Import upstream 1.6.0.upstream/1.6.0

Change-Id: Icf52b556470cac8677297f2ef14ded16684f7887
Signed-off-by: Junfeng Dong <junfeng.dong@intel.com>

1 files changed, 63 insertions, 76 deletions

diff --git a/aio-posix.c b/aio-posix.c
index 05cc84e12..b68eccd40 100644
--- a/aio-posix.c
+++ b/aio-posix.c
@@ -14,9 +14,9 @@
  */
 
 #include "qemu-common.h"
-#include "block.h"
-#include "qemu-queue.h"
-#include "qemu_socket.h"
+#include "block/block.h"
+#include "qemu/queue.h"
+#include "qemu/sockets.h"
 
 struct AioHandler
 {
@@ -25,6 +25,7 @@ struct AioHandler
     IOHandler *io_write;
     AioFlushHandler *io_flush;
     int deleted;
+    int pollfds_idx;
     void *opaque;
     QLIST_ENTRY(AioHandler) node;
 };
@@ -85,9 +86,10 @@ void aio_set_fd_handler(AioContext *ctx,
         node->io_write = io_write;
         node->io_flush = io_flush;
         node->opaque = opaque;
+        node->pollfds_idx = -1;
 
-        node->pfd.events = (io_read ? G_IO_IN | G_IO_HUP : 0);
-        node->pfd.events |= (io_write ? G_IO_OUT : 0);
+        node->pfd.events = (io_read ? G_IO_IN | G_IO_HUP | G_IO_ERR : 0);
+        node->pfd.events |= (io_write ? G_IO_OUT | G_IO_ERR : 0);
     }
 
     aio_notify(ctx);
@@ -110,13 +112,6 @@ bool aio_pending(AioContext *ctx)
     QLIST_FOREACH(node, &ctx->aio_handlers, node) {
         int revents;
 
-        /*
-         * FIXME: right now we cannot get G_IO_HUP and G_IO_ERR because
-         * main-loop.c is still select based (due to the slirp legacy).
-         * If main-loop.c ever switches to poll, G_IO_ERR should be
-         * tested too.  Dispatching G_IO_ERR to both handlers should be
-         * okay, since handlers need to be ready for spurious wakeups.
-         */
         revents = node->pfd.revents & node->pfd.events;
         if (revents & (G_IO_IN | G_IO_HUP | G_IO_ERR) && node->io_read) {
             return true;
@@ -129,30 +124,12 @@ bool aio_pending(AioContext *ctx)
     return false;
 }
 
-bool aio_poll(AioContext *ctx, bool blocking)
+static bool aio_dispatch(AioContext *ctx)
 {
-    static struct timeval tv0;
     AioHandler *node;
-    fd_set rdfds, wrfds;
-    int max_fd = -1;
-    int ret;
-    bool busy, progress;
-
-    progress = false;
+    bool progress = false;
 
     /*
-     * If there are callbacks left that have been queued, we need to call then.
-     * Do not call select in this case, because it is possible that the caller
-     * does not need a complete flush (as is the case for qemu_aio_wait loops).
-     */
-    if (aio_bh_poll(ctx)) {
-        blocking = false;
-        progress = true;
-    }
-
-    /*
-     * Then dispatch any pending callbacks from the GSource.
-     *
      * We have to walk very carefully in case qemu_aio_set_fd_handler is
      * called while we're walking.
      */
@@ -166,12 +143,15 @@ bool aio_poll(AioContext *ctx, bool blocking)
         revents = node->pfd.revents & node->pfd.events;
         node->pfd.revents = 0;
 
-        /* See comment in aio_pending.  */
-        if (revents & (G_IO_IN | G_IO_HUP | G_IO_ERR) && node->io_read) {
+        if (!node->deleted &&
+            (revents & (G_IO_IN | G_IO_HUP | G_IO_ERR)) &&
+            node->io_read) {
             node->io_read(node->opaque);
             progress = true;
         }
-        if (revents & (G_IO_OUT | G_IO_ERR) && node->io_write) {
+        if (!node->deleted &&
+            (revents & (G_IO_OUT | G_IO_ERR)) &&
+            node->io_write) {
             node->io_write(node->opaque);
             progress = true;
         }
@@ -186,6 +166,30 @@ bool aio_poll(AioContext *ctx, bool blocking)
             g_free(tmp);
         }
     }
+    return progress;
+}
+
+bool aio_poll(AioContext *ctx, bool blocking)
+{
+    AioHandler *node;
+    int ret;
+    bool busy, progress;
+
+    progress = false;
+
+    /*
+     * If there are callbacks left that have been queued, we need to call them.
+     * Do not call select in this case, because it is possible that the caller
+     * does not need a complete flush (as is the case for qemu_aio_wait loops).
+     */
+    if (aio_bh_poll(ctx)) {
+        blocking = false;
+        progress = true;
+    }
+
+    if (aio_dispatch(ctx)) {
+        progress = true;
+    }
 
     if (progress && !blocking) {
         return true;
@@ -193,12 +197,13 @@ bool aio_poll(AioContext *ctx, bool blocking)
 
     ctx->walking_handlers++;
 
-    FD_ZERO(&rdfds);
-    FD_ZERO(&wrfds);
+    g_array_set_size(ctx->pollfds, 0);
 
-    /* fill fd sets */
+    /* fill pollfds */
     busy = false;
     QLIST_FOREACH(node, &ctx->aio_handlers, node) {
+        node->pollfds_idx = -1;
+
         /* If there aren't pending AIO operations, don't invoke callbacks.
          * Otherwise, if there are no AIO requests, qemu_aio_wait() would
          * wait indefinitely.
@@ -209,13 +214,13 @@ bool aio_poll(AioContext *ctx, bool blocking)
             }
             busy = true;
         }
-        if (!node->deleted && node->io_read) {
-            FD_SET(node->pfd.fd, &rdfds);
-            max_fd = MAX(max_fd, node->pfd.fd + 1);
-        }
-        if (!node->deleted && node->io_write) {
-            FD_SET(node->pfd.fd, &wrfds);
-            max_fd = MAX(max_fd, node->pfd.fd + 1);
+        if (!node->deleted && node->pfd.events) {
+            GPollFD pfd = {
+                .fd = node->pfd.fd,
+                .events = node->pfd.events,
+            };
+            node->pollfds_idx = ctx->pollfds->len;
+            g_array_append_val(ctx->pollfds, pfd);
         }
     }
 
@@ -227,42 +232,24 @@ bool aio_poll(AioContext *ctx, bool blocking)
     }
 
     /* wait until next event */
-    ret = select(max_fd, &rdfds, &wrfds, NULL, blocking ? NULL : &tv0);
+    ret = g_poll((GPollFD *)ctx->pollfds->data,
+                 ctx->pollfds->len,
+                 blocking ? -1 : 0);
 
     /* if we have any readable fds, dispatch event */
     if (ret > 0) {
-        /* we have to walk very carefully in case
-         * qemu_aio_set_fd_handler is called while we're walking */
-        node = QLIST_FIRST(&ctx->aio_handlers);
-        while (node) {
-            AioHandler *tmp;
-
-            ctx->walking_handlers++;
-
-            if (!node->deleted &&
-                FD_ISSET(node->pfd.fd, &rdfds) &&
-                node->io_read) {
-                node->io_read(node->opaque);
-                progress = true;
-            }
-            if (!node->deleted &&
-                FD_ISSET(node->pfd.fd, &wrfds) &&
-                node->io_write) {
-                node->io_write(node->opaque);
-                progress = true;
-            }
-
-            tmp = node;
-            node = QLIST_NEXT(node, node);
-
-            ctx->walking_handlers--;
-
-            if (!ctx->walking_handlers && tmp->deleted) {
-                QLIST_REMOVE(tmp, node);
-                g_free(tmp);
+        QLIST_FOREACH(node, &ctx->aio_handlers, node) {
+            if (node->pollfds_idx != -1) {
+                GPollFD *pfd = &g_array_index(ctx->pollfds, GPollFD,
+                                              node->pollfds_idx);
+                node->pfd.revents = pfd->revents;
             }
         }
+        if (aio_dispatch(ctx)) {
+            progress = true;
+        }
     }
 
-    return progress;
+    assert(progress || busy);
+    return true;
 }