Hi Matthias,
> Any news here? I know a lot of people who are eager to run OpenWrt on
> their UAP Outdoor+ ...
oh well so nearly two months have passed
I must admit no news. I am/was planning to run the code in ARMSim to analyze
what is sent to the filter and from where.
Still want to do th
Hi Kirill,
On Dienstag, 19. Mai 2015 15:11:31 Kirill Berezin wrote:
> I think that they tune an external receiver via some sort of SPI bus (GPIO
> pins are 5,6,7,8). Write procedure is quite simple, but I can't still get
> how enable works (for sure it sends a disable); it is quite possible that
Hi,
> -Trapping ar5416GpioSet with some hand crafted assembly code that traces
> calls and logs them somewhere
this works. I've trapped ar5416GpioSet, putting my code into space occupied by
the unused function ar5416GpioCfgInput. ar5416GpioGet is work to do. I can
send the code, but it will req
Hi Kirill,
> Updated version is in the attachment. I tried it on a real device, but I
> can't say that the problem was solved, but something is definitely happens.
>
> Unfortunately I couldn't find where a bunch of zeros were generated, so
> I skipped this part ))) (I tried a version with zeros,
Hi,
> Updated version is in the attachment. I tried it on a real device, but I
> can't say that the problem was solved, but something is definitely happens.
related question: After installing OpenWRT on the access point, do you know
whether I can revert using the instructions from:
https://comm
Hi Kirill,
> Updated version is in the attachment. I tried it on a real device, but I
> can't say that the problem was solved, but something is definitely happens.
which OpenWrt image did you use? I flashed openwrt-15.05-rc1-ar71xx-generic-
ubnt-unifi-outdoor-plus-squashfs-factory.bin and did not
Hi,
> It seems that ath9k does not even export its pins to the GPIO
> subsystem.
ok, here we go. For the brave ones who know their pci wifi chip is at
0x1000 and don't mind bit banging io registers.
Surely still all usleep() timers wrong and full of unknown side effects.
But tunes the hsr
Hi Kirill,
> So I made a patch for atheros driver and it works. A signal level on
> different frequencies is more or less equal and seems adequate (about
> -30 dbm if a client within a meter from ap).
so you reverse engineered the timer values, the response format and the 's'
command I had plann
Hi Kirill,
> > An Unifi specific user space daemon listens to scan and channel change
> > events via netlink (will check next weekend if channel change is
> > broadcasted reliably) and tune the filter via sysfs interface
> > accordingly like hsr.c tried.
>
> I like this idea because in this case
+ return 0;
+}
+
+static void hsr_tune(struct ath_hw* ah, int bw, int fq) {
+ static int initialized;
+
+ if (!initialized) {
+ initialized = 1;
+ hsr_init(ah);
+ }
+ hsr_enable(ah, bw, fq);
+ hsr_status(ah);
+}
+
+
+
+static int __init hsr_mod_init(void)
+{
+ rtnl_lock(); /* Should lock a
ialized) {
++ initialized = 1;
++ hsr_init(ah);
++ }
++
++ if (!hsr_enable(ah, bw, fq)) {
++ hsr_status(ah);
++ last_bw = bw;
++ last_fq = fq;
++ } else {
++ /* Tuning failed - make sure that we try again */
++
Hi Kirill, hi Arjen,
thanks for your feedback. I'll create a new version of the patch with your
updates and suggestions.
I will also try to move the HSR tuner module into it's own subdirectory below
package/kernel to minimize the mac80211 patch that Felix will have to
maintain.
> > I also tr
Hi,
this patchset adds support for the HSR filter of the Ubiquiti Unifi Outdoor
Plus access point.
The HSR is a configurable RF filter in the receive path that must be tuned
according to the selected Wifi channel for the access point to work.
Patch 1 adds a "channel change helper" callback to
Add a "channel change helper" callback to the mac80211 package and
export the ath9k kernel API for other packages.
Signed-off-by: Stefan Rompf
---
Index: package/kernel/mac80211/Makefile
===
--- package/kernel/mac8021
Add kmod-ubnt-hsr package that tunes the HSR filter of the access point as a
ath9k channel change helper
Signed-off-by: Stefan Rompf
---
Index: package/kernel/ubnt-hsr/Makefile
===
--- package/kernel/ubnt-hsr/Makefile(Revision 0
Hi,
this is the third iteration of the patchset to support the HSR filter of the
Ubiquiti Unifi Outdoor Plus access point.
The HSR is a configurable RF filter in the receive path that must be tuned
according to the selected Wifi channel for the access point to work.
Patch 1 adds a "channel set
This patch adds a "channel set helper" callback to the ath9k driver and
exports the ath9k kernel API for other packages. The registered function
is called whenever ath9k changes the channel.
Signed-off-by: Stefan Rompf
---
Index: package/kernel/mac8021
Add the tuner for the Ubiquiti Outdoor Plus HSR filter as a self contained
kmod package. When loaded, it registers with ath9k as a channel set helper and
tunes the HSR on every channel change.
Signed-off-by: Stefan Rompf
---
Index: package/kernel/ubnt-hsr/Makefile
Hi Felix,
thanks for your feedback!
> NACK from me for the ath9k_set_channel_helper part. This helper function
> needs to be passed in from the platform data, similar to how external
> reset is handled.
> From previous emails you pointed out that you guys chose this
> design/structure simply beca
Hi Felix,
> Maybe in this case it would be
> better to put the code in ath9k.ko and enable it via a Kconfig option in
> the mac80211 package. Then add a bool flag to the ath9k_platform_data
> struct to enable it only for devices that need it.
sounds good to me, will create a patch...
Stefan
Hi,
so we arrive at the fourth iteration of the patchset to support the HSR of the
Ubiquiti Unifi Outdoor Plus access point. This version implements the
suggestions from Felix. We lost the feature to avoid tuning the same channel
multiple times in a row, but this should not be critical.
The HS
Add ubnt_hsr flag to struct ath9k_platform_data and enable it for the Outdoor
Plus access point so that the availability ot the HSR filter can be detected
by the ath9k driver at runtime.
Signed-off-by: Stefan Rompf
---
Index: package/kernel/mac80211/patches/150-ath9k_ubnt_hsr_filter.patch
Add tuner for the HSR filter of the UniFi Outdoor Plus access point. Usage of
the tuner is controlled at runtime by ath9k_platform_data. The code can be
enabled or disabled by a compile time option.
Signed-off-by: Stefan Rompf
---
Index: package/kernel/mac80211/Makefile
On Dienstag, 16. Juni 2015 23:14:05 Stefan Rompf wrote:
> so we arrive at the fourth iteration of the patchset to support the HSR of
> the Ubiquiti Unifi Outdoor Plus access point. This version implements the
> suggestions from Felix.
Ping
Hi,
On Montag, 11. Januar 2016 23:45:39 Rafał Miłecki wrote:
> > Any comments to this usage/syntax? It doesn't look too nice (this long
> > quoted string as a value), but swconfig is strongly focused on simple
> > values.
>
> One more RFC... anyone?
I am not an OpenWRT core developer so I canno
Hi Rafał,
> int (*get_port_link)(struct switch_dev *dev, int port,
>struct switch_port_link *link);
> + int (*set_port_link)(struct switch_dev *dev, int port,
> + enum switch_port_speed speed);
this creates an assymetric API. I think
Hi Tim,
> The auto-built image of 9 December 2015 for this device results in a
> bricked state. TFTP recovery with a stock recovery firmware from
> Ubiquity does not allow the device to boot either.
try flashing the chaos calmer factory image with TFTP
(openwrt-15.05-ar71xx-generic-ubnt-unifi-out
Hi Rafał,
On Mittwoch, 16. Dezember 2015 18:19:01 Rafał Miłecki wrote:
> So far we were sending link data as a string. It got some drawbacks:
> 1) Didn't allow writing clean user space apps reading link state. It was
>needed to do some screen scraping.
> 2) Forced whole PORT_LINK communicatio
Hi Rafał,
> Supported syntax is inspired by ethtool. Example usage:
> swconfig dev switch0 port 2 set link "duplex half speed 100 autoneg off"
>
> Signed-off-by: Rafał Miłecki
I think that's the way to go, will try implementing set_port_link() for my
good old 1043ND.
Ack for this and the kern
Hi,
On Mittwoch, 11. März 2015 15:14:21 Shvedov Yury wrote:
> Thank you for your answer. We will try to solve this issue with our own
> forces.
any progress on this? I have one of these access points available right now
for hacking (as long as I do not brick it). The last hours I made those
fi
Hi Federico,
On Donnerstag, 12. März 2015 13:37:37 Nemesis wrote:
> I went at this URL:
> https://www.ubnt.com/download/?group=nanobeam-ac
>
> Clicked on "GPL Archive", downloaded it, extracted it, and took a look
> in it and found out a readme file which states:
>
> that the archive contains a
Hi Sergey,
> If they built SPIoverGPIO or I2CoverGPIO or some other serial bus over
> GPIO, then you do not see any changes between channel switches. Or
> they could use some non GPIO interface to communicate with external
> filter (embedded SPI, I2C or even USB or PCI of SoC).
well, GPIO has bee
On Dienstag, 17. März 2015 23:24:26 Sergey Ryazanov wrote:
> > Interesting enough: When stopping all wifi related software on the device
> > and creating a monitor interface, the RF filter seems to follow
> > "iwconfig channel" commands (yes it is the madwifi driver).
>
> How do you realized that
Hi Sergey,
> You should be a wizard to reveal hw protocol without touching hw :)
I cannot say 'Mischief managed!' by now, but it's slowly making progress ;-)
> Unlikely that iwconfig changes something directly, I would bet on the
> driver.
Indeed. So far
-The RF filter is controlled by the GPI
34 matches
Mail list logo
