kernel-packages team mailing list archive

["Christopher M. Penalver" <christopher.m.penalver@xxxxxxxxx>]

Nico Schlömer:
>"The problem here is that we are dealing with with closed-source software, so I cannot point to a place in the code that should be implemented differently."

Nobody expects you to point to code for closed source software, that
isn't the focus of this report, nor a requirement to get his bug
triaged.

>"I'm also not capable of running production tests for counting the
kernel oopses per day, or whatever metric the quality of a driver is
measured by."

Well, without some sort of number of oopses per time interval, the
release(s) this happened in, and a stack trace from your logs, this bug
report is useless for the purposes of dealing with your issue. If you
aren't willing or capable of providing the information requested, please
feel free to toggle the Status Invalid.

>" All I can say is that wl oopses more frequently now than it did
before."

Remaining detailed and specific:
1) How many times per time interval did it oops in which release(s) exactly?
2) Could you please provide a stack trace from your logs?
2) What release was the earliest release you have tested with your hardware?

>"My perception is that the old driver used to work better than
Broadcom's newer release,"

Vague perceptions are unhelpful, and doesn't have a place here on
Launchpad, a development platform.

>"and the same seems to be true for many other people; see the links in
the OP. This is why I opened this bug."

Again, what someone else claimed with some other hardware on some forum,
and linking to some trash URL is useless for the purpose of triaging
your problem.

>"I realize that this isn't much for you to work with. In particular, I
don't expect anyone to track down the bug. This is why I suggest to not
immediately discard old drivers on the arrival of a newer version."

Speculation, suggestions, subjective comments, etc. are simply unhelpful
here.

>"The new driver still accounts for about one wifi outage per hour here.
One OFF-ON on WiFi switch brings it back on."

So when the WiFi goes out in Utopic, is there anything in your logs
capturing a crash of some sort?

-- 
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to bcmwl in Ubuntu.
https://bugs.launchpad.net/bugs/1289857

Title:
  offer 6.20.155.1 build in trusty

Status in “bcmwl” package in Ubuntu:
  Incomplete

Bug description:
  As opposed to open-source software, proprietary software has the USP
  that newer versions are not always better. This seems to be the case
  for the 6.30.223.141 driver which offers considerably worse
  performance for many users than its predecessor 6.20.155.1. This
  assumption is based on the reactions
  <https://plus.google.com/+NickGroenen/posts/HF828QcPK6k> to
  instructions for a "forthport" of 6.20.155.1 to saucy
  <https://zoni.nl/posts/2013/11/09/fixing-the-broadcom-bcm4331
  -wireless-drivers-on-ubuntu-1310/>.

  Please continue to provide older BCM packages in the tree to make it
  easier for users to switch.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/bcmwl/+bug/1289857/+subscriptions