[Docs] Add user-guide page for qd.simt.block.* primitives by hughperkins · Pull Request #638 · Genesis-Embodied-AI/quadrants

hughperkins · 2026-05-06T23:06:34Z

Summary

Adds a comprehensive user-guide page for qd.simt.block.* operations as docs/source/user_guide/block.md, plus a section on the closely related qd.simt.grid.memfence() so users picking between a block-scope and a device-scope fence can see both side by side.

Coverage:

block.sync() / block.mem_sync() — barrier-vs-fence semantics, including the Barrier vs fence: the distinction that matters section.
block.sync_all_nonzero / _any_nonzero / _count_nonzero — CUDA-only barrier-and-reduce predicate ops.
block.SharedArray(shape, dtype) — shared-memory allocation.
block.global_thread_idx() / block.thread_idx() — backend-specific thread indexing.
qd.simt.grid.memfence() — device-scope memory fence.

Each op gets a row in the What's available support matrix, a Semantics section, and at least one worked Examples entry.

Notes

The CUDA mem_sync() caveat references the in-flight fix in [CUDA] Map qd.simt.block.mem_sync() to __threadfence_block() #637, so the divergent-branch example is honest about current behavior on main. Once [CUDA] Map qd.simt.block.mem_sync() to __threadfence_block() #637 lands, the caveat in the support-table footnote can be deleted.
Also updates docs/source/user_guide/index.md to add block.md to the SIMT primitives toctree, and docs/source/user_guide/parallelization.md to fix the stale qd.stmt.SharedArray reference and cross-reference the new page.
A separate PR adds the analogous expanded subgroup.md.

Test plan

python python/tools/markdown_link_check.py docs passes (verified locally).
Sphinx HTML build (make html in docs/) succeeds in CI.

Adds docs/source/user_guide/block.md covering every operation currently in python/quadrants/lang/simt/block.py, plus the closely related qd.simt.grid.memfence() (so users picking between block-scope and device-scope fences see both side by side): - block.sync() / block.mem_sync() — barrier vs memory fence, including the section that explains the distinction that matters most. - block.sync_all_nonzero / sync_any_nonzero / sync_count_nonzero (CUDA-only barrier-and-reduce ops). - block.SharedArray(shape, dtype) — shared memory allocation. - block.global_thread_idx() / block.thread_idx() — backend-specific thread indexing helpers. - qd.simt.grid.memfence() — device-scope memory fence. Each op has a What's available row in the support matrix, a Semantics section, and at least one worked Examples entry. The CUDA mem_sync divergent-branch caveat references the in-flight fix in quadrants#637 so the doc's example is honest about current behavior on main. Updates docs/source/user_guide/index.md to add block.md to the SIMT primitives toctree, and docs/source/user_guide/parallelization.md to fix the stale qd.stmt.SharedArray reference and cross-reference the new page.

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: a5b1dc0b55

ℹ️ About Codex in GitHub

Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".

chatgpt-codex-connector · 2026-05-06T23:09:13Z

+        if tid == 0:
+            buf[i // 32] = i // 32
+            qd.simt.grid.memfence()
+            qd.atomic_add(flag, 0, 1)


Fix the atomic_add call in the publish example

In this example flag is declared as a 1-D ndarray, but qd.atomic_add only accepts the writable target and the increment value; the rest of the repo uses indexed targets such as counter[None] or arr[i]. As written, users copying the cross-block publish sample will call atomic_add with three arguments and fail before they can exercise grid.memfence(); this should index the target, e.g. flag[0], or make flag a scalar tensor and use flag[()].

Useful? React with 👍 / 👎.

Project convention: .md docs do not hard-wrap prose lines. Reflow every paragraph in block.md into single lines, matching the existing unwrapped style in subgroup.md / parallelization.md. Code blocks, tables, and bullet lists are unchanged.

github-actions · 2026-05-07T00:33:29Z

Diff coverage: 0% · 0 lines, 0 missing

hughperkins · 2026-05-07T08:46:42Z

+
+There is no built-in grid-wide barrier (cooperative-groups-style); the only way to converge threads across blocks is to finish the kernel and launch a new one.
+
+## Examples


lets remove examples, so I dont have to review them...

hughperkins · 2026-05-07T08:47:23Z

+
+The `memfence()` here ensures that the data write to `buf` becomes visible to other blocks before the atomic increment of `flag` does, so any block reading `flag > N` is guaranteed to see at least `N` published `buf` entries. This is the building block of decoupled-look-back scan and other single-pass multi-block algorithms.
+
+## Performance and portability notes


lets remove this section. less stuff to review/maintain.

Trims block.md down to: intro, support table, barrier-vs-fence overview, per-op semantics, grid-scope fence, and a Related-pages list. The Examples section and the Performance and portability notes section are removed in this revision. Two in-page anchor references that pointed into the now-gone sections are also rewritten to no longer rely on those anchors: - The CUDA support-table footnote previously cited the "memory-fence example" anchor; it now refers to the divergent-branch pattern shown inline in the `block.mem_sync()` semantics subsection. - The `block.thread_idx()` semantics description previously linked to "the examples below"; it now states the `global_thread_idx() % block_dim` substitute pattern inline. No changes to the table, semantics, or grid-scope sections; the inline `block.mem_sync()` divergent-publish snippet is kept since it is part of that op's semantics description, not under the removed Examples heading.

github-actions · 2026-05-07T10:38:19Z

Diff coverage: 0% · 0 lines, 0 missing

alanray-tech

Review Summary

Well-written documentation page that accurately reflects the current source code, including the honest CUDA mem_sync() caveat referencing #637. The "Barrier vs fence" section is especially valuable. The parallelization.md fix for the stale qd.stmt.SharedArray reference is a welcome cleanup.

While reviewing the docs against the source, I found three pre-existing source-code issues (not introduced by this PR). I will file them as separate issues so they can be tracked independently:

global_thread_idx() has an or-precedence bug in block.py:61 -- if arch == _qd_core.cuda or _qd_core.amdgpu: always evaluates to true because _qd_core.amdgpu is a truthy enum value, making the SPIR-V branch on line 63-64 unreachable. The doc describes this as "a control-flow quirk in the dispatch", which is accurate in effect but understates the root cause.
grid.memfence() has no Python-level arch guard -- unlike every other op in block.py, grid.py:memfence() calls the internal directly without checking the arch. The doc says "CUDA only" but a user calling it on a non-CUDA backend would get a less-than-helpful error from the C++ layer rather than a clean ValueError.
sync() error message says "shared_array" -- block.py:19 raises ValueError(f"qd.block.shared_array is not supported for arch {arch}") inside the sync() function; the message should say sync.

Documentation-only nit: the anchor link in parallelization.md (block.md#grid-scope-fence-qdsimtgridmemfence) should be verified against the actual Sphinx-generated anchor to avoid a dead link.

Overall: LGTM as a docs PR. The content is accurate, well-structured, and covers every public op in block.py / grid.py.

Add a rename notice anywhere in block.md that introduces or references mem_sync(), so a reader landing on any section of the page sees the upcoming rename: - A standalone "Naming note:" paragraph immediately under the support table caveat, visible to anyone scanning the table. - Inline parenthetical on the Barrier-vs-fence bullet that introduces mem_sync() in that section. - A bold "**Planned rename: block.mem_fence().**" lead line in the dedicated `### block.mem_sync()` semantics subsection, the most likely jump-target for someone clicking through from a TOC. - Inline parenthetical on the Grid-scope-fence cross-reference. Code samples are intentionally not annotated (annotating inside a code block would alter or break the example); the surrounding prose carries the notice in each case.

Extend the rename annotations added in the previous commit to also cover the upcoming grid.memfence -> grid.mem_fence rename (note the underscore). Same per-section approach: - Generalise the standalone Naming-note paragraph under the support table to list both renames as bullets. - Inline parenthetical on the Barrier-vs-fence section's device-scope cross-reference. - Bold "**Planned rename: qd.simt.grid.mem_fence()**" lead line as the first content of the dedicated Grid-scope fence subsection, mirroring the treatment of block.mem_sync(). The block.mem_sync() notices added previously are unchanged; this commit only adds the grid.memfence() coverage and rewrites the shared Naming-note paragraph to enumerate both renames.

alanray-tech · 2026-05-07T13:48:29Z

OK to merge

github-actions · 2026-05-07T14:33:57Z

Diff coverage: 0% · 0 lines, 0 missing

* [Misc] Warn user to disable caching when print_ir/QD_DUMP_IR enabled (Genesis-Embodied-AI#425) Co-authored-by: v01dxyz <v01dxyz@v01d.xyz> * [Build] Pin torch version to CUDA 12.8 for CUDA tests (Genesis-Embodied-AI#428) * [Misc] Fixing up taichi-dev urls (Genesis-Embodied-AI#429) * [Perf] Rename cuda_graph to gpu_graph across the codebase (Genesis-Embodied-AI#430) * Misc: fix typo integeral -> integral (Genesis-Embodied-AI#434) Co-authored-by: v01dxyz <v01dxyz@v01d.xyz> * [Perf] CUDA graph 4: call from multiple locations (Genesis-Embodied-AI#420) * [Bug] Fix fastcache not restoring graph_do_while_arg (Genesis-Embodied-AI#435) * [Perf] Cache last-call result in perf_dispatch for single-compatible case (Genesis-Embodied-AI#438) * Fix gpu_graph fallback on old Nvidia GPU. (Genesis-Embodied-AI#443) * Fix shared memory offset not reset between CUDA kernels. (Genesis-Embodied-AI#442) * [Misc] Allow disabling GPU graph via QD_GPU_GRAPH=0 env var (Genesis-Embodied-AI#439) * [Misc] Add named top-level loops (Genesis-Embodied-AI#440) * [Misc] Rename gpu_graph to graph (Genesis-Embodied-AI#446) * [Misc] Add cross-platform shuffle (Genesis-Embodied-AI#447) * [Bug] Fix graph_do_while on Windows: search for cudadevrt.lib (Genesis-Embodied-AI#456) * [Bug] Also search default CUDA toolkit install location on Windows (Genesis-Embodied-AI#461) * [SPIRV] Feature Parity Atomics & Shared Array (Genesis-Embodied-AI#432) * [Misc] Change clang format to 120 characters (Genesis-Embodied-AI#463) * [Misc] CUDA graph 5 Add fatbin (Genesis-Embodied-AI#464) * [Bug] Reuse VkInstance across init/reset cycles (Genesis-Embodied-AI#465) * [Perf] Tiles 1: _load, _store, _eye_ (Genesis-Embodied-AI#466) * [Misc] Remove dead InternalFuncStmt type_check override (Genesis-Embodied-AI#471) * [Perf] Tiles 2: add cholesky and ger (Genesis-Embodied-AI#472) * [Perf] Tiles 2b: add triangular solve (Genesis-Embodied-AI#474) * [Misc] Refactor: use _get_col/_set_col in tiles load/store/init (Genesis-Embodied-AI#475) * [Build] Fix flaky test_clock_accuracy (Genesis-Embodied-AI#436) * Fix AARCH64 emitting invalid asm in CUDA kernels. (Genesis-Embodied-AI#473) Co-authored-by: Hugh Perkins <hughperkins@gmail.com> * [AMDGPU] Enable HIP memory pool and surface pool-exhaustion errors. (Genesis-Embodied-AI#485) * [AMDGPU] Scope hsaco tmp dir per-user to avoid collisions. (Genesis-Embodied-AI#484) * [Perf] Tiles 3: Add slice syntax, qd.outer() and initial doc (Genesis-Embodied-AI#477) * [AMDGPU] Fix gradient computation. (Genesis-Embodied-AI#486) * Enable all backends that are supported in unit tests. (Genesis-Embodied-AI#488) * Fix SPIRV ID overflow for large kernels due to autodiff. (Genesis-Embodied-AI#489) * [Misc] Fix purity checker to allow accessing constants from quadrants modules (Genesis-Embodied-AI#487) * [Misc] Increase tolerance for clock monotonic test (Genesis-Embodied-AI#492) * [CI] Serialize api doc workflow (Genesis-Embodied-AI#494) * [CI] Increase tolerance for clock test (Genesis-Embodied-AI#506) * [CI] Increase clock test tolerance to 20% (Genesis-Embodied-AI#509) * [Perf] Add tensor_type parametrization to tile16 tests (Genesis-Embodied-AI#504) * [Perf] Tiles 4b: Migrate tiles16 tests to enable fastcache (Genesis-Embodied-AI#505) * [Perf] Tiles 4c: add Tiles16x16 proxy (Genesis-Embodied-AI#507) * [Perf] Tiles 4d: Consolidate slice error tests using parametrize (Genesis-Embodied-AI#508) * [Perf] Tiles 4: add SharedArray slice support (Genesis-Embodied-AI#482) * [Perf] Tiles 5: add Cholesky benchmark demo (Genesis-Embodied-AI#483) * [Doc] Add user guide page for subgroup shuffle (Genesis-Embodied-AI#512) * [Perf] Implement cross-platform shuffle_down (Genesis-Embodied-AI#510) * [Perf] Add portable subgroup reduce_add and reduce_all_add (Genesis-Embodied-AI#511) * [Perf] Add first warmup config to perf dispatch (Genesis-Embodied-AI#422) * [AutoDiff] Autodiff 1: Add baseline adstack regression test for unary_collections (Genesis-Embodied-AI#500) * [AutoDiff] Autodiff 2: Implement derivative for tan (Genesis-Embodied-AI#501) * [AutoDiff] Autodiff 3: Recompute tanh/exp on the operand in the reverse pass (Genesis-Embodied-AI#502) * [AutoDiff] Autodiff 4: Mark rsqrt as non-linear for adstack promotion (Genesis-Embodied-AI#503) * [AutoDiff] Autodiff 5: Fix adjoint-alloca placement for GlobalLoads outside the current range-for (Genesis-Embodied-AI#496) * [AutoDiff] Autodiff 6: Adstack regression tests (Genesis-Embodied-AI#491) * [AutoDiff] Autodiff 7: Fix header size in AdStackAllocaStmt to match u64 runtime layout (Genesis-Embodied-AI#534) * [AutoDiff] Autodiff 8: Surface LLVM adstack push/pop overflow as a Python exception (Genesis-Embodied-AI#535) * [AutoDiff] Autodiff 9: Guard against LLVM worker-thread stack overflow from large per-task adstack budget (Genesis-Embodied-AI#495) * [AutoDiff] Autodiff 10: Implement adstack for SPIR-V (Genesis-Embodied-AI#490) * [AutoDiff] Autodiff 11: Latent adstack-adjacent fixes (AMDGPU hipFree, flush() keeps ctx_buffers_, always-preallocate) (Genesis-Embodied-AI#536) * [Doc] Add AGENTS.md with instructions for AI agents (Genesis-Embodied-AI#541) * [Bug] Abort kernel execution on assertion failure instead of segfaulting (Genesis-Embodied-AI#419) * [Type] ndarray typing 1: Add eval_str=True to inspect.signature() calls (Genesis-Embodied-AI#411) * [CI] Suppress reportPrivateImportUsage in torch-using files (Genesis-Embodied-AI#552) * [Misc] QD_DUMP_IR dumps to files with the task_id added to the filename (Genesis-Embodied-AI#441) * [Type] ndarray typing 2: Fix NDArray single-arg subscript crash (Genesis-Embodied-AI#412) * [Test] Flush xdist channel before worker exit so test failure reports are visible (Genesis-Embodied-AI#555) * [CI] Reduce test retries on CI from 3 to 1. (Genesis-Embodied-AI#554) * [AutoDiff] Autodiff 12: Heap-backed adstack on LLVM backends (CPU/CUDA/AMDGPU) (Genesis-Embodied-AI#537) * [AutoDiff] Autodiff 13: Heap-backed adstack on SPIR-V backends (Metal, Vulkan) (Genesis-Embodied-AI#493) * [AutoDiff] Autodiff 14: Resolve bounded-inner-loop adstacks without default_ad_stack_size fallback (Genesis-Embodied-AI#539) * [SPIRV] Vulkan SPIR-V correctness: atomic-view aliasing, PSB stride, narrow storage caps, u1 cast, per-init layer recheck (Genesis-Embodied-AI#513) * [Build] Autodiff 15: Replace 2022 MoltenVK pin with LunarG Vulkan SDK fetch and sanitise MoltenVK cap advertisement (Genesis-Embodied-AI#551) * [Test] Suppress stock pytest-timeout to avoid conflict with pytest_hardtle (Genesis-Embodied-AI#557) * [Vulkan] Use SDK validation layer for debugPrintf instead of apt package (Genesis-Embodied-AI#562) * [Test] Fix flaky perf_dispatch tests by increasing work amounts (Genesis-Embodied-AI#559) * [Test] Add --maxfail CLI option to run_tests.py (default 20) (Genesis-Embodied-AI#558) * [CI] Vulkan debug printf fix to address flaky tests (Genesis-Embodied-AI#563) * [Docs] Add a new page to help for first time contributors (Genesis-Embodied-AI#426) Authored-by: v01dxyz <v01dxyz@v01d.xyz> * [AutoDiff] Autodiff 16: Resolve reverse-mode adstack depths per-launch via runtime-evaluated SizeExpr (Genesis-Embodied-AI#543) * Fix: raise error if device memory allocation fails (Genesis-Embodied-AI#451) (Genesis-Embodied-AI#453) Co-authored-by: v01dxyz <v01dxyz@v01d.xyz> Co-authored-by: Hugh Perkins <hughperkins@gmail.com> * [CI] Add CI job to check line wrapping of comments and docs (Genesis-Embodied-AI#564) * [Misc] Add coverage report to PRs, including kernels (Genesis-Embodied-AI#470) * [CI] CI wrap check feeds only diffs to agent (Genesis-Embodied-AI#567) * Skip 'flaky' test on MacOS CI. (Genesis-Embodied-AI#573) * [Test] Fix missing `import sys` in test_fail_device_memory_allocation (Genesis-Embodied-AI#574) * [CI] Fix Vulkan debugPrintf flake with session-scoped warmup (Genesis-Embodied-AI#571) * [AutoDiff] determine_ad_stack_size: replace whole-CFG Bellman-Ford with SCC + DAG DP (Genesis-Embodied-AI#575) * [Test] Fix macOS OOM skip reason to describe actual root cause (Genesis-Embodied-AI#576) * [Lang] whole_kernel_cse: 2.5x compile time speedup on large kernels (Genesis-Embodied-AI#577) * [CI] Add CI check for unnecessarily deleted comments (Genesis-Embodied-AI#570) * [CI] Migrate coverage report to github Check page (Genesis-Embodied-AI#566) * [Lang] Skip IR verifier between passes unless debug=true (Genesis-Embodied-AI#579) * [Lang] Inline AdStack ops on release LLVM codegen: dramatically reduces compile time for adstack-enabled reverse-mode kernels (Genesis-Embodied-AI#584) * [CUDA] Honor offline_cache=False end-to-end so QD_OFFLINE_CACHE=0 actually gives a cold compile (Genesis-Embodied-AI#580) * [Type] Tensor 24 (Genesis-Embodied-AI#561) Co-authored-by: hugh <hugh@slurm-login-0.slurm-login.tenant-slurm.svc.cluster.local> * [Lang] auto_diff host-walk reductions: dramatically faster front-end compile time on adstack-enabled reverse-mode kernels (Genesis-Embodied-AI#587) * [AutoDiff] Speed up reverse-mode kernel launches on GPU backends (Genesis-Embodied-AI#578) * [Vulkan] Move adstack-sizer scratch out of Function-scope memory to fix SPIR-V pipeline build failures (Genesis-Embodied-AI#588) * [AutoDiff] Improve diagnosis of unsupported reverse-mode AD patterns (Genesis-Embodied-AI#590) * [Bug] Fix: promote Ndarray to AnyArray in build_Name for flattened struct fields (Genesis-Embodied-AI#592) * [SPIR-V] Shrink reverse-grad kernel MSL by ~50% (Genesis-Embodied-AI#591) * [CI] Add CI check that PR changes have test coverage (Genesis-Embodied-AI#596) * [Perf] Enable zero-copy in to_torch() and to_numpy() (Genesis-Embodied-AI#450) * Add BufferView: safe sub-range ndarray access for kernels (Genesis-Embodied-AI#585) Co-authored-by: alanray-tech <alanray-tech@users.noreply.github.com> Co-authored-by: Hugh Perkins <hughperkins@gmail.com> * [Doc] Add user-facing fastcache documentation (Genesis-Embodied-AI#597) Co-authored-by: hugh <hugh@slurm-login-0.slurm-login.tenant-slurm.svc.cluster.local> * [Misc] Upgrade to enable v1 dlpack so to_numpy(copy=False) writable (Genesis-Embodied-AI#598) Co-authored-by: root <root@rtx-209-201.slurm-compute.tenant-slurm.svc.cluster.local> * [AutoDiff] Cut reverse-mode adstack memory usage 10x on all backends (Genesis-Embodied-AI#599) * [Misc] Add CI check for feature file factorization (Genesis-Embodied-AI#606) * [Perf] Skip _recursive_set_args for all-Field frozen dataclass structs (Genesis-Embodied-AI#607) Co-authored-by: Cursor <cursoragent@cursor.com> * [AutoDiff] SNode-arm bound-expr capture rejects fold-attack gate indices (Genesis-Embodied-AI#610) * [Misc] Suppress field fastcache warning for qd.Tensor (Genesis-Embodied-AI#615) Co-authored-by: Cursor <cursoragent@cursor.com> * [AutoDiff] Adstack heap: clip reducer count by per-task loop trip count (compile-time and SizeExpr-evaluated) (Genesis-Embodied-AI#611) * [Misc] Forward copy= through qd.Tensor, add copy=None option (Genesis-Embodied-AI#616) Co-authored-by: Cursor <cursoragent@cursor.com> * [Doc] Update README (Genesis-Embodied-AI#617) Co-authored-by: Cursor <cursoragent@cursor.com> * [CI] Fix coverage report showing def lines as uncovered (Genesis-Embodied-AI#623) Co-authored-by: Cursor <cursoragent@cursor.com> * [Perf] Generic launcher: persistent context, JIT-pointer reuse, Metal compute encoder, LLVM-GPU async memory ops (Part 1/2) (Genesis-Embodied-AI#619) * [CI] Encode Python-first testing policy in coverage-check prompt (Genesis-Embodied-AI#622) Co-authored-by: Cursor <cursoragent@cursor.com> * [CI] Add PR Line change report (Genesis-Embodied-AI#624) Co-authored-by: Cursor <cursoragent@cursor.com> * [CI] Disable quadrants pytest plugin during quadrants internal coverage runs (Genesis-Embodied-AI#629) Co-authored-by: Cursor <cursoragent@cursor.com> * [AutoDiff] Adstack load+store eliminations: EliminateRecomputableAdStackPushes pass + leaf extensions (Genesis-Embodied-AI#621) * [CI] Simplify coverage PR comment to a single linked line (Genesis-Embodied-AI#630) * [CUDA] Add AGX Thor, SM_110 (Genesis-Embodied-AI#631) Co-authored-by: Johnny Nunez and Hugh Perkins * [CI] Lines changed report: collapse PR comment to a single linked totals line (Genesis-Embodied-AI#632) * [FEATURE] Support external Metal command queue via qd.init (Genesis-Embodied-AI#618) Co-authored-by: Cursor <cursoragent@cursor.com> * [Perf] Cache adstack-sizer metadata per task across SPIR-V + LLVM-GPU; per-snode / DeviceAllocation invalidation (Part 2/2) (Genesis-Embodied-AI#620) * [AutoDiff] Disable EliminateRecomputableAdStackPushes pending mutated-SNode chain-leaf fix (Genesis-Embodied-AI#633) * [AutoDiff] Adstack chain-clone safety: mutated-SNode leaf reject + load_top consumer-aware guard (Genesis-Embodied-AI#634) * [Docs] Add user-guide page for qd.simt.block.* primitives (Genesis-Embodied-AI#638) * [Docs] Expand qd.simt.subgroup user-guide page to cover every op (Genesis-Embodied-AI#639) * [Perf] Streams 1-4 (Genesis-Embodied-AI#410) * [Docs] Add user-guide page for matrix decompositions and solvers (Genesis-Embodied-AI#643) * [Bug] Revert "[Perf] Streams 1-4 (Genesis-Embodied-AI#410)" (Genesis-Embodied-AI#650) * [Docs] Add user-guide page for atomics and bit operations (Genesis-Embodied-AI#640) * [Docs] Add user-guide page for qd.simt.grid.* primitives (Genesis-Embodied-AI#641) * [AutoDiff] Adstack max-reducer: parallel multi-axis MaxOverRange dispatch (Genesis-Embodied-AI#635) * [AMDGPU] Fix amdgpu parallel rand init (Genesis-Embodied-AI#658) * [Perf] Adstack: skip max-reducer recognizer on CPU + lift host-eval cap (Genesis-Embodied-AI#655) * [Perf] Re-land Streams 1-4 with bug fixes (Genesis-Embodied-AI#653) * [AMDGPU] Apply device_memory_GB=0.3 cap to AMDGPU tests (Genesis-Embodied-AI#659) * [Perf] Per-launch host sync: drop wait_idle on SPIR-V, pin stream and drop stream_synchronize on CUDA/AMDGPU (Genesis-Embodied-AI#654) * [AMDGPU] Unload hipModule_t in JITModuleAMDGPU destructor (Genesis-Embodied-AI#660) * [AMDGPU] Trim default mempool on qd.reset() (Genesis-Embodied-AI#669) * [AMDGPU] Hoist rand-state buffer to process lifetime (Genesis-Embodied-AI#668) * [Streams] Use events for streams serialization on AMDGPU and CUDA (Genesis-Embodied-AI#667) * [Perf] Adstack max-reducer: launch cache + zero-copy result map; content-stable registry_id (Genesis-Embodied-AI#671) * [SPIR-V] dispatch_max_reducers: register each task with the real kernel name (Genesis-Embodied-AI#675) * [AutoDiff] Debug-mode field/grad/dual: dtype, layout, and access-time invariants (Genesis-Embodied-AI#677) * [Docs] Add user-guide page for qd.algorithms.* device-wide algorithms (Genesis-Embodied-AI#642) Co-authored-by: alanray-tech <alan.ray@genesis-ai.company> * [Docs] Doc for existing atomics: switch support table to per-backend columns (Genesis-Embodied-AI#657) Co-authored-by: alanray-tech <alan.ray@genesis-ai.company> * [GPU] Cross gpu atomics (Genesis-Embodied-AI#666) Co-authored-by: alanray-tech <alan.ray@genesis-ai.company> * [GPU] Make block operations portable cross-gpu (Genesis-Embodied-AI#664) * [Perf] CPU LLVM adstack-cache: skip per-launch bump-writes + ndarray_shapes capture on forward-only handles (Genesis-Embodied-AI#685) * [GPU] Cross-GPU for grid ops (Genesis-Embodied-AI#670) * [Math] Make bitop operations portable cross-gpu (Genesis-Embodied-AI#662) * [AMDGPU] Always use wave64, on both RDNA and CDNA (Genesis-Embodied-AI#687) * [AMDGPU] Use syncscope("agent") for atomix xor to avoid CAS livelock (Genesis-Embodied-AI#672) * [GPU] New bit ops for QIPC (Genesis-Embodied-AI#679) * [GPU] Subgroup ops cross-gpu (Genesis-Embodied-AI#665) * [Graph] Rename CUDA Graph to Graph in docs (Genesis-Embodied-AI#691) * [SPIR-V] Fix FIFO-queue ordering when sharing command queue. (Genesis-Embodied-AI#694) * [Atomics] New QIPC ops for atomics (Genesis-Embodied-AI#690) * Pass dataclass sub-structs into qd.func (Genesis-Embodied-AI#698) * [AMDGPU] HIP graph runtime support for @qd.kernel(graph=True) (Genesis-Embodied-AI#692) * [CI] Add per-file timing report to Mac Metal test job (Genesis-Embodied-AI#695) Co-authored-by: Cursor <cursoragent@cursor.com> * [CI] Enable kernel disk cache during tests (Genesis-Embodied-AI#696) * [Math] New QIPC ops for single-threaded linalg (Genesis-Embodied-AI#683) * [BREAKING][GPU] New QIPC ops for subgroups (Genesis-Embodied-AI#676) * [GPU] New QIPC ops for block (Genesis-Embodied-AI#684) * [GPU] New device-level ops for QIPC (Genesis-Embodied-AI#693) * [algorithms] PrefixSumExecutor: drop unused GRID_SZ local (Genesis-Embodied-AI#701) * [block] sync(): fix unsupported-arch error message (Genesis-Embodied-AI#700) * [volatile_load] add qd.volatile_load primitive (closes Genesis-Embodied-AI#648) (Genesis-Embodied-AI#702) * [AutoDiff] Reject recycled identity_key in AdStackCache::register_adstack_sizing_info (Genesis-Embodied-AI#708) * [Vulkan] Declare GroupNonUniform SPIR-V caps and enable shaderSubgroupExtendedTypes (Genesis-Embodied-AI#707) * Fix duplicate HIP graph driver-function declarations after v1.0.0 merge The amd-integration fork had cherry-picked the HIP graph driver functions (graph_create / graph_destroy / graph_add_kernel_node / graph_instantiate / graph_exec_destroy / graph_launch), and upstream v1.0.0 added the same set. The per-file 3-way merge appended both copies into amdgpu_driver_functions.inc.h, producing redeclaration errors that broke the AMDGPU RHI/runtime compile. Drop the upstream duplicate block; the signatures are identical to the fork's existing declarations. Co-authored-by: Cursor <cursoragent@cursor.com> * Fix AMDGPU launcher coherence and num_instructions visibility after v1.0.0 merge - kernel_launcher.cpp: the 3-way merge spliced upstream v1.0.0's launch_llvm_kernel rewrite (ephemeral arg/context buffers, explicit-stream path, AmdgpuDefaultStream PinGuard) onto the AMD fork's kernarg-by-value + persistent-scratch design, leaving references to undefined `ephemeral_context_ptr`. Restore the fork's coherent launch_llvm_kernel verbatim; it calls the (already merged) enhanced launch_offloaded_tasks, which keeps the max-reducer dispatch and stream-parallel groups adapted onto the AMD launch path. - llvm_context.h: both the fork and upstream added `num_instructions`; the merge kept upstream's private placement, but the AMDGPU codegen force-inline heuristic calls it statically from outside the class. Move it back to the public section. Co-authored-by: Cursor <cursoragent@cursor.com> * Restore async result D2H and hoist kernarg vectors in AMDGPU launcher The v1.0.0 merge resolution regressed two amd-integration baseline optimizations in launch_llvm_kernel / launch_offloaded_tasks: - The per-launch result-buffer copy was a blocking memcpy_device_to_host, forcing a host stall on every value-returning launch and serializing the GPU pipeline. Restore the async D2H (the caller synchronizes lazily when it needs the value); external-array transfers still stream_synchronize once before reading back. - launch_task constructed the kernarg std::vectors from initializer lists ({kernarg_payload} / {kernarg_size}) on every dispatch (heap alloc + free per launch). Hoist arg_ptrs/arg_sizes out of the per-task launch and reuse. Co-authored-by: Cursor <cursoragent@cursor.com> * amdgpu: default to LDS permlane64 emulation; drop host-x86 barrier asm on retarget Two AMDGPU JIT-compile crashes surfaced after the v1.0.0 merge pulled in the QIPC subgroup ops (Genesis-Embodied-AI#676), which made the rigid constraint solver's wave-cooperative reductions route through `amdgpu_cross_half_shuffle_i32`. Both manifested as a SIGSEGV inside `llvm::SIInstrInfo::getInstSizeInBytes` during `JITSessionAMDGPU::compile_module_to_hsaco` (i.e. at first kernel launch), and reproduce on gfx942 / MI300X. Baseline 0.4.6 never emitted these constructs, which is why it was unaffected. 1. Native `llvm.amdgcn.permlane64` lowering crashes the bundled LLVM 22.1.0 AMDGPU backend. Default `amdgpu_permlane64` to the existing LDS-roundtrip software emulation on every target (it produces identical results). Add `QD_AMDGPU_USE_NATIVE_PERMLANE64=1` to opt back into the native instruction once the backend bug is fixed; the old `QD_AMDGPU_FORCE_PERMLANE64_FALLBACK` is now the default and still honored. This is the actual crash fix. 2. The runtime module is compiled by the host x86_64 clang and only retargeted to amdgcn here, so `amdgpu_cross_half_shuffle_i32`'s `__asm__ volatile("" : "+v"(byte))` optimization barrier carries x86 flag clobbers (`~{dirflag},~{fpsr},~{flags}`) that are meaningless on AMDGPU. The IR verifies but the empty-body INLINEASM is invalid on the amdgcn target. Neutralize empty-body barrier asm during retarget (forward the tied value, then erase) so no stale host asm reaches codegen. On the wave64 targets we ship `ds_bpermute` already addresses the full wave, so the hint is a no-op. Co-authored-by: Cursor <cursoragent@cursor.com> * style: apply clang-format (v19.1.7) to AMDGPU fn_attrs and launcher sources CI pre-commit's clang-format hook reformatted these files (long declarations/lambda signatures collapsed onto single lines per the repo's clang-format config). Apply the same formatting so the hook passes. No functional changes. Co-authored-by: Cursor <cursoragent@cursor.com> * fix(amdgpu): use CreateNeg for branchless i32 sgn instead of CreateSub(0, input) clang-tidy (modernize-use-nullptr, -warnings-as-errors) flagged `builder->CreateSub(0, input)` in the i32 sgn path: the literal `0` binds to the `llvm::Value*` LHS parameter as a null pointer, not an integer zero. Replace with `builder->CreateNeg(input)`, which emits `0 - input` with a proper zero constant -- identical intended semantics, and clang-tidy clean. Co-authored-by: Cursor <cursoragent@cursor.com> --------- Co-authored-by: Robert Dazi <14996868+v01dXYZ@users.noreply.github.com> Co-authored-by: v01dxyz <v01dxyz@v01d.xyz> Co-authored-by: Hugh Perkins <hughperkins@gmail.com> Co-authored-by: Alexis DUBURCQ <alexis.duburcq@gmail.com> Co-authored-by: hugh <hugh@slurm-login-0.slurm-login.tenant-slurm.svc.cluster.local> Co-authored-by: alanray-tech <alan.ray@genesis-ai.company> Co-authored-by: alanray-tech <alanray-tech@users.noreply.github.com> Co-authored-by: root <root@rtx-209-201.slurm-compute.tenant-slurm.svc.cluster.local> Co-authored-by: Cursor <cursoragent@cursor.com> Co-authored-by: Johnny <johnnynuca14@gmail.com>

hughperkins mentioned this pull request May 6, 2026

[Docs] Expand qd.simt.subgroup user-guide page to cover every op #639

Merged

2 tasks

chatgpt-codex-connector Bot reviewed May 6, 2026

View reviewed changes

[Docs] Unwrap prose in block.md

75cf28b

Project convention: .md docs do not hard-wrap prose lines. Reflow every paragraph in block.md into single lines, matching the existing unwrapped style in subgroup.md / parallelization.md. Code blocks, tables, and bullet lists are unchanged.

hughperkins commented May 7, 2026

View reviewed changes

alanray-tech reviewed May 7, 2026

View reviewed changes

This was referenced May 7, 2026

[Bug] global_thread_idx() or-precedence bug makes SPIR-V branch unreachable #644

Closed

[Bug] grid.memfence() missing Python-level arch guard #645

Closed

[Bug] sync() error message incorrectly says shared_array #646

Closed

hughperkins added 2 commits May 7, 2026 06:10

alanray-tech approved these changes May 7, 2026

View reviewed changes

hughperkins merged commit 331aec2 into main May 7, 2026
52 of 54 checks passed

hughperkins deleted the hp/docs-block branch May 7, 2026 14:50

This comment was marked as off-topic.

Sign in to view

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

[Docs] Add user-guide page for qd.simt.block.* primitives#638

[Docs] Add user-guide page for qd.simt.block.* primitives#638
hughperkins merged 5 commits into
mainfrom
hp/docs-block

hughperkins commented May 6, 2026

Uh oh!

chatgpt-codex-connector Bot left a comment

Uh oh!

chatgpt-codex-connector Bot May 6, 2026

Uh oh!

github-actions Bot commented May 7, 2026

Uh oh!

hughperkins May 7, 2026

Uh oh!

hughperkins May 7, 2026

Uh oh!

github-actions Bot commented May 7, 2026

Uh oh!

alanray-tech left a comment

Uh oh!

alanray-tech commented May 7, 2026

Uh oh!

github-actions Bot commented May 7, 2026

Uh oh!

Uh oh!

This comment was marked as off-topic.

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants


		There is no built-in grid-wide barrier (cooperative-groups-style); the only way to converge threads across blocks is to finish the kernel and launch a new one.

		## Examples


		The `memfence()` here ensures that the data write to `buf` becomes visible to other blocks before the atomic increment of `flag` does, so any block reading `flag > N` is guaranteed to see at least `N` published `buf` entries. This is the building block of decoupled-look-back scan and other single-pass multi-block algorithms.

		## Performance and portability notes

Uh oh!

Conversation

hughperkins commented May 6, 2026

Summary

Notes

Test plan

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

chatgpt-codex-connector Bot May 6, 2026

Choose a reason for hiding this comment

Uh oh!

github-actions Bot commented May 7, 2026

Uh oh!

hughperkins May 7, 2026

Choose a reason for hiding this comment

Uh oh!

hughperkins May 7, 2026

Choose a reason for hiding this comment

Uh oh!

github-actions Bot commented May 7, 2026

Uh oh!

alanray-tech left a comment

Choose a reason for hiding this comment

Review Summary

Uh oh!

alanray-tech commented May 7, 2026

Uh oh!

github-actions Bot commented May 7, 2026

Uh oh!

Uh oh!

This comment was marked as off-topic.

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants