Merge tag 'for_upstream' of https://git.kernel.org/pub/scm/virt/kvm/mst/qemu into staging

virtio,pc,pci: features, fixes, cleanups

CXL now can use Generic Port Affinity Structures.
CXL now allows control of link speed and width
vhost-user-blk now supports live resize, by means of
a new device-sync-config command
amd iommu now supports interrupt remapping
pcie devices now report extended tag field support
intel_iommu dropped support for Transient Mapping, to match VTD spec
arch agnostic ACPI infrastructure for vCPU Hotplug

Fixes, cleanups all over the place.

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

# -----BEGIN PGP SIGNATURE-----
#
# iQFDBAABCAAtFiEEXQn9CHHI+FuUyooNKB8NuNKNVGkFAmcpNqUPHG1zdEByZWRo
# YXQuY29tAAoJECgfDbjSjVRp/2oH/0qO33prhDa48J5mqT9NuJzzYwp5QHKF9Zjv
# fDAplMUEmfxZIEgJchcyDWPYTGX2geT4pCFhRWioZMIR/0JyzrFgSwsk1kL88cMh
# 46gzhNVD6ybyPJ7O0Zq3GLy5jo7rlw/n+fFxKAuRCzcbK/fmH8gNC+RwW1IP64Na
# HDczYilHUhnO7yKZFQzQNQVbK4BckrG1bu0Fcx0EMUQBf4V6x7GLOrT+3hkKYcr6
# +DG5DmUmv20or/FXnu2Ye+MzR8Ebx6JVK3A3sXEE4Ns2CCzK9QLzeeyc2aU13jWN
# OpZ6WcKF8HqYprIwnSsMTxhPcq0/c7TvrGrazVwna5RUBMyjjvc=
# =zSX4
# -----END PGP SIGNATURE-----
# gpg: Signature made Mon 04 Nov 2024 21:03:33 GMT
# gpg:                using RSA key 5D09FD0871C8F85B94CA8A0D281F0DB8D28D5469
# gpg:                issuer "mst@redhat.com"
# gpg: Good signature from "Michael S. Tsirkin <mst@kernel.org>" [full]
# gpg:                 aka "Michael S. Tsirkin <mst@redhat.com>" [full]
# Primary key fingerprint: 0270 606B 6F3C DF3D 0B17  0970 C350 3912 AFBE 8E67
#      Subkey fingerprint: 5D09 FD08 71C8 F85B 94CA  8A0D 281F 0DB8 D28D 5469

* tag 'for_upstream' of https://git.kernel.org/pub/scm/virt/kvm/mst/qemu: (65 commits)
  intel_iommu: Add missed reserved bit check for IEC descriptor
  intel_iommu: Add missed sanity check for 256-bit invalidation queue
  intel_iommu: Send IQE event when setting reserved bit in IQT_TAIL
  hw/acpi: Update GED with vCPU Hotplug VMSD for migration
  tests/qtest/bios-tables-test: Update DSDT golden masters for x86/{pc,q35}
  hw/acpi: Update ACPI `_STA` method with QOM vCPU ACPI Hotplug states
  qtest: allow ACPI DSDT Table changes
  hw/acpi: Make CPUs ACPI `presence` conditional during vCPU hot-unplug
  hw/pci: Add parenthesis to PCI_BUILD_BDF macro
  hw/cxl: Ensure there is enough data to read the input header in cmd_get_physical_port_state()
  hw/cxl: Ensure there is enough data for the header in cmd_ccls_set_lsa()
  hw/cxl: Check that writes do not go beyond end of target attributes
  hw/cxl: Ensuring enough data to read parameters in cmd_tunnel_management_cmd()
  hw/cxl: Avoid accesses beyond the end of cel_log.
  hw/cxl: Check the length of data requested fits in get_log()
  hw/cxl: Check enough data in cmd_firmware_update_transfer()
  hw/cxl: Check input length is large enough in cmd_events_clear_records()
  hw/cxl: Check input includes at least the header in cmd_features_set_feature()
  hw/cxl: Check size of input data to dynamic capacity mailbox commands
  hw/cxl/cxl-mailbox-util: Fix output buffer index update when retrieving DC extents
  ...

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

1 files changed, 79 insertions, 19 deletions

diff --git a/hw/i386/amd_iommu.c b/hw/i386/amd_iommu.c
index 464f0b6..13af721 100644
--- a/hw/i386/amd_iommu.c
+++ b/hw/i386/amd_iommu.c
@@ -32,6 +32,7 @@
 #include "trace.h"
 #include "hw/i386/apic-msidef.h"
 #include "hw/qdev-properties.h"
+#include "kvm/kvm_i386.h"
 
 /* used AMD-Vi MMIO registers */
 const char *amdvi_mmio_low[] = {
@@ -60,8 +61,9 @@ struct AMDVIAddressSpace {
     uint8_t bus_num;            /* bus number                           */
     uint8_t devfn;              /* device function                      */
     AMDVIState *iommu_state;    /* AMDVI - one per machine              */
-    MemoryRegion root;          /* AMDVI Root memory map region */
+    MemoryRegion root;          /* AMDVI Root memory map region         */
     IOMMUMemoryRegion iommu;    /* Device's address translation region  */
+    MemoryRegion iommu_nodma;   /* Alias of shared nodma memory region  */
     MemoryRegion iommu_ir;      /* Device's interrupt remapping region  */
     AddressSpace as;            /* device's corresponding address space */
 };
@@ -430,6 +432,12 @@ static void amdvi_complete_ppr(AMDVIState *s, uint64_t *cmd)
     trace_amdvi_ppr_exec();
 }
 
+static void amdvi_intremap_inval_notify_all(AMDVIState *s, bool global,
+                               uint32_t index, uint32_t mask)
+{
+    x86_iommu_iec_notify_all(X86_IOMMU_DEVICE(s), global, index, mask);
+}
+
 static void amdvi_inval_all(AMDVIState *s, uint64_t *cmd)
 {
     if (extract64(cmd[0], 0, 60) || cmd[1]) {
@@ -437,6 +445,9 @@ static void amdvi_inval_all(AMDVIState *s, uint64_t *cmd)
                                    s->cmdbuf + s->cmdbuf_head);
     }
 
+    /* Notify global invalidation */
+    amdvi_intremap_inval_notify_all(s, true, 0, 0);
+
     amdvi_iotlb_reset(s);
     trace_amdvi_all_inval();
 }
@@ -485,6 +496,9 @@ static void amdvi_inval_inttable(AMDVIState *s, uint64_t *cmd)
         return;
     }
 
+    /* Notify global invalidation */
+    amdvi_intremap_inval_notify_all(s, true, 0, 0);
+
     trace_amdvi_intr_inval();
 }
 
@@ -1412,6 +1426,7 @@ static AddressSpace *amdvi_host_dma_iommu(PCIBus *bus, void *opaque, int devfn)
     AMDVIState *s = opaque;
     AMDVIAddressSpace **iommu_as, *amdvi_dev_as;
     int bus_num = pci_bus_num(bus);
+    X86IOMMUState *x86_iommu = X86_IOMMU_DEVICE(s);
 
     iommu_as = s->address_spaces[bus_num];
 
@@ -1436,13 +1451,13 @@ static AddressSpace *amdvi_host_dma_iommu(PCIBus *bus, void *opaque, int devfn)
          * Memory region relationships looks like (Address range shows
          * only lower 32 bits to make it short in length...):
          *
-         * |-----------------+-------------------+----------|
-         * | Name            | Address range     | Priority |
-         * |-----------------+-------------------+----------+
-         * | amdvi_root      | 00000000-ffffffff |        0 |
-         * |  amdvi_iommu    | 00000000-ffffffff |        1 |
-         * |  amdvi_iommu_ir | fee00000-feefffff |       64 |
-         * |-----------------+-------------------+----------|
+         * |--------------------+-------------------+----------|
+         * | Name               | Address range     | Priority |
+         * |--------------------+-------------------+----------+
+         * | amdvi-root         | 00000000-ffffffff |        0 |
+         * |  amdvi-iommu_nodma  | 00000000-ffffffff |       0 |
+         * |  amdvi-iommu_ir     | fee00000-feefffff |       1 |
+         * |--------------------+-------------------+----------|
          */
         memory_region_init_iommu(&amdvi_dev_as->iommu,
                                  sizeof(amdvi_dev_as->iommu),
@@ -1452,16 +1467,34 @@ static AddressSpace *amdvi_host_dma_iommu(PCIBus *bus, void *opaque, int devfn)
         memory_region_init(&amdvi_dev_as->root, OBJECT(s),
                            "amdvi_root", UINT64_MAX);
         address_space_init(&amdvi_dev_as->as, &amdvi_dev_as->root, name);
-        memory_region_init_io(&amdvi_dev_as->iommu_ir, OBJECT(s),
-                              &amdvi_ir_ops, s, "amd_iommu_ir",
-                              AMDVI_INT_ADDR_SIZE);
-        memory_region_add_subregion_overlap(&amdvi_dev_as->root,
-                                            AMDVI_INT_ADDR_FIRST,
-                                            &amdvi_dev_as->iommu_ir,
-                                            64);
         memory_region_add_subregion_overlap(&amdvi_dev_as->root, 0,
                                             MEMORY_REGION(&amdvi_dev_as->iommu),
-                                            1);
+                                            0);
+
+        /* Build the DMA Disabled alias to shared memory */
+        memory_region_init_alias(&amdvi_dev_as->iommu_nodma, OBJECT(s),
+                                 "amdvi-sys", &s->mr_sys, 0,
+                                 memory_region_size(&s->mr_sys));
+        memory_region_add_subregion_overlap(&amdvi_dev_as->root, 0,
+                                            &amdvi_dev_as->iommu_nodma,
+                                            0);
+        /* Build the Interrupt Remapping alias to shared memory */
+        memory_region_init_alias(&amdvi_dev_as->iommu_ir, OBJECT(s),
+                                 "amdvi-ir", &s->mr_ir, 0,
+                                 memory_region_size(&s->mr_ir));
+        memory_region_add_subregion_overlap(MEMORY_REGION(&amdvi_dev_as->iommu),
+                                            AMDVI_INT_ADDR_FIRST,
+                                            &amdvi_dev_as->iommu_ir, 1);
+
+        if (!x86_iommu->pt_supported) {
+            memory_region_set_enabled(&amdvi_dev_as->iommu_nodma, false);
+            memory_region_set_enabled(MEMORY_REGION(&amdvi_dev_as->iommu),
+                                      true);
+        } else {
+            memory_region_set_enabled(MEMORY_REGION(&amdvi_dev_as->iommu),
+                                      false);
+            memory_region_set_enabled(&amdvi_dev_as->iommu_nodma, true);
+        }
     }
     return &iommu_as[devfn]->as;
 }
@@ -1598,10 +1631,37 @@ static void amdvi_sysbus_realize(DeviceState *dev, Error **errp)
     x86ms->ioapic_as = amdvi_host_dma_iommu(bus, s, AMDVI_IOAPIC_SB_DEVID);
 
     /* set up MMIO */
-    memory_region_init_io(&s->mmio, OBJECT(s), &mmio_mem_ops, s, "amdvi-mmio",
-                          AMDVI_MMIO_SIZE);
+    memory_region_init_io(&s->mr_mmio, OBJECT(s), &mmio_mem_ops, s,
+                          "amdvi-mmio", AMDVI_MMIO_SIZE);
     memory_region_add_subregion(get_system_memory(), AMDVI_BASE_ADDR,
-                                &s->mmio);
+                                &s->mr_mmio);
+
+    /* Create the share memory regions by all devices */
+    memory_region_init(&s->mr_sys, OBJECT(s), "amdvi-sys", UINT64_MAX);
+
+    /* set up the DMA disabled memory region */
+    memory_region_init_alias(&s->mr_nodma, OBJECT(s),
+                             "amdvi-nodma", get_system_memory(), 0,
+                             memory_region_size(get_system_memory()));
+    memory_region_add_subregion_overlap(&s->mr_sys, 0,
+                                        &s->mr_nodma, 0);
+
+    /* set up the Interrupt Remapping memory region */
+    memory_region_init_io(&s->mr_ir, OBJECT(s), &amdvi_ir_ops,
+                          s, "amdvi-ir", AMDVI_INT_ADDR_SIZE);
+    memory_region_add_subregion_overlap(&s->mr_sys, AMDVI_INT_ADDR_FIRST,
+                                        &s->mr_ir, 1);
+
+    /* AMD IOMMU with x2APIC mode requires xtsup=on */
+    if (x86ms->apic_id_limit > 255 && !s->xtsup) {
+        error_report("AMD IOMMU with x2APIC confguration requires xtsup=on");
+        exit(EXIT_FAILURE);
+    }
+    if (s->xtsup && kvm_irqchip_is_split() && !kvm_enable_x2apic()) {
+        error_report("AMD IOMMU xtsup=on requires support on the KVM side");
+        exit(EXIT_FAILURE);
+    }
+
     pci_setup_iommu(bus, &amdvi_iommu_ops, s);
     amdvi_init(s);
 }