CVE-2026-72046 PUBLISHED

gve: fix header buffer corruption with header-split and HW-GRO

Assigner: Linux
Reserved: 09.08.2026 Published: 15.08.2026 Updated: 17.08.2026

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

gve: fix header buffer corruption with header-split and HW-GRO

The DQO RX datapath programs a per-buffer-queue-descriptor header_buf_addr at post time and reads the split header back at completion time. Both the post and the read currently index the header buffer by queue position rather than by the buffer's identity:

  • post (gve_rx_post_buffers_dqo): header_buf_addr is computed from bufq->tail
  • read (gve_rx_dqo): the header is read from desc_idx (the completion queue head index)

This relies on the buffer-queue index and the completion-queue index being equal for the start of every packet, i.e. on the device consuming posted buffers and returning completions in the exact same order. That assumption does not hold once HW-GRO is enabled with multiple flows: coalesced segments are accepted and completed in an order that may differ from the order buffers were posted, and segments from different flows may interleave.

That results in two problems:

  1. Wrong header slot on read. Because the read offset is derived from the completion index (desc_idx) while the device wrote the header to the address programmed for the buffer's buf_id, the driver can copy a header belonging to a different packet. This shows up as throughput drop (about 30% drop and large numbers of TCP retransmissions) with header-split and HW-GRO both enabled and many streams.

  2. Header buffer reused while still owned by the device. The driver advances bufq->head by one per completion and re-posts buffers based on that. Arrival of N RX completions only guarantees that at least N RX buffer descriptors have been read by the device. It does not guarantee that the device has relinquished the ownership of all the buffers corresponding to those N descriptors. With out-of-order completions (e.g. the completion for a packet copied into buffer N arrives before the completion for a packet copied into buffer N-1), the driver can re-post and overwrite a header buffer that the device is still going to write into, corrupting the header of a packet whose completion has not yet been processed.

Fix both issues by indexing the header buffer by buf_id on both the post and read paths. Reading from buf_id's slot is therefore always correct regardless of completion ordering (fixes problem 1).

Indexing by buf_id also ties each header slot to the lifetime of its buffer state. A buffer state is only returned to the free/recycle lists when its own completion (buf_id) is processed, so its header slot can only be re-posted after the device is done with it. This makes header slot reuse safe under out-of-order completions (fixes problem 2).

Allocate (gve_rx_alloc_hdr_bufs) and free (gve_rx_free_hdr_bufs) the header buffers based on num_buf_states to match the buf_id indexing.

Metrics

CVSS Vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
CVSS Score: 9.8

AV:N - The flaw is in the GVE driver RX NAPI path triggered by incoming network frames; a remote peer can reach gve_rx_dqo()/gve_rx_post_buffers_dqo() by sending TCP traffic to an internet-facing GVE interface (e.g., GCP VM) without local access. AC:L - Once header-split and HW-GRO are enabled, an attacker can reliably trigger out-of-order HW-GRO completions by opening many concurrent TCP flows to the target; no race against uncontrollable victim state is required beyond normal multi-stream traffic patterns. PR:N - Exploitation requires only the ability to send network packets to the host; no local account, capabilities, or authentication on the victim is needed. Header-split being enabled is victim configuration, not attacker privilege. UI:N - No victim user action is required during exploitation beyond the host already receiving network traffic on its GVE interface; the attacker does not depend on the user mounting filesystems or opening files. S:U - Impact is confined to the guest kernel network stack and driver DMA header buffers on the affected VM; it does not cross a VM/host or IOMMU security boundary to affect the hypervisor or other tenants. C:H - Out-of-order completions cause the driver to read the wrong header slot, mixing headers between packets/flows and disclosing other connections' L3/L4 header bytes; premature header-buffer reuse also corrupts DMA data the driver later copies into skbs. I:H - The driver can repost and overwrite header DMA buffers while the device is still writing them, corrupting packet headers fed into the stack; wrong headers paired with payloads can break protocol parsing and constitutes exploitable kernel memory corruption. A:H - The bug causes severe TCP throughput collapse, massive retransmissions, and dropped/corrupted packets; header-buffer corruption and malformed skbs can also destabilize the kernel RX path and cause oopses or loss of network availability on the host.

Product Status

Vendor Linux
Product Linux
Versions Default: unaffected
  • affected from 5e37d8254e7f551dda62e7590e819d69c7491845 to 84d3753d4bf284ef770ead6dee2270aaabb3ef41 (excl.)
  • affected from 5e37d8254e7f551dda62e7590e819d69c7491845 to 35267819b25074084130b6a7be18bbaf44d3ae74 (excl.)
  • affected from 5e37d8254e7f551dda62e7590e819d69c7491845 to 9f8e7f59b0c2f466be74bd923726b0f5496c27ad (excl.)
  • affected from 5e37d8254e7f551dda62e7590e819d69c7491845 to d676c9a73bdcd8237425dbb826f2bd1a25c36e40 (excl.)
Vendor Linux
Product Linux
Versions Default: affected
  • Version 6.9 is affected
  • unaffected from 0 to 6.9 (excl.)
  • unaffected from 6.12.101 to 6.12.* (incl.)
  • unaffected from 6.18.40 to 6.18.* (incl.)
  • unaffected from 7.1.5 to 7.1.* (incl.)
  • unaffected from 7.2 to * (incl.)

References