LinuxKPI: 802.11: lock MO tx/wake_tx_queue() downcalls

[bz]

Summary

Lock the two TX MO downcalls into driver/firmware in
lkpi_80211_txq_tx_one() to make sure they cannot happen in the
middle of other (net80211 triggered) updates calling down into
the driver/firmware.

MFC after: 3 days

[cc]

I think LKPI_80211_LHW_LOCK_ASSERT can help remind us of such LKPI_80211_LHW_LOCK requirement for lkpi_80211_mo_* functions.

sys/compat/linuxkpi/common/src/linux_80211.c

3773–3774

Please start by adding the LKPI_80211_LHW_LOCK_ASSERT for each lkpi_80211_mo_*. Here we shall add LKPI_80211_LHW_LOCK_ASSERT in lkpi_80211_mo_wake_tx_queue().

3788–3789

Here we shall add LKPI_80211_LHW_LOCK_ASSERT in lkpi_80211_mo_tx().

[bz]

@cc Adding LOCK asserts is a very good idea; not all the functions need them if I remember correctly. Should I really add it to this "fix" or should I do a separate full pass on the linux_80211_macops.c file?

[cc]

In D43966#1023788, @bz wrote:

@cc Adding LOCK asserts is a very good idea; not all the functions need them if I remember correctly. Should I really add it to this "fix" or should I do a separate full pass on the linux_80211_macops.c file?

If you put a separate "full pass of asserts" on a plan, with more scrutiny and tests, then I think we can move forward with this fix.