* [dts] [PATCH] jumboframes: Fix bug for fm10k NIC
@ 2015-10-14 3:15 Michael Qiu
2015-10-14 10:19 ` Liu, Yong
0 siblings, 1 reply; 4+ messages in thread
From: Michael Qiu @ 2015-10-14 3:15 UTC (permalink / raw)
To: dts
For Ruby Rapid, in Testpoint should run "set port config 1,5 max_frame_size 9000"
and "set port config 1,5,20,22 rx_cut_through off"
Singed-off-by: Michael Qiu <michael.qiu@intel.com>
---
tests/TestSuite_jumboframes.py | 12 ++++++++----
1 file changed, 8 insertions(+), 4 deletions(-)
diff --git a/tests/TestSuite_jumboframes.py b/tests/TestSuite_jumboframes.py
index c5c70b9..e9a6abf 100644
--- a/tests/TestSuite_jumboframes.py
+++ b/tests/TestSuite_jumboframes.py
@@ -93,7 +93,9 @@ class TestJumboframes(TestCase):
p1rx_err -= gp1rx_err
if received:
- self.verify(p0tx_pkts == p1rx_pkts and p0tx_bytes == pktsize and p1rx_bytes == pktsize,
+ #some nic like RRC always strip CRC, so it should be pktsize - 4
+ size_equal = p0tx_bytes == p1rx_bytes and (p1rx_bytes == pktsize or p1rx_bytes == pktsize - 4)
+ self.verify(p0tx_pkts == p1rx_pkts and size_equal,
"packet pass assert error")
else:
#self.verify(p0tx_pkts == p1rx_pkts and (p1rx_err == 1 or p1rx_pkts == 0),
@@ -152,10 +154,12 @@ class TestJumboframes(TestCase):
This case aims to test transmitting jumbo frame packet on testpmd without
jumbo frame support.
"""
- self.pmdout.start_testpmd("Default", "--max-pkt-len=%d" % (ETHER_STANDARD_MTU))
+ # RRC has no ability to set the max pkt len to hardware
if self.nic == "redrockcanyou":
- self.dut.send_expect("set promisc all off", "testpmd> ")
- self.dut.send_expect("set fwd mac", "testpmd> ")
+ print dts.RED("fm10k not support this case\n")
+ return
+ self.pmdout.start_testpmd("Default", "--max-pkt-len=%d" % (ETHER_STANDARD_MTU))
+ self.dut.send_expect("set fwd mac", "testpmd> ")
self.dut.send_expect("start", "testpmd> ")
self.jumboframes_send_packet(ETHER_STANDARD_MTU + 1, False)
--
1.9.3
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [dts] [PATCH] jumboframes: Fix bug for fm10k NIC
2015-10-14 3:15 [dts] [PATCH] jumboframes: Fix bug for fm10k NIC Michael Qiu
@ 2015-10-14 10:19 ` Liu, Yong
2015-11-04 5:45 ` Qiu, Michael
0 siblings, 1 reply; 4+ messages in thread
From: Liu, Yong @ 2015-10-14 10:19 UTC (permalink / raw)
To: Qiu, Michael, dts
Hi Michael,
Same comment, we have excel file tracking those features what we do not support.
For the RRC Testpoint, I think we need implement one structure for this. We cannot reply on manual commands to run DTS.
> -----Original Message-----
> From: dts [mailto:dts-bounces@dpdk.org] On Behalf Of Michael Qiu
> Sent: Wednesday, October 14, 2015 4:15 AM
> To: dts@dpdk.org
> Subject: [dts] [PATCH] jumboframes: Fix bug for fm10k NIC
>
> For Ruby Rapid, in Testpoint should run "set port config 1,5
> max_frame_size 9000"
> and "set port config 1,5,20,22 rx_cut_through off"
>
> Singed-off-by: Michael Qiu <michael.qiu@intel.com>
> ---
> tests/TestSuite_jumboframes.py | 12 ++++++++----
> 1 file changed, 8 insertions(+), 4 deletions(-)
>
> diff --git a/tests/TestSuite_jumboframes.py
> b/tests/TestSuite_jumboframes.py
> index c5c70b9..e9a6abf 100644
> --- a/tests/TestSuite_jumboframes.py
> +++ b/tests/TestSuite_jumboframes.py
> @@ -93,7 +93,9 @@ class TestJumboframes(TestCase):
> p1rx_err -= gp1rx_err
>
> if received:
> - self.verify(p0tx_pkts == p1rx_pkts and p0tx_bytes == pktsize
> and p1rx_bytes == pktsize,
> + #some nic like RRC always strip CRC, so it should be pktsize
> - 4
> + size_equal = p0tx_bytes == p1rx_bytes and (p1rx_bytes ==
> pktsize or p1rx_bytes == pktsize - 4)
> + self.verify(p0tx_pkts == p1rx_pkts and size_equal,
> "packet pass assert error")
> else:
> #self.verify(p0tx_pkts == p1rx_pkts and (p1rx_err == 1 or
> p1rx_pkts == 0),
> @@ -152,10 +154,12 @@ class TestJumboframes(TestCase):
> This case aims to test transmitting jumbo frame packet on testpmd
> without
> jumbo frame support.
> """
> - self.pmdout.start_testpmd("Default", "--max-pkt-len=%d" %
> (ETHER_STANDARD_MTU))
> + # RRC has no ability to set the max pkt len to hardware
> if self.nic == "redrockcanyou":
> - self.dut.send_expect("set promisc all off", "testpmd> ")
> - self.dut.send_expect("set fwd mac", "testpmd> ")
> + print dts.RED("fm10k not support this case\n")
> + return
> + self.pmdout.start_testpmd("Default", "--max-pkt-len=%d" %
> (ETHER_STANDARD_MTU))
> + self.dut.send_expect("set fwd mac", "testpmd> ")
> self.dut.send_expect("start", "testpmd> ")
>
> self.jumboframes_send_packet(ETHER_STANDARD_MTU + 1, False)
> --
> 1.9.3
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [dts] [PATCH] jumboframes: Fix bug for fm10k NIC
2015-10-14 10:19 ` Liu, Yong
@ 2015-11-04 5:45 ` Qiu, Michael
2015-11-04 7:10 ` Liu, Yong
0 siblings, 1 reply; 4+ messages in thread
From: Qiu, Michael @ 2015-11-04 5:45 UTC (permalink / raw)
To: Liu, Yong, dts
Hi, Marvin
Just as I said in other thread, I will make a patch to update the list.
And what about this patch?
Thanks,
Michael
On 2015/10/14 18:19, Liu, Yong wrote:
> Hi Michael,
> Same comment, we have excel file tracking those features what we do not support.
> For the RRC Testpoint, I think we need implement one structure for this. We cannot reply on manual commands to run DTS.
>
>> -----Original Message-----
>> From: dts [mailto:dts-bounces@dpdk.org] On Behalf Of Michael Qiu
>> Sent: Wednesday, October 14, 2015 4:15 AM
>> To: dts@dpdk.org
>> Subject: [dts] [PATCH] jumboframes: Fix bug for fm10k NIC
>>
>> For Ruby Rapid, in Testpoint should run "set port config 1,5
>> max_frame_size 9000"
>> and "set port config 1,5,20,22 rx_cut_through off"
>>
>> Singed-off-by: Michael Qiu <michael.qiu@intel.com>
>> ---
>> tests/TestSuite_jumboframes.py | 12 ++++++++----
>> 1 file changed, 8 insertions(+), 4 deletions(-)
>>
>> diff --git a/tests/TestSuite_jumboframes.py
>> b/tests/TestSuite_jumboframes.py
>> index c5c70b9..e9a6abf 100644
>> --- a/tests/TestSuite_jumboframes.py
>> +++ b/tests/TestSuite_jumboframes.py
>> @@ -93,7 +93,9 @@ class TestJumboframes(TestCase):
>> p1rx_err -= gp1rx_err
>>
>> if received:
>> - self.verify(p0tx_pkts == p1rx_pkts and p0tx_bytes == pktsize
>> and p1rx_bytes == pktsize,
>> + #some nic like RRC always strip CRC, so it should be pktsize
>> - 4
>> + size_equal = p0tx_bytes == p1rx_bytes and (p1rx_bytes ==
>> pktsize or p1rx_bytes == pktsize - 4)
>> + self.verify(p0tx_pkts == p1rx_pkts and size_equal,
>> "packet pass assert error")
>> else:
>> #self.verify(p0tx_pkts == p1rx_pkts and (p1rx_err == 1 or
>> p1rx_pkts == 0),
>> @@ -152,10 +154,12 @@ class TestJumboframes(TestCase):
>> This case aims to test transmitting jumbo frame packet on testpmd
>> without
>> jumbo frame support.
>> """
>> - self.pmdout.start_testpmd("Default", "--max-pkt-len=%d" %
>> (ETHER_STANDARD_MTU))
>> + # RRC has no ability to set the max pkt len to hardware
>> if self.nic == "redrockcanyou":
>> - self.dut.send_expect("set promisc all off", "testpmd> ")
>> - self.dut.send_expect("set fwd mac", "testpmd> ")
>> + print dts.RED("fm10k not support this case\n")
>> + return
>> + self.pmdout.start_testpmd("Default", "--max-pkt-len=%d" %
>> (ETHER_STANDARD_MTU))
>> + self.dut.send_expect("set fwd mac", "testpmd> ")
>> self.dut.send_expect("start", "testpmd> ")
>>
>> self.jumboframes_send_packet(ETHER_STANDARD_MTU + 1, False)
>> --
>> 1.9.3
>
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [dts] [PATCH] jumboframes: Fix bug for fm10k NIC
2015-11-04 5:45 ` Qiu, Michael
@ 2015-11-04 7:10 ` Liu, Yong
0 siblings, 0 replies; 4+ messages in thread
From: Liu, Yong @ 2015-11-04 7:10 UTC (permalink / raw)
To: Qiu, Michael, dts
Hi Michael,
That's fine. Applied this patch, thanks.
On 11/04/2015 01:45 PM, Qiu, Michael wrote:
> Hi, Marvin
>
> Just as I said in other thread, I will make a patch to update the list.
>
> And what about this patch?
>
> Thanks,
> Michael
>
> On 2015/10/14 18:19, Liu, Yong wrote:
>> Hi Michael,
>> Same comment, we have excel file tracking those features what we do not support.
>> For the RRC Testpoint, I think we need implement one structure for this. We cannot reply on manual commands to run DTS.
>>
>>> -----Original Message-----
>>> From: dts [mailto:dts-bounces@dpdk.org] On Behalf Of Michael Qiu
>>> Sent: Wednesday, October 14, 2015 4:15 AM
>>> To: dts@dpdk.org
>>> Subject: [dts] [PATCH] jumboframes: Fix bug for fm10k NIC
>>>
>>> For Ruby Rapid, in Testpoint should run "set port config 1,5
>>> max_frame_size 9000"
>>> and "set port config 1,5,20,22 rx_cut_through off"
>>>
>>> Singed-off-by: Michael Qiu <michael.qiu@intel.com>
>>> ---
>>> tests/TestSuite_jumboframes.py | 12 ++++++++----
>>> 1 file changed, 8 insertions(+), 4 deletions(-)
>>>
>>> diff --git a/tests/TestSuite_jumboframes.py
>>> b/tests/TestSuite_jumboframes.py
>>> index c5c70b9..e9a6abf 100644
>>> --- a/tests/TestSuite_jumboframes.py
>>> +++ b/tests/TestSuite_jumboframes.py
>>> @@ -93,7 +93,9 @@ class TestJumboframes(TestCase):
>>> p1rx_err -= gp1rx_err
>>>
>>> if received:
>>> - self.verify(p0tx_pkts == p1rx_pkts and p0tx_bytes == pktsize
>>> and p1rx_bytes == pktsize,
>>> + #some nic like RRC always strip CRC, so it should be pktsize
>>> - 4
>>> + size_equal = p0tx_bytes == p1rx_bytes and (p1rx_bytes ==
>>> pktsize or p1rx_bytes == pktsize - 4)
>>> + self.verify(p0tx_pkts == p1rx_pkts and size_equal,
>>> "packet pass assert error")
>>> else:
>>> #self.verify(p0tx_pkts == p1rx_pkts and (p1rx_err == 1 or
>>> p1rx_pkts == 0),
>>> @@ -152,10 +154,12 @@ class TestJumboframes(TestCase):
>>> This case aims to test transmitting jumbo frame packet on testpmd
>>> without
>>> jumbo frame support.
>>> """
>>> - self.pmdout.start_testpmd("Default", "--max-pkt-len=%d" %
>>> (ETHER_STANDARD_MTU))
>>> + # RRC has no ability to set the max pkt len to hardware
>>> if self.nic == "redrockcanyou":
>>> - self.dut.send_expect("set promisc all off", "testpmd> ")
>>> - self.dut.send_expect("set fwd mac", "testpmd> ")
>>> + print dts.RED("fm10k not support this case\n")
>>> + return
>>> + self.pmdout.start_testpmd("Default", "--max-pkt-len=%d" %
>>> (ETHER_STANDARD_MTU))
>>> + self.dut.send_expect("set fwd mac", "testpmd> ")
>>> self.dut.send_expect("start", "testpmd> ")
>>>
>>> self.jumboframes_send_packet(ETHER_STANDARD_MTU + 1, False)
>>> --
>>> 1.9.3
>
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2015-11-04 7:08 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2015-10-14 3:15 [dts] [PATCH] jumboframes: Fix bug for fm10k NIC Michael Qiu
2015-10-14 10:19 ` Liu, Yong
2015-11-04 5:45 ` Qiu, Michael
2015-11-04 7:10 ` Liu, Yong
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).