linux-kernel - nvme, allocation failures, and aborted commands

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-next>] [day] [month] [year] [list]

Date:   Fri, 03 Mar 2017 08:27:40 -0700
From:   Thomas Fjellstrom <thomas@...llstrom.ca>
To:     linux-kernel <linux-kernel@...r.kernel.org>
Subject: nvme, allocation failures, and aborted commands

I just got a new Samsung 960 Evo 256GB m.2 nvme drive, and put it through some 
tests.

When I run `iozone -a -s 60G` on my desktop (that has 32GB ram and a amd FX 
8320), i run into some errors. One time the drive completely dropped from the 
system (ok, only the dev device did, it still appeared in lspci) and I had to 
reboot. The later attempt's results can be seen in the attached file (its 5000 
lines, so I didn't think it was good to post inline).

I realize that iozone isn't a good indicator of real world performance, I just 
wanted to see some numbers :D and to stress test the card a bit.

I doubt this is expected behavior regardless of how pointless the exercise is.

I'm running kernel 4.9.13 from debian sid/unstable at the moment.

Please CC me, as I'm not currently subscribed to the list.

Thanks.

-- 
Thomas Fjellstrom
thomas@...llstrom.ca
View attachment "nvme-dmesg-info-trimmed.log" of type "text/plain" (408820 bytes)