I'm using STM32F401 for the test. The openocd config file is in the attached
zip.
To help understanding this problem better, I captured the SWD sequence using a
logic analyzer. The capture files are attached and can be viewed by
[DSView](https://www.dreamsourcelab.com/download/). Ch0 in the file is SWCLK,
Ch1 is SRST, Ch2 is SWDIO. You can see that in the NG file the reset line is
released earlier than the OK file, which is certainly not expected.
I tried some debugging, and found that if you comment out line 108 (`adapter
speed 2000
` in the `reset-start` event block) in tcl/target/stm32f4x.cfg, the premature
release of srst line will no longer happen. Also I if you comment out
`jlink_speed` at all, the problem also disappears. So it is clear that
`jlink_speed` is deasserting the reset line.
Included in the attachment zip are two logic analyzer capture files (.dsl), two
log files, and the config. `openocd-ng.dsl` and `openocd-ng.log` is produced
with the original OpenOCD, which fails to connect to the MCU; `openocd-ok.dsl`
and `openocd-ok.log` is produced with [the
patch](https://github.com/t123yh/openocd/commit/fd1997bc21c1c7199f4715c0e5ab24f64bb8bf90)
applied (no other files were changed).
I doubt this could also be a Segger bug, of which I sent an email to them (they
did not reply yet). I hope this can be reproduced or analyzed on your side.
Attachments:
-
[attach.zip](https://sourceforge.net/p/openocd/tickets/_discuss/thread/05dde6e6a8/eefa/attachment/attach.zip)
(665.4 kB; application/zip)
---
** [tickets:#272] J-Link v9 + STM32: failed to connect under reset**
**Status:** new
**Milestone:** 0.9.0
**Created:** Fri Jun 26, 2020 01:13 PM UTC by Yunhao Tian
**Last Updated:** Fri Jun 26, 2020 01:13 PM UTC
**Owner:** nobody
Hi,
I'm using J-Link v9 with STM32. It works fine except connect_assert_reset.
After some investigation I noticed that the jlink_speed function deasserts the
reset line, which caused the STM32 to boot too early when it's supposed to be
in reset state, and leads to `Error: timed out while waiting for target halted`
problem.
This commit
(https://github.com/t123yh/openocd/commit/fd1997bc21c1c7199f4715c0e5ab24f64bb8bf90)
temproarily fixes the problem, but I wonder if there's a more elegant solution.
---
Sent from sourceforge.net because [email protected] is
subscribed to https://sourceforge.net/p/openocd/tickets/
To unsubscribe from further messages, a project admin can change settings at
https://sourceforge.net/p/openocd/admin/tickets/options. Or, if this is a
mailing list, you can unsubscribe from the mailing list.
_______________________________________________
OpenOCD-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/openocd-devel
