[dts] [patch] ironpond , twinpond and springfountain 's mbuf is 2048, not 1024

Liu, Yong yong.liu at intel.com
Tue Oct 20 05:19:04 CEST 2015


I'm not sure why use 2048 for some NICs. Anybody known about this?
Scatter feature is used when received packets over the size of mbuf. 
I think 1024 and 2048 should be the same to measure this feature.

> -----Original Message-----
> From: Qiu, Michael
> Sent: Tuesday, October 20, 2015 11:05 AM
> To: Liu, Yong
> Cc: Tu, LijuanX A; dts at dpdk.org
> Subject: Re: [dts] [patch] ironpond , twinpond and springfountain 's mbuf
> is 2048, not 1024
> 
> I have a question, which nic's default mbuf is 1024?
> Thanks,
> Michael
> 
> 发自我的 iPhone
> 
> > 在 2015年10月20日,上午10:23,Liu, Yong <yong.liu at intel.com> 写道:
> >
> > Applied. Thx.
> >
> >> -----Original Message-----
> >> From: dts [mailto:dts-bounces at dpdk.org] On Behalf Of lijuan tu
> >> Sent: Thursday, October 15, 2015 12:25 PM
> >> To: dts at dpdk.org
> >> Subject: [dts] [patch] ironpond , twinpond and springfountain 's mbuf
> is
> >> 2048, not 1024
> >>
> >> Signed-off-by: lijuan tu <lijuanx.a.tu at intel.com>
> >> ---
> >> tests/TestSuite_scatter.py | 2 +-
> >> 1 file changed, 1 insertion(+), 1 deletion(-)
> >>
> >> diff --git a/tests/TestSuite_scatter.py b/tests/TestSuite_scatter.py
> >> index abb1a08..1bec2c8 100644
> >> --- a/tests/TestSuite_scatter.py
> >> +++ b/tests/TestSuite_scatter.py
> >> @@ -55,7 +55,7 @@ class TestScatter(TestCase):
> >>         # Verify that enough ports are available
> >>         self.verify(len(dutPorts) >= 2, "Insufficient ports")
> >>         self.pmdout = PmdOutput(self.dut)
> >> -        if self.nic in ["niantic", "fortville_eagle",
> "fortville_spirit",
> >> "fortville_spirit_single", "redrockcanyou"]:
> >> +        if self.nic in ["niantic", "fortville_eagle",
> "fortville_spirit",
> >> "fortville_spirit_single", "redrockcanyou", "ironpond", "twinpond",
> >> "springfountain"]:
> >>             self.mbsize = 2048
> >>         else:
> >>             self.mbsize = 1024
> >> --
> >> 2.1.0
> >


More information about the dts mailing list