* Intel FAST_FREE offload question
@ 2025-02-21 16:58 Morten Brørup
2025-02-21 17:09 ` Bruce Richardson
0 siblings, 1 reply; 3+ messages in thread
From: Morten Brørup @ 2025-02-21 16:58 UTC (permalink / raw)
To: bruce.richardson, Anatoly Burakov, Vladimir Medvedkin,
Ian Stokes, Jingjing Wu, Praveen Shetty
Cc: dev
Intel NIC folks,
Why do the Intel network drivers, when using RTE_ETH_TX_OFFLOAD_MBUF_FAST_FREE, fall back to normal freeing when the mempool cache is empty (cache->len == 0)? It doesn't make sense to me.
Example:
https://git.dpdk.org/dpdk/tree/drivers/net/intel/common/tx.h#n146
Med venlig hilsen / Kind regards,
-Morten Brørup
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: Intel FAST_FREE offload question
2025-02-21 16:58 Intel FAST_FREE offload question Morten Brørup
@ 2025-02-21 17:09 ` Bruce Richardson
2025-02-21 17:16 ` Morten Brørup
0 siblings, 1 reply; 3+ messages in thread
From: Bruce Richardson @ 2025-02-21 17:09 UTC (permalink / raw)
To: Morten Brørup
Cc: Anatoly Burakov, Vladimir Medvedkin, Ian Stokes, Jingjing Wu,
Praveen Shetty, dev
On Fri, Feb 21, 2025 at 05:58:21PM +0100, Morten Brørup wrote:
> Intel NIC folks,
>
> Why do the Intel network drivers, when using
> RTE_ETH_TX_OFFLOAD_MBUF_FAST_FREE, fall back to normal freeing when the
> mempool cache is empty (cache->len == 0)? It doesn't make sense to me.
>
> Example:
> https://git.dpdk.org/dpdk/tree/drivers/net/intel/common/tx.h#n146
>
Good question. I suspect that it may be a bug and that we meant to check
for size == 0 rather than len == 0.
^ permalink raw reply [flat|nested] 3+ messages in thread
* RE: Intel FAST_FREE offload question
2025-02-21 17:09 ` Bruce Richardson
@ 2025-02-21 17:16 ` Morten Brørup
0 siblings, 0 replies; 3+ messages in thread
From: Morten Brørup @ 2025-02-21 17:16 UTC (permalink / raw)
To: Bruce Richardson
Cc: Anatoly Burakov, Vladimir Medvedkin, Ian Stokes, Jingjing Wu,
Praveen Shetty, dev
> From: Bruce Richardson [mailto:bruce.richardson@intel.com]
> Sent: Friday, 21 February 2025 18.10
>
> On Fri, Feb 21, 2025 at 05:58:21PM +0100, Morten Brørup wrote:
> > Intel NIC folks,
> >
> > Why do the Intel network drivers, when using
> > RTE_ETH_TX_OFFLOAD_MBUF_FAST_FREE, fall back to normal freeing when
> the
> > mempool cache is empty (cache->len == 0)? It doesn't make sense to
> me.
> >
> > Example:
> > https://git.dpdk.org/dpdk/tree/drivers/net/intel/common/tx.h#n146
> >
> Good question. I suspect that it may be a bug and that we meant to
> check
> for size == 0 rather than len == 0.
Then checking for cache == NULL suffices, because rte_mempool_default_cache() returns NULL if the cache size is 0:
https://elixir.bootlin.com/dpdk/v24.11.1/source/lib/mempool/rte_mempool.h#L1333
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2025-02-21 17:16 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2025-02-21 16:58 Intel FAST_FREE offload question Morten Brørup
2025-02-21 17:09 ` Bruce Richardson
2025-02-21 17:16 ` Morten Brørup
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).