Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.debian.kernel > #94322 > unrolled thread

Bug#1149790: linux-image-6.12.111+deb13-amd64: NIC ports (bnxt_en) die after a few hours with AMD-Vi IO_PAGE_FAULT

Back to article view | Back to linux.debian.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Bug#1149790: linux-image-6.12.111+deb13-amd64: NIC ports (bnxt_en) die after a few hours with AMD-Vi IO_PAGE_FAULT Salvatore Bonaccorso <carnil@debian.org> - 2026-10-03 17:00 +0200

#94322 — Bug#1149790: linux-image-6.12.111+deb13-amd64: NIC ports (bnxt_en) die after a few hours with AMD-Vi IO_PAGE_FAULT

Control: forcemerge 149564 1149790

Hi Rudolph,

On Sat, Oct 03, 2026 at 08:25:43AM +0200, Rudolph Bott wrote:
> Package: src:linux
> Version: 6.12.111-1
> Severity: important
> 
> Dear Maintainer,
> 
> we have a bunch of exactly similar systems which were upgraded to the latest
> Debian Linux security patch (6.12.111+deb13). Since then we are
> observing NIC failures (bnxt_en) every few hours. It occurs both on
> systems under load and also on completely idle systems.
> 
> The systems are using dual port "BCM57414 NetXtreme-E 10Gb/25Gb RDMA
> Ethernet Controller" and at some point one of the ports dies and several
> hours later the other port.
> 
> Since these are Dell PowerEdge R6615 servers we have tried updating host
> firmware to the latest versions (including NIC, BIOS etc.) but that did
> not change anything (we now have a mix of different versions across the
> fleet, but all behave the same). What did help: downgrading the kernel
> to 6.12.107+deb13. Systems with the older kernel release are working
> stable.
> 
> The problems usually start with messages like theese:
> kernel: bnxt_en 0000:c3:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x000b address=0xf68cc000 flags=0x0000]
> kernel: bnxt_en 0000:c3:00.0 linkA: Abandoning msg {0x23 0x1335} len: 0 due to firmware status: 0x2000001
> kernel: bnxt_en 0000:c3:00.0 linkA: Abandoning msg {0xb4 0x1336} len: 0 due to firmware status: 0x2000001
> kernel: bnxt_en 0000:c3:00.0 linkA: hwrm req_type 0x23 seq id 0x1337 error 0xf
> kernel: bnxt_en 0000:c3:00.0 linkA: hwrm req_type 0xb4 seq id 0x1338 error 0xf
> kernel: bnxt_en 0000:c3:00.0 linkA: Abandoning msg {0x23 0x133e} len: 0 due to firmware status: 0x2000001
> kernel: bnxt_en 0000:c3:00.0 linkA: Abandoning msg {0xb4 0x133f} len: 0 due to firmware status: 0x2000001
> 
> A complete log from one system is attached, as well as CPU/platform
> information. Also the entire boot log (from the affected kernel).

This is the same as #149564, so merging.

It would be great if you can check as well https://bugs.debian.org/1149564#42

Regards,
Salvatore

[toc] | [standalone]

Back to top | Article view | linux.debian.kernel

csiph-web