[0/1] MHI fix for v5.16 - Take 2

Message

Manivannan Sadhasivam Dec. 8, 2021, 8:57 a.m. UTC

Hi Greg,

Here is the final MHI fix for the v5.16 cycle. Kalle reported a regression with
ath11k stack due to 020d3b26c07a, and that's been affecting multiple WLAN
products like QCA6390 and WCN6855. The regression doesn't let the device to
resume from suspend and that has been due to a firmware bug. The device goes
to a bad MHI state during suspend and that can be seen in the MHI state during
resume. But that doesn't seem to affect the functionality of the device as it
continues to work fine if we allow it to resume.

Ideally, this should be fixed in the firmware but as like most of the products
out in market, getting a firmware fix for this issue not doable. So we have to
workaround for it in the MHI stack by allowing it to resume irrespective of the
MHI state.

Since the fix patch also touches ath11k stack, Kalle has given his Ack so that
the patch can be taken via mhi/char-misc tree.

Thanks,
Mani

Loic Poulain (1):
  bus: mhi: core: Add support for forced PM resume

 drivers/bus/mhi/core/pm.c             | 10 +++++++---
 drivers/bus/mhi/pci_generic.c         |  2 +-
 drivers/net/wireless/ath/ath11k/mhi.c |  6 +++++-
 include/linux/mhi.h                   |  3 ++-
 4 files changed, 15 insertions(+), 6 deletions(-)