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

virtio,pci,pc: features,fixes,cleanups

New virtio mem options.
A vhost-user cleanup.
Control over smbios entry point type.
Config interrupt support for vdpa.
Fixes, cleanups all over the place.

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

# gpg: Signature made Fri 07 Jan 2022 04:30:41 PM PST
# gpg:                using RSA key 5D09FD0871C8F85B94CA8A0D281F0DB8D28D5469
# gpg:                issuer "mst@redhat.com"
# gpg: Good signature from "Michael S. Tsirkin <mst@kernel.org>" [undefined]
# gpg:                 aka "Michael S. Tsirkin <mst@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: 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 git://git.kernel.org/pub/scm/virt/kvm/mst/qemu: (55 commits)
  tests: acpi: Add updated TPM related tables
  acpi: tpm: Add missing device identification objects
  tests: acpi: prepare for updated TPM related tables
  virtio/vhost-vsock: don't double close vhostfd, remove redundant cleanup
  hw/scsi/vhost-scsi: don't double close vhostfd on error
  hw/scsi/vhost-scsi: don't leak vqs on error
  docs: reSTify virtio-balloon-stats documentation and move to docs/interop
  hw/i386/pc: Add missing property descriptions
  acpihp: simplify acpi_pcihp_disable_root_bus
  tests: acpi: SLIC: update expected blobs
  tests: acpi: add SLIC table test
  tests: acpi: whitelist expected blobs before changing them
  acpi: fix QEMU crash when started with SLIC table
  intel-iommu: correctly check passthrough during translation
  virtio-mem: Set "unplugged-inaccessible=auto" for the 7.0 machine on x86
  virtio-mem: Support VIRTIO_MEM_F_UNPLUGGED_INACCESSIBLE
  linux-headers: sync VIRTIO_MEM_F_UNPLUGGED_INACCESSIBLE
  MAINTAINERS: Add a separate entry for acpi/VIOT tables
  virtio: signal after wrapping packed used_idx
  virtio-mem: Support "prealloc=on" option
  ...

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

1 files changed, 177 insertions, 55 deletions

diff --git a/util/oslib-posix.c b/util/oslib-posix.c
index e8bdb02..9efdc74 100644
--- a/util/oslib-posix.c
+++ b/util/oslib-posix.c
@@ -35,11 +35,13 @@
 #include "sysemu/sysemu.h"
 #include "trace.h"
 #include "qapi/error.h"
+#include "qemu/error-report.h"
 #include "qemu/sockets.h"
 #include "qemu/thread.h"
 #include <libgen.h>
 #include "qemu/cutils.h"
 #include "qemu/compiler.h"
+#include "qemu/units.h"
 
 #ifdef CONFIG_LINUX
 #include <sys/syscall.h>
@@ -73,22 +75,32 @@
 
 #define MAX_MEM_PREALLOC_THREAD_COUNT 16
 
+struct MemsetThread;
+
+typedef struct MemsetContext {
+    bool all_threads_created;
+    bool any_thread_failed;
+    struct MemsetThread *threads;
+    int num_threads;
+} MemsetContext;
+
 struct MemsetThread {
     char *addr;
     size_t numpages;
     size_t hpagesize;
     QemuThread pgthread;
     sigjmp_buf env;
+    MemsetContext *context;
 };
 typedef struct MemsetThread MemsetThread;
 
-static MemsetThread *memset_thread;
-static int memset_num_threads;
-static bool memset_thread_failed;
+/* used by sigbus_handler() */
+static MemsetContext *sigbus_memset_context;
+struct sigaction sigbus_oldact;
+static QemuMutex sigbus_mutex;
 
 static QemuMutex page_mutex;
 static QemuCond page_cond;
-static bool threads_created_flag;
 
 int qemu_get_thread_id(void)
 {
@@ -436,22 +448,50 @@ const char *qemu_get_exec_dir(void)
     return exec_dir;
 }
 
+#ifdef CONFIG_LINUX
+static void sigbus_handler(int signal, siginfo_t *siginfo, void *ctx)
+#else /* CONFIG_LINUX */
 static void sigbus_handler(int signal)
+#endif /* CONFIG_LINUX */
 {
     int i;
-    if (memset_thread) {
-        for (i = 0; i < memset_num_threads; i++) {
-            if (qemu_thread_is_self(&memset_thread[i].pgthread)) {
-                siglongjmp(memset_thread[i].env, 1);
+
+    if (sigbus_memset_context) {
+        for (i = 0; i < sigbus_memset_context->num_threads; i++) {
+            MemsetThread *thread = &sigbus_memset_context->threads[i];
+
+            if (qemu_thread_is_self(&thread->pgthread)) {
+                siglongjmp(thread->env, 1);
             }
         }
     }
+
+#ifdef CONFIG_LINUX
+    /*
+     * We assume that the MCE SIGBUS handler could have been registered. We
+     * should never receive BUS_MCEERR_AO on any of our threads, but only on
+     * the main thread registered for PR_MCE_KILL_EARLY. Further, we should not
+     * receive BUS_MCEERR_AR triggered by action of other threads on one of
+     * our threads. So, no need to check for unrelated SIGBUS when seeing one
+     * for our threads.
+     *
+     * We will forward to the MCE handler, which will either handle the SIGBUS
+     * or reinstall the default SIGBUS handler and reraise the SIGBUS. The
+     * default SIGBUS handler will crash the process, so we don't care.
+     */
+    if (sigbus_oldact.sa_flags & SA_SIGINFO) {
+        sigbus_oldact.sa_sigaction(signal, siginfo, ctx);
+        return;
+    }
+#endif /* CONFIG_LINUX */
+    warn_report("os_mem_prealloc: unrelated SIGBUS detected and ignored");
 }
 
 static void *do_touch_pages(void *arg)
 {
     MemsetThread *memset_args = (MemsetThread *)arg;
     sigset_t set, oldset;
+    int ret = 0;
 
     /*
      * On Linux, the page faults from the loop below can cause mmap_sem
@@ -459,7 +499,7 @@ static void *do_touch_pages(void *arg)
      * clearing until all threads have been created.
      */
     qemu_mutex_lock(&page_mutex);
-    while(!threads_created_flag){
+    while (!memset_args->context->all_threads_created) {
         qemu_cond_wait(&page_cond, &page_mutex);
     }
     qemu_mutex_unlock(&page_mutex);
@@ -470,7 +510,7 @@ static void *do_touch_pages(void *arg)
     pthread_sigmask(SIG_UNBLOCK, &set, &oldset);
 
     if (sigsetjmp(memset_args->env, 1)) {
-        memset_thread_failed = true;
+        ret = -EFAULT;
     } else {
         char *addr = memset_args->addr;
         size_t numpages = memset_args->numpages;
@@ -484,20 +524,37 @@ static void *do_touch_pages(void *arg)
              *
              * 'volatile' to stop compiler optimizing this away
              * to a no-op
-             *
-             * TODO: get a better solution from kernel so we
-             * don't need to write at all so we don't cause
-             * wear on the storage backing the region...
              */
             *(volatile char *)addr = *addr;
             addr += hpagesize;
         }
     }
     pthread_sigmask(SIG_SETMASK, &oldset, NULL);
-    return NULL;
+    return (void *)(uintptr_t)ret;
 }
 
-static inline int get_memset_num_threads(int smp_cpus)
+static void *do_madv_populate_write_pages(void *arg)
+{
+    MemsetThread *memset_args = (MemsetThread *)arg;
+    const size_t size = memset_args->numpages * memset_args->hpagesize;
+    char * const addr = memset_args->addr;
+    int ret = 0;
+
+    /* See do_touch_pages(). */
+    qemu_mutex_lock(&page_mutex);
+    while (!memset_args->context->all_threads_created) {
+        qemu_cond_wait(&page_cond, &page_mutex);
+    }
+    qemu_mutex_unlock(&page_mutex);
+
+    if (size && qemu_madvise(addr, size, QEMU_MADV_POPULATE_WRITE)) {
+        ret = -errno;
+    }
+    return (void *)(uintptr_t)ret;
+}
+
+static inline int get_memset_num_threads(size_t hpagesize, size_t numpages,
+                                         int smp_cpus)
 {
     long host_procs = sysconf(_SC_NPROCESSORS_ONLN);
     int ret = 1;
@@ -505,17 +562,27 @@ static inline int get_memset_num_threads(int smp_cpus)
     if (host_procs > 0) {
         ret = MIN(MIN(host_procs, MAX_MEM_PREALLOC_THREAD_COUNT), smp_cpus);
     }
+
+    /* Especially with gigantic pages, don't create more threads than pages. */
+    ret = MIN(ret, numpages);
+    /* Don't start threads to prealloc comparatively little memory. */
+    ret = MIN(ret, MAX(1, hpagesize * numpages / (64 * MiB)));
+
     /* In case sysconf() fails, we fall back to single threaded */
     return ret;
 }
 
-static bool touch_all_pages(char *area, size_t hpagesize, size_t numpages,
-                            int smp_cpus)
+static int touch_all_pages(char *area, size_t hpagesize, size_t numpages,
+                           int smp_cpus, bool use_madv_populate_write)
 {
     static gsize initialized = 0;
+    MemsetContext context = {
+        .num_threads = get_memset_num_threads(hpagesize, numpages, smp_cpus),
+    };
     size_t numpages_per_thread, leftover;
+    void *(*touch_fn)(void *);
+    int ret = 0, i = 0;
     char *addr = area;
-    int i = 0;
 
     if (g_once_init_enter(&initialized)) {
         qemu_mutex_init(&page_mutex);
@@ -523,66 +590,121 @@ static bool touch_all_pages(char *area, size_t hpagesize, size_t numpages,
         g_once_init_leave(&initialized, 1);
     }
 
-    memset_thread_failed = false;
-    threads_created_flag = false;
-    memset_num_threads = get_memset_num_threads(smp_cpus);
-    memset_thread = g_new0(MemsetThread, memset_num_threads);
-    numpages_per_thread = numpages / memset_num_threads;
-    leftover = numpages % memset_num_threads;
-    for (i = 0; i < memset_num_threads; i++) {
-        memset_thread[i].addr = addr;
-        memset_thread[i].numpages = numpages_per_thread + (i < leftover);
-        memset_thread[i].hpagesize = hpagesize;
-        qemu_thread_create(&memset_thread[i].pgthread, "touch_pages",
-                           do_touch_pages, &memset_thread[i],
+    if (use_madv_populate_write) {
+        /* Avoid creating a single thread for MADV_POPULATE_WRITE */
+        if (context.num_threads == 1) {
+            if (qemu_madvise(area, hpagesize * numpages,
+                             QEMU_MADV_POPULATE_WRITE)) {
+                return -errno;
+            }
+            return 0;
+        }
+        touch_fn = do_madv_populate_write_pages;
+    } else {
+        touch_fn = do_touch_pages;
+    }
+
+    context.threads = g_new0(MemsetThread, context.num_threads);
+    numpages_per_thread = numpages / context.num_threads;
+    leftover = numpages % context.num_threads;
+    for (i = 0; i < context.num_threads; i++) {
+        context.threads[i].addr = addr;
+        context.threads[i].numpages = numpages_per_thread + (i < leftover);
+        context.threads[i].hpagesize = hpagesize;
+        context.threads[i].context = &context;
+        qemu_thread_create(&context.threads[i].pgthread, "touch_pages",
+                           touch_fn, &context.threads[i],
                            QEMU_THREAD_JOINABLE);
-        addr += memset_thread[i].numpages * hpagesize;
+        addr += context.threads[i].numpages * hpagesize;
+    }
+
+    if (!use_madv_populate_write) {
+        sigbus_memset_context = &context;
     }
 
     qemu_mutex_lock(&page_mutex);
-    threads_created_flag = true;
+    context.all_threads_created = true;
     qemu_cond_broadcast(&page_cond);
     qemu_mutex_unlock(&page_mutex);
 
-    for (i = 0; i < memset_num_threads; i++) {
-        qemu_thread_join(&memset_thread[i].pgthread);
+    for (i = 0; i < context.num_threads; i++) {
+        int tmp = (uintptr_t)qemu_thread_join(&context.threads[i].pgthread);
+
+        if (tmp) {
+            ret = tmp;
+        }
     }
-    g_free(memset_thread);
-    memset_thread = NULL;
 
-    return memset_thread_failed;
+    if (!use_madv_populate_write) {
+        sigbus_memset_context = NULL;
+    }
+    g_free(context.threads);
+
+    return ret;
+}
+
+static bool madv_populate_write_possible(char *area, size_t pagesize)
+{
+    return !qemu_madvise(area, pagesize, QEMU_MADV_POPULATE_WRITE) ||
+           errno != EINVAL;
 }
 
 void os_mem_prealloc(int fd, char *area, size_t memory, int smp_cpus,
                      Error **errp)
 {
+    static gsize initialized;
     int ret;
-    struct sigaction act, oldact;
     size_t hpagesize = qemu_fd_getpagesize(fd);
     size_t numpages = DIV_ROUND_UP(memory, hpagesize);
+    bool use_madv_populate_write;
+    struct sigaction act;
 
-    memset(&act, 0, sizeof(act));
-    act.sa_handler = &sigbus_handler;
-    act.sa_flags = 0;
+    /*
+     * Sense on every invocation, as MADV_POPULATE_WRITE cannot be used for
+     * some special mappings, such as mapping /dev/mem.
+     */
+    use_madv_populate_write = madv_populate_write_possible(area, hpagesize);
 
-    ret = sigaction(SIGBUS, &act, &oldact);
-    if (ret) {
-        error_setg_errno(errp, errno,
-            "os_mem_prealloc: failed to install signal handler");
-        return;
+    if (!use_madv_populate_write) {
+        if (g_once_init_enter(&initialized)) {
+            qemu_mutex_init(&sigbus_mutex);
+            g_once_init_leave(&initialized, 1);
+        }
+
+        qemu_mutex_lock(&sigbus_mutex);
+        memset(&act, 0, sizeof(act));
+#ifdef CONFIG_LINUX
+        act.sa_sigaction = &sigbus_handler;
+        act.sa_flags = SA_SIGINFO;
+#else /* CONFIG_LINUX */
+        act.sa_handler = &sigbus_handler;
+        act.sa_flags = 0;
+#endif /* CONFIG_LINUX */
+
+        ret = sigaction(SIGBUS, &act, &sigbus_oldact);
+        if (ret) {
+            error_setg_errno(errp, errno,
+                "os_mem_prealloc: failed to install signal handler");
+            return;
+        }
     }
 
     /* touch pages simultaneously */
-    if (touch_all_pages(area, hpagesize, numpages, smp_cpus)) {
-        error_setg(errp, "os_mem_prealloc: Insufficient free host memory "
-            "pages available to allocate guest RAM");
+    ret = touch_all_pages(area, hpagesize, numpages, smp_cpus,
+                          use_madv_populate_write);
+    if (ret) {
+        error_setg_errno(errp, -ret,
+                         "os_mem_prealloc: preallocating memory failed");
     }
 
-    ret = sigaction(SIGBUS, &oldact, NULL);
-    if (ret) {
-        /* Terminate QEMU since it can't recover from error */
-        perror("os_mem_prealloc: failed to reinstall signal handler");
-        exit(1);
+    if (!use_madv_populate_write) {
+        ret = sigaction(SIGBUS, &sigbus_oldact, NULL);
+        if (ret) {
+            /* Terminate QEMU since it can't recover from error */
+            perror("os_mem_prealloc: failed to reinstall signal handler");
+            exit(1);
+        }
+        qemu_mutex_unlock(&sigbus_mutex);
     }
 }