From: Jakub Kicinski <kuba@kernel.org>
To: Taehee Yoo <ap420073@gmail.com>, Mina Almasry <almasrymina@google.com>
Cc: davem@davemloft.net, pabeni@redhat.com, edumazet@google.com,
netdev@vger.kernel.org, linux-doc@vger.kernel.org,
donald.hunter@gmail.com, corbet@lwn.net,
michael.chan@broadcom.com, kory.maincent@bootlin.com,
andrew@lunn.ch, maxime.chevallier@bootlin.com,
danieller@nvidia.com, hengqi@linux.alibaba.com,
ecree.xilinx@gmail.com, przemyslaw.kitszel@intel.com,
hkallweit1@gmail.com, ahmed.zaki@intel.com,
paul.greenwalt@intel.com, rrameshbabu@nvidia.com,
idosch@nvidia.com, asml.silence@gmail.com, kaiyuanz@google.com,
willemb@google.com, aleksander.lobakin@intel.com, dw@davidwei.uk,
sridhar.samudrala@intel.com, bcreeley@amd.com
Subject: Re: [PATCH net-next v3 7/7] bnxt_en: add support for device memory tcp
Date: Wed, 9 Oct 2024 17:01:02 -0700 [thread overview]
Message-ID: <20241009170102.1980ed1d@kernel.org> (raw)
In-Reply-To: <CAMArcTWVrQ7KWPt+c0u7X=jvBd2VZGVLwjWYCjMYhWZTymMRTg@mail.gmail.com>
On Thu, 10 Oct 2024 00:37:49 +0900 Taehee Yoo wrote:
> > Yes, but netmem_ref can be either a net_iov or a normal page,
> > and skb_add_rx_frag_netmem() and similar helpers should automatically
> > set skb->unreadable or not.
> >
> > IOW you should be able to always use netmem-aware APIs, no?
>
> I'm not sure the update skb->unreadable flag is possible because
> frag API like skb_add_rx_frag_netmem(), receives only frag, not skb.
> How about an additional API to update skb->unreadable flag?
> skb_update_unreadable() or skb_update_netmem()?
Ah, the case where we don't get skb is because we're just building XDP
frame at that stage. And XDP can't be netmem.
In that case switching to skb_frag_fill_netmem_desc() should be enough.
> > > The reason why the branch exists here is the PP_FLAG_ALLOW_UNREADABLE_NETMEM
> > > flag can't be used with PP_FLAG_DMA_SYNC_DEV.
> >
> > Hm. Isn't the existing check the wrong way around? Is the driver
> > supposed to sync the buffers for device before passing them down?
>
> I haven't thought the failure of PP_FLAG_DMA_SYNC_DEV
> for dmabuf may be wrong.
> I think device memory TCP is not related to this flag.
> So device memory TCP core API should not return failure when
> PP_FLAG_DMA_SYNC_DEV flag is set.
> How about removing this condition check code in device memory TCP core?
I think we need to invert the check..
Mina, WDYT?
diff --git a/net/core/devmem.c b/net/core/devmem.c
index 11b91c12ee11..c5cace3f9831 100644
--- a/net/core/devmem.c
+++ b/net/core/devmem.c
@@ -331,12 +331,6 @@ int mp_dmabuf_devmem_init(struct page_pool *pool)
if (!binding)
return -EINVAL;
- if (!pool->dma_map)
- return -EOPNOTSUPP;
-
- if (pool->dma_sync)
- return -EOPNOTSUPP;
-
if (pool->p.order != 0)
return -E2BIG;
diff --git a/net/core/page_pool.c b/net/core/page_pool.c
index a813d30d2135..c8dbbf262de3 100644
--- a/net/core/page_pool.c
+++ b/net/core/page_pool.c
@@ -287,6 +287,12 @@ static int page_pool_init(struct page_pool *pool,
}
if (pool->mp_priv) {
+ if (!pool->dma_map || !pool->dma_sync)
+ return -EOPNOTSUPP;
+
+ /* Memory provider is responsible for syncing the pages. */
+ pool->dma_sync = 0;
+
err = mp_dmabuf_devmem_init(pool);
if (err) {
pr_warn("%s() mem-provider init failed %d\n", __func__,
next prev parent reply other threads:[~2024-10-10 0:01 UTC|newest]
Thread overview: 73+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-10-03 16:06 [PATCH net-next v3 0/7] bnxt_en: implement device memory TCP for bnxt Taehee Yoo
2024-10-03 16:06 ` [PATCH net-next v3 1/7] bnxt_en: add support for rx-copybreak ethtool command Taehee Yoo
2024-10-03 16:57 ` Brett Creeley
2024-10-03 17:15 ` Taehee Yoo
2024-10-03 17:13 ` Michael Chan
2024-10-03 17:22 ` Taehee Yoo
2024-10-03 17:43 ` Michael Chan
2024-10-03 18:28 ` Taehee Yoo
2024-10-03 18:34 ` Andrew Lunn
2024-10-05 6:29 ` Taehee Yoo
2024-10-08 18:10 ` Jakub Kicinski
2024-10-08 19:38 ` Michael Chan
2024-10-08 19:53 ` Jakub Kicinski
2024-10-08 20:35 ` Michael Chan
2024-10-03 16:06 ` [PATCH net-next v3 2/7] bnxt_en: add support for tcp-data-split " Taehee Yoo
2024-10-08 18:19 ` Jakub Kicinski
2024-10-09 13:54 ` Taehee Yoo
2024-10-09 15:28 ` Jakub Kicinski
2024-10-09 17:47 ` Taehee Yoo
2024-10-31 17:34 ` Taehee Yoo
2024-10-31 23:56 ` Jakub Kicinski
2024-11-01 17:11 ` Taehee Yoo
2024-10-03 16:06 ` [PATCH net-next v3 3/7] net: ethtool: add support for configuring tcp-data-split-thresh Taehee Yoo
2024-10-03 18:25 ` Mina Almasry
2024-10-03 19:33 ` Taehee Yoo
2024-10-04 1:47 ` Mina Almasry
2024-10-05 6:11 ` Taehee Yoo
2024-10-08 18:33 ` Jakub Kicinski
2024-10-09 14:25 ` Taehee Yoo
2024-10-09 15:46 ` Jakub Kicinski
2024-10-09 17:49 ` Taehee Yoo
2024-10-03 16:06 ` [PATCH net-next v3 4/7] bnxt_en: add support for tcp-data-split-thresh ethtool command Taehee Yoo
2024-10-03 18:13 ` Brett Creeley
2024-10-03 19:13 ` Taehee Yoo
2024-10-08 18:35 ` Jakub Kicinski
2024-10-09 14:31 ` Taehee Yoo
2024-10-03 16:06 ` [PATCH net-next v3 5/7] net: devmem: add ring parameter filtering Taehee Yoo
2024-10-03 18:29 ` Mina Almasry
2024-10-04 3:57 ` Taehee Yoo
2024-10-03 18:35 ` Brett Creeley
2024-10-03 18:49 ` Mina Almasry
2024-10-08 19:28 ` Jakub Kicinski
2024-10-09 14:35 ` Taehee Yoo
2024-10-04 4:01 ` Taehee Yoo
2024-10-03 16:06 ` [PATCH net-next v3 6/7] net: ethtool: " Taehee Yoo
2024-10-03 18:32 ` Mina Almasry
2024-10-03 19:35 ` Taehee Yoo
2024-10-03 16:06 ` [PATCH net-next v3 7/7] bnxt_en: add support for device memory tcp Taehee Yoo
2024-10-03 18:43 ` Mina Almasry
2024-10-04 10:34 ` Taehee Yoo
2024-10-08 2:57 ` David Wei
2024-10-09 15:02 ` Taehee Yoo
2024-10-08 19:50 ` Jakub Kicinski
2024-10-09 15:37 ` Taehee Yoo
2024-10-10 0:01 ` Jakub Kicinski [this message]
2024-10-10 17:44 ` Mina Almasry
2024-10-11 1:34 ` Jakub Kicinski
2024-10-11 17:33 ` Mina Almasry
2024-10-11 23:42 ` Jason Gunthorpe
2024-10-14 22:38 ` Mina Almasry
2024-10-15 0:16 ` Jakub Kicinski
2024-10-15 1:10 ` Mina Almasry
2024-10-15 12:44 ` Jason Gunthorpe
2024-10-18 8:25 ` Mina Almasry
2024-10-19 13:55 ` Taehee Yoo
2024-10-15 14:29 ` Pavel Begunkov
2024-10-15 17:38 ` David Wei
2024-10-05 3:48 ` kernel test robot
2024-10-08 2:45 ` David Wei
2024-10-08 3:54 ` Taehee Yoo
2024-10-08 3:58 ` Taehee Yoo
2024-10-16 20:17 ` [PATCH net-next v3 0/7] bnxt_en: implement device memory TCP for bnxt Stanislav Fomichev
2024-10-17 8:58 ` Taehee Yoo
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20241009170102.1980ed1d@kernel.org \
--to=kuba@kernel.org \
--cc=ahmed.zaki@intel.com \
--cc=aleksander.lobakin@intel.com \
--cc=almasrymina@google.com \
--cc=andrew@lunn.ch \
--cc=ap420073@gmail.com \
--cc=asml.silence@gmail.com \
--cc=bcreeley@amd.com \
--cc=corbet@lwn.net \
--cc=danieller@nvidia.com \
--cc=davem@davemloft.net \
--cc=donald.hunter@gmail.com \
--cc=dw@davidwei.uk \
--cc=ecree.xilinx@gmail.com \
--cc=edumazet@google.com \
--cc=hengqi@linux.alibaba.com \
--cc=hkallweit1@gmail.com \
--cc=idosch@nvidia.com \
--cc=kaiyuanz@google.com \
--cc=kory.maincent@bootlin.com \
--cc=linux-doc@vger.kernel.org \
--cc=maxime.chevallier@bootlin.com \
--cc=michael.chan@broadcom.com \
--cc=netdev@vger.kernel.org \
--cc=pabeni@redhat.com \
--cc=paul.greenwalt@intel.com \
--cc=przemyslaw.kitszel@intel.com \
--cc=rrameshbabu@nvidia.com \
--cc=sridhar.samudrala@intel.com \
--cc=willemb@google.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).