summaryrefslogtreecommitdiffstats
path: root/vpp-api/lua/README.md
diff options
context:
space:
mode:
authorAndrew Yourtchenko <ayourtch@gmail.com>2016-11-25 13:29:30 +0000
committerDamjan Marion <dmarion.lists@gmail.com>2016-11-28 11:55:34 +0000
commit985f3d11fae7f93eb9a9e96d59b8e7a066819672 (patch)
treed44d3888ca6118384d7ea79b85411aa0b8539b5c /vpp-api/lua/README.md
parent085f5c00661ef00a038d3d54be7be6dafd0a6d89 (diff)
Clean up the Lua API, make it luajit-only, no C
This removes the libcough which was a temporary layer atop the libpneum to make it synchronous. Now Lua API only requres luajit, and has no dependencies on compiling any C. Also comment out a couple of debug print functions. Change-Id: I35fd1c8088c6fd1b10b9e4d903ad241ab32fd91a Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
Diffstat (limited to 'vpp-api/lua/README.md')
-rw-r--r--vpp-api/lua/README.md47
1 files changed, 21 insertions, 26 deletions
diff --git a/vpp-api/lua/README.md b/vpp-api/lua/README.md
index 3fc93b888a9..4ecdb34d2f3 100644
--- a/vpp-api/lua/README.md
+++ b/vpp-api/lua/README.md
@@ -6,44 +6,39 @@ To run the examples here:
1) install luajit - "sudo apt-get install luajit" on ubuntu
-2) make build-vpp-api in the top directory
+2) "make build-vpp-api" in the top VPP directory
-3) "make" in this directory to build libcough.so
+3) "make run" in a separate terminal window
+ This ensures you have an instance of VPP running
-4) "make run" in a separate terminal window
-
-5) sudo luajit examples/example-cli.lua
+4) sudo luajit examples/example-cli.lua
This will result in something like this:
-libcough detected
-
-Version: 17.01-rc0~37-g8b3191e
-00000000 31 37 2E 30 31 2D 72 63 30 7E 33 37 2D 67 38 62 17.01-rc0~37-g8b
-00000010 33 31 39 31 65 00 00 00 00 00 00 00 00 00 00 00 3191e...........
+Version:
+00000000 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
+00000010 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
-{ [1] = { ["luaapi_message_name"] = show_version_reply,["program"] = vpe,["version"] = 17.01-rc0~37-g8b3191e,["build_date"] = Fri Nov 11 15:30:21 UTC 2016,["retval"] = 0,["build_directory"] = /home/ubuntu/vpp,["_vl_msg_id"] = 166,["context"] = 0,} ,}
+{ [1] = { ["luaapi_message_name"] = show_version_reply,["program"] = vpe,["version"] = ,["build_date"] = Fri Nov 25 10:58:48 UTC 2016,["retval"] = 0,["build_directory"] = /home/ubuntu/vpp,["_vl_msg_id"] = 170,["context"] = 0,} ,}
---
-{ [1] = { ["luaapi_message_name"] = cli_inband_reply,["_vl_msg_id"] = 90,["length"] = 94,["reply"] = vpp v17.01-rc0~37-g8b3191e built by ubuntu on vpp-lapi-commit at Fri Nov 11 15:30:21 UTC 2016
+{ [1] = { ["luaapi_message_name"] = cli_inband_reply,["_vl_msg_id"] = 94,["length"] = 66,["reply"] = vpp v built by ubuntu on vpp-toys at Fri Nov 25 10:58:48 UTC 2016
,["retval"] = 0,["context"] = 0,} ,}
---
-6) You can also run the performance test bench:
+5) You can also run the performance test bench:
$ sudo luajit bench.lua
-libcough detected
-
-10001 iterations, average speed 4108LL per second
-10001 iterations, average speed 4660LL per second
-10001 iterations, average speed 4095LL per second
-10001 iterations, average speed 4542LL per second
-10001 iterations, average speed 8048LL per second
-10001 iterations, average speed 6805LL per second
-10001 iterations, average speed 5170LL per second
-10001 iterations, average speed 6585LL per second
-10001 iterations, average speed 6714LL per second
-10001 iterations, average speed 6942LL per second
-Average tps across the tests: 5766LL
+10001 iterations, average speed 5624LL per second
+10001 iterations, average speed 6650LL per second
+10001 iterations, average speed 6053LL per second
+10001 iterations, average speed 7056LL per second
+10001 iterations, average speed 6388LL per second
+10001 iterations, average speed 5849LL per second
+10001 iterations, average speed 6321LL per second
+10001 iterations, average speed 6368LL per second
+10001 iterations, average speed 5958LL per second
+10001 iterations, average speed 6482LL per second
+Average tps across the tests: 6274LL
Note: the above is run in an lxd container running inside 2-core
xhyve VM on a Macbook Pro, so I would not take the performance numbers for granted :)