[kernel] Update utrace.patch from Oleg Nesterov
Josh Boyer
jwboyer at fedoraproject.org
Wed Jan 25 21:48:47 UTC 2012
commit 62e9e5ab948df1986d0c054176b6712e1475d9a3
Author: Josh Boyer <jwboyer at redhat.com>
Date: Wed Jan 25 16:48:32 2012 -0500
Update utrace.patch from Oleg Nesterov
kernel.spec | 7 +-
utrace.patch | 8297 ++++++++++++++++++++--------------------------------------
2 files changed, 2905 insertions(+), 5399 deletions(-)
---
diff --git a/kernel.spec b/kernel.spec
index ce5061d..9d029d2 100644
--- a/kernel.spec
+++ b/kernel.spec
@@ -54,7 +54,7 @@ Summary: The Linux kernel
# For non-released -rc kernels, this will be appended after the rcX and
# gitX tags, so a 3 here would become part of release "0.rcX.gitX.3"
#
-%global baserelease 1
+%global baserelease 2
%global fedora_build %{baserelease}
# base_sublevel is the kernel version we're starting with and patching
@@ -1417,7 +1417,7 @@ ApplyPatch revert-efi-rtclock.patch
ApplyPatch efi-dont-map-boot-services-on-32bit.patch
# utrace.
-# pplyPatch utrace.patch
+ApplyPatch utrace.patch
ApplyPatch ext4-Support-check-none-nocheck-mount-options.patch
@@ -2266,6 +2266,9 @@ fi
# ||----w |
# || ||
%changelog
+* Wed Jan 25 2012 Josh Boyer <jwboyer at redhat.com>
+- Update utrace.patch from Oleg Nesterov
+
* Wed Jan 25 2012 Josh Boyer <jwboyer at redhat.com> - 3.3.0-0.rc1.git2.1
- Linux 3.3-rc1-git2 (upstream f8275f9694b8adf9f3498e747ea4c3e8b984499b)
diff --git a/utrace.patch b/utrace.patch
index 060d351..0e34971 100644
--- a/utrace.patch
+++ b/utrace.patch
@@ -1,81 +1,70 @@
-From davej Wed Aug 3 15:16:11 2011
-From oleg at redhat.com Mon Nov 21 15:06:14 2011
-Return-Path: oleg at redhat.com
-Received: from zmta01.collab.prod.int.phx2.redhat.com (LHLO
- zmta01.collab.prod.int.phx2.redhat.com) (10.5.5.31) by
- zmail13.collab.prod.int.phx2.redhat.com with LMTP; Mon, 21 Nov 2011
- 15:06:14 -0500 (EST)
-Received: from localhost (localhost.localdomain [127.0.0.1])
- by zmta01.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 9A1F8D50F1;
- Mon, 21 Nov 2011 15:06:14 -0500 (EST)
-Received: from zmta01.collab.prod.int.phx2.redhat.com ([127.0.0.1])
- by localhost (zmta01.collab.prod.int.phx2.redhat.com [127.0.0.1]) (amavisd-new, port 10024)
- with ESMTP id gJzpul4rDNnA; Mon, 21 Nov 2011 15:06:14 -0500 (EST)
-Received: from int-mx12.intmail.prod.int.phx2.redhat.com (int-mx12.intmail.prod.int.phx2.redhat.com [10.5.11.25])
- by zmta01.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 7DD4BD50EA;
- Mon, 21 Nov 2011 15:06:14 -0500 (EST)
-Received: from tranklukator.englab.brq.redhat.com (dhcp-1-232.brq.redhat.com [10.34.1.232])
- by int-mx12.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with SMTP id pALK6BJJ022074;
- Mon, 21 Nov 2011 15:06:12 -0500
-Received: by tranklukator.englab.brq.redhat.com (nbSMTP-1.00) for uid 500
- oleg at redhat.com; Mon, 21 Nov 2011 21:01:32 +0100 (CET)
-Date: Mon, 21 Nov 2011 21:01:30 +0100
-From: Oleg Nesterov <oleg at redhat.com>
-To: Dave Jones <davej at redhat.com>, "Frank Ch. Eigler" <fche at redhat.com>,
- Josh Boyer <jwboyer at redhat.com>, Josh Stone <jistone at redhat.com>,
- Kyle McMartin <kmcmartin at redhat.com>
-Cc: kernel at lists.fedoraproject.org
-Subject: [PATCH 01/33] utrace core
-Message-ID: <20111121200130.GA27756 at redhat.com>
-MIME-Version: 1.0
-Content-Type: text/plain; charset=us-ascii
-Content-Disposition: inline
-In-Reply-To: <20111121200039.GA27699 at redhat.com>
-User-Agent: Mutt/1.5.18 (2008-05-17)
-X-Scanned-By: MIMEDefang 2.68 on 10.5.11.25
-Status: RO
-Content-Length: 139951
-Lines: 3867
+utrace for 3.3-rc1 kernel, on top of dcd6c92267155e70a94b3927bce681ce74b80d1f.
+
+The split-out series is available in the git repository at:
+
+ git at github.com:utrace/linux.git utrace-3.3
+
+Oleg Nesterov (31):
+ utrace: add utrace_init_task/utrace_free_task calls
+ tracehooks: add utrace hooks
+ tracehooks: reintroduce tracehook_consider_fatal_signal()
+ add utrace hooks into sig_ignored() and recalc_sigpending()
+ restore the EXEC/EXIT/CLONE utrace hooks
+ utrace: utrace_report_death() can use task_utrace_struct()
+ restore the DEATH/REAP utrace hooks
+ utrace: remove jobctl bits
+ ptrace: take ->siglock around s/TRACED/RUNNING/
+ introduce wake_up_quiescent()
+ introduce ptrace_signal_wake_up()
+ wait_task_inactive: treat task->state and match_state as bitmasks
+ introduce TASK_UTRACED state
+ utrace: use TASK_UTRACED instead of TASK_TRACED
+ reintroduce tracehook_finish_jctl() as utrace_end_stop()
+ teach wake_up_quiescent() to do "selective" wake_up
+ ptrace_stop: do not assume the task is running after wake_up_quiescent()
+ get_signal_to_deliver: restore/restructure utrace/ptrace signal reporting
+ utrace_get_signal: s/JOBCTL_STOP_PENDING/JOBCTL_PENDING_MASK/
+ introduce ptrace_set_syscall_trace()
+ introduce PT_SYSCALL_TRACE flag
+ utrace: don't clear TIF_SYSCALL_TRACE if it is needed by ptrace
+ introduce task_utrace_lock/task_utrace_unlock
+ teach ptrace_set_syscall_trace() to play well with utrace
+ introduce PT_SINGLE_STEP and PT_SINGLE_BLOCK
+ utrace: finish_resume_report: don't do user_xxx_step() if ptrace_wants_step()
+ ptrace: shift user_*_step() from ptrace_resume() to ptrace_stop()
+ ptrace_disable: no need to disable stepping
+ ptrace_report_syscall: check TIF_SYSCALL_EMU
+ utrace_resume: check irqs_disabled() to shut up lockdep
+ utrace: s390: fix the compile problem with traps.c
+
+Roland McGrath (1):
+ utrace core
+
+Tony Breeds (1):
+ ptrace_report_syscall: check if TIF_SYSCALL_EMU is defined
-From: Roland McGrath <roland at redhat.com>
-
-This adds the utrace facility, a new modular interface in the kernel for
-implementing user thread tracing and debugging. This fits on top of the
-tracehook_* layer, so the new code is well-isolated.
-
-The new interface is in <linux/utrace.h> and the DocBook utrace book
-describes it. It allows for multiple separate tracing engines to work in
-parallel without interfering with each other. Higher-level tracing
-facilities can be implemented as loadable kernel modules using this layer.
-
-The new facility is made optional under CONFIG_UTRACE.
-When this is not enabled, no new code is added.
-It can only be enabled on machines that have all the
-prerequisites and select CONFIG_HAVE_ARCH_TRACEHOOK.
-
-In this initial version, utrace and ptrace do not play together at all,
-the next patches try to fix this.
-
-This is is same/old utrace-core patch except:
-
- - use task->jobctl/JOBCTL_STOP_DEQUEUED instead of removed
- signal->flags/SIGNAL_STOP_DEQUEUED in utrace_get_signal()
-
- - do not include the tracehook changes, this comes with the
- next patches
-
-Signed-off-by: Roland McGrath <roland at redhat.com>
Signed-off-by: Oleg Nesterov <oleg at redhat.com>
---
Documentation/DocBook/Makefile | 2 +-
Documentation/DocBook/utrace.tmpl | 589 +++++++++
- fs/proc/array.c | 3 +
- include/linux/sched.h | 5 +
- include/linux/utrace.h | 692 +++++++++++
+ arch/s390/kernel/traps.c | 6 +-
+ arch/x86/kernel/ptrace.c | 1 -
+ fs/exec.c | 5 +-
+ fs/proc/array.c | 14 +-
+ include/linux/ptrace.h | 7 +
+ include/linux/sched.h | 25 +-
+ include/linux/signal.h | 2 +
+ include/linux/tracehook.h | 59 +-
+ include/linux/utrace.h | 773 ++++++++++++
init/Kconfig | 9 +
kernel/Makefile | 1 +
- kernel/utrace.c | 2440 +++++++++++++++++++++++++++++++++++++
- 8 files changed, 3740 insertions(+), 1 deletions(-)
+ kernel/exit.c | 5 +
+ kernel/fork.c | 9 +
+ kernel/ptrace.c | 57 +-
+ kernel/sched/core.c | 2 +-
+ kernel/signal.c | 97 ++-
+ kernel/utrace.c | 2462 +++++++++++++++++++++++++++++++++++++
+ 19 files changed, 4069 insertions(+), 56 deletions(-)
create mode 100644 Documentation/DocBook/utrace.tmpl
create mode 100644 include/linux/utrace.h
create mode 100644 kernel/utrace.c
@@ -688,8 +677,69 @@ index 0000000..0c40add
+</chapter>
+
+</book>
+diff --git a/arch/s390/kernel/traps.c b/arch/s390/kernel/traps.c
+index 5ce3750..4f0b32f 100644
+--- a/arch/s390/kernel/traps.c
++++ b/arch/s390/kernel/traps.c
+@@ -18,7 +18,7 @@
+ #include <linux/kernel.h>
+ #include <linux/string.h>
+ #include <linux/errno.h>
+-#include <linux/ptrace.h>
++#include <linux/tracehook.h>
+ #include <linux/timer.h>
+ #include <linux/mm.h>
+ #include <linux/smp.h>
+@@ -330,7 +330,7 @@ void __kprobes do_per_trap(struct pt_regs *regs)
+
+ if (notify_die(DIE_SSTEP, "sstep", regs, 0, 0, SIGTRAP) == NOTIFY_STOP)
+ return;
+- if (!current->ptrace)
++ if (!tracehook_consider_fatal_signal(current, SIGTRAP))
+ return;
+ info.si_signo = SIGTRAP;
+ info.si_errno = 0;
+@@ -415,7 +415,7 @@ static void __kprobes illegal_op(struct pt_regs *regs)
+ if (get_user(*((__u16 *) opcode), (__u16 __user *) location))
+ return;
+ if (*((__u16 *) opcode) == S390_BREAKPOINT_U16) {
+- if (current->ptrace) {
++ if (tracehook_consider_fatal_signal(current, SIGTRAP)) {
+ info.si_signo = SIGTRAP;
+ info.si_errno = 0;
+ info.si_code = TRAP_BRKPT;
+diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c
+index 5026738..97687f3 100644
+--- a/arch/x86/kernel/ptrace.c
++++ b/arch/x86/kernel/ptrace.c
+@@ -809,7 +809,6 @@ static int ioperm_get(struct task_struct *target,
+ */
+ void ptrace_disable(struct task_struct *child)
+ {
+- user_disable_single_step(child);
+ #ifdef TIF_SYSCALL_EMU
+ clear_tsk_thread_flag(child, TIF_SYSCALL_EMU);
+ #endif
+diff --git a/fs/exec.c b/fs/exec.c
+index aeb135c..36a0cbe 100644
+--- a/fs/exec.c
++++ b/fs/exec.c
+@@ -1401,9 +1401,12 @@ int search_binary_handler(struct linux_binprm *bprm,struct pt_regs *regs)
+ */
+ bprm->recursion_depth = depth;
+ if (retval >= 0) {
+- if (depth == 0)
++ if (depth == 0) {
++ UTRACE_HOOK(current, EXEC,
++ report_exec(fmt, bprm, regs));
+ ptrace_event(PTRACE_EVENT_EXEC,
+ old_pid);
++ }
+ put_binfmt(fmt);
+ allow_write_access(bprm->file);
+ if (bprm->file)
diff --git a/fs/proc/array.c b/fs/proc/array.c
-index 3a1dafd..f0c0ea2 100644
+index c602b8d..a700b78 100644
--- a/fs/proc/array.c
+++ b/fs/proc/array.c
@@ -81,6 +81,7 @@
@@ -700,7 +750,25 @@ index 3a1dafd..f0c0ea2 100644
#include <asm/pgtable.h>
#include <asm/processor.h>
-@@ -192,6 +193,8 @@ static inline void task_state(struct seq_file *m, struct pid_namespace *ns,
+@@ -137,11 +138,12 @@ static const char * const task_state_array[] = {
+ "D (disk sleep)", /* 2 */
+ "T (stopped)", /* 4 */
+ "t (tracing stop)", /* 8 */
+- "Z (zombie)", /* 16 */
+- "X (dead)", /* 32 */
+- "x (dead)", /* 64 */
+- "K (wakekill)", /* 128 */
+- "W (waking)", /* 256 */
++ "t (tracing stop)", /* 16 (stopped by utrace) */
++ "Z (zombie)", /* 32 */
++ "X (dead)", /* 64 */
++ "x (dead)", /* 128 */
++ "K (wakekill)", /* 256 */
++ "W (waking)", /* 512 */
+ };
+
+ static inline const char *get_task_state(struct task_struct *tsk)
+@@ -192,6 +194,8 @@ static inline void task_state(struct seq_file *m, struct pid_namespace *ns,
cred->uid, cred->euid, cred->suid, cred->fsuid,
cred->gid, cred->egid, cred->sgid, cred->fsgid);
@@ -709,11 +777,87 @@ index 3a1dafd..f0c0ea2 100644
task_lock(p);
if (p->files)
fdt = files_fdtable(p->files);
+diff --git a/include/linux/ptrace.h b/include/linux/ptrace.h
+index c2f1f6a..236b920 100644
+--- a/include/linux/ptrace.h
++++ b/include/linux/ptrace.h
+@@ -104,6 +104,10 @@
+
+ #define PT_TRACE_MASK 0x000003f4
+
++#define PT_SYSCALL_TRACE 0x00020000
++#define PT_SINGLE_STEP 0x00040000
++#define PT_SINGLE_BLOCK 0x00080000
++
+ /* single stepping state bits (used on ARM and PA-RISC) */
+ #define PT_SINGLESTEP_BIT 31
+ #define PT_SINGLESTEP (1<<PT_SINGLESTEP_BIT)
+@@ -114,6 +118,7 @@
+ #include <linux/sched.h> /* For struct task_struct. */
+ #include <linux/err.h> /* for IS_ERR_VALUE */
+
++extern void ptrace_signal_wake_up(struct task_struct *p, int quiescent);
+
+ extern long arch_ptrace(struct task_struct *child, long request,
+ unsigned long addr, unsigned long data);
+@@ -228,6 +233,8 @@ static inline void ptrace_init_task(struct task_struct *child, bool ptrace)
+
+ if (unlikely(ptrace) && current->ptrace) {
+ child->ptrace = current->ptrace;
++ child->ptrace &=
++ ~(PT_SYSCALL_TRACE | PT_SINGLE_STEP | PT_SINGLE_BLOCK);
+ __ptrace_link(child, current->parent);
+
+ if (child->ptrace & PT_SEIZED)
diff --git a/include/linux/sched.h b/include/linux/sched.h
-index 68daf4f..4d45f93 100644
+index 4032ec1..e95c1dc 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
-@@ -1403,6 +1403,11 @@ struct task_struct {
+@@ -185,16 +185,17 @@ print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq)
+ #define TASK_UNINTERRUPTIBLE 2
+ #define __TASK_STOPPED 4
+ #define __TASK_TRACED 8
++#define __TASK_UTRACED 16
+ /* in tsk->exit_state */
+-#define EXIT_ZOMBIE 16
+-#define EXIT_DEAD 32
++#define EXIT_ZOMBIE 32
++#define EXIT_DEAD 64
+ /* in tsk->state again */
+-#define TASK_DEAD 64
+-#define TASK_WAKEKILL 128
+-#define TASK_WAKING 256
+-#define TASK_STATE_MAX 512
++#define TASK_DEAD 128
++#define TASK_WAKEKILL 256
++#define TASK_WAKING 512
++#define TASK_STATE_MAX 1024
+
+-#define TASK_STATE_TO_CHAR_STR "RSDTtZXxKW"
++#define TASK_STATE_TO_CHAR_STR "RSDTtUZXxKW"
+
+ extern char ___assert_task_state[1 - 2*!!(
+ sizeof(TASK_STATE_TO_CHAR_STR)-1 != ilog2(TASK_STATE_MAX)+1)];
+@@ -203,15 +204,16 @@ extern char ___assert_task_state[1 - 2*!!(
+ #define TASK_KILLABLE (TASK_WAKEKILL | TASK_UNINTERRUPTIBLE)
+ #define TASK_STOPPED (TASK_WAKEKILL | __TASK_STOPPED)
+ #define TASK_TRACED (TASK_WAKEKILL | __TASK_TRACED)
++#define TASK_UTRACED (TASK_WAKEKILL | __TASK_UTRACED)
+
+ /* Convenience macros for the sake of wake_up */
+ #define TASK_NORMAL (TASK_INTERRUPTIBLE | TASK_UNINTERRUPTIBLE)
+-#define TASK_ALL (TASK_NORMAL | __TASK_STOPPED | __TASK_TRACED)
++#define TASK_ALL (TASK_NORMAL | __TASK_STOPPED | __TASK_TRACED | __TASK_UTRACED)
+
+ /* get_task_state() */
+ #define TASK_REPORT (TASK_RUNNING | TASK_INTERRUPTIBLE | \
+ TASK_UNINTERRUPTIBLE | __TASK_STOPPED | \
+- __TASK_TRACED)
++ __TASK_TRACED | __TASK_UTRACED)
+
+ #define task_is_traced(task) ((task->state & __TASK_TRACED) != 0)
+ #define task_is_stopped(task) ((task->state & __TASK_STOPPED) != 0)
+@@ -1420,6 +1422,11 @@ struct task_struct {
#endif
seccomp_t seccomp;
@@ -725,60 +869,188 @@ index 68daf4f..4d45f93 100644
/* Thread group tracking */
u32 parent_exec_id;
u32 self_exec_id;
-diff --git a/include/linux/utrace.h b/include/linux/utrace.h
-new file mode 100644
-index 0000000..f251efe
---- /dev/null
-+++ b/include/linux/utrace.h
-@@ -0,0 +1,692 @@
-+/*
-+ * utrace infrastructure interface for debugging user processes
-+ *
-+ * Copyright (C) 2006-2009 Red Hat, Inc. All rights reserved.
-+ *
-+ * This copyrighted material is made available to anyone wishing to use,
-+ * modify, copy, or redistribute it subject to the terms and conditions
-+ * of the GNU General Public License v.2.
-+ *
-+ * Red Hat Author: Roland McGrath.
-+ *
-+ * This interface allows for notification of interesting events in a
-+ * thread. It also mediates access to thread state such as registers.
-+ * Multiple unrelated users can be associated with a single thread.
-+ * We call each of these a tracing engine.
-+ *
-+ * A tracing engine starts by calling utrace_attach_task() or
-+ * utrace_attach_pid() on the chosen thread, passing in a set of hooks
-+ * (&struct utrace_engine_ops), and some associated data. This produces a
-+ * &struct utrace_engine, which is the handle used for all other
-+ * operations. An attached engine has its ops vector, its data, and an
-+ * event mask controlled by utrace_set_events().
-+ *
-+ * For each event bit that is set, that engine will get the
-+ * appropriate ops->report_*() callback when the event occurs. The
-+ * &struct utrace_engine_ops need not provide callbacks for an event
-+ * unless the engine sets one of the associated event bits.
-+ */
-+
-+#ifndef _LINUX_UTRACE_H
-+#define _LINUX_UTRACE_H 1
+diff --git a/include/linux/signal.h b/include/linux/signal.h
+index 7987ce7..c320549 100644
+--- a/include/linux/signal.h
++++ b/include/linux/signal.h
+@@ -239,6 +239,8 @@ static inline int valid_signal(unsigned long sig)
+ struct timespec;
+ struct pt_regs;
+
++extern int wake_up_quiescent(struct task_struct *p, unsigned int state);
+
-+#include <linux/list.h>
-+#include <linux/kref.h>
-+#include <linux/signal.h>
-+#include <linux/sched.h>
+ extern int next_signal(struct sigpending *pending, sigset_t *mask);
+ extern int do_send_sig_info(int sig, struct siginfo *info,
+ struct task_struct *p, bool group);
+diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h
+index a71a292..a1bac95 100644
+--- a/include/linux/tracehook.h
++++ b/include/linux/tracehook.h
+@@ -49,6 +49,7 @@
+ #include <linux/sched.h>
+ #include <linux/ptrace.h>
+ #include <linux/security.h>
++#include <linux/utrace.h>
+ struct linux_binprm;
+
+ /*
+@@ -58,8 +59,12 @@ static inline void ptrace_report_syscall(struct pt_regs *regs)
+ {
+ int ptrace = current->ptrace;
+
+- if (!(ptrace & PT_PTRACED))
+- return;
++ if (!(ptrace & PT_SYSCALL_TRACE)) {
++#ifdef TIF_SYSCALL_EMU
++ if (!test_thread_flag(TIF_SYSCALL_EMU))
++#endif
++ return;
++ }
+
+ ptrace_notify(SIGTRAP | ((ptrace & PT_TRACESYSGOOD) ? 0x80 : 0));
+
+@@ -96,10 +101,16 @@ static inline void ptrace_report_syscall(struct pt_regs *regs)
+ static inline __must_check int tracehook_report_syscall_entry(
+ struct pt_regs *regs)
+ {
++ if ((task_utrace_flags(current) & UTRACE_EVENT(SYSCALL_ENTRY)) &&
++ utrace_report_syscall_entry(regs))
++ return 1;
+ ptrace_report_syscall(regs);
+ return 0;
+ }
+
++#define ptrace_wants_step(task) \
++ ((task)->ptrace & (PT_SINGLE_STEP | PT_SINGLE_BLOCK))
+
-+struct linux_binprm;
-+struct pt_regs;
-+struct utrace;
-+struct user_regset;
-+struct user_regset_view;
+ /**
+ * tracehook_report_syscall_exit - task has just finished a system call
+ * @regs: user register state of current task
+@@ -119,7 +130,10 @@ static inline __must_check int tracehook_report_syscall_entry(
+ */
+ static inline void tracehook_report_syscall_exit(struct pt_regs *regs, int step)
+ {
+- if (step) {
++ if (task_utrace_flags(current) & UTRACE_EVENT(SYSCALL_EXIT))
++ utrace_report_syscall_exit(regs);
+
-+/*
-+ * Event bits passed to utrace_set_events().
-+ * These appear in &struct task_struct. at utrace_flags
-+ * and &struct utrace_engine. at flags.
-+ */
++ if (step && ptrace_wants_step(current)) {
+ siginfo_t info;
+ user_single_step_siginfo(current, regs, &info);
+ force_sig_info(SIGTRAP, &info, current);
+@@ -148,10 +162,34 @@ static inline void tracehook_signal_handler(int sig, siginfo_t *info,
+ const struct k_sigaction *ka,
+ struct pt_regs *regs, int stepping)
+ {
+- if (stepping)
++ if (task_utrace_flags(current))
++ utrace_signal_handler(current, stepping);
++ if (stepping && ptrace_wants_step(current))
+ ptrace_notify(SIGTRAP);
+ }
+
++/**
++ * tracehook_consider_fatal_signal - suppress special handling of fatal signal
++ * @task: task receiving the signal
++ * @sig: signal number being sent
++ *
++ * Return nonzero to prevent special handling of this termination signal.
++ * Normally handler for signal is %SIG_DFL. It can be %SIG_IGN if @sig is
++ * ignored, in which case force_sig() is about to reset it to %SIG_DFL.
++ * When this returns zero, this signal might cause a quick termination
++ * that does not give the debugger a chance to intercept the signal.
++ *
++ * Called with or without @task->sighand->siglock held.
++ */
++static inline int tracehook_consider_fatal_signal(struct task_struct *task,
++ int sig)
++{
++ if (unlikely(task_utrace_flags(task) & (UTRACE_EVENT(SIGNAL_TERM) |
++ UTRACE_EVENT(SIGNAL_CORE))))
++ return 1;
++ return task->ptrace != 0;
++}
++
+ #ifdef TIF_NOTIFY_RESUME
+ /**
+ * set_notify_resume - cause tracehook_notify_resume() to be called
+@@ -179,10 +217,21 @@ static inline void set_notify_resume(struct task_struct *task)
+ * asynchronously, this will be called again before we return to
+ * user mode.
+ *
+- * Called without locks.
++ * Called without locks. However, on some machines this may be
++ * called with interrupts disabled.
+ */
+ static inline void tracehook_notify_resume(struct pt_regs *regs)
+ {
++ struct task_struct *task = current;
++ /*
++ * Prevent the following store/load from getting ahead of the
++ * caller which clears TIF_NOTIFY_RESUME. This pairs with the
++ * implicit mb() before setting TIF_NOTIFY_RESUME in
++ * set_notify_resume().
++ */
++ smp_mb();
++ if (task_utrace_flags(task))
++ utrace_resume(task, regs);
+ }
+ #endif /* TIF_NOTIFY_RESUME */
+
+diff --git a/include/linux/utrace.h b/include/linux/utrace.h
+new file mode 100644
+index 0000000..f37373b
+--- /dev/null
++++ b/include/linux/utrace.h
+@@ -0,0 +1,773 @@
++/*
++ * utrace infrastructure interface for debugging user processes
++ *
++ * Copyright (C) 2006-2009 Red Hat, Inc. All rights reserved.
++ *
++ * This copyrighted material is made available to anyone wishing to use,
++ * modify, copy, or redistribute it subject to the terms and conditions
++ * of the GNU General Public License v.2.
++ *
++ * Red Hat Author: Roland McGrath.
++ *
++ * This interface allows for notification of interesting events in a
++ * thread. It also mediates access to thread state such as registers.
++ * Multiple unrelated users can be associated with a single thread.
++ * We call each of these a tracing engine.
++ *
++ * A tracing engine starts by calling utrace_attach_task() or
++ * utrace_attach_pid() on the chosen thread, passing in a set of hooks
++ * (&struct utrace_engine_ops), and some associated data. This produces a
++ * &struct utrace_engine, which is the handle used for all other
++ * operations. An attached engine has its ops vector, its data, and an
++ * event mask controlled by utrace_set_events().
++ *
++ * For each event bit that is set, that engine will get the
++ * appropriate ops->report_*() callback when the event occurs. The
++ * &struct utrace_engine_ops need not provide callbacks for an event
++ * unless the engine sets one of the associated event bits.
++ */
++
++#ifndef _LINUX_UTRACE_H
++#define _LINUX_UTRACE_H 1
++
++#include <linux/list.h>
++#include <linux/kref.h>
++#include <linux/signal.h>
++#include <linux/sched.h>
++
++struct linux_binprm;
++struct pt_regs;
++struct utrace;
++struct user_regset;
++struct user_regset_view;
++
++/*
++ * Event bits passed to utrace_set_events().
++ * These appear in &struct task_struct. at utrace_flags
++ * and &struct utrace_engine. at flags.
++ */
+enum utrace_events {
+ _UTRACE_EVENT_QUIESCE, /* Thread is available for examination. */
+ _UTRACE_EVENT_REAP, /* Zombie reaped, no more tracing possible. */
@@ -832,7 +1104,7 @@ index 0000000..f251efe
+void utrace_report_clone(unsigned long, struct task_struct *);
+void utrace_finish_vfork(struct task_struct *);
+void utrace_report_exit(long *exit_code);
-+void utrace_report_death(struct task_struct *, struct utrace *, bool, int);
++void utrace_report_death(struct task_struct *, bool, int);
+void utrace_report_jctl(int notify, int type);
+void utrace_report_exec(struct linux_binfmt *, struct linux_binprm *,
+ struct pt_regs *regs);
@@ -840,8 +1112,22 @@ index 0000000..f251efe
+void utrace_report_syscall_exit(struct pt_regs *);
+void utrace_signal_handler(struct task_struct *, int);
+
++#define UTRACE_FLAG(task, ev) (task_utrace_flags(task) & UTRACE_EVENT(ev))
++
++#define UTRACE_HOOK(task, ev, callback) \
++ do { \
++ if (UTRACE_FLAG(task, ev)) \
++ utrace_ ## callback; \
++ } while (0)
++
+#ifndef CONFIG_UTRACE
+
++static inline void task_utrace_lock(struct task_struct *task)
++{
++}
++static inline void task_utrace_unlock(struct task_struct *task)
++{
++}
+/*
+ * <linux/tracehook.h> uses these accessors to avoid #ifdef CONFIG_UTRACE.
+ */
@@ -864,6 +1150,9 @@ index 0000000..f251efe
+
+#else /* CONFIG_UTRACE */
+
++extern void task_utrace_lock(struct task_struct *task);
++extern void task_utrace_unlock(struct task_struct *task);
++
+static inline unsigned long task_utrace_flags(struct task_struct *task)
+{
+ return task->utrace_flags;
@@ -1422,14 +1711,78 @@ index 0000000..f251efe
+
+#endif /* CONFIG_UTRACE */
+
++static inline void utrace_release_task(struct task_struct *task)
++{
++ /* see utrace_add_engine() about this barrier */
++ smp_mb();
++ if (task_utrace_flags(task))
++ utrace_maybe_reap(task, task_utrace_struct(task), true);
++}
++
++static inline void utrace_exit_notify(struct task_struct *task,
++ int signal, int group_dead)
++{
++ /*
++ * If utrace_set_events() was just called to enable
++ * UTRACE_EVENT(DEATH), then we are obliged to call
++ * utrace_report_death() and not miss it. utrace_set_events()
++ * checks @task->exit_state under tasklist_lock to synchronize
++ * with exit_notify(), the caller.
++ */
++ if (task_utrace_flags(task) & _UTRACE_DEATH_EVENTS)
++ utrace_report_death(task, group_dead, signal);
++}
++
++/**
++ * utrace_end_stop - report about return from STOPPED/TRACED
++ *
++ * This is called by do_signal_stop() and ptrace_stop after wakeup.
++ */
++static inline void utrace_end_stop(void)
++{
++ if (task_utrace_flags(current))
++ utrace_finish_stop();
++}
++
++/**
++ * utrace_hook_signal - deliver synthetic signal to traced task
++ * @task: @current
++ * @regs: task_pt_regs(@current)
++ * @info: details of synthetic signal
++ * @return_ka: sigaction for synthetic signal
++ *
++ * Return zero to check for a real pending signal normally.
++ * Return -1 after releasing the siglock to repeat the check.
++ * Return a signal number to induce an artificial signal delivery,
++ * setting *@info and *@return_ka to specify its details and behavior.
++ *
++ * The @return_ka->sa_handler value controls the disposition of the
++ * signal, no matter the signal number. For %SIG_DFL, the return value
++ * is a representative signal to indicate the behavior (e.g. %SIGTERM
++ * for death, %SIGQUIT for core dump, %SIGSTOP for job control stop,
++ * %SIGTSTP for stop unless in an orphaned pgrp), but the signal number
++ * reported will be @info->si_signo instead.
++ *
++ * Called with @task->sighand->siglock held, before dequeuing pending signals.
++ */
++static inline int utrace_hook_signal(struct task_struct *task,
++ struct pt_regs *regs,
++ siginfo_t *info,
++ struct k_sigaction *return_ka)
++{
++ if (unlikely(task_utrace_flags(task)))
++ return utrace_get_signal(task, regs, info, return_ka);
++ return 0;
++}
++
+#endif /* linux/utrace.h */
diff --git a/init/Kconfig b/init/Kconfig
-index 43298f9..3f670e5 100644
+index 3f42cd6..5f4f92e 100644
--- a/init/Kconfig
+++ b/init/Kconfig
-@@ -372,6 +372,15 @@ config AUDIT_TREE
- depends on AUDITSYSCALL
- select FSNOTIFY
+@@ -386,6 +386,15 @@ config AUDIT_LOGINUID_IMMUTABLE
+ one to drop potentially dangerous capabilites from the login tasks,
+ but may not be backwards compatible with older init systems.
+config UTRACE
+ bool "Infrastructure for tracing and debugging user processes"
@@ -1444,10 +1797,10 @@ index 43298f9..3f670e5 100644
menu "RCU Subsystem"
diff --git a/kernel/Makefile b/kernel/Makefile
-index e898c5b..e43bbfb 100644
+index 2d9de86..6c6749d 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
-@@ -68,6 +68,7 @@ obj-$(CONFIG_IKCONFIG) += configs.o
+@@ -67,6 +67,7 @@ obj-$(CONFIG_IKCONFIG) += configs.o
obj-$(CONFIG_RESOURCE_COUNTERS) += res_counter.o
obj-$(CONFIG_SMP) += stop_machine.o
obj-$(CONFIG_KPROBES_SANITY_TEST) += test_kprobes.o
@@ -1455,5728 +1808,2878 @@ index e898c5b..e43bbfb 100644
obj-$(CONFIG_AUDIT) += audit.o auditfilter.o
obj-$(CONFIG_AUDITSYSCALL) += auditsc.o
obj-$(CONFIG_AUDIT_WATCH) += audit_watch.o
-diff --git a/kernel/utrace.c b/kernel/utrace.c
-new file mode 100644
-index 0000000..ef856c9
---- /dev/null
-+++ b/kernel/utrace.c
-@@ -0,0 +1,2440 @@
-+/*
-+ * utrace infrastructure interface for debugging user processes
-+ *
-+ * Copyright (C) 2006-2010 Red Hat, Inc. All rights reserved.
-+ *
-+ * This copyrighted material is made available to anyone wishing to use,
-+ * modify, copy, or redistribute it subject to the terms and conditions
-+ * of the GNU General Public License v.2.
-+ *
-+ * Red Hat Author: Roland McGrath.
-+ */
-+
-+#include <linux/utrace.h>
-+#include <linux/tracehook.h>
-+#include <linux/regset.h>
-+#include <asm/syscall.h>
-+#include <linux/ptrace.h>
-+#include <linux/err.h>
-+#include <linux/sched.h>
-+#include <linux/freezer.h>
-+#include <linux/module.h>
-+#include <linux/init.h>
-+#include <linux/slab.h>
-+#include <linux/seq_file.h>
+diff --git a/kernel/exit.c b/kernel/exit.c
+index 294b170..16108a5 100644
+--- a/kernel/exit.c
++++ b/kernel/exit.c
+@@ -169,6 +169,8 @@ void release_task(struct task_struct * p)
+ struct task_struct *leader;
+ int zap_leader;
+ repeat:
++ utrace_release_task(p);
+
+ /* don't need to get the RCU readlock here - the process is dead and
+ * can't be modifying its own credentials. But shut RCU-lockdep up */
+ rcu_read_lock();
+@@ -857,6 +859,8 @@ static void exit_notify(struct task_struct *tsk, int group_dead)
+ wake_up_process(tsk->signal->group_exit_task);
+ write_unlock_irq(&tasklist_lock);
+
++ utrace_exit_notify(tsk, autoreap ? -1 : SIGCHLD, group_dead);
+
-+/*
-+ * Per-thread structure private to utrace implementation.
-+ * If task_struct.utrace_flags is nonzero, task_struct.utrace
-+ * has always been allocated first. Once allocated, it is
-+ * never freed until free_task().
-+ *
-+ * The common event reporting loops are done by the task making the
-+ * report without ever taking any locks. To facilitate this, the two
-+ * lists @attached and @attaching work together for smooth asynchronous
-+ * attaching with low overhead. Modifying either list requires @lock.
-+ * The @attaching list can be modified any time while holding @lock.
-+ * New engines being attached always go on this list.
-+ *
-+ * The @attached list is what the task itself uses for its reporting
-+ * loops. When the task itself is not quiescent, it can use the
-+ * @attached list without taking any lock. Nobody may modify the list
-+ * when the task is not quiescent. When it is quiescent, that means
-+ * that it won't run again without taking @lock itself before using
-+ * the list.
-+ *
-+ * At each place where we know the task is quiescent (or it's current),
-+ * while holding @lock, we call splice_attaching(), below. This moves
-+ * the @attaching list members on to the end of the @attached list.
-+ * Since this happens at the start of any reporting pass, any new
-+ * engines attached asynchronously go on the stable @attached list
-+ * in time to have their callbacks seen.
-+ */
-+struct utrace {
-+ spinlock_t lock;
-+ struct list_head attached, attaching;
-+
-+ struct task_struct *cloning;
-+
-+ struct utrace_engine *reporting;
-+
-+ enum utrace_resume_action resume:UTRACE_RESUME_BITS;
-+ unsigned int signal_handler:1;
-+ unsigned int vfork_stop:1; /* need utrace_stop() before vfork wait */
-+ unsigned int death:1; /* in utrace_report_death() now */
-+ unsigned int reap:1; /* release_task() has run */
-+ unsigned int pending_attach:1; /* need splice_attaching() */
-+};
-+
-+static struct kmem_cache *utrace_cachep;
-+static struct kmem_cache *utrace_engine_cachep;
-+static const struct utrace_engine_ops utrace_detached_ops; /* forward decl */
+ /* If the process is dead, release it - nobody will wait for it */
+ if (autoreap)
+ release_task(tsk);
+@@ -910,6 +914,7 @@ void do_exit(long code)
+ */
+ set_fs(USER_DS);
+
++ UTRACE_HOOK(current, EXIT, report_exit(&code));
+ ptrace_event(PTRACE_EVENT_EXIT, code);
+
+ validate_creds_for_do_exit(tsk);
+diff --git a/kernel/fork.c b/kernel/fork.c
+index 051f090..e103101 100644
+--- a/kernel/fork.c
++++ b/kernel/fork.c
+@@ -66,6 +66,7 @@
+ #include <linux/user-return-notifier.h>
+ #include <linux/oom.h>
+ #include <linux/khugepaged.h>
++#include <linux/utrace.h>
+
+ #include <asm/pgtable.h>
+ #include <asm/pgalloc.h>
+@@ -169,6 +170,8 @@ void free_task(struct task_struct *tsk)
+ free_thread_info(tsk->stack);
+ rt_mutex_debug_task_free(tsk);
+ ftrace_graph_exit_task(tsk);
++ if (task_utrace_struct(tsk))
++ utrace_free_task(tsk);
+ free_task_struct(tsk);
+ }
+ EXPORT_SYMBOL(free_task);
+@@ -1092,6 +1095,8 @@ static struct task_struct *copy_process(unsigned long clone_flags,
+ if (!p)
+ goto fork_out;
+
++ utrace_init_task(p);
+
-+static int __init utrace_init(void)
-+{
-+ utrace_cachep = KMEM_CACHE(utrace, SLAB_PANIC);
-+ utrace_engine_cachep = KMEM_CACHE(utrace_engine, SLAB_PANIC);
-+ return 0;
-+}
-+module_init(utrace_init);
+ ftrace_graph_init_task(p);
+
+ rt_mutex_init_task(p);
+@@ -1527,6 +1532,8 @@ long do_fork(unsigned long clone_flags,
+ init_completion(&vfork);
+ }
+
++ UTRACE_HOOK(current, CLONE, report_clone(clone_flags, p));
+
-+/*
-+ * Set up @task.utrace for the first time. We can have races
-+ * between two utrace_attach_task() calls here. The task_lock()
-+ * governs installing the new pointer. If another one got in first,
-+ * we just punt the new one we allocated.
-+ *
-+ * This returns false only in case of a memory allocation failure.
-+ */
-+static bool utrace_task_alloc(struct task_struct *task)
+ /*
+ * We set PF_STARTING at creation in case tracing wants to
+ * use this to distinguish a fully live task from one that
+@@ -1538,6 +1545,8 @@ long do_fork(unsigned long clone_flags,
+ wake_up_new_task(p);
+
+ /* forking complete and child started to run, tell ptracer */
++ if (clone_flags & CLONE_VFORK)
++ UTRACE_HOOK(current, CLONE, finish_vfork(current));
+ if (unlikely(trace))
+ ptrace_event(trace, nr);
+
+diff --git a/kernel/ptrace.c b/kernel/ptrace.c
+index 00ab2ca..a7024b8 100644
+--- a/kernel/ptrace.c
++++ b/kernel/ptrace.c
+@@ -24,7 +24,34 @@
+ #include <linux/regset.h>
+ #include <linux/hw_breakpoint.h>
+ #include <linux/cn_proc.h>
++#include <linux/utrace.h>
+
++void ptrace_signal_wake_up(struct task_struct *p, int quiescent)
+{
-+ struct utrace *utrace = kmem_cache_zalloc(utrace_cachep, GFP_KERNEL);
-+ if (unlikely(!utrace))
-+ return false;
-+ spin_lock_init(&utrace->lock);
-+ INIT_LIST_HEAD(&utrace->attached);
-+ INIT_LIST_HEAD(&utrace->attaching);
-+ utrace->resume = UTRACE_RESUME;
-+ task_lock(task);
-+ if (likely(!task->utrace)) {
-+ /*
-+ * This barrier makes sure the initialization of the struct
-+ * precedes the installation of the pointer. This pairs
-+ * with smp_read_barrier_depends() in task_utrace_struct().
-+ */
-+ smp_wmb();
-+ task->utrace = utrace;
-+ }
-+ task_unlock(task);
-+
-+ if (unlikely(task->utrace != utrace))
-+ kmem_cache_free(utrace_cachep, utrace);
-+ return true;
-+}
++ unsigned int state;
+
-+/*
-+ * This is called via tracehook_free_task() from free_task()
-+ * when @task is being deallocated.
-+ */
-+void utrace_free_task(struct task_struct *task)
-+{
-+ kmem_cache_free(utrace_cachep, task->utrace);
-+}
++ set_tsk_thread_flag(p, TIF_SIGPENDING);
+
-+/*
-+ * This is calledwhen the task is safely quiescent, i.e. it won't consult
-+ * utrace->attached without the lock. Move any engines attached
-+ * asynchronously from @utrace->attaching onto the @utrace->attached list.
-+ */
-+static void splice_attaching(struct utrace *utrace)
-+{
-+ lockdep_assert_held(&utrace->lock);
-+ list_splice_tail_init(&utrace->attaching, &utrace->attached);
-+ utrace->pending_attach = 0;
++ state = TASK_INTERRUPTIBLE;
++ if (quiescent)
++ state |= (__TASK_STOPPED | __TASK_TRACED);
++ if (!wake_up_quiescent(p, state))
++ kick_process(p);
+}
+
-+/*
-+ * This is the exported function used by the utrace_engine_put() inline.
-+ */
-+void __utrace_engine_release(struct kref *kref)
++static void ptrace_set_syscall_trace(struct task_struct *p, bool on)
+{
-+ struct utrace_engine *engine = container_of(kref, struct utrace_engine,
-+ kref);
-+ BUG_ON(!list_empty(&engine->entry));
-+ if (engine->release)
-+ (*engine->release)(engine->data);
-+ kmem_cache_free(utrace_engine_cachep, engine);
++ task_utrace_lock(p);
++ if (on) {
++ p->ptrace |= PT_SYSCALL_TRACE;
++ set_tsk_thread_flag(p, TIF_SYSCALL_TRACE);
++ } else {
++ p->ptrace &= ~PT_SYSCALL_TRACE;
++ if (!(task_utrace_flags(p) & UTRACE_EVENT_SYSCALL))
++ clear_tsk_thread_flag(p, TIF_SYSCALL_TRACE);
++ }
++ task_utrace_unlock(p);
+}
-+EXPORT_SYMBOL_GPL(__utrace_engine_release);
+
+ static int ptrace_trapping_sleep_fn(void *flags)
+ {
+@@ -117,7 +144,7 @@ void __ptrace_unlink(struct task_struct *child)
+ * TASK_KILLABLE sleeps.
+ */
+ if (child->jobctl & JOBCTL_STOP_PENDING || task_is_traced(child))
+- signal_wake_up(child, task_is_traced(child));
++ ptrace_signal_wake_up(child, task_is_traced(child));
+
+ spin_unlock(&child->sighand->siglock);
+ }
+@@ -315,7 +342,7 @@ static int ptrace_attach(struct task_struct *task, long request,
+ */
+ if (task_is_stopped(task) &&
+ task_set_jobctl_pending(task, JOBCTL_TRAP_STOP | JOBCTL_TRAPPING))
+- signal_wake_up(task, 1);
++ ptrace_signal_wake_up(task, 1);
+
+ spin_unlock(&task->sighand->siglock);
+
+@@ -425,7 +452,7 @@ static int ptrace_detach(struct task_struct *child, unsigned int data)
+
+ /* Architecture-specific hardware disable .. */
+ ptrace_disable(child);
+- clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
++ ptrace_set_syscall_trace(child, false);
+
+ write_lock_irq(&tasklist_lock);
+ /*
+@@ -608,13 +635,12 @@ static int ptrace_setsiginfo(struct task_struct *child, const siginfo_t *info)
+ static int ptrace_resume(struct task_struct *child, long request,
+ unsigned long data)
+ {
++ unsigned long flags;
+
-+static bool engine_matches(struct utrace_engine *engine, int flags,
-+ const struct utrace_engine_ops *ops, void *data)
-+{
-+ if ((flags & UTRACE_ATTACH_MATCH_OPS) && engine->ops != ops)
-+ return false;
-+ if ((flags & UTRACE_ATTACH_MATCH_DATA) && engine->data != data)
-+ return false;
-+ return engine->ops && engine->ops != &utrace_detached_ops;
-+}
+ if (!valid_signal(data))
+ return -EIO;
+
+- if (request == PTRACE_SYSCALL)
+- set_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
+- else
+- clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
++ ptrace_set_syscall_trace(child, request == PTRACE_SYSCALL);
+
+ #ifdef TIF_SYSCALL_EMU
+ if (request == PTRACE_SYSEMU || request == PTRACE_SYSEMU_SINGLESTEP)
+@@ -623,20 +649,23 @@ static int ptrace_resume(struct task_struct *child, long request,
+ clear_tsk_thread_flag(child, TIF_SYSCALL_EMU);
+ #endif
+
++ child->ptrace &= ~(PT_SINGLE_STEP | PT_SINGLE_BLOCK);
+ if (is_singleblock(request)) {
+ if (unlikely(!arch_has_block_step()))
+ return -EIO;
+- user_enable_block_step(child);
++ child->ptrace |= PT_SINGLE_BLOCK;
+ } else if (is_singlestep(request) || is_sysemu_singlestep(request)) {
+ if (unlikely(!arch_has_single_step()))
+ return -EIO;
+- user_enable_single_step(child);
+- } else {
+- user_disable_single_step(child);
++ child->ptrace |= PT_SINGLE_STEP;
+ }
+
+ child->exit_code = data;
+- wake_up_state(child, __TASK_TRACED);
+
-+static struct utrace_engine *find_matching_engine(
-+ struct utrace *utrace, int flags,
-+ const struct utrace_engine_ops *ops, void *data)
-+{
-+ struct utrace_engine *engine;
-+ list_for_each_entry(engine, &utrace->attached, entry)
-+ if (engine_matches(engine, flags, ops, data))
-+ return engine;
-+ list_for_each_entry(engine, &utrace->attaching, entry)
-+ if (engine_matches(engine, flags, ops, data))
-+ return engine;
-+ return NULL;
-+}
++ if (lock_task_sighand(child, &flags)) {
++ wake_up_quiescent(child, __TASK_TRACED);
++ unlock_task_sighand(child, &flags);
++ }
+
+ return 0;
+ }
+@@ -744,7 +773,7 @@ int ptrace_request(struct task_struct *child, long request,
+ * tracee into STOP.
+ */
+ if (likely(task_set_jobctl_pending(child, JOBCTL_TRAP_STOP)))
+- signal_wake_up(child, child->jobctl & JOBCTL_LISTENING);
++ ptrace_signal_wake_up(child, child->jobctl & JOBCTL_LISTENING);
+
+ unlock_task_sighand(child, &flags);
+ ret = 0;
+@@ -770,7 +799,7 @@ int ptrace_request(struct task_struct *child, long request,
+ * start of this trap and now. Trigger re-trap.
+ */
+ if (child->jobctl & JOBCTL_TRAP_NOTIFY)
+- signal_wake_up(child, true);
++ ptrace_signal_wake_up(child, true);
+ ret = 0;
+ }
+ unlock_task_sighand(child, &flags);
+diff --git a/kernel/sched/core.c b/kernel/sched/core.c
+index df00cb0..24dfee4 100644
+--- a/kernel/sched/core.c
++++ b/kernel/sched/core.c
+@@ -1172,7 +1172,7 @@ unsigned long wait_task_inactive(struct task_struct *p, long match_state)
+ * is actually now running somewhere else!
+ */
+ while (task_running(rq, p)) {
+- if (match_state && unlikely(p->state != match_state))
++ if (match_state && !likely(p->state & match_state))
+ return 0;
+ cpu_relax();
+ }
+diff --git a/kernel/signal.c b/kernel/signal.c
+index c73c428..0508d93 100644
+--- a/kernel/signal.c
++++ b/kernel/signal.c
+@@ -88,7 +88,7 @@ static int sig_ignored(struct task_struct *t, int sig, int from_ancestor_ns)
+ /*
+ * Tracers may want to know about even ignored signals.
+ */
+- return !t->ptrace;
++ return !t->ptrace && !UTRACE_FLAG(t, SIGNAL_IGN);
+ }
+
+ /*
+@@ -151,6 +151,11 @@ void recalc_sigpending_and_wake(struct task_struct *t)
+
+ void recalc_sigpending(void)
+ {
++ if (task_utrace_flags(current) && utrace_interrupt_pending()) {
++ set_thread_flag(TIF_SIGPENDING);
++ return;
++ }
+
+ if (!recalc_sigpending_tsk(current) && !freezing(current))
+ clear_thread_flag(TIF_SIGPENDING);
+
+@@ -495,7 +500,7 @@ int unhandled_signal(struct task_struct *tsk, int sig)
+ if (handler != SIG_IGN && handler != SIG_DFL)
+ return 0;
+ /* if ptraced, let the tracer determine */
+- return !tsk->ptrace;
++ return !tracehook_consider_fatal_signal(tsk, sig);
+ }
+
+ /*
+@@ -697,6 +702,29 @@ void signal_wake_up(struct task_struct *t, int resume)
+ kick_process(t);
+ }
+
++#define STATE_QUIESCENT (__TASK_STOPPED | __TASK_TRACED | __TASK_UTRACED)
+/*
-+ * Enqueue @engine, or maybe don't if UTRACE_ATTACH_EXCLUSIVE.
++ * wakes up the STOPPED/TRACED task, must be called with ->siglock held.
+ */
-+static int utrace_add_engine(struct task_struct *target,
-+ struct utrace *utrace,
-+ struct utrace_engine *engine,
-+ int flags,
-+ const struct utrace_engine_ops *ops,
-+ void *data)
++int wake_up_quiescent(struct task_struct *p, unsigned int state)
+{
-+ int ret;
-+
-+ spin_lock(&utrace->lock);
++ unsigned int quiescent = (p->state & STATE_QUIESCENT);
+
-+ ret = -EEXIST;
-+ if ((flags & UTRACE_ATTACH_EXCLUSIVE) &&
-+ unlikely(find_matching_engine(utrace, flags, ops, data)))
-+ goto unlock;
++ WARN_ON(state & ~(STATE_QUIESCENT | TASK_INTERRUPTIBLE));
+
-+ /*
-+ * In case we had no engines before, make sure that
-+ * utrace_flags is not zero. Since we did unlock+lock
-+ * at least once after utrace_task_alloc() installed
-+ * ->utrace, we have the necessary barrier which pairs
-+ * with rmb() in task_utrace_struct().
-+ */
-+ ret = -ESRCH;
-+ if (!target->utrace_flags) {
-+ target->utrace_flags = UTRACE_EVENT(REAP);
-+ /*
-+ * If we race with tracehook_prepare_release_task()
-+ * make sure that either it sees utrace_flags != 0
-+ * or we see exit_state == EXIT_DEAD.
-+ */
-+ smp_mb();
-+ if (unlikely(target->exit_state == EXIT_DEAD)) {
-+ target->utrace_flags = 0;
-+ goto unlock;
++ if (quiescent) {
++ state &= ~TASK_INTERRUPTIBLE;
++ if ((quiescent & ~state) != 0) {
++ p->state &= ~state;
++ WARN_ON(!(p->state & STATE_QUIESCENT));
++ WARN_ON(!(p->state & TASK_WAKEKILL));
++ return 0;
+ }
+ }
+
-+ /*
-+ * Put the new engine on the pending ->attaching list.
-+ * Make sure it gets onto the ->attached list by the next
-+ * time it's examined. Setting ->pending_attach ensures
-+ * that start_report() takes the lock and splices the lists
-+ * before the next new reporting pass.
-+ *
-+ * When target == current, it would be safe just to call
-+ * splice_attaching() right here. But if we're inside a
-+ * callback, that would mean the new engine also gets
-+ * notified about the event that precipitated its own
-+ * creation. This is not what the user wants.
-+ */
-+ list_add_tail(&engine->entry, &utrace->attaching);
-+ utrace->pending_attach = 1;
-+ utrace_engine_get(engine);
-+ ret = 0;
-+unlock:
-+ spin_unlock(&utrace->lock);
-+
-+ return ret;
++ return wake_up_state(p, state);
+}
+
-+/**
-+ * utrace_attach_task - attach new engine, or look up an attached engine
-+ * @target: thread to attach to
-+ * @flags: flag bits combined with OR, see below
-+ * @ops: callback table for new engine
-+ * @data: engine private data pointer
-+ *
-+ * The caller must ensure that the @target thread does not get freed,
-+ * i.e. hold a ref or be its parent. It is always safe to call this
-+ * on @current, or on the @child pointer in a @report_clone callback.
-+ * For most other cases, it's easier to use utrace_attach_pid() instead.
-+ *
-+ * UTRACE_ATTACH_CREATE:
-+ * Create a new engine. If %UTRACE_ATTACH_CREATE is not specified, you
-+ * only look up an existing engine already attached to the thread.
-+ *
-+ * UTRACE_ATTACH_EXCLUSIVE:
-+ * Attempting to attach a second (matching) engine fails with -%EEXIST.
-+ *
-+ * UTRACE_ATTACH_MATCH_OPS: Only consider engines matching @ops.
-+ * UTRACE_ATTACH_MATCH_DATA: Only consider engines matching @data.
-+ *
-+ * Calls with neither %UTRACE_ATTACH_MATCH_OPS nor %UTRACE_ATTACH_MATCH_DATA
-+ * match the first among any engines attached to @target. That means that
-+ * %UTRACE_ATTACH_EXCLUSIVE in such a call fails with -%EEXIST if there
-+ * are any engines on @target at all.
-+ */
-+struct utrace_engine *utrace_attach_task(
-+ struct task_struct *target, int flags,
-+ const struct utrace_engine_ops *ops, void *data)
-+{
-+ struct utrace *utrace = task_utrace_struct(target);
-+ struct utrace_engine *engine;
-+ int ret;
-+
-+ if (!(flags & UTRACE_ATTACH_CREATE)) {
-+ if (unlikely(!utrace))
-+ return ERR_PTR(-ENOENT);
-+ spin_lock(&utrace->lock);
-+ engine = find_matching_engine(utrace, flags, ops, data);
-+ if (engine)
-+ utrace_engine_get(engine);
-+ spin_unlock(&utrace->lock);
-+ return engine ?: ERR_PTR(-ENOENT);
-+ }
+ /*
+ * Remove signals in mask from the pending set and queue.
+ * Returns 1 if any signals were found.
+@@ -842,7 +870,7 @@ static void ptrace_trap_notify(struct task_struct *t)
+ assert_spin_locked(&t->sighand->siglock);
+
+ task_set_jobctl_pending(t, JOBCTL_TRAP_NOTIFY);
+- signal_wake_up(t, t->jobctl & JOBCTL_LISTENING);
++ ptrace_signal_wake_up(t, t->jobctl & JOBCTL_LISTENING);
+ }
+
+ /*
+@@ -884,7 +912,7 @@ static int prepare_signal(int sig, struct task_struct *p, int from_ancestor_ns)
+ task_clear_jobctl_pending(t, JOBCTL_STOP_PENDING);
+ rm_from_queue(SIG_KERNEL_STOP_MASK, &t->pending);
+ if (likely(!(t->ptrace & PT_SEIZED)))
+- wake_up_state(t, __TASK_STOPPED);
++ wake_up_quiescent(t, __TASK_STOPPED);
+ else
+ ptrace_trap_notify(t);
+ } while_each_thread(p, t);
+@@ -983,7 +1011,7 @@ static void complete_signal(int sig, struct task_struct *p, int group)
+ if (sig_fatal(p, sig) &&
+ !(signal->flags & (SIGNAL_UNKILLABLE | SIGNAL_GROUP_EXIT)) &&
+ !sigismember(&t->real_blocked, sig) &&
+- (sig == SIGKILL || !t->ptrace)) {
++ (sig == SIGKILL || !tracehook_consider_fatal_signal(t, sig))) {
+ /*
+ * This signal will be fatal to the whole group.
+ */
+@@ -1913,10 +1941,34 @@ static void ptrace_stop(int exit_code, int why, int clear_code, siginfo_t *info)
+ if (gstop_done)
+ do_notify_parent_cldstop(current, false, why);
+
+- __set_current_state(TASK_RUNNING);
++ spin_lock_irq(¤t->sighand->siglock);
++ wake_up_quiescent(current, __TASK_TRACED);
++ spin_unlock_irq(¤t->sighand->siglock);
+
-+ if (unlikely(!ops) || unlikely(ops == &utrace_detached_ops))
-+ return ERR_PTR(-EINVAL);
+ if (clear_code)
+ current->exit_code = 0;
+ read_unlock(&tasklist_lock);
+
-+ if (unlikely(target->flags & PF_KTHREAD))
+ /*
-+ * Silly kernel, utrace is for users!
++ * It is possible that __TASK_UTRACED was added by utrace
++ * while we were __TASK_TRACED and before we take ->siglock
++ * for wake_up_quiescent(), we need to block in this case.
++ * Otherwise this is unnecessary but absolutely harmless.
+ */
-+ return ERR_PTR(-EPERM);
-+
-+ if (!utrace) {
-+ if (unlikely(!utrace_task_alloc(target)))
-+ return ERR_PTR(-ENOMEM);
-+ utrace = task_utrace_struct(target);
++ schedule();
+ }
+
-+ engine = kmem_cache_alloc(utrace_engine_cachep, GFP_KERNEL);
-+ if (unlikely(!engine))
-+ return ERR_PTR(-ENOMEM);
++ utrace_end_stop();
+
-+ /*
-+ * Initialize the new engine structure. It starts out with one ref
-+ * to return. utrace_add_engine() adds another for being attached.
-+ */
-+ kref_init(&engine->kref);
-+ engine->flags = 0;
-+ engine->ops = ops;
-+ engine->data = data;
-+ engine->release = ops->release;
-+
-+ ret = utrace_add_engine(target, utrace, engine, flags, ops, data);
-+
-+ if (unlikely(ret)) {
-+ kmem_cache_free(utrace_engine_cachep, engine);
-+ engine = ERR_PTR(ret);
-+ }
++ if (current->ptrace & PT_SINGLE_BLOCK)
++ user_enable_block_step(current);
++ else if (current->ptrace & PT_SINGLE_STEP)
++ user_enable_single_step(current);
++ else {
++ user_disable_single_step(current);
++ /* if utrace needs the stepping it should reassert */
++ if (task_utrace_flags(current))
++ set_thread_flag(TIF_NOTIFY_RESUME);
+ }
+
+ /*
+@@ -2081,6 +2133,9 @@ static bool do_signal_stop(int signr)
+
+ /* Now we don't run again until woken by SIGCONT or SIGKILL */
+ schedule();
+
++ utrace_end_stop();
+
-+ return engine;
-+}
-+EXPORT_SYMBOL_GPL(utrace_attach_task);
+ return true;
+ } else {
+ /*
+@@ -2231,17 +2286,27 @@ relock:
+ for (;;) {
+ struct k_sigaction *ka;
+
+- if (unlikely(current->jobctl & JOBCTL_STOP_PENDING) &&
+- do_signal_stop(0))
++ signr = utrace_hook_signal(current, regs, info, return_ka);
++ if (unlikely(signr < 0))
+ goto relock;
+
+- if (unlikely(current->jobctl & JOBCTL_TRAP_MASK)) {
+- do_jobctl_trap();
+- spin_unlock_irq(&sighand->siglock);
+- goto relock;
+- }
++ if (unlikely(signr != 0))
++ ka = return_ka;
++ else {
++ if (unlikely(current->jobctl & JOBCTL_STOP_PENDING) &&
++ do_signal_stop(0))
++ goto relock;
+
-+/**
-+ * utrace_attach_pid - attach new engine, or look up an attached engine
-+ * @pid: &struct pid pointer representing thread to attach to
-+ * @flags: flag bits combined with OR, see utrace_attach_task()
-+ * @ops: callback table for new engine
-+ * @data: engine private data pointer
-+ *
-+ * This is the same as utrace_attach_task(), but takes a &struct pid
-+ * pointer rather than a &struct task_struct pointer. The caller must
-+ * hold a ref on @pid, but does not need to worry about the task
-+ * staying valid. If it's been reaped so that @pid points nowhere,
-+ * then this call returns -%ESRCH.
-+ */
-+struct utrace_engine *utrace_attach_pid(
-+ struct pid *pid, int flags,
-+ const struct utrace_engine_ops *ops, void *data)
-+{
-+ struct utrace_engine *engine = ERR_PTR(-ESRCH);
-+ struct task_struct *task = get_pid_task(pid, PIDTYPE_PID);
-+ if (task) {
-+ engine = utrace_attach_task(task, flags, ops, data);
-+ put_task_struct(task);
-+ }
-+ return engine;
-+}
-+EXPORT_SYMBOL_GPL(utrace_attach_pid);
++ if (unlikely(current->jobctl & JOBCTL_TRAP_MASK)) {
++ do_jobctl_trap();
++ spin_unlock_irq(&sighand->siglock);
++ goto relock;
++ }
+
+- signr = dequeue_signal(current, ¤t->blocked, info);
++ signr = dequeue_signal(current, ¤t->blocked, info);
+
++ ka = &sighand->action[signr-1];
++ }
+
+ if (!signr)
+ break; /* will return 0 */
+@@ -2251,9 +2316,9 @@ relock:
+ regs, cookie);
+ if (!signr)
+ continue;
+- }
+
+- ka = &sighand->action[signr-1];
++ ka = &sighand->action[signr-1];
++ }
+
+ /* Trace actually delivered signals. */
+ trace_signal_deliver(signr, info, ka);
+diff --git a/kernel/utrace.c b/kernel/utrace.c
+new file mode 100644
+index 0000000..c817a46
+--- /dev/null
++++ b/kernel/utrace.c
+@@ -0,0 +1,2462 @@
+/*
-+ * When an engine is detached, the target thread may still see it and
-+ * make callbacks until it quiesces. We install a special ops vector
-+ * with these two callbacks. When the target thread quiesces, it can
-+ * safely free the engine itself. For any event we will always get
-+ * the report_quiesce() callback first, so we only need this one
-+ * pointer to be set. The only exception is report_reap(), so we
-+ * supply that callback too.
++ * utrace infrastructure interface for debugging user processes
++ *
++ * Copyright (C) 2006-2010 Red Hat, Inc. All rights reserved.
++ *
++ * This copyrighted material is made available to anyone wishing to use,
++ * modify, copy, or redistribute it subject to the terms and conditions
++ * of the GNU General Public License v.2.
++ *
++ * Red Hat Author: Roland McGrath.
+ */
-+static u32 utrace_detached_quiesce(u32 action, struct utrace_engine *engine,
-+ unsigned long event)
-+{
-+ return UTRACE_DETACH;
-+}
+
-+static void utrace_detached_reap(struct utrace_engine *engine,
-+ struct task_struct *task)
-+{
-+}
++#include <linux/utrace.h>
++#include <linux/tracehook.h>
++#include <linux/regset.h>
++#include <asm/syscall.h>
++#include <linux/ptrace.h>
++#include <linux/err.h>
++#include <linux/sched.h>
++#include <linux/freezer.h>
++#include <linux/module.h>
++#include <linux/init.h>
++#include <linux/slab.h>
++#include <linux/seq_file.h>
+
-+static const struct utrace_engine_ops utrace_detached_ops = {
-+ .report_quiesce = &utrace_detached_quiesce,
-+ .report_reap = &utrace_detached_reap
-+};
+
+/*
-+ * The caller has to hold a ref on the engine. If the attached flag is
-+ * true (all but utrace_barrier() calls), the engine is supposed to be
-+ * attached. If the attached flag is false (utrace_barrier() only),
-+ * then return -ERESTARTSYS for an engine marked for detach but not yet
-+ * fully detached. The task pointer can be invalid if the engine is
-+ * detached.
++ * Per-thread structure private to utrace implementation.
++ * If task_struct.utrace_flags is nonzero, task_struct.utrace
++ * has always been allocated first. Once allocated, it is
++ * never freed until free_task().
+ *
-+ * Get the utrace lock for the target task.
-+ * Returns the struct if locked, or ERR_PTR(-errno).
++ * The common event reporting loops are done by the task making the
++ * report without ever taking any locks. To facilitate this, the two
++ * lists @attached and @attaching work together for smooth asynchronous
++ * attaching with low overhead. Modifying either list requires @lock.
++ * The @attaching list can be modified any time while holding @lock.
++ * New engines being attached always go on this list.
+ *
-+ * This has to be robust against races with:
-+ * utrace_control(target, UTRACE_DETACH) calls
-+ * UTRACE_DETACH after reports
-+ * utrace_report_death
-+ * utrace_release_task
++ * The @attached list is what the task itself uses for its reporting
++ * loops. When the task itself is not quiescent, it can use the
++ * @attached list without taking any lock. Nobody may modify the list
++ * when the task is not quiescent. When it is quiescent, that means
++ * that it won't run again without taking @lock itself before using
++ * the list.
++ *
++ * At each place where we know the task is quiescent (or it's current),
++ * while holding @lock, we call splice_attaching(), below. This moves
++ * the @attaching list members on to the end of the @attached list.
++ * Since this happens at the start of any reporting pass, any new
++ * engines attached asynchronously go on the stable @attached list
++ * in time to have their callbacks seen.
+ */
-+static struct utrace *get_utrace_lock(struct task_struct *target,
-+ struct utrace_engine *engine,
-+ bool attached)
-+ __acquires(utrace->lock)
++struct utrace {
++ spinlock_t lock;
++ struct list_head attached, attaching;
++
++ struct task_struct *cloning;
++
++ struct utrace_engine *reporting;
++
++ enum utrace_resume_action resume:UTRACE_RESUME_BITS;
++ unsigned int signal_handler:1;
++ unsigned int vfork_stop:1; /* need utrace_stop() before vfork wait */
++ unsigned int death:1; /* in utrace_report_death() now */
++ unsigned int reap:1; /* release_task() has run */
++ unsigned int pending_attach:1; /* need splice_attaching() */
++};
++
++static struct kmem_cache *utrace_cachep;
++static struct kmem_cache *utrace_engine_cachep;
++static const struct utrace_engine_ops utrace_detached_ops; /* forward decl */
++
++static int __init utrace_init(void)
+{
-+ struct utrace *utrace;
++ utrace_cachep = KMEM_CACHE(utrace, SLAB_PANIC);
++ utrace_engine_cachep = KMEM_CACHE(utrace_engine, SLAB_PANIC);
++ return 0;
++}
++module_init(utrace_init);
+
-+ rcu_read_lock();
++void task_utrace_lock(struct task_struct *task)
++{
++ struct utrace *utrace = task_utrace_struct(task);
+
-+ /*
-+ * If this engine was already detached, bail out before we look at
-+ * the task_struct pointer at all. If it's detached after this
-+ * check, then RCU is still keeping this task_struct pointer valid.
-+ *
-+ * The ops pointer is NULL when the engine is fully detached.
-+ * It's &utrace_detached_ops when it's marked detached but still
-+ * on the list. In the latter case, utrace_barrier() still works,
-+ * since the target might be in the middle of an old callback.
-+ */
-+ if (unlikely(!engine->ops)) {
-+ rcu_read_unlock();
-+ return ERR_PTR(-ESRCH);
-+ }
++ if (!utrace) {
++ task_lock(task);
++ utrace = task_utrace_struct(task);
++ if (!utrace)
++ return;
+
-+ if (unlikely(engine->ops == &utrace_detached_ops)) {
-+ rcu_read_unlock();
-+ return attached ? ERR_PTR(-ESRCH) : ERR_PTR(-ERESTARTSYS);
++ task_unlock(task);
+ }
+
-+ utrace = task_utrace_struct(target);
+ spin_lock(&utrace->lock);
-+ if (unlikely(utrace->reap) || unlikely(!engine->ops) ||
-+ unlikely(engine->ops == &utrace_detached_ops)) {
-+ /*
-+ * By the time we got the utrace lock,
-+ * it had been reaped or detached already.
-+ */
-+ spin_unlock(&utrace->lock);
-+ utrace = ERR_PTR(-ESRCH);
-+ if (!attached && engine->ops == &utrace_detached_ops)
-+ utrace = ERR_PTR(-ERESTARTSYS);
-+ }
-+ rcu_read_unlock();
-+
-+ return utrace;
+}
+
-+/*
-+ * Now that we don't hold any locks, run through any
-+ * detached engines and free their references. Each
-+ * engine had one implicit ref while it was attached.
-+ */
-+static void put_detached_list(struct list_head *list)
++void task_utrace_unlock(struct task_struct *task)
+{
-+ struct utrace_engine *engine, *next;
-+ list_for_each_entry_safe(engine, next, list, entry) {
-+ list_del_init(&engine->entry);
-+ utrace_engine_put(engine);
-+ }
++ struct utrace *utrace = task_utrace_struct(task);
++
++ if (utrace)
++ spin_unlock(&utrace->lock);
++ else
++ task_unlock(task);
+}
+
+/*
-+ * We use an extra bit in utrace_engine.flags past the event bits,
-+ * to record whether the engine is keeping the target thread stopped.
++ * Set up @task.utrace for the first time. We can have races
++ * between two utrace_attach_task() calls here. The task_lock()
++ * governs installing the new pointer. If another one got in first,
++ * we just punt the new one we allocated.
+ *
-+ * This bit is set in task_struct.utrace_flags whenever it is set in any
-+ * engine's flags. Only utrace_reset() resets it in utrace_flags.
++ * This returns false only in case of a memory allocation failure.
+ */
-+#define ENGINE_STOP (1UL << _UTRACE_NEVENTS)
-+
-+static void mark_engine_wants_stop(struct task_struct *task,
-+ struct utrace_engine *engine)
++static bool utrace_task_alloc(struct task_struct *task)
+{
-+ engine->flags |= ENGINE_STOP;
-+ task->utrace_flags |= ENGINE_STOP;
++ struct utrace *utrace = kmem_cache_zalloc(utrace_cachep, GFP_KERNEL);
++ if (unlikely(!utrace))
++ return false;
++ spin_lock_init(&utrace->lock);
++ INIT_LIST_HEAD(&utrace->attached);
++ INIT_LIST_HEAD(&utrace->attaching);
++ utrace->resume = UTRACE_RESUME;
++ task_lock(task);
++ if (likely(!task->utrace)) {
++ /*
++ * This barrier makes sure the initialization of the struct
++ * precedes the installation of the pointer. This pairs
++ * with smp_read_barrier_depends() in task_utrace_struct().
++ */
++ smp_wmb();
++ task->utrace = utrace;
++ }
++ task_unlock(task);
++
++ if (unlikely(task->utrace != utrace))
++ kmem_cache_free(utrace_cachep, utrace);
++ return true;
+}
+
-+static void clear_engine_wants_stop(struct utrace_engine *engine)
++/*
++ * This is called via tracehook_free_task() from free_task()
++ * when @task is being deallocated.
++ */
++void utrace_free_task(struct task_struct *task)
+{
-+ engine->flags &= ~ENGINE_STOP;
++ kmem_cache_free(utrace_cachep, task->utrace);
+}
+
-+static bool engine_wants_stop(struct utrace_engine *engine)
++/*
++ * This is calledwhen the task is safely quiescent, i.e. it won't consult
++ * utrace->attached without the lock. Move any engines attached
++ * asynchronously from @utrace->attaching onto the @utrace->attached list.
++ */
++static void splice_attaching(struct utrace *utrace)
+{
-+ return (engine->flags & ENGINE_STOP) != 0;
++ lockdep_assert_held(&utrace->lock);
++ list_splice_tail_init(&utrace->attaching, &utrace->attached);
++ utrace->pending_attach = 0;
+}
+
-+/**
-+ * utrace_set_events - choose which event reports a tracing engine gets
-+ * @target: thread to affect
-+ * @engine: attached engine to affect
-+ * @events: new event mask
-+ *
-+ * This changes the set of events for which @engine wants callbacks made.
-+ *
-+ * This fails with -%EALREADY and does nothing if you try to clear
-+ * %UTRACE_EVENT(%DEATH) when the @report_death callback may already have
-+ * begun, or if you try to newly set %UTRACE_EVENT(%DEATH) or
-+ * %UTRACE_EVENT(%QUIESCE) when @target is already dead or dying.
-+ *
-+ * This fails with -%ESRCH if you try to clear %UTRACE_EVENT(%REAP) when
-+ * the @report_reap callback may already have begun, or when @target has
-+ * already been detached, including forcible detach on reaping.
-+ *
-+ * If @target was stopped before the call, then after a successful call,
-+ * no event callbacks not requested in @events will be made; if
-+ * %UTRACE_EVENT(%QUIESCE) is included in @events, then a
-+ * @report_quiesce callback will be made when @target resumes.
-+ *
-+ * If @target was not stopped and @events excludes some bits that were
-+ * set before, this can return -%EINPROGRESS to indicate that @target
-+ * may have been making some callback to @engine. When this returns
-+ * zero, you can be sure that no event callbacks you've disabled in
-+ * @events can be made. If @events only sets new bits that were not set
-+ * before on @engine, then -%EINPROGRESS will never be returned.
-+ *
-+ * To synchronize after an -%EINPROGRESS return, see utrace_barrier().
-+ *
-+ * When @target is @current, -%EINPROGRESS is not returned. But note
-+ * that a newly-created engine will not receive any callbacks related to
-+ * an event notification already in progress. This call enables @events
-+ * callbacks to be made as soon as @engine becomes eligible for any
-+ * callbacks, see utrace_attach_task().
-+ *
-+ * These rules provide for coherent synchronization based on %UTRACE_STOP,
-+ * even when %SIGKILL is breaking its normal simple rules.
++/*
++ * This is the exported function used by the utrace_engine_put() inline.
+ */
-+int utrace_set_events(struct task_struct *target,
-+ struct utrace_engine *engine,
-+ unsigned long events)
++void __utrace_engine_release(struct kref *kref)
+{
-+ struct utrace *utrace;
-+ unsigned long old_flags, old_utrace_flags;
-+ int ret = -EALREADY;
++ struct utrace_engine *engine = container_of(kref, struct utrace_engine,
++ kref);
++ BUG_ON(!list_empty(&engine->entry));
++ if (engine->release)
++ (*engine->release)(engine->data);
++ kmem_cache_free(utrace_engine_cachep, engine);
++}
++EXPORT_SYMBOL_GPL(__utrace_engine_release);
+
-+ /*
-+ * We just ignore the internal bit, so callers can use
-+ * engine->flags to seed bitwise ops for our argument.
-+ */
-+ events &= ~ENGINE_STOP;
++static bool engine_matches(struct utrace_engine *engine, int flags,
++ const struct utrace_engine_ops *ops, void *data)
++{
++ if ((flags & UTRACE_ATTACH_MATCH_OPS) && engine->ops != ops)
++ return false;
++ if ((flags & UTRACE_ATTACH_MATCH_DATA) && engine->data != data)
++ return false;
++ return engine->ops && engine->ops != &utrace_detached_ops;
++}
+
-+ utrace = get_utrace_lock(target, engine, true);
-+ if (unlikely(IS_ERR(utrace)))
-+ return PTR_ERR(utrace);
++static struct utrace_engine *find_matching_engine(
++ struct utrace *utrace, int flags,
++ const struct utrace_engine_ops *ops, void *data)
++{
++ struct utrace_engine *engine;
++ list_for_each_entry(engine, &utrace->attached, entry)
++ if (engine_matches(engine, flags, ops, data))
++ return engine;
++ list_for_each_entry(engine, &utrace->attaching, entry)
++ if (engine_matches(engine, flags, ops, data))
++ return engine;
++ return NULL;
++}
+
-+ old_utrace_flags = target->utrace_flags;
-+ old_flags = engine->flags & ~ENGINE_STOP;
++/*
++ * Enqueue @engine, or maybe don't if UTRACE_ATTACH_EXCLUSIVE.
++ */
++static int utrace_add_engine(struct task_struct *target,
++ struct utrace *utrace,
++ struct utrace_engine *engine,
++ int flags,
++ const struct utrace_engine_ops *ops,
++ void *data)
++{
++ int ret;
+
-+ /*
-+ * If utrace_report_death() is already progress now,
-+ * it's too late to clear the death event bits.
-+ */
-+ if (((old_flags & ~events) & _UTRACE_DEATH_EVENTS) && utrace->death)
++ spin_lock(&utrace->lock);
++
++ ret = -EEXIST;
++ if ((flags & UTRACE_ATTACH_EXCLUSIVE) &&
++ unlikely(find_matching_engine(utrace, flags, ops, data)))
+ goto unlock;
+
+ /*
-+ * When setting these flags, it's essential that we really
-+ * synchronize with exit_notify(). They cannot be set after
-+ * exit_notify() takes the tasklist_lock. By holding the read
-+ * lock here while setting the flags, we ensure that the calls
-+ * to tracehook_notify_death() and tracehook_report_death() will
-+ * see the new flags. This ensures that utrace_release_task()
-+ * knows positively that utrace_report_death() will be called or
-+ * that it won't.
++ * In case we had no engines before, make sure that
++ * utrace_flags is not zero. Since we did unlock+lock
++ * at least once after utrace_task_alloc() installed
++ * ->utrace, we have the necessary barrier which pairs
++ * with rmb() in task_utrace_struct().
+ */
-+ if ((events & ~old_flags) & _UTRACE_DEATH_EVENTS) {
-+ read_lock(&tasklist_lock);
-+ if (unlikely(target->exit_state)) {
-+ read_unlock(&tasklist_lock);
++ ret = -ESRCH;
++ if (!target->utrace_flags) {
++ target->utrace_flags = UTRACE_EVENT(REAP);
++ /*
++ * If we race with tracehook_prepare_release_task()
++ * make sure that either it sees utrace_flags != 0
++ * or we see exit_state == EXIT_DEAD.
++ */
++ smp_mb();
++ if (unlikely(target->exit_state == EXIT_DEAD)) {
++ target->utrace_flags = 0;
+ goto unlock;
+ }
-+ target->utrace_flags |= events;
-+ read_unlock(&tasklist_lock);
+ }
+
-+ engine->flags = events | (engine->flags & ENGINE_STOP);
-+ target->utrace_flags |= events;
-+
-+ if ((events & UTRACE_EVENT_SYSCALL) &&
-+ !(old_utrace_flags & UTRACE_EVENT_SYSCALL))
-+ set_tsk_thread_flag(target, TIF_SYSCALL_TRACE);
-+
++ /*
++ * Put the new engine on the pending ->attaching list.
++ * Make sure it gets onto the ->attached list by the next
++ * time it's examined. Setting ->pending_attach ensures
++ * that start_report() takes the lock and splices the lists
++ * before the next new reporting pass.
++ *
++ * When target == current, it would be safe just to call
++ * splice_attaching() right here. But if we're inside a
++ * callback, that would mean the new engine also gets
++ * notified about the event that precipitated its own
++ * creation. This is not what the user wants.
++ */
++ list_add_tail(&engine->entry, &utrace->attaching);
++ utrace->pending_attach = 1;
++ utrace_engine_get(engine);
+ ret = 0;
-+ if ((old_flags & ~events) && target != current &&
-+ !task_is_stopped_or_traced(target) && !target->exit_state) {
-+ /*
-+ * This barrier ensures that our engine->flags changes
-+ * have hit before we examine utrace->reporting,
-+ * pairing with the barrier in start_callback(). If
-+ * @target has not yet hit finish_callback() to clear
-+ * utrace->reporting, we might be in the middle of a
-+ * callback to @engine.
-+ */
-+ smp_mb();
-+ if (utrace->reporting == engine)
-+ ret = -EINPROGRESS;
-+ }
+unlock:
+ spin_unlock(&utrace->lock);
+
+ return ret;
+}
-+EXPORT_SYMBOL_GPL(utrace_set_events);
+
-+/*
-+ * Asynchronously mark an engine as being detached.
++/**
++ * utrace_attach_task - attach new engine, or look up an attached engine
++ * @target: thread to attach to
++ * @flags: flag bits combined with OR, see below
++ * @ops: callback table for new engine
++ * @data: engine private data pointer
+ *
-+ * This must work while the target thread races with us doing
-+ * start_callback(), defined below. It uses smp_rmb() between checking
-+ * @engine->flags and using @engine->ops. Here we change @engine->ops
-+ * first, then use smp_wmb() before changing @engine->flags. This ensures
-+ * it can check the old flags before using the old ops, or check the old
-+ * flags before using the new ops, or check the new flags before using the
-+ * new ops, but can never check the new flags before using the old ops.
-+ * Hence, utrace_detached_ops might be used with any old flags in place.
-+ * It has report_quiesce() and report_reap() callbacks to handle all cases.
-+ */
-+static void mark_engine_detached(struct utrace_engine *engine)
-+{
-+ engine->ops = &utrace_detached_ops;
-+ smp_wmb();
-+ engine->flags = UTRACE_EVENT(QUIESCE);
-+}
-+
-+/*
-+ * Get @target to stop and return true if it is already stopped now.
-+ * If we return false, it will make some event callback soonish.
-+ * Called with @utrace locked.
++ * The caller must ensure that the @target thread does not get freed,
++ * i.e. hold a ref or be its parent. It is always safe to call this
++ * on @current, or on the @child pointer in a @report_clone callback.
++ * For most other cases, it's easier to use utrace_attach_pid() instead.
++ *
++ * UTRACE_ATTACH_CREATE:
++ * Create a new engine. If %UTRACE_ATTACH_CREATE is not specified, you
++ * only look up an existing engine already attached to the thread.
++ *
++ * UTRACE_ATTACH_EXCLUSIVE:
++ * Attempting to attach a second (matching) engine fails with -%EEXIST.
++ *
++ * UTRACE_ATTACH_MATCH_OPS: Only consider engines matching @ops.
++ * UTRACE_ATTACH_MATCH_DATA: Only consider engines matching @data.
++ *
++ * Calls with neither %UTRACE_ATTACH_MATCH_OPS nor %UTRACE_ATTACH_MATCH_DATA
++ * match the first among any engines attached to @target. That means that
++ * %UTRACE_ATTACH_EXCLUSIVE in such a call fails with -%EEXIST if there
++ * are any engines on @target at all.
+ */
-+static bool utrace_do_stop(struct task_struct *target, struct utrace *utrace)
++struct utrace_engine *utrace_attach_task(
++ struct task_struct *target, int flags,
++ const struct utrace_engine_ops *ops, void *data)
+{
-+ if (task_is_stopped(target)) {
++ struct utrace *utrace = task_utrace_struct(target);
++ struct utrace_engine *engine;
++ int ret;
++
++ if (!(flags & UTRACE_ATTACH_CREATE)) {
++ if (unlikely(!utrace))
++ return ERR_PTR(-ENOENT);
++ spin_lock(&utrace->lock);
++ engine = find_matching_engine(utrace, flags, ops, data);
++ if (engine)
++ utrace_engine_get(engine);
++ spin_unlock(&utrace->lock);
++ return engine ?: ERR_PTR(-ENOENT);
++ }
++
++ if (unlikely(!ops) || unlikely(ops == &utrace_detached_ops))
++ return ERR_PTR(-EINVAL);
++
++ if (unlikely(target->flags & PF_KTHREAD))
+ /*
-+ * Stopped is considered quiescent; when it wakes up, it will
-+ * go through utrace_finish_stop() before doing anything else.
++ * Silly kernel, utrace is for users!
+ */
-+ spin_lock_irq(&target->sighand->siglock);
-+ if (likely(task_is_stopped(target)))
-+ __set_task_state(target, TASK_TRACED);
-+ spin_unlock_irq(&target->sighand->siglock);
-+ } else if (utrace->resume > UTRACE_REPORT) {
-+ utrace->resume = UTRACE_REPORT;
-+ set_notify_resume(target);
++ return ERR_PTR(-EPERM);
++
++ if (!utrace) {
++ if (unlikely(!utrace_task_alloc(target)))
++ return ERR_PTR(-ENOMEM);
++ utrace = task_utrace_struct(target);
++ }
++
++ engine = kmem_cache_alloc(utrace_engine_cachep, GFP_KERNEL);
++ if (unlikely(!engine))
++ return ERR_PTR(-ENOMEM);
++
++ /*
++ * Initialize the new engine structure. It starts out with one ref
++ * to return. utrace_add_engine() adds another for being attached.
++ */
++ kref_init(&engine->kref);
++ engine->flags = 0;
++ engine->ops = ops;
++ engine->data = data;
++ engine->release = ops->release;
++
++ ret = utrace_add_engine(target, utrace, engine, flags, ops, data);
++
++ if (unlikely(ret)) {
++ kmem_cache_free(utrace_engine_cachep, engine);
++ engine = ERR_PTR(ret);
+ }
+
-+ return task_is_traced(target);
++
++ return engine;
++}
++EXPORT_SYMBOL_GPL(utrace_attach_task);
++
++/**
++ * utrace_attach_pid - attach new engine, or look up an attached engine
++ * @pid: &struct pid pointer representing thread to attach to
++ * @flags: flag bits combined with OR, see utrace_attach_task()
++ * @ops: callback table for new engine
++ * @data: engine private data pointer
++ *
++ * This is the same as utrace_attach_task(), but takes a &struct pid
++ * pointer rather than a &struct task_struct pointer. The caller must
++ * hold a ref on @pid, but does not need to worry about the task
++ * staying valid. If it's been reaped so that @pid points nowhere,
++ * then this call returns -%ESRCH.
++ */
++struct utrace_engine *utrace_attach_pid(
++ struct pid *pid, int flags,
++ const struct utrace_engine_ops *ops, void *data)
++{
++ struct utrace_engine *engine = ERR_PTR(-ESRCH);
++ struct task_struct *task = get_pid_task(pid, PIDTYPE_PID);
++ if (task) {
++ engine = utrace_attach_task(task, flags, ops, data);
++ put_task_struct(task);
++ }
++ return engine;
+}
++EXPORT_SYMBOL_GPL(utrace_attach_pid);
+
+/*
-+ * If the target is not dead it should not be in tracing
-+ * stop any more. Wake it unless it's in job control stop.
++ * When an engine is detached, the target thread may still see it and
++ * make callbacks until it quiesces. We install a special ops vector
++ * with these two callbacks. When the target thread quiesces, it can
++ * safely free the engine itself. For any event we will always get
++ * the report_quiesce() callback first, so we only need this one
++ * pointer to be set. The only exception is report_reap(), so we
++ * supply that callback too.
+ */
-+static void utrace_wakeup(struct task_struct *target, struct utrace *utrace)
++static u32 utrace_detached_quiesce(u32 action, struct utrace_engine *engine,
++ unsigned long event)
++{
++ return UTRACE_DETACH;
++}
++
++static void utrace_detached_reap(struct utrace_engine *engine,
++ struct task_struct *task)
+{
-+ lockdep_assert_held(&utrace->lock);
-+ spin_lock_irq(&target->sighand->siglock);
-+ if (target->signal->flags & SIGNAL_STOP_STOPPED ||
-+ target->signal->group_stop_count)
-+ target->state = TASK_STOPPED;
-+ else
-+ wake_up_state(target, __TASK_TRACED);
-+ spin_unlock_irq(&target->sighand->siglock);
+}
+
++static const struct utrace_engine_ops utrace_detached_ops = {
++ .report_quiesce = &utrace_detached_quiesce,
++ .report_reap = &utrace_detached_reap
++};
++
+/*
-+ * This is called when there might be some detached engines on the list or
-+ * some stale bits in @task->utrace_flags. Clean them up and recompute the
-+ * flags. Returns true if we're now fully detached.
++ * The caller has to hold a ref on the engine. If the attached flag is
++ * true (all but utrace_barrier() calls), the engine is supposed to be
++ * attached. If the attached flag is false (utrace_barrier() only),
++ * then return -ERESTARTSYS for an engine marked for detach but not yet
++ * fully detached. The task pointer can be invalid if the engine is
++ * detached.
+ *
-+ * Called with @utrace->lock held, returns with it released.
-+ * After this returns, @utrace might be freed if everything detached.
++ * Get the utrace lock for the target task.
++ * Returns the struct if locked, or ERR_PTR(-errno).
++ *
++ * This has to be robust against races with:
++ * utrace_control(target, UTRACE_DETACH) calls
++ * UTRACE_DETACH after reports
++ * utrace_report_death
++ * utrace_release_task
+ */
-+static bool utrace_reset(struct task_struct *task, struct utrace *utrace)
-+ __releases(utrace->lock)
++static struct utrace *get_utrace_lock(struct task_struct *target,
++ struct utrace_engine *engine,
++ bool attached)
++ __acquires(utrace->lock)
+{
-+ struct utrace_engine *engine, *next;
-+ unsigned long flags = 0;
-+ LIST_HEAD(detached);
++ struct utrace *utrace;
+
-+ splice_attaching(utrace);
++ rcu_read_lock();
+
+ /*
-+ * Update the set of events of interest from the union
-+ * of the interests of the remaining tracing engines.
-+ * For any engine marked detached, remove it from the list.
-+ * We'll collect them on the detached list.
++ * If this engine was already detached, bail out before we look at
++ * the task_struct pointer at all. If it's detached after this
++ * check, then RCU is still keeping this task_struct pointer valid.
++ *
++ * The ops pointer is NULL when the engine is fully detached.
++ * It's &utrace_detached_ops when it's marked detached but still
++ * on the list. In the latter case, utrace_barrier() still works,
++ * since the target might be in the middle of an old callback.
+ */
-+ list_for_each_entry_safe(engine, next, &utrace->attached, entry) {
-+ if (engine->ops == &utrace_detached_ops) {
-+ engine->ops = NULL;
-+ list_move(&engine->entry, &detached);
-+ } else {
-+ flags |= engine->flags | UTRACE_EVENT(REAP);
-+ }
-+ }
-+
-+ if (task->exit_state) {
-+ /*
-+ * Once it's already dead, we never install any flags
-+ * except REAP. When ->exit_state is set and events
-+ * like DEATH are not set, then they never can be set.
-+ * This ensures that utrace_release_task() knows
-+ * positively that utrace_report_death() can never run.
-+ */
-+ BUG_ON(utrace->death);
-+ flags &= UTRACE_EVENT(REAP);
-+ } else if (!(flags & UTRACE_EVENT_SYSCALL) &&
-+ test_tsk_thread_flag(task, TIF_SYSCALL_TRACE)) {
-+ clear_tsk_thread_flag(task, TIF_SYSCALL_TRACE);
++ if (unlikely(!engine->ops)) {
++ rcu_read_unlock();
++ return ERR_PTR(-ESRCH);
+ }
+
-+ if (!flags) {
-+ /*
-+ * No more engines, cleared out the utrace.
-+ */
-+ utrace->resume = UTRACE_RESUME;
-+ utrace->signal_handler = 0;
++ if (unlikely(engine->ops == &utrace_detached_ops)) {
++ rcu_read_unlock();
++ return attached ? ERR_PTR(-ESRCH) : ERR_PTR(-ERESTARTSYS);
+ }
+
-+ /*
-+ * If no more engines want it stopped, wake it up.
-+ */
-+ if (task_is_traced(task) && !(flags & ENGINE_STOP)) {
++ utrace = task_utrace_struct(target);
++ spin_lock(&utrace->lock);
++ if (unlikely(utrace->reap) || unlikely(!engine->ops) ||
++ unlikely(engine->ops == &utrace_detached_ops)) {
+ /*
-+ * It just resumes, so make sure single-step
-+ * is not left set.
++ * By the time we got the utrace lock,
++ * it had been reaped or detached already.
+ */
-+ if (utrace->resume == UTRACE_RESUME)
-+ user_disable_single_step(task);
-+ utrace_wakeup(task, utrace);
++ spin_unlock(&utrace->lock);
++ utrace = ERR_PTR(-ESRCH);
++ if (!attached && engine->ops == &utrace_detached_ops)
++ utrace = ERR_PTR(-ERESTARTSYS);
+ }
-+
-+ /*
-+ * In theory spin_lock() doesn't imply rcu_read_lock().
-+ * Once we clear ->utrace_flags this task_struct can go away
-+ * because tracehook_prepare_release_task() path does not take
-+ * utrace->lock when ->utrace_flags == 0.
-+ */
-+ rcu_read_lock();
-+ task->utrace_flags = flags;
-+ spin_unlock(&utrace->lock);
+ rcu_read_unlock();
+
-+ put_detached_list(&detached);
-+
-+ return !flags;
++ return utrace;
+}
+
-+void utrace_finish_stop(void)
++/*
++ * Now that we don't hold any locks, run through any
++ * detached engines and free their references. Each
++ * engine had one implicit ref while it was attached.
++ */
++static void put_detached_list(struct list_head *list)
+{
-+ /*
-+ * If we were task_is_traced() and then SIGKILL'ed, make
-+ * sure we do nothing until the tracer drops utrace->lock.
-+ */
-+ if (unlikely(__fatal_signal_pending(current))) {
-+ struct utrace *utrace = task_utrace_struct(current);
-+ spin_unlock_wait(&utrace->lock);
++ struct utrace_engine *engine, *next;
++ list_for_each_entry_safe(engine, next, list, entry) {
++ list_del_init(&engine->entry);
++ utrace_engine_put(engine);
+ }
+}
+
+/*
-+ * Perform %UTRACE_STOP, i.e. block in TASK_TRACED until woken up.
-+ * @task == current, @utrace == current->utrace, which is not locked.
-+ * Return true if we were woken up by SIGKILL even though some utrace
-+ * engine may still want us to stay stopped.
++ * We use an extra bit in utrace_engine.flags past the event bits,
++ * to record whether the engine is keeping the target thread stopped.
++ *
++ * This bit is set in task_struct.utrace_flags whenever it is set in any
++ * engine's flags. Only utrace_reset() resets it in utrace_flags.
+ */
-+static void utrace_stop(struct task_struct *task, struct utrace *utrace,
-+ enum utrace_resume_action action)
-+{
-+relock:
-+ spin_lock(&utrace->lock);
++#define ENGINE_STOP (1UL << _UTRACE_NEVENTS)
+
-+ if (action < utrace->resume) {
-+ /*
-+ * Ensure a reporting pass when we're resumed.
-+ */
-+ utrace->resume = action;
-+ if (action == UTRACE_INTERRUPT)
-+ set_thread_flag(TIF_SIGPENDING);
-+ else
-+ set_thread_flag(TIF_NOTIFY_RESUME);
-+ }
++#define task_is_utraced(task) ((task->state & __TASK_UTRACED) != 0)
+
-+ /*
-+ * If the ENGINE_STOP bit is clear in utrace_flags, that means
-+ * utrace_reset() ran after we processed some UTRACE_STOP return
-+ * values from callbacks to get here. If all engines have detached
-+ * or resumed us, we don't stop. This check doesn't require
-+ * siglock, but it should follow the interrupt/report bookkeeping
-+ * steps (this can matter for UTRACE_RESUME but not UTRACE_DETACH).
-+ */
-+ if (unlikely(!(task->utrace_flags & ENGINE_STOP))) {
-+ utrace_reset(task, utrace);
-+ if (task->utrace_flags & ENGINE_STOP)
-+ goto relock;
-+ return;
-+ }
++static void mark_engine_wants_stop(struct task_struct *task,
++ struct utrace_engine *engine)
++{
++ engine->flags |= ENGINE_STOP;
++ task->utrace_flags |= ENGINE_STOP;
++}
+
-+ /*
-+ * The siglock protects us against signals. As well as SIGKILL
-+ * waking us up, we must synchronize with the signal bookkeeping
-+ * for stop signals and SIGCONT.
-+ */
-+ spin_lock_irq(&task->sighand->siglock);
++static void clear_engine_wants_stop(struct utrace_engine *engine)
++{
++ engine->flags &= ~ENGINE_STOP;
++}
+
-+ if (unlikely(__fatal_signal_pending(task))) {
-+ spin_unlock_irq(&task->sighand->siglock);
-+ spin_unlock(&utrace->lock);
-+ return;
-+ }
++static bool engine_wants_stop(struct utrace_engine *engine)
++{
++ return (engine->flags & ENGINE_STOP) != 0;
++}
+
-+ __set_current_state(TASK_TRACED);
++/**
++ * utrace_set_events - choose which event reports a tracing engine gets
++ * @target: thread to affect
++ * @engine: attached engine to affect
++ * @events: new event mask
++ *
++ * This changes the set of events for which @engine wants callbacks made.
++ *
++ * This fails with -%EALREADY and does nothing if you try to clear
++ * %UTRACE_EVENT(%DEATH) when the @report_death callback may already have
++ * begun, or if you try to newly set %UTRACE_EVENT(%DEATH) or
++ * %UTRACE_EVENT(%QUIESCE) when @target is already dead or dying.
++ *
++ * This fails with -%ESRCH if you try to clear %UTRACE_EVENT(%REAP) when
++ * the @report_reap callback may already have begun, or when @target has
++ * already been detached, including forcible detach on reaping.
++ *
++ * If @target was stopped before the call, then after a successful call,
++ * no event callbacks not requested in @events will be made; if
++ * %UTRACE_EVENT(%QUIESCE) is included in @events, then a
++ * @report_quiesce callback will be made when @target resumes.
++ *
++ * If @target was not stopped and @events excludes some bits that were
++ * set before, this can return -%EINPROGRESS to indicate that @target
++ * may have been making some callback to @engine. When this returns
++ * zero, you can be sure that no event callbacks you've disabled in
++ * @events can be made. If @events only sets new bits that were not set
++ * before on @engine, then -%EINPROGRESS will never be returned.
++ *
++ * To synchronize after an -%EINPROGRESS return, see utrace_barrier().
++ *
++ * When @target is @current, -%EINPROGRESS is not returned. But note
++ * that a newly-created engine will not receive any callbacks related to
++ * an event notification already in progress. This call enables @events
++ * callbacks to be made as soon as @engine becomes eligible for any
++ * callbacks, see utrace_attach_task().
++ *
++ * These rules provide for coherent synchronization based on %UTRACE_STOP,
++ * even when %SIGKILL is breaking its normal simple rules.
++ */
++int utrace_set_events(struct task_struct *target,
++ struct utrace_engine *engine,
++ unsigned long events)
++{
++ struct utrace *utrace;
++ unsigned long old_flags, old_utrace_flags;
++ int ret = -EALREADY;
+
+ /*
-+ * If there is a group stop in progress,
-+ * we must participate in the bookkeeping.
++ * We just ignore the internal bit, so callers can use
++ * engine->flags to seed bitwise ops for our argument.
+ */
-+ if (unlikely(task->signal->group_stop_count) &&
-+ !--task->signal->group_stop_count)
-+ task->signal->flags = SIGNAL_STOP_STOPPED;
-+
-+ spin_unlock_irq(&task->sighand->siglock);
-+ spin_unlock(&utrace->lock);
++ events &= ~ENGINE_STOP;
+
-+ schedule();
++ utrace = get_utrace_lock(target, engine, true);
++ if (unlikely(IS_ERR(utrace)))
++ return PTR_ERR(utrace);
+
-+ utrace_finish_stop();
++ old_utrace_flags = target->utrace_flags;
++ old_flags = engine->flags & ~ENGINE_STOP;
+
+ /*
-+ * While in TASK_TRACED, we were considered "frozen enough".
-+ * Now that we woke up, it's crucial if we're supposed to be
-+ * frozen that we freeze now before running anything substantial.
++ * If utrace_report_death() is already progress now,
++ * it's too late to clear the death event bits.
+ */
-+ try_to_freeze();
++ if (((old_flags & ~events) & _UTRACE_DEATH_EVENTS) && utrace->death)
++ goto unlock;
+
+ /*
-+ * While we were in TASK_TRACED, complete_signal() considered
-+ * us "uninterested" in signal wakeups. Now make sure our
-+ * TIF_SIGPENDING state is correct for normal running.
++ * When setting these flags, it's essential that we really
++ * synchronize with exit_notify(). They cannot be set after
++ * exit_notify() takes the tasklist_lock. By holding the read
++ * lock here while setting the flags, we ensure that the calls
++ * to tracehook_notify_death() and tracehook_report_death() will
++ * see the new flags. This ensures that utrace_release_task()
++ * knows positively that utrace_report_death() will be called or
++ * that it won't.
+ */
-+ spin_lock_irq(&task->sighand->siglock);
-+ recalc_sigpending();
-+ spin_unlock_irq(&task->sighand->siglock);
-+}
-+
-+/*
-+ * Called by release_task() with @reap set to true.
-+ * Called by utrace_report_death() with @reap set to false.
-+ * On reap, make report_reap callbacks and clean out @utrace
-+ * unless still making callbacks. On death, update bookkeeping
-+ * and handle the reap work if release_task() came in first.
-+ */
-+void utrace_maybe_reap(struct task_struct *target, struct utrace *utrace,
-+ bool reap)
-+{
-+ struct utrace_engine *engine, *next;
-+ struct list_head attached;
++ if ((events & ~old_flags) & _UTRACE_DEATH_EVENTS) {
++ read_lock(&tasklist_lock);
++ if (unlikely(target->exit_state)) {
++ read_unlock(&tasklist_lock);
++ goto unlock;
++ }
++ target->utrace_flags |= events;
++ read_unlock(&tasklist_lock);
++ }
+
-+ spin_lock(&utrace->lock);
++ engine->flags = events | (engine->flags & ENGINE_STOP);
++ target->utrace_flags |= events;
+
-+ if (reap) {
-+ /*
-+ * If the target will do some final callbacks but hasn't
-+ * finished them yet, we know because it clears these event
-+ * bits after it's done. Instead of cleaning up here and
-+ * requiring utrace_report_death() to cope with it, we
-+ * delay the REAP report and the teardown until after the
-+ * target finishes its death reports.
-+ */
-+ utrace->reap = 1;
++ if ((events & UTRACE_EVENT_SYSCALL) &&
++ !(old_utrace_flags & UTRACE_EVENT_SYSCALL))
++ set_tsk_thread_flag(target, TIF_SYSCALL_TRACE);
+
-+ if (target->utrace_flags & _UTRACE_DEATH_EVENTS) {
-+ spin_unlock(&utrace->lock);
-+ return;
-+ }
-+ } else {
++ ret = 0;
++ if ((old_flags & ~events) && target != current &&
++ !task_is_utraced(target) && !target->exit_state) {
+ /*
-+ * After we unlock with this flag clear, any competing
-+ * utrace_control/utrace_set_events calls know that we've
-+ * finished our callbacks and any detach bookkeeping.
++ * This barrier ensures that our engine->flags changes
++ * have hit before we examine utrace->reporting,
++ * pairing with the barrier in start_callback(). If
++ * @target has not yet hit finish_callback() to clear
++ * utrace->reporting, we might be in the middle of a
++ * callback to @engine.
+ */
-+ utrace->death = 0;
-+
-+ if (!utrace->reap) {
-+ /*
-+ * We're just dead, not reaped yet. This will
-+ * reset @target->utrace_flags so the later call
-+ * with @reap set won't hit the check above.
-+ */
-+ utrace_reset(target, utrace);
-+ return;
-+ }
++ smp_mb();
++ if (utrace->reporting == engine)
++ ret = -EINPROGRESS;
+ }
-+
-+ /*
-+ * utrace_add_engine() checks ->utrace_flags != 0. Since
-+ * @utrace->reap is set, nobody can set or clear UTRACE_EVENT(REAP)
-+ * in @engine->flags or change @engine->ops and nobody can change
-+ * @utrace->attached after we drop the lock.
-+ */
-+ target->utrace_flags = 0;
-+
-+ /*
-+ * We clear out @utrace->attached before we drop the lock so
-+ * that find_matching_engine() can't come across any old engine
-+ * while we are busy tearing it down.
-+ */
-+ list_replace_init(&utrace->attached, &attached);
-+ list_splice_tail_init(&utrace->attaching, &attached);
-+
++unlock:
+ spin_unlock(&utrace->lock);
+
-+ list_for_each_entry_safe(engine, next, &attached, entry) {
-+ if (engine->flags & UTRACE_EVENT(REAP))
-+ engine->ops->report_reap(engine, target);
-+
-+ engine->ops = NULL;
-+ engine->flags = 0;
-+ list_del_init(&engine->entry);
-+
-+ utrace_engine_put(engine);
-+ }
++ return ret;
+}
++EXPORT_SYMBOL_GPL(utrace_set_events);
+
+/*
-+ * You can't do anything to a dead task but detach it.
-+ * If release_task() has been called, you can't do that.
++ * Asynchronously mark an engine as being detached.
+ *
-+ * On the exit path, DEATH and QUIESCE event bits are set only
-+ * before utrace_report_death() has taken the lock. At that point,
-+ * the death report will come soon, so disallow detach until it's
-+ * done. This prevents us from racing with it detaching itself.
-+ *
-+ * Called only when @target->exit_state is nonzero.
++ * This must work while the target thread races with us doing
++ * start_callback(), defined below. It uses smp_rmb() between checking
++ * @engine->flags and using @engine->ops. Here we change @engine->ops
++ * first, then use smp_wmb() before changing @engine->flags. This ensures
++ * it can check the old flags before using the old ops, or check the old
++ * flags before using the new ops, or check the new flags before using the
++ * new ops, but can never check the new flags before using the old ops.
++ * Hence, utrace_detached_ops might be used with any old flags in place.
++ * It has report_quiesce() and report_reap() callbacks to handle all cases.
+ */
-+static inline int utrace_control_dead(struct task_struct *target,
-+ struct utrace *utrace,
-+ enum utrace_resume_action action)
++static void mark_engine_detached(struct utrace_engine *engine)
+{
-+ lockdep_assert_held(&utrace->lock);
-+
-+ if (action != UTRACE_DETACH || unlikely(utrace->reap))
-+ return -ESRCH;
++ engine->ops = &utrace_detached_ops;
++ smp_wmb();
++ engine->flags = UTRACE_EVENT(QUIESCE);
++}
+
-+ if (unlikely(utrace->death))
++/*
++ * Get @target to stop and return true if it is already stopped now.
++ * If we return false, it will make some event callback soonish.
++ * Called with @utrace locked.
++ */
++static bool utrace_do_stop(struct task_struct *target, struct utrace *utrace)
++{
++ if (task_is_stopped_or_traced(target)) {
+ /*
-+ * We have already started the death report. We can't
-+ * prevent the report_death and report_reap callbacks,
-+ * so tell the caller they will happen.
++ * Stopped is considered quiescent; when it wakes up, it will
++ * go through utrace_finish_stop() before doing anything else.
+ */
-+ return -EALREADY;
++ spin_lock_irq(&target->sighand->siglock);
++ if (likely(task_is_stopped_or_traced(target)))
++ target->state |= TASK_UTRACED;
++ spin_unlock_irq(&target->sighand->siglock);
++ } else if (utrace->resume > UTRACE_REPORT) {
++ utrace->resume = UTRACE_REPORT;
++ set_notify_resume(target);
++ }
+
-+ return 0;
++ return task_is_utraced(target);
+}
+
-+/**
-+ * utrace_control - control a thread being traced by a tracing engine
-+ * @target: thread to affect
-+ * @engine: attached engine to affect
-+ * @action: &enum utrace_resume_action for thread to do
-+ *
-+ * This is how a tracing engine asks a traced thread to do something.
-+ * This call is controlled by the @action argument, which has the
-+ * same meaning as the &enum utrace_resume_action value returned by
-+ * event reporting callbacks.
-+ *
-+ * If @target is already dead (@target->exit_state nonzero),
-+ * all actions except %UTRACE_DETACH fail with -%ESRCH.
-+ *
-+ * The following sections describe each option for the @action argument.
-+ *
-+ * UTRACE_DETACH:
-+ *
-+ * After this, the @engine data structure is no longer accessible,
-+ * and the thread might be reaped. The thread will start running
-+ * again if it was stopped and no longer has any attached engines
-+ * that want it stopped.
-+ *
-+ * If the @report_reap callback may already have begun, this fails
-+ * with -%ESRCH. If the @report_death callback may already have
-+ * begun, this fails with -%EALREADY.
-+ *
-+ * If @target is not already stopped, then a callback to this engine
-+ * might be in progress or about to start on another CPU. If so,
-+ * then this returns -%EINPROGRESS; the detach happens as soon as
-+ * the pending callback is finished. To synchronize after an
-+ * -%EINPROGRESS return, see utrace_barrier().
-+ *
-+ * If @target is properly stopped before utrace_control() is called,
-+ * then after successful return it's guaranteed that no more callbacks
-+ * to the @engine->ops vector will be made.
-+ *
-+ * The only exception is %SIGKILL (and exec or group-exit by another
-+ * thread in the group), which can cause asynchronous @report_death
-+ * and/or @report_reap callbacks even when %UTRACE_STOP was used.
-+ * (In that event, this fails with -%ESRCH or -%EALREADY, see above.)
-+ *
-+ * UTRACE_STOP:
-+ *
-+ * This asks that @target stop running. This returns 0 only if
-+ * @target is already stopped, either for tracing or for job
-+ * control. Then @target will remain stopped until another
-+ * utrace_control() call is made on @engine; @target can be woken
-+ * only by %SIGKILL (or equivalent, such as exec or termination by
-+ * another thread in the same thread group).
-+ *
-+ * This returns -%EINPROGRESS if @target is not already stopped.
-+ * Then the effect is like %UTRACE_REPORT. A @report_quiesce or
-+ * @report_signal callback will be made soon. Your callback can
-+ * then return %UTRACE_STOP to keep @target stopped.
-+ *
-+ * This does not interrupt system calls in progress, including ones
-+ * that sleep for a long time. For that, use %UTRACE_INTERRUPT.
-+ * To interrupt system calls and then keep @target stopped, your
-+ * @report_signal callback can return %UTRACE_STOP.
-+ *
-+ * UTRACE_RESUME:
-+ *
-+ * Just let @target continue running normally, reversing the effect
-+ * of a previous %UTRACE_STOP. If another engine is keeping @target
-+ * stopped, then it remains stopped until all engines let it resume.
-+ * If @target was not stopped, this has no effect.
-+ *
-+ * UTRACE_REPORT:
-+ *
-+ * This is like %UTRACE_RESUME, but also ensures that there will be
-+ * a @report_quiesce or @report_signal callback made soon. If
-+ * @target had been stopped, then there will be a callback before it
-+ * resumes running normally. If another engine is keeping @target
-+ * stopped, then there might be no callbacks until all engines let
-+ * it resume.
-+ *
-+ * Since this is meaningless unless @report_quiesce callbacks will
-+ * be made, it returns -%EINVAL if @engine lacks %UTRACE_EVENT(%QUIESCE).
-+ *
-+ * UTRACE_INTERRUPT:
-+ *
-+ * This is like %UTRACE_REPORT, but ensures that @target will make a
-+ * @report_signal callback before it resumes or delivers signals.
-+ * If @target was in a system call or about to enter one, work in
-+ * progress will be interrupted as if by %SIGSTOP. If another
-+ * engine is keeping @target stopped, then there might be no
-+ * callbacks until all engines let it resume.
-+ *
-+ * This gives @engine an opportunity to introduce a forced signal
-+ * disposition via its @report_signal callback.
-+ *
-+ * UTRACE_SINGLESTEP:
-+ *
-+ * It's invalid to use this unless arch_has_single_step() returned true.
-+ * This is like %UTRACE_RESUME, but resumes for one user instruction only.
-+ *
-+ * Note that passing %UTRACE_SINGLESTEP or %UTRACE_BLOCKSTEP to
-+ * utrace_control() or returning it from an event callback alone does
-+ * not necessarily ensure that stepping will be enabled. If there are
-+ * more callbacks made to any engine before returning to user mode,
-+ * then the resume action is chosen only by the last set of callbacks.
-+ * To be sure, enable %UTRACE_EVENT(%QUIESCE) and look for the
-+ * @report_quiesce callback with a zero event mask, or the
-+ * @report_signal callback with %UTRACE_SIGNAL_REPORT.
-+ *
-+ * Since this is not robust unless @report_quiesce callbacks will
-+ * be made, it returns -%EINVAL if @engine lacks %UTRACE_EVENT(%QUIESCE).
-+ *
-+ * UTRACE_BLOCKSTEP:
-+ *
-+ * It's invalid to use this unless arch_has_block_step() returned true.
-+ * This is like %UTRACE_SINGLESTEP, but resumes for one whole basic
-+ * block of user instructions.
-+ *
-+ * Since this is not robust unless @report_quiesce callbacks will
-+ * be made, it returns -%EINVAL if @engine lacks %UTRACE_EVENT(%QUIESCE).
++/*
++ * If the target is not dead it should not be in tracing
++ * stop any more. Wake it unless it's in job control stop.
++ */
++static void utrace_wakeup(struct task_struct *target, struct utrace *utrace)
++{
++ lockdep_assert_held(&utrace->lock);
++ spin_lock_irq(&target->sighand->siglock);
++ wake_up_quiescent(target, __TASK_UTRACED);
++ spin_unlock_irq(&target->sighand->siglock);
++}
++
++/*
++ * This is called when there might be some detached engines on the list or
++ * some stale bits in @task->utrace_flags. Clean them up and recompute the
++ * flags. Returns true if we're now fully detached.
+ *
-+ * %UTRACE_BLOCKSTEP devolves to %UTRACE_SINGLESTEP when another
-+ * tracing engine is using %UTRACE_SINGLESTEP at the same time.
++ * Called with @utrace->lock held, returns with it released.
++ * After this returns, @utrace might be freed if everything detached.
+ */
-+int utrace_control(struct task_struct *target,
-+ struct utrace_engine *engine,
-+ enum utrace_resume_action action)
++static bool utrace_reset(struct task_struct *task, struct utrace *utrace)
++ __releases(utrace->lock)
+{
-+ struct utrace *utrace;
-+ bool reset;
-+ int ret;
++ struct utrace_engine *engine, *next;
++ unsigned long flags = 0;
++ LIST_HEAD(detached);
+
-+ if (unlikely(action >= UTRACE_RESUME_MAX)) {
-+ WARN(1, "invalid action argument to utrace_control()!");
-+ return -EINVAL;
-+ }
++ splice_attaching(utrace);
+
+ /*
-+ * This is a sanity check for a programming error in the caller.
-+ * Their request can only work properly in all cases by relying on
-+ * a follow-up callback, but they didn't set one up! This check
-+ * doesn't do locking, but it shouldn't matter. The caller has to
-+ * be synchronously sure the callback is set up to be operating the
-+ * interface properly.
++ * Update the set of events of interest from the union
++ * of the interests of the remaining tracing engines.
++ * For any engine marked detached, remove it from the list.
++ * We'll collect them on the detached list.
+ */
-+ if (action >= UTRACE_REPORT && action < UTRACE_RESUME &&
-+ unlikely(!(engine->flags & UTRACE_EVENT(QUIESCE)))) {
-+ WARN(1, "utrace_control() with no QUIESCE callback in place!");
-+ return -EINVAL;
++ list_for_each_entry_safe(engine, next, &utrace->attached, entry) {
++ if (engine->ops == &utrace_detached_ops) {
++ engine->ops = NULL;
++ list_move(&engine->entry, &detached);
++ } else {
++ flags |= engine->flags | UTRACE_EVENT(REAP);
++ }
+ }
+
-+ utrace = get_utrace_lock(target, engine, true);
-+ if (unlikely(IS_ERR(utrace)))
-+ return PTR_ERR(utrace);
++ if (task->exit_state) {
++ /*
++ * Once it's already dead, we never install any flags
++ * except REAP. When ->exit_state is set and events
++ * like DEATH are not set, then they never can be set.
++ * This ensures that utrace_release_task() knows
++ * positively that utrace_report_death() can never run.
++ */
++ BUG_ON(utrace->death);
++ flags &= UTRACE_EVENT(REAP);
++ } else if (!(flags & UTRACE_EVENT_SYSCALL) &&
++ !(task->ptrace & PT_SYSCALL_TRACE) &&
++ test_tsk_thread_flag(task, TIF_SYSCALL_TRACE)) {
++ clear_tsk_thread_flag(task, TIF_SYSCALL_TRACE);
++ }
+
-+ reset = task_is_traced(target);
-+ ret = 0;
++ if (!flags) {
++ /*
++ * No more engines, cleared out the utrace.
++ */
++ utrace->resume = UTRACE_RESUME;
++ utrace->signal_handler = 0;
++ }
+
+ /*
-+ * ->exit_state can change under us, this doesn't matter.
-+ * We do not care about ->exit_state in fact, but we do
-+ * care about ->reap and ->death. If either flag is set,
-+ * we must also see ->exit_state != 0.
++ * If no more engines want it stopped, wake it up.
+ */
-+ if (unlikely(target->exit_state)) {
-+ ret = utrace_control_dead(target, utrace, action);
-+ if (ret) {
-+ spin_unlock(&utrace->lock);
-+ return ret;
-+ }
-+ reset = true;
++ if (task_is_utraced(task) && !(flags & ENGINE_STOP)) {
++ /*
++ * It just resumes, so make sure single-step
++ * is not left set.
++ */
++ if (utrace->resume == UTRACE_RESUME)
++ user_disable_single_step(task);
++ utrace_wakeup(task, utrace);
+ }
+
-+ switch (action) {
-+ case UTRACE_STOP:
-+ mark_engine_wants_stop(target, engine);
-+ if (!reset && !utrace_do_stop(target, utrace))
-+ ret = -EINPROGRESS;
-+ reset = false;
-+ break;
++ /*
++ * In theory spin_lock() doesn't imply rcu_read_lock().
++ * Once we clear ->utrace_flags this task_struct can go away
++ * because tracehook_prepare_release_task() path does not take
++ * utrace->lock when ->utrace_flags == 0.
++ */
++ rcu_read_lock();
++ task->utrace_flags = flags;
++ spin_unlock(&utrace->lock);
++ rcu_read_unlock();
+
-+ case UTRACE_DETACH:
-+ if (engine_wants_stop(engine))
-+ target->utrace_flags &= ~ENGINE_STOP;
-+ mark_engine_detached(engine);
-+ reset = reset || utrace_do_stop(target, utrace);
-+ if (!reset) {
-+ /*
-+ * As in utrace_set_events(), this barrier ensures
-+ * that our engine->flags changes have hit before we
-+ * examine utrace->reporting, pairing with the barrier
-+ * in start_callback(). If @target has not yet hit
-+ * finish_callback() to clear utrace->reporting, we
-+ * might be in the middle of a callback to @engine.
-+ */
-+ smp_mb();
-+ if (utrace->reporting == engine)
-+ ret = -EINPROGRESS;
-+ }
-+ break;
++ put_detached_list(&detached);
+
-+ case UTRACE_RESUME:
-+ clear_engine_wants_stop(engine);
-+ break;
++ return !flags;
++}
+
-+ case UTRACE_BLOCKSTEP:
-+ /*
-+ * Resume from stopped, step one block.
-+ * We fall through to treat it like UTRACE_SINGLESTEP.
-+ */
-+ if (unlikely(!arch_has_block_step())) {
-+ WARN(1, "UTRACE_BLOCKSTEP when !arch_has_block_step()");
-+ action = UTRACE_SINGLESTEP;
-+ }
++void utrace_finish_stop(void)
++{
++ /*
++ * If we were task_is_utraced() and then SIGKILL'ed, make
++ * sure we do nothing until the tracer drops utrace->lock.
++ */
++ if (unlikely(__fatal_signal_pending(current))) {
++ struct utrace *utrace = task_utrace_struct(current);
++ spin_unlock_wait(&utrace->lock);
++ }
++}
+
-+ case UTRACE_SINGLESTEP:
-+ /*
-+ * Resume from stopped, step one instruction.
-+ * We fall through to the UTRACE_REPORT case.
-+ */
-+ if (unlikely(!arch_has_single_step())) {
-+ WARN(1,
-+ "UTRACE_SINGLESTEP when !arch_has_single_step()");
-+ reset = false;
-+ ret = -EOPNOTSUPP;
-+ break;
-+ }
++/*
++ * Perform %UTRACE_STOP, i.e. block in TASK_UTRACED until woken up.
++ * @task == current, @utrace == current->utrace, which is not locked.
++ * Return true if we were woken up by SIGKILL even though some utrace
++ * engine may still want us to stay stopped.
++ */
++static void utrace_stop(struct task_struct *task, struct utrace *utrace,
++ enum utrace_resume_action action)
++{
++relock:
++ spin_lock(&utrace->lock);
+
-+ case UTRACE_REPORT:
++ if (action < utrace->resume) {
+ /*
-+ * Make the thread call tracehook_notify_resume() soon.
-+ * But don't bother if it's already been interrupted.
-+ * In that case, utrace_get_signal() will be reporting soon.
++ * Ensure a reporting pass when we're resumed.
+ */
-+ clear_engine_wants_stop(engine);
-+ if (action < utrace->resume) {
-+ utrace->resume = action;
-+ set_notify_resume(target);
-+ }
-+ break;
++ utrace->resume = action;
++ if (action == UTRACE_INTERRUPT)
++ set_thread_flag(TIF_SIGPENDING);
++ else
++ set_thread_flag(TIF_NOTIFY_RESUME);
++ }
+
-+ case UTRACE_INTERRUPT:
-+ /*
-+ * Make the thread call tracehook_get_signal() soon.
-+ */
-+ clear_engine_wants_stop(engine);
-+ if (utrace->resume == UTRACE_INTERRUPT)
-+ break;
-+ utrace->resume = UTRACE_INTERRUPT;
++ /*
++ * If the ENGINE_STOP bit is clear in utrace_flags, that means
++ * utrace_reset() ran after we processed some UTRACE_STOP return
++ * values from callbacks to get here. If all engines have detached
++ * or resumed us, we don't stop. This check doesn't require
++ * siglock, but it should follow the interrupt/report bookkeeping
++ * steps (this can matter for UTRACE_RESUME but not UTRACE_DETACH).
++ */
++ if (unlikely(!(task->utrace_flags & ENGINE_STOP))) {
++ utrace_reset(task, utrace);
++ if (task->utrace_flags & ENGINE_STOP)
++ goto relock;
++ return;
++ }
+
-+ /*
-+ * If it's not already stopped, interrupt it now. We need
-+ * the siglock here in case it calls recalc_sigpending()
-+ * and clears its own TIF_SIGPENDING. By taking the lock,
-+ * we've serialized any later recalc_sigpending() after our
-+ * setting of utrace->resume to force it on.
-+ */
-+ if (reset) {
-+ /*
-+ * This is really just to keep the invariant that
-+ * TIF_SIGPENDING is set with UTRACE_INTERRUPT.
-+ * When it's stopped, we know it's always going
-+ * through utrace_get_signal() and will recalculate.
-+ */
-+ set_tsk_thread_flag(target, TIF_SIGPENDING);
-+ } else {
-+ struct sighand_struct *sighand;
-+ unsigned long irqflags;
-+ sighand = lock_task_sighand(target, &irqflags);
-+ if (likely(sighand)) {
-+ signal_wake_up(target, 0);
-+ unlock_task_sighand(target, &irqflags);
-+ }
-+ }
-+ break;
++ /*
++ * The siglock protects us against signals. As well as SIGKILL
++ * waking us up, we must synchronize with the signal bookkeeping
++ * for stop signals and SIGCONT.
++ */
++ spin_lock_irq(&task->sighand->siglock);
+
-+ default:
-+ BUG(); /* We checked it on entry. */
++ if (unlikely(__fatal_signal_pending(task))) {
++ spin_unlock_irq(&task->sighand->siglock);
++ spin_unlock(&utrace->lock);
++ return;
+ }
+
++ __set_current_state(TASK_UTRACED);
++
++ spin_unlock_irq(&task->sighand->siglock);
++ spin_unlock(&utrace->lock);
++
++ schedule();
++
++ utrace_finish_stop();
++
+ /*
-+ * Let the thread resume running. If it's not stopped now,
-+ * there is nothing more we need to do.
++ * While in TASK_UTRACED, we were considered "frozen enough".
++ * Now that we woke up, it's crucial if we're supposed to be
++ * frozen that we freeze now before running anything substantial.
+ */
-+ if (reset)
-+ utrace_reset(target, utrace);
-+ else
-+ spin_unlock(&utrace->lock);
++ try_to_freeze();
+
-+ return ret;
++ /*
++ * While we were in TASK_UTRACED, complete_signal() considered
++ * us "uninterested" in signal wakeups. Now make sure our
++ * TIF_SIGPENDING state is correct for normal running.
++ */
++ spin_lock_irq(&task->sighand->siglock);
++ recalc_sigpending();
++ spin_unlock_irq(&task->sighand->siglock);
+}
-+EXPORT_SYMBOL_GPL(utrace_control);
+
-+/**
-+ * utrace_barrier - synchronize with simultaneous tracing callbacks
-+ * @target: thread to affect
-+ * @engine: engine to affect (can be detached)
-+ *
-+ * This blocks while @target might be in the midst of making a callback to
-+ * @engine. It can be interrupted by signals and will return -%ERESTARTSYS.
-+ * A return value of zero means no callback from @target to @engine was
-+ * in progress. Any effect of its return value (such as %UTRACE_STOP) has
-+ * already been applied to @engine.
-+ *
-+ * It's not necessary to keep the @target pointer alive for this call.
-+ * It's only necessary to hold a ref on @engine. This will return
-+ * safely even if @target has been reaped and has no task refs.
-+ *
-+ * A successful return from utrace_barrier() guarantees its ordering
-+ * with respect to utrace_set_events() and utrace_control() calls. If
-+ * @target was not properly stopped, event callbacks just disabled might
-+ * still be in progress; utrace_barrier() waits until there is no chance
-+ * an unwanted callback can be in progress.
++/*
++ * Called by release_task() with @reap set to true.
++ * Called by utrace_report_death() with @reap set to false.
++ * On reap, make report_reap callbacks and clean out @utrace
++ * unless still making callbacks. On death, update bookkeeping
++ * and handle the reap work if release_task() came in first.
+ */
-+int utrace_barrier(struct task_struct *target, struct utrace_engine *engine)
++void utrace_maybe_reap(struct task_struct *target, struct utrace *utrace,
++ bool reap)
+{
-+ struct utrace *utrace;
-+ int ret = -ERESTARTSYS;
++ struct utrace_engine *engine, *next;
++ struct list_head attached;
+
-+ if (unlikely(target == current))
-+ return 0;
++ spin_lock(&utrace->lock);
+
-+ do {
-+ utrace = get_utrace_lock(target, engine, false);
-+ if (unlikely(IS_ERR(utrace))) {
-+ ret = PTR_ERR(utrace);
-+ if (ret != -ERESTARTSYS)
-+ break;
-+ } else {
-+ /*
-+ * All engine state changes are done while
-+ * holding the lock, i.e. before we get here.
-+ * Since we have the lock, we only need to
-+ * worry about @target making a callback.
-+ * When it has entered start_callback() but
-+ * not yet gotten to finish_callback(), we
-+ * will see utrace->reporting == @engine.
-+ * When @target doesn't take the lock, it uses
-+ * barriers to order setting utrace->reporting
-+ * before it examines the engine state.
-+ */
-+ if (utrace->reporting != engine)
-+ ret = 0;
++ if (reap) {
++ /*
++ * If the target will do some final callbacks but hasn't
++ * finished them yet, we know because it clears these event
++ * bits after it's done. Instead of cleaning up here and
++ * requiring utrace_report_death() to cope with it, we
++ * delay the REAP report and the teardown until after the
++ * target finishes its death reports.
++ */
++ utrace->reap = 1;
++
++ if (target->utrace_flags & _UTRACE_DEATH_EVENTS) {
+ spin_unlock(&utrace->lock);
-+ if (!ret)
-+ break;
++ return;
+ }
-+ schedule_timeout_interruptible(1);
-+ } while (!signal_pending(current));
-+
-+ return ret;
-+}
-+EXPORT_SYMBOL_GPL(utrace_barrier);
-+
-+/*
-+ * This is local state used for reporting loops, perhaps optimized away.
-+ */
-+struct utrace_report {
-+ u32 result;
-+ enum utrace_resume_action action;
-+ enum utrace_resume_action resume_action;
-+ bool detaches;
-+ bool spurious;
-+};
-+
-+#define INIT_REPORT(var) \
-+ struct utrace_report var = { \
-+ .action = UTRACE_RESUME, \
-+ .resume_action = UTRACE_RESUME, \
-+ .spurious = true \
-+ }
-+
-+/*
-+ * We are now making the report, so clear the flag saying we need one.
-+ * When there is a new attach, ->pending_attach is set just so we will
-+ * know to do splice_attaching() here before the callback loop.
-+ */
-+static enum utrace_resume_action start_report(struct utrace *utrace)
-+{
-+ enum utrace_resume_action resume = utrace->resume;
-+ if (utrace->pending_attach ||
-+ (resume > UTRACE_INTERRUPT && resume < UTRACE_RESUME)) {
-+ spin_lock(&utrace->lock);
-+ splice_attaching(utrace);
-+ resume = utrace->resume;
-+ if (resume > UTRACE_INTERRUPT)
-+ utrace->resume = UTRACE_RESUME;
-+ spin_unlock(&utrace->lock);
-+ }
-+ return resume;
-+}
-+
-+static inline void finish_report_reset(struct task_struct *task,
-+ struct utrace *utrace,
-+ struct utrace_report *report)
-+{
-+ if (unlikely(report->spurious || report->detaches)) {
-+ spin_lock(&utrace->lock);
-+ if (utrace_reset(task, utrace))
-+ report->action = UTRACE_RESUME;
-+ }
-+}
-+
-+/*
-+ * Complete a normal reporting pass, pairing with a start_report() call.
-+ * This handles any UTRACE_DETACH or UTRACE_REPORT or UTRACE_INTERRUPT
-+ * returns from engine callbacks. If @will_not_stop is true and any
-+ * engine's last callback used UTRACE_STOP, we do UTRACE_REPORT here to
-+ * ensure we stop before user mode. If there were no callbacks made, it
-+ * will recompute @task->utrace_flags to avoid another false-positive.
-+ */
-+static void finish_report(struct task_struct *task, struct utrace *utrace,
-+ struct utrace_report *report, bool will_not_stop)
-+{
-+ enum utrace_resume_action resume = report->action;
-+
-+ if (resume == UTRACE_STOP)
-+ resume = will_not_stop ? UTRACE_REPORT : UTRACE_RESUME;
-+
-+ if (resume < utrace->resume) {
-+ spin_lock(&utrace->lock);
-+ utrace->resume = resume;
-+ if (resume == UTRACE_INTERRUPT)
-+ set_tsk_thread_flag(task, TIF_SIGPENDING);
-+ else
-+ set_tsk_thread_flag(task, TIF_NOTIFY_RESUME);
-+ spin_unlock(&utrace->lock);
-+ }
-+
-+ finish_report_reset(task, utrace, report);
-+}
-+
-+static void finish_callback_report(struct task_struct *task,
-+ struct utrace *utrace,
-+ struct utrace_report *report,
-+ struct utrace_engine *engine,
-+ enum utrace_resume_action action)
-+{
-+ if (action == UTRACE_DETACH) {
++ } else {
+ /*
-+ * By holding the lock here, we make sure that
-+ * utrace_barrier() (really get_utrace_lock()) sees the
-+ * effect of this detach. Otherwise utrace_barrier() could
-+ * return 0 after this callback had returned UTRACE_DETACH.
-+ * This way, a 0 return is an unambiguous indicator that any
-+ * callback returning UTRACE_DETACH has indeed caused detach.
++ * After we unlock with this flag clear, any competing
++ * utrace_control/utrace_set_events calls know that we've
++ * finished our callbacks and any detach bookkeeping.
+ */
-+ spin_lock(&utrace->lock);
-+ engine->ops = &utrace_detached_ops;
-+ spin_unlock(&utrace->lock);
++ utrace->death = 0;
++
++ if (!utrace->reap) {
++ /*
++ * We're just dead, not reaped yet. This will
++ * reset @target->utrace_flags so the later call
++ * with @reap set won't hit the check above.
++ */
++ utrace_reset(target, utrace);
++ return;
++ }
+ }
+
+ /*
-+ * If utrace_control() was used, treat that like UTRACE_DETACH here.
++ * utrace_add_engine() checks ->utrace_flags != 0. Since
++ * @utrace->reap is set, nobody can set or clear UTRACE_EVENT(REAP)
++ * in @engine->flags or change @engine->ops and nobody can change
++ * @utrace->attached after we drop the lock.
+ */
-+ if (engine->ops == &utrace_detached_ops) {
-+ report->detaches = true;
-+ return;
-+ }
++ target->utrace_flags = 0;
+
-+ if (action < report->action)
-+ report->action = action;
++ /*
++ * We clear out @utrace->attached before we drop the lock so
++ * that find_matching_engine() can't come across any old engine
++ * while we are busy tearing it down.
++ */
++ list_replace_init(&utrace->attached, &attached);
++ list_splice_tail_init(&utrace->attaching, &attached);
+
-+ if (action != UTRACE_STOP) {
-+ if (action < report->resume_action)
-+ report->resume_action = action;
++ spin_unlock(&utrace->lock);
+
-+ if (engine_wants_stop(engine)) {
-+ spin_lock(&utrace->lock);
-+ clear_engine_wants_stop(engine);
-+ spin_unlock(&utrace->lock);
-+ }
++ list_for_each_entry_safe(engine, next, &attached, entry) {
++ if (engine->flags & UTRACE_EVENT(REAP))
++ engine->ops->report_reap(engine, target);
+
-+ return;
-+ }
++ engine->ops = NULL;
++ engine->flags = 0;
++ list_del_init(&engine->entry);
+
-+ if (!engine_wants_stop(engine)) {
-+ spin_lock(&utrace->lock);
-+ /*
-+ * If utrace_control() came in and detached us
-+ * before we got the lock, we must not stop now.
-+ */
-+ if (unlikely(engine->ops == &utrace_detached_ops))
-+ report->detaches = true;
-+ else
-+ mark_engine_wants_stop(task, engine);
-+ spin_unlock(&utrace->lock);
++ utrace_engine_put(engine);
+ }
+}
+
+/*
-+ * Apply the return value of one engine callback to @report.
-+ * Returns true if @engine detached and should not get any more callbacks.
++ * You can't do anything to a dead task but detach it.
++ * If release_task() has been called, you can't do that.
++ *
++ * On the exit path, DEATH and QUIESCE event bits are set only
++ * before utrace_report_death() has taken the lock. At that point,
++ * the death report will come soon, so disallow detach until it's
++ * done. This prevents us from racing with it detaching itself.
++ *
++ * Called only when @target->exit_state is nonzero.
+ */
-+static bool finish_callback(struct task_struct *task, struct utrace *utrace,
-+ struct utrace_report *report,
-+ struct utrace_engine *engine,
-+ u32 ret)
++static inline int utrace_control_dead(struct task_struct *target,
++ struct utrace *utrace,
++ enum utrace_resume_action action)
+{
-+ report->result = ret & ~UTRACE_RESUME_MASK;
-+ finish_callback_report(task, utrace, report, engine,
-+ utrace_resume_action(ret));
++ lockdep_assert_held(&utrace->lock);
+
-+ /*
-+ * Now that we have applied the effect of the return value,
-+ * clear this so that utrace_barrier() can stop waiting.
-+ * A subsequent utrace_control() can stop or resume @engine
-+ * and know this was ordered after its callback's action.
-+ *
-+ * We don't need any barriers here because utrace_barrier()
-+ * takes utrace->lock. If we touched engine->flags above,
-+ * the lock guaranteed this change was before utrace_barrier()
-+ * examined utrace->reporting.
-+ */
-+ utrace->reporting = NULL;
++ if (action != UTRACE_DETACH || unlikely(utrace->reap))
++ return -ESRCH;
+
-+ /*
-+ * We've just done an engine callback. These are allowed to sleep,
-+ * though all well-behaved ones restrict that to blocking kalloc()
-+ * or quickly-acquired mutex_lock() and the like. This is a good
-+ * place to make sure tracing engines don't introduce too much
-+ * latency under voluntary preemption.
-+ */
-+ might_sleep();
++ if (unlikely(utrace->death))
++ /*
++ * We have already started the death report. We can't
++ * prevent the report_death and report_reap callbacks,
++ * so tell the caller they will happen.
++ */
++ return -EALREADY;
+
-+ return engine->ops == &utrace_detached_ops;
++ return 0;
+}
+
-+/*
-+ * Start the callbacks for @engine to consider @event (a bit mask).
-+ * This makes the report_quiesce() callback first. If @engine wants
-+ * a specific callback for @event, we return the ops vector to use.
-+ * If not, we return NULL. The return value from the ops->callback
-+ * function called should be passed to finish_callback().
-+ */
-+static const struct utrace_engine_ops *start_callback(
-+ struct utrace *utrace, struct utrace_report *report,
-+ struct utrace_engine *engine, struct task_struct *task,
-+ unsigned long event)
-+{
-+ const struct utrace_engine_ops *ops;
-+ unsigned long want;
-+
-+ /*
-+ * This barrier ensures that we've set utrace->reporting before
-+ * we examine engine->flags or engine->ops. utrace_barrier()
-+ * relies on this ordering to indicate that the effect of any
-+ * utrace_control() and utrace_set_events() calls is in place
-+ * by the time utrace->reporting can be seen to be NULL.
++/**
++ * utrace_control - control a thread being traced by a tracing engine
++ * @target: thread to affect
++ * @engine: attached engine to affect
++ * @action: &enum utrace_resume_action for thread to do
++ *
++ * This is how a tracing engine asks a traced thread to do something.
++ * This call is controlled by the @action argument, which has the
++ * same meaning as the &enum utrace_resume_action value returned by
++ * event reporting callbacks.
++ *
++ * If @target is already dead (@target->exit_state nonzero),
++ * all actions except %UTRACE_DETACH fail with -%ESRCH.
++ *
++ * The following sections describe each option for the @action argument.
++ *
++ * UTRACE_DETACH:
++ *
++ * After this, the @engine data structure is no longer accessible,
++ * and the thread might be reaped. The thread will start running
++ * again if it was stopped and no longer has any attached engines
++ * that want it stopped.
++ *
++ * If the @report_reap callback may already have begun, this fails
++ * with -%ESRCH. If the @report_death callback may already have
++ * begun, this fails with -%EALREADY.
++ *
++ * If @target is not already stopped, then a callback to this engine
++ * might be in progress or about to start on another CPU. If so,
++ * then this returns -%EINPROGRESS; the detach happens as soon as
++ * the pending callback is finished. To synchronize after an
++ * -%EINPROGRESS return, see utrace_barrier().
++ *
++ * If @target is properly stopped before utrace_control() is called,
++ * then after successful return it's guaranteed that no more callbacks
++ * to the @engine->ops vector will be made.
++ *
++ * The only exception is %SIGKILL (and exec or group-exit by another
++ * thread in the group), which can cause asynchronous @report_death
++ * and/or @report_reap callbacks even when %UTRACE_STOP was used.
++ * (In that event, this fails with -%ESRCH or -%EALREADY, see above.)
++ *
++ * UTRACE_STOP:
++ *
++ * This asks that @target stop running. This returns 0 only if
++ * @target is already stopped, either for tracing or for job
++ * control. Then @target will remain stopped until another
++ * utrace_control() call is made on @engine; @target can be woken
++ * only by %SIGKILL (or equivalent, such as exec or termination by
++ * another thread in the same thread group).
++ *
++ * This returns -%EINPROGRESS if @target is not already stopped.
++ * Then the effect is like %UTRACE_REPORT. A @report_quiesce or
++ * @report_signal callback will be made soon. Your callback can
++ * then return %UTRACE_STOP to keep @target stopped.
++ *
++ * This does not interrupt system calls in progress, including ones
++ * that sleep for a long time. For that, use %UTRACE_INTERRUPT.
++ * To interrupt system calls and then keep @target stopped, your
++ * @report_signal callback can return %UTRACE_STOP.
++ *
++ * UTRACE_RESUME:
++ *
++ * Just let @target continue running normally, reversing the effect
++ * of a previous %UTRACE_STOP. If another engine is keeping @target
++ * stopped, then it remains stopped until all engines let it resume.
++ * If @target was not stopped, this has no effect.
++ *
++ * UTRACE_REPORT:
++ *
++ * This is like %UTRACE_RESUME, but also ensures that there will be
++ * a @report_quiesce or @report_signal callback made soon. If
++ * @target had been stopped, then there will be a callback before it
++ * resumes running normally. If another engine is keeping @target
++ * stopped, then there might be no callbacks until all engines let
++ * it resume.
++ *
++ * Since this is meaningless unless @report_quiesce callbacks will
++ * be made, it returns -%EINVAL if @engine lacks %UTRACE_EVENT(%QUIESCE).
++ *
++ * UTRACE_INTERRUPT:
++ *
++ * This is like %UTRACE_REPORT, but ensures that @target will make a
++ * @report_signal callback before it resumes or delivers signals.
++ * If @target was in a system call or about to enter one, work in
++ * progress will be interrupted as if by %SIGSTOP. If another
++ * engine is keeping @target stopped, then there might be no
++ * callbacks until all engines let it resume.
++ *
++ * This gives @engine an opportunity to introduce a forced signal
++ * disposition via its @report_signal callback.
++ *
++ * UTRACE_SINGLESTEP:
++ *
++ * It's invalid to use this unless arch_has_single_step() returned true.
++ * This is like %UTRACE_RESUME, but resumes for one user instruction only.
++ *
++ * Note that passing %UTRACE_SINGLESTEP or %UTRACE_BLOCKSTEP to
++ * utrace_control() or returning it from an event callback alone does
++ * not necessarily ensure that stepping will be enabled. If there are
++ * more callbacks made to any engine before returning to user mode,
++ * then the resume action is chosen only by the last set of callbacks.
++ * To be sure, enable %UTRACE_EVENT(%QUIESCE) and look for the
++ * @report_quiesce callback with a zero event mask, or the
++ * @report_signal callback with %UTRACE_SIGNAL_REPORT.
++ *
++ * Since this is not robust unless @report_quiesce callbacks will
++ * be made, it returns -%EINVAL if @engine lacks %UTRACE_EVENT(%QUIESCE).
++ *
++ * UTRACE_BLOCKSTEP:
++ *
++ * It's invalid to use this unless arch_has_block_step() returned true.
++ * This is like %UTRACE_SINGLESTEP, but resumes for one whole basic
++ * block of user instructions.
++ *
++ * Since this is not robust unless @report_quiesce callbacks will
++ * be made, it returns -%EINVAL if @engine lacks %UTRACE_EVENT(%QUIESCE).
++ *
++ * %UTRACE_BLOCKSTEP devolves to %UTRACE_SINGLESTEP when another
++ * tracing engine is using %UTRACE_SINGLESTEP at the same time.
++ */
++int utrace_control(struct task_struct *target,
++ struct utrace_engine *engine,
++ enum utrace_resume_action action)
++{
++ struct utrace *utrace;
++ bool reset;
++ int ret;
++
++ if (unlikely(action >= UTRACE_RESUME_MAX)) {
++ WARN(1, "invalid action argument to utrace_control()!");
++ return -EINVAL;
++ }
++
++ /*
++ * This is a sanity check for a programming error in the caller.
++ * Their request can only work properly in all cases by relying on
++ * a follow-up callback, but they didn't set one up! This check
++ * doesn't do locking, but it shouldn't matter. The caller has to
++ * be synchronously sure the callback is set up to be operating the
++ * interface properly.
+ */
-+ utrace->reporting = engine;
-+ smp_mb();
++ if (action >= UTRACE_REPORT && action < UTRACE_RESUME &&
++ unlikely(!(engine->flags & UTRACE_EVENT(QUIESCE)))) {
++ WARN(1, "utrace_control() with no QUIESCE callback in place!");
++ return -EINVAL;
++ }
++
++ utrace = get_utrace_lock(target, engine, true);
++ if (unlikely(IS_ERR(utrace)))
++ return PTR_ERR(utrace);
++
++ reset = task_is_utraced(target);
++ ret = 0;
+
+ /*
-+ * This pairs with the barrier in mark_engine_detached().
-+ * It makes sure that we never see the old ops vector with
-+ * the new flags, in case the original vector had no report_quiesce.
++ * ->exit_state can change under us, this doesn't matter.
++ * We do not care about ->exit_state in fact, but we do
++ * care about ->reap and ->death. If either flag is set,
++ * we must also see ->exit_state != 0.
+ */
-+ want = engine->flags;
-+ smp_rmb();
-+ ops = engine->ops;
++ if (unlikely(target->exit_state)) {
++ ret = utrace_control_dead(target, utrace, action);
++ if (ret) {
++ spin_unlock(&utrace->lock);
++ return ret;
++ }
++ reset = true;
++ }
+
-+ if ((want & UTRACE_EVENT(QUIESCE)) || ops == &utrace_detached_ops) {
-+ if (finish_callback(task, utrace, report, engine,
-+ (*ops->report_quiesce)(report->action,
-+ engine, event)))
-+ return NULL;
++ switch (action) {
++ case UTRACE_STOP:
++ mark_engine_wants_stop(target, engine);
++ if (!reset && !utrace_do_stop(target, utrace))
++ ret = -EINPROGRESS;
++ reset = false;
++ break;
+
-+ if (!event) {
-+ /* We only got here to report QUIESCE */
-+ report->spurious = false;
-+ return NULL;
++ case UTRACE_DETACH:
++ if (engine_wants_stop(engine))
++ target->utrace_flags &= ~ENGINE_STOP;
++ mark_engine_detached(engine);
++ reset = reset || utrace_do_stop(target, utrace);
++ if (!reset) {
++ /*
++ * As in utrace_set_events(), this barrier ensures
++ * that our engine->flags changes have hit before we
++ * examine utrace->reporting, pairing with the barrier
++ * in start_callback(). If @target has not yet hit
++ * finish_callback() to clear utrace->reporting, we
++ * might be in the middle of a callback to @engine.
++ */
++ smp_mb();
++ if (utrace->reporting == engine)
++ ret = -EINPROGRESS;
+ }
++ break;
++
++ case UTRACE_RESUME:
++ clear_engine_wants_stop(engine);
++ break;
+
++ case UTRACE_BLOCKSTEP:
+ /*
-+ * finish_callback() reset utrace->reporting after the
-+ * quiesce callback. Now we set it again (as above)
-+ * before re-examining engine->flags, which could have
-+ * been changed synchronously by ->report_quiesce or
-+ * asynchronously by utrace_control() or utrace_set_events().
++ * Resume from stopped, step one block.
++ * We fall through to treat it like UTRACE_SINGLESTEP.
+ */
-+ utrace->reporting = engine;
-+ smp_mb();
-+ want = engine->flags;
-+ }
-+
-+ if (want & ENGINE_STOP)
-+ report->action = UTRACE_STOP;
-+
-+ if (want & event) {
-+ report->spurious = false;
-+ return ops;
-+ }
++ if (unlikely(!arch_has_block_step())) {
++ WARN(1, "UTRACE_BLOCKSTEP when !arch_has_block_step()");
++ action = UTRACE_SINGLESTEP;
++ }
+
-+ utrace->reporting = NULL;
-+ return NULL;
-+}
++ case UTRACE_SINGLESTEP:
++ /*
++ * Resume from stopped, step one instruction.
++ * We fall through to the UTRACE_REPORT case.
++ */
++ if (unlikely(!arch_has_single_step())) {
++ WARN(1,
++ "UTRACE_SINGLESTEP when !arch_has_single_step()");
++ reset = false;
++ ret = -EOPNOTSUPP;
++ break;
++ }
+
-+/*
-+ * Do a normal reporting pass for engines interested in @event.
-+ * @callback is the name of the member in the ops vector, and remaining
-+ * args are the extras it takes after the standard three args.
-+ */
-+#define REPORT_CALLBACKS(rev, task, utrace, report, event, callback, ...) \
-+ do { \
-+ struct utrace_engine *engine; \
-+ const struct utrace_engine_ops *ops; \
-+ list_for_each_entry##rev(engine, &utrace->attached, entry) { \
-+ ops = start_callback(utrace, report, engine, task, \
-+ event); \
-+ if (!ops) \
-+ continue; \
-+ finish_callback(task, utrace, report, engine, \
-+ (*ops->callback)(__VA_ARGS__)); \
-+ } \
-+ } while (0)
-+#define REPORT(task, utrace, report, event, callback, ...) \
-+ do { \
-+ start_report(utrace); \
-+ REPORT_CALLBACKS(, task, utrace, report, event, callback, \
-+ (report)->action, engine, ## __VA_ARGS__); \
-+ finish_report(task, utrace, report, true); \
-+ } while (0)
-+
-+/*
-+ * Called iff UTRACE_EVENT(EXEC) flag is set.
-+ */
-+void utrace_report_exec(struct linux_binfmt *fmt, struct linux_binprm *bprm,
-+ struct pt_regs *regs)
-+{
-+ struct task_struct *task = current;
-+ struct utrace *utrace = task_utrace_struct(task);
-+ INIT_REPORT(report);
-+
-+ REPORT(task, utrace, &report, UTRACE_EVENT(EXEC),
-+ report_exec, fmt, bprm, regs);
-+}
-+
-+static u32 do_report_syscall_entry(struct pt_regs *regs,
-+ struct task_struct *task,
-+ struct utrace *utrace,
-+ struct utrace_report *report,
-+ u32 resume_report)
-+{
-+ start_report(utrace);
-+ REPORT_CALLBACKS(_reverse, task, utrace, report,
-+ UTRACE_EVENT(SYSCALL_ENTRY), report_syscall_entry,
-+ resume_report | report->result | report->action,
-+ engine, regs);
-+ finish_report(task, utrace, report, false);
-+
-+ if (report->action != UTRACE_STOP)
-+ return 0;
-+
-+ utrace_stop(task, utrace, report->resume_action);
++ case UTRACE_REPORT:
++ /*
++ * Make the thread call tracehook_notify_resume() soon.
++ * But don't bother if it's already been interrupted.
++ * In that case, utrace_get_signal() will be reporting soon.
++ */
++ clear_engine_wants_stop(engine);
++ if (action < utrace->resume) {
++ utrace->resume = action;
++ set_notify_resume(target);
++ }
++ break;
+
-+ if (fatal_signal_pending(task)) {
++ case UTRACE_INTERRUPT:
+ /*
-+ * We are continuing despite UTRACE_STOP because of a
-+ * SIGKILL. Don't let the system call actually proceed.
++ * Make the thread call tracehook_get_signal() soon.
+ */
-+ report->result = UTRACE_SYSCALL_ABORT;
-+ } else if (utrace->resume <= UTRACE_REPORT) {
++ clear_engine_wants_stop(engine);
++ if (utrace->resume == UTRACE_INTERRUPT)
++ break;
++ utrace->resume = UTRACE_INTERRUPT;
++
+ /*
-+ * If we've been asked for another report after our stop,
-+ * go back to report (and maybe stop) again before we run
-+ * the system call. The second (and later) reports are
-+ * marked with the UTRACE_SYSCALL_RESUMED flag so that
-+ * engines know this is a second report at the same
-+ * entry. This gives them the chance to examine the
-+ * registers anew after they might have been changed
-+ * while we were stopped.
++ * If it's not already stopped, interrupt it now. We need
++ * the siglock here in case it calls recalc_sigpending()
++ * and clears its own TIF_SIGPENDING. By taking the lock,
++ * we've serialized any later recalc_sigpending() after our
++ * setting of utrace->resume to force it on.
+ */
-+ report->detaches = false;
-+ report->spurious = true;
-+ report->action = report->resume_action = UTRACE_RESUME;
-+ return UTRACE_SYSCALL_RESUMED;
++ if (reset) {
++ /*
++ * This is really just to keep the invariant that
++ * TIF_SIGPENDING is set with UTRACE_INTERRUPT.
++ * When it's stopped, we know it's always going
++ * through utrace_get_signal() and will recalculate.
++ */
++ set_tsk_thread_flag(target, TIF_SIGPENDING);
++ } else {
++ struct sighand_struct *sighand;
++ unsigned long irqflags;
++ sighand = lock_task_sighand(target, &irqflags);
++ if (likely(sighand)) {
++ signal_wake_up(target, 0);
++ unlock_task_sighand(target, &irqflags);
++ }
++ }
++ break;
++
++ default:
++ BUG(); /* We checked it on entry. */
+ }
+
-+ return 0;
++ /*
++ * Let the thread resume running. If it's not stopped now,
++ * there is nothing more we need to do.
++ */
++ if (reset)
++ utrace_reset(target, utrace);
++ else
++ spin_unlock(&utrace->lock);
++
++ return ret;
+}
++EXPORT_SYMBOL_GPL(utrace_control);
+
-+/*
-+ * Called iff UTRACE_EVENT(SYSCALL_ENTRY) flag is set.
-+ * Return true to prevent the system call.
++/**
++ * utrace_barrier - synchronize with simultaneous tracing callbacks
++ * @target: thread to affect
++ * @engine: engine to affect (can be detached)
++ *
++ * This blocks while @target might be in the midst of making a callback to
++ * @engine. It can be interrupted by signals and will return -%ERESTARTSYS.
++ * A return value of zero means no callback from @target to @engine was
++ * in progress. Any effect of its return value (such as %UTRACE_STOP) has
++ * already been applied to @engine.
++ *
++ * It's not necessary to keep the @target pointer alive for this call.
++ * It's only necessary to hold a ref on @engine. This will return
++ * safely even if @target has been reaped and has no task refs.
++ *
++ * A successful return from utrace_barrier() guarantees its ordering
++ * with respect to utrace_set_events() and utrace_control() calls. If
++ * @target was not properly stopped, event callbacks just disabled might
++ * still be in progress; utrace_barrier() waits until there is no chance
++ * an unwanted callback can be in progress.
+ */
-+bool utrace_report_syscall_entry(struct pt_regs *regs)
++int utrace_barrier(struct task_struct *target, struct utrace_engine *engine)
+{
-+ struct task_struct *task = current;
-+ struct utrace *utrace = task_utrace_struct(task);
-+ INIT_REPORT(report);
-+ u32 resume_report = 0;
++ struct utrace *utrace;
++ int ret = -ERESTARTSYS;
++
++ if (unlikely(target == current))
++ return 0;
+
+ do {
-+ resume_report = do_report_syscall_entry(regs, task, utrace,
-+ &report, resume_report);
-+ } while (resume_report);
++ utrace = get_utrace_lock(target, engine, false);
++ if (unlikely(IS_ERR(utrace))) {
++ ret = PTR_ERR(utrace);
++ if (ret != -ERESTARTSYS)
++ break;
++ } else {
++ /*
++ * All engine state changes are done while
++ * holding the lock, i.e. before we get here.
++ * Since we have the lock, we only need to
++ * worry about @target making a callback.
++ * When it has entered start_callback() but
++ * not yet gotten to finish_callback(), we
++ * will see utrace->reporting == @engine.
++ * When @target doesn't take the lock, it uses
++ * barriers to order setting utrace->reporting
++ * before it examines the engine state.
++ */
++ if (utrace->reporting != engine)
++ ret = 0;
++ spin_unlock(&utrace->lock);
++ if (!ret)
++ break;
++ }
++ schedule_timeout_interruptible(1);
++ } while (!signal_pending(current));
+
-+ return utrace_syscall_action(report.result) == UTRACE_SYSCALL_ABORT;
++ return ret;
+}
++EXPORT_SYMBOL_GPL(utrace_barrier);
+
+/*
-+ * Called iff UTRACE_EVENT(SYSCALL_EXIT) flag is set.
++ * This is local state used for reporting loops, perhaps optimized away.
+ */
-+void utrace_report_syscall_exit(struct pt_regs *regs)
-+{
-+ struct task_struct *task = current;
-+ struct utrace *utrace = task_utrace_struct(task);
-+ INIT_REPORT(report);
++struct utrace_report {
++ u32 result;
++ enum utrace_resume_action action;
++ enum utrace_resume_action resume_action;
++ bool detaches;
++ bool spurious;
++};
+
-+ REPORT(task, utrace, &report, UTRACE_EVENT(SYSCALL_EXIT),
-+ report_syscall_exit, regs);
-+}
++#define INIT_REPORT(var) \
++ struct utrace_report var = { \
++ .action = UTRACE_RESUME, \
++ .resume_action = UTRACE_RESUME, \
++ .spurious = true \
++ }
+
+/*
-+ * Called iff UTRACE_EVENT(CLONE) flag is set.
-+ * This notification call blocks the wake_up_new_task call on the child.
-+ * So we must not quiesce here. tracehook_report_clone_complete will do
-+ * a quiescence check momentarily.
++ * We are now making the report, so clear the flag saying we need one.
++ * When there is a new attach, ->pending_attach is set just so we will
++ * know to do splice_attaching() here before the callback loop.
+ */
-+void utrace_report_clone(unsigned long clone_flags, struct task_struct *child)
++static enum utrace_resume_action start_report(struct utrace *utrace)
+{
-+ struct task_struct *task = current;
-+ struct utrace *utrace = task_utrace_struct(task);
-+ INIT_REPORT(report);
-+
-+ /*
-+ * We don't use the REPORT() macro here, because we need
-+ * to clear utrace->cloning before finish_report().
-+ * After finish_report(), utrace can be a stale pointer
-+ * in cases when report.action is still UTRACE_RESUME.
-+ */
-+ start_report(utrace);
-+ utrace->cloning = child;
-+
-+ REPORT_CALLBACKS(, task, utrace, &report,
-+ UTRACE_EVENT(CLONE), report_clone,
-+ report.action, engine, clone_flags, child);
++ enum utrace_resume_action resume = utrace->resume;
++ if (utrace->pending_attach ||
++ (resume > UTRACE_INTERRUPT && resume < UTRACE_RESUME)) {
++ spin_lock(&utrace->lock);
++ splice_attaching(utrace);
++ resume = utrace->resume;
++ if (resume > UTRACE_INTERRUPT)
++ utrace->resume = UTRACE_RESUME;
++ spin_unlock(&utrace->lock);
++ }
++ return resume;
++}
+
-+ utrace->cloning = NULL;
-+ finish_report(task, utrace, &report, !(clone_flags & CLONE_VFORK));
-+
-+ /*
-+ * For a vfork, we will go into an uninterruptible block waiting
-+ * for the child. We need UTRACE_STOP to happen before this, not
-+ * after. For CLONE_VFORK, utrace_finish_vfork() will be called.
-+ */
-+ if (report.action == UTRACE_STOP && (clone_flags & CLONE_VFORK)) {
++static inline void finish_report_reset(struct task_struct *task,
++ struct utrace *utrace,
++ struct utrace_report *report)
++{
++ if (unlikely(report->spurious || report->detaches)) {
+ spin_lock(&utrace->lock);
-+ utrace->vfork_stop = 1;
-+ spin_unlock(&utrace->lock);
++ if (utrace_reset(task, utrace))
++ report->action = UTRACE_RESUME;
+ }
+}
+
+/*
-+ * We're called after utrace_report_clone() for a CLONE_VFORK.
-+ * If UTRACE_STOP was left from the clone report, we stop here.
-+ * After this, we'll enter the uninterruptible wait_for_completion()
-+ * waiting for the child.
++ * Complete a normal reporting pass, pairing with a start_report() call.
++ * This handles any UTRACE_DETACH or UTRACE_REPORT or UTRACE_INTERRUPT
++ * returns from engine callbacks. If @will_not_stop is true and any
++ * engine's last callback used UTRACE_STOP, we do UTRACE_REPORT here to
++ * ensure we stop before user mode. If there were no callbacks made, it
++ * will recompute @task->utrace_flags to avoid another false-positive.
+ */
-+void utrace_finish_vfork(struct task_struct *task)
++static void finish_report(struct task_struct *task, struct utrace *utrace,
++ struct utrace_report *report, bool will_not_stop)
+{
-+ struct utrace *utrace = task_utrace_struct(task);
++ enum utrace_resume_action resume = report->action;
+
-+ if (utrace->vfork_stop) {
++ if (resume == UTRACE_STOP)
++ resume = will_not_stop ? UTRACE_REPORT : UTRACE_RESUME;
++
++ if (resume < utrace->resume) {
+ spin_lock(&utrace->lock);
-+ utrace->vfork_stop = 0;
++ utrace->resume = resume;
++ if (resume == UTRACE_INTERRUPT)
++ set_tsk_thread_flag(task, TIF_SIGPENDING);
++ else
++ set_tsk_thread_flag(task, TIF_NOTIFY_RESUME);
+ spin_unlock(&utrace->lock);
-+ utrace_stop(task, utrace, UTRACE_RESUME); /* XXX */
+ }
++
++ finish_report_reset(task, utrace, report);
+}
+
-+/*
-+ * Called iff UTRACE_EVENT(JCTL) flag is set.
-+ *
-+ * Called with siglock held.
-+ */
-+void utrace_report_jctl(int notify, int what)
++static void finish_callback_report(struct task_struct *task,
++ struct utrace *utrace,
++ struct utrace_report *report,
++ struct utrace_engine *engine,
++ enum utrace_resume_action action)
+{
-+ struct task_struct *task = current;
-+ struct utrace *utrace = task_utrace_struct(task);
-+ INIT_REPORT(report);
++ if (action == UTRACE_DETACH) {
++ /*
++ * By holding the lock here, we make sure that
++ * utrace_barrier() (really get_utrace_lock()) sees the
++ * effect of this detach. Otherwise utrace_barrier() could
++ * return 0 after this callback had returned UTRACE_DETACH.
++ * This way, a 0 return is an unambiguous indicator that any
++ * callback returning UTRACE_DETACH has indeed caused detach.
++ */
++ spin_lock(&utrace->lock);
++ engine->ops = &utrace_detached_ops;
++ spin_unlock(&utrace->lock);
++ }
+
-+ spin_unlock_irq(&task->sighand->siglock);
++ /*
++ * If utrace_control() was used, treat that like UTRACE_DETACH here.
++ */
++ if (engine->ops == &utrace_detached_ops) {
++ report->detaches = true;
++ return;
++ }
+
-+ REPORT(task, utrace, &report, UTRACE_EVENT(JCTL),
-+ report_jctl, what, notify);
++ if (action < report->action)
++ report->action = action;
+
-+ spin_lock_irq(&task->sighand->siglock);
-+}
++ if (action != UTRACE_STOP) {
++ if (action < report->resume_action)
++ report->resume_action = action;
+
-+/*
-+ * Called iff UTRACE_EVENT(EXIT) flag is set.
-+ */
-+void utrace_report_exit(long *exit_code)
-+{
-+ struct task_struct *task = current;
-+ struct utrace *utrace = task_utrace_struct(task);
-+ INIT_REPORT(report);
-+ long orig_code = *exit_code;
++ if (engine_wants_stop(engine)) {
++ spin_lock(&utrace->lock);
++ clear_engine_wants_stop(engine);
++ spin_unlock(&utrace->lock);
++ }
+
-+ REPORT(task, utrace, &report, UTRACE_EVENT(EXIT),
-+ report_exit, orig_code, exit_code);
++ return;
++ }
+
-+ if (report.action == UTRACE_STOP)
-+ utrace_stop(task, utrace, report.resume_action);
++ if (!engine_wants_stop(engine)) {
++ spin_lock(&utrace->lock);
++ /*
++ * If utrace_control() came in and detached us
++ * before we got the lock, we must not stop now.
++ */
++ if (unlikely(engine->ops == &utrace_detached_ops))
++ report->detaches = true;
++ else
++ mark_engine_wants_stop(task, engine);
++ spin_unlock(&utrace->lock);
++ }
+}
+
+/*
-+ * Called iff UTRACE_EVENT(DEATH) or UTRACE_EVENT(QUIESCE) flag is set.
-+ *
-+ * It is always possible that we are racing with utrace_release_task here.
-+ * For this reason, utrace_release_task checks for the event bits that get
-+ * us here, and delays its cleanup for us to do.
++ * Apply the return value of one engine callback to @report.
++ * Returns true if @engine detached and should not get any more callbacks.
+ */
-+void utrace_report_death(struct task_struct *task, struct utrace *utrace,
-+ bool group_dead, int signal)
++static bool finish_callback(struct task_struct *task, struct utrace *utrace,
++ struct utrace_report *report,
++ struct utrace_engine *engine,
++ u32 ret)
+{
-+ INIT_REPORT(report);
-+
-+ BUG_ON(!task->exit_state);
++ report->result = ret & ~UTRACE_RESUME_MASK;
++ finish_callback_report(task, utrace, report, engine,
++ utrace_resume_action(ret));
+
+ /*
-+ * We are presently considered "quiescent"--which is accurate
-+ * inasmuch as we won't run any more user instructions ever again.
-+ * But for utrace_control and utrace_set_events to be robust, they
-+ * must be sure whether or not we will run any more callbacks. If
-+ * a call comes in before we do, taking the lock here synchronizes
-+ * us so we don't run any callbacks just disabled. Calls that come
-+ * in while we're running the callbacks will see the exit.death
-+ * flag and know that we are not yet fully quiescent for purposes
-+ * of detach bookkeeping.
++ * Now that we have applied the effect of the return value,
++ * clear this so that utrace_barrier() can stop waiting.
++ * A subsequent utrace_control() can stop or resume @engine
++ * and know this was ordered after its callback's action.
++ *
++ * We don't need any barriers here because utrace_barrier()
++ * takes utrace->lock. If we touched engine->flags above,
++ * the lock guaranteed this change was before utrace_barrier()
++ * examined utrace->reporting.
+ */
-+ spin_lock(&utrace->lock);
-+ BUG_ON(utrace->death);
-+ utrace->death = 1;
-+ utrace->resume = UTRACE_RESUME;
-+ splice_attaching(utrace);
-+ spin_unlock(&utrace->lock);
++ utrace->reporting = NULL;
+
-+ REPORT_CALLBACKS(, task, utrace, &report, UTRACE_EVENT(DEATH),
-+ report_death, engine, group_dead, signal);
++ /*
++ * We've just done an engine callback. These are allowed to sleep,
++ * though all well-behaved ones restrict that to blocking kalloc()
++ * or quickly-acquired mutex_lock() and the like. This is a good
++ * place to make sure tracing engines don't introduce too much
++ * latency under voluntary preemption.
++ */
++ might_sleep();
+
-+ utrace_maybe_reap(task, utrace, false);
++ return engine->ops == &utrace_detached_ops;
+}
+
+/*
-+ * Finish the last reporting pass before returning to user mode.
++ * Start the callbacks for @engine to consider @event (a bit mask).
++ * This makes the report_quiesce() callback first. If @engine wants
++ * a specific callback for @event, we return the ops vector to use.
++ * If not, we return NULL. The return value from the ops->callback
++ * function called should be passed to finish_callback().
+ */
-+static void finish_resume_report(struct task_struct *task,
-+ struct utrace *utrace,
-+ struct utrace_report *report)
++static const struct utrace_engine_ops *start_callback(
++ struct utrace *utrace, struct utrace_report *report,
++ struct utrace_engine *engine, struct task_struct *task,
++ unsigned long event)
+{
-+ finish_report_reset(task, utrace, report);
++ const struct utrace_engine_ops *ops;
++ unsigned long want;
+
-+ switch (report->action) {
-+ case UTRACE_STOP:
-+ utrace_stop(task, utrace, report->resume_action);
-+ break;
++ /*
++ * This barrier ensures that we've set utrace->reporting before
++ * we examine engine->flags or engine->ops. utrace_barrier()
++ * relies on this ordering to indicate that the effect of any
++ * utrace_control() and utrace_set_events() calls is in place
++ * by the time utrace->reporting can be seen to be NULL.
++ */
++ utrace->reporting = engine;
++ smp_mb();
+
-+ case UTRACE_INTERRUPT:
-+ if (!signal_pending(task))
-+ set_tsk_thread_flag(task, TIF_SIGPENDING);
-+ break;
++ /*
++ * This pairs with the barrier in mark_engine_detached().
++ * It makes sure that we never see the old ops vector with
++ * the new flags, in case the original vector had no report_quiesce.
++ */
++ want = engine->flags;
++ smp_rmb();
++ ops = engine->ops;
+
-+ case UTRACE_BLOCKSTEP:
-+ if (likely(arch_has_block_step())) {
-+ user_enable_block_step(task);
-+ break;
++ if ((want & UTRACE_EVENT(QUIESCE)) || ops == &utrace_detached_ops) {
++ if (finish_callback(task, utrace, report, engine,
++ (*ops->report_quiesce)(report->action,
++ engine, event)))
++ return NULL;
++
++ if (!event) {
++ /* We only got here to report QUIESCE */
++ report->spurious = false;
++ return NULL;
+ }
+
+ /*
-+ * This means some callback is to blame for failing
-+ * to check arch_has_block_step() itself. Warn and
-+ * then fall through to treat it as SINGLESTEP.
++ * finish_callback() reset utrace->reporting after the
++ * quiesce callback. Now we set it again (as above)
++ * before re-examining engine->flags, which could have
++ * been changed synchronously by ->report_quiesce or
++ * asynchronously by utrace_control() or utrace_set_events().
+ */
-+ WARN(1, "UTRACE_BLOCKSTEP when !arch_has_block_step()");
++ utrace->reporting = engine;
++ smp_mb();
++ want = engine->flags;
++ }
+
-+ case UTRACE_SINGLESTEP:
-+ if (likely(arch_has_single_step())) {
-+ user_enable_single_step(task);
-+ } else {
-+ /*
-+ * This means some callback is to blame for failing
-+ * to check arch_has_single_step() itself. Spew
-+ * about it so the loser will fix his module.
-+ */
-+ WARN(1,
-+ "UTRACE_SINGLESTEP when !arch_has_single_step()");
-+ }
-+ break;
++ if (want & ENGINE_STOP)
++ report->action = UTRACE_STOP;
+
-+ case UTRACE_REPORT:
-+ case UTRACE_RESUME:
-+ default:
-+ user_disable_single_step(task);
-+ break;
++ if (want & event) {
++ report->spurious = false;
++ return ops;
+ }
++
++ utrace->reporting = NULL;
++ return NULL;
+}
+
+/*
-+ * This is called when TIF_NOTIFY_RESUME had been set (and is now clear).
-+ * We are close to user mode, and this is the place to report or stop.
-+ * When we return, we're going to user mode or into the signals code.
++ * Do a normal reporting pass for engines interested in @event.
++ * @callback is the name of the member in the ops vector, and remaining
++ * args are the extras it takes after the standard three args.
+ */
-+void utrace_resume(struct task_struct *task, struct pt_regs *regs)
++#define REPORT_CALLBACKS(rev, task, utrace, report, event, callback, ...) \
++ do { \
++ struct utrace_engine *engine; \
++ const struct utrace_engine_ops *ops; \
++ list_for_each_entry##rev(engine, &utrace->attached, entry) { \
++ ops = start_callback(utrace, report, engine, task, \
++ event); \
++ if (!ops) \
++ continue; \
++ finish_callback(task, utrace, report, engine, \
++ (*ops->callback)(__VA_ARGS__)); \
++ } \
++ } while (0)
++#define REPORT(task, utrace, report, event, callback, ...) \
++ do { \
++ start_report(utrace); \
++ REPORT_CALLBACKS(, task, utrace, report, event, callback, \
++ (report)->action, engine, ## __VA_ARGS__); \
++ finish_report(task, utrace, report, true); \
++ } while (0)
++
++/*
++ * Called iff UTRACE_EVENT(EXEC) flag is set.
++ */
++void utrace_report_exec(struct linux_binfmt *fmt, struct linux_binprm *bprm,
++ struct pt_regs *regs)
+{
++ struct task_struct *task = current;
+ struct utrace *utrace = task_utrace_struct(task);
+ INIT_REPORT(report);
-+ struct utrace_engine *engine;
+
-+ /*
-+ * Some machines get here with interrupts disabled. The same arch
-+ * code path leads to calling into get_signal_to_deliver(), which
-+ * implicitly reenables them by virtue of spin_unlock_irq.
-+ */
-+ local_irq_enable();
++ REPORT(task, utrace, &report, UTRACE_EVENT(EXEC),
++ report_exec, fmt, bprm, regs);
++}
+
-+ /*
-+ * If this flag is still set it's because there was a signal
-+ * handler setup done but no report_signal following it. Clear
-+ * the flag before we get to user so it doesn't confuse us later.
-+ */
-+ if (unlikely(utrace->signal_handler)) {
-+ spin_lock(&utrace->lock);
-+ utrace->signal_handler = 0;
-+ spin_unlock(&utrace->lock);
-+ }
++static u32 do_report_syscall_entry(struct pt_regs *regs,
++ struct task_struct *task,
++ struct utrace *utrace,
++ struct utrace_report *report,
++ u32 resume_report)
++{
++ start_report(utrace);
++ REPORT_CALLBACKS(_reverse, task, utrace, report,
++ UTRACE_EVENT(SYSCALL_ENTRY), report_syscall_entry,
++ resume_report | report->result | report->action,
++ engine, regs);
++ finish_report(task, utrace, report, false);
+
-+ /*
-+ * Update our bookkeeping even if there are no callbacks made here.
-+ */
-+ report.action = start_report(utrace);
++ if (report->action != UTRACE_STOP)
++ return 0;
+
-+ switch (report.action) {
-+ case UTRACE_RESUME:
-+ /*
-+ * Anything we might have done was already handled by
-+ * utrace_get_signal(), or this is an entirely spurious
-+ * call. (The arch might use TIF_NOTIFY_RESUME for other
-+ * purposes as well as calling us.)
-+ */
-+ return;
-+ case UTRACE_REPORT:
-+ if (unlikely(!(task->utrace_flags & UTRACE_EVENT(QUIESCE))))
-+ break;
++ utrace_stop(task, utrace, report->resume_action);
++
++ if (fatal_signal_pending(task)) {
+ /*
-+ * Do a simple reporting pass, with no specific
-+ * callback after report_quiesce.
++ * We are continuing despite UTRACE_STOP because of a
++ * SIGKILL. Don't let the system call actually proceed.
+ */
-+ report.action = UTRACE_RESUME;
-+ list_for_each_entry(engine, &utrace->attached, entry)
-+ start_callback(utrace, &report, engine, task, 0);
-+ break;
-+ default:
++ report->result = UTRACE_SYSCALL_ABORT;
++ } else if (utrace->resume <= UTRACE_REPORT) {
+ /*
-+ * Even if this report was truly spurious, there is no need
-+ * for utrace_reset() now. TIF_NOTIFY_RESUME was already
-+ * cleared--it doesn't stay spuriously set.
++ * If we've been asked for another report after our stop,
++ * go back to report (and maybe stop) again before we run
++ * the system call. The second (and later) reports are
++ * marked with the UTRACE_SYSCALL_RESUMED flag so that
++ * engines know this is a second report at the same
++ * entry. This gives them the chance to examine the
++ * registers anew after they might have been changed
++ * while we were stopped.
+ */
-+ report.spurious = false;
-+ break;
++ report->detaches = false;
++ report->spurious = true;
++ report->action = report->resume_action = UTRACE_RESUME;
++ return UTRACE_SYSCALL_RESUMED;
+ }
+
-+ /*
-+ * Finish the report and either stop or get ready to resume.
-+ * If utrace->resume was not UTRACE_REPORT, this applies its
-+ * effect now (i.e. step or interrupt).
-+ */
-+ finish_resume_report(task, utrace, &report);
++ return 0;
+}
+
+/*
-+ * Return true if current has forced signal_pending().
-+ *
-+ * This is called only when current->utrace_flags is nonzero, so we know
-+ * that current->utrace must be set. It's not inlined in tracehook.h
-+ * just so that struct utrace can stay opaque outside this file.
++ * Called iff UTRACE_EVENT(SYSCALL_ENTRY) flag is set.
++ * Return true to prevent the system call.
+ */
-+bool utrace_interrupt_pending(void)
++bool utrace_report_syscall_entry(struct pt_regs *regs)
+{
-+ return task_utrace_struct(current)->resume == UTRACE_INTERRUPT;
++ struct task_struct *task = current;
++ struct utrace *utrace = task_utrace_struct(task);
++ INIT_REPORT(report);
++ u32 resume_report = 0;
++
++ do {
++ resume_report = do_report_syscall_entry(regs, task, utrace,
++ &report, resume_report);
++ } while (resume_report);
++
++ return utrace_syscall_action(report.result) == UTRACE_SYSCALL_ABORT;
+}
+
+/*
-+ * Take the siglock and push @info back on our queue.
-+ * Returns with @task->sighand->siglock held.
++ * Called iff UTRACE_EVENT(SYSCALL_EXIT) flag is set.
+ */
-+static void push_back_signal(struct task_struct *task, siginfo_t *info)
-+ __acquires(task->sighand->siglock)
++void utrace_report_syscall_exit(struct pt_regs *regs)
+{
-+ struct sigqueue *q;
-+
-+ if (unlikely(!info->si_signo)) { /* Oh, a wise guy! */
-+ spin_lock_irq(&task->sighand->siglock);
-+ return;
-+ }
-+
-+ q = sigqueue_alloc();
-+ if (likely(q)) {
-+ q->flags = 0;
-+ copy_siginfo(&q->info, info);
-+ }
-+
-+ spin_lock_irq(&task->sighand->siglock);
-+
-+ sigaddset(&task->pending.signal, info->si_signo);
-+ if (likely(q))
-+ list_add(&q->list, &task->pending.list);
++ struct task_struct *task = current;
++ struct utrace *utrace = task_utrace_struct(task);
++ INIT_REPORT(report);
+
-+ set_tsk_thread_flag(task, TIF_SIGPENDING);
++ REPORT(task, utrace, &report, UTRACE_EVENT(SYSCALL_EXIT),
++ report_syscall_exit, regs);
+}
+
+/*
-+ * This is the hook from the signals code, called with the siglock held.
-+ * Here is the ideal place to stop. We also dequeue and intercept signals.
++ * Called iff UTRACE_EVENT(CLONE) flag is set.
++ * This notification call blocks the wake_up_new_task call on the child.
++ * So we must not quiesce here. tracehook_report_clone_complete will do
++ * a quiescence check momentarily.
+ */
-+int utrace_get_signal(struct task_struct *task, struct pt_regs *regs,
-+ siginfo_t *info, struct k_sigaction *return_ka)
-+ __releases(task->sighand->siglock)
-+ __acquires(task->sighand->siglock)
++void utrace_report_clone(unsigned long clone_flags, struct task_struct *child)
+{
-+ struct utrace *utrace;
-+ struct k_sigaction *ka;
++ struct task_struct *task = current;
++ struct utrace *utrace = task_utrace_struct(task);
+ INIT_REPORT(report);
-+ struct utrace_engine *engine;
-+ const struct utrace_engine_ops *ops;
-+ unsigned long event, want;
-+ u32 ret;
-+ int signr;
+
-+ utrace = task_utrace_struct(task);
-+ if (utrace->resume < UTRACE_RESUME ||
-+ utrace->pending_attach || utrace->signal_handler) {
-+ enum utrace_resume_action resume;
++ /*
++ * We don't use the REPORT() macro here, because we need
++ * to clear utrace->cloning before finish_report().
++ * After finish_report(), utrace can be a stale pointer
++ * in cases when report.action is still UTRACE_RESUME.
++ */
++ start_report(utrace);
++ utrace->cloning = child;
+
-+ /*
-+ * We've been asked for an explicit report before we
-+ * even check for pending signals.
-+ */
++ REPORT_CALLBACKS(, task, utrace, &report,
++ UTRACE_EVENT(CLONE), report_clone,
++ report.action, engine, clone_flags, child);
+
-+ spin_unlock_irq(&task->sighand->siglock);
++ utrace->cloning = NULL;
++ finish_report(task, utrace, &report, !(clone_flags & CLONE_VFORK));
+
++ /*
++ * For a vfork, we will go into an uninterruptible block waiting
++ * for the child. We need UTRACE_STOP to happen before this, not
++ * after. For CLONE_VFORK, utrace_finish_vfork() will be called.
++ */
++ if (report.action == UTRACE_STOP && (clone_flags & CLONE_VFORK)) {
+ spin_lock(&utrace->lock);
++ utrace->vfork_stop = 1;
++ spin_unlock(&utrace->lock);
++ }
++}
+
-+ splice_attaching(utrace);
-+
-+ report.result = utrace->signal_handler ?
-+ UTRACE_SIGNAL_HANDLER : UTRACE_SIGNAL_REPORT;
-+ utrace->signal_handler = 0;
-+
-+ resume = utrace->resume;
-+ utrace->resume = UTRACE_RESUME;
++/*
++ * We're called after utrace_report_clone() for a CLONE_VFORK.
++ * If UTRACE_STOP was left from the clone report, we stop here.
++ * After this, we'll enter the uninterruptible wait_for_completion()
++ * waiting for the child.
++ */
++void utrace_finish_vfork(struct task_struct *task)
++{
++ struct utrace *utrace = task_utrace_struct(task);
+
++ if (utrace->vfork_stop) {
++ spin_lock(&utrace->lock);
++ utrace->vfork_stop = 0;
+ spin_unlock(&utrace->lock);
++ utrace_stop(task, utrace, UTRACE_RESUME); /* XXX */
++ }
++}
+
-+ /*
-+ * Make sure signal_pending() only returns true
-+ * if there are real signals pending.
-+ */
-+ if (signal_pending(task)) {
-+ spin_lock_irq(&task->sighand->siglock);
-+ recalc_sigpending();
-+ spin_unlock_irq(&task->sighand->siglock);
-+ }
++/*
++ * Called iff UTRACE_EVENT(JCTL) flag is set.
++ *
++ * Called with siglock held.
++ */
++void utrace_report_jctl(int notify, int what)
++{
++ struct task_struct *task = current;
++ struct utrace *utrace = task_utrace_struct(task);
++ INIT_REPORT(report);
+
-+ if (resume > UTRACE_REPORT) {
-+ /*
-+ * We only got here to process utrace->resume.
-+ * Despite no callbacks, this report is not spurious.
-+ */
-+ report.action = resume;
-+ report.spurious = false;
-+ finish_resume_report(task, utrace, &report);
-+ return -1;
-+ } else if (!(task->utrace_flags & UTRACE_EVENT(QUIESCE))) {
-+ /*
-+ * We only got here to clear utrace->signal_handler.
-+ */
-+ return -1;
-+ }
++ spin_unlock_irq(&task->sighand->siglock);
+
-+ /*
-+ * Do a reporting pass for no signal, just for EVENT(QUIESCE).
-+ * The engine callbacks can fill in *info and *return_ka.
-+ * We'll pass NULL for the @orig_ka argument to indicate
-+ * that there was no original signal.
-+ */
-+ event = 0;
-+ ka = NULL;
-+ memset(return_ka, 0, sizeof *return_ka);
-+ } else if (!(task->utrace_flags & UTRACE_EVENT_SIGNAL_ALL) ||
-+ unlikely(task->signal->group_stop_count)) {
-+ /*
-+ * If no engine is interested in intercepting signals or
-+ * we must stop, let the caller just dequeue them normally
-+ * or participate in group-stop.
-+ */
-+ return 0;
-+ } else {
-+ /*
-+ * Steal the next signal so we can let tracing engines
-+ * examine it. From the signal number and sigaction,
-+ * determine what normal delivery would do. If no
-+ * engine perturbs it, we'll do that by returning the
-+ * signal number after setting *return_ka.
-+ */
-+ signr = dequeue_signal(task, &task->blocked, info);
-+ if (signr == 0)
-+ return signr;
-+ BUG_ON(signr != info->si_signo);
++ REPORT(task, utrace, &report, UTRACE_EVENT(JCTL),
++ report_jctl, what, notify);
+
-+ ka = &task->sighand->action[signr - 1];
-+ *return_ka = *ka;
++ spin_lock_irq(&task->sighand->siglock);
++}
+
-+ /*
-+ * We are never allowed to interfere with SIGKILL.
-+ * Just punt after filling in *return_ka for our caller.
-+ */
-+ if (signr == SIGKILL)
-+ return signr;
++/*
++ * Called iff UTRACE_EVENT(EXIT) flag is set.
++ */
++void utrace_report_exit(long *exit_code)
++{
++ struct task_struct *task = current;
++ struct utrace *utrace = task_utrace_struct(task);
++ INIT_REPORT(report);
++ long orig_code = *exit_code;
+
-+ if (ka->sa.sa_handler == SIG_IGN) {
-+ event = UTRACE_EVENT(SIGNAL_IGN);
-+ report.result = UTRACE_SIGNAL_IGN;
-+ } else if (ka->sa.sa_handler != SIG_DFL) {
-+ event = UTRACE_EVENT(SIGNAL);
-+ report.result = UTRACE_SIGNAL_DELIVER;
-+ } else if (sig_kernel_coredump(signr)) {
-+ event = UTRACE_EVENT(SIGNAL_CORE);
-+ report.result = UTRACE_SIGNAL_CORE;
-+ } else if (sig_kernel_ignore(signr)) {
-+ event = UTRACE_EVENT(SIGNAL_IGN);
-+ report.result = UTRACE_SIGNAL_IGN;
-+ } else if (signr == SIGSTOP) {
-+ event = UTRACE_EVENT(SIGNAL_STOP);
-+ report.result = UTRACE_SIGNAL_STOP;
-+ } else if (sig_kernel_stop(signr)) {
-+ event = UTRACE_EVENT(SIGNAL_STOP);
-+ report.result = UTRACE_SIGNAL_TSTP;
-+ } else {
-+ event = UTRACE_EVENT(SIGNAL_TERM);
-+ report.result = UTRACE_SIGNAL_TERM;
-+ }
++ REPORT(task, utrace, &report, UTRACE_EVENT(EXIT),
++ report_exit, orig_code, exit_code);
+
-+ /*
-+ * Now that we know what event type this signal is, we
-+ * can short-circuit if no engines care about those.
-+ */
-+ if ((task->utrace_flags & (event | UTRACE_EVENT(QUIESCE))) == 0)
-+ return signr;
++ if (report.action == UTRACE_STOP)
++ utrace_stop(task, utrace, report.resume_action);
++}
+
-+ /*
-+ * We have some interested engines, so tell them about
-+ * the signal and let them change its disposition.
-+ */
-+ spin_unlock_irq(&task->sighand->siglock);
-+ }
++/*
++ * Called iff UTRACE_EVENT(DEATH) or UTRACE_EVENT(QUIESCE) flag is set.
++ *
++ * It is always possible that we are racing with utrace_release_task here.
++ * For this reason, utrace_release_task checks for the event bits that get
++ * us here, and delays its cleanup for us to do.
++ */
++void utrace_report_death(struct task_struct *task, bool group_dead, int signal)
++{
++ struct utrace *utrace = task_utrace_struct(task);
++
++ INIT_REPORT(report);
++
++ BUG_ON(!task->exit_state);
+
+ /*
-+ * This reporting pass chooses what signal disposition we'll act on.
++ * We are presently considered "quiescent"--which is accurate
++ * inasmuch as we won't run any more user instructions ever again.
++ * But for utrace_control and utrace_set_events to be robust, they
++ * must be sure whether or not we will run any more callbacks. If
++ * a call comes in before we do, taking the lock here synchronizes
++ * us so we don't run any callbacks just disabled. Calls that come
++ * in while we're running the callbacks will see the exit.death
++ * flag and know that we are not yet fully quiescent for purposes
++ * of detach bookkeeping.
+ */
-+ list_for_each_entry(engine, &utrace->attached, entry) {
-+ /*
-+ * See start_callback() comment about this barrier.
-+ */
-+ utrace->reporting = engine;
-+ smp_mb();
++ spin_lock(&utrace->lock);
++ BUG_ON(utrace->death);
++ utrace->death = 1;
++ utrace->resume = UTRACE_RESUME;
++ splice_attaching(utrace);
++ spin_unlock(&utrace->lock);
+
-+ /*
-+ * This pairs with the barrier in mark_engine_detached(),
-+ * see start_callback() comments.
-+ */
-+ want = engine->flags;
-+ smp_rmb();
-+ ops = engine->ops;
++ REPORT_CALLBACKS(, task, utrace, &report, UTRACE_EVENT(DEATH),
++ report_death, engine, group_dead, signal);
+
-+ if ((want & (event | UTRACE_EVENT(QUIESCE))) == 0) {
-+ utrace->reporting = NULL;
-+ continue;
-+ }
++ utrace_maybe_reap(task, utrace, false);
++}
+
-+ if (ops->report_signal)
-+ ret = (*ops->report_signal)(
-+ report.result | report.action, engine,
-+ regs, info, ka, return_ka);
-+ else
-+ ret = (report.result | (*ops->report_quiesce)(
-+ report.action, engine, event));
++/*
++ * Finish the last reporting pass before returning to user mode.
++ */
++static void finish_resume_report(struct task_struct *task,
++ struct utrace *utrace,
++ struct utrace_report *report)
++{
++ finish_report_reset(task, utrace, report);
+
-+ /*
-+ * Avoid a tight loop reporting again and again if some
-+ * engine is too stupid.
-+ */
-+ switch (utrace_resume_action(ret)) {
-+ default:
-+ break;
-+ case UTRACE_INTERRUPT:
-+ case UTRACE_REPORT:
-+ ret = (ret & ~UTRACE_RESUME_MASK) | UTRACE_RESUME;
++ switch (report->action) {
++ case UTRACE_STOP:
++ utrace_stop(task, utrace, report->resume_action);
++ break;
++
++ case UTRACE_INTERRUPT:
++ if (!signal_pending(task))
++ set_tsk_thread_flag(task, TIF_SIGPENDING);
++ break;
++
++ case UTRACE_BLOCKSTEP:
++ if (likely(arch_has_block_step())) {
++ if (!ptrace_wants_step(task))
++ user_enable_block_step(task);
+ break;
+ }
+
-+ finish_callback(task, utrace, &report, engine, ret);
-+ }
++ /*
++ * This means some callback is to blame for failing
++ * to check arch_has_block_step() itself. Warn and
++ * then fall through to treat it as SINGLESTEP.
++ */
++ WARN(1, "UTRACE_BLOCKSTEP when !arch_has_block_step()");
+
-+ /*
-+ * We express the chosen action to the signals code in terms
-+ * of a representative signal whose default action does it.
-+ * Our caller uses our return value (signr) to decide what to
-+ * do, but uses info->si_signo as the signal number to report.
-+ */
-+ switch (utrace_signal_action(report.result)) {
-+ case UTRACE_SIGNAL_TERM:
-+ signr = SIGTERM;
++ case UTRACE_SINGLESTEP:
++ if (likely(arch_has_single_step())) {
++ if (!ptrace_wants_step(task))
++ user_enable_single_step(task);
++ } else {
++ /*
++ * This means some callback is to blame for failing
++ * to check arch_has_single_step() itself. Spew
++ * about it so the loser will fix his module.
++ */
++ WARN(1,
++ "UTRACE_SINGLESTEP when !arch_has_single_step()");
++ }
+ break;
+
-+ case UTRACE_SIGNAL_CORE:
-+ signr = SIGQUIT;
++ case UTRACE_REPORT:
++ case UTRACE_RESUME:
++ default:
++ if (!ptrace_wants_step(task))
++ user_disable_single_step(task);
+ break;
++ }
++}
+
-+ case UTRACE_SIGNAL_STOP:
-+ signr = SIGSTOP;
-+ break;
-+
-+ case UTRACE_SIGNAL_TSTP:
-+ signr = SIGTSTP;
-+ break;
-+
-+ case UTRACE_SIGNAL_DELIVER:
-+ signr = info->si_signo;
-+
-+ if (return_ka->sa.sa_handler == SIG_DFL) {
-+ /*
-+ * We'll do signr's normal default action.
-+ * For ignore, we'll fall through below.
-+ * For stop/death, break locks and returns it.
-+ */
-+ if (likely(signr) && !sig_kernel_ignore(signr))
-+ break;
-+ } else if (return_ka->sa.sa_handler != SIG_IGN &&
-+ likely(signr)) {
-+ /*
-+ * Complete the bookkeeping after the report.
-+ * The handler will run. If an engine wanted to
-+ * stop or step, then make sure we do another
-+ * report after signal handler setup.
-+ */
-+ if (report.action != UTRACE_RESUME)
-+ report.action = UTRACE_INTERRUPT;
-+ finish_report(task, utrace, &report, true);
-+
-+ if (unlikely(report.result & UTRACE_SIGNAL_HOLD))
-+ push_back_signal(task, info);
-+ else
-+ spin_lock_irq(&task->sighand->siglock);
++/*
++ * This is called when TIF_NOTIFY_RESUME had been set (and is now clear).
++ * We are close to user mode, and this is the place to report or stop.
++ * When we return, we're going to user mode or into the signals code.
++ */
++void utrace_resume(struct task_struct *task, struct pt_regs *regs)
++{
++ struct utrace *utrace = task_utrace_struct(task);
++ INIT_REPORT(report);
++ struct utrace_engine *engine;
+
-+ /*
-+ * We do the SA_ONESHOT work here since the
-+ * normal path will only touch *return_ka now.
-+ */
-+ if (unlikely(return_ka->sa.sa_flags & SA_ONESHOT)) {
-+ return_ka->sa.sa_flags &= ~SA_ONESHOT;
-+ if (likely(valid_signal(signr))) {
-+ ka = &task->sighand->action[signr - 1];
-+ ka->sa.sa_handler = SIG_DFL;
-+ }
-+ }
++ /*
++ * Some machines get here with interrupts disabled. The same arch
++ * code path leads to calling into get_signal_to_deliver(), which
++ * implicitly reenables them by virtue of spin_unlock_irq.
++ */
++ if (irqs_disabled()) /* make trace_hardirqs_on() happy */
++ local_irq_enable();
+
-+ return signr;
-+ }
++ /*
++ * If this flag is still set it's because there was a signal
++ * handler setup done but no report_signal following it. Clear
++ * the flag before we get to user so it doesn't confuse us later.
++ */
++ if (unlikely(utrace->signal_handler)) {
++ spin_lock(&utrace->lock);
++ utrace->signal_handler = 0;
++ spin_unlock(&utrace->lock);
++ }
+
-+ /* Fall through for an ignored signal. */
++ /*
++ * Update our bookkeeping even if there are no callbacks made here.
++ */
++ report.action = start_report(utrace);
+
-+ case UTRACE_SIGNAL_IGN:
-+ case UTRACE_SIGNAL_REPORT:
++ switch (report.action) {
++ case UTRACE_RESUME:
++ /*
++ * Anything we might have done was already handled by
++ * utrace_get_signal(), or this is an entirely spurious
++ * call. (The arch might use TIF_NOTIFY_RESUME for other
++ * purposes as well as calling us.)
++ */
++ return;
++ case UTRACE_REPORT:
++ if (unlikely(!(task->utrace_flags & UTRACE_EVENT(QUIESCE))))
++ break;
++ /*
++ * Do a simple reporting pass, with no specific
++ * callback after report_quiesce.
++ */
++ report.action = UTRACE_RESUME;
++ list_for_each_entry(engine, &utrace->attached, entry)
++ start_callback(utrace, &report, engine, task, 0);
++ break;
+ default:
+ /*
-+ * If the signal is being ignored, then we are on the way
-+ * directly back to user mode. We can stop here, or step,
-+ * as in utrace_resume(), above. After we've dealt with that,
-+ * our caller will relock and come back through here.
++ * Even if this report was truly spurious, there is no need
++ * for utrace_reset() now. TIF_NOTIFY_RESUME was already
++ * cleared--it doesn't stay spuriously set.
+ */
-+ finish_resume_report(task, utrace, &report);
-+
-+ if (unlikely(fatal_signal_pending(task))) {
-+ /*
-+ * The only reason we woke up now was because of a
-+ * SIGKILL. Don't do normal dequeuing in case it
-+ * might get a signal other than SIGKILL. That would
-+ * perturb the death state so it might differ from
-+ * what the debugger would have allowed to happen.
-+ * Instead, pluck out just the SIGKILL to be sure
-+ * we'll die immediately with nothing else different
-+ * from the quiescent state the debugger wanted us in.
-+ */
-+ sigset_t sigkill_only;
-+ siginitsetinv(&sigkill_only, sigmask(SIGKILL));
-+ spin_lock_irq(&task->sighand->siglock);
-+ signr = dequeue_signal(task, &sigkill_only, info);
-+ BUG_ON(signr != SIGKILL);
-+ *return_ka = task->sighand->action[SIGKILL - 1];
-+ return signr;
-+ }
-+
-+ if (unlikely(report.result & UTRACE_SIGNAL_HOLD)) {
-+ push_back_signal(task, info);
-+ spin_unlock_irq(&task->sighand->siglock);
-+ }
-+
-+ return -1;
++ report.spurious = false;
++ break;
+ }
+
+ /*
-+ * Complete the bookkeeping after the report.
-+ * This sets utrace->resume if UTRACE_STOP was used.
++ * Finish the report and either stop or get ready to resume.
++ * If utrace->resume was not UTRACE_REPORT, this applies its
++ * effect now (i.e. step or interrupt).
+ */
-+ finish_report(task, utrace, &report, true);
++ finish_resume_report(task, utrace, &report);
++}
+
-+ return_ka->sa.sa_handler = SIG_DFL;
++/*
++ * Return true if current has forced signal_pending().
++ *
++ * This is called only when current->utrace_flags is nonzero, so we know
++ * that current->utrace must be set. It's not inlined in tracehook.h
++ * just so that struct utrace can stay opaque outside this file.
++ */
++bool utrace_interrupt_pending(void)
++{
++ return task_utrace_struct(current)->resume == UTRACE_INTERRUPT;
++}
+
-+ /*
-+ * If this signal is fatal, si_signo gets through as exit_code.
-+ * We can't allow a completely bogus value there or else core
-+ * kernel code can freak out. (If an engine wants to control
-+ * the exit_code value exactly, it can do so in report_exit.)
-+ * We'll produce a big complaint in dmesg, but won't crash.
-+ * That's nicer for debugging your utrace engine.
-+ */
-+ if (unlikely(info->si_signo & 0x80)) {
-+ WARN(1, "utrace engine left bogus si_signo value!");
-+ info->si_signo = SIGTRAP;
-+ }
++/*
++ * Take the siglock and push @info back on our queue.
++ * Returns with @task->sighand->siglock held.
++ */
++static void push_back_signal(struct task_struct *task, siginfo_t *info)
++ __acquires(task->sighand->siglock)
++{
++ struct sigqueue *q;
+
-+ if (unlikely(report.result & UTRACE_SIGNAL_HOLD))
-+ push_back_signal(task, info);
-+ else
++ if (unlikely(!info->si_signo)) { /* Oh, a wise guy! */
+ spin_lock_irq(&task->sighand->siglock);
++ return;
++ }
+
-+ if (sig_kernel_stop(signr))
-+ task->jobctl |= JOBCTL_STOP_DEQUEUED;
++ q = sigqueue_alloc();
++ if (likely(q)) {
++ q->flags = 0;
++ copy_siginfo(&q->info, info);
++ }
+
-+ return signr;
++ spin_lock_irq(&task->sighand->siglock);
++
++ sigaddset(&task->pending.signal, info->si_signo);
++ if (likely(q))
++ list_add(&q->list, &task->pending.list);
++
++ set_tsk_thread_flag(task, TIF_SIGPENDING);
+}
+
+/*
-+ * This gets called after a signal handler has been set up.
-+ * We set a flag so the next report knows it happened.
-+ * If we're already stepping, make sure we do a report_signal.
-+ * If not, make sure we get into utrace_resume() where we can
-+ * clear the signal_handler flag before resuming.
++ * This is the hook from the signals code, called with the siglock held.
++ * Here is the ideal place to stop. We also dequeue and intercept signals.
+ */
-+void utrace_signal_handler(struct task_struct *task, int stepping)
++int utrace_get_signal(struct task_struct *task, struct pt_regs *regs,
++ siginfo_t *info, struct k_sigaction *return_ka)
++ __releases(task->sighand->siglock)
++ __acquires(task->sighand->siglock)
+{
-+ struct utrace *utrace = task_utrace_struct(task);
++ struct utrace *utrace;
++ struct k_sigaction *ka;
++ INIT_REPORT(report);
++ struct utrace_engine *engine;
++ const struct utrace_engine_ops *ops;
++ unsigned long event, want;
++ u32 ret;
++ int signr;
+
-+ spin_lock(&utrace->lock);
++ utrace = task_utrace_struct(task);
++ if (utrace->resume < UTRACE_RESUME ||
++ utrace->pending_attach || utrace->signal_handler) {
++ enum utrace_resume_action resume;
+
-+ utrace->signal_handler = 1;
-+ if (utrace->resume > UTRACE_INTERRUPT) {
-+ if (stepping) {
-+ utrace->resume = UTRACE_INTERRUPT;
-+ set_tsk_thread_flag(task, TIF_SIGPENDING);
-+ } else if (utrace->resume == UTRACE_RESUME) {
-+ set_tsk_thread_flag(task, TIF_NOTIFY_RESUME);
-+ }
-+ }
++ /*
++ * We've been asked for an explicit report before we
++ * even check for pending signals.
++ */
+
-+ spin_unlock(&utrace->lock);
-+}
++ spin_unlock_irq(&task->sighand->siglock);
+
-+/**
-+ * utrace_prepare_examine - prepare to examine thread state
-+ * @target: thread of interest, a &struct task_struct pointer
-+ * @engine: engine pointer returned by utrace_attach_task()
-+ * @exam: temporary state, a &struct utrace_examiner pointer
-+ *
-+ * This call prepares to safely examine the thread @target using
-+ * &struct user_regset calls, or direct access to thread-synchronous fields.
-+ *
-+ * When @target is current, this call is superfluous. When @target is
-+ * another thread, it must be held stopped via %UTRACE_STOP by @engine.
-+ *
-+ * This call may block the caller until @target stays stopped, so it must
-+ * be called only after the caller is sure @target is about to unschedule.
-+ * This means a zero return from a utrace_control() call on @engine giving
-+ * %UTRACE_STOP, or a report_quiesce() or report_signal() callback to
-+ * @engine that used %UTRACE_STOP in its return value.
-+ *
-+ * Returns -%ESRCH if @target is dead or -%EINVAL if %UTRACE_STOP was
-+ * not used. If @target has started running again despite %UTRACE_STOP
-+ * (for %SIGKILL or a spurious wakeup), this call returns -%EAGAIN.
-+ *
-+ * When this call returns zero, it's safe to use &struct user_regset
-+ * calls and task_user_regset_view() on @target and to examine some of
-+ * its fields directly. When the examination is complete, a
-+ * utrace_finish_examine() call must follow to check whether it was
-+ * completed safely.
-+ */
-+int utrace_prepare_examine(struct task_struct *target,
-+ struct utrace_engine *engine,
-+ struct utrace_examiner *exam)
-+{
-+ int ret = 0;
++ spin_lock(&utrace->lock);
+
-+ if (unlikely(target == current))
-+ return 0;
++ splice_attaching(utrace);
+
-+ rcu_read_lock();
-+ if (unlikely(!engine_wants_stop(engine)))
-+ ret = -EINVAL;
-+ else if (unlikely(target->exit_state))
-+ ret = -ESRCH;
-+ else {
-+ exam->state = target->state;
-+ if (unlikely(exam->state == TASK_RUNNING))
-+ ret = -EAGAIN;
-+ else
-+ get_task_struct(target);
-+ }
-+ rcu_read_unlock();
++ report.result = utrace->signal_handler ?
++ UTRACE_SIGNAL_HANDLER : UTRACE_SIGNAL_REPORT;
++ utrace->signal_handler = 0;
+
-+ if (likely(!ret)) {
-+ exam->ncsw = wait_task_inactive(target, exam->state);
-+ put_task_struct(target);
-+ if (unlikely(!exam->ncsw))
-+ ret = -EAGAIN;
-+ }
++ resume = utrace->resume;
++ utrace->resume = UTRACE_RESUME;
+
-+ return ret;
-+}
-+EXPORT_SYMBOL_GPL(utrace_prepare_examine);
++ spin_unlock(&utrace->lock);
+
-+/**
-+ * utrace_finish_examine - complete an examination of thread state
-+ * @target: thread of interest, a &struct task_struct pointer
-+ * @engine: engine pointer returned by utrace_attach_task()
-+ * @exam: pointer passed to utrace_prepare_examine() call
-+ *
-+ * This call completes an examination on the thread @target begun by a
-+ * paired utrace_prepare_examine() call with the same arguments that
-+ * returned success (zero).
-+ *
-+ * When @target is current, this call is superfluous. When @target is
-+ * another thread, this returns zero if @target has remained unscheduled
-+ * since the paired utrace_prepare_examine() call returned zero.
-+ *
-+ * When this returns an error, any examination done since the paired
-+ * utrace_prepare_examine() call is unreliable and the data extracted
-+ * should be discarded. The error is -%EINVAL if @engine is not
-+ * keeping @target stopped, or -%EAGAIN if @target woke up unexpectedly.
-+ */
-+int utrace_finish_examine(struct task_struct *target,
-+ struct utrace_engine *engine,
-+ struct utrace_examiner *exam)
-+{
-+ int ret = 0;
++ /*
++ * Make sure signal_pending() only returns true
++ * if there are real signals pending.
++ */
++ if (signal_pending(task)) {
++ spin_lock_irq(&task->sighand->siglock);
++ recalc_sigpending();
++ spin_unlock_irq(&task->sighand->siglock);
++ }
+
-+ if (unlikely(target == current))
++ if (resume > UTRACE_REPORT) {
++ /*
++ * We only got here to process utrace->resume.
++ * Despite no callbacks, this report is not spurious.
++ */
++ report.action = resume;
++ report.spurious = false;
++ finish_resume_report(task, utrace, &report);
++ return -1;
++ } else if (!(task->utrace_flags & UTRACE_EVENT(QUIESCE))) {
++ /*
++ * We only got here to clear utrace->signal_handler.
++ */
++ return -1;
++ }
++
++ /*
++ * Do a reporting pass for no signal, just for EVENT(QUIESCE).
++ * The engine callbacks can fill in *info and *return_ka.
++ * We'll pass NULL for the @orig_ka argument to indicate
++ * that there was no original signal.
++ */
++ event = 0;
++ ka = NULL;
++ memset(return_ka, 0, sizeof *return_ka);
++ } else if (!(task->utrace_flags & UTRACE_EVENT_SIGNAL_ALL) ||
++ unlikely(task->jobctl & JOBCTL_PENDING_MASK)) {
++ /*
++ * If no engine is interested in intercepting signals or
++ * we must stop, let the caller just dequeue them normally
++ * or participate in group-stop.
++ */
+ return 0;
++ } else {
++ /*
++ * Steal the next signal so we can let tracing engines
++ * examine it. From the signal number and sigaction,
++ * determine what normal delivery would do. If no
++ * engine perturbs it, we'll do that by returning the
++ * signal number after setting *return_ka.
++ */
++ signr = dequeue_signal(task, &task->blocked, info);
++ if (signr == 0)
++ return signr;
++ BUG_ON(signr != info->si_signo);
+
-+ rcu_read_lock();
-+ if (unlikely(!engine_wants_stop(engine)))
-+ ret = -EINVAL;
-+ else if (unlikely(target->state != exam->state))
-+ ret = -EAGAIN;
-+ else
-+ get_task_struct(target);
-+ rcu_read_unlock();
++ ka = &task->sighand->action[signr - 1];
++ *return_ka = *ka;
+
-+ if (likely(!ret)) {
-+ unsigned long ncsw = wait_task_inactive(target, exam->state);
-+ if (unlikely(ncsw != exam->ncsw))
-+ ret = -EAGAIN;
-+ put_task_struct(target);
++ /*
++ * We are never allowed to interfere with SIGKILL.
++ * Just punt after filling in *return_ka for our caller.
++ */
++ if (signr == SIGKILL)
++ return signr;
++
++ if (ka->sa.sa_handler == SIG_IGN) {
++ event = UTRACE_EVENT(SIGNAL_IGN);
++ report.result = UTRACE_SIGNAL_IGN;
++ } else if (ka->sa.sa_handler != SIG_DFL) {
++ event = UTRACE_EVENT(SIGNAL);
++ report.result = UTRACE_SIGNAL_DELIVER;
++ } else if (sig_kernel_coredump(signr)) {
++ event = UTRACE_EVENT(SIGNAL_CORE);
++ report.result = UTRACE_SIGNAL_CORE;
++ } else if (sig_kernel_ignore(signr)) {
++ event = UTRACE_EVENT(SIGNAL_IGN);
++ report.result = UTRACE_SIGNAL_IGN;
++ } else if (signr == SIGSTOP) {
++ event = UTRACE_EVENT(SIGNAL_STOP);
++ report.result = UTRACE_SIGNAL_STOP;
++ } else if (sig_kernel_stop(signr)) {
++ event = UTRACE_EVENT(SIGNAL_STOP);
++ report.result = UTRACE_SIGNAL_TSTP;
++ } else {
++ event = UTRACE_EVENT(SIGNAL_TERM);
++ report.result = UTRACE_SIGNAL_TERM;
++ }
++
++ /*
++ * Now that we know what event type this signal is, we
++ * can short-circuit if no engines care about those.
++ */
++ if ((task->utrace_flags & (event | UTRACE_EVENT(QUIESCE))) == 0)
++ return signr;
++
++ /*
++ * We have some interested engines, so tell them about
++ * the signal and let them change its disposition.
++ */
++ spin_unlock_irq(&task->sighand->siglock);
+ }
+
-+ return ret;
-+}
-+EXPORT_SYMBOL_GPL(utrace_finish_examine);
++ /*
++ * This reporting pass chooses what signal disposition we'll act on.
++ */
++ list_for_each_entry(engine, &utrace->attached, entry) {
++ /*
++ * See start_callback() comment about this barrier.
++ */
++ utrace->reporting = engine;
++ smp_mb();
+
-+/*
-+ * This is declared in linux/regset.h and defined in machine-dependent
-+ * code. We put the export here to ensure no machine forgets it.
-+ */
-+EXPORT_SYMBOL_GPL(task_user_regset_view);
++ /*
++ * This pairs with the barrier in mark_engine_detached(),
++ * see start_callback() comments.
++ */
++ want = engine->flags;
++ smp_rmb();
++ ops = engine->ops;
+
-+/*
-+ * Called with rcu_read_lock() held.
-+ */
-+void task_utrace_proc_status(struct seq_file *m, struct task_struct *p)
-+{
-+ seq_printf(m, "Utrace:\t%lx\n", p->utrace_flags);
-+}
---
-1.5.5.1
-
-
-From oleg at redhat.com Mon Nov 21 15:06:17 2011
-Return-Path: oleg at redhat.com
-Received: from zmta02.collab.prod.int.phx2.redhat.com (LHLO
- zmta02.collab.prod.int.phx2.redhat.com) (10.5.5.32) by
- zmail13.collab.prod.int.phx2.redhat.com with LMTP; Mon, 21 Nov 2011
- 15:06:17 -0500 (EST)
-Received: from localhost (localhost.localdomain [127.0.0.1])
- by zmta02.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 1CCED128D6B;
- Mon, 21 Nov 2011 15:06:17 -0500 (EST)
-Received: from zmta02.collab.prod.int.phx2.redhat.com ([127.0.0.1])
- by localhost (zmta02.collab.prod.int.phx2.redhat.com [127.0.0.1]) (amavisd-new, port 10024)
- with ESMTP id FWdJDV74o-ay; Mon, 21 Nov 2011 15:06:17 -0500 (EST)
-Received: from int-mx10.intmail.prod.int.phx2.redhat.com (int-mx10.intmail.prod.int.phx2.redhat.com [10.5.11.23])
- by zmta02.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 082081285E4;
- Mon, 21 Nov 2011 15:06:17 -0500 (EST)
-Received: from tranklukator.englab.brq.redhat.com (dhcp-1-232.brq.redhat.com [10.34.1.232])
- by int-mx10.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with SMTP id pALK6E7d015251;
- Mon, 21 Nov 2011 15:06:15 -0500
-Received: by tranklukator.englab.brq.redhat.com (nbSMTP-1.00) for uid 500
- oleg at redhat.com; Mon, 21 Nov 2011 21:01:35 +0100 (CET)
-Date: Mon, 21 Nov 2011 21:01:33 +0100
-From: Oleg Nesterov <oleg at redhat.com>
-To: Dave Jones <davej at redhat.com>, "Frank Ch. Eigler" <fche at redhat.com>,
- Josh Boyer <jwboyer at redhat.com>, Josh Stone <jistone at redhat.com>,
- Kyle McMartin <kmcmartin at redhat.com>
-Cc: kernel at lists.fedoraproject.org
-Subject: [PATCH 02/33] utrace: add utrace_init_task/utrace_free_task calls
-Message-ID: <20111121200133.GA27759 at redhat.com>
-MIME-Version: 1.0
-Content-Type: text/plain; charset=us-ascii
-Content-Disposition: inline
-In-Reply-To: <20111121200039.GA27699 at redhat.com>
-User-Agent: Mutt/1.5.18 (2008-05-17)
-X-Scanned-By: MIMEDefang 2.68 on 10.5.11.23
-Status: RO
-Content-Length: 1295
-Lines: 47
-
-Add the necessary copy_process()->utrace_init_task() and
-free_task()->utrace_free_task() calls.
-
-Originally this was the part of "utrace core" patch, but since
-tracehooks are dying it doesn't make sense to reintroduce them.
-Instead, just call the utrace_ helpers directly. This is fine
-even without CONFIG_UTRACE, gcc is smart enough to optimize out
-the code in free_task().
-
-Signed-off-by: Oleg Nesterov <oleg at redhat.com>
----
- kernel/fork.c | 5 +++++
- 1 files changed, 5 insertions(+), 0 deletions(-)
-
-diff --git a/kernel/fork.c b/kernel/fork.c
-index ba0d172..b8719c2 100644
---- a/kernel/fork.c
-+++ b/kernel/fork.c
-@@ -66,6 +66,7 @@
- #include <linux/user-return-notifier.h>
- #include <linux/oom.h>
- #include <linux/khugepaged.h>
-+#include <linux/utrace.h>
-
- #include <asm/pgtable.h>
- #include <asm/pgalloc.h>
-@@ -167,6 +168,8 @@ void free_task(struct task_struct *tsk)
- free_thread_info(tsk->stack);
- rt_mutex_debug_task_free(tsk);
- ftrace_graph_exit_task(tsk);
-+ if (task_utrace_struct(tsk))
-+ utrace_free_task(tsk);
- free_task_struct(tsk);
- }
- EXPORT_SYMBOL(free_task);
-@@ -1093,6 +1096,8 @@ static struct task_struct *copy_process(unsigned long clone_flags,
- if (!p)
- goto fork_out;
-
-+ utrace_init_task(p);
++ if ((want & (event | UTRACE_EVENT(QUIESCE))) == 0) {
++ utrace->reporting = NULL;
++ continue;
++ }
+
- ftrace_graph_init_task(p);
-
- rt_mutex_init_task(p);
---
-1.5.5.1
-
-
-From oleg at redhat.com Mon Nov 21 15:06:20 2011
-Return-Path: oleg at redhat.com
-Received: from zmta03.collab.prod.int.phx2.redhat.com (LHLO
- zmta03.collab.prod.int.phx2.redhat.com) (10.5.5.33) by
- zmail13.collab.prod.int.phx2.redhat.com with LMTP; Mon, 21 Nov 2011
- 15:06:20 -0500 (EST)
-Received: from localhost (localhost.localdomain [127.0.0.1])
- by zmta03.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 8779AD87BC;
- Mon, 21 Nov 2011 15:06:20 -0500 (EST)
-Received: from zmta03.collab.prod.int.phx2.redhat.com ([127.0.0.1])
- by localhost (zmta03.collab.prod.int.phx2.redhat.com [127.0.0.1]) (amavisd-new, port 10024)
- with ESMTP id umzAMwRd7rMt; Mon, 21 Nov 2011 15:06:20 -0500 (EST)
-Received: from int-mx10.intmail.prod.int.phx2.redhat.com (int-mx10.intmail.prod.int.phx2.redhat.com [10.5.11.23])
- by zmta03.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 72C98D8707;
- Mon, 21 Nov 2011 15:06:20 -0500 (EST)
-Received: from tranklukator.englab.brq.redhat.com (dhcp-1-232.brq.redhat.com [10.34.1.232])
- by int-mx10.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with SMTP id pALK6HAj015261;
- Mon, 21 Nov 2011 15:06:18 -0500
-Received: by tranklukator.englab.brq.redhat.com (nbSMTP-1.00) for uid 500
- oleg at redhat.com; Mon, 21 Nov 2011 21:01:39 +0100 (CET)
-Date: Mon, 21 Nov 2011 21:01:36 +0100
-From: Oleg Nesterov <oleg at redhat.com>
-To: Dave Jones <davej at redhat.com>, "Frank Ch. Eigler" <fche at redhat.com>,
- Josh Boyer <jwboyer at redhat.com>, Josh Stone <jistone at redhat.com>,
- Kyle McMartin <kmcmartin at redhat.com>
-Cc: kernel at lists.fedoraproject.org
-Subject: [PATCH 03/33] tracehooks: add utrace hooks
-Message-ID: <20111121200136.GA27766 at redhat.com>
-MIME-Version: 1.0
-Content-Type: text/plain; charset=us-ascii
-Content-Disposition: inline
-In-Reply-To: <20111121200039.GA27699 at redhat.com>
-User-Agent: Mutt/1.5.18 (2008-05-17)
-X-Scanned-By: MIMEDefang 2.68 on 10.5.11.23
-Status: RO
-Content-Length: 2310
-Lines: 75
-
-Add the necessary utrace hooks in the tracehooks which were not
-removed yet.
-
-Signed-off-by: Oleg Nesterov <oleg at redhat.com>
----
- include/linux/tracehook.h | 22 +++++++++++++++++++++-
- 1 files changed, 21 insertions(+), 1 deletions(-)
-
-diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h
-index a71a292..8cc28bc 100644
---- a/include/linux/tracehook.h
-+++ b/include/linux/tracehook.h
-@@ -49,6 +49,7 @@
- #include <linux/sched.h>
- #include <linux/ptrace.h>
- #include <linux/security.h>
-+#include <linux/utrace.h>
- struct linux_binprm;
-
- /*
-@@ -96,6 +97,9 @@ static inline void ptrace_report_syscall(struct pt_regs *regs)
- static inline __must_check int tracehook_report_syscall_entry(
- struct pt_regs *regs)
- {
-+ if ((task_utrace_flags(current) & UTRACE_EVENT(SYSCALL_ENTRY)) &&
-+ utrace_report_syscall_entry(regs))
-+ return 1;
- ptrace_report_syscall(regs);
- return 0;
- }
-@@ -119,6 +123,9 @@ static inline __must_check int tracehook_report_syscall_entry(
- */
- static inline void tracehook_report_syscall_exit(struct pt_regs *regs, int step)
- {
-+ if (task_utrace_flags(current) & UTRACE_EVENT(SYSCALL_EXIT))
-+ utrace_report_syscall_exit(regs);
++ if (ops->report_signal)
++ ret = (*ops->report_signal)(
++ report.result | report.action, engine,
++ regs, info, ka, return_ka);
++ else
++ ret = (report.result | (*ops->report_quiesce)(
++ report.action, engine, event));
++
++ /*
++ * Avoid a tight loop reporting again and again if some
++ * engine is too stupid.
++ */
++ switch (utrace_resume_action(ret)) {
++ default:
++ break;
++ case UTRACE_INTERRUPT:
++ case UTRACE_REPORT:
++ ret = (ret & ~UTRACE_RESUME_MASK) | UTRACE_RESUME;
++ break;
++ }
++
++ finish_callback(task, utrace, &report, engine, ret);
++ }
+
- if (step) {
- siginfo_t info;
- user_single_step_siginfo(current, regs, &info);
-@@ -148,6 +155,8 @@ static inline void tracehook_signal_handler(int sig, siginfo_t *info,
- const struct k_sigaction *ka,
- struct pt_regs *regs, int stepping)
- {
-+ if (task_utrace_flags(current))
-+ utrace_signal_handler(current, stepping);
- if (stepping)
- ptrace_notify(SIGTRAP);
- }
-@@ -179,10 +188,21 @@ static inline void set_notify_resume(struct task_struct *task)
- * asynchronously, this will be called again before we return to
- * user mode.
- *
-- * Called without locks.
-+ * Called without locks. However, on some machines this may be
-+ * called with interrupts disabled.
- */
- static inline void tracehook_notify_resume(struct pt_regs *regs)
- {
-+ struct task_struct *task = current;
+ /*
-+ * Prevent the following store/load from getting ahead of the
-+ * caller which clears TIF_NOTIFY_RESUME. This pairs with the
-+ * implicit mb() before setting TIF_NOTIFY_RESUME in
-+ * set_notify_resume().
++ * We express the chosen action to the signals code in terms
++ * of a representative signal whose default action does it.
++ * Our caller uses our return value (signr) to decide what to
++ * do, but uses info->si_signo as the signal number to report.
+ */
-+ smp_mb();
-+ if (task_utrace_flags(task))
-+ utrace_resume(task, regs);
- }
- #endif /* TIF_NOTIFY_RESUME */
-
---
-1.5.5.1
-
-
-From oleg at redhat.com Mon Nov 21 15:06:23 2011
-Return-Path: oleg at redhat.com
-Received: from zmta01.collab.prod.int.phx2.redhat.com (LHLO
- zmta01.collab.prod.int.phx2.redhat.com) (10.5.5.31) by
- zmail13.collab.prod.int.phx2.redhat.com with LMTP; Mon, 21 Nov 2011
- 15:06:23 -0500 (EST)
-Received: from localhost (localhost.localdomain [127.0.0.1])
- by zmta01.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 6559BD4FDE;
- Mon, 21 Nov 2011 15:06:23 -0500 (EST)
-Received: from zmta01.collab.prod.int.phx2.redhat.com ([127.0.0.1])
- by localhost (zmta01.collab.prod.int.phx2.redhat.com [127.0.0.1]) (amavisd-new, port 10024)
- with ESMTP id 6BjISiO+cS1U; Mon, 21 Nov 2011 15:06:23 -0500 (EST)
-Received: from int-mx09.intmail.prod.int.phx2.redhat.com (int-mx09.intmail.prod.int.phx2.redhat.com [10.5.11.22])
- by zmta01.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 50366D4C39;
- Mon, 21 Nov 2011 15:06:23 -0500 (EST)
-Received: from tranklukator.englab.brq.redhat.com (dhcp-1-232.brq.redhat.com [10.34.1.232])
- by int-mx09.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with SMTP id pALK6KT9017533;
- Mon, 21 Nov 2011 15:06:21 -0500
-Received: by tranklukator.englab.brq.redhat.com (nbSMTP-1.00) for uid 500
- oleg at redhat.com; Mon, 21 Nov 2011 21:01:42 +0100 (CET)
-Date: Mon, 21 Nov 2011 21:01:39 +0100
-From: Oleg Nesterov <oleg at redhat.com>
-To: Dave Jones <davej at redhat.com>, "Frank Ch. Eigler" <fche at redhat.com>,
- Josh Boyer <jwboyer at redhat.com>, Josh Stone <jistone at redhat.com>,
- Kyle McMartin <kmcmartin at redhat.com>
-Cc: kernel at lists.fedoraproject.org
-Subject: [PATCH 04/33] tracehooks: reintroduce
- tracehook_consider_fatal_signal()
-Message-ID: <20111121200139.GA27769 at redhat.com>
-MIME-Version: 1.0
-Content-Type: text/plain; charset=us-ascii
-Content-Disposition: inline
-In-Reply-To: <20111121200039.GA27699 at redhat.com>
-User-Agent: Mutt/1.5.18 (2008-05-17)
-X-Scanned-By: MIMEDefang 2.68 on 10.5.11.22
-Status: RO
-Content-Length: 3257
-Lines: 90
-
-Add the killed tracehook_consider_fatal_signal() back. It has multiple
-callers and it is not easy add the necessary checks inline.
-
-Signed-off-by: Oleg Nesterov <oleg at redhat.com>
----
- arch/s390/kernel/traps.c | 4 ++--
- include/linux/tracehook.h | 22 ++++++++++++++++++++++
- kernel/signal.c | 4 ++--
- 3 files changed, 26 insertions(+), 4 deletions(-)
-
-diff --git a/arch/s390/kernel/traps.c b/arch/s390/kernel/traps.c
-index a9807dd..f506e1b 100644
---- a/arch/s390/kernel/traps.c
-+++ b/arch/s390/kernel/traps.c
-@@ -329,7 +329,7 @@ void __kprobes do_per_trap(struct pt_regs *regs)
-
- if (notify_die(DIE_SSTEP, "sstep", regs, 0, 0, SIGTRAP) == NOTIFY_STOP)
- return;
-- if (!current->ptrace)
-+ if (!tracehook_consider_fatal_signal(current, SIGTRAP))
- return;
- info.si_signo = SIGTRAP;
- info.si_errno = 0;
-@@ -429,7 +429,7 @@ static void __kprobes illegal_op(struct pt_regs *regs, long pgm_int_code,
- if (get_user(*((__u16 *) opcode), (__u16 __user *) location))
- return;
- if (*((__u16 *) opcode) == S390_BREAKPOINT_U16) {
-- if (current->ptrace) {
-+ if (tracehook_consider_fatal_signal(current, SIGTRAP)) {
- info.si_signo = SIGTRAP;
- info.si_errno = 0;
- info.si_code = TRAP_BRKPT;
-diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h
-index 8cc28bc..ec2af67 100644
---- a/include/linux/tracehook.h
-+++ b/include/linux/tracehook.h
-@@ -161,6 +161,28 @@ static inline void tracehook_signal_handler(int sig, siginfo_t *info,
- ptrace_notify(SIGTRAP);
- }
-
-+/**
-+ * tracehook_consider_fatal_signal - suppress special handling of fatal signal
-+ * @task: task receiving the signal
-+ * @sig: signal number being sent
-+ *
-+ * Return nonzero to prevent special handling of this termination signal.
-+ * Normally handler for signal is %SIG_DFL. It can be %SIG_IGN if @sig is
-+ * ignored, in which case force_sig() is about to reset it to %SIG_DFL.
-+ * When this returns zero, this signal might cause a quick termination
-+ * that does not give the debugger a chance to intercept the signal.
-+ *
-+ * Called with or without @task->sighand->siglock held.
-+ */
-+static inline int tracehook_consider_fatal_signal(struct task_struct *task,
-+ int sig)
-+{
-+ if (unlikely(task_utrace_flags(task) & (UTRACE_EVENT(SIGNAL_TERM) |
-+ UTRACE_EVENT(SIGNAL_CORE))))
-+ return 1;
-+ return task->ptrace != 0;
-+}
++ switch (utrace_signal_action(report.result)) {
++ case UTRACE_SIGNAL_TERM:
++ signr = SIGTERM;
++ break;
+
- #ifdef TIF_NOTIFY_RESUME
- /**
- * set_notify_resume - cause tracehook_notify_resume() to be called
-diff --git a/kernel/signal.c b/kernel/signal.c
-index b3f78d0..d7b90cd 100644
---- a/kernel/signal.c
-+++ b/kernel/signal.c
-@@ -494,7 +494,7 @@ int unhandled_signal(struct task_struct *tsk, int sig)
- if (handler != SIG_IGN && handler != SIG_DFL)
- return 0;
- /* if ptraced, let the tracer determine */
-- return !tsk->ptrace;
-+ return !tracehook_consider_fatal_signal(tsk, sig);
- }
-
- /*
-@@ -982,7 +982,7 @@ static void complete_signal(int sig, struct task_struct *p, int group)
- if (sig_fatal(p, sig) &&
- !(signal->flags & (SIGNAL_UNKILLABLE | SIGNAL_GROUP_EXIT)) &&
- !sigismember(&t->real_blocked, sig) &&
-- (sig == SIGKILL || !t->ptrace)) {
-+ (sig == SIGKILL || !tracehook_consider_fatal_signal(t, sig))) {
- /*
- * This signal will be fatal to the whole group.
- */
---
-1.5.5.1
-
-
-From oleg at redhat.com Mon Nov 21 15:06:26 2011
-Return-Path: oleg at redhat.com
-Received: from zmta01.collab.prod.int.phx2.redhat.com (LHLO
- zmta01.collab.prod.int.phx2.redhat.com) (10.5.5.31) by
- zmail13.collab.prod.int.phx2.redhat.com with LMTP; Mon, 21 Nov 2011
- 15:06:25 -0500 (EST)
-Received: from localhost (localhost.localdomain [127.0.0.1])
- by zmta01.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id E7A2DD498A;
- Mon, 21 Nov 2011 15:06:25 -0500 (EST)
-Received: from zmta01.collab.prod.int.phx2.redhat.com ([127.0.0.1])
- by localhost (zmta01.collab.prod.int.phx2.redhat.com [127.0.0.1]) (amavisd-new, port 10024)
- with ESMTP id 4L9zskFVsr4I; Mon, 21 Nov 2011 15:06:25 -0500 (EST)
-Received: from int-mx09.intmail.prod.int.phx2.redhat.com (int-mx09.intmail.prod.int.phx2.redhat.com [10.5.11.22])
- by zmta01.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id D3487D4C39;
- Mon, 21 Nov 2011 15:06:25 -0500 (EST)
-Received: from tranklukator.englab.brq.redhat.com (dhcp-1-232.brq.redhat.com [10.34.1.232])
- by int-mx09.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with SMTP id pALK6Nnp017552;
- Mon, 21 Nov 2011 15:06:24 -0500
-Received: by tranklukator.englab.brq.redhat.com (nbSMTP-1.00) for uid 500
- oleg at redhat.com; Mon, 21 Nov 2011 21:01:44 +0100 (CET)
-Date: Mon, 21 Nov 2011 21:01:42 +0100
-From: Oleg Nesterov <oleg at redhat.com>
-To: Dave Jones <davej at redhat.com>, "Frank Ch. Eigler" <fche at redhat.com>,
- Josh Boyer <jwboyer at redhat.com>, Josh Stone <jistone at redhat.com>,
- Kyle McMartin <kmcmartin at redhat.com>
-Cc: kernel at lists.fedoraproject.org
-Subject: [PATCH 05/33] add utrace hooks into sig_ignored() and
- recalc_sigpending()
-Message-ID: <20111121200142.GA27777 at redhat.com>
-MIME-Version: 1.0
-Content-Type: text/plain; charset=us-ascii
-Content-Disposition: inline
-In-Reply-To: <20111121200039.GA27699 at redhat.com>
-User-Agent: Mutt/1.5.18 (2008-05-17)
-X-Scanned-By: MIMEDefang 2.68 on 10.5.11.22
-Status: RO
-Content-Length: 1508
-Lines: 51
-
-Add the necessary and somewhat "special" hooks into sig_ignored() and
-recalc_sigpending(). Basically this restores _force_sigpending() and
-_consider_ignored_signal() tracehook logic.
-
-Signed-off-by: Oleg Nesterov <oleg at redhat.com>
----
- include/linux/utrace.h | 2 ++
- kernel/signal.c | 7 ++++++-
- 2 files changed, 8 insertions(+), 1 deletions(-)
-
-diff --git a/include/linux/utrace.h b/include/linux/utrace.h
-index f251efe..1b8da1c 100644
---- a/include/linux/utrace.h
-+++ b/include/linux/utrace.h
-@@ -107,6 +107,8 @@ bool utrace_report_syscall_entry(struct pt_regs *);
- void utrace_report_syscall_exit(struct pt_regs *);
- void utrace_signal_handler(struct task_struct *, int);
-
-+#define UTRACE_FLAG(task, ev) (task_utrace_flags(task) & UTRACE_EVENT(ev))
++ case UTRACE_SIGNAL_CORE:
++ signr = SIGQUIT;
++ break;
+
- #ifndef CONFIG_UTRACE
-
- /*
-diff --git a/kernel/signal.c b/kernel/signal.c
-index d7b90cd..8594cb2 100644
---- a/kernel/signal.c
-+++ b/kernel/signal.c
-@@ -87,7 +87,7 @@ static int sig_ignored(struct task_struct *t, int sig, int from_ancestor_ns)
- /*
- * Tracers may want to know about even ignored signals.
- */
-- return !t->ptrace;
-+ return !t->ptrace && !UTRACE_FLAG(t, SIGNAL_IGN);
- }
-
- /*
-@@ -150,6 +150,11 @@ void recalc_sigpending_and_wake(struct task_struct *t)
-
- void recalc_sigpending(void)
- {
-+ if (task_utrace_flags(current) && utrace_interrupt_pending()) {
-+ set_thread_flag(TIF_SIGPENDING);
-+ return;
-+ }
++ case UTRACE_SIGNAL_STOP:
++ signr = SIGSTOP;
++ break;
+
- if (!recalc_sigpending_tsk(current) && !freezing(current))
- clear_thread_flag(TIF_SIGPENDING);
-
---
-1.5.5.1
-
-
-From oleg at redhat.com Mon Nov 21 15:06:28 2011
-Return-Path: oleg at redhat.com
-Received: from zmta01.collab.prod.int.phx2.redhat.com (LHLO
- zmta01.collab.prod.int.phx2.redhat.com) (10.5.5.31) by
- zmail13.collab.prod.int.phx2.redhat.com with LMTP; Mon, 21 Nov 2011
- 15:06:28 -0500 (EST)
-Received: from localhost (localhost.localdomain [127.0.0.1])
- by zmta01.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 6E4A6D4F4A;
- Mon, 21 Nov 2011 15:06:28 -0500 (EST)
-Received: from zmta01.collab.prod.int.phx2.redhat.com ([127.0.0.1])
- by localhost (zmta01.collab.prod.int.phx2.redhat.com [127.0.0.1]) (amavisd-new, port 10024)
- with ESMTP id vEeGLyK1+rX9; Mon, 21 Nov 2011 15:06:28 -0500 (EST)
-Received: from int-mx02.intmail.prod.int.phx2.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12])
- by zmta01.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 5AD50D4C39;
- Mon, 21 Nov 2011 15:06:28 -0500 (EST)
-Received: from tranklukator.englab.brq.redhat.com (dhcp-1-232.brq.redhat.com [10.34.1.232])
- by int-mx02.intmail.prod.int.phx2.redhat.com (8.13.8/8.13.8) with SMTP id pALK6QOB029543;
- Mon, 21 Nov 2011 15:06:26 -0500
-Received: by tranklukator.englab.brq.redhat.com (nbSMTP-1.00) for uid 500
- oleg at redhat.com; Mon, 21 Nov 2011 21:01:47 +0100 (CET)
-Date: Mon, 21 Nov 2011 21:01:45 +0100
-From: Oleg Nesterov <oleg at redhat.com>
-To: Dave Jones <davej at redhat.com>, "Frank Ch. Eigler" <fche at redhat.com>,
- Josh Boyer <jwboyer at redhat.com>, Josh Stone <jistone at redhat.com>,
- Kyle McMartin <kmcmartin at redhat.com>
-Cc: kernel at lists.fedoraproject.org
-Subject: [PATCH 06/33] restore the EXEC/EXIT/CLONE utrace hooks
-Message-ID: <20111121200145.GA27780 at redhat.com>
-MIME-Version: 1.0
-Content-Type: text/plain; charset=us-ascii
-Content-Disposition: inline
-In-Reply-To: <20111121200039.GA27699 at redhat.com>
-User-Agent: Mutt/1.5.18 (2008-05-17)
-X-Scanned-By: MIMEDefang 2.67 on 10.5.11.12
-Status: RO
-Content-Length: 2383
-Lines: 83
-
-Restore the "trivial" EXEC/EXIT/CLONE utrace hooks. Add the
-simple helper, UTRACE_HOOK(), to minimize the changes.
-
-Signed-off-by: Oleg Nesterov <oleg at redhat.com>
----
- fs/exec.c | 5 ++++-
- include/linux/utrace.h | 6 ++++++
- kernel/exit.c | 1 +
- kernel/fork.c | 4 ++++
- 4 files changed, 15 insertions(+), 1 deletions(-)
-
-diff --git a/fs/exec.c b/fs/exec.c
-index 3625464..7d8f8bd 100644
---- a/fs/exec.c
-+++ b/fs/exec.c
-@@ -1397,9 +1397,12 @@ int search_binary_handler(struct linux_binprm *bprm,struct pt_regs *regs)
- */
- bprm->recursion_depth = depth;
- if (retval >= 0) {
-- if (depth == 0)
-+ if (depth == 0) {
-+ UTRACE_HOOK(current, EXEC,
-+ report_exec(fmt, bprm, regs));
- ptrace_event(PTRACE_EVENT_EXEC,
- old_pid);
-+ }
- put_binfmt(fmt);
- allow_write_access(bprm->file);
- if (bprm->file)
-diff --git a/include/linux/utrace.h b/include/linux/utrace.h
-index 1b8da1c..9ac0b1b 100644
---- a/include/linux/utrace.h
-+++ b/include/linux/utrace.h
-@@ -109,6 +109,12 @@ void utrace_signal_handler(struct task_struct *, int);
-
- #define UTRACE_FLAG(task, ev) (task_utrace_flags(task) & UTRACE_EVENT(ev))
-
-+#define UTRACE_HOOK(task, ev, callback) \
-+ do { \
-+ if (UTRACE_FLAG(task, ev)) \
-+ utrace_ ## callback; \
-+ } while (0)
++ case UTRACE_SIGNAL_TSTP:
++ signr = SIGTSTP;
++ break;
+
- #ifndef CONFIG_UTRACE
-
- /*
-diff --git a/kernel/exit.c b/kernel/exit.c
-index d0b7d98..f5eba3b 100644
---- a/kernel/exit.c
-+++ b/kernel/exit.c
-@@ -911,6 +911,7 @@ NORET_TYPE void do_exit(long code)
- */
- set_fs(USER_DS);
-
-+ UTRACE_HOOK(current, EXIT, report_exit(&code));
- ptrace_event(PTRACE_EVENT_EXIT, code);
-
- validate_creds_for_do_exit(tsk);
-diff --git a/kernel/fork.c b/kernel/fork.c
-index b8719c2..09d5049 100644
---- a/kernel/fork.c
-+++ b/kernel/fork.c
-@@ -1535,6 +1535,8 @@ long do_fork(unsigned long clone_flags,
-
- audit_finish_fork(p);
-
-+ UTRACE_HOOK(current, CLONE, report_clone(clone_flags, p));
++ case UTRACE_SIGNAL_DELIVER:
++ signr = info->si_signo;
+
- /*
- * We set PF_STARTING at creation in case tracing wants to
- * use this to distinguish a fully live task from one that
-@@ -1546,6 +1548,8 @@ long do_fork(unsigned long clone_flags,
- wake_up_new_task(p);
-
- /* forking complete and child started to run, tell ptracer */
-+ if (clone_flags & CLONE_VFORK)
-+ UTRACE_HOOK(current, CLONE, finish_vfork(current));
- if (unlikely(trace))
- ptrace_event(trace, nr);
-
---
-1.5.5.1
-
-
-From oleg at redhat.com Mon Nov 21 15:06:30 2011
-Return-Path: oleg at redhat.com
-Received: from zmta02.collab.prod.int.phx2.redhat.com (LHLO
- zmta02.collab.prod.int.phx2.redhat.com) (10.5.5.32) by
- zmail13.collab.prod.int.phx2.redhat.com with LMTP; Mon, 21 Nov 2011
- 15:06:30 -0500 (EST)
-Received: from localhost (localhost.localdomain [127.0.0.1])
- by zmta02.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id D248D1284E0;
- Mon, 21 Nov 2011 15:06:30 -0500 (EST)
-Received: from zmta02.collab.prod.int.phx2.redhat.com ([127.0.0.1])
- by localhost (zmta02.collab.prod.int.phx2.redhat.com [127.0.0.1]) (amavisd-new, port 10024)
- with ESMTP id divvekTmLScS; Mon, 21 Nov 2011 15:06:30 -0500 (EST)
-Received: from int-mx02.intmail.prod.int.phx2.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12])
- by zmta02.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id BDC9412800E;
- Mon, 21 Nov 2011 15:06:30 -0500 (EST)
-Received: from tranklukator.englab.brq.redhat.com (dhcp-1-232.brq.redhat.com [10.34.1.232])
- by int-mx02.intmail.prod.int.phx2.redhat.com (8.13.8/8.13.8) with SMTP id pALK6Shj029546;
- Mon, 21 Nov 2011 15:06:29 -0500
-Received: by tranklukator.englab.brq.redhat.com (nbSMTP-1.00) for uid 500
- oleg at redhat.com; Mon, 21 Nov 2011 21:01:49 +0100 (CET)
-Date: Mon, 21 Nov 2011 21:01:47 +0100
-From: Oleg Nesterov <oleg at redhat.com>
-To: Dave Jones <davej at redhat.com>, "Frank Ch. Eigler" <fche at redhat.com>,
- Josh Boyer <jwboyer at redhat.com>, Josh Stone <jistone at redhat.com>,
- Kyle McMartin <kmcmartin at redhat.com>
-Cc: kernel at lists.fedoraproject.org
-Subject: [PATCH 07/33] utrace: utrace_report_death() can use
- task_utrace_struct()
-Message-ID: <20111121200147.GA27787 at redhat.com>
-MIME-Version: 1.0
-Content-Type: text/plain; charset=us-ascii
-Content-Disposition: inline
-In-Reply-To: <20111121200039.GA27699 at redhat.com>
-User-Agent: Mutt/1.5.18 (2008-05-17)
-X-Scanned-By: MIMEDefang 2.67 on 10.5.11.12
-Status: RO
-Content-Length: 1692
-Lines: 44
-
-utrace_report_death() assumes that the caller (exit_notify) should
-pass task->utrace as an argument. This is no longer needed, it can
-safely do task_utrace_struct(). This way we avoid the nasty changes
-in exit_notify().
-
-Signed-off-by: Oleg Nesterov <oleg at redhat.com>
----
- include/linux/utrace.h | 2 +-
- kernel/utrace.c | 5 +++--
- 2 files changed, 4 insertions(+), 3 deletions(-)
-
-diff --git a/include/linux/utrace.h b/include/linux/utrace.h
-index 9ac0b1b..9a2e2f4 100644
---- a/include/linux/utrace.h
-+++ b/include/linux/utrace.h
-@@ -99,7 +99,7 @@ int utrace_get_signal(struct task_struct *, struct pt_regs *,
- void utrace_report_clone(unsigned long, struct task_struct *);
- void utrace_finish_vfork(struct task_struct *);
- void utrace_report_exit(long *exit_code);
--void utrace_report_death(struct task_struct *, struct utrace *, bool, int);
-+void utrace_report_death(struct task_struct *, bool, int);
- void utrace_report_jctl(int notify, int type);
- void utrace_report_exec(struct linux_binfmt *, struct linux_binprm *,
- struct pt_regs *regs);
-diff --git a/kernel/utrace.c b/kernel/utrace.c
-index ef856c9..1e750ad 100644
---- a/kernel/utrace.c
-+++ b/kernel/utrace.c
-@@ -1759,9 +1759,10 @@ void utrace_report_exit(long *exit_code)
- * For this reason, utrace_release_task checks for the event bits that get
- * us here, and delays its cleanup for us to do.
- */
--void utrace_report_death(struct task_struct *task, struct utrace *utrace,
-- bool group_dead, int signal)
-+void utrace_report_death(struct task_struct *task, bool group_dead, int signal)
- {
-+ struct utrace *utrace = task_utrace_struct(task);
++ if (return_ka->sa.sa_handler == SIG_DFL) {
++ /*
++ * We'll do signr's normal default action.
++ * For ignore, we'll fall through below.
++ * For stop/death, break locks and returns it.
++ */
++ if (likely(signr) && !sig_kernel_ignore(signr))
++ break;
++ } else if (return_ka->sa.sa_handler != SIG_IGN &&
++ likely(signr)) {
++ /*
++ * Complete the bookkeeping after the report.
++ * The handler will run. If an engine wanted to
++ * stop or step, then make sure we do another
++ * report after signal handler setup.
++ */
++ if (report.action != UTRACE_RESUME)
++ report.action = UTRACE_INTERRUPT;
++ finish_report(task, utrace, &report, true);
+
- INIT_REPORT(report);
-
- BUG_ON(!task->exit_state);
---
-1.5.5.1
-
-
-From oleg at redhat.com Mon Nov 21 15:06:33 2011
-Return-Path: oleg at redhat.com
-Received: from zmta01.collab.prod.int.phx2.redhat.com (LHLO
- zmta01.collab.prod.int.phx2.redhat.com) (10.5.5.31) by
- zmail13.collab.prod.int.phx2.redhat.com with LMTP; Mon, 21 Nov 2011
- 15:06:33 -0500 (EST)
-Received: from localhost (localhost.localdomain [127.0.0.1])
- by zmta01.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 62E1BD50A3;
- Mon, 21 Nov 2011 15:06:33 -0500 (EST)
-Received: from zmta01.collab.prod.int.phx2.redhat.com ([127.0.0.1])
- by localhost (zmta01.collab.prod.int.phx2.redhat.com [127.0.0.1]) (amavisd-new, port 10024)
- with ESMTP id FB36YkOAHDc5; Mon, 21 Nov 2011 15:06:33 -0500 (EST)
-Received: from int-mx12.intmail.prod.int.phx2.redhat.com (int-mx12.intmail.prod.int.phx2.redhat.com [10.5.11.25])
- by zmta01.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 4D161D4C39;
- Mon, 21 Nov 2011 15:06:33 -0500 (EST)
-Received: from tranklukator.englab.brq.redhat.com (dhcp-1-232.brq.redhat.com [10.34.1.232])
- by int-mx12.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with SMTP id pALK6VBW022130;
- Mon, 21 Nov 2011 15:06:31 -0500
-Received: by tranklukator.englab.brq.redhat.com (nbSMTP-1.00) for uid 500
- oleg at redhat.com; Mon, 21 Nov 2011 21:01:52 +0100 (CET)
-Date: Mon, 21 Nov 2011 21:01:50 +0100
-From: Oleg Nesterov <oleg at redhat.com>
-To: Dave Jones <davej at redhat.com>, "Frank Ch. Eigler" <fche at redhat.com>,
- Josh Boyer <jwboyer at redhat.com>, Josh Stone <jistone at redhat.com>,
- Kyle McMartin <kmcmartin at redhat.com>
-Cc: kernel at lists.fedoraproject.org
-Subject: [PATCH 08/33] restore the DEATH/REAP utrace hooks
-Message-ID: <20111121200150.GA27790 at redhat.com>
-MIME-Version: 1.0
-Content-Type: text/plain; charset=us-ascii
-Content-Disposition: inline
-In-Reply-To: <20111121200039.GA27699 at redhat.com>
-User-Agent: Mutt/1.5.18 (2008-05-17)
-X-Scanned-By: MIMEDefang 2.68 on 10.5.11.25
-Status: RO
-Content-Length: 2346
-Lines: 70
-
-Restore the necessary hooks in release_task() and exit_notify(),
-add the corresponding helpers into utrace.h.
-
-Note: the @signal argument passed to ->report_death() does not
-match the previous behaviour. I think this shouldn't affect the
-current users, and I bet nobody can really understand what this
-magic argument should actually mean anyway.
-
-Signed-off-by: Oleg Nesterov <oleg at redhat.com>
----
- include/linux/utrace.h | 22 ++++++++++++++++++++++
- kernel/exit.c | 4 ++++
- 2 files changed, 26 insertions(+), 0 deletions(-)
-
-diff --git a/include/linux/utrace.h b/include/linux/utrace.h
-index 9a2e2f4..cf13839 100644
---- a/include/linux/utrace.h
-+++ b/include/linux/utrace.h
-@@ -697,4 +697,26 @@ static inline __must_check int utrace_barrier_pid(struct pid *pid,
-
- #endif /* CONFIG_UTRACE */
-
-+static inline void utrace_release_task(struct task_struct *task)
-+{
-+ /* see utrace_add_engine() about this barrier */
-+ smp_mb();
-+ if (task_utrace_flags(task))
-+ utrace_maybe_reap(task, task_utrace_struct(task), true);
-+}
++ if (unlikely(report.result & UTRACE_SIGNAL_HOLD))
++ push_back_signal(task, info);
++ else
++ spin_lock_irq(&task->sighand->siglock);
+
-+static inline void utrace_exit_notify(struct task_struct *task,
-+ int signal, int group_dead)
-+{
-+ /*
-+ * If utrace_set_events() was just called to enable
-+ * UTRACE_EVENT(DEATH), then we are obliged to call
-+ * utrace_report_death() and not miss it. utrace_set_events()
-+ * checks @task->exit_state under tasklist_lock to synchronize
-+ * with exit_notify(), the caller.
-+ */
-+ if (task_utrace_flags(task) & _UTRACE_DEATH_EVENTS)
-+ utrace_report_death(task, group_dead, signal);
-+}
++ /*
++ * We do the SA_ONESHOT work here since the
++ * normal path will only touch *return_ka now.
++ */
++ if (unlikely(return_ka->sa.sa_flags & SA_ONESHOT)) {
++ return_ka->sa.sa_flags &= ~SA_ONESHOT;
++ if (likely(valid_signal(signr))) {
++ ka = &task->sighand->action[signr - 1];
++ ka->sa.sa_handler = SIG_DFL;
++ }
++ }
+
- #endif /* linux/utrace.h */
-diff --git a/kernel/exit.c b/kernel/exit.c
-index f5eba3b..746c5df 100644
---- a/kernel/exit.c
-+++ b/kernel/exit.c
-@@ -168,6 +168,8 @@ void release_task(struct task_struct * p)
- struct task_struct *leader;
- int zap_leader;
- repeat:
-+ utrace_release_task(p);
++ return signr;
++ }
+
- /* don't need to get the RCU readlock here - the process is dead and
- * can't be modifying its own credentials. But shut RCU-lockdep up */
- rcu_read_lock();
-@@ -858,6 +860,8 @@ static void exit_notify(struct task_struct *tsk, int group_dead)
- wake_up_process(tsk->signal->group_exit_task);
- write_unlock_irq(&tasklist_lock);
-
-+ utrace_exit_notify(tsk, autoreap ? -1 : SIGCHLD, group_dead);
++ /* Fall through for an ignored signal. */
+
- /* If the process is dead, release it - nobody will wait for it */
- if (autoreap)
- release_task(tsk);
---
-1.5.5.1
-
-
-From oleg at redhat.com Mon Nov 21 15:06:35 2011
-Return-Path: oleg at redhat.com
-Received: from zmta03.collab.prod.int.phx2.redhat.com (LHLO
- zmta03.collab.prod.int.phx2.redhat.com) (10.5.5.33) by
- zmail13.collab.prod.int.phx2.redhat.com with LMTP; Mon, 21 Nov 2011
- 15:06:35 -0500 (EST)
-Received: from localhost (localhost.localdomain [127.0.0.1])
- by zmta03.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id D4AB3D8761;
- Mon, 21 Nov 2011 15:06:35 -0500 (EST)
-Received: from zmta03.collab.prod.int.phx2.redhat.com ([127.0.0.1])
- by localhost (zmta03.collab.prod.int.phx2.redhat.com [127.0.0.1]) (amavisd-new, port 10024)
- with ESMTP id TiShkH1f6rnj; Mon, 21 Nov 2011 15:06:35 -0500 (EST)
-Received: from int-mx10.intmail.prod.int.phx2.redhat.com (int-mx10.intmail.prod.int.phx2.redhat.com [10.5.11.23])
- by zmta03.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id BCB85D877F;
- Mon, 21 Nov 2011 15:06:35 -0500 (EST)
-Received: from tranklukator.englab.brq.redhat.com (dhcp-1-232.brq.redhat.com [10.34.1.232])
- by int-mx10.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with SMTP id pALK6XNu015312;
- Mon, 21 Nov 2011 15:06:34 -0500
-Received: by tranklukator.englab.brq.redhat.com (nbSMTP-1.00) for uid 500
- oleg at redhat.com; Mon, 21 Nov 2011 21:01:54 +0100 (CET)
-Date: Mon, 21 Nov 2011 21:01:52 +0100
-From: Oleg Nesterov <oleg at redhat.com>
-To: Dave Jones <davej at redhat.com>, "Frank Ch. Eigler" <fche at redhat.com>,
- Josh Boyer <jwboyer at redhat.com>, Josh Stone <jistone at redhat.com>,
- Kyle McMartin <kmcmartin at redhat.com>
-Cc: kernel at lists.fedoraproject.org
-Subject: [PATCH 09/33] utrace: remove jobctl bits
-Message-ID: <20111121200152.GA27793 at redhat.com>
-MIME-Version: 1.0
-Content-Type: text/plain; charset=us-ascii
-Content-Disposition: inline
-In-Reply-To: <20111121200039.GA27699 at redhat.com>
-User-Agent: Mutt/1.5.18 (2008-05-17)
-X-Scanned-By: MIMEDefang 2.68 on 10.5.11.23
-Status: RO
-Content-Length: 1888
-Lines: 56
-
-- change utrace_get_signal() to check JOBCTL_STOP_PENDING instead of
- signal->group_stop_count. With the recent changes group_stop_count
- doesn't necessarily mean this task should participate in group stop.
-
-- remove the "participate in group stop" code from utrace_wakeup() and
- utrace_stop(), this is no longer needed and wrong.
-
-Signed-off-by: Oleg Nesterov <oleg at redhat.com>
----
- kernel/utrace.c | 16 ++--------------
- 1 files changed, 2 insertions(+), 14 deletions(-)
-
-diff --git a/kernel/utrace.c b/kernel/utrace.c
-index 1e750ad..5d3974e 100644
---- a/kernel/utrace.c
-+++ b/kernel/utrace.c
-@@ -648,11 +648,7 @@ static void utrace_wakeup(struct task_struct *target, struct utrace *utrace)
- {
- lockdep_assert_held(&utrace->lock);
- spin_lock_irq(&target->sighand->siglock);
-- if (target->signal->flags & SIGNAL_STOP_STOPPED ||
-- target->signal->group_stop_count)
-- target->state = TASK_STOPPED;
-- else
-- wake_up_state(target, __TASK_TRACED);
-+ wake_up_state(target, __TASK_TRACED);
- spin_unlock_irq(&target->sighand->siglock);
- }
-
-@@ -805,14 +801,6 @@ relock:
-
- __set_current_state(TASK_TRACED);
-
-- /*
-- * If there is a group stop in progress,
-- * we must participate in the bookkeeping.
-- */
-- if (unlikely(task->signal->group_stop_count) &&
-- !--task->signal->group_stop_count)
-- task->signal->flags = SIGNAL_STOP_STOPPED;
--
- spin_unlock_irq(&task->sighand->siglock);
- spin_unlock(&utrace->lock);
-
-@@ -2037,7 +2025,7 @@ int utrace_get_signal(struct task_struct *task, struct pt_regs *regs,
- ka = NULL;
- memset(return_ka, 0, sizeof *return_ka);
- } else if (!(task->utrace_flags & UTRACE_EVENT_SIGNAL_ALL) ||
-- unlikely(task->signal->group_stop_count)) {
-+ unlikely(task->jobctl & JOBCTL_STOP_PENDING)) {
- /*
- * If no engine is interested in intercepting signals or
- * we must stop, let the caller just dequeue them normally
---
-1.5.5.1
-
-
-From oleg at redhat.com Mon Nov 21 15:06:38 2011
-Return-Path: oleg at redhat.com
-Received: from zmta03.collab.prod.int.phx2.redhat.com (LHLO
- zmta03.collab.prod.int.phx2.redhat.com) (10.5.5.33) by
- zmail13.collab.prod.int.phx2.redhat.com with LMTP; Mon, 21 Nov 2011
- 15:06:38 -0500 (EST)
-Received: from localhost (localhost.localdomain [127.0.0.1])
- by zmta03.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 425CED8707;
- Mon, 21 Nov 2011 15:06:38 -0500 (EST)
-Received: from zmta03.collab.prod.int.phx2.redhat.com ([127.0.0.1])
- by localhost (zmta03.collab.prod.int.phx2.redhat.com [127.0.0.1]) (amavisd-new, port 10024)
- with ESMTP id 9paRC+dwSflh; Mon, 21 Nov 2011 15:06:38 -0500 (EST)
-Received: from int-mx01.intmail.prod.int.phx2.redhat.com (int-mx01.intmail.prod.int.phx2.redhat.com [10.5.11.11])
- by zmta03.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 2DDBCD8410;
- Mon, 21 Nov 2011 15:06:38 -0500 (EST)
-Received: from tranklukator.englab.brq.redhat.com (dhcp-1-232.brq.redhat.com [10.34.1.232])
- by int-mx01.intmail.prod.int.phx2.redhat.com (8.13.8/8.13.8) with SMTP id pALK6a31031355;
- Mon, 21 Nov 2011 15:06:36 -0500
-Received: by tranklukator.englab.brq.redhat.com (nbSMTP-1.00) for uid 500
- oleg at redhat.com; Mon, 21 Nov 2011 21:01:56 +0100 (CET)
-Date: Mon, 21 Nov 2011 21:01:55 +0100
-From: Oleg Nesterov <oleg at redhat.com>
-To: Dave Jones <davej at redhat.com>, "Frank Ch. Eigler" <fche at redhat.com>,
- Josh Boyer <jwboyer at redhat.com>, Josh Stone <jistone at redhat.com>,
- Kyle McMartin <kmcmartin at redhat.com>
-Cc: kernel at lists.fedoraproject.org
-Subject: [PATCH 10/33] ptrace: take ->siglock around s/TRACED/RUNNING/
-Message-ID: <20111121200155.GA27801 at redhat.com>
-MIME-Version: 1.0
-Content-Type: text/plain; charset=us-ascii
-Content-Disposition: inline
-In-Reply-To: <20111121200039.GA27699 at redhat.com>
-User-Agent: Mutt/1.5.18 (2008-05-17)
-X-Scanned-By: MIMEDefang 2.67 on 10.5.11.11
-Status: RO
-Content-Length: 1624
-Lines: 56
-
-change ptrace_resume() and ptrace_stop() to take ->siglock around changing
-task->state from TRACED to RUNNING.
-
-With this patch __TASK_TRACED/STOPPED bits are fully protected by ->siglock,
-nobody can set or clear these bits without ->siglock held.
-
-Signed-off-by: Oleg Nesterov <oleg at redhat.com>
----
- kernel/ptrace.c | 8 +++++++-
- kernel/signal.c | 3 +++
- 2 files changed, 10 insertions(+), 1 deletions(-)
-
-diff --git a/kernel/ptrace.c b/kernel/ptrace.c
-index 24d0447..daf47bc 100644
---- a/kernel/ptrace.c
-+++ b/kernel/ptrace.c
-@@ -589,6 +589,8 @@ static int ptrace_setsiginfo(struct task_struct *child, const siginfo_t *info)
- static int ptrace_resume(struct task_struct *child, long request,
- unsigned long data)
- {
-+ unsigned long flags;
++ case UTRACE_SIGNAL_IGN:
++ case UTRACE_SIGNAL_REPORT:
++ default:
++ /*
++ * If the signal is being ignored, then we are on the way
++ * directly back to user mode. We can stop here, or step,
++ * as in utrace_resume(), above. After we've dealt with that,
++ * our caller will relock and come back through here.
++ */
++ finish_resume_report(task, utrace, &report);
+
- if (!valid_signal(data))
- return -EIO;
-
-@@ -617,7 +619,11 @@ static int ptrace_resume(struct task_struct *child, long request,
- }
-
- child->exit_code = data;
-- wake_up_state(child, __TASK_TRACED);
++ if (unlikely(fatal_signal_pending(task))) {
++ /*
++ * The only reason we woke up now was because of a
++ * SIGKILL. Don't do normal dequeuing in case it
++ * might get a signal other than SIGKILL. That would
++ * perturb the death state so it might differ from
++ * what the debugger would have allowed to happen.
++ * Instead, pluck out just the SIGKILL to be sure
++ * we'll die immediately with nothing else different
++ * from the quiescent state the debugger wanted us in.
++ */
++ sigset_t sigkill_only;
++ siginitsetinv(&sigkill_only, sigmask(SIGKILL));
++ spin_lock_irq(&task->sighand->siglock);
++ signr = dequeue_signal(task, &sigkill_only, info);
++ BUG_ON(signr != SIGKILL);
++ *return_ka = task->sighand->action[SIGKILL - 1];
++ return signr;
++ }
+
-+ if (lock_task_sighand(child, &flags)) {
-+ wake_up_state(child, __TASK_TRACED);
-+ unlock_task_sighand(child, &flags);
++ if (unlikely(report.result & UTRACE_SIGNAL_HOLD)) {
++ push_back_signal(task, info);
++ spin_unlock_irq(&task->sighand->siglock);
++ }
++
++ return -1;
+ }
-
- return 0;
- }
-diff --git a/kernel/signal.c b/kernel/signal.c
-index 8594cb2..4512bb4 100644
---- a/kernel/signal.c
-+++ b/kernel/signal.c
-@@ -1886,7 +1886,10 @@ static void ptrace_stop(int exit_code, int why, int clear_code, siginfo_t *info)
- if (gstop_done)
- do_notify_parent_cldstop(current, false, why);
-
-+ spin_lock_irq(¤t->sighand->siglock);
- __set_current_state(TASK_RUNNING);
-+ spin_unlock_irq(¤t->sighand->siglock);
+
- if (clear_code)
- current->exit_code = 0;
- read_unlock(&tasklist_lock);
---
-1.5.5.1
-
-
-From oleg at redhat.com Mon Nov 21 15:06:40 2011
-Return-Path: oleg at redhat.com
-Received: from zmta03.collab.prod.int.phx2.redhat.com (LHLO
- zmta03.collab.prod.int.phx2.redhat.com) (10.5.5.33) by
- zmail13.collab.prod.int.phx2.redhat.com with LMTP; Mon, 21 Nov 2011
- 15:06:40 -0500 (EST)
-Received: from localhost (localhost.localdomain [127.0.0.1])
- by zmta03.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id B0D19D877F;
- Mon, 21 Nov 2011 15:06:40 -0500 (EST)
-Received: from zmta03.collab.prod.int.phx2.redhat.com ([127.0.0.1])
- by localhost (zmta03.collab.prod.int.phx2.redhat.com [127.0.0.1]) (amavisd-new, port 10024)
- with ESMTP id 3jv-cDBpMP5B; Mon, 21 Nov 2011 15:06:40 -0500 (EST)
-Received: from int-mx02.intmail.prod.int.phx2.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12])
- by zmta03.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 9C98DD8410;
- Mon, 21 Nov 2011 15:06:40 -0500 (EST)
-Received: from tranklukator.englab.brq.redhat.com (dhcp-1-232.brq.redhat.com [10.34.1.232])
- by int-mx02.intmail.prod.int.phx2.redhat.com (8.13.8/8.13.8) with SMTP id pALK6cHF029573;
- Mon, 21 Nov 2011 15:06:39 -0500
-Received: by tranklukator.englab.brq.redhat.com (nbSMTP-1.00) for uid 500
- oleg at redhat.com; Mon, 21 Nov 2011 21:01:59 +0100 (CET)
-Date: Mon, 21 Nov 2011 21:01:57 +0100
-From: Oleg Nesterov <oleg at redhat.com>
-To: Dave Jones <davej at redhat.com>, "Frank Ch. Eigler" <fche at redhat.com>,
- Josh Boyer <jwboyer at redhat.com>, Josh Stone <jistone at redhat.com>,
- Kyle McMartin <kmcmartin at redhat.com>
-Cc: kernel at lists.fedoraproject.org
-Subject: [PATCH 11/33] introduce wake_up_quiescent()
-Message-ID: <20111121200157.GA27805 at redhat.com>
-MIME-Version: 1.0
-Content-Type: text/plain; charset=us-ascii
-Content-Disposition: inline
-In-Reply-To: <20111121200039.GA27699 at redhat.com>
-User-Agent: Mutt/1.5.18 (2008-05-17)
-X-Scanned-By: MIMEDefang 2.67 on 10.5.11.12
-Status: RO
-Content-Length: 3186
-Lines: 93
-
-No functional changes. Add the new helper, wake_up_quiescent(task, state),
-which simply returns wake_up_state(task, state). Change all callers which
-do wake_up_state(STOPPED/TRACED) to use the new helper. ptrace_stop() is
-a bit special, it does __set_current_state(RUNNING) in the very unlikely
-case, change it as well.
-
-Signed-off-by: Oleg Nesterov <oleg at redhat.com>
----
- include/linux/signal.h | 2 ++
- kernel/ptrace.c | 2 +-
- kernel/signal.c | 12 ++++++++++--
- kernel/utrace.c | 2 +-
- 4 files changed, 14 insertions(+), 4 deletions(-)
-
-diff --git a/include/linux/signal.h b/include/linux/signal.h
-index a822300..2be3712 100644
---- a/include/linux/signal.h
-+++ b/include/linux/signal.h
-@@ -239,6 +239,8 @@ static inline int valid_signal(unsigned long sig)
- struct timespec;
- struct pt_regs;
-
-+extern int wake_up_quiescent(struct task_struct *p, unsigned int state);
++ /*
++ * Complete the bookkeeping after the report.
++ * This sets utrace->resume if UTRACE_STOP was used.
++ */
++ finish_report(task, utrace, &report, true);
+
- extern int next_signal(struct sigpending *pending, sigset_t *mask);
- extern int do_send_sig_info(int sig, struct siginfo *info,
- struct task_struct *p, bool group);
-diff --git a/kernel/ptrace.c b/kernel/ptrace.c
-index daf47bc..8439ef1 100644
---- a/kernel/ptrace.c
-+++ b/kernel/ptrace.c
-@@ -621,7 +621,7 @@ static int ptrace_resume(struct task_struct *child, long request,
- child->exit_code = data;
-
- if (lock_task_sighand(child, &flags)) {
-- wake_up_state(child, __TASK_TRACED);
-+ wake_up_quiescent(child, __TASK_TRACED);
- unlock_task_sighand(child, &flags);
- }
-
-diff --git a/kernel/signal.c b/kernel/signal.c
-index 4512bb4..99a6008 100644
---- a/kernel/signal.c
-+++ b/kernel/signal.c
-@@ -702,6 +702,14 @@ void signal_wake_up(struct task_struct *t, int resume)
- }
-
- /*
-+ * wakes up the STOPPED/TRACED task, must be called with ->siglock held.
-+ */
-+int wake_up_quiescent(struct task_struct *p, unsigned int state)
-+{
-+ return wake_up_state(p, state);
-+}
++ return_ka->sa.sa_handler = SIG_DFL;
+
-+/*
- * Remove signals in mask from the pending set and queue.
- * Returns 1 if any signals were found.
- *
-@@ -888,7 +896,7 @@ static int prepare_signal(int sig, struct task_struct *p, int from_ancestor_ns)
- task_clear_jobctl_pending(t, JOBCTL_STOP_PENDING);
- rm_from_queue(SIG_KERNEL_STOP_MASK, &t->pending);
- if (likely(!(t->ptrace & PT_SEIZED)))
-- wake_up_state(t, __TASK_STOPPED);
-+ wake_up_quiescent(t, __TASK_STOPPED);
- else
- ptrace_trap_notify(t);
- } while_each_thread(p, t);
-@@ -1887,7 +1895,7 @@ static void ptrace_stop(int exit_code, int why, int clear_code, siginfo_t *info)
- do_notify_parent_cldstop(current, false, why);
-
- spin_lock_irq(¤t->sighand->siglock);
-- __set_current_state(TASK_RUNNING);
-+ wake_up_quiescent(current, __TASK_TRACED);
- spin_unlock_irq(¤t->sighand->siglock);
-
- if (clear_code)
-diff --git a/kernel/utrace.c b/kernel/utrace.c
-index 5d3974e..cebc390 100644
---- a/kernel/utrace.c
-+++ b/kernel/utrace.c
-@@ -648,7 +648,7 @@ static void utrace_wakeup(struct task_struct *target, struct utrace *utrace)
- {
- lockdep_assert_held(&utrace->lock);
- spin_lock_irq(&target->sighand->siglock);
-- wake_up_state(target, __TASK_TRACED);
-+ wake_up_quiescent(target, __TASK_TRACED);
- spin_unlock_irq(&target->sighand->siglock);
- }
-
---
-1.5.5.1
-
-
-From oleg at redhat.com Mon Nov 21 15:06:43 2011
-Return-Path: oleg at redhat.com
-Received: from zmta01.collab.prod.int.phx2.redhat.com (LHLO
- zmta01.collab.prod.int.phx2.redhat.com) (10.5.5.31) by
- zmail13.collab.prod.int.phx2.redhat.com with LMTP; Mon, 21 Nov 2011
- 15:06:43 -0500 (EST)
-Received: from localhost (localhost.localdomain [127.0.0.1])
- by zmta01.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 6488BD514C;
- Mon, 21 Nov 2011 15:06:43 -0500 (EST)
-Received: from zmta01.collab.prod.int.phx2.redhat.com ([127.0.0.1])
- by localhost (zmta01.collab.prod.int.phx2.redhat.com [127.0.0.1]) (amavisd-new, port 10024)
- with ESMTP id TKQUj5mU4NYM; Mon, 21 Nov 2011 15:06:43 -0500 (EST)
-Received: from int-mx10.intmail.prod.int.phx2.redhat.com (int-mx10.intmail.prod.int.phx2.redhat.com [10.5.11.23])
- by zmta01.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 4EC82D5000;
- Mon, 21 Nov 2011 15:06:43 -0500 (EST)
-Received: from tranklukator.englab.brq.redhat.com (dhcp-1-232.brq.redhat.com [10.34.1.232])
- by int-mx10.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with SMTP id pALK6fp6015343;
- Mon, 21 Nov 2011 15:06:41 -0500
-Received: by tranklukator.englab.brq.redhat.com (nbSMTP-1.00) for uid 500
- oleg at redhat.com; Mon, 21 Nov 2011 21:02:02 +0100 (CET)
-Date: Mon, 21 Nov 2011 21:01:59 +0100
-From: Oleg Nesterov <oleg at redhat.com>
-To: Dave Jones <davej at redhat.com>, "Frank Ch. Eigler" <fche at redhat.com>,
- Josh Boyer <jwboyer at redhat.com>, Josh Stone <jistone at redhat.com>,
- Kyle McMartin <kmcmartin at redhat.com>
-Cc: kernel at lists.fedoraproject.org
-Subject: [PATCH 12/33] introduce ptrace_signal_wake_up()
-Message-ID: <20111121200159.GA27812 at redhat.com>
-MIME-Version: 1.0
-Content-Type: text/plain; charset=us-ascii
-Content-Disposition: inline
-In-Reply-To: <20111121200039.GA27699 at redhat.com>
-User-Agent: Mutt/1.5.18 (2008-05-17)
-X-Scanned-By: MIMEDefang 2.68 on 10.5.11.23
-Status: RO
-Content-Length: 3670
-Lines: 113
-
-Add the new helper, ptrace_signal_wake_up(), change ptrace.c/signal.c
-to use it instead of signal_wake_up() to wake up a STOPPED/TRACED task.
-
-The new helper does almost the same, except:
-
- - it doesn't use the TASK_WAKEKILL bit to wake up the TRACED
- or STOPPED task, it uses __TASK_STOPPED | __TASK_TRACED
- explicitly. This is what ptrace actually wants, it should
- never wake up a TASK_KILLABLE task.
-
- This should be cleanuped upatream, signal_wake_up() should
- take the state as an argument, not a boolean. Until then
- we add a new static helper.
-
- - it uses wake_up_quiescent() instead of wake_up_state().
-
-Thereafter every change from STOPPED/TRACED to RUNNING is done via
-wake_up_quiescent().
-
-Signed-off-by: Oleg Nesterov <oleg at redhat.com>
----
- include/linux/ptrace.h | 1 +
- kernel/ptrace.c | 20 ++++++++++++++++----
- kernel/signal.c | 2 +-
- 3 files changed, 18 insertions(+), 5 deletions(-)
-
-diff --git a/include/linux/ptrace.h b/include/linux/ptrace.h
-index 800f113..6d9282a 100644
---- a/include/linux/ptrace.h
-+++ b/include/linux/ptrace.h
-@@ -113,6 +113,7 @@
- #include <linux/compiler.h> /* For unlikely. */
- #include <linux/sched.h> /* For struct task_struct. */
-
-+extern void ptrace_signal_wake_up(struct task_struct *p, int quiescent);
-
- extern long arch_ptrace(struct task_struct *child, long request,
- unsigned long addr, unsigned long data);
-diff --git a/kernel/ptrace.c b/kernel/ptrace.c
-index 8439ef1..a464ab5 100644
---- a/kernel/ptrace.c
-+++ b/kernel/ptrace.c
-@@ -25,6 +25,18 @@
- #include <linux/hw_breakpoint.h>
- #include <linux/cn_proc.h>
-
-+void ptrace_signal_wake_up(struct task_struct *p, int quiescent)
++ /*
++ * If this signal is fatal, si_signo gets through as exit_code.
++ * We can't allow a completely bogus value there or else core
++ * kernel code can freak out. (If an engine wants to control
++ * the exit_code value exactly, it can do so in report_exit.)
++ * We'll produce a big complaint in dmesg, but won't crash.
++ * That's nicer for debugging your utrace engine.
++ */
++ if (unlikely(info->si_signo & 0x80)) {
++ WARN(1, "utrace engine left bogus si_signo value!");
++ info->si_signo = SIGTRAP;
++ }
++
++ if (unlikely(report.result & UTRACE_SIGNAL_HOLD))
++ push_back_signal(task, info);
++ else
++ spin_lock_irq(&task->sighand->siglock);
++
++ if (sig_kernel_stop(signr))
++ task->jobctl |= JOBCTL_STOP_DEQUEUED;
++
++ return signr;
++}
++
++/*
++ * This gets called after a signal handler has been set up.
++ * We set a flag so the next report knows it happened.
++ * If we're already stepping, make sure we do a report_signal.
++ * If not, make sure we get into utrace_resume() where we can
++ * clear the signal_handler flag before resuming.
++ */
++void utrace_signal_handler(struct task_struct *task, int stepping)
+{
-+ unsigned int state;
++ struct utrace *utrace = task_utrace_struct(task);
+
-+ set_tsk_thread_flag(p, TIF_SIGPENDING);
++ spin_lock(&utrace->lock);
+
-+ state = TASK_INTERRUPTIBLE;
-+ if (quiescent)
-+ state |= (__TASK_STOPPED | __TASK_TRACED);
-+ if (!wake_up_quiescent(p, state))
-+ kick_process(p);
++ utrace->signal_handler = 1;
++ if (utrace->resume > UTRACE_INTERRUPT) {
++ if (stepping) {
++ utrace->resume = UTRACE_INTERRUPT;
++ set_tsk_thread_flag(task, TIF_SIGPENDING);
++ } else if (utrace->resume == UTRACE_RESUME) {
++ set_tsk_thread_flag(task, TIF_NOTIFY_RESUME);
++ }
++ }
++
++ spin_unlock(&utrace->lock);
+}
-
- static int ptrace_trapping_sleep_fn(void *flags)
- {
-@@ -106,7 +118,7 @@ void __ptrace_unlink(struct task_struct *child)
- * TASK_KILLABLE sleeps.
- */
- if (child->jobctl & JOBCTL_STOP_PENDING || task_is_traced(child))
-- signal_wake_up(child, task_is_traced(child));
-+ ptrace_signal_wake_up(child, task_is_traced(child));
-
- spin_unlock(&child->sighand->siglock);
- }
-@@ -296,7 +308,7 @@ static int ptrace_attach(struct task_struct *task, long request,
- */
- if (task_is_stopped(task) &&
- task_set_jobctl_pending(task, JOBCTL_TRAP_STOP | JOBCTL_TRAPPING))
-- signal_wake_up(task, 1);
-+ ptrace_signal_wake_up(task, 1);
-
- spin_unlock(&task->sighand->siglock);
-
-@@ -731,7 +743,7 @@ int ptrace_request(struct task_struct *child, long request,
- * tracee into STOP.
- */
- if (likely(task_set_jobctl_pending(child, JOBCTL_TRAP_STOP)))
-- signal_wake_up(child, child->jobctl & JOBCTL_LISTENING);
-+ ptrace_signal_wake_up(child, child->jobctl & JOBCTL_LISTENING);
-
- unlock_task_sighand(child, &flags);
- ret = 0;
-@@ -757,7 +769,7 @@ int ptrace_request(struct task_struct *child, long request,
- * start of this trap and now. Trigger re-trap.
- */
- if (child->jobctl & JOBCTL_TRAP_NOTIFY)
-- signal_wake_up(child, true);
-+ ptrace_signal_wake_up(child, true);
- ret = 0;
- }
- unlock_task_sighand(child, &flags);
-diff --git a/kernel/signal.c b/kernel/signal.c
-index 99a6008..7a47a93 100644
---- a/kernel/signal.c
-+++ b/kernel/signal.c
-@@ -854,7 +854,7 @@ static void ptrace_trap_notify(struct task_struct *t)
- assert_spin_locked(&t->sighand->siglock);
-
- task_set_jobctl_pending(t, JOBCTL_TRAP_NOTIFY);
-- signal_wake_up(t, t->jobctl & JOBCTL_LISTENING);
-+ ptrace_signal_wake_up(t, t->jobctl & JOBCTL_LISTENING);
- }
-
- /*
---
-1.5.5.1
-
-
-From oleg at redhat.com Mon Nov 21 15:06:46 2011
-Return-Path: oleg at redhat.com
-Received: from zmta03.collab.prod.int.phx2.redhat.com (LHLO
- zmta03.collab.prod.int.phx2.redhat.com) (10.5.5.33) by
- zmail13.collab.prod.int.phx2.redhat.com with LMTP; Mon, 21 Nov 2011
- 15:06:46 -0500 (EST)
-Received: from localhost (localhost.localdomain [127.0.0.1])
- by zmta03.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 3A7AED8410;
- Mon, 21 Nov 2011 15:06:46 -0500 (EST)
-Received: from zmta03.collab.prod.int.phx2.redhat.com ([127.0.0.1])
- by localhost (zmta03.collab.prod.int.phx2.redhat.com [127.0.0.1]) (amavisd-new, port 10024)
- with ESMTP id jhYAwDyYmdhg; Mon, 21 Nov 2011 15:06:46 -0500 (EST)
-Received: from int-mx02.intmail.prod.int.phx2.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12])
- by zmta03.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 25D5FD889A;
- Mon, 21 Nov 2011 15:06:46 -0500 (EST)
-Received: from tranklukator.englab.brq.redhat.com (dhcp-1-232.brq.redhat.com [10.34.1.232])
- by int-mx02.intmail.prod.int.phx2.redhat.com (8.13.8/8.13.8) with SMTP id pALK6h29029619;
- Mon, 21 Nov 2011 15:06:44 -0500
-Received: by tranklukator.englab.brq.redhat.com (nbSMTP-1.00) for uid 500
- oleg at redhat.com; Mon, 21 Nov 2011 21:02:04 +0100 (CET)
-Date: Mon, 21 Nov 2011 21:02:02 +0100
-From: Oleg Nesterov <oleg at redhat.com>
-To: Dave Jones <davej at redhat.com>, "Frank Ch. Eigler" <fche at redhat.com>,
- Josh Boyer <jwboyer at redhat.com>, Josh Stone <jistone at redhat.com>,
- Kyle McMartin <kmcmartin at redhat.com>
-Cc: kernel at lists.fedoraproject.org
-Subject: [PATCH 13/33] wait_task_inactive: treat task->state and
- match_state as bitmasks
-Message-ID: <20111121200202.GA27816 at redhat.com>
-MIME-Version: 1.0
-Content-Type: text/plain; charset=us-ascii
-Content-Disposition: inline
-In-Reply-To: <20111121200039.GA27699 at redhat.com>
-User-Agent: Mutt/1.5.18 (2008-05-17)
-X-Scanned-By: MIMEDefang 2.67 on 10.5.11.12
-Status: RO
-Content-Length: 1558
-Lines: 39
-
-Change wait_task_inactive() to check "state & match_state" instead of
-"state == match_state". This should not make any difference, but this
-allows us to add more "stopped" bits which can be set or cleared
-independently.
-
-IOW. wait_task_inactive() assumes that if task->state != 0, it can
-only be changed to TASK_RUNNING. Currently this is true, and in this
-case "state & match_state" continues to work. But, unlike the current
-check, it also works if task->state has other bits set while the caller
-is only interested in, say, __TASK_TRACED.
-
-Note: I think wait_task_inactive() should be cleanuped upstrean anyway,
-nowadays we have TASK_WAKING and task->state != 0 doesn't necessarily
-mean it is TASK_RUNNING. It also makes sense to exclude the !TASK_REPORT
-bits during the check. Finally, probably this patch makes sense anyway
-even without utrace. For example, a stopped _and_ traced thread could
-have task->state = TASK_STOPPED | TASK_TRACED, this can be useful.
-
-Signed-off-by: Oleg Nesterov <oleg at redhat.com>
----
- kernel/sched.c | 2 +-
- 1 files changed, 1 insertions(+), 1 deletions(-)
-
-diff --git a/kernel/sched.c b/kernel/sched.c
-index 0e9344a..84ecc0a 100644
---- a/kernel/sched.c
-+++ b/kernel/sched.c
-@@ -2443,7 +2443,7 @@ unsigned long wait_task_inactive(struct task_struct *p, long match_state)
- * is actually now running somewhere else!
- */
- while (task_running(rq, p)) {
-- if (match_state && unlikely(p->state != match_state))
-+ if (match_state && !likely(p->state & match_state))
- return 0;
- cpu_relax();
- }
---
-1.5.5.1
-
-
-From oleg at redhat.com Mon Nov 21 15:06:48 2011
-Return-Path: oleg at redhat.com
-Received: from zmta01.collab.prod.int.phx2.redhat.com (LHLO
- zmta01.collab.prod.int.phx2.redhat.com) (10.5.5.31) by
- zmail13.collab.prod.int.phx2.redhat.com with LMTP; Mon, 21 Nov 2011
- 15:06:48 -0500 (EST)
-Received: from localhost (localhost.localdomain [127.0.0.1])
- by zmta01.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id C2934D5000;
- Mon, 21 Nov 2011 15:06:48 -0500 (EST)
-Received: from zmta01.collab.prod.int.phx2.redhat.com ([127.0.0.1])
- by localhost (zmta01.collab.prod.int.phx2.redhat.com [127.0.0.1]) (amavisd-new, port 10024)
- with ESMTP id ThT2PyhztDpQ; Mon, 21 Nov 2011 15:06:48 -0500 (EST)
-Received: from int-mx12.intmail.prod.int.phx2.redhat.com (int-mx12.intmail.prod.int.phx2.redhat.com [10.5.11.25])
- by zmta01.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id ACE02D4F84;
- Mon, 21 Nov 2011 15:06:48 -0500 (EST)
-Received: from tranklukator.englab.brq.redhat.com (dhcp-1-232.brq.redhat.com [10.34.1.232])
- by int-mx12.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with SMTP id pALK6kdb022201;
- Mon, 21 Nov 2011 15:06:47 -0500
-Received: by tranklukator.englab.brq.redhat.com (nbSMTP-1.00) for uid 500
- oleg at redhat.com; Mon, 21 Nov 2011 21:02:07 +0100 (CET)
-Date: Mon, 21 Nov 2011 21:02:05 +0100
-From: Oleg Nesterov <oleg at redhat.com>
-To: Dave Jones <davej at redhat.com>, "Frank Ch. Eigler" <fche at redhat.com>,
- Josh Boyer <jwboyer at redhat.com>, Josh Stone <jistone at redhat.com>,
- Kyle McMartin <kmcmartin at redhat.com>
-Cc: kernel at lists.fedoraproject.org
-Subject: [PATCH 14/33] introduce TASK_UTRACED state
-Message-ID: <20111121200205.GA27823 at redhat.com>
-MIME-Version: 1.0
-Content-Type: text/plain; charset=us-ascii
-Content-Disposition: inline
-In-Reply-To: <20111121200039.GA27699 at redhat.com>
-User-Agent: Mutt/1.5.18 (2008-05-17)
-X-Scanned-By: MIMEDefang 2.68 on 10.5.11.25
-Status: RO
-Content-Length: 2913
-Lines: 84
-
-Introduce TASK_UTRACED state, will be used by utrace instead of TASK_TRACED.
-
-Note: this state is reported as "t (tracing stop)" to the user-space to
-avoid the confusion. IOW, it looks like TASK_TRACED in /proc/pid/status.
-
-Signed-off-by: Oleg Nesterov <oleg at redhat.com>
----
- fs/proc/array.c | 11 ++++++-----
- include/linux/sched.h | 20 +++++++++++---------
- 2 files changed, 17 insertions(+), 14 deletions(-)
-
-diff --git a/fs/proc/array.c b/fs/proc/array.c
-index f0c0ea2..e0daec4 100644
---- a/fs/proc/array.c
-+++ b/fs/proc/array.c
-@@ -138,11 +138,12 @@ static const char * const task_state_array[] = {
- "D (disk sleep)", /* 2 */
- "T (stopped)", /* 4 */
- "t (tracing stop)", /* 8 */
-- "Z (zombie)", /* 16 */
-- "X (dead)", /* 32 */
-- "x (dead)", /* 64 */
-- "K (wakekill)", /* 128 */
-- "W (waking)", /* 256 */
-+ "t (tracing stop)", /* 16 (stopped by utrace) */
-+ "Z (zombie)", /* 32 */
-+ "X (dead)", /* 64 */
-+ "x (dead)", /* 128 */
-+ "K (wakekill)", /* 256 */
-+ "W (waking)", /* 512 */
- };
-
- static inline const char *get_task_state(struct task_struct *tsk)
-diff --git a/include/linux/sched.h b/include/linux/sched.h
-index 4d45f93..746b4d3 100644
---- a/include/linux/sched.h
-+++ b/include/linux/sched.h
-@@ -185,16 +185,17 @@ print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq)
- #define TASK_UNINTERRUPTIBLE 2
- #define __TASK_STOPPED 4
- #define __TASK_TRACED 8
-+#define __TASK_UTRACED 16
- /* in tsk->exit_state */
--#define EXIT_ZOMBIE 16
--#define EXIT_DEAD 32
-+#define EXIT_ZOMBIE 32
-+#define EXIT_DEAD 64
- /* in tsk->state again */
--#define TASK_DEAD 64
--#define TASK_WAKEKILL 128
--#define TASK_WAKING 256
--#define TASK_STATE_MAX 512
-+#define TASK_DEAD 128
-+#define TASK_WAKEKILL 256
-+#define TASK_WAKING 512
-+#define TASK_STATE_MAX 1024
-
--#define TASK_STATE_TO_CHAR_STR "RSDTtZXxKW"
-+#define TASK_STATE_TO_CHAR_STR "RSDTtUZXxKW"
-
- extern char ___assert_task_state[1 - 2*!!(
- sizeof(TASK_STATE_TO_CHAR_STR)-1 != ilog2(TASK_STATE_MAX)+1)];
-@@ -203,15 +204,16 @@ extern char ___assert_task_state[1 - 2*!!(
- #define TASK_KILLABLE (TASK_WAKEKILL | TASK_UNINTERRUPTIBLE)
- #define TASK_STOPPED (TASK_WAKEKILL | __TASK_STOPPED)
- #define TASK_TRACED (TASK_WAKEKILL | __TASK_TRACED)
-+#define TASK_UTRACED (TASK_WAKEKILL | __TASK_UTRACED)
-
- /* Convenience macros for the sake of wake_up */
- #define TASK_NORMAL (TASK_INTERRUPTIBLE | TASK_UNINTERRUPTIBLE)
--#define TASK_ALL (TASK_NORMAL | __TASK_STOPPED | __TASK_TRACED)
-+#define TASK_ALL (TASK_NORMAL | __TASK_STOPPED | __TASK_TRACED | __TASK_UTRACED)
-
- /* get_task_state() */
- #define TASK_REPORT (TASK_RUNNING | TASK_INTERRUPTIBLE | \
- TASK_UNINTERRUPTIBLE | __TASK_STOPPED | \
-- __TASK_TRACED)
-+ __TASK_TRACED | __TASK_UTRACED)
-
- #define task_is_traced(task) ((task->state & __TASK_TRACED) != 0)
- #define task_is_stopped(task) ((task->state & __TASK_STOPPED) != 0)
---
-1.5.5.1
-
-
-From oleg at redhat.com Mon Nov 21 15:06:51 2011
-Return-Path: oleg at redhat.com
-Received: from zmta03.collab.prod.int.phx2.redhat.com (LHLO
- zmta03.collab.prod.int.phx2.redhat.com) (10.5.5.33) by
- zmail13.collab.prod.int.phx2.redhat.com with LMTP; Mon, 21 Nov 2011
- 15:06:51 -0500 (EST)
-Received: from localhost (localhost.localdomain [127.0.0.1])
- by zmta03.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 48F35D889C;
- Mon, 21 Nov 2011 15:06:51 -0500 (EST)
-Received: from zmta03.collab.prod.int.phx2.redhat.com ([127.0.0.1])
- by localhost (zmta03.collab.prod.int.phx2.redhat.com [127.0.0.1]) (amavisd-new, port 10024)
- with ESMTP id ZXkDvXaDg0cN; Mon, 21 Nov 2011 15:06:51 -0500 (EST)
-Received: from int-mx12.intmail.prod.int.phx2.redhat.com (int-mx12.intmail.prod.int.phx2.redhat.com [10.5.11.25])
- by zmta03.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 34485D889A;
- Mon, 21 Nov 2011 15:06:51 -0500 (EST)
-Received: from tranklukator.englab.brq.redhat.com (dhcp-1-232.brq.redhat.com [10.34.1.232])
- by int-mx12.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with SMTP id pALK6nlT022211;
- Mon, 21 Nov 2011 15:06:49 -0500
-Received: by tranklukator.englab.brq.redhat.com (nbSMTP-1.00) for uid 500
- oleg at redhat.com; Mon, 21 Nov 2011 21:02:09 +0100 (CET)
-Date: Mon, 21 Nov 2011 21:02:08 +0100
-From: Oleg Nesterov <oleg at redhat.com>
-To: Dave Jones <davej at redhat.com>, "Frank Ch. Eigler" <fche at redhat.com>,
- Josh Boyer <jwboyer at redhat.com>, Josh Stone <jistone at redhat.com>,
- Kyle McMartin <kmcmartin at redhat.com>
-Cc: kernel at lists.fedoraproject.org
-Subject: [PATCH 15/33] utrace: use TASK_UTRACED instead of TASK_TRACED
-Message-ID: <20111121200208.GA27826 at redhat.com>
-MIME-Version: 1.0
-Content-Type: text/plain; charset=us-ascii
-Content-Disposition: inline
-In-Reply-To: <20111121200039.GA27699 at redhat.com>
-User-Agent: Mutt/1.5.18 (2008-05-17)
-X-Scanned-By: MIMEDefang 2.68 on 10.5.11.25
-Status: RO
-Content-Length: 4316
-Lines: 130
-
-Change utrace.c to use TASK_UTRACED instead of TASK_TRACED.
-
-- utrace_stop/utrace_wakeup: simply use the new state
-
-- utrace_do_stop: do not clear STOPPED/TRACED, but add the new
- __TASK_UTRACED bit to state the fact that both ptrace and utrace
- want this task to be stopped
-
-- naturally, do not use task_is_traced() to check if this task was
- stopped by utrace, use the new task_is_utraced() helper which
- checks __TASK_UTRACED.
-
-Signed-off-by: Oleg Nesterov <oleg at redhat.com>
----
- kernel/utrace.c | 26 ++++++++++++++------------
- 1 files changed, 14 insertions(+), 12 deletions(-)
-
-diff --git a/kernel/utrace.c b/kernel/utrace.c
-index cebc390..2097103 100644
---- a/kernel/utrace.c
-+++ b/kernel/utrace.c
-@@ -462,6 +462,8 @@ static void put_detached_list(struct list_head *list)
- */
- #define ENGINE_STOP (1UL << _UTRACE_NEVENTS)
-
-+#define task_is_utraced(task) ((task->state & __TASK_UTRACED) != 0)
+
- static void mark_engine_wants_stop(struct task_struct *task,
- struct utrace_engine *engine)
- {
-@@ -576,7 +578,7 @@ int utrace_set_events(struct task_struct *target,
-
- ret = 0;
- if ((old_flags & ~events) && target != current &&
-- !task_is_stopped_or_traced(target) && !target->exit_state) {
-+ !task_is_utraced(target) && !target->exit_state) {
- /*
- * This barrier ensures that our engine->flags changes
- * have hit before we examine utrace->reporting,
-@@ -623,21 +625,21 @@ static void mark_engine_detached(struct utrace_engine *engine)
- */
- static bool utrace_do_stop(struct task_struct *target, struct utrace *utrace)
- {
-- if (task_is_stopped(target)) {
-+ if (task_is_stopped_or_traced(target)) {
- /*
- * Stopped is considered quiescent; when it wakes up, it will
- * go through utrace_finish_stop() before doing anything else.
- */
- spin_lock_irq(&target->sighand->siglock);
-- if (likely(task_is_stopped(target)))
-- __set_task_state(target, TASK_TRACED);
-+ if (likely(task_is_stopped_or_traced(target)))
-+ target->state |= TASK_UTRACED;
- spin_unlock_irq(&target->sighand->siglock);
- } else if (utrace->resume > UTRACE_REPORT) {
- utrace->resume = UTRACE_REPORT;
- set_notify_resume(target);
- }
-
-- return task_is_traced(target);
-+ return task_is_utraced(target);
- }
-
- /*
-@@ -648,7 +650,7 @@ static void utrace_wakeup(struct task_struct *target, struct utrace *utrace)
- {
- lockdep_assert_held(&utrace->lock);
- spin_lock_irq(&target->sighand->siglock);
-- wake_up_quiescent(target, __TASK_TRACED);
-+ wake_up_quiescent(target, __TASK_UTRACED);
- spin_unlock_irq(&target->sighand->siglock);
- }
-
-@@ -710,7 +712,7 @@ static bool utrace_reset(struct task_struct *task, struct utrace *utrace)
- /*
- * If no more engines want it stopped, wake it up.
- */
-- if (task_is_traced(task) && !(flags & ENGINE_STOP)) {
-+ if (task_is_utraced(task) && !(flags & ENGINE_STOP)) {
- /*
- * It just resumes, so make sure single-step
- * is not left set.
-@@ -749,7 +751,7 @@ void utrace_finish_stop(void)
- }
-
- /*
-- * Perform %UTRACE_STOP, i.e. block in TASK_TRACED until woken up.
-+ * Perform %UTRACE_STOP, i.e. block in TASK_UTRACED until woken up.
- * @task == current, @utrace == current->utrace, which is not locked.
- * Return true if we were woken up by SIGKILL even though some utrace
- * engine may still want us to stay stopped.
-@@ -799,7 +801,7 @@ relock:
- return;
- }
-
-- __set_current_state(TASK_TRACED);
-+ __set_current_state(TASK_UTRACED);
-
- spin_unlock_irq(&task->sighand->siglock);
- spin_unlock(&utrace->lock);
-@@ -809,14 +811,14 @@ relock:
- utrace_finish_stop();
-
- /*
-- * While in TASK_TRACED, we were considered "frozen enough".
-+ * While in TASK_UTRACED, we were considered "frozen enough".
- * Now that we woke up, it's crucial if we're supposed to be
- * frozen that we freeze now before running anything substantial.
- */
- try_to_freeze();
-
- /*
-- * While we were in TASK_TRACED, complete_signal() considered
-+ * While we were in TASK_UTRACED, complete_signal() considered
- * us "uninterested" in signal wakeups. Now make sure our
- * TIF_SIGPENDING state is correct for normal running.
- */
-@@ -1087,7 +1089,7 @@ int utrace_control(struct task_struct *target,
- if (unlikely(IS_ERR(utrace)))
- return PTR_ERR(utrace);
-
-- reset = task_is_traced(target);
-+ reset = task_is_utraced(target);
- ret = 0;
-
- /*
---
-1.5.5.1
-
-
-From oleg at redhat.com Mon Nov 21 15:06:53 2011
-Return-Path: oleg at redhat.com
-Received: from zmta01.collab.prod.int.phx2.redhat.com (LHLO
- zmta01.collab.prod.int.phx2.redhat.com) (10.5.5.31) by
- zmail13.collab.prod.int.phx2.redhat.com with LMTP; Mon, 21 Nov 2011
- 15:06:53 -0500 (EST)
-Received: from localhost (localhost.localdomain [127.0.0.1])
- by zmta01.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id B8F54D5296;
- Mon, 21 Nov 2011 15:06:53 -0500 (EST)
-Received: from zmta01.collab.prod.int.phx2.redhat.com ([127.0.0.1])
- by localhost (zmta01.collab.prod.int.phx2.redhat.com [127.0.0.1]) (amavisd-new, port 10024)
- with ESMTP id PX4Y3H1mw0hD; Mon, 21 Nov 2011 15:06:53 -0500 (EST)
-Received: from int-mx02.intmail.prod.int.phx2.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12])
- by zmta01.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id A4B5CD528E;
- Mon, 21 Nov 2011 15:06:53 -0500 (EST)
-Received: from tranklukator.englab.brq.redhat.com (dhcp-1-232.brq.redhat.com [10.34.1.232])
- by int-mx02.intmail.prod.int.phx2.redhat.com (8.13.8/8.13.8) with SMTP id pALK6pXd029643;
- Mon, 21 Nov 2011 15:06:52 -0500
-Received: by tranklukator.englab.brq.redhat.com (nbSMTP-1.00) for uid 500
- oleg at redhat.com; Mon, 21 Nov 2011 21:02:12 +0100 (CET)
-Date: Mon, 21 Nov 2011 21:02:10 +0100
-From: Oleg Nesterov <oleg at redhat.com>
-To: Dave Jones <davej at redhat.com>, "Frank Ch. Eigler" <fche at redhat.com>,
- Josh Boyer <jwboyer at redhat.com>, Josh Stone <jistone at redhat.com>,
- Kyle McMartin <kmcmartin at redhat.com>
-Cc: kernel at lists.fedoraproject.org
-Subject: [PATCH 16/33] reintroduce tracehook_finish_jctl() as
- utrace_end_stop()
-Message-ID: <20111121200210.GA27829 at redhat.com>
-MIME-Version: 1.0
-Content-Type: text/plain; charset=us-ascii
-Content-Disposition: inline
-In-Reply-To: <20111121200039.GA27699 at redhat.com>
-User-Agent: Mutt/1.5.18 (2008-05-17)
-X-Scanned-By: MIMEDefang 2.67 on 10.5.11.12
-Status: RO
-Content-Length: 2401
-Lines: 77
-
-utrace_finish_stop() is needed to avoid the races with SIGKILL which
-wakes up UTRACED task, and thus it should be called every time after
-the STOPPED/TRACED/UTRACED returns from schedule(), remember that
-TASK_UTRACED can be added while the task is STOPPED/UTRACED.
-
-- change do_signal_state() to call this helper right after schedule(),
- otherwise this logic is broken by the upstream changes
-
-- now that utrace doesn't control TASK_TRACED bit, ptrace_stop() must
- call this helper too.
-
-Signed-off-by: Oleg Nesterov <oleg at redhat.com>
----
- include/linux/utrace.h | 11 +++++++++++
- kernel/signal.c | 5 +++++
- kernel/utrace.c | 2 +-
- 3 files changed, 17 insertions(+), 1 deletions(-)
-
-diff --git a/include/linux/utrace.h b/include/linux/utrace.h
-index cf13839..0279c74 100644
---- a/include/linux/utrace.h
-+++ b/include/linux/utrace.h
-@@ -719,4 +719,15 @@ static inline void utrace_exit_notify(struct task_struct *task,
- utrace_report_death(task, group_dead, signal);
- }
-
+/**
-+ * utrace_end_stop - report about return from STOPPED/TRACED
++ * utrace_prepare_examine - prepare to examine thread state
++ * @target: thread of interest, a &struct task_struct pointer
++ * @engine: engine pointer returned by utrace_attach_task()
++ * @exam: temporary state, a &struct utrace_examiner pointer
+ *
-+ * This is called by do_signal_stop() and ptrace_stop after wakeup.
++ * This call prepares to safely examine the thread @target using
++ * &struct user_regset calls, or direct access to thread-synchronous fields.
++ *
++ * When @target is current, this call is superfluous. When @target is
++ * another thread, it must be held stopped via %UTRACE_STOP by @engine.
++ *
++ * This call may block the caller until @target stays stopped, so it must
++ * be called only after the caller is sure @target is about to unschedule.
++ * This means a zero return from a utrace_control() call on @engine giving
++ * %UTRACE_STOP, or a report_quiesce() or report_signal() callback to
++ * @engine that used %UTRACE_STOP in its return value.
++ *
++ * Returns -%ESRCH if @target is dead or -%EINVAL if %UTRACE_STOP was
++ * not used. If @target has started running again despite %UTRACE_STOP
++ * (for %SIGKILL or a spurious wakeup), this call returns -%EAGAIN.
++ *
++ * When this call returns zero, it's safe to use &struct user_regset
++ * calls and task_user_regset_view() on @target and to examine some of
++ * its fields directly. When the examination is complete, a
++ * utrace_finish_examine() call must follow to check whether it was
++ * completed safely.
+ */
-+static inline void utrace_end_stop(void)
++int utrace_prepare_examine(struct task_struct *target,
++ struct utrace_engine *engine,
++ struct utrace_examiner *exam)
+{
-+ if (task_utrace_flags(current))
-+ utrace_finish_stop();
-+}
-+
- #endif /* linux/utrace.h */
-diff --git a/kernel/signal.c b/kernel/signal.c
-index 7a47a93..ba46eab 100644
---- a/kernel/signal.c
-+++ b/kernel/signal.c
-@@ -1903,6 +1903,8 @@ static void ptrace_stop(int exit_code, int why, int clear_code, siginfo_t *info)
- read_unlock(&tasklist_lock);
- }
-
-+ utrace_end_stop();
-+
- /*
- * While in TASK_TRACED, we were considered "frozen enough".
- * Now that we woke up, it's crucial if we're supposed to be
-@@ -2067,6 +2069,9 @@ static bool do_signal_stop(int signr)
-
- /* Now we don't run again until woken by SIGCONT or SIGKILL */
- schedule();
-+
-+ utrace_end_stop();
++ int ret = 0;
+
- return true;
- } else {
- /*
-diff --git a/kernel/utrace.c b/kernel/utrace.c
-index 2097103..d41b982 100644
---- a/kernel/utrace.c
-+++ b/kernel/utrace.c
-@@ -741,7 +741,7 @@ static bool utrace_reset(struct task_struct *task, struct utrace *utrace)
- void utrace_finish_stop(void)
- {
- /*
-- * If we were task_is_traced() and then SIGKILL'ed, make
-+ * If we were task_is_utraced() and then SIGKILL'ed, make
- * sure we do nothing until the tracer drops utrace->lock.
- */
- if (unlikely(__fatal_signal_pending(current))) {
---
-1.5.5.1
-
-
-From oleg at redhat.com Mon Nov 21 15:06:56 2011
-Return-Path: oleg at redhat.com
-Received: from zmta03.collab.prod.int.phx2.redhat.com (LHLO
- zmta03.collab.prod.int.phx2.redhat.com) (10.5.5.33) by
- zmail13.collab.prod.int.phx2.redhat.com with LMTP; Mon, 21 Nov 2011
- 15:06:56 -0500 (EST)
-Received: from localhost (localhost.localdomain [127.0.0.1])
- by zmta03.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 4A487D83A1;
- Mon, 21 Nov 2011 15:06:56 -0500 (EST)
-Received: from zmta03.collab.prod.int.phx2.redhat.com ([127.0.0.1])
- by localhost (zmta03.collab.prod.int.phx2.redhat.com [127.0.0.1]) (amavisd-new, port 10024)
- with ESMTP id 3vHogPyhojlJ; Mon, 21 Nov 2011 15:06:56 -0500 (EST)
-Received: from int-mx09.intmail.prod.int.phx2.redhat.com (int-mx09.intmail.prod.int.phx2.redhat.com [10.5.11.22])
- by zmta03.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 358A0D88B0;
- Mon, 21 Nov 2011 15:06:56 -0500 (EST)
-Received: from tranklukator.englab.brq.redhat.com (dhcp-1-232.brq.redhat.com [10.34.1.232])
- by int-mx09.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with SMTP id pALK6sHk017679;
- Mon, 21 Nov 2011 15:06:54 -0500
-Received: by tranklukator.englab.brq.redhat.com (nbSMTP-1.00) for uid 500
- oleg at redhat.com; Mon, 21 Nov 2011 21:02:14 +0100 (CET)
-Date: Mon, 21 Nov 2011 21:02:12 +0100
-From: Oleg Nesterov <oleg at redhat.com>
-To: Dave Jones <davej at redhat.com>, "Frank Ch. Eigler" <fche at redhat.com>,
- Josh Boyer <jwboyer at redhat.com>, Josh Stone <jistone at redhat.com>,
- Kyle McMartin <kmcmartin at redhat.com>
-Cc: kernel at lists.fedoraproject.org
-Subject: [PATCH 17/33] teach wake_up_quiescent() to do "selective" wake_up
-Message-ID: <20111121200212.GA27836 at redhat.com>
-MIME-Version: 1.0
-Content-Type: text/plain; charset=us-ascii
-Content-Disposition: inline
-In-Reply-To: <20111121200039.GA27699 at redhat.com>
-User-Agent: Mutt/1.5.18 (2008-05-17)
-X-Scanned-By: MIMEDefang 2.68 on 10.5.11.22
-Status: RO
-Content-Length: 1335
-Lines: 46
-
-Both utrace and ptrace can want the same thread to be quiescent, in this
-case its state is TASK_TRACED | TASK_UTRACED. And this also means that
-this task must not run unless both utrace and ptrace resume it.
-
-Change wake_up_quiescent(p, state) to do "p->state &= ~state" and return
-false unless there is no more "quiescent" bits in task->state.
-
-Signed-off-by: Oleg Nesterov <oleg at redhat.com>
----
- kernel/signal.c | 15 +++++++++++++++
- 1 files changed, 15 insertions(+), 0 deletions(-)
-
-diff --git a/kernel/signal.c b/kernel/signal.c
-index ba46eab..e06f795 100644
---- a/kernel/signal.c
-+++ b/kernel/signal.c
-@@ -701,11 +701,26 @@ void signal_wake_up(struct task_struct *t, int resume)
- kick_process(t);
- }
-
-+#define STATE_QUIESCENT (__TASK_STOPPED | __TASK_TRACED | __TASK_UTRACED)
- /*
- * wakes up the STOPPED/TRACED task, must be called with ->siglock held.
- */
- int wake_up_quiescent(struct task_struct *p, unsigned int state)
- {
-+ unsigned int quiescent = (p->state & STATE_QUIESCENT);
++ if (unlikely(target == current))
++ return 0;
+
-+ WARN_ON(state & ~(STATE_QUIESCENT | TASK_INTERRUPTIBLE));
++ rcu_read_lock();
++ if (unlikely(!engine_wants_stop(engine)))
++ ret = -EINVAL;
++ else if (unlikely(target->exit_state))
++ ret = -ESRCH;
++ else {
++ exam->state = target->state;
++ if (unlikely(exam->state == TASK_RUNNING))
++ ret = -EAGAIN;
++ else
++ get_task_struct(target);
++ }
++ rcu_read_unlock();
+
-+ if (quiescent) {
-+ state &= ~TASK_INTERRUPTIBLE;
-+ if ((quiescent & ~state) != 0) {
-+ p->state &= ~state;
-+ WARN_ON(!(p->state & STATE_QUIESCENT));
-+ WARN_ON(!(p->state & TASK_WAKEKILL));
-+ return 0;
-+ }
++ if (likely(!ret)) {
++ exam->ncsw = wait_task_inactive(target, exam->state);
++ put_task_struct(target);
++ if (unlikely(!exam->ncsw))
++ ret = -EAGAIN;
+ }
+
- return wake_up_state(p, state);
- }
-
---
-1.5.5.1
-
-
-From oleg at redhat.com Mon Nov 21 15:06:59 2011
-Return-Path: oleg at redhat.com
-Received: from zmta01.collab.prod.int.phx2.redhat.com (LHLO
- zmta01.collab.prod.int.phx2.redhat.com) (10.5.5.31) by
- zmail13.collab.prod.int.phx2.redhat.com with LMTP; Mon, 21 Nov 2011
- 15:06:58 -0500 (EST)
-Received: from localhost (localhost.localdomain [127.0.0.1])
- by zmta01.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id C88E2D528B;
- Mon, 21 Nov 2011 15:06:58 -0500 (EST)
-Received: from zmta01.collab.prod.int.phx2.redhat.com ([127.0.0.1])
- by localhost (zmta01.collab.prod.int.phx2.redhat.com [127.0.0.1]) (amavisd-new, port 10024)
- with ESMTP id nNrAxBD5KOME; Mon, 21 Nov 2011 15:06:58 -0500 (EST)
-Received: from int-mx01.intmail.prod.int.phx2.redhat.com (int-mx01.intmail.prod.int.phx2.redhat.com [10.5.11.11])
- by zmta01.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id B2114D51C1;
- Mon, 21 Nov 2011 15:06:58 -0500 (EST)
-Received: from tranklukator.englab.brq.redhat.com (dhcp-1-232.brq.redhat.com [10.34.1.232])
- by int-mx01.intmail.prod.int.phx2.redhat.com (8.13.8/8.13.8) with SMTP id pALK6uQQ031440;
- Mon, 21 Nov 2011 15:06:57 -0500
-Received: by tranklukator.englab.brq.redhat.com (nbSMTP-1.00) for uid 500
- oleg at redhat.com; Mon, 21 Nov 2011 21:02:17 +0100 (CET)
-Date: Mon, 21 Nov 2011 21:02:15 +0100
-From: Oleg Nesterov <oleg at redhat.com>
-To: Dave Jones <davej at redhat.com>, "Frank Ch. Eigler" <fche at redhat.com>,
- Josh Boyer <jwboyer at redhat.com>, Josh Stone <jistone at redhat.com>,
- Kyle McMartin <kmcmartin at redhat.com>
-Cc: kernel at lists.fedoraproject.org
-Subject: [PATCH 18/33] ptrace_stop: do not assume the task is running after
- wake_up_quiescent()
-Message-ID: <20111121200215.GA27839 at redhat.com>
-MIME-Version: 1.0
-Content-Type: text/plain; charset=us-ascii
-Content-Disposition: inline
-In-Reply-To: <20111121200039.GA27699 at redhat.com>
-User-Agent: Mutt/1.5.18 (2008-05-17)
-X-Scanned-By: MIMEDefang 2.67 on 10.5.11.11
-Status: RO
-Content-Length: 956
-Lines: 31
-
-If ptrace_stop() sets TASK_TRACED and then detects we should not stop,
-it can race with utrace_do_stop() which can see TASK_TRACED and add
-TASK_UTRACED. In this case we should stop for utrace needs.
-
-Signed-off-by: Oleg Nesterov <oleg at redhat.com>
----
- kernel/signal.c | 8 ++++++++
- 1 files changed, 8 insertions(+), 0 deletions(-)
-
-diff --git a/kernel/signal.c b/kernel/signal.c
-index e06f795..9348da6 100644
---- a/kernel/signal.c
-+++ b/kernel/signal.c
-@@ -1916,6 +1916,14 @@ static void ptrace_stop(int exit_code, int why, int clear_code, siginfo_t *info)
- if (clear_code)
- current->exit_code = 0;
- read_unlock(&tasklist_lock);
++ return ret;
++}
++EXPORT_SYMBOL_GPL(utrace_prepare_examine);
+
-+ /*
-+ * It is possible that __TASK_UTRACED was added by utrace
-+ * while we were __TASK_TRACED and before we take ->siglock
-+ * for wake_up_quiescent(), we need to block in this case.
-+ * Otherwise this is unnecessary but absolutely harmless.
-+ */
-+ schedule();
- }
-
- utrace_end_stop();
---
-1.5.5.1
-
-
-From oleg at redhat.com Mon Nov 21 15:07:01 2011
-Return-Path: oleg at redhat.com
-Received: from zmta03.collab.prod.int.phx2.redhat.com (LHLO
- zmta03.collab.prod.int.phx2.redhat.com) (10.5.5.33) by
- zmail13.collab.prod.int.phx2.redhat.com with LMTP; Mon, 21 Nov 2011
- 15:07:01 -0500 (EST)
-Received: from localhost (localhost.localdomain [127.0.0.1])
- by zmta03.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 4BF37D889A;
- Mon, 21 Nov 2011 15:07:01 -0500 (EST)
-Received: from zmta03.collab.prod.int.phx2.redhat.com ([127.0.0.1])
- by localhost (zmta03.collab.prod.int.phx2.redhat.com [127.0.0.1]) (amavisd-new, port 10024)
- with ESMTP id 5p8tvlXAqUHr; Mon, 21 Nov 2011 15:07:01 -0500 (EST)
-Received: from int-mx01.intmail.prod.int.phx2.redhat.com (int-mx01.intmail.prod.int.phx2.redhat.com [10.5.11.11])
- by zmta03.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 37C22D8757;
- Mon, 21 Nov 2011 15:07:01 -0500 (EST)
-Received: from tranklukator.englab.brq.redhat.com (dhcp-1-232.brq.redhat.com [10.34.1.232])
- by int-mx01.intmail.prod.int.phx2.redhat.com (8.13.8/8.13.8) with SMTP id pALK6xFJ031449;
- Mon, 21 Nov 2011 15:06:59 -0500
-Received: by tranklukator.englab.brq.redhat.com (nbSMTP-1.00) for uid 500
- oleg at redhat.com; Mon, 21 Nov 2011 21:02:19 +0100 (CET)
-Date: Mon, 21 Nov 2011 21:02:18 +0100
-From: Oleg Nesterov <oleg at redhat.com>
-To: Dave Jones <davej at redhat.com>, "Frank Ch. Eigler" <fche at redhat.com>,
- Josh Boyer <jwboyer at redhat.com>, Josh Stone <jistone at redhat.com>,
- Kyle McMartin <kmcmartin at redhat.com>
-Cc: kernel at lists.fedoraproject.org
-Subject: [PATCH 19/33] get_signal_to_deliver: restore/restructure
- utrace/ptrace signal reporting
-Message-ID: <20111121200218.GA27846 at redhat.com>
-MIME-Version: 1.0
-Content-Type: text/plain; charset=us-ascii
-Content-Disposition: inline
-In-Reply-To: <20111121200039.GA27699 at redhat.com>
-User-Agent: Mutt/1.5.18 (2008-05-17)
-X-Scanned-By: MIMEDefang 2.67 on 10.5.11.11
-Status: RO
-Content-Length: 3556
-Lines: 115
-
-- Reintroduce tracehook_get_signal() as utrace_hook_signal().
-
-- Change get_signal_to_deliver() to call utrace_hook_signal() first,
- before dequeue_signal()
-
-- Always call ptrace_signal() if signal != SIGKILL, no matter whether
- this signal comes from utrace or not.
-
- Since this can change signr again, update "struct k_sigaction *ka"
- in this case.
-
-IOW, roughly, ptrace acts as if it is the last attached engine, it
-takes the final decision about the signal.
-
-Signed-off-by: Oleg Nesterov <oleg at redhat.com>
----
- include/linux/utrace.h | 31 +++++++++++++++++++++++++++++++
- kernel/signal.c | 30 ++++++++++++++++++++----------
- 2 files changed, 51 insertions(+), 10 deletions(-)
-
-diff --git a/include/linux/utrace.h b/include/linux/utrace.h
-index 0279c74..63103e2 100644
---- a/include/linux/utrace.h
-+++ b/include/linux/utrace.h
-@@ -730,4 +730,35 @@ static inline void utrace_end_stop(void)
- utrace_finish_stop();
- }
-
+/**
-+ * utrace_hook_signal - deliver synthetic signal to traced task
-+ * @task: @current
-+ * @regs: task_pt_regs(@current)
-+ * @info: details of synthetic signal
-+ * @return_ka: sigaction for synthetic signal
++ * utrace_finish_examine - complete an examination of thread state
++ * @target: thread of interest, a &struct task_struct pointer
++ * @engine: engine pointer returned by utrace_attach_task()
++ * @exam: pointer passed to utrace_prepare_examine() call
+ *
-+ * Return zero to check for a real pending signal normally.
-+ * Return -1 after releasing the siglock to repeat the check.
-+ * Return a signal number to induce an artificial signal delivery,
-+ * setting *@info and *@return_ka to specify its details and behavior.
++ * This call completes an examination on the thread @target begun by a
++ * paired utrace_prepare_examine() call with the same arguments that
++ * returned success (zero).
+ *
-+ * The @return_ka->sa_handler value controls the disposition of the
-+ * signal, no matter the signal number. For %SIG_DFL, the return value
-+ * is a representative signal to indicate the behavior (e.g. %SIGTERM
-+ * for death, %SIGQUIT for core dump, %SIGSTOP for job control stop,
-+ * %SIGTSTP for stop unless in an orphaned pgrp), but the signal number
-+ * reported will be @info->si_signo instead.
++ * When @target is current, this call is superfluous. When @target is
++ * another thread, this returns zero if @target has remained unscheduled
++ * since the paired utrace_prepare_examine() call returned zero.
+ *
-+ * Called with @task->sighand->siglock held, before dequeuing pending signals.
++ * When this returns an error, any examination done since the paired
++ * utrace_prepare_examine() call is unreliable and the data extracted
++ * should be discarded. The error is -%EINVAL if @engine is not
++ * keeping @target stopped, or -%EAGAIN if @target woke up unexpectedly.
+ */
-+static inline int utrace_hook_signal(struct task_struct *task,
-+ struct pt_regs *regs,
-+ siginfo_t *info,
-+ struct k_sigaction *return_ka)
++int utrace_finish_examine(struct task_struct *target,
++ struct utrace_engine *engine,
++ struct utrace_examiner *exam)
+{
-+ if (unlikely(task_utrace_flags(task)))
-+ return utrace_get_signal(task, regs, info, return_ka);
-+ return 0;
-+}
++ int ret = 0;
+
- #endif /* linux/utrace.h */
-diff --git a/kernel/signal.c b/kernel/signal.c
-index 9348da6..38ea4e6 100644
---- a/kernel/signal.c
-+++ b/kernel/signal.c
-@@ -2242,17 +2242,27 @@ relock:
- for (;;) {
- struct k_sigaction *ka;
-
-- if (unlikely(current->jobctl & JOBCTL_STOP_PENDING) &&
-- do_signal_stop(0))
-+ signr = utrace_hook_signal(current, regs, info, return_ka);
-+ if (unlikely(signr < 0))
- goto relock;
-
-- if (unlikely(current->jobctl & JOBCTL_TRAP_MASK)) {
-- do_jobctl_trap();
-- spin_unlock_irq(&sighand->siglock);
-- goto relock;
-- }
-+ if (unlikely(signr != 0))
-+ ka = return_ka;
-+ else {
-+ if (unlikely(current->jobctl & JOBCTL_STOP_PENDING) &&
-+ do_signal_stop(0))
-+ goto relock;
-
-- signr = dequeue_signal(current, ¤t->blocked, info);
-+ if (unlikely(current->jobctl & JOBCTL_TRAP_MASK)) {
-+ do_jobctl_trap();
-+ spin_unlock_irq(&sighand->siglock);
-+ goto relock;
-+ }
-+
-+ signr = dequeue_signal(current, ¤t->blocked, info);
++ if (unlikely(target == current))
++ return 0;
+
-+ ka = &sighand->action[signr-1];
-+ }
-
- if (!signr)
- break; /* will return 0 */
-@@ -2262,9 +2272,9 @@ relock:
- regs, cookie);
- if (!signr)
- continue;
-- }
-
-- ka = &sighand->action[signr-1];
-+ ka = &sighand->action[signr-1];
-+ }
-
- /* Trace actually delivered signals. */
- trace_signal_deliver(signr, info, ka);
---
-1.5.5.1
-
-
-From oleg at redhat.com Mon Nov 21 15:07:03 2011
-Return-Path: oleg at redhat.com
-Received: from zmta03.collab.prod.int.phx2.redhat.com (LHLO
- zmta03.collab.prod.int.phx2.redhat.com) (10.5.5.33) by
- zmail13.collab.prod.int.phx2.redhat.com with LMTP; Mon, 21 Nov 2011
- 15:07:03 -0500 (EST)
-Received: from localhost (localhost.localdomain [127.0.0.1])
- by zmta03.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id CB362D89E9;
- Mon, 21 Nov 2011 15:07:03 -0500 (EST)
-Received: from zmta03.collab.prod.int.phx2.redhat.com ([127.0.0.1])
- by localhost (zmta03.collab.prod.int.phx2.redhat.com [127.0.0.1]) (amavisd-new, port 10024)
- with ESMTP id wm2JqccPnhsV; Mon, 21 Nov 2011 15:07:03 -0500 (EST)
-Received: from int-mx01.intmail.prod.int.phx2.redhat.com (int-mx01.intmail.prod.int.phx2.redhat.com [10.5.11.11])
- by zmta03.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id B6D25D89DE;
- Mon, 21 Nov 2011 15:07:03 -0500 (EST)
-Received: from tranklukator.englab.brq.redhat.com (dhcp-1-232.brq.redhat.com [10.34.1.232])
- by int-mx01.intmail.prod.int.phx2.redhat.com (8.13.8/8.13.8) with SMTP id pALK715f031455;
- Mon, 21 Nov 2011 15:07:02 -0500
-Received: by tranklukator.englab.brq.redhat.com (nbSMTP-1.00) for uid 500
- oleg at redhat.com; Mon, 21 Nov 2011 21:02:22 +0100 (CET)
-Date: Mon, 21 Nov 2011 21:02:20 +0100
-From: Oleg Nesterov <oleg at redhat.com>
-To: Dave Jones <davej at redhat.com>, "Frank Ch. Eigler" <fche at redhat.com>,
- Josh Boyer <jwboyer at redhat.com>, Josh Stone <jistone at redhat.com>,
- Kyle McMartin <kmcmartin at redhat.com>
-Cc: kernel at lists.fedoraproject.org
-Subject: [PATCH 20/33] utrace_get_signal:
- s/JOBCTL_STOP_PENDING/JOBCTL_PENDING_MASK/
-Message-ID: <20111121200220.GA27849 at redhat.com>
-MIME-Version: 1.0
-Content-Type: text/plain; charset=us-ascii
-Content-Disposition: inline
-In-Reply-To: <20111121200039.GA27699 at redhat.com>
-User-Agent: Mutt/1.5.18 (2008-05-17)
-X-Scanned-By: MIMEDefang 2.67 on 10.5.11.11
-Status: RO
-Content-Length: 1204
-Lines: 33
-
-utrace_get_signal() checks JOBCTL_STOP_PENDING to detect the
-case when we should not try to dequeue the signal but should
-try to participate in the group-stop.
-
-With the recent changes this is not enough, everything which
-contrbutes to recalc_sigpending_tsk() should be respected.
-
-Check JOBCTL_PENDING_MASK instead. This matches the
-JOBCTL_STOP_PENDING | JOBCTL_TRAP_MASK code in the caller,
-get_signal_to_deliver(). Note that this code won't run if
-utrace_get_signal() returns signr > 0.
-
-Signed-off-by: Oleg Nesterov <oleg at redhat.com>
----
- kernel/utrace.c | 2 +-
- 1 files changed, 1 insertions(+), 1 deletions(-)
-
-diff --git a/kernel/utrace.c b/kernel/utrace.c
-index d41b982..0bb0a06 100644
---- a/kernel/utrace.c
-+++ b/kernel/utrace.c
-@@ -2027,7 +2027,7 @@ int utrace_get_signal(struct task_struct *task, struct pt_regs *regs,
- ka = NULL;
- memset(return_ka, 0, sizeof *return_ka);
- } else if (!(task->utrace_flags & UTRACE_EVENT_SIGNAL_ALL) ||
-- unlikely(task->jobctl & JOBCTL_STOP_PENDING)) {
-+ unlikely(task->jobctl & JOBCTL_PENDING_MASK)) {
- /*
- * If no engine is interested in intercepting signals or
- * we must stop, let the caller just dequeue them normally
---
-1.5.5.1
-
-
-From oleg at redhat.com Mon Nov 21 15:07:07 2011
-Return-Path: oleg at redhat.com
-Received: from zmta01.collab.prod.int.phx2.redhat.com (LHLO
- zmta01.collab.prod.int.phx2.redhat.com) (10.5.5.31) by
- zmail13.collab.prod.int.phx2.redhat.com with LMTP; Mon, 21 Nov 2011
- 15:07:06 -0500 (EST)
-Received: from localhost (localhost.localdomain [127.0.0.1])
- by zmta01.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 8C14BD4F84;
- Mon, 21 Nov 2011 15:07:06 -0500 (EST)
-Received: from zmta01.collab.prod.int.phx2.redhat.com ([127.0.0.1])
- by localhost (zmta01.collab.prod.int.phx2.redhat.com [127.0.0.1]) (amavisd-new, port 10024)
- with ESMTP id 39eM-KKZxrZ3; Mon, 21 Nov 2011 15:07:06 -0500 (EST)
-Received: from int-mx12.intmail.prod.int.phx2.redhat.com (int-mx12.intmail.prod.int.phx2.redhat.com [10.5.11.25])
- by zmta01.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 7628FD4FF4;
- Mon, 21 Nov 2011 15:07:06 -0500 (EST)
-Received: from tranklukator.englab.brq.redhat.com (dhcp-1-232.brq.redhat.com [10.34.1.232])
- by int-mx12.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with SMTP id pALK74dN022253;
- Mon, 21 Nov 2011 15:07:04 -0500
-Received: by tranklukator.englab.brq.redhat.com (nbSMTP-1.00) for uid 500
- oleg at redhat.com; Mon, 21 Nov 2011 21:02:25 +0100 (CET)
-Date: Mon, 21 Nov 2011 21:02:23 +0100
-From: Oleg Nesterov <oleg at redhat.com>
-To: Dave Jones <davej at redhat.com>, "Frank Ch. Eigler" <fche at redhat.com>,
- Josh Boyer <jwboyer at redhat.com>, Josh Stone <jistone at redhat.com>,
- Kyle McMartin <kmcmartin at redhat.com>
-Cc: kernel at lists.fedoraproject.org
-Subject: [PATCH 21/33] introduce ptrace_set_syscall_trace()
-Message-ID: <20111121200223.GA27856 at redhat.com>
-MIME-Version: 1.0
-Content-Type: text/plain; charset=us-ascii
-Content-Disposition: inline
-In-Reply-To: <20111121200039.GA27699 at redhat.com>
-User-Agent: Mutt/1.5.18 (2008-05-17)
-X-Scanned-By: MIMEDefang 2.68 on 10.5.11.25
-Status: RO
-Content-Length: 1549
-Lines: 52
-
-No functional changes. Add the new helper, ptrace_set_syscall_trace(),
-which should be used to set/clear TIF_SYSCALL_TRACE in ptrace code.
-Currently it does nothing more.
-
-Signed-off-by: Oleg Nesterov <oleg at redhat.com>
----
- kernel/ptrace.c | 15 ++++++++++-----
- 1 files changed, 10 insertions(+), 5 deletions(-)
-
-diff --git a/kernel/ptrace.c b/kernel/ptrace.c
-index a464ab5..43357e5 100644
---- a/kernel/ptrace.c
-+++ b/kernel/ptrace.c
-@@ -38,6 +38,14 @@ void ptrace_signal_wake_up(struct task_struct *p, int quiescent)
- kick_process(p);
- }
-
-+static void ptrace_set_syscall_trace(struct task_struct *p, bool on)
-+{
-+ if (on)
-+ set_tsk_thread_flag(p, TIF_SYSCALL_TRACE);
++ rcu_read_lock();
++ if (unlikely(!engine_wants_stop(engine)))
++ ret = -EINVAL;
++ else if (unlikely(target->state != exam->state))
++ ret = -EAGAIN;
+ else
-+ clear_tsk_thread_flag(p, TIF_SYSCALL_TRACE);
-+}
-+
- static int ptrace_trapping_sleep_fn(void *flags)
- {
- schedule();
-@@ -418,7 +426,7 @@ static int ptrace_detach(struct task_struct *child, unsigned int data)
-
- /* Architecture-specific hardware disable .. */
- ptrace_disable(child);
-- clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
-+ ptrace_set_syscall_trace(child, false);
-
- write_lock_irq(&tasklist_lock);
- /*
-@@ -606,10 +614,7 @@ static int ptrace_resume(struct task_struct *child, long request,
- if (!valid_signal(data))
- return -EIO;
-
-- if (request == PTRACE_SYSCALL)
-- set_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
-- else
-- clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
-+ ptrace_set_syscall_trace(child, request == PTRACE_SYSCALL);
-
- #ifdef TIF_SYSCALL_EMU
- if (request == PTRACE_SYSEMU || request == PTRACE_SYSEMU_SINGLESTEP)
---
-1.5.5.1
-
-
-From oleg at redhat.com Mon Nov 21 15:07:09 2011
-Return-Path: oleg at redhat.com
-Received: from zmta02.collab.prod.int.phx2.redhat.com (LHLO
- zmta02.collab.prod.int.phx2.redhat.com) (10.5.5.32) by
- zmail13.collab.prod.int.phx2.redhat.com with LMTP; Mon, 21 Nov 2011
- 15:07:09 -0500 (EST)
-Received: from localhost (localhost.localdomain [127.0.0.1])
- by zmta02.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 1ADF01285E4;
- Mon, 21 Nov 2011 15:07:09 -0500 (EST)
-Received: from zmta02.collab.prod.int.phx2.redhat.com ([127.0.0.1])
- by localhost (zmta02.collab.prod.int.phx2.redhat.com [127.0.0.1]) (amavisd-new, port 10024)
- with ESMTP id rF+XegsxVXnL; Mon, 21 Nov 2011 15:07:09 -0500 (EST)
-Received: from int-mx12.intmail.prod.int.phx2.redhat.com (int-mx12.intmail.prod.int.phx2.redhat.com [10.5.11.25])
- by zmta02.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 067E41287D0;
- Mon, 21 Nov 2011 15:07:09 -0500 (EST)
-Received: from tranklukator.englab.brq.redhat.com (dhcp-1-232.brq.redhat.com [10.34.1.232])
- by int-mx12.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with SMTP id pALK76tD022270;
- Mon, 21 Nov 2011 15:07:07 -0500
-Received: by tranklukator.englab.brq.redhat.com (nbSMTP-1.00) for uid 500
- oleg at redhat.com; Mon, 21 Nov 2011 21:02:27 +0100 (CET)
-Date: Mon, 21 Nov 2011 21:02:25 +0100
-From: Oleg Nesterov <oleg at redhat.com>
-To: Dave Jones <davej at redhat.com>, "Frank Ch. Eigler" <fche at redhat.com>,
- Josh Boyer <jwboyer at redhat.com>, Josh Stone <jistone at redhat.com>,
- Kyle McMartin <kmcmartin at redhat.com>
-Cc: kernel at lists.fedoraproject.org
-Subject: [PATCH 22/33] introduce PT_SYSCALL_TRACE flag
-Message-ID: <20111121200225.GA27860 at redhat.com>
-MIME-Version: 1.0
-Content-Type: text/plain; charset=us-ascii
-Content-Disposition: inline
-In-Reply-To: <20111121200039.GA27699 at redhat.com>
-User-Agent: Mutt/1.5.18 (2008-05-17)
-X-Scanned-By: MIMEDefang 2.68 on 10.5.11.25
-Status: RO
-Content-Length: 2268
-Lines: 73
-
-Currently tracehooks assume that if the ptraced task has
-TIF_SYSCALL_TRACE set, the tracee should report the syscall.
-This is not true, this thread flag can be set by utrace.
-
-Add the new internal ptrace flag, PT_SYSCALL_TRACE. Change
-ptrace_set_syscall_trace() to set/clear this bit along with
-TIF_SYSCALL_TRACE, change ptrace_report_syscall() to check
-this flag instead of PT_PTRACED.
-
-Signed-off-by: Oleg Nesterov <oleg at redhat.com>
----
- include/linux/ptrace.h | 3 +++
- include/linux/tracehook.h | 2 +-
- kernel/ptrace.c | 7 +++++--
- 3 files changed, 9 insertions(+), 3 deletions(-)
-
-diff --git a/include/linux/ptrace.h b/include/linux/ptrace.h
-index 6d9282a..c10f610 100644
---- a/include/linux/ptrace.h
-+++ b/include/linux/ptrace.h
-@@ -104,6 +104,8 @@
-
- #define PT_TRACE_MASK 0x000003f4
-
-+#define PT_SYSCALL_TRACE 0x00020000
-+
- /* single stepping state bits (used on ARM and PA-RISC) */
- #define PT_SINGLESTEP_BIT 31
- #define PT_SINGLESTEP (1<<PT_SINGLESTEP_BIT)
-@@ -227,6 +229,7 @@ static inline void ptrace_init_task(struct task_struct *child, bool ptrace)
-
- if (unlikely(ptrace) && current->ptrace) {
- child->ptrace = current->ptrace;
-+ child->ptrace &= ~PT_SYSCALL_TRACE;
- __ptrace_link(child, current->parent);
-
- if (child->ptrace & PT_SEIZED)
-diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h
-index ec2af67..eb9fe30 100644
---- a/include/linux/tracehook.h
-+++ b/include/linux/tracehook.h
-@@ -59,7 +59,7 @@ static inline void ptrace_report_syscall(struct pt_regs *regs)
- {
- int ptrace = current->ptrace;
-
-- if (!(ptrace & PT_PTRACED))
-+ if (!(ptrace & PT_SYSCALL_TRACE))
- return;
-
- ptrace_notify(SIGTRAP | ((ptrace & PT_TRACESYSGOOD) ? 0x80 : 0));
-diff --git a/kernel/ptrace.c b/kernel/ptrace.c
-index 43357e5..1ac03eb 100644
---- a/kernel/ptrace.c
-+++ b/kernel/ptrace.c
-@@ -40,10 +40,13 @@ void ptrace_signal_wake_up(struct task_struct *p, int quiescent)
-
- static void ptrace_set_syscall_trace(struct task_struct *p, bool on)
- {
-- if (on)
-+ if (on) {
-+ p->ptrace |= PT_SYSCALL_TRACE;
- set_tsk_thread_flag(p, TIF_SYSCALL_TRACE);
-- else
-+ } else {
-+ p->ptrace &= ~PT_SYSCALL_TRACE;
- clear_tsk_thread_flag(p, TIF_SYSCALL_TRACE);
-+ }
- }
-
- static int ptrace_trapping_sleep_fn(void *flags)
---
-1.5.5.1
-
-
-From oleg at redhat.com Mon Nov 21 15:07:11 2011
-Return-Path: oleg at redhat.com
-Received: from zmta01.collab.prod.int.phx2.redhat.com (LHLO
- zmta01.collab.prod.int.phx2.redhat.com) (10.5.5.31) by
- zmail13.collab.prod.int.phx2.redhat.com with LMTP; Mon, 21 Nov 2011
- 15:07:11 -0500 (EST)
-Received: from localhost (localhost.localdomain [127.0.0.1])
- by zmta01.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id B7D62D4DCC;
- Mon, 21 Nov 2011 15:07:11 -0500 (EST)
-Received: from zmta01.collab.prod.int.phx2.redhat.com ([127.0.0.1])
- by localhost (zmta01.collab.prod.int.phx2.redhat.com [127.0.0.1]) (amavisd-new, port 10024)
- with ESMTP id 1ENpGO166I-W; Mon, 21 Nov 2011 15:07:11 -0500 (EST)
-Received: from int-mx09.intmail.prod.int.phx2.redhat.com (int-mx09.intmail.prod.int.phx2.redhat.com [10.5.11.22])
- by zmta01.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id A3A55D498A;
- Mon, 21 Nov 2011 15:07:11 -0500 (EST)
-Received: from tranklukator.englab.brq.redhat.com (dhcp-1-232.brq.redhat.com [10.34.1.232])
- by int-mx09.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with SMTP id pALK79AS017728;
- Mon, 21 Nov 2011 15:07:10 -0500
-Received: by tranklukator.englab.brq.redhat.com (nbSMTP-1.00) for uid 500
- oleg at redhat.com; Mon, 21 Nov 2011 21:02:30 +0100 (CET)
-Date: Mon, 21 Nov 2011 21:02:28 +0100
-From: Oleg Nesterov <oleg at redhat.com>
-To: Dave Jones <davej at redhat.com>, "Frank Ch. Eigler" <fche at redhat.com>,
- Josh Boyer <jwboyer at redhat.com>, Josh Stone <jistone at redhat.com>,
- Kyle McMartin <kmcmartin at redhat.com>
-Cc: kernel at lists.fedoraproject.org
-Subject: [PATCH 23/33] utrace: don't clear TIF_SYSCALL_TRACE if it is
- needed by ptrace
-Message-ID: <20111121200228.GA27863 at redhat.com>
-MIME-Version: 1.0
-Content-Type: text/plain; charset=us-ascii
-Content-Disposition: inline
-In-Reply-To: <20111121200039.GA27699 at redhat.com>
-User-Agent: Mutt/1.5.18 (2008-05-17)
-X-Scanned-By: MIMEDefang 2.68 on 10.5.11.22
-Status: RO
-Content-Length: 797
-Lines: 24
-
-TIF_SYSCALL_TRACE should be cleared only if both ptrace and utrace do
-not want it, change utrace_reset() to check PT_SYSCALL_TRACE before
-clear_tsk_thread_flag(TIF_SYSCALL_TRACE).
-
-Signed-off-by: Oleg Nesterov <oleg at redhat.com>
----
- kernel/utrace.c | 1 +
- 1 files changed, 1 insertions(+), 0 deletions(-)
-
-diff --git a/kernel/utrace.c b/kernel/utrace.c
-index 0bb0a06..bebf6de 100644
---- a/kernel/utrace.c
-+++ b/kernel/utrace.c
-@@ -697,6 +697,7 @@ static bool utrace_reset(struct task_struct *task, struct utrace *utrace)
- BUG_ON(utrace->death);
- flags &= UTRACE_EVENT(REAP);
- } else if (!(flags & UTRACE_EVENT_SYSCALL) &&
-+ !(task->ptrace & PT_SYSCALL_TRACE) &&
- test_tsk_thread_flag(task, TIF_SYSCALL_TRACE)) {
- clear_tsk_thread_flag(task, TIF_SYSCALL_TRACE);
- }
---
-1.5.5.1
-
-
-From oleg at redhat.com Mon Nov 21 15:07:14 2011
-Return-Path: oleg at redhat.com
-Received: from zmta03.collab.prod.int.phx2.redhat.com (LHLO
- zmta03.collab.prod.int.phx2.redhat.com) (10.5.5.33) by
- zmail13.collab.prod.int.phx2.redhat.com with LMTP; Mon, 21 Nov 2011
- 15:07:14 -0500 (EST)
-Received: from localhost (localhost.localdomain [127.0.0.1])
- by zmta03.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 5376AD8774;
- Mon, 21 Nov 2011 15:07:14 -0500 (EST)
-Received: from zmta03.collab.prod.int.phx2.redhat.com ([127.0.0.1])
- by localhost (zmta03.collab.prod.int.phx2.redhat.com [127.0.0.1]) (amavisd-new, port 10024)
- with ESMTP id 5Oc8ErfFfJSs; Mon, 21 Nov 2011 15:07:14 -0500 (EST)
-Received: from int-mx10.intmail.prod.int.phx2.redhat.com (int-mx10.intmail.prod.int.phx2.redhat.com [10.5.11.23])
- by zmta03.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 3F2CFD81E9;
- Mon, 21 Nov 2011 15:07:14 -0500 (EST)
-Received: from tranklukator.englab.brq.redhat.com (dhcp-1-232.brq.redhat.com [10.34.1.232])
- by int-mx10.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with SMTP id pALK7Cw3015467;
- Mon, 21 Nov 2011 15:07:12 -0500
-Received: by tranklukator.englab.brq.redhat.com (nbSMTP-1.00) for uid 500
- oleg at redhat.com; Mon, 21 Nov 2011 21:02:32 +0100 (CET)
-Date: Mon, 21 Nov 2011 21:02:30 +0100
-From: Oleg Nesterov <oleg at redhat.com>
-To: Dave Jones <davej at redhat.com>, "Frank Ch. Eigler" <fche at redhat.com>,
- Josh Boyer <jwboyer at redhat.com>, Josh Stone <jistone at redhat.com>,
- Kyle McMartin <kmcmartin at redhat.com>
-Cc: kernel at lists.fedoraproject.org
-Subject: [PATCH 24/33] introduce task_utrace_lock/task_utrace_unlock
-Message-ID: <20111121200230.GA27870 at redhat.com>
-MIME-Version: 1.0
-Content-Type: text/plain; charset=us-ascii
-Content-Disposition: inline
-In-Reply-To: <20111121200039.GA27699 at redhat.com>
-User-Agent: Mutt/1.5.18 (2008-05-17)
-X-Scanned-By: MIMEDefang 2.68 on 10.5.11.23
-Status: RO
-Content-Length: 2190
-Lines: 81
-
-- Add task_utrace_lock(task). It simply takes task->utrace->lock if
- this task was ever utraced. Otherwise it takes task_lock(), this
- serializes with utrace_attach_task()->utrace_task_alloc(). In both
- case the caller can be sure it can't race with anything which needs
- utrace->lock.
-
-- Add task_utrace_unlock(task), it releases the corresponding lock.
-
-Signed-off-by: Oleg Nesterov <oleg at redhat.com>
----
- include/linux/utrace.h | 9 +++++++++
- kernel/utrace.c | 26 ++++++++++++++++++++++++++
- 2 files changed, 35 insertions(+), 0 deletions(-)
-
-diff --git a/include/linux/utrace.h b/include/linux/utrace.h
-index 63103e2..f37373b 100644
---- a/include/linux/utrace.h
-+++ b/include/linux/utrace.h
-@@ -117,6 +117,12 @@ void utrace_signal_handler(struct task_struct *, int);
-
- #ifndef CONFIG_UTRACE
-
-+static inline void task_utrace_lock(struct task_struct *task)
-+{
-+}
-+static inline void task_utrace_unlock(struct task_struct *task)
-+{
-+}
- /*
- * <linux/tracehook.h> uses these accessors to avoid #ifdef CONFIG_UTRACE.
- */
-@@ -139,6 +145,9 @@ static inline void task_utrace_proc_status(struct seq_file *m,
-
- #else /* CONFIG_UTRACE */
-
-+extern void task_utrace_lock(struct task_struct *task);
-+extern void task_utrace_unlock(struct task_struct *task);
-+
- static inline unsigned long task_utrace_flags(struct task_struct *task)
- {
- return task->utrace_flags;
-diff --git a/kernel/utrace.c b/kernel/utrace.c
-index bebf6de..960dd9e 100644
---- a/kernel/utrace.c
-+++ b/kernel/utrace.c
-@@ -79,6 +79,32 @@ static int __init utrace_init(void)
- }
- module_init(utrace_init);
-
-+void task_utrace_lock(struct task_struct *task)
-+{
-+ struct utrace *utrace = task_utrace_struct(task);
-+
-+ if (!utrace) {
-+ task_lock(task);
-+ utrace = task_utrace_struct(task);
-+ if (!utrace)
-+ return;
++ get_task_struct(target);
++ rcu_read_unlock();
+
-+ task_unlock(task);
++ if (likely(!ret)) {
++ unsigned long ncsw = wait_task_inactive(target, exam->state);
++ if (unlikely(ncsw != exam->ncsw))
++ ret = -EAGAIN;
++ put_task_struct(target);
+ }
+
-+ spin_lock(&utrace->lock);
++ return ret;
+}
++EXPORT_SYMBOL_GPL(utrace_finish_examine);
+
-+void task_utrace_unlock(struct task_struct *task)
-+{
-+ struct utrace *utrace = task_utrace_struct(task);
++/*
++ * This is declared in linux/regset.h and defined in machine-dependent
++ * code. We put the export here to ensure no machine forgets it.
++ */
++EXPORT_SYMBOL_GPL(task_user_regset_view);
+
-+ if (utrace)
-+ spin_unlock(&utrace->lock);
-+ else
-+ task_unlock(task);
++/*
++ * Called with rcu_read_lock() held.
++ */
++void task_utrace_proc_status(struct seq_file *m, struct task_struct *p)
++{
++ seq_printf(m, "Utrace:\t%lx\n", p->utrace_flags);
+}
-+
- /*
- * Set up @task.utrace for the first time. We can have races
- * between two utrace_attach_task() calls here. The task_lock()
---
-1.5.5.1
-
-
-From oleg at redhat.com Mon Nov 21 15:07:17 2011
-Return-Path: oleg at redhat.com
-Received: from zmta01.collab.prod.int.phx2.redhat.com (LHLO
- zmta01.collab.prod.int.phx2.redhat.com) (10.5.5.31) by
- zmail13.collab.prod.int.phx2.redhat.com with LMTP; Mon, 21 Nov 2011
- 15:07:17 -0500 (EST)
-Received: from localhost (localhost.localdomain [127.0.0.1])
- by zmta01.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 01034D498A;
- Mon, 21 Nov 2011 15:07:17 -0500 (EST)
-Received: from zmta01.collab.prod.int.phx2.redhat.com ([127.0.0.1])
- by localhost (zmta01.collab.prod.int.phx2.redhat.com [127.0.0.1]) (amavisd-new, port 10024)
- with ESMTP id BglmrU3H2hJr; Mon, 21 Nov 2011 15:07:16 -0500 (EST)
-Received: from int-mx10.intmail.prod.int.phx2.redhat.com (int-mx10.intmail.prod.int.phx2.redhat.com [10.5.11.23])
- by zmta01.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id E0D31D4FF4;
- Mon, 21 Nov 2011 15:07:16 -0500 (EST)
-Received: from tranklukator.englab.brq.redhat.com (dhcp-1-232.brq.redhat.com [10.34.1.232])
- by int-mx10.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with SMTP id pALK7Egx015474;
- Mon, 21 Nov 2011 15:07:15 -0500
-Received: by tranklukator.englab.brq.redhat.com (nbSMTP-1.00) for uid 500
- oleg at redhat.com; Mon, 21 Nov 2011 21:02:35 +0100 (CET)
-Date: Mon, 21 Nov 2011 21:02:33 +0100
-From: Oleg Nesterov <oleg at redhat.com>
-To: Dave Jones <davej at redhat.com>, "Frank Ch. Eigler" <fche at redhat.com>,
- Josh Boyer <jwboyer at redhat.com>, Josh Stone <jistone at redhat.com>,
- Kyle McMartin <kmcmartin at redhat.com>
-Cc: kernel at lists.fedoraproject.org
-Subject: [PATCH 25/33] teach ptrace_set_syscall_trace() to play well with
- utrace
-Message-ID: <20111121200233.GA27873 at redhat.com>
-MIME-Version: 1.0
-Content-Type: text/plain; charset=us-ascii
-Content-Disposition: inline
-In-Reply-To: <20111121200039.GA27699 at redhat.com>
-User-Agent: Mutt/1.5.18 (2008-05-17)
-X-Scanned-By: MIMEDefang 2.68 on 10.5.11.23
-Status: RO
-Content-Length: 1545
-Lines: 49
-
-1. ptrace_set_syscall_trace(true)->set_tsk_thread_flag(TIF_SYSCALL_TRACE)
- can race with utrace_control()->utrace_reset() path which can miss
- PT_SYSCALL_TRACE and clear TIF_SYSCALL_TRACE after it was already set.
-
-2. ptrace_set_syscall_trace(false) clears TIF_SYSCALL_TRACE and this is
- not utrace-friendly, it can need this flag.
-
-Change ptrace_set_syscall_trace() to take task_utrace_lock(), this is
-enough to fix the 1st problem. Check task_utrace_flags() before clearing
-TIF_SYSCALL_TRACE, this fixes 2.
-
-Signed-off-by: Oleg Nesterov <oleg at redhat.com>
----
- kernel/ptrace.c | 6 +++++-
- 1 files changed, 5 insertions(+), 1 deletions(-)
-
-diff --git a/kernel/ptrace.c b/kernel/ptrace.c
-index 1ac03eb..739183a 100644
---- a/kernel/ptrace.c
-+++ b/kernel/ptrace.c
-@@ -24,6 +24,7 @@
- #include <linux/regset.h>
- #include <linux/hw_breakpoint.h>
- #include <linux/cn_proc.h>
-+#include <linux/utrace.h>
-
- void ptrace_signal_wake_up(struct task_struct *p, int quiescent)
- {
-@@ -40,13 +41,16 @@ void ptrace_signal_wake_up(struct task_struct *p, int quiescent)
-
- static void ptrace_set_syscall_trace(struct task_struct *p, bool on)
- {
-+ task_utrace_lock(p);
- if (on) {
- p->ptrace |= PT_SYSCALL_TRACE;
- set_tsk_thread_flag(p, TIF_SYSCALL_TRACE);
- } else {
- p->ptrace &= ~PT_SYSCALL_TRACE;
-- clear_tsk_thread_flag(p, TIF_SYSCALL_TRACE);
-+ if (!(task_utrace_flags(p) & UTRACE_EVENT_SYSCALL))
-+ clear_tsk_thread_flag(p, TIF_SYSCALL_TRACE);
- }
-+ task_utrace_unlock(p);
- }
-
- static int ptrace_trapping_sleep_fn(void *flags)
---
-1.5.5.1
-
-
-From oleg at redhat.com Mon Nov 21 15:07:19 2011
-Return-Path: oleg at redhat.com
-Received: from zmta02.collab.prod.int.phx2.redhat.com (LHLO
- zmta02.collab.prod.int.phx2.redhat.com) (10.5.5.32) by
- zmail13.collab.prod.int.phx2.redhat.com with LMTP; Mon, 21 Nov 2011
- 15:07:19 -0500 (EST)
-Received: from localhost (localhost.localdomain [127.0.0.1])
- by zmta02.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 8F87F128EAA;
- Mon, 21 Nov 2011 15:07:19 -0500 (EST)
-Received: from zmta02.collab.prod.int.phx2.redhat.com ([127.0.0.1])
- by localhost (zmta02.collab.prod.int.phx2.redhat.com [127.0.0.1]) (amavisd-new, port 10024)
- with ESMTP id yittEjCd+X4F; Mon, 21 Nov 2011 15:07:19 -0500 (EST)
-Received: from int-mx01.intmail.prod.int.phx2.redhat.com (int-mx01.intmail.prod.int.phx2.redhat.com [10.5.11.11])
- by zmta02.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 7BC701287D0;
- Mon, 21 Nov 2011 15:07:19 -0500 (EST)
-Received: from tranklukator.englab.brq.redhat.com (dhcp-1-232.brq.redhat.com [10.34.1.232])
- by int-mx01.intmail.prod.int.phx2.redhat.com (8.13.8/8.13.8) with SMTP id pALK7HbH031553;
- Mon, 21 Nov 2011 15:07:18 -0500
-Received: by tranklukator.englab.brq.redhat.com (nbSMTP-1.00) for uid 500
- oleg at redhat.com; Mon, 21 Nov 2011 21:02:38 +0100 (CET)
-Date: Mon, 21 Nov 2011 21:02:36 +0100
-From: Oleg Nesterov <oleg at redhat.com>
-To: Dave Jones <davej at redhat.com>, "Frank Ch. Eigler" <fche at redhat.com>,
- Josh Boyer <jwboyer at redhat.com>, Josh Stone <jistone at redhat.com>,
- Kyle McMartin <kmcmartin at redhat.com>
-Cc: kernel at lists.fedoraproject.org
-Subject: [PATCH 26/33] introduce PT_SINGLE_STEP and PT_SINGLE_BLOCK
-Message-ID: <20111121200236.GA27880 at redhat.com>
-MIME-Version: 1.0
-Content-Type: text/plain; charset=us-ascii
-Content-Disposition: inline
-In-Reply-To: <20111121200039.GA27699 at redhat.com>
-User-Agent: Mutt/1.5.18 (2008-05-17)
-X-Scanned-By: MIMEDefang 2.67 on 10.5.11.11
-Status: RO
-Content-Length: 3723
-Lines: 108
-
-Add the new internal ptrace flags, PT_SINGLE_STEP and PT_SINGLE_BLOCK.
-
-Like PT_SYSCALL_TRACE, this is needed to avoid the unnecessary ptrace
-reports when TIF_SINGLESTEP was set by another engine, not by ptrace.
-Also, we need these bits to coordinate the user_*_single_step() calls
-from ptrace and utrace.
-
-TODO: update the !x86 ptrace code which does user_disable_single_step().
-
-Signed-off-by: Oleg Nesterov <oleg at redhat.com>
----
- arch/x86/kernel/ptrace.c | 1 +
- include/linux/ptrace.h | 5 ++++-
- include/linux/tracehook.h | 7 +++++--
- kernel/ptrace.c | 3 +++
- 4 files changed, 13 insertions(+), 3 deletions(-)
-
-diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c
-index 8252879..d1557dc 100644
---- a/arch/x86/kernel/ptrace.c
-+++ b/arch/x86/kernel/ptrace.c
-@@ -808,6 +808,7 @@ static int ioperm_get(struct task_struct *target,
- */
- void ptrace_disable(struct task_struct *child)
- {
-+ child->ptrace &= ~(PT_SINGLE_STEP | PT_SINGLE_BLOCK);
- user_disable_single_step(child);
- #ifdef TIF_SYSCALL_EMU
- clear_tsk_thread_flag(child, TIF_SYSCALL_EMU);
-diff --git a/include/linux/ptrace.h b/include/linux/ptrace.h
-index c10f610..2743315 100644
---- a/include/linux/ptrace.h
-+++ b/include/linux/ptrace.h
-@@ -105,6 +105,8 @@
- #define PT_TRACE_MASK 0x000003f4
-
- #define PT_SYSCALL_TRACE 0x00020000
-+#define PT_SINGLE_STEP 0x00040000
-+#define PT_SINGLE_BLOCK 0x00080000
-
- /* single stepping state bits (used on ARM and PA-RISC) */
- #define PT_SINGLESTEP_BIT 31
-@@ -229,7 +231,8 @@ static inline void ptrace_init_task(struct task_struct *child, bool ptrace)
-
- if (unlikely(ptrace) && current->ptrace) {
- child->ptrace = current->ptrace;
-- child->ptrace &= ~PT_SYSCALL_TRACE;
-+ child->ptrace &=
-+ ~(PT_SYSCALL_TRACE | PT_SINGLE_STEP | PT_SINGLE_BLOCK);
- __ptrace_link(child, current->parent);
-
- if (child->ptrace & PT_SEIZED)
-diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h
-index eb9fe30..21c8ca2 100644
---- a/include/linux/tracehook.h
-+++ b/include/linux/tracehook.h
-@@ -104,6 +104,9 @@ static inline __must_check int tracehook_report_syscall_entry(
- return 0;
- }
-
-+#define ptrace_wants_step() \
-+ (current->ptrace & (PT_SINGLE_STEP | PT_SINGLE_BLOCK))
-+
- /**
- * tracehook_report_syscall_exit - task has just finished a system call
- * @regs: user register state of current task
-@@ -126,7 +129,7 @@ static inline void tracehook_report_syscall_exit(struct pt_regs *regs, int step)
- if (task_utrace_flags(current) & UTRACE_EVENT(SYSCALL_EXIT))
- utrace_report_syscall_exit(regs);
-
-- if (step) {
-+ if (step && ptrace_wants_step()) {
- siginfo_t info;
- user_single_step_siginfo(current, regs, &info);
- force_sig_info(SIGTRAP, &info, current);
-@@ -157,7 +160,7 @@ static inline void tracehook_signal_handler(int sig, siginfo_t *info,
- {
- if (task_utrace_flags(current))
- utrace_signal_handler(current, stepping);
-- if (stepping)
-+ if (stepping && ptrace_wants_step())
- ptrace_notify(SIGTRAP);
- }
-
-diff --git a/kernel/ptrace.c b/kernel/ptrace.c
-index 739183a..792080d 100644
---- a/kernel/ptrace.c
-+++ b/kernel/ptrace.c
-@@ -630,13 +630,16 @@ static int ptrace_resume(struct task_struct *child, long request,
- clear_tsk_thread_flag(child, TIF_SYSCALL_EMU);
- #endif
-
-+ child->ptrace &= ~(PT_SINGLE_STEP | PT_SINGLE_BLOCK);
- if (is_singleblock(request)) {
- if (unlikely(!arch_has_block_step()))
- return -EIO;
-+ child->ptrace |= PT_SINGLE_BLOCK;
- user_enable_block_step(child);
- } else if (is_singlestep(request) || is_sysemu_singlestep(request)) {
- if (unlikely(!arch_has_single_step()))
- return -EIO;
-+ child->ptrace |= PT_SINGLE_STEP;
- user_enable_single_step(child);
- } else {
- user_disable_single_step(child);
---
-1.5.5.1
-
-
-From oleg at redhat.com Mon Nov 21 15:07:22 2011
-Return-Path: oleg at redhat.com
-Received: from zmta02.collab.prod.int.phx2.redhat.com (LHLO
- zmta02.collab.prod.int.phx2.redhat.com) (10.5.5.32) by
- zmail13.collab.prod.int.phx2.redhat.com with LMTP; Mon, 21 Nov 2011
- 15:07:22 -0500 (EST)
-Received: from localhost (localhost.localdomain [127.0.0.1])
- by zmta02.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 0C136128FB0;
- Mon, 21 Nov 2011 15:07:22 -0500 (EST)
-Received: from zmta02.collab.prod.int.phx2.redhat.com ([127.0.0.1])
- by localhost (zmta02.collab.prod.int.phx2.redhat.com [127.0.0.1]) (amavisd-new, port 10024)
- with ESMTP id GNbS20gY5Jq9; Mon, 21 Nov 2011 15:07:21 -0500 (EST)
-Received: from int-mx02.intmail.prod.int.phx2.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12])
- by zmta02.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id EB5821287D0;
- Mon, 21 Nov 2011 15:07:21 -0500 (EST)
-Received: from tranklukator.englab.brq.redhat.com (dhcp-1-232.brq.redhat.com [10.34.1.232])
- by int-mx02.intmail.prod.int.phx2.redhat.com (8.13.8/8.13.8) with SMTP id pALK7Kn1029766;
- Mon, 21 Nov 2011 15:07:20 -0500
-Received: by tranklukator.englab.brq.redhat.com (nbSMTP-1.00) for uid 500
- oleg at redhat.com; Mon, 21 Nov 2011 21:02:40 +0100 (CET)
-Date: Mon, 21 Nov 2011 21:02:38 +0100
-From: Oleg Nesterov <oleg at redhat.com>
-To: Dave Jones <davej at redhat.com>, "Frank Ch. Eigler" <fche at redhat.com>,
- Josh Boyer <jwboyer at redhat.com>, Josh Stone <jistone at redhat.com>,
- Kyle McMartin <kmcmartin at redhat.com>
-Cc: kernel at lists.fedoraproject.org
-Subject: [PATCH 27/33] utrace: finish_resume_report: don't do
- user_xxx_step() if ptrace_wants_step()
-Message-ID: <20111121200238.GA27883 at redhat.com>
-MIME-Version: 1.0
-Content-Type: text/plain; charset=us-ascii
-Content-Disposition: inline
-In-Reply-To: <20111121200039.GA27699 at redhat.com>
-User-Agent: Mutt/1.5.18 (2008-05-17)
-X-Scanned-By: MIMEDefang 2.67 on 10.5.11.12
-Status: RO
-Content-Length: 2741
-Lines: 84
-
-finish_resume_report() should not enable/disable the stepping if
-ptrace_wants_step() == T. If ptrace wants block_step while utrace
-wants single_step we could "promote" the stepping, but I do not
-think this really makes sense.
-
-Unless the tracee is killed this can't race with ptrace, this is
-called by the tracee itself. If it is killed we do not care.
-
-Signed-off-by: Oleg Nesterov <oleg at redhat.com>
----
- include/linux/tracehook.h | 8 ++++----
- kernel/utrace.c | 9 ++++++---
- 2 files changed, 10 insertions(+), 7 deletions(-)
-
-diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h
-index 21c8ca2..b6812d4 100644
---- a/include/linux/tracehook.h
-+++ b/include/linux/tracehook.h
-@@ -104,8 +104,8 @@ static inline __must_check int tracehook_report_syscall_entry(
- return 0;
- }
-
--#define ptrace_wants_step() \
-- (current->ptrace & (PT_SINGLE_STEP | PT_SINGLE_BLOCK))
-+#define ptrace_wants_step(task) \
-+ ((task)->ptrace & (PT_SINGLE_STEP | PT_SINGLE_BLOCK))
-
- /**
- * tracehook_report_syscall_exit - task has just finished a system call
-@@ -129,7 +129,7 @@ static inline void tracehook_report_syscall_exit(struct pt_regs *regs, int step)
- if (task_utrace_flags(current) & UTRACE_EVENT(SYSCALL_EXIT))
- utrace_report_syscall_exit(regs);
-
-- if (step && ptrace_wants_step()) {
-+ if (step && ptrace_wants_step(current)) {
- siginfo_t info;
- user_single_step_siginfo(current, regs, &info);
- force_sig_info(SIGTRAP, &info, current);
-@@ -160,7 +160,7 @@ static inline void tracehook_signal_handler(int sig, siginfo_t *info,
- {
- if (task_utrace_flags(current))
- utrace_signal_handler(current, stepping);
-- if (stepping && ptrace_wants_step())
-+ if (stepping && ptrace_wants_step(current))
- ptrace_notify(SIGTRAP);
- }
-
-diff --git a/kernel/utrace.c b/kernel/utrace.c
-index 960dd9e..05e8532 100644
---- a/kernel/utrace.c
-+++ b/kernel/utrace.c
-@@ -1829,7 +1829,8 @@ static void finish_resume_report(struct task_struct *task,
-
- case UTRACE_BLOCKSTEP:
- if (likely(arch_has_block_step())) {
-- user_enable_block_step(task);
-+ if (!ptrace_wants_step(task))
-+ user_enable_block_step(task);
- break;
- }
-
-@@ -1842,7 +1843,8 @@ static void finish_resume_report(struct task_struct *task,
-
- case UTRACE_SINGLESTEP:
- if (likely(arch_has_single_step())) {
-- user_enable_single_step(task);
-+ if (!ptrace_wants_step(task))
-+ user_enable_single_step(task);
- } else {
- /*
- * This means some callback is to blame for failing
-@@ -1857,7 +1859,8 @@ static void finish_resume_report(struct task_struct *task,
- case UTRACE_REPORT:
- case UTRACE_RESUME:
- default:
-- user_disable_single_step(task);
-+ if (!ptrace_wants_step(task))
-+ user_disable_single_step(task);
- break;
- }
- }
---
-1.5.5.1
-
-
-From oleg at redhat.com Mon Nov 21 15:07:24 2011
-Return-Path: oleg at redhat.com
-Received: from zmta02.collab.prod.int.phx2.redhat.com (LHLO
- zmta02.collab.prod.int.phx2.redhat.com) (10.5.5.32) by
- zmail13.collab.prod.int.phx2.redhat.com with LMTP; Mon, 21 Nov 2011
- 15:07:24 -0500 (EST)
-Received: from localhost (localhost.localdomain [127.0.0.1])
- by zmta02.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 8954F12914D;
- Mon, 21 Nov 2011 15:07:24 -0500 (EST)
-Received: from zmta02.collab.prod.int.phx2.redhat.com ([127.0.0.1])
- by localhost (zmta02.collab.prod.int.phx2.redhat.com [127.0.0.1]) (amavisd-new, port 10024)
- with ESMTP id 0lU8wXKC78yu; Mon, 21 Nov 2011 15:07:24 -0500 (EST)
-Received: from int-mx12.intmail.prod.int.phx2.redhat.com (int-mx12.intmail.prod.int.phx2.redhat.com [10.5.11.25])
- by zmta02.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 759B61287D0;
- Mon, 21 Nov 2011 15:07:24 -0500 (EST)
-Received: from tranklukator.englab.brq.redhat.com (dhcp-1-232.brq.redhat.com [10.34.1.232])
- by int-mx12.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with SMTP id pALK7Mi7022329;
- Mon, 21 Nov 2011 15:07:23 -0500
-Received: by tranklukator.englab.brq.redhat.com (nbSMTP-1.00) for uid 500
- oleg at redhat.com; Mon, 21 Nov 2011 21:02:43 +0100 (CET)
-Date: Mon, 21 Nov 2011 21:02:41 +0100
-From: Oleg Nesterov <oleg at redhat.com>
-To: Dave Jones <davej at redhat.com>, "Frank Ch. Eigler" <fche at redhat.com>,
- Josh Boyer <jwboyer at redhat.com>, Josh Stone <jistone at redhat.com>,
- Kyle McMartin <kmcmartin at redhat.com>
-Cc: kernel at lists.fedoraproject.org
-Subject: [PATCH 28/33] ptrace: shift user_*_step() from ptrace_resume() to
- ptrace_stop()
-Message-ID: <20111121200241.GA27890 at redhat.com>
-MIME-Version: 1.0
-Content-Type: text/plain; charset=us-ascii
-Content-Disposition: inline
-In-Reply-To: <20111121200039.GA27699 at redhat.com>
-User-Agent: Mutt/1.5.18 (2008-05-17)
-X-Scanned-By: MIMEDefang 2.68 on 10.5.11.25
-Status: RO
-Content-Length: 2374
-Lines: 70
-
-1. ptrace_resume() plays with the stopped task which can be also
- task_is_utraced(). In this case user_enable_xxx_step() can race
- with utrace_reset()->user_disable_single_step().
-
- We could change utrace_reset() to check ptrace_wants_step() and
- add the task_utrace_lock + unlock barrier after ptrace_resume()
- sets PT_SINGLE_STEP.
-
- But it is better to reassign enable/desable from the tracer to
- the tracee, it can check its PT_SINGLE_ bits after wakeup. This
- also makes sense because enable_step(task) can be faster if
- task == current.
-
-2. ptrace can do user_disable_single_step() while utrace needs the
- stepping.
-
- Set TIF_NOTIFY_RESUME after user_disable_single_step() to ensure
- the tracee can't return to the user-space without utrace_resume().
- Any correct engine which wants the stepping should reassert it.
-
-Signed-off-by: Oleg Nesterov <oleg at redhat.com>
----
- kernel/ptrace.c | 4 ----
- kernel/signal.c | 11 +++++++++++
- 2 files changed, 11 insertions(+), 4 deletions(-)
-
-diff --git a/kernel/ptrace.c b/kernel/ptrace.c
-index 792080d..eba9a22 100644
---- a/kernel/ptrace.c
-+++ b/kernel/ptrace.c
-@@ -635,14 +635,10 @@ static int ptrace_resume(struct task_struct *child, long request,
- if (unlikely(!arch_has_block_step()))
- return -EIO;
- child->ptrace |= PT_SINGLE_BLOCK;
-- user_enable_block_step(child);
- } else if (is_singlestep(request) || is_sysemu_singlestep(request)) {
- if (unlikely(!arch_has_single_step()))
- return -EIO;
- child->ptrace |= PT_SINGLE_STEP;
-- user_enable_single_step(child);
-- } else {
-- user_disable_single_step(child);
- }
-
- child->exit_code = data;
-diff --git a/kernel/signal.c b/kernel/signal.c
-index 38ea4e6..b13d2bc 100644
---- a/kernel/signal.c
-+++ b/kernel/signal.c
-@@ -1928,6 +1928,17 @@ static void ptrace_stop(int exit_code, int why, int clear_code, siginfo_t *info)
-
- utrace_end_stop();
-
-+ if (current->ptrace & PT_SINGLE_BLOCK)
-+ user_enable_block_step(current);
-+ else if (current->ptrace & PT_SINGLE_STEP)
-+ user_enable_single_step(current);
-+ else {
-+ user_disable_single_step(current);
-+ /* if utrace needs the stepping it should reassert */
-+ if (task_utrace_flags(current))
-+ set_thread_flag(TIF_NOTIFY_RESUME);
-+ }
-+
- /*
- * While in TASK_TRACED, we were considered "frozen enough".
- * Now that we woke up, it's crucial if we're supposed to be
---
-1.5.5.1
-
-
-From oleg at redhat.com Mon Nov 21 15:07:27 2011
-Return-Path: oleg at redhat.com
-Received: from zmta02.collab.prod.int.phx2.redhat.com (LHLO
- zmta02.collab.prod.int.phx2.redhat.com) (10.5.5.32) by
- zmail13.collab.prod.int.phx2.redhat.com with LMTP; Mon, 21 Nov 2011
- 15:07:27 -0500 (EST)
-Received: from localhost (localhost.localdomain [127.0.0.1])
- by zmta02.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 043581287D0;
- Mon, 21 Nov 2011 15:07:27 -0500 (EST)
-Received: from zmta02.collab.prod.int.phx2.redhat.com ([127.0.0.1])
- by localhost (zmta02.collab.prod.int.phx2.redhat.com [127.0.0.1]) (amavisd-new, port 10024)
- with ESMTP id 4WO1El1jiIrJ; Mon, 21 Nov 2011 15:07:26 -0500 (EST)
-Received: from int-mx09.intmail.prod.int.phx2.redhat.com (int-mx09.intmail.prod.int.phx2.redhat.com [10.5.11.22])
- by zmta02.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id E3FCA129045;
- Mon, 21 Nov 2011 15:07:26 -0500 (EST)
-Received: from tranklukator.englab.brq.redhat.com (dhcp-1-232.brq.redhat.com [10.34.1.232])
- by int-mx09.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with SMTP id pALK7OdV017836;
- Mon, 21 Nov 2011 15:07:25 -0500
-Received: by tranklukator.englab.brq.redhat.com (nbSMTP-1.00) for uid 500
- oleg at redhat.com; Mon, 21 Nov 2011 21:02:45 +0100 (CET)
-Date: Mon, 21 Nov 2011 21:02:43 +0100
-From: Oleg Nesterov <oleg at redhat.com>
-To: Dave Jones <davej at redhat.com>, "Frank Ch. Eigler" <fche at redhat.com>,
- Josh Boyer <jwboyer at redhat.com>, Josh Stone <jistone at redhat.com>,
- Kyle McMartin <kmcmartin at redhat.com>
-Cc: kernel at lists.fedoraproject.org
-Subject: [PATCH 29/33] ptrace_disable: no need to disable stepping
-Message-ID: <20111121200243.GA27893 at redhat.com>
-MIME-Version: 1.0
-Content-Type: text/plain; charset=us-ascii
-Content-Disposition: inline
-In-Reply-To: <20111121200039.GA27699 at redhat.com>
-User-Agent: Mutt/1.5.18 (2008-05-17)
-X-Scanned-By: MIMEDefang 2.68 on 10.5.11.22
-Status: RO
-Content-Length: 945
-Lines: 29
-
-ptrace_disable() is called when the tracee is quiescent and we
-are going to untrace. This means we are going to clear ->ptrace
-and wake up the tracee. Now that ptrace_stop() checks PT_ bits
-and does user_disable_single_step() we can remove this code from
-ptrace_disable(), it is unneeded and not utrace-friendly.
-
-TODO: change !x86 code.
-
-Signed-off-by: Oleg Nesterov <oleg at redhat.com>
----
- arch/x86/kernel/ptrace.c | 2 --
- 1 files changed, 0 insertions(+), 2 deletions(-)
-
-diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c
-index d1557dc..96d315a 100644
---- a/arch/x86/kernel/ptrace.c
-+++ b/arch/x86/kernel/ptrace.c
-@@ -808,8 +808,6 @@ static int ioperm_get(struct task_struct *target,
- */
- void ptrace_disable(struct task_struct *child)
- {
-- child->ptrace &= ~(PT_SINGLE_STEP | PT_SINGLE_BLOCK);
-- user_disable_single_step(child);
- #ifdef TIF_SYSCALL_EMU
- clear_tsk_thread_flag(child, TIF_SYSCALL_EMU);
- #endif
---
-1.5.5.1
-
-
-From oleg at redhat.com Mon Nov 21 15:07:29 2011
-Return-Path: oleg at redhat.com
-Received: from zmta01.collab.prod.int.phx2.redhat.com (LHLO
- zmta01.collab.prod.int.phx2.redhat.com) (10.5.5.31) by
- zmail13.collab.prod.int.phx2.redhat.com with LMTP; Mon, 21 Nov 2011
- 15:07:29 -0500 (EST)
-Received: from localhost (localhost.localdomain [127.0.0.1])
- by zmta01.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 8A6E8D4C39;
- Mon, 21 Nov 2011 15:07:29 -0500 (EST)
-Received: from zmta01.collab.prod.int.phx2.redhat.com ([127.0.0.1])
- by localhost (zmta01.collab.prod.int.phx2.redhat.com [127.0.0.1]) (amavisd-new, port 10024)
- with ESMTP id NLURBx2yQes5; Mon, 21 Nov 2011 15:07:29 -0500 (EST)
-Received: from int-mx10.intmail.prod.int.phx2.redhat.com (int-mx10.intmail.prod.int.phx2.redhat.com [10.5.11.23])
- by zmta01.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 7660DD498A;
- Mon, 21 Nov 2011 15:07:29 -0500 (EST)
-Received: from tranklukator.englab.brq.redhat.com (dhcp-1-232.brq.redhat.com [10.34.1.232])
- by int-mx10.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with SMTP id pALK7RnP015578;
- Mon, 21 Nov 2011 15:07:27 -0500
-Received: by tranklukator.englab.brq.redhat.com (nbSMTP-1.00) for uid 500
- oleg at redhat.com; Mon, 21 Nov 2011 21:02:48 +0100 (CET)
-Date: Mon, 21 Nov 2011 21:02:46 +0100
-From: Oleg Nesterov <oleg at redhat.com>
-To: Dave Jones <davej at redhat.com>, "Frank Ch. Eigler" <fche at redhat.com>,
- Josh Boyer <jwboyer at redhat.com>, Josh Stone <jistone at redhat.com>,
- Kyle McMartin <kmcmartin at redhat.com>
-Cc: kernel at lists.fedoraproject.org
-Subject: [PATCH 30/33] ptrace_report_syscall: check TIF_SYSCALL_EMU
-Message-ID: <20111121200246.GA27896 at redhat.com>
-MIME-Version: 1.0
-Content-Type: text/plain; charset=us-ascii
-Content-Disposition: inline
-In-Reply-To: <20111121200039.GA27699 at redhat.com>
-User-Agent: Mutt/1.5.18 (2008-05-17)
-X-Scanned-By: MIMEDefang 2.68 on 10.5.11.23
-Status: RO
-Content-Length: 858
-Lines: 28
-
-4d16a64 "introduce PT_SYSCALL_TRACE flag" breaks PTRACE_SYSEMU
-which doesn't set PT_SYSCALL_TRACE.
-
-Change ptrace_report_syscall() to check TIF_SYSCALL_EMU as well.
-This can't conflict with utrace, this flag can only be set by
-ptrace.
-
-Signed-off-by: Oleg Nesterov <oleg at redhat.com>
----
- include/linux/tracehook.h | 2 +-
- 1 files changed, 1 insertions(+), 1 deletions(-)
-
-diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h
-index b6812d4..90ca578 100644
---- a/include/linux/tracehook.h
-+++ b/include/linux/tracehook.h
-@@ -59,7 +59,7 @@ static inline void ptrace_report_syscall(struct pt_regs *regs)
- {
- int ptrace = current->ptrace;
-
-- if (!(ptrace & PT_SYSCALL_TRACE))
-+ if (!(ptrace & PT_SYSCALL_TRACE) && !test_thread_flag(TIF_SYSCALL_EMU))
- return;
-
- ptrace_notify(SIGTRAP | ((ptrace & PT_TRACESYSGOOD) ? 0x80 : 0));
---
-1.5.5.1
-
-
-From oleg at redhat.com Mon Nov 21 15:07:32 2011
-Return-Path: oleg at redhat.com
-Received: from zmta02.collab.prod.int.phx2.redhat.com (LHLO
- zmta02.collab.prod.int.phx2.redhat.com) (10.5.5.32) by
- zmail13.collab.prod.int.phx2.redhat.com with LMTP; Mon, 21 Nov 2011
- 15:07:32 -0500 (EST)
-Received: from localhost (localhost.localdomain [127.0.0.1])
- by zmta02.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 10D1312916C;
- Mon, 21 Nov 2011 15:07:32 -0500 (EST)
-Received: from zmta02.collab.prod.int.phx2.redhat.com ([127.0.0.1])
- by localhost (zmta02.collab.prod.int.phx2.redhat.com [127.0.0.1]) (amavisd-new, port 10024)
- with ESMTP id Z14plX1oxaYL; Mon, 21 Nov 2011 15:07:32 -0500 (EST)
-Received: from int-mx12.intmail.prod.int.phx2.redhat.com (int-mx12.intmail.prod.int.phx2.redhat.com [10.5.11.25])
- by zmta02.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id F12FF128E1D;
- Mon, 21 Nov 2011 15:07:31 -0500 (EST)
-Received: from tranklukator.englab.brq.redhat.com (dhcp-1-232.brq.redhat.com [10.34.1.232])
- by int-mx12.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with SMTP id pALK7T0q022460;
- Mon, 21 Nov 2011 15:07:30 -0500
-Received: by tranklukator.englab.brq.redhat.com (nbSMTP-1.00) for uid 500
- oleg at redhat.com; Mon, 21 Nov 2011 21:02:50 +0100 (CET)
-Date: Mon, 21 Nov 2011 21:02:48 +0100
-From: Oleg Nesterov <oleg at redhat.com>
-To: Dave Jones <davej at redhat.com>, "Frank Ch. Eigler" <fche at redhat.com>,
- Josh Boyer <jwboyer at redhat.com>, Josh Stone <jistone at redhat.com>,
- Kyle McMartin <kmcmartin at redhat.com>
-Cc: kernel at lists.fedoraproject.org
-Subject: [PATCH 31/33] utrace_resume: check irqs_disabled() to shut up
- lockdep
-Message-ID: <20111121200248.GA27903 at redhat.com>
-MIME-Version: 1.0
-Content-Type: text/plain; charset=us-ascii
-Content-Disposition: inline
-In-Reply-To: <20111121200039.GA27699 at redhat.com>
-User-Agent: Mutt/1.5.18 (2008-05-17)
-X-Scanned-By: MIMEDefang 2.68 on 10.5.11.25
-Status: RO
-Content-Length: 823
-Lines: 26
-
-utrace_resume() enables irqs unconditionally. With the recent changes
-in lockdep.c this triggers the warning. Check irqs_disabled() before
-local_irq_enable().
-
-Signed-off-by: Oleg Nesterov <oleg at redhat.com>
----
- kernel/utrace.c | 3 ++-
- 1 files changed, 2 insertions(+), 1 deletions(-)
-
-diff --git a/kernel/utrace.c b/kernel/utrace.c
-index 05e8532..c817a46 100644
---- a/kernel/utrace.c
-+++ b/kernel/utrace.c
-@@ -1881,7 +1881,8 @@ void utrace_resume(struct task_struct *task, struct pt_regs *regs)
- * code path leads to calling into get_signal_to_deliver(), which
- * implicitly reenables them by virtue of spin_unlock_irq.
- */
-- local_irq_enable();
-+ if (irqs_disabled()) /* make trace_hardirqs_on() happy */
-+ local_irq_enable();
-
- /*
- * If this flag is still set it's because there was a signal
---
-1.5.5.1
-
-
-From oleg at redhat.com Mon Nov 21 15:07:34 2011
-Return-Path: oleg at redhat.com
-Received: from zmta03.collab.prod.int.phx2.redhat.com (LHLO
- zmta03.collab.prod.int.phx2.redhat.com) (10.5.5.33) by
- zmail13.collab.prod.int.phx2.redhat.com with LMTP; Mon, 21 Nov 2011
- 15:07:34 -0500 (EST)
-Received: from localhost (localhost.localdomain [127.0.0.1])
- by zmta03.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 860BCD8A3E;
- Mon, 21 Nov 2011 15:07:34 -0500 (EST)
-Received: from zmta03.collab.prod.int.phx2.redhat.com ([127.0.0.1])
- by localhost (zmta03.collab.prod.int.phx2.redhat.com [127.0.0.1]) (amavisd-new, port 10024)
- with ESMTP id pVTbN1oF62dg; Mon, 21 Nov 2011 15:07:34 -0500 (EST)
-Received: from int-mx02.intmail.prod.int.phx2.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12])
- by zmta03.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 6FBFFD87A3;
- Mon, 21 Nov 2011 15:07:34 -0500 (EST)
-Received: from tranklukator.englab.brq.redhat.com (dhcp-1-232.brq.redhat.com [10.34.1.232])
- by int-mx02.intmail.prod.int.phx2.redhat.com (8.13.8/8.13.8) with SMTP id pALK7W0B029871;
- Mon, 21 Nov 2011 15:07:32 -0500
-Received: by tranklukator.englab.brq.redhat.com (nbSMTP-1.00) for uid 500
- oleg at redhat.com; Mon, 21 Nov 2011 21:02:53 +0100 (CET)
-Date: Mon, 21 Nov 2011 21:02:51 +0100
-From: Oleg Nesterov <oleg at redhat.com>
-To: Dave Jones <davej at redhat.com>, "Frank Ch. Eigler" <fche at redhat.com>,
- Josh Boyer <jwboyer at redhat.com>, Josh Stone <jistone at redhat.com>,
- Kyle McMartin <kmcmartin at redhat.com>
-Cc: kernel at lists.fedoraproject.org
-Subject: [PATCH 32/33] ptrace_report_syscall: check if TIF_SYSCALL_EMU is
- defined
-Message-ID: <20111121200251.GA27906 at redhat.com>
-MIME-Version: 1.0
-Content-Type: text/plain; charset=us-ascii
-Content-Disposition: inline
-In-Reply-To: <20111121200039.GA27699 at redhat.com>
-User-Agent: Mutt/1.5.18 (2008-05-17)
-X-Scanned-By: MIMEDefang 2.67 on 10.5.11.12
-Status: RO
-Content-Length: 839
-Lines: 31
-
-From: Tony Breeds <tony at bakeyournoodle.com>
-
-TIF_SYSCALL_EMU is x86 only, add ifdef into ptrace_report_syscall().
-
-Signed-off-by: Oleg Nesterov <oleg at redhat.com>
----
- include/linux/tracehook.h | 8 ++++++--
- 1 files changed, 6 insertions(+), 2 deletions(-)
-
-diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h
-index 90ca578..a1bac95 100644
---- a/include/linux/tracehook.h
-+++ b/include/linux/tracehook.h
-@@ -59,8 +59,12 @@ static inline void ptrace_report_syscall(struct pt_regs *regs)
- {
- int ptrace = current->ptrace;
-
-- if (!(ptrace & PT_SYSCALL_TRACE) && !test_thread_flag(TIF_SYSCALL_EMU))
-- return;
-+ if (!(ptrace & PT_SYSCALL_TRACE)) {
-+#ifdef TIF_SYSCALL_EMU
-+ if (!test_thread_flag(TIF_SYSCALL_EMU))
-+#endif
-+ return;
-+ }
-
- ptrace_notify(SIGTRAP | ((ptrace & PT_TRACESYSGOOD) ? 0x80 : 0));
-
---
-1.5.5.1
-
-
-From oleg at redhat.com Mon Nov 21 15:07:37 2011
-Return-Path: oleg at redhat.com
-Received: from zmta03.collab.prod.int.phx2.redhat.com (LHLO
- zmta03.collab.prod.int.phx2.redhat.com) (10.5.5.33) by
- zmail13.collab.prod.int.phx2.redhat.com with LMTP; Mon, 21 Nov 2011
- 15:07:37 -0500 (EST)
-Received: from localhost (localhost.localdomain [127.0.0.1])
- by zmta03.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 33405D8A50;
- Mon, 21 Nov 2011 15:07:37 -0500 (EST)
-Received: from zmta03.collab.prod.int.phx2.redhat.com ([127.0.0.1])
- by localhost (zmta03.collab.prod.int.phx2.redhat.com [127.0.0.1]) (amavisd-new, port 10024)
- with ESMTP id 3WH8FhvizjEA; Mon, 21 Nov 2011 15:07:37 -0500 (EST)
-Received: from int-mx09.intmail.prod.int.phx2.redhat.com (int-mx09.intmail.prod.int.phx2.redhat.com [10.5.11.22])
- by zmta03.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id 1274FD89BB;
- Mon, 21 Nov 2011 15:07:37 -0500 (EST)
-Received: from tranklukator.englab.brq.redhat.com (dhcp-1-232.brq.redhat.com [10.34.1.232])
- by int-mx09.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with SMTP id pALK7YHr017898;
- Mon, 21 Nov 2011 15:07:35 -0500
-Received: by tranklukator.englab.brq.redhat.com (nbSMTP-1.00) for uid 500
- oleg at redhat.com; Mon, 21 Nov 2011 21:02:55 +0100 (CET)
-Date: Mon, 21 Nov 2011 21:02:53 +0100
-From: Oleg Nesterov <oleg at redhat.com>
-To: Dave Jones <davej at redhat.com>, "Frank Ch. Eigler" <fche at redhat.com>,
- Josh Boyer <jwboyer at redhat.com>, Josh Stone <jistone at redhat.com>,
- Kyle McMartin <kmcmartin at redhat.com>
-Cc: kernel at lists.fedoraproject.org
-Subject: [PATCH 33/33] utrace: s390: fix the compile problem with traps.c
-Message-ID: <20111121200253.GA27913 at redhat.com>
-MIME-Version: 1.0
-Content-Type: text/plain; charset=us-ascii
-Content-Disposition: inline
-In-Reply-To: <20111121200039.GA27699 at redhat.com>
-User-Agent: Mutt/1.5.18 (2008-05-17)
-X-Scanned-By: MIMEDefang 2.68 on 10.5.11.22
-Status: RO
-Content-Length: 725
-Lines: 25
-
-d99e60e5 "tracehooks: reintroduce tracehook_consider_fatal_signal()"
-breaks the compilation of arch/s390/kernel/traps.c. Restore the
-necessary include removed by upstream 73b7d40f commit.
-
-Signed-off-by: Oleg Nesterov <oleg at redhat.com>
----
- arch/s390/kernel/traps.c | 2 +-
- 1 files changed, 1 insertions(+), 1 deletions(-)
-
-diff --git a/arch/s390/kernel/traps.c b/arch/s390/kernel/traps.c
-index f506e1b..3132498 100644
---- a/arch/s390/kernel/traps.c
-+++ b/arch/s390/kernel/traps.c
-@@ -18,7 +18,7 @@
- #include <linux/kernel.h>
- #include <linux/string.h>
- #include <linux/errno.h>
--#include <linux/ptrace.h>
-+#include <linux/tracehook.h>
- #include <linux/timer.h>
- #include <linux/mm.h>
- #include <linux/smp.h>
---
-1.5.5.1
-
More information about the scm-commits
mailing list