projects/linux·commit b81a618d

Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6

b81a618d · Linus Torvalds <[email protected]> · 2011-03-23 20:51:42 -07:00

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6:
  deal with races in /proc/*/{syscall,stack,personality}
  proc: enable writing to /proc/pid/mem
  proc: make check_mem_permission() return an mm_struct on success
  proc: hold cred_guard_mutex in check_mem_permission()
  proc: disable mem_write after exec
  mm: implement access_remote_vm
  mm: factor out main logic of access_process_vm
  mm: use mm_struct to resolve gate vma's in __get_user_pages
  mm: arch: rename in_gate_area_no_task to in_gate_area_no_mm
  mm: arch: make in_gate_area take an mm_struct instead of a task_struct
  mm: arch: make get_gate_vma take an mm_struct instead of a task_struct
  x86: mark associated mm when running a task in 32 bit compatibility mode
  x86: add context tag to mark mm when running a task in 32-bit compatibility mode
  auxv: require the target to be tracable (or yourself)
  close race in /proc/*/environ
  report errors in /proc/*/*map* sanely
  pagemap: close races with suid execve
  make sessionid permissions in /proc/*/task/* match those in /proc/*
  fix leaks in path_lookupat()

Fix up trivial conflicts in fs/proc/base.c
Commit
b81a618dcd3ea99de292dbe624f41ca68f464376
Parents
2f284c84 · a9712bc1

Merge commit. Changes below are compared with the first parent: 2f284c84.

Changes

18 files changed · +244 −134

FileStatusAddDelContent
arch/powerpc/kernel/vdso.cmodified+3−3View file
arch/s390/kernel/vdso.cmodified+3−3View file
arch/sh/kernel/vsyscall/vsyscall.cmodified+3−3View file
arch/x86/ia32/ia32_aout.cmodified+1−0View file
arch/x86/include/asm/mmu.hmodified+6−0View file
arch/x86/kernel/process_64.cmodified+8−0View file
arch/x86/mm/init_64.cmodified+8−8View file
arch/x86/vdso/vdso32-setup.cmodified+8−7View file
fs/binfmt_elf.cmodified+1−1View file
fs/namei.cmodified+5−2View file
fs/proc/base.cmodified+118−63View file
fs/proc/task_mmu.cmodified+11−12View file
fs/proc/task_nommu.cmodified+3−3View file
include/linux/mm.hmodified+7−5View file
kernel/kallsyms.cmodified+2−2View file
mm/memory.cmodified+54−19View file
mm/mlock.cmodified+2−2View file
mm/nommu.cmodified+1−1View file

Only the first 600 diff lines are shown. Download the complete patch.

arch/powerpc/kernel/vdso.c

View file
diff --git a/arch/powerpc/kernel/vdso.c b/arch/powerpc/kernel/vdso.c
index fd87287..142ab10 100644
--- a/arch/powerpc/kernel/vdso.c
+++ b/arch/powerpc/kernel/vdso.c
@@ -820,17 +820,17 @@ static int __init vdso_init(void)
}
arch_initcall(vdso_init);
-int in_gate_area_no_task(unsigned long addr)
+int in_gate_area_no_mm(unsigned long addr)
{
return 0;
}
-int in_gate_area(struct task_struct *task, unsigned long addr)
+int in_gate_area(struct mm_struct *mm, unsigned long addr)
{
return 0;
}
-struct vm_area_struct *get_gate_vma(struct task_struct *tsk)
+struct vm_area_struct *get_gate_vma(struct mm_struct *mm)
{
return NULL;
}

arch/s390/kernel/vdso.c

View file
diff --git a/arch/s390/kernel/vdso.c b/arch/s390/kernel/vdso.c
index f438d74..d73630b 100644
--- a/arch/s390/kernel/vdso.c
+++ b/arch/s390/kernel/vdso.c
@@ -337,17 +337,17 @@ static int __init vdso_init(void)
}
arch_initcall(vdso_init);
-int in_gate_area_no_task(unsigned long addr)
+int in_gate_area_no_mm(unsigned long addr)
{
return 0;
}
-int in_gate_area(struct task_struct *task, unsigned long addr)
+int in_gate_area(struct mm_struct *mm, unsigned long addr)
{
return 0;
}
-struct vm_area_struct *get_gate_vma(struct task_struct *tsk)
+struct vm_area_struct *get_gate_vma(struct mm_struct *mm)
{
return NULL;
}

arch/sh/kernel/vsyscall/vsyscall.c

View file
diff --git a/arch/sh/kernel/vsyscall/vsyscall.c b/arch/sh/kernel/vsyscall/vsyscall.c
index 242117c..1d6d51a 100644
--- a/arch/sh/kernel/vsyscall/vsyscall.c
+++ b/arch/sh/kernel/vsyscall/vsyscall.c
@@ -94,17 +94,17 @@ const char *arch_vma_name(struct vm_area_struct *vma)
return NULL;
}
-struct vm_area_struct *get_gate_vma(struct task_struct *task)
+struct vm_area_struct *get_gate_vma(struct mm_struct *mm)
{
return NULL;
}
-int in_gate_area(struct task_struct *task, unsigned long address)
+int in_gate_area(struct mm_struct *mm, unsigned long address)
{
return 0;
}
-int in_gate_area_no_task(unsigned long address)
+int in_gate_area_no_mm(unsigned long address)
{
return 0;
}

arch/x86/ia32/ia32_aout.c

View file
diff --git a/arch/x86/ia32/ia32_aout.c b/arch/x86/ia32/ia32_aout.c
index 2d93bdb..fd84387 100644
--- a/arch/x86/ia32/ia32_aout.c
+++ b/arch/x86/ia32/ia32_aout.c
@@ -298,6 +298,7 @@ static int load_aout_binary(struct linux_binprm *bprm, struct pt_regs *regs)
/* OK, This is the point of no return */
set_personality(PER_LINUX);
set_thread_flag(TIF_IA32);
+ current->mm->context.ia32_compat = 1;
setup_new_exec(bprm);

arch/x86/include/asm/mmu.h

View file
diff --git a/arch/x86/include/asm/mmu.h b/arch/x86/include/asm/mmu.h
index 80a1dee..aeff3e8 100644
--- a/arch/x86/include/asm/mmu.h
+++ b/arch/x86/include/asm/mmu.h
@@ -13,6 +13,12 @@ typedef struct {
int size;
struct mutex lock;
void *vdso;
+
+#ifdef CONFIG_X86_64
+ /* True if mm supports a task running in 32 bit compatibility mode. */
+ unsigned short ia32_compat;
+#endif
+
} mm_context_t;
#ifdef CONFIG_SMP

arch/x86/kernel/process_64.c

View file
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index bd387e8..6c9dd92 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -501,6 +501,10 @@ void set_personality_64bit(void)
/* Make sure to be in 64bit mode */
clear_thread_flag(TIF_IA32);
+ /* Ensure the corresponding mm is not marked. */
+ if (current->mm)
+ current->mm->context.ia32_compat = 0;
+
/* TBD: overwrites user setup. Should have two bits.
But 64bit processes have always behaved this way,
so it's not too bad. The main problem is just that
@@ -516,6 +520,10 @@ void set_personality_ia32(void)
set_thread_flag(TIF_IA32);
current->personality |= force_personality32;
+ /* Mark the associated mm as containing 32-bit tasks. */
+ if (current->mm)
+ current->mm->context.ia32_compat = 1;
+
/* Prepare the first "return" to user space */
current_thread_info()->status |= TS_COMPAT;
}

arch/x86/mm/init_64.c

View file
diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index 2362b64..7942335 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -862,18 +862,18 @@ static struct vm_area_struct gate_vma = {
.vm_flags = VM_READ | VM_EXEC
};
-struct vm_area_struct *get_gate_vma(struct task_struct *tsk)
+struct vm_area_struct *get_gate_vma(struct mm_struct *mm)
{
#ifdef CONFIG_IA32_EMULATION
- if (test_tsk_thread_flag(tsk, TIF_IA32))
+ if (!mm || mm->context.ia32_compat)
return NULL;
#endif
return &gate_vma;
}
-int in_gate_area(struct task_struct *task, unsigned long addr)
+int in_gate_area(struct mm_struct *mm, unsigned long addr)
{
- struct vm_area_struct *vma = get_gate_vma(task);
+ struct vm_area_struct *vma = get_gate_vma(mm);
if (!vma)
return 0;
@@ -882,11 +882,11 @@ int in_gate_area(struct task_struct *task, unsigned long addr)
}
/*
- * Use this when you have no reliable task/vma, typically from interrupt
- * context. It is less reliable than using the task's vma and may give
- * false positives:
+ * Use this when you have no reliable mm, typically from interrupt
+ * context. It is less reliable than using a task's mm and may give
+ * false positives.
*/
-int in_gate_area_no_task(unsigned long addr)
+int in_gate_area_no_mm(unsigned long addr)
{
return (addr >= VSYSCALL_START) && (addr < VSYSCALL_END);
}

arch/x86/vdso/vdso32-setup.c

View file
diff --git a/arch/x86/vdso/vdso32-setup.c b/arch/x86/vdso/vdso32-setup.c
index 36df991..468d591 100644
--- a/arch/x86/vdso/vdso32-setup.c
+++ b/arch/x86/vdso/vdso32-setup.c
@@ -417,24 +417,25 @@ const char *arch_vma_name(struct vm_area_struct *vma)
return NULL;
}
-struct vm_area_struct *get_gate_vma(struct task_struct *tsk)
+struct vm_area_struct *get_gate_vma(struct mm_struct *mm)
{
- struct mm_struct *mm = tsk->mm;
-
- /* Check to see if this task was created in compat vdso mode */
+ /*
+ * Check to see if the corresponding task was created in compat vdso
+ * mode.
+ */
if (mm && mm->context.vdso == (void *)VDSO_HIGH_BASE)
return &gate_vma;
return NULL;
}
-int in_gate_area(struct task_struct *task, unsigned long addr)
+int in_gate_area(struct mm_struct *mm, unsigned long addr)
{
- const struct vm_area_struct *vma = get_gate_vma(task);
+ const struct vm_area_struct *vma = get_gate_vma(mm);
return vma && addr >= vma->vm_start && addr < vma->vm_end;
}
-int in_gate_area_no_task(unsigned long addr)
+int in_gate_area_no_mm(unsigned long addr)
{
return 0;
}

fs/binfmt_elf.c

View file
diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
index b2fae00..f34078d 100644
--- a/fs/binfmt_elf.c
+++ b/fs/binfmt_elf.c
@@ -1906,7 +1906,7 @@ static int elf_core_dump(struct coredump_params *cprm)
segs = current->mm->map_count;
segs += elf_core_extra_phdrs();
- gate_vma = get_gate_vma(current);
+ gate_vma = get_gate_vma(current->mm);
if (gate_vma != NULL)
segs++;

fs/namei.c

View file
diff --git a/fs/namei.c b/fs/namei.c
index fc858b1..d0066e1 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -1650,13 +1650,16 @@ static int path_lookupat(int dfd, const char *name,
err = -ECHILD;
}
- if (!err)
+ if (!err) {
err = handle_reval_path(nd);
+ if (err)
+ path_put(&nd->path);
+ }
if (!err && nd->flags & LOOKUP_DIRECTORY) {
if (!nd->inode->i_op->lookup) {
path_put(&nd->path);
- return -ENOTDIR;
+ err = -ENOTDIR;
}
}

fs/proc/base.c

View file
diff --git a/fs/proc/base.c b/fs/proc/base.c
index daba136..5a670c1 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -191,17 +191,20 @@ static int proc_root_link(struct inode *inode, struct path *path)
return result;
}
-/*
- * Return zero if current may access user memory in @task, -error if not.
- */
-static int check_mem_permission(struct task_struct *task)
+static struct mm_struct *__check_mem_permission(struct task_struct *task)
{
+ struct mm_struct *mm;
+
+ mm = get_task_mm(task);
+ if (!mm)
+ return ERR_PTR(-EINVAL);
+
/*
* A task can always look at itself, in case it chooses
* to use system calls instead of load instructions.
*/
if (task == current)
- return 0;
+ return mm;
/*
* If current is actively ptrace'ing, and would also be
@@ -213,27 +216,53 @@ static int check_mem_permission(struct task_struct *task)
match = (tracehook_tracer_task(task) == current);
rcu_read_unlock();
if (match && ptrace_may_access(task, PTRACE_MODE_ATTACH))
- return 0;
+ return mm;
}
/*
* Noone else is allowed.
*/
- return -EPERM;
+ mmput(mm);
+ return ERR_PTR(-EPERM);
+}
+
+/*
+ * If current may access user memory in @task return a reference to the
+ * corresponding mm, otherwise ERR_PTR.
+ */
+static struct mm_struct *check_mem_permission(struct task_struct *task)
+{
+ struct mm_struct *mm;
+ int err;
+
+ /*
+ * Avoid racing if task exec's as we might get a new mm but validate
+ * against old credentials.
+ */
+ err = mutex_lock_killable(&task->signal->cred_guard_mutex);
+ if (err)
+ return ERR_PTR(err);
+
+ mm = __check_mem_permission(task);
+ mutex_unlock(&task->signal->cred_guard_mutex);
+
+ return mm;
}
struct mm_struct *mm_for_maps(struct task_struct *task)
{
struct mm_struct *mm;
+ int err;
- if (mutex_lock_killable(&task->signal->cred_guard_mutex))
- return NULL;
+ err = mutex_lock_killable(&task->signal->cred_guard_mutex);
+ if (err)
+ return ERR_PTR(err);
mm = get_task_mm(task);
if (mm && mm != current->mm &&
!ptrace_may_access(task, PTRACE_MODE_READ)) {
mmput(mm);
- mm = NULL;
+ mm = ERR_PTR(-EACCES);
}
mutex_unlock(&task->signal->cred_guard_mutex);
@@ -279,9 +308,9 @@ out:
static int proc_pid_auxv(struct task_struct *task, char *buffer)
{
- int res = 0;
- struct mm_struct *mm = get_task_mm(task);
- if (mm) {
+ struct mm_struct *mm = mm_for_maps(task);
+ int res = PTR_ERR(mm);
+ if (mm && !IS_ERR(mm)) {
unsigned int nwords = 0;
do {
nwords += 2;
@@ -318,6 +347,23 @@ static int proc_pid_wchan(struct task_struct *task, char *buffer)
}
#endif /* CONFIG_KALLSYMS */
+static int lock_trace(struct task_struct *task)
+{
+ int err = mutex_lock_killable(&task->signal->cred_guard_mutex);
+ if (err)
+ return err;
+ if (!ptrace_may_access(task, PTRACE_MODE_ATTACH)) {
+ mutex_unlock(&task->signal->cred_guard_mutex);
+ return -EPERM;
+ }
+ return 0;
+}
+
+static void unlock_trace(struct task_struct *task)
+{
+ mutex_unlock(&task->signal->cred_guard_mutex);
+}
+
#ifdef CONFIG_STACKTRACE
#define MAX_STACK_TRACE_DEPTH 64
@@ -327,6 +373,7 @@ static int proc_pid_stack(struct seq_file *m, struct pid_namespace *ns,
{
struct stack_trace trace;
unsigned long *entries;
+ int err;
int i;
entries = kmalloc(MAX_STACK_TRACE_DEPTH * sizeof(*entries), GFP_KERNEL);
@@ -337,15 +384,20 @@ static int proc_pid_stack(struct seq_file *m, struct pid_namespace *ns,
trace.max_entries = MAX_STACK_TRACE_DEPTH;
trace.entries = entries;
trace.skip = 0;
- save_stack_trace_tsk(task, &trace);
- for (i = 0; i < trace.nr_entries; i++) {
- seq_printf(m, "[<%pK>] %pS\n",
- (void *)entries[i], (void *)entries[i]);
+ err = lock_trace(task);
+ if (!err) {
+ save_stack_trace_tsk(task, &trace);
+
+ for (i = 0; i < trace.nr_entries; i++) {
+ seq_printf(m, "[<%pK>] %pS\n",
+ (void *)entries[i], (void *)entries[i]);
+ }
+ unlock_trace(task);
}
kfree(entries);
- return 0;
+ return err;
}
#endif
@@ -508,18 +560,22 @@ static int proc_pid_syscall(struct task_struct *task, char *buffer)
{
long nr;
unsigned long args[6], sp, pc;
+ int res = lock_trace(task);
+ if (res)
+ return res;
if (task_current_syscall(task, &nr, args, 6, &sp, &pc))
- return sprintf(buffer, "running\n");
-
- if (nr < 0)
- return sprintf(buffer, "%ld 0x%lx 0x%lx\n", nr, sp, pc);
-
- return sprintf(buffer,
+ res = sprintf(buffer, "running\n");
+ else if (nr < 0)
+ res = sprintf(buffer, "%ld 0x%lx 0x%lx\n", nr, sp, pc);
+ else
+ res = sprintf(buffer,
"%ld 0x%lx 0x%lx 0x%lx 0x%lx 0x%lx 0x%lx 0x%lx 0x%lx\n",
nr,
args[0], args[1], args[2], args[3], args[4], args[5],
sp, pc);
+ unlock_trace(task);
+ return res;
}
#endif /* CONFIG_HAVE_ARCH_TRACEHOOK */
@@ -775,19 +831,14 @@ static ssize_t mem_read(struct file * file, char __user * buf,
if (!task)
goto out_no_task;
- ret = check_mem_permission(task);
- if (ret)
- goto out;
-
ret = -ENOMEM;
page = (char *)__get_free_page(GFP_TEMPORARY);
if (!page)
goto out;
- ret = 0;
-
- mm = get_task_mm(task);
- if (!mm)
+ mm = check_mem_permission(task);
+ ret = PTR_ERR(mm);
+ if (IS_ERR(mm))
goto out_free;
ret = -EIO;
@@ -801,8 +852,8 @@ static ssize_t mem_read(struct file * file, char __user * buf,
int this_len, retval;
this_len = (count > PAGE_SIZE) ? PAGE_SIZE : count;
- retval = access_process_vm(task, src, page, this_len, 0);
- if (!retval || check_mem_permission(task)) {
+ retval = access_remote_vm(mm, src, page, this_len, 0);
+ if (!retval) {
if (!ret)
ret = -EIO;
break;
@@ -830,10 +881,6 @@ out_no_task:
return ret;
}
-#define mem_write NULL
-
-#ifndef mem_write
-/* This is a security hazard */
static ssize_t mem_write(struct file * file, const char __user *buf,
size_t count, loff_t *ppos)
{
@@ -841,19 +888,25 @@ static ssize_t mem_write(struct file * file, const char __user *buf,
char *page;
struct task_struct *task = get_proc_task(file->f_path.dentry->d_inode);
unsigned long dst = *ppos;
+ struct mm_struct *mm;
copied = -ESRCH;
if (!task)
goto out_no_task;
- copied = check_mem_permission(task);
- if (copied)
- goto out;
+ mm = check_mem_permission(task);
+ copied = PTR_ERR(mm);
+ if (IS_ERR(mm))
+ goto out_task;
+
+ copied = -EIO;
+ if (file->private_data != (void *)((long)current->self_exec_id))
+ goto out_mm;
copied = -ENOMEM;
page = (char *)__get_free_page(GFP_TEMPORARY);
if (!page)
- goto out;
+ goto out_mm;
copied = 0;
while (count > 0) {
@@ -864,7 +917,7 @@ static ssize_t mem_write(struct file * file, const char __user *buf,
copied = -EFAULT;
break;
}
- retval = access_process_vm(task, dst, page, this_len, 1);
+ retval = access_remote_vm(mm, dst, page, this_len, 1);
if (!retval) {
if (!copied)
copied = -EIO;
@@ -877,12 +930,13 @@ static ssize_t mem_write(struct file * file, const char __user *buf,
}
*ppos = dst;
free_page((unsigned long) page);
-out:
+out_mm:
+ mmput(mm);
+out_task:
put_task_struct(task);
out_no_task:
return copied;
}
-#endif
loff_t mem_lseek(struct file *file, loff_t offset, int orig)
{
@@ -919,21 +973,18 @@ static ssize_t environ_read(struct file *file, char __user *buf,
if (!task)
goto out_no_task;
- ret = -EPERM;
- if (!ptrace_may_access(task, PTRACE_MODE_READ))
- goto out;
-
ret = -ENOMEM;
page = (char *)__get_free_page(GFP_TEMPORARY);
if (!page)
goto out;
- ret = 0;
- mm = get_task_mm(task);
- if (!mm)
+ mm = mm_for_maps(task);
+ ret = PTR_ERR(mm);
+ if (!mm || IS_ERR(mm))
goto out_free;
+ ret = 0;
while (count > 0) {
int this_len, retval, max_len;
@@ -2751,8 +2802,12 @@ static int proc_tgid_io_accounting(struct task_struct *task, char *buffer)
static int proc_pid_personality(struct seq_file *m, struct pid_namespace *ns,
struct pid *pid, struct task_struct *task)
{
- seq_printf(m, "%08x\n", task->personality);
- return 0;
+ int err = lock_trace(task);
+ if (!err) {
+ seq_printf(m, "%08x\n", task->personality);
+ unlock_trace(task);
+ }
+ return err;
}
/*
@@ -2771,7 +2826,7 @@ static const struct pid_entry tgid_base_stuff[] = {
REG("environ", S_IRUSR, proc_environ_operations),
INF("auxv", S_IRUSR, proc_pid_auxv),
ONE("status", S_IRUGO, proc_pid_status),
- ONE("personality", S_IRUSR, proc_pid_personality),
+ ONE("personality", S_IRUGO, proc_pid_personality),
INF("limits", S_IRUGO, proc_pid_limits),
#ifdef CONFIG_SCHED_DEBUG
REG("sched", S_IRUGO|S_IWUSR, proc_pid_sched_operations),
@@ -2781,7 +2836,7 @@ static const struct pid_entry tgid_base_stuff[] = {
#endif
REG("comm", S_IRUGO|S_IWUSR, proc_pid_set_comm_operations),
#ifdef CONFIG_HAVE_ARCH_TRACEHOOK
- INF("syscall", S_IRUSR, proc_pid_syscall),
+ INF("syscall", S_IRUGO, proc_pid_syscall),
#endif
INF("cmdline", S_IRUGO, proc_pid_cmdline),
ONE("stat", S_IRUGO, proc_tgid_stat),
@@ -2800,7 +2855,7 @@ static const struct pid_entry tgid_base_stuff[] = {
#ifdef CONFIG_PROC_PAGE_MONITOR
REG("clear_refs", S_IWUSR, proc_clear_refs_operations),
REG("smaps", S_IRUGO, proc_smaps_operations),
- REG("pagemap", S_IRUSR, proc_pagemap_operations),
+ REG("pagemap", S_IRUGO, proc_pagemap_operations),

This diff is incomplete. Download the complete patch.

fs/proc/task_mmu.c

View file

This diff is incomplete. Download the complete patch.

fs/proc/task_nommu.c

View file

This diff is incomplete. Download the complete patch.

include/linux/mm.h

View file

This diff is incomplete. Download the complete patch.

kernel/kallsyms.c

View file

This diff is incomplete. Download the complete patch.

mm/memory.c

View file

This diff is incomplete. Download the complete patch.

mm/mlock.c

View file

This diff is incomplete. Download the complete patch.

mm/nommu.c

View file

This diff is incomplete. Download the complete patch.