2014-08-26 07:44 AM
Dear community,
one of our products uses a STM32F417IGH6 controller and has previously been build using silicone revisions A and Z. The current batch is the first with silicone revision 1 and serveral problems came up. 90% of the boards will not connect to a debugger (e.g. STLink/V2 using STLink Utility V3.4.0) and therefore cannot be programmed or tested, no matter if using SWD or JTAG. 5% are able to connect from time to time and 5% can connect reliably to the debugger. Reading these devices, I can verify, that the device is empty (0xFF on all addresses), but STLink Utility shows Revision ID: Rev. X, while the package is marked ''1''. All devices have the datecode 427. There are no changes to PCB or other populated parts, the only change is the STM32's revision. The errata sheet has also no item, that seems to be related to debugging. #stm32f4172014-08-26 08:09 AM
Do you have connectivity to NRST? How about the BOOTx pins?
Are you using the most current ST-LINK firmware, and Utilities? Have you tried ''Connect Under Reset'' mode? Have you tried other more commercial JTAG/SWD pods?2014-08-26 11:39 PM
Dear Clive1,
we are using the ST-Link Utility v3.4.0 with USBDriver.dll v4.3.3.0 and the firmware in the ST-Link/V2 is V2.J21.S4. All was downloaded directly from ST's website and reinstalled, so I'm quite sure, this is the latest version. NRST is connected to a reset controller, which works fine, verified on an oscilloscope, and is also connected to a legacy 20pin ARM debug connector. BOOT0 is pulled up and also connected to a switch to allow to pull it high or low. Both modes made no difference. BOOT1 is pulled down by a 10k resistor. No option in the ST-Link does make a difference, ''connect under reset'' was the first thing I tried, since the utility recommended it.2014-08-26 11:52 PM
2014-08-27 12:10 AM
Thanks for the remark, but I don't know how to change the topic titel, and I'm also quite sure all in this forum know what I ment.