提交 9786a8f3 编写于 作者: B Bodo Stroesser 提交者: Linus Torvalds

[PATCH] uml: Proper clone support for skas0

This patch implements the clone-stub mechanism, which allows skas0 to run
with proc_mm==0, even if the clib in UML uses modify_ldt.

Note: There is a bug in skas3.v7 host patch, that avoids UML-skas from
running properly on a SMP-box.  In full skas3, I never really saw problems,
but in skas0 they showed up.

More commentary by jdike - What this patch does is makes sure that the host
parent of each new host process matches the UML parent of the corresponding
UML process.  This ensures that any changed LDTs are inherited.  This is
done by having clone actually called by the UML process from its stub,
rather than by the kernel.  We have special syscall stubs that are loaded
onto the stub code page because that code must be completely
self-contained.  These stubs are given C interfaces, and used like normal C
functions, but there are subtleties.  Principally, we have to be careful
about stack variables in stub_clone_handler after the clone.  The code is
written so that there aren't any - everything boils down to a fixed
address.  If there were any locals, references to them after the clone
would be wrong because the stack just changed.
Signed-off-by: NBodo Stroesser <bstroesser@fujitsu-siemens.com>
Signed-off-by: NJeff Dike <jdike@addtoit.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>
上级 d67b569f
...@@ -10,9 +10,56 @@ ...@@ -10,9 +10,56 @@
#include <asm/unistd.h> #include <asm/unistd.h>
extern void stub_segv_handler(int sig); extern void stub_segv_handler(int sig);
extern void stub_clone_handler(void);
#define STUB_SYSCALL_RET EAX #define STUB_SYSCALL_RET EAX
#define STUB_MMAP_NR __NR_mmap2 #define STUB_MMAP_NR __NR_mmap2
#define MMAP_OFFSET(o) ((o) >> PAGE_SHIFT) #define MMAP_OFFSET(o) ((o) >> PAGE_SHIFT)
static inline long stub_syscall2(long syscall, long arg1, long arg2)
{
long ret;
__asm__("movl %0, %%ecx; " : : "g" (arg2) : "%ecx");
__asm__("movl %0, %%ebx; " : : "g" (arg1) : "%ebx");
__asm__("movl %0, %%eax; " : : "g" (syscall) : "%eax");
__asm__("int $0x80;" : : : "%eax");
__asm__ __volatile__("movl %%eax, %0; " : "=g" (ret) :);
return(ret);
}
static inline long stub_syscall3(long syscall, long arg1, long arg2, long arg3)
{
__asm__("movl %0, %%edx; " : : "g" (arg3) : "%edx");
return(stub_syscall2(syscall, arg1, arg2));
}
static inline long stub_syscall4(long syscall, long arg1, long arg2, long arg3,
long arg4)
{
__asm__("movl %0, %%esi; " : : "g" (arg4) : "%esi");
return(stub_syscall3(syscall, arg1, arg2, arg3));
}
static inline long stub_syscall6(long syscall, long arg1, long arg2, long arg3,
long arg4, long arg5, long arg6)
{
long ret;
__asm__("movl %0, %%eax; " : : "g" (syscall) : "%eax");
__asm__("movl %0, %%ebx; " : : "g" (arg1) : "%ebx");
__asm__("movl %0, %%ecx; " : : "g" (arg2) : "%ecx");
__asm__("movl %0, %%edx; " : : "g" (arg3) : "%edx");
__asm__("movl %0, %%esi; " : : "g" (arg4) : "%esi");
__asm__("movl %0, %%edi; " : : "g" (arg5) : "%edi");
__asm__ __volatile__("pushl %%ebp ; movl %1, %%ebp; "
"int $0x80; popl %%ebp ; "
"movl %%eax, %0; " : "=g" (ret) : "g" (arg6) : "%eax");
return(ret);
}
static inline void trap_myself(void)
{
__asm("int3");
}
#endif #endif
...@@ -11,9 +11,48 @@ ...@@ -11,9 +11,48 @@
#include <sysdep/ptrace_user.h> #include <sysdep/ptrace_user.h>
extern void stub_segv_handler(int sig); extern void stub_segv_handler(int sig);
extern void stub_clone_handler(void);
#define STUB_SYSCALL_RET PT_INDEX(RAX) #define STUB_SYSCALL_RET PT_INDEX(RAX)
#define STUB_MMAP_NR __NR_mmap #define STUB_MMAP_NR __NR_mmap
#define MMAP_OFFSET(o) (o) #define MMAP_OFFSET(o) (o)
static inline long stub_syscall2(long syscall, long arg1, long arg2)
{
long ret;
__asm__("movq %0, %%rsi; " : : "g" (arg2) : "%rsi");
__asm__("movq %0, %%rdi; " : : "g" (arg1) : "%rdi");
__asm__("movq %0, %%rax; " : : "g" (syscall) : "%rax");
__asm__("syscall;" : : : "%rax", "%r11", "%rcx");
__asm__ __volatile__("movq %%rax, %0; " : "=g" (ret) :);
return(ret);
}
static inline long stub_syscall3(long syscall, long arg1, long arg2, long arg3)
{
__asm__("movq %0, %%rdx; " : : "g" (arg3) : "%rdx");
return(stub_syscall2(syscall, arg1, arg2));
}
static inline long stub_syscall4(long syscall, long arg1, long arg2, long arg3,
long arg4)
{
__asm__("movq %0, %%r10; " : : "g" (arg4) : "%r10");
return(stub_syscall3(syscall, arg1, arg2, arg3));
}
static inline long stub_syscall6(long syscall, long arg1, long arg2, long arg3,
long arg4, long arg5, long arg6)
{
__asm__("movq %0, %%r9; " : : "g" (arg6) : "%r9");
__asm__("movq %0, %%r8; " : : "g" (arg5) : "%r8");
return(stub_syscall4(syscall, arg1, arg2, arg3, arg4));
}
static inline void trap_myself(void)
{
__asm("int3");
}
#endif #endif
...@@ -10,6 +10,7 @@ extern void timer(void); ...@@ -10,6 +10,7 @@ extern void timer(void);
extern void switch_timers(int to_real); extern void switch_timers(int to_real);
extern void idle_sleep(int secs); extern void idle_sleep(int secs);
extern void enable_timer(void); extern void enable_timer(void);
extern void prepare_timer(void * ptr);
extern void disable_timer(void); extern void disable_timer(void);
extern unsigned long time_lock(void); extern unsigned long time_lock(void);
extern void time_unlock(unsigned long); extern void time_unlock(unsigned long);
......
...@@ -3,11 +3,14 @@ ...@@ -3,11 +3,14 @@
# Licensed under the GPL # Licensed under the GPL
# #
obj-y := exec_kern.o mem.o mem_user.o mmu.o process.o process_kern.o \ obj-y := clone.o exec_kern.o mem.o mem_user.o mmu.o process.o process_kern.o \
syscall_kern.o syscall_user.o tlb.o trap_user.o uaccess.o \ syscall_kern.o syscall_user.o tlb.o trap_user.o uaccess.o \
subdir- := util subdir- := util
USER_OBJS := process.o USER_OBJS := process.o clone.o
include arch/um/scripts/Makefile.rules include arch/um/scripts/Makefile.rules
# clone.o is in the stub, so it can't be built with profiling
$(obj)/clone.o : c_flags = -Wp,-MD,$(depfile) $(call unprofile,$(USER_CFLAGS))
#include <sched.h>
#include <signal.h>
#include <sys/mman.h>
#include <sys/time.h>
#include <asm/unistd.h>
#include <asm/page.h>
#include "ptrace_user.h"
#include "skas.h"
#include "stub-data.h"
#include "uml-config.h"
#include "sysdep/stub.h"
/* This is in a separate file because it needs to be compiled with any
* extraneous gcc flags (-pg, -fprofile-arcs, -ftest-coverage) disabled
*/
void __attribute__ ((__section__ (".__syscall_stub")))
stub_clone_handler(void)
{
long err;
struct stub_data *from = (struct stub_data *) UML_CONFIG_STUB_DATA;
err = stub_syscall2(__NR_clone, CLONE_PARENT | CLONE_FILES | SIGCHLD,
UML_CONFIG_STUB_DATA + PAGE_SIZE / 2 -
sizeof(void *));
if(err != 0)
goto out;
err = stub_syscall4(__NR_ptrace, PTRACE_TRACEME, 0, 0, 0);
if(err)
goto out;
err = stub_syscall3(__NR_setitimer, ITIMER_VIRTUAL,
(long) &from->timer, 0);
if(err)
goto out;
err = stub_syscall6(STUB_MMAP_NR, UML_CONFIG_STUB_DATA, PAGE_SIZE,
PROT_READ | PROT_WRITE, MAP_FIXED | MAP_SHARED,
from->fd, from->offset);
out:
/* save current result. Parent: pid; child: retcode of mmap */
from->err = err;
trap_myself();
}
...@@ -32,6 +32,7 @@ extern int protect(struct mm_id * mm_idp, unsigned long addr, ...@@ -32,6 +32,7 @@ extern int protect(struct mm_id * mm_idp, unsigned long addr,
extern void user_signal(int sig, union uml_pt_regs *regs, int pid); extern void user_signal(int sig, union uml_pt_regs *regs, int pid);
extern int new_mm(int from); extern int new_mm(int from);
extern int start_userspace(unsigned long stub_stack); extern int start_userspace(unsigned long stub_stack);
extern int copy_context_skas0(unsigned long stack, int pid);
extern void get_skas_faultinfo(int pid, struct faultinfo * fi); extern void get_skas_faultinfo(int pid, struct faultinfo * fi);
extern long execute_syscall_skas(void *r); extern long execute_syscall_skas(void *r);
extern unsigned long current_stub_stack(void); extern unsigned long current_stub_stack(void);
......
/*
* Copyright (C) 2005 Jeff Dike (jdike@karaya.com)
* Licensed under the GPL
*/
#ifndef __STUB_DATA_H
#define __STUB_DATA_H
#include <sys/time.h>
struct stub_data {
long offset;
int fd;
struct itimerval timer;
long err;
};
#endif
...@@ -75,6 +75,7 @@ static int init_stub_pte(struct mm_struct *mm, unsigned long proc, ...@@ -75,6 +75,7 @@ static int init_stub_pte(struct mm_struct *mm, unsigned long proc,
int init_new_context_skas(struct task_struct *task, struct mm_struct *mm) int init_new_context_skas(struct task_struct *task, struct mm_struct *mm)
{ {
struct mm_struct *cur_mm = current->mm; struct mm_struct *cur_mm = current->mm;
struct mm_id *cur_mm_id = &cur_mm->context.skas.id;
struct mm_id *mm_id = &mm->context.skas.id; struct mm_id *mm_id = &mm->context.skas.id;
unsigned long stack; unsigned long stack;
int from, ret; int from, ret;
...@@ -115,7 +116,11 @@ int init_new_context_skas(struct task_struct *task, struct mm_struct *mm) ...@@ -115,7 +116,11 @@ int init_new_context_skas(struct task_struct *task, struct mm_struct *mm)
goto out_free; goto out_free;
mm->nr_ptes--; mm->nr_ptes--;
mm_id->u.pid = start_userspace(stack);
if((cur_mm != NULL) && (cur_mm != &init_mm))
mm_id->u.pid = copy_context_skas0(stack,
cur_mm_id->u.pid);
else mm_id->u.pid = start_userspace(stack);
} }
return 0; return 0;
......
...@@ -13,6 +13,7 @@ ...@@ -13,6 +13,7 @@
#include <sys/wait.h> #include <sys/wait.h>
#include <sys/mman.h> #include <sys/mman.h>
#include <sys/user.h> #include <sys/user.h>
#include <sys/time.h>
#include <asm/unistd.h> #include <asm/unistd.h>
#include <asm/types.h> #include <asm/types.h>
#include "user.h" #include "user.h"
...@@ -22,6 +23,7 @@ ...@@ -22,6 +23,7 @@
#include "user_util.h" #include "user_util.h"
#include "kern_util.h" #include "kern_util.h"
#include "skas.h" #include "skas.h"
#include "stub-data.h"
#include "mm_id.h" #include "mm_id.h"
#include "sysdep/sigcontext.h" #include "sysdep/sigcontext.h"
#include "sysdep/stub.h" #include "sysdep/stub.h"
...@@ -296,6 +298,67 @@ void userspace(union uml_pt_regs *regs) ...@@ -296,6 +298,67 @@ void userspace(union uml_pt_regs *regs)
#define INIT_JMP_HALT 3 #define INIT_JMP_HALT 3
#define INIT_JMP_REBOOT 4 #define INIT_JMP_REBOOT 4
int copy_context_skas0(unsigned long new_stack, int pid)
{
int err;
unsigned long regs[MAX_REG_NR];
unsigned long current_stack = current_stub_stack();
struct stub_data *data = (struct stub_data *) current_stack;
struct stub_data *child_data = (struct stub_data *) new_stack;
__u64 new_offset;
int new_fd = phys_mapping(to_phys((void *)new_stack), &new_offset);
/* prepare offset and fd of child's stack as argument for parent's
* and child's mmap2 calls
*/
*data = ((struct stub_data) { .offset = MMAP_OFFSET(new_offset),
.fd = new_fd,
.timer = ((struct itimerval)
{ { 0, 1000000 / hz() },
{ 0, 1000000 / hz() }})});
get_safe_registers(regs);
/* Set parent's instruction pointer to start of clone-stub */
regs[REGS_IP_INDEX] = UML_CONFIG_STUB_CODE +
(unsigned long) stub_clone_handler -
(unsigned long) &__syscall_stub_start;
regs[REGS_SP_INDEX] = UML_CONFIG_STUB_DATA + PAGE_SIZE -
sizeof(void *);
err = ptrace_setregs(pid, regs);
if(err < 0)
panic("copy_context_skas0 : PTRACE_SETREGS failed, "
"pid = %d, errno = %d\n", pid, errno);
/* set a well known return code for detection of child write failure */
child_data->err = 12345678;
/* Wait, until parent has finished its work: read child's pid from
* parent's stack, and check, if bad result.
*/
wait_stub_done(pid, 0, "copy_context_skas0");
pid = data->err;
if(pid < 0)
panic("copy_context_skas0 - stub-parent reports error %d\n",
pid);
/* Wait, until child has finished too: read child's result from
* child's stack and check it.
*/
wait_stub_done(pid, -1, "copy_context_skas0");
if (child_data->err != UML_CONFIG_STUB_DATA)
panic("copy_context_skas0 - stub-child reports error %d\n",
child_data->err);
if (ptrace(PTRACE_OLDSETOPTIONS, pid, NULL,
(void *)PTRACE_O_TRACESYSGOOD) < 0)
panic("copy_context_skas0 : PTRACE_SETOPTIONS failed, "
"errno = %d\n", errno);
return pid;
}
void new_thread(void *stack, void **switch_buf_ptr, void **fork_buf_ptr, void new_thread(void *stack, void **switch_buf_ptr, void **fork_buf_ptr,
void (*handler)(int)) void (*handler)(int))
{ {
......
...@@ -48,6 +48,13 @@ void enable_timer(void) ...@@ -48,6 +48,13 @@ void enable_timer(void)
set_interval(ITIMER_VIRTUAL); set_interval(ITIMER_VIRTUAL);
} }
void prepare_timer(void * ptr)
{
int usec = 1000000/hz();
*(struct itimerval *)ptr = ((struct itimerval) { { 0, usec },
{ 0, usec }});
}
void disable_timer(void) void disable_timer(void)
{ {
struct itimerval disable = ((struct itimerval) { { 0, 0 }, { 0, 0 }}); struct itimerval disable = ((struct itimerval) { { 0, 0 }, { 0, 0 }});
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册