https://searchfox.org/mozilla-central/rev/588814f2edddf0e132d77d326ddae50911e8bad1/docshell/base/nsDocShell.cpp#8520

The issue here is that the docshell tree (same-type) is walked up, stopping and returning true at the first ancestor docshell which has a principal which is subsumed by the subject principal.

I suspect this information is unlikely to be shared cross process. Do we have some mechanism to query this via IPC already, or does an implementation need to be added?

(In reply to Kannan Vijayan [:djvj] from comment #0)

The issue here is that the docshell tree (same-type) is walked up, stopping and returning true at the first ancestor docshell which has a principal which is subsumed by the subject principal.

This should be straightforward enough to fix. the subject principal will never subsume the principal of an iframe in a different process, so we can walk up the BrowsingContext tree, and assume that the subsume check fails for BCs with missing DocShells.

Kannan says replacing nsIDocShellTreeItem calls should block enabling Fission in Nightly (M6).

There's a r+ patch which didn't land and no activity in this bug for 2 weeks.
:djvj, could you have a look please?
For more information, please visit auto_nag documentation.

The patch was r+ but I was running into testing errors that I could not resolve before I went on PTO. I'll take a look at this when I get back to work. Keeping needinfo.

I can't really get back to this - I have one major and another minor thing on my plate currently which I should focus on. Anyone should feel free to take/work on this.

Auditing whether this use of nsIDocShellTreeItem breaks when Fission is enabled blocks Fission Nightly.

https://hg.mozilla.org/mozilla-central/rev/2496ca6ee332