| Age | Commit message (Collapse) | Author |
|---|
|
The in kernel texture check fails because of both
bit11 flags being set on 16x16 textures. It tuns out
that these bits are still set and not cleared in the
pp_txpitch field of the texture. The attached patch
at least helps for this case on my machine. It clears
the bit 11 from the pitch field if the texture is smaller
and masks out that hight bits on the conventional width
and height field.
Fixes bug 24584
|
|
|
|
The non-KMS interface is to blame here. In theory, a proper fix
could be produced that works for the KMS interface only, but it
require cleaning a lot of mess. Easier to just do it right in r300g.
Signed-off-by: Nicolai Hähnle <nhaehnle@gmail.com>
|
|
Fixes #22741
|
|
Signed-off-by: Nicolai Hähnle <nhaehnle@gmail.com>
|
|
|
|
Signed-off-by: Nicolai Hähnle <nhaehnle@gmail.com>
|
|
The handling is a bit inefficient, unfortunately, but I don't want to make
any intrusive changes for Mesa 7.6.
Signed-off-by: Nicolai Hähnle <nhaehnle@gmail.com>
|
|
Signed-off-by: Nicolai Hähnle <nhaehnle@gmail.com>
|
|
We obviously need to move the code addr register backwards because their may
be overlap.
This bug affected in particular the Compiz water plugin.
Signed-off-by: Nicolai Hähnle <nhaehnle@gmail.com>
|
|
|
|
|
|
These function are aiming to make it very simple to add and keep large amount
of debugging code without having runtime impact in relase builds. Basic idea
is to expose simple printf style debugging functions that are inlined.
Level parameter will be evalueted in compile time so compiler can optimise
some of debugging functions out if compile time request for debug level is too
tight.
|
|
Kernel side doesn't have required register in safe list so can't send it kernel.
|
|
All that state stuff should really be in state atoms :/
|
|
|
|
Problem was to find the correct place to run prediction. Only place that is
called for every primitive is ALLOC_VERTS so we have to do prediction there
before allocation.
|
|
this makes gnome-shell work on r300 for me
|
|
|
|
The fragment program native limits are actually higher on R4xx
as reported by Alex Deucher, but using those requires some
bit-twiddling changes to program handling.
Signed-off-by: Nicolai Hähnle <nhaehnle@gmail.com>
|
|
Signed-off-by: Nicolai Hähnle <nhaehnle@gmail.com>
|
|
Swtcl calls flush everytime primitive changes so prediction has to made again
after flushing.
|
|
Only use the macro variant of MAD when absolutely necessary.
Apparently it cannot deal with relative addressing.
Signed-off-by: Nicolai Hähnle <nhaehnle@gmail.com>
|
|
Find used temporaries even if they are only written to in dead code.
This fixes a bug in the NQSSADCE stage.
Signed-off-by: Nicolai Hähnle <nhaehnle@gmail.com>
|
|
We were check command buffer sizes too alte so allocated dma regions
were freed before relocations so space checking failed.
|
|
|
|
|
|
r600_state_predict
|
|
The s3tc extensions are properly enabled now, when force_s3tc_enable option is set in driconf.
|
|
|
|
Prediction code making too small prediction may cause space check aserttion
failure later in rendering. So warning about any failure to predict correctly
should be fixed.
|
|
Needed for occulsion queries on rv530 chips
Signed-off-by: Alex Deucher <alexdeucher@gmail.com>
|
|
r600_state_predict
|
|
|
|
r600_state_predict
Conflicts:
src/mesa/drivers/dri/r300/r300_cmdbuf.c
src/mesa/drivers/dri/radeon/radeon_cmdbuf.h
|
|
|
|
use properly implemented OUT_BATCH_TABLE where possible
|
|
|
|
|
|
|
|
|
|
All rendering is checked in r300PredictTryDrawPrimsSize which mamde these calls useless.
|
|
Trying to make understanding code easier with small refactoring and renaming.
|
|
|
|
We do flush for cmd buffer in case there isn't enough space left for whole
rendering operation. This protects dma regions from getting released in middle
of state emit.
Signed-off-by: Pauli Nieminen <suokkos@gmail.com>
|
|
|
|
This makes it easier to predict size of next rendering operation so we
can do early flush.
|
|
This fixes some state atom check functions from returing wrong emit size.
There is emit code cleanup so that emit function selection is done in init
time instead of runtime.
Signed-off-by: Pauli Nieminen <suokkos@gmail.com>
|
|
Signed-off-by: Pauli Nieminen <suokkos@gmail.com>
|
|
We keep dma buffer objects in list untill they have been unused for many
draw operations. Current limit of having 100 flushes is just guess for
good performance/memory trade off.
Moving WARN_ONCE macro to common context because it is used in multiple drivers.
Signed-off-by: Pauli Nieminen <suokkos@gmail.com>
|
