ports/stm: bootloader entry on STM32F4 - #11270
Merged
Merged
Conversation
mikeysklar
changed the title
mikeysklar
force-pushed
the
stm32f4-dfu-retry
branch
2 times, most recently
from
August 29, 2026 03:06
243c173 to
78db9aa
Compare
mikeysklar
force-pushed
the
stm32f4-dfu-retry
branch
from
August 29, 2026 03:37
78db9aa to
1e0e328
Compare
The ROM clocks USB from the HSE but does not know which crystal is fitted, so it measures one against the HSI. On a miss it resets the part instead of starting DFU: AN2606 Figure 32/33, "HSE detected" -> no -> "Generate System reset". On a Feather STM32F405 Express and its 12 MHz crystal a single jump reached DFU 5 times in 12. Record the request in a backup register and take a real reset, then jump from the top of port_init() and retry when the ROM bounces us. That reaches DFU 12 times in 12. Also drops the HAL_RCC_DeInit()/HAL_DeInit()/NVIC teardown, which a system reset supersedes. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01TehTMf9ApHxxU5UNHosXKj
mikeysklar
force-pushed
the
stm32f4-dfu-retry
branch
from
August 29, 2026 05:24
1e0e328 to
7717926
Compare
Collaborator
Author
|
The |
tannewt
requested changes
Aug 31, 2026
tannewt
left a comment
tannewt
left a comment
Member
There was a problem hiding this comment.
One question. Thanks for looking into this.
Move the pending ST system bootloader request from RTC->BKP1R to port_set_saved_word(), which is RAM that startup does not clear. This drops the backup domain clock enable and access unlock at both call sites, and removes the ordering constraint against the __HAL_RCC_BACKUPRESET_FORCE() further down port_init(). The RTC staleness window stays. It only needs a read of RTC->TR, which does not require unlocking the backup domain for write. safe_mode.c shares the saved word. Its guard is 0xad0000af under mask 0xff0000ff and the bootloader magic is 0xf05a0000 under mask 0xffff0000, so neither can be read as the other, and the word is only zeroed when the magic is ours. Co-Authored-By: Claude <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What
RunMode.BOOTLOADERand 1200 baud touch reach the bootloader 40% of the time on STM32F405. Retrying gives 12/12.Why
ROM HSE detection is marginal at 12 MHz (AN2606) and resets on a miss. Firmware cannot fix it, only retry.
History
#3444 requested, #6919 implemented, #6961 fixed. Verified only on NUCLEO-F446RE, whose HSE is 8 MHz.
How
reset_to_bootloader()sets a backup register and resets.check_enter_bootloader()inport_init()jumps, retrying when the ROM resets us.Notes
BKP1R matches other ports' double-tap state; BKP0R already used. Requests older than 5 seconds are dropped as stale.
Evidence
Feather STM32F405, 12 MHz crystal, 10.3.0-rc.0, Linux host. Each trial power cycles USB, triggers over REPL, reads idProduct.
maindfu-util -s 0x08000000:leave -D firmware.bindfu-util --leaveDecisions
AI assistance
Claude Code was used. All hardware trials run on real boards and reviewed by hand.