aboutsummaryrefslogtreecommitdiffstats
path: root/GPL/tools/trex
diff options
context:
space:
mode:
authorVratko Polak <vrpolak@cisco.com>2020-10-27 19:09:44 +0100
committerVratko Polak <vrpolak@cisco.com>2020-10-29 20:25:50 +0000
commit023fa41e51c966a1956bda6b915ffd894ff10e84 (patch)
treecdb96c99a8ade4855176c43969cbd9a06adf693b /GPL/tools/trex
parente31998ea56c55879fbaae8e58b0dad0bc6549dae (diff)
Support existing test types with ASTF
+ Add UDP_CPS, TCP_CPS, UDP_PPS and TCP_PPS suites. + Update existing cps traffic profiles. + Add missing traffic profiles. + UDP: + Single burst of 32 packets was confirmed as safe enough for TRex. + Maybe 64 could work, but not enough testing for that. + Multiple bursts have lead to reduced TRex performance, as overlaping bursts (from different client instances) tend to fill up the buffers. + TCP: + Data size set to 11111 bytes, completely arbitrarily. + Results look reasonable, so I have kept that. - MSS not set at all - No tested support for frame size other than 64B. - Frame size does not even factor into TCP profiles. + So other frame sizes are skipped in autogen. + Update tags in related suites. - HOSTS_{n} and SRC_USER_{n} should be unified. - Questionable clarification on difference between IP4BASE and SCALE. + Add NAT state resetters to tests that need them. + Resetter is called (if set) before each measurement. + If ramp-up is detected, resetter is not set. + Rename "mult" argument to "multiplier". + Abstracted from packets to transactions. + Transaction corresponds to profile. + TRex multiplier argument sets target rate in transactions per second. + The familiar STL traffic: + Bidirectional is considered to be 2 packets per transaction. + Unidirectional is considered to be 1 packet per transaction. + The newer ASTF traffic: + 4 subtypes, each has different number of packets per transaction. + For max rate computation: + Packets in the more numerous direction are considered. + Rely on TRex reported traffic duration for ASTF: + Use the server side value. - Client side value is higher by an overhead. - TRex is not sending traffic during that time. + Remove delays from traffic profiles. - Those delays would increase the reprted traffic time. + Support for scale lmited trials. + Only for ASTF profiles, each ASTF profile has limited scale. + Scale defined in suite variables. + For TRex to send all transactions provided duration value is ignored. + The appropriate value is computed in TrafficGenerator. + An ad-hoc time constant is added to match the TRex client side time overhead. + The profile driver receives the computed duration. + Measurement for PLRsearch add a sleep if the computed duration is smaller. + Alternative argument for search algos if scale is limited. + Both need higher timeout to accomodate big scales. + MLRsearch can afford fewer phases. + Added a parameter to optionally shorten the duration. + Use short duration for runtime stats trial and failure stats trial. + Use very large keepalive values in udp profiles to avoid ka packets. + No polling in ASTF profile driver. - Polling could eliminate the time overhead value. + But polling proved to introduce some loss, affecting the results. + Handle duration stretching in ASTF by stopping traffic. + The stop has several steps so that: + The traffic is really stopped entirely. + Late packets do not count (maybe as errors). + Stats are preserved to read for results (and cleared afterwards). + Several quantities added to ReceiveRateMeasurement: + Original target duration is preserved (algos need that). + Input estimate (tps) for early search iterations. + Output estimate (maybe pps) for MRR output. + Strict result (unsent counts as loss) for NDR. + Use L2 counters (opackets, ipackets) where possible. - TRex has trouble processing packets for the L7 ones at high loads. + Remove warmup from profile drivers and keywords. + Suites should call "Send ramp-up traffic" explicitly if needed. + Added parsing for few more counters. + Both to use in formulas or just for debug purposes. - Only 64B cases in autogen, framesize support to be added later. + Latency streams during search can be enabled via PERF_USE_LATENCY env var. + MLRsearch improvments: + Rename argument names to min_rate and max_rate. + Use relative receive rate in initial phase. + PLRsearch improvements: + Careful computation when output (pps) does not match input (tps). + Use geometric distribution (instead of Poisson). + Helps agains math errors. + This should improve estimate stability. - But in practice big losses still lead to significant jumps. + Traffic generator improvements: + send_traffic_on_tg now calls the full set_rate_provider_defaults. + _send_traffic_on_tg_internal for the logic without provider defaults. + As the internal function is re-used by measure() without affecting defaults. + Move _parse_traffic_results just before get_measurement_result. + As the latter uses fields set bu the former, it is now easier to read. + Multiple sources for approximate duration. + Tried from more precise to more available. + Includes logic for _pps tests (added in later change). + Move explicit type conversions to earlier occurences. + Profile driver output field uses semicolons to simplify parsing. + Performance Robot lib file split to several smaller ones. + performance_actions.robot: + Hosts Additional Statistics Action For * keywords. + performance_display.robot: + Hosts keyword for displaying and verifying results. + Change test message to use the correct unit (pps or cps). + performance_limits.robot renamed to performance_vars.robot + Added many keywords, mostly for accessing test variables. + Moved variables for Policer into a new keyword there. + Some keywords need sophisticated logic. - Other are basically Get Variable Value. + But in future more logic can be added, without editing callers. + Documentation for the new keywords acts as a documentation for test variables. + performance_utils.robot has the rest. + Eliminated arguments if the value is in test variable. + Small improvements to documentation. - Still not enough cleanup with respect to arguments and test variables. + Keywords are sorted alphabetically now in each one. + Suites: + Unified variables table: + No colons in comments. + ${n_hosts}, ${n_ports} and use them instead hardcoded numbers. + Add -cps to existing cps suite names. + Remove "trial data overwrite". + Compute max rate as in STL suites. + Each NAT suite has ip4base suite to compare results to. - Those act as indirect TRex calibration. - VPP does not lose packets in those. + Latency in ASTF suites is disabled hard. - As we do not support latency in ASTF profiles yet. + Unidirectional tests governed by suite variable, not an argument. + Write long argument lists vertically. + Prefer to use argument names. + In Python, also the last argument is followed by comma. + It makes renaming and reordering easier. + Similarly applies to prints with long lists of values. + A TODO to update api crc file comments. Change-Id: I84729355edbec051298a9de1162107f88ff5737d Signed-off-by: Vratko Polak <vrpolak@cisco.com>
Diffstat (limited to 'GPL/tools/trex')
-rw-r--r--GPL/tools/trex/trex_astf_profile.py247
-rw-r--r--GPL/tools/trex/trex_astf_stop.py13
-rw-r--r--GPL/tools/trex/trex_stl_profile.py96
-rw-r--r--GPL/tools/trex/trex_stl_stop.py15
4 files changed, 173 insertions, 198 deletions
diff --git a/GPL/tools/trex/trex_astf_profile.py b/GPL/tools/trex/trex_astf_profile.py
index 0542486105..63831e4bf4 100644
--- a/GPL/tools/trex/trex_astf_profile.py
+++ b/GPL/tools/trex/trex_astf_profile.py
@@ -15,7 +15,7 @@
"""This module gets T-Rex advanced stateful (astf) traffic profile together
with other parameters, reads the profile and sends the traffic. At the end, it
-measures the packet loss and latency.
+parses for various counters.
"""
import argparse
@@ -60,8 +60,16 @@ def fmt_latency(lat_min, lat_avg, lat_max, hdrh):
def simple_burst(
- profile_file, duration, framesize, mult, warmup_time, port_0, port_1,
- latency, async_start=False, traffic_directions=2):
+ profile_file,
+ duration,
+ framesize,
+ multiplier,
+ port_0,
+ port_1,
+ latency,
+ async_start=False,
+ traffic_directions=2,
+ ):
"""Send traffic and measure packet loss and latency.
Procedure:
@@ -70,20 +78,36 @@ def simple_burst(
- resets the ports,
- removes all existing streams,
- adds streams from the traffic profile to the ports,
- - if the warm-up time is more than 0, sends the warm-up traffic, reads the
- statistics,
- clears the statistics from the client,
- starts the traffic,
- waits for the defined time (or runs forever if async mode is defined),
- - stops the traffic,
+ - explicitly stops the traffic,
- reads and displays the statistics and
- disconnects from the client.
+ Duration details:
+ Contrary to stateless mode, ASTF profiles typically limit the number
+ of flows/transactions that can happen.
+ The caller is expected to set the duration parameter accordingly to
+ this limit and multiplier, including any overheads.
+ See *_traffic_duration output fields for TRex's measurement
+ of the real traffic duration (should be without any inactivity overheads).
+ If traffic has not ended by the final time, the traffic
+ is stopped explicitly, counters reflect the state just after the stop.
+
+ TODO: Support tests which focus only on some transaction phases,
+ e.g. TCP tests ignoring init and teardown separated by delays.
+ Currently, approximated time measures the whole traffic duration.
+
:param profile_file: A python module with T-rex traffic profile.
- :param duration: Duration of traffic run in seconds (-1=infinite).
+ :param duration: Expected duration for all transactions to finish,
+ assuming only tolerable duration stretching happens.
+ This includes later start of later transactions
+ (according to TPS multiplier) and expected duration of each transaction.
+ Critically, this also includes any delay TRex shows when starting
+ traffic (but not the similar delay during stopping).
:param framesize: Frame size.
- :param mult: Multiplier of profile CPS.
- :param warmup_time: Traffic warm-up time in seconds, 0 = disable.
+ :param multiplier: Multiplier of profile CPS.
:param port_0: Port 0 on the traffic generator.
:param port_1: Port 1 on the traffic generator.
:param latency: With latency stats.
@@ -92,8 +116,7 @@ def simple_burst(
:type profile_file: str
:type duration: float
:type framesize: int or str
- :type mult: int
- :type warmup_time: float
+ :type multiplier: int
:type port_0: int
:type port_1: int
:type latency: bool
@@ -101,7 +124,7 @@ def simple_burst(
:type traffic_directions: int
"""
client = None
- total_rcvd = 0
+ total_received = 0
total_sent = 0
lost_a = 0
lost_b = 0
@@ -111,7 +134,6 @@ def simple_burst(
lat_b_hist = u""
l7_data = u""
stats = dict()
- stats_sampling = 1.0
approximated_duration = 0
# Read the profile.
@@ -139,37 +161,6 @@ def simple_burst(
if traffic_directions > 1:
ports.append(port_1)
- # Warm-up phase.
- if warmup_time > 0:
- # Clear the stats before injecting.
- client.clear_stats()
- # Choose CPS and start traffic.
- client.start(mult=mult, duration=warmup_time)
- time_start = time.monotonic()
-
- # Read the stats after the warmup duration (no sampling needed).
- time.sleep(warmup_time)
- stats[time.monotonic()-time_start] = client.get_stats()
-
- if client.get_warnings():
- for warning in client.get_warnings():
- print(warning)
-
- client.reset()
-
- print(u"##### Warmup Statistics #####")
- print(json.dumps(stats, indent=4, separators=(u",", u": ")))
-
- # TODO: check stats format
- stats = stats[sorted(stats.keys())[-1]]
- lost_a = stats[port_0][u"opackets"] - stats[port_1][u"ipackets"]
- if traffic_directions > 1:
- lost_b = stats[port_1][u"opackets"] - stats[port_0][u"ipackets"]
-
- print(f"packets lost from {port_0} --> {port_1}: {lost_a} pkts")
- if traffic_directions > 1:
- print(f"packets lost from {port_1} --> {port_0}: {lost_b} pkts")
-
# Clear the stats before injecting.
lost_a = 0
lost_b = 0
@@ -177,21 +168,15 @@ def simple_burst(
# Choose CPS and start traffic.
client.start(
- mult=mult, duration=duration, nc=True,
- latency_pps=mult if latency else 0, client_mask=2**len(ports)-1
+ mult=multiplier,
+ # Increase the input duration slightly,
+ # to ensure it does not end before sleep&stop below happens.
+ duration=duration + 0.1 if duration > 0 else duration,
+ nc=True,
+ latency_pps=int(multiplier) if latency else 0,
+ client_mask=2**len(ports)-1,
)
time_start = time.monotonic()
- # t-rex starts the packet flow with the delay
- stats[time.monotonic()-time_start] = client.get_stats(ports=[port_0])
- while stats[sorted(stats.keys())[-1]][port_0][u"opackets"] == 0:
- stats.clear()
- time.sleep(0.001)
- stats[time.monotonic() - time_start] = \
- client.get_stats(ports=[port_0])
- else:
- trex_start_time = list(sorted(stats.keys()))[-1]
- time_start += trex_start_time
- stats.clear()
if async_start:
# For async stop, we need to export the current snapshot.
@@ -201,26 +186,27 @@ def simple_burst(
xsnap1 = client.ports[port_1].get_xstats().reference_stats
print(f"Xstats snapshot 1: {xsnap1!r}")
else:
- time.sleep(
- stats_sampling if stats_sampling < duration else duration
+ time.sleep(duration)
+
+ # Do not block yet, the existing transactions may take long time
+ # to finish. We need an action that is almost reset(),
+ # but without clearing stats.
+ client.stop(block=False)
+ client.stop_latency()
+ client.remove_rx_queue(client.get_all_ports())
+ # Now we can wait for the real traffic stop.
+ client.stop(block=True)
+
+ # Read the stats after the traffic stopped (or time up).
+ stats[time.monotonic() - time_start] = client.get_stats(
+ ports=ports
)
- # Do not block until done.
- while client.is_traffic_active(ports=ports):
- # Sample the stats.
- stats[time.monotonic()-time_start] = \
- client.get_stats(ports=ports)
- time.sleep(
- stats_sampling if stats_sampling < duration else duration
- )
- else:
- # Read the stats after the test
- stats[time.monotonic()-time_start] = \
- client.get_stats(ports=ports)
if client.get_warnings():
for warning in client.get_warnings():
print(warning)
+ # Now finish the complete reset.
client.reset()
print(u"##### Statistics #####")
@@ -253,8 +239,10 @@ def simple_burst(
if traffic_directions > 1:
total_sent = \
stats[port_0][u"opackets"] + stats[port_1][u"opackets"]
- total_rcvd = \
+ total_received = \
stats[port_0][u"ipackets"] + stats[port_1][u"ipackets"]
+ client_sent = stats[port_0][u"opackets"]
+ client_received = stats[port_0][u"ipackets"]
client_stats = stats[u"traffic"][u"client"]
server_stats = stats[u"traffic"][u"server"]
# Some zero counters are not sent
@@ -263,109 +251,112 @@ def simple_burst(
c_act_flows = client_stats[u"m_active_flows"]
c_est_flows = client_stats[u"m_est_flows"]
c_traffic_duration = client_stats.get(u"m_traffic_duration", 0)
- l7_data = f"client_active_flows={c_act_flows}, "
- l7_data += f"client_established_flows={c_est_flows}, "
- l7_data += f"client_traffic_duration={c_traffic_duration}, "
+ l7_data = f"client_active_flows={c_act_flows}; "
+ l7_data += f"client_established_flows={c_est_flows}; "
+ l7_data += f"client_traffic_duration={c_traffic_duration}; "
# Possible errors
# Too many packets in NIC rx queue
c_err_rx_throttled = client_stats.get(u"err_rx_throttled", 0)
- l7_data += f"client_err_rx_throttled={c_err_rx_throttled}, "
+ l7_data += f"client_err_rx_throttled={c_err_rx_throttled}; "
# Number of client side flows that were not opened
# due to flow-table overflow
c_err_nf_throttled = client_stats.get(u"err_c_nf_throttled", 0)
- l7_data += f"client_err_nf_throttled={c_err_nf_throttled}, "
+ l7_data += f"client_err_nf_throttled={c_err_nf_throttled}; "
# Too many flows
c_err_flow_overflow = client_stats.get(u"err_flow_overflow", 0)
- l7_data += f"client_err_flow_overflow={c_err_flow_overflow}, "
+ l7_data += f"client_err_flow_overflow={c_err_flow_overflow}; "
# Server
s_act_flows = server_stats[u"m_active_flows"]
s_est_flows = server_stats[u"m_est_flows"]
s_traffic_duration = server_stats.get(u"m_traffic_duration", 0)
- l7_data += f"server_active_flows={s_act_flows}, "
- l7_data += f"server_established_flows={s_est_flows}, "
- l7_data += f"server_traffic_duration={s_traffic_duration}, "
+ l7_data += f"server_active_flows={s_act_flows}; "
+ l7_data += f"server_established_flows={s_est_flows}; "
+ l7_data += f"server_traffic_duration={s_traffic_duration}; "
# Possible errors
# Too many packets in NIC rx queue
s_err_rx_throttled = server_stats.get(u"err_rx_throttled", 0)
- l7_data += f"client_err_rx_throttled={s_err_rx_throttled}, "
+ l7_data += f"client_err_rx_throttled={s_err_rx_throttled}; "
if u"udp" in profile_file:
# Client
# Established connections
c_udp_connects = client_stats.get(u"udps_connects", 0)
- l7_data += f"client_udp_connects={c_udp_connects}, "
+ l7_data += f"client_udp_connects={c_udp_connects}; "
# Closed connections
c_udp_closed = client_stats.get(u"udps_closed", 0)
- l7_data += f"client_udp_closed={c_udp_closed}, "
+ l7_data += f"client_udp_closed={c_udp_closed}; "
# Sent bytes
c_udp_sndbyte = client_stats.get(u"udps_sndbyte", 0)
- l7_data += f"client_udp_tx_bytes={c_udp_sndbyte}, "
+ l7_data += f"client_udp_tx_bytes={c_udp_sndbyte}; "
# Sent packets
c_udp_sndpkt = client_stats.get(u"udps_sndpkt", 0)
- l7_data += f"client_udp_tx_packets={c_udp_sndpkt}, "
+ l7_data += f"client_udp_tx_packets={c_udp_sndpkt}; "
# Received bytes
c_udp_rcvbyte = client_stats.get(u"udps_rcvbyte", 0)
- l7_data += f"client_udp_rx_bytes={c_udp_rcvbyte}, "
+ l7_data += f"client_udp_rx_bytes={c_udp_rcvbyte}; "
# Received packets
c_udp_rcvpkt = client_stats.get(u"udps_rcvpkt", 0)
- l7_data += f"client_udp_rx_packets={c_udp_rcvpkt}, "
+ l7_data += f"client_udp_rx_packets={c_udp_rcvpkt}; "
# Keep alive drops
c_udp_keepdrops = client_stats.get(u"udps_keepdrops", 0)
- l7_data += f"client_udp_keep_drops={c_udp_keepdrops}, "
+ l7_data += f"client_udp_keep_drops={c_udp_keepdrops}; "
+ # Client without flow
+ c_err_cwf = client_stats.get(u"err_cwf", 0)
+ l7_data += f"client_err_cwf={c_err_cwf}; "
# Server
# Accepted connections
s_udp_accepts = server_stats.get(u"udps_accepts", 0)
- l7_data += f"server_udp_accepts={s_udp_accepts}, "
+ l7_data += f"server_udp_accepts={s_udp_accepts}; "
# Closed connections
s_udp_closed = server_stats.get(u"udps_closed", 0)
- l7_data += f"server_udp_closed={s_udp_closed}, "
+ l7_data += f"server_udp_closed={s_udp_closed}; "
# Sent bytes
s_udp_sndbyte = server_stats.get(u"udps_sndbyte", 0)
- l7_data += f"server_udp_tx_bytes={s_udp_sndbyte}, "
+ l7_data += f"server_udp_tx_bytes={s_udp_sndbyte}; "
# Sent packets
s_udp_sndpkt = server_stats.get(u"udps_sndpkt", 0)
- l7_data += f"server_udp_tx_packets={s_udp_sndpkt}, "
+ l7_data += f"server_udp_tx_packets={s_udp_sndpkt}; "
# Received bytes
s_udp_rcvbyte = server_stats.get(u"udps_rcvbyte", 0)
- l7_data += f"server_udp_rx_bytes={s_udp_rcvbyte}, "
+ l7_data += f"server_udp_rx_bytes={s_udp_rcvbyte}; "
# Received packets
s_udp_rcvpkt = server_stats.get(u"udps_rcvpkt", 0)
- l7_data += f"server_udp_rx_packets={s_udp_rcvpkt}, "
+ l7_data += f"server_udp_rx_packets={s_udp_rcvpkt}; "
elif u"tcp" in profile_file:
# Client
- # Initiated connections
- c_tcp_connatt = client_stats.get(u"tcps_connattempt", 0)
- l7_data += f"client_tcp_connect_inits={c_tcp_connatt}, "
+ # Connection attempts
+ c_tcp_connattempt = client_stats.get(u"tcps_connattempt", 0)
+ l7_data += f"client_tcp_connattempt={c_tcp_connattempt}; "
# Established connections
c_tcp_connects = client_stats.get(u"tcps_connects", 0)
- l7_data += f"client_tcp_connects={c_tcp_connects}, "
+ l7_data += f"client_tcp_connects={c_tcp_connects}; "
# Closed connections
c_tcp_closed = client_stats.get(u"tcps_closed", 0)
- l7_data += f"client_tcp_closed={c_tcp_closed}, "
+ l7_data += f"client_tcp_closed={c_tcp_closed}; "
# Send bytes
c_tcp_sndbyte = client_stats.get(u"tcps_sndbyte", 0)
- l7_data += f"client_tcp_tx_bytes={c_tcp_sndbyte}, "
+ l7_data += f"client_tcp_tx_bytes={c_tcp_sndbyte}; "
# Received bytes
c_tcp_rcvbyte = client_stats.get(u"tcps_rcvbyte", 0)
- l7_data += f"client_tcp_rx_bytes={c_tcp_rcvbyte}, "
+ l7_data += f"client_tcp_rx_bytes={c_tcp_rcvbyte}; "
# Server
# Accepted connections
s_tcp_accepts = server_stats.get(u"tcps_accepts", 0)
- l7_data += f"server_tcp_accepts={s_tcp_accepts}, "
+ l7_data += f"server_tcp_accepts={s_tcp_accepts}; "
# Established connections
s_tcp_connects = server_stats.get(u"tcps_connects", 0)
- l7_data += f"server_tcp_connects={s_tcp_connects}, "
+ l7_data += f"server_tcp_connects={s_tcp_connects}; "
# Closed connections
s_tcp_closed = server_stats.get(u"tcps_closed", 0)
- l7_data += f"server_tcp_closed={s_tcp_closed}, "
+ l7_data += f"server_tcp_closed={s_tcp_closed}; "
# Sent bytes
s_tcp_sndbyte = server_stats.get(u"tcps_sndbyte", 0)
- l7_data += f"server_tcp_tx_bytes={s_tcp_sndbyte}, "
+ l7_data += f"server_tcp_tx_bytes={s_tcp_sndbyte}; "
# Received bytes
s_tcp_rcvbyte = server_stats.get(u"tcps_rcvbyte", 0)
- l7_data += f"server_tcp_rx_bytes={s_tcp_rcvbyte}, "
+ l7_data += f"server_tcp_rx_bytes={s_tcp_rcvbyte}; "
else:
total_sent = stats[port_0][u"opackets"]
- total_rcvd = stats[port_1][u"ipackets"]
+ total_received = stats[port_1][u"ipackets"]
print(f"packets lost from {port_0} --> {port_1}: {lost_a} pkts")
if traffic_directions > 1:
@@ -383,14 +374,17 @@ def simple_burst(
client.clear_profile()
client.disconnect()
print(
- f"trex_start_time={trex_start_time}, "
- f"cps={mult!r}, total_received={total_rcvd}, "
- f"total_sent={total_sent}, frame_loss={lost_a + lost_b}, "
- f"approximated_duration={approximated_duration}, "
- f"latency_stream_0(usec)={lat_a}, "
- f"latency_stream_1(usec)={lat_b}, "
- f"latency_hist_stream_0={lat_a_hist}, "
- f"latency_hist_stream_1={lat_b_hist}, "
+ f"multiplier={multiplier!r}; "
+ f"total_received={total_received}; "
+ f"total_sent={total_sent}; "
+ f"frame_loss={lost_a + lost_b}; "
+ f"approximated_duration={approximated_duration}; "
+ f"latency_stream_0(usec)={lat_a}; "
+ f"latency_stream_1(usec)={lat_b}; "
+ f"latency_hist_stream_0={lat_a_hist}; "
+ f"latency_hist_stream_1={lat_b_hist}; "
+ f"client_sent={client_sent}; "
+ f"client_received={client_received}; "
f"{l7_data}"
)
@@ -408,21 +402,17 @@ def main():
)
parser.add_argument(
u"-d", u"--duration", required=True, type=float,
- help=u"Duration of traffic run."
+ help=u"Duration of the whole traffic run, including overheads."
)
parser.add_argument(
u"-s", u"--frame_size", required=True,
help=u"Size of a Frame without padding and IPG."
)
parser.add_argument(
- u"-m", u"--mult", required=True, type=int,
+ u"-m", u"--multiplier", required=True, type=float,
help=u"Multiplier of profile CPS."
)
parser.add_argument(
- u"-w", u"--warmup_time", type=float, default=5.0,
- help=u"Traffic warm-up time in seconds, 0 = disable."
- )
- parser.add_argument(
u"--port_0", required=True, type=int,
help=u"Port 0 on the traffic generator."
)
@@ -451,10 +441,15 @@ def main():
framesize = args.frame_size
simple_burst(
- profile_file=args.profile, duration=args.duration, framesize=framesize,
- mult=args.mult, warmup_time=args.warmup_time, port_0=args.port_0,
- port_1=args.port_1, latency=args.latency, async_start=args.async_start,
- traffic_directions=args.traffic_directions
+ profile_file=args.profile,
+ duration=args.duration,
+ framesize=framesize,
+ multiplier=args.multiplier,
+ port_0=args.port_0,
+ port_1=args.port_1,
+ latency=args.latency,
+ async_start=args.async_start,
+ traffic_directions=args.traffic_directions,
)
diff --git a/GPL/tools/trex/trex_astf_stop.py b/GPL/tools/trex/trex_astf_stop.py
index 655bdf1a97..d9784e661a 100644
--- a/GPL/tools/trex/trex_astf_stop.py
+++ b/GPL/tools/trex/trex_astf_stop.py
@@ -92,12 +92,15 @@ def main():
total_rcvd, total_sent = rx_0 + rx_1, tx_0 + tx_1
total_lost = total_sent - total_rcvd
- # TODO: Add latency.
print(
- f"cps='unknown', total_received={total_rcvd}, total_sent={total_sent}, "
- f"frame_loss={total_lost}, "
- f"latency_stream_0(usec)=-1/-1/-1, latency_stream_1(usec)=-1/-1/-1, "
- u"latency_hist_stream_0={}, latency_hist_stream_1={}, "
+ f"cps='unknown'; "
+ f"total_received={total_rcvd}; "
+ f"total_sent={total_sent}; "
+ f"frame_loss={total_lost}; "
+ f"latency_stream_0(usec)=-1/-1/-1; "
+ f"latency_stream_1(usec)=-1/-1/-1; "
+ f"latency_hist_stream_0={}; "
+ f"latency_hist_stream_1={}; "
)
diff --git a/GPL/tools/trex/trex_stl_profile.py b/GPL/tools/trex/trex_stl_profile.py
index ac09b8dda1..ea001b58b6 100644
--- a/GPL/tools/trex/trex_stl_profile.py
+++ b/GPL/tools/trex/trex_stl_profile.py
@@ -60,8 +60,17 @@ def fmt_latency(lat_min, lat_avg, lat_max, hdrh):
def simple_burst(
- profile_file, duration, framesize, rate, warmup_time, port_0, port_1,
- latency, async_start=False, traffic_directions=2, force=False):
+ profile_file,
+ duration,
+ framesize,
+ rate,
+ port_0,
+ port_1,
+ latency,
+ async_start=False,
+ traffic_directions=2,
+ force=False,
+ ):
"""Send traffic and measure packet loss and latency.
Procedure:
@@ -83,7 +92,6 @@ def simple_burst(
:param framesize: Frame size.
:param duration: Duration of traffic run in seconds (-1=infinite).
:param rate: Traffic rate [percentage, pps, bps].
- :param warmup_time: Traffic warm-up time in seconds, 0 = disable.
:param port_0: Port 0 on the traffic generator.
:param port_1: Port 1 on the traffic generator.
:param latency: With latency stats.
@@ -94,7 +102,6 @@ def simple_burst(
:type framesize: int or str
:type duration: float
:type rate: str
- :type warmup_time: float
:type port_0: int
:type port_1: int
:type latency: bool
@@ -105,8 +112,7 @@ def simple_burst(
client = None
total_rcvd = 0
total_sent = 0
- approximated_duration = 0
- approximated_rate = 0
+ approximated_duration = 0.0
lost_a = 0
lost_b = 0
lat_a = u"-1/-1/-1/"
@@ -162,40 +168,6 @@ def simple_burst(
ports = [port_0]
if traffic_directions > 1:
ports.append(port_1)
- # Warm-up phase:
- if warmup_time > 0:
- # Clear the stats before injecting:
- client.clear_stats()
-
- # Choose rate and start traffic:
- client.start(
- ports=ports, mult=rate, duration=warmup_time, force=force,
- core_mask=STLClient.CORE_MASK_PIN
- )
-
- # Block until done:
- time_start = time.monotonic()
- client.wait_on_traffic(ports=ports, timeout=warmup_time+30)
- time_stop = time.monotonic()
- approximated_duration = time_stop - time_start
-
- if client.get_warnings():
- for warning in client.get_warnings():
- print(warning)
-
- # Read the stats after the test:
- stats = client.get_stats()
-
- print(u"##### Warmup statistics #####")
- print(json.dumps(stats, indent=4, separators=(u",", u": ")))
-
- lost_a = stats[port_0][u"opackets"] - stats[port_1][u"ipackets"]
- if traffic_directions > 1:
- lost_b = stats[port_1][u"opackets"] - stats[port_0][u"ipackets"]
-
- print(f"\npackets lost from {port_0} --> {port_1}: {lost_a} pkts")
- if traffic_directions > 1:
- print(f"packets lost from {port_1} --> {port_0}: {lost_b} pkts")
# Clear the stats before injecting:
client.clear_stats()
@@ -204,8 +176,11 @@ def simple_burst(
# Choose rate and start traffic:
client.start(
- ports=ports, mult=rate, duration=duration, force=force,
- core_mask=STLClient.CORE_MASK_PIN
+ ports=ports,
+ mult=rate,
+ duration=duration,
+ force=force,
+ core_mask=STLClient.CORE_MASK_PIN,
)
if async_start:
@@ -254,10 +229,6 @@ def simple_burst(
else:
total_sent = stats[port_0][u"opackets"]
total_rcvd = stats[port_1][u"ipackets"]
- try:
- approximated_rate = total_sent / approximated_duration
- except ZeroDivisionError:
- pass
print(f"\npackets lost from {port_0} --> {port_1}: {lost_a} pkts")
if traffic_directions > 1:
@@ -275,13 +246,14 @@ def simple_burst(
if client:
client.disconnect()
print(
- f"rate={rate!r}, total_received={total_rcvd}, "
- f"total_sent={total_sent}, frame_loss={lost_a + lost_b}, "
- f"target_duration={duration!r}, "
- f"approximated_duration={approximated_duration!r}, "
- f"approximated_rate={approximated_rate}, "
- f"latency_stream_0(usec)={lat_a}, "
- f"latency_stream_1(usec)={lat_b}, "
+ f"rate={rate!r}; "
+ f"total_received={total_rcvd}; "
+ f"total_sent={total_sent}; "
+ f"frame_loss={lost_a + lost_b}; "
+ f"target_duration={duration!r}; "
+ f"approximated_duration={approximated_duration!r}; "
+ f"latency_stream_0(usec)={lat_a}; "
+ f"latency_stream_1(usec)={lat_b}; "
)
@@ -309,10 +281,6 @@ def main():
help=u"Traffic rate with included units (pps)."
)
parser.add_argument(
- u"-w", u"--warmup_time", type=float, default=5.0,
- help=u"Traffic warm-up time in seconds, 0 = disable."
- )
- parser.add_argument(
u"--port_0", required=True, type=int,
help=u"Port 0 on the traffic generator."
)
@@ -345,10 +313,16 @@ def main():
framesize = args.frame_size
simple_burst(
- profile_file=args.profile, duration=args.duration, framesize=framesize,
- rate=args.rate, warmup_time=args.warmup_time, port_0=args.port_0,
- port_1=args.port_1, latency=args.latency, async_start=args.async_start,
- traffic_directions=args.traffic_directions, force=args.force
+ profile_file=args.profile,
+ duration=args.duration,
+ framesize=framesize,
+ rate=args.rate,
+ port_0=args.port_0,
+ port_1=args.port_1,
+ latency=args.latency,
+ async_start=args.async_start,
+ traffic_directions=args.traffic_directions,
+ force=args.force,
)
diff --git a/GPL/tools/trex/trex_stl_stop.py b/GPL/tools/trex/trex_stl_stop.py
index dcdae7c10a..f07479da8a 100644
--- a/GPL/tools/trex/trex_stl_stop.py
+++ b/GPL/tools/trex/trex_stl_stop.py
@@ -91,13 +91,16 @@ def main():
total_rcvd, total_sent = rx_0 + rx_1, tx_0 + tx_1
total_lost = total_sent - total_rcvd
- # TODO: Add latency.
print(
- f"rate='unknown', total_received={total_rcvd}, "
- f"total_sent={total_sent}, frame_loss={total_lost}, "
- f"target_duration='manual', approximated_duration='manual', "
- f"approximated_rate='unknown', "
- f"latency_stream_0(usec)=-1/-1/-1, latency_stream_1(usec)=-1/-1/-1"
+ f"rate='unknown'; "
+ f"total_received={total_rcvd}; "
+ f"total_sent={total_sent}; "
+ f"frame_loss={total_lost}; "
+ f"target_duration='manual'; "
+ f"approximated_duration='manual'; "
+ f"approximated_rate='unknown'; "
+ f"latency_stream_0(usec)=-1/-1/-1; "
+ f"latency_stream_1(usec)=-1/-1/-1; "
)