ejolson
Posts: 1644
Joined: Tue Mar 18, 2014 11:47 am

Re: RPI3 USB Hardware Bug?

Fri Apr 27, 2018 11:09 pm

user12345 wrote:
Fri Apr 27, 2018 9:10 pm
I read the BananaPI support is bad but this outclass all.
The Pi 3B+ contains two internal USB hubs. One is cascaded to the other. Not all of the USB sockets attach to the same hub. Use

$ lsusb -t

to see which physical connections go where.

My experience with a USB WiFi dongle indicated that it didn't work as well from the cascaded hub as from the other. I think there was someone reporting issues with a disconnecting USB harddisk last week.

User avatar
Greg Erskine
Posts: 71
Joined: Sat Sep 15, 2012 4:20 am

Re: RPI3 USB Hardware Bug?

Sat Apr 28, 2018 12:00 am

user12345 wrote:
Tue Apr 17, 2018 12:19 pm
Cause RPI2 and RPI3 are a different Hardware Design the only thing thats stays is a long standing Kernel Bug.
I'm lucky that this Kernel Bug hasn't reached the southern hemisphere yet because all my Raspberry Pi's on wifi have worked properly for years! :mrgreen:

drgeoff
Posts: 8194
Joined: Wed Jan 25, 2012 6:39 pm

Re: RPI3 USB Hardware Bug?

Sat Apr 28, 2018 7:54 am

user12345 wrote:
Tue Apr 17, 2018 12:19 pm
.. RPI2 and RPI3 are a different Hardware Design ..
AIUI the USB aspects of RPi2B and RPi3B are identical. (RPi3B+ is different.)

user12345
Posts: 58
Joined: Mon Dec 22, 2014 7:06 pm

Re: RPI3 USB Hardware Bug?

Thu May 03, 2018 6:24 am

ejolson wrote:
Fri Apr 27, 2018 11:09 pm
My experience with a USB WiFi dongle indicated that it didn't work as well from the cascaded hub as from the other. I think there was someone reporting issues with a disconnecting USB harddisk last week.
I mixed the Ports already, dont really understand the output, right now its the RPI2 i see root_hub->hub (hub/5p).
1/3/0 Network
3/4/2 Audio
3/4/X Webcam
4/5/X Webcam
5/6/0 Wifi
Greg Erskine wrote:
Sat Apr 28, 2018 12:00 am
I'm lucky that this Kernel Bug hasn't reached the southern hemisphere yet because all my Raspberry Pi's on wifi have worked properly for years! :mrgreen:
This help not much if you dont tell witch Kernel and Wifi stick you use.

Cause i already used a 0.2 Metre cable and now the 5th wifi stick there are only a
-Kernel Bug
-missing watch dog ieee80211 for example
-another rt2800usb driver bug(.bin firmware is fine same file like on real linux that works 24/7 since years without a error)
-A already confirmed Kernel Bug is of course that the device is still shown under lsusb/ifconfig/iwconfig, a working Linux/OS drop the
Device if it doesnt respond after some seconds, this is too called "NAK Timeout".
-Hardware Design error(this transistors with a much higer Voltage Drop than the Polyfuse, or a noisy Line for example in the USB Power switch, or similar)

If you Power off the RPI and on again the Error is not so often like with using "modprobe -r rt2800usb". Power down the rpi is of course not
acceptable. But this again speaks for a Kernel Bug. And meanwhile 2 webcams works without error, but low bandwitdth wifi fails.

user12345
Posts: 58
Joined: Mon Dec 22, 2014 7:06 pm

Re: RPI3 USB Hardware Bug?

Wed Jun 06, 2018 6:44 pm

Very bad support.
Still there the Bug with newest kernel 4.14.44-v7+.
Is it possible to use 3.12.22 wireless modules? they are not accepted if i simply copy them over but maybe there exists some trick to use it?

jamesh
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 19809
Joined: Sat Jul 30, 2011 7:41 pm

Re: RPI3 USB Hardware Bug?

Wed Jun 06, 2018 8:32 pm

I just read through this thread again to see if there is anything I can suggest, but without having the particular wifi dongle being used it difficult to make any assessment. I'm pretty sure its the stick or the driver for the stick, if only because huge numbers of users of the 3B and 3B+ have no problems at all with the onboard Wifi (and wifi dongles on the 2). If you use the onboard Wifi of the 3B does that work OK?
Principal Software Engineer at Raspberry Pi (Trading) Ltd.
Please direct all questions to the forum, I do not do support via PM.

user12345
Posts: 58
Joined: Mon Dec 22, 2014 7:06 pm

Re: RPI3 USB Hardware Bug?

Thu Jun 07, 2018 11:44 am

jamesh wrote:
Wed Jun 06, 2018 8:32 pm
I just read through this thread again to see if there is anything I can suggest, but without having the particular wifi dongle being used it difficult to make any assessment. I'm pretty sure its the stick or the driver for the stick, if only because huge numbers of users of the 3B and 3B+ have no problems at all with the onboard Wifi (and wifi dongles on the 2). If you use the onboard Wifi of the 3B does that work OK?
I tryed 3 Alpha AWUS(148f:3070 Ralink Technology, Corp. RT2870/RT3070) and a 1 Tenda Stick, this has the same RT Base Chip but other addtitional Chips. The Alpha sticks are very common and very famous cause they have the best sensitivity&Signal quality.
Cause its was 4 different Sticks(3 of them running fine 24/7 on Desktop Linux) it cant be the stick.
It may be the driver (the driver firmware is the same like on Desktop Linux CRC32 compared). The Desktop runned different Linux(Backtrack/2 Kali versions), with none there was a problem.

The Answer for my Question why a "dead" device is listed by ifconfig/iwconfig/lsusb would too intresting. If a device doesnt responds it should be dropped by Linux.

I disabled onboard Wifi cause of this bug and Random wlan0/wlan1. Monitor Mode doesnt work with it and it has very bad sensitivity&Signal.
Other People that have working wifi may often use the "Managed Mode" or use small wifi sticks/onboard that doesnt get much wifi traffic.

-It may possible that the bug has todo with the Monitor Mode part.
-Cause the bug frequency has increased over the Time, it has todo with increased wifi devices&traffic
-It seems sticks with less sensitivity trigger the bug a bit less.
-It may be something like the driver/packed filter that get a malicous 802.11 frame and crash (driver may modified from original Debian sources or not ported well, eg. datatype "long" is today usual 64Bit but on rpi its 32 Bit).

user12345
Posts: 58
Joined: Mon Dec 22, 2014 7:06 pm

Re: RPI3 USB Hardware Bug?

Tue Jun 12, 2018 8:56 pm

Found something intresting, tryed to add some debug output to airodump-ng.
It has a Keyboard Task, a Channel switch Task and Main.
While the Kernel Bug the Keyboard&Channel switch Tasks works, the display output not really(ixconfig show different frequencys and keys works)
Added file debug output after wi_read in the main.
1) Got very few and impossible output, wi_read returns all 0, and a number that only counts++, counts backwarts in the debug output. Means the library that does "wi_read" overwrites the memory area of the Program&after 0-20 counts it dont return. Cause i cant build in a watchdog that do "modeprobe -r" for bug workaround.
Shouldnt Linux have MMU&Memory Protection? It dont seem work well.

2) usual you use airtools in that way:
airmon-ng start wlan0 <-creates a mon0 device, this way you can use wlan0 managed and mon0 as monitor device
airodump-ng -i mon0 --update 1 -w /tmp/output.csv (--update 1 for slow terminal bug of rpi)

Newer Versions or airodump can work direct on wlan0 too by setting it into monitor mode.
While the Kernel Bug was active i tryed "airodump-ng -i wlan0", it worked!
That means mon0 is crashed, not the Wifi Stick, not usb, not rt2800usb stuff.
First i think the created interface mon0 by using "iw" is buggy.
I restartet the RPI, used wlan0 only. wlan0 crashed too after 1 day. It crash airodump-ng with it too, got no debug output.
After "ifconfig wlan0 down" and "ifconfig wlan0 up" airodump works again, but this dont work on the fly.
Means no bug workaround.

So Developers, now you know where the Bug is placed, can you FINALLY FIX IT PLEEEASE?!

ejolson
Posts: 1644
Joined: Tue Mar 18, 2014 11:47 am

Re: RPI3 USB Hardware Bug?

Wed Jun 13, 2018 1:08 am

user12345 wrote:
Tue Jun 12, 2018 8:56 pm
Found something intresting
You seem to have multiple USB webcams and WiFi adapters connected. Webcams and WiFi can be surprisingly power hungry. Drawing more current than the official power supply provides could cause WiFi devices, for example, to go offline without any other observable issues. I experienced this with a battery powered Pi mounted in a remote control car.

I recently set up a Pi 3B+ as a WiFi to WiFi forwarding masquerade firewall. In this case, the USB WiFi device would occasionally go offline but a reboot would restore it. The problem was fixed by disabling power saving mode.

Sorry I can't help more, this seems outside my level.

user12345
Posts: 58
Joined: Mon Dec 22, 2014 7:06 pm

Re: RPI3 USB Hardware Bug?

Wed Jun 13, 2018 7:28 am

@ejolson its a Kernel Bug, powersaving is disabled and multible PSUs are tested(even a 3 A continues),too direct powering usb.

jamesh
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 19809
Joined: Sat Jul 30, 2011 7:41 pm

Re: RPI3 USB Hardware Bug?

Wed Jun 13, 2018 9:59 am

user12345 wrote:
Tue Jun 12, 2018 8:56 pm
Found something intresting, tryed to add some debug output to airodump-ng.
It has a Keyboard Task, a Channel switch Task and Main.
While the Kernel Bug the Keyboard&Channel switch Tasks works, the display output not really(ixconfig show different frequencys and keys works)
Added file debug output after wi_read in the main.
1) Got very few and impossible output, wi_read returns all 0, and a number that only counts++, counts backwarts in the debug output. Means the library that does "wi_read" overwrites the memory area of the Program&after 0-20 counts it dont return. Cause i cant build in a watchdog that do "modeprobe -r" for bug workaround.
Shouldnt Linux have MMU&Memory Protection? It dont seem work well.

2) usual you use airtools in that way:
airmon-ng start wlan0 <-creates a mon0 device, this way you can use wlan0 managed and mon0 as monitor device
airodump-ng -i mon0 --update 1 -w /tmp/output.csv (--update 1 for slow terminal bug of rpi)

Newer Versions or airodump can work direct on wlan0 too by setting it into monitor mode.
While the Kernel Bug was active i tryed "airodump-ng -i wlan0", it worked!
That means mon0 is crashed, not the Wifi Stick, not usb, not rt2800usb stuff.
First i think the created interface mon0 by using "iw" is buggy.
I restartet the RPI, used wlan0 only. wlan0 crashed too after 1 day. It crash airodump-ng with it too, got no debug output.
After "ifconfig wlan0 down" and "ifconfig wlan0 up" airodump works again, but this dont work on the fly.
Means no bug workaround.

So Developers, now you know where the Bug is placed, can you FINALLY FIX IT PLEEEASE?!
Sorry, I do not understand most of this post. As far as I can tell it doesn't really give any idea as to where the bug might even be, although I am presuming it's in the driver for the stick somewhere, which is NOT OUR CODE. We cannot generally provide support drivers for devices that are not produced by ourselves. There is the kernel wireless mailing list that might be able to help, there are Realtek driver developers on there.
Principal Software Engineer at Raspberry Pi (Trading) Ltd.
Please direct all questions to the forum, I do not do support via PM.

user12345
Posts: 58
Joined: Mon Dec 22, 2014 7:06 pm

Re: RPI3 USB Hardware Bug?

Wed Jun 13, 2018 4:38 pm

jamesh wrote:
Wed Jun 13, 2018 9:59 am
Sorry, I do not understand most of this post. As far as I can tell it doesn't really give any idea as to where the bug might even be, although I am presuming it's in the driver for the stick somewhere, which is NOT OUR CODE. We cannot generally provide support drivers for devices that are not produced by ourselves. There is the kernel wireless mailing list that might be able to help, there are Realtek driver developers on there.
The device wlan0/mon0 inself is dead. If mon0 is crashed, wlan0 can still be used. But wlan0 can crash too if using it allone.
If the device or driver would be dead you could not use mon0 AND not use wlan0 both would be inaccessable.
Means if i create 20 interfaces i could use one after another until all 20 are crashed.
No, it cant be a Problem of the original driver developers. As the drivers work on Linux x86/x64/rpi1/bananaPI 24h/7days per week
Even if it is rt2800usb, the raspberry developer had ported it wrong to ARM, or modified it.
As far i understand iw creates too wlan0, its possible thats (Ported by Raspbian authors, works fine on other systems) iw contains the bug.

Its and stays on Rasbian same thing works on other systems 24/7.

jamesh
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 19809
Joined: Sat Jul 30, 2011 7:41 pm

Re: RPI3 USB Hardware Bug?

Wed Jun 13, 2018 8:01 pm

user12345 wrote:
Wed Jun 13, 2018 4:38 pm
jamesh wrote:
Wed Jun 13, 2018 9:59 am
Sorry, I do not understand most of this post. As far as I can tell it doesn't really give any idea as to where the bug might even be, although I am presuming it's in the driver for the stick somewhere, which is NOT OUR CODE. We cannot generally provide support drivers for devices that are not produced by ourselves. There is the kernel wireless mailing list that might be able to help, there are Realtek driver developers on there.
The device wlan0/mon0 inself is dead. If mon0 is crashed, wlan0 can still be used. But wlan0 can crash too if using it allone.
If the device or driver would be dead you could not use mon0 AND not use wlan0 both would be inaccessable.
Means if i create 20 interfaces i could use one after another until all 20 are crashed.
No, it cant be a Problem of the original driver developers. As the drivers work on Linux x86/x64/rpi1/bananaPI 24h/7days per week
Even if it is rt2800usb, the raspberry developer had ported it wrong to ARM, or modified it.
As far i understand iw creates too wlan0, its possible thats (Ported by Raspbian authors, works fine on other systems) iw contains the bug.

Its and stays on Rasbian same thing works on other systems 24/7.
You seem to think we port (ie change for Pi) a lot of these drivers. We don't. The huge majority of code running on the Pi is upstream code, simply recompiled.

Only drivers for specific Pi devices will have been changed by us, and the platform code to bring everything up.

Worth noting, that just because a device and it's driver work on one architecture/platform (e.g. x86) that is no guarantee they will work correctly on another architecture/platform. They usually do, but often a different combination of things can cause drivers to behave slightly differently, often exposing bugs as new code paths are suddenly used.

Also worth noting that if all these drivers were bug free, why do they continually get bug fixes released for them? You only need to subscribe to the netdev mailing list to see that Realtek drivers (and many others) have a lot of churn.
Principal Software Engineer at Raspberry Pi (Trading) Ltd.
Please direct all questions to the forum, I do not do support via PM.

User avatar
scruss
Posts: 1701
Joined: Sat Jun 09, 2012 12:25 pm
Location: Toronto, ON
Contact: Website

Re: RPI3 USB Hardware Bug?

Wed Jun 13, 2018 10:41 pm

I've used one of these before, but ISTR having to use a USB hub on the Raspberry Pi because they draw a lot of power.

Do an lsusb -v on the device, and see what its MaxPower entry is. You may be trying to draw more current than you can deliver from a Raspberry Pi - especially if you have other USB devices plugged in.
‘Remember the Golden Rule of Selling: “Do not resort to violence.”’ — McGlashan.

user12345
Posts: 58
Joined: Mon Dec 22, 2014 7:06 pm

Re: RPI3 USB Hardware Bug?

Thu Jun 14, 2018 4:44 pm

jamesh wrote:
Wed Jun 13, 2018 8:01 pm
You seem to think we port (ie change for Pi) a lot of these drivers. We don't. The huge majority of code running on the Pi is upstream code, simply recompiled.

Also worth noting that if all these drivers were bug free, why do they continually get bug fixes released for them? You only need to subscribe to the netdev mailing list to see that Realtek drivers (and many others) have a lot of churn.
I think it isnt the driver just the (virtual)device(wlan0/mon0) goes down, the drivers&device are alive.

Fact is it worked with Kernel 3.12, 24/7. And dont with 4 other Kernels, if you build in buggy stuff you should Rollback!

I too dont understand why "apt-get update", "apt-get upgrade", "rpi-update" dont update the /sbin dir, there are udev, ifconfig, iw stuff from 2012 in there that maybe incompatible with Kernel 4.1 and up.

We get continually updates of rasbian to drive the people crazy. Kernel 3.12 worked 24/7 but PI1 had only 1 core and is slow. People buy rpi2/3 cause slowness and think it work like pi1 and then fight years with the bugs they never would expect. Developers dont care, device is selled already.

User avatar
rpdom
Posts: 12582
Joined: Sun May 06, 2012 5:17 am
Location: Essex, UK

Re: RPI3 USB Hardware Bug?

Thu Jun 14, 2018 5:36 pm

user12345 wrote:
Thu Jun 14, 2018 4:44 pm
I too dont understand why "apt-get update", "apt-get upgrade", "rpi-update" dont update the /sbin dir, there are udev, ifconfig, iw stuff from 2012 in there that maybe incompatible with Kernel 4.1 and up.
That has nothing to do with specific Raspberry Pi builds. I am running a fully updated Ubuntu PC and the dates on the files are similar. There are no newer versions in the "stable" (Stretch) release. The ones provided do work with the current kernels.

pws
Posts: 52
Joined: Mon Apr 11, 2016 4:16 pm

Re: RPI3 USB Hardware Bug?

Thu Jun 14, 2018 5:46 pm

So...of the hundreds of boards out there, ONE has demonstrated this "issue", so everything MUST stop because it has been declared a major "Bug" by ONE person... :roll:

ejolson
Posts: 1644
Joined: Tue Mar 18, 2014 11:47 am

Re: RPI3 USB Hardware Bug?

Thu Jun 14, 2018 9:12 pm

user12345 wrote:
Thu Jun 14, 2018 4:44 pm
I too dont understand why "apt-get update", "apt-get upgrade", "rpi-update" dont update the /sbin dir, there are udev, ifconfig, iw stuff from 2012 in there that maybe incompatible with Kernel 4.1 and up.
This post has been answered by people affiliated with the Raspberry Pi. If the developers didn't care, they would be locking the thread and banning the person who made the original post.

You mention upgrading from a 3.x kernel. Have you tried a fresh install of the latest Raspbian?

jamesh
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 19809
Joined: Sat Jul 30, 2011 7:41 pm

Re: RPI3 USB Hardware Bug?

Fri Jun 15, 2018 7:45 am

user12345 wrote:
Thu Jun 14, 2018 4:44 pm
jamesh wrote:
Wed Jun 13, 2018 8:01 pm
You seem to think we port (ie change for Pi) a lot of these drivers. We don't. The huge majority of code running on the Pi is upstream code, simply recompiled.

Also worth noting that if all these drivers were bug free, why do they continually get bug fixes released for them? You only need to subscribe to the netdev mailing list to see that Realtek drivers (and many others) have a lot of churn.
I think it isnt the driver just the (virtual)device(wlan0/mon0) goes down, the drivers&device are alive.

Fact is it worked with Kernel 3.12, 24/7. And dont with 4 other Kernels, if you build in buggy stuff you should Rollback!

I too dont understand why "apt-get update", "apt-get upgrade", "rpi-update" dont update the /sbin dir, there are udev, ifconfig, iw stuff from 2012 in there that maybe incompatible with Kernel 4.1 and up.

We get continually updates of rasbian to drive the people crazy. Kernel 3.12 worked 24/7 but PI1 had only 1 core and is slow. People buy rpi2/3 cause slowness and think it work like pi1 and then fight years with the bugs they never would expect. Developers dont care, device is selled already.
If we didn't care we wouldn't have spent the last week debugging odd ethernet issues on the Pi3b+, or investigating a Wifi bug, or any other of the myriad things we do every day to make the product work better. In fact, I think you would be hard pressed to find a more dedicated team. Just because your very specific issue has not yet been solved, does not mean we are all sitting back doing nothing. It just means, in our list of stuff that needs fixing, its not very high on the list. It's not even our HW or driver!
Principal Software Engineer at Raspberry Pi (Trading) Ltd.
Please direct all questions to the forum, I do not do support via PM.

user12345
Posts: 58
Joined: Mon Dec 22, 2014 7:06 pm

Re: RPI3 USB Hardware Bug?

Fri Jun 22, 2018 2:12 pm

rpdom wrote:
Thu Jun 14, 2018 5:36 pm
That has nothing to do with specific Raspberry Pi builds. I am running a fully updated Ubuntu PC and the dates on the files are similar. There are no newer versions in the "stable" (Stretch) release. The ones provided do work with the current kernels.
The crc of the updated/upgraded card has old Filedates, different crc´s and tell a different Version if you enter "iw --version" for example compared
to the wheezy or stretch image.
I replaced iw* in /sbin with the newest Version. The Bug stays the same but was worth a try.

user12345
Posts: 58
Joined: Mon Dec 22, 2014 7:06 pm

Re: RPI3 USB Hardware Bug?

Fri Jun 22, 2018 2:16 pm

ejolson wrote:
Thu Jun 14, 2018 9:12 pm
You mention upgrading from a 3.x kernel. Have you tried a fresh install of the latest Raspbian?
Its very difficult to get a new image run all Software that is on the old card.
Stretch is buggy and cant run dolphin&kwrite.
But i prepared a bit a wheezy Version but no time to complete.

user12345
Posts: 58
Joined: Mon Dec 22, 2014 7:06 pm

Re: RPI3 USB Hardware Bug?

Fri Jun 22, 2018 2:35 pm

jamesh wrote:
Fri Jun 15, 2018 7:45 am
If we didn't care we wouldn't have spent the last week debugging odd ethernet issues on the Pi3b+, or investigating a Wifi bug, or any other of the myriad things we do every day to make the product work better. In fact, I think you would be hard pressed to find a more dedicated team. Just because your very specific issue has not yet been solved, does not mean we are all sitting back doing nothing. It just means, in our list of stuff that needs fixing, its not very high on the list. It's not even our HW or driver!
I reported the Bug in March 2018, the Bug exists at least since RPI3 exists(tryed old Kernels).
RPI2/3 is advertised to support Wifi sticks and special to support MY wifi sticks:
https://elinux.org/RPi_USB_Wi-Fi_Adapters
It doesnt help me much if you fix new Hardware while old is still broken.

Meanwhile i reprogrammed airodump-ng with an watchdog, it sits in the channelhopper Task and looks if the incomming packets increases.
If not it inhibit the main Task(its frozen anyway but for the follow), it does ifconfig wlan0 down, ifconfig wlan0 up.
The Main Task freeze at wi_read, cause of ifconfig it wakes up now, waits cause of the inhibit.
The other things it does allone cause it well develop and recognize wlan0 was down and reopen it, set it to monitor mode.
It survive already 14 times the Bug, today 7 times of course it loose about 38 seconds every time caused by delays but i dont have
interrupted files.


Today i get this (the Queue 14 is on every down/up while the Bug occur, the WARNING: is the first time, seems another bug but may help anyway).
[175562.872966] ieee80211 phy0: rt2x00queue_flush_queue: Warning - Queue 14 failed to flush
[176482.908830] ------------[ cut here ]------------
[176482.908920] WARNING: CPU: 0 PID: 7028 at drivers/usb/core/urb.c:363 usb_submit_urb+0x2e0/0x4cc
[176482.908927] URB afc5d500 submitted while active
[176482.908933] Modules linked in: snd_bcm2835(C) arc4 rt2800usb rt2800lib rt2x00usb rt2x00lib mac80211 evdev cfg80211 rfkill crc_ccitt snd_usb_audio snd_hwdep snd_usbmidi_lib snd_seq_midi snd_seq_midi_event snd_rawmidi snd_pcm snd_seq snd_seq_device snd_timer uvcvideo videobuf2_vmalloc videobuf2_memops videobuf2_v4l2 snd videobuf2_core videodev fixed uio_pdrv_genirq uio media
[176482.909269] CPU: 0 PID: 7028 Comm: kworker/u8:0 Tainted: G C 4.14.44-v7+ #1117
[176482.909275] Hardware name: BCM2835
[176482.909305] Workqueue: phy0 rt2x00usb_work_rxdone [rt2x00usb]
[176482.909340] [<8010ffd8>] (unwind_backtrace) from [<8010c240>] (show_stack+0x20/0x24)
[176482.909360] [<8010c240>] (show_stack) from [<80785424>] (dump_stack+0xd4/0x118)
[176482.909379] [<80785424>] (dump_stack) from [<8011da4c>] (__warn+0xf8/0x110)
[176482.909394] [<8011da4c>] (__warn) from [<8011daac>] (warn_slowpath_fmt+0x48/0x50)
[176482.909412] [<8011daac>] (warn_slowpath_fmt) from [<805b2fb4>] (usb_submit_urb+0x2e0/0x4cc)
[176482.909442] [<805b2fb4>] (usb_submit_urb) from [<7f35d6b8>] (rt2x00usb_kick_rx_entry+0xc0/0x100 [rt2x00usb])
[176482.909498] [<7f35d6b8>] (rt2x00usb_kick_rx_entry [rt2x00usb]) from [<7f35d728>] (rt2x00usb_clear_entry+0x30/0x34 [rt2x00usb])
[176482.909563] [<7f35d728>] (rt2x00usb_clear_entry [rt2x00usb]) from [<7f348ac0>] (rt2x00lib_rxdone+0x154/0x5b0 [rt2x00lib])
[176482.909629] [<7f348ac0>] (rt2x00lib_rxdone [rt2x00lib]) from [<7f35e028>] (rt2x00usb_work_rxdone+0x58/0x8c [rt2x00usb])
[176482.909662] [<7f35e028>] (rt2x00usb_work_rxdone [rt2x00usb]) from [<80137528>] (process_one_work+0x158/0x454)
[176482.909685] [<80137528>] (process_one_work) from [<80137888>] (worker_thread+0x64/0x5b8)
[176482.909705] [<80137888>] (worker_thread) from [<8013d8f8>] (kthread+0x13c/0x16c)
[176482.909725] [<8013d8f8>] (kthread) from [<8010810c>] (ret_from_fork+0x14/0x28)
[176482.910047] ---[ end trace cd7654f098b82474 ]---
[176529.608738] ieee80211 phy0: rt2x00queue_flush_queue: Warning - Queue 14 failed to flush
[177935.186516] ieee80211 phy0: rt2x00queue_flush_queue: Warning - Queue 14 failed to flush
[180793.450893] ieee80211 phy0: rt2x00queue_flush_queue: Warning - Queue 14 failed to flush
[180855.661209] ieee80211 phy0: rt2x00queue_flush_queue: Warning - Queue 14 failed to flush
Last edited by user12345 on Fri Jun 22, 2018 2:39 pm, edited 1 time in total.

jamesh
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 19809
Joined: Sat Jul 30, 2011 7:41 pm

Re: RPI3 USB Hardware Bug?

Fri Jun 22, 2018 2:37 pm

user12345 wrote:
Fri Jun 22, 2018 2:16 pm
ejolson wrote:
Thu Jun 14, 2018 9:12 pm
You mention upgrading from a 3.x kernel. Have you tried a fresh install of the latest Raspbian?
Its very difficult to get a new image run all Software that is on the old card.
Stretch is buggy and cant run dolphin&kwrite.
But i prepared a bit a wheezy Version but no time to complete.
Just tried dolphin and kwrite on my 3B+ running latest Raspbian plus kernel 4.14.50, both worked fine.

I doubt Stretch is particularly buggy.
Principal Software Engineer at Raspberry Pi (Trading) Ltd.
Please direct all questions to the forum, I do not do support via PM.

SurferTim
Posts: 1648
Joined: Sat Sep 14, 2013 9:27 am
Location: Miramar Beach, Florida

Re: RPI3 USB Hardware Bug?

Fri Jun 22, 2018 2:40 pm

user12345 wrote:
Fri Jun 22, 2018 2:16 pm
Stretch is buggy and cant run dolphin&kwrite.
Wrong. I can run both on my Stretch box.

user12345
Posts: 58
Joined: Mon Dec 22, 2014 7:06 pm

Re: RPI3 USB Hardware Bug?

Fri Jun 22, 2018 2:43 pm

jamesh wrote:
Fri Jun 22, 2018 2:37 pm
Just tried dolphin and kwrite on my 3B+ running latest Raspbian plus kernel 4.14.50, both worked fine.
4.14 from April. All images are missing if you run it as root. (both required in root mode). I made photos but already deleted it.
With lastest wheezy it works.

Return to “Troubleshooting”

Who is online

Users browsing this forum: No registered users and 32 guests