mirror of
https://github.com/AuxXxilium/linux_dsm_epyc7002.git
synced 2024-12-21 11:09:29 +07:00
ce3dc44749
With virtually mapped kernel stacks the kernel stack overflow detection is now fault based, every stack has a guard page in the vmalloc space. The panic_stack is renamed to nodat_stack and is used for all function that need to run without DAT, e.g. memcpy_real or do_start_kdump. The main effect is a reduction in the kernel image size as with vmap stacks the old style overflow checking that adds two instructions per function is not needed anymore. Result from bloat-o-meter: add/remove: 20/1 grow/shrink: 13/26854 up/down: 2198/-216240 (-214042) In regard to performance the micro-benchmark for fork has a hit of a few microseconds, allocating 4 pages in vmalloc space is more expensive compare to an order-2 page allocation. But with real workload I could not find a noticeable difference. Acked-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
229 lines
5.5 KiB
C
229 lines
5.5 KiB
C
// SPDX-License-Identifier: GPL-2.0
|
|
/*
|
|
* Access kernel memory without faulting -- s390 specific implementation.
|
|
*
|
|
* Copyright IBM Corp. 2009, 2015
|
|
*
|
|
* Author(s): Heiko Carstens <heiko.carstens@de.ibm.com>,
|
|
*
|
|
*/
|
|
|
|
#include <linux/uaccess.h>
|
|
#include <linux/kernel.h>
|
|
#include <linux/types.h>
|
|
#include <linux/errno.h>
|
|
#include <linux/gfp.h>
|
|
#include <linux/cpu.h>
|
|
#include <asm/ctl_reg.h>
|
|
#include <asm/io.h>
|
|
|
|
static notrace long s390_kernel_write_odd(void *dst, const void *src, size_t size)
|
|
{
|
|
unsigned long aligned, offset, count;
|
|
char tmp[8];
|
|
|
|
aligned = (unsigned long) dst & ~7UL;
|
|
offset = (unsigned long) dst & 7UL;
|
|
size = min(8UL - offset, size);
|
|
count = size - 1;
|
|
asm volatile(
|
|
" bras 1,0f\n"
|
|
" mvc 0(1,%4),0(%5)\n"
|
|
"0: mvc 0(8,%3),0(%0)\n"
|
|
" ex %1,0(1)\n"
|
|
" lg %1,0(%3)\n"
|
|
" lra %0,0(%0)\n"
|
|
" sturg %1,%0\n"
|
|
: "+&a" (aligned), "+&a" (count), "=m" (tmp)
|
|
: "a" (&tmp), "a" (&tmp[offset]), "a" (src)
|
|
: "cc", "memory", "1");
|
|
return size;
|
|
}
|
|
|
|
/*
|
|
* s390_kernel_write - write to kernel memory bypassing DAT
|
|
* @dst: destination address
|
|
* @src: source address
|
|
* @size: number of bytes to copy
|
|
*
|
|
* This function writes to kernel memory bypassing DAT and possible page table
|
|
* write protection. It writes to the destination using the sturg instruction.
|
|
* Therefore we have a read-modify-write sequence: the function reads eight
|
|
* bytes from destination at an eight byte boundary, modifies the bytes
|
|
* requested and writes the result back in a loop.
|
|
*
|
|
* Note: this means that this function may not be called concurrently on
|
|
* several cpus with overlapping words, since this may potentially
|
|
* cause data corruption.
|
|
*/
|
|
void notrace s390_kernel_write(void *dst, const void *src, size_t size)
|
|
{
|
|
long copied;
|
|
|
|
while (size) {
|
|
copied = s390_kernel_write_odd(dst, src, size);
|
|
dst += copied;
|
|
src += copied;
|
|
size -= copied;
|
|
}
|
|
}
|
|
|
|
static int __memcpy_real(void *dest, void *src, size_t count)
|
|
{
|
|
register unsigned long _dest asm("2") = (unsigned long) dest;
|
|
register unsigned long _len1 asm("3") = (unsigned long) count;
|
|
register unsigned long _src asm("4") = (unsigned long) src;
|
|
register unsigned long _len2 asm("5") = (unsigned long) count;
|
|
int rc = -EFAULT;
|
|
|
|
asm volatile (
|
|
"0: mvcle %1,%2,0x0\n"
|
|
"1: jo 0b\n"
|
|
" lhi %0,0x0\n"
|
|
"2:\n"
|
|
EX_TABLE(1b,2b)
|
|
: "+d" (rc), "+d" (_dest), "+d" (_src), "+d" (_len1),
|
|
"+d" (_len2), "=m" (*((long *) dest))
|
|
: "m" (*((long *) src))
|
|
: "cc", "memory");
|
|
return rc;
|
|
}
|
|
|
|
static unsigned long _memcpy_real(unsigned long dest, unsigned long src,
|
|
unsigned long count)
|
|
{
|
|
int irqs_disabled, rc;
|
|
unsigned long flags;
|
|
|
|
if (!count)
|
|
return 0;
|
|
flags = __arch_local_irq_stnsm(0xf8UL);
|
|
irqs_disabled = arch_irqs_disabled_flags(flags);
|
|
if (!irqs_disabled)
|
|
trace_hardirqs_off();
|
|
rc = __memcpy_real((void *) dest, (void *) src, (size_t) count);
|
|
if (!irqs_disabled)
|
|
trace_hardirqs_on();
|
|
__arch_local_irq_ssm(flags);
|
|
return rc;
|
|
}
|
|
|
|
/*
|
|
* Copy memory in real mode (kernel to kernel)
|
|
*/
|
|
int memcpy_real(void *dest, void *src, size_t count)
|
|
{
|
|
if (S390_lowcore.nodat_stack != 0)
|
|
return CALL_ON_STACK(_memcpy_real, S390_lowcore.nodat_stack,
|
|
3, dest, src, count);
|
|
/*
|
|
* This is a really early memcpy_real call, the stacks are
|
|
* not set up yet. Just call _memcpy_real on the early boot
|
|
* stack
|
|
*/
|
|
return _memcpy_real((unsigned long) dest,(unsigned long) src,
|
|
(unsigned long) count);
|
|
}
|
|
|
|
/*
|
|
* Copy memory in absolute mode (kernel to kernel)
|
|
*/
|
|
void memcpy_absolute(void *dest, void *src, size_t count)
|
|
{
|
|
unsigned long cr0, flags, prefix;
|
|
|
|
flags = arch_local_irq_save();
|
|
__ctl_store(cr0, 0, 0);
|
|
__ctl_clear_bit(0, 28); /* disable lowcore protection */
|
|
prefix = store_prefix();
|
|
if (prefix) {
|
|
local_mcck_disable();
|
|
set_prefix(0);
|
|
memcpy(dest, src, count);
|
|
set_prefix(prefix);
|
|
local_mcck_enable();
|
|
} else {
|
|
memcpy(dest, src, count);
|
|
}
|
|
__ctl_load(cr0, 0, 0);
|
|
arch_local_irq_restore(flags);
|
|
}
|
|
|
|
/*
|
|
* Copy memory from kernel (real) to user (virtual)
|
|
*/
|
|
int copy_to_user_real(void __user *dest, void *src, unsigned long count)
|
|
{
|
|
int offs = 0, size, rc;
|
|
char *buf;
|
|
|
|
buf = (char *) __get_free_page(GFP_KERNEL);
|
|
if (!buf)
|
|
return -ENOMEM;
|
|
rc = -EFAULT;
|
|
while (offs < count) {
|
|
size = min(PAGE_SIZE, count - offs);
|
|
if (memcpy_real(buf, src + offs, size))
|
|
goto out;
|
|
if (copy_to_user(dest + offs, buf, size))
|
|
goto out;
|
|
offs += size;
|
|
}
|
|
rc = 0;
|
|
out:
|
|
free_page((unsigned long) buf);
|
|
return rc;
|
|
}
|
|
|
|
/*
|
|
* Check if physical address is within prefix or zero page
|
|
*/
|
|
static int is_swapped(unsigned long addr)
|
|
{
|
|
unsigned long lc;
|
|
int cpu;
|
|
|
|
if (addr < sizeof(struct lowcore))
|
|
return 1;
|
|
for_each_online_cpu(cpu) {
|
|
lc = (unsigned long) lowcore_ptr[cpu];
|
|
if (addr > lc + sizeof(struct lowcore) - 1 || addr < lc)
|
|
continue;
|
|
return 1;
|
|
}
|
|
return 0;
|
|
}
|
|
|
|
/*
|
|
* Convert a physical pointer for /dev/mem access
|
|
*
|
|
* For swapped prefix pages a new buffer is returned that contains a copy of
|
|
* the absolute memory. The buffer size is maximum one page large.
|
|
*/
|
|
void *xlate_dev_mem_ptr(phys_addr_t addr)
|
|
{
|
|
void *bounce = (void *) addr;
|
|
unsigned long size;
|
|
|
|
get_online_cpus();
|
|
preempt_disable();
|
|
if (is_swapped(addr)) {
|
|
size = PAGE_SIZE - (addr & ~PAGE_MASK);
|
|
bounce = (void *) __get_free_page(GFP_ATOMIC);
|
|
if (bounce)
|
|
memcpy_absolute(bounce, (void *) addr, size);
|
|
}
|
|
preempt_enable();
|
|
put_online_cpus();
|
|
return bounce;
|
|
}
|
|
|
|
/*
|
|
* Free converted buffer for /dev/mem access (if necessary)
|
|
*/
|
|
void unxlate_dev_mem_ptr(phys_addr_t addr, void *buf)
|
|
{
|
|
if ((void *) addr != buf)
|
|
free_page((unsigned long) buf);
|
|
}
|