We need SocketTransportService to work on socket process for socket I/O
To be more specific, the main thread of socket process should work like the current STS thread on chrome process.
Whiteboard: [necko-triaged] → [necko-triaged][SOCK-PROC-M1]
(In reply to Shih-Chiang Chien [:schien] (UTC+8) (use ni? plz) from comment #1) > To be more specific, the main thread of socket process should work like the > current STS thread on chrome process. I personally don't agree with this but. I would leave the main thread as is to be able to process OS events, if necessary. STS thread still can be a worker thread, as is now on the parent process. These days we are capable of delivering IPC directly to background threads, so I don't see a problem with that.
W/F, we are already successfully running an STS thread on the socket process as a worker thread and it's a good and sufficient architecture. Trying to poll sockets on the main thread is IMO an unnecessary overkill. There is no rational for the bug outlined here, nor I have found it in any notes from SC.
Status: NEW → RESOLVED
Last Resolved: 5 months ago
Resolution: --- → WONTFIX
You need to log in before you can comment on or make changes to this bug.