From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-wr1-x432.google.com (mail-wr1-x432.google.com [IPv6:2a00:1450:4864:20::432]) by sourceware.org (Postfix) with ESMTPS id F31B43858C50 for ; Sat, 14 Jan 2023 06:11:32 +0000 (GMT) DMARC-Filter: OpenDMARC Filter v1.4.2 sourceware.org F31B43858C50 Authentication-Results: sourceware.org; dmarc=pass (p=none dis=none) header.from=adacore.com Authentication-Results: sourceware.org; spf=pass smtp.mailfrom=adacore.com Received: by mail-wr1-x432.google.com with SMTP id t5so18432016wrq.1 for ; Fri, 13 Jan 2023 22:11:32 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=adacore.com; s=google; h=in-reply-to:content-disposition:mime-version:references:message-id :subject:cc:to:from:date:from:to:cc:subject:date:message-id:reply-to; bh=fgRu4lewk+3EEr+iKIp8ziInXqbsmb6W3X6f5TU0FtE=; b=M3TZJCG9r2glqeGFayWSg26JiIrzuxMbYg2809tQuhvqYXDqWVlMSl74sG9iHmrahQ CRYCpvsOObs8o8ZUi5BUQ7sZK4QN/lQTsvSyOnF8wZuxM/hvfhFNG71OtrmigF5RAaOk Th4vz64gxyJ2d3lTdQ93loiL+mV6rgrBRfoK6PXaMaPG+TCRS+64gvqo2UKEh5teJeZB M6ONuZITqzrT/OcyVsuwURIhQtLWZf5TLENgkX+1fQhG7/EV8HWq1CHvfE7cWZFmzcUd yIVoZ9jiEKBK+IzOQMKNtbL1L/GubPOVflq1/RXgCmOrEL6skWeX2F3YP40ZdgEzzkwc BheA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=in-reply-to:content-disposition:mime-version:references:message-id :subject:cc:to:from:date:x-gm-message-state:from:to:cc:subject:date :message-id:reply-to; bh=fgRu4lewk+3EEr+iKIp8ziInXqbsmb6W3X6f5TU0FtE=; b=yUWRvby6baExzMsefEDXVzXxDV76woaPnDp4EFE1aWpbqYfKjWug/PBZQdPJG38vSj xItl+0aPh84ZcIyB4yzzjQVIzHqE8nkz/tTrfrs+Ft9DpDoVGKSjI9GxcPOFFBBfydZ8 S/I+wKX+0yUEYJlrYNdpXZfIvQmY9oixga8SM7Z4KYHvjroEqOu0v35PfCiG94Pbo0U9 cORh4bq28M5OlQYDBHeBMrni4JzYRa02t+1MxNI2VXeS38KiB10AiYVMrsOZ5CmFrifu FZHTHUtmtWBFjQ32+5G159dijySB8s8gCmHr8LgKUgSTWZ5r/WsPxD6qa651c/vFnXtR AxXw== X-Gm-Message-State: AFqh2krPAR5uO7+ZYxJzeTafGeRaHLmorIAobEpyi6shfBjpGmkTW1Xo z3KGh9HTYM+pP/mwjhUgM/DM X-Google-Smtp-Source: AMrXdXs6iu2fjEkr6i2EZVkuu8TPquqzF0LPKYOIhiOuX9AUeR5oBMDiHTLTHCOYceyISAWvmiNL8Q== X-Received: by 2002:a05:6000:124e:b0:2bb:e9e6:ac26 with SMTP id j14-20020a056000124e00b002bbe9e6ac26mr13597546wrx.57.1673676691647; Fri, 13 Jan 2023 22:11:31 -0800 (PST) Received: from takamaka.gnat.com ([2a01:cb22:1d5:1100:6616:27c7:2935:afbc]) by smtp.gmail.com with ESMTPSA id u14-20020adfdd4e000000b002366e3f1497sm20820189wrm.6.2023.01.13.22.11.30 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Fri, 13 Jan 2023 22:11:31 -0800 (PST) Received: by takamaka.gnat.com (Postfix, from userid 1000) id E433C8219E; Sat, 14 Jan 2023 10:11:28 +0400 (+04) Date: Sat, 14 Jan 2023 10:11:28 +0400 From: Joel Brobecker To: Tom Tromey via Gdb-patches Cc: Tom Tromey , Joel Brobecker Subject: Re: [PATCH v2 4/4] Fix parameter-less template regression in new DWARF reader Message-ID: References: <20230110183338.2623088-1-tromey@adacore.com> <20230110183338.2623088-5-tromey@adacore.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20230110183338.2623088-5-tromey@adacore.com> X-Spam-Status: No, score=-11.4 required=5.0 tests=BAYES_00,DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,DKIM_VALID_EF,GIT_PATCH_0,KAM_SHORT,RCVD_IN_DNSWL_NONE,SPF_HELO_NONE,SPF_PASS,TXREP autolearn=ham autolearn_force=no version=3.4.6 X-Spam-Checker-Version: SpamAssassin 3.4.6 (2021-04-09) on server2.sourceware.org List-Id: Hi Tom, On Tue, Jan 10, 2023 at 11:33:38AM -0700, Tom Tromey via Gdb-patches wrote: > PR c++/29896 points out a regression in the new DWARF reader. It does > not properly handle a case like "break fn", where "fn" is a template > function. > > This happens because the new index uses strncasecmp to compare. > However, to make this work correctly, we need a custom function that > ignores template parameters. > > This patch adds a custom comparison function and fixes the bug. A new > test case is included. > > Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=29896 Thanks for updating the copyright years. I believe it should be 2022-2023, though. I remember being told that the start of the copyright year range corresponds to the year the file was committed to medium. The patch is still OK for me with the minor correction above. > --- > gdb/dwarf2/cooked-index.c | 143 +++++++++++++++++++++++++---- > gdb/dwarf2/cooked-index.h | 15 ++- > gdb/dwarf2/read.c | 3 +- > gdb/testsuite/gdb.cp/paramless.cc | 46 ++++++++++ > gdb/testsuite/gdb.cp/paramless.exp | 41 +++++++++ > 5 files changed, 226 insertions(+), 22 deletions(-) > create mode 100644 gdb/testsuite/gdb.cp/paramless.cc > create mode 100644 gdb/testsuite/gdb.cp/paramless.exp > > diff --git a/gdb/dwarf2/cooked-index.c b/gdb/dwarf2/cooked-index.c > index c711e3b9b2a..09b3fd70b26 100644 > --- a/gdb/dwarf2/cooked-index.c > +++ b/gdb/dwarf2/cooked-index.c > @@ -25,6 +25,114 @@ > #include "ada-lang.h" > #include "split-name.h" > #include > +#include "safe-ctype.h" > +#include "gdbsupport/selftest.h" > + > +/* See cooked-index.h. */ > + > +bool > +cooked_index_entry::compare (const char *stra, const char *strb, > + bool completing) > +{ > + /* If we've ever matched "<" in both strings, then we disable the > + special template parameter handling. */ > + bool seen_lt = false; > + > + while (*stra != '\0' > + && *strb != '\0' > + && (TOLOWER ((unsigned char) *stra) > + == TOLOWER ((unsigned char ) *strb))) > + { > + if (*stra == '<') > + seen_lt = true; > + ++stra; > + ++strb; > + } > + > + unsigned c1 = TOLOWER ((unsigned char) *stra); > + unsigned c2 = TOLOWER ((unsigned char) *strb); > + > + if (completing) > + { > + /* When completing, if one string ends earlier than the other, > + consider them as equal. Also, completion mode ignores the > + special '<' handling. */ > + if (c1 == '\0' || c2 == '\0') > + return false; > + /* Fall through to the generic case. */ > + } > + else if (seen_lt) > + { > + /* Fall through to the generic case. */ > + } > + else if (c1 == '\0' || c1 == '<') > + { > + /* Maybe they both end at the same spot. */ > + if (c2 == '\0' || c2 == '<') > + return false; > + /* First string ended earlier. */ > + return true; > + } > + else if (c2 == '\0' || c2 == '<') > + { > + /* Second string ended earlier. */ > + return false; > + } > + > + return c1 < c2; > +} > + > +#if GDB_SELF_TEST > + > +namespace { > + > +void > +test_compare () > +{ > + SELF_CHECK (!cooked_index_entry::compare ("abcd", "abcd", false)); > + SELF_CHECK (!cooked_index_entry::compare ("abcd", "abcd", false)); > + SELF_CHECK (!cooked_index_entry::compare ("abcd", "abcd", true)); > + SELF_CHECK (!cooked_index_entry::compare ("abcd", "abcd", true)); > + > + SELF_CHECK (cooked_index_entry::compare ("abcd", "ABCDE", false)); > + SELF_CHECK (!cooked_index_entry::compare ("ABCDE", "abcd", false)); > + SELF_CHECK (!cooked_index_entry::compare ("abcd", "ABCDE", true)); > + SELF_CHECK (!cooked_index_entry::compare ("ABCDE", "abcd", true)); > + > + SELF_CHECK (!cooked_index_entry::compare ("name", "name<>", false)); > + SELF_CHECK (!cooked_index_entry::compare ("name<>", "name", false)); > + SELF_CHECK (!cooked_index_entry::compare ("name", "name<>", true)); > + SELF_CHECK (!cooked_index_entry::compare ("name<>", "name", true)); > + > + SELF_CHECK (!cooked_index_entry::compare ("name", "name", false)); > + SELF_CHECK (!cooked_index_entry::compare ("name", "name", false)); > + SELF_CHECK (!cooked_index_entry::compare ("name", "name", true)); > + SELF_CHECK (!cooked_index_entry::compare ("name", "name", true)); > + > + SELF_CHECK (!cooked_index_entry::compare ("name>", > + "name>", false)); > + > + SELF_CHECK (!cooked_index_entry::compare ("name", "name>", false)); > + SELF_CHECK (!cooked_index_entry::compare ("name>", "name", false)); > + SELF_CHECK (cooked_index_entry::compare ("name>", > + false)); > + SELF_CHECK (!cooked_index_entry::compare ("name + "name>", > + true)); > + SELF_CHECK (!cooked_index_entry::compare ("name>", "name + false)); > + SELF_CHECK (!cooked_index_entry::compare ("name>", "name + true)); > + > + SELF_CHECK (cooked_index_entry::compare ("", "abcd", false)); > + SELF_CHECK (!cooked_index_entry::compare ("", "abcd", true)); > + SELF_CHECK (!cooked_index_entry::compare ("abcd", "", false)); > + SELF_CHECK (!cooked_index_entry::compare ("abcd", "", true)); > +} > + > +} /* anonymous namespace */ > + > +#endif /* GDB_SELF_TEST */ > > /* See cooked-index.h. */ > > @@ -247,30 +355,24 @@ cooked_index::do_finalize () > /* See cooked-index.h. */ > > cooked_index::range > -cooked_index::find (gdb::string_view name, bool completing) > +cooked_index::find (const std::string &name, bool completing) > { > wait (); > > - auto lower = std::lower_bound (m_entries.begin (), m_entries.end (), > - name, > + auto lower = std::lower_bound (m_entries.begin (), m_entries.end (), name, > [=] (const cooked_index_entry *entry, > - const gdb::string_view &n) > + const std::string &n) > { > - int cmp = strncasecmp (entry->canonical, n.data (), n.length ()); > - if (cmp != 0 || completing) > - return cmp < 0; > - return strlen (entry->canonical) < n.length (); > + return cooked_index_entry::compare (entry->canonical, n.c_str (), > + completing); > }); > > - auto upper = std::upper_bound (m_entries.begin (), m_entries.end (), > - name, > - [=] (const gdb::string_view &n, > + auto upper = std::upper_bound (m_entries.begin (), m_entries.end (), name, > + [=] (const std::string &n, > const cooked_index_entry *entry) > { > - int cmp = strncasecmp (n.data (), entry->canonical, n.length ()); > - if (cmp != 0 || completing) > - return cmp < 0; > - return n.length () < strlen (entry->canonical); > + return cooked_index_entry::compare (n.c_str (), entry->canonical, > + completing); > }); > > return range (lower, upper); > @@ -311,7 +413,7 @@ cooked_index_vector::get_addrmaps () > /* See cooked-index.h. */ > > cooked_index_vector::range > -cooked_index_vector::find (gdb::string_view name, bool completing) > +cooked_index_vector::find (const std::string &name, bool completing) > { > std::vector result_range; > result_range.reserve (m_vector.size ()); > @@ -339,3 +441,12 @@ cooked_index_vector::get_main () const > > return result; > } > + > +void _initialize_cooked_index (); > +void > +_initialize_cooked_index () > +{ > +#if GDB_SELF_TEST > + selftests::register_test ("cooked_index_entry::compare", test_compare); > +#endif > +} > diff --git a/gdb/dwarf2/cooked-index.h b/gdb/dwarf2/cooked-index.h > index 607e07745f9..2b34a6534e8 100644 > --- a/gdb/dwarf2/cooked-index.h > +++ b/gdb/dwarf2/cooked-index.h > @@ -143,11 +143,16 @@ struct cooked_index_entry : public allocate_on_obstack > STORAGE. */ > const char *full_name (struct obstack *storage) const; > > - /* Entries must be sorted case-insensitively; this compares two > - entries. */ > + /* Compare two strings, case-insensitively. Return true if STRA is > + less than STRB. If one string has template parameters, but the > + other does not, then they are considered to be equal; so for > + example "t" == "t", "t" < "t", but "t" == "t". */ > + static bool compare (const char *stra, const char *strb, bool completing); > + > + /* Compare two entries by canonical name. */ > bool operator< (const cooked_index_entry &other) const > { > - return strcasecmp (canonical, other.canonical) < 0; > + return compare (canonical, other.canonical, false); > } > > /* The name as it appears in DWARF. This always points into one of > @@ -232,7 +237,7 @@ class cooked_index > /* Look up an entry by name. Returns a range of all matching > results. If COMPLETING is true, then a larger range, suitable > for completion, will be returned. */ > - range find (gdb::string_view name, bool completing); > + range find (const std::string &name, bool completing); > > private: > > @@ -335,7 +340,7 @@ class cooked_index_vector : public dwarf_scanner_base > /* Look up an entry by name. Returns a range of all matching > results. If COMPLETING is true, then a larger range, suitable > for completion, will be returned. */ > - range find (gdb::string_view name, bool completing); > + range find (const std::string &name, bool completing); > > /* Return a range of all the entries. */ > range all_entries () > diff --git a/gdb/dwarf2/read.c b/gdb/dwarf2/read.c > index c3f246fedf7..44b54f77de9 100644 > --- a/gdb/dwarf2/read.c > +++ b/gdb/dwarf2/read.c > @@ -18741,8 +18741,9 @@ cooked_index_functions::expand_symtabs_matching > { > std::vector name_vec > = lookup_name_without_params.split_name (lang); > + std::string last_name = gdb::to_string (name_vec.back ()); > > - for (const cooked_index_entry *entry : table->find (name_vec.back (), > + for (const cooked_index_entry *entry : table->find (last_name, > completing)) > { > QUIT; > diff --git a/gdb/testsuite/gdb.cp/paramless.cc b/gdb/testsuite/gdb.cp/paramless.cc > new file mode 100644 > index 00000000000..392b15f4256 > --- /dev/null > +++ b/gdb/testsuite/gdb.cp/paramless.cc > @@ -0,0 +1,46 @@ > +/* Test case for template breakpoint test. > + > + Copyright 2023 Free Software Foundation, Inc. > + > + This file is part of GDB. > + > + This program is free software; you can redistribute it and/or modify > + it under the terms of the GNU General Public License as published by > + the Free Software Foundation; either version 3 of the License, or > + (at your option) any later version. > + > + This program is distributed in the hope that it will be useful, > + but WITHOUT ANY WARRANTY; without even the implied warranty of > + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the > + GNU General Public License for more details. > + > + You should have received a copy of the GNU General Public License > + along with this program. If not, see . */ > + > +template > +struct outer > +{ > + template > + int fn (int x) > + { > + return x + 23; > + } > +}; > + > +template > +int toplev (int y) > +{ > + return y; > +} > + > +outer outer1; > +outer outer2; > + > +int main () > +{ > + int x1 = outer1.fn (0); > + int x2 = outer2.fn (-46); > + int x3 = toplev (0); > + int x4 = toplev (0); > + return x1 + x2; > +} > diff --git a/gdb/testsuite/gdb.cp/paramless.exp b/gdb/testsuite/gdb.cp/paramless.exp > new file mode 100644 > index 00000000000..4fc8fd9e015 > --- /dev/null > +++ b/gdb/testsuite/gdb.cp/paramless.exp > @@ -0,0 +1,41 @@ > +# Copyright 2023 Free Software Foundation, Inc. > + > +# This program is free software; you can redistribute it and/or modify > +# it under the terms of the GNU General Public License as published by > +# the Free Software Foundation; either version 3 of the License, or > +# (at your option) any later version. > +# > +# This program is distributed in the hope that it will be useful, > +# but WITHOUT ANY WARRANTY; without even the implied warranty of > +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the > +# GNU General Public License for more details. > +# > +# You should have received a copy of the GNU General Public License > +# along with this program. If not, see . > + > +# This file is part of the gdb testsuite. > + > +# Test template breakpoints without parameters. > + > +if { [skip_cplus_tests] } { continue } > + > +standard_testfile .cc > + > +if {[prepare_for_testing "failed to prepare" $testfile $srcfile {debug c++}]} { > + return -1 > +} > + > +gdb_breakpoint "outer::fn" message > +delete_breakpoints > + > +gdb_breakpoint "outer::fn" message > +delete_breakpoints > + > +gdb_test "break outer::fn" "Breakpoint $decimal at .*2 locations." > +delete_breakpoints > + > +gdb_test "break toplev" "Breakpoint $decimal at .*2 locations." > +delete_breakpoints > + > +gdb_breakpoint "toplev" message > +delete_breakpoints > -- > 2.38.1 > -- Joel