1112827 - Cannot use base::ProcessId as a type in IPDL declarations

Status: RESOLVED FIXED

(Core :: IPC, defect)

Description

[(No longer employed by Mozilla) Aaron Klotz]

IPDL implicitly generates typedefs for base::ProcessId by default, but since it is not explicitly declared, using ProcessId in an IPDL file will fail. OTOH, explicitly declaring ProcessId via the |using| statement will generate two identical typedefs for base::ProcessId, causing C++ compilation errors.

I'd like to add the implicit typedefs to the set of explicit typedefs so that we can declare these types if needed without the duplicate typedefs in the generated code.

Comment 1

[(No longer employed by Mozilla) Aaron Klotz]

Attachment: Prevent IPDL from generating duplicate typedefs

Comment 2

[Ben Turner (not reading bugmail, use the needinfo flag!)]

Comment on attachment 8538147 [details] [diff] [review]
Prevent IPDL from generating duplicate typedefs

Review of attachment 8538147 [details] [diff] [review]:
-----------------------------------------------------------------

This looks ok to me, but would you mind attaching a diff of the generated code?

Comment 3

[Ben Turner (not reading bugmail, use the needinfo flag!)]

(In reply to ben turner [:bent] (use the needinfo? flag!) from comment #2)
> This looks ok to me, but would you mind attaching a diff of the generated
> code?

... for just one header!

Comment 4

[(No longer employed by Mozilla) Aaron Klotz]

Attachment: PContentChild.h diff

I had edited PContent.ipdl with a |using| declaration for base::ProcessId, which fails C++ compilation without the patch.

This is a diff of the generated PContentChild.h demonstrating the effect of the patch on the generated header.

Comment 5

[Ben Turner (not reading bugmail, use the needinfo flag!)]

Comment on attachment 8538147 [details] [diff] [review]
Prevent IPDL from generating duplicate typedefs

Review of attachment 8538147 [details] [diff] [review]:
-----------------------------------------------------------------

Great, thanks!

Comment 6

[(No longer employed by Mozilla) Aaron Klotz]

https://hg.mozilla.org/integration/mozilla-inbound/rev/f0d4cb5f206c

Comment 7

[Ryan VanderMeulen [:RyanVM]]

Backed out for Mulet permacrashing.
https://hg.mozilla.org/integration/mozilla-inbound/rev/2cfe0238dedc

https://treeherder.mozilla.org/logviewer.html#?job_id=4769926&repo=mozilla-inbound

Comment 8

[(No longer employed by Mozilla) Aaron Klotz]

Mulet when green again before the above checkin was backed out.
The crashes were not IPC related, so I'm not sure what was going on there.
Discussed with RyanVM, trying to reland:
https://hg.mozilla.org/integration/mozilla-inbound/rev/841573344c3f

Comment 9

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/841573344c3f