This patch set adds AF_XDP selftests based on veth to selftests/bpf.
# Topology: # --------- # ----------- # _ | Process | _ # / ----------- \ # / | \ # / | \ # ----------- | ----------- # | Thread1 | | | Thread2 | # ----------- | ----------- # | | | # ----------- | ----------- # | xskX | | | xskY | # ----------- | ----------- # | | | # ----------- | ---------- # | vethX | --------- | vethY | # ----------- peer ---------- # | | | # namespaceX | namespaceY
These selftests test AF_XDP SKB and Native/DRV modes using veth Virtual Ethernet interfaces.
The test program contains two threads, each thread is single socket with a unique UMEM. It validates in-order packet delivery and packet content by sending packets to each other.
Prerequisites setup by script test_xsk_prerequisites.sh:
Set up veth interfaces as per the topology shown ^^: * setup two veth interfaces and one namespace ** veth<xxxx> in root namespace ** veth<yyyy> in af_xdp<xxxx> namespace ** namespace af_xdp<xxxx> * create a spec file veth.spec that includes this run-time configuration that is read by test scripts - filenames prefixed with test_xsk_ *** xxxx and yyyy are randomly generated 4 digit numbers used to avoid conflict with any existing interface
The following tests are provided:
1. AF_XDP SKB mode Generic mode XDP is driver independent, used when the driver does not have support for XDP. Works on any netdevice using sockets and generic XDP path. XDP hook from netif_receive_skb(). a. nopoll - soft-irq processing b. poll - using poll() syscall c. Socket Teardown Create a Tx and a Rx socket, Tx from one socket, Rx on another. Destroy both sockets, then repeat multiple times. Only nopoll mode is used d. Bi-directional Sockets Configure sockets as bi-directional tx/rx sockets, sets up fill and completion rings on each socket, tx/rx in both directions. Only nopoll mode is used
2. AF_XDP DRV/Native mode Works on any netdevice with XDP_REDIRECT support, driver dependent. Processes packets before SKB allocation. Provides better performance than SKB. Driver hook available just after DMA of buffer descriptor. a. nopoll b. poll c. Socket Teardown d. Bi-directional Sockets * Only copy mode is supported because veth does not currently support zero-copy mode
Total tests: 8
Flow: * Single process spawns two threads: Tx and Rx * Each of these two threads attach to a veth interface within their assigned namespaces * Each thread creates one AF_XDP socket connected to a unique umem for each veth interface * Tx thread transmits 10k packets from veth<xxxx> to veth<yyyy> * Rx thread verifies if all 10k packets were received and delivered in-order, and have the right content
v2 changes: * Move selftests/xsk to selftests/bpf * Remove Makefiles under selftests/xsk, and utilize selftests/bpf/Makefile
Structure of the patch set:
Patch 1: This patch adds XSK Selftests framework under selftests/bpf Patch 2: Adds tests: SKB poll and nopoll mode, and mac-ip-udp debug Patch 3: Adds tests: DRV poll and nopoll mode Patch 4: Adds tests: SKB and DRV Socket Teardown Patch 5: Adds tests: SKB and DRV Bi-directional Sockets
Thanks: Weqaar
Weqaar Janjua (5): selftests/bpf: xsk selftests framework selftests/bpf: xsk selftests - SKB POLL, NOPOLL selftests/bpf: xsk selftests - DRV POLL, NOPOLL selftests/bpf: xsk selftests - Socket Teardown - SKB, DRV selftests/bpf: xsk selftests - Bi-directional Sockets - SKB, DRV
tools/testing/selftests/bpf/Makefile | 15 +- .../bpf/test_xsk_drv_bidirectional.sh | 23 + .../selftests/bpf/test_xsk_drv_nopoll.sh | 20 + .../selftests/bpf/test_xsk_drv_poll.sh | 20 + .../selftests/bpf/test_xsk_drv_teardown.sh | 20 + .../selftests/bpf/test_xsk_prerequisites.sh | 127 ++ .../bpf/test_xsk_skb_bidirectional.sh | 20 + .../selftests/bpf/test_xsk_skb_nopoll.sh | 20 + .../selftests/bpf/test_xsk_skb_poll.sh | 20 + .../selftests/bpf/test_xsk_skb_teardown.sh | 20 + tools/testing/selftests/bpf/xdpxceiver.c | 1056 +++++++++++++++++ tools/testing/selftests/bpf/xdpxceiver.h | 158 +++ tools/testing/selftests/bpf/xsk_env.sh | 28 + tools/testing/selftests/bpf/xsk_prereqs.sh | 119 ++ 14 files changed, 1664 insertions(+), 2 deletions(-) create mode 100755 tools/testing/selftests/bpf/test_xsk_drv_bidirectional.sh create mode 100755 tools/testing/selftests/bpf/test_xsk_drv_nopoll.sh create mode 100755 tools/testing/selftests/bpf/test_xsk_drv_poll.sh create mode 100755 tools/testing/selftests/bpf/test_xsk_drv_teardown.sh create mode 100755 tools/testing/selftests/bpf/test_xsk_prerequisites.sh create mode 100755 tools/testing/selftests/bpf/test_xsk_skb_bidirectional.sh create mode 100755 tools/testing/selftests/bpf/test_xsk_skb_nopoll.sh create mode 100755 tools/testing/selftests/bpf/test_xsk_skb_poll.sh create mode 100755 tools/testing/selftests/bpf/test_xsk_skb_teardown.sh create mode 100644 tools/testing/selftests/bpf/xdpxceiver.c create mode 100644 tools/testing/selftests/bpf/xdpxceiver.h create mode 100755 tools/testing/selftests/bpf/xsk_env.sh create mode 100755 tools/testing/selftests/bpf/xsk_prereqs.sh
This patch adds AF_XDP selftests framework under selftests/bpf.
Prerequisites setup by script test_xsk_prerequisites.sh:
Set up veth interfaces as per the topology shown ^^: * setup two veth interfaces and one namespace ** veth<xxxx> in root namespace ** veth<yyyy> in af_xdp<xxxx> namespace ** namespace af_xdp<xxxx> * create a spec file veth.spec that includes this run-time configuration that is read by test scripts - filenames prefixed with test_xsk_ *** xxxx and yyyy are randomly generated 4 digit numbers used to avoid conflict with any existing interface * tests the veth and xsk layers of the topology
Signed-off-by: Weqaar Janjua weqaar.a.janjua@intel.com --- tools/testing/selftests/bpf/Makefile | 6 +- .../selftests/bpf/test_xsk_framework.sh | 17 +++ .../selftests/bpf/test_xsk_prerequisites.sh | 116 +++++++++++++++++ tools/testing/selftests/bpf/xsk_env.sh | 11 ++ tools/testing/selftests/bpf/xsk_prereqs.sh | 119 ++++++++++++++++++ 5 files changed, 268 insertions(+), 1 deletion(-) create mode 100755 tools/testing/selftests/bpf/test_xsk_framework.sh create mode 100755 tools/testing/selftests/bpf/test_xsk_prerequisites.sh create mode 100755 tools/testing/selftests/bpf/xsk_env.sh create mode 100755 tools/testing/selftests/bpf/xsk_prereqs.sh
diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile index 3d5940cd110d..51436db24f32 100644 --- a/tools/testing/selftests/bpf/Makefile +++ b/tools/testing/selftests/bpf/Makefile @@ -46,7 +46,9 @@ endif
TEST_GEN_FILES = TEST_FILES = test_lwt_ip_encap.o \ - test_tc_edt.o + test_tc_edt.o \ + xsk_prereqs.sh \ + xsk_env.sh
# Order correspond to 'make run_tests' order TEST_PROGS := test_kmod.sh \ @@ -70,6 +72,8 @@ TEST_PROGS := test_kmod.sh \ test_bpftool_build.sh \ test_bpftool.sh \ test_bpftool_metadata.sh \ + test_xsk_prerequisites.sh \ + test_xsk_framework.sh
TEST_PROGS_EXTENDED := with_addr.sh \ with_tunnels.sh \ diff --git a/tools/testing/selftests/bpf/test_xsk_framework.sh b/tools/testing/selftests/bpf/test_xsk_framework.sh new file mode 100755 index 000000000000..2e3f099d001c --- /dev/null +++ b/tools/testing/selftests/bpf/test_xsk_framework.sh @@ -0,0 +1,17 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 +# Copyright(c) 2020 Intel Corporation. + +# See test_xsk_prerequisites.sh for detailed information on tests + +. xsk_prereqs.sh +. xsk_env.sh + +TEST_NAME="XSK FRAMEWORK" + +test_status $ksft_pass "${TEST_NAME}" + +# Must be called in the last test to execute +cleanup_exit ${VETH0} ${VETH1} ${NS1} + +test_exit $ksft_pass 0 diff --git a/tools/testing/selftests/bpf/test_xsk_prerequisites.sh b/tools/testing/selftests/bpf/test_xsk_prerequisites.sh new file mode 100755 index 000000000000..00bfcf53127c --- /dev/null +++ b/tools/testing/selftests/bpf/test_xsk_prerequisites.sh @@ -0,0 +1,116 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 +# Copyright(c) 2020 Intel Corporation, Weqaar Janjua weqaar.a.janjua@intel.com + +# AF_XDP selftests based on veth +# +# End-to-end AF_XDP over Veth test +# +# Topology: +# --------- +# ----------- ----------- +# | xskX | --------- | xskY | +# ----------- | ----------- +# | | | +# ----------- | ---------- +# | vethX | --------- | vethY | +# ----------- peer ---------- +# | | | +# namespaceX | namespaceY +# +# AF_XDP is an address family optimized for high performance packet processing, +# it is XDP’s user-space interface. +# +# An AF_XDP socket is linked to a single UMEM which is a region of virtual +# contiguous memory, divided into equal-sized frames. +# +# Refer to AF_XDP Kernel Documentation for detailed information: +# https://www.kernel.org/doc/html/latest/networking/af_xdp.html +# +# Prerequisites setup by script test_xsk_prerequisites.sh: +# +# Set up veth interfaces as per the topology shown ^^: +# * setup two veth interfaces and one namespace +# ** veth<xxxx> in root namespace +# ** veth<yyyy> in af_xdp<xxxx> namespace +# ** namespace af_xdp<xxxx> +# * create a spec file veth.spec that includes this run-time configuration +# that is read by test scripts - filenames prefixed with test_xsk_ +# *** xxxx and yyyy are randomly generated 4 digit numbers used to avoid +# conflict with any existing interface +# * tests the veth and xsk layers of the topology +# +# Kernel configuration: +# --------------------- +# See "config" file for recommended kernel config options. +# +# Turn on XDP sockets and veth support when compiling i.e. +# Networking support --> +# Networking options --> +# [ * ] XDP sockets +# +# Executing Tests: +# ---------------- +# Must run with CAP_NET_ADMIN capability. +# +# Run (summary only): +# sudo make summary=1 run_tests +# +# Run (full color-coded output): +# sudo make colorconsole=1 run_tests +# +# Run (full output without color-coding): +# sudo make run_tests +# +# Clean: +# sudo make clean + +. xsk_prereqs.sh + +TEST_NAME="PREREQUISITES" + +URANDOM=/dev/urandom +[ ! -e "${URANDOM}" ] && { echo "${URANDOM} not found. Skipping tests."; test_exit 1 1; } + +VETH0_POSTFIX=$(cat ${URANDOM} | tr -dc '0-9' | fold -w 256 | head -n 1 | head --bytes 4) +VETH0=ve${VETH0_POSTFIX} +VETH1_POSTFIX=$(cat ${URANDOM} | tr -dc '0-9' | fold -w 256 | head -n 1 | head --bytes 4) +VETH1=ve${VETH1_POSTFIX} +NS1=af_xdp${VETH1_POSTFIX} +IPADDR_VETH0=192.168.222.1/30 +IPADDR_VETH1=192.168.222.2/30 +MTU=1500 + +setup_vethPairs() { + echo "setting up ${VETH0}: root: ${IPADDR_VETH0}" + ip netns add ${NS1} + ip link add ${VETH0} type veth peer name ${VETH1} + ip addr add dev ${VETH0} ${IPADDR_VETH0} + echo "setting up ${VETH1}: ${NS1}: ${IPADDR_VETH1}" + ip link set ${VETH1} netns ${NS1} + ip netns exec ${NS1} ip addr add dev ${VETH1} ${IPADDR_VETH1} + ip netns exec ${NS1} ip link set ${VETH1} mtu ${MTU} + ip netns exec ${NS1} ip link set ${VETH1} up + ip link set ${VETH0} mtu ${MTU} + ip link set ${VETH0} up +} + +validate_root_exec +validate_veth_support ${VETH0} +validate_configs +setup_vethPairs + +retval=$? +if [ $retval -ne 0 ]; then + test_status $retval "${TEST_NAME}" + cleanup_exit ${VETH0} ${VETH1} ${NS1} + exit $retval +fi + +echo "${VETH0}:${VETH1},${NS1}" > ${SPECFILE} + +echo "Spec file created: ${SPECFILE}" + +test_status $retval "${TEST_NAME}" + +exit $retval diff --git a/tools/testing/selftests/bpf/xsk_env.sh b/tools/testing/selftests/bpf/xsk_env.sh new file mode 100755 index 000000000000..2c41b4284cae --- /dev/null +++ b/tools/testing/selftests/bpf/xsk_env.sh @@ -0,0 +1,11 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 +# Copyright(c) 2020 Intel Corporation. + +. xsk_prereqs.sh + +validate_veth_spec_file + +VETH0=$(cat ${SPECFILE} | cut -d':' -f 1) +VETH1=$(cat ${SPECFILE} | cut -d':' -f 2 | cut -d',' -f 1) +NS1=$(cat ${SPECFILE} | cut -d':' -f 2 | cut -d',' -f 2) diff --git a/tools/testing/selftests/bpf/xsk_prereqs.sh b/tools/testing/selftests/bpf/xsk_prereqs.sh new file mode 100755 index 000000000000..694c5f5ab5e3 --- /dev/null +++ b/tools/testing/selftests/bpf/xsk_prereqs.sh @@ -0,0 +1,119 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 +# Copyright(c) 2020 Intel Corporation. + +ksft_pass=0 +ksft_fail=1 +ksft_xfail=2 +ksft_xpass=3 +ksft_skip=4 + +GREEN='\033[0;92m' +YELLOW='\033[0;93m' +RED='\033[0;31m' +NC='\033[0m' +STACK_LIM=131072 +SPECFILE=veth.spec + +validate_root_exec() +{ + msg="skip all tests:" + if [ $UID != 0 ]; then + echo $msg must be run as root >&2 + test_exit $ksft_fail 2 + else + return $ksft_pass + fi +} + +validate_veth_support() +{ + msg="skip all tests:" + if [ $(ip link add $1 type veth 2>/dev/null; echo $?;) != 0 ]; then + echo $msg veth kernel support not available >&2 + test_exit $ksft_skip 1 + else + ip link del $1 + return $ksft_pass + fi +} + +validate_veth_spec_file() +{ + if [ ! -f ${SPECFILE} ]; then + test_exit $ksft_skip 1 + fi +} + +test_status() +{ + statusval=$1 + if [ -n "${colorconsole+set}" ]; then + if [ $statusval -eq 2 ]; then + echo -e "${YELLOW}$2${NC}: [ ${RED}FAIL${NC} ]" + elif [ $statusval -eq 1 ]; then + echo -e "${YELLOW}$2${NC}: [ ${RED}SKIPPED${NC} ]" + elif [ $statusval -eq 0 ]; then + echo -e "${YELLOW}$2${NC}: [ ${GREEN}PASS${NC} ]" + fi + else + if [ $statusval -eq 2 ]; then + echo -e "$2: [ FAIL ]" + elif [ $statusval -eq 1 ]; then + echo -e "$2: [ SKIPPED ]" + elif [ $statusval -eq 0 ]; then + echo -e "$2: [ PASS ]" + fi + fi +} + +test_exit() +{ + retval=$1 + if [ $2 -ne 0 ]; then + test_status $2 $(basename $0) + fi + exit $retval +} + +clear_configs() +{ + if [ $(ip netns show | grep $3 &>/dev/null; echo $?;) == 0 ]; then + [ $(ip netns exec $3 ip link show $2 &>/dev/null; echo $?;) == 0 ] && + { echo "removing link $2"; ip netns exec $3 ip link del $2; } + echo "removing ns $3" + ip netns del $3 + fi + #Once we delete a veth pair node, the entire veth pair is removed, + #this is just to be cautious just incase the NS does not exist then + #veth node inside NS won't get removed so we explicitly remove it + [ $(ip link show $1 &>/dev/null; echo $?;) == 0 ] && + { echo "removing link $1"; ip link del $1; } + if [ -f ${SPECFILE} ]; then + echo "removing spec file:" ${SPECFILE} + rm -f ${SPECFILE} + fi +} + +cleanup_exit() +{ + echo "cleaning up..." + clear_configs $1 $2 $3 +} + +validate_configs() +{ + [ ! $(type -P ip) ] && { echo "'ip' not found. Skipping tests."; test_exit $ksft_skip 1; } +} + +vethXDPgeneric() +{ + ip link set dev $1 xdpdrv off + ip netns exec $3 ip link set dev $2 xdpdrv off +} + +vethXDPnative() +{ + ip link set dev $1 xdpgeneric off + ip netns exec $3 ip link set dev $2 xdpgeneric off +}
Adds following tests:
1. AF_XDP SKB mode Generic mode XDP is driver independent, used when the driver does not have support for XDP. Works on any netdevice using sockets and generic XDP path. XDP hook from netif_receive_skb(). a. nopoll - soft-irq processing b. poll - using poll() syscall
Signed-off-by: Weqaar Janjua weqaar.a.janjua@intel.com --- tools/testing/selftests/bpf/Makefile | 5 +- .../selftests/bpf/test_xsk_prerequisites.sh | 15 +- .../selftests/bpf/test_xsk_skb_nopoll.sh | 20 + ..._xsk_framework.sh => test_xsk_skb_poll.sh} | 12 +- tools/testing/selftests/bpf/xdpxceiver.c | 961 ++++++++++++++++++ tools/testing/selftests/bpf/xdpxceiver.h | 151 +++ tools/testing/selftests/bpf/xsk_env.sh | 17 + 7 files changed, 1174 insertions(+), 7 deletions(-) create mode 100755 tools/testing/selftests/bpf/test_xsk_skb_nopoll.sh rename tools/testing/selftests/bpf/{test_xsk_framework.sh => test_xsk_skb_poll.sh} (61%) create mode 100644 tools/testing/selftests/bpf/xdpxceiver.c create mode 100644 tools/testing/selftests/bpf/xdpxceiver.h
diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile index 51436db24f32..17af570a32d7 100644 --- a/tools/testing/selftests/bpf/Makefile +++ b/tools/testing/selftests/bpf/Makefile @@ -73,7 +73,8 @@ TEST_PROGS := test_kmod.sh \ test_bpftool.sh \ test_bpftool_metadata.sh \ test_xsk_prerequisites.sh \ - test_xsk_framework.sh + test_xsk_skb_nopoll.sh \ + test_xsk_skb_poll.sh
TEST_PROGS_EXTENDED := with_addr.sh \ with_tunnels.sh \ @@ -84,7 +85,7 @@ TEST_PROGS_EXTENDED := with_addr.sh \ # Compile but not part of 'make run_tests' TEST_GEN_PROGS_EXTENDED = test_sock_addr test_skb_cgroup_id_user \ flow_dissector_load test_flow_dissector test_tcp_check_syncookie_user \ - test_lirc_mode2_user xdping test_cpp runqslower bench + test_lirc_mode2_user xdping test_cpp runqslower bench xdpxceiver
TEST_CUSTOM_PROGS = urandom_read
diff --git a/tools/testing/selftests/bpf/test_xsk_prerequisites.sh b/tools/testing/selftests/bpf/test_xsk_prerequisites.sh index 00bfcf53127c..a9ce8887dffc 100755 --- a/tools/testing/selftests/bpf/test_xsk_prerequisites.sh +++ b/tools/testing/selftests/bpf/test_xsk_prerequisites.sh @@ -8,8 +8,17 @@ # # Topology: # --------- -# ----------- ----------- -# | xskX | --------- | xskY | +# ----------- +# _ | Process | _ +# / ----------- \ +# / | \ +# / | \ +# ----------- | ----------- +# | Thread1 | | | Thread2 | +# ----------- | ----------- +# | | | +# ----------- | ----------- +# | xskX | | | xskY | # ----------- | ----------- # | | | # ----------- | ---------- @@ -40,6 +49,8 @@ # conflict with any existing interface # * tests the veth and xsk layers of the topology # +# See the source xdpxceiver.c for information on each test +# # Kernel configuration: # --------------------- # See "config" file for recommended kernel config options. diff --git a/tools/testing/selftests/bpf/test_xsk_skb_nopoll.sh b/tools/testing/selftests/bpf/test_xsk_skb_nopoll.sh new file mode 100755 index 000000000000..96600b0f5136 --- /dev/null +++ b/tools/testing/selftests/bpf/test_xsk_skb_nopoll.sh @@ -0,0 +1,20 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 +# Copyright(c) 2020 Intel Corporation. + +# See test_xsk_prerequisites.sh for detailed information on tests + +. xsk_prereqs.sh +. xsk_env.sh + +TEST_NAME="SKB NOPOLL" + +vethXDPgeneric ${VETH0} ${VETH1} ${NS1} + +params=("-S") +execxdpxceiver params + +retval=$? +test_status $retval "${TEST_NAME}" + +test_exit $retval 0 diff --git a/tools/testing/selftests/bpf/test_xsk_framework.sh b/tools/testing/selftests/bpf/test_xsk_skb_poll.sh similarity index 61% rename from tools/testing/selftests/bpf/test_xsk_framework.sh rename to tools/testing/selftests/bpf/test_xsk_skb_poll.sh index 2e3f099d001c..d152c8a24251 100755 --- a/tools/testing/selftests/bpf/test_xsk_framework.sh +++ b/tools/testing/selftests/bpf/test_xsk_skb_poll.sh @@ -7,11 +7,17 @@ . xsk_prereqs.sh . xsk_env.sh
-TEST_NAME="XSK FRAMEWORK" +TEST_NAME="SKB POLL"
-test_status $ksft_pass "${TEST_NAME}" +vethXDPgeneric ${VETH0} ${VETH1} ${NS1} + +params=("-S" "-p") +execxdpxceiver params + +retval=$? +test_status $retval "${TEST_NAME}"
# Must be called in the last test to execute cleanup_exit ${VETH0} ${VETH1} ${NS1}
-test_exit $ksft_pass 0 +test_exit $retval 0 diff --git a/tools/testing/selftests/bpf/xdpxceiver.c b/tools/testing/selftests/bpf/xdpxceiver.c new file mode 100644 index 000000000000..106307155bbe --- /dev/null +++ b/tools/testing/selftests/bpf/xdpxceiver.c @@ -0,0 +1,961 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright(c) 2020 Intel Corporation. */ + +/* + * Some functions in this program are taken from + * Linux kernel samples/bpf/xdpsock* and modified + * for use. + * + * See test_xsk_prerequisites.sh for detailed information on test topology + * and prerequisite network setup. + * + * This test program contains two threads, each thread is single socket with + * a unique UMEM. It validates in-order packet delivery and packet content + * by sending packets to each other. + * + * Tests Information: + * ------------------ + * These selftests test AF_XDP SKB and Native/DRV modes using veth + * Virtual Ethernet interfaces. + * + * The following tests are run: + * + * 1. AF_XDP SKB mode + * Generic mode XDP is driver independent, used when the driver does + * not have support for XDP. Works on any netdevice using sockets and + * generic XDP path. XDP hook from netif_receive_skb(). + * a. nopoll - soft-irq processing + * b. poll - using poll() syscall + * + * Total tests: 2 + * + * Flow: + * ----- + * - Single process spawns two threads: Tx and Rx + * - Each of these two threads attach to a veth interface within their assigned + * namespaces + * - Each thread Creates one AF_XDP socket connected to a unique umem for each + * veth interface + * - Tx thread Transmits 10k packets from veth<xxxx> to veth<yyyy> + * - Rx thread verifies if all 10k packets were received and delivered in-order, + * and have the right content + * + * Enable/disable debug mode: + * -------------------------- + * To enable L2 - L4 headers and payload dump of each packet on STDOUT, add + * parameter -D to params array in TEST_XSK_<TEST>.sh, i.e. params=("-S" "-D") + */ + +#define _GNU_SOURCE +#include <fcntl.h> +#include <errno.h> +#include <getopt.h> +#include <asm/barrier.h> +typedef __u16 __sum16; +#include <linux/if_link.h> +#include <linux/if_ether.h> +#include <linux/ip.h> +#include <linux/udp.h> +#include <arpa/inet.h> +#include <net/if.h> +#include <locale.h> +#include <poll.h> +#include <pthread.h> +#include <signal.h> +#include <stdbool.h> +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +#include <sys/mman.h> +#include <sys/resource.h> +#include <sys/types.h> +#include <sys/queue.h> +#include <time.h> +#include <unistd.h> +#include <stdatomic.h> +#include <bpf/xsk.h> +#include "xdpxceiver.h" +#include "../kselftest.h" + +static void __exit_with_error(int error, const char *file, const char *func, int line) +{ + ksft_test_result_fail + ("[%s:%s:%i]: ERROR: %d/"%s"\n", file, func, line, error, strerror(error)); + ksft_exit_xfail(); +} + +#define exit_with_error(error) __exit_with_error(error, __FILE__, __func__, __LINE__) + +#define print_ksft_result(void)\ + (ksft_test_result_pass("PASS: %s %s\n", uut ? "" : "SKB", opt_poll ? "POLL" : "NOPOLL")) + +static void pthread_init_mutex(void) +{ + pthread_mutex_init(&sync_mutex, NULL); + pthread_mutex_init(&sync_mutex_tx, NULL); + pthread_cond_init(&signal_rx_condition, NULL); + pthread_cond_init(&signal_tx_condition, NULL); +} + +static void pthread_destroy_mutex(void) +{ + pthread_mutex_destroy(&sync_mutex); + pthread_mutex_destroy(&sync_mutex_tx); + pthread_cond_destroy(&signal_rx_condition); + pthread_cond_destroy(&signal_tx_condition); +} + +static void *memset32_htonl(void *dest, u32 val, u32 size) +{ + u32 *ptr = (u32 *)dest; + int i; + + val = htonl(val); + + for (i = 0; i < (size & (~0x3)); i += 4) + ptr[i >> 2] = val; + + for (; i < size; i++) + ((char *)dest)[i] = ((char *)&val)[i & 3]; + + return dest; +} + +/* + * This function code has been taken from + * Linux kernel lib/checksum.c + */ +static inline unsigned short from32to16(unsigned int x) +{ + /* add up 16-bit and 16-bit for 16+c bit */ + x = (x & 0xffff) + (x >> 16); + /* add up carry.. */ + x = (x & 0xffff) + (x >> 16); + return x; +} + +/* + * Fold a partial checksum + * This function code has been taken from + * Linux kernel include/asm-generic/checksum.h + */ +static inline __u16 csum_fold(__u32 csum) +{ + u32 sum = (__force u32)csum; + + sum = (sum & 0xffff) + (sum >> 16); + sum = (sum & 0xffff) + (sum >> 16); + return (__force __u16)~sum; +} + +/* + * This function code has been taken from + * Linux kernel lib/checksum.c + */ +static inline u32 from64to32(u64 x) +{ + /* add up 32-bit and 32-bit for 32+c bit */ + x = (x & 0xffffffff) + (x >> 32); + /* add up carry.. */ + x = (x & 0xffffffff) + (x >> 32); + return (u32)x; +} + +__u32 csum_tcpudp_nofold(__be32 saddr, __be32 daddr, __u32 len, __u8 proto, __u32 sum); + +/* + * This function code has been taken from + * Linux kernel lib/checksum.c + */ +__u32 csum_tcpudp_nofold(__be32 saddr, __be32 daddr, __u32 len, __u8 proto, __u32 sum) +{ + unsigned long long s = (__force u32)sum; + + s += (__force u32)saddr; + s += (__force u32)daddr; +#ifdef __BIG_ENDIAN__ + s += proto + len; +#else + s += (proto + len) << 8; +#endif + return (__force __u32)from64to32(s); +} + +/* + * This function has been taken from + * Linux kernel include/asm-generic/checksum.h + */ +static inline __u16 +csum_tcpudp_magic(__be32 saddr, __be32 daddr, __u32 len, __u8 proto, __u32 sum) +{ + return csum_fold(csum_tcpudp_nofold(saddr, daddr, len, proto, sum)); +} + +static inline u16 udp_csum(u32 saddr, u32 daddr, u32 len, u8 proto, u16 *udp_pkt) +{ + u32 csum = 0; + u32 cnt = 0; + + /* udp hdr and data */ + for (; cnt < len; cnt += 2) + csum += udp_pkt[cnt >> 1]; + + return csum_tcpudp_magic(saddr, daddr, len, proto, csum); +} + +static void gen_eth_hdr(void *data, struct ethhdr *eth_hdr) +{ + memcpy(eth_hdr->h_dest, ((struct ifobject *)data)->dst_mac, ETH_ALEN); + memcpy(eth_hdr->h_source, ((struct ifobject *)data)->src_mac, ETH_ALEN); + eth_hdr->h_proto = htons(ETH_P_IP); +} + +static void gen_ip_hdr(void *data, struct iphdr *ip_hdr) +{ + ip_hdr->version = IPVERSION; + ip_hdr->ihl = 0x5; + ip_hdr->tos = 0x0; + ip_hdr->tot_len = htons(IP_PKT_SIZE); + ip_hdr->id = 0; + ip_hdr->frag_off = 0; + ip_hdr->ttl = IPDEFTTL; + ip_hdr->protocol = IPPROTO_UDP; + ip_hdr->saddr = ((struct ifobject *)data)->src_ip; + ip_hdr->daddr = ((struct ifobject *)data)->dst_ip; + ip_hdr->check = 0; +} + +static void gen_udp_hdr(void *data, void *arg, struct udphdr *udp_hdr) +{ + udp_hdr->source = htons(((struct ifobject *)arg)->src_port); + udp_hdr->dest = htons(((struct ifobject *)arg)->dst_port); + udp_hdr->len = htons(UDP_PKT_SIZE); + memset32_htonl(pkt_data + PKT_HDR_SIZE, + htonl(((struct generic_data *)data)->seqnum), UDP_PKT_DATA_SIZE); +} + +static void gen_udp_csum(struct udphdr *udp_hdr, struct iphdr *ip_hdr) +{ + udp_hdr->check = 0; + udp_hdr->check = + udp_csum(ip_hdr->saddr, ip_hdr->daddr, UDP_PKT_SIZE, IPPROTO_UDP, (u16 *)udp_hdr); +} + +static void gen_eth_frame(struct xsk_umem_info *umem, u64 addr) +{ + memcpy(xsk_umem__get_data(umem->buffer, addr), pkt_data, PKT_SIZE); +} + +static void xsk_configure_umem(struct ifobject *data, void *buffer, u64 size) +{ + int ret; + + data->umem = calloc(1, sizeof(struct xsk_umem_info)); + if (!data->umem) + exit_with_error(errno); + + ret = xsk_umem__create(&data->umem->umem, buffer, size, + &data->umem->fq, &data->umem->cq, NULL); + if (ret) + exit_with_error(ret); + + data->umem->buffer = buffer; +} + +static void xsk_populate_fill_ring(struct xsk_umem_info *umem) +{ + int ret, i; + u32 idx; + + ret = xsk_ring_prod__reserve(&umem->fq, XSK_RING_PROD__DEFAULT_NUM_DESCS, &idx); + if (ret != XSK_RING_PROD__DEFAULT_NUM_DESCS) + exit_with_error(ret); + for (i = 0; i < XSK_RING_PROD__DEFAULT_NUM_DESCS; i++) + *xsk_ring_prod__fill_addr(&umem->fq, idx++) = i * XSK_UMEM__DEFAULT_FRAME_SIZE; + xsk_ring_prod__submit(&umem->fq, XSK_RING_PROD__DEFAULT_NUM_DESCS); +} + +static int xsk_configure_socket(struct ifobject *ifobject) +{ + struct xsk_socket_config cfg; + struct xsk_ring_cons *rxr; + struct xsk_ring_prod *txr; + int ret; + + ifobject->xsk = calloc(1, sizeof(struct xsk_socket_info)); + if (!ifobject->xsk) + exit_with_error(errno); + + ifobject->xsk->umem = ifobject->umem; + cfg.rx_size = XSK_RING_CONS__DEFAULT_NUM_DESCS; + cfg.tx_size = XSK_RING_PROD__DEFAULT_NUM_DESCS; + cfg.libbpf_flags = 0; + cfg.xdp_flags = opt_xdp_flags; + cfg.bind_flags = opt_xdp_bind_flags; + + rxr = (ifobject->fv.vector == rx) ? &ifobject->xsk->rx : NULL; + txr = (ifobject->fv.vector == tx) ? &ifobject->xsk->tx : NULL; + + ret = xsk_socket__create(&ifobject->xsk->xsk, ifobject->ifname, + opt_queue, ifobject->umem->umem, rxr, txr, &cfg); + + if (ret) + return 1; + + return 0; +} + +static struct option long_options[] = { + {"interface", required_argument, 0, 'i'}, + {"queue", optional_argument, 0, 'q'}, + {"poll", no_argument, 0, 'p'}, + {"xdp-skb", no_argument, 0, 'S'}, + {"copy", no_argument, 0, 'c'}, + {"debug", optional_argument, 0, 'D'}, + {"tx-pkt-count", optional_argument, 0, 'C'}, + {0, 0, 0, 0} +}; + +static void usage(const char *prog) +{ + const char *str = + " Usage: %s [OPTIONS]\n" + " Options:\n" + " -i, --interface Use interface\n" + " -q, --queue=n Use queue n (default 0)\n" + " -p, --poll Use poll syscall\n" + " -S, --xdp-skb=n Use XDP SKB mode\n" + " -c, --copy Force copy mode\n" + " -D, --debug Debug mode - dump packets L2 - L5\n" + " -C, --tx-pkt-count=n Number of packets to send\n"; + ksft_print_msg(str, prog); +} + +static bool switch_namespace(int idx) +{ + char fqns[26] = "/var/run/netns/"; + int nsfd; + + strncat(fqns, ifdict[idx]->nsname, sizeof(fqns) - strlen(fqns) - 1); + nsfd = open(fqns, O_RDONLY); + + if (nsfd == -1) + exit_with_error(errno); + + if (setns(nsfd, 0) == -1) + exit_with_error(errno); + + return true; +} + +static void *nsswitchthread(void *args) +{ + if (switch_namespace(((struct targs *)args)->idx)) { + ifdict[((struct targs *)args)->idx]->ifindex = + if_nametoindex(ifdict[((struct targs *)args)->idx]->ifname); + if (!ifdict[((struct targs *)args)->idx]->ifindex) { + ksft_test_result_fail + ("ERROR: [%s] interface "%s" does not exist\n", + __func__, ifdict[((struct targs *)args)->idx]->ifname); + ((struct targs *)args)->retptr = false; + } else { + ksft_print_msg("Interface found: %s\n", + ifdict[((struct targs *)args)->idx]->ifname); + ((struct targs *)args)->retptr = true; + } + } else { + ((struct targs *)args)->retptr = false; + } + pthread_exit(NULL); +} + +static int validate_interfaces(void) +{ + bool ret = true; + + for (int i = 0; i < MAX_INTERFACES; i++) { + if (!strcmp(ifdict[i]->ifname, "")) { + ret = false; + ksft_test_result_fail("ERROR: interfaces: -i <int>,<ns> -i <int>,<ns>."); + } + if (strcmp(ifdict[i]->nsname, "")) { + struct targs *targs; + + targs = (struct targs *)malloc(sizeof(struct targs)); + if (!targs) + exit_with_error(errno); + + targs->idx = i; + if (pthread_create(&ns_thread, NULL, nsswitchthread, (void *)targs)) + exit_with_error(errno); + + pthread_join(ns_thread, NULL); + + if (targs->retptr) + ksft_print_msg("NS switched: %s\n", ifdict[i]->nsname); + + free(targs); + } else { + ifdict[i]->ifindex = if_nametoindex(ifdict[i]->ifname); + if (!ifdict[i]->ifindex) { + ksft_test_result_fail + ("ERROR: interface "%s" does not exist\n", ifdict[i]->ifname); + ret = false; + } else { + ksft_print_msg("Interface found: %s\n", ifdict[i]->ifname); + } + } + } + return ret; +} + +static void parse_command_line(int argc, char **argv) +{ + int option_index, interface_index = 0, c; + + opterr = 0; + + for (;;) { + c = getopt_long(argc, argv, "i:q:pScDC:", long_options, &option_index); + + if (c == -1) + break; + + switch (c) { + case 'i': + if (interface_index == MAX_INTERFACES) + break; + char *sptr, *token; + + memcpy(ifdict[interface_index]->ifname, + strtok_r(optarg, ",", &sptr), MAX_INTERFACE_NAME_CHARS); + token = strtok_r(NULL, ",", &sptr); + if (token) + memcpy(ifdict[interface_index]->nsname, token, + MAX_INTERFACES_NAMESPACE_CHARS); + interface_index++; + break; + case 'q': + opt_queue = atoi(optarg); + break; + case 'p': + opt_poll = 1; + break; + case 'S': + opt_xdp_flags |= XDP_FLAGS_SKB_MODE; + opt_xdp_bind_flags |= XDP_COPY; + uut = ORDER_CONTENT_VALIDATE_XDP_SKB; + break; + case 'c': + opt_xdp_bind_flags |= XDP_COPY; + break; + case 'D': + debug_pkt_dump = 1; + break; + case 'C': + opt_pkt_count = atoi(optarg); + break; + default: + usage(basename(argv[0])); + ksft_exit_xfail(); + } + } + + if (!validate_interfaces()) { + usage(basename(argv[0])); + ksft_exit_xfail(); + } +} + +static void kick_tx(struct xsk_socket_info *xsk) +{ + int ret; + + ret = sendto(xsk_socket__fd(xsk->xsk), NULL, 0, MSG_DONTWAIT, NULL, 0); + if (ret >= 0 || errno == ENOBUFS || errno == EAGAIN || errno == EBUSY || errno == ENETDOWN) + return; + exit_with_error(errno); +} + +static inline void complete_tx_only(struct xsk_socket_info *xsk, int batch_size) +{ + unsigned int rcvd; + u32 idx; + + if (!xsk->outstanding_tx) + return; + + if (!NEED_WAKEUP || xsk_ring_prod__needs_wakeup(&xsk->tx)) + kick_tx(xsk); + + rcvd = xsk_ring_cons__peek(&xsk->umem->cq, batch_size, &idx); + if (rcvd) { + xsk_ring_cons__release(&xsk->umem->cq, rcvd); + xsk->outstanding_tx -= rcvd; + xsk->tx_npkts += rcvd; + } +} + +static void rx_pkt(struct xsk_socket_info *xsk, struct pollfd *fds) +{ + unsigned int rcvd, i; + u32 idx_rx = 0, idx_fq = 0; + int ret; + + rcvd = xsk_ring_cons__peek(&xsk->rx, BATCH_SIZE, &idx_rx); + if (!rcvd) { + if (xsk_ring_prod__needs_wakeup(&xsk->umem->fq)) { + ret = poll(fds, 1, POLL_TMOUT); + if (ret < 0) + exit_with_error(ret); + } + return; + } + + ret = xsk_ring_prod__reserve(&xsk->umem->fq, rcvd, &idx_fq); + while (ret != rcvd) { + if (ret < 0) + exit_with_error(ret); + if (xsk_ring_prod__needs_wakeup(&xsk->umem->fq)) { + ret = poll(fds, 1, POLL_TMOUT); + if (ret < 0) + exit_with_error(ret); + } + ret = xsk_ring_prod__reserve(&xsk->umem->fq, rcvd, &idx_fq); + } + + for (i = 0; i < rcvd; i++) { + u64 addr = xsk_ring_cons__rx_desc(&xsk->rx, idx_rx)->addr; + (void)xsk_ring_cons__rx_desc(&xsk->rx, idx_rx++)->len; + u64 orig = xsk_umem__extract_addr(addr); + + addr = xsk_umem__add_offset_to_addr(addr); + pkt_node_rx = malloc(sizeof(struct pkt) + PKT_SIZE); + if (!pkt_node_rx) + exit_with_error(errno); + + pkt_node_rx->pkt_frame = (char *)malloc(PKT_SIZE); + if (!pkt_node_rx->pkt_frame) + exit_with_error(errno); + + memcpy(pkt_node_rx->pkt_frame, xsk_umem__get_data(xsk->umem->buffer, addr), + PKT_SIZE); + + TAILQ_INSERT_HEAD(&head, pkt_node_rx, pkt_nodes); + + *xsk_ring_prod__fill_addr(&xsk->umem->fq, idx_fq++) = orig; + } + + xsk_ring_prod__submit(&xsk->umem->fq, rcvd); + xsk_ring_cons__release(&xsk->rx, rcvd); + xsk->rx_npkts += rcvd; +} + +static void tx_only(struct xsk_socket_info *xsk, u32 *frameptr, int batch_size) +{ + u32 idx; + unsigned int i; + + while (xsk_ring_prod__reserve(&xsk->tx, batch_size, &idx) < batch_size) + complete_tx_only(xsk, batch_size); + + for (i = 0; i < batch_size; i++) { + struct xdp_desc *tx_desc = xsk_ring_prod__tx_desc(&xsk->tx, idx + i); + + tx_desc->addr = (*frameptr + i) << XSK_UMEM__DEFAULT_FRAME_SHIFT; + tx_desc->len = PKT_SIZE; + } + + xsk_ring_prod__submit(&xsk->tx, batch_size); + xsk->outstanding_tx += batch_size; + *frameptr += batch_size; + *frameptr %= num_frames; + complete_tx_only(xsk, batch_size); +} + +static inline int get_batch_size(int pkt_cnt) +{ + if (!opt_pkt_count) + return BATCH_SIZE; + + if (pkt_cnt + BATCH_SIZE <= opt_pkt_count) + return BATCH_SIZE; + + return opt_pkt_count - pkt_cnt; +} + +static void complete_tx_only_all(void *arg) +{ + bool pending; + + do { + pending = false; + if (((struct ifobject *)arg)->xsk->outstanding_tx) { + complete_tx_only(((struct ifobject *) + arg)->xsk, BATCH_SIZE); + pending = !!((struct ifobject *)arg)->xsk->outstanding_tx; + } + } while (pending); +} + +static void tx_only_all(void *arg) +{ + struct pollfd fds[MAX_SOCKS] = { }; + u32 frame_nb = 0; + int pkt_cnt = 0; + int ret; + + fds[0].fd = xsk_socket__fd(((struct ifobject *)arg)->xsk->xsk); + fds[0].events = POLLOUT; + + while ((opt_pkt_count && pkt_cnt < opt_pkt_count) || !opt_pkt_count) { + int batch_size = get_batch_size(pkt_cnt); + + if (opt_poll) { + ret = poll(fds, 1, POLL_TMOUT); + if (ret <= 0) + continue; + + if (!(fds[0].revents & POLLOUT)) + continue; + } + + tx_only(((struct ifobject *)arg)->xsk, &frame_nb, batch_size); + pkt_cnt += batch_size; + } + + if (opt_pkt_count) + complete_tx_only_all(arg); +} + +static void worker_pkt_dump(void) +{ + struct in_addr ipaddr; + + fprintf(stdout, "---------------------------------------\n"); + for (int iter = 0; iter < num_frames - 1; iter++) { + /*extract L2 frame */ + fprintf(stdout, "DEBUG>> L2: dst mac: "); + for (int i = 0; i < ETH_ALEN; i++) + fprintf(stdout, "%02X", ((struct ethhdr *) + pkt_buf[iter]->payload)->h_dest[i]); + + fprintf(stdout, "\nDEBUG>> L2: src mac: "); + for (int i = 0; i < ETH_ALEN; i++) + fprintf(stdout, "%02X", ((struct ethhdr *) + pkt_buf[iter]->payload)->h_source[i]); + + /*extract L3 frame */ + fprintf(stdout, "\nDEBUG>> L3: ip_hdr->ihl: %02X\n", + ((struct iphdr *)(pkt_buf[iter]->payload + sizeof(struct ethhdr)))->ihl); + + ipaddr.s_addr = + ((struct iphdr *)(pkt_buf[iter]->payload + sizeof(struct ethhdr)))->saddr; + fprintf(stdout, "DEBUG>> L3: ip_hdr->saddr: %s\n", inet_ntoa(ipaddr)); + + ipaddr.s_addr = + ((struct iphdr *)(pkt_buf[iter]->payload + sizeof(struct ethhdr)))->daddr; + fprintf(stdout, "DEBUG>> L3: ip_hdr->daddr: %s\n", inet_ntoa(ipaddr)); + + /*extract L4 frame */ + fprintf(stdout, "DEBUG>> L4: udp_hdr->src: %d\n", + ntohs(((struct udphdr *)(pkt_buf[iter]->payload + + sizeof(struct ethhdr) + + sizeof(struct iphdr)))->source)); + + fprintf(stdout, "DEBUG>> L4: udp_hdr->dst: %d\n", + ntohs(((struct udphdr *)(pkt_buf[iter]->payload + + sizeof(struct ethhdr) + + sizeof(struct iphdr)))->dest)); + /*extract L5 frame */ + int payload = *((uint32_t *)(pkt_buf[iter]->payload + PKT_HDR_SIZE)); + + if (payload == EOT) { + ksft_print_msg("End-of-tranmission frame received\n"); + fprintf(stdout, "---------------------------------------\n"); + break; + } + fprintf(stdout, "DEBUG>> L5: payload: %d\n", payload); + fprintf(stdout, "---------------------------------------\n"); + } +} + +static void worker_pkt_validate(void) +{ + u32 payloadseqnum = -2; + + while (1) { + pkt_node_rx_q = malloc(sizeof(struct pkt)); + pkt_node_rx_q = TAILQ_LAST(&head, head_s); + if (!pkt_node_rx_q) + break; + + payloadseqnum = *((uint32_t *)(pkt_node_rx_q->pkt_frame + PKT_HDR_SIZE)); + if (debug_pkt_dump && payloadseqnum != EOT) { + pkt_obj = (struct pkt_frame *)malloc(sizeof(struct pkt_frame)); + pkt_obj->payload = (char *)malloc(PKT_SIZE); + memcpy(pkt_obj->payload, pkt_node_rx_q->pkt_frame, PKT_SIZE); + pkt_buf[payloadseqnum] = pkt_obj; + } + + if (payloadseqnum == EOT) { + ksft_print_msg("End-of-tranmission frame received: PASS\n"); + sigvar = 1; + break; + } + + if (prev_pkt + 1 != payloadseqnum) { + ksft_test_result_fail + ("ERROR: [%s] prev_pkt [%d], payloadseqnum [%d]\n", + __func__, prev_pkt, payloadseqnum); + ksft_exit_xfail(); + } + + TAILQ_REMOVE(&head, pkt_node_rx_q, pkt_nodes); + free(pkt_node_rx_q->pkt_frame); + free(pkt_node_rx_q); + pkt_node_rx_q = NULL; + prev_pkt = payloadseqnum; + pkt_counter++; + } +} + +static void thread_common_ops(void *arg, void *bufs, pthread_mutex_t *mutexptr, + atomic_int *spinningptr) +{ + int ctr = 0; + int ret; + + xsk_configure_umem((struct ifobject *)arg, bufs, num_frames * XSK_UMEM__DEFAULT_FRAME_SIZE); + ret = xsk_configure_socket((struct ifobject *)arg); + + /* Retry Create Socket if it fails as xsk_socket__create() + * is asynchronous + * + * Essential to lock Mutex here to prevent Tx thread from + * entering before Rx and causing a deadlock + */ + pthread_mutex_lock(mutexptr); + while (ret && ctr < SOCK_RECONF_CTR) { + atomic_store(spinningptr, 1); + xsk_configure_umem((struct ifobject *)arg, + bufs, num_frames * XSK_UMEM__DEFAULT_FRAME_SIZE); + ret = xsk_configure_socket((struct ifobject *)arg); + usleep(USLEEP_MAX); + ctr++; + } + atomic_store(spinningptr, 0); + pthread_mutex_unlock(mutexptr); + + if (ctr >= SOCK_RECONF_CTR) + exit_with_error(ret); +} + +static void *worker_testapp_validate(void *arg) +{ + struct udphdr *udp_hdr = + (struct udphdr *)(pkt_data + sizeof(struct ethhdr) + sizeof(struct iphdr)); + struct generic_data *data = (struct generic_data *)malloc(sizeof(struct generic_data)); + struct iphdr *ip_hdr = (struct iphdr *)(pkt_data + sizeof(struct ethhdr)); + struct ethhdr *eth_hdr = (struct ethhdr *)pkt_data; + void *bufs; + + pthread_attr_setstacksize(&attr, THREAD_STACK); + + bufs = mmap(NULL, num_frames * XSK_UMEM__DEFAULT_FRAME_SIZE, + PROT_READ | PROT_WRITE, MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); + if (bufs == MAP_FAILED) + exit_with_error(errno); + + if (strcmp(((struct ifobject *)arg)->nsname, "")) + switch_namespace(((struct ifobject *)arg)->ifdict_index); + + if (((struct ifobject *)arg)->fv.vector == tx) { + int spinningrxctr = 0; + + thread_common_ops(arg, bufs, &sync_mutex_tx, &spinning_tx); + + while (atomic_load(&spinning_rx) && spinningrxctr < SOCK_RECONF_CTR) { + spinningrxctr++; + usleep(USLEEP_MAX); + } + + ksft_print_msg("Interface [%s] vector [Tx]\n", ((struct ifobject *)arg)->ifname); + for (int i = 0; i < num_frames; i++) { + /*send EOT frame */ + if (i == (num_frames - 1)) + data->seqnum = -1; + else + data->seqnum = i; + gen_udp_hdr((void *)data, (void *)arg, udp_hdr); + gen_ip_hdr((void *)arg, ip_hdr); + gen_udp_csum(udp_hdr, ip_hdr); + gen_eth_hdr((void *)arg, eth_hdr); + gen_eth_frame(((struct ifobject *)arg)->umem, + i * XSK_UMEM__DEFAULT_FRAME_SIZE); + } + + free(data); + ksft_print_msg("Sending %d packets on interface %s\n", + (opt_pkt_count - 1), ((struct ifobject *)arg)->ifname); + tx_only_all(arg); + } else if (((struct ifobject *)arg)->fv.vector == rx) { + struct pollfd fds[MAX_SOCKS] = { }; + int ret; + + thread_common_ops(arg, bufs, &sync_mutex_tx, &spinning_rx); + + ksft_print_msg("Interface [%s] vector [Rx]\n", ((struct ifobject *)arg)->ifname); + xsk_populate_fill_ring(((struct ifobject *)arg)->umem); + + TAILQ_INIT(&head); + if (debug_pkt_dump) { + pkt_buf = malloc(sizeof(struct pkt_frame **) * num_frames); + if (!pkt_buf) + exit_with_error(errno); + } + + fds[0].fd = xsk_socket__fd(((struct ifobject *)arg)->xsk->xsk); + fds[0].events = POLLIN; + + pthread_mutex_lock(&sync_mutex); + pthread_cond_signal(&signal_rx_condition); + pthread_mutex_unlock(&sync_mutex); + + while (1) { + if (opt_poll) { + ret = poll(fds, 1, POLL_TMOUT); + if (ret <= 0) + continue; + } + rx_pkt(((struct ifobject *)arg)->xsk, fds); + worker_pkt_validate(); + + if (sigvar) + break; + } + + ksft_print_msg("Received %d packets on interface %s\n", + pkt_counter, ((struct ifobject *)arg)->ifname); + } + + xsk_socket__delete(((struct ifobject *)arg)->xsk->xsk); + (void)xsk_umem__delete(((struct ifobject *)arg)->umem->umem); + pthread_exit(NULL); +} + +static void testapp_validate(void) +{ + pthread_attr_init(&attr); + pthread_attr_setstacksize(&attr, THREAD_STACK); + + pthread_mutex_lock(&sync_mutex); + + /*Spawn RX thread */ + if (pthread_create(&t0, &attr, worker_testapp_validate, (void *)ifdict[1])) + exit_with_error(errno); + + struct timespec max_wait = { 0, 0 }; + + if (clock_gettime(CLOCK_REALTIME, &max_wait)) + exit_with_error(errno); + max_wait.tv_sec += TMOUT_SEC; + + if (pthread_cond_timedwait(&signal_rx_condition, &sync_mutex, &max_wait) == ETIMEDOUT) + exit_with_error(errno); + + pthread_mutex_unlock(&sync_mutex); + + /*Spawn TX thread */ + if (pthread_create(&t1, &attr, worker_testapp_validate, (void *)ifdict[0])) + exit_with_error(errno); + + pthread_join(t1, NULL); + pthread_join(t0, NULL); + + if (debug_pkt_dump) { + worker_pkt_dump(); + for (int iter = 0; iter < num_frames - 1; iter++) { + free(pkt_buf[iter]->payload); + free(pkt_buf[iter]); + } + free(pkt_buf); + } + + print_ksft_result(); +} + +static void init_iface_config(void *ifaceconfig) +{ + /*Init interface0 */ + ifdict[0]->fv.vector = tx; + memcpy(ifdict[0]->dst_mac, ((struct ifaceconfigobj *)ifaceconfig)->dst_mac, ETH_ALEN); + memcpy(ifdict[0]->src_mac, ((struct ifaceconfigobj *)ifaceconfig)->src_mac, ETH_ALEN); + ifdict[0]->dst_ip = ((struct ifaceconfigobj *)ifaceconfig)->dst_ip.s_addr; + ifdict[0]->src_ip = ((struct ifaceconfigobj *)ifaceconfig)->src_ip.s_addr; + ifdict[0]->dst_port = ((struct ifaceconfigobj *)ifaceconfig)->dst_port; + ifdict[0]->src_port = ((struct ifaceconfigobj *)ifaceconfig)->src_port; + + /*Init interface1 */ + ifdict[1]->fv.vector = rx; + memcpy(ifdict[1]->dst_mac, ((struct ifaceconfigobj *)ifaceconfig)->src_mac, ETH_ALEN); + memcpy(ifdict[1]->src_mac, ((struct ifaceconfigobj *)ifaceconfig)->dst_mac, ETH_ALEN); + ifdict[1]->dst_ip = ((struct ifaceconfigobj *)ifaceconfig)->src_ip.s_addr; + ifdict[1]->src_ip = ((struct ifaceconfigobj *)ifaceconfig)->dst_ip.s_addr; + ifdict[1]->dst_port = ((struct ifaceconfigobj *)ifaceconfig)->src_port; + ifdict[1]->src_port = ((struct ifaceconfigobj *)ifaceconfig)->dst_port; +} + +int main(int argc, char **argv) +{ + struct rlimit _rlim = { RLIM_INFINITY, RLIM_INFINITY }; + + if (setrlimit(RLIMIT_MEMLOCK, &_rlim)) + exit_with_error(errno); + + const char *MAC1 = "\x00\x0A\x56\x9E\xEE\x62"; + const char *MAC2 = "\x00\x0A\x56\x9E\xEE\x61"; + const char *IP1 = "192.168.100.162"; + const char *IP2 = "192.168.100.161"; + u16 UDP_DST_PORT = 2020; + u16 UDP_SRC_PORT = 2121; + + ifaceconfig = (struct ifaceconfigobj *)malloc(sizeof(struct ifaceconfigobj)); + memcpy(ifaceconfig->dst_mac, MAC1, ETH_ALEN); + memcpy(ifaceconfig->src_mac, MAC2, ETH_ALEN); + inet_aton(IP1, &ifaceconfig->dst_ip); + inet_aton(IP2, &ifaceconfig->src_ip); + ifaceconfig->dst_port = UDP_DST_PORT; + ifaceconfig->src_port = UDP_SRC_PORT; + + for (int i = 0; i < MAX_INTERFACES; i++) { + ifdict[i] = (struct ifobject *)malloc(sizeof(struct ifobject)); + if (!ifdict[i]) + exit_with_error(errno); + + ifdict[i]->ifdict_index = i; + } + + setlocale(LC_ALL, ""); + + parse_command_line(argc, argv); + + num_frames = ++opt_pkt_count; + + init_iface_config((void *)ifaceconfig); + + pthread_init_mutex(); + + ksft_set_plan(1); + + testapp_validate(); + + for (int i = 0; i < MAX_INTERFACES; i++) + free(ifdict[i]); + + pthread_destroy_mutex(); + + ksft_exit_pass(); + + return 0; +} diff --git a/tools/testing/selftests/bpf/xdpxceiver.h b/tools/testing/selftests/bpf/xdpxceiver.h new file mode 100644 index 000000000000..32ee33311141 --- /dev/null +++ b/tools/testing/selftests/bpf/xdpxceiver.h @@ -0,0 +1,151 @@ +/* SPDX-License-Identifier: GPL-2.0 + * Copyright(c) 2020 Intel Corporation. + */ + +#ifndef XDPXCEIVER_H_ +#define XDPXCEIVER_H_ + +#ifndef SOL_XDP +#define SOL_XDP 283 +#endif + +#ifndef AF_XDP +#define AF_XDP 44 +#endif + +#ifndef PF_XDP +#define PF_XDP AF_XDP +#endif + +#define MAX_INTERFACES 2 +#define MAX_INTERFACE_NAME_CHARS 7 +#define MAX_INTERFACES_NAMESPACE_CHARS 10 +#define MAX_SOCKS 1 +#define PKT_HDR_SIZE (sizeof(struct ethhdr) + sizeof(struct iphdr) + \ + sizeof(struct udphdr)) +#define MIN_PKT_SIZE 64 +#define ETH_FCS_SIZE 4 +#define PKT_SIZE (MIN_PKT_SIZE - ETH_FCS_SIZE) +#define IP_PKT_SIZE (PKT_SIZE - sizeof(struct ethhdr)) +#define UDP_PKT_SIZE (IP_PKT_SIZE - sizeof(struct iphdr)) +#define UDP_PKT_DATA_SIZE (UDP_PKT_SIZE - sizeof(struct udphdr)) +#define TMOUT_SEC (3) +#define EOT (-1) +#define USLEEP_MAX 200000 +#define THREAD_STACK 60000000 +#define SOCK_RECONF_CTR 10 +#define BATCH_SIZE 64 +#define POLL_TMOUT 1000 +#define NEED_WAKEUP true + +typedef __u32 u32; +typedef __u16 u16; +typedef __u8 u8; + +enum TESTS { + ORDER_CONTENT_VALIDATE_XDP_SKB = 0, +}; + +u8 uut; +u8 debug_pkt_dump; +u32 num_frames; + +static u32 opt_xdp_flags = XDP_FLAGS_UPDATE_IF_NOEXIST; +static int opt_queue; +static int opt_pkt_count; +static int opt_poll; +static u32 opt_xdp_bind_flags = XDP_USE_NEED_WAKEUP; +static u8 pkt_data[XSK_UMEM__DEFAULT_FRAME_SIZE]; +static u32 pkt_counter; +static u32 prev_pkt = -1; +static int sigvar; + +struct xsk_umem_info { + struct xsk_ring_prod fq; + struct xsk_ring_cons cq; + struct xsk_umem *umem; + void *buffer; +}; + +struct xsk_socket_info { + struct xsk_ring_cons rx; + struct xsk_ring_prod tx; + struct xsk_umem_info *umem; + struct xsk_socket *xsk; + unsigned long rx_npkts; + unsigned long tx_npkts; + unsigned long prev_rx_npkts; + unsigned long prev_tx_npkts; + u32 outstanding_tx; +}; + +struct flow_vector { + enum fvector { + tx, + rx, + bidi, + undef, + } vector; +}; + +struct generic_data { + u32 seqnum; +}; + +struct ifaceconfigobj { + u8 dst_mac[ETH_ALEN]; + u8 src_mac[ETH_ALEN]; + struct in_addr dst_ip; + struct in_addr src_ip; + u16 src_port; + u16 dst_port; +} *ifaceconfig; + +struct ifobject { + int ifindex; + int ifdict_index; + char ifname[MAX_INTERFACE_NAME_CHARS]; + char nsname[MAX_INTERFACES_NAMESPACE_CHARS]; + struct flow_vector fv; + struct xsk_socket_info *xsk; + struct xsk_umem_info *umem; + u8 dst_mac[ETH_ALEN]; + u8 src_mac[ETH_ALEN]; + u32 dst_ip; + u32 src_ip; + u16 src_port; + u16 dst_port; +}; + +static struct ifobject *ifdict[MAX_INTERFACES]; + +/*threads*/ +atomic_int spinning_tx; +atomic_int spinning_rx; +pthread_mutex_t sync_mutex; +pthread_mutex_t sync_mutex_tx; +pthread_cond_t signal_rx_condition; +pthread_cond_t signal_tx_condition; +pthread_t t0, t1, ns_thread; +pthread_attr_t attr; + +struct targs { + bool retptr; + int idx; +}; + +TAILQ_HEAD(head_s, pkt) head = TAILQ_HEAD_INITIALIZER(head); +struct head_s *head_p; +struct pkt { + char *pkt_frame; + + TAILQ_ENTRY(pkt) pkt_nodes; +} *pkt_node_rx, *pkt_node_rx_q; + +struct pkt_frame { + char *payload; +} *pkt_obj; + +struct pkt_frame **pkt_buf; + +#endif /* XDPXCEIVER_H */ diff --git a/tools/testing/selftests/bpf/xsk_env.sh b/tools/testing/selftests/bpf/xsk_env.sh index 2c41b4284cae..1490bae406e8 100755 --- a/tools/testing/selftests/bpf/xsk_env.sh +++ b/tools/testing/selftests/bpf/xsk_env.sh @@ -4,8 +4,25 @@
. xsk_prereqs.sh
+XSKOBJ=xdpxceiver +NUMPKTS=10000 + validate_veth_spec_file
VETH0=$(cat ${SPECFILE} | cut -d':' -f 1) VETH1=$(cat ${SPECFILE} | cut -d':' -f 2 | cut -d',' -f 1) NS1=$(cat ${SPECFILE} | cut -d':' -f 2 | cut -d',' -f 2) + +execxdpxceiver() +{ + local -a 'paramkeys=("${!'"$1"'[@]}")' copy + paramkeysstr=${paramkeys[*]} + + for index in $paramkeysstr; + do + current=$1"[$index]" + copy[$index]=${!current} + done + + ./${XSKOBJ} -i ${VETH0} -i ${VETH1},${NS1} ${copy[*]} -C ${NUMPKTS} +}
On 11/20/20 5:00 AM, Weqaar Janjua wrote:
Adds following tests:
- AF_XDP SKB mode Generic mode XDP is driver independent, used when the driver does not have support for XDP. Works on any netdevice using sockets and generic XDP path. XDP hook from netif_receive_skb(). a. nopoll - soft-irq processing b. poll - using poll() syscall
Signed-off-by: Weqaar Janjua weqaar.a.janjua@intel.com
tools/testing/selftests/bpf/Makefile | 5 +- .../selftests/bpf/test_xsk_prerequisites.sh | 15 +- .../selftests/bpf/test_xsk_skb_nopoll.sh | 20 + ..._xsk_framework.sh => test_xsk_skb_poll.sh} | 12 +- tools/testing/selftests/bpf/xdpxceiver.c | 961 ++++++++++++++++++ tools/testing/selftests/bpf/xdpxceiver.h | 151 +++ tools/testing/selftests/bpf/xsk_env.sh | 17 + 7 files changed, 1174 insertions(+), 7 deletions(-) create mode 100755 tools/testing/selftests/bpf/test_xsk_skb_nopoll.sh rename tools/testing/selftests/bpf/{test_xsk_framework.sh => test_xsk_skb_poll.sh} (61%) create mode 100644 tools/testing/selftests/bpf/xdpxceiver.c create mode 100644 tools/testing/selftests/bpf/xdpxceiver.h
diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile index 51436db24f32..17af570a32d7 100644 --- a/tools/testing/selftests/bpf/Makefile +++ b/tools/testing/selftests/bpf/Makefile @@ -73,7 +73,8 @@ TEST_PROGS := test_kmod.sh \ test_bpftool.sh \ test_bpftool_metadata.sh \ test_xsk_prerequisites.sh \
- test_xsk_framework.sh
- test_xsk_skb_nopoll.sh \
- test_xsk_skb_poll.sh
TEST_PROGS_EXTENDED := with_addr.sh \ with_tunnels.sh \ @@ -84,7 +85,7 @@ TEST_PROGS_EXTENDED := with_addr.sh \ # Compile but not part of 'make run_tests' TEST_GEN_PROGS_EXTENDED = test_sock_addr test_skb_cgroup_id_user \ flow_dissector_load test_flow_dissector test_tcp_check_syncookie_user \
- test_lirc_mode2_user xdping test_cpp runqslower bench
- test_lirc_mode2_user xdping test_cpp runqslower bench xdpxceiver
TEST_CUSTOM_PROGS = urandom_read diff --git a/tools/testing/selftests/bpf/test_xsk_prerequisites.sh b/tools/testing/selftests/bpf/test_xsk_prerequisites.sh index 00bfcf53127c..a9ce8887dffc 100755 --- a/tools/testing/selftests/bpf/test_xsk_prerequisites.sh +++ b/tools/testing/selftests/bpf/test_xsk_prerequisites.sh @@ -8,8 +8,17 @@ # # Topology: # --------- -# ----------- ----------- -# | xskX | --------- | xskY | +# ----------- +# _ | Process | _ +# / ----------- \ +# / | \ +# / | \ +# ----------- | ----------- +# | Thread1 | | | Thread2 | +# ----------- | ----------- +# | | | +# ----------- | ----------- +# | xskX | | | xskY | # ----------- | ----------- # | | | # ----------- | ---------- @@ -40,6 +49,8 @@ # conflict with any existing interface # * tests the veth and xsk layers of the topology # +# See the source xdpxceiver.c for information on each test +# # Kernel configuration: # --------------------- # See "config" file for recommended kernel config options. diff --git a/tools/testing/selftests/bpf/test_xsk_skb_nopoll.sh b/tools/testing/selftests/bpf/test_xsk_skb_nopoll.sh new file mode 100755 index 000000000000..96600b0f5136 --- /dev/null +++ b/tools/testing/selftests/bpf/test_xsk_skb_nopoll.sh @@ -0,0 +1,20 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 +# Copyright(c) 2020 Intel Corporation.
+# See test_xsk_prerequisites.sh for detailed information on tests
+. xsk_prereqs.sh +. xsk_env.sh
+TEST_NAME="SKB NOPOLL"
+vethXDPgeneric ${VETH0} ${VETH1} ${NS1}
+params=("-S") +execxdpxceiver params
+retval=$? +test_status $retval "${TEST_NAME}"
+test_exit $retval 0 diff --git a/tools/testing/selftests/bpf/test_xsk_framework.sh b/tools/testing/selftests/bpf/test_xsk_skb_poll.sh similarity index 61% rename from tools/testing/selftests/bpf/test_xsk_framework.sh rename to tools/testing/selftests/bpf/test_xsk_skb_poll.sh index 2e3f099d001c..d152c8a24251 100755 --- a/tools/testing/selftests/bpf/test_xsk_framework.sh +++ b/tools/testing/selftests/bpf/test_xsk_skb_poll.sh @@ -7,11 +7,17 @@ . xsk_prereqs.sh . xsk_env.sh
Here both xsk_prereqs.sh and xsk_env.sh are executed. But xsk_env.sh also calls xsk_prereqs.sh. This double execution of xsk_prereqs.sh is required or is an oversight?
-TEST_NAME="XSK FRAMEWORK" +TEST_NAME="SKB POLL" -test_status $ksft_pass "${TEST_NAME}" +vethXDPgeneric ${VETH0} ${VETH1} ${NS1}
+params=("-S" "-p") +execxdpxceiver params
+retval=$? +test_status $retval "${TEST_NAME}" # Must be called in the last test to execute cleanup_exit ${VETH0} ${VETH1} ${NS1} -test_exit $ksft_pass 0 +test_exit $retval 0 diff --git a/tools/testing/selftests/bpf/xdpxceiver.c b/tools/testing/selftests/bpf/xdpxceiver.c
[...]
On Fri, 20 Nov 2020 at 18:54, Yonghong Song yhs@fb.com wrote:
On 11/20/20 5:00 AM, Weqaar Janjua wrote:
Adds following tests:
- AF_XDP SKB mode Generic mode XDP is driver independent, used when the driver does not have support for XDP. Works on any netdevice using sockets and generic XDP path. XDP hook from netif_receive_skb(). a. nopoll - soft-irq processing b. poll - using poll() syscall
Signed-off-by: Weqaar Janjua weqaar.a.janjua@intel.com
tools/testing/selftests/bpf/Makefile | 5 +- .../selftests/bpf/test_xsk_prerequisites.sh | 15 +- .../selftests/bpf/test_xsk_skb_nopoll.sh | 20 + ..._xsk_framework.sh => test_xsk_skb_poll.sh} | 12 +- tools/testing/selftests/bpf/xdpxceiver.c | 961 ++++++++++++++++++ tools/testing/selftests/bpf/xdpxceiver.h | 151 +++ tools/testing/selftests/bpf/xsk_env.sh | 17 + 7 files changed, 1174 insertions(+), 7 deletions(-) create mode 100755 tools/testing/selftests/bpf/test_xsk_skb_nopoll.sh rename tools/testing/selftests/bpf/{test_xsk_framework.sh => test_xsk_skb_poll.sh} (61%) create mode 100644 tools/testing/selftests/bpf/xdpxceiver.c create mode 100644 tools/testing/selftests/bpf/xdpxceiver.h
diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile index 51436db24f32..17af570a32d7 100644 --- a/tools/testing/selftests/bpf/Makefile +++ b/tools/testing/selftests/bpf/Makefile @@ -73,7 +73,8 @@ TEST_PROGS := test_kmod.sh \ test_bpftool.sh \ test_bpftool_metadata.sh \ test_xsk_prerequisites.sh \
test_xsk_framework.sh
test_xsk_skb_nopoll.sh \test_xsk_skb_poll.shTEST_PROGS_EXTENDED := with_addr.sh \ with_tunnels.sh \
@@ -84,7 +85,7 @@ TEST_PROGS_EXTENDED := with_addr.sh \ # Compile but not part of 'make run_tests' TEST_GEN_PROGS_EXTENDED = test_sock_addr test_skb_cgroup_id_user \ flow_dissector_load test_flow_dissector test_tcp_check_syncookie_user \
test_lirc_mode2_user xdping test_cpp runqslower bench
test_lirc_mode2_user xdping test_cpp runqslower bench xdpxceiverTEST_CUSTOM_PROGS = urandom_read
diff --git a/tools/testing/selftests/bpf/test_xsk_prerequisites.sh b/tools/testing/selftests/bpf/test_xsk_prerequisites.sh index 00bfcf53127c..a9ce8887dffc 100755 --- a/tools/testing/selftests/bpf/test_xsk_prerequisites.sh +++ b/tools/testing/selftests/bpf/test_xsk_prerequisites.sh @@ -8,8 +8,17 @@ # # Topology: # --------- -# ----------- ----------- -# | xskX | --------- | xskY | +# ----------- +# _ | Process | _ +# / ----------- \ +# / | \ +# / | \ +# ----------- | ----------- +# | Thread1 | | | Thread2 | +# ----------- | ----------- +# | | | +# ----------- | ----------- +# | xskX | | | xskY | # ----------- | ----------- # | | | # ----------- | ---------- @@ -40,6 +49,8 @@ # conflict with any existing interface # * tests the veth and xsk layers of the topology # +# See the source xdpxceiver.c for information on each test +# # Kernel configuration: # --------------------- # See "config" file for recommended kernel config options. diff --git a/tools/testing/selftests/bpf/test_xsk_skb_nopoll.sh b/tools/testing/selftests/bpf/test_xsk_skb_nopoll.sh new file mode 100755 index 000000000000..96600b0f5136 --- /dev/null +++ b/tools/testing/selftests/bpf/test_xsk_skb_nopoll.sh @@ -0,0 +1,20 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 +# Copyright(c) 2020 Intel Corporation.
+# See test_xsk_prerequisites.sh for detailed information on tests
+. xsk_prereqs.sh +. xsk_env.sh
+TEST_NAME="SKB NOPOLL"
+vethXDPgeneric ${VETH0} ${VETH1} ${NS1}
+params=("-S") +execxdpxceiver params
+retval=$? +test_status $retval "${TEST_NAME}"
+test_exit $retval 0 diff --git a/tools/testing/selftests/bpf/test_xsk_framework.sh b/tools/testing/selftests/bpf/test_xsk_skb_poll.sh similarity index 61% rename from tools/testing/selftests/bpf/test_xsk_framework.sh rename to tools/testing/selftests/bpf/test_xsk_skb_poll.sh index 2e3f099d001c..d152c8a24251 100755 --- a/tools/testing/selftests/bpf/test_xsk_framework.sh +++ b/tools/testing/selftests/bpf/test_xsk_skb_poll.sh @@ -7,11 +7,17 @@ . xsk_prereqs.sh . xsk_env.sh
Here both xsk_prereqs.sh and xsk_env.sh are executed. But xsk_env.sh also calls xsk_prereqs.sh. This double execution of xsk_prereqs.sh is required or is an oversight?
Oversight, will fix as v3 - in all 5/5 test_xsk_*.sh, thanks
-TEST_NAME="XSK FRAMEWORK" +TEST_NAME="SKB POLL"
-test_status $ksft_pass "${TEST_NAME}" +vethXDPgeneric ${VETH0} ${VETH1} ${NS1}
+params=("-S" "-p") +execxdpxceiver params
+retval=$? +test_status $retval "${TEST_NAME}"
# Must be called in the last test to execute cleanup_exit ${VETH0} ${VETH1} ${NS1}
-test_exit $ksft_pass 0 +test_exit $retval 0 diff --git a/tools/testing/selftests/bpf/xdpxceiver.c b/tools/testing/selftests/bpf/xdpxceiver.c
[...]
Adds following tests:
2. AF_XDP DRV/Native mode Works on any netdevice with XDP_REDIRECT support, driver dependent. Processes packets before SKB allocation. Provides better performance than SKB. Driver hook available just after DMA of buffer descriptor. a. nopoll b. poll * Only copy mode is supported because veth does not currently support zero-copy mode
Signed-off-by: Weqaar Janjua weqaar.a.janjua@intel.com --- tools/testing/selftests/bpf/Makefile | 4 +++- .../selftests/bpf/test_xsk_drv_nopoll.sh | 20 ++++++++++++++++ .../selftests/bpf/test_xsk_drv_poll.sh | 23 +++++++++++++++++++ .../selftests/bpf/test_xsk_skb_poll.sh | 3 --- tools/testing/selftests/bpf/xdpxceiver.c | 22 +++++++++++++++--- tools/testing/selftests/bpf/xdpxceiver.h | 1 + 6 files changed, 66 insertions(+), 7 deletions(-) create mode 100755 tools/testing/selftests/bpf/test_xsk_drv_nopoll.sh create mode 100755 tools/testing/selftests/bpf/test_xsk_drv_poll.sh
diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile index 17af570a32d7..9dd3f3b9014f 100644 --- a/tools/testing/selftests/bpf/Makefile +++ b/tools/testing/selftests/bpf/Makefile @@ -74,7 +74,9 @@ TEST_PROGS := test_kmod.sh \ test_bpftool_metadata.sh \ test_xsk_prerequisites.sh \ test_xsk_skb_nopoll.sh \ - test_xsk_skb_poll.sh + test_xsk_skb_poll.sh \ + test_xsk_drv_nopoll.sh \ + test_xsk_drv_poll.sh
TEST_PROGS_EXTENDED := with_addr.sh \ with_tunnels.sh \ diff --git a/tools/testing/selftests/bpf/test_xsk_drv_nopoll.sh b/tools/testing/selftests/bpf/test_xsk_drv_nopoll.sh new file mode 100755 index 000000000000..a7e895bc4bfd --- /dev/null +++ b/tools/testing/selftests/bpf/test_xsk_drv_nopoll.sh @@ -0,0 +1,20 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 +# Copyright(c) 2020 Intel Corporation. + +# See test_xsk_prerequisites.sh for detailed information on tests + +. xsk_prereqs.sh +. xsk_env.sh + +TEST_NAME="DRV NOPOLL" + +vethXDPnative ${VETH0} ${VETH1} ${NS1} + +params=("-N") +execxdpxceiver params + +retval=$? +test_status $retval "${TEST_NAME}" + +test_exit $retval 0 diff --git a/tools/testing/selftests/bpf/test_xsk_drv_poll.sh b/tools/testing/selftests/bpf/test_xsk_drv_poll.sh new file mode 100755 index 000000000000..1fe488d5794a --- /dev/null +++ b/tools/testing/selftests/bpf/test_xsk_drv_poll.sh @@ -0,0 +1,23 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 +# Copyright(c) 2020 Intel Corporation. + +# See test_xsk_prerequisites.sh for detailed information on tests + +. xsk_prereqs.sh +. xsk_env.sh + +TEST_NAME="DRV POLL" + +vethXDPnative ${VETH0} ${VETH1} ${NS1} + +params=("-N" "-p") +execxdpxceiver params + +retval=$? +test_status $retval "${TEST_NAME}" + +# Must be called in the last test to execute +cleanup_exit ${VETH0} ${VETH1} ${NS1} + +test_exit $retval 0 diff --git a/tools/testing/selftests/bpf/test_xsk_skb_poll.sh b/tools/testing/selftests/bpf/test_xsk_skb_poll.sh index d152c8a24251..962a89b40a32 100755 --- a/tools/testing/selftests/bpf/test_xsk_skb_poll.sh +++ b/tools/testing/selftests/bpf/test_xsk_skb_poll.sh @@ -17,7 +17,4 @@ execxdpxceiver params retval=$? test_status $retval "${TEST_NAME}"
-# Must be called in the last test to execute -cleanup_exit ${VETH0} ${VETH1} ${NS1} - test_exit $retval 0 diff --git a/tools/testing/selftests/bpf/xdpxceiver.c b/tools/testing/selftests/bpf/xdpxceiver.c index 106307155bbe..e998200502de 100644 --- a/tools/testing/selftests/bpf/xdpxceiver.c +++ b/tools/testing/selftests/bpf/xdpxceiver.c @@ -27,7 +27,16 @@ * a. nopoll - soft-irq processing * b. poll - using poll() syscall * - * Total tests: 2 + * 2. AF_XDP DRV/Native mode + * Works on any netdevice with XDP_REDIRECT support, driver dependent. Processes + * packets before SKB allocation. Provides better performance than SKB. Driver + * hook available just after DMA of buffer descriptor. + * a. nopoll + * b. poll + * - Only copy mode is supported because veth does not currently support + * zero-copy mode + * + * Total tests: 4 * * Flow: * ----- @@ -87,7 +96,7 @@ static void __exit_with_error(int error, const char *file, const char *func, int #define exit_with_error(error) __exit_with_error(error, __FILE__, __func__, __LINE__)
#define print_ksft_result(void)\ - (ksft_test_result_pass("PASS: %s %s\n", uut ? "" : "SKB", opt_poll ? "POLL" : "NOPOLL")) + (ksft_test_result_pass("PASS: %s %s\n", uut ? "DRV" : "SKB", opt_poll ? "POLL" : "NOPOLL"))
static void pthread_init_mutex(void) { @@ -310,6 +319,7 @@ static struct option long_options[] = { {"queue", optional_argument, 0, 'q'}, {"poll", no_argument, 0, 'p'}, {"xdp-skb", no_argument, 0, 'S'}, + {"xdp-native", no_argument, 0, 'N'}, {"copy", no_argument, 0, 'c'}, {"debug", optional_argument, 0, 'D'}, {"tx-pkt-count", optional_argument, 0, 'C'}, @@ -325,6 +335,7 @@ static void usage(const char *prog) " -q, --queue=n Use queue n (default 0)\n" " -p, --poll Use poll syscall\n" " -S, --xdp-skb=n Use XDP SKB mode\n" + " -N, --xdp-native=n Enforce XDP DRV (native) mode\n" " -c, --copy Force copy mode\n" " -D, --debug Debug mode - dump packets L2 - L5\n" " -C, --tx-pkt-count=n Number of packets to send\n"; @@ -416,7 +427,7 @@ static void parse_command_line(int argc, char **argv) opterr = 0;
for (;;) { - c = getopt_long(argc, argv, "i:q:pScDC:", long_options, &option_index); + c = getopt_long(argc, argv, "i:q:pSNcDC:", long_options, &option_index);
if (c == -1) break; @@ -446,6 +457,11 @@ static void parse_command_line(int argc, char **argv) opt_xdp_bind_flags |= XDP_COPY; uut = ORDER_CONTENT_VALIDATE_XDP_SKB; break; + case 'N': + opt_xdp_flags |= XDP_FLAGS_DRV_MODE; + opt_xdp_bind_flags |= XDP_COPY; + uut = ORDER_CONTENT_VALIDATE_XDP_DRV; + break; case 'c': opt_xdp_bind_flags |= XDP_COPY; break; diff --git a/tools/testing/selftests/bpf/xdpxceiver.h b/tools/testing/selftests/bpf/xdpxceiver.h index 32ee33311141..dba47e818466 100644 --- a/tools/testing/selftests/bpf/xdpxceiver.h +++ b/tools/testing/selftests/bpf/xdpxceiver.h @@ -44,6 +44,7 @@ typedef __u8 u8;
enum TESTS { ORDER_CONTENT_VALIDATE_XDP_SKB = 0, + ORDER_CONTENT_VALIDATE_XDP_DRV = 1, };
u8 uut;
Adds following tests:
1. AF_XDP SKB mode c. Socket Teardown Create a Tx and a Rx socket, Tx from one socket, Rx on another. Destroy both sockets, then repeat multiple times. Only nopoll mode is used
2. AF_XDP DRV/Native mode c. Socket Teardown * Only copy mode is supported because veth does not currently support zero-copy mode
Signed-off-by: Weqaar Janjua weqaar.a.janjua@intel.com --- tools/testing/selftests/bpf/Makefile | 4 ++- .../selftests/bpf/test_xsk_drv_poll.sh | 3 -- .../selftests/bpf/test_xsk_drv_teardown.sh | 23 ++++++++++++ .../selftests/bpf/test_xsk_skb_teardown.sh | 20 +++++++++++ tools/testing/selftests/bpf/xdpxceiver.c | 35 ++++++++++++++++--- tools/testing/selftests/bpf/xdpxceiver.h | 2 ++ 6 files changed, 79 insertions(+), 8 deletions(-) create mode 100755 tools/testing/selftests/bpf/test_xsk_drv_teardown.sh create mode 100755 tools/testing/selftests/bpf/test_xsk_skb_teardown.sh
diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile index 9dd3f3b9014f..515b29d321d7 100644 --- a/tools/testing/selftests/bpf/Makefile +++ b/tools/testing/selftests/bpf/Makefile @@ -76,7 +76,9 @@ TEST_PROGS := test_kmod.sh \ test_xsk_skb_nopoll.sh \ test_xsk_skb_poll.sh \ test_xsk_drv_nopoll.sh \ - test_xsk_drv_poll.sh + test_xsk_drv_poll.sh \ + test_xsk_skb_teardown.sh \ + test_xsk_drv_teardown.sh
TEST_PROGS_EXTENDED := with_addr.sh \ with_tunnels.sh \ diff --git a/tools/testing/selftests/bpf/test_xsk_drv_poll.sh b/tools/testing/selftests/bpf/test_xsk_drv_poll.sh index 1fe488d5794a..46e0ae0cabed 100755 --- a/tools/testing/selftests/bpf/test_xsk_drv_poll.sh +++ b/tools/testing/selftests/bpf/test_xsk_drv_poll.sh @@ -17,7 +17,4 @@ execxdpxceiver params retval=$? test_status $retval "${TEST_NAME}"
-# Must be called in the last test to execute -cleanup_exit ${VETH0} ${VETH1} ${NS1} - test_exit $retval 0 diff --git a/tools/testing/selftests/bpf/test_xsk_drv_teardown.sh b/tools/testing/selftests/bpf/test_xsk_drv_teardown.sh new file mode 100755 index 000000000000..28bf730b589e --- /dev/null +++ b/tools/testing/selftests/bpf/test_xsk_drv_teardown.sh @@ -0,0 +1,23 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 +# Copyright(c) 2020 Intel Corporation. + +# See test_xsk_prerequisites.sh for detailed information on tests + +. xsk_prereqs.sh +. xsk_env.sh + +TEST_NAME="DRV SOCKET TEARDOWN" + +vethXDPnative ${VETH0} ${VETH1} ${NS1} + +params=("-N" "-T") +execxdpxceiver params + +retval=$? +test_status $retval "${TEST_NAME}" + +# Must be called in the last test to execute +cleanup_exit ${VETH0} ${VETH1} ${NS1} + +test_exit $retval 0 diff --git a/tools/testing/selftests/bpf/test_xsk_skb_teardown.sh b/tools/testing/selftests/bpf/test_xsk_skb_teardown.sh new file mode 100755 index 000000000000..3ceda125647b --- /dev/null +++ b/tools/testing/selftests/bpf/test_xsk_skb_teardown.sh @@ -0,0 +1,20 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 +# Copyright(c) 2020 Intel Corporation. + +# See test_xsk_prerequisites.sh for detailed information on tests + +. xsk_prereqs.sh +. xsk_env.sh + +TEST_NAME="SKB SOCKET TEARDOWN" + +vethXDPgeneric ${VETH0} ${VETH1} ${NS1} + +params=("-S" "-T") +execxdpxceiver params + +retval=$? +test_status $retval "${TEST_NAME}" + +test_exit $retval 0 diff --git a/tools/testing/selftests/bpf/xdpxceiver.c b/tools/testing/selftests/bpf/xdpxceiver.c index e998200502de..ba5de1ef9f64 100644 --- a/tools/testing/selftests/bpf/xdpxceiver.c +++ b/tools/testing/selftests/bpf/xdpxceiver.c @@ -26,6 +26,9 @@ * generic XDP path. XDP hook from netif_receive_skb(). * a. nopoll - soft-irq processing * b. poll - using poll() syscall + * c. Socket Teardown + * Create a Tx and a Rx socket, Tx from one socket, Rx on another. Destroy + * both sockets, then repeat multiple times. Only nopoll mode is used * * 2. AF_XDP DRV/Native mode * Works on any netdevice with XDP_REDIRECT support, driver dependent. Processes @@ -33,10 +36,11 @@ * hook available just after DMA of buffer descriptor. * a. nopoll * b. poll + * c. Socket Teardown * - Only copy mode is supported because veth does not currently support * zero-copy mode * - * Total tests: 4 + * Total tests: 6 * * Flow: * ----- @@ -96,7 +100,8 @@ static void __exit_with_error(int error, const char *file, const char *func, int #define exit_with_error(error) __exit_with_error(error, __FILE__, __func__, __LINE__)
#define print_ksft_result(void)\ - (ksft_test_result_pass("PASS: %s %s\n", uut ? "DRV" : "SKB", opt_poll ? "POLL" : "NOPOLL")) + (ksft_test_result_pass("PASS: %s %s %s\n", uut ? "DRV" : "SKB", opt_poll ? "POLL" :\ + "NOPOLL", opt_teardown ? "Socket Teardown" : ""))
static void pthread_init_mutex(void) { @@ -321,6 +326,7 @@ static struct option long_options[] = { {"xdp-skb", no_argument, 0, 'S'}, {"xdp-native", no_argument, 0, 'N'}, {"copy", no_argument, 0, 'c'}, + {"tear-down", no_argument, 0, 'T'}, {"debug", optional_argument, 0, 'D'}, {"tx-pkt-count", optional_argument, 0, 'C'}, {0, 0, 0, 0} @@ -337,6 +343,7 @@ static void usage(const char *prog) " -S, --xdp-skb=n Use XDP SKB mode\n" " -N, --xdp-native=n Enforce XDP DRV (native) mode\n" " -c, --copy Force copy mode\n" + " -T, --tear-down Tear down sockets by repeatedly recreating them\n" " -D, --debug Debug mode - dump packets L2 - L5\n" " -C, --tx-pkt-count=n Number of packets to send\n"; ksft_print_msg(str, prog); @@ -427,7 +434,7 @@ static void parse_command_line(int argc, char **argv) opterr = 0;
for (;;) { - c = getopt_long(argc, argv, "i:q:pSNcDC:", long_options, &option_index); + c = getopt_long(argc, argv, "i:q:pSNcTDC:", long_options, &option_index);
if (c == -1) break; @@ -465,6 +472,9 @@ static void parse_command_line(int argc, char **argv) case 'c': opt_xdp_bind_flags |= XDP_COPY; break; + case 'T': + opt_teardown = 1; + break; case 'D': debug_pkt_dump = 1; break; @@ -853,6 +863,9 @@ static void *worker_testapp_validate(void *arg)
ksft_print_msg("Received %d packets on interface %s\n", pkt_counter, ((struct ifobject *)arg)->ifname); + + if (opt_teardown) + ksft_print_msg("Destroying socket\n"); }
xsk_socket__delete(((struct ifobject *)arg)->xsk->xsk); @@ -898,6 +911,20 @@ static void testapp_validate(void) free(pkt_buf); }
+ if (!opt_teardown) + print_ksft_result(); +} + +static void testapp_sockets(void) +{ + for (int i = 0; i < MAX_TEARDOWN_ITER; i++) { + pkt_counter = 0; + prev_pkt = -1; + sigvar = 0; + ksft_print_msg("Creating socket\n"); + testapp_validate(); + } + print_ksft_result(); }
@@ -964,7 +991,7 @@ int main(int argc, char **argv)
ksft_set_plan(1);
- testapp_validate(); + opt_teardown ? testapp_sockets() : testapp_validate();
for (int i = 0; i < MAX_INTERFACES; i++) free(ifdict[i]); diff --git a/tools/testing/selftests/bpf/xdpxceiver.h b/tools/testing/selftests/bpf/xdpxceiver.h index dba47e818466..9d2670f28d86 100644 --- a/tools/testing/selftests/bpf/xdpxceiver.h +++ b/tools/testing/selftests/bpf/xdpxceiver.h @@ -21,6 +21,7 @@ #define MAX_INTERFACE_NAME_CHARS 7 #define MAX_INTERFACES_NAMESPACE_CHARS 10 #define MAX_SOCKS 1 +#define MAX_TEARDOWN_ITER 10 #define PKT_HDR_SIZE (sizeof(struct ethhdr) + sizeof(struct iphdr) + \ sizeof(struct udphdr)) #define MIN_PKT_SIZE 64 @@ -55,6 +56,7 @@ static u32 opt_xdp_flags = XDP_FLAGS_UPDATE_IF_NOEXIST; static int opt_queue; static int opt_pkt_count; static int opt_poll; +static int opt_teardown; static u32 opt_xdp_bind_flags = XDP_USE_NEED_WAKEUP; static u8 pkt_data[XSK_UMEM__DEFAULT_FRAME_SIZE]; static u32 pkt_counter;
Adds following tests:
1. AF_XDP SKB mode d. Bi-directional Sockets Configure sockets as bi-directional tx/rx sockets, sets up fill and completion rings on each socket, tx/rx in both directions. Only nopoll mode is used
2. AF_XDP DRV/Native mode d. Bi-directional Sockets * Only copy mode is supported because veth does not currently support zero-copy mode
Signed-off-by: Weqaar Janjua weqaar.a.janjua@intel.com --- tools/testing/selftests/bpf/Makefile | 4 +- .../bpf/test_xsk_drv_bidirectional.sh | 23 ++++ .../selftests/bpf/test_xsk_drv_teardown.sh | 3 - .../bpf/test_xsk_skb_bidirectional.sh | 20 ++++ tools/testing/selftests/bpf/xdpxceiver.c | 100 +++++++++++++----- tools/testing/selftests/bpf/xdpxceiver.h | 4 + 6 files changed, 126 insertions(+), 28 deletions(-) create mode 100755 tools/testing/selftests/bpf/test_xsk_drv_bidirectional.sh create mode 100755 tools/testing/selftests/bpf/test_xsk_skb_bidirectional.sh
diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile index 515b29d321d7..258bd72812e0 100644 --- a/tools/testing/selftests/bpf/Makefile +++ b/tools/testing/selftests/bpf/Makefile @@ -78,7 +78,9 @@ TEST_PROGS := test_kmod.sh \ test_xsk_drv_nopoll.sh \ test_xsk_drv_poll.sh \ test_xsk_skb_teardown.sh \ - test_xsk_drv_teardown.sh + test_xsk_drv_teardown.sh \ + test_xsk_skb_bidirectional.sh \ + test_xsk_drv_bidirectional.sh
TEST_PROGS_EXTENDED := with_addr.sh \ with_tunnels.sh \ diff --git a/tools/testing/selftests/bpf/test_xsk_drv_bidirectional.sh b/tools/testing/selftests/bpf/test_xsk_drv_bidirectional.sh new file mode 100755 index 000000000000..d3a7e2934d83 --- /dev/null +++ b/tools/testing/selftests/bpf/test_xsk_drv_bidirectional.sh @@ -0,0 +1,23 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 +# Copyright(c) 2020 Intel Corporation. + +# See test_xsk_prerequisites.sh for detailed information on tests + +. xsk_prereqs.sh +. xsk_env.sh + +TEST_NAME="DRV BIDIRECTIONAL SOCKETS" + +vethXDPnative ${VETH0} ${VETH1} ${NS1} + +params=("-N" "-B") +execxdpxceiver params + +retval=$? +test_status $retval "${TEST_NAME}" + +# Must be called in the last test to execute +cleanup_exit ${VETH0} ${VETH1} ${NS1} + +test_exit $retval 0 diff --git a/tools/testing/selftests/bpf/test_xsk_drv_teardown.sh b/tools/testing/selftests/bpf/test_xsk_drv_teardown.sh index 28bf730b589e..7f11033747ac 100755 --- a/tools/testing/selftests/bpf/test_xsk_drv_teardown.sh +++ b/tools/testing/selftests/bpf/test_xsk_drv_teardown.sh @@ -17,7 +17,4 @@ execxdpxceiver params retval=$? test_status $retval "${TEST_NAME}"
-# Must be called in the last test to execute -cleanup_exit ${VETH0} ${VETH1} ${NS1} - test_exit $retval 0 diff --git a/tools/testing/selftests/bpf/test_xsk_skb_bidirectional.sh b/tools/testing/selftests/bpf/test_xsk_skb_bidirectional.sh new file mode 100755 index 000000000000..8ae1b6694f02 --- /dev/null +++ b/tools/testing/selftests/bpf/test_xsk_skb_bidirectional.sh @@ -0,0 +1,20 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 +# Copyright(c) 2020 Intel Corporation. + +# See test_xsk_prerequisites.sh for detailed information on tests + +. xsk_prereqs.sh +. xsk_env.sh + +TEST_NAME="SKB BIDIRECTIONAL SOCKETS" + +vethXDPgeneric ${VETH0} ${VETH1} ${NS1} + +params=("-S" "-B") +execxdpxceiver params + +retval=$? +test_status $retval "${TEST_NAME}" + +test_exit $retval 0 diff --git a/tools/testing/selftests/bpf/xdpxceiver.c b/tools/testing/selftests/bpf/xdpxceiver.c index ba5de1ef9f64..277d9344ce34 100644 --- a/tools/testing/selftests/bpf/xdpxceiver.c +++ b/tools/testing/selftests/bpf/xdpxceiver.c @@ -29,6 +29,10 @@ * c. Socket Teardown * Create a Tx and a Rx socket, Tx from one socket, Rx on another. Destroy * both sockets, then repeat multiple times. Only nopoll mode is used + * d. Bi-directional sockets + * Configure sockets as bi-directional tx/rx sockets, sets up fill and + * completion rings on each socket, tx/rx in both directions. Only nopoll + * mode is used * * 2. AF_XDP DRV/Native mode * Works on any netdevice with XDP_REDIRECT support, driver dependent. Processes @@ -37,10 +41,11 @@ * a. nopoll * b. poll * c. Socket Teardown + * d. Bi-directional sockets * - Only copy mode is supported because veth does not currently support * zero-copy mode * - * Total tests: 6 + * Total tests: 8 * * Flow: * ----- @@ -100,8 +105,9 @@ static void __exit_with_error(int error, const char *file, const char *func, int #define exit_with_error(error) __exit_with_error(error, __FILE__, __func__, __LINE__)
#define print_ksft_result(void)\ - (ksft_test_result_pass("PASS: %s %s %s\n", uut ? "DRV" : "SKB", opt_poll ? "POLL" :\ - "NOPOLL", opt_teardown ? "Socket Teardown" : "")) + (ksft_test_result_pass("PASS: %s %s %s%s\n", uut ? "DRV" : "SKB", opt_poll ? "POLL" :\ + "NOPOLL", opt_teardown ? "Socket Teardown" : "",\ + opt_bidi ? "Bi-directional Sockets" : ""))
static void pthread_init_mutex(void) { @@ -307,8 +313,13 @@ static int xsk_configure_socket(struct ifobject *ifobject) cfg.xdp_flags = opt_xdp_flags; cfg.bind_flags = opt_xdp_bind_flags;
- rxr = (ifobject->fv.vector == rx) ? &ifobject->xsk->rx : NULL; - txr = (ifobject->fv.vector == tx) ? &ifobject->xsk->tx : NULL; + if (!opt_bidi) { + rxr = (ifobject->fv.vector == rx) ? &ifobject->xsk->rx : NULL; + txr = (ifobject->fv.vector == tx) ? &ifobject->xsk->tx : NULL; + } else { + rxr = &ifobject->xsk->rx; + txr = &ifobject->xsk->tx; + }
ret = xsk_socket__create(&ifobject->xsk->xsk, ifobject->ifname, opt_queue, ifobject->umem->umem, rxr, txr, &cfg); @@ -327,6 +338,7 @@ static struct option long_options[] = { {"xdp-native", no_argument, 0, 'N'}, {"copy", no_argument, 0, 'c'}, {"tear-down", no_argument, 0, 'T'}, + {"bidi", optional_argument, 0, 'B'}, {"debug", optional_argument, 0, 'D'}, {"tx-pkt-count", optional_argument, 0, 'C'}, {0, 0, 0, 0} @@ -344,6 +356,7 @@ static void usage(const char *prog) " -N, --xdp-native=n Enforce XDP DRV (native) mode\n" " -c, --copy Force copy mode\n" " -T, --tear-down Tear down sockets by repeatedly recreating them\n" + " -B, --bidi Bi-directional sockets test\n" " -D, --debug Debug mode - dump packets L2 - L5\n" " -C, --tx-pkt-count=n Number of packets to send\n"; ksft_print_msg(str, prog); @@ -434,7 +447,7 @@ static void parse_command_line(int argc, char **argv) opterr = 0;
for (;;) { - c = getopt_long(argc, argv, "i:q:pSNcTDC:", long_options, &option_index); + c = getopt_long(argc, argv, "i:q:pSNcTBDC:", long_options, &option_index);
if (c == -1) break; @@ -475,6 +488,9 @@ static void parse_command_line(int argc, char **argv) case 'T': opt_teardown = 1; break; + case 'B': + opt_bidi = 1; + break; case 'D': debug_pkt_dump = 1; break; @@ -784,22 +800,25 @@ static void *worker_testapp_validate(void *arg) struct generic_data *data = (struct generic_data *)malloc(sizeof(struct generic_data)); struct iphdr *ip_hdr = (struct iphdr *)(pkt_data + sizeof(struct ethhdr)); struct ethhdr *eth_hdr = (struct ethhdr *)pkt_data; - void *bufs; + void *bufs = NULL;
pthread_attr_setstacksize(&attr, THREAD_STACK);
- bufs = mmap(NULL, num_frames * XSK_UMEM__DEFAULT_FRAME_SIZE, - PROT_READ | PROT_WRITE, MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); - if (bufs == MAP_FAILED) - exit_with_error(errno); + if (!bidi_pass) { + bufs = mmap(NULL, num_frames * XSK_UMEM__DEFAULT_FRAME_SIZE, + PROT_READ | PROT_WRITE, MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); + if (bufs == MAP_FAILED) + exit_with_error(errno);
- if (strcmp(((struct ifobject *)arg)->nsname, "")) - switch_namespace(((struct ifobject *)arg)->ifdict_index); + if (strcmp(((struct ifobject *)arg)->nsname, "")) + switch_namespace(((struct ifobject *)arg)->ifdict_index); + }
if (((struct ifobject *)arg)->fv.vector == tx) { int spinningrxctr = 0;
- thread_common_ops(arg, bufs, &sync_mutex_tx, &spinning_tx); + if (!bidi_pass) + thread_common_ops(arg, bufs, &sync_mutex_tx, &spinning_tx);
while (atomic_load(&spinning_rx) && spinningrxctr < SOCK_RECONF_CTR) { spinningrxctr++; @@ -829,7 +848,8 @@ static void *worker_testapp_validate(void *arg) struct pollfd fds[MAX_SOCKS] = { }; int ret;
- thread_common_ops(arg, bufs, &sync_mutex_tx, &spinning_rx); + if (!bidi_pass) + thread_common_ops(arg, bufs, &sync_mutex_tx, &spinning_rx);
ksft_print_msg("Interface [%s] vector [Rx]\n", ((struct ifobject *)arg)->ifname); xsk_populate_fill_ring(((struct ifobject *)arg)->umem); @@ -868,8 +888,10 @@ static void *worker_testapp_validate(void *arg) ksft_print_msg("Destroying socket\n"); }
- xsk_socket__delete(((struct ifobject *)arg)->xsk->xsk); - (void)xsk_umem__delete(((struct ifobject *)arg)->umem->umem); + if (!opt_bidi || (opt_bidi && bidi_pass)) { + xsk_socket__delete(((struct ifobject *)arg)->xsk->xsk); + (void)xsk_umem__delete(((struct ifobject *)arg)->umem->umem); + } pthread_exit(NULL); }
@@ -878,11 +900,26 @@ static void testapp_validate(void) pthread_attr_init(&attr); pthread_attr_setstacksize(&attr, THREAD_STACK);
+ if (opt_bidi && bidi_pass) { + pthread_init_mutex(); + if (!switching_notify) { + ksft_print_msg("Switching Tx/Rx vectors\n"); + switching_notify++; + } + } + pthread_mutex_lock(&sync_mutex);
/*Spawn RX thread */ - if (pthread_create(&t0, &attr, worker_testapp_validate, (void *)ifdict[1])) - exit_with_error(errno); + if (!opt_bidi || (opt_bidi && !bidi_pass)) { + if (pthread_create(&t0, &attr, worker_testapp_validate, (void *)ifdict[1])) + exit_with_error(errno); + } else if (opt_bidi && bidi_pass) { + /*switch Tx/Rx vectors */ + ifdict[0]->fv.vector = rx; + if (pthread_create(&t0, &attr, worker_testapp_validate, (void *)ifdict[0])) + exit_with_error(errno); + }
struct timespec max_wait = { 0, 0 };
@@ -896,8 +933,15 @@ static void testapp_validate(void) pthread_mutex_unlock(&sync_mutex);
/*Spawn TX thread */ - if (pthread_create(&t1, &attr, worker_testapp_validate, (void *)ifdict[0])) - exit_with_error(errno); + if (!opt_bidi || (opt_bidi && !bidi_pass)) { + if (pthread_create(&t1, &attr, worker_testapp_validate, (void *)ifdict[0])) + exit_with_error(errno); + } else if (opt_bidi && bidi_pass) { + /*switch Tx/Rx vectors */ + ifdict[1]->fv.vector = tx; + if (pthread_create(&t1, &attr, worker_testapp_validate, (void *)ifdict[1])) + exit_with_error(errno); + }
pthread_join(t1, NULL); pthread_join(t0, NULL); @@ -911,18 +955,19 @@ static void testapp_validate(void) free(pkt_buf); }
- if (!opt_teardown) + if (!opt_teardown && !opt_bidi) print_ksft_result(); }
static void testapp_sockets(void) { - for (int i = 0; i < MAX_TEARDOWN_ITER; i++) { + for (int i = 0; i < (opt_teardown ? MAX_TEARDOWN_ITER : MAX_BIDI_ITER); i++) { pkt_counter = 0; prev_pkt = -1; sigvar = 0; ksft_print_msg("Creating socket\n"); testapp_validate(); + opt_bidi ? bidi_pass++ : bidi_pass; }
print_ksft_result(); @@ -991,7 +1036,14 @@ int main(int argc, char **argv)
ksft_set_plan(1);
- opt_teardown ? testapp_sockets() : testapp_validate(); + if (!opt_teardown && !opt_bidi) { + testapp_validate(); + } else if (opt_teardown && opt_bidi) { + ksft_test_result_fail("ERROR: parameters -T and -B cannot be used together\n"); + ksft_exit_xfail(); + } else { + testapp_sockets(); + }
for (int i = 0; i < MAX_INTERFACES; i++) free(ifdict[i]); diff --git a/tools/testing/selftests/bpf/xdpxceiver.h b/tools/testing/selftests/bpf/xdpxceiver.h index 9d2670f28d86..d6630a19140b 100644 --- a/tools/testing/selftests/bpf/xdpxceiver.h +++ b/tools/testing/selftests/bpf/xdpxceiver.h @@ -22,6 +22,7 @@ #define MAX_INTERFACES_NAMESPACE_CHARS 10 #define MAX_SOCKS 1 #define MAX_TEARDOWN_ITER 10 +#define MAX_BIDI_ITER 2 #define PKT_HDR_SIZE (sizeof(struct ethhdr) + sizeof(struct iphdr) + \ sizeof(struct udphdr)) #define MIN_PKT_SIZE 64 @@ -51,12 +52,15 @@ enum TESTS { u8 uut; u8 debug_pkt_dump; u32 num_frames; +u8 switching_notify; +u8 bidi_pass;
static u32 opt_xdp_flags = XDP_FLAGS_UPDATE_IF_NOEXIST; static int opt_queue; static int opt_pkt_count; static int opt_poll; static int opt_teardown; +static int opt_bidi; static u32 opt_xdp_bind_flags = XDP_USE_NEED_WAKEUP; static u8 pkt_data[XSK_UMEM__DEFAULT_FRAME_SIZE]; static u32 pkt_counter;
On 11/20/20 5:00 AM, Weqaar Janjua wrote:
Adds following tests:
AF_XDP SKB mode d. Bi-directional Sockets Configure sockets as bi-directional tx/rx sockets, sets up fill and completion rings on each socket, tx/rx in both directions. Only nopoll mode is used
AF_XDP DRV/Native mode d. Bi-directional Sockets
- Only copy mode is supported because veth does not currently support zero-copy mode
Signed-off-by: Weqaar Janjua weqaar.a.janjua@intel.com
tools/testing/selftests/bpf/Makefile | 4 +- .../bpf/test_xsk_drv_bidirectional.sh | 23 ++++ .../selftests/bpf/test_xsk_drv_teardown.sh | 3 - .../bpf/test_xsk_skb_bidirectional.sh | 20 ++++ tools/testing/selftests/bpf/xdpxceiver.c | 100 +++++++++++++----- tools/testing/selftests/bpf/xdpxceiver.h | 4 + 6 files changed, 126 insertions(+), 28 deletions(-) create mode 100755 tools/testing/selftests/bpf/test_xsk_drv_bidirectional.sh create mode 100755 tools/testing/selftests/bpf/test_xsk_skb_bidirectional.sh
diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile index 515b29d321d7..258bd72812e0 100644 --- a/tools/testing/selftests/bpf/Makefile +++ b/tools/testing/selftests/bpf/Makefile @@ -78,7 +78,9 @@ TEST_PROGS := test_kmod.sh \ test_xsk_drv_nopoll.sh \ test_xsk_drv_poll.sh \ test_xsk_skb_teardown.sh \
- test_xsk_drv_teardown.sh
- test_xsk_drv_teardown.sh \
- test_xsk_skb_bidirectional.sh \
- test_xsk_drv_bidirectional.sh
TEST_PROGS_EXTENDED := with_addr.sh \ with_tunnels.sh \ diff --git a/tools/testing/selftests/bpf/test_xsk_drv_bidirectional.sh b/tools/testing/selftests/bpf/test_xsk_drv_bidirectional.sh new file mode 100755 index 000000000000..d3a7e2934d83 --- /dev/null +++ b/tools/testing/selftests/bpf/test_xsk_drv_bidirectional.sh @@ -0,0 +1,23 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 +# Copyright(c) 2020 Intel Corporation.
+# See test_xsk_prerequisites.sh for detailed information on tests
+. xsk_prereqs.sh +. xsk_env.sh
+TEST_NAME="DRV BIDIRECTIONAL SOCKETS"
+vethXDPnative ${VETH0} ${VETH1} ${NS1}
+params=("-N" "-B") +execxdpxceiver params
+retval=$? +test_status $retval "${TEST_NAME}"
+# Must be called in the last test to execute +cleanup_exit ${VETH0} ${VETH1} ${NS1}
This also makes hard to run tests as users will not know this unless they are familiar with the details of the tests.
How about you have another scripts test_xsk.sh which includes all these individual tests and pull the above cleanup_exit into test_xsk.sh? User just need to run test_xsk.sh will be able to run all tests you implemented here.
+test_exit $retval 0 diff --git a/tools/testing/selftests/bpf/test_xsk_drv_teardown.sh b/tools/testing/selftests/bpf/test_xsk_drv_teardown.sh
[...]
On Fri, 20 Nov 2020 at 20:45, Yonghong Song yhs@fb.com wrote:
On 11/20/20 5:00 AM, Weqaar Janjua wrote:
Adds following tests:
AF_XDP SKB mode d. Bi-directional Sockets Configure sockets as bi-directional tx/rx sockets, sets up fill and completion rings on each socket, tx/rx in both directions. Only nopoll mode is used
AF_XDP DRV/Native mode d. Bi-directional Sockets
- Only copy mode is supported because veth does not currently support zero-copy mode
Signed-off-by: Weqaar Janjua weqaar.a.janjua@intel.com
tools/testing/selftests/bpf/Makefile | 4 +- .../bpf/test_xsk_drv_bidirectional.sh | 23 ++++ .../selftests/bpf/test_xsk_drv_teardown.sh | 3 - .../bpf/test_xsk_skb_bidirectional.sh | 20 ++++ tools/testing/selftests/bpf/xdpxceiver.c | 100 +++++++++++++----- tools/testing/selftests/bpf/xdpxceiver.h | 4 + 6 files changed, 126 insertions(+), 28 deletions(-) create mode 100755 tools/testing/selftests/bpf/test_xsk_drv_bidirectional.sh create mode 100755 tools/testing/selftests/bpf/test_xsk_skb_bidirectional.sh
diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile index 515b29d321d7..258bd72812e0 100644 --- a/tools/testing/selftests/bpf/Makefile +++ b/tools/testing/selftests/bpf/Makefile @@ -78,7 +78,9 @@ TEST_PROGS := test_kmod.sh \ test_xsk_drv_nopoll.sh \ test_xsk_drv_poll.sh \ test_xsk_skb_teardown.sh \
test_xsk_drv_teardown.sh
test_xsk_drv_teardown.sh \test_xsk_skb_bidirectional.sh \test_xsk_drv_bidirectional.shTEST_PROGS_EXTENDED := with_addr.sh \ with_tunnels.sh \
diff --git a/tools/testing/selftests/bpf/test_xsk_drv_bidirectional.sh b/tools/testing/selftests/bpf/test_xsk_drv_bidirectional.sh new file mode 100755 index 000000000000..d3a7e2934d83 --- /dev/null +++ b/tools/testing/selftests/bpf/test_xsk_drv_bidirectional.sh @@ -0,0 +1,23 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 +# Copyright(c) 2020 Intel Corporation.
+# See test_xsk_prerequisites.sh for detailed information on tests
+. xsk_prereqs.sh +. xsk_env.sh
+TEST_NAME="DRV BIDIRECTIONAL SOCKETS"
+vethXDPnative ${VETH0} ${VETH1} ${NS1}
+params=("-N" "-B") +execxdpxceiver params
+retval=$? +test_status $retval "${TEST_NAME}"
+# Must be called in the last test to execute +cleanup_exit ${VETH0} ${VETH1} ${NS1}
This also makes hard to run tests as users will not know this unless they are familiar with the details of the tests.
How about you have another scripts test_xsk.sh which includes all these individual tests and pull the above cleanup_exit into test_xsk.sh? User just need to run test_xsk.sh will be able to run all tests you implemented here.
This works, test_xsk_* >> test_xsk.sh, will ship out as v3.
+test_exit $retval 0 diff --git a/tools/testing/selftests/bpf/test_xsk_drv_teardown.sh b/tools/testing/selftests/bpf/test_xsk_drv_teardown.sh
[...]
On Sat, 21 Nov 2020 at 20:14, Weqaar Janjua weqaar.janjua@gmail.com wrote:
On Fri, 20 Nov 2020 at 20:45, Yonghong Song yhs@fb.com wrote:
On 11/20/20 5:00 AM, Weqaar Janjua wrote:
Adds following tests:
AF_XDP SKB mode d. Bi-directional Sockets Configure sockets as bi-directional tx/rx sockets, sets up fill and completion rings on each socket, tx/rx in both directions. Only nopoll mode is used
AF_XDP DRV/Native mode d. Bi-directional Sockets
- Only copy mode is supported because veth does not currently support zero-copy mode
Signed-off-by: Weqaar Janjua weqaar.a.janjua@intel.com
tools/testing/selftests/bpf/Makefile | 4 +- .../bpf/test_xsk_drv_bidirectional.sh | 23 ++++ .../selftests/bpf/test_xsk_drv_teardown.sh | 3 - .../bpf/test_xsk_skb_bidirectional.sh | 20 ++++ tools/testing/selftests/bpf/xdpxceiver.c | 100 +++++++++++++----- tools/testing/selftests/bpf/xdpxceiver.h | 4 + 6 files changed, 126 insertions(+), 28 deletions(-) create mode 100755 tools/testing/selftests/bpf/test_xsk_drv_bidirectional.sh create mode 100755 tools/testing/selftests/bpf/test_xsk_skb_bidirectional.sh
diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile index 515b29d321d7..258bd72812e0 100644 --- a/tools/testing/selftests/bpf/Makefile +++ b/tools/testing/selftests/bpf/Makefile @@ -78,7 +78,9 @@ TEST_PROGS := test_kmod.sh \ test_xsk_drv_nopoll.sh \ test_xsk_drv_poll.sh \ test_xsk_skb_teardown.sh \
test_xsk_drv_teardown.sh
test_xsk_drv_teardown.sh \test_xsk_skb_bidirectional.sh \test_xsk_drv_bidirectional.shTEST_PROGS_EXTENDED := with_addr.sh \ with_tunnels.sh \
diff --git a/tools/testing/selftests/bpf/test_xsk_drv_bidirectional.sh b/tools/testing/selftests/bpf/test_xsk_drv_bidirectional.sh new file mode 100755 index 000000000000..d3a7e2934d83 --- /dev/null +++ b/tools/testing/selftests/bpf/test_xsk_drv_bidirectional.sh @@ -0,0 +1,23 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 +# Copyright(c) 2020 Intel Corporation.
+# See test_xsk_prerequisites.sh for detailed information on tests
+. xsk_prereqs.sh +. xsk_env.sh
+TEST_NAME="DRV BIDIRECTIONAL SOCKETS"
+vethXDPnative ${VETH0} ${VETH1} ${NS1}
+params=("-N" "-B") +execxdpxceiver params
+retval=$? +test_status $retval "${TEST_NAME}"
+# Must be called in the last test to execute +cleanup_exit ${VETH0} ${VETH1} ${NS1}
This also makes hard to run tests as users will not know this unless they are familiar with the details of the tests.
How about you have another scripts test_xsk.sh which includes all these individual tests and pull the above cleanup_exit into test_xsk.sh? User just need to run test_xsk.sh will be able to run all tests you implemented here.
This works, test_xsk_* >> test_xsk.sh, will ship out as v3.
An issue with merging all tests in a single test_xsk.sh is reporting number of test failures, with this approach a single test status is printed by kselftest:
# PREREQUISITES: [ PASS ] # SKB NOPOLL: [ FAIL ] # SKB POLL: [ PASS ] ok 1 selftests: xsk-patch2: test_xsk.sh
This is due to the fact Makefile has one TEST_PROGS = test_xsk.sh (thus kselftest considers it one test?), where in the original approach all tests have separate TEST_PROGS .sh which makes reporting match each test and status. This can be a problem for automation.
An alternative would be to exit each test with failure status but then the tests will stop execution at the failed test without executing the rest of xsk tests, which we probably wouldn't want.
Suggestions please?
+test_exit $retval 0 diff --git a/tools/testing/selftests/bpf/test_xsk_drv_teardown.sh b/tools/testing/selftests/bpf/test_xsk_drv_teardown.sh
[...]
On 11/24/20 7:11 AM, Weqaar Janjua wrote:
On Sat, 21 Nov 2020 at 20:14, Weqaar Janjua weqaar.janjua@gmail.com wrote:
On Fri, 20 Nov 2020 at 20:45, Yonghong Song yhs@fb.com wrote:
On 11/20/20 5:00 AM, Weqaar Janjua wrote:
Adds following tests:
AF_XDP SKB mode d. Bi-directional Sockets Configure sockets as bi-directional tx/rx sockets, sets up fill and completion rings on each socket, tx/rx in both directions. Only nopoll mode is used
AF_XDP DRV/Native mode d. Bi-directional Sockets
- Only copy mode is supported because veth does not currently support zero-copy mode
Signed-off-by: Weqaar Janjua weqaar.a.janjua@intel.com
tools/testing/selftests/bpf/Makefile | 4 +- .../bpf/test_xsk_drv_bidirectional.sh | 23 ++++ .../selftests/bpf/test_xsk_drv_teardown.sh | 3 - .../bpf/test_xsk_skb_bidirectional.sh | 20 ++++ tools/testing/selftests/bpf/xdpxceiver.c | 100 +++++++++++++----- tools/testing/selftests/bpf/xdpxceiver.h | 4 + 6 files changed, 126 insertions(+), 28 deletions(-) create mode 100755 tools/testing/selftests/bpf/test_xsk_drv_bidirectional.sh create mode 100755 tools/testing/selftests/bpf/test_xsk_skb_bidirectional.sh
diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile index 515b29d321d7..258bd72812e0 100644 --- a/tools/testing/selftests/bpf/Makefile +++ b/tools/testing/selftests/bpf/Makefile @@ -78,7 +78,9 @@ TEST_PROGS := test_kmod.sh \ test_xsk_drv_nopoll.sh \ test_xsk_drv_poll.sh \ test_xsk_skb_teardown.sh \
test_xsk_drv_teardown.sh
test_xsk_drv_teardown.sh \test_xsk_skb_bidirectional.sh \test_xsk_drv_bidirectional.shTEST_PROGS_EXTENDED := with_addr.sh \ with_tunnels.sh \
diff --git a/tools/testing/selftests/bpf/test_xsk_drv_bidirectional.sh b/tools/testing/selftests/bpf/test_xsk_drv_bidirectional.sh new file mode 100755 index 000000000000..d3a7e2934d83 --- /dev/null +++ b/tools/testing/selftests/bpf/test_xsk_drv_bidirectional.sh @@ -0,0 +1,23 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 +# Copyright(c) 2020 Intel Corporation.
+# See test_xsk_prerequisites.sh for detailed information on tests
+. xsk_prereqs.sh +. xsk_env.sh
+TEST_NAME="DRV BIDIRECTIONAL SOCKETS"
+vethXDPnative ${VETH0} ${VETH1} ${NS1}
+params=("-N" "-B") +execxdpxceiver params
+retval=$? +test_status $retval "${TEST_NAME}"
+# Must be called in the last test to execute +cleanup_exit ${VETH0} ${VETH1} ${NS1}
This also makes hard to run tests as users will not know this unless they are familiar with the details of the tests.
How about you have another scripts test_xsk.sh which includes all these individual tests and pull the above cleanup_exit into test_xsk.sh? User just need to run test_xsk.sh will be able to run all tests you implemented here.
This works, test_xsk_* >> test_xsk.sh, will ship out as v3.
An issue with merging all tests in a single test_xsk.sh is reporting number of test failures, with this approach a single test status is printed by kselftest:
# PREREQUISITES: [ PASS ] # SKB NOPOLL: [ FAIL ] # SKB POLL: [ PASS ] ok 1 selftests: xsk-patch2: test_xsk.sh
This is due to the fact Makefile has one TEST_PROGS = test_xsk.sh (thus kselftest considers it one test?), where in the original approach all tests have separate TEST_PROGS .sh which makes reporting match each test and status. This can be a problem for automation.
An alternative would be to exit each test with failure status but then the tests will stop execution at the failed test without executing the rest of xsk tests, which we probably wouldn't want.
Suggestions please?
I think it is okay to put everything xsk related to one test. If later on the test becomes more complex, you can have test_xsk_<1>.sh test_xsk_<2>.sh etc. But each .sh should be able to run independently without any particular order.
You can have subtests inside the .sh file. See test_offload.py as an example. You do not need to exit after one subtest fails, you can continue to run the next one. currently test_offload.py may exit when some subtest failed, but I think you don't have to.
+test_exit $retval 0 diff --git a/tools/testing/selftests/bpf/test_xsk_drv_teardown.sh b/tools/testing/selftests/bpf/test_xsk_drv_teardown.sh
[...]
On Tue, 24 Nov 2020 at 17:10, Yonghong Song yhs@fb.com wrote:
On 11/24/20 7:11 AM, Weqaar Janjua wrote:
On Sat, 21 Nov 2020 at 20:14, Weqaar Janjua weqaar.janjua@gmail.com wrote:
On Fri, 20 Nov 2020 at 20:45, Yonghong Song yhs@fb.com wrote:
On 11/20/20 5:00 AM, Weqaar Janjua wrote:
Adds following tests:
AF_XDP SKB mode d. Bi-directional Sockets Configure sockets as bi-directional tx/rx sockets, sets up fill and completion rings on each socket, tx/rx in both directions. Only nopoll mode is used
AF_XDP DRV/Native mode d. Bi-directional Sockets
- Only copy mode is supported because veth does not currently support zero-copy mode
Signed-off-by: Weqaar Janjua weqaar.a.janjua@intel.com
tools/testing/selftests/bpf/Makefile | 4 +- .../bpf/test_xsk_drv_bidirectional.sh | 23 ++++ .../selftests/bpf/test_xsk_drv_teardown.sh | 3 - .../bpf/test_xsk_skb_bidirectional.sh | 20 ++++ tools/testing/selftests/bpf/xdpxceiver.c | 100 +++++++++++++----- tools/testing/selftests/bpf/xdpxceiver.h | 4 + 6 files changed, 126 insertions(+), 28 deletions(-) create mode 100755 tools/testing/selftests/bpf/test_xsk_drv_bidirectional.sh create mode 100755 tools/testing/selftests/bpf/test_xsk_skb_bidirectional.sh
diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile index 515b29d321d7..258bd72812e0 100644 --- a/tools/testing/selftests/bpf/Makefile +++ b/tools/testing/selftests/bpf/Makefile @@ -78,7 +78,9 @@ TEST_PROGS := test_kmod.sh \ test_xsk_drv_nopoll.sh \ test_xsk_drv_poll.sh \ test_xsk_skb_teardown.sh \
test_xsk_drv_teardown.sh
test_xsk_drv_teardown.sh \test_xsk_skb_bidirectional.sh \test_xsk_drv_bidirectional.shTEST_PROGS_EXTENDED := with_addr.sh \ with_tunnels.sh \
diff --git a/tools/testing/selftests/bpf/test_xsk_drv_bidirectional.sh b/tools/testing/selftests/bpf/test_xsk_drv_bidirectional.sh new file mode 100755 index 000000000000..d3a7e2934d83 --- /dev/null +++ b/tools/testing/selftests/bpf/test_xsk_drv_bidirectional.sh @@ -0,0 +1,23 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 +# Copyright(c) 2020 Intel Corporation.
+# See test_xsk_prerequisites.sh for detailed information on tests
+. xsk_prereqs.sh +. xsk_env.sh
+TEST_NAME="DRV BIDIRECTIONAL SOCKETS"
+vethXDPnative ${VETH0} ${VETH1} ${NS1}
+params=("-N" "-B") +execxdpxceiver params
+retval=$? +test_status $retval "${TEST_NAME}"
+# Must be called in the last test to execute +cleanup_exit ${VETH0} ${VETH1} ${NS1}
This also makes hard to run tests as users will not know this unless they are familiar with the details of the tests.
How about you have another scripts test_xsk.sh which includes all these individual tests and pull the above cleanup_exit into test_xsk.sh? User just need to run test_xsk.sh will be able to run all tests you implemented here.
This works, test_xsk_* >> test_xsk.sh, will ship out as v3.
An issue with merging all tests in a single test_xsk.sh is reporting number of test failures, with this approach a single test status is printed by kselftest:
# PREREQUISITES: [ PASS ] # SKB NOPOLL: [ FAIL ] # SKB POLL: [ PASS ] ok 1 selftests: xsk-patch2: test_xsk.sh
This is due to the fact Makefile has one TEST_PROGS = test_xsk.sh (thus kselftest considers it one test?), where in the original approach all tests have separate TEST_PROGS .sh which makes reporting match each test and status. This can be a problem for automation.
An alternative would be to exit each test with failure status but then the tests will stop execution at the failed test without executing the rest of xsk tests, which we probably wouldn't want.
Suggestions please?
I think it is okay to put everything xsk related to one test. If later on the test becomes more complex, you can have test_xsk_<1>.sh test_xsk_<2>.sh etc. But each .sh should be able to run independently without any particular order.
You can have subtests inside the .sh file. See test_offload.py as an example. You do not need to exit after one subtest fails, you can continue to run the next one. currently test_offload.py may exit when some subtest failed, but I think you don't have to.
ACK, I will go ahead and merge all test_xsk_*.sh into test_xsk.sh.
Just to clarify that all current xsk tests are independent, there is no subtest at present, and do not need to run in any order.
Thanks, /Weqaar
+test_exit $retval 0 diff --git a/tools/testing/selftests/bpf/test_xsk_drv_teardown.sh b/tools/testing/selftests/bpf/test_xsk_drv_teardown.sh
[...]
On 11/20/20 5:00 AM, Weqaar Janjua wrote:
This patch set adds AF_XDP selftests based on veth to selftests/bpf.
# Topology: # --------- # ----------- # _ | Process | _ # / ----------- \ # / | \ # / | \ # ----------- | ----------- # | Thread1 | | | Thread2 | # ----------- | ----------- # | | | # ----------- | ----------- # | xskX | | | xskY | # ----------- | ----------- # | | | # ----------- | ---------- # | vethX | --------- | vethY | # ----------- peer ---------- # | | | # namespaceX | namespaceY
These selftests test AF_XDP SKB and Native/DRV modes using veth Virtual Ethernet interfaces.
The test program contains two threads, each thread is single socket with a unique UMEM. It validates in-order packet delivery and packet content by sending packets to each other.
Prerequisites setup by script test_xsk_prerequisites.sh:
Set up veth interfaces as per the topology shown ^^: * setup two veth interfaces and one namespace ** veth<xxxx> in root namespace ** veth<yyyy> in af_xdp<xxxx> namespace ** namespace af_xdp<xxxx> * create a spec file veth.spec that includes this run-time configuration that is read by test scripts - filenames prefixed with test_xsk_ *** xxxx and yyyy are randomly generated 4 digit numbers used to avoid conflict with any existing interfaceThe following tests are provided:
AF_XDP SKB mode Generic mode XDP is driver independent, used when the driver does not have support for XDP. Works on any netdevice using sockets and generic XDP path. XDP hook from netif_receive_skb(). a. nopoll - soft-irq processing b. poll - using poll() syscall c. Socket Teardown Create a Tx and a Rx socket, Tx from one socket, Rx on another. Destroy both sockets, then repeat multiple times. Only nopoll mode is used d. Bi-directional Sockets Configure sockets as bi-directional tx/rx sockets, sets up fill and completion rings on each socket, tx/rx in both directions. Only nopoll mode is used
AF_XDP DRV/Native mode Works on any netdevice with XDP_REDIRECT support, driver dependent. Processes packets before SKB allocation. Provides better performance than SKB. Driver hook available just after DMA of buffer descriptor. a. nopoll b. poll c. Socket Teardown d. Bi-directional Sockets
- Only copy mode is supported because veth does not currently support zero-copy mode
Total tests: 8
Flow:
- Single process spawns two threads: Tx and Rx
- Each of these two threads attach to a veth interface within their assigned namespaces
- Each thread creates one AF_XDP socket connected to a unique umem for each veth interface
- Tx thread transmits 10k packets from veth<xxxx> to veth<yyyy>
- Rx thread verifies if all 10k packets were received and delivered in-order, and have the right content
v2 changes:
- Move selftests/xsk to selftests/bpf
- Remove Makefiles under selftests/xsk, and utilize selftests/bpf/Makefile
Structure of the patch set:
Patch 1: This patch adds XSK Selftests framework under selftests/bpf Patch 2: Adds tests: SKB poll and nopoll mode, and mac-ip-udp debug Patch 3: Adds tests: DRV poll and nopoll mode Patch 4: Adds tests: SKB and DRV Socket Teardown Patch 5: Adds tests: SKB and DRV Bi-directional Sockets
I just want to report that after applying the above 5 patches on top of bpf-next commit 450d060e8f75 ("bpftool: Add {i,d}tlb_misses support for bpftool profile"), I hit the following error with below command sequences:
$ ./test_xsk_prerequisites.sh $ ./test_xsk_skb_poll.sh # Interface found: ve1480 # Interface found: ve9258 # NS switched: af_xdp9258 1..1 # Interface [ve9258] vector [Rx] # Interface [ve1480] vector [Tx] # Sending 10000 packets on interface ve1480 [ 331.741244] ------------[ cut here ]------------ [ 331.741741] kernel BUG at net/core/skbuff.c:1621! [ 331.742265] invalid opcode: 0000 [#1] PREEMPT SMP PTI [ 331.742837] CPU: 0 PID: 1883 Comm: xdpxceiver Not tainted 5.10.0-rc3+ #1037 [ 331.743468] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.9.3 -1.el7.centos 04/01/2014 [ 331.744300] RIP: 0010:pskb_expand_head+0x27b/0x310 [ 331.744747] Code: df e8 69 fc ff ff e9 ab fe ff ff 44 2b 6c 24 04 44 01 ab d0 00 00 00 48 83 c4 08 31 c0 5b 5d 41 5c 41 5d 41 5e 41 5f c3 0f 0b <0f> 0b be 02 00 00 00 e8 89 4e ad ff 8b 83 bc 00 00 00 48 03 83 c0 [ 331.746414] RSP: 0018:ffffbae4c0003d08 EFLAGS: 00010202 [ 331.746892] RAX: 000000000000013f RBX: ffff9e0a8367ad00 RCX: 0000000000000a20 [ 331.747534] RDX: 0000000000000002 RSI: 0000000000000100 RDI: ffff9e0a8367ad00 [ 331.748192] RBP: ffffbae4c00b2000 R08: 0000000000000001 R09: 000000000000000e [ 331.748834] R10: ffffbae4c0003eb8 R11: 00000000ef974e19 R12: ffff9e0a86ecf000 [ 331.749472] R13: 0000000000000001 R14: ffff9e0a8367ad00 R15: ffff9e0a8367ad00 [ 331.750119] FS: 00007ff0806c5e00(0000) GS:ffff9e0abae00000(0000) knlGS:00000 00000000000 [ 331.750848] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 331.751379] CR2: 00007ff0806c01d8 CR3: 0000000106e00006 CR4: 0000000000370ef0 [ 331.752022] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 331.752665] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 331.753307] Call Trace: [ 331.753535] <IRQ> [ 331.753736] do_xdp_generic.part.157+0xa3/0x550 [ 331.754151] __netif_receive_skb_core+0x67e/0x12b0 [ 331.754588] ? process_backlog+0x86/0x250 [ 331.754961] ? __netif_receive_skb_one_core+0x3c/0xa0 [ 331.755419] __netif_receive_skb_one_core+0x3c/0xa0 [ 331.755865] process_backlog+0xf5/0x250 [ 331.756215] net_rx_action+0x144/0x440 [ 331.756559] __do_softirq+0xe4/0x493 [ 331.756894] asm_call_irq_on_stack+0x12/0x20 [ 331.757282] </IRQ> [ 331.757478] ? dev_direct_xmit+0x1e8/0x230 [ 331.757856] do_softirq_own_stack+0x81/0xa0 [ 331.758244] do_softirq.part.16+0x3c/0x80 [ 331.758611] __local_bh_enable_ip+0xda/0xe0 [ 331.758995] dev_direct_xmit+0x20d/0x230 [ 331.759356] __xsk_sendmsg+0x314/0x3d0 [ 331.759704] sock_sendmsg+0x5b/0x60 [ 331.760025] __sys_sendto+0xf1/0x160 [ 331.760355] ? lockdep_hardirqs_on+0xbf/0x130 [ 331.760759] ? syscall_enter_from_user_mode+0x1c/0x50 [ 331.761216] __x64_sys_sendto+0x24/0x30 [ 331.761563] do_syscall_64+0x33/0x40 [ 331.761895] entry_SYSCALL_64_after_hwframe+0x44/0xa9 [ 331.762357] RIP: 0033:0x7ff08c8b9633 [ 331.762689] Code: 49 89 ca b8 2c 00 00 00 0f 05 48 3d 01 f0 ff ff 73 34 c3 48 83 ec 08 e8 1b f7 ff ff 48 89 04 24 49 89 ca b8 2c 00 00 00 0f 05 <48> 8b 3c 24 48 89 c2 e8 61 f7 ff ff 48 89 d0 48 83 c4 08 48 3d 01 [ 331.764356] RSP: 002b:00007ff0806c55c0 EFLAGS: 00000293 ORIG_RAX: 00000000000 0002c [ 331.765038] RAX: ffffffffffffffda RBX: 00007ff0780009b0 RCX: 00007ff08c8b9633 [ 331.765684] RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000008 [ 331.766324] RBP: 0000000000000040 R08: 0000000000000000 R09: 0000000000000000 [ 331.766969] R10: 0000000000000040 R11: 0000000000000293 R12: 0000000000000040 [ 331.767608] R13: 0000000000000040 R14: 0000000000000000 R15: 0000000000cd4030 [ 331.768261] Modules linked in: [ 331.768596] ---[ end trace d9ca37a7957928dd ]--- [ 331.769126] RIP: 0010:pskb_expand_head+0x27b/0x310 [ 331.769678] Code: df e8 69 fc ff ff e9 ab fe ff ff 44 2b 6c 24 04 44 01 ab d0 00 00 00 48 83 c4 08 31 c0 5b 5d 41 5c 41 5d 41 5e 41 5f c3 0f 0b <0f> 0b be 02 00 00 00 e8 89 4e ad ff 8b 83 bc 00 00 00 48 03 83 c0 [ 331.771459] RSP: 0018:ffffbae4c0003d08 EFLAGS: 00010202 [ 331.772043] RAX: 000000000000013f RBX: ffff9e0a8367ad00 RCX: 0000000000000a20 [ 331.772784] RDX: 0000000000000002 RSI: 0000000000000100 RDI: ffff9e0a8367ad00 [ 331.773526] RBP: ffffbae4c00b2000 R08: 0000000000000001 R09: 000000000000000e [ 331.774293] R10: ffffbae4c0003eb8 R11: 00000000ef974e19 R12: ffff9e0a86ecf000 [ 331.775049] R13: 0000000000000001 R14: ffff9e0a8367ad00 R15: ffff9e0a8367ad00 [ 331.775901] FS: 00007ff0806c5e00(0000) GS:ffff9e0abae00000(0000) knlGS:00000 00000000000 [ 331.776809] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 331.777455] CR2: 00007ff0806c01d8 CR3: 0000000106e00006 CR4: 0000000000370ef0 [ 331.778232] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 331.778989] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 331.779739] Kernel panic - not syncing: Fatal exception in interrupt [ 331.780523] Kernel Offset: 0xb800000 from 0xffffffff81000000 (relocation rang e: 0xffffffff80000000-0xffffffffbfffffff) [ 331.781488] ---[ end Kernel panic - not syncing: Fatal exception in interrupt ]---
In any case, kernel should not panic. You or somebody familiar with xsk may want to take a look.
Thanks: Weqaar
Weqaar Janjua (5): selftests/bpf: xsk selftests framework selftests/bpf: xsk selftests - SKB POLL, NOPOLL selftests/bpf: xsk selftests - DRV POLL, NOPOLL selftests/bpf: xsk selftests - Socket Teardown - SKB, DRV selftests/bpf: xsk selftests - Bi-directional Sockets - SKB, DRV
tools/testing/selftests/bpf/Makefile | 15 +- .../bpf/test_xsk_drv_bidirectional.sh | 23 + .../selftests/bpf/test_xsk_drv_nopoll.sh | 20 + .../selftests/bpf/test_xsk_drv_poll.sh | 20 + .../selftests/bpf/test_xsk_drv_teardown.sh | 20 + .../selftests/bpf/test_xsk_prerequisites.sh | 127 ++ .../bpf/test_xsk_skb_bidirectional.sh | 20 + .../selftests/bpf/test_xsk_skb_nopoll.sh | 20 + .../selftests/bpf/test_xsk_skb_poll.sh | 20 + .../selftests/bpf/test_xsk_skb_teardown.sh | 20 + tools/testing/selftests/bpf/xdpxceiver.c | 1056 +++++++++++++++++ tools/testing/selftests/bpf/xdpxceiver.h | 158 +++ tools/testing/selftests/bpf/xsk_env.sh | 28 + tools/testing/selftests/bpf/xsk_prereqs.sh | 119 ++ 14 files changed, 1664 insertions(+), 2 deletions(-)
[...]
On 2020-11-21 01:31, Yonghong Song wrote:
On 11/20/20 5:00 AM, Weqaar Janjua wrote:
This patch set adds AF_XDP selftests based on veth to selftests/bpf.
# Topology: # --------- # ----------- # _ | Process | _ # / ----------- \ # / | \ # / | \ # ----------- | ----------- # | Thread1 | | | Thread2 | # ----------- | ----------- # | | | # ----------- | ----------- # | xskX | | | xskY | # ----------- | ----------- # | | | # ----------- | ---------- # | vethX | --------- | vethY | # ----------- peer ---------- # | | | # namespaceX | namespaceY
These selftests test AF_XDP SKB and Native/DRV modes using veth Virtual Ethernet interfaces.
The test program contains two threads, each thread is single socket with a unique UMEM. It validates in-order packet delivery and packet content by sending packets to each other.
Prerequisites setup by script test_xsk_prerequisites.sh:
Set up veth interfaces as per the topology shown ^^: * setup two veth interfaces and one namespace ** veth<xxxx> in root namespace ** veth<yyyy> in af_xdp<xxxx> namespace ** namespace af_xdp<xxxx> * create a spec file veth.spec that includes this run-time configuration that is read by test scripts - filenames prefixed with test_xsk_ *** xxxx and yyyy are randomly generated 4 digit numbers used to avoid conflict with any existing interface
The following tests are provided:
- AF_XDP SKB mode
Generic mode XDP is driver independent, used when the driver does not have support for XDP. Works on any netdevice using sockets and generic XDP path. XDP hook from netif_receive_skb(). a. nopoll - soft-irq processing b. poll - using poll() syscall c. Socket Teardown Create a Tx and a Rx socket, Tx from one socket, Rx on another. Destroy both sockets, then repeat multiple times. Only nopoll mode is used d. Bi-directional Sockets Configure sockets as bi-directional tx/rx sockets, sets up fill and completion rings on each socket, tx/rx in both directions. Only nopoll mode is used
- AF_XDP DRV/Native mode
Works on any netdevice with XDP_REDIRECT support, driver dependent. Processes packets before SKB allocation. Provides better performance than SKB. Driver hook available just after DMA of buffer descriptor. a. nopoll b. poll c. Socket Teardown d. Bi-directional Sockets * Only copy mode is supported because veth does not currently support zero-copy mode
Total tests: 8
Flow:
- Single process spawns two threads: Tx and Rx
- Each of these two threads attach to a veth interface within their
assigned namespaces
- Each thread creates one AF_XDP socket connected to a unique umem
for each veth interface
- Tx thread transmits 10k packets from veth<xxxx> to veth<yyyy>
- Rx thread verifies if all 10k packets were received and delivered
in-order, and have the right content
v2 changes:
- Move selftests/xsk to selftests/bpf
- Remove Makefiles under selftests/xsk, and utilize
selftests/bpf/Makefile
Structure of the patch set:
Patch 1: This patch adds XSK Selftests framework under selftests/bpf Patch 2: Adds tests: SKB poll and nopoll mode, and mac-ip-udp debug Patch 3: Adds tests: DRV poll and nopoll mode Patch 4: Adds tests: SKB and DRV Socket Teardown Patch 5: Adds tests: SKB and DRV Bi-directional Sockets
I just want to report that after applying the above 5 patches on top of bpf-next commit 450d060e8f75 ("bpftool: Add {i,d}tlb_misses support for bpftool profile"), I hit the following error with below command sequences:
$ ./test_xsk_prerequisites.sh $ ./test_xsk_skb_poll.sh # Interface found: ve1480 # Interface found: ve9258 # NS switched: af_xdp9258 1..1 # Interface [ve9258] vector [Rx] # Interface [ve1480] vector [Tx] # Sending 10000 packets on interface ve1480 [ 331.741244] ------------[ cut here ]------------ [ 331.741741] kernel BUG at net/core/skbuff.c:1621! [ 331.742265] invalid opcode: 0000 [#1] PREEMPT SMP PTI [ 331.742837] CPU: 0 PID: 1883 Comm: xdpxceiver Not tainted 5.10.0-rc3+ #1037 [ 331.743468] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.9.3 -1.el7.centos 04/01/2014 [ 331.744300] RIP: 0010:pskb_expand_head+0x27b/0x310
Ugh, looks like the tests are working. :-P
This is a BUG_ON(skb_shared(skb)) trigger, related to the skbuff refcount changes done recently in AF_XDP.
I'll cook a patch! Thanks for the report!
Björn
On 2020-11-23 13:20, Björn Töpel wrote:
On 2020-11-21 01:31, Yonghong Song wrote:
On 11/20/20 5:00 AM, Weqaar Janjua wrote:
This patch set adds AF_XDP selftests based on veth to selftests/bpf.
# Topology: # --------- # ----------- # _ | Process | _ # / ----------- \ # / | \ # / | \ # ----------- | ----------- # | Thread1 | | | Thread2 | # ----------- | ----------- # | | | # ----------- | ----------- # | xskX | | | xskY | # ----------- | ----------- # | | | # ----------- | ---------- # | vethX | --------- | vethY | # ----------- peer ---------- # | | | # namespaceX | namespaceY
These selftests test AF_XDP SKB and Native/DRV modes using veth Virtual Ethernet interfaces.
The test program contains two threads, each thread is single socket with a unique UMEM. It validates in-order packet delivery and packet content by sending packets to each other.
Prerequisites setup by script test_xsk_prerequisites.sh:
Set up veth interfaces as per the topology shown ^^: * setup two veth interfaces and one namespace ** veth<xxxx> in root namespace ** veth<yyyy> in af_xdp<xxxx> namespace ** namespace af_xdp<xxxx> * create a spec file veth.spec that includes this run-time configuration that is read by test scripts - filenames prefixed with test_xsk_ *** xxxx and yyyy are randomly generated 4 digit numbers used to avoid conflict with any existing interface
The following tests are provided:
- AF_XDP SKB mode
Generic mode XDP is driver independent, used when the driver does not have support for XDP. Works on any netdevice using sockets and generic XDP path. XDP hook from netif_receive_skb(). a. nopoll - soft-irq processing b. poll - using poll() syscall c. Socket Teardown Create a Tx and a Rx socket, Tx from one socket, Rx on another. Destroy both sockets, then repeat multiple times. Only nopoll mode is used d. Bi-directional Sockets Configure sockets as bi-directional tx/rx sockets, sets up fill and completion rings on each socket, tx/rx in both directions. Only nopoll mode is used
- AF_XDP DRV/Native mode
Works on any netdevice with XDP_REDIRECT support, driver dependent. Processes packets before SKB allocation. Provides better performance than SKB. Driver hook available just after DMA of buffer descriptor. a. nopoll b. poll c. Socket Teardown d. Bi-directional Sockets * Only copy mode is supported because veth does not currently support zero-copy mode
Total tests: 8
Flow:
- Single process spawns two threads: Tx and Rx
- Each of these two threads attach to a veth interface within their
assigned namespaces
- Each thread creates one AF_XDP socket connected to a unique umem
for each veth interface
- Tx thread transmits 10k packets from veth<xxxx> to veth<yyyy>
- Rx thread verifies if all 10k packets were received and delivered
in-order, and have the right content
v2 changes:
- Move selftests/xsk to selftests/bpf
- Remove Makefiles under selftests/xsk, and utilize
selftests/bpf/Makefile
Structure of the patch set:
Patch 1: This patch adds XSK Selftests framework under selftests/bpf Patch 2: Adds tests: SKB poll and nopoll mode, and mac-ip-udp debug Patch 3: Adds tests: DRV poll and nopoll mode Patch 4: Adds tests: SKB and DRV Socket Teardown Patch 5: Adds tests: SKB and DRV Bi-directional Sockets
I just want to report that after applying the above 5 patches on top of bpf-next commit 450d060e8f75 ("bpftool: Add {i,d}tlb_misses support for bpftool profile"), I hit the following error with below command sequences:
$ ./test_xsk_prerequisites.sh $ ./test_xsk_skb_poll.sh # Interface found: ve1480 # Interface found: ve9258 # NS switched: af_xdp9258 1..1 # Interface [ve9258] vector [Rx] # Interface [ve1480] vector [Tx] # Sending 10000 packets on interface ve1480 [ 331.741244] ------------[ cut here ]------------ [ 331.741741] kernel BUG at net/core/skbuff.c:1621! [ 331.742265] invalid opcode: 0000 [#1] PREEMPT SMP PTI [ 331.742837] CPU: 0 PID: 1883 Comm: xdpxceiver Not tainted 5.10.0-rc3+ #1037 [ 331.743468] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.9.3 -1.el7.centos 04/01/2014 [ 331.744300] RIP: 0010:pskb_expand_head+0x27b/0x310
Ugh, looks like the tests are working. :-P
This is a BUG_ON(skb_shared(skb)) trigger, related to the skbuff refcount changes done recently in AF_XDP.
I'll cook a patch! Thanks for the report!
Posted a fix [1].
Please not that it's for the bpf tree, so when Weqaar pushes the v3 of the selftests to bpf-next, [1] needs to be pulled in.
Björn
[1] https://lore.kernel.org/bpf/20201123131215.136131-1-bjorn.topel@gmail.com/
Björn
linux-kselftest-mirror@lists.linaro.org
-
Björn Töpel -
Weqaar Janjua -
Yonghong Song