Merge tag 'pull-riscv-to-apply-20241220' of https://github.com/alistair23/qemu into staging

RISC-V PR for 10.0

* Correct the validness check of iova
* Fix APLIC in_clrip and clripnum write emulation
* Support riscv-iommu-sys device
* Add Tenstorrent Ascalon CPU
* Add AIA userspace irqchip_split support
* Add Microblaze V generic board
* Upgrade ACPI SPCR table to support SPCR table revision 4 format
* Remove tswap64() calls from HTIF
* Support 64-bit address of initrd
* Introduce svukte ISA extension
* Support ssstateen extension
* Support for RV64 Xiangshan Nanhu CPU

 # -----BEGIN PGP SIGNATURE-----
 #
 # iQIzBAABCAAdFiEEaukCtqfKh31tZZKWr3yVEwxTgBMFAmdkzjgACgkQr3yVEwxT
 # gBOcyA//e0XhAQciQglCZZCfINdOyI8qSh+P2K0qtrXZ4VERHEMp7UoD5CQr2cZv
 # h8ij1EkatXCwukVELx0rNckxG33bEFgG1oESnQSrwGE0Iu4csNW24nK5WlUS0/r+
 # A5oD2wtzEF+cbhTKrVSDBN/PvlnWTKGEoJRkuXWfz5d4uR9eyQhfED0S2j36lNEC
 # X1x/OZoKM89XuXtOFe9g55Z5UNzAatcdTISozL0FydiPh7QeVjTLHh28/tt559MX
 # 7v5aJFlQuZ78z1mIHkZmPSorSrJ0zqhkP6NWe1ae06oMgzwRQQhYLppDILV4ZgUF
 # 3mSDRoXmBycQXiYNPcHep3LdXfvxr+PpWHSevx8gH1jwm93On7Y/H7Uol6TDXzfC
 # mrFjalfV5tzrD90ZvB+s5bCMF1q5Z8Dlj0pYF9aN9P1ILoWy3dndFAPJB6uKKDP7
 # Qd4qOQ3dVyHAX9jLmVkB6QvAV/vTDrYTsAxaF/EaoLOy0IoKhjTvgda3XzE1MFKA
 # gVafLluADIfSEdqa2QR2ExL8d1SZVoiObCp5TMLRer0HIpg/vQZwjfdbo4BgQKL3
 # 7Q6wBxcZUNqrFgspXjm5WFIrdk2rfS/79OmvpNM6SZaK6BnklntdJHJHtAWujGsm
 # EM310AUFpHMp2h6Nqnemb3qr5l4d20KSt8DhoPAUq1IE59Kb8XY=
 # =0iQW
 # -----END PGP SIGNATURE-----
 # gpg: Signature made Thu 19 Dec 2024 20:54:00 EST
 # gpg:                using RSA key 6AE902B6A7CA877D6D659296AF7C95130C538013
 # gpg: Good signature from "Alistair Francis <alistair@alistair23.me>" [unknown]
 # gpg: WARNING: This key is not certified with a trusted signature!
 # gpg:          There is no indication that the signature belongs to the owner.
 # Primary key fingerprint: 6AE9 02B6 A7CA 877D 6D65  9296 AF7C 9513 0C53 8013

* tag 'pull-riscv-to-apply-20241220' of https://github.com/alistair23/qemu: (39 commits)
  target/riscv: add support for RV64 Xiangshan Nanhu CPU
  target/riscv: add ssstateen
  target/riscv/tcg: hide warn for named feats when disabling via priv_ver
  target/riscv: Include missing headers in 'internals.h'
  target/riscv: Include missing headers in 'vector_internals.h'
  target/riscv: Check svukte is not enabled in RV32
  target/riscv: Expose svukte ISA extension
  target/riscv: Check memory access to meet svukte rule
  target/riscv: Support hstatus[HUKTE] bit when svukte extension is enabled
  target/riscv: Support senvcfg[UKTE] bit when svukte extension is enabled
  target/riscv: Add svukte extension capability variable
  hw/riscv: Add the checking if DTB overlaps to kernel or initrd
  hw/riscv: Add a new struct RISCVBootInfo
  hw/riscv: Support to load DTB after 3GB memory on 64-bit system.
  hw/char/riscv_htif: Clarify MemoryRegionOps expect 32-bit accesses
  hw/char/riscv_htif: Explicit little-endian implementation
  MAINTAINERS: Cover RISC-V HTIF interface
  tests/qtest/bios-tables-test: Update virt SPCR golden reference for RISC-V
  hw/acpi: Upgrade ACPI SPCR table to support SPCR table revision 4 format
  qtest: allow SPCR acpi table changes
  ...

Conflicts:
  target/riscv/cpu.c

  Merge conflict with DEFINE_PROP_END_OF_LIST() removal. No Property
  array terminator is needed anymore.

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

1 files changed, 140 insertions, 19 deletions

diff --git a/hw/riscv/virt.c b/hw/riscv/virt.c
index 2feb851f15..d2bfdec56a 100644
--- a/hw/riscv/virt.c
+++ b/hw/riscv/virt.c
@@ -33,6 +33,7 @@
 #include "target/riscv/pmu.h"
 #include "hw/riscv/riscv_hart.h"
 #include "hw/riscv/iommu.h"
+#include "hw/riscv/riscv-iommu-bits.h"
 #include "hw/riscv/virt.h"
 #include "hw/riscv/boot.h"
 #include "hw/riscv/numa.h"
@@ -58,9 +59,18 @@
 #include "hw/virtio/virtio-iommu.h"
 
 /* KVM AIA only supports APLIC MSI. APLIC Wired is always emulated by QEMU. */
-static bool virt_use_kvm_aia(RISCVVirtState *s)
+static bool virt_use_kvm_aia_aplic_imsic(RISCVVirtAIAType aia_type)
 {
-    return kvm_irqchip_in_kernel() && s->aia_type == VIRT_AIA_TYPE_APLIC_IMSIC;
+    bool msimode = aia_type == VIRT_AIA_TYPE_APLIC_IMSIC;
+
+    return riscv_is_kvm_aia_aplic_imsic(msimode);
+}
+
+static bool virt_use_emulated_aplic(RISCVVirtAIAType aia_type)
+{
+    bool msimode = aia_type == VIRT_AIA_TYPE_APLIC_IMSIC;
+
+    return riscv_use_emulated_aplic(msimode);
 }
 
 static bool virt_aclint_allowed(void)
@@ -76,6 +86,7 @@ static const MemMapEntry virt_memmap[] = {
     [VIRT_CLINT] =        {  0x2000000,       0x10000 },
     [VIRT_ACLINT_SSWI] =  {  0x2F00000,        0x4000 },
     [VIRT_PCIE_PIO] =     {  0x3000000,       0x10000 },
+    [VIRT_IOMMU_SYS] =    {  0x3010000,        0x1000 },
     [VIRT_PLATFORM_BUS] = {  0x4000000,     0x2000000 },
     [VIRT_PLIC] =         {  0xc000000, VIRT_PLIC_SIZE(VIRT_CPUS_MAX * 2) },
     [VIRT_APLIC_M] =      {  0xc000000, APLIC_SIZE(VIRT_CPUS_MAX) },
@@ -775,12 +786,19 @@ static void create_fdt_sockets(RISCVVirtState *s, const MemMapEntry *memmap,
         *msi_pcie_phandle = msi_s_phandle;
     }
 
-    /* KVM AIA only has one APLIC instance */
-    if (kvm_enabled() && virt_use_kvm_aia(s)) {
+    /*
+     * With KVM AIA aplic-imsic, using an irqchip without split
+     * mode, we'll use only one APLIC instance.
+     */
+    if (!virt_use_emulated_aplic(s->aia_type)) {
         create_fdt_socket_aplic(s, memmap, 0,
                                 msi_m_phandle, msi_s_phandle, phandle,
                                 &intc_phandles[0], xplic_phandles,
                                 ms->smp.cpus);
+
+        *irq_mmio_phandle = xplic_phandles[0];
+        *irq_virtio_phandle = xplic_phandles[0];
+        *irq_pcie_phandle = xplic_phandles[0];
     } else {
         phandle_pos = ms->smp.cpus;
         for (socket = (socket_count - 1); socket >= 0; socket--) {
@@ -798,13 +816,7 @@ static void create_fdt_sockets(RISCVVirtState *s, const MemMapEntry *memmap,
                                         s->soc[socket].num_harts);
             }
         }
-    }
 
-    if (kvm_enabled() && virt_use_kvm_aia(s)) {
-        *irq_mmio_phandle = xplic_phandles[0];
-        *irq_virtio_phandle = xplic_phandles[0];
-        *irq_pcie_phandle = xplic_phandles[0];
-    } else {
         for (socket = 0; socket < socket_count; socket++) {
             if (socket == 0) {
                 *irq_mmio_phandle = xplic_phandles[socket];
@@ -853,7 +865,8 @@ static void create_fdt_virtio(RISCVVirtState *s, const MemMapEntry *memmap,
 
 static void create_fdt_pcie(RISCVVirtState *s, const MemMapEntry *memmap,
                             uint32_t irq_pcie_phandle,
-                            uint32_t msi_pcie_phandle)
+                            uint32_t msi_pcie_phandle,
+                            uint32_t iommu_sys_phandle)
 {
     g_autofree char *name = NULL;
     MachineState *ms = MACHINE(s);
@@ -887,6 +900,12 @@ static void create_fdt_pcie(RISCVVirtState *s, const MemMapEntry *memmap,
         2, virt_high_pcie_memmap.base,
         2, virt_high_pcie_memmap.base, 2, virt_high_pcie_memmap.size);
 
+    if (virt_is_iommu_sys_enabled(s)) {
+        qemu_fdt_setprop_cells(ms->fdt, name, "iommu-map",
+                               0, iommu_sys_phandle, 0, 0, 0,
+                               iommu_sys_phandle, 0, 0xffff);
+    }
+
     create_pcie_irq_map(s, ms->fdt, name, irq_pcie_phandle);
 }
 
@@ -1033,6 +1052,47 @@ static void create_fdt_virtio_iommu(RISCVVirtState *s, uint16_t bdf)
                            bdf + 1, iommu_phandle, bdf + 1, 0xffff - bdf);
 }
 
+static void create_fdt_iommu_sys(RISCVVirtState *s, uint32_t irq_chip,
+                                 uint32_t msi_phandle,
+                                 uint32_t *iommu_sys_phandle)
+{
+    const char comp[] = "riscv,iommu";
+    void *fdt = MACHINE(s)->fdt;
+    uint32_t iommu_phandle;
+    g_autofree char *iommu_node = NULL;
+    hwaddr addr = s->memmap[VIRT_IOMMU_SYS].base;
+    hwaddr size = s->memmap[VIRT_IOMMU_SYS].size;
+    uint32_t iommu_irq_map[RISCV_IOMMU_INTR_COUNT] = {
+        IOMMU_SYS_IRQ + RISCV_IOMMU_INTR_CQ,
+        IOMMU_SYS_IRQ + RISCV_IOMMU_INTR_FQ,
+        IOMMU_SYS_IRQ + RISCV_IOMMU_INTR_PM,
+        IOMMU_SYS_IRQ + RISCV_IOMMU_INTR_PQ,
+    };
+
+    iommu_node = g_strdup_printf("/soc/iommu@%x",
+                               (unsigned int) s->memmap[VIRT_IOMMU_SYS].base);
+    iommu_phandle = qemu_fdt_alloc_phandle(fdt);
+    qemu_fdt_add_subnode(fdt, iommu_node);
+
+    qemu_fdt_setprop(fdt, iommu_node, "compatible", comp, sizeof(comp));
+    qemu_fdt_setprop_cell(fdt, iommu_node, "#iommu-cells", 1);
+    qemu_fdt_setprop_cell(fdt, iommu_node, "phandle", iommu_phandle);
+
+    qemu_fdt_setprop_cells(fdt, iommu_node, "reg",
+                           addr >> 32, addr, size >> 32, size);
+    qemu_fdt_setprop_cell(fdt, iommu_node, "interrupt-parent", irq_chip);
+
+    qemu_fdt_setprop_cells(fdt, iommu_node, "interrupts",
+        iommu_irq_map[0], FDT_IRQ_TYPE_EDGE_LOW,
+        iommu_irq_map[1], FDT_IRQ_TYPE_EDGE_LOW,
+        iommu_irq_map[2], FDT_IRQ_TYPE_EDGE_LOW,
+        iommu_irq_map[3], FDT_IRQ_TYPE_EDGE_LOW);
+
+    qemu_fdt_setprop_cell(fdt, iommu_node, "msi-parent", msi_phandle);
+
+    *iommu_sys_phandle = iommu_phandle;
+}
+
 static void create_fdt_iommu(RISCVVirtState *s, uint16_t bdf)
 {
     const char comp[] = "riscv,pci-iommu";
@@ -1061,6 +1121,7 @@ static void finalize_fdt(RISCVVirtState *s)
 {
     uint32_t phandle = 1, irq_mmio_phandle = 1, msi_pcie_phandle = 1;
     uint32_t irq_pcie_phandle = 1, irq_virtio_phandle = 1;
+    uint32_t iommu_sys_phandle = 1;
 
     create_fdt_sockets(s, virt_memmap, &phandle, &irq_mmio_phandle,
                        &irq_pcie_phandle, &irq_virtio_phandle,
@@ -1068,7 +1129,12 @@ static void finalize_fdt(RISCVVirtState *s)
 
     create_fdt_virtio(s, virt_memmap, irq_virtio_phandle);
 
-    create_fdt_pcie(s, virt_memmap, irq_pcie_phandle, msi_pcie_phandle);
+    if (virt_is_iommu_sys_enabled(s)) {
+        create_fdt_iommu_sys(s, irq_mmio_phandle, msi_pcie_phandle,
+                             &iommu_sys_phandle);
+    }
+    create_fdt_pcie(s, virt_memmap, irq_pcie_phandle, msi_pcie_phandle,
+                    iommu_sys_phandle);
 
     create_fdt_reset(s, virt_memmap, &phandle);
 
@@ -1234,7 +1300,7 @@ static DeviceState *virt_create_aia(RISCVVirtAIAType aia_type, int aia_guests,
                                     int base_hartid, int hart_count)
 {
     int i;
-    hwaddr addr;
+    hwaddr addr = 0;
     uint32_t guest_bits;
     DeviceState *aplic_s = NULL;
     DeviceState *aplic_m = NULL;
@@ -1284,6 +1350,10 @@ static DeviceState *virt_create_aia(RISCVVirtAIAType aia_type, int aia_guests,
                                  VIRT_IRQCHIP_NUM_PRIO_BITS,
                                  msimode, false, aplic_m);
 
+    if (kvm_enabled() && msimode) {
+        riscv_aplic_set_kvm_msicfgaddr(RISCV_APLIC(aplic_s), addr);
+    }
+
     return kvm_enabled() ? aplic_s : aplic_m;
 }
 
@@ -1364,6 +1434,7 @@ static void virt_machine_done(Notifier *notifier, void *data)
     uint64_t fdt_load_addr;
     uint64_t kernel_entry = 0;
     BlockBackend *pflash_blk0;
+    RISCVBootInfo boot_info;
 
     /*
      * An user provided dtb must include everything, including
@@ -1412,17 +1483,19 @@ static void virt_machine_done(Notifier *notifier, void *data)
         }
     }
 
+    riscv_boot_info_init(&boot_info, &s->soc[0]);
+
     if (machine->kernel_filename && !kernel_entry) {
-        kernel_start_addr = riscv_calc_kernel_start_addr(&s->soc[0],
+        kernel_start_addr = riscv_calc_kernel_start_addr(&boot_info,
                                                          firmware_end_addr);
-
-        kernel_entry = riscv_load_kernel(machine, &s->soc[0],
-                                         kernel_start_addr, true, NULL);
+        riscv_load_kernel(machine, &boot_info, kernel_start_addr,
+                          true, NULL);
+        kernel_entry = boot_info.image_low_addr;
     }
 
     fdt_load_addr = riscv_compute_fdt_addr(memmap[VIRT_DRAM].base,
                                            memmap[VIRT_DRAM].size,
-                                           machine);
+                                           machine, &boot_info);
     riscv_load_fdt(fdt_load_addr, machine->fdt);
 
     /* load the reset vector */
@@ -1565,7 +1638,7 @@ static void virt_machine_init(MachineState *machine)
         }
     }
 
-    if (kvm_enabled() && virt_use_kvm_aia(s)) {
+    if (kvm_enabled() && virt_use_kvm_aia_aplic_imsic(s->aia_type)) {
         kvm_riscv_aia_create(machine, IMSIC_MMIO_GROUP_MIN_SHIFT,
                              VIRT_IRQCHIP_NUM_SOURCES, VIRT_IRQCHIP_NUM_MSIS,
                              memmap[VIRT_APLIC_S].base,
@@ -1648,6 +1721,22 @@ static void virt_machine_init(MachineState *machine)
         create_fdt(s, memmap);
     }
 
+    if (virt_is_iommu_sys_enabled(s)) {
+        DeviceState *iommu_sys = qdev_new(TYPE_RISCV_IOMMU_SYS);
+
+        object_property_set_uint(OBJECT(iommu_sys), "addr",
+                                 s->memmap[VIRT_IOMMU_SYS].base,
+                                 &error_fatal);
+        object_property_set_uint(OBJECT(iommu_sys), "base-irq",
+                                 IOMMU_SYS_IRQ,
+                                 &error_fatal);
+        object_property_set_link(OBJECT(iommu_sys), "irqchip",
+                                 OBJECT(mmio_irqchip),
+                                 &error_fatal);
+
+        sysbus_realize_and_unref(SYS_BUS_DEVICE(iommu_sys), &error_fatal);
+    }
+
     s->machine_done.notify = virt_machine_done;
     qemu_add_machine_init_done_notifier(&s->machine_done);
 }
@@ -1661,6 +1750,7 @@ static void virt_machine_instance_init(Object *obj)
     s->oem_id = g_strndup(ACPI_BUILD_APPNAME6, 6);
     s->oem_table_id = g_strndup(ACPI_BUILD_APPNAME8, 8);
     s->acpi = ON_OFF_AUTO_AUTO;
+    s->iommu_sys = ON_OFF_AUTO_AUTO;
 }
 
 static char *virt_get_aia_guests(Object *obj, Error **errp)
@@ -1733,6 +1823,28 @@ static void virt_set_aclint(Object *obj, bool value, Error **errp)
     s->have_aclint = value;
 }
 
+bool virt_is_iommu_sys_enabled(RISCVVirtState *s)
+{
+    return s->iommu_sys == ON_OFF_AUTO_ON;
+}
+
+static void virt_get_iommu_sys(Object *obj, Visitor *v, const char *name,
+                               void *opaque, Error **errp)
+{
+    RISCVVirtState *s = RISCV_VIRT_MACHINE(obj);
+    OnOffAuto iommu_sys = s->iommu_sys;
+
+    visit_type_OnOffAuto(v, name, &iommu_sys, errp);
+}
+
+static void virt_set_iommu_sys(Object *obj, Visitor *v, const char *name,
+                               void *opaque, Error **errp)
+{
+    RISCVVirtState *s = RISCV_VIRT_MACHINE(obj);
+
+    visit_type_OnOffAuto(v, name, &s->iommu_sys, errp);
+}
+
 bool virt_is_acpi_enabled(RISCVVirtState *s)
 {
     return s->acpi != ON_OFF_AUTO_OFF;
@@ -1759,10 +1871,12 @@ static HotplugHandler *virt_machine_get_hotplug_handler(MachineState *machine,
                                                         DeviceState *dev)
 {
     MachineClass *mc = MACHINE_GET_CLASS(machine);
+    RISCVVirtState *s = RISCV_VIRT_MACHINE(machine);
 
     if (device_is_dynamic_sysbus(mc, dev) ||
         object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_IOMMU_PCI) ||
         object_dynamic_cast(OBJECT(dev), TYPE_RISCV_IOMMU_PCI)) {
+        s->iommu_sys = ON_OFF_AUTO_OFF;
         return HOTPLUG_HANDLER(machine);
     }
 
@@ -1789,6 +1903,7 @@ static void virt_machine_device_plug_cb(HotplugHandler *hotplug_dev,
 
     if (object_dynamic_cast(OBJECT(dev), TYPE_RISCV_IOMMU_PCI)) {
         create_fdt_iommu(s, pci_get_bdf(PCI_DEVICE(dev)));
+        s->iommu_sys = ON_OFF_AUTO_OFF;
     }
 }
 
@@ -1851,6 +1966,12 @@ static void virt_machine_class_init(ObjectClass *oc, void *data)
                               NULL, NULL);
     object_class_property_set_description(oc, "acpi",
                                           "Enable ACPI");
+
+    object_class_property_add(oc, "iommu-sys", "OnOffAuto",
+                              virt_get_iommu_sys, virt_set_iommu_sys,
+                              NULL, NULL);
+    object_class_property_set_description(oc, "iommu-sys",
+                                          "Enable IOMMU platform device");
 }
 
 static const TypeInfo virt_machine_typeinfo = {