Merge tag 'for-upstream' of https://gitlab.com/bonzini/qemu into staging

* kvm: enable dirty ring for arm64
* target/i386: new features
* target/i386: AVX fixes
* configure: create a python venv unconditionally
* meson: bump to 0.63.0 and move tests from configure
* meson: Pass -j option to sphinx
* drop support for Python 3.6
* fix check-python-tox
* fix "make clean" in the source directory

# -----BEGIN PGP SIGNATURE-----
#
# iQFIBAABCAAyFiEE8TM4V0tmI4mGbHaCv/vSX3jHroMFAmRmDYQUHHBib256aW5p
# QHJlZGhhdC5jb20ACgkQv/vSX3jHroOXSwf/WKmYPe09yHfxfVSFsSz83QpB3e+f
# KJx6FdyMMt26ZQJpcqorobrDV23R8FyxngXPkwoxqobAEtXB/AH0/S/u8RUZ46Qt
# IrF8FXr4ZdyLW7CW6nmIejmlul0iRmFD7D98E6dZ3QXfype3Ifra7gG74spZ1B44
# ZNvaomJKUK8Ga8rbChs9KtgrxlOC5q8IfTWF5ZExmZszPC9NRnZmU5Oncnuwek9T
# Ic6zDPoAeF3jDtovZhxg1HAB9e/ENZX/V9NjO92yZa8u/TITQ88l4tJctf7uiLxO
# 2oGY12ln8i//pbjyUe4iM+bNh5+reAChEI8iv7WxEsj9s2HBUJ68f3tpbQ==
# =Zg00
# -----END PGP SIGNATURE-----
# gpg: Signature made Thu 18 May 2023 04:35:32 AM PDT
# gpg:                using RSA key F13338574B662389866C7682BFFBD25F78C7AE83
# gpg:                issuer "pbonzini@redhat.com"
# gpg: Good signature from "Paolo Bonzini <bonzini@gnu.org>" [undefined]
# gpg:                 aka "Paolo Bonzini <pbonzini@redhat.com>" [undefined]
# 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: 46F5 9FBD 57D6 12E7 BFD4  E2F7 7E15 100C CD36 69B1
#      Subkey fingerprint: F133 3857 4B66 2389 866C  7682 BFFB D25F 78C7 AE83

* tag 'for-upstream' of https://gitlab.com/bonzini/qemu: (68 commits)
  docs/devel: update build system docs
  configure: remove unnecessary check
  configure: reorder option parsing code
  configure: remove unnecessary mkdir
  configure: do not rerun the tests with -Werror
  configure: remove compiler sanity check
  build: move --disable-debug-info to meson
  build: move compiler version check to meson
  build: move remaining compiler flag tests to meson
  build: move warning flag selection to meson
  build: move stack protector flag selection to meson
  build: move coroutine backend selection to meson
  build: move SafeStack tests to meson
  build: move sanitizer tests to meson
  meson: prepare move of QEMU_CFLAGS to meson
  configure, meson: move --enable-modules to Meson
  configure: remove pkg-config functions
  build: move glib detection and workarounds to meson
  meson: drop unnecessary declare_dependency()
  meson: add more version numbers to the summary
  ...

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>

2 files changed, 85 insertions, 34 deletions

diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c
index cf3a88d..7679f39 100644
--- a/accel/kvm/kvm-all.c
+++ b/accel/kvm/kvm-all.c
@@ -1361,6 +1361,10 @@ static void kvm_set_phys_mem(KVMMemoryListener *kml,
                  */
                 if (kvm_state->kvm_dirty_ring_size) {
                     kvm_dirty_ring_reap_locked(kvm_state, NULL);
+                    if (kvm_state->kvm_dirty_ring_with_bitmap) {
+                        kvm_slot_sync_dirty_pages(mem);
+                        kvm_slot_get_dirty_log(kvm_state, mem);
+                    }
                 } else {
                     kvm_slot_get_dirty_log(kvm_state, mem);
                 }
@@ -1458,6 +1462,69 @@ static int kvm_dirty_ring_reaper_init(KVMState *s)
     return 0;
 }
 
+static int kvm_dirty_ring_init(KVMState *s)
+{
+    uint32_t ring_size = s->kvm_dirty_ring_size;
+    uint64_t ring_bytes = ring_size * sizeof(struct kvm_dirty_gfn);
+    unsigned int capability = KVM_CAP_DIRTY_LOG_RING;
+    int ret;
+
+    s->kvm_dirty_ring_size = 0;
+    s->kvm_dirty_ring_bytes = 0;
+
+    /* Bail if the dirty ring size isn't specified */
+    if (!ring_size) {
+        return 0;
+    }
+
+    /*
+     * Read the max supported pages. Fall back to dirty logging mode
+     * if the dirty ring isn't supported.
+     */
+    ret = kvm_vm_check_extension(s, capability);
+    if (ret <= 0) {
+        capability = KVM_CAP_DIRTY_LOG_RING_ACQ_REL;
+        ret = kvm_vm_check_extension(s, capability);
+    }
+
+    if (ret <= 0) {
+        warn_report("KVM dirty ring not available, using bitmap method");
+        return 0;
+    }
+
+    if (ring_bytes > ret) {
+        error_report("KVM dirty ring size %" PRIu32 " too big "
+                     "(maximum is %ld).  Please use a smaller value.",
+                     ring_size, (long)ret / sizeof(struct kvm_dirty_gfn));
+        return -EINVAL;
+    }
+
+    ret = kvm_vm_enable_cap(s, capability, 0, ring_bytes);
+    if (ret) {
+        error_report("Enabling of KVM dirty ring failed: %s. "
+                     "Suggested minimum value is 1024.", strerror(-ret));
+        return -EIO;
+    }
+
+    /* Enable the backup bitmap if it is supported */
+    ret = kvm_vm_check_extension(s, KVM_CAP_DIRTY_LOG_RING_WITH_BITMAP);
+    if (ret > 0) {
+        ret = kvm_vm_enable_cap(s, KVM_CAP_DIRTY_LOG_RING_WITH_BITMAP, 0);
+        if (ret) {
+            error_report("Enabling of KVM dirty ring's backup bitmap failed: "
+                         "%s. ", strerror(-ret));
+            return -EIO;
+        }
+
+        s->kvm_dirty_ring_with_bitmap = true;
+    }
+
+    s->kvm_dirty_ring_size = ring_size;
+    s->kvm_dirty_ring_bytes = ring_bytes;
+
+    return 0;
+}
+
 static void kvm_region_add(MemoryListener *listener,
                            MemoryRegionSection *section)
 {
@@ -1563,7 +1630,7 @@ static void kvm_log_sync(MemoryListener *listener,
     kvm_slots_unlock();
 }
 
-static void kvm_log_sync_global(MemoryListener *l)
+static void kvm_log_sync_global(MemoryListener *l, bool last_stage)
 {
     KVMMemoryListener *kml = container_of(l, KVMMemoryListener, listener);
     KVMState *s = kvm_state;
@@ -1582,6 +1649,12 @@ static void kvm_log_sync_global(MemoryListener *l)
         mem = &kml->slots[i];
         if (mem->memory_size && mem->flags & KVM_MEM_LOG_DIRTY_PAGES) {
             kvm_slot_sync_dirty_pages(mem);
+
+            if (s->kvm_dirty_ring_with_bitmap && last_stage &&
+                kvm_slot_get_dirty_log(s, mem)) {
+                kvm_slot_sync_dirty_pages(mem);
+            }
+
             /*
              * This is not needed by KVM_GET_DIRTY_LOG because the
              * ioctl will unconditionally overwrite the whole region.
@@ -2521,35 +2594,9 @@ static int kvm_init(MachineState *ms)
      * Enable KVM dirty ring if supported, otherwise fall back to
      * dirty logging mode
      */
-    if (s->kvm_dirty_ring_size > 0) {
-        uint64_t ring_bytes;
-
-        ring_bytes = s->kvm_dirty_ring_size * sizeof(struct kvm_dirty_gfn);
-
-        /* Read the max supported pages */
-        ret = kvm_vm_check_extension(s, KVM_CAP_DIRTY_LOG_RING);
-        if (ret > 0) {
-            if (ring_bytes > ret) {
-                error_report("KVM dirty ring size %" PRIu32 " too big "
-                             "(maximum is %ld).  Please use a smaller value.",
-                             s->kvm_dirty_ring_size,
-                             (long)ret / sizeof(struct kvm_dirty_gfn));
-                ret = -EINVAL;
-                goto err;
-            }
-
-            ret = kvm_vm_enable_cap(s, KVM_CAP_DIRTY_LOG_RING, 0, ring_bytes);
-            if (ret) {
-                error_report("Enabling of KVM dirty ring failed: %s. "
-                             "Suggested minimum value is 1024.", strerror(-ret));
-                goto err;
-            }
-
-            s->kvm_dirty_ring_bytes = ring_bytes;
-         } else {
-             warn_report("KVM dirty ring not available, using bitmap method");
-             s->kvm_dirty_ring_size = 0;
-        }
+    ret = kvm_dirty_ring_init(s);
+    if (ret < 0) {
+        goto err;
     }
 
     /*
@@ -3710,6 +3757,7 @@ static void kvm_accel_instance_init(Object *obj)
     s->kernel_irqchip_split = ON_OFF_AUTO_AUTO;
     /* KVM dirty ring is by default off */
     s->kvm_dirty_ring_size = 0;
+    s->kvm_dirty_ring_with_bitmap = false;
     s->notify_vmexit = NOTIFY_VMEXIT_OPTION_RUN;
     s->notify_window = 0;
     s->xen_version = 0;
diff --git a/accel/tcg/tcg-accel-ops-rr.c b/accel/tcg/tcg-accel-ops-rr.c
index 5788efa..b6d10fa 100644
--- a/accel/tcg/tcg-accel-ops-rr.c
+++ b/accel/tcg/tcg-accel-ops-rr.c
@@ -72,11 +72,13 @@ static void rr_kick_next_cpu(void)
 {
     CPUState *cpu;
     do {
-        cpu = qatomic_mb_read(&rr_current_cpu);
+        cpu = qatomic_read(&rr_current_cpu);
         if (cpu) {
             cpu_exit(cpu);
         }
-    } while (cpu != qatomic_mb_read(&rr_current_cpu));
+        /* Finish kicking this cpu before reading again.  */
+        smp_mb();
+    } while (cpu != qatomic_read(&rr_current_cpu));
 }
 
 static void rr_kick_thread(void *opaque)
@@ -241,8 +243,9 @@ static void *rr_cpu_thread_fn(void *arg)
         }
 
         while (cpu && cpu_work_list_empty(cpu) && !cpu->exit_request) {
-
+            /* Store rr_current_cpu before evaluating cpu_can_run().  */
             qatomic_mb_set(&rr_current_cpu, cpu);
+
             current_cpu = cpu;
 
             qemu_clock_enable(QEMU_CLOCK_VIRTUAL,
@@ -280,7 +283,7 @@ static void *rr_cpu_thread_fn(void *arg)
             cpu = CPU_NEXT(cpu);
         } /* while (cpu && !cpu->exit_request).. */
 
-        /* Does not need qatomic_mb_set because a spurious wakeup is okay.  */
+        /* Does not need a memory barrier because a spurious wakeup is okay.  */
         qatomic_set(&rr_current_cpu, NULL);
 
         if (cpu && cpu->exit_request) {