From: Martin Lau <kafai@fb.com>
To: Magnus Karlsson <magnus.karlsson@intel.com>
Cc: "bjorn.topel@intel.com" <bjorn.topel@intel.com>,
"ast@kernel.org" <ast@kernel.org>,
"daniel@iogearbox.net" <daniel@iogearbox.net>,
"netdev@vger.kernel.org" <netdev@vger.kernel.org>,
"jonathan.lemon@gmail.com" <jonathan.lemon@gmail.com>,
"bpf@vger.kernel.org" <bpf@vger.kernel.org>,
"saeedm@mellanox.com" <saeedm@mellanox.com>,
"jeffrey.t.kirsher@intel.com" <jeffrey.t.kirsher@intel.com>,
"maciej.fijalkowski@intel.com" <maciej.fijalkowski@intel.com>,
"maciejromanfijalkowski@gmail.com"
<maciejromanfijalkowski@gmail.com>
Subject: Re: [PATCH bpf-next 02/12] xsk: consolidate to one single cached producer pointer
Date: Tue, 10 Dec 2019 00:42:58 +0000 [thread overview]
Message-ID: <20191210004254.m5cicj3tkc2bhlrd@kafai-mbp> (raw)
In-Reply-To: <1575878189-31860-3-git-send-email-magnus.karlsson@intel.com>
On Mon, Dec 09, 2019 at 08:56:19AM +0100, Magnus Karlsson wrote:
> Currently, the xsk ring code has two cached producer pointers:
> prod_head and prod_tail. This patch consolidates these two into a
> single one called cached_prod to make the code simpler and easier to
> maintain. This will be in line with the user space part of the the
> code found in libbpf, that only uses a single cached pointer.
>
> The Rx path only uses the two top level functions
> xskq_produce_batch_desc and xskq_produce_flush_desc and they both use
> prod_head and never prod_tail. So just move them over to
> cached_prod.
>
> The Tx XDP_DRV path uses xskq_produce_addr_lazy and
> xskq_produce_flush_addr_n and unnecessarily operates on both prod_tail
> and prod_cons, so move them over to just use cached_prod by skipping
prod_cons or prod_head?
> the intermediate step of updating prod_tail.
>
> The Tx path in XDP_SKB mode uses xskq_reserve_addr and
> xskq_produce_addr. They currently use both cached pointers, but we can
> operate on the global producer pointer in xskq_produce_addr since it
> has to be updated anyway, thus eliminating the use of both cached
> pointers. We can also remove the xskq_nb_free in xskq_produce_addr
> since it is already called in xskq_reserve_addr. No need to do it
> twice.
>
> When there is only one cached producer pointer, we can also simplify
> xskq_nb_free by removing one argument.
next prev parent reply other threads:[~2019-12-10 0:43 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-12-09 7:56 [PATCH bpf-next 00/12] xsk: clean up ring access functions Magnus Karlsson
2019-12-09 7:56 ` [PATCH bpf-next 01/12] xsk: eliminate the lazy update threshold Magnus Karlsson
2019-12-09 7:56 ` [PATCH bpf-next 02/12] xsk: consolidate to one single cached producer pointer Magnus Karlsson
2019-12-10 0:42 ` Martin Lau [this message]
2019-12-10 9:04 ` Magnus Karlsson
2019-12-13 18:04 ` Maxim Mikityanskiy
2019-12-16 8:46 ` Magnus Karlsson
2019-12-19 14:35 ` Maxim Mikityanskiy
2019-12-19 16:21 ` Magnus Karlsson
2019-12-09 7:56 ` [PATCH bpf-next 03/12] xsk: standardize naming of producer ring access functions Magnus Karlsson
2019-12-09 7:56 ` [PATCH bpf-next 04/12] xsk: simplify detection of empty and full rings Magnus Karlsson
2019-12-09 7:56 ` [PATCH bpf-next 05/12] xsk: eliminate the RX batch size Magnus Karlsson
2019-12-09 10:16 ` Sergei Shtylyov
2019-12-09 13:07 ` Magnus Karlsson
2019-12-09 7:56 ` [PATCH bpf-next 06/12] xsk: simplify xskq_nb_avail and xskq_nb_free Magnus Karlsson
2019-12-09 7:56 ` [PATCH bpf-next 07/12] xsk: simplify the consumer ring access functions Magnus Karlsson
2019-12-09 7:56 ` [PATCH bpf-next 08/12] xsk: change names of validation functions Magnus Karlsson
2019-12-09 7:56 ` [PATCH bpf-next 09/12] xsk: ixgbe: i40e: ice: mlx5: xsk_umem_discard_addr to xsk_umem_release_addr Magnus Karlsson
2019-12-09 7:56 ` [PATCH bpf-next 10/12] xsk: remove unnecessary READ_ONCE of data Magnus Karlsson
2019-12-09 7:56 ` [PATCH bpf-next 11/12] xsk: add function naming comments and reorder functions Magnus Karlsson
2019-12-09 7:56 ` [PATCH bpf-next 12/12] xsk: use struct_size() helper Magnus Karlsson
2019-12-10 1:02 ` [PATCH bpf-next 00/12] xsk: clean up ring access functions Martin Lau
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=20191210004254.m5cicj3tkc2bhlrd@kafai-mbp \
--to=kafai@fb.com \
--cc=ast@kernel.org \
--cc=bjorn.topel@intel.com \
--cc=bpf@vger.kernel.org \
--cc=daniel@iogearbox.net \
--cc=jeffrey.t.kirsher@intel.com \
--cc=jonathan.lemon@gmail.com \
--cc=maciej.fijalkowski@intel.com \
--cc=maciejromanfijalkowski@gmail.com \
--cc=magnus.karlsson@intel.com \
--cc=netdev@vger.kernel.org \
--cc=saeedm@mellanox.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).