public inbox for gdb-prs@sourceware.org
help / color / mirror / Atom feed
* [Bug fortran/30180] New: FAIL: gdb.fortran/module.exp: print var_d
@ 2023-02-28 13:05 vries at gcc dot gnu.org
  0 siblings, 0 replies; only message in thread
From: vries at gcc dot gnu.org @ 2023-02-28 13:05 UTC (permalink / raw)
  To: gdb-prs

https://sourceware.org/bugzilla/show_bug.cgi?id=30180

            Bug ID: 30180
           Summary: FAIL: gdb.fortran/module.exp: print var_d
           Product: gdb
           Version: HEAD
            Status: NEW
          Severity: normal
          Priority: P2
         Component: fortran
          Assignee: unassigned at sourceware dot org
          Reporter: vries at gcc dot gnu.org
  Target Milestone: ---

On powerpc64le-linux (centos 7, with gcc 4.8.5), I run into:
...
(gdb) PASS: gdb.fortran/module.exp: print var_c
print var_d^M
Missing ELF symbol "var_d".^M
(gdb) FAIL: gdb.fortran/module.exp: print var_d
...

Looking at the debuginfo, we have an imported_declaration:
...
 <2><19b>: Abbrev Number: 12 (DW_TAG_imported_declaration)
    <19c>   DW_AT_decl_file   : 1
    <19d>   DW_AT_decl_line   : 59
    <19e>   DW_AT_name        : var_d
    <1a2>   DW_AT_import      : <0x20b> [Abbrev Number: 14 (DW_TAG_variable)]
...
referring to a declaration:
...
 <2><20b>: Abbrev Number: 14 (DW_TAG_variable)
    <20c>   DW_AT_name        : var_c
    <210>   DW_AT_decl_file   : 1
    <211>   DW_AT_decl_line   : 33
    <212>   DW_AT_linkage_name: __modmany_MOD_var_c
    <216>   DW_AT_type        : <0x60>
    <21a>   DW_AT_external    : 1
    <21a>   DW_AT_declaration : 1
...

In contrast, on x86_64-linux (openSUSE Leap 15.4, with gcc 7.5.0), I have:
...
(gdb) PASS: gdb.fortran/module.exp: print var_c
print var_d^M
$8 = 12^M
(gdb) PASS: gdb.fortran/module.exp: print var_d
...
with debuginfo also mentioning an imported_declaration:
...
 <2><2df>: Abbrev Number: 12 (DW_TAG_imported_declaration)
    <2e0>   DW_AT_decl_file   : 1
    <2e1>   DW_AT_decl_line   : 59
    <2e2>   DW_AT_name        : var_d
    <2e6>   DW_AT_import      : <0x1b7> [Abbrev Number: 3 (DW_TAG_variable)]
...
but referring to a definition, with location attribute:
...
 <2><1b7>: Abbrev Number: 3 (DW_TAG_variable)
    <1b8>   DW_AT_name        : var_c
    <1bc>   DW_AT_decl_file   : 1
    <1bd>   DW_AT_decl_line   : 33
    <1be>   DW_AT_linkage_name: __modmany_MOD_var_c
    <1c2>   DW_AT_type        : <0xf2>
    <1c6>   DW_AT_external    : 1
    <1c6>   DW_AT_location    : 9 byte block: 3 44 20 40 0 0 0 0 0
                                (DW_OP_addr: 402044)
...

So, I'd classify this as a testsuite issue and xfail it for some compiler
versions, but I wonder.  We have this trick in gdb where we combine a decl with
a minimal symbol in order to get the location, so why doesn't this work?  Is
this not enabled for fortran?

The minimal symbol seems to be there:
...
$ nm  outputs/gdb.fortran/module/module | grep __modmany_MOD_var_c
0000000010020058 D __modmany_MOD_var_c
...
so it could work.

-- 
You are receiving this mail because:
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] only message in thread

only message in thread, other threads:[~2023-02-28 13:05 UTC | newest]

Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-02-28 13:05 [Bug fortran/30180] New: FAIL: gdb.fortran/module.exp: print var_d vries 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).