[reactos] 01/01: [MOUNTMGR] Misc. fixes for WorkerThread()

[Pierre Schweitzer]

https://git.reactos.org/?p=reactos.git;a=commitdiff;h=7608ac9f71956c2ef60a1…

commit 7608ac9f71956c2ef60a18b1e52f4b4f12749d05
Author:     Pierre Schweitzer <pierre(a)reactos.org>
AuthorDate: Mon Oct 21 10:54:51 2019 +0200
Commit:     Pierre Schweitzer <pierre(a)reactos.org>
CommitDate: Mon Oct 21 10:58:14 2019 +0200

    [MOUNTMGR] Misc. fixes for WorkerThread()
    
    - Properly quit the active loop when we're out of work items;
    - Fix timeout duration (setting it to 1s);
    - Fix handling the "Unloading" variable in case of a shutdown
      so that waiting loop is properly stopped;
    - Documented why we're waiting on VolumesSafeForWriteAccess.
    
    This fixes shutting down ReactOS with work items queued.
    This is needed here because no one ever sets that event (properly)
    created by SMSS though. A. Ionescu was explaining in 2018 that it's
    autochk responsibility, but it doesn't seem to be the case in W2K3.
    To be investigated.
    
    This fix with all the previous ones and more uncommitted stuff (yet ;-))
    allows reaching the first steps towards a NT5 storage stack:
    https://twitter.com/HeisSpiter/status/1186199631740506112
---
 drivers/filters/mountmgr/database.c | 40 ++++++++++++++++++++++++++++---------
 1 file changed, 31 insertions(+), 9 deletions(-)

diff --git a/drivers/filters/mountmgr/database.c b/drivers/filters/mountmgr/database.c
index aa0c6a42ef5..694c98a25fa 100644
--- a/drivers/filters/mountmgr/database.c
+++ b/drivers/filters/mountmgr/database.c
@@ -1168,21 +1168,40 @@ WorkerThread(IN PDEVICE_OBJECT DeviceObject,
                                NULL,
                                NULL);
     KeInitializeEvent(&Event, NotificationEvent, FALSE);
-    Timeout.LowPart = 0xFFFFFFFF;
-    Timeout.HighPart = 0xFF676980;
+    Timeout.QuadPart = -10000000LL; /* Wait for 1 second */
 
-    /* Try to wait as long as possible */
-    for (i = (Unloading ? 999 : 0); i < 1000; i++)
+    /* Wait as long as possible for clearance from autochk
+     * We will write remote databases only if it is safe
+     * to access volumes.
+     * First, given we start before SMSS, wait for the
+     * event creation.
+     */
+    i = 0;
+    do
     {
-        Status = ZwOpenEvent(&SafeEvent, EVENT_ALL_ACCESS, &ObjectAttributes);
-        if (NT_SUCCESS(Status))
+        /* If we started to shutdown, stop waiting forever and jump to last attempt */
+        if (Unloading)
         {
-            break;
+            i = 999;
+        }
+        else
+        {
+            /* Attempt to open the event */
+            Status = ZwOpenEvent(&SafeEvent, EVENT_ALL_ACCESS,
&ObjectAttributes);
+            if (NT_SUCCESS(Status))
+            {
+                break;
+            }
+
+            /* Wait a bit to give SMSS a chance to create the event */
+            KeWaitForSingleObject(&Event, Executive, KernelMode, FALSE,
&Timeout);
         }
 
-        KeWaitForSingleObject(&Event, Executive, KernelMode, FALSE, &Timeout);
+        ++i;
     }
+    while (i < 1000);
 
+    /* We managed to open the event, wait until autochk signals it */
     if (i < 1000)
     {
         do
@@ -1220,7 +1239,7 @@ WorkerThread(IN PDEVICE_OBJECT DeviceObject,
         IoFreeWorkItem(WorkItem->WorkItem);
         FreePool(WorkItem);
 
-        if (InterlockedDecrement(&(DeviceExtension->WorkerReferences)) == 0)
+        if (InterlockedDecrement(&(DeviceExtension->WorkerReferences)) < 0)
         {
             return;
         }
@@ -1603,6 +1622,9 @@ ReconcileThisDatabaseWithMaster(IN PDEVICE_EXTENSION
DeviceExtension,
         return;
     }
 
+    UNIMPLEMENTED;
+    return;
+
     /* Allocate a work item */
     WorkItem = AllocatePool(sizeof(RECONCILE_WORK_ITEM));
     if (!WorkItem)