This restores old behaviour, where we end up doing a DRI2GetBuffers()
call from intelMakeCurrent(). The idea was that we could do this
lazily, just before we start rendering. However, if we don't do the
DRI2GetBuffers() round-trip we don't get the drawable size and higher
level mesa ends up short-cutting a number of GL calls, such as glClear().
This way, if we get an invalidate as we update the buffers, we
don't clobber the drawable stamp and ignore the invalidate.
Pointed-out-by: Francisco Jerez
This uses a stamp mechanisms to mark the DRI drawable as invalid.
Instead of immediately updating the buffers we just bump the drawable
stamp and call out to DRI2GetBuffers "later".
"Later" used to be at LOCK_HARDWARE time, and this patch brings back
callouts at the points where we used to call LOCK_HARDWARE. A new function,
intel_prepare_render(), is called where we used to call LOCK_HARDWARE,
and if the buffers are invalid, we call out to DRI2GetBuffers there.
This lets us invalidate buffers only when notified instead of on
every glViewport() call. If the loader calls the DRI invalidate
entrypoint, we disable viewport triggered buffer invalidation.
Additionally, we can clean up the old viewport mechanism a bit,
since we can just invalidate the buffers and not worry about
reentrancy and whatnot.
When a buffer invalidation event is received from the X server, the
"invalidate" hook of the DRI2 flush extension is executed: A generic
implementation (dri2InvalidateDrawable) is provided that just bumps
the "pStamp" sequence number in __DRIdrawableRec.
For old servers not supporting buffer invalidation events, the
invalidate hook will be called before flushing the fake front/back
buffer (that's typically once per frame -- not a lot worse than the
situation we were in before).
No effort has been made on preserving backwards compatibility with
version 2 of the flush extension, but I think it's acceptable because
AFAIK no released stack is making use of it.
Signed-off-by: Kristian Høgsberg <krh@bitplanet.net>
The server side does the throttling on our behalf now by putting the
client to sleep, so we don't need our previous hacks for limiting the
number of outstanding frames. Same effect as
7d4e674b21.
Previously for frame throttling we would wait on the first batch after
a swap before emitting another swap, because we had no hook after a
swap was emitted. This meant that if an app managed to squeeze
everything it for a frame had into one batch, it would lock-step with
the GPU. With the swapbuffers changes, we now have the entrypoint we
want.
This takes the WoW intro screen from 25% GPU idle and visibly jerky to
4-5% GPU idle and rather smooth. Other apps such as OpenArena have
run into this problem as well.
As part of the DRI driver interface rewrite I merged __DRIscreenPrivate
and __DRIscreen, and likewise for __DRIdrawablePrivate and
__DRIcontextPrivate. I left typedefs in place though, to avoid renaming
all the *Private use internal to the driver. That was probably a
mistake, and it turns out a one-line find+sed combo can do the mass
rename. Better late than never.
gen2/3/4 are easier to say than "8xx, 915-945/g33/pineview, 965/g45/misc",
and compares on generation are often easier than stringing together a bunch
of chipset checks.
We only need it when drawing to the front buffer, which we never do for
DRI2. No significant performance difference, but the flush is definitely
gone from the end of every batchbuffer.
_ActualFormat is replaced by Format (MESA_FORMAT_x).
ColorEncoding, ComponentType, RedBits, GreenBits, BlueBits, etc. are
all replaced by MESA_FORMAT_x queries.
Before, if we called glXMakeCurrent() to change a context's window binding
while an FBO was bound, we weren't updating the intel->driDrawable and
intel->driReadDrawable fields. This could cause us to dereference a null
pointer elsewhere.
Since commit 2921a2555d ('intel: Deassociated
drawables from private context struct in intelUnbindContext'),
intel->driDrawable may be NULL in intel_flush().
