Modify

Opened 5 years ago

Last modified 23 months ago

#12073 reopened defect

AR9331 / hornet(-ub) >= r33031 is dangerous?!

Reported by: Halo2 Owned by: developers
Priority: normal Milestone: Barrier Breaker 14.07
Component: packages Version: Trunk
Keywords: Cc:

Description

Its the second time I upgrade a Alfa Hornet-UB and have a defect one the day after.
The defect is that it does not receive anything (except for a one meter range or so).
Directly after upgrading its okay. Downgrade after defect does not help. Receive side is really dead.

1st one was upgraded from r32764 to r32764 but with mac80211 from r33032.
2nd one from r32764 to r33264.

So in my opinion there are 2 possiblities:
1st: r33031
2nd: lightning damage

To the possible lightning damage:
There were just some discharges kilometres away (before and after the update). Before upgrading there was no performance degrade (for weeks). In the night of the day I've upgraded, at least the 2nd one, was dead a few hours later.
So it would be a unbelievable sick coincidence if this happened 2 times exactly after upgrading, and never before. I'm not sure if I want to kill a third one, but by the way I don't have one left now, so i currently can't test it anyhow.
Antenna is protected by lambda/4 surge protection. A other router on the same pole is still okay after a year (and many discharges in really close proximity).

So I hope of input from other people who have their hornet-ub / ar933x on >= r33031 (and propably with an real external antenna).

Thank you!

---

this
antispam
is
a
little
bit
sick

Attachments (0)

Change History (6)

comment:1 follow-up: Changed 5 years ago by nbd

  • Resolution set to fixed
  • Status changed from new to closed

I found a serious issue in the PA predistortion code in ath9k which was enabled in r33031. This was causing crashes on AR9330 and might also have been responsible for the hw issues that you're seeing.

The driver did a dummy PAPRD calibration transmission while the gain tables were not initialized properly, which might have caused the rx path damage, especially on high power devices.

This is fixed in r33278

comment:2 in reply to: ↑ 1 Changed 5 years ago by anonymous

Replying to nbd:

I found a serious issue in the PA predistortion code in ath9k which was enabled in r33031. This was causing crashes on AR9330 and might also have been responsible for the hw issues that you're seeing.

The driver did a dummy PAPRD calibration transmission while the gain tables were not initialized properly, which might have caused the rx path damage, especially on high power devices.

This is fixed in r33278

Thanks for the quick fix!
Will be able to retest this when new hardware arrives.. .
This only influenced AR9485 and AR9330?
Or is PAPRD now on for ar938x/939x too (but not ar92xx)?

comment:3 Changed 5 years ago by nbd

PAPRD is on for AR938x as well, however not all cards support it (there is an EEPROM flag for it).

comment:4 Changed 4 years ago by jow

  • Milestone changed from Attitude Adjustment 12.09 to Barrier Breaker 14.07

Milestone Attitude Adjustment 12.09 deleted

comment:5 Changed 23 months ago by anonymous

problem not fixed. the issues described in https://dev.openwrt.org/ticket/16543#comment:7

comment:6 Changed 23 months ago by anonymous

  • Resolution fixed deleted
  • Status changed from closed to reopened

Add Comment

Modify Ticket

Action
as reopened .
Author


E-mail address and user name can be saved in the Preferences.

 
Note: See TracTickets for help on using tickets.