Project

General

Profile

Actions

Bug #11953

open

PCI Fatal Error encountered on AMD B450 platform

Added by Rick V over 2 years ago. Updated over 1 year ago.

Status:
New
Priority:
Normal
Category:
-
Start date:
Due date:
% Done:

90%

Estimated time:
Difficulty:
Medium
Tags:
Gerrit CR:

Description

Recently put together a new build on a GIGABYTE AORUS PRO based on the AMD B450 chipset to replace an aging Core2 build from 2007 (which got struck by lightning a while ago. PCI cards and storage were spared.)

  • Ryzen 7 2700X, 8 CPUs, 16 threads
  • 32GB DDR4-2400
  • 512GB SSD
  • 1000GB HDD
  • two spare 2000GB HDDs for RAID1 (to replace single HDD above eventually)

Tested both GTX1650 and GT1030, as there is no builtin ATI Radeon Vega on the processor. Booting from OI DVD forcing VESA driver, or forcing text console made no difference, the faulting controller was moved from pcieb-9 to pcieb-0

  • Booted OpenBSD/i386 miniroot correctly.
  • Booted OpenBSD/amd64 miniroot correctly
  • Booted Fedora Core 31/amd64 Live correctly
  • failed to boot existing OI on hard driive (pcie fatal error on pcieb-9)
  • failed to boot OI Live (pcie fatal error on pcieb-0)
  • failed to boot OI live with VESA driver
  • failed to boot OI live with text console only
  • failed to boot OI live in EFI mode as above (vesa, text-only)

There does not seem to be any kind of serial port present, which may hinder debugging efforts.
Will try to provide more information as needed.


Files

fmdump.sun-pc.log (29.6 KB) fmdump.sun-pc.log output of fmdump -eVp Rick V, 2019-11-11 12:15 AM
messages.log (154 KB) messages.log system log output Rick V, 2019-11-11 12:15 AM
prtconf.txt (323 KB) prtconf.txt prtconf -vDd output Rick V, 2019-11-11 01:19 AM
Screenshot from 2019-10-13 11-56-40.png (191 KB) Screenshot from 2019-10-13 11-56-40.png Harmen R, 2019-11-13 06:45 PM
video_30-9-2019_8-55-26_part1.m4v (679 KB) video_30-9-2019_8-55-26_part1.m4v Harmen R, 2019-11-14 01:36 PM
kmdb.txt (898 Bytes) kmdb.txt János Zsitvai, 2019-11-15 10:38 PM
console.log (18.1 KB) console.log János Zsitvai, 2019-11-16 08:51 PM
prtconf2.txt (301 KB) prtconf2.txt largely similar to the original prtconf dump less a few USB devices Rick V, 2019-11-17 04:03 AM
fmdump2.txt (95.6 KB) fmdump2.txt fmdump of PCIEX-8000-G2 Rick V, 2019-11-17 04:03 AM
fmdump.txt (158 KB) fmdump.txt János Zsitvai, 2019-11-17 01:52 PM
prtconf.txt (270 KB) prtconf.txt János Zsitvai, 2019-11-17 01:52 PM
prtconf.txt (544 KB) prtconf.txt Harmen R, 2019-11-17 08:38 PM
fmdump.txt (2.98 MB) fmdump.txt huge fmdump Rick V, 2019-11-26 09:40 PM
prtconf-aspm.txt (245 KB) prtconf-aspm.txt current prtconf dump with ASPM status Rick V, 2019-11-27 05:17 AM
fmdump.txt (8.7 KB) fmdump.txt Harmen R, 2019-11-28 05:55 AM
prtconf.txt (574 KB) prtconf.txt Harmen R, 2019-11-28 05:55 AM
prtconf_vp.txt (54.5 KB) prtconf_vp.txt Harmen R, 2020-01-04 08:02 PM
err.log (8.34 KB) err.log Rick V, 2020-02-11 01:12 AM

Related issues

Related to illumos gate - Feature #12134: Capture PCIe aspm statusClosedRobert Mustacchi

Actions
Related to illumos gate - Bug #12135: ECRC PCIe errors shouldn't be fatalClosedRobert Mustacchi

Actions
Related to illumos gate - Bug #12136: Want hook to disable PCIe link monitoringClosedRobert Mustacchi

Actions
Actions

Also available in: Atom PDF