Month: December 2005

Fix for smart card startup races

The rules mentioned in my last post to avoid race conditions by handling the entire setup of USB smart card readers within udev and /dev: SUBSYSTEM==”usb_device”, SYSFS{idVendor}==”04e6″, SYSFS{idProduct}==”e003″, GROUP=”scard”, MODE=”0664″ SUBSYSTEM==”usb_device”, SYSFS{bDeviceClass}==”0x0B”, GROUP=”scard”, MODE=”0664″ SUBSYSTEM==”usb_device”, SYSFS{idVendor}==”04e6″, SYSFS{idProduct}==”5115″, GROUP=”scard”, MODE=”0664″ This requires 2.6.14 and an udev that was current for that kernel. With more recent udev […]

Bugs is bugs is bugs

Ingo, I fear that you are overstating the situation m68k is now in. Non release critical bugs do tend to get fixed – if they didn’t there would scarcely be any point in having them in the BTS at all. My guess would be that any difference will be felt mainly when it comes to […]

Smart card readers at startup

Turns out that as documented in Debian bug 336297 it turns out that there is a problem with using a script invoked by udev to modify permissions on USB entries in /proc (at the time I wrote that post it happened to work due to timing issues). An alternative solution is in the works but […]

Debconf 07

Lars, it’s probably worth pointing out that most of the people who are promoting a Debconf in Edinburgh are actually from various places in England. Some of them do appear to be listed on the Volunteers wanted page (and many of the remaining slots look like they mostly need attendees).

Natsemi NAPI patch

I’ve just sent off a patch implemeting NAPI support for the natsemi driver in Linux, origianlly based on one by Harald Welte. For those that don’t know, NAPI provides an API allowing Linux network drivers to use generic interrupt mitigation and work throttling capabilities provided by the kernel, improving the performance of the network drivers.