Repository navigation
Retire boards/: every board is upstream's, plus fragments and two small variants - #29
Merged
Merged
Conversation
…heir repos, the LCD-7 and T-Embed as ESP32_GENERIC_S3 variants
… step 5 was proven with
bdbarnett
added a commit
that referenced
this pull request
Oct 19, 2026
…ts a wedge (#69) host_stop() gave teardown 200 ticks before calling the host wedged. At CONFIG_FREERTOS_HZ=100 that is 2 s; at 1000 Hz, which the ESP32-S3-Touch-LCD-7's firmware runs, it is 200 ms, while teardown with a device attached takes about 310 ms. Every such stop was reported as a wedge and the next host_start() refused, whether the device had been opened or not (#29). The stop wait is now USBIF_HOST_STOP_WAIT_MS (6 s, above teardown's own worst case), and every other wait that counted vTaskDelay(1) calls (interface release retries, CDC and MIDI OUT waits, MSC's per-stage timeout) is bounded in milliseconds through usbif_ticks.h. Teardown's log lines carry the time since it began. Fixes #29
bdbarnett
added a commit
that referenced
this pull request
Oct 19, 2026
…#29) An esp32 build with audioif and audiodsp checked out anywhere but side by side (git worktrees linked into a module directory, say) left _audioif out of the image with only a CMake status line, so audiodev quietly fell back to machine.I2S. Both build files now look for audiodsp first among the user C modules the build was given (USER_C_MODULES, which on 1.29 includes the manifests' c_module()s; on a Make port also one level inside each entry, for the parent-directory form), then beside this directory. When it is in neither place the build fails, naming the fix: add audiodsp to the modules, or set AUDIOPUMP_AUDIODSP_DIR / AUDIODSP_DIR. A non-esp32 CMake port still skips this module as before. Fixes #28.
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
1 participant
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.
Step 5 of the build plan.
boards/esp32/is gone, and building for any board we use is onebuild_mp.pyline on an upstream board. The newcomer's guide has the command for each.The four old boards' settings sorted into three homes:
variants/esp32/sdkconfig.esp32s3(PSRAM, Wi-Fi and lwIP tuning, flash coredump) andsdkconfig.esp32p4(NimBLE over the C6, chunked cache writeback). The S3 also getspartitions.esp32s3.csv, upstream's table plus a coredump partition.build_mp.pynow finds apartitions.csvby the same convention as the fragments, and autosize grows it.ESP32_GENERIC_S3, each anmpconfigvariant.cmakethat includes upstream'sSPIRAM_OCTplus a fragment:LCD_7(cache and tick tuning) andT_EMBED(flash auto-suspend). The plan said one variant. The T-Embed needs its own because IDF supports auto-suspend only on particular flash chips, so it can't be a chip-wide setting.Proven on the bench, 2026-10-05, all built with
--modules all:C6_WIFI, 16 MB) and P4 Touch-LCD-4B (PRE_REV3_C6_WIFI, 32 MB): every module imports, castif included. Wi-Fi through the C6 scans and BLE comes up. The panel detects its OV5647 and captures a JPEG.LCD_7, 8 MB): every module imports except castif (P4-only), the coredump partition is there, and Wi-Fi and BLE work.lv_test_timer.pyandboxlines.pyrun clean on the glass, after DotClockFramebuffer: keep both panel buffers in step with what was painted displayif#46.lv_test_timer.pyruns clean on the glass after wt32sc01-plus: pins from the datasheet, freq=, and a 10 MHz bus pydevices#122.The T-Embed was built (its sdkconfig has auto-suspend and keeps the default 100 Hz tick) but wasn't attached, so it hasn't booted.
modules.lockmoves cameraif, displayif, usbif and pydevices to the commits this was proven with.