[llvm-dev] Experiment on how to improve our temporary file handing.

UE US via llvm-dev llvm-dev at lists.llvm.org
Fri Nov 10 17:04:55 PST 2017

NTFS journals writes and keeps a fairly large buffer so it'd have to be a
very well timed, fast power loss.  Software should do what it could to
alleviate this (especially in long runtime cases) but a cheap UPS or some
form of hardware battery backed hardware RAID is a better solution if the
data is that important, IMO.

As an aside, Windows 10 with enough RAM can run for at least 3 minutes with
the system drive cable pulled.  I did it accidentally once while
hot-swapping another drive that snagged it.  No data corruption, NTFS
journals in memory finished the write when I noticed my clumsy and and
plugged it back in.  Just anecdotal but funny to me.


On Fri, Nov 10, 2017 at 3:35 PM, Rafael Avila de Espindola via llvm-dev <
llvm-dev at lists.llvm.org> wrote:

> Mark Kettenis <mark.kettenis at xs4all.nl> writes:
> > I'm actually surprised Linux allows this as there are some serious
> > security implications as this allows programs to create an entry in
> > the filesystem for file descriptors passed over a socket.
> BTW, would you mind expanding on what is the security problem of that?
> Thanks,
> Rafael
> _______________________________________________
> LLVM Developers mailing list
> llvm-dev at lists.llvm.org
> http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20171110/27c7ed55/attachment.html>

More information about the llvm-dev mailing list