Re: [PATCH net-next v5 09/14] net: ethernet: cortina: Recycle claimed free queue pages

[netdev-bot+sashiko@kernel.org]

Thank you for your contribution! Sashiko AI review found more severe
issue(s) in other patches of this series. This mail is only an FYI, in case
you want to address the nit picks below when you respin.

Low priority / nit picks for your consideration:

- [Low] The stable dependency on this patch is not recorded.

--

net: ethernet: cortina: Recycle claimed free queue pages

This patch keeps the base page reference and streaming DMA mapping of a
free queue page after all its fragments have been claimed, and records
such slots in a new recycle bitmap. During refill, a recycled page is
reposted if its refcount is back to one. Otherwise it is unmapped and
released before a replacement is allocated.

> net: ethernet: cortina: Recycle claimed free queue pages
>
> The preceding "Preserve in-flight free queue pages" fix releases each
> mapping after its last descriptor is claimed. Refilling therefore
> allocates and maps a fresh page at line rate even after the network stack
> has released all references to an old page.

[Severity: Low]
Should this patch be tied to the earlier fix for stable?

The earlier patch in this series, "net: ethernet: cortina: Preserve
in-flight free queue pages", carries:

    Fixes: 4d5ae32f5e1e ("net: ethernet: Add a driver for Gemini gigabit ethernet")

so it is a stable candidate. Its commit message also says:

    A later patch restores recycling independently of queue position.

This patch is that later patch, and the paragraph quoted above describes
the earlier fix's behaviour as a problem. This patch has no Fixes: tag,
no Cc: stable and no stable dependency note.

Mainline is not affected because both patches land together. Suppose a
stable backport takes only the Fixes-tagged patch, plus the untagged
prerequisites it needs to apply, such as "Index free queue fragments
with XArray". Wouldn't geth_fill_freeq() then do one alloc_page(GFP_ATOMIC)
and one full-page dma_map_single() for every refilled page, with nothing
pointing out that recycling is missing?

Would it make sense to give this patch the same Fixes: tag? The
alternative would be a stable dependency note on the earlier patch that
points to this one.

-- 
Sashiko AI review · https://netdev-ai.bots.linux.dev/sashiko/#/patchset/20261015-gemini-ethernet-fixes-3-v5-0-b5c9e2f7b8a3%40kernel.org