Topic: davidc502 1900ac 3200acm builds

The content of this topic has been archived between 26 Feb 2018 and 7 May 2018. Unfortunately there are posts – most likely complete pages – missing.

Page 122 of 229

Post #3026

starcms

11 Aug 2017, 23:21

kyzhk wrote:

Thanks starcms.
I looked at the source code for reghack (http://luci.subsignal.org/~jow/reghack/) and looks like I only need to change these two lines replacing 30 to 4:
struct ieee80211_reg_rule r2 = REG_RULE(2400, 2483, 40, 0, 30, 0);
struct ieee80211_reg_rule r5 = REG_RULE(5140, 5860, 160, 0, 30, 0);
and recompile an arm version to patch the powertable without recompiling cfg80211.ko

Or you could just apply the reghack2 file I had mentioned earlier in the thread (which is already compiled to work with the wrt1200/1900 series; you could apply it to the 3200, as well as the 1200AC V2 and 1900ACS V2 also but it wouldn't do anything), and then you can change the transmit power level to anything you want (from 4 to 30 dBm). It would work perfectly on your WRT1900AC V1 (mamba).

The only reason you are having difficulty changing the power level as-is is because I never noticed that in the power table, the values for the mamba (WRT1900AC V1) are more restricted than those of the WRT1900AC V2, WRT1900ACS V1, and WRT1200AC V1 (the models which read the power table from firmware, not eeprom). Applying reghack2 the way it is would fix that on the WRT1900AC V1.

So there's really 3 cases your router can fall under, not 2 as previously thought:

a) 1900AC V1: all by itself. It has a restricted power table which doesn't allow changing of transmit power level, but it is in firmware and can be modified and reghack2 will do so and work just fine. Can be modified by other means also.

b) 1900AC V2, 1900ACS V1, 1200AC V1: has a power table that allows the transmit power and country code to be modified. Power table is in firmware and can be modified further using reghack2 or other means.

c) 1200AC V2, 1900ACS V2, and 3200ACM: power table is in eeprom and you can't modify anything, period.

Edit: Also, reghack doesn't simply change the two lines to:

struct ieee80211_reg_rule r2 = REG_RULE(2400, 2483, 40, 0, 30, 0);
struct ieee80211_reg_rule r5 = REG_RULE(5140, 5860, 160, 0, 30, 0);

it also removes any other restrictions preventing you from lowering the transmit power. So if you just apply it the way it is, it will allow you to change the transmit power from 4dBm to 30dBm on any channel and negate the limited power table on your mamba.

(Last edited by starcms on 11 Aug 2017, 23:46)

Post #3027

oldchip

12 Aug 2017, 15:21

starcms wrote:

beachbum wrote:
Just a quick follow up to the page loading problem, I went back to LEDE 17.01.0 to make sure I wasn't remembering wrong and that the page loading problem wasn't present in it, spent the past few days running it with dnscrypt and without dnscrypt, no page loading problems either way, and like starcms mentioned nonwildcard is set to 0 by default, so if you're still having problems with page loading just go back to 17.01.0 and you should be good.
Or just use @david's latest build and set nonwildcard equal to 0 in /etc/config/dhcp. And/or if you are using custom DNS servers, enter them in DHCP and DNS settings and check the box for Ignore Resolv File as I had mentioned. Either or both of those things should fix the DNS resolving issue. It's been confirmed that this solves any resolving issues a few people were having.
LEDE 17.01.0 is built from the "stable" branch/not the development branch like @david's builds, is a few months out of date (which puts it several months behind the development branch, especially regarding mwlwifi driver), and most importantly isn't compiled by @david with all the added features/packages and the amsdu fix in the patched driver. Doesn't belong in this thread. Pretty much anyone would see decreased performance with 17.01.0, especially those with a 3200ACM.

the page loading problem is still having problem i think because sw. when i add new vlan tag eth1 and wan. after creat new dhcp interface connect and reboot router, the page loading very slow. I disconect the new created interface, everything come back nomal. In my case, @starcms guide don't work.

(Last edited by oldchip on 12 Aug 2017, 16:16)

Post #3028

davidc502

12 Aug 2017, 18:04

Driver version 10.3.4.0-20170810, which includes disabling amsdu is now ready. I will start building the wifi driver for those on kernel 4.4.x, and should be ready this afternoon.

The driver below is for those who are using the latest build with Kernel 4.9.40.
https://davidc502sis.dynamic-dns.net/re … _vfpv3.ipk

So far I've tested on the 1900acs, and is working well.

(Last edited by davidc502 on 12 Aug 2017, 18:23)

Post #3029

starcms

12 Aug 2017, 18:59

davidc502 wrote:

Driver version 10.3.4.0-20170810, which includes disabling amsdu is now ready. I will start building the wifi driver for those on kernel 4.4.x, and should be ready this afternoon.

The driver below is for those who are using the latest build with Kernel 4.9.40.
https://davidc502sis.dynamic-dns.net/re … _vfpv3.ipk
So far I've tested on the 1900acs, and is working well.

Tested on 1200ac, working great! amsdu confirmed disabled and ping issue fixed. driver version confirmed to be 0810. Thanks @david!

For those who need instructions, it is very simple

a) You must be running @david's latest build (r4651) with kernel 4.9 (this is everyone but those with a mamba, 1900AC V1)
b) Login via SSH
c) cd /tmp
d) wget https://davidc502sis.dynamic-dns.net/re … _vfpv3.ipk
e) opkg install kmod-mwlwifi_4.9.40+10.3.4.0-20170810-1_arm_cortex-a9_vfpv3.ipk
f) reboot
g) enjoy!

For those with a mamba (1900AC V1) running the kernel 4.4 version of @david's build r4651, as he said above, he will release a compatible updated driver for you shortly. Same instructions as above once it comes out, except change the url in step d) to point to the link he gives for the kernel 4.4 version, and step e) to match the name of the file in the url.

(Last edited by starcms on 12 Aug 2017, 19:35)

Post #3030

kyzhk

12 Aug 2017, 19:00

starcms wrote:

kyzhk wrote:
Thanks starcms.
I looked at the source code for reghack (http://luci.subsignal.org/~jow/reghack/) and looks like I only need to change these two lines replacing 30 to 4:
struct ieee80211_reg_rule r2 = REG_RULE(2400, 2483, 40, 0, 30, 0);
struct ieee80211_reg_rule r5 = REG_RULE(5140, 5860, 160, 0, 30, 0);
and recompile an arm version to patch the powertable without recompiling cfg80211.ko
Or you could just apply the reghack2 file I had mentioned earlier in the thread (which is already compiled to work with the wrt1200/1900 series; you could apply it to the 3200, as well as the 1200AC V2 and 1900ACS V2 also but it wouldn't do anything), and then you can change the transmit power level to anything you want (from 4 to 30 dBm). It would work perfectly on your WRT1900AC V1 (mamba).
The only reason you are having difficulty changing the power level as-is is because I never noticed that in the power table, the values for the mamba (WRT1900AC V1) are more restricted than those of the WRT1900AC V2, WRT1900ACS V1, and WRT1200AC V1 (the models which read the power table from firmware, not eeprom). Applying reghack2 the way it is would fix that on the WRT1900AC V1.
So there's really 3 cases your router can fall under, not 2 as previously thought:
a) 1900AC V1: all by itself. It has a restricted power table which doesn't allow changing of transmit power level, but it is in firmware and can be modified and reghack2 will do so and work just fine. Can be modified by other means also.
b) 1900AC V2, 1900ACS V1, 1200AC V1: has a power table that allows the transmit power and country code to be modified. Power table is in firmware and can be modified further using reghack2 or other means.
c) 1200AC V2, 1900ACS V2, and 3200ACM: power table is in eeprom and you can't modify anything, period.
Edit: Also, reghack doesn't simply change the two lines to:
struct ieee80211_reg_rule r2 = REG_RULE(2400, 2483, 40, 0, 30, 0);
struct ieee80211_reg_rule r5 = REG_RULE(5140, 5860, 160, 0, 30, 0);
it also removes any other restrictions preventing you from lowering the transmit power. So if you just apply it the way it is, it will allow you to change the transmit power from 4dBm to 30dBm on any channel and negate the limited power table on your mamba.

I applied reghack2 as is, it spits out the following log when ran:

Patching @ 0x00029974: core world8(?) regdomain in cfg80211/reg.o (new rule)
Patching @ 0x0002b2d8: embedded US regdomain in cfg80211/regdb.o (new rule)
Patching @ 0x00030d70: embedded 00 regdomain in cfg80211/regdb.o

after a reboot, iw reg get returns
global
country US: DFS-UNSET
(2400 - 2483 @ 40), (N/A, 30), (N/A)
(5140 - 5860 @ 160), (N/A, 30), (N/A)

However, the RSSI observed at phone follows the same pattern as before applying the patch. I did a quick test:

channel, txpower setting, rssi
149, 10, -36 dbm
149, 20, -36 dbm
36, 10, -54 dbm
36, 20, -55 dbm

furthermore, I tested with manual selection of dfs channel. The wifi ssid doesn't come up until about a minute after wifi reload, but logread has no logs about testing for radar interference that previously shows up when selecting a dfs channel.

This seems to indicate that the mwlwifi driver is doing things on its own - the patch on cfg80211.ko seems to be cosmetic only.

I found a commit back in 2015 in mwlwifi repo that switched from an ini file to reading dts power table couple days ago, I recall that when I was looking at the current master head of mwlwifi repo, it doesn't contain the changed files in that commit anymore. I will look and see whats up there.

(Last edited by kyzhk on 12 Aug 2017, 19:02)

Post #3031

starcms

12 Aug 2017, 19:06

oldchip wrote:

the page loading problem is still having problem i think because sw. when i add new vlan tag eth1 and wan. after creat new dhcp interface connect and reboot router, the page loading very slow. I disconect the new created interface, everything come back nomal. In my case, @starcms guide don't work.

Just curious, why would you need to make a vlan on the wan connection? eth1 and eth1.2 exist by default.

Also, just fyi, I have vlans on the LAN connection, and there is no problem, slowdown, etc, with dns lookups. That is with and without using dnscrypt.

Either you have a very special, unique case, and/or it's user error. If it is indeed because you have a very unique setup, post an issue on the LEDE forums and/or github so it can be looked at and resolved. @david cannot fix it.

(Last edited by starcms on 12 Aug 2017, 19:06)

Post #3032

starcms

12 Aug 2017, 19:11

kyzhk wrote:

I applied reghack2 as is, it spits out the following log when ran:
Patching @ 0x00029974: core world8(?) regdomain in cfg80211/reg.o (new rule)
Patching @ 0x0002b2d8: embedded US regdomain in cfg80211/regdb.o (new rule)
Patching @ 0x00030d70: embedded 00 regdomain in cfg80211/regdb.o
after a reboot, iw reg get returns
global
country US: DFS-UNSET
(2400 - 2483 @ 40), (N/A, 30), (N/A)
(5140 - 5860 @ 160), (N/A, 30), (N/A)
However, the RSSI observed at phone follows the same pattern as before applying the patch. I did a quick test:
channel, txpower setting, rssi
149, 10, -36 dbm
149, 20, -36 dbm
36, 10, -54 dbm
36, 20, -55 dbm
furthermore, I tested with manual selection of dfs channel. The wifi ssid doesn't come up until about a minute after wifi reload, but logread has no logs about testing for radar interference that previously shows up when selecting a dfs channel.
This seems to indicate that the mwlwifi driver is doing things on its own - the patch on cfg80211.ko seems to be cosmetic only.
I found a commit back in 2015 in mwlwifi repo that switched from an ini file to reading dts power table couple days ago, I recall that when I was looking at the current master head of mwlwifi repo, it doesn't contain the changed files in that commit anymore. I will look and see whats up there.

First, dealing with DFS channels, it is built into mwlwifi to pause for 1 minute before activating an AP on a DFS channel while it scans for interference. But as you said, with reghack2 applied, after 1 minute, the wifi ssid comes up and there are no log entries about testing for radar interference. Therefore you have disabled DFS. Unlike the normal operation where after less than an hour it will jump off a DFS channel and switch to channel 36, yours will now stay on whatever DFS channel you set indefinitely.

The output reghack2 gave you is correct.

If you really still cannot adjust your transmit power, then I have to say it is a unique case for the mamba. It works fine on my caimen (WRT1200AC V1) as tested with my phone and a wifi scanner software and the latest mwlwifi driver. The transmit power level most definitely has an effect.

(Last edited by starcms on 12 Aug 2017, 19:16)

Post #3033

kyzhk

12 Aug 2017, 19:16

https://github.com/kaloz/mwlwifi/blob/f … mwl_main.c

This is the snapshot of mwl_main.c right before the commit that made the driver read from dts rather than ini when loading the driver.

It used
static int mwl_load_tx_pwr_tbl(struct mwl_priv *priv, char *pwr_tbl)
to read the ini and load settings.

It became:
static void mwl_reg_notifier(struct wiphy *wiphy,
struct regulatory_request *request)

after the commit.

i.e. it used to be like this:
insmod mwlwifi.ko fw_name=88W8864.bin pwr_tbl=Mamba_FCC_v1.2_5G4TX.ini
after changing to read from dts it becomes this:
insmod mwlwifi.ko

This is the commit that did the switch.
https://github.com/kaloz/mwlwifi/commit … a92b5a976c

Add:
At the current master head, this function has been moved to
core.c

I will look more and edit this post as I go.

(Last edited by kyzhk on 12 Aug 2017, 19:30)

Post #3034

starcms

12 Aug 2017, 19:19

kyzhk wrote:

https://github.com/kaloz/mwlwifi/blob/f … mwl_main.c
This is the snapshot of mwl_main.c right before the commit that made the driver read from dts rather than ini when loading the driver.
i.e. it used to be like this:
insmod mwlwifi.ko fw_name=88W8864.bin pwr_tbl=Mamba_FCC_v1.2_5G4TX.ini
after changing to read from dts it becomes this:
insmod mwlwifi.ko
This is the commit that did the switch.
https://github.com/kaloz/mwlwifi/commit … a92b5a976c
I will look more and edit this post as I go.

The way mwlwifi is now, on the 1900AC V1, 1200AC V1, 1900AC V2, and 1900ACS V1, it reads from the power table which is stored in firmware that you had linked to earlier. That is where it gets all it's regulatory requirements from. You can edit that power table file to be whatever you want and recompile. The regulatory requirements are not in the mwlwifi driver, at least not anymore and haven't been for a long, long, long time.

(Last edited by starcms on 12 Aug 2017, 19:28)

Post #3035

kyzhk

12 Aug 2017, 19:21

starcms wrote:

kyzhk wrote:
https://github.com/kaloz/mwlwifi/blob/f … mwl_main.c
This is the snapshot of mwl_main.c right before the commit that made the driver read from dts rather than ini when loading the driver.
i.e. it used to be like this:
insmod mwlwifi.ko fw_name=88W8864.bin pwr_tbl=Mamba_FCC_v1.2_5G4TX.ini
after changing to read from dts it becomes this:
insmod mwlwifi.ko
This is the commit that did the switch.
https://github.com/kaloz/mwlwifi/commit … a92b5a976c
I will look more and edit this post as I go.
But on the 1900AC V1, 1200AC V1, 1900AC V2, and 1900ACS V2, it still reads from the power table which is stored in firmware that you had linked to earlier. That is where it gets all it's regulatory requirements from. You can edit that power table file to be whatever you want and recompile. The regulatory requirements are not in the mwlwifi driver!

Yes. I am only trying to determine the code path and see what it does currently at the master head.

I guess my current suspicion is on the fact that changing the channel still produces a noticeable dBm difference there, when the 5g low channels and high channels should have already been changed to 30 dBm max - which means switching between 36/149 channel should not have produced what I observed.

There are couple points I want to highlight:
1) The router is physically capable of changing tx power, as seen by changing channels between 36 and 149.
2) If the router is not respecting the dts power table setting (as both are changed now to max 30 dbm?), how does it determine the txpower for 36 and 149?

Driver info:
root@WRT1900ACv1:/sys/kernel/debug/ieee80211/phy1/mwlwifi# cat info

driver name: mwlwifi
chip type: 88W8864
hw version: 7
driver version: 10.3.4.0-20170606
firmware version: 0x0702091a
power table loaded from dts: yes
firmware region code: 0x0
mac address: 00:25:9c:13:19:0f
2g: disable
5g: enable
antenna: 4 4
irq number: 91
ap macid support: 0000ffff
sta macid support: 00010000
macid used: 00000001
radio: enable
iobase0: d1300000
iobase1: d1480000
tx limit: 768
rx limit: 64
qe trigger number: 1817815

curiously I have both 2.4g and 5g wifi enabled, but it says 2g: disable

edit:
this is logread for power table dump from presumably the wiphy_debug call

Sat Aug 12 11:52:45 2017 kern.debug kernel: [   10.144258] ieee80211 phy0: priv->pcmd_buf = ce404000  priv->pphys_cmd_buf = 0e404000
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   10.151886] ieee80211 phy0: fw download start
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   10.269581] ieee80211 phy0: FwSize = 118776 downloaded Size = 118776 curr_iteration 65510
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   11.132736] ieee80211 phy0: fw download complete
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   11.134097] ieee80211 phy0: pcmd = ce404000
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   11.203095] ieee80211 phy0: regdomain: FCC
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   11.203132] ieee80211 phy0: Channel: 1: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   11.203163] ieee80211 phy0: 17 11 11 11 10 10 10 10 f f f f 0 0 0 0
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   11.203180] ieee80211 phy0: Channel: 2: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   11.203208] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   11.203224] ieee80211 phy0: Channel: 3: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   11.203251] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   11.203267] ieee80211 phy0: Channel: 4: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   11.203295] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   11.203311] ieee80211 phy0: Channel: 5: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   11.203339] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   11.203354] ieee80211 phy0: Channel: 6: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   11.203382] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   11.203398] ieee80211 phy0: Channel: 7: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   11.203426] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   11.203480] ieee80211 phy0: Channel: 8: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   11.203557] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   11.203614] ieee80211 phy0: Channel: 9: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   11.203718] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   11.203753] ieee80211 phy0: Channel: 10: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   11.203837] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   11.203860] ieee80211 phy0: Channel: 11: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   11.203898] ieee80211 phy0: 17 11 11 11 11 11 11 11 10 10 10 10 0 0 0 0
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   11.228787] ieee80211 phy1: priv->pcmd_buf = ce630000  priv->pphys_cmd_buf = 0e630000
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   11.232715] ieee80211 phy1: fw download start
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   11.349752] ieee80211 phy1: FwSize = 118776 downloaded Size = 118776 curr_iteration 65501
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.222729] ieee80211 phy1: fw download complete
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.224062] ieee80211 phy1: pcmd = ce630000
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.295815] ieee80211 phy1: regdomain: FCC
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.295840] ieee80211 phy1: Channel: 36: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.295869] ieee80211 phy1: 8 8 8 8 8 8 8 8 9 9 9 9 9 9 9 9
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.295886] ieee80211 phy1: Channel: 40: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.295913] ieee80211 phy1: 8 8 8 8 9 9 9 9 9 9 9 9 9 9 9 9
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.295930] ieee80211 phy1: Channel: 44: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.295957] ieee80211 phy1: 8 8 8 8 9 9 9 9 9 9 9 9 9 9 9 9
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.295972] ieee80211 phy1: Channel: 48: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.295999] ieee80211 phy1: 8 8 8 8 9 9 9 9 9 9 9 9 9 9 9 9
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296015] ieee80211 phy1: Channel: 52: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296043] ieee80211 phy1: f f f f f f f f 12 12 12 12 12 12 12 12
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296059] ieee80211 phy1: Channel: 56: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296086] ieee80211 phy1: f f f f f f f f 12 12 12 12 12 12 12 12
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296102] ieee80211 phy1: Channel: 60: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296130] ieee80211 phy1: f f f f f f f f 12 12 12 12 12 12 12 12
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296146] ieee80211 phy1: Channel: 64: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296173] ieee80211 phy1: f f f f f f f f 12 12 12 12 12 12 12 12
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296189] ieee80211 phy1: Channel: 100: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296232] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296250] ieee80211 phy1: Channel: 104: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296278] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296294] ieee80211 phy1: Channel: 108: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296322] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296337] ieee80211 phy1: Channel: 112: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296365] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296381] ieee80211 phy1: Channel: 116: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296408] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296424] ieee80211 phy1: Channel: 120: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296452] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296467] ieee80211 phy1: Channel: 124: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296495] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296511] ieee80211 phy1: Channel: 128: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296538] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296554] ieee80211 phy1: Channel: 132: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296582] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296597] ieee80211 phy1: Channel: 136: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296625] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296641] ieee80211 phy1: Channel: 140: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296668] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296684] ieee80211 phy1: Channel: 149: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296712] ieee80211 phy1: 16 16 16 16 14 14 14 14 15 15 15 15 14 14 14 14
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296728] ieee80211 phy1: Channel: 153: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296755] ieee80211 phy1: 15 15 15 15 15 15 15 15 15 15 15 15 14 14 14 14
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296771] ieee80211 phy1: Channel: 157: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296799] ieee80211 phy1: 15 15 15 15 15 15 15 15 15 15 15 15 14 14 14 14
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296814] ieee80211 phy1: Channel: 161: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296842] ieee80211 phy1: 15 15 15 15 15 15 15 15 15 15 15 15 14 14 14 14
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296858] ieee80211 phy1: Channel: 165: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296885] ieee80211 phy1: 16 16 16 16 16 16 16 16 15 15 15 15 14 14 14 14
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179050] ieee80211 phy1: regdomain: FCC
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179103] ieee80211 phy1: Channel: 36: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179128] ieee80211 phy1: 8 8 8 8 8 8 8 8 9 9 9 9 9 9 9 9
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179137] ieee80211 phy1: Channel: 40: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179202] ieee80211 phy1: 8 8 8 8 9 9 9 9 9 9 9 9 9 9 9 9
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179224] ieee80211 phy1: Channel: 44: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179237] ieee80211 phy1: 8 8 8 8 9 9 9 9 9 9 9 9 9 9 9 9
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179246] ieee80211 phy1: Channel: 48: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179260] ieee80211 phy1: 8 8 8 8 9 9 9 9 9 9 9 9 9 9 9 9
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179269] ieee80211 phy1: Channel: 52: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179284] ieee80211 phy1: f f f f f f f f 12 12 12 12 12 12 12 12
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179292] ieee80211 phy1: Channel: 56: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179307] ieee80211 phy1: f f f f f f f f 12 12 12 12 12 12 12 12
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179349] ieee80211 phy1: Channel: 60: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179389] ieee80211 phy1: f f f f f f f f 12 12 12 12 12 12 12 12
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179403] ieee80211 phy1: Channel: 64: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179418] ieee80211 phy1: f f f f f f f f 12 12 12 12 12 12 12 12
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179426] ieee80211 phy1: Channel: 100: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179441] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179450] ieee80211 phy1: Channel: 104: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179465] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179473] ieee80211 phy1: Channel: 108: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179488] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179496] ieee80211 phy1: Channel: 112: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179511] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179520] ieee80211 phy1: Channel: 116: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179534] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179543] ieee80211 phy1: Channel: 120: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179557] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179565] ieee80211 phy1: Channel: 124: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179580] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179589] ieee80211 phy1: Channel: 128: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179603] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179611] ieee80211 phy1: Channel: 132: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179626] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179634] ieee80211 phy1: Channel: 136: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179664] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179672] ieee80211 phy1: Channel: 140: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179745] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179795] ieee80211 phy1: Channel: 149: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179832] ieee80211 phy1: 16 16 16 16 14 14 14 14 15 15 15 15 14 14 14 14
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179845] ieee80211 phy1: Channel: 153: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179894] ieee80211 phy1: 15 15 15 15 15 15 15 15 15 15 15 15 14 14 14 14
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.179934] ieee80211 phy1: Channel: 157: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.180002] ieee80211 phy1: 15 15 15 15 15 15 15 15 15 15 15 15 14 14 14 14
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.180055] ieee80211 phy1: Channel: 161: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.180114] ieee80211 phy1: 15 15 15 15 15 15 15 15 15 15 15 15 14 14 14 14
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.180150] ieee80211 phy1: Channel: 165: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.180213] ieee80211 phy1: 16 16 16 16 16 16 16 16 15 15 15 15 14 14 14 14
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.180283] ieee80211 phy0: regdomain: FCC
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.180302] ieee80211 phy0: Channel: 1: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.180322] ieee80211 phy0: 17 11 11 11 10 10 10 10 f f f f 0 0 0 0
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.180371] ieee80211 phy0: Channel: 2: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.180404] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.180416] ieee80211 phy0: Channel: 3: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.180430] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.180467] ieee80211 phy0: Channel: 4: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.180499] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.180510] ieee80211 phy0: Channel: 5: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.180525] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.180533] ieee80211 phy0: Channel: 6: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.180547] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.180555] ieee80211 phy0: Channel: 7: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.180570] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.180579] ieee80211 phy0: Channel: 8: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.180593] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.180601] ieee80211 phy0: Channel: 9: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.180616] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.180624] ieee80211 phy0: Channel: 10: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.180638] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.180646] ieee80211 phy0: Channel: 11: 0x0 0x0 0xf
Sat Aug 12 11:52:49 2017 kern.debug kernel: [   17.180661] ieee80211 phy0: 17 11 11 11 11 11 11 11 10 10 10 10 0 0 0 0

(Last edited by kyzhk on 12 Aug 2017, 19:57)

Post #3036

davidc502

12 Aug 2017, 19:34

Wifi driver for kernel 4.4.79 is ready now.

Copying the post from above into this one so they are both together.

Driver version 10.3.4.0-20170810, which includes disabling amsdu is now ready. I will start building the wifi driver for those on kernel 4.4.x, and should be ready this afternoon.

The driver below is for those who are using the latest build with Kernel 4.9.40.
https://davidc502sis.dynamic-dns.net/re … _vfpv3.ipk

Driver below is for those who are using the latest build with Kernel 4.4.79.
https://davidc502sis.dynamic-dns.net/re … _vfpv3.ipk

*EDIT

Since it looks like we will be updating the driver more often it might be worth while to update the website to include the latest driver for a particular build.

(Last edited by davidc502 on 12 Aug 2017, 19:36)

Post #3037

starcms

12 Aug 2017, 19:39

davidc502 wrote:

*EDIT
Since it looks like we will be updating the driver more often it might be worth while to update the website to include the latest driver for a particular build.

Sounds like a great idea! Feel free to grab my instructions from the post under your post with the updated driver for kernel 4.9 and put on the website along with it.

Post #3038

starcms

12 Aug 2017, 19:50

kyzhk wrote:

Yes. I am only trying to determine the code path and see what it does currently at the master head.
I guess my current suspicion is on the fact that changing the channel still produces a noticeable dBm difference there, when the 5g low channels and high channels should have already been changed to 30 dBm max - which means switching between 36/149 channel should not have produced what I observed.
There are couple points I want to highlight:
1) The router is physically capable of changing tx power, as seen by changing channels between 36 and 149.
2) If the router is not respecting the dts power table setting (as both are changed now to max 30 dbm?), how does it determine the txpower for 36 and 149?
Driver info:
root@WRT1900ACv1:/sys/kernel/debug/ieee80211/phy1/mwlwifi# cat info
driver name: mwlwifi
chip type: 88W8864
hw version: 7
driver version: 10.3.4.0-20170606
firmware version: 0x0702091a
power table loaded from dts: yes
firmware region code: 0x0
mac address: 00:25:9c:13:19:0f
2g: disable
5g: enable
antenna: 4 4
irq number: 91
ap macid support: 0000ffff
sta macid support: 00010000
macid used: 00000001
radio: enable
iobase0: d1300000
iobase1: d1480000
tx limit: 768
rx limit: 64
qe trigger number: 1817815
curiously I have both 2.4g and 5g wifi enabled, but it says 2g: disable

Easy part first. It says 2g disable because you are in the phy1 directory. If you go to phy0 directory, you'll see the opposite

On all other models, phy0 is 5ghz and phy1 is 2.4ghz. But I believe on the mamba, phy0 is 2.4ghz and phy1 is 5ghz (which seems to be confirmed by your results above).

Also "power table loaded from dts: yes" means the power table is being loaded from firmware, not the eeprom chip of the newer models (1200V2, 1900ACS V2, 3200ACM).

As for why the mamba isn't respecting the changes made to the power table by reghack2, I don't understand. As I said before, it works fine on my caimen and has been tested. I'm not just changing the transmit power number and assuming it works. I have tested it.

I don't understand why it doesn't work for you on the mamba.

Edit: You did reboot after applying reghack2, correct?

Nm, you already said you did.

Edit2: Why are you running an old driver version?

(Last edited by starcms on 12 Aug 2017, 20:04)

Post #3039

kyzhk

12 Aug 2017, 20:04

starcms wrote:

kyzhk wrote:
Yes. I am only trying to determine the code path and see what it does currently at the master head.
I guess my current suspicion is on the fact that changing the channel still produces a noticeable dBm difference there, when the 5g low channels and high channels should have already been changed to 30 dBm max - which means switching between 36/149 channel should not have produced what I observed.
There are couple points I want to highlight:
1) The router is physically capable of changing tx power, as seen by changing channels between 36 and 149.
2) If the router is not respecting the dts power table setting (as both are changed now to max 30 dbm?), how does it determine the txpower for 36 and 149?
Driver info:
root@WRT1900ACv1:/sys/kernel/debug/ieee80211/phy1/mwlwifi# cat info
driver name: mwlwifi
chip type: 88W8864
hw version: 7
driver version: 10.3.4.0-20170606
firmware version: 0x0702091a
power table loaded from dts: yes
firmware region code: 0x0
mac address: 00:25:9c:13:19:0f
2g: disable
5g: enable
antenna: 4 4
irq number: 91
ap macid support: 0000ffff
sta macid support: 00010000
macid used: 00000001
radio: enable
iobase0: d1300000
iobase1: d1480000
tx limit: 768
rx limit: 64
qe trigger number: 1817815
curiously I have both 2.4g and 5g wifi enabled, but it says 2g: disable
Easy part first. It says 2g disable because you are in the phy1 directory. If you go to phy0 directory, you'll see the opposite
On all other models, phy0 is 5ghz and phy1 is 2.4ghz. But I believe on the mamba, phy0 is 2.4ghz and phy1 is 5ghz (which seems to be confirmed by your results above).
Also "power table loaded from dts: yes" means the power table is being loaded from firmware, not the eeprom chip of the newer models (1200V2, 1900ACS V2, 3200ACM).
As for why the mamba isn't respecting the changes made to the power table by reghack2, I don't understand. As I said before, it works fine on my caimen and has been tested. I'm not just changing the transmit power number and assuming it works. I have tested it.
I don't understand why it doesn't work for you on the mamba.
Edit: You did reboot after applying reghack2, correct?
Nm, you already said you did.

Hey starcms, I posted the power table dump from logread in an earlier post. examples:

Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.295840] ieee80211 phy1: Channel: 36: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.295869] ieee80211 phy1: 8 8 8 8 8 8 8 8 9 9 9 9 9 9 9 9

Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296684] ieee80211 phy1: Channel: 149: 0x0 0x0 0xf
Sat Aug 12 11:52:45 2017 kern.debug kernel: [   12.296712] ieee80211 phy1: 16 16 16 16 14 14 14 14 15 15 15 15 14 14 14 14

This dump seems to be from mwlwifi core.c, line 329-354

            /* Dump loaded power tabel */
            wiphy_debug(hw->wiphy, "regdomain: %s\n", prop->name);
            for (i = 0; i < SYSADPT_MAX_NUM_CHANNELS; i++) {
                struct mwl_tx_pwr_tbl *pwr_tbl;
                char disp_buf[64];
                char *disp_ptr;

                pwr_tbl = &priv->tx_pwr_tbl[i];
                if (pwr_tbl->channel == 0)
                    break;
                wiphy_debug(hw->wiphy,
                        "Channel: %d: 0x%x 0x%x 0x%x\n",
                        pwr_tbl->channel,
                        pwr_tbl->setcap,
                        pwr_tbl->cdd,
                        pwr_tbl->txantenna2);
                disp_ptr = disp_buf;
                for (j = 0; j < SYSADPT_TX_POWER_LEVEL_TOTAL;
                     j++) {
                    disp_ptr +=
                        sprintf(disp_ptr, "%x ",
                            pwr_tbl->tx_power[j]);
                }
                wiphy_debug(hw->wiphy, "%s\n", disp_buf);
}

So this power table should match exactly what the driver actually read. If all 5ghz channel are set to 30 dbm max, shouldn't we see matching numbers for channel 36 and channel 149 in the power table dump?

Do you mind reloading your mwlwifi driver / rebooting your wrt and post the power table dump? Thanks

edit:
Let me check my sanity by reverting the patched cfg80211.ko, reboot, and check the power table dump again.

power table dump with original cfg80211.ko:

Sat Aug 12 11:52:57 2017 kern.debug kernel: [   10.902219] ieee80211 phy0: priv->pcmd_buf = ce07c000  priv->pphys_cmd_buf = 0e07c000
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   10.910944] ieee80211 phy0: fw download start
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   11.029472] ieee80211 phy0: FwSize = 118776 downloaded Size = 118776 curr_iteration 65510
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   11.892682] ieee80211 phy0: fw download complete
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   11.894028] ieee80211 phy0: pcmd = ce07c000
Sat Aug 12 11:52:57 2017 kern.info kernel: [   11.911919] ieee80211 phy0: firmware version: 0x702091a
Sat Aug 12 11:52:57 2017 kern.info kernel: [   11.935698] ieee80211 phy0: firmware region code: 0
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   11.963013] ieee80211 phy0: regdomain: FCC
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   11.963061] ieee80211 phy0: Channel: 1: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   11.963092] ieee80211 phy0: 17 11 11 11 10 10 10 10 f f f f 0 0 0 0
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   11.963109] ieee80211 phy0: Channel: 2: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   11.963136] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   11.963152] ieee80211 phy0: Channel: 3: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   11.963192] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   11.963216] ieee80211 phy0: Channel: 4: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   11.963284] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   11.963306] ieee80211 phy0: Channel: 5: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   11.963334] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   11.963350] ieee80211 phy0: Channel: 6: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   11.963389] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   11.963412] ieee80211 phy0: Channel: 7: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   11.963483] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   11.963506] ieee80211 phy0: Channel: 8: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   11.963534] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   11.963549] ieee80211 phy0: Channel: 9: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   11.963578] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   11.963610] ieee80211 phy0: Channel: 10: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   11.963657] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   11.963687] ieee80211 phy0: Channel: 11: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   11.963716] ieee80211 phy0: 17 11 11 11 11 11 11 11 10 10 10 10 0 0 0 0
Sat Aug 12 11:52:57 2017 kern.info kernel: [   11.965913] ieee80211 phy0: 2G enabled, 5G disabled
Sat Aug 12 11:52:57 2017 kern.info kernel: [   11.970851] ieee80211 phy0: 4 TX antennas, 4 RX antennas
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   11.988533] ieee80211 phy1: priv->pcmd_buf = ce640000  priv->pphys_cmd_buf = 0e640000
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   11.992671] ieee80211 phy1: fw download start
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   12.109767] ieee80211 phy1: FwSize = 118776 downloaded Size = 118776 curr_iteration 65501
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   12.982692] ieee80211 phy1: fw download complete
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   12.984042] ieee80211 phy1: pcmd = ce640000
Sat Aug 12 11:52:57 2017 kern.info kernel: [   13.001920] ieee80211 phy1: firmware version: 0x702091a
Sat Aug 12 11:52:57 2017 kern.info kernel: [   13.025706] ieee80211 phy1: firmware region code: 0
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.055561] ieee80211 phy1: regdomain: FCC
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.055588] ieee80211 phy1: Channel: 36: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.055616] ieee80211 phy1: 8 8 8 8 8 8 8 8 9 9 9 9 9 9 9 9
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.055632] ieee80211 phy1: Channel: 40: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.055659] ieee80211 phy1: 8 8 8 8 9 9 9 9 9 9 9 9 9 9 9 9
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.055675] ieee80211 phy1: Channel: 44: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.055701] ieee80211 phy1: 8 8 8 8 9 9 9 9 9 9 9 9 9 9 9 9
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.055717] ieee80211 phy1: Channel: 48: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.055743] ieee80211 phy1: 8 8 8 8 9 9 9 9 9 9 9 9 9 9 9 9
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.055758] ieee80211 phy1: Channel: 52: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.055786] ieee80211 phy1: f f f f f f f f 12 12 12 12 12 12 12 12
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.055802] ieee80211 phy1: Channel: 56: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.055830] ieee80211 phy1: f f f f f f f f 12 12 12 12 12 12 12 12
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.055846] ieee80211 phy1: Channel: 60: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.055873] ieee80211 phy1: f f f f f f f f 12 12 12 12 12 12 12 12
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.055889] ieee80211 phy1: Channel: 64: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.055916] ieee80211 phy1: f f f f f f f f 12 12 12 12 12 12 12 12
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.055933] ieee80211 phy1: Channel: 100: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.055975] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.055994] ieee80211 phy1: Channel: 104: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.056022] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.056037] ieee80211 phy1: Channel: 108: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.056065] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.056081] ieee80211 phy1: Channel: 112: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.056108] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.056124] ieee80211 phy1: Channel: 116: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.056152] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.056168] ieee80211 phy1: Channel: 120: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.056195] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.056211] ieee80211 phy1: Channel: 124: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.056239] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.056254] ieee80211 phy1: Channel: 128: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.056282] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.056298] ieee80211 phy1: Channel: 132: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.056326] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.056341] ieee80211 phy1: Channel: 136: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.056369] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.056385] ieee80211 phy1: Channel: 140: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.056412] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.056428] ieee80211 phy1: Channel: 149: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.056456] ieee80211 phy1: 16 16 16 16 14 14 14 14 15 15 15 15 14 14 14 14
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.056471] ieee80211 phy1: Channel: 153: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.056499] ieee80211 phy1: 15 15 15 15 15 15 15 15 15 15 15 15 14 14 14 14
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.056515] ieee80211 phy1: Channel: 157: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.056543] ieee80211 phy1: 15 15 15 15 15 15 15 15 15 15 15 15 14 14 14 14
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.056558] ieee80211 phy1: Channel: 161: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.056586] ieee80211 phy1: 15 15 15 15 15 15 15 15 15 15 15 15 14 14 14 14
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.056602] ieee80211 phy1: Channel: 165: 0x0 0x0 0xf
Sat Aug 12 11:52:57 2017 kern.debug kernel: [   13.056629] ieee80211 phy1: 16 16 16 16 16 16 16 16 15 15 15 15 14 14 14 14
Sat Aug 12 11:52:57 2017 kern.info kernel: [   13.058673] ieee80211 phy1: 2G disabled, 5G enabled
Sat Aug 12 11:52:57 2017 kern.info kernel: [   13.063666] ieee80211 phy1: 4 TX antennas, 4 RX antennas
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743124] ieee80211 phy1: regdomain: FCC
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743143] ieee80211 phy1: Channel: 36: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743159] ieee80211 phy1: 8 8 8 8 8 8 8 8 9 9 9 9 9 9 9 9
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743167] ieee80211 phy1: Channel: 40: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743181] ieee80211 phy1: 8 8 8 8 9 9 9 9 9 9 9 9 9 9 9 9
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743189] ieee80211 phy1: Channel: 44: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743202] ieee80211 phy1: 8 8 8 8 9 9 9 9 9 9 9 9 9 9 9 9
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743210] ieee80211 phy1: Channel: 48: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743224] ieee80211 phy1: 8 8 8 8 9 9 9 9 9 9 9 9 9 9 9 9
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743232] ieee80211 phy1: Channel: 52: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743246] ieee80211 phy1: f f f f f f f f 12 12 12 12 12 12 12 12
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743254] ieee80211 phy1: Channel: 56: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743268] ieee80211 phy1: f f f f f f f f 12 12 12 12 12 12 12 12
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743276] ieee80211 phy1: Channel: 60: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743290] ieee80211 phy1: f f f f f f f f 12 12 12 12 12 12 12 12
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743298] ieee80211 phy1: Channel: 64: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743311] ieee80211 phy1: f f f f f f f f 12 12 12 12 12 12 12 12
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743320] ieee80211 phy1: Channel: 100: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743334] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743342] ieee80211 phy1: Channel: 104: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743356] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743365] ieee80211 phy1: Channel: 108: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743379] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743387] ieee80211 phy1: Channel: 112: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743401] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743409] ieee80211 phy1: Channel: 116: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743423] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743432] ieee80211 phy1: Channel: 120: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743446] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743454] ieee80211 phy1: Channel: 124: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743468] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743476] ieee80211 phy1: Channel: 128: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743491] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743499] ieee80211 phy1: Channel: 132: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743513] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743521] ieee80211 phy1: Channel: 136: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743535] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743544] ieee80211 phy1: Channel: 140: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743558] ieee80211 phy1: 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743566] ieee80211 phy1: Channel: 149: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743580] ieee80211 phy1: 16 16 16 16 14 14 14 14 15 15 15 15 14 14 14 14
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743588] ieee80211 phy1: Channel: 153: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743602] ieee80211 phy1: 15 15 15 15 15 15 15 15 15 15 15 15 14 14 14 14
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743611] ieee80211 phy1: Channel: 157: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743625] ieee80211 phy1: 15 15 15 15 15 15 15 15 15 15 15 15 14 14 14 14
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743633] ieee80211 phy1: Channel: 161: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743647] ieee80211 phy1: 15 15 15 15 15 15 15 15 15 15 15 15 14 14 14 14
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743655] ieee80211 phy1: Channel: 165: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743670] ieee80211 phy1: 16 16 16 16 16 16 16 16 15 15 15 15 14 14 14 14
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743694] ieee80211 phy0: regdomain: FCC
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743703] ieee80211 phy0: Channel: 1: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743717] ieee80211 phy0: 17 11 11 11 10 10 10 10 f f f f 0 0 0 0
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743725] ieee80211 phy0: Channel: 2: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743739] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743747] ieee80211 phy0: Channel: 3: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743761] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743769] ieee80211 phy0: Channel: 4: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743783] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743791] ieee80211 phy0: Channel: 5: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743805] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743813] ieee80211 phy0: Channel: 6: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743827] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743835] ieee80211 phy0: Channel: 7: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743849] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743857] ieee80211 phy0: Channel: 8: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743871] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743879] ieee80211 phy0: Channel: 9: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743893] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743901] ieee80211 phy0: Channel: 10: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743915] ieee80211 phy0: 17 16 16 16 16 16 16 14 10 10 10 10 0 0 0 0
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743924] ieee80211 phy0: Channel: 11: 0x0 0x0 0xf
Sat Aug 12 11:53:00 2017 kern.debug kernel: [   17.743938] ieee80211 phy0: 17 11 11 11 11 11 11 11 10 10 10 10 0 0 0 0
Sat Aug 12 11:53:01 2017 kern.debug kernel: [   19.321853] ieee80211 phy0: change: 0xffffffff
Sat Aug 12 11:53:02 2017 kern.debug kernel: [   19.627643] ieee80211 phy1: change: 0xffffffff
Sat Aug 12 11:53:02 2017 kern.debug kernel: [   19.764694] ieee80211 phy0: change: 0x100
Sat Aug 12 11:53:02 2017 kern.debug kernel: [   19.772639] ieee80211 phy0: change: 0x42
Sat Aug 12 11:53:02 2017 kern.debug kernel: [   20.199763] ieee80211 phy0: change: 0x20
Sat Aug 12 11:53:02 2017 kern.debug kernel: [   20.256451] ieee80211 phy1: change: 0x100
Sat Aug 12 11:53:02 2017 kern.debug kernel: [   20.265832] ieee80211 phy1: change: 0x42
Sat Aug 12 11:53:03 2017 kern.debug kernel: [   20.590275] ieee80211 phy1: change: 0x20

The power table dump seems to be the same for original cfg80211.ko vs patched version.
Somehow patching cfg80211.ko doesn't seem to affect the power table read by mwlwifi driver on my wrt1900acv1...

(Last edited by kyzhk on 12 Aug 2017, 20:15)

Post #3040

davidc502

12 Aug 2017, 20:42

starcms wrote:

davidc502 wrote:
*EDIT
Since it looks like we will be updating the driver more often it might be worth while to update the website to include the latest driver for a particular build.
Sounds like a great idea! Feel free to grab my instructions from the post under your post with the updated driver for kernel 4.9 and put on the website along with it.

Question: In the instructions should the switch be --force-reinstall ?

Post #3041

oldchip

12 Aug 2017, 21:05

starcms wrote:

oldchip wrote:
the page loading problem is still having problem i think because sw. when i add new vlan tag eth1 and wan. after creat new dhcp interface connect and reboot router, the page loading very slow. I disconect the new created interface, everything come back nomal. In my case, @starcms guide don't work.
Just curious, why would you need to make a vlan on the wan connection? eth1 and eth1.2 exist by default.
Also, just fyi, I have vlans on the LAN connection, and there is no problem, slowdown, etc, with dns lookups. That is with and without using dnscrypt.
Either you have a very special, unique case, and/or it's user error. If it is indeed because you have a very unique setup, post an issue on the LEDE forums and/or github so it can be looked at and resolved. @david cannot fix it.

@Starcms. My main isp has two vlan one for internet and one for iptv, and other have not. I use mwan3 too. But the problem is solved. when i comeback linksys firm, i reset by manual to clear all setting, come back david firm again, everything be ok. my command to update firm is:
fw_setenv boot_part 1 or 2
mtd -r write <linksysfirmware.img> kernel 2 or kenel 1 ( should be 2 to avoid brick if something wrong)
reupdate linksys firm again ( it should be on partion 1) and factory reset.
To keep linksys firm on partion 1. we could use this command when update openwrt
mtd -r write <openwrt_firmware.img or .bin> kernel 2
best regard,
P/s so sad to know that the new wifi driver is not effect with 1900acs v2.

Post #3042

bill1228

12 Aug 2017, 22:59

Running Lede Reboot SNAPSHOT r4651-a6f6f8d / LuCI Master (git-17.209.51293-4c9ae3f) from download site dated 8/12 on my WRT3200. I then installed the driver version 10.3.4.0-20170810, which includes disabling amsdu update. Has been running fine on 5GHz wireless until just now. Was browsing web on wireless attached PC and suddenly lost the 5GHz connection. Checked a second PC that had also lost the 5GHz wireless connection. A couple of minutes later, before I could get to a wire attached PC the link came back up on both PC's. Looking at the up time on the wrt3200 I would say the router rebooted for some reason. Of course that means any logs were lost.

Before someone asks, yes I rebooted the router after installing the driver update.

Not off to a good start here. Since I was running the r4651 from the earlier date with the then current patched files without any issue, it looks like to me there is either something up with the new wireless drivers or the build on the web site. Guessing the r4651 posted with today's date is just the same build with the new drivers. Did not check before I installed the ampdu patch file.

(Last edited by bill1228 on 12 Aug 2017, 23:02)

Post #3043

sean.p.hopkins

13 Aug 2017, 01:13

starcms wrote:

a) You must be running @david's latest build (r4651) with kernel 4.9 (this is everyone but those with a mamba, 1900AC V1)

Pardon my ignorance as I am not sure how these things work, but I am running r4576 and it feels stable on my 3200ACM. Is it worthwhile to upgrade to the new release? Or do versions of the driver get built for the other releases or only for the current release? Other than power issues (damn t-storms!) I do not detect any issues with my install (and am not a gamer).

Edit: I missed the post right above mine about the driver. Will investigate.

(Last edited by sean.p.hopkins on 13 Aug 2017, 02:45)

Post #3044

davidc502

13 Aug 2017, 02:36

bill1228 wrote:

Running Lede Reboot SNAPSHOT r4651-a6f6f8d / LuCI Master (git-17.209.51293-4c9ae3f) from download site dated 8/12 on my WRT3200. I then installed the driver version 10.3.4.0-20170810, which includes disabling amsdu update. Has been running fine on 5GHz wireless until just now. Was browsing web on wireless attached PC and suddenly lost the 5GHz connection. Checked a second PC that had also lost the 5GHz wireless connection. A couple of minutes later, before I could get to a wire attached PC the link came back up on both PC's. Looking at the up time on the wrt3200 I would say the router rebooted for some reason. Of course that means any logs were lost.
Before someone asks, yes I rebooted the router after installing the driver update.
Not off to a good start here. Since I was running the r4651 from the earlier date with the then current patched files without any issue, it looks like to me there is either something up with the new wireless drivers or the build on the web site. Guessing the r4651 posted with today's date is just the same build with the new drivers. Did not check before I installed the ampdu patch file.

Thanks for the report.

Will probably continue to run the 1900acs for the next few days, and then will switch back to the 3200. Hopefully we have a few more reports from 3200acm owners.

Post #3045

bill1228

13 Aug 2017, 02:50

sean.p.hopkins wrote:

starcms wrote:
a) You must be running @david's latest build (r4651) with kernel 4.9 (this is everyone but those with a mamba, 1900AC V1)
Pardon my ignorance as I am not sure how these things work, but I am running r4576 and it feels stable on my 3200ACM. Is it worthwhile to upgrade to the new release? Or do versions of the driver get built for the other releases or only for the current release? Other than power issues (damn t-storms!) I do not detect any issues with my install (and am not a gamer).

Some folks want to always be on the bleeding edge and others subscribe to the 'if it ain't broke don't fix it". Think you fit the later group. Running well no issues no real need to upgrade if you do not want to, in my opinion.

Post #3046

bill1228

13 Aug 2017, 02:54

davidc502 wrote:

bill1228 wrote:
Running Lede Reboot SNAPSHOT r4651-a6f6f8d / LuCI Master (git-17.209.51293-4c9ae3f) from download site dated 8/12 on my WRT3200. I then installed the driver version 10.3.4.0-20170810, which includes disabling amsdu update. Has been running fine on 5GHz wireless until just now. Was browsing web on wireless attached PC and suddenly lost the 5GHz connection. Checked a second PC that had also lost the 5GHz wireless connection. A couple of minutes later, before I could get to a wire attached PC the link came back up on both PC's. Looking at the up time on the wrt3200 I would say the router rebooted for some reason. Of course that means any logs were lost.
Before someone asks, yes I rebooted the router after installing the driver update.
Not off to a good start here. Since I was running the r4651 from the earlier date with the then current patched files without any issue, it looks like to me there is either something up with the new wireless drivers or the build on the web site. Guessing the r4651 posted with today's date is just the same build with the new drivers. Did not check before I installed the ampdu patch file.
Thanks for the report.
Will probably continue to run the 1900acs for the next few days, and then will switch back to the 3200. Hopefully we have a few more reports from 3200acm owners.

@david
Now up for over 4 hours without anymore reboots. Maybe a fluke? First reboot was after only about 20 minutes of run time. Will update if it happens again..

Post #3047

sean.p.hopkins

13 Aug 2017, 03:13

bill1228 wrote:

Some folks want to always be on the bleeding edge and others subscribe to the 'if it ain't broke don't fix it". Think you fit the later group. Running well no issues no real need to upgrade if you do not want to, in my opinion.

Thanks Bill. Seeing as how I am running an open-source firmware instead of safe stock firmware I would say that I am getting close to the bleeding edge but I agree not on the complete bleeding edge ;-) My family can sometimes not be impressed with what I do to "their" internet... and are not as tolerant of downtime!

Post #3048

bill1228

13 Aug 2017, 04:04

sean.p.hopkins wrote:

bill1228 wrote:
Some folks want to always be on the bleeding edge and others subscribe to the 'if it ain't broke don't fix it". Think you fit the later group. Running well no issues no real need to upgrade if you do not want to, in my opinion.
Thanks Bill. Seeing as how I am running an open-source firmware instead of safe stock firmware I would say that I am getting close to the bleeding edge but I agree not on the complete bleeding edge ;-) My family can sometimes not be impressed with what I do to "their" internet... and are not as tolerant of downtime!

Understand well. Wife and teen girl that lives with us think that Internet is a right not a privilege and do not understand any downtime.
Anymore I think most of folks playing with routers fit into this situation with family. .

Post #3049

dktshudy

13 Aug 2017, 09:12

Running 4651 with the patched mwlwifi driver on a 1900ACS and so far I'm very pleased with the performance and stability.

Thank you David for your time and effort.

If we are going to be updating the wifi driver, download links and installation steps would be a good idea. Also include a step to verify the driver version. lsmod/modinfo don't seem to display version number.

Post #3050

floydburgermcdahm

13 Aug 2017, 11:31

dktshudy wrote:

If we are going to be updating the wifi driver, download links and installation steps would be a good idea. Also include a step to verify the driver version. lsmod/modinfo don't seem to display version number.

opkg list-installed | grep mwlwifi

Page 122 of 229