Custom Query (242 matches)

Filters
 
Or
 
  
 
Columns

Show under each result:


Results (34 - 36 of 242)

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
Ticket Resolution Summary Owner Reporter
#134 fixed Lars' Installer Brendan Oakley
Description

Lars Erdmann wrote a replacement installer for Uniaud. It, and its source, are in a package on Hobbes. It is an improvement over the existing one in that it provides a reliable uninstall. Additionally, he has made the source freely distributable. Mike Greene created makefiles so that it builds with OpenWatcom.

The problem is that the new minstall in eComStation does not support all of the API's used by this installer, so ironically it only works in Warp 4 at present, until minstall is fixed to support it.

The new minstall source is available here: http://svn.netlabs.org/minstall/browser

#136 too-old Build Uniaud32 without DDK Brendan Oakley
Description

This is closely related to Ticket #51.

While not entirely impossible, the DDK is hard to get these days, so it does keep some people from building Uniaud32 who might otherwise.

This makes it extremely difficult to debug Uniaud in the same way that ALSA does, as they rely on testers and bug reporters to build the driver themselves for testing.

It is important to note that under the GPL, we can include the DDK headers as-is, but any replacement code must be GPL, so essentially it must be implemented anew in Uniaud.

#138 too-old Linux version is 2.3.0 Brendan Oakley
Description

When ALSA compiles, it includes different code depending on which version of Linux is reported. At present in Uniaud this is 2.3.0, which, we can be sure, is untested with the latest ALSA code. This could be a source of some problems on the new hardware.

Setting the value to 2.6.0 causes the compile to fail, so clearly there is some work needed to update the Linux code. It should be a matter of setting this and finding the relevant bits in Linux and porting it.

One thing to check with this is the macro used to figure the version number in Uniaud uses a bit shift, whereas I have the distinct impression that Linux uses a direct decimal conversion. So the numbers might only work out the same when all digits are low enough. Just something to verify at the same time this is done.

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
Note: See TracQuery for help on using queries.