Testing with AR9271 USB device I was able to start 8 VAPs, but one client device could connect only if 7 VAPs were configured. This might need more work to make the feature fully usable, but even at present situation it enables functionality that was previously disabled.
Updated ath9k_htc firmware is required too, PR already opened: qca/open-ath9k-htc-firmware#149

Signed-off-by: Tomislav Požega <pozega.tomislav@gmail.com>

Move dynack debug code from debug.c to common-debug.c and adjust ath9k/ath9k_htc debug for common dynack output.

Signed-off-by: Tomislav Požega <pozega.tomislav@gmail.com>

Enable ANI debug output as well as disabling or enabling the feature.

Signed-off-by: Tomislav Požega <pozega.tomislav@gmail.com>

The increase of stations limit eliminates problem of only 7-8 clients able to connect to an AP.

Signed-off-by: Tomislav Požega <pozega.tomislav@gmail.com>

This enables 5/10 MHz wide channels via debug file for both ath9k and ath9k_htc drivers. Connection tested and works in legacy and HT20 modes. This changeset is based on OpenWrt patch: 512-ath9k_channelbw_debugfs.patch

Signed-off-by: Tomislav Požega <pozega.tomislav@gmail.com>

Signed-off-by: Tomislav Požega <pozega.tomislav@gmail.com>

Signed-off-by: Tomislav Požega <pozega.tomislav@gmail.com>

Signed-off-by: Tomislav Požega <pozega.tomislav@gmail.com>

not tested

…vior" (#11)

fix fan slowdown on temp. decrease, v4.9, compaq 6735b, reverts https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/drivers/thermal/step_wise.c?id=07209fcf33542c1ff1e29df2dbdf8f29cdaacb10

fix fan speedup on temp. increase, v4.9, compaq 6735b, similar as the acer issue in https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/drivers/thermal/step_wise.c?h=v3.12.74&id=bca746037a308e2fbe742248d34635f07be2bfd9

commit 212e155b14636edec1aebe9a003a727e1250b242
Author: Bjorn Helgaas <bhelgaas@google.com>
Date:   Tue Aug 23 11:56:26 2011 -0600

    x86, ioapic: Reserve less space for IOAPICs
    
    Previously we reserved 1024 bytes, but that's more space than the IOAPIC
    consumes, and it can cause conflicts with nearby devices.
    
    Also, delay the IOAPIC output a bit so we can print the actual range we
    claim.

note: breaks CCFL backlight...