From: Doug Evans <dje@google.com>
To: Yao Qi <yao@codesourcery.com>
Cc: Tom Tromey <tromey@redhat.com>,
gdb-patches <gdb-patches@sourceware.org>
Subject: Re: [PATCH v3 03/15] Read CTF by the ctf target
Date: Tue, 09 Apr 2013 18:41:00 -0000 [thread overview]
Message-ID: <20836.14726.472470.645674@ruffy2.mtv.corp.google.com> (raw)
In-Reply-To: <5163D6FE.1090106@codesourcery.com>
Yao Qi writes:
> On 04/09/2013 12:58 AM, Doug Evans wrote:
> >> + if (ret < 0)
> > > + {
> > > + ctf_destroy ();
> > > + error (_("Unable to use libbabeltrace open \"%s\""), dirname);
> >
> > The wording of this error message (here and below) is awkward.
> > I don't know libbabeltrace enough to know if this is accurate, but
> > it reads better.
> >
> > error (_("Unable to use libbabeltrace on \"%s\""), dirname);
> >
> > or
> >
> > error (_("Unable to use libbabeltrace on directory \"%s\""), dirname);
>
> I changed the message to
>
> error (_("Unable to use libbabeltrace on directory \"%s\""),
> dirname);
>
> >
> > Also, IWBN to add a more specific reason for the failure,
> > but I'm not sure how easy that would be.
> > Is there a bt_foo function akin to strerror?
> >
>
> Unfortunately, there isn't such function.
>
>
> > > +
> > > + if (bt_iter_next (bt_ctf_get_iter (ctf_iter)) < 0)
> > > + break;
> > > + }
> > > +
> > > + /* Restore the position. */
> > > + bt_iter_set_pos (bt_ctf_get_iter (ctf_iter), pos);
> > > +
> >
> > Remove blank line.
> >
> > > +}
>
> Removed.
>
>
> > > +
> > > + scope = bt_ctf_get_top_level_scope (event,
> > > + BT_EVENT_FIELDS);
> > > +
> > > + def = bt_ctf_get_field (event, scope, "address");
> > > + maddr = bt_ctf_get_uint64 (def);
> > > + def = bt_ctf_get_field (event, scope, "length");
> > > + mlen = (uint16_t) bt_ctf_get_uint64 (def);
> >
> > I don't know libbabeltrace, I'm assuming length can be at most 16 bits.
> >
>
> Right. "length" is defined as uint16_t in metadata, on the other hand,
> "length" is got from the remote target, which is 16-bit as well.
>
> > > +
> > > +/* This is the implementation of target_ops method to_traceframe_info.
> > > + Iterate the events whose name is "memory", in current
> > > + frame, extract memory range information, and return them in
> > > + traceframe_info. */
> > > +
> > > +static struct traceframe_info *
> > > +ctf_traceframe_info (void)
> > > +{
> > > + struct traceframe_info *info = XCNEW (struct traceframe_info);
> > > + const char *name;
> > > + struct bt_iter_pos *pos;
> > > +
> > > + gdb_assert (ctf_iter != NULL);
> > > + /* Save the current position. */
> > > + pos = bt_iter_get_pos (bt_ctf_get_iter (ctf_iter));
> > > + gdb_assert (pos->type == BT_SEEK_RESTORE);
> > > +
> > > + do
> > > + {
> > > + struct bt_ctf_event *event
> > > + = bt_ctf_iter_read_event (ctf_iter);
> > > +
> > > + name = bt_ctf_event_name (event);
> > > +
> > > + if (name == NULL || strcmp (name, "register") == 0
> > > + || strcmp (name, "frame") == 0)
> > > + ;
> > > + else if (strcmp (name, "memory") == 0)
> > > + {
> > > + const struct bt_definition *scope
> > > + = bt_ctf_get_top_level_scope (event,
> > > + BT_EVENT_FIELDS);
> > > + const struct bt_definition *def;
> > > + struct mem_range *r;
> > > +
> > > + r = VEC_safe_push (mem_range_s, info->memory, NULL);
> > > + def = bt_ctf_get_field (event, scope, "address");
> > > + r->start = bt_ctf_get_uint64 (def);
> > > +
> > > + def = bt_ctf_get_field (event, scope, "length");
> > > + r->length = (uint16_t) bt_ctf_get_uint64 (def);
> > > + }
> > > + else
> > > + warning (_("Unhandled trace block type (%s) "
> > > + "while building trace frame info."),
> > > + name);
> >
> > These days we prefer multi-line single statements for if/else/etc.
> > to be wrapped in braces. i.e.,
> >
> > {
> > warning (_("Unhandled trace block type (%s) "
> > "while building trace frame info."),
> > name);
> > }
> >
>
> Fixed. I knew that single statement with comments should be warped by
> braces, like:
>
> if (foo)
> /* Comments. */
> bar ();
>
> after reading GDB internals, I realize that "warning message in
> multiple lines" falls into this category as well. You are right.
>
> > > +
> > > +/* module initialization */
> >
> > blank line
> >
> > > +void
> > > +_initialize_ctf (void)
>
> Fixed.
>
> --
> Yao ( )
>
> gdb:
>
> 2013-04-09 Hui Zhu <hui@codesourcery.com>
> Yao Qi <yao@codesourcery.com>
>
> * configure.ac: Check libbabeltrace is installed.
> * config.in: Regenerate.
> * configure: Regenerate.
> * Makefile.in (LIBBABELTRACE): New.
> (CLIBS): Add LIBBABELTRACE.
> * ctf.c (ctx, ctf_iter, trace_dirname): New.
> (ctf_destroy, ctf_open_dir, ctf_open): New.
> (ctf_close, ctf_files_info): New.
> (ctf_fetch_registers, ctf_xfer_partial): New.
> (ctf_get_trace_state_variable_value): New.
> (ctf_get_tpnum_from_frame_event): New.
> (ctf_get_traceframe_address): New.
> (ctf_trace_find, ctf_has_stack): New.
> (ctf_has_registers, ctf_traceframe_info, init_ctf_ops): New.
> (_initialize_ctf): New.
> * tracepoint.c (get_tracepoint_number): New
> (struct traceframe_info, trace_regblock_size): Move it to ...
> * tracepoint.h: ... here.
> (get_tracepoint_number): Declare it.
Thanks.
The patch is ok.
next prev parent reply other threads:[~2013-04-09 15:53 UTC|newest]
Thread overview: 60+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-03-09 3:48 [PATCH v3 00/15] CTF Support Yao Qi
2013-03-09 3:49 ` [PATCH v3 13/15] Test on saving tracepoint defs Yao Qi
2013-03-12 20:10 ` Tom Tromey
2013-03-13 9:48 ` Yao Qi
2013-03-13 14:58 ` Tom Tromey
2013-03-09 3:49 ` [PATCH v3 06/15] Write status to CTF and read Yao Qi
2013-03-12 19:31 ` Tom Tromey
2013-03-14 18:06 ` Doug Evans
2013-03-29 14:46 ` Yao Qi
2013-03-29 16:47 ` Doug Evans
2013-03-09 3:49 ` [PATCH v3 11/15] Write tsv definition in " Yao Qi
2013-03-12 19:56 ` Tom Tromey
2013-03-09 3:49 ` [PATCH v3 01/15] Refactor 'tsave' Yao Qi
2013-03-12 18:30 ` Tom Tromey
2013-03-13 10:33 ` Yao Qi
2013-03-13 19:26 ` Tom Tromey
2013-03-14 9:17 ` Yao Qi
2013-03-09 3:49 ` [PATCH v3 08/15] Write 'stop_desc' of trace status to tfile Yao Qi
2013-03-12 19:15 ` Tom Tromey
2013-03-09 3:49 ` [PATCH v3 09/15] Check the tstatus output on tfile target Yao Qi
2013-03-12 19:45 ` Tom Tromey
2013-03-09 3:49 ` [PATCH v3 07/15] Write trace notes and username into tfile Yao Qi
2013-03-12 19:35 ` Tom Tromey
2013-03-09 3:49 ` [PATCH v3 05/15] ctf test: report.exp Yao Qi
2013-03-12 19:25 ` Tom Tromey
2013-03-09 3:49 ` [PATCH v3 14/15] Test on saving tracepoint defs: CTF Yao Qi
2013-03-12 20:23 ` Tom Tromey
2013-03-13 9:55 ` Yao Qi
2013-03-13 15:49 ` Tom Tromey
2013-03-09 3:49 ` [PATCH v3 15/15] MAINTAINERS Yao Qi
2013-03-09 3:49 ` [PATCH v3 12/15] Write tracepoint definition in CTF and read Yao Qi
2013-03-12 20:02 ` Tom Tromey
2013-03-14 18:34 ` Doug Evans
2013-03-09 3:49 ` [PATCH v3 10/15] tstatus.exp: ctf Yao Qi
2013-03-12 19:48 ` Tom Tromey
2013-03-09 3:49 ` [PATCH v3 02/15] Save trace into CTF format Yao Qi
2013-03-12 18:49 ` Tom Tromey
2013-03-13 10:33 ` Yao Qi
2013-03-13 19:30 ` Tom Tromey
2013-03-14 17:49 ` Doug Evans
2013-03-09 3:49 ` [PATCH v3 04/15] ctf doc and NEWS Yao Qi
2013-03-09 3:49 ` [PATCH v3 03/15] Read CTF by the ctf target Yao Qi
2013-03-13 20:09 ` Tom Tromey
2013-03-14 13:23 ` Yao Qi
2013-03-14 14:59 ` Tom Tromey
2013-03-14 16:57 ` Doug Evans
2013-03-14 17:39 ` Doug Evans
2013-03-25 13:33 ` Yao Qi
2013-03-25 17:14 ` Doug Evans
2013-03-26 16:16 ` Yao Qi
2013-03-29 17:56 ` Doug Evans
2013-04-08 14:19 ` Yao Qi
2013-04-08 21:48 ` Doug Evans
2013-04-09 15:23 ` Yao Qi
2013-04-09 18:41 ` Doug Evans [this message]
2013-04-10 19:16 ` [PATCH v3 00/15] CTF Support Yao Qi
2013-04-11 22:59 ` [patch] Regenerate config.in [Re: [PATCH v3 00/15] CTF Support] Jan Kratochvil
2013-04-12 22:27 ` [commit] " Jan Kratochvil
2014-08-04 18:59 ` Incorrect placement of babeltrace gdb/NEWS item " Jan Kratochvil
2014-08-06 1:30 ` Yao Qi
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=20836.14726.472470.645674@ruffy2.mtv.corp.google.com \
--to=dje@google.com \
--cc=gdb-patches@sourceware.org \
--cc=tromey@redhat.com \
--cc=yao@codesourcery.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