From: prasad@linux.vnet.ibm.com
To: mingo@elte.hu
Cc: Andrew Morton <akpm@linux-foundation.org>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
Alan Stern <stern@rowland.harvard.edu>,
Roland McGrath <roland@redhat.com>,
"K.Prasad" <prasad@linux.vnet.ibm.com>
Subject: [patch 03/11] Modifying generic debug exception to use virtual debug registers
Date: Thu, 5 Mar 2009 10:08:23 +0530 [thread overview]
Message-ID: <20090305043823.GD17747@in.ibm.com> (raw)
In-Reply-To: 20090305043440.189041194@linux.vnet.ibm.com
[-- Attachment #1: 3 --]
[-- Type: text/plain, Size: 3646 bytes --]
From: Alan Stern <stern@rowland.harvard.edu>
This patch modifies the breakpoint exception handler code to use the abstract
register names.
[K.Prasad: Split-out from the bigger patch and minor changes following
re-basing]
Signed-off-by: K.Prasad <prasad@linux.vnet.ibm.com>
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
---
arch/x86/kernel/traps.c | 73 ++++++++++++++++--------------------------------
1 file changed, 25 insertions(+), 48 deletions(-)
Index: linux-2.6-tip/arch/x86/kernel/traps.c
===================================================================
--- linux-2.6-tip.orig/arch/x86/kernel/traps.c
+++ linux-2.6-tip/arch/x86/kernel/traps.c
@@ -576,13 +576,14 @@ asmlinkage __kprobes struct pt_regs *syn
dotraplinkage void __kprobes do_debug(struct pt_regs *regs, long error_code)
{
struct task_struct *tsk = current;
- unsigned long condition;
+ unsigned long dr6;
int si_code;
- get_debugreg(condition, 6);
+ get_debugreg(dr6, 6);
+ set_debugreg(0, 6); /* DR6 may or may not be cleared by the CPU */
/* Catch kmemcheck conditions first of all! */
- if (condition & DR_STEP && kmemcheck_trap(regs))
+ if (dr6 & DR_STEP && kmemcheck_trap(regs))
return;
/*
@@ -591,61 +592,37 @@ dotraplinkage void __kprobes do_debug(st
clear_tsk_thread_flag(tsk, TIF_DEBUGCTLMSR);
tsk->thread.debugctlmsr = 0;
- if (notify_die(DIE_DEBUG, "debug", regs, condition, error_code,
+ /* Store the virtualized DR6 value */
+ tsk->thread.vdr6 = dr6;
+
+ if (notify_die(DIE_DEBUG, "debug", regs, dr6, error_code,
SIGTRAP) == NOTIFY_STOP)
return;
/* It's safe to allow irq's after DR6 has been saved */
preempt_conditional_sti(regs);
- /* Mask out spurious debug traps due to lazy DR7 setting */
- if (condition & (DR_TRAP0|DR_TRAP1|DR_TRAP2|DR_TRAP3)) {
- if (!tsk->thread.debugreg7)
- goto clear_dr7;
- }
-
-#ifdef CONFIG_X86_32
- if (regs->flags & X86_VM_MASK)
- goto debug_vm86;
-#endif
-
- /* Save debug status register where ptrace can see it */
- tsk->thread.debugreg6 = condition;
-
- /*
- * Single-stepping through TF: make sure we ignore any events in
- * kernel space (but re-enable TF when returning to user mode).
- */
- if (condition & DR_STEP) {
- if (!user_mode(regs))
- goto clear_TF_reenable;
+ if (regs->flags & X86_VM_MASK) {
+ handle_vm86_trap((struct kernel_vm86_regs *) regs,
+ error_code, 1);
+ return;
}
- si_code = get_si_code(condition);
- /* Ok, finally something we can handle */
- send_sigtrap(tsk, regs, error_code, si_code);
-
/*
- * Disable additional traps. They'll be re-enabled when
- * the signal is delivered.
+ * Single-stepping through system calls: ignore any exceptions in
+ * kernel space, but re-enable TF when returning to user mode.
+ *
+ * We already checked v86 mode above, so we can check for kernel mode
+ * by just checking the CPL of CS.
*/
-clear_dr7:
- set_debugreg(0, 7);
- preempt_conditional_cli(regs);
- return;
-
-#ifdef CONFIG_X86_32
-debug_vm86:
- /* reenable preemption: handle_vm86_trap() might sleep */
- dec_preempt_count();
- handle_vm86_trap((struct kernel_vm86_regs *) regs, error_code, 1);
- conditional_cli(regs);
- return;
-#endif
-
-clear_TF_reenable:
- set_tsk_thread_flag(tsk, TIF_SINGLESTEP);
- regs->flags &= ~X86_EFLAGS_TF;
+ if ((dr6 & DR_STEP) && !user_mode(regs)) {
+ tsk->thread.vdr6 &= ~DR_STEP;
+ set_tsk_thread_flag(tsk, TIF_SINGLESTEP);
+ regs->flags &= ~X86_EFLAGS_TF;
+ }
+ si_code = get_si_code(dr6);
+ if (tsk->thread.vdr6 & (DR_STEP|DR_TRAP0|DR_TRAP1|DR_TRAP2|DR_TRAP3))
+ send_sigtrap(tsk, regs, error_code, si_code);
preempt_conditional_cli(regs);
return;
}
next prev parent reply other threads:[~2009-03-05 4:38 UTC|newest]
Thread overview: 69+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20090305043440.189041194@linux.vnet.ibm.com>
2009-03-05 4:37 ` [patch 01/11] Introducing generic hardware breakpoint handler interfaces prasad
2009-03-10 13:50 ` Ingo Molnar
2009-03-10 14:19 ` Alan Stern
2009-03-10 14:50 ` Ingo Molnar
2009-03-11 12:57 ` K.Prasad
2009-03-11 13:35 ` Ingo Molnar
2009-03-05 4:38 ` [patch 02/11] x86 architecture implementation of Hardware Breakpoint interfaces prasad
2009-03-10 14:09 ` Ingo Molnar
2009-03-10 14:59 ` Alan Stern
2009-03-10 15:18 ` Ingo Molnar
2009-03-10 17:11 ` Alan Stern
2009-03-10 17:26 ` Ingo Molnar
2009-03-10 20:30 ` Alan Stern
2009-03-11 12:12 ` Ingo Molnar
2009-03-11 12:50 ` K.Prasad
2009-03-11 13:10 ` Ingo Molnar
2009-03-14 3:46 ` Benjamin Herrenschmidt
2009-03-11 16:39 ` Alan Stern
2009-03-11 16:32 ` Alan Stern
2009-03-11 17:41 ` K.Prasad
2009-03-14 3:47 ` Benjamin Herrenschmidt
2009-03-14 3:43 ` Benjamin Herrenschmidt
2009-03-14 3:41 ` Benjamin Herrenschmidt
2009-03-14 3:40 ` Benjamin Herrenschmidt
2009-03-12 2:46 ` Roland McGrath
2009-03-13 3:43 ` Ingo Molnar
2009-03-13 14:04 ` Alan Stern
2009-03-13 14:13 ` Ingo Molnar
2009-03-13 19:01 ` K.Prasad
2009-03-13 21:21 ` Alan Stern
2009-03-14 12:24 ` Ingo Molnar
2009-03-14 16:10 ` Alan Stern
2009-03-14 16:39 ` Ingo Molnar
2009-03-14 3:51 ` Benjamin Herrenschmidt
2009-03-05 4:38 ` prasad [this message]
2009-03-05 4:38 ` [patch 04/11] Introduce virtual debug register in thread_struct and wrapper-routines around process related functions prasad
2009-03-10 14:35 ` Ingo Molnar
2009-03-10 15:53 ` Alan Stern
2009-03-10 17:06 ` Ingo Molnar
2009-03-12 2:26 ` Roland McGrath
2009-03-05 4:38 ` [patch 05/11] Use wrapper routines around debug registers in processor " prasad
2009-03-05 4:40 ` [patch 06/11] Use virtual debug registers in process/thread handling code prasad
2009-03-10 14:49 ` Ingo Molnar
2009-03-10 16:05 ` Alan Stern
2009-03-10 16:58 ` Ingo Molnar
2009-03-10 17:07 ` Ingo Molnar
2009-03-10 20:10 ` Alan Stern
2009-03-11 11:53 ` Ingo Molnar
2009-03-05 4:40 ` [patch 07/11] Modify signal handling code to refrain from re-enabling HW Breakpoints prasad
2009-03-05 4:40 ` [patch 08/11] Modify Ptrace routines to access breakpoint registers prasad
2009-03-10 14:40 ` Ingo Molnar
2009-03-10 15:54 ` Alan Stern
2009-03-12 3:14 ` Roland McGrath
2009-03-05 4:41 ` [patch 09/11] Cleanup HW Breakpoint registers before kexec prasad
2009-03-10 14:42 ` Ingo Molnar
2009-03-05 4:41 ` [patch 10/11] Sample HW breakpoint over kernel data address prasad
2009-03-05 4:43 ` prasad
2009-03-05 4:43 ` [patch 11/11] ftrace plugin for kernel symbol tracing using HW Breakpoint interfaces prasad
2009-03-05 6:37 ` Frederic Weisbecker
2009-03-05 9:16 ` Ingo Molnar
2009-03-05 13:15 ` K.Prasad
2009-03-05 13:28 ` Ingo Molnar
2009-03-05 11:33 ` K.Prasad
2009-03-05 12:19 ` K.Prasad
2009-03-05 12:30 ` Frederic Weisbecker
2009-03-05 12:28 ` Frederic Weisbecker
2009-03-05 15:00 ` Steven Rostedt
2009-03-05 14:54 ` Steven Rostedt
[not found] <20090307045120.039324630@linux.vnet.ibm.com>
2009-03-07 5:06 ` [Patch 03/11] Modifying generic debug exception to use virtual debug registers prasad
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=20090305043823.GD17747@in.ibm.com \
--to=prasad@linux.vnet.ibm.com \
--cc=akpm@linux-foundation.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=roland@redhat.com \
--cc=stern@rowland.harvard.edu \
/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