Dave Airlie <[email protected]> writes: I'm just going to do review here, reading from the git tree diffs.
> randr: add provider object and provider property support (v6) Given that the _rrProvider structs are per-screen, any devPrivate stuff could live in the screen private instead; otherwise, I'd suggest making this a 'real' private structure, which seems like a pain. Remove the typedef for RRProviderSetRoleProcPtr (Hrm. Having NULL for SetProviderOffloadSink and SetProviderOutputSource means returning BadRequest for those; I know they'll just be no-ops for now, but this patch should probably have stubs that return BadMatch instead? We'll let that slide though; bisecting across this probably won't break any apps). (Hrm, would also be nice if this patch created provider objects for the screens so that we could test the protocol support here) Otherwise, this patch is Reviewed-by: Keith Packard <[email protected]> > xfree86: add framework for provider support in ddx. (v3) Need to free 'name' when the screen is deleted. Otherwise, Reviewed-by: Keith Packard <[email protected]> > dix: add unattached list for attaching screens to initially. (v1.1) > dix: attach unbound screens to protocol screen 0 (v2) > randr: expose unattached providers to the clients. > xf86dga: handle DGAAvailable for gpu screens. (v2) > dix: add ability to link output slave gpus to the current gpu (v1.1) > dix: pixmap sharing infrastructure (v3) Reviewed-by: Keith Packard <[email protected]> > exa: pixmap sharing infrastructure (v3) > randr: add initial scanout pixmap support (v3) > xf86: add initial scanout pixmap support (v2) (already reviewed) > randr: fixup constrain to work with slave screens. > randr: fix tellchanged to work with slave scanouts > randr: check the screen size bounds against the master Reviewed-by: Keith Packard <[email protected]> > dix/pixmap: track dirty pixmaps in server. (v3) There are bunch of spurious whitespace fixes in dix/pixmap.c Fails to check DamageCreate return value for NULL PixmapSyncDirtyHelper might be made faster by clipping to the dirty_region instead of iterating over the boxes in it? Setting SourceValidate to NULL is going to pick up the software cursor, I assume this is intentional? A comment might be nice here. Of course, a cleaner fix would be to just have the cursor code drive a cursor on the monitor separately; that way, we'd get hw cursors on the 'main' screen and only have a sw cursor on the output provider. The rest of this patch is Reviewed-by: Keith Packard <[email protected]> > xf86/crtc: add new interface to detach slave crtcs Reviewed-by: Keith Packard <[email protected]> > randr: hook up output slave to screen resources return Calling RRModesForScreen twice is ugly, but it's probably less code than creating a separate function to compute the needed values. It seems like rrGetMultiScreenResources would work fine for screens without connected output providers. If so, please just replace the old function instead of copying it. It'll get more testing that way anyways. > randr: add output source setup Reviewed-by: Keith Packard <[email protected]> > xf86: add output source setting callback Not sure you need to mess with the RootClip of the current master screen when disconnecting the output; you'll get a screen flash as a result. Otherwise: Reviewed-by: Keith Packard <[email protected]> > xf86/cursor: fallback to sw cursor if we have slaves present. Would be better to draw the cursor on the slave than have it rendered to the master and copied to the slave? As you say, all providers can draw X stuff. > xf86: make sure rotate calcs are done on the right screen boundaries Reviewed-by: Keith Packard <[email protected]> -- [email protected]
pgp2zNdmIUe3A.pgp
Description: PGP signature
_______________________________________________ [email protected]: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
