linux-kernel - Re: [PATCH v4 10/11] fs/9p: writeback mode fixes

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <Y/E1MCTVslQzozEb@codewreck.org>
Date:   Sun, 19 Feb 2023 05:29:36 +0900
From:   asmadeus@...ewreck.org
To:     Eric Van Hensbergen <ericvh@...il.com>
Cc:     Eric Van Hensbergen <ericvh@...nel.org>,
        v9fs-developer@...ts.sourceforge.net, rminnich@...il.com,
        lucho@...kov.net, linux-kernel@...r.kernel.org,
        linux-fsdevel@...r.kernel.org, linux_oss@...debyte.com
Subject: Re: [PATCH v4 10/11] fs/9p: writeback mode fixes

Eric Van Hensbergen wrote on Sat, Feb 18, 2023 at 10:40:27AM -0600:
> This is stupidity on my part, but can you send me your setup for
> fscache so I can test the way you are testing it?  It is the one thing
> I still haven't incorporated into my test matrix so definitely a blind
> spot I appreciate you exposing.

I mounted with fscache but I didn't actually have a backend running when
testing, so there's really not setup involved -- just qemu's
'-virtfs local,path=/tmp/linux-test,mount_tag=tmp,security_model=mapped-file'
and
'mount -t 9p -o debug=1,cache=fscache,trans=virtio tmp /mnt'

In that case, fscache basically acts like loose, except it also does all
its cookies stuff so it's sort of useful to test anyway.


I also have a setup that runs cachefilesd on top but it wasn't involved
here; for completeness if you want to test:
- add an extra disk to qemu
qemu-img create -f qcow2 /tmp/disk 1G
'-device virtio-blk-pci,drive=hd0 -drive if=none,file=/tmp/disk,snapshot=on,id=hd0'
- just runs cachefilesd (ignore hardcoded nix paths, that's my easy way out
of running straight out of initrd without making it huge):
---
if [ -e /dev/vda ]; then
        /run/current-system/sw/bin/mkfs.ext4 -q /dev/vda
        mkdir /cache /var/run
        mount /dev/vda /cache
        echo -e 'dir /cache\ntag cache' > /etc/cachefilesd.conf
        echo 'modprobe cachefiles'
        echo '/nix/store/4vy0z1ygfidfmbzaxblkmzv7j0irhhwc-cachefilesd-0.10.10/bin/cachefilesd -s -d > /cache/log 2>&1'
fi
---
And that's basically it; you should see files poping up in /dev/vda when
you do reads (iirc writes weren't cached through last I looked)
-- 
Dominique