This hasn't been used for a very, very, very long time, (since before
OpenBSD had dri support, for example) and it causes segfaults on dri
drivers when sigio is disabled. Now we don't need to do context swaps on sigio
nor are we trying to do interrupts in userland (thank fuck for that) this
function can die the death that I intended it to die about two years ago, may it
burn.
The kernel support the the sigio ioctl will be removed in a couple of
weeks to give people time to update (right now it accepts it, then
ignores it).
ok kettenis@, matthieu@.
attempt to open that first before trying /dev/ttyC[0-7].
This makes X autoconfiguration a tad bit more intuitive on machines
with multiple SBus or UPA framebuffers, where wsdisplay0 isn't the
console. PCI framebuffers are still busted though.
ok matthieu@
The code to add sunffb unconditionally on !solaris for __sparc__
systems is incorrect for openbsd. More specifically, due to interactions
between hardware drivers and wsfb in preinit we can't unconditionally
add wsfb to the list of fallbacks, so we only add wsfb if no other
option was found. Additionally sunffb does not need to be
unconditionally added because the bus probing code will find these
devices already.
So, long story short: make that code chunk conditional on __sparc__ &&
defined(__linux__) instead.
change from !openbsd to __linux__ requested by kettenis@.
Tested by at least myself and stsp@.
ok matthieu@, kettenis@.
- don't list the dead i810 driver anymore
- blacklist the poulsbo chipset which isn't supported by
xf86-video-intel. Gives vesa a chance. ok oga@, kettenis@.
XAA PixmapOps: Sync before accessing unwrapped callbacks.
When using any XAAPixmapOps, we call into unknown but freshly
unwrapped callbacks (like fb ones). Unlike the XAA*Fallback calls,
we did so without syncing first, exposing us to all kinds of
synchronisation issues.
I believe that the rendering errors appeared now because *PaintWindow
vanished (e4d11e58), and we just use miPaintWindow instead. This
takes a less direct route to the hw and ends up at
PolyFillRectPixmap, which very often left drawing artifacts.
We now sync accordingly, and no longer get the rendering artifacts i
was methodically reproducing on radeonhd, radeon, unichrome...
Also, in order to allow driver authors to remove extensive syncing
or flushing to hide this issue, create XAA_VERSION_ defines, put
them in xaa.h and bump the patchlevel.
tested by naddy@ and Edd Barrett. ok oga@.
PCI address space is completely seperate, so we don't have to claim any
resources to avoid clashes with main memory and firmware. Makes accelerated
graphics work on the Tadpole SPARCLE.
ok matthieu@, oga@
