public inbox for gdb-patches@sourceware.org
 help / color / mirror / Atom feed
From: Felix Willgerodt <felix.willgerodt@intel.com>
To: gdb-patches@sourceware.org, markus.t.metzger@intel.com, simark@simark.ca
Subject: [PATCH v10 04/10] btrace: Handle stepping and goto for auxiliary instructions.
Date: Tue, 18 Jul 2023 13:56:31 +0200	[thread overview]
Message-ID: <20230718115637.3531-5-felix.willgerodt@intel.com> (raw)
In-Reply-To: <20230718115637.3531-1-felix.willgerodt@intel.com>

Print the auxiliary data when stepping. Don't allow to goto an auxiliary
instruction.

This patch is in preparation for the new ptwrite feature, which is based on
auxiliary instructions.
---
 gdb/record-btrace.c | 65 +++++++++++++++++++++++++++++++++++++--------
 1 file changed, 54 insertions(+), 11 deletions(-)

diff --git a/gdb/record-btrace.c b/gdb/record-btrace.c
index 02e71889eaf..c93b3d7c8de 100644
--- a/gdb/record-btrace.c
+++ b/gdb/record-btrace.c
@@ -2380,9 +2380,13 @@ record_btrace_single_step_forward (struct thread_info *tp)
     return btrace_step_stopped ();
 
   /* Skip gaps during replay.  If we end up at a gap (at the end of the trace),
-     jump back to the instruction at which we started.  */
+     jump back to the instruction at which we started.  If we're stepping a
+     BTRACE_INSN_AUX instruction, print the auxiliary data and skip the
+     instruction.  */
+
   start = *replay;
-  do
+
+  for (;;)
     {
       unsigned int steps;
 
@@ -2394,8 +2398,23 @@ record_btrace_single_step_forward (struct thread_info *tp)
 	  *replay = start;
 	  return btrace_step_no_history ();
 	}
+
+      const struct btrace_insn *insn = btrace_insn_get (replay);
+      if (insn == nullptr)
+	continue;
+
+      /* If we're stepping a BTRACE_INSN_AUX instruction, print the auxiliary
+	 data and skip the instruction.  */
+      if (insn->iclass == BTRACE_INSN_AUX)
+	{
+	  gdb_printf ("[%s]\n",
+		      btinfo->aux_data.at (insn->aux_data_index).c_str ());
+	  continue;
+	}
+
+      /* We have an instruction, we are done.  */
+      break;
     }
-  while (btrace_insn_get (replay) == NULL);
 
   /* Determine the end of the instruction trace.  */
   btrace_insn_end (&end, btinfo);
@@ -2426,9 +2445,12 @@ record_btrace_single_step_backward (struct thread_info *tp)
 
   /* If we can't step any further, we reached the end of the history.
      Skip gaps during replay.  If we end up at a gap (at the beginning of
-     the trace), jump back to the instruction at which we started.  */
+     the trace), jump back to the instruction at which we started.
+     If we're stepping a BTRACE_INSN_AUX instruction, print the auxiliary
+     data and skip the instruction.  */
   start = *replay;
-  do
+
+  for (;;)
     {
       unsigned int steps;
 
@@ -2438,8 +2460,22 @@ record_btrace_single_step_backward (struct thread_info *tp)
 	  *replay = start;
 	  return btrace_step_no_history ();
 	}
+
+      const struct btrace_insn *insn = btrace_insn_get (replay);
+      if (insn == nullptr)
+	continue;
+
+      /* Check if we're stepping a BTRACE_INSN_AUX instruction and skip it.  */
+      if (insn->iclass == BTRACE_INSN_AUX)
+	{
+	  gdb_printf ("[%s]\n",
+		      btinfo->aux_data.at (insn->aux_data_index).c_str ());
+	  continue;
+	}
+
+      /* We have an instruction, we are done.  */
+      break;
     }
-  while (btrace_insn_get (replay) == NULL);
 
   /* Check if we're stepping a breakpoint.
 
@@ -2861,26 +2897,33 @@ record_btrace_target::goto_record_end ()
 /* The goto_record method of target record-btrace.  */
 
 void
-record_btrace_target::goto_record (ULONGEST insn)
+record_btrace_target::goto_record (ULONGEST insn_number)
 {
   struct thread_info *tp;
   struct btrace_insn_iterator it;
   unsigned int number;
   int found;
 
-  number = insn;
+  number = insn_number;
 
   /* Check for wrap-arounds.  */
-  if (number != insn)
+  if (number != insn_number)
     error (_("Instruction number out of range."));
 
   tp = require_btrace_thread ();
 
   found = btrace_find_insn_by_number (&it, &tp->btrace, number);
 
-  /* Check if the instruction could not be found or is a gap.  */
-  if (found == 0 || btrace_insn_get (&it) == NULL)
+  /* Check if the instruction could not be found or is a gap or an
+     auxiliary instruction.  */
+  if (found == 0)
+    error (_("No such instruction."));
+
+  const struct btrace_insn *insn = btrace_insn_get (&it);
+  if (insn == NULL)
     error (_("No such instruction."));
+  if (insn->iclass == BTRACE_INSN_AUX)
+    error (_("Can't go to an auxiliary instruction."));
 
   record_btrace_set_replay (tp, &it);
 }
-- 
2.34.1

Intel Deutschland GmbH
Registered Address: Am Campeon 10, 85579 Neubiberg, Germany
Tel: +49 89 99 8853-0, www.intel.de <http://www.intel.de>
Managing Directors: Christin Eisenschmid, Sharon Heck, Tiffany Doon Silva  
Chairperson of the Supervisory Board: Nicole Lau
Registered Office: Munich
Commercial Register: Amtsgericht Muenchen HRB 186928


  parent reply	other threads:[~2023-07-18 11:57 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-07-18 11:56 [PATCH v10 00/10] Extensions for PTWRITE Felix Willgerodt
2023-07-18 11:56 ` [PATCH v10 01/10] btrace: Introduce auxiliary instructions Felix Willgerodt
2023-07-18 11:56 ` [PATCH v10 02/10] btrace: Enable auxiliary instructions in record instruction-history Felix Willgerodt
2023-07-18 11:56 ` [PATCH v10 03/10] btrace: Enable auxiliary instructions in record function-call-history Felix Willgerodt
2023-07-18 11:56 ` Felix Willgerodt [this message]
2023-07-18 11:56 ` [PATCH v10 05/10] python: Introduce gdb.RecordAuxiliary class Felix Willgerodt
2023-07-25 13:30   ` Metzger, Markus T
2023-08-07  9:08     ` Willgerodt, Felix
2023-08-08  9:37       ` Metzger, Markus T
2023-08-09  9:32         ` Willgerodt, Felix
2023-08-09 10:00           ` Metzger, Markus T
2023-08-30  8:39             ` Willgerodt, Felix
2023-07-18 11:56 ` [PATCH v10 06/10] python: Add clear() to gdb.Record Felix Willgerodt
2023-07-18 13:00   ` Eli Zaretskii
2023-07-18 11:56 ` [PATCH v10 07/10] btrace, gdbserver: Add ptwrite to btrace_config_pt Felix Willgerodt
2023-07-18 11:56 ` [PATCH v10 08/10] btrace, linux: Enable ptwrite packets Felix Willgerodt
2023-07-25 13:30   ` Metzger, Markus T
2023-08-07  9:10     ` Willgerodt, Felix
2023-08-08 10:13       ` Metzger, Markus T
2023-08-09  9:31         ` Willgerodt, Felix
2023-08-09  9:48           ` Metzger, Markus T
2023-08-30  8:42             ` Willgerodt, Felix
2023-09-06 14:07               ` Metzger, Markus T
2023-07-18 11:56 ` [PATCH v10 09/10] btrace, python: Enable ptwrite filter registration Felix Willgerodt
2023-07-18 11:56 ` [PATCH v10 10/10] btrace: Extend ptwrite event decoding Felix Willgerodt
2023-07-18 13:01   ` Eli Zaretskii

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=20230718115637.3531-5-felix.willgerodt@intel.com \
    --to=felix.willgerodt@intel.com \
    --cc=gdb-patches@sourceware.org \
    --cc=markus.t.metzger@intel.com \
    --cc=simark@simark.ca \
    /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).