From: Yao Qi <yao@codesourcery.com>
To: Ulrich Weigand <uweigand@de.ibm.com>
Cc: gdb-patches@sourceware.org
Subject: Re: [PATCH] Fix that different function breakpoints are set@same pc address (PR gdb/12703)
Date: Wed, 12 Oct 2011 00:47:00 -0000 [thread overview]
Message-ID: <4E94E367.3040109@codesourcery.com> (raw)
In-Reply-To: <201110101446.p9AEklVX022612@d06av02.portsmouth.uk.ibm.com>
On 10/10/2011 10:46 PM, Ulrich Weigand wrote:
>> > - else if ((insn & 0xe000) == 0xe000)
>> > + else if ((insn & 0xe000) == 0xe000 && (insn & 0x1800) != 0)
>> > {
> Instead of open-coding the check, I think it would be preferable to
> use the thumb_insn_size routine instead.
>
> Note that there are a number of other places that either already open-
> code the correct check, or -worse- use the same incorrect check as the
> code above:
>
> - thumb_in_function_epilogue_p
> - thumb_get_next_pc_raw
> - arm_breakpoint_from_pc
>
> Would you mind converting them all to thumb_insn_size?
Yeah, we should replace these checks around many places with
thumb_insn_size, I agree. Here is the patch for this purpose.
Regression tested on arm-linux-gnueabi with both -marm and -mthumb.
OK for mainline?
--
Yao (é½å°§)
PR gdb/12703
* arm-tdep.c (thumb_analyze_prologue): Call thumb_insn_size to check
whether insn is a 32-bit Thumb-2 instruction.
(thumb_in_function_epilogue_p): Likewise.
(thumb_get_next_pc_raw): Likewise.
(arm_breakpoint_from_pc): Likewise.
diff --git a/gdb/arm-tdep.c b/gdb/arm-tdep.c
index 278e6e9..0db8b5f 100644
--- a/gdb/arm-tdep.c
+++ b/gdb/arm-tdep.c
@@ -231,6 +231,8 @@ static void arm_neon_quad_write (struct gdbarch *gdbarch,
struct regcache *regcache,
int regnum, const gdb_byte *buf);
+static int thumb_insn_size (unsigned short inst1);
+
struct arm_prologue_cache
{
/* The stack pointer at the time this frame was created; i.e. the
@@ -836,7 +838,7 @@ thumb_analyze_prologue (struct gdbarch *gdbarch,
constant = read_memory_unsigned_integer (loc, 4, byte_order);
regs[bits (insn, 8, 10)] = pv_constant (constant);
}
- else if ((insn & 0xe000) == 0xe000)
+ else if (thumb_insn_size (insn) == 4) /* 32-bit Thumb-2 instructions. */
{
unsigned short inst2;
@@ -3093,7 +3095,7 @@ thumb_in_function_epilogue_p (struct gdbarch *gdbarch, CORE_ADDR pc)
if (insn & 0x0100) /* <registers> include PC. */
found_return = 1;
}
- else if ((insn & 0xe000) == 0xe000) /* 32-bit Thumb-2 instruction */
+ else if (thumb_insn_size (insn) == 4) /* 32-bit Thumb-2 instruction */
{
if (target_read_memory (scan_pc, buf, 2))
break;
@@ -4335,14 +4337,9 @@ thumb_get_next_pc_raw (struct frame_info *frame, CORE_ADDR pc)
int cond = itstate >> 4;
if (! condition_true (cond, status))
- {
- /* Advance to the next instruction. All the 32-bit
- instructions share a common prefix. */
- if ((inst1 & 0xe000) == 0xe000 && (inst1 & 0x1800) != 0)
- return MAKE_THUMB_ADDR (pc + 4);
- else
- return MAKE_THUMB_ADDR (pc + 2);
- }
+ /* Advance to the next instruction. All the 32-bit
+ instructions share a common prefix. */
+ return MAKE_THUMB_ADDR (pc + thumb_insn_size (inst1));
/* Otherwise, handle the instruction normally. */
}
@@ -4376,7 +4373,7 @@ thumb_get_next_pc_raw (struct frame_info *frame, CORE_ADDR pc)
{
nextpc = pc_val + (sbits (inst1, 0, 10) << 1);
}
- else if ((inst1 & 0xe000) == 0xe000) /* 32-bit instruction */
+ else if (thumb_insn_size (inst1) == 4) /* 32-bit instruction */
{
unsigned short inst2;
inst2 = read_memory_unsigned_integer (pc + 2, 2, byte_order_for_code);
@@ -8473,7 +8470,7 @@ arm_breakpoint_from_pc (struct gdbarch *gdbarch, CORE_ADDR *pcptr, int *lenptr)
{
unsigned short inst1;
inst1 = extract_unsigned_integer (buf, 2, byte_order_for_code);
- if ((inst1 & 0xe000) == 0xe000 && (inst1 & 0x1800) != 0)
+ if (thumb_insn_size (inst1) == 4)
{
*lenptr = tdep->thumb2_breakpoint_size;
return tdep->thumb2_breakpoint;
next prev parent reply other threads:[~2011-10-12 0:47 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-06-24 2:31 [PATCH] Fix that different function breakpoints are set at same " Terry Guo
2011-06-24 3:55 ` Yao Qi
2011-06-24 8:59 ` Pedro Alves
2011-06-24 10:39 ` Yao Qi
2011-09-27 12:53 ` [ping]: " Yao Qi
2011-10-09 15:06 ` [ping 2]: " Yao Qi
2011-10-10 14:47 ` [PATCH] Fix that different function breakpoints are set@same " Ulrich Weigand
2011-10-12 0:47 ` Yao Qi [this message]
2011-10-12 11:58 ` Ulrich Weigand
2011-10-13 8:19 ` [committed] : " Yao Qi
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=4E94E367.3040109@codesourcery.com \
--to=yao@codesourcery.com \
--cc=gdb-patches@sourceware.org \
--cc=uweigand@de.ibm.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