A CPU read of DISPATCH_INDIRECT arguments waited only through the flush
hook, which waits for the ranges recorded work noted as pending writes.
An address-based store (FLAT/GLOBAL, or through a descriptor the
recompiler cannot resolve to a static buffer) is not recorded as a
pending write: its batch leases the registered allocations, and only the
guest pin waiter finished that batch. So the thread-dimension,
misaligned and disabled paths, the fill-kernel path and the device's CPU
fallback read the arguments from before an address-based producer ran.
GTA V (PPSA04264) dies loading Story on such a read: a 64-thread clear
kernel read {160, 288, 0xffffffff} as thread dimensions
(3x288x4294967295 groups), bytes its 1-thread setup kernel had not yet
overwritten with {n, 1, 1}, and threw "workgroup count
3x288x4294967295 exceeds device limits".
Every CPU read of dispatch arguments at execution now first finishes the
recorder up to the newest batch recorded with a lease
(Graphics::FinishLeaseWork). The group read-ahead still reads without
waiting; a dispatch it resolved is adopted only when the execution-time
read gives the same counts, and one that differs is dropped and counted
in the [groupcap] line. The device limit error names the argument
address and cpuReason, and the driver adds the queue, path and
read-ahead outcome. The [indirect] read time includes the lease wait,
and the [recorder] sync-site table records the offset of
FinishLeaseWork's caller. The CPU reads that still miss address-based
writes (indirect draw arguments and count, the copy kernel's count
record, the awaits-title checks) are listed in the technical debt.
IndirectArgumentsPendingProducer covers the execution-time CPU argument
reads on compute queues and the pending-label CPU fallback in the Vulkan
device. Its sixth case records a real argument label behind a host Vulkan
event in an earlier batch and an address-based producer in the later open
batch. The event releases only after that producer batch is submitted;
a bounded watchdog releases for cleanup but fails the case. It requires
a deferred producer lease, an unsubmitted producer at the argument
decision, cpuReason 2 and the fresh arguments and counter result. Small
valid stale counts keep a broken-wait control finite on every driver.
Exact CPU-reason and watchdog diagnostics identify the failing path. This
replaces a spinning workload whose earlier label could retire before
the fallback decision on lavapipe, allowing a false green with cpuReason 0.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Co-Authored-By: Codex <noreply@openai.com>
Filled rectangles (VGT_PRIMITIVE_TYPE 7 and 17) use the same single-sample MSAA_ENABLE path as triangles; PPSA03541 stops on a rect-list draw otherwise.
Co-authored-by: Cursor <cursoragent@cursor.com>
RADV initializes and fast-clears DCC keys with its fill_memory compute
kernel, which stores through a global address. Run as an ordinary
dispatch, its stores name no range. The DCC key tracking therefore saw
a key fill only once it had landed in guest memory, which could be
after later writes to the surface had already been recorded.
vkQuake (PPSA99010) hits this on every item pickup:
- the scene renders into a second color buffer;
- RADV's UNDEFINED -> GENERAL barrier fills color buffer 0's keys with
DCC_CLEAR_0000;
- a screen-effects compute pass writes color buffer 0;
- the UI pass draws on color buffer 0.
The keys landed while the UI pass was being recorded. The refresh took
them for a clear made after the compute pass, dropped its results and
cleared the image to black. For the length of the pickup flash the
scene was black, with only the HUD drawn.
The fill HLE now matches the 16-byte variant of the kernel as a fill of
min(threads * 16, user[4] + 16) bytes. The kernel takes its address in
user[2..3], its largest store offset in user[4] and its value in
user[5]. A key fill through it is handled where it is in the command
stream.
Primitive type 19 (DI_PT_QUADLIST, 0x13) is used by Unity and other engines for UI quads, sprites and font glyphs. Because Vulkan lacks native quad-list primitive topology, quad draws are expanded on the CPU into indexed triangle lists with 6 indices per 4-vertex quad, respecting the first/last provoking vertex configuration and discarding incomplete trailing vertices. Adds unit tests and documents the functional debt.
A dispatch over maxComputeWorkGroupCount was split into vkCmdDispatchBase
chunks whose base reached the limit, which VUID-vkCmdDispatchBase-baseGroupX-00421
forbids, and NVIDIA refused the split outright.
Compute modules now declare WorkgroupOffset specialization constants (default 0)
that the recompiler adds to WorkgroupId and GlobalInvocationId and keeps
unmaterialized. Each chunk is recorded with vkCmdDispatch from group 0, bound to
a pipeline variant specialized with the chunk's base. The NVIDIA throw is gone;
device-memory LDS splits still throw.
Fixes#7375
PPSA08270 (Eternights) stopped after its story intro with "AGC graphics:
CB resolve from a target whose sample count differs from the
rasterizer's". It resolves a 4x 512x512 target into a single-sample one
with PA_SC_MODE_CNTL_0.MSAA_ENABLE off while PA_SC_AA_CONFIG still holds
2x from earlier draws. Without MSAA_ENABLE the scan converter does not
use PA_SC_AA_CONFIG's sample count, so the resolve takes its sample
count from CB0 as it already does when PA_SC_AA_CONFIG is single-sample.
Imports, type-3 mirrors and the buffers and textures placed in host memory share the driver's system memory heap. On Windows that heap is half the physical memory, so on a 16 GiB machine the old budget (the heap less 2 GiB) let imports hold about 6 of its 8 GiB, and host buffer allocations then overflowed into the 256 MiB device-local host-visible heap and failed. The budget now reserves the larger of 2 GiB and half the largest host heap.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The custom MSAA sample-location debt entry keeps this branch's grid stand-in text and takes main's sentence on ANCILLARY and the position inputs of a once-per-pixel shader.
Resolves the shader debt file bullet by bullet: main's EmitSetAttribute entry, and this branch's PERSP_SAMPLE/LINEAR_SAMPLE and pixel-position entries.
Repacked copies and depth stencil copies allocated device-local memory
directly and stopped the title on a refusal; they now take the same
allocation path as the other images. Device buffers choose their host
fallback type the same way, the host placement line carries the cumulative
MiB, and the debt file records that a placed image stays in host memory
for its lifetime.
Adrià