summaryrefslogtreecommitdiff
path: root/docs/reference/device_issues.rst
diff options
context:
space:
mode:
authorHiFiPhile <[email protected]>2026-06-23 23:12:51 +0200
committerHiFiPhile <[email protected]>2026-06-23 23:12:51 +0200
commit033dc6bb77e8a9e1c1172d74d47707848b5c5bf5 (patch)
treee7672f9bda54aa1a243b2a1c0f74a48055569c63 /docs/reference/device_issues.rst
parentb5e63ca44c846813ece9ad9df684cae0d1d5b542 (diff)
parentcd3561bf158afd5a5718904b8139a338d1e3b67c (diff)
Merge remote-tracking branch 'tinyusb/master' into fix-stm32-usbc
Diffstat (limited to 'docs/reference/device_issues.rst')
-rw-r--r--docs/reference/device_issues.rst35
1 files changed, 35 insertions, 0 deletions
diff --git a/docs/reference/device_issues.rst b/docs/reference/device_issues.rst
new file mode 100644
index 000000000..ae9cd55f1
--- /dev/null
+++ b/docs/reference/device_issues.rst
@@ -0,0 +1,35 @@
+Device specific known issues and workarounds
+===============================================
+This page lists known issues and workarounds for specific devices.
+
+NXP LPC54600
+----------------
+**Severity: High**
+
+**Not recommended for USB device applications (except high-speed host controller)**
+
+Reference: `LPC54600 Errata Sheet`_
+
+.. _LPC54600 Errata Sheet: https://www.nxp.com/docs/en/errata/ES_LPC546XX.pdf
+
+The LPC54600 series have a very buggy USB controller, with 17 issues listed in the errata which is more than half of the total issues.
+
+Most severe issues are:
+
+- USB.2: In USB high-speed device mode, the NBytes field is not correct after BULK IN transfer
+- USB.5: In USB full-speed host mode, linked list on done queue is broken.
+- USB.15: USB high-speed device in endpoint TX data corruption
+
+WCH CH32F20x/CH32V20x/CH32V30x
+---------------------------------
+**Severity: Medium**
+
+**Not recommended for USB audio applications**
+
+Reference: `CH32V30X Reference Manual`_ USBFS/USBHS controller chapter
+
+.. _CH32V30X Reference Manual: https://www.wch-ic.com/downloads/CH32FV2x_V3xRM_PDF.html
+
+Data corruption may occur on isochronous endpoints. Due to the lacking of FIFO for interrupt status registers, later completed transfer will overwrite `INT_ST` and `RX_LEN` register if previous transfer processing is not completed.
+
+Other types of transfers are not affected.