* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ * along with this program; if not, see <http://www.gnu.org/licenses/>.
*/
-#include <stdlib.h>
-#include <stdio.h>
-#include <stdarg.h>
-#include <string.h>
-#include <unistd.h>
-#include <errno.h>
+#include "qemu/osdep.h"
#include <sys/mman.h>
+#include <linux/mman.h>
+#include <linux/unistd.h>
#include "qemu.h"
+#include "qemu-common.h"
+#include "translate-all.h"
//#define DEBUG_MMAP
-#if defined(USE_NPTL)
-pthread_mutex_t mmap_mutex;
-static int __thread mmap_lock_count;
+static pthread_mutex_t mmap_mutex = PTHREAD_MUTEX_INITIALIZER;
+static __thread int mmap_lock_count;
void mmap_lock(void)
{
else
pthread_mutex_unlock(&mmap_mutex);
}
-#else
-/* We aren't threadsafe to start with, so no need to worry about locking. */
-void mmap_lock(void)
-{
-}
-
-void mmap_unlock(void)
-{
-}
-#endif
-
-void *qemu_vmalloc(size_t size)
-{
- void *p;
- unsigned long addr;
- mmap_lock();
- /* Use map and mark the pages as used. */
- p = mmap(NULL, size, PROT_READ | PROT_WRITE,
- MAP_PRIVATE | MAP_ANONYMOUS, -1, 0);
-
- addr = (unsigned long)p;
- if (addr == (target_ulong) addr) {
- /* Allocated region overlaps guest address space.
- This may recurse. */
- page_set_flags(addr & TARGET_PAGE_MASK, TARGET_PAGE_ALIGN(addr + size),
- PAGE_RESERVED);
- }
-
- mmap_unlock();
- return p;
-}
-
-void *qemu_malloc(size_t size)
-{
- char * p;
- size += 16;
- p = qemu_vmalloc(size);
- *(size_t *)p = size;
- return p + 16;
-}
-
-/* We use map, which is always zero initialized. */
-void * qemu_mallocz(size_t size)
-{
- return qemu_malloc(size);
-}
-
-void qemu_free(void *ptr)
-{
- /* FIXME: We should unmark the reserved pages here. However this gets
- complicated when one target page spans multiple host pages, so we
- don't bother. */
- size_t *p;
- p = (size_t *)((char *)ptr - 16);
- munmap(p, *p);
-}
/* NOTE: all the constants are the HOST ones, but addresses are target. */
int target_mprotect(abi_ulong start, abi_ulong len, int prot)
int prot1, ret;
#ifdef DEBUG_MMAP
- printf("mprotect: start=0x" TARGET_FMT_lx
- "len=0x" TARGET_FMT_lx " prot=%c%c%c\n", start, len,
+ printf("mprotect: start=0x" TARGET_ABI_FMT_lx
+ "len=0x" TARGET_ABI_FMT_lx " prot=%c%c%c\n", start, len,
prot & PROT_READ ? 'r' : '-',
prot & PROT_WRITE ? 'w' : '-',
prot & PROT_EXEC ? 'x' : '-');
possible while it is a shared mapping */
if ((flags & MAP_TYPE) == MAP_SHARED &&
(prot & PROT_WRITE))
- return -EINVAL;
+ return -1;
/* adjust protection to be able to read */
if (!(prot1 & PROT_WRITE))
mprotect(host_start, qemu_host_page_size, prot1 | PROT_WRITE);
/* read the corresponding file data */
- pread(fd, g2h(start), end - start, offset);
+ if (pread(fd, g2h(start), end - start, offset) == -1)
+ return -1;
/* put final protection */
if (prot_new != (prot1 | PROT_WRITE))
mprotect(host_start, qemu_host_page_size, prot_new);
} else {
- /* just update the protection */
if (prot_new != prot1) {
mprotect(host_start, qemu_host_page_size, prot_new);
}
+ if (prot_new & PROT_WRITE) {
+ memset(g2h(start), 0, end - start);
+ }
}
return 0;
}
-#if defined(__CYGWIN__)
+#if HOST_LONG_BITS == 64 && TARGET_ABI_BITS == 64
+# define TASK_UNMAPPED_BASE (1ul << 38)
+#elif defined(__CYGWIN__)
/* Cygwin doesn't have a whole lot of address space. */
-static abi_ulong mmap_next_start = 0x18000000;
+# define TASK_UNMAPPED_BASE 0x18000000
#else
-static abi_ulong mmap_next_start = 0x40000000;
+# define TASK_UNMAPPED_BASE 0x40000000
#endif
+abi_ulong mmap_next_start = TASK_UNMAPPED_BASE;
unsigned long last_brk;
-/* find a free memory area of size 'size'. The search starts at
- 'start'. If 'start' == 0, then a default start address is used.
- Return -1 if error.
-*/
-/* page_init() marks pages used by the host as reserved to be sure not
- to use them. */
-static abi_ulong mmap_find_vma(abi_ulong start, abi_ulong size)
+/* Subroutine of mmap_find_vma, used when we have pre-allocated a chunk
+ of guest address space. */
+static abi_ulong mmap_find_vma_reserved(abi_ulong start, abi_ulong size)
{
- abi_ulong addr, addr1, addr_start;
+ abi_ulong addr;
+ abi_ulong end_addr;
int prot;
- unsigned long new_brk;
-
- new_brk = (unsigned long)sbrk(0);
- if (last_brk && last_brk < new_brk && last_brk == (target_ulong)last_brk) {
- /* This is a hack to catch the host allocating memory with brk().
- If it uses mmap then we loose.
- FIXME: We really want to avoid the host allocating memory in
- the first place, and maybe leave some slack to avoid switching
- to mmap. */
- page_set_flags(last_brk & TARGET_PAGE_MASK,
- TARGET_PAGE_ALIGN(new_brk),
- PAGE_RESERVED);
+ int looped = 0;
+
+ if (size > reserved_va) {
+ return (abi_ulong)-1;
}
- last_brk = new_brk;
size = HOST_PAGE_ALIGN(size);
- start = start & qemu_host_page_mask;
- addr = start;
- if (addr == 0)
- addr = mmap_next_start;
- addr_start = addr;
- for(;;) {
- prot = 0;
- for(addr1 = addr; addr1 < (addr + size); addr1 += TARGET_PAGE_SIZE) {
- prot |= page_get_flags(addr1);
+ end_addr = start + size;
+ if (end_addr > reserved_va) {
+ end_addr = reserved_va;
+ }
+ addr = end_addr - qemu_host_page_size;
+
+ while (1) {
+ if (addr > end_addr) {
+ if (looped) {
+ return (abi_ulong)-1;
+ }
+ end_addr = reserved_va;
+ addr = end_addr - qemu_host_page_size;
+ looped = 1;
+ continue;
+ }
+ prot = page_get_flags(addr);
+ if (prot) {
+ end_addr = addr;
}
- if (prot == 0)
+ if (addr + size == end_addr) {
break;
- addr += qemu_host_page_size;
- /* we found nothing */
- if (addr == addr_start)
- return (abi_ulong)-1;
+ }
+ addr -= qemu_host_page_size;
}
- if (start == 0)
- mmap_next_start = addr + size;
+
+ if (start == mmap_next_start) {
+ mmap_next_start = addr;
+ }
+
return addr;
}
+/*
+ * Find and reserve a free memory area of size 'size'. The search
+ * starts at 'start'.
+ * It must be called with mmap_lock() held.
+ * Return -1 if error.
+ */
+abi_ulong mmap_find_vma(abi_ulong start, abi_ulong size)
+{
+ void *ptr, *prev;
+ abi_ulong addr;
+ int wrapped, repeat;
+
+ /* If 'start' == 0, then a default start address is used. */
+ if (start == 0) {
+ start = mmap_next_start;
+ } else {
+ start &= qemu_host_page_mask;
+ }
+
+ size = HOST_PAGE_ALIGN(size);
+
+ if (reserved_va) {
+ return mmap_find_vma_reserved(start, size);
+ }
+
+ addr = start;
+ wrapped = repeat = 0;
+ prev = 0;
+
+ for (;; prev = ptr) {
+ /*
+ * Reserve needed memory area to avoid a race.
+ * It should be discarded using:
+ * - mmap() with MAP_FIXED flag
+ * - mremap() with MREMAP_FIXED flag
+ * - shmat() with SHM_REMAP flag
+ */
+ ptr = mmap(g2h(addr), size, PROT_NONE,
+ MAP_ANONYMOUS|MAP_PRIVATE|MAP_NORESERVE, -1, 0);
+
+ /* ENOMEM, if host address space has no memory */
+ if (ptr == MAP_FAILED) {
+ return (abi_ulong)-1;
+ }
+
+ /* Count the number of sequential returns of the same address.
+ This is used to modify the search algorithm below. */
+ repeat = (ptr == prev ? repeat + 1 : 0);
+
+ if (h2g_valid(ptr + size - 1)) {
+ addr = h2g(ptr);
+
+ if ((addr & ~TARGET_PAGE_MASK) == 0) {
+ /* Success. */
+ if (start == mmap_next_start && addr >= TASK_UNMAPPED_BASE) {
+ mmap_next_start = addr + size;
+ }
+ return addr;
+ }
+
+ /* The address is not properly aligned for the target. */
+ switch (repeat) {
+ case 0:
+ /* Assume the result that the kernel gave us is the
+ first with enough free space, so start again at the
+ next higher target page. */
+ addr = TARGET_PAGE_ALIGN(addr);
+ break;
+ case 1:
+ /* Sometimes the kernel decides to perform the allocation
+ at the top end of memory instead. */
+ addr &= TARGET_PAGE_MASK;
+ break;
+ case 2:
+ /* Start over at low memory. */
+ addr = 0;
+ break;
+ default:
+ /* Fail. This unaligned block must the last. */
+ addr = -1;
+ break;
+ }
+ } else {
+ /* Since the result the kernel gave didn't fit, start
+ again at low memory. If any repetition, fail. */
+ addr = (repeat ? -1 : 0);
+ }
+
+ /* Unmap and try again. */
+ munmap(ptr, size);
+
+ /* ENOMEM if we checked the whole of the target address space. */
+ if (addr == (abi_ulong)-1) {
+ return (abi_ulong)-1;
+ } else if (addr == 0) {
+ if (wrapped) {
+ return (abi_ulong)-1;
+ }
+ wrapped = 1;
+ /* Don't actually use 0 when wrapping, instead indicate
+ that we'd truly like an allocation in low memory. */
+ addr = (mmap_min_addr > TARGET_PAGE_SIZE
+ ? TARGET_PAGE_ALIGN(mmap_min_addr)
+ : TARGET_PAGE_SIZE);
+ } else if (wrapped && addr >= start) {
+ return (abi_ulong)-1;
+ }
+ }
+}
+
/* NOTE: all the constants are the HOST ones */
abi_long target_mmap(abi_ulong start, abi_ulong len, int prot,
int flags, int fd, abi_ulong offset)
{
abi_ulong ret, end, real_start, real_end, retaddr, host_offset, host_len;
- unsigned long host_start;
mmap_lock();
#ifdef DEBUG_MMAP
{
- printf("mmap: start=0x" TARGET_FMT_lx
- " len=0x" TARGET_FMT_lx " prot=%c%c%c flags=",
+ printf("mmap: start=0x" TARGET_ABI_FMT_lx
+ " len=0x" TARGET_ABI_FMT_lx " prot=%c%c%c flags=",
start, len,
prot & PROT_READ ? 'r' : '-',
prot & PROT_WRITE ? 'w' : '-',
printf("[MAP_TYPE=0x%x] ", flags & MAP_TYPE);
break;
}
- printf("fd=%d offset=" TARGET_FMT_lx "\n", fd, offset);
+ printf("fd=%d offset=" TARGET_ABI_FMT_lx "\n", fd, offset);
}
#endif
if (len == 0)
goto the_end;
real_start = start & qemu_host_page_mask;
+ host_offset = offset & qemu_host_page_mask;
+ /* If the user is asking for the kernel to find a location, do that
+ before we truncate the length for mapping files below. */
if (!(flags & MAP_FIXED)) {
- abi_ulong mmap_start;
- void *p;
- host_offset = offset & qemu_host_page_mask;
host_len = len + offset - host_offset;
host_len = HOST_PAGE_ALIGN(host_len);
- mmap_start = mmap_find_vma(real_start, host_len);
- if (mmap_start == (abi_ulong)-1) {
+ start = mmap_find_vma(real_start, host_len);
+ if (start == (abi_ulong)-1) {
errno = ENOMEM;
goto fail;
}
+ }
+
+ /* When mapping files into a memory area larger than the file, accesses
+ to pages beyond the file size will cause a SIGBUS.
+
+ For example, if mmaping a file of 100 bytes on a host with 4K pages
+ emulating a target with 8K pages, the target expects to be able to
+ access the first 8K. But the host will trap us on any access beyond
+ 4K.
+
+ When emulating a target with a larger page-size than the hosts, we
+ may need to truncate file maps at EOF and add extra anonymous pages
+ up to the targets page boundary. */
+
+ if ((qemu_real_host_page_size < TARGET_PAGE_SIZE)
+ && !(flags & MAP_ANONYMOUS)) {
+ struct stat sb;
+
+ if (fstat (fd, &sb) == -1)
+ goto fail;
+
+ /* Are we trying to create a map beyond EOF?. */
+ if (offset + len > sb.st_size) {
+ /* If so, truncate the file map at eof aligned with
+ the hosts real pagesize. Additional anonymous maps
+ will be created beyond EOF. */
+ len = REAL_HOST_PAGE_ALIGN(sb.st_size - offset);
+ }
+ }
+
+ if (!(flags & MAP_FIXED)) {
+ unsigned long host_start;
+ void *p;
+
+ host_len = len + offset - host_offset;
+ host_len = HOST_PAGE_ALIGN(host_len);
+
/* Note: we prefer to control the mapping address. It is
especially important if qemu_host_page_size >
qemu_real_host_page_size */
- p = mmap(g2h(mmap_start),
- host_len, prot, flags | MAP_FIXED, fd, host_offset);
+ p = mmap(g2h(start), host_len, prot,
+ flags | MAP_FIXED | MAP_ANONYMOUS, -1, 0);
if (p == MAP_FAILED)
goto fail;
/* update start so that it points to the file position at 'offset' */
host_start = (unsigned long)p;
- if (!(flags & MAP_ANONYMOUS))
+ if (!(flags & MAP_ANONYMOUS)) {
+ p = mmap(g2h(start), len, prot,
+ flags | MAP_FIXED, fd, host_offset);
+ if (p == MAP_FAILED) {
+ munmap(g2h(start), host_len);
+ goto fail;
+ }
host_start += offset - host_offset;
+ }
start = h2g(host_start);
} else {
- int flg;
- target_ulong addr;
-
if (start & ~TARGET_PAGE_MASK) {
errno = EINVAL;
goto fail;
end = start + len;
real_end = HOST_PAGE_ALIGN(end);
- for(addr = real_start; addr < real_end; addr += TARGET_PAGE_SIZE) {
- flg = page_get_flags(addr);
- if (flg & PAGE_RESERVED) {
- errno = ENXIO;
- goto fail;
- }
+ /*
+ * Test if requested memory area fits target address space
+ * It can fail only on 64-bit host with 32-bit target.
+ * On any other target/host host mmap() handles this error correctly.
+ */
+ if ((unsigned long)start + len - 1 > (abi_ulong) -1) {
+ errno = EINVAL;
+ goto fail;
}
/* worst case: we cannot map the file because the offset is not
-1, 0);
if (retaddr == -1)
goto fail;
- pread(fd, g2h(start), len, offset);
+ if (pread(fd, g2h(start), len, offset) == -1)
+ goto fail;
if (!(prot & PROT_WRITE)) {
ret = target_mprotect(start, len, prot);
- if (ret != 0) {
- start = ret;
- goto the_end;
- }
+ assert(ret == 0);
}
goto the_end;
}
/* handle the end of the mapping */
if (end < real_end) {
ret = mmap_frag(real_end - qemu_host_page_size,
- real_end - qemu_host_page_size, real_end,
+ real_end - qemu_host_page_size, end,
prot, flags, fd,
offset + real_end - qemu_host_page_size - start);
if (ret == -1)
page_set_flags(start, start + len, prot | PAGE_VALID);
the_end:
#ifdef DEBUG_MMAP
- printf("ret=0x" TARGET_FMT_lx "\n", start);
+ printf("ret=0x" TARGET_ABI_FMT_lx "\n", start);
page_dump(stdout);
printf("\n");
#endif
+ tb_invalidate_phys_range(start, start + len);
mmap_unlock();
return start;
fail:
return -1;
}
+static void mmap_reserve(abi_ulong start, abi_ulong size)
+{
+ abi_ulong real_start;
+ abi_ulong real_end;
+ abi_ulong addr;
+ abi_ulong end;
+ int prot;
+
+ real_start = start & qemu_host_page_mask;
+ real_end = HOST_PAGE_ALIGN(start + size);
+ end = start + size;
+ if (start > real_start) {
+ /* handle host page containing start */
+ prot = 0;
+ for (addr = real_start; addr < start; addr += TARGET_PAGE_SIZE) {
+ prot |= page_get_flags(addr);
+ }
+ if (real_end == real_start + qemu_host_page_size) {
+ for (addr = end; addr < real_end; addr += TARGET_PAGE_SIZE) {
+ prot |= page_get_flags(addr);
+ }
+ end = real_end;
+ }
+ if (prot != 0)
+ real_start += qemu_host_page_size;
+ }
+ if (end < real_end) {
+ prot = 0;
+ for (addr = end; addr < real_end; addr += TARGET_PAGE_SIZE) {
+ prot |= page_get_flags(addr);
+ }
+ if (prot != 0)
+ real_end -= qemu_host_page_size;
+ }
+ if (real_start != real_end) {
+ mmap(g2h(real_start), real_end - real_start, PROT_NONE,
+ MAP_FIXED | MAP_ANONYMOUS | MAP_PRIVATE | MAP_NORESERVE,
+ -1, 0);
+ }
+}
+
int target_munmap(abi_ulong start, abi_ulong len)
{
abi_ulong end, real_start, real_end, addr;
int prot, ret;
#ifdef DEBUG_MMAP
- printf("munmap: start=0x%lx len=0x%lx\n", start, len);
+ printf("munmap: start=0x" TARGET_ABI_FMT_lx " len=0x"
+ TARGET_ABI_FMT_lx "\n",
+ start, len);
#endif
if (start & ~TARGET_PAGE_MASK)
return -EINVAL;
ret = 0;
/* unmap what we can */
if (real_start < real_end) {
- ret = munmap(g2h(real_start), real_end - real_start);
+ if (reserved_va) {
+ mmap_reserve(real_start, real_end - real_start);
+ } else {
+ ret = munmap(g2h(real_start), real_end - real_start);
+ }
}
- if (ret == 0)
+ if (ret == 0) {
page_set_flags(start, start + len, 0);
+ tb_invalidate_phys_range(start, start + len);
+ }
mmap_unlock();
return ret;
}
-/* XXX: currently, we only handle MAP_ANONYMOUS and not MAP_FIXED
- blocks which have been allocated starting on a host page */
abi_long target_mremap(abi_ulong old_addr, abi_ulong old_size,
abi_ulong new_size, unsigned long flags,
abi_ulong new_addr)
{
int prot;
- unsigned long host_addr;
+ void *host_addr;
mmap_lock();
- /* XXX: use 5 args syscall */
- host_addr = (long)mremap(g2h(old_addr), old_size, new_size, flags);
- if (host_addr == -1) {
+
+ if (flags & MREMAP_FIXED) {
+ host_addr = (void *) syscall(__NR_mremap, g2h(old_addr),
+ old_size, new_size,
+ flags,
+ g2h(new_addr));
+
+ if (reserved_va && host_addr != MAP_FAILED) {
+ /* If new and old addresses overlap then the above mremap will
+ already have failed with EINVAL. */
+ mmap_reserve(old_addr, old_size);
+ }
+ } else if (flags & MREMAP_MAYMOVE) {
+ abi_ulong mmap_start;
+
+ mmap_start = mmap_find_vma(0, new_size);
+
+ if (mmap_start == -1) {
+ errno = ENOMEM;
+ host_addr = MAP_FAILED;
+ } else {
+ host_addr = (void *) syscall(__NR_mremap, g2h(old_addr),
+ old_size, new_size,
+ flags | MREMAP_FIXED,
+ g2h(mmap_start));
+ if (reserved_va) {
+ mmap_reserve(old_addr, old_size);
+ }
+ }
+ } else {
+ int prot = 0;
+ if (reserved_va && old_size < new_size) {
+ abi_ulong addr;
+ for (addr = old_addr + old_size;
+ addr < old_addr + new_size;
+ addr++) {
+ prot |= page_get_flags(addr);
+ }
+ }
+ if (prot == 0) {
+ host_addr = mremap(g2h(old_addr), old_size, new_size, flags);
+ if (host_addr != MAP_FAILED && reserved_va && old_size > new_size) {
+ mmap_reserve(old_addr + old_size, new_size - old_size);
+ }
+ } else {
+ errno = ENOMEM;
+ host_addr = MAP_FAILED;
+ }
+ /* Check if address fits target address space */
+ if ((unsigned long)host_addr + new_size > (abi_ulong)-1) {
+ /* Revert mremap() changes */
+ host_addr = mremap(g2h(old_addr), new_size, old_size, flags);
+ errno = ENOMEM;
+ host_addr = MAP_FAILED;
+ }
+ }
+
+ if (host_addr == MAP_FAILED) {
new_addr = -1;
} else {
new_addr = h2g(host_addr);
page_set_flags(old_addr, old_addr + old_size, 0);
page_set_flags(new_addr, new_addr + new_size, prot | PAGE_VALID);
}
+ tb_invalidate_phys_range(new_addr, new_addr + new_size);
mmap_unlock();
return new_addr;
}
projects / qemu.git / blobdiff