[??] IAR 5.20 and pre silicon rev Y

[adriano2]

Posted on December 10, 2008 at 10:26

[??] IAR 5.20 and pre silicon rev Y

[adriano2]

Posted on May 17, 2011 at 12:55

Dear All

we are using the combination IAR 5.20, STM32 Rev Z and we have a strage behaviour: by using the YELLOW JLINK, we are unable to flash correctly the device, instead by using old BLACK JLINK, all work fine. Do someone has faced with the same behaviour? Bye Genius1974

[jj]

Posted on May 17, 2011 at 12:55

Greetings- (and glad to meet someone of such stature...)

This from ST's TN0067 (14898.pdf listed in tech notes): While running the tests with this new toolchain revision, STMicro and IAR found some incompatibility with revisions B and Z of the STM32F101xx and STM32F103xx medium-density devices. ST recommends to use the latest revision of the STM32F101xx and STM32F103xx medium-density, that is revision ''Y.'' ST closed this post but there is quite a history ''readable'' for your review/understanding. Basically - with the newer versions of IAR (& other IDEs) you should NOT be using Rev ''Z.'' We too had black & yellow JLinks - black one is way old & may not be suitable for use with IAR 5.2. Check w/Segger for ''source'' confirmation. There are regular updates from Segger - if your yellow JLink is marked, ''JTAG/SWD'' that's the best. (SWD saves you 2-3 pins while supporting full-speed Debug) (s) Dull Normal-1950

[relaxe]

Posted on May 17, 2011 at 12:55

Does this affect -ONLY- medium density devices?

We currently use the yellow Rev6 Jlink (jtag/swd), IAR 5.20 and STM32F103ZET6. To my understanding, this is a rev Z high density device.

[adriano2]

Posted on May 17, 2011 at 12:55

Thank you for your fast reply.

To complete the information from our side: we are using STM3210E-EVAL board MB672 revD where is present STM32F103ZET6 Rev.Z. We are able to program this board with IAR 5.2 and BLACK JLINK and Segger JFlash v3.96D. At the same time, we are not able to program it with the YELLOW JLINK. Bye Genius1974

[jj]

Posted on May 17, 2011 at 12:55

@relaxe - I recall that TN0067 specified only med. density devices. Urge you to read to learn all details. (14898.pdf listed in ST tech notes) By the way - we've enjoyed/benefited from many of your recent posts.

@genius - would have thought just

[jj]

Posted on May 17, 2011 at 12:55

@genius-

What do you mean by, ''unable to flash correctly the device?'' Give us more details, s'il vous plait...

[relaxe]

Posted on May 17, 2011 at 12:55

I just got scared with all of this!

A quick trip to the segger web site showed some pieces of software to read/write the STM32 outside or IAR. I suggest anybody having doubts over their code flashing integrity should flash their board, then read the flash back with one of segger's utility, then compare the hex dump with what should have been there. Also, in IAR, you can fiddle with the JLink setting in the project options. Maybe trying ''Fixed'' speed at a very low value (20khz?), or trying the SWD to see if you get the same behaviour from your target... -Relaxe

[adriano2]

Posted on May 17, 2011 at 12:55

Dear jj.sprague

I receive this message when I use Yellow Jlink (v6 07-35): Tue Dec 09 18:03:48 2008: Loaded macro file: C:\Program Files\IAR Systems\Embedded Workbench 5.20\ARM\config\flashloader\ST\FlashSTM32F10xxx.mac Tue Dec 09 18:03:48 2008: DLL version: V3.96d, compiled Nov 21 2008 18:59:52 Tue Dec 09 18:03:48 2008: Firmware: J-Link ARM V6 compiled Nov 21 2008 18:12:00 Tue Dec 09 18:03:48 2008: JTAG speed is initially set to: 32 kHz Tue Dec 09 18:03:48 2008: TotalIRLen = ?, IRPrint = 0x..050104EEFFFFFFC008EFFFFFFFFFFFFF Tue Dec 09 18:03:48 2008: TotalIRLen = 121, IRPrint = 0x19050104EEFFFFFFFFFFFFFFC008EFFF Tue Dec 09 18:03:48 2008: TotalIRLen = ?, IRPrint = 0x..050104EEFFFFFFC008EFFFFFFFFFFFFF Tue Dec 09 18:03:48 2008: Resetting target using RESET pin Tue Dec 09 18:03:54 2008: Halting CPU core Tue Dec 09 18:04:00 2008: Fatal error: Could not find supported CPU core on JTAG chain Bad JTAG communication: Write to IR: Expected 0x1, got 0x7 (TAP Command : 2) @ Off 0x5. Since, I'm unable to flash. Instead by a simple change of JLINK I'm able to flash correctly the micro. Bye

[jj]

Posted on May 17, 2011 at 12:55

Hmmmm-

Like relaxe's suggestion - change to slower JTAG speed via IAR's Project>Options... Should this fail - ''roll back'' to dll ver 3.90 - which works for us/many of our clients. If this action still fails then urge you to call/email Segger support - they seem quite capable... Let us know, please.