From: Pedro Alves <palves@redhat.com>
To: Sergio Durigan Junior <sergiodj@redhat.com>,
GDB Patches <gdb-patches@sourceware.org>
Cc: Tom Tromey <tom@tromey.com>, Eli Zaretskii <eliz@gnu.org>,
Simon Marchi <simon.marchi@ericsson.com>,
Keith Seitz <keiths@redhat.com>
Subject: Re: [PATCH v6 2/2] Implement pahole-like 'ptype /o' option
Date: Thu, 14 Dec 2017 14:50:00 -0000 [thread overview]
Message-ID: <afe48f6e-54af-6e64-fa73-786c8c3fa2a6@redhat.com> (raw)
In-Reply-To: <20171214024816.29629-3-sergiodj@redhat.com>
On 12/14/2017 02:48 AM, Sergio Durigan Junior wrote:
> A big part of this patch handles the formatting logic of 'ptype',
> which is a bit messy. The code to handle bitfield offsets, however,
> took some time to craft. My thanks to Pedro Alves for figuring things
> out and pointing me to the right direction, as well as coming up with
> a way to inspect the layout of structs with bitfields (see testcase
> for comments).
Why, thanks. :-)
> +Issuing a @kbd{ptype /o struct tuv} command would print:
> +
> +@smallexample
> +(@value{GDBP}) ptype /o struct tuv
> +/* offset | size */
> +struct tuv @{
> +/* 0 | 4 */ int a1;
> +/* XXX 4-byte hole */
> +/* 8 | 8 */ char *a2;
> +/* 16 | 4 */ int a3;
> +@} /* total size: 24 bytes */
> +@end smallexample
These examples need to be updated per the new output format.
> + You should have received a copy of the GNU General Public License
> + along with this program. If not, see <http://www.gnu.org/licenses/>. */
> +
> +/* This file will be used to test 'ptype /o' on x86_64 only. */
No longer true...
> +
> +#include <stdint.h>
> +
> +
> +# Test only works on x86_64 LP64 targets. That's how we guarantee
Remove reference to x86_64; it's no longer true.
> +# that the expected holes will be present in the struct.
> +if { ![is_lp64_target] } {
> + untested "test work only on lp64 targets"
> + return 0
> +}
> +
> +if { [prepare_for_testing "failed to prepare" $testfile $srcfile \
> + { debug c++ }] } {
> + return -1
> +}
I think we're missing a test for "ptype /oTM", to make sure that
we can still override the fact that /o implies /tm ?
I'd add at least one for "/oTM" and one for "/TMo". The
latter ends up being the same as "/o".
> +/* The default values for a struct print_offset_data. */
> +
> +struct print_offset_data print_offset_default_data =
> +{
> + 0, /* offset_bitpos */
> + 0, /* endpos */
> +};
> +
Do we really need this object ? How about just defining
the default values in-class ? See below.
> --- a/gdb/typeprint.h
> +++ b/gdb/typeprint.h
> @@ -24,6 +24,22 @@ struct ui_file;
> struct typedef_hash_table;
> struct ext_lang_type_printers;
>
> +struct print_offset_data
> +{
> + /* The offset to be applied to bitpos when PRINT_OFFSETS is true.
> + This is needed for when we are printing nested structs and want
> + to make sure that the printed offset for each field carries over
> + the offset of the outter struct. */
> + unsigned int offset_bitpos;
So here:
unsigned int offset_bitpos = 0;
> +
> + /* ENDPOS is the one-past-the-end bit position of the previous field
> + (where we expect the current field to be if there is no
> + hole). */
> + unsigned int endpos;
and
unsigned int endpos = 0;
Then you default-constructed print_offset_data objects have
the fields automatically zeroed:
print_offset_data podata;
while at it, wouldn't it be better to name this one "end_bitpos" or
something like that with "bit" in it as well?
Thanks,
Pedro Alves
next prev parent reply other threads:[~2017-12-14 14:50 UTC|newest]
Thread overview: 75+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-11-21 16:07 [PATCH] " Sergio Durigan Junior
2017-11-21 16:16 ` Sergio Durigan Junior
2017-11-21 16:50 ` Eli Zaretskii
2017-11-21 17:00 ` Sergio Durigan Junior
2017-11-21 19:14 ` Eli Zaretskii
2017-11-26 19:27 ` Tom Tromey
2017-11-27 19:54 ` Sergio Durigan Junior
2017-11-27 22:20 ` Tom Tromey
2017-11-28 0:39 ` Sergio Durigan Junior
2017-11-28 21:21 ` [PATCH v2] " Sergio Durigan Junior
2017-11-29 3:28 ` Eli Zaretskii
2017-12-04 15:03 ` Sergio Durigan Junior
2017-12-04 15:41 ` Eli Zaretskii
2017-12-04 16:47 ` Sergio Durigan Junior
2017-12-08 21:32 ` Sergio Durigan Junior
2017-12-11 15:43 ` Simon Marchi
2017-12-11 18:59 ` Sergio Durigan Junior
2017-12-11 20:45 ` Simon Marchi
2017-12-11 21:07 ` Sergio Durigan Junior
2017-12-11 22:42 ` Pedro Alves
2017-12-11 22:50 ` Sergio Durigan Junior
2017-12-11 23:46 ` Pedro Alves
2017-12-12 0:25 ` Sergio Durigan Junior
2017-12-12 0:52 ` Pedro Alves
2017-12-12 1:25 ` Simon Marchi
2017-12-12 15:50 ` John Baldwin
2017-12-12 17:04 ` Sergio Durigan Junior
2017-12-11 19:58 ` [PATCH v3 0/2] Implement pahole-like 'ptype /o' option (and do some code reorg) Sergio Durigan Junior
2017-12-11 19:58 ` [PATCH v3 2/2] Implement pahole-like 'ptype /o' option Sergio Durigan Junior
2017-12-11 21:50 ` Simon Marchi
2017-12-11 23:24 ` Sergio Durigan Junior
2017-12-12 1:32 ` Simon Marchi
2017-12-12 6:19 ` Sergio Durigan Junior
2017-12-12 18:14 ` Pedro Alves
2017-12-12 18:40 ` Sergio Durigan Junior
2017-12-12 20:12 ` Pedro Alves
2017-12-11 19:58 ` [PATCH v3 1/2] Reorganize code to handle TYPE_CODE_{STRUCT,UNION} on 'c_type_print_base' Sergio Durigan Junior
2017-12-11 20:55 ` Simon Marchi
2017-12-11 23:05 ` Sergio Durigan Junior
2017-12-11 23:43 ` [PATCH v4 0/2] Implement pahole-like 'ptype /o' option (and do some code reorg) Sergio Durigan Junior
2017-12-11 23:44 ` [PATCH v4 1/2] Reorganize code to handle TYPE_CODE_{STRUCT,UNION} on 'c_type_print_base' Sergio Durigan Junior
2017-12-11 23:44 ` [PATCH v4 2/2] Implement pahole-like 'ptype /o' option Sergio Durigan Junior
2017-12-12 0:27 ` Sergio Durigan Junior
2017-12-12 0:29 ` Sergio Durigan Junior
2017-12-12 1:59 ` Simon Marchi
2017-12-12 3:39 ` Eli Zaretskii
2017-12-13 3:17 ` [PATCH v5 0/2] Implement pahole-like 'ptype /o' option (and do some code reorg) Sergio Durigan Junior
2017-12-13 3:17 ` [PATCH v5 2/2] Implement pahole-like 'ptype /o' option Sergio Durigan Junior
2017-12-13 4:50 ` Simon Marchi
2017-12-13 16:42 ` Sergio Durigan Junior
2017-12-13 16:17 ` Eli Zaretskii
2017-12-13 17:14 ` Sergio Durigan Junior
2017-12-13 16:19 ` Pedro Alves
2017-12-13 17:13 ` Sergio Durigan Junior
2017-12-13 20:36 ` Sergio Durigan Junior
2017-12-13 21:22 ` Pedro Alves
2017-12-13 21:30 ` Pedro Alves
2017-12-13 21:34 ` Sergio Durigan Junior
2017-12-13 16:20 ` Pedro Alves
2017-12-13 17:41 ` Sergio Durigan Junior
2017-12-13 3:17 ` [PATCH v5 1/2] Reorganize code to handle TYPE_CODE_{STRUCT,UNION} on 'c_type_print_base' Sergio Durigan Junior
2017-12-14 2:48 ` [PATCH v6 0/2] Implement pahole-like 'ptype /o' option (and do some code reorg) Sergio Durigan Junior
2017-12-14 2:48 ` [PATCH v6 2/2] Implement pahole-like 'ptype /o' option Sergio Durigan Junior
2017-12-14 14:19 ` Pedro Alves
2017-12-14 20:31 ` Sergio Durigan Junior
2017-12-14 14:50 ` Pedro Alves [this message]
2017-12-14 20:29 ` Sergio Durigan Junior
2017-12-14 16:30 ` Eli Zaretskii
2017-12-14 2:48 ` [PATCH v6 1/2] Reorganize code to handle TYPE_CODE_{STRUCT,UNION} on 'c_type_print_base' Sergio Durigan Junior
2017-12-15 1:12 ` [PATCH v7 0/2] Implement pahole-like 'ptype /o' option (and do some code reorg) Sergio Durigan Junior
2017-12-15 1:12 ` [PATCH v7 1/2] Reorganize code to handle TYPE_CODE_{STRUCT,UNION} on 'c_type_print_base' Sergio Durigan Junior
2017-12-15 1:13 ` [PATCH v7 2/2] Implement pahole-like 'ptype /o' option Sergio Durigan Junior
2017-12-15 17:24 ` Pedro Alves
2017-12-15 20:04 ` Sergio Durigan Junior
2017-12-15 20:11 ` [PATCH v7 0/2] Implement pahole-like 'ptype /o' option (and do some code reorg) Sergio Durigan Junior
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=afe48f6e-54af-6e64-fa73-786c8c3fa2a6@redhat.com \
--to=palves@redhat.com \
--cc=eliz@gnu.org \
--cc=gdb-patches@sourceware.org \
--cc=keiths@redhat.com \
--cc=sergiodj@redhat.com \
--cc=simon.marchi@ericsson.com \
--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;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).