Skip to content

feat(rccl/gin-sdma): GIN SDMA Broadcast (-D3 hybrid Anvil-SDMA) - #5

Open
dlamd1dai wants to merge 18 commits into
users/dondai/gin-stage3b-sdma-agfrom
users/dondai/gin-stage3c-sdma-bc
Open

feat(rccl/gin-sdma): GIN SDMA Broadcast (-D3 hybrid Anvil-SDMA)#5
dlamd1dai wants to merge 18 commits into
users/dondai/gin-stage3b-sdma-agfrom
users/dondai/gin-stage3c-sdma-bc

Conversation

@dlamd1dai

@dlamd1dai dlamd1dai commented Aug 29, 2026 •
edited
Loading

Copy link
Copy Markdown
Owner

Summary

Stacked on #4 — adds GIN Anvil-SDMA hybrid Broadcast (-D 3, NCCL_GIN_TYPE=5) on top of the AllGather work in #4.

Broadcast

  • Ported from users/dondai/gin-stage3c-sdma-bcast: gin_sdma_broadcast_policy.h, tiered kernels (LL / LSA / flat / scatter+AG / ring), devtime hook via the shared augment-line scaffold, and broadcast RDC link
  • Host policy unit test (rccl-UnitTestsGinSdmaBroadcastPolicy) and opt-in pytest multi-segment regressions (test_Broadcast.py)
  • GIN CI matrix entry: broadcast_perf -D 3 with NCCL_GIN_TYPE=5
  • Rebased onto latest feat(rccl/gin-sdma): GIN SDMA AllGather Implementation (-D3 hybrid xGMI-SDMA) #4 base (alloc.h host-build fix + AllGather GetDevCommRequirements API alignment)
  • broadcast.cu aligned with common.h testEngine API (GetBw size_t, ncclCommQueryProperties, getCollImplInfo slot)

4 commits on top of #4 base (+ formatting), 7 files changed.

Depends on: #4

Test plan

  • rccl-UnitTestsGinSdmaBroadcastPolicy — broadcast tier policy (host GTest wired in CMakeLists; policy UT not in docker image — validated via tier logic in pytest + SUT gates)
  • broadcast_perf -D 3 with NCCL_GIN_ENABLE=1 NCCL_GIN_TYPE=5 (128 B – multi-segment) — PASS on 8× MI355X (smci355-ccs-aus-m03-17): BC-C1 host baseline NP=8 128M; BC-C2 GIN NP=8 128M #wrong=0
  • pytest projects/rccl-tests/test/test_Broadcast.py -k gin_sdma (opt-in) — includes 2 GiB + 4 GiB hang guards; SUT BC-C2-L PASS at 4G (~373 GB/s, #wrong=0, root=0)
  • GIN CI: broadcast_perf entry in gin-tests.json via run-gin-ci.sh — pending CI re-run after rebase

SUT evidence (MI355X, docker rccl-gin-gda-sdma-713:latest, COLLECTIVE=bcast rebuild)

# Formal gate (128M)
RUN_POLICY_UT=0 bash ddai-artifacts/scripts/gin-sdma-bcast-test.bash 8 128M  → PASS

# Large GIN (4G)
RUN_BC_C2=0 RUN_HOST_BASELINE=0 RUN_BC_C2_LARGE=1 \
  bash ddai-artifacts/scripts/gin-sdma-bcast-test.bash 8 128M  → PASS (~373 GB/s)

dlamd1dai and others added 12 commits August 28, 2026 19:56
@dlamd1dai @cursoragent
rccl-tests/gin-sdma: add Anvil-SDMA hybrid AllGather (-D 3)
860c4b4 
Port the GIN Anvil-SDMA hybrid AllGather kernel into stage3b, at parity
with the AllReduce path: per-rank chunk size gates LSA-direct peer stores
(<= sdmaThreshold) vs. all-peers GIN copy-engine puts (> threshold).

- all_gather.cu: GinHybridAllGatherKernel (NCCL_GIN_TYPE=5) + host launcher;
  sizing/threshold/tier/bandwidth logic factored into a testable policy header.
- gin_sdma_allgather_policy.h: pure host/device policy helpers (single source
  of truth for the LSA<->SDMA crossover) shared by kernel and unit tests.
- rccl/test: host policy unit test + on-GPU tier-predicate/addressing test.
- gin_plugin_anvil_sdma.cc: honor explicit NCCL_GIN_ANVIL_SDMA_THRESHOLD=0.
- common.cu: don't report host cputime column for device-impl kernels.

Validated on 8x MI355X across 128 B-128 MB: 0 wrong in both tiers, crossover
lands exactly at 2 MiB/rank, SDMA tier ~390 GB/s bus bw.

Co-authored-by: Cursor <cursoragent@cursor.com>
(cherry picked from commit ffe0d60)
@dlamd1dai @cursoragent
rccl-tests/gin-sdma: add AllGather multi-segment regression tests
9db0239 
Mirror the AllToAll GIN-SDMA >1 GiB hang-fix tests (PR ROCm#9927) for AllGather so
the multi-channel/SDMA ordering concern is covered by an automated test, not
just manual runs. Drives GinHybridAllGatherKernel (-D 3, NCCL_GIN_TYPE=5) with
the SDMA tier forced (threshold=0) at per-rank chunks that cross the <=128 MiB
segmentation boundary (256 MiB -> 2 seg, 2 GiB -> 16 seg), plus a 2 GiB-total
completion guard with a subprocess timeout that turns a reintroduced SDMA hang
into a failure. Exact-integer dtypes + built-in data check (-c 1) so a truncated
/stale tail can't hide under float tolerance.

Opt-in via RCCL_TESTS_GIN_SDMA_AG (needs an 8x MI355X-class node + MPI + a
GIN-SDMA build); config mirrors test_AllToAll.py (NP/launcher/xenv/exe/timeout).
Includes the harness's intermittent gfx950 cuMem-VMM connectivity-gate retry so
the transient abort doesn't flake the test; a genuine data-check mismatch is
never retried.

Verified on 8x MI355X: 2 GiB-total hang guard passes, 0 wrong, ~425 GB/s
(SDMA/SIMPLE tier, 2x128 MiB segments).

Co-authored-by: Cursor <cursoragent@cursor.com>
(cherry picked from commit 38b741a)
@dlamd1dai @cursoragent
rccl-tests/gin-sdma: per-collective AllGather SDMA threshold
bf1b836 
Give the GIN Anvil-SDMA hybrid AllGather its own LSA<->SDMA crossover knob,
NCCL_GIN_ANVIL_SDMA_THRESHOLD_ALLGATHER, decoupled from the backend gin.put
inline-vs-copy-engine threshold (the global NCCL_GIN_ANVIL_SDMA_THRESHOLD /
ctx->sdmaThreshold). The device API is untouched: the threshold is host-resolved
and passed to GinHybridAllGatherKernel as a kernel argument, consumed only by the
host-testable tier predicate (chunkUsesLsaTier); gin.put's signature is unchanged.

- gin_sdma_allgather_policy.h: replace the context-based resolveSdmaThreshold()
  with a pure, precedence-based pickSdmaThreshold() (per-collective env > global
  env > compiled default) plus the 32 KiB/rank compiled default constant.
- all_gather.cu: host AllGatherResolveSdmaThreshold() reads the per-collective
  env, falls back to the global env then the compiled default, parsed as 64-bit
  (explicit 0 honored -> all-SDMA tier; >2 GiB values do not wrap). Thread the
  resolved value through a new AllGather-specific launch helper into the kernel;
  guard the RunColl -D 3 case with ENABLE_DEVICE_API so a device-API-off build
  cleanly falls through to testNotImplemented (matches the shared launcher's
  prior behavior).
- gin_sdma_allgather_policy_test.cpp: cover pickSdmaThreshold precedence,
  explicit-0, and the 64-bit no-wrap path.

Validated: 13/13 host policy unit tests; docker device compile + A2A and
AllGather GIN smoke gates pass on 8x MI355X (0 wrong).

Co-authored-by: Cursor <cursoragent@cursor.com>
(cherry picked from commit 212d3fc)
@dlamd1dai @cursoragent
rccl-tests/gin-sdma: size-adaptive CTA selection for AllGather (-D 3)
1be099b 
AllGather launched a fixed deviceCtaCount (-V) grid for all sizes, which over-
subscribes the GIN-put/SDMA tier: only nRanks threads issue the puts (the copy
engines move the bytes), so extra CTAs are pure barrier/signal overhead. Mirror
the ReduceScatter design and self-select a size-adaptive CTA count decoupled from
-V, keyed off the same LSA<->SDMA tier predicate the kernel evaluates:
  * LSA-direct tier (chunk <= threshold): 16 CTAs (grid-stride store scales with
    threads; peaks ~16 on 8x MI355X, tiny sizes are latency-bound/CTA-indifferent).
  * GIN-put/SDMA tier (chunk > threshold): 4 CTAs (stable near-peak 1 MiB-128 MiB).

Measured gains vs the old fixed -V 32 launch: 512 KiB +21%, 2 MiB +18%, 8 MiB
+13%, 32 MiB +5%; LSA tier unchanged. Datacheck bit-exact 128 B-256 MiB; A2A/RS
unaffected. Add allGatherCtas()/allGatherMaxCtas() to the policy header (host-
unit-testable), NCCL_GIN_ANVIL_AG_CTAS to pin a fixed count (diagnostic), an
explicit-grid launcher, and size the devComm barrier/signal pool to
max(deviceCtaCount, allGatherMaxCtas()) so the self-selected grid never exceeds it.

Co-authored-by: Cursor <cursoragent@cursor.com>
(cherry picked from commit e5ce620)
@dlamd1dai @cursoragent
rccl-tests/gin-sdma: port AllGather device-time (devtime) hook from s…
d099936 
…tage3e

Bring the in-kernel wall_clock64 device-timing hook (AllGatherDeviceTime, opt-in
via NCCL_GIN_ANVIL_DEVICE_TIMING) to the AG branch, reconciled with the size-
adaptive CTA selection just cherry-picked here. To share one implementation
between the production and timed launches, factor the LSA/SDMA logic out of
GinHybridAllGatherKernel into ginAllGatherBody<T>() and add GinHybridAllGather-
TimedKernel that loops the body under one persistent launch. The devtime path
self-selects the same allGatherCtas() count as the perf path (verified: the
resolved per-collective sdmaThreshold so the timed tier matches the launched cfg.

common.h already carries the testColl.deviceTime field and gin_sdma_devtime.h
exists on this branch (used by AllToAll), so only all_gather.cu changes: include
the scaffold, add the body/timed kernel, add AllGatherDeviceTime, and wire it as
the testColl.deviceTime member. Builds clean; datacheck bit-exact; devtime prints.

Co-authored-by: Cursor <cursoragent@cursor.com>
(cherry picked from commit 4435b34)
@dlamd1dai @cursoragent
rccl-tests/gin-sdma: drop ReduceScatter cross-references from AllGather
1c664a2 
Reword the size-adaptive CTA comments so the AllGather design does not reference
the ReduceScatter design (this is the AG branch). No functional change.

Co-authored-by: Cursor <cursoragent@cursor.com>
(cherry picked from commit 3ae778c)
@dlamd1dai @cursoragent
rccl-tests/gin-sdma: unit-test the AllGather size-adaptive CTA ladder
12ab993 
Cover the new allGatherCtas()/allGatherMaxCtas() policy helpers: LSA vs SDMA tier
selection keyed off the crossover predicate, tracking of a moved threshold, env
pin + 128 clamp + 0-falls-back-to-ladder, and max-covers-both-tiers. 18/18 host
policy tests pass.

Co-authored-by: Cursor <cursoragent@cursor.com>
(cherry picked from commit e35adb7)
@dlamd1dai @cursoragent
rccl/gin-sdma: address AllGather PR #4 review feedback
983612b 
Resolves the seven review comments on #4 (GIN SDMA AllGather):

- rccl-tests: clamp NCCL_GIN_ANVIL_AG_CTAS to the launched barrier/signal pool
  (allGatherPoolCtas = max(-V, ladder peak)); a pin above the pool previously
  launched more CTAs than slots -> OOB on the blockIdx.x-indexed pools.
- rccl: fold Anvil-SDMA conn-check infra failures (write/verify kernel, D2H
  copy, slot-reset memset) into the collective decision instead of a per-rank
  goto, so a single-rank failure aborts all ranks together rather than leaving
  peers hung on bootstrapBarrier/allgather.
- rccl: scope the gfx950 cuMem free-skip to the peer-aperture unmap in
  ncclCuMemFreeAddr only; ncclCuMemFree (physical handle) and ncclCudaFree
  (ordinary buffers) now free normally, so long-lived procs no longer leak on
  every ncclCommDestroy. Leak is confined to the peer VA aperture.
- rccl: replace __builtin_unreachable() in the ANVIL_SDMA/ROCSHMEM_GDA get()
  specializations with __builtin_trap() so a future get() fails loudly. Uses
  the compiler builtin (no HIP runtime include) so it also compiles in the
  symmetric gensrc TU that pulls these headers via nccl_device.h.
- rccl-tests: share allGatherRecvSliceOffset() between AllGatherLsaDirect and
  the GPU addressing test so a wrong slice formula is caught, not just
  self-consistent; scope the test comment to what it does/doesn't cover.
- rccl: fix the rocshmem-gda template test stub to define atomic_add /
  atomic_add_single (was stale atomic_nofetch, broke --rocshmem-gin builds).
- rccl-tests: replace tautological policy asserts with a behavioral
  "grid never exceeds pool" regression guarding the CTA-pin clamp.

Validated on 8x MI355X (gfx950): --rocshmem-gin image builds clean; host policy
unit tests pass (AllGather 19/19 incl. the new clamp+grid guards, SDMA 5/5);
all_gather_perf -D3 (NCCL_GIN_TYPE=5) 128 B-128 MiB is bit-exact (#wrong=0) in
both the LSA-direct and SDMA tiers with no hang.

Co-authored-by: Cursor <cursoragent@cursor.com>
@dlamd1dai @cursoragent
rccl-tests/gin-sdma: drive AllGather device-time via shared CLI flags
a30179d 
Replace stale NCCL_GIN_ANVIL_AG_DEVTIME_* env vars with the harness
--device_timing/-L/-P tier knobs (matching A2A), gate to -D 3, and return
the -1.0 mode-2 sentinel when no measurement is produced.

Co-authored-by: Cursor <cursoragent@cursor.com>
@dlamd1dai @cursoragent
rccl/gin-sdma: port conn-check hang fix from NCCL 2.30.7 wip; add AG …
984c4fa 
…sync check

Add LSA signal connectivity self-test (localFail collective abort) and the
gin_anvil_sdma_oss7_device launchers so the PR branch matches the wip branch's
plugin safety path. Include the AllGather manifest sync script used to keep the
six core AG files identical across gin-stage3b-sdma-ag and the NCCL 2.30.7 wip.

Co-authored-by: Cursor <cursoragent@cursor.com>
@dlamd1dai @cursoragent
rccl/gin-sdma: drop cross-branch AG sync tooling from PR branch
4a2af89 
Remove the AllGather manifest and sync-check scripts; that validation
tooling belongs on the NCCL 2.30.7 wip branch only. The conn-check
plugin fix from the prior commit is unchanged.

Co-authored-by: Cursor <cursoragent@cursor.com>
@dlamd1dai @cursoragent
rccl/gin-sdma: port upstream devtime scaffold and GIN unit test templ…
12975a5 
…ates

Reconcile PR #4 with merged upstream ROCm#10658 (A2A devtime host helpers,
BenchTime skip/augment/check path) and ROCm#10672 (GDA/SDMA device template
tests + GIN CI wiring). AllGather devtime now buffers via devtimeAugmentLine
to match the A2A augment pattern; AllGather-specific unit test targets remain.

Co-authored-by: Cursor <cursoragent@cursor.com>
@dlamd1dai
dlamd1dai force-pushed the users/dondai/gin-stage3c-sdma-bc branch from 04450da to abd91c5 Compare August 29, 2026 00:28
@dlamd1dai dlamd1dai changed the title feat(rccl/gin-sdma): GIN SDMA AllGather + Broadcast (-D3 hybrid Anvil-SDMA) feat(rccl/gin-sdma): GIN SDMA Broadcast (-D3 hybrid Anvil-SDMA) Aug 29, 2026
@dlamd1dai
dlamd1dai changed the base branch from develop to users/dondai/gin-stage3b-sdma-ag August 29, 2026 00:29
@github-actions

github-actions Bot commented Aug 29, 2026 •
edited
Loading

Copy link
Copy Markdown

rocprofiler-systems CI Summary

Test Results (for commit a31c234)

Workflow jobs covered: 35

Workflow Runs Avg Test Time Tests Skipped Failed
Ubuntu 22.04 5 30m 33s 2,570 100 0
Ubuntu 24.04 5 32m 23s 2,590 110 0
Debian 4 37m 00s 2,168 78 0
RHEL 12 36m 08s 6,168 277 0
System deps 9 29m 19s 4,822 197 0
Total 35 33m 09s 18,318 762 0

Failed Tests

No failed tests reported.

Sanitizers

Thread Undefined Address
Passed Passed Passed

@github-actions

github-actions Bot commented Aug 29, 2026

Copy link
Copy Markdown

✅ MINOR ABI BREAKAGE resolved - ABI compliance check is now passing!

@github-actions

github-actions Bot commented Aug 29, 2026

Copy link
Copy Markdown

Code Coverage Report

Code Coverage Report

Unit Tests

Code Coverage: unit-tests

Lines: 🔴 23.1% — 8,148/35,268 across 321 files
Functions: 🔴 25.2% — 922/3,653

🔴 0-20% (191 files)
Lines Functions File
0.0% 0/10 0/2 build/coverage/_deps/nlohmann_json-src/include/nlohmann/byte_container_with_subtype.hpp
0.0% 0/1 0/1 build/coverage/_deps/spdlog-src/include/spdlog/details/log_msg_buffer.h
0.0% 0/1 0/1 build/coverage/_deps/spdlog-src/include/spdlog/sinks/ansicolor_sink.h
0.0% 0/2 0/1 build/coverage/include/profiler-hub/shared_types.hpp
0.0% 0/105 0/9 source/bin/common/argument_registration.hpp
0.0% 0/254 0/28 source/bin/common/tool_runner.cpp
0.0% 0/61 0/6 source/bin/rocprof-sys-avail/avail.hpp
0.0% 0/258 0/12 source/bin/rocprof-sys-avail/generate_config.cpp
0.0% 0/102 0/6 source/bin/rocprof-sys-avail/hw_counter_query.cpp
0.0% 0/446 0/47 source/bin/rocprof-sys-causal/impl.cpp
0.0% 0/80 0/2 source/bin/rocprof-sys-causal/rocprof-sys-causal.cpp
0.0% 0/673 0/46 source/bin/rocprof-sys-instrument/details.cpp
0.0% 0/91 0/7 source/bin/rocprof-sys-instrument/function_signature.cpp
0.0% 0/20 0/5 source/bin/rocprof-sys-instrument/function_signature.hpp
0.0% 0/158 0/7 source/bin/rocprof-sys-instrument/info.hpp
0.0% 0/264 0/22 source/bin/rocprof-sys-instrument/internal_libs.cpp
0.0% 0/64 0/6 source/bin/rocprof-sys-instrument/log.cpp
0.0% 0/5 0/1 source/bin/rocprof-sys-instrument/log.hpp
0.0% 0/533 0/41 source/bin/rocprof-sys-instrument/module_function.cpp
0.0% 0/93 0/7 source/bin/rocprof-sys-instrument/module_function.hpp
0.0% 0/1536 0/87 source/bin/rocprof-sys-instrument/rocprof-sys-instrument.cpp
0.0% 0/157 0/12 source/bin/rocprof-sys-instrument/rocprof-sys-instrument.hpp
0.0% 0/2 0/1 source/bin/rocprof-sys-run/rocprof-sys-run.cpp
0.0% 0/2 0/1 source/bin/rocprof-sys-sample/rocprof-sys-sample.cpp
0.0% 0/24 0/1 source/lib/backends/amd_smi/wrapper.hpp
0.0% 0/16 0/4 source/lib/binary/address_multirange.cpp
0.0% 0/9 0/1 source/lib/binary/address_multirange.hpp
0.0% 0/117 0/5 source/lib/binary/analysis.cpp
0.0% 0/7 0/1 source/lib/binary/binary_info.hpp
0.0% 0/102 0/9 source/lib/binary/dwarf_entry.cpp
0.0% 0/1 0/1 source/lib/binary/dwarf_entry.hpp
0.0% 0/69 0/6 source/lib/binary/link_map.cpp
0.0% 0/3 0/1 source/lib/binary/link_map.hpp
0.0% 0/10 0/1 source/lib/binary/scope_filter.cpp
0.0% 0/5 0/2 source/lib/binary/scope_filter.hpp
0.0% 0/158 0/18 source/lib/binary/symbol.cpp
0.0% 0/8 0/8 source/lib/binary/symbol.hpp
0.0% 0/9 0/2 source/lib/common/env_vars.hpp
0.0% 0/40 0/3 source/lib/common/invoke.hpp
0.0% 0/150 0/10 source/lib/common/md5sum.hpp
0.0% 0/22 0/2 source/lib/common/setup.hpp
0.0% 0/3 0/1 source/lib/common/units.hpp
0.0% 0/166 0/2 source/lib/core/agent_info.hpp
0.0% 0/80 0/18 source/lib/core/binary/address_range.cpp
0.0% 0/9 0/3 source/lib/core/binary/address_range.hpp
0.0% 0/49 0/10 source/lib/core/categories.cpp
0.0% 0/11 0/11 source/lib/core/components/fwd.hpp
0.0% 0/14 0/2 source/lib/core/containers/aligned_static_vector.hpp
0.0% 0/11 0/3 source/lib/core/containers/c_array.hpp
0.0% 0/1 0/1 source/lib/core/containers/operators.hpp
0.0% 0/46 0/12 source/lib/core/containers/stable_vector.hpp
0.0% 0/23 0/5 source/lib/core/containers/static_vector.hpp
0.0% 0/89 0/6 source/lib/core/cpu.cpp
0.0% 0/1 0/1 source/lib/core/cpu.hpp
0.0% 0/53 0/6 source/lib/core/dynamic_library.cpp
0.0% 0/28 0/7 source/lib/core/exception.cpp
0.0% 0/160 0/12 source/lib/core/gpu_metrics.cpp
0.0% 0/33 0/6 source/lib/core/mpi.hpp
0.0% 0/24 0/2 source/lib/core/node_info.cpp
0.0% 0/2 0/2 source/lib/core/node_info.hpp
0.0% 0/79 0/5 source/lib/core/perf.cpp
0.0% 0/51 0/5 source/lib/core/perfetto/cached_perfetto_session.cpp
0.0% 0/33 0/4 source/lib/core/perfetto/counter_track.hpp
0.0% 0/29 0/7 source/lib/core/perfetto/emitter.cpp
0.0% 0/48 0/9 source/lib/core/perfetto/emitter.hpp
0.0% 0/7 0/1 source/lib/core/perfetto/sinks/append_mode.hpp
0.0% 0/13 0/3 source/lib/core/perfetto/sinks/tee_sink.cpp
0.0% 0/1 0/1 source/lib/core/perfetto/sinks/tee_sink.hpp
0.0% 0/1 0/1 source/lib/core/progress/bar.hpp
0.0% 0/8 0/3 source/lib/core/sdk/tracing-config-deps.hpp
0.0% 0/54 0/3 source/lib/core/trace_cache/cache_manager.cpp
0.0% 0/7 0/3 source/lib/core/trace_cache/cache_manager.hpp
0.0% 0/516 0/71 source/lib/core/trace_cache/metadata_registry.cpp
0.0% 0/764 0/103 source/lib/core/trace_cache/perfetto_processor.cpp
0.0% 0/119 0/12 source/lib/core/trace_cache/post_processor.cpp
0.0% 0/817 0/32 source/lib/core/trace_cache/rocpd_processor.cpp
0.0% 0/77 0/1 source/lib/core/trace_cache/sample_processor.hpp
0.0% 0/14 0/6 source/lib/rocprof-sys-causal-api/causal_api.cpp
0.0% 0/539 0/85 source/lib/rocprof-sys-dl/dl.cpp
0.0% 0/16 0/1 source/lib/rocprof-sys-dl/main.c
0.0% 0/280 0/31 source/lib/rocprof-sys-rt/src/RTcommon.c
0.0% 0/29 0/4 source/lib/rocprof-sys-rt/src/RTheap-linux.c
0.0% 0/65 0/4 source/lib/rocprof-sys-rt/src/RTheap.c
0.0% 0/212 0/22 source/lib/rocprof-sys-rt/src/RTlinux.c
0.0% 0/97 0/8 source/lib/rocprof-sys-rt/src/RTposix.c
0.0% 0/6 0/2 source/lib/rocprof-sys-rt/src/RTsignal.c
0.0% 0/11 0/2 source/lib/rocprof-sys-rt/src/RTthread-x86-64.c
0.0% 0/3 0/1 source/lib/rocprof-sys-rt/src/RTthread.c
0.0% 0/84 0/20 source/lib/rocprof-sys/api.cpp
0.0% 0/620 0/36 source/lib/rocprof-sys/library.cpp
0.0% 0/77 0/6 source/lib/rocprof-sys/library/causal/components/backtrace.cpp
0.0% 0/7 0/2 source/lib/rocprof-sys/library/causal/components/backtrace.hpp
0.0% 0/93 0/11 source/lib/rocprof-sys/library/causal/components/blocking_gotcha.cpp
0.0% 0/33 0/7 source/lib/rocprof-sys/library/causal/components/causal_gotcha.cpp
0.0% 0/85 0/22 source/lib/rocprof-sys/library/causal/components/progress_point.cpp
0.0% 0/23 0/3 source/lib/rocprof-sys/library/causal/components/progress_point.hpp
0.0% 0/44 0/8 source/lib/rocprof-sys/library/causal/components/unblocking_gotcha.cpp
0.0% 0/553 0/33 source/lib/rocprof-sys/library/causal/data.cpp
0.0% 0/87 0/16 source/lib/rocprof-sys/library/causal/delay.cpp
0.0% 0/373 0/29 source/lib/rocprof-sys/library/causal/experiment.cpp
0.0% 0/3 0/2 source/lib/rocprof-sys/library/causal/experiment.hpp
0.0% 0/24 0/5 source/lib/rocprof-sys/library/causal/sample_data.cpp
0.0% 0/268 0/25 source/lib/rocprof-sys/library/causal/sampling.cpp
0.0% 0/4 0/1 source/lib/rocprof-sys/library/causal/selected_entry.cpp
0.0% 0/5 0/3 source/lib/rocprof-sys/library/causal/selected_entry.hpp
0.0% 0/71 0/11 source/lib/rocprof-sys/library/components/backtrace.cpp
0.0% 0/1 0/1 source/lib/rocprof-sys/library/components/backtrace.hpp
0.0% 0/215 0/30 source/lib/rocprof-sys/library/components/backtrace_metrics.cpp
0.0% 0/11 0/2 source/lib/rocprof-sys/library/components/backtrace_metrics.hpp
0.0% 0/9 0/3 source/lib/rocprof-sys/library/components/backtrace_timestamp.cpp
0.0% 0/4 0/2 source/lib/rocprof-sys/library/components/backtrace_timestamp.hpp
0.0% 0/71 0/10 source/lib/rocprof-sys/library/components/callchain.cpp
0.0% 0/1 0/1 source/lib/rocprof-sys/library/components/callchain.hpp
0.0% 0/367 0/30 source/lib/rocprof-sys/library/components/comm_data.cpp
0.0% 0/19 0/5 source/lib/rocprof-sys/library/components/comm_data.hpp
0.0% 0/8 0/2 source/lib/rocprof-sys/library/components/ensure_storage.hpp
0.0% 0/28 0/6 source/lib/rocprof-sys/library/components/exit_gotcha.cpp
0.0% 0/1 0/1 source/lib/rocprof-sys/library/components/exit_gotcha.hpp
0.0% 0/68 0/8 source/lib/rocprof-sys/library/components/fork_gotcha.cpp
0.0% 0/1 0/1 source/lib/rocprof-sys/library/components/fork_gotcha.hpp
0.0% 0/13 0/3 source/lib/rocprof-sys/library/components/kill_gotcha.cpp
0.0% 0/1 0/1 source/lib/rocprof-sys/library/components/kill_gotcha.hpp
0.0% 0/158 0/22 source/lib/rocprof-sys/library/components/mpi_gotcha.cpp
0.0% 0/2 0/1 source/lib/rocprof-sys/library/components/mpi_gotcha.hpp
0.0% 0/533 0/12 source/lib/rocprof-sys/library/components/mpip.hpp
0.0% 0/78 0/19 source/lib/rocprof-sys/library/components/numa_gotcha.cpp
0.0% 0/2 0/1 source/lib/rocprof-sys/library/components/numa_gotcha.hpp
0.0% 0/336 0/20 source/lib/rocprof-sys/library/components/pthread_create_gotcha.cpp
0.0% 0/3 0/1 source/lib/rocprof-sys/library/components/pthread_create_gotcha.hpp
0.0% 0/34 0/9 source/lib/rocprof-sys/library/components/pthread_gotcha.cpp
0.0% 0/24 0/2 source/lib/rocprof-sys/library/components/pthread_mutex_gotcha_policy.hpp
0.0% 0/164 0/31 source/lib/rocprof-sys/library/components/vaapi_gotcha.cpp
0.0% 0/2 0/1 source/lib/rocprof-sys/library/components/vaapi_gotcha.hpp
0.0% 0/145 0/10 source/lib/rocprof-sys/library/coverage.cpp
0.0% 0/29 0/5 source/lib/rocprof-sys/library/coverage.hpp
0.0% 0/60 0/16 source/lib/rocprof-sys/library/coverage/impl.hpp
0.0% 0/252 0/37 source/lib/rocprof-sys/library/kokkosp.cpp
0.0% 0/245 0/31 source/lib/rocprof-sys/library/perf.cpp
0.0% 0/11 0/3 source/lib/rocprof-sys/library/perf.hpp
0.0% 0/186 0/12 source/lib/rocprof-sys/library/pmc/collectors/common/settings.hpp
0.0% 0/66 0/5 source/lib/rocprof-sys/library/pmc/collectors/cpu/cache_policy.hpp
0.0% 0/47 0/6 source/lib/rocprof-sys/library/pmc/collectors/cpu/cpu_traits.hpp
0.0% 0/52 0/9 source/lib/rocprof-sys/library/pmc/collectors/cpu/perfetto_policy.hpp
0.0% 0/100 0/6 source/lib/rocprof-sys/library/pmc/collectors/gpu/cache_policy.hpp
0.0% 0/222 0/36 source/lib/rocprof-sys/library/pmc/collectors/gpu/perfetto_policy.hpp
0.0% 0/25 0/3 source/lib/rocprof-sys/library/pmc/collectors/gpu/sample.hpp
0.0% 0/32 0/3 source/lib/rocprof-sys/library/pmc/collectors/gpu_perf_counter/cache_policy.hpp
0.0% 0/39 0/7 source/lib/rocprof-sys/library/pmc/collectors/gpu_perf_counter/perfetto_policy.hpp
0.0% 0/16 0/2 source/lib/rocprof-sys/library/pmc/collectors/nic/sample.hpp
0.0% 0/14 0/2 source/lib/rocprof-sys/library/pmc/device_providers/procfs/provider.hpp
0.0% 0/148 0/18 source/lib/rocprof-sys/library/pmc/sampler.cpp
0.0% 0/113 0/11 source/lib/rocprof-sys/library/process_sampler.cpp
0.0% 0/13 0/1 source/lib/rocprof-sys/library/process_sampler.hpp
0.0% 0/2 0/1 source/lib/rocprof-sys/library/rocm.cpp
0.0% 0/1268 0/88 source/lib/rocprof-sys/library/rocprofiler-sdk.cpp
0.0% 0/111 0/13 source/lib/rocprof-sys/library/rocprofiler-sdk/counters.cpp
0.0% 0/17 0/6 source/lib/rocprof-sys/library/rocprofiler-sdk/counters.hpp
0.0% 0/129 0/9 source/lib/rocprof-sys/library/rocprofiler-sdk/fwd.cpp
0.0% 0/34 0/7 source/lib/rocprof-sys/library/rocprofiler-sdk/fwd.hpp
0.0% 0/36 0/9 source/lib/rocprof-sys/library/rocprofiler-sdk/marker_writer.cpp
0.0% 0/136 0/13 source/lib/rocprof-sys/library/rocprofiler-sdk/rccl.cpp
0.0% 0/112 0/10 source/lib/rocprof-sys/library/rocprofiler-sdk/roctx_client.cpp
0.0% 0/118 0/20 source/lib/rocprof-sys/library/runtime.cpp
0.0% 0/2 0/1 source/lib/rocprof-sys/library/runtime.hpp
0.0% 0/994 0/65 source/lib/rocprof-sys/library/sampling.cpp
0.0% 0/145 0/31 source/lib/rocprof-sys/library/thread_data.hpp
0.0% 0/10 0/3 source/lib/rocprof-sys/library/thread_data_growth.hpp
0.0% 0/10 0/1 source/lib/rocprof-sys/library/thread_deleter.cpp
0.0% 0/5 0/1 source/lib/rocprof-sys/library/thread_deleter.hpp
0.0% 0/161 0/27 source/lib/rocprof-sys/library/thread_info.cpp
0.0% 0/2 0/1 source/lib/rocprof-sys/library/thread_info.hpp
0.0% 0/113 0/11 source/lib/rocprof-sys/library/tracing.cpp
0.0% 0/109 0/30 source/lib/rocprof-sys/library/tracing.hpp
0.0% 0/3 0/1 source/lib/rocprof-sys/library/tracing/annotation.cpp
0.0% 0/45 0/4 source/lib/rocprof-sys/library/tracing/annotation.hpp
0.0% 0/8 0/2 source/lib/rocprof-sys/progress.cpp
0.0% 0/74 0/16 source/lib/rocprof-sys/regions.cpp
0.0% 0/82 0/4 source/lib/rocprof-sys/timeout.cpp
0.0% 0/440 0/45 source/python/libpyrocprofsys.cpp
0.6% 1/166 1/13 source/lib/core/perfetto.cpp
0.8% 7/859 1/74 source/lib/core/argparse.cpp
6.1% 2/33 1/3 source/lib/core/amd_smi.cpp
7.1% 1/14 0/1 build/coverage/_deps/nlohmann_json-src/include/nlohmann/detail/iterators/primitive_iterator.hpp
9.6% 17/178 4/25 source/lib/core/constraint.cpp
13.2% 30/227 8/25 source/bin/rocprof-sys-avail/common.cpp
15.4% 4/26 0/6 source/lib/core/trace_cache/metadata_registry.hpp
18.0% 32/178 6/23 source/lib/core/gpu.cpp
18.2% 4/22 1/3 source/lib/common/static_object.hpp
18.8% 9/48 2/5 source/lib/core/output_file_registry.cpp
19.0% 4/21 0/2 source/lib/backends/rocprofiler_sdk/wrapper.hpp
19.4% 6/31 2/5 source/bin/common/common_utils.hpp
🟠 20-50% (14 files)
Lines Functions File
20.0% 1/5 0/2 source/lib/core/argparse.hpp
21.3% 13/61 3/9 source/lib/core/utility.hpp
22.2% 2/9 0/6 source/lib/core/constraint.hpp
23.2% 13/56 1/4 source/lib/core/mproc.cpp
32.1% 52/162 7/17 build/coverage/_deps/nlohmann_json-src/include/nlohmann/detail/input/json_sax.hpp
34.8% 147/423 14/46 build/coverage/_deps/fmt-src/include/fmt/format.h
35.3% 612/1732 156/302 source/lib/core/config.cpp
36.5% 23/63 6/12 source/lib/core/agent_manager.cpp
38.2% 13/34 1/4 source/lib/core/perfetto/sinks/io_helpers.hpp
40.3% 195/484 14/17 build/coverage/_deps/nlohmann_json-src/include/nlohmann/detail/input/lexer.hpp
40.5% 17/42 5/12 build/coverage/_deps/nlohmann_json-src/include/nlohmann/detail/exceptions.hpp
46.0% 433/941 10/48 source/bin/rocprof-sys-avail/avail.cpp
47.4% 222/468 17/39 build/coverage/_deps/fmt-src/include/fmt/base.h
49.6% 127/256 14/32 source/lib/rocprof-sys/library/components/category_region.hpp
🟡 50-80% (35 files)
Lines Functions File
50.0% 153/306 9/11 build/coverage/_deps/nlohmann_json-src/include/nlohmann/detail/output/serializer.hpp
50.0% 1/2 0/1 source/lib/rocprof-sys/library/pmc/collectors/nic/types.hpp
51.9% 220/424 25/37 source/bin/common/common_utils.cpp
52.0% 13/25 2/3 source/lib/rocprof-sys/library/pmc/collectors/gpu_perf_counter/gpu_perf_counter_traits.hpp
53.3% 8/15 3/3 source/lib/core/perfetto/sinks/per_pid_file_sink.cpp
55.0% 22/40 4/6 build/coverage/_deps/fmt-src/include/fmt/ranges.h
55.6% 5/9 1/3 build/coverage/_deps/fmt-src/include/fmt/chrono.h
55.7% 34/61 5/8 source/lib/rocprof-sys/library/pmc/collectors/cpu/sample.hpp
58.3% 14/24 3/8 build/coverage/_deps/nlohmann_json-src/include/nlohmann/detail/output/output_adapters.hpp
58.3% 7/12 0/4 build/coverage/_deps/spdlog-src/include/spdlog/common.h
60.0% 3/5 1/1 source/lib/core/agent.hpp
62.1% 18/29 1/1 source/lib/core/perfetto/locked_file_append.cpp
62.1% 41/66 7/8 source/lib/rocprof-sys/library/pmc/collectors/base/collector.hpp
62.8% 49/78 4/5 source/lib/core/trace_cache/discovery.cpp
63.1% 70/111 7/10 source/lib/core/progress/bar.cpp
63.4% 272/429 33/40 build/coverage/_deps/nlohmann_json-src/include/nlohmann/json.hpp
64.1% 336/524 26/40 source/bin/common/json_config.cpp
65.4% 34/52 4/4 build/coverage/_deps/nlohmann_json-src/include/nlohmann/detail/conversions/from_json.hpp
65.4% 17/26 2/3 source/lib/core/config.hpp
66.2% 47/71 5/5 source/lib/rocprof-sys/library/pmc/collectors/gpu/gpu_traits.hpp
66.4% 87/131 4/5 build/coverage/_deps/nlohmann_json-src/include/nlohmann/detail/input/parser.hpp
66.7% 2/3 0/2 source/lib/rocprof-sys/library/pmc/collectors/cpu/types.hpp
67.2% 43/64 11/14 source/lib/core/trace_cache/data_types.cpp
68.4% 39/57 4/6 source/lib/core/perfetto.hpp
68.8% 22/32 3/8 source/lib/rocprof-sys/library/pmc/collectors/gpu_perf_counter/types.hpp
71.1% 69/97 8/8 build/coverage/_deps/nlohmann_json-src/include/nlohmann/detail/iterators/iter_impl.hpp
71.9% 220/306 15/17 source/bin/common/preset_registry.cpp
72.3% 86/119 14/17 source/lib/core/perfetto/engine.cpp
73.7% 28/38 6/7 build/coverage/_deps/nlohmann_json-src/include/nlohmann/detail/iterators/iteration_proxy.hpp
75.0% 3/4 0/1 source/lib/backends/amd_smi/nic_types.hpp
75.0% 3/4 1/2 source/lib/core/perfetto/sinks/single_file_sink.hpp
76.5% 78/102 4/4 source/lib/core/perfetto/sinks/single_file_sink.cpp
77.8% 7/9 0/1 source/bin/rocprof-sys-avail/info_type.hpp
77.8% 77/99 10/11 source/lib/core/perfetto/engine_impl.hpp
79.3% 88/111 13/15 source/lib/common/path.hpp
🟢 80-100% (81 files)
Lines Functions File
80.0% 8/10 1/2 build/coverage/include/profiler-hub/writer_types.hpp
80.5% 62/77 9/10 source/lib/rocprof-sys/library/pmc/device_providers/rocprofiler_sdk/provider.hpp
82.8% 48/58 6/7 build/coverage/_deps/nlohmann_json-src/include/nlohmann/detail/conversions/to_json.hpp
83.9% 135/161 12/16 source/lib/backends/procfs/backend.hpp
84.6% 11/13 2/4 source/lib/rocprof-sys/library/rocprofiler-sdk/roctx_client.hpp
85.0% 85/100 11/15 source/lib/logger/logger.hpp
85.1% 40/47 2/2 source/lib/core/trace_cache/storage_parser.hpp
86.9% 213/245 15/16 build/coverage/_deps/nlohmann_json-src/include/nlohmann/detail/conversions/to_chars.hpp
89.0% 276/310 30/35 source/lib/common/environment.hpp
89.1% 49/55 4/4 source/lib/rocprof-sys/library/pmc/collectors/gpu_perf_counter/device.hpp
89.1% 221/248 23/33 source/lib/core/sdk/tracing-config.hpp
89.5% 34/38 5/8 build/coverage/_deps/nlohmann_json-src/include/nlohmann/detail/input/input_adapters.hpp
90.0% 54/60 4/4 source/lib/core/trace_cache/buffer_storage.cpp
90.1% 100/111 8/8 source/lib/rocprof-sys/library/pmc/collectors/cpu/device.hpp
90.7% 303/334 28/31 source/lib/core/trace_cache/sample_type.hpp
90.9% 10/11 1/2 build/coverage/_deps/nlohmann_json-src/include/nlohmann/detail/json_ref.hpp
91.0% 243/267 19/20 source/lib/core/trace_cache/unified_memory_processor.cpp
91.3% 178/195 14/18 source/lib/rocprof-sys/library/components/shmem_gotcha.hpp
91.7% 11/12 2/2 source/lib/core/perfetto/session_backend.hpp
91.7% 44/48 4/4 source/lib/core/utility.cpp
92.6% 163/176 9/9 source/lib/rocprof-sys/library/pmc/collectors/gpu/device.hpp
92.9% 13/14 0/1 source/lib/rocprof-sys/library/pmc/collectors/common/collector_slice.hpp
93.0% 40/43 6/11 source/lib/core/demangler.hpp
93.2% 68/73 7/8 source/lib/core/trace_cache/buffer_storage.hpp
94.1% 16/17 3/3 source/bin/rocprof-sys-avail/component_categories.hpp
94.2% 129/137 29/32 source/lib/backends/rocprofiler_sdk/backend.hpp
94.7% 36/38 7/7 build/coverage/_deps/spdlog-src/include/spdlog/logger.h
94.7% 54/57 6/8 source/lib/rocprof-sys/library/components/pthread_mutex_gotcha.hpp
94.8% 55/58 1/5 source/lib/core/trace_cache/cacheable.hpp
94.8% 92/97 9/10 source/lib/rocprof-sys/library/rocprofiler-sdk/trace_control.cpp
95.0% 19/20 4/6 build/coverage/_deps/nlohmann_json-src/include/nlohmann/detail/string_concat.hpp
95.7% 22/23 4/6 source/lib/common/synchronized.hpp
95.8% 23/24 2/4 source/lib/common/pci_bdf.hpp
95.9% 162/169 19/22 source/lib/rocprof-sys/library/components/ucx_gotcha.hpp
96.2% 25/26 1/2 source/lib/common/delimit.hpp
96.5% 82/85 11/12 source/lib/core/state.hpp
96.8% 30/31 3/4 source/lib/rocprof-sys/library/pmc/collectors/gpu_perf_counter/sample.hpp
97.3% 73/75 3/3 source/lib/core/perfetto/packet_framing.hpp
97.4% 74/76 6/6 source/lib/core/common_types.hpp
97.5% 39/40 6/7 source/lib/core/trace_cache/rocpd_helpers.hpp
97.8% 45/46 7/7 source/lib/rocprof-sys/library/rocprofiler-sdk/rccl_internal.hpp
97.9% 47/48 3/3 source/lib/rocprof-sys/library/pmc/collectors/nic/device.hpp
98.1% 51/52 7/7 source/lib/rocprof-sys/library/pmc/device_providers/amd_smi/provider.hpp
98.7% 78/79 15/16 source/lib/backends/amd_smi/backend.hpp
100.0% 14/14 1/1 build/coverage/_deps/fmt-src/include/fmt/ostream.h
100.0% 6/6 0/3 build/coverage/_deps/nlohmann_json-src/include/nlohmann/adl_serializer.hpp
100.0% 2/2 0/1 build/coverage/_deps/nlohmann_json-src/include/nlohmann/detail/input/position_t.hpp
100.0% 2/2 0/1 build/coverage/_deps/spdlog-src/include/spdlog/details/backtracer.h
100.0% 2/2 0/1 build/coverage/_deps/spdlog-src/include/spdlog/details/circular_q.h
100.0% 1/1 0/1 build/coverage/include/profiler-hub/reader_types.hpp
100.0% 1/1 0/1 source/bin/common/json_config.hpp
100.0% 39/39 2/2 source/bin/rocprof-sys-avail/get_availability.hpp
100.0% 10/10 1/1 source/bin/rocprof-sys-avail/get_categories.hpp
100.0% 4/4 1/1 source/bin/rocprof-sys-avail/info_type.cpp
100.0% 107/107 13/14 source/lib/backends/amd_smi/device.hpp
100.0% 7/7 0/1 source/lib/backends/amd_smi/gpu_types.hpp
100.0% 1/1 0/1 source/lib/backends/rocprofiler_sdk/types.hpp
100.0% 12/12 1/1 source/lib/common/join.hpp
100.0% 5/5 1/1 source/lib/common/string_utility.hpp
100.0% 8/8 0/2 source/lib/common/version.hpp
100.0% 2/2 1/2 source/lib/core/agent_manager.hpp
100.0% 1/1 0/1 source/lib/core/output_file_registry.hpp
100.0% 1/1 0/1 source/lib/core/perfetto/category_registry.hpp
100.0% 1/1 0/1 source/lib/core/perfetto/engine.hpp
100.0% 7/7 0/1 source/lib/core/perfetto/locked_file_append.hpp
100.0% 9/9 2/3 source/lib/core/perfetto/sinks/polymorphic_sink_view.hpp
100.0% 6/6 2/2 source/lib/core/perfetto/sinks/recording_sink.cpp
100.0% 5/5 0/2 source/lib/core/perfetto/sinks/recording_sink.hpp
100.0% 6/6 2/2 source/lib/core/progress/tracker.cpp
100.0% 1/1 0/1 source/lib/core/progress/tracker.hpp
100.0% 4/4 0/3 source/lib/core/trace_cache/data_types.hpp
100.0% 9/9 3/3 source/lib/core/trace_cache/type_registry.hpp
100.0% 24/24 4/5 source/lib/core/trace_cache/unified_memory_processor.hpp
100.0% 5/5 2/2 source/lib/core/track_registry.cpp
100.0% 4/4 0/2 source/lib/core/track_registry.hpp
100.0% 10/10 1/1 source/lib/logger/logger.cpp
100.0% 9/9 0/2 source/lib/rocprof-sys/library/pmc/collectors/gpu/types.hpp
100.0% 20/20 4/4 source/lib/rocprof-sys/library/pmc/collectors/nic/perfetto_policy.hpp
100.0% 2/2 0/2 source/lib/rocprof-sys/library/pmc/common/types.hpp
100.0% 34/34 3/4 source/lib/rocprof-sys/library/rocprofiler-sdk/marker_writer.hpp
100.0% 3/3 1/2 source/lib/rocprof-sys/library/rocprofiler-sdk/trace_control.hpp

Distribution

Range Files
0-20% 191
20-50% 14
50-80% 35
80-100% 81

E2E Tests

Code Coverage: e2e-tests

Lines: 🟡 64.3% — 22,674/35,268 across 321 files
Functions: 🟡 64.2% — 2,347/3,653

🔴 0-20% (49 files)
Lines Functions File
0.0% 0/10 0/2 build/coverage/_deps/nlohmann_json-src/include/nlohmann/byte_container_with_subtype.hpp
0.0% 0/1 0/1 build/coverage/_deps/spdlog-src/include/spdlog/details/log_msg_buffer.h
0.0% 0/1 0/1 build/coverage/_deps/spdlog-src/include/spdlog/sinks/ansicolor_sink.h
0.0% 0/2 0/1 build/coverage/include/profiler-hub/shared_types.hpp
0.0% 0/102 0/6 source/bin/rocprof-sys-avail/hw_counter_query.cpp
0.0% 0/24 0/1 source/lib/backends/amd_smi/wrapper.hpp
0.0% 0/166 0/2 source/lib/core/agent_info.hpp
0.0% 0/89 0/6 source/lib/core/cpu.cpp
0.0% 0/1 0/1 source/lib/core/cpu.hpp
0.0% 0/53 0/6 source/lib/core/dynamic_library.cpp
0.0% 0/28 0/7 source/lib/core/exception.cpp
0.0% 0/160 0/12 source/lib/core/gpu_metrics.cpp
0.0% 0/13 0/3 source/lib/core/perfetto/sinks/tee_sink.cpp
0.0% 0/1 0/1 source/lib/core/perfetto/sinks/tee_sink.hpp
0.0% 0/8 0/3 source/lib/core/sdk/tracing-config-deps.hpp
0.0% 0/6 0/2 source/lib/rocprof-sys-rt/src/RTsignal.c
0.0% 0/11 0/2 source/lib/rocprof-sys-rt/src/RTthread-x86-64.c
0.0% 0/3 0/1 source/lib/rocprof-sys-rt/src/RTthread.c
0.0% 0/1 0/1 source/lib/rocprof-sys/library/components/exit_gotcha.hpp
0.0% 0/1 0/1 source/lib/rocprof-sys/library/components/kill_gotcha.hpp
0.0% 0/164 0/31 source/lib/rocprof-sys/library/components/vaapi_gotcha.cpp
0.0% 0/52 0/9 source/lib/rocprof-sys/library/pmc/collectors/cpu/perfetto_policy.hpp
0.0% 0/100 0/6 source/lib/rocprof-sys/library/pmc/collectors/gpu/cache_policy.hpp
0.0% 0/222 0/36 source/lib/rocprof-sys/library/pmc/collectors/gpu/perfetto_policy.hpp
0.0% 0/25 0/3 source/lib/rocprof-sys/library/pmc/collectors/gpu/sample.hpp
0.0% 0/32 0/3 source/lib/rocprof-sys/library/pmc/collectors/gpu_perf_counter/cache_policy.hpp
0.0% 0/39 0/7 source/lib/rocprof-sys/library/pmc/collectors/gpu_perf_counter/perfetto_policy.hpp
0.0% 0/16 0/2 source/lib/rocprof-sys/library/pmc/collectors/nic/sample.hpp
0.0% 0/2 0/1 source/lib/rocprof-sys/library/rocm.cpp
0.0% 0/111 0/13 source/lib/rocprof-sys/library/rocprofiler-sdk/counters.cpp
0.0% 0/17 0/6 source/lib/rocprof-sys/library/rocprofiler-sdk/counters.hpp
0.0% 0/129 0/9 source/lib/rocprof-sys/library/rocprofiler-sdk/fwd.cpp
0.0% 0/34 0/7 source/lib/rocprof-sys/library/rocprofiler-sdk/fwd.hpp
0.0% 0/136 0/13 source/lib/rocprof-sys/library/rocprofiler-sdk/rccl.cpp
0.0% 0/112 0/10 source/lib/rocprof-sys/library/rocprofiler-sdk/roctx_client.cpp
0.0% 0/3 0/1 source/lib/rocprof-sys/library/tracing/annotation.cpp
0.0% 0/45 0/4 source/lib/rocprof-sys/library/tracing/annotation.hpp
0.0% 0/440 0/45 source/python/libpyrocprofsys.cpp
0.6% 1/166 1/13 source/lib/core/perfetto.cpp
2.6% 33/1268 3/88 source/lib/rocprof-sys/library/rocprofiler-sdk.cpp
6.1% 2/33 1/3 source/lib/core/amd_smi.cpp
7.1% 1/14 0/1 build/coverage/_deps/nlohmann_json-src/include/nlohmann/detail/iterators/primitive_iterator.hpp
8.3% 3/36 1/9 source/lib/rocprof-sys/library/rocprofiler-sdk/marker_writer.cpp
13.6% 38/280 7/31 source/lib/rocprof-sys-rt/src/RTcommon.c
13.8% 4/29 1/4 source/lib/rocprof-sys-rt/src/RTheap-linux.c
14.2% 30/212 4/22 source/lib/rocprof-sys-rt/src/RTlinux.c
16.2% 12/74 4/16 source/lib/rocprof-sys/regions.cpp
18.0% 32/178 6/23 source/lib/core/gpu.cpp
19.0% 4/21 0/2 source/lib/backends/rocprofiler_sdk/wrapper.hpp
🟠 20-50% (25 files)
Lines Functions File
21.5% 17/79 3/5 source/lib/core/perf.cpp
23.3% 190/817 9/32 source/lib/core/trace_cache/rocpd_processor.cpp
23.6% 122/516 24/71 source/lib/core/trace_cache/metadata_registry.cpp
27.8% 27/97 4/8 source/lib/rocprof-sys-rt/src/RTposix.c
29.8% 228/764 32/103 source/lib/core/trace_cache/perfetto_processor.cpp
31.7% 59/186 6/12 source/lib/rocprof-sys/library/pmc/collectors/common/settings.hpp
32.1% 52/162 7/17 build/coverage/_deps/nlohmann_json-src/include/nlohmann/detail/input/json_sax.hpp
35.7% 30/84 10/20 source/lib/rocprof-sys/api.cpp
35.7% 10/28 3/6 source/lib/rocprof-sys/library/components/exit_gotcha.cpp
38.5% 5/13 2/3 source/lib/rocprof-sys/library/components/kill_gotcha.cpp
38.8% 31/80 1/2 source/bin/rocprof-sys-causal/rocprof-sys-causal.cpp
39.8% 37/93 5/11 source/lib/rocprof-sys/library/causal/components/blocking_gotcha.cpp
40.3% 195/484 14/17 build/coverage/_deps/nlohmann_json-src/include/nlohmann/detail/input/lexer.hpp
40.5% 17/42 5/12 build/coverage/_deps/nlohmann_json-src/include/nlohmann/detail/exceptions.hpp
41.3% 26/63 7/12 source/lib/core/agent_manager.cpp
43.8% 21/48 6/9 source/lib/core/perfetto/emitter.hpp
43.9% 36/82 2/4 source/lib/rocprof-sys/timeout.cpp
44.4% 4/9 1/3 source/lib/rocprof-sys/library/components/backtrace_timestamp.cpp
45.0% 27/60 6/16 source/lib/rocprof-sys/library/coverage/impl.hpp
45.6% 72/158 3/7 source/bin/rocprof-sys-instrument/info.hpp
46.2% 36/78 6/19 source/lib/rocprof-sys/library/components/numa_gotcha.cpp
47.7% 31/65 3/4 source/lib/rocprof-sys-rt/src/RTheap.c
47.7% 52/109 12/30 source/lib/rocprof-sys/library/tracing.hpp
48.3% 14/29 5/7 source/lib/core/perfetto/emitter.cpp
48.5% 178/367 16/30 source/lib/rocprof-sys/library/components/comm_data.cpp
🟡 50-80% (72 files)
Lines Functions File
50.0% 8/16 2/4 source/lib/binary/address_multirange.cpp
50.0% 7/14 2/6 source/lib/rocprof-sys-causal-api/causal_api.cpp
50.0% 1/2 0/1 source/lib/rocprof-sys/library/components/numa_gotcha.hpp
50.0% 1/2 0/1 source/lib/rocprof-sys/library/components/vaapi_gotcha.hpp
50.0% 1/2 0/1 source/lib/rocprof-sys/library/pmc/collectors/nic/types.hpp
50.0% 1/2 0/1 source/lib/rocprof-sys/library/thread_info.hpp
50.0% 4/8 1/2 source/lib/rocprof-sys/progress.cpp
50.4% 213/423 23/46 build/coverage/_deps/fmt-src/include/fmt/format.h
50.8% 128/252 25/37 source/lib/rocprof-sys/library/kokkosp.cpp
52.0% 13/25 2/3 source/lib/rocprof-sys/library/pmc/collectors/gpu_perf_counter/gpu_perf_counter_traits.hpp
53.2% 41/77 0/1 source/lib/core/trace_cache/sample_processor.hpp
53.4% 79/148 12/18 source/lib/rocprof-sys/library/pmc/sampler.cpp
55.0% 22/40 4/6 build/coverage/_deps/fmt-src/include/fmt/ranges.h
55.6% 5/9 1/3 build/coverage/_deps/fmt-src/include/fmt/chrono.h
56.2% 172/306 9/11 build/coverage/_deps/nlohmann_json-src/include/nlohmann/detail/output/serializer.hpp
57.5% 23/40 1/3 source/lib/common/invoke.hpp
58.3% 14/24 3/8 build/coverage/_deps/nlohmann_json-src/include/nlohmann/detail/output/output_adapters.hpp
58.3% 7/12 0/4 build/coverage/_deps/spdlog-src/include/spdlog/common.h
60.0% 63/105 5/9 source/bin/common/argument_registration.hpp
60.0% 3/5 1/1 source/lib/core/agent.hpp
60.9% 131/215 12/30 source/lib/rocprof-sys/library/components/backtrace_metrics.cpp
61.8% 289/468 21/39 build/coverage/_deps/fmt-src/include/fmt/base.h
61.8% 42/68 5/8 source/lib/rocprof-sys/library/components/fork_gotcha.cpp
62.1% 18/29 1/1 source/lib/core/perfetto/locked_file_append.cpp
62.5% 50/80 13/18 source/lib/core/binary/address_range.cpp
62.8% 49/78 4/5 source/lib/core/trace_cache/discovery.cpp
63.1% 70/111 7/10 source/lib/core/progress/bar.cpp
63.3% 155/245 23/31 source/lib/rocprof-sys/library/perf.cpp
64.7% 22/34 3/4 source/lib/core/perfetto/sinks/io_helpers.hpp
65.4% 34/52 4/4 build/coverage/_deps/nlohmann_json-src/include/nlohmann/detail/conversions/from_json.hpp
65.5% 281/429 33/40 build/coverage/_deps/nlohmann_json-src/include/nlohmann/json.hpp
66.1% 37/56 3/4 source/lib/core/mproc.cpp
66.2% 47/71 5/5 source/lib/rocprof-sys/library/pmc/collectors/gpu/gpu_traits.hpp
66.4% 87/131 4/5 build/coverage/_deps/nlohmann_json-src/include/nlohmann/detail/input/parser.hpp
66.9% 79/118 15/20 source/lib/rocprof-sys/library/runtime.cpp
67.1% 106/158 14/22 source/lib/rocprof-sys/library/components/mpi_gotcha.cpp
68.8% 22/32 3/8 source/lib/rocprof-sys/library/pmc/collectors/gpu_perf_counter/types.hpp
69.1% 465/673 34/46 source/bin/rocprof-sys-instrument/details.cpp
69.2% 373/539 40/85 source/lib/rocprof-sys-dl/dl.cpp
69.2% 18/26 2/6 source/lib/core/trace_cache/metadata_registry.hpp
69.2% 9/13 0/1 source/lib/rocprof-sys/library/process_sampler.hpp
69.7% 46/66 7/8 source/lib/rocprof-sys/library/pmc/collectors/base/collector.hpp
70.2% 40/57 4/6 source/lib/core/perfetto.hpp
70.3% 64/91 5/7 source/bin/rocprof-sys-instrument/function_signature.cpp
70.6% 24/34 5/9 source/lib/rocprof-sys/library/components/pthread_gotcha.cpp
71.1% 69/97 8/8 build/coverage/_deps/nlohmann_json-src/include/nlohmann/detail/iterators/iter_impl.hpp
71.4% 10/14 1/2 source/lib/core/containers/aligned_static_vector.hpp
72.6% 722/994 52/65 source/lib/rocprof-sys/library/sampling.cpp
72.7% 24/33 5/7 source/lib/rocprof-sys/library/causal/components/causal_gotcha.cpp
72.7% 32/44 5/8 source/lib/rocprof-sys/library/causal/components/unblocking_gotcha.cpp
73.7% 28/38 6/7 build/coverage/_deps/nlohmann_json-src/include/nlohmann/detail/iterators/iteration_proxy.hpp
74.2% 227/306 16/17 source/bin/common/preset_registry.cpp
74.3% 411/553 30/33 source/lib/rocprof-sys/library/causal/data.cpp
74.5% 35/47 4/6 source/lib/rocprof-sys/library/pmc/collectors/cpu/cpu_traits.hpp
75.0% 15/20 5/5 source/bin/rocprof-sys-instrument/function_signature.hpp
75.0% 3/4 0/1 source/lib/backends/amd_smi/nic_types.hpp
75.0% 18/24 4/5 source/lib/rocprof-sys/library/causal/sample_data.cpp
75.2% 394/524 30/40 source/bin/common/json_config.cpp
75.5% 1159/1536 51/87 source/bin/rocprof-sys-instrument/rocprof-sys-instrument.cpp
75.6% 90/119 11/12 source/lib/core/trace_cache/post_processor.cpp
76.5% 408/533 40/41 source/bin/rocprof-sys-instrument/module_function.cpp
76.6% 49/64 5/6 source/bin/rocprof-sys-instrument/log.cpp
76.8% 53/69 5/6 source/lib/binary/link_map.cpp
77.3% 17/22 3/3 source/lib/common/static_object.hpp
77.4% 328/424 35/37 source/bin/common/common_utils.cpp
77.7% 482/620 27/36 source/lib/rocprof-sys/library.cpp
77.9% 113/145 26/31 source/lib/rocprof-sys/library/thread_data.hpp
78.0% 209/268 23/25 source/lib/rocprof-sys/library/causal/sampling.cpp
78.1% 139/178 19/25 source/lib/core/constraint.cpp
79.1% 201/254 24/28 source/bin/common/tool_runner.cpp
79.2% 19/24 2/2 source/lib/core/node_info.cpp
79.9% 211/264 20/22 source/bin/rocprof-sys-instrument/internal_libs.cpp
🟢 80-100% (175 files)
Lines Functions File
80.0% 4/5 0/1 source/bin/rocprof-sys-instrument/log.hpp
80.5% 62/77 9/10 source/lib/rocprof-sys/library/pmc/device_providers/rocprofiler_sdk/provider.hpp
81.2% 69/85 17/22 source/lib/rocprof-sys/library/causal/components/progress_point.cpp
81.2% 13/16 1/1 source/lib/rocprof-sys-dl/main.c
81.4% 131/161 24/27 source/lib/rocprof-sys/library/thread_info.cpp
81.8% 1417/1732 277/302 source/lib/core/config.cpp
81.8% 81/99 11/11 source/lib/core/perfetto/engine_impl.hpp
82.4% 42/51 5/5 source/lib/core/perfetto/cached_perfetto_session.cpp
82.4% 187/227 22/25 source/bin/rocprof-sys-avail/common.cpp
83.3% 85/102 4/4 source/lib/core/perfetto/sinks/single_file_sink.cpp
83.4% 131/157 12/12 source/bin/rocprof-sys-instrument/rocprof-sys-instrument.hpp
84.2% 16/19 3/5 source/lib/rocprof-sys/library/components/comm_data.hpp
84.3% 86/102 8/9 source/lib/binary/dwarf_entry.cpp
84.6% 11/13 2/4 source/lib/rocprof-sys/library/rocprofiler-sdk/roctx_client.hpp
84.8% 134/158 17/18 source/lib/binary/symbol.cpp
85.0% 317/373 18/29 source/lib/rocprof-sys/library/causal/experiment.cpp
85.1% 74/87 14/16 source/lib/rocprof-sys/library/causal/delay.cpp
86.0% 86/100 11/15 source/lib/logger/logger.hpp
86.0% 80/93 5/7 source/bin/rocprof-sys-instrument/module_function.hpp
86.7% 98/113 9/11 source/lib/rocprof-sys/library/process_sampler.cpp
87.0% 47/54 3/3 source/lib/core/trace_cache/cache_manager.cpp
87.4% 390/446 35/47 source/bin/rocprof-sys-causal/impl.cpp
87.5% 42/48 4/5 source/lib/core/output_file_registry.cpp
87.6% 141/161 13/16 source/lib/backends/procfs/backend.hpp
87.6% 99/113 10/11 source/lib/rocprof-sys/library/tracing.cpp
88.9% 8/9 5/6 source/lib/core/constraint.hpp
89.0% 129/145 10/10 source/lib/rocprof-sys/library/coverage.cpp
89.0% 299/336 16/20 source/lib/rocprof-sys/library/components/pthread_create_gotcha.cpp
89.0% 276/310 31/35 source/lib/common/environment.hpp
89.1% 49/55 4/4 source/lib/rocprof-sys/library/pmc/collectors/gpu_perf_counter/device.hpp
89.1% 221/248 23/33 source/lib/core/sdk/tracing-config.hpp
89.4% 768/859 59/74 source/lib/core/argparse.cpp
89.8% 220/245 15/16 build/coverage/_deps/nlohmann_json-src/include/nlohmann/detail/conversions/to_chars.hpp
90.0% 9/10 1/1 source/lib/binary/scope_filter.cpp
90.0% 54/60 4/4 source/lib/core/trace_cache/buffer_storage.cpp
90.0% 9/10 1/1 source/lib/rocprof-sys/library/thread_deleter.cpp
90.1% 64/71 9/11 source/lib/rocprof-sys/library/components/backtrace.cpp
90.1% 64/71 7/10 source/lib/rocprof-sys/library/components/callchain.cpp
90.7% 234/258 11/12 source/bin/rocprof-sys-avail/generate_config.cpp
90.7% 303/334 28/31 source/lib/core/trace_cache/sample_type.hpp
90.9% 10/11 1/2 build/coverage/_deps/nlohmann_json-src/include/nlohmann/detail/json_ref.hpp
90.9% 20/22 2/2 source/lib/common/setup.hpp
91.0% 243/267 19/20 source/lib/core/trace_cache/unified_memory_processor.cpp
91.3% 178/195 14/18 source/lib/rocprof-sys/library/components/shmem_gotcha.hpp
91.3% 21/23 4/5 source/lib/core/containers/static_vector.hpp
91.5% 107/117 5/5 source/lib/binary/analysis.cpp
91.7% 11/12 2/2 source/lib/core/perfetto/session_backend.hpp
91.9% 102/111 8/8 source/lib/rocprof-sys/library/pmc/collectors/cpu/device.hpp
92.6% 163/176 9/9 source/lib/rocprof-sys/library/pmc/collectors/gpu/device.hpp
92.8% 873/941 41/48 source/bin/rocprof-sys-avail/avail.cpp
92.9% 13/14 0/1 source/lib/rocprof-sys/library/pmc/collectors/common/collector_slice.hpp
92.9% 13/14 2/2 source/lib/rocprof-sys/library/pmc/device_providers/procfs/provider.hpp
93.1% 27/29 3/5 source/lib/rocprof-sys/library/coverage.hpp
93.2% 68/73 7/8 source/lib/core/trace_cache/buffer_storage.hpp
93.3% 111/119 16/17 source/lib/core/perfetto/engine.cpp
93.3% 14/15 3/3 source/lib/core/perfetto/sinks/per_pid_file_sink.cpp
93.4% 239/256 28/32 source/lib/rocprof-sys/library/components/category_region.hpp
93.9% 31/33 6/6 source/lib/core/mpi.hpp
94.1% 16/17 3/3 source/bin/rocprof-sys-avail/component_categories.hpp
94.2% 129/137 29/32 source/lib/backends/rocprofiler_sdk/backend.hpp
94.6% 105/111 14/15 source/lib/common/path.hpp
94.7% 36/38 7/7 build/coverage/_deps/spdlog-src/include/spdlog/logger.h
94.8% 55/58 1/5 source/lib/core/trace_cache/cacheable.hpp
95.0% 19/20 4/6 build/coverage/_deps/nlohmann_json-src/include/nlohmann/detail/string_concat.hpp
95.1% 58/61 7/8 source/lib/rocprof-sys/library/pmc/collectors/cpu/sample.hpp
95.7% 44/46 10/12 source/lib/core/containers/stable_vector.hpp
95.7% 45/47 2/2 source/lib/core/trace_cache/storage_parser.hpp
95.8% 23/24 2/4 source/lib/common/pci_bdf.hpp
95.9% 162/169 19/22 source/lib/rocprof-sys/library/components/ucx_gotcha.hpp
96.1% 74/77 6/6 source/lib/rocprof-sys/library/causal/components/backtrace.cpp
96.2% 25/26 1/2 source/lib/common/delimit.hpp
96.6% 56/58 7/7 build/coverage/_deps/nlohmann_json-src/include/nlohmann/detail/conversions/to_json.hpp
96.7% 59/61 9/9 source/lib/core/utility.hpp
96.8% 30/31 3/4 source/lib/rocprof-sys/library/pmc/collectors/gpu_perf_counter/sample.hpp
96.9% 62/64 14/14 source/lib/core/trace_cache/data_types.cpp
96.9% 94/97 10/10 source/lib/rocprof-sys/library/rocprofiler-sdk/trace_control.cpp
97.0% 32/33 4/4 source/lib/core/perfetto/counter_track.hpp
97.3% 73/75 3/3 source/lib/core/perfetto/packet_framing.hpp
97.4% 74/76 6/6 source/lib/core/common_types.hpp
97.5% 39/40 6/7 source/lib/core/trace_cache/rocpd_helpers.hpp
97.7% 42/43 7/11 source/lib/core/demangler.hpp
97.8% 45/46 7/7 source/lib/rocprof-sys/library/rocprofiler-sdk/rccl_internal.hpp
97.9% 47/48 4/4 source/lib/core/utility.cpp
97.9% 47/48 3/3 source/lib/rocprof-sys/library/pmc/collectors/nic/device.hpp
98.0% 147/150 9/10 source/lib/common/md5sum.hpp
98.1% 51/52 7/7 source/lib/rocprof-sys/library/pmc/device_providers/amd_smi/provider.hpp
98.2% 56/57 6/8 source/lib/rocprof-sys/library/components/pthread_mutex_gotcha.hpp
98.7% 78/79 15/16 source/lib/backends/amd_smi/backend.hpp
100.0% 14/14 1/1 build/coverage/_deps/fmt-src/include/fmt/ostream.h
100.0% 6/6 0/3 build/coverage/_deps/nlohmann_json-src/include/nlohmann/adl_serializer.hpp
100.0% 38/38 7/8 build/coverage/_deps/nlohmann_json-src/include/nlohmann/detail/input/input_adapters.hpp
100.0% 2/2 0/1 build/coverage/_deps/nlohmann_json-src/include/nlohmann/detail/input/position_t.hpp
100.0% 2/2 0/1 build/coverage/_deps/spdlog-src/include/spdlog/details/backtracer.h
100.0% 2/2 0/1 build/coverage/_deps/spdlog-src/include/spdlog/details/circular_q.h
100.0% 1/1 0/1 build/coverage/include/profiler-hub/reader_types.hpp
100.0% 10/10 1/2 build/coverage/include/profiler-hub/writer_types.hpp
100.0% 31/31 4/5 source/bin/common/common_utils.hpp
100.0% 1/1 0/1 source/bin/common/json_config.hpp
100.0% 61/61 6/6 source/bin/rocprof-sys-avail/avail.hpp
100.0% 39/39 2/2 source/bin/rocprof-sys-avail/get_availability.hpp
100.0% 10/10 1/1 source/bin/rocprof-sys-avail/get_categories.hpp
100.0% 4/4 1/1 source/bin/rocprof-sys-avail/info_type.cpp
100.0% 9/9 0/1 source/bin/rocprof-sys-avail/info_type.hpp
100.0% 2/2 1/1 source/bin/rocprof-sys-run/rocprof-sys-run.cpp
100.0% 2/2 1/1 source/bin/rocprof-sys-sample/rocprof-sys-sample.cpp
100.0% 107/107 13/14 source/lib/backends/amd_smi/device.hpp
100.0% 7/7 0/1 source/lib/backends/amd_smi/gpu_types.hpp
100.0% 1/1 0/1 source/lib/backends/rocprofiler_sdk/types.hpp
100.0% 9/9 0/1 source/lib/binary/address_multirange.hpp
100.0% 7/7 1/1 source/lib/binary/binary_info.hpp
100.0% 1/1 0/1 source/lib/binary/dwarf_entry.hpp
100.0% 3/3 0/1 source/lib/binary/link_map.hpp
100.0% 5/5 1/2 source/lib/binary/scope_filter.hpp
100.0% 8/8 7/8 source/lib/binary/symbol.hpp
100.0% 9/9 2/2 source/lib/common/env_vars.hpp
100.0% 12/12 1/1 source/lib/common/join.hpp
100.0% 5/5 1/1 source/lib/common/string_utility.hpp
100.0% 23/23 5/6 source/lib/common/synchronized.hpp
100.0% 3/3 1/1 source/lib/common/units.hpp
100.0% 8/8 0/2 source/lib/common/version.hpp
100.0% 2/2 1/2 source/lib/core/agent_manager.hpp
100.0% 5/5 0/2 source/lib/core/argparse.hpp
100.0% 9/9 1/3 source/lib/core/binary/address_range.hpp
100.0% 49/49 10/10 source/lib/core/categories.cpp
100.0% 11/11 11/11 source/lib/core/components/fwd.hpp
100.0% 26/26 3/3 source/lib/core/config.hpp
100.0% 11/11 1/3 source/lib/core/containers/c_array.hpp
100.0% 1/1 0/1 source/lib/core/containers/operators.hpp
100.0% 2/2 2/2 source/lib/core/node_info.hpp
100.0% 1/1 0/1 source/lib/core/output_file_registry.hpp
100.0% 1/1 0/1 source/lib/core/perfetto/category_registry.hpp
100.0% 1/1 0/1 source/lib/core/perfetto/engine.hpp
100.0% 7/7 0/1 source/lib/core/perfetto/locked_file_append.hpp
100.0% 7/7 1/1 source/lib/core/perfetto/sinks/append_mode.hpp
100.0% 9/9 2/3 source/lib/core/perfetto/sinks/polymorphic_sink_view.hpp
100.0% 6/6 2/2 source/lib/core/perfetto/sinks/recording_sink.cpp
100.0% 5/5 0/2 source/lib/core/perfetto/sinks/recording_sink.hpp
100.0% 4/4 2/2 source/lib/core/perfetto/sinks/single_file_sink.hpp
100.0% 1/1 0/1 source/lib/core/progress/bar.hpp
100.0% 6/6 2/2 source/lib/core/progress/tracker.cpp
100.0% 1/1 0/1 source/lib/core/progress/tracker.hpp
100.0% 85/85 11/12 source/lib/core/state.hpp
100.0% 7/7 1/3 source/lib/core/trace_cache/cache_manager.hpp
100.0% 4/4 0/3 source/lib/core/trace_cache/data_types.hpp
100.0% 9/9 3/3 source/lib/core/trace_cache/type_registry.hpp
100.0% 24/24 4/5 source/lib/core/trace_cache/unified_memory_processor.hpp
100.0% 5/5 2/2 source/lib/core/track_registry.cpp
100.0% 4/4 0/2 source/lib/core/track_registry.hpp
100.0% 10/10 1/1 source/lib/logger/logger.cpp
100.0% 7/7 1/2 source/lib/rocprof-sys/library/causal/components/backtrace.hpp
100.0% 23/23 2/3 source/lib/rocprof-sys/library/causal/components/progress_point.hpp
100.0% 3/3 1/2 source/lib/rocprof-sys/library/causal/experiment.hpp
100.0% 4/4 1/1 source/lib/rocprof-sys/library/causal/selected_entry.cpp
100.0% 5/5 2/3 source/lib/rocprof-sys/library/causal/selected_entry.hpp
100.0% 1/1 1/1 source/lib/rocprof-sys/library/components/backtrace.hpp
100.0% 11/11 1/2 source/lib/rocprof-sys/library/components/backtrace_metrics.hpp
100.0% 4/4 0/2 source/lib/rocprof-sys/library/components/backtrace_timestamp.hpp
100.0% 1/1 1/1 source/lib/rocprof-sys/library/components/callchain.hpp
100.0% 8/8 2/2 source/lib/rocprof-sys/library/components/ensure_storage.hpp
100.0% 1/1 0/1 source/lib/rocprof-sys/library/components/fork_gotcha.hpp
100.0% 2/2 0/1 source/lib/rocprof-sys/library/components/mpi_gotcha.hpp
100.0% 533/533 12/12 source/lib/rocprof-sys/library/components/mpip.hpp
100.0% 3/3 0/1 source/lib/rocprof-sys/library/components/pthread_create_gotcha.hpp
100.0% 24/24 1/2 source/lib/rocprof-sys/library/components/pthread_mutex_gotcha_policy.hpp
100.0% 11/11 0/3 source/lib/rocprof-sys/library/perf.hpp
100.0% 66/66 5/5 source/lib/rocprof-sys/library/pmc/collectors/cpu/cache_policy.hpp
100.0% 3/3 0/2 source/lib/rocprof-sys/library/pmc/collectors/cpu/types.hpp
100.0% 9/9 0/2 source/lib/rocprof-sys/library/pmc/collectors/gpu/types.hpp
100.0% 20/20 4/4 source/lib/rocprof-sys/library/pmc/collectors/nic/perfetto_policy.hpp
100.0% 2/2 0/2 source/lib/rocprof-sys/library/pmc/common/types.hpp
100.0% 34/34 3/4 source/lib/rocprof-sys/library/rocprofiler-sdk/marker_writer.hpp
100.0% 3/3 1/2 source/lib/rocprof-sys/library/rocprofiler-sdk/trace_control.hpp
100.0% 2/2 0/1 source/lib/rocprof-sys/library/runtime.hpp
100.0% 10/10 3/3 source/lib/rocprof-sys/library/thread_data_growth.hpp
100.0% 5/5 1/1 source/lib/rocprof-sys/library/thread_deleter.hpp

Distribution

Range Files
0-20% 49
20-50% 25
50-80% 72
80-100% 175

dlamd1dai and others added 6 commits August 29, 2026 01:04
@dlamd1dai @cursoragent
fix(rccl/gin-sdma): address review on alloc.h and AllGather devComm reqs
cbdc463 
Remove a stray closing brace in alloc.h that broke RCCL host builds, and
align AllGatherGetDevCommRequirements with common.h by taking ncclComm_t and
querying comm properties internally (matching alltoall/all_reduce).

Co-authored-by: Cursor <cursoragent@cursor.com>
@dlamd1dai @cursoragent
rccl-tests/gin-sdma: port Broadcast -D 3 from stage3c-sdma-bcast
5edee72 
Port the GIN Anvil-SDMA hybrid Broadcast design: policy header, tiered
broadcast.cu kernels (LL/LSA/flat/scatter+AG/ring), devtime hook via the
shared augment-line scaffold, broadcast_perf RDC link, and opt-in pytest
multi-segment regression tests (-D 3).

Co-authored-by: Cursor <cursoragent@cursor.com>
@dlamd1dai @cursoragent
rccl/gin-sdma: add Broadcast policy unit test and GIN CI entry
cb90904 
Wire rccl-UnitTestsGinSdmaBroadcastPolicy (host GTest for
gin_sdma_broadcast_policy.h) and add broadcast_perf -D 3 to the GIN CI
test matrix.

Co-authored-by: Cursor <cursoragent@cursor.com>
@dlamd1dai @cursoragent
rccl-tests/gin-sdma: align broadcast_perf with NCCL 2.30.7 testEngine…
ecd845b 
… API.

Fix GetBw/GetDevCommRequirements signatures and testColl init so broadcast_perf -D 3 builds against common.h on the wip branch.

Co-authored-by: Cursor <cursoragent@cursor.com>
@dlamd1dai @cursoragent
rccl-tests/gin-sdma: add 4 GiB Broadcast GIN hybrid hang guard pytest.
bcdd42d 
Mirrors the existing 2 GiB completion test with default tier selection for multi-GB ring/SAG coverage.

Co-authored-by: Cursor <cursoragent@cursor.com>
@dlamd1dai @cursoragent
style(rccl-tests): apply black formatting to test_Broadcast.py.
445c18d 
Co-authored-by: Cursor <cursoragent@cursor.com>
@dlamd1dai
dlamd1dai force-pushed the users/dondai/gin-stage3c-sdma-bc branch from abd91c5 to bcdd42d Compare August 29, 2026 05:55
@dlamd1dai
dlamd1dai force-pushed the users/dondai/gin-stage3b-sdma-ag branch from cbdc463 to b9b0cb4 Compare August 29, 2026 19:23
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

project: rccl project: rccl-tests

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

1 participant

@dlamd1dai