From 826eba4db0c643cc0e705c8365902c63fac2b7ee Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 3 Aug 2012 12:14:44 +0400 Subject: the only place that needs to include asm/exec.h is linux/binfmts.h Signed-off-by: Al Viro --- include/linux/binfmts.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/linux/binfmts.h b/include/linux/binfmts.h index 366422bc1633..8938beabad7a 100644 --- a/include/linux/binfmts.h +++ b/include/linux/binfmts.h @@ -19,6 +19,7 @@ struct pt_regs; #ifdef __KERNEL__ #include +#include #define CORENAME_MAX_SIZE 128 -- cgit v1.2.3 From 2aa3a7f8660355c3dddead17e224545c1a3d5a5f Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 21 Sep 2012 19:55:31 -0400 Subject: preparation for generic kernel_thread() Let architectures select GENERIC_KERNEL_THREAD and have their copy_thread() treat NULL regs as "it came from kernel_thread(), sp argument contains the function new thread will be calling and stack_size - the argument for that function". Switching the architectures begins shortly... Signed-off-by: Al Viro --- arch/Kconfig | 3 +++ include/linux/sched.h | 3 +++ kernel/fork.c | 13 ++++++++++++- 3 files changed, 18 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/arch/Kconfig b/arch/Kconfig index 72f2fa189cc5..d397e11d167d 100644 --- a/arch/Kconfig +++ b/arch/Kconfig @@ -258,6 +258,9 @@ config ARCH_WANT_OLD_COMPAT_IPC select ARCH_WANT_COMPAT_IPC_PARSE_VERSION bool +config GENERIC_KERNEL_THREAD + bool + config HAVE_ARCH_SECCOMP_FILTER bool help diff --git a/include/linux/sched.h b/include/linux/sched.h index 23bddac4bad8..34da9340c6a4 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -2325,6 +2325,9 @@ extern int do_execve(const char *, const char __user * const __user *, struct pt_regs *); extern long do_fork(unsigned long, unsigned long, struct pt_regs *, unsigned long, int __user *, int __user *); struct task_struct *fork_idle(int); +#ifdef CONFIG_GENERIC_KERNEL_THREAD +extern pid_t kernel_thread(int (*fn)(void *), void *arg, unsigned long flags); +#endif extern void set_task_comm(struct task_struct *tsk, char *from); extern char *get_task_comm(char *to, struct task_struct *tsk); diff --git a/kernel/fork.c b/kernel/fork.c index 2c8857e12855..a42c62a8eb24 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1609,7 +1609,7 @@ long do_fork(unsigned long clone_flags, * requested, no event is reported; otherwise, report if the event * for the type of forking is enabled. */ - if (likely(user_mode(regs)) && !(clone_flags & CLONE_UNTRACED)) { + if (!(clone_flags & CLONE_UNTRACED) && likely(user_mode(regs))) { if (clone_flags & CLONE_VFORK) trace = PTRACE_EVENT_VFORK; else if ((clone_flags & CSIGNAL) != SIGCHLD) @@ -1659,6 +1659,17 @@ long do_fork(unsigned long clone_flags, return nr; } +#ifdef CONFIG_GENERIC_KERNEL_THREAD +/* + * Create a kernel thread. + */ +pid_t kernel_thread(int (*fn)(void *), void *arg, unsigned long flags) +{ + return do_fork(flags|CLONE_VM|CLONE_UNTRACED, (unsigned long)fn, NULL, + (unsigned long)arg, NULL, NULL); +} +#endif + #ifndef ARCH_MIN_MMSTRUCT_ALIGN #define ARCH_MIN_MMSTRUCT_ALIGN 0 #endif -- cgit v1.2.3 From a3460a59747cfddfa7be4758e5ef08bf5d751d59 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 30 Sep 2012 13:12:36 -0400 Subject: new helper: current_pt_regs() Normally (and that's the default) it's just task_pt_regs(current). However, if an architecture can optimize that, it can do so by making a macro of its own available from asm/ptrace.h. More importantly, some architectures have task_pt_regs() working only for traced tasks blocked on signal delivery. current_pt_regs() needs to work for *all* processes, so before those architectures start using stuff relying on current_pt_regs() they'll need a properly working variant. Signed-off-by: Al Viro --- include/linux/ptrace.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include') diff --git a/include/linux/ptrace.h b/include/linux/ptrace.h index 597e4fdb97fe..07fd922d6928 100644 --- a/include/linux/ptrace.h +++ b/include/linux/ptrace.h @@ -403,6 +403,10 @@ static inline void user_single_step_siginfo(struct task_struct *tsk, #define arch_ptrace_stop(code, info) do { } while (0) #endif +#ifndef current_pt_regs +#define current_pt_regs() task_pt_regs(current) +#endif + extern int task_current_syscall(struct task_struct *target, long *callno, unsigned long args[6], unsigned int maxargs, unsigned long *sp, unsigned long *pc); -- cgit v1.2.3 From 282124d18626379a20b41d25e0c580f290cd09d4 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 30 Sep 2012 13:20:09 -0400 Subject: generic kernel_execve() based mostly on arm and alpha versions. Architectures can define __ARCH_WANT_KERNEL_EXECVE and use it, provided that * they have working current_pt_regs(), even for kernel threads. * kernel_thread-spawned threads do have space for pt_regs in the normal location. Normally that's as simple as switching to generic kernel_thread() and making sure that kernel threads do *not* go through return from syscall path; call the payload from equivalent of ret_from_fork if we are in a kernel thread (or just have separate ret_from_kernel_thread and make copy_thread() use it instead of ret_from_fork in kernel thread case). * they have ret_from_kernel_execve(); it is called after successful do_execve() done by kernel_execve() and gets normal pt_regs location passed to it as argument. It's essentially a longjmp() analog - it should set sp, etc. to the situation expected at the return for syscall and go there. Eventually the need for that sucker will disappear, but that'll take some surgery on kernel_thread() payloads. Signed-off-by: Al Viro --- fs/exec.c | 22 ++++++++++++++++++++++ include/linux/binfmts.h | 5 +++++ 2 files changed, 27 insertions(+) (limited to 'include') diff --git a/fs/exec.c b/fs/exec.c index d7f9e14f8977..48c525115fe4 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -2318,3 +2318,25 @@ int dump_seek(struct file *file, loff_t off) return ret; } EXPORT_SYMBOL(dump_seek); + +#ifdef __ARCH_WANT_KERNEL_EXECVE +int kernel_execve(const char *filename, + const char *const argv[], + const char *const envp[]) +{ + struct pt_regs *p = current_pt_regs(); + int ret; + + ret = do_execve(filename, + (const char __user *const __user *)argv, + (const char __user *const __user *)envp, p); + if (ret < 0) + return ret; + + /* + * We were successful. We won't be returning to our caller, but + * instead to user space by manipulating the kernel stack. + */ + ret_from_kernel_execve(p); +} +#endif diff --git a/include/linux/binfmts.h b/include/linux/binfmts.h index 8938beabad7a..f9c9d08f4f7c 100644 --- a/include/linux/binfmts.h +++ b/include/linux/binfmts.h @@ -19,6 +19,7 @@ struct pt_regs; #ifdef __KERNEL__ #include +#include #include #define CORENAME_MAX_SIZE 128 @@ -137,5 +138,9 @@ extern void do_coredump(long signr, int exit_code, struct pt_regs *regs); extern void set_binfmt(struct linux_binfmt *new); extern void free_bprm(struct linux_binprm *); +#ifdef __ARCH_WANT_KERNEL_EXECVE +extern void ret_from_kernel_execve(struct pt_regs *normal) __noreturn; +#endif + #endif /* __KERNEL__ */ #endif /* _LINUX_BINFMTS_H */ -- cgit v1.2.3 From 38b983b3461e7d3c64a981e2338bb34605c46f30 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 30 Sep 2012 13:38:55 -0400 Subject: generic sys_execve() Selected by __ARCH_WANT_SYS_EXECVE in unistd.h. Requires * working current_pt_regs() * *NOT* doing a syscall-in-kernel kind of kernel_execve() implementation. Using generic kernel_execve() is fine. Signed-off-by: Al Viro --- fs/exec.c | 38 ++++++++++++++++++++++++++++++++++---- include/linux/compat.h | 8 ++++++-- 2 files changed, 40 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/fs/exec.c b/fs/exec.c index 48c525115fe4..50a1270da95b 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -401,7 +401,7 @@ struct user_arg_ptr { union { const char __user *const __user *native; #ifdef CONFIG_COMPAT - compat_uptr_t __user *compat; + const compat_uptr_t __user *compat; #endif } ptr; }; @@ -1600,9 +1600,9 @@ int do_execve(const char *filename, } #ifdef CONFIG_COMPAT -int compat_do_execve(char *filename, - compat_uptr_t __user *__argv, - compat_uptr_t __user *__envp, +int compat_do_execve(const char *filename, + const compat_uptr_t __user *__argv, + const compat_uptr_t __user *__envp, struct pt_regs *regs) { struct user_arg_ptr argv = { @@ -2319,6 +2319,36 @@ int dump_seek(struct file *file, loff_t off) } EXPORT_SYMBOL(dump_seek); +#ifdef __ARCH_WANT_SYS_EXECVE +SYSCALL_DEFINE3(execve, + const char __user *, filename, + const char __user *const __user *, argv, + const char __user *const __user *, envp) +{ + const char *path = getname(filename); + int error = PTR_ERR(path); + if (!IS_ERR(path)) { + error = do_execve(path, argv, envp, current_pt_regs()); + putname(path); + } + return error; +} +#ifdef CONFIG_COMPAT +asmlinkage long compat_sys_execve(const char __user * filename, + const compat_uptr_t __user * argv, + const compat_uptr_t __user * envp) +{ + const char *path = getname(filename); + int error = PTR_ERR(path); + if (!IS_ERR(path)) { + error = compat_do_execve(path, argv, envp, current_pt_regs()); + putname(path); + } + return error; +} +#endif +#endif + #ifdef __ARCH_WANT_KERNEL_EXECVE int kernel_execve(const char *filename, const char *const argv[], diff --git a/include/linux/compat.h b/include/linux/compat.h index 09b28b7369d7..c4be3f55511d 100644 --- a/include/linux/compat.h +++ b/include/linux/compat.h @@ -289,8 +289,12 @@ asmlinkage ssize_t compat_sys_pwritev(unsigned long fd, const struct compat_iovec __user *vec, unsigned long vlen, u32 pos_low, u32 pos_high); -int compat_do_execve(char *filename, compat_uptr_t __user *argv, - compat_uptr_t __user *envp, struct pt_regs *regs); +int compat_do_execve(const char *filename, const compat_uptr_t __user *argv, + const compat_uptr_t __user *envp, struct pt_regs *regs); +#ifdef __ARCH_WANT_SYS_EXECVE +asmlinkage long compat_sys_execve(const char __user *filename, const compat_uptr_t __user *argv, + const compat_uptr_t __user *envp); +#endif asmlinkage long compat_sys_select(int n, compat_ulong_t __user *inp, compat_ulong_t __user *outp, compat_ulong_t __user *exp, -- cgit v1.2.3