From: Pedro Alves <palves@redhat.com>
To: Jan Kratochvil <jan.kratochvil@redhat.com>, gdb-patches@sourceware.org
Cc: Phil Muldoon <pmuldoon@redhat.com>
Subject: Re: [PATCH v3 1/9] Code cleanup: Make parts of print_command_1 public
Date: Wed, 29 Apr 2015 15:44:00 -0000 [thread overview]
Message-ID: <5540FC34.3050200@redhat.com> (raw)
In-Reply-To: <20150411194333.29128.30245.stgit@host1.jankratochvil.net>
On 04/11/2015 08:43 PM, Jan Kratochvil wrote:
> +/* Parse print command format string and update *EXPP, return it allocated,
> + caller has to call xfree for it. Return NULL if no format string has been
> + found. CMDNAME should name the current command. */
> +
> +struct format_data *
> +print_command_parse_format (const char **expp, const char *cmdname)
> +{
I read the series, and AFAICS, this will be used by
compile_print_command in patch #7. But then AFAICS, compile_print_command
leaks fmtp.
I think this all ends up simpler if it follows the pattern that
the current code already follows. That is, instead of having
print_command_parse_format "struct format_data" on the heap,
make the function take an output format parameter:
void
print_command_parse_format (const char **expp, const char *cmdname,
struct format_data *fmt)
{
> + struct format_data *fmtp;
> + const char *exp = *expp;
> + struct cleanup *cleanup;
> +
> + if (exp == NULL || *exp != '/')
> + return NULL;
and in this case set the defaults in the passed in fmt, like the
original code does:
> - if (exp && *exp == '/')
> - {
...
> - }
> - else
> - {
> - fmt.count = 1;
> - fmt.format = 0;
> - fmt.size = 0;
> - fmt.raw = 0;
> - }
(the else branch)
> + exp++;
> +
> + fmtp = xmalloc (sizeof (*fmtp));
> + cleanup = make_cleanup (xfree, fmtp);
> + *fmtp = decode_format (&exp, last_format, 0);
> + validate_format (*fmtp, cmdname);
> + last_format = fmtp->format;
> +
> + discard_cleanups (cleanup);
... and then no need for the xmalloc and cleanups...
> + *expp = exp;
> + return fmtp;
> +}
> +
> +
> +/* Print VAL to console, including recording it to the history. */
> +
> +void
> +print_value (struct value *val, const struct format_data *fmtp)
> +{
> + struct value_print_options opts;
> + int histindex = record_latest_value (val);
> +
> + annotate_value_history_begin (histindex, value_type (val));
> +
> + printf_filtered ("$%d = ", histindex);
> +
> + annotate_value_history_value ();
> +
> + get_formatted_print_options (&opts, (fmtp == NULL ? 0 : fmtp->format));
> + opts.raw = (fmtp == NULL ? 0 : fmtp->raw);
> +
> + print_formatted (val, (fmtp == NULL ? 0 : fmtp->size), &opts, gdb_stdout);
... and no need for the NULL checks here...
> + printf_filtered ("\n");
> +
> + annotate_value_history_end ();
> +}
> +
> /* Evaluate string EXP as an expression in the current language and
> print the resulting value. EXP may contain a format specifier as the
> first argument ("/x myvar" for example, to print myvar in hex). */
> @@ -947,25 +997,9 @@ static void
> print_command_1 (const char *exp, int voidprint)
> {
> struct expression *expr;
> - struct cleanup *old_chain = make_cleanup (null_cleanup, NULL);
> - char format = 0;
> struct value *val;
> - struct format_data fmt;
... and this 'fmt' variable stays on the stack ...
> -
> - if (exp && *exp == '/')
> - {
> - exp++;
> - fmt = decode_format (&exp, last_format, 0);
> - validate_format (fmt, "print");
> - last_format = format = fmt.format;
> - }
> - else
> - {
> - fmt.count = 1;
> - fmt.format = 0;
> - fmt.size = 0;
> - fmt.raw = 0;
> - }
> + struct format_data *fmtp = print_command_parse_format (&exp, "print");
> + struct cleanup *old_chain = make_cleanup (xfree, fmtp);
... and no need for this cleanup either.
>
> if (exp && *exp)
> {
> @@ -978,24 +1012,7 @@ print_command_1 (const char *exp, int voidprint)
>
> if (voidprint || (val && value_type (val) &&
> TYPE_CODE (value_type (val)) != TYPE_CODE_VOID))
> - {
> - struct value_print_options opts;
> - int histindex = record_latest_value (val);
> -
> - annotate_value_history_begin (histindex, value_type (val));
> -
> - printf_filtered ("$%d = ", histindex);
> -
> - annotate_value_history_value ();
> -
> - get_formatted_print_options (&opts, format);
> - opts.raw = fmt.raw;
> -
> - print_formatted (val, fmt.size, &opts, gdb_stdout);
> - printf_filtered ("\n");
> -
> - annotate_value_history_end ();
> - }
> + print_value (val, fmtp);
>
> do_cleanups (old_chain);
> }
> diff --git a/gdb/valprint.h b/gdb/valprint.h
> index e3d0137..3ab531f 100644
> --- a/gdb/valprint.h
> +++ b/gdb/valprint.h
> @@ -217,4 +217,9 @@ extern void output_command_const (const char *args, int from_tty);
>
> extern int val_print_scalar_type_p (struct type *type);
>
> +struct format_data;
> +extern struct format_data *print_command_parse_format (const char **expp,
> + const char *cmdname);
> +extern void print_value (struct value *val, const struct format_data *fmtp);
> +
> #endif
>
Thanks,
Pedro Alves
next prev parent reply other threads:[~2015-04-29 15:43 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-04-11 19:43 [PATCH v3 0/9] compile: compile print&printf Jan Kratochvil
2015-04-11 19:43 ` [PATCH v3 2/9] compile: Distribute scope, add scope_data Jan Kratochvil
2015-04-29 15:44 ` Pedro Alves
2015-04-11 19:43 ` [PATCH v3 1/9] Code cleanup: Make parts of print_command_1 public Jan Kratochvil
2015-04-29 15:44 ` Pedro Alves [this message]
2015-04-30 0:24 ` Jan Kratochvil
2015-04-11 19:43 ` [PATCH v3 3/9] Code cleanup: compile: Constify some parameters Jan Kratochvil
2015-04-29 15:47 ` Pedro Alves
2015-05-06 18:58 ` [commit] " Jan Kratochvil
2015-04-11 19:44 ` [PATCH v3 8/9] compile: New compile printf Jan Kratochvil
2015-04-29 15:54 ` Pedro Alves
2015-05-03 14:06 ` Jan Kratochvil
2015-05-06 10:22 ` Pedro Alves
2015-05-06 11:30 ` Jan Kratochvil
2015-05-06 11:47 ` Pedro Alves
2015-04-11 19:44 ` [PATCH v3 5/9] compile: Use -Wall, not -w Jan Kratochvil
2015-04-29 15:49 ` Pedro Alves
2015-05-03 14:05 ` Jan Kratochvil
2015-05-06 10:21 ` Pedro Alves
2015-04-11 19:44 ` [PATCH v3 9/9] compile: compile printf: gdbserver support Jan Kratochvil
2015-04-26 9:33 ` Jan Kratochvil
2015-04-29 18:19 ` Pedro Alves
2015-05-03 14:06 ` Jan Kratochvil
2015-05-06 10:22 ` Pedro Alves
2015-04-29 16:12 ` Pedro Alves
2015-04-11 19:44 ` [PATCH v3 6/9] Code cleanup: compile: func_addr -> func_sym Jan Kratochvil
2015-04-29 15:52 ` Pedro Alves
2015-04-11 19:44 ` [PATCH v3 4/9] compile: Support relocation to GNU-IFUNCs Jan Kratochvil
2015-04-29 15:48 ` Pedro Alves
2015-05-06 19:00 ` [commit] " Jan Kratochvil
2015-04-11 19:44 ` [PATCH v3 7/9] compile: New 'compile print' Jan Kratochvil
2015-04-29 15:53 ` Pedro Alves
2015-05-03 14:06 ` Jan Kratochvil
2015-05-06 10:22 ` Pedro Alves
2015-05-06 12:23 ` Jan Kratochvil
2015-05-06 14:11 ` Pedro Alves
2015-05-06 19:18 ` Jan Kratochvil
2015-05-15 16:35 ` Pedro Alves
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=5540FC34.3050200@redhat.com \
--to=palves@redhat.com \
--cc=gdb-patches@sourceware.org \
--cc=jan.kratochvil@redhat.com \
--cc=pmuldoon@redhat.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