| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID:
<AS1PR10MB5675D13A436CB20FFE5E1082EBD52@AS1PR10MB5675.EURPRD10.PROD.OUTLOOK.COM>
Date: Fri, 7 Mar 2025 13:25:17 +0000
From: "Bouska, Zdenek" <zdenek.bouska@...mens.com>
To: Song Yoong Siang <yoong.siang.song@...el.com>, "David S . Miller"
<davem@...emloft.net>, Eric Dumazet <edumazet@...gle.com>, Jakub Kicinski
<kuba@...nel.org>, Paolo Abeni <pabeni@...hat.com>, Simon Horman
<horms@...nel.org>, Willem de Bruijn <willemb@...gle.com>, "Bezdeka, Florian"
<florian.bezdeka@...mens.com>, Donald Hunter <donald.hunter@...il.com>,
Jonathan Corbet <corbet@....net>, Bjorn Topel <bjorn@...nel.org>, Magnus
Karlsson <magnus.karlsson@...el.com>, Maciej Fijalkowski
<maciej.fijalkowski@...el.com>, Jonathan Lemon <jonathan.lemon@...il.com>,
Andrew Lunn <andrew+netdev@...n.ch>, Alexei Starovoitov <ast@...nel.org>,
Daniel Borkmann <daniel@...earbox.net>, Jesper Dangaard Brouer
<hawk@...nel.org>, John Fastabend <john.fastabend@...il.com>, Joe Damato
<jdamato@...tly.com>, Stanislav Fomichev <sdf@...ichev.me>, Xuan Zhuo
<xuanzhuo@...ux.alibaba.com>, Mina Almasry <almasrymina@...gle.com>, Daniel
Jurgens <danielj@...dia.com>, Andrii Nakryiko <andrii@...nel.org>, Eduard
Zingerman <eddyz87@...il.com>, Mykola Lysenko <mykolal@...com>, Martin KaFai
Lau <martin.lau@...ux.dev>, Song Liu <song@...nel.org>, Yonghong Song
<yonghong.song@...ux.dev>, KP Singh <kpsingh@...nel.org>, Hao Luo
<haoluo@...gle.com>, Jiri Olsa <jolsa@...nel.org>, Shuah Khan
<shuah@...nel.org>, Alexandre Torgue <alexandre.torgue@...s.st.com>, Jose
Abreu <joabreu@...opsys.com>, Maxime Coquelin <mcoquelin.stm32@...il.com>,
Tony Nguyen <anthony.l.nguyen@...el.com>, Przemek Kitszel
<przemyslaw.kitszel@...el.com>, Faizal Rahim
<faizal.abdul.rahim@...ux.intel.com>, Choong Yong Liang
<yong.liang.choong@...ux.intel.com>
CC: "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
"linux-doc@...r.kernel.org" <linux-doc@...r.kernel.org>,
"bpf@...r.kernel.org" <bpf@...r.kernel.org>,
"linux-kselftest@...r.kernel.org" <linux-kselftest@...r.kernel.org>,
"linux-stm32@...md-mailman.stormreply.com"
<linux-stm32@...md-mailman.stormreply.com>,
"linux-arm-kernel@...ts.infradead.org"
<linux-arm-kernel@...ts.infradead.org>, "intel-wired-lan@...ts.osuosl.org"
<intel-wired-lan@...ts.osuosl.org>, "xdp-hints@...-project.net"
<xdp-hints@...-project.net>
Subject: RE: [PATCH bpf-next v12 5/5] igc: Add launch time support to XDP ZC
> -----Original Message-----
> From: Song Yoong Siang <yoong.siang.song@...el.com>
>
> Enable Launch Time Control (LTC) support for XDP zero copy via XDP Tx
> metadata framework.
>
> This patch has been tested with tools/testing/selftests/bpf/xdp_hw_metadata
> on Intel I225-LM Ethernet controller. Below are the test steps and result.
>
> Test 1: Send a single packet with the launch time set to 1 s in the future.
>
> Test steps:
> 1. On the DUT, start the xdp_hw_metadata selftest application:
> $ sudo ./xdp_hw_metadata enp2s0 -l 1000000000 -L 1
>
> 2. On the Link Partner, send a UDP packet with VLAN priority 1 to port 9091
> of the DUT.
>
> Result:
> When the launch time is set to 1 s in the future, the delta between the
> launch time and the transmit hardware timestamp is 0.016 us, as shown in
> printout of the xdp_hw_metadata application below.
> 0x562ff5dc8880: rx_desc[4]->addr=84110 addr=84110 comp_addr=84110 EoP
> rx_hash: 0xE343384 with RSS type:0x1
> HW RX-time: 1734578015467548904 (sec:1734578015.4675)
> delta to User RX-time sec:0.0002 (183.103 usec)
> XDP RX-time: 1734578015467651698 (sec:1734578015.4677)
> delta to User RX-time sec:0.0001 (80.309 usec)
> No rx_vlan_tci or rx_vlan_proto, err=-95
> 0x562ff5dc8880: ping-pong with csum=561c (want c7dd)
> csum_start=34 csum_offset=6
> HW RX-time: 1734578015467548904 (sec:1734578015.4675)
> delta to HW Launch-time sec:1.0000 (1000000.000 usec)
> 0x562ff5dc8880: complete tx idx=4 addr=4018
> HW Launch-time: 1734578016467548904 (sec:1734578016.4675)
> delta to HW TX-complete-time sec:0.0000 (0.016 usec)
> HW TX-complete-time: 1734578016467548920 (sec:1734578016.4675)
> delta to User TX-complete-time sec:0.0000
> (32.546 usec)
> XDP RX-time: 1734578015467651698 (sec:1734578015.4677)
> delta to User TX-complete-time sec:0.9999
> (999929.768 usec)
> HW RX-time: 1734578015467548904 (sec:1734578015.4675)
> delta to HW TX-complete-time sec:1.0000 (1000000.016 usec)
> 0x562ff5dc8880: complete rx idx=132 addr=84110
>
> Test 2: Send 1000 packets with a 10 ms interval and the launch time set to
> 500 us in the future.
>
> Test steps:
> 1. On the DUT, start the xdp_hw_metadata selftest application:
> $ sudo chrt -f 99 ./xdp_hw_metadata enp2s0 -l 500000 -L 1 > \
> /dev/shm/result.log
>
> 2. On the Link Partner, send 1000 UDP packets with a 10 ms interval and
> VLAN priority 1 to port 9091 of the DUT.
>
> Result:
> When the launch time is set to 500 us in the future, the average delta
> between the launch time and the transmit hardware timestamp is 0.016 us,
> as shown in the analysis of /dev/shm/result.log below. The XDP launch time
> works correctly in sending 1000 packets continuously.
> Min delta: 0.005 us
> Avr delta: 0.016 us
> Max delta: 0.031 us
> Total packets forwarded: 1000
>
> Reviewed-by: Faizal Rahim <faizal.abdul.rahim@...ux.intel.com>
> Reviewed-by: Maciej Fijalkowski <maciej.fijalkowski@...el.com>
> Signed-off-by: Song Yoong Siang <yoong.siang.song@...el.com>
> ---
> drivers/net/ethernet/intel/igc/igc.h | 1 +
> drivers/net/ethernet/intel/igc/igc_main.c | 61 ++++++++++++++++++++++-
> 2 files changed, 60 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/net/ethernet/intel/igc/igc.h b/drivers/net/ethernet/intel/igc/igc.h
> index b8111ad9a9a8..cd1d7b6c1782 100644
> --- a/drivers/net/ethernet/intel/igc/igc.h
> +++ b/drivers/net/ethernet/intel/igc/igc.h
> @@ -579,6 +579,7 @@ struct igc_metadata_request {
> struct xsk_tx_metadata *meta;
> struct igc_ring *tx_ring;
> u32 cmd_type;
> + u16 used_desc;
> };
>
> struct igc_q_vector {
> diff --git a/drivers/net/ethernet/intel/igc/igc_main.c
> b/drivers/net/ethernet/intel/igc/igc_main.c
> index 1bfa71545e37..3044392e8ded 100644
> --- a/drivers/net/ethernet/intel/igc/igc_main.c
> +++ b/drivers/net/ethernet/intel/igc/igc_main.c
> @@ -2971,9 +2971,48 @@ static u64 igc_xsk_fill_timestamp(void *_priv)
> return *(u64 *)_priv;
> }
>
> +static void igc_xsk_request_launch_time(u64 launch_time, void *_priv)
> +{
> + struct igc_metadata_request *meta_req = _priv;
> + struct igc_ring *tx_ring = meta_req->tx_ring;
> + __le32 launch_time_offset;
> + bool insert_empty = false;
> + bool first_flag = false;
> + u16 used_desc = 0;
> +
> + if (!tx_ring->launchtime_enable)
> + return;
> +
> + launch_time_offset = igc_tx_launchtime(tx_ring,
> + ns_to_ktime(launch_time),
> + &first_flag, &insert_empty);
> + if (insert_empty) {
> + /* Disregard the launch time request if the required empty frame
> + * fails to be inserted.
> + */
> + if (igc_insert_empty_frame(tx_ring))
> + return;
> +
> + meta_req->tx_buffer =
> + &tx_ring->tx_buffer_info[tx_ring->next_to_use];
> + /* Inserting an empty packet requires two descriptors:
> + * one data descriptor and one context descriptor.
> + */
> + used_desc += 2;
> + }
> +
> + /* Use one context descriptor to specify launch time and first flag. */
> + igc_tx_ctxtdesc(tx_ring, launch_time_offset, first_flag, 0, 0, 0);
> + used_desc += 1;
> +
> + /* Update the number of used descriptors in this request */
> + meta_req->used_desc += used_desc;
> +}
> +
> const struct xsk_tx_metadata_ops igc_xsk_tx_metadata_ops = {
> .tmo_request_timestamp = igc_xsk_request_timestamp,
> .tmo_fill_timestamp = igc_xsk_fill_timestamp,
> + .tmo_request_launch_time = igc_xsk_request_launch_time,
> };
>
> static void igc_xdp_xmit_zc(struct igc_ring *ring)
> @@ -2996,7 +3035,13 @@ static void igc_xdp_xmit_zc(struct igc_ring *ring)
> ntu = ring->next_to_use;
> budget = igc_desc_unused(ring);
>
> - while (xsk_tx_peek_desc(pool, &xdp_desc) && budget--) {
> + /* Packets with launch time require one data descriptor and one context
> + * descriptor. When the launch time falls into the next Qbv cycle, we
> + * may need to insert an empty packet, which requires two more
> + * descriptors. Therefore, to be safe, we always ensure we have at least
> + * 4 descriptors available.
> + */
> + while (xsk_tx_peek_desc(pool, &xdp_desc) && budget >= 4) {
I think that here is a bug: some frames could be missed if budget < 4.
I was able to reproduce it by sending 100000x 60 B frames with minimal IPG
(672 ns between starts of frames) on 1Gbit/s. Always 1026 frames were not sent
and were missing a AF_XDP competition. Interesting was that then even when I sent more
frames for hours it still was 1026 frames not sent and missing competition.
Bug seems to be fixed when I change this line to:
while (budget >= 4 && xsk_tx_peek_desc(pool, &xdp_desc)) {
Do you think this is a good fix?
I think this bug is also in original code base, but I was only able to reproduce
it with launch time.
> struct igc_metadata_request meta_req;
> struct xsk_tx_metadata *meta = NULL;
> struct igc_tx_buffer *bi;
> @@ -3017,9 +3062,19 @@ static void igc_xdp_xmit_zc(struct igc_ring *ring)
> meta_req.tx_ring = ring;
> meta_req.tx_buffer = bi;
> meta_req.meta = meta;
> + meta_req.used_desc = 0;
> xsk_tx_metadata_request(meta, &igc_xsk_tx_metadata_ops,
> &meta_req);
>
> + /* xsk_tx_metadata_request() may have updated next_to_use */
> + ntu = ring->next_to_use;
> +
> + /* xsk_tx_metadata_request() may have updated Tx buffer info */
> + bi = meta_req.tx_buffer;
> +
> + /* xsk_tx_metadata_request() may use a few descriptors */
> + budget -= meta_req.used_desc;
> +
> tx_desc = IGC_TX_DESC(ring, ntu);
> tx_desc->read.cmd_type_len = cpu_to_le32(meta_req.cmd_type);
> tx_desc->read.olinfo_status = cpu_to_le32(olinfo_status);
> @@ -3037,9 +3092,11 @@ static void igc_xdp_xmit_zc(struct igc_ring *ring)
> ntu++;
> if (ntu == ring->count)
> ntu = 0;
> +
> + ring->next_to_use = ntu;
> + budget--;
> }
>
> - ring->next_to_use = ntu;
> if (tx_desc) {
> igc_flush_tx_descriptors(ring);
> xsk_tx_release(pool);
> --
> 2.34.1
Best regards,
Zdenek Bouska
--
Siemens, s.r.o
Foundational Technologies
Powered by blists - more mailing lists