[netperf-talk] 回复：Re: netperf UDP RR problem

[Rick Jones]

Andrew Gallatin wrote:
> Rick Jones wrote:
> 
> 
>>svn co http://www.netperf.org/svn/netperf2/trunk  netperf2_trunk
>>
>>and see if you get the same error when you run that on both ends rather than the 
>>2.4.4 bits.
> 
> 
> Linux hides its errno definitions in many (9 on RHEL5) files.
> I'd do:
> 
> % find /usr/include/ -name '*errno*' | xargs exec grep 14
> xargs: exec: No such file or directory
> <10:29am>asus1/gallatin:~>find /usr/include/ -name '*errno*' | xargs grep 14
> /usr/include/asm-generic/errno.h:#define        EALREADY        114 
> /* Operation already in progress */
> /usr/include/asm-generic/errno-base.h:#define   EFAULT          14 
> /* Bad address */

I thought it might be EFAULT based on what HP-UX does but couldn't recall if the 
lower EMUMBLE values were based on accident or actual specification :)

> So the "14" is the "Bad Address" quoted in the original message,
> and it sounds like something is really wrong.

Indeed.  I have vague recollections of remote netservers efaulting and being 
fixed in top-of-trunk so  your expeience with UDP_RR below:

> 
> On my RHEL5 x64 machines, svn works fine:
> 
> % build.rhel5>./src/netperf -Hasus2-m -t UDP_RR -l 20 -- -r 128,8192
> UDP REQUEST/RESPONSE TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 
> asus2-m.sw.myri.com (10.0.130.217) port 0 AF_INET
> Local /Remote
> Socket Size   Request  Resp.   Elapsed  Trans.
> Send   Recv   Size     Size    Time     Rate
> bytes  Bytes  bytes    bytes   secs.    per sec
> 
> 126976 126976 128      8192    20.00    20272.35
> 126976 126976

is consistent with that.

happy benchmarking,

rick