Pawel Salek wrote: > > On 2001.10.25 10:25 Daniel T. Chen wrote: > > Actually I just bypassed it by putting in a #define in host.def (thanks > > [EMAIL PROTECTED]) and a few extra twiddles, and it compiled and installed just > > fine. Thanks. > > [...] > > On 25 Oct 2001, Michel [ISO-8859-1] D�nzer wrote: > > > > > Have you tried copying it to $ProjectRoot/lib instead? > > I wonder: why is this not documented? More, why is it not possible to run > a ./configure script that will detect and setup all needed things? IMHO, > Answer "you should know what you are doing" is only an excuse: if a > certain setup algorithm cannot be expressed via a script, there is > somethin wrong with the algorithm. This also makes reproducing the results > more difficult - instead of saying: I did "cvs co dri; ./configure && make > World.LOG && make install" one has to explain "I did cvs co dri, run lndir > ...., edited host.def; oh, I applied extra patches from xxx, then I got an > error during the compilation so I copied binaries from A to B, and finally > I got something". > I am interested in mach64 development (apparently same reason as most > people: a laptop) and could contribute some of my spare time on testing, > but it seems I should start from spending my time on testing the > compilation, not the driver itself...
These are pretty good points. There are a lot of things I'd rather see if we were to build a repository purely for the purpose of developing and distributing dri drivers. At the moment what we have isn't that at all: it's a subtree of the XFree cvs and build structure, and all the descisions have been made by people with somewhat different goals and history. Most of them were made decades ago... Keith _________________________________________________________ Do You Yahoo!? Get your free @yahoo.com address at http://mail.yahoo.com _______________________________________________ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
