CVE-2026-98965: ice: Restore Ordered MMIO Writes for Tx Doorbells

In the Linux kernel, the following vulnerability has been resolved:

ice: Restore Ordered MMIO Writes for Tx Doorbells

The DQL accounting state which tracks the number of bytes queued for the NIC to transmit, namely dql->numqueued, is updated by the ICE driver when it invokes netdevtxsentqueue(). Subsequently the driver updates the hardware queue tail allowing the NIC to begin processing the work queue. The queue accounting update should be ordered before the NIC begins transmitting descriptors.

The transmit path currently updates the hardware doorbell using writelrelaxed(), which (on arm64) does not provide the same ordering guarantees between writes to normal memory and writes to MMIO registers that writel() does. This introduces a potential race where the NIC begins transmitting descriptors before the dql->numqueued update is globally visible. If the NIC finishes before the update is observed by dqlcompleted(), it detects an invalid state where more bytes have been completed than have been queued. When this happens the following BUGON is triggered.

BUGON(count > numqueued - dql->numcompleted);

This has been observed and manifests as the following crash (note that the trace has been trimmed) in an environment with sustained network load that uses an Intel Corporation Ethernet Controller E810-XXV for SFP (rev 02) on an arm64 machine. Replacing writelrelaxed() with writel() in a test kernel eliminated the crash in the user's workload, which previously reproduced the issue reliably.

kernel BUG at lib/dynamicqueuelimits.c:99

Internal error: Oops - BUG: 00000000f2000800 [#1] SMP pc : dqlcompleted+0x268/0x2a0 lr : icecleantxirq+0x1d4/0x620 [ice] Call trace: dqlcompleted+0x268/0x2a0 (P) icenapipoll+0x94/0x520 [ice] napipoll+0x48/0x3f0 netrxaction+0x194/0x420

This restores the behaviour prior to commit ccde82e90946 ("ice: add E830 Earliest TxTime First Offload support"), which changed the notification mechanism from writel() to writelrelaxed().