You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@mynewt.apache.org by Pierre Kircher <pk...@me.com> on 2017/05/30 15:32:20 UTC
segger jlink
nrf52 hal_os_tick prevents starting of a debugging session
running command$ newt debug bleprph -n
[/Users/pkircher/dev/puck/repos/apache-mynewt-core/hw/bsp/nrf52dk/nrf52dk_debug.sh /Users/pkircher/dev/puck/repos/apache-mynewt-core/hw/bsp/nrf52dk /Users/pkircher/dev/puck/bin/targets/bleprph/app/apps/bleprph/bleprph]
/Users/pkircher/dev/puck/repos/apache-mynewt-core/hw/scripts/common.sh: line 38: [: =: unary operator expected
SEGGER J-Link GDB Server V6.14c Command Line Version
JLinkARM.dll V6.14c (DLL compiled Mar 31 2017 17:42:10)
-----GDB Server start settings-----
GDBInit file: none
GDB Server Listening port: 3333
SWO raw output listening port: 2332
Terminal I/O port: 2333
Accept remote connection: yes
Generate logfile: off
Verify download: off
Init regs on start: off
Silent mode: off
Single run mode: on
Target connection timeout: 0 ms
------J-Link related settings------
J-Link Host interface: USB
J-Link script: none
J-Link settings file: none
------Target related settings------
Target device: nRF52
Target interface: SWD
Target interface speed: 4000kHz
Target endian: little
Connecting to J-Link...
J-Link is connected.
Firmware: J-Link ARM V8 compiled Nov 28 2014 13:44:46
Hardware: V8.00
S/N: 261282027
Feature(s): RDI,FlashDL,FlashBP,JFlash,GDB
Checking target voltage...
Target voltage: 2.90 V
Listening on TCP/IP port 3333
Connecting to target...Connected to target
Waiting for GDB connection...Connected to 127.0.0.1
Reading all registers
Read 4 bytes @ address 0x0001C6FA (Data = 0xE8BDB11C)
Reading 64 bytes @ address 0x20002F40
Read 4 bytes @ address 0x000089DE (Data = 0xF0014638)
Read 2 bytes @ address 0x00008410 (Data = 0x2100)
Read 2 bytes @ address 0x00008482 (Data = 0x9003)
Reading 64 bytes @ address 0x20002F80
Read 4 bytes @ address 0x00000000 (Data = 0x20010000)
Read 4 bytes @ address 0x00000000 (Data = 0x20010000)
Setting breakpoint @ address 0x00008482, Size = 2, BPHandle = 0x0001
Starting target CPU…
the breakpoint points on main.c 129 > apps/bleprph/src
Reading symbols from /Users/pkircher/dev/puck/bin/targets/bleprph/app/apps/bleprph/bleprph.elf...
done.
os_tick_idle (ticks=595) at repos/apache-mynewt-core/hw/mcu/nordic/nrf52xxx/src/hal_os_tick.c:200
200 if (ticks > 0) {
Not implemented stop reason (assuming exception): undefined
never hits that breakpoint - is that a known issue ?
and yes im aware im asking a lot questions .. but im more then willing to contribute back once i got the hang of it
thanks once again
pierre
Re: segger jlink
Posted by Justin Mclean <ju...@classsoftware.com>.
Hi,
> Well, I’m on mynewt.slack.com - if you care to be a guinea pig! :-)
>
> Invite link here: https://join.slack.com/mynewt/shared_invite/MTkwMTg1ODM1NTg5LTE0OTYxNzQ4NzQtZTU1YmNhYjhkMg
You might want to consider adding a room to https://the-asf.slack.com/ ?
Thanks,
Justin
Re: segger jlink
Posted by Jacob Rosenthal <ja...@gmail.com>.
I dislike slack, but Id happily use whatever it took to find the decision
making conversations.
On Tue, May 30, 2017 at 1:24 PM, Sterling Hughes <
sterling.hughes.public@gmail.com> wrote:
> Well, I’m on mynewt.slack.com - if you care to be a guinea pig! :-)
>
> Invite link here: https://join.slack.com/mynewt/
> shared_invite/MTkwMTg1ODM1NTg5LTE0OTYxNzQ4NzQtZTU1YmNhYjhkMg
>
> Sterling
>
>
> On 30 May 2017, at 13:22, Pierre Kircher wrote:
>
> i dont mind using slack ..
>>
>> a quick chat is just way faster to communicate and has a better flow then
>> a mailing list
>>
>> thats why i asked in the first place
>>
>> pierre
>>
>> On 30 May 2017, at 21:19, Sterling Hughes <sterling.hughes.public@gmail.
>>> com> wrote:
>>>
>>> Hi Jacob,
>>>
>>> I agree. Some of the Mynewt core devs have an IRC culture, but most use
>>> less open, more ..friendly.. communications tools like Slack.
>>>
>>> I’m looking up to see if there is a Slack->IRC gateway, which might make
>>> it easier. Alternatively, what do folks think of just pointing people to
>>> Slack —> if it’s better attended by developers, is it too much of a hurdle
>>> even these days?
>>>
>>> Sterling
>>>
>>> On 30 May 2017, at 12:27, Jacob Rosenthal wrote:
>>>
>>> The mailing list has been gracious to help me fix a bunch of stuff
>>>> upstream
>>>> and taken my changes, and Ive really enjoyed the tooling on mynewt.
>>>>
>>>> However, no offense to the mynewt core, but I would push back on the
>>>> idea
>>>> you have a functioning IRC.
>>>>
>>>> Ive been idling there for well over 6 months and have tried asking
>>>> questions there. Any questions there go unanswered, or are told to be
>>>> taken
>>>> here.
>>>>
>>>> Further in all that time the room has been absolutely silent because the
>>>> core team has literally no public discussion there. It would be nice to
>>>> hear that kind of core discussion that really doesnt happen here
>>>> either...
>>>>
>>>>
>>>>
>>>>
>>>> On Tue, May 30, 2017 at 12:07 PM, aditi hilbert <ad...@runtime.io>
>>>> wrote:
>>>>
>>>> at exchange - and we would love it if you could contribute back to the
>>>>> documentation (https://github.com/apache/incubator-mynewt-site,
>>>>> develop
>>>>> branch) :)
>>>>>
>>>>>
Re: segger jlink
Posted by Sterling Hughes <st...@gmail.com>.
Well, I’m on mynewt.slack.com - if you care to be a guinea pig! :-)
Invite link here:
https://join.slack.com/mynewt/shared_invite/MTkwMTg1ODM1NTg5LTE0OTYxNzQ4NzQtZTU1YmNhYjhkMg
Sterling
On 30 May 2017, at 13:22, Pierre Kircher wrote:
> i dont mind using slack ..
>
> a quick chat is just way faster to communicate and has a better flow
> then a mailing list
>
> thats why i asked in the first place
>
> pierre
>
>> On 30 May 2017, at 21:19, Sterling Hughes
>> <st...@gmail.com> wrote:
>>
>> Hi Jacob,
>>
>> I agree. Some of the Mynewt core devs have an IRC culture, but most
>> use less open, more ..friendly.. communications tools like Slack.
>>
>> I’m looking up to see if there is a Slack->IRC gateway, which might
>> make it easier. Alternatively, what do folks think of just pointing
>> people to Slack —> if it’s better attended by developers, is it
>> too much of a hurdle even these days?
>>
>> Sterling
>>
>> On 30 May 2017, at 12:27, Jacob Rosenthal wrote:
>>
>>> The mailing list has been gracious to help me fix a bunch of stuff
>>> upstream
>>> and taken my changes, and Ive really enjoyed the tooling on mynewt.
>>>
>>> However, no offense to the mynewt core, but I would push back on the
>>> idea
>>> you have a functioning IRC.
>>>
>>> Ive been idling there for well over 6 months and have tried asking
>>> questions there. Any questions there go unanswered, or are told to
>>> be taken
>>> here.
>>>
>>> Further in all that time the room has been absolutely silent because
>>> the
>>> core team has literally no public discussion there. It would be nice
>>> to
>>> hear that kind of core discussion that really doesnt happen here
>>> either...
>>>
>>>
>>>
>>>
>>> On Tue, May 30, 2017 at 12:07 PM, aditi hilbert <ad...@runtime.io>
>>> wrote:
>>>
>>>> at exchange - and we would love it if you could contribute back to
>>>> the
>>>> documentation (https://github.com/apache/incubator-mynewt-site,
>>>> develop
>>>> branch) :)
>>>>
Re: segger jlink
Posted by Pierre Kircher <pk...@me.com>.
i dont mind using slack ..
a quick chat is just way faster to communicate and has a better flow then a mailing list
thats why i asked in the first place
pierre
> On 30 May 2017, at 21:19, Sterling Hughes <st...@gmail.com> wrote:
>
> Hi Jacob,
>
> I agree. Some of the Mynewt core devs have an IRC culture, but most use less open, more ..friendly.. communications tools like Slack.
>
> I’m looking up to see if there is a Slack->IRC gateway, which might make it easier. Alternatively, what do folks think of just pointing people to Slack —> if it’s better attended by developers, is it too much of a hurdle even these days?
>
> Sterling
>
> On 30 May 2017, at 12:27, Jacob Rosenthal wrote:
>
>> The mailing list has been gracious to help me fix a bunch of stuff upstream
>> and taken my changes, and Ive really enjoyed the tooling on mynewt.
>>
>> However, no offense to the mynewt core, but I would push back on the idea
>> you have a functioning IRC.
>>
>> Ive been idling there for well over 6 months and have tried asking
>> questions there. Any questions there go unanswered, or are told to be taken
>> here.
>>
>> Further in all that time the room has been absolutely silent because the
>> core team has literally no public discussion there. It would be nice to
>> hear that kind of core discussion that really doesnt happen here either...
>>
>>
>>
>>
>> On Tue, May 30, 2017 at 12:07 PM, aditi hilbert <ad...@runtime.io> wrote:
>>
>>> at exchange - and we would love it if you could contribute back to the
>>> documentation (https://github.com/apache/incubator-mynewt-site, develop
>>> branch) :)
>>>
Re: segger jlink
Posted by Szymon Janc <sz...@codecoup.pl>.
Hi Sterling,
On 30 May 2017 at 22:19, Sterling Hughes
<st...@gmail.com> wrote:
>
> Hi Jacob,
>
> I agree. Some of the Mynewt core devs have an IRC culture, but most use less open, more ..friendly.. communications tools like Slack.
>
> I’m looking up to see if there is a Slack->IRC gateway, which might make it easier. Alternatively, what do folks think of just pointing people to Slack —> if it’s better attended by developers, is it too much of a hurdle even these days?
I can live with Slack as long as IRC gateway is enabled :)
--
pozdrawiam
Szymon K. Janc
Re: segger jlink
Posted by Sterling Hughes <st...@gmail.com>.
Hi Jacob,
I agree. Some of the Mynewt core devs have an IRC culture, but most use
less open, more ..friendly.. communications tools like Slack.
I’m looking up to see if there is a Slack->IRC gateway, which might
make it easier. Alternatively, what do folks think of just pointing
people to Slack —> if it’s better attended by developers, is it too
much of a hurdle even these days?
Sterling
On 30 May 2017, at 12:27, Jacob Rosenthal wrote:
> The mailing list has been gracious to help me fix a bunch of stuff
> upstream
> and taken my changes, and Ive really enjoyed the tooling on mynewt.
>
> However, no offense to the mynewt core, but I would push back on the
> idea
> you have a functioning IRC.
>
> Ive been idling there for well over 6 months and have tried asking
> questions there. Any questions there go unanswered, or are told to be
> taken
> here.
>
> Further in all that time the room has been absolutely silent because
> the
> core team has literally no public discussion there. It would be nice
> to
> hear that kind of core discussion that really doesnt happen here
> either...
>
>
>
>
> On Tue, May 30, 2017 at 12:07 PM, aditi hilbert <ad...@runtime.io>
> wrote:
>
>> at exchange - and we would love it if you could contribute back to
>> the
>> documentation (https://github.com/apache/incubator-mynewt-site,
>> develop
>> branch) :)
>>
Re: segger jlink
Posted by Jacob Rosenthal <ja...@gmail.com>.
The mailing list has been gracious to help me fix a bunch of stuff upstream
and taken my changes, and Ive really enjoyed the tooling on mynewt.
However, no offense to the mynewt core, but I would push back on the idea
you have a functioning IRC.
Ive been idling there for well over 6 months and have tried asking
questions there. Any questions there go unanswered, or are told to be taken
here.
Further in all that time the room has been absolutely silent because the
core team has literally no public discussion there. It would be nice to
hear that kind of core discussion that really doesnt happen here either...
On Tue, May 30, 2017 at 12:07 PM, aditi hilbert <ad...@runtime.io> wrote:
> at exchange - and we would love it if you could contribute back to the
> documentation (https://github.com/apache/incubator-mynewt-site, develop
> branch) :)
>
Re: segger jlink
Posted by aditi hilbert <ad...@runtime.io>.
Pierre,
All this is great exchange - and we would love it if you could contribute back to the documentation (https://github.com/apache/incubator-mynewt-site, develop branch) :)
We do have a chat channel on irc.freenode.net <http://irc.freenode.net/>, channel name #mynewt.
thanks,
aditi
> On May 31, 2017, at 1:54 AM, Pierre Kircher <pk...@me.com> wrote:
>
> might worth mentioning that just happens in gdb ..
> if i use ozone (segger tool) it debugs straight away
>
> i check those recommendations as soon as im back on the desk
>
> thanks again
>
> love the project but could use some more documentation or a chatroom irc/gitter
>
> pierre
>
>> On 30 May 2017, at 18:37, Christopher Collins <ch...@runtime.io> wrote:
>>
>> Hi Pierre,
>>
>> First, don't worry about asking a lot of questions. There are probably
>> others with the same questions who will benefit from your asking them!
>>
>> On Tue, May 30, 2017 at 04:32:20PM +0100, Pierre Kircher wrote:
>> [...]
>>> the breakpoint points on main.c 129 > apps/bleprph/src
>>>
>>> Reading symbols from /Users/pkircher/dev/puck/bin/targets/bleprph/app/apps/bleprph/bleprph.elf...
>>> done.
>>> os_tick_idle (ticks=595) at repos/apache-mynewt-core/hw/mcu/nordic/nrf52xxx/src/hal_os_tick.c:200
>>> 200 if (ticks > 0) {
>>> Not implemented stop reason (assuming exception): undefined
>>>
>>> never hits that breakpoint - is that a known issue ?
>>
>> That is not a known issue. My guess is that the API call immediately
>> before that point (ble_gap_adv_set_fields) is failing for some reason.
>> One common reason is that the app is trying to put too much data into
>> the advertisement. For example, this happens if you are using a very
>> long name. I would backtrack a bit; set a breakpoint on
>> bleprph_advertise() and see if your app gets there.
>>
>> FYI, here is the expected sequence of events:
>> 1. main()
>> 2. bleprph_on_sync() (communication established among host and
>> controller).
>> 3. bleprph_advertise (called from bleprph_on_sync()).
>>
>> Chris
>
Re: segger jlink
Posted by Pierre Kircher <pk...@me.com>.
might worth mentioning that just happens in gdb ..
if i use ozone (segger tool) it debugs straight away
i check those recommendations as soon as im back on the desk
thanks again
love the project but could use some more documentation or a chatroom irc/gitter
pierre
> On 30 May 2017, at 18:37, Christopher Collins <ch...@runtime.io> wrote:
>
> Hi Pierre,
>
> First, don't worry about asking a lot of questions. There are probably
> others with the same questions who will benefit from your asking them!
>
> On Tue, May 30, 2017 at 04:32:20PM +0100, Pierre Kircher wrote:
> [...]
>> the breakpoint points on main.c 129 > apps/bleprph/src
>>
>> Reading symbols from /Users/pkircher/dev/puck/bin/targets/bleprph/app/apps/bleprph/bleprph.elf...
>> done.
>> os_tick_idle (ticks=595) at repos/apache-mynewt-core/hw/mcu/nordic/nrf52xxx/src/hal_os_tick.c:200
>> 200 if (ticks > 0) {
>> Not implemented stop reason (assuming exception): undefined
>>
>> never hits that breakpoint - is that a known issue ?
>
> That is not a known issue. My guess is that the API call immediately
> before that point (ble_gap_adv_set_fields) is failing for some reason.
> One common reason is that the app is trying to put too much data into
> the advertisement. For example, this happens if you are using a very
> long name. I would backtrack a bit; set a breakpoint on
> bleprph_advertise() and see if your app gets there.
>
> FYI, here is the expected sequence of events:
> 1. main()
> 2. bleprph_on_sync() (communication established among host and
> controller).
> 3. bleprph_advertise (called from bleprph_on_sync()).
>
> Chris
Re: segger jlink
Posted by Christopher Collins <ch...@runtime.io>.
Hi Pierre,
First, don't worry about asking a lot of questions. There are probably
others with the same questions who will benefit from your asking them!
On Tue, May 30, 2017 at 04:32:20PM +0100, Pierre Kircher wrote:
[...]
> the breakpoint points on main.c 129 > apps/bleprph/src
>
> Reading symbols from /Users/pkircher/dev/puck/bin/targets/bleprph/app/apps/bleprph/bleprph.elf...
> done.
> os_tick_idle (ticks=595) at repos/apache-mynewt-core/hw/mcu/nordic/nrf52xxx/src/hal_os_tick.c:200
> 200 if (ticks > 0) {
> Not implemented stop reason (assuming exception): undefined
>
> never hits that breakpoint - is that a known issue ?
That is not a known issue. My guess is that the API call immediately
before that point (ble_gap_adv_set_fields) is failing for some reason.
One common reason is that the app is trying to put too much data into
the advertisement. For example, this happens if you are using a very
long name. I would backtrack a bit; set a breakpoint on
bleprph_advertise() and see if your app gets there.
FYI, here is the expected sequence of events:
1. main()
2. bleprph_on_sync() (communication established among host and
controller).
3. bleprph_advertise (called from bleprph_on_sync()).
Chris