On Thu, Sep 20, 2018 at 05:50:39PM +0100, Kevin Traynor wrote:
On 09/18/2018 10:35 AM, Jens Freimann wrote:This adds a new forwarding mode to testpmd to simulate more realistic behavior of a guest machine engaged in receiving and sending packets performing Virtual Network Function (VNF).The goal is to enable a simple way of measuring performance impact on cache and memory footprint utilization from various VNF co-located on the same host machine. For this it does: * Buffer packets in a FIFO: Create a fifo to buffer received packets. Once it flows over put those packets into the actual tx queue. The fifo is created per tx queue and its size can be set with the --noisy-tx-sw-buffer-flushtime commandline parameter. A second commandline parameter is used to set a timeout in milliseconds after which the fifo is flushed. --noisy-tx-sw-buffer-size [packet numbers] Keep the mbuf in a FIFO and forward the over flooding packets from the FIFO. This queue is per TX-queue (after all other packet processing). --noisy-tx-sw-buffer-flushtime [delay] Flush the packet queue if no packets have been seen during [delay]. As long as packets are seen, the timer is reset. Add several options to simulate route lookups (memory reads) in tables that can be quite large, as well as route hit statistics update. These options simulates the while stack traversal and will trash the cache. Memory access is random. * simulate route lookups: Allocate a buffer and perform reads and writes on it as specified by commandline options: --noisy-lkup-memory [size] Size of the VNF internal memory (MB), in which the random read/write will be done, allocated by rte_malloc (hugepages). --noisy-lkup-num-writes [num] Number of random writes in memory per packet should be performed, simulating hit-flags update. 64 bits per write, all write in different cache lines. --noisy-lkup-num-reads [num] Number of random reads in memory per packet should be performed, simulating FIB/table lookups. 64 bits per read, all write in different cache lines. --noisy-lkup-num-reads-writes [num] Number of random reads and writes in memory per packet should be performed, simulating stats update. 64 bits per read-write, all reads and writes in different cache lines. Signed-off-by: Jens Freimann <jfreim...@redhat.com> ---Hi Jens, thanks for the new version. A small few remaining comments below, Kevin. <snip>+ +static void +noisy_fwd_begin(portid_t pi) +{ + struct noisy_config *n; + char name[NOISY_STRSIZE]; + + noisy_cfg[pi] = rte_zmalloc("testpmd noisy fifo and timers", + sizeof(struct noisy_config), + RTE_CACHE_LINE_SIZE); + if (noisy_cfg == NULL) {Looks like it should be 'if (noisy_cfg[pi] == NULL)'
yep, fixed.
+ rte_exit(EXIT_FAILURE, + "rte_zmalloc(%d) struct noisy_config) \ + failed\n", (int) pi); + } + n = noisy_cfg[pi]; + n->do_buffering = noisy_tx_sw_bufsz > 0; + n->do_sim = noisy_lkup_num_writes + noisy_lkup_num_reads + + noisy_lkup_num_reads_writes; + n->do_flush = noisy_tx_sw_buf_flush_time > 0; + + if (n->do_buffering) { + snprintf(name, NOISY_STRSIZE, NOISY_RING, pi); + n->f = rte_ring_create(name, noisy_tx_sw_bufsz, + rte_socket_id(), 0); + if (!n->f) + rte_exit(EXIT_FAILURE, + "rte_ring_create(%d), size %d) \ + failed\n", (int) pi, + noisy_tx_sw_bufsz); + } + if (noisy_lkup_mem_sz > 0) { + n->vnf_mem = (char *) rte_zmalloc("vnf sim memory", + noisy_lkup_mem_sz * 1024 * 1024, + RTE_CACHE_LINE_SIZE); + if (!n->vnf_mem) + rte_exit(EXIT_FAILURE, + "rte_zmalloc(%" PRIu64 ") for vnf \ + memory) failed\n", noisy_lkup_mem_sz); + } else if (n->do_sim) { + rte_exit(EXIT_FAILURE, "--noisy-lkup-memory-size \ + must be > 0\n"); + } +} + +struct fwd_engine noisy_vnf_engine = { + .fwd_mode_name = "noisy", + .port_fwd_begin = noisy_fwd_begin, + .port_fwd_end = noisy_fwd_end, + .packet_fwd = pkt_burst_noisy_vnf, +}; +new blank line at EOF. + warning: 1 line adds whitespace errors.
hmpf
diff --git a/app/test-pmd/parameters.c b/app/test-pmd/parameters.c index 9220e1c1b..3231b0c51 100644 --- a/app/test-pmd/parameters.c +++ b/app/test-pmd/parameters.c @@ -625,6 +625,12 @@ launch_args_parse(int argc, char** argv) { "vxlan-gpe-port", 1, 0, 0 }, { "mlockall", 0, 0, 0 }, { "no-mlockall", 0, 0, 0 }, + { "noisy-tx-sw-buffer-size", 1, 0, 0 }, + { "noisy-tx-sw-buffer-flushtime",1, 0, 0 }, + { "noisy-lkup-memory", 1, 0, 0 }, + { "noisy-lkup-num-writes", 1, 0, 0 }, + { "noisy-lkup-num-reads", 1, 0, 0 }, + { "noisy-lkup-num-reads-writes",1, 0, 0 }, { 0, 0, 0, 0 }, }; @@ -1145,6 +1151,60 @@ launch_args_parse(int argc, char** argv) do_mlockall = 1; if (!strcmp(lgopts[opt_idx].name, "no-mlockall")) do_mlockall = 0; + if (!strcmp(lgopts[opt_idx].name, + "noisy-tx-sw-buffer-size")) { + n = atoi(optarg); + if (n >= 0) + noisy_tx_sw_bufsz = n; + else + rte_exit(EXIT_FAILURE, + "noisy-tx-sw-buffer-size must be >= 0\n"); + } + if (!strcmp(lgopts[opt_idx].name, + "noisy-tx-sw-buffer-flushtime")) { + n = atoi(optarg); + if (n >= 0) + noisy_tx_sw_buf_flush_time = n; + else + rte_exit(EXIT_FAILURE, + "noisy-tx-sw-buffer-flushtime must be >= 0\n"); + } + if (!strcmp(lgopts[opt_idx].name, + "noisy-lkup-memory")) { + n = atoi(optarg); + if (n > 0)I thought this and below ones would also be '>=' also?
I wasn't sure if it's a good idea or if I should choose a different default... and then forgot about it. sorry. I checked that it's safe to allow 0, will keep the defaults at 0 and allow it to be set here. Thanks for the review Kevin! regards,Jens
