linux-kernel - OT: How to report a HW bug

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>] [day] [month] [year] [list]

Message-ID: <20210321184437.56ae13a7@Zen-II-x12.niklas.com>
Date:   Sun, 21 Mar 2021 18:44:37 -0400
From:   David Niklas <Hgntkwis@...mail.net>
To:     LKML <linux-kernel@...r.kernel.org>
Subject: OT: How to report a HW bug

Hello,

I'm asking how to report a bug in the physical HW here because I can't
think of anywhere else to ask where there might be people who have done
this before.

The HW in question is an AMD 3900X and ASRock Taichi B550 MB. I have
turned both CPU and MB in for warranty and the defect remains.

What happens is this:
1: I get my ASM1166 6 port SATA PCIe 2.0 expansion card and plug it into
the second 16 lane wide PCIe port (my GPU is in the top one). It is
connected to the CPU's PCIe lanes.
2: I try to boot. I get several errors regarding memory and chipset
initialization errors. In particular. The last is error code 94. The
system fails to POST and reboots.
3: I place the AIC into the lowest PCIe port (also 16 lanes wide), and
the system boots without a problem. The AIC works correctly. The AIC is
then electrically connected to the chipset, not the CPU.

I'm attaching the output of lspci for the AIC. Here's the AIC I purchased:
https://www.ebay.com/itm/PCIE-SATA-3-0-PCI-E-SATA-Card-PCI-E-PCI-Express-SATA-Controller-6-Ports-SATA3-X4/114295190520

IMHO, the problem could be:
The Chipset, because it doesn't initialize.
The CPU, because it's PCIe lanes go as far as the second PCIe 16 lane
port).
The PCIe 4.0 re-drivers, because they might be malfunctioning.
The card, because it might abuse the PCIe 2.0 protocol.

So, who should I report this to? AMD or ASRock?
Is there anything I can do to further isolate this problem?
Is what I have written sufficient, or would you recommend that I add some
other info?
Is there any difficulty to trying to file a report with AMD and ASRock
which you would have any suggestions about working around?

Thanks,
David

View attachment "lspci-aic-ASM1166.txt" of type "text/plain" (2749 bytes)