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

virtio,pc,pci: features, cleanups, fixes

vhost-user support without ioeventfd
word replacements in vhost user spec
shpc improvements

cleanups, fixes all over the place

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

# -----BEGIN PGP SIGNATURE-----
#
# iQFDBAABCAAtFiEEXQn9CHHI+FuUyooNKB8NuNKNVGkFAmQBO8QPHG1zdEByZWRo
# YXQuY29tAAoJECgfDbjSjVRpMUMH/3/FVp4qaF4CDwCHn7xWFRJpOREIhX/iWfUu
# lGkwxnB7Lfyqdg7i4CAfgMf2emWKZchEE2DamfCo5bIX0IgRU3DWcOdR9ePvJ29J
# cKwIYpxZcB4RYSoWL5OUakQLCT3JOu4XWaXeVjyHABjQhf3lGpwN4KmIOBGOy/N6
# 0YHOQScW2eW62wIOwhAEuYQceMt6KU32Uw3tLnMbJliiBf3a/hPctVNM9TFY9pcd
# UYHGfBx/zD45owf1lTVEQFDg0eqPZKWW29g5haiOd5oAyXHHolzu+bt3bU7lH46b
# f7iP12LqDudyrgoF5YWv3NJ4HaGm5V3kPqNqLLF/mjF7alxG+N8=
# =hN3h
# -----END PGP SIGNATURE-----
# gpg: Signature made Fri 03 Mar 2023 00:13:56 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: (53 commits)
  tests/data/acpi/virt: drop (most) duplicate files.
  hw/cxl/mailbox: Use new UUID network order define for cel_uuid
  qemu/uuid: Add UUID static initializer
  qemu/bswap: Add const_le64()
  tests: acpi: Update q35/DSDT.cxl for removed duplicate UID
  hw/i386/acpi: Drop duplicate _UID entry for CXL root bridge
  tests/acpi: Allow update of q35/DSDT.cxl
  hw/cxl: Add CXL_CAPACITY_MULTIPLIER definition
  hw/cxl: set cxl-type3 device type to PCI_CLASS_MEMORY_CXL
  hw/pci-bridge/cxl_downstream: Fix type naming mismatch
  hw/mem/cxl_type3: Improve error handling in realize()
  MAINTAINERS: Add Fan Ni as Compute eXpress Link QEMU reviewer
  intel-iommu: send UNMAP notifications for domain or global inv desc
  smmu: switch to use memory_region_unmap_iommu_notifier_range()
  memory: introduce memory_region_unmap_iommu_notifier_range()
  intel-iommu: fail DEVIOTLB_UNMAP without dt mode
  intel-iommu: fail MAP notifier without caching mode
  memory: Optimize replay of guest mapping
  chardev/char-socket: set s->listener = NULL in char_socket_finalize
  hw/pci: Trace IRQ routing on PCI topology
  ...

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

4 files changed, 84 insertions, 62 deletions

diff --git a/hw/pci/pci.c b/hw/pci/pci.c
index 10c980b..034fe49 100644
--- a/hw/pci/pci.c
+++ b/hw/pci/pci.c
@@ -282,9 +282,13 @@ static void pci_change_irq_level(PCIDevice *pci_dev, int irq_num, int change)
 {
     PCIBus *bus;
     for (;;) {
+        int dev_irq = irq_num;
         bus = pci_get_bus(pci_dev);
         assert(bus->map_irq);
         irq_num = bus->map_irq(pci_dev, irq_num);
+        trace_pci_route_irq(dev_irq, DEVICE(pci_dev)->canonical_path, irq_num,
+                            pci_bus_is_root(bus) ? "root-complex"
+                                    : DEVICE(bus->parent_dev)->canonical_path);
         if (bus->set_irq)
             break;
         pci_dev = bus->parent_dev;
@@ -1617,8 +1621,12 @@ PCIINTxRoute pci_device_route_intx_to_irq(PCIDevice *dev, int pin)
     PCIBus *bus;
 
     do {
+        int dev_irq = pin;
         bus = pci_get_bus(dev);
         pin = bus->map_irq(dev, pin);
+        trace_pci_route_irq(dev_irq, DEVICE(dev)->canonical_path, pin,
+                            pci_bus_is_root(bus) ? "root-complex"
+                                    : DEVICE(bus->parent_dev)->canonical_path);
         dev = bus->parent_dev;
     } while (dev);
 
diff --git a/hw/pci/pcie.c b/hw/pci/pcie.c
index 924fdabd..b8c24cf 100644
--- a/hw/pci/pcie.c
+++ b/hw/pci/pcie.c
@@ -39,6 +39,11 @@
 #define PCIE_DEV_PRINTF(dev, fmt, ...)                                  \
     PCIE_DPRINTF("%s:%x "fmt, (dev)->name, (dev)->devfn, ## __VA_ARGS__)
 
+static bool pcie_sltctl_powered_off(uint16_t sltctl)
+{
+    return (sltctl & PCI_EXP_SLTCTL_PCC) == PCI_EXP_SLTCTL_PWR_OFF
+        && (sltctl & PCI_EXP_SLTCTL_PIC) == PCI_EXP_SLTCTL_PWR_IND_OFF;
+}
 
 /***************************************************************************
  * pci express capability helper functions
@@ -373,8 +378,8 @@ void pcie_cap_slot_enable_power(PCIDevice *dev)
     uint32_t sltcap = pci_get_long(exp_cap + PCI_EXP_SLTCAP);
 
     if (sltcap & PCI_EXP_SLTCAP_PCP) {
-        pci_set_word_by_mask(exp_cap + PCI_EXP_SLTCTL,
-                             PCI_EXP_SLTCTL_PCC, PCI_EXP_SLTCTL_PWR_ON);
+        pci_word_test_and_clear_mask(exp_cap + PCI_EXP_SLTCTL,
+                                     PCI_EXP_SLTCTL_PCC);
     }
 }
 
@@ -395,6 +400,7 @@ static void pcie_cap_update_power(PCIDevice *hotplug_dev)
 
     if (sltcap & PCI_EXP_SLTCAP_PCP) {
         power = (sltctl & PCI_EXP_SLTCTL_PCC) == PCI_EXP_SLTCTL_PWR_ON;
+        /* Don't we need to check also (sltctl & PCI_EXP_SLTCTL_PIC) ? */
     }
 
     pci_for_each_device(sec_bus, pci_bus_num(sec_bus),
@@ -579,8 +585,7 @@ void pcie_cap_slot_unplug_request_cb(HotplugHandler *hotplug_dev,
         return;
     }
 
-    if (((sltctl & PCI_EXP_SLTCTL_PIC) == PCI_EXP_SLTCTL_PWR_IND_OFF) &&
-        ((sltctl & PCI_EXP_SLTCTL_PCC) == PCI_EXP_SLTCTL_PWR_OFF)) {
+    if (pcie_sltctl_powered_off(sltctl)) {
         /* slot is powered off -> unplug without round-trip to the guest */
         pcie_cap_slot_do_unplug(hotplug_pdev);
         hotplug_event_notify(hotplug_pdev);
@@ -634,8 +639,8 @@ void pcie_cap_slot_init(PCIDevice *dev, PCIESlot *s)
                                  PCI_EXP_SLTCTL_PIC |
                                  PCI_EXP_SLTCTL_AIC);
     pci_word_test_and_set_mask(dev->config + pos + PCI_EXP_SLTCTL,
-                               PCI_EXP_SLTCTL_PIC_OFF |
-                               PCI_EXP_SLTCTL_AIC_OFF);
+                               PCI_EXP_SLTCTL_PWR_IND_OFF |
+                               PCI_EXP_SLTCTL_ATTN_IND_OFF);
     pci_word_test_and_set_mask(dev->wmask + pos + PCI_EXP_SLTCTL,
                                PCI_EXP_SLTCTL_PIC |
                                PCI_EXP_SLTCTL_AIC |
@@ -679,7 +684,8 @@ void pcie_cap_slot_reset(PCIDevice *dev)
                                  PCI_EXP_SLTCTL_PDCE |
                                  PCI_EXP_SLTCTL_ABPE);
     pci_word_test_and_set_mask(exp_cap + PCI_EXP_SLTCTL,
-                               PCI_EXP_SLTCTL_AIC_OFF);
+                               PCI_EXP_SLTCTL_PWR_IND_OFF |
+                               PCI_EXP_SLTCTL_ATTN_IND_OFF);
 
     if (dev->cap_present & QEMU_PCIE_SLTCAP_PCP) {
         /* Downstream ports enforce device number 0. */
@@ -694,7 +700,8 @@ void pcie_cap_slot_reset(PCIDevice *dev)
                                        PCI_EXP_SLTCTL_PCC);
         }
 
-        pic = populated ? PCI_EXP_SLTCTL_PIC_ON : PCI_EXP_SLTCTL_PIC_OFF;
+        pic = populated ?
+                PCI_EXP_SLTCTL_PWR_IND_ON : PCI_EXP_SLTCTL_PWR_IND_OFF;
         pci_word_test_and_set_mask(exp_cap + PCI_EXP_SLTCTL, pic);
     }
 
@@ -769,10 +776,9 @@ void pcie_cap_slot_write_config(PCIDevice *dev,
      * this is a work around for guests that overwrite
      * control of powered off slots before powering them on.
      */
-    if ((sltsta & PCI_EXP_SLTSTA_PDS) && (val & PCI_EXP_SLTCTL_PCC) &&
-        (val & PCI_EXP_SLTCTL_PIC_OFF) == PCI_EXP_SLTCTL_PIC_OFF &&
-        (!(old_slt_ctl & PCI_EXP_SLTCTL_PCC) ||
-        (old_slt_ctl & PCI_EXP_SLTCTL_PIC_OFF) != PCI_EXP_SLTCTL_PIC_OFF)) {
+    if ((sltsta & PCI_EXP_SLTSTA_PDS) && pcie_sltctl_powered_off(val) &&
+        !pcie_sltctl_powered_off(old_slt_ctl))
+    {
         pcie_cap_slot_do_unplug(dev);
     }
     pcie_cap_update_power(dev);
diff --git a/hw/pci/shpc.c b/hw/pci/shpc.c
index fca7f66..e7bc719 100644
--- a/hw/pci/shpc.c
+++ b/hw/pci/shpc.c
@@ -123,10 +123,13 @@
 #define SHPC_PCI_TO_IDX(pci_slot) ((pci_slot) - 1)
 #define SHPC_IDX_TO_PHYSICAL(slot) ((slot) + 1)
 
-static uint16_t shpc_get_status(SHPCDevice *shpc, int slot, uint16_t msk)
+static uint8_t shpc_get_status(SHPCDevice *shpc, int slot, uint16_t msk)
 {
     uint8_t *status = shpc->config + SHPC_SLOT_STATUS(slot);
-    return (pci_get_word(status) & msk) >> ctz32(msk);
+    uint16_t result = (pci_get_word(status) & msk) >> ctz32(msk);
+
+    assert(result <= UINT8_MAX);
+    return result;
 }
 
 static void shpc_set_status(SHPCDevice *shpc,
@@ -223,6 +226,7 @@ void shpc_reset(PCIDevice *d)
                             SHPC_SLOT_STATUS_PRSNT_MASK);
             shpc_set_status(shpc, i, SHPC_LED_OFF, SHPC_SLOT_PWR_LED_MASK);
         }
+        shpc_set_status(shpc, i, SHPC_LED_OFF, SHPC_SLOT_ATTN_LED_MASK);
         shpc_set_status(shpc, i, 0, SHPC_SLOT_STATUS_66);
     }
     shpc_set_sec_bus_speed(shpc, SHPC_SEC_BUS_33);
@@ -254,60 +258,66 @@ static void shpc_free_devices_in_slot(SHPCDevice *shpc, int slot)
     }
 }
 
-static void shpc_slot_command(SHPCDevice *shpc, uint8_t target,
+static bool shpc_slot_is_off(uint8_t state, uint8_t power, uint8_t attn)
+{
+    return state == SHPC_STATE_DISABLED && power == SHPC_LED_OFF;
+}
+
+static void shpc_slot_command(PCIDevice *d, uint8_t target,
                               uint8_t state, uint8_t power, uint8_t attn)
 {
-    uint8_t current_state;
+    SHPCDevice *shpc = d->shpc;
     int slot = SHPC_LOGICAL_TO_IDX(target);
+    uint8_t old_state = shpc_get_status(shpc, slot, SHPC_SLOT_STATE_MASK);
+    uint8_t old_power = shpc_get_status(shpc, slot, SHPC_SLOT_PWR_LED_MASK);
+    uint8_t old_attn = shpc_get_status(shpc, slot, SHPC_SLOT_ATTN_LED_MASK);
+
     if (target < SHPC_CMD_TRGT_MIN || slot >= shpc->nslots) {
         shpc_invalid_command(shpc);
         return;
     }
-    current_state = shpc_get_status(shpc, slot, SHPC_SLOT_STATE_MASK);
-    if (current_state == SHPC_STATE_ENABLED && state == SHPC_STATE_PWRONLY) {
+
+    if (old_state == SHPC_STATE_ENABLED && state == SHPC_STATE_PWRONLY) {
         shpc_invalid_command(shpc);
         return;
     }
 
-    switch (power) {
-    case SHPC_LED_NO:
-        break;
-    default:
+    if (power == SHPC_LED_NO) {
+        power = old_power;
+    } else {
         /* TODO: send event to monitor */
         shpc_set_status(shpc, slot, power, SHPC_SLOT_PWR_LED_MASK);
     }
-    switch (attn) {
-    case SHPC_LED_NO:
-        break;
-    default:
+
+    if (attn == SHPC_LED_NO) {
+        attn = old_attn;
+    } else {
         /* TODO: send event to monitor */
         shpc_set_status(shpc, slot, attn, SHPC_SLOT_ATTN_LED_MASK);
     }
 
-    if ((current_state == SHPC_STATE_DISABLED && state == SHPC_STATE_PWRONLY) ||
-        (current_state == SHPC_STATE_DISABLED && state == SHPC_STATE_ENABLED)) {
-        shpc_set_status(shpc, slot, state, SHPC_SLOT_STATE_MASK);
-    } else if ((current_state == SHPC_STATE_ENABLED ||
-                current_state == SHPC_STATE_PWRONLY) &&
-               state == SHPC_STATE_DISABLED) {
+    if (state == SHPC_STATE_NO) {
+        state = old_state;
+    } else {
         shpc_set_status(shpc, slot, state, SHPC_SLOT_STATE_MASK);
-        power = shpc_get_status(shpc, slot, SHPC_SLOT_PWR_LED_MASK);
-        /* TODO: track what monitor requested. */
-        /* Look at LED to figure out whether it's ok to remove the device. */
-        if (power == SHPC_LED_OFF) {
-            shpc_free_devices_in_slot(shpc, slot);
-            shpc_set_status(shpc, slot, 1, SHPC_SLOT_STATUS_MRL_OPEN);
-            shpc_set_status(shpc, slot, SHPC_SLOT_STATUS_PRSNT_EMPTY,
-                            SHPC_SLOT_STATUS_PRSNT_MASK);
-            shpc->config[SHPC_SLOT_EVENT_LATCH(slot)] |=
-                SHPC_SLOT_EVENT_MRL |
-                SHPC_SLOT_EVENT_PRESENCE;
-        }
+    }
+
+    if (!shpc_slot_is_off(old_state, old_power, old_attn) &&
+        shpc_slot_is_off(state, power, attn))
+    {
+        shpc_free_devices_in_slot(shpc, slot);
+        shpc_set_status(shpc, slot, 1, SHPC_SLOT_STATUS_MRL_OPEN);
+        shpc_set_status(shpc, slot, SHPC_SLOT_STATUS_PRSNT_EMPTY,
+                        SHPC_SLOT_STATUS_PRSNT_MASK);
+        shpc->config[SHPC_SLOT_EVENT_LATCH(slot)] |=
+            SHPC_SLOT_EVENT_MRL |
+            SHPC_SLOT_EVENT_PRESENCE;
     }
 }
 
-static void shpc_command(SHPCDevice *shpc)
+static void shpc_command(PCIDevice *d)
 {
+    SHPCDevice *shpc = d->shpc;
     uint8_t code = pci_get_byte(shpc->config + SHPC_CMD_CODE);
     uint8_t speed;
     uint8_t target;
@@ -328,7 +338,7 @@ static void shpc_command(SHPCDevice *shpc)
         state = (code & SHPC_SLOT_STATE_MASK) >> SHPC_SLOT_STATE_SHIFT;
         power = (code & SHPC_SLOT_PWR_LED_MASK) >> SHPC_SLOT_PWR_LED_SHIFT;
         attn = (code & SHPC_SLOT_ATTN_LED_MASK) >> SHPC_SLOT_ATTN_LED_SHIFT;
-        shpc_slot_command(shpc, target, state, power, attn);
+        shpc_slot_command(d, target, state, power, attn);
         break;
     case 0x40 ... 0x47:
         speed = code & SHPC_SEC_BUS_MASK;
@@ -346,10 +356,10 @@ static void shpc_command(SHPCDevice *shpc)
         }
         for (i = 0; i < shpc->nslots; ++i) {
             if (!(shpc_get_status(shpc, i, SHPC_SLOT_STATUS_MRL_OPEN))) {
-                shpc_slot_command(shpc, i + SHPC_CMD_TRGT_MIN,
+                shpc_slot_command(d, i + SHPC_CMD_TRGT_MIN,
                                   SHPC_STATE_PWRONLY, SHPC_LED_ON, SHPC_LED_NO);
             } else {
-                shpc_slot_command(shpc, i + SHPC_CMD_TRGT_MIN,
+                shpc_slot_command(d, i + SHPC_CMD_TRGT_MIN,
                                   SHPC_STATE_NO, SHPC_LED_OFF, SHPC_LED_NO);
             }
         }
@@ -367,10 +377,10 @@ static void shpc_command(SHPCDevice *shpc)
         }
         for (i = 0; i < shpc->nslots; ++i) {
             if (!(shpc_get_status(shpc, i, SHPC_SLOT_STATUS_MRL_OPEN))) {
-                shpc_slot_command(shpc, i + SHPC_CMD_TRGT_MIN,
+                shpc_slot_command(d, i + SHPC_CMD_TRGT_MIN,
                                   SHPC_STATE_ENABLED, SHPC_LED_ON, SHPC_LED_NO);
             } else {
-                shpc_slot_command(shpc, i + SHPC_CMD_TRGT_MIN,
+                shpc_slot_command(d, i + SHPC_CMD_TRGT_MIN,
                                   SHPC_STATE_NO, SHPC_LED_OFF, SHPC_LED_NO);
             }
         }
@@ -402,7 +412,7 @@ static void shpc_write(PCIDevice *d, unsigned addr, uint64_t val, int l)
         shpc->config[a] &= ~(val & w1cmask); /* W1C: Write 1 to Clear */
     }
     if (ranges_overlap(addr, l, SHPC_CMD_CODE, 2)) {
-        shpc_command(shpc);
+        shpc_command(d);
     }
     shpc_interrupt_update(d);
 }
@@ -486,8 +496,9 @@ static const MemoryRegionOps shpc_mmio_ops = {
         .max_access_size = 4,
     },
 };
-static void shpc_device_plug_common(PCIDevice *affected_dev, int *slot,
-                                    SHPCDevice *shpc, Error **errp)
+
+static bool shpc_device_get_slot(PCIDevice *affected_dev, int *slot,
+                                 SHPCDevice *shpc, Error **errp)
 {
     int pci_slot = PCI_SLOT(affected_dev->devfn);
     *slot = SHPC_PCI_TO_IDX(pci_slot);
@@ -497,21 +508,20 @@ static void shpc_device_plug_common(PCIDevice *affected_dev, int *slot,
                    "controller. Valid slots are between %d and %d.",
                    pci_slot, SHPC_IDX_TO_PCI(0),
                    SHPC_IDX_TO_PCI(shpc->nslots) - 1);
-        return;
+        return false;
     }
+
+    return true;
 }
 
 void shpc_device_plug_cb(HotplugHandler *hotplug_dev, DeviceState *dev,
                             Error **errp)
 {
-    Error *local_err = NULL;
     PCIDevice *pci_hotplug_dev = PCI_DEVICE(hotplug_dev);
     SHPCDevice *shpc = pci_hotplug_dev->shpc;
     int slot;
 
-    shpc_device_plug_common(PCI_DEVICE(dev), &slot, shpc, &local_err);
-    if (local_err) {
-        error_propagate(errp, local_err);
+    if (!shpc_device_get_slot(PCI_DEVICE(dev), &slot, shpc, errp)) {
         return;
     }
 
@@ -553,16 +563,13 @@ void shpc_device_unplug_cb(HotplugHandler *hotplug_dev, DeviceState *dev,
 void shpc_device_unplug_request_cb(HotplugHandler *hotplug_dev,
                                    DeviceState *dev, Error **errp)
 {
-    Error *local_err = NULL;
     PCIDevice *pci_hotplug_dev = PCI_DEVICE(hotplug_dev);
     SHPCDevice *shpc = pci_hotplug_dev->shpc;
     uint8_t state;
     uint8_t led;
     int slot;
 
-    shpc_device_plug_common(PCI_DEVICE(dev), &slot, shpc, &local_err);
-    if (local_err) {
-        error_propagate(errp, local_err);
+    if (!shpc_device_get_slot(PCI_DEVICE(dev), &slot, shpc, errp)) {
         return;
     }
 
diff --git a/hw/pci/trace-events b/hw/pci/trace-events
index aaf46bc..4243086 100644
--- a/hw/pci/trace-events
+++ b/hw/pci/trace-events
@@ -3,6 +3,7 @@
 # pci.c
 pci_update_mappings_del(const char *dev, uint32_t bus, uint32_t slot, uint32_t func, int bar, uint64_t addr, uint64_t size) "%s %02x:%02x.%x %d,0x%"PRIx64"+0x%"PRIx64
 pci_update_mappings_add(const char *dev, uint32_t bus, uint32_t slot, uint32_t func, int bar, uint64_t addr, uint64_t size) "%s %02x:%02x.%x %d,0x%"PRIx64"+0x%"PRIx64
+pci_route_irq(int dev_irq, const char *dev_path, int parent_irq, const char *parent_path) "IRQ %d @%s -> IRQ %d @%s"
 
 # pci_host.c
 pci_cfg_read(const char *dev, uint32_t bus, uint32_t slot, uint32_t func, unsigned offs, unsigned val) "%s %02x:%02x.%x @0x%x -> 0x%x"