PING for review. (And obviously, the release notes needs to be for 26.11, not 26.07.)
> -----Original Message----- > From: Morten Brørup [mailto:[email protected]] > Sent: Friday, 3 July 2026 17.44 > > graph: add optional profiling stats > > Added graph node profiling stats, build time configurable by enabling > RTE_GRAPH_PROFILE in rte_config.h. > > Signed-off-by: Morten Brørup <[email protected]> > --- > v9: > * Fixed comment still mentioning 32 objects. > * Moved sample size array outside loop. (AI) > * Added release note. (AI) > v8: > * Added static const array as local variable, instead of indexing > directly > into const array. (AI) > This also eliminates the space required between "} [idx];" weirdness. > * Added build time configurable RTE_GRAPH_PROFILE_BURST_SIZE to replace > the hardcoded burst size of 32. (AI) > v7: > * Use RTE_DIM() in histogram for loop. > * Added static_assert for histogram index values. > * Minor details to please checkpatch. > Although I disagree with requiring a space when indexing into > a constant array "(const type []){values} [idx];", > I have changed the code to comply. > v6: > * Consolidate the four histogram entries into one array. (Saeed > Bishara) > * Sample at 32 objs instead of a half burst. (Saeed Bishara) > * Moved stats to different location in rte_node structure. (Jerin) > * Minor details to please checkpatch. > v5: > * Added stats for a half burst and a full burst. > v4: > * Added documentation. (AI) > * Added more comments. (AI) > * Improved dump. (AI) > * Debug shows both cycles/call and cycles/obj. > v3: > * Debug shows cycles/obj instead of cycles/call. > * Fixed missing --in-reply-to. > v2: > * Fixed indentation. > --- > config/rte_config.h | 2 ++ > doc/guides/prog_guide/graph_lib.rst | 3 ++ > doc/guides/rel_notes/release_26_07.rst | 7 ++++ > lib/graph/graph_debug.c | 50 ++++++++++++++++++++++++++ > lib/graph/node.c | 2 ++ > lib/graph/rte_graph_worker_common.h | 32 +++++++++++++++-- > 6 files changed, 93 insertions(+), 3 deletions(-) > > diff --git a/config/rte_config.h b/config/rte_config.h > index 0447cdf2ad..7703a6325b 100644 > --- a/config/rte_config.h > +++ b/config/rte_config.h > @@ -106,6 +106,8 @@ > /* rte_graph defines */ > #define RTE_GRAPH_BURST_SIZE 256 > #define RTE_LIBRTE_GRAPH_STATS 1 > +/* RTE_GRAPH_PROFILE is not set */ > +#define RTE_GRAPH_PROFILE_BURST_SIZE 32 > > /****** driver defines ********/ > > diff --git a/doc/guides/prog_guide/graph_lib.rst > b/doc/guides/prog_guide/graph_lib.rst > index 8dd49c19d2..4311c37cc2 100644 > --- a/doc/guides/prog_guide/graph_lib.rst > +++ b/doc/guides/prog_guide/graph_lib.rst > @@ -49,6 +49,9 @@ Performance tuning parameters > RTE_GRAPH_BURST_SIZE config option. > The testing shows, on x86 and arm64 servers, The sweet spot is 256 > burst > size. While on arm64 embedded SoCs, it is either 64 or 128. > +- Enable the ``RTE_GRAPH_PROFILE`` config option for more profiling > details. > + Set the ``RTE_GRAPH_PROFILE_BURST_SIZE`` config option to sample a > specific > + burst size. > - Disable node statistics (using ``RTE_LIBRTE_GRAPH_STATS`` config > option) > if not needed. > > diff --git a/doc/guides/rel_notes/release_26_07.rst > b/doc/guides/rel_notes/release_26_07.rst > index 8b1bdada1a..cb7d48eeac 100644 > --- a/doc/guides/rel_notes/release_26_07.rst > +++ b/doc/guides/rel_notes/release_26_07.rst > @@ -83,6 +83,13 @@ New Features > * The size of the ``struct rte_mempool_cache`` was kept > for API/ABI compatibility purposes. > > +* **Added optional graph profiling statistics.** > + > + Added build-time configurable graph node profiling statistics via > + ``RTE_GRAPH_PROFILE`` in ``rte_config.h``. When enabled, tracks > cycles > + spent processing bursts of 0, 1, ``RTE_GRAPH_PROFILE_BURST_SIZE``, > + and ``RTE_GRAPH_BURST_SIZE`` objects per node. > + > * **Added RISC-V vector paths.** > > * Increased the default SIMD bitwidth to allow using the vector > extension. > diff --git a/lib/graph/graph_debug.c b/lib/graph/graph_debug.c > index e3b8cccdc1..b999fa4140 100644 > --- a/lib/graph/graph_debug.c > +++ b/lib/graph/graph_debug.c > @@ -93,6 +93,56 @@ rte_graph_obj_dump(FILE *f, struct rte_graph *g, > bool all) > n->dispatch.total_sched_fail); > } > fprintf(f, " total_calls=%" PRId64 "\n", n- > >total_calls); > + if (rte_graph_has_stats_feature()) > + fprintf(f, " total_cycles=%" PRIu64 ", avg > cycles/call=%.1f\n", > + n->total_cycles, > + n->total_calls == 0 ? 0.0 : > + (double)n->total_cycles / (double)n- > >total_calls); > +#ifdef RTE_GRAPH_PROFILE > + static const uint16_t profile_sample_sizes[] = { > + 0, 1, RTE_GRAPH_PROFILE_BURST_SIZE, > RTE_GRAPH_BURST_SIZE}; > + static_assert(RTE_DIM(profile_sample_sizes) == RTE_DIM(n- > >usage_stats), > + "usage_stats array size mismatch"); > + int64_t calls_other = n->total_calls; > + int64_t cycles_other = n->total_cycles; > + int64_t objs_other = n->total_objs; > + for (int idx = 0; idx < RTE_DIM(n->usage_stats) + 1; idx++) > { > + uint64_t calls; > + uint64_t cycles; > + double objs_per_call; > + if (idx < RTE_DIM(n->usage_stats)) { > + uint16_t idx_objs = profile_sample_sizes[idx]; > + fprintf(f, " objs[%u]\n", idx_objs); > + calls = n->usage_stats[idx].calls; > + cycles = n->usage_stats[idx].cycles; > + objs_per_call = (double)idx_objs; > + calls_other -= calls; > + cycles_other -= cycles; > + objs_other -= idx_objs * calls; > + } else { > + fprintf(f, " objs[other]\n"); > + if (calls_other > 0 && cycles_other > 0 && > objs_other > 0) { > + calls = calls_other; > + cycles = cycles_other; > + objs_per_call = (double)objs_other / > (double)calls_other; > + fprintf(f, " avg > objs/call=%.1f\n", objs_per_call); > + } else { > + calls = 0; > + cycles = 0; > + objs_per_call = 0.0; > + } > + } > + fprintf(f, " calls=%" PRIu64, calls); > + if (calls != 0) > + fprintf(f, ", cycles=%" PRIu64 ", avg > cycles/call=%.1f", > + cycles, > + (double)cycles / (double)calls); > + if (calls != 0 && objs_per_call != 0.0) > + fprintf(f, ", avg cycles/obj=%.1f", > + (double)cycles / (double)calls / > objs_per_call); > + fprintf(f, "\n"); > + } > +#endif > for (i = 0; i < n->nb_edges; i++) > fprintf(f, " edge[%d] <%s>\n", i, > n->nodes[i]->name); > diff --git a/lib/graph/node.c b/lib/graph/node.c > index 1fce3e6632..19b38881ae 100644 > --- a/lib/graph/node.c > +++ b/lib/graph/node.c > @@ -110,10 +110,12 @@ __rte_node_register(const struct > rte_node_register *reg) > rte_edge_t i; > size_t sz; > > +#ifndef RTE_GRAPH_PROFILE > /* Limit Node specific metadata to one cacheline on 64B CL > machine */ > RTE_BUILD_BUG_ON((offsetof(struct rte_node, nodes) - > offsetof(struct rte_node, ctx)) != > RTE_CACHE_LINE_MIN_SIZE); > +#endif > > graph_spinlock_lock(); > > diff --git a/lib/graph/rte_graph_worker_common.h > b/lib/graph/rte_graph_worker_common.h > index 4ab53a533e..00e8f5859d 100644 > --- a/lib/graph/rte_graph_worker_common.h > +++ b/lib/graph/rte_graph_worker_common.h > @@ -121,6 +121,17 @@ struct __rte_cache_aligned rte_node { > rte_graph_off_t xstat_off; /**< Offset to xstat counters. */ > > /** Fast path area cache line 2. */ > +#ifdef RTE_GRAPH_PROFILE > + /** > + * Usage when this node processed 0, 1, > RTE_GRAPH_PROFILE_BURST_SIZE, > + * or RTE_GRAPH_BURST_SIZE objects. > + */ > + struct __rte_cache_aligned { > + uint64_t calls; /**< Calls done. */ > + uint64_t cycles; /**< Cycles spent. */ > + } usage_stats[4]; > + /** Fast path area cache line 3. */ > +#endif > __extension__ struct __rte_cache_aligned { > #define RTE_NODE_CTX_SZ 16 > union { > @@ -148,8 +159,10 @@ struct __rte_cache_aligned rte_node { > }; > }; > > +#ifndef RTE_GRAPH_PROFILE > static_assert(offsetof(struct rte_node, nodes) - offsetof(struct > rte_node, ctx) > == RTE_CACHE_LINE_MIN_SIZE, "rte_node fast path area must fit in > 64 bytes"); > +#endif > > /** > * @internal > @@ -197,7 +210,7 @@ void __rte_node_stream_alloc_size(struct rte_graph > *graph, > static __rte_always_inline void > __rte_node_process(struct rte_graph *graph, struct rte_node *node) > { > - uint64_t start; > + uint64_t cycles; > uint16_t rc; > void **objs; > > @@ -206,11 +219,24 @@ __rte_node_process(struct rte_graph *graph, > struct rte_node *node) > rte_prefetch0(objs); > > if (rte_graph_has_stats_feature()) { > - start = rte_rdtsc(); > + cycles = -rte_rdtsc(); > rc = node->process(graph, node, objs, node->idx); > - node->total_cycles += rte_rdtsc() - start; > + cycles += rte_rdtsc(); > + node->total_cycles += cycles; > node->total_calls++; > node->total_objs += rc; > +#ifdef RTE_GRAPH_PROFILE > + if (rc <= 1) { > + node->usage_stats[rc].calls++; > + node->usage_stats[rc].cycles += cycles; > + } else if (rc == RTE_GRAPH_PROFILE_BURST_SIZE) { > + node->usage_stats[2].calls++; > + node->usage_stats[2].cycles += cycles; > + } else if (rc == RTE_GRAPH_BURST_SIZE) { > + node->usage_stats[3].calls++; > + node->usage_stats[3].cycles += cycles; > + } > +#endif > } else { > node->process(graph, node, objs, node->idx); > } > -- > 2.43.0
