From: Markus Metzger <markus.t.metzger@intel.com>
To: palves@redhat.com
Cc: gdb-patches@sourceware.org
Subject: [PATCH v2 10/17] btrace: temporarily set inferior_ptid in record_btrace_start_replaying
Date: Fri, 11 Sep 2015 06:52:00 -0000 [thread overview]
Message-ID: <1441954298-25298-11-git-send-email-markus.t.metzger@intel.com> (raw)
In-Reply-To: <1441954298-25298-1-git-send-email-markus.t.metzger@intel.com>
Get_current_frame uses inferior_ptid. In record_btrace_start_replaying,
we need to get the current frame of the argument thread. So far, this
has always been inferior_ptid. With non-stop, this is not guaranteed.
Temporarily set inferior_ptid to the ptid of the argument thread.
We already temporarily set the argument thread's executing flag to false.
Move both into a new function get_thread_current_frame that does the temporary
adjustments, calls get_current_frame, and restores the previous values.
2015-09-11 Markus Metzger <markus.t.metzger@intel.com>
gdb/
* record-btrace.c (get_thread_current_frame): New.
(record_btrace_start_replaying): Call get_thread_current_frame.
---
gdb/record-btrace.c | 70 ++++++++++++++++++++++++++++++++++++++---------------
1 file changed, 51 insertions(+), 19 deletions(-)
diff --git a/gdb/record-btrace.c b/gdb/record-btrace.c
index c25b8f4..0514471 100644
--- a/gdb/record-btrace.c
+++ b/gdb/record-btrace.c
@@ -1707,6 +1707,55 @@ record_btrace_resume_thread (struct thread_info *tp,
btinfo->flags |= flag;
}
+/* Get the current frame for TP. */
+
+static struct frame_info *
+get_thread_current_frame (struct thread_info *tp)
+{
+ struct frame_info *frame;
+ ptid_t old_inferior_ptid;
+ int executing;
+
+ /* Set INFERIOR_PTID, which is implicitly used by get_current_frame. */
+ old_inferior_ptid = inferior_ptid;
+ inferior_ptid = tp->ptid;
+
+ /* Clear the executing flag to allow changes to the current frame.
+ We are not actually running, yet. We just started a reverse execution
+ command or a record goto command.
+ For the latter, EXECUTING is false and this has no effect.
+ For the former, EXECUTING is true and we're in to_wait, about to
+ move the thread. Since we need to recompute the stack, we temporarily
+ set EXECUTING to flase. */
+ executing = is_executing (inferior_ptid);
+ set_executing (inferior_ptid, 0);
+
+ frame = NULL;
+ TRY
+ {
+ frame = get_current_frame ();
+ }
+ CATCH (except, RETURN_MASK_ALL)
+ {
+ /* Restore the previous execution state. */
+ set_executing (inferior_ptid, executing);
+
+ /* Restore the previous inferior_ptid. */
+ inferior_ptid = old_inferior_ptid;
+
+ throw_exception (except);
+ }
+ END_CATCH
+
+ /* Restore the previous execution state. */
+ set_executing (inferior_ptid, executing);
+
+ /* Restore the previous inferior_ptid. */
+ inferior_ptid = old_inferior_ptid;
+
+ return frame;
+}
+
/* Start replaying a thread. */
static struct btrace_insn_iterator *
@@ -1714,7 +1763,6 @@ record_btrace_start_replaying (struct thread_info *tp)
{
struct btrace_insn_iterator *replay;
struct btrace_thread_info *btinfo;
- int executing;
btinfo = &tp->btrace;
replay = NULL;
@@ -1723,16 +1771,6 @@ record_btrace_start_replaying (struct thread_info *tp)
if (btinfo->begin == NULL)
return NULL;
- /* Clear the executing flag to allow changes to the current frame.
- We are not actually running, yet. We just started a reverse execution
- command or a record goto command.
- For the latter, EXECUTING is false and this has no effect.
- For the former, EXECUTING is true and we're in to_wait, about to
- move the thread. Since we need to recompute the stack, we temporarily
- set EXECUTING to flase. */
- executing = is_executing (tp->ptid);
- set_executing (tp->ptid, 0);
-
/* GDB stores the current frame_id when stepping in order to detects steps
into subroutines.
Since frames are computed differently when we're replaying, we need to
@@ -1745,7 +1783,7 @@ record_btrace_start_replaying (struct thread_info *tp)
int upd_step_frame_id, upd_step_stack_frame_id;
/* The current frame without replaying - computed via normal unwind. */
- frame = get_current_frame ();
+ frame = get_thread_current_frame (tp);
frame_id = get_frame_id (frame);
/* Check if we need to update any stepping-related frame id's. */
@@ -1777,7 +1815,7 @@ record_btrace_start_replaying (struct thread_info *tp)
registers_changed_ptid (tp->ptid);
/* The current frame with replaying - computed via btrace unwind. */
- frame = get_current_frame ();
+ frame = get_thread_current_frame (tp);
frame_id = get_frame_id (frame);
/* Replace stepping related frames where necessary. */
@@ -1788,9 +1826,6 @@ record_btrace_start_replaying (struct thread_info *tp)
}
CATCH (except, RETURN_MASK_ALL)
{
- /* Restore the previous execution state. */
- set_executing (tp->ptid, executing);
-
xfree (btinfo->replay);
btinfo->replay = NULL;
@@ -1800,9 +1835,6 @@ record_btrace_start_replaying (struct thread_info *tp)
}
END_CATCH
- /* Restore the previous execution state. */
- set_executing (tp->ptid, executing);
-
return replay;
}
--
1.8.3.1
next prev parent reply other threads:[~2015-09-11 6:51 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-09-11 6:52 [PATCH v2 00/17] record btrace: non-stop and ASNS Markus Metzger
2015-09-11 6:51 ` [PATCH v2 07/17] btrace: add missing NO_HISTORY Markus Metzger
2015-09-11 6:51 ` [PATCH v2 11/17] btrace: async Markus Metzger
2015-09-11 6:51 ` [PATCH v2 04/17] btrace: extract the breakpoint check from record_btrace_step_thread Markus Metzger
2015-09-11 6:51 ` [PATCH v2 12/17] infrun: switch to NO_HISTORY thread Markus Metzger
2015-09-11 6:51 ` [PATCH v2 08/17] btrace: lock-step Markus Metzger
2015-09-11 6:51 ` [PATCH v2 01/17] btrace: fix non-stop check in to_wait Markus Metzger
2015-09-11 6:51 ` [PATCH v2 06/17] btrace: move breakpoint checking into stepping functions Markus Metzger
2015-09-11 6:51 ` [PATCH v2 05/17] btrace: split record_btrace_step_thread Markus Metzger
2015-09-11 6:51 ` [PATCH v2 03/17] btrace: improve stepping debugging Markus Metzger
2015-09-11 6:52 ` [PATCH v2 13/17] btrace: non-stop Markus Metzger
2015-09-11 6:58 ` Eli Zaretskii
2015-09-11 6:52 ` [PATCH v2 15/17] btrace: allow full memory and register access for non-replaying threads Markus Metzger
2015-09-11 6:52 ` [PATCH v2 09/17] btrace: resume all requested threads Markus Metzger
2015-09-11 6:52 ` Markus Metzger [this message]
2015-09-11 6:52 ` [PATCH v2 14/17] target, record: add PTID argument to to_record_is_replaying Markus Metzger
2015-09-11 6:52 ` [PATCH v2 02/17] btrace: support to_stop Markus Metzger
2015-09-11 6:52 ` [PATCH v2 16/17] target: add to_record_stop_replaying target method Markus Metzger
2015-09-11 6:52 ` [PATCH v2 17/17] infrun: scheduler-locking reverse Markus Metzger
2015-09-11 7:01 ` Eli Zaretskii
2015-09-11 8:55 ` Pedro Alves
2015-09-11 9:02 ` Eli Zaretskii
2015-09-16 12:28 ` Metzger, Markus T
2015-09-16 12:54 ` Pedro Alves
2015-09-16 13:32 ` Pedro Alves
2015-09-16 13:56 ` Metzger, Markus T
2015-09-16 14:25 ` Pedro Alves
2015-09-11 7:56 ` [PATCH v2 00/17] record btrace: non-stop and ASNS Pedro Alves
2015-09-11 8:02 ` Metzger, Markus T
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=1441954298-25298-11-git-send-email-markus.t.metzger@intel.com \
--to=markus.t.metzger@intel.com \
--cc=gdb-patches@sourceware.org \
--cc=palves@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