mirror of
https://github.com/AuxXxilium/linux_dsm_epyc7002.git
synced 2024-12-22 19:24:27 +07:00
0b38b5e1d0
When userspace executes a syscall or gets interrupted,
BEAR contains a kernel address when returning to userspace.
This make it pretty easy to figure out where the kernel is
mapped even with KASLR enabled. To fix this, add lpswe to
lowcore and always execute it there, so userspace sees only
the lowcore address of lpswe. For this we have to extend
both critical_cleanup and the SWITCH_ASYNC macro to also check
for lpswe addresses in lowcore.
Fixes: b2d24b97b2
("s390/kernel: add support for kernel address space layout randomization (KASLR)")
Cc: <stable@vger.kernel.org> # v5.2+
Reviewed-by: Gerald Schaefer <gerald.schaefer@de.ibm.com>
Signed-off-by: Sven Schnelle <svens@linux.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
242 lines
6.0 KiB
C
242 lines
6.0 KiB
C
// SPDX-License-Identifier: GPL-2.0
|
|
/*
|
|
* This file handles the architecture dependent parts of process handling.
|
|
*
|
|
* Copyright IBM Corp. 1999, 2009
|
|
* Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com>,
|
|
* Hartmut Penner <hp@de.ibm.com>,
|
|
* Denis Joseph Barrow,
|
|
*/
|
|
|
|
#include <linux/elf-randomize.h>
|
|
#include <linux/compiler.h>
|
|
#include <linux/cpu.h>
|
|
#include <linux/sched.h>
|
|
#include <linux/sched/debug.h>
|
|
#include <linux/sched/task.h>
|
|
#include <linux/sched/task_stack.h>
|
|
#include <linux/kernel.h>
|
|
#include <linux/mm.h>
|
|
#include <linux/elfcore.h>
|
|
#include <linux/smp.h>
|
|
#include <linux/slab.h>
|
|
#include <linux/interrupt.h>
|
|
#include <linux/tick.h>
|
|
#include <linux/personality.h>
|
|
#include <linux/syscalls.h>
|
|
#include <linux/compat.h>
|
|
#include <linux/kprobes.h>
|
|
#include <linux/random.h>
|
|
#include <linux/export.h>
|
|
#include <linux/init_task.h>
|
|
#include <asm/cpu_mf.h>
|
|
#include <asm/io.h>
|
|
#include <asm/processor.h>
|
|
#include <asm/vtimer.h>
|
|
#include <asm/exec.h>
|
|
#include <asm/irq.h>
|
|
#include <asm/nmi.h>
|
|
#include <asm/smp.h>
|
|
#include <asm/stacktrace.h>
|
|
#include <asm/switch_to.h>
|
|
#include <asm/runtime_instr.h>
|
|
#include <asm/unwind.h>
|
|
#include "entry.h"
|
|
|
|
asmlinkage void ret_from_fork(void) asm ("ret_from_fork");
|
|
|
|
extern void kernel_thread_starter(void);
|
|
|
|
void flush_thread(void)
|
|
{
|
|
}
|
|
|
|
void arch_setup_new_exec(void)
|
|
{
|
|
if (S390_lowcore.current_pid != current->pid) {
|
|
S390_lowcore.current_pid = current->pid;
|
|
if (test_facility(40))
|
|
lpp(&S390_lowcore.lpp);
|
|
}
|
|
}
|
|
|
|
void arch_release_task_struct(struct task_struct *tsk)
|
|
{
|
|
runtime_instr_release(tsk);
|
|
guarded_storage_release(tsk);
|
|
}
|
|
|
|
int arch_dup_task_struct(struct task_struct *dst, struct task_struct *src)
|
|
{
|
|
/*
|
|
* Save the floating-point or vector register state of the current
|
|
* task and set the CIF_FPU flag to lazy restore the FPU register
|
|
* state when returning to user space.
|
|
*/
|
|
save_fpu_regs();
|
|
|
|
memcpy(dst, src, arch_task_struct_size);
|
|
dst->thread.fpu.regs = dst->thread.fpu.fprs;
|
|
return 0;
|
|
}
|
|
|
|
int copy_thread_tls(unsigned long clone_flags, unsigned long new_stackp,
|
|
unsigned long arg, struct task_struct *p, unsigned long tls)
|
|
{
|
|
struct fake_frame
|
|
{
|
|
struct stack_frame sf;
|
|
struct pt_regs childregs;
|
|
} *frame;
|
|
|
|
frame = container_of(task_pt_regs(p), struct fake_frame, childregs);
|
|
p->thread.ksp = (unsigned long) frame;
|
|
/* Save access registers to new thread structure. */
|
|
save_access_regs(&p->thread.acrs[0]);
|
|
/* start new process with ar4 pointing to the correct address space */
|
|
p->thread.mm_segment = get_fs();
|
|
/* Don't copy debug registers */
|
|
memset(&p->thread.per_user, 0, sizeof(p->thread.per_user));
|
|
memset(&p->thread.per_event, 0, sizeof(p->thread.per_event));
|
|
clear_tsk_thread_flag(p, TIF_SINGLE_STEP);
|
|
p->thread.per_flags = 0;
|
|
/* Initialize per thread user and system timer values */
|
|
p->thread.user_timer = 0;
|
|
p->thread.guest_timer = 0;
|
|
p->thread.system_timer = 0;
|
|
p->thread.hardirq_timer = 0;
|
|
p->thread.softirq_timer = 0;
|
|
p->thread.last_break = 1;
|
|
|
|
frame->sf.back_chain = 0;
|
|
/* new return point is ret_from_fork */
|
|
frame->sf.gprs[8] = (unsigned long) ret_from_fork;
|
|
/* fake return stack for resume(), don't go back to schedule */
|
|
frame->sf.gprs[9] = (unsigned long) frame;
|
|
|
|
/* Store access registers to kernel stack of new process. */
|
|
if (unlikely(p->flags & PF_KTHREAD)) {
|
|
/* kernel thread */
|
|
memset(&frame->childregs, 0, sizeof(struct pt_regs));
|
|
frame->childregs.psw.mask = PSW_KERNEL_BITS | PSW_MASK_DAT |
|
|
PSW_MASK_IO | PSW_MASK_EXT | PSW_MASK_MCHECK;
|
|
frame->childregs.psw.addr =
|
|
(unsigned long) kernel_thread_starter;
|
|
frame->childregs.gprs[9] = new_stackp; /* function */
|
|
frame->childregs.gprs[10] = arg;
|
|
frame->childregs.gprs[11] = (unsigned long) do_exit;
|
|
frame->childregs.orig_gpr2 = -1;
|
|
|
|
return 0;
|
|
}
|
|
frame->childregs = *current_pt_regs();
|
|
frame->childregs.gprs[2] = 0; /* child returns 0 on fork. */
|
|
frame->childregs.flags = 0;
|
|
if (new_stackp)
|
|
frame->childregs.gprs[15] = new_stackp;
|
|
|
|
/* Don't copy runtime instrumentation info */
|
|
p->thread.ri_cb = NULL;
|
|
frame->childregs.psw.mask &= ~PSW_MASK_RI;
|
|
/* Don't copy guarded storage control block */
|
|
p->thread.gs_cb = NULL;
|
|
p->thread.gs_bc_cb = NULL;
|
|
|
|
/* Set a new TLS ? */
|
|
if (clone_flags & CLONE_SETTLS) {
|
|
if (is_compat_task()) {
|
|
p->thread.acrs[0] = (unsigned int)tls;
|
|
} else {
|
|
p->thread.acrs[0] = (unsigned int)(tls >> 32);
|
|
p->thread.acrs[1] = (unsigned int)tls;
|
|
}
|
|
}
|
|
return 0;
|
|
}
|
|
|
|
asmlinkage void execve_tail(void)
|
|
{
|
|
current->thread.fpu.fpc = 0;
|
|
asm volatile("sfpc %0" : : "d" (0));
|
|
}
|
|
|
|
/*
|
|
* fill in the FPU structure for a core dump.
|
|
*/
|
|
int dump_fpu (struct pt_regs * regs, s390_fp_regs *fpregs)
|
|
{
|
|
save_fpu_regs();
|
|
fpregs->fpc = current->thread.fpu.fpc;
|
|
fpregs->pad = 0;
|
|
if (MACHINE_HAS_VX)
|
|
convert_vx_to_fp((freg_t *)&fpregs->fprs,
|
|
current->thread.fpu.vxrs);
|
|
else
|
|
memcpy(&fpregs->fprs, current->thread.fpu.fprs,
|
|
sizeof(fpregs->fprs));
|
|
return 1;
|
|
}
|
|
EXPORT_SYMBOL(dump_fpu);
|
|
|
|
unsigned long get_wchan(struct task_struct *p)
|
|
{
|
|
struct unwind_state state;
|
|
unsigned long ip = 0;
|
|
|
|
if (!p || p == current || p->state == TASK_RUNNING || !task_stack_page(p))
|
|
return 0;
|
|
|
|
if (!try_get_task_stack(p))
|
|
return 0;
|
|
|
|
unwind_for_each_frame(&state, p, NULL, 0) {
|
|
if (state.stack_info.type != STACK_TYPE_TASK) {
|
|
ip = 0;
|
|
break;
|
|
}
|
|
|
|
ip = unwind_get_return_address(&state);
|
|
if (!ip)
|
|
break;
|
|
|
|
if (!in_sched_functions(ip))
|
|
break;
|
|
}
|
|
|
|
put_task_stack(p);
|
|
return ip;
|
|
}
|
|
|
|
unsigned long arch_align_stack(unsigned long sp)
|
|
{
|
|
if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space)
|
|
sp -= get_random_int() & ~PAGE_MASK;
|
|
return sp & ~0xf;
|
|
}
|
|
|
|
static inline unsigned long brk_rnd(void)
|
|
{
|
|
return (get_random_int() & BRK_RND_MASK) << PAGE_SHIFT;
|
|
}
|
|
|
|
unsigned long arch_randomize_brk(struct mm_struct *mm)
|
|
{
|
|
unsigned long ret;
|
|
|
|
ret = PAGE_ALIGN(mm->brk + brk_rnd());
|
|
return (ret > mm->brk) ? ret : mm->brk;
|
|
}
|
|
|
|
void set_fs_fixup(void)
|
|
{
|
|
struct pt_regs *regs = current_pt_regs();
|
|
static bool warned;
|
|
|
|
set_fs(USER_DS);
|
|
if (warned)
|
|
return;
|
|
WARN(1, "Unbalanced set_fs - int code: 0x%x\n", regs->int_code);
|
|
show_registers(regs);
|
|
warned = true;
|
|
}
|