Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Driver crashes running kernel 5.0 #299

Closed
ZerBea opened this issue Mar 16, 2019 · 8 comments
Closed

Driver crashes running kernel 5.0 #299

ZerBea opened this issue Mar 16, 2019 · 8 comments

Comments

@ZerBea
Copy link

ZerBea commented Mar 16, 2019

While the driver is running fine, using kernel 4.19.23-1-lts, it crashes running kernel 5.0.2-arch1-1-ARCH.

4.19.23-1-lts:
$ hcxdumptool -i wlp0s29f7u3 --ignore_warning
initialization...

start capturing (stop with ctrl+c)
INTERFACE................: wlp0s29f7u3
ERRORMAX.................: 100 errors
FILTERLIST...............: 0 entries
MAC CLIENT...............: b0ece170192e
MAC ACCESS POINT.........: 00221c066d88 (incremented on every new client)
EAPOL TIMEOUT............: 150000
REPLAYCOUNT..............: 64235
ANONCE...................: 5b6c86960af3887e50bf88622cac56620a41d22304db91927863f5541b0be8f6

INFO: cha=11, rx=1266, rx(dropped)=433, tx=79, powned=2, err=0

5.0.2-arch1-1-ARCH:
$ hcxdumptool -i wlp3s0f0u1 ---ignore_warning
initialization...

start capturing (stop with ctrl+c)
INTERFACE................: wlp3s0f0u1
ERRORMAX.................: 100 errors
FILTERLIST...............: 0 entries
MAC CLIENT...............: b0febdb30d78
MAC ACCESS POINT.........: 544e4585957f (incremented on every new client)
EAPOL TIMEOUT............: 150000
REPLAYCOUNT..............: 61451
ANONCE...................: 7f34b17ce8dcd85c287ca7bc39e55d524a490c997243a513e94f0a2eec29d06c

INFO: cha=1, rx=0, rx(dropped)=0, tx=10, powned=0, err=0

Sometimes a few packets are delivered via raw_socket, before the driver stops working:
INFO: cha=1, rx=233, rx(dropped)=31, tx=99, powned=0, err=0

But mostly no packets are delivered
INFO: cha=1, rx=0, rx(dropped)=0, tx=10, powned=0, err=0

Maybe something went wrong in combination with kernel 5.0, because the driver was working fine running 4.19

Do you see the same behavior running aireplay-ng?

@kimocoder
Copy link
Collaborator

which branch? I've spotted some kernel patches in the wild but haven't really got time to look at it right away.

@kimocoder
Copy link
Collaborator

Take a look https://github.com/lwfinger/rtl8812au/commits/v5.2.6 if you need a quick fix until I'm back

@ZerBea
Copy link
Author

ZerBea commented Mar 16, 2019

Hi Christian.
Running the default branch, here. Maybe we have some important changes within kernel 5.0. Right now, I noticed that behavior on the mt76x0, too, while mt7601u and rt2800usb are working fine on 5.0.
I'll check https://github.com/lwfinger/rtl8812au/commits/v5.2.6.
BTW:
The rtl8188eu seems to work fine, too (after I noticed that the driver reports the frequency in Hz instead of MHz):
ZerBea/hcxdumptool#48
I had to add a new option to hcxdumptool (--ignore_warning), because it looks like both drivers
doesn't support all SIOCGIWMODE flags.

Cheers
Mike

@ZerBea
Copy link
Author

ZerBea commented Mar 16, 2019

Looks like the driver from here
https://github.com/lwfinger/rtl8812au/commits/v5.2.6
has a regression, too (or it reports wrong tx power):
Interface wlp3s0f0u1
ifindex 3
wdev 0x1
addr 74:da:38:06:45:e7
type managed
wiphy 0
txpower 12.00 dBm

this driver (aircrack-ng/rtl8812au):
Interface wlp3s0f0u1
ifindex 5
wdev 0x200000001
addr 74:da:38:06:45:e7
type managed
wiphy 2
txpower 18.00 dBm

@kimocoder
Copy link
Collaborator

@kcdtv join me and Alfa Networks om Telegram.
@Re4son should also join the party.

Fins me by the Nick

@ZerBea
Copy link
Author

ZerBea commented Mar 28, 2019

Just tested RTL8187 (included in kernel source) and noticed that the driver showing the same issues running on kernel >= 5.0.
Even Wireshark doesn't receive packets.
ZerBea/hcxdumptool#42 (comment)
That's very interesting, because kernel <= 4.20 working fine.

@ZerBea
Copy link
Author

ZerBea commented Mar 30, 2019

Running some further going tests and found the cause of the crash. It is related to the radiotap header.
Thanks for your effort.
Cheers
Mike

@ZerBea
Copy link
Author

ZerBea commented Apr 2, 2019

Closed that issue, because it is possible related to xhci subsystem in combination with USB3.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants