Lisa Hsu wrote:
where did you get your disk image from? is it the old image from m5
1.1, or did you make it off of linux-dist? if it's the old one,
that's why there's a mismatch....you should make your image with
linux-dist.
lisa
No I used the image from the following links on the download page.
* M5 2.0b1 <http://www.m5sim.org/dist/current/m5_2.0b1.tar.bz2> --
Includes testing infrastructure, and Alpha PAL/Console code
* M5 2.0b1 Full System Files
<http://www.m5sim.org/dist/current/m5_system_2.0b1.tar.bz2> --
Pre-compiled Linux kernels, PALcode/Console code, and a filesystem
I'll make an image with linux-dist and try it again. Thanks a lot.
Ranjith
On 11/30/06, *Ranjith Subramanian* <[EMAIL PROTECTED]
<mailto:[EMAIL PROTECTED]>> wrote:
Yes, I didn't realize that there were two machines being simulated. I
tried running the benchmark again connecting to both consoles and this
the output I got on the other console.
**** output fromt the other console
mounting filesystems...
EXT2-fs warning: checktime reached, running e2fsck is recommended
loading script...
setting up network...
eth0: link now 1000F mbps, full duplex and up.
running netserver...
/tmp/script: line 22: /benchmarks/netperf-bin/netserver: No such
file or
directory
signal client to begin...done.
starting bash...
****
I checked up and found that there is no directory called
/benchmarks/netperf-bin on the disk but when I grep for netperf-bin in
the boot directory containing the .rCS files I get a lot of references
to it.
-bash-3.00$ grep netperf-bin *
nat-netperf-maerts-client.rcS:BINARY=/benchmarks/netperf-bin/netperf
nat-netperf-server.rcS:/benchmarks/netperf-bin/netserver
nat-netperf-stream-client.rcS:BINARY=/benchmarks/netperf-bin/netperf
netperf-rr.rcS:BINARY=/benchmarks/netperf-bin/netperf
netperf-server.rcS:/benchmarks/netperf-bin/netserver
netperf-stream-client.rcS:BINARY=/benchmarks/netperf-bin/netperf
netperf-stream-nt-client.rcS:BINARY=/benchmarks/netperf-bin/netperf
I changed all occurences of netperf-bin to netperf and the benchmark
runs to completion without a problem. Has anyone else faced this
problem
before?
Ranjith
Steve Reinhardt wrote:
> Do you get two console.*.sim_console output files? If so, what does
> the other one say?
>
> Ranjith Subramanian wrote:
>
>> Yes, it is 2.0 beta1.
>>
>> My command line is
>> ./m5.opt ../configs/examples/fs.py -b NetperfMaerts
>>
>> My disk image is in /tmp/ranji/dist/m5/system. I modified
>> SysPaths.py to point to this directory.
>>
>> Ranjith
>>
>> Lisa Hsu wrote:
>>
>>> i've never seen that error before...
>>>
>>> can you provide some extra details - is this the
2.0beta1? where is
>>> your disk image from? what is your command line?
>>>
>>> lisa
>>>
>>> On 11/29/06, *Ranjith Subramanian* < [EMAIL PROTECTED]
<mailto:[EMAIL PROTECTED]>
>>> <mailto:[EMAIL PROTECTED] <mailto:[EMAIL PROTECTED]>>> wrote:
>>>
>>> I've installed m52.0 and am trying to run the NetperfMaerts
>>> benchmark in
>>> FS mode.
>>> I simulation starts and hangs with the following output.
I'm not
>>> sure
>>> what is going wong here.
>>>
>>> mounting filesystems...
>>> EXT2-fs warning: checktime reached, running e2fsck is
recommended
>>> loading script...
>>> setting up network...
>>> eth0: link now 1000F mbps, full duplex and up.
>>> waiting for server...server ready
>>> starting test...
>>> netperf warmup
>>> /benchmarks/netperf/netperf -H 10.0.0.1 <http://10.0.0.1>
< http://10.0.0.1> -t
>>> TCP_MAERTS -l -100k
>>> establish_control: control socket connect failed: Connection
>>> refused
>>> Are you sure there is a netserver running on 10.0.0.1
<http://10.0.0.1>
>>> <http://10.0.0.1> at port 12865?
>>> netperf benchmark... starting 1...
>>> /benchmarks/netperf/netperf -H 10.0.0.1 <http://10.0.0.1>
<http://10.0.0.1> -t
>>> TCP_MAERTS -k16384,0
>>> -K16384,0 -- -m 65536 -M 65536 -s 262144 -S 262144
>>> configuring checkpointing at 1400000000...
>>> done configuring checkpointing...
>>> establish_control: control socket connect failed: Connection
>>> refused
>>> Are you sure there is a netserver running on 10.0.0.1
<http://10.0.0.1>
>>> <http://10.0.0.1> at port 12865?
>>>
>>>
>>> Ranjith
>>>
>>>
>>> _______________________________________________
>>> m5-users mailing list
>>> m5-users@m5sim.org <mailto:m5-users@m5sim.org> <mailto:
m5-users@m5sim.org <mailto:m5-users@m5sim.org>>
>>> http://m5sim.org/cgi-bin/mailman/listinfo/m5-users
>>>
>>>
>>>
------------------------------------------------------------------------
>>>
>>>
>>> _______________________________________________
>>> m5-users mailing list
>>> m5-users@m5sim.org <mailto:m5-users@m5sim.org>
>>> http://m5sim.org/cgi-bin/mailman/listinfo/m5-users
>>>
>>
>> _______________________________________________
>> m5-users mailing list
>> m5-users@m5sim.org <mailto:m5-users@m5sim.org>
>> http://m5sim.org/cgi-bin/mailman/listinfo/m5-users
>
> _______________________________________________
> m5-users mailing list
> m5-users@m5sim.org <mailto:m5-users@m5sim.org>
> http://m5sim.org/cgi-bin/mailman/listinfo/m5-users
_______________________________________________
m5-users mailing list
m5-users@m5sim.org <mailto:m5-users@m5sim.org>
http://m5sim.org/cgi-bin/mailman/listinfo/m5-users
<http://m5sim.org/cgi-bin/mailman/listinfo/m5-users>
------------------------------------------------------------------------
_______________________________________________
m5-users mailing list
m5-users@m5sim.org
http://m5sim.org/cgi-bin/mailman/listinfo/m5-users
_______________________________________________
m5-users mailing list
m5-users@m5sim.org
http://m5sim.org/cgi-bin/mailman/listinfo/m5-users