On Wed, 2003-09-24 at 17:43, Linus Torvalds wrote: > On Tue, 9 Sep 2003, Michel Dänzer wrote: > > > > Absolutely, otherwise it's a bug. FWIW, the DRM from DRI CVS seems to > > work the same here as the one from the linuxppc-2.5 tree shortly before > > the -test5 merge. The only recent change I see which might cause > > problems is my Radeon AGP => GART cleanup; I tried hard not to break > > anything but may have missed something. > > Ok. I've merged the Radeon bits - they looked clean, but my main machine > no longer has a radeon, so I'll just have to take your word for it > working. > > (Btw, the shadows in tuxracer are still broken on i830. That's always been > true, as far as I can tell. Oh, well..) > > During the merge I noticed that there are big whitespace cleanups to the > sis driver that hide the rest of the changes. Can somebody please stand up > and tell what the changes are, and tell me whether the DRI CVS SiS driver > should be merged into the kernel? > > The new whitespace looks like proper indentation, but I'd like to hear > from somebody that the current DRI CVS is worth merging to wrt SiS. ?
I redid the whitespace in the SiS DRM while working on it. There were other changes -- SiS DRM now works without sisfb (the sisfb usage is dependent on whether CONFIG_FB_SIS is defined when the DRM is compiled). Also, there was an opportunity for userland to panic the kernel by passing in a bogus pointer as far as I could tell (I didn't actually test), which I believe I've fixed. I still feel like the SiS DRM could be much smaller, but I haven't thought about it yet. -- Eric Anholt [EMAIL PROTECTED] http://people.freebsd.org/~anholt/ [EMAIL PROTECTED] ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf _______________________________________________ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
