Since spir-v needs actual bools for its conditional instructions, the
time to do bool properly finally came. As expected, the changes caused
quite a mess, but Ruamoko now does bool/true/false.
Unfortunately, the function isn't used yet, so it's easy to miss. I just
happened to notice things didn't look right when looking at my
projection matrices.
BSP textures have pre-generated mips, including the transparent
textures, but mips don't play nicely with the transparency color (255):
it gets blended to other colors (at least in ad tears). Regenerating the
mips from the rgba texture (using transparent black) produces nice
results.
Also, needed to discard on 0 alpha to avoid weird interactions with fog
(I didn't know ad tears had fog: I guess I had last loaded it before
implementing fog).
One is potentially legit (depends on what really happens when allocating
0-length arrays on the stack), but the offset null pointer ones are
dubious as `a` should never be null when `size` is non-zero, and when
`size` is zero, everything gets corrected.
Switching to native ruamoko bool for vkgen resulted in bool getting
incorrectly parsed as a string (which mean getting the string's address
written to the value). Fixes the randomly enabled debug pipelines (and
probably a few randomly disabled normal pipelines, though less likely as
they're usually implicitly initialized). The one bit of actual legit UB
(ie, that which wasn't immediately mitigated one way or another) found
by ubsan, and it was a data error.
Recent vulkan validation layers started complaining about vertex
processing stages having read-write buffers without atomics enabled
(fair enough), but I decided to get all the buffers that don't need to
be written.
Thanks to validation layers showing command buffer debug regions, it was
pretty easy to find the offending buffers. Did need to modify
QFV_PacketCopyBuffer to take a source barrier as well as the destination
barrier, but this is probably for the best.
Now all pipelines and any tasks that have a command buffer attached get
a region using their names (tasks use the function name). I don't know
when it happened, or if I failed to notice last time, but (sync)
validation layers now include the debug region for command buffers: very
nice.
The particle renderer uses the palette texture in the vertex shader, so
updating the palette needs the vertex shader stage included in the
barrier, but I imagine not all texture updates will need it, so add a
parameter to Vulkan_UpdateTex to select inclusion.
Recent nvidia drivers now crash with more than 17 views when building
pipelines. Still no idea if it's me or the drivers. However, I really
need to come up with a way to use the render graph render passes as a
template so things aren't so hard-coded.
There's still a lot of work needed to separate out quake from
quakeforge, but this lets my test scene get a rather mangled scene
rendering (weird translucency: not sure what I've done wrong: probably
bad clear).
It turns out what I thought was a cascade selection bug was just very
bad choice of cascade steps: factors of 8 just don't work nicely. I'm
not sure that simple factors work all that well, either. I need to make
the cascade system configurable and probably support more cascades.
This puts pixels that have not been rendered at infinity. I was rather
surprised to see fog in my test scene, but it depended on my position
relative to the origin, so something was definitely off (the pixels were
at the origin).
I'm not sure why the final range was only a factor of 4 instead of 8.
There are still issues with range selection, but I'll look into them in
a bit (flying around my little test scene really shows the problems).
While every possible subsystem needs an initialization call, all that
does is add the actual initialization task to the render graph system.
This allows the render graph to be fully configurable, initializing only
those subsystems that the graph needs.
Scripted initialization is still separated from startup as render graph
creation needs various resources (eg, attachments) defined before
creating render and compute passes, but all those need to be created
before the subsystems can actually start up.
Finally. However, it has effect only when no render config is provided.
When a config is provided, things will break currently as nothing is
done yet, but getting a config in will take some work in qwaq and also
the render graph system as I want to make the startup functions
configurable.
The config is a pre-parsed property list. Currently unsupported by
anything but Vulkan (but only a warning is given, not a hard error at
this stage), and Vulkan doesn't use it yet.
Even the comment says it's 8.8, so no need for 32 bits for each value.
It seems to have made a very small improvement to my glsl stub test, but
it's probably just noise (< 0.5%). However, having it "officially" 16
bits means that cached values can be 16 bits thus reducing struct sizes
when I rework lightmap surface data (taking the cache from 16 to 8
bytes).
I had gotten confused about how dynamic lights were calculated and thus
used the wrong radius in the final intensity calculation. Takes care of
the scruffy corners often visible on the dynamic lights.
This gets dynamic lights working again (well, minus the bad updates, I
need to figure out what's up there, but they're nothing new). I guess I
checked only for things running, not that dynamic lights worked.
And clean up the resulting errors. While some were tricky, there weren't
all that many: just some attachment issues and the multi-stage image
copy for scraps.
Fixing scraps required a barrier between copies. It might be overkill,
but a transfer_dst to transfer_dst image barrier worked.
Fixing attachments was a bit trickier:
- depth needed early and late fragment tests to be treated as one stage
- all attachments that were read later needed storeOp = none (using the
extension)
- and then finalLayout needed to be correct to avoid ghost transitions
- as well, for some reason the deffered gbuffer subpass needed a depth
dependency on the translucent pass even though neither one writes to
the depth attachment (possibly a validation bug, needs more
investigation).
It's not perfect (double fog on translucent surfaces, the
scatter/absorption isn't right, and no local lighting on the fog
itself), but it at least seems to look ok.
I think has been one of the biggest roadblocks to breaking free of
quake, so having dual render paths and thus the different new scene load
sequence has proven to be unexpected helpful. There's a lot more to be
done to make the render graph actually usable by anyone but me, but just
making scene load configurable frees up a lot. I think there needs to be
renderer startup/shutdown configuration too, but this seems to be enough
for now.
