10.04.2018 08:35, Oleksandr Natalenko wrote:
- does it reproduce _without_ hardened usercopy? (I would assume yes,
but you'd just not get any warning until the hangs started.) If it
does reproduce without hardened usercopy, then a new bisect run could
narrow the search even more.
Looks like it cannot be disabled via kernel cmdline, so I have to
re-compile the kernel, right? I can certainly do that anyway.
Okay, I've recompiled the kernel without hardened usercopy:
[root@archlinux ~]# zgrep USERCOPY /proc/config.gz
# CONFIG_HARDENED_USERCOPY is not set
and I cannot reproduce the issue anymore. I/O doesn't hang regardless of
how long I hammer it.
Eeeh? Maybe, this is a matter of some cleanup code path once the
warn/bug condition is hit with hardening enabled? I'm just guessing here
Will work towards checking Linus' master branch now…