This is the mail archive of the
mailing list for the Archer project.
[delayed-symfile] [commit] Fix a testcase regression due to: (nodebugging symbols found)
- From: Jan Kratochvil <jan dot kratochvil at redhat dot com>
- To: Tom Tromey <tromey at redhat dot com>
- Cc: archer at sourceware dot org
- Date: Wed, 25 Feb 2009 00:13:42 +0100
- Subject: [delayed-symfile] [commit] Fix a testcase regression due to: (nodebugging symbols found)
FSF GDB code prints `(no debugging symbols found)' when at none of the loaded
objfiles has any debug info.
Former archer-tromey-delayed-symfile branch did print `(no debugging symbols
found)' on each objfile that has no debug info. At least on
(modified)kernel-184.108.40.206-170.2.24.fc10.x86_64 with no kernel-debuginfo
installed its vDSO is missing the debug info. Thus gdb.base/annota1.exp
regressed on it as it does not expect `(no debugging symbols found)'.
This change is disputable as patched Archer is not completely compatible with
former FSF GDB behavior.
* It could be made compatible but it would be more compatible code.
* I find the behavior with this patch more meaningful than the FSF GDB one.
* It has no regression for the current FSF GDB testsuite.
* symfile.c (symbol_file_add_with_addrs_or_offsets): Permit the
optional printing of `(no debugging symbols found)' only for MAINLINE.
gdb/symfile.c | 4 +++-
1 files changed, 3 insertions(+), 1 deletions(-)
diff --git a/gdb/symfile.c b/gdb/symfile.c
index f86c216..cd34a7b 100644
@@ -1064,7 +1064,9 @@ symbol_file_add_with_addrs_or_offsets (bfd *abfd, int from_tty,
- if (print_symbol_loading && !has_any_debug_symbols (objfile))
+ /* has_any_debug_symbols is not fully compatible with the former calls which
+ would just be needlessly expensive here. */
+ if (print_symbol_loading && !has_any_debug_symbols (objfile) && mainline)
printf_unfiltered (_("(no debugging symbols found)"));