Atmel Studio 6.2 beta released

Go To Last Post
76 posts / 0 new

Pages

Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Atmel Studio 6.2 beta is now available!

Downloads and release notes can be found here:
http://www.atmel.com/tools/ATMEL...

Brief presentation of new trace functionality:
http://www.youtube.com/watch?v=m...
http://www.youtube.com/watch?v=G...

As always, enjoy! :D

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Still ~800MB download, does this mean that I will ned to download and install garbage I don't want as I will NEVER use it?

John Samperi

Ampertronics Pty. Ltd.

www.ampertronics.com.au

* Electronic Design * Custom Products * Contract Assembly

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Installed 6.2 with no issue. Opened extension manager to add extensions i have been using in 6.1 and there is only one page of available extensions. How do i fix this issue ?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Extensions need to be explicitly set as compatible in their manifest to show up, to avoid compatibility problems if something has changed internally. You can edit the manifest yourself if you grab the VSIX files from gallery.atmel.com or use my Python script:

https://github.com/abcminiuser/a...

Since I wasn't given any prerelease builds I haven't updated the manifests on my extensons yet -- I'll make a note to download and test.

One of the improvements is better debugging, which I'm VERY excited about; it uses GDB under the hood for better run control.

- Dean :twisted:

Make Atmel Studio better with my free extensions. Open source and feedback welcome!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

rev wrote:
Installed 6.2 with no issue. Opened extension manager to add extensions i have been using in 6.1 and there is only one page of available extensions. How do i fix this issue ?

Many extensions are yet to be updated to latest 6.2.

Same info is forwarded to all the authors of extension by atmel a month ago.

Might be they didn't noticed.

Almost all 6.1 extensions will work fine on 6.2 with following modification in extension manifest

      AtmelStudio

to

AtmelStudio

After that create short cut to that extension and paste it under 6.2 installed location

For example:
"C:\Program Files (x86)\Atmel\Atmel Studio 6.2\extensions\Extension Name\"

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Try the data breakpoint feature, which is missed for long time in atmel studio.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The AS 6.2 is supposed to add not only data breakpoints (AVR8, AVR32 and Cortex?) that but also trace and OS-awareness (all three architectures?). I do not know what "live watch" is, I suppose they called this way a data breakpoint from DWT of Cortex emitting events via ITM. Similar to "viewing MCU event counters".

No RSTDISBL, no fun!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Live watch - similar to watch window. Difference is watch window will show the latest value when debug mode is in break state. Live watch will show the value even in run mode. But you can add an item(variable) to live watch window only in break mode.
Live watch internally uses the data breakpoint registers present in device.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Live feedback as I install and test (will update):

- Install smooth, new USB installer package which took a million years to install each driver (despite being on a high-end laptop with SSD). One brazillion unsigned driver confirmation requests, will be a problem with Win8 users.

- Dean :twisted:

Make Atmel Studio better with my free extensions. Open source and feedback welcome!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

abcminiuser wrote:
One brazillion unsigned driver confirmation requests, will be a problem with Win8 users.
Atmel!!! Get your act together.

I'll try to check it out over the weekend. As soon as I get my main dev machine back. Building FreeBSD from source is so much fun...

"I may make you feel but I can't make you think" - Jethro Tull - Thick As A Brick

"void transmigratus(void) {transmigratus();} // recursio infinitus" - larryvc

"It's much more practical to rely on the processing powers of the real debugger, i.e. the one between the keyboard and chair." - JW wek3

"When you arise in the morning think of what a privilege it is to be alive: to breathe, to think, to enjoy, to love." -  Marcus Aurelius

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Running win 8.1 and usb drivers took ages as you said dean. just did some other stuff while it took its time

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

- New Example Project wizard: shows much faster, until you try to expand out the treeview to show the examples, then it freezes for a few seconds.

- Dean :twisted:

Make Atmel Studio better with my free extensions. Open source and feedback welcome!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hey dean, so are your free extensions compatible with AS6.2

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Literally updating them as we speak - but they'll need to go through the update process (~2 days). Grab them here in the meantime:

http://fourwalledcubicle.com/fil...

Note that the LSS classifier is now deprecated, as I donated the code to Atmel and it's now built into 6.2.

- Dean :twisted:

Make Atmel Studio better with my free extensions. Open source and feedback welcome!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The new USB installer will take some time to install, this is most visisble for users of Windows 7. This because Windows 7 and 8 requires that some drivers must be registered in Windows Driver Store for them to work automatically when tool is plugged in. In XP the INF-files for the drivers are just copied to windows/inf folder. Windows trigger a Restore point for each driver. Take some time on Windows 7, a progressbar is displayed. Does not take THAT long. More some minutes than years. On Windows 8/8.1 is is much faster. This need only be done once on the machine

Lars Kvenild
Atmel Norway

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

As i said previously. no big deal. Make a coffee....have a beer...all good :wink:

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Regarding GDB, we've only enabled it for ARM projects for now, while we try and iron out the bugs in avr-gdb and avr32-gdb.

You can switch back to non-GDB by going to the Advanced tab in the project settings.

Also, the watch view has changed when in GDB mode. The Hexadecimal display toggle does not work yet, and the watch value modifiers have not yet been implemented for values computed by gdb. To print a variable in hex in GDB mode, you need to watch "var,hexadecimal".
Similarly, you now need to use the GDB syntax when forcing an expression to be an array:
value@array_length

-dan

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Oh joyous delight - see picture. When I click through to the log link it says:

[00D8:0C88][2014-02-28T16:42:48]i360: Creating a system restore point.
[00D8:0C88][2014-02-28T16:42:56]i361: Created a system restore point.
[00D8:0C88][2014-02-28T16:42:56]i000: Caching bundle from: 'C:\DOCUME~1\asl\LOCALS~1\Temp\{13fdfa7e-1bec-46cf-8231-3dd7bf6d0f3c}\.be\AtmelUSBInstaller.exe' to: 'C:\Documents and Settings\All Users\Application Data\Package Cache\{13fdfa7e-1bec-46cf-8231-3dd7bf6d0f3c}\AtmelUSBInstaller.exe'
[00D8:0C88][2014-02-28T16:42:57]i320: Registering bundle dependency provider: {13fdfa7e-1bec-46cf-8231-3dd7bf6d0f3c}, version: 6.2.223.0
[00D8:0E44][2014-02-28T16:43:27]e000: Error 0x800b010a: Failed authenticode verification of payload: C:\Documents and Settings\All Users\Application Data\Package Cache\.unverified\JungoUSBInstaller.msi
[00D8:0E44][2014-02-28T16:43:27]e000: Error 0x800b010a: Failed to verify signature of payload: JungoUSBInstaller.msi
[00D8:0E44][2014-02-28T16:43:27]e310: Failed to verify payload: JungoUSBInstaller.msi at path: C:\Documents and Settings\All Users\Application Data\Package Cache\.unverified\JungoUSBInstaller.msi, error: 0x800b010a. Deleting file.
[00D8:0E44][2014-02-28T16:43:27]e000: Error 0x800b010a: Failed to cache payload: JungoUSBInstaller.msi
[0E04:0968][2014-02-28T16:43:27]e314: Failed to cache payload: JungoUSBInstaller.msi from working path: C:\DOCUME~1\asl\LOCALS~1\Temp\{13fdfa7e-1bec-46cf-8231-3dd7bf6d0f3c}\JungoUSBInstaller.msi, error: 0x800b010a.
[0E04:0968][2014-02-28T16:43:27]e349: Application requested retry of payload: JungoUSBInstaller.msi, encountered error: 0x800b010a. Retrying...
[00D8:0E44][2014-02-28T16:43:28]e000: Error 0x800b010a: Failed authenticode verification of payload: C:\Documents and Settings\All Users\Application Data\Package Cache\.unverified\JungoUSBInstaller.msi
[00D8:0E44][2014-02-28T16:43:28]e000: Error 0x800b010a: Failed to verify signature of payload: JungoUSBInstaller.msi
[00D8:0E44][2014-02-28T16:43:28]e310: Failed to verify payload: JungoUSBInstaller.msi at path: C:\Documents and Settings\All Users\Application Data\Package Cache\.unverified\JungoUSBInstaller.msi, error: 0x800b010a. Deleting file.
[00D8:0E44][2014-02-28T16:43:28]e000: Error 0x800b010a: Failed to cache payload: JungoUSBInstaller.msi
[0E04:0968][2014-02-28T16:43:28]e314: Failed to cache payload: JungoUSBInstaller.msi from working path: C:\DOCUME~1\asl\LOCALS~1\Temp\{13fdfa7e-1bec-46cf-8231-3dd7bf6d0f3c}\JungoUSBInstaller.msi, error: 0x800b010a.
[0E04:0968][2014-02-28T16:43:28]e349: Application requested retry of payload: JungoUSBInstaller.msi, encountered error: 0x800b010a. Retrying...
[00D8:0E44][2014-02-28T16:43:28]e000: Error 0x800b010a: Failed authenticode verification of payload: C:\Documents and Settings\All Users\Application Data\Package Cache\.unverified\JungoUSBInstaller.msi
[00D8:0E44][2014-02-28T16:43:28]e000: Error 0x800b010a: Failed to verify signature of payload: JungoUSBInstaller.msi
[00D8:0E44][2014-02-28T16:43:28]e310: Failed to verify payload: JungoUSBInstaller.msi at path: C:\Documents and Settings\All Users\Application Data\Package Cache\.unverified\JungoUSBInstaller.msi, error: 0x800b010a. Deleting file.
[00D8:0E44][2014-02-28T16:43:28]e000: Error 0x800b010a: Failed to cache payload: JungoUSBInstaller.msi
[0E04:0968][2014-02-28T16:43:28]e314: Failed to cache payload: JungoUSBInstaller.msi from working path: C:\DOCUME~1\asl\LOCALS~1\Temp\{13fdfa7e-1bec-46cf-8231-3dd7bf6d0f3c}\JungoUSBInstaller.msi, error: 0x800b010a.
[0E04:0660][2014-02-28T16:43:28]e000: Error 0x800b010a: Failed while caching, aborting execution.
[00D8:0C88][2014-02-28T16:43:28]i330: Removed bundle dependency provider: {13fdfa7e-1bec-46cf-8231-3dd7bf6d0f3c}
[00D8:0C88][2014-02-28T16:43:28]i352: Removing cached bundle: {13fdfa7e-1bec-46cf-8231-3dd7bf6d0f3c}, from path: C:\Documents and Settings\All Users\Application Data\Package Cache\{13fdfa7e-1bec-46cf-8231-3dd7bf6d0f3c}\
[0E04:0660][2014-02-28T16:43:28]i399: Apply complete, result: 0x800b010a, restart: None, ba requested restart:  No

Attachment(s): 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

clawson wrote:
Oh joyous delight...
Are you an administrator? :wink:

"I may make you feel but I can't make you think" - Jethro Tull - Thick As A Brick

"void transmigratus(void) {transmigratus();} // recursio infinitus" - larryvc

"It's much more practical to rely on the processing powers of the real debugger, i.e. the one between the keyboard and chair." - JW wek3

"When you arise in the morning think of what a privilege it is to be alive: to breathe, to think, to enjoy, to love." -  Marcus Aurelius

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Xp needs the new strong keychain that Ms added in win8. This is either automatically downloaded if the machine is online during install, or it has been deployed by Ms in security patch some time ago. So either be online during the USB installation, or make sure that you have updated Xp (might be optional, not really sure). Currently on Kastrup, but will post some summaries of all the things I have read the last week (hotel net in Germany is slower than our Web team) when I get back to Trondheim and have had some sleep...

:: Morten

 

(yes, I work for Atmel, yes, I do this in my spare time, now stop sending PMs)

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I already got Cliff his Birthday present. :P He's going to just love it.

"I may make you feel but I can't make you think" - Jethro Tull - Thick As A Brick

"void transmigratus(void) {transmigratus();} // recursio infinitus" - larryvc

"It's much more practical to rely on the processing powers of the real debugger, i.e. the one between the keyboard and chair." - JW wek3

"When you arise in the morning think of what a privilege it is to be alive: to breathe, to think, to enjoy, to love." -  Marcus Aurelius

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

My XP VM was online when I tried to install but it's true I have it set to just tell me about updates rather than apply automatically so maybe that was it.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hmh, I thought it would force the key-chain to install if it did not have it...

Anyway, you need updated root certificates etc to verify the installer.

:: Morten

 

(yes, I work for Atmel, yes, I do this in my spare time, now stop sending PMs)

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I'll see what's waiting in the security update queue.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I had problems debugging a ATtiny2313A/ATtiny4313 with debugWire and JTAGICE3. Atmel guys confirmed me the issue is connected to a bug in Atmel Studio 6.

Do you know if this bug is resolved in Atmel Studio 6.2?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

Do you know if this bug is resolved in Atmel Studio 6.2?

Not without some specifics (e.g. bug number or something I can look up).

:: Morten

 

(yes, I work for Atmel, yes, I do this in my spare time, now stop sending PMs)

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

clawson wrote:
I'll see what's waiting in the security update queue.
300 updates later... Good golly, take advantage of them while they still exist. :shock: Now you have absolutely no right to complain any more, for now. You can start complaining all over again when you receive my gift. :P

"I may make you feel but I can't make you think" - Jethro Tull - Thick As A Brick

"void transmigratus(void) {transmigratus();} // recursio infinitus" - larryvc

"It's much more practical to rely on the processing powers of the real debugger, i.e. the one between the keyboard and chair." - JW wek3

"When you arise in the morning think of what a privilege it is to be alive: to breathe, to think, to enjoy, to love." -  Marcus Aurelius

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Checking out the new "live watch" feature. Pretty cool!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

That's Morten's baby - so thank him.

- Dean :twisted:

Make Atmel Studio better with my free extensions. Open source and feedback welcome!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Ran the beta on my avr32 with freertos/lwip.
Huge improvements all over!
Only remaining problem is that the watchdog driver still block the cycle counter.

- jan

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Ok, sum up.

6.2 Beta Full installer: 728MB
MD5 4a6843e8f09b82fdd3d870ba2917aa24
SHA1 aa2a7206dfda778eaedba4bdf26424a12c5fe69b

6.2 Beta Normal installer: 513MB
MD5 1b3ca3a913ac5ea484724ed8fb30ca56
SHA1 0681174b217b25ccae0c60c2cb5ae4f98106b1af

New Features
Major features by my head is

  • Atmel-ICE
  • Improved debug using internal GDB
  • OSCCAL
  • Data Breakpoints
  • Create ELF File
  • Cortex-M4/M3 Trace
  • IDR is finally shown
  • Live Watch

Atmel-ICE
Documentation

New all-round debugger. Supports AVR8/32 and SAM3/4 and SAM D/G. Supports full program and debug, and streaming SWO trace from Cortex-M3/M4. Priced from $32-$85 depending on kit version. See doc for differences.

Improved debug
Default for ARM now is to spin up a gdb instance internally that helps in stepping, running and evaulation of expressions. Plan is to do this for other architectures as well, with avr8 beeing the next. Still some issues before the beta, so gdb is not enabled for any other architecture than ARM.

Changes that could be visible is different strings when variables are watched, and slightly different syntax for array expansion etc. There is also a GDB console window available under View->Other Windows, but doing anything with the run state may screw up the sync between gdb and Atmel Studio.

Another good thing is that now, it is usually possible to recover the call stack in exception handlers for ARM, meaning that especially on Cortex-M0+ it is possible to find the reason for a HARDFAULT. This information is available in the callstack window.

OSCCAL
Same feature as in studio 4. Available as a page in the programming dialog for devices that has this. Shouldn't bee any functional difference.

Data Breakpoints
It's finaly here again! Databreakpoints which exposes the specific features of the selected device OCD. Available either by right clicking a variable, or through Debug->Data Breakpoints. The properties of a data breakpoint depends on the device, so a mega will for instance be limited relative to UC3. Link to documentation.

Create ELF File
Again, much asked for feature. Available through the Production File page in the programming dialog.

M3/M4 Trace
Shown at EW14. We are collaborating with Percepio AB to bring the power of their Tracealyzer into Atmel Studio. This provides live non-intrusive trace on the Cortex-M3 and M4. Plans are to extend this to also cover M0+ using normal data breakpoints for the full release. Other features planned are code profiling using PC sampler and other coole statistics.

IDR
Remeber IDR events from Studio 4.17? It's back! If a IDR event is recieved by the debugger, a tab in the output window is openend showing the hex value that was recieved. Note that in brown-out conditions the debugger may find many spurious events that are shown here.

Live Watch
A nice gui on top of databreakpoints, showing the update history of variables (break,read,run cycle). Enables you just to watch, show history of a watched variable and also provides a way to save the variables history to csv for further excel-ing.

List of bugs that are fixed is listed in the release notes as usual. Other things is a much rewamped ASF wizard that should (tm) be faster that it was. There has also been many optimizations on startup time of studio, so even cold startups should be noticeably faster than for 6.1.

:: Morten

 

(yes, I work for Atmel, yes, I do this in my spare time, now stop sending PMs)

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

Remeber IDR events from Studio 4.17?

You mean there was actually a version of 4 were this once worked? I remember trying this a few times but I could never get it to work so I just assumed it was something that had never worked (except for the spurious IDR 0xFF message people get when their JTAG cables were too long).

Oh deep joy.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
Remeber IDR events from Studio 4.17?

Some of us are old enough to remember IDR :)

I wrote:
If that 100ms is a limit, the throughput of 10 Bps is $%^&&* and not worth the effort.

No RSTDISBL, no fun!

Last Edited: Mon. Mar 3, 2014 - 09:35 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

old enough to remember

As I get older I realise how oxymoronic that phrase is!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

clawson wrote:
Quote:

old enough to remember

As I get older I realise how oxymoronic that phrase is!

Maybe "Old enough to know but young enough to remember" is a phrase better suited then?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hehe, I have used it even with the 10 pull rate, to do poor-mans function entry and spurious interrupt trace. Used it more or less as 256 leds for entry and exit markers. Still, better than actually having to wire up 256 leds.

:: Morten

 

(yes, I work for Atmel, yes, I do this in my spare time, now stop sending PMs)

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
Still, better than actually having to wire up 256 leds.

I am not THAT old (or desperate).

No RSTDISBL, no fun!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

meolsen wrote:
pozzugno wrote:

Do you know if this bug is resolved in Atmel Studio 6.2?

Not without some specifics (e.g. bug number or something I can look up).

I have a ATtiny4313 board with ISP/debugWire 6-poles connector and I have a JTAGICE3.

With a new/fresh device, the following are the steps to reproduce the problem:

- Launch Atmel Studio 6.2 (the behaviour is identical to Atmel Studio 6.1).
- Open Device Programming tool and check correct communication (I can read fuses, Flash memory and so on).
- Open a solution in Atmel Studio 6.2 with a single project for ATtiny4313.
- Launch debug session (Debug->Start Debugging and Break) and receive the message "Failed to launch debug session with debugWire... Do you want to use SPI to enable DWEN fuse?". Click Yes.
- I receive the message "debugWire is enabled. Toggle target power then click OK".
- Toggle the power on target board and click OK. I receive the message "Setting debugWire fuse seems to have failed...". Click OK.
- Debug session doesn't start and now I can't use Device Programming anymore.

It seems the fuse DWEN is programmed (but I'm not sure), but the IDE isn't able to reset it (I tried with "Start Debugging and Break", with "Start without Debugging", with "Continue", without success).

The only solution I have found is to start a debug session with Atmel Studio 5.1 (installed in a virtual machine) using the same JTAGICE3. Atmel Studio 5.1 is able to start debug session and disable it. From now on I can read fuses and memories from Atmel Studio 6 again.

I have also a ATmega32A board with JTAG interface and I can program and debug it with the same JTAGICE3 and Atmel Studio IDE.

IMHO, it's evident it's a bug in Atmel Studio 6/JTAGICE3/debugWire and I was hoping 6.2 solved the problem, but nothing :(

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi Pozzugno,

Do you by any chance have a capacitor on the RESET line ?
When using debugWire, there should be no capacitor on RESET.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

No capacitor or anything else on RESET line of ATtiny4313.

I don't think it's a hw problem, because AS 5.1 is able to start and stop debug sessions. It's a problem related only to AS 6 and JATGICE3 and debugWire.

Can someone here with the same hw/sw try the same sequence?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I just installed 6.2 and it can't find either of my programmers at all. I'm using a JTAGICE3 and a JTAGICE MKII. This is a brand new Windows 7 install as well. Both devices are listed in device manager without any exclamation points or obvious issues. AS6 simply sees no tools in any of the tool selection menus. Does anyone have any advice?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I have JTAGICE3 and Atmel Studio 6.2, installed in a Windows 7 system, detects it without problem (I have just a problem for debugWire).

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Bandtank: verify that the atbackend process is running and able to communicate over TCP (firewalls and antivirus may prevent this).
Also, try plugging/unplugging one of the tools.

If all that fails, turn on Diagnostic logging in Tools Options Tools and send me the log that comes in the Backend agent section of the Output view.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

http://asf.atmel.com/bugzilla doesn't have a section for AS6.2 Beta. I like to file a bug report. I get the following error message from the Backend Agent Output window. I'm using the SAMD20 Xplained board.

03 58 12 750: dap DAP_Commands reports error 0x1003
03 58 12 750: usb Error  in JTAGIceMk3::readAndDispatchEvents
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

how does one use the Percepio Trace functionality? I start and break, init sysclk, then add a Control-Flow Trace view. All I see, however, is "No data source".

Is there an Atmel note, or a guide of any sort for the trace functionality?

Also:
Do I have to update JTAGICE3 to latest firmware? I downgraded to 2.x

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

just found this in the help file:

Percepio Trace requires a debug probe with Serial Wire Debug (SWD) support and that the project is configured to use the SWD protocol, rather than JTAG.

Does that mean I can't do tracing with AS 6.2 + JTAGICE3 ?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

fuuuuuu wrote:
Does that mean I can't do tracing with AS 6.2 + JTAGICE3 ?

JTAGICE3 can do SWD with recent firmwares.

But I don't understand the source of confusion. Right now chips that support SWD (SAMDx) do not support JTAG and vice versa.

What MCU do you have?

NOTE: I no longer actively read this forum. Please ask your question on www.eevblog.com/forum if you want my answer.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
Percepio Trace requires a debug probe with Serial Wire Debug (SWD) support

I suppose the author meant the ITM trace which requires SWV or ETM interface. The SWV is just a regular uart (or actually uat) that can do NRZI or Manchester. ETM is a bit more complicated (and faster) and supported by bigger ARMs.
So for tinies your dongle must support SWV or, if ASes didn't lock this possibility, a serial2USB cable.

No RSTDISBL, no fun!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

Right now chips that support SWD (SAMDx) do not support JTAG and vice versa.

Not really applicable, as the D does not have SWO (also nown as SWV, but that's just something lost in translation). Also, most of our devices have both JTAG and SWD, but only SWD enables the SWO for trace output (this is the TDO pin in JTAG).

The new firmware for the jtagice3 in studio 6.2 enables swo capturing. However, currently the jtagice3 only supports the D and G family, both which to my knowledge lacks a ITM and TPIU which is needed for SWV. We are working with Percepio to get some more usable feedback on this, as Atmel Studio reports trace capabilities based on the DAP ROM table of the device. Currently Percepio just does nothing if we report that a device does not support anything.

There are some options to uncheck the "Non-Intrusive" flag, which then will use databreakpoints instead of watchpoints, with all the slowdown that that will cause.

:: Morten

 

(yes, I work for Atmel, yes, I do this in my spare time, now stop sending PMs)

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

so, correct me if i am wrong:

the new debug features in 6.2 will work only for:
-Cortex M3/M4 with
-AVR One!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Non-intrusive trace for Cortex-M3/4 with EDBG or Atmel-Ice or Sam-Ice or some other segger.
Intrusive trace uses databreakpoints, so anything that has those (mainly cortex, not tried on uc3 yet but should work).

Currently, none of the trace specific features of the One has been implemented.

:: Morten

 

(yes, I work for Atmel, yes, I do this in my spare time, now stop sending PMs)

Pages