Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
From: Yao Qi <qiyaoltc@gmail.com>
To: Yao Qi <qiyaoltc@gmail.com>
Cc: Antoine Tremblay <antoine.tremblay@ericsson.com>,
	 <gdb-patches@sourceware.org>
Subject: Re: [PATCH v8 3/7] Refactor arm_software_single_step to use regcache
Date: Mon, 18 Jan 2016 13:07:00 -0000	[thread overview]
Message-ID: <86h9ibt4sg.fsf@gmail.com> (raw)
In-Reply-To: <86oacjtbue.fsf@gmail.com> (Yao Qi's message of "Mon, 18 Jan 2016	10:34:49 +0000")

Yao Qi <qiyaoltc@gmail.com> writes:

> As a record, this patch causes regressions at least in
> gdb.base/sigstep.exp,
>
> (gdb) PASS: gdb.base/sigstep.exp: continue to handler, si+advance in
> handler, step from handler: advance in handler
> step^M
> 39      } /* handler */^M
> 1: x/i $pc^M
> => 0x8740 <handler+80>: sub     sp, r11, #0^M
> (gdb) step^M
> ^M
> Program received signal SIGSEGV, Segmentation fault.^M
> setitimer () at ../sysdeps/unix/syscall-template.S:81^M
> 81      ../sysdeps/unix/syscall-template.S: No such file or directory.^M
> 1: x/i $pc^M
> => 0xb6eff9c0 <setitimer>:      push    {r7}^M
> (gdb) FAIL: gdb.base/sigstep.exp: continue to handler, si+advance in
> handler, step from handler: leave handler
>
> Could you take a look at it?

Here is my patch fixing the regression in sigstep.exp.  The regression
test is running now.

-- 
Yao (齐尧)
From 1c46ba5ca260289788ed2db3c22f110b44481856 Mon Sep 17 00:00:00 2001
From: Yao Qi <yao.qi@linaro.org>
Date: Mon, 18 Jan 2016 12:55:26 +0000
Subject: [PATCH] Detect the arm/thumb mode of code SIGRETURN or RT_SIGRETURN
 returns to

This patch fixes the following regression introduced by commit d0e59a68

step^M
39      } /* handler */^M
1: x/i $pc^M
=> 0x8740 <handler+80>: sub     sp, r11, #0^M
(gdb) step^M
^M
Program received signal SIGSEGV, Segmentation fault.^M
setitimer () at ../sysdeps/unix/syscall-template.S:81^M
81      ../sysdeps/unix/syscall-template.S: No such file or directory.^M
1: x/i $pc^M
=> 0xb6eff9c0 <setitimer>:      push    {r7}^M
(gdb) FAIL: gdb.base/sigstep.exp: continue to handler, si+advance in handler, step from handler: leave handler

in my test setting, program is compiled in arm mode, but the glibc
is built in thumb mode, so when we do 'step' to step over syscall
instruction svc for SIGRETURN, GDB should set breakpoint for arm mode
in the program, even though the current program in glibc is in thumb
mode.  Current GDB doesn't consider the case that the mode of program
SIGRETURN goes to can be different from current program mode.

In fact, GDB has taken care of this arm/thumb mode changes already,
see

/* Copy the value of next pc of sigreturn and rt_sigrturn into PC,
   return 1.  In addition, set IS_THUMB depending on whether we
   will return to ARM or Thumb code.  Return 0 if it is not a
   rt_sigreturn/sigreturn syscall.  */
static int
arm_linux_sigreturn_return_addr (struct frame_info *frame,
				 unsigned long svc_number,
				 CORE_ADDR *pc, int *is_thumb)

but in the commit d0e59a68

> -  arm_linux_sigreturn_return_addr (frame, svc_number, &return_addr, &is_thumb);
> +  if (svc_number == ARM_SIGRETURN || svc_number == ARM_RT_SIGRETURN)
> +    next_pc = arm_linux_sigreturn_next_pc (regcache, svc_number);

the IS_THUMB setting is lost, so it is a regression.

gdb:

2016-01-18  Yao Qi  <yao.qi@linaro.org>

	* arm-linux-tdep.c (arm_linux_sigreturn_next_pc): Add parameter
	is_thumb and set it according to CPSR saved on the stack.
	(arm_linux_get_next_pcs_syscall_next_pc): Pass is_thumb to
	arm_linux_sigreturn_next_pc.

gdb/gdbserver:

2016-01-18  Yao Qi  <yao.qi@linaro.org>

	* linux-arm-low.c (arm_sigreturn_next_pc): Add parameter
	is_thumb and set it according to CPSR saved on the stack.
	(get_next_pcs_syscall_next_pc): Pass is_thumb to
	arm_sigreturn_next_pc.

diff --git a/gdb/arm-linux-tdep.c b/gdb/arm-linux-tdep.c
index 2306bda..f6a831a 100644
--- a/gdb/arm-linux-tdep.c
+++ b/gdb/arm-linux-tdep.c
@@ -782,10 +782,12 @@ arm_linux_sigreturn_return_addr (struct frame_info *frame,
 }
 
 /* Find the value of the next PC after a sigreturn or rt_sigreturn syscall
-   based on current processor state.  */
+   based on current processor state.  In addition, set IS_THUMB depending
+   on whether we will return to ARM or Thumb code.  */
+
 static CORE_ADDR
 arm_linux_sigreturn_next_pc (struct regcache *regcache,
-			     unsigned long svc_number)
+			     unsigned long svc_number, int *is_thumb)
 {
   ULONGEST sp;
   unsigned long sp_data;
@@ -794,6 +796,7 @@ arm_linux_sigreturn_next_pc (struct regcache *regcache,
   enum bfd_endian byte_order = gdbarch_byte_order (gdbarch);
   int pc_offset = 0;
   int is_sigreturn = 0;
+  CORE_ADDR cpsr;
 
   gdb_assert (svc_number == ARM_SIGRETURN
 	      || svc_number == ARM_RT_SIGRETURN);
@@ -807,6 +810,10 @@ arm_linux_sigreturn_next_pc (struct regcache *regcache,
 
   next_pc = read_memory_unsigned_integer (sp + pc_offset, 4, byte_order);
 
+  /* Set IS_THUMB according the CPSR saved on the stack.  */
+  cpsr = read_memory_unsigned_integer (sp + pc_offset + 4, 4, byte_order);
+  *is_thumb = ((cpsr & arm_psr_thumb_bit (gdbarch)) != 0);
+
   return next_pc;
 }
 
@@ -899,7 +906,12 @@ arm_linux_get_next_pcs_syscall_next_pc (struct arm_get_next_pcs *self,
     }
 
   if (svc_number == ARM_SIGRETURN || svc_number == ARM_RT_SIGRETURN)
-    next_pc = arm_linux_sigreturn_next_pc (self->regcache, svc_number);
+    {
+      /* SIGRETURN or RT_SIGRETURN may affect the arm thumb mode, so
+	 update IS_THUMB.   */
+      next_pc = arm_linux_sigreturn_next_pc (self->regcache, svc_number,
+					     &is_thumb);
+    }
 
   /* Addresses for calling Thumb functions have the bit 0 set.  */
   if (is_thumb)
diff --git a/gdb/gdbserver/linux-arm-low.c b/gdb/gdbserver/linux-arm-low.c
index 927a6fa..01a3bc0 100644
--- a/gdb/gdbserver/linux-arm-low.c
+++ b/gdb/gdbserver/linux-arm-low.c
@@ -769,16 +769,20 @@ arm_prepare_to_resume (struct lwp_info *lwp)
       }
 }
 
-/* Find the next pc for a sigreturn or rt_sigreturn syscall.
+/* Find the next pc for a sigreturn or rt_sigreturn syscall.  In
+   addition, set IS_THUMB depending on whether we will return to ARM
+   or Thumb code.
    See arm-linux.h for stack layout details.  */
 static CORE_ADDR
-arm_sigreturn_next_pc (struct regcache *regcache, int svc_number)
+arm_sigreturn_next_pc (struct regcache *regcache, int svc_number,
+		       int *is_thumb)
 {
   unsigned long sp;
   unsigned long sp_data;
   /* Offset of PC register.  */
   int pc_offset = 0;
   CORE_ADDR next_pc = 0;
+  CORE_ADDR cpsr;
 
   gdb_assert (svc_number == __NR_sigreturn || svc_number == __NR_rt_sigreturn);
 
@@ -790,6 +794,10 @@ arm_sigreturn_next_pc (struct regcache *regcache, int svc_number)
 
   (*the_target->read_memory) (sp + pc_offset, (unsigned char *) &next_pc, 4);
 
+  /* Set IS_THUMB according the CPSR saved on the stack.  */
+  (*the_target->read_memory) (sp + pc_offset + 4, (unsigned char *) &cpsr, 4);
+  *is_thumb = ((cpsr & CPSR_T) != 0);
+
   return next_pc;
 }
 
@@ -831,7 +839,9 @@ get_next_pcs_syscall_next_pc (struct arm_get_next_pcs *self, CORE_ADDR pc)
   /* This is a sigreturn or sigreturn_rt syscall.  */
   if (svc_number == __NR_sigreturn || svc_number == __NR_rt_sigreturn)
     {
-      next_pc = arm_sigreturn_next_pc (regcache, svc_number);
+      /* SIGRETURN or RT_SIGRETURN may affect the arm thumb mode, so
+	 update IS_THUMB.   */
+      next_pc = arm_sigreturn_next_pc (regcache, svc_number, &is_thumb);
     }
 
   /* Addresses for calling Thumb functions have the bit 0 set.  */


  reply	other threads:[~2016-01-18 13:07 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-12-17 14:14 [PATCH v8 0/7] Support software single step and conditional breakpoints on ARM in GDBServer Antoine Tremblay
2015-12-17 14:15 ` [PATCH v8 5/7] Support software single step " Antoine Tremblay
2015-12-21 13:58   ` Yao Qi
2016-01-04 12:59     ` Antoine Tremblay
2016-01-06 14:23   ` Yao Qi
2016-01-06 14:36     ` Antoine Tremblay
2016-01-06 14:56       ` Yao Qi
2016-01-06 14:58         ` Antoine Tremblay
2016-01-06 14:42   ` Yao Qi
2016-01-06 14:50     ` Antoine Tremblay
2016-01-06 15:04       ` Yao Qi
2016-01-13 16:13   ` Yao Qi
2016-01-13 19:10     ` Antoine Tremblay
2015-12-17 14:15 ` [PATCH v8 7/7] Enable conditional breakpoints for targets that support software single step " Antoine Tremblay
2015-12-17 14:15 ` [PATCH v8 4/7] Share regcache function regcache_raw_read_unsigned Antoine Tremblay
2015-12-17 14:15 ` [PATCH v8 2/7] Share some ARM target dependent code from GDB with GDBServer Antoine Tremblay
2015-12-17 14:15 ` [PATCH v8 6/7] Enable software single stepping for while-stepping actions in GDBServer Antoine Tremblay
2015-12-17 14:15 ` [PATCH v8 3/7] Refactor arm_software_single_step to use regcache Antoine Tremblay
2016-01-18 10:35   ` Yao Qi
2016-01-18 13:07     ` Yao Qi [this message]
2016-01-18 14:02       ` Antoine Tremblay
2016-01-21  7:52       ` Yao Qi
2015-12-17 14:15 ` [PATCH v8 1/7] Replace breakpoint_reinsert_addr by get_next_pcs operation in GDBServer Antoine Tremblay
2015-12-18 12:56 ` [PATCH v8 0/7] Support software single step and conditional breakpoints on ARM " Antoine Tremblay
2015-12-18 16:45 ` Antoine Tremblay

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=86h9ibt4sg.fsf@gmail.com \
    --to=qiyaoltc@gmail.com \
    --cc=antoine.tremblay@ericsson.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