public inbox for gdb-patches@sourceware.org
 help / color / mirror / Atom feed
From: Don Breazeal <donb@codesourcery.com>
To: <gdb-patches@sourceware.org>
Subject: [PATCH 07/16 v3] Extended-remote arch-specific follow fork
Date: Fri, 31 Oct 2014 23:29:00 -0000	[thread overview]
Message-ID: <1414798134-11536-5-git-send-email-donb@codesourcery.com> (raw)
In-Reply-To: <1408580964-27916-1-git-send-email-donb@codesourcery.com>

This patch implements the architecture-specific pieces of follow-fork,
which in the current implementation copyies the parent's debug register
state into the new child's data structures.  This is required for x86,
arm, aarch64, and mips.

I followed the native implementation as closely as I could by
implementing a new linux_target_ops function 'new_fork', which is
analogous to 'linux_nat_new_fork' in linux-nat.c.  In gdbserver, the debug
registers are stored in the process list, instead of an
architecture-specific list, so the function arguments are process_info
pointers instead of an lwp_info and a pid as in the native implementation.

In the MIPS implementation the debug register mirror is stored differently
from x86, ARM, and aarch64, so instead of doing a simple structure assignment
I had to clone the list of watchpoint structures.

Tested using gdb.threads/watchpoint-fork.exp on x86, and ran manual tests
on a MIPS board and an ARM board.

I don't currently have access to an aarch64 board, so again if someone is
able to test this easily, please do.

Thanks
--Don

gdb/gdbserver/
2014-10-31  Don Breazeal  <donb@codesourcery.com>

	* linux-aarch64-low.c (aarch64_linux_new_fork): New function.
	(the_low_target) <new_fork>: Initialize new member.
	* linux-arm-low.c (arm_new_fork): New function.
	(the_low_target) <new_fork>: Initialize new member.
	* linux-low.c (handle_extended_wait): Call new target function
	new_fork.
	* linux-low.h (struct linux_target_ops) <new_fork>: New member.
	* linux-mips-low.c (mips_add_watchpoint): New function
	extracted from mips_insert_point.
	(the_low_target) <new_fork>: Initialize new member.
	(mips_linux_new_fork): New function.
	(mips_insert_point): Call mips_add_watchpoint.
	* linux-x86-low.c (x86_linux_new_fork): New function.
	(the_low_target) <new_fork>: Initialize new member.

---
 gdb/gdbserver/linux-aarch64-low.c |   28 +++++++++++++
 gdb/gdbserver/linux-arm-low.c     |   42 ++++++++++++++++++++
 gdb/gdbserver/linux-low.c         |    4 ++
 gdb/gdbserver/linux-low.h         |    3 +
 gdb/gdbserver/linux-mips-low.c    |   76 +++++++++++++++++++++++++++++++------
 gdb/gdbserver/linux-x86-low.c     |   29 ++++++++++++++
 6 files changed, 170 insertions(+), 12 deletions(-)

diff --git a/gdb/gdbserver/linux-aarch64-low.c b/gdb/gdbserver/linux-aarch64-low.c
index 654b319..31ec0d7 100644
--- a/gdb/gdbserver/linux-aarch64-low.c
+++ b/gdb/gdbserver/linux-aarch64-low.c
@@ -1129,6 +1129,33 @@ aarch64_linux_new_thread (void)
   return info;
 }
 
+static void
+aarch64_linux_new_fork (struct process_info *parent,
+			struct process_info *child)
+{
+  /* These are allocated by linux_add_process.  */
+  gdb_assert (parent->private != NULL
+	      && parent->private->arch_private != NULL);
+  gdb_assert (child->private != NULL
+	      && child->private->arch_private != NULL);
+
+  /* Linux kernel before 2.6.33 commit
+     72f674d203cd230426437cdcf7dd6f681dad8b0d
+     will inherit hardware debug registers from parent
+     on fork/vfork/clone.  Newer Linux kernels create such tasks with
+     zeroed debug registers.
+
+     GDB core assumes the child inherits the watchpoints/hw
+     breakpoints of the parent, and will remove them all from the
+     forked off process.  Copy the debug registers mirrors into the
+     new process so that all breakpoints and watchpoints can be
+     removed together.  The debug registers mirror will become zeroed
+     in the end before detaching the forked off process, thus making
+     this compatible with older Linux kernels too.  */
+
+  *child->private->arch_private = *parent->private->arch_private;
+}
+
 /* Called when resuming a thread.
    If the debug regs have changed, update the thread's copies.  */
 
@@ -1293,6 +1320,7 @@ struct linux_target_ops the_low_target =
   NULL,
   aarch64_linux_new_process,
   aarch64_linux_new_thread,
+  aarch64_linux_new_fork,
   aarch64_linux_prepare_to_resume,
 };
 
diff --git a/gdb/gdbserver/linux-arm-low.c b/gdb/gdbserver/linux-arm-low.c
index 8b72523..fe9c34e 100644
--- a/gdb/gdbserver/linux-arm-low.c
+++ b/gdb/gdbserver/linux-arm-low.c
@@ -717,6 +717,47 @@ arm_new_thread (void)
   return info;
 }
 
+static void
+arm_new_fork (struct process_info *parent, struct process_info *child)
+{
+  struct arch_process_info *parent_proc_info = parent->private->arch_private;
+  struct arch_process_info *child_proc_info = child->private->arch_private;
+  struct lwp_info *child_lwp;
+  struct arch_lwp_info *child_lwp_info;
+  int i;
+
+  /* These are allocated by linux_add_process.  */
+  gdb_assert (parent->private != NULL
+	      && parent->private->arch_private != NULL);
+  gdb_assert (child->private != NULL
+	      && child->private->arch_private != NULL);
+
+  /* Linux kernel before 2.6.33 commit
+     72f674d203cd230426437cdcf7dd6f681dad8b0d
+     will inherit hardware debug registers from parent
+     on fork/vfork/clone.  Newer Linux kernels create such tasks with
+     zeroed debug registers.
+
+     GDB core assumes the child inherits the watchpoints/hw
+     breakpoints of the parent, and will remove them all from the
+     forked off process.  Copy the debug registers mirrors into the
+     new process so that all breakpoints and watchpoints can be
+     removed together.  The debug registers mirror will become zeroed
+     in the end before detaching the forked off process, thus making
+     this compatible with older Linux kernels too.  */
+
+  *child_proc_info = *parent_proc_info;
+
+  /* Mark all the hardware breakpoints and watchpoints as changed to
+     make sure that the registers will be updated.  */
+  child_lwp = find_lwp_pid (ptid_of (child));
+  child_lwp_info = child_lwp->arch_private;
+  for (i = 0; i < MAX_BPTS; i++)
+    child_lwp_info->bpts_changed[i] = 1;
+  for (i = 0; i < MAX_WPTS; i++)
+    child_lwp_info->wpts_changed[i] = 1;
+}
+
 /* Called when resuming a thread.
    If the debug regs have changed, update the thread's copies.  */
 static void
@@ -920,6 +961,7 @@ struct linux_target_ops the_low_target = {
   NULL, /* siginfo_fixup */
   arm_new_process,
   arm_new_thread,
+  arm_new_fork,
   arm_prepare_to_resume,
 };
 
diff --git a/gdb/gdbserver/linux-low.c b/gdb/gdbserver/linux-low.c
index db7ef09..4f6d664 100644
--- a/gdb/gdbserver/linux-low.c
+++ b/gdb/gdbserver/linux-low.c
@@ -438,6 +438,10 @@ handle_extended_wait (struct lwp_info *event_child, int wstat)
 	  child_proc->tdesc = tdesc;
 	  child_lwp->must_set_ptrace_flags = 1;
 
+	  /* Clone arch-specific process data.  */
+	  if (the_low_target.new_fork != NULL)
+	    the_low_target.new_fork (parent_proc, child_proc);
+
 	  /* Save fork info for target processing.  */
 	  current_thread->pending_follow.kind = TARGET_WAITKIND_FORKED;
 	  current_thread->pending_follow.value.related_pid = ptid;
diff --git a/gdb/gdbserver/linux-low.h b/gdb/gdbserver/linux-low.h
index a903430..53d1c24 100644
--- a/gdb/gdbserver/linux-low.h
+++ b/gdb/gdbserver/linux-low.h
@@ -185,6 +185,9 @@ struct linux_target_ops
      allocate it here.  */
   struct arch_lwp_info * (*new_thread) (void);
 
+  /* Hook to call, if any, when a new fork is attached.  */
+  void (*new_fork) (struct process_info *parent, struct process_info *child);
+
   /* Hook to call prior to resuming a thread.  */
   void (*prepare_to_resume) (struct lwp_info *);
 
diff --git a/gdb/gdbserver/linux-mips-low.c b/gdb/gdbserver/linux-mips-low.c
index 0fc8cb4..d5db79d 100644
--- a/gdb/gdbserver/linux-mips-low.c
+++ b/gdb/gdbserver/linux-mips-low.c
@@ -344,6 +344,68 @@ mips_linux_new_thread (void)
   return info;
 }
 
+/* Create a new mips_watchpoint and add it to the list.  */
+
+static void
+mips_add_watchpoint (struct arch_process_info *private, CORE_ADDR addr,
+		     int len, int watch_type)
+{
+  struct mips_watchpoint *new_watch;
+  struct mips_watchpoint **pw;
+
+  new_watch = xmalloc (sizeof (struct mips_watchpoint));
+  new_watch->addr = addr;
+  new_watch->len = len;
+  new_watch->type = watch_type;
+  new_watch->next = NULL;
+
+  pw = &private->current_watches;
+  while (*pw != NULL)
+    pw = &(*pw)->next;
+  *pw = new_watch;
+}
+
+/* Hook to call when a new fork is attached.  */
+
+static void
+mips_linux_new_fork (struct process_info *parent,
+			struct process_info *child)
+{
+  struct arch_process_info *parent_private;
+  struct arch_process_info *child_private;
+  struct mips_watchpoint *wp;
+
+  /* These are allocated by linux_add_process.  */
+  gdb_assert (parent->private != NULL
+	      && parent->private->arch_private != NULL);
+  gdb_assert (child->private != NULL
+	      && child->private->arch_private != NULL);
+
+  /* Linux kernel before 2.6.33 commit
+     72f674d203cd230426437cdcf7dd6f681dad8b0d
+     will inherit hardware debug registers from parent
+     on fork/vfork/clone.  Newer Linux kernels create such tasks with
+     zeroed debug registers.
+
+     GDB core assumes the child inherits the watchpoints/hw
+     breakpoints of the parent, and will remove them all from the
+     forked off process.  Copy the debug registers mirrors into the
+     new process so that all breakpoints and watchpoints can be
+     removed together.  The debug registers mirror will become zeroed
+     in the end before detaching the forked off process, thus making
+     this compatible with older Linux kernels too.  */
+
+  parent_private = parent->private->arch_private;
+  child_private = child->private->arch_private;
+
+  child_private->watch_readback_valid = parent_private->watch_readback_valid;
+  child_private->watch_readback = parent_private->watch_readback;
+
+  for (wp = parent_private->current_watches; wp != NULL; wp = wp->next)
+    mips_add_watchpoint (child_private, wp->addr, wp->len, wp->type);
+
+  child_private->watch_mirror = parent_private->watch_mirror;
+}
 /* This is the implementation of linux_target_ops method
    prepare_to_resume.  If the watch regs have changed, update the
    thread's copies.  */
@@ -397,8 +459,6 @@ mips_insert_point (enum raw_bkpt_type type, CORE_ADDR addr,
   struct process_info *proc = current_process ();
   struct arch_process_info *private = proc->private->arch_private;
   struct pt_watch_regs regs;
-  struct mips_watchpoint *new_watch;
-  struct mips_watchpoint **pw;
   int pid;
   long lwpid;
   enum target_hw_bp_type watch_type;
@@ -425,16 +485,7 @@ mips_insert_point (enum raw_bkpt_type type, CORE_ADDR addr,
     return -1;
 
   /* It fit.  Stick it on the end of the list.  */
-  new_watch = xmalloc (sizeof (struct mips_watchpoint));
-  new_watch->addr = addr;
-  new_watch->len = len;
-  new_watch->type = watch_type;
-  new_watch->next = NULL;
-
-  pw = &private->current_watches;
-  while (*pw != NULL)
-    pw = &(*pw)->next;
-  *pw = new_watch;
+  mips_add_watchpoint (private, addr, len, watch_type);
 
   private->watch_mirror = regs;
 
@@ -845,6 +896,7 @@ struct linux_target_ops the_low_target = {
   NULL, /* siginfo_fixup */
   mips_linux_new_process,
   mips_linux_new_thread,
+  mips_linux_new_fork,
   mips_linux_prepare_to_resume
 };
 
diff --git a/gdb/gdbserver/linux-x86-low.c b/gdb/gdbserver/linux-x86-low.c
index 838e7c9..10b8a56 100644
--- a/gdb/gdbserver/linux-x86-low.c
+++ b/gdb/gdbserver/linux-x86-low.c
@@ -768,6 +768,34 @@ x86_linux_new_thread (void)
   return info;
 }
 
+/* Target routine for linux_new_fork.  */
+
+static void
+x86_linux_new_fork (struct process_info *parent, struct process_info *child)
+{
+  /* These are allocated by linux_add_process.  */
+  gdb_assert (parent->private != NULL
+	      && parent->private->arch_private != NULL);
+  gdb_assert (child->private != NULL
+	      && child->private->arch_private != NULL);
+
+  /* Linux kernel before 2.6.33 commit
+     72f674d203cd230426437cdcf7dd6f681dad8b0d
+     will inherit hardware debug registers from parent
+     on fork/vfork/clone.  Newer Linux kernels create such tasks with
+     zeroed debug registers.
+
+     GDB core assumes the child inherits the watchpoints/hw
+     breakpoints of the parent, and will remove them all from the
+     forked off process.  Copy the debug registers mirrors into the
+     new process so that all breakpoints and watchpoints can be
+     removed together.  The debug registers mirror will become zeroed
+     in the end before detaching the forked off process, thus making
+     this compatible with older Linux kernels too.  */
+
+  *child->private->arch_private = *parent->private->arch_private;
+}
+
 /* Called when resuming a thread.
    If the debug regs have changed, update the thread's copies.  */
 
@@ -3428,6 +3456,7 @@ struct linux_target_ops the_low_target =
   x86_siginfo_fixup,
   x86_linux_new_process,
   x86_linux_new_thread,
+  x86_linux_new_fork,
   x86_linux_prepare_to_resume,
   x86_linux_process_qsupported,
   x86_supports_tracepoints,
-- 
1.7.0.4

  parent reply	other threads:[~2014-10-31 23:29 UTC|newest]

Thread overview: 110+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-08-07 18:00 [PATCH 00/10] Linux extended-remote fork events Don Breazeal
2014-08-07 18:00 ` [PATCH 04/10] Enhance extended ptrace event setup Don Breazeal
2014-08-13 17:50   ` Breazeal, Don
2014-08-07 18:00 ` [PATCH 01/10] Refactor native follow-fork Don Breazeal
2014-08-07 18:00 ` [PATCH 02/10] Refactor follow-fork message printing Don Breazeal
2014-08-07 18:00 ` [PATCH 06/10] Extended-remote follow fork Don Breazeal
2014-08-07 18:00 ` [PATCH 05/10] GDBserver clone breakpoint list Don Breazeal
2014-08-07 18:00 ` [PATCH 07/10] Extended-remote arch-specific follow fork Don Breazeal
2014-08-07 18:00 ` [PATCH 03/10] Refactor extended ptrace event status Don Breazeal
2014-08-07 18:01 ` [PATCH 10/10] Extended-remote fork event documentation Don Breazeal
2014-08-07 19:31   ` Eli Zaretskii
2014-08-08 15:35     ` Breazeal, Don
2014-08-07 18:01 ` [PATCH 09/10] Extended-remote fork catchpoints Don Breazeal
2014-08-07 18:01 ` [PATCH 08/10] Extended-remote follow vfork Don Breazeal
2014-08-21  0:29 ` [PATCH 01/16 v2] Refactor native follow-fork Don Breazeal
2014-09-05 14:20   ` Pedro Alves
2014-09-05 18:56     ` Breazeal, Don
2014-09-05 20:20       ` Breazeal, Don
2014-09-09 10:57       ` Pedro Alves
2014-09-08 23:54     ` Breazeal, Don
2014-09-09 11:09       ` Pedro Alves
2014-09-12 16:50         ` Breazeal, Don
2014-09-22 15:53           ` Breazeal, Don
2014-09-26 18:13           ` Pedro Alves
2014-09-29 18:08             ` Breazeal, Don
2014-09-30 10:56               ` Pedro Alves
2014-09-30 18:43                 ` Breazeal, Don
2014-08-21  0:29 ` [Patch 00/16 v2] Linux extended-remote fork and exec events Don Breazeal
2014-09-04 20:57   ` Breazeal, Don
2014-10-31 23:29   ` Don Breazeal [this message]
2014-10-31 23:29   ` [PATCH 04/16 v3] Determine supported extended-remote features Don Breazeal
2014-11-13 12:59     ` Pedro Alves
2014-11-13 18:28       ` Breazeal, Don
2014-11-13 18:33         ` Pedro Alves
2014-11-13 19:08           ` Pedro Alves
2014-11-13 18:37         ` Breazeal, Don
2014-11-13 18:48           ` Pedro Alves
2014-12-06  0:30             ` Breazeal, Don
2015-01-12 22:36           ` Don Breazeal
2015-01-21 21:02             ` Breazeal, Don
2014-10-31 23:29   ` [PATCH 08/16 v3] Extended-remote follow vfork Don Breazeal
2014-10-31 23:29   ` [PATCH 00/16 v3] Linux extended-remote fork and exec events Don Breazeal
2014-11-12 15:54     ` Pedro Alves
2014-11-13 13:41     ` Pedro Alves
2014-11-13 13:51       ` Pedro Alves
2014-11-13 14:58         ` Pedro Alves
2014-11-13 19:14     ` Pedro Alves
2014-10-31 23:29   ` [PATCH 05/16 v3] GDBserver clone breakpoint list Don Breazeal
2014-10-31 23:29   ` [PATCH 06/16 v3] Extended-remote Linux follow fork Don Breazeal
2014-11-13 13:00     ` Pedro Alves
2014-11-13 18:53       ` Breazeal, Don
2014-11-13 18:59         ` Pedro Alves
2014-11-13 19:06           ` Breazeal, Don
2014-12-06  0:31             ` Breazeal, Don
2015-01-23 12:53               ` Pedro Alves
2015-01-23 17:18                 ` Breazeal, Don
     [not found]                 ` <1422222420-25421-1-git-send-email-donb@codesourcery.com>
2015-01-25 21:49                   ` [PATCH v4 6/7] Remote follow vfork Don Breazeal
2015-02-10 16:39                     ` Pedro Alves
2015-01-25 21:49                   ` [PATCH v4 5/7] Arch-specific remote follow fork Don Breazeal
2015-02-10 16:37                     ` Pedro Alves
2015-01-25 21:50                   ` [PATCH v4 1/7] Identify remote fork event support Don Breazeal
2015-02-10 16:34                     ` Pedro Alves
2015-01-25 21:50                   ` [PATCH v4 2/7] Clone remote breakpoints Don Breazeal
2015-01-25 21:58                   ` [PATCH v4 7/7] Remote fork catch Don Breazeal
2015-01-26  0:07                   ` [PATCH v4 3/7 v3] Extended-remote Linux follow fork Don Breazeal
2015-02-10 16:36                     ` Pedro Alves
2015-01-26  0:20                   ` [PATCH v4 4/7] Target remote " Don Breazeal
2015-01-12 22:39             ` [PATCH 06/16 v3] Extended-remote Linux " Don Breazeal
2015-01-12 22:49               ` Breazeal, Don
2014-10-31 23:30   ` [PATCH 11/16 v3] Extended-remote Linux exit events Don Breazeal
2014-11-13 19:18     ` Pedro Alves
2014-10-31 23:30   ` [PATCH 09/16 v3] Extended-remote fork catchpoints Don Breazeal
2014-10-31 23:30   ` [PATCH 13/16 v3] Extended-remote exec catchpoints Don Breazeal
2014-10-31 23:30   ` [PATCH 12/16 v3] Extended-remote follow exec Don Breazeal
2014-10-31 23:30   ` [PATCH 10/16 v3] Extended-remote fork event documentation Don Breazeal
2014-10-31 23:31   ` [PATCH 16/16 v3] Non-stop follow exec tests Don Breazeal
2014-10-31 23:31   ` [PATCH 15/16 v3] Extended-remote exec event documentation Don Breazeal
2014-10-31 23:31   ` [PATCH 14/16 v3] Suppress spurious warnings with extended-remote follow exec Don Breazeal
2014-08-21  0:30 ` [PATCH 03/16 v2] Refactor ptrace extended event status Don Breazeal
2014-09-09 11:31   ` Pedro Alves
2014-09-19 18:14     ` [pushed] " Breazeal, Don
2014-08-21  0:30 ` [PATCH 02/16 v2] Refactor follow-fork message printing Don Breazeal
2014-09-26 19:52   ` Pedro Alves
2014-09-26 20:14     ` Breazeal, Don
2014-10-03 23:51       ` Breazeal, Don
2014-10-15 16:08       ` Pedro Alves
2014-10-22 23:47         ` Breazeal, Don
2014-10-24 12:35           ` Pedro Alves
2014-10-24 18:45             ` Breazeal, Don
2014-08-21  0:30 ` [PATCH 04/16 v2] Determine supported extended-remote features Don Breazeal
2014-10-15 16:17   ` Pedro Alves
2014-10-21 23:23     ` Breazeal, Don
2014-10-22 21:48       ` Pedro Alves
2014-10-31 23:38         ` Breazeal, Don
2014-08-21  0:31 ` [PATCH 06/16 v2] Extended-remote Linux follow fork Don Breazeal
2014-09-19 20:57   ` Breazeal, Don
2014-08-21  0:31 ` [PATCH 07/16 v2] Extended-remote arch-specific " Don Breazeal
2014-09-19 21:26   ` Breazeal, Don
2014-08-21  0:31 ` [PATCH 05/16 v2] GDBserver clone breakpoint list Don Breazeal
2014-10-15 17:40   ` Pedro Alves
2014-10-31 23:44     ` Breazeal, Don
2014-08-21  0:32 ` [PATCH 08/16 v2] Extended-remote follow vfork Don Breazeal
2014-08-21  0:33 ` [PATCH 09/16 v2] Extended-remote fork catchpoints Don Breazeal
2014-08-21  0:33 ` [PATCH 10/16 v2] Extended-remote fork event documentation Don Breazeal
2014-08-21  0:33 ` [PATCH 11/16 v2] Extended-remote Linux exit events Don Breazeal
2014-08-21  0:34 ` [PATCH 13/16 v2] Extended-remote exec catchpoints Don Breazeal
2014-08-21  0:34 ` [PATCH 12/16 v2] Extended-remote follow exec Don Breazeal
2014-08-21  0:35 ` [PATCH 14/16 v2] Suppress spurious warnings with extended-remote " Don Breazeal
2014-08-21  0:36 ` [PATCH 15/16 v2] Extended-remote exec event documentation Don Breazeal
2014-08-21  0:36 ` [PATCH 16/16 v2] Non-stop follow exec tests Don Breazeal

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=1414798134-11536-5-git-send-email-donb@codesourcery.com \
    --to=donb@codesourcery.com \
    --cc=gdb-patches@sourceware.org \
    /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).