Re: [PATCH net-next v1 05/11] xsk: Install UMEM page pool providers in the core
From: Björn Töpel
Date: Sun Nov 15 2026 - 14:05:17 EST
netdev-bot+sashiko@xxxxxxxxxx writes:
[...]
> Should xsk_pp_detach() be skipped, or the DMA unmap and UMEM release be
> deferred until pp->nr_pools drops to zero, when the close fails?
Deferred. A failed restart leaves the old queue running on the provider.
I have a net fix that keeps the provider on such a queue and retries the
removal, which also covers netif_rxq_cleanup_unlease(); I might add a
fix in v2 that unmaps and releases the UMEM only once the queue DMA has
stopped.
[...]
> Should the xsk_pp pin the buffer pool and the UMEM, and do the DMA
> unmap in xsk_pp_free(), so they outlive the last page pool? Or should
> xsk_bind() defer releasing the pool in this case?
This case can't happen. The page pools of a failed open never hold a
chunk. fbnic fills its rings only in ndo_queue_start(), which can't
fail, so a failed open frees empty pools, and they're destroyed at once.
pool->pp protects nothing there, so v2 drops that branch for a WARN.
Björn