From nobody Fri Sep 25 13:19:28 2026 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-alma10-1.taild15c8.ts.net [100.103.45.18]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id D2E5D13D53C; Sat, 12 Sep 2026 05:47:10 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=100.103.45.18 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1789192032; cv=none; b=YF/0tWOWZnz3rV2na5MCSDE3lggk/pThvFoqkAAxLu1oMEw1XcXPaS1FfoCuUTAwy5tKW2FMmUzWyXNfK9pcsrty6MW2+zZaygnfpDU71RnmTOLh1JvRb/z/e8F3LR8NO5J5SlmQeQlGZcP4Pb2kqf4T25oqzMjFB0GZERxvOJ4= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1789192032; c=relaxed/simple; bh=b0y4fppm3IVxVlSekfro69fhCaMWcqXPtXouG2ysrW4=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=OmUsYXP3uTMLl0Yp2rcFtBLigVICps6vew41szgVK9qfJ16uJmhdFI7a5jyuts9iGJV3eT35DltVbqpU0jamW602NL1Imjj0iFSwB9DxVn1bW2e8MCiN3M6PyxF3S4qvd6k2hBvWP3H6MZpBlfnUgOozZTEinrArZ04XfABJweM= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b=jQZN1RCZ; arc=none smtp.client-ip=100.103.45.18 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b="jQZN1RCZ" Received: by smtp.kernel.org (Postfix) with ESMTPSA id 45B631F00899; Sat, 12 Sep 2026 05:47:10 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel.org; s=k20260515; t=1789192030; bh=hhnNuVevy2qsUjBNwvkDhXpLkiD54VsA4F5CMHnUpQQ=; h=From:To:Cc:Subject:Date:In-Reply-To:References; b=jQZN1RCZiH8szfsc1ig1FAew6LzV0P/jVrGy2KI+u4XVbBfsNrFf4AsFEixF68zza /ObxnIqORgheZ+MNVxe6IQzMNQByc1KlBmcbP2vvKfvMBCQoUQk8b5UlFOkxnDEWdW R1LRDnT4Wfo6bpw8NigvlWU1t1sMXBwnxkX0z1ctCpEGFfVeunRboLqv79yn9zg6B0 c2uMEUogE5lsUegLw2vSfLkDCsOlWDVQ/712678MNWh2g7CuuZUK7hESfJSW3GqSaw MFFXDHFZufFULNGqKhSgz9trHZBTSZKj1K3fa6DsZJoutT8V7HPf9Bm6s3dFhFJeXl 8PqciKjnSlnyQ== From: Namhyung Kim To: Arnaldo Carvalho de Melo Cc: Ian Rogers , Jiri Olsa , Adrian Hunter , James Clark , Peter Zijlstra , Ingo Molnar , LKML , linux-perf-users@vger.kernel.org, Zecheng Li , Yanbo Zhao , Tengda Wu , Shuai Xue Subject: [PATCH 1/4] perf annotate-data: Convert type histogram to hashmap Date: Fri, 11 Sep 2026 22:47:03 -0700 Message-ID: <20260912054706.1475583-2-namhyung@kernel.org> X-Mailer: git-send-email 2.55.0.1032.g73a4cd73de-goog In-Reply-To: <20260912054706.1475583-1-namhyung@kernel.org> References: <20260912054706.1475583-1-namhyung@kernel.org> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" The type histogram maintains sample counts and periods per offset. Use a hashmap instead of an array to reduce the memory overhead. No functional changes intended. Signed-off-by: Namhyung Kim --- tools/perf/ui/browsers/annotate-data.c | 11 ++-- tools/perf/util/annotate-data.c | 72 ++++++++++++++++---------- tools/perf/util/annotate-data.h | 8 +-- 3 files changed, 58 insertions(+), 33 deletions(-) diff --git a/tools/perf/ui/browsers/annotate-data.c b/tools/perf/ui/browser= s/annotate-data.c index aa8c89fe2e82c1c5..82667c6ae2e746c6 100644 --- a/tools/perf/ui/browsers/annotate-data.c +++ b/tools/perf/ui/browsers/annotate-data.c @@ -62,12 +62,17 @@ static int get_member_overhead(struct annotated_data_ty= pe *adt, =20 k =3D 0; for_each_group_evsel(evsel, leader) { + struct type_hist_entry *hist; + if (symbol_conf.skip_empty && evsel__hists(evsel)->stats.nr_samples =3D=3D 0) continue; =20 - h =3D adt->histograms[evsel->core.idx]; - update_hist_entry(&entry->hists[k++], &h->addr[offset]); + h =3D &adt->histograms[evsel->core.idx]; + if (!hashmap__find(&h->samples, offset, &hist)) + continue; + + update_hist_entry(&entry->hists[k++], hist); } } return 0; @@ -416,7 +421,7 @@ static void browser__write(struct ui_browser *uib, void= *entry, int row) =20 /* print the number */ for_each_group_evsel(evsel, leader) { - struct type_hist *h =3D adt->histograms[evsel->core.idx]; + struct type_hist *h =3D &adt->histograms[evsel->core.idx]; =20 if (symbol_conf.skip_empty && evsel__hists(evsel)->stats.nr_samples =3D=3D 0) diff --git a/tools/perf/util/annotate-data.c b/tools/perf/util/annotate-dat= a.c index 4e4c587640823c81..aff60a630fd05b01 100644 --- a/tools/perf/util/annotate-data.c +++ b/tools/perf/util/annotate-data.c @@ -1750,42 +1750,45 @@ struct annotated_data_type *find_data_type(struct d= ata_loc_info *dloc) return dso__findnew_data_type(dso, &type_die); } =20 +static size_t data_type_hash(long key, void *ctx __maybe_unused) +{ + return key; +} + +static bool data_type_equal(long key1, long key2, void *ctx __maybe_unused) +{ + return key1 =3D=3D key2; +} + static int alloc_data_type_histograms(struct annotated_data_type *adt, int= nr_entries) { int i; - size_t sz =3D sizeof(struct type_hist); =20 - sz +=3D sizeof(struct type_hist_entry) * adt->self.size; - - /* Allocate a table of pointers for each event */ + /* Allocate a histogram for each event */ adt->histograms =3D calloc(nr_entries, sizeof(*adt->histograms)); if (adt->histograms =3D=3D NULL) return -ENOMEM; =20 - /* - * Each histogram is allocated for the whole size of the type. - * TODO: Probably we can move the histogram to members. - */ for (i =3D 0; i < nr_entries; i++) { - adt->histograms[i] =3D zalloc(sz); - if (adt->histograms[i] =3D=3D NULL) - goto err; + hashmap__init(&adt->histograms[i].samples, data_type_hash, + data_type_equal, /*ctx=3D*/NULL); } =20 adt->nr_histograms =3D nr_entries; return 0; - -err: - while (--i >=3D 0) - zfree(&(adt->histograms[i])); - zfree(&adt->histograms); - return -ENOMEM; } =20 static void delete_data_type_histograms(struct annotated_data_type *adt) { - for (int i =3D 0; i < adt->nr_histograms; i++) - zfree(&(adt->histograms[i])); + for (int i =3D 0; i < adt->nr_histograms; i++) { + struct hashmap *map =3D &adt->histograms[i].samples; + struct hashmap_entry *pos, *tmp; + size_t bkt; + + hashmap__for_each_entry_safe(map, pos, tmp, bkt) + free(pos->pvalue); + hashmap__clear(map); + } =20 zfree(&adt->histograms); adt->nr_histograms =3D 0; @@ -1824,6 +1827,7 @@ int annotated_data_type__update_samples(struct annota= ted_data_type *adt, int nr_samples, u64 period) { struct type_hist *h; + struct type_hist_entry *entry; =20 if (adt =3D=3D NULL) return 0; @@ -1838,12 +1842,23 @@ int annotated_data_type__update_samples(struct anno= tated_data_type *adt, if (offset < 0 || offset >=3D adt->self.size) return -1; =20 - h =3D adt->histograms[evsel->core.idx]; + h =3D &adt->histograms[evsel->core.idx]; =20 h->nr_samples +=3D nr_samples; - h->addr[offset].nr_samples +=3D nr_samples; h->period +=3D period; - h->addr[offset].period +=3D period; + + if (!hashmap__find(&h->samples, offset, &entry)) { + entry =3D zalloc(sizeof(*entry)); + if (entry =3D=3D NULL) + return -1; + + if (hashmap__append(&h->samples, offset, entry) < 0) { + free(entry); + return -1; + } + } + entry->nr_samples +=3D nr_samples; + entry->period +=3D period; return 0; } =20 @@ -1911,14 +1926,14 @@ static void print_annotated_data_type(struct annota= ted_data_type *mem_type, struct evsel *evsel, int indent) { struct annotated_member *child; - struct type_hist *h =3D mem_type->histograms[evsel->core.idx]; + struct type_hist *h; int i, nr_events =3D 0, samples =3D 0; u64 period =3D 0; int width =3D symbol_conf.show_total_period ? 11 : 7; struct evsel *pos; =20 for_each_group_evsel(pos, evsel) { - h =3D mem_type->histograms[pos->core.idx]; + h =3D &mem_type->histograms[pos->core.idx]; =20 if (symbol_conf.skip_empty && evsel__hists(pos)->stats.nr_samples =3D=3D 0) @@ -1927,8 +1942,13 @@ static void print_annotated_data_type(struct annotat= ed_data_type *mem_type, samples =3D 0; period =3D 0; for (i =3D 0; i < member->size; i++) { - samples +=3D h->addr[member->offset + i].nr_samples; - period +=3D h->addr[member->offset + i].period; + struct type_hist_entry *entry; + + if (!hashmap__find(&h->samples, member->offset + i, &entry)) + continue; + + samples +=3D entry->nr_samples; + period +=3D entry->period; } print_annotated_data_value(h, period, samples); nr_events++; diff --git a/tools/perf/util/annotate-data.h b/tools/perf/util/annotate-dat= a.h index c26130744260955f..ca2096a9ee62cbfe 100644 --- a/tools/perf/util/annotate-data.h +++ b/tools/perf/util/annotate-data.h @@ -73,12 +73,12 @@ struct type_hist_entry { * struct type_hist - Type histogram for each event * @nr_samples: Total number of samples in this data type * @period: Total count of the event in this data type - * @offset: Array of histogram entry + * @samples: Hashmap of (offset, type_hist_entry) */ struct type_hist { u64 nr_samples; u64 period; - struct type_hist_entry addr[]; + struct hashmap samples; }; =20 /** @@ -86,7 +86,7 @@ struct type_hist { * @node: RB-tree node for dso->type_tree * @self: Actual type information * @nr_histogram: Number of histogram entries - * @histograms: An array of pointers to histograms + * @histograms: An array of histograms * * This represents a data type accessed by samples in the profile data. */ @@ -94,7 +94,7 @@ struct annotated_data_type { struct rb_node node; struct annotated_member self; int nr_histograms; - struct type_hist **histograms; + struct type_hist *histograms; }; =20 extern struct annotated_data_type unknown_type; --=20 2.55.0.1032.g73a4cd73de-goog From nobody Fri Sep 25 13:19:28 2026 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-alma10-1.taild15c8.ts.net [100.103.45.18]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 9AE343438A0; Sat, 12 Sep 2026 05:47:11 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=100.103.45.18 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1789192033; cv=none; b=RJ7zXrOF3aq65K8mzB/rHO+Twq7c6pqkcgoiitXiwrKLL4z9KKUd1s6IxLaMBtgNF6+RJiAiede+1SgGL6kFvKwvZX78qPuPpcap9stP5VPpQ3VgsTwoAEvVXAU9E6ujlqyZLALol4H2La8jd0cmv+zb8lig0YbqFZWkqRzoB0c= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1789192033; c=relaxed/simple; bh=9rTPBRQNFbhkPpWu07uefnMLUJIgUw0iqFOmDICck7Q=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=EGMBE7BVNjZgFG4Uf+T9NiCcQMgMe1e7zoMTelMsfQqT8uWMTXWDoeol4p43WVpyReI5smiHfL5X69VBc4r3WEb+QPP3RCjRiZ4ABAaB3/D3BSScimUA3ZMC/f1Cm/3fv+lkHDwVXjcatLIRtTHDfih5nQZEjwI9gOEJUfbLYeM= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b=kExom4ei; arc=none smtp.client-ip=100.103.45.18 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b="kExom4ei" Received: by smtp.kernel.org (Postfix) with ESMTPSA id DA7511F0089D; Sat, 12 Sep 2026 05:47:10 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel.org; s=k20260515; t=1789192031; bh=/lbcpL34xjVfLdT41e2SVefAYol68GxosbShhiJAp/s=; h=From:To:Cc:Subject:Date:In-Reply-To:References; b=kExom4eiDy6NqWDOFMkx8p3L1m9z2e+hhx2452ObCDnAuAlCY8rL9ZSHm4lgya3PA gXxTFfjuCXjmWCUKfpJliLtd0+AY+jwjhVNIZA9q9JtNjJgSOKx99UR6RLRfWH2bIw hr5zLy8Dedm2epBtKbNI2Z77uHAv9O3fdurwZYMNYLQj2brLDuLWROD3pE2sksvrqz t4v9NdQqeeiI1dTDeoqi9ouoheRQqBJDyg+2CvvT8kvW+gGYny1O+1aJ6WDP/9264V 2OABpyFJDGWVjShN0+IK0DtmFCM4DTdN/89ygfS9g/M+W6VkAY/gbR0zYGgvNMFkuP 8RuWQjNVPfrrg== From: Namhyung Kim To: Arnaldo Carvalho de Melo Cc: Ian Rogers , Jiri Olsa , Adrian Hunter , James Clark , Peter Zijlstra , Ingo Molnar , LKML , linux-perf-users@vger.kernel.org, Zecheng Li , Yanbo Zhao , Tengda Wu , Shuai Xue , Masami Hiramatsu Subject: [PATCH 2/4] perf dwarf-aux: Add die_has_flex_array() helper Date: Fri, 11 Sep 2026 22:47:04 -0700 Message-ID: <20260912054706.1475583-3-namhyung@kernel.org> X-Mailer: git-send-email 2.55.0.1032.g73a4cd73de-goog In-Reply-To: <20260912054706.1475583-1-namhyung@kernel.org> References: <20260912054706.1475583-1-namhyung@kernel.org> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" The die_has_flex_array() returns true when the given type is a struct and contains an array at the end of the struct. Cc: Masami Hiramatsu Signed-off-by: Namhyung Kim --- tools/perf/util/dwarf-aux.c | 57 +++++++++++++++++++++++++++++++++++++ tools/perf/util/dwarf-aux.h | 3 ++ 2 files changed, 60 insertions(+) diff --git a/tools/perf/util/dwarf-aux.c b/tools/perf/util/dwarf-aux.c index d7160f87ac7d7ab3..98d018798fa750ae 100644 --- a/tools/perf/util/dwarf-aux.c +++ b/tools/perf/util/dwarf-aux.c @@ -2180,3 +2180,60 @@ Dwarf_Die *die_deref_ptr_type(Dwarf_Die *ptr_die, in= t offset, =20 return die_get_member_type(&type_die, offset, die_mem); } + +static bool __die_find_last_member(Dwarf_Die *parent_die, Dwarf_Die *die_m= em) +{ + Dwarf_Die tmp_die; + bool found =3D false; + + if (dwarf_child(parent_die, &tmp_die)) + return false; + + do { + if (dwarf_tag(&tmp_die) =3D=3D DW_TAG_member) { + memcpy(die_mem, &tmp_die, sizeof(tmp_die)); + found =3D true; + } + } while (dwarf_siblingof(&tmp_die, &tmp_die) =3D=3D 0); + + return found; +} + +/** + * die_has_flex_array - Check if the given type has a flex-array at the end + * @parent_die: a pointer to type DIE + * + * This function returns %true iff @parent_die is a struct type and has an + * array at the end. Note that the flex-array has no element, it should h= ave + * no size and the parent size doesn't include the flex-array. So it shou= ld + * locate at the offset of the parent size. + * + * For simplicity, it assumes the parent size of aligned with the flex-arr= ay. + */ +bool die_has_flex_array(Dwarf_Die *parent_die) +{ + Dwarf_Die die_mem, type_die; + Dwarf_Word size, loc; + + if (dwarf_tag(parent_die) !=3D DW_TAG_structure_type) + return false; + + if (dwarf_aggregate_size(parent_die, &size) < 0) + return false; + + /* get the member field at the end of the struct */ + if (!__die_find_last_member(parent_die, &die_mem)) + return false; + + /* get the type of the member */ + if (die_get_real_type(&die_mem, &type_die) =3D=3D NULL) + return false; + + if (dwarf_tag(&type_die) =3D=3D DW_TAG_array_type) + return die_get_data_member_location(&die_mem, &loc) =3D=3D 0 && loc =3D= =3D size; + + if (dwarf_tag(&type_die) =3D=3D DW_TAG_structure_type) + return die_has_flex_array(&type_die); + + return false; +} diff --git a/tools/perf/util/dwarf-aux.h b/tools/perf/util/dwarf-aux.h index 161f0bf980b6ee6a..9b662db710220522 100644 --- a/tools/perf/util/dwarf-aux.h +++ b/tools/perf/util/dwarf-aux.h @@ -189,4 +189,7 @@ void die_collect_global_vars(Dwarf_Die *cu_die, struct = die_var_type **var_types) /* Get the frame base information from CFA */ int die_get_cfa(Dwarf *dwarf, u64 pc, int *preg, int *poffset); =20 +/* Check whether given type has a flex array */ +bool die_has_flex_array(Dwarf_Die *parent_die); + #endif /* _DWARF_AUX_H */ --=20 2.55.0.1032.g73a4cd73de-goog From nobody Fri Sep 25 13:19:28 2026 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-alma10-1.taild15c8.ts.net [100.103.45.18]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 4E874369D54; Sat, 12 Sep 2026 05:47:12 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=100.103.45.18 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1789192033; cv=none; b=loGpz8QnP+tA9n1dSklHsbyMU9Iecq2tMTPYWfPevPanTmXilppSsmMWyrCwfkeKag0cNWnctJL5jutOGJMG1BR4JamFT+0TAH4lJOgz1tuQgOtVg94y21IdvMBZEcibg+OddbqtPJifv5x9lypyFYQWiBmRhM5IUgWoQH47Gpg= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1789192033; c=relaxed/simple; bh=WNKbhKgqiUM7cYedLfYZJc322IJnvKwdL9zCUyBNm7k=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=D7gZswASimUjmsxHmhXocaTOCJKW1uyl3b54Gpgp6ptMzEW6Q2EL2rRyFor/KbGU++WDVy6xrmnTyEU/5s92XxXMNWE6dqVQHxTyFlhLq5sS3yobcBrM9w/ZBqxosFlAX2jVRERr0LH2pR657nUtEZhImgpvaER+XoR13Bgrpuc= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b=BkyL/Isd; arc=none smtp.client-ip=100.103.45.18 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b="BkyL/Isd" Received: by smtp.kernel.org (Postfix) with ESMTPSA id 8358E1F00898; Sat, 12 Sep 2026 05:47:11 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel.org; s=k20260515; t=1789192032; bh=JTsJ9/vKQrACbY3Fqgy68jekiTz8o+GkXh3fRunTm0I=; h=From:To:Cc:Subject:Date:In-Reply-To:References; b=BkyL/IsdCPmIX36A0TuBRbReN3Qh8QepQfh2Xy+tFcoNtEWl3pXZjRmE64dC8/yA6 CBBmq0YG6L37+7Ofeake2SwJdXX4Wm7ojwb3V5dGQQPb3MUxzrQrlc2vzAPNshk/bO ZiShdVH7q6MvX4A4tKxbPXg5Gs/FWE7657dP+ZeOfGnxnBucY8TpPRb/1pzcjBTiWK KXBZU44kx5aBi6WWPHb3oWk8QNZeN1ZTZGCVs/sCFOgXtOYXu/N2aMDePmp4o+0K8w XlKNY+Ul5QeXtaZaRPw5TV7A3TuX9FVvXvdmYC50GWHw2wwT8R9OwcPgEU1goahn/5 zMaDcZ4e99y6A== From: Namhyung Kim To: Arnaldo Carvalho de Melo Cc: Ian Rogers , Jiri Olsa , Adrian Hunter , James Clark , Peter Zijlstra , Ingo Molnar , LKML , linux-perf-users@vger.kernel.org, Zecheng Li , Yanbo Zhao , Tengda Wu , Shuai Xue Subject: [PATCH 3/4] perf annotate-date: Allow out-of-size access for flex-array types Date: Fri, 11 Sep 2026 22:47:05 -0700 Message-ID: <20260912054706.1475583-4-namhyung@kernel.org> X-Mailer: git-send-email 2.55.0.1032.g73a4cd73de-goog In-Reply-To: <20260912054706.1475583-1-namhyung@kernel.org> References: <20260912054706.1475583-1-namhyung@kernel.org> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" Structs that have a flex array will have accesses beyond its original size as the array was declared as 0 sized. For now, it just allow any offset bigger than the size. It could be refined later. Signed-off-by: Namhyung Kim --- tools/perf/util/annotate-data.c | 22 +++++++++++++++------- tools/perf/util/annotate-data.h | 2 ++ 2 files changed, 17 insertions(+), 7 deletions(-) diff --git a/tools/perf/util/annotate-data.c b/tools/perf/util/annotate-dat= a.c index aff60a630fd05b01..ad043403ced58d98 100644 --- a/tools/perf/util/annotate-data.c +++ b/tools/perf/util/annotate-data.c @@ -248,8 +248,15 @@ static int __add_member_cb(Dwarf_Die *die, void *arg) else die_mem =3D member_type; =20 - if (dwarf_aggregate_size(&die_mem, &size) < 0) - size =3D 0; + if (dwarf_aggregate_size(&die_mem, &size) < 0) { + if (dwarf_tag(&die_mem) =3D=3D DW_TAG_array_type) { /* flex-array? */ + __die_get_real_type(&die_mem, &die_mem); + if (dwarf_aggregate_size(&die_mem, &size) < 0) + size =3D 0; + } else { + size =3D 0; + } + } =20 if (dwarf_attr_integrate(die, DW_AT_data_member_location, &attr)) { if (dwarf_formudata(&attr, &loc) !=3D 0) { @@ -399,6 +406,7 @@ static struct annotated_data_type *dso__findnew_data_ty= pe(struct dso *dso, result->self.type_name =3D type_name; result->self.size =3D size; INIT_LIST_HEAD(&result->self.children); + result->flex_array =3D die_has_flex_array(type_die); =20 if (symbol_conf.annotate_data_member) add_member_types(result, type_die); @@ -559,7 +567,7 @@ static enum type_match_result check_variable(struct dat= a_loc_info *dloc, return PERF_TMR_NO_SIZE; =20 /* Minimal sanity check */ - if ((unsigned)offset >=3D size) + if ((unsigned)offset >=3D size && !die_has_flex_array(&sized_type)) return PERF_TMR_BAD_OFFSET; =20 return PERF_TMR_OK; @@ -1168,7 +1176,7 @@ static enum type_match_result check_matching_type(str= uct type_state *state, =20 /* Get the size of the actual type */ if (dwarf_aggregate_size(&sized_type, &size) < 0 || - (unsigned)dloc->type_offset >=3D size) + ((unsigned)dloc->type_offset >=3D size && !die_has_flex_array(&sized= _type))) return PERF_TMR_BAD_OFFSET; =20 return PERF_TMR_OK; @@ -1192,7 +1200,7 @@ static enum type_match_result check_matching_type(str= uct type_state *state, =20 /* Get the size of the actual type */ if (dwarf_aggregate_size(type_die, &size) < 0 || - (unsigned)dloc->type_offset >=3D size) + ((unsigned)dloc->type_offset >=3D size && !die_has_flex_array(type_d= ie))) return PERF_TMR_BAD_OFFSET; =20 return PERF_TMR_OK; @@ -1211,7 +1219,7 @@ static enum type_match_result check_matching_type(str= uct type_state *state, =20 /* Get the size of the actual type */ if (dwarf_aggregate_size(type_die, &size) < 0 || - (unsigned)dloc->type_offset >=3D size) + ((unsigned)dloc->type_offset >=3D size && !die_has_flex_array(type_d= ie))) return PERF_TMR_BAIL_OUT; =20 return PERF_TMR_OK; @@ -1839,7 +1847,7 @@ int annotated_data_type__update_samples(struct annota= ted_data_type *adt, return -1; } =20 - if (offset < 0 || offset >=3D adt->self.size) + if (offset < 0 || (offset >=3D adt->self.size && !adt->flex_array)) return -1; =20 h =3D &adt->histograms[evsel->core.idx]; diff --git a/tools/perf/util/annotate-data.h b/tools/perf/util/annotate-dat= a.h index ca2096a9ee62cbfe..27b7148b64f60350 100644 --- a/tools/perf/util/annotate-data.h +++ b/tools/perf/util/annotate-data.h @@ -85,6 +85,7 @@ struct type_hist { * struct annotated_data_type - Data type to profile * @node: RB-tree node for dso->type_tree * @self: Actual type information + * @flex_array: Whethere it has a flex array * @nr_histogram: Number of histogram entries * @histograms: An array of histograms * @@ -93,6 +94,7 @@ struct type_hist { struct annotated_data_type { struct rb_node node; struct annotated_member self; + bool flex_array; int nr_histograms; struct type_hist *histograms; }; --=20 2.55.0.1032.g73a4cd73de-goog From nobody Fri Sep 25 13:19:28 2026 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-alma10-1.taild15c8.ts.net [100.103.45.18]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id A65B025776; Sat, 12 Sep 2026 05:47:12 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=100.103.45.18 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1789192033; cv=none; b=gscNLUltwr1ou/J61gRw14hHDsQklAtat4pel0J7hSPjedjnPFsZNg4ylePUXCJYLnuOYRWXN3323y+nEQWVMVaO7lap6i5QTsBw0ra+94NB9i9XCQHioVI8KYG3SSYlehAy51dgjT4BJN0HOxYdtwFj7gI9LH0kaqlpFaNkORs= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1789192033; c=relaxed/simple; bh=rxJHyYBFUF4d3+RIhzPWcaffFWFaNkaQyoI6io9Iq0A=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=q7RuMO56i8fWl3/pj03WvZb4TGZxUZNr4wMpRa5WKl9A0IuAzvUo81gRJeezHYIWhrQbF/yaHsbSVWjCKjmmcamFQWc7Ha5bykX0DJcCHQrYtNcpAqgsWNeVpte0RPFN3Oq3/QNs3CzrDB74+D8e4gaIfDtGeEjGYgkaZW+RetM= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b=g8FtxHkq; arc=none smtp.client-ip=100.103.45.18 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b="g8FtxHkq" Received: by smtp.kernel.org (Postfix) with ESMTPSA id 1EC421F0089C; Sat, 12 Sep 2026 05:47:12 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel.org; s=k20260515; t=1789192032; bh=rruhuPrNVDhLgAmVl7i/uXVuPDrgS4PaGgLoND0pxEA=; h=From:To:Cc:Subject:Date:In-Reply-To:References; b=g8FtxHkqmYMaUI12boMRp2k1bEbcUcKS1/fE8RrHrJSNj/Qezx4bkbzBmPbn2SKy3 ZrRx2LBZUD1xqkKRpNHCNxfGW1eqNxyVeaum6TPGzpbcMRe+QjBZd/ixOyi3aJrWw6 HWf4LL2BpkueQviQovgtk9K39IRPbI/gsfmDpTXq2d5zPk4c6ZDE+oN+/aYaJQvikE DuA7TBQmVZ8i5md7vFY9GIaDNhCEa49/DlIQAlHMwgTdsI29z32f6r4QIN5v7OJRy+ NSWaTWem3Xm4Ki3yypQy2CtT78uMi6Et6HOUeQXQ4ryuI10CeS0zAWJvcNZz9YAk1z MmO1SQEKB0HYg== From: Namhyung Kim To: Arnaldo Carvalho de Melo Cc: Ian Rogers , Jiri Olsa , Adrian Hunter , James Clark , Peter Zijlstra , Ingo Molnar , LKML , linux-perf-users@vger.kernel.org, Zecheng Li , Yanbo Zhao , Tengda Wu , Shuai Xue Subject: [PATCH 4/4] perf annotate-data: Adjust type offset for flex-array Date: Fri, 11 Sep 2026 22:47:06 -0700 Message-ID: <20260912054706.1475583-5-namhyung@kernel.org> X-Mailer: git-send-email 2.55.0.1032.g73a4cd73de-goog In-Reply-To: <20260912054706.1475583-1-namhyung@kernel.org> References: <20260912054706.1475583-1-namhyung@kernel.org> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" The flex array members are located beyond the original type size. Also it needs to adjust the offset in an array to find a corresponding element using module operation. Note that we focus on access to type and field, so array index is not important. Make sure to find a field name for flex arrays. Signed-off-by: Namhyung Kim --- tools/perf/util/annotate-data.c | 73 ++++++++++++++++++++++++++++++--- 1 file changed, 67 insertions(+), 6 deletions(-) diff --git a/tools/perf/util/annotate-data.c b/tools/perf/util/annotate-dat= a.c index ad043403ced58d98..0789a8cbea7c0355 100644 --- a/tools/perf/util/annotate-data.c +++ b/tools/perf/util/annotate-data.c @@ -332,19 +332,40 @@ static void delete_members(struct annotated_member *m= ember) } =20 static int fill_member_name(char *buf, size_t sz, struct annotated_member = *m, - int offset, bool first) + int offset, bool first, bool has_flex_array) { struct annotated_member *child; + bool found =3D false; + int len; =20 if (list_empty(&m->children)) return 0; =20 list_for_each_entry(child, &m->children, node) { - int len; - if (offset < child->offset || offset >=3D child->offset + child->size) continue; =20 + found =3D true; + break; + } + + if (!found && has_flex_array) { + /* + * It may have an intermediate struct that has another struct that + * contains a flex array. In that case, the outer struct itself is + * has no array and the size is less than the offset so the above + * logic won't find the outer struct at the offset. Let's use the + * last struct if it couldn't find a member for the flex array. + */ + child =3D list_last_entry(&m->children, struct annotated_member, node); + + if (!strstr(child->type_name, "struct ")) + return 0; + + found =3D true; + } + + if (found) { /* It can have anonymous struct/union members */ if (child->var_name) { len =3D scnprintf(buf, sz, "%s%s", @@ -354,15 +375,37 @@ static int fill_member_name(char *buf, size_t sz, str= uct annotated_member *m, len =3D 0; } =20 - return fill_member_name(buf + len, sz - len, child, offset, first) + len; + return fill_member_name(buf + len, sz - len, child, offset, first, + has_flex_array) + len; } + return 0; } =20 int annotated_data_type__get_member_name(struct annotated_data_type *adt, char *buf, size_t sz, int member_offset) { - return fill_member_name(buf, sz, &adt->self, member_offset, /*first=3D*/t= rue); + return fill_member_name(buf, sz, &adt->self, member_offset, /*first=3D*/t= rue, + adt->flex_array); +} + +static struct annotated_member *find_flex_array(struct annotated_member *m= , int offset) +{ + struct annotated_member *child; + + if (list_empty(&m->children)) + return NULL; + + child =3D list_last_entry(&m->children, struct annotated_member, node); + if (offset =3D=3D child->offset && strstr(child->type_name, "[]")) + return child; + + return find_flex_array(child, offset); +} + +static struct annotated_member *get_flex_array_member(struct annotated_dat= a_type *adt) +{ + return find_flex_array(&adt->self, adt->self.size); } =20 static struct annotated_data_type *dso__findnew_data_type(struct dso *dso, @@ -1743,6 +1786,7 @@ struct annotated_data_type *find_data_type(struct dat= a_loc_info *dloc) { struct dso *dso =3D map__dso(dloc->ms->map); Dwarf_Die type_die; + struct annotated_data_type *result; =20 /* * The type offset is the same as instruction offset by default. @@ -1755,7 +1799,24 @@ struct annotated_data_type *find_data_type(struct da= ta_loc_info *dloc) if (find_data_type_die(dloc, &type_die) < 0) return NULL; =20 - return dso__findnew_data_type(dso, &type_die); + result =3D dso__findnew_data_type(dso, &type_die); + if (result =3D=3D NULL) + return NULL; + + if (result->flex_array && dloc->type_offset > result->self.size) { + struct annotated_member *flex_array =3D get_flex_array_member(result); + if (flex_array) { + int offset =3D dloc->type_offset; + + /* adjust offset in the flex array */ + offset -=3D result->self.size; + offset %=3D flex_array->size; + offset +=3D result->self.size; + + dloc->type_offset =3D offset; + } + } + return result; } =20 static size_t data_type_hash(long key, void *ctx __maybe_unused) --=20 2.55.0.1032.g73a4cd73de-goog