Skip to content

Commit 20b807d

Browse files
danish-tidanish-ti
committed
feat(linux): Add Ethernet performance numbers for 12.01 release
Add netperf Ethernet performance data for AM62X, AM62PX, AM62LX and AM64X in their respective performance guide RST files. Each section covers CPSW TCP bidirectional throughput (with and without interrupt pacing) and UDP egress/ingress 0-loss throughput across 64/128/256/ 1024/1518B frame sizes. AM64X additionally includes ICSSG results. Signed-off-by: MD Danish Anwar <danishanwar@ti.com>
1 parent 9aa0202 commit 20b807d

4 files changed

Lines changed: 509 additions & 0 deletions

File tree

‎source/devices/AM62LX/linux/Linux_Performance_Guide.rst‎

Lines changed: 115 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -429,6 +429,121 @@ ALSA SoC Audio Driver
429429
"96000","3073222.00 (min 3073203.00, max 3073253.00)","0.57 (min 0.31, max 0.97)"
430430

431431

432+
Ethernet
433+
--------
434+
435+
Ethernet performance benchmarks were measured using :command:`netperf` 2.7.1 https://hewlettpackard.github.io/netperf/doc/netperf.html
436+
Test procedures were modeled after those defined in RFC-2544:
437+
https://tools.ietf.org/html/rfc2544, where the DUT is the TI device
438+
and the "tester" used was a Linux PC. To produce consistent results,
439+
it is recommended to carry out performance tests in a private network and to avoid
440+
running NFS on the same interface used in the test. In these results,
441+
CPU utilization was captured as the total percentage used across all cores on the device,
442+
while running the performance test over one external interface.
443+
444+
UDP Throughput (0% loss) was measured by the procedure defined in RFC-2544 section 26.1: Throughput.
445+
In this scenario, :command:`netperf` options burst_size (-b) and wait_time (-w) are used to limit bandwidth
446+
during different trials of the test, with the goal of finding the highest rate at which
447+
no loss is seen. For example, to limit bandwidth to 500Mbits/sec with 1472B datagram:
448+
449+
.. code-block:: console
450+
451+
burst_size = <bandwidth (bits/sec)> / 8 (bits -> bytes) / <UDP datagram size> / 100 (seconds -> 10 ms)
452+
burst_size = 500000000 / 8 / 1472 / 100 = 425
453+
454+
wait_time = 10 milliseconds (minimum supported by Linux PC used for testing)
455+
456+
UDP Throughput (possible loss) was measured by capturing throughput and packet loss statistics when
457+
running the :command:`netperf` test with no bandwidth limit (remove -b/-w options).
458+
459+
The following commands were used to tune the socket buffer sizes on the DUT before running the performance tests:
460+
461+
.. code-block:: console
462+
463+
sysctl -w net.core.rmem_default=33554432
464+
sysctl -w net.core.rmem_max=67108864
465+
466+
In order to start a :command:`netperf` client on one device, the other device must have :command:`netserver` running.
467+
To start :command:`netserver`:
468+
469+
.. code-block:: console
470+
471+
netserver [-p <port_number>] [-4 (IPv4 addressing)] [-6 (IPv6 addressing)]
472+
473+
Running the following shell script from the DUT will trigger :command:`netperf` clients to measure
474+
bidirectional TCP performance for 60 seconds and report CPU utilization. Parameter -k is used in
475+
client commands to summarize selected statistics on their own line and -j is used to gain
476+
additional timing measurements during the test.
477+
478+
.. code-block:: console
479+
480+
#!/bin/bash
481+
for i in 1
482+
do
483+
netperf -H <tester ip> -j -c -l 60 -t TCP_STREAM --
484+
-k DIRECTION,THROUGHPUT,MEAN_LATENCY,LOCAL_CPU_UTIL,REMOTE_CPU_UTIL,LOCAL_BYTES_SENT,REMOTE_BYTES_RECVD,LOCAL_SEND_SIZE &
485+
486+
netperf -H <tester ip> -j -c -l 60 -t TCP_MAERTS --
487+
-k DIRECTION,THROUGHPUT,MEAN_LATENCY,LOCAL_CPU_UTIL,REMOTE_CPU_UTIL,LOCAL_BYTES_SENT,REMOTE_BYTES_RECVD,LOCAL_SEND_SIZE &
488+
done
489+
490+
Running the following commands will trigger :command:`netperf` clients to measure UDP burst performance for
491+
60 seconds at various burst/datagram sizes and report CPU utilization.
492+
493+
- For UDP egress tests, run :command:`netperf` client from DUT and start :command:`netserver` on tester.
494+
495+
.. code-block:: console
496+
497+
netperf -H <tester ip> -j -c -l 60 -t UDP_STREAM -b <burst_size> -w <wait_time> -- -m <UDP datagram size>
498+
-k DIRECTION,THROUGHPUT,MEAN_LATENCY,LOCAL_CPU_UTIL,REMOTE_CPU_UTIL,LOCAL_BYTES_SENT,REMOTE_BYTES_RECVD,LOCAL_SEND_SIZE
499+
500+
- For UDP ingress tests, run :command:`netperf` client from tester and start :command:`netserver` on DUT.
501+
502+
.. code-block:: console
503+
504+
netperf -H <DUT ip> -j -C -l 60 -t UDP_STREAM -b <burst_size> -w <wait_time> -- -m <UDP datagram size>
505+
-k DIRECTION,THROUGHPUT,MEAN_LATENCY,LOCAL_CPU_UTIL,REMOTE_CPU_UTIL,LOCAL_BYTES_SENT,REMOTE_BYTES_RECVD,LOCAL_SEND_SIZE
506+
507+
CPSW/CPSW2g/CPSW3g Ethernet
508+
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
509+
510+
TCP Bidirectional Throughput
511+
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
512+
513+
.. csv-table:: CPSW2g TCP Bidirectional Throughput
514+
:header: "Command Used","am62lxx_evm-fs: THROUGHPUT (Mbits/sec)","am62lxx_evm-fs: CPU Load % (LOCAL_CPU_UTIL)"
515+
516+
"netperf -H 192.168.0.1 -j -c -C -l 60 -t TCP_STREAM; netperf -H 192.168.0.1 -j -c -C -l 60 -t TCP_MAERTS","1215.82","98.99"
517+
518+
TCP Bidirectional Throughput Interrupt Pacing
519+
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
520+
521+
.. csv-table:: CPSW2g TCP Bidirectional Throughput Interrupt Pacing
522+
:header: "Command Used","am62lxx_evm-fs: THROUGHPUT (Mbits/sec)","am62lxx_evm-fs: CPU Load % (LOCAL_CPU_UTIL)"
523+
524+
"netperf -H 192.168.0.1 -j -c -C -l 60 -t TCP_STREAM; netperf -H 192.168.0.1 -j -c -C -l 60 -t TCP_MAERTS","1266.92","99.09"
525+
526+
UDP Throughput
527+
~~~~~~~~~~~~~~
528+
529+
.. csv-table:: CPSW2g UDP Egress Throughput 0 loss
530+
:header: "Frame Size(bytes)", "am62lxx_evm-fs: THROUGHPUT (Mbits/sec)", "am62lxx_evm-fs: Packets Per Second (kPPS)", "am62lxx_evm-fs: CPU Load % (LOCAL_CPU_UTIL)"
531+
532+
"64","41.23","81","77.62"
533+
"128","80.45","79","77.20"
534+
"256","145.58","71","75.10"
535+
"1024","571.23","70","75.05"
536+
"1518","810.12","69","74.91"
537+
538+
.. csv-table:: CPSW2g UDP Ingress Throughput 0 loss
539+
:header: "Frame Size(bytes)", "am62lxx_evm-fs: THROUGHPUT (Mbits/sec)", "am62lxx_evm-fs: Packets Per Second (kPPS)", "am62lxx_evm-fs: CPU Load % (LOCAL_CPU_UTIL)"
540+
541+
"64","48.36","94","77.82"
542+
"128","104.64","102","76.83"
543+
"256","205.60","100","75.71"
544+
"1024","651.14","79","73.67"
545+
"1518","810.12","69","74.91"
546+
432547
|
433548
434549
Linux OSPI Flash Driver

‎source/devices/AM62PX/linux/Linux_Performance_Guide.rst‎

Lines changed: 117 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -412,6 +412,123 @@ Run Glmark2 and capture performance reported (Score). All display outputs (HDMI,
412412
"Glmark2-Wayland","776.40 (min 765.00, max 786.00)"
413413

414414

415+
Ethernet
416+
--------
417+
418+
Ethernet performance benchmarks were measured using :command:`netperf` 2.7.1 https://hewlettpackard.github.io/netperf/doc/netperf.html
419+
Test procedures were modeled after those defined in RFC-2544:
420+
https://tools.ietf.org/html/rfc2544, where the DUT is the TI device
421+
and the "tester" used was a Linux PC. To produce consistent results,
422+
it is recommended to carry out performance tests in a private network and to avoid
423+
running NFS on the same interface used in the test. In these results,
424+
CPU utilization was captured as the total percentage used across all cores on the device,
425+
while running the performance test over one external interface.
426+
427+
UDP Throughput (0% loss) was measured by the procedure defined in RFC-2544 section 26.1: Throughput.
428+
In this scenario, :command:`netperf` options burst_size (-b) and wait_time (-w) are used to limit bandwidth
429+
during different trials of the test, with the goal of finding the highest rate at which
430+
no loss is seen. For example, to limit bandwidth to 500Mbits/sec with 1472B datagram:
431+
432+
.. code-block:: console
433+
434+
burst_size = <bandwidth (bits/sec)> / 8 (bits -> bytes) / <UDP datagram size> / 100 (seconds -> 10 ms)
435+
burst_size = 500000000 / 8 / 1472 / 100 = 425
436+
437+
wait_time = 10 milliseconds (minimum supported by Linux PC used for testing)
438+
439+
UDP Throughput (possible loss) was measured by capturing throughput and packet loss statistics when
440+
running the :command:`netperf` test with no bandwidth limit (remove -b/-w options).
441+
442+
The following commands were used to tune the socket buffer sizes on the DUT before running the performance tests:
443+
444+
.. code-block:: console
445+
446+
sysctl -w net.core.rmem_default=33554432
447+
sysctl -w net.core.rmem_max=67108864
448+
449+
In order to start a :command:`netperf` client on one device, the other device must have :command:`netserver` running.
450+
To start :command:`netserver`:
451+
452+
.. code-block:: console
453+
454+
netserver [-p <port_number>] [-4 (IPv4 addressing)] [-6 (IPv6 addressing)]
455+
456+
Running the following shell script from the DUT will trigger :command:`netperf` clients to measure
457+
bidirectional TCP performance for 60 seconds and report CPU utilization. Parameter -k is used in
458+
client commands to summarize selected statistics on their own line and -j is used to gain
459+
additional timing measurements during the test.
460+
461+
.. code-block:: console
462+
463+
#!/bin/bash
464+
for i in 1
465+
do
466+
netperf -H <tester ip> -j -c -l 60 -t TCP_STREAM --
467+
-k DIRECTION,THROUGHPUT,MEAN_LATENCY,LOCAL_CPU_UTIL,REMOTE_CPU_UTIL,LOCAL_BYTES_SENT,REMOTE_BYTES_RECVD,LOCAL_SEND_SIZE &
468+
469+
netperf -H <tester ip> -j -c -l 60 -t TCP_MAERTS --
470+
-k DIRECTION,THROUGHPUT,MEAN_LATENCY,LOCAL_CPU_UTIL,REMOTE_CPU_UTIL,LOCAL_BYTES_SENT,REMOTE_BYTES_RECVD,LOCAL_SEND_SIZE &
471+
done
472+
473+
Running the following commands will trigger :command:`netperf` clients to measure UDP burst performance for
474+
60 seconds at various burst/datagram sizes and report CPU utilization.
475+
476+
- For UDP egress tests, run :command:`netperf` client from DUT and start :command:`netserver` on tester.
477+
478+
.. code-block:: console
479+
480+
netperf -H <tester ip> -j -c -l 60 -t UDP_STREAM -b <burst_size> -w <wait_time> -- -m <UDP datagram size>
481+
-k DIRECTION,THROUGHPUT,MEAN_LATENCY,LOCAL_CPU_UTIL,REMOTE_CPU_UTIL,LOCAL_BYTES_SENT,REMOTE_BYTES_RECVD,LOCAL_SEND_SIZE
482+
483+
- For UDP ingress tests, run :command:`netperf` client from tester and start :command:`netserver` on DUT.
484+
485+
.. code-block:: console
486+
487+
netperf -H <DUT ip> -j -C -l 60 -t UDP_STREAM -b <burst_size> -w <wait_time> -- -m <UDP datagram size>
488+
-k DIRECTION,THROUGHPUT,MEAN_LATENCY,LOCAL_CPU_UTIL,REMOTE_CPU_UTIL,LOCAL_BYTES_SENT,REMOTE_BYTES_RECVD,LOCAL_SEND_SIZE
489+
490+
CPSW/CPSW2g/CPSW3g Ethernet
491+
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
492+
493+
- CPSW3g: AM62px
494+
495+
TCP Bidirectional Throughput
496+
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
497+
498+
.. csv-table:: CPSW2g TCP Bidirectional Throughput
499+
:header: "Command Used","am62pxx_sk-fs: THROUGHPUT (Mbits/sec)","am62pxx_sk-fs: CPU Load % (LOCAL_CPU_UTIL)"
500+
501+
"netperf -H 192.168.0.1 -j -c -C -l 60 -t TCP_STREAM; netperf -H 192.168.0.1 -j -c -C -l 60 -t TCP_MAERTS","1828.16","43.22"
502+
503+
TCP Bidirectional Throughput Interrupt Pacing
504+
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
505+
506+
.. csv-table:: CPSW2g TCP Bidirectional Throughput Interrupt Pacing
507+
:header: "Command Used","am62pxx_sk-fs: THROUGHPUT (Mbits/sec)","am62pxx_sk-fs: CPU Load % (LOCAL_CPU_UTIL)"
508+
509+
"netperf -H 192.168.0.1 -j -c -C -l 60 -t TCP_STREAM; netperf -H 192.168.0.1 -j -c -C -l 60 -t TCP_MAERTS","1789.37","41.18"
510+
511+
UDP Throughput
512+
~~~~~~~~~~~~~~
513+
514+
.. csv-table:: CPSW2g UDP Egress Throughput 0 loss
515+
:header: "Frame Size(bytes)", "am62pxx_sk-fs: THROUGHPUT (Mbits/sec)", "am62pxx_sk-fs: Packets Per Second (kPPS)", "am62pxx_sk-fs: CPU Load % (LOCAL_CPU_UTIL)"
516+
517+
"64","46.50","91","37.58"
518+
"128","90.35","88","37.33"
519+
"256","179.38","88","37.18"
520+
"1024","673.26","82","35.55"
521+
"1518","956.86","81","40.26"
522+
523+
.. csv-table:: CPSW2g UDP Ingress Throughput 0 loss
524+
:header: "Frame Size(bytes)", "am62pxx_sk-fs: THROUGHPUT (Mbits/sec)", "am62pxx_sk-fs: Packets Per Second (kPPS)", "am62pxx_sk-fs: CPU Load % (LOCAL_CPU_UTIL)"
525+
526+
"64","30.46","59","9.72"
527+
"128","64.61","63","12.30"
528+
"256","156.26","76","16.65"
529+
"1024","939.47","115","36.24"
530+
"1518","908.82","77","36.28"
531+
415532
|
416533
417534
Linux OSPI Flash Driver

‎source/devices/AM62X/linux/Linux_Performance_Guide.rst‎

Lines changed: 117 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -468,6 +468,123 @@ Run Glmark2 and capture performance reported (Score). All display outputs (HDMI,
468468

469469

470470

471+
Ethernet
472+
--------
473+
474+
Ethernet performance benchmarks were measured using :command:`netperf` 2.7.1 https://hewlettpackard.github.io/netperf/doc/netperf.html
475+
Test procedures were modeled after those defined in RFC-2544:
476+
https://tools.ietf.org/html/rfc2544, where the DUT is the TI device
477+
and the "tester" used was a Linux PC. To produce consistent results,
478+
it is recommended to carry out performance tests in a private network and to avoid
479+
running NFS on the same interface used in the test. In these results,
480+
CPU utilization was captured as the total percentage used across all cores on the device,
481+
while running the performance test over one external interface.
482+
483+
UDP Throughput (0% loss) was measured by the procedure defined in RFC-2544 section 26.1: Throughput.
484+
In this scenario, :command:`netperf` options burst_size (-b) and wait_time (-w) are used to limit bandwidth
485+
during different trials of the test, with the goal of finding the highest rate at which
486+
no loss is seen. For example, to limit bandwidth to 500Mbits/sec with 1472B datagram:
487+
488+
.. code-block:: console
489+
490+
burst_size = <bandwidth (bits/sec)> / 8 (bits -> bytes) / <UDP datagram size> / 100 (seconds -> 10 ms)
491+
burst_size = 500000000 / 8 / 1472 / 100 = 425
492+
493+
wait_time = 10 milliseconds (minimum supported by Linux PC used for testing)
494+
495+
UDP Throughput (possible loss) was measured by capturing throughput and packet loss statistics when
496+
running the :command:`netperf` test with no bandwidth limit (remove -b/-w options).
497+
498+
The following commands were used to tune the socket buffer sizes on the DUT before running the performance tests:
499+
500+
.. code-block:: console
501+
502+
sysctl -w net.core.rmem_default=33554432
503+
sysctl -w net.core.rmem_max=67108864
504+
505+
In order to start a :command:`netperf` client on one device, the other device must have :command:`netserver` running.
506+
To start :command:`netserver`:
507+
508+
.. code-block:: console
509+
510+
netserver [-p <port_number>] [-4 (IPv4 addressing)] [-6 (IPv6 addressing)]
511+
512+
Running the following shell script from the DUT will trigger :command:`netperf` clients to measure
513+
bidirectional TCP performance for 60 seconds and report CPU utilization. Parameter -k is used in
514+
client commands to summarize selected statistics on their own line and -j is used to gain
515+
additional timing measurements during the test.
516+
517+
.. code-block:: console
518+
519+
#!/bin/bash
520+
for i in 1
521+
do
522+
netperf -H <tester ip> -j -c -l 60 -t TCP_STREAM --
523+
-k DIRECTION,THROUGHPUT,MEAN_LATENCY,LOCAL_CPU_UTIL,REMOTE_CPU_UTIL,LOCAL_BYTES_SENT,REMOTE_BYTES_RECVD,LOCAL_SEND_SIZE &
524+
525+
netperf -H <tester ip> -j -c -l 60 -t TCP_MAERTS --
526+
-k DIRECTION,THROUGHPUT,MEAN_LATENCY,LOCAL_CPU_UTIL,REMOTE_CPU_UTIL,LOCAL_BYTES_SENT,REMOTE_BYTES_RECVD,LOCAL_SEND_SIZE &
527+
done
528+
529+
Running the following commands will trigger :command:`netperf` clients to measure UDP burst performance for
530+
60 seconds at various burst/datagram sizes and report CPU utilization.
531+
532+
- For UDP egress tests, run :command:`netperf` client from DUT and start :command:`netserver` on tester.
533+
534+
.. code-block:: console
535+
536+
netperf -H <tester ip> -j -c -l 60 -t UDP_STREAM -b <burst_size> -w <wait_time> -- -m <UDP datagram size>
537+
-k DIRECTION,THROUGHPUT,MEAN_LATENCY,LOCAL_CPU_UTIL,REMOTE_CPU_UTIL,LOCAL_BYTES_SENT,REMOTE_BYTES_RECVD,LOCAL_SEND_SIZE
538+
539+
- For UDP ingress tests, run :command:`netperf` client from tester and start :command:`netserver` on DUT.
540+
541+
.. code-block:: console
542+
543+
netperf -H <DUT ip> -j -C -l 60 -t UDP_STREAM -b <burst_size> -w <wait_time> -- -m <UDP datagram size>
544+
-k DIRECTION,THROUGHPUT,MEAN_LATENCY,LOCAL_CPU_UTIL,REMOTE_CPU_UTIL,LOCAL_BYTES_SENT,REMOTE_BYTES_RECVD,LOCAL_SEND_SIZE
545+
546+
CPSW/CPSW2g/CPSW3g Ethernet
547+
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
548+
549+
- CPSW3g: AM62x
550+
551+
TCP Bidirectional Throughput
552+
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
553+
554+
.. csv-table:: CPSW2g TCP Bidirectional Throughput
555+
:header: "Command Used","am62xx_lp_sk-fs: THROUGHPUT (Mbits/sec)","am62xx_lp_sk-fs: CPU Load % (LOCAL_CPU_UTIL)","am62xx_sk-fs: THROUGHPUT (Mbits/sec)","am62xx_sk-fs: CPU Load % (LOCAL_CPU_UTIL)","am62xxsip_sk-fs: THROUGHPUT (Mbits/sec)","am62xxsip_sk-fs: CPU Load % (LOCAL_CPU_UTIL)"
556+
557+
"netperf -H 192.168.0.1 -j -c -C -l 60 -t TCP_STREAM; netperf -H 192.168.0.1 -j -c -C -l 60 -t TCP_MAERTS","1657.01","67.11","1822.62","66.72","1763.18","67.48"
558+
559+
TCP Bidirectional Throughput Interrupt Pacing
560+
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
561+
562+
.. csv-table:: CPSW2g TCP Bidirectional Throughput Interrupt Pacing
563+
:header: "Command Used","am62xx_lp_sk-fs: THROUGHPUT (Mbits/sec)","am62xx_lp_sk-fs: CPU Load % (LOCAL_CPU_UTIL)","am62xx_sk-fs: THROUGHPUT (Mbits/sec)","am62xx_sk-fs: CPU Load % (LOCAL_CPU_UTIL)","am62xxsip_sk-fs: THROUGHPUT (Mbits/sec)","am62xxsip_sk-fs: CPU Load % (LOCAL_CPU_UTIL)"
564+
565+
"netperf -H 192.168.0.1 -j -c -C -l 60 -t TCP_STREAM; netperf -H 192.168.0.1 -j -c -C -l 60 -t TCP_MAERTS","1794.07","63.53","1828.80","57.91","1543.11","49.30"
566+
567+
UDP Throughput
568+
~~~~~~~~~~~~~~
569+
570+
.. csv-table:: CPSW2g UDP Egress Throughput 0 loss
571+
:header: "Frame Size(bytes)", "am62xx_lp_sk-fs: THROUGHPUT (Mbits/sec)", "am62xx_lp_sk-fs: Packets Per Second (kPPS)", "am62xx_lp_sk-fs: CPU Load % (LOCAL_CPU_UTIL)", "am62xx_sk-fs: THROUGHPUT (Mbits/sec)", "am62xx_sk-fs: Packets Per Second (kPPS)", "am62xx_sk-fs: CPU Load % (LOCAL_CPU_UTIL)", "am62xxsip_sk-fs: THROUGHPUT (Mbits/sec)", "am62xxsip_sk-fs: Packets Per Second (kPPS)", "am62xxsip_sk-fs: CPU Load % (LOCAL_CPU_UTIL)"
572+
573+
"64","38.61","75","36.53","44.13","86","37.14","45.71","89","37.74"
574+
"128","76.11","74","36.58","85.34","83","36.72","91.47","89","37.33"
575+
"256","148.89","73","36.41","169.41","83","36.51","170.76","83","36.89"
576+
"1024","592.94","72","36.75","659.08","80","36.45","704.41","86","37.13"
577+
"1518","846.54","72","36.56","208.42","18","11.87","916.55","78","39.59"
578+
579+
.. csv-table:: CPSW2g UDP Ingress Throughput 0 loss
580+
:header: "Frame Size(bytes)", "am62xx_lp_sk-fs: THROUGHPUT (Mbits/sec)", "am62xx_lp_sk-fs: Packets Per Second (kPPS)", "am62xx_lp_sk-fs: CPU Load % (LOCAL_CPU_UTIL)", "am62xx_sk-fs: THROUGHPUT (Mbits/sec)", "am62xx_sk-fs: Packets Per Second (kPPS)", "am62xx_sk-fs: CPU Load % (LOCAL_CPU_UTIL)", "am62xxsip_sk-fs: THROUGHPUT (Mbits/sec)", "am62xxsip_sk-fs: Packets Per Second (kPPS)", "am62xxsip_sk-fs: CPU Load % (LOCAL_CPU_UTIL)"
581+
582+
"64","33.53","65","15.91","41.16","80","19.91","30.41","59","12.23"
583+
"128","63.90","62","15.10","66.05","65","15.68","61.95","60","14.96"
584+
"256","170.41","83","26.12","148.27","72","18.20","142.13","69","13.55"
585+
"1024","611.97","75","26.40","754.92","92","31.48","681.07","83","41.67"
586+
"1518","954.54","81","39.92","956.88","81","38.55","710.47","60","31.15"
587+
471588
|
472589
473590
Linux OSPI Flash Driver

0 commit comments

Comments
 (0)