UPGRADE YOUR BROWSER

We have detected your current browser version is not the latest one. Xilinx.com uses the latest web technologies to bring you the best online experience possible. Please upgrade to a Xilinx.com supported browser:Chrome, Firefox, Internet Explorer 11, Safari. Thank you!

cancel
Showing results for 
Search instead for 
Did you mean: 
Visitor base13
Visitor
10,316 Views
Registered: ‎02-25-2015

dumping memory from SDK fails - jtag issues??

Jump to solution

Hi All,

 

I've been trying to pull data out of a memory on my Artix 7 DDR3.  Reads/Writes work fine with mrd/mwr and my hw seems to be writting memory correctly (ish), but getting that data out of memory has been unwieldy.   It seems like there are 3 methods, with the last method working very inconsistently (meaning I don't understand how to use it - I guess :)

 

1) mrd/mwr - good for small batches - "hangs" for the 100MB I'd like to extract (hang in quote because it's probably running but no data comes out until maybe it's done - whenever that is if ever).

 

2) jtag-axi and it's create_hw_axi instructions - similar sort of issues - dumping stuff to a screen is not feasiable and there doesn't seem to be an option for either of these two commands to dump to a file.

 

3) use SDK -> Xilinx Tools -> Dump/Restore Data File - which worked a few times for me, but obviously I'm missing something because I can't quite sort out how to "connect".   I've been bringing up vivado to connect via jtag to the blaze/etc, and to launch the SDK, but then trying to dump memory I end up with some flavor of this error (and no dump)

 

15:56:50 INFO : Dumping Memory Contents ...
15:56:51 ERROR : Failed to Restore Memory, Reason: Command [C 21 Memory get "JTAG-jsn-JTAG-SMT2-210251985863-13636093-0.MDM1.0" 107...] aborted

 

The other flavor is something to the effect of  

 

16:04:12 ERROR : Failed to Connect to HW Server, Reason: TCF error report:

Command: Memory getContext "JTAG-jsn-JTAG-SMT2-210251985863-13636093-0.MDM1"
Time: 2015-06-08 16:04:12.544
Error text: Invalid context
Error code: 16

 

Another possible hint is that the "select peer and context" dialog says my blaze is "Access denied" implying that I need to do something to allow SDK to acess it (see attached pic).

 

Any hints or dope slaps appreciated.

 

As an aside, is there a cmd line place to do this command?  in XMD console?

 

 

Screenshot-3.png
0 Kudos
1 Solution

Accepted Solutions
Visitor base13
Visitor
18,899 Views
Registered: ‎02-25-2015

Re: dumping memory from SDK fails - jtag issues??

Jump to solution

Ah, it ends up there was a zombie xmd client on my system (probably from a previous hang/kill)

 

steve@moby 498> ps aux | grep xmd
steve 2886 0.0 0.0 12528 1604 pts/14 S 15:55 0:00 /bin/bash /tools/Xilinx.2015.1/SDK/2015.1/bin/xmd -nx -json -ipcport 2350
steve 2909 0.0 0.0 12524 1576 pts/14 S 15:55 0:00 /bin/bash /tools/Xilinx.2015.1/SDK/2015.1/bin/loader -exec rdi_xmd -nx -json -ipcport 2350
steve 2919 0.4 0.1 356484 59772 pts/14 Sl 15:55 0:10 /tools/Xilinx.2015.1/SDK/2015.1/bin/unwrapped/lnx64.o/rdi_xmd -nx -json -ipcport 2350
steve 6216 0.0 0.0 11744 932 pts/6 S+ 16:31 0:00 grep --colour=auto xmd
steve@moby 498> killall xmd

 

So I killed them and now it seems like this is working:

 

xsdb% mrd -bin -file mem.bin 0x40000000 0x08000000
Debug context is claimed by another client
xsdb% Info: MicroBlaze #0 (target 3) Stopped at 0x0 (Suspended)  << after kill
xsdb% mrd -bin -file mem.bin 0x40000000 0x08000000

 

I'm detailing this because it hit me and I couldn't find a solution here (so maybe someone in the future will stumble upon this).

 

I'll mark this as solution accepted as soon as it finishes (a few more minutes).

 

3 Replies
Visitor base13
Visitor
10,315 Views
Registered: ‎02-25-2015

Re: dumping memory from SDK fails - jtag issues??

Jump to solution

Oh, yes, i'm using vivado 2015.1 and the matching SDK for it..

0 Kudos
Visitor base13
Visitor
10,308 Views
Registered: ‎02-25-2015

Re: dumping memory from SDK fails - jtag issues??

Jump to solution

At least mrd has a -file option in XSDB console: but fundumentally I this this is the issue (I guess this is becoming even more of a noob question).

 

xsdb% targets
1 xc7a200t
  2 Debug Module at USER2  
    3 MicroBlaze #0 (Access denied) (Running)
4 xc7a100t
  5 Debug Module at USER2
    6 MicroBlaze #0 (Suspended)

 

xsdb% mrd -bin -file mem.bin 0x40000000 0x08000000
Debug context is claimed by another client
xsdb%

 

I've closed vivado, I don't think there's another job/client running ??

 

0 Kudos
Visitor base13
Visitor
18,900 Views
Registered: ‎02-25-2015

Re: dumping memory from SDK fails - jtag issues??

Jump to solution

Ah, it ends up there was a zombie xmd client on my system (probably from a previous hang/kill)

 

steve@moby 498> ps aux | grep xmd
steve 2886 0.0 0.0 12528 1604 pts/14 S 15:55 0:00 /bin/bash /tools/Xilinx.2015.1/SDK/2015.1/bin/xmd -nx -json -ipcport 2350
steve 2909 0.0 0.0 12524 1576 pts/14 S 15:55 0:00 /bin/bash /tools/Xilinx.2015.1/SDK/2015.1/bin/loader -exec rdi_xmd -nx -json -ipcport 2350
steve 2919 0.4 0.1 356484 59772 pts/14 Sl 15:55 0:10 /tools/Xilinx.2015.1/SDK/2015.1/bin/unwrapped/lnx64.o/rdi_xmd -nx -json -ipcport 2350
steve 6216 0.0 0.0 11744 932 pts/6 S+ 16:31 0:00 grep --colour=auto xmd
steve@moby 498> killall xmd

 

So I killed them and now it seems like this is working:

 

xsdb% mrd -bin -file mem.bin 0x40000000 0x08000000
Debug context is claimed by another client
xsdb% Info: MicroBlaze #0 (target 3) Stopped at 0x0 (Suspended)  << after kill
xsdb% mrd -bin -file mem.bin 0x40000000 0x08000000

 

I'm detailing this because it hit me and I couldn't find a solution here (so maybe someone in the future will stumble upon this).

 

I'll mark this as solution accepted as soon as it finishes (a few more minutes).