qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Jan Kiszka <jan.kiszka@siemens.com>
To: qemu-devel@nongnu.org
Subject: [Qemu-devel] [PATCH 12/13] Introduce BP_CPU as a breakpoint type
Date: Thu, 03 Jul 2008 18:03:11 +0200	[thread overview]
Message-ID: <486CF83F.9010604@siemens.com> (raw)
In-Reply-To: <486CF559.5090805@siemens.com>

Add another breakpoint/watchpoint type to BP_GDB: BP_CPU. This type is
intended for hardware-assisted break/watchpoint emulations like the x86
architecture requires.

To keep the highest priority for BP_GDB breakpoints, this type is
always inserted at the head of break/watchpoint lists, thus is found
first when looking up the origin of a debug interruption.

Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
---
 cpu-all.h |    1 +
 exec.c    |   46 ++++++++++++++++++++++++++++++++++++++--------
 2 files changed, 39 insertions(+), 8 deletions(-)

Index: b/cpu-all.h
===================================================================
--- a/cpu-all.h
+++ b/cpu-all.h
@@ -806,6 +806,7 @@ void cpu_reset_interrupt(CPUState *env,
 #define BP_STOP_BEFORE_ACCESS 0x04
 #define BP_WATCHPOINT_HIT     0x08
 #define BP_GDB                0x10
+#define BP_CPU                0x20
 
 int cpu_breakpoint_insert(CPUState *env, target_ulong pc, int flags,
                           CPUBreakpoint **breakpoint);
Index: b/exec.c
===================================================================
--- a/exec.c
+++ b/exec.c
@@ -1236,7 +1236,7 @@ int cpu_watchpoint_insert(CPUState *env,
                           int flags, CPUWatchpoint **watchpoint)
 {
     target_ulong len_mask = ~(len - 1);
-    CPUWatchpoint *wp;
+    CPUWatchpoint *wp, *prev_wp;
 
     /* sanity checks: allow power-of-2 lengths, deny unaligned watchpoints */
     if ((len != 1 && len != 2 && len != 4) || (addr & ~len_mask))
@@ -1250,11 +1250,26 @@ int cpu_watchpoint_insert(CPUState *env,
     wp->len_mask = len_mask;
     wp->flags = flags;
 
-    wp->next = env->watchpoints;
-    wp->prev = NULL;
+    /* keep all GDB-injected watchpoints in front */
+    if (!(flags & BP_GDB) && env->watchpoints) {
+        prev_wp = env->watchpoints;
+        while (prev_wp->next != NULL && (prev_wp->next->flags & BP_GDB))
+            prev_wp = prev_wp->next;
+    } else {
+        prev_wp = NULL;
+    }
+
+    /* Insert new watchpoint */
+    if (prev_wp) {
+        wp->next = prev_wp->next;
+        prev_wp->next = wp;
+    } else {
+        wp->next = env->watchpoints;
+        env->watchpoints = wp;
+    }
     if (wp->next)
         wp->next->prev = wp;
-    env->watchpoints = wp;
+    wp->prev = prev_wp;
 
     tlb_flush_page(env, addr);
 
@@ -1310,7 +1325,7 @@ int cpu_breakpoint_insert(CPUState *env,
                           CPUBreakpoint **breakpoint)
 {
 #if defined(TARGET_HAS_ICE)
-    CPUBreakpoint *bp;
+    CPUBreakpoint *bp, *prev_bp;
 
     bp = qemu_malloc(sizeof(*bp));
     if (!bp)
@@ -1319,11 +1334,26 @@ int cpu_breakpoint_insert(CPUState *env,
     bp->pc = pc;
     bp->flags = flags;
 
-    bp->next = env->breakpoints;
-    bp->prev = NULL;
+    /* keep all GDB-injected breakpoints in front */
+    if (!(flags & BP_GDB) && env->breakpoints) {
+        prev_bp = env->breakpoints;
+        while (prev_bp->next != NULL && (prev_bp->next->flags & BP_GDB))
+            prev_bp = prev_bp->next;
+    } else {
+        prev_bp = NULL;
+    }
+
+    /* Insert new breakpoint */
+    if (prev_bp) {
+        bp->next = prev_bp->next;
+        prev_bp->next = bp;
+    } else {
+        bp->next = env->breakpoints;
+        env->breakpoints = bp;
+    }
     if (bp->next)
         bp->next->prev = bp;
-    env->breakpoints = bp;
+    bp->prev = prev_bp;
 
     breakpoint_invalidate(env, pc);
 

  parent reply	other threads:[~2008-07-03 16:04 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-07-03 15:50 [Qemu-devel] [PATCH 0/13] Enhance debugging support - 2nd take Jan Kiszka
2008-07-03 15:56 ` [Qemu-devel] [PATCH 1/13] Return appropriate watch message to gdb Jan Kiszka
2008-07-03 15:57 ` [Qemu-devel] [PATCH 2/13] Refactor and enhance break/watchpoint API Jan Kiszka
2008-07-03 15:58 ` [Qemu-devel] [PATCH 3/13] Set mem_io_vaddr on io_read Jan Kiszka
2008-07-03 15:58 ` [Qemu-devel] [PATCH 4/13] Respect length of watchpoints Jan Kiszka
2008-07-03 15:59 ` [Qemu-devel] [PATCH 5/13] Introduce next_cflags Jan Kiszka
2008-07-04 22:03   ` [Qemu-devel] [PATCH 5/13] Introduce next_cflags - v2 Jan Kiszka
2008-07-03 15:59 ` [Qemu-devel] [PATCH 6/13] Switch self-modified code recompilation to next_cflags Jan Kiszka
2008-07-03 16:00 ` [Qemu-devel] [PATCH 7/13] Restore pc on watchpoint hits Jan Kiszka
2008-07-04 20:49   ` Paul Brook
2008-07-04 21:15     ` [Qemu-devel] " Jan Kiszka
2008-07-03 16:00 ` [Qemu-devel] [PATCH 8/13] Remove premature memop TB terminations Jan Kiszka
2008-07-03 16:01 ` [Qemu-devel] [PATCH 9/13] Improve debugging of SMP guests Jan Kiszka
2008-07-03 16:02 ` [Qemu-devel] [PATCH 10/13] Introduce BP_WATCHPOINT_HIT flag Jan Kiszka
2008-07-03 16:02 ` [Qemu-devel] [PATCH 11/13] Add debug exception hook Jan Kiszka
2008-07-03 16:03 ` Jan Kiszka [this message]
2008-07-03 16:03 ` [Qemu-devel] [PATCH 13/13] x86: Debug register emulation Jan Kiszka
2008-07-11 10:40 ` [Qemu-devel] Re: [PATCH 0/13] Enhance debugging support - 2nd take Jan Kiszka
2008-08-20 14:45 ` [Qemu-devel] [RESEND][PATCH " Jan Kiszka
2008-08-20 14:47   ` [Qemu-devel] [RESEND][PATCH 1/13] Return appropriate watch message to gdb Jan Kiszka
2008-08-20 14:49   ` [Qemu-devel] [RESEND][PATCH 2/13] Refactor and enhance break/watchpoint API Jan Kiszka
2008-09-07  2:14     ` Anthony Liguori
2008-09-07  7:15       ` Aurelien Jarno
2008-09-08  8:42         ` [Qemu-devel] " Jan Kiszka
2008-09-07 19:11       ` [Qemu-devel] " Blue Swirl
2008-09-08  8:43         ` [Qemu-devel] " Jan Kiszka
2008-09-15 13:22       ` Jan Kiszka
2008-09-08 20:07     ` Jan Kiszka
2008-08-20 14:50   ` [Qemu-devel] [RESEND][PATCH 3/13] Set mem_io_vaddr on io_read Jan Kiszka
2008-08-20 14:51   ` [Qemu-devel] [RESEND][PATCH 4/13] Respect length of watchpoints Jan Kiszka
2008-08-20 14:52   ` [Qemu-devel] [RESEND][PATCH 5/13] Introduce next_cflags Jan Kiszka
2008-08-20 14:54   ` [Qemu-devel] [RESEND][PATCH 6/13] Switch self-modified code recompilation to next_cflags Jan Kiszka
2008-08-20 14:56   ` [Qemu-devel] [RESEND][PATCH 7/13] Restore pc on watchpoint hits Jan Kiszka
2008-08-20 14:57   ` [Qemu-devel] [RESEND][PATCH 8/13] Remove premature memop TB terminations Jan Kiszka
2008-08-20 14:59   ` [Qemu-devel] [RESEND][PATCH 10/13] Introduce BP_WATCHPOINT_HIT flag Jan Kiszka
2008-08-20 15:00   ` [Qemu-devel] [RESEND][PATCH 11/13] Add debug exception hook Jan Kiszka
2008-08-20 15:01   ` [Qemu-devel] [RESEND][PATCH 12/13] Introduce BP_CPU as a breakpoint type Jan Kiszka
2008-08-20 15:02   ` [Qemu-devel] [RESEND][PATCH 13/13] x86: Debug register emulation Jan Kiszka
2008-08-20 15:04   ` [Qemu-devel] [RESEND][PATCH 9/13] Improve debugging of SMP guests Jan Kiszka
2008-08-21 20:19   ` [Qemu-devel] [RESEND][PATCH 0/13] Enhance debugging support - 2nd take Anthony Liguori
2008-08-21 22:53     ` [Qemu-devel] " Jan Kiszka
  -- strict thread matches above, loose matches on Subject: below --
2008-10-14  9:12 [Qemu-devel] [PATCH 00/13] Enhance debugging support - 3rd take Jan Kiszka
2008-10-14  9:12 ` [Qemu-devel] [PATCH 12/13] Introduce BP_CPU as a breakpoint type Jan Kiszka

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=486CF83F.9010604@siemens.com \
    --to=jan.kiszka@siemens.com \
    --cc=qemu-devel@nongnu.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 NNTP newsgroup(s).