oss-sec mailing list archives
Re: Linux: DMA-after-unmap race in ZCRX via netif_rxq_cleanup_unlease() ordering inversion (netkit + page_pool)
From: Jacob Bachmeyer <jcb62281 () gmail com>
Date: Wed, 27 May 2026 23:09:28 -0500
On 5/27/26 17:53, Prénom? Ahmed wrote:
This report reads like the product of an "AI" system. What "AI" assisted you in preparing this report?Hello,I would like to report a source-proven teardown ordering bug in the Linux kernel that can lead to a DMA-after-unmap race condition involving ZCRX (io_uring zero-copy receive), page_pool, and netkit queue leasing.***Reporter:** Ahmed Abdelmoemen **Discovery Date:** 2026-05-26 **Kernel Version:** Linux 7.1.0-rc3*Executive Summary [...] I am happy to provide more details or assist with testing.
[I am sending this to the lists to ensure that everyone will know that this question has been asked, in an effort to avoid deluging the reporter with this question.]
-- Jacob
Current thread:
- Linux: DMA-after-unmap race in ZCRX via netif_rxq_cleanup_unlease() ordering inversion (netkit + page_pool) Prénom? Ahmed (May 27)
- Re: Linux: DMA-after-unmap race in ZCRX via netif_rxq_cleanup_unlease() ordering inversion (netkit + page_pool) Jacob Bachmeyer (May 27)
- Re: Linux: DMA-after-unmap race in ZCRX via netif_rxq_cleanup_unlease() ordering inversion (netkit + page_pool) Solar Designer (May 28)
