Merge tag 'mem-2024-12-21' of https://github.com/davidhildenbrand/qemu into staging

Hi,

"Host Memory Backends" and "Memory devices" queue ("mem"):
- Fixup handling of virtio-mem unplug during system resets, as
  preparation for s390x support (especially kdump in the Linux guest)
- virtio-mem support for s390x

 # -----BEGIN PGP SIGNATURE-----
 #
 # iQJFBAABCAAvFiEEG9nKrXNcTDpGDfzKTd4Q9wD/g1oFAmdnFD4RHGRhdmlkQHJl
 # ZGhhdC5jb20ACgkQTd4Q9wD/g1rWBBAAp7WkYaNAjRy1PgpjNZ3z1gUJc/vk+skJ
 # xVgGodA8txrJOFpNrbTyfhrdLs2TV4oWDvB/zrZRRtuxvur3O1EhFd9k6EqXuydr
 # 0FunvLvVJwRHfEZycjN4aacQMRH3CJw07OaTzexeSl5UR/6w5PRofwUK4HX7W/Ka
 # arqomGa3OJrs1+WgkV0Qcn4vh9HLRVv3iNC2Xo4W1wOCr1Du9zSPn9oC7zOQ0EO4
 # ZC//7QsdkNRjUX/yMXMkhlSXx3b/RmRg2DBrxo7BZXg27VwGu4uHxL4LRBZiB2A7
 # V9MqFOcVKzPMkXKTRjrgZ0vXQx1MPJ6WprEihMzMpYU6DrpA7KN/l8Ca8H24B2ln
 # h7+bmkDsHVVcWovE9ii/9cMRfws6uWXXg3KoA8RQ8IbX1tU02lblw2uHhXEzcoge
 # npqp/Z5LAiKVMetEnNnLH5thjut5PAEjuqD00cmZAMy4DNngLX2bGSdzMeVBkDMa
 # 78ehLGRplm3t7ibUfaZaMKe6UD9tFrcD6XKsvUTXXHNbYO8ynbx58WOxSZmY98zU
 # n3JNQRqtXYjBVlH3Dqm47vOTZHgOzFv3raa8BmSLpcBDeTXCTcUIl20s77dGw/vT
 # r5YNCMN7O4YPFKUoRK9604QTgw6qlYaRTQlJD09usprGqVylb6gQtfZZuZkYDMp8
 # sEI77QHsePA=
 # =HDxr
 # -----END PGP SIGNATURE-----
 # gpg: Signature made Sat 21 Dec 2024 14:17:18 EST
 # gpg:                using RSA key 1BD9CAAD735C4C3A460DFCCA4DDE10F700FF835A
 # gpg:                issuer "david@redhat.com"
 # gpg: Good signature from "David Hildenbrand <david@redhat.com>" [unknown]
 # gpg:                 aka "David Hildenbrand <davidhildenbrand@gmail.com>" [full]
 # gpg:                 aka "David Hildenbrand <hildenbr@in.tum.de>" [unknown]
 # gpg: WARNING: The key's User ID is not certified with a trusted signature!
 # gpg:          There is no indication that the signature belongs to the owner.
 # Primary key fingerprint: 1BD9 CAAD 735C 4C3A 460D  FCCA 4DDE 10F7 00FF 835A

* tag 'mem-2024-12-21' of https://github.com/davidhildenbrand/qemu:
  s390x: virtio-mem support
  s390x/virtio-ccw: add support for virtio based memory devices
  s390x: remember the maximum page size
  s390x/pv: prepare for memory devices
  s390x/s390-virtio-ccw: prepare for memory devices
  s390x/s390-skeys: prepare for memory devices
  s390x/s390-stattrib-kvm: prepare for memory devices and sparse memory layouts
  s390x/s390-hypercall: introduce DIAG500 STORAGE_LIMIT
  s390x: introduce s390_get_memory_limit()
  s390x/s390-virtio-ccw: move setting the maximum guest size from sclp to machine code
  s390x: rename s390-virtio-hcall* to s390-hypercall*
  s390x/s390-virtio-hcall: prepare for more diag500 hypercalls
  s390x/s390-virtio-hcall: remove hypercall registration mechanism
  s390x/s390-virtio-ccw: don't crash on weird RAM sizes
  virtio-mem: unplug memory only during system resets, not device resets

Conflicts:
- hw/s390x/s390-stattrib-kvm.c
  sysemu/ -> system/ header rename conflict.
- hw/s390x/virtio-ccw-mem.c
  Make Property array const and removed DEFINE_PROP_END_OF_LIST() to
  conform to the latest conventions.

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

MAINTAINERS

@@ -2384,6 +2384,9 @@ F: include/hw/virtio/virtio-crypto.h
 virtio based memory device
 M: David Hildenbrand <david@redhat.com>
 S: Supported
+F: hw/s390x/virtio-ccw-md.c
+F: hw/s390x/virtio-ccw-md.h
+F: hw/s390x/virtio-ccw-md-stubs.c
 F: hw/virtio/virtio-md-pci.c
 F: include/hw/virtio/virtio-md-pci.h
 F: stubs/virtio-md-pci.c
@@ -2395,6 +2398,8 @@ W: https://virtio-mem.gitlab.io/
 F: hw/virtio/virtio-mem.c
 F: hw/virtio/virtio-mem-pci.h
 F: hw/virtio/virtio-mem-pci.c
+F: hw/s390x/virtio-ccw-mem.c
+F: hw/s390x/virtio-ccw-mem.h
 F: include/hw/virtio/virtio-mem.h
 
 virtio-snd

hw/s390x/Kconfig

@@ -16,3 +16,4 @@ config S390_CCW_VIRTIO
     select SCLPCONSOLE
     select VIRTIO_CCW
     select MSI_NONBROKEN
+    select VIRTIO_MEM_SUPPORTED

hw/s390x/meson.build

@@ -12,7 +12,6 @@ s390x_ss.add(files(
   's390-pci-inst.c',
   's390-skeys.c',
   's390-stattrib.c',
-  's390-virtio-hcall.c',
   'sclp.c',
   'sclpcpu.c',
   'sclpquiesce.c',
@@ -28,7 +27,10 @@ s390x_ss.add(when: 'CONFIG_KVM', if_true: files(
 s390x_ss.add(when: 'CONFIG_TCG', if_true: files(
   'tod-tcg.c',
 ))
-s390x_ss.add(when: 'CONFIG_S390_CCW_VIRTIO', if_true: files('s390-virtio-ccw.c'))
+s390x_ss.add(when: 'CONFIG_S390_CCW_VIRTIO', if_true: files(
+  's390-virtio-ccw.c',
+  's390-hypercall.c',
+))
 s390x_ss.add(when: 'CONFIG_TERMINAL3270', if_true: files('3270-ccw.c'))
 s390x_ss.add(when: 'CONFIG_VFIO', if_true: files('s390-pci-vfio.c'))
 
@@ -48,8 +50,12 @@ endif
 virtio_ss.add(when: 'CONFIG_VHOST_SCSI', if_true: files('vhost-scsi-ccw.c'))
 virtio_ss.add(when: 'CONFIG_VHOST_VSOCK', if_true: files('vhost-vsock-ccw.c'))
 virtio_ss.add(when: 'CONFIG_VHOST_USER_FS', if_true: files('vhost-user-fs-ccw.c'))
+virtio_ss.add(when: 'CONFIG_VIRTIO_MD', if_true: files('virtio-ccw-md.c'))
+virtio_ss.add(when: 'CONFIG_VIRTIO_MEM', if_true: files('virtio-ccw-mem.c'))
 s390x_ss.add_all(when: 'CONFIG_VIRTIO_CCW', if_true: virtio_ss)
 
+s390x_ss.add(when: 'CONFIG_VIRTIO_MD', if_false: files('virtio-ccw-md-stubs.c'))
+
 hw_arch += {'s390x': s390x_ss}
 
 hw_s390x_modules = {}

hw/s390x/s390-hypercall.c

@@ -0,0 +1,85 @@
+/*
+ * Support for QEMU/KVM hypercalls on s390
+ *
+ * Copyright 2012 IBM Corp.
+ * Author(s): Cornelia Huck <cornelia.huck@de.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 "cpu.h"
+#include "hw/s390x/s390-virtio-ccw.h"
+#include "hw/s390x/s390-hypercall.h"
+#include "hw/s390x/ioinst.h"
+#include "hw/s390x/css.h"
+#include "virtio-ccw.h"
+
+static int handle_virtio_notify(uint64_t mem)
+{
+    MachineState *ms = MACHINE(qdev_get_machine());
+
+    if (mem < ms->ram_size) {
+        /* Early printk */
+        return 0;
+    }
+    return -EINVAL;
+}
+
+static int handle_virtio_ccw_notify(uint64_t subch_id, uint64_t data)
+{
+    SubchDev *sch;
+    VirtIODevice *vdev;
+    int cssid, ssid, schid, m;
+    uint16_t vq_idx = data;
+
+    if (ioinst_disassemble_sch_ident(subch_id, &m, &cssid, &ssid, &schid)) {
+        return -EINVAL;
+    }
+    sch = css_find_subch(m, cssid, ssid, schid);
+    if (!sch || !css_subch_visible(sch)) {
+        return -EINVAL;
+    }
+
+    vdev = virtio_ccw_get_vdev(sch);
+    if (vq_idx >= VIRTIO_QUEUE_MAX || !virtio_queue_get_num(vdev, vq_idx)) {
+        return -EINVAL;
+    }
+
+    if (virtio_vdev_has_feature(vdev, VIRTIO_F_NOTIFICATION_DATA)) {
+        virtio_queue_set_shadow_avail_idx(virtio_get_queue(vdev, vq_idx),
+                                          (data >> 16) & 0xFFFF);
+    }
+
+    virtio_queue_notify(vdev, vq_idx);
+    return 0;
+}
+
+static uint64_t handle_storage_limit(void)
+{
+    S390CcwMachineState *s390ms = S390_CCW_MACHINE(qdev_get_machine());
+
+    return s390_get_memory_limit(s390ms) - 1;
+}
+
+void handle_diag_500(S390CPU *cpu, uintptr_t ra)
+{
+    CPUS390XState *env = &cpu->env;
+    const uint64_t subcode = env->regs[1];
+
+    switch (subcode) {
+    case DIAG500_VIRTIO_NOTIFY:
+        env->regs[2] = handle_virtio_notify(env->regs[2]);
+        break;
+    case DIAG500_VIRTIO_CCW_NOTIFY:
+        env->regs[2] = handle_virtio_ccw_notify(env->regs[2], env->regs[3]);
+        break;
+    case DIAG500_STORAGE_LIMIT:
+        env->regs[2] = handle_storage_limit();
+        break;
+    default:
+        s390_program_interrupt(env, PGM_SPECIFICATION, ra);
+    }
+}

hw/s390x/s390-hypercall.h

@@ -0,0 +1,25 @@
+/*
+ * Support for QEMU/KVM hypercalls on s390x
+ *
+ * Copyright IBM Corp. 2012, 2017
+ * Author(s): Cornelia Huck <cornelia.huck@de.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.
+ */
+
+#ifndef HW_S390_HYPERCALL_H
+#define HW_S390_HYPERCALL_H
+
+#include "cpu.h"
+
+#define DIAG500_VIRTIO_NOTIFY           0 /* legacy, implemented as a NOP */
+#define DIAG500_VIRTIO_RESET            1 /* legacy */
+#define DIAG500_VIRTIO_SET_STATUS       2 /* legacy */
+#define DIAG500_VIRTIO_CCW_NOTIFY       3 /* KVM_S390_VIRTIO_CCW_NOTIFY */
+#define DIAG500_STORAGE_LIMIT           4
+
+void handle_diag_500(S390CPU *cpu, uintptr_t ra);
+
+#endif /* HW_S390_HYPERCALL_H */

hw/s390x/s390-skeys.c

@@ -11,7 +11,7 @@
 
 #include "qemu/osdep.h"
 #include "qemu/units.h"
-#include "hw/boards.h"
+#include "hw/s390x/s390-virtio-ccw.h"
 #include "hw/qdev-properties.h"
 #include "hw/s390x/storage-keys.h"
 #include "qapi/error.h"
@@ -251,9 +251,9 @@ static bool qemu_s390_enable_skeys(S390SKeysState *ss)
      *    g_once_init_enter() is good enough.
      */
     if (g_once_init_enter(&initialized)) {
-        MachineState *machine = MACHINE(qdev_get_machine());
+        S390CcwMachineState *s390ms = S390_CCW_MACHINE(qdev_get_machine());
 
-        skeys->key_count = machine->ram_size / TARGET_PAGE_SIZE;
+        skeys->key_count = s390_get_memory_limit(s390ms) / TARGET_PAGE_SIZE;
         skeys->keydata = g_malloc0(skeys->key_count);
         g_once_init_leave(&initialized, 1);
     }

hw/s390x/s390-stattrib-kvm.c

@@ -10,11 +10,12 @@
  */
 
 #include "qemu/osdep.h"
-#include "hw/boards.h"
+#include "hw/s390x/s390-virtio-ccw.h"
 #include "migration/qemu-file.h"
 #include "hw/s390x/storage-attributes.h"
 #include "qemu/error-report.h"
 #include "system/kvm.h"
+#include "system/memory_mapping.h"
 #include "exec/ram_addr.h"
 #include "kvm/kvm_s390x.h"
 #include "qapi/error.h"
@@ -84,8 +85,8 @@ static int kvm_s390_stattrib_set_stattr(S390StAttribState *sa,
                                         uint8_t *values)
 {
     KVMS390StAttribState *sas = KVM_S390_STATTRIB(sa);
-    MachineState *machine = MACHINE(qdev_get_machine());
-    unsigned long max = machine->ram_size / TARGET_PAGE_SIZE;
+    S390CcwMachineState *s390ms = S390_CCW_MACHINE(qdev_get_machine());
+    unsigned long max = s390_get_memory_limit(s390ms) / TARGET_PAGE_SIZE;
 
     if (start_gfn + count > max) {
         error_report("Out of memory bounds when setting storage attributes");
@@ -103,39 +104,57 @@ static int kvm_s390_stattrib_set_stattr(S390StAttribState *sa,
 static void kvm_s390_stattrib_synchronize(S390StAttribState *sa)
 {
     KVMS390StAttribState *sas = KVM_S390_STATTRIB(sa);
-    MachineState *machine = MACHINE(qdev_get_machine());
-    unsigned long max = machine->ram_size / TARGET_PAGE_SIZE;
-    /* We do not need to reach the maximum buffer size allowed */
-    unsigned long cx, len = KVM_S390_SKEYS_MAX / 2;
+    S390CcwMachineState *s390ms = S390_CCW_MACHINE(qdev_get_machine());
+    unsigned long max = s390_get_memory_limit(s390ms) / TARGET_PAGE_SIZE;
+    unsigned long start_gfn, end_gfn, pages;
+    GuestPhysBlockList guest_phys_blocks;
+    GuestPhysBlock *block;
     int r;
     struct kvm_s390_cmma_log clog = {
         .flags = 0,
         .mask = ~0ULL,
     };
 
-    if (sas->incoming_buffer) {
-        for (cx = 0; cx + len <= max; cx += len) {
-            clog.start_gfn = cx;
-            clog.count = len;
-            clog.values = (uint64_t)(sas->incoming_buffer + cx);
-            r = kvm_vm_ioctl(kvm_state, KVM_S390_SET_CMMA_BITS, &clog);
-            if (r) {
-                error_report("KVM_S390_SET_CMMA_BITS failed: %s", strerror(-r));
-                return;
-            }
-        }
-        if (cx < max) {
-            clog.start_gfn = cx;
-            clog.count = max - cx;
-            clog.values = (uint64_t)(sas->incoming_buffer + cx);
+    if (!sas->incoming_buffer) {
+        return;
+    }
+    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));
+
+        start_gfn = block->target_start / TARGET_PAGE_SIZE;
+        end_gfn = block->target_end / TARGET_PAGE_SIZE;
+
+        while (start_gfn < end_gfn) {
+            /* Don't exceed the maximum buffer size. */
+            pages = MIN(end_gfn - start_gfn, KVM_S390_SKEYS_MAX / 2);
+
+            /*
+             * If we ever get guest physical memory beyond the configured
+             * memory limit, something went very wrong.
+             */
+            assert(start_gfn + pages <= max);
+
+            clog.start_gfn = start_gfn;
+            clog.count = pages;
+            clog.values = (uint64_t)(sas->incoming_buffer + start_gfn);
             r = kvm_vm_ioctl(kvm_state, KVM_S390_SET_CMMA_BITS, &clog);
             if (r) {
                 error_report("KVM_S390_SET_CMMA_BITS failed: %s", strerror(-r));
+                goto out;
             }
+
+            start_gfn += pages;
         }
-        g_free(sas->incoming_buffer);
-        sas->incoming_buffer = NULL;
     }
+
+out:
+    guest_phys_blocks_free(&guest_phys_blocks);
+    g_free(sas->incoming_buffer);
+    sas->incoming_buffer = NULL;
 }
 
 static int kvm_s390_stattrib_set_migrationmode(S390StAttribState *sa, bool val,

hw/s390x/s390-virtio-ccw.c

@@ -16,11 +16,8 @@
 #include "exec/ram_addr.h"
 #include "system/confidential-guest-support.h"
 #include "hw/boards.h"
-#include "hw/s390x/s390-virtio-hcall.h"
 #include "hw/s390x/sclp.h"
 #include "hw/s390x/s390_flic.h"
-#include "hw/s390x/ioinst.h"
-#include "hw/s390x/css.h"
 #include "virtio-ccw.h"
 #include "qemu/config-file.h"
 #include "qemu/ctype.h"
@@ -48,6 +45,9 @@
 #include "migration/blocker.h"
 #include "qapi/visitor.h"
 #include "hw/s390x/cpu-topology.h"
+#include "kvm/kvm_s390x.h"
+#include "hw/virtio/virtio-md-pci.h"
+#include "hw/s390x/virtio-ccw-md.h"
 #include CONFIG_DEVICES
 
 static Error *pv_mig_blocker;
@@ -124,70 +124,86 @@ static void subsystem_reset(void)
     }
 }
 
-static int virtio_ccw_hcall_notify(const uint64_t *args)
+static void s390_set_memory_limit(S390CcwMachineState *s390ms,
+                                  uint64_t new_limit)
 {
-    uint64_t subch_id = args[0];
-    uint64_t data = args[1];
-    SubchDev *sch;
-    VirtIODevice *vdev;
-    int cssid, ssid, schid, m;
-    uint16_t vq_idx = data;
+    uint64_t hw_limit = 0;
+    int ret = 0;
 
-    if (ioinst_disassemble_sch_ident(subch_id, &m, &cssid, &ssid, &schid)) {
-        return -EINVAL;
+    assert(!s390ms->memory_limit && new_limit);
+    if (kvm_enabled()) {
+        ret = kvm_s390_set_mem_limit(new_limit, &hw_limit);
     }
-    sch = css_find_subch(m, cssid, ssid, schid);
-    if (!sch || !css_subch_visible(sch)) {
-        return -EINVAL;
-    }
-
-    vdev = virtio_ccw_get_vdev(sch);
-    if (vq_idx >= VIRTIO_QUEUE_MAX || !virtio_queue_get_num(vdev, vq_idx)) {
-        return -EINVAL;
+    if (ret == -E2BIG) {
+        error_report("host supports a maximum of %" PRIu64 " GB",
+                     hw_limit / GiB);
+        exit(EXIT_FAILURE);
+    } else if (ret) {
+        error_report("setting the guest size failed");
+        exit(EXIT_FAILURE);
     }
+    s390ms->memory_limit = new_limit;
+}
 
-    if (virtio_vdev_has_feature(vdev, VIRTIO_F_NOTIFICATION_DATA)) {
-        virtio_queue_set_shadow_avail_idx(virtio_get_queue(vdev, vq_idx),
-                                          (data >> 16) & 0xFFFF);
+static void s390_set_max_pagesize(S390CcwMachineState *s390ms,
+                                  uint64_t pagesize)
+{
+    assert(!s390ms->max_pagesize && pagesize);
+    if (kvm_enabled()) {
+        kvm_s390_set_max_pagesize(pagesize, &error_fatal);
     }
-
-    virtio_queue_notify(vdev, vq_idx);
-    return 0;
+    s390ms->max_pagesize = pagesize;
 }
 
-static int virtio_ccw_hcall_early_printk(const uint64_t *args)
+static void s390_memory_init(MachineState *machine)
 {
-    uint64_t mem = args[0];
-    MachineState *ms = MACHINE(qdev_get_machine());
+    S390CcwMachineState *s390ms = S390_CCW_MACHINE(machine);
+    MemoryRegion *sysmem = get_system_memory();
+    MemoryRegion *ram = machine->ram;
+    uint64_t ram_size = memory_region_size(ram);
+    uint64_t devmem_base, devmem_size;
 
-    if (mem < ms->ram_size) {
-        /* Early printk */
-        return 0;
+    if (!QEMU_IS_ALIGNED(ram_size, 1 * MiB)) {
+        /*
+         * SCLP cannot possibly expose smaller granularity right now and KVM
+         * cannot handle smaller granularity. As we don't support NUMA, the
+         * region size directly corresponds to machine->ram_size, and the region
+         * is a single RAM memory region.
+         */
+        error_report("ram size must be multiples of 1 MiB");
+        exit(EXIT_FAILURE);
     }
-    return -EINVAL;
-}
 
-static void virtio_ccw_register_hcalls(void)
-{
-    s390_register_virtio_hypercall(KVM_S390_VIRTIO_CCW_NOTIFY,
-                                   virtio_ccw_hcall_notify);
-    /* Tolerate early printk. */
-    s390_register_virtio_hypercall(KVM_S390_VIRTIO_NOTIFY,
-                                   virtio_ccw_hcall_early_printk);
-}
+    devmem_size = 0;
+    devmem_base = ram_size;
+#ifdef CONFIG_MEM_DEVICE
+    if (machine->ram_size < machine->maxram_size) {
 
-static void s390_memory_init(MemoryRegion *ram)
-{
-    MemoryRegion *sysmem = get_system_memory();
+        /*
+         * Make sure memory devices have a sane default alignment, even
+         * when weird initial memory sizes are specified.
+         */
+        devmem_base = QEMU_ALIGN_UP(devmem_base, 1 * GiB);
+        devmem_size = machine->maxram_size - machine->ram_size;
+    }
+#endif
+    s390_set_memory_limit(s390ms, devmem_base + devmem_size);
 
-    /* allocate RAM for core */
+    /* Map the initial memory. Must happen after setting the memory limit. */
     memory_region_add_subregion(sysmem, 0, ram);
 
+    /* Initialize address space for memory devices. */
+#ifdef CONFIG_MEM_DEVICE
+    if (devmem_size) {
+        machine_memory_devices_init(machine, devmem_base, devmem_size);
+    }
+#endif /* CONFIG_MEM_DEVICE */
+
     /*
      * Configure the maximum page size. As no memory devices were created
      * yet, this is the page size of initial memory only.
      */
-    s390_set_max_pagesize(qemu_maxrampagesize(), &error_fatal);
+    s390_set_max_pagesize(s390ms, qemu_maxrampagesize());
     /* Initialize storage key device */
     s390_skeys_init();
     /* Initialize storage attributes device */
@@ -255,7 +271,7 @@ static void ccw_init(MachineState *machine)
     qdev_realize_and_unref(DEVICE(ms->sclp), NULL, &error_fatal);
 
     /* init memory + setup max page size. Required for the CPU model */
-    s390_memory_init(machine->ram);
+    s390_memory_init(machine);
 
     /* init CPUs (incl. CPU model) early so s390_has_feature() works */
     s390_init_cpus(machine);
@@ -285,9 +301,6 @@ static void ccw_init(MachineState *machine)
                               OBJECT(dev));
     sysbus_realize_and_unref(SYS_BUS_DEVICE(dev), &error_fatal);
 
-    /* register hypercalls */
-    virtio_ccw_register_hcalls();
-
     s390_enable_css_support(s390_cpu_addr2state(0));
 
     ret = css_create_css_image(VIRTUAL_CSSID, true);
@@ -535,11 +548,39 @@ static void s390_machine_reset(MachineState *machine, ResetType type)
     s390_ipl_clear_reset_request();
 }
 
+static void s390_machine_device_pre_plug(HotplugHandler *hotplug_dev,
+                                         DeviceState *dev, Error **errp)
+{
+    if (object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_MD_CCW)) {
+        virtio_ccw_md_pre_plug(VIRTIO_MD_CCW(dev), MACHINE(hotplug_dev), errp);
+    } else if (object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_MD_PCI)) {
+        error_setg(errp,
+                   "PCI-attached virtio based memory devices not supported");
+    }
+}
+
 static void s390_machine_device_plug(HotplugHandler *hotplug_dev,
                                      DeviceState *dev, Error **errp)
 {
+    S390CcwMachineState *s390ms = S390_CCW_MACHINE(hotplug_dev);
+
     if (object_dynamic_cast(OBJECT(dev), TYPE_CPU)) {
         s390_cpu_plug(hotplug_dev, dev, errp);
+    } else if (object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_MD_CCW)) {
+        /*
+         * At this point, the device is realized and set all memdevs mapped, so
+         * qemu_maxrampagesize() will pick up the page sizes of these memdevs
+         * as well. Before we plug the device and expose any RAM memory regions
+         * to the system, make sure we don't exceed the previously set max page
+         * size. While only relevant for KVM, there is not really any use case
+         * for this with TCG, so we'll unconditionally reject it.
+         */
+        if (qemu_maxrampagesize() != s390ms->max_pagesize) {
+            error_setg(errp, "Memory device uses a bigger page size than"
+                       " initial memory");
+            return;
+        }
+        virtio_ccw_md_plug(VIRTIO_MD_CCW(dev), MACHINE(hotplug_dev), errp);
     }
 }
 
@@ -549,9 +590,20 @@ static void s390_machine_device_unplug_request(HotplugHandler *hotplug_dev,
     if (object_dynamic_cast(OBJECT(dev), TYPE_CPU)) {
         error_setg(errp, "CPU hot unplug not supported on this machine");
         return;
+    } else if (object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_MD_CCW)) {
+        virtio_ccw_md_unplug_request(VIRTIO_MD_CCW(dev), MACHINE(hotplug_dev),
+                                     errp);
     }
 }
 
+static void s390_machine_device_unplug(HotplugHandler *hotplug_dev,
+                                       DeviceState *dev, Error **errp)
+{
+    if (object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_MD_CCW)) {
+        virtio_ccw_md_unplug(VIRTIO_MD_CCW(dev), MACHINE(hotplug_dev), errp);
+     }
+ }
+
 static CpuInstanceProperties s390_cpu_index_to_props(MachineState *ms,
                                                      unsigned cpu_index)
 {
@@ -598,7 +650,9 @@ static const CPUArchIdList *s390_possible_cpu_arch_ids(MachineState *ms)
 static HotplugHandler *s390_get_hotplug_handler(MachineState *machine,
                                                 DeviceState *dev)
 {
-    if (object_dynamic_cast(OBJECT(dev), TYPE_CPU)) {
+    if (object_dynamic_cast(OBJECT(dev), TYPE_CPU) ||
+        object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_MD_CCW) ||
+        object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_MD_PCI)) {
         return HOTPLUG_HANDLER(machine);
     }
     return NULL;
@@ -758,8 +812,10 @@ static void ccw_machine_class_init(ObjectClass *oc, void *data)
     mc->possible_cpu_arch_ids = s390_possible_cpu_arch_ids;
     /* it is overridden with 'host' cpu *in kvm_arch_init* */
     mc->default_cpu_type = S390_CPU_TYPE_NAME("qemu");
+    hc->pre_plug = s390_machine_device_pre_plug;
     hc->plug = s390_machine_device_plug;
     hc->unplug_request = s390_machine_device_unplug_request;
+    hc->unplug = s390_machine_device_unplug;
     nc->nmi_monitor_handler = s390_nmi;
     mc->default_ram_id = "s390.ram";
     mc->default_nic = "virtio-net-ccw";

hw/s390x/s390-virtio-hcall.c

@@ -1,41 +0,0 @@
-/*
- * Support for virtio hypercalls on s390
- *
- * Copyright 2012 IBM Corp.
- * Author(s): Cornelia Huck <cornelia.huck@de.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 "cpu.h"
-#include "hw/s390x/s390-virtio-hcall.h"
-
-#define MAX_DIAG_SUBCODES 255
-
-static s390_virtio_fn s390_diag500_table[MAX_DIAG_SUBCODES];
-
-void s390_register_virtio_hypercall(uint64_t code, s390_virtio_fn fn)
-{
-    assert(code < MAX_DIAG_SUBCODES);
-    assert(!s390_diag500_table[code]);
-
-    s390_diag500_table[code] = fn;
-}
-
-int s390_virtio_hypercall(CPUS390XState *env)
-{
-    s390_virtio_fn fn;
-
-    if (env->regs[1] < MAX_DIAG_SUBCODES) {
-        fn = s390_diag500_table[env->regs[1]];
-        if (fn) {
-            env->regs[2] = fn(&env->regs[2]);
-            return 0;
-        }
-    }
-
-    return -EINVAL;
-}

hw/s390x/s390-virtio-hcall.h

@@ -1,25 +0,0 @@
-/*
- * Support for virtio hypercalls on s390x
- *
- * Copyright IBM Corp. 2012, 2017
- * Author(s): Cornelia Huck <cornelia.huck@de.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.
- */
-
-#ifndef HW_S390_VIRTIO_HCALL_H
-#define HW_S390_VIRTIO_HCALL_H
-
-#include "standard-headers/asm-s390/virtio-ccw.h"
-#include "cpu.h"
-
-/* The only thing that we need from the old kvm_virtio.h file */
-#define KVM_S390_VIRTIO_NOTIFY 0
-
-typedef int (*s390_virtio_fn)(const uint64_t *args);
-void s390_register_virtio_hypercall(uint64_t code, s390_virtio_fn fn);
-int s390_virtio_hypercall(CPUS390XState *env);
-
-#endif /* HW_S390_VIRTIO_HCALL_H */

hw/s390x/sclp.c

@@ -161,7 +161,11 @@ static void read_SCP_info(SCLPDevice *sclp, SCCB *sccb)
         read_info->rnsize2 = cpu_to_be32(rnsize);
     }
 
-    /* we don't support standby memory, maxram_size is never exposed */
+    /*
+     * We don't support standby memory. maxram_size is used for sizing the
+     * memory device region, which is not exposed through SCLP but through
+     * diag500.
+     */
     rnmax = machine->ram_size >> sclp->increment_size;
     if (rnmax < 0x10000) {
         read_info->rnmax = cpu_to_be16(rnmax);
@@ -376,10 +380,7 @@ void sclp_service_interrupt(uint32_t sccb)
 /* qemu object creation and initialization functions */
 static void sclp_realize(DeviceState *dev, Error **errp)
 {
-    MachineState *machine = MACHINE(qdev_get_machine());
     SCLPDevice *sclp = SCLP(dev);
-    uint64_t hw_limit;
-    int ret;
 
     /*
      * qdev_device_add searches the sysbus for TYPE_SCLP_EVENTS_BUS. As long
@@ -389,14 +390,6 @@ static void sclp_realize(DeviceState *dev, Error **errp)
     if (!sysbus_realize(SYS_BUS_DEVICE(sclp->event_facility), errp)) {
         return;
     }
-
-    ret = s390_set_memory_limit(machine->maxram_size, &hw_limit);
-    if (ret == -E2BIG) {
-        error_setg(errp, "host supports a maximum of %" PRIu64 " GB",
-                   hw_limit / GiB);
-    } else if (ret) {
-        error_setg(errp, "setting the guest size failed");
-    }
 }
 
 static void sclp_memory_init(SCLPDevice *sclp)

hw/s390x/virtio-ccw-md-stubs.c

@@ -0,0 +1,24 @@
+#include "qemu/osdep.h"
+#include "qapi/error.h"
+#include "hw/s390x/virtio-ccw-md.h"
+
+void virtio_ccw_md_pre_plug(VirtIOMDCcw *vmd, MachineState *ms, Error **errp)
+{
+    error_setg(errp, "virtio based memory devices not supported");
+}
+
+void virtio_ccw_md_plug(VirtIOMDCcw *vmd, MachineState *ms, Error **errp)
+{
+    error_setg(errp, "virtio based memory devices not supported");
+}
+
+void virtio_ccw_md_unplug_request(VirtIOMDCcw *vmd, MachineState *ms,
+                                  Error **errp)
+{
+    error_setg(errp, "virtio based memory devices not supported");
+}
+
+void virtio_ccw_md_unplug(VirtIOMDCcw *vmd, MachineState *ms, Error **errp)
+{
+    error_setg(errp, "virtio based memory devices not supported");
+}

hw/s390x/virtio-ccw-md.c

@@ -0,0 +1,153 @@
+/*
+ * Virtio CCW support for abstract virtio based memory device
+ *
+ * Copyright (C) 2024 Red Hat, Inc.
+ *
+ * Authors:
+ *  David Hildenbrand <david@redhat.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.
+ * See the COPYING file in the top-level directory.
+ */
+
+#include "qemu/osdep.h"
+#include "hw/s390x/virtio-ccw-md.h"
+#include "hw/mem/memory-device.h"
+#include "qapi/error.h"
+#include "qemu/error-report.h"
+
+void virtio_ccw_md_pre_plug(VirtIOMDCcw *vmd, MachineState *ms, Error **errp)
+{
+    DeviceState *dev = DEVICE(vmd);
+    HotplugHandler *bus_handler = qdev_get_bus_hotplug_handler(dev);
+    MemoryDeviceState *md = MEMORY_DEVICE(vmd);
+    Error *local_err = NULL;
+
+    if (!bus_handler && dev->hotplugged) {
+        /*
+         * Without a bus hotplug handler, we cannot control the plug/unplug
+         * order. We should never reach this point when hotplugging, but
+         * better add a safety net.
+         */
+        error_setg(errp, "hotplug of virtio based memory devices not supported"
+                   " on this bus.");
+        return;
+    }
+
+    /*
+     * First, see if we can plug this memory device at all. If that
+     * succeeds, branch of to the actual hotplug handler.
+     */
+    memory_device_pre_plug(md, ms, &local_err);
+    if (!local_err && bus_handler) {
+        hotplug_handler_pre_plug(bus_handler, dev, &local_err);
+    }
+    error_propagate(errp, local_err);
+}
+
+void virtio_ccw_md_plug(VirtIOMDCcw *vmd, MachineState *ms, Error **errp)
+{
+    DeviceState *dev = DEVICE(vmd);
+    HotplugHandler *bus_handler = qdev_get_bus_hotplug_handler(dev);
+    MemoryDeviceState *md = MEMORY_DEVICE(vmd);
+    Error *local_err = NULL;
+
+    /*
+     * Plug the memory device first and then branch off to the actual
+     * hotplug handler. If that one fails, we can easily undo the memory
+     * device bits.
+     */
+    memory_device_plug(md, ms);
+    if (bus_handler) {
+        hotplug_handler_plug(bus_handler, dev, &local_err);
+        if (local_err) {
+            memory_device_unplug(md, ms);
+        }
+    }
+    error_propagate(errp, local_err);
+}
+
+void virtio_ccw_md_unplug_request(VirtIOMDCcw *vmd, MachineState *ms,
+                                  Error **errp)
+{
+    VirtIOMDCcwClass *vmdc = VIRTIO_MD_CCW_GET_CLASS(vmd);
+    DeviceState *dev = DEVICE(vmd);
+    HotplugHandler *bus_handler = qdev_get_bus_hotplug_handler(dev);
+    HotplugHandlerClass *hdc;
+    Error *local_err = NULL;
+
+    if (!vmdc->unplug_request_check) {
+        error_setg(errp,
+                   "this virtio based memory devices cannot be unplugged");
+        return;
+    }
+
+    if (!bus_handler) {
+        error_setg(errp, "hotunplug of virtio based memory devices not"
+                   "supported on this bus");
+        return;
+    }
+
+    vmdc->unplug_request_check(vmd, &local_err);
+    if (local_err) {
+        error_propagate(errp, local_err);
+        return;
+    }
+
+    /*
+     * Forward the async request or turn it into a sync request (handling it
+     * like qdev_unplug()).
+     */
+    hdc = HOTPLUG_HANDLER_GET_CLASS(bus_handler);
+    if (hdc->unplug_request) {
+        hotplug_handler_unplug_request(bus_handler, dev, &local_err);
+    } else {
+        virtio_ccw_md_unplug(vmd, ms, &local_err);
+        if (!local_err) {
+            object_unparent(OBJECT(dev));
+        }
+    }
+}
+
+void virtio_ccw_md_unplug(VirtIOMDCcw *vmd, MachineState *ms, Error **errp)
+{
+    DeviceState *dev = DEVICE(vmd);
+    HotplugHandler *bus_handler = qdev_get_bus_hotplug_handler(dev);
+    MemoryDeviceState *md = MEMORY_DEVICE(vmd);
+    Error *local_err = NULL;
+
+    /* Unplug the memory device while it is still realized. */
+    memory_device_unplug(md, ms);
+
+    if (bus_handler) {
+        hotplug_handler_unplug(bus_handler, dev, &local_err);
+        if (local_err) {
+            /* Not expected to fail ... but still try to recover. */
+            memory_device_plug(md, ms);
+            error_propagate(errp, local_err);
+            return;
+        }
+    } else {
+        /* Very unexpected, but let's just try to do the right thing. */
+        warn_report("Unexpected unplug of virtio based memory device");
+        qdev_unrealize(dev);
+    }
+}
+
+static const TypeInfo virtio_ccw_md_info = {
+    .name = TYPE_VIRTIO_MD_CCW,
+    .parent = TYPE_VIRTIO_CCW_DEVICE,
+    .instance_size = sizeof(VirtIOMDCcw),
+    .class_size = sizeof(VirtIOMDCcwClass),
+    .abstract = true,
+    .interfaces = (InterfaceInfo[]) {
+        { TYPE_MEMORY_DEVICE },
+        { }
+    },
+};
+
+static void virtio_ccw_md_register(void)
+{
+    type_register_static(&virtio_ccw_md_info);
+}
+type_init(virtio_ccw_md_register)

hw/s390x/virtio-ccw-md.h

@@ -0,0 +1,44 @@
+/*
+ * Virtio CCW support for abstract virtio based memory device
+ *
+ * Copyright (C) 2024 Red Hat, Inc.
+ *
+ * Authors:
+ *  David Hildenbrand <david@redhat.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.
+ * See the COPYING file in the top-level directory.
+ */
+
+#ifndef HW_S390X_VIRTIO_CCW_MD_H
+#define HW_S390X_VIRTIO_CCW_MD_H
+
+#include "virtio-ccw.h"
+#include "qom/object.h"
+
+/*
+ * virtio-md-ccw: This extends VirtioCcwDevice.
+ */
+#define TYPE_VIRTIO_MD_CCW "virtio-md-ccw"
+
+OBJECT_DECLARE_TYPE(VirtIOMDCcw, VirtIOMDCcwClass, VIRTIO_MD_CCW)
+
+struct VirtIOMDCcwClass {
+    /* private */
+    VirtIOCCWDeviceClass parent;
+
+    /* public */
+    void (*unplug_request_check)(VirtIOMDCcw *vmd, Error **errp);
+};
+
+struct VirtIOMDCcw {
+    VirtioCcwDevice parent_obj;
+};
+
+void virtio_ccw_md_pre_plug(VirtIOMDCcw *vmd, MachineState *ms, Error **errp);
+void virtio_ccw_md_plug(VirtIOMDCcw *vmd, MachineState *ms, Error **errp);
+void virtio_ccw_md_unplug_request(VirtIOMDCcw *vmd, MachineState *ms,
+                                  Error **errp);
+void virtio_ccw_md_unplug(VirtIOMDCcw *vmd, MachineState *ms, Error **errp);
+
+#endif /* HW_S390X_VIRTIO_CCW_MD_H */

hw/s390x/virtio-ccw-mem.c

@@ -0,0 +1,225 @@
+/*
+ * virtio-mem CCW implementation
+ *
+ * Copyright (C) 2024 Red Hat, Inc.
+ *
+ * Authors:
+ *  David Hildenbrand <david@redhat.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.
+ * See the COPYING file in the top-level directory.
+ */
+
+#include "qemu/osdep.h"
+#include "hw/qdev-properties.h"
+#include "qapi/error.h"
+#include "qemu/module.h"
+#include "virtio-ccw-mem.h"
+#include "hw/mem/memory-device.h"
+#include "qapi/qapi-events-machine.h"
+#include "qapi/qapi-events-misc.h"
+
+static void virtio_ccw_mem_realize(VirtioCcwDevice *ccw_dev, Error **errp)
+{
+    VirtIOMEMCcw *dev = VIRTIO_MEM_CCW(ccw_dev);
+    DeviceState *vdev = DEVICE(&dev->vdev);
+
+    qdev_realize(vdev, BUS(&ccw_dev->bus), errp);
+}
+
+static void virtio_ccw_mem_set_addr(MemoryDeviceState *md, uint64_t addr,
+                                    Error **errp)
+{
+    object_property_set_uint(OBJECT(md), VIRTIO_MEM_ADDR_PROP, addr, errp);
+}
+
+static uint64_t virtio_ccw_mem_get_addr(const MemoryDeviceState *md)
+{
+    return object_property_get_uint(OBJECT(md), VIRTIO_MEM_ADDR_PROP,
+                                    &error_abort);
+}
+
+static MemoryRegion *virtio_ccw_mem_get_memory_region(MemoryDeviceState *md,
+                                                      Error **errp)
+{
+    VirtIOMEMCcw *dev = VIRTIO_MEM_CCW(md);
+    VirtIOMEM *vmem = &dev->vdev;
+    VirtIOMEMClass *vmc = VIRTIO_MEM_GET_CLASS(vmem);
+
+    return vmc->get_memory_region(vmem, errp);
+}
+
+static void virtio_ccw_mem_decide_memslots(MemoryDeviceState *md,
+                                           unsigned int limit)
+{
+    VirtIOMEMCcw *dev = VIRTIO_MEM_CCW(md);
+    VirtIOMEM *vmem = VIRTIO_MEM(&dev->vdev);
+    VirtIOMEMClass *vmc = VIRTIO_MEM_GET_CLASS(vmem);
+
+    vmc->decide_memslots(vmem, limit);
+}
+
+static unsigned int virtio_ccw_mem_get_memslots(MemoryDeviceState *md)
+{
+    VirtIOMEMCcw *dev = VIRTIO_MEM_CCW(md);
+    VirtIOMEM *vmem = VIRTIO_MEM(&dev->vdev);
+    VirtIOMEMClass *vmc = VIRTIO_MEM_GET_CLASS(vmem);
+
+    return vmc->get_memslots(vmem);
+}
+
+static uint64_t virtio_ccw_mem_get_plugged_size(const MemoryDeviceState *md,
+                                                Error **errp)
+{
+    return object_property_get_uint(OBJECT(md), VIRTIO_MEM_SIZE_PROP,
+                                    errp);
+}
+
+static void virtio_ccw_mem_fill_device_info(const MemoryDeviceState *md,
+                                            MemoryDeviceInfo *info)
+{
+    VirtioMEMDeviceInfo *vi = g_new0(VirtioMEMDeviceInfo, 1);
+    VirtIOMEMCcw *dev = VIRTIO_MEM_CCW(md);
+    VirtIOMEM *vmem = &dev->vdev;
+    VirtIOMEMClass *vpc = VIRTIO_MEM_GET_CLASS(vmem);
+    DeviceState *vdev = DEVICE(md);
+
+    if (vdev->id) {
+        vi->id = g_strdup(vdev->id);
+    }
+
+    /* let the real device handle everything else */
+    vpc->fill_device_info(vmem, vi);
+
+    info->u.virtio_mem.data = vi;
+    info->type = MEMORY_DEVICE_INFO_KIND_VIRTIO_MEM;
+}
+
+static uint64_t virtio_ccw_mem_get_min_alignment(const MemoryDeviceState *md)
+{
+    return object_property_get_uint(OBJECT(md), VIRTIO_MEM_BLOCK_SIZE_PROP,
+                                    &error_abort);
+}
+
+static void virtio_ccw_mem_size_change_notify(Notifier *notifier, void *data)
+{
+    VirtIOMEMCcw *dev = container_of(notifier, VirtIOMEMCcw,
+                                         size_change_notifier);
+    DeviceState *vdev = DEVICE(dev);
+    char *qom_path = object_get_canonical_path(OBJECT(dev));
+    const uint64_t * const size_p = data;
+
+    qapi_event_send_memory_device_size_change(vdev->id, *size_p, qom_path);
+    g_free(qom_path);
+}
+
+static void virtio_ccw_mem_unplug_request_check(VirtIOMDCcw *vmd, Error **errp)
+{
+    VirtIOMEMCcw *dev = VIRTIO_MEM_CCW(vmd);
+    VirtIOMEM *vmem = &dev->vdev;
+    VirtIOMEMClass *vpc = VIRTIO_MEM_GET_CLASS(vmem);
+
+    vpc->unplug_request_check(vmem, errp);
+}
+
+static void virtio_ccw_mem_get_requested_size(Object *obj, Visitor *v,
+                                              const char *name, void *opaque,
+                                              Error **errp)
+{
+    VirtIOMEMCcw *dev = VIRTIO_MEM_CCW(obj);
+
+    object_property_get(OBJECT(&dev->vdev), name, v, errp);
+}
+
+static void virtio_ccw_mem_set_requested_size(Object *obj, Visitor *v,
+                                              const char *name, void *opaque,
+                                              Error **errp)
+{
+    VirtIOMEMCcw *dev = VIRTIO_MEM_CCW(obj);
+    DeviceState *vdev = DEVICE(obj);
+
+    /*
+     * If we passed virtio_ccw_mem_unplug_request_check(), making sure that
+     * the requested size is 0, don't allow modifying the requested size
+     * anymore, otherwise the VM might end up hotplugging memory before
+     * handling the unplug request.
+     */
+    if (vdev->pending_deleted_event) {
+        error_setg(errp, "'%s' cannot be changed if the device is in the"
+                   " process of unplug", name);
+        return;
+    }
+
+    object_property_set(OBJECT(&dev->vdev), name, v, errp);
+}
+
+static const Property virtio_ccw_mem_properties[] = {
+    DEFINE_PROP_BIT("ioeventfd", VirtioCcwDevice, flags,
+                    VIRTIO_CCW_FLAG_USE_IOEVENTFD_BIT, true),
+    DEFINE_PROP_UINT32("max_revision", VirtioCcwDevice, max_rev,
+                       VIRTIO_CCW_MAX_REV),
+};
+
+static void virtio_ccw_mem_class_init(ObjectClass *klass, void *data)
+{
+    DeviceClass *dc = DEVICE_CLASS(klass);
+    VirtIOCCWDeviceClass *k = VIRTIO_CCW_DEVICE_CLASS(klass);
+    MemoryDeviceClass *mdc = MEMORY_DEVICE_CLASS(klass);
+    VirtIOMDCcwClass *vmdc = VIRTIO_MD_CCW_CLASS(klass);
+
+    k->realize = virtio_ccw_mem_realize;
+    set_bit(DEVICE_CATEGORY_MISC, dc->categories);
+    device_class_set_props(dc, virtio_ccw_mem_properties);
+
+    mdc->get_addr = virtio_ccw_mem_get_addr;
+    mdc->set_addr = virtio_ccw_mem_set_addr;
+    mdc->get_plugged_size = virtio_ccw_mem_get_plugged_size;
+    mdc->get_memory_region = virtio_ccw_mem_get_memory_region;
+    mdc->decide_memslots = virtio_ccw_mem_decide_memslots;
+    mdc->get_memslots = virtio_ccw_mem_get_memslots;
+    mdc->fill_device_info = virtio_ccw_mem_fill_device_info;
+    mdc->get_min_alignment = virtio_ccw_mem_get_min_alignment;
+
+    vmdc->unplug_request_check = virtio_ccw_mem_unplug_request_check;
+}
+
+static void virtio_ccw_mem_instance_init(Object *obj)
+{
+    VirtIOMEMCcw *dev = VIRTIO_MEM_CCW(obj);
+    VirtIOMEMClass *vmc;
+    VirtIOMEM *vmem;
+
+    virtio_instance_init_common(obj, &dev->vdev, sizeof(dev->vdev),
+                                TYPE_VIRTIO_MEM);
+
+    dev->size_change_notifier.notify = virtio_ccw_mem_size_change_notify;
+    vmem = &dev->vdev;
+    vmc = VIRTIO_MEM_GET_CLASS(vmem);
+    /*
+     * We never remove the notifier again, as we expect both devices to
+     * disappear at the same time.
+     */
+    vmc->add_size_change_notifier(vmem, &dev->size_change_notifier);
+
+    object_property_add_alias(obj, VIRTIO_MEM_BLOCK_SIZE_PROP,
+                              OBJECT(&dev->vdev), VIRTIO_MEM_BLOCK_SIZE_PROP);
+    object_property_add_alias(obj, VIRTIO_MEM_SIZE_PROP, OBJECT(&dev->vdev),
+                              VIRTIO_MEM_SIZE_PROP);
+    object_property_add(obj, VIRTIO_MEM_REQUESTED_SIZE_PROP, "size",
+                        virtio_ccw_mem_get_requested_size,
+                        virtio_ccw_mem_set_requested_size, NULL, NULL);
+}
+
+static const TypeInfo virtio_ccw_mem = {
+    .name = TYPE_VIRTIO_MEM_CCW,
+    .parent = TYPE_VIRTIO_MD_CCW,
+    .instance_size = sizeof(VirtIOMEMCcw),
+    .instance_init = virtio_ccw_mem_instance_init,
+    .class_init = virtio_ccw_mem_class_init,
+};
+
+static void virtio_ccw_mem_register_types(void)
+{
+    type_register_static(&virtio_ccw_mem);
+}
+type_init(virtio_ccw_mem_register_types)

hw/s390x/virtio-ccw-mem.h

@@ -0,0 +1,34 @@
+/*
+ * Virtio MEM CCW device
+ *
+ * Copyright (C) 2024 Red Hat, Inc.
+ *
+ * Authors:
+ *  David Hildenbrand <david@redhat.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.
+ * See the COPYING file in the top-level directory.
+ */
+
+#ifndef HW_S390X_VIRTIO_CCW_MEM_H
+#define HW_S390X_VIRTIO_CCW_MEM_H
+
+#include "virtio-ccw-md.h"
+#include "hw/virtio/virtio-mem.h"
+#include "qom/object.h"
+
+typedef struct VirtIOMEMCcw VirtIOMEMCcw;
+
+/*
+ * virtio-mem-ccw: This extends VirtIOMDCcw
+ */
+#define TYPE_VIRTIO_MEM_CCW "virtio-mem-ccw"
+DECLARE_INSTANCE_CHECKER(VirtIOMEMCcw, VIRTIO_MEM_CCW, TYPE_VIRTIO_MEM_CCW)
+
+struct VirtIOMEMCcw {
+    VirtIOMDCcw parent_obj;
+    VirtIOMEM vdev;
+    Notifier size_change_notifier;
+};
+
+#endif /* HW_S390X_VIRTIO_CCW_MEM_H */

hw/virtio/Kconfig

@@ -29,6 +29,7 @@ config VIRTIO_MMIO
 config VIRTIO_CCW
     bool
     select VIRTIO
+    select VIRTIO_MD_SUPPORTED
 
 config VIRTIO_BALLOON
     bool

hw/virtio/virtio-mem.c

@@ -61,6 +61,8 @@ static uint32_t virtio_mem_default_thp_size(void)
     } else if (qemu_real_host_page_size() == 64 * KiB) {
         default_thp_size = 512 * MiB;
     }
+#elif defined(__s390x__)
+    default_thp_size = 1 * MiB;
 #endif
 
     return default_thp_size;
@@ -168,7 +170,7 @@ static bool virtio_mem_has_shared_zeropage(RAMBlock *rb)
  * necessary (as the section size can change). But it's more likely that the
  * section size will rather get smaller and not bigger over time.
  */
-#if defined(TARGET_X86_64) || defined(TARGET_I386)
+#if defined(TARGET_X86_64) || defined(TARGET_I386) || defined(TARGET_S390X)
 #define VIRTIO_MEM_USABLE_EXTENT (2 * (128 * MiB))
 #elif defined(TARGET_ARM)
 #define VIRTIO_MEM_USABLE_EXTENT (2 * (512 * MiB))
@@ -956,6 +958,7 @@ static void virtio_mem_device_realize(DeviceState *dev, Error **errp)
     VirtIOMEM *vmem = VIRTIO_MEM(dev);
     uint64_t page_size;
     RAMBlock *rb;
+    Object *obj;
     int ret;
 
     if (!vmem->memdev) {
@@ -1121,7 +1124,28 @@ static void virtio_mem_device_realize(DeviceState *dev, Error **errp)
         vmstate_register_any(VMSTATE_IF(vmem),
                              &vmstate_virtio_mem_device_early, vmem);
     }
-    qemu_register_resettable(OBJECT(vmem));
+
+    /*
+     * We only want to unplug all memory to start with a clean slate when
+     * it is safe for the guest -- during system resets that call
+     * qemu_devices_reset().
+     *
+     * We'll filter out selected qemu_devices_reset() calls used for other
+     * purposes, like resetting all devices during wakeup from suspend on
+     * x86 based on the reset type passed to qemu_devices_reset().
+     *
+     * Unplugging all memory during simple device resets can result in the VM
+     * unexpectedly losing RAM, corrupting VM state.
+     *
+     * Simple device resets (or resets triggered by getting a parent device
+     * reset) must not change the state of plugged memory blocks. Therefore,
+     * we need a dedicated reset object that only gets called during
+     * qemu_devices_reset().
+     */
+    obj = object_new(TYPE_VIRTIO_MEM_SYSTEM_RESET);
+    vmem->system_reset = VIRTIO_MEM_SYSTEM_RESET(obj);
+    vmem->system_reset->vmem = vmem;
+    qemu_register_resettable(obj);
 
     /*
      * Set ourselves as RamDiscardManager before the plug handler maps the
@@ -1141,7 +1165,10 @@ static void virtio_mem_device_unrealize(DeviceState *dev)
      * found via an address space anymore. Unset ourselves.
      */
     memory_region_set_ram_discard_manager(&vmem->memdev->mr, NULL);
-    qemu_unregister_resettable(OBJECT(vmem));
+
+    qemu_unregister_resettable(OBJECT(vmem->system_reset));
+    object_unref(OBJECT(vmem->system_reset));
+
     if (vmem->early_migration) {
         vmstate_unregister(VMSTATE_IF(vmem), &vmstate_virtio_mem_device_early,
                            vmem);
@@ -1840,38 +1867,12 @@ static void virtio_mem_unplug_request_check(VirtIOMEM *vmem, Error **errp)
     }
 }
 
-static ResettableState *virtio_mem_get_reset_state(Object *obj)
-{
-    VirtIOMEM *vmem = VIRTIO_MEM(obj);
-    return &vmem->reset_state;
-}
-
-static void virtio_mem_system_reset_hold(Object *obj, ResetType type)
-{
-    VirtIOMEM *vmem = VIRTIO_MEM(obj);
-
-    /*
-     * When waking up from standby/suspend-to-ram, do not unplug any memory.
-     */
-    if (type == RESET_TYPE_WAKEUP) {
-        return;
-    }
-
-    /*
-     * During usual resets, we will unplug all memory and shrink the usable
-     * region size. This is, however, not possible in all scenarios. Then,
-     * the guest has to deal with this manually (VIRTIO_MEM_REQ_UNPLUG_ALL).
-     */
-    virtio_mem_unplug_all(vmem);
-}
-
 static void virtio_mem_class_init(ObjectClass *klass, void *data)
 {
     DeviceClass *dc = DEVICE_CLASS(klass);
     VirtioDeviceClass *vdc = VIRTIO_DEVICE_CLASS(klass);
     VirtIOMEMClass *vmc = VIRTIO_MEM_CLASS(klass);
     RamDiscardManagerClass *rdmc = RAM_DISCARD_MANAGER_CLASS(klass);
-    ResettableClass *rc = RESETTABLE_CLASS(klass);
 
     device_class_set_props(dc, virtio_mem_properties);
     dc->vmsd = &vmstate_virtio_mem;
@@ -1898,9 +1899,6 @@ static void virtio_mem_class_init(ObjectClass *klass, void *data)
     rdmc->replay_discarded = virtio_mem_rdm_replay_discarded;
     rdmc->register_listener = virtio_mem_rdm_register_listener;
     rdmc->unregister_listener = virtio_mem_rdm_unregister_listener;
-
-    rc->get_state = virtio_mem_get_reset_state;
-    rc->phases.hold = virtio_mem_system_reset_hold;
 }
 
 static const TypeInfo virtio_mem_info = {
@@ -1923,3 +1921,48 @@ static void virtio_register_types(void)
 }
 
 type_init(virtio_register_types)
+
+OBJECT_DEFINE_SIMPLE_TYPE_WITH_INTERFACES(VirtioMemSystemReset, virtio_mem_system_reset, VIRTIO_MEM_SYSTEM_RESET, OBJECT, { TYPE_RESETTABLE_INTERFACE }, { })
+
+static void virtio_mem_system_reset_init(Object *obj)
+{
+}
+
+static void virtio_mem_system_reset_finalize(Object *obj)
+{
+}
+
+static ResettableState *virtio_mem_system_reset_get_state(Object *obj)
+{
+    VirtioMemSystemReset *vmem_reset = VIRTIO_MEM_SYSTEM_RESET(obj);
+
+    return &vmem_reset->reset_state;
+}
+
+static void virtio_mem_system_reset_hold(Object *obj, ResetType type)
+{
+    VirtioMemSystemReset *vmem_reset = VIRTIO_MEM_SYSTEM_RESET(obj);
+    VirtIOMEM *vmem = vmem_reset->vmem;
+
+    /*
+     * When waking up from standby/suspend-to-ram, do not unplug any memory.
+     */
+    if (type == RESET_TYPE_WAKEUP) {
+        return;
+    }
+
+    /*
+     * During usual resets, we will unplug all memory and shrink the usable
+     * region size. This is, however, not possible in all scenarios. Then,
+     * the guest has to deal with this manually (VIRTIO_MEM_REQ_UNPLUG_ALL).
+     */
+    virtio_mem_unplug_all(vmem);
+}
+
+static void virtio_mem_system_reset_class_init(ObjectClass *klass, void *data)
+{
+    ResettableClass *rc = RESETTABLE_CLASS(klass);
+
+    rc->get_state = virtio_mem_system_reset_get_state;
+    rc->phases.hold = virtio_mem_system_reset_hold;
+}

include/hw/s390x/s390-virtio-ccw.h

@@ -29,10 +29,19 @@ struct S390CcwMachineState {
     bool dea_key_wrap;
     bool pv;
     uint8_t loadparm[8];
+    uint64_t memory_limit;
+    uint64_t max_pagesize;
 
     SCLPDevice *sclp;
 };
 
+static inline uint64_t s390_get_memory_limit(S390CcwMachineState *s390ms)
+{
+    /* We expect to be called only after the limit was set. */
+    assert(s390ms->memory_limit);
+    return s390ms->memory_limit;
+}
+
 #define S390_PTF_REASON_NONE (0x00 << 8)
 #define S390_PTF_REASON_DONE (0x01 << 8)
 #define S390_PTF_REASON_BUSY (0x02 << 8)

include/hw/virtio/virtio-mem.h

@@ -25,6 +25,10 @@
 OBJECT_DECLARE_TYPE(VirtIOMEM, VirtIOMEMClass,
                     VIRTIO_MEM)
 
+#define TYPE_VIRTIO_MEM_SYSTEM_RESET "virtio-mem-system-reset"
+
+OBJECT_DECLARE_SIMPLE_TYPE(VirtioMemSystemReset, VIRTIO_MEM_SYSTEM_RESET)
+
 #define VIRTIO_MEM_MEMDEV_PROP "memdev"
 #define VIRTIO_MEM_NODE_PROP "node"
 #define VIRTIO_MEM_SIZE_PROP "size"
@@ -117,8 +121,15 @@ struct VirtIOMEM {
     /* listeners to notify on plug/unplug activity. */
     QLIST_HEAD(, RamDiscardListener) rdl_list;
 
-    /* State of the resettable container */
+    /* Catch system resets -> qemu_devices_reset() only. */
+    VirtioMemSystemReset *system_reset;
+};
+
+struct VirtioMemSystemReset {
+    Object parent;
+
     ResettableState reset_state;
+    VirtIOMEM *vmem;
 };
 
 struct VirtIOMEMClass {

target/s390x/cpu-system.c

@@ -255,21 +255,6 @@ unsigned int s390_cpu_set_state(uint8_t cpu_state, S390CPU *cpu)
     return s390_count_running_cpus();
 }
 
-int s390_set_memory_limit(uint64_t new_limit, uint64_t *hw_limit)
-{
-    if (kvm_enabled()) {
-        return kvm_s390_set_mem_limit(new_limit, hw_limit);
-    }
-    return 0;
-}
-
-void s390_set_max_pagesize(uint64_t pagesize, Error **errp)
-{
-    if (kvm_enabled()) {
-        kvm_s390_set_max_pagesize(pagesize, errp);
-    }
-}
-
 void s390_cmma_reset(void)
 {
     if (kvm_enabled()) {

target/s390x/cpu.h

@@ -881,8 +881,6 @@ static inline void s390_do_cpu_load_normal(CPUState *cs, run_on_cpu_data arg)
 
 /* cpu.c */
 void s390_crypto_reset(void);
-int s390_set_memory_limit(uint64_t new_limit, uint64_t *hw_limit);
-void s390_set_max_pagesize(uint64_t pagesize, Error **errp);
 void s390_cmma_reset(void);
 void s390_enable_css_support(S390CPU *cpu);
 void s390_do_cpu_set_diag318(CPUState *cs, run_on_cpu_data arg);

target/s390x/kvm/kvm.c

@@ -49,8 +49,9 @@
 #include "hw/s390x/ebcdic.h"
 #include "exec/memattrs.h"
 #include "hw/s390x/s390-virtio-ccw.h"
-#include "hw/s390x/s390-virtio-hcall.h"
+#include "hw/s390x/s390-hypercall.h"
 #include "target/s390x/kvm/pv.h"
+#include CONFIG_DEVICES
 
 #define kvm_vm_check_mem_attr(s, attr) \
     kvm_vm_check_attr(s, KVM_S390_VM_MEM_CTRL, attr)
@@ -1491,20 +1492,6 @@ static int handle_e3(S390CPU *cpu, struct kvm_run *run, uint8_t ipbl)
     return r;
 }
 
-static int handle_hypercall(S390CPU *cpu, struct kvm_run *run)
-{
-    CPUS390XState *env = &cpu->env;
-    int ret;
-
-    ret = s390_virtio_hypercall(env);
-    if (ret == -EINVAL) {
-        kvm_s390_program_interrupt(cpu, PGM_SPECIFICATION);
-        return 0;
-    }
-
-    return ret;
-}
-
 static void kvm_handle_diag_288(S390CPU *cpu, struct kvm_run *run)
 {
     uint64_t r1, r3;
@@ -1600,9 +1587,11 @@ static int handle_diag(S390CPU *cpu, struct kvm_run *run, uint32_t ipb)
     case DIAG_SET_CONTROL_PROGRAM_CODES:
         handle_diag_318(cpu, run);
         break;
+#ifdef CONFIG_S390_CCW_VIRTIO
     case DIAG_KVM_HYPERCALL:
-        r = handle_hypercall(cpu, run);
+        handle_diag_500(cpu, RA_IGNORED);
         break;
+#endif /* CONFIG_S390_CCW_VIRTIO */
     case DIAG_KVM_BREAKPOINT:
         r = handle_sw_breakpoint(cpu, run);
         break;

target/s390x/kvm/pv.c

@@ -133,7 +133,7 @@ bool s390_pv_vm_try_disable_async(S390CcwMachineState *ms)
      * If the feature is not present or if the VM is not larger than 2 GiB,
      * KVM_PV_ASYNC_CLEANUP_PREPARE fill fail; no point in attempting it.
      */
-    if ((MACHINE(ms)->maxram_size <= 2 * GiB) ||
+    if (s390_get_memory_limit(ms) <= 2 * GiB ||
         !kvm_check_extension(kvm_state, KVM_CAP_S390_PROTECTED_ASYNC_DISABLE)) {
         return false;
     }

target/s390x/tcg/misc_helper.c

@@ -36,13 +36,14 @@
 #include "system/cpus.h"
 #include "system/system.h"
 #include "hw/s390x/ebcdic.h"
-#include "hw/s390x/s390-virtio-hcall.h"
+#include "hw/s390x/s390-hypercall.h"
 #include "hw/s390x/sclp.h"
 #include "hw/s390x/s390_flic.h"
 #include "hw/s390x/ioinst.h"
 #include "hw/s390x/s390-pci-inst.h"
 #include "hw/boards.h"
 #include "hw/s390x/tod.h"
+#include CONFIG_DEVICES
 #endif
 
 /* #define DEBUG_HELPER */
@@ -116,12 +117,15 @@ void HELPER(diag)(CPUS390XState *env, uint32_t r1, uint32_t r3, uint32_t num)
     uint64_t r;
 
     switch (num) {
+#ifdef CONFIG_S390_CCW_VIRTIO
     case 0x500:
-        /* KVM hypercall */
+        /* QEMU/KVM hypercall */
         bql_lock();
-        r = s390_virtio_hypercall(env);
+        handle_diag_500(env_archcpu(env), GETPC());
         bql_unlock();
+        r = 0;
         break;
+#endif /* CONFIG_S390_CCW_VIRTIO */
     case 0x44:
         /* yield */
         r = 0;