mirror of
https://github.com/xemu-project/xemu.git
synced 2024-11-25 20:49:49 +00:00
097a50d0d8
The comment I put in mmap-alloc to document the ppc64 rules refers to the previous revision of the patch: we don't look at memory alignment anymore, we check the fs from which the fd is mapped, instead. It's also not clear what does "in this case" refer to, rearrange text to make it clearer. Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
112 lines
3.0 KiB
C
112 lines
3.0 KiB
C
/*
|
|
* Support for RAM backed by mmaped host memory.
|
|
*
|
|
* Copyright (c) 2015 Red Hat, Inc.
|
|
*
|
|
* Authors:
|
|
* Michael S. Tsirkin <mst@redhat.com>
|
|
*
|
|
* This work is licensed under the terms of the GNU GPL, version 2 or
|
|
* later. See the COPYING file in the top-level directory.
|
|
*/
|
|
#include <qemu/mmap-alloc.h>
|
|
#include <sys/types.h>
|
|
#include <sys/mman.h>
|
|
#include <assert.h>
|
|
|
|
#define HUGETLBFS_MAGIC 0x958458f6
|
|
|
|
#ifdef CONFIG_LINUX
|
|
#include <sys/vfs.h>
|
|
#endif
|
|
|
|
size_t qemu_fd_getpagesize(int fd)
|
|
{
|
|
#ifdef CONFIG_LINUX
|
|
struct statfs fs;
|
|
int ret;
|
|
|
|
if (fd != -1) {
|
|
do {
|
|
ret = fstatfs(fd, &fs);
|
|
} while (ret != 0 && errno == EINTR);
|
|
|
|
if (ret == 0 && fs.f_type == HUGETLBFS_MAGIC) {
|
|
return fs.f_bsize;
|
|
}
|
|
}
|
|
#endif
|
|
|
|
return getpagesize();
|
|
}
|
|
|
|
void *qemu_ram_mmap(int fd, size_t size, size_t align, bool shared)
|
|
{
|
|
/*
|
|
* Note: this always allocates at least one extra page of virtual address
|
|
* space, even if size is already aligned.
|
|
*/
|
|
size_t total = size + align;
|
|
#if defined(__powerpc64__) && defined(__linux__)
|
|
/* On ppc64 mappings in the same segment (aka slice) must share the same
|
|
* page size. Since we will be re-allocating part of this segment
|
|
* from the supplied fd, we should make sure to use the same page size, to
|
|
* this end we mmap the supplied fd. In this case, set MAP_NORESERVE to
|
|
* avoid allocating backing store memory.
|
|
* We do this unless we are using the system page size, in which case
|
|
* anonymous memory is OK.
|
|
*/
|
|
int anonfd = fd == -1 || qemu_fd_getpagesize(fd) == getpagesize() ? -1 : fd;
|
|
int flags = anonfd == -1 ? MAP_ANONYMOUS : MAP_NORESERVE;
|
|
void *ptr = mmap(0, total, PROT_NONE, flags | MAP_PRIVATE, anonfd, 0);
|
|
#else
|
|
void *ptr = mmap(0, total, PROT_NONE, MAP_ANONYMOUS | MAP_PRIVATE, -1, 0);
|
|
#endif
|
|
size_t offset = QEMU_ALIGN_UP((uintptr_t)ptr, align) - (uintptr_t)ptr;
|
|
void *ptr1;
|
|
|
|
if (ptr == MAP_FAILED) {
|
|
return MAP_FAILED;
|
|
}
|
|
|
|
/* Make sure align is a power of 2 */
|
|
assert(!(align & (align - 1)));
|
|
/* Always align to host page size */
|
|
assert(align >= getpagesize());
|
|
|
|
ptr1 = mmap(ptr + offset, size, PROT_READ | PROT_WRITE,
|
|
MAP_FIXED |
|
|
(fd == -1 ? MAP_ANONYMOUS : 0) |
|
|
(shared ? MAP_SHARED : MAP_PRIVATE),
|
|
fd, 0);
|
|
if (ptr1 == MAP_FAILED) {
|
|
munmap(ptr, total);
|
|
return MAP_FAILED;
|
|
}
|
|
|
|
ptr += offset;
|
|
total -= offset;
|
|
|
|
if (offset > 0) {
|
|
munmap(ptr - offset, offset);
|
|
}
|
|
|
|
/*
|
|
* Leave a single PROT_NONE page allocated after the RAM block, to serve as
|
|
* a guard page guarding against potential buffer overflows.
|
|
*/
|
|
if (total > size + getpagesize()) {
|
|
munmap(ptr + size + getpagesize(), total - size - getpagesize());
|
|
}
|
|
|
|
return ptr;
|
|
}
|
|
|
|
void qemu_ram_munmap(void *ptr, size_t size)
|
|
{
|
|
if (ptr) {
|
|
/* Unmap both the RAM block and the guard page */
|
|
munmap(ptr, size + getpagesize());
|
|
}
|
|
}
|