Separate the job of reading the line number info statement program
authorJim Blandy <jimb@codesourcery.com>
Mon, 6 May 2002 21:00:21 +0000 (21:00 +0000)
committerJim Blandy <jimb@codesourcery.com>
Mon, 6 May 2002 21:00:21 +0000 (21:00 +0000)
commitdebd256d71bf60d1285d31c5742909f535ccfab3
tree98cc88f3b9a381f7e93b5b240708e864bffb0e0b
parentac3aafc7a6f8a61724529611901b112abbdb5861
Separate the job of reading the line number info statement program
header (...expialidocious) out into its own function.
* dwarf2read.c (struct line_head, struct filenames, struct
directories): Replace with...
(struct line_header): New structure, containing the full
contents of the statement program header, including the
include directory and file name tables.
(read_file_scope): If we have line number info, instead of just
calling dwarf_decode_lines to do all the work, call
dwarf_decode_line_header first to get a `struct line_header'
containing the data in the statement program header, and then
pass that to dwarf_decode_lines, which will pick up where that
left off.  Be sure to clean up the `struct line_header' object.
(dwarf_decode_line_header, free_line_header, add_include_dir,
add_file_name): New functions.
(dwarf_decode_lines): Move all the code to read the statement
program header into dwarf_decode_line_header.  Take the line
header it built as the first argument, instead of the offset to
the compilation unit's line number info.  Use the new `struct
line_header' type instead of the old structures.  No need to do
cleanups here now, since we don't allocate anything.
(dwarf2_statement_list_fits_in_line_number_section,
dwarf2_line_header_too_long): New complaints.
gdb/ChangeLog
gdb/dwarf2read.c
This page took 0.027442 seconds and 4 git commands to generate.