From: Tom de Vries <tdevries@suse.de>
To: Tom Tromey <tom@tromey.com>
Cc: gdb-patches@sourceware.org
Subject: Re: [PATCH 3/4] [gdb/testsuite] Add missing type_offset in gdb.dwarf/debug-names-tu.exp
Date: Sat, 23 May 2026 10:15:59 +0200 [thread overview]
Message-ID: <a0be9c8d-6ca0-440d-954a-91debbb90a2f@suse.de> (raw)
In-Reply-To: <87se7j7ahe.fsf@tromey.com>
On 5/22/26 5:03 PM, Tom Tromey wrote:
>>>>>> "Tom" == Tom de Vries <tdevries@suse.de> writes:
>
> Tom> With llvm-dwarfdump we run into this warning:
> Tom> ...
> Tom> $ llvm-dwarfdump -a -v debug-names-tu > /dev/null
> Tom> warning: DWARF type unit at offset 0x00000028 has its relocated type_offset \
> Tom> 0x00000028 pointing inside the header
> Tom> ...
>
> Tom> Fix this by adding the missing type_offset.
>
> Tom> Likewise in gdb.dwarf2/debug-names-bad-cu-index.exp.
>
> Tom> Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=34157
>
> This looks good but I wonder if the type offset should simply be
> mandatory to 'proc tu'? Like should it reject an empty string and
> require a real label?
>
I came across fission-type-unit-locexpr.exp, which has:
...
# Stub for "int" type unit.
tu {
version 4
} 0xCAFE1 "" {
type_unit {
DW_AT_GNU_dwo_name ${::gdb_test_file_name}.dwo DW_FORM_strp
DW_AT_comp_dir ${outdir}
} {
}
}
...
We could fix this by doing:
...
+ declare_labels none1
# Stub for "int" type unit.
tu {
version 4
- } 0xCAFE1 "" {
+ } 0xCAFE1 none1 {
type_unit {
DW_AT_GNU_dwo_name ${::gdb_test_file_name}.dwo DW_FORM_strp
DW_AT_comp_dir ${outdir}
} {
+ none1:
}
}
...
and indeed that fixes the warning, but eventually there might be another
warning complaining about that (points to abbrev number 0 instead of type).
So I left this as is.
Thanks,
- Tom
> Approved-By: Tom Tromey <tom@tromey.com>
>
> Tom
next prev parent reply other threads:[~2026-05-23 8:16 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-05-22 14:37 [PATCH 0/4] [gdb/testsuite] Fix some more dwarf assembly test-cases Tom de Vries
2026-05-22 14:37 ` [PATCH 1/4] [gdb/testsuite] Remove invalid .debug_line section in gdb.dwarf2/dw2-icycle.exp Tom de Vries
2026-05-22 14:55 ` Tom Tromey
2026-05-22 14:37 ` [PATCH 2/4] [gdb/testsuite] Complete 64-bit port of gdb.dwarf2/dw2-skip-prologue.exp Tom de Vries
2026-05-22 15:07 ` Tom Tromey
2026-05-22 14:37 ` [PATCH 3/4] [gdb/testsuite] Add missing type_offset in gdb.dwarf/debug-names-tu.exp Tom de Vries
2026-05-22 15:03 ` Tom Tromey
2026-05-23 8:15 ` Tom de Vries [this message]
2026-05-22 14:37 ` [PATCH 4/4] [gdb/testsuite] Improve dwarf assembly for implicit const in .debug_names Tom de Vries
2026-05-22 14:54 ` Tom Tromey
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=a0be9c8d-6ca0-440d-954a-91debbb90a2f@suse.de \
--to=tdevries@suse.de \
--cc=gdb-patches@sourceware.org \
--cc=tom@tromey.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox