[Bug 297426] linuxulator: linprocfs /proc/self/exe host path breaks execve under Linux ABI
| Newsgroups | gmane.os.freebsd.bugs |
|---|---|
| Message-ID | <[email protected]/bugzilla/> |
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=297426 --- Comment #23 from Devin Teske <[email protected]> --- (In reply to Gleb Popov from comment #22) Your statement flies in the face of good people that have worked very hard to get rid of procfs in the base and we have operated without procfs for many releases without a problem. I have not had to mount procfs in over 10 years. Not for Firefox, not for anything in base, not for any natively compiled program. Then I attend one BSDCan and get the bug to start improving relations with the Framework folks and the Foundation -- working toward having Framework officially (on the record) state that they support FreeBSD as an OS and I go ahead and start polishing the OS to support all manner of use-cases for the laptop. I start my own window manager. I implement touchscreen gestures. I develop handicapable interfaces for disabled people. I test the platform as a portable gaming rig. A pro audio station. Then I get to the use case of using the laptop as a Set Top Box (STB) to replace my Apple TV and of course, I have to get things like Netflix going. I really didn't want to use native chromium because I don't trust Google (spent some time with tcpdump and I can verifiably prove that their mesh hardware doesn't play "nice" with things like PiHole; notably when DNS gets a response like 0.0.0.0 for a domain, their mesh routers will call home to the Google infra over DoH and resolve the domain on the auspice that the DNS gave a "bad" result, thus neutering the use of block lists). So, when it came to running Widevine to run sites like Netflix, I wanted to use a browser I could trust. I am aware that Brave is built atop chromium anyhow, but at least it's more trustworthy than Chromium in my opinion. So I started playing with linux-brave (the only way I am aware of running brave on FreeBSD) and that's the first time in over a decade that I've had to load procfs. So saying that it's unlikely we will ever get rid of procfs ignores the fact that I have been living procfs free for a decade plus and only when I had to load a single linux program did I have to bring it back onto my systems that have been happily procfs-free for a long time. -- You are receiving this mail because: You are the assignee for the bug.