Alec,

That seems to be the problem:

?chdir /usr/bof
!chdir ok
?progdev casper_test_2013_Aug_21_1251.bof
#log info 948058172898 raw attempting\_to\_empty\_fpga
#log info 948058172899 raw
attempting\_to\_program\_casper_test_2013_Aug_21_1251
.bof
#log info 948058172988 raw
attempting\_to\_program\_bitstream\_of\_19586188\_byt
es\_to\_device\_/dev/roach/config
!progdev ok

Also on the ROACH board,

/dev/roach # cd /proc/458
/proc/458 # ls -la
total 0
dr-xr-xr-x    7 root     root             0 Jan 16 21:10 .
dr-xr-xr-x   45 root     root             0 Jan  1  1970 ..
-r--------    1 root     root             0 Jan 16 21:28 auxv
--w-------    1 root     root             0 Jan 16 21:28 clear_refs
-r--r--r--    1 root     root             0 Jan 16 21:10 cmdline
-rw-r--r--    1 root     root             0 Jan 16 21:28 comm
-rw-r--r--    1 root     root             0 Jan 16 21:28 coredump_filter
lrwxrwxrwx    1 root     root             0 Jan 16 21:28 cwd -> /var/tmp
...

so it may be the case that cwd is pointing to the wrong directory.  How do
I change cwd permanently?

-Joe Kujawski


On Fri, Sep 6, 2013 at 10:35 AM, Alec Rust <[email protected]> wrote:

> Ok, so either your file is broken or the present working directory of
> tcpborphserver3 is not correct. You can see the current working directory
> by:
> cd /proc/PID
> ls -la
> cwd will be a link to the current working directory.
>
> In the telnet session run ?chdir /usr/bof
> Then try again. If this does not work try md5sum on your boffile and
> compare it to the original.
>
>
>
>
> Alec,
>
> Here is the output:
>
> /dev/roach # ps ax
> PID   USER     TIME   COMMAND
>     1 root       0:01 init
>     2 root       0:00 [kthreadd]
>     3 root       0:00 [ksoftirqd/0]
>     4 root       0:00 [kworker/0:0]
>     5 root       0:00 [kworker/u:0]
>     6 root       0:00 [khelper]
>     7 root       0:00 [kworker/u:1]
>    92 root       0:00 [sync_supers]
>    94 root       0:00 [bdi-default]
>    95 root       0:00 [kintegrityd]
>    97 root       0:00 [kblockd]
>   138 root       0:00 [rpciod]
>   140 root       0:00 [kworker/0:1]
>   166 root       0:00 [khungtaskd]
>   170 root       0:00 [kswapd0]
>   171 root       0:00 [ksmd]
>   233 root       0:00 [fsnotify_mark]
>    237 root       0:00 [nfsiod]
>   336 root       0:00 [mtdblock0]
>   341 root       0:00 [mtdblock1]
>   346 root       0:00 [mtdblock2]
>   351 root       0:00 [mtdblock3]
>   356 root       0:00 [mtdblock4]
>   361 root       0:00 [mtdblock5]
>   436 root       0:00 [deferwq]
>   443 root       0:00 [jffs2_gcd_mtd2]
>   454 root       0:00 {rc.tcpborphserv} /bin/sh /etc/rc.tcpborphserver3
>   455 root       0:00 /sbin/getty 115200 ttyS0
>   456 root       0:00 /sbin/getty 115200 ttyS1
>   458 root       0:00 tcpborphserver3 -f -b /usr/bof
>   466 root       0:00 telnetd
>   467 root       0:00 -sh
>   470 root       0:00 ps ax
> /dev/roach # strace -p 458
> Process 458 attached - interrupt to quit
> select(27, [25 26], [], NULL, NULL)     = 1 (in [26])
> rt_sigprocmask(SIG_SETMASK, [CHLD], NULL, 8) = 0
> read(26, "?progdev casper_test_2013_Aug_21"..., 492) = 43
> gettimeofday({948057315, 506178}, NULL) = 0
> gettimeofday({948057315, 507016}, NULL) = 0
> open("./casper_test_2013_Aug_21_1251.bof", O_RDONLY) = -1 ENOENT (No such
> file or directory)
> gettimeofday({948057315, 508973}, NULL) = 0
> rt_sigprocmask(SIG_SETMASK, [], [CHLD], 8) = 0
> select(27, [25 26], [26], NULL, NULL)   = 1 (out [26])
> rt_sigprocmask(SIG_SETMASK, [CHLD], NULL, 8) = 0
> send(26, "#log info 948057315506 raw attem"..., 276,
> MSG_DONTWAIT|MSG_NOSIGNAL) = 276
> rt_sigprocmask(SIG_SETMASK, [], [CHLD], 8) = 0
> select(27, [25 26], [], NULL, NULL
>
> -Joe Kujawski
>
>
> On Fri, Sep 6, 2013 at 10:17 AM, Alec Rust <[email protected]> wrote:
>
>> Please see if there are 2 device files in /dev/roach. config and mem
>> should be there. If they are not there you are running a kernel with an
>> incorrect build option.
>>
>> If this is the case you can quickly fix it by following steps 3.3 to 3.5
>> under "Steps to follow" on this wiki page:
>> https://casper.berkeley.edu/wiki/FPGA_Device_Driver_Memo
>>
>> If you do have those files please telnet into the roach and find the PID
>> of tcpborphserver3 as follows:
>> ps ax
>> then look for: 467 root       0:00 tcpborphserver3 -f -l /dev/null -i
>> /etc/tcpborphserver3.i
>> In this case 467 is the pid.
>> Strace this process:
>> strace -p PID
>> Do not press enter yet, open another telnet session to the roach on port
>> 7147, then press enter.
>> now run ?progdev boffile in the telnet session and send me the strace
>> output.
>>
>> If the file really can't be read you will find a line like this:
>> open("/usr/bof/fubarorso", O_RDONLY)    = -1 ENOENT (No such file or
>> directory)
>> where fubarorso does not exist
>>
>> There might be other reasons, but we should be able to figure it out with
>> strace.
>>
>>
>>
>>
>>
>> On Fri, Sep 6, 2013 at 3:19 PM, Alec Rust <[email protected]> wrote:
>>
>>> Hi Joe, the bof file should be uploaded in to /usr/bof (note with
>>> tcpborphserver3 the boffiles are not executable any more, you have to use
>>> progdev).
>>>
>>> Telnet to the ROACH and run progdev:
>>> telnet roach_ip 7147
>>>   ?progdev boffile
>>>
>>> You should get !progdev ok
>>> ?listdev should list all your registers
>>>
>>> type ?help for a list of commands.
>>>
>>> You can transfer and program the boffile in one move using ?upload port,
>>> or just transfer using ?uploadbof port filename.
>>>
>>> #help upload
>>> upload\_and\_program\_a\_(possibly\_compressed)\_boffile\_(?upload\_[port\_[length\_[timeout]]])
>>> #help uploadbof
>>> upload\_a\_(possibly\_compressed)\_boffile\_(?uploadbof\_port\_filename\_[length\_[timeout]])
>>>
>>> Note the file then needs to be sent across using a separate TCP stream
>>> to the specified port (in linux use netcat, not sure what to use in windows
>>> but there is probably some netcat client available)
>>>
>>> Please let me know if you need more info.
>>>
>>> Regards
>>> Alec
>>>
>>>
>>> On Fri, Sep 6, 2013 at 1:25 AM, Kujawski, Joseph <[email protected]>wrote:
>>>
>>>> Laura,
>>>>
>>>> I am not using KATCP as the instruction set that I have access to does
>>>> not explain how to use it.  Do you have a link to some 'using KATCP'
>>>> instructions that I can follow?
>>>>
>>>> My question is simply this:
>>>>
>>>> Given that I have a .bof file on my PC, how do I program the FPGA on
>>>> the ROACH2 system to execute the program?
>>>>
>>>> I already know that:
>>>> 1) The ROACH2 boots into soloboot.
>>>> 2) I can physically connect to the ROACH2 (i.e. I have the proper
>>>> hardware in place).
>>>> 3) I can Telnet into the ROACH2 (i.e. the proper software is in place
>>>> on both my Windows machine and the ROACH2).
>>>> 4) I have a valid .bof file on the Windows machine.
>>>> 5) I can transfer this file to the ROACH2.
>>>>
>>>> -Joe Kujawski
>>>>
>>>>
>>>> On Thu, Sep 5, 2013 at 6:47 PM, Laura Vertatschitsch 
>>>> <[email protected]>wrote:
>>>>
>>>>> hey joe,
>>>>>
>>>>> how far did you get with KATCP and progdev?
>>>>>
>>>>> is your question perhaps about using the uploadbof command (mail list
>>>>> archive<http://www.mail-archive.com/[email protected]/msg03726.html>
>>>>> )?
>>>>>
>>>>> --Laura
>>>>>
>>>>>
>>>>> On Thu, Sep 5, 2013 at 11:21 PM, Kujawski, Joseph <[email protected]
>>>>> > wrote:
>>>>>
>>>>>> I have a ROACH2 that is in the following configuration:
>>>>>>
>>>>>> 1) My external computer is a Windows 7 (64 bit) computer.
>>>>>> 2) The Windows machine is connected to the ROACH2 via a direct cable
>>>>>> connection from the PC to the PPC ethernet port on the ROACH2.
>>>>>> 3) I am able to telnet into the ROACH2 and transfer files from the PC
>>>>>> to the ROACH2.
>>>>>>
>>>>>> According to the instructions on the CASPER website, I should be able
>>>>>> to program the system by executing the .bof file, however, this does not
>>>>>> work.  One individual associated with the project has indicated that the
>>>>>> ROACH2 cannot be executed and instead needs to be programmed via KATCP's
>>>>>> "progdev" command.  What do I do next?
>>>>>>
>>>>>> Thanks,
>>>>>>
>>>>>> -Joe Kujawski
>>>>>>
>>>>>> --
>>>>>> **************************************
>>>>>> * Joe Kujawski
>>>>>> * Siena College
>>>>>> * Dept. of Physics and Astronomy, RB 113
>>>>>> * 515 Loudon Road
>>>>>> * Loudonville, NY 12211-1462
>>>>>> *
>>>>>> * Email: [email protected]
>>>>>> * Phone: 518-782-6885
>>>>>> * Fax: 518-783-2986
>>>>>> **************************************
>>>>>>
>>>>>>
>>>>>
>>>>
>>>>
>>>> --
>>>> **************************************
>>>> * Joe Kujawski
>>>> * Siena College
>>>> * Dept. of Physics and Astronomy, RB 113
>>>> * 515 Loudon Road
>>>> * Loudonville, NY 12211-1462
>>>> *
>>>> * Email: [email protected]
>>>> * Phone: 518-782-6885
>>>> * Fax: 518-783-2986
>>>> **************************************
>>>>
>>>>
>>>
>>
>
>
> --
> **************************************
> * Joe Kujawski
> * Siena College
> * Dept. of Physics and Astronomy, RB 113
> * 515 Loudon Road
> * Loudonville, NY 12211-1462
> *
> * Email: [email protected]
> * Phone: 518-782-6885
> * Fax: 518-783-2986
> **************************************
>
>
>


-- 
**************************************
* Joe Kujawski
* Siena College
* Dept. of Physics and Astronomy, RB 113
* 515 Loudon Road
* Loudonville, NY 12211-1462
*
* Email: [email protected]
* Phone: 518-782-6885
* Fax: 518-783-2986
**************************************

Reply via email to