http://bugs.freedesktop.org/show_bug.cgi?id=15203
--- Comment #22 from Roland Scheidegger <[EMAIL PROTECTED]> 2008-04-01 05:23:51 PST --- I fail to see what's wrong with this commit. Clearly we shouldn't update the ring pointer for every command read (which just causes almost as much data traffic for writing the ring pointer as for reading commands at least in some cases), and increasing fetch size should be fine too - I'm sure fglrx does that too. Maybe you could try doing only one or the other to see what's problematic. I suspect though the reason this commit breaks things is somewhat more complicated (e.g. the small delays introduced by writing ring ptr could fix things). RADEON_CP_RB_CNTL is documented in older docs - I guess it's not in the new docs because it's not a 3d reg. There is a section on ring buffer management in the r5xx docs (which is likely exactly the same as for r3xx), though without the register defines it seems. Masks should be unnecessary (but certainly don't hurt) since the values calculated don't exceed the maximum anyway. -- Configure bugmail: http://bugs.freedesktop.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug. ------------------------------------------------------------------------- Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace -- _______________________________________________ Dri-devel mailing list Dri-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
