Skip to content

Bluetooth: btmtk: fix subsystem reset error reporting

Ismail Tarim linux-bluetooth linux-mediatek lkml 2 versions Applied

Status from this message: This series was applied to bluetooth/bluetooth-next.git (master)

Get series with b4
b4 am [email protected]

Versions newest first · linked by author and title

  1. v2 Latest2 patches 1 replyApplied 2/2
  2. v12 patches 1 replySuperseded

Changes in v2 from the cover letter

- 1/2: shorten the subject so the CI gitlint check passes (was 81 > 80).
- 1/2: document how the failing path can be reached on demand.
- No functional change; the diff is identical to v1.

Patches in v2 All 2 applied

#Subject and filesSizeRepliesApplied
1/2Bluetooth: btmtk: Do not report success when subsys reset failsdrivers/bluetooth/btmtk.c+3 -11 replyApplied
2/2Bluetooth: btmtk: Do not discard the subsystem reset timeoutdrivers/bluetooth/btmtk.c+5 -20 repliesApplied

Reviews trailers per patch

No Reviewed-by, Acked-by or Tested-by trailers for v2 yet.

Read the v2 thread

Cover letter [PATCH v2 0/2] Bluetooth: btmtk: fix subsystem reset error reporting

Two bugs in btmtk_usb_subsys_reset() that make it report success to its caller when the subsystem reset has actually failed.

Patch 1 covers the case where the post-reset chip id reads back as zero:
the function logs "subsys reset fail." and then returns 0.

Patch 2 restores propagation of the MTK_BT_RST_DONE poll timeout, which has been silently overwritten since commit 3dcb122b3064 ("Bluetooth:
btusb: mediatek: return error for failed reg access") started assigning err at the chip id read.

Found while investigating an MT7902 [13d3:3579] whose controller stops responding and never returns to the USB bus. On that device the failing path can be reached on demand by requesting a coredump, since btmtk asks for a reset once the dump completes:

  # echo 1 > /sys/class/bluetooth/hci0/device/coredump

The captured firmware coredump shows the controller asserting inside its ROM rather than in the downloadable blob, which is consistent with the failure surviving a firmware update:

<ASSERT> system/rom/transport/tra_usb3.c #764 - BTSYS, id=0x4 idle

Both patches only correct the error reporting; neither changes the recovery behaviour or makes that controller recoverable. They are a prerequisite for anything that wants to react to a failed reset, since today the failure is not visible to the caller.

Compile tested on x86_64 (W=1, no new warnings); checkpatch --strict clean.

Changes in v2:
- 1/2: shorten the subject so the CI gitlint check passes (was 81 > 80).
- 1/2: document how the failing path can be reached on demand.
- No functional change; the diff is identical to v1.

v1: https://lore.kernel.org/linux-bluetooth/[email protected]/

Ismail Tarim (2):
Bluetooth: btmtk: Do not report success when subsys reset fails
Bluetooth: btmtk: Do not discard the subsystem reset timeout

 drivers/bluetooth/btmtk.c | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

base-commit: 15ef2f78c49d20d53ec7c0f1c9b40b02e089f2d6

Signature · 1 line
2.55.0