qemu/hw/s390x/s390-skeys.c

500 lines
15 KiB
C
Raw Normal View History

/*
* s390 storage key device
*
* Copyright 2015 IBM Corp.
* Author(s): Jason J. Herne <jjherne@linux.vnet.ibm.com>
*
* This work is licensed under the terms of the GNU GPL, version 2 or (at
* your option) any later version. See the COPYING file in the top-level
* directory.
*/
#include "qemu/osdep.h"
#include "qemu/units.h"
#include "hw/boards.h"
#include "hw/s390x/storage-keys.h"
#include "qapi/error.h"
#include "qapi/qapi-commands-misc-target.h"
#include "qapi/qmp/qdict.h"
#include "qemu/error-report.h"
#include "sysemu/memory_mapping.h"
#include "exec/address-spaces.h"
#include "sysemu/kvm.h"
#include "migration/qemu-file-types.h"
#include "migration/register.h"
#define S390_SKEYS_BUFFER_SIZE (128 * KiB) /* Room for 128k storage keys */
#define S390_SKEYS_SAVE_FLAG_EOS 0x01
#define S390_SKEYS_SAVE_FLAG_SKEYS 0x02
#define S390_SKEYS_SAVE_FLAG_ERROR 0x04
S390SKeysState *s390_get_skeys_device(void)
{
S390SKeysState *ss;
ss = S390_SKEYS(object_resolve_path_type("", TYPE_S390_SKEYS, NULL));
assert(ss);
return ss;
}
void s390_skeys_init(void)
{
Object *obj;
if (kvm_enabled()) {
obj = object_new(TYPE_KVM_S390_SKEYS);
} else {
obj = object_new(TYPE_QEMU_S390_SKEYS);
}
object_property_add_child(qdev_get_machine(), TYPE_S390_SKEYS,
qom: Drop parameter @errp of object_property_add() & friends The only way object_property_add() can fail is when a property with the same name already exists. Since our property names are all hardcoded, failure is a programming error, and the appropriate way to handle it is passing &error_abort. Same for its variants, except for object_property_add_child(), which additionally fails when the child already has a parent. Parentage is also under program control, so this is a programming error, too. We have a bit over 500 callers. Almost half of them pass &error_abort, slightly fewer ignore errors, one test case handles errors, and the remaining few callers pass them to their own callers. The previous few commits demonstrated once again that ignoring programming errors is a bad idea. Of the few ones that pass on errors, several violate the Error API. The Error ** argument must be NULL, &error_abort, &error_fatal, or a pointer to a variable containing NULL. Passing an argument of the latter kind twice without clearing it in between is wrong: if the first call sets an error, it no longer points to NULL for the second call. ich9_pm_add_properties(), sparc32_ledma_realize(), sparc32_dma_realize(), xilinx_axidma_realize(), xilinx_enet_realize() are wrong that way. When the one appropriate choice of argument is &error_abort, letting users pick the argument is a bad idea. Drop parameter @errp and assert the preconditions instead. There's one exception to "duplicate property name is a programming error": the way object_property_add() implements the magic (and undocumented) "automatic arrayification". Don't drop @errp there. Instead, rename object_property_add() to object_property_try_add(), and add the obvious wrapper object_property_add(). Signed-off-by: Markus Armbruster <armbru@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Paolo Bonzini <pbonzini@redhat.com> Message-Id: <20200505152926.18877-15-armbru@redhat.com> [Two semantic rebase conflicts resolved]
2020-05-05 23:29:22 +08:00
obj);
object_unref(obj);
qdev_realize(DEVICE(obj), NULL, &error_fatal);
}
static void write_keys(FILE *f, uint8_t *keys, uint64_t startgfn,
uint64_t count, Error **errp)
{
uint64_t curpage = startgfn;
uint64_t maxpage = curpage + count - 1;
for (; curpage <= maxpage; curpage++) {
uint8_t acc = (*keys & 0xF0) >> 4;
int fp = (*keys & 0x08);
int ref = (*keys & 0x04);
int ch = (*keys & 0x02);
int res = (*keys & 0x01);
fprintf(f, "page=%03" PRIx64 ": key(%d) => ACC=%X, FP=%d, REF=%d,"
" ch=%d, reserved=%d\n",
curpage, *keys, acc, fp, ref, ch, res);
keys++;
}
}
void hmp_info_skeys(Monitor *mon, const QDict *qdict)
{
S390SKeysState *ss = s390_get_skeys_device();
S390SKeysClass *skeyclass = S390_SKEYS_GET_CLASS(ss);
uint64_t addr = qdict_get_int(qdict, "addr");
uint8_t key;
int r;
/* Quick check to see if guest is using storage keys*/
if (!skeyclass->skeys_are_enabled(ss)) {
monitor_printf(mon, "Error: This guest is not using storage keys\n");
return;
}
if (!address_space_access_valid(&address_space_memory,
addr & TARGET_PAGE_MASK, TARGET_PAGE_SIZE,
false, MEMTXATTRS_UNSPECIFIED)) {
monitor_printf(mon, "Error: The given address is not valid\n");
return;
}
r = skeyclass->get_skeys(ss, addr / TARGET_PAGE_SIZE, 1, &key);
if (r < 0) {
monitor_printf(mon, "Error: %s\n", strerror(-r));
return;
}
monitor_printf(mon, " key: 0x%X\n", key);
}
void hmp_dump_skeys(Monitor *mon, const QDict *qdict)
{
const char *filename = qdict_get_str(qdict, "filename");
Error *err = NULL;
qmp_dump_skeys(filename, &err);
if (err) {
error: Use error_report_err() instead of monitor_printf() Both error_report_err() and monitor_printf() print to the same destination when monitor_printf() is used correctly, i.e. within an HMP monitor. Elsewhere, monitor_printf() does nothing, while error_report_err() reports to stderr. Most changed functions are HMP command handlers. These should only run within an HMP monitor. The one exception is bdrv_password_cb(), which should also only run within an HMP monitor. Four command handlers prefix the error message with the command name: balloon, migrate_set_capability, migrate_set_parameter, migrate. Pointless, drop. Unlike monitor_printf(), error_report_err() uses the error whole instead of just its message obtained with error_get_pretty(). This avoids suppressing its hint (see commit 50b7b00). Example: (qemu) device_add ivshmem,id=666 Parameter 'id' expects an identifier Identifiers consist of letters, digits, '-', '.', '_', starting with a letter. Try "help device_add" for more information The "Identifiers consist of..." line is new with this patch. Coccinelle semantic patch: @@ expression M, E; @@ - monitor_printf(M, "%s\n", error_get_pretty(E)); - error_free(E); + error_report_err(E); @r1@ expression M, E; format F; position p; @@ - monitor_printf(M, "...%@F@\n", error_get_pretty(E));@p - error_free(E); + error_report_err(E); @script:python@ p << r1.p; @@ print "%s:%s:%s: prefix dropped" % (p[0].file, p[0].line, p[0].column) Signed-off-by: Markus Armbruster <armbru@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Message-Id: <1450452927-8346-4-git-send-email-armbru@redhat.com>
2015-12-18 23:35:06 +08:00
error_report_err(err);
}
}
void qmp_dump_skeys(const char *filename, Error **errp)
{
S390SKeysState *ss = s390_get_skeys_device();
S390SKeysClass *skeyclass = S390_SKEYS_GET_CLASS(ss);
GuestPhysBlockList guest_phys_blocks;
GuestPhysBlock *block;
uint64_t pages, gfn;
Error *lerr = NULL;
uint8_t *buf;
int ret;
int fd;
FILE *f;
/* Quick check to see if guest is using storage keys*/
if (!skeyclass->skeys_are_enabled(ss)) {
error_setg(errp, "This guest is not using storage keys - "
"nothing to dump");
return;
}
fd = qemu_open_old(filename, O_WRONLY | O_CREAT | O_TRUNC, 0600);
if (fd < 0) {
error_setg_file_open(errp, errno, filename);
return;
}
f = fdopen(fd, "wb");
if (!f) {
close(fd);
error_setg_file_open(errp, errno, filename);
return;
}
buf = g_try_malloc(S390_SKEYS_BUFFER_SIZE);
if (!buf) {
error_setg(errp, "Could not allocate memory");
goto out;
}
assert(qemu_mutex_iothread_locked());
guest_phys_blocks_init(&guest_phys_blocks);
guest_phys_blocks_append(&guest_phys_blocks);
QTAILQ_FOREACH(block, &guest_phys_blocks.head, next) {
assert(QEMU_IS_ALIGNED(block->target_start, TARGET_PAGE_SIZE));
assert(QEMU_IS_ALIGNED(block->target_end, TARGET_PAGE_SIZE));
gfn = block->target_start / TARGET_PAGE_SIZE;
pages = (block->target_end - block->target_start) / TARGET_PAGE_SIZE;
while (pages) {
const uint64_t cur_pages = MIN(pages, S390_SKEYS_BUFFER_SIZE);
ret = skeyclass->get_skeys(ss, gfn, cur_pages, buf);
if (ret < 0) {
error_setg_errno(errp, -ret, "get_keys error");
goto out_free;
}
/* write keys to stream */
write_keys(f, buf, gfn, cur_pages, &lerr);
if (lerr) {
goto out_free;
}
gfn += cur_pages;
pages -= cur_pages;
}
}
out_free:
guest_phys_blocks_free(&guest_phys_blocks);
error_propagate(errp, lerr);
g_free(buf);
out:
fclose(f);
}
static bool qemu_s390_skeys_are_enabled(S390SKeysState *ss)
{
QEMUS390SKeysState *skeys = QEMU_S390_SKEYS(ss);
/* Lockless check is sufficient. */
return !!skeys->keydata;
}
static bool qemu_s390_enable_skeys(S390SKeysState *ss)
{
QEMUS390SKeysState *skeys = QEMU_S390_SKEYS(ss);
static gsize initialized;
if (likely(skeys->keydata)) {
return true;
}
/*
* TODO: Modern Linux doesn't use storage keys unless running KVM guests
* that use storage keys. Therefore, we keep it simple for now.
*
* 1) We should initialize to "referenced+changed" for an initial
* over-indication. Let's avoid touching megabytes of data for now and
* assume that any sane user will issue a storage key instruction before
* actually relying on this data.
* 2) Relying on ram_size and allocating a big array is ugly. We should
* allocate and manage storage key data per RAMBlock or optimally using
* some sparse data structure.
* 3) We only ever have a single S390SKeysState, so relying on
* g_once_init_enter() is good enough.
*/
if (g_once_init_enter(&initialized)) {
MachineState *machine = MACHINE(qdev_get_machine());
skeys->key_count = machine->ram_size / TARGET_PAGE_SIZE;
skeys->keydata = g_malloc0(skeys->key_count);
g_once_init_leave(&initialized, 1);
}
return false;
}
static int qemu_s390_skeys_set(S390SKeysState *ss, uint64_t start_gfn,
uint64_t count, uint8_t *keys)
{
QEMUS390SKeysState *skeydev = QEMU_S390_SKEYS(ss);
int i;
/* Check for uint64 overflow and access beyond end of key data */
if (unlikely(!skeydev->keydata || start_gfn + count > skeydev->key_count ||
start_gfn + count < count)) {
error_report("Error: Setting storage keys for pages with unallocated "
"storage key memory: gfn=%" PRIx64 " count=%" PRId64,
start_gfn, count);
return -EINVAL;
}
for (i = 0; i < count; i++) {
skeydev->keydata[start_gfn + i] = keys[i];
}
return 0;
}
static int qemu_s390_skeys_get(S390SKeysState *ss, uint64_t start_gfn,
uint64_t count, uint8_t *keys)
{
QEMUS390SKeysState *skeydev = QEMU_S390_SKEYS(ss);
int i;
/* Check for uint64 overflow and access beyond end of key data */
if (unlikely(!skeydev->keydata || start_gfn + count > skeydev->key_count ||
start_gfn + count < count)) {
error_report("Error: Getting storage keys for pages with unallocated "
"storage key memory: gfn=%" PRIx64 " count=%" PRId64,
start_gfn, count);
return -EINVAL;
}
for (i = 0; i < count; i++) {
keys[i] = skeydev->keydata[start_gfn + i];
}
return 0;
}
static void qemu_s390_skeys_class_init(ObjectClass *oc, void *data)
{
S390SKeysClass *skeyclass = S390_SKEYS_CLASS(oc);
DeviceClass *dc = DEVICE_CLASS(oc);
skeyclass->skeys_are_enabled = qemu_s390_skeys_are_enabled;
skeyclass->enable_skeys = qemu_s390_enable_skeys;
skeyclass->get_skeys = qemu_s390_skeys_get;
skeyclass->set_skeys = qemu_s390_skeys_set;
/* Reason: Internal device (only one skeys device for the whole memory) */
dc->user_creatable = false;
}
static const TypeInfo qemu_s390_skeys_info = {
.name = TYPE_QEMU_S390_SKEYS,
.parent = TYPE_S390_SKEYS,
.instance_size = sizeof(QEMUS390SKeysState),
.class_init = qemu_s390_skeys_class_init,
.class_size = sizeof(S390SKeysClass),
};
static void s390_storage_keys_save(QEMUFile *f, void *opaque)
{
S390SKeysState *ss = S390_SKEYS(opaque);
S390SKeysClass *skeyclass = S390_SKEYS_GET_CLASS(ss);
GuestPhysBlockList guest_phys_blocks;
GuestPhysBlock *block;
uint64_t pages, gfn;
int error = 0;
uint8_t *buf;
if (!skeyclass->skeys_are_enabled(ss)) {
goto end_stream;
}
buf = g_try_malloc(S390_SKEYS_BUFFER_SIZE);
if (!buf) {
error_report("storage key save could not allocate memory");
goto end_stream;
}
guest_phys_blocks_init(&guest_phys_blocks);
guest_phys_blocks_append(&guest_phys_blocks);
/* Send each contiguous physical memory range separately. */
QTAILQ_FOREACH(block, &guest_phys_blocks.head, next) {
assert(QEMU_IS_ALIGNED(block->target_start, TARGET_PAGE_SIZE));
assert(QEMU_IS_ALIGNED(block->target_end, TARGET_PAGE_SIZE));
gfn = block->target_start / TARGET_PAGE_SIZE;
pages = (block->target_end - block->target_start) / TARGET_PAGE_SIZE;
qemu_put_be64(f, block->target_start | S390_SKEYS_SAVE_FLAG_SKEYS);
qemu_put_be64(f, pages);
while (pages) {
const uint64_t cur_pages = MIN(pages, S390_SKEYS_BUFFER_SIZE);
if (!error) {
error = skeyclass->get_skeys(ss, gfn, cur_pages, buf);
if (error) {
/*
* Create a valid stream with all 0x00 and indicate
* S390_SKEYS_SAVE_FLAG_ERROR to the destination.
*/
error_report("S390_GET_KEYS error %d", error);
memset(buf, 0, S390_SKEYS_BUFFER_SIZE);
}
}
qemu_put_buffer(f, buf, cur_pages);
gfn += cur_pages;
pages -= cur_pages;
}
if (error) {
break;
}
}
guest_phys_blocks_free(&guest_phys_blocks);
g_free(buf);
end_stream:
if (error) {
qemu_put_be64(f, S390_SKEYS_SAVE_FLAG_ERROR);
} else {
qemu_put_be64(f, S390_SKEYS_SAVE_FLAG_EOS);
}
}
static int s390_storage_keys_load(QEMUFile *f, void *opaque, int version_id)
{
S390SKeysState *ss = S390_SKEYS(opaque);
S390SKeysClass *skeyclass = S390_SKEYS_GET_CLASS(ss);
int ret = 0;
/*
* Make sure to lazy-enable if required to be done explicitly. No need to
* flush any TLB as the VM is not running yet.
*/
if (skeyclass->enable_skeys) {
skeyclass->enable_skeys(ss);
}
while (!ret) {
ram_addr_t addr;
int flags;
addr = qemu_get_be64(f);
flags = addr & ~TARGET_PAGE_MASK;
addr &= TARGET_PAGE_MASK;
switch (flags) {
case S390_SKEYS_SAVE_FLAG_SKEYS: {
const uint64_t total_count = qemu_get_be64(f);
uint64_t handled_count = 0, cur_count;
uint64_t cur_gfn = addr / TARGET_PAGE_SIZE;
uint8_t *buf = g_try_malloc(S390_SKEYS_BUFFER_SIZE);
if (!buf) {
error_report("storage key load could not allocate memory");
ret = -ENOMEM;
break;
}
while (handled_count < total_count) {
cur_count = MIN(total_count - handled_count,
S390_SKEYS_BUFFER_SIZE);
qemu_get_buffer(f, buf, cur_count);
ret = skeyclass->set_skeys(ss, cur_gfn, cur_count, buf);
if (ret < 0) {
error_report("S390_SET_KEYS error %d", ret);
break;
}
handled_count += cur_count;
cur_gfn += cur_count;
}
g_free(buf);
break;
}
case S390_SKEYS_SAVE_FLAG_ERROR: {
error_report("Storage key data is incomplete");
ret = -EINVAL;
break;
}
case S390_SKEYS_SAVE_FLAG_EOS:
/* normal exit */
return 0;
default:
error_report("Unexpected storage key flag data: %#x", flags);
ret = -EINVAL;
}
}
return ret;
}
static inline bool s390_skeys_get_migration_enabled(Object *obj, Error **errp)
{
S390SKeysState *ss = S390_SKEYS(obj);
return ss->migration_enabled;
}
static SaveVMHandlers savevm_s390_storage_keys = {
.save_state = s390_storage_keys_save,
.load_state = s390_storage_keys_load,
};
static inline void s390_skeys_set_migration_enabled(Object *obj, bool value,
Error **errp)
{
S390SKeysState *ss = S390_SKEYS(obj);
/* Prevent double registration of savevm handler */
if (ss->migration_enabled == value) {
return;
}
ss->migration_enabled = value;
if (ss->migration_enabled) {
register_savevm_live(TYPE_S390_SKEYS, 0, 1,
&savevm_s390_storage_keys, ss);
} else {
unregister_savevm(VMSTATE_IF(ss), TYPE_S390_SKEYS, ss);
}
}
static void s390_skeys_instance_init(Object *obj)
{
object_property_add_bool(obj, "migration-enabled",
s390_skeys_get_migration_enabled,
qom: Drop parameter @errp of object_property_add() & friends The only way object_property_add() can fail is when a property with the same name already exists. Since our property names are all hardcoded, failure is a programming error, and the appropriate way to handle it is passing &error_abort. Same for its variants, except for object_property_add_child(), which additionally fails when the child already has a parent. Parentage is also under program control, so this is a programming error, too. We have a bit over 500 callers. Almost half of them pass &error_abort, slightly fewer ignore errors, one test case handles errors, and the remaining few callers pass them to their own callers. The previous few commits demonstrated once again that ignoring programming errors is a bad idea. Of the few ones that pass on errors, several violate the Error API. The Error ** argument must be NULL, &error_abort, &error_fatal, or a pointer to a variable containing NULL. Passing an argument of the latter kind twice without clearing it in between is wrong: if the first call sets an error, it no longer points to NULL for the second call. ich9_pm_add_properties(), sparc32_ledma_realize(), sparc32_dma_realize(), xilinx_axidma_realize(), xilinx_enet_realize() are wrong that way. When the one appropriate choice of argument is &error_abort, letting users pick the argument is a bad idea. Drop parameter @errp and assert the preconditions instead. There's one exception to "duplicate property name is a programming error": the way object_property_add() implements the magic (and undocumented) "automatic arrayification". Don't drop @errp there. Instead, rename object_property_add() to object_property_try_add(), and add the obvious wrapper object_property_add(). Signed-off-by: Markus Armbruster <armbru@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Paolo Bonzini <pbonzini@redhat.com> Message-Id: <20200505152926.18877-15-armbru@redhat.com> [Two semantic rebase conflicts resolved]
2020-05-05 23:29:22 +08:00
s390_skeys_set_migration_enabled);
qom: Put name parameter before value / visitor parameter The object_property_set_FOO() setters take property name and value in an unusual order: void object_property_set_FOO(Object *obj, FOO_TYPE value, const char *name, Error **errp) Having to pass value before name feels grating. Swap them. Same for object_property_set(), object_property_get(), and object_property_parse(). Convert callers with this Coccinelle script: @@ identifier fun = { object_property_get, object_property_parse, object_property_set_str, object_property_set_link, object_property_set_bool, object_property_set_int, object_property_set_uint, object_property_set, object_property_set_qobject }; expression obj, v, name, errp; @@ - fun(obj, v, name, errp) + fun(obj, name, v, errp) Chokes on hw/arm/musicpal.c's lcd_refresh() with the unhelpful error message "no position information". Convert that one manually. Fails to convert hw/arm/armsse.c, because Coccinelle gets confused by ARMSSE being used both as typedef and function-like macro there. Convert manually. Fails to convert hw/rx/rx-gdbsim.c, because Coccinelle gets confused by RXCPU being used both as typedef and function-like macro there. Convert manually. The other files using RXCPU that way don't need conversion. Signed-off-by: Markus Armbruster <armbru@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Message-Id: <20200707160613.848843-27-armbru@redhat.com> [Straightforwad conflict with commit 2336172d9b "audio: set default value for pcspk.iobase property" resolved]
2020-07-08 00:05:54 +08:00
object_property_set_bool(obj, "migration-enabled", true, NULL);
}
static void s390_skeys_class_init(ObjectClass *oc, void *data)
{
DeviceClass *dc = DEVICE_CLASS(oc);
dc->hotpluggable = false;
set_bit(DEVICE_CATEGORY_MISC, dc->categories);
}
static const TypeInfo s390_skeys_info = {
.name = TYPE_S390_SKEYS,
.parent = TYPE_DEVICE,
.instance_init = s390_skeys_instance_init,
.instance_size = sizeof(S390SKeysState),
.class_init = s390_skeys_class_init,
.class_size = sizeof(S390SKeysClass),
.abstract = true,
};
static void qemu_s390_skeys_register_types(void)
{
type_register_static(&s390_skeys_info);
type_register_static(&qemu_s390_skeys_info);
}
type_init(qemu_s390_skeys_register_types)