[dts] [PATCH]tests vf_port_start_stop: update test case

test suite reviews and discussions
 help / color / mirror / Atom feed

* [dts] [PATCH]tests vf_port_start_stop: update test case
@ 2016-02-26  5:13 Lijuan Tu
  2016-02-26  6:15 ` Pei, Yulong
  0 siblings, 1 reply; 5+ messages in thread
From: Lijuan Tu @ 2016-02-26  5:13 UTC (permalink / raw)
  To: yong.liu, dts, yulong.pei, qian.q.xu

Forwarding checking RX-packets, RX-Bytes, Tx-packets, Tx-bytes,
Do not checking Rx-errors
fortville NIC:
    when not start, the port is ready to RX but NOT RX packets,
    if you send 119 packets at 64B, then it will be RXed at the NIC's hardware buffer
    if the buffer is full, it will show in RX-errors
    so we do not verify Rx-errors.

Signed-off-by: Lijuan Tu <lijuanx.a.tu@intel.com>
---
 tests/TestSuite_vf_port_start_stop.py | 48 ++++++++++++++++++++++-------------
 1 file changed, 30 insertions(+), 18 deletions(-)

diff --git a/tests/TestSuite_vf_port_start_stop.py b/tests/TestSuite_vf_port_start_stop.py
index 0cc7ffd..17aa589 100644
--- a/tests/TestSuite_vf_port_start_stop.py
+++ b/tests/TestSuite_vf_port_start_stop.py
@@ -114,11 +114,19 @@ class TestVfPortStartStop(TestCase):
         testpmd.execute_cmd('port stop all')
         testpmd.execute_cmd('clear port stats all')
 
-    def check_port_start_stop(self, testpmd, times=10):
+    def check_port_start_stop(self, testpmd, times=1000):
         """
         VF port start/stop several times , check if it work well.
         """
+        port_id_0 = 0
+        port_id_1 = 1
+
         for i in range(times):
+            vf0_stats = self.vm0_testpmd.get_pmd_stats(port_id_0)
+            b_vf0_rx_pkts, b_vf0_rx_bytes = vf0_stats['RX-packets'], vf0_stats['RX-bytes']
+            vf1_stats = self.vm0_testpmd.get_pmd_stats(port_id_1)
+            b_vf1_tx_pkts, b_vf1_tx_bytes = vf1_stats['TX-packets'], vf1_stats['TX-bytes']
+
             out = testpmd.execute_cmd('port start all')
             self.verify("Checking link statuses" in out, "ERROR: port start all")
             testpmd.execute_cmd('start')
@@ -127,22 +135,20 @@ class TestVfPortStartStop(TestCase):
             out = testpmd.execute_cmd('port stop all')
             self.verify("Checking link statuses" in out, "ERROR: port stop all")
 
-        port_id_0 = 0
-        port_id_1 = 1
-        vf0_stats = self.vm0_testpmd.get_pmd_stats(port_id_0)
-        vf1_stats = self.vm0_testpmd.get_pmd_stats(port_id_1)
-
-        vf0_rx_cnt = vf0_stats['RX-packets']
-        self.verify(vf0_rx_cnt != 0, "no packet was received by vm0_VF0")
-
-        vf0_rx_err = vf0_stats['RX-errors']
-        self.verify(vf0_rx_err == 0, "vm0_VF0 rx-errors")
-    
-        vf1_tx_cnt = vf1_stats['TX-packets']
-        self.verify(vf1_tx_cnt != 0, "no packet was transmitted by vm0_VF1")
-
-        vf1_tx_err = vf1_stats['TX-errors']
-        self.verify(vf1_tx_err == 0, "vm0_VF0 tx-errors")
+            vf0_stats = self.vm0_testpmd.get_pmd_stats(port_id_0)
+            c_vf0_rx_pkts, c_vf0_rx_bytes = vf0_stats['RX-packets'], vf0_stats['RX-bytes']
+            vf1_stats = self.vm0_testpmd.get_pmd_stats(port_id_1)
+            c_vf1_tx_pkts, c_vf1_tx_bytes = vf1_stats['TX-packets'], vf1_stats['TX-bytes']
+
+            #Fortville NIC :when not start, the port is ready to RX but NOT RX packets, 
+            #if you send  packets , then it will be RXed at the NIC's hardware buffer
+            #if buffer is full ,it will show RX-errors
+            #so, we do not check RX-errors
+            #compare vf0 RX-packets RX-bytes ,vf1 Tx-packets TX-bytes
+            self.verify(c_vf0_rx_pkts > b_vf0_rx_pkts, "Packets received error")
+            self.verify(c_vf0_rx_bytes > b_vf0_rx_bytes, "Bytes received error")
+            self.verify(c_vf1_tx_pkts > b_vf1_tx_pkts, "Packets forwarding error")
+            self.verify(c_vf1_tx_bytes > b_vf1_tx_bytes, "Bytes forvwarding error")
 
     def setup_1pf_2vf_1vm_env(self, driver='default'):
 
@@ -215,11 +221,17 @@ class TestVfPortStartStop(TestCase):
 
         self.vm0_testpmd = PmdOutput(self.vm_dut_0)
         self.vm0_testpmd.start_testpmd(VM_CORES_MASK)
+
+        self.vm0_testpmd.execute_cmd('port stop all')
+        self.vm0_testpmd.execute_cmd('port config all crc-strip on')
+        self.vm0_testpmd.execute_cmd('port start all')
         self.vm0_testpmd.execute_cmd('set fwd mac')
+#        self.vm0_testpmd.execute_cmd('start')
 
         time.sleep(2)
 
-        dst_mac = self.vm_dut_0.get_mac_address(self.vm0_dut_ports[0])
+        port_id_0 = 0
+        dst_mac = self.vm0_testpmd.get_port_mac(port_id_0)
         self.send_and_verify(dst_mac, self.vm0_testpmd) 
 
     def tear_down(self):
-- 
1.9.3

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [dts] [PATCH]tests vf_port_start_stop: update test case
  2016-02-26  5:13 [dts] [PATCH]tests vf_port_start_stop: update test case Lijuan Tu
@ 2016-02-26  6:15 ` Pei, Yulong
  2016-02-26  6:22   ` Tu, LijuanX A
  0 siblings, 1 reply; 5+ messages in thread
From: Pei, Yulong @ 2016-02-26  6:15 UTC (permalink / raw)
  To: Tu, LijuanX A, Liu, Yong, dts, Xu, Qian Q

Hi Lijuan,

I think that maybe we keep checking Rx-errors and ignore this issue for Fortville NIC and regard it as a known issue is better.

Best Regards
Yulong Pei

-----Original Message-----
From: Tu, LijuanX A 
Sent: Friday, February 26, 2016 1:13 PM
To: Liu, Yong <yong.liu@intel.com>; dts@dpdk.org; Pei, Yulong <yulong.pei@intel.com>; Xu, Qian Q <qian.q.xu@intel.com>
Cc: Tu, LijuanX A <lijuanx.a.tu@intel.com>
Subject: [dts][PATCH]tests vf_port_start_stop: update test case

Forwarding checking RX-packets, RX-Bytes, Tx-packets, Tx-bytes, Do not checking Rx-errors fortville NIC:
    when not start, the port is ready to RX but NOT RX packets,
    if you send 119 packets at 64B, then it will be RXed at the NIC's hardware buffer
    if the buffer is full, it will show in RX-errors
    so we do not verify Rx-errors.

Signed-off-by: Lijuan Tu <lijuanx.a.tu@intel.com>
---
 tests/TestSuite_vf_port_start_stop.py | 48 ++++++++++++++++++++++-------------
 1 file changed, 30 insertions(+), 18 deletions(-)

diff --git a/tests/TestSuite_vf_port_start_stop.py b/tests/TestSuite_vf_port_start_stop.py
index 0cc7ffd..17aa589 100644
--- a/tests/TestSuite_vf_port_start_stop.py
+++ b/tests/TestSuite_vf_port_start_stop.py
@@ -114,11 +114,19 @@ class TestVfPortStartStop(TestCase):
         testpmd.execute_cmd('port stop all')
         testpmd.execute_cmd('clear port stats all')
 
-    def check_port_start_stop(self, testpmd, times=10):
+    def check_port_start_stop(self, testpmd, times=1000):
         """
         VF port start/stop several times , check if it work well.
         """
+        port_id_0 = 0
+        port_id_1 = 1
+
         for i in range(times):
+            vf0_stats = self.vm0_testpmd.get_pmd_stats(port_id_0)
+            b_vf0_rx_pkts, b_vf0_rx_bytes = vf0_stats['RX-packets'], vf0_stats['RX-bytes']
+            vf1_stats = self.vm0_testpmd.get_pmd_stats(port_id_1)
+            b_vf1_tx_pkts, b_vf1_tx_bytes = vf1_stats['TX-packets'], 
+ vf1_stats['TX-bytes']
+
             out = testpmd.execute_cmd('port start all')
             self.verify("Checking link statuses" in out, "ERROR: port start all")
             testpmd.execute_cmd('start') @@ -127,22 +135,20 @@ class TestVfPortStartStop(TestCase):
             out = testpmd.execute_cmd('port stop all')
             self.verify("Checking link statuses" in out, "ERROR: port stop all")
 
-        port_id_0 = 0
-        port_id_1 = 1
-        vf0_stats = self.vm0_testpmd.get_pmd_stats(port_id_0)
-        vf1_stats = self.vm0_testpmd.get_pmd_stats(port_id_1)
-
-        vf0_rx_cnt = vf0_stats['RX-packets']
-        self.verify(vf0_rx_cnt != 0, "no packet was received by vm0_VF0")
-
-        vf0_rx_err = vf0_stats['RX-errors']
-        self.verify(vf0_rx_err == 0, "vm0_VF0 rx-errors")
-    
-        vf1_tx_cnt = vf1_stats['TX-packets']
-        self.verify(vf1_tx_cnt != 0, "no packet was transmitted by vm0_VF1")
-
-        vf1_tx_err = vf1_stats['TX-errors']
-        self.verify(vf1_tx_err == 0, "vm0_VF0 tx-errors")
+            vf0_stats = self.vm0_testpmd.get_pmd_stats(port_id_0)
+            c_vf0_rx_pkts, c_vf0_rx_bytes = vf0_stats['RX-packets'], vf0_stats['RX-bytes']
+            vf1_stats = self.vm0_testpmd.get_pmd_stats(port_id_1)
+            c_vf1_tx_pkts, c_vf1_tx_bytes = vf1_stats['TX-packets'], 
+ vf1_stats['TX-bytes']
+
+            #Fortville NIC :when not start, the port is ready to RX but NOT RX packets, 
+            #if you send  packets , then it will be RXed at the NIC's hardware buffer
+            #if buffer is full ,it will show RX-errors
+            #so, we do not check RX-errors
+            #compare vf0 RX-packets RX-bytes ,vf1 Tx-packets TX-bytes
+            self.verify(c_vf0_rx_pkts > b_vf0_rx_pkts, "Packets received error")
+            self.verify(c_vf0_rx_bytes > b_vf0_rx_bytes, "Bytes received error")
+            self.verify(c_vf1_tx_pkts > b_vf1_tx_pkts, "Packets forwarding error")
+            self.verify(c_vf1_tx_bytes > b_vf1_tx_bytes, "Bytes 
+ forvwarding error")
 
     def setup_1pf_2vf_1vm_env(self, driver='default'):
 
@@ -215,11 +221,17 @@ class TestVfPortStartStop(TestCase):
 
         self.vm0_testpmd = PmdOutput(self.vm_dut_0)
         self.vm0_testpmd.start_testpmd(VM_CORES_MASK)
+
+        self.vm0_testpmd.execute_cmd('port stop all')
+        self.vm0_testpmd.execute_cmd('port config all crc-strip on')
+        self.vm0_testpmd.execute_cmd('port start all')
         self.vm0_testpmd.execute_cmd('set fwd mac')
+#        self.vm0_testpmd.execute_cmd('start')
 
         time.sleep(2)
 
-        dst_mac = self.vm_dut_0.get_mac_address(self.vm0_dut_ports[0])
+        port_id_0 = 0
+        dst_mac = self.vm0_testpmd.get_port_mac(port_id_0)
         self.send_and_verify(dst_mac, self.vm0_testpmd) 
 
     def tear_down(self):
--
1.9.3

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [dts] [PATCH]tests vf_port_start_stop: update test case
  2016-02-26  6:15 ` Pei, Yulong
@ 2016-02-26  6:22   ` Tu, LijuanX A
  2016-02-26  6:35     ` Xu, Qian Q
  0 siblings, 1 reply; 5+ messages in thread
From: Tu, LijuanX A @ 2016-02-26  6:22 UTC (permalink / raw)
  To: Pei, Yulong, Liu, Yong, dts, Xu, Qian Q

OK ,I Will verify Rx-errors, except fortville

> -----Original Message-----
> From: Pei, Yulong
> Sent: Friday, February 26, 2016 2:16 PM
> To: Tu, LijuanX A; Liu, Yong; dts@dpdk.org; Xu, Qian Q
> Subject: RE: [dts][PATCH]tests vf_port_start_stop: update test case
> 
> Hi Lijuan,
> 
> I think that maybe we keep checking Rx-errors and ignore this issue for
> Fortville NIC and regard it as a known issue is better.
> 
> Best Regards
> Yulong Pei
> 
> -----Original Message-----
> From: Tu, LijuanX A
> Sent: Friday, February 26, 2016 1:13 PM
> To: Liu, Yong <yong.liu@intel.com>; dts@dpdk.org; Pei, Yulong
> <yulong.pei@intel.com>; Xu, Qian Q <qian.q.xu@intel.com>
> Cc: Tu, LijuanX A <lijuanx.a.tu@intel.com>
> Subject: [dts][PATCH]tests vf_port_start_stop: update test case
> 
> Forwarding checking RX-packets, RX-Bytes, Tx-packets, Tx-bytes, Do not
> checking Rx-errors fortville NIC:
>     when not start, the port is ready to RX but NOT RX packets,
>     if you send 119 packets at 64B, then it will be RXed at the NIC's
> hardware buffer
>     if the buffer is full, it will show in RX-errors
>     so we do not verify Rx-errors.
> 
> Signed-off-by: Lijuan Tu <lijuanx.a.tu@intel.com>
> ---
>  tests/TestSuite_vf_port_start_stop.py | 48 ++++++++++++++++++++++-----
> --------
>  1 file changed, 30 insertions(+), 18 deletions(-)
> 
> diff --git a/tests/TestSuite_vf_port_start_stop.py
> b/tests/TestSuite_vf_port_start_stop.py
> index 0cc7ffd..17aa589 100644
> --- a/tests/TestSuite_vf_port_start_stop.py
> +++ b/tests/TestSuite_vf_port_start_stop.py
> @@ -114,11 +114,19 @@ class TestVfPortStartStop(TestCase):
>          testpmd.execute_cmd('port stop all')
>          testpmd.execute_cmd('clear port stats all')
> 
> -    def check_port_start_stop(self, testpmd, times=10):
> +    def check_port_start_stop(self, testpmd, times=1000):
>          """
>          VF port start/stop several times , check if it work well.
>          """
> +        port_id_0 = 0
> +        port_id_1 = 1
> +
>          for i in range(times):
> +            vf0_stats = self.vm0_testpmd.get_pmd_stats(port_id_0)
> +            b_vf0_rx_pkts, b_vf0_rx_bytes = vf0_stats['RX-packets'],
> vf0_stats['RX-bytes']
> +            vf1_stats = self.vm0_testpmd.get_pmd_stats(port_id_1)
> +            b_vf1_tx_pkts, b_vf1_tx_bytes = vf1_stats['TX-packets'],
> + vf1_stats['TX-bytes']
> +
>              out = testpmd.execute_cmd('port start all')
>              self.verify("Checking link statuses" in out, "ERROR: port
> start all")
>              testpmd.execute_cmd('start') @@ -127,22 +135,20 @@ class
> TestVfPortStartStop(TestCase):
>              out = testpmd.execute_cmd('port stop all')
>              self.verify("Checking link statuses" in out, "ERROR: port
> stop all")
> 
> -        port_id_0 = 0
> -        port_id_1 = 1
> -        vf0_stats = self.vm0_testpmd.get_pmd_stats(port_id_0)
> -        vf1_stats = self.vm0_testpmd.get_pmd_stats(port_id_1)
> -
> -        vf0_rx_cnt = vf0_stats['RX-packets']
> -        self.verify(vf0_rx_cnt != 0, "no packet was received by
> vm0_VF0")
> -
> -        vf0_rx_err = vf0_stats['RX-errors']
> -        self.verify(vf0_rx_err == 0, "vm0_VF0 rx-errors")
> -
> -        vf1_tx_cnt = vf1_stats['TX-packets']
> -        self.verify(vf1_tx_cnt != 0, "no packet was transmitted by
> vm0_VF1")
> -
> -        vf1_tx_err = vf1_stats['TX-errors']
> -        self.verify(vf1_tx_err == 0, "vm0_VF0 tx-errors")
> +            vf0_stats = self.vm0_testpmd.get_pmd_stats(port_id_0)
> +            c_vf0_rx_pkts, c_vf0_rx_bytes = vf0_stats['RX-packets'],
> vf0_stats['RX-bytes']
> +            vf1_stats = self.vm0_testpmd.get_pmd_stats(port_id_1)
> +            c_vf1_tx_pkts, c_vf1_tx_bytes = vf1_stats['TX-packets'],
> + vf1_stats['TX-bytes']
> +
> +            #Fortville NIC :when not start, the port is ready to RX
> but NOT RX packets,
> +            #if you send  packets , then it will be RXed at the NIC's
> hardware buffer
> +            #if buffer is full ,it will show RX-errors
> +            #so, we do not check RX-errors
> +            #compare vf0 RX-packets RX-bytes ,vf1 Tx-packets TX-bytes
> +            self.verify(c_vf0_rx_pkts > b_vf0_rx_pkts, "Packets
> received error")
> +            self.verify(c_vf0_rx_bytes > b_vf0_rx_bytes, "Bytes
> received error")
> +            self.verify(c_vf1_tx_pkts > b_vf1_tx_pkts, "Packets
> forwarding error")
> +            self.verify(c_vf1_tx_bytes > b_vf1_tx_bytes, "Bytes
> + forvwarding error")
> 
>      def setup_1pf_2vf_1vm_env(self, driver='default'):
> 
> @@ -215,11 +221,17 @@ class TestVfPortStartStop(TestCase):
> 
>          self.vm0_testpmd = PmdOutput(self.vm_dut_0)
>          self.vm0_testpmd.start_testpmd(VM_CORES_MASK)
> +
> +        self.vm0_testpmd.execute_cmd('port stop all')
> +        self.vm0_testpmd.execute_cmd('port config all crc-strip on')
> +        self.vm0_testpmd.execute_cmd('port start all')
>          self.vm0_testpmd.execute_cmd('set fwd mac')
> +#        self.vm0_testpmd.execute_cmd('start')
> 
>          time.sleep(2)
> 
> -        dst_mac = self.vm_dut_0.get_mac_address(self.vm0_dut_ports[0])
> +        port_id_0 = 0
> +        dst_mac = self.vm0_testpmd.get_port_mac(port_id_0)
>          self.send_and_verify(dst_mac, self.vm0_testpmd)
> 
>      def tear_down(self):
> --
> 1.9.3

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [dts] [PATCH]tests vf_port_start_stop: update test case
  2016-02-26  6:22   ` Tu, LijuanX A
@ 2016-02-26  6:35     ` Xu, Qian Q
  2016-02-26  6:54       ` Tu, LijuanX A
  0 siblings, 1 reply; 5+ messages in thread
From: Xu, Qian Q @ 2016-02-26  6:35 UTC (permalink / raw)
  To: Tu, LijuanX A, Pei, Yulong, Liu, Yong, dts

In fact, maybe we can't call it as an issue, it's a DPDK behavior. Because DPDK has a state that port is ready to RX/TX but not start Polling, so it's a state that Linux driver don't have, then we can't take Linux driver's behavior as reference. 

Thanks
Qian


-----Original Message-----
From: Tu, LijuanX A 
Sent: Friday, February 26, 2016 2:22 PM
To: Pei, Yulong; Liu, Yong; dts@dpdk.org; Xu, Qian Q
Subject: RE: [dts][PATCH]tests vf_port_start_stop: update test case

OK ,I Will verify Rx-errors, except fortville

> -----Original Message-----
> From: Pei, Yulong
> Sent: Friday, February 26, 2016 2:16 PM
> To: Tu, LijuanX A; Liu, Yong; dts@dpdk.org; Xu, Qian Q
> Subject: RE: [dts][PATCH]tests vf_port_start_stop: update test case
> 
> Hi Lijuan,
> 
> I think that maybe we keep checking Rx-errors and ignore this issue 
> for Fortville NIC and regard it as a known issue is better.
> 
> Best Regards
> Yulong Pei
> 
> -----Original Message-----
> From: Tu, LijuanX A
> Sent: Friday, February 26, 2016 1:13 PM
> To: Liu, Yong <yong.liu@intel.com>; dts@dpdk.org; Pei, Yulong 
> <yulong.pei@intel.com>; Xu, Qian Q <qian.q.xu@intel.com>
> Cc: Tu, LijuanX A <lijuanx.a.tu@intel.com>
> Subject: [dts][PATCH]tests vf_port_start_stop: update test case
> 
> Forwarding checking RX-packets, RX-Bytes, Tx-packets, Tx-bytes, Do not 
> checking Rx-errors fortville NIC:
>     when not start, the port is ready to RX but NOT RX packets,
>     if you send 119 packets at 64B, then it will be RXed at the NIC's 
> hardware buffer
>     if the buffer is full, it will show in RX-errors
>     so we do not verify Rx-errors.
> 
> Signed-off-by: Lijuan Tu <lijuanx.a.tu@intel.com>
> ---
>  tests/TestSuite_vf_port_start_stop.py | 48 
> ++++++++++++++++++++++-----
> --------
>  1 file changed, 30 insertions(+), 18 deletions(-)
> 
> diff --git a/tests/TestSuite_vf_port_start_stop.py
> b/tests/TestSuite_vf_port_start_stop.py
> index 0cc7ffd..17aa589 100644
> --- a/tests/TestSuite_vf_port_start_stop.py
> +++ b/tests/TestSuite_vf_port_start_stop.py
> @@ -114,11 +114,19 @@ class TestVfPortStartStop(TestCase):
>          testpmd.execute_cmd('port stop all')
>          testpmd.execute_cmd('clear port stats all')
> 
> -    def check_port_start_stop(self, testpmd, times=10):
> +    def check_port_start_stop(self, testpmd, times=1000):
>          """
>          VF port start/stop several times , check if it work well.
>          """
> +        port_id_0 = 0
> +        port_id_1 = 1
> +
>          for i in range(times):
> +            vf0_stats = self.vm0_testpmd.get_pmd_stats(port_id_0)
> +            b_vf0_rx_pkts, b_vf0_rx_bytes = vf0_stats['RX-packets'],
> vf0_stats['RX-bytes']
> +            vf1_stats = self.vm0_testpmd.get_pmd_stats(port_id_1)
> +            b_vf1_tx_pkts, b_vf1_tx_bytes = vf1_stats['TX-packets'], 
> + vf1_stats['TX-bytes']
> +
>              out = testpmd.execute_cmd('port start all')
>              self.verify("Checking link statuses" in out, "ERROR: port 
> start all")
>              testpmd.execute_cmd('start') @@ -127,22 +135,20 @@ class
> TestVfPortStartStop(TestCase):
>              out = testpmd.execute_cmd('port stop all')
>              self.verify("Checking link statuses" in out, "ERROR: port 
> stop all")
> 
> -        port_id_0 = 0
> -        port_id_1 = 1
> -        vf0_stats = self.vm0_testpmd.get_pmd_stats(port_id_0)
> -        vf1_stats = self.vm0_testpmd.get_pmd_stats(port_id_1)
> -
> -        vf0_rx_cnt = vf0_stats['RX-packets']
> -        self.verify(vf0_rx_cnt != 0, "no packet was received by
> vm0_VF0")
> -
> -        vf0_rx_err = vf0_stats['RX-errors']
> -        self.verify(vf0_rx_err == 0, "vm0_VF0 rx-errors")
> -
> -        vf1_tx_cnt = vf1_stats['TX-packets']
> -        self.verify(vf1_tx_cnt != 0, "no packet was transmitted by
> vm0_VF1")
> -
> -        vf1_tx_err = vf1_stats['TX-errors']
> -        self.verify(vf1_tx_err == 0, "vm0_VF0 tx-errors")
> +            vf0_stats = self.vm0_testpmd.get_pmd_stats(port_id_0)
> +            c_vf0_rx_pkts, c_vf0_rx_bytes = vf0_stats['RX-packets'],
> vf0_stats['RX-bytes']
> +            vf1_stats = self.vm0_testpmd.get_pmd_stats(port_id_1)
> +            c_vf1_tx_pkts, c_vf1_tx_bytes = vf1_stats['TX-packets'], 
> + vf1_stats['TX-bytes']
> +
> +            #Fortville NIC :when not start, the port is ready to RX
> but NOT RX packets,
> +            #if you send  packets , then it will be RXed at the NIC's
> hardware buffer
> +            #if buffer is full ,it will show RX-errors
> +            #so, we do not check RX-errors
> +            #compare vf0 RX-packets RX-bytes ,vf1 Tx-packets TX-bytes
> +            self.verify(c_vf0_rx_pkts > b_vf0_rx_pkts, "Packets
> received error")
> +            self.verify(c_vf0_rx_bytes > b_vf0_rx_bytes, "Bytes
> received error")
> +            self.verify(c_vf1_tx_pkts > b_vf1_tx_pkts, "Packets
> forwarding error")
> +            self.verify(c_vf1_tx_bytes > b_vf1_tx_bytes, "Bytes 
> + forvwarding error")
> 
>      def setup_1pf_2vf_1vm_env(self, driver='default'):
> 
> @@ -215,11 +221,17 @@ class TestVfPortStartStop(TestCase):
> 
>          self.vm0_testpmd = PmdOutput(self.vm_dut_0)
>          self.vm0_testpmd.start_testpmd(VM_CORES_MASK)
> +
> +        self.vm0_testpmd.execute_cmd('port stop all')
> +        self.vm0_testpmd.execute_cmd('port config all crc-strip on')
> +        self.vm0_testpmd.execute_cmd('port start all')
>          self.vm0_testpmd.execute_cmd('set fwd mac')
> +#        self.vm0_testpmd.execute_cmd('start')
> 
>          time.sleep(2)
> 
> -        dst_mac = self.vm_dut_0.get_mac_address(self.vm0_dut_ports[0])
> +        port_id_0 = 0
> +        dst_mac = self.vm0_testpmd.get_port_mac(port_id_0)
>          self.send_and_verify(dst_mac, self.vm0_testpmd)
> 
>      def tear_down(self):
> --
> 1.9.3

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [dts] [PATCH]tests vf_port_start_stop: update test case
  2016-02-26  6:35     ` Xu, Qian Q
@ 2016-02-26  6:54       ` Tu, LijuanX A
  0 siblings, 0 replies; 5+ messages in thread
From: Tu, LijuanX A @ 2016-02-26  6:54 UTC (permalink / raw)
  To: Xu, Qian Q, Pei, Yulong, Liu, Yong, dts

Hi Qian,
  In my opinion, the behavior is only Fortville's . 
  Because I do not get Rx-errors with Niantic  

> -----Original Message-----
> From: Xu, Qian Q
> Sent: Friday, February 26, 2016 2:35 PM
> To: Tu, LijuanX A; Pei, Yulong; Liu, Yong; dts@dpdk.org
> Subject: RE: [dts][PATCH]tests vf_port_start_stop: update test case
> 
> In fact, maybe we can't call it as an issue, it's a DPDK behavior.
> Because DPDK has a state that port is ready to RX/TX but not start
> Polling, so it's a state that Linux driver don't have, then we can't
> take Linux driver's behavior as reference.
> 
> Thanks
> Qian
> 
> 
> -----Original Message-----
> From: Tu, LijuanX A
> Sent: Friday, February 26, 2016 2:22 PM
> To: Pei, Yulong; Liu, Yong; dts@dpdk.org; Xu, Qian Q
> Subject: RE: [dts][PATCH]tests vf_port_start_stop: update test case
> 
> OK ,I Will verify Rx-errors, except fortville
> 
> > -----Original Message-----
> > From: Pei, Yulong
> > Sent: Friday, February 26, 2016 2:16 PM
> > To: Tu, LijuanX A; Liu, Yong; dts@dpdk.org; Xu, Qian Q
> > Subject: RE: [dts][PATCH]tests vf_port_start_stop: update test case
> >
> > Hi Lijuan,
> >
> > I think that maybe we keep checking Rx-errors and ignore this issue
> > for Fortville NIC and regard it as a known issue is better.
> >
> > Best Regards
> > Yulong Pei
> >
> > -----Original Message-----
> > From: Tu, LijuanX A
> > Sent: Friday, February 26, 2016 1:13 PM
> > To: Liu, Yong <yong.liu@intel.com>; dts@dpdk.org; Pei, Yulong
> > <yulong.pei@intel.com>; Xu, Qian Q <qian.q.xu@intel.com>
> > Cc: Tu, LijuanX A <lijuanx.a.tu@intel.com>
> > Subject: [dts][PATCH]tests vf_port_start_stop: update test case
> >
> > Forwarding checking RX-packets, RX-Bytes, Tx-packets, Tx-bytes, Do
> not
> > checking Rx-errors fortville NIC:
> >     when not start, the port is ready to RX but NOT RX packets,
> >     if you send 119 packets at 64B, then it will be RXed at the NIC's
> > hardware buffer
> >     if the buffer is full, it will show in RX-errors
> >     so we do not verify Rx-errors.
> >
> > Signed-off-by: Lijuan Tu <lijuanx.a.tu@intel.com>
> > ---
> >  tests/TestSuite_vf_port_start_stop.py | 48
> > ++++++++++++++++++++++-----
> > --------
> >  1 file changed, 30 insertions(+), 18 deletions(-)
> >
> > diff --git a/tests/TestSuite_vf_port_start_stop.py
> > b/tests/TestSuite_vf_port_start_stop.py
> > index 0cc7ffd..17aa589 100644
> > --- a/tests/TestSuite_vf_port_start_stop.py
> > +++ b/tests/TestSuite_vf_port_start_stop.py
> > @@ -114,11 +114,19 @@ class TestVfPortStartStop(TestCase):
> >          testpmd.execute_cmd('port stop all')
> >          testpmd.execute_cmd('clear port stats all')
> >
> > -    def check_port_start_stop(self, testpmd, times=10):
> > +    def check_port_start_stop(self, testpmd, times=1000):
> >          """
> >          VF port start/stop several times , check if it work well.
> >          """
> > +        port_id_0 = 0
> > +        port_id_1 = 1
> > +
> >          for i in range(times):
> > +            vf0_stats = self.vm0_testpmd.get_pmd_stats(port_id_0)
> > +            b_vf0_rx_pkts, b_vf0_rx_bytes = vf0_stats['RX-packets'],
> > vf0_stats['RX-bytes']
> > +            vf1_stats = self.vm0_testpmd.get_pmd_stats(port_id_1)
> > +            b_vf1_tx_pkts, b_vf1_tx_bytes = vf1_stats['TX-packets'],
> > + vf1_stats['TX-bytes']
> > +
> >              out = testpmd.execute_cmd('port start all')
> >              self.verify("Checking link statuses" in out, "ERROR:
> port
> > start all")
> >              testpmd.execute_cmd('start') @@ -127,22 +135,20 @@ class
> > TestVfPortStartStop(TestCase):
> >              out = testpmd.execute_cmd('port stop all')
> >              self.verify("Checking link statuses" in out, "ERROR:
> port
> > stop all")
> >
> > -        port_id_0 = 0
> > -        port_id_1 = 1
> > -        vf0_stats = self.vm0_testpmd.get_pmd_stats(port_id_0)
> > -        vf1_stats = self.vm0_testpmd.get_pmd_stats(port_id_1)
> > -
> > -        vf0_rx_cnt = vf0_stats['RX-packets']
> > -        self.verify(vf0_rx_cnt != 0, "no packet was received by
> > vm0_VF0")
> > -
> > -        vf0_rx_err = vf0_stats['RX-errors']
> > -        self.verify(vf0_rx_err == 0, "vm0_VF0 rx-errors")
> > -
> > -        vf1_tx_cnt = vf1_stats['TX-packets']
> > -        self.verify(vf1_tx_cnt != 0, "no packet was transmitted by
> > vm0_VF1")
> > -
> > -        vf1_tx_err = vf1_stats['TX-errors']
> > -        self.verify(vf1_tx_err == 0, "vm0_VF0 tx-errors")
> > +            vf0_stats = self.vm0_testpmd.get_pmd_stats(port_id_0)
> > +            c_vf0_rx_pkts, c_vf0_rx_bytes = vf0_stats['RX-packets'],
> > vf0_stats['RX-bytes']
> > +            vf1_stats = self.vm0_testpmd.get_pmd_stats(port_id_1)
> > +            c_vf1_tx_pkts, c_vf1_tx_bytes = vf1_stats['TX-packets'],
> > + vf1_stats['TX-bytes']
> > +
> > +            #Fortville NIC :when not start, the port is ready to RX
> > but NOT RX packets,
> > +            #if you send  packets , then it will be RXed at the
> NIC's
> > hardware buffer
> > +            #if buffer is full ,it will show RX-errors
> > +            #so, we do not check RX-errors
> > +            #compare vf0 RX-packets RX-bytes ,vf1 Tx-packets TX-
> bytes
> > +            self.verify(c_vf0_rx_pkts > b_vf0_rx_pkts, "Packets
> > received error")
> > +            self.verify(c_vf0_rx_bytes > b_vf0_rx_bytes, "Bytes
> > received error")
> > +            self.verify(c_vf1_tx_pkts > b_vf1_tx_pkts, "Packets
> > forwarding error")
> > +            self.verify(c_vf1_tx_bytes > b_vf1_tx_bytes, "Bytes
> > + forvwarding error")
> >
> >      def setup_1pf_2vf_1vm_env(self, driver='default'):
> >
> > @@ -215,11 +221,17 @@ class TestVfPortStartStop(TestCase):
> >
> >          self.vm0_testpmd = PmdOutput(self.vm_dut_0)
> >          self.vm0_testpmd.start_testpmd(VM_CORES_MASK)
> > +
> > +        self.vm0_testpmd.execute_cmd('port stop all')
> > +        self.vm0_testpmd.execute_cmd('port config all crc-strip on')
> > +        self.vm0_testpmd.execute_cmd('port start all')
> >          self.vm0_testpmd.execute_cmd('set fwd mac')
> > +#        self.vm0_testpmd.execute_cmd('start')
> >
> >          time.sleep(2)
> >
> > -        dst_mac =
> self.vm_dut_0.get_mac_address(self.vm0_dut_ports[0])
> > +        port_id_0 = 0
> > +        dst_mac = self.vm0_testpmd.get_port_mac(port_id_0)
> >          self.send_and_verify(dst_mac, self.vm0_testpmd)
> >
> >      def tear_down(self):
> > --
> > 1.9.3

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2016-02-26  6:54 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2016-02-26  5:13 [dts] [PATCH]tests vf_port_start_stop: update test case Lijuan Tu
2016-02-26  6:15 ` Pei, Yulong
2016-02-26  6:22   ` Tu, LijuanX A
2016-02-26  6:35     ` Xu, Qian Q
2016-02-26  6:54       ` Tu, LijuanX A

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).