You are not logged in.
Pages: 1
Hello,
I recently upgraded my hardware to following:
-"Asus TUF B450-PLUS GAMING", "AMD Ryzen 7 2700" and the culprit NVMe "INTEL SSDPEKKW256G8".
So far I'm quite happy with the hardware, but the NVMe causes some minor issues.
Rootfs is mounted as BTRFS. If I run
fstrim -v /
Then I get following in the dmesg (kernel: 4.19.18-1-lts):
[ 3803.418461] nvme 0000:01:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0000 address=0x00000000feb21000 flags=0x0000]
This time it logged only few messages but I have got tens messages spammed from single fstrim run.
Is is safe for me to enable daily fstrim.timer service with this hardware?
I also have trouble booting with lvmcached LVs, but I will probably post this problem separately.
If it ain't broken yet, some time passes and I'm doomed to learn its secrets.
Now it is half broken but twice as fast than before.
Offline
I've just bought 2400g + asus tuf b450m pro + sx8200, installed recent gentoo with gcc 8.2, kernel 4.20.12, linux-firmware 20190221, added microcodes.
I've received the same issue after mounted nvme ssd with "discard" option (just after first reboot).
[ 1323.137873] nvme 0000:06:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0000 address=0x00000000fcabc000 flags=0x0000]
...
[ 1323.153346] amd_iommu_report_page_fault: 2 callbacks suppressed
[ 1323.153347] AMD-Vi: Event logged [IO_PAGE_FAULT device=06:00.0 domain=0x0000 address=0x00000000ffa6e000 flags=0x0000]
...
Last edited by puchuu (2019-02-23 22:40:05)
Offline
I can confirm that it is an iommu issue. I can't reproduce it after booting kernel with iommu=soft. I will report this issue later. It looks like nvme is not isolated properly and discard/trim provides issues with iommu. Maybe you can help me with that.
Last edited by puchuu (2019-02-23 23:25:35)
Offline
I will report this issue later.
Please do report the issue upstream not to the arch bug tracker as arch can not fix upstream issues.
Offline
I've reported issue here https://bugzilla.kernel.org/show_bug.cgi?id=202665. I have no skills on reporting issues on kernel.org. Please help me and provide your logs and assumptions to this issue. Thank you.
Offline
Pages: 1