MemoryIndexBelongsTo walked every instruction of the program and asked
BufferAccessOf, AddressOpcodeInfoOf and ImageOpcodeInfoOf about each one.
TryMakeTableImage calls it for each of a table handle's eight descriptor
dwords and once more for a material key, and PlanIndirectImages calls
TryMakeTableImage for every image instruction whose handle is not planned
yet, so each such instruction could cost up to nine walks of the whole
program. PlanIndirectImages now indexes the instructions that access
memory by their memory index once, before it plans, and
MemoryIndexBelongsTo looks the index up while that pass runs. The
planning pass adds, removes and changes no instruction, so the index
stays valid until the pass drops it at its end. Outside the pass, where
the function has no caller, it walks the program as before.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Fnrk5Dazjb2tLooVSpdRQq
A sample whose T# is chosen in one if/else and whose S# is chosen in a later one has descriptor Phis in two blocks. The web walk required every Phi of the tuple to sit in one block, so it gave up and the plan threw "GetImageResource dword 0 is not a valid runtime value". The walk now splits at the Phis of the latest block in program order, and Phis of earlier blocks pass through each edge unchanged until the walk reaches their block.
A compute shader in Astro Bot samples an image as 2D while the bound
T# is 1D, and the recompiler refused it. Offer a 1D runtime mode for
sampled images used only by image_sample_lz and mip-less image_load,
which reads the texel the first coordinate selects, as measured on
RDNA2 in #2237.
A 3D image access bound to a 2D T# matched no runtime image mode and
threw "image descriptor is incompatible with the static runtime image
interface" (Sonic Origins, #2023). CouchRiv measured RDNA2 (gfx1036)
with the hardware oracle: image_sample_lz with dim 3D ignores r and
reads the (u,v) texel, and image_load reads the (x,y) texel when z is
0 and returns zero otherwise. Add a 2D mode for 3D accesses whose only
uses are those two forms, and make the load return zero for z != 0.
Every other access keeps throwing.
Fixes#2023
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
When every dst_sel of a sampled view selects constant 0 or 1, a sample
or gather returns that constant whatever the texel, filter, LOD or
coordinates. Such placeholder views can have a dimension the static
runtime image interface has no mode for (a 3D view under a 2D sample,
a 1D view under a 2D gather), and a 3D one paired with an unnormalized
sampler fails the unnormalized proof.
A direct sampled image whose accesses are all non-compare samples or
gathers gets three constant runtime modes (float, uint, sint). The mode
is selected when all four dst_sel are 0 or 1, unless the view is FMASK,
depth bits, converted or sRGB-decoded. In it the emitter returns each
channel's constant (the dmask channel's, in all four texels, for a
gather) from the image's swizzle specialization constants, and the
image is left unbound. ProveUnnormalized skips the mode after the
run-time-selected check, so table roots still throw.
A path whose T# or S# dwords hold a program-counter, EXEC or compare-mask
value is no longer folded onto the first descriptor: the web split gives up
and the read throws as on main. Cache the invariant-Phi resolution per web,
lower the state limit to 512 and test the program-counter, EXEC and
compare-mask rows, unwritten registers, and the arm and state limits.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SplitDescriptorPhi (#414) handles a T# or S# that is a Phi of SRT reads in
one block. A descriptor picked inside a loop through a jump table reaches
the read as a chain of Phis rooted at the loop header, so the plan threw
"GetImageResource dword 0 is not a valid runtime value".
When the one-block split does not apply, walk the descriptor dwords
together through the Phi web, give each distinct SRT descriptor an arm,
build a U32 selector Phi web carrying the arm index, and select between one
copy of the read per arm. Paths whose dwords hold a program-counter, EXEC
or ballot value fold onto the first arm (listed in TechnicalDebt); any
other descriptor that cannot be recomputed still throws.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
image_atomic_* with a two-dword data mask (four for cmpswap) is the
64-bit form of an integer image atomic. The decoder rejected it with
"invalid MIMG data mask". PPSA14632 (Unreal Engine) uses
image_atomic_umax this way in a compute shader during its first frames.
Swap, add, sub, signed and unsigned min/max, and, or, xor and cmpswap
now decode with 64-bit data and translate to new ImageAtomic*64 IR
opcodes. An image updated by them gets its own descriptor binding
range, declared as an R64ui image of 64-bit integers, and the module
uses Int64ImageEXT from SPV_EXT_shader_image_int64. The driver enables
shaderImageInt64Atomics when the device offers it and binds an
R64_UINT view of the 32_32 UINT, SINT or FLOAT surface. Without the
feature, recompiling such a shader throws.
64-bit inc, dec and the float atomics still throw, as does any other
access in the same shader to an image it updates with 64-bit atomics.
The new agc_driver_image_atomics64 test runs every operation on 32_32
UINT, SINT and FLOAT surfaces and checks the returned and stored values
against edge cases across the dword boundary.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The resource tracker refused a program's 33rd distinct V# with "buffer
resource limit exceeded". The 32 was a recompiler constant, not a
Vulkan limit: every buffer is one constant-indexed element of the guest
buffers binding, which ShaderResources checks against the device's
storage-buffer limits.
Drop ShaderInfo::MaxBuffers and the tracker's refusal, and size the
SPIR-V emitter's per-buffer offset list from the binding layout. When
the user data and offsets do not fit in the push constants, they all go
to the shader data buffer as before. ShaderResources counted only
storage buffers against maxPerStageResources (200 on Intel UHD) and now
adds the stage's sampled and storage images.
agc_driver_many_buffers reads and writes 52 and 124 buffers, and
agc_driver_graphics rejects buffers plus an image above the limit.
flat_store_* and global_store_* decoded and translated, but a store whose
segment is not scratch was refused when the SPIR-V was emitted ("writable
FLAT/GLOBAL addresses require GPU ownership tracking").
Such a store now goes through the BDA page table like a store through a
GPU-selected V#: the write lookup needs Write permission, which only a
writable registered range imported in place has, and the written page is
noted for the host to mark GPU-written. Byte and short stores, and dword
stores to an address that is not dword aligned, are stored byte by byte.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Return the raw texel bits: dmask bit i selects texel dword i, dwords past the
texel read 0 and the _sgn forms sign-extend texels narrower than 32 bits
(measured on RDNA2 hardware). Only formats whose bits are exactly recoverable
from the sampled view are accepted (unsigned/signed integer, unorm up to 16
bits, 32-bit float components), on a direct sampled image with identity
swizzle; anything else throws.
Co-Authored-By: Claude <noreply@anthropic.com>
A sampler (or texture) chosen by scalar control flow reaches the image
operation as a Phi of two ReadConst values. The SRT walker only accepts
invariant Phis, so GetHandle threw "GetSamplerResource dword 0 is not a
valid runtime value" and the draw was skipped (G-buffer and HDR draws
in PPSA21564).
Before tracking, a read-only image operation (sample, gather, LOD query,
size query, load) whose descriptor dwords are Phis of one block is
copied once per distinct incoming descriptor. Each copy gets its own
memory entry and descriptor handles whose dwords are the edge's values,
recomputed before the operation (SRT reads, user data, the shader base
and pure uniform arithmetic over them only). The results are selected
by a Bool Phi of the edge that entered the Phi block. The copies run
where the original did, so implicit derivatives are unchanged, and the
host binds every edge's descriptor. Any other shape (a nested Phi, a
dynamic read on an edge, a store or atomic) is left alone and still
throws.
Only the lines this branch added lose their comments; the commits before
keep them, with the reasoning behind each change, for whoever digs into
that code later.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Every ray query returned a miss (a stub). The node test is now the GFX10
(RT IP 1.1) one AMD's GPURT software fallback defines (IntersectCommon.hlsl)
with the TriangleNode1_0 and BoxNode1_0 layouts, as SharpEmu ports it: the T#
(base, size, box grow and sort, triangle return mode) and the node pointer
are run-time values, triangle nodes return t over its denominator with the
barycentric numerators or the triangle id, float16/float32 box nodes the hit
children (clipped slabs, NaN misses, optionally sorted by entry time); an
inactive lane, a null BVH, an out-of-range node or another type returns four
invalid dwords.
The decoder addresses 11 dwords (8 with A16, the directions as halves), the
translation emits ImageBvhIntersectRay with the T#, MakeImageAddress and exec,
and the program takes the BDA path: one probe covers the node and direct loads
read it (a node the table does not map as one range records an Unmapped
fault); EmitBdaDwordReads can read every dword for the per-byte fallback.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A traversal builds the V# of a bottom-level structure from a pointer it loads
from an instance record, and reads it with s_buffer_load: no binding holds
such a V#, and TakeGpuDescriptor took vector raw dword loads and stores only,
so the resource analysis failed on the SGPR chain.
A ReadConstBuffer through a V# the SRT walker cannot compute is a
MemoryInfo::gpuDescriptor access too: the dword at the V#'s base plus the
dword-aligned offset, read through the BDA page table, zero past
STRIDE * NUM_RECORDS bytes (NUM_RECORDS for stride 0).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A V# the SRT walker cannot compute (loaded from a descriptor heap at a
GPU offset, or built by the program) has no binding to hold it. Raw
dword loads and stores of 1-4 dwords through one now decode the V# in the
shader, apply the RDNA2 bounds rules (OOB_SELECT, stride, swizzle,
ADD_TID, format 0) and go through the BDA page table, as Kyty does.
Stores need Write permission, which only ranges that are a writable
allocation's host import get, so a store into a mirror or a copy faults
rather than being lost. The shader notes each written page in an
open-addressed set behind the fault record; once the work completed the
host marks those pages GPU-written, and an overflowing set throws.
Formatted, typed, sub-dword and atomic access through such a V#, and
their use outside compute shaders, still fail.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Dispatch cache with per-key value variants and data-only hits (flat-SRT leaves the recompiler
proves pure are refreshed into the resource template instead of re-captured), dispatch and draw
recipes replaying a proven prepare, register-keyed draw entries, template data refresh, unit
shadows in device-local slabs, 64 KiB block-unit storage tracking with windowed detile, DCC
clear-key proofs, per-class GPU timing, a CPU copy path with batch read tracking, present slot
ring, rate-limited reaps and lazy completion counting, plan-failure and poisoned source memos,
bindless image tables (runtime-indexed descriptor arrays from a CPU-read table) so the
per-frame compute program that failed its plan now compiles. Worker affinity opt-in, per-draw
logs behind APS5_TRACE_DRAWS, wait-timeout reports include the recorder's view of the label.
Fixes a use-after-free behind BDA-miss dispatch snapshots and the CPU-write stamp of
UnchangedSinceCollected. Intro-video stage: 154 -> ~500 presents per 10 s at 4K.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0177gthrWEBBfBu5xJDXgBfo
- s_*_saveexec reads its source before writing the old EXEC into the
destination, so the aliased form (vcc, vcc) computes exec & vcc; the
per-lane exec bit comes from the lane's own bit. Every Bink 2 decoder
kernel uses that form; this was the video corruption.
- buffer atomics add/sub/or/xor with a zero operand become an atomic
load or nothing.
- DescriptorBinding carries bufferWritten/bufferAtomic/imageWritten;
results expose a variantId and share SPIR-V copy-on-write; source
keys use a length plus 64-bit hash; captures can be reused.
- wave-LDS ordering barriers, v_movrels/v_movreld lowering, SMEM with a
VCC base, f16 conversion opcode fix, a per-function SPIRV-Tools pass
list, env-gated IR dumps and probes.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0177gthrWEBBfBu5xJDXgBfo
Adrià
Jan Šlégl
Dean Galvin
Nahuel Genchi
Daan Dehing
Gjon Gojcevic
Adrià Franch
Chad
Petre
boykopovar
alfalfa-star
Celegans12
DotDebian
STL1te
devilismyfriend