Red Hat Bugzilla – Bug 155258
xircom_cb much slower than xircom_tulip_cb
Last modified: 2015-01-04 17:18:59 EST
From Bugzilla Helper:
User-Agent: Mozilla/5.0 Galeon/1.2.10 (X11; Linux i686; U;) Gecko/20030412
Description of problem:
I recently upgraded my old Inspiron 5000 from FC1 to FC3. Since the system
doesn't have built-in Ethernet, I've had a Xircom Cardbus Ethernet PCMCIA card
in it for years. In FC3, the card uses the xircom_cb driver.
After the upgrade, I discovered that my network throughput had dropped significantly. Whereas in 2.4.x, I could wget from another system on the LAN at
8-9MB/s (64-72Mbit) in FC3 I'm getting 2-3MB/s (16-24Mbit) from the same system
downloading the same test file. Out of curiousity, I installed vanilla 188.8.131.52
and observed the same issue using the xircom_cb driver. However, switching over
to the xircom_tulip_cb driver, my throughput is right back where it was in 2.4.x.
I've reported the issue to the linux-pcmcia list at infradead, but wanted to ask
that you consider enabling the xircom_tulip_cb module in a future kernel update
to address this issue.
Version-Release number of selected component (if applicable):
Steps to Reproduce:
1. Install FC3 on an Inspiron 5000 with a Xircom Cardbus Ethernet PCMCIA card
2. Test throughput
Actual Results: Throughput is much lower than it was in FC1. (2.4.x)
Expected Results: Throughput should be as good as 2.4.x. (xircom_tulip_cb module should be
installed for now as a fallback)
An update has been released for Fedora Core 3 (kernel-2.6.12-1.1372_FC3) which
may contain a fix for your problem. Please update to this new kernel, and
report whether or not it fixes your problem.
If you have updated to Fedora Core 4 since this bug was opened, and the problem
still occurs with the latest updates for that release, please change the version
field of this bug to 'fc4'.
Unfortunately, this system is now in production using a different card, so I am
currently unable to test the fix in the new kernel. I don't know that I'll have
an opportunity in the near future, so if you'd like to close the bug for now, I
have no objection.