Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
From: Uros Bizjak <ubizjak@gmail.com>
To: gdb-patches@sourceware.org
Cc: Alan Modra <amodra@gmail.com>, Richard Henderson <rth@redhat.com>
Subject: [PATCH v2, alpha]: Handle single-stepping of atomic sequences
Date: Fri, 18 Nov 2011 12:19:00 -0000	[thread overview]
Message-ID: <CAFULd4ZZXmnvzEXMZ-GRpDeFkSwXTmm9p+Xoz5N9rsdb3xng8w@mail.gmail.com> (raw)
In-Reply-To: <CAFULd4Ysi5pmL6AHNsvy2qAEk4_WrEo4vLKQvgHgvXXqd1-TRg@mail.gmail.com>

[-- Attachment #1: Type: text/plain, Size: 1145 bytes --]

On Wed, Nov 16, 2011 at 11:47 PM, Uros Bizjak <ubizjak@gmail.com> wrote:

> The gcc simulate-thread.exp testsuite exposed the problem in handling
> of alpha atomic sequences. Single-stepping broke link in alpha's LL/SC
> atomic sequences.
>
> Attached patch copies handling of atomic sequences from rs6000-tdep.c
> and adapts the functionality for alpha.
>
> 2011-11-16  Uros Bizjak  <ubizjak@gmail.com>
>
>        * alpha-tdep.c (beq_opcode): New.
>        (ldl_l_opcode, ldq_l_opcode, stl_c_opcode, stq_c_opcode): Ditto.
>        (alpha_deal_with_atomic_sequence): New function.
>        (alpha_gdbarch_init): Handle single stepping of atomic sequences
>        with alpha_deal_with_atomic_sequence.
>
> Patched gdb survives simulate-thread.exp testsuite on native
> alphaev68-pc-linux-gnu.
>
> I don't have write access, so If OK, please someone commit this patch
> to gdb CVS.

[v2]: Since this patch is almost a copy of rs6000-tdep.c atomic
sequence handling, also take into account proposed changes [1] to
original rs6000 implementation.

[1] http://sourceware.org/ml/gdb-patches/2011-11/msg00468.html

Uros,

[-- Attachment #2: gdb-7.3.50.20111114.diff --]
[-- Type: text/x-patch, Size: 4141 bytes --]

--- alpha-tdep.c	2011-11-18 12:20:08.000000000 +0100
+++ alpha-tdep.c.ub	2011-11-18 12:18:37.000000000 +0100
@@ -65,6 +65,7 @@ static const int stq_opcode = 0x2d;
 /* Branch instruction format */
 #define BR_RA(insn) MEM_RA(insn)
 
+static const int beq_opcode = 0x39;
 static const int bne_opcode = 0x3d;
 
 /* Operate instruction format */
@@ -762,6 +763,94 @@ alpha_skip_prologue (struct gdbarch *gdb
 }
 
 \f
+static const int ldl_l_opcode = 0x2a;
+static const int ldq_l_opcode = 0x2b;
+static const int stl_c_opcode = 0x2e;
+static const int stq_c_opcode = 0x2f;
+
+/* Checks for an atomic sequence of instructions beginning with a LDL_L/LDQ_L
+   instruction and ending with a STL_C/STQ_C instruction.  If such a sequence
+   is found, attempt to step through it.  A breakpoint is placed at the end of 
+   the sequence.  */
+
+int 
+alpha_deal_with_atomic_sequence (struct frame_info *frame)
+{
+  struct gdbarch *gdbarch = get_frame_arch (frame);
+  struct address_space *aspace = get_frame_address_space (frame);
+  CORE_ADDR pc = get_frame_pc (frame);
+  CORE_ADDR breaks[2] = {-1, -1};
+  CORE_ADDR loc = pc;
+  CORE_ADDR closing_insn; /* Instruction that closes the atomic sequence.  */
+  unsigned int insn = alpha_read_insn (gdbarch, loc);
+  int insn_count;
+  int index;
+  int last_breakpoint = 0; /* Defaults to 0 (no breakpoints placed).  */  
+  const int atomic_sequence_length = 16; /* Instruction sequence length.  */
+  int bc_insn_count = 0; /* Conditional branch instruction count.  */
+
+  /* Assume all atomic sequences start with a LDL_L/LDQ_L instruction.  */
+  if (INSN_OPCODE (insn) != ldl_l_opcode
+      && INSN_OPCODE (insn) != ldq_l_opcode)
+    return 0;
+
+  /* Assume that no atomic sequence is longer than "atomic_sequence_length" 
+     instructions.  */
+  for (insn_count = 0; insn_count < atomic_sequence_length; ++insn_count)
+    {
+      loc += ALPHA_INSN_SIZE;
+      insn = alpha_read_insn (gdbarch, loc);
+
+      /* Assume that there is at most one conditional branch in the atomic
+         sequence.  If a conditional branch is found, put a breakpoint in 
+         its destination address.  */
+      if (INSN_OPCODE (insn) >= beq_opcode)
+        {
+          int immediate = (insn & 0x001fffff) << 2;
+
+	  immediate = (immediate ^ 0x400000) - 0x400000;
+
+          if (bc_insn_count >= 1)
+            return 0; /* More than one conditional branch found, fallback 
+                         to the standard single-step code.  */
+ 
+	  breaks[1] = loc + ALPHA_INSN_SIZE + immediate;
+
+	  bc_insn_count++;
+	  last_breakpoint++;
+        }
+
+      if (INSN_OPCODE (insn) == stl_c_opcode
+	  || INSN_OPCODE (insn) == stq_c_opcode)
+	break;
+    }
+
+  /* Assume that the atomic sequence ends with a STL_C/STQ_C instruction.  */
+  if (INSN_OPCODE (insn) != stl_c_opcode
+      && INSN_OPCODE (insn) != stq_c_opcode)
+    return 0;
+
+  closing_insn = loc;
+  loc += ALPHA_INSN_SIZE;
+
+  /* Insert a breakpoint right after the end of the atomic sequence.  */
+  breaks[0] = loc;
+
+  /* Check for duplicated breakpoints.  Check also for a breakpoint
+     placed (branch instruction's destination) anywhere in sequence.  */ 
+  if (last_breakpoint
+      && (breaks[1] == breaks[0]
+	  || (breaks[1] >= pc && breaks[1] <= closing_insn)))
+    last_breakpoint = 0;
+
+  /* Effectively inserts the breakpoints.  */
+  for (index = 0; index <= last_breakpoint; index++)
+    insert_single_step_breakpoint (gdbarch, aspace, breaks[index]);
+
+  return 1;
+}
+
+\f
 /* Figure out where the longjmp will land.
    We expect the first arg to be a pointer to the jmp_buf structure from
    which we extract the PC (JB_PC) that we will land at.  The PC is copied
@@ -1749,6 +1838,9 @@ alpha_gdbarch_init (struct gdbarch_info
   set_gdbarch_decr_pc_after_break (gdbarch, ALPHA_INSN_SIZE);
   set_gdbarch_cannot_step_breakpoint (gdbarch, 1);
 
+  /* Handles single stepping of atomic sequences.  */
+  set_gdbarch_software_single_step (gdbarch, alpha_deal_with_atomic_sequence);
+
   /* Hook in ABI-specific overrides, if they have been registered.  */
   gdbarch_init_osabi (info, gdbarch);
 

  reply	other threads:[~2011-11-18 12:19 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-11-16 22:47 [PATCH, " Uros Bizjak
2011-11-18 12:19 ` Uros Bizjak [this message]
2011-11-24  9:21   ` [PATCH v2, " Uros Bizjak
2011-11-24 12:37     ` Ulrich Weigand
2011-11-24 13:29       ` [PATCH v3, " Uros Bizjak
2011-11-25 17:09         ` Ulrich Weigand

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=CAFULd4ZZXmnvzEXMZ-GRpDeFkSwXTmm9p+Xoz5N9rsdb3xng8w@mail.gmail.com \
    --to=ubizjak@gmail.com \
    --cc=amodra@gmail.com \
    --cc=gdb-patches@sourceware.org \
    --cc=rth@redhat.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