[Bug 269234] www/chromium: Sandboxing cleanup and basic Capsicum support for renderer processes
- Reply: bugzilla-noreply_a_freebsd.org: "maintainer-feedback requested: [Bug 269234] www/chromium: Sandboxing cleanup and basic Capsicum support for renderer processes"
- Reply: bugzilla-noreply_a_freebsd.org: "[Bug 269234] www/chromium: Sandboxing cleanup and basic Capsicum support for renderer processes"
- Reply: bugzilla-noreply_a_freebsd.org: "[Bug 269234] www/chromium: Sandboxing cleanup and basic Capsicum support for renderer processes"
- Go to: [ bottom of page ] [ top of archives ] [ this month ]
Date: Sun, 29 Jan 2023 19:39:28 UTC
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=269234 Bug ID: 269234 Summary: www/chromium: Sandboxing cleanup and basic Capsicum support for renderer processes Product: Ports & Packages Version: Latest Hardware: Any OS: Any Status: New Severity: Affects Only Me Priority: --- Component: Individual Port(s) Assignee: chromium@FreeBSD.org Reporter: sigsys@gmail.com Assignee: chromium@FreeBSD.org Flags: maintainer-feedback?(chromium@FreeBSD.org) Created attachment 239789 --> https://bugs.freebsd.org/bugzilla/attachment.cgi?id=239789&action=edit Chromium port basic Capsicum support The patchset already supports different backends for OpenBSD and FreeBSD sandboxing, but some files were still including the OpenBSD-specific headers and the preprocessor guards in the FreeBSD header were the same as the OpenBSD ones. So this patch clears that up. And it adds rudimentary Capsicum support for the renderer processes (which IIUC should be the most important processes to sandbox). It limits the stdio FDs (important since they could be TTYs), but does not limit any other FDs. And tbh, I do not know what kind of FDs they could be passed and how dangerous their ioctls could be. But it seems to work without issues (so far) and should be better than nothing. -- You are receiving this mail because: You are the assignee for the bug.