Eclipse with SDK5.2 nan-29

RE: Eclipse with SDK5.2 nan-29

Jerod — Fri, 05 Jun 2015 01:28:15 GMT

For the record -- struggled with this one for far too long.

I was getting " No source available for "main() at 0x177b4" using SDK 7.2.1 and SD 7.1.0 with Eclipse and the generic examples.

Finally, found a good article here that suggested turning off optimizations and adding debugging: CFLAGS += -O0 -g3

But also needed to get rid of the followng line (flags): CFLAGS += -flto -fno-builtin

Didn't find "both" bits of info anywhere, so sharing here in hopes of helping ..

RE: Eclipse with SDK5.2 nan-29

Jamie Simonsen — Thu, 19 Jun 2014 03:16:08 GMT

Sorry resolved it myself didn't see the suggestion about setting the device name to nRF51822 in the Jlink settings which resolved it.

RE: Eclipse with SDK5.2 nan-29

Jamie Simonsen — Wed, 18 Jun 2014 22:46:47 GMT

I've followed everything in this thread but im still unable to debug, Im using the nRFGo Motherboard with the nRF51822 Development kit board and I'm able to build the blinky demo fine but when i go to debug i get a similar issue to what pascal posted above but i think the issue is with the J-Link as i get the following in the jlink log

SEGGER J-Link GDB Server V4.86a GUI Version

JLinkARM.dll V4.86a (DLL compiled Jun 11 2014 17:22:49)

-----GDB Server start settings-----
GDBInit file:                  none
GDB Server Listening port:     2331
SWO raw output listening port: 2332
Terminal I/O port:             2333
Accept remote connection:      localhost only
Generate logfile:              off
Verify download:               on
Init regs on start:            on
Silent mode:                   off
Single run mode:               on
Target connection timeout:     5 sec.
------J-Link related settings------
J-Link Host interface:         USB
J-Link script:                 none
J-Link settings file:          none
------Target related settings------
Target device:                 Cortex-M0
Target interface:              SWD
Target interface speed:        1000kHz
Target endian:                 little

Connecting to J-Link...
J-Link is connected.
Firmware: J-Link Lite-Cortex-M V8 compiled Aug 29 2012 15:24:23
Hardware: V8.00
S/N: 538000310
Feature(s): GDB
Checking target voltage...
Target voltage: 3.30 V
Listening on TCP/IP port 2331
Connecting to target...Connected to target
Waiting for GDB connection...Connected to 127.0.0.1
Reading all registers
Read 4 bytes @ address 0x00000000 (Data = 0xFFFFFFFF)
Target interface speed set to 1000 kHz
Resetting target
Halting target CPU...
...Target halted (PC = 0xFFFFFFFE)
R0 = FFFFFFFF, R1 = FFFFFFFF, R2 = FFFFFFFF, R3 = FFFFFFFF
R4 = FFFFFFFF, R5 = FFFFFFFF, R6 = FFFFFFFF, R7 = FFFFFFFF
R8 = FFFFFFFF, R9 = FFFFFFFF, R10= FFFFFFFF, R11= FFFFFFFF
R12= FFFFFFFF, R13= FFFFFFFC, MSP= FFFFFFFC, PSP= FFFFFFFC
R14(LR) = FFFFFFFF, R15(PC) = FFFFFFFE
XPSR C1000000, APSR C0000000, EPSR 01000000, IPSR 00000000
CFBP 00000000, CONTROL 00, FAULTMASK 00, BASEPRI 00, PRIMASK 00
Reading all registers
Select auto target interface speed (2000 kHz)
Flash breakpoints enabled
Semi-hosting enabled (VectorAddr = 0x08)
Semihosting I/O set to TELNET Client
SWO disabled succesfully.
ERROR: Could not detect CPU clock speed. Verification of test code downloaded into RAM failed.

Failed to enable SWO. Could not measure target CPU frequency.
Downloading 12 bytes @ address 0x00008000 - Verify failed
Downloading 2124 bytes @ address 0x0000800C - Verify failed
Downloading 12 bytes @ address 0x00008858 - Verify failed
Downloading 12 bytes @ address 0x00008864 - Verify failed
Downloading 8 bytes @ address 0x00008870 - Verify failed
Downloading 4 bytes @ address 0x00008878 - Verify failed
Downloading 8 bytes @ address 0x0001087C - Verify failed
Downloading 4 bytes @ address 0x00010884 - Verify failed
Downloading 4 bytes @ address 0x00010888 - Verify failed
Downloading 1336 bytes @ address 0x00010890 - Verify failed
Writing register (PC = 0x000080a5)
Read 4 bytes @ address 0x000080A4 (Data = 0xFFFFFFFF)
Resetting target
Halting target CPU...
...Target halted (PC = 0xFFFFFFFE)
R0 = FFFFFFFF, R1 = FFFFFFFF, R2 = FFFFFFFF, R3 = FFFFFFFF
R4 = FFFFFFFF, R5 = FFFFFFFF, R6 = FFFFFFFF, R7 = FFFFFFFF
R8 = FFFFFFFF, R9 = FFFFFFFF, R10= FFFFFFFF, R11= FFFFFFFF
R12= FFFFFFFF, R13= FFFFFFFC, MSP= FFFFFFFC, PSP= FFFFFFFC
R14(LR) = FFFFFFFF, R15(PC) = FFFFFFFE
XPSR C1000000, APSR C0000000, EPSR 01000000, IPSR 00000000
CFBP 00000000, CONTROL 00, FAULTMASK 00, BASEPRI 00, PRIMASK 00
Reading all registers
Read 4 bytes @ address 0xFFFFFFFE (Data = 0xFFFF0000)
Starting target CPU...
...Target halted (PC = 0xFFFFFFFE)
Reading all registers
Read 4 bytes @ address 0xFFFFFFFE (Data = 0xFFFFFFFF)
Read 4 bytes @ address 0xFFFFFFF4 (Data = 0x00000000)
Reading 64 bytes @ address 0xFFFFFFC0
Read 4 bytes @ address 0x00000000 (Data = 0xFFFFFFFF)

It looks like its having issue getting the CPU clock?

RE: Eclipse with SDK5.2 nan-29

pascal defesche — Wed, 16 Apr 2014 08:52:02 GMT

Or you could set the device name to: Cortex-M0

check this link, i've got debugging working following this setup: http://gnuarmeclipse.livius.net/blog/jlink-debugging/

Only i have to start the JLinkGDBServerCL.exe located at: C:\Program Files (x86)\SEGGER\JLinkARM_V484 before i start debugging.

RE: Eclipse with SDK5.2 nan-29

Nguyen Hoan Hoang — Wed, 16 Apr 2014 08:44:40 GMT

You set wrong device name. It should be just nRF51822.

RE: Eclipse with SDK5.2 nan-29

julian — Wed, 16 Apr 2014 03:57:31 GMT

Thanks for the suggestions: https://www.dropbox.com/s/mds4tkwgneb8kiq/Screenshot%202014-04-14%2009.30.08.png https://www.dropbox.com/s/f32ktwe19b9dt3y/Screenshot%202014-04-14%2009.35.52.png is eclipse settings https://www.dropbox.com/s/hbitsyecckw067z/1.jpg is keil settings

Read 4 bytes @ address 0x00000000 (Data = 0x20001C48) Target interface speed set to 30 kHz Resetting target Halting target CPU... ...Target halted (PC = 0x00012B98) R0 = FFFFFFFF, R1 = FFFFFFFF, R2 = FFFFFFFF, R3 = FFFFFFFF R4 = FFFFFFFF, R5 = FFFFFFFF, R6 = FFFFFFFF, R7 = FFFFFFFF R8 = FFFFFFFF, R9 = FFFFFFFF, R10= FFFFFFFF, R11= FFFFFFFF R12= FFFFFFFF, R13= 20001C48, MSP= 20001C48, PSP= FFFFFFFC R14(LR) = FFFFFFFF, R15(PC) = 00012B98 XPSR C1000000, APSR C0000000, EPSR 01000000, IPSR 00000000 CFBP 00000000, CONTROL 00, FAULTMASK 00, BASEPRI 00, PRIMASK 00 Reading all registers Target interface speed set to 2000 kHz Flash breakpoints enabled Read 4 bytes @ address 0x00012B98 (Data = 0x681B4B15) Downloading 16192 bytes @ address 0x00014000 - Verify failed Downloading 15968 bytes @ address 0x00017F40 - Verify failed Downloading 16160 bytes @ address 0x0001BDA0 - Verify failed Downloading 16096 bytes @ address 0x0001FCC0 - Verify failed Downloading 16080 bytes @ address 0x00023BA0 - Verify failed Downloading 16176 bytes @ address 0x00027A70 - Verify failed Downloading 12208 bytes @ address 0x0002B9A0 - Verify failed Downloading 68 bytes @ address 0x0002E950 - Verify failed Downloading 8 bytes @ address 0x0002E994 - Verify failed Downloading 2396 bytes @ address 0x0002E99C - Verify failed Writing register (PC = 0x0002e951) Read 4 bytes @ address 0x0002E950 (Data = 0xFFFFFFFF) Read 4 bytes @ address 0xFFFFFFFF (Data = 0x001C4800) Read 4 bytes @ address 0xFFFFFFFF (Data = 0x001C4800) Resetting target Halting target CPU... ...Target halted (PC = 0x00012B98) R0 = FFFFFFFF, R1 = FFFFFFFF, R2 = FFFFFFFF, R3 = FFFFFFFF R4 = FFFFFFFF, R5 = FFFFFFFF, R6 = FFFFFFFF, R7 = FFFFFFFF R8 = FFFFFFFF, R9 = FFFFFFFF, R10= FFFFFFFF, R11= FFFFFFFF R12= FFFFFFFF, R13= 20001C48, MSP= 20001C48, PSP= FFFFFFFC R14(LR) = FFFFFFFF, R15(PC) = 00012B98 XPSR C1000000, APSR C0000000, EPSR 01000000, IPSR 00000000 CFBP 00000000, CONTROL 00, FAULTMASK 00, BASEPRI 00, PRIMASK 00 Reading all registers Read 4 bytes @ address 0x00012B98 (Data = 0x681B4B15) Starting target CPU... Debugger requested to halt target... ...Target halted (PC = 0x0000FE3E) Reading all registers Read 4 bytes @ address 0x0000FE3E (Data = 0xBF20BF40) Read 4 bytes @ address 0x0000016A (Data = 0x9801B401) Read 4 bytes @ address 0x0000016A (Data = 0x9801B401) GDB closed TCP/IP connection

RE: Eclipse with SDK5.2 nan-29

Nguyen Hoan Hoang — Mon, 14 Apr 2014 10:38:03 GMT

try disable SWO and Semihosting settings. Try different jtag/swd speed. It's on the startup tab of the debug configuration.

RE: Eclipse with SDK5.2 nan-29

julian — Mon, 14 Apr 2014 04:51:50 GMT

I use Nordic dev kit, and able to flash in keil without problem

RE: Eclipse with SDK5.2 nan-29

Nguyen Hoan Hoang — Sat, 12 Apr 2014 17:56:37 GMT

I see a bunch of Verify failed. Are you using the Nordic Kit or your own board ? I seems unable to flash your app properly.

RE: Eclipse with SDK5.2 nan-29

julian — Sat, 12 Apr 2014 00:24:37 GMT

my thread goes too long, I can't find your reply in the thread, so I post it on your thread Thanks for help out first!

my current issue is able to start debugger, but can't stop at breakpoint, below is log: www.dropbox.com/.../Screenshot 2014-04-11 17.19.53.png

Connecting to J-Link... J-Link is connected. Firmware: J-Link OB-SAM3U128 V1 compiled Dec 11 2013 20:20:11 Hardware: V1.00 S/N: 480205216 Checking target voltage... Target voltage: 3.30 V Listening on TCP/IP port 2331 Connecting to target...Connected to target Waiting for GDB connection...Connected to 127.0.0.1 Reading all registers Read 4 bytes @ address 0x00000000 (Data = 0x20001C48) Target interface speed set to 30 kHz Resetting target Halting target CPU... ...Target halted (PC = 0x00012B98) R0 = FFFFFFFF, R1 = FFFFFFFF, R2 = FFFFFFFF, R3 = FFFFFFFF R4 = FFFFFFFF, R5 = FFFFFFFF, R6 = FFFFFFFF, R7 = FFFFFFFF R8 = FFFFFFFF, R9 = FFFFFFFF, R10= FFFFFFFF, R11= FFFFFFFF R12= FFFFFFFF, R13= 20001C48, MSP= 20001C48, PSP= FFFFFFFC R14(LR) = FFFFFFFF, R15(PC) = 00012B98 XPSR C1000000, APSR C0000000, EPSR 01000000, IPSR 00000000 CFBP 00000000, CONTROL 00, FAULTMASK 00, BASEPRI 00, PRIMASK 00 Reading all registers Select auto target interface speed (2000 kHz) Flash breakpoints enabled Semi-hosting enabled (VectorAddr = 0x08) Semihosting I/O set to TELNET Client Failed to disable SWO. ERROR: Could not detect CPU clock speed. Verification of test code downloaded into RAM failed.

Failed to enable SWO. Could not measure target CPU frequency. Read 4 bytes @ address 0x00012B98 (Data = 0x681B4B15) Downloading 16192 bytes @ address 0x00014000 - Verify failed Downloading 15968 bytes @ address 0x00017F40 - Verify failed Downloading 16160 bytes @ address 0x0001BDA0 - Verify failed Downloading 16096 bytes @ address 0x0001FCC0 - Verify failed Downloading 16080 bytes @ address 0x00023BA0 - Verify failed Downloading 16176 bytes @ address 0x00027A70 - Verify failed Downloading 12208 bytes @ address 0x0002B9A0 - Verify failed Downloading 68 bytes @ address 0x0002E950 - Verify failed Downloading 8 bytes @ address 0x0002E994 - Verify failed Downloading 2396 bytes @ address 0x0002E99C - Verify failed Writing register (PC = 0x0002e951) Read 4 bytes @ address 0x0002E950 (Data = 0xFFFFFFFF) Read 4 bytes @ address 0xFFFFFFFF (Data = 0x001C4800) Read 4 bytes @ address 0xFFFFFFFF (Data = 0x001C4800) Resetting target Halting target CPU... ...Target halted (PC = 0x00012B98) R0 = FFFFFFFF, R1 = FFFFFFFF, R2 = FFFFFFFF, R3 = FFFFFFFF R4 = FFFFFFFF, R5 = FFFFFFFF, R6 = FFFFFFFF, R7 = FFFFFFFF R8 = FFFFFFFF, R9 = FFFFFFFF, R10= FFFFFFFF, R11= FFFFFFFF R12= FFFFFFFF, R13= 20001C48, MSP= 20001C48, PSP= FFFFFFFC R14(LR) = FFFFFFFF, R15(PC) = 00012B98 XPSR C1000000, APSR C0000000, EPSR 01000000, IPSR 00000000 CFBP 00000000, CONTROL 00, FAULTMASK 00, BASEPRI 00, PRIMASK 00 Reading all registers Read 4 bytes @ address 0x00012B98 (Data = 0x681B4B15) Starting target CPU...

if I suspend it www.dropbox.com/.../Screenshot 2014-04-11 17.21.12.png

do you have any hint on this?

RE: Eclipse with SDK5.2 nan-29

Nguyen Hoan Hoang — Sat, 05 Apr 2014 17:53:09 GMT

You're welcome. What I meant was in the gcc_startup_nrf51.S, instead of the call to _start, replace it by a call to _main. If you are writing C++ code. The runtime lib need to be initialized call cling _libc_init_array. See in my code. I put a conditional compilation for it. C++ code will generate a bigger file due to it runtime library.

RE: Eclipse with SDK5.2 nan-29

leo — Sat, 05 Apr 2014 17:46:36 GMT

You are my hero! A huge thanks! I can debug it now (with your files). But what do you mean by "call _main directly" and if i use the c++ runtime, i can use the default startup code? Thanks again!! :)

RE: Eclipse with SDK5.2 nan-29

Nguyen Hoan Hoang — Sat, 05 Apr 2014 17:29:51 GMT

Ah yes, forgot about that bug in the gcc_startup. You can call _main directly. the libc_init_arry is needed only require for C++ runtime. You can also use my startup code with my LD script instead. Files attached here

nrf51_startup_ld.zip

RE: Eclipse with SDK5.2 nan-29

leo — Sat, 05 Apr 2014 17:18:10 GMT

Hi Nguyen, you were right. Its possible to debug. I can debug the "gcc_startup_nrf5.S" file, and step through "SystemInit()" in "system_nrf51.c". But in the function "_start" there is a call to __libc_init_array, and here my application crashes (i guess.. here is picture: http://postimg.org/image/w7hwk0a71/). Have no clue why.. Another question i have is, which gcc compiler flags should i use .. are there some important? (for the linker i use: " -mabi=aapcs --specs=nano.specs -lc -lnosys -Wl,--gc-sections" and for the target processor i use: "-ffunction-sections -fdata-sections". Thanks a lot in advance!

RE: Eclipse with SDK5.2 nan-29

Nguyen Hoan Hoang — Sat, 05 Apr 2014 14:48:29 GMT

There is probably an error somewhere in your code. The sdk generate an exception when there is an error. Find APP_ERROR_HANDLER in the code and put a break point on it. Then you can back trace to where the error was generated.

RE: Eclipse with SDK5.2 nan-29

leo — Sat, 05 Apr 2014 10:23:36 GMT

So you say i have to flash it with nFRGo (and not with JLinkExe) one time and after that, i can debug it with eclipse without any problems. Thanks for the hint, i'll test it as soon i'm on a win pc. But, do you have a clue why i stuck in the HardFaultHandler-loop?

RE: Eclipse with SDK5.2 nan-29

Nguyen Hoan Hoang — Fri, 04 Apr 2014 18:26:05 GMT

Hi Leo,

 Here is what I know works for me.  Development on OSX with Elcipse Kepler.  Without DFU loaded, debug works directly with s110 pre-flashed.  If DFU loaded.  i need first upload my app using dfu.  After that I can use Eclipse to debug.  It seem that DFU sets some flag to make s110 jumps to the App.  Flashing with Eclipse does not set the flag so s110 always jumps to DFU.  That is what I suspected, need to be confirmed.  When DFU is not loaded, S110 don't see it so it jumps to the App directly.  So if you work with ble, then you need to Flash the s110 first into the chip using nFRGo.

RE: Eclipse with SDK5.2 nan-29

leo — Fri, 04 Apr 2014 15:59:52 GMT

hoi nguyen, thanks for your reply. No, i work with the ble_hrs example. It compiles in eclipse, but when i start the debugger i'm in an endless loop (the application is running, no breakpoint, but when i hit the pause button i see this: http://postimg.org/image/w5t48o99v/) in HardFaultHandler.

RE: Eclipse with SDK5.2 nan-29

Nguyen Hoan Hoang — Fri, 04 Apr 2014 15:52:36 GMT

Leo, are you working with blinky_exemple too? That exemple is not ble. You don't need s110 and dfu. Use s110 and dfu only when you do ble application.

RE: Eclipse with SDK5.2 nan-29

leo — Fri, 04 Apr 2014 15:32:22 GMT

hi nguyen,

i have the same problem as pascal. So, i need additionally to the softdevice (flashed via Makefile from http://hg.cmason.com/nrf) a dfu bootloader?

RE: Eclipse with SDK5.2 nan-29

Nguyen Hoan Hoang — Fri, 04 Apr 2014 15:02:18 GMT

The blinky_example does not use the s110. I should use the blank linker script for this one. Get a feel how the environment works first.

When to start working with s110. You'll need need to flash in the soft device first then DFU using nRFGo. Then use DFU to upload your app. After that you can debug with Eclipse. That's the work around s110 issue for debugging. I have posted full Eclipse project for DFU, see previous comment. Use the serial version to upload your app via segger cdc.

RE: Eclipse with SDK5.2 nan-29

pascal defesche — Fri, 04 Apr 2014 14:46:42 GMT

Ok in my case its with capital S, and i think its correct i can see the files in the debug/startup folder. I've attached my new updated blinky_example.rar above, could you check for me if you can run this example? Thanks a lot for your help so far Nguyen.

RE: Eclipse with SDK5.2 nan-29

Nguyen Hoan Hoang — Fri, 04 Apr 2014 14:35:59 GMT

I think it has to be capital S.

RE: Eclipse with SDK5.2 nan-29

pascal defesche — Fri, 04 Apr 2014 14:33:37 GMT

Yes i did add the system_nrf51.c & gcc_startup_nrf51.s (this needs to be captital S or lower case?) see my attachment: gcc_debug.png

RE: Eclipse with SDK5.2 nan-29

leo — Fri, 04 Apr 2014 14:29:32 GMT

i sometimes get the same error. i thought i fixed it, with erasing the chip before starting debugging. Unfortunately it doesn't work with the ble_hrs app.. so i would be happy, if you're getting an answer to this problem :)