lkml.org 
[lkml]   [2004]   [Nov]   [3]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
Subject[patch 1/1] uml: use PTRACE_SYSEMU also for TT mode
From
Date

From: Jeff Dike <jdike@addtoit.com>, Paolo 'Blaisorblade' Giarrusso<blaisorblade_spam@yahoo.it>
Cc: Roland McGrath <roland@redhat.com>

As Jeff Dike noted, even if SYSEMU was originally written for use in SKAS
mode, there is nothing SKAS specific in SYSEMU, so use it for TT mode, too.

SYSEMU simply allows one to ptrace(PTRACE_SYSEMU) a process, that will stop on
the syscall entry path like with PTRACE_SYSCALL, but with the difference that
it makes sure that the syscall execution is totally skipped and that the child
will see the return value set by the ptracer when the child was stopped.

We even hope that the SYSEMU patch can be merged in mainline sooner than SKAS.

Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade_spam@yahoo.it>
---

vanilla-linux-2.6.9-paolo/arch/um/include/ptrace_user.h | 11 +
vanilla-linux-2.6.9-paolo/arch/um/kernel/process.c | 3
vanilla-linux-2.6.9-paolo/arch/um/kernel/process_kern.c | 56 ++++++++++
vanilla-linux-2.6.9-paolo/arch/um/kernel/skas/include/ptrace-skas.h | 10 -
vanilla-linux-2.6.9-paolo/arch/um/kernel/skas/process.c | 18 +--
vanilla-linux-2.6.9-paolo/arch/um/kernel/skas/process_kern.c | 55 ---------
vanilla-linux-2.6.9-paolo/arch/um/kernel/tt/include/tt.h | 2
vanilla-linux-2.6.9-paolo/arch/um/kernel/tt/syscall_user.c | 5
vanilla-linux-2.6.9-paolo/arch/um/kernel/tt/tracer.c | 9 +
9 files changed, 86 insertions(+), 83 deletions(-)

diff -puN arch/um/include/ptrace_user.h~uml-use-sysemu-for-tt arch/um/include/ptrace_user.h
--- vanilla-linux-2.6.9/arch/um/include/ptrace_user.h~uml-use-sysemu-for-tt 2004-11-04 05:33:06.050276464 +0100
+++ vanilla-linux-2.6.9-paolo/arch/um/include/ptrace_user.h 2004-11-04 05:33:06.105268104 +0100
@@ -15,4 +15,15 @@ extern void arch_enter_kernel(void *task
extern void arch_leave_kernel(void *task, int pid);
extern void ptrace_pokeuser(unsigned long addr, unsigned long data);

+
+/* syscall emulation path in ptrace */
+
+#ifndef PTRACE_SYSEMU
+#define PTRACE_SYSEMU 31
+#endif
+
+void set_using_sysemu(int value);
+int get_using_sysemu(void);
+extern int sysemu_supported;
+
#endif
diff -puN arch/um/kernel/skas/include/ptrace-skas.h~uml-use-sysemu-for-tt arch/um/kernel/skas/include/ptrace-skas.h
--- vanilla-linux-2.6.9/arch/um/kernel/skas/include/ptrace-skas.h~uml-use-sysemu-for-tt 2004-11-04 05:33:06.052276160 +0100
+++ vanilla-linux-2.6.9-paolo/arch/um/kernel/skas/include/ptrace-skas.h 2004-11-04 05:33:06.105268104 +0100
@@ -10,16 +10,6 @@

#ifdef UML_CONFIG_MODE_SKAS

-/* syscall emulation path in ptrace */
-
-#ifndef PTRACE_SYSEMU
-#define PTRACE_SYSEMU 31
-#endif
-
-void set_using_sysemu(int value);
-int get_using_sysemu(void);
-extern int sysemu_supported;
-
#include "skas_ptregs.h"

#define HOST_FRAME_SIZE 17
diff -puN arch/um/kernel/process.c~uml-use-sysemu-for-tt arch/um/kernel/process.c
--- vanilla-linux-2.6.9/arch/um/kernel/process.c~uml-use-sysemu-for-tt 2004-11-04 05:33:06.054275856 +0100
+++ vanilla-linux-2.6.9-paolo/arch/um/kernel/process.c 2004-11-04 05:33:06.111267192 +0100
@@ -242,9 +242,6 @@ static void __init check_sysemu(void)
void *stack;
int pid, n, status;

- if (mode_tt)
- return;
-
printk("Checking syscall emulation patch for ptrace...");
sysemu_supported = 0;
pid = start_ptraced_child(&stack);
diff -puN arch/um/kernel/tt/syscall_user.c~uml-use-sysemu-for-tt arch/um/kernel/tt/syscall_user.c
--- vanilla-linux-2.6.9/arch/um/kernel/tt/syscall_user.c~uml-use-sysemu-for-tt 2004-11-04 05:33:06.056275552 +0100
+++ vanilla-linux-2.6.9-paolo/arch/um/kernel/tt/syscall_user.c 2004-11-04 05:33:06.111267192 +0100
@@ -50,7 +50,7 @@ void syscall_handler_tt(int sig, union u
record_syscall_end(index, result);
}

-int do_syscall(void *task, int pid)
+int do_syscall(void *task, int pid, int local_using_sysemu)
{
unsigned long proc_regs[FRAME_SIZE];
union uml_pt_regs *regs;
@@ -70,6 +70,9 @@ int do_syscall(void *task, int pid)
((unsigned long *) PT_IP(proc_regs) <= &_etext))
tracer_panic("I'm tracing myself and I can't get out");

+ if(local_using_sysemu)
+ return(1);
+
if(ptrace(PTRACE_POKEUSER, pid, PT_SYSCALL_NR_OFFSET,
__NR_getpid) < 0)
tracer_panic("do_syscall : Nullifying syscall failed, "
diff -puN arch/um/kernel/tt/tracer.c~uml-use-sysemu-for-tt arch/um/kernel/tt/tracer.c
--- vanilla-linux-2.6.9/arch/um/kernel/tt/tracer.c~uml-use-sysemu-for-tt 2004-11-04 05:33:06.086270992 +0100
+++ vanilla-linux-2.6.9-paolo/arch/um/kernel/tt/tracer.c 2004-11-04 05:33:06.111267192 +0100
@@ -184,6 +184,7 @@ int tracer(int (*init_proc)(void *), voi
unsigned long eip = 0;
int status, pid = 0, sig = 0, cont_type, tracing = 0, op = 0;
int last_index, proc_id = 0, n, err, old_tracing = 0, strace = 0;
+ int pt_syscall_parm, local_using_sysemu;

capture_signal_stack();
signal(SIGPIPE, SIG_IGN);
@@ -297,6 +298,9 @@ int tracer(int (*init_proc)(void *), voi
tracing = is_tracing(task);
old_tracing = tracing;

+ local_using_sysemu = get_using_sysemu();
+ pt_syscall_parm = local_using_sysemu ? PTRACE_SYSEMU : PTRACE_SYSCALL;
+
switch(sig){
case SIGUSR1:
sig = 0;
@@ -330,7 +334,7 @@ int tracer(int (*init_proc)(void *), voi
continue;
}
tracing = 0;
- if(do_syscall(task, pid))
+ if(do_syscall(task, pid, local_using_sysemu))
sig = SIGUSR2;
else clear_singlestep(task);
break;
@@ -349,6 +353,7 @@ int tracer(int (*init_proc)(void *), voi
case SIGBUS:
case SIGILL:
case SIGWINCH:
+
default:
tracing = 0;
break;
@@ -370,7 +375,7 @@ int tracer(int (*init_proc)(void *), voi
if(tracing){
if(singlestepping_tt(task))
cont_type = PTRACE_SINGLESTEP;
- else cont_type = PTRACE_SYSCALL;
+ else cont_type = pt_syscall_parm;
}
else cont_type = PTRACE_CONT;

diff -puN arch/um/kernel/skas/process.c~uml-use-sysemu-for-tt arch/um/kernel/skas/process.c
--- vanilla-linux-2.6.9/arch/um/kernel/skas/process.c~uml-use-sysemu-for-tt 2004-11-04 05:33:06.089270536 +0100
+++ vanilla-linux-2.6.9-paolo/arch/um/kernel/skas/process.c 2004-11-04 05:33:06.112267040 +0100
@@ -139,17 +139,16 @@ void start_userspace(int cpu)

void userspace(union uml_pt_regs *regs)
{
- int err, status, op, pid = userspace_pid[0];
+ int err, status, op, pt_syscall_parm, pid = userspace_pid[0];
int local_using_sysemu; /*To prevent races if using_sysemu changes under us.*/

restore_registers(regs);

local_using_sysemu = get_using_sysemu();

- if (local_using_sysemu)
- err = ptrace(PTRACE_SYSEMU, pid, 0, 0);
- else
- err = ptrace(PTRACE_SYSCALL, pid, 0, 0);
+ pt_syscall_parm = local_using_sysemu ? PTRACE_SYSEMU : PTRACE_SYSCALL;
+ err = ptrace(pt_syscall_parm, pid, 0, 0);
+
if(err)
panic("userspace - PTRACE_%s failed, errno = %d\n",
local_using_sysemu ? "SYSEMU" : "SYSCALL", errno);
@@ -189,13 +188,10 @@ void userspace(union uml_pt_regs *regs)

/*Now we ended the syscall, so re-read local_using_sysemu.*/
local_using_sysemu = get_using_sysemu();
+ pt_syscall_parm = local_using_sysemu ? PTRACE_SYSEMU : PTRACE_SYSCALL;

- if (local_using_sysemu)
- op = singlestepping_skas() ? PTRACE_SINGLESTEP :
- PTRACE_SYSEMU;
- else
- op = singlestepping_skas() ? PTRACE_SINGLESTEP :
- PTRACE_SYSCALL;
+ op = singlestepping_skas() ? PTRACE_SINGLESTEP :
+ pt_syscall_parm;

err = ptrace(op, pid, 0, 0);
if(err)
diff -puN arch/um/kernel/skas/process_kern.c~uml-use-sysemu-for-tt arch/um/kernel/skas/process_kern.c
--- vanilla-linux-2.6.9/arch/um/kernel/skas/process_kern.c~uml-use-sysemu-for-tt 2004-11-04 05:33:06.093269928 +0100
+++ vanilla-linux-2.6.9-paolo/arch/um/kernel/skas/process_kern.c 2004-11-04 05:33:06.112267040 +0100
@@ -24,61 +24,6 @@
#include "mode.h"
#include "proc_mm.h"

-static atomic_t using_sysemu = ATOMIC_INIT(0);
-int sysemu_supported;
-
-void set_using_sysemu(int value)
-{
- atomic_set(&using_sysemu, sysemu_supported && value);
-}
-
-int get_using_sysemu(void)
-{
- return atomic_read(&using_sysemu);
-}
-
-int proc_read_sysemu(char *buf, char **start, off_t offset, int size,int *eof, void *data)
-{
- if (snprintf(buf, size, "%d\n", get_using_sysemu()) < size) /*No overflow*/
- *eof = 1;
-
- return strlen(buf);
-}
-
-int proc_write_sysemu(struct file *file,const char *buf, unsigned long count,void *data)
-{
- char tmp[2];
-
- if (copy_from_user(tmp, buf, 1))
- return -EFAULT;
-
- if (tmp[0] == '0' || tmp[0] == '1')
- set_using_sysemu(tmp[0] - '0');
- return count; /*We use the first char, but pretend to write everything*/
-}
-
-int __init make_proc_sysemu(void)
-{
- struct proc_dir_entry *ent;
- if (mode_tt || !sysemu_supported)
- return 0;
-
- ent = create_proc_entry("sysemu", 0600, &proc_root);
-
- if (ent == NULL)
- {
- printk("Failed to register /proc/sysemu\n");
- return(0);
- }
-
- ent->read_proc = proc_read_sysemu;
- ent->write_proc = proc_write_sysemu;
-
- return 0;
-}
-
-late_initcall(make_proc_sysemu);
-
int singlestepping_skas(void)
{
int ret = current->ptrace & PT_DTRACE;
diff -puN arch/um/kernel/process_kern.c~uml-use-sysemu-for-tt arch/um/kernel/process_kern.c
--- vanilla-linux-2.6.9/arch/um/kernel/process_kern.c~uml-use-sysemu-for-tt 2004-11-04 05:33:06.095269624 +0100
+++ vanilla-linux-2.6.9-paolo/arch/um/kernel/process_kern.c 2004-11-04 05:33:06.113266888 +0100
@@ -18,6 +18,7 @@
#include "linux/capability.h"
#include "linux/vmalloc.h"
#include "linux/spinlock.h"
+#include "linux/proc_fs.h"
#include "asm/unistd.h"
#include "asm/mman.h"
#include "asm/segment.h"
@@ -398,6 +399,61 @@ int cpu(void)
return(current_thread->cpu);
}

+static atomic_t using_sysemu = ATOMIC_INIT(0);
+int sysemu_supported;
+
+void set_using_sysemu(int value)
+{
+ atomic_set(&using_sysemu, sysemu_supported && value);
+}
+
+int get_using_sysemu(void)
+{
+ return atomic_read(&using_sysemu);
+}
+
+static int proc_read_sysemu(char *buf, char **start, off_t offset, int size,int *eof, void *data)
+{
+ if (snprintf(buf, size, "%d\n", get_using_sysemu()) < size) /*No overflow*/
+ *eof = 1;
+
+ return strlen(buf);
+}
+
+static int proc_write_sysemu(struct file *file,const char *buf, unsigned long count,void *data)
+{
+ char tmp[2];
+
+ if (copy_from_user(tmp, buf, 1))
+ return -EFAULT;
+
+ if (tmp[0] == '0' || tmp[0] == '1')
+ set_using_sysemu(tmp[0] - '0');
+ return count; /*We use the first char, but pretend to write everything*/
+}
+
+int __init make_proc_sysemu(void)
+{
+ struct proc_dir_entry *ent;
+ if (!sysemu_supported)
+ return 0;
+
+ ent = create_proc_entry("sysemu", 0600, &proc_root);
+
+ if (ent == NULL)
+ {
+ printk("Failed to register /proc/sysemu\n");
+ return(0);
+ }
+
+ ent->read_proc = proc_read_sysemu;
+ ent->write_proc = proc_write_sysemu;
+
+ return 0;
+}
+
+late_initcall(make_proc_sysemu);
+
/*
* Overrides for Emacs so that we follow Linus's tabbing style.
* Emacs will notice this stuff at the end of the file and automatically
diff -puN arch/um/kernel/tt/include/tt.h~uml-use-sysemu-for-tt arch/um/kernel/tt/include/tt.h
--- vanilla-linux-2.6.9/arch/um/kernel/tt/include/tt.h~uml-use-sysemu-for-tt 2004-11-04 05:33:06.097269320 +0100
+++ vanilla-linux-2.6.9-paolo/arch/um/kernel/tt/include/tt.h 2004-11-04 05:33:06.113266888 +0100
@@ -28,7 +28,7 @@ extern int singlestepping_tt(void *t);
extern void clear_singlestep(void *t);
extern void syscall_handler(int sig, union uml_pt_regs *regs);
extern void exit_kernel(int pid, void *task);
-extern int do_syscall(void *task, int pid);
+extern int do_syscall(void *task, int pid, int local_using_sysemu);
extern int is_valid_pid(int pid);
extern void remap_data(void *segment_start, void *segment_end, int w);

_
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
\
 
 \ /
  Last update: 2005-03-22 14:07    [W:0.025 / U:1.152 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site