ImportWatched kept a copy of every non-zero page of the range being imported, so that a page the
write watch reported during the import could be told from one that still held its bytes. The copy
had the range's footprint: a first import of a 0.5-3 GiB guest heap held that much host memory
while it ran.
The window now keeps a 64-bit digest per 4 KiB page (four xxHash64 lanes) and compares digests:
8 bytes per page, 1/512 of the range. Same decision for every page, same reads of guest memory.
CheckImportWindowFootprint imports a 64 MiB watched range and requires the heap held during the
import to stay under 1 MiB. It failed with "held 64 MiB" before this change.
- Drain the draw pipeline before recording the copy, as the CB metadata
pass does, so the copy follows the queued draws that render the plane.
- Address the plane as decodeDepth registers it: a multisampled plane's
pipe-bank XOR bits are stripped and a DB_DEPTH_VIEW slice is added.
Mipmapped or partially resident planes are refused.
- Refuse a color target with CMASK fast clears, as the CB resolve does.
- Find the source through the depth owner (planeSurface) and require a
32-bit float surface with the decoded sample count.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
A draw with DB_RENDER_CONTROL.DEPTH_COPY copies one sample of the
Z_32_FLOAT depth surface into color target 0. It now runs as a compute
pass that fetches that sample and stores it into the R32 float target,
when the draw covers the whole surface without depth or stencil tests.
With COPY_CENTROID the sample is the first one the draw covers, which is
sample 0 under the full sample mask the driver requires, as radeonsi
programs its per-sample DB to CB copies.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
PA_SU_SC_MODE_CNTL with POLY_MODE set to a point or line type was refused,
and every triangle was drawn filled. The state decode now takes the polygon
type of the faces that are not culled into State::polygonMode, ignores the
types when POLY_MODE is clear or the primitives are points or lines, and
keeps refusing front and back types that differ while both faces are drawn.
Points are drawn one pixel wide (PA_SU_POINT_SIZE 0x00080008 inside
PA_SU_POINT_MINMAX), the size VK_KHR_maintenance5 gives a point whose
shader does not write PointSize, and lines one pixel wide (PA_SU_LINE_CNTL).
The device enables fillModeNonSolid and maintenance5 when they are
available. The polygon mode joins the pipeline, library and draw keys and
the pipeline journal (format 5).
Co-Authored-By: Claude Opus 5.5 <[email protected]>
main 7ce3b1df3e took disk-cache format 43 (#8295) for its own layout, so this
branch's format moves to 44. The request serializer stays at 42: main
still writes 41.
enabling it broke agc_driver_layered_volume_draw, agc_driver_layered_volume_vertex
and agc_driver_subvector_loop on an RX 6700 XT. With #7623 and #7828 on
main, those tests pass with the feature on, in either library-mode order.
This reverts 48b394d1e and keeps the record of #7623's empty-mask test in
the s_subvector_loop debt entry.
Co-Authored-By: Claude <[email protected]>
When a draw's stage fails to prepare, capture or recompile, the packet
worker fails the queue and the next Submit or WaitIdle rethrows the
error, which ends the title. With APS5_SKIP_REFUSED_DRAWS set,
materializeDrawStage rethrows that error as a RefusedDraw; the packet
loop skips the draw, counts it as a thrown skip, logs the shader once
per address and runs the rest of the submission. The frame then lacks
that draw. Compute refusals and errors outside the stage build fail as
before, and without the switch nothing changes.
agc_driver_draw_unsupported_buffer_type_skip_refused submits a draw
whose pixel shader is refused twice with the switch: the WRITE_DATA
after it lands both times and the color target stays untouched.
AI-assisted: yes
The recompiler asked for DenormPreserve only for 16-bit and 64-bit floats,
so a device that flushes 32-bit denormals by default (RADV, MoltenVK) flushed
them whatever the guest's float mode said. target.float32DenormPreserve is set
from shaderDenormPreserveFloat32 when denormBehaviorIndependence allows a 32-bit
mode, is part of the cache key and of the request serialization (version 41),
and makes the module declare DenormPreserve 32 unless the 32-bit flush is
already taken in hardware. APS5_NO_HW_F32_PRESERVE=1 turns it off.
With APS5_DUMP_FRAMES, FrameDumpWriter::Enqueue threw "frame dump queue exceeded 64 MiB" when the
writer thread fell more than 64 MiB behind, which failed the presentation and ended the title.
#4320 made that rare at the default APS5_DUMP_SCALE=4; at APS5_DUMP_SCALE=1 a 1920x1080 title
reaches it within nine frames and a 3840x2160 one on the third.
Enqueue now waits on the writer's condition variable until the frame fits, and the writer
notifies after each file; a single frame larger than the queue still throws. The VideoOut frame
dump queue test gets a wait mode at scale 1 whose third frame has to wait for the held first one.
Co-Authored-By: Claude Fable 5.1 <[email protected]>
The startup probe that lets shaders use the device's f32 denormal flush
instead of the emulated one only ran on devices that declare
shaderDenormFlushToZeroFloat32 with independent denormal behavior.
MoltenVK declares neither flushing nor preserving f32 denormals, so every
shader kept the emulated flush on every f32 add, multiply, fma, min, max
and comparison.
On MoltenVK the probe now runs too. Since the device declares no
DenormFlushToZero capability there, a module compiled for the hardware
flush declares no DenormFlushToZero execution mode and relies on the
native behavior the probe measured; when any probe result differs from
the emulated flush, shaders keep the emulation as before. Devices that
declare the capability are unchanged.
Six vkCreateShaderModule calls added to main since this branch's base
still passed their SPIR-V unchanged, so on AMD's Windows driver they
shared on-disk cache entries across the two pipeline library modes:
splitComputePipeline's recompiled compute programs, the multisample
readback modules, the dispatch-arguments module, the texture repack
module and the two performance overlay modules.
Graphics::CreateShaderModule now keys the generator word and creates the
module in one place. Every module the driver creates goes through it,
including the recompiled graphics and compute programs that copied the
key inline, so new call sites can't skip it.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
blu