From: Christoph Hellwig <hch@lst.de>
To: roland@redhat.com, oleg@redhat.com, akpm@linux-foundation.org,
starvik@axis.com, jesper.nilsson@axis.com
Cc: linux-kernel@vger.kernel.org, linux-arch@vger.kernel.org
Subject: [PATCH, RFC 13/14] cris arch-v32: use generic ptrace_resume code
Date: Tue, 2 Feb 2010 20:00:42 +0100 [thread overview]
Message-ID: <20100202190042.GM3630@lst.de> (raw)
In-Reply-To: <20100202185755.GA3630@lst.de>
Use the generic ptrace_resume code for PTRACE_SYSCALL, PTRACE_CONT,
PTRACE_KILL and PTRACE_SINGLESTEP. This implies defining
arch_has_single_step in <asm/ptrace.h> and implementing the
user_enable_single_step and user_disable_single_step functions, which
also causes the breakpoint information to be cleared on fork, which
could be considered a bug fix.
Also the TIF_SYSCALL_TRACE thread flag is now cleared on PTRACE_KILL
which it previously wasn't which is consistent with all architectures
using the modern ptrace code.
The way breakpoints are disabled is entirely inconsistent
currently, I tried to make some sense of it, but I suspect all of the
content of ptrace_disable should be moved into user_disable_single_step,
this defintively needs some revisting as the current patch changes
behaviour in not quite designed ways.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Index: linux-2.6/arch/cris/arch-v32/kernel/ptrace.c
===================================================================
--- linux-2.6.orig/arch/cris/arch-v32/kernel/ptrace.c 2010-02-02 19:23:12.314003057 +0100
+++ linux-2.6/arch/cris/arch-v32/kernel/ptrace.c 2010-02-02 19:25:27.396253967 +0100
@@ -78,6 +78,35 @@ int put_reg(struct task_struct *task, un
return 0;
}
+void user_enable_single_step(struct task_struct *child)
+{
+ unsigned long tmp;
+
+ /*
+ * Set up SPC if not set already (in which case we have no other
+ * choice but to trust it).
+ */
+ if (!get_reg(child, PT_SPC)) {
+ /* In case we're stopped in a delay slot. */
+ tmp = get_reg(child, PT_ERP) & ~1;
+ put_reg(child, PT_SPC, tmp);
+ }
+ tmp = get_reg(child, PT_CCS) | SBIT_USER;
+ put_reg(child, PT_CCS, tmp);
+}
+
+void user_disable_single_step(struct task_struct *child)
+{
+ put_reg(child, PT_SPC, 0);
+
+ if (!get_debugreg(child->pid, PT_BP_CTRL)) {
+ unsigned long tmp;
+ /* If no h/w bp configured, disable S bit. */
+ tmp = get_reg(child, PT_CCS) & ~SBIT_USER;
+ put_reg(child, PT_CCS, tmp);
+ }
+}
+
/*
* Called by kernel/ptrace.c when detaching.
*
@@ -89,8 +118,7 @@ ptrace_disable(struct task_struct *child
unsigned long tmp;
/* Deconfigure SPC and S-bit. */
- tmp = get_reg(child, PT_CCS) & ~SBIT_USER;
- put_reg(child, PT_CCS, tmp);
+ user_disable_single_step(child);
put_reg(child, PT_SPC, 0);
/* Deconfigure any watchpoints associated with the child. */
@@ -163,83 +191,6 @@ long arch_ptrace(struct task_struct *chi
ret = 0;
break;
- case PTRACE_SYSCALL:
- case PTRACE_CONT:
- ret = -EIO;
-
- if (!valid_signal(data))
- break;
-
- /* Continue means no single-step. */
- put_reg(child, PT_SPC, 0);
-
- if (!get_debugreg(child->pid, PT_BP_CTRL)) {
- unsigned long tmp;
- /* If no h/w bp configured, disable S bit. */
- tmp = get_reg(child, PT_CCS) & ~SBIT_USER;
- put_reg(child, PT_CCS, tmp);
- }
-
- if (request == PTRACE_SYSCALL) {
- set_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
- }
- else {
- clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
- }
-
- child->exit_code = data;
-
- /* TODO: make sure any pending breakpoint is killed */
- wake_up_process(child);
- ret = 0;
-
- break;
-
- /* Make the child exit by sending it a sigkill. */
- case PTRACE_KILL:
- ret = 0;
-
- if (child->exit_state == EXIT_ZOMBIE)
- break;
-
- child->exit_code = SIGKILL;
-
- /* Deconfigure single-step and h/w bp. */
- ptrace_disable(child);
-
- /* TODO: make sure any pending breakpoint is killed */
- wake_up_process(child);
- break;
-
- /* Set the trap flag. */
- case PTRACE_SINGLESTEP: {
- unsigned long tmp;
- ret = -EIO;
-
- /* Set up SPC if not set already (in which case we have
- no other choice but to trust it). */
- if (!get_reg(child, PT_SPC)) {
- /* In case we're stopped in a delay slot. */
- tmp = get_reg(child, PT_ERP) & ~1;
- put_reg(child, PT_SPC, tmp);
- }
- tmp = get_reg(child, PT_CCS) | SBIT_USER;
- put_reg(child, PT_CCS, tmp);
-
- if (!valid_signal(data))
- break;
-
- clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
-
- /* TODO: set some clever breakpoint mechanism... */
-
- child->exit_code = data;
- wake_up_process(child);
- ret = 0;
- break;
-
- }
-
/* Get all GP registers from the child. */
case PTRACE_GETREGS: {
int i;
Index: linux-2.6/arch/cris/include/arch-v32/arch/ptrace.h
===================================================================
--- linux-2.6.orig/arch/cris/include/arch-v32/arch/ptrace.h 2010-02-02 19:25:39.092254418 +0100
+++ linux-2.6/arch/cris/include/arch-v32/arch/ptrace.h 2010-02-02 19:25:46.173254455 +0100
@@ -108,6 +108,7 @@ struct switch_stack {
#ifdef __KERNEL__
+#define arch_has_single_step() (1)
#define user_mode(regs) (((regs)->ccs & (1 << (U_CCS_BITNR + CCS_SHIFT))) != 0)
#define instruction_pointer(regs) ((regs)->erp)
extern void show_regs(struct pt_regs *);
next prev parent reply other threads:[~2010-02-02 19:01 UTC|newest]
Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-02-02 18:57 [PATCH 1/14] move user_enable_single_step & co prototypes to linux/ptrace.h Christoph Hellwig
2010-02-02 18:58 ` [PATCH 2/14] alpha: use generic ptrace_resume code Christoph Hellwig
2010-02-03 4:35 ` Matt Turner
2010-02-02 18:58 ` [PATCH 3/14] arm: " Christoph Hellwig
2010-02-02 18:58 ` [PATCH 4/14] avr32: " Christoph Hellwig
2010-02-03 3:17 ` Haavard Skinnemoen
2010-02-03 8:36 ` Christoph Hellwig
2010-02-03 19:22 ` Oleg Nesterov
2010-02-03 19:28 ` Christoph Hellwig
2010-02-02 18:59 ` [PATCH 5/14] blackfin: " Christoph Hellwig
2010-02-02 20:29 ` Mike Frysinger
2010-02-03 19:36 ` Mike Frysinger
2010-02-03 19:42 ` Christoph Hellwig
2010-02-11 9:43 ` [PATCH 0/2] Blackfin: " Mike Frysinger
2010-02-11 9:43 ` [PATCH 1/2] Blackfin: initial tracehook support Mike Frysinger
2010-02-11 20:46 ` Roland McGrath
2010-02-11 23:54 ` Mike Frysinger
2010-02-12 3:24 ` Roland McGrath
2010-02-12 4:33 ` Mike Frysinger
2010-02-12 15:24 ` Oleg Nesterov
2010-02-12 20:44 ` Roland McGrath
2010-02-13 9:41 ` Mike Frysinger
2010-02-15 7:36 ` Mike Frysinger
2010-02-15 20:07 ` Roland McGrath
2010-02-11 9:43 ` [PATCH 2/2] Blackfin: use generic ptrace_resume code Mike Frysinger
2010-02-02 18:59 ` [PATCH 6/14] h8300: " Christoph Hellwig
2010-02-02 18:59 ` [PATCH 7/14] m68knommu: " Christoph Hellwig
2010-02-03 6:54 ` Greg Ungerer
2010-02-02 18:59 ` [PATCH 8/14] microblaze: " Christoph Hellwig
2010-02-03 11:00 ` Michal Simek
2010-02-02 18:59 ` [PATCH 9/14] mips: " Christoph Hellwig
2010-02-02 19:19 ` Ralf Baechle
2010-02-02 19:00 ` [PATCH 10/14] um: " Christoph Hellwig
2010-02-02 19:00 ` [PATCH 11/14] xtensa: " Christoph Hellwig
2010-02-02 19:00 ` [PATCH 12/14] cris arch-v10: " Christoph Hellwig
2010-02-02 19:00 ` Christoph Hellwig [this message]
2010-02-02 19:00 ` [PATCH, RFC 14/14] m32r: " Christoph Hellwig
2010-02-03 8:42 ` [PATCH 1/14] move user_enable_single_step & co prototypes to linux/ptrace.h Mike Frysinger
2010-02-03 8:56 ` Christoph Hellwig
2010-02-08 10:50 ` David Howells
2010-02-08 19:51 ` Roland McGrath
2010-02-10 22:03 ` Christoph Hellwig
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=20100202190042.GM3630@lst.de \
--to=hch@lst.de \
--cc=akpm@linux-foundation.org \
--cc=jesper.nilsson@axis.com \
--cc=linux-arch@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=oleg@redhat.com \
--cc=roland@redhat.com \
--cc=starvik@axis.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;
as well as URLs for NNTP newsgroup(s).