qemu/memory.c

3270 lines
101 KiB
C
Raw Normal View History

/*
* Physical memory management
*
* Copyright 2011 Red Hat, Inc. and/or its affiliates
*
* Authors:
* Avi Kivity <avi@redhat.com>
*
* This work is licensed under the terms of the GNU GPL, version 2. See
* the COPYING file in the top-level directory.
*
* Contributions after 2012-01-13 are licensed under the terms of the
* GNU GPL, version 2 or (at your option) any later version.
*/
#include "qemu/osdep.h"
2016-03-14 11:01:28 +03:00
#include "qapi/error.h"
#include "cpu.h"
#include "exec/memory.h"
#include "exec/address-spaces.h"
#include "qapi/visitor.h"
#include "qemu/bitops.h"
#include "qemu/error-report.h"
#include "qemu/main-loop.h"
#include "qemu/qemu-print.h"
#include "qom/object.h"
#include "trace-root.h"
#include "exec/memory-internal.h"
#include "exec/ram_addr.h"
#include "sysemu/kvm.h"
#include "sysemu/runstate.h"
#include "sysemu/tcg.h"
#include "sysemu/accel.h"
#include "hw/boards.h"
#include "migration/vmstate.h"
//#define DEBUG_UNASSIGNED
static unsigned memory_region_transaction_depth;
static bool memory_region_update_pending;
static bool ioeventfd_update_pending;
bool global_dirty_log;
static QTAILQ_HEAD(, MemoryListener) memory_listeners
= QTAILQ_HEAD_INITIALIZER(memory_listeners);
static QTAILQ_HEAD(, AddressSpace) address_spaces
= QTAILQ_HEAD_INITIALIZER(address_spaces);
static GHashTable *flat_views;
typedef struct AddrRange AddrRange;
/*
* Note that signed integers are needed for negative offsetting in aliases
* (large MemoryRegion::alias_offset).
*/
struct AddrRange {
Int128 start;
Int128 size;
};
static AddrRange addrrange_make(Int128 start, Int128 size)
{
return (AddrRange) { start, size };
}
static bool addrrange_equal(AddrRange r1, AddrRange r2)
{
return int128_eq(r1.start, r2.start) && int128_eq(r1.size, r2.size);
}
static Int128 addrrange_end(AddrRange r)
{
return int128_add(r.start, r.size);
}
static AddrRange addrrange_shift(AddrRange range, Int128 delta)
{
int128_addto(&range.start, delta);
return range;
}
static bool addrrange_contains(AddrRange range, Int128 addr)
{
return int128_ge(addr, range.start)
&& int128_lt(addr, addrrange_end(range));
}
static bool addrrange_intersects(AddrRange r1, AddrRange r2)
{
return addrrange_contains(r1, r2.start)
|| addrrange_contains(r2, r1.start);
}
static AddrRange addrrange_intersection(AddrRange r1, AddrRange r2)
{
Int128 start = int128_max(r1.start, r2.start);
Int128 end = int128_min(addrrange_end(r1), addrrange_end(r2));
return addrrange_make(start, int128_sub(end, start));
}
enum ListenerDirection { Forward, Reverse };
#define MEMORY_LISTENER_CALL_GLOBAL(_callback, _direction, _args...) \
do { \
MemoryListener *_listener; \
\
switch (_direction) { \
case Forward: \
QTAILQ_FOREACH(_listener, &memory_listeners, link) { \
if (_listener->_callback) { \
_listener->_callback(_listener, ##_args); \
} \
} \
break; \
case Reverse: \
QTAILQ_FOREACH_REVERSE(_listener, &memory_listeners, link) { \
if (_listener->_callback) { \
_listener->_callback(_listener, ##_args); \
} \
} \
break; \
default: \
abort(); \
} \
} while (0)
#define MEMORY_LISTENER_CALL(_as, _callback, _direction, _section, _args...) \
do { \
MemoryListener *_listener; \
\
switch (_direction) { \
case Forward: \
QTAILQ_FOREACH(_listener, &(_as)->listeners, link_as) { \
if (_listener->_callback) { \
_listener->_callback(_listener, _section, ##_args); \
} \
} \
break; \
case Reverse: \
QTAILQ_FOREACH_REVERSE(_listener, &(_as)->listeners, link_as) { \
if (_listener->_callback) { \
_listener->_callback(_listener, _section, ##_args); \
} \
} \
break; \
default: \
abort(); \
} \
} while (0)
/* No need to ref/unref .mr, the FlatRange keeps it alive. */
#define MEMORY_LISTENER_UPDATE_REGION(fr, as, dir, callback, _args...) \
do { \
MemoryRegionSection mrs = section_from_flat_range(fr, \
address_space_to_flatview(as)); \
MEMORY_LISTENER_CALL(as, callback, dir, &mrs, ##_args); \
} while(0)
struct CoalescedMemoryRange {
AddrRange addr;
QTAILQ_ENTRY(CoalescedMemoryRange) link;
};
struct MemoryRegionIoeventfd {
AddrRange addr;
bool match_data;
uint64_t data;
EventNotifier *e;
};
static bool memory_region_ioeventfd_before(MemoryRegionIoeventfd *a,
MemoryRegionIoeventfd *b)
{
if (int128_lt(a->addr.start, b->addr.start)) {
return true;
} else if (int128_gt(a->addr.start, b->addr.start)) {
return false;
} else if (int128_lt(a->addr.size, b->addr.size)) {
return true;
} else if (int128_gt(a->addr.size, b->addr.size)) {
return false;
} else if (a->match_data < b->match_data) {
return true;
} else if (a->match_data > b->match_data) {
return false;
} else if (a->match_data) {
if (a->data < b->data) {
return true;
} else if (a->data > b->data) {
return false;
}
}
if (a->e < b->e) {
return true;
} else if (a->e > b->e) {
return false;
}
return false;
}
static bool memory_region_ioeventfd_equal(MemoryRegionIoeventfd *a,
MemoryRegionIoeventfd *b)
{
return !memory_region_ioeventfd_before(a, b)
&& !memory_region_ioeventfd_before(b, a);
}
/* Range of memory in the global map. Addresses are absolute. */
struct FlatRange {
MemoryRegion *mr;
hwaddr offset_in_region;
AddrRange addr;
uint8_t dirty_log_mask;
bool romd_mode;
bool readonly;
bool nonvolatile;
};
#define FOR_EACH_FLAT_RANGE(var, view) \
for (var = (view)->ranges; var < (view)->ranges + (view)->nr; ++var)
static inline MemoryRegionSection
section_from_flat_range(FlatRange *fr, FlatView *fv)
{
return (MemoryRegionSection) {
.mr = fr->mr,
.fv = fv,
.offset_within_region = fr->offset_in_region,
.size = fr->addr.size,
.offset_within_address_space = int128_get64(fr->addr.start),
.readonly = fr->readonly,
.nonvolatile = fr->nonvolatile,
};
}
static bool flatrange_equal(FlatRange *a, FlatRange *b)
{
return a->mr == b->mr
&& addrrange_equal(a->addr, b->addr)
&& a->offset_in_region == b->offset_in_region
&& a->romd_mode == b->romd_mode
&& a->readonly == b->readonly
&& a->nonvolatile == b->nonvolatile;
}
static FlatView *flatview_new(MemoryRegion *mr_root)
{
FlatView *view;
view = g_new0(FlatView, 1);
view->ref = 1;
view->root = mr_root;
memory_region_ref(mr_root);
trace_flatview_new(view, mr_root);
return view;
}
/* Insert a range into a given position. Caller is responsible for maintaining
* sorting order.
*/
static void flatview_insert(FlatView *view, unsigned pos, FlatRange *range)
{
if (view->nr == view->nr_allocated) {
view->nr_allocated = MAX(2 * view->nr, 10);
view->ranges = g_realloc(view->ranges,
view->nr_allocated * sizeof(*view->ranges));
}
memmove(view->ranges + pos + 1, view->ranges + pos,
(view->nr - pos) * sizeof(FlatRange));
view->ranges[pos] = *range;
memory_region_ref(range->mr);
++view->nr;
}
static void flatview_destroy(FlatView *view)
{
int i;
trace_flatview_destroy(view, view->root);
if (view->dispatch) {
address_space_dispatch_free(view->dispatch);
}
for (i = 0; i < view->nr; i++) {
memory_region_unref(view->ranges[i].mr);
}
g_free(view->ranges);
memory_region_unref(view->root);
g_free(view);
}
static bool flatview_ref(FlatView *view)
{
return atomic_fetch_inc_nonzero(&view->ref) > 0;
}
void flatview_unref(FlatView *view)
{
if (atomic_fetch_dec(&view->ref) == 1) {
trace_flatview_destroy_rcu(view, view->root);
assert(view->root);
call_rcu(view, flatview_destroy, rcu);
}
}
static bool can_merge(FlatRange *r1, FlatRange *r2)
{
return int128_eq(addrrange_end(r1->addr), r2->addr.start)
&& r1->mr == r2->mr
&& int128_eq(int128_add(int128_make64(r1->offset_in_region),
r1->addr.size),
int128_make64(r2->offset_in_region))
&& r1->dirty_log_mask == r2->dirty_log_mask
&& r1->romd_mode == r2->romd_mode
&& r1->readonly == r2->readonly
&& r1->nonvolatile == r2->nonvolatile;
}
/* Attempt to simplify a view by merging adjacent ranges */
static void flatview_simplify(FlatView *view)
{
memory: unref the memory region in simplify flatview The memory region reference is increased when insert a range into flatview range array, then decreased by destroy flatview. If some flat range merged by flatview_simplify, the memory region reference can not be decreased by destroy flatview any more. In this case, start virtual machine by the command line: qemu-system-x86_64 -name guest=ubuntu,debug-threads=on -machine pc,accel=kvm,usb=off,dump-guest-core=off -cpu host -m 16384 -realtime mlock=off -smp 8,sockets=2,cores=4,threads=1 -object memory-backend-file,id=ram-node0,prealloc=yes,mem-path=/dev/hugepages,share=yes,size=8589934592 -numa node,nodeid=0,cpus=0-3,memdev=ram-node0 -object memory-backend-file,id=ram-node1,prealloc=yes,mem-path=/dev/hugepages,share=yes,size=8589934592 -numa node,nodeid=1,cpus=4-7,memdev=ram-node1 -no-user-config -nodefaults -rtc base=utc -no-shutdown -boot strict=on -device piix3-usb-uhci,id=usb,bus=pci.0,addr=0x1.0x2 -device virtio-scsi-pci,id=scsi0,bus=pci.0,addr=0x2 -device virtio-serial-pci,id=virtio-serial0,bus=pci.0,addr=0x3 -drive file=ubuntu.qcow2,format=qcow2,if=none,id=drive-virtio-disk0,cache=none,aio=native -device virtio-blk-pci,scsi=off,bus=pci.0,addr=0x4,drive=drive-virtio-disk0,id=virtio-disk0,bootindex=1 -chardev pty,id=charserial0 -device isa-serial,chardev=charserial0,id=serial0 -device usb-tablet,id=input0,bus=usb.0,port=1 -vnc 0.0.0.0:0 -device VGA,id=video0,vgamem_mb=16,bus=pci.0,addr=0x5 -device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x6 -msg timestamp=on And run the script in guest OS: while true do setpci -s 00:06.0 04.b=03 setpci -s 00:06.0 04.b=07 done I found the reference of node0 HostMemoryBackendFile is a big one. (gdb) p numa_info[0]->node_memdev->parent.ref $6 = 1636278 (gdb) Signed-off-by: King Wang<king.wang@huawei.com> Message-Id: <20190712065241.11784-1-king.wang@huawei.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2019-07-12 09:52:41 +03:00
unsigned i, j, k;
i = 0;
while (i < view->nr) {
j = i + 1;
while (j < view->nr
&& can_merge(&view->ranges[j-1], &view->ranges[j])) {
int128_addto(&view->ranges[i].addr.size, view->ranges[j].addr.size);
++j;
}
++i;
memory: unref the memory region in simplify flatview The memory region reference is increased when insert a range into flatview range array, then decreased by destroy flatview. If some flat range merged by flatview_simplify, the memory region reference can not be decreased by destroy flatview any more. In this case, start virtual machine by the command line: qemu-system-x86_64 -name guest=ubuntu,debug-threads=on -machine pc,accel=kvm,usb=off,dump-guest-core=off -cpu host -m 16384 -realtime mlock=off -smp 8,sockets=2,cores=4,threads=1 -object memory-backend-file,id=ram-node0,prealloc=yes,mem-path=/dev/hugepages,share=yes,size=8589934592 -numa node,nodeid=0,cpus=0-3,memdev=ram-node0 -object memory-backend-file,id=ram-node1,prealloc=yes,mem-path=/dev/hugepages,share=yes,size=8589934592 -numa node,nodeid=1,cpus=4-7,memdev=ram-node1 -no-user-config -nodefaults -rtc base=utc -no-shutdown -boot strict=on -device piix3-usb-uhci,id=usb,bus=pci.0,addr=0x1.0x2 -device virtio-scsi-pci,id=scsi0,bus=pci.0,addr=0x2 -device virtio-serial-pci,id=virtio-serial0,bus=pci.0,addr=0x3 -drive file=ubuntu.qcow2,format=qcow2,if=none,id=drive-virtio-disk0,cache=none,aio=native -device virtio-blk-pci,scsi=off,bus=pci.0,addr=0x4,drive=drive-virtio-disk0,id=virtio-disk0,bootindex=1 -chardev pty,id=charserial0 -device isa-serial,chardev=charserial0,id=serial0 -device usb-tablet,id=input0,bus=usb.0,port=1 -vnc 0.0.0.0:0 -device VGA,id=video0,vgamem_mb=16,bus=pci.0,addr=0x5 -device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x6 -msg timestamp=on And run the script in guest OS: while true do setpci -s 00:06.0 04.b=03 setpci -s 00:06.0 04.b=07 done I found the reference of node0 HostMemoryBackendFile is a big one. (gdb) p numa_info[0]->node_memdev->parent.ref $6 = 1636278 (gdb) Signed-off-by: King Wang<king.wang@huawei.com> Message-Id: <20190712065241.11784-1-king.wang@huawei.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2019-07-12 09:52:41 +03:00
for (k = i; k < j; k++) {
memory_region_unref(view->ranges[k].mr);
}
memmove(&view->ranges[i], &view->ranges[j],
(view->nr - j) * sizeof(view->ranges[j]));
view->nr -= j - i;
}
}
static bool memory_region_big_endian(MemoryRegion *mr)
{
#ifdef TARGET_WORDS_BIGENDIAN
return mr->ops->endianness != DEVICE_LITTLE_ENDIAN;
#else
return mr->ops->endianness == DEVICE_BIG_ENDIAN;
#endif
}
static void adjust_endianness(MemoryRegion *mr, uint64_t *data, MemOp op)
{
if ((op & MO_BSWAP) != devend_memop(mr->ops->endianness)) {
switch (op & MO_SIZE) {
case MO_8:
break;
case MO_16:
*data = bswap16(*data);
break;
case MO_32:
*data = bswap32(*data);
break;
case MO_64:
*data = bswap64(*data);
break;
default:
g_assert_not_reached();
}
}
}
static inline void memory_region_shift_read_access(uint64_t *value,
signed shift,
uint64_t mask,
uint64_t tmp)
{
if (shift >= 0) {
*value |= (tmp & mask) << shift;
} else {
*value |= (tmp & mask) >> -shift;
}
}
static inline uint64_t memory_region_shift_write_access(uint64_t *value,
signed shift,
uint64_t mask)
{
uint64_t tmp;
if (shift >= 0) {
tmp = (*value >> shift) & mask;
} else {
tmp = (*value << -shift) & mask;
}
return tmp;
}
static hwaddr memory_region_to_absolute_addr(MemoryRegion *mr, hwaddr offset)
{
MemoryRegion *root;
hwaddr abs_addr = offset;
abs_addr += mr->addr;
for (root = mr; root->container; ) {
root = root->container;
abs_addr += root->addr;
}
return abs_addr;
}
static int get_cpu_index(void)
{
if (current_cpu) {
return current_cpu->cpu_index;
}
return -1;
}
static MemTxResult memory_region_read_accessor(MemoryRegion *mr,
hwaddr addr,
uint64_t *value,
unsigned size,
signed shift,
uint64_t mask,
MemTxAttrs attrs)
{
uint64_t tmp;
tmp = mr->ops->read(mr->opaque, addr, size);
if (mr->subpage) {
trace_memory_region_subpage_read(get_cpu_index(), mr, addr, tmp, size);
} else if (mr == &io_mem_notdirty) {
/* Accesses to code which has previously been translated into a TB show
* up in the MMIO path, as accesses to the io_mem_notdirty
* MemoryRegion. */
trace_memory_region_tb_read(get_cpu_index(), addr, tmp, size);
} else if (TRACE_MEMORY_REGION_OPS_READ_ENABLED) {
hwaddr abs_addr = memory_region_to_absolute_addr(mr, addr);
trace_memory_region_ops_read(get_cpu_index(), mr, abs_addr, tmp, size);
}
memory_region_shift_read_access(value, shift, mask, tmp);
return MEMTX_OK;
}
static MemTxResult memory_region_read_with_attrs_accessor(MemoryRegion *mr,
hwaddr addr,
uint64_t *value,
unsigned size,
signed shift,
uint64_t mask,
MemTxAttrs attrs)
{
uint64_t tmp = 0;
MemTxResult r;
r = mr->ops->read_with_attrs(mr->opaque, addr, &tmp, size, attrs);
if (mr->subpage) {
trace_memory_region_subpage_read(get_cpu_index(), mr, addr, tmp, size);
} else if (mr == &io_mem_notdirty) {
/* Accesses to code which has previously been translated into a TB show
* up in the MMIO path, as accesses to the io_mem_notdirty
* MemoryRegion. */
trace_memory_region_tb_read(get_cpu_index(), addr, tmp, size);
} else if (TRACE_MEMORY_REGION_OPS_READ_ENABLED) {
hwaddr abs_addr = memory_region_to_absolute_addr(mr, addr);
trace_memory_region_ops_read(get_cpu_index(), mr, abs_addr, tmp, size);
}
memory_region_shift_read_access(value, shift, mask, tmp);
return r;
}
static MemTxResult memory_region_write_accessor(MemoryRegion *mr,
hwaddr addr,
uint64_t *value,
unsigned size,
signed shift,
uint64_t mask,
MemTxAttrs attrs)
{
uint64_t tmp = memory_region_shift_write_access(value, shift, mask);
if (mr->subpage) {
trace_memory_region_subpage_write(get_cpu_index(), mr, addr, tmp, size);
} else if (mr == &io_mem_notdirty) {
/* Accesses to code which has previously been translated into a TB show
* up in the MMIO path, as accesses to the io_mem_notdirty
* MemoryRegion. */
trace_memory_region_tb_write(get_cpu_index(), addr, tmp, size);
} else if (TRACE_MEMORY_REGION_OPS_WRITE_ENABLED) {
hwaddr abs_addr = memory_region_to_absolute_addr(mr, addr);
trace_memory_region_ops_write(get_cpu_index(), mr, abs_addr, tmp, size);
}
mr->ops->write(mr->opaque, addr, tmp, size);
return MEMTX_OK;
}
static MemTxResult memory_region_write_with_attrs_accessor(MemoryRegion *mr,
hwaddr addr,
uint64_t *value,
unsigned size,
signed shift,
uint64_t mask,
MemTxAttrs attrs)
{
uint64_t tmp = memory_region_shift_write_access(value, shift, mask);
if (mr->subpage) {
trace_memory_region_subpage_write(get_cpu_index(), mr, addr, tmp, size);
} else if (mr == &io_mem_notdirty) {
/* Accesses to code which has previously been translated into a TB show
* up in the MMIO path, as accesses to the io_mem_notdirty
* MemoryRegion. */
trace_memory_region_tb_write(get_cpu_index(), addr, tmp, size);
} else if (TRACE_MEMORY_REGION_OPS_WRITE_ENABLED) {
hwaddr abs_addr = memory_region_to_absolute_addr(mr, addr);
trace_memory_region_ops_write(get_cpu_index(), mr, abs_addr, tmp, size);
}
return mr->ops->write_with_attrs(mr->opaque, addr, tmp, size, attrs);
}
static MemTxResult access_with_adjusted_size(hwaddr addr,
uint64_t *value,
unsigned size,
unsigned access_size_min,
unsigned access_size_max,
MemTxResult (*access_fn)
(MemoryRegion *mr,
hwaddr addr,
uint64_t *value,
unsigned size,
signed shift,
uint64_t mask,
MemTxAttrs attrs),
MemoryRegion *mr,
MemTxAttrs attrs)
{
uint64_t access_mask;
unsigned access_size;
unsigned i;
MemTxResult r = MEMTX_OK;
if (!access_size_min) {
access_size_min = 1;
}
if (!access_size_max) {
access_size_max = 4;
}
/* FIXME: support unaligned access? */
access_size = MAX(MIN(size, access_size_max), access_size_min);
access_mask = MAKE_64BIT_MASK(0, access_size * 8);
if (memory_region_big_endian(mr)) {
for (i = 0; i < size; i += access_size) {
r |= access_fn(mr, addr + i, value, access_size,
(size - access_size - i) * 8, access_mask, attrs);
}
} else {
for (i = 0; i < size; i += access_size) {
r |= access_fn(mr, addr + i, value, access_size, i * 8,
access_mask, attrs);
}
}
return r;
}
static AddressSpace *memory_region_to_address_space(MemoryRegion *mr)
{
AddressSpace *as;
while (mr->container) {
mr = mr->container;
}
QTAILQ_FOREACH(as, &address_spaces, address_spaces_link) {
if (mr == as->root) {
return as;
}
}
return NULL;
}
/* Render a memory region into the global view. Ranges in @view obscure
* ranges in @mr.
*/
static void render_memory_region(FlatView *view,
MemoryRegion *mr,
Int128 base,
AddrRange clip,
bool readonly,
bool nonvolatile)
{
MemoryRegion *subregion;
unsigned i;
hwaddr offset_in_region;
Int128 remain;
Int128 now;
FlatRange fr;
AddrRange tmp;
if (!mr->enabled) {
return;
}
int128_addto(&base, int128_make64(mr->addr));
readonly |= mr->readonly;
nonvolatile |= mr->nonvolatile;
tmp = addrrange_make(base, mr->size);
if (!addrrange_intersects(tmp, clip)) {
return;
}
clip = addrrange_intersection(tmp, clip);
if (mr->alias) {
int128_subfrom(&base, int128_make64(mr->alias->addr));
int128_subfrom(&base, int128_make64(mr->alias_offset));
render_memory_region(view, mr->alias, base, clip,
readonly, nonvolatile);
return;
}
/* Render subregions in priority order. */
QTAILQ_FOREACH(subregion, &mr->subregions, subregions_link) {
render_memory_region(view, subregion, base, clip,
readonly, nonvolatile);
}
if (!mr->terminates) {
return;
}
offset_in_region = int128_get64(int128_sub(clip.start, base));
base = clip.start;
remain = clip.size;
fr.mr = mr;
fr.dirty_log_mask = memory_region_get_dirty_log_mask(mr);
fr.romd_mode = mr->romd_mode;
fr.readonly = readonly;
fr.nonvolatile = nonvolatile;
/* Render the region itself into any gaps left by the current view. */
for (i = 0; i < view->nr && int128_nz(remain); ++i) {
if (int128_ge(base, addrrange_end(view->ranges[i].addr))) {
continue;
}
if (int128_lt(base, view->ranges[i].addr.start)) {
now = int128_min(remain,
int128_sub(view->ranges[i].addr.start, base));
fr.offset_in_region = offset_in_region;
fr.addr = addrrange_make(base, now);
flatview_insert(view, i, &fr);
++i;
int128_addto(&base, now);
offset_in_region += int128_get64(now);
int128_subfrom(&remain, now);
}
now = int128_sub(int128_min(int128_add(base, remain),
addrrange_end(view->ranges[i].addr)),
base);
int128_addto(&base, now);
offset_in_region += int128_get64(now);
int128_subfrom(&remain, now);
}
if (int128_nz(remain)) {
fr.offset_in_region = offset_in_region;
fr.addr = addrrange_make(base, remain);
flatview_insert(view, i, &fr);
}
}
static MemoryRegion *memory_region_get_flatview_root(MemoryRegion *mr)
{
while (mr->enabled) {
if (mr->alias) {
if (!mr->alias_offset && int128_ge(mr->size, mr->alias->size)) {
/* The alias is included in its entirety. Use it as
* the "real" root, so that we can share more FlatViews.
*/
mr = mr->alias;
continue;
}
} else if (!mr->terminates) {
unsigned int found = 0;
MemoryRegion *child, *next = NULL;
QTAILQ_FOREACH(child, &mr->subregions, subregions_link) {
if (child->enabled) {
if (++found > 1) {
next = NULL;
break;
}
if (!child->addr && int128_ge(mr->size, child->size)) {
/* A child is included in its entirety. If it's the only
* enabled one, use it in the hope of finding an alias down the
* way. This will also let us share FlatViews.
*/
next = child;
}
}
}
if (found == 0) {
return NULL;
}
if (next) {
mr = next;
continue;
}
}
return mr;
}
return NULL;
}
/* Render a memory topology into a list of disjoint absolute ranges. */
static FlatView *generate_memory_topology(MemoryRegion *mr)
{
int i;
FlatView *view;
view = flatview_new(mr);
if (mr) {
render_memory_region(view, mr, int128_zero(),
addrrange_make(int128_zero(), int128_2_64()),
false, false);
}
flatview_simplify(view);
view->dispatch = address_space_dispatch_new(view);
for (i = 0; i < view->nr; i++) {
MemoryRegionSection mrs =
section_from_flat_range(&view->ranges[i], view);
flatview_add_to_dispatch(view, &mrs);
}
address_space_dispatch_compact(view->dispatch);
g_hash_table_replace(flat_views, mr, view);
return view;
}
static void address_space_add_del_ioeventfds(AddressSpace *as,
MemoryRegionIoeventfd *fds_new,
unsigned fds_new_nb,
MemoryRegionIoeventfd *fds_old,
unsigned fds_old_nb)
{
unsigned iold, inew;
MemoryRegionIoeventfd *fd;
MemoryRegionSection section;
/* Generate a symmetric difference of the old and new fd sets, adding
* and deleting as necessary.
*/
iold = inew = 0;
while (iold < fds_old_nb || inew < fds_new_nb) {
if (iold < fds_old_nb
&& (inew == fds_new_nb
|| memory_region_ioeventfd_before(&fds_old[iold],
&fds_new[inew]))) {
fd = &fds_old[iold];
section = (MemoryRegionSection) {
.fv = address_space_to_flatview(as),
.offset_within_address_space = int128_get64(fd->addr.start),
.size = fd->addr.size,
};
MEMORY_LISTENER_CALL(as, eventfd_del, Forward, &section,
fd->match_data, fd->data, fd->e);
++iold;
} else if (inew < fds_new_nb
&& (iold == fds_old_nb
|| memory_region_ioeventfd_before(&fds_new[inew],
&fds_old[iold]))) {
fd = &fds_new[inew];
section = (MemoryRegionSection) {
.fv = address_space_to_flatview(as),
.offset_within_address_space = int128_get64(fd->addr.start),
.size = fd->addr.size,
};
MEMORY_LISTENER_CALL(as, eventfd_add, Reverse, &section,
fd->match_data, fd->data, fd->e);
++inew;
} else {
++iold;
++inew;
}
}
}
FlatView *address_space_get_flatview(AddressSpace *as)
{
FlatView *view;
rcu_read_lock();
do {
view = address_space_to_flatview(as);
/* If somebody has replaced as->current_map concurrently,
* flatview_ref returns false.
*/
} while (!flatview_ref(view));
rcu_read_unlock();
return view;
}
static void address_space_update_ioeventfds(AddressSpace *as)
{
FlatView *view;
FlatRange *fr;
unsigned ioeventfd_nb = 0;
MemoryRegionIoeventfd *ioeventfds = NULL;
AddrRange tmp;
unsigned i;
view = address_space_get_flatview(as);
FOR_EACH_FLAT_RANGE(fr, view) {
for (i = 0; i < fr->mr->ioeventfd_nb; ++i) {
tmp = addrrange_shift(fr->mr->ioeventfds[i].addr,
int128_sub(fr->addr.start,
int128_make64(fr->offset_in_region)));
if (addrrange_intersects(fr->addr, tmp)) {
++ioeventfd_nb;
ioeventfds = g_realloc(ioeventfds,
ioeventfd_nb * sizeof(*ioeventfds));
ioeventfds[ioeventfd_nb-1] = fr->mr->ioeventfds[i];
ioeventfds[ioeventfd_nb-1].addr = tmp;
}
}
}
address_space_add_del_ioeventfds(as, ioeventfds, ioeventfd_nb,
as->ioeventfds, as->ioeventfd_nb);
g_free(as->ioeventfds);
as->ioeventfds = ioeventfds;
as->ioeventfd_nb = ioeventfd_nb;
flatview_unref(view);
}
/*
* Notify the memory listeners about the coalesced IO change events of
* range `cmr'. Only the part that has intersection of the specified
* FlatRange will be sent.
*/
static void flat_range_coalesced_io_notify(FlatRange *fr, AddressSpace *as,
CoalescedMemoryRange *cmr, bool add)
{
AddrRange tmp;
tmp = addrrange_shift(cmr->addr,
int128_sub(fr->addr.start,
int128_make64(fr->offset_in_region)));
if (!addrrange_intersects(tmp, fr->addr)) {
return;
}
tmp = addrrange_intersection(tmp, fr->addr);
if (add) {
MEMORY_LISTENER_UPDATE_REGION(fr, as, Forward, coalesced_io_add,
int128_get64(tmp.start),
int128_get64(tmp.size));
} else {
MEMORY_LISTENER_UPDATE_REGION(fr, as, Reverse, coalesced_io_del,
int128_get64(tmp.start),
int128_get64(tmp.size));
}
}
static void flat_range_coalesced_io_del(FlatRange *fr, AddressSpace *as)
{
CoalescedMemoryRange *cmr;
QTAILQ_FOREACH(cmr, &fr->mr->coalesced, link) {
flat_range_coalesced_io_notify(fr, as, cmr, false);
}
}
static void flat_range_coalesced_io_add(FlatRange *fr, AddressSpace *as)
{
MemoryRegion *mr = fr->mr;
CoalescedMemoryRange *cmr;
if (QTAILQ_EMPTY(&mr->coalesced)) {
return;
}
QTAILQ_FOREACH(cmr, &mr->coalesced, link) {
flat_range_coalesced_io_notify(fr, as, cmr, true);
}
}
static void address_space_update_topology_pass(AddressSpace *as,
const FlatView *old_view,
const FlatView *new_view,
bool adding)
{
unsigned iold, inew;
FlatRange *frold, *frnew;
/* Generate a symmetric difference of the old and new memory maps.
* Kill ranges in the old map, and instantiate ranges in the new map.
*/
iold = inew = 0;
while (iold < old_view->nr || inew < new_view->nr) {
if (iold < old_view->nr) {
frold = &old_view->ranges[iold];
} else {
frold = NULL;
}
if (inew < new_view->nr) {
frnew = &new_view->ranges[inew];
} else {
frnew = NULL;
}
if (frold
&& (!frnew
|| int128_lt(frold->addr.start, frnew->addr.start)
|| (int128_eq(frold->addr.start, frnew->addr.start)
&& !flatrange_equal(frold, frnew)))) {
/* In old but not in new, or in both but attributes changed. */
if (!adding) {
flat_range_coalesced_io_del(frold, as);
MEMORY_LISTENER_UPDATE_REGION(frold, as, Reverse, region_del);
}
++iold;
} else if (frold && frnew && flatrange_equal(frold, frnew)) {
/* In both and unchanged (except logging may have changed) */
if (adding) {
MEMORY_LISTENER_UPDATE_REGION(frnew, as, Forward, region_nop);
if (frnew->dirty_log_mask & ~frold->dirty_log_mask) {
MEMORY_LISTENER_UPDATE_REGION(frnew, as, Forward, log_start,
frold->dirty_log_mask,
frnew->dirty_log_mask);
}
if (frold->dirty_log_mask & ~frnew->dirty_log_mask) {
MEMORY_LISTENER_UPDATE_REGION(frnew, as, Reverse, log_stop,
frold->dirty_log_mask,
frnew->dirty_log_mask);
}
}
++iold;
++inew;
} else {
/* In new */
if (adding) {
MEMORY_LISTENER_UPDATE_REGION(frnew, as, Forward, region_add);
flat_range_coalesced_io_add(frnew, as);
}
++inew;
}
}
}
static void flatviews_init(void)
{
static FlatView *empty_view;
if (flat_views) {
return;
}
flat_views = g_hash_table_new_full(g_direct_hash, g_direct_equal, NULL,
(GDestroyNotify) flatview_unref);
if (!empty_view) {
empty_view = generate_memory_topology(NULL);
/* We keep it alive forever in the global variable. */
flatview_ref(empty_view);
} else {
g_hash_table_replace(flat_views, NULL, empty_view);
flatview_ref(empty_view);
}
}
static void flatviews_reset(void)
{
AddressSpace *as;
if (flat_views) {
g_hash_table_unref(flat_views);
flat_views = NULL;
}
flatviews_init();
/* Render unique FVs */
QTAILQ_FOREACH(as, &address_spaces, address_spaces_link) {
MemoryRegion *physmr = memory_region_get_flatview_root(as->root);
if (g_hash_table_lookup(flat_views, physmr)) {
continue;
}
generate_memory_topology(physmr);
}
}
static void address_space_set_flatview(AddressSpace *as)
{
FlatView *old_view = address_space_to_flatview(as);
MemoryRegion *physmr = memory_region_get_flatview_root(as->root);
FlatView *new_view = g_hash_table_lookup(flat_views, physmr);
assert(new_view);
if (old_view == new_view) {
return;
}
if (old_view) {
flatview_ref(old_view);
}
flatview_ref(new_view);
if (!QTAILQ_EMPTY(&as->listeners)) {
FlatView tmpview = { .nr = 0 }, *old_view2 = old_view;
if (!old_view2) {
old_view2 = &tmpview;
}
address_space_update_topology_pass(as, old_view2, new_view, false);
address_space_update_topology_pass(as, old_view2, new_view, true);
}
/* Writes are protected by the BQL. */
atomic_rcu_set(&as->current_map, new_view);
if (old_view) {
flatview_unref(old_view);
}
/* Note that all the old MemoryRegions are still alive up to this
* point. This relieves most MemoryListeners from the need to
* ref/unref the MemoryRegions they get---unless they use them
* outside the iothread mutex, in which case precise reference
* counting is necessary.
*/
if (old_view) {
flatview_unref(old_view);
}
}
static void address_space_update_topology(AddressSpace *as)
{
MemoryRegion *physmr = memory_region_get_flatview_root(as->root);
flatviews_init();
if (!g_hash_table_lookup(flat_views, physmr)) {
generate_memory_topology(physmr);
}
address_space_set_flatview(as);
}
void memory_region_transaction_begin(void)
{
qemu_flush_coalesced_mmio_buffer();
++memory_region_transaction_depth;
}
void memory_region_transaction_commit(void)
{
AddressSpace *as;
assert(memory_region_transaction_depth);
tcg: drop global lock during TCG code execution This finally allows TCG to benefit from the iothread introduction: Drop the global mutex while running pure TCG CPU code. Reacquire the lock when entering MMIO or PIO emulation, or when leaving the TCG loop. We have to revert a few optimization for the current TCG threading model, namely kicking the TCG thread in qemu_mutex_lock_iothread and not kicking it in qemu_cpu_kick. We also need to disable RAM block reordering until we have a more efficient locking mechanism at hand. Still, a Linux x86 UP guest and my Musicpal ARM model boot fine here. These numbers demonstrate where we gain something: 20338 jan 20 0 331m 75m 6904 R 99 0.9 0:50.95 qemu-system-arm 20337 jan 20 0 331m 75m 6904 S 20 0.9 0:26.50 qemu-system-arm The guest CPU was fully loaded, but the iothread could still run mostly independent on a second core. Without the patch we don't get beyond 32206 jan 20 0 330m 73m 7036 R 82 0.9 1:06.00 qemu-system-arm 32204 jan 20 0 330m 73m 7036 S 21 0.9 0:17.03 qemu-system-arm We don't benefit significantly, though, when the guest is not fully loading a host CPU. Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com> Message-Id: <1439220437-23957-10-git-send-email-fred.konrad@greensocs.com> [FK: Rebase, fix qemu_devices_reset deadlock, rm address_space_* mutex] Signed-off-by: KONRAD Frederic <fred.konrad@greensocs.com> [EGC: fixed iothread lock for cpu-exec IRQ handling] Signed-off-by: Emilio G. Cota <cota@braap.org> [AJB: -smp single-threaded fix, clean commit msg, BQL fixes] Signed-off-by: Alex Bennée <alex.bennee@linaro.org> Reviewed-by: Richard Henderson <rth@twiddle.net> Reviewed-by: Pranith Kumar <bobby.prani@gmail.com> [PM: target-arm changes] Acked-by: Peter Maydell <peter.maydell@linaro.org>
2017-02-23 21:29:11 +03:00
assert(qemu_mutex_iothread_locked());
--memory_region_transaction_depth;
if (!memory_region_transaction_depth) {
if (memory_region_update_pending) {
flatviews_reset();
MEMORY_LISTENER_CALL_GLOBAL(begin, Forward);
QTAILQ_FOREACH(as, &address_spaces, address_spaces_link) {
address_space_set_flatview(as);
address_space_update_ioeventfds(as);
}
memory_region_update_pending = false;
ioeventfd_update_pending = false;
MEMORY_LISTENER_CALL_GLOBAL(commit, Forward);
} else if (ioeventfd_update_pending) {
QTAILQ_FOREACH(as, &address_spaces, address_spaces_link) {
address_space_update_ioeventfds(as);
}
ioeventfd_update_pending = false;
}
}
}
static void memory_region_destructor_none(MemoryRegion *mr)
{
}
static void memory_region_destructor_ram(MemoryRegion *mr)
{
qemu_ram_free(mr->ram_block);
}
static bool memory_region_need_escape(char c)
{
return c == '/' || c == '[' || c == '\\' || c == ']';
}
static char *memory_region_escape_name(const char *name)
{
const char *p;
char *escaped, *q;
uint8_t c;
size_t bytes = 0;
for (p = name; *p; p++) {
bytes += memory_region_need_escape(*p) ? 4 : 1;
}
if (bytes == p - name) {
return g_memdup(name, bytes + 1);
}
escaped = g_malloc(bytes + 1);
for (p = name, q = escaped; *p; p++) {
c = *p;
if (unlikely(memory_region_need_escape(c))) {
*q++ = '\\';
*q++ = 'x';
*q++ = "0123456789abcdef"[c >> 4];
c = "0123456789abcdef"[c & 15];
}
*q++ = c;
}
*q = 0;
return escaped;
}
static void memory_region_do_init(MemoryRegion *mr,
Object *owner,
const char *name,
uint64_t size)
{
mr->size = int128_make64(size);
if (size == UINT64_MAX) {
mr->size = int128_2_64();
}
mr->name = g_strdup(name);
mr->owner = owner;
mr->ram_block = NULL;
if (name) {
char *escaped_name = memory_region_escape_name(name);
char *name_array = g_strdup_printf("%s[*]", escaped_name);
if (!owner) {
owner = container_get(qdev_get_machine(), "/unattached");
}
object_property_add_child(owner, name_array, OBJECT(mr), &error_abort);
object_unref(OBJECT(mr));
g_free(name_array);
g_free(escaped_name);
}
}
void memory_region_init(MemoryRegion *mr,
Object *owner,
const char *name,
uint64_t size)
{
object_initialize(mr, sizeof(*mr), TYPE_MEMORY_REGION);
memory_region_do_init(mr, owner, name, size);
}
static void memory_region_get_addr(Object *obj, Visitor *v, const char *name,
void *opaque, Error **errp)
{
MemoryRegion *mr = MEMORY_REGION(obj);
uint64_t value = mr->addr;
qapi: Swap visit_* arguments for consistent 'name' placement JSON uses "name":value, but many of our visitor interfaces were called with visit_type_FOO(v, &value, name, errp). This can be a bit confusing to have to mentally swap the parameter order to match JSON order. It's particularly bad for visit_start_struct(), where the 'name' parameter is smack in the middle of the otherwise-related group of 'obj, kind, size' parameters! It's time to do a global swap of the parameter ordering, so that the 'name' parameter is always immediately after the Visitor argument. Additional reason in favor of the swap: the existing include/qjson.h prefers listing 'name' first in json_prop_*(), and I have plans to unify that file with the qapi visitors; listing 'name' first in qapi will minimize churn to the (admittedly few) qjson.h clients. Later patches will then fix docs, object.h, visitor-impl.h, and those clients to match. Done by first patching scripts/qapi*.py by hand to make generated files do what I want, then by running the following Coccinelle script to affect the rest of the code base: $ spatch --sp-file script `git grep -l '\bvisit_' -- '**/*.[ch]'` I then had to apply some touchups (Coccinelle insisted on TAB indentation in visitor.h, and botched the signature of visit_type_enum() by rewriting 'const char *const strings[]' to the syntactically invalid 'const char*const[] strings'). The movement of parameters is sufficient to provoke compiler errors if any callers were missed. // Part 1: Swap declaration order @@ type TV, TErr, TObj, T1, T2; identifier OBJ, ARG1, ARG2; @@ void visit_start_struct -(TV v, TObj OBJ, T1 ARG1, const char *name, T2 ARG2, TErr errp) +(TV v, const char *name, TObj OBJ, T1 ARG1, T2 ARG2, TErr errp) { ... } @@ type bool, TV, T1; identifier ARG1; @@ bool visit_optional -(TV v, T1 ARG1, const char *name) +(TV v, const char *name, T1 ARG1) { ... } @@ type TV, TErr, TObj, T1; identifier OBJ, ARG1; @@ void visit_get_next_type -(TV v, TObj OBJ, T1 ARG1, const char *name, TErr errp) +(TV v, const char *name, TObj OBJ, T1 ARG1, TErr errp) { ... } @@ type TV, TErr, TObj, T1, T2; identifier OBJ, ARG1, ARG2; @@ void visit_type_enum -(TV v, TObj OBJ, T1 ARG1, T2 ARG2, const char *name, TErr errp) +(TV v, const char *name, TObj OBJ, T1 ARG1, T2 ARG2, TErr errp) { ... } @@ type TV, TErr, TObj; identifier OBJ; identifier VISIT_TYPE =~ "^visit_type_"; @@ void VISIT_TYPE -(TV v, TObj OBJ, const char *name, TErr errp) +(TV v, const char *name, TObj OBJ, TErr errp) { ... } // Part 2: swap caller order @@ expression V, NAME, OBJ, ARG1, ARG2, ERR; identifier VISIT_TYPE =~ "^visit_type_"; @@ ( -visit_start_struct(V, OBJ, ARG1, NAME, ARG2, ERR) +visit_start_struct(V, NAME, OBJ, ARG1, ARG2, ERR) | -visit_optional(V, ARG1, NAME) +visit_optional(V, NAME, ARG1) | -visit_get_next_type(V, OBJ, ARG1, NAME, ERR) +visit_get_next_type(V, NAME, OBJ, ARG1, ERR) | -visit_type_enum(V, OBJ, ARG1, ARG2, NAME, ERR) +visit_type_enum(V, NAME, OBJ, ARG1, ARG2, ERR) | -VISIT_TYPE(V, OBJ, NAME, ERR) +VISIT_TYPE(V, NAME, OBJ, ERR) ) Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <1454075341-13658-19-git-send-email-eblake@redhat.com> Signed-off-by: Markus Armbruster <armbru@redhat.com>
2016-01-29 16:48:54 +03:00
visit_type_uint64(v, name, &value, errp);
}
static void memory_region_get_container(Object *obj, Visitor *v,
const char *name, void *opaque,
Error **errp)
{
MemoryRegion *mr = MEMORY_REGION(obj);
gchar *path = (gchar *)"";
if (mr->container) {
path = object_get_canonical_path(OBJECT(mr->container));
}
qapi: Swap visit_* arguments for consistent 'name' placement JSON uses "name":value, but many of our visitor interfaces were called with visit_type_FOO(v, &value, name, errp). This can be a bit confusing to have to mentally swap the parameter order to match JSON order. It's particularly bad for visit_start_struct(), where the 'name' parameter is smack in the middle of the otherwise-related group of 'obj, kind, size' parameters! It's time to do a global swap of the parameter ordering, so that the 'name' parameter is always immediately after the Visitor argument. Additional reason in favor of the swap: the existing include/qjson.h prefers listing 'name' first in json_prop_*(), and I have plans to unify that file with the qapi visitors; listing 'name' first in qapi will minimize churn to the (admittedly few) qjson.h clients. Later patches will then fix docs, object.h, visitor-impl.h, and those clients to match. Done by first patching scripts/qapi*.py by hand to make generated files do what I want, then by running the following Coccinelle script to affect the rest of the code base: $ spatch --sp-file script `git grep -l '\bvisit_' -- '**/*.[ch]'` I then had to apply some touchups (Coccinelle insisted on TAB indentation in visitor.h, and botched the signature of visit_type_enum() by rewriting 'const char *const strings[]' to the syntactically invalid 'const char*const[] strings'). The movement of parameters is sufficient to provoke compiler errors if any callers were missed. // Part 1: Swap declaration order @@ type TV, TErr, TObj, T1, T2; identifier OBJ, ARG1, ARG2; @@ void visit_start_struct -(TV v, TObj OBJ, T1 ARG1, const char *name, T2 ARG2, TErr errp) +(TV v, const char *name, TObj OBJ, T1 ARG1, T2 ARG2, TErr errp) { ... } @@ type bool, TV, T1; identifier ARG1; @@ bool visit_optional -(TV v, T1 ARG1, const char *name) +(TV v, const char *name, T1 ARG1) { ... } @@ type TV, TErr, TObj, T1; identifier OBJ, ARG1; @@ void visit_get_next_type -(TV v, TObj OBJ, T1 ARG1, const char *name, TErr errp) +(TV v, const char *name, TObj OBJ, T1 ARG1, TErr errp) { ... } @@ type TV, TErr, TObj, T1, T2; identifier OBJ, ARG1, ARG2; @@ void visit_type_enum -(TV v, TObj OBJ, T1 ARG1, T2 ARG2, const char *name, TErr errp) +(TV v, const char *name, TObj OBJ, T1 ARG1, T2 ARG2, TErr errp) { ... } @@ type TV, TErr, TObj; identifier OBJ; identifier VISIT_TYPE =~ "^visit_type_"; @@ void VISIT_TYPE -(TV v, TObj OBJ, const char *name, TErr errp) +(TV v, const char *name, TObj OBJ, TErr errp) { ... } // Part 2: swap caller order @@ expression V, NAME, OBJ, ARG1, ARG2, ERR; identifier VISIT_TYPE =~ "^visit_type_"; @@ ( -visit_start_struct(V, OBJ, ARG1, NAME, ARG2, ERR) +visit_start_struct(V, NAME, OBJ, ARG1, ARG2, ERR) | -visit_optional(V, ARG1, NAME) +visit_optional(V, NAME, ARG1) | -visit_get_next_type(V, OBJ, ARG1, NAME, ERR) +visit_get_next_type(V, NAME, OBJ, ARG1, ERR) | -visit_type_enum(V, OBJ, ARG1, ARG2, NAME, ERR) +visit_type_enum(V, NAME, OBJ, ARG1, ARG2, ERR) | -VISIT_TYPE(V, OBJ, NAME, ERR) +VISIT_TYPE(V, NAME, OBJ, ERR) ) Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <1454075341-13658-19-git-send-email-eblake@redhat.com> Signed-off-by: Markus Armbruster <armbru@redhat.com>
2016-01-29 16:48:54 +03:00
visit_type_str(v, name, &path, errp);
if (mr->container) {
g_free(path);
}
}
static Object *memory_region_resolve_container(Object *obj, void *opaque,
const char *part)
{
MemoryRegion *mr = MEMORY_REGION(obj);
return OBJECT(mr->container);
}
static void memory_region_get_priority(Object *obj, Visitor *v,
const char *name, void *opaque,
Error **errp)
{
MemoryRegion *mr = MEMORY_REGION(obj);
int32_t value = mr->priority;
qapi: Swap visit_* arguments for consistent 'name' placement JSON uses "name":value, but many of our visitor interfaces were called with visit_type_FOO(v, &value, name, errp). This can be a bit confusing to have to mentally swap the parameter order to match JSON order. It's particularly bad for visit_start_struct(), where the 'name' parameter is smack in the middle of the otherwise-related group of 'obj, kind, size' parameters! It's time to do a global swap of the parameter ordering, so that the 'name' parameter is always immediately after the Visitor argument. Additional reason in favor of the swap: the existing include/qjson.h prefers listing 'name' first in json_prop_*(), and I have plans to unify that file with the qapi visitors; listing 'name' first in qapi will minimize churn to the (admittedly few) qjson.h clients. Later patches will then fix docs, object.h, visitor-impl.h, and those clients to match. Done by first patching scripts/qapi*.py by hand to make generated files do what I want, then by running the following Coccinelle script to affect the rest of the code base: $ spatch --sp-file script `git grep -l '\bvisit_' -- '**/*.[ch]'` I then had to apply some touchups (Coccinelle insisted on TAB indentation in visitor.h, and botched the signature of visit_type_enum() by rewriting 'const char *const strings[]' to the syntactically invalid 'const char*const[] strings'). The movement of parameters is sufficient to provoke compiler errors if any callers were missed. // Part 1: Swap declaration order @@ type TV, TErr, TObj, T1, T2; identifier OBJ, ARG1, ARG2; @@ void visit_start_struct -(TV v, TObj OBJ, T1 ARG1, const char *name, T2 ARG2, TErr errp) +(TV v, const char *name, TObj OBJ, T1 ARG1, T2 ARG2, TErr errp) { ... } @@ type bool, TV, T1; identifier ARG1; @@ bool visit_optional -(TV v, T1 ARG1, const char *name) +(TV v, const char *name, T1 ARG1) { ... } @@ type TV, TErr, TObj, T1; identifier OBJ, ARG1; @@ void visit_get_next_type -(TV v, TObj OBJ, T1 ARG1, const char *name, TErr errp) +(TV v, const char *name, TObj OBJ, T1 ARG1, TErr errp) { ... } @@ type TV, TErr, TObj, T1, T2; identifier OBJ, ARG1, ARG2; @@ void visit_type_enum -(TV v, TObj OBJ, T1 ARG1, T2 ARG2, const char *name, TErr errp) +(TV v, const char *name, TObj OBJ, T1 ARG1, T2 ARG2, TErr errp) { ... } @@ type TV, TErr, TObj; identifier OBJ; identifier VISIT_TYPE =~ "^visit_type_"; @@ void VISIT_TYPE -(TV v, TObj OBJ, const char *name, TErr errp) +(TV v, const char *name, TObj OBJ, TErr errp) { ... } // Part 2: swap caller order @@ expression V, NAME, OBJ, ARG1, ARG2, ERR; identifier VISIT_TYPE =~ "^visit_type_"; @@ ( -visit_start_struct(V, OBJ, ARG1, NAME, ARG2, ERR) +visit_start_struct(V, NAME, OBJ, ARG1, ARG2, ERR) | -visit_optional(V, ARG1, NAME) +visit_optional(V, NAME, ARG1) | -visit_get_next_type(V, OBJ, ARG1, NAME, ERR) +visit_get_next_type(V, NAME, OBJ, ARG1, ERR) | -visit_type_enum(V, OBJ, ARG1, ARG2, NAME, ERR) +visit_type_enum(V, NAME, OBJ, ARG1, ARG2, ERR) | -VISIT_TYPE(V, OBJ, NAME, ERR) +VISIT_TYPE(V, NAME, OBJ, ERR) ) Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <1454075341-13658-19-git-send-email-eblake@redhat.com> Signed-off-by: Markus Armbruster <armbru@redhat.com>
2016-01-29 16:48:54 +03:00
visit_type_int32(v, name, &value, errp);
}
static void memory_region_get_size(Object *obj, Visitor *v, const char *name,
void *opaque, Error **errp)
{
MemoryRegion *mr = MEMORY_REGION(obj);
uint64_t value = memory_region_size(mr);
qapi: Swap visit_* arguments for consistent 'name' placement JSON uses "name":value, but many of our visitor interfaces were called with visit_type_FOO(v, &value, name, errp). This can be a bit confusing to have to mentally swap the parameter order to match JSON order. It's particularly bad for visit_start_struct(), where the 'name' parameter is smack in the middle of the otherwise-related group of 'obj, kind, size' parameters! It's time to do a global swap of the parameter ordering, so that the 'name' parameter is always immediately after the Visitor argument. Additional reason in favor of the swap: the existing include/qjson.h prefers listing 'name' first in json_prop_*(), and I have plans to unify that file with the qapi visitors; listing 'name' first in qapi will minimize churn to the (admittedly few) qjson.h clients. Later patches will then fix docs, object.h, visitor-impl.h, and those clients to match. Done by first patching scripts/qapi*.py by hand to make generated files do what I want, then by running the following Coccinelle script to affect the rest of the code base: $ spatch --sp-file script `git grep -l '\bvisit_' -- '**/*.[ch]'` I then had to apply some touchups (Coccinelle insisted on TAB indentation in visitor.h, and botched the signature of visit_type_enum() by rewriting 'const char *const strings[]' to the syntactically invalid 'const char*const[] strings'). The movement of parameters is sufficient to provoke compiler errors if any callers were missed. // Part 1: Swap declaration order @@ type TV, TErr, TObj, T1, T2; identifier OBJ, ARG1, ARG2; @@ void visit_start_struct -(TV v, TObj OBJ, T1 ARG1, const char *name, T2 ARG2, TErr errp) +(TV v, const char *name, TObj OBJ, T1 ARG1, T2 ARG2, TErr errp) { ... } @@ type bool, TV, T1; identifier ARG1; @@ bool visit_optional -(TV v, T1 ARG1, const char *name) +(TV v, const char *name, T1 ARG1) { ... } @@ type TV, TErr, TObj, T1; identifier OBJ, ARG1; @@ void visit_get_next_type -(TV v, TObj OBJ, T1 ARG1, const char *name, TErr errp) +(TV v, const char *name, TObj OBJ, T1 ARG1, TErr errp) { ... } @@ type TV, TErr, TObj, T1, T2; identifier OBJ, ARG1, ARG2; @@ void visit_type_enum -(TV v, TObj OBJ, T1 ARG1, T2 ARG2, const char *name, TErr errp) +(TV v, const char *name, TObj OBJ, T1 ARG1, T2 ARG2, TErr errp) { ... } @@ type TV, TErr, TObj; identifier OBJ; identifier VISIT_TYPE =~ "^visit_type_"; @@ void VISIT_TYPE -(TV v, TObj OBJ, const char *name, TErr errp) +(TV v, const char *name, TObj OBJ, TErr errp) { ... } // Part 2: swap caller order @@ expression V, NAME, OBJ, ARG1, ARG2, ERR; identifier VISIT_TYPE =~ "^visit_type_"; @@ ( -visit_start_struct(V, OBJ, ARG1, NAME, ARG2, ERR) +visit_start_struct(V, NAME, OBJ, ARG1, ARG2, ERR) | -visit_optional(V, ARG1, NAME) +visit_optional(V, NAME, ARG1) | -visit_get_next_type(V, OBJ, ARG1, NAME, ERR) +visit_get_next_type(V, NAME, OBJ, ARG1, ERR) | -visit_type_enum(V, OBJ, ARG1, ARG2, NAME, ERR) +visit_type_enum(V, NAME, OBJ, ARG1, ARG2, ERR) | -VISIT_TYPE(V, OBJ, NAME, ERR) +VISIT_TYPE(V, NAME, OBJ, ERR) ) Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <1454075341-13658-19-git-send-email-eblake@redhat.com> Signed-off-by: Markus Armbruster <armbru@redhat.com>
2016-01-29 16:48:54 +03:00
visit_type_uint64(v, name, &value, errp);
}
static void memory_region_initfn(Object *obj)
{
MemoryRegion *mr = MEMORY_REGION(obj);
ObjectProperty *op;
mr->ops = &unassigned_mem_ops;
mr->enabled = true;
mr->romd_mode = true;
mr->global_locking = true;
mr->destructor = memory_region_destructor_none;
QTAILQ_INIT(&mr->subregions);
QTAILQ_INIT(&mr->coalesced);
op = object_property_add(OBJECT(mr), "container",
"link<" TYPE_MEMORY_REGION ">",
memory_region_get_container,
NULL, /* memory_region_set_container */
NULL, NULL, &error_abort);
op->resolve = memory_region_resolve_container;
object_property_add(OBJECT(mr), "addr", "uint64",
memory_region_get_addr,
NULL, /* memory_region_set_addr */
NULL, NULL, &error_abort);
object_property_add(OBJECT(mr), "priority", "uint32",
memory_region_get_priority,
NULL, /* memory_region_set_priority */
NULL, NULL, &error_abort);
object_property_add(OBJECT(mr), "size", "uint64",
memory_region_get_size,
NULL, /* memory_region_set_size, */
NULL, NULL, &error_abort);
}
static void iommu_memory_region_initfn(Object *obj)
{
MemoryRegion *mr = MEMORY_REGION(obj);
mr->is_iommu = true;
}
static uint64_t unassigned_mem_read(void *opaque, hwaddr addr,
unsigned size)
{
#ifdef DEBUG_UNASSIGNED
printf("Unassigned mem read " TARGET_FMT_plx "\n", addr);
#endif
if (current_cpu != NULL) {
bool is_exec = current_cpu->mem_io_access_type == MMU_INST_FETCH;
cpu_unassigned_access(current_cpu, addr, false, is_exec, 0, size);
}
return 0;
}
static void unassigned_mem_write(void *opaque, hwaddr addr,
uint64_t val, unsigned size)
{
#ifdef DEBUG_UNASSIGNED
printf("Unassigned mem write " TARGET_FMT_plx " = 0x%"PRIx64"\n", addr, val);
#endif
if (current_cpu != NULL) {
cpu_unassigned_access(current_cpu, addr, true, false, 0, size);
}
}
static bool unassigned_mem_accepts(void *opaque, hwaddr addr,
unsigned size, bool is_write,
MemTxAttrs attrs)
{
return false;
}
const MemoryRegionOps unassigned_mem_ops = {
.valid.accepts = unassigned_mem_accepts,
.endianness = DEVICE_NATIVE_ENDIAN,
};
memory: Don't use memcpy for ram_device regions With a vfio assigned device we lay down a base MemoryRegion registered as an IO region, giving us read & write accessors. If the region supports mmap, we lay down a higher priority sub-region MemoryRegion on top of the base layer initialized as a RAM device pointer to the mmap. Finally, if we have any quirks for the device (ie. address ranges that need additional virtualization support), we put another IO sub-region on top of the mmap MemoryRegion. When this is flattened, we now potentially have sub-page mmap MemoryRegions exposed which cannot be directly mapped through KVM. This is as expected, but a subtle detail of this is that we end up with two different access mechanisms through QEMU. If we disable the mmap MemoryRegion, we make use of the IO MemoryRegion and service accesses using pread and pwrite to the vfio device file descriptor. If the mmap MemoryRegion is enabled and results in one of these sub-page gaps, QEMU handles the access as RAM, using memcpy to the mmap. Using either pread/pwrite or the mmap directly should be correct, but using memcpy causes us problems. I expect that not only does memcpy not necessarily honor the original width and alignment in performing a copy, but it potentially also uses processor instructions not intended for MMIO spaces. It turns out that this has been a problem for Realtek NIC assignment, which has such a quirk that creates a sub-page mmap MemoryRegion access. To resolve this, we disable memory_access_is_direct() for ram_device regions since QEMU assumes that it can use memcpy for those regions. Instead we access through MemoryRegionOps, which replaces the memcpy with simple de-references of standard sizes to the host memory. With this patch we attempt to provide unrestricted access to the RAM device, allowing byte through qword access as well as unaligned access. The assumption here is that accesses initiated by the VM are driven by a device specific driver, which knows the device capabilities. If unaligned accesses are not supported by the device, we don't want them to work in a VM by performing multiple aligned accesses to compose the unaligned access. A down-side of this philosophy is that the xp command from the monitor attempts to use the largest available access weidth, unaware of the underlying device. Using memcpy had this same restriction, but at least now an operator can dump individual registers, even if blocks of device memory may result in access widths beyond the capabilities of a given device (RTL NICs only support up to dword). Reported-by: Thorsten Kohfeldt <thorsten.kohfeldt@gmx.de> Signed-off-by: Alex Williamson <alex.williamson@redhat.com> Acked-by: Paolo Bonzini <pbonzini@redhat.com>
2016-10-31 18:53:03 +03:00
static uint64_t memory_region_ram_device_read(void *opaque,
hwaddr addr, unsigned size)
{
MemoryRegion *mr = opaque;
uint64_t data = (uint64_t)~0;
switch (size) {
case 1:
data = *(uint8_t *)(mr->ram_block->host + addr);
break;
case 2:
data = *(uint16_t *)(mr->ram_block->host + addr);
break;
case 4:
data = *(uint32_t *)(mr->ram_block->host + addr);
break;
case 8:
data = *(uint64_t *)(mr->ram_block->host + addr);
break;
}
trace_memory_region_ram_device_read(get_cpu_index(), mr, addr, data, size);
return data;
}
static void memory_region_ram_device_write(void *opaque, hwaddr addr,
uint64_t data, unsigned size)
{
MemoryRegion *mr = opaque;
trace_memory_region_ram_device_write(get_cpu_index(), mr, addr, data, size);
switch (size) {
case 1:
*(uint8_t *)(mr->ram_block->host + addr) = (uint8_t)data;
break;
case 2:
*(uint16_t *)(mr->ram_block->host + addr) = (uint16_t)data;
break;
case 4:
*(uint32_t *)(mr->ram_block->host + addr) = (uint32_t)data;
break;
case 8:
*(uint64_t *)(mr->ram_block->host + addr) = data;
break;
}
}
static const MemoryRegionOps ram_device_mem_ops = {
.read = memory_region_ram_device_read,
.write = memory_region_ram_device_write,
.endianness = DEVICE_HOST_ENDIAN,
memory: Don't use memcpy for ram_device regions With a vfio assigned device we lay down a base MemoryRegion registered as an IO region, giving us read & write accessors. If the region supports mmap, we lay down a higher priority sub-region MemoryRegion on top of the base layer initialized as a RAM device pointer to the mmap. Finally, if we have any quirks for the device (ie. address ranges that need additional virtualization support), we put another IO sub-region on top of the mmap MemoryRegion. When this is flattened, we now potentially have sub-page mmap MemoryRegions exposed which cannot be directly mapped through KVM. This is as expected, but a subtle detail of this is that we end up with two different access mechanisms through QEMU. If we disable the mmap MemoryRegion, we make use of the IO MemoryRegion and service accesses using pread and pwrite to the vfio device file descriptor. If the mmap MemoryRegion is enabled and results in one of these sub-page gaps, QEMU handles the access as RAM, using memcpy to the mmap. Using either pread/pwrite or the mmap directly should be correct, but using memcpy causes us problems. I expect that not only does memcpy not necessarily honor the original width and alignment in performing a copy, but it potentially also uses processor instructions not intended for MMIO spaces. It turns out that this has been a problem for Realtek NIC assignment, which has such a quirk that creates a sub-page mmap MemoryRegion access. To resolve this, we disable memory_access_is_direct() for ram_device regions since QEMU assumes that it can use memcpy for those regions. Instead we access through MemoryRegionOps, which replaces the memcpy with simple de-references of standard sizes to the host memory. With this patch we attempt to provide unrestricted access to the RAM device, allowing byte through qword access as well as unaligned access. The assumption here is that accesses initiated by the VM are driven by a device specific driver, which knows the device capabilities. If unaligned accesses are not supported by the device, we don't want them to work in a VM by performing multiple aligned accesses to compose the unaligned access. A down-side of this philosophy is that the xp command from the monitor attempts to use the largest available access weidth, unaware of the underlying device. Using memcpy had this same restriction, but at least now an operator can dump individual registers, even if blocks of device memory may result in access widths beyond the capabilities of a given device (RTL NICs only support up to dword). Reported-by: Thorsten Kohfeldt <thorsten.kohfeldt@gmx.de> Signed-off-by: Alex Williamson <alex.williamson@redhat.com> Acked-by: Paolo Bonzini <pbonzini@redhat.com>
2016-10-31 18:53:03 +03:00
.valid = {
.min_access_size = 1,
.max_access_size = 8,
.unaligned = true,
},
.impl = {
.min_access_size = 1,
.max_access_size = 8,
.unaligned = true,
},
};
bool memory_region_access_valid(MemoryRegion *mr,
hwaddr addr,
unsigned size,
bool is_write,
MemTxAttrs attrs)
{
int access_size_min, access_size_max;
int access_size, i;
if (!mr->ops->valid.unaligned && (addr & (size - 1))) {
return false;
}
if (!mr->ops->valid.accepts) {
return true;
}
access_size_min = mr->ops->valid.min_access_size;
if (!mr->ops->valid.min_access_size) {
access_size_min = 1;
}
access_size_max = mr->ops->valid.max_access_size;
if (!mr->ops->valid.max_access_size) {
access_size_max = 4;
}
access_size = MAX(MIN(size, access_size_max), access_size_min);
for (i = 0; i < size; i += access_size) {
if (!mr->ops->valid.accepts(mr->opaque, addr + i, access_size,
is_write, attrs)) {
return false;
}
}
return true;
}
static MemTxResult memory_region_dispatch_read1(MemoryRegion *mr,
hwaddr addr,
uint64_t *pval,
unsigned size,
MemTxAttrs attrs)
{
*pval = 0;
if (mr->ops->read) {
return access_with_adjusted_size(addr, pval, size,
mr->ops->impl.min_access_size,
mr->ops->impl.max_access_size,
memory_region_read_accessor,
mr, attrs);
} else {
return access_with_adjusted_size(addr, pval, size,
mr->ops->impl.min_access_size,
mr->ops->impl.max_access_size,
memory_region_read_with_attrs_accessor,
mr, attrs);
}
}
MemTxResult memory_region_dispatch_read(MemoryRegion *mr,
hwaddr addr,
uint64_t *pval,
MemOp op,
MemTxAttrs attrs)
{
unsigned size = memop_size(op);
MemTxResult r;
if (!memory_region_access_valid(mr, addr, size, false, attrs)) {
*pval = unassigned_mem_read(mr, addr, size);
return MEMTX_DECODE_ERROR;
}
r = memory_region_dispatch_read1(mr, addr, pval, size, attrs);
adjust_endianness(mr, pval, op);
return r;
}
/* Return true if an eventfd was signalled */
static bool memory_region_dispatch_write_eventfds(MemoryRegion *mr,
hwaddr addr,
uint64_t data,
unsigned size,
MemTxAttrs attrs)
{
MemoryRegionIoeventfd ioeventfd = {
.addr = addrrange_make(int128_make64(addr), int128_make64(size)),
.data = data,
};
unsigned i;
for (i = 0; i < mr->ioeventfd_nb; i++) {
ioeventfd.match_data = mr->ioeventfds[i].match_data;
ioeventfd.e = mr->ioeventfds[i].e;
if (memory_region_ioeventfd_equal(&ioeventfd, &mr->ioeventfds[i])) {
event_notifier_set(ioeventfd.e);
return true;
}
}
return false;
}
MemTxResult memory_region_dispatch_write(MemoryRegion *mr,
hwaddr addr,
uint64_t data,
MemOp op,
MemTxAttrs attrs)
{
unsigned size = memop_size(op);
if (!memory_region_access_valid(mr, addr, size, true, attrs)) {
unassigned_mem_write(mr, addr, data, size);
return MEMTX_DECODE_ERROR;
}
adjust_endianness(mr, &data, op);
if ((!kvm_eventfds_enabled()) &&
memory_region_dispatch_write_eventfds(mr, addr, data, size, attrs)) {
return MEMTX_OK;
}
if (mr->ops->write) {
return access_with_adjusted_size(addr, &data, size,
mr->ops->impl.min_access_size,
mr->ops->impl.max_access_size,
memory_region_write_accessor, mr,
attrs);
} else {
return
access_with_adjusted_size(addr, &data, size,
mr->ops->impl.min_access_size,
mr->ops->impl.max_access_size,
memory_region_write_with_attrs_accessor,
mr, attrs);
}
}
void memory_region_init_io(MemoryRegion *mr,
Object *owner,
const MemoryRegionOps *ops,
void *opaque,
const char *name,
uint64_t size)
{
memory_region_init(mr, owner, name, size);
mr->ops = ops ? ops : &unassigned_mem_ops;
mr->opaque = opaque;
mr->terminates = true;
}
void memory_region_init_ram_nomigrate(MemoryRegion *mr,
Object *owner,
const char *name,
uint64_t size,
Error **errp)
{
memory_region_init_ram_shared_nomigrate(mr, owner, name, size, false, errp);
}
void memory_region_init_ram_shared_nomigrate(MemoryRegion *mr,
Object *owner,
const char *name,
uint64_t size,
bool share,
Error **errp)
{
Error *err = NULL;
memory_region_init(mr, owner, name, size);
mr->ram = true;
mr->terminates = true;
mr->destructor = memory_region_destructor_ram;
mr->ram_block = qemu_ram_alloc(size, share, mr, &err);
mr->dirty_log_mask = tcg_enabled() ? (1 << DIRTY_MEMORY_CODE) : 0;
if (err) {
mr->size = int128_zero();
object_unparent(OBJECT(mr));
error_propagate(errp, err);
}
}
void memory_region_init_resizeable_ram(MemoryRegion *mr,
Object *owner,
const char *name,
uint64_t size,
uint64_t max_size,
void (*resized)(const char*,
uint64_t length,
void *host),
Error **errp)
{
Error *err = NULL;
memory_region_init(mr, owner, name, size);
mr->ram = true;
mr->terminates = true;
mr->destructor = memory_region_destructor_ram;
mr->ram_block = qemu_ram_alloc_resizeable(size, max_size, resized,
mr, &err);
mr->dirty_log_mask = tcg_enabled() ? (1 << DIRTY_MEMORY_CODE) : 0;
if (err) {
mr->size = int128_zero();
object_unparent(OBJECT(mr));
error_propagate(errp, err);
}
}
#ifdef CONFIG_POSIX
void memory_region_init_ram_from_file(MemoryRegion *mr,
struct Object *owner,
const char *name,
uint64_t size,
uint64_t align,
uint32_t ram_flags,
const char *path,
Error **errp)
{
Error *err = NULL;
memory_region_init(mr, owner, name, size);
mr->ram = true;
mr->terminates = true;
mr->destructor = memory_region_destructor_ram;
mr->align = align;
mr->ram_block = qemu_ram_alloc_from_file(size, mr, ram_flags, path, &err);
mr->dirty_log_mask = tcg_enabled() ? (1 << DIRTY_MEMORY_CODE) : 0;
if (err) {
mr->size = int128_zero();
object_unparent(OBJECT(mr));
error_propagate(errp, err);
}
}
void memory_region_init_ram_from_fd(MemoryRegion *mr,
struct Object *owner,
const char *name,
uint64_t size,
bool share,
int fd,
Error **errp)
{
Error *err = NULL;
memory_region_init(mr, owner, name, size);
mr->ram = true;
mr->terminates = true;
mr->destructor = memory_region_destructor_ram;
mr->ram_block = qemu_ram_alloc_from_fd(size, mr,
share ? RAM_SHARED : 0,
fd, &err);
mr->dirty_log_mask = tcg_enabled() ? (1 << DIRTY_MEMORY_CODE) : 0;
if (err) {
mr->size = int128_zero();
object_unparent(OBJECT(mr));
error_propagate(errp, err);
}
}
#endif
void memory_region_init_ram_ptr(MemoryRegion *mr,
Object *owner,
const char *name,
uint64_t size,
void *ptr)
{
memory_region_init(mr, owner, name, size);
mr->ram = true;
mr->terminates = true;
mr->destructor = memory_region_destructor_ram;
mr->dirty_log_mask = tcg_enabled() ? (1 << DIRTY_MEMORY_CODE) : 0;
/* qemu_ram_alloc_from_ptr cannot fail with ptr != NULL. */
assert(ptr != NULL);
mr->ram_block = qemu_ram_alloc_from_ptr(size, ptr, mr, &error_fatal);
}
void memory_region_init_ram_device_ptr(MemoryRegion *mr,
Object *owner,
const char *name,
uint64_t size,
void *ptr)
{
memory_region_init(mr, owner, name, size);
mr->ram = true;
mr->terminates = true;
mr->ram_device = true;
memory: Don't use memcpy for ram_device regions With a vfio assigned device we lay down a base MemoryRegion registered as an IO region, giving us read & write accessors. If the region supports mmap, we lay down a higher priority sub-region MemoryRegion on top of the base layer initialized as a RAM device pointer to the mmap. Finally, if we have any quirks for the device (ie. address ranges that need additional virtualization support), we put another IO sub-region on top of the mmap MemoryRegion. When this is flattened, we now potentially have sub-page mmap MemoryRegions exposed which cannot be directly mapped through KVM. This is as expected, but a subtle detail of this is that we end up with two different access mechanisms through QEMU. If we disable the mmap MemoryRegion, we make use of the IO MemoryRegion and service accesses using pread and pwrite to the vfio device file descriptor. If the mmap MemoryRegion is enabled and results in one of these sub-page gaps, QEMU handles the access as RAM, using memcpy to the mmap. Using either pread/pwrite or the mmap directly should be correct, but using memcpy causes us problems. I expect that not only does memcpy not necessarily honor the original width and alignment in performing a copy, but it potentially also uses processor instructions not intended for MMIO spaces. It turns out that this has been a problem for Realtek NIC assignment, which has such a quirk that creates a sub-page mmap MemoryRegion access. To resolve this, we disable memory_access_is_direct() for ram_device regions since QEMU assumes that it can use memcpy for those regions. Instead we access through MemoryRegionOps, which replaces the memcpy with simple de-references of standard sizes to the host memory. With this patch we attempt to provide unrestricted access to the RAM device, allowing byte through qword access as well as unaligned access. The assumption here is that accesses initiated by the VM are driven by a device specific driver, which knows the device capabilities. If unaligned accesses are not supported by the device, we don't want them to work in a VM by performing multiple aligned accesses to compose the unaligned access. A down-side of this philosophy is that the xp command from the monitor attempts to use the largest available access weidth, unaware of the underlying device. Using memcpy had this same restriction, but at least now an operator can dump individual registers, even if blocks of device memory may result in access widths beyond the capabilities of a given device (RTL NICs only support up to dword). Reported-by: Thorsten Kohfeldt <thorsten.kohfeldt@gmx.de> Signed-off-by: Alex Williamson <alex.williamson@redhat.com> Acked-by: Paolo Bonzini <pbonzini@redhat.com>
2016-10-31 18:53:03 +03:00
mr->ops = &ram_device_mem_ops;
mr->opaque = mr;
mr->destructor = memory_region_destructor_ram;
mr->dirty_log_mask = tcg_enabled() ? (1 << DIRTY_MEMORY_CODE) : 0;
/* qemu_ram_alloc_from_ptr cannot fail with ptr != NULL. */
assert(ptr != NULL);
mr->ram_block = qemu_ram_alloc_from_ptr(size, ptr, mr, &error_fatal);
}
void memory_region_init_alias(MemoryRegion *mr,
Object *owner,
const char *name,
MemoryRegion *orig,
hwaddr offset,
uint64_t size)
{
memory_region_init(mr, owner, name, size);
mr->alias = orig;
mr->alias_offset = offset;
}
void memory_region_init_rom_nomigrate(MemoryRegion *mr,
struct Object *owner,
const char *name,
uint64_t size,
Error **errp)
{
Error *err = NULL;
memory_region_init(mr, owner, name, size);
mr->ram = true;
mr->readonly = true;
mr->terminates = true;
mr->destructor = memory_region_destructor_ram;
mr->ram_block = qemu_ram_alloc(size, false, mr, &err);
mr->dirty_log_mask = tcg_enabled() ? (1 << DIRTY_MEMORY_CODE) : 0;
if (err) {
mr->size = int128_zero();
object_unparent(OBJECT(mr));
error_propagate(errp, err);
}
}
void memory_region_init_rom_device_nomigrate(MemoryRegion *mr,
Object *owner,
const MemoryRegionOps *ops,
void *opaque,
const char *name,
uint64_t size,
Error **errp)
{
Error *err = NULL;
assert(ops);
memory_region_init(mr, owner, name, size);
mr->ops = ops;
mr->opaque = opaque;
mr->terminates = true;
mr->rom_device = true;
mr->destructor = memory_region_destructor_ram;
mr->ram_block = qemu_ram_alloc(size, false, mr, &err);
if (err) {
mr->size = int128_zero();
object_unparent(OBJECT(mr));
error_propagate(errp, err);
}
}
void memory_region_init_iommu(void *_iommu_mr,
size_t instance_size,
const char *mrtypename,
Object *owner,
const char *name,
uint64_t size)
{
struct IOMMUMemoryRegion *iommu_mr;
struct MemoryRegion *mr;
object_initialize(_iommu_mr, instance_size, mrtypename);
mr = MEMORY_REGION(_iommu_mr);
memory_region_do_init(mr, owner, name, size);
iommu_mr = IOMMU_MEMORY_REGION(mr);
mr->terminates = true; /* then re-forwards */
QLIST_INIT(&iommu_mr->iommu_notify);
iommu_mr->iommu_notify_flags = IOMMU_NOTIFIER_NONE;
}
static void memory_region_finalize(Object *obj)
{
MemoryRegion *mr = MEMORY_REGION(obj);
assert(!mr->container);
/* We know the region is not visible in any address space (it
* does not have a container and cannot be a root either because
* it has no references, so we can blindly clear mr->enabled.
* memory_region_set_enabled instead could trigger a transaction
* and cause an infinite loop.
*/
mr->enabled = false;
memory_region_transaction_begin();
while (!QTAILQ_EMPTY(&mr->subregions)) {
MemoryRegion *subregion = QTAILQ_FIRST(&mr->subregions);
memory_region_del_subregion(mr, subregion);
}
memory_region_transaction_commit();
mr->destructor(mr);
memory_region_clear_coalescing(mr);
g_free((char *)mr->name);
g_free(mr->ioeventfds);
}
Object *memory_region_owner(MemoryRegion *mr)
{
Object *obj = OBJECT(mr);
return obj->parent;
}
void memory_region_ref(MemoryRegion *mr)
{
/* MMIO callbacks most likely will access data that belongs
* to the owner, hence the need to ref/unref the owner whenever
* the memory region is in use.
*
* The memory region is a child of its owner. As long as the
* owner doesn't call unparent itself on the memory region,
* ref-ing the owner will also keep the memory region alive.
* Memory regions without an owner are supposed to never go away;
* we do not ref/unref them because it slows down DMA sensibly.
*/
if (mr && mr->owner) {
object_ref(mr->owner);
}
}
void memory_region_unref(MemoryRegion *mr)
{
if (mr && mr->owner) {
object_unref(mr->owner);
}
}
uint64_t memory_region_size(MemoryRegion *mr)
{
if (int128_eq(mr->size, int128_2_64())) {
return UINT64_MAX;
}
return int128_get64(mr->size);
}
const char *memory_region_name(const MemoryRegion *mr)
{
if (!mr->name) {
((MemoryRegion *)mr)->name =
object_get_canonical_path_component(OBJECT(mr));
}
return mr->name;
}
bool memory_region_is_ram_device(MemoryRegion *mr)
{
return mr->ram_device;
}
uint8_t memory_region_get_dirty_log_mask(MemoryRegion *mr)
{
uint8_t mask = mr->dirty_log_mask;
if (global_dirty_log && mr->ram_block) {
mask |= (1 << DIRTY_MEMORY_MIGRATION);
}
return mask;
}
bool memory_region_is_logging(MemoryRegion *mr, uint8_t client)
{
return memory_region_get_dirty_log_mask(mr) & (1 << client);
}
static void memory_region_update_iommu_notify_flags(IOMMUMemoryRegion *iommu_mr)
{
IOMMUNotifierFlag flags = IOMMU_NOTIFIER_NONE;
IOMMUNotifier *iommu_notifier;
IOMMUMemoryRegionClass *imrc = IOMMU_MEMORY_REGION_GET_CLASS(iommu_mr);
IOMMU_NOTIFIER_FOREACH(iommu_notifier, iommu_mr) {
flags |= iommu_notifier->notifier_flags;
}
if (flags != iommu_mr->iommu_notify_flags && imrc->notify_flag_changed) {
imrc->notify_flag_changed(iommu_mr,
iommu_mr->iommu_notify_flags,
flags);
}
iommu_mr->iommu_notify_flags = flags;
}
void memory_region_register_iommu_notifier(MemoryRegion *mr,
IOMMUNotifier *n)
{
IOMMUMemoryRegion *iommu_mr;
if (mr->alias) {
memory_region_register_iommu_notifier(mr->alias, n);
return;
}
/* We need to register for at least one bitfield */
iommu_mr = IOMMU_MEMORY_REGION(mr);
assert(n->notifier_flags != IOMMU_NOTIFIER_NONE);
assert(n->start <= n->end);
assert(n->iommu_idx >= 0 &&
n->iommu_idx < memory_region_iommu_num_indexes(iommu_mr));
QLIST_INSERT_HEAD(&iommu_mr->iommu_notify, n, node);
memory_region_update_iommu_notify_flags(iommu_mr);
}
uint64_t memory_region_iommu_get_min_page_size(IOMMUMemoryRegion *iommu_mr)
memory: Allow replay of IOMMU mapping notifications When we have guest visible IOMMUs, we allow notifiers to be registered which will be informed of all changes to IOMMU mappings. This is used by vfio to keep the host IOMMU mappings in sync with guest IOMMU mappings. However, unlike with a memory region listener, an iommu notifier won't be told about any mappings which already exist in the (guest) IOMMU at the time it is registered. This can cause problems if hotplugging a VFIO device onto a guest bus which had existing guest IOMMU mappings, but didn't previously have an VFIO devices (and hence no host IOMMU mappings). This adds a memory_region_iommu_replay() function to handle this case. It replays any existing mappings in an IOMMU memory region to a specified notifier. Because the IOMMU memory region doesn't internally remember the granularity of the guest IOMMU it has a small hack where the caller must specify a granularity at which to replay mappings. If there are finer mappings in the guest IOMMU these will be reported in the iotlb structures passed to the notifier which it must handle (probably causing it to flag an error). This isn't new - the VFIO iommu notifier must already handle notifications about guest IOMMU mappings too short for it to represent in the host IOMMU. Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Reviewed-by: Laurent Vivier <lvivier@redhat.com> Acked-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
2015-09-30 05:13:55 +03:00
{
IOMMUMemoryRegionClass *imrc = IOMMU_MEMORY_REGION_GET_CLASS(iommu_mr);
if (imrc->get_min_page_size) {
return imrc->get_min_page_size(iommu_mr);
}
return TARGET_PAGE_SIZE;
}
void memory_region_iommu_replay(IOMMUMemoryRegion *iommu_mr, IOMMUNotifier *n)
{
MemoryRegion *mr = MEMORY_REGION(iommu_mr);
IOMMUMemoryRegionClass *imrc = IOMMU_MEMORY_REGION_GET_CLASS(iommu_mr);
hwaddr addr, granularity;
memory: Allow replay of IOMMU mapping notifications When we have guest visible IOMMUs, we allow notifiers to be registered which will be informed of all changes to IOMMU mappings. This is used by vfio to keep the host IOMMU mappings in sync with guest IOMMU mappings. However, unlike with a memory region listener, an iommu notifier won't be told about any mappings which already exist in the (guest) IOMMU at the time it is registered. This can cause problems if hotplugging a VFIO device onto a guest bus which had existing guest IOMMU mappings, but didn't previously have an VFIO devices (and hence no host IOMMU mappings). This adds a memory_region_iommu_replay() function to handle this case. It replays any existing mappings in an IOMMU memory region to a specified notifier. Because the IOMMU memory region doesn't internally remember the granularity of the guest IOMMU it has a small hack where the caller must specify a granularity at which to replay mappings. If there are finer mappings in the guest IOMMU these will be reported in the iotlb structures passed to the notifier which it must handle (probably causing it to flag an error). This isn't new - the VFIO iommu notifier must already handle notifications about guest IOMMU mappings too short for it to represent in the host IOMMU. Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Reviewed-by: Laurent Vivier <lvivier@redhat.com> Acked-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
2015-09-30 05:13:55 +03:00
IOMMUTLBEntry iotlb;
/* If the IOMMU has its own replay callback, override */
if (imrc->replay) {
imrc->replay(iommu_mr, n);
return;
}
granularity = memory_region_iommu_get_min_page_size(iommu_mr);
memory: Allow replay of IOMMU mapping notifications When we have guest visible IOMMUs, we allow notifiers to be registered which will be informed of all changes to IOMMU mappings. This is used by vfio to keep the host IOMMU mappings in sync with guest IOMMU mappings. However, unlike with a memory region listener, an iommu notifier won't be told about any mappings which already exist in the (guest) IOMMU at the time it is registered. This can cause problems if hotplugging a VFIO device onto a guest bus which had existing guest IOMMU mappings, but didn't previously have an VFIO devices (and hence no host IOMMU mappings). This adds a memory_region_iommu_replay() function to handle this case. It replays any existing mappings in an IOMMU memory region to a specified notifier. Because the IOMMU memory region doesn't internally remember the granularity of the guest IOMMU it has a small hack where the caller must specify a granularity at which to replay mappings. If there are finer mappings in the guest IOMMU these will be reported in the iotlb structures passed to the notifier which it must handle (probably causing it to flag an error). This isn't new - the VFIO iommu notifier must already handle notifications about guest IOMMU mappings too short for it to represent in the host IOMMU. Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Reviewed-by: Laurent Vivier <lvivier@redhat.com> Acked-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
2015-09-30 05:13:55 +03:00
for (addr = 0; addr < memory_region_size(mr); addr += granularity) {
iotlb = imrc->translate(iommu_mr, addr, IOMMU_NONE, n->iommu_idx);
memory: Allow replay of IOMMU mapping notifications When we have guest visible IOMMUs, we allow notifiers to be registered which will be informed of all changes to IOMMU mappings. This is used by vfio to keep the host IOMMU mappings in sync with guest IOMMU mappings. However, unlike with a memory region listener, an iommu notifier won't be told about any mappings which already exist in the (guest) IOMMU at the time it is registered. This can cause problems if hotplugging a VFIO device onto a guest bus which had existing guest IOMMU mappings, but didn't previously have an VFIO devices (and hence no host IOMMU mappings). This adds a memory_region_iommu_replay() function to handle this case. It replays any existing mappings in an IOMMU memory region to a specified notifier. Because the IOMMU memory region doesn't internally remember the granularity of the guest IOMMU it has a small hack where the caller must specify a granularity at which to replay mappings. If there are finer mappings in the guest IOMMU these will be reported in the iotlb structures passed to the notifier which it must handle (probably causing it to flag an error). This isn't new - the VFIO iommu notifier must already handle notifications about guest IOMMU mappings too short for it to represent in the host IOMMU. Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Reviewed-by: Laurent Vivier <lvivier@redhat.com> Acked-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
2015-09-30 05:13:55 +03:00
if (iotlb.perm != IOMMU_NONE) {
n->notify(n, &iotlb);
}
/* if (2^64 - MR size) < granularity, it's possible to get an
* infinite loop here. This should catch such a wraparound */
if ((addr + granularity) < addr) {
break;
}
}
}
void memory_region_unregister_iommu_notifier(MemoryRegion *mr,
IOMMUNotifier *n)
{
IOMMUMemoryRegion *iommu_mr;
if (mr->alias) {
memory_region_unregister_iommu_notifier(mr->alias, n);
return;
}
QLIST_REMOVE(n, node);
iommu_mr = IOMMU_MEMORY_REGION(mr);
memory_region_update_iommu_notify_flags(iommu_mr);
}
void memory_region_notify_one(IOMMUNotifier *notifier,
IOMMUTLBEntry *entry)
{
IOMMUNotifierFlag request_flags;
hwaddr entry_end = entry->iova + entry->addr_mask;
/*
* Skip the notification if the notification does not overlap
* with registered range.
*/
if (notifier->start > entry_end || notifier->end < entry->iova) {
return;
}
assert(entry->iova >= notifier->start && entry_end <= notifier->end);
if (entry->perm & IOMMU_RW) {
request_flags = IOMMU_NOTIFIER_MAP;
} else {
request_flags = IOMMU_NOTIFIER_UNMAP;
}
if (notifier->notifier_flags & request_flags) {
notifier->notify(notifier, entry);
}
}
void memory_region_notify_iommu(IOMMUMemoryRegion *iommu_mr,
int iommu_idx,
IOMMUTLBEntry entry)
{
IOMMUNotifier *iommu_notifier;
assert(memory_region_is_iommu(MEMORY_REGION(iommu_mr)));
IOMMU_NOTIFIER_FOREACH(iommu_notifier, iommu_mr) {
if (iommu_notifier->iommu_idx == iommu_idx) {
memory_region_notify_one(iommu_notifier, &entry);
}
}
}
int memory_region_iommu_get_attr(IOMMUMemoryRegion *iommu_mr,
enum IOMMUMemoryRegionAttr attr,
void *data)
{
IOMMUMemoryRegionClass *imrc = IOMMU_MEMORY_REGION_GET_CLASS(iommu_mr);
if (!imrc->get_attr) {
return -EINVAL;
}
return imrc->get_attr(iommu_mr, attr, data);
}
int memory_region_iommu_attrs_to_index(IOMMUMemoryRegion *iommu_mr,
MemTxAttrs attrs)
{
IOMMUMemoryRegionClass *imrc = IOMMU_MEMORY_REGION_GET_CLASS(iommu_mr);
if (!imrc->attrs_to_index) {
return 0;
}
return imrc->attrs_to_index(iommu_mr, attrs);
}
int memory_region_iommu_num_indexes(IOMMUMemoryRegion *iommu_mr)
{
IOMMUMemoryRegionClass *imrc = IOMMU_MEMORY_REGION_GET_CLASS(iommu_mr);
if (!imrc->num_indexes) {
return 1;
}
return imrc->num_indexes(iommu_mr);
}
void memory_region_set_log(MemoryRegion *mr, bool log, unsigned client)
{
uint8_t mask = 1 << client;
uint8_t old_logging;
assert(client == DIRTY_MEMORY_VGA);
old_logging = mr->vga_logging_count;
mr->vga_logging_count += log ? 1 : -1;
if (!!old_logging == !!mr->vga_logging_count) {
return;
}
memory_region_transaction_begin();
mr->dirty_log_mask = (mr->dirty_log_mask & ~mask) | (log * mask);
memory_region_update_pending |= mr->enabled;
memory_region_transaction_commit();
}
void memory_region_set_dirty(MemoryRegion *mr, hwaddr addr,
hwaddr size)
{
assert(mr->ram_block);
cpu_physical_memory_set_dirty_range(memory_region_get_ram_addr(mr) + addr,
size,
memory_region_get_dirty_log_mask(mr));
}
static void memory_region_sync_dirty_bitmap(MemoryRegion *mr)
{
MemoryListener *listener;
AddressSpace *as;
FlatView *view;
FlatRange *fr;
/* If the same address space has multiple log_sync listeners, we
* visit that address space's FlatView multiple times. But because
* log_sync listeners are rare, it's still cheaper than walking each
* address space once.
*/
QTAILQ_FOREACH(listener, &memory_listeners, link) {
if (!listener->log_sync) {
continue;
}
as = listener->address_space;
view = address_space_get_flatview(as);
FOR_EACH_FLAT_RANGE(fr, view) {
if (fr->dirty_log_mask && (!mr || fr->mr == mr)) {
MemoryRegionSection mrs = section_from_flat_range(fr, view);
listener->log_sync(listener, &mrs);
}
}
flatview_unref(view);
}
}
memory: Introduce memory listener hook log_clear() Introduce a new memory region listener hook log_clear() to allow the listeners to hook onto the points where the dirty bitmap is cleared by the bitmap users. Previously log_sync() contains two operations: - dirty bitmap collection, and, - dirty bitmap clear on remote site. Let's take KVM as example - log_sync() for KVM will first copy the kernel dirty bitmap to userspace, and at the same time we'll clear the dirty bitmap there along with re-protecting all the guest pages again. We add this new log_clear() interface only to split the old log_sync() into two separated procedures: - use log_sync() to collect the collection only, and, - use log_clear() to clear the remote dirty bitmap. With the new interface, the memory listener users will still be able to decide how to implement the log synchronization procedure, e.g., they can still only provide log_sync() method only and put all the two procedures within log_sync() (that's how the old KVM works before KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2 is introduced). However with this new interface the memory listener users will start to have a chance to postpone the log clear operation explicitly if the module supports. That can really benefit users like KVM at least for host kernels that support KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2. There are three places that can clear dirty bits in any one of the dirty bitmap in the ram_list.dirty_memory[3] array: cpu_physical_memory_snapshot_and_clear_dirty cpu_physical_memory_test_and_clear_dirty cpu_physical_memory_sync_dirty_bitmap Currently we hook directly into each of the functions to notify about the log_clear(). Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Peter Xu <peterx@redhat.com> Message-Id: <20190603065056.25211-7-peterx@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
2019-06-03 09:50:51 +03:00
void memory_region_clear_dirty_bitmap(MemoryRegion *mr, hwaddr start,
hwaddr len)
{
MemoryRegionSection mrs;
MemoryListener *listener;
AddressSpace *as;
FlatView *view;
FlatRange *fr;
hwaddr sec_start, sec_end, sec_size;
QTAILQ_FOREACH(listener, &memory_listeners, link) {
if (!listener->log_clear) {
continue;
}
as = listener->address_space;
view = address_space_get_flatview(as);
FOR_EACH_FLAT_RANGE(fr, view) {
if (!fr->dirty_log_mask || fr->mr != mr) {
/*
* Clear dirty bitmap operation only applies to those
* regions whose dirty logging is at least enabled
*/
continue;
}
mrs = section_from_flat_range(fr, view);
sec_start = MAX(mrs.offset_within_region, start);
sec_end = mrs.offset_within_region + int128_get64(mrs.size);
sec_end = MIN(sec_end, start + len);
if (sec_start >= sec_end) {
/*
* If this memory region section has no intersection
* with the requested range, skip.
*/
continue;
}
/* Valid case; shrink the section if needed */
mrs.offset_within_address_space +=
sec_start - mrs.offset_within_region;
mrs.offset_within_region = sec_start;
sec_size = sec_end - sec_start;
mrs.size = int128_make64(sec_size);
listener->log_clear(listener, &mrs);
}
flatview_unref(view);
}
}
DirtyBitmapSnapshot *memory_region_snapshot_and_clear_dirty(MemoryRegion *mr,
hwaddr addr,
hwaddr size,
unsigned client)
{
DirtyBitmapSnapshot *snapshot;
assert(mr->ram_block);
memory_region_sync_dirty_bitmap(mr);
snapshot = cpu_physical_memory_snapshot_and_clear_dirty(mr, addr, size, client);
memory_global_after_dirty_log_sync();
return snapshot;
}
bool memory_region_snapshot_get_dirty(MemoryRegion *mr, DirtyBitmapSnapshot *snap,
hwaddr addr, hwaddr size)
{
assert(mr->ram_block);
return cpu_physical_memory_snapshot_get_dirty(snap,
memory_region_get_ram_addr(mr) + addr, size);
}
void memory_region_set_readonly(MemoryRegion *mr, bool readonly)
{
if (mr->readonly != readonly) {
memory_region_transaction_begin();
mr->readonly = readonly;
memory_region_update_pending |= mr->enabled;
memory_region_transaction_commit();
}
}
void memory_region_set_nonvolatile(MemoryRegion *mr, bool nonvolatile)
{
if (mr->nonvolatile != nonvolatile) {
memory_region_transaction_begin();
mr->nonvolatile = nonvolatile;
memory_region_update_pending |= mr->enabled;
memory_region_transaction_commit();
}
}
void memory_region_rom_device_set_romd(MemoryRegion *mr, bool romd_mode)
{
if (mr->romd_mode != romd_mode) {
memory_region_transaction_begin();
mr->romd_mode = romd_mode;
memory_region_update_pending |= mr->enabled;
memory_region_transaction_commit();
}
}
void memory_region_reset_dirty(MemoryRegion *mr, hwaddr addr,
hwaddr size, unsigned client)
{
assert(mr->ram_block);
cpu_physical_memory_test_and_clear_dirty(
memory_region_get_ram_addr(mr) + addr, size, client);
}
int memory_region_get_fd(MemoryRegion *mr)
{
int fd;
rcu_read_lock();
while (mr->alias) {
mr = mr->alias;
}
fd = mr->ram_block->fd;
rcu_read_unlock();
return fd;
}
void *memory_region_get_ram_ptr(MemoryRegion *mr)
{
void *ptr;
uint64_t offset = 0;
rcu_read_lock();
while (mr->alias) {
offset += mr->alias_offset;
mr = mr->alias;
}
assert(mr->ram_block);
ptr = qemu_map_ram_ptr(mr->ram_block, offset);
rcu_read_unlock();
return ptr;
}
MemoryRegion *memory_region_from_host(void *ptr, ram_addr_t *offset)
{
RAMBlock *block;
block = qemu_ram_block_from_host(ptr, false, offset);
if (!block) {
return NULL;
}
return block->mr;
}
ram_addr_t memory_region_get_ram_addr(MemoryRegion *mr)
{
return mr->ram_block ? mr->ram_block->offset : RAM_ADDR_INVALID;
}
void memory_region_ram_resize(MemoryRegion *mr, ram_addr_t newsize, Error **errp)
{
assert(mr->ram_block);
qemu_ram_resize(mr->ram_block, newsize, errp);
}
/*
* Call proper memory listeners about the change on the newly
* added/removed CoalescedMemoryRange.
*/
static void memory_region_update_coalesced_range(MemoryRegion *mr,
CoalescedMemoryRange *cmr,
bool add)
{
AddressSpace *as;
FlatView *view;
FlatRange *fr;
QTAILQ_FOREACH(as, &address_spaces, address_spaces_link) {
view = address_space_get_flatview(as);
FOR_EACH_FLAT_RANGE(fr, view) {
if (fr->mr == mr) {
flat_range_coalesced_io_notify(fr, as, cmr, add);
}
}
flatview_unref(view);
}
}
void memory_region_set_coalescing(MemoryRegion *mr)
{
memory_region_clear_coalescing(mr);
memory_region_add_coalescing(mr, 0, int128_get64(mr->size));
}
void memory_region_add_coalescing(MemoryRegion *mr,
hwaddr offset,
uint64_t size)
{
CoalescedMemoryRange *cmr = g_malloc(sizeof(*cmr));
cmr->addr = addrrange_make(int128_make64(offset), int128_make64(size));
QTAILQ_INSERT_TAIL(&mr->coalesced, cmr, link);
memory_region_update_coalesced_range(mr, cmr, true);
memory_region_set_flush_coalesced(mr);
}
void memory_region_clear_coalescing(MemoryRegion *mr)
{
CoalescedMemoryRange *cmr;
if (QTAILQ_EMPTY(&mr->coalesced)) {
return;
}
qemu_flush_coalesced_mmio_buffer();
mr->flush_coalesced_mmio = false;
while (!QTAILQ_EMPTY(&mr->coalesced)) {
cmr = QTAILQ_FIRST(&mr->coalesced);
QTAILQ_REMOVE(&mr->coalesced, cmr, link);
memory_region_update_coalesced_range(mr, cmr, false);
g_free(cmr);
}
}
void memory_region_set_flush_coalesced(MemoryRegion *mr)
{
mr->flush_coalesced_mmio = true;
}
void memory_region_clear_flush_coalesced(MemoryRegion *mr)
{
qemu_flush_coalesced_mmio_buffer();
if (QTAILQ_EMPTY(&mr->coalesced)) {
mr->flush_coalesced_mmio = false;
}
}
void memory_region_clear_global_locking(MemoryRegion *mr)
{
mr->global_locking = false;
}
static bool userspace_eventfd_warning;
void memory_region_add_eventfd(MemoryRegion *mr,
hwaddr addr,
unsigned size,
bool match_data,
uint64_t data,
EventNotifier *e)
{
MemoryRegionIoeventfd mrfd = {
.addr.start = int128_make64(addr),
.addr.size = int128_make64(size),
.match_data = match_data,
.data = data,
.e = e,
};
unsigned i;
if (kvm_enabled() && (!(kvm_eventfds_enabled() ||
userspace_eventfd_warning))) {
userspace_eventfd_warning = true;
error_report("Using eventfd without MMIO binding in KVM. "
"Suboptimal performance expected");
}
if (size) {
adjust_endianness(mr, &mrfd.data, size_memop(size) | MO_TE);
}
memory_region_transaction_begin();
for (i = 0; i < mr->ioeventfd_nb; ++i) {
if (memory_region_ioeventfd_before(&mrfd, &mr->ioeventfds[i])) {
break;
}
}
++mr->ioeventfd_nb;
mr->ioeventfds = g_realloc(mr->ioeventfds,
sizeof(*mr->ioeventfds) * mr->ioeventfd_nb);
memmove(&mr->ioeventfds[i+1], &mr->ioeventfds[i],
sizeof(*mr->ioeventfds) * (mr->ioeventfd_nb-1 - i));
mr->ioeventfds[i] = mrfd;
ioeventfd_update_pending |= mr->enabled;
memory_region_transaction_commit();
}
void memory_region_del_eventfd(MemoryRegion *mr,
hwaddr addr,
unsigned size,
bool match_data,
uint64_t data,
EventNotifier *e)
{
MemoryRegionIoeventfd mrfd = {
.addr.start = int128_make64(addr),
.addr.size = int128_make64(size),
.match_data = match_data,
.data = data,
.e = e,
};
unsigned i;
if (size) {
adjust_endianness(mr, &mrfd.data, size_memop(size) | MO_TE);
}
memory_region_transaction_begin();
for (i = 0; i < mr->ioeventfd_nb; ++i) {
if (memory_region_ioeventfd_equal(&mrfd, &mr->ioeventfds[i])) {
break;
}
}
assert(i != mr->ioeventfd_nb);
memmove(&mr->ioeventfds[i], &mr->ioeventfds[i+1],
sizeof(*mr->ioeventfds) * (mr->ioeventfd_nb - (i+1)));
--mr->ioeventfd_nb;
mr->ioeventfds = g_realloc(mr->ioeventfds,
sizeof(*mr->ioeventfds)*mr->ioeventfd_nb + 1);
ioeventfd_update_pending |= mr->enabled;
memory_region_transaction_commit();
}
static void memory_region_update_container_subregions(MemoryRegion *subregion)
{
MemoryRegion *mr = subregion->container;
MemoryRegion *other;
memory_region_transaction_begin();
memory_region_ref(subregion);
QTAILQ_FOREACH(other, &mr->subregions, subregions_link) {
if (subregion->priority >= other->priority) {
QTAILQ_INSERT_BEFORE(other, subregion, subregions_link);
goto done;
}
}
QTAILQ_INSERT_TAIL(&mr->subregions, subregion, subregions_link);
done:
memory_region_update_pending |= mr->enabled && subregion->enabled;
memory_region_transaction_commit();
}
static void memory_region_add_subregion_common(MemoryRegion *mr,
hwaddr offset,
MemoryRegion *subregion)
{
assert(!subregion->container);
subregion->container = mr;
subregion->addr = offset;
memory_region_update_container_subregions(subregion);
}
void memory_region_add_subregion(MemoryRegion *mr,
hwaddr offset,
MemoryRegion *subregion)
{
subregion->priority = 0;
memory_region_add_subregion_common(mr, offset, subregion);
}
void memory_region_add_subregion_overlap(MemoryRegion *mr,
hwaddr offset,
MemoryRegion *subregion,
int priority)
{
subregion->priority = priority;
memory_region_add_subregion_common(mr, offset, subregion);
}
void memory_region_del_subregion(MemoryRegion *mr,
MemoryRegion *subregion)
{
memory_region_transaction_begin();
assert(subregion->container == mr);
subregion->container = NULL;
QTAILQ_REMOVE(&mr->subregions, subregion, subregions_link);
memory_region_unref(subregion);
memory_region_update_pending |= mr->enabled && subregion->enabled;
memory_region_transaction_commit();
}
void memory_region_set_enabled(MemoryRegion *mr, bool enabled)
{
if (enabled == mr->enabled) {
return;
}
memory_region_transaction_begin();
mr->enabled = enabled;
memory_region_update_pending = true;
memory_region_transaction_commit();
}
void memory_region_set_size(MemoryRegion *mr, uint64_t size)
{
Int128 s = int128_make64(size);
if (size == UINT64_MAX) {
s = int128_2_64();
}
if (int128_eq(s, mr->size)) {
return;
}
memory_region_transaction_begin();
mr->size = s;
memory_region_update_pending = true;
memory_region_transaction_commit();
}
static void memory_region_readd_subregion(MemoryRegion *mr)
{
MemoryRegion *container = mr->container;
if (container) {
memory_region_transaction_begin();
memory_region_ref(mr);
memory_region_del_subregion(container, mr);
mr->container = container;
memory_region_update_container_subregions(mr);
memory_region_unref(mr);
memory_region_transaction_commit();
}
}
void memory_region_set_address(MemoryRegion *mr, hwaddr addr)
{
if (addr != mr->addr) {
mr->addr = addr;
memory_region_readd_subregion(mr);
}
}
void memory_region_set_alias_offset(MemoryRegion *mr, hwaddr offset)
{
assert(mr->alias);
if (offset == mr->alias_offset) {
return;
}
memory_region_transaction_begin();
mr->alias_offset = offset;
memory_region_update_pending |= mr->enabled;
memory_region_transaction_commit();
}
uint64_t memory_region_get_alignment(const MemoryRegion *mr)
{
return mr->align;
}
static int cmp_flatrange_addr(const void *addr_, const void *fr_)
{
const AddrRange *addr = addr_;
const FlatRange *fr = fr_;
if (int128_le(addrrange_end(*addr), fr->addr.start)) {
return -1;
} else if (int128_ge(addr->start, addrrange_end(fr->addr))) {
return 1;
}
return 0;
}
static FlatRange *flatview_lookup(FlatView *view, AddrRange addr)
{
return bsearch(&addr, view->ranges, view->nr,
sizeof(FlatRange), cmp_flatrange_addr);
}
bool memory_region_is_mapped(MemoryRegion *mr)
{
return mr->container ? true : false;
}
/* Same as memory_region_find, but it does not add a reference to the
* returned region. It must be called from an RCU critical section.
*/
static MemoryRegionSection memory_region_find_rcu(MemoryRegion *mr,
hwaddr addr, uint64_t size)
{
MemoryRegionSection ret = { .mr = NULL };
MemoryRegion *root;
AddressSpace *as;
AddrRange range;
FlatView *view;
FlatRange *fr;
addr += mr->addr;
for (root = mr; root->container; ) {
root = root->container;
addr += root->addr;
}
as = memory_region_to_address_space(root);
if (!as) {
return ret;
}
range = addrrange_make(int128_make64(addr), int128_make64(size));
view = address_space_to_flatview(as);
fr = flatview_lookup(view, range);
if (!fr) {
return ret;
}
while (fr > view->ranges && addrrange_intersects(fr[-1].addr, range)) {
--fr;
}
ret.mr = fr->mr;
ret.fv = view;
range = addrrange_intersection(range, fr->addr);
ret.offset_within_region = fr->offset_in_region;
ret.offset_within_region += int128_get64(int128_sub(range.start,
fr->addr.start));
ret.size = range.size;
ret.offset_within_address_space = int128_get64(range.start);
ret.readonly = fr->readonly;
ret.nonvolatile = fr->nonvolatile;
return ret;
}
MemoryRegionSection memory_region_find(MemoryRegion *mr,
hwaddr addr, uint64_t size)
{
MemoryRegionSection ret;
rcu_read_lock();
ret = memory_region_find_rcu(mr, addr, size);
if (ret.mr) {
memory_region_ref(ret.mr);
}
rcu_read_unlock();
return ret;
}
bool memory_region_present(MemoryRegion *container, hwaddr addr)
{
MemoryRegion *mr;
rcu_read_lock();
mr = memory_region_find_rcu(container, addr, 1).mr;
rcu_read_unlock();
return mr && mr != container;
}
void memory_global_dirty_log_sync(void)
{
memory_region_sync_dirty_bitmap(NULL);
}
void memory_global_after_dirty_log_sync(void)
{
MEMORY_LISTENER_CALL_GLOBAL(log_global_after_sync, Forward);
}
static VMChangeStateEntry *vmstate_change;
void memory_global_dirty_log_start(void)
{
if (vmstate_change) {
qemu_del_vm_change_state_handler(vmstate_change);
vmstate_change = NULL;
}
global_dirty_log = true;
MEMORY_LISTENER_CALL_GLOBAL(log_global_start, Forward);
/* Refresh DIRTY_MEMORY_MIGRATION bit. */
memory_region_transaction_begin();
memory_region_update_pending = true;
memory_region_transaction_commit();
}
static void memory_global_dirty_log_do_stop(void)
{
global_dirty_log = false;
/* Refresh DIRTY_MEMORY_MIGRATION bit. */
memory_region_transaction_begin();
memory_region_update_pending = true;
memory_region_transaction_commit();
MEMORY_LISTENER_CALL_GLOBAL(log_global_stop, Reverse);
}
static void memory_vm_change_state_handler(void *opaque, int running,
RunState state)
{
if (running) {
memory_global_dirty_log_do_stop();
if (vmstate_change) {
qemu_del_vm_change_state_handler(vmstate_change);
vmstate_change = NULL;
}
}
}
void memory_global_dirty_log_stop(void)
{
if (!runstate_is_running()) {
if (vmstate_change) {
return;
}
vmstate_change = qemu_add_vm_change_state_handler(
memory_vm_change_state_handler, NULL);
return;
}
memory_global_dirty_log_do_stop();
}
static void listener_add_address_space(MemoryListener *listener,
AddressSpace *as)
{
FlatView *view;
FlatRange *fr;
if (listener->begin) {
listener->begin(listener);
}
if (global_dirty_log) {
if (listener->log_global_start) {
listener->log_global_start(listener);
}
}
view = address_space_get_flatview(as);
FOR_EACH_FLAT_RANGE(fr, view) {
MemoryRegionSection section = section_from_flat_range(fr, view);
if (listener->region_add) {
listener->region_add(listener, &section);
}
if (fr->dirty_log_mask && listener->log_start) {
listener->log_start(listener, &section, 0, fr->dirty_log_mask);
}
}
if (listener->commit) {
listener->commit(listener);
}
flatview_unref(view);
}
static void listener_del_address_space(MemoryListener *listener,
AddressSpace *as)
{
FlatView *view;
FlatRange *fr;
if (listener->begin) {
listener->begin(listener);
}
view = address_space_get_flatview(as);
FOR_EACH_FLAT_RANGE(fr, view) {
MemoryRegionSection section = section_from_flat_range(fr, view);
if (fr->dirty_log_mask && listener->log_stop) {
listener->log_stop(listener, &section, fr->dirty_log_mask, 0);
}
if (listener->region_del) {
listener->region_del(listener, &section);
}
}
if (listener->commit) {
listener->commit(listener);
}
flatview_unref(view);
}
void memory_listener_register(MemoryListener *listener, AddressSpace *as)
{
MemoryListener *other = NULL;
listener->address_space = as;
if (QTAILQ_EMPTY(&memory_listeners)
|| listener->priority >= QTAILQ_LAST(&memory_listeners)->priority) {
QTAILQ_INSERT_TAIL(&memory_listeners, listener, link);
} else {
QTAILQ_FOREACH(other, &memory_listeners, link) {
if (listener->priority < other->priority) {
break;
}
}
QTAILQ_INSERT_BEFORE(other, listener, link);
}
if (QTAILQ_EMPTY(&as->listeners)
|| listener->priority >= QTAILQ_LAST(&as->listeners)->priority) {
QTAILQ_INSERT_TAIL(&as->listeners, listener, link_as);
} else {
QTAILQ_FOREACH(other, &as->listeners, link_as) {
if (listener->priority < other->priority) {
break;
}
}
QTAILQ_INSERT_BEFORE(other, listener, link_as);
}
listener_add_address_space(listener, as);
}
void memory_listener_unregister(MemoryListener *listener)
{
if (!listener->address_space) {
return;
}
listener_del_address_space(listener, listener->address_space);
QTAILQ_REMOVE(&memory_listeners, listener, link);
QTAILQ_REMOVE(&listener->address_space->listeners, listener, link_as);
listener->address_space = NULL;
}
2019-06-21 12:27:33 +03:00
void address_space_remove_listeners(AddressSpace *as)
{
while (!QTAILQ_EMPTY(&as->listeners)) {
memory_listener_unregister(QTAILQ_FIRST(&as->listeners));
}
}
void address_space_init(AddressSpace *as, MemoryRegion *root, const char *name)
{
memory_region_ref(root);
as->root = root;
as->current_map = NULL;
as->ioeventfd_nb = 0;
as->ioeventfds = NULL;
QTAILQ_INIT(&as->listeners);
QTAILQ_INSERT_TAIL(&address_spaces, as, address_spaces_link);
as->name = g_strdup(name ? name : "anonymous");
address_space_update_topology(as);
address_space_update_ioeventfds(as);
}
static void do_address_space_destroy(AddressSpace *as)
{
assert(QTAILQ_EMPTY(&as->listeners));
flatview_unref(as->current_map);
g_free(as->name);
g_free(as->ioeventfds);
memory_region_unref(as->root);
}
void address_space_destroy(AddressSpace *as)
{
MemoryRegion *root = as->root;
/* Flush out anything from MemoryListeners listening in on this */
memory_region_transaction_begin();
as->root = NULL;
memory_region_transaction_commit();
QTAILQ_REMOVE(&address_spaces, as, address_spaces_link);
/* At this point, as->dispatch and as->current_map are dummy
* entries that the guest should never use. Wait for the old
* values to expire before freeing the data.
*/
as->root = root;
call_rcu(as, do_address_space_destroy, rcu);
}
static const char *memory_region_type(MemoryRegion *mr)
{
if (memory_region_is_ram_device(mr)) {
return "ramd";
} else if (memory_region_is_romd(mr)) {
return "romd";
} else if (memory_region_is_rom(mr)) {
return "rom";
} else if (memory_region_is_ram(mr)) {
return "ram";
} else {
return "i/o";
}
}
typedef struct MemoryRegionList MemoryRegionList;
struct MemoryRegionList {
const MemoryRegion *mr;
QTAILQ_ENTRY(MemoryRegionList) mrqueue;
};
typedef QTAILQ_HEAD(, MemoryRegionList) MemoryRegionListHead;
#define MR_SIZE(size) (int128_nz(size) ? (hwaddr)int128_get64( \
int128_sub((size), int128_one())) : 0)
#define MTREE_INDENT " "
static void mtree_expand_owner(const char *label, Object *obj)
{
DeviceState *dev = (DeviceState *) object_dynamic_cast(obj, TYPE_DEVICE);
qemu_printf(" %s:{%s", label, dev ? "dev" : "obj");
if (dev && dev->id) {
qemu_printf(" id=%s", dev->id);
} else {
gchar *canonical_path = object_get_canonical_path(obj);
if (canonical_path) {
qemu_printf(" path=%s", canonical_path);
g_free(canonical_path);
} else {
qemu_printf(" type=%s", object_get_typename(obj));
}
}
qemu_printf("}");
}
static void mtree_print_mr_owner(const MemoryRegion *mr)
{
Object *owner = mr->owner;
Object *parent = memory_region_owner((MemoryRegion *)mr);
if (!owner && !parent) {
qemu_printf(" orphan");
return;
}
if (owner) {
mtree_expand_owner("owner", owner);
}
if (parent && parent != owner) {
mtree_expand_owner("parent", parent);
}
}
static void mtree_print_mr(const MemoryRegion *mr, unsigned int level,
hwaddr base,
MemoryRegionListHead *alias_print_queue,
bool owner)
{
MemoryRegionList *new_ml, *ml, *next_ml;
MemoryRegionListHead submr_print_queue;
const MemoryRegion *submr;
unsigned int i;
hwaddr cur_start, cur_end;
if (!mr) {
return;
}
for (i = 0; i < level; i++) {
qemu_printf(MTREE_INDENT);
}
cur_start = base + mr->addr;
cur_end = cur_start + MR_SIZE(mr->size);
/*
* Try to detect overflow of memory region. This should never
* happen normally. When it happens, we dump something to warn the
* user who is observing this.
*/
if (cur_start < base || cur_end < cur_start) {
qemu_printf("[DETECTED OVERFLOW!] ");
}
if (mr->alias) {
MemoryRegionList *ml;
bool found = false;
/* check if the alias is already in the queue */
QTAILQ_FOREACH(ml, alias_print_queue, mrqueue) {
if (ml->mr == mr->alias) {
found = true;
}
}
if (!found) {
ml = g_new(MemoryRegionList, 1);
ml->mr = mr->alias;
QTAILQ_INSERT_TAIL(alias_print_queue, ml, mrqueue);
}
qemu_printf(TARGET_FMT_plx "-" TARGET_FMT_plx
" (prio %d, %s%s): alias %s @%s " TARGET_FMT_plx
"-" TARGET_FMT_plx "%s",
cur_start, cur_end,
mr->priority,
mr->nonvolatile ? "nv-" : "",
memory_region_type((MemoryRegion *)mr),
memory_region_name(mr),
memory_region_name(mr->alias),
mr->alias_offset,
mr->alias_offset + MR_SIZE(mr->size),
mr->enabled ? "" : " [disabled]");
if (owner) {
mtree_print_mr_owner(mr);
}
} else {
qemu_printf(TARGET_FMT_plx "-" TARGET_FMT_plx
" (prio %d, %s%s): %s%s",
cur_start, cur_end,
mr->priority,
mr->nonvolatile ? "nv-" : "",
memory_region_type((MemoryRegion *)mr),
memory_region_name(mr),
mr->enabled ? "" : " [disabled]");
if (owner) {
mtree_print_mr_owner(mr);
}
}
qemu_printf("\n");
QTAILQ_INIT(&submr_print_queue);
QTAILQ_FOREACH(submr, &mr->subregions, subregions_link) {
new_ml = g_new(MemoryRegionList, 1);
new_ml->mr = submr;
QTAILQ_FOREACH(ml, &submr_print_queue, mrqueue) {
if (new_ml->mr->addr < ml->mr->addr ||
(new_ml->mr->addr == ml->mr->addr &&
new_ml->mr->priority > ml->mr->priority)) {
QTAILQ_INSERT_BEFORE(ml, new_ml, mrqueue);
new_ml = NULL;
break;
}
}
if (new_ml) {
QTAILQ_INSERT_TAIL(&submr_print_queue, new_ml, mrqueue);
}
}
QTAILQ_FOREACH(ml, &submr_print_queue, mrqueue) {
mtree_print_mr(ml->mr, level + 1, cur_start,
alias_print_queue, owner);
}
QTAILQ_FOREACH_SAFE(ml, &submr_print_queue, mrqueue, next_ml) {
g_free(ml);
}
}
struct FlatViewInfo {
int counter;
bool dispatch_tree;
bool owner;
AccelClass *ac;
const char *ac_name;
};
static void mtree_print_flatview(gpointer key, gpointer value,
gpointer user_data)
{
FlatView *view = key;
GArray *fv_address_spaces = value;
struct FlatViewInfo *fvi = user_data;
FlatRange *range = &view->ranges[0];
MemoryRegion *mr;
int n = view->nr;
int i;
AddressSpace *as;
qemu_printf("FlatView #%d\n", fvi->counter);
++fvi->counter;
for (i = 0; i < fv_address_spaces->len; ++i) {
as = g_array_index(fv_address_spaces, AddressSpace*, i);
qemu_printf(" AS \"%s\", root: %s",
as->name, memory_region_name(as->root));
if (as->root->alias) {
qemu_printf(", alias %s", memory_region_name(as->root->alias));
}
qemu_printf("\n");
}
qemu_printf(" Root memory region: %s\n",
view->root ? memory_region_name(view->root) : "(none)");
if (n <= 0) {
qemu_printf(MTREE_INDENT "No rendered FlatView\n\n");
return;
}
while (n--) {
mr = range->mr;
memory: show region offset and ROM/RAM type in "info mtree -f" "info mtree -f" output is currently hard to use for large RAM regions, because there is no hint as to what part of the region is being mapped. Add the offset if it is nonzero. Secondly, FlatView has a readonly field, that can override the MemoryRegion in the presence of aliases. Take it into account. Together, with this patch this: address-space (flat view): KVM-SMRAM 0000000000000000-00000000000bffff (prio 0, ram): pc.ram 00000000000c0000-00000000000c9fff (prio 0, ram): pc.ram 00000000000ca000-00000000000ccfff (prio 0, ram): pc.ram 00000000000cd000-00000000000ebfff (prio 0, ram): pc.ram 00000000000ec000-00000000000effff (prio 0, ram): pc.ram 00000000000f0000-00000000000fffff (prio 0, ram): pc.ram 0000000000100000-00000000bfffffff (prio 0, ram): pc.ram 00000000fd000000-00000000fdffffff (prio 1, ram): vga.vram 00000000febc0000-00000000febdffff (prio 1, i/o): e1000-mmio 00000000febf0400-00000000febf041f (prio 0, i/o): vga ioports remapped 00000000febf0500-00000000febf0515 (prio 0, i/o): bochs dispi interface 00000000febf0600-00000000febf0607 (prio 0, i/o): qemu extended regs 00000000fec00000-00000000fec00fff (prio 0, i/o): kvm-ioapic 00000000fed00000-00000000fed003ff (prio 0, i/o): hpet 00000000fee00000-00000000feefffff (prio 4096, i/o): kvm-apic-msi 00000000fffc0000-00000000ffffffff (prio 0, rom): pc.bios 0000000100000000-000000013fffffff (prio 0, ram): pc.ram becomes this: address-space (flat view): KVM-SMRAM 0000000000000000-00000000000bffff (prio 0, ram): pc.ram 00000000000c0000-00000000000c9fff (prio 0, rom): pc.ram @00000000000c0000 00000000000ca000-00000000000ccfff (prio 0, ram): pc.ram @00000000000ca000 00000000000cd000-00000000000ebfff (prio 0, rom): pc.ram @00000000000cd000 00000000000ec000-00000000000effff (prio 0, ram): pc.ram @00000000000ec000 00000000000f0000-00000000000fffff (prio 0, rom): pc.ram @00000000000f0000 0000000000100000-00000000bfffffff (prio 0, ram): pc.ram @0000000000100000 00000000fd000000-00000000fdffffff (prio 1, ram): vga.vram 00000000febc0000-00000000febdffff (prio 1, i/o): e1000-mmio 00000000febf0400-00000000febf041f (prio 0, i/o): vga ioports remapped 00000000febf0500-00000000febf0515 (prio 0, i/o): bochs dispi interface 00000000febf0600-00000000febf0607 (prio 0, i/o): qemu extended regs 00000000fec00000-00000000fec00fff (prio 0, i/o): kvm-ioapic 00000000fed00000-00000000fed003ff (prio 0, i/o): hpet 00000000fee00000-00000000feefffff (prio 4096, i/o): kvm-apic-msi 00000000fffc0000-00000000ffffffff (prio 0, rom): pc.bios 0000000100000000-000000013fffffff (prio 0, ram): pc.ram @00000000c0000000 This should make it easier to understand what's going on. Cc: Peter Xu <peterx@redhat.com> Cc: "William Tambe" <tambewilliam@gmail.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2017-03-03 00:49:41 +03:00
if (range->offset_in_region) {
qemu_printf(MTREE_INDENT TARGET_FMT_plx "-" TARGET_FMT_plx
" (prio %d, %s%s): %s @" TARGET_FMT_plx,
int128_get64(range->addr.start),
int128_get64(range->addr.start)
+ MR_SIZE(range->addr.size),
mr->priority,
range->nonvolatile ? "nv-" : "",
range->readonly ? "rom" : memory_region_type(mr),
memory_region_name(mr),
range->offset_in_region);
memory: show region offset and ROM/RAM type in "info mtree -f" "info mtree -f" output is currently hard to use for large RAM regions, because there is no hint as to what part of the region is being mapped. Add the offset if it is nonzero. Secondly, FlatView has a readonly field, that can override the MemoryRegion in the presence of aliases. Take it into account. Together, with this patch this: address-space (flat view): KVM-SMRAM 0000000000000000-00000000000bffff (prio 0, ram): pc.ram 00000000000c0000-00000000000c9fff (prio 0, ram): pc.ram 00000000000ca000-00000000000ccfff (prio 0, ram): pc.ram 00000000000cd000-00000000000ebfff (prio 0, ram): pc.ram 00000000000ec000-00000000000effff (prio 0, ram): pc.ram 00000000000f0000-00000000000fffff (prio 0, ram): pc.ram 0000000000100000-00000000bfffffff (prio 0, ram): pc.ram 00000000fd000000-00000000fdffffff (prio 1, ram): vga.vram 00000000febc0000-00000000febdffff (prio 1, i/o): e1000-mmio 00000000febf0400-00000000febf041f (prio 0, i/o): vga ioports remapped 00000000febf0500-00000000febf0515 (prio 0, i/o): bochs dispi interface 00000000febf0600-00000000febf0607 (prio 0, i/o): qemu extended regs 00000000fec00000-00000000fec00fff (prio 0, i/o): kvm-ioapic 00000000fed00000-00000000fed003ff (prio 0, i/o): hpet 00000000fee00000-00000000feefffff (prio 4096, i/o): kvm-apic-msi 00000000fffc0000-00000000ffffffff (prio 0, rom): pc.bios 0000000100000000-000000013fffffff (prio 0, ram): pc.ram becomes this: address-space (flat view): KVM-SMRAM 0000000000000000-00000000000bffff (prio 0, ram): pc.ram 00000000000c0000-00000000000c9fff (prio 0, rom): pc.ram @00000000000c0000 00000000000ca000-00000000000ccfff (prio 0, ram): pc.ram @00000000000ca000 00000000000cd000-00000000000ebfff (prio 0, rom): pc.ram @00000000000cd000 00000000000ec000-00000000000effff (prio 0, ram): pc.ram @00000000000ec000 00000000000f0000-00000000000fffff (prio 0, rom): pc.ram @00000000000f0000 0000000000100000-00000000bfffffff (prio 0, ram): pc.ram @0000000000100000 00000000fd000000-00000000fdffffff (prio 1, ram): vga.vram 00000000febc0000-00000000febdffff (prio 1, i/o): e1000-mmio 00000000febf0400-00000000febf041f (prio 0, i/o): vga ioports remapped 00000000febf0500-00000000febf0515 (prio 0, i/o): bochs dispi interface 00000000febf0600-00000000febf0607 (prio 0, i/o): qemu extended regs 00000000fec00000-00000000fec00fff (prio 0, i/o): kvm-ioapic 00000000fed00000-00000000fed003ff (prio 0, i/o): hpet 00000000fee00000-00000000feefffff (prio 4096, i/o): kvm-apic-msi 00000000fffc0000-00000000ffffffff (prio 0, rom): pc.bios 0000000100000000-000000013fffffff (prio 0, ram): pc.ram @00000000c0000000 This should make it easier to understand what's going on. Cc: Peter Xu <peterx@redhat.com> Cc: "William Tambe" <tambewilliam@gmail.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2017-03-03 00:49:41 +03:00
} else {
qemu_printf(MTREE_INDENT TARGET_FMT_plx "-" TARGET_FMT_plx
" (prio %d, %s%s): %s",
int128_get64(range->addr.start),
int128_get64(range->addr.start)
+ MR_SIZE(range->addr.size),
mr->priority,
range->nonvolatile ? "nv-" : "",
range->readonly ? "rom" : memory_region_type(mr),
memory_region_name(mr));
memory: show region offset and ROM/RAM type in "info mtree -f" "info mtree -f" output is currently hard to use for large RAM regions, because there is no hint as to what part of the region is being mapped. Add the offset if it is nonzero. Secondly, FlatView has a readonly field, that can override the MemoryRegion in the presence of aliases. Take it into account. Together, with this patch this: address-space (flat view): KVM-SMRAM 0000000000000000-00000000000bffff (prio 0, ram): pc.ram 00000000000c0000-00000000000c9fff (prio 0, ram): pc.ram 00000000000ca000-00000000000ccfff (prio 0, ram): pc.ram 00000000000cd000-00000000000ebfff (prio 0, ram): pc.ram 00000000000ec000-00000000000effff (prio 0, ram): pc.ram 00000000000f0000-00000000000fffff (prio 0, ram): pc.ram 0000000000100000-00000000bfffffff (prio 0, ram): pc.ram 00000000fd000000-00000000fdffffff (prio 1, ram): vga.vram 00000000febc0000-00000000febdffff (prio 1, i/o): e1000-mmio 00000000febf0400-00000000febf041f (prio 0, i/o): vga ioports remapped 00000000febf0500-00000000febf0515 (prio 0, i/o): bochs dispi interface 00000000febf0600-00000000febf0607 (prio 0, i/o): qemu extended regs 00000000fec00000-00000000fec00fff (prio 0, i/o): kvm-ioapic 00000000fed00000-00000000fed003ff (prio 0, i/o): hpet 00000000fee00000-00000000feefffff (prio 4096, i/o): kvm-apic-msi 00000000fffc0000-00000000ffffffff (prio 0, rom): pc.bios 0000000100000000-000000013fffffff (prio 0, ram): pc.ram becomes this: address-space (flat view): KVM-SMRAM 0000000000000000-00000000000bffff (prio 0, ram): pc.ram 00000000000c0000-00000000000c9fff (prio 0, rom): pc.ram @00000000000c0000 00000000000ca000-00000000000ccfff (prio 0, ram): pc.ram @00000000000ca000 00000000000cd000-00000000000ebfff (prio 0, rom): pc.ram @00000000000cd000 00000000000ec000-00000000000effff (prio 0, ram): pc.ram @00000000000ec000 00000000000f0000-00000000000fffff (prio 0, rom): pc.ram @00000000000f0000 0000000000100000-00000000bfffffff (prio 0, ram): pc.ram @0000000000100000 00000000fd000000-00000000fdffffff (prio 1, ram): vga.vram 00000000febc0000-00000000febdffff (prio 1, i/o): e1000-mmio 00000000febf0400-00000000febf041f (prio 0, i/o): vga ioports remapped 00000000febf0500-00000000febf0515 (prio 0, i/o): bochs dispi interface 00000000febf0600-00000000febf0607 (prio 0, i/o): qemu extended regs 00000000fec00000-00000000fec00fff (prio 0, i/o): kvm-ioapic 00000000fed00000-00000000fed003ff (prio 0, i/o): hpet 00000000fee00000-00000000feefffff (prio 4096, i/o): kvm-apic-msi 00000000fffc0000-00000000ffffffff (prio 0, rom): pc.bios 0000000100000000-000000013fffffff (prio 0, ram): pc.ram @00000000c0000000 This should make it easier to understand what's going on. Cc: Peter Xu <peterx@redhat.com> Cc: "William Tambe" <tambewilliam@gmail.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2017-03-03 00:49:41 +03:00
}
if (fvi->owner) {
mtree_print_mr_owner(mr);
}
if (fvi->ac) {
for (i = 0; i < fv_address_spaces->len; ++i) {
as = g_array_index(fv_address_spaces, AddressSpace*, i);
if (fvi->ac->has_memory(current_machine, as,
int128_get64(range->addr.start),
MR_SIZE(range->addr.size) + 1)) {
qemu_printf(" %s", fvi->ac_name);
}
}
}
qemu_printf("\n");
range++;
}
#if !defined(CONFIG_USER_ONLY)
if (fvi->dispatch_tree && view->root) {
mtree_print_dispatch(view->dispatch, view->root);
}
#endif
qemu_printf("\n");
}
static gboolean mtree_info_flatview_free(gpointer key, gpointer value,
gpointer user_data)
{
FlatView *view = key;
GArray *fv_address_spaces = value;
g_array_unref(fv_address_spaces);
flatview_unref(view);
return true;
}
void mtree_info(bool flatview, bool dispatch_tree, bool owner)
{
MemoryRegionListHead ml_head;
MemoryRegionList *ml, *ml2;
AddressSpace *as;
if (flatview) {
FlatView *view;
struct FlatViewInfo fvi = {
.counter = 0,
.dispatch_tree = dispatch_tree,
.owner = owner,
};
GArray *fv_address_spaces;
GHashTable *views = g_hash_table_new(g_direct_hash, g_direct_equal);
AccelClass *ac = ACCEL_GET_CLASS(current_machine->accelerator);
if (ac->has_memory) {
fvi.ac = ac;
fvi.ac_name = current_machine->accel ? current_machine->accel :
object_class_get_name(OBJECT_CLASS(ac));
}
/* Gather all FVs in one table */
QTAILQ_FOREACH(as, &address_spaces, address_spaces_link) {
view = address_space_get_flatview(as);
fv_address_spaces = g_hash_table_lookup(views, view);
if (!fv_address_spaces) {
fv_address_spaces = g_array_new(false, false, sizeof(as));
g_hash_table_insert(views, view, fv_address_spaces);
}
g_array_append_val(fv_address_spaces, as);
}
/* Print */
g_hash_table_foreach(views, mtree_print_flatview, &fvi);
/* Free */
g_hash_table_foreach_remove(views, mtree_info_flatview_free, 0);
g_hash_table_unref(views);
return;
}
QTAILQ_INIT(&ml_head);
QTAILQ_FOREACH(as, &address_spaces, address_spaces_link) {
qemu_printf("address-space: %s\n", as->name);
mtree_print_mr(as->root, 1, 0, &ml_head, owner);
qemu_printf("\n");
}
/* print aliased regions */
QTAILQ_FOREACH(ml, &ml_head, mrqueue) {
qemu_printf("memory-region: %s\n", memory_region_name(ml->mr));
mtree_print_mr(ml->mr, 1, 0, &ml_head, owner);
qemu_printf("\n");
}
QTAILQ_FOREACH_SAFE(ml, &ml_head, mrqueue, ml2) {
g_free(ml);
}
}
void memory_region_init_ram(MemoryRegion *mr,
struct Object *owner,
const char *name,
uint64_t size,
Error **errp)
{
DeviceState *owner_dev;
Error *err = NULL;
memory_region_init_ram_nomigrate(mr, owner, name, size, &err);
if (err) {
error_propagate(errp, err);
return;
}
/* This will assert if owner is neither NULL nor a DeviceState.
* We only want the owner here for the purposes of defining a
* unique name for migration. TODO: Ideally we should implement
* a naming scheme for Objects which are not DeviceStates, in
* which case we can relax this restriction.
*/
owner_dev = DEVICE(owner);
vmstate_register_ram(mr, owner_dev);
}
void memory_region_init_rom(MemoryRegion *mr,
struct Object *owner,
const char *name,
uint64_t size,
Error **errp)
{
DeviceState *owner_dev;
Error *err = NULL;
memory_region_init_rom_nomigrate(mr, owner, name, size, &err);
if (err) {
error_propagate(errp, err);
return;
}
/* This will assert if owner is neither NULL nor a DeviceState.
* We only want the owner here for the purposes of defining a
* unique name for migration. TODO: Ideally we should implement
* a naming scheme for Objects which are not DeviceStates, in
* which case we can relax this restriction.
*/
owner_dev = DEVICE(owner);
vmstate_register_ram(mr, owner_dev);
}
void memory_region_init_rom_device(MemoryRegion *mr,
struct Object *owner,
const MemoryRegionOps *ops,
void *opaque,
const char *name,
uint64_t size,
Error **errp)
{
DeviceState *owner_dev;
Error *err = NULL;
memory_region_init_rom_device_nomigrate(mr, owner, ops, opaque,
name, size, &err);
if (err) {
error_propagate(errp, err);
return;
}
/* This will assert if owner is neither NULL nor a DeviceState.
* We only want the owner here for the purposes of defining a
* unique name for migration. TODO: Ideally we should implement
* a naming scheme for Objects which are not DeviceStates, in
* which case we can relax this restriction.
*/
owner_dev = DEVICE(owner);
vmstate_register_ram(mr, owner_dev);
}
static const TypeInfo memory_region_info = {
.parent = TYPE_OBJECT,
.name = TYPE_MEMORY_REGION,
.class_size = sizeof(MemoryRegionClass),
.instance_size = sizeof(MemoryRegion),
.instance_init = memory_region_initfn,
.instance_finalize = memory_region_finalize,
};
static const TypeInfo iommu_memory_region_info = {
.parent = TYPE_MEMORY_REGION,
.name = TYPE_IOMMU_MEMORY_REGION,
.class_size = sizeof(IOMMUMemoryRegionClass),
.instance_size = sizeof(IOMMUMemoryRegion),
.instance_init = iommu_memory_region_initfn,
.abstract = true,
};
static void memory_register_types(void)
{
type_register_static(&memory_region_info);
type_register_static(&iommu_memory_region_info);
}
type_init(memory_register_types)