From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.129.124]) by sourceware.org (Postfix) with ESMTPS id 2BCF6385C6DB for ; Tue, 19 Sep 2023 14:35:49 +0000 (GMT) DMARC-Filter: OpenDMARC Filter v1.4.2 sourceware.org 2BCF6385C6DB Authentication-Results: sourceware.org; dmarc=pass (p=none dis=none) header.from=redhat.com Authentication-Results: sourceware.org; spf=pass smtp.mailfrom=redhat.com DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1695134148; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version:content-type:content-type: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=1rvLJicqTNL2QAOmoJCotQqKOgPWoBCf7oIGJh0q4wM=; b=YgSpq6IcPW+HR6B+eVE7QcTaGviQNlk9V7ufommex/5bfs6xCkZeRi5jn9rp1rVrv1E1gb OhtaNe5ENkuyZzeKHxgjnhRW9pSknjV/FsdqyvJrVE9CYhDHC9FCb91ie1pIc3UIil2BBD xdSAIvrCm3XyxmQ7QaB/cKQ7Kn8bvLk= Received: from mail-pj1-f69.google.com (mail-pj1-f69.google.com [209.85.216.69]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-553-dp1IaEXtOYG5sDxQjzVtJw-1; Tue, 19 Sep 2023 10:35:46 -0400 X-MC-Unique: dp1IaEXtOYG5sDxQjzVtJw-1 Received: by mail-pj1-f69.google.com with SMTP id 98e67ed59e1d1-274ac6d570aso3338297a91.2 for ; Tue, 19 Sep 2023 07:35:46 -0700 (PDT) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1695134146; x=1695738946; h=content-transfer-encoding:cc:to:subject:message-id:date:from :in-reply-to:references:mime-version:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=1rvLJicqTNL2QAOmoJCotQqKOgPWoBCf7oIGJh0q4wM=; b=IKf17yf9Ddtvmwtne5Z/CnWTIObOqOu1GLY/UtzpRscEv6Si8pPV3OEpEyNppOV3at JajZ3Dy/M1jIVNXb1RxrR66etesqNVUP3lTFTUPznCo5Azj7t9I6rEsgK1oEh4p26jFQ tgVVmsxIldmHucVrqPcfmE4bMzjK1ufx7v+7Laqvr2sQezr2tgY9rLjCzuuuDrQWxPXd TOxO6WpyAdutompsA1d1xDGncfhZN4usUziKn9NUkw9UTyuN9wlUFp3P71Qq2MngvkMY fj3GkjEvGsK0pZ6o37ecnJsOMUe7TOkNZhG91wzlNgSoMbNd//vebWaYywQmI4gHXZxK RbDA== X-Gm-Message-State: AOJu0Ywouh30hTo9/pADF7AyPonYpRQnjUZyBhwG9TK6yZI+F01BLjl0 9XCSDOurC1CddJAzbSCRrrNQIjfZI2XVZy7/JGpVsV/5gVQiaEsPg28Y9Up6zsyF2HWfKblIuxj oTJnxj4EObCzWFXLmzW7jpQunWkxEQhEPMW+h3k9FEqWrvXk= X-Received: by 2002:a17:90a:8d04:b0:268:ef84:b0cd with SMTP id c4-20020a17090a8d0400b00268ef84b0cdmr10708493pjo.18.1695134145472; Tue, 19 Sep 2023 07:35:45 -0700 (PDT) X-Google-Smtp-Source: AGHT+IHhXx5sXEAXX3kggP75+fJlSfaBAG7dulR1Gz6RgU08U2CTH3L8cPPmOlf7ddIF8i2xdsrHMf/ghnYQ2jfE+08= X-Received: by 2002:a17:90a:8d04:b0:268:ef84:b0cd with SMTP id c4-20020a17090a8d0400b00268ef84b0cdmr10708466pjo.18.1695134144833; Tue, 19 Sep 2023 07:35:44 -0700 (PDT) MIME-Version: 1.0 References: <20230816044259.2675531-1-amerey@redhat.com> <20230816044259.2675531-8-amerey@redhat.com> In-Reply-To: <20230816044259.2675531-8-amerey@redhat.com> From: Aaron Merey Date: Tue, 19 Sep 2023 10:35:33 -0400 Message-ID: Subject: Re: [PATCH 7/7 v4] gdb/debuginfod: Add .debug_line downloading To: gdb-patches@sourceware.org Cc: aburgess@redhat.com X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: redhat.com Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Spam-Status: No, score=-11.2 required=5.0 tests=BAYES_00,DKIMWL_WL_HIGH,DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,DKIM_VALID_EF,GIT_PATCH_0,RCVD_IN_DNSWL_NONE,RCVD_IN_MSPIKE_H4,RCVD_IN_MSPIKE_WL,SPF_HELO_NONE,SPF_NONE,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: Ping Thanks, Aaron On Wed, Aug 16, 2023 at 12:43=E2=80=AFAM Aaron Merey wr= ote: > > v3: https://sourceware.org/pipermail/gdb-patches/2023-June/199988.html > > v4 adds .debug_line downloading during command autocompletion. > > Commit message: > > ELF/DWARF section downloading allows gdb to download .gdb_index files in > order to defer full debuginfo downloads. However .gdb_index does not > contain any information regarding source filenames. When a gdb command > includes a filename argument (ex. 'break main.c:50'), this results in > the mass downloading of all deferred debuginfo so that gdb can search the > debuginfo for matching source filenames. This can result in unnecessary > downloads. > > To improve this, have gdb instead download each debuginfo's .debug_line > (and .debug_line_str if using DWARF5) when executing these commands. > Download full debuginfo only when its .debug_line contains a matching > filename. > > Since the combined size of .debug_line and .debug_line_str is only about > 1% the size of the corresponding debuginfo, significant time can be saved > by checking these sections before choosing to download an entire debuginf= o. > > This patch also redirects stdout and stderr of the debuginfod server > used by testsuite/gdb.debuginfod tests to a server_log standard output > file. While adding tests for this patch I ran into an issue where the > test server would block when logging to stderr, presumably because the > stderr buffer filled up and wasn't being read from. Redirecting the > log to a file fixes this and also makes the server log more accessible > when debugging test failures. > --- > gdb/completer.c | 18 +- > gdb/dwarf2/line-header.c | 215 +++++++++++++++-------- > gdb/dwarf2/line-header.h | 10 ++ > gdb/dwarf2/read-gdb-index.c | 60 +++++++ > gdb/dwarf2/read.c | 208 ++++++++++++++++++++++ > gdb/dwarf2/read.h | 37 ++++ > gdb/testsuite/gdb.debuginfod/section.exp | 21 +++ > 7 files changed, 491 insertions(+), 78 deletions(-) > > diff --git a/gdb/completer.c b/gdb/completer.c > index 91a29daee36..7832c21b638 100644 > --- a/gdb/completer.c > +++ b/gdb/completer.c > @@ -1345,6 +1345,10 @@ complete_line_internal_1 (completion_tracker &trac= ker, > { > /* We've recognized a full command. */ > > + /* Disable pagination since responding to the pagination prompt > + overwrites rl_line_buffer. */ > + scoped_restore pag_restore =3D make_scoped_restore (&pagination_en= abled, false); > + > if (p =3D=3D tmp_command + point) > { > /* There is no non-whitespace in the line beyond the > @@ -1444,7 +1448,8 @@ complete_line_internal_1 (completion_tracker &track= er, > } > > /* Wrapper around complete_line_internal_1 to handle > - MAX_COMPLETIONS_REACHED_ERROR. */ > + MAX_COMPLETIONS_REACHED_ERROR and possible progress update > + interactions. */ > > static void > complete_line_internal (completion_tracker &tracker, > @@ -1452,6 +1457,11 @@ complete_line_internal (completion_tracker &tracke= r, > const char *line_buffer, int point, > complete_line_internal_reason reason) > { > + scoped_restore restore_prefix_state > + =3D make_scoped_restore > + (&cur_prefix_state, > + ui_out::progress_update::prefix_state::NEWLINE_NEEDED); > + > try > { > complete_line_internal_1 (tracker, text, line_buffer, point, reaso= n); > @@ -1461,6 +1471,12 @@ complete_line_internal (completion_tracker &tracke= r, > if (except.error !=3D MAX_COMPLETIONS_REACHED_ERROR) > throw; > } > + > + /* If progress update messages printed, then the text being completed > + needs to be printed again. */ > + if (cur_prefix_state > + =3D=3D ui_out::progress_update::prefix_state::NEWLINE_PRINTED) > + rl_forced_update_display (); > } > > /* See completer.h. */ > diff --git a/gdb/dwarf2/line-header.c b/gdb/dwarf2/line-header.c > index d072a91bac9..b9210d84f6b 100644 > --- a/gdb/dwarf2/line-header.c > +++ b/gdb/dwarf2/line-header.c > @@ -102,50 +102,57 @@ read_checked_initial_length_and_offset (bfd *abfd, = const gdb_byte *buf, > { > LONGEST length =3D read_initial_length (abfd, buf, bytes_read); > > - gdb_assert (cu_header->initial_length_size =3D=3D 4 > - || cu_header->initial_length_size =3D=3D 8 > - || cu_header->initial_length_size =3D=3D 12); > + if (cu_header !=3D nullptr) > + { > + gdb_assert (cu_header->initial_length_size =3D=3D 4 > + || cu_header->initial_length_size =3D=3D 8 > + || cu_header->initial_length_size =3D=3D 12); > > - if (cu_header->initial_length_size !=3D *bytes_read) > - complaint (_("intermixed 32-bit and 64-bit DWARF sections")); > + if (cu_header->initial_length_size !=3D *bytes_read) > + complaint (_("intermixed 32-bit and 64-bit DWARF sections")); > + } > > *offset_size =3D (*bytes_read =3D=3D 4) ? 4 : 8; > return length; > } > > -/* Read directory or file name entry format, starting with byte of > - format count entries, ULEB128 pairs of entry formats, ULEB128 of > - entries count and the entries themselves in the described entry > - format. */ > + > +/* Like read_formatted_entries but the .debug_line and .debug_line_str > + are stored in LINE_BUFP and LINE_STR_DATA. This is used for cases > + where these sections are read from separate files without necessarily > + having access to the entire debuginfo file they originate from. */ > > static void > -read_formatted_entries (dwarf2_per_objfile *per_objfile, bfd *abfd, > - const gdb_byte **bufp, struct line_header *lh, > - unsigned int offset_size, > - void (*callback) (struct line_header *lh, > - const char *name, > - dir_index d_index, > - unsigned int mod_time, > - unsigned int length)) > +read_formatted_entries > + (bfd *parent_bfd, const gdb_byte **line_bufp, > + const gdb::array_view line_str_data, > + struct line_header *lh, > + unsigned int offset_size, > + void (*callback) (struct line_header *lh, > + const char *name, > + dir_index d_index, > + unsigned int mod_time, > + unsigned int length)) > { > gdb_byte format_count, formati; > ULONGEST data_count, datai; > - const gdb_byte *buf =3D *bufp; > + const gdb_byte *buf =3D *line_bufp; > + const gdb_byte *str_buf =3D line_str_data.data (); > const gdb_byte *format_header_data; > unsigned int bytes_read; > > - format_count =3D read_1_byte (abfd, buf); > + format_count =3D read_1_byte (parent_bfd, buf); > buf +=3D 1; > format_header_data =3D buf; > for (formati =3D 0; formati < format_count; formati++) > { > - read_unsigned_leb128 (abfd, buf, &bytes_read); > + read_unsigned_leb128 (parent_bfd, buf, &bytes_read); > buf +=3D bytes_read; > - read_unsigned_leb128 (abfd, buf, &bytes_read); > + read_unsigned_leb128 (parent_bfd, buf, &bytes_read); > buf +=3D bytes_read; > } > > - data_count =3D read_unsigned_leb128 (abfd, buf, &bytes_read); > + data_count =3D read_unsigned_leb128 (parent_bfd, buf, &bytes_read); > buf +=3D bytes_read; > for (datai =3D 0; datai < data_count; datai++) > { > @@ -154,10 +161,10 @@ read_formatted_entries (dwarf2_per_objfile *per_obj= file, bfd *abfd, > > for (formati =3D 0; formati < format_count; formati++) > { > - ULONGEST content_type =3D read_unsigned_leb128 (abfd, format, &= bytes_read); > + ULONGEST content_type =3D read_unsigned_leb128 (parent_bfd, for= mat, &bytes_read); > format +=3D bytes_read; > > - ULONGEST form =3D read_unsigned_leb128 (abfd, format, &bytes_r= ead); > + ULONGEST form =3D read_unsigned_leb128 (parent_bfd, format, &b= ytes_read); > format +=3D bytes_read; > > gdb::optional string; > @@ -166,36 +173,48 @@ read_formatted_entries (dwarf2_per_objfile *per_obj= file, bfd *abfd, > switch (form) > { > case DW_FORM_string: > - string.emplace (read_direct_string (abfd, buf, &bytes_read)= ); > + string.emplace (read_direct_string (parent_bfd, buf, &bytes= _read)); > buf +=3D bytes_read; > break; > > case DW_FORM_line_strp: > { > - const char *str > - =3D per_objfile->read_line_string (buf, offset_size); > + if (line_str_data.empty ()) > + error (_("Dwarf Error: DW_FORM_line_strp used without "= \ > + "required section")); > + if (line_str_data.size () <=3D offset_size) > + error (_("Dwarf Error: DW_FORM_line_strp pointing outsi= de " \ > + "of section .debug_line")); > + > + ULONGEST str_offset =3D read_offset (parent_bfd, buf, off= set_size); > + > + const char *str; > + if (str_buf[str_offset] =3D=3D '\0') > + str =3D nullptr; > + else > + str =3D (const char *) (str_buf + str_offset); > string.emplace (str); > buf +=3D offset_size; > + break; > } > - break; > > case DW_FORM_data1: > - uint.emplace (read_1_byte (abfd, buf)); > + uint.emplace (read_1_byte (parent_bfd, buf)); > buf +=3D 1; > break; > > case DW_FORM_data2: > - uint.emplace (read_2_bytes (abfd, buf)); > + uint.emplace (read_2_bytes (parent_bfd, buf)); > buf +=3D 2; > break; > > case DW_FORM_data4: > - uint.emplace (read_4_bytes (abfd, buf)); > + uint.emplace (read_4_bytes (parent_bfd, buf)); > buf +=3D 4; > break; > > case DW_FORM_data8: > - uint.emplace (read_8_bytes (abfd, buf)); > + uint.emplace (read_8_bytes (parent_bfd, buf)); > buf +=3D 8; > break; > > @@ -205,7 +224,7 @@ read_formatted_entries (dwarf2_per_objfile *per_objfi= le, bfd *abfd, > break; > > case DW_FORM_udata: > - uint.emplace (read_unsigned_leb128 (abfd, buf, &bytes_read)= ); > + uint.emplace (read_unsigned_leb128 (parent_bfd, buf, &bytes= _read)); > buf +=3D bytes_read; > break; > > @@ -248,28 +267,30 @@ read_formatted_entries (dwarf2_per_objfile *per_obj= file, bfd *abfd, > callback (lh, fe.name, fe.d_index, fe.mod_time, fe.length); > } > > - *bufp =3D buf; > + *line_bufp =3D buf; > } > > /* See line-header.h. */ > > line_header_up > -dwarf_decode_line_header (sect_offset sect_off, bool is_dwz, > - dwarf2_per_objfile *per_objfile, > - struct dwarf2_section_info *section, > - const struct comp_unit_head *cu_header, > - const char *comp_dir) > +dwarf_decode_line_header (bfd *parent_bfd, > + gdb::array_view line_data, > + gdb::array_view line_str_data, > + const gdb_byte **debug_line_ptr, > + bool is_dwz, > + const struct comp_unit_head *cu_header, > + const char *comp_dir) > { > - const gdb_byte *line_ptr; > + const gdb_byte *line_ptr, *buf; > unsigned int bytes_read, offset_size; > int i; > const char *cur_dir, *cur_file; > > - bfd *abfd =3D section->get_bfd_owner (); > + buf =3D *debug_line_ptr; > > /* Make sure that at least there's room for the total_length field. > That could be 12 bytes long, but we're just going to fudge that. *= / > - if (to_underlying (sect_off) + 4 >=3D section->size) > + if (buf + 4 >=3D line_data.data () + line_data.size ()) > { > dwarf2_statement_list_fits_in_line_number_section_complaint (); > return 0; > @@ -277,62 +298,65 @@ dwarf_decode_line_header (sect_offset sect_off, bo= ol is_dwz, > > line_header_up lh (new line_header (comp_dir)); > > - lh->sect_off =3D sect_off; > + lh->sect_off =3D (sect_offset) (buf - line_data.data ()); > lh->offset_in_dwz =3D is_dwz; > > - line_ptr =3D section->buffer + to_underlying (sect_off); > + line_ptr =3D buf; > > /* Read in the header. */ > LONGEST unit_length > - =3D read_checked_initial_length_and_offset (abfd, line_ptr, cu_heade= r, > + =3D read_checked_initial_length_and_offset (parent_bfd, buf, cu_head= er, > &bytes_read, &offset_size); > - line_ptr +=3D bytes_read; > > - const gdb_byte *start_here =3D line_ptr; > + line_ptr +=3D bytes_read; > > - if (line_ptr + unit_length > (section->buffer + section->size)) > + if (line_ptr + unit_length > buf + line_data.size ()) > { > dwarf2_statement_list_fits_in_line_number_section_complaint (); > return 0; > } > + > + const gdb_byte *start_here =3D line_ptr; > + > lh->statement_program_end =3D start_here + unit_length; > - lh->version =3D read_2_bytes (abfd, line_ptr); > + lh->version =3D read_2_bytes (parent_bfd, line_ptr); > line_ptr +=3D 2; > if (lh->version > 5) > { > /* This is a version we don't understand. The format could have > changed in ways we don't handle properly so just punt. */ > complaint (_("unsupported version in .debug_line section")); > - return NULL; > + return nullptr; > } > if (lh->version >=3D 5) > { > gdb_byte segment_selector_size; > > /* Skip address size. */ > - read_1_byte (abfd, line_ptr); > + read_1_byte (parent_bfd, line_ptr); > line_ptr +=3D 1; > > - segment_selector_size =3D read_1_byte (abfd, line_ptr); > + segment_selector_size =3D read_1_byte (parent_bfd, line_ptr); > line_ptr +=3D 1; > if (segment_selector_size !=3D 0) > { > complaint (_("unsupported segment selector size %u " > "in .debug_line section"), > segment_selector_size); > - return NULL; > + return nullptr; > } > } > > - LONGEST header_length =3D read_offset (abfd, line_ptr, offset_size); > + LONGEST header_length =3D read_offset (parent_bfd, line_ptr, offset_si= ze); > line_ptr +=3D offset_size; > lh->statement_program_start =3D line_ptr + header_length; > - lh->minimum_instruction_length =3D read_1_byte (abfd, line_ptr); > + > + lh->minimum_instruction_length =3D read_1_byte (parent_bfd, line_ptr); > line_ptr +=3D 1; > > if (lh->version >=3D 4) > { > - lh->maximum_ops_per_instruction =3D read_1_byte (abfd, line_ptr); > + lh->maximum_ops_per_instruction =3D read_1_byte (parent_bfd, line_= ptr); > line_ptr +=3D 1; > } > else > @@ -345,41 +369,47 @@ dwarf_decode_line_header (sect_offset sect_off, bo= ol is_dwz, > "in `.debug_line' section")); > } > > - lh->default_is_stmt =3D read_1_byte (abfd, line_ptr); > + lh->default_is_stmt =3D read_1_byte (parent_bfd, line_ptr); > line_ptr +=3D 1; > - lh->line_base =3D read_1_signed_byte (abfd, line_ptr); > + > + lh->line_base =3D read_1_signed_byte (parent_bfd, line_ptr); > line_ptr +=3D 1; > - lh->line_range =3D read_1_byte (abfd, line_ptr); > + > + lh->line_range =3D read_1_byte (parent_bfd, line_ptr); > line_ptr +=3D 1; > - lh->opcode_base =3D read_1_byte (abfd, line_ptr); > + > + lh->opcode_base =3D read_1_byte (parent_bfd, line_ptr); > line_ptr +=3D 1; > + > lh->standard_opcode_lengths.reset (new unsigned char[lh->opcode_base])= ; > > lh->standard_opcode_lengths[0] =3D 1; /* This should never be used an= yway. */ > for (i =3D 1; i < lh->opcode_base; ++i) > { > - lh->standard_opcode_lengths[i] =3D read_1_byte (abfd, line_ptr); > + lh->standard_opcode_lengths[i] =3D read_1_byte (parent_bfd, line_p= tr); > line_ptr +=3D 1; > } > > if (lh->version >=3D 5) > { > /* Read directory table. */ > - read_formatted_entries (per_objfile, abfd, &line_ptr, lh.get (), > - offset_size, > - [] (struct line_header *header, const char = *name, > - dir_index d_index, unsigned int mod_tim= e, > - unsigned int length) > + read_formatted_entries > + (parent_bfd, &line_ptr, line_str_data, > + lh.get (), offset_size, > + [] (struct line_header *header, const char *name, > + dir_index d_index, unsigned int mod_time, > + unsigned int length) > { > header->add_include_dir (name); > }); > > /* Read file name table. */ > - read_formatted_entries (per_objfile, abfd, &line_ptr, lh.get (), > - offset_size, > - [] (struct line_header *header, const char = *name, > - dir_index d_index, unsigned int mod_tim= e, > - unsigned int length) > + read_formatted_entries > + (parent_bfd, &line_ptr, line_str_data, > + lh.get (), offset_size, > + [] (struct line_header *header, const char *name, > + dir_index d_index, unsigned int mod_time, > + unsigned int length) > { > header->add_file_name (name, d_index, mod_time, length); > }); > @@ -387,7 +417,7 @@ dwarf_decode_line_header (sect_offset sect_off, bool= is_dwz, > else > { > /* Read directory table. */ > - while ((cur_dir =3D read_direct_string (abfd, line_ptr, &bytes_rea= d)) !=3D NULL) > + while ((cur_dir =3D read_direct_string (parent_bfd, line_ptr, &byt= es_read)) !=3D nullptr) > { > line_ptr +=3D bytes_read; > lh->add_include_dir (cur_dir); > @@ -395,17 +425,17 @@ dwarf_decode_line_header (sect_offset sect_off, bo= ol is_dwz, > line_ptr +=3D bytes_read; > > /* Read file name table. */ > - while ((cur_file =3D read_direct_string (abfd, line_ptr, &bytes_re= ad)) !=3D NULL) > + while ((cur_file =3D read_direct_string (parent_bfd, line_ptr, &by= tes_read)) !=3D nullptr) > { > unsigned int mod_time, length; > dir_index d_index; > > line_ptr +=3D bytes_read; > - d_index =3D (dir_index) read_unsigned_leb128 (abfd, line_ptr, &= bytes_read); > + d_index =3D (dir_index) read_unsigned_leb128 (parent_bfd, line_= ptr, &bytes_read); > line_ptr +=3D bytes_read; > - mod_time =3D read_unsigned_leb128 (abfd, line_ptr, &bytes_read)= ; > + mod_time =3D read_unsigned_leb128 (parent_bfd, line_ptr, &bytes= _read); > line_ptr +=3D bytes_read; > - length =3D read_unsigned_leb128 (abfd, line_ptr, &bytes_read); > + length =3D read_unsigned_leb128 (parent_bfd, line_ptr, &bytes_r= ead); > line_ptr +=3D bytes_read; > > lh->add_file_name (cur_file, d_index, mod_time, length); > @@ -413,9 +443,40 @@ dwarf_decode_line_header (sect_offset sect_off, boo= l is_dwz, > line_ptr +=3D bytes_read; > } > > - if (line_ptr > (section->buffer + section->size)) > + if (line_ptr > (buf + line_data.size ())) > complaint (_("line number info header doesn't " > "fit in `.debug_line' section")); > > + *debug_line_ptr +=3D unit_length + offset_size; > return lh; > } > + > +line_header_up > +dwarf_decode_line_header (sect_offset sect_off, bool is_dwz, > + dwarf2_per_objfile *per_objfile, > + struct dwarf2_section_info *section, > + const struct comp_unit_head *cu_header, > + const char *comp_dir) > +{ > + struct objfile *objfile =3D per_objfile->objfile; > + struct dwarf2_per_bfd *per_bfd =3D per_objfile->per_bfd; > + > + /* Read .debug_line. */ > + dwarf2_section_info *line_sec =3D &per_bfd->line; > + bfd_size_type line_size =3D line_sec->get_size (objfile); > + > + gdb::array_view line (line_sec->buffer, line_size); > + > + /* Read .debug_line_str. */ > + dwarf2_section_info *line_str_sec =3D &per_bfd->line_str; > + bfd_size_type line_str_size =3D line_str_sec->get_size (objfile); > + > + gdb::array_view line_str (line_str_sec->buffer, > + line_str_size); > + > + const gdb_byte *line_ptr =3D line.data () + to_underlying (sect_off); > + > + return dwarf_decode_line_header > + (per_bfd->obfd, line, line_str, &line_ptr, > + is_dwz, cu_header, comp_dir); > +} > diff --git a/gdb/dwarf2/line-header.h b/gdb/dwarf2/line-header.h > index 06d2eec573b..22db9f9aa78 100644 > --- a/gdb/dwarf2/line-header.h > +++ b/gdb/dwarf2/line-header.h > @@ -217,4 +217,14 @@ extern line_header_up dwarf_decode_line_header > struct dwarf2_section_info *section, const struct comp_unit_head *cu_= header, > const char *comp_dir); > > +/* Like above but the .debug_line and .debug_line_str are stored in > + LINE_DATA and LINE_STR_DATA. *DEBUG_LINE_PTR should point to a > + statement program header within LINE_DATA. */ > + > +extern line_header_up dwarf_decode_line_header > + (bfd *parent_bfd, gdb::array_view line_data, > + gdb::array_view line_str_data, > + const gdb_byte **debug_line_ptr, bool is_dwz, > + const comp_unit_head *cu_header, const char *comp_dir); > + > #endif /* DWARF2_LINE_HEADER_H */ > diff --git a/gdb/dwarf2/read-gdb-index.c b/gdb/dwarf2/read-gdb-index.c > index 2aae83ea794..dc4823b83d4 100644 > --- a/gdb/dwarf2/read-gdb-index.c > +++ b/gdb/dwarf2/read-gdb-index.c > @@ -128,6 +128,9 @@ struct mapped_gdb_index final : public mapped_index_b= ase > } > }; > > +struct mapped_debug_line; > +typedef std::unique_ptr mapped_debug_line_up; > + > struct dwarf2_gdb_index : public dwarf2_base_index_functions > { > /* This dumps minimal information about the index. > @@ -172,6 +175,15 @@ struct dwarf2_gdb_index : public dwarf2_base_index_f= unctions > domain_enum domain, > enum search_domain kind); > > + /* If OBJFILE's debuginfo download has been deferred, use a mapped_debu= g_line > + to generate filenames. > + > + Otherwise call dwarf2_base_index_functions::map_symbol_filenames. *= / > + > + void map_symbol_filenames (struct objfile *objfile, > + gdb::function_view fu= n, > + bool need_fullname) override; > + > /* Calls dwarf2_base_index_functions::expand_all_symtabs and downloads > debuginfo if necessary. */ > void expand_all_symtabs (struct objfile *objfile) override; > @@ -179,6 +191,15 @@ struct dwarf2_gdb_index : public dwarf2_base_index_f= unctions > /* Calls dwarf2_base_index_functions::find_last_source_symtab and down= loads > debuginfo if necessary. */ > struct symtab *find_last_source_symtab (struct objfile *objfile) overr= ide; > + > + /* Filename information related to this .gdb_index. */ > + mapped_debug_line_up mdl; > + > + /* Return true if any of the filenames in this .gdb_index's .debug_lin= e > + mapping match FILE_MATCHER. Initializes the mapping if necessary. = */ > + bool filename_in_debug_line > + (objfile *objfile, > + gdb::function_view file_matcher)= ; > }; > > void > @@ -214,6 +235,30 @@ dwarf2_gdb_index::find_last_source_symtab (struct ob= jfile *objfile) > } > } > > +void > +dwarf2_gdb_index::map_symbol_filenames > + (struct objfile *objfile, > + gdb::function_view fun, > + bool need_fullname) > +{ > + try > + { > + dwarf2_base_index_functions::map_symbol_filenames (objfile, fun, > + need_fullname); > + } > + catch (const gdb_exception &e) > + { > + if ((objfile->flags & OBJF_DOWNLOAD_DEFERRED) =3D=3D 0) > + exception_print (gdb_stderr, e); > + else > + { > + if (mdl =3D=3D nullptr) > + mdl.reset (new mapped_debug_line (objfile)); > + mdl->map_filenames (fun, need_fullname); > + } > + } > +} > + > /* This dumps minimal information about the index. > It is called via "mt print objfiles". > One use is to verify .gdb_index has been loaded by the > @@ -587,6 +632,17 @@ dwarf2_gdb_index::do_expand_symtabs_matching > return result; > } > > +bool > +dwarf2_gdb_index::filename_in_debug_line > + (objfile *objfile, > + gdb::function_view file_matcher) > +{ > + if (mdl =3D=3D nullptr) > + mdl.reset (new mapped_debug_line (objfile)); > + > + return mdl->contains_matching_filename (file_matcher); > +} > + > bool > dwarf2_gdb_index::expand_symtabs_matching > (struct objfile *objfile, > @@ -615,6 +671,10 @@ dwarf2_gdb_index::expand_symtabs_matching > return false; > } > > + if (file_matcher !=3D nullptr > + && !filename_in_debug_line (objfile, file_matcher)) > + return true; > + > read_full_dwarf_from_debuginfod (objfile, this); > return true; > } > diff --git a/gdb/dwarf2/read.c b/gdb/dwarf2/read.c > index 62886e07013..97c85009242 100644 > --- a/gdb/dwarf2/read.c > +++ b/gdb/dwarf2/read.c > @@ -81,6 +81,7 @@ > #include "gdbsupport/gdb_optional.h" > #include "gdbsupport/underlying.h" > #include "gdbsupport/hash_enum.h" > +#include "gdbsupport/scoped_mmap.h" > #include "filename-seen-cache.h" > #include "producer.h" > #include > @@ -2135,6 +2136,213 @@ dw2_get_file_names (dwarf2_per_cu_data *this_cu, > return this_cu->file_names; > } > > +#if !HAVE_SYS_MMAN_H > + > +bool > +mapped_debug_line::contains_matching_filename > + (gdb::function_view file_matcher) > +{ > + return false; > +} > + > +gdb::array_view > +mapped_debug_line::read_debug_line_separate > + (char *filename, std::unique_ptr *resource) > +{ > + return {}; > +} > + > +bool > +mapped_debug_line::read_debug_line_from_debuginfod (objfile *objfile) > +{ > + return false; > +} > + > +void > +mapped_debug_line::map_filenames > + (gdb::function_view fun, > + bool need_fullname) > +{ > + return; > +} > + > +#else /* !HAVE_SYS_MMAN_H */ > + > +struct line_resource_mmap final : public index_cache_resource > +{ > + /* Try to mmap FILENAME. Throw an exception on failure, including if = the > + file doesn't exist. */ > + line_resource_mmap (const char *filename) > + : mapping (mmap_file (filename)) > + {} > + > + scoped_mmap mapping; > +}; > + > +/* See read.h. */ > + > +bool > +mapped_debug_line::contains_matching_filename > + (gdb::function_view file_matcher) > +{ > + for (line_header_up &lh : line_headers) > + for (file_entry &fe : lh->file_names ()) > + { > + const char *filename =3D fe.name; > + > + if (file_matcher (fe.name, false)) > + return true; > + > + bool basename_match =3D file_matcher (lbasename (fe.name), true); > + > + if (!basenames_may_differ && !basename_match) > + continue; > + > + /* DW_AT_comp_dir is not explicitly mentioned in the .debug_line > + until DWARF5. Since we don't have access to the CU at this > + point we just check for a partial match on the filename. > + If there is a match, the full debuginfo will be downloaded > + ane the match will be re-evalute with DW_AT_comp_dir. */ > + if (lh->version < 5 && fe.d_index =3D=3D 0) > + return basename_match; > + > + const char *dirname =3D fe.include_dir (&*lh); > + std::string fullname; > + > + if (dirname =3D=3D nullptr || IS_ABSOLUTE_PATH (filename)) > + fullname =3D filename; > + else > + fullname =3D std::string (dirname) + SLASH_STRING + filename; > + > + gdb::unique_xmalloc_ptr rewritten > + =3D rewrite_source_path (fullname.c_str ()); > + if (rewritten !=3D nullptr) > + fullname =3D rewritten.release (); > + > + if (file_matcher (fullname.c_str (), false)) > + return true; > + } > + > + return false; > +} > + > +/* See read.h. */ > + > +void > +mapped_debug_line::map_filenames > + (gdb::function_view fun, > + bool need_fullname) > +{ > + for (line_header_up &lh : line_headers) > + for (file_entry &fe : lh->file_names ()) > + { > + const char *filename =3D fe.name; > + > + if (!need_fullname) > + { > + fun (filename, nullptr); > + continue; > + } > + > + const char *dirname =3D fe.include_dir (&*lh); > + std::string fullname; > + > + if (dirname =3D=3D nullptr || IS_ABSOLUTE_PATH (filename)) > + fullname =3D filename; > + else > + fullname =3D std::string (dirname) + SLASH_STRING + filename; > + > + gdb::unique_xmalloc_ptr rewritten > + =3D rewrite_source_path (fullname.c_str ()); > + if (rewritten !=3D nullptr) > + fullname =3D rewritten.release (); > + > + fun (filename, fullname.c_str ()); > + } > +} > + > +/* See read.h. */ > + > +gdb::array_view > +mapped_debug_line::read_debug_line_separate > + (char *filename, std::unique_ptr *resource) > +{ > + if (filename =3D=3D nullptr) > + return {}; > + > + try > + { > + line_resource_mmap *mmap_resource > + =3D new line_resource_mmap (filename); > + > + resource->reset (mmap_resource); > + > + return gdb::array_view > + ((const gdb_byte *) mmap_resource->mapping.get (), > + mmap_resource->mapping.size ()); > + } > + catch (const gdb_exception &except) > + { > + exception_print (gdb_stderr, except); > + } > + > + return {}; > +} > + > +/* See read.h. */ > + > +bool > +mapped_debug_line::read_debug_line_from_debuginfod (objfile *objfile) > +{ > + const bfd_build_id *build_id =3D build_id_bfd_get (objfile->obfd.get (= )); > + if (build_id =3D=3D nullptr) > + return false; > + > + gdb::unique_xmalloc_ptr line_path; > + scoped_fd line_fd =3D debuginfod_section_query (build_id->data, > + build_id->size, > + bfd_get_filename > + (objfile->obfd.get ()), > + ".debug_line", > + &line_path); > + > + if (line_fd.get () < 0) > + return false; > + > + gdb::unique_xmalloc_ptr line_str_path; > + scoped_fd line_str_fd =3D debuginfod_section_query (build_id->data, > + build_id->size, > + bfd_get_filename > + (objfile->obfd.get = ()), > + ".debug_line_str", > + &line_str_path); > + > + line_data =3D read_debug_line_separate (line_path.get (), &line_resour= ce); > + line_str_data =3D read_debug_line_separate (line_str_path.get (), > + &line_str_resource); > + > + const gdb_byte *line_ptr =3D line_data.data (); > + > + while (line_ptr < line_data.data () + line_data.size ()) > + { > + line_header_up lh > + =3D dwarf_decode_line_header (objfile->obfd.get (), > + line_data, line_str_data, > + &line_ptr, false, > + nullptr, nullptr); > + line_headers.emplace_back (lh.release ()); > + } > + > + return true; > +} > +#endif /* !HAVE_SYS_MMAN_H */ > + > +mapped_debug_line::mapped_debug_line (objfile *objfile) > +{ > + if (!read_debug_line_from_debuginfod (objfile)) > + line_headers.clear (); > +} > + > /* A helper for the "quick" functions which computes and caches the > real path for a given file name from the line table. */ > > diff --git a/gdb/dwarf2/read.h b/gdb/dwarf2/read.h > index f15a2746586..8edda4d1de4 100644 > --- a/gdb/dwarf2/read.h > +++ b/gdb/dwarf2/read.h > @@ -34,6 +34,7 @@ > #include "gdbsupport/hash_enum.h" > #include "gdbsupport/function-view.h" > #include "gdbsupport/packed.h" > +#include "dwarf2/line-header.h" > > /* Hold 'maintenance (set|show) dwarf' commands. */ > extern struct cmd_list_element *set_dwarf_cmdlist; > @@ -961,4 +962,40 @@ extern bool read_addrmap_from_aranges (dwarf2_per_ob= jfile *per_objfile, > extern void read_full_dwarf_from_debuginfod (struct objfile *, > dwarf2_base_index_functions = *); > > +struct mapped_debug_line > +{ > + mapped_debug_line (objfile *objfile); > + > + /* Return true if any of the mapped .debug_line's filenames match > + FILE_MATCHER. */ > + > + bool contains_matching_filename > + (gdb::function_view file_matcher)= ; > + > + /* Call FUN with each filename in this mapped .debug_line. Include > + each file's fullname if NEED_FULLNAME is true. */ > + > + void map_filenames (gdb::function_view fun, > + bool need_fullname); > + > +private: > + std::vector line_headers; > + > + gdb::array_view line_data; > + gdb::array_view line_str_data; > + > + std::unique_ptr line_resource; > + std::unique_ptr line_str_resource; > + > + /* Download the .debug_line and .debug_line_str associated with OBJFIL= E > + and populate line_headers. */ > + > + bool read_debug_line_from_debuginfod (objfile *objfile); > + > + /* Initialize line_data and line_str_data with the .debug_line and > + .debug_line_str downloaded read_debug_line_from_debuginfod. */ > + > + gdb::array_view read_debug_line_separate > + (char *filename, std::unique_ptr *resource); > +}; > #endif /* DWARF2READ_H */ > diff --git a/gdb/testsuite/gdb.debuginfod/section.exp b/gdb/testsuite/gdb= .debuginfod/section.exp > index ff57c6e32b7..b5c6929fcf7 100644 > --- a/gdb/testsuite/gdb.debuginfod/section.exp > +++ b/gdb/testsuite/gdb.debuginfod/section.exp > @@ -171,6 +171,27 @@ proc_with_prefix local_url { } { > # during backtrace. > set res ".*debug info for $lib_sl1\.\.\.\r\n\#1.*" > gdb_test "bt" $res "break backtrace" > + > + clean_restart_with_prompt $sectexec "" "line 1" > + > + # List source file using .debug_line download. > + set res ".*\.debug_line.*$lib_sl1.*21.*extern void libsection2_test.= *" > + gdb_test "list $libsrc1:21" $res "line 1 list" > + > + clean_restart_with_prompt $sectexec "" "line 2" > + > + # Set breakpoint using .debug_line download. > + set res ".*section \.debug_line for $lib_sl1.*Breakpoint 2 at.*$libs= rc1.*" > + gdb_test "br $libsrc1:37" $res "line 2 br" > + > + # Continue to breakpoint. > + set res "Breakpoint 2, libsection1_test.*\"Cancelling thread\\\\n\".= *" > + gdb_test "c" $res "line 2 continue" > + > + # Check that download progress message is correctly formatted > + # when printing threads. > + set res ".*separate debug info for $lib_sl2\.\.\.\r\n.* 2 Thread.= *" > + gdb_test "info thr" $res "line thread" > } > > # Create CACHE and DB directories ready for debuginfod to use. > -- > 2.41.0 >