Handle MIPS Linux SIGTRAP siginfo.si_code values

This unbreaks pending/delayed breakpoints handling, as well as
hardware watchpoints, on MIPS.

Ref: https://sourceware.org/ml/gdb-patches/2016-02/msg00681.html

The MIPS kernel reports SI_KERNEL for all kernel generated traps,
instead of TRAP_BRKPT / TRAP_HWBKPT, but GDB isn't aware of this.

Basically, this commit:

- Folds watchpoints logic into check_stopped_by_breakpoint, and
  renames it to save_stop_reason.

- Adds GDB_ARCH_IS_TRAP_HWBKPT.

- Makes MIPS set both GDB_ARCH_IS_TRAP_BRPT and
  GDB_ARCH_IS_TRAP_HWBKPT to SI_KERNEL.  In save_stop_reason, we
  handle the case of the same si_code returning true for both
  TRAP_BRPT and TRAP_HWBKPT by looking at what the debug registers
  say.

Tested on x86-64 Fedora 20, native and gdbserver.

gdb/ChangeLog:
2016-02-24  Pedro Alves  <palves@redhat.com>

	* linux-nat.c (save_sigtrap) Delete.
	(stop_wait_callback): Call save_stop_reason instead of
	save_sigtrap.
	(check_stopped_by_breakpoint): Rename to ...
	(save_stop_reason): ... this.  Bits of save_sigtrap folded here.
	Use GDB_ARCH_IS_TRAP_HWBKPT and handle ambiguous
	GDB_ARCH_IS_TRAP_BRKPT / GDB_ARCH_IS_TRAP_HWBKPT.  Factor out
	common code between the USE_SIGTRAP_SIGINFO and
	!USE_SIGTRAP_SIGINFO blocks.
	(linux_nat_filter_event): Call save_stop_reason instead of
	save_sigtrap.
	* nat/linux-ptrace.h: Check for both SI_KERNEL and TRAP_BRKPT
	si_code for MIPS.
	* nat/linux-ptrace.h: Fix "TRAP_HWBPT" typo in x86 table.  Add
	comments on MIPS behavior.
	(GDB_ARCH_IS_TRAP_HWBKPT): Define for all archs.

gdb/gdbserver/ChangeLog:
2016-02-24  Pedro Alves  <palves@redhat.com>

	* linux-low.c (check_stopped_by_breakpoint): Rename to ...
	(save_stop_reason): ... this.  Use GDB_ARCH_IS_TRAP_HWBKPT and
	handle ambiguous GDB_ARCH_IS_TRAP_BRKPT / GDB_ARCH_IS_TRAP_HWBKPT.
	Factor out common code between the USE_SIGTRAP_SIGINFO and
	!USE_SIGTRAP_SIGINFO blocks.
	(linux_low_filter_event): Call save_stop_reason instead of
	check_stopped_by_breakpoint and check_stopped_by_watchpoint.
	Update comments.
	(linux_wait_1): Update comments.

1 files changed, 28 insertions, 10 deletions

diff --git a/gdb/nat/linux-ptrace.h b/gdb/nat/linux-ptrace.h
index ba58717..b9123c9 100644
--- a/gdb/nat/linux-ptrace.h
+++ b/gdb/nat/linux-ptrace.h
@@ -119,14 +119,14 @@ struct buffer;
 /* The x86 kernel gets some of the si_code values backwards, like
    this:
 
-   | what                                     | si_code    |
-   |------------------------------------------+------------|
-   | software breakpoints (int3)              | SI_KERNEL  |
-   | single-steps                             | TRAP_TRACE |
-   | single-stepping a syscall                | TRAP_BRKPT |
-   | user sent SIGTRAP                        | 0          |
-   | exec SIGTRAP (when no PTRACE_EVENT_EXEC) | 0          |
-   | hardware breakpoints/watchpoints         | TRAP_HWBPT |
+   | what                                     | si_code     |
+   |------------------------------------------+-------------|
+   | software breakpoints (int3)              | SI_KERNEL   |
+   | single-steps                             | TRAP_TRACE  |
+   | single-stepping a syscall                | TRAP_BRKPT  |
+   | user sent SIGTRAP                        | 0           |
+   | exec SIGTRAP (when no PTRACE_EVENT_EXEC) | 0           |
+   | hardware breakpoints/watchpoints         | TRAP_HWBKPT |
 
    That is, it reports SI_KERNEL for software breakpoints (and only
    for those), and TRAP_BRKPT for single-stepping a syscall...  If the
@@ -140,14 +140,32 @@ struct buffer;
    in SPU code on a Cell/B.E.  However, SI_KERNEL is never seen
    on a SIGTRAP for any other reason.
 
-   The generic Linux target code should use GDB_ARCH_IS_TRAP_BRKPT
-   instead of TRAP_BRKPT to abstract out these peculiarities.  */
+   The MIPS kernel uses SI_KERNEL for all kernel generated traps.
+   Since:
+
+     - MIPS doesn't do hardware single-step.
+     - We don't need to care about exec SIGTRAPs --- we assume
+       PTRACE_EVENT_EXEC is available.
+     - The MIPS kernel doesn't support hardware breakpoints.
+
+   on MIPS, all we need to care about is distinguishing between
+   software breakpoints and hardware watchpoints, which can be done by
+   peeking the debug registers.
+
+   The generic Linux target code should use GDB_ARCH_IS_TRAP_* instead
+   of TRAP_* to abstract out these peculiarities.  */
 #if defined __i386__ || defined __x86_64__
 # define GDB_ARCH_IS_TRAP_BRKPT(X) ((X) == SI_KERNEL)
+# define GDB_ARCH_IS_TRAP_HWBKPT(X) ((X) == TRAP_HWBKPT)
 #elif defined __powerpc__
 # define GDB_ARCH_IS_TRAP_BRKPT(X) ((X) == SI_KERNEL || (X) == TRAP_BRKPT)
+# define GDB_ARCH_IS_TRAP_HWBKPT(X) ((X) == TRAP_HWBKPT)
+#elif defined __mips__
+# define GDB_ARCH_IS_TRAP_BRKPT(X) ((X) == SI_KERNEL)
+# define GDB_ARCH_IS_TRAP_HWBKPT(X) ((X) == SI_KERNEL)
 #else
 # define GDB_ARCH_IS_TRAP_BRKPT(X) ((X) == TRAP_BRKPT)
+# define GDB_ARCH_IS_TRAP_HWBKPT(X) ((X) == TRAP_HWBKPT)
 #endif
 
 #ifndef TRAP_HWBKPT