key h as no usage.
[Describe test coverage new/current, TBPL]: includes a test that works fine on m-c
[Risks and why]: Low risk.
[String/UUID change made/needed]: None.

We need this for bug 1034856.

Comment 7

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/releases/mozilla-aurora/rev/a0b472565f9e