public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug debug/54734] New: Debug info for C++ and LTO misses types
@ 2012-09-28 10:37 rguenth at gcc dot gnu.org
2020-07-16 19:07 ` [Bug debug/54734] " mark at gcc dot gnu.org
2020-07-17 6:29 ` rguenth at gcc dot gnu.org
0 siblings, 2 replies; 3+ messages in thread
From: rguenth at gcc dot gnu.org @ 2012-09-28 10:37 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54734
Bug #: 54734
Summary: Debug info for C++ and LTO misses types
Classification: Unclassified
Product: gcc
Version: 4.8.0
Status: UNCONFIRMED
Keywords: lto
Severity: normal
Priority: P3
Component: debug
AssignedTo: unassigned@gcc.gnu.org
ReportedBy: rguenth@gcc.gnu.org
Consider the following reduced testcase (from standard library use):
template <class T> class basic_string;
typedef basic_string<char> sstring;
template <class T>
class basic_string
{
public:
typedef T type;
};
extern template class basic_string<char>;
int main()
{
sstring s;
return sstring::type (0);
}
without LTO we have
<1><130>: Abbrev Number: 2 (DW_TAG_typedef)
<131> DW_AT_name : (indirect string, offset: 0x98): sstring
<135> DW_AT_decl_file : 1
<136> DW_AT_decl_line : 2
<137> DW_AT_type : <0x13b>
<1><13b>: Abbrev Number: 3 (DW_TAG_class_type)
<13c> DW_AT_name : (indirect string, offset: 0xa0):
basic_string<char>
<140> DW_AT_byte_size : 1
<141> DW_AT_decl_file : 1
<142> DW_AT_decl_line : 5
<143> DW_AT_sibling : <0x15a>
<2><147>: Abbrev Number: 2 (DW_TAG_typedef)
<148> DW_AT_name : (indirect string, offset: 0x8e): type
<14c> DW_AT_decl_file : 1
<14d> DW_AT_decl_line : 8
<14e> DW_AT_type : <0x15a>
while with LTO we get only
<1><137>: Abbrev Number: 3 (DW_TAG_typedef)
<138> DW_AT_name : (indirect string, offset: 0xf8): sstring
<13c> DW_AT_decl_file : 1
<13d> DW_AT_decl_line : 11
<13e> DW_AT_type : <0x142>
<1><142>: Abbrev Number: 4 (DW_TAG_structure_type)
<143> DW_AT_name : (indirect string, offset: 0xeb): basic_string
<147> DW_AT_declaration : 1
which results from gen_struct_or_union_type_die on the type seeing
a TYPE_STUB_DECL with TYPE_DECL_SUPPRESS_DEBUG set.
Somehow the C++ frontend "manually" outputs the desired info and then for
later processing directs dwarf2out.c to not output anything?
This probably all boils down to LTO "ignoring" debug-hook calls from the
frontend, but is there a way to recover from that? For example by
clearing TYPE_DECL_SUPPRESS_DEBUG for LTO? Or is that used in "meaningful"
ways as well?
^ permalink raw reply [flat|nested] 3+ messages in thread
* [Bug debug/54734] Debug info for C++ and LTO misses types
2012-09-28 10:37 [Bug debug/54734] New: Debug info for C++ and LTO misses types rguenth at gcc dot gnu.org
@ 2020-07-16 19:07 ` mark at gcc dot gnu.org
2020-07-17 6:29 ` rguenth at gcc dot gnu.org
1 sibling, 0 replies; 3+ messages in thread
From: mark at gcc dot gnu.org @ 2020-07-16 19:07 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=54734
Mark Wielaard <mark at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |mark at gcc dot gnu.org
--- Comment #1 from Mark Wielaard <mark at gcc dot gnu.org> ---
This seems fixed with recent GCC. At least with GNU C++14 10.1.1 20200507 (Red
Hat 10.1.1-1) I now get:
[ 6b] typedef abbrev: 2
name (strp) "sstring"
decl_file (data1) t.cc (1)
decl_line (data1) 2
decl_column (data1) 28
type (ref4) [ 77]
[ 77] class_type abbrev: 3
name (strp) "basic_string<char>"
byte_size (data1) 1
decl_file (data1) t.cc (1)
decl_line (data1) 5
decl_column (data1) 7
sibling (ref4) [ 99]
[ 84] typedef abbrev: 4
name (strp) "type"
decl_file (data1) t.cc (1)
decl_line (data1) 8
decl_column (data1) 13
type (ref4) [ 99]
accessibility (data1) public (1)
[ 91] template_type_parameter abbrev: 5
name (string) "T"
type (ref4) [ 99]
[ 99] base_type abbrev: 6
byte_size (data1) 1
encoding (data1) signed_char (6)
name (strp) "char"
^ permalink raw reply [flat|nested] 3+ messages in thread
* [Bug debug/54734] Debug info for C++ and LTO misses types
2012-09-28 10:37 [Bug debug/54734] New: Debug info for C++ and LTO misses types rguenth at gcc dot gnu.org
2020-07-16 19:07 ` [Bug debug/54734] " mark at gcc dot gnu.org
@ 2020-07-17 6:29 ` rguenth at gcc dot gnu.org
1 sibling, 0 replies; 3+ messages in thread
From: rguenth at gcc dot gnu.org @ 2020-07-17 6:29 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=54734
Richard Biener <rguenth at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|UNCONFIRMED |RESOLVED
Resolution|--- |FIXED
Known to work| |8.1.0
--- Comment #2 from Richard Biener <rguenth at gcc dot gnu.org> ---
This was fixed with LTO early debug in GCC 8.
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2020-07-17 6:29 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2012-09-28 10:37 [Bug debug/54734] New: Debug info for C++ and LTO misses types rguenth at gcc dot gnu.org
2020-07-16 19:07 ` [Bug debug/54734] " mark at gcc dot gnu.org
2020-07-17 6:29 ` rguenth at gcc dot gnu.org
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).