From: Lorenzo Bianconi <lorenzo.bianconi@redhat.com>
To: Yunsheng Lin <linyunsheng@huawei.com>
Cc: Lorenzo Bianconi <lorenzo@kernel.org>,
netdev@vger.kernel.org, davem@davemloft.net, kuba@kernel.org,
edumazet@google.com, pabeni@redhat.com, bpf@vger.kernel.org,
toke@redhat.com, willemdebruijn.kernel@gmail.com,
jasowang@redhat.com, sdf@google.com, hawk@kernel.org,
ilias.apalodimas@linaro.org
Subject: Re: [PATCH v6 net-next 1/5] net: add generic per-cpu page_pool allocator
Date: Tue, 30 Jan 2024 15:26:22 +0100 [thread overview]
Message-ID: <ZbkHDo4bxcWtGP9X@lore-desk> (raw)
In-Reply-To: <7343292d-3273-a10a-9167-420f3232dbdd@huawei.com>
[-- Attachment #1: Type: text/plain, Size: 2021 bytes --]
> On 2024/1/29 21:04, Lorenzo Bianconi wrote:
> >> On 2024/1/28 22:20, Lorenzo Bianconi wrote:
> >>
> >>> #ifdef CONFIG_LOCKDEP
> >>> /*
> >>> * register_netdevice() inits txq->_xmit_lock and sets lockdep class
> >>> @@ -11686,6 +11690,27 @@ static void __init net_dev_struct_check(void)
> >>> *
> >>> */
> >>>
> >>> +#define SD_PAGE_POOL_RING_SIZE 256
> >>
> >> I might missed that if there is a reason we choose 256 here, do we
> >> need to use different value for differe page size, for 64K page size,
> >> it means we might need to reserve 16MB memory for each CPU.
> >
> > honestly I have not spent time on it, most of the current page_pool users set
> > pool_size to 256. Anyway, do you mean something like:
> >
> > diff --git a/net/core/dev.c b/net/core/dev.c
> > index f70fb6cad2b2..3934a3fc5c45 100644
> > --- a/net/core/dev.c
> > +++ b/net/core/dev.c
> > @@ -11806,12 +11806,11 @@ static void __init net_dev_struct_check(void)
> > *
> > */
> >
> > -#define SD_PAGE_POOL_RING_SIZE 256
> > static int net_page_pool_alloc(int cpuid)
> > {
> > #if IS_ENABLED(CONFIG_PAGE_POOL)
>
> Isn't better to have a config like CONFIG_PER_CPU_PAGE_POOL to enable
> this feature? and this config can be selected by whoever needs this
> feature?
since it will be used for generic xdp (at least) I think this will be 99%
enabled when we have bpf enabled, right?
>
> > struct page_pool_params page_pool_params = {
> > - .pool_size = SD_PAGE_POOL_RING_SIZE,
> > + .pool_size = PAGE_SIZE < SZ_64K ? 256 : 16,
>
> What about other page size? like 16KB?
> How about something like below:
> PAGE_SIZE << get_order(PER_CPU_PAGE_POOL_MAX_SIZE)
since pool_size is the number of elements in the ptr_ring associated to the pool,
assuming we want to consume PER_CPU_PAGE_POOL_MAX_SIZE for each cpu, something
like:
PER_CPU_PAGE_POOL_MAX_SIZE / PAGE_SIZE
Regards,
Lorenzo
>
> > .nid = NUMA_NO_NODE,
> > };
> > struct page_pool *pp_ptr;
>
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 228 bytes --]
next prev parent reply other threads:[~2024-01-30 14:26 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-01-28 14:20 [PATCH v6 net-next 0/5] add multi-buff support for xdp running in generic mode Lorenzo Bianconi
2024-01-28 14:20 ` [PATCH v6 net-next 1/5] net: add generic per-cpu page_pool allocator Lorenzo Bianconi
2024-01-29 6:37 ` kernel test robot
2024-01-29 12:05 ` Yunsheng Lin
2024-01-29 13:04 ` Lorenzo Bianconi
2024-01-30 11:02 ` Yunsheng Lin
2024-01-30 14:26 ` Lorenzo Bianconi [this message]
2024-02-01 11:49 ` Lorenzo Bianconi
2024-02-01 12:14 ` Yunsheng Lin
2024-01-29 12:45 ` Toke Høiland-Jørgensen
2024-01-29 12:52 ` Lorenzo Bianconi
2024-01-29 15:44 ` Toke Høiland-Jørgensen
2024-01-31 12:51 ` Jesper Dangaard Brouer
2024-01-31 12:28 ` Jesper Dangaard Brouer
2024-01-31 13:36 ` Lorenzo Bianconi
2024-01-28 14:20 ` [PATCH v6 net-next 2/5] xdp: rely on skb pointer reference in do_xdp_generic and netif_receive_generic_xdp Lorenzo Bianconi
2024-01-29 8:05 ` Ilias Apalodimas
2024-01-29 9:58 ` Lorenzo Bianconi
2024-01-28 14:20 ` [PATCH v6 net-next 3/5] xdp: add multi-buff support for xdp running in generic mode Lorenzo Bianconi
2024-01-31 23:47 ` Jakub Kicinski
2024-02-01 11:34 ` Lorenzo Bianconi
2024-02-01 15:15 ` Jakub Kicinski
2024-02-01 16:41 ` Lorenzo Bianconi
2024-01-28 14:20 ` [PATCH v6 net-next 4/5] net: page_pool: make stats available just for global pools Lorenzo Bianconi
2024-01-29 12:06 ` Yunsheng Lin
2024-01-29 13:07 ` Lorenzo Bianconi
2024-01-30 11:23 ` Yunsheng Lin
2024-01-30 13:52 ` Lorenzo Bianconi
2024-01-30 15:11 ` Jesper Dangaard Brouer
2024-01-30 16:01 ` Lorenzo Bianconi
2024-01-31 15:32 ` Toke Høiland-Jørgensen
2024-01-31 23:52 ` Jakub Kicinski
2024-02-01 10:54 ` Lorenzo Bianconi
2024-01-28 14:20 ` [PATCH v6 net-next 5/5] veth: rely on netif_skb_segment_for_xdp utility routine Lorenzo Bianconi
2024-01-29 12:43 ` [PATCH v6 net-next 0/5] add multi-buff support for xdp running in generic mode Toke Høiland-Jørgensen
2024-01-29 13:05 ` Lorenzo Bianconi
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=ZbkHDo4bxcWtGP9X@lore-desk \
--to=lorenzo.bianconi@redhat.com \
--cc=bpf@vger.kernel.org \
--cc=davem@davemloft.net \
--cc=edumazet@google.com \
--cc=hawk@kernel.org \
--cc=ilias.apalodimas@linaro.org \
--cc=jasowang@redhat.com \
--cc=kuba@kernel.org \
--cc=linyunsheng@huawei.com \
--cc=lorenzo@kernel.org \
--cc=netdev@vger.kernel.org \
--cc=pabeni@redhat.com \
--cc=sdf@google.com \
--cc=toke@redhat.com \
--cc=willemdebruijn.kernel@gmail.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).