public inbox for gdb-patches@sourceware.org
 help / color / mirror / Atom feed
From: Andrew Burgess <aburgess@redhat.com>
To: gdb-patches@sourceware.org
Cc: Andrew Burgess <aburgess@redhat.com>
Subject: [PATCHv4 3/6] gdb: select suitable thread for gdbarch_adjust_breakpoint_address
Date: Tue, 28 Jun 2022 15:28:54 +0100	[thread overview]
Message-ID: <218bb38b88a6149cc6d1a327851fa00ceccf4fc1.1656426157.git.aburgess@redhat.com> (raw)
In-Reply-To: <cover.1656426157.git.aburgess@redhat.com>

The three targets that implement gdbarch_adjust_breakpoint_address are
arm, frv, and mips.  In each of these targets the adjust breakpoint
address function does some combination of reading the symbol table, or
reading memory at the location the breakpoint could be placed.

The problem is that performing these actions requires that the current
inferior and program space be the one in which the breakpoint will be
placed, and this is not currently always the case.

Consider a GDB session with multiple inferiors.  One inferior might be
a native target while another could be a remote target of a completely
different architecture.  Alternatively, if we consider ARM and
AArch64, one native inferior might be AArch64, while a second native
inferior could be ARM.

In these cases it is possible, and valid, for a user to have one
inferior selected, and place a breakpoint in the other inferior by
placing a breakpoint on a particular symbol.

If this happens, then currently, when
gdbarch_adjust_breakpoint_address is called, the wrong inferior (and
program space) will be selected, and memory reads, and symbol look
ups, will not return the expected results, this could lead to
breakpoints being placed in the wrong location.

There are currently two places where gdbarch_adjust_breakpoint_address
is called:

  1. In infrun.c, in the function handle_step_into_function.  In this
  case, I believe that the correct inferior and program space will
  already be selected as this is called as part of the stop event
  handling, so I don't think we need to worry about this case, and

  2. In breakpoint.c, in the function adjust_breakpoint_address, which
  is itself called from code_breakpoint::add_location and
  watch_command_1.

  The watch_command_1 case I don't think we need to worry about, this
  is for when a local watch expression is created, which can only be
  in the currently selected inferior, so this case should be fine.

  The code_breakpoint::add_location case is the one that needs fixing,
  this is what allows a breakpoint to be created between inferiors.

To fix the code_breakpoint::add_location case, I propose that we pass
the "correct" program_space (i.e. the program space in which the
breakpoint will be created) to the adjust_breakpoint_address function.
Then in adjust_breakpoint_address we can make use of
switch_to_program_space_and_thread to switch program_space and
inferior before calling gdbarch_adjust_breakpoint_address.

I discovered this issue while working on a later patch in this
series.  This later patch will detect when we cast the result of
gdbarch_tdep to the wrong type.

With this later patch in place I ran gdb.multi/multi-arch.exp on an
AArch64 target.  In this situation, two inferiors are created, an
AArch64 inferior, and an ARM inferior.  The test selected the AArch64
inferior and tries to create a breakpoint in the ARM inferior.

As a result of this we end up in arm_adjust_breakpoint_address, which
calls arm_pc_is_thumb.  Before this commit the AArch64 inferior would
be current.  As a result, all of the checks in arm_pc_is_thumb would
fail (they rely on reading symbols from the current program space),
and so, at the end of arm_pc_is_thumb we would call
arm_frame_is_thumb.  However, remember, at this point the current
inferior is the AArch64 inferior, so the current frame is an AArch64
frame.

In arm_frame_is_thumb we call arm_psr_thumb_bit, which calls
gdbarch_tdep and casts the result to arm_gdbarch_tdep.  This is wrong,
the tdep field is of type aarch64_gdbarch_tdep.  After this we have
undefined behaviour.

With this patch in place, we will have switched to a thread in the ARM
program space before calling arm_adjust_breakpoint_address.  As a
result, we now succeed in looking up the required symbols in
arm_pc_is_thumb, and so we never call arm_frame_is_thumb.

However, in the worst case scenario, if we did end up calling
arm_frame_is_thumb, as the current inferior should now be the ARM
inferior, the current frame should be an ARM frame, so we still should
not hit undefined behaviour.

I have added an assert to arm_frame_is_thumb.
---
 gdb/arm-tdep.c   | 12 ++++++++----
 gdb/breakpoint.c | 27 ++++++++++++++++++++-------
 2 files changed, 28 insertions(+), 11 deletions(-)

diff --git a/gdb/arm-tdep.c b/gdb/arm-tdep.c
index 86d870048cf..994af630f34 100644
--- a/gdb/arm-tdep.c
+++ b/gdb/arm-tdep.c
@@ -577,20 +577,24 @@ arm_is_thumb (struct regcache *regcache)
   return (cpsr & t_bit) != 0;
 }
 
-/* Determine if FRAME is executing in Thumb mode.  */
+/* Determine if FRAME is executing in Thumb mode.  FRAME must be an ARM
+   frame.  */
 
 int
 arm_frame_is_thumb (struct frame_info *frame)
 {
-  CORE_ADDR cpsr;
-  ULONGEST t_bit = arm_psr_thumb_bit (get_frame_arch (frame));
+  /* Check the architecture of FRAME.  */
+  struct gdbarch *gdbarch = get_frame_arch (frame);
+  gdb_assert (gdbarch_bfd_arch_info (gdbarch)->arch == bfd_arch_arm);
 
   /* Every ARM frame unwinder can unwind the T bit of the CPSR, either
      directly (from a signal frame or dummy frame) or by interpreting
      the saved LR (from a prologue or DWARF frame).  So consult it and
      trust the unwinders.  */
-  cpsr = get_frame_register_unsigned (frame, ARM_PS_REGNUM);
+  CORE_ADDR cpsr = get_frame_register_unsigned (frame, ARM_PS_REGNUM);
 
+  /* Find and extract the thumb bit.  */
+  ULONGEST t_bit = arm_psr_thumb_bit (gdbarch);
   return (cpsr & t_bit) != 0;
 }
 
diff --git a/gdb/breakpoint.c b/gdb/breakpoint.c
index a3be12557f6..d11e87713f4 100644
--- a/gdb/breakpoint.c
+++ b/gdb/breakpoint.c
@@ -124,7 +124,8 @@ static void breakpoint_adjustment_warning (CORE_ADDR, CORE_ADDR, int, int);
 
 static CORE_ADDR adjust_breakpoint_address (struct gdbarch *gdbarch,
 					    CORE_ADDR bpaddr,
-					    enum bptype bptype);
+					    enum bptype bptype,
+					    struct program_space *pspace);
 
 static int watchpoint_locations_match (struct bp_location *loc1,
 				       struct bp_location *loc2);
@@ -7103,8 +7104,11 @@ breakpoint_adjustment_warning (CORE_ADDR from_addr, CORE_ADDR to_addr,
 
 static CORE_ADDR
 adjust_breakpoint_address (struct gdbarch *gdbarch,
-			   CORE_ADDR bpaddr, enum bptype bptype)
+			   CORE_ADDR bpaddr, enum bptype bptype,
+			   struct program_space *pspace)
 {
+  gdb_assert (pspace != nullptr);
+
   if (bptype == bp_watchpoint
       || bptype == bp_hardware_watchpoint
       || bptype == bp_read_watchpoint
@@ -7129,11 +7133,18 @@ adjust_breakpoint_address (struct gdbarch *gdbarch,
     {
       CORE_ADDR adjusted_bpaddr = bpaddr;
 
+      /* Some targets have architectural constraints on the placement
+	 of breakpoint instructions.  Obtain the adjusted address.  */
       if (gdbarch_adjust_breakpoint_address_p (gdbarch))
 	{
-	  /* Some targets have architectural constraints on the placement
-	     of breakpoint instructions.  Obtain the adjusted address.  */
-	  adjusted_bpaddr = gdbarch_adjust_breakpoint_address (gdbarch, bpaddr);
+	  /* Targets that implement this adjustment function will likely
+	     inspect either the symbol table, target memory at BPADDR, or
+	     even state registers, so ensure a suitable thread (and its
+	     associated program space) are currently selected.  */
+	  scoped_restore_current_pspace_and_thread restore_pspace_thread;
+	  switch_to_program_space_and_thread (pspace);
+	  adjusted_bpaddr
+	    = gdbarch_adjust_breakpoint_address (gdbarch, bpaddr);
 	}
 
       adjusted_bpaddr = address_significant (gdbarch, adjusted_bpaddr);
@@ -8087,7 +8098,8 @@ code_breakpoint::add_location (const symtab_and_line &sal)
      not want its scan of the location chain to find a breakpoint and
      location that's only been partially initialized.  */
   adjusted_address = adjust_breakpoint_address (loc_gdbarch,
-						sal.pc, type);
+						sal.pc, type,
+						sal.pspace);
 
   /* Sort the locations by their ADDRESS.  */
   new_loc = allocate_location ();
@@ -10077,7 +10089,8 @@ watch_command_1 (const char *arg, int accessflag, int from_tty,
 	  scope_breakpoint->loc->address
 	    = adjust_breakpoint_address (scope_breakpoint->loc->gdbarch,
 					 scope_breakpoint->loc->requested_address,
-					 scope_breakpoint->type);
+					 scope_breakpoint->type,
+					 current_program_space);
 	}
     }
 
-- 
2.25.4


  parent reply	other threads:[~2022-06-28 14:29 UTC|newest]

Thread overview: 83+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-05-31 14:30 [PATCH 0/5] Handle trying to use a native target with the wrong binary Andrew Burgess
2022-05-31 14:30 ` [PATCH 1/5] gdb/arm: move fetch of arm_gdbarch_tdep to a more inner scope Andrew Burgess
2022-06-01  7:58   ` Luis Machado
2022-05-31 14:30 ` [PATCH 2/5] gdb/mips: rewrite show_mask_address Andrew Burgess
2022-05-31 14:30 ` [PATCH 3/5] gdb: move the type cast into gdbarch_tdep Andrew Burgess
2022-06-01  8:01   ` Luis Machado
2022-05-31 14:30 ` [PATCH 4/5] gdb: ensure the cast in gdbarch_tdep is valid Andrew Burgess
2022-05-31 16:04   ` John Baldwin
2022-05-31 17:22     ` Andrew Burgess
2022-05-31 14:30 ` [PATCH 5/5] gdb: native target invalid architecture detection Andrew Burgess
2022-05-31 16:08   ` John Baldwin
2022-05-31 16:51     ` Andrew Burgess
2022-06-01  8:25       ` Luis Machado
2022-06-01 21:06         ` John Baldwin
2022-06-01 21:21           ` Christophe Lyon
2022-06-02 14:56             ` John Baldwin
2022-06-06 14:38         ` Andrew Burgess
2022-06-06 17:48           ` Andrew Burgess
2022-06-07 11:03             ` Luis Machado
2022-06-07 18:42               ` Pedro Alves
2022-06-07 20:15                 ` Pedro Alves
2022-06-08  8:18                   ` Luis Machado
2022-06-08 10:17                     ` Pedro Alves
2022-06-08  7:54                 ` Luis Machado
2022-06-08 10:12                   ` Pedro Alves
2022-06-08 11:20                     ` [PATCH v2] aarch64: Add fallback if ARM_CC_FOR_TARGET not set (was: Re: [PATCH 5/5] gdb: native target invalid architecture detection) Pedro Alves
2022-06-08 12:50                       ` Luis Machado
2022-06-08 13:23                         ` Pedro Alves
2022-06-08 13:38                       ` Andrew Burgess
2022-06-08 19:01                       ` John Baldwin
2022-06-08 21:48                         ` Pedro Alves
2022-06-09 16:31                           ` John Baldwin
2022-06-10 13:08 ` [PATCHv2 0/6] Handle trying to use a native target with the wrong binary Andrew Burgess
2022-06-10 13:08   ` [PATCHv2 1/6] gdb/arm: move fetch of arm_gdbarch_tdep to a more inner scope Andrew Burgess
2022-06-10 13:08   ` [PATCHv2 2/6] gdb/mips: rewrite show_mask_address Andrew Burgess
2022-06-10 13:08   ` [PATCHv2 3/6] gdb/arm: avoid undefined behaviour in arm_frame_is_thumb Andrew Burgess
2022-06-10 15:21     ` Luis Machado
2022-06-10 15:49       ` Andrew Burgess
2022-06-10 16:29         ` Luis Machado
2022-06-10 13:08   ` [PATCHv2 4/6] gdb: move the type cast into gdbarch_tdep Andrew Burgess
2022-06-10 16:35     ` Luis Machado
2022-06-10 13:08   ` [PATCHv2 5/6] gdb: ensure the cast in gdbarch_tdep is valid Andrew Burgess
2022-06-10 13:08   ` [PATCHv2 6/6] gdb: native target invalid architecture detection Andrew Burgess
2022-06-10 16:20     ` John Baldwin
2022-06-10 16:31     ` Luis Machado
2022-06-13 16:15   ` [PATCHv3 0/6] Handle trying to use a native target with the wrong binary Andrew Burgess
2022-06-13 16:15     ` [PATCHv3 1/6] gdb/arm: move fetch of arm_gdbarch_tdep to a more inner scope Andrew Burgess
2022-06-13 16:15     ` [PATCHv3 2/6] gdb/mips: rewrite show_mask_address Andrew Burgess
2022-06-13 16:15     ` [PATCHv3 3/6] gdb: select suitable thread for gdbarch_adjust_breakpoint_address Andrew Burgess
2022-06-14  9:45       ` Luis Machado
2022-06-14 14:05         ` Andrew Burgess
2022-06-24 16:58       ` Pedro Alves
2022-06-13 16:15     ` [PATCHv3 4/6] gdb: move the type cast into gdbarch_tdep Andrew Burgess
2022-06-13 16:15     ` [PATCHv3 5/6] gdb: ensure the cast in gdbarch_tdep is valid Andrew Burgess
2022-06-24 18:15       ` Pedro Alves
2022-06-13 16:15     ` [PATCHv3 6/6] gdb: native target invalid architecture detection Andrew Burgess
2022-06-24 19:23       ` Pedro Alves
2022-06-27 16:27         ` Andrew Burgess
2022-06-27 21:38           ` Pedro Alves
2022-06-28 10:37             ` Andrew Burgess
2022-06-28 12:42               ` [PATCH v2] gdb+gdbserver/Linux: avoid reading registers while going through shell (was: Re: [PATCHv3 6/6] gdb: native target invalid architecture detection) Pedro Alves
2022-06-28 14:21                 ` Andrew Burgess
2022-06-29 15:17                 ` Simon Marchi
2022-06-29 16:22                   ` [PATCH] Fix GDBserver regression due to change to avoid reading shell registers Pedro Alves
2022-06-29 16:38                     ` Simon Marchi
2022-06-30  9:33             ` [PATCHv3 6/6] gdb: native target invalid architecture detection Andrew Burgess
2022-06-30 11:44               ` Pedro Alves
2022-07-11 10:47                 ` Andrew Burgess
2022-06-24 10:15     ` [PATCHv3 0/6] Handle trying to use a native target with the wrong binary Andrew Burgess
2022-06-28 14:28     ` [PATCHv4 0/6] Detect invalid casts of gdbarch_tdep structures Andrew Burgess
2022-06-28 14:28       ` [PATCHv4 1/6] gdb/arm: move fetch of arm_gdbarch_tdep to a more inner scope Andrew Burgess
2022-06-28 14:28       ` [PATCHv4 2/6] gdb/mips: rewrite show_mask_address Andrew Burgess
2022-06-28 14:28       ` Andrew Burgess [this message]
2022-06-28 14:28       ` [PATCHv4 4/6] gdb: move the type cast into gdbarch_tdep Andrew Burgess
2022-06-28 14:28       ` [PATCHv4 5/6] gdbsupport: add checked_static_cast Andrew Burgess
2022-06-28 14:28       ` [PATCHv4 6/6] gdb: ensure the cast in gdbarch_tdep is valid Andrew Burgess
2022-07-11 10:46       ` [PATCHv4 0/6] Detect invalid casts of gdbarch_tdep structures Andrew Burgess
2022-07-21 18:21         ` Andrew Burgess
2022-07-22  0:50           ` Luis Machado
2022-07-23  0:02             ` [PATCH] Rename gdbarch_tdep template function to gdbarch_tdep_cast for g++ 4.8 Mark Wielaard
2022-07-25 11:19               ` Andrew Burgess
2022-07-25 11:27                 ` Mark Wielaard
2022-07-26 11:05                   ` Andrew Burgess

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=218bb38b88a6149cc6d1a327851fa00ceccf4fc1.1656426157.git.aburgess@redhat.com \
    --to=aburgess@redhat.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).